diff --git a/AGENTS.md b/AGENTS.md
index 9e3dc76fa249f0244eefc1135e5f7f3ff30b184f..b7e136226cb7219b1dc1cb40ccf6731c890c8ddb 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -6,11 +6,13 @@ Guidance for AI coding agents working in this repository.
 
 Frontend code for the TabArena leaderboard, deployed as a Gradio app on a Hugging Face Space. The leaderboard ranks tabular ML models across tasks, splits, and dataset subsets.
 
-- Entrypoint: `main.py` (Gradio app; HF Space runs it via the `app_file` declared in `README.md` frontmatter).
+- Entrypoint: `main.py` (Gradio app plus the page's CSS/JS/head; HF Space runs it via the `app_file` declared in `README.md` frontmatter).
+- Pages and their layout: `pages.py` (one entry per leaderboard in `PAGES`); component builders: `views.py`.
+- Data paths and CSV parsing: `data_loading.py`. Agent-facing endpoints: `api.py`.
 - UI text and copy: `website_texts.py`.
 - Static constants (e.g. model-type emoji map): `constants.py`.
-- Precomputed artifacts (CSVs + zipped PNGs): `data/imputation_{no,yes}/splits_{...}/tasks_{...}/datasets_{...}/`.
-- Python: requires 3.12+. Dependencies pinned via `pyproject.toml`.
+- Precomputed artifacts (CSVs + zipped PNGs): `data/imputation_{no,yes}/splits_{...}/tasks_{...}/datasets_{...}/` for TabArena, `data_beyondarena/subsets/{subset}/` for BeyondArena.
+- Python: requires 3.12+. Gradio 6 (`sdk_version` in `README.md` must match the pin in `requirements.txt`).
 
 ## Local dev
 
@@ -31,9 +33,45 @@ Do not edit files under `data/` by hand — they are generated artifacts.
 
 ## Architecture notes
 
-- `LBMatrix` (in `main.py`) enumerates the leaderboard variants along four axes: `imputation`, `splits`, `tasks`, `datasets`. The first element of each list is the default selection for that axis — changing list order changes both the default tab and the rendering order.
-- `LBContainer` loads `website_leaderboard.csv` and lazily unzips the per-view PNGs from `data/...`.
-- The reactive Gradio render is driven by four `gr.State` values, one per axis; selecting a tab updates the corresponding state and triggers `render_details`.
+- A leaderboard view is one cell of a grid, described by the frozen `Subset` dataclass (`imputation`, `splits`, `tasks`, `datasets`) whose `rel_path` *is* the artifact directory. BeyondArena uses `BeyondSubset` with a single `subset` axis. The axis dicts in `data_loading.py` (`TASK_LABELS`, `DATASET_LABELS`, `BEYOND_SUBSET_LABELS`) define both the valid values and the tab order; the first entry of each is the default.
+- `LBContainer` (`data_loading.py`) loads one subset's `website_leaderboard.csv`, lazily unzips its PNGs, and inlines its interactive HTML plots.
+- The reactive render is driven by one `gr.State` per axis: selecting a tab or toggle updates that state, and the `@gr.render` block in `pages.py` rebuilds the subset's figures and table.
+- The full leaderboard table is `views.make_leaderboard`: a native `gr.Dataframe` (sorting, search, fullscreen, copy) inside one card with its filters, column picker and column key. Filtering is `views.filter_leaderboard`, kept pure so it can be tested without a Gradio context. It replaced the `gradio_leaderboard` custom component, which pinned Gradio < 6 — do not reintroduce that dependency.
+
+## The agent-facing API (`api.py`)
+
+Hugging Face serves a generated `agents.md` for every Gradio Space at
+`huggingface.co/spaces/<id>/agents.md`, reachable from the **Agents** button on the Space page. It is not
+this file and cannot be overridden from the repo: it tells an agent to read `/gradio_api/info` and call
+what it finds. So `/gradio_api/info` is the real contract, and `api.py` owns it.
+
+- Four endpoints, registered by `register_api()` from inside the `gr.Blocks` context in `main.py`:
+  `list_leaderboards`, `get_tabarena_leaderboard`, `get_beyondarena_leaderboard` and
+  `get_pareto_frontier`. They read the same published CSVs the UI reads and reuse
+  `data_loading.parse_model`, so the API cannot disagree with the table.
+- `get_pareto_frontier` is the one to point "which model should I use" at: the leaderboard's top row
+  is often orders of magnitude slower than a model just behind it, so it returns the outright best,
+  the non-dominated frontier on quality vs train/predict time (each entry with its speedup and
+  quality gap), the best model inside a stated time budget, and a summary sentence to relay.
+- `kind` separates individual models from whole AutoML systems (rows the artifacts mark
+  `Reference Pipeline`, i.e. AutoGluon). It defaults to `models`, because "the best tabular model"
+  answered with an AutoML system is a wrong answer.
+- **Every UI event listener must pass `api_visibility="private"`.** Without it, each `.click` / `.change` /
+  `.select` lambda lands in the public schema as `/lambda_7` with a parameter named `value_11`. Before this
+  rule the Space advertised 23 such endpoints and nothing callable.
+- The same three functions are served as MCP tools at `/gradio_api/mcp/` (`launch(mcp_server=True)`,
+  `gradio[mcp]` extra). Gradio derives each tool's schema from the type hints and its description from the
+  docstring, so those are load-bearing for agents, not just for readers.
+- Record keys are the CSV headers with the sort-direction marker stripped and snake_cased
+  (`Improvability (%) [⬇️]` becomes `improvability_pct`); `Model` is split into `model`, `variant` and
+  `model_url`; `Verified` becomes a bool. `_api_key` in `api.py` is the single place that mapping lives.
+- Bulk readers should be pointed at the raw CSV instead, which `list_leaderboards` returns as a URL
+  template. It serves identical numbers with no queue and no token.
+
+The generated `agents.md` tells agents to POST to `/gradio_api/call/v2/<endpoint>` with named parameters.
+That route only exists in Gradio 6, which is why the Space is pinned there; on 5.x it returned 405 and
+agents had to fall back to `/gradio_api/call/<endpoint>` with a positional `{"data": [...]}` body. Both
+work now.
 
 ## Conventions
 
diff --git a/README.md b/README.md
index d519794195cbe99fde7af59ae051fbf9dd43035f..7192016c06aa11decfeca9bee4e7985cacfaf1f8 100644
--- a/README.md
+++ b/README.md
@@ -7,8 +7,8 @@ sdk: gradio
 app_file: main.py
 pinned: true
 license: apache-2.0
-short_description: 'TabArena'
-sdk_version: 5.49.1
+short_description: 'Elo-ranked leaderboards for tabular ML models, IID and beyond'
+sdk_version: 6.22.0
 ---
 
 # TabArena Leaderboard Code
@@ -29,6 +29,25 @@ pip install -e ".[dev]"
 uv pip install -r pyproject.toml
 ```
 
+# Reading the leaderboard programmatically
+
+The published numbers are plain CSVs in this repo, so the cheapest way to read
+them is a direct fetch (no queue, no token, `-L` because the raw path redirects
+to the resolve cache):
+
+```bash
+curl -sL https://huggingface.co/spaces/TabArena/leaderboard/resolve/main/data/imputation_yes/splits_all/tasks_all/datasets_all/website_leaderboard.csv
+```
+
+The path is `data/imputation_{yes,no}/splits_{all,lite}/tasks_{...}/datasets_{...}/`
+for TabArena and `data_beyondarena/subsets/{subset}/` for BeyondArena.
+
+The Space also serves three JSON endpoints for agents, defined in `api.py`:
+`list_leaderboards`, `get_tabarena_leaderboard`, `get_beyondarena_leaderboard`.
+Hugging Face advertises them through the **Agents** button on the Space page,
+which points at a generated `agents.md`. See AGENTS.md for the contract and its
+one caveat.
+
 # Current Steps to get results:
 1. Run https://github.com/autogluon/tabarena/blob/main/scripts/run_generate_website_artifacts.py
 2. Delete the current `data` folder contents in this repo to remove old results.
diff --git a/api.py b/api.py
new file mode 100644
index 0000000000000000000000000000000000000000..8d841d8febc2251049fe5ff0eaea1b9ac18e09c4
--- /dev/null
+++ b/api.py
@@ -0,0 +1,592 @@
+"""The Space's machine-readable API, for agents rather than browsers.
+
+Hugging Face serves a generated ``agents.md`` for every Gradio Space
+(``huggingface.co/spaces/<id>/agents.md``). It is not a repo file and cannot be
+overridden: it tells an agent to read ``/gradio_api/info`` and call whatever it
+finds there. So the endpoints registered here *are* the agent-facing contract.
+
+Two rules keep that contract readable. Everything in this file is a named,
+described, typed endpoint; and every UI event listener elsewhere in the app
+passes ``api_visibility="private"``, so render callbacks stay out of the schema
+(a public listener would otherwise show up as ``/lambda_7`` with a parameter
+called ``value_11``).
+
+The same three functions are also served as MCP tools at ``/gradio_api/mcp/``,
+from these type hints and docstrings, because ``main.py`` launches with
+``mcp_server=True``. Anything written here is read by an agent twice over.
+
+Records are the published CSVs with agent-friendly keys, not a second source of
+truth: the same files the site reads, the same model-cell parser the table uses.
+For bulk access, skip the queue and fetch the CSV directly (see
+:func:`list_leaderboards`, which hands out the URL template).
+"""
+
+from __future__ import annotations
+
+import os
+import re
+from pathlib import Path
+from typing import Any, Literal, get_args
+
+import gradio as gr
+import pandas as pd
+
+from constants import Constants
+from data_loading import (
+    BEYOND_DATA_DIR,
+    BEYOND_SUBSET_LABELS,
+    DATA_DIR,
+    DATASET_LABELS,
+    BeyondSubset,
+    LBContainer,
+    Subset,
+    TASK_LABELS,
+    parse_model,
+)
+
+# Axis values as Literals, so /gradio_api/info carries an enum per parameter.
+# That schema is the only place an agent reliably learns the valid values: a
+# rejected call comes back over the REST route as a bare `event: error` with a
+# null payload, message dropped. _validate_axes below keeps these in step with
+# the label dicts that define the data layout.
+# A leaderboard row is either a single model (evaluated in its default / tuned /
+# tuned+ensembled variants) or a whole AutoML system such as AutoGluon, which the
+# artifacts mark as a reference pipeline. Agents conflate the two otherwise, and
+# "the best tabular model" answered with an AutoML system is a wrong answer, so the
+# endpoints return models only unless asked for something else.
+KindAxis = Literal["models", "systems", "all"]
+SYSTEM_TYPE = Constants.reference
+
+# Which benchmark a cross-benchmark endpoint should read.
+BenchmarkAxis = Literal["tabarena", "beyondarena"]
+# What "best" means. The leaderboards rank by Elo; the rest are complementary.
+QualityAxis = Literal["elo", "score", "rank", "harmonic_rank", "improvability_pct"]
+_QUALITY_HIGHER_IS_BETTER = {
+    "elo": True,
+    "score": True,
+    "rank": False,
+    "harmonic_rank": False,
+    "improvability_pct": False,
+}
+# What "cheap" means, and the record key each maps to.
+CostAxis = Literal["predict_time", "train_time"]
+_COST_KEY = {
+    "predict_time": "median_predict_time_s_per_1k",
+    "train_time": "median_train_time_s_per_1k",
+}
+_COST_UNITS = {
+    "median_predict_time_s_per_1k": "seconds per 1000 rows to predict",
+    "median_train_time_s_per_1k": "seconds per 1000 rows to train",
+}
+
+TasksAxis = Literal["all", "classification", "regression", "binary", "multiclass"]
+DatasetsAxis = Literal["all", "small", "medium"]
+ImputationAxis = Literal["yes", "no"]
+SplitsAxis = Literal["all", "lite"]
+BeyondSubsetAxis = Literal[
+    "full",
+    "random",
+    "temporal",
+    "grouped",
+    "tiny",
+    "small",
+    "medium",
+    "large",
+    "low-dim",
+    "high-dim",
+    "text",
+    "high-cardinality",
+]
+
+IMPUTATION_VALUES = list(get_args(ImputationAxis))
+SPLITS_VALUES = list(get_args(SplitsAxis))
+KIND_VALUES = list(get_args(KindAxis))
+
+
+def _validate_axes() -> None:
+    """Fail at import if an axis Literal has drifted from the data layout."""
+    for name, literal, allowed in (
+        ("TasksAxis", TasksAxis, TASK_LABELS),
+        ("DatasetsAxis", DatasetsAxis, DATASET_LABELS),
+        ("BeyondSubsetAxis", BeyondSubsetAxis, BEYOND_SUBSET_LABELS),
+    ):
+        if set(get_args(literal)) != set(allowed):
+            raise RuntimeError(
+                f"api.{name} is out of sync with data_loading: "
+                f"{sorted(set(get_args(literal)) ^ set(allowed))} on one side only."
+            )
+
+
+_validate_axes()
+
+# The Space serving this app; HF sets SPACE_ID in the container.
+SPACE_ID = os.environ.get("SPACE_ID", "TabArena/leaderboard")
+_RAW_URL = f"https://huggingface.co/spaces/{SPACE_ID}/resolve/main"
+
+# Column headers carry a sort-direction marker for the table widget; it is not
+# part of the data, so it is stripped before the header becomes a JSON key.
+_DIRECTION_RE = re.compile(r"\s*\[[⬆⬇]️?\]$")
+_KEY_OVERRIDES = {"#": "position", "TypeName": "type_name"}
+
+# Bulk-download URLs, templated on the same rel_path the app reads from so the
+# two cannot drift.
+_TABARENA_CSV_TEMPLATE = (
+    f"{_RAW_URL}/{DATA_DIR.name}/"
+    + Subset("{imputation}", "{splits}", "{tasks}", "{datasets}").rel_path
+    + "/website_leaderboard.csv"
+)
+_BEYOND_CSV_TEMPLATE = (
+    f"{_RAW_URL}/{BEYOND_DATA_DIR.name}/"
+    + BeyondSubset("{subset}").rel_path
+    + "/website_leaderboard.csv"
+)
+
+
+def _api_key(column: str) -> str:
+    """Turn a CSV header into a JSON key: ``Improvability (%) [⬇️]`` -> ``improvability_pct``."""
+    if column in _KEY_OVERRIDES:
+        return _KEY_OVERRIDES[column]
+    name = _DIRECTION_RE.sub("", column).replace("%", "pct").replace("/", "_per_")
+    return re.sub(r"[^0-9a-zA-Z]+", "_", name).strip("_").lower()
+
+
+def _records(df: pd.DataFrame) -> list[dict[str, Any]]:
+    """One JSON record per leaderboard row."""
+    # NaN is not JSON; None is. Casting to object first keeps ints from turning
+    # into floats on the way through.
+    df = df.astype(object).where(pd.notna(df), None)
+    records = []
+    for row in df.to_dict("records"):
+        record = {_api_key(key): value for key, value in row.items()}
+        # The Model cell is markdown (`[TabFM (default)](url)`); agents want the
+        # three parts separately.
+        name, variant, url = parse_model(str(row["Model"]))
+        record["model"] = name
+        record["variant"] = variant
+        record["model_url"] = url
+        record["verified"] = str(row.get("Verified") or "").strip() == "✔️"
+        record["kind"] = "system" if row.get("TypeName") == SYSTEM_TYPE else "model"
+        records.append(record)
+    return records
+
+
+def _select_kind(records: list[dict[str, Any]], kind: str) -> list[dict[str, Any]]:
+    if kind == "all":
+        return records
+    wanted = "system" if kind == "systems" else "model"
+    return [r for r in records if r["kind"] == wanted]
+
+
+def _check(axis: str, value: str, allowed: list[str]) -> None:
+    if value not in allowed:
+        raise gr.Error(f"Unknown {axis}={value!r}. Valid values: {', '.join(allowed)}.")
+
+
+def _pareto_frontier(
+    records: list[dict[str, Any]], quality: str, cost_key: str
+) -> list[dict[str, Any]]:
+    """The non-dominated records: nothing is both better and cheaper.
+
+    Walks cheapest-first and keeps a record only when it beats every cheaper one on
+    quality, which is the 2-D skyline. Ties on cost keep the better-quality record.
+    """
+    higher_is_better = _QUALITY_HIGHER_IS_BETTER[quality]
+    usable = [
+        r
+        for r in records
+        if isinstance(r.get(quality), (int, float)) and isinstance(r.get(cost_key), (int, float))
+    ]
+    ordered = sorted(
+        usable,
+        key=lambda r: (r[cost_key], -r[quality] if higher_is_better else r[quality]),
+    )
+    frontier: list[dict[str, Any]] = []
+    best = None
+    for record in ordered:
+        value = record[quality]
+        if best is None or (value > best if higher_is_better else value < best):
+            frontier.append(record)
+            best = value
+    return frontier
+
+
+def _best(records: list[dict[str, Any]], quality: str) -> dict[str, Any] | None:
+    """The single best record on `quality`, ignoring rows that lack it."""
+    usable = [r for r in records if isinstance(r.get(quality), (int, float))]
+    if not usable:
+        return None
+    return max(
+        usable,
+        key=lambda r: r[quality] if _QUALITY_HIGHER_IS_BETTER[quality] else -r[quality],
+    )
+
+
+def _brief(record: dict[str, Any] | None, quality: str, cost_key: str) -> dict[str, Any] | None:
+    """The fields worth returning for one model in a Pareto answer."""
+    if record is None:
+        return None
+    return {
+        "model": record.get("model"),
+        "variant": record.get("variant"),
+        "kind": record.get("kind"),
+        "type_name": record.get("type_name"),
+        "verified": record.get("verified"),
+        "model_url": record.get("model_url"),
+        quality: record.get(quality),
+        cost_key: record.get(cost_key),
+        "median_train_time_s_per_1k": record.get("median_train_time_s_per_1k"),
+        "median_predict_time_s_per_1k": record.get("median_predict_time_s_per_1k"),
+    }
+
+
+def _describe(name: str | None, variant: str | None) -> str:
+    return f"{name} ({variant})" if variant else str(name)
+
+
+def _load(data_root: Path, subset: Subset | BeyondSubset) -> list[dict[str, Any]]:
+    lb = LBContainer(data_root=data_root, subset=subset, name="")
+    try:
+        return _records(lb.load_df())
+    except FileNotFoundError:
+        raise gr.Error(
+            f"No results for {subset.rel_path!r} in this deployment. "
+            "Call list_leaderboards for the subsets that are available."
+        ) from None
+
+
+def get_tabarena_leaderboard(
+    tasks: TasksAxis = "all",
+    datasets: DatasetsAxis = "all",
+    kind: KindAxis = "models",
+    imputation: ImputationAxis = "yes",
+    splits: SplitsAxis = "all",
+) -> list[dict[str, Any]]:
+    """Ranked TabArena results for one subset of the IID benchmark.
+
+    Args:
+        tasks: Task subset, one of ``all``, ``classification``, ``regression``,
+            ``binary``, ``multiclass``.
+        datasets: Dataset-size subset, one of ``all``, ``small``, ``medium``.
+        kind: What to rank. ``models`` (the default) returns individual models
+            such as TabPFN, RealMLP or LightGBM. ``systems`` returns whole AutoML
+            systems such as AutoGluon, which tune and ensemble many models
+            internally and are therefore not comparable to a single model.
+            ``all`` returns both, as the website's table does. Ask for ``models``
+            when the question is "which model should I use" and ``systems`` when
+            it is "which AutoML framework should I use"; every record also
+            carries its own ``kind``.
+        imputation: ``yes`` includes models that cannot run on every dataset,
+            imputing the missing runs with a default RandomForest (this is what
+            the website shows by default); ``no`` drops them.
+        splits: ``all`` for the full repeated-CV protocol, ``lite`` for the
+            single-split TabArena-Lite results.
+
+    Returns:
+        One record per model variant, best first, with ``model``, ``variant``,
+        ``elo``, ``score``, ``rank``, ``median_train_time_s_per_1k`` and the
+        remaining leaderboard columns. ``variant`` is ``default``, ``tuned`` or
+        ``tuned + ensembled``, and empty for reference pipelines such as
+        AutoGluon, which are whole pipelines rather than one model's variant
+        (``type_name`` identifies those). Note the website's variant filter
+        groups them with the tuned ensembles instead.
+    """
+    _check("tasks", tasks, list(TASK_LABELS))
+    _check("datasets", datasets, list(DATASET_LABELS))
+    _check("kind", kind, KIND_VALUES)
+    _check("imputation", imputation, IMPUTATION_VALUES)
+    _check("splits", splits, SPLITS_VALUES)
+    subset = Subset(imputation=imputation, splits=splits, tasks=tasks, datasets=datasets)
+    return _select_kind(_load(DATA_DIR, subset), kind)
+
+
+def get_beyondarena_leaderboard(
+    subset: BeyondSubsetAxis = "full", kind: KindAxis = "models"
+) -> list[dict[str, Any]]:
+    """Ranked BeyondArena results for one subset, on the recommended core protocol.
+
+    BeyondArena goes beyond the IID assumption: it spans random, temporal and
+    grouped splits. It has a single subset axis rather than TabArena's grid.
+
+    Args:
+        subset: One of ``full``, ``random``, ``temporal``, ``grouped``, ``tiny``,
+            ``small``, ``medium``, ``large``, ``low-dim``, ``high-dim``, ``text``,
+            ``high-cardinality``.
+        kind: ``models`` (the default), ``systems`` for whole AutoML systems, or
+            ``all``. See :func:`get_tabarena_leaderboard` for the distinction.
+
+    Returns:
+        One record per model variant, best first, in the same shape as
+        :func:`get_tabarena_leaderboard`.
+    """
+    _check("subset", subset, list(BEYOND_SUBSET_LABELS))
+    _check("kind", kind, KIND_VALUES)
+    return _select_kind(_load(BEYOND_DATA_DIR, BeyondSubset(subset=subset)), kind)
+
+
+def get_pareto_frontier(
+    benchmark: BenchmarkAxis = "tabarena",
+    quality: QualityAxis = "elo",
+    cost: CostAxis = "predict_time",
+    kind: KindAxis = "models",
+    tasks: TasksAxis = "all",
+    datasets: DatasetsAxis = "all",
+    beyond_subset: BeyondSubsetAxis = "full",
+    max_train_time_s_per_1k: float = 0.0,
+    max_predict_time_s_per_1k: float = 0.0,
+    imputation: ImputationAxis = "yes",
+    splits: SplitsAxis = "all",
+) -> dict[str, Any]:
+    """The accuracy-versus-time trade-off, for answering "which model should I use".
+
+    The top of a leaderboard is only one answer, and often the wrong one: the
+    highest-Elo model can be orders of magnitude slower than one a hair behind it.
+    This returns the whole trade-off instead — the outright best, the models that
+    nothing beats on both axes at once, and the best model that fits a time budget.
+
+    Args:
+        benchmark: ``tabarena`` (IID) or ``beyondarena`` (temporal / grouped splits).
+        quality: What "better" means: ``elo`` (default) or ``score``, where higher
+            wins, or ``rank`` / ``harmonic_rank`` / ``improvability_pct``, where
+            lower wins.
+        cost: What "cheaper" means: ``predict_time`` (default, what matters for
+            serving) or ``train_time`` (what matters for retraining).
+        kind: ``models`` (default), ``systems``, or ``all``. See
+            :func:`get_tabarena_leaderboard`.
+        tasks: TabArena task subset; ignored when `benchmark` is ``beyondarena``.
+        datasets: TabArena dataset-size subset; ignored for ``beyondarena``.
+        beyond_subset: BeyondArena subset; ignored when `benchmark` is ``tabarena``.
+        max_train_time_s_per_1k: Optional ceiling on median train seconds per 1000
+            rows. 0 means no limit.
+        max_predict_time_s_per_1k: Optional ceiling on median predict seconds per
+            1000 rows. 0 means no limit. Use this for "what is the best model I can
+            afford to serve".
+        imputation: TabArena only; see :func:`get_tabarena_leaderboard`.
+        splits: TabArena only; see :func:`get_tabarena_leaderboard`.
+
+    Returns:
+        ``best_overall`` (the outright leader, ignoring cost), ``frontier`` (the
+        non-dominated models, cheapest first, each with ``speedup_vs_best`` and
+        ``quality_gap_vs_best``), ``best_within_limits`` (the leader among models
+        satisfying the time ceilings, absent when no ceiling was given),
+        ``dominated_count``, and a ``summary`` sentence to relay.
+    """
+    _check("benchmark", benchmark, list(get_args(BenchmarkAxis)))
+    _check("quality", quality, list(_QUALITY_HIGHER_IS_BETTER))
+    _check("cost", cost, list(_COST_KEY))
+    _check("kind", kind, KIND_VALUES)
+    if benchmark == "beyondarena":
+        records = get_beyondarena_leaderboard(subset=beyond_subset, kind=kind)
+        where = f"BeyondArena ({beyond_subset}, core protocol)"
+    else:
+        records = get_tabarena_leaderboard(
+            tasks=tasks, datasets=datasets, kind=kind, imputation=imputation, splits=splits
+        )
+        where = f"TabArena ({tasks} tasks, {datasets} datasets)"
+
+    cost_key = _COST_KEY[cost]
+    best_overall = _best(records, quality)
+    frontier = _pareto_frontier(records, quality, cost_key)
+
+    limits = {
+        "median_train_time_s_per_1k": max_train_time_s_per_1k,
+        "median_predict_time_s_per_1k": max_predict_time_s_per_1k,
+    }
+    active_limits = {key: value for key, value in limits.items() if value and value > 0}
+    within = records
+    for key, ceiling in active_limits.items():
+        within = [
+            r for r in within if isinstance(r.get(key), (int, float)) and r[key] <= ceiling
+        ]
+    best_within = _best(within, quality) if active_limits else None
+
+    entries = []
+    for record in frontier:
+        entry = _brief(record, quality, cost_key)
+        if best_overall and isinstance(best_overall.get(cost_key), (int, float)):
+            reference = best_overall[cost_key]
+            entry["speedup_vs_best"] = (
+                round(reference / record[cost_key], 2) if record[cost_key] else None
+            )
+            entry["quality_gap_vs_best"] = round(
+                abs(best_overall[quality] - record[quality]), 4
+            )
+        entries.append(entry)
+
+    return {
+        "benchmark": benchmark,
+        "subset": where,
+        "quality": quality,
+        "higher_is_better": _QUALITY_HIGHER_IS_BETTER[quality],
+        "cost": cost_key,
+        "kind": kind,
+        "best_overall": _brief(best_overall, quality, cost_key),
+        "frontier": entries,
+        "dominated_count": max(0, len(records) - len(frontier)),
+        "limits": active_limits or None,
+        "best_within_limits": _brief(best_within, quality, cost_key),
+        "summary": _pareto_summary(
+            where, quality, cost_key, best_overall, entries, active_limits, best_within
+        ),
+    }
+
+
+def _pareto_summary(
+    where: str,
+    quality: str,
+    cost_key: str,
+    best_overall: dict[str, Any] | None,
+    frontier: list[dict[str, Any]],
+    limits: dict[str, float],
+    best_within: dict[str, Any] | None,
+) -> str:
+    """One paragraph an agent can relay instead of reciting the whole frontier."""
+    if best_overall is None:
+        return f"No {quality} values are available for {where}."
+    lead = _describe(best_overall.get("model"), best_overall.get("variant"))
+    units = _COST_UNITS.get(cost_key, cost_key)
+    parts = [
+        f"On {where}, the best {quality} is {lead} at {best_overall[quality]} "
+        f"({best_overall.get(cost_key)} {units})."
+    ]
+    if len(frontier) > 1:
+        cheapest = frontier[0]
+        parts.append(
+            f"{len(frontier)} models are on the accuracy/time frontier. The cheapest, "
+            f"{_describe(cheapest.get('model'), cheapest.get('variant'))}, is "
+            f"{cheapest.get('speedup_vs_best')}x faster but {cheapest.get('quality_gap_vs_best')} "
+            f"{quality} behind."
+        )
+        # The trade-off worth naming is the smallest sacrifice that still buys a real
+        # speedup, not the biggest speedup (that is just the cheapest model again).
+        bargains = [
+            entry
+            for entry in frontier
+            if (entry.get("speedup_vs_best") or 0) >= 2 and entry.get("quality_gap_vs_best")
+        ]
+        if bargains:
+            pick = min(bargains, key=lambda e: e["quality_gap_vs_best"])
+            parts.append(
+                # "at a cost of" rather than "less", which reads backwards for the
+                # metrics where lower is better.
+                f"The best compromise is {_describe(pick.get('model'), pick.get('variant'))}: "
+                f"{pick.get('speedup_vs_best')}x faster at a cost of "
+                f"{pick.get('quality_gap_vs_best')} {quality}."
+            )
+    if limits:
+        stated = ", ".join(f"{key} <= {value}" for key, value in limits.items())
+        if best_within:
+            parts.append(
+                f"Within {stated}, the best is "
+                f"{_describe(best_within.get('model'), best_within.get('variant'))} at "
+                f"{best_within[quality]}."
+            )
+        else:
+            parts.append(f"No model satisfies {stated}.")
+    return " ".join(parts)
+
+
+def list_leaderboards() -> dict[str, Any]:
+    """Describe the available leaderboards, their subset axes, and the bulk-download URLs.
+
+    Call this first: it lists the valid argument values for the other endpoints
+    and the record keys they return. It also gives the raw CSV URL template for
+    each benchmark, which serves the identical numbers over plain HTTP with no
+    queue or session, and is the better choice for reading many subsets.
+    """
+    default_rows = _load(DATA_DIR, Subset())
+    return {
+        "leaderboards": [
+            {
+                "name": "tabarena",
+                "endpoint": "/get_tabarena_leaderboard",
+                "description": "Tabular ML on 51 curated IID datasets, ranked by Elo.",
+                "axes": {
+                    "tasks": list(TASK_LABELS),
+                    "datasets": list(DATASET_LABELS),
+                    "kind": KIND_VALUES,
+                    "imputation": IMPUTATION_VALUES,
+                    "splits": SPLITS_VALUES,
+                },
+                "csv_url_template": _TABARENA_CSV_TEMPLATE,
+            },
+            {
+                "name": "beyondarena",
+                "endpoint": "/get_beyondarena_leaderboard",
+                "description": (
+                    "Tabular ML beyond the IID assumption: random, temporal and grouped "
+                    "splits across dataset sizes and feature types."
+                ),
+                "axes": {"subset": list(BEYOND_SUBSET_LABELS), "kind": KIND_VALUES},
+                "csv_url_template": _BEYOND_CSV_TEMPLATE,
+            },
+        ],
+        "record_keys": sorted(default_rows[0]) if default_rows else [],
+        "kinds": {
+            "models": "Individual models (TabPFN, RealMLP, LightGBM, ...). The default.",
+            "systems": (
+                f"Whole AutoML systems (AutoGluon, ...), marked '{SYSTEM_TYPE}' in the "
+                "artifacts. They tune and ensemble many models internally, so ranking them "
+                "against a single model compares different things."
+            ),
+            "all": "Both, as the website's table shows them.",
+        },
+        "choosing_a_model": (
+            "For 'which model should I use', call get_pareto_frontier rather than reading the "
+            "top row here: it returns the accuracy/time trade-off, since the highest-Elo model "
+            "is often orders of magnitude slower than one just behind it. It also takes "
+            "train- and predict-time budgets."
+        ),
+        "notes": (
+            "Scores are read from the published artifacts, so they match the website exactly. "
+            "Higher is better for elo and score; lower is better for rank, harmonic_rank, "
+            "improvability_pct and the time columns. Every record carries a `kind` of "
+            "'model' or 'system'; the get_* endpoints return models only unless asked."
+        ),
+    }
+
+
+def register_api() -> None:
+    """Register the endpoints. Call inside the app's ``gr.Blocks`` context."""
+    gr.api(
+        list_leaderboards,
+        api_name="list_leaderboards",
+        api_description=(
+            "List the available TabArena leaderboards, the valid subset values for each, "
+            "the difference between models and systems, the keys every record carries, and "
+            "the raw CSV URLs for bulk download. Call this before the get_* endpoints."
+        ),
+    )
+    gr.api(
+        get_tabarena_leaderboard,
+        api_name="get_tabarena_leaderboard",
+        api_description=(
+            "Ranked TabArena results (Elo, score, ranks, train/predict time) for one subset "
+            "of the IID benchmark, selected by tasks, datasets, imputation and splits. "
+            "Pass kind='models' (the default) for individual models such as TabPFN or "
+            "LightGBM, kind='systems' for whole AutoML systems such as AutoGluon, or "
+            "kind='all' for both. A question about the best model wants 'models'; a "
+            "question about the best AutoML framework wants 'systems'."
+        ),
+    )
+    gr.api(
+        get_beyondarena_leaderboard,
+        api_name="get_beyondarena_leaderboard",
+        api_description=(
+            "Ranked BeyondArena results for one subset (random, temporal, grouped splits; "
+            "size and feature-type buckets), on the recommended core protocol. Takes the "
+            "same kind='models' | 'systems' | 'all' distinction as the TabArena endpoint."
+        ),
+    )
+    gr.api(
+        get_pareto_frontier,
+        api_name="get_pareto_frontier",
+        api_description=(
+            "The accuracy-versus-time trade-off for one subset: the outright best model, "
+            "the models nothing beats on both accuracy and speed at once (each with its "
+            "speedup and accuracy gap versus the leader), and the best model that fits a "
+            "train- or predict-time budget. Prefer this over get_*_leaderboard for "
+            "'which model should I use', 'what is the best model', or any question about "
+            "models that are faster to fit or to serve: the top of the leaderboard is "
+            "often far slower than a model just behind it. Returns a summary sentence."
+        ),
+    )
diff --git a/constants.py b/constants.py
index ab3a7ade87aea03b050d85c6b15d7cbd6d0354ed..64133bfcd35f3fc5fca3f801f21fb30e67b194f0 100644
--- a/constants.py
+++ b/constants.py
@@ -19,6 +19,15 @@ model_type_emoji = {
     Constants.reference:"📊",
 }
 
+# Per-variant accent color, taken from the interactive Leaderboard Overview
+# explorer's --var-* (dark theme) so its variant toggles and the leaderboard
+# table's agree. Keys must match views.VARIANT_VALUES.
+variant_color = {
+    "default": "#4386d5",
+    "tuned": "#c05f38",
+    "tuned + ensembled": "#289972",
+}
+
 # Per-type accent color (readable as text/border on the dark theme). Used for the
 # type legend and to color-code model names in the cross-subset overview.
 model_type_color = {
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_all/leaderboard_overview_explorer.html
index 783d81be83287d2adb3a5a66951a1a2602a85f39..8082ae98441a143bb05149e783bd598587d1c54a 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | All Tasks | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_all/leaderboard_table.html b/data/imputation_no/splits_all/tasks_all/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..e6b878991a11032cb3fb4d282c2c78e67f74db9e
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_all/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1792.0,"score":0.889,"rank":6.17,"hrank":1.72,"imp":4.074,"train_time":38.81,"predict_time":6.985,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-110","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1691.0,"score":0.662,"rank":9.39,"hrank":4.03,"imp":7.979,"train_time":289.07,"predict_time":4.031,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-68","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1665.0,"score":0.67,"rank":10.41,"hrank":4.06,"imp":9.328,"train_time":3.66,"predict_time":0.399,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-61","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1620.0,"score":0.583,"rank":12.42,"hrank":7.14,"imp":10.751,"train_time":5.48,"predict_time":0.555,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-51","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1595.0,"score":0.549,"rank":13.64,"hrank":6.22,"imp":10.427,"train_time":2040.22,"predict_time":8.908,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-56","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1594.0,"score":0.606,"rank":13.71,"hrank":4.97,"imp":10.175,"train_time":2.05,"predict_time":0.151,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-61","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1554.0,"score":0.504,"rank":15.81,"hrank":7.33,"imp":11.116,"train_time":2040.22,"predict_time":1.218,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-55","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1524.0,"score":0.474,"rank":17.55,"hrank":9.62,"imp":11.66,"train_time":5.81,"predict_time":0.645,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-44","imputed":false},{"position":8,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1514.0,"score":0.401,"rank":18.2,"hrank":8.44,"imp":12.789,"train_time":1735.72,"predict_time":2.558,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-46","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1508.0,"score":0.405,"rank":18.55,"hrank":11.52,"imp":12.881,"train_time":2950.72,"predict_time":11.975,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-46","imputed":false},{"position":10,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1459.0,"score":0.38,"rank":21.7,"hrank":9.59,"imp":14.0,"train_time":2.07,"predict_time":0.182,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-51","imputed":false},{"position":11,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1458.0,"score":0.375,"rank":21.81,"hrank":8.16,"imp":13.788,"train_time":4910.38,"predict_time":286.537,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-51","imputed":false},{"position":12,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1450.0,"score":0.315,"rank":22.33,"hrank":12.48,"imp":14.081,"train_time":2450.13,"predict_time":2.247,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-43","imputed":false},{"position":13,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1433.0,"score":0.262,"rank":23.53,"hrank":17.2,"imp":15.033,"train_time":417.05,"predict_time":2.639,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+33\/-31","imputed":false},{"position":14,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1426.0,"score":0.295,"rank":23.99,"hrank":13.67,"imp":14.202,"train_time":2950.72,"predict_time":0.663,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-45","imputed":false},{"position":15,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1422.0,"score":0.261,"rank":24.34,"hrank":14.77,"imp":14.656,"train_time":1346.21,"predict_time":0.344,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-38","imputed":false},{"position":16,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1408.0,"score":0.245,"rank":25.3,"hrank":15.31,"imp":14.885,"train_time":1346.21,"predict_time":0.038,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-43","imputed":false},{"position":17,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1404.0,"score":0.235,"rank":25.56,"hrank":17.78,"imp":15.304,"train_time":12685.08,"predict_time":464.37,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+40\/-42","imputed":false},{"position":18,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1403.0,"score":0.316,"rank":25.69,"hrank":11.3,"imp":14.936,"train_time":4910.38,"predict_time":39.96,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-56","imputed":false},{"position":19,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1394.0,"score":0.251,"rank":26.35,"hrank":14.47,"imp":14.952,"train_time":2450.13,"predict_time":0.251,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-42","imputed":false},{"position":20,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1389.0,"score":0.277,"rank":26.66,"hrank":12.79,"imp":15.599,"train_time":4621.67,"predict_time":8.138,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-52","imputed":false},{"position":21,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1386.0,"score":0.194,"rank":26.93,"hrank":20.71,"imp":15.688,"train_time":417.05,"predict_time":0.334,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+29\/-29","imputed":false},{"position":22,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1378.0,"score":0.224,"rank":27.47,"hrank":16.11,"imp":15.524,"train_time":5.88,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-42","imputed":false},{"position":23,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1375.0,"score":0.198,"rank":27.73,"hrank":19.98,"imp":15.754,"train_time":693.49,"predict_time":1.689,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+31\/-33","imputed":false},{"position":24,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1363.0,"score":0.299,"rank":28.66,"hrank":9.55,"imp":15.526,"train_time":26.46,"predict_time":6.242,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-59","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1355.0,"score":0.189,"rank":29.21,"hrank":16.82,"imp":16.044,"train_time":4621.67,"predict_time":0.475,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+37\/-37","imputed":false},{"position":26,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1350.0,"score":0.251,"rank":29.58,"hrank":17.68,"imp":16.081,"train_time":1.18,"predict_time":0.072,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-52","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1350.0,"score":0.16,"rank":29.6,"hrank":22.22,"imp":16.065,"train_time":693.49,"predict_time":0.308,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+29\/-32","imputed":false},{"position":28,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1348.0,"score":0.209,"rank":29.79,"hrank":18.96,"imp":16.284,"train_time":846.89,"predict_time":2.549,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-43","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1333.0,"score":0.186,"rank":30.87,"hrank":19.34,"imp":17.213,"train_time":427.39,"predict_time":0.404,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-60","imputed":false},{"position":30,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1325.0,"score":0.245,"rank":31.53,"hrank":14.33,"imp":17.225,"train_time":47.65,"predict_time":43.738,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-55","imputed":false},{"position":31,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1308.0,"score":0.156,"rank":32.79,"hrank":21.32,"imp":17.551,"train_time":427.39,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-60","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1300.0,"score":0.158,"rank":33.43,"hrank":22.28,"imp":17.449,"train_time":846.89,"predict_time":0.13,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+45\/-39","imputed":false},{"position":33,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1298.0,"score":0.161,"rank":33.63,"hrank":21.84,"imp":17.0,"train_time":7.49,"predict_time":0.244,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-44","imputed":false},{"position":34,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1295.0,"score":0.111,"rank":33.83,"hrank":23.47,"imp":17.147,"train_time":12685.08,"predict_time":62.13,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+40\/-33","imputed":false},{"position":35,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1285.0,"score":0.126,"rank":34.58,"hrank":25.37,"imp":17.161,"train_time":2875.52,"predict_time":1.952,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-47","imputed":false},{"position":36,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1283.0,"score":0.194,"rank":34.72,"hrank":14.11,"imp":18.278,"train_time":14.11,"predict_time":2.075,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-55","imputed":false},{"position":37,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1272.0,"score":0.116,"rank":35.64,"hrank":25.8,"imp":18.672,"train_time":1711.25,"predict_time":0.141,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-40","imputed":false},{"position":38,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1264.0,"score":0.121,"rank":36.23,"hrank":23.23,"imp":18.657,"train_time":1.99,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-55","imputed":false},{"position":39,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1248.0,"score":0.09,"rank":37.44,"hrank":23.68,"imp":19.194,"train_time":14.87,"predict_time":0.307,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+41\/-39","imputed":false},{"position":40,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1235.0,"score":0.082,"rank":38.43,"hrank":28.17,"imp":19.314,"train_time":1711.25,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-40","imputed":false},{"position":41,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1230.0,"score":0.066,"rank":38.85,"hrank":29.23,"imp":18.569,"train_time":10.06,"predict_time":1.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+36\/-37","imputed":false},{"position":42,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1214.0,"score":0.053,"rank":40.04,"hrank":30.76,"imp":18.792,"train_time":1.94,"predict_time":0.123,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-36","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1212.0,"score":0.069,"rank":40.24,"hrank":30.96,"imp":18.877,"train_time":2875.52,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-42","imputed":false},{"position":44,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1208.0,"score":0.083,"rank":40.47,"hrank":28.16,"imp":19.777,"train_time":263.04,"predict_time":0.766,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-43","imputed":false},{"position":45,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1207.0,"score":0.105,"rank":40.58,"hrank":25.38,"imp":19.677,"train_time":593.24,"predict_time":4.467,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-59","imputed":false},{"position":46,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1204.0,"score":0.069,"rank":40.82,"hrank":28.54,"imp":20.255,"train_time":6.67,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-49","imputed":false},{"position":47,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1186.0,"score":0.038,"rank":42.12,"hrank":36.15,"imp":19.344,"train_time":1.96,"predict_time":0.142,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-29","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1180.0,"score":0.074,"rank":42.56,"hrank":26.73,"imp":20.564,"train_time":263.04,"predict_time":0.071,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-52","imputed":false},{"position":49,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1171.0,"score":0.06,"rank":43.28,"hrank":29.04,"imp":20.828,"train_time":373.24,"predict_time":0.771,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-47","imputed":false},{"position":50,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1147.0,"score":0.048,"rank":45.01,"hrank":31.19,"imp":21.096,"train_time":593.24,"predict_time":0.306,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-59","imputed":false},{"position":51,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1137.0,"score":0.041,"rank":45.71,"hrank":28.56,"imp":21.535,"train_time":373.24,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-44","imputed":false},{"position":52,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1097.0,"score":0.125,"rank":48.47,"hrank":15.95,"imp":25.584,"train_time":28729.74,"predict_time":4.268,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-83","imputed":false},{"position":53,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1096.0,"score":0.123,"rank":48.56,"hrank":15.35,"imp":25.6,"train_time":28729.74,"predict_time":18.811,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-83","imputed":false},{"position":54,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1094.0,"score":0.037,"rank":48.66,"hrank":33.93,"imp":23.024,"train_time":296.64,"predict_time":68.173,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-54","imputed":false},{"position":55,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1089.0,"score":0.024,"rank":49.04,"hrank":38.83,"imp":26.446,"train_time":185.31,"predict_time":0.6,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-45","imputed":false},{"position":56,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1074.0,"score":0.011,"rank":49.99,"hrank":39.42,"imp":22.815,"train_time":9.99,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-42","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1050.0,"score":0.011,"rank":51.52,"hrank":39.96,"imp":27.802,"train_time":185.31,"predict_time":0.258,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-50","imputed":false},{"position":58,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1043.0,"score":0.029,"rank":51.95,"hrank":44.11,"imp":26.445,"train_time":3.23,"predict_time":0.919,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-70","imputed":false},{"position":59,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1009.0,"score":0.012,"rank":54.0,"hrank":47.38,"imp":26.176,"train_time":1.87,"predict_time":0.283,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-66","imputed":false},{"position":60,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1008.0,"score":0.009,"rank":54.07,"hrank":39.97,"imp":25.682,"train_time":2.86,"predict_time":0.373,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-62","imputed":false},{"position":61,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.003,"rank":54.51,"hrank":49.81,"imp":26.265,"train_time":0.43,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-47","imputed":false},{"position":62,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":997.0,"score":0.017,"rank":54.67,"hrank":38.92,"imp":27.822,"train_time":129.08,"predict_time":1.801,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-76","imputed":false},{"position":63,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":993.0,"score":0.077,"rank":54.89,"hrank":24.31,"imp":30.456,"train_time":384.75,"predict_time":5.348,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-104","imputed":false},{"position":64,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":965.0,"score":0.031,"rank":56.39,"hrank":40.87,"imp":33.667,"train_time":237.63,"predict_time":0.424,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-100","imputed":false},{"position":65,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":940.0,"score":0.022,"rank":57.67,"hrank":42.64,"imp":34.211,"train_time":237.63,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-105","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":936.0,"score":0.0,"rank":57.84,"hrank":45.44,"imp":31.748,"train_time":27.32,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-57","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":890.0,"score":0.008,"rank":59.92,"hrank":46.93,"imp":32.782,"train_time":129.08,"predict_time":0.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-100","imputed":false},{"position":68,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":861.0,"score":0.014,"rank":61.12,"hrank":47.51,"imp":36.822,"train_time":1.19,"predict_time":0.12,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-118","imputed":false},{"position":69,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":654.0,"score":0.0,"rank":66.89,"hrank":60.42,"imp":45.987,"train_time":0.19,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-91","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_all/pareto_front_explorer.html
index 724b708a6f94c870b8515c4037ce395da6e39fd6..50944cf6de8cdcd296eeddf3010e4066e27f7067 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"imp":15.5244408181,"elo":1378.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0380928654,"imp":14.8850630485,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3442282032,"imp":14.655679873,"elo":1421.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0325848325,"imp":18.6570877535,"elo":1263.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0336512776,"imp":17.5510939543,"elo":1308.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4040760969,"imp":17.2125514645,"elo":1333.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"imp":20.2553633192,"elo":1203.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0150308987,"imp":19.3136682806,"elo":1235.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1412900124,"imp":18.6722157117,"elo":1271.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2834081321,"imp":26.1764963285,"elo":1009.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0712129939,"imp":20.5638085151,"elo":1180.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7662402325,"imp":19.7773891991,"elo":1208.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3731781389,"imp":25.6818545229,"elo":1007.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3063915412,"imp":21.0963185218,"elo":1146.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4666950703,"imp":19.6768711642,"elo":1207.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1416787942,"imp":19.3442915921,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3337603531,"imp":15.6876116943,"elo":1385.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"imp":15.0334967342,"elo":1432.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.2423742585,"imp":15.5263025459,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3073466784,"imp":19.1942294206,"elo":1248.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4747242315,"imp":16.0439590519,"elo":1355.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.1379168675,"imp":15.5985458077,"elo":1389.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"imp":31.7480551938,"elo":936.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2581384468,"imp":27.801643493,"elo":1050.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5997539679,"imp":26.4456425257,"elo":1088.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0535405758,"imp":26.2650042304,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852279186,"imp":21.5350457807,"elo":1137.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7709478321,"imp":20.8284764332,"elo":1170.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.6887954948,"imp":18.5692266729,"elo":1229.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.663132055,"imp":14.2021817976,"elo":1426.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":11.9749853765,"imp":12.8814924929,"elo":1507.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"imp":11.6604512027,"elo":1524.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2178786994,"imp":11.1161483399,"elo":1554.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9076602204,"imp":10.4265657302,"elo":1594.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0745579697,"imp":18.2783205087,"elo":1283.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"imp":17.2252827702,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9597364603,"imp":14.9360532408,"elo":1402.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":286.5373788357,"imp":13.7879250832,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"imp":13.9998432493,"elo":1459.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.9854290718,"imp":4.0744823648,"elo":1791.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1505492146,"imp":10.1751538315,"elo":1593.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2435252328,"imp":17.0001433318,"elo":1297.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2512939462,"imp":14.951713351,"elo":1393.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.2468658586,"imp":14.0806210836,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5548034626,"imp":10.7505455244,"elo":1619.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"imp":9.3282966338,"elo":1665.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3480868421,"imp":30.4556099407,"elo":993.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2680534632,"imp":25.584307557,"elo":1097.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"imp":25.6000754281,"elo":1096.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719683769,"imp":16.0807432886,"elo":1350.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1258046788,"imp":22.8151308708,"elo":1074.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1311128328,"imp":18.876878914,"elo":1211.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.95162543,"imp":17.1611523587,"elo":1285.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1226232881,"imp":18.7922233221,"elo":1214.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3080760155,"imp":16.0650443256,"elo":1350.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.689445163,"imp":15.7538195428,"elo":1375.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":68.173296937,"imp":23.0236433289,"elo":1094.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":62.1299021172,"imp":17.1472473791,"elo":1295.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":464.3696034625,"imp":15.3035551324,"elo":1404.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"imp":26.4448633944,"elo":1043.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1302743502,"imp":17.4492208687,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.5492595349,"imp":16.2842179754,"elo":1347.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"x_train":5.876676442,"imp":15.5244408181,"elo":1378.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0380928654,"x_train":1346.2106911803,"imp":14.8850630485,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3442282032,"x_train":1346.2106911803,"imp":14.655679873,"elo":1421.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0325848325,"x_train":1.9886321895,"imp":18.6570877535,"elo":1263.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0336512776,"x_train":427.391750384,"imp":17.5510939543,"elo":1308.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4040760969,"x_train":427.391750384,"imp":17.2125514645,"elo":1333.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"x_train":6.6731468509,"imp":20.2553633192,"elo":1203.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0150308987,"x_train":1711.2458519871,"imp":19.3136682806,"elo":1235.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1412900124,"x_train":1711.2458519871,"imp":18.6722157117,"elo":1271.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2834081321,"x_train":1.8719729433,"imp":26.1764963285,"elo":1009.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0712129939,"x_train":263.0379305179,"imp":20.5638085151,"elo":1180.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7662402325,"x_train":263.0379305179,"imp":19.7773891991,"elo":1208.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3731781389,"x_train":2.8561491959,"imp":25.6818545229,"elo":1007.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3063915412,"x_train":593.2377884029,"imp":21.0963185218,"elo":1146.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4666950703,"x_train":593.2377884029,"imp":19.6768711642,"elo":1207.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1416787942,"x_train":1.960032688,"imp":19.3442915921,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3337603531,"x_train":417.0474735649,"imp":15.6876116943,"elo":1385.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"x_train":417.0474735649,"imp":15.0334967342,"elo":1432.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.2423742585,"x_train":26.4597512682,"imp":15.5263025459,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3073466784,"x_train":14.8696863446,"imp":19.1942294206,"elo":1248.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4747242315,"x_train":4621.6656335635,"imp":16.0439590519,"elo":1355.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.1379168675,"x_train":4621.6656335635,"imp":15.5985458077,"elo":1389.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"x_train":27.322832323,"imp":31.7480551938,"elo":936.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2581384468,"x_train":185.3067056521,"imp":27.801643493,"elo":1050.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5997539679,"x_train":185.3067056521,"imp":26.4456425257,"elo":1088.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0535405758,"x_train":0.4343250827,"imp":26.2650042304,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852279186,"x_train":373.2371055096,"imp":21.5350457807,"elo":1137.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7709478321,"x_train":373.2371055096,"imp":20.8284764332,"elo":1170.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.6887954948,"x_train":10.055649238,"imp":18.5692266729,"elo":1229.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.663132055,"x_train":2950.7244476948,"imp":14.2021817976,"elo":1426.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":11.9749853765,"x_train":2950.7244476948,"imp":12.8814924929,"elo":1507.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"x_train":5.8097985299,"imp":11.6604512027,"elo":1524.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2178786994,"x_train":2040.2228420973,"imp":11.1161483399,"elo":1554.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9076602204,"x_train":2040.2228420973,"imp":10.4265657302,"elo":1594.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0745579697,"x_train":14.1132668455,"imp":18.2783205087,"elo":1283.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"x_train":47.6466878891,"imp":17.2252827702,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9597364603,"x_train":4910.3773062373,"imp":14.9360532408,"elo":1402.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":286.5373788357,"x_train":4910.3773062373,"imp":13.7879250832,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"x_train":2.0676840515,"imp":13.9998432493,"elo":1459.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.9854290718,"x_train":38.8059088678,"imp":4.0744823648,"elo":1791.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1505492146,"x_train":2.0524528172,"imp":10.1751538315,"elo":1593.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2435252328,"x_train":7.4864639685,"imp":17.0001433318,"elo":1297.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2512939462,"x_train":2450.1270774379,"imp":14.951713351,"elo":1393.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.2468658586,"x_train":2450.1270774379,"imp":14.0806210836,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5548034626,"x_train":5.4825798778,"imp":10.7505455244,"elo":1619.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"x_train":3.65846503,"imp":9.3282966338,"elo":1665.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3480868421,"x_train":384.7456301814,"imp":30.4556099407,"elo":993.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2680534632,"x_train":28729.7373655693,"imp":25.584307557,"elo":1097.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"x_train":28729.7373655693,"imp":25.6000754281,"elo":1096.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719683769,"x_train":1.1753394519,"imp":16.0807432886,"elo":1350.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1258046788,"x_train":9.9925632138,"imp":22.8151308708,"elo":1074.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1311128328,"x_train":2875.5203407747,"imp":18.876878914,"elo":1211.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.95162543,"x_train":2875.5203407747,"imp":17.1611523587,"elo":1285.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1226232881,"x_train":1.9409389396,"imp":18.7922233221,"elo":1214.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3080760155,"x_train":693.4907982506,"imp":16.0650443256,"elo":1350.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.689445163,"x_train":693.4907982506,"imp":15.7538195428,"elo":1375.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":68.173296937,"x_train":296.6392917616,"imp":23.0236433289,"elo":1094.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":62.1299021172,"x_train":12685.0815501834,"imp":17.1472473791,"elo":1295.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":464.3696034625,"x_train":12685.0815501834,"imp":15.3035551324,"elo":1404.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"x_train":3.2328671519,"imp":26.4448633944,"elo":1043.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1302743502,"x_train":846.8860702337,"imp":17.4492208687,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.5492595349,"x_train":846.8860702337,"imp":16.2842179754,"elo":1347.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_all/tuning_trajectories_explorer.html
index e55eb92c49d4a5b925d9a9d05f74e91178f81925..51687b24fb6486868a48eff82610fdc8c17968c9 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":5.876676442,"imp":15.5457736304,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":11.6678251256,"imp":15.1182419551,"elo":1413.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":31.9106517624,"imp":14.848578649,"elo":1430.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":62.5721251757,"imp":14.7477568558,"elo":1435.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":155.9049355669,"imp":14.6406177695,"elo":1441.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":317.345916891,"imp":14.6311446381,"elo":1441.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":662.4830210734,"imp":14.6478465613,"elo":1440.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":998.8498259224,"imp":14.6673563694,"elo":1438.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1346.2106911803,"imp":14.674240095,"elo":1437.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.9886321895,"imp":18.6780631139,"elo":1273.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.4383850203,"imp":18.3366328085,"elo":1294.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.0701848574,"imp":18.116797306,"elo":1307.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":20.4734178164,"imp":17.9259739747,"elo":1318.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":52.7173757556,"imp":17.6537027938,"elo":1330.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":106.3941874905,"imp":17.5022920851,"elo":1338.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":212.8207478205,"imp":17.339314755,"elo":1346.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":318.9957179838,"imp":17.2700034244,"elo":1348.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":427.391750384,"imp":17.2326457261,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":6.6731468509,"imp":20.2759337918,"elo":1208.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":16.5765343457,"imp":19.8786837436,"elo":1222.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":47.7385589033,"imp":19.4534296615,"elo":1243.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":89.4110724699,"imp":19.2726704565,"elo":1252.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":218.448124708,"imp":19.0438192816,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":440.2545735965,"imp":18.9023411836,"elo":1271.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":856.7647171053,"imp":18.780391801,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1266.0910927829,"imp":18.7369843166,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1711.2458519871,"imp":18.6945455008,"elo":1284.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.8719729433,"imp":26.1948484129,"elo":1009.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1259596548,"imp":24.3488542313,"elo":1053.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.7721912257,"imp":21.6826802288,"elo":1124.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":13.288092342,"imp":20.760558803,"elo":1165.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":32.8419339595,"imp":20.1069857393,"elo":1197.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":65.4938634865,"imp":19.9445772606,"elo":1207.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":130.8369715186,"imp":19.8340399339,"elo":1214.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":196.2244804971,"imp":19.7910126465,"elo":1218.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":263.0379305179,"imp":19.7984513883,"elo":1219.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.8561491959,"imp":25.7013457941,"elo":1012.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.5343795004,"imp":24.5020141267,"elo":1049.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.5510344275,"imp":22.6534704395,"elo":1116.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":29.723343154,"imp":21.7464416276,"elo":1149.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":72.5444980812,"imp":20.8458197285,"elo":1180.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":145.754371191,"imp":20.3916263193,"elo":1196.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":293.9475008279,"imp":20.0488916705,"elo":1207.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":441.8525399102,"imp":19.826636643,"elo":1215.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":593.2377884029,"imp":19.6935269069,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":1.960032688,"imp":19.3647921745,"elo":1188.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":4.1465368891,"imp":17.8536668429,"elo":1267.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":10.3907433568,"imp":16.6855138703,"elo":1346.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":20.2701675373,"imp":16.0840498218,"elo":1386.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":51.487448334,"imp":15.5280964347,"elo":1421.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":103.6855742013,"imp":15.3158433015,"elo":1437.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":207.5435114784,"imp":15.1552157574,"elo":1447.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":310.2926532568,"imp":15.0933557354,"elo":1451.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":417.0474735649,"imp":15.0561804683,"elo":1453.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":26.4597512682,"imp":15.4843630349,"elo":1381.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.8696863446,"imp":19.2146348547,"elo":1258.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":40.5171687704,"imp":17.9235444693,"elo":1296.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":111.4498310885,"imp":16.0052009744,"elo":1381.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":224.929440425,"imp":15.5140651615,"elo":1406.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":554.8880523029,"imp":15.0054551931,"elo":1431.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1145.4516650488,"imp":15.0529606984,"elo":1428.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2295.4928354707,"imp":15.3121547879,"elo":1419.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3433.7924746694,"imp":15.5114179492,"elo":1411.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4621.6656335635,"imp":15.6253232633,"elo":1406.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":27.322832323,"imp":32.2510059823,"elo":936.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":60.3304655332,"imp":30.0030285719,"elo":1001.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":157.3958954151,"imp":26.6320054165,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":185.3067056521,"imp":26.4602162594,"elo":1095.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4343250827,"imp":26.2932402488,"elo":999.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.3006968127,"imp":23.6312931455,"elo":1058.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":7.8694051654,"imp":22.2945002711,"elo":1101.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":17.2473724895,"imp":21.598133622,"elo":1131.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":45.0373615697,"imp":21.1082364807,"elo":1162.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":91.3917233329,"imp":20.9763766453,"elo":1170.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":184.6996741609,"imp":20.8893896531,"elo":1176.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":277.8099565315,"imp":20.8512167326,"elo":1179.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":373.2371055096,"imp":20.8463533192,"elo":1181.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.055649238,"imp":18.5902394535,"elo":1235.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.6705594408,"imp":16.2369194285,"elo":1345.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":69.643527334,"imp":14.5607377501,"elo":1438.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":141.1529977589,"imp":13.9427929116,"elo":1472.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":364.6011469598,"imp":13.4671732129,"elo":1502.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":736.9754853516,"imp":13.2471506373,"elo":1515.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1469.2653973047,"imp":13.0647764231,"elo":1524.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2197.5320041687,"imp":12.9722032627,"elo":1529.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2950.7244476948,"imp":12.9046673494,"elo":1532.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.8097985299,"imp":11.6839820929,"elo":1538.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.6950892358,"imp":11.6143468084,"elo":1544.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":45.8180610371,"imp":11.292487148,"elo":1561.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":96.056645332,"imp":11.1111899349,"elo":1578.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":245.1410909168,"imp":10.868476471,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":498.5667116885,"imp":10.6868240527,"elo":1608.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1015.1487598759,"imp":10.5680156723,"elo":1613.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1526.7130408425,"imp":10.5076542069,"elo":1614.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2040.2228420973,"imp":10.4555406579,"elo":1613.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":14.1132668455,"imp":18.2986171534,"elo":1297.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":47.6466878891,"imp":17.2447142424,"elo":1337.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.407219275,"imp":16.4273759647,"elo":1364.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":132.9785409439,"imp":15.6129784957,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":243.7764468154,"imp":15.2584743138,"elo":1415.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":599.1452639259,"imp":14.5879282112,"elo":1442.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1224.6320092925,"imp":14.2465622774,"elo":1456.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2440.7040454743,"imp":14.025515153,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3652.693914187,"imp":13.8836168304,"elo":1472.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4910.3773062373,"imp":13.8120012204,"elo":1474.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0676840515,"imp":14.0203361292,"elo":1472.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.8059088678,"imp":4.0944582803,"elo":1807.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":2.0524528172,"imp":10.1989756415,"elo":1611.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":7.4864639685,"imp":17.0218795202,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":16.6863025317,"imp":16.4957522836,"elo":1334.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":56.5960636953,"imp":15.8572340345,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":118.7995551511,"imp":15.2854538522,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":301.7168598673,"imp":14.7930237659,"elo":1431.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":606.795758935,"imp":14.5049392091,"elo":1447.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1211.551032833,"imp":14.301458535,"elo":1459.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1826.0696865509,"imp":14.1570213529,"elo":1465.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2450.1270774379,"imp":14.1019482991,"elo":1471.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.4825798778,"imp":10.7738922898,"elo":1643.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.65846503,"imp":9.3502735666,"elo":1683.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":384.7456301814,"imp":30.4741238844,"elo":997.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":579.5314694238,"imp":30.347038883,"elo":1006.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":986.9188087142,"imp":30.0809439358,"elo":1014.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1626.2057458405,"imp":29.4210570456,"elo":1032.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3523.6508469471,"imp":28.2250520331,"elo":1055.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":7090.4373429841,"imp":27.0261952401,"elo":1076.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":14236.6030369928,"imp":26.3553702322,"elo":1092.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":21385.328840253,"imp":25.9429054894,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":28729.7373655693,"imp":25.6201849214,"elo":1114.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1753394519,"imp":16.1020832631,"elo":1367.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.9925632138,"imp":22.8295561037,"elo":1082.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":23.3976730308,"imp":21.109494918,"elo":1136.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":67.83522273,"imp":19.5850220402,"elo":1188.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":150.5961763502,"imp":18.8591735804,"elo":1222.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":358.8914252792,"imp":18.1777335918,"elo":1252.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":711.8242682675,"imp":17.7628541744,"elo":1273.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1402.3570975877,"imp":17.4684976164,"elo":1287.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2125.0611186926,"imp":17.3117399788,"elo":1296.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2875.5203407747,"imp":17.1778661155,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.9409389396,"imp":18.8126923396,"elo":1218.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.7736976246,"imp":17.1651628717,"elo":1301.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":17.1658020762,"imp":16.6325079028,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":34.0813643193,"imp":16.3148977734,"elo":1358.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.4155038646,"imp":16.0688154505,"elo":1372.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":172.6999358448,"imp":15.965588217,"elo":1379.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":344.2808053206,"imp":15.8832179383,"elo":1385.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":515.1408947248,"imp":15.8239279243,"elo":1389.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":693.4907982506,"imp":15.7757129878,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":296.6392917616,"imp":23.0430171,"elo":1103.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":761.0284006744,"imp":20.349313351,"elo":1173.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2323.5909795411,"imp":18.8122133517,"elo":1278.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4948.4562156001,"imp":17.541943154,"elo":1341.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12120.2560802151,"imp":15.3539794882,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12685.0815501834,"imp":15.3249586647,"elo":1423.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.2328671519,"imp":26.4634580018,"elo":1050.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":7.7757953049,"imp":23.2274871521,"elo":1112.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":20.6204304219,"imp":19.273412113,"elo":1230.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":41.9962580807,"imp":17.5247895428,"elo":1312.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":105.704979404,"imp":16.6937711198,"elo":1350.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":210.3194448262,"imp":16.424264842,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":422.0870812379,"imp":16.3215950454,"elo":1365.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":631.9086612835,"imp":16.3002737638,"elo":1366.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":846.8860702337,"imp":16.2968984962,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"x_train":5.876676442,"imp":15.5457736304,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0633653548,"x_train":11.6678251256,"imp":15.1182419551,"elo":1413.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1177165074,"x_train":31.9106517624,"imp":14.848578649,"elo":1430.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1714353465,"x_train":62.5721251757,"imp":14.7477568558,"elo":1435.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2216220526,"x_train":155.9049355669,"imp":14.6406177695,"elo":1441.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2666794695,"x_train":317.345916891,"imp":14.6311446381,"elo":1441.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3105276001,"x_train":662.4830210734,"imp":14.6478465613,"elo":1440.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3407350551,"x_train":998.8498259224,"imp":14.6673563694,"elo":1438.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3448570028,"x_train":1346.2106911803,"imp":14.674240095,"elo":1437.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0325848325,"x_train":1.9886321895,"imp":18.6780631139,"elo":1273.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0609444401,"x_train":4.4383850203,"imp":18.3366328085,"elo":1294.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1129020181,"x_train":10.0701848574,"imp":18.116797306,"elo":1307.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1771034691,"x_train":20.4734178164,"imp":17.9259739747,"elo":1318.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2454903367,"x_train":52.7173757556,"imp":17.6537027938,"elo":1330.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2926890222,"x_train":106.3941874905,"imp":17.5022920851,"elo":1338.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.312025413,"x_train":212.8207478205,"imp":17.339314755,"elo":1346.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3237207716,"x_train":318.9957179838,"imp":17.2700034244,"elo":1348.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4040760969,"x_train":427.391750384,"imp":17.2326457261,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"x_train":6.6731468509,"imp":20.2759337918,"elo":1208.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0304247242,"x_train":16.5765343457,"imp":19.8786837436,"elo":1222.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0528424778,"x_train":47.7385589033,"imp":19.4534296615,"elo":1243.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0702878348,"x_train":89.4110724699,"imp":19.2726704565,"elo":1252.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0974716988,"x_train":218.448124708,"imp":19.0438192816,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1141121872,"x_train":440.2545735965,"imp":18.9023411836,"elo":1271.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1277215511,"x_train":856.7647171053,"imp":18.780391801,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1376216405,"x_train":1266.0910927829,"imp":18.7369843166,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1384442291,"x_train":1711.2458519871,"imp":18.6945455008,"elo":1284.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2834081321,"x_train":1.8719729433,"imp":26.1948484129,"elo":1009.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2992675583,"x_train":3.1259596548,"imp":24.3488542313,"elo":1053.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3233863551,"x_train":6.7721912257,"imp":21.6826802288,"elo":1124.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3916757047,"x_train":13.288092342,"imp":20.760558803,"elo":1165.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4883309222,"x_train":32.8419339595,"imp":20.1069857393,"elo":1197.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5677931699,"x_train":65.4938634865,"imp":19.9445772606,"elo":1207.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6638268741,"x_train":130.8369715186,"imp":19.8340399339,"elo":1214.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7241674769,"x_train":196.2244804971,"imp":19.7910126465,"elo":1218.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7662402325,"x_train":263.0379305179,"imp":19.7984513883,"elo":1219.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3731781389,"x_train":2.8561491959,"imp":25.7013457941,"elo":1012.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6641420919,"x_train":5.5343795004,"imp":24.5020141267,"elo":1049.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1863927783,"x_train":14.5510344275,"imp":22.6534704395,"elo":1116.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.732344048,"x_train":29.723343154,"imp":21.7464416276,"elo":1149.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.3920814156,"x_train":72.5444980812,"imp":20.8458197285,"elo":1180.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1392189601,"x_train":145.754371191,"imp":20.3916263193,"elo":1196.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.8048101374,"x_train":293.9475008279,"imp":20.0488916705,"elo":1207.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.2073695134,"x_train":441.8525399102,"imp":19.826636643,"elo":1215.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4558194315,"x_train":593.2377884029,"imp":19.6935269069,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1416787942,"x_train":1.960032688,"imp":19.3647921745,"elo":1188.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6924533472,"x_train":4.1465368891,"imp":17.8536668429,"elo":1267.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9306682769,"x_train":10.3907433568,"imp":16.6855138703,"elo":1346.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1975461537,"x_train":20.2701675373,"imp":16.0840498218,"elo":1386.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5075319607,"x_train":51.487448334,"imp":15.5280964347,"elo":1421.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6740993927,"x_train":103.6855742013,"imp":15.3158433015,"elo":1437.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.0960239899,"x_train":207.5435114784,"imp":15.1552157574,"elo":1447.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.295770736,"x_train":310.2926532568,"imp":15.0933557354,"elo":1451.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"x_train":417.0474735649,"imp":15.0561804683,"elo":1453.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":6.2423742585,"x_train":26.4597512682,"imp":15.4843630349,"elo":1381.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3073466784,"x_train":14.8696863446,"imp":19.2146348547,"elo":1258.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8515440488,"x_train":40.5171687704,"imp":17.9235444693,"elo":1296.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.746831802,"x_train":111.4498310885,"imp":16.0052009744,"elo":1381.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.5416509891,"x_train":224.929440425,"imp":15.5140651615,"elo":1406.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.0322964811,"x_train":554.8880523029,"imp":15.0054551931,"elo":1431.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.4485225391,"x_train":1145.4516650488,"imp":15.0529606984,"elo":1428.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.931406374,"x_train":2295.4928354707,"imp":15.3121547879,"elo":1419.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.7285966831,"x_train":3433.7924746694,"imp":15.5114179492,"elo":1411.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.1379168675,"x_train":4621.6656335635,"imp":15.6253232633,"elo":1406.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"x_train":27.322832323,"imp":32.2510059823,"elo":936.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2337632354,"x_train":60.3304655332,"imp":30.0030285719,"elo":1001.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5732905746,"x_train":157.3958954151,"imp":26.6320054165,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5997539679,"x_train":185.3067056521,"imp":26.4602162594,"elo":1095.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0535405758,"x_train":0.4343250827,"imp":26.2932402488,"elo":999.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1187728047,"x_train":2.3006968127,"imp":23.6312931455,"elo":1058.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.235791088,"x_train":7.8694051654,"imp":22.2945002711,"elo":1101.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3406296664,"x_train":17.2473724895,"imp":21.598133622,"elo":1131.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5043278961,"x_train":45.0373615697,"imp":21.1082364807,"elo":1162.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5878338984,"x_train":91.3917233329,"imp":20.9763766453,"elo":1170.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6965050642,"x_train":184.6996741609,"imp":20.8893896531,"elo":1176.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7595438385,"x_train":277.8099565315,"imp":20.8512167326,"elo":1179.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7784693874,"x_train":373.2371055096,"imp":20.8463533192,"elo":1181.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6887954948,"x_train":10.055649238,"imp":18.5902394535,"elo":1235.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.459301494,"x_train":24.6705594408,"imp":16.2369194285,"elo":1345.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.716219813,"x_train":69.643527334,"imp":14.5607377501,"elo":1438.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.6764824634,"x_train":141.1529977589,"imp":13.9427929116,"elo":1472.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.8477187358,"x_train":364.6011469598,"imp":13.4671732129,"elo":1502.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.7676765361,"x_train":736.9754853516,"imp":13.2471506373,"elo":1515.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.259146773,"x_train":1469.2653973047,"imp":13.0647764231,"elo":1524.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.0531789205,"x_train":2197.5320041687,"imp":12.9722032627,"elo":1529.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.3778248403,"x_train":2950.7244476948,"imp":12.9046673494,"elo":1532.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"x_train":5.8097985299,"imp":11.6839820929,"elo":1538.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4031929556,"x_train":16.6950892358,"imp":11.6143468084,"elo":1544.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8742272233,"x_train":45.8180610371,"imp":11.292487148,"elo":1561.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.0955253444,"x_train":96.056645332,"imp":11.1111899349,"elo":1578.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7534074644,"x_train":245.1410909168,"imp":10.868476471,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8500761173,"x_train":498.5667116885,"imp":10.6868240527,"elo":1608.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.7920736053,"x_train":1015.1487598759,"imp":10.5680156723,"elo":1613.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9997599059,"x_train":1526.7130408425,"imp":10.5076542069,"elo":1614.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.0306819261,"x_train":2040.2228420973,"imp":10.4555406579,"elo":1613.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0745579697,"x_train":14.1132668455,"imp":18.2986171534,"elo":1297.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"x_train":47.6466878891,"imp":17.2447142424,"elo":1337.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":57.2989818274,"x_train":70.407219275,"imp":16.4273759647,"elo":1364.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":81.7804361754,"x_train":132.9785409439,"imp":15.6129784957,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":113.4133274069,"x_train":243.7764468154,"imp":15.2584743138,"elo":1415.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":170.4030706025,"x_train":599.1452639259,"imp":14.5879282112,"elo":1442.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":211.9243576754,"x_train":1224.6320092925,"imp":14.2465622774,"elo":1456.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":252.4669965135,"x_train":2440.7040454743,"imp":14.025515153,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":264.8770310031,"x_train":3652.693914187,"imp":13.8836168304,"elo":1472.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":289.5741517329,"x_train":4910.3773062373,"imp":13.8120012204,"elo":1474.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"x_train":2.0676840515,"imp":14.0203361292,"elo":1472.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.9854290718,"x_train":38.8059088678,"imp":4.0944582803,"elo":1807.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1505492146,"x_train":2.0524528172,"imp":10.1989756415,"elo":1611.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2435252328,"x_train":7.4864639685,"imp":17.0218795202,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.486302031,"x_train":16.6863025317,"imp":16.4957522836,"elo":1334.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.8905707593,"x_train":56.5960636953,"imp":15.8572340345,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.969667724,"x_train":118.7995551511,"imp":15.2854538522,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.2874446083,"x_train":301.7168598673,"imp":14.7930237659,"elo":1431.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.5958040708,"x_train":606.795758935,"imp":14.5049392091,"elo":1447.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.9425270856,"x_train":1211.551032833,"imp":14.301458535,"elo":1459.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.1383865311,"x_train":1826.0696865509,"imp":14.1570213529,"elo":1465.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.2361524691,"x_train":2450.1270774379,"imp":14.1019482991,"elo":1471.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5548034626,"x_train":5.4825798778,"imp":10.7738922898,"elo":1643.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"x_train":3.65846503,"imp":9.3502735666,"elo":1683.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3480868421,"x_train":384.7456301814,"imp":30.4741238844,"elo":997.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.8647718906,"x_train":579.5314694238,"imp":30.347038883,"elo":1006.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8842610276,"x_train":986.9188087142,"imp":30.0809439358,"elo":1014.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.4371098465,"x_train":1626.2057458405,"imp":29.4210570456,"elo":1032.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.4116860286,"x_train":3523.6508469471,"imp":28.2250520331,"elo":1055.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.581168526,"x_train":7090.4373429841,"imp":27.0261952401,"elo":1076.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.0586264094,"x_train":14236.6030369928,"imp":26.3553702322,"elo":1092.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3998340801,"x_train":21385.328840253,"imp":25.9429054894,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"x_train":28729.7373655693,"imp":25.6201849214,"elo":1114.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719683769,"x_train":1.1753394519,"imp":16.1020832631,"elo":1367.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1258046788,"x_train":9.9925632138,"imp":22.8295561037,"elo":1082.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.250263975,"x_train":23.3976730308,"imp":21.109494918,"elo":1136.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5082028825,"x_train":67.83522273,"imp":19.5850220402,"elo":1188.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7461273476,"x_train":150.5961763502,"imp":18.8591735804,"elo":1222.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1062947766,"x_train":358.8914252792,"imp":18.1777335918,"elo":1252.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3608592506,"x_train":711.8242682675,"imp":17.7628541744,"elo":1273.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5999240551,"x_train":1402.3570975877,"imp":17.4684976164,"elo":1287.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.7712720949,"x_train":2125.0611186926,"imp":17.3117399788,"elo":1296.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.95162543,"x_train":2875.5203407747,"imp":17.1778661155,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1226232881,"x_train":1.9409389396,"imp":18.8126923396,"elo":1218.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5491058954,"x_train":5.7736976246,"imp":17.1651628717,"elo":1301.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9903293256,"x_train":17.1658020762,"imp":16.6325079028,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1808822728,"x_train":34.0813643193,"imp":16.3148977734,"elo":1358.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.5144136771,"x_train":86.4155038646,"imp":16.0688154505,"elo":1372.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.6869916704,"x_train":172.6999358448,"imp":15.965588217,"elo":1379.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.732994711,"x_train":344.2808053206,"imp":15.8832179383,"elo":1385.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.809783177,"x_train":515.1408947248,"imp":15.8239279243,"elo":1389.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7397156138,"x_train":693.4907982506,"imp":15.7757129878,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":68.173296937,"x_train":296.6392917616,"imp":23.0430171,"elo":1103.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":133.585345468,"x_train":761.0284006744,"imp":20.349313351,"elo":1173.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":253.2846359591,"x_train":2323.5909795411,"imp":18.8122133517,"elo":1278.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":348.3200767517,"x_train":4948.4562156001,"imp":17.541943154,"elo":1341.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":469.7064233571,"x_train":12120.2560802151,"imp":15.3539794882,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":463.2685781743,"x_train":12685.0815501834,"imp":15.3249586647,"elo":1423.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"x_train":3.2328671519,"imp":26.4634580018,"elo":1050.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8711410437,"x_train":7.7757953049,"imp":23.2274871521,"elo":1112.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1335512608,"x_train":20.6204304219,"imp":19.273412113,"elo":1230.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.4441868365,"x_train":41.9962580807,"imp":17.5247895428,"elo":1312.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8072467962,"x_train":105.704979404,"imp":16.6937711198,"elo":1350.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.1086840459,"x_train":210.3194448262,"imp":16.424264842,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.3155613238,"x_train":422.0870812379,"imp":16.3215950454,"elo":1365.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.5338332871,"x_train":631.9086612835,"imp":16.3002737638,"elo":1366.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.5660475163,"x_train":846.8860702337,"imp":16.2968984962,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_all/winrate_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..6d5bef00a30a8b18b93b777eab21865e0806b344
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_all/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (default)", "RealMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "AutoGluon 1.4 (best, 4h)", "TabDPT (tuned + ensembled)", "TabM (tuned + ensembled)", "TabDPT (tuned)", "RealMLP (tuned)", "LightGBM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "TabM (tuned)", "CatBoost (tuned)", "LimiX (default)", "iLTM (tuned + ensembled)", "TabSwift (default)", "TabDPT (default)", "LightGBM (tuned)", "xRFM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "CatBoost (default)", "ModernNCA (tuned + ensembled)", "ModernNCA (tuned)", "XGBoost (tuned)", "SAP-RPT-OSS (default)", "xRFM (tuned)", "TabM (default)", "ChimeraBoost (tuned + ensembled)", "EBM (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "iLTM (tuned)", "ModernNCA (default)", "ChimeraBoost (tuned)", "RealMLP (default)", "EBM (tuned)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (tuned + ensembled)", "EBM (default)", "ChimeraBoost (default)", "TorchMLP (tuned)", "ExtraTrees (tuned)", "FastaiMLP (tuned)", "XGBoost (default)", "RandomForest (tuned + ensembled)", "LightGBM (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned)", "TorchMLP (default)", "PerpetualBooster (tuned + ensembled)", "iLTM (default)", "xRFM (default)", "PerpetualBooster (tuned)", "TabSTAR (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "RandomForest (default)", "Linear (tuned)", "PerpetualBooster (default)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7708333134651184, 0.751225471496582, 0.8088235259056091, 0.8174019455909729, 0.7873774766921997, 0.7977941036224365, 0.8357843160629272, 0.8651960492134094, 0.8799019455909729, 0.8897058963775635, 0.8823529481887817, 0.9031862616539001, 0.8958333134651184, 0.875, 0.9154411554336548, 0.9154411554336548, 0.9264705777168274, 0.9276960492134094, 0.90625, 0.9240196347236633, 0.8860294222831726, 0.8995097875595093, 0.9276960492134094, 0.9203431606292725, 0.9350489974021912, 0.9227941036224365, 0.9424019455909729, 0.9313725233078003, 0.9411764740943909, 0.8786764740943909, 0.9154411554336548, 0.9325980544090271, 0.9350489974021912, 0.9264705777168274, 0.9595588445663452, 0.9424019455909729, 0.938725471496582, 0.9399510025978088, 0.9473039507865906, 0.9375, 0.9264705777168274, 0.9534313678741455, 0.9350489974021912, 0.9546568393707275, 0.969362735748291, 0.9485294222831726, 0.9497548937797546, 0.9718137383460999, 0.969362735748291, 0.9718137383460999, 0.8725489974021912, 0.8713235259056091, 0.968137264251709, 0.9754902124404907, 0.9865196347236633, 0.9571078419685364, 0.970588207244873, 0.9852941036224365, 0.904411792755127, 0.9865196347236633, 0.9583333134651184, 0.9767156839370728, 0.969362735748291, 0.9852941036224365, 0.9632353186607361, 0.9877451062202454, 0.9754902124404907, 0.9718137383460999, 0.9938725233078003], [0.2291666716337204, NaN, 0.5796568393707275, 0.5551470518112183, 0.6348039507865906, 0.6188725233078003, 0.6703431606292725, 0.7071078419685364, 0.7450980544090271, 0.7745097875595093, 0.751225471496582, 0.7585784196853638, 0.7977941036224365, 0.7916666865348816, 0.779411792755127, 0.8259803652763367, 0.8345588445663452, 0.8370097875595093, 0.8406862616539001, 0.7843137383460999, 0.8639705777168274, 0.8823529481887817, 0.8272058963775635, 0.8615196347236633, 0.8529411554336548, 0.8486519455909729, 0.8541666865348816, 0.8786764740943909, 0.8651960492134094, 0.8590686321258545, 0.7965686321258545, 0.8639705777168274, 0.8700980544090271, 0.8860294222831726, 0.8590686321258545, 0.9105392098426819, 0.9258578419685364, 0.905024528503418, 0.8860294222831726, 0.9252451062202454, 0.876225471496582, 0.8841911554336548, 0.9350489974021912, 0.8964460492134094, 0.8958333134651184, 0.938725471496582, 0.936274528503418, 0.9209558963775635, 0.9509803652763367, 0.938725471496582, 0.9595588445663452, 0.8511029481887817, 0.8492646813392639, 0.9546568393707275, 0.9656862616539001, 0.9473039507865906, 0.9607843160629272, 0.9620097875595093, 0.9644607901573181, 0.8897058963775635, 0.9754902124404907, 0.9325980544090271, 0.966911792755127, 0.9485294222831726, 0.9852941036224365, 0.938725471496582, 0.9779411554336548, 0.9632353186607361, 0.9656862616539001, 0.9950980544090271], [0.24877451360225677, 0.4203431308269501, NaN, 0.4791666567325592, 0.6200980544090271, 0.5539215803146362, 0.6090686321258545, 0.6348039507865906, 0.7775735259056091, 0.6715686321258545, 0.8051470518112183, 0.7377451062202454, 0.8259803652763367, 0.7610294222831726, 0.8186274766921997, 0.8615196347236633, 0.8639705777168274, 0.8388480544090271, 0.8670343160629272, 0.7769607901573181, 0.8909313678741455, 0.8357843160629272, 0.7659313678741455, 0.876225471496582, 0.875, 0.8884803652763367, 0.8792892098426819, 0.8860294222831726, 0.8970588445663452, 0.8946078419685364, 0.7720588445663452, 0.8694853186607361, 0.8835784196853638, 0.9166666865348816, 0.8897058963775635, 0.9252451062202454, 0.9154411554336548, 0.8958333134651184, 0.9240196347236633, 0.9117646813392639, 0.8982843160629272, 0.8982843160629272, 0.9240196347236633, 0.8995097875595093, 0.9301470518112183, 0.9448529481887817, 0.9166666865348816, 0.9166666865348816, 0.9620097875595093, 0.9436274766921997, 0.9607843160629272, 0.8357843160629272, 0.8357843160629272, 0.9460784196853638, 0.969362735748291, 0.9620097875595093, 0.9583333134651184, 0.9595588445663452, 0.967524528503418, 0.873774528503418, 0.9583333134651184, 0.9276960492134094, 0.9656862616539001, 0.9509803652763367, 0.9730392098426819, 0.9436274766921997, 0.9742646813392639, 0.9601715803146362, 0.9644607901573181, 0.9938725233078003], [0.19117647409439087, 0.44485294818878174, 0.5208333134651184, NaN, 0.5870097875595093, 0.5674019455909729, 0.6090686321258545, 0.6525735259056091, 0.7156862616539001, 0.7316176295280457, 0.7450980544090271, 0.7665441036224365, 0.7867646813392639, 0.7683823704719543, 0.7524510025978088, 0.8088235259056091, 0.811274528503418, 0.8088235259056091, 0.8149510025978088, 0.7861519455909729, 0.843137264251709, 0.8370097875595093, 0.7928921580314636, 0.8382353186607361, 0.8247548937797546, 0.8247548937797546, 0.8382353186607361, 0.843137264251709, 0.8480392098426819, 0.8296568393707275, 0.780637264251709, 0.8204656839370728, 0.8566176295280457, 0.8566176295280457, 0.845588207244873, 0.8799019455909729, 0.875, 0.8841911554336548, 0.8664215803146362, 0.9019607901573181, 0.8639705777168274, 0.8700980544090271, 0.8976715803146362, 0.8897058963775635, 0.876225471496582, 0.908088207244873, 0.8995097875595093, 0.906862735748291, 0.9178921580314636, 0.9252451062202454, 0.9411764740943909, 0.844362735748291, 0.8394607901573181, 0.9276960492134094, 0.9460784196853638, 0.9460784196853638, 0.9399510025978088, 0.9448529481887817, 0.9497548937797546, 0.8921568393707275, 0.9583333134651184, 0.9411764740943909, 0.9497548937797546, 0.9546568393707275, 0.968137264251709, 0.9473039507865906, 0.9718137383460999, 0.9662989974021912, 0.9644607901573181, 0.9950980544090271], [0.1825980395078659, 0.3651960790157318, 0.3799019753932953, 0.41299018263816833, NaN, 0.4828431308269501, 0.5435048937797546, 0.6139705777168274, 0.6911764740943909, 0.6433823704719543, 0.7095588445663452, 0.6727941036224365, 0.7757353186607361, 0.7022058963775635, 0.7634803652763367, 0.8075980544090271, 0.813725471496582, 0.8149510025978088, 0.8149510025978088, 0.751225471496582, 0.841911792755127, 0.7990196347236633, 0.7218137383460999, 0.8529411554336548, 0.8235294222831726, 0.8492646813392639, 0.8517156839370728, 0.8590686321258545, 0.8602941036224365, 0.8713235259056091, 0.7916666865348816, 0.843137264251709, 0.8682597875595093, 0.8946078419685364, 0.8615196347236633, 0.8958333134651184, 0.906862735748291, 0.8982843160629272, 0.9031862616539001, 0.9093137383460999, 0.8915441036224365, 0.8725489974021912, 0.908088207244873, 0.8872548937797546, 0.9105392098426819, 0.9350489974021912, 0.905024528503418, 0.9093137383460999, 0.9620097875595093, 0.9313725233078003, 0.9583333134651184, 0.8370097875595093, 0.8406862616539001, 0.9485294222831726, 0.9632353186607361, 0.9558823704719543, 0.9460784196853638, 0.9485294222831726, 0.9534313678741455, 0.8688725233078003, 0.9607843160629272, 0.9252451062202454, 0.9509803652763367, 0.9485294222831726, 0.9779411554336548, 0.9399510025978088, 0.9730392098426819, 0.9571078419685364, 0.9607843160629272, 0.9926470518112183], [0.21262255311012268, 0.3811274468898773, 0.44607841968536377, 0.4325980246067047, 0.5171568393707275, NaN, 0.6666666865348816, 0.6642156839370728, 0.6936274766921997, 0.623774528503418, 0.7352941036224365, 0.6642156839370728, 0.7610294222831726, 0.7022058963775635, 0.7475489974021912, 0.7928921580314636, 0.7977941036224365, 0.7990196347236633, 0.8088235259056091, 0.7598039507865906, 0.7922794222831726, 0.7965686321258545, 0.6985294222831726, 0.8204656839370728, 0.8174019455909729, 0.8284313678741455, 0.8284313678741455, 0.8088235259056091, 0.8517156839370728, 0.842524528503418, 0.78125, 0.8272058963775635, 0.8529411554336548, 0.8486519455909729, 0.8566176295280457, 0.8872548937797546, 0.873774528503418, 0.8627451062202454, 0.8664215803146362, 0.8823529481887817, 0.8780637383460999, 0.8682597875595093, 0.8774510025978088, 0.8799019455909729, 0.8872548937797546, 0.9203431606292725, 0.8799019455909729, 0.9093137383460999, 0.9301470518112183, 0.9178921580314636, 0.9485294222831726, 0.8296568393707275, 0.8296568393707275, 0.9325980544090271, 0.9571078419685364, 0.9571078419685364, 0.936274528503418, 0.936274528503418, 0.9583333134651184, 0.8639705777168274, 0.9558823704719543, 0.9203431606292725, 0.9522058963775635, 0.9301470518112183, 0.9656862616539001, 0.9313725233078003, 0.9742646813392639, 0.9491421580314636, 0.9485294222831726, 0.9914215803146362], [0.20220588147640228, 0.3296568691730499, 0.3909313678741455, 0.3909313678741455, 0.45649510622024536, 0.3333333432674408, NaN, 0.5759803652763367, 0.6421568393707275, 0.5894607901573181, 0.6752451062202454, 0.6213235259056091, 0.7267156839370728, 0.6623774766921997, 0.7071078419685364, 0.7450980544090271, 0.7585784196853638, 0.780024528503418, 0.7647058963775635, 0.7242646813392639, 0.75, 0.7640931606292725, 0.6740196347236633, 0.7953431606292725, 0.7910539507865906, 0.7941176295280457, 0.7855392098426819, 0.7738970518112183, 0.7990196347236633, 0.8063725233078003, 0.7524510025978088, 0.8100489974021912, 0.8308823704719543, 0.8094362616539001, 0.8370097875595093, 0.8639705777168274, 0.8541666865348816, 0.8394607901573181, 0.8345588445663452, 0.8664215803146362, 0.8529411554336548, 0.8504902124404907, 0.8670343160629272, 0.8651960492134094, 0.8639705777168274, 0.904411792755127, 0.875, 0.9031862616539001, 0.9117646813392639, 0.8897058963775635, 0.9375, 0.8051470518112183, 0.8100489974021912, 0.9117646813392639, 0.9558823704719543, 0.938112735748291, 0.9301470518112183, 0.9264705777168274, 0.9448529481887817, 0.8566176295280457, 0.938725471496582, 0.9093137383460999, 0.9448529481887817, 0.9178921580314636, 0.9558823704719543, 0.9240196347236633, 0.9742646813392639, 0.9515931606292725, 0.9509803652763367, 0.9865196347236633], [0.16421568393707275, 0.2928921580314636, 0.3651960790157318, 0.34742647409439087, 0.3860294222831726, 0.33578431606292725, 0.42401960492134094, NaN, 0.5894607901573181, 0.5502451062202454, 0.6593137383460999, 0.5796568393707275, 0.7132353186607361, 0.6280637383460999, 0.6764705777168274, 0.7303921580314636, 0.7475489974021912, 0.7585784196853638, 0.7647058963775635, 0.7279411554336548, 0.7267156839370728, 0.7708333134651184, 0.6691176295280457, 0.7855392098426819, 0.7843137383460999, 0.7745097875595093, 0.7904411554336548, 0.7365196347236633, 0.783088207244873, 0.8039215803146362, 0.7634803652763367, 0.813725471496582, 0.8308823704719543, 0.8161764740943909, 0.8198529481887817, 0.8639705777168274, 0.8584558963775635, 0.844362735748291, 0.8272058963775635, 0.8602941036224365, 0.8480392098426819, 0.873774528503418, 0.8700980544090271, 0.8639705777168274, 0.8664215803146362, 0.904411792755127, 0.8823529481887817, 0.905637264251709, 0.9007353186607361, 0.876225471496582, 0.9338235259056091, 0.8308823704719543, 0.8302696347236633, 0.9117646813392639, 0.9718137383460999, 0.9350489974021912, 0.9399510025978088, 0.9338235259056091, 0.9424019455909729, 0.8713235259056091, 0.9424019455909729, 0.9215686321258545, 0.9485294222831726, 0.908088207244873, 0.9546568393707275, 0.9264705777168274, 0.969362735748291, 0.9620097875595093, 0.9473039507865906, 0.9852941036224365], [0.1348039209842682, 0.2549019753932953, 0.22242647409439087, 0.28431373834609985, 0.30882352590560913, 0.3063725531101227, 0.3578431308269501, 0.4105392098426819, NaN, 0.49754902720451355, 0.53125, 0.4908088147640228, 0.6151960492134094, 0.5349264740943909, 0.6942402124404907, 0.7034313678741455, 0.719362735748291, 0.6825980544090271, 0.7254902124404907, 0.5906862616539001, 0.7401960492134094, 0.595588207244873, 0.5882353186607361, 0.7579656839370728, 0.7671568393707275, 0.7953431606292725, 0.7438725233078003, 0.7034313678741455, 0.7291666865348816, 0.813725471496582, 0.6715686321258545, 0.7720588445663452, 0.8149510025978088, 0.7696078419685364, 0.7977941036224365, 0.781862735748291, 0.8259803652763367, 0.7965686321258545, 0.7910539507865906, 0.9013480544090271, 0.8174019455909729, 0.8474264740943909, 0.8480392098426819, 0.8388480544090271, 0.8149510025978088, 0.8321078419685364, 0.8572303652763367, 0.8609068393707275, 0.9105392098426819, 0.9117646813392639, 0.9276960492134094, 0.781862735748291, 0.7855392098426819, 0.9313725233078003, 0.936274528503418, 0.9436274766921997, 0.9117646813392639, 0.938725471496582, 0.9509803652763367, 0.8370097875595093, 0.9607843160629272, 0.905637264251709, 0.9522058963775635, 0.9252451062202454, 0.9656862616539001, 0.906862735748291, 0.9754902124404907, 0.9203431606292725, 0.9460784196853638, 0.9938725233078003], [0.1200980395078659, 0.22549019753932953, 0.3284313678741455, 0.26838234066963196, 0.35661765933036804, 0.3762255012989044, 0.4105392098426819, 0.44975489377975464, 0.5024510025978088, NaN, 0.5563725233078003, 0.5147058963775635, 0.5741421580314636, 0.5808823704719543, 0.5839460492134094, 0.6286764740943909, 0.6654411554336548, 0.6323529481887817, 0.6642156839370728, 0.6593137383460999, 0.7022058963775635, 0.7077205777168274, 0.7028186321258545, 0.6654411554336548, 0.6936274766921997, 0.6875, 0.6936274766921997, 0.6850489974021912, 0.7046568393707275, 0.7022058963775635, 0.6629902124404907, 0.7365196347236633, 0.716911792755127, 0.7450980544090271, 0.7120097875595093, 0.7757353186607361, 0.7984068393707275, 0.7990196347236633, 0.7671568393707275, 0.8051470518112183, 0.7469362616539001, 0.813725471496582, 0.8486519455909729, 0.7904411554336548, 0.7916666865348816, 0.8284313678741455, 0.8713235259056091, 0.8627451062202454, 0.8553921580314636, 0.8713235259056091, 0.8823529481887817, 0.7647058963775635, 0.7671568393707275, 0.8823529481887817, 0.904411792755127, 0.9019607901573181, 0.8811274766921997, 0.906862735748291, 0.9276960492134094, 0.845588207244873, 0.9454656839370728, 0.8725489974021912, 0.938725471496582, 0.9252451062202454, 0.9620097875595093, 0.8811274766921997, 0.969362735748291, 0.9436274766921997, 0.9571078419685364, 0.9901960492134094], [0.11029411852359772, 0.24877451360225677, 0.19485294818878174, 0.2549019753932953, 0.2904411852359772, 0.2647058963775635, 0.32475489377975464, 0.34068626165390015, 0.46875, 0.4436274468898773, NaN, 0.4436274468898773, 0.5674019455909729, 0.47549018263816833, 0.5882353186607361, 0.720588207244873, 0.686274528503418, 0.6372548937797546, 0.7046568393707275, 0.5453431606292725, 0.7009803652763367, 0.5814951062202454, 0.5563725233078003, 0.7598039507865906, 0.6917892098426819, 0.7696078419685364, 0.75, 0.688725471496582, 0.6997548937797546, 0.7696078419685364, 0.6176470518112183, 0.7377451062202454, 0.7463235259056091, 0.7971813678741455, 0.7732843160629272, 0.8186274766921997, 0.8075980544090271, 0.7855392098426819, 0.8149510025978088, 0.7977941036224365, 0.7928921580314636, 0.8247548937797546, 0.783088207244873, 0.8284313678741455, 0.8504902124404907, 0.8578431606292725, 0.8063725233078003, 0.8627451062202454, 0.9105392098426819, 0.8566176295280457, 0.9240196347236633, 0.7769607901573181, 0.779411792755127, 0.8817402124404907, 0.9215686321258545, 0.936274528503418, 0.9129902124404907, 0.904411792755127, 0.9375, 0.8321078419685364, 0.9301470518112183, 0.9025735259056091, 0.9473039507865906, 0.904411792755127, 0.9411764740943909, 0.9197303652763367, 0.9595588445663452, 0.9411764740943909, 0.9375, 0.9901960492134094], [0.11764705926179886, 0.24142156541347504, 0.26225489377975464, 0.23345588147640228, 0.3272058963775635, 0.33578431606292725, 0.37867647409439087, 0.4203431308269501, 0.5091911554336548, 0.4852941036224365, 0.5563725233078003, NaN, 0.5729166865348816, 0.6593137383460999, 0.5827205777168274, 0.6323529481887817, 0.6372548937797546, 0.624387264251709, 0.6470588445663452, 0.5986519455909729, 0.6973039507865906, 0.6439951062202454, 0.7046568393707275, 0.6617646813392639, 0.6991421580314636, 0.6734068393707275, 0.686274528503418, 0.6936274766921997, 0.6960784196853638, 0.7034313678741455, 0.6605392098426819, 0.7212010025978088, 0.7101715803146362, 0.7303921580314636, 0.7077205777168274, 0.749387264251709, 0.7892156839370728, 0.7787989974021912, 0.75, 0.7781862616539001, 0.7322303652763367, 0.7781862616539001, 0.8174019455909729, 0.748774528503418, 0.7879902124404907, 0.8100489974021912, 0.8370097875595093, 0.8247548937797546, 0.841911792755127, 0.8700980544090271, 0.8774510025978088, 0.751225471496582, 0.7524510025978088, 0.8786764740943909, 0.8946078419685364, 0.9117646813392639, 0.8713235259056091, 0.9172794222831726, 0.9209558963775635, 0.8357843160629272, 0.9509803652763367, 0.8406862616539001, 0.9203431606292725, 0.9172794222831726, 0.9558823704719543, 0.8541666865348816, 0.9571078419685364, 0.904411792755127, 0.9436274766921997, 0.9938725233078003], [0.09681372344493866, 0.20220588147640228, 0.17401960492134094, 0.2132352888584137, 0.2242647111415863, 0.2389705926179886, 0.27328431606292725, 0.2867647111415863, 0.3848039209842682, 0.4258578419685364, 0.4325980246067047, 0.4270833432674408, NaN, 0.469362735748291, 0.49754902720451355, 0.591911792755127, 0.6017156839370728, 0.719362735748291, 0.6188725233078003, 0.5098039507865906, 0.6011029481887817, 0.5343137383460999, 0.5404411554336548, 0.6513480544090271, 0.5968137383460999, 0.6727941036224365, 0.6838235259056091, 0.6286764740943909, 0.6403186321258545, 0.7120097875595093, 0.623774528503418, 0.625, 0.8609068393707275, 0.7218137383460999, 0.7328431606292725, 0.7340686321258545, 0.7267156839370728, 0.7267156839370728, 0.7585784196853638, 0.8229166865348816, 0.7769607901573181, 0.7757353186607361, 0.7426470518112183, 0.7861519455909729, 0.8075980544090271, 0.814338207244873, 0.748774528503418, 0.8308823704719543, 0.8578431606292725, 0.7867646813392639, 0.8829656839370728, 0.7732843160629272, 0.7757353186607361, 0.8210784196853638, 0.9227941036224365, 0.8860294222831726, 0.905637264251709, 0.8590686321258545, 0.905637264251709, 0.8296568393707275, 0.8921568393707275, 0.9031862616539001, 0.9178921580314636, 0.8909313678741455, 0.9203431606292725, 0.9148284196853638, 0.9558823704719543, 0.9350489974021912, 0.9276960492134094, 0.9754902124404907], [0.1041666641831398, 0.2083333283662796, 0.2389705926179886, 0.23161764442920685, 0.2977941036224365, 0.2977941036224365, 0.3376225531101227, 0.37193626165390015, 0.46507352590560913, 0.41911765933036804, 0.5245097875595093, 0.34068626165390015, 0.530637264251709, NaN, 0.5257353186607361, 0.5857843160629272, 0.595588207244873, 0.5870097875595093, 0.6060048937797546, 0.5759803652763367, 0.6311274766921997, 0.5882353186607361, 0.6458333134651184, 0.6225489974021912, 0.654411792755127, 0.6439951062202454, 0.6421568393707275, 0.6464460492134094, 0.6470588445663452, 0.654411792755127, 0.6139705777168274, 0.6691176295280457, 0.6703431606292725, 0.6960784196853638, 0.6666666865348816, 0.7095588445663452, 0.7340686321258545, 0.7242646813392639, 0.7095588445663452, 0.7340686321258545, 0.7058823704719543, 0.7365196347236633, 0.7732843160629272, 0.7242646813392639, 0.7475489974021912, 0.7634803652763367, 0.7916666865348816, 0.7953431606292725, 0.7855392098426819, 0.811274528503418, 0.8321078419685364, 0.7224264740943909, 0.7218137383460999, 0.8394607901573181, 0.8725489974021912, 0.8933823704719543, 0.8345588445663452, 0.8933823704719543, 0.9019607901573181, 0.8235294222831726, 0.9227941036224365, 0.8265931606292725, 0.905637264251709, 0.9105392098426819, 0.9301470518112183, 0.841911792755127, 0.9558823704719543, 0.8921568393707275, 0.9350489974021912, 0.9914215803146362], [0.125, 0.22058823704719543, 0.18137255311012268, 0.24754901230335236, 0.23651960492134094, 0.25245097279548645, 0.2928921580314636, 0.3235294222831726, 0.30575981736183167, 0.4160539209842682, 0.4117647111415863, 0.4172794222831726, 0.5024510025978088, 0.4742647111415863, NaN, 0.5833333134651184, 0.5882353186607361, 0.5857843160629272, 0.5968137383460999, 0.53125, 0.59375, 0.5183823704719543, 0.5355392098426819, 0.6421568393707275, 0.6200980544090271, 0.6764705777168274, 0.6421568393707275, 0.6078431606292725, 0.6421568393707275, 0.7040441036224365, 0.6078431606292725, 0.6721813678741455, 0.7316176295280457, 0.6666666865348816, 0.720588207244873, 0.7340686321258545, 0.7303921580314636, 0.7414215803146362, 0.7034313678741455, 0.8026960492134094, 0.7585784196853638, 0.7732843160629272, 0.7622548937797546, 0.7585784196853638, 0.75, 0.8051470518112183, 0.7843137383460999, 0.8296568393707275, 0.845588207244873, 0.844362735748291, 0.8566176295280457, 0.7659313678741455, 0.7671568393707275, 0.8627451062202454, 0.905637264251709, 0.9007353186607361, 0.876225471496582, 0.8811274766921997, 0.908088207244873, 0.8210784196853638, 0.8897058963775635, 0.8651960492134094, 0.9307597875595093, 0.8897058963775635, 0.9203431606292725, 0.8719362616539001, 0.9620097875595093, 0.8995097875595093, 0.9166666865348816, 0.9828431606292725], [0.08455882221460342, 0.17401960492134094, 0.13848039507865906, 0.19117647409439087, 0.1924019604921341, 0.20710784196853638, 0.2549019753932953, 0.2696078419685364, 0.2965686321258545, 0.37132352590560913, 0.27941176295280457, 0.36764705181121826, 0.40808823704719543, 0.41421568393707275, 0.4166666567325592, NaN, 0.5183823704719543, 0.4944852888584137, 0.5520833134651184, 0.44485294818878174, 0.5275735259056091, 0.5257353186607361, 0.5134803652763367, 0.655024528503418, 0.5539215803146362, 0.6801470518112183, 0.6709558963775635, 0.5465686321258545, 0.5772058963775635, 0.716911792755127, 0.5563725233078003, 0.6151960492134094, 0.688112735748291, 0.7224264740943909, 0.7040441036224365, 0.6397058963775635, 0.6715686321258545, 0.6924019455909729, 0.7634803652763367, 0.7230392098426819, 0.7475489974021912, 0.7144607901573181, 0.7555146813392639, 0.8174019455909729, 0.8321078419685364, 0.7340686321258545, 0.7585784196853638, 0.7904411554336548, 0.8958333134651184, 0.8345588445663452, 0.9571078419685364, 0.7591911554336548, 0.7598039507865906, 0.873774528503418, 0.8492646813392639, 0.9142156839370728, 0.9191176295280457, 0.8321078419685364, 0.9289215803146362, 0.8284313678741455, 0.9178921580314636, 0.8811274766921997, 0.8823529481887817, 0.8982843160629272, 0.9485294222831726, 0.8958333134651184, 0.9595588445663452, 0.936274528503418, 0.9344362616539001, 0.9865196347236633], [0.08455882221460342, 0.16544117033481598, 0.1360294073820114, 0.18872548639774323, 0.18627451360225677, 0.20220588147640228, 0.24142156541347504, 0.25245097279548645, 0.280637264251709, 0.3345588147640228, 0.3137255012989044, 0.36274510622024536, 0.39828431606292725, 0.40441176295280457, 0.4117647111415863, 0.48161765933036804, NaN, 0.46568626165390015, 0.594975471496582, 0.45588234066963196, 0.5232843160629272, 0.5171568393707275, 0.47181373834609985, 0.5508578419685364, 0.5404411554336548, 0.626838207244873, 0.6482843160629272, 0.5747548937797546, 0.5851715803146362, 0.657475471496582, 0.5808823704719543, 0.6200980544090271, 0.6372548937797546, 0.6875, 0.6642156839370728, 0.655637264251709, 0.6948529481887817, 0.6740196347236633, 0.7242646813392639, 0.7071078419685364, 0.7064951062202454, 0.6924019455909729, 0.7561274766921997, 0.7561274766921997, 0.780637264251709, 0.7438725233078003, 0.7787989974021912, 0.7610294222831726, 0.8872548937797546, 0.843137264251709, 0.8921568393707275, 0.7585784196853638, 0.7549019455909729, 0.8780637383460999, 0.8566176295280457, 0.875, 0.8854166865348816, 0.8553921580314636, 0.8933823704719543, 0.811274528503418, 0.9019607901573181, 0.8590686321258545, 0.8774510025978088, 0.8995097875595093, 0.9424019455909729, 0.8713235259056091, 0.9325980544090271, 0.8958333134651184, 0.9264705777168274, 0.9877451062202454], [0.07352941483259201, 0.16299019753932953, 0.1611519604921341, 0.19117647409439087, 0.18504901230335236, 0.20098039507865906, 0.21997548639774323, 0.24142156541347504, 0.3174019753932953, 0.36764705181121826, 0.36274510622024536, 0.375612735748291, 0.280637264251709, 0.41299018263816833, 0.41421568393707275, 0.5055146813392639, 0.5343137383460999, NaN, 0.5502451062202454, 0.48713234066963196, 0.531862735748291, 0.4828431308269501, 0.4944852888584137, 0.5502451062202454, 0.5269607901573181, 0.5802696347236633, 0.6029411554336548, 0.5674019455909729, 0.5833333134651184, 0.6041666865348816, 0.5465686321258545, 0.5735294222831726, 0.720588207244873, 0.6593137383460999, 0.6433823704719543, 0.6740196347236633, 0.655637264251709, 0.6703431606292725, 0.6629902124404907, 0.7634803652763367, 0.6973039507865906, 0.7236519455909729, 0.6924019455909729, 0.7328431606292725, 0.7285539507865906, 0.7542892098426819, 0.7095588445663452, 0.7928921580314636, 0.8051470518112183, 0.751225471496582, 0.8002451062202454, 0.7561274766921997, 0.7604166865348816, 0.7696078419685364, 0.904411792755127, 0.8651960492134094, 0.8848039507865906, 0.8186274766921997, 0.875, 0.8100489974021912, 0.844362735748291, 0.8811274766921997, 0.8774510025978088, 0.8615196347236633, 0.8884803652763367, 0.8897058963775635, 0.9375, 0.9178921580314636, 0.908088207244873, 0.968137264251709], [0.07230392098426819, 0.15931372344493866, 0.13296568393707275, 0.18504901230335236, 0.18504901230335236, 0.19117647409439087, 0.23529411852359772, 0.23529411852359772, 0.27450981736183167, 0.33578431606292725, 0.2953431308269501, 0.3529411852359772, 0.3811274468898773, 0.39399510622024536, 0.40318626165390015, 0.4479166567325592, 0.4050244987010956, 0.44975489377975464, NaN, 0.4534313678741455, 0.4950980246067047, 0.4883578419685364, 0.46078431606292725, 0.531862735748291, 0.5214460492134094, 0.6029411554336548, 0.6072303652763367, 0.5600489974021912, 0.563725471496582, 0.6409313678741455, 0.5698529481887817, 0.6017156839370728, 0.625612735748291, 0.6501225233078003, 0.655637264251709, 0.6495097875595093, 0.658088207244873, 0.656862735748291, 0.6991421580314636, 0.6948529481887817, 0.6936274766921997, 0.6740196347236633, 0.7365196347236633, 0.7549019455909729, 0.7659313678741455, 0.7310048937797546, 0.748774528503418, 0.751838207244873, 0.8823529481887817, 0.8284313678741455, 0.8664215803146362, 0.7585784196853638, 0.7573529481887817, 0.8602941036224365, 0.8406862616539001, 0.8651960492134094, 0.876225471496582, 0.8468137383460999, 0.8860294222831726, 0.8125, 0.8884803652763367, 0.8553921580314636, 0.8651960492134094, 0.8909313678741455, 0.9338235259056091, 0.8658088445663452, 0.936274528503418, 0.8909313678741455, 0.9301470518112183, 0.9865196347236633], [0.09375, 0.21568627655506134, 0.22303920984268188, 0.2138480395078659, 0.24877451360225677, 0.2401960790157318, 0.2757352888584137, 0.2720588147640228, 0.40931373834609985, 0.34068626165390015, 0.4546568691730499, 0.4013480246067047, 0.4901960790157318, 0.42401960492134094, 0.46875, 0.5551470518112183, 0.5441176295280457, 0.5128676295280457, 0.5465686321258545, NaN, 0.5765931606292725, 0.4840686321258545, 0.46078431606292725, 0.5784313678741455, 0.5453431606292725, 0.5882353186607361, 0.6090686321258545, 0.5698529481887817, 0.6053921580314636, 0.6041666865348816, 0.5465686321258545, 0.5655637383460999, 0.5870097875595093, 0.6740196347236633, 0.6703431606292725, 0.6397058963775635, 0.6176470518112183, 0.6654411554336548, 0.688725471496582, 0.6642156839370728, 0.6985294222831726, 0.686887264251709, 0.6985294222831726, 0.7022058963775635, 0.7022058963775635, 0.6997548937797546, 0.7022058963775635, 0.7542892098426819, 0.7536764740943909, 0.7524510025978088, 0.7438725233078003, 0.7285539507865906, 0.7254902124404907, 0.7708333134651184, 0.781862735748291, 0.780637264251709, 0.8039215803146362, 0.8308823704719543, 0.7953431606292725, 0.7775735259056091, 0.813725471496582, 0.8566176295280457, 0.8394607901573181, 0.8370097875595093, 0.8602941036224365, 0.8713235259056091, 0.8713235259056091, 0.8909313678741455, 0.876225471496582, 0.968137264251709], [0.07598039507865906, 0.1360294073820114, 0.1090686246752739, 0.1568627506494522, 0.15808823704719543, 0.2077205926179886, 0.25, 0.27328431606292725, 0.2598039209842682, 0.2977941036224365, 0.29901960492134094, 0.3026960790157318, 0.39889705181121826, 0.3688725531101227, 0.40625, 0.47242647409439087, 0.47671568393707275, 0.468137264251709, 0.5049019455909729, 0.4234068691730499, NaN, 0.4803921580314636, 0.44607841968536377, 0.5453431606292725, 0.5496323704719543, 0.5680146813392639, 0.5502451062202454, 0.5514705777168274, 0.5894607901573181, 0.6164215803146362, 0.5575980544090271, 0.6182597875595093, 0.5710784196853638, 0.6531862616539001, 0.6360294222831726, 0.6654411554336548, 0.7371323704719543, 0.7009803652763367, 0.6715686321258545, 0.7279411554336548, 0.6740196347236633, 0.7352941036224365, 0.7420343160629272, 0.7107843160629272, 0.71875, 0.7450980544090271, 0.7683823704719543, 0.7861519455909729, 0.8308823704719543, 0.8272058963775635, 0.8700980544090271, 0.7218137383460999, 0.7224264740943909, 0.8553921580314636, 0.8688725233078003, 0.8541666865348816, 0.8823529481887817, 0.8946078419685364, 0.8884803652763367, 0.8063725233078003, 0.9093137383460999, 0.813725471496582, 0.8970588445663452, 0.8774510025978088, 0.9276960492134094, 0.8394607901573181, 0.9313725233078003, 0.8786764740943909, 0.9087010025978088, 0.9828431606292725], [0.11397058516740799, 0.11764705926179886, 0.16421568393707275, 0.16299019753932953, 0.20098039507865906, 0.2034313678741455, 0.23590686917304993, 0.2291666716337204, 0.40441176295280457, 0.2922794222831726, 0.41850489377975464, 0.35600489377975464, 0.46568626165390015, 0.4117647111415863, 0.48161765933036804, 0.4742647111415863, 0.4828431308269501, 0.5171568393707275, 0.5116421580314636, 0.5159313678741455, 0.5196078419685364, NaN, 0.5061274766921997, 0.5208333134651184, 0.5428921580314636, 0.5441176295280457, 0.5367646813392639, 0.5453431606292725, 0.5710784196853638, 0.5833333134651184, 0.5821078419685364, 0.5796568393707275, 0.6176470518112183, 0.6017156839370728, 0.6225489974021912, 0.6666666865348816, 0.6666666865348816, 0.6776960492134094, 0.6323529481887817, 0.720588207244873, 0.658088207244873, 0.7671568393707275, 0.6850489974021912, 0.6789215803146362, 0.7046568393707275, 0.7254902124404907, 0.7071078419685364, 0.7928921580314636, 0.7389705777168274, 0.718137264251709, 0.7837010025978088, 0.7622548937797546, 0.7634803652763367, 0.7585784196853638, 0.8529411554336548, 0.8541666865348816, 0.8321078419685364, 0.8345588445663452, 0.8725489974021912, 0.8345588445663452, 0.8615196347236633, 0.8259803652763367, 0.8946078419685364, 0.8823529481887817, 0.8897058963775635, 0.845588207244873, 0.9325980544090271, 0.9172794222831726, 0.9534313678741455, 0.9938725233078003], [0.10049019753932953, 0.17279411852359772, 0.2340686321258545, 0.20710784196853638, 0.27818626165390015, 0.3014705777168274, 0.32598039507865906, 0.33088234066963196, 0.4117647111415863, 0.2971813678741455, 0.4436274468898773, 0.2953431308269501, 0.4595588147640228, 0.3541666567325592, 0.4644607901573181, 0.48651960492134094, 0.5281862616539001, 0.5055146813392639, 0.5392156839370728, 0.5392156839370728, 0.5539215803146362, 0.4938725531101227, NaN, 0.5343137383460999, 0.5698529481887817, 0.5674019455909729, 0.5477941036224365, 0.5625, 0.5710784196853638, 0.5882353186607361, 0.5735294222831726, 0.5968137383460999, 0.6200980544090271, 0.6299019455909729, 0.5833333134651184, 0.6335784196853638, 0.6495097875595093, 0.6642156839370728, 0.6397058963775635, 0.6378676295280457, 0.6121323704719543, 0.6715686321258545, 0.7303921580314636, 0.656862735748291, 0.6813725233078003, 0.6911764740943909, 0.751225471496582, 0.7316176295280457, 0.718137264251709, 0.7536764740943909, 0.7598039507865906, 0.688725471496582, 0.6899510025978088, 0.7916666865348816, 0.8161764740943909, 0.8259803652763367, 0.811274528503418, 0.8394607901573181, 0.8615196347236633, 0.7855392098426819, 0.8848039507865906, 0.7781862616539001, 0.876225471496582, 0.875, 0.876225471496582, 0.7990196347236633, 0.9215686321258545, 0.876225471496582, 0.906862735748291, 0.9926470518112183], [0.07230392098426819, 0.13848039507865906, 0.12377450615167618, 0.1617647111415863, 0.14705882966518402, 0.17953431606292725, 0.20465686917304993, 0.21446079015731812, 0.24203431606292725, 0.3345588147640228, 0.2401960790157318, 0.3382352888584137, 0.3486519753932953, 0.37745097279548645, 0.3578431308269501, 0.3449755012989044, 0.4491421580314636, 0.44975489377975464, 0.468137264251709, 0.4215686321258545, 0.4546568691730499, 0.4791666567325592, 0.46568626165390015, NaN, 0.499387264251709, 0.560661792755127, 0.5900735259056091, 0.5024510025978088, 0.5196078419685364, 0.6041666865348816, 0.5183823704719543, 0.5502451062202454, 0.6397058963775635, 0.6470588445663452, 0.6642156839370728, 0.6066176295280457, 0.6066176295280457, 0.6507353186607361, 0.6776960492134094, 0.6942402124404907, 0.7144607901573181, 0.6807597875595093, 0.7077205777168274, 0.7598039507865906, 0.7696078419685364, 0.688112735748291, 0.7322303652763367, 0.7414215803146362, 0.8609068393707275, 0.7769607901573181, 0.9160539507865906, 0.75, 0.751225471496582, 0.8284313678741455, 0.8284313678741455, 0.876838207244873, 0.9037989974021912, 0.8204656839370728, 0.8982843160629272, 0.813725471496582, 0.8933823704719543, 0.8615196347236633, 0.8676470518112183, 0.876225471496582, 0.9252451062202454, 0.8805146813392639, 0.9546568393707275, 0.9227941036224365, 0.9264705777168274, 0.9816176295280457], [0.07965686172246933, 0.14705882966518402, 0.125, 0.17524509131908417, 0.1764705926179886, 0.1825980395078659, 0.2089460790157318, 0.21568627655506134, 0.23284313082695007, 0.3063725531101227, 0.3082107901573181, 0.3008578419685364, 0.40318626165390015, 0.34558823704719543, 0.3799019753932953, 0.44607841968536377, 0.4595588147640228, 0.4730392098426819, 0.4785539209842682, 0.4546568691730499, 0.45036765933036804, 0.4571078419685364, 0.43014705181121826, 0.500612735748291, NaN, 0.5165441036224365, 0.5349264740943909, 0.5147058963775635, 0.5165441036224365, 0.5490196347236633, 0.5085784196853638, 0.6427696347236633, 0.5710784196853638, 0.6115196347236633, 0.6066176295280457, 0.591911792755127, 0.6207107901573181, 0.6225489974021912, 0.6299019455909729, 0.6397058963775635, 0.6605392098426819, 0.6636029481887817, 0.6838235259056091, 0.6850489974021912, 0.6593137383460999, 0.6764705777168274, 0.7230392098426819, 0.7291666865348816, 0.7573529481887817, 0.7671568393707275, 0.7622548937797546, 0.7162989974021912, 0.7144607901573181, 0.7990196347236633, 0.8468137383460999, 0.8161764740943909, 0.8100489974021912, 0.9497548937797546, 0.8498774766921997, 0.781862735748291, 0.9350489974021912, 0.8676470518112183, 0.905637264251709, 0.90625, 0.9411764740943909, 0.8799019455909729, 0.9215686321258545, 0.9037989974021912, 0.9289215803146362, 0.9889705777168274], [0.06495098024606705, 0.1513480395078659, 0.11151960492134094, 0.17524509131908417, 0.1507352888584137, 0.1715686321258545, 0.20588235557079315, 0.22549019753932953, 0.20465686917304993, 0.3125, 0.23039215803146362, 0.3265931308269501, 0.3272058963775635, 0.35600489377975464, 0.3235294222831726, 0.31985294818878174, 0.37316176295280457, 0.41973039507865906, 0.3970588147640228, 0.4117647111415863, 0.4319852888584137, 0.45588234066963196, 0.4325980246067047, 0.43933823704719543, 0.4834558963775635, NaN, 0.5275735259056091, 0.4779411852359772, 0.5183823704719543, 0.6047794222831726, 0.5147058963775635, 0.5686274766921997, 0.5857843160629272, 0.6029411554336548, 0.623774528503418, 0.6121323704719543, 0.6102941036224365, 0.6397058963775635, 0.6397058963775635, 0.6642156839370728, 0.6789215803146362, 0.6678921580314636, 0.6899510025978088, 0.7328431606292725, 0.7150735259056091, 0.6789215803146362, 0.7291666865348816, 0.720588207244873, 0.8995097875595093, 0.8180146813392639, 0.9105392098426819, 0.7377451062202454, 0.7340686321258545, 0.8412989974021912, 0.8186274766921997, 0.8321078419685364, 0.8541666865348816, 0.8272058963775635, 0.8609068393707275, 0.7916666865348816, 0.8786764740943909, 0.8504902124404907, 0.845588207244873, 0.8664215803146362, 0.9350489974021912, 0.8590686321258545, 0.9227941036224365, 0.8897058963775635, 0.9148284196853638, 0.9791666865348816], [0.07720588147640228, 0.1458333283662796, 0.12071078270673752, 0.1617647111415863, 0.14828431606292725, 0.1715686321258545, 0.21446079015731812, 0.20955882966518402, 0.2561274468898773, 0.3063725531101227, 0.25, 0.3137255012989044, 0.31617647409439087, 0.3578431308269501, 0.3578431308269501, 0.3290441036224365, 0.35171568393707275, 0.3970588147640228, 0.39276960492134094, 0.3909313678741455, 0.44975489377975464, 0.4632352888584137, 0.4522058963775635, 0.40992647409439087, 0.46507352590560913, 0.47242647409439087, NaN, 0.4779411852359772, 0.5036764740943909, 0.5128676295280457, 0.5122548937797546, 0.5575980544090271, 0.5723039507865906, 0.6280637383460999, 0.6286764740943909, 0.5992646813392639, 0.6115196347236633, 0.594362735748291, 0.6752451062202454, 0.656862735748291, 0.6838235259056091, 0.6654411554336548, 0.6875, 0.7328431606292725, 0.7757353186607361, 0.6924019455909729, 0.7046568393707275, 0.7328431606292725, 0.783088207244873, 0.7769607901573181, 0.8161764740943909, 0.750612735748291, 0.751225471496582, 0.8063725233078003, 0.8125, 0.8480392098426819, 0.9093137383460999, 0.7941176295280457, 0.8799019455909729, 0.8149510025978088, 0.8664215803146362, 0.8394607901573181, 0.876225471496582, 0.8725489974021912, 0.9019607901573181, 0.8553921580314636, 0.9276960492134094, 0.905637264251709, 0.9325980544090271, 0.9840686321258545], [0.057598039507865906, 0.12132352590560913, 0.11397058516740799, 0.1568627506494522, 0.1409313678741455, 0.19117647409439087, 0.22610294818878174, 0.26348039507865906, 0.2965686321258545, 0.31495097279548645, 0.3112744987010956, 0.3063725531101227, 0.37132352590560913, 0.3535539209842682, 0.3921568691730499, 0.4534313678741455, 0.42524510622024536, 0.4325980246067047, 0.43995097279548645, 0.43014705181121826, 0.4485294222831726, 0.4546568691730499, 0.4375, 0.49754902720451355, 0.4852941036224365, 0.5220588445663452, 0.5220588445663452, NaN, 0.6011029481887817, 0.5281862616539001, 0.49142158031463623, 0.5784313678741455, 0.5428921580314636, 0.5772058963775635, 0.5980392098426819, 0.5784313678741455, 0.5808823704719543, 0.7107843160629272, 0.6164215803146362, 0.6715686321258545, 0.6366421580314636, 0.6519607901573181, 0.6629902124404907, 0.654411792755127, 0.6348039507865906, 0.658088207244873, 0.6924019455909729, 0.7107843160629272, 0.7622548937797546, 0.7303921580314636, 0.7426470518112183, 0.6789215803146362, 0.6764705777168274, 0.7414215803146362, 0.7769607901573181, 0.783088207244873, 0.7965686321258545, 0.8596813678741455, 0.7916666865348816, 0.7377451062202454, 0.843137264251709, 0.7671568393707275, 0.8210784196853638, 0.8235294222831726, 0.8553921580314636, 0.7781862616539001, 0.8578431606292725, 0.813725471496582, 0.8517156839370728, 0.9583333134651184], [0.06862745434045792, 0.1348039209842682, 0.10294117778539658, 0.15196079015731812, 0.13970588147640228, 0.14828431606292725, 0.20098039507865906, 0.21691176295280457, 0.2708333432674408, 0.2953431308269501, 0.30024510622024536, 0.30392158031463623, 0.3596813678741455, 0.3529411852359772, 0.3578431308269501, 0.4227941036224365, 0.41482841968536377, 0.4166666567325592, 0.4362744987010956, 0.3946078419685364, 0.4105392098426819, 0.42892158031463623, 0.42892158031463623, 0.4803921580314636, 0.4834558963775635, 0.48161765933036804, 0.49632352590560913, 0.39889705181121826, NaN, 0.49693626165390015, 0.49142158031463623, 0.5416666865348816, 0.5416666865348816, 0.5808823704719543, 0.5821078419685364, 0.5833333134651184, 0.5698529481887817, 0.6924019455909729, 0.59375, 0.6531862616539001, 0.6274510025978088, 0.6648284196853638, 0.6482843160629272, 0.654411792755127, 0.6482843160629272, 0.6997548937797546, 0.6691176295280457, 0.7463235259056091, 0.718137264251709, 0.7414215803146362, 0.7254902124404907, 0.7218137383460999, 0.7254902124404907, 0.7598039507865906, 0.8063725233078003, 0.7904411554336548, 0.8492646813392639, 0.8553921580314636, 0.8259803652763367, 0.7879902124404907, 0.8468137383460999, 0.8223039507865906, 0.8645833134651184, 0.8627451062202454, 0.8774510025978088, 0.844362735748291, 0.8958333134651184, 0.8884803652763367, 0.905637264251709, 0.9779411554336548], [0.05882352963089943, 0.1409313678741455, 0.10539215803146362, 0.17034313082695007, 0.12867647409439087, 0.15747548639774323, 0.19362744688987732, 0.19607843458652496, 0.18627451360225677, 0.2977941036224365, 0.23039215803146362, 0.2965686321258545, 0.28799018263816833, 0.34558823704719543, 0.2959558963775635, 0.28308823704719543, 0.3425244987010956, 0.3958333432674408, 0.3590686321258545, 0.3958333432674408, 0.38357841968536377, 0.4166666567325592, 0.4117647111415863, 0.3958333432674408, 0.45098039507865906, 0.3952205777168274, 0.48713234066963196, 0.47181373834609985, 0.5030637383460999, NaN, 0.5073529481887817, 0.532475471496582, 0.5514705777168274, 0.5484068393707275, 0.5980392098426819, 0.5857843160629272, 0.5747548937797546, 0.6078431606292725, 0.594362735748291, 0.6360294222831726, 0.656862735748291, 0.6433823704719543, 0.6832107901573181, 0.7022058963775635, 0.6727941036224365, 0.6593137383460999, 0.716911792755127, 0.7156862616539001, 0.8529411554336548, 0.7892156839370728, 0.8676470518112183, 0.7242646813392639, 0.7261029481887817, 0.8198529481887817, 0.8063725233078003, 0.8051470518112183, 0.8480392098426819, 0.8161764740943909, 0.845588207244873, 0.7892156839370728, 0.876225471496582, 0.841911792755127, 0.8357843160629272, 0.8633578419685364, 0.9203431606292725, 0.8578431606292725, 0.9215686321258545, 0.8848039507865906, 0.9105392098426819, 0.9767156839370728], [0.12132352590560913, 0.2034313678741455, 0.22794117033481598, 0.2193627506494522, 0.2083333283662796, 0.21875, 0.24754901230335236, 0.23651960492134094, 0.3284313678741455, 0.33700981736183167, 0.38235294818878174, 0.3394607901573181, 0.3762255012989044, 0.3860294222831726, 0.3921568691730499, 0.4436274468898773, 0.41911765933036804, 0.4534313678741455, 0.43014705181121826, 0.4534313678741455, 0.4424019753932953, 0.4178921580314636, 0.4264705777168274, 0.48161765933036804, 0.49142158031463623, 0.4852941036224365, 0.48774510622024536, 0.5085784196853638, 0.5085784196853638, 0.49264705181121826, NaN, 0.4950980246067047, 0.5159313678741455, 0.5232843160629272, 0.5563725233078003, 0.5857843160629272, 0.530024528503418, 0.5833333134651184, 0.5441176295280457, 0.6225489974021912, 0.593137264251709, 0.6703431606292725, 0.625, 0.6329656839370728, 0.594362735748291, 0.626225471496582, 0.6433823704719543, 0.7267156839370728, 0.6470588445663452, 0.6770833134651184, 0.6801470518112183, 0.6507353186607361, 0.6458333134651184, 0.688112735748291, 0.7549019455909729, 0.7683823704719543, 0.7267156839370728, 0.7622548937797546, 0.7775735259056091, 0.7953431606292725, 0.7904411554336548, 0.8247548937797546, 0.8161764740943909, 0.7904411554336548, 0.7990196347236633, 0.8308823704719543, 0.8382353186607361, 0.8627451062202454, 0.8884803652763367, 0.9852941036224365], [0.08455882221460342, 0.1360294073820114, 0.1305147111415863, 0.17953431606292725, 0.1568627506494522, 0.17279411852359772, 0.18995098769664764, 0.18627451360225677, 0.22794117033481598, 0.26348039507865906, 0.26225489377975464, 0.27879902720451355, 0.375, 0.33088234066963196, 0.3278186321258545, 0.3848039209842682, 0.3799019753932953, 0.4264705777168274, 0.39828431606292725, 0.43443626165390015, 0.38174018263816833, 0.4203431308269501, 0.40318626165390015, 0.44975489377975464, 0.35723039507865906, 0.4313725531101227, 0.4424019753932953, 0.4215686321258545, 0.4583333432674408, 0.4675244987010956, 0.5049019455909729, NaN, 0.531862735748291, 0.5447303652763367, 0.563112735748291, 0.5698529481887817, 0.5661764740943909, 0.593137264251709, 0.5870097875595093, 0.5992646813392639, 0.6164215803146362, 0.6305146813392639, 0.6372548937797546, 0.6397058963775635, 0.623774528503418, 0.6458333134651184, 0.6850489974021912, 0.7058823704719543, 0.6764705777168274, 0.7022058963775635, 0.7046568393707275, 0.716911792755127, 0.719362735748291, 0.7438725233078003, 0.8382353186607361, 0.780024528503418, 0.813725471496582, 0.9019607901573181, 0.8204656839370728, 0.7904411554336548, 0.8688725233078003, 0.8627451062202454, 0.8860294222831726, 0.8933823704719543, 0.8909313678741455, 0.8799019455909729, 0.908088207244873, 0.9178921580314636, 0.9276960492134094, 0.9865196347236633], [0.0674019604921341, 0.1299019604921341, 0.11642156541347504, 0.14338235557079315, 0.13174019753932953, 0.14705882966518402, 0.16911764442920685, 0.16911764442920685, 0.18504901230335236, 0.28308823704719543, 0.25367647409439087, 0.28982841968536377, 0.13909313082695007, 0.3296568691730499, 0.26838234066963196, 0.311887264251709, 0.36274510622024536, 0.27941176295280457, 0.374387264251709, 0.41299018263816833, 0.42892158031463623, 0.38235294818878174, 0.3799019753932953, 0.3602941036224365, 0.42892158031463623, 0.41421568393707275, 0.4276960790157318, 0.4571078419685364, 0.4583333432674408, 0.4485294222831726, 0.4840686321258545, 0.468137264251709, NaN, 0.4779411852359772, 0.5024510025978088, 0.5723039507865906, 0.5361519455909729, 0.5674019455909729, 0.5049019455909729, 0.6102941036224365, 0.5710784196853638, 0.6433823704719543, 0.6188725233078003, 0.5998774766921997, 0.5759803652763367, 0.658088207244873, 0.6397058963775635, 0.7242646813392639, 0.6715686321258545, 0.6875, 0.654411792755127, 0.7352941036224365, 0.7377451062202454, 0.6985294222831726, 0.7965686321258545, 0.7628676295280457, 0.8014705777168274, 0.7291666865348816, 0.7745097875595093, 0.8125, 0.7916666865348816, 0.8272058963775635, 0.8247548937797546, 0.8186274766921997, 0.8321078419685364, 0.8388480544090271, 0.8823529481887817, 0.8823529481887817, 0.9007353186607361, 0.9522058963775635], [0.06495098024606705, 0.11397058516740799, 0.0833333358168602, 0.14338235557079315, 0.10539215803146362, 0.1513480395078659, 0.19056372344493866, 0.18382352590560913, 0.23039215803146362, 0.2549019753932953, 0.2028186321258545, 0.2696078419685364, 0.27818626165390015, 0.30392158031463623, 0.3333333432674408, 0.27757352590560913, 0.3125, 0.34068626165390015, 0.3498774468898773, 0.32598039507865906, 0.34681373834609985, 0.39828431606292725, 0.3700980246067047, 0.3529411852359772, 0.38848039507865906, 0.3970588147640228, 0.37193626165390015, 0.4227941036224365, 0.41911765933036804, 0.4515931308269501, 0.47671568393707275, 0.45526960492134094, 0.5220588445663452, NaN, 0.5539215803146362, 0.5281862616539001, 0.5159313678741455, 0.47181373834609985, 0.6072303652763367, 0.5698529481887817, 0.595588207244873, 0.561274528503418, 0.5747548937797546, 0.6311274766921997, 0.7965686321258545, 0.591911792755127, 0.6102941036224365, 0.6084558963775635, 0.6776960492134094, 0.6776960492134094, 0.7322303652763367, 0.7095588445663452, 0.7046568393707275, 0.7156862616539001, 0.716911792755127, 0.7867646813392639, 0.8174019455909729, 0.7426470518112183, 0.8149510025978088, 0.7855392098426819, 0.7879902124404907, 0.7855392098426819, 0.8161764740943909, 0.7953431606292725, 0.8210784196853638, 0.8002451062202454, 0.8700980544090271, 0.8700980544090271, 0.8860294222831726, 0.9736519455909729], [0.07352941483259201, 0.1409313678741455, 0.11029411852359772, 0.15441176295280457, 0.13848039507865906, 0.14338235557079315, 0.16299019753932953, 0.18014705181121826, 0.20220588147640228, 0.28799018263816833, 0.22671568393707275, 0.2922794222831726, 0.2671568691730499, 0.3333333432674408, 0.27941176295280457, 0.2959558963775635, 0.33578431606292725, 0.35661765933036804, 0.344362735748291, 0.3296568691730499, 0.3639705777168274, 0.37745097279548645, 0.4166666567325592, 0.33578431606292725, 0.39338234066963196, 0.3762255012989044, 0.37132352590560913, 0.4019607901573181, 0.4178921580314636, 0.4019607901573181, 0.4436274468898773, 0.436887264251709, 0.49754902720451355, 0.44607841968536377, NaN, 0.5257353186607361, 0.4828431308269501, 0.5257353186607361, 0.48651960492134094, 0.5625, 0.7212010025978088, 0.5508578419685364, 0.6023284196853638, 0.7683823704719543, 0.5600489974021912, 0.6158088445663452, 0.6127451062202454, 0.623161792755127, 0.6415441036224365, 0.658088207244873, 0.6360294222831726, 0.7230392098426819, 0.7144607901573181, 0.6838235259056091, 0.7463235259056091, 0.7549019455909729, 0.8026960492134094, 0.7083333134651184, 0.7726715803146362, 0.8100489974021912, 0.7585784196853638, 0.8615196347236633, 0.8321078419685364, 0.8296568393707275, 0.7953431606292725, 0.875, 0.8848039507865906, 0.9007353186607361, 0.8878676295280457, 0.9522058963775635], [0.040441177785396576, 0.08946078270673752, 0.07475490123033524, 0.1200980395078659, 0.1041666641831398, 0.11274509876966476, 0.1360294073820114, 0.1360294073820114, 0.2181372493505478, 0.2242647111415863, 0.18137255311012268, 0.250612735748291, 0.2659313678741455, 0.2904411852359772, 0.2659313678741455, 0.3602941036224365, 0.344362735748291, 0.32598039507865906, 0.35049018263816833, 0.3602941036224365, 0.3345588147640228, 0.3333333432674408, 0.36642158031463623, 0.39338234066963196, 0.40808823704719543, 0.38786765933036804, 0.4007352888584137, 0.4215686321258545, 0.4166666567325592, 0.41421568393707275, 0.41421568393707275, 0.43014705181121826, 0.4276960790157318, 0.47181373834609985, 0.4742647111415863, NaN, 0.5067402124404907, 0.5441176295280457, 0.5085784196853638, 0.5551470518112183, 0.5392156839370728, 0.6384803652763367, 0.6323529481887817, 0.5796568393707275, 0.5281862616539001, 0.748774528503418, 0.6470588445663452, 0.7432597875595093, 0.6531862616539001, 0.6875, 0.6482843160629272, 0.7034313678741455, 0.7101715803146362, 0.7120097875595093, 0.8921568393707275, 0.7218137383460999, 0.8069853186607361, 0.7683823704719543, 0.7708333134651184, 0.8002451062202454, 0.7647058963775635, 0.811887264251709, 0.8584558963775635, 0.8198529481887817, 0.7965686321258545, 0.8370097875595093, 0.8823529481887817, 0.8982843160629272, 0.8933823704719543, 0.9595588445663452], [0.057598039507865906, 0.07414215803146362, 0.08455882221460342, 0.125, 0.09313725680112839, 0.12622548639774323, 0.1458333283662796, 0.14154411852359772, 0.17401960492134094, 0.20159313082695007, 0.1924019604921341, 0.21078431606292725, 0.27328431606292725, 0.2659313678741455, 0.2696078419685364, 0.3284313678741455, 0.30514705181121826, 0.344362735748291, 0.34191176295280457, 0.38235294818878174, 0.26286765933036804, 0.3333333432674408, 0.35049018263816833, 0.39338234066963196, 0.3792892098426819, 0.3897058963775635, 0.38848039507865906, 0.41911765933036804, 0.43014705181121826, 0.42524510622024536, 0.4699755012989044, 0.43382352590560913, 0.4638480246067047, 0.4840686321258545, 0.5171568393707275, 0.49325981736183167, NaN, 0.560661792755127, 0.5349264740943909, 0.5710784196853638, 0.5698529481887817, 0.6188725233078003, 0.594362735748291, 0.6041666865348816, 0.5747548937797546, 0.6041666865348816, 0.6446078419685364, 0.6703431606292725, 0.6666666865348816, 0.6960784196853638, 0.6924019455909729, 0.6917892098426819, 0.688725471496582, 0.7291666865348816, 0.8100489974021912, 0.7230392098426819, 0.810661792755127, 0.811274528503418, 0.7708333134651184, 0.7598039507865906, 0.8161764740943909, 0.7616421580314636, 0.8468137383460999, 0.8149510025978088, 0.8468137383460999, 0.7916666865348816, 0.8651960492134094, 0.8492646813392639, 0.873161792755127, 0.9620097875595093], [0.061274509876966476, 0.09497549384832382, 0.1041666641831398, 0.11580882221460342, 0.10171568393707275, 0.13725490868091583, 0.16053920984268188, 0.1556372493505478, 0.2034313678741455, 0.20098039507865906, 0.21446079015731812, 0.22120098769664764, 0.27328431606292725, 0.2757352888584137, 0.25857841968536377, 0.3075980246067047, 0.32598039507865906, 0.3296568691730499, 0.343137264251709, 0.3345588147640228, 0.29901960492134094, 0.3223039209842682, 0.33578431606292725, 0.3492647111415863, 0.37745097279548645, 0.3602941036224365, 0.405637264251709, 0.28921568393707275, 0.3075980246067047, 0.3921568691730499, 0.4166666567325592, 0.406862735748291, 0.4325980246067047, 0.5281862616539001, 0.4742647111415863, 0.45588234066963196, 0.43933823704719543, NaN, 0.5441176295280457, 0.49264705181121826, 0.5196078419685364, 0.5477941036224365, 0.5686274766921997, 0.563725471496582, 0.5857843160629272, 0.5416666865348816, 0.6041666865348816, 0.6200980544090271, 0.5882353186607361, 0.6617646813392639, 0.623774528503418, 0.6789215803146362, 0.6838235259056091, 0.7218137383460999, 0.7732843160629272, 0.7242646813392639, 0.7928921580314636, 0.7647058963775635, 0.7769607901573181, 0.779411792755127, 0.813112735748291, 0.7977941036224365, 0.8247548937797546, 0.8651960492134094, 0.8578431606292725, 0.8180146813392639, 0.8639705777168274, 0.8982843160629272, 0.9093137383460999, 0.9803921580314636], [0.06004901975393295, 0.11397058516740799, 0.07598039507865906, 0.13357843458652496, 0.09681372344493866, 0.13357843458652496, 0.16544117033481598, 0.17279411852359772, 0.2089460790157318, 0.23284313082695007, 0.18504901230335236, 0.25, 0.24142156541347504, 0.2904411852359772, 0.2965686321258545, 0.23651960492134094, 0.2757352888584137, 0.33700981736183167, 0.3008578419685364, 0.3112744987010956, 0.3284313678741455, 0.36764705181121826, 0.3602941036224365, 0.3223039209842682, 0.3700980246067047, 0.3602941036224365, 0.32475489377975464, 0.38357841968536377, 0.40625, 0.405637264251709, 0.45588234066963196, 0.41299018263816833, 0.4950980246067047, 0.39276960492134094, 0.5134803652763367, 0.49142158031463623, 0.46507352590560913, 0.45588234066963196, NaN, 0.5526960492134094, 0.5667892098426819, 0.5379902124404907, 0.5526960492134094, 0.595588207244873, 0.7156862616539001, 0.5625, 0.593137264251709, 0.5876225233078003, 0.6501225233078003, 0.6495097875595093, 0.686274528503418, 0.6973039507865906, 0.6960784196853638, 0.6899510025978088, 0.7071078419685364, 0.7469362616539001, 0.7916666865348816, 0.7218137383460999, 0.7763480544090271, 0.7708333134651184, 0.7769607901573181, 0.7732843160629272, 0.7941176295280457, 0.7769607901573181, 0.7904411554336548, 0.779411792755127, 0.8633578419685364, 0.8566176295280457, 0.8639705777168274, 0.9620097875595093], [0.05269607901573181, 0.07475490123033524, 0.0882352963089943, 0.09803921729326248, 0.09068627655506134, 0.11764705926179886, 0.13357843458652496, 0.13970588147640228, 0.0986519604921341, 0.19485294818878174, 0.20220588147640228, 0.22181372344493866, 0.1770833283662796, 0.2659313678741455, 0.1973039209842682, 0.2769607901573181, 0.2928921580314636, 0.23651960492134094, 0.30514705181121826, 0.33578431606292725, 0.2720588147640228, 0.27941176295280457, 0.36213234066963196, 0.30575981736183167, 0.3602941036224365, 0.33578431606292725, 0.343137264251709, 0.3284313678741455, 0.34681373834609985, 0.3639705777168274, 0.37745097279548645, 0.4007352888584137, 0.3897058963775635, 0.43014705181121826, 0.4375, 0.44485294818878174, 0.42892158031463623, 0.5073529481887817, 0.4473039209842682, NaN, 0.5049019455909729, 0.5422794222831726, 0.5735294222831726, 0.5343137383460999, 0.5110294222831726, 0.5833333134651184, 0.6066176295280457, 0.6531862616539001, 0.6139705777168274, 0.6531862616539001, 0.6102941036224365, 0.6740196347236633, 0.6813725233078003, 0.6875, 0.7763480544090271, 0.7156862616539001, 0.7720588445663452, 0.720588207244873, 0.7720588445663452, 0.7781862616539001, 0.7953431606292725, 0.7953431606292725, 0.8725489974021912, 0.8284313678741455, 0.8014705777168274, 0.8259803652763367, 0.876225471496582, 0.8872548937797546, 0.906862735748291, 0.9791666865348816], [0.0625, 0.12377450615167618, 0.10171568393707275, 0.1360294073820114, 0.10845588147640228, 0.12193627655506134, 0.14705882966518402, 0.15196079015731812, 0.1825980395078659, 0.25306373834609985, 0.20710784196853638, 0.26776960492134094, 0.22303920984268188, 0.29411765933036804, 0.24142156541347504, 0.25245097279548645, 0.29350489377975464, 0.3026960790157318, 0.3063725531101227, 0.3014705777168274, 0.32598039507865906, 0.34191176295280457, 0.38786765933036804, 0.2855392098426819, 0.3394607901573181, 0.32107841968536377, 0.31617647409439087, 0.3633578419685364, 0.37254902720451355, 0.343137264251709, 0.406862735748291, 0.38357841968536377, 0.42892158031463623, 0.40441176295280457, 0.27879902720451355, 0.46078431606292725, 0.43014705181121826, 0.4803921580314636, 0.4332107901573181, 0.4950980246067047, NaN, 0.5024510025978088, 0.5477941036224365, 0.6623774766921997, 0.5036764740943909, 0.5674019455909729, 0.5821078419685364, 0.594975471496582, 0.591911792755127, 0.6115196347236633, 0.5845588445663452, 0.7034313678741455, 0.7003676295280457, 0.6519607901573181, 0.716911792755127, 0.7218137383460999, 0.7598039507865906, 0.686274528503418, 0.7463235259056091, 0.7953431606292725, 0.7426470518112183, 0.8345588445663452, 0.7990196347236633, 0.8026960492134094, 0.7683823704719543, 0.8504902124404907, 0.8639705777168274, 0.8823529481887817, 0.8860294222831726, 0.9460784196853638], [0.07352941483259201, 0.11580882221460342, 0.10171568393707275, 0.1299019604921341, 0.12745098769664764, 0.13174019753932953, 0.14950980246067047, 0.12622548639774323, 0.15257352590560913, 0.18627451360225677, 0.17524509131908417, 0.22181372344493866, 0.2242647111415863, 0.26348039507865906, 0.22671568393707275, 0.2855392098426819, 0.3075980246067047, 0.2763480246067047, 0.32598039507865906, 0.313112735748291, 0.2647058963775635, 0.23284313082695007, 0.3284313678741455, 0.31924018263816833, 0.33639705181121826, 0.3321078419685364, 0.3345588147640228, 0.3480392098426819, 0.33517158031463623, 0.35661765933036804, 0.3296568691730499, 0.3694852888584137, 0.35661765933036804, 0.4387255012989044, 0.4491421580314636, 0.36151960492134094, 0.3811274468898773, 0.4522058963775635, 0.46200981736183167, 0.4577205777168274, 0.49754902720451355, NaN, 0.5085784196853638, 0.5171568393707275, 0.4852941036224365, 0.4901960790157318, 0.5257353186607361, 0.6850489974021912, 0.5404411554336548, 0.6047794222831726, 0.5337010025978088, 0.6311274766921997, 0.6348039507865906, 0.6225489974021912, 0.6942402124404907, 0.6384803652763367, 0.7328431606292725, 0.7034313678741455, 0.6727941036224365, 0.7426470518112183, 0.7450980544090271, 0.8272058963775635, 0.9509803652763367, 0.7855392098426819, 0.7303921580314636, 0.8480392098426819, 0.8186274766921997, 0.8970588445663452, 0.8602941036224365, 0.9546568393707275], [0.046568628400564194, 0.06495098024606705, 0.07598039507865906, 0.10232843458652496, 0.09191176295280457, 0.12254901975393295, 0.13296568393707275, 0.1299019604921341, 0.15196079015731812, 0.1513480395078659, 0.21691176295280457, 0.1825980395078659, 0.25735294818878174, 0.22671568393707275, 0.23774509131908417, 0.2444852888584137, 0.24387255311012268, 0.3075980246067047, 0.26348039507865906, 0.3014705777168274, 0.25796568393707275, 0.31495097279548645, 0.2696078419685364, 0.2922794222831726, 0.31617647409439087, 0.31004902720451355, 0.3125, 0.33700981736183167, 0.35171568393707275, 0.3167892098426819, 0.375, 0.36274510622024536, 0.3811274468898773, 0.42524510622024536, 0.39767158031463623, 0.36764705181121826, 0.405637264251709, 0.4313725531101227, 0.4473039209842682, 0.4264705777168274, 0.4522058963775635, 0.49142158031463623, NaN, 0.4803921580314636, 0.49754902720451355, 0.44424018263816833, 0.6807597875595093, 0.5392156839370728, 0.5196078419685364, 0.6911764740943909, 0.5379902124404907, 0.595588207244873, 0.6053921580314636, 0.7438725233078003, 0.6127451062202454, 0.6507353186607361, 0.6985294222831726, 0.7132353186607361, 0.6838235259056091, 0.7322303652763367, 0.9191176295280457, 0.7420343160629272, 0.75, 0.8688725233078003, 0.8860294222831726, 0.7579656839370728, 0.780637264251709, 0.8308823704719543, 0.9129902124404907, 0.9914215803146362], [0.06495098024606705, 0.10355392098426819, 0.10049019753932953, 0.11029411852359772, 0.11274509876966476, 0.1200980395078659, 0.1348039209842682, 0.1360294073820114, 0.1611519604921341, 0.20955882966518402, 0.1715686321258545, 0.2512255012989044, 0.2138480395078659, 0.2757352888584137, 0.24142156541347504, 0.1825980395078659, 0.24387255311012268, 0.2671568691730499, 0.2450980395078659, 0.2977941036224365, 0.28921568393707275, 0.32107841968536377, 0.343137264251709, 0.2401960790157318, 0.31495097279548645, 0.2671568691730499, 0.2671568691730499, 0.34558823704719543, 0.34558823704719543, 0.2977941036224365, 0.36703431606292725, 0.3602941036224365, 0.4001225531101227, 0.3688725531101227, 0.23161764442920685, 0.4203431308269501, 0.3958333432674408, 0.4362744987010956, 0.40441176295280457, 0.46568626165390015, 0.3376225531101227, 0.4828431308269501, 0.5196078419685364, NaN, 0.4632352888584137, 0.5379902124404907, 0.5379902124404907, 0.5686274766921997, 0.5281862616539001, 0.5772058963775635, 0.5428921580314636, 0.6727941036224365, 0.6703431606292725, 0.6090686321258545, 0.7058823704719543, 0.6801470518112183, 0.7561274766921997, 0.6433823704719543, 0.7071078419685364, 0.7781862616539001, 0.7279411554336548, 0.8198529481887817, 0.7879902124404907, 0.8186274766921997, 0.7414215803146362, 0.844975471496582, 0.8382353186607361, 0.8995097875595093, 0.8897058963775635, 0.9485294222831726], [0.04534313827753067, 0.1041666641831398, 0.06985294073820114, 0.12377450615167618, 0.08946078270673752, 0.11274509876966476, 0.1360294073820114, 0.13357843458652496, 0.18504901230335236, 0.2083333283662796, 0.14950980246067047, 0.21200980246067047, 0.1924019604921341, 0.25245097279548645, 0.25, 0.16789215803146362, 0.2193627506494522, 0.2714460790157318, 0.2340686321258545, 0.2977941036224365, 0.28125, 0.2953431308269501, 0.3186274468898773, 0.23039215803146362, 0.34068626165390015, 0.28492647409439087, 0.2242647111415863, 0.3651960790157318, 0.35171568393707275, 0.3272058963775635, 0.405637264251709, 0.3762255012989044, 0.42401960492134094, 0.2034313678741455, 0.43995097279548645, 0.47181373834609985, 0.42524510622024536, 0.41421568393707275, 0.28431373834609985, 0.4889705777168274, 0.49632352590560913, 0.5147058963775635, 0.5024510025978088, 0.5367646813392639, NaN, 0.531862735748291, 0.5232843160629272, 0.5526960492134094, 0.5863970518112183, 0.6127451062202454, 0.5912989974021912, 0.6727941036224365, 0.6776960492134094, 0.6286764740943909, 0.656862735748291, 0.6924019455909729, 0.7561274766921997, 0.6776960492134094, 0.7279411554336548, 0.7426470518112183, 0.6850489974021912, 0.7659313678741455, 0.7475489974021912, 0.7334558963775635, 0.7573529481887817, 0.7757353186607361, 0.8198529481887817, 0.8553921580314636, 0.8639705777168274, 0.9607843160629272], [0.030637254938483238, 0.061274509876966476, 0.05514705926179886, 0.09191176295280457, 0.06495098024606705, 0.07965686172246933, 0.09558823704719543, 0.09558823704719543, 0.16789215803146362, 0.1715686321258545, 0.14215686917304993, 0.18995098769664764, 0.18566176295280457, 0.23651960492134094, 0.19485294818878174, 0.2659313678741455, 0.2561274468898773, 0.24571079015731812, 0.26899510622024536, 0.30024510622024536, 0.2549019753932953, 0.27450981736183167, 0.30882352590560913, 0.311887264251709, 0.3235294222831726, 0.32107841968536377, 0.3075980246067047, 0.34191176295280457, 0.30024510622024536, 0.34068626165390015, 0.3737744987010956, 0.3541666567325592, 0.34191176295280457, 0.40808823704719543, 0.3841911852359772, 0.2512255012989044, 0.3958333432674408, 0.4583333432674408, 0.4375, 0.4166666567325592, 0.4325980246067047, 0.5098039507865906, 0.5557597875595093, 0.46200981736183167, 0.468137264251709, NaN, 0.5747548937797546, 0.6066176295280457, 0.5367646813392639, 0.6078431606292725, 0.5545343160629272, 0.6850489974021912, 0.686274528503418, 0.6372548937797546, 0.8082107901573181, 0.6531862616539001, 0.7438725233078003, 0.7083333134651184, 0.7009803652763367, 0.7781862616539001, 0.718137264251709, 0.7426470518112183, 0.781862735748291, 0.7775735259056091, 0.7450980544090271, 0.7604166865348816, 0.8235294222831726, 0.8700980544090271, 0.8566176295280457, 0.9473039507865906], [0.05147058889269829, 0.06372549384832382, 0.0833333358168602, 0.10049019753932953, 0.09497549384832382, 0.1200980395078659, 0.125, 0.11764705926179886, 0.14276960492134094, 0.12867647409439087, 0.19362744688987732, 0.16299019753932953, 0.2512255012989044, 0.2083333283662796, 0.21568627655506134, 0.24142156541347504, 0.22120098769664764, 0.2904411852359772, 0.2512255012989044, 0.2977941036224365, 0.23161764442920685, 0.2928921580314636, 0.24877451360225677, 0.26776960492134094, 0.2769607901573181, 0.2708333432674408, 0.2953431308269501, 0.3075980246067047, 0.33088234066963196, 0.28308823704719543, 0.35661765933036804, 0.31495097279548645, 0.3602941036224365, 0.3897058963775635, 0.38725489377975464, 0.3529411852359772, 0.3553921580314636, 0.3958333432674408, 0.406862735748291, 0.39338234066963196, 0.4178921580314636, 0.4742647111415863, 0.31924018263816833, 0.46200981736183167, 0.47671568393707275, 0.42524510622024536, NaN, 0.5220588445663452, 0.4644607901573181, 0.593137264251709, 0.4950980246067047, 0.5962010025978088, 0.6017156839370728, 0.6611519455909729, 0.5857843160629272, 0.6286764740943909, 0.6764705777168274, 0.6776960492134094, 0.6642156839370728, 0.7279411554336548, 0.8504902124404907, 0.7279411554336548, 0.7291666865348816, 0.8382353186607361, 0.8486519455909729, 0.7377451062202454, 0.7683823704719543, 0.8235294222831726, 0.9178921580314636, 0.9901960492134094], [0.050245098769664764, 0.07904411852359772, 0.0833333358168602, 0.09313725680112839, 0.09068627655506134, 0.09068627655506134, 0.09681372344493866, 0.09436274319887161, 0.13909313082695007, 0.13725490868091583, 0.13725490868091583, 0.17524509131908417, 0.16911764442920685, 0.20465686917304993, 0.17034313082695007, 0.20955882966518402, 0.2389705926179886, 0.20710784196853638, 0.24816176295280457, 0.24571079015731812, 0.2138480395078659, 0.20710784196853638, 0.26838234066963196, 0.25857841968536377, 0.2708333432674408, 0.27941176295280457, 0.2671568691730499, 0.28921568393707275, 0.25367647409439087, 0.28431373834609985, 0.27328431606292725, 0.29411765933036804, 0.2757352888584137, 0.3915441036224365, 0.37683823704719543, 0.25674018263816833, 0.3296568691730499, 0.3799019753932953, 0.4123774468898773, 0.34681373834609985, 0.4050244987010956, 0.31495097279548645, 0.46078431606292725, 0.4313725531101227, 0.4473039209842682, 0.39338234066963196, 0.4779411852359772, NaN, 0.469362735748291, 0.532475471496582, 0.4644607901573181, 0.591911792755127, 0.5894607901573181, 0.5514705777168274, 0.6121323704719543, 0.5833333134651184, 0.6740196347236633, 0.6348039507865906, 0.626225471496582, 0.7218137383460999, 0.7009803652763367, 0.7647058963775635, 0.8333333134651184, 0.7426470518112183, 0.688725471496582, 0.7873774766921997, 0.7561274766921997, 0.8651960492134094, 0.8321078419685364, 0.9460784196853638], [0.02818627469241619, 0.04901960864663124, 0.03799019753932953, 0.08210784196853638, 0.03799019753932953, 0.06985294073820114, 0.0882352963089943, 0.0992647036910057, 0.08946078270673752, 0.14460784196853638, 0.08946078270673752, 0.15808823704719543, 0.14215686917304993, 0.21446079015731812, 0.15441176295280457, 0.1041666641831398, 0.11274509876966476, 0.19485294818878174, 0.11764705926179886, 0.24632352590560913, 0.16911764442920685, 0.2610294222831726, 0.281862735748291, 0.13909313082695007, 0.24264705181121826, 0.10049019753932953, 0.21691176295280457, 0.23774509131908417, 0.281862735748291, 0.14705882966518402, 0.3529411852359772, 0.3235294222831726, 0.3284313678741455, 0.3223039209842682, 0.3584558963775635, 0.34681373834609985, 0.3333333432674408, 0.4117647111415863, 0.3498774468898773, 0.3860294222831726, 0.40808823704719543, 0.4595588147640228, 0.4803921580314636, 0.47181373834609985, 0.41360294818878174, 0.4632352888584137, 0.5355392098426819, 0.530637264251709, NaN, 0.5490196347236633, 0.48100489377975464, 0.6397058963775635, 0.6433823704719543, 0.6066176295280457, 0.6360294222831726, 0.5735294222831726, 0.75, 0.7046568393707275, 0.6507353186607361, 0.7242646813392639, 0.751225471496582, 0.7218137383460999, 0.718137264251709, 0.7696078419685364, 0.8100489974021912, 0.7365196347236633, 0.7745097875595093, 0.8223039507865906, 0.844362735748291, 0.9448529481887817], [0.030637254938483238, 0.061274509876966476, 0.05637254938483238, 0.07475490123033524, 0.06862745434045792, 0.08210784196853638, 0.11029411852359772, 0.12377450615167618, 0.0882352963089943, 0.12867647409439087, 0.14338235557079315, 0.1299019604921341, 0.2132352888584137, 0.18872548639774323, 0.1556372493505478, 0.16544117033481598, 0.1568627506494522, 0.24877451360225677, 0.1715686321258545, 0.24754901230335236, 0.17279411852359772, 0.281862735748291, 0.24632352590560913, 0.22303920984268188, 0.23284313082695007, 0.1819852888584137, 0.22303920984268188, 0.2696078419685364, 0.25857841968536377, 0.21078431606292725, 0.3229166567325592, 0.2977941036224365, 0.3125, 0.3223039209842682, 0.34191176295280457, 0.3125, 0.30392158031463623, 0.3382352888584137, 0.35049018263816833, 0.34681373834609985, 0.38848039507865906, 0.3952205777168274, 0.30882352590560913, 0.4227941036224365, 0.38725489377975464, 0.3921568691730499, 0.406862735748291, 0.4675244987010956, 0.45098039507865906, NaN, 0.468137264251709, 0.5661764740943909, 0.5674019455909729, 0.6789215803146362, 0.5698529481887817, 0.593137264251709, 0.6274510025978088, 0.6605392098426819, 0.6286764740943909, 0.6954656839370728, 0.781862735748291, 0.7156862616539001, 0.7022058963775635, 0.8014705777168274, 0.8958333134651184, 0.7254902124404907, 0.7267156839370728, 0.8057597875595093, 0.8860294222831726, 0.9718137383460999], [0.02818627469241619, 0.040441177785396576, 0.03921568766236305, 0.05882352963089943, 0.0416666679084301, 0.05147058889269829, 0.0625, 0.06617647409439087, 0.07230392098426819, 0.11764705926179886, 0.07598039507865906, 0.12254901975393295, 0.11703431606292725, 0.16789215803146362, 0.14338235557079315, 0.04289215803146362, 0.10784313827753067, 0.19975490868091583, 0.13357843458652496, 0.2561274468898773, 0.1299019604921341, 0.21629901230335236, 0.2401960790157318, 0.08394607901573181, 0.23774509131908417, 0.08946078270673752, 0.18382352590560913, 0.25735294818878174, 0.27450981736183167, 0.13235294818878174, 0.31985294818878174, 0.2953431308269501, 0.34558823704719543, 0.26776960492134094, 0.3639705777168274, 0.35171568393707275, 0.3075980246067047, 0.3762255012989044, 0.3137255012989044, 0.3897058963775635, 0.4154411852359772, 0.46629902720451355, 0.46200981736183167, 0.4571078419685364, 0.40870097279548645, 0.44546568393707275, 0.5049019455909729, 0.5355392098426819, 0.5189951062202454, 0.531862735748291, NaN, 0.6335784196853638, 0.6299019455909729, 0.5741421580314636, 0.6482843160629272, 0.6078431606292725, 0.7058823704719543, 0.686274528503418, 0.6691176295280457, 0.718137264251709, 0.7303921580314636, 0.6973039507865906, 0.7291666865348816, 0.7524510025978088, 0.7726715803146362, 0.7120097875595093, 0.8051470518112183, 0.813725471496582, 0.8492646813392639, 0.9552696347236633], [0.12745098769664764, 0.14889705181121826, 0.16421568393707275, 0.1556372493505478, 0.16299019753932953, 0.17034313082695007, 0.19485294818878174, 0.16911764442920685, 0.2181372493505478, 0.23529411852359772, 0.22303920984268188, 0.24877451360225677, 0.22671568393707275, 0.27757352590560913, 0.2340686321258545, 0.24080882966518402, 0.24142156541347504, 0.24387255311012268, 0.24142156541347504, 0.2714460790157318, 0.27818626165390015, 0.23774509131908417, 0.3112744987010956, 0.25, 0.28370097279548645, 0.26225489377975464, 0.2493872493505478, 0.32107841968536377, 0.27818626165390015, 0.2757352888584137, 0.3492647111415863, 0.28308823704719543, 0.2647058963775635, 0.2904411852359772, 0.2769607901573181, 0.2965686321258545, 0.3082107901573181, 0.32107841968536377, 0.3026960790157318, 0.32598039507865906, 0.2965686321258545, 0.3688725531101227, 0.40441176295280457, 0.3272058963775635, 0.3272058963775635, 0.31495097279548645, 0.40379902720451355, 0.40808823704719543, 0.3602941036224365, 0.43382352590560913, 0.36642158031463623, NaN, 0.53125, 0.47181373834609985, 0.4264705777168274, 0.5551470518112183, 0.43014705181121826, 0.5257353186607361, 0.5772058963775635, 0.9534313678741455, 0.6225489974021912, 0.6629902124404907, 0.623774528503418, 0.6948529481887817, 0.6102941036224365, 0.6850489974021912, 0.6139705777168274, 0.7892156839370728, 0.8259803652763367, 0.9031862616539001], [0.12867647409439087, 0.1507352888584137, 0.16421568393707275, 0.16053920984268188, 0.15931372344493866, 0.17034313082695007, 0.18995098769664764, 0.16973039507865906, 0.21446079015731812, 0.23284313082695007, 0.22058823704719543, 0.24754901230335236, 0.2242647111415863, 0.27818626165390015, 0.23284313082695007, 0.2401960790157318, 0.2450980395078659, 0.2395833283662796, 0.24264705181121826, 0.27450981736183167, 0.27757352590560913, 0.23651960492134094, 0.31004902720451355, 0.24877451360225677, 0.2855392098426819, 0.2659313678741455, 0.24877451360225677, 0.3235294222831726, 0.27450981736183167, 0.27389705181121826, 0.3541666567325592, 0.280637264251709, 0.26225489377975464, 0.2953431308269501, 0.2855392098426819, 0.28982841968536377, 0.3112744987010956, 0.31617647409439087, 0.30392158031463623, 0.3186274468898773, 0.29963234066963196, 0.3651960790157318, 0.3946078419685364, 0.3296568691730499, 0.3223039209842682, 0.3137255012989044, 0.39828431606292725, 0.4105392098426819, 0.35661765933036804, 0.4325980246067047, 0.3700980246067047, 0.46875, NaN, 0.4730392098426819, 0.42524510622024536, 0.5514705777168274, 0.4325980246067047, 0.5232843160629272, 0.5723039507865906, 0.966911792755127, 0.623774528503418, 0.655637264251709, 0.6200980544090271, 0.686274528503418, 0.6127451062202454, 0.6825980544090271, 0.6127451062202454, 0.7892156839370728, 0.8210784196853638, 0.9031862616539001], [0.03186274692416191, 0.04534313827753067, 0.053921569138765335, 0.07230392098426819, 0.05147058889269829, 0.0674019604921341, 0.0882352963089943, 0.0882352963089943, 0.06862745434045792, 0.11764705926179886, 0.11825980246067047, 0.12132352590560913, 0.17892156541347504, 0.16053920984268188, 0.13725490868091583, 0.12622548639774323, 0.12193627655506134, 0.23039215803146362, 0.13970588147640228, 0.2291666716337204, 0.14460784196853638, 0.24142156541347504, 0.2083333283662796, 0.1715686321258545, 0.20098039507865906, 0.15870098769664764, 0.19362744688987732, 0.25857841968536377, 0.2401960790157318, 0.18014705181121826, 0.311887264251709, 0.2561274468898773, 0.3014705777168274, 0.28431373834609985, 0.31617647409439087, 0.28799018263816833, 0.2708333432674408, 0.27818626165390015, 0.31004902720451355, 0.3125, 0.3480392098426819, 0.37745097279548645, 0.2561274468898773, 0.3909313678741455, 0.37132352590560913, 0.36274510622024536, 0.3388480246067047, 0.4485294222831726, 0.39338234066963196, 0.32107841968536377, 0.4258578419685364, 0.5281862616539001, 0.5269607901573181, NaN, 0.5220588445663452, 0.5674019455909729, 0.6078431606292725, 0.6164215803146362, 0.6053921580314636, 0.6776960492134094, 0.7058823704719543, 0.6948529481887817, 0.6948529481887817, 0.7702205777168274, 0.7647058963775635, 0.7071078419685364, 0.7083333134651184, 0.7953431606292725, 0.8786764740943909, 0.970588207244873], [0.02450980432331562, 0.03431372717022896, 0.030637254938483238, 0.053921569138765335, 0.036764707416296005, 0.04289215803146362, 0.04411764815449715, 0.02818627469241619, 0.06372549384832382, 0.09558823704719543, 0.0784313753247261, 0.10539215803146362, 0.07720588147640228, 0.12745098769664764, 0.09436274319887161, 0.1507352888584137, 0.14338235557079315, 0.09558823704719543, 0.15931372344493866, 0.2181372493505478, 0.13112744688987732, 0.14705882966518402, 0.18382352590560913, 0.1715686321258545, 0.15318627655506134, 0.18137255311012268, 0.1875, 0.22303920984268188, 0.19362744688987732, 0.19362744688987732, 0.2450980395078659, 0.1617647111415863, 0.2034313678741455, 0.28308823704719543, 0.25367647409439087, 0.10784313827753067, 0.18995098769664764, 0.22671568393707275, 0.2928921580314636, 0.2236519604921341, 0.28308823704719543, 0.30575981736183167, 0.38725489377975464, 0.29411765933036804, 0.343137264251709, 0.19178920984268188, 0.41421568393707275, 0.38786765933036804, 0.3639705777168274, 0.43014705181121826, 0.35171568393707275, 0.5735294222831726, 0.5747548937797546, 0.4779411852359772, NaN, 0.5122548937797546, 0.5453431606292725, 0.5845588445663452, 0.5539215803146362, 0.6911764740943909, 0.6446078419685364, 0.6715686321258545, 0.6470588445663452, 0.7083333134651184, 0.6421568393707275, 0.6924019455909729, 0.6678921580314636, 0.8026960492134094, 0.8272058963775635, 0.9301470518112183], [0.013480392284691334, 0.05269607901573181, 0.03799019753932953, 0.053921569138765335, 0.04411764815449715, 0.04289215803146362, 0.06188725307583809, 0.06495098024606705, 0.05637254938483238, 0.09803921729326248, 0.06372549384832382, 0.0882352963089943, 0.11397058516740799, 0.10661764442920685, 0.0992647036910057, 0.08578431606292725, 0.125, 0.1348039209842682, 0.1348039209842682, 0.2193627506494522, 0.1458333283662796, 0.1458333283662796, 0.17401960492134094, 0.12316176295280457, 0.18382352590560913, 0.16789215803146362, 0.15196079015731812, 0.21691176295280457, 0.20955882966518402, 0.19485294818878174, 0.23161764442920685, 0.21997548639774323, 0.23713235557079315, 0.2132352888584137, 0.2450980395078659, 0.27818626165390015, 0.2769607901573181, 0.2757352888584137, 0.25306373834609985, 0.28431373834609985, 0.27818626165390015, 0.36151960492134094, 0.3492647111415863, 0.31985294818878174, 0.3075980246067047, 0.34681373834609985, 0.37132352590560913, 0.4166666567325592, 0.4264705777168274, 0.406862735748291, 0.3921568691730499, 0.44485294818878174, 0.4485294222831726, 0.4325980246067047, 0.48774510622024536, NaN, 0.561274528503418, 0.563725471496582, 0.7395833134651184, 0.530637264251709, 0.5906862616539001, 0.6507353186607361, 0.5894607901573181, 0.6047794222831726, 0.6139705777168274, 0.6629902124404907, 0.843137264251709, 0.716911792755127, 0.7450980544090271, 0.9350489974021912], [0.04289215803146362, 0.03921568766236305, 0.0416666679084301, 0.06004901975393295, 0.053921569138765335, 0.06372549384832382, 0.06985294073820114, 0.06004901975393295, 0.0882352963089943, 0.11887254565954208, 0.08700980246067047, 0.12867647409439087, 0.09436274319887161, 0.16544117033481598, 0.12377450615167618, 0.08088235557079315, 0.1145833358168602, 0.11519607901573181, 0.12377450615167618, 0.19607843458652496, 0.11764705926179886, 0.16789215803146362, 0.18872548639774323, 0.09620098024606705, 0.18995098769664764, 0.1458333283662796, 0.09068627655506134, 0.2034313678741455, 0.1507352888584137, 0.15196079015731812, 0.27328431606292725, 0.18627451360225677, 0.1985294073820114, 0.1825980395078659, 0.1973039209842682, 0.1930147111415863, 0.18933823704719543, 0.20710784196853638, 0.2083333283662796, 0.22794117033481598, 0.2401960790157318, 0.2671568691730499, 0.3014705777168274, 0.24387255311012268, 0.24387255311012268, 0.2561274468898773, 0.3235294222831726, 0.32598039507865906, 0.25, 0.37254902720451355, 0.29411765933036804, 0.5698529481887817, 0.5674019455909729, 0.3921568691730499, 0.4546568691730499, 0.4387255012989044, NaN, 0.4901960790157318, 0.46078431606292725, 0.6531862616539001, 0.5024510025978088, 0.654411792755127, 0.593137264251709, 0.6924019455909729, 0.5674019455909729, 0.6715686321258545, 0.5735294222831726, 0.7610294222831726, 0.7953431606292725, 0.8970588445663452], [0.029411764815449715, 0.03799019753932953, 0.040441177785396576, 0.05514705926179886, 0.05147058889269829, 0.06372549384832382, 0.07352941483259201, 0.06617647409439087, 0.061274509876966476, 0.09313725680112839, 0.09558823704719543, 0.08272058516740799, 0.1409313678741455, 0.10661764442920685, 0.11887254565954208, 0.16789215803146362, 0.14460784196853638, 0.18137255311012268, 0.15318627655506134, 0.16911764442920685, 0.10539215803146362, 0.16544117033481598, 0.16053920984268188, 0.17953431606292725, 0.050245098769664764, 0.17279411852359772, 0.20588235557079315, 0.1403186321258545, 0.14460784196853638, 0.18382352590560913, 0.23774509131908417, 0.09803921729326248, 0.2708333432674408, 0.25735294818878174, 0.2916666567325592, 0.23161764442920685, 0.18872548639774323, 0.23529411852359772, 0.27818626165390015, 0.27941176295280457, 0.3137255012989044, 0.2965686321258545, 0.2867647111415863, 0.35661765933036804, 0.3223039209842682, 0.2916666567325592, 0.3223039209842682, 0.3651960790157318, 0.2953431308269501, 0.3394607901573181, 0.3137255012989044, 0.4742647111415863, 0.47671568393707275, 0.38357841968536377, 0.4154411852359772, 0.4362744987010956, 0.5098039507865906, NaN, 0.46629902720451355, 0.5735294222831726, 0.47549018263816833, 0.5870097875595093, 0.530637264251709, 0.5821078419685364, 0.5257353186607361, 0.593137264251709, 0.5906862616539001, 0.6752451062202454, 0.7083333134651184, 0.9203431606292725], [0.014705882407724857, 0.03553921729326248, 0.032475490123033524, 0.050245098769664764, 0.046568628400564194, 0.0416666679084301, 0.05514705926179886, 0.057598039507865906, 0.04901960864663124, 0.07230392098426819, 0.0625, 0.07904411852359772, 0.09436274319887161, 0.09803921729326248, 0.09191176295280457, 0.07107843458652496, 0.10661764442920685, 0.125, 0.11397058516740799, 0.20465686917304993, 0.11151960492134094, 0.12745098769664764, 0.13848039507865906, 0.10171568393707275, 0.15012255311012268, 0.13909313082695007, 0.1200980395078659, 0.2083333283662796, 0.17401960492134094, 0.15441176295280457, 0.22242647409439087, 0.17953431606292725, 0.22549019753932953, 0.18504901230335236, 0.22732843458652496, 0.2291666716337204, 0.2291666716337204, 0.22303920984268188, 0.2236519604921341, 0.22794117033481598, 0.25367647409439087, 0.3272058963775635, 0.31617647409439087, 0.2928921580314636, 0.2720588147640228, 0.29901960492134094, 0.33578431606292725, 0.3737744987010956, 0.3492647111415863, 0.37132352590560913, 0.33088234066963196, 0.4227941036224365, 0.4276960790157318, 0.3946078419685364, 0.44607841968536377, 0.2604166567325592, 0.5392156839370728, 0.5337010025978088, NaN, 0.5085784196853638, 0.5465686321258545, 0.6335784196853638, 0.5502451062202454, 0.5674019455909729, 0.5845588445663452, 0.6446078419685364, 0.7469362616539001, 0.6985294222831726, 0.716911792755127, 0.9142156839370728], [0.09558823704719543, 0.11029411852359772, 0.12622548639774323, 0.10784313827753067, 0.13112744688987732, 0.1360294073820114, 0.14338235557079315, 0.12867647409439087, 0.16299019753932953, 0.15441176295280457, 0.16789215803146362, 0.16421568393707275, 0.17034313082695007, 0.1764705926179886, 0.17892156541347504, 0.1715686321258545, 0.18872548639774323, 0.18995098769664764, 0.1875, 0.22242647409439087, 0.19362744688987732, 0.16544117033481598, 0.21446079015731812, 0.18627451360225677, 0.2181372493505478, 0.2083333283662796, 0.18504901230335236, 0.26225489377975464, 0.21200980246067047, 0.21078431606292725, 0.20465686917304993, 0.20955882966518402, 0.1875, 0.21446079015731812, 0.18995098769664764, 0.19975490868091583, 0.2401960790157318, 0.22058823704719543, 0.2291666716337204, 0.22181372344493866, 0.20465686917304993, 0.25735294818878174, 0.26776960492134094, 0.22181372344493866, 0.25735294818878174, 0.22181372344493866, 0.2720588147640228, 0.27818626165390015, 0.2757352888584137, 0.30453431606292725, 0.281862735748291, 0.046568628400564194, 0.033088237047195435, 0.3223039209842682, 0.30882352590560913, 0.469362735748291, 0.34681373834609985, 0.4264705777168274, 0.49142158031463623, NaN, 0.46568626165390015, 0.5575980544090271, 0.4522058963775635, 0.5526960492134094, 0.43382352590560913, 0.5674019455909729, 0.531862735748291, 0.6617646813392639, 0.6911764740943909, 0.8725489974021912], [0.013480392284691334, 0.02450980432331562, 0.0416666679084301, 0.0416666679084301, 0.03921568766236305, 0.04411764815449715, 0.061274509876966476, 0.057598039507865906, 0.03921568766236305, 0.05453431233763695, 0.06985294073820114, 0.04901960864663124, 0.10784313827753067, 0.07720588147640228, 0.11029411852359772, 0.08210784196853638, 0.09803921729326248, 0.1556372493505478, 0.11151960492134094, 0.18627451360225677, 0.09068627655506134, 0.13848039507865906, 0.11519607901573181, 0.10661764442920685, 0.06495098024606705, 0.12132352590560913, 0.13357843458652496, 0.1568627506494522, 0.15318627655506134, 0.12377450615167618, 0.20955882966518402, 0.13112744688987732, 0.2083333283662796, 0.21200980246067047, 0.24142156541347504, 0.23529411852359772, 0.18382352590560913, 0.1868872493505478, 0.22303920984268188, 0.20465686917304993, 0.25735294818878174, 0.2549019753932953, 0.08088235557079315, 0.2720588147640228, 0.31495097279548645, 0.281862735748291, 0.14950980246067047, 0.29901960492134094, 0.24877451360225677, 0.2181372493505478, 0.2696078419685364, 0.37745097279548645, 0.3762255012989044, 0.29411765933036804, 0.3553921580314636, 0.40931373834609985, 0.49754902720451355, 0.5245097875595093, 0.4534313678741455, 0.5343137383460999, NaN, 0.595588207244873, 0.5575980544090271, 0.623774528503418, 0.5453431606292725, 0.6090686321258545, 0.5759803652763367, 0.7009803652763367, 0.7573529481887817, 0.9718137383460999], [0.0416666679084301, 0.0674019604921341, 0.07230392098426819, 0.05882352963089943, 0.07475490123033524, 0.07965686172246933, 0.09068627655506134, 0.0784313753247261, 0.09436274319887161, 0.12745098769664764, 0.09742647409439087, 0.15931372344493866, 0.09681372344493866, 0.17340686917304993, 0.1348039209842682, 0.11887254565954208, 0.1409313678741455, 0.11887254565954208, 0.14460784196853638, 0.14338235557079315, 0.18627451360225677, 0.17401960492134094, 0.22181372344493866, 0.13848039507865906, 0.13235294818878174, 0.14950980246067047, 0.16053920984268188, 0.23284313082695007, 0.1776960790157318, 0.15808823704719543, 0.17524509131908417, 0.13725490868091583, 0.17279411852359772, 0.21446079015731812, 0.13848039507865906, 0.1881127506494522, 0.23835784196853638, 0.20220588147640228, 0.22671568393707275, 0.20465686917304993, 0.16544117033481598, 0.17279411852359772, 0.25796568393707275, 0.18014705181121826, 0.2340686321258545, 0.25735294818878174, 0.2720588147640228, 0.23529411852359772, 0.27818626165390015, 0.28431373834609985, 0.3026960790157318, 0.33700981736183167, 0.344362735748291, 0.30514705181121826, 0.3284313678741455, 0.3492647111415863, 0.34558823704719543, 0.41299018263816833, 0.36642158031463623, 0.4424019753932953, 0.40441176295280457, NaN, 0.36642158031463623, 0.44485294818878174, 0.406862735748291, 0.6672794222831726, 0.46078431606292725, 0.8125, 0.5490196347236633, 0.7585784196853638], [0.023284314200282097, 0.033088237047195435, 0.03431372717022896, 0.050245098769664764, 0.04901960864663124, 0.04779411852359772, 0.05514705926179886, 0.05147058889269829, 0.04779411852359772, 0.061274509876966476, 0.05269607901573181, 0.07965686172246933, 0.08210784196853638, 0.09436274319887161, 0.06924019753932953, 0.11764705926179886, 0.12254901975393295, 0.12254901975393295, 0.1348039209842682, 0.16053920984268188, 0.10294117778539658, 0.10539215803146362, 0.12377450615167618, 0.13235294818878174, 0.09436274319887161, 0.15441176295280457, 0.12377450615167618, 0.17892156541347504, 0.1354166716337204, 0.16421568393707275, 0.18382352590560913, 0.11397058516740799, 0.17524509131908417, 0.18382352590560913, 0.16789215803146362, 0.14154411852359772, 0.15318627655506134, 0.17524509131908417, 0.20588235557079315, 0.12745098769664764, 0.20098039507865906, 0.04901960864663124, 0.25, 0.21200980246067047, 0.25245097279548645, 0.2181372493505478, 0.2708333432674408, 0.1666666716337204, 0.281862735748291, 0.2977941036224365, 0.2708333432674408, 0.3762255012989044, 0.3799019753932953, 0.30514705181121826, 0.3529411852359772, 0.4105392098426819, 0.406862735748291, 0.469362735748291, 0.44975489377975464, 0.5477941036224365, 0.4424019753932953, 0.6335784196853638, NaN, 0.5735294222831726, 0.47732841968536377, 0.6531862616539001, 0.5882353186607361, 0.7475489974021912, 0.7450980544090271, 0.9129902124404907], [0.030637254938483238, 0.05147058889269829, 0.04901960864663124, 0.04534313827753067, 0.05147058889269829, 0.06985294073820114, 0.08210784196853638, 0.09191176295280457, 0.07475490123033524, 0.07475490123033524, 0.09558823704719543, 0.08272058516740799, 0.1090686246752739, 0.08946078270673752, 0.11029411852359772, 0.10171568393707275, 0.10049019753932953, 0.13848039507865906, 0.1090686246752739, 0.16299019753932953, 0.12254901975393295, 0.11764705926179886, 0.125, 0.12377450615167618, 0.09375, 0.13357843458652496, 0.12745098769664764, 0.1764705926179886, 0.13725490868091583, 0.13664215803146362, 0.20955882966518402, 0.10661764442920685, 0.18137255311012268, 0.20465686917304993, 0.17034313082695007, 0.18014705181121826, 0.18504901230335236, 0.1348039209842682, 0.22303920984268188, 0.1715686321258545, 0.1973039209842682, 0.21446079015731812, 0.13112744688987732, 0.18137255311012268, 0.2665441036224365, 0.22242647409439087, 0.1617647111415863, 0.25735294818878174, 0.23039215803146362, 0.1985294073820114, 0.24754901230335236, 0.30514705181121826, 0.3137255012989044, 0.2297794073820114, 0.2916666567325592, 0.3952205777168274, 0.3075980246067047, 0.4178921580314636, 0.4325980246067047, 0.4473039209842682, 0.3762255012989044, 0.5551470518112183, 0.4264705777168274, NaN, 0.4117647111415863, 0.5827205777168274, 0.5392156839370728, 0.6985294222831726, 0.8866421580314636, 0.9742646813392639], [0.014705882407724857, 0.014705882407724857, 0.026960784569382668, 0.03186274692416191, 0.022058824077248573, 0.03431372717022896, 0.04411764815449715, 0.04534313827753067, 0.03431372717022896, 0.03799019753932953, 0.05882352963089943, 0.04411764815449715, 0.07965686172246933, 0.06985294073820114, 0.07965686172246933, 0.05147058889269829, 0.057598039507865906, 0.11151960492134094, 0.06617647409439087, 0.13970588147640228, 0.07230392098426819, 0.11029411852359772, 0.12377450615167618, 0.07475490123033524, 0.05882352963089943, 0.06495098024606705, 0.09803921729326248, 0.14460784196853638, 0.12254901975393295, 0.07965686172246933, 0.20098039507865906, 0.1090686246752739, 0.16789215803146362, 0.17892156541347504, 0.20465686917304993, 0.2034313678741455, 0.15318627655506134, 0.14215686917304993, 0.20955882966518402, 0.1985294073820114, 0.23161764442920685, 0.2696078419685364, 0.11397058516740799, 0.25857841968536377, 0.24264705181121826, 0.2549019753932953, 0.1513480395078659, 0.3112744987010956, 0.18995098769664764, 0.1041666641831398, 0.22732843458652496, 0.3897058963775635, 0.38725489377975464, 0.23529411852359772, 0.3578431308269501, 0.3860294222831726, 0.4325980246067047, 0.4742647111415863, 0.4154411852359772, 0.5661764740943909, 0.4546568691730499, 0.593137264251709, 0.5226715803146362, 0.5882353186607361, NaN, 0.6151960492134094, 0.5674019455909729, 0.6960784196853638, 0.748774528503418, 0.9460784196853638], [0.036764707416296005, 0.061274509876966476, 0.05637254938483238, 0.05269607901573181, 0.06004901975393295, 0.06862745434045792, 0.07598039507865906, 0.07352941483259201, 0.09313725680112839, 0.11887254565954208, 0.08026960492134094, 0.1458333283662796, 0.08517156541347504, 0.15808823704719543, 0.12806372344493866, 0.1041666641831398, 0.12867647409439087, 0.11029411852359772, 0.13419117033481598, 0.12867647409439087, 0.16053920984268188, 0.15441176295280457, 0.20098039507865906, 0.1194852963089943, 0.1200980395078659, 0.1409313678741455, 0.14460784196853638, 0.22181372344493866, 0.1556372493505478, 0.14215686917304993, 0.16911764442920685, 0.1200980395078659, 0.1611519604921341, 0.19975490868091583, 0.125, 0.16299019753932953, 0.2083333283662796, 0.1819852888584137, 0.22058823704719543, 0.17401960492134094, 0.14950980246067047, 0.15196079015731812, 0.24203431606292725, 0.15502451360225677, 0.2242647111415863, 0.2395833283662796, 0.26225489377975464, 0.21262255311012268, 0.26348039507865906, 0.27450981736183167, 0.28799018263816833, 0.31495097279548645, 0.3174019753932953, 0.2928921580314636, 0.3075980246067047, 0.33700981736183167, 0.3284313678741455, 0.406862735748291, 0.3553921580314636, 0.4325980246067047, 0.3909313678741455, 0.3327205777168274, 0.34681373834609985, 0.4172794222831726, 0.3848039209842682, NaN, 0.4546568691730499, 0.7438725233078003, 0.5367646813392639, 0.7524510025978088], [0.01225490216165781, 0.022058824077248573, 0.025735294446349144, 0.02818627469241619, 0.026960784569382668, 0.025735294446349144, 0.025735294446349144, 0.030637254938483238, 0.02450980432331562, 0.030637254938483238, 0.040441177785396576, 0.04289215803146362, 0.04411764815449715, 0.04411764815449715, 0.03799019753932953, 0.040441177785396576, 0.0674019604921341, 0.0625, 0.06372549384832382, 0.12867647409439087, 0.06862745434045792, 0.0674019604921341, 0.0784313753247261, 0.04534313827753067, 0.0784313753247261, 0.07720588147640228, 0.07230392098426819, 0.14215686917304993, 0.1041666641831398, 0.0784313753247261, 0.1617647111415863, 0.09191176295280457, 0.11764705926179886, 0.1299019604921341, 0.11519607901573181, 0.11764705926179886, 0.1348039209842682, 0.1360294073820114, 0.13664215803146362, 0.12377450615167618, 0.1360294073820114, 0.18137255311012268, 0.2193627506494522, 0.1617647111415863, 0.18014705181121826, 0.1764705926179886, 0.23161764442920685, 0.24387255311012268, 0.22549019753932953, 0.27328431606292725, 0.19485294818878174, 0.3860294222831726, 0.38725489377975464, 0.2916666567325592, 0.3321078419685364, 0.1568627506494522, 0.4264705777168274, 0.40931373834609985, 0.25306373834609985, 0.468137264251709, 0.42401960492134094, 0.5392156839370728, 0.4117647111415863, 0.46078431606292725, 0.4325980246067047, 0.5453431606292725, NaN, 0.6158088445663452, 0.6078431606292725, 0.8553921580314636], [0.02450980432331562, 0.036764707416296005, 0.039828430861234665, 0.03370098024606705, 0.04289215803146362, 0.05085784196853638, 0.04840686172246933, 0.03799019753932953, 0.07965686172246933, 0.05637254938483238, 0.05882352963089943, 0.09558823704719543, 0.06495098024606705, 0.10784313827753067, 0.10049019753932953, 0.06372549384832382, 0.1041666641831398, 0.08210784196853638, 0.1090686246752739, 0.1090686246752739, 0.12132352590560913, 0.08272058516740799, 0.12377450615167618, 0.07720588147640228, 0.09620098024606705, 0.11029411852359772, 0.09436274319887161, 0.18627451360225677, 0.11151960492134094, 0.11519607901573181, 0.13725490868091583, 0.08210784196853638, 0.11764705926179886, 0.1299019604921341, 0.0992647036910057, 0.10171568393707275, 0.1507352888584137, 0.10171568393707275, 0.14338235557079315, 0.11274509876966476, 0.11764705926179886, 0.10294117778539658, 0.16911764442920685, 0.10049019753932953, 0.14460784196853638, 0.1299019604921341, 0.1764705926179886, 0.1348039209842682, 0.1776960790157318, 0.19424019753932953, 0.18627451360225677, 0.21078431606292725, 0.21078431606292725, 0.20465686917304993, 0.1973039209842682, 0.28308823704719543, 0.2389705926179886, 0.32475489377975464, 0.3014705777168274, 0.3382352888584137, 0.29901960492134094, 0.1875, 0.25245097279548645, 0.3014705777168274, 0.30392158031463623, 0.2561274468898773, 0.3841911852359772, NaN, 0.46568626165390015, 0.7071078419685364], [0.02818627469241619, 0.03431372717022896, 0.03553921729326248, 0.03553921729326248, 0.03921568766236305, 0.05147058889269829, 0.04901960864663124, 0.05269607901573181, 0.053921569138765335, 0.04289215803146362, 0.0625, 0.05637254938483238, 0.07230392098426819, 0.06495098024606705, 0.0833333358168602, 0.06556372344493866, 0.07352941483259201, 0.09191176295280457, 0.06985294073820114, 0.12377450615167618, 0.09129901975393295, 0.046568628400564194, 0.09313725680112839, 0.07352941483259201, 0.07107843458652496, 0.08517156541347504, 0.0674019604921341, 0.14828431606292725, 0.09436274319887161, 0.08946078270673752, 0.11151960492134094, 0.07230392098426819, 0.0992647036910057, 0.11397058516740799, 0.11213235557079315, 0.10661764442920685, 0.12683823704719543, 0.09068627655506134, 0.1360294073820114, 0.09313725680112839, 0.11397058516740799, 0.13970588147640228, 0.08700980246067047, 0.11029411852359772, 0.1360294073820114, 0.14338235557079315, 0.08210784196853638, 0.16789215803146362, 0.1556372493505478, 0.11397058516740799, 0.1507352888584137, 0.17401960492134094, 0.17892156541347504, 0.12132352590560913, 0.17279411852359772, 0.2549019753932953, 0.20465686917304993, 0.2916666567325592, 0.28308823704719543, 0.30882352590560913, 0.24264705181121826, 0.45098039507865906, 0.2549019753932953, 0.11335784196853638, 0.2512255012989044, 0.4632352888584137, 0.3921568691730499, 0.5343137383460999, NaN, 0.9375], [0.006127451080828905, 0.0049019609577953815, 0.006127451080828905, 0.0049019609577953815, 0.007352941203862429, 0.00857843179255724, 0.013480392284691334, 0.014705882407724857, 0.006127451080828905, 0.009803921915590763, 0.009803921915590763, 0.006127451080828905, 0.02450980432331562, 0.00857843179255724, 0.01715686358511448, 0.013480392284691334, 0.01225490216165781, 0.03186274692416191, 0.013480392284691334, 0.03186274692416191, 0.01715686358511448, 0.006127451080828905, 0.007352941203862429, 0.018382353708148003, 0.011029412038624287, 0.02083333395421505, 0.015931373462080956, 0.0416666679084301, 0.022058824077248573, 0.023284314200282097, 0.014705882407724857, 0.013480392284691334, 0.04779411852359772, 0.026348039507865906, 0.04779411852359772, 0.040441177785396576, 0.03799019753932953, 0.019607843831181526, 0.03799019753932953, 0.02083333395421505, 0.053921569138765335, 0.04534313827753067, 0.00857843179255724, 0.05147058889269829, 0.03921568766236305, 0.05269607901573181, 0.009803921915590763, 0.053921569138765335, 0.05514705926179886, 0.02818627469241619, 0.04473039135336876, 0.09681372344493866, 0.09681372344493866, 0.029411764815449715, 0.06985294073820114, 0.06495098024606705, 0.10294117778539658, 0.07965686172246933, 0.08578431606292725, 0.12745098769664764, 0.02818627469241619, 0.24142156541347504, 0.08700980246067047, 0.025735294446349144, 0.053921569138765335, 0.24754901230335236, 0.14460784196853638, 0.2928921580314636, 0.0625, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9224033779},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8560226618},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.853340794},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.833342211},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8255896549},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8180857495},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7878658718},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7785947703},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7414393269},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7230747368},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7178264437},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7095055417},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6769057268},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6700056846},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6691620473},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.645842213},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6313672201},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6195030576},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6160041878},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.615169439},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6137929815},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6133400791},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.6108446989},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6008010082},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5890700477},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5822321705},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5781560805},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5651019476},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.56345908},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5587080845},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5541524572},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.551230816},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5205580439},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5129653309},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5102035389},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5060208862},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5014208592},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4869458646},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4857470123},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4680484501},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4679063685},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.449861466},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.441487284},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.437135907},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4355818392},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4324737143},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.414251208},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3882583833},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3835873133},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3751420845},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3704266143},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3692366429},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3675848969},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3413523013},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3066567213},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3052980257},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2795538525},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2788256617},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2718723358},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2661356197},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2480107987},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2389350676},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2373366011},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2272929088},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2248596905},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2188565638},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1902973164},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1557082974},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1469700202},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0452543339}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_all/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9261755675eb34f63619a5264a42a2779862da0b
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_all/winrate_matrix.csv
@@ -0,0 +1,71 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (default),RealMLP (tuned + ensembled),TabDPT-Turbo (default),"AutoGluon 1.4 (best, 4h)",TabDPT (tuned + ensembled),TabM (tuned + ensembled),TabDPT (tuned),RealMLP (tuned),LightGBM (tuned + ensembled),CatBoost (tuned + ensembled),TabM (tuned),CatBoost (tuned),LimiX (default),iLTM (tuned + ensembled),TabSwift (default),TabDPT (default),LightGBM (tuned),xRFM (tuned + ensembled),XGBoost (tuned + ensembled),CatBoost (default),ModernNCA (tuned + ensembled),ModernNCA (tuned),XGBoost (tuned),SAP-RPT-OSS (default),xRFM (tuned),TabM (default),ChimeraBoost (tuned + ensembled),EBM (tuned + ensembled),TorchMLP (tuned + ensembled),iLTM (tuned),ModernNCA (default),ChimeraBoost (tuned),RealMLP (default),EBM (tuned),FastaiMLP (tuned + ensembled),ExtraTrees (tuned + ensembled),EBM (default),ChimeraBoost (default),TorchMLP (tuned),ExtraTrees (tuned),FastaiMLP (tuned),XGBoost (default),RandomForest (tuned + ensembled),LightGBM (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),RandomForest (tuned),TorchMLP (default),PerpetualBooster (tuned + ensembled),iLTM (default),xRFM (default),PerpetualBooster (tuned),TabSTAR (default),ExtraTrees (default),Linear (tuned + ensembled),FastaiMLP (default),KNN (tuned + ensembled),RandomForest (default),Linear (tuned),PerpetualBooster (default),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.7708333,0.7512255,0.8088235,0.81740195,0.7873775,0.7977941,0.8357843,0.86519605,0.87990195,0.8897059,0.88235295,0.90318626,0.8958333,0.875,0.91544116,0.91544116,0.9264706,0.92769605,0.90625,0.92401963,0.8860294,0.8995098,0.92769605,0.92034316,0.935049,0.9227941,0.94240195,0.9313725,0.9411765,0.8786765,0.91544116,0.93259805,0.935049,0.9264706,0.95955884,0.94240195,0.9387255,0.939951,0.94730395,0.9375,0.9264706,0.95343137,0.935049,0.95465684,0.96936274,0.9485294,0.9497549,0.97181374,0.96936274,0.97181374,0.872549,0.8713235,0.96813726,0.9754902,0.98651963,0.95710784,0.9705882,0.9852941,0.9044118,0.98651963,0.9583333,0.9767157,0.96936274,0.9852941,0.9632353,0.9877451,0.9754902,0.97181374,0.9938725
+TabPFN-3 (default),0.22916667,,0.57965684,0.55514705,0.63480395,0.6188725,0.67034316,0.70710784,0.74509805,0.7745098,0.7512255,0.7585784,0.7977941,0.7916667,0.7794118,0.82598037,0.83455884,0.8370098,0.84068626,0.78431374,0.8639706,0.88235295,0.8272059,0.86151963,0.85294116,0.84865195,0.8541667,0.8786765,0.86519605,0.85906863,0.79656863,0.8639706,0.87009805,0.8860294,0.85906863,0.9105392,0.92585784,0.9050245,0.8860294,0.9252451,0.8762255,0.88419116,0.935049,0.89644605,0.8958333,0.9387255,0.9362745,0.9209559,0.95098037,0.9387255,0.95955884,0.85110295,0.8492647,0.95465684,0.96568626,0.94730395,0.9607843,0.9620098,0.9644608,0.8897059,0.9754902,0.93259805,0.9669118,0.9485294,0.9852941,0.9387255,0.97794116,0.9632353,0.96568626,0.99509805
+"AutoGluon 1.5 (extreme, 4h)",0.24877451,0.42034313,,0.47916666,0.62009805,0.5539216,0.60906863,0.63480395,0.7775735,0.67156863,0.80514705,0.7377451,0.82598037,0.7610294,0.8186275,0.86151963,0.8639706,0.83884805,0.8670343,0.7769608,0.89093137,0.8357843,0.76593137,0.8762255,0.875,0.88848037,0.8792892,0.8860294,0.89705884,0.89460784,0.77205884,0.8694853,0.8835784,0.9166667,0.8897059,0.9252451,0.91544116,0.8958333,0.92401963,0.9117647,0.8982843,0.8982843,0.92401963,0.8995098,0.93014705,0.94485295,0.9166667,0.9166667,0.9620098,0.9436275,0.9607843,0.8357843,0.8357843,0.9460784,0.96936274,0.9620098,0.9583333,0.95955884,0.9675245,0.8737745,0.9583333,0.92769605,0.96568626,0.95098037,0.9730392,0.9436275,0.9742647,0.9601716,0.9644608,0.9938725
+TabICLv2 (default),0.19117647,0.44485295,0.5208333,,0.5870098,0.56740195,0.60906863,0.6525735,0.71568626,0.7316176,0.74509805,0.7665441,0.7867647,0.7683824,0.752451,0.8088235,0.8112745,0.8088235,0.814951,0.78615195,0.84313726,0.8370098,0.79289216,0.8382353,0.8247549,0.8247549,0.8382353,0.84313726,0.8480392,0.82965684,0.78063726,0.8204657,0.8566176,0.8566176,0.8455882,0.87990195,0.875,0.88419116,0.8664216,0.9019608,0.8639706,0.87009805,0.8976716,0.8897059,0.8762255,0.9080882,0.8995098,0.90686274,0.91789216,0.9252451,0.9411765,0.84436274,0.8394608,0.92769605,0.9460784,0.9460784,0.939951,0.94485295,0.9497549,0.89215684,0.9583333,0.9411765,0.9497549,0.95465684,0.96813726,0.94730395,0.97181374,0.966299,0.9644608,0.99509805
+TabPFN-2.6 (default),0.18259804,0.36519608,0.37990198,0.41299018,,0.48284313,0.5435049,0.6139706,0.6911765,0.6433824,0.70955884,0.6727941,0.7757353,0.7022059,0.76348037,0.80759805,0.8137255,0.814951,0.814951,0.7512255,0.8419118,0.79901963,0.72181374,0.85294116,0.8235294,0.8492647,0.8517157,0.85906863,0.8602941,0.8713235,0.7916667,0.84313726,0.8682598,0.89460784,0.86151963,0.8958333,0.90686274,0.8982843,0.90318626,0.90931374,0.8915441,0.872549,0.9080882,0.8872549,0.9105392,0.935049,0.9050245,0.90931374,0.9620098,0.9313725,0.9583333,0.8370098,0.84068626,0.9485294,0.9632353,0.9558824,0.9460784,0.9485294,0.95343137,0.8688725,0.9607843,0.9252451,0.95098037,0.9485294,0.97794116,0.939951,0.9730392,0.95710784,0.9607843,0.99264705
+RealTabPFN-2.5 (tuned + ensembled),0.21262255,0.38112745,0.44607842,0.43259802,0.51715684,,0.6666667,0.6642157,0.6936275,0.6237745,0.7352941,0.6642157,0.7610294,0.7022059,0.747549,0.79289216,0.7977941,0.79901963,0.8088235,0.75980395,0.7922794,0.79656863,0.6985294,0.8204657,0.81740195,0.82843137,0.82843137,0.8088235,0.8517157,0.8425245,0.78125,0.8272059,0.85294116,0.84865195,0.8566176,0.8872549,0.8737745,0.8627451,0.8664216,0.88235295,0.87806374,0.8682598,0.877451,0.87990195,0.8872549,0.92034316,0.87990195,0.90931374,0.93014705,0.91789216,0.9485294,0.82965684,0.82965684,0.93259805,0.95710784,0.95710784,0.9362745,0.9362745,0.9583333,0.8639706,0.9558824,0.92034316,0.9522059,0.93014705,0.96568626,0.9313725,0.9742647,0.94914216,0.9485294,0.9914216
+RealTabPFN-2.5 (tuned),0.20220588,0.32965687,0.39093137,0.39093137,0.4564951,0.33333334,,0.57598037,0.64215684,0.5894608,0.6752451,0.6213235,0.7267157,0.6623775,0.70710784,0.74509805,0.7585784,0.7800245,0.7647059,0.7242647,0.75,0.76409316,0.67401963,0.79534316,0.79105395,0.7941176,0.7855392,0.77389705,0.79901963,0.8063725,0.752451,0.810049,0.8308824,0.80943626,0.8370098,0.8639706,0.8541667,0.8394608,0.83455884,0.8664216,0.85294116,0.8504902,0.8670343,0.86519605,0.8639706,0.9044118,0.875,0.90318626,0.9117647,0.8897059,0.9375,0.80514705,0.810049,0.9117647,0.9558824,0.93811274,0.93014705,0.9264706,0.94485295,0.8566176,0.9387255,0.90931374,0.94485295,0.91789216,0.9558824,0.92401963,0.9742647,0.95159316,0.95098037,0.98651963
+RealTabPFN-2.5 (default),0.16421568,0.29289216,0.36519608,0.34742647,0.38602942,0.33578432,0.4240196,,0.5894608,0.5502451,0.65931374,0.57965684,0.7132353,0.62806374,0.6764706,0.73039216,0.747549,0.7585784,0.7647059,0.72794116,0.7267157,0.7708333,0.6691176,0.7855392,0.78431374,0.7745098,0.79044116,0.73651963,0.7830882,0.8039216,0.76348037,0.8137255,0.8308824,0.8161765,0.81985295,0.8639706,0.8584559,0.84436274,0.8272059,0.8602941,0.8480392,0.8737745,0.87009805,0.8639706,0.8664216,0.9044118,0.88235295,0.90563726,0.9007353,0.8762255,0.9338235,0.8308824,0.83026963,0.9117647,0.97181374,0.935049,0.939951,0.9338235,0.94240195,0.8713235,0.94240195,0.92156863,0.9485294,0.9080882,0.95465684,0.9264706,0.96936274,0.9620098,0.94730395,0.9852941
+RealMLP (tuned + ensembled),0.13480392,0.25490198,0.22242647,0.28431374,0.30882353,0.30637255,0.35784313,0.4105392,,0.49754903,0.53125,0.4908088,0.61519605,0.5349265,0.6942402,0.70343137,0.71936274,0.68259805,0.7254902,0.59068626,0.74019605,0.5955882,0.5882353,0.7579657,0.76715684,0.79534316,0.7438725,0.70343137,0.7291667,0.8137255,0.67156863,0.77205884,0.814951,0.76960784,0.7977941,0.78186274,0.82598037,0.79656863,0.79105395,0.90134805,0.81740195,0.8474265,0.8480392,0.83884805,0.814951,0.83210784,0.85723037,0.86090684,0.9105392,0.9117647,0.92769605,0.78186274,0.7855392,0.9313725,0.9362745,0.9436275,0.9117647,0.9387255,0.95098037,0.8370098,0.9607843,0.90563726,0.9522059,0.9252451,0.96568626,0.90686274,0.9754902,0.92034316,0.9460784,0.9938725
+TabDPT-Turbo (default),0.12009804,0.2254902,0.32843137,0.26838234,0.35661766,0.3762255,0.4105392,0.4497549,0.502451,,0.5563725,0.5147059,0.57414216,0.5808824,0.58394605,0.6286765,0.66544116,0.63235295,0.6642157,0.65931374,0.7022059,0.7077206,0.70281863,0.66544116,0.6936275,0.6875,0.6936275,0.685049,0.70465684,0.7022059,0.6629902,0.73651963,0.7169118,0.74509805,0.7120098,0.7757353,0.79840684,0.79901963,0.76715684,0.80514705,0.74693626,0.8137255,0.84865195,0.79044116,0.7916667,0.82843137,0.8713235,0.8627451,0.85539216,0.8713235,0.88235295,0.7647059,0.76715684,0.88235295,0.9044118,0.9019608,0.8811275,0.90686274,0.92769605,0.8455882,0.9454657,0.872549,0.9387255,0.9252451,0.9620098,0.8811275,0.96936274,0.9436275,0.95710784,0.99019605
+"AutoGluon 1.4 (best, 4h)",0.11029412,0.24877451,0.19485295,0.25490198,0.2904412,0.2647059,0.3247549,0.34068626,0.46875,0.44362745,,0.44362745,0.56740195,0.47549018,0.5882353,0.7205882,0.6862745,0.6372549,0.70465684,0.54534316,0.70098037,0.5814951,0.5563725,0.75980395,0.6917892,0.76960784,0.75,0.6887255,0.6997549,0.76960784,0.61764705,0.7377451,0.7463235,0.79718137,0.7732843,0.8186275,0.80759805,0.7855392,0.814951,0.7977941,0.79289216,0.8247549,0.7830882,0.82843137,0.8504902,0.85784316,0.8063725,0.8627451,0.9105392,0.8566176,0.92401963,0.7769608,0.7794118,0.8817402,0.92156863,0.9362745,0.9129902,0.9044118,0.9375,0.83210784,0.93014705,0.9025735,0.94730395,0.9044118,0.9411765,0.91973037,0.95955884,0.9411765,0.9375,0.99019605
+TabDPT (tuned + ensembled),0.11764706,0.24142157,0.2622549,0.23345588,0.3272059,0.33578432,0.37867647,0.42034313,0.50919116,0.4852941,0.5563725,,0.5729167,0.65931374,0.5827206,0.63235295,0.6372549,0.62438726,0.64705884,0.59865195,0.69730395,0.6439951,0.70465684,0.6617647,0.69914216,0.67340684,0.6862745,0.6936275,0.6960784,0.70343137,0.6605392,0.721201,0.7101716,0.73039216,0.7077206,0.74938726,0.7892157,0.778799,0.75,0.77818626,0.73223037,0.77818626,0.81740195,0.7487745,0.7879902,0.810049,0.8370098,0.8247549,0.8419118,0.87009805,0.877451,0.7512255,0.752451,0.8786765,0.89460784,0.9117647,0.8713235,0.9172794,0.9209559,0.8357843,0.95098037,0.84068626,0.92034316,0.9172794,0.9558824,0.8541667,0.95710784,0.9044118,0.9436275,0.9938725
+TabM (tuned + ensembled),0.09681372,0.20220588,0.1740196,0.21323529,0.22426471,0.2389706,0.27328432,0.2867647,0.38480392,0.42585784,0.43259802,0.42708334,,0.46936274,0.49754903,0.5919118,0.6017157,0.71936274,0.6188725,0.50980395,0.60110295,0.53431374,0.54044116,0.65134805,0.59681374,0.6727941,0.6838235,0.6286765,0.64031863,0.7120098,0.6237745,0.625,0.86090684,0.72181374,0.73284316,0.73406863,0.7267157,0.7267157,0.7585784,0.8229167,0.7769608,0.7757353,0.74264705,0.78615195,0.80759805,0.8143382,0.7487745,0.8308824,0.85784316,0.7867647,0.8829657,0.7732843,0.7757353,0.8210784,0.9227941,0.8860294,0.90563726,0.85906863,0.90563726,0.82965684,0.89215684,0.90318626,0.91789216,0.89093137,0.92034316,0.9148284,0.9558824,0.935049,0.92769605,0.9754902
+TabDPT (tuned),0.104166664,0.20833333,0.2389706,0.23161764,0.2977941,0.2977941,0.33762255,0.37193626,0.46507353,0.41911766,0.5245098,0.34068626,0.53063726,,0.5257353,0.5857843,0.5955882,0.5870098,0.6060049,0.57598037,0.6311275,0.5882353,0.6458333,0.622549,0.6544118,0.6439951,0.64215684,0.64644605,0.64705884,0.6544118,0.6139706,0.6691176,0.67034316,0.6960784,0.6666667,0.70955884,0.73406863,0.7242647,0.70955884,0.73406863,0.7058824,0.73651963,0.7732843,0.7242647,0.747549,0.76348037,0.7916667,0.79534316,0.7855392,0.8112745,0.83210784,0.7224265,0.72181374,0.8394608,0.872549,0.8933824,0.83455884,0.8933824,0.9019608,0.8235294,0.9227941,0.82659316,0.90563726,0.9105392,0.93014705,0.8419118,0.9558824,0.89215684,0.935049,0.9914216
+RealMLP (tuned),0.125,0.22058824,0.18137255,0.24754901,0.2365196,0.25245097,0.29289216,0.32352942,0.30575982,0.41605392,0.4117647,0.41727942,0.502451,0.4742647,,0.5833333,0.5882353,0.5857843,0.59681374,0.53125,0.59375,0.5183824,0.5355392,0.64215684,0.62009805,0.6764706,0.64215684,0.60784316,0.64215684,0.7040441,0.60784316,0.67218137,0.7316176,0.6666667,0.7205882,0.73406863,0.73039216,0.7414216,0.70343137,0.80269605,0.7585784,0.7732843,0.7622549,0.7585784,0.75,0.80514705,0.78431374,0.82965684,0.8455882,0.84436274,0.8566176,0.76593137,0.76715684,0.8627451,0.90563726,0.9007353,0.8762255,0.8811275,0.9080882,0.8210784,0.8897059,0.86519605,0.9307598,0.8897059,0.92034316,0.87193626,0.9620098,0.8995098,0.9166667,0.98284316
+LightGBM (tuned + ensembled),0.08455882,0.1740196,0.1384804,0.19117647,0.19240196,0.20710784,0.25490198,0.26960784,0.29656863,0.37132353,0.27941176,0.36764705,0.40808824,0.41421568,0.41666666,,0.5183824,0.4944853,0.5520833,0.44485295,0.5275735,0.5257353,0.51348037,0.6550245,0.5539216,0.68014705,0.6709559,0.54656863,0.5772059,0.7169118,0.5563725,0.61519605,0.68811274,0.7224265,0.7040441,0.6397059,0.67156863,0.69240195,0.76348037,0.7230392,0.747549,0.7144608,0.7555147,0.81740195,0.83210784,0.73406863,0.7585784,0.79044116,0.8958333,0.83455884,0.95710784,0.75919116,0.75980395,0.8737745,0.8492647,0.9142157,0.9191176,0.83210784,0.9289216,0.82843137,0.91789216,0.8811275,0.88235295,0.8982843,0.9485294,0.8958333,0.95955884,0.9362745,0.93443626,0.98651963
+CatBoost (tuned + ensembled),0.08455882,0.16544117,0.1360294,0.18872549,0.18627451,0.20220588,0.24142157,0.25245097,0.28063726,0.3345588,0.3137255,0.3627451,0.39828432,0.40441176,0.4117647,0.48161766,,0.46568626,0.5949755,0.45588234,0.5232843,0.51715684,0.47181374,0.55085784,0.54044116,0.6268382,0.6482843,0.5747549,0.5851716,0.6574755,0.5808824,0.62009805,0.6372549,0.6875,0.6642157,0.65563726,0.69485295,0.67401963,0.7242647,0.70710784,0.7064951,0.69240195,0.7561275,0.7561275,0.78063726,0.7438725,0.778799,0.7610294,0.8872549,0.84313726,0.89215684,0.7585784,0.75490195,0.87806374,0.8566176,0.875,0.8854167,0.85539216,0.8933824,0.8112745,0.9019608,0.85906863,0.877451,0.8995098,0.94240195,0.8713235,0.93259805,0.8958333,0.9264706,0.9877451
+TabM (tuned),0.073529415,0.1629902,0.16115196,0.19117647,0.18504901,0.2009804,0.21997549,0.24142157,0.31740198,0.36764705,0.3627451,0.37561274,0.28063726,0.41299018,0.41421568,0.5055147,0.53431374,,0.5502451,0.48713234,0.53186274,0.48284313,0.4944853,0.5502451,0.5269608,0.58026963,0.60294116,0.56740195,0.5833333,0.6041667,0.54656863,0.5735294,0.7205882,0.65931374,0.6433824,0.67401963,0.65563726,0.67034316,0.6629902,0.76348037,0.69730395,0.72365195,0.69240195,0.73284316,0.72855395,0.7542892,0.70955884,0.79289216,0.80514705,0.7512255,0.8002451,0.7561275,0.7604167,0.76960784,0.9044118,0.86519605,0.88480395,0.8186275,0.875,0.810049,0.84436274,0.8811275,0.877451,0.86151963,0.88848037,0.8897059,0.9375,0.91789216,0.9080882,0.96813726
+CatBoost (tuned),0.07230392,0.15931372,0.13296568,0.18504901,0.18504901,0.19117647,0.23529412,0.23529412,0.27450982,0.33578432,0.29534313,0.3529412,0.38112745,0.3939951,0.40318626,0.44791666,0.4050245,0.4497549,,0.45343137,0.49509802,0.48835784,0.46078432,0.53186274,0.52144605,0.60294116,0.60723037,0.560049,0.5637255,0.64093137,0.56985295,0.6017157,0.62561274,0.6501225,0.65563726,0.6495098,0.6580882,0.65686274,0.69914216,0.69485295,0.6936275,0.67401963,0.73651963,0.75490195,0.76593137,0.7310049,0.7487745,0.7518382,0.88235295,0.82843137,0.8664216,0.7585784,0.75735295,0.8602941,0.84068626,0.86519605,0.8762255,0.84681374,0.8860294,0.8125,0.88848037,0.85539216,0.86519605,0.89093137,0.9338235,0.86580884,0.9362745,0.89093137,0.93014705,0.98651963
+LimiX (default),0.09375,0.21568628,0.22303921,0.21384804,0.24877451,0.24019608,0.2757353,0.2720588,0.40931374,0.34068626,0.45465687,0.40134802,0.49019608,0.4240196,0.46875,0.55514705,0.5441176,0.5128676,0.54656863,,0.57659316,0.48406863,0.46078432,0.57843137,0.54534316,0.5882353,0.60906863,0.56985295,0.60539216,0.6041667,0.54656863,0.56556374,0.5870098,0.67401963,0.67034316,0.6397059,0.61764705,0.66544116,0.6887255,0.6642157,0.6985294,0.68688726,0.6985294,0.7022059,0.7022059,0.6997549,0.7022059,0.7542892,0.7536765,0.752451,0.7438725,0.72855395,0.7254902,0.7708333,0.78186274,0.78063726,0.8039216,0.8308824,0.79534316,0.7775735,0.8137255,0.8566176,0.8394608,0.8370098,0.8602941,0.8713235,0.8713235,0.89093137,0.8762255,0.96813726
+iLTM (tuned + ensembled),0.075980395,0.1360294,0.109068625,0.15686275,0.15808824,0.2077206,0.25,0.27328432,0.25980392,0.2977941,0.2990196,0.30269608,0.39889705,0.36887255,0.40625,0.47242647,0.47671568,0.46813726,0.50490195,0.42340687,,0.48039216,0.44607842,0.54534316,0.5496324,0.5680147,0.5502451,0.5514706,0.5894608,0.6164216,0.55759805,0.6182598,0.5710784,0.65318626,0.6360294,0.66544116,0.7371324,0.70098037,0.67156863,0.72794116,0.67401963,0.7352941,0.7420343,0.7107843,0.71875,0.74509805,0.7683824,0.78615195,0.8308824,0.8272059,0.87009805,0.72181374,0.7224265,0.85539216,0.8688725,0.8541667,0.88235295,0.89460784,0.88848037,0.8063725,0.90931374,0.8137255,0.89705884,0.877451,0.92769605,0.8394608,0.9313725,0.8786765,0.908701,0.98284316
+TabSwift (default),0.113970585,0.11764706,0.16421568,0.1629902,0.2009804,0.20343137,0.23590687,0.22916667,0.40441176,0.29227942,0.4185049,0.3560049,0.46568626,0.4117647,0.48161766,0.4742647,0.48284313,0.51715684,0.51164216,0.51593137,0.51960784,,0.5061275,0.5208333,0.54289216,0.5441176,0.5367647,0.54534316,0.5710784,0.5833333,0.58210784,0.57965684,0.61764705,0.6017157,0.622549,0.6666667,0.6666667,0.67769605,0.63235295,0.7205882,0.6580882,0.76715684,0.685049,0.6789216,0.70465684,0.7254902,0.70710784,0.79289216,0.7389706,0.71813726,0.783701,0.7622549,0.76348037,0.7585784,0.85294116,0.8541667,0.83210784,0.83455884,0.872549,0.83455884,0.86151963,0.82598037,0.89460784,0.88235295,0.8897059,0.8455882,0.93259805,0.9172794,0.95343137,0.9938725
+TabDPT (default),0.1004902,0.17279412,0.23406863,0.20710784,0.27818626,0.30147058,0.3259804,0.33088234,0.4117647,0.29718137,0.44362745,0.29534313,0.4595588,0.35416666,0.4644608,0.4865196,0.52818626,0.5055147,0.5392157,0.5392157,0.5539216,0.49387255,,0.53431374,0.56985295,0.56740195,0.5477941,0.5625,0.5710784,0.5882353,0.5735294,0.59681374,0.62009805,0.62990195,0.5833333,0.6335784,0.6495098,0.6642157,0.6397059,0.6378676,0.6121324,0.67156863,0.73039216,0.65686274,0.6813725,0.6911765,0.7512255,0.7316176,0.71813726,0.7536765,0.75980395,0.6887255,0.689951,0.7916667,0.8161765,0.82598037,0.8112745,0.8394608,0.86151963,0.7855392,0.88480395,0.77818626,0.8762255,0.875,0.8762255,0.79901963,0.92156863,0.8762255,0.90686274,0.99264705
+LightGBM (tuned),0.07230392,0.1384804,0.123774506,0.16176471,0.14705883,0.17953432,0.20465687,0.21446079,0.24203432,0.3345588,0.24019608,0.3382353,0.34865198,0.37745097,0.35784313,0.3449755,0.44914216,0.4497549,0.46813726,0.42156863,0.45465687,0.47916666,0.46568626,,0.49938726,0.5606618,0.5900735,0.502451,0.51960784,0.6041667,0.5183824,0.5502451,0.6397059,0.64705884,0.6642157,0.6066176,0.6066176,0.6507353,0.67769605,0.6942402,0.7144608,0.6807598,0.7077206,0.75980395,0.76960784,0.68811274,0.73223037,0.7414216,0.86090684,0.7769608,0.91605395,0.75,0.7512255,0.82843137,0.82843137,0.8768382,0.903799,0.8204657,0.8982843,0.8137255,0.8933824,0.86151963,0.86764705,0.8762255,0.9252451,0.8805147,0.95465684,0.9227941,0.9264706,0.9816176
+xRFM (tuned + ensembled),0.07965686,0.14705883,0.125,0.17524509,0.1764706,0.18259804,0.20894608,0.21568628,0.23284313,0.30637255,0.3082108,0.30085784,0.40318626,0.34558824,0.37990198,0.44607842,0.4595588,0.4730392,0.47855392,0.45465687,0.45036766,0.45710784,0.43014705,0.50061274,,0.5165441,0.5349265,0.5147059,0.5165441,0.54901963,0.5085784,0.64276963,0.5710784,0.61151963,0.6066176,0.5919118,0.6207108,0.622549,0.62990195,0.6397059,0.6605392,0.66360295,0.6838235,0.685049,0.65931374,0.6764706,0.7230392,0.7291667,0.75735295,0.76715684,0.7622549,0.716299,0.7144608,0.79901963,0.84681374,0.8161765,0.810049,0.9497549,0.8498775,0.78186274,0.935049,0.86764705,0.90563726,0.90625,0.9411765,0.87990195,0.92156863,0.903799,0.9289216,0.9889706
+XGBoost (tuned + ensembled),0.06495098,0.15134804,0.111519605,0.17524509,0.15073529,0.17156863,0.20588236,0.2254902,0.20465687,0.3125,0.23039216,0.32659313,0.3272059,0.3560049,0.32352942,0.31985295,0.37316176,0.4197304,0.3970588,0.4117647,0.4319853,0.45588234,0.43259802,0.43933824,0.4834559,,0.5275735,0.4779412,0.5183824,0.6047794,0.5147059,0.5686275,0.5857843,0.60294116,0.6237745,0.6121324,0.6102941,0.6397059,0.6397059,0.6642157,0.6789216,0.66789216,0.689951,0.73284316,0.7150735,0.6789216,0.7291667,0.7205882,0.8995098,0.8180147,0.9105392,0.7377451,0.73406863,0.841299,0.8186275,0.83210784,0.8541667,0.8272059,0.86090684,0.7916667,0.8786765,0.8504902,0.8455882,0.8664216,0.935049,0.85906863,0.9227941,0.8897059,0.9148284,0.9791667
+CatBoost (default),0.07720588,0.14583333,0.12071078,0.16176471,0.14828432,0.17156863,0.21446079,0.20955883,0.25612745,0.30637255,0.25,0.3137255,0.31617647,0.35784313,0.35784313,0.3290441,0.35171568,0.3970588,0.3927696,0.39093137,0.4497549,0.4632353,0.4522059,0.40992647,0.46507353,0.47242647,,0.4779412,0.5036765,0.5128676,0.5122549,0.55759805,0.57230395,0.62806374,0.6286765,0.5992647,0.61151963,0.59436274,0.6752451,0.65686274,0.6838235,0.66544116,0.6875,0.73284316,0.7757353,0.69240195,0.70465684,0.73284316,0.7830882,0.7769608,0.8161765,0.75061274,0.7512255,0.8063725,0.8125,0.8480392,0.90931374,0.7941176,0.87990195,0.814951,0.8664216,0.8394608,0.8762255,0.872549,0.9019608,0.85539216,0.92769605,0.90563726,0.93259805,0.98406863
+ModernNCA (tuned + ensembled),0.05759804,0.121323526,0.113970585,0.15686275,0.14093137,0.19117647,0.22610295,0.2634804,0.29656863,0.31495097,0.3112745,0.30637255,0.37132353,0.35355392,0.39215687,0.45343137,0.4252451,0.43259802,0.43995097,0.43014705,0.44852942,0.45465687,0.4375,0.49754903,0.4852941,0.52205884,0.52205884,,0.60110295,0.52818626,0.49142158,0.57843137,0.54289216,0.5772059,0.5980392,0.57843137,0.5808824,0.7107843,0.6164216,0.67156863,0.63664216,0.6519608,0.6629902,0.6544118,0.63480395,0.6580882,0.69240195,0.7107843,0.7622549,0.73039216,0.74264705,0.6789216,0.6764706,0.7414216,0.7769608,0.7830882,0.79656863,0.85968137,0.7916667,0.7377451,0.84313726,0.76715684,0.8210784,0.8235294,0.85539216,0.77818626,0.85784316,0.8137255,0.8517157,0.9583333
+ModernNCA (tuned),0.068627454,0.13480392,0.10294118,0.15196079,0.13970588,0.14828432,0.2009804,0.21691176,0.27083334,0.29534313,0.3002451,0.30392158,0.35968137,0.3529412,0.35784313,0.4227941,0.41482842,0.41666666,0.4362745,0.39460784,0.4105392,0.42892158,0.42892158,0.48039216,0.4834559,0.48161766,0.49632353,0.39889705,,0.49693626,0.49142158,0.5416667,0.5416667,0.5808824,0.58210784,0.5833333,0.56985295,0.69240195,0.59375,0.65318626,0.627451,0.6648284,0.6482843,0.6544118,0.6482843,0.6997549,0.6691176,0.7463235,0.71813726,0.7414216,0.7254902,0.72181374,0.7254902,0.75980395,0.8063725,0.79044116,0.8492647,0.85539216,0.82598037,0.7879902,0.84681374,0.82230395,0.8645833,0.8627451,0.877451,0.84436274,0.8958333,0.88848037,0.90563726,0.97794116
+XGBoost (tuned),0.05882353,0.14093137,0.10539216,0.17034313,0.12867647,0.15747549,0.19362745,0.19607843,0.18627451,0.2977941,0.23039216,0.29656863,0.28799018,0.34558824,0.2959559,0.28308824,0.3425245,0.39583334,0.35906863,0.39583334,0.38357842,0.41666666,0.4117647,0.39583334,0.4509804,0.39522058,0.48713234,0.47181374,0.50306374,,0.50735295,0.5324755,0.5514706,0.54840684,0.5980392,0.5857843,0.5747549,0.60784316,0.59436274,0.6360294,0.65686274,0.6433824,0.6832108,0.7022059,0.6727941,0.65931374,0.7169118,0.71568626,0.85294116,0.7892157,0.86764705,0.7242647,0.72610295,0.81985295,0.8063725,0.80514705,0.8480392,0.8161765,0.8455882,0.7892157,0.8762255,0.8419118,0.8357843,0.86335784,0.92034316,0.85784316,0.92156863,0.88480395,0.9105392,0.9767157
+SAP-RPT-OSS (default),0.121323526,0.20343137,0.22794117,0.21936275,0.20833333,0.21875,0.24754901,0.2365196,0.32843137,0.33700982,0.38235295,0.3394608,0.3762255,0.38602942,0.39215687,0.44362745,0.41911766,0.45343137,0.43014705,0.45343137,0.44240198,0.41789216,0.42647058,0.48161766,0.49142158,0.4852941,0.4877451,0.5085784,0.5085784,0.49264705,,0.49509802,0.51593137,0.5232843,0.5563725,0.5857843,0.5300245,0.5833333,0.5441176,0.622549,0.59313726,0.67034316,0.625,0.6329657,0.59436274,0.6262255,0.6433824,0.7267157,0.64705884,0.6770833,0.68014705,0.6507353,0.6458333,0.68811274,0.75490195,0.7683824,0.7267157,0.7622549,0.7775735,0.79534316,0.79044116,0.8247549,0.8161765,0.79044116,0.79901963,0.8308824,0.8382353,0.8627451,0.88848037,0.9852941
+xRFM (tuned),0.08455882,0.1360294,0.13051471,0.17953432,0.15686275,0.17279412,0.18995099,0.18627451,0.22794117,0.2634804,0.2622549,0.27879903,0.375,0.33088234,0.32781863,0.38480392,0.37990198,0.42647058,0.39828432,0.43443626,0.38174018,0.42034313,0.40318626,0.4497549,0.3572304,0.43137255,0.44240198,0.42156863,0.45833334,0.4675245,0.50490195,,0.53186274,0.54473037,0.56311274,0.56985295,0.5661765,0.59313726,0.5870098,0.5992647,0.6164216,0.6305147,0.6372549,0.6397059,0.6237745,0.6458333,0.685049,0.7058824,0.6764706,0.7022059,0.70465684,0.7169118,0.71936274,0.7438725,0.8382353,0.7800245,0.8137255,0.9019608,0.8204657,0.79044116,0.8688725,0.8627451,0.8860294,0.8933824,0.89093137,0.87990195,0.9080882,0.91789216,0.92769605,0.98651963
+TabM (default),0.06740196,0.12990196,0.116421565,0.14338236,0.1317402,0.14705883,0.16911764,0.16911764,0.18504901,0.28308824,0.25367647,0.28982842,0.13909313,0.32965687,0.26838234,0.31188726,0.3627451,0.27941176,0.37438726,0.41299018,0.42892158,0.38235295,0.37990198,0.3602941,0.42892158,0.41421568,0.42769608,0.45710784,0.45833334,0.44852942,0.48406863,0.46813726,,0.4779412,0.502451,0.57230395,0.53615195,0.56740195,0.50490195,0.6102941,0.5710784,0.6433824,0.6188725,0.5998775,0.57598037,0.6580882,0.6397059,0.7242647,0.67156863,0.6875,0.6544118,0.7352941,0.7377451,0.6985294,0.79656863,0.7628676,0.8014706,0.7291667,0.7745098,0.8125,0.7916667,0.8272059,0.8247549,0.8186275,0.83210784,0.83884805,0.88235295,0.88235295,0.9007353,0.9522059
+ChimeraBoost (tuned + ensembled),0.06495098,0.113970585,0.083333336,0.14338236,0.10539216,0.15134804,0.19056372,0.18382353,0.23039216,0.25490198,0.20281863,0.26960784,0.27818626,0.30392158,0.33333334,0.27757353,0.3125,0.34068626,0.34987745,0.3259804,0.34681374,0.39828432,0.37009802,0.3529412,0.3884804,0.3970588,0.37193626,0.4227941,0.41911766,0.45159313,0.47671568,0.4552696,0.52205884,,0.5539216,0.52818626,0.51593137,0.47181374,0.60723037,0.56985295,0.5955882,0.5612745,0.5747549,0.6311275,0.79656863,0.5919118,0.6102941,0.6084559,0.67769605,0.67769605,0.73223037,0.70955884,0.70465684,0.71568626,0.7169118,0.7867647,0.81740195,0.74264705,0.814951,0.7855392,0.7879902,0.7855392,0.8161765,0.79534316,0.8210784,0.8002451,0.87009805,0.87009805,0.8860294,0.97365195
+EBM (tuned + ensembled),0.073529415,0.14093137,0.11029412,0.15441176,0.1384804,0.14338236,0.1629902,0.18014705,0.20220588,0.28799018,0.22671568,0.29227942,0.26715687,0.33333334,0.27941176,0.2959559,0.33578432,0.35661766,0.34436274,0.32965687,0.36397058,0.37745097,0.41666666,0.33578432,0.39338234,0.3762255,0.37132353,0.4019608,0.41789216,0.4019608,0.44362745,0.43688726,0.49754903,0.44607842,,0.5257353,0.48284313,0.5257353,0.4865196,0.5625,0.721201,0.55085784,0.6023284,0.7683824,0.560049,0.61580884,0.6127451,0.6231618,0.6415441,0.6580882,0.6360294,0.7230392,0.7144608,0.6838235,0.7463235,0.75490195,0.80269605,0.7083333,0.7726716,0.810049,0.7585784,0.86151963,0.83210784,0.82965684,0.79534316,0.875,0.88480395,0.9007353,0.8878676,0.9522059
+TorchMLP (tuned + ensembled),0.040441178,0.08946078,0.0747549,0.12009804,0.104166664,0.1127451,0.1360294,0.1360294,0.21813725,0.22426471,0.18137255,0.25061274,0.26593137,0.2904412,0.26593137,0.3602941,0.34436274,0.3259804,0.35049018,0.3602941,0.3345588,0.33333334,0.36642158,0.39338234,0.40808824,0.38786766,0.4007353,0.42156863,0.41666666,0.41421568,0.41421568,0.43014705,0.42769608,0.47181374,0.4742647,,0.5067402,0.5441176,0.5085784,0.55514705,0.5392157,0.63848037,0.63235295,0.57965684,0.52818626,0.7487745,0.64705884,0.7432598,0.65318626,0.6875,0.6482843,0.70343137,0.7101716,0.7120098,0.89215684,0.72181374,0.8069853,0.7683824,0.7708333,0.8002451,0.7647059,0.81188726,0.8584559,0.81985295,0.79656863,0.8370098,0.88235295,0.8982843,0.8933824,0.95955884
+iLTM (tuned),0.05759804,0.07414216,0.08455882,0.125,0.09313726,0.12622549,0.14583333,0.14154412,0.1740196,0.20159313,0.19240196,0.21078432,0.27328432,0.26593137,0.26960784,0.32843137,0.30514705,0.34436274,0.34191176,0.38235295,0.26286766,0.33333334,0.35049018,0.39338234,0.3792892,0.3897059,0.3884804,0.41911766,0.43014705,0.4252451,0.4699755,0.43382353,0.46384802,0.48406863,0.51715684,0.49325982,,0.5606618,0.5349265,0.5710784,0.56985295,0.6188725,0.59436274,0.6041667,0.5747549,0.6041667,0.64460784,0.67034316,0.6666667,0.6960784,0.69240195,0.6917892,0.6887255,0.7291667,0.810049,0.7230392,0.8106618,0.8112745,0.7708333,0.75980395,0.8161765,0.76164216,0.84681374,0.814951,0.84681374,0.7916667,0.86519605,0.8492647,0.8731618,0.9620098
+ModernNCA (default),0.06127451,0.094975494,0.104166664,0.11580882,0.101715684,0.13725491,0.16053921,0.15563725,0.20343137,0.2009804,0.21446079,0.22120099,0.27328432,0.2757353,0.25857842,0.30759802,0.3259804,0.32965687,0.34313726,0.3345588,0.2990196,0.32230392,0.33578432,0.3492647,0.37745097,0.3602941,0.40563726,0.28921568,0.30759802,0.39215687,0.41666666,0.40686274,0.43259802,0.52818626,0.4742647,0.45588234,0.43933824,,0.5441176,0.49264705,0.51960784,0.5477941,0.5686275,0.5637255,0.5857843,0.5416667,0.6041667,0.62009805,0.5882353,0.6617647,0.6237745,0.6789216,0.6838235,0.72181374,0.7732843,0.7242647,0.79289216,0.7647059,0.7769608,0.7794118,0.81311274,0.7977941,0.8247549,0.86519605,0.85784316,0.8180147,0.8639706,0.8982843,0.90931374,0.98039216
+ChimeraBoost (tuned),0.06004902,0.113970585,0.075980395,0.13357843,0.09681372,0.13357843,0.16544117,0.17279412,0.20894608,0.23284313,0.18504901,0.25,0.24142157,0.2904412,0.29656863,0.2365196,0.2757353,0.33700982,0.30085784,0.3112745,0.32843137,0.36764705,0.3602941,0.32230392,0.37009802,0.3602941,0.3247549,0.38357842,0.40625,0.40563726,0.45588234,0.41299018,0.49509802,0.3927696,0.51348037,0.49142158,0.46507353,0.45588234,,0.55269605,0.5667892,0.5379902,0.55269605,0.5955882,0.71568626,0.5625,0.59313726,0.5876225,0.6501225,0.6495098,0.6862745,0.69730395,0.6960784,0.689951,0.70710784,0.74693626,0.7916667,0.72181374,0.77634805,0.7708333,0.7769608,0.7732843,0.7941176,0.7769608,0.79044116,0.7794118,0.86335784,0.8566176,0.8639706,0.9620098
+RealMLP (default),0.05269608,0.0747549,0.0882353,0.09803922,0.09068628,0.11764706,0.13357843,0.13970588,0.09865196,0.19485295,0.20220588,0.22181372,0.17708333,0.26593137,0.19730392,0.2769608,0.29289216,0.2365196,0.30514705,0.33578432,0.2720588,0.27941176,0.36213234,0.30575982,0.3602941,0.33578432,0.34313726,0.32843137,0.34681374,0.36397058,0.37745097,0.4007353,0.3897059,0.43014705,0.4375,0.44485295,0.42892158,0.50735295,0.44730392,,0.50490195,0.5422794,0.5735294,0.53431374,0.5110294,0.5833333,0.6066176,0.65318626,0.6139706,0.65318626,0.6102941,0.67401963,0.6813725,0.6875,0.77634805,0.71568626,0.77205884,0.7205882,0.77205884,0.77818626,0.79534316,0.79534316,0.872549,0.82843137,0.8014706,0.82598037,0.8762255,0.8872549,0.90686274,0.9791667
+EBM (tuned),0.0625,0.123774506,0.101715684,0.1360294,0.10845588,0.12193628,0.14705883,0.15196079,0.18259804,0.25306374,0.20710784,0.2677696,0.22303921,0.29411766,0.24142157,0.25245097,0.2935049,0.30269608,0.30637255,0.30147058,0.3259804,0.34191176,0.38786766,0.2855392,0.3394608,0.32107842,0.31617647,0.36335784,0.37254903,0.34313726,0.40686274,0.38357842,0.42892158,0.40441176,0.27879903,0.46078432,0.43014705,0.48039216,0.4332108,0.49509802,,0.502451,0.5477941,0.6623775,0.5036765,0.56740195,0.58210784,0.5949755,0.5919118,0.61151963,0.58455884,0.70343137,0.7003676,0.6519608,0.7169118,0.72181374,0.75980395,0.6862745,0.7463235,0.79534316,0.74264705,0.83455884,0.79901963,0.80269605,0.7683824,0.8504902,0.8639706,0.88235295,0.8860294,0.9460784
+FastaiMLP (tuned + ensembled),0.073529415,0.11580882,0.101715684,0.12990196,0.12745099,0.1317402,0.1495098,0.12622549,0.15257353,0.18627451,0.17524509,0.22181372,0.22426471,0.2634804,0.22671568,0.2855392,0.30759802,0.27634802,0.3259804,0.31311274,0.2647059,0.23284313,0.32843137,0.31924018,0.33639705,0.33210784,0.3345588,0.3480392,0.33517158,0.35661766,0.32965687,0.3694853,0.35661766,0.4387255,0.44914216,0.3615196,0.38112745,0.4522059,0.46200982,0.45772058,0.49754903,,0.5085784,0.51715684,0.4852941,0.49019608,0.5257353,0.685049,0.54044116,0.6047794,0.533701,0.6311275,0.63480395,0.622549,0.6942402,0.63848037,0.73284316,0.70343137,0.6727941,0.74264705,0.74509805,0.8272059,0.95098037,0.7855392,0.73039216,0.8480392,0.8186275,0.89705884,0.8602941,0.95465684
+ExtraTrees (tuned + ensembled),0.04656863,0.06495098,0.075980395,0.102328435,0.09191176,0.12254902,0.13296568,0.12990196,0.15196079,0.15134804,0.21691176,0.18259804,0.25735295,0.22671568,0.23774509,0.24448529,0.24387255,0.30759802,0.2634804,0.30147058,0.25796568,0.31495097,0.26960784,0.29227942,0.31617647,0.31004903,0.3125,0.33700982,0.35171568,0.3167892,0.375,0.3627451,0.38112745,0.4252451,0.39767158,0.36764705,0.40563726,0.43137255,0.44730392,0.42647058,0.4522059,0.49142158,,0.48039216,0.49754903,0.44424018,0.6807598,0.5392157,0.51960784,0.6911765,0.5379902,0.5955882,0.60539216,0.7438725,0.6127451,0.6507353,0.6985294,0.7132353,0.6838235,0.73223037,0.9191176,0.7420343,0.75,0.8688725,0.8860294,0.7579657,0.78063726,0.8308824,0.9129902,0.9914216
+EBM (default),0.06495098,0.10355392,0.1004902,0.11029412,0.1127451,0.12009804,0.13480392,0.1360294,0.16115196,0.20955883,0.17156863,0.2512255,0.21384804,0.2757353,0.24142157,0.18259804,0.24387255,0.26715687,0.24509804,0.2977941,0.28921568,0.32107842,0.34313726,0.24019608,0.31495097,0.26715687,0.26715687,0.34558824,0.34558824,0.2977941,0.36703432,0.3602941,0.40012255,0.36887255,0.23161764,0.42034313,0.39583334,0.4362745,0.40441176,0.46568626,0.33762255,0.48284313,0.51960784,,0.4632353,0.5379902,0.5379902,0.5686275,0.52818626,0.5772059,0.54289216,0.6727941,0.67034316,0.60906863,0.7058824,0.68014705,0.7561275,0.6433824,0.70710784,0.77818626,0.72794116,0.81985295,0.7879902,0.8186275,0.7414216,0.8449755,0.8382353,0.8995098,0.8897059,0.9485294
+ChimeraBoost (default),0.04534314,0.104166664,0.06985294,0.123774506,0.08946078,0.1127451,0.1360294,0.13357843,0.18504901,0.20833333,0.1495098,0.2120098,0.19240196,0.25245097,0.25,0.16789216,0.21936275,0.27144608,0.23406863,0.2977941,0.28125,0.29534313,0.31862745,0.23039216,0.34068626,0.28492647,0.22426471,0.36519608,0.35171568,0.3272059,0.40563726,0.3762255,0.4240196,0.20343137,0.43995097,0.47181374,0.4252451,0.41421568,0.28431374,0.48897058,0.49632353,0.5147059,0.502451,0.5367647,,0.53186274,0.5232843,0.55269605,0.58639705,0.6127451,0.591299,0.6727941,0.67769605,0.6286765,0.65686274,0.69240195,0.7561275,0.67769605,0.72794116,0.74264705,0.685049,0.76593137,0.747549,0.7334559,0.75735295,0.7757353,0.81985295,0.85539216,0.8639706,0.9607843
+TorchMLP (tuned),0.030637255,0.06127451,0.05514706,0.09191176,0.06495098,0.07965686,0.09558824,0.09558824,0.16789216,0.17156863,0.14215687,0.18995099,0.18566176,0.2365196,0.19485295,0.26593137,0.25612745,0.24571079,0.2689951,0.3002451,0.25490198,0.27450982,0.30882353,0.31188726,0.32352942,0.32107842,0.30759802,0.34191176,0.3002451,0.34068626,0.3737745,0.35416666,0.34191176,0.40808824,0.3841912,0.2512255,0.39583334,0.45833334,0.4375,0.41666666,0.43259802,0.50980395,0.5557598,0.46200982,0.46813726,,0.5747549,0.6066176,0.5367647,0.60784316,0.5545343,0.685049,0.6862745,0.6372549,0.8082108,0.65318626,0.7438725,0.7083333,0.70098037,0.77818626,0.71813726,0.74264705,0.78186274,0.7775735,0.74509805,0.7604167,0.8235294,0.87009805,0.8566176,0.94730395
+ExtraTrees (tuned),0.05147059,0.063725494,0.083333336,0.1004902,0.094975494,0.12009804,0.125,0.11764706,0.1427696,0.12867647,0.19362745,0.1629902,0.2512255,0.20833333,0.21568628,0.24142157,0.22120099,0.2904412,0.2512255,0.2977941,0.23161764,0.29289216,0.24877451,0.2677696,0.2769608,0.27083334,0.29534313,0.30759802,0.33088234,0.28308824,0.35661766,0.31495097,0.3602941,0.3897059,0.3872549,0.3529412,0.35539216,0.39583334,0.40686274,0.39338234,0.41789216,0.4742647,0.31924018,0.46200982,0.47671568,0.4252451,,0.52205884,0.4644608,0.59313726,0.49509802,0.596201,0.6017157,0.66115195,0.5857843,0.6286765,0.6764706,0.67769605,0.6642157,0.72794116,0.8504902,0.72794116,0.7291667,0.8382353,0.84865195,0.7377451,0.7683824,0.8235294,0.91789216,0.99019605
+FastaiMLP (tuned),0.0502451,0.07904412,0.083333336,0.09313726,0.09068628,0.09068628,0.09681372,0.09436274,0.13909313,0.13725491,0.13725491,0.17524509,0.16911764,0.20465687,0.17034313,0.20955883,0.2389706,0.20710784,0.24816176,0.24571079,0.21384804,0.20710784,0.26838234,0.25857842,0.27083334,0.27941176,0.26715687,0.28921568,0.25367647,0.28431374,0.27328432,0.29411766,0.2757353,0.3915441,0.37683824,0.25674018,0.32965687,0.37990198,0.41237745,0.34681374,0.4050245,0.31495097,0.46078432,0.43137255,0.44730392,0.39338234,0.4779412,,0.46936274,0.5324755,0.4644608,0.5919118,0.5894608,0.5514706,0.6121324,0.5833333,0.67401963,0.63480395,0.6262255,0.72181374,0.70098037,0.7647059,0.8333333,0.74264705,0.6887255,0.7873775,0.7561275,0.86519605,0.83210784,0.9460784
+XGBoost (default),0.028186275,0.04901961,0.037990198,0.08210784,0.037990198,0.06985294,0.0882353,0.0992647,0.08946078,0.14460784,0.08946078,0.15808824,0.14215687,0.21446079,0.15441176,0.104166664,0.1127451,0.19485295,0.11764706,0.24632353,0.16911764,0.26102942,0.28186274,0.13909313,0.24264705,0.1004902,0.21691176,0.23774509,0.28186274,0.14705883,0.3529412,0.32352942,0.32843137,0.32230392,0.3584559,0.34681374,0.33333334,0.4117647,0.34987745,0.38602942,0.40808824,0.4595588,0.48039216,0.47181374,0.41360295,0.4632353,0.5355392,0.53063726,,0.54901963,0.4810049,0.6397059,0.6433824,0.6066176,0.6360294,0.5735294,0.75,0.70465684,0.6507353,0.7242647,0.7512255,0.72181374,0.71813726,0.76960784,0.810049,0.73651963,0.7745098,0.82230395,0.84436274,0.94485295
+RandomForest (tuned + ensembled),0.030637255,0.06127451,0.05637255,0.0747549,0.068627454,0.08210784,0.11029412,0.123774506,0.0882353,0.12867647,0.14338236,0.12990196,0.21323529,0.18872549,0.15563725,0.16544117,0.15686275,0.24877451,0.17156863,0.24754901,0.17279412,0.28186274,0.24632353,0.22303921,0.23284313,0.18198529,0.22303921,0.26960784,0.25857842,0.21078432,0.32291666,0.2977941,0.3125,0.32230392,0.34191176,0.3125,0.30392158,0.3382353,0.35049018,0.34681374,0.3884804,0.39522058,0.30882353,0.4227941,0.3872549,0.39215687,0.40686274,0.4675245,0.4509804,,0.46813726,0.5661765,0.56740195,0.6789216,0.56985295,0.59313726,0.627451,0.6605392,0.6286765,0.6954657,0.78186274,0.71568626,0.7022059,0.8014706,0.8958333,0.7254902,0.7267157,0.8057598,0.8860294,0.97181374
+LightGBM (default),0.028186275,0.040441178,0.039215688,0.05882353,0.041666668,0.05147059,0.0625,0.066176474,0.07230392,0.11764706,0.075980395,0.12254902,0.117034316,0.16789216,0.14338236,0.042892158,0.10784314,0.19975491,0.13357843,0.25612745,0.12990196,0.21629901,0.24019608,0.08394608,0.23774509,0.08946078,0.18382353,0.25735295,0.27450982,0.13235295,0.31985295,0.29534313,0.34558824,0.2677696,0.36397058,0.35171568,0.30759802,0.3762255,0.3137255,0.3897059,0.4154412,0.46629903,0.46200982,0.45710784,0.40870097,0.44546568,0.50490195,0.5355392,0.5189951,0.53186274,,0.6335784,0.62990195,0.57414216,0.6482843,0.60784316,0.7058824,0.6862745,0.6691176,0.71813726,0.73039216,0.69730395,0.7291667,0.752451,0.7726716,0.7120098,0.80514705,0.8137255,0.8492647,0.95526963
+TabSTAR (tuned),0.12745099,0.14889705,0.16421568,0.15563725,0.1629902,0.17034313,0.19485295,0.16911764,0.21813725,0.23529412,0.22303921,0.24877451,0.22671568,0.27757353,0.23406863,0.24080883,0.24142157,0.24387255,0.24142157,0.27144608,0.27818626,0.23774509,0.3112745,0.25,0.28370097,0.2622549,0.24938725,0.32107842,0.27818626,0.2757353,0.3492647,0.28308824,0.2647059,0.2904412,0.2769608,0.29656863,0.3082108,0.32107842,0.30269608,0.3259804,0.29656863,0.36887255,0.40441176,0.3272059,0.3272059,0.31495097,0.40379903,0.40808824,0.3602941,0.43382353,0.36642158,,0.53125,0.47181374,0.42647058,0.55514705,0.43014705,0.5257353,0.5772059,0.95343137,0.622549,0.6629902,0.6237745,0.69485295,0.6102941,0.685049,0.6139706,0.7892157,0.82598037,0.90318626
+TabSTAR (tuned + ensembled),0.12867647,0.15073529,0.16421568,0.16053921,0.15931372,0.17034313,0.18995099,0.1697304,0.21446079,0.23284313,0.22058824,0.24754901,0.22426471,0.27818626,0.23284313,0.24019608,0.24509804,0.23958333,0.24264705,0.27450982,0.27757353,0.2365196,0.31004903,0.24877451,0.2855392,0.26593137,0.24877451,0.32352942,0.27450982,0.27389705,0.35416666,0.28063726,0.2622549,0.29534313,0.2855392,0.28982842,0.3112745,0.31617647,0.30392158,0.31862745,0.29963234,0.36519608,0.39460784,0.32965687,0.32230392,0.3137255,0.39828432,0.4105392,0.35661766,0.43259802,0.37009802,0.46875,,0.4730392,0.4252451,0.5514706,0.43259802,0.5232843,0.57230395,0.9669118,0.6237745,0.65563726,0.62009805,0.6862745,0.6127451,0.68259805,0.6127451,0.7892157,0.8210784,0.90318626
+RandomForest (tuned),0.031862747,0.04534314,0.05392157,0.07230392,0.05147059,0.06740196,0.0882353,0.0882353,0.068627454,0.11764706,0.1182598,0.121323526,0.17892157,0.16053921,0.13725491,0.12622549,0.12193628,0.23039216,0.13970588,0.22916667,0.14460784,0.24142157,0.20833333,0.17156863,0.2009804,0.15870099,0.19362745,0.25857842,0.24019608,0.18014705,0.31188726,0.25612745,0.30147058,0.28431374,0.31617647,0.28799018,0.27083334,0.27818626,0.31004903,0.3125,0.3480392,0.37745097,0.25612745,0.39093137,0.37132353,0.3627451,0.33884802,0.44852942,0.39338234,0.32107842,0.42585784,0.52818626,0.5269608,,0.52205884,0.56740195,0.60784316,0.6164216,0.60539216,0.67769605,0.7058824,0.69485295,0.69485295,0.7702206,0.7647059,0.70710784,0.7083333,0.79534316,0.8786765,0.9705882
+TorchMLP (default),0.024509804,0.034313727,0.030637255,0.05392157,0.036764707,0.042892158,0.04411765,0.028186275,0.063725494,0.09558824,0.078431375,0.10539216,0.07720588,0.12745099,0.09436274,0.15073529,0.14338236,0.09558824,0.15931372,0.21813725,0.13112745,0.14705883,0.18382353,0.17156863,0.15318628,0.18137255,0.1875,0.22303921,0.19362745,0.19362745,0.24509804,0.16176471,0.20343137,0.28308824,0.25367647,0.10784314,0.18995099,0.22671568,0.29289216,0.22365196,0.28308824,0.30575982,0.3872549,0.29411766,0.34313726,0.19178921,0.41421568,0.38786766,0.36397058,0.43014705,0.35171568,0.5735294,0.5747549,0.4779412,,0.5122549,0.54534316,0.58455884,0.5539216,0.6911765,0.64460784,0.67156863,0.64705884,0.7083333,0.64215684,0.69240195,0.66789216,0.80269605,0.8272059,0.93014705
+PerpetualBooster (tuned + ensembled),0.013480392,0.05269608,0.037990198,0.05392157,0.04411765,0.042892158,0.061887253,0.06495098,0.05637255,0.09803922,0.063725494,0.0882353,0.113970585,0.106617644,0.0992647,0.085784316,0.125,0.13480392,0.13480392,0.21936275,0.14583333,0.14583333,0.1740196,0.12316176,0.18382353,0.16789216,0.15196079,0.21691176,0.20955883,0.19485295,0.23161764,0.21997549,0.23713236,0.21323529,0.24509804,0.27818626,0.2769608,0.2757353,0.25306374,0.28431374,0.27818626,0.3615196,0.3492647,0.31985295,0.30759802,0.34681374,0.37132353,0.41666666,0.42647058,0.40686274,0.39215687,0.44485295,0.44852942,0.43259802,0.4877451,,0.5612745,0.5637255,0.7395833,0.53063726,0.59068626,0.6507353,0.5894608,0.6047794,0.6139706,0.6629902,0.84313726,0.7169118,0.74509805,0.935049
+iLTM (default),0.042892158,0.039215688,0.041666668,0.06004902,0.05392157,0.063725494,0.06985294,0.06004902,0.0882353,0.118872546,0.0870098,0.12867647,0.09436274,0.16544117,0.123774506,0.080882356,0.114583336,0.11519608,0.123774506,0.19607843,0.11764706,0.16789216,0.18872549,0.09620098,0.18995099,0.14583333,0.09068628,0.20343137,0.15073529,0.15196079,0.27328432,0.18627451,0.1985294,0.18259804,0.19730392,0.19301471,0.18933824,0.20710784,0.20833333,0.22794117,0.24019608,0.26715687,0.30147058,0.24387255,0.24387255,0.25612745,0.32352942,0.3259804,0.25,0.37254903,0.29411766,0.56985295,0.56740195,0.39215687,0.45465687,0.4387255,,0.49019608,0.46078432,0.65318626,0.502451,0.6544118,0.59313726,0.69240195,0.56740195,0.67156863,0.5735294,0.7610294,0.79534316,0.89705884
+xRFM (default),0.029411765,0.037990198,0.040441178,0.05514706,0.05147059,0.063725494,0.073529415,0.066176474,0.06127451,0.09313726,0.09558824,0.082720585,0.14093137,0.106617644,0.118872546,0.16789216,0.14460784,0.18137255,0.15318628,0.16911764,0.10539216,0.16544117,0.16053921,0.17953432,0.0502451,0.17279412,0.20588236,0.14031863,0.14460784,0.18382353,0.23774509,0.09803922,0.27083334,0.25735295,0.29166666,0.23161764,0.18872549,0.23529412,0.27818626,0.27941176,0.3137255,0.29656863,0.2867647,0.35661766,0.32230392,0.29166666,0.32230392,0.36519608,0.29534313,0.3394608,0.3137255,0.4742647,0.47671568,0.38357842,0.4154412,0.4362745,0.50980395,,0.46629903,0.5735294,0.47549018,0.5870098,0.53063726,0.58210784,0.5257353,0.59313726,0.59068626,0.6752451,0.7083333,0.92034316
+PerpetualBooster (tuned),0.014705882,0.035539217,0.03247549,0.0502451,0.04656863,0.041666668,0.05514706,0.05759804,0.04901961,0.07230392,0.0625,0.07904412,0.09436274,0.09803922,0.09191176,0.071078435,0.106617644,0.125,0.113970585,0.20465687,0.111519605,0.12745099,0.1384804,0.101715684,0.15012255,0.13909313,0.12009804,0.20833333,0.1740196,0.15441176,0.22242647,0.17953432,0.2254902,0.18504901,0.22732843,0.22916667,0.22916667,0.22303921,0.22365196,0.22794117,0.25367647,0.3272059,0.31617647,0.29289216,0.2720588,0.2990196,0.33578432,0.3737745,0.3492647,0.37132353,0.33088234,0.4227941,0.42769608,0.39460784,0.44607842,0.26041666,0.5392157,0.533701,,0.5085784,0.54656863,0.6335784,0.5502451,0.56740195,0.58455884,0.64460784,0.74693626,0.6985294,0.7169118,0.9142157
+TabSTAR (default),0.09558824,0.11029412,0.12622549,0.10784314,0.13112745,0.1360294,0.14338236,0.12867647,0.1629902,0.15441176,0.16789216,0.16421568,0.17034313,0.1764706,0.17892157,0.17156863,0.18872549,0.18995099,0.1875,0.22242647,0.19362745,0.16544117,0.21446079,0.18627451,0.21813725,0.20833333,0.18504901,0.2622549,0.2120098,0.21078432,0.20465687,0.20955883,0.1875,0.21446079,0.18995099,0.19975491,0.24019608,0.22058824,0.22916667,0.22181372,0.20465687,0.25735295,0.2677696,0.22181372,0.25735295,0.22181372,0.2720588,0.27818626,0.2757353,0.30453432,0.28186274,0.04656863,0.033088237,0.32230392,0.30882353,0.46936274,0.34681374,0.42647058,0.49142158,,0.46568626,0.55759805,0.4522059,0.55269605,0.43382353,0.56740195,0.53186274,0.6617647,0.6911765,0.872549
+ExtraTrees (default),0.013480392,0.024509804,0.041666668,0.041666668,0.039215688,0.04411765,0.06127451,0.05759804,0.039215688,0.054534312,0.06985294,0.04901961,0.10784314,0.07720588,0.11029412,0.08210784,0.09803922,0.15563725,0.111519605,0.18627451,0.09068628,0.1384804,0.11519608,0.106617644,0.06495098,0.121323526,0.13357843,0.15686275,0.15318628,0.123774506,0.20955883,0.13112745,0.20833333,0.2120098,0.24142157,0.23529412,0.18382353,0.18688725,0.22303921,0.20465687,0.25735295,0.25490198,0.080882356,0.2720588,0.31495097,0.28186274,0.1495098,0.2990196,0.24877451,0.21813725,0.26960784,0.37745097,0.3762255,0.29411766,0.35539216,0.40931374,0.49754903,0.5245098,0.45343137,0.53431374,,0.5955882,0.55759805,0.6237745,0.54534316,0.60906863,0.57598037,0.70098037,0.75735295,0.97181374
+Linear (tuned + ensembled),0.041666668,0.06740196,0.07230392,0.05882353,0.0747549,0.07965686,0.09068628,0.078431375,0.09436274,0.12745099,0.097426474,0.15931372,0.09681372,0.17340687,0.13480392,0.118872546,0.14093137,0.118872546,0.14460784,0.14338236,0.18627451,0.1740196,0.22181372,0.1384804,0.13235295,0.1495098,0.16053921,0.23284313,0.17769608,0.15808824,0.17524509,0.13725491,0.17279412,0.21446079,0.1384804,0.18811275,0.23835784,0.20220588,0.22671568,0.20465687,0.16544117,0.17279412,0.25796568,0.18014705,0.23406863,0.25735295,0.2720588,0.23529412,0.27818626,0.28431374,0.30269608,0.33700982,0.34436274,0.30514705,0.32843137,0.3492647,0.34558824,0.41299018,0.36642158,0.44240198,0.40441176,,0.36642158,0.44485295,0.40686274,0.6672794,0.46078432,0.8125,0.54901963,0.7585784
+FastaiMLP (default),0.023284314,0.033088237,0.034313727,0.0502451,0.04901961,0.04779412,0.05514706,0.05147059,0.04779412,0.06127451,0.05269608,0.07965686,0.08210784,0.09436274,0.0692402,0.11764706,0.12254902,0.12254902,0.13480392,0.16053921,0.10294118,0.10539216,0.123774506,0.13235295,0.09436274,0.15441176,0.123774506,0.17892157,0.13541667,0.16421568,0.18382353,0.113970585,0.17524509,0.18382353,0.16789216,0.14154412,0.15318628,0.17524509,0.20588236,0.12745099,0.2009804,0.04901961,0.25,0.2120098,0.25245097,0.21813725,0.27083334,0.16666667,0.28186274,0.2977941,0.27083334,0.3762255,0.37990198,0.30514705,0.3529412,0.4105392,0.40686274,0.46936274,0.4497549,0.5477941,0.44240198,0.6335784,,0.5735294,0.47732842,0.65318626,0.5882353,0.747549,0.74509805,0.9129902
+KNN (tuned + ensembled),0.030637255,0.05147059,0.04901961,0.04534314,0.05147059,0.06985294,0.08210784,0.09191176,0.0747549,0.0747549,0.09558824,0.082720585,0.109068625,0.08946078,0.11029412,0.101715684,0.1004902,0.1384804,0.109068625,0.1629902,0.12254902,0.11764706,0.125,0.123774506,0.09375,0.13357843,0.12745099,0.1764706,0.13725491,0.13664216,0.20955883,0.106617644,0.18137255,0.20465687,0.17034313,0.18014705,0.18504901,0.13480392,0.22303921,0.17156863,0.19730392,0.21446079,0.13112745,0.18137255,0.2665441,0.22242647,0.16176471,0.25735295,0.23039216,0.1985294,0.24754901,0.30514705,0.3137255,0.2297794,0.29166666,0.39522058,0.30759802,0.41789216,0.43259802,0.44730392,0.3762255,0.55514705,0.42647058,,0.4117647,0.5827206,0.5392157,0.6985294,0.88664216,0.9742647
+RandomForest (default),0.014705882,0.014705882,0.026960785,0.031862747,0.022058824,0.034313727,0.04411765,0.04534314,0.034313727,0.037990198,0.05882353,0.04411765,0.07965686,0.06985294,0.07965686,0.05147059,0.05759804,0.111519605,0.066176474,0.13970588,0.07230392,0.11029412,0.123774506,0.0747549,0.05882353,0.06495098,0.09803922,0.14460784,0.12254902,0.07965686,0.2009804,0.109068625,0.16789216,0.17892157,0.20465687,0.20343137,0.15318628,0.14215687,0.20955883,0.1985294,0.23161764,0.26960784,0.113970585,0.25857842,0.24264705,0.25490198,0.15134804,0.3112745,0.18995099,0.104166664,0.22732843,0.3897059,0.3872549,0.23529412,0.35784313,0.38602942,0.43259802,0.4742647,0.4154412,0.5661765,0.45465687,0.59313726,0.5226716,0.5882353,,0.61519605,0.56740195,0.6960784,0.7487745,0.9460784
+Linear (tuned),0.036764707,0.06127451,0.05637255,0.05269608,0.06004902,0.068627454,0.075980395,0.073529415,0.09313726,0.118872546,0.080269605,0.14583333,0.085171565,0.15808824,0.12806372,0.104166664,0.12867647,0.11029412,0.13419117,0.12867647,0.16053921,0.15441176,0.2009804,0.1194853,0.12009804,0.14093137,0.14460784,0.22181372,0.15563725,0.14215687,0.16911764,0.12009804,0.16115196,0.19975491,0.125,0.1629902,0.20833333,0.18198529,0.22058824,0.1740196,0.1495098,0.15196079,0.24203432,0.15502451,0.22426471,0.23958333,0.2622549,0.21262255,0.2634804,0.27450982,0.28799018,0.31495097,0.31740198,0.29289216,0.30759802,0.33700982,0.32843137,0.40686274,0.35539216,0.43259802,0.39093137,0.33272058,0.34681374,0.41727942,0.38480392,,0.45465687,0.7438725,0.5367647,0.752451
+PerpetualBooster (default),0.012254902,0.022058824,0.025735294,0.028186275,0.026960785,0.025735294,0.025735294,0.030637255,0.024509804,0.030637255,0.040441178,0.042892158,0.04411765,0.04411765,0.037990198,0.040441178,0.06740196,0.0625,0.063725494,0.12867647,0.068627454,0.06740196,0.078431375,0.04534314,0.078431375,0.07720588,0.07230392,0.14215687,0.104166664,0.078431375,0.16176471,0.09191176,0.11764706,0.12990196,0.11519608,0.11764706,0.13480392,0.1360294,0.13664216,0.123774506,0.1360294,0.18137255,0.21936275,0.16176471,0.18014705,0.1764706,0.23161764,0.24387255,0.2254902,0.27328432,0.19485295,0.38602942,0.3872549,0.29166666,0.33210784,0.15686275,0.42647058,0.40931374,0.25306374,0.46813726,0.4240196,0.5392157,0.4117647,0.46078432,0.43259802,0.54534316,,0.61580884,0.60784316,0.85539216
+Linear (default),0.024509804,0.036764707,0.03982843,0.03370098,0.042892158,0.050857842,0.04840686,0.037990198,0.07965686,0.05637255,0.05882353,0.09558824,0.06495098,0.10784314,0.1004902,0.063725494,0.104166664,0.08210784,0.109068625,0.109068625,0.121323526,0.082720585,0.123774506,0.07720588,0.09620098,0.11029412,0.09436274,0.18627451,0.111519605,0.11519608,0.13725491,0.08210784,0.11764706,0.12990196,0.0992647,0.101715684,0.15073529,0.101715684,0.14338236,0.1127451,0.11764706,0.10294118,0.16911764,0.1004902,0.14460784,0.12990196,0.1764706,0.13480392,0.17769608,0.1942402,0.18627451,0.21078432,0.21078432,0.20465687,0.19730392,0.28308824,0.2389706,0.3247549,0.30147058,0.3382353,0.2990196,0.1875,0.25245097,0.30147058,0.30392158,0.25612745,0.3841912,,0.46568626,0.70710784
+KNN (tuned),0.028186275,0.034313727,0.035539217,0.035539217,0.039215688,0.05147059,0.04901961,0.05269608,0.05392157,0.042892158,0.0625,0.05637255,0.07230392,0.06495098,0.083333336,0.06556372,0.073529415,0.09191176,0.06985294,0.123774506,0.09129902,0.04656863,0.09313726,0.073529415,0.071078435,0.085171565,0.06740196,0.14828432,0.09436274,0.08946078,0.111519605,0.07230392,0.0992647,0.113970585,0.112132356,0.106617644,0.12683824,0.09068628,0.1360294,0.09313726,0.113970585,0.13970588,0.0870098,0.11029412,0.1360294,0.14338236,0.08210784,0.16789216,0.15563725,0.113970585,0.15073529,0.1740196,0.17892157,0.121323526,0.17279412,0.25490198,0.20465687,0.29166666,0.28308824,0.30882353,0.24264705,0.4509804,0.25490198,0.11335784,0.2512255,0.4632353,0.39215687,0.53431374,,0.9375
+KNN (default),0.006127451,0.004901961,0.006127451,0.004901961,0.007352941,0.008578432,0.013480392,0.014705882,0.006127451,0.009803922,0.009803922,0.006127451,0.024509804,0.008578432,0.017156864,0.013480392,0.012254902,0.031862747,0.013480392,0.031862747,0.017156864,0.006127451,0.007352941,0.018382354,0.011029412,0.020833334,0.015931373,0.041666668,0.022058824,0.023284314,0.014705882,0.013480392,0.04779412,0.02634804,0.04779412,0.040441178,0.037990198,0.019607844,0.037990198,0.020833334,0.05392157,0.04534314,0.008578432,0.05147059,0.039215688,0.05269608,0.009803922,0.05392157,0.05514706,0.028186275,0.04473039,0.09681372,0.09681372,0.029411765,0.06985294,0.06495098,0.10294118,0.07965686,0.085784316,0.12745099,0.028186275,0.24142157,0.0870098,0.025735294,0.05392157,0.24754901,0.14460784,0.29289216,0.0625,
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_medium/leaderboard_overview_explorer.html
index f8c36f958721dc20cd59b00998031224ecaeb8b1..7efbb1be60b9328bff30da17511273c94fe8ba03 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | All Tasks | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_all/tasks_all/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..d1b8aa1a1b8dbacbb77d14594915172595900d4e
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_all/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1986.0,"score":0.867,"rank":4.41,"hrank":1.69,"imp":2.683,"train_time":27.29,"predict_time":13.119,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+306\/-145","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1899.0,"score":0.71,"rank":6.33,"hrank":2.98,"imp":4.858,"train_time":191.18,"predict_time":2.207,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-49","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1794.0,"score":0.659,"rank":9.55,"hrank":3.26,"imp":5.878,"train_time":1.01,"predict_time":0.215,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+214\/-135","imputed":false},{"position":3,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1775.0,"score":0.579,"rank":10.23,"hrank":4.46,"imp":6.048,"train_time":432.36,"predict_time":4.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-51","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1722.0,"score":0.504,"rank":12.39,"hrank":8.78,"imp":7.846,"train_time":2.76,"predict_time":0.701,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-75","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1697.0,"score":0.541,"rank":13.53,"hrank":4.41,"imp":6.851,"train_time":0.6,"predict_time":0.119,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+186\/-110","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1646.0,"score":0.446,"rank":16.07,"hrank":8.48,"imp":8.576,"train_time":735.58,"predict_time":11.736,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+142\/-104","imputed":false},{"position":7,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1625.0,"score":0.412,"rank":17.17,"hrank":11.86,"imp":8.917,"train_time":1719.82,"predict_time":1.675,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-92","imputed":false},{"position":8,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1615.0,"score":0.397,"rank":17.73,"hrank":12.25,"imp":9.07,"train_time":2526.31,"predict_time":6.013,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-105","imputed":false},{"position":9,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1611.0,"score":0.379,"rank":17.97,"hrank":9.32,"imp":8.932,"train_time":489.93,"predict_time":0.249,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-63","imputed":false},{"position":10,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1600.0,"score":0.403,"rank":18.55,"hrank":10.47,"imp":9.218,"train_time":735.58,"predict_time":1.386,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+126\/-99","imputed":false},{"position":11,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1596.0,"score":0.372,"rank":18.8,"hrank":9.43,"imp":9.14,"train_time":489.93,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-66","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1589.0,"score":0.327,"rank":19.19,"hrank":13.5,"imp":10.466,"train_time":131.56,"predict_time":2.639,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-51","imputed":false},{"position":13,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1576.0,"score":0.378,"rank":19.95,"hrank":9.61,"imp":9.095,"train_time":2.47,"predict_time":0.01,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-101","imputed":false},{"position":14,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1558.0,"score":0.322,"rank":20.99,"hrank":15.67,"imp":10.221,"train_time":2067.95,"predict_time":65.195,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-52","imputed":false},{"position":15,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1548.0,"score":0.295,"rank":21.61,"hrank":14.86,"imp":10.546,"train_time":282.13,"predict_time":0.563,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-55","imputed":false},{"position":16,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1544.0,"score":0.292,"rank":21.87,"hrank":13.14,"imp":10.254,"train_time":1236.88,"predict_time":0.579,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-87","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1533.0,"score":0.301,"rank":22.52,"hrank":14.08,"imp":10.104,"train_time":1719.82,"predict_time":0.081,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-89","imputed":false},{"position":18,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1530.0,"score":0.324,"rank":22.69,"hrank":14.72,"imp":10.077,"train_time":1.88,"predict_time":0.645,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-79","imputed":false},{"position":19,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1526.0,"score":0.355,"rank":22.94,"hrank":10.98,"imp":10.809,"train_time":0.75,"predict_time":0.18,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-139","imputed":false},{"position":20,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1522.0,"score":0.319,"rank":23.22,"hrank":6.34,"imp":10.287,"train_time":4786.6,"predict_time":444.544,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-119","imputed":false},{"position":21,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1517.0,"score":0.314,"rank":23.53,"hrank":13.31,"imp":11.684,"train_time":95.92,"predict_time":0.297,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-121","imputed":false},{"position":22,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1506.0,"score":0.216,"rank":24.2,"hrank":18.44,"imp":11.603,"train_time":131.56,"predict_time":0.132,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-54","imputed":false},{"position":23,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1501.0,"score":0.234,"rank":24.53,"hrank":20.07,"imp":11.102,"train_time":282.13,"predict_time":0.066,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-55","imputed":false},{"position":24,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1481.0,"score":0.234,"rank":25.86,"hrank":15.63,"imp":11.089,"train_time":1236.88,"predict_time":0.054,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-76","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1472.0,"score":0.217,"rank":26.42,"hrank":20.56,"imp":11.215,"train_time":2526.31,"predict_time":0.349,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-81","imputed":false},{"position":26,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1470.0,"score":0.266,"rank":26.58,"hrank":15.47,"imp":12.146,"train_time":95.92,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-115","imputed":false},{"position":27,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1423.0,"score":0.224,"rank":29.66,"hrank":21.18,"imp":12.482,"train_time":566.74,"predict_time":0.951,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-114","imputed":false},{"position":28,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1423.0,"score":0.155,"rank":29.69,"hrank":18.8,"imp":12.112,"train_time":2067.95,"predict_time":10.881,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-63","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1418.0,"score":0.203,"rank":30.0,"hrank":19.33,"imp":13.176,"train_time":0.48,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-119","imputed":false},{"position":30,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1404.0,"score":0.221,"rank":30.97,"hrank":12.53,"imp":12.662,"train_time":4786.6,"predict_time":42.641,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+123\/-118","imputed":false},{"position":31,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1377.0,"score":0.128,"rank":32.79,"hrank":26.34,"imp":12.45,"train_time":1770.56,"predict_time":1.069,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-61","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1373.0,"score":0.159,"rank":33.01,"hrank":27.27,"imp":13.982,"train_time":566.74,"predict_time":0.06,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-101","imputed":false},{"position":33,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1372.0,"score":0.137,"rank":33.13,"hrank":21.73,"imp":12.907,"train_time":0.49,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-103","imputed":false},{"position":34,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1344.0,"score":0.157,"rank":34.99,"hrank":24.16,"imp":13.337,"train_time":0.51,"predict_time":0.127,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-102","imputed":false},{"position":35,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1332.0,"score":0.174,"rank":35.84,"hrank":16.46,"imp":15.311,"train_time":46.62,"predict_time":43.738,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+119\/-134","imputed":false},{"position":36,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1326.0,"score":0.096,"rank":36.23,"hrank":29.85,"imp":13.982,"train_time":2.13,"predict_time":0.047,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-88","imputed":false},{"position":37,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1326.0,"score":0.134,"rank":36.24,"hrank":27.26,"imp":15.595,"train_time":695.61,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-128","imputed":false},{"position":38,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1298.0,"score":0.073,"rank":38.1,"hrank":33.85,"imp":14.33,"train_time":8.06,"predict_time":0.25,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-79","imputed":false},{"position":39,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1279.0,"score":0.17,"rank":39.33,"hrank":9.92,"imp":16.418,"train_time":6.55,"predict_time":5.409,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-160","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1272.0,"score":0.049,"rank":39.78,"hrank":35.01,"imp":14.114,"train_time":1770.56,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-75","imputed":false},{"position":41,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1272.0,"score":0.034,"rank":39.84,"hrank":36.58,"imp":15.091,"train_time":6.75,"predict_time":0.235,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-68","imputed":false},{"position":42,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1265.0,"score":0.089,"rank":40.24,"hrank":32.83,"imp":16.441,"train_time":695.61,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-133","imputed":false},{"position":43,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1258.0,"score":0.028,"rank":40.73,"hrank":38.32,"imp":14.82,"train_time":0.29,"predict_time":0.045,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-64","imputed":false},{"position":44,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1244.0,"score":0.03,"rank":41.64,"hrank":35.97,"imp":15.1,"train_time":66.0,"predict_time":0.341,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-85","imputed":false},{"position":45,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1221.0,"score":0.057,"rank":43.14,"hrank":37.72,"imp":17.87,"train_time":2.18,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-136","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1217.0,"score":0.022,"rank":43.4,"hrank":40.85,"imp":15.304,"train_time":49.61,"predict_time":0.199,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-89","imputed":false},{"position":47,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1214.0,"score":0.031,"rank":43.54,"hrank":37.85,"imp":15.744,"train_time":66.0,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-84","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1201.0,"score":0.06,"rank":44.39,"hrank":36.49,"imp":17.192,"train_time":351.94,"predict_time":1.658,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-138","imputed":false},{"position":49,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1201.0,"score":0.028,"rank":44.41,"hrank":39.08,"imp":16.933,"train_time":11.44,"predict_time":10.416,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-95","imputed":false},{"position":50,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1179.0,"score":0.033,"rank":45.73,"hrank":41.33,"imp":18.488,"train_time":88.27,"predict_time":9.62,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-111","imputed":false},{"position":51,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1174.0,"score":0.008,"rank":46.07,"hrank":43.76,"imp":16.303,"train_time":49.61,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-83","imputed":false},{"position":52,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1126.0,"score":0.001,"rank":48.9,"hrank":47.16,"imp":18.145,"train_time":2.18,"predict_time":0.039,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-72","imputed":false},{"position":53,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1110.0,"score":0.019,"rank":49.83,"hrank":43.02,"imp":21.633,"train_time":20.26,"predict_time":0.515,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+101\/-125","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1101.0,"score":0.015,"rank":50.32,"hrank":45.23,"imp":19.294,"train_time":351.94,"predict_time":0.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-105","imputed":false},{"position":55,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1060.0,"score":0.004,"rank":52.57,"hrank":48.1,"imp":22.747,"train_time":20.26,"predict_time":0.2,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-116","imputed":false},{"position":56,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1046.0,"score":0.018,"rank":53.27,"hrank":47.13,"imp":25.98,"train_time":2.49,"predict_time":0.145,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+147\/-184","imputed":false},{"position":57,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1031.0,"score":0.018,"rank":54.04,"hrank":48.76,"imp":22.864,"train_time":1.51,"predict_time":0.111,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+136\/-171","imputed":false},{"position":58,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":55.55,"hrank":54.46,"imp":21.197,"train_time":0.2,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-92","imputed":false},{"position":59,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":989.0,"score":0.005,"rank":56.06,"hrank":49.75,"imp":24.353,"train_time":150.3,"predict_time":10.071,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-128","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":988.0,"score":0.0,"rank":56.13,"hrank":54.65,"imp":25.807,"train_time":3381.77,"predict_time":1.724,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-90","imputed":false},{"position":61,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":987.0,"score":0.0,"rank":56.18,"hrank":54.79,"imp":25.781,"train_time":3381.77,"predict_time":5.283,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-86","imputed":false},{"position":62,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":976.0,"score":0.0,"rank":56.64,"hrank":54.71,"imp":22.789,"train_time":0.48,"predict_time":0.073,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-121","imputed":false},{"position":63,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":935.0,"score":0.004,"rank":58.45,"hrank":53.87,"imp":32.872,"train_time":43.64,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+153\/-226","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":901.0,"score":0.0,"rank":59.81,"hrank":57.82,"imp":29.424,"train_time":1.5,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-153","imputed":false},{"position":65,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":892.0,"score":0.002,"rank":60.16,"hrank":56.48,"imp":33.73,"train_time":43.64,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+154\/-245","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":848.0,"score":0.0,"rank":61.7,"hrank":61.11,"imp":30.685,"train_time":110.78,"predict_time":1.964,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-88","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":814.0,"score":0.0,"rank":62.78,"hrank":62.5,"imp":29.973,"train_time":150.3,"predict_time":0.482,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-106","imputed":false},{"position":68,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":797.0,"score":0.0,"rank":63.27,"hrank":61.49,"imp":35.794,"train_time":0.19,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-244","imputed":false},{"position":69,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":593.0,"score":0.0,"rank":67.65,"hrank":67.52,"imp":43.501,"train_time":0.1,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-146","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_medium/pareto_front_explorer.html
index ce8c550e1916c282e84f575941fc5a3030cb7df4..15a350f91b1b763e5d984899cd3cea99ad3bb168 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0102827644,"imp":9.0952095474,"elo":1575.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0196975147,"imp":9.1399516875,"elo":1595.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2485703935,"imp":8.9317687118,"elo":1610.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0225370945,"imp":13.175735213,"elo":1418.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0265970701,"imp":12.1456953066,"elo":1469.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2967122121,"imp":11.6842157983,"elo":1517.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"imp":17.8697973966,"elo":1220.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0061458745,"imp":16.4411619269,"elo":1265.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0851920607,"imp":15.5948171656,"elo":1325.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0733045085,"imp":22.7894839202,"elo":976.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0153132761,"imp":16.3030865094,"elo":1173.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1990111469,"imp":15.3042054388,"elo":1216.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"imp":22.8643090483,"elo":1031.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1141755222,"imp":19.2937753734,"elo":1101.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6581625315,"imp":17.1916474397,"elo":1200.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0447579838,"imp":14.819832856,"elo":1258.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1322357268,"imp":11.6033224024,"elo":1506.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"imp":10.4658385154,"elo":1588.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.4085730775,"imp":16.4177044552,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"imp":14.3302915146,"elo":1297.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3490410688,"imp":11.2147479695,"elo":1472.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.0133340655,"imp":9.0700293864,"elo":1614.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0149545776,"imp":29.4235354386,"elo":900.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1995710638,"imp":22.7471582873,"elo":1059.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5149235805,"imp":21.6330567052,"elo":1109.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.020899149,"imp":21.1973502954,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0243850814,"imp":15.7438782463,"elo":1214.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3407391303,"imp":15.0995978866,"elo":1244.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"imp":15.0910499881,"elo":1271.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0807033173,"imp":10.103937795,"elo":1533.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.6746674551,"imp":8.9170097602,"elo":1625.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"imp":10.077321119,"elo":1530.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3859818406,"imp":9.2178209823,"elo":1600.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.7359861516,"imp":8.5761380952,"elo":1645.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.4162023366,"imp":16.9330805211,"elo":1200.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"imp":15.3106058247,"elo":1331.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.6414441586,"imp":12.6616447819,"elo":1403.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":444.5439481393,"imp":10.2874153003,"elo":1522.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1795394575,"imp":10.8085465808,"elo":1526.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":13.1188916889,"imp":2.6827874305,"elo":1985.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1186281919,"imp":6.8505424386,"elo":1697.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0470610966,"imp":13.9816848431,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0538978015,"imp":11.0894895039,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5790353938,"imp":10.2538835957,"elo":1543.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"imp":7.8460750331,"elo":1722.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2147265792,"imp":5.8782765424,"elo":1793.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"imp":30.6853633341,"elo":847.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.723789106,"imp":25.807196939,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"imp":25.7806935994,"elo":986.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1271670825,"imp":13.3366678662,"elo":1344.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0388490278,"imp":18.1452782924,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.050592863,"imp":14.1135657835,"elo":1272.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0691781292,"imp":12.4495254577,"elo":1376.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"imp":12.9074172378,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0664942176,"imp":11.1023313966,"elo":1501.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.5626253891,"imp":10.5460036421,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":9.6204056883,"imp":18.4877413347,"elo":1179.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.8813375711,"imp":12.1121479202,"elo":1422.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.1951531214,"imp":10.2205385819,"elo":1558.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1448756257,"imp":25.9796305322,"elo":1046.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0599221336,"imp":13.9821871779,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.9511369135,"imp":12.4822741841,"elo":1423.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0102827644,"x_train":2.4694138287,"imp":9.0952095474,"elo":1575.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0196975147,"x_train":489.9274101037,"imp":9.1399516875,"elo":1595.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2485703935,"x_train":489.9274101037,"imp":8.9317687118,"elo":1610.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0225370945,"x_train":0.4789788316,"imp":13.175735213,"elo":1418.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0265970701,"x_train":95.9169155889,"imp":12.1456953066,"elo":1469.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2967122121,"x_train":95.9169155889,"imp":11.6842157983,"elo":1517.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"x_train":2.1751921611,"imp":17.8697973966,"elo":1220.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0061458745,"x_train":695.6092947225,"imp":16.4411619269,"elo":1265.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0851920607,"x_train":695.6092947225,"imp":15.5948171656,"elo":1325.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0733045085,"x_train":0.4842583839,"imp":22.7894839202,"elo":976.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0153132761,"x_train":49.6127891307,"imp":16.3030865094,"elo":1173.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1990111469,"x_train":49.6127891307,"imp":15.3042054388,"elo":1216.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"x_train":1.5145310857,"imp":22.8643090483,"elo":1031.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1141755222,"x_train":351.9412405462,"imp":19.2937753734,"elo":1101.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6581625315,"x_train":351.9412405462,"imp":17.1916474397,"elo":1200.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0447579838,"x_train":0.2873206973,"imp":14.819832856,"elo":1258.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1322357268,"x_train":131.5620070704,"imp":11.6033224024,"elo":1506.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"x_train":131.5620070704,"imp":10.4658385154,"elo":1588.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.4085730775,"x_train":6.5532925928,"imp":16.4177044552,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"x_train":8.0637708772,"imp":14.3302915146,"elo":1297.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3490410688,"x_train":2526.3066806002,"imp":11.2147479695,"elo":1472.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.0133340655,"x_train":2526.3066806002,"imp":9.0700293864,"elo":1614.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0149545776,"x_train":1.4985629713,"imp":29.4235354386,"elo":900.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1995710638,"x_train":20.2619047675,"imp":22.7471582873,"elo":1059.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5149235805,"x_train":20.2619047675,"imp":21.6330567052,"elo":1109.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.020899149,"x_train":0.1991527253,"imp":21.1973502954,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0243850814,"x_train":65.9990639089,"imp":15.7438782463,"elo":1214.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3407391303,"x_train":65.9990639089,"imp":15.0995978866,"elo":1244.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"x_train":6.7532462417,"imp":15.0910499881,"elo":1271.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0807033173,"x_train":1719.8202901016,"imp":10.103937795,"elo":1533.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.6746674551,"x_train":1719.8202901016,"imp":8.9170097602,"elo":1625.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"x_train":1.8760518371,"imp":10.077321119,"elo":1530.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3859818406,"x_train":735.5848736649,"imp":9.2178209823,"elo":1600.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.7359861516,"x_train":735.5848736649,"imp":8.5761380952,"elo":1645.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.4162023366,"x_train":11.4380569036,"imp":16.9330805211,"elo":1200.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"x_train":46.6209023207,"imp":15.3106058247,"elo":1331.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.6414441586,"x_train":4786.5989331939,"imp":12.6616447819,"elo":1403.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":444.5439481393,"x_train":4786.5989331939,"imp":10.2874153003,"elo":1522.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1795394575,"x_train":0.7535798582,"imp":10.8085465808,"elo":1526.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":13.1188916889,"x_train":27.2884140106,"imp":2.6827874305,"elo":1985.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1186281919,"x_train":0.6015671442,"imp":6.8505424386,"elo":1697.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0470610966,"x_train":2.1306936636,"imp":13.9816848431,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0538978015,"x_train":1236.8778796591,"imp":11.0894895039,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5790353938,"x_train":1236.8778796591,"imp":10.2538835957,"elo":1543.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"x_train":2.7558592937,"imp":7.8460750331,"elo":1722.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2147265792,"x_train":1.0099025842,"imp":5.8782765424,"elo":1793.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"x_train":110.7816266415,"imp":30.6853633341,"elo":847.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.723789106,"x_train":3381.7712162192,"imp":25.807196939,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"x_train":3381.7712162192,"imp":25.7806935994,"elo":986.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1271670825,"x_train":0.5069992699,"imp":13.3366678662,"elo":1344.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0388490278,"x_train":2.1848199639,"imp":18.1452782924,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.050592863,"x_train":1770.5620854906,"imp":14.1135657835,"elo":1272.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0691781292,"x_train":1770.5620854906,"imp":12.4495254577,"elo":1376.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"x_train":0.4869354634,"imp":12.9074172378,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0664942176,"x_train":282.1320041804,"imp":11.1023313966,"elo":1501.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.5626253891,"x_train":282.1320041804,"imp":10.5460036421,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":9.6204056883,"x_train":88.2701787746,"imp":18.4877413347,"elo":1179.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.8813375711,"x_train":2067.9528996121,"imp":12.1121479202,"elo":1422.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.1951531214,"x_train":2067.9528996121,"imp":10.2205385819,"elo":1558.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1448756257,"x_train":2.4940806965,"imp":25.9796305322,"elo":1046.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0599221336,"x_train":566.7369825686,"imp":13.9821871779,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.9511369135,"x_train":566.7369825686,"imp":12.4822741841,"elo":1423.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_medium/tuning_trajectories_explorer.html
index 1b2c5db16e86fd9c3d8087f2ce7da5471b7125f0..b9316dfa275f52bd2ed35a1fbbd3a0c3cf628866 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":2.4694138287,"imp":9.1083854124,"elo":1592.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":4.4162445685,"imp":9.1159938977,"elo":1597.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.8766122224,"imp":9.0779495775,"elo":1614.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":24.258974221,"imp":9.0588594541,"elo":1619.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":62.0516636378,"imp":8.9467666446,"elo":1630.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":122.5800174334,"imp":8.9070623506,"elo":1636.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":242.6410763635,"imp":8.9234658367,"elo":1635.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":361.1283580261,"imp":8.9312030363,"elo":1634.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":489.9274101037,"imp":8.939773823,"elo":1632.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4789788316,"imp":13.1894545746,"elo":1436.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.9589536975,"imp":12.7804689305,"elo":1466.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.356217589,"imp":12.4744183072,"elo":1487.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.7221268742,"imp":12.2391809253,"elo":1504.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.9255203586,"imp":12.0497339763,"elo":1519.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":23.8768976028,"imp":11.913101781,"elo":1529.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":47.751833203,"imp":11.7755666991,"elo":1538.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":71.5860615637,"imp":11.7150159217,"elo":1542.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":95.9169155889,"imp":11.6957695725,"elo":1545.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.1751921611,"imp":17.8803428496,"elo":1226.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":6.6636925925,"imp":17.1941915211,"elo":1252.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":17.2339258054,"imp":16.6116876896,"elo":1279.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":34.5777558766,"imp":16.3560470223,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":87.9245165634,"imp":16.0521398506,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":174.9652611714,"imp":15.8994205919,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":345.0671578857,"imp":15.7386714692,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":520.7521321905,"imp":15.6825400755,"elo":1338.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":695.6092947225,"imp":15.6013580723,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.4842583839,"imp":22.7998895097,"elo":973.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.846613752,"imp":20.6597293066,"elo":1029.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.6923175211,"imp":17.4621722506,"elo":1121.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9291798058,"imp":16.3110647365,"elo":1167.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.6147364724,"imp":15.5601823286,"elo":1201.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.7108167547,"imp":15.3929982195,"elo":1213.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.9416823722,"imp":15.2669930228,"elo":1223.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.1384318226,"imp":15.2655490595,"elo":1226.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":49.6127891307,"imp":15.3160959667,"elo":1228.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.5145310857,"imp":22.8738415301,"elo":1039.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.1588915157,"imp":21.8537906596,"elo":1065.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.5956312383,"imp":20.5238950426,"elo":1113.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":17.7818624821,"imp":19.7509391562,"elo":1138.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":42.5300885718,"imp":18.7952090418,"elo":1166.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":85.6762822632,"imp":18.216855293,"elo":1183.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":173.75989872,"imp":17.7296246052,"elo":1197.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":261.7851807625,"imp":17.4018458318,"elo":1210.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":351.9412405462,"imp":17.1977570828,"elo":1219.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.2873206973,"imp":14.831406831,"elo":1265.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.9316502247,"imp":14.2272850125,"elo":1316.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":2.7286702854,"imp":13.0771089423,"elo":1416.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":6.1477936876,"imp":12.2165878048,"elo":1486.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":16.5209724855,"imp":11.2062569778,"elo":1552.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":32.6797765909,"imp":10.7760390499,"elo":1588.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":65.3538276873,"imp":10.5834838941,"elo":1603.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":98.4212833586,"imp":10.5224169036,"elo":1608.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":131.5620070704,"imp":10.4847692361,"elo":1615.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":6.5532925928,"imp":16.1415758528,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":8.0637708772,"imp":14.341711205,"elo":1311.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":21.7538395214,"imp":12.8782704772,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.9177628832,"imp":11.027959171,"elo":1499.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":121.8592858048,"imp":10.3441528707,"elo":1546.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":292.3137710756,"imp":9.6118899681,"elo":1599.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":629.8900400305,"imp":9.3077354843,"elo":1621.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1243.5591974675,"imp":9.1817703032,"elo":1633.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1862.5519447854,"imp":9.1222507032,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2526.3066806002,"imp":9.0917386137,"elo":1640.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.4985629713,"imp":29.6890705123,"elo":914.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.4317773957,"imp":27.150126118,"elo":995.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":16.9493056602,"imp":21.7611748075,"elo":1110.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.2619047675,"imp":21.637241668,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1991527253,"imp":21.2131448299,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.5021873111,"imp":17.6947708454,"elo":1106.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.4894550512,"imp":16.2525252885,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":3.1307773192,"imp":15.7036900771,"elo":1208.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":8.0202567964,"imp":15.3404069654,"elo":1234.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":16.2377871584,"imp":15.2141741302,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":32.6710934428,"imp":15.1267895029,"elo":1254.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":49.1719658028,"imp":15.0853480414,"elo":1259.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":65.9990639089,"imp":15.1111406521,"elo":1262.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.7532462417,"imp":15.1005671513,"elo":1281.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.5275786908,"imp":12.9072260047,"elo":1413.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.1241530408,"imp":11.3635581269,"elo":1512.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":82.7422408171,"imp":10.5674293143,"elo":1553.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":213.2761385894,"imp":9.8481783226,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":427.9091537602,"imp":9.5297368118,"elo":1614.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":848.9386425085,"imp":9.1601227196,"elo":1634.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1278.8937139162,"imp":9.0036425298,"elo":1645.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1719.8202901016,"imp":8.9287868289,"elo":1654.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.8760518371,"imp":10.0918725787,"elo":1542.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.1278647155,"imp":10.018663163,"elo":1552.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.2069927888,"imp":9.8482067878,"elo":1566.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":34.2536042268,"imp":9.7015761863,"elo":1583.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":88.9628219424,"imp":9.3584259435,"elo":1623.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":181.3952243454,"imp":9.000340936,"elo":1645.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":363.9172361826,"imp":8.7779080241,"elo":1656.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":547.7650992279,"imp":8.6576561267,"elo":1660.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":735.5848736649,"imp":8.5856394282,"elo":1660.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":11.4380569036,"imp":16.9413141811,"elo":1215.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.6209023207,"imp":15.3200410441,"elo":1343.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.0348846629,"imp":14.1956495304,"elo":1371.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":129.2960112707,"imp":13.0943832644,"elo":1413.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":230.7293949461,"imp":12.6784457778,"elo":1431.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":589.0616829286,"imp":11.7243258163,"elo":1477.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1187.7578308406,"imp":11.1912530117,"elo":1502.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2376.2279314343,"imp":10.8135073436,"elo":1519.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3574.1919132683,"imp":10.4831511393,"elo":1533.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4786.5989331939,"imp":10.296860814,"elo":1538.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.7535798582,"imp":10.8199022081,"elo":1540.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":27.2884140106,"imp":2.6842884759,"elo":1990.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6015671442,"imp":6.8659561097,"elo":1713.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":2.1306936636,"imp":13.9939671673,"elo":1336.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.7201323021,"imp":13.1386024727,"elo":1384.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":26.071931656,"imp":12.07671084,"elo":1440.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":56.6403905326,"imp":11.465419546,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":151.6325084544,"imp":10.9150875666,"elo":1514.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":305.4050159003,"imp":10.6347881155,"elo":1536.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":598.166050268,"imp":10.4092043336,"elo":1551.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":909.6246432443,"imp":10.3205639671,"elo":1559.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1236.8778796591,"imp":10.2623069414,"elo":1568.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.7558592937,"imp":7.8605289769,"elo":1746.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.0099025842,"imp":5.8834864036,"elo":1798.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":110.7816266415,"imp":30.6917614221,"elo":842.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":125.5711234695,"imp":30.609802755,"elo":858.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":166.4991276653,"imp":30.4571522335,"elo":864.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":216.8260972869,"imp":30.0179135156,"elo":881.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":435.6268489691,"imp":29.0532719309,"elo":906.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":841.7488888841,"imp":27.8654166105,"elo":933.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1663.6865760139,"imp":27.0808278838,"elo":956.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2487.2049124756,"imp":26.3846780548,"elo":981.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3381.7712162192,"imp":25.7887166563,"elo":1007.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.5069992699,"imp":13.3506819035,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":2.1848199639,"imp":18.1535807635,"elo":1135.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.222603335,"imp":16.9138550883,"elo":1182.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":41.2363997061,"imp":15.5421029397,"elo":1240.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":91.3237652004,"imp":14.7262690578,"elo":1284.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":218.5473202022,"imp":13.8435434487,"elo":1325.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":447.0507548682,"imp":13.3314756024,"elo":1355.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":866.7256429369,"imp":12.9284009506,"elo":1377.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1331.6601195387,"imp":12.685478784,"elo":1390.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1770.5620854906,"imp":12.4629978646,"elo":1401.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4869354634,"imp":12.9207034717,"elo":1387.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.4341782987,"imp":11.9175965636,"elo":1466.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":7.341922179,"imp":11.3659122277,"elo":1515.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":13.4253918933,"imp":11.0012314263,"elo":1536.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":35.1370362218,"imp":10.7494419639,"elo":1552.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":69.8388019133,"imp":10.6449945049,"elo":1560.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":138.6843072833,"imp":10.5779558918,"elo":1567.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":209.6003170598,"imp":10.5660972584,"elo":1571.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":282.1320041804,"imp":10.5606176671,"elo":1574.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":88.2701787746,"imp":18.4983781039,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":168.2367198699,"imp":16.3890454014,"elo":1247.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":410.883351941,"imp":16.2186243252,"elo":1345.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":805.5219751868,"imp":13.5594974448,"elo":1442.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1992.2815124104,"imp":10.2704282323,"elo":1578.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":2067.9528996121,"imp":10.2338269266,"elo":1584.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.4940806965,"imp":25.9875577867,"elo":1051.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.3133350441,"imp":21.3468517179,"elo":1127.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":13.8020981623,"imp":16.1353961932,"elo":1255.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":28.3270720096,"imp":13.9002791385,"elo":1372.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":70.6047440508,"imp":12.9743181235,"elo":1424.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":140.5453323732,"imp":12.6248944307,"elo":1441.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":281.1244162931,"imp":12.4891082036,"elo":1446.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":422.3532460518,"imp":12.4480188978,"elo":1447.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":566.7369825686,"imp":12.4628187149,"elo":1446.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0102827644,"x_train":2.4694138287,"imp":9.1083854124,"elo":1592.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0406807345,"x_train":4.4162445685,"imp":9.1159938977,"elo":1597.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0688636844,"x_train":12.8766122224,"imp":9.0779495775,"elo":1614.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0926006369,"x_train":24.258974221,"imp":9.0588594541,"elo":1619.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1480097263,"x_train":62.0516636378,"imp":8.9467666446,"elo":1630.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1788978917,"x_train":122.5800174334,"imp":8.9070623506,"elo":1636.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2123832568,"x_train":242.6410763635,"imp":8.9234658367,"elo":1635.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.228573636,"x_train":361.1283580261,"imp":8.9312030363,"elo":1634.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2570153277,"x_train":489.9274101037,"imp":8.939773823,"elo":1632.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0225370945,"x_train":0.4789788316,"imp":13.1894545746,"elo":1436.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0470795501,"x_train":0.9589536975,"imp":12.7804689305,"elo":1466.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0965149191,"x_train":2.356217589,"imp":12.4744183072,"elo":1487.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.14469563,"x_train":4.7221268742,"imp":12.2391809253,"elo":1504.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2016668384,"x_train":11.9255203586,"imp":12.0497339763,"elo":1519.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2410485968,"x_train":23.8768976028,"imp":11.913101781,"elo":1529.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2647783876,"x_train":47.751833203,"imp":11.7755666991,"elo":1538.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2890240067,"x_train":71.5860615637,"imp":11.7150159217,"elo":1542.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3064184086,"x_train":95.9169155889,"imp":11.6957695725,"elo":1545.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"x_train":2.1751921611,"imp":17.8803428496,"elo":1226.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0116608277,"x_train":6.6636925925,"imp":17.1941915211,"elo":1252.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0226990233,"x_train":17.2339258054,"imp":16.6116876896,"elo":1279.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333135949,"x_train":34.5777558766,"imp":16.3560470223,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0497571688,"x_train":87.9245165634,"imp":16.0521398506,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0593008852,"x_train":174.9652611714,"imp":15.8994205919,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0709126161,"x_train":345.0671578857,"imp":15.7386714692,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0763826625,"x_train":520.7521321905,"imp":15.6825400755,"elo":1338.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0851920607,"x_train":695.6092947225,"imp":15.6013580723,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0733045085,"x_train":0.4842583839,"imp":22.7998895097,"elo":973.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0752221241,"x_train":0.846613752,"imp":20.6597293066,"elo":1029.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0926460136,"x_train":1.6923175211,"imp":17.4621722506,"elo":1121.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1010934443,"x_train":2.9291798058,"imp":16.3110647365,"elo":1167.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1082818488,"x_train":6.6147364724,"imp":15.5601823286,"elo":1201.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.143858289,"x_train":12.7108167547,"imp":15.3929982195,"elo":1213.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1608782018,"x_train":24.9416823722,"imp":15.2669930228,"elo":1223.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1877156331,"x_train":37.1384318226,"imp":15.2655490595,"elo":1226.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1990111469,"x_train":49.6127891307,"imp":15.3160959667,"elo":1228.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"x_train":1.5145310857,"imp":22.8738415301,"elo":1039.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2219956901,"x_train":3.1588915157,"imp":21.8537906596,"elo":1065.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4051013545,"x_train":8.5956312383,"imp":20.5238950426,"elo":1113.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6441505685,"x_train":17.7818624821,"imp":19.7509391562,"elo":1138.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9551913221,"x_train":42.5300885718,"imp":18.7952090418,"elo":1166.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2793194595,"x_train":85.6762822632,"imp":18.216855293,"elo":1183.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.5591641512,"x_train":173.75989872,"imp":17.7296246052,"elo":1197.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6900218907,"x_train":261.7851807625,"imp":17.4018458318,"elo":1210.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6589060596,"x_train":351.9412405462,"imp":17.1977570828,"elo":1219.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0447579838,"x_train":0.2873206973,"imp":14.831406831,"elo":1265.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4234159248,"x_train":0.9316502247,"imp":14.2272850125,"elo":1316.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7433885464,"x_train":2.7286702854,"imp":13.0771089423,"elo":1416.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9410923145,"x_train":6.1477936876,"imp":12.2165878048,"elo":1486.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3202136893,"x_train":16.5209724855,"imp":11.2062569778,"elo":1552.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4195738396,"x_train":32.6797765909,"imp":10.7760390499,"elo":1588.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.0760086735,"x_train":65.3538276873,"imp":10.5834838941,"elo":1603.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2773590032,"x_train":98.4212833586,"imp":10.5224169036,"elo":1608.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"x_train":131.5620070704,"imp":10.4847692361,"elo":1615.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":5.4085730775,"x_train":6.5532925928,"imp":16.1415758528,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"x_train":8.0637708772,"imp":14.341711205,"elo":1311.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5785744608,"x_train":21.7538395214,"imp":12.8782704772,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.2947480655,"x_train":60.9177628832,"imp":11.027959171,"elo":1499.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1427076773,"x_train":121.8592858048,"imp":10.3441528707,"elo":1546.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7736882796,"x_train":292.3137710756,"imp":9.6118899681,"elo":1599.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.9630762477,"x_train":629.8900400305,"imp":9.3077354843,"elo":1621.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1679213756,"x_train":1243.5591974675,"imp":9.1817703032,"elo":1633.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.8768173804,"x_train":1862.5519447854,"imp":9.1222507032,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.10416073,"x_train":2526.3066806002,"imp":9.0917386137,"elo":1640.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0149545776,"x_train":1.4985629713,"imp":29.6890705123,"elo":914.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1460145839,"x_train":6.4317773957,"imp":27.150126118,"elo":995.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4447962725,"x_train":16.9493056602,"imp":21.7611748075,"elo":1110.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5149235805,"x_train":20.2619047675,"imp":21.637241668,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.020899149,"x_train":0.1991527253,"imp":21.2131448299,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.045901974,"x_train":0.5021873111,"imp":17.6947708454,"elo":1106.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1128248591,"x_train":1.4894550512,"imp":16.2525252885,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1367392873,"x_train":3.1307773192,"imp":15.7036900771,"elo":1208.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2025401714,"x_train":8.0202567964,"imp":15.3404069654,"elo":1234.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2523558649,"x_train":16.2377871584,"imp":15.2141741302,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2972209594,"x_train":32.6710934428,"imp":15.1267895029,"elo":1254.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3310669279,"x_train":49.1719658028,"imp":15.0853480414,"elo":1259.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3414833614,"x_train":65.9990639089,"imp":15.1111406521,"elo":1262.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"x_train":6.7532462417,"imp":15.1005671513,"elo":1281.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3080259999,"x_train":15.5275786908,"imp":12.9072260047,"elo":1413.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5106934439,"x_train":39.1241530408,"imp":11.3635581269,"elo":1512.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.7594361579,"x_train":82.7422408171,"imp":10.5674293143,"elo":1553.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.9843404045,"x_train":213.2761385894,"imp":9.8481783226,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1860326968,"x_train":427.9091537602,"imp":9.5297368118,"elo":1614.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4580786339,"x_train":848.9386425085,"imp":9.1601227196,"elo":1634.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5889939537,"x_train":1278.8937139162,"imp":9.0036425298,"elo":1645.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6682080328,"x_train":1719.8202901016,"imp":8.9287868289,"elo":1654.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"x_train":1.8760518371,"imp":10.0918725787,"elo":1542.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4937099645,"x_train":5.1278647155,"imp":10.018663163,"elo":1552.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2513268047,"x_train":16.2069927888,"imp":9.8482067878,"elo":1566.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.9918356458,"x_train":34.2536042268,"imp":9.7015761863,"elo":1583.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4620925312,"x_train":88.9628219424,"imp":9.3584259435,"elo":1623.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.4114827703,"x_train":181.3952243454,"imp":9.000340936,"elo":1645.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6236180593,"x_train":363.9172361826,"imp":8.7779080241,"elo":1656.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.9955169691,"x_train":547.7650992279,"imp":8.6576561267,"elo":1660.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.7359861516,"x_train":735.5848736649,"imp":8.5856394282,"elo":1660.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.4162023366,"x_train":11.4380569036,"imp":16.9413141811,"elo":1215.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"x_train":46.6209023207,"imp":15.3200410441,"elo":1343.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":60.7634572276,"x_train":70.0348846629,"imp":14.1956495304,"elo":1371.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":97.7966669957,"x_train":129.2960112707,"imp":13.0943832644,"elo":1413.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":126.1317081836,"x_train":230.7293949461,"imp":12.6784457778,"elo":1431.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":211.892083719,"x_train":589.0616829286,"imp":11.7243258163,"elo":1477.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":283.7834460479,"x_train":1187.7578308406,"imp":11.1912530117,"elo":1502.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":356.1524517375,"x_train":2376.2279314343,"imp":10.8135073436,"elo":1519.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":402.585313069,"x_train":3574.1919132683,"imp":10.4831511393,"elo":1533.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":433.0154440766,"x_train":4786.5989331939,"imp":10.296860814,"elo":1538.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1795394575,"x_train":0.7535798582,"imp":10.8199022081,"elo":1540.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":13.1188916889,"x_train":27.2884140106,"imp":2.6842884759,"elo":1990.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1186281919,"x_train":0.6015671442,"imp":6.8659561097,"elo":1713.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0470610966,"x_train":2.1306936636,"imp":13.9939671673,"elo":1336.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1016642016,"x_train":6.7201323021,"imp":13.1386024727,"elo":1384.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.1965833708,"x_train":26.071931656,"imp":12.07671084,"elo":1440.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2690374607,"x_train":56.6403905326,"imp":11.465419546,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3453390813,"x_train":151.6325084544,"imp":10.9150875666,"elo":1514.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4641659542,"x_train":305.4050159003,"imp":10.6347881155,"elo":1536.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.5371776272,"x_train":598.166050268,"imp":10.4092043336,"elo":1551.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.5567799537,"x_train":909.6246432443,"imp":10.3205639671,"elo":1559.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5682121276,"x_train":1236.8778796591,"imp":10.2623069414,"elo":1568.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"x_train":2.7558592937,"imp":7.8605289769,"elo":1746.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2147265792,"x_train":1.0099025842,"imp":5.8834864036,"elo":1798.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"x_train":110.7816266415,"imp":30.6917614221,"elo":842.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9334778523,"x_train":125.5711234695,"imp":30.609802755,"elo":858.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1406475162,"x_train":166.4991276653,"imp":30.4571522335,"elo":864.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.4886555763,"x_train":216.8260972869,"imp":30.0179135156,"elo":881.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.967901931,"x_train":435.6268489691,"imp":29.0532719309,"elo":906.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4310027255,"x_train":841.7488888841,"imp":27.8654166105,"elo":933.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.4436810211,"x_train":1663.6865760139,"imp":27.0808278838,"elo":956.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9435530344,"x_train":2487.2049124756,"imp":26.3846780548,"elo":981.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"x_train":3381.7712162192,"imp":25.7887166563,"elo":1007.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1271670825,"x_train":0.5069992699,"imp":13.3506819035,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0388490278,"x_train":2.1848199639,"imp":18.1535807635,"elo":1135.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0914656589,"x_train":9.222603335,"imp":16.9138550883,"elo":1182.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2165503089,"x_train":41.2363997061,"imp":15.5421029397,"elo":1240.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.357854022,"x_train":91.3237652004,"imp":14.7262690578,"elo":1284.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5218944893,"x_train":218.5473202022,"imp":13.8435434487,"elo":1325.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7429074477,"x_train":447.0507548682,"imp":13.3314756024,"elo":1355.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9478582394,"x_train":866.7256429369,"imp":12.9284009506,"elo":1377.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0346365316,"x_train":1331.6601195387,"imp":12.685478784,"elo":1390.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0691781292,"x_train":1770.5620854906,"imp":12.4629978646,"elo":1401.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"x_train":0.4869354634,"imp":12.9207034717,"elo":1387.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1793184212,"x_train":2.4341782987,"imp":11.9175965636,"elo":1466.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2826887094,"x_train":7.341922179,"imp":11.3659122277,"elo":1515.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2968304513,"x_train":13.4253918933,"imp":11.0012314263,"elo":1536.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4028365541,"x_train":35.1370362218,"imp":10.7494419639,"elo":1552.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4362798234,"x_train":69.8388019133,"imp":10.6449945049,"elo":1560.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4803960764,"x_train":138.6843072833,"imp":10.5779558918,"elo":1567.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5237722573,"x_train":209.6003170598,"imp":10.5660972584,"elo":1571.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5608426218,"x_train":282.1320041804,"imp":10.5606176671,"elo":1574.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":9.6204056883,"x_train":88.2701787746,"imp":18.4983781039,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":18.9558180407,"x_train":168.2367198699,"imp":16.3890454014,"elo":1247.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":33.2102432394,"x_train":410.883351941,"imp":16.2186243252,"elo":1345.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":46.9242331846,"x_train":805.5219751868,"imp":13.5594974448,"elo":1442.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":66.5740410026,"x_train":1992.2815124104,"imp":10.2704282323,"elo":1578.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":65.1951531214,"x_train":2067.9528996121,"imp":10.2338269266,"elo":1584.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1448756257,"x_train":2.4940806965,"imp":25.9875577867,"elo":1051.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.1903662412,"x_train":5.3133350441,"imp":21.3468517179,"elo":1127.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.2632687113,"x_train":13.8020981623,"imp":16.1353961932,"elo":1255.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.381885611,"x_train":28.3270720096,"imp":13.9002791385,"elo":1372.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.5413359982,"x_train":70.6047440508,"imp":12.9743181235,"elo":1424.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.6886364123,"x_train":140.5453323732,"imp":12.6248944307,"elo":1441.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.8346908135,"x_train":281.1244162931,"imp":12.4891082036,"elo":1446.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":0.8992040717,"x_train":422.3532460518,"imp":12.4480188978,"elo":1447.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":0.9679699005,"x_train":566.7369825686,"imp":12.4628187149,"elo":1446.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..bb1c1cec7a31fab8c6b4cbf2d9d3b432f37a9a26
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_all/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "AutoGluon 1.4 (best, 4h)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealMLP (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "CatBoost (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "CatBoost (tuned)", "LightGBM (tuned + ensembled)", "CatBoost (default)", "iLTM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabM (tuned + ensembled)", "RealMLP (tuned)", "RealTabPFN-2.5 (default)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "LightGBM (tuned)", "XGBoost (tuned)", "TabM (tuned)", "ModernNCA (tuned)", "ChimeraBoost (tuned)", "xRFM (tuned + ensembled)", "iLTM (tuned)", "ChimeraBoost (default)", "TabDPT (tuned)", "TorchMLP (tuned + ensembled)", "xRFM (tuned)", "XGBoost (default)", "TabSwift (default)", "TabDPT (default)", "TabM (default)", "EBM (tuned + ensembled)", "ModernNCA (default)", "LimiX (default)", "TorchMLP (tuned)", "RealMLP (default)", "EBM (tuned)", "LightGBM (default)", "RandomForest (tuned + ensembled)", "EBM (default)", "ExtraTrees (tuned + ensembled)", "RandomForest (tuned)", "FastaiMLP (tuned + ensembled)", "SAP-RPT-OSS (default)", "iLTM (default)", "ExtraTrees (tuned)", "TorchMLP (default)", "PerpetualBooster (tuned + ensembled)", "FastaiMLP (tuned)", "PerpetualBooster (tuned)", "xRFM (default)", "FastaiMLP (default)", "RandomForest (default)", "KNN (tuned + ensembled)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned)", "TabSTAR (default)", "KNN (tuned)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.6666666865348816, 0.7851851582527161, 0.7555555701255798, 0.9333333373069763, 0.9037036895751953, 0.9111111164093018, 0.8814814686775208, 0.9259259104728699, 0.8518518805503845, 0.9185185432434082, 0.8814814686775208, 0.9037036895751953, 0.8592592477798462, 0.8962963223457336, 0.9185185432434082, 0.9925925731658936, 0.8888888955116272, 0.9851852059364319, 0.9629629850387573, 0.8740741014480591, 0.9185185432434082, 0.9777777791023254, 0.9407407641410828, 1.0, 0.9629629850387573, 0.9185185432434082, 0.9259259104728699, 0.9333333373069763, 0.9333333373069763, 0.9259259104728699, 0.9777777791023254, 0.9407407641410828, 0.9851852059364319, 0.970370352268219, 0.9407407641410828, 1.0, 0.9925925731658936, 1.0, 0.9037036895751953, 1.0, 1.0, 1.0, 1.0, 0.9851852059364319, 1.0, 0.9925925731658936, 0.9851852059364319, 0.9407407641410828, 0.9851852059364319, 1.0, 0.9925925731658936, 1.0, 1.0, 0.9851852059364319, 1.0, 0.970370352268219, 0.9851852059364319, 0.9925925731658936, 0.970370352268219, 1.0, 1.0, 0.9925925731658936, 0.9851852059364319, 1.0, 0.9851852059364319, 1.0, 1.0, 1.0, 1.0], [0.3333333432674408, NaN, 0.3629629611968994, 0.8222222328186035, 0.7333333492279053, 0.5185185074806213, 0.7111111283302307, 0.8962963223457336, 0.8296296000480652, 0.9481481313705444, 0.770370364189148, 0.9407407641410828, 0.9555555582046509, 0.9333333373069763, 0.9555555582046509, 0.970370352268219, 0.8740741014480591, 0.9111111164093018, 0.7851851582527161, 0.7259259223937988, 0.7333333492279053, 0.9481481313705444, 0.9851852059364319, 0.9925925731658936, 0.9111111164093018, 0.9777777791023254, 0.9555555582046509, 0.9629629850387573, 0.9259259104728699, 0.9777777791023254, 0.7777777910232544, 0.9925925731658936, 0.9925925731658936, 0.9777777791023254, 0.9629629850387573, 0.8518518805503845, 1.0, 0.9925925731658936, 0.9851852059364319, 0.8592592477798462, 0.9851852059364319, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 0.970370352268219, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.21481481194496155, 0.6370370388031006, NaN, 0.6518518328666687, 0.770370364189148, 0.6592592597007751, 0.8296296000480652, 0.7629629373550415, 0.8592592477798462, 0.7333333492279053, 0.8444444537162781, 0.7259259223937988, 0.7555555701255798, 0.7259259223937988, 0.7851851582527161, 0.7629629373550415, 0.8592592477798462, 0.770370364189148, 0.9185185432434082, 0.9185185432434082, 0.7333333492279053, 0.7851851582527161, 0.8666666746139526, 0.7851851582527161, 0.9037036895751953, 0.8518518805503845, 0.7777777910232544, 0.800000011920929, 0.9111111164093018, 0.7777777910232544, 0.8222222328186035, 0.8888888955116272, 0.8666666746139526, 0.8740741014480591, 0.9777777791023254, 0.8740741014480591, 0.9925925731658936, 0.8444444537162781, 0.9851852059364319, 0.8666666746139526, 0.9777777791023254, 0.9629629850387573, 0.8740741014480591, 0.9925925731658936, 0.9629629850387573, 0.9185185432434082, 0.9777777791023254, 0.9777777791023254, 0.9333333373069763, 0.9481481313705444, 0.9925925731658936, 0.9777777791023254, 0.9777777791023254, 0.9333333373069763, 0.9555555582046509, 0.9851852059364319, 0.9481481313705444, 0.9555555582046509, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 0.9481481313705444, 1.0, 0.9555555582046509, 1.0, 1.0, 0.970370352268219, 1.0], [0.24444444477558136, 0.17777778208255768, 0.3481481373310089, NaN, 0.5925925970077515, 0.5037037134170532, 0.5037037134170532, 0.7037037014961243, 0.6666666865348816, 0.8222222328186035, 0.6000000238418579, 0.8296296000480652, 0.8740741014480591, 0.8074073791503906, 0.8888888955116272, 0.9111111164093018, 0.7925925850868225, 0.8740741014480591, 0.644444465637207, 0.6296296119689941, 0.6296296119689941, 0.8518518805503845, 0.9111111164093018, 0.9259259104728699, 0.8518518805503845, 0.8370370268821716, 0.9111111164093018, 0.9185185432434082, 0.8962963223457336, 0.9259259104728699, 0.770370364189148, 0.9407407641410828, 0.970370352268219, 0.9481481313705444, 0.8444444537162781, 0.8296296000480652, 0.9481481313705444, 0.970370352268219, 0.9629629850387573, 0.8518518805503845, 0.9851852059364319, 0.9777777791023254, 0.970370352268219, 0.970370352268219, 0.9925925731658936, 0.970370352268219, 0.9925925731658936, 0.9851852059364319, 0.9777777791023254, 0.9407407641410828, 0.9777777791023254, 1.0, 0.9925925731658936, 0.9851852059364319, 0.9851852059364319, 0.9777777791023254, 1.0, 0.9925925731658936, 1.0, 0.9925925731658936, 1.0, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.2666666805744171, 0.22962963581085205, 0.40740740299224854, NaN, 0.4592592716217041, 0.5555555820465088, 0.6888889074325562, 0.6740740537643433, 0.6518518328666687, 0.644444465637207, 0.6518518328666687, 0.7333333492279053, 0.6518518328666687, 0.7037037014961243, 0.7185184955596924, 0.7333333492279053, 0.7925925850868225, 0.7777777910232544, 0.7185184955596924, 0.7407407164573669, 0.770370364189148, 0.8296296000480652, 0.8222222328186035, 0.8222222328186035, 0.8814814686775208, 0.800000011920929, 0.8074073791503906, 0.8370370268821716, 0.7851851582527161, 0.8296296000480652, 0.8962963223457336, 0.8666666746139526, 0.9407407641410828, 0.9111111164093018, 0.8296296000480652, 0.9481481313705444, 0.9185185432434082, 0.9851852059364319, 0.8518518805503845, 0.970370352268219, 1.0, 0.9777777791023254, 0.9629629850387573, 0.970370352268219, 0.9777777791023254, 0.9851852059364319, 0.9925925731658936, 0.9111111164093018, 0.970370352268219, 0.9629629850387573, 0.9925925731658936, 1.0, 0.9925925731658936, 0.9629629850387573, 0.9925925731658936, 0.9481481313705444, 0.9481481313705444, 1.0, 0.9629629850387573, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 0.9777777791023254, 1.0, 1.0, 0.9777777791023254, 1.0], [0.0962962955236435, 0.48148149251937866, 0.34074074029922485, 0.4962962865829468, 0.5407407283782959, NaN, 0.7333333492279053, 0.644444465637207, 0.7407407164573669, 0.6740740537643433, 0.7555555701255798, 0.6666666865348816, 0.7111111283302307, 0.6740740537643433, 0.6888889074325562, 0.6814814805984497, 0.7407407164573669, 0.6814814805984497, 0.8222222328186035, 0.7851851582527161, 0.770370364189148, 0.6814814805984497, 0.7777777910232544, 0.6962962746620178, 0.8148148059844971, 0.800000011920929, 0.6814814805984497, 0.7333333492279053, 0.770370364189148, 0.6962962746620178, 0.8222222328186035, 0.8074073791503906, 0.7481481432914734, 0.8444444537162781, 0.9407407641410828, 0.8222222328186035, 0.9629629850387573, 0.770370364189148, 0.9037036895751953, 0.8740741014480591, 0.8814814686775208, 0.9407407641410828, 0.8666666746139526, 0.9481481313705444, 0.9185185432434082, 0.8888888955116272, 0.9925925731658936, 0.9259259104728699, 0.8962963223457336, 0.9481481313705444, 0.9481481313705444, 0.9925925731658936, 0.9851852059364319, 0.9333333373069763, 0.9555555582046509, 0.9333333373069763, 0.9333333373069763, 0.9481481313705444, 1.0, 0.970370352268219, 0.9851852059364319, 0.9851852059364319, 1.0, 0.9481481313705444, 0.9333333373069763, 0.9629629850387573, 1.0, 1.0, 0.9777777791023254, 1.0], [0.08888889104127884, 0.2888889014720917, 0.17037037014961243, 0.4962962865829468, 0.4444444477558136, 0.2666666805744171, NaN, 0.6000000238418579, 0.5777778029441833, 0.585185170173645, 0.8222222328186035, 0.6074073910713196, 0.6296296119689941, 0.6000000238418579, 0.614814817905426, 0.6592592597007751, 0.614814817905426, 0.6888889074325562, 0.8074073791503906, 0.6296296119689941, 0.6592592597007751, 0.6074073910713196, 0.6962962746620178, 0.6814814805984497, 0.6962962746620178, 0.7925925850868225, 0.6666666865348816, 0.7629629373550415, 0.7481481432914734, 0.7407407164573669, 0.7925925850868225, 0.8074073791503906, 0.8074073791503906, 0.8296296000480652, 0.9037036895751953, 0.800000011920929, 0.8296296000480652, 0.8222222328186035, 0.8814814686775208, 0.7777777910232544, 0.8814814686775208, 0.8814814686775208, 0.8814814686775208, 0.9481481313705444, 0.9407407641410828, 0.9111111164093018, 0.9777777791023254, 0.970370352268219, 0.8222222328186035, 0.9037036895751953, 0.9037036895751953, 0.9777777791023254, 0.970370352268219, 0.9777777791023254, 0.9185185432434082, 0.9851852059364319, 0.9481481313705444, 0.9333333373069763, 1.0, 0.9407407641410828, 1.0, 1.0, 0.9925925731658936, 0.8962963223457336, 1.0, 0.9037036895751953, 1.0, 1.0, 0.970370352268219, 1.0], [0.11851851642131805, 0.10370370000600815, 0.2370370328426361, 0.29629629850387573, 0.31111112236976624, 0.35555556416511536, 0.4000000059604645, NaN, 0.5333333611488342, 0.5185185074806213, 0.4592592716217041, 0.5481481552124023, 0.6074073910713196, 0.5185185074806213, 0.6370370388031006, 0.6370370388031006, 0.5925925970077515, 0.770370364189148, 0.5925925970077515, 0.5925925970077515, 0.5777778029441833, 0.5185185074806213, 0.7037037014961243, 0.6888889074325562, 0.644444465637207, 0.7185184955596924, 0.614814817905426, 0.8592592477798462, 0.7481481432914734, 0.644444465637207, 0.7185184955596924, 0.8370370268821716, 0.9185185432434082, 0.7851851582527161, 0.7555555701255798, 0.7481481432914734, 0.9333333373069763, 0.8222222328186035, 0.9037036895751953, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8814814686775208, 0.9555555582046509, 0.9185185432434082, 0.8962963223457336, 0.9333333373069763, 0.9111111164093018, 0.9851852059364319, 0.9259259104728699, 0.8740741014480591, 0.9185185432434082, 0.9925925731658936, 0.9851852059364319, 0.9851852059364319, 0.9925925731658936, 1.0, 0.9777777791023254, 0.9555555582046509, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07407407462596893, 0.17037037014961243, 0.14074073731899261, 0.3333333432674408, 0.32592591643333435, 0.25925925374031067, 0.42222222685813904, 0.46666666865348816, NaN, 0.5259259343147278, 0.4962962865829468, 0.5185185074806213, 0.5777778029441833, 0.5629629492759705, 0.614814817905426, 0.6296296119689941, 0.6000000238418579, 0.6740740537643433, 0.6370370388031006, 0.5481481552124023, 0.6370370388031006, 0.6222222447395325, 0.6296296119689941, 0.644444465637207, 0.6962962746620178, 0.8296296000480652, 0.6962962746620178, 0.7555555701255798, 0.7555555701255798, 0.7481481432914734, 0.7629629373550415, 0.8222222328186035, 0.8444444537162781, 0.7629629373550415, 0.7851851582527161, 0.8148148059844971, 0.8814814686775208, 0.8222222328186035, 0.970370352268219, 0.8814814686775208, 0.8962963223457336, 0.9259259104728699, 0.8370370268821716, 0.9333333373069763, 0.8592592477798462, 0.8814814686775208, 0.9333333373069763, 0.8518518805503845, 0.9111111164093018, 0.9037036895751953, 0.8962963223457336, 0.9259259104728699, 0.9259259104728699, 0.9925925731658936, 0.9555555582046509, 0.9925925731658936, 1.0, 0.9777777791023254, 1.0, 0.9925925731658936, 0.9481481313705444, 0.9481481313705444, 0.9851852059364319, 0.9333333373069763, 1.0, 0.9407407641410828, 1.0, 1.0, 0.9851852059364319, 1.0], [0.14814814925193787, 0.051851850003004074, 0.2666666805744171, 0.17777778208255768, 0.3481481373310089, 0.32592591643333435, 0.4148148000240326, 0.48148149251937866, 0.4740740656852722, NaN, 0.4296296238899231, 0.614814817905426, 0.5185185074806213, 0.6666666865348816, 0.6000000238418579, 0.6888889074325562, 0.4888888895511627, 0.614814817905426, 0.5555555820465088, 0.46666666865348816, 0.5333333611488342, 0.6962962746620178, 0.644444465637207, 0.7629629373550415, 0.5555555820465088, 0.6703703999519348, 0.8074073791503906, 0.7185184955596924, 0.7481481432914734, 0.8592592477798462, 0.6370370388031006, 0.7925925850868225, 0.7925925850868225, 0.8740741014480591, 0.7481481432914734, 0.6888889074325562, 0.8222222328186035, 0.8962963223457336, 0.8222222328186035, 0.6814814805984497, 0.9037036895751953, 0.8888888955116272, 0.8814814686775208, 0.9629629850387573, 0.9259259104728699, 0.9629629850387573, 0.970370352268219, 0.9185185432434082, 0.9111111164093018, 0.9185185432434082, 0.9481481313705444, 0.970370352268219, 0.9629629850387573, 0.9777777791023254, 0.9777777791023254, 0.9925925731658936, 0.9259259104728699, 0.9851852059364319, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 1.0], [0.08148147910833359, 0.22962963581085205, 0.15555556118488312, 0.4000000059604645, 0.35555556416511536, 0.24444444477558136, 0.17777778208255768, 0.5407407283782959, 0.5037037134170532, 0.5703703761100769, NaN, 0.5777778029441833, 0.6000000238418579, 0.5777778029441833, 0.585185170173645, 0.6222222447395325, 0.6074073910713196, 0.6296296119689941, 0.7111111283302307, 0.5777778029441833, 0.6370370388031006, 0.5703703761100769, 0.7037037014961243, 0.6370370388031006, 0.6888889074325562, 0.7407407164573669, 0.6370370388031006, 0.7481481432914734, 0.7185184955596924, 0.7111111283302307, 0.7407407164573669, 0.770370364189148, 0.7925925850868225, 0.7851851582527161, 0.8888888955116272, 0.7555555701255798, 0.7851851582527161, 0.7777777910232544, 0.8592592477798462, 0.770370364189148, 0.8740741014480591, 0.8444444537162781, 0.8222222328186035, 0.9259259104728699, 0.9037036895751953, 0.8888888955116272, 0.9407407641410828, 0.9555555582046509, 0.800000011920929, 0.8888888955116272, 0.8814814686775208, 0.9555555582046509, 0.9407407641410828, 0.9777777791023254, 0.9185185432434082, 0.9629629850387573, 0.9333333373069763, 0.9259259104728699, 0.9925925731658936, 0.9259259104728699, 0.9925925731658936, 1.0, 0.9851852059364319, 0.8814814686775208, 1.0, 0.9111111164093018, 1.0, 1.0, 0.9555555582046509, 1.0], [0.11851851642131805, 0.05925925821065903, 0.27407407760620117, 0.17037037014961243, 0.3481481373310089, 0.3333333432674408, 0.39259257912635803, 0.45185184478759766, 0.48148149251937866, 0.385185182094574, 0.42222222685813904, NaN, 0.4444444477558136, 0.585185170173645, 0.585185170173645, 0.6592592597007751, 0.4962962865829468, 0.6074073910713196, 0.5259259343147278, 0.4592592716217041, 0.5333333611488342, 0.7185184955596924, 0.6370370388031006, 0.7259259223937988, 0.5407407283782959, 0.6370370388031006, 0.8074073791503906, 0.7259259223937988, 0.7037037014961243, 0.8296296000480652, 0.6296296119689941, 0.7851851582527161, 0.7851851582527161, 0.8666666746139526, 0.7407407164573669, 0.6814814805984497, 0.8444444537162781, 0.8888888955116272, 0.8074073791503906, 0.6814814805984497, 0.8962963223457336, 0.8962963223457336, 0.8814814686775208, 0.9407407641410828, 0.9185185432434082, 0.9777777791023254, 0.9481481313705444, 0.9185185432434082, 0.8814814686775208, 0.9111111164093018, 0.9555555582046509, 0.9629629850387573, 0.9555555582046509, 0.9777777791023254, 0.970370352268219, 0.9925925731658936, 0.9333333373069763, 0.9629629850387573, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9851852059364319, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0962962955236435, 0.04444444552063942, 0.24444444477558136, 0.1259259283542633, 0.2666666805744171, 0.2888889014720917, 0.37037035822868347, 0.39259257912635803, 0.42222222685813904, 0.48148149251937866, 0.4000000059604645, 0.5555555820465088, NaN, 0.5777778029441833, 0.5407407283782959, 0.6888889074325562, 0.4740740656852722, 0.5259259343147278, 0.4962962865829468, 0.5185185074806213, 0.4962962865829468, 0.644444465637207, 0.8222222328186035, 0.7777777910232544, 0.5925925970077515, 0.614814817905426, 0.7407407164573669, 0.644444465637207, 0.7259259223937988, 0.7481481432914734, 0.6296296119689941, 0.7555555701255798, 0.7185184955596924, 0.8814814686775208, 0.7259259223937988, 0.7481481432914734, 0.8518518805503845, 0.8444444537162781, 0.8444444537162781, 0.6592592597007751, 0.8740741014480591, 0.8814814686775208, 0.9111111164093018, 0.9851852059364319, 0.9333333373069763, 0.9777777791023254, 0.9851852059364319, 0.9407407641410828, 0.8444444537162781, 0.9037036895751953, 0.9851852059364319, 0.9851852059364319, 0.9555555582046509, 0.9851852059364319, 0.9407407641410828, 0.9851852059364319, 0.9037036895751953, 0.9111111164093018, 1.0, 0.9851852059364319, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9851852059364319, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0], [0.14074073731899261, 0.06666667014360428, 0.27407407760620117, 0.192592591047287, 0.3481481373310089, 0.32592591643333435, 0.4000000059604645, 0.48148149251937866, 0.43703705072402954, 0.3333333432674408, 0.42222222685813904, 0.4148148000240326, 0.42222222685813904, NaN, 0.5629629492759705, 0.5777778029441833, 0.4888888895511627, 0.585185170173645, 0.5407407283782959, 0.4592592716217041, 0.5481481552124023, 0.6074073910713196, 0.6074073910713196, 0.6740740537643433, 0.5703703761100769, 0.6518518328666687, 0.7333333492279053, 0.7259259223937988, 0.7333333492279053, 0.800000011920929, 0.6370370388031006, 0.7925925850868225, 0.800000011920929, 0.7925925850868225, 0.770370364189148, 0.6962962746620178, 0.770370364189148, 0.8740741014480591, 0.8444444537162781, 0.7037037014961243, 0.9037036895751953, 0.8074073791503906, 0.8740741014480591, 0.8740741014480591, 0.8518518805503845, 0.9259259104728699, 0.8814814686775208, 0.8518518805503845, 0.8888888955116272, 0.8444444537162781, 0.9629629850387573, 0.8962963223457336, 0.9185185432434082, 0.9555555582046509, 0.9851852059364319, 0.970370352268219, 0.9259259104728699, 0.970370352268219, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 0.9777777791023254, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10370370000600815, 0.04444444552063942, 0.21481481194496155, 0.1111111119389534, 0.29629629850387573, 0.31111112236976624, 0.385185182094574, 0.3629629611968994, 0.385185182094574, 0.4000000059604645, 0.4148148000240326, 0.4148148000240326, 0.4592592716217041, 0.43703705072402954, NaN, 0.5407407283782959, 0.4888888895511627, 0.4962962865829468, 0.5555555820465088, 0.43703705072402954, 0.5111111402511597, 0.5037037134170532, 0.5777778029441833, 0.6592592597007751, 0.614814817905426, 0.6000000238418579, 0.5777778029441833, 0.6296296119689941, 0.7555555701255798, 0.6222222447395325, 0.6296296119689941, 0.8814814686775208, 0.7333333492279053, 0.7481481432914734, 0.7037037014961243, 0.6592592597007751, 0.7925925850868225, 0.800000011920929, 0.8074073791503906, 0.6888889074325562, 0.9555555582046509, 0.9407407641410828, 0.8074073791503906, 0.9111111164093018, 0.9111111164093018, 0.8666666746139526, 0.9333333373069763, 0.9185185432434082, 0.970370352268219, 0.9111111164093018, 0.9851852059364319, 0.9555555582046509, 0.9925925731658936, 0.8666666746139526, 0.970370352268219, 0.9333333373069763, 0.9777777791023254, 0.9925925731658936, 0.9851852059364319, 0.9851852059364319, 0.9925925731658936, 0.9925925731658936, 0.9629629850387573, 0.9851852059364319, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936], [0.08148147910833359, 0.029629629105329514, 0.2370370328426361, 0.08888889104127884, 0.28148147463798523, 0.3185185194015503, 0.34074074029922485, 0.3629629611968994, 0.37037035822868347, 0.31111112236976624, 0.3777777850627899, 0.34074074029922485, 0.31111112236976624, 0.42222222685813904, 0.4592592716217041, NaN, 0.4888888895511627, 0.48148149251937866, 0.5037037134170532, 0.48148149251937866, 0.5259259343147278, 0.5333333611488342, 0.43703705072402954, 0.7555555701255798, 0.5407407283782959, 0.5629629492759705, 0.5777778029441833, 0.644444465637207, 0.6666666865348816, 0.7259259223937988, 0.6370370388031006, 0.800000011920929, 0.7407407164573669, 0.8814814686775208, 0.6518518328666687, 0.6814814805984497, 0.770370364189148, 0.8666666746139526, 0.8074073791503906, 0.6518518328666687, 0.8296296000480652, 0.8666666746139526, 0.8962963223457336, 0.970370352268219, 0.9333333373069763, 0.9851852059364319, 0.9481481313705444, 0.9333333373069763, 0.8666666746139526, 0.9037036895751953, 0.9777777791023254, 0.9629629850387573, 0.970370352268219, 0.9037036895751953, 0.9185185432434082, 0.9629629850387573, 0.9111111164093018, 0.9333333373069763, 1.0, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.007407407276332378, 0.1259259283542633, 0.14074073731899261, 0.2074074000120163, 0.2666666805744171, 0.25925925374031067, 0.385185182094574, 0.40740740299224854, 0.4000000059604645, 0.5111111402511597, 0.39259257912635803, 0.5037037134170532, 0.5259259343147278, 0.5111111402511597, 0.5111111402511597, 0.5111111402511597, NaN, 0.5111111402511597, 0.39259257912635803, 0.48148149251937866, 0.46666666865348816, 0.5185185074806213, 0.6592592597007751, 0.5925925970077515, 0.8074073791503906, 0.614814817905426, 0.5703703761100769, 0.5629629492759705, 0.6888889074325562, 0.614814817905426, 0.614814817905426, 0.6666666865348816, 0.585185170173645, 0.7111111283302307, 0.7259259223937988, 0.6814814805984497, 0.9777777791023254, 0.7111111283302307, 0.770370364189148, 0.7111111283302307, 0.7851851582527161, 0.9851852059364319, 0.8296296000480652, 0.9333333373069763, 0.8222222328186035, 0.8222222328186035, 0.8888888955116272, 0.8222222328186035, 0.8814814686775208, 0.8666666746139526, 0.8888888955116272, 0.9111111164093018, 0.9851852059364319, 0.9185185432434082, 0.9407407641410828, 0.9333333373069763, 0.8888888955116272, 0.9333333373069763, 0.9925925731658936, 0.9629629850387573, 1.0, 1.0, 1.0, 0.9555555582046509, 0.9333333373069763, 0.9629629850387573, 1.0, 1.0, 0.9851852059364319, 1.0], [0.1111111119389534, 0.08888889104127884, 0.22962963581085205, 0.1259259283542633, 0.2074074000120163, 0.3185185194015503, 0.31111112236976624, 0.22962963581085205, 0.32592591643333435, 0.385185182094574, 0.37037035822868347, 0.39259257912635803, 0.4740740656852722, 0.4148148000240326, 0.5037037134170532, 0.5185185074806213, 0.4888888895511627, NaN, 0.4888888895511627, 0.4592592716217041, 0.5333333611488342, 0.39259257912635803, 0.6000000238418579, 0.6000000238418579, 0.5259259343147278, 0.6222222447395325, 0.4592592716217041, 0.6962962746620178, 0.6518518328666687, 0.5629629492759705, 0.6666666865348816, 0.7777777910232544, 0.800000011920929, 0.7259259223937988, 0.6074073910713196, 0.6888889074325562, 0.8148148059844971, 0.7481481432914734, 0.8296296000480652, 0.8370370268821716, 0.8740741014480591, 0.9185185432434082, 0.800000011920929, 0.8444444537162781, 0.8888888955116272, 0.8148148059844971, 0.8148148059844971, 0.8814814686775208, 0.9185185432434082, 0.8592592477798462, 0.8222222328186035, 0.8592592477798462, 0.9555555582046509, 0.9481481313705444, 0.9777777791023254, 0.970370352268219, 0.9481481313705444, 0.970370352268219, 0.9407407641410828, 0.9777777791023254, 0.9851852059364319, 0.9851852059364319, 0.9629629850387573, 0.9851852059364319, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0], [0.014814814552664757, 0.21481481194496155, 0.08148147910833359, 0.35555556416511536, 0.2222222238779068, 0.17777778208255768, 0.192592591047287, 0.40740740299224854, 0.3629629611968994, 0.4444444477558136, 0.2888889014720917, 0.4740740656852722, 0.5037037134170532, 0.4592592716217041, 0.4444444477558136, 0.4962962865829468, 0.6074073910713196, 0.5111111402511597, NaN, 0.3777777850627899, 0.5111111402511597, 0.5111111402511597, 0.5925925970077515, 0.5555555820465088, 0.6592592597007751, 0.614814817905426, 0.5481481552124023, 0.6814814805984497, 0.7111111283302307, 0.644444465637207, 0.6370370388031006, 0.7407407164573669, 0.7185184955596924, 0.7259259223937988, 0.9111111164093018, 0.6962962746620178, 0.7259259223937988, 0.6222222447395325, 0.800000011920929, 0.7259259223937988, 0.8666666746139526, 0.8074073791503906, 0.6888889074325562, 0.9185185432434082, 0.8296296000480652, 0.7481481432914734, 0.9481481313705444, 0.8666666746139526, 0.7777777910232544, 0.9333333373069763, 0.8888888955116272, 0.9629629850387573, 0.970370352268219, 0.9185185432434082, 0.8666666746139526, 0.9259259104728699, 0.9333333373069763, 0.8962963223457336, 0.9851852059364319, 0.9333333373069763, 1.0, 1.0, 0.9925925731658936, 0.8962963223457336, 0.9407407641410828, 0.8962963223457336, 1.0, 1.0, 0.9481481313705444, 1.0], [0.03703703731298447, 0.27407407760620117, 0.08148147910833359, 0.37037035822868347, 0.28148147463798523, 0.21481481194496155, 0.37037035822868347, 0.40740740299224854, 0.45185184478759766, 0.5333333611488342, 0.42222222685813904, 0.5407407283782959, 0.48148149251937866, 0.5407407283782959, 0.5629629492759705, 0.5185185074806213, 0.5185185074806213, 0.5407407283782959, 0.6222222447395325, NaN, 0.46666666865348816, 0.5259259343147278, 0.5259259343147278, 0.5481481552124023, 0.6000000238418579, 0.5703703761100769, 0.585185170173645, 0.6592592597007751, 0.6666666865348816, 0.5777778029441833, 0.6592592597007751, 0.7629629373550415, 0.7407407164573669, 0.6666666865348816, 0.8074073791503906, 0.7925925850868225, 0.7111111283302307, 0.6074073910713196, 0.7777777910232544, 0.7777777910232544, 0.8222222328186035, 0.800000011920929, 0.6518518328666687, 0.8592592477798462, 0.7851851582527161, 0.7555555701255798, 0.8444444537162781, 0.8370370268821716, 0.7629629373550415, 0.7851851582527161, 0.7925925850868225, 0.8814814686775208, 0.8666666746139526, 0.9111111164093018, 0.8296296000480652, 0.9629629850387573, 0.9407407641410828, 0.9185185432434082, 0.970370352268219, 0.9037036895751953, 0.9259259104728699, 0.9259259104728699, 0.9481481313705444, 0.8222222328186035, 1.0, 0.8370370268821716, 1.0, 1.0, 0.9185185432434082, 1.0], [0.1259259283542633, 0.2666666805744171, 0.2666666805744171, 0.37037035822868347, 0.25925925374031067, 0.22962963581085205, 0.34074074029922485, 0.42222222685813904, 0.3629629611968994, 0.46666666865348816, 0.3629629611968994, 0.46666666865348816, 0.5037037134170532, 0.45185184478759766, 0.4888888895511627, 0.4740740656852722, 0.5333333611488342, 0.46666666865348816, 0.4888888895511627, 0.5333333611488342, NaN, 0.46666666865348816, 0.5259259343147278, 0.5185185074806213, 0.614814817905426, 0.5555555820465088, 0.5037037134170532, 0.7037037014961243, 0.6296296119689941, 0.6000000238418579, 0.9111111164093018, 0.7185184955596924, 0.7333333492279053, 0.6370370388031006, 0.7111111283302307, 0.9555555582046509, 0.7259259223937988, 0.6296296119689941, 0.8370370268821716, 0.7259259223937988, 0.8148148059844971, 0.7555555701255798, 0.6814814805984497, 0.8518518805503845, 0.7777777910232544, 0.7037037014961243, 0.7851851582527161, 0.770370364189148, 0.7925925850868225, 0.8296296000480652, 0.7555555701255798, 0.8592592477798462, 0.8518518805503845, 0.9851852059364319, 0.8666666746139526, 0.9851852059364319, 0.9481481313705444, 0.8888888955116272, 0.9925925731658936, 0.8962963223457336, 0.9037036895751953, 0.9037036895751953, 0.9925925731658936, 0.8370370268821716, 1.0, 0.8370370268821716, 1.0, 1.0, 0.9259259104728699, 1.0], [0.08148147910833359, 0.051851850003004074, 0.21481481194496155, 0.14814814925193787, 0.22962963581085205, 0.3185185194015503, 0.39259257912635803, 0.48148149251937866, 0.3777777850627899, 0.3037036955356598, 0.4296296238899231, 0.28148147463798523, 0.35555556416511536, 0.39259257912635803, 0.4962962865829468, 0.46666666865348816, 0.48148149251937866, 0.6074073910713196, 0.4888888895511627, 0.4740740656852722, 0.5333333611488342, NaN, 0.4000000059604645, 0.5481481552124023, 0.5481481552124023, 0.6074073910713196, 0.7777777910232544, 0.7037037014961243, 0.6592592597007751, 0.9037036895751953, 0.6222222447395325, 0.6740740537643433, 0.7407407164573669, 0.7111111283302307, 0.644444465637207, 0.6222222447395325, 0.7407407164573669, 0.8148148059844971, 0.7407407164573669, 0.6296296119689941, 0.7777777910232544, 0.8148148059844971, 0.8370370268821716, 0.8444444537162781, 0.8592592477798462, 0.8888888955116272, 0.8370370268821716, 0.8592592477798462, 0.8370370268821716, 0.8518518805503845, 0.9037036895751953, 0.8444444537162781, 0.8222222328186035, 0.8444444537162781, 0.8592592477798462, 0.8592592477798462, 0.8592592477798462, 0.8518518805503845, 0.9333333373069763, 0.9629629850387573, 0.9333333373069763, 0.9333333373069763, 0.9259259104728699, 1.0, 0.8814814686775208, 1.0, 0.9481481313705444, 1.0, 1.0, 1.0], [0.02222222276031971, 0.014814814552664757, 0.13333334028720856, 0.08888889104127884, 0.17037037014961243, 0.2222222238779068, 0.3037036955356598, 0.29629629850387573, 0.37037035822868347, 0.35555556416511536, 0.29629629850387573, 0.3629629611968994, 0.17777778208255768, 0.39259257912635803, 0.42222222685813904, 0.5629629492759705, 0.34074074029922485, 0.4000000059604645, 0.40740740299224854, 0.4740740656852722, 0.4740740656852722, 0.6000000238418579, NaN, 0.6592592597007751, 0.48148149251937866, 0.5259259343147278, 0.6518518328666687, 0.6074073910713196, 0.6296296119689941, 0.6888889074325562, 0.585185170173645, 0.6518518328666687, 0.6222222447395325, 0.8296296000480652, 0.6222222447395325, 0.644444465637207, 0.8074073791503906, 0.7851851582527161, 0.7851851582527161, 0.6296296119689941, 0.7333333492279053, 0.8444444537162781, 0.8222222328186035, 0.9777777791023254, 0.8592592477798462, 0.9037036895751953, 0.8888888955116272, 0.8962963223457336, 0.8370370268821716, 0.8148148059844971, 0.9555555582046509, 0.9481481313705444, 0.9333333373069763, 0.9555555582046509, 0.8592592477798462, 0.9777777791023254, 0.8666666746139526, 0.8740741014480591, 0.9925925731658936, 0.9407407641410828, 1.0, 1.0, 0.9925925731658936, 0.9111111164093018, 1.0, 0.9407407641410828, 1.0, 1.0, 0.9777777791023254, 1.0], [0.05925925821065903, 0.007407407276332378, 0.21481481194496155, 0.07407407462596893, 0.17777778208255768, 0.3037036955356598, 0.3185185194015503, 0.31111112236976624, 0.35555556416511536, 0.2370370328426361, 0.3629629611968994, 0.27407407760620117, 0.2222222238779068, 0.32592591643333435, 0.34074074029922485, 0.24444444477558136, 0.40740740299224854, 0.4000000059604645, 0.4444444477558136, 0.45185184478759766, 0.48148149251937866, 0.45185184478759766, 0.34074074029922485, NaN, 0.4962962865829468, 0.5555555820465088, 0.5037037134170532, 0.5925925970077515, 0.5925925970077515, 0.6740740537643433, 0.614814817905426, 0.7111111283302307, 0.6740740537643433, 0.7777777910232544, 0.6296296119689941, 0.6592592597007751, 0.7333333492279053, 0.8074073791503906, 0.7851851582527161, 0.6370370388031006, 0.8148148059844971, 0.8444444537162781, 0.8740741014480591, 0.9111111164093018, 0.9111111164093018, 0.9481481313705444, 0.9259259104728699, 0.9037036895751953, 0.8370370268821716, 0.8962963223457336, 0.9555555582046509, 0.9555555582046509, 0.970370352268219, 0.8592592477798462, 0.9111111164093018, 0.9555555582046509, 0.8666666746139526, 0.9037036895751953, 1.0, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.08888889104127884, 0.0962962955236435, 0.14814814925193787, 0.17777778208255768, 0.18518517911434174, 0.3037036955356598, 0.35555556416511536, 0.3037036955356598, 0.4444444477558136, 0.31111112236976624, 0.4592592716217041, 0.40740740299224854, 0.4296296238899231, 0.385185182094574, 0.4592592716217041, 0.192592591047287, 0.4740740656852722, 0.34074074029922485, 0.4000000059604645, 0.385185182094574, 0.45185184478759766, 0.5185185074806213, 0.5037037134170532, NaN, 0.5185185074806213, 0.5111111402511597, 0.4888888895511627, 0.6518518328666687, 0.5259259343147278, 0.5407407283782959, 0.5777778029441833, 0.5777778029441833, 0.6888889074325562, 0.644444465637207, 0.585185170173645, 0.8888888955116272, 0.6370370388031006, 0.6888889074325562, 0.7111111283302307, 0.7333333492279053, 0.9333333373069763, 0.7777777910232544, 0.8296296000480652, 0.7851851582527161, 0.7925925850868225, 0.8296296000480652, 0.7925925850868225, 0.8666666746139526, 0.8074073791503906, 0.8370370268821716, 0.8888888955116272, 0.9851852059364319, 0.9185185432434082, 0.9185185432434082, 0.9259259104728699, 0.8814814686775208, 0.9407407641410828, 0.9777777791023254, 0.9259259104728699, 1.0, 1.0, 0.9851852059364319, 0.9259259104728699, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9555555582046509, 1.0], [0.03703703731298447, 0.02222222276031971, 0.14814814925193787, 0.16296295821666718, 0.11851851642131805, 0.20000000298023224, 0.2074074000120163, 0.28148147463798523, 0.17037037014961243, 0.3296296298503876, 0.25925925374031067, 0.3629629611968994, 0.385185182094574, 0.3481481373310089, 0.4000000059604645, 0.43703705072402954, 0.385185182094574, 0.3777777850627899, 0.385185182094574, 0.4296296238899231, 0.4444444477558136, 0.39259257912635803, 0.4740740656852722, 0.4444444477558136, 0.48148149251937866, NaN, 0.4444444477558136, 0.5703703761100769, 0.644444465637207, 0.5259259343147278, 0.6000000238418579, 0.7185184955596924, 0.6370370388031006, 0.585185170173645, 0.7037037014961243, 0.7407407164573669, 0.7111111283302307, 0.6222222447395325, 0.8888888955116272, 0.7333333492279053, 0.8148148059844971, 0.7851851582527161, 0.7037037014961243, 0.7851851582527161, 0.7851851582527161, 0.7555555701255798, 0.8296296000480652, 0.800000011920929, 0.7777777910232544, 0.8222222328186035, 0.8444444537162781, 0.8370370268821716, 0.9037036895751953, 0.9037036895751953, 0.9481481313705444, 0.9481481313705444, 0.9185185432434082, 0.9407407641410828, 0.9777777791023254, 0.970370352268219, 0.970370352268219, 0.970370352268219, 0.9407407641410828, 0.9111111164093018, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9925925731658936, 1.0], [0.08148147910833359, 0.04444444552063942, 0.2222222238779068, 0.08888889104127884, 0.20000000298023224, 0.3185185194015503, 0.3333333432674408, 0.385185182094574, 0.3037036955356598, 0.192592591047287, 0.3629629611968994, 0.192592591047287, 0.25925925374031067, 0.2666666805744171, 0.42222222685813904, 0.42222222685813904, 0.4296296238899231, 0.5407407283782959, 0.45185184478759766, 0.4148148000240326, 0.4962962865829468, 0.2222222238779068, 0.3481481373310089, 0.4962962865829468, 0.4888888895511627, 0.5555555820465088, NaN, 0.6666666865348816, 0.6000000238418579, 0.7555555701255798, 0.6000000238418579, 0.6296296119689941, 0.7185184955596924, 0.6592592597007751, 0.6222222447395325, 0.585185170173645, 0.7037037014961243, 0.7333333492279053, 0.7037037014961243, 0.6222222447395325, 0.7481481432914734, 0.8074073791503906, 0.7851851582527161, 0.7851851582527161, 0.8444444537162781, 0.8370370268821716, 0.8074073791503906, 0.8444444537162781, 0.8074073791503906, 0.8222222328186035, 0.8666666746139526, 0.8222222328186035, 0.8148148059844971, 0.8296296000480652, 0.8370370268821716, 0.8444444537162781, 0.8518518805503845, 0.8444444537162781, 0.9259259104728699, 0.9481481313705444, 0.9259259104728699, 0.9259259104728699, 0.9185185432434082, 1.0, 0.8888888955116272, 1.0, 0.9481481313705444, 1.0, 1.0, 1.0], [0.07407407462596893, 0.03703703731298447, 0.20000000298023224, 0.08148147910833359, 0.192592591047287, 0.2666666805744171, 0.2370370328426361, 0.14074073731899261, 0.24444444477558136, 0.28148147463798523, 0.2518518567085266, 0.27407407760620117, 0.35555556416511536, 0.27407407760620117, 0.37037035822868347, 0.35555556416511536, 0.43703705072402954, 0.3037036955356598, 0.3185185194015503, 0.34074074029922485, 0.29629629850387573, 0.29629629850387573, 0.39259257912635803, 0.40740740299224854, 0.5111111402511597, 0.4296296238899231, 0.3333333432674408, NaN, 0.5037037134170532, 0.39259257912635803, 0.35555556416511536, 0.6000000238418579, 0.7851851582527161, 0.5111111402511597, 0.5111111402511597, 0.5333333611488342, 0.6962962746620178, 0.6296296119689941, 0.6370370388031006, 0.7037037014961243, 0.7185184955596924, 0.7481481432914734, 0.6888889074325562, 0.6370370388031006, 0.7629629373550415, 0.7185184955596924, 0.8074073791503906, 0.770370364189148, 0.8222222328186035, 0.7259259223937988, 0.6962962746620178, 0.8074073791503906, 0.8370370268821716, 0.8370370268821716, 0.8666666746139526, 0.9037036895751953, 1.0, 0.9111111164093018, 0.9185185432434082, 0.9777777791023254, 0.9037036895751953, 0.9037036895751953, 0.9851852059364319, 0.9851852059364319, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9777777791023254, 1.0, 1.0], [0.06666667014360428, 0.07407407462596893, 0.08888889104127884, 0.10370370000600815, 0.16296295821666718, 0.22962963581085205, 0.2518518567085266, 0.2518518567085266, 0.24444444477558136, 0.2518518567085266, 0.28148147463798523, 0.29629629850387573, 0.27407407760620117, 0.2666666805744171, 0.24444444477558136, 0.3333333432674408, 0.31111112236976624, 0.3481481373310089, 0.2888889014720917, 0.3333333432674408, 0.37037035822868347, 0.34074074029922485, 0.37037035822868347, 0.40740740299224854, 0.3481481373310089, 0.35555556416511536, 0.4000000059604645, 0.4962962865829468, NaN, 0.48148149251937866, 0.5259259343147278, 0.5629629492759705, 0.5407407283782959, 0.4962962865829468, 0.5481481552124023, 0.5777778029441833, 0.614814817905426, 0.6074073910713196, 0.6296296119689941, 0.7111111283302307, 0.6814814805984497, 0.7259259223937988, 0.6518518328666687, 0.7851851582527161, 0.8222222328186035, 0.6666666865348816, 0.7555555701255798, 0.8370370268821716, 0.8222222328186035, 0.8296296000480652, 0.8370370268821716, 0.8444444537162781, 0.9259259104728699, 0.770370364189148, 0.9037036895751953, 0.8222222328186035, 0.9407407641410828, 0.9333333373069763, 0.9777777791023254, 0.9925925731658936, 0.9629629850387573, 0.9629629850387573, 0.9481481313705444, 0.9333333373069763, 0.9407407641410828, 0.9555555582046509, 0.9925925731658936, 1.0, 1.0, 1.0], [0.06666667014360428, 0.02222222276031971, 0.2222222238779068, 0.07407407462596893, 0.21481481194496155, 0.3037036955356598, 0.25925925374031067, 0.35555556416511536, 0.2518518567085266, 0.14074073731899261, 0.2888889014720917, 0.17037037014961243, 0.2518518567085266, 0.20000000298023224, 0.3777777850627899, 0.27407407760620117, 0.385185182094574, 0.43703705072402954, 0.35555556416511536, 0.42222222685813904, 0.4000000059604645, 0.0962962955236435, 0.31111112236976624, 0.32592591643333435, 0.4740740656852722, 0.4740740656852722, 0.24444444477558136, 0.6074073910713196, 0.5185185074806213, NaN, 0.5925925970077515, 0.614814817905426, 0.7037037014961243, 0.5629629492759705, 0.5925925970077515, 0.614814817905426, 0.6074073910713196, 0.6666666865348816, 0.6666666865348816, 0.6222222447395325, 0.6962962746620178, 0.6962962746620178, 0.7481481432914734, 0.6666666865348816, 0.7481481432914734, 0.7777777910232544, 0.6592592597007751, 0.7407407164573669, 0.8074073791503906, 0.7777777910232544, 0.8222222328186035, 0.7185184955596924, 0.7851851582527161, 0.7925925850868225, 0.8444444537162781, 0.8296296000480652, 0.8518518805503845, 0.8444444537162781, 0.9333333373069763, 0.9333333373069763, 0.9259259104728699, 0.9333333373069763, 0.9185185432434082, 1.0, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9777777791023254, 1.0, 1.0], [0.07407407462596893, 0.2222222238779068, 0.17777778208255768, 0.22962963581085205, 0.17037037014961243, 0.17777778208255768, 0.2074074000120163, 0.28148147463798523, 0.2370370328426361, 0.3629629611968994, 0.25925925374031067, 0.37037035822868347, 0.37037035822868347, 0.3629629611968994, 0.37037035822868347, 0.3629629611968994, 0.385185182094574, 0.3333333432674408, 0.3629629611968994, 0.34074074029922485, 0.08888889104127884, 0.3777777850627899, 0.4148148000240326, 0.385185182094574, 0.4592592716217041, 0.4000000059604645, 0.4000000059604645, 0.644444465637207, 0.4740740656852722, 0.40740740299224854, NaN, 0.5259259343147278, 0.6222222447395325, 0.4888888895511627, 0.5481481552124023, 0.7740740776062012, 0.5629629492759705, 0.48148149251937866, 0.614814817905426, 0.6814814805984497, 0.6740740537643433, 0.6074073910713196, 0.6000000238418579, 0.7037037014961243, 0.6592592597007751, 0.6074073910713196, 0.7111111283302307, 0.7259259223937988, 0.6518518328666687, 0.6592592597007751, 0.6296296119689941, 0.7629629373550415, 0.7851851582527161, 0.9629629850387573, 0.7851851582527161, 0.970370352268219, 0.9333333373069763, 0.8518518805503845, 0.9333333373069763, 0.8666666746139526, 0.7777777910232544, 0.7851851582527161, 0.9555555582046509, 0.7925925850868225, 1.0, 0.8296296000480652, 0.9407407641410828, 0.9851852059364319, 0.8666666746139526, 1.0], [0.02222222276031971, 0.007407407276332378, 0.1111111119389534, 0.05925925821065903, 0.10370370000600815, 0.192592591047287, 0.192592591047287, 0.16296295821666718, 0.17777778208255768, 0.2074074000120163, 0.22962963581085205, 0.21481481194496155, 0.24444444477558136, 0.2074074000120163, 0.11851851642131805, 0.20000000298023224, 0.3333333432674408, 0.2222222238779068, 0.25925925374031067, 0.2370370328426361, 0.28148147463798523, 0.32592591643333435, 0.3481481373310089, 0.2888889014720917, 0.42222222685813904, 0.28148147463798523, 0.37037035822868347, 0.4000000059604645, 0.43703705072402954, 0.385185182094574, 0.4740740656852722, NaN, 0.45185184478759766, 0.4740740656852722, 0.5185185074806213, 0.5111111402511597, 0.5703703761100769, 0.5481481552124023, 0.5259259343147278, 0.6000000238418579, 0.9111111164093018, 0.6296296119689941, 0.6592592597007751, 0.6370370388031006, 0.7259259223937988, 0.6888889074325562, 0.7555555701255798, 0.7851851582527161, 0.7407407164573669, 0.8148148059844971, 0.8148148059844971, 0.800000011920929, 0.9925925731658936, 0.7333333492279053, 0.8814814686775208, 0.7555555701255798, 0.8222222328186035, 0.8888888955116272, 0.9407407641410828, 0.9407407641410828, 0.9481481313705444, 0.9481481313705444, 0.9037036895751953, 0.9407407641410828, 0.9259259104728699, 0.9407407641410828, 1.0, 1.0, 0.970370352268219, 1.0], [0.05925925821065903, 0.007407407276332378, 0.13333334028720856, 0.029629629105329514, 0.13333334028720856, 0.2518518567085266, 0.192592591047287, 0.08148147910833359, 0.15555556118488312, 0.2074074000120163, 0.2074074000120163, 0.21481481194496155, 0.28148147463798523, 0.20000000298023224, 0.2666666805744171, 0.25925925374031067, 0.4148148000240326, 0.20000000298023224, 0.28148147463798523, 0.25925925374031067, 0.2666666805744171, 0.25925925374031067, 0.3777777850627899, 0.32592591643333435, 0.42222222685813904, 0.3629629611968994, 0.28148147463798523, 0.21481481194496155, 0.4592592716217041, 0.29629629850387573, 0.3777777850627899, 0.5481481552124023, NaN, 0.45185184478759766, 0.5185185074806213, 0.4296296238899231, 0.644444465637207, 0.5407407283782959, 0.5777778029441833, 0.614814817905426, 0.6888889074325562, 0.7185184955596924, 0.6370370388031006, 0.5925925970077515, 0.7333333492279053, 0.614814817905426, 0.7037037014961243, 0.770370364189148, 0.7777777910232544, 0.800000011920929, 0.6518518328666687, 0.7851851582527161, 0.8222222328186035, 0.8148148059844971, 0.8814814686775208, 0.8444444537162781, 0.9629629850387573, 0.9111111164093018, 0.8666666746139526, 0.9851852059364319, 0.8888888955116272, 0.8962963223457336, 0.9481481313705444, 0.970370352268219, 0.9259259104728699, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0], [0.014814814552664757, 0.02222222276031971, 0.1259259283542633, 0.051851850003004074, 0.05925925821065903, 0.15555556118488312, 0.17037037014961243, 0.21481481194496155, 0.2370370328426361, 0.1259259283542633, 0.21481481194496155, 0.13333334028720856, 0.11851851642131805, 0.2074074000120163, 0.2518518567085266, 0.11851851642131805, 0.2888889014720917, 0.27407407760620117, 0.27407407760620117, 0.3333333432674408, 0.3629629611968994, 0.2888889014720917, 0.17037037014961243, 0.2222222238779068, 0.31111112236976624, 0.4148148000240326, 0.34074074029922485, 0.4888888895511627, 0.5037037134170532, 0.43703705072402954, 0.5111111402511597, 0.5259259343147278, 0.5481481552124023, NaN, 0.5259259343147278, 0.5185185074806213, 0.4740740656852722, 0.585185170173645, 0.614814817905426, 0.585185170173645, 0.6962962746620178, 0.5629629492759705, 0.6592592597007751, 0.5111111402511597, 0.7629629373550415, 0.7629629373550415, 0.7259259223937988, 0.7851851582527161, 0.7925925850868225, 0.7629629373550415, 0.8666666746139526, 0.7555555701255798, 0.8814814686775208, 0.7481481432914734, 0.8074073791503906, 0.8814814686775208, 0.7777777910232544, 0.8444444537162781, 0.9481481313705444, 0.9259259104728699, 0.9629629850387573, 0.970370352268219, 0.8740741014480591, 0.9185185432434082, 1.0, 0.9259259104728699, 1.0, 1.0, 0.9333333373069763, 1.0], [0.029629629105329514, 0.03703703731298447, 0.02222222276031971, 0.15555556118488312, 0.08888889104127884, 0.05925925821065903, 0.0962962955236435, 0.24444444477558136, 0.21481481194496155, 0.2518518567085266, 0.1111111119389534, 0.25925925374031067, 0.27407407760620117, 0.22962963581085205, 0.29629629850387573, 0.3481481373310089, 0.27407407760620117, 0.39259257912635803, 0.08888889104127884, 0.192592591047287, 0.2888889014720917, 0.35555556416511536, 0.3777777850627899, 0.37037035822868347, 0.35555556416511536, 0.29629629850387573, 0.3777777850627899, 0.4888888895511627, 0.45185184478759766, 0.40740740299224854, 0.45185184478759766, 0.48148149251937866, 0.48148149251937866, 0.4740740656852722, NaN, 0.6222222447395325, 0.4888888895511627, 0.5481481552124023, 0.5629629492759705, 0.5925925970077515, 0.5777778029441833, 0.585185170173645, 0.585185170173645, 0.5925925970077515, 0.5555555820465088, 0.6000000238418579, 0.6222222447395325, 0.5777778029441833, 0.7481481432914734, 0.6666666865348816, 0.6962962746620178, 0.6666666865348816, 0.7185184955596924, 0.7925925850868225, 0.7851851582527161, 0.8296296000480652, 0.800000011920929, 0.7851851582527161, 0.8370370268821716, 0.8518518805503845, 0.8666666746139526, 0.8666666746139526, 0.8814814686775208, 0.8148148059844971, 0.9333333373069763, 0.8370370268821716, 0.9481481313705444, 0.9777777791023254, 0.8740741014480591, 1.0], [0.05925925821065903, 0.14814814925193787, 0.1259259283542633, 0.17037037014961243, 0.17037037014961243, 0.17777778208255768, 0.20000000298023224, 0.2518518567085266, 0.18518517911434174, 0.31111112236976624, 0.24444444477558136, 0.3185185194015503, 0.2518518567085266, 0.3037036955356598, 0.34074074029922485, 0.3185185194015503, 0.3185185194015503, 0.31111112236976624, 0.3037036955356598, 0.2074074000120163, 0.04444444552063942, 0.3777777850627899, 0.35555556416511536, 0.34074074029922485, 0.4148148000240326, 0.25925925374031067, 0.4148148000240326, 0.46666666865348816, 0.42222222685813904, 0.385185182094574, 0.22592592239379883, 0.4888888895511627, 0.5703703761100769, 0.48148149251937866, 0.3777777850627899, NaN, 0.5555555820465088, 0.48148149251937866, 0.5407407283782959, 0.5703703761100769, 0.585185170173645, 0.5555555820465088, 0.5333333611488342, 0.6000000238418579, 0.6296296119689941, 0.5407407283782959, 0.6222222447395325, 0.6592592597007751, 0.585185170173645, 0.5555555820465088, 0.614814817905426, 0.644444465637207, 0.7481481432914734, 0.770370364189148, 0.6740740537643433, 0.8666666746139526, 0.8148148059844971, 0.7555555701255798, 0.7851851582527161, 0.7407407164573669, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.800000011920929, 0.9851852059364319, 0.8074073791503906, 0.8074073791503906, 0.8222222328186035, 0.8296296000480652, 1.0], [0.0, 0.0, 0.007407407276332378, 0.051851850003004074, 0.051851850003004074, 0.03703703731298447, 0.17037037014961243, 0.06666667014360428, 0.11851851642131805, 0.17777778208255768, 0.21481481194496155, 0.15555556118488312, 0.14814814925193787, 0.22962963581085205, 0.2074074000120163, 0.22962963581085205, 0.02222222276031971, 0.18518517911434174, 0.27407407760620117, 0.2888889014720917, 0.27407407760620117, 0.25925925374031067, 0.192592591047287, 0.2666666805744171, 0.1111111119389534, 0.2888889014720917, 0.29629629850387573, 0.3037036955356598, 0.385185182094574, 0.39259257912635803, 0.43703705072402954, 0.4296296238899231, 0.35555556416511536, 0.5259259343147278, 0.5111111402511597, 0.4444444477558136, NaN, 0.5333333611488342, 0.5629629492759705, 0.5703703761100769, 0.5407407283782959, 0.6814814805984497, 0.6222222447395325, 0.614814817905426, 0.6888889074325562, 0.6518518328666687, 0.6370370388031006, 0.6962962746620178, 0.770370364189148, 0.6814814805984497, 0.6740740537643433, 0.6592592597007751, 0.7259259223937988, 0.8592592477798462, 0.8518518805503845, 0.8814814686775208, 0.7407407164573669, 0.8740741014480591, 0.8666666746139526, 0.8962963223457336, 0.9333333373069763, 0.9333333373069763, 0.9185185432434082, 0.8666666746139526, 0.9333333373069763, 0.8592592477798462, 1.0, 1.0, 0.9333333373069763, 1.0], [0.007407407276332378, 0.007407407276332378, 0.15555556118488312, 0.029629629105329514, 0.08148147910833359, 0.22962963581085205, 0.17777778208255768, 0.17777778208255768, 0.17777778208255768, 0.10370370000600815, 0.2222222238779068, 0.1111111119389534, 0.15555556118488312, 0.1259259283542633, 0.20000000298023224, 0.13333334028720856, 0.2888889014720917, 0.2518518567085266, 0.3777777850627899, 0.39259257912635803, 0.37037035822868347, 0.18518517911434174, 0.21481481194496155, 0.192592591047287, 0.3629629611968994, 0.3777777850627899, 0.2666666805744171, 0.37037035822868347, 0.39259257912635803, 0.3333333432674408, 0.5185185074806213, 0.45185184478759766, 0.4592592716217041, 0.4148148000240326, 0.45185184478759766, 0.5185185074806213, 0.46666666865348816, NaN, 0.5481481552124023, 0.46666666865348816, 0.644444465637207, 0.5481481552124023, 0.8296296000480652, 0.5185185074806213, 0.614814817905426, 0.9481481313705444, 0.6666666865348816, 0.644444465637207, 0.644444465637207, 0.6666666865348816, 0.644444465637207, 0.6962962746620178, 0.800000011920929, 0.6518518328666687, 0.7037037014961243, 0.644444465637207, 0.6740740537643433, 0.7851851582527161, 0.7185184955596924, 0.8370370268821716, 0.9777777791023254, 0.9407407641410828, 0.7259259223937988, 0.9333333373069763, 0.7555555701255798, 0.9481481313705444, 1.0, 0.9259259104728699, 0.9629629850387573, 0.9333333373069763], [0.0, 0.014814814552664757, 0.014814814552664757, 0.03703703731298447, 0.014814814552664757, 0.0962962955236435, 0.11851851642131805, 0.0962962955236435, 0.029629629105329514, 0.17777778208255768, 0.14074073731899261, 0.192592591047287, 0.15555556118488312, 0.15555556118488312, 0.192592591047287, 0.192592591047287, 0.22962963581085205, 0.17037037014961243, 0.20000000298023224, 0.2222222238779068, 0.16296295821666718, 0.25925925374031067, 0.21481481194496155, 0.21481481194496155, 0.31111112236976624, 0.1111111119389534, 0.29629629850387573, 0.3629629611968994, 0.37037035822868347, 0.3333333432674408, 0.385185182094574, 0.4740740656852722, 0.42222222685813904, 0.385185182094574, 0.43703705072402954, 0.4592592716217041, 0.43703705072402954, 0.45185184478759766, NaN, 0.5555555820465088, 0.5555555820465088, 0.5259259343147278, 0.48148149251937866, 0.5925925970077515, 0.644444465637207, 0.5037037134170532, 0.5777778029441833, 0.7111111283302307, 0.585185170173645, 0.6740740537643433, 0.6222222447395325, 0.6370370388031006, 0.8370370268821716, 0.770370364189148, 0.6962962746620178, 0.8444444537162781, 0.8666666746139526, 0.7851851582527161, 0.9555555582046509, 0.8888888955116272, 0.8296296000480652, 0.8370370268821716, 0.8888888955116272, 0.800000011920929, 0.9481481313705444, 0.8296296000480652, 1.0, 1.0, 0.9111111164093018, 1.0], [0.0962962955236435, 0.14074073731899261, 0.13333334028720856, 0.14814814925193787, 0.14814814925193787, 0.1259259283542633, 0.2222222238779068, 0.1111111119389534, 0.11851851642131805, 0.3185185194015503, 0.22962963581085205, 0.3185185194015503, 0.34074074029922485, 0.29629629850387573, 0.31111112236976624, 0.3481481373310089, 0.2888889014720917, 0.16296295821666718, 0.27407407760620117, 0.2222222238779068, 0.27407407760620117, 0.37037035822868347, 0.37037035822868347, 0.3629629611968994, 0.2888889014720917, 0.2666666805744171, 0.3777777850627899, 0.29629629850387573, 0.2888889014720917, 0.3777777850627899, 0.3185185194015503, 0.4000000059604645, 0.385185182094574, 0.4148148000240326, 0.40740740299224854, 0.4296296238899231, 0.4296296238899231, 0.5333333611488342, 0.4444444477558136, NaN, 0.5333333611488342, 0.4888888895511627, 0.5407407283782959, 0.40740740299224854, 0.5111111402511597, 0.5407407283782959, 0.48148149251937866, 0.5333333611488342, 0.6222222447395325, 0.6000000238418579, 0.5259259343147278, 0.5037037134170532, 0.5925925970077515, 0.5629629492759705, 0.7111111283302307, 0.6074073910713196, 0.6666666865348816, 0.7407407164573669, 0.770370364189148, 0.7111111283302307, 0.6814814805984497, 0.6888889074325562, 0.6962962746620178, 0.7629629373550415, 0.7925925850868225, 0.770370364189148, 0.7481481432914734, 0.7851851582527161, 0.770370364189148, 0.9259259104728699], [0.0, 0.014814814552664757, 0.02222222276031971, 0.014814814552664757, 0.029629629105329514, 0.11851851642131805, 0.11851851642131805, 0.1111111119389534, 0.10370370000600815, 0.0962962955236435, 0.1259259283542633, 0.10370370000600815, 0.1259259283542633, 0.0962962955236435, 0.04444444552063942, 0.17037037014961243, 0.21481481194496155, 0.1259259283542633, 0.13333334028720856, 0.17777778208255768, 0.18518517911434174, 0.2222222238779068, 0.2666666805744171, 0.18518517911434174, 0.2666666805744171, 0.18518517911434174, 0.2518518567085266, 0.28148147463798523, 0.3185185194015503, 0.3037036955356598, 0.32592591643333435, 0.08888889104127884, 0.31111112236976624, 0.3037036955356598, 0.42222222685813904, 0.4148148000240326, 0.4592592716217041, 0.35555556416511536, 0.4444444477558136, 0.46666666865348816, NaN, 0.43703705072402954, 0.4740740656852722, 0.5185185074806213, 0.5703703761100769, 0.5185185074806213, 0.614814817905426, 0.6296296119689941, 0.6370370388031006, 0.7037037014961243, 0.7259259223937988, 0.6962962746620178, 0.9259259104728699, 0.7037037014961243, 0.7481481432914734, 0.7333333492279053, 0.770370364189148, 0.7777777910232544, 0.8518518805503845, 0.8962963223457336, 0.8962963223457336, 0.8962963223457336, 0.8148148059844971, 0.9185185432434082, 0.9037036895751953, 0.9185185432434082, 0.9851852059364319, 1.0, 0.9481481313705444, 1.0], [0.0, 0.0, 0.03703703731298447, 0.02222222276031971, 0.0, 0.05925925821065903, 0.11851851642131805, 0.0, 0.07407407462596893, 0.1111111119389534, 0.15555556118488312, 0.10370370000600815, 0.11851851642131805, 0.192592591047287, 0.05925925821065903, 0.13333334028720856, 0.014814814552664757, 0.08148147910833359, 0.192592591047287, 0.20000000298023224, 0.24444444477558136, 0.18518517911434174, 0.15555556118488312, 0.15555556118488312, 0.06666667014360428, 0.21481481194496155, 0.192592591047287, 0.2518518567085266, 0.27407407760620117, 0.3037036955356598, 0.39259257912635803, 0.37037035822868347, 0.28148147463798523, 0.43703705072402954, 0.4148148000240326, 0.4444444477558136, 0.3185185194015503, 0.45185184478759766, 0.4740740656852722, 0.5111111402511597, 0.5629629492759705, NaN, 0.5333333611488342, 0.48148149251937866, 0.614814817905426, 0.5555555820465088, 0.6222222447395325, 0.6740740537643433, 0.6666666865348816, 0.6296296119689941, 0.6740740537643433, 0.6518518328666687, 0.7259259223937988, 0.7481481432914734, 0.8814814686775208, 0.7925925850868225, 0.6888889074325562, 0.8962963223457336, 0.8222222328186035, 0.8592592477798462, 0.8444444537162781, 0.8444444537162781, 0.9185185432434082, 0.8888888955116272, 0.9333333373069763, 0.9185185432434082, 0.9777777791023254, 1.0, 0.9407407641410828, 1.0], [0.0, 0.0, 0.1259259283542633, 0.029629629105329514, 0.02222222276031971, 0.13333334028720856, 0.11851851642131805, 0.11851851642131805, 0.16296295821666718, 0.11851851642131805, 0.17777778208255768, 0.11851851642131805, 0.08888889104127884, 0.1259259283542633, 0.192592591047287, 0.10370370000600815, 0.17037037014961243, 0.20000000298023224, 0.31111112236976624, 0.3481481373310089, 0.3185185194015503, 0.16296295821666718, 0.17777778208255768, 0.1259259283542633, 0.2222222238779068, 0.29629629850387573, 0.21481481194496155, 0.31111112236976624, 0.3481481373310089, 0.2518518567085266, 0.4000000059604645, 0.34074074029922485, 0.3629629611968994, 0.34074074029922485, 0.4148148000240326, 0.46666666865348816, 0.3777777850627899, 0.17037037014961243, 0.5185185074806213, 0.4592592716217041, 0.5259259343147278, 0.46666666865348816, NaN, 0.4592592716217041, 0.5333333611488342, 0.8074073791503906, 0.5925925970077515, 0.5555555820465088, 0.5703703761100769, 0.614814817905426, 0.5555555820465088, 0.6370370388031006, 0.7407407164573669, 0.6296296119689941, 0.6888889074325562, 0.6370370388031006, 0.6592592597007751, 0.7481481432914734, 0.6740740537643433, 0.7925925850868225, 0.9481481313705444, 0.9185185432434082, 0.6888889074325562, 0.9037036895751953, 0.7333333492279053, 0.9259259104728699, 1.0, 0.9185185432434082, 0.9481481313705444, 0.9333333373069763], [0.0, 0.0, 0.007407407276332378, 0.029629629105329514, 0.03703703731298447, 0.051851850003004074, 0.051851850003004074, 0.04444444552063942, 0.06666667014360428, 0.03703703731298447, 0.07407407462596893, 0.05925925821065903, 0.014814814552664757, 0.1259259283542633, 0.08888889104127884, 0.029629629105329514, 0.06666667014360428, 0.15555556118488312, 0.08148147910833359, 0.14074073731899261, 0.14814814925193787, 0.15555556118488312, 0.02222222276031971, 0.08888889104127884, 0.17037037014961243, 0.21481481194496155, 0.21481481194496155, 0.3629629611968994, 0.21481481194496155, 0.3333333432674408, 0.29629629850387573, 0.3629629611968994, 0.40740740299224854, 0.4888888895511627, 0.40740740299224854, 0.4000000059604645, 0.385185182094574, 0.48148149251937866, 0.40740740299224854, 0.5925925970077515, 0.48148149251937866, 0.5185185074806213, 0.5407407283782959, NaN, 0.5925925970077515, 0.5629629492759705, 0.5703703761100769, 0.6000000238418579, 0.6518518328666687, 0.5777778029441833, 0.6740740537643433, 0.6370370388031006, 0.7629629373550415, 0.7333333492279053, 0.7333333492279053, 0.8148148059844971, 0.7925925850868225, 0.800000011920929, 0.9259259104728699, 0.8444444537162781, 0.8962963223457336, 0.9037036895751953, 0.9407407641410828, 0.7925925850868225, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 0.8740741014480591, 1.0], [0.014814814552664757, 0.0, 0.03703703731298447, 0.007407407276332378, 0.029629629105329514, 0.08148147910833359, 0.05925925821065903, 0.08148147910833359, 0.14074073731899261, 0.07407407462596893, 0.0962962955236435, 0.08148147910833359, 0.06666667014360428, 0.14814814925193787, 0.08888889104127884, 0.06666667014360428, 0.17777778208255768, 0.1111111119389534, 0.17037037014961243, 0.21481481194496155, 0.2222222238779068, 0.14074073731899261, 0.14074073731899261, 0.08888889104127884, 0.21481481194496155, 0.21481481194496155, 0.15555556118488312, 0.2370370328426361, 0.17777778208255768, 0.2518518567085266, 0.34074074029922485, 0.27407407760620117, 0.2666666805744171, 0.2370370328426361, 0.4444444477558136, 0.37037035822868347, 0.31111112236976624, 0.385185182094574, 0.35555556416511536, 0.4888888895511627, 0.4296296238899231, 0.385185182094574, 0.46666666865348816, 0.40740740299224854, NaN, 0.5185185074806213, 0.5555555820465088, 0.6962962746620178, 0.5925925970077515, 0.6666666865348816, 0.6370370388031006, 0.6370370388031006, 0.7111111283302307, 0.6518518328666687, 0.7333333492279053, 0.6814814805984497, 0.6814814805984497, 0.800000011920929, 0.970370352268219, 0.9259259104728699, 0.8666666746139526, 0.8740741014480591, 0.8592592477798462, 0.8666666746139526, 0.8444444537162781, 0.9037036895751953, 0.9555555582046509, 1.0, 0.970370352268219, 1.0], [0.0, 0.0, 0.08148147910833359, 0.029629629105329514, 0.02222222276031971, 0.1111111119389534, 0.08888889104127884, 0.10370370000600815, 0.11851851642131805, 0.03703703731298447, 0.1111111119389534, 0.02222222276031971, 0.02222222276031971, 0.07407407462596893, 0.13333334028720856, 0.014814814552664757, 0.17777778208255768, 0.18518517911434174, 0.2518518567085266, 0.24444444477558136, 0.29629629850387573, 0.1111111119389534, 0.0962962955236435, 0.051851850003004074, 0.2074074000120163, 0.24444444477558136, 0.16296295821666718, 0.28148147463798523, 0.3333333432674408, 0.2222222238779068, 0.39259257912635803, 0.31111112236976624, 0.385185182094574, 0.2370370328426361, 0.4000000059604645, 0.4592592716217041, 0.3481481373310089, 0.051851850003004074, 0.4962962865829468, 0.4592592716217041, 0.48148149251937866, 0.4444444477558136, 0.192592591047287, 0.43703705072402954, 0.48148149251937866, NaN, 0.585185170173645, 0.5185185074806213, 0.5333333611488342, 0.5185185074806213, 0.5629629492759705, 0.6370370388031006, 0.7333333492279053, 0.614814817905426, 0.6888889074325562, 0.6370370388031006, 0.6222222447395325, 0.7111111283302307, 0.6592592597007751, 0.7851851582527161, 0.8592592477798462, 0.8444444537162781, 0.6814814805984497, 0.8740741014480591, 0.7259259223937988, 0.8962963223457336, 1.0, 0.8814814686775208, 0.9407407641410828, 0.9333333373069763], [0.007407407276332378, 0.0, 0.02222222276031971, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.02222222276031971, 0.06666667014360428, 0.06666667014360428, 0.029629629105329514, 0.05925925821065903, 0.051851850003004074, 0.014814814552664757, 0.11851851642131805, 0.06666667014360428, 0.051851850003004074, 0.1111111119389534, 0.18518517911434174, 0.051851850003004074, 0.15555556118488312, 0.21481481194496155, 0.16296295821666718, 0.1111111119389534, 0.07407407462596893, 0.17037037014961243, 0.17037037014961243, 0.192592591047287, 0.192592591047287, 0.24444444477558136, 0.34074074029922485, 0.2888889014720917, 0.24444444477558136, 0.29629629850387573, 0.27407407760620117, 0.3777777850627899, 0.3777777850627899, 0.3629629611968994, 0.3333333432674408, 0.42222222685813904, 0.5185185074806213, 0.385185182094574, 0.3777777850627899, 0.40740740299224854, 0.4296296238899231, 0.4444444477558136, 0.4148148000240326, NaN, 0.46666666865348816, 0.5407407283782959, 0.5925925970077515, 0.6074073910713196, 0.8074073791503906, 0.5703703761100769, 0.6370370388031006, 0.6222222447395325, 0.6888889074325562, 0.6814814805984497, 0.7629629373550415, 0.9259259104728699, 0.8296296000480652, 0.8148148059844971, 0.8148148059844971, 0.9851852059364319, 0.8148148059844971, 0.8666666746139526, 0.8592592477798462, 0.9037036895751953, 0.9333333373069763, 0.9333333373069763, 1.0], [0.014814814552664757, 0.007407407276332378, 0.02222222276031971, 0.014814814552664757, 0.007407407276332378, 0.07407407462596893, 0.029629629105329514, 0.08888889104127884, 0.14814814925193787, 0.08148147910833359, 0.04444444552063942, 0.08148147910833359, 0.05925925821065903, 0.14814814925193787, 0.08148147910833359, 0.06666667014360428, 0.17777778208255768, 0.11851851642131805, 0.13333334028720856, 0.16296295821666718, 0.22962963581085205, 0.14074073731899261, 0.10370370000600815, 0.0962962955236435, 0.2074074000120163, 0.20000000298023224, 0.15555556118488312, 0.22962963581085205, 0.16296295821666718, 0.25925925374031067, 0.27407407760620117, 0.21481481194496155, 0.22962963581085205, 0.21481481194496155, 0.42222222685813904, 0.34074074029922485, 0.3037036955356598, 0.35555556416511536, 0.2888889014720917, 0.46666666865348816, 0.37037035822868347, 0.32592591643333435, 0.4444444477558136, 0.4000000059604645, 0.3037036955356598, 0.48148149251937866, 0.5333333611488342, NaN, 0.5555555820465088, 0.644444465637207, 0.6296296119689941, 0.6074073910713196, 0.6222222447395325, 0.6222222447395325, 0.6888889074325562, 0.6370370388031006, 0.6518518328666687, 0.770370364189148, 0.8629629611968994, 0.9037036895751953, 0.8592592477798462, 0.8666666746139526, 0.8222222328186035, 0.8074073791503906, 0.8370370268821716, 0.8518518805503845, 0.9555555582046509, 1.0, 0.9407407641410828, 1.0], [0.05925925821065903, 0.0, 0.06666667014360428, 0.02222222276031971, 0.08888889104127884, 0.10370370000600815, 0.17777778208255768, 0.014814814552664757, 0.08888889104127884, 0.08888889104127884, 0.20000000298023224, 0.11851851642131805, 0.15555556118488312, 0.1111111119389534, 0.029629629105329514, 0.13333334028720856, 0.11851851642131805, 0.08148147910833359, 0.2222222238779068, 0.2370370328426361, 0.2074074000120163, 0.16296295821666718, 0.16296295821666718, 0.16296295821666718, 0.13333334028720856, 0.2222222238779068, 0.192592591047287, 0.17777778208255768, 0.17777778208255768, 0.192592591047287, 0.3481481373310089, 0.25925925374031067, 0.2222222238779068, 0.2074074000120163, 0.2518518567085266, 0.4148148000240326, 0.22962963581085205, 0.35555556416511536, 0.4148148000240326, 0.3777777850627899, 0.3629629611968994, 0.3333333432674408, 0.4296296238899231, 0.3481481373310089, 0.40740740299224854, 0.46666666865348816, 0.4592592716217041, 0.4444444477558136, NaN, 0.46666666865348816, 0.45185184478759766, 0.48148149251937866, 0.5333333611488342, 0.5703703761100769, 0.8296296000480652, 0.585185170173645, 0.6666666865348816, 0.9777777791023254, 0.6000000238418579, 0.7185184955596924, 0.7185184955596924, 0.7185184955596924, 0.7555555701255798, 0.9259259104728699, 0.8074073791503906, 0.9333333373069763, 0.8444444537162781, 0.7925925850868225, 0.9851852059364319, 1.0], [0.014814814552664757, 0.029629629105329514, 0.051851850003004074, 0.05925925821065903, 0.029629629105329514, 0.051851850003004074, 0.0962962955236435, 0.07407407462596893, 0.0962962955236435, 0.08148147910833359, 0.1111111119389534, 0.08888889104127884, 0.0962962955236435, 0.15555556118488312, 0.08888889104127884, 0.0962962955236435, 0.13333334028720856, 0.14074073731899261, 0.06666667014360428, 0.21481481194496155, 0.17037037014961243, 0.14814814925193787, 0.18518517911434174, 0.10370370000600815, 0.192592591047287, 0.17777778208255768, 0.17777778208255768, 0.27407407760620117, 0.17037037014961243, 0.2222222238779068, 0.34074074029922485, 0.18518517911434174, 0.20000000298023224, 0.2370370328426361, 0.3333333432674408, 0.4444444477558136, 0.3185185194015503, 0.3333333432674408, 0.32592591643333435, 0.4000000059604645, 0.29629629850387573, 0.37037035822868347, 0.385185182094574, 0.42222222685813904, 0.3333333432674408, 0.48148149251937866, 0.40740740299224854, 0.35555556416511536, 0.5333333611488342, NaN, 0.40740740299224854, 0.5185185074806213, 0.6000000238418579, 0.6814814805984497, 0.6962962746620178, 0.7037037014961243, 0.6222222447395325, 0.7777777910232544, 0.6888889074325562, 0.8592592477798462, 0.7111111283302307, 0.7259259223937988, 0.770370364189148, 0.8740741014480591, 0.800000011920929, 0.8962963223457336, 0.970370352268219, 1.0, 0.9851852059364319, 1.0], [0.0, 0.0, 0.007407407276332378, 0.02222222276031971, 0.03703703731298447, 0.051851850003004074, 0.0962962955236435, 0.1259259283542633, 0.10370370000600815, 0.051851850003004074, 0.11851851642131805, 0.04444444552063942, 0.014814814552664757, 0.03703703731298447, 0.014814814552664757, 0.02222222276031971, 0.1111111119389534, 0.17777778208255768, 0.1111111119389534, 0.2074074000120163, 0.24444444477558136, 0.0962962955236435, 0.04444444552063942, 0.04444444552063942, 0.16296295821666718, 0.15555556118488312, 0.13333334028720856, 0.3037036955356598, 0.16296295821666718, 0.17777778208255768, 0.37037035822868347, 0.18518517911434174, 0.3481481373310089, 0.13333334028720856, 0.3037036955356598, 0.385185182094574, 0.32592591643333435, 0.35555556416511536, 0.3777777850627899, 0.4740740656852722, 0.27407407760620117, 0.32592591643333435, 0.4444444477558136, 0.32592591643333435, 0.3629629611968994, 0.43703705072402954, 0.39259257912635803, 0.37037035822868347, 0.5481481552124023, 0.5925925970077515, NaN, 0.4962962865829468, 0.5703703761100769, 0.5037037134170532, 0.6740740537643433, 0.4962962865829468, 0.6000000238418579, 0.7851851582527161, 0.6740740537643433, 0.800000011920929, 0.8740741014480591, 0.8814814686775208, 0.6370370388031006, 0.8370370268821716, 0.6222222447395325, 0.8666666746139526, 0.9925925731658936, 0.8666666746139526, 0.9259259104728699, 0.9555555582046509], [0.007407407276332378, 0.0, 0.02222222276031971, 0.0, 0.007407407276332378, 0.007407407276332378, 0.02222222276031971, 0.08148147910833359, 0.07407407462596893, 0.029629629105329514, 0.04444444552063942, 0.03703703731298447, 0.014814814552664757, 0.10370370000600815, 0.04444444552063942, 0.03703703731298447, 0.08888889104127884, 0.14074073731899261, 0.03703703731298447, 0.11851851642131805, 0.14074073731899261, 0.15555556118488312, 0.051851850003004074, 0.04444444552063942, 0.1111111119389534, 0.16296295821666718, 0.17777778208255768, 0.192592591047287, 0.15555556118488312, 0.28148147463798523, 0.2370370328426361, 0.20000000298023224, 0.21481481194496155, 0.24444444477558136, 0.3333333432674408, 0.35555556416511536, 0.34074074029922485, 0.3037036955356598, 0.3629629611968994, 0.4962962865829468, 0.3037036955356598, 0.3481481373310089, 0.3629629611968994, 0.3629629611968994, 0.3629629611968994, 0.3629629611968994, 0.192592591047287, 0.39259257912635803, 0.5185185074806213, 0.48148149251937866, 0.5037037134170532, NaN, 0.5111111402511597, 0.6074073910713196, 0.5925925970077515, 0.6666666865348816, 0.6592592597007751, 0.6962962746620178, 0.9037036895751953, 0.7925925850868225, 0.8074073791503906, 0.800000011920929, 0.9629629850387573, 0.8074073791503906, 0.8592592477798462, 0.8296296000480652, 0.8962963223457336, 0.9407407641410828, 0.9259259104728699, 1.0], [0.0, 0.0, 0.02222222276031971, 0.007407407276332378, 0.0, 0.014814814552664757, 0.029629629105329514, 0.007407407276332378, 0.07407407462596893, 0.03703703731298447, 0.05925925821065903, 0.04444444552063942, 0.04444444552063942, 0.08148147910833359, 0.007407407276332378, 0.029629629105329514, 0.014814814552664757, 0.04444444552063942, 0.029629629105329514, 0.13333334028720856, 0.14814814925193787, 0.17777778208255768, 0.06666667014360428, 0.029629629105329514, 0.014814814552664757, 0.0962962955236435, 0.18518517911434174, 0.16296295821666718, 0.07407407462596893, 0.21481481194496155, 0.21481481194496155, 0.007407407276332378, 0.17777778208255768, 0.11851851642131805, 0.28148147463798523, 0.2518518567085266, 0.27407407760620117, 0.20000000298023224, 0.16296295821666718, 0.40740740299224854, 0.07407407462596893, 0.27407407760620117, 0.25925925374031067, 0.2370370328426361, 0.2888889014720917, 0.2666666805744171, 0.4296296238899231, 0.3777777850627899, 0.46666666865348816, 0.4000000059604645, 0.4296296238899231, 0.4888888895511627, NaN, 0.6000000238418579, 0.6000000238418579, 0.614814817905426, 0.6074073910713196, 0.6370370388031006, 0.6888889074325562, 0.8370370268821716, 0.770370364189148, 0.7629629373550415, 0.7777777910232544, 0.8222222328186035, 0.770370364189148, 0.8444444537162781, 0.9185185432434082, 0.9925925731658936, 0.9185185432434082, 1.0], [0.0, 0.007407407276332378, 0.06666667014360428, 0.014814814552664757, 0.007407407276332378, 0.06666667014360428, 0.02222222276031971, 0.014814814552664757, 0.007407407276332378, 0.02222222276031971, 0.02222222276031971, 0.02222222276031971, 0.014814814552664757, 0.04444444552063942, 0.13333334028720856, 0.0962962955236435, 0.08148147910833359, 0.051851850003004074, 0.08148147910833359, 0.08888889104127884, 0.014814814552664757, 0.15555556118488312, 0.04444444552063942, 0.14074073731899261, 0.08148147910833359, 0.0962962955236435, 0.17037037014961243, 0.16296295821666718, 0.22962963581085205, 0.2074074000120163, 0.03703703731298447, 0.2666666805744171, 0.18518517911434174, 0.2518518567085266, 0.2074074000120163, 0.22962963581085205, 0.14074073731899261, 0.3481481373310089, 0.22962963581085205, 0.43703705072402954, 0.29629629850387573, 0.2518518567085266, 0.37037035822868347, 0.2666666805744171, 0.3481481373310089, 0.385185182094574, 0.3629629611968994, 0.3777777850627899, 0.4296296238899231, 0.3185185194015503, 0.4962962865829468, 0.39259257912635803, 0.4000000059604645, NaN, 0.46666666865348816, 0.8962963223457336, 0.644444465637207, 0.5333333611488342, 0.6740740537643433, 0.6296296119689941, 0.5111111402511597, 0.5111111402511597, 0.7111111283302307, 0.7037037014961243, 0.9481481313705444, 0.7111111283302307, 0.5777778029441833, 0.770370364189148, 0.7111111283302307, 0.970370352268219], [0.014814814552664757, 0.0, 0.04444444552063942, 0.014814814552664757, 0.03703703731298447, 0.04444444552063942, 0.08148147910833359, 0.014814814552664757, 0.04444444552063942, 0.02222222276031971, 0.08148147910833359, 0.029629629105329514, 0.05925925821065903, 0.014814814552664757, 0.029629629105329514, 0.08148147910833359, 0.05925925821065903, 0.02222222276031971, 0.13333334028720856, 0.17037037014961243, 0.13333334028720856, 0.14074073731899261, 0.14074073731899261, 0.08888889104127884, 0.08148147910833359, 0.051851850003004074, 0.16296295821666718, 0.13333334028720856, 0.0962962955236435, 0.15555556118488312, 0.21481481194496155, 0.11851851642131805, 0.11851851642131805, 0.192592591047287, 0.21481481194496155, 0.32592591643333435, 0.14814814925193787, 0.29629629850387573, 0.3037036955356598, 0.2888889014720917, 0.2518518567085266, 0.11851851642131805, 0.31111112236976624, 0.2666666805744171, 0.2666666805744171, 0.31111112236976624, 0.3777777850627899, 0.31111112236976624, 0.17037037014961243, 0.3037036955356598, 0.32592591643333435, 0.40740740299224854, 0.4000000059604645, 0.5333333611488342, NaN, 0.5629629492759705, 0.5925925970077515, 0.7777777910232544, 0.5333333611488342, 0.6074073910713196, 0.6370370388031006, 0.6370370388031006, 0.6666666865348816, 0.7777777910232544, 0.7333333492279053, 0.8222222328186035, 0.8518518805503845, 0.7851851582527161, 0.9333333373069763, 1.0], [0.0, 0.0, 0.014814814552664757, 0.02222222276031971, 0.007407407276332378, 0.06666667014360428, 0.014814814552664757, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.03703703731298447, 0.007407407276332378, 0.014814814552664757, 0.029629629105329514, 0.06666667014360428, 0.03703703731298447, 0.06666667014360428, 0.029629629105329514, 0.07407407462596893, 0.03703703731298447, 0.014814814552664757, 0.14074073731899261, 0.02222222276031971, 0.04444444552063942, 0.07407407462596893, 0.051851850003004074, 0.15555556118488312, 0.0962962955236435, 0.17777778208255768, 0.17037037014961243, 0.029629629105329514, 0.24444444477558136, 0.15555556118488312, 0.11851851642131805, 0.17037037014961243, 0.13333334028720856, 0.11851851642131805, 0.35555556416511536, 0.15555556118488312, 0.39259257912635803, 0.2666666805744171, 0.2074074000120163, 0.3629629611968994, 0.18518517911434174, 0.3185185194015503, 0.3629629611968994, 0.31111112236976624, 0.3629629611968994, 0.4148148000240326, 0.29629629850387573, 0.5037037134170532, 0.3333333432674408, 0.385185182094574, 0.10370370000600815, 0.43703705072402954, NaN, 0.585185170173645, 0.5185185074806213, 0.6518518328666687, 0.6000000238418579, 0.5111111402511597, 0.5111111402511597, 0.614814817905426, 0.6814814805984497, 0.9259259104728699, 0.6962962746620178, 0.5407407283782959, 0.7111111283302307, 0.6888889074325562, 0.970370352268219], [0.029629629105329514, 0.0, 0.051851850003004074, 0.0, 0.051851850003004074, 0.06666667014360428, 0.051851850003004074, 0.0, 0.0, 0.07407407462596893, 0.06666667014360428, 0.06666667014360428, 0.0962962955236435, 0.07407407462596893, 0.02222222276031971, 0.08888889104127884, 0.1111111119389534, 0.051851850003004074, 0.06666667014360428, 0.05925925821065903, 0.051851850003004074, 0.14074073731899261, 0.13333334028720856, 0.13333334028720856, 0.11851851642131805, 0.08148147910833359, 0.14814814925193787, 0.0, 0.05925925821065903, 0.14814814925193787, 0.06666667014360428, 0.17777778208255768, 0.03703703731298447, 0.2222222238779068, 0.20000000298023224, 0.18518517911434174, 0.25925925374031067, 0.32592591643333435, 0.13333334028720856, 0.3333333432674408, 0.22962963581085205, 0.31111112236976624, 0.34074074029922485, 0.2074074000120163, 0.3185185194015503, 0.3777777850627899, 0.3185185194015503, 0.3481481373310089, 0.3333333432674408, 0.3777777850627899, 0.4000000059604645, 0.34074074029922485, 0.39259257912635803, 0.35555556416511536, 0.40740740299224854, 0.4148148000240326, NaN, 0.4592592716217041, 0.4444444477558136, 0.5259259343147278, 0.5333333611488342, 0.5333333611488342, 0.4888888895511627, 0.48148149251937866, 0.614814817905426, 0.5185185074806213, 0.5777778029441833, 0.6592592597007751, 0.5259259343147278, 0.9111111164093018], [0.014814814552664757, 0.0, 0.04444444552063942, 0.007407407276332378, 0.051851850003004074, 0.051851850003004074, 0.06666667014360428, 0.02222222276031971, 0.02222222276031971, 0.014814814552664757, 0.07407407462596893, 0.03703703731298447, 0.08888889104127884, 0.029629629105329514, 0.007407407276332378, 0.06666667014360428, 0.06666667014360428, 0.029629629105329514, 0.10370370000600815, 0.08148147910833359, 0.1111111119389534, 0.14814814925193787, 0.1259259283542633, 0.0962962955236435, 0.05925925821065903, 0.05925925821065903, 0.15555556118488312, 0.08888889104127884, 0.06666667014360428, 0.15555556118488312, 0.14814814925193787, 0.1111111119389534, 0.08888889104127884, 0.15555556118488312, 0.21481481194496155, 0.24444444477558136, 0.1259259283542633, 0.21481481194496155, 0.21481481194496155, 0.25925925374031067, 0.2222222238779068, 0.10370370000600815, 0.2518518567085266, 0.20000000298023224, 0.20000000298023224, 0.2888889014720917, 0.2370370328426361, 0.22962963581085205, 0.02222222276031971, 0.2222222238779068, 0.21481481194496155, 0.3037036955356598, 0.3629629611968994, 0.46666666865348816, 0.2222222238779068, 0.48148149251937866, 0.5407407283782959, NaN, 0.4148148000240326, 0.5333333611488342, 0.5703703761100769, 0.5777778029441833, 0.43703705072402954, 0.7037037014961243, 0.614814817905426, 0.7555555701255798, 0.7333333492279053, 0.6740740537643433, 0.7851851582527161, 0.8666666746139526], [0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.0, 0.007407407276332378, 0.014814814552664757, 0.0, 0.007407407276332378, 0.05925925821065903, 0.014814814552664757, 0.029629629105329514, 0.007407407276332378, 0.06666667014360428, 0.007407407276332378, 0.0, 0.02222222276031971, 0.02222222276031971, 0.07407407462596893, 0.08148147910833359, 0.02222222276031971, 0.06666667014360428, 0.06666667014360428, 0.05925925821065903, 0.13333334028720856, 0.051851850003004074, 0.16296295821666718, 0.21481481194496155, 0.13333334028720856, 0.28148147463798523, 0.04444444552063942, 0.22962963581085205, 0.14814814925193787, 0.17777778208255768, 0.32592591643333435, 0.07407407462596893, 0.029629629105329514, 0.34074074029922485, 0.07407407462596893, 0.13703703880310059, 0.4000000059604645, 0.31111112236976624, 0.32592591643333435, 0.0962962955236435, 0.31111112236976624, 0.32592591643333435, 0.46666666865348816, 0.3481481373310089, 0.5555555820465088, 0.585185170173645, NaN, 0.5185185074806213, 0.5925925970077515, 0.5925925970077515, 0.585185170173645, 0.6296296119689941, 0.7629629373550415, 0.6592592597007751, 0.6740740537643433, 0.770370364189148, 0.6740740537643433, 1.0], [0.029629629105329514, 0.007407407276332378, 0.04444444552063942, 0.007407407276332378, 0.03703703731298447, 0.029629629105329514, 0.05925925821065903, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.07407407462596893, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.03703703731298447, 0.02222222276031971, 0.06666667014360428, 0.0962962955236435, 0.10370370000600815, 0.03703703731298447, 0.05925925821065903, 0.007407407276332378, 0.07407407462596893, 0.029629629105329514, 0.051851850003004074, 0.02222222276031971, 0.007407407276332378, 0.06666667014360428, 0.13333334028720856, 0.05925925821065903, 0.014814814552664757, 0.07407407462596893, 0.14814814925193787, 0.25925925374031067, 0.10370370000600815, 0.16296295821666718, 0.1111111119389534, 0.2888889014720917, 0.10370370000600815, 0.14074073731899261, 0.2074074000120163, 0.15555556118488312, 0.07407407462596893, 0.21481481194496155, 0.17037037014961243, 0.0962962955236435, 0.28148147463798523, 0.14074073731899261, 0.20000000298023224, 0.2074074000120163, 0.16296295821666718, 0.37037035822868347, 0.39259257912635803, 0.4000000059604645, 0.4740740656852722, 0.46666666865348816, 0.48148149251937866, NaN, 0.39259257912635803, 0.39259257912635803, 0.46666666865348816, 0.7185184955596924, 0.6666666865348816, 0.7555555701255798, 0.5925925970077515, 1.0, 0.8148148059844971, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.014814814552664757, 0.0, 0.0, 0.051851850003004074, 0.0, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.0, 0.007407407276332378, 0.007407407276332378, 0.0, 0.014814814552664757, 0.0, 0.07407407462596893, 0.0962962955236435, 0.06666667014360428, 0.0, 0.007407407276332378, 0.0, 0.029629629105329514, 0.07407407462596893, 0.0962962955236435, 0.03703703731298447, 0.07407407462596893, 0.2222222238779068, 0.051851850003004074, 0.1111111119389534, 0.03703703731298447, 0.13333334028720856, 0.2222222238779068, 0.06666667014360428, 0.02222222276031971, 0.17037037014961243, 0.3185185194015503, 0.10370370000600815, 0.15555556118488312, 0.051851850003004074, 0.10370370000600815, 0.13333334028720856, 0.14074073731899261, 0.18518517911434174, 0.14074073731899261, 0.28148147463798523, 0.2888889014720917, 0.1259259283542633, 0.192592591047287, 0.22962963581085205, 0.4888888895511627, 0.3629629611968994, 0.4888888895511627, 0.46666666865348816, 0.4296296238899231, 0.40740740299224854, 0.6074073910713196, NaN, 0.5185185074806213, 0.5555555820465088, 0.7111111283302307, 0.5333333611488342, 0.7185184955596924, 1.0, 0.7851851582527161, 0.8296296000480652, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.014814814552664757, 0.0, 0.0, 0.051851850003004074, 0.0, 0.0, 0.007407407276332378, 0.007407407276332378, 0.0, 0.007407407276332378, 0.007407407276332378, 0.0, 0.014814814552664757, 0.0, 0.07407407462596893, 0.0962962955236435, 0.06666667014360428, 0.0, 0.007407407276332378, 0.0, 0.029629629105329514, 0.07407407462596893, 0.0962962955236435, 0.03703703731298447, 0.06666667014360428, 0.21481481194496155, 0.051851850003004074, 0.10370370000600815, 0.029629629105329514, 0.13333334028720856, 0.2222222238779068, 0.06666667014360428, 0.05925925821065903, 0.16296295821666718, 0.31111112236976624, 0.10370370000600815, 0.15555556118488312, 0.08148147910833359, 0.0962962955236435, 0.1259259283542633, 0.15555556118488312, 0.18518517911434174, 0.13333334028720856, 0.28148147463798523, 0.27407407760620117, 0.11851851642131805, 0.20000000298023224, 0.2370370328426361, 0.4888888895511627, 0.3629629611968994, 0.4888888895511627, 0.46666666865348816, 0.42222222685813904, 0.40740740299224854, 0.6074073910713196, 0.48148149251937866, NaN, 0.5555555820465088, 0.7111111283302307, 0.5333333611488342, 0.7185184955596924, 1.0, 0.7851851582527161, 0.8296296000480652, 0.800000011920929], [0.007407407276332378, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.007407407276332378, 0.0, 0.014814814552664757, 0.007407407276332378, 0.014814814552664757, 0.014814814552664757, 0.007407407276332378, 0.02222222276031971, 0.03703703731298447, 0.007407407276332378, 0.0, 0.03703703731298447, 0.007407407276332378, 0.051851850003004074, 0.007407407276332378, 0.07407407462596893, 0.007407407276332378, 0.007407407276332378, 0.014814814552664757, 0.05925925821065903, 0.08148147910833359, 0.014814814552664757, 0.051851850003004074, 0.08148147910833359, 0.04444444552063942, 0.0962962955236435, 0.051851850003004074, 0.1259259283542633, 0.11851851642131805, 0.2222222238779068, 0.08148147910833359, 0.27407407760620117, 0.1111111119389534, 0.3037036955356598, 0.18518517911434174, 0.08148147910833359, 0.31111112236976624, 0.05925925821065903, 0.14074073731899261, 0.3185185194015503, 0.014814814552664757, 0.17777778208255768, 0.24444444477558136, 0.22962963581085205, 0.3629629611968994, 0.03703703731298447, 0.2222222238779068, 0.2888889014720917, 0.3333333432674408, 0.385185182094574, 0.5111111402511597, 0.5629629492759705, 0.4148148000240326, 0.5333333611488342, 0.4444444477558136, 0.4444444477558136, NaN, 0.6370370388031006, 0.7259259223937988, 0.6592592597007751, 0.5703703761100769, 0.7037037014961243, 0.6814814805984497, 1.0], [0.014814814552664757, 0.0, 0.051851850003004074, 0.0, 0.02222222276031971, 0.051851850003004074, 0.10370370000600815, 0.0, 0.06666667014360428, 0.007407407276332378, 0.11851851642131805, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.014814814552664757, 0.0, 0.04444444552063942, 0.014814814552664757, 0.10370370000600815, 0.17777778208255768, 0.16296295821666718, 0.0, 0.08888889104127884, 0.0, 0.07407407462596893, 0.08888889104127884, 0.0, 0.014814814552664757, 0.06666667014360428, 0.0, 0.2074074000120163, 0.05925925821065903, 0.029629629105329514, 0.08148147910833359, 0.18518517911434174, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.2370370328426361, 0.08148147910833359, 0.1111111119389534, 0.0962962955236435, 0.2074074000120163, 0.13333334028720856, 0.1259259283542633, 0.18518517911434174, 0.192592591047287, 0.07407407462596893, 0.1259259283542633, 0.16296295821666718, 0.192592591047287, 0.17777778208255768, 0.29629629850387573, 0.2222222238779068, 0.3185185194015503, 0.5185185074806213, 0.29629629850387573, 0.37037035822868347, 0.28148147463798523, 0.2888889014720917, 0.2888889014720917, 0.3629629611968994, NaN, 0.5111111402511597, 0.8370370268821716, 0.3333333432674408, 0.4148148000240326, 0.9555555582046509, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.06666667014360428, 0.0, 0.05925925821065903, 0.0, 0.0, 0.11851851642131805, 0.0, 0.0, 0.06666667014360428, 0.0, 0.1111111119389534, 0.06666667014360428, 0.05925925821065903, 0.13333334028720856, 0.0, 0.07407407462596893, 0.07407407462596893, 0.0, 0.06666667014360428, 0.014814814552664757, 0.06666667014360428, 0.24444444477558136, 0.051851850003004074, 0.2074074000120163, 0.0962962955236435, 0.06666667014360428, 0.2666666805744171, 0.06666667014360428, 0.15555556118488312, 0.27407407760620117, 0.13333334028720856, 0.16296295821666718, 0.192592591047287, 0.20000000298023224, 0.3777777850627899, 0.14074073731899261, 0.22962963581085205, 0.051851850003004074, 0.2666666805744171, 0.07407407462596893, 0.385185182094574, 0.385185182094574, 0.2370370328426361, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.27407407760620117, 0.4888888895511627, NaN, 0.5037037134170532, 0.46666666865348816, 0.45185184478759766, 0.5037037134170532, 0.9185185432434082], [0.014814814552664757, 0.0, 0.04444444552063942, 0.0, 0.02222222276031971, 0.03703703731298447, 0.0962962955236435, 0.0, 0.05925925821065903, 0.0, 0.08888889104127884, 0.0, 0.007407407276332378, 0.0, 0.007407407276332378, 0.0, 0.03703703731298447, 0.007407407276332378, 0.10370370000600815, 0.16296295821666718, 0.16296295821666718, 0.0, 0.05925925821065903, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.04444444552063942, 0.0, 0.17037037014961243, 0.05925925821065903, 0.007407407276332378, 0.07407407462596893, 0.16296295821666718, 0.192592591047287, 0.14074073731899261, 0.051851850003004074, 0.17037037014961243, 0.22962963581085205, 0.08148147910833359, 0.08148147910833359, 0.07407407462596893, 0.20000000298023224, 0.0962962955236435, 0.10370370000600815, 0.14074073731899261, 0.14814814925193787, 0.06666667014360428, 0.10370370000600815, 0.13333334028720856, 0.17037037014961243, 0.15555556118488312, 0.2888889014720917, 0.17777778208255768, 0.3037036955356598, 0.48148149251937866, 0.24444444477558136, 0.34074074029922485, 0.24444444477558136, 0.28148147463798523, 0.28148147463798523, 0.34074074029922485, 0.16296295821666718, 0.4962962865829468, NaN, 0.3333333432674408, 0.4148148000240326, 0.8814814686775208, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.051851850003004074, 0.0, 0.0, 0.0, 0.0, 0.051851850003004074, 0.06666667014360428, 0.007407407276332378, 0.06666667014360428, 0.05925925821065903, 0.0, 0.007407407276332378, 0.0, 0.051851850003004074, 0.192592591047287, 0.0, 0.0, 0.0, 0.2518518567085266, 0.014814814552664757, 0.02222222276031971, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0962962955236435, 0.04444444552063942, 0.15555556118488312, 0.029629629105329514, 0.007407407276332378, 0.10370370000600815, 0.08148147910833359, 0.42222222685813904, 0.14814814925193787, 0.4592592716217041, 0.42222222685813904, 0.2666666805744171, 0.32592591643333435, 0.40740740299224854, 0.0, 0.0, 0.4296296238899231, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, NaN, 0.6666666865348816, 0.6814814805984497, 0.7851851582527161], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.014814814552664757, 0.0, 0.0, 0.0, 0.02222222276031971, 0.17777778208255768, 0.0, 0.07407407462596893, 0.0, 0.21481481194496155, 0.0, 0.0, 0.08148147910833359, 0.0, 0.0, 0.11851851642131805, 0.06666667014360428, 0.0, 0.2074074000120163, 0.0, 0.13333334028720856, 0.05925925821065903, 0.007407407276332378, 0.22962963581085205, 0.21481481194496155, 0.2888889014720917, 0.34074074029922485, 0.32592591643333435, 0.22962963581085205, 0.0, 0.21481481194496155, 0.21481481194496155, 0.29629629850387573, 0.585185170173645, 0.5481481552124023, 0.585185170173645, 0.3333333432674408, NaN, 0.585185170173645, 0.9925925731658936], [0.0, 0.0, 0.029629629105329514, 0.0, 0.02222222276031971, 0.02222222276031971, 0.029629629105329514, 0.0, 0.014814814552664757, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.014814814552664757, 0.0, 0.051851850003004074, 0.08148147910833359, 0.07407407462596893, 0.0, 0.02222222276031971, 0.0, 0.04444444552063942, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.13333334028720856, 0.029629629105329514, 0.0, 0.06666667014360428, 0.1259259283542633, 0.17037037014961243, 0.06666667014360428, 0.03703703731298447, 0.08888889104127884, 0.22962963581085205, 0.051851850003004074, 0.05925925821065903, 0.051851850003004074, 0.1259259283542633, 0.029629629105329514, 0.05925925821065903, 0.06666667014360428, 0.05925925821065903, 0.014814814552664757, 0.014814814552664757, 0.07407407462596893, 0.07407407462596893, 0.08148147910833359, 0.2888889014720917, 0.06666667014360428, 0.31111112236976624, 0.4740740656852722, 0.21481481194496155, 0.32592591643333435, 0.18518517911434174, 0.17037037014961243, 0.17037037014961243, 0.3185185194015503, 0.04444444552063942, 0.4962962865829468, 0.11851851642131805, 0.3185185194015503, 0.4148148000240326, NaN, 0.6296296119689941], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.07407407462596893, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.029629629105329514, 0.0, 0.029629629105329514, 0.08888889104127884, 0.13333334028720856, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.3333333432674408, 0.08148147910833359, 0.3333333432674408, 0.21481481194496155, 0.007407407276332378, 0.37037035822868347, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9506172875},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9228126679},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8761137961},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8662372525},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.834997315},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8184648422},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7816425088},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7656468085},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7575952743},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.754106281},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7456790141},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7420289829},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7363392413},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7253891578},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.710359636},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.701341922},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6975845458},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6881374142},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6856682808},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6820182487},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6779388094},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6734299509},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.663768115},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6589371964},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6397208817},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6316156723},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.629307567},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5846484163},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5842190019},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5797101451},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5655931285},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5393451423},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5360171768},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5342995161},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5074610853},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4950617294},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4894256596},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4892109504},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4622651647},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4444444459},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4380032221},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4371443926},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4312399366},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4242619443},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.410950081},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.3892646288},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3855072452},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3834138501},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3711218459},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3709071414},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3517981738},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3468599022},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3058507796},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2923242102},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2852388644},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2526033287},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2425120795},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2313472906},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2093934502},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.20203972},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2009661834},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2003220614},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1935587769},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1673644662},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1477187349},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1426731077},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1202361794},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1046698867},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.097477187},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.034031133}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_all/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9800e71a9690a2f769315f986f1cfa82f717ad45
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,71 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),"AutoGluon 1.4 (best, 4h)",TabPFN-2.6 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),RealMLP (tuned + ensembled),ModernNCA (tuned + ensembled),CatBoost (tuned + ensembled),RealTabPFN-2.5 (tuned),CatBoost (tuned),LightGBM (tuned + ensembled),CatBoost (default),iLTM (tuned + ensembled),XGBoost (tuned + ensembled),TabM (tuned + ensembled),RealMLP (tuned),RealTabPFN-2.5 (default),TabDPT-Turbo (default),TabDPT (tuned + ensembled),ChimeraBoost (tuned + ensembled),LightGBM (tuned),XGBoost (tuned),TabM (tuned),ModernNCA (tuned),ChimeraBoost (tuned),xRFM (tuned + ensembled),iLTM (tuned),ChimeraBoost (default),TabDPT (tuned),TorchMLP (tuned + ensembled),xRFM (tuned),XGBoost (default),TabSwift (default),TabDPT (default),TabM (default),EBM (tuned + ensembled),ModernNCA (default),LimiX (default),TorchMLP (tuned),RealMLP (default),EBM (tuned),LightGBM (default),RandomForest (tuned + ensembled),EBM (default),ExtraTrees (tuned + ensembled),RandomForest (tuned),FastaiMLP (tuned + ensembled),SAP-RPT-OSS (default),iLTM (default),ExtraTrees (tuned),TorchMLP (default),PerpetualBooster (tuned + ensembled),FastaiMLP (tuned),PerpetualBooster (tuned),xRFM (default),FastaiMLP (default),RandomForest (default),KNN (tuned + ensembled),TabSTAR (tuned),TabSTAR (tuned + ensembled),ExtraTrees (default),Linear (tuned + ensembled),PerpetualBooster (default),Linear (tuned),TabSTAR (default),KNN (tuned),Linear (default),KNN (default)
+TabFM (default),,0.6666667,0.78518516,0.75555557,0.93333334,0.9037037,0.9111111,0.88148147,0.9259259,0.8518519,0.91851854,0.88148147,0.9037037,0.85925925,0.8962963,0.91851854,0.9925926,0.8888889,0.9851852,0.962963,0.8740741,0.91851854,0.9777778,0.94074076,1.0,0.962963,0.91851854,0.9259259,0.93333334,0.93333334,0.9259259,0.9777778,0.94074076,0.9851852,0.97037035,0.94074076,1.0,0.9925926,1.0,0.9037037,1.0,1.0,1.0,1.0,0.9851852,1.0,0.9925926,0.9851852,0.94074076,0.9851852,1.0,0.9925926,1.0,1.0,0.9851852,1.0,0.97037035,0.9851852,0.9925926,0.97037035,1.0,1.0,0.9925926,0.9851852,1.0,0.9851852,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.33333334,,0.36296296,0.82222223,0.73333335,0.5185185,0.7111111,0.8962963,0.8296296,0.94814813,0.77037036,0.94074076,0.95555556,0.93333334,0.95555556,0.97037035,0.8740741,0.9111111,0.78518516,0.7259259,0.73333335,0.94814813,0.9851852,0.9925926,0.9111111,0.9777778,0.95555556,0.962963,0.9259259,0.9777778,0.7777778,0.9925926,0.9925926,0.9777778,0.962963,0.8518519,1.0,0.9925926,0.9851852,0.85925925,0.9851852,1.0,1.0,1.0,1.0,1.0,1.0,0.9925926,1.0,0.97037035,1.0,1.0,1.0,0.9925926,1.0,1.0,1.0,1.0,1.0,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.21481481,0.63703704,,0.65185183,0.77037036,0.65925926,0.8296296,0.76296294,0.85925925,0.73333335,0.84444445,0.7259259,0.75555557,0.7259259,0.78518516,0.76296294,0.85925925,0.77037036,0.91851854,0.91851854,0.73333335,0.78518516,0.8666667,0.78518516,0.9037037,0.8518519,0.7777778,0.8,0.9111111,0.7777778,0.82222223,0.8888889,0.8666667,0.8740741,0.9777778,0.8740741,0.9925926,0.84444445,0.9851852,0.8666667,0.9777778,0.962963,0.8740741,0.9925926,0.962963,0.91851854,0.9777778,0.9777778,0.93333334,0.94814813,0.9925926,0.9777778,0.9777778,0.93333334,0.95555556,0.9851852,0.94814813,0.95555556,1.0,0.95555556,1.0,1.0,1.0,0.94814813,1.0,0.95555556,1.0,1.0,0.97037035,1.0
+"AutoGluon 1.4 (best, 4h)",0.24444444,0.17777778,0.34814814,,0.5925926,0.5037037,0.5037037,0.7037037,0.6666667,0.82222223,0.6,0.8296296,0.8740741,0.8074074,0.8888889,0.9111111,0.7925926,0.8740741,0.64444447,0.6296296,0.6296296,0.8518519,0.9111111,0.9259259,0.8518519,0.837037,0.9111111,0.91851854,0.8962963,0.9259259,0.77037036,0.94074076,0.97037035,0.94814813,0.84444445,0.8296296,0.94814813,0.97037035,0.962963,0.8518519,0.9851852,0.9777778,0.97037035,0.97037035,0.9925926,0.97037035,0.9925926,0.9851852,0.9777778,0.94074076,0.9777778,1.0,0.9925926,0.9851852,0.9851852,0.9777778,1.0,0.9925926,1.0,0.9925926,1.0,1.0,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.06666667,0.26666668,0.22962964,0.4074074,,0.45925927,0.5555556,0.6888889,0.67407405,0.65185183,0.64444447,0.65185183,0.73333335,0.65185183,0.7037037,0.7185185,0.73333335,0.7925926,0.7777778,0.7185185,0.7407407,0.77037036,0.8296296,0.82222223,0.82222223,0.88148147,0.8,0.8074074,0.837037,0.78518516,0.8296296,0.8962963,0.8666667,0.94074076,0.9111111,0.8296296,0.94814813,0.91851854,0.9851852,0.8518519,0.97037035,1.0,0.9777778,0.962963,0.97037035,0.9777778,0.9851852,0.9925926,0.9111111,0.97037035,0.962963,0.9925926,1.0,0.9925926,0.962963,0.9925926,0.94814813,0.94814813,1.0,0.962963,1.0,1.0,1.0,0.9777778,1.0,0.9777778,1.0,1.0,0.9777778,1.0
+TabICLv2 (default),0.096296296,0.4814815,0.34074074,0.4962963,0.5407407,,0.73333335,0.64444447,0.7407407,0.67407405,0.75555557,0.6666667,0.7111111,0.67407405,0.6888889,0.6814815,0.7407407,0.6814815,0.82222223,0.78518516,0.77037036,0.6814815,0.7777778,0.6962963,0.8148148,0.8,0.6814815,0.73333335,0.77037036,0.6962963,0.82222223,0.8074074,0.74814814,0.84444445,0.94074076,0.82222223,0.962963,0.77037036,0.9037037,0.8740741,0.88148147,0.94074076,0.8666667,0.94814813,0.91851854,0.8888889,0.9925926,0.9259259,0.8962963,0.94814813,0.94814813,0.9925926,0.9851852,0.93333334,0.95555556,0.93333334,0.93333334,0.94814813,1.0,0.97037035,0.9851852,0.9851852,1.0,0.94814813,0.93333334,0.962963,1.0,1.0,0.9777778,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.08888889,0.2888889,0.17037037,0.4962963,0.44444445,0.26666668,,0.6,0.5777778,0.58518517,0.82222223,0.6074074,0.6296296,0.6,0.6148148,0.65925926,0.6148148,0.6888889,0.8074074,0.6296296,0.65925926,0.6074074,0.6962963,0.6814815,0.6962963,0.7925926,0.6666667,0.76296294,0.74814814,0.7407407,0.7925926,0.8074074,0.8074074,0.8296296,0.9037037,0.8,0.8296296,0.82222223,0.88148147,0.7777778,0.88148147,0.88148147,0.88148147,0.94814813,0.94074076,0.9111111,0.9777778,0.97037035,0.82222223,0.9037037,0.9037037,0.9777778,0.97037035,0.9777778,0.91851854,0.9851852,0.94814813,0.93333334,1.0,0.94074076,1.0,1.0,0.9925926,0.8962963,1.0,0.9037037,1.0,1.0,0.97037035,1.0
+RealMLP (tuned + ensembled),0.11851852,0.1037037,0.23703703,0.2962963,0.31111112,0.35555556,0.4,,0.53333336,0.5185185,0.45925927,0.54814816,0.6074074,0.5185185,0.63703704,0.63703704,0.5925926,0.77037036,0.5925926,0.5925926,0.5777778,0.5185185,0.7037037,0.6888889,0.64444447,0.7185185,0.6148148,0.85925925,0.74814814,0.64444447,0.7185185,0.837037,0.91851854,0.78518516,0.75555557,0.74814814,0.93333334,0.82222223,0.9037037,0.8888889,0.8888889,1.0,0.88148147,0.95555556,0.91851854,0.8962963,0.93333334,0.9111111,0.9851852,0.9259259,0.8740741,0.91851854,0.9925926,0.9851852,0.9851852,0.9925926,1.0,0.9777778,0.95555556,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.074074075,0.17037037,0.14074074,0.33333334,0.32592592,0.25925925,0.42222223,0.46666667,,0.52592593,0.4962963,0.5185185,0.5777778,0.56296295,0.6148148,0.6296296,0.6,0.67407405,0.63703704,0.54814816,0.63703704,0.62222224,0.6296296,0.64444447,0.6962963,0.8296296,0.6962963,0.75555557,0.75555557,0.74814814,0.76296294,0.82222223,0.84444445,0.76296294,0.78518516,0.8148148,0.88148147,0.82222223,0.97037035,0.88148147,0.8962963,0.9259259,0.837037,0.93333334,0.85925925,0.88148147,0.93333334,0.8518519,0.9111111,0.9037037,0.8962963,0.9259259,0.9259259,0.9925926,0.95555556,0.9925926,1.0,0.9777778,1.0,0.9925926,0.94814813,0.94814813,0.9851852,0.93333334,1.0,0.94074076,1.0,1.0,0.9851852,1.0
+CatBoost (tuned + ensembled),0.14814815,0.05185185,0.26666668,0.17777778,0.34814814,0.32592592,0.4148148,0.4814815,0.47407407,,0.42962962,0.6148148,0.5185185,0.6666667,0.6,0.6888889,0.4888889,0.6148148,0.5555556,0.46666667,0.53333336,0.6962963,0.64444447,0.76296294,0.5555556,0.6703704,0.8074074,0.7185185,0.74814814,0.85925925,0.63703704,0.7925926,0.7925926,0.8740741,0.74814814,0.6888889,0.82222223,0.8962963,0.82222223,0.6814815,0.9037037,0.8888889,0.88148147,0.962963,0.9259259,0.962963,0.97037035,0.91851854,0.9111111,0.91851854,0.94814813,0.97037035,0.962963,0.9777778,0.9777778,0.9925926,0.9259259,0.9851852,0.9925926,0.9925926,1.0,1.0,0.9925926,0.9925926,1.0,1.0,1.0,0.9925926,1.0,1.0
+RealTabPFN-2.5 (tuned),0.08148148,0.22962964,0.15555556,0.4,0.35555556,0.24444444,0.17777778,0.5407407,0.5037037,0.5703704,,0.5777778,0.6,0.5777778,0.58518517,0.62222224,0.6074074,0.6296296,0.7111111,0.5777778,0.63703704,0.5703704,0.7037037,0.63703704,0.6888889,0.7407407,0.63703704,0.74814814,0.7185185,0.7111111,0.7407407,0.77037036,0.7925926,0.78518516,0.8888889,0.75555557,0.78518516,0.7777778,0.85925925,0.77037036,0.8740741,0.84444445,0.82222223,0.9259259,0.9037037,0.8888889,0.94074076,0.95555556,0.8,0.8888889,0.88148147,0.95555556,0.94074076,0.9777778,0.91851854,0.962963,0.93333334,0.9259259,0.9925926,0.9259259,0.9925926,1.0,0.9851852,0.88148147,1.0,0.9111111,1.0,1.0,0.95555556,1.0
+CatBoost (tuned),0.11851852,0.05925926,0.27407408,0.17037037,0.34814814,0.33333334,0.39259258,0.45185184,0.4814815,0.38518518,0.42222223,,0.44444445,0.58518517,0.58518517,0.65925926,0.4962963,0.6074074,0.52592593,0.45925927,0.53333336,0.7185185,0.63703704,0.7259259,0.5407407,0.63703704,0.8074074,0.7259259,0.7037037,0.8296296,0.6296296,0.78518516,0.78518516,0.8666667,0.7407407,0.6814815,0.84444445,0.8888889,0.8074074,0.6814815,0.8962963,0.8962963,0.88148147,0.94074076,0.91851854,0.9777778,0.94814813,0.91851854,0.88148147,0.9111111,0.95555556,0.962963,0.95555556,0.9777778,0.97037035,0.9925926,0.93333334,0.962963,0.9925926,0.9925926,0.9925926,0.9925926,0.9851852,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.096296296,0.044444446,0.24444444,0.12592593,0.26666668,0.2888889,0.37037036,0.39259258,0.42222223,0.4814815,0.4,0.5555556,,0.5777778,0.5407407,0.6888889,0.47407407,0.52592593,0.4962963,0.5185185,0.4962963,0.64444447,0.82222223,0.7777778,0.5925926,0.6148148,0.7407407,0.64444447,0.7259259,0.74814814,0.6296296,0.75555557,0.7185185,0.88148147,0.7259259,0.74814814,0.8518519,0.84444445,0.84444445,0.65925926,0.8740741,0.88148147,0.9111111,0.9851852,0.93333334,0.9777778,0.9851852,0.94074076,0.84444445,0.9037037,0.9851852,0.9851852,0.95555556,0.9851852,0.94074076,0.9851852,0.9037037,0.9111111,1.0,0.9851852,0.9925926,0.9925926,0.9925926,0.9851852,1.0,0.9925926,1.0,1.0,1.0,1.0
+CatBoost (default),0.14074074,0.06666667,0.27407408,0.19259259,0.34814814,0.32592592,0.4,0.4814815,0.43703705,0.33333334,0.42222223,0.4148148,0.42222223,,0.56296295,0.5777778,0.4888889,0.58518517,0.5407407,0.45925927,0.54814816,0.6074074,0.6074074,0.67407405,0.5703704,0.65185183,0.73333335,0.7259259,0.73333335,0.8,0.63703704,0.7925926,0.8,0.7925926,0.77037036,0.6962963,0.77037036,0.8740741,0.84444445,0.7037037,0.9037037,0.8074074,0.8740741,0.8740741,0.8518519,0.9259259,0.88148147,0.8518519,0.8888889,0.84444445,0.962963,0.8962963,0.91851854,0.95555556,0.9851852,0.97037035,0.9259259,0.97037035,0.9925926,0.9925926,1.0,1.0,0.9777778,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.1037037,0.044444446,0.21481481,0.11111111,0.2962963,0.31111112,0.38518518,0.36296296,0.38518518,0.4,0.4148148,0.4148148,0.45925927,0.43703705,,0.5407407,0.4888889,0.4962963,0.5555556,0.43703705,0.51111114,0.5037037,0.5777778,0.65925926,0.6148148,0.6,0.5777778,0.6296296,0.75555557,0.62222224,0.6296296,0.88148147,0.73333335,0.74814814,0.7037037,0.65925926,0.7925926,0.8,0.8074074,0.6888889,0.95555556,0.94074076,0.8074074,0.9111111,0.9111111,0.8666667,0.93333334,0.91851854,0.97037035,0.9111111,0.9851852,0.95555556,0.9925926,0.8666667,0.97037035,0.93333334,0.9777778,0.9925926,0.9851852,0.9851852,0.9925926,0.9925926,0.962963,0.9851852,0.9925926,0.9925926,0.9925926,0.9925926,0.9925926,0.9925926
+XGBoost (tuned + ensembled),0.08148148,0.02962963,0.23703703,0.08888889,0.28148147,0.31851852,0.34074074,0.36296296,0.37037036,0.31111112,0.37777779,0.34074074,0.31111112,0.42222223,0.45925927,,0.4888889,0.4814815,0.5037037,0.4814815,0.52592593,0.53333336,0.43703705,0.75555557,0.5407407,0.56296295,0.5777778,0.64444447,0.6666667,0.7259259,0.63703704,0.8,0.7407407,0.88148147,0.65185183,0.6814815,0.77037036,0.8666667,0.8074074,0.65185183,0.8296296,0.8666667,0.8962963,0.97037035,0.93333334,0.9851852,0.94814813,0.93333334,0.8666667,0.9037037,0.9777778,0.962963,0.97037035,0.9037037,0.91851854,0.962963,0.9111111,0.93333334,1.0,0.9925926,0.9925926,0.9925926,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0074074073,0.12592593,0.14074074,0.2074074,0.26666668,0.25925925,0.38518518,0.4074074,0.4,0.51111114,0.39259258,0.5037037,0.52592593,0.51111114,0.51111114,0.51111114,,0.51111114,0.39259258,0.4814815,0.46666667,0.5185185,0.65925926,0.5925926,0.8074074,0.6148148,0.5703704,0.56296295,0.6888889,0.6148148,0.6148148,0.6666667,0.58518517,0.7111111,0.7259259,0.6814815,0.9777778,0.7111111,0.77037036,0.7111111,0.78518516,0.9851852,0.8296296,0.93333334,0.82222223,0.82222223,0.8888889,0.82222223,0.88148147,0.8666667,0.8888889,0.9111111,0.9851852,0.91851854,0.94074076,0.93333334,0.8888889,0.93333334,0.9925926,0.962963,1.0,1.0,1.0,0.95555556,0.93333334,0.962963,1.0,1.0,0.9851852,1.0
+RealMLP (tuned),0.11111111,0.08888889,0.22962964,0.12592593,0.2074074,0.31851852,0.31111112,0.22962964,0.32592592,0.38518518,0.37037036,0.39259258,0.47407407,0.4148148,0.5037037,0.5185185,0.4888889,,0.4888889,0.45925927,0.53333336,0.39259258,0.6,0.6,0.52592593,0.62222224,0.45925927,0.6962963,0.65185183,0.56296295,0.6666667,0.7777778,0.8,0.7259259,0.6074074,0.6888889,0.8148148,0.74814814,0.8296296,0.837037,0.8740741,0.91851854,0.8,0.84444445,0.8888889,0.8148148,0.8148148,0.88148147,0.91851854,0.85925925,0.82222223,0.85925925,0.95555556,0.94814813,0.9777778,0.97037035,0.94814813,0.97037035,0.94074076,0.9777778,0.9851852,0.9851852,0.962963,0.9851852,1.0,0.9925926,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.014814815,0.21481481,0.08148148,0.35555556,0.22222222,0.17777778,0.19259259,0.4074074,0.36296296,0.44444445,0.2888889,0.47407407,0.5037037,0.45925927,0.44444445,0.4962963,0.6074074,0.51111114,,0.37777779,0.51111114,0.51111114,0.5925926,0.5555556,0.65925926,0.6148148,0.54814816,0.6814815,0.7111111,0.64444447,0.63703704,0.7407407,0.7185185,0.7259259,0.9111111,0.6962963,0.7259259,0.62222224,0.8,0.7259259,0.8666667,0.8074074,0.6888889,0.91851854,0.8296296,0.74814814,0.94814813,0.8666667,0.7777778,0.93333334,0.8888889,0.962963,0.97037035,0.91851854,0.8666667,0.9259259,0.93333334,0.8962963,0.9851852,0.93333334,1.0,1.0,0.9925926,0.8962963,0.94074076,0.8962963,1.0,1.0,0.94814813,1.0
+TabDPT-Turbo (default),0.037037037,0.27407408,0.08148148,0.37037036,0.28148147,0.21481481,0.37037036,0.4074074,0.45185184,0.53333336,0.42222223,0.5407407,0.4814815,0.5407407,0.56296295,0.5185185,0.5185185,0.5407407,0.62222224,,0.46666667,0.52592593,0.52592593,0.54814816,0.6,0.5703704,0.58518517,0.65925926,0.6666667,0.5777778,0.65925926,0.76296294,0.7407407,0.6666667,0.8074074,0.7925926,0.7111111,0.6074074,0.7777778,0.7777778,0.82222223,0.8,0.65185183,0.85925925,0.78518516,0.75555557,0.84444445,0.837037,0.76296294,0.78518516,0.7925926,0.88148147,0.8666667,0.9111111,0.8296296,0.962963,0.94074076,0.91851854,0.97037035,0.9037037,0.9259259,0.9259259,0.94814813,0.82222223,1.0,0.837037,1.0,1.0,0.91851854,1.0
+TabDPT (tuned + ensembled),0.12592593,0.26666668,0.26666668,0.37037036,0.25925925,0.22962964,0.34074074,0.42222223,0.36296296,0.46666667,0.36296296,0.46666667,0.5037037,0.45185184,0.4888889,0.47407407,0.53333336,0.46666667,0.4888889,0.53333336,,0.46666667,0.52592593,0.5185185,0.6148148,0.5555556,0.5037037,0.7037037,0.6296296,0.6,0.9111111,0.7185185,0.73333335,0.63703704,0.7111111,0.95555556,0.7259259,0.6296296,0.837037,0.7259259,0.8148148,0.75555557,0.6814815,0.8518519,0.7777778,0.7037037,0.78518516,0.77037036,0.7925926,0.8296296,0.75555557,0.85925925,0.8518519,0.9851852,0.8666667,0.9851852,0.94814813,0.8888889,0.9925926,0.8962963,0.9037037,0.9037037,0.9925926,0.837037,1.0,0.837037,1.0,1.0,0.9259259,1.0
+ChimeraBoost (tuned + ensembled),0.08148148,0.05185185,0.21481481,0.14814815,0.22962964,0.31851852,0.39259258,0.4814815,0.37777779,0.3037037,0.42962962,0.28148147,0.35555556,0.39259258,0.4962963,0.46666667,0.4814815,0.6074074,0.4888889,0.47407407,0.53333336,,0.4,0.54814816,0.54814816,0.6074074,0.7777778,0.7037037,0.65925926,0.9037037,0.62222224,0.67407405,0.7407407,0.7111111,0.64444447,0.62222224,0.7407407,0.8148148,0.7407407,0.6296296,0.7777778,0.8148148,0.837037,0.84444445,0.85925925,0.8888889,0.837037,0.85925925,0.837037,0.8518519,0.9037037,0.84444445,0.82222223,0.84444445,0.85925925,0.85925925,0.85925925,0.8518519,0.93333334,0.962963,0.93333334,0.93333334,0.9259259,1.0,0.88148147,1.0,0.94814813,1.0,1.0,1.0
+LightGBM (tuned),0.022222223,0.014814815,0.13333334,0.08888889,0.17037037,0.22222222,0.3037037,0.2962963,0.37037036,0.35555556,0.2962963,0.36296296,0.17777778,0.39259258,0.42222223,0.56296295,0.34074074,0.4,0.4074074,0.47407407,0.47407407,0.6,,0.65925926,0.4814815,0.52592593,0.65185183,0.6074074,0.6296296,0.6888889,0.58518517,0.65185183,0.62222224,0.8296296,0.62222224,0.64444447,0.8074074,0.78518516,0.78518516,0.6296296,0.73333335,0.84444445,0.82222223,0.9777778,0.85925925,0.9037037,0.8888889,0.8962963,0.837037,0.8148148,0.95555556,0.94814813,0.93333334,0.95555556,0.85925925,0.9777778,0.8666667,0.8740741,0.9925926,0.94074076,1.0,1.0,0.9925926,0.9111111,1.0,0.94074076,1.0,1.0,0.9777778,1.0
+XGBoost (tuned),0.05925926,0.0074074073,0.21481481,0.074074075,0.17777778,0.3037037,0.31851852,0.31111112,0.35555556,0.23703703,0.36296296,0.27407408,0.22222222,0.32592592,0.34074074,0.24444444,0.4074074,0.4,0.44444445,0.45185184,0.4814815,0.45185184,0.34074074,,0.4962963,0.5555556,0.5037037,0.5925926,0.5925926,0.67407405,0.6148148,0.7111111,0.67407405,0.7777778,0.6296296,0.65925926,0.73333335,0.8074074,0.78518516,0.63703704,0.8148148,0.84444445,0.8740741,0.9111111,0.9111111,0.94814813,0.9259259,0.9037037,0.837037,0.8962963,0.95555556,0.95555556,0.97037035,0.85925925,0.9111111,0.95555556,0.8666667,0.9037037,1.0,0.9925926,0.9925926,0.9925926,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.08888889,0.096296296,0.14814815,0.17777778,0.18518518,0.3037037,0.35555556,0.3037037,0.44444445,0.31111112,0.45925927,0.4074074,0.42962962,0.38518518,0.45925927,0.19259259,0.47407407,0.34074074,0.4,0.38518518,0.45185184,0.5185185,0.5037037,,0.5185185,0.51111114,0.4888889,0.65185183,0.52592593,0.5407407,0.5777778,0.5777778,0.6888889,0.64444447,0.58518517,0.8888889,0.63703704,0.6888889,0.7111111,0.73333335,0.93333334,0.7777778,0.8296296,0.78518516,0.7925926,0.8296296,0.7925926,0.8666667,0.8074074,0.837037,0.8888889,0.9851852,0.91851854,0.91851854,0.9259259,0.88148147,0.94074076,0.9777778,0.9259259,1.0,1.0,0.9851852,0.9259259,0.93333334,0.93333334,1.0,1.0,0.95555556,1.0
+ModernNCA (tuned),0.037037037,0.022222223,0.14814815,0.16296296,0.11851852,0.2,0.2074074,0.28148147,0.17037037,0.32962963,0.25925925,0.36296296,0.38518518,0.34814814,0.4,0.43703705,0.38518518,0.37777779,0.38518518,0.42962962,0.44444445,0.39259258,0.47407407,0.44444445,0.4814815,,0.44444445,0.5703704,0.64444447,0.52592593,0.6,0.7185185,0.63703704,0.58518517,0.7037037,0.7407407,0.7111111,0.62222224,0.8888889,0.73333335,0.8148148,0.78518516,0.7037037,0.78518516,0.78518516,0.75555557,0.8296296,0.8,0.7777778,0.82222223,0.84444445,0.837037,0.9037037,0.9037037,0.94814813,0.94814813,0.91851854,0.94074076,0.9777778,0.97037035,0.97037035,0.97037035,0.94074076,0.9111111,1.0,0.93333334,1.0,1.0,0.9925926,1.0
+ChimeraBoost (tuned),0.08148148,0.044444446,0.22222222,0.08888889,0.2,0.31851852,0.33333334,0.38518518,0.3037037,0.19259259,0.36296296,0.19259259,0.25925925,0.26666668,0.42222223,0.42222223,0.42962962,0.5407407,0.45185184,0.4148148,0.4962963,0.22222222,0.34814814,0.4962963,0.4888889,0.5555556,,0.6666667,0.6,0.75555557,0.6,0.6296296,0.7185185,0.65925926,0.62222224,0.58518517,0.7037037,0.73333335,0.7037037,0.62222224,0.74814814,0.8074074,0.78518516,0.78518516,0.84444445,0.837037,0.8074074,0.84444445,0.8074074,0.82222223,0.8666667,0.82222223,0.8148148,0.8296296,0.837037,0.84444445,0.8518519,0.84444445,0.9259259,0.94814813,0.9259259,0.9259259,0.91851854,1.0,0.8888889,1.0,0.94814813,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.074074075,0.037037037,0.2,0.08148148,0.19259259,0.26666668,0.23703703,0.14074074,0.24444444,0.28148147,0.25185186,0.27407408,0.35555556,0.27407408,0.37037036,0.35555556,0.43703705,0.3037037,0.31851852,0.34074074,0.2962963,0.2962963,0.39259258,0.4074074,0.51111114,0.42962962,0.33333334,,0.5037037,0.39259258,0.35555556,0.6,0.78518516,0.51111114,0.51111114,0.53333336,0.6962963,0.6296296,0.63703704,0.7037037,0.7185185,0.74814814,0.6888889,0.63703704,0.76296294,0.7185185,0.8074074,0.77037036,0.82222223,0.7259259,0.6962963,0.8074074,0.837037,0.837037,0.8666667,0.9037037,1.0,0.9111111,0.91851854,0.9777778,0.9037037,0.9037037,0.9851852,0.9851852,0.93333334,1.0,0.93333334,0.9777778,1.0,1.0
+iLTM (tuned),0.06666667,0.074074075,0.08888889,0.1037037,0.16296296,0.22962964,0.25185186,0.25185186,0.24444444,0.25185186,0.28148147,0.2962963,0.27407408,0.26666668,0.24444444,0.33333334,0.31111112,0.34814814,0.2888889,0.33333334,0.37037036,0.34074074,0.37037036,0.4074074,0.34814814,0.35555556,0.4,0.4962963,,0.4814815,0.52592593,0.56296295,0.5407407,0.4962963,0.54814816,0.5777778,0.6148148,0.6074074,0.6296296,0.7111111,0.6814815,0.7259259,0.65185183,0.78518516,0.82222223,0.6666667,0.75555557,0.837037,0.82222223,0.8296296,0.837037,0.84444445,0.9259259,0.77037036,0.9037037,0.82222223,0.94074076,0.93333334,0.9777778,0.9925926,0.962963,0.962963,0.94814813,0.93333334,0.94074076,0.95555556,0.9925926,1.0,1.0,1.0
+ChimeraBoost (default),0.06666667,0.022222223,0.22222222,0.074074075,0.21481481,0.3037037,0.25925925,0.35555556,0.25185186,0.14074074,0.2888889,0.17037037,0.25185186,0.2,0.37777779,0.27407408,0.38518518,0.43703705,0.35555556,0.42222223,0.4,0.096296296,0.31111112,0.32592592,0.47407407,0.47407407,0.24444444,0.6074074,0.5185185,,0.5925926,0.6148148,0.7037037,0.56296295,0.5925926,0.6148148,0.6074074,0.6666667,0.6666667,0.62222224,0.6962963,0.6962963,0.74814814,0.6666667,0.74814814,0.7777778,0.65925926,0.7407407,0.8074074,0.7777778,0.82222223,0.7185185,0.78518516,0.7925926,0.84444445,0.8296296,0.8518519,0.84444445,0.93333334,0.93333334,0.9259259,0.93333334,0.91851854,1.0,0.8666667,1.0,0.93333334,0.9777778,1.0,1.0
+TabDPT (tuned),0.074074075,0.22222222,0.17777778,0.22962964,0.17037037,0.17777778,0.2074074,0.28148147,0.23703703,0.36296296,0.25925925,0.37037036,0.37037036,0.36296296,0.37037036,0.36296296,0.38518518,0.33333334,0.36296296,0.34074074,0.08888889,0.37777779,0.4148148,0.38518518,0.45925927,0.4,0.4,0.64444447,0.47407407,0.4074074,,0.52592593,0.62222224,0.4888889,0.54814816,0.7740741,0.56296295,0.4814815,0.6148148,0.6814815,0.67407405,0.6074074,0.6,0.7037037,0.65925926,0.6074074,0.7111111,0.7259259,0.65185183,0.65925926,0.6296296,0.76296294,0.78518516,0.962963,0.78518516,0.97037035,0.93333334,0.8518519,0.93333334,0.8666667,0.7777778,0.78518516,0.95555556,0.7925926,1.0,0.8296296,0.94074076,0.9851852,0.8666667,1.0
+TorchMLP (tuned + ensembled),0.022222223,0.0074074073,0.11111111,0.05925926,0.1037037,0.19259259,0.19259259,0.16296296,0.17777778,0.2074074,0.22962964,0.21481481,0.24444444,0.2074074,0.11851852,0.2,0.33333334,0.22222222,0.25925925,0.23703703,0.28148147,0.32592592,0.34814814,0.2888889,0.42222223,0.28148147,0.37037036,0.4,0.43703705,0.38518518,0.47407407,,0.45185184,0.47407407,0.5185185,0.51111114,0.5703704,0.54814816,0.52592593,0.6,0.9111111,0.6296296,0.65925926,0.63703704,0.7259259,0.6888889,0.75555557,0.78518516,0.7407407,0.8148148,0.8148148,0.8,0.9925926,0.73333335,0.88148147,0.75555557,0.82222223,0.8888889,0.94074076,0.94074076,0.94814813,0.94814813,0.9037037,0.94074076,0.9259259,0.94074076,1.0,1.0,0.97037035,1.0
+xRFM (tuned),0.05925926,0.0074074073,0.13333334,0.02962963,0.13333334,0.25185186,0.19259259,0.08148148,0.15555556,0.2074074,0.2074074,0.21481481,0.28148147,0.2,0.26666668,0.25925925,0.4148148,0.2,0.28148147,0.25925925,0.26666668,0.25925925,0.37777779,0.32592592,0.42222223,0.36296296,0.28148147,0.21481481,0.45925927,0.2962963,0.37777779,0.54814816,,0.45185184,0.5185185,0.42962962,0.64444447,0.5407407,0.5777778,0.6148148,0.6888889,0.7185185,0.63703704,0.5925926,0.73333335,0.6148148,0.7037037,0.77037036,0.7777778,0.8,0.65185183,0.78518516,0.82222223,0.8148148,0.88148147,0.84444445,0.962963,0.9111111,0.8666667,0.9851852,0.8888889,0.8962963,0.94814813,0.97037035,0.9259259,0.9925926,0.9925926,1.0,1.0,1.0
+XGBoost (default),0.014814815,0.022222223,0.12592593,0.05185185,0.05925926,0.15555556,0.17037037,0.21481481,0.23703703,0.12592593,0.21481481,0.13333334,0.11851852,0.2074074,0.25185186,0.11851852,0.2888889,0.27407408,0.27407408,0.33333334,0.36296296,0.2888889,0.17037037,0.22222222,0.31111112,0.4148148,0.34074074,0.4888889,0.5037037,0.43703705,0.51111114,0.52592593,0.54814816,,0.52592593,0.5185185,0.47407407,0.58518517,0.6148148,0.58518517,0.6962963,0.56296295,0.65925926,0.51111114,0.76296294,0.76296294,0.7259259,0.78518516,0.7925926,0.76296294,0.8666667,0.75555557,0.88148147,0.74814814,0.8074074,0.88148147,0.7777778,0.84444445,0.94814813,0.9259259,0.962963,0.97037035,0.8740741,0.91851854,1.0,0.9259259,1.0,1.0,0.93333334,1.0
+TabSwift (default),0.02962963,0.037037037,0.022222223,0.15555556,0.08888889,0.05925926,0.096296296,0.24444444,0.21481481,0.25185186,0.11111111,0.25925925,0.27407408,0.22962964,0.2962963,0.34814814,0.27407408,0.39259258,0.08888889,0.19259259,0.2888889,0.35555556,0.37777779,0.37037036,0.35555556,0.2962963,0.37777779,0.4888889,0.45185184,0.4074074,0.45185184,0.4814815,0.4814815,0.47407407,,0.62222224,0.4888889,0.54814816,0.56296295,0.5925926,0.5777778,0.58518517,0.58518517,0.5925926,0.5555556,0.6,0.62222224,0.5777778,0.74814814,0.6666667,0.6962963,0.6666667,0.7185185,0.7925926,0.78518516,0.8296296,0.8,0.78518516,0.837037,0.8518519,0.8666667,0.8666667,0.88148147,0.8148148,0.93333334,0.837037,0.94814813,0.9777778,0.8740741,1.0
+TabDPT (default),0.05925926,0.14814815,0.12592593,0.17037037,0.17037037,0.17777778,0.2,0.25185186,0.18518518,0.31111112,0.24444444,0.31851852,0.25185186,0.3037037,0.34074074,0.31851852,0.31851852,0.31111112,0.3037037,0.2074074,0.044444446,0.37777779,0.35555556,0.34074074,0.4148148,0.25925925,0.4148148,0.46666667,0.42222223,0.38518518,0.22592592,0.4888889,0.5703704,0.4814815,0.37777779,,0.5555556,0.4814815,0.5407407,0.5703704,0.58518517,0.5555556,0.53333336,0.6,0.6296296,0.5407407,0.62222224,0.65925926,0.58518517,0.5555556,0.6148148,0.64444447,0.74814814,0.77037036,0.67407405,0.8666667,0.8148148,0.75555557,0.78518516,0.7407407,0.7777778,0.7777778,0.7777778,0.8,0.9851852,0.8074074,0.8074074,0.82222223,0.8296296,1.0
+TabM (default),0.0,0.0,0.0074074073,0.05185185,0.05185185,0.037037037,0.17037037,0.06666667,0.11851852,0.17777778,0.21481481,0.15555556,0.14814815,0.22962964,0.2074074,0.22962964,0.022222223,0.18518518,0.27407408,0.2888889,0.27407408,0.25925925,0.19259259,0.26666668,0.11111111,0.2888889,0.2962963,0.3037037,0.38518518,0.39259258,0.43703705,0.42962962,0.35555556,0.52592593,0.51111114,0.44444445,,0.53333336,0.56296295,0.5703704,0.5407407,0.6814815,0.62222224,0.6148148,0.6888889,0.65185183,0.63703704,0.6962963,0.77037036,0.6814815,0.67407405,0.65925926,0.7259259,0.85925925,0.8518519,0.88148147,0.7407407,0.8740741,0.8666667,0.8962963,0.93333334,0.93333334,0.91851854,0.8666667,0.93333334,0.85925925,1.0,1.0,0.93333334,1.0
+EBM (tuned + ensembled),0.0074074073,0.0074074073,0.15555556,0.02962963,0.08148148,0.22962964,0.17777778,0.17777778,0.17777778,0.1037037,0.22222222,0.11111111,0.15555556,0.12592593,0.2,0.13333334,0.2888889,0.25185186,0.37777779,0.39259258,0.37037036,0.18518518,0.21481481,0.19259259,0.36296296,0.37777779,0.26666668,0.37037036,0.39259258,0.33333334,0.5185185,0.45185184,0.45925927,0.4148148,0.45185184,0.5185185,0.46666667,,0.54814816,0.46666667,0.64444447,0.54814816,0.8296296,0.5185185,0.6148148,0.94814813,0.6666667,0.64444447,0.64444447,0.6666667,0.64444447,0.6962963,0.8,0.65185183,0.7037037,0.64444447,0.67407405,0.78518516,0.7185185,0.837037,0.9777778,0.94074076,0.7259259,0.93333334,0.75555557,0.94814813,1.0,0.9259259,0.962963,0.93333334
+ModernNCA (default),0.0,0.014814815,0.014814815,0.037037037,0.014814815,0.096296296,0.11851852,0.096296296,0.02962963,0.17777778,0.14074074,0.19259259,0.15555556,0.15555556,0.19259259,0.19259259,0.22962964,0.17037037,0.2,0.22222222,0.16296296,0.25925925,0.21481481,0.21481481,0.31111112,0.11111111,0.2962963,0.36296296,0.37037036,0.33333334,0.38518518,0.47407407,0.42222223,0.38518518,0.43703705,0.45925927,0.43703705,0.45185184,,0.5555556,0.5555556,0.52592593,0.4814815,0.5925926,0.64444447,0.5037037,0.5777778,0.7111111,0.58518517,0.67407405,0.62222224,0.63703704,0.837037,0.77037036,0.6962963,0.84444445,0.8666667,0.78518516,0.95555556,0.8888889,0.8296296,0.837037,0.8888889,0.8,0.94814813,0.8296296,1.0,1.0,0.9111111,1.0
+LimiX (default),0.096296296,0.14074074,0.13333334,0.14814815,0.14814815,0.12592593,0.22222222,0.11111111,0.11851852,0.31851852,0.22962964,0.31851852,0.34074074,0.2962963,0.31111112,0.34814814,0.2888889,0.16296296,0.27407408,0.22222222,0.27407408,0.37037036,0.37037036,0.36296296,0.2888889,0.26666668,0.37777779,0.2962963,0.2888889,0.37777779,0.31851852,0.4,0.38518518,0.4148148,0.4074074,0.42962962,0.42962962,0.53333336,0.44444445,,0.53333336,0.4888889,0.5407407,0.4074074,0.51111114,0.5407407,0.4814815,0.53333336,0.62222224,0.6,0.52592593,0.5037037,0.5925926,0.56296295,0.7111111,0.6074074,0.6666667,0.7407407,0.77037036,0.7111111,0.6814815,0.6888889,0.6962963,0.76296294,0.7925926,0.77037036,0.74814814,0.78518516,0.77037036,0.9259259
+TorchMLP (tuned),0.0,0.014814815,0.022222223,0.014814815,0.02962963,0.11851852,0.11851852,0.11111111,0.1037037,0.096296296,0.12592593,0.1037037,0.12592593,0.096296296,0.044444446,0.17037037,0.21481481,0.12592593,0.13333334,0.17777778,0.18518518,0.22222222,0.26666668,0.18518518,0.26666668,0.18518518,0.25185186,0.28148147,0.31851852,0.3037037,0.32592592,0.08888889,0.31111112,0.3037037,0.42222223,0.4148148,0.45925927,0.35555556,0.44444445,0.46666667,,0.43703705,0.47407407,0.5185185,0.5703704,0.5185185,0.6148148,0.6296296,0.63703704,0.7037037,0.7259259,0.6962963,0.9259259,0.7037037,0.74814814,0.73333335,0.77037036,0.7777778,0.8518519,0.8962963,0.8962963,0.8962963,0.8148148,0.91851854,0.9037037,0.91851854,0.9851852,1.0,0.94814813,1.0
+RealMLP (default),0.0,0.0,0.037037037,0.022222223,0.0,0.05925926,0.11851852,0.0,0.074074075,0.11111111,0.15555556,0.1037037,0.11851852,0.19259259,0.05925926,0.13333334,0.014814815,0.08148148,0.19259259,0.2,0.24444444,0.18518518,0.15555556,0.15555556,0.06666667,0.21481481,0.19259259,0.25185186,0.27407408,0.3037037,0.39259258,0.37037036,0.28148147,0.43703705,0.4148148,0.44444445,0.31851852,0.45185184,0.47407407,0.51111114,0.56296295,,0.53333336,0.4814815,0.6148148,0.5555556,0.62222224,0.67407405,0.6666667,0.6296296,0.67407405,0.65185183,0.7259259,0.74814814,0.88148147,0.7925926,0.6888889,0.8962963,0.82222223,0.85925925,0.84444445,0.84444445,0.91851854,0.8888889,0.93333334,0.91851854,0.9777778,1.0,0.94074076,1.0
+EBM (tuned),0.0,0.0,0.12592593,0.02962963,0.022222223,0.13333334,0.11851852,0.11851852,0.16296296,0.11851852,0.17777778,0.11851852,0.08888889,0.12592593,0.19259259,0.1037037,0.17037037,0.2,0.31111112,0.34814814,0.31851852,0.16296296,0.17777778,0.12592593,0.22222222,0.2962963,0.21481481,0.31111112,0.34814814,0.25185186,0.4,0.34074074,0.36296296,0.34074074,0.4148148,0.46666667,0.37777779,0.17037037,0.5185185,0.45925927,0.52592593,0.46666667,,0.45925927,0.53333336,0.8074074,0.5925926,0.5555556,0.5703704,0.6148148,0.5555556,0.63703704,0.7407407,0.6296296,0.6888889,0.63703704,0.65925926,0.74814814,0.67407405,0.7925926,0.94814813,0.91851854,0.6888889,0.9037037,0.73333335,0.9259259,1.0,0.91851854,0.94814813,0.93333334
+LightGBM (default),0.0,0.0,0.0074074073,0.02962963,0.037037037,0.05185185,0.05185185,0.044444446,0.06666667,0.037037037,0.074074075,0.05925926,0.014814815,0.12592593,0.08888889,0.02962963,0.06666667,0.15555556,0.08148148,0.14074074,0.14814815,0.15555556,0.022222223,0.08888889,0.17037037,0.21481481,0.21481481,0.36296296,0.21481481,0.33333334,0.2962963,0.36296296,0.4074074,0.4888889,0.4074074,0.4,0.38518518,0.4814815,0.4074074,0.5925926,0.4814815,0.5185185,0.5407407,,0.5925926,0.56296295,0.5703704,0.6,0.65185183,0.5777778,0.67407405,0.63703704,0.76296294,0.73333335,0.73333335,0.8148148,0.7925926,0.8,0.9259259,0.84444445,0.8962963,0.9037037,0.94074076,0.7925926,0.93333334,0.8,1.0,1.0,0.8740741,1.0
+RandomForest (tuned + ensembled),0.014814815,0.0,0.037037037,0.0074074073,0.02962963,0.08148148,0.05925926,0.08148148,0.14074074,0.074074075,0.096296296,0.08148148,0.06666667,0.14814815,0.08888889,0.06666667,0.17777778,0.11111111,0.17037037,0.21481481,0.22222222,0.14074074,0.14074074,0.08888889,0.21481481,0.21481481,0.15555556,0.23703703,0.17777778,0.25185186,0.34074074,0.27407408,0.26666668,0.23703703,0.44444445,0.37037036,0.31111112,0.38518518,0.35555556,0.4888889,0.42962962,0.38518518,0.46666667,0.4074074,,0.5185185,0.5555556,0.6962963,0.5925926,0.6666667,0.63703704,0.63703704,0.7111111,0.65185183,0.73333335,0.6814815,0.6814815,0.8,0.97037035,0.9259259,0.8666667,0.8740741,0.85925925,0.8666667,0.84444445,0.9037037,0.95555556,1.0,0.97037035,1.0
+EBM (default),0.0,0.0,0.08148148,0.02962963,0.022222223,0.11111111,0.08888889,0.1037037,0.11851852,0.037037037,0.11111111,0.022222223,0.022222223,0.074074075,0.13333334,0.014814815,0.17777778,0.18518518,0.25185186,0.24444444,0.2962963,0.11111111,0.096296296,0.05185185,0.2074074,0.24444444,0.16296296,0.28148147,0.33333334,0.22222222,0.39259258,0.31111112,0.38518518,0.23703703,0.4,0.45925927,0.34814814,0.05185185,0.4962963,0.45925927,0.4814815,0.44444445,0.19259259,0.43703705,0.4814815,,0.58518517,0.5185185,0.53333336,0.5185185,0.56296295,0.63703704,0.73333335,0.6148148,0.6888889,0.63703704,0.62222224,0.7111111,0.65925926,0.78518516,0.85925925,0.84444445,0.6814815,0.8740741,0.7259259,0.8962963,1.0,0.88148147,0.94074076,0.93333334
+ExtraTrees (tuned + ensembled),0.0074074073,0.0,0.022222223,0.0074074073,0.014814815,0.0074074073,0.022222223,0.06666667,0.06666667,0.02962963,0.05925926,0.05185185,0.014814815,0.11851852,0.06666667,0.05185185,0.11111111,0.18518518,0.05185185,0.15555556,0.21481481,0.16296296,0.11111111,0.074074075,0.17037037,0.17037037,0.19259259,0.19259259,0.24444444,0.34074074,0.2888889,0.24444444,0.2962963,0.27407408,0.37777779,0.37777779,0.36296296,0.33333334,0.42222223,0.5185185,0.38518518,0.37777779,0.4074074,0.42962962,0.44444445,0.4148148,,0.46666667,0.5407407,0.5925926,0.6074074,0.8074074,0.5703704,0.63703704,0.62222224,0.6888889,0.6814815,0.76296294,0.9259259,0.8296296,0.8148148,0.8148148,0.9851852,0.8148148,0.8666667,0.85925925,0.9037037,0.93333334,0.93333334,1.0
+RandomForest (tuned),0.014814815,0.0074074073,0.022222223,0.014814815,0.0074074073,0.074074075,0.02962963,0.08888889,0.14814815,0.08148148,0.044444446,0.08148148,0.05925926,0.14814815,0.08148148,0.06666667,0.17777778,0.11851852,0.13333334,0.16296296,0.22962964,0.14074074,0.1037037,0.096296296,0.2074074,0.2,0.15555556,0.22962964,0.16296296,0.25925925,0.27407408,0.21481481,0.22962964,0.21481481,0.42222223,0.34074074,0.3037037,0.35555556,0.2888889,0.46666667,0.37037036,0.32592592,0.44444445,0.4,0.3037037,0.4814815,0.53333336,,0.5555556,0.64444447,0.6296296,0.6074074,0.62222224,0.62222224,0.6888889,0.63703704,0.65185183,0.77037036,0.86296296,0.9037037,0.85925925,0.8666667,0.82222223,0.8074074,0.837037,0.8518519,0.95555556,1.0,0.94074076,1.0
+FastaiMLP (tuned + ensembled),0.05925926,0.0,0.06666667,0.022222223,0.08888889,0.1037037,0.17777778,0.014814815,0.08888889,0.08888889,0.2,0.11851852,0.15555556,0.11111111,0.02962963,0.13333334,0.11851852,0.08148148,0.22222222,0.23703703,0.2074074,0.16296296,0.16296296,0.16296296,0.13333334,0.22222222,0.19259259,0.17777778,0.17777778,0.19259259,0.34814814,0.25925925,0.22222222,0.2074074,0.25185186,0.4148148,0.22962964,0.35555556,0.4148148,0.37777779,0.36296296,0.33333334,0.42962962,0.34814814,0.4074074,0.46666667,0.45925927,0.44444445,,0.46666667,0.45185184,0.4814815,0.53333336,0.5703704,0.8296296,0.58518517,0.6666667,0.9777778,0.6,0.7185185,0.7185185,0.7185185,0.75555557,0.9259259,0.8074074,0.93333334,0.84444445,0.7925926,0.9851852,1.0
+SAP-RPT-OSS (default),0.014814815,0.02962963,0.05185185,0.05925926,0.02962963,0.05185185,0.096296296,0.074074075,0.096296296,0.08148148,0.11111111,0.08888889,0.096296296,0.15555556,0.08888889,0.096296296,0.13333334,0.14074074,0.06666667,0.21481481,0.17037037,0.14814815,0.18518518,0.1037037,0.19259259,0.17777778,0.17777778,0.27407408,0.17037037,0.22222222,0.34074074,0.18518518,0.2,0.23703703,0.33333334,0.44444445,0.31851852,0.33333334,0.32592592,0.4,0.2962963,0.37037036,0.38518518,0.42222223,0.33333334,0.4814815,0.4074074,0.35555556,0.53333336,,0.4074074,0.5185185,0.6,0.6814815,0.6962963,0.7037037,0.62222224,0.7777778,0.6888889,0.85925925,0.7111111,0.7259259,0.77037036,0.8740741,0.8,0.8962963,0.97037035,1.0,0.9851852,1.0
+iLTM (default),0.0,0.0,0.0074074073,0.022222223,0.037037037,0.05185185,0.096296296,0.12592593,0.1037037,0.05185185,0.11851852,0.044444446,0.014814815,0.037037037,0.014814815,0.022222223,0.11111111,0.17777778,0.11111111,0.2074074,0.24444444,0.096296296,0.044444446,0.044444446,0.16296296,0.15555556,0.13333334,0.3037037,0.16296296,0.17777778,0.37037036,0.18518518,0.34814814,0.13333334,0.3037037,0.38518518,0.32592592,0.35555556,0.37777779,0.47407407,0.27407408,0.32592592,0.44444445,0.32592592,0.36296296,0.43703705,0.39259258,0.37037036,0.54814816,0.5925926,,0.4962963,0.5703704,0.5037037,0.67407405,0.4962963,0.6,0.78518516,0.67407405,0.8,0.8740741,0.88148147,0.63703704,0.837037,0.62222224,0.8666667,0.9925926,0.8666667,0.9259259,0.95555556
+ExtraTrees (tuned),0.0074074073,0.0,0.022222223,0.0,0.0074074073,0.0074074073,0.022222223,0.08148148,0.074074075,0.02962963,0.044444446,0.037037037,0.014814815,0.1037037,0.044444446,0.037037037,0.08888889,0.14074074,0.037037037,0.11851852,0.14074074,0.15555556,0.05185185,0.044444446,0.11111111,0.16296296,0.17777778,0.19259259,0.15555556,0.28148147,0.23703703,0.2,0.21481481,0.24444444,0.33333334,0.35555556,0.34074074,0.3037037,0.36296296,0.4962963,0.3037037,0.34814814,0.36296296,0.36296296,0.36296296,0.36296296,0.19259259,0.39259258,0.5185185,0.4814815,0.5037037,,0.51111114,0.6074074,0.5925926,0.6666667,0.65925926,0.6962963,0.9037037,0.7925926,0.8074074,0.8,0.962963,0.8074074,0.85925925,0.8296296,0.8962963,0.94074076,0.9259259,1.0
+TorchMLP (default),0.0,0.0,0.022222223,0.0074074073,0.0,0.014814815,0.02962963,0.0074074073,0.074074075,0.037037037,0.05925926,0.044444446,0.044444446,0.08148148,0.0074074073,0.02962963,0.014814815,0.044444446,0.02962963,0.13333334,0.14814815,0.17777778,0.06666667,0.02962963,0.014814815,0.096296296,0.18518518,0.16296296,0.074074075,0.21481481,0.21481481,0.0074074073,0.17777778,0.11851852,0.28148147,0.25185186,0.27407408,0.2,0.16296296,0.4074074,0.074074075,0.27407408,0.25925925,0.23703703,0.2888889,0.26666668,0.42962962,0.37777779,0.46666667,0.4,0.42962962,0.4888889,,0.6,0.6,0.6148148,0.6074074,0.63703704,0.6888889,0.837037,0.77037036,0.76296294,0.7777778,0.82222223,0.77037036,0.84444445,0.91851854,0.9925926,0.91851854,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0074074073,0.06666667,0.014814815,0.0074074073,0.06666667,0.022222223,0.014814815,0.0074074073,0.022222223,0.022222223,0.022222223,0.014814815,0.044444446,0.13333334,0.096296296,0.08148148,0.05185185,0.08148148,0.08888889,0.014814815,0.15555556,0.044444446,0.14074074,0.08148148,0.096296296,0.17037037,0.16296296,0.22962964,0.2074074,0.037037037,0.26666668,0.18518518,0.25185186,0.2074074,0.22962964,0.14074074,0.34814814,0.22962964,0.43703705,0.2962963,0.25185186,0.37037036,0.26666668,0.34814814,0.38518518,0.36296296,0.37777779,0.42962962,0.31851852,0.4962963,0.39259258,0.4,,0.46666667,0.8962963,0.64444447,0.53333336,0.67407405,0.6296296,0.51111114,0.51111114,0.7111111,0.7037037,0.94814813,0.7111111,0.5777778,0.77037036,0.7111111,0.97037035
+FastaiMLP (tuned),0.014814815,0.0,0.044444446,0.014814815,0.037037037,0.044444446,0.08148148,0.014814815,0.044444446,0.022222223,0.08148148,0.02962963,0.05925926,0.014814815,0.02962963,0.08148148,0.05925926,0.022222223,0.13333334,0.17037037,0.13333334,0.14074074,0.14074074,0.08888889,0.08148148,0.05185185,0.16296296,0.13333334,0.096296296,0.15555556,0.21481481,0.11851852,0.11851852,0.19259259,0.21481481,0.32592592,0.14814815,0.2962963,0.3037037,0.2888889,0.25185186,0.11851852,0.31111112,0.26666668,0.26666668,0.31111112,0.37777779,0.31111112,0.17037037,0.3037037,0.32592592,0.4074074,0.4,0.53333336,,0.56296295,0.5925926,0.7777778,0.53333336,0.6074074,0.63703704,0.63703704,0.6666667,0.7777778,0.73333335,0.82222223,0.8518519,0.78518516,0.93333334,1.0
+PerpetualBooster (tuned),0.0,0.0,0.014814815,0.022222223,0.0074074073,0.06666667,0.014814815,0.0074074073,0.0074074073,0.0074074073,0.037037037,0.0074074073,0.014814815,0.02962963,0.06666667,0.037037037,0.06666667,0.02962963,0.074074075,0.037037037,0.014814815,0.14074074,0.022222223,0.044444446,0.074074075,0.05185185,0.15555556,0.096296296,0.17777778,0.17037037,0.02962963,0.24444444,0.15555556,0.11851852,0.17037037,0.13333334,0.11851852,0.35555556,0.15555556,0.39259258,0.26666668,0.2074074,0.36296296,0.18518518,0.31851852,0.36296296,0.31111112,0.36296296,0.4148148,0.2962963,0.5037037,0.33333334,0.38518518,0.1037037,0.43703705,,0.58518517,0.5185185,0.65185183,0.6,0.51111114,0.51111114,0.6148148,0.6814815,0.9259259,0.6962963,0.5407407,0.7111111,0.6888889,0.97037035
+xRFM (default),0.02962963,0.0,0.05185185,0.0,0.05185185,0.06666667,0.05185185,0.0,0.0,0.074074075,0.06666667,0.06666667,0.096296296,0.074074075,0.022222223,0.08888889,0.11111111,0.05185185,0.06666667,0.05925926,0.05185185,0.14074074,0.13333334,0.13333334,0.11851852,0.08148148,0.14814815,0.0,0.05925926,0.14814815,0.06666667,0.17777778,0.037037037,0.22222222,0.2,0.18518518,0.25925925,0.32592592,0.13333334,0.33333334,0.22962964,0.31111112,0.34074074,0.2074074,0.31851852,0.37777779,0.31851852,0.34814814,0.33333334,0.37777779,0.4,0.34074074,0.39259258,0.35555556,0.4074074,0.4148148,,0.45925927,0.44444445,0.52592593,0.53333336,0.53333336,0.4888889,0.4814815,0.6148148,0.5185185,0.5777778,0.65925926,0.52592593,0.9111111
+FastaiMLP (default),0.014814815,0.0,0.044444446,0.0074074073,0.05185185,0.05185185,0.06666667,0.022222223,0.022222223,0.014814815,0.074074075,0.037037037,0.08888889,0.02962963,0.0074074073,0.06666667,0.06666667,0.02962963,0.1037037,0.08148148,0.11111111,0.14814815,0.12592593,0.096296296,0.05925926,0.05925926,0.15555556,0.08888889,0.06666667,0.15555556,0.14814815,0.11111111,0.08888889,0.15555556,0.21481481,0.24444444,0.12592593,0.21481481,0.21481481,0.25925925,0.22222222,0.1037037,0.25185186,0.2,0.2,0.2888889,0.23703703,0.22962964,0.022222223,0.22222222,0.21481481,0.3037037,0.36296296,0.46666667,0.22222222,0.4814815,0.5407407,,0.4148148,0.53333336,0.5703704,0.5777778,0.43703705,0.7037037,0.6148148,0.75555557,0.73333335,0.67407405,0.78518516,0.8666667
+RandomForest (default),0.0074074073,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.0074074073,0.0074074073,0.0074074073,0.0,0.0074074073,0.014814815,0.0,0.0074074073,0.05925926,0.014814815,0.02962963,0.0074074073,0.06666667,0.0074074073,0.0,0.022222223,0.022222223,0.074074075,0.08148148,0.022222223,0.06666667,0.06666667,0.05925926,0.13333334,0.05185185,0.16296296,0.21481481,0.13333334,0.28148147,0.044444446,0.22962964,0.14814815,0.17777778,0.32592592,0.074074075,0.02962963,0.34074074,0.074074075,0.13703704,0.4,0.31111112,0.32592592,0.096296296,0.31111112,0.32592592,0.46666667,0.34814814,0.5555556,0.58518517,,0.5185185,0.5925926,0.5925926,0.58518517,0.6296296,0.76296294,0.65925926,0.67407405,0.77037036,0.67407405,1.0
+KNN (tuned + ensembled),0.02962963,0.0074074073,0.044444446,0.0074074073,0.037037037,0.02962963,0.05925926,0.0074074073,0.0074074073,0.0074074073,0.074074075,0.0074074073,0.014814815,0.0074074073,0.014814815,0.0074074073,0.037037037,0.022222223,0.06666667,0.096296296,0.1037037,0.037037037,0.05925926,0.0074074073,0.074074075,0.02962963,0.05185185,0.022222223,0.0074074073,0.06666667,0.13333334,0.05925926,0.014814815,0.074074075,0.14814815,0.25925925,0.1037037,0.16296296,0.11111111,0.2888889,0.1037037,0.14074074,0.2074074,0.15555556,0.074074075,0.21481481,0.17037037,0.096296296,0.28148147,0.14074074,0.2,0.2074074,0.16296296,0.37037036,0.39259258,0.4,0.47407407,0.46666667,0.4814815,,0.39259258,0.39259258,0.46666667,0.7185185,0.6666667,0.75555557,0.5925926,1.0,0.8148148,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.014814815,0.0,0.0,0.05185185,0.0,0.0074074073,0.0074074073,0.0074074073,0.0,0.0074074073,0.0074074073,0.0,0.014814815,0.0,0.074074075,0.096296296,0.06666667,0.0,0.0074074073,0.0,0.02962963,0.074074075,0.096296296,0.037037037,0.074074075,0.22222222,0.05185185,0.11111111,0.037037037,0.13333334,0.22222222,0.06666667,0.022222223,0.17037037,0.31851852,0.1037037,0.15555556,0.05185185,0.1037037,0.13333334,0.14074074,0.18518518,0.14074074,0.28148147,0.2888889,0.12592593,0.19259259,0.22962964,0.4888889,0.36296296,0.4888889,0.46666667,0.42962962,0.4074074,0.6074074,,0.5185185,0.5555556,0.7111111,0.53333336,0.7185185,1.0,0.78518516,0.8296296,0.8
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.014814815,0.0,0.0,0.05185185,0.0,0.0,0.0074074073,0.0074074073,0.0,0.0074074073,0.0074074073,0.0,0.014814815,0.0,0.074074075,0.096296296,0.06666667,0.0,0.0074074073,0.0,0.02962963,0.074074075,0.096296296,0.037037037,0.06666667,0.21481481,0.05185185,0.1037037,0.02962963,0.13333334,0.22222222,0.06666667,0.05925926,0.16296296,0.31111112,0.1037037,0.15555556,0.08148148,0.096296296,0.12592593,0.15555556,0.18518518,0.13333334,0.28148147,0.27407408,0.11851852,0.2,0.23703703,0.4888889,0.36296296,0.4888889,0.46666667,0.42222223,0.4074074,0.6074074,0.4814815,,0.5555556,0.7111111,0.53333336,0.7185185,1.0,0.78518516,0.8296296,0.8
+ExtraTrees (default),0.0074074073,0.0,0.0,0.0074074073,0.0,0.0,0.0074074073,0.0,0.014814815,0.0074074073,0.014814815,0.014814815,0.0074074073,0.022222223,0.037037037,0.0074074073,0.0,0.037037037,0.0074074073,0.05185185,0.0074074073,0.074074075,0.0074074073,0.0074074073,0.014814815,0.05925926,0.08148148,0.014814815,0.05185185,0.08148148,0.044444446,0.096296296,0.05185185,0.12592593,0.11851852,0.22222222,0.08148148,0.27407408,0.11111111,0.3037037,0.18518518,0.08148148,0.31111112,0.05925926,0.14074074,0.31851852,0.014814815,0.17777778,0.24444444,0.22962964,0.36296296,0.037037037,0.22222222,0.2888889,0.33333334,0.38518518,0.51111114,0.56296295,0.4148148,0.53333336,0.44444445,0.44444445,,0.63703704,0.7259259,0.65925926,0.5703704,0.7037037,0.6814815,1.0
+Linear (tuned + ensembled),0.014814815,0.0,0.05185185,0.0,0.022222223,0.05185185,0.1037037,0.0,0.06666667,0.0074074073,0.11851852,0.0074074073,0.014814815,0.0074074073,0.014814815,0.0,0.044444446,0.014814815,0.1037037,0.17777778,0.16296296,0.0,0.08888889,0.0,0.074074075,0.08888889,0.0,0.014814815,0.06666667,0.0,0.2074074,0.05925926,0.02962963,0.08148148,0.18518518,0.2,0.13333334,0.06666667,0.2,0.23703703,0.08148148,0.11111111,0.096296296,0.2074074,0.13333334,0.12592593,0.18518518,0.19259259,0.074074075,0.12592593,0.16296296,0.19259259,0.17777778,0.2962963,0.22222222,0.31851852,0.5185185,0.2962963,0.37037036,0.28148147,0.2888889,0.2888889,0.36296296,,0.51111114,0.837037,0.33333334,0.4148148,0.95555556,0.6666667
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.06666667,0.0,0.05925926,0.0,0.0,0.11851852,0.0,0.0,0.06666667,0.0,0.11111111,0.06666667,0.05925926,0.13333334,0.0,0.074074075,0.074074075,0.0,0.06666667,0.014814815,0.06666667,0.24444444,0.05185185,0.2074074,0.096296296,0.06666667,0.26666668,0.06666667,0.15555556,0.27407408,0.13333334,0.16296296,0.19259259,0.2,0.37777779,0.14074074,0.22962964,0.05185185,0.26666668,0.074074075,0.38518518,0.38518518,0.23703703,0.33333334,0.46666667,0.46666667,0.27407408,0.4888889,,0.5037037,0.46666667,0.45185184,0.5037037,0.91851854
+Linear (tuned),0.014814815,0.0,0.044444446,0.0,0.022222223,0.037037037,0.096296296,0.0,0.05925926,0.0,0.08888889,0.0,0.0074074073,0.0,0.0074074073,0.0,0.037037037,0.0074074073,0.1037037,0.16296296,0.16296296,0.0,0.05925926,0.0,0.06666667,0.06666667,0.0,0.0,0.044444446,0.0,0.17037037,0.05925926,0.0074074073,0.074074075,0.16296296,0.19259259,0.14074074,0.05185185,0.17037037,0.22962964,0.08148148,0.08148148,0.074074075,0.2,0.096296296,0.1037037,0.14074074,0.14814815,0.06666667,0.1037037,0.13333334,0.17037037,0.15555556,0.2888889,0.17777778,0.3037037,0.4814815,0.24444444,0.34074074,0.24444444,0.28148147,0.28148147,0.34074074,0.16296296,0.4962963,,0.33333334,0.4148148,0.88148147,0.6666667
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.0,0.0,0.0,0.0,0.0,0.05185185,0.0,0.0,0.0,0.0,0.05185185,0.06666667,0.0074074073,0.06666667,0.05925926,0.0,0.0074074073,0.0,0.05185185,0.19259259,0.0,0.0,0.0,0.25185186,0.014814815,0.022222223,0.0,0.0,0.044444446,0.0,0.096296296,0.044444446,0.15555556,0.02962963,0.0074074073,0.1037037,0.08148148,0.42222223,0.14814815,0.45925927,0.42222223,0.26666668,0.32592592,0.4074074,0.0,0.0,0.42962962,0.6666667,0.53333336,0.6666667,,0.6666667,0.6814815,0.78518516
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.022222223,0.014814815,0.0,0.0,0.0,0.022222223,0.17777778,0.0,0.074074075,0.0,0.21481481,0.0,0.0,0.08148148,0.0,0.0,0.11851852,0.06666667,0.0,0.2074074,0.0,0.13333334,0.05925926,0.0074074073,0.22962964,0.21481481,0.2888889,0.34074074,0.32592592,0.22962964,0.0,0.21481481,0.21481481,0.2962963,0.58518517,0.54814816,0.58518517,0.33333334,,0.58518517,0.9925926
+Linear (default),0.0,0.0,0.02962963,0.0,0.022222223,0.022222223,0.02962963,0.0,0.014814815,0.0,0.044444446,0.0,0.0,0.0,0.0074074073,0.0,0.014814815,0.0,0.05185185,0.08148148,0.074074075,0.0,0.022222223,0.0,0.044444446,0.0074074073,0.0,0.0,0.0,0.0,0.13333334,0.02962963,0.0,0.06666667,0.12592593,0.17037037,0.06666667,0.037037037,0.08888889,0.22962964,0.05185185,0.05925926,0.05185185,0.12592593,0.02962963,0.05925926,0.06666667,0.05925926,0.014814815,0.014814815,0.074074075,0.074074075,0.08148148,0.2888889,0.06666667,0.31111112,0.47407407,0.21481481,0.32592592,0.18518518,0.17037037,0.17037037,0.31851852,0.044444446,0.4962963,0.11851852,0.31851852,0.4148148,,0.6296296
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.074074075,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.044444446,0.0,0.0,0.02962963,0.0,0.02962963,0.08888889,0.13333334,0.0,0.0,0.2,0.2,0.0,0.33333334,0.08148148,0.33333334,0.21481481,0.0074074073,0.37037036,
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_small/leaderboard_overview_explorer.html
index e185d70aaf56aa10455efd06c8a075b673a39b9f..259579a172b5fde83d52c73cbb56f961b75b84eb 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | All Tasks | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_small/leaderboard_table.html b/data/imputation_no/splits_all/tasks_all/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..9df0f30fc0c1eb11b7a8a7128d6751e5b14df15a
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_all/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1744.0,"score":0.897,"rank":6.9,"hrank":1.74,"imp":4.654,"train_time":39.57,"predict_time":5.358,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+142\/-93","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1638.0,"score":0.642,"rank":10.67,"hrank":4.73,"imp":9.279,"train_time":346.58,"predict_time":6.557,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-52","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1636.0,"score":0.675,"rank":10.77,"hrank":4.53,"imp":10.766,"train_time":9.17,"predict_time":0.416,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-56","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1599.0,"score":0.616,"rank":12.44,"hrank":6.63,"imp":11.961,"train_time":7.03,"predict_time":0.548,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-47","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1595.0,"score":0.593,"rank":12.63,"hrank":5.6,"imp":11.198,"train_time":2289.39,"predict_time":8.048,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-62","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1572.0,"score":0.633,"rank":13.78,"hrank":5.25,"imp":11.56,"train_time":3.93,"predict_time":0.152,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-78","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1555.0,"score":0.547,"rank":14.67,"hrank":6.51,"imp":11.907,"train_time":2289.39,"predict_time":1.143,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-56","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1542.0,"score":0.536,"rank":15.41,"hrank":8.41,"imp":12.32,"train_time":6.76,"predict_time":0.636,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-49","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1480.0,"score":0.402,"rank":19.13,"hrank":11.38,"imp":14.533,"train_time":3771.85,"predict_time":21.902,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-44","imputed":false},{"position":9,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1448.0,"score":0.391,"rank":21.19,"hrank":9.11,"imp":15.33,"train_time":4.69,"predict_time":0.192,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-54","imputed":false},{"position":10,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1448.0,"score":0.398,"rank":21.22,"hrank":9.27,"imp":15.246,"train_time":5119.78,"predict_time":218.453,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-60","imputed":false},{"position":11,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1444.0,"score":0.326,"rank":21.52,"hrank":13.45,"imp":15.598,"train_time":3716.57,"predict_time":2.547,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-46","imputed":false},{"position":12,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1429.0,"score":0.325,"rank":22.52,"hrank":12.22,"imp":15.675,"train_time":2585.58,"predict_time":3.838,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-48","imputed":false},{"position":13,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1415.0,"score":0.356,"rank":23.49,"hrank":10.85,"imp":15.884,"train_time":5119.78,"predict_time":28.351,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-58","imputed":false},{"position":14,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1405.0,"score":0.353,"rank":24.21,"hrank":9.41,"imp":15.155,"train_time":35.51,"predict_time":6.828,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-72","imputed":false},{"position":15,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1400.0,"score":0.293,"rank":24.6,"hrank":13.5,"imp":15.91,"train_time":3771.85,"predict_time":1.778,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+51\/-43","imputed":false},{"position":16,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1390.0,"score":0.235,"rank":25.34,"hrank":19.43,"imp":16.937,"train_time":892.49,"predict_time":2.57,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-34","imputed":false},{"position":17,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1373.0,"score":0.258,"rank":26.55,"hrank":14.04,"imp":16.561,"train_time":2585.58,"predict_time":0.436,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-48","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1368.0,"score":0.211,"rank":26.99,"hrank":19.52,"imp":17.041,"train_time":1739.17,"predict_time":0.396,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-39","imputed":false},{"position":19,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1363.0,"score":0.29,"rank":27.32,"hrank":15.91,"imp":17.224,"train_time":2.76,"predict_time":0.058,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-49","imputed":false},{"position":20,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1361.0,"score":0.199,"rank":27.46,"hrank":18.84,"imp":17.421,"train_time":22811.74,"predict_time":841.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+44\/-39","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1354.0,"score":0.192,"rank":28.01,"hrank":20.68,"imp":17.279,"train_time":1739.17,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-36","imputed":false},{"position":22,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1353.0,"score":0.185,"rank":28.07,"hrank":21.83,"imp":17.389,"train_time":892.49,"predict_time":0.346,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-31","imputed":false},{"position":23,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1332.0,"score":0.275,"rank":29.74,"hrank":13.6,"imp":18.023,"train_time":50.33,"predict_time":43.707,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-67","imputed":false},{"position":24,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1330.0,"score":0.203,"rank":29.84,"hrank":18.17,"imp":17.868,"train_time":1126.97,"predict_time":2.94,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-42","imputed":false},{"position":25,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1325.0,"score":0.158,"rank":30.28,"hrank":23.33,"imp":17.924,"train_time":884.48,"predict_time":2.367,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+33\/-37","imputed":false},{"position":26,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1323.0,"score":0.178,"rank":30.38,"hrank":15.63,"imp":18.056,"train_time":6523.55,"predict_time":0.505,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+40\/-42","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1323.0,"score":0.227,"rank":30.39,"hrank":13.02,"imp":18.319,"train_time":6523.55,"predict_time":8.39,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-56","imputed":false},{"position":28,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1320.0,"score":0.159,"rank":30.6,"hrank":22.44,"imp":18.203,"train_time":8.46,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+35\/-35","imputed":false},{"position":29,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1319.0,"score":0.263,"rank":30.69,"hrank":11.14,"imp":18.839,"train_time":16.23,"predict_time":1.277,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-57","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1306.0,"score":0.13,"rank":31.71,"hrank":23.26,"imp":18.133,"train_time":884.48,"predict_time":0.393,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-40","imputed":false},{"position":31,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1296.0,"score":0.188,"rank":32.54,"hrank":19.64,"imp":18.258,"train_time":9.94,"predict_time":0.445,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-45","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1282.0,"score":0.157,"rank":33.6,"hrank":20.71,"imp":18.894,"train_time":1126.97,"predict_time":0.163,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+42\/-42","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1278.0,"score":0.133,"rank":33.92,"hrank":23.84,"imp":19.516,"train_time":1218.87,"predict_time":0.455,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-46","imputed":false},{"position":34,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1260.0,"score":0.125,"rank":35.33,"hrank":24.98,"imp":19.124,"train_time":3865.95,"predict_time":2.157,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-49","imputed":false},{"position":35,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1260.0,"score":0.11,"rank":35.37,"hrank":25.32,"imp":19.803,"train_time":1218.87,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-50","imputed":false},{"position":36,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1260.0,"score":0.109,"rank":35.38,"hrank":25.23,"imp":19.954,"train_time":2499.4,"predict_time":0.153,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-45","imputed":false},{"position":37,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1258.0,"score":0.092,"rank":35.55,"hrank":26.17,"imp":19.245,"train_time":22811.74,"predict_time":161.221,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+37\/-35","imputed":false},{"position":38,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1237.0,"score":0.097,"rank":37.17,"hrank":21.05,"imp":21.221,"train_time":16.28,"predict_time":0.321,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-51","imputed":false},{"position":39,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1231.0,"score":0.079,"rank":37.67,"hrank":26.59,"imp":20.511,"train_time":2499.4,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-44","imputed":false},{"position":40,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1221.0,"score":0.079,"rank":38.44,"hrank":26.97,"imp":20.018,"train_time":15.69,"predict_time":4.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+39\/-43","imputed":false},{"position":41,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1216.0,"score":0.087,"rank":38.82,"hrank":25.36,"imp":20.941,"train_time":5.49,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-58","imputed":false},{"position":42,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1214.0,"score":0.124,"rank":38.99,"hrank":22.53,"imp":20.712,"train_time":1679.4,"predict_time":8.069,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-60","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1211.0,"score":0.109,"rank":39.24,"hrank":24.93,"imp":21.641,"train_time":822.58,"predict_time":1.501,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-63","imputed":false},{"position":44,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1203.0,"score":0.074,"rank":39.86,"hrank":25.92,"imp":21.249,"train_time":7.39,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-53","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1196.0,"score":0.078,"rank":40.43,"hrank":29.54,"imp":20.862,"train_time":3865.95,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-49","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1187.0,"score":0.101,"rank":41.1,"hrank":23.0,"imp":22.339,"train_time":822.58,"predict_time":0.214,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-66","imputed":false},{"position":47,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1166.0,"score":0.043,"rank":42.71,"hrank":35.31,"imp":21.229,"train_time":3.61,"predict_time":0.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-44","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1165.0,"score":0.061,"rank":42.8,"hrank":27.62,"imp":21.847,"train_time":1679.4,"predict_time":0.826,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-52","imputed":false},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1163.0,"score":0.018,"rank":42.91,"hrank":37.21,"imp":21.244,"train_time":3.29,"predict_time":0.253,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+27\/-37","imputed":false},{"position":50,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1150.0,"score":0.073,"rank":43.96,"hrank":26.88,"imp":23.216,"train_time":590.04,"predict_time":1.46,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-70","imputed":false},{"position":51,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1132.0,"score":0.178,"rank":45.27,"hrank":12.31,"imp":25.491,"train_time":40848.87,"predict_time":5.606,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-83","imputed":false},{"position":52,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1130.0,"score":0.174,"rank":45.38,"hrank":11.81,"imp":25.525,"train_time":40848.87,"predict_time":25.936,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-84","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1113.0,"score":0.046,"rank":46.61,"hrank":25.91,"imp":23.948,"train_time":590.04,"predict_time":0.15,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-68","imputed":false},{"position":54,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1084.0,"score":0.026,"rank":48.72,"hrank":37.31,"imp":28.451,"train_time":342.04,"predict_time":0.895,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-59","imputed":false},{"position":55,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1066.0,"score":0.039,"rank":49.88,"hrank":31.58,"imp":24.914,"train_time":524.48,"predict_time":161.813,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-74","imputed":false},{"position":56,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1058.0,"score":0.015,"rank":50.45,"hrank":36.89,"imp":24.761,"train_time":13.69,"predict_time":0.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-50","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1048.0,"score":0.014,"rank":51.08,"hrank":37.33,"imp":29.908,"train_time":342.04,"predict_time":0.313,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-57","imputed":false},{"position":58,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1043.0,"score":0.034,"rank":51.4,"hrank":42.96,"imp":26.639,"train_time":5.45,"predict_time":1.098,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-86","imputed":false},{"position":59,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1033.0,"score":0.109,"rank":52.04,"hrank":19.43,"imp":30.36,"train_time":471.66,"predict_time":7.214,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-124","imputed":false},{"position":60,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1020.0,"score":0.017,"rank":52.89,"hrank":44.87,"imp":27.588,"train_time":4.19,"predict_time":0.566,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-69","imputed":false},{"position":61,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.004,"rank":54.07,"hrank":48.1,"imp":28.377,"train_time":0.51,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-71","imputed":false},{"position":62,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1000.0,"score":0.005,"rank":54.08,"hrank":37.17,"imp":26.856,"train_time":5.48,"predict_time":0.684,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-62","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1000.0,"score":0.022,"rank":54.08,"hrank":35.68,"imp":29.267,"train_time":111.87,"predict_time":1.214,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-94","imputed":false},{"position":64,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":974.0,"score":0.042,"rank":55.53,"hrank":37.13,"imp":33.999,"train_time":521.07,"predict_time":0.646,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-95","imputed":false},{"position":65,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":954.0,"score":0.03,"rank":56.63,"hrank":38.68,"imp":34.411,"train_time":521.07,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-99","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":947.0,"score":0.0,"rank":57.02,"hrank":41.72,"imp":32.717,"train_time":39.09,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-58","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":912.0,"score":0.012,"rank":58.72,"hrank":42.52,"imp":33.953,"train_time":111.87,"predict_time":0.111,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-89","imputed":false},{"position":68,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":879.0,"score":0.019,"rank":60.23,"hrank":43.4,"imp":37.25,"train_time":2.59,"predict_time":0.138,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-118","imputed":false},{"position":69,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":670.0,"score":0.0,"rank":66.58,"hrank":57.89,"imp":47.022,"train_time":0.37,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-134","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_small/pareto_front_explorer.html
index e77f20d0284063cc5040e1a683da2f6c1209b7cf..ac1e9ba3f9d3d753a71cb1f6ba384b1a03566fad 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0455896074,"imp":18.2032871809,"elo":1320.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0508532668,"imp":17.2788594488,"elo":1354.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"imp":17.0406428568,"elo":1367.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0413889487,"imp":20.9409846453,"elo":1216.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0359833093,"imp":19.8033433908,"elo":1259.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.454903865,"imp":19.5160246588,"elo":1278.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0171560587,"imp":21.2493491202,"elo":1203.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0172938338,"imp":20.510545928,"elo":1230.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1529693961,"imp":19.9544651059,"elo":1259.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5663988088,"imp":27.5877514986,"elo":1019.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2136923717,"imp":22.3391093508,"elo":1187.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5011279394,"imp":21.6412157659,"elo":1210.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6842938244,"imp":26.8558318039,"elo":999.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.82578692,"imp":21.847378167,"elo":1164.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.0688374767,"imp":20.7123810494,"elo":1214.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1695753133,"imp":21.2294827321,"elo":1166.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3458375858,"imp":17.3893988993,"elo":1353.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.5704534161,"imp":16.9366876587,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.8284826783,"imp":15.1548850837,"elo":1405.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.320600689,"imp":21.2208702147,"elo":1237.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5053503954,"imp":18.0561303362,"elo":1323.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3903226253,"imp":18.3187609832,"elo":1323.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0406570787,"imp":32.7166050918,"elo":946.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3125464363,"imp":29.9076789954,"elo":1048.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.8948102984,"imp":28.4508866176,"elo":1083.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0885648768,"imp":28.3765267033,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1503007896,"imp":23.9480322533,"elo":1113.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.4602306271,"imp":23.2155091609,"elo":1149.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.6888676463,"imp":20.0184669583,"elo":1221.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.7776750936,"imp":15.9097834653,"elo":1399.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.9024962791,"imp":14.5333602982,"elo":1479.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6362272808,"imp":12.3200887376,"elo":1541.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1425914845,"imp":11.9071180723,"elo":1555.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0475948482,"imp":11.1975772448,"elo":1595.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2773508677,"imp":18.8388371702,"elo":1319.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"imp":18.0230648308,"elo":1331.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":28.3513403832,"imp":15.883723432,"elo":1415.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":218.4533640735,"imp":15.2464708261,"elo":1448.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1922706752,"imp":15.3295501945,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.3578381333,"imp":4.6543552541,"elo":1743.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1520731745,"imp":11.5604085785,"elo":1572.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.444937792,"imp":18.2578343688,"elo":1295.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4361574183,"imp":16.5609732873,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.8381409686,"imp":15.6750950369,"elo":1429.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.548288662,"imp":11.9607415625,"elo":1599.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4158216639,"imp":10.7658050052,"elo":1636.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.2137264423,"imp":30.3598793602,"elo":1033.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.6062080426,"imp":25.4914369812,"elo":1131.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9362072534,"imp":25.5248178568,"elo":1130.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0579126899,"imp":17.2241080479,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1550906392,"imp":24.7609027784,"elo":1057.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1474394112,"imp":20.8615927184,"elo":1195.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1570271396,"imp":19.1243302341,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2525411892,"imp":21.2442258572,"elo":1163.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3929683731,"imp":18.1328413794,"elo":1306.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.3674863949,"imp":17.9237428347,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":161.8134383214,"imp":24.9136024932,"elo":1066.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":161.2206474003,"imp":19.245205487,"elo":1257.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"imp":17.4214786951,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.0976028857,"imp":26.6387104203,"elo":1043.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1632446807,"imp":18.8938182398,"elo":1282.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9395549707,"imp":17.8683612217,"elo":1330.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0455896074,"x_train":8.4625903355,"imp":18.2032871809,"elo":1320.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0508532668,"x_train":1739.1690497298,"imp":17.2788594488,"elo":1354.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"x_train":1739.1690497298,"imp":17.0406428568,"elo":1367.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0413889487,"x_train":5.4899520078,"imp":20.9409846453,"elo":1216.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0359833093,"x_train":1218.8724153796,"imp":19.8033433908,"elo":1259.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.454903865,"x_train":1218.8724153796,"imp":19.5160246588,"elo":1278.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0171560587,"x_train":7.3884202915,"imp":21.2493491202,"elo":1203.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0172938338,"x_train":2499.3997394449,"imp":20.510545928,"elo":1230.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1529693961,"x_train":2499.3997394449,"imp":19.9544651059,"elo":1259.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5663988088,"x_train":4.1941752843,"imp":27.5877514986,"elo":1019.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2136923717,"x_train":822.5846846043,"imp":22.3391093508,"elo":1187.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5011279394,"x_train":822.5846846043,"imp":21.6412157659,"elo":1210.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6842938244,"x_train":5.483894574,"imp":26.8558318039,"elo":999.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.82578692,"x_train":1679.3959423676,"imp":21.847378167,"elo":1164.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.0688374767,"x_train":1679.3959423676,"imp":20.7123810494,"elo":1214.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1695753133,"x_train":3.6086515992,"imp":21.2294827321,"elo":1166.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3458375858,"x_train":892.485407479,"imp":17.3893988993,"elo":1353.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.5704534161,"x_train":892.485407479,"imp":16.9366876587,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.8284826783,"x_train":35.5096105631,"imp":15.1548850837,"elo":1405.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.320600689,"x_train":16.2796239102,"imp":21.2208702147,"elo":1237.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5053503954,"x_train":6523.5503361576,"imp":18.0561303362,"elo":1323.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3903226253,"x_train":6523.5503361576,"imp":18.3187609832,"elo":1323.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0406570787,"x_train":39.0907116856,"imp":32.7166050918,"elo":946.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3125464363,"x_train":342.0390094668,"imp":29.9076789954,"elo":1048.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.8948102984,"x_train":342.0390094668,"imp":28.4508866176,"elo":1083.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0885648768,"x_train":0.5088535014,"imp":28.3765267033,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1503007896,"x_train":590.0394439851,"imp":23.9480322533,"elo":1113.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.4602306271,"x_train":590.0394439851,"imp":23.2155091609,"elo":1149.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.6888676463,"x_train":15.693235166,"imp":20.0184669583,"elo":1221.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.7776750936,"x_train":3771.8540221068,"imp":15.9097834653,"elo":1399.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.9024962791,"x_train":3771.8540221068,"imp":14.5333602982,"elo":1479.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6362272808,"x_train":6.7587899978,"imp":12.3200887376,"elo":1541.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1425914845,"x_train":2289.3887157667,"imp":11.9071180723,"elo":1555.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0475948482,"x_train":2289.3887157667,"imp":11.1975772448,"elo":1595.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2773508677,"x_train":16.2304597591,"imp":18.8388371702,"elo":1319.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"x_train":50.3250241527,"imp":18.0230648308,"elo":1331.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":28.3513403832,"x_train":5119.7769851031,"imp":15.883723432,"elo":1415.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":218.4533640735,"x_train":5119.7769851031,"imp":15.2464708261,"elo":1448.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1922706752,"x_train":4.6911130946,"imp":15.3295501945,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.3578381333,"x_train":39.5720422516,"imp":4.6543552541,"elo":1743.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1520731745,"x_train":3.9320097545,"imp":11.5604085785,"elo":1572.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.444937792,"x_train":9.9360092189,"imp":18.2578343688,"elo":1295.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4361574183,"x_train":2585.5798636409,"imp":16.5609732873,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.8381409686,"x_train":2585.5798636409,"imp":15.6750950369,"elo":1429.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.548288662,"x_train":7.0342890773,"imp":11.9607415625,"elo":1599.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4158216639,"x_train":9.1699299544,"imp":10.7658050052,"elo":1636.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.2137264423,"x_train":471.6614585494,"imp":30.3598793602,"elo":1033.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.6062080426,"x_train":40848.8665748301,"imp":25.4914369812,"elo":1131.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9362072534,"x_train":40848.8665748301,"imp":25.5248178568,"elo":1130.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0579126899,"x_train":2.7560512439,"imp":17.2241080479,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1550906392,"x_train":13.688894957,"imp":24.7609027784,"elo":1057.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1474394112,"x_train":3865.9507169384,"imp":20.8615927184,"elo":1195.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1570271396,"x_train":3865.9507169384,"imp":19.1243302341,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2525411892,"x_train":3.2873922338,"imp":21.2442258572,"elo":1163.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3929683731,"x_train":884.4808277373,"imp":18.1328413794,"elo":1306.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.3674863949,"x_train":884.4808277373,"imp":17.9237428347,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":161.8134383214,"x_train":524.4765295004,"imp":24.9136024932,"elo":1066.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":161.2206474003,"x_train":22811.7440512431,"imp":19.245205487,"elo":1257.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"x_train":22811.7440512431,"imp":17.4214786951,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.0976028857,"x_train":5.4460301782,"imp":26.6387104203,"elo":1043.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1632446807,"x_train":1126.9686387245,"imp":18.8938182398,"elo":1282.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9395549707,"x_train":1126.9686387245,"imp":17.8683612217,"elo":1330.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_small/tuning_trajectories_explorer.html
index f201d35200b9e0ff87289983c594c36564d77819..efaf4c10a0a2cf8195fd9eba6e92a84efb138b64 100644
--- a/data/imputation_no/splits_all/tasks_all/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_all/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":8.4625903355,"imp":18.2280187213,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":20.3461935557,"imp":17.6191786457,"elo":1359.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":49.7230862409,"imp":17.2530074288,"elo":1378.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":92.4850850095,"imp":17.1181307732,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":216.0131399202,"imp":17.0130557382,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":428.4029173435,"imp":17.0161789246,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":862.883798881,"imp":17.0330051965,"elo":1384.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1291.8438396447,"imp":17.0574202583,"elo":1383.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1739.1690497298,"imp":17.0636010417,"elo":1382.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.4899520078,"imp":20.9649833386,"elo":1223.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.3341031504,"imp":20.651701091,"elo":1240.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":28.8927967722,"imp":20.4677885555,"elo":1251.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":58.8720319788,"imp":20.2954710786,"elo":1259.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":150.365699836,"imp":19.988689801,"elo":1271.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":302.4787198089,"imp":19.8311213784,"elo":1278.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":606.008218265,"imp":19.6575431116,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":908.7557510718,"imp":19.5845815506,"elo":1288.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1218.8724153796,"imp":19.5396774568,"elo":1291.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.3884202915,"imp":21.2740966844,"elo":1208.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":19.1397994387,"imp":20.9972221697,"elo":1219.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":59.6568017021,"imp":20.6374888164,"elo":1238.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":123.907431834,"imp":20.4879302207,"elo":1246.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":318.4600106407,"imp":20.2903523778,"elo":1254.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":623.1113495163,"imp":20.1535580968,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1256.5403166692,"imp":20.0477752726,"elo":1267.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1885.0271107114,"imp":20.0096694171,"elo":1270.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2499.3997394449,"imp":19.9833735961,"elo":1271.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.1941752843,"imp":27.6094146225,"elo":1021.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":8.7200784389,"imp":25.8859896167,"elo":1062.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":21.6852759737,"imp":23.4412252197,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.1231699977,"imp":22.614514664,"elo":1170.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":107.3953873718,"imp":22.0014871604,"elo":1202.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":210.0602463393,"imp":21.8410685277,"elo":1212.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":412.076486211,"imp":21.7369761469,"elo":1218.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":614.2406759256,"imp":21.6766224744,"elo":1223.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":822.5846846043,"imp":21.6660994807,"elo":1223.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.483894574,"imp":26.8794725708,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.5302504742,"imp":25.6054405713,"elo":1045.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":42.0685403004,"imp":23.5407935215,"elo":1120.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":79.1969033107,"imp":22.5779009907,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":202.916509576,"imp":21.7002408479,"elo":1190.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":413.0147999434,"imp":21.2977809136,"elo":1207.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":829.4495951905,"imp":21.0152529476,"elo":1217.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1248.1221717005,"imp":20.8369661477,"elo":1223.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1679.3959423676,"imp":20.7334310003,"elo":1228.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":3.6086515992,"imp":21.2537027344,"elo":1168.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":8.8659554243,"imp":19.3646592722,"elo":1257.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":21.9215058782,"imp":18.1890159236,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":43.5933949927,"imp":17.6954923289,"elo":1361.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":110.831400314,"imp":17.3288628751,"elo":1386.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":222.3578942114,"imp":17.2074284064,"elo":1395.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":444.7316933954,"imp":17.0601040338,"elo":1404.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":664.9958936779,"imp":16.997913582,"elo":1408.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":892.485407479,"imp":16.9609351484,"elo":1409.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":35.5096105631,"imp":15.2105243607,"elo":1422.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":16.2796239102,"imp":21.2450197087,"elo":1248.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":54.2164838254,"imp":20.025741966,"elo":1273.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":158.539941202,"imp":18.0790517258,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":318.0880622722,"imp":17.6681952827,"elo":1366.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":784.3400464924,"imp":17.2527740369,"elo":1384.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1619.6093744476,"imp":17.4468045376,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3239.1118632438,"imp":17.8664816565,"elo":1357.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4838.9886069116,"imp":18.1735709683,"elo":1345.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6523.5503361576,"imp":18.3476502006,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":39.0907116856,"imp":33.3184790948,"elo":943.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":114.1968017072,"imp":31.1917379277,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":285.3525964278,"imp":28.6615181702,"elo":1083.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":342.0390094668,"imp":28.4697890058,"elo":1090.2,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.5088535014,"imp":28.4099466734,"elo":999.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.3556051422,"imp":26.1048441038,"elo":1044.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":12.849220133,"imp":24.8119898471,"elo":1078.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":27.9677019589,"imp":24.0541517658,"elo":1108.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":70.7546625501,"imp":23.5114987788,"elo":1142.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":143.9864831887,"imp":23.3772943599,"elo":1150.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":290.4393098655,"imp":23.290473049,"elo":1155.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":439.3706578063,"imp":23.2536620206,"elo":1157.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":590.0394439851,"imp":23.2360252638,"elo":1158.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":15.693235166,"imp":20.0442695794,"elo":1226.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":33.8496808632,"imp":17.6242916884,"elo":1331.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":89.0225754036,"imp":15.8928959265,"elo":1424.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":186.9144346811,"imp":15.3491944105,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":469.4672918765,"imp":14.9750877506,"elo":1481.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":941.4479936377,"imp":14.7960730646,"elo":1494.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1881.365821255,"imp":14.6917154663,"elo":1499.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2818.5782940854,"imp":14.6257702347,"elo":1503.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3771.8540221068,"imp":14.561284233,"elo":1503.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.7587899978,"imp":12.3473610571,"elo":1560.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":20.0220773915,"imp":12.2792149939,"elo":1565.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":54.3500923902,"imp":11.8942706314,"elo":1583.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":108.9879012879,"imp":11.6985289969,"elo":1602.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":284.7684460337,"imp":11.4976641908,"elo":1612.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":566.4138152119,"imp":11.3895253513,"elo":1616.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1137.2039338041,"imp":11.313893859,"elo":1617.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1710.4116369596,"imp":11.2784867403,"elo":1617.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2289.3887157667,"imp":11.2346661703,"elo":1616.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":16.2304597591,"imp":18.8641600585,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":50.3250241527,"imp":18.0466614084,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":73.7698941637,"imp":17.3572619789,"elo":1371.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":139.8184770834,"imp":16.6623931755,"elo":1406.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":256.4808826589,"imp":16.3334862039,"elo":1422.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":631.0078513376,"imp":15.7810958758,"elo":1444.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1275.2299790654,"imp":15.5196078047,"elo":1455.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2539.0709530625,"imp":15.3638517402,"elo":1463.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3811.2080246053,"imp":15.300477535,"elo":1465.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":5119.7769851031,"imp":15.2766430564,"elo":1465.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":4.6911130946,"imp":15.3538502629,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":39.5720422516,"imp":4.6820290321,"elo":1759.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":3.9320097545,"imp":11.5877337798,"elo":1591.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.9360092189,"imp":18.2835096672,"elo":1306.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":21.7462127279,"imp":17.8945647047,"elo":1326.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":59.9879738748,"imp":17.4324520322,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":127.1136519479,"imp":16.8771348131,"elo":1388.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":318.7083632316,"imp":16.4088305156,"elo":1413.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":637.1067463953,"imp":16.1175021648,"elo":1428.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1278.8911702563,"imp":15.9232311189,"elo":1439.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1925.3471149571,"imp":15.7555452636,"elo":1445.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2585.5798636409,"imp":15.7017988648,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":7.0342890773,"imp":11.9877936702,"elo":1622.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":9.1699299544,"imp":10.7947682179,"elo":1658.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":471.6614585494,"imp":30.3834415771,"elo":1039.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":797.2338293006,"imp":30.2375539363,"elo":1046.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1402.6890753282,"imp":29.9241904785,"elo":1055.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2175.0878519494,"imp":29.1723668497,"elo":1073.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5045.5225760468,"imp":27.879960409,"elo":1096.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":10226.3553477645,"imp":26.6765196691,"elo":1116.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":20731.065214537,"imp":26.0530962107,"elo":1130.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":30617.5877879932,"imp":25.7588335871,"elo":1138.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":40848.8665748301,"imp":25.5499633652,"elo":1146.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":2.7560512439,"imp":17.2485004962,"elo":1378.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.688894957,"imp":24.7778791622,"elo":1066.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":32.1799419222,"imp":22.8576781803,"elo":1125.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":92.7470191468,"imp":21.2695716655,"elo":1176.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":198.4037882463,"imp":20.5812171314,"elo":1207.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":483.4739388196,"imp":19.9836461514,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":974.0069917832,"imp":19.6092619128,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1885.3999284296,"imp":19.3602045605,"elo":1264.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2882.8865617337,"imp":19.23934881,"elo":1271.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3865.9507169384,"imp":19.1423945534,"elo":1276.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":3.2873922338,"imp":21.2676877013,"elo":1165.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":8.2975106968,"imp":19.3516488334,"elo":1250.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":21.7087779063,"imp":18.8269227674,"elo":1282.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":42.6971413089,"imp":18.528925418,"elo":1303.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":109.8273979632,"imp":18.2852210699,"elo":1318.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":219.7348179637,"imp":18.1825022637,"elo":1324.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":439.6020314301,"imp":18.093743791,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":659.3117486382,"imp":18.0146907017,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":884.4808277373,"imp":17.9486693714,"elo":1340.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":524.4765295004,"imp":24.9366166817,"elo":1075.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1309.3726391156,"imp":21.9994249967,"elo":1153.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":3961.6553268098,"imp":19.8928754461,"elo":1262.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":8841.9143914963,"imp":19.2012955328,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":21737.9715611002,"imp":17.4721258448,"elo":1374.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":22811.7440512431,"imp":17.4462635555,"elo":1377.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":5.4460301782,"imp":26.6617497581,"elo":1050.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":11.0404626494,"imp":24.0110852496,"elo":1111.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":27.9509507616,"imp":20.5809187462,"elo":1229.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":55.9413410319,"imp":19.0350022113,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":140.091009561,"imp":18.2435432016,"elo":1333.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":280.2176062527,"imp":18.0073358467,"elo":1344.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":560.4680551302,"imp":17.9184645628,"elo":1346.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":840.8353293445,"imp":17.905379958,"elo":1346.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1126.9686387245,"imp":17.8944317384,"elo":1346.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0455896074,"x_train":8.4625903355,"imp":18.2280187213,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0966097048,"x_train":20.3461935557,"imp":17.6191786457,"elo":1359.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1806528582,"x_train":49.7230862409,"imp":17.2530074288,"elo":1378.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2398344067,"x_train":92.4850850095,"imp":17.1181307732,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2815893316,"x_train":216.0131399202,"imp":17.0130557382,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3179145648,"x_train":428.4029173435,"imp":17.0161789246,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806002107,"x_train":862.883798881,"imp":17.0330051965,"elo":1384.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3873615123,"x_train":1291.8438396447,"imp":17.0574202583,"elo":1383.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"x_train":1739.1690497298,"imp":17.0636010417,"elo":1382.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0413889487,"x_train":5.4899520078,"imp":20.9649833386,"elo":1223.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0801821127,"x_train":11.3341031504,"imp":20.651701091,"elo":1240.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1457038997,"x_train":28.8927967722,"imp":20.4677885555,"elo":1251.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2046887662,"x_train":58.8720319788,"imp":20.2954710786,"elo":1259.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2874258032,"x_train":150.365699836,"imp":19.988689801,"elo":1271.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3447163363,"x_train":302.4787198089,"imp":19.8311213784,"elo":1278.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4006058605,"x_train":606.008218265,"imp":19.6575431116,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4327969131,"x_train":908.7557510718,"imp":19.5845815506,"elo":1288.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4496610991,"x_train":1218.8724153796,"imp":19.5396774568,"elo":1291.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0171560587,"x_train":7.3884202915,"imp":21.2740966844,"elo":1208.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333155319,"x_train":19.1397994387,"imp":20.9972221697,"elo":1219.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.060202021,"x_train":59.6568017021,"imp":20.6374888164,"elo":1238.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0794108609,"x_train":123.907431834,"imp":20.4879302207,"elo":1246.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1075090355,"x_train":318.4600106407,"imp":20.2903523778,"elo":1254.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1237830888,"x_train":623.1113495163,"imp":20.1535580968,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1440522632,"x_train":1256.5403166692,"imp":20.0477752726,"elo":1267.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1469719753,"x_train":1885.0271107114,"imp":20.0096694171,"elo":1270.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1440681396,"x_train":2499.3997394449,"imp":19.9833735961,"elo":1271.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5663988088,"x_train":4.1941752843,"imp":27.6094146225,"elo":1021.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7011757792,"x_train":8.7200784389,"imp":25.8859896167,"elo":1062.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7330007013,"x_train":21.6852759737,"imp":23.4412252197,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7428670659,"x_train":43.1231699977,"imp":22.614514664,"elo":1170.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0188559546,"x_train":107.3953873718,"imp":22.0014871604,"elo":1202.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2214373513,"x_train":210.0602463393,"imp":21.8410685277,"elo":1212.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4038021731,"x_train":412.076486211,"imp":21.7369761469,"elo":1218.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4626957587,"x_train":614.2406759256,"imp":21.6766224744,"elo":1223.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5011279394,"x_train":822.5846846043,"imp":21.6660994807,"elo":1223.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6842938244,"x_train":5.483894574,"imp":26.8794725708,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4234649143,"x_train":14.5302504742,"imp":25.6054405713,"elo":1045.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7086381558,"x_train":42.0685403004,"imp":23.5407935215,"elo":1120.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.9064906779,"x_train":79.1969033107,"imp":22.5779009907,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.5176917438,"x_train":202.916509576,"imp":21.7002408479,"elo":1190.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.8255235607,"x_train":413.0147999434,"imp":21.2977809136,"elo":1207.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.5799140147,"x_train":829.4495951905,"imp":21.0152529476,"elo":1217.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.9133058467,"x_train":1248.1221717005,"imp":20.8369661477,"elo":1223.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.0772277344,"x_train":1679.3959423676,"imp":20.7334310003,"elo":1228.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1695753133,"x_train":3.6086515992,"imp":21.2537027344,"elo":1168.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7271936335,"x_train":8.8659554243,"imp":19.3646592722,"elo":1257.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1652455249,"x_train":21.9215058782,"imp":18.1890159236,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4098790834,"x_train":43.5933949927,"imp":17.6954923289,"elo":1361.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.8172136654,"x_train":110.831400314,"imp":17.3288628751,"elo":1386.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.0744273861,"x_train":222.3578942114,"imp":17.2074284064,"elo":1395.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.3211352586,"x_train":444.7316933954,"imp":17.0601040338,"elo":1404.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.4252767385,"x_train":664.9958936779,"imp":16.997913582,"elo":1408.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.5817474842,"x_train":892.485407479,"imp":16.9609351484,"elo":1409.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":6.8284826783,"x_train":35.5096105631,"imp":15.2105243607,"elo":1422.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.320600689,"x_train":16.2796239102,"imp":21.2450197087,"elo":1248.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8565664695,"x_train":54.2164838254,"imp":20.025741966,"elo":1273.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9156693463,"x_train":158.539941202,"imp":18.0790517258,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7858437586,"x_train":318.0880622722,"imp":17.6681952827,"elo":1366.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.2195144755,"x_train":784.3400464924,"imp":17.2527740369,"elo":1384.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5749647915,"x_train":1619.6093744476,"imp":17.4468045376,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.0637124303,"x_train":3239.1118632438,"imp":17.8664816565,"elo":1357.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8596602241,"x_train":4838.9886069116,"imp":18.1735709683,"elo":1345.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3848981892,"x_train":6523.5503361576,"imp":18.3476502006,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0406570787,"x_train":39.0907116856,"imp":33.3184790948,"elo":943.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3252722558,"x_train":114.1968017072,"imp":31.1917379277,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.783304892,"x_train":285.3525964278,"imp":28.6615181702,"elo":1083.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.8948102984,"x_train":342.0390094668,"imp":28.4697890058,"elo":1090.2,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0885648768,"x_train":0.5088535014,"imp":28.4099466734,"elo":999.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2741812716,"x_train":3.3556051422,"imp":26.1048441038,"elo":1044.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4818339926,"x_train":12.849220133,"imp":24.8119898471,"elo":1078.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6789925729,"x_train":27.9677019589,"imp":24.0541517658,"elo":1108.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8688854,"x_train":70.7546625501,"imp":23.5114987788,"elo":1142.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0449947563,"x_train":143.9864831887,"imp":23.3772943599,"elo":1150.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2523875096,"x_train":290.4393098655,"imp":23.290473049,"elo":1155.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.3737687487,"x_train":439.3706578063,"imp":23.2536620206,"elo":1157.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4599601668,"x_train":590.0394439851,"imp":23.2360252638,"elo":1158.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.6888676463,"x_train":15.693235166,"imp":20.0442695794,"elo":1226.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.699336819,"x_train":33.8496808632,"imp":17.6242916884,"elo":1331.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.361104313,"x_train":89.0225754036,"imp":15.8928959265,"elo":1424.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.4140492169,"x_train":186.9144346811,"imp":15.3491944105,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.9271276382,"x_train":469.4672918765,"imp":14.9750877506,"elo":1481.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.2043254803,"x_train":941.4479936377,"imp":14.7960730646,"elo":1494.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.3143880217,"x_train":1881.365821255,"imp":14.6917154663,"elo":1499.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":20.5026820042,"x_train":2818.5782940854,"imp":14.6257702347,"elo":1503.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.9024962791,"x_train":3771.8540221068,"imp":14.561284233,"elo":1503.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6362272808,"x_train":6.7587899978,"imp":12.3473610571,"elo":1560.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3637258308,"x_train":20.0220773915,"imp":12.2792149939,"elo":1565.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.7767959403,"x_train":54.3500923902,"imp":11.8942706314,"elo":1583.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.0029832468,"x_train":108.9879012879,"imp":11.6985289969,"elo":1602.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7106402076,"x_train":284.7684460337,"imp":11.4976641908,"elo":1612.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.7935919179,"x_train":566.4138152119,"imp":11.3895253513,"elo":1616.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.3891270541,"x_train":1137.2039338041,"imp":11.313893859,"elo":1617.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8718643944,"x_train":1710.4116369596,"imp":11.2784867403,"elo":1617.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3533448181,"x_train":2289.3887157667,"imp":11.2346661703,"elo":1616.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2773508677,"x_train":16.2304597591,"imp":18.8641600585,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"x_train":50.3250241527,"imp":18.0466614084,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.1120162901,"x_train":73.7698941637,"imp":17.3572619789,"elo":1371.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":66.0787878177,"x_train":139.8184770834,"imp":16.6623931755,"elo":1406.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":83.6221921837,"x_train":256.4808826589,"imp":16.3334862039,"elo":1422.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":110.2479942847,"x_train":631.0078513376,"imp":15.7810958758,"elo":1444.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":138.475374244,"x_train":1275.2299790654,"imp":15.5196078047,"elo":1455.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":167.4276056736,"x_train":2539.0709530625,"imp":15.3638517402,"elo":1463.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":188.4316559068,"x_train":3811.2080246053,"imp":15.300477535,"elo":1465.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":214.0824864784,"x_train":5119.7769851031,"imp":15.2766430564,"elo":1465.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1922706752,"x_train":4.6911130946,"imp":15.3538502629,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":5.3578381333,"x_train":39.5720422516,"imp":4.6820290321,"elo":1759.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1520731745,"x_train":3.9320097545,"imp":11.5877337798,"elo":1591.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.444937792,"x_train":9.9360092189,"imp":18.2835096672,"elo":1306.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.8480171202,"x_train":21.7462127279,"imp":17.8945647047,"elo":1326.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.6121786403,"x_train":59.9879738748,"imp":17.4324520322,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.0828680317,"x_train":127.1136519479,"imp":16.8771348131,"elo":1388.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.6698479453,"x_train":318.7083632316,"imp":16.4088305156,"elo":1413.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.1238178806,"x_train":637.1067463953,"imp":16.1175021648,"elo":1428.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.4844153286,"x_train":1278.8911702563,"imp":15.9232311189,"elo":1439.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":3.6365060902,"x_train":1925.3471149571,"imp":15.7555452636,"elo":1445.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":3.8381409686,"x_train":2585.5798636409,"imp":15.7017988648,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.548288662,"x_train":7.0342890773,"imp":11.9877936702,"elo":1622.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4158216639,"x_train":9.1699299544,"imp":10.7947682179,"elo":1658.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.2137264423,"x_train":471.6614585494,"imp":30.3834415771,"elo":1039.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.2461770633,"x_train":797.2338293006,"imp":30.2375539363,"elo":1046.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.3473767381,"x_train":1402.6890753282,"imp":29.9241904785,"elo":1055.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":11.9018815775,"x_train":2175.0878519494,"imp":29.1723668497,"elo":1073.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.2259261458,"x_train":5045.5225760468,"imp":27.879960409,"elo":1096.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.6020662841,"x_train":10226.3553477645,"imp":26.6765196691,"elo":1116.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.9769855841,"x_train":20731.065214537,"imp":26.0530962107,"elo":1130.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":24.3740533341,"x_train":30617.5877879932,"imp":25.7588335871,"elo":1138.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9362072534,"x_train":40848.8665748301,"imp":25.5499633652,"elo":1146.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0579126899,"x_train":2.7560512439,"imp":17.2485004962,"elo":1378.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1550906392,"x_train":13.688894957,"imp":24.7778791622,"elo":1066.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3108419787,"x_train":32.1799419222,"imp":22.8576781803,"elo":1125.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6204009277,"x_train":92.7470191468,"imp":21.2695716655,"elo":1176.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8859862453,"x_train":198.4037882463,"imp":20.5812171314,"elo":1207.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2789048063,"x_train":483.4739388196,"imp":19.9836461514,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6522358142,"x_train":974.0069917832,"imp":19.6092619128,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8962723592,"x_train":1885.3999284296,"imp":19.3602045605,"elo":1264.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0769295602,"x_train":2882.8865617337,"imp":19.23934881,"elo":1271.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1462178945,"x_train":3865.9507169384,"imp":19.1423945534,"elo":1276.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2525411892,"x_train":3.2873922338,"imp":21.2676877013,"elo":1165.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.668721793,"x_train":8.2975106968,"imp":19.3516488334,"elo":1250.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.3135984732,"x_train":21.7087779063,"imp":18.8269227674,"elo":1282.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4736641563,"x_train":42.6971413089,"imp":18.528925418,"elo":1303.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8610108955,"x_train":109.8273979632,"imp":18.2852210699,"elo":1318.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0831926002,"x_train":219.7348179637,"imp":18.1825022637,"elo":1324.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.2439698975,"x_train":439.6020314301,"imp":18.093743791,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3416077193,"x_train":659.3117486382,"imp":18.0146907017,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.4462956888,"x_train":884.4808277373,"imp":17.9486693714,"elo":1340.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":161.8134383214,"x_train":524.4765295004,"imp":24.9366166817,"elo":1075.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":315.1785851675,"x_train":1309.3726391156,"imp":21.9994249967,"elo":1153.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":530.068709775,"x_train":3961.6553268098,"imp":19.8928754461,"elo":1262.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":748.7519406764,"x_train":8841.9143914963,"imp":19.2012955328,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":828.5067468066,"x_train":21737.9715611002,"imp":17.4721258448,"elo":1374.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"x_train":22811.7440512431,"imp":17.4462635555,"elo":1377.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.0976028857,"x_train":5.4460301782,"imp":26.6617497581,"elo":1050.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.2469331487,"x_train":11.0404626494,"imp":24.0110852496,"elo":1111.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.4842749757,"x_train":27.9509507616,"imp":20.5809187462,"elo":1229.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.8164722651,"x_train":55.9413410319,"imp":19.0350022113,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8569378726,"x_train":140.091009561,"imp":18.2435432016,"elo":1333.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.3361472689,"x_train":280.2176062527,"imp":18.0073358467,"elo":1344.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.5518696321,"x_train":560.4680551302,"imp":17.9184645628,"elo":1346.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.795048319,"x_train":840.8353293445,"imp":17.905379958,"elo":1346.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.9237006693,"x_train":1126.9686387245,"imp":17.8944317384,"elo":1346.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_small/winrate_explorer.html b/data/imputation_no/splits_all/tasks_all/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..045a6a8c94f5cb1af0e3c3bb2afa641e894b448d
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_all/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "TabDPT (tuned)", "AutoGluon 1.4 (best, 4h)", "TabM (tuned + ensembled)", "RealMLP (tuned)", "LimiX (default)", "TabSwift (default)", "TabDPT (default)", "LightGBM (tuned + ensembled)", "TabM (tuned)", "CatBoost (tuned + ensembled)", "iLTM (tuned + ensembled)", "CatBoost (tuned)", "SAP-RPT-OSS (default)", "xRFM (tuned + ensembled)", "LightGBM (tuned)", "XGBoost (tuned + ensembled)", "xRFM (tuned)", "ModernNCA (tuned)", "CatBoost (default)", "XGBoost (tuned)", "ModernNCA (tuned + ensembled)", "TabM (default)", "EBM (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "ModernNCA (default)", "iLTM (tuned)", "ChimeraBoost (tuned + ensembled)", "EBM (tuned)", "RealMLP (default)", "FastaiMLP (tuned + ensembled)", "ChimeraBoost (tuned)", "ExtraTrees (tuned + ensembled)", "EBM (default)", "TorchMLP (tuned)", "ExtraTrees (tuned)", "FastaiMLP (tuned)", "ChimeraBoost (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned + ensembled)", "LightGBM (default)", "XGBoost (default)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "TorchMLP (default)", "TabSTAR (default)", "xRFM (default)", "PerpetualBooster (tuned)", "iLTM (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "FastaiMLP (default)", "Linear (tuned)", "KNN (tuned + ensembled)", "RandomForest (default)", "PerpetualBooster (default)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7679882645606995, 0.7679882645606995, 0.7900146842002869, 0.7628487348556519, 0.7944199442863464, 0.8061674237251282, 0.7738619446754456, 0.8619676828384399, 0.8634361028671265, 0.8839941024780273, 0.8898678421974182, 0.91629958152771, 0.8854625821113586, 0.8722466826438904, 0.9067547917366028, 0.8693098425865173, 0.8913362622261047, 0.9177680015563965, 0.9118942618370056, 0.9280470013618469, 0.9295154213905334, 0.9368575811386108, 0.8575624227523804, 0.919236421585083, 0.9177680015563965, 0.9383260011672974, 0.9104258418083191, 0.9251101613044739, 0.9353891611099243, 0.9412628412246704, 0.9456681609153748, 0.919236421585083, 0.9133626818656921, 0.9559471607208252, 0.9265785813331604, 0.9441996812820435, 0.9383260011672974, 0.9251101613044739, 0.9368575811386108, 0.9236416816711426, 0.9441996812820435, 0.9456681609153748, 0.922173261642456, 0.9632892608642578, 0.9397944211959839, 0.9427312612533569, 0.9588840007781982, 0.8472834229469299, 0.8458150029182434, 0.9662261605262756, 0.9662261605262756, 0.9691630005836487, 0.9647576808929443, 0.9838472604751587, 0.9706314206123352, 0.8854625821113586, 0.9706314206123352, 0.9823788404464722, 0.9486050009727478, 0.98531574010849, 0.9530102610588074, 0.9750367403030396, 0.9588840007781982, 0.9691630005836487, 0.9838472604751587, 0.98531574010849, 0.9706314206123352, 0.9662261605262756, 0.9926578402519226], [0.23201175034046173, NaN, 0.5682819485664368, 0.5345080494880676, 0.5770925283432007, 0.607929527759552, 0.6651982665061951, 0.6358296871185303, 0.7415565252304077, 0.7459618449211121, 0.7635829448699951, 0.7856094241142273, 0.7709251046180725, 0.7856094241142273, 0.781204104423523, 0.7679882645606995, 0.8634361028671265, 0.8179148435592651, 0.8399412631988525, 0.8237885236740112, 0.8546255230903625, 0.8795888423919678, 0.8634361028671265, 0.7665198445320129, 0.8634361028671265, 0.8604992628097534, 0.865638792514801, 0.8634361028671265, 0.8678414225578308, 0.8795888423919678, 0.8737151026725769, 0.8825256824493408, 0.8458150029182434, 0.8619676828384399, 0.9148311018943787, 0.889133632183075, 0.9287812113761902, 0.9060205817222595, 0.8766520023345947, 0.9177680015563965, 0.8744493126869202, 0.907489001750946, 0.9265785813331604, 0.892070472240448, 0.93098384141922, 0.9280470013618469, 0.9140968918800354, 0.919236421585083, 0.8215858936309814, 0.8193832635879517, 0.933920681476593, 0.9530102610588074, 0.9662261605262756, 0.9500734210014343, 0.9500734210014343, 0.9632892608642578, 0.8678414225578308, 0.9647576808929443, 0.9603524208068848, 0.9544786810874939, 0.9706314206123352, 0.9295154213905334, 0.9691630005836487, 0.9353891611099243, 0.9471365809440613, 0.9823788404464722, 0.9735682606697083, 0.9618208408355713, 0.9588840007781982, 0.9941262602806091], [0.23201175034046173, 0.43171805143356323, NaN, 0.47136563062667847, 0.5227606296539307, 0.5976505279541016, 0.6049926280975342, 0.5770925283432007, 0.7540381550788879, 0.6607929468154907, 0.7386196851730347, 0.757709264755249, 0.8017621040344238, 0.8164464235305786, 0.8002936840057373, 0.7606461048126221, 0.8105726838111877, 0.7488986849784851, 0.8428781032562256, 0.8245227336883545, 0.8472834229469299, 0.8781204223632812, 0.8524228930473328, 0.7327459454536438, 0.8575624227523804, 0.8546255230903625, 0.8722466826438904, 0.8450807929039001, 0.8810572624206543, 0.8685756325721741, 0.8751835823059082, 0.8972100019454956, 0.8604992628097534, 0.8693098425865173, 0.9118942618370056, 0.8781204223632812, 0.9133626818656921, 0.9104258418083191, 0.8781204223632812, 0.8942731022834778, 0.8781204223632812, 0.9177680015563965, 0.9089574217796326, 0.8795888423919678, 0.9368575811386108, 0.9001468420028687, 0.9001468420028687, 0.9207048416137695, 0.8032305240631104, 0.8032305240631104, 0.9324522614479065, 0.9530102610588074, 0.9588840007781982, 0.9368575811386108, 0.9559471607208252, 0.9632892608642578, 0.8487518429756165, 0.9515418410301208, 0.961086630821228, 0.9500734210014343, 0.9500734210014343, 0.9133626818656921, 0.9588840007781982, 0.9324522614479065, 0.9427312612533569, 0.9676945805549622, 0.9691630005836487, 0.9522760510444641, 0.9574155807495117, 0.9926578402519226], [0.20998531579971313, 0.46549192070961, 0.5286343693733215, NaN, 0.5345080494880676, 0.596182107925415, 0.6189427375793457, 0.5800293684005737, 0.7298091053962708, 0.7209985256195068, 0.7657856345176697, 0.757709264755249, 0.7944199442863464, 0.7958884239196777, 0.7665198445320129, 0.7687224745750427, 0.8164464235305786, 0.7870778441429138, 0.8281938433647156, 0.8076358437538147, 0.838472843170166, 0.8737151026725769, 0.8443465232849121, 0.7474302649497986, 0.8428781032562256, 0.850220263004303, 0.853157103061676, 0.8348017334938049, 0.8575624227523804, 0.8707782626152039, 0.8560940027236938, 0.8634361028671265, 0.835536003112793, 0.8604992628097534, 0.8942731022834778, 0.880323052406311, 0.8957415819168091, 0.8913362622261047, 0.8634361028671265, 0.8942731022834778, 0.864904522895813, 0.9030836820602417, 0.8788546323776245, 0.8898678421974182, 0.9133626818656921, 0.8810572624206543, 0.8972100019454956, 0.9118942618370056, 0.8164464235305786, 0.8105726838111877, 0.9265785813331604, 0.9397944211959839, 0.9324522614479065, 0.9280470013618469, 0.9486050009727478, 0.9383260011672974, 0.8707782626152039, 0.9471365809440613, 0.9530102610588074, 0.9383260011672974, 0.9500734210014343, 0.9397944211959839, 0.9500734210014343, 0.9441996812820435, 0.9515418410301208, 0.9618208408355713, 0.9794420003890991, 0.9640234708786011, 0.9574155807495117, 0.9941262602806091], [0.23715125024318695, 0.4229075014591217, 0.47723934054374695, 0.46549192070961, NaN, 0.5315712094306946, 0.6358296871185303, 0.6358296871185303, 0.7121879458427429, 0.622613787651062, 0.6651982665061951, 0.6842877864837646, 0.7826725244522095, 0.7900146842002869, 0.7591776847839355, 0.7562408447265625, 0.7753304243087769, 0.6784141063690186, 0.8252570033073425, 0.8193832635879517, 0.8399412631988525, 0.8274596333503723, 0.8487518429756165, 0.7569750547409058, 0.8281938433647156, 0.8450807929039001, 0.8619676828384399, 0.8311306834220886, 0.8634361028671265, 0.8737151026725769, 0.8744493126869202, 0.8546255230903625, 0.8575624227523804, 0.8634361028671265, 0.9030836820602417, 0.8590308427810669, 0.8986784219741821, 0.8964757919311523, 0.877386212348938, 0.8825256824493408, 0.877386212348938, 0.9060205817222595, 0.8575624227523804, 0.8737151026725769, 0.9280470013618469, 0.8604992628097534, 0.907489001750946, 0.91629958152771, 0.7958884239196777, 0.7958884239196777, 0.9133626818656921, 0.9486050009727478, 0.9500734210014343, 0.9251101613044739, 0.9530102610588074, 0.9544786810874939, 0.8370044231414795, 0.933920681476593, 0.9530102610588074, 0.9427312612533569, 0.9486050009727478, 0.9251101613044739, 0.9559471607208252, 0.9368575811386108, 0.9280470013618469, 0.9588840007781982, 0.9691630005836487, 0.9449338912963867, 0.9383260011672974, 0.9897210001945496], [0.20558002591133118, 0.392070472240448, 0.40234947204589844, 0.40381792187690735, 0.4684287905693054, NaN, 0.5814977884292603, 0.5234948396682739, 0.691629946231842, 0.6284875273704529, 0.6593245267868042, 0.676945686340332, 0.769456684589386, 0.784140944480896, 0.757709264755249, 0.7312775254249573, 0.7767988443374634, 0.700440526008606, 0.8223201036453247, 0.8135095238685608, 0.8458150029182434, 0.8693098425865173, 0.8472834229469299, 0.7562408447265625, 0.826725423336029, 0.8575624227523804, 0.8751835823059082, 0.838472843170166, 0.8560940027236938, 0.8913362622261047, 0.8810572624206543, 0.8957415819168091, 0.8524228930473328, 0.850220263004303, 0.8957415819168091, 0.8810572624206543, 0.9207048416137695, 0.919236421585083, 0.8744493126869202, 0.8913362622261047, 0.864904522895813, 0.9236416816711426, 0.8928046822547913, 0.8693098425865173, 0.9280470013618469, 0.8876652121543884, 0.8986784219741821, 0.9353891611099243, 0.8046989440917969, 0.8091042637825012, 0.9236416816711426, 0.9574155807495117, 0.9662261605262756, 0.9397944211959839, 0.9486050009727478, 0.9559471607208252, 0.8428781032562256, 0.9486050009727478, 0.9456681609153748, 0.9427312612533569, 0.9530102610588074, 0.9148311018943787, 0.9515418410301208, 0.9324522614479065, 0.9456681609153748, 0.9735682606697083, 0.9676945805549622, 0.9530102610588074, 0.9530102610588074, 0.9911894202232361], [0.1938326060771942, 0.3348017632961273, 0.39500734210014343, 0.3810572624206543, 0.3641703426837921, 0.41850221157073975, NaN, 0.4508076310157776, 0.6255506873130798, 0.5844346284866333, 0.5932452082633972, 0.6262848973274231, 0.7195301055908203, 0.7342143654823303, 0.7092511057853699, 0.7283406853675842, 0.7430249452590942, 0.6637297868728638, 0.7753304243087769, 0.7782672643661499, 0.8076358437538147, 0.7826725244522095, 0.8223201036453247, 0.7298091053962708, 0.8046989440917969, 0.8237885236740112, 0.8296622633934021, 0.8325991034507751, 0.8164464235305786, 0.8560940027236938, 0.853157103061676, 0.8105726838111877, 0.8516886830329895, 0.8590308427810669, 0.8883994221687317, 0.853157103061676, 0.8876652121543884, 0.8766520023345947, 0.8795888423919678, 0.8707782626152039, 0.8928046822547913, 0.8825256824493408, 0.8546255230903625, 0.8869310021400452, 0.9118942618370056, 0.8663730025291443, 0.9133626818656921, 0.9104258418083191, 0.7973568439483643, 0.796622633934021, 0.8854625821113586, 0.9368575811386108, 0.9353891611099243, 0.9207048416137695, 0.9383260011672974, 0.9720998406410217, 0.8458150029182434, 0.933920681476593, 0.9456681609153748, 0.9500734210014343, 0.9324522614479065, 0.9265785813331604, 0.9588840007781982, 0.9324522614479065, 0.9030836820602417, 0.9486050009727478, 0.9750367403030396, 0.9647576808929443, 0.9368575811386108, 0.9823788404464722], [0.22613802552223206, 0.3641703426837921, 0.4229075014591217, 0.41997063159942627, 0.3641703426837921, 0.4765051305294037, 0.5491923689842224, NaN, 0.6622613668441772, 0.5917767882347107, 0.6182085275650024, 0.646842896938324, 0.7298091053962708, 0.7503671050071716, 0.7224669456481934, 0.7151248455047607, 0.7393538951873779, 0.6578561067581177, 0.7738619446754456, 0.7980910539627075, 0.7958884239196777, 0.7826725244522095, 0.8017621040344238, 0.7254038453102112, 0.799559473991394, 0.8135095238685608, 0.8281938433647156, 0.8135095238685608, 0.8105726838111877, 0.826725423336029, 0.8399412631988525, 0.8274596333503723, 0.8399412631988525, 0.8487518429756165, 0.8825256824493408, 0.835536003112793, 0.8810572624206543, 0.8568282127380371, 0.8590308427810669, 0.8707782626152039, 0.8604992628097534, 0.8737151026725769, 0.8524228930473328, 0.8604992628097534, 0.9104258418083191, 0.8590308427810669, 0.9001468420028687, 0.8942731022834778, 0.7679882645606995, 0.772393524646759, 0.8869310021400452, 0.9397944211959839, 0.9368575811386108, 0.9030836820602417, 0.9302496314048767, 0.9588840007781982, 0.8281938433647156, 0.9251101613044739, 0.9412628412246704, 0.9397944211959839, 0.9295154213905334, 0.9148311018943787, 0.9486050009727478, 0.9265785813331604, 0.91629958152771, 0.9486050009727478, 0.9691630005836487, 0.9508076310157776, 0.9412628412246704, 0.9838472604751587], [0.13803230226039886, 0.2584434747695923, 0.24596181511878967, 0.27019089460372925, 0.2878120541572571, 0.30837005376815796, 0.37444934248924255, 0.33773863315582275, NaN, 0.47870779037475586, 0.47356829047203064, 0.4985315799713135, 0.577826738357544, 0.619676947593689, 0.6791483163833618, 0.5315712094306946, 0.5638766288757324, 0.5565345287322998, 0.7224669456481934, 0.6901615262031555, 0.7591776847839355, 0.7606461048126221, 0.7606461048126221, 0.6211453676223755, 0.7488986849784851, 0.7687224745750427, 0.826725423336029, 0.7430249452590942, 0.7312775254249573, 0.7885462641716003, 0.838472843170166, 0.7371512651443481, 0.7914831042289734, 0.7929515242576599, 0.7709251046180725, 0.7753304243087769, 0.8414096832275391, 0.8193832635879517, 0.8046989440917969, 0.8817914724349976, 0.8201174736022949, 0.8259912133216858, 0.8311306834220886, 0.8274596333503723, 0.8208516836166382, 0.8450807929039001, 0.8362702131271362, 0.8487518429756165, 0.7386196851730347, 0.7430249452590942, 0.9104258418083191, 0.922173261642456, 0.9353891611099243, 0.9353891611099243, 0.9353891611099243, 0.9251101613044739, 0.8046989440917969, 0.9265785813331604, 0.9427312612533569, 0.919236421585083, 0.9530102610588074, 0.8869310021400452, 0.9471365809440613, 0.8883994221687317, 0.9118942618370056, 0.9676945805549622, 0.9706314206123352, 0.9045521020889282, 0.9353891611099243, 0.9926578402519226], [0.13656388223171234, 0.2540381848812103, 0.3392070531845093, 0.27900147438049316, 0.3773861825466156, 0.3715124726295471, 0.4155653417110443, 0.4082232117652893, 0.5212922096252441, NaN, 0.5242290496826172, 0.5653451085090637, 0.5932452082633972, 0.5851688981056213, 0.592510998249054, 0.6358296871185303, 0.6879588961601257, 0.6850220561027527, 0.6578561067581177, 0.6387665271759033, 0.691629946231842, 0.7298091053962708, 0.688693106174469, 0.6387665271759033, 0.700440526008606, 0.6930983662605286, 0.7209985256195068, 0.7356828451156616, 0.7312775254249573, 0.7239353656768799, 0.7327459454536438, 0.7312775254249573, 0.7180616855621338, 0.7327459454536438, 0.7782672643661499, 0.8032305240631104, 0.8245227336883545, 0.7885462641716003, 0.7657856345176697, 0.8061674237251282, 0.8237885236740112, 0.8032305240631104, 0.8494860529899597, 0.7973568439483643, 0.8296622633934021, 0.8693098425865173, 0.8693098425865173, 0.8340675234794617, 0.7327459454536438, 0.7356828451156616, 0.8883994221687317, 0.8869310021400452, 0.8928046822547913, 0.8913362622261047, 0.9001468420028687, 0.9118942618370056, 0.8149779438972473, 0.9001468420028687, 0.9207048416137695, 0.8986784219741821, 0.9449338912963867, 0.8825256824493408, 0.9427312612533569, 0.8898678421974182, 0.9295154213905334, 0.9603524208068848, 0.9632892608642578, 0.9486050009727478, 0.9486050009727478, 0.988252580165863], [0.11600587517023087, 0.2364170402288437, 0.26138031482696533, 0.2342143952846527, 0.3348017632961273, 0.3406754732131958, 0.4067547619342804, 0.38179147243499756, 0.5264317393302917, 0.4757709205150604, NaN, 0.6093979477882385, 0.5932452082633972, 0.580763578414917, 0.6057268977165222, 0.5734214186668396, 0.6306901574134827, 0.6549192070960999, 0.6578561067581177, 0.6262848973274231, 0.6710719466209412, 0.7386196851730347, 0.6828193664550781, 0.6270191073417664, 0.6982378959655762, 0.688693106174469, 0.7129221558570862, 0.718795895576477, 0.7239353656768799, 0.7327459454536438, 0.7400881052017212, 0.7591776847839355, 0.7070484757423401, 0.7232011556625366, 0.7555066347122192, 0.7672540545463562, 0.8208516836166382, 0.7826725244522095, 0.742290735244751, 0.7826725244522095, 0.7753304243087769, 0.7988252639770508, 0.8237885236740112, 0.757709264755249, 0.8091042637825012, 0.8325991034507751, 0.8164464235305786, 0.8252570033073425, 0.7209985256195068, 0.7224669456481934, 0.8883994221687317, 0.8825256824493408, 0.8825256824493408, 0.9001468420028687, 0.8972100019454956, 0.9030836820602417, 0.8032305240631104, 0.9111600518226624, 0.9082232117652893, 0.8942731022834778, 0.9427312612533569, 0.8414096832275391, 0.9265785813331604, 0.8575624227523804, 0.9214390516281128, 0.9486050009727478, 0.9486050009727478, 0.9001468420028687, 0.9324522614479065, 0.9926578402519226], [0.11013215780258179, 0.2143906056880951, 0.24229075014591217, 0.24229075014591217, 0.31571218371391296, 0.32305434346199036, 0.3737151324748993, 0.3531571328639984, 0.5014684200286865, 0.43465492129325867, 0.3906020522117615, NaN, 0.5829662084579468, 0.5594713687896729, 0.5638766288757324, 0.5550661087036133, 0.596182107925415, 0.6204111576080322, 0.6284875273704529, 0.6123347878456116, 0.6417033672332764, 0.6828193664550781, 0.6527165770530701, 0.6049926280975342, 0.6563876867294312, 0.6637297868728638, 0.6997063159942627, 0.6784141063690186, 0.6960352659225464, 0.6975036859512329, 0.7077826857566833, 0.727606475353241, 0.691629946231842, 0.703377366065979, 0.7459618449211121, 0.7459618449211121, 0.7856094241142273, 0.7591776847839355, 0.7268722653388977, 0.7591776847839355, 0.7533039450645447, 0.7709251046180725, 0.7856094241142273, 0.7474302649497986, 0.781204104423523, 0.7973568439483643, 0.7973568439483643, 0.8149779438972473, 0.7114537358283997, 0.7092511057853699, 0.8414096832275391, 0.8575624227523804, 0.8443465232849121, 0.8619676828384399, 0.8795888423919678, 0.8898678421974182, 0.8002936840057373, 0.8854625821113586, 0.8883994221687317, 0.8751835823059082, 0.91629958152771, 0.8333333134651184, 0.91629958152771, 0.8443465232849121, 0.919236421585083, 0.9295154213905334, 0.9471365809440613, 0.8972100019454956, 0.9251101613044739, 0.9897210001945496], [0.08370044082403183, 0.2290748953819275, 0.19823788106441498, 0.20558002591133118, 0.21732746064662933, 0.230543315410614, 0.2804698944091797, 0.27019089460372925, 0.42217326164245605, 0.4067547619342804, 0.4067547619342804, 0.41703376173973083, NaN, 0.5227606296539307, 0.5315712094306946, 0.48458150029182434, 0.5293685793876648, 0.5022026300430298, 0.6901615262031555, 0.5947136282920837, 0.6593245267868042, 0.6637297868728638, 0.6798825263977051, 0.553597629070282, 0.646842896938324, 0.7298091053962708, 0.7415565252304077, 0.691629946231842, 0.6725403666496277, 0.7386196851730347, 0.7386196851730347, 0.6930983662605286, 0.7063142657279968, 0.7342143654823303, 0.7944199442863464, 0.7503671050071716, 0.7900146842002869, 0.7863436341285706, 0.757709264755249, 0.7621145248413086, 0.7944199442863464, 0.7958884239196777, 0.7415565252304077, 0.8002936840057373, 0.8325991034507751, 0.7679882645606995, 0.838472843170166, 0.835536003112793, 0.7327459454536438, 0.7356828451156616, 0.8296622633934021, 0.9148311018943787, 0.9030836820602417, 0.8612334728240967, 0.9265785813331604, 0.907489001750946, 0.7988252639770508, 0.8854625821113586, 0.9295154213905334, 0.9001468420028687, 0.9177680015563965, 0.8832598924636841, 0.9383260011672974, 0.903817892074585, 0.8869310021400452, 0.9295154213905334, 0.9515418410301208, 0.9295154213905334, 0.9251101613044739, 0.988252580165863], [0.11453744769096375, 0.2143906056880951, 0.18355359137058258, 0.20411160588264465, 0.20998531579971313, 0.21585902571678162, 0.2657856047153473, 0.24963289499282837, 0.38032305240631104, 0.41483113169670105, 0.419236421585083, 0.44052863121032715, 0.47723934054374695, NaN, 0.4948605000972748, 0.46989721059799194, 0.4963289201259613, 0.5124816298484802, 0.6049926280975342, 0.7019089460372925, 0.619676947593689, 0.6189427375793457, 0.6417033672332764, 0.5756241083145142, 0.6035242080688477, 0.649779736995697, 0.7048457860946655, 0.6328927874565125, 0.6453744769096375, 0.7180616855621338, 0.7356828451156616, 0.6740087866783142, 0.8377386331558228, 0.7371512651443481, 0.7474302649497986, 0.7180616855621338, 0.7342143654823303, 0.7621145248413086, 0.7665198445320129, 0.7907488942146301, 0.7547723650932312, 0.7958884239196777, 0.7136563658714294, 0.7790014743804932, 0.8201174736022949, 0.7165932655334473, 0.8091042637825012, 0.8458150029182434, 0.7283406853675842, 0.7312775254249573, 0.7797356843948364, 0.8729808926582336, 0.8869310021400452, 0.8208516836166382, 0.8795888423919678, 0.9104258418083191, 0.7958884239196777, 0.853157103061676, 0.9001468420028687, 0.9089574217796326, 0.8707782626152039, 0.8928046822547913, 0.9148311018943787, 0.9052863717079163, 0.8766520023345947, 0.9060205817222595, 0.9603524208068848, 0.9251101613044739, 0.9133626818656921, 0.9706314206123352], [0.12775330245494843, 0.21879589557647705, 0.1997063159942627, 0.23348017036914825, 0.24082231521606445, 0.24229075014591217, 0.2907488942146301, 0.27753305435180664, 0.3208516836166382, 0.40748897194862366, 0.39427313208580017, 0.4361233413219452, 0.4684287905693054, 0.5051395297050476, NaN, 0.4706314206123352, 0.5007342100143433, 0.5051395297050476, 0.6049926280975342, 0.5976505279541016, 0.6284875273704529, 0.6116005778312683, 0.6372981071472168, 0.5580029487609863, 0.6049926280975342, 0.6505139470100403, 0.7077826857566833, 0.646842896938324, 0.6461086869239807, 0.6872246861457825, 0.7246695756912231, 0.6637297868728638, 0.7151248455047607, 0.7151248455047607, 0.7254038453102112, 0.7239353656768799, 0.7459618449211121, 0.7209985256195068, 0.7503671050071716, 0.7797356843948364, 0.7444933652877808, 0.7518355250358582, 0.7518355250358582, 0.7474302649497986, 0.7914831042289734, 0.769456684589386, 0.8002936840057373, 0.7870778441429138, 0.7224669456481934, 0.7239353656768799, 0.835536003112793, 0.8590308427810669, 0.8693098425865173, 0.8590308427810669, 0.8913362622261047, 0.8957415819168091, 0.7856094241142273, 0.8678414225578308, 0.8957415819168091, 0.8869310021400452, 0.8751835823059082, 0.8414096832275391, 0.9229074716567993, 0.8480176329612732, 0.8722466826438904, 0.91629958152771, 0.9544786810874939, 0.8795888423919678, 0.9001468420028687, 0.9794420003890991], [0.09324523061513901, 0.23201175034046173, 0.23935389518737793, 0.23127754032611847, 0.2437591850757599, 0.2687224745750427, 0.27165931463241577, 0.28487518429756165, 0.4684287905693054, 0.3641703426837921, 0.426578551530838, 0.4449339210987091, 0.515418529510498, 0.5301027894020081, 0.5293685793876648, NaN, 0.49926578998565674, 0.4669603407382965, 0.5976505279541016, 0.5572687387466431, 0.5888399481773376, 0.6292217373847961, 0.5917767882347107, 0.5359765291213989, 0.5947136282920837, 0.619676947593689, 0.6358296871185303, 0.6013215780258179, 0.6725403666496277, 0.6710719466209412, 0.6519823670387268, 0.6593245267868042, 0.6182085275650024, 0.6975036859512329, 0.6872246861457825, 0.7092511057853699, 0.6828193664550781, 0.7342143654823303, 0.7298091053962708, 0.6989721059799194, 0.6997063159942627, 0.7503671050071716, 0.7415565252304077, 0.7342143654823303, 0.7327459454536438, 0.7415565252304077, 0.7628487348556519, 0.7665198445320129, 0.7378854751586914, 0.7327459454536438, 0.8002936840057373, 0.8105726838111877, 0.8208516836166382, 0.8179148435592651, 0.8237885236740112, 0.8193832635879517, 0.7834067344665527, 0.8634361028671265, 0.8325991034507751, 0.8590308427810669, 0.8370044231414795, 0.8751835823059082, 0.8590308427810669, 0.8913362622261047, 0.8619676828384399, 0.8781204223632812, 0.8869310021400452, 0.9148311018943787, 0.8942731022834778, 0.9765051603317261], [0.13069015741348267, 0.13656388223171234, 0.18942731618881226, 0.18355359137058258, 0.22466960549354553, 0.22320117056369781, 0.2569750249385834, 0.26064610481262207, 0.4361233413219452, 0.31204110383987427, 0.36930984258651733, 0.40381792187690735, 0.4706314206123352, 0.5036710500717163, 0.49926578998565674, 0.5007342100143433, NaN, 0.4831130802631378, 0.5139500498771667, 0.5491923689842224, 0.5286343693733215, 0.5638766288757324, 0.5616739988327026, 0.5653451085090637, 0.553597629070282, 0.5491923689842224, 0.5829662084579468, 0.5991189479827881, 0.6255506873130798, 0.5976505279541016, 0.6255506873130798, 0.610866367816925, 0.6431717872619629, 0.6372981071472168, 0.703377366065979, 0.700440526008606, 0.7092511057853699, 0.6505139470100403, 0.6725403666496277, 0.7474302649497986, 0.7709251046180725, 0.6828193664550781, 0.6975036859512329, 0.6945667862892151, 0.7547723650932312, 0.7151248455047607, 0.7944199442863464, 0.7635829448699951, 0.7415565252304077, 0.7430249452590942, 0.7503671050071716, 0.8215858936309814, 0.7914831042289734, 0.7944199442863464, 0.8663730025291443, 0.8795888423919678, 0.8120411038398743, 0.8414096832275391, 0.8810572624206543, 0.8590308427810669, 0.8575624227523804, 0.8281938433647156, 0.91629958152771, 0.8472834229469299, 0.8883994221687317, 0.9001468420028687, 0.9324522614479065, 0.9258443713188171, 0.9486050009727478, 0.9926578402519226], [0.10866373032331467, 0.18208517134189606, 0.2511013150215149, 0.21292217075824738, 0.32158589363098145, 0.29955947399139404, 0.33627018332481384, 0.3421438932418823, 0.4434655010700226, 0.3149779736995697, 0.34508076310157776, 0.3795888423919678, 0.4977973699569702, 0.4875183701515198, 0.4948605000972748, 0.5330396294593811, 0.5168869495391846, NaN, 0.5330396294593811, 0.5234948396682739, 0.5712187886238098, 0.596182107925415, 0.5829662084579468, 0.5770925283432007, 0.5903083682060242, 0.5697503685951233, 0.6167401075363159, 0.6020557880401611, 0.6328927874565125, 0.596182107925415, 0.6372981071472168, 0.6372981071472168, 0.6328927874565125, 0.6035242080688477, 0.6622613668441772, 0.688693106174469, 0.6945667862892151, 0.6798825263977051, 0.6277533173561096, 0.6541849970817566, 0.688693106174469, 0.6842877864837646, 0.7518355250358582, 0.6798825263977051, 0.7121879458427429, 0.772393524646759, 0.7430249452590942, 0.7400881052017212, 0.6710719466209412, 0.6725403666496277, 0.7782672643661499, 0.7914831042289734, 0.7650514245033264, 0.8179148435592651, 0.8370044231414795, 0.8296622633934021, 0.781204104423523, 0.8443465232849121, 0.8604992628097534, 0.850220263004303, 0.9060205817222595, 0.7738619446754456, 0.9001468420028687, 0.7973568439483643, 0.9016152620315552, 0.8942731022834778, 0.9089574217796326, 0.8854625821113586, 0.9236416816711426, 0.9911894202232361], [0.08223201334476471, 0.16005873680114746, 0.15712188184261322, 0.17180617153644562, 0.17474302649497986, 0.1776798814535141, 0.22466960549354553, 0.22613802552223206, 0.27753305435180664, 0.3421438932418823, 0.3421438932418823, 0.3715124726295471, 0.3098384737968445, 0.39500734210014343, 0.39500734210014343, 0.40234947204589844, 0.48604992032051086, 0.4669603407382965, NaN, 0.47503671050071716, 0.5256975293159485, 0.5249632596969604, 0.5513949990272522, 0.4875183701515198, 0.5359765291213989, 0.6218795776367188, 0.6784141063690186, 0.5947136282920837, 0.5697503685951233, 0.6894273161888123, 0.7048457860946655, 0.5712187886238098, 0.6556534767150879, 0.6762114763259888, 0.6167401075363159, 0.6622613668441772, 0.6607929468154907, 0.7378854751586914, 0.7151248455047607, 0.691629946231842, 0.688693106174469, 0.7679882645606995, 0.7099853157997131, 0.7856094241142273, 0.7063142657279968, 0.7136563658714294, 0.7606461048126221, 0.8487518429756165, 0.7129221558570862, 0.7136563658714294, 0.8149779438972473, 0.9515418410301208, 0.8986784219741821, 0.8604992628097534, 0.9001468420028687, 0.8281938433647156, 0.7944199442863464, 0.8179148435592651, 0.9177680015563965, 0.9060205817222595, 0.9030836820602417, 0.8604992628097534, 0.8766520023345947, 0.8766520023345947, 0.8810572624206543, 0.9383260011672974, 0.9515418410301208, 0.9236416816711426, 0.9214390516281128, 0.9838472604751587], [0.08810572326183319, 0.17621144652366638, 0.17547723650932312, 0.1923641711473465, 0.18061673641204834, 0.18649046123027802, 0.2217327505350113, 0.20190896093845367, 0.3098384737968445, 0.3612334728240967, 0.3737151324748993, 0.3876652121543884, 0.40528634190559387, 0.2980910539627075, 0.40234947204589844, 0.4427312910556793, 0.4508076310157776, 0.4765051305294037, 0.5249632596969604, NaN, 0.5521292090415955, 0.5609397888183594, 0.5682819485664368, 0.4948605000972748, 0.5345080494880676, 0.5565345287322998, 0.6042584180831909, 0.5726872086524963, 0.596182107925415, 0.6372981071472168, 0.6240822076797485, 0.619676947593689, 0.6872246861457825, 0.6446402072906494, 0.6930983662605286, 0.6666666865348816, 0.6563876867294312, 0.700440526008606, 0.6813509464263916, 0.7298091053962708, 0.6953010559082031, 0.6930983662605286, 0.6651982665061951, 0.7209985256195068, 0.7584434747695923, 0.6740087866783142, 0.7679882645606995, 0.7687224745750427, 0.7077826857566833, 0.7129221558570862, 0.7444933652877808, 0.7944199442863464, 0.8281938433647156, 0.7650514245033264, 0.8546255230903625, 0.8883994221687317, 0.772393524646759, 0.8061674237251282, 0.864904522895813, 0.8942731022834778, 0.8164464235305786, 0.8722466826438904, 0.864904522895813, 0.8810572624206543, 0.8487518429756165, 0.8707782626152039, 0.9383260011672974, 0.9104258418083191, 0.8898678421974182, 0.9618208408355713], [0.07195301353931427, 0.14537444710731506, 0.15271659195423126, 0.16152717173099518, 0.16005873680114746, 0.15418502688407898, 0.1923641711473465, 0.20411160588264465, 0.24082231521606445, 0.30837005376815796, 0.32892805337905884, 0.35829663276672363, 0.3406754732131958, 0.38032305240631104, 0.3715124726295471, 0.41116005182266235, 0.47136563062667847, 0.4287812113761902, 0.4743025004863739, 0.44787079095840454, NaN, 0.5080763697624207, 0.5910425782203674, 0.5139500498771667, 0.5051395297050476, 0.5323054194450378, 0.6145374178886414, 0.5859031081199646, 0.5682819485664368, 0.6446402072906494, 0.6365638971328735, 0.5947136282920837, 0.6005873680114746, 0.6182085275650024, 0.6284875273704529, 0.6446402072906494, 0.6842877864837646, 0.685756266117096, 0.6718061566352844, 0.6710719466209412, 0.6490455269813538, 0.7077826857566833, 0.7136563658714294, 0.7151248455047607, 0.7121879458427429, 0.7408223152160645, 0.7180616855621338, 0.7650514245033264, 0.7107195258140564, 0.7063142657279968, 0.826725423336029, 0.8781204223632812, 0.8898678421974182, 0.8700440526008606, 0.8546255230903625, 0.835536003112793, 0.7738619446754456, 0.8414096832275391, 0.8737151026725769, 0.8729808926582336, 0.8839941024780273, 0.8325991034507751, 0.8560940027236938, 0.8458150029182434, 0.8810572624206543, 0.9324522614479065, 0.919236421585083, 0.8751835823059082, 0.9133626818656921, 0.98531574010849], [0.07048457860946655, 0.12041115760803223, 0.12187959253787994, 0.1262848824262619, 0.17254038155078888, 0.13069015741348267, 0.21732746064662933, 0.21732746064662933, 0.23935389518737793, 0.27019089460372925, 0.26138031482696533, 0.3171806037425995, 0.33627018332481384, 0.3810572624206543, 0.3883994221687317, 0.37077826261520386, 0.4361233413219452, 0.40381792187690735, 0.47503671050071716, 0.4390602111816406, 0.49192363023757935, NaN, 0.5227606296539307, 0.4875183701515198, 0.5337738394737244, 0.538913369178772, 0.5734214186668396, 0.5954478979110718, 0.5873715281486511, 0.5726872086524963, 0.607929527759552, 0.5844346284866333, 0.527165949344635, 0.6035242080688477, 0.622613787651062, 0.6798825263977051, 0.7334801554679871, 0.6828193664550781, 0.6475771069526672, 0.685756266117096, 0.688693106174469, 0.6901615262031555, 0.7041115760803223, 0.6798825263977051, 0.703377366065979, 0.7312775254249573, 0.7496328949928284, 0.7378854751586914, 0.6681351065635681, 0.6688693165779114, 0.8105726838111877, 0.8619676828384399, 0.8472834229469299, 0.8428781032562256, 0.8516886830329895, 0.8443465232849121, 0.769456684589386, 0.8781204223632812, 0.8795888423919678, 0.8619676828384399, 0.8986784219741821, 0.7797356843948364, 0.8781204223632812, 0.8091042637825012, 0.8560940027236938, 0.91629958152771, 0.919236421585083, 0.8560940027236938, 0.892070472240448, 0.9809104204177856], [0.06314244121313095, 0.13656388223171234, 0.14757709205150604, 0.1556534469127655, 0.15124815702438354, 0.15271659195423126, 0.1776798814535141, 0.19823788106441498, 0.23935389518737793, 0.311306893825531, 0.3171806037425995, 0.34728339314460754, 0.3201174736022949, 0.35829663276672363, 0.3627019226551056, 0.4082232117652893, 0.43832600116729736, 0.41703376173973083, 0.4486050009727478, 0.43171805143356323, 0.40895742177963257, 0.47723934054374695, NaN, 0.5022026300430298, 0.48091042041778564, 0.5110132098197937, 0.5917767882347107, 0.5653451085090637, 0.5491923689842224, 0.6116005778312683, 0.6240822076797485, 0.5756241083145142, 0.5822319984436035, 0.6093979477882385, 0.622613787651062, 0.6270191073417664, 0.6490455269813538, 0.6365638971328735, 0.6563876867294312, 0.6549192070960999, 0.6328927874565125, 0.6776798963546753, 0.6945667862892151, 0.7107195258140564, 0.6982378959655762, 0.7063142657279968, 0.7085168957710266, 0.7533039450645447, 0.7121879458427429, 0.7107195258140564, 0.8105726838111877, 0.8516886830329895, 0.8854625821113586, 0.8487518429756165, 0.8428781032562256, 0.8179148435592651, 0.7753304243087769, 0.8296622633934021, 0.864904522895813, 0.8604992628097534, 0.8693098425865173, 0.8281938433647156, 0.8458150029182434, 0.8392070531845093, 0.8707782626152039, 0.922173261642456, 0.9236416816711426, 0.8693098425865173, 0.91629958152771, 0.9838472604751587], [0.14243759214878082, 0.23348017036914825, 0.2672540247440338, 0.2525697648525238, 0.24302496016025543, 0.2437591850757599, 0.27019089460372925, 0.2745961844921112, 0.3788546323776245, 0.3612334728240967, 0.37298092246055603, 0.39500734210014343, 0.44640234112739563, 0.4243759214878082, 0.44199705123901367, 0.46402350068092346, 0.43465492129325867, 0.4229075014591217, 0.5124816298484802, 0.5051395297050476, 0.48604992032051086, 0.5124816298484802, 0.4977973699569702, NaN, 0.5345080494880676, 0.5403817892074585, 0.5624082088470459, 0.553597629070282, 0.5741556286811829, 0.553597629070282, 0.5697503685951233, 0.5903083682060242, 0.5550661087036133, 0.6005873680114746, 0.6651982665061951, 0.634361207485199, 0.6013215780258179, 0.5976505279541016, 0.634361207485199, 0.6725403666496277, 0.6975036859512329, 0.6167401075363159, 0.6681351065635681, 0.6629955768585205, 0.691629946231842, 0.6681351065635681, 0.7327459454536438, 0.6681351065635681, 0.6387665271759033, 0.6299559473991394, 0.7452276349067688, 0.7312775254249573, 0.7283406853675842, 0.7540381550788879, 0.7856094241142273, 0.7856094241142273, 0.7606461048126221, 0.7900146842002869, 0.7922173142433167, 0.7900146842002869, 0.7944199442863464, 0.8149779438972473, 0.8237885236740112, 0.8179148435592651, 0.7767988443374634, 0.8208516836166382, 0.8458150029182434, 0.838472843170166, 0.8663730025291443, 0.9823788404464722], [0.08076358586549759, 0.13656388223171234, 0.14243759214878082, 0.15712188184261322, 0.17180617153644562, 0.17327459156513214, 0.19530102610588074, 0.20044052600860596, 0.2511013150215149, 0.29955947399139404, 0.30176210403442383, 0.34361234307289124, 0.3531571328639984, 0.39647576212882996, 0.39500734210014343, 0.40528634190559387, 0.44640234112739563, 0.40969163179397583, 0.46402350068092346, 0.46549192070961, 0.4948605000972748, 0.46622613072395325, 0.5190895795822144, 0.46549192070961, NaN, 0.5220264196395874, 0.5484581589698792, 0.6145374178886414, 0.5337738394737244, 0.5866373181343079, 0.5770925283432007, 0.5682819485664368, 0.5462555289268494, 0.6020557880401611, 0.5903083682060242, 0.619676947593689, 0.6439059972763062, 0.6740087866783142, 0.6549192070960999, 0.6182085275650024, 0.6321585774421692, 0.688693106174469, 0.6593245267868042, 0.6784141063690186, 0.6681351065635681, 0.7063142657279968, 0.7019089460372925, 0.7121879458427429, 0.6791483163833618, 0.676945686340332, 0.7679882645606995, 0.7870778441429138, 0.8061674237251282, 0.8046989440917969, 0.8120411038398743, 0.8487518429756165, 0.7518355250358582, 0.9397944211959839, 0.8392070531845093, 0.8325991034507751, 0.9251101613044739, 0.8443465232849121, 0.9045521020889282, 0.8560940027236938, 0.892070472240448, 0.9456681609153748, 0.919236421585083, 0.8847283124923706, 0.919236421585083, 0.9867841601371765], [0.08223201334476471, 0.13950073719024658, 0.14537444710731506, 0.14977973699569702, 0.15491923689842224, 0.14243759214878082, 0.17621144652366638, 0.18649046123027802, 0.23127754032611847, 0.30690160393714905, 0.311306893825531, 0.33627018332481384, 0.27019089460372925, 0.350220263004303, 0.3494860529899597, 0.38032305240631104, 0.4508076310157776, 0.4302496314048767, 0.37812042236328125, 0.4434655010700226, 0.46769458055496216, 0.461086630821228, 0.4889867901802063, 0.4596182107925415, 0.4779735803604126, NaN, 0.5602055788040161, 0.5359765291213989, 0.5183553695678711, 0.6292217373847961, 0.5932452082633972, 0.5286343693733215, 0.6064611077308655, 0.6402349472045898, 0.5976505279541016, 0.6240822076797485, 0.6020557880401611, 0.6563876867294312, 0.6930983662605286, 0.664463996887207, 0.649779736995697, 0.6828193664550781, 0.6718061566352844, 0.7312775254249573, 0.6791483163833618, 0.6894273161888123, 0.7180616855621338, 0.7856094241142273, 0.700440526008606, 0.7019089460372925, 0.7606461048126221, 0.903817892074585, 0.8671072125434875, 0.8149779438972473, 0.8612334728240967, 0.8076358437538147, 0.7767988443374634, 0.811306893825531, 0.8825256824493408, 0.8935388922691345, 0.8737151026725769, 0.8516886830329895, 0.8663730025291443, 0.8685756325721741, 0.8634361028671265, 0.9118942618370056, 0.9456681609153748, 0.9118942618370056, 0.9118942618370056, 0.9779735803604126], [0.06167401000857353, 0.13436123728752136, 0.12775330245494843, 0.14684288203716278, 0.13803230226039886, 0.12481644749641418, 0.1703377366065979, 0.17180617153644562, 0.17327459156513214, 0.27900147438049316, 0.28707781434059143, 0.3002936840057373, 0.2584434747695923, 0.2951541841030121, 0.29221731424331665, 0.3641703426837921, 0.41703376173973083, 0.38325992226600647, 0.32158589363098145, 0.3957415521144867, 0.38546255230903625, 0.426578551530838, 0.4082232117652893, 0.4375917911529541, 0.45154184103012085, 0.4397944211959839, NaN, 0.5345080494880676, 0.5095447897911072, 0.5484581589698792, 0.5748898386955261, 0.49926578998565674, 0.5491923689842224, 0.5756241083145142, 0.5748898386955261, 0.6064611077308655, 0.5991189479827881, 0.6167401075363159, 0.6358296871185303, 0.6240822076797485, 0.6284875273704529, 0.6519823670387268, 0.6387665271759033, 0.6828193664550781, 0.6490455269813538, 0.6828193664550781, 0.6813509464263916, 0.7129221558570862, 0.6872246861457825, 0.6828193664550781, 0.7951542139053345, 0.8986784219741821, 0.9030836820602417, 0.823054313659668, 0.8179148435592651, 0.7885462641716003, 0.7503671050071716, 0.8105726838111877, 0.8406754732131958, 0.8296622633934021, 0.8560940027236938, 0.8208516836166382, 0.8281938433647156, 0.8311306834220886, 0.8414096832275391, 0.922173261642456, 0.907489001750946, 0.8678414225578308, 0.8979442119598389, 0.9750367403030396], [0.08957415819168091, 0.13656388223171234, 0.15491923689842224, 0.16519823670387268, 0.16886931657791138, 0.16152717173099518, 0.16740088164806366, 0.18649046123027802, 0.2569750249385834, 0.26431718468666077, 0.28120410442352295, 0.32158589363098145, 0.30837005376815796, 0.36710718274116516, 0.3531571328639984, 0.39867842197418213, 0.4008810520172119, 0.39794421195983887, 0.40528634190559387, 0.4273127615451813, 0.4140969216823578, 0.4045521318912506, 0.43465492129325867, 0.44640234112739563, 0.38546255230903625, 0.46402350068092346, 0.46549192070961, NaN, 0.47723934054374695, 0.4904552102088928, 0.49559471011161804, 0.4743025004863739, 0.5095447897911072, 0.5675477385520935, 0.5741556286811829, 0.596182107925415, 0.5873715281486511, 0.6013215780258179, 0.6123347878456116, 0.5756241083145142, 0.6013215780258179, 0.6475771069526672, 0.6240822076797485, 0.6446402072906494, 0.6372981071472168, 0.6651982665061951, 0.6710719466209412, 0.688693106174469, 0.6828193664550781, 0.6842877864837646, 0.6960352659225464, 0.7268722653388977, 0.7209985256195068, 0.7386196851730347, 0.7731277346611023, 0.8414096832275391, 0.7503671050071716, 0.8898678421974182, 0.8157122135162354, 0.8458150029182434, 0.853157103061676, 0.8414096832275391, 0.8810572624206543, 0.8575624227523804, 0.8751835823059082, 0.8957415819168091, 0.9045521020889282, 0.9016152620315552, 0.9133626818656921, 0.9838472604751587], [0.07488986849784851, 0.13215859234333038, 0.1189427301287651, 0.14243759214878082, 0.13656388223171234, 0.14390602707862854, 0.18355359137058258, 0.18942731618881226, 0.2687224745750427, 0.2687224745750427, 0.27606460452079773, 0.303964763879776, 0.3274596035480499, 0.35462555289268494, 0.3538913428783417, 0.3274596035480499, 0.37444934248924255, 0.36710718274116516, 0.4302496314048767, 0.40381792187690735, 0.43171805143356323, 0.41262850165367126, 0.4508076310157776, 0.42584434151649475, 0.46622613072395325, 0.4816446304321289, 0.4904552102088928, 0.5227606296539307, NaN, 0.5256975293159485, 0.5073421597480774, 0.44419971108436584, 0.5080763697624207, 0.5741556286811829, 0.5565345287322998, 0.6534507870674133, 0.5550661087036133, 0.6182085275650024, 0.6123347878456116, 0.6270191073417664, 0.6424375772476196, 0.6233479976654053, 0.6123347878456116, 0.634361207485199, 0.676945686340332, 0.6358296871185303, 0.7063142657279968, 0.6725403666496277, 0.6725403666496277, 0.676945686340332, 0.7327459454536438, 0.7136563658714294, 0.7444933652877808, 0.7518355250358582, 0.7679882645606995, 0.7870778441429138, 0.7459618449211121, 0.8428781032562256, 0.8017621040344238, 0.850220263004303, 0.8281938433647156, 0.8046989440917969, 0.8494860529899597, 0.826725423336029, 0.8414096832275391, 0.8575624227523804, 0.8751835823059082, 0.8678414225578308, 0.8869310021400452, 0.9735682606697083], [0.06461086869239807, 0.12041115760803223, 0.13142438232898712, 0.12922173738479614, 0.1262848824262619, 0.10866373032331467, 0.14390602707862854, 0.17327459156513214, 0.21145375072956085, 0.27606460452079773, 0.2672540247440338, 0.3024963140487671, 0.26138031482696533, 0.2819383144378662, 0.3127753436565399, 0.32892805337905884, 0.40234947204589844, 0.40381792187690735, 0.31057268381118774, 0.3627019226551056, 0.3553597629070282, 0.4273127615451813, 0.3883994221687317, 0.44640234112739563, 0.4133627116680145, 0.37077826261520386, 0.45154184103012085, 0.5095447897911072, 0.4743025004863739, NaN, 0.48091042041778564, 0.48604992032051086, 0.5330396294593811, 0.5800293684005737, 0.5609397888183594, 0.5447870492935181, 0.5873715281486511, 0.6321585774421692, 0.6461086869239807, 0.6270191073417664, 0.6211453676223755, 0.6637297868728638, 0.6490455269813538, 0.6945667862892151, 0.6505139470100403, 0.6666666865348816, 0.6828193664550781, 0.7709251046180725, 0.7011747360229492, 0.7019089460372925, 0.7621145248413086, 0.8046989440917969, 0.781204104423523, 0.7973568439483643, 0.826725423336029, 0.7914831042289734, 0.7782672643661499, 0.7679882645606995, 0.8619676828384399, 0.8986784219741821, 0.8443465232849121, 0.8091042637825012, 0.8575624227523804, 0.826725423336029, 0.8487518429756165, 0.8839941024780273, 0.9133626818656921, 0.8869310021400452, 0.919236421585083, 0.9809104204177856], [0.05873715132474899, 0.1262848824262619, 0.12481644749641418, 0.14390602707862854, 0.12555065751075745, 0.1189427301287651, 0.14684288203716278, 0.16005873680114746, 0.16152717173099518, 0.2672540247440338, 0.2599118947982788, 0.29221731424331665, 0.26138031482696533, 0.26431718468666077, 0.27533039450645447, 0.3480176329612732, 0.37444934248924255, 0.3627019226551056, 0.2951541841030121, 0.3759177625179291, 0.36343613266944885, 0.392070472240448, 0.3759177625179291, 0.4302496314048767, 0.4229075014591217, 0.4067547619342804, 0.4251101315021515, 0.5044052600860596, 0.492657870054245, 0.5190895795822144, NaN, 0.4948605000972748, 0.515418529510498, 0.5565345287322998, 0.5609397888183594, 0.5726872086524963, 0.5712187886238098, 0.5675477385520935, 0.6138032078742981, 0.5947136282920837, 0.6049926280975342, 0.6123347878456116, 0.635095477104187, 0.6534507870674133, 0.6284875273704529, 0.6696035265922546, 0.676945686340332, 0.6725403666496277, 0.6710719466209412, 0.673274576663971, 0.7650514245033264, 0.8590308427810669, 0.8678414225578308, 0.8032305240631104, 0.7944199442863464, 0.7738619446754456, 0.7474302649497986, 0.8061674237251282, 0.8237885236740112, 0.826725423336029, 0.853157103061676, 0.8105726838111877, 0.8223201036453247, 0.8296622633934021, 0.8377386331558228, 0.9045521020889282, 0.9060205817222595, 0.8619676828384399, 0.8928046822547913, 0.9720998406410217], [0.05433186516165733, 0.11747430264949799, 0.10279001295566559, 0.13656388223171234, 0.14537444710731506, 0.10425844043493271, 0.18942731618881226, 0.17254038155078888, 0.26284876465797424, 0.2687224745750427, 0.24082231521606445, 0.27239352464675903, 0.30690160393714905, 0.3259911835193634, 0.33627018332481384, 0.3406754732131958, 0.38913363218307495, 0.3627019226551056, 0.4287812113761902, 0.38032305240631104, 0.40528634190559387, 0.4155653417110443, 0.4243759214878082, 0.40969163179397583, 0.43171805143356323, 0.47136563062667847, 0.5007342100143433, 0.5256975293159485, 0.5558003187179565, 0.5139500498771667, 0.5051395297050476, NaN, 0.4757709205150604, 0.553597629070282, 0.5301027894020081, 0.6593245267868042, 0.5462555289268494, 0.5682819485664368, 0.5969163179397583, 0.6211453676223755, 0.6005873680114746, 0.6005873680114746, 0.6093979477882385, 0.6093979477882385, 0.610866367816925, 0.6461086869239807, 0.6622613668441772, 0.6123347878456116, 0.6255506873130798, 0.622613787651062, 0.7048457860946655, 0.7048457860946655, 0.7621145248413086, 0.7195301055908203, 0.7415565252304077, 0.7474302649497986, 0.685756266117096, 0.8318648934364319, 0.7518355250358582, 0.7767988443374634, 0.8149779438972473, 0.7342143654823303, 0.7900146842002869, 0.7459618449211121, 0.7900146842002869, 0.826725423336029, 0.8296622633934021, 0.7797356843948364, 0.8223201036453247, 0.9500734210014343], [0.08076358586549759, 0.15418502688407898, 0.13950073719024658, 0.16446402668952942, 0.14243759214878082, 0.14757709205150604, 0.1483113020658493, 0.16005873680114746, 0.20851688086986542, 0.2819383144378662, 0.2929515540599823, 0.30837005376815796, 0.29368576407432556, 0.16226138174533844, 0.28487518429756165, 0.38179147243499756, 0.3568281829357147, 0.36710718274116516, 0.3443465530872345, 0.3127753436565399, 0.3994126319885254, 0.4728340804576874, 0.4177680015563965, 0.4449339210987091, 0.453744500875473, 0.3935389220714569, 0.4508076310157776, 0.4904552102088928, 0.49192363023757935, 0.4669603407382965, 0.48458150029182434, 0.5242290496826172, NaN, 0.4963289201259613, 0.6005873680114746, 0.5682819485664368, 0.566079318523407, 0.5212922096252441, 0.5609397888183594, 0.596182107925415, 0.6182085275650024, 0.5462555289268494, 0.6152716875076294, 0.5895741581916809, 0.6813509464263916, 0.6358296871185303, 0.6989721059799194, 0.6123347878456116, 0.6960352659225464, 0.6989721059799194, 0.6872246861457825, 0.6622613668441772, 0.700440526008606, 0.6989721059799194, 0.7437591552734375, 0.8105726838111877, 0.7753304243087769, 0.7268722653388977, 0.7533039450645447, 0.826725423336029, 0.7665198445320129, 0.8193832635879517, 0.8149779438972473, 0.8348017334938049, 0.8032305240631104, 0.8252570033073425, 0.8722466826438904, 0.8722466826438904, 0.8810572624206543, 0.9427312612533569], [0.08663729578256607, 0.13803230226039886, 0.13069015741348267, 0.13950073719024658, 0.13656388223171234, 0.14977973699569702, 0.1409691572189331, 0.15124815702438354, 0.2070484608411789, 0.2672540247440338, 0.276798814535141, 0.2966226041316986, 0.2657856047153473, 0.26284876465797424, 0.28487518429756165, 0.3024963140487671, 0.3627019226551056, 0.39647576212882996, 0.3237885534763336, 0.3553597629070282, 0.38179147243499756, 0.39647576212882996, 0.3906020522117615, 0.3994126319885254, 0.39794421195983887, 0.35976505279541016, 0.4243759214878082, 0.4324522614479065, 0.42584434151649475, 0.41997063159942627, 0.4434655010700226, 0.44640234112739563, 0.5036710500717163, NaN, 0.5403817892074585, 0.5212922096252441, 0.5007342100143433, 0.4977973699569702, 0.6997063159942627, 0.5653451085090637, 0.5323054194450378, 0.5301027894020081, 0.5895741581916809, 0.7327459454536438, 0.6101321578025818, 0.596182107925415, 0.6071953177452087, 0.6049926280975342, 0.6725403666496277, 0.6696035265922546, 0.6666666865348816, 0.6593245267868042, 0.6864904761314392, 0.691629946231842, 0.7753304243087769, 0.7356828451156616, 0.772393524646759, 0.7151248455047607, 0.7980910539627075, 0.8340675234794617, 0.7650514245033264, 0.8472834229469299, 0.8414096832275391, 0.8604992628097534, 0.8281938433647156, 0.8105726838111877, 0.9104258418083191, 0.8883994221687317, 0.880323052406311, 0.9559471607208252], [0.044052861630916595, 0.08516886830329895, 0.08810572326183319, 0.10572687536478043, 0.0969163030385971, 0.10425844043493271, 0.11160058528184891, 0.11747430264949799, 0.2290748953819275, 0.2217327505350113, 0.24449339509010315, 0.2540381848812103, 0.20558002591133118, 0.2525697648525238, 0.2745961844921112, 0.3127753436565399, 0.2966226041316986, 0.33773863315582275, 0.38325992226600647, 0.30690160393714905, 0.3715124726295471, 0.3773861825466156, 0.3773861825466156, 0.3348017632961273, 0.40969163179397583, 0.40234947204589844, 0.4251101315021515, 0.42584434151649475, 0.4434655010700226, 0.4390602111816406, 0.4390602111816406, 0.46989721059799194, 0.3994126319885254, 0.4596182107925415, NaN, 0.5477239489555359, 0.5205579996109009, 0.5007342100143433, 0.515418529510498, 0.5403817892074585, 0.6182085275650024, 0.5359765291213989, 0.607929527759552, 0.5580029487609863, 0.7165932655334473, 0.6167401075363159, 0.715859055519104, 0.5565345287322998, 0.6549192070960999, 0.6629955768585205, 0.6798825263977051, 0.6505139470100403, 0.688693106174469, 0.6975036859512329, 0.7195301055908203, 0.8722466826438904, 0.7606461048126221, 0.757709264755249, 0.7738619446754456, 0.8054332137107849, 0.7371512651443481, 0.7863436341285706, 0.8524228930473328, 0.8164464235305786, 0.7958884239196777, 0.7679882645606995, 0.8737151026725769, 0.8839941024780273, 0.8722466826438904, 0.9515418410301208], [0.07342144101858139, 0.11086637526750565, 0.12187959253787994, 0.11967694759368896, 0.1409691572189331, 0.1189427301287651, 0.14684288203716278, 0.16446402668952942, 0.22466960549354553, 0.19676946103572845, 0.232745960354805, 0.2540381848812103, 0.24963289499282837, 0.2819383144378662, 0.27606460452079773, 0.2907488942146301, 0.29955947399139404, 0.311306893825531, 0.33773863315582275, 0.3333333432674408, 0.3553597629070282, 0.3201174736022949, 0.37298092246055603, 0.36563876271247864, 0.38032305240631104, 0.3759177625179291, 0.3935389220714569, 0.40381792187690735, 0.3465491831302643, 0.45521292090415955, 0.4273127615451813, 0.3406754732131958, 0.43171805143356323, 0.47870779037475586, 0.4522760510444641, NaN, 0.45301029086112976, 0.5814977884292603, 0.527165949344635, 0.48604992032051086, 0.5403817892074585, 0.5932452082633972, 0.5668135285377502, 0.5756241083145142, 0.538913369178772, 0.5976505279541016, 0.6049926280975342, 0.6358296871185303, 0.6490455269813538, 0.6534507870674133, 0.6651982665061951, 0.6299559473991394, 0.6284875273704529, 0.7239353656768799, 0.7151248455047607, 0.7606461048126221, 0.7356828451156616, 0.7444933652877808, 0.7635829448699951, 0.826725423336029, 0.7980910539627075, 0.7973568439483643, 0.8325991034507751, 0.8157122135162354, 0.8604992628097534, 0.838472843170166, 0.8472834229469299, 0.8957415819168091, 0.8913362622261047, 0.9765051603317261], [0.055800292640924454, 0.07121879607439041, 0.08663729578256607, 0.10425844043493271, 0.10132158547639847, 0.07929515093564987, 0.11233480274677277, 0.1189427301287651, 0.15859030187129974, 0.17547723650932312, 0.17914831638336182, 0.2143906056880951, 0.20998531579971313, 0.2657856047153473, 0.2540381848812103, 0.3171806037425995, 0.2907488942146301, 0.3054331839084625, 0.3392070531845093, 0.34361234307289124, 0.31571218371391296, 0.26651981472969055, 0.35095447301864624, 0.39867842197418213, 0.35609397292137146, 0.39794421195983887, 0.4008810520172119, 0.41262850165367126, 0.4449339210987091, 0.41262850165367126, 0.4287812113761902, 0.453744500875473, 0.4339207112789154, 0.49926578998565674, 0.4794420003890991, 0.5469897389411926, NaN, 0.5124816298484802, 0.553597629070282, 0.5403817892074585, 0.5785609483718872, 0.5616739988327026, 0.5624082088470459, 0.5917767882347107, 0.5888399481773376, 0.6049926280975342, 0.6240822076797485, 0.5932452082633972, 0.6380323171615601, 0.634361207485199, 0.6710719466209412, 0.6740087866783142, 0.700440526008606, 0.7077826857566833, 0.7136563658714294, 0.7870778441429138, 0.7136563658714294, 0.7856094241142273, 0.7606461048126221, 0.8054332137107849, 0.7900146842002869, 0.727606475353241, 0.8296622633934021, 0.7591776847839355, 0.7797356843948364, 0.8208516836166382, 0.850220263004303, 0.8193832635879517, 0.8480176329612732, 0.9544786810874939], [0.06167401000857353, 0.09397944062948227, 0.08957415819168091, 0.10866373032331467, 0.10352423042058945, 0.08076358586549759, 0.12334802001714706, 0.14317180216312408, 0.18061673641204834, 0.21145375072956085, 0.21732746064662933, 0.24082231521606445, 0.21365638077259064, 0.2378854602575302, 0.27900147438049316, 0.2657856047153473, 0.3494860529899597, 0.3201174736022949, 0.2621145248413086, 0.29955947399139404, 0.31424376368522644, 0.3171806037425995, 0.36343613266944885, 0.40234947204589844, 0.3259911835193634, 0.34361234307289124, 0.38325992226600647, 0.39867842197418213, 0.38179147243499756, 0.3678414225578308, 0.4324522614479065, 0.43171805143356323, 0.47870779037475586, 0.5022026300430298, 0.49926578998565674, 0.41850221157073975, 0.4875183701515198, NaN, 0.5477239489555359, 0.5212922096252441, 0.5066079497337341, 0.5734214186668396, 0.5227606296539307, 0.5800293684005737, 0.5550661087036133, 0.5638766288757324, 0.5587371587753296, 0.7753304243087769, 0.6651982665061951, 0.6593245267868042, 0.6417033672332764, 0.7099853157997131, 0.6710719466209412, 0.6872246861457825, 0.7753304243087769, 0.6960352659225464, 0.7533039450645447, 0.7195301055908203, 0.8061674237251282, 0.8002936840057373, 0.7606461048126221, 0.7430249452590942, 0.8091042637825012, 0.7606461048126221, 0.7621145248413086, 0.7988252639770508, 0.8678414225578308, 0.8443465232849121, 0.8634361028671265, 0.9684287905693054], [0.07488986849784851, 0.12334802001714706, 0.12187959253787994, 0.13656388223171234, 0.1226138025522232, 0.12555065751075745, 0.12041115760803223, 0.1409691572189331, 0.19530102610588074, 0.2342143952846527, 0.257709264755249, 0.2731277644634247, 0.24229075014591217, 0.23348017036914825, 0.24963289499282837, 0.27019089460372925, 0.3274596035480499, 0.3722466826438904, 0.28487518429756165, 0.3186490535736084, 0.3281938433647156, 0.35242289304733276, 0.34361234307289124, 0.36563876271247864, 0.34508076310157776, 0.30690160393714905, 0.3641703426837921, 0.3876652121543884, 0.3876652121543884, 0.3538913428783417, 0.3861967623233795, 0.4030837118625641, 0.4390602111816406, 0.3002936840057373, 0.48458150029182434, 0.4728340804576874, 0.44640234112739563, 0.4522760510444641, NaN, 0.5007342100143433, 0.4889867901802063, 0.4765051305294037, 0.538913369178772, 0.6336269974708557, 0.5756241083145142, 0.5712187886238098, 0.5763583183288574, 0.553597629070282, 0.6549192070960999, 0.6571218967437744, 0.6270191073417664, 0.6093979477882385, 0.6417033672332764, 0.6710719466209412, 0.7400881052017212, 0.7121879458427429, 0.7547723650932312, 0.691629946231842, 0.7679882645606995, 0.8002936840057373, 0.7533039450645447, 0.8208516836166382, 0.8091042637825012, 0.835536003112793, 0.8046989440917969, 0.7870778441429138, 0.8898678421974182, 0.8693098425865173, 0.8795888423919678, 0.9486050009727478], [0.06314244121313095, 0.08223201334476471, 0.10572687536478043, 0.10572687536478043, 0.11747430264949799, 0.10866373032331467, 0.12922173738479614, 0.12922173738479614, 0.11820852011442184, 0.1938326060771942, 0.21732746064662933, 0.24082231521606445, 0.2378854602575302, 0.20925110578536987, 0.22026431560516357, 0.30102789402008057, 0.2525697648525238, 0.345814973115921, 0.30837005376815796, 0.27019089460372925, 0.32892805337905884, 0.31424376368522644, 0.34508076310157776, 0.3274596035480499, 0.38179147243499756, 0.3355359733104706, 0.3759177625179291, 0.4243759214878082, 0.37298092246055603, 0.37298092246055603, 0.40528634190559387, 0.3788546323776245, 0.40381792187690735, 0.43465492129325867, 0.4596182107925415, 0.5139500498771667, 0.4596182107925415, 0.47870779037475586, 0.49926578998565674, NaN, 0.5176211595535278, 0.4977973699569702, 0.5638766288757324, 0.5301027894020081, 0.5873715281486511, 0.5976505279541016, 0.607929527759552, 0.5521292090415955, 0.6402349472045898, 0.6490455269813538, 0.6607929468154907, 0.6358296871185303, 0.6490455269813538, 0.6901615262031555, 0.7092511057853699, 0.7863436341285706, 0.7386196851730347, 0.7268722653388977, 0.7679882645606995, 0.7914831042289734, 0.7709251046180725, 0.7767988443374634, 0.8678414225578308, 0.8076358437538147, 0.8223201036453247, 0.7973568439483643, 0.864904522895813, 0.8766520023345947, 0.8883994221687317, 0.9750367403030396], [0.07635829597711563, 0.12555065751075745, 0.12187959253787994, 0.13509544730186462, 0.1226138025522232, 0.13509544730186462, 0.10719530284404755, 0.13950073719024658, 0.17988252639770508, 0.17621144652366638, 0.22466960549354553, 0.24669604003429413, 0.20558002591133118, 0.2452276051044464, 0.25550660490989685, 0.3002936840057373, 0.2290748953819275, 0.311306893825531, 0.311306893825531, 0.30469897389411926, 0.35095447301864624, 0.311306893825531, 0.36710718274116516, 0.3024963140487671, 0.3678414225578308, 0.350220263004303, 0.3715124726295471, 0.39867842197418213, 0.35756242275238037, 0.3788546323776245, 0.39500734210014343, 0.3994126319885254, 0.38179147243499756, 0.46769458055496216, 0.38179147243499756, 0.4596182107925415, 0.4214390516281128, 0.49339208006858826, 0.5110132098197937, 0.48237884044647217, NaN, 0.515418529510498, 0.5183553695678711, 0.527165949344635, 0.515418529510498, 0.5345080494880676, 0.6563876867294312, 0.5433186292648315, 0.6138032078742981, 0.6182085275650024, 0.6439059972763062, 0.5704845786094666, 0.6064611077308655, 0.6578561067581177, 0.6519823670387268, 0.7261380553245544, 0.7224669456481934, 0.7107195258140564, 0.6901615262031555, 0.7885462641716003, 0.7430249452590942, 0.8076358437538147, 0.9456681609153748, 0.8311306834220886, 0.7988252639770508, 0.7562408447265625, 0.8208516836166382, 0.8795888423919678, 0.8737151026725769, 0.9456681609153748], [0.055800292640924454, 0.09251101315021515, 0.08223201334476471, 0.0969163030385971, 0.09397944062948227, 0.07635829597711563, 0.11747430264949799, 0.1262848824262619, 0.1740088164806366, 0.19676946103572845, 0.20117473602294922, 0.2290748953819275, 0.20411160588264465, 0.20411160588264465, 0.24816446006298065, 0.24963289499282837, 0.3171806037425995, 0.31571218371391296, 0.23201175034046173, 0.30690160393714905, 0.29221731424331665, 0.3098384737968445, 0.3223201036453247, 0.38325992226600647, 0.311306893825531, 0.3171806037425995, 0.3480176329612732, 0.35242289304733276, 0.37665197253227234, 0.33627018332481384, 0.3876652121543884, 0.3994126319885254, 0.453744500875473, 0.46989721059799194, 0.46402350068092346, 0.4067547619342804, 0.43832600116729736, 0.426578551530838, 0.5234948396682739, 0.5022026300430298, 0.48458150029182434, NaN, 0.5022026300430298, 0.5477239489555359, 0.5256975293159485, 0.5477239489555359, 0.5381791591644287, 0.7077826857566833, 0.6519823670387268, 0.6505139470100403, 0.610866367816925, 0.6666666865348816, 0.6483113169670105, 0.6593245267868042, 0.730543315410614, 0.685756266117096, 0.7356828451156616, 0.6960352659225464, 0.7628487348556519, 0.7767988443374634, 0.7488986849784851, 0.7283406853675842, 0.784140944480896, 0.7356828451156616, 0.7430249452590942, 0.7635829448699951, 0.8582966327667236, 0.8281938433647156, 0.8370044231414795, 0.9544786810874939], [0.05433186516165733, 0.07342144101858139, 0.09104258567094803, 0.12114537507295609, 0.14243759214878082, 0.10719530284404755, 0.14537444710731506, 0.14757709205150604, 0.16886931657791138, 0.15051394701004028, 0.17621144652366638, 0.2143906056880951, 0.2584434747695923, 0.28634360432624817, 0.24816446006298065, 0.2584434747695923, 0.3024963140487671, 0.24816446006298065, 0.29001468420028687, 0.3348017632961273, 0.28634360432624817, 0.29588839411735535, 0.3054331839084625, 0.3318648934364319, 0.3406754732131958, 0.3281938433647156, 0.3612334728240967, 0.3759177625179291, 0.3876652121543884, 0.35095447301864624, 0.3649045526981354, 0.3906020522117615, 0.384728342294693, 0.4104258418083191, 0.392070472240448, 0.43318650126457214, 0.4375917911529541, 0.47723934054374695, 0.461086630821228, 0.4361233413219452, 0.4816446304321289, 0.4977973699569702, NaN, 0.49339208006858826, 0.4559471309185028, 0.6556534767150879, 0.5227606296539307, 0.5286343693733215, 0.5521292090415955, 0.5638766288757324, 0.7400881052017212, 0.5594713687896729, 0.5682819485664368, 0.7988252639770508, 0.6534507870674133, 0.6211453676223755, 0.6982378959655762, 0.7195301055908203, 0.6828193664550781, 0.7165932655334473, 0.9060205817222595, 0.727606475353241, 0.7474302649497986, 0.7378854751586914, 0.8766520023345947, 0.8781204223632812, 0.7635829448699951, 0.8105726838111877, 0.9089574217796326, 0.9897210001945496], [0.07782672345638275, 0.10792951285839081, 0.12041115760803223, 0.11013215780258179, 0.1262848824262619, 0.13069015741348267, 0.11306901276111603, 0.13950073719024658, 0.17254038155078888, 0.20264317095279694, 0.24229075014591217, 0.2525697648525238, 0.1997063159942627, 0.22099852561950684, 0.2525697648525238, 0.2657856047153473, 0.3054331839084625, 0.3201174736022949, 0.2143906056880951, 0.27900147438049316, 0.28487518429756165, 0.3201174736022949, 0.2892804741859436, 0.3370043933391571, 0.32158589363098145, 0.2687224745750427, 0.3171806037425995, 0.3553597629070282, 0.36563876271247864, 0.3054331839084625, 0.3465491831302643, 0.3906020522117615, 0.4104258418083191, 0.2672540247440338, 0.44199705123901367, 0.4243759214878082, 0.4082232117652893, 0.41997063159942627, 0.3663729727268219, 0.46989721059799194, 0.4728340804576874, 0.4522760510444641, 0.5066079497337341, NaN, 0.5491923689842224, 0.5183553695678711, 0.5447870492935181, 0.5110132098197937, 0.6358296871185303, 0.6358296871185303, 0.596182107925415, 0.5638766288757324, 0.5859031081199646, 0.6270191073417664, 0.6930983662605286, 0.700440526008606, 0.7342143654823303, 0.6475771069526672, 0.7209985256195068, 0.7944199442863464, 0.7371512651443481, 0.8091042637825012, 0.8032305240631104, 0.8348017334938049, 0.8252570033073425, 0.757709264755249, 0.8604992628097534, 0.8913362622261047, 0.8913362622261047, 0.9515418410301208], [0.036710720509290695, 0.06901615113019943, 0.06314244121313095, 0.08663729578256607, 0.07195301353931427, 0.07195301353931427, 0.08810572326183319, 0.08957415819168091, 0.17914831638336182, 0.1703377366065979, 0.19089573621749878, 0.21879589557647705, 0.16740088164806366, 0.17988252639770508, 0.20851688086986542, 0.2672540247440338, 0.2452276051044464, 0.2878120541572571, 0.29368576407432556, 0.2415565401315689, 0.2878120541572571, 0.2966226041316986, 0.30176210403442383, 0.30837005376815796, 0.3318648934364319, 0.3208516836166382, 0.35095447301864624, 0.3627019226551056, 0.32305434346199036, 0.3494860529899597, 0.3715124726295471, 0.38913363218307495, 0.3186490535736084, 0.3898678421974182, 0.2834067642688751, 0.461086630821228, 0.41116005182266235, 0.4449339210987091, 0.4243759214878082, 0.41262850165367126, 0.48458150029182434, 0.4743025004863739, 0.5440528392791748, 0.4508076310157776, NaN, 0.5506607890129089, 0.5785609483718872, 0.5007342100143433, 0.6431717872619629, 0.6446402072906494, 0.6152716875076294, 0.5616739988327026, 0.5829662084579468, 0.6387665271759033, 0.6431717872619629, 0.7848751544952393, 0.7371512651443481, 0.6960352659225464, 0.6945667862892151, 0.7474302649497986, 0.6989721059799194, 0.7077826857566833, 0.7826725244522095, 0.7290748953819275, 0.7540381550788879, 0.7239353656768799, 0.8076358437538147, 0.8546255230903625, 0.8281938433647156, 0.9368575811386108], [0.06020557880401611, 0.07195301353931427, 0.09985315799713135, 0.1189427301287651, 0.13950073719024658, 0.11233480274677277, 0.1336270123720169, 0.1409691572189331, 0.15491923689842224, 0.13069015741348267, 0.16740088164806366, 0.20264317095279694, 0.23201175034046173, 0.2834067642688751, 0.230543315410614, 0.2584434747695923, 0.28487518429756165, 0.22760646045207977, 0.28634360432624817, 0.3259911835193634, 0.25917768478393555, 0.2687224745750427, 0.29368576407432556, 0.3318648934364319, 0.29368576407432556, 0.31057268381118774, 0.3171806037425995, 0.3348017632961273, 0.3641703426837921, 0.3333333432674408, 0.33039647340774536, 0.3538913428783417, 0.3641703426837921, 0.40381792187690735, 0.38325992226600647, 0.40234947204589844, 0.39500734210014343, 0.4361233413219452, 0.4287812113761902, 0.40234947204589844, 0.46549192070961, 0.4522760510444641, 0.3443465530872345, 0.4816446304321289, 0.44933921098709106, NaN, 0.5080763697624207, 0.515418529510498, 0.5543318390846252, 0.5624082088470459, 0.6387665271759033, 0.5212922096252441, 0.5080763697624207, 0.7143905758857727, 0.6328927874565125, 0.6005873680114746, 0.6945667862892151, 0.6813509464263916, 0.6637297868728638, 0.7107195258140564, 0.8281938433647156, 0.7121879458427429, 0.7356828451156616, 0.7195301055908203, 0.8472834229469299, 0.8377386331558228, 0.7503671050071716, 0.8032305240631104, 0.9133626818656921, 0.988252580165863], [0.05726872384548187, 0.08590308576822281, 0.09985315799713135, 0.10279001295566559, 0.09251101315021515, 0.10132158547639847, 0.08663729578256607, 0.09985315799713135, 0.16372980177402496, 0.13069015741348267, 0.18355359137058258, 0.20264317095279694, 0.16152717173099518, 0.19089573621749878, 0.1997063159942627, 0.23715125024318695, 0.20558002591133118, 0.2569750249385834, 0.23935389518737793, 0.23201175034046173, 0.2819383144378662, 0.25036710500717163, 0.2914831042289734, 0.2672540247440338, 0.2980910539627075, 0.2819383144378662, 0.3186490535736084, 0.32892805337905884, 0.29368576407432556, 0.3171806037425995, 0.32305434346199036, 0.33773863315582275, 0.30102789402008057, 0.39280471205711365, 0.2841409742832184, 0.39500734210014343, 0.3759177625179291, 0.4412628412246704, 0.42364171147346497, 0.392070472240448, 0.34361234307289124, 0.4618208408355713, 0.47723934054374695, 0.45521292090415955, 0.4214390516281128, 0.49192363023757935, NaN, 0.5051395297050476, 0.5829662084579468, 0.5800293684005737, 0.5851688981056213, 0.5036710500717163, 0.5242290496826172, 0.5991189479827881, 0.5932452082633972, 0.6541849970817566, 0.6960352659225464, 0.6431717872619629, 0.6387665271759033, 0.7430249452590942, 0.7077826857566833, 0.7621145248413086, 0.8443465232849121, 0.7804698944091797, 0.769456684589386, 0.7195301055908203, 0.7606461048126221, 0.8516886830329895, 0.8414096832275391, 0.9353891611099243], [0.041116006672382355, 0.08076358586549759, 0.07929515093564987, 0.08810572326183319, 0.08370044082403183, 0.06461086869239807, 0.08957415819168091, 0.10572687536478043, 0.15124815702438354, 0.16593244671821594, 0.17474302649497986, 0.1850220263004303, 0.16446402668952942, 0.15418502688407898, 0.21292217075824738, 0.23348017036914825, 0.2364170402288437, 0.2599118947982788, 0.15124815702438354, 0.23127754032611847, 0.23494860529899597, 0.2621145248413086, 0.24669604003429413, 0.3318648934364319, 0.2878120541572571, 0.2143906056880951, 0.28707781434059143, 0.311306893825531, 0.3274596035480499, 0.2290748953819275, 0.3274596035480499, 0.3876652121543884, 0.3876652121543884, 0.39500734210014343, 0.4434655010700226, 0.3641703426837921, 0.4067547619342804, 0.22466960549354553, 0.44640234112739563, 0.44787079095840454, 0.45668134093284607, 0.29221731424331665, 0.47136563062667847, 0.4889867901802063, 0.49926578998565674, 0.48458150029182434, 0.4948605000972748, NaN, 0.622613787651062, 0.6270191073417664, 0.5859031081199646, 0.5763583183288574, 0.5910425782203674, 0.6064611077308655, 0.6725403666496277, 0.6314243674278259, 0.7048457860946655, 0.6431717872619629, 0.7077826857566833, 0.7430249452590942, 0.6387665271759033, 0.7195301055908203, 0.7283406853675842, 0.7312775254249573, 0.6938325762748718, 0.7224669456481934, 0.8105726838111877, 0.826725423336029, 0.8414096832275391, 0.9530102610588074], [0.15271659195423126, 0.17841409146785736, 0.19676946103572845, 0.18355359137058258, 0.20411160588264465, 0.19530102610588074, 0.20264317095279694, 0.23201175034046173, 0.26138031482696533, 0.2672540247440338, 0.27900147438049316, 0.28854626417160034, 0.2672540247440338, 0.27165931463241577, 0.27753305435180664, 0.2621145248413086, 0.2584434747695923, 0.32892805337905884, 0.28707781434059143, 0.29221731424331665, 0.2892804741859436, 0.3318648934364319, 0.2878120541572571, 0.3612334728240967, 0.3208516836166382, 0.29955947399139404, 0.3127753436565399, 0.3171806037425995, 0.3274596035480499, 0.2988252639770508, 0.32892805337905884, 0.37444934248924255, 0.303964763879776, 0.3274596035480499, 0.34508076310157776, 0.35095447301864624, 0.36196768283843994, 0.3348017632961273, 0.34508076310157776, 0.35976505279541016, 0.3861967623233795, 0.3480176329612732, 0.44787079095840454, 0.3641703426837921, 0.3568281829357147, 0.44566813111305237, 0.41703376173973083, 0.3773861825466156, NaN, 0.5337738394737244, 0.49339208006858826, 0.41850221157073975, 0.4243759214878082, 0.5374449491500854, 0.5682819485664368, 0.46549192070961, 0.9441996812820435, 0.5374449491500854, 0.5947136282920837, 0.4904552102088928, 0.6358296871185303, 0.6534507870674133, 0.6622613668441772, 0.6784141063690186, 0.7121879458427429, 0.6505139470100403, 0.6299559473991394, 0.781204104423523, 0.8340675234794617, 0.9236416816711426], [0.15418502688407898, 0.18061673641204834, 0.19676946103572845, 0.18942731618881226, 0.20411160588264465, 0.19089573621749878, 0.2033773809671402, 0.22760646045207977, 0.2569750249385834, 0.26431718468666077, 0.27753305435180664, 0.2907488942146301, 0.26431718468666077, 0.2687224745750427, 0.27606460452079773, 0.2672540247440338, 0.2569750249385834, 0.3274596035480499, 0.28634360432624817, 0.28707781434059143, 0.29368576407432556, 0.3311306834220886, 0.2892804741859436, 0.3700440526008606, 0.32305434346199036, 0.2980910539627075, 0.3171806037425995, 0.31571218371391296, 0.32305434346199036, 0.2980910539627075, 0.32672539353370667, 0.3773861825466156, 0.30102789402008057, 0.33039647340774536, 0.3370043933391571, 0.3465491831302643, 0.36563876271247864, 0.3406754732131958, 0.342878133058548, 0.35095447301864624, 0.38179147243499756, 0.3494860529899597, 0.4361233413219452, 0.3641703426837921, 0.3553597629070282, 0.4375917911529541, 0.41997063159942627, 0.37298092246055603, 0.46622613072395325, NaN, 0.49339208006858826, 0.4243759214878082, 0.4214390516281128, 0.5403817892074585, 0.5638766288757324, 0.46255508065223694, 0.9603524208068848, 0.5345080494880676, 0.5888399481773376, 0.4948605000972748, 0.6372981071472168, 0.6446402072906494, 0.6593245267868042, 0.6754772663116455, 0.7019089460372925, 0.6534507870674133, 0.6284875273704529, 0.781204104423523, 0.8281938433647156, 0.9236416816711426], [0.033773861825466156, 0.06607929617166519, 0.06754772365093231, 0.07342144101858139, 0.08663729578256607, 0.07635829597711563, 0.11453744769096375, 0.11306901276111603, 0.08957415819168091, 0.11160058528184891, 0.11160058528184891, 0.15859030187129974, 0.1703377366065979, 0.22026431560516357, 0.16446402668952942, 0.1997063159942627, 0.24963289499282837, 0.2217327505350113, 0.1850220263004303, 0.25550660490989685, 0.17327459156513214, 0.18942731618881226, 0.18942731618881226, 0.2547723948955536, 0.23201175034046173, 0.23935389518737793, 0.20484581589698792, 0.303964763879776, 0.2672540247440338, 0.2378854602575302, 0.23494860529899597, 0.2951541841030121, 0.3127753436565399, 0.3333333432674408, 0.3201174736022949, 0.3348017632961273, 0.32892805337905884, 0.35829663276672363, 0.37298092246055603, 0.3392070531845093, 0.35609397292137146, 0.38913363218307495, 0.2599118947982788, 0.40381792187690735, 0.384728342294693, 0.3612334728240967, 0.41483113169670105, 0.4140969216823578, 0.5066079497337341, 0.5066079497337341, NaN, 0.4801762104034424, 0.49339208006858826, 0.6754772663116455, 0.5814977884292603, 0.541850209236145, 0.6439059972763062, 0.6563876867294312, 0.6182085275650024, 0.6255506873130798, 0.7665198445320129, 0.685756266117096, 0.6828193664550781, 0.6901615262031555, 0.7767988443374634, 0.8810572624206543, 0.703377366065979, 0.7731277346611023, 0.8634361028671265, 0.9662261605262756], [0.033773861825466156, 0.046989720314741135, 0.046989720314741135, 0.06020557880401611, 0.051395006477832794, 0.042584434151649475, 0.06314244121313095, 0.06020557880401611, 0.07782672345638275, 0.11306901276111603, 0.11747430264949799, 0.14243759214878082, 0.08516886830329895, 0.12701909244060516, 0.1409691572189331, 0.18942731618881226, 0.17841409146785736, 0.20851688086986542, 0.04845815151929855, 0.20558002591133118, 0.12187959253787994, 0.13803230226039886, 0.1483113020658493, 0.2687224745750427, 0.21292217075824738, 0.09618208557367325, 0.10132158547639847, 0.2731277644634247, 0.28634360432624817, 0.19530102610588074, 0.1409691572189331, 0.2951541841030121, 0.33773863315582275, 0.3406754732131958, 0.3494860529899597, 0.3700440526008606, 0.3259911835193634, 0.29001468420028687, 0.3906020522117615, 0.3641703426837921, 0.42951542139053345, 0.3333333432674408, 0.44052863121032715, 0.4361233413219452, 0.43832600116729736, 0.47870779037475586, 0.4963289201259613, 0.42364171147346497, 0.5814977884292603, 0.5756241083145142, 0.5198237895965576, NaN, 0.5249632596969604, 0.56901615858078, 0.5829662084579468, 0.6255506873130798, 0.6622613668441772, 0.6651982665061951, 0.6402349472045898, 0.7121879458427429, 0.688693106174469, 0.6784141063690186, 0.7151248455047607, 0.6945667862892151, 0.7342143654823303, 0.742290735244751, 0.7797356843948364, 0.8017621040344238, 0.8193832635879517, 0.946402370929718], [0.030837005004286766, 0.033773861825466156, 0.041116006672382355, 0.06754772365093231, 0.049926578998565674, 0.033773861825466156, 0.06461086869239807, 0.06314244121313095, 0.06461086869239807, 0.10719530284404755, 0.11747430264949799, 0.1556534469127655, 0.0969163030385971, 0.11306901276111603, 0.13069015741348267, 0.17914831638336182, 0.20851688086986542, 0.23494860529899597, 0.10132158547639847, 0.17180617153644562, 0.11013215780258179, 0.15271659195423126, 0.11453744769096375, 0.27165931463241577, 0.1938326060771942, 0.13289280235767365, 0.0969163030385971, 0.27900147438049316, 0.25550660490989685, 0.21879589557647705, 0.13215859234333038, 0.2378854602575302, 0.29955947399139404, 0.3135095536708832, 0.311306893825531, 0.3715124726295471, 0.29955947399139404, 0.32892805337905884, 0.35829663276672363, 0.35095447301864624, 0.3935389220714569, 0.3516886830329895, 0.43171805143356323, 0.4140969216823578, 0.41703376173973083, 0.49192363023757935, 0.4757709205150604, 0.40895742177963257, 0.5756241083145142, 0.5785609483718872, 0.5066079497337341, 0.47503671050071716, NaN, 0.5712187886238098, 0.538913369178772, 0.5873715281486511, 0.6696035265922546, 0.6901615262031555, 0.6049926280975342, 0.7268722653388977, 0.7268722653388977, 0.6828193664550781, 0.6930983662605286, 0.6989721059799194, 0.7386196851730347, 0.7826725244522095, 0.7298091053962708, 0.8002936840057373, 0.8135095238685608, 0.933920681476593], [0.035242289304733276, 0.049926578998565674, 0.06314244121313095, 0.07195301353931427, 0.07488986849784851, 0.06020557880401611, 0.07929515093564987, 0.0969163030385971, 0.06461086869239807, 0.10866373032331467, 0.09985315799713135, 0.13803230226039886, 0.13876652717590332, 0.17914831638336182, 0.1409691572189331, 0.18208517134189606, 0.20558002591133118, 0.18208517134189606, 0.13950073719024658, 0.23494860529899597, 0.1299559473991394, 0.15712188184261322, 0.15124815702438354, 0.24596181511878967, 0.19530102610588074, 0.1850220263004303, 0.17694567143917084, 0.26138031482696533, 0.24816446006298065, 0.20264317095279694, 0.19676946103572845, 0.2804698944091797, 0.30102789402008057, 0.30837005376815796, 0.3024963140487671, 0.27606460452079773, 0.29221731424331665, 0.3127753436565399, 0.32892805337905884, 0.3098384737968445, 0.3421438932418823, 0.3406754732131958, 0.20117473602294922, 0.37298092246055603, 0.3612334728240967, 0.2856093943119049, 0.4008810520172119, 0.3935389220714569, 0.46255508065223694, 0.4596182107925415, 0.3245227634906769, 0.43098384141921997, 0.4287812113761902, NaN, 0.5565345287322998, 0.5022026300430298, 0.622613787651062, 0.6093979477882385, 0.5991189479827881, 0.6035242080688477, 0.6828193664550781, 0.6725403666496277, 0.6798825263977051, 0.6784141063690186, 0.7437591552734375, 0.7452276349067688, 0.6828193664550781, 0.7665198445320129, 0.8546255230903625, 0.9647576808929443], [0.016152717173099518, 0.049926578998565674, 0.044052861630916595, 0.051395006477832794, 0.046989720314741135, 0.051395006477832794, 0.06167401000857353, 0.06975036859512329, 0.06461086869239807, 0.09985315799713135, 0.10279001295566559, 0.12041115760803223, 0.07342144101858139, 0.12041115760803223, 0.10866373032331467, 0.17621144652366638, 0.1336270123720169, 0.1629955917596817, 0.09985315799713135, 0.14537444710731506, 0.14537444710731506, 0.1483113020658493, 0.15712188184261322, 0.2143906056880951, 0.18795888125896454, 0.13876652717590332, 0.18208517134189606, 0.2268722504377365, 0.23201175034046173, 0.17327459156513214, 0.20558002591133118, 0.2584434747695923, 0.2562408149242401, 0.22466960549354553, 0.2804698944091797, 0.28487518429756165, 0.28634360432624817, 0.22466960549354553, 0.2599118947982788, 0.2907488942146301, 0.3480176329612732, 0.269456684589386, 0.3465491831302643, 0.30690160393714905, 0.3568281829357147, 0.36710718274116516, 0.4067547619342804, 0.3274596035480499, 0.43171805143356323, 0.4361233413219452, 0.41850221157073975, 0.41703376173973083, 0.461086630821228, 0.4434655010700226, NaN, 0.5051395297050476, 0.5212922096252441, 0.5477239489555359, 0.7085168957710266, 0.5741556286811829, 0.5668135285377502, 0.6402349472045898, 0.6005873680114746, 0.6534507870674133, 0.5998531579971313, 0.6020557880401611, 0.8223201036453247, 0.7180616855621338, 0.7400881052017212, 0.9280470013618469], [0.029368575662374496, 0.036710720509290695, 0.036710720509290695, 0.06167401000857353, 0.045521292835474014, 0.044052861630916595, 0.027900146320462227, 0.041116006672382355, 0.07488986849784851, 0.08810572326183319, 0.0969163030385971, 0.11013215780258179, 0.09251101315021515, 0.08957415819168091, 0.10425844043493271, 0.18061673641204834, 0.12041115760803223, 0.1703377366065979, 0.17180617153644562, 0.11160058528184891, 0.16446402668952942, 0.1556534469127655, 0.18208517134189606, 0.2143906056880951, 0.15124815702438354, 0.1923641711473465, 0.21145375072956085, 0.15859030187129974, 0.21292217075824738, 0.20851688086986542, 0.22613802552223206, 0.2525697648525238, 0.18942731618881226, 0.26431718468666077, 0.12775330245494843, 0.23935389518737793, 0.21292217075824738, 0.303964763879776, 0.2878120541572571, 0.21365638077259064, 0.27386197447776794, 0.31424376368522644, 0.3788546323776245, 0.29955947399139404, 0.21512481570243835, 0.3994126319885254, 0.345814973115921, 0.3685756325721741, 0.5345080494880676, 0.5374449491500854, 0.458149790763855, 0.37444934248924255, 0.41262850165367126, 0.4977973699569702, 0.4948605000972748, NaN, 0.6461086869239807, 0.5800293684005737, 0.541850209236145, 0.5682819485664368, 0.6182085275650024, 0.6417033672332764, 0.6490455269813538, 0.6622613668441772, 0.6828193664550781, 0.6328927874565125, 0.6475771069526672, 0.7797356843948364, 0.7944199442863464, 0.91629958152771], [0.11453744769096375, 0.13215859234333038, 0.15124815702438354, 0.12922173738479614, 0.1629955917596817, 0.15712188184261322, 0.15418502688407898, 0.17180617153644562, 0.19530102610588074, 0.1850220263004303, 0.19676946103572845, 0.1997063159942627, 0.20117473602294922, 0.20411160588264465, 0.2143906056880951, 0.21659325063228607, 0.18795888125896454, 0.21879589557647705, 0.20558002591133118, 0.22760646045207977, 0.22613802552223206, 0.230543315410614, 0.22466960549354553, 0.23935389518737793, 0.24816446006298065, 0.22320117056369781, 0.24963289499282837, 0.24963289499282837, 0.2540381848812103, 0.2217327505350113, 0.2525697648525238, 0.31424376368522644, 0.22466960549354553, 0.22760646045207977, 0.23935389518737793, 0.26431718468666077, 0.28634360432624817, 0.24669604003429413, 0.2452276051044464, 0.26138031482696533, 0.27753305435180664, 0.26431718468666077, 0.30176210403442383, 0.2657856047153473, 0.26284876465797424, 0.3054331839084625, 0.303964763879776, 0.2951541841030121, 0.055800292640924454, 0.039647575467824936, 0.35609397292137146, 0.33773863315582275, 0.33039647340774536, 0.3773861825466156, 0.47870779037475586, 0.3538913428783417, NaN, 0.4273127615451813, 0.4977973699569702, 0.4140969216823578, 0.4728340804576874, 0.5359765291213989, 0.4889867901802063, 0.5477239489555359, 0.5814977884292603, 0.45521292090415955, 0.5315712094306946, 0.6578561067581177, 0.6960352659225464, 0.8898678421974182], [0.029368575662374496, 0.035242289304733276, 0.04845815151929855, 0.05286343768239021, 0.06607929617166519, 0.051395006477832794, 0.06607929617166519, 0.07488986849784851, 0.07342144101858139, 0.09985315799713135, 0.08883994072675705, 0.11453744769096375, 0.11453744769096375, 0.14684288203716278, 0.13215859234333038, 0.13656388223171234, 0.15859030187129974, 0.1556534469127655, 0.18208517134189606, 0.1938326060771942, 0.15859030187129974, 0.12187959253787994, 0.1703377366065979, 0.20998531579971313, 0.06020557880401611, 0.1886930912733078, 0.18942731618881226, 0.11013215780258179, 0.15712188184261322, 0.23201175034046173, 0.1938326060771942, 0.16813509166240692, 0.2731277644634247, 0.28487518429756165, 0.24229075014591217, 0.25550660490989685, 0.2143906056880951, 0.2804698944091797, 0.30837005376815796, 0.2731277644634247, 0.2892804741859436, 0.303964763879776, 0.2804698944091797, 0.35242289304733276, 0.303964763879776, 0.3186490535736084, 0.3568281829357147, 0.3568281829357147, 0.46255508065223694, 0.46549192070961, 0.34361234307289124, 0.3348017632961273, 0.3098384737968445, 0.3906020522117615, 0.4522760510444641, 0.41997063159942627, 0.5726872086524963, NaN, 0.4765051305294037, 0.5315712094306946, 0.4728340804576874, 0.607929527759552, 0.5447870492935181, 0.607929527759552, 0.5932452082633972, 0.541850209236145, 0.5859031081199646, 0.7048457860946655, 0.7180616855621338, 0.922173261642456], [0.017621144652366638, 0.039647575467824936, 0.038913361728191376, 0.046989720314741135, 0.046989720314741135, 0.05433186516165733, 0.05433186516165733, 0.05873715132474899, 0.05726872384548187, 0.07929515093564987, 0.09177679568529129, 0.11160058528184891, 0.07048457860946655, 0.09985315799713135, 0.10425844043493271, 0.16740088164806366, 0.1189427301287651, 0.13950073719024658, 0.08223201334476471, 0.13509544730186462, 0.1262848824262619, 0.12041115760803223, 0.13509544730186462, 0.20778267085552216, 0.16079294681549072, 0.11747430264949799, 0.1593245267868042, 0.18428781628608704, 0.19823788106441498, 0.13803230226039886, 0.17621144652366638, 0.24816446006298065, 0.24669604003429413, 0.20190896093845367, 0.22613802552223206, 0.2364170402288437, 0.23935389518737793, 0.1938326060771942, 0.23201175034046173, 0.23201175034046173, 0.3098384737968445, 0.23715125024318695, 0.3171806037425995, 0.27900147438049316, 0.3054331839084625, 0.33627018332481384, 0.3612334728240967, 0.29221731424331665, 0.40528634190559387, 0.41116005182266235, 0.38179147243499756, 0.35976505279541016, 0.39500734210014343, 0.4008810520172119, 0.2914831042289734, 0.458149790763855, 0.5022026300430298, 0.5234948396682739, NaN, 0.5462555289268494, 0.5330396294593811, 0.6240822076797485, 0.5565345287322998, 0.634361207485199, 0.5609397888183594, 0.5712187886238098, 0.7114537358283997, 0.700440526008606, 0.7180616855621338, 0.9030836820602417], [0.051395006477832794, 0.045521292835474014, 0.049926578998565674, 0.06167401000857353, 0.05726872384548187, 0.05726872384548187, 0.049926578998565674, 0.06020557880401611, 0.08076358586549759, 0.10132158547639847, 0.10572687536478043, 0.12481644749641418, 0.09985315799713135, 0.09104258567094803, 0.11306901276111603, 0.1409691572189331, 0.1409691572189331, 0.14977973699569702, 0.09397944062948227, 0.10572687536478043, 0.12701909244060516, 0.13803230226039886, 0.13950073719024658, 0.20998531579971313, 0.16740088164806366, 0.10646108537912369, 0.1703377366065979, 0.15418502688407898, 0.14977973699569702, 0.10132158547639847, 0.17327459156513214, 0.22320117056369781, 0.17327459156513214, 0.16593244671821594, 0.19456681609153748, 0.17327459156513214, 0.19456681609153748, 0.1997063159942627, 0.1997063159942627, 0.20851688086986542, 0.21145375072956085, 0.22320117056369781, 0.2834067642688751, 0.20558002591133118, 0.2525697648525238, 0.2892804741859436, 0.2569750249385834, 0.2569750249385834, 0.5095447897911072, 0.5051395297050476, 0.37444934248924255, 0.2878120541572571, 0.2731277644634247, 0.39647576212882996, 0.42584434151649475, 0.43171805143356323, 0.5859031081199646, 0.4684287905693054, 0.453744500875473, NaN, 0.4757709205150604, 0.6182085275650024, 0.5550661087036133, 0.6328927874565125, 0.6710719466209412, 0.5462555289268494, 0.5638766288757324, 0.7283406853675842, 0.781204104423523, 0.8854625821113586], [0.014684287831187248, 0.029368575662374496, 0.049926578998565674, 0.049926578998565674, 0.051395006477832794, 0.046989720314741135, 0.06754772365093231, 0.07048457860946655, 0.046989720314741135, 0.055066078901290894, 0.05726872384548187, 0.08370044082403183, 0.08223201334476471, 0.12922173738479614, 0.12481644749641418, 0.1629955917596817, 0.14243759214878082, 0.09397944062948227, 0.0969163030385971, 0.18355359137058258, 0.11600587517023087, 0.10132158547639847, 0.13069015741348267, 0.20558002591133118, 0.07488986849784851, 0.1262848824262619, 0.14390602707862854, 0.14684288203716278, 0.17180617153644562, 0.1556534469127655, 0.14684288203716278, 0.1850220263004303, 0.23348017036914825, 0.23494860529899597, 0.26284876465797424, 0.20190896093845367, 0.20998531579971313, 0.23935389518737793, 0.24669604003429413, 0.2290748953819275, 0.2569750249385834, 0.2511013150215149, 0.09397944062948227, 0.26284876465797424, 0.30102789402008057, 0.17180617153644562, 0.29221731424331665, 0.3612334728240967, 0.3641703426837921, 0.3627019226551056, 0.23348017036914825, 0.311306893825531, 0.2731277644634247, 0.3171806037425995, 0.43318650126457214, 0.38179147243499756, 0.527165949344635, 0.527165949344635, 0.4669603407382965, 0.5242290496826172, NaN, 0.5873715281486511, 0.5565345287322998, 0.5991189479827881, 0.6417033672332764, 0.5712187886238098, 0.5462555289268494, 0.7048457860946655, 0.7679882645606995, 0.9662261605262756], [0.046989720314741135, 0.07048457860946655, 0.08663729578256607, 0.06020557880401611, 0.07488986849784851, 0.08516886830329895, 0.07342144101858139, 0.08516886830329895, 0.11306901276111603, 0.11747430264949799, 0.15859030187129974, 0.1666666716337204, 0.11674008518457413, 0.10719530284404755, 0.15859030187129974, 0.12481644749641418, 0.17180617153644562, 0.22613802552223206, 0.13950073719024658, 0.12775330245494843, 0.16740088164806366, 0.22026431560516357, 0.17180617153644562, 0.1850220263004303, 0.1556534469127655, 0.1483113020658493, 0.17914831638336182, 0.15859030187129974, 0.19530102610588074, 0.19089573621749878, 0.18942731618881226, 0.2657856047153473, 0.18061673641204834, 0.15271659195423126, 0.21365638077259064, 0.20264317095279694, 0.27239352464675903, 0.2569750249385834, 0.17914831638336182, 0.22320117056369781, 0.1923641711473465, 0.27165931463241577, 0.27239352464675903, 0.19089573621749878, 0.29221731424331665, 0.2878120541572571, 0.2378854602575302, 0.2804698944091797, 0.3465491831302643, 0.3553597629070282, 0.31424376368522644, 0.32158589363098145, 0.3171806037425995, 0.3274596035480499, 0.35976505279541016, 0.35829663276672363, 0.46402350068092346, 0.392070472240448, 0.3759177625179291, 0.38179147243499756, 0.41262850165367126, NaN, 0.38032305240631104, 0.6336269974708557, 0.47723934054374695, 0.4140969216823578, 0.4508076310157776, 0.784140944480896, 0.5756241083145142, 0.7767988443374634], [0.024963289499282837, 0.030837005004286766, 0.041116006672382355, 0.049926578998565674, 0.044052861630916595, 0.04845815151929855, 0.041116006672382355, 0.051395006477832794, 0.05286343768239021, 0.05726872384548187, 0.07342144101858139, 0.08370044082403183, 0.06167401000857353, 0.08516886830329895, 0.07709251344203949, 0.1409691572189331, 0.08370044082403183, 0.09985315799713135, 0.12334802001714706, 0.13509544730186462, 0.14390602707862854, 0.12187959253787994, 0.15418502688407898, 0.17621144652366638, 0.09544786810874939, 0.1336270123720169, 0.17180617153644562, 0.1189427301287651, 0.15051394701004028, 0.14243759214878082, 0.1776798814535141, 0.20998531579971313, 0.1850220263004303, 0.15859030187129974, 0.14757709205150604, 0.16740088164806366, 0.1703377366065979, 0.19089573621749878, 0.19089573621749878, 0.13215859234333038, 0.05433186516165733, 0.21585902571678162, 0.2525697648525238, 0.19676946103572845, 0.21732746064662933, 0.26431718468666077, 0.1556534469127655, 0.27165931463241577, 0.33773863315582275, 0.3406754732131958, 0.3171806037425995, 0.28487518429756165, 0.30690160393714905, 0.3201174736022949, 0.3994126319885254, 0.35095447301864624, 0.5110132098197937, 0.45521292090415955, 0.4434655010700226, 0.4449339210987091, 0.4434655010700226, 0.619676947593689, NaN, 0.6328927874565125, 0.5814977884292603, 0.48972100019454956, 0.5829662084579468, 0.7400881052017212, 0.7591776847839355, 0.922173261642456], [0.041116006672382355, 0.06461086869239807, 0.06754772365093231, 0.055800292640924454, 0.06314244121313095, 0.06754772365093231, 0.06754772365093231, 0.07342144101858139, 0.11160058528184891, 0.11013215780258179, 0.14243759214878082, 0.1556534469127655, 0.09618208557367325, 0.09471365809440613, 0.151982381939888, 0.10866373032331467, 0.15271659195423126, 0.20264317095279694, 0.12334802001714706, 0.1189427301287651, 0.15418502688407898, 0.19089573621749878, 0.16079294681549072, 0.18208517134189606, 0.14390602707862854, 0.13142438232898712, 0.16886931657791138, 0.14243759214878082, 0.17327459156513214, 0.17327459156513214, 0.1703377366065979, 0.2540381848812103, 0.16519823670387268, 0.13950073719024658, 0.18355359137058258, 0.18428781628608704, 0.24082231521606445, 0.23935389518737793, 0.16446402668952942, 0.1923641711473465, 0.16886931657791138, 0.26431718468666077, 0.2621145248413086, 0.16519823670387268, 0.2709251046180725, 0.2804698944091797, 0.2195301055908203, 0.2687224745750427, 0.32158589363098145, 0.3245227634906769, 0.3098384737968445, 0.3054331839084625, 0.30102789402008057, 0.32158589363098145, 0.3465491831302643, 0.33773863315582275, 0.4522760510444641, 0.392070472240448, 0.36563876271247864, 0.36710718274116516, 0.4008810520172119, 0.3663729727268219, 0.36710718274116516, NaN, 0.45154184103012085, 0.3935389220714569, 0.44640234112739563, 0.7165932655334473, 0.5609397888183594, 0.769456684589386], [0.030837005004286766, 0.05286343768239021, 0.05726872384548187, 0.04845815151929855, 0.07195301353931427, 0.05433186516165733, 0.0969163030385971, 0.08370044082403183, 0.08810572326183319, 0.07048457860946655, 0.07856094092130661, 0.08076358586549759, 0.11306901276111603, 0.12334802001714706, 0.12775330245494843, 0.13803230226039886, 0.11160058528184891, 0.09838473051786423, 0.1189427301287651, 0.15124815702438354, 0.1189427301287651, 0.14390602707862854, 0.12922173738479614, 0.22320117056369781, 0.10792951285839081, 0.13656388223171234, 0.15859030187129974, 0.12481644749641418, 0.15859030187129974, 0.15124815702438354, 0.16226138174533844, 0.20998531579971313, 0.19676946103572845, 0.17180617153644562, 0.20411160588264465, 0.13950073719024658, 0.22026431560516357, 0.2378854602575302, 0.19530102610588074, 0.1776798814535141, 0.20117473602294922, 0.2569750249385834, 0.12334802001714706, 0.17474302649497986, 0.24596181511878967, 0.15271659195423126, 0.230543315410614, 0.3061673939228058, 0.2878120541572571, 0.2980910539627075, 0.22320117056369781, 0.2657856047153473, 0.26138031482696533, 0.2562408149242401, 0.40014684200286865, 0.3171806037425995, 0.41850221157073975, 0.4067547619342804, 0.4390602111816406, 0.32892805337905884, 0.35829663276672363, 0.5227606296539307, 0.41850221157073975, 0.5484581589698792, NaN, 0.39794421195983887, 0.5139500498771667, 0.6754772663116455, 0.8641703128814697, 0.9691630005836487], [0.016152717173099518, 0.017621144652366638, 0.032305434346199036, 0.038179147988557816, 0.041116006672382355, 0.026431718841195107, 0.051395006477832794, 0.051395006477832794, 0.032305434346199036, 0.039647575467824936, 0.051395006477832794, 0.07048457860946655, 0.07048457860946655, 0.09397944062948227, 0.08370044082403183, 0.12187959253787994, 0.09985315799713135, 0.10572687536478043, 0.06167401000857353, 0.12922173738479614, 0.06754772365093231, 0.08370044082403183, 0.07782672345638275, 0.17914831638336182, 0.05433186516165733, 0.08810572326183319, 0.07782672345638275, 0.10425844043493271, 0.14243759214878082, 0.11600587517023087, 0.09544786810874939, 0.17327459156513214, 0.17474302649497986, 0.18942731618881226, 0.23201175034046173, 0.16152717173099518, 0.17914831638336182, 0.20117473602294922, 0.21292217075824738, 0.20264317095279694, 0.2437591850757599, 0.2364170402288437, 0.12187959253787994, 0.24229075014591217, 0.27606460452079773, 0.16226138174533844, 0.2804698944091797, 0.27753305435180664, 0.3494860529899597, 0.3465491831302643, 0.1189427301287651, 0.257709264755249, 0.21732746064662933, 0.2547723948955536, 0.39794421195983887, 0.36710718274116516, 0.5447870492935181, 0.458149790763855, 0.4287812113761902, 0.453744500875473, 0.4287812113761902, 0.5859031081199646, 0.5102789998054504, 0.6064611077308655, 0.6020557880401611, NaN, 0.5286343693733215, 0.700440526008606, 0.7444933652877808, 0.9353891611099243], [0.014684287831187248, 0.026431718841195107, 0.030837005004286766, 0.020558003336191177, 0.030837005004286766, 0.032305434346199036, 0.024963289499282837, 0.030837005004286766, 0.029368575662374496, 0.036710720509290695, 0.051395006477832794, 0.05286343768239021, 0.04845815151929855, 0.039647575467824936, 0.045521292835474014, 0.11306901276111603, 0.06754772365093231, 0.09104258567094803, 0.04845815151929855, 0.06167401000857353, 0.08076358586549759, 0.08076358586549759, 0.07635829597711563, 0.15418502688407898, 0.08076358586549759, 0.05433186516165733, 0.09251101315021515, 0.09544786810874939, 0.12481644749641418, 0.08663729578256607, 0.09397944062948227, 0.1703377366065979, 0.12775330245494843, 0.08957415819168091, 0.1262848824262619, 0.15271659195423126, 0.14977973699569702, 0.13215859234333038, 0.11013215780258179, 0.13509544730186462, 0.17914831638336182, 0.14170338213443756, 0.2364170402288437, 0.13950073719024658, 0.1923641711473465, 0.24963289499282837, 0.23935389518737793, 0.18942731618881226, 0.3700440526008606, 0.3715124726295471, 0.2966226041316986, 0.22026431560516357, 0.27019089460372925, 0.3171806037425995, 0.1776798814535141, 0.35242289304733276, 0.4684287905693054, 0.4140969216823578, 0.28854626417160034, 0.4361233413219452, 0.453744500875473, 0.5491923689842224, 0.41703376173973083, 0.553597629070282, 0.48604992032051086, 0.47136563062667847, NaN, 0.6380323171615601, 0.6387665271759033, 0.8428781032562256], [0.029368575662374496, 0.038179147988557816, 0.047723934054374695, 0.035976506769657135, 0.055066078901290894, 0.046989720314741135, 0.035242289304733276, 0.049192365258932114, 0.09544786810874939, 0.051395006477832794, 0.09985315799713135, 0.10279001295566559, 0.07048457860946655, 0.07488986849784851, 0.12041115760803223, 0.08516886830329895, 0.07415565103292465, 0.11453744769096375, 0.07635829597711563, 0.08957415819168091, 0.12481644749641418, 0.14390602707862854, 0.13069015741348267, 0.16152717173099518, 0.115271657705307, 0.08810572326183319, 0.13215859234333038, 0.09838473051786423, 0.13215859234333038, 0.11306901276111603, 0.13803230226039886, 0.22026431560516357, 0.12775330245494843, 0.11160058528184891, 0.11600587517023087, 0.10425844043493271, 0.18061673641204834, 0.1556534469127655, 0.13069015741348267, 0.12334802001714706, 0.12041115760803223, 0.17180617153644562, 0.18942731618881226, 0.10866373032331467, 0.14537444710731506, 0.19676946103572845, 0.1483113020658493, 0.17327459156513214, 0.21879589557647705, 0.21879589557647705, 0.2268722504377365, 0.19823788106441498, 0.1997063159942627, 0.23348017036914825, 0.2819383144378662, 0.22026431560516357, 0.3421438932418823, 0.2951541841030121, 0.29955947399139404, 0.27165931463241577, 0.2951541841030121, 0.21585902571678162, 0.2599118947982788, 0.2834067642688751, 0.3245227634906769, 0.29955947399139404, 0.36196768283843994, NaN, 0.4757709205150604, 0.7224669456481934], [0.033773861825466156, 0.041116006672382355, 0.042584434151649475, 0.042584434151649475, 0.06167401000857353, 0.046989720314741135, 0.06314244121313095, 0.05873715132474899, 0.06461086869239807, 0.051395006477832794, 0.06754772365093231, 0.07488986849784851, 0.07488986849784851, 0.08663729578256607, 0.09985315799713135, 0.10572687536478043, 0.051395006477832794, 0.07635829597711563, 0.07856094092130661, 0.11013215780258179, 0.08663729578256607, 0.10792951285839081, 0.08370044082403183, 0.1336270123720169, 0.08076358586549759, 0.08810572326183319, 0.10205580294132233, 0.08663729578256607, 0.11306901276111603, 0.08076358586549759, 0.10719530284404755, 0.1776798814535141, 0.1189427301287651, 0.11967694759368896, 0.12775330245494843, 0.10866373032331467, 0.151982381939888, 0.13656388223171234, 0.12041115760803223, 0.11160058528184891, 0.1262848824262619, 0.1629955917596817, 0.09104258567094803, 0.10866373032331467, 0.17180617153644562, 0.08663729578256607, 0.15859030187129974, 0.15859030187129974, 0.16593244671821594, 0.17180617153644562, 0.13656388223171234, 0.18061673641204834, 0.18649046123027802, 0.14537444710731506, 0.2599118947982788, 0.20558002591133118, 0.303964763879776, 0.2819383144378662, 0.2819383144378662, 0.21879589557647705, 0.23201175034046173, 0.4243759214878082, 0.24082231521606445, 0.4390602111816406, 0.13582965731620789, 0.25550660490989685, 0.3612334728240967, 0.5242290496826172, NaN, 0.9265785813331604], [0.007342143915593624, 0.005873715039342642, 0.007342143915593624, 0.005873715039342642, 0.010279001668095589, 0.008810572326183319, 0.017621144652366638, 0.016152717173099518, 0.007342143915593624, 0.011747430078685284, 0.007342143915593624, 0.010279001668095589, 0.011747430078685284, 0.029368575662374496, 0.020558003336191177, 0.023494860157370567, 0.007342143915593624, 0.008810572326183319, 0.016152717173099518, 0.038179147988557816, 0.014684287831187248, 0.019089573994278908, 0.016152717173099518, 0.017621144652366638, 0.013215859420597553, 0.022026430815458298, 0.024963289499282837, 0.016152717173099518, 0.026431718841195107, 0.019089573994278908, 0.027900146320462227, 0.049926578998565674, 0.05726872384548187, 0.044052861630916595, 0.04845815151929855, 0.023494860157370567, 0.045521292835474014, 0.031571220606565475, 0.051395006477832794, 0.024963289499282837, 0.05433186516165733, 0.045521292835474014, 0.010279001668095589, 0.04845815151929855, 0.06314244121313095, 0.011747430078685284, 0.06461086869239807, 0.046989720314741135, 0.07635829597711563, 0.07635829597711563, 0.033773861825466156, 0.05359765142202377, 0.06607929617166519, 0.035242289304733276, 0.07195301353931427, 0.08370044082403183, 0.11013215780258179, 0.07782672345638275, 0.0969163030385971, 0.11453744769096375, 0.033773861825466156, 0.22320117056369781, 0.07782672345638275, 0.230543315410614, 0.030837005004286766, 0.06461086869239807, 0.15712188184261322, 0.27753305435180664, 0.07342144101858139, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9168103208},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8520398384},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8395688337},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8362914717},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8253101821},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8237247018},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7970163235},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.796228906},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7366404933},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7312136873},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7157632625},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6907042105},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6884057957},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6728064018},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6654004159},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6490135987},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6343293093},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.6337972697},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6279022745},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.615495114},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6070356896},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5946498109},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5910213025},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5904786222},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.589946583},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5883185419},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.558620103},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5542467388},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5499478603},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5489689059},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5388388755},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5269424758},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5267296614},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5143650652},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4994147593},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4918385177},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.485007128},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4811551649},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4751750409},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4741748104},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4654706428},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.457287877},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4525846479},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4466258056},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4313775556},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4276107162},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4086807531},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4070101502},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4025942217},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.4007427254},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3680435869},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3597544107},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3537104428},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3330141082},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3078699258},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3068164899},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.29505842},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2860243884},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2756921811},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2652322899},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2588052532},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.253123069},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.238523909},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2339590108},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2322990483},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2279256859},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1987380023},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.167251909},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1553555083},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0474791977}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_all/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_all/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..c1d3dc0f5c5da752c676aaf9eeaa7a495ddf03c7
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_all/datasets_small/winrate_matrix.csv
@@ -0,0 +1,71 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),TabPFN-2.6 (default),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),TabDPT-Turbo (default),TabDPT (tuned + ensembled),TabDPT (tuned),"AutoGluon 1.4 (best, 4h)",TabM (tuned + ensembled),RealMLP (tuned),LimiX (default),TabSwift (default),TabDPT (default),LightGBM (tuned + ensembled),TabM (tuned),CatBoost (tuned + ensembled),iLTM (tuned + ensembled),CatBoost (tuned),SAP-RPT-OSS (default),xRFM (tuned + ensembled),LightGBM (tuned),XGBoost (tuned + ensembled),xRFM (tuned),ModernNCA (tuned),CatBoost (default),XGBoost (tuned),ModernNCA (tuned + ensembled),TabM (default),EBM (tuned + ensembled),TorchMLP (tuned + ensembled),ModernNCA (default),iLTM (tuned),ChimeraBoost (tuned + ensembled),EBM (tuned),RealMLP (default),FastaiMLP (tuned + ensembled),ChimeraBoost (tuned),ExtraTrees (tuned + ensembled),EBM (default),TorchMLP (tuned),ExtraTrees (tuned),FastaiMLP (tuned),ChimeraBoost (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),RandomForest (tuned + ensembled),LightGBM (default),XGBoost (default),RandomForest (tuned),PerpetualBooster (tuned + ensembled),TorchMLP (default),TabSTAR (default),xRFM (default),PerpetualBooster (tuned),iLTM (default),ExtraTrees (default),Linear (tuned + ensembled),FastaiMLP (default),Linear (tuned),KNN (tuned + ensembled),RandomForest (default),PerpetualBooster (default),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.76798826,0.76798826,0.7900147,0.76284873,0.79441994,0.8061674,0.77386194,0.8619677,0.8634361,0.8839941,0.88986784,0.9162996,0.8854626,0.8722467,0.9067548,0.86930984,0.89133626,0.917768,0.91189426,0.928047,0.9295154,0.9368576,0.8575624,0.9192364,0.917768,0.938326,0.91042584,0.92511016,0.93538916,0.94126284,0.94566816,0.9192364,0.9133627,0.95594716,0.9265786,0.9441997,0.938326,0.92511016,0.9368576,0.9236417,0.9441997,0.94566816,0.92217326,0.96328926,0.9397944,0.94273126,0.958884,0.8472834,0.845815,0.96622616,0.96622616,0.969163,0.9647577,0.98384726,0.9706314,0.8854626,0.9706314,0.98237884,0.948605,0.98531574,0.95301026,0.97503674,0.958884,0.969163,0.98384726,0.98531574,0.9706314,0.96622616,0.99265784
+TabPFN-3 (default),0.23201175,,0.56828195,0.53450805,0.5770925,0.6079295,0.66519827,0.6358297,0.7415565,0.74596184,0.76358294,0.7856094,0.7709251,0.7856094,0.7812041,0.76798826,0.8634361,0.81791484,0.83994126,0.8237885,0.8546255,0.87958884,0.8634361,0.76651984,0.8634361,0.86049926,0.8656388,0.8634361,0.8678414,0.87958884,0.8737151,0.8825257,0.845815,0.8619677,0.9148311,0.88913363,0.9287812,0.9060206,0.876652,0.917768,0.8744493,0.907489,0.9265786,0.8920705,0.93098384,0.928047,0.9140969,0.9192364,0.8215859,0.81938326,0.9339207,0.95301026,0.96622616,0.9500734,0.9500734,0.96328926,0.8678414,0.9647577,0.9603524,0.9544787,0.9706314,0.9295154,0.969163,0.93538916,0.9471366,0.98237884,0.97356826,0.96182084,0.958884,0.99412626
+"AutoGluon 1.5 (extreme, 4h)",0.23201175,0.43171805,,0.47136563,0.5227606,0.5976505,0.6049926,0.5770925,0.75403816,0.66079295,0.7386197,0.75770926,0.8017621,0.8164464,0.8002937,0.7606461,0.8105727,0.7488987,0.8428781,0.82452273,0.8472834,0.8781204,0.8524229,0.73274595,0.8575624,0.8546255,0.8722467,0.8450808,0.88105726,0.86857563,0.8751836,0.89721,0.86049926,0.86930984,0.91189426,0.8781204,0.9133627,0.91042584,0.8781204,0.8942731,0.8781204,0.917768,0.9089574,0.87958884,0.9368576,0.90014684,0.90014684,0.92070484,0.8032305,0.8032305,0.93245226,0.95301026,0.958884,0.9368576,0.95594716,0.96328926,0.84875184,0.95154184,0.96108663,0.9500734,0.9500734,0.9133627,0.958884,0.93245226,0.94273126,0.9676946,0.969163,0.95227605,0.9574156,0.99265784
+TabICLv2 (default),0.20998532,0.46549192,0.52863437,,0.53450805,0.5961821,0.61894274,0.58002937,0.7298091,0.7209985,0.76578563,0.75770926,0.79441994,0.7958884,0.76651984,0.7687225,0.8164464,0.78707784,0.82819384,0.80763584,0.83847284,0.8737151,0.8443465,0.74743026,0.8428781,0.85022026,0.8531571,0.83480173,0.8575624,0.87077826,0.856094,0.8634361,0.835536,0.86049926,0.8942731,0.88032305,0.8957416,0.89133626,0.8634361,0.8942731,0.8649045,0.9030837,0.87885463,0.88986784,0.9133627,0.88105726,0.89721,0.91189426,0.8164464,0.8105727,0.9265786,0.9397944,0.93245226,0.928047,0.948605,0.938326,0.87077826,0.9471366,0.95301026,0.938326,0.9500734,0.9397944,0.9500734,0.9441997,0.95154184,0.96182084,0.979442,0.9640235,0.9574156,0.99412626
+RealTabPFN-2.5 (tuned + ensembled),0.23715125,0.4229075,0.47723934,0.46549192,,0.5315712,0.6358297,0.6358297,0.71218795,0.6226138,0.66519827,0.6842878,0.7826725,0.7900147,0.7591777,0.75624084,0.7753304,0.6784141,0.825257,0.81938326,0.83994126,0.82745963,0.84875184,0.75697505,0.82819384,0.8450808,0.8619677,0.8311307,0.8634361,0.8737151,0.8744493,0.8546255,0.8575624,0.8634361,0.9030837,0.85903084,0.8986784,0.8964758,0.8773862,0.8825257,0.8773862,0.9060206,0.8575624,0.8737151,0.928047,0.86049926,0.907489,0.9162996,0.7958884,0.7958884,0.9133627,0.948605,0.9500734,0.92511016,0.95301026,0.9544787,0.8370044,0.9339207,0.95301026,0.94273126,0.948605,0.92511016,0.95594716,0.9368576,0.928047,0.958884,0.969163,0.9449339,0.938326,0.989721
+TabPFN-2.6 (default),0.20558003,0.39207047,0.40234947,0.40381792,0.4684288,,0.5814978,0.52349484,0.69162995,0.6284875,0.6593245,0.6769457,0.7694567,0.78414094,0.75770926,0.7312775,0.77679884,0.7004405,0.8223201,0.8135095,0.845815,0.86930984,0.8472834,0.75624084,0.8267254,0.8575624,0.8751836,0.83847284,0.856094,0.89133626,0.88105726,0.8957416,0.8524229,0.85022026,0.8957416,0.88105726,0.92070484,0.9192364,0.8744493,0.89133626,0.8649045,0.9236417,0.8928047,0.86930984,0.928047,0.8876652,0.8986784,0.93538916,0.80469894,0.80910426,0.9236417,0.9574156,0.96622616,0.9397944,0.948605,0.95594716,0.8428781,0.948605,0.94566816,0.94273126,0.95301026,0.9148311,0.95154184,0.93245226,0.94566816,0.97356826,0.9676946,0.95301026,0.95301026,0.9911894
+RealTabPFN-2.5 (default),0.1938326,0.33480176,0.39500734,0.38105726,0.36417034,0.4185022,,0.45080763,0.6255507,0.5844346,0.5932452,0.6262849,0.7195301,0.73421437,0.7092511,0.7283407,0.74302495,0.6637298,0.7753304,0.77826726,0.80763584,0.7826725,0.8223201,0.7298091,0.80469894,0.8237885,0.82966226,0.8325991,0.8164464,0.856094,0.8531571,0.8105727,0.8516887,0.85903084,0.8883994,0.8531571,0.8876652,0.876652,0.87958884,0.87077826,0.8928047,0.8825257,0.8546255,0.886931,0.91189426,0.866373,0.9133627,0.91042584,0.79735684,0.79662263,0.8854626,0.9368576,0.93538916,0.92070484,0.938326,0.97209984,0.845815,0.9339207,0.94566816,0.9500734,0.93245226,0.9265786,0.958884,0.93245226,0.9030837,0.948605,0.97503674,0.9647577,0.9368576,0.98237884
+RealTabPFN-2.5 (tuned),0.22613803,0.36417034,0.4229075,0.41997063,0.36417034,0.47650513,0.54919237,,0.66226137,0.5917768,0.6182085,0.6468429,0.7298091,0.7503671,0.72246695,0.71512485,0.7393539,0.6578561,0.77386194,0.79809105,0.7958884,0.7826725,0.8017621,0.72540385,0.7995595,0.8135095,0.82819384,0.8135095,0.8105727,0.8267254,0.83994126,0.82745963,0.83994126,0.84875184,0.8825257,0.835536,0.88105726,0.8568282,0.85903084,0.87077826,0.86049926,0.8737151,0.8524229,0.86049926,0.91042584,0.85903084,0.90014684,0.8942731,0.76798826,0.7723935,0.886931,0.9397944,0.9368576,0.9030837,0.93024963,0.958884,0.82819384,0.92511016,0.94126284,0.9397944,0.9295154,0.9148311,0.948605,0.9265786,0.9162996,0.948605,0.969163,0.95080763,0.94126284,0.98384726
+RealMLP (tuned + ensembled),0.1380323,0.25844347,0.24596182,0.2701909,0.28781205,0.30837005,0.37444934,0.33773863,,0.4787078,0.4735683,0.49853158,0.57782674,0.61967695,0.6791483,0.5315712,0.5638766,0.5565345,0.72246695,0.6901615,0.7591777,0.7606461,0.7606461,0.62114537,0.7488987,0.7687225,0.8267254,0.74302495,0.7312775,0.78854626,0.83847284,0.73715127,0.7914831,0.7929515,0.7709251,0.7753304,0.8414097,0.81938326,0.80469894,0.8817915,0.8201175,0.8259912,0.8311307,0.82745963,0.8208517,0.8450808,0.8362702,0.84875184,0.7386197,0.74302495,0.91042584,0.92217326,0.93538916,0.93538916,0.93538916,0.92511016,0.80469894,0.9265786,0.94273126,0.9192364,0.95301026,0.886931,0.9471366,0.8883994,0.91189426,0.9676946,0.9706314,0.9045521,0.93538916,0.99265784
+TabDPT-Turbo (default),0.13656388,0.25403818,0.33920705,0.27900147,0.37738618,0.37151247,0.41556534,0.4082232,0.5212922,,0.52422905,0.5653451,0.5932452,0.5851689,0.592511,0.6358297,0.6879589,0.68502206,0.6578561,0.6387665,0.69162995,0.7298091,0.6886931,0.6387665,0.7004405,0.69309837,0.7209985,0.73568285,0.7312775,0.72393537,0.73274595,0.7312775,0.7180617,0.73274595,0.77826726,0.8032305,0.82452273,0.78854626,0.76578563,0.8061674,0.8237885,0.8032305,0.84948605,0.79735684,0.82966226,0.86930984,0.86930984,0.8340675,0.73274595,0.73568285,0.8883994,0.886931,0.8928047,0.89133626,0.90014684,0.91189426,0.81497794,0.90014684,0.92070484,0.8986784,0.9449339,0.8825257,0.94273126,0.88986784,0.9295154,0.9603524,0.96328926,0.948605,0.948605,0.9882526
+TabDPT (tuned + ensembled),0.116005875,0.23641704,0.2613803,0.2342144,0.33480176,0.34067547,0.40675476,0.38179147,0.52643174,0.47577092,,0.60939795,0.5932452,0.5807636,0.6057269,0.5734214,0.63069016,0.6549192,0.6578561,0.6262849,0.67107195,0.7386197,0.68281937,0.6270191,0.6982379,0.6886931,0.71292216,0.7187959,0.72393537,0.73274595,0.7400881,0.7591777,0.7070485,0.72320116,0.75550663,0.76725405,0.8208517,0.7826725,0.74229074,0.7826725,0.7753304,0.79882526,0.8237885,0.75770926,0.80910426,0.8325991,0.8164464,0.825257,0.7209985,0.72246695,0.8883994,0.8825257,0.8825257,0.90014684,0.89721,0.9030837,0.8032305,0.91116005,0.9082232,0.8942731,0.94273126,0.8414097,0.9265786,0.8575624,0.92143905,0.948605,0.948605,0.90014684,0.93245226,0.99265784
+TabDPT (tuned),0.11013216,0.2143906,0.24229075,0.24229075,0.31571218,0.32305434,0.37371513,0.35315713,0.5014684,0.43465492,0.39060205,,0.5829662,0.55947137,0.5638766,0.5550661,0.5961821,0.62041116,0.6284875,0.6123348,0.64170337,0.68281937,0.6527166,0.6049926,0.6563877,0.6637298,0.6997063,0.6784141,0.69603527,0.6975037,0.7077827,0.7276065,0.69162995,0.70337737,0.74596184,0.74596184,0.7856094,0.7591777,0.72687227,0.7591777,0.75330395,0.7709251,0.7856094,0.74743026,0.7812041,0.79735684,0.79735684,0.81497794,0.71145374,0.7092511,0.8414097,0.8575624,0.8443465,0.8619677,0.87958884,0.88986784,0.8002937,0.8854626,0.8883994,0.8751836,0.9162996,0.8333333,0.9162996,0.8443465,0.9192364,0.9295154,0.9471366,0.89721,0.92511016,0.989721
+"AutoGluon 1.4 (best, 4h)",0.08370044,0.2290749,0.19823788,0.20558003,0.21732746,0.23054332,0.2804699,0.2701909,0.42217326,0.40675476,0.40675476,0.41703376,,0.5227606,0.5315712,0.4845815,0.5293686,0.50220263,0.6901615,0.5947136,0.6593245,0.6637298,0.6798825,0.5535976,0.6468429,0.7298091,0.7415565,0.69162995,0.67254037,0.7386197,0.7386197,0.69309837,0.70631427,0.73421437,0.79441994,0.7503671,0.7900147,0.78634363,0.75770926,0.7621145,0.79441994,0.7958884,0.7415565,0.8002937,0.8325991,0.76798826,0.83847284,0.835536,0.73274595,0.73568285,0.82966226,0.9148311,0.9030837,0.8612335,0.9265786,0.907489,0.79882526,0.8854626,0.9295154,0.90014684,0.917768,0.8832599,0.938326,0.9038179,0.886931,0.9295154,0.95154184,0.9295154,0.92511016,0.9882526
+TabM (tuned + ensembled),0.11453745,0.2143906,0.18355359,0.2041116,0.20998532,0.21585903,0.2657856,0.2496329,0.38032305,0.41483113,0.41923642,0.44052863,0.47723934,,0.4948605,0.4698972,0.49632892,0.51248163,0.6049926,0.70190895,0.61967695,0.61894274,0.64170337,0.5756241,0.6035242,0.64977974,0.7048458,0.6328928,0.6453745,0.7180617,0.73568285,0.6740088,0.83773863,0.73715127,0.74743026,0.7180617,0.73421437,0.7621145,0.76651984,0.7907489,0.75477237,0.7958884,0.71365637,0.7790015,0.8201175,0.71659327,0.80910426,0.845815,0.7283407,0.7312775,0.7797357,0.8729809,0.886931,0.8208517,0.87958884,0.91042584,0.7958884,0.8531571,0.90014684,0.9089574,0.87077826,0.8928047,0.9148311,0.9052864,0.876652,0.9060206,0.9603524,0.92511016,0.9133627,0.9706314
+RealMLP (tuned),0.1277533,0.2187959,0.19970632,0.23348017,0.24082232,0.24229075,0.2907489,0.27753305,0.32085168,0.40748897,0.39427313,0.43612334,0.4684288,0.5051395,,0.47063142,0.5007342,0.5051395,0.6049926,0.5976505,0.6284875,0.6116006,0.6372981,0.55800295,0.6049926,0.65051395,0.7077827,0.6468429,0.6461087,0.6872247,0.7246696,0.6637298,0.71512485,0.71512485,0.72540385,0.72393537,0.74596184,0.7209985,0.7503671,0.7797357,0.74449337,0.7518355,0.7518355,0.74743026,0.7914831,0.7694567,0.8002937,0.78707784,0.72246695,0.72393537,0.835536,0.85903084,0.86930984,0.85903084,0.89133626,0.8957416,0.7856094,0.8678414,0.8957416,0.886931,0.8751836,0.8414097,0.9229075,0.84801763,0.8722467,0.9162996,0.9544787,0.87958884,0.90014684,0.979442
+LimiX (default),0.09324523,0.23201175,0.2393539,0.23127754,0.24375919,0.26872247,0.2716593,0.28487518,0.4684288,0.36417034,0.42657855,0.44493392,0.5154185,0.5301028,0.5293686,,0.4992658,0.46696034,0.5976505,0.55726874,0.58883995,0.62922174,0.5917768,0.5359765,0.5947136,0.61967695,0.6358297,0.6013216,0.67254037,0.67107195,0.65198237,0.6593245,0.6182085,0.6975037,0.6872247,0.7092511,0.68281937,0.73421437,0.7298091,0.6989721,0.6997063,0.7503671,0.7415565,0.73421437,0.73274595,0.7415565,0.76284873,0.76651984,0.7378855,0.73274595,0.8002937,0.8105727,0.8208517,0.81791484,0.8237885,0.81938326,0.78340673,0.8634361,0.8325991,0.85903084,0.8370044,0.8751836,0.85903084,0.89133626,0.8619677,0.8781204,0.886931,0.9148311,0.8942731,0.97650516
+TabSwift (default),0.13069016,0.13656388,0.18942732,0.18355359,0.2246696,0.22320117,0.25697502,0.2606461,0.43612334,0.3120411,0.36930984,0.40381792,0.47063142,0.50367105,0.4992658,0.5007342,,0.48311308,0.51395005,0.54919237,0.52863437,0.5638766,0.561674,0.5653451,0.5535976,0.54919237,0.5829662,0.59911895,0.6255507,0.5976505,0.6255507,0.61086637,0.6431718,0.6372981,0.70337737,0.7004405,0.7092511,0.65051395,0.67254037,0.74743026,0.7709251,0.68281937,0.6975037,0.6945668,0.75477237,0.71512485,0.79441994,0.76358294,0.7415565,0.74302495,0.7503671,0.8215859,0.7914831,0.79441994,0.866373,0.87958884,0.8120411,0.8414097,0.88105726,0.85903084,0.8575624,0.82819384,0.9162996,0.8472834,0.8883994,0.90014684,0.93245226,0.9258444,0.948605,0.99265784
+TabDPT (default),0.10866373,0.18208517,0.25110132,0.21292217,0.3215859,0.29955947,0.33627018,0.3421439,0.4434655,0.31497797,0.34508076,0.37958884,0.49779737,0.48751837,0.4948605,0.5330396,0.51688695,,0.5330396,0.52349484,0.5712188,0.5961821,0.5829662,0.5770925,0.59030837,0.56975037,0.6167401,0.6020558,0.6328928,0.5961821,0.6372981,0.6372981,0.6328928,0.6035242,0.66226137,0.6886931,0.6945668,0.6798825,0.6277533,0.654185,0.6886931,0.6842878,0.7518355,0.6798825,0.71218795,0.7723935,0.74302495,0.7400881,0.67107195,0.67254037,0.77826726,0.7914831,0.7650514,0.81791484,0.8370044,0.82966226,0.7812041,0.8443465,0.86049926,0.85022026,0.9060206,0.77386194,0.90014684,0.79735684,0.90161526,0.8942731,0.9089574,0.8854626,0.9236417,0.9911894
+LightGBM (tuned + ensembled),0.08223201,0.16005874,0.15712188,0.17180617,0.17474303,0.17767988,0.2246696,0.22613803,0.27753305,0.3421439,0.3421439,0.37151247,0.30983847,0.39500734,0.39500734,0.40234947,0.48604992,0.46696034,,0.4750367,0.5256975,0.52496326,0.551395,0.48751837,0.5359765,0.6218796,0.6784141,0.5947136,0.56975037,0.6894273,0.7048458,0.5712188,0.6556535,0.6762115,0.6167401,0.66226137,0.66079295,0.7378855,0.71512485,0.69162995,0.6886931,0.76798826,0.7099853,0.7856094,0.70631427,0.71365637,0.7606461,0.84875184,0.71292216,0.71365637,0.81497794,0.95154184,0.8986784,0.86049926,0.90014684,0.82819384,0.79441994,0.81791484,0.917768,0.9060206,0.9030837,0.86049926,0.876652,0.876652,0.88105726,0.938326,0.95154184,0.9236417,0.92143905,0.98384726
+TabM (tuned),0.08810572,0.17621145,0.17547724,0.19236417,0.18061674,0.18649046,0.22173275,0.20190896,0.30983847,0.36123347,0.37371513,0.3876652,0.40528634,0.29809105,0.40234947,0.4427313,0.45080763,0.47650513,0.52496326,,0.5521292,0.5609398,0.56828195,0.4948605,0.53450805,0.5565345,0.6042584,0.5726872,0.5961821,0.6372981,0.6240822,0.61967695,0.6872247,0.6446402,0.69309837,0.6666667,0.6563877,0.7004405,0.68135095,0.7298091,0.69530106,0.69309837,0.66519827,0.7209985,0.7584435,0.6740088,0.76798826,0.7687225,0.7077827,0.71292216,0.74449337,0.79441994,0.82819384,0.7650514,0.8546255,0.8883994,0.7723935,0.8061674,0.8649045,0.8942731,0.8164464,0.8722467,0.8649045,0.88105726,0.84875184,0.87077826,0.938326,0.91042584,0.88986784,0.96182084
+CatBoost (tuned + ensembled),0.07195301,0.14537445,0.15271659,0.16152717,0.16005874,0.15418503,0.19236417,0.2041116,0.24082232,0.30837005,0.32892805,0.35829663,0.34067547,0.38032305,0.37151247,0.41116005,0.47136563,0.4287812,0.4743025,0.4478708,,0.50807637,0.5910426,0.51395005,0.5051395,0.5323054,0.6145374,0.5859031,0.56828195,0.6446402,0.6365639,0.5947136,0.60058737,0.6182085,0.6284875,0.6446402,0.6842878,0.68575627,0.67180616,0.67107195,0.6490455,0.7077827,0.71365637,0.71512485,0.71218795,0.7408223,0.7180617,0.7650514,0.7107195,0.70631427,0.8267254,0.8781204,0.88986784,0.87004405,0.8546255,0.835536,0.77386194,0.8414097,0.8737151,0.8729809,0.8839941,0.8325991,0.856094,0.845815,0.88105726,0.93245226,0.9192364,0.8751836,0.9133627,0.98531574
+iLTM (tuned + ensembled),0.07048458,0.12041116,0.12187959,0.12628488,0.17254038,0.13069016,0.21732746,0.21732746,0.2393539,0.2701909,0.2613803,0.3171806,0.33627018,0.38105726,0.38839942,0.37077826,0.43612334,0.40381792,0.4750367,0.4390602,0.49192363,,0.5227606,0.48751837,0.53377384,0.53891337,0.5734214,0.5954479,0.5873715,0.5726872,0.6079295,0.5844346,0.52716595,0.6035242,0.6226138,0.6798825,0.73348016,0.68281937,0.6475771,0.68575627,0.6886931,0.6901615,0.7041116,0.6798825,0.70337737,0.7312775,0.7496329,0.7378855,0.6681351,0.6688693,0.8105727,0.8619677,0.8472834,0.8428781,0.8516887,0.8443465,0.7694567,0.8781204,0.87958884,0.8619677,0.8986784,0.7797357,0.8781204,0.80910426,0.856094,0.9162996,0.9192364,0.856094,0.8920705,0.9809104
+CatBoost (tuned),0.06314244,0.13656388,0.14757709,0.15565345,0.15124816,0.15271659,0.17767988,0.19823788,0.2393539,0.3113069,0.3171806,0.3472834,0.32011747,0.35829663,0.36270192,0.4082232,0.438326,0.41703376,0.448605,0.43171805,0.40895742,0.47723934,,0.50220263,0.48091042,0.5110132,0.5917768,0.5653451,0.54919237,0.6116006,0.6240822,0.5756241,0.582232,0.60939795,0.6226138,0.6270191,0.6490455,0.6365639,0.6563877,0.6549192,0.6328928,0.6776799,0.6945668,0.7107195,0.6982379,0.70631427,0.7085169,0.75330395,0.71218795,0.7107195,0.8105727,0.8516887,0.8854626,0.84875184,0.8428781,0.81791484,0.7753304,0.82966226,0.8649045,0.86049926,0.86930984,0.82819384,0.845815,0.83920705,0.87077826,0.92217326,0.9236417,0.86930984,0.9162996,0.98384726
+SAP-RPT-OSS (default),0.14243759,0.23348017,0.26725402,0.25256976,0.24302496,0.24375919,0.2701909,0.27459618,0.37885463,0.36123347,0.37298092,0.39500734,0.44640234,0.42437592,0.44199705,0.4640235,0.43465492,0.4229075,0.51248163,0.5051395,0.48604992,0.51248163,0.49779737,,0.53450805,0.5403818,0.5624082,0.5535976,0.5741556,0.5535976,0.56975037,0.59030837,0.5550661,0.60058737,0.66519827,0.6343612,0.6013216,0.5976505,0.6343612,0.67254037,0.6975037,0.6167401,0.6681351,0.6629956,0.69162995,0.6681351,0.73274595,0.6681351,0.6387665,0.62995595,0.74522763,0.7312775,0.7283407,0.75403816,0.7856094,0.7856094,0.7606461,0.7900147,0.7922173,0.7900147,0.79441994,0.81497794,0.8237885,0.81791484,0.77679884,0.8208517,0.845815,0.83847284,0.866373,0.98237884
+xRFM (tuned + ensembled),0.080763586,0.13656388,0.14243759,0.15712188,0.17180617,0.17327459,0.19530103,0.20044053,0.25110132,0.29955947,0.3017621,0.34361234,0.35315713,0.39647576,0.39500734,0.40528634,0.44640234,0.40969163,0.4640235,0.46549192,0.4948605,0.46622613,0.5190896,0.46549192,,0.5220264,0.54845816,0.6145374,0.53377384,0.5866373,0.5770925,0.56828195,0.5462555,0.6020558,0.59030837,0.61967695,0.643906,0.6740088,0.6549192,0.6182085,0.6321586,0.6886931,0.6593245,0.6784141,0.6681351,0.70631427,0.70190895,0.71218795,0.6791483,0.6769457,0.76798826,0.78707784,0.8061674,0.80469894,0.8120411,0.84875184,0.7518355,0.9397944,0.83920705,0.8325991,0.92511016,0.8443465,0.9045521,0.856094,0.8920705,0.94566816,0.9192364,0.8847283,0.9192364,0.98678416
+LightGBM (tuned),0.08223201,0.13950074,0.14537445,0.14977974,0.15491924,0.14243759,0.17621145,0.18649046,0.23127754,0.3069016,0.3113069,0.33627018,0.2701909,0.35022026,0.34948605,0.38032305,0.45080763,0.43024963,0.37812042,0.4434655,0.46769458,0.46108663,0.4889868,0.4596182,0.47797358,,0.5602056,0.5359765,0.51835537,0.62922174,0.5932452,0.52863437,0.6064611,0.64023495,0.5976505,0.6240822,0.6020558,0.6563877,0.69309837,0.664464,0.64977974,0.68281937,0.67180616,0.7312775,0.6791483,0.6894273,0.7180617,0.7856094,0.7004405,0.70190895,0.7606461,0.9038179,0.8671072,0.81497794,0.8612335,0.80763584,0.77679884,0.8113069,0.8825257,0.8935389,0.8737151,0.8516887,0.866373,0.86857563,0.8634361,0.91189426,0.94566816,0.91189426,0.91189426,0.9779736
+XGBoost (tuned + ensembled),0.06167401,0.13436124,0.1277533,0.14684288,0.1380323,0.12481645,0.17033774,0.17180617,0.17327459,0.27900147,0.2870778,0.30029368,0.25844347,0.29515418,0.2922173,0.36417034,0.41703376,0.38325992,0.3215859,0.39574155,0.38546255,0.42657855,0.4082232,0.4375918,0.45154184,0.43979442,,0.53450805,0.5095448,0.54845816,0.57488984,0.4992658,0.54919237,0.5756241,0.57488984,0.6064611,0.59911895,0.6167401,0.6358297,0.6240822,0.6284875,0.65198237,0.6387665,0.68281937,0.6490455,0.68281937,0.68135095,0.71292216,0.6872247,0.68281937,0.7951542,0.8986784,0.9030837,0.8230543,0.81791484,0.78854626,0.7503671,0.8105727,0.8406755,0.82966226,0.856094,0.8208517,0.82819384,0.8311307,0.8414097,0.92217326,0.907489,0.8678414,0.8979442,0.97503674
+xRFM (tuned),0.08957416,0.13656388,0.15491924,0.16519824,0.16886932,0.16152717,0.16740088,0.18649046,0.25697502,0.26431718,0.2812041,0.3215859,0.30837005,0.36710718,0.35315713,0.39867842,0.40088105,0.3979442,0.40528634,0.42731276,0.41409692,0.40455213,0.43465492,0.44640234,0.38546255,0.4640235,0.46549192,,0.47723934,0.4904552,0.4955947,0.4743025,0.5095448,0.56754774,0.5741556,0.5961821,0.5873715,0.6013216,0.6123348,0.5756241,0.6013216,0.6475771,0.6240822,0.6446402,0.6372981,0.66519827,0.67107195,0.6886931,0.68281937,0.6842878,0.69603527,0.72687227,0.7209985,0.7386197,0.77312773,0.8414097,0.7503671,0.88986784,0.8157122,0.845815,0.8531571,0.8414097,0.88105726,0.8575624,0.8751836,0.8957416,0.9045521,0.90161526,0.9133627,0.98384726
+ModernNCA (tuned),0.07488987,0.13215859,0.11894273,0.14243759,0.13656388,0.14390603,0.18355359,0.18942732,0.26872247,0.26872247,0.2760646,0.30396476,0.3274596,0.35462555,0.35389134,0.3274596,0.37444934,0.36710718,0.43024963,0.40381792,0.43171805,0.4126285,0.45080763,0.42584434,0.46622613,0.48164463,0.4904552,0.5227606,,0.5256975,0.50734216,0.4441997,0.50807637,0.5741556,0.5565345,0.6534508,0.5550661,0.6182085,0.6123348,0.6270191,0.6424376,0.623348,0.6123348,0.6343612,0.6769457,0.6358297,0.70631427,0.67254037,0.67254037,0.6769457,0.73274595,0.71365637,0.74449337,0.7518355,0.76798826,0.78707784,0.74596184,0.8428781,0.8017621,0.85022026,0.82819384,0.80469894,0.84948605,0.8267254,0.8414097,0.8575624,0.8751836,0.8678414,0.886931,0.97356826
+CatBoost (default),0.06461087,0.12041116,0.13142438,0.12922174,0.12628488,0.10866373,0.14390603,0.17327459,0.21145375,0.2760646,0.26725402,0.3024963,0.2613803,0.2819383,0.31277534,0.32892805,0.40234947,0.40381792,0.31057268,0.36270192,0.35535976,0.42731276,0.38839942,0.44640234,0.4133627,0.37077826,0.45154184,0.5095448,0.4743025,,0.48091042,0.48604992,0.5330396,0.58002937,0.5609398,0.54478705,0.5873715,0.6321586,0.6461087,0.6270191,0.62114537,0.6637298,0.6490455,0.6945668,0.65051395,0.6666667,0.68281937,0.7709251,0.70117474,0.70190895,0.7621145,0.80469894,0.7812041,0.79735684,0.8267254,0.7914831,0.77826726,0.76798826,0.8619677,0.8986784,0.8443465,0.80910426,0.8575624,0.8267254,0.84875184,0.8839941,0.9133627,0.886931,0.9192364,0.9809104
+XGBoost (tuned),0.05873715,0.12628488,0.12481645,0.14390603,0.12555066,0.11894273,0.14684288,0.16005874,0.16152717,0.26725402,0.2599119,0.2922173,0.2613803,0.26431718,0.2753304,0.34801763,0.37444934,0.36270192,0.29515418,0.37591776,0.36343613,0.39207047,0.37591776,0.43024963,0.4229075,0.40675476,0.42511013,0.50440526,0.49265787,0.5190896,,0.4948605,0.5154185,0.5565345,0.5609398,0.5726872,0.5712188,0.56754774,0.6138032,0.5947136,0.6049926,0.6123348,0.6350955,0.6534508,0.6284875,0.6696035,0.6769457,0.67254037,0.67107195,0.6732746,0.7650514,0.85903084,0.8678414,0.8032305,0.79441994,0.77386194,0.74743026,0.8061674,0.8237885,0.8267254,0.8531571,0.8105727,0.8223201,0.82966226,0.83773863,0.9045521,0.9060206,0.8619677,0.8928047,0.97209984
+ModernNCA (tuned + ensembled),0.054331865,0.1174743,0.10279001,0.13656388,0.14537445,0.10425844,0.18942732,0.17254038,0.26284876,0.26872247,0.24082232,0.27239352,0.3069016,0.32599118,0.33627018,0.34067547,0.38913363,0.36270192,0.4287812,0.38032305,0.40528634,0.41556534,0.42437592,0.40969163,0.43171805,0.47136563,0.5007342,0.5256975,0.5558003,0.51395005,0.5051395,,0.47577092,0.5535976,0.5301028,0.6593245,0.5462555,0.56828195,0.5969163,0.62114537,0.60058737,0.60058737,0.60939795,0.60939795,0.61086637,0.6461087,0.66226137,0.6123348,0.6255507,0.6226138,0.7048458,0.7048458,0.7621145,0.7195301,0.7415565,0.74743026,0.68575627,0.8318649,0.7518355,0.77679884,0.81497794,0.73421437,0.7900147,0.74596184,0.7900147,0.8267254,0.82966226,0.7797357,0.8223201,0.9500734
+TabM (default),0.080763586,0.15418503,0.13950074,0.16446403,0.14243759,0.14757709,0.1483113,0.16005874,0.20851688,0.2819383,0.29295155,0.30837005,0.29368576,0.16226138,0.28487518,0.38179147,0.35682818,0.36710718,0.34434655,0.31277534,0.39941263,0.47283408,0.417768,0.44493392,0.4537445,0.39353892,0.45080763,0.4904552,0.49192363,0.46696034,0.4845815,0.52422905,,0.49632892,0.60058737,0.56828195,0.5660793,0.5212922,0.5609398,0.5961821,0.6182085,0.5462555,0.6152717,0.58957416,0.68135095,0.6358297,0.6989721,0.6123348,0.69603527,0.6989721,0.6872247,0.66226137,0.7004405,0.6989721,0.74375916,0.8105727,0.7753304,0.72687227,0.75330395,0.8267254,0.76651984,0.81938326,0.81497794,0.83480173,0.8032305,0.825257,0.8722467,0.8722467,0.88105726,0.94273126
+EBM (tuned + ensembled),0.086637296,0.1380323,0.13069016,0.13950074,0.13656388,0.14977974,0.14096916,0.15124816,0.20704846,0.26725402,0.2767988,0.2966226,0.2657856,0.26284876,0.28487518,0.3024963,0.36270192,0.39647576,0.32378855,0.35535976,0.38179147,0.39647576,0.39060205,0.39941263,0.3979442,0.35976505,0.42437592,0.43245226,0.42584434,0.41997063,0.4434655,0.44640234,0.50367105,,0.5403818,0.5212922,0.5007342,0.49779737,0.6997063,0.5653451,0.5323054,0.5301028,0.58957416,0.73274595,0.61013216,0.5961821,0.6071953,0.6049926,0.67254037,0.6696035,0.6666667,0.6593245,0.6864905,0.69162995,0.7753304,0.73568285,0.7723935,0.71512485,0.79809105,0.8340675,0.7650514,0.8472834,0.8414097,0.86049926,0.82819384,0.8105727,0.91042584,0.8883994,0.88032305,0.95594716
+TorchMLP (tuned + ensembled),0.04405286,0.08516887,0.08810572,0.105726875,0.0969163,0.10425844,0.111600585,0.1174743,0.2290749,0.22173275,0.2444934,0.25403818,0.20558003,0.25256976,0.27459618,0.31277534,0.2966226,0.33773863,0.38325992,0.3069016,0.37151247,0.37738618,0.37738618,0.33480176,0.40969163,0.40234947,0.42511013,0.42584434,0.4434655,0.4390602,0.4390602,0.4698972,0.39941263,0.4596182,,0.54772395,0.520558,0.5007342,0.5154185,0.5403818,0.6182085,0.5359765,0.6079295,0.55800295,0.71659327,0.6167401,0.71585906,0.5565345,0.6549192,0.6629956,0.6798825,0.65051395,0.6886931,0.6975037,0.7195301,0.8722467,0.7606461,0.75770926,0.77386194,0.8054332,0.73715127,0.78634363,0.8524229,0.8164464,0.7958884,0.76798826,0.8737151,0.8839941,0.8722467,0.95154184
+ModernNCA (default),0.07342144,0.110866375,0.12187959,0.11967695,0.14096916,0.11894273,0.14684288,0.16446403,0.2246696,0.19676946,0.23274596,0.25403818,0.2496329,0.2819383,0.2760646,0.2907489,0.29955947,0.3113069,0.33773863,0.33333334,0.35535976,0.32011747,0.37298092,0.36563876,0.38032305,0.37591776,0.39353892,0.40381792,0.34654918,0.45521292,0.42731276,0.34067547,0.43171805,0.4787078,0.45227605,,0.4530103,0.5814978,0.52716595,0.48604992,0.5403818,0.5932452,0.5668135,0.5756241,0.53891337,0.5976505,0.6049926,0.6358297,0.6490455,0.6534508,0.66519827,0.62995595,0.6284875,0.72393537,0.71512485,0.7606461,0.73568285,0.74449337,0.76358294,0.8267254,0.79809105,0.79735684,0.8325991,0.8157122,0.86049926,0.83847284,0.8472834,0.8957416,0.89133626,0.97650516
+iLTM (tuned),0.055800293,0.071218796,0.086637296,0.10425844,0.101321585,0.07929515,0.1123348,0.11894273,0.1585903,0.17547724,0.17914832,0.2143906,0.20998532,0.2657856,0.25403818,0.3171806,0.2907489,0.30543318,0.33920705,0.34361234,0.31571218,0.2665198,0.35095447,0.39867842,0.35609397,0.3979442,0.40088105,0.4126285,0.44493392,0.4126285,0.4287812,0.4537445,0.4339207,0.4992658,0.479442,0.54698974,,0.51248163,0.5535976,0.5403818,0.57856095,0.561674,0.5624082,0.5917768,0.58883995,0.6049926,0.6240822,0.5932452,0.6380323,0.6343612,0.67107195,0.6740088,0.7004405,0.7077827,0.71365637,0.78707784,0.71365637,0.7856094,0.7606461,0.8054332,0.7900147,0.7276065,0.82966226,0.7591777,0.7797357,0.8208517,0.85022026,0.81938326,0.84801763,0.9544787
+ChimeraBoost (tuned + ensembled),0.06167401,0.09397944,0.08957416,0.10866373,0.10352423,0.080763586,0.12334802,0.1431718,0.18061674,0.21145375,0.21732746,0.24082232,0.21365638,0.23788546,0.27900147,0.2657856,0.34948605,0.32011747,0.26211452,0.29955947,0.31424376,0.3171806,0.36343613,0.40234947,0.32599118,0.34361234,0.38325992,0.39867842,0.38179147,0.36784142,0.43245226,0.43171805,0.4787078,0.50220263,0.4992658,0.4185022,0.48751837,,0.54772395,0.5212922,0.50660795,0.5734214,0.5227606,0.58002937,0.5550661,0.5638766,0.55873716,0.7753304,0.66519827,0.6593245,0.64170337,0.7099853,0.67107195,0.6872247,0.7753304,0.69603527,0.75330395,0.7195301,0.8061674,0.8002937,0.7606461,0.74302495,0.80910426,0.7606461,0.7621145,0.79882526,0.8678414,0.8443465,0.8634361,0.9684288
+EBM (tuned),0.07488987,0.12334802,0.12187959,0.13656388,0.1226138,0.12555066,0.12041116,0.14096916,0.19530103,0.2342144,0.25770926,0.27312776,0.24229075,0.23348017,0.2496329,0.2701909,0.3274596,0.37224668,0.28487518,0.31864905,0.32819384,0.3524229,0.34361234,0.36563876,0.34508076,0.3069016,0.36417034,0.3876652,0.3876652,0.35389134,0.38619676,0.4030837,0.4390602,0.30029368,0.4845815,0.47283408,0.44640234,0.45227605,,0.5007342,0.4889868,0.47650513,0.53891337,0.633627,0.5756241,0.5712188,0.5763583,0.5535976,0.6549192,0.6571219,0.6270191,0.60939795,0.64170337,0.67107195,0.7400881,0.71218795,0.75477237,0.69162995,0.76798826,0.8002937,0.75330395,0.8208517,0.80910426,0.835536,0.80469894,0.78707784,0.88986784,0.86930984,0.87958884,0.948605
+RealMLP (default),0.06314244,0.08223201,0.105726875,0.105726875,0.1174743,0.10866373,0.12922174,0.12922174,0.11820852,0.1938326,0.21732746,0.24082232,0.23788546,0.2092511,0.22026432,0.3010279,0.25256976,0.34581497,0.30837005,0.2701909,0.32892805,0.31424376,0.34508076,0.3274596,0.38179147,0.33553597,0.37591776,0.42437592,0.37298092,0.37298092,0.40528634,0.37885463,0.40381792,0.43465492,0.4596182,0.51395005,0.4596182,0.4787078,0.4992658,,0.51762116,0.49779737,0.5638766,0.5301028,0.5873715,0.5976505,0.6079295,0.5521292,0.64023495,0.6490455,0.66079295,0.6358297,0.6490455,0.6901615,0.7092511,0.78634363,0.7386197,0.72687227,0.76798826,0.7914831,0.7709251,0.77679884,0.8678414,0.80763584,0.8223201,0.79735684,0.8649045,0.876652,0.8883994,0.97503674
+FastaiMLP (tuned + ensembled),0.076358296,0.12555066,0.12187959,0.13509545,0.1226138,0.13509545,0.1071953,0.13950074,0.17988253,0.17621145,0.2246696,0.24669604,0.20558003,0.2452276,0.2555066,0.30029368,0.2290749,0.3113069,0.3113069,0.30469897,0.35095447,0.3113069,0.36710718,0.3024963,0.36784142,0.35022026,0.37151247,0.39867842,0.35756242,0.37885463,0.39500734,0.39941263,0.38179147,0.46769458,0.38179147,0.4596182,0.42143905,0.49339208,0.5110132,0.48237884,,0.5154185,0.51835537,0.52716595,0.5154185,0.53450805,0.6563877,0.5433186,0.6138032,0.6182085,0.643906,0.5704846,0.6064611,0.6578561,0.65198237,0.72613806,0.72246695,0.7107195,0.6901615,0.78854626,0.74302495,0.80763584,0.94566816,0.8311307,0.79882526,0.75624084,0.8208517,0.87958884,0.8737151,0.94566816
+ChimeraBoost (tuned),0.055800293,0.09251101,0.08223201,0.0969163,0.09397944,0.076358296,0.1174743,0.12628488,0.17400882,0.19676946,0.20117474,0.2290749,0.2041116,0.2041116,0.24816446,0.2496329,0.3171806,0.31571218,0.23201175,0.3069016,0.2922173,0.30983847,0.3223201,0.38325992,0.3113069,0.3171806,0.34801763,0.3524229,0.37665197,0.33627018,0.3876652,0.39941263,0.4537445,0.4698972,0.4640235,0.40675476,0.438326,0.42657855,0.52349484,0.50220263,0.4845815,,0.50220263,0.54772395,0.5256975,0.54772395,0.53817916,0.7077827,0.65198237,0.65051395,0.61086637,0.6666667,0.6483113,0.6593245,0.7305433,0.68575627,0.73568285,0.69603527,0.76284873,0.77679884,0.7488987,0.7283407,0.78414094,0.73568285,0.74302495,0.76358294,0.85829663,0.82819384,0.8370044,0.9544787
+ExtraTrees (tuned + ensembled),0.054331865,0.07342144,0.091042586,0.121145375,0.14243759,0.1071953,0.14537445,0.14757709,0.16886932,0.15051395,0.17621145,0.2143906,0.25844347,0.2863436,0.24816446,0.25844347,0.3024963,0.24816446,0.29001468,0.33480176,0.2863436,0.2958884,0.30543318,0.3318649,0.34067547,0.32819384,0.36123347,0.37591776,0.3876652,0.35095447,0.36490455,0.39060205,0.38472834,0.41042584,0.39207047,0.4331865,0.4375918,0.47723934,0.46108663,0.43612334,0.48164463,0.49779737,,0.49339208,0.45594713,0.6556535,0.5227606,0.52863437,0.5521292,0.5638766,0.7400881,0.55947137,0.56828195,0.79882526,0.6534508,0.62114537,0.6982379,0.7195301,0.68281937,0.71659327,0.9060206,0.7276065,0.74743026,0.7378855,0.876652,0.8781204,0.76358294,0.8105727,0.9089574,0.989721
+EBM (default),0.07782672,0.10792951,0.12041116,0.11013216,0.12628488,0.13069016,0.11306901,0.13950074,0.17254038,0.20264317,0.24229075,0.25256976,0.19970632,0.22099853,0.25256976,0.2657856,0.30543318,0.32011747,0.2143906,0.27900147,0.28487518,0.32011747,0.28928047,0.3370044,0.3215859,0.26872247,0.3171806,0.35535976,0.36563876,0.30543318,0.34654918,0.39060205,0.41042584,0.26725402,0.44199705,0.42437592,0.4082232,0.41997063,0.36637297,0.4698972,0.47283408,0.45227605,0.50660795,,0.54919237,0.51835537,0.54478705,0.5110132,0.6358297,0.6358297,0.5961821,0.5638766,0.5859031,0.6270191,0.69309837,0.7004405,0.73421437,0.6475771,0.7209985,0.79441994,0.73715127,0.80910426,0.8032305,0.83480173,0.825257,0.75770926,0.86049926,0.89133626,0.89133626,0.95154184
+TorchMLP (tuned),0.03671072,0.06901615,0.06314244,0.086637296,0.07195301,0.07195301,0.08810572,0.08957416,0.17914832,0.17033774,0.19089574,0.2187959,0.16740088,0.17988253,0.20851688,0.26725402,0.2452276,0.28781205,0.29368576,0.24155654,0.28781205,0.2966226,0.3017621,0.30837005,0.3318649,0.32085168,0.35095447,0.36270192,0.32305434,0.34948605,0.37151247,0.38913363,0.31864905,0.38986784,0.28340676,0.46108663,0.41116005,0.44493392,0.42437592,0.4126285,0.4845815,0.4743025,0.54405284,0.45080763,,0.5506608,0.57856095,0.5007342,0.6431718,0.6446402,0.6152717,0.561674,0.5829662,0.6387665,0.6431718,0.78487515,0.73715127,0.69603527,0.6945668,0.74743026,0.6989721,0.7077827,0.7826725,0.7290749,0.75403816,0.72393537,0.80763584,0.8546255,0.82819384,0.9368576
+ExtraTrees (tuned),0.06020558,0.07195301,0.09985316,0.11894273,0.13950074,0.1123348,0.13362701,0.14096916,0.15491924,0.13069016,0.16740088,0.20264317,0.23201175,0.28340676,0.23054332,0.25844347,0.28487518,0.22760646,0.2863436,0.32599118,0.25917768,0.26872247,0.29368576,0.3318649,0.29368576,0.31057268,0.3171806,0.33480176,0.36417034,0.33333334,0.33039647,0.35389134,0.36417034,0.40381792,0.38325992,0.40234947,0.39500734,0.43612334,0.4287812,0.40234947,0.46549192,0.45227605,0.34434655,0.48164463,0.4493392,,0.50807637,0.5154185,0.55433184,0.5624082,0.6387665,0.5212922,0.50807637,0.7143906,0.6328928,0.60058737,0.6945668,0.68135095,0.6637298,0.7107195,0.82819384,0.71218795,0.73568285,0.7195301,0.8472834,0.83773863,0.7503671,0.8032305,0.9133627,0.9882526
+FastaiMLP (tuned),0.057268724,0.085903086,0.09985316,0.10279001,0.09251101,0.101321585,0.086637296,0.09985316,0.1637298,0.13069016,0.18355359,0.20264317,0.16152717,0.19089574,0.19970632,0.23715125,0.20558003,0.25697502,0.2393539,0.23201175,0.2819383,0.2503671,0.2914831,0.26725402,0.29809105,0.2819383,0.31864905,0.32892805,0.29368576,0.3171806,0.32305434,0.33773863,0.3010279,0.3928047,0.28414097,0.39500734,0.37591776,0.44126284,0.4236417,0.39207047,0.34361234,0.46182084,0.47723934,0.45521292,0.42143905,0.49192363,,0.5051395,0.5829662,0.58002937,0.5851689,0.50367105,0.52422905,0.59911895,0.5932452,0.654185,0.69603527,0.6431718,0.6387665,0.74302495,0.7077827,0.7621145,0.8443465,0.7804699,0.7694567,0.7195301,0.7606461,0.8516887,0.8414097,0.93538916
+ChimeraBoost (default),0.041116007,0.080763586,0.07929515,0.08810572,0.08370044,0.06461087,0.08957416,0.105726875,0.15124816,0.16593245,0.17474303,0.18502203,0.16446403,0.15418503,0.21292217,0.23348017,0.23641704,0.2599119,0.15124816,0.23127754,0.2349486,0.26211452,0.24669604,0.3318649,0.28781205,0.2143906,0.2870778,0.3113069,0.3274596,0.2290749,0.3274596,0.3876652,0.3876652,0.39500734,0.4434655,0.36417034,0.40675476,0.2246696,0.44640234,0.4478708,0.45668134,0.2922173,0.47136563,0.4889868,0.4992658,0.4845815,0.4948605,,0.6226138,0.6270191,0.5859031,0.5763583,0.5910426,0.6064611,0.67254037,0.63142437,0.7048458,0.6431718,0.7077827,0.74302495,0.6387665,0.7195301,0.7283407,0.7312775,0.6938326,0.72246695,0.8105727,0.8267254,0.8414097,0.95301026
+TabSTAR (tuned),0.15271659,0.17841409,0.19676946,0.18355359,0.2041116,0.19530103,0.20264317,0.23201175,0.2613803,0.26725402,0.27900147,0.28854626,0.26725402,0.2716593,0.27753305,0.26211452,0.25844347,0.32892805,0.2870778,0.2922173,0.28928047,0.3318649,0.28781205,0.36123347,0.32085168,0.29955947,0.31277534,0.3171806,0.3274596,0.29882526,0.32892805,0.37444934,0.30396476,0.3274596,0.34508076,0.35095447,0.36196768,0.33480176,0.34508076,0.35976505,0.38619676,0.34801763,0.4478708,0.36417034,0.35682818,0.44566813,0.41703376,0.37738618,,0.53377384,0.49339208,0.4185022,0.42437592,0.53744495,0.56828195,0.46549192,0.9441997,0.53744495,0.5947136,0.4904552,0.6358297,0.6534508,0.66226137,0.6784141,0.71218795,0.65051395,0.62995595,0.7812041,0.8340675,0.9236417
+TabSTAR (tuned + ensembled),0.15418503,0.18061674,0.19676946,0.18942732,0.2041116,0.19089574,0.20337738,0.22760646,0.25697502,0.26431718,0.27753305,0.2907489,0.26431718,0.26872247,0.2760646,0.26725402,0.25697502,0.3274596,0.2863436,0.2870778,0.29368576,0.33113068,0.28928047,0.37004405,0.32305434,0.29809105,0.3171806,0.31571218,0.32305434,0.29809105,0.3267254,0.37738618,0.3010279,0.33039647,0.3370044,0.34654918,0.36563876,0.34067547,0.34287813,0.35095447,0.38179147,0.34948605,0.43612334,0.36417034,0.35535976,0.4375918,0.41997063,0.37298092,0.46622613,,0.49339208,0.42437592,0.42143905,0.5403818,0.5638766,0.46255508,0.9603524,0.53450805,0.58883995,0.4948605,0.6372981,0.6446402,0.6593245,0.67547727,0.70190895,0.6534508,0.6284875,0.7812041,0.82819384,0.9236417
+RandomForest (tuned + ensembled),0.03377386,0.066079296,0.06754772,0.07342144,0.086637296,0.076358296,0.11453745,0.11306901,0.08957416,0.111600585,0.111600585,0.1585903,0.17033774,0.22026432,0.16446403,0.19970632,0.2496329,0.22173275,0.18502203,0.2555066,0.17327459,0.18942732,0.18942732,0.2547724,0.23201175,0.2393539,0.20484582,0.30396476,0.26725402,0.23788546,0.2349486,0.29515418,0.31277534,0.33333334,0.32011747,0.33480176,0.32892805,0.35829663,0.37298092,0.33920705,0.35609397,0.38913363,0.2599119,0.40381792,0.38472834,0.36123347,0.41483113,0.41409692,0.50660795,0.50660795,,0.4801762,0.49339208,0.67547727,0.5814978,0.5418502,0.643906,0.6563877,0.6182085,0.6255507,0.76651984,0.68575627,0.68281937,0.6901615,0.77679884,0.88105726,0.70337737,0.77312773,0.8634361,0.96622616
+LightGBM (default),0.03377386,0.04698972,0.04698972,0.06020558,0.051395006,0.042584434,0.06314244,0.06020558,0.07782672,0.11306901,0.1174743,0.14243759,0.08516887,0.12701909,0.14096916,0.18942732,0.17841409,0.20851688,0.04845815,0.20558003,0.12187959,0.1380323,0.1483113,0.26872247,0.21292217,0.096182086,0.101321585,0.27312776,0.2863436,0.19530103,0.14096916,0.29515418,0.33773863,0.34067547,0.34948605,0.37004405,0.32599118,0.29001468,0.39060205,0.36417034,0.42951542,0.33333334,0.44052863,0.43612334,0.438326,0.4787078,0.49632892,0.4236417,0.5814978,0.5756241,0.5198238,,0.52496326,0.56901616,0.5829662,0.6255507,0.66226137,0.66519827,0.64023495,0.71218795,0.6886931,0.6784141,0.71512485,0.6945668,0.73421437,0.74229074,0.7797357,0.8017621,0.81938326,0.9464024
+XGBoost (default),0.030837005,0.03377386,0.041116007,0.06754772,0.04992658,0.03377386,0.06461087,0.06314244,0.06461087,0.1071953,0.1174743,0.15565345,0.0969163,0.11306901,0.13069016,0.17914832,0.20851688,0.2349486,0.101321585,0.17180617,0.11013216,0.15271659,0.11453745,0.2716593,0.1938326,0.1328928,0.0969163,0.27900147,0.2555066,0.2187959,0.13215859,0.23788546,0.29955947,0.31350955,0.3113069,0.37151247,0.29955947,0.32892805,0.35829663,0.35095447,0.39353892,0.35168868,0.43171805,0.41409692,0.41703376,0.49192363,0.47577092,0.40895742,0.5756241,0.57856095,0.50660795,0.4750367,,0.5712188,0.53891337,0.5873715,0.6696035,0.6901615,0.6049926,0.72687227,0.72687227,0.68281937,0.69309837,0.6989721,0.7386197,0.7826725,0.7298091,0.8002937,0.8135095,0.9339207
+RandomForest (tuned),0.03524229,0.04992658,0.06314244,0.07195301,0.07488987,0.06020558,0.07929515,0.0969163,0.06461087,0.10866373,0.09985316,0.1380323,0.13876653,0.17914832,0.14096916,0.18208517,0.20558003,0.18208517,0.13950074,0.2349486,0.12995595,0.15712188,0.15124816,0.24596182,0.19530103,0.18502203,0.17694567,0.2613803,0.24816446,0.20264317,0.19676946,0.2804699,0.3010279,0.30837005,0.3024963,0.2760646,0.2922173,0.31277534,0.32892805,0.30983847,0.3421439,0.34067547,0.20117474,0.37298092,0.36123347,0.2856094,0.40088105,0.39353892,0.46255508,0.4596182,0.32452276,0.43098384,0.4287812,,0.5565345,0.50220263,0.6226138,0.60939795,0.59911895,0.6035242,0.68281937,0.67254037,0.6798825,0.6784141,0.74375916,0.74522763,0.68281937,0.76651984,0.8546255,0.9647577
+PerpetualBooster (tuned + ensembled),0.016152717,0.04992658,0.04405286,0.051395006,0.04698972,0.051395006,0.06167401,0.06975037,0.06461087,0.09985316,0.10279001,0.12041116,0.07342144,0.12041116,0.10866373,0.17621145,0.13362701,0.16299559,0.09985316,0.14537445,0.14537445,0.1483113,0.15712188,0.2143906,0.18795888,0.13876653,0.18208517,0.22687225,0.23201175,0.17327459,0.20558003,0.25844347,0.25624081,0.2246696,0.2804699,0.28487518,0.2863436,0.2246696,0.2599119,0.2907489,0.34801763,0.26945668,0.34654918,0.3069016,0.35682818,0.36710718,0.40675476,0.3274596,0.43171805,0.43612334,0.4185022,0.41703376,0.46108663,0.4434655,,0.5051395,0.5212922,0.54772395,0.7085169,0.5741556,0.5668135,0.64023495,0.60058737,0.6534508,0.59985316,0.6020558,0.8223201,0.7180617,0.7400881,0.928047
+TorchMLP (default),0.029368576,0.03671072,0.03671072,0.06167401,0.045521293,0.04405286,0.027900146,0.041116007,0.07488987,0.08810572,0.0969163,0.11013216,0.09251101,0.08957416,0.10425844,0.18061674,0.12041116,0.17033774,0.17180617,0.111600585,0.16446403,0.15565345,0.18208517,0.2143906,0.15124816,0.19236417,0.21145375,0.1585903,0.21292217,0.20851688,0.22613803,0.25256976,0.18942732,0.26431718,0.1277533,0.2393539,0.21292217,0.30396476,0.28781205,0.21365638,0.27386197,0.31424376,0.37885463,0.29955947,0.21512482,0.39941263,0.34581497,0.36857563,0.53450805,0.53744495,0.4581498,0.37444934,0.4126285,0.49779737,0.4948605,,0.6461087,0.58002937,0.5418502,0.56828195,0.6182085,0.64170337,0.6490455,0.66226137,0.68281937,0.6328928,0.6475771,0.7797357,0.79441994,0.9162996
+TabSTAR (default),0.11453745,0.13215859,0.15124816,0.12922174,0.16299559,0.15712188,0.15418503,0.17180617,0.19530103,0.18502203,0.19676946,0.19970632,0.20117474,0.2041116,0.2143906,0.21659325,0.18795888,0.2187959,0.20558003,0.22760646,0.22613803,0.23054332,0.2246696,0.2393539,0.24816446,0.22320117,0.2496329,0.2496329,0.25403818,0.22173275,0.25256976,0.31424376,0.2246696,0.22760646,0.2393539,0.26431718,0.2863436,0.24669604,0.2452276,0.2613803,0.27753305,0.26431718,0.3017621,0.2657856,0.26284876,0.30543318,0.30396476,0.29515418,0.055800293,0.039647575,0.35609397,0.33773863,0.33039647,0.37738618,0.4787078,0.35389134,,0.42731276,0.49779737,0.41409692,0.47283408,0.5359765,0.4889868,0.54772395,0.5814978,0.45521292,0.5315712,0.6578561,0.69603527,0.88986784
+xRFM (default),0.029368576,0.03524229,0.04845815,0.052863438,0.066079296,0.051395006,0.066079296,0.07488987,0.07342144,0.09985316,0.08883994,0.11453745,0.11453745,0.14684288,0.13215859,0.13656388,0.1585903,0.15565345,0.18208517,0.1938326,0.1585903,0.12187959,0.17033774,0.20998532,0.06020558,0.18869309,0.18942732,0.11013216,0.15712188,0.23201175,0.1938326,0.16813509,0.27312776,0.28487518,0.24229075,0.2555066,0.2143906,0.2804699,0.30837005,0.27312776,0.28928047,0.30396476,0.2804699,0.3524229,0.30396476,0.31864905,0.35682818,0.35682818,0.46255508,0.46549192,0.34361234,0.33480176,0.30983847,0.39060205,0.45227605,0.41997063,0.5726872,,0.47650513,0.5315712,0.47283408,0.6079295,0.54478705,0.6079295,0.5932452,0.5418502,0.5859031,0.7048458,0.7180617,0.92217326
+PerpetualBooster (tuned),0.017621145,0.039647575,0.03891336,0.04698972,0.04698972,0.054331865,0.054331865,0.05873715,0.057268724,0.07929515,0.091776796,0.111600585,0.07048458,0.09985316,0.10425844,0.16740088,0.11894273,0.13950074,0.08223201,0.13509545,0.12628488,0.12041116,0.13509545,0.20778267,0.16079295,0.1174743,0.15932453,0.18428782,0.19823788,0.1380323,0.17621145,0.24816446,0.24669604,0.20190896,0.22613803,0.23641704,0.2393539,0.1938326,0.23201175,0.23201175,0.30983847,0.23715125,0.3171806,0.27900147,0.30543318,0.33627018,0.36123347,0.2922173,0.40528634,0.41116005,0.38179147,0.35976505,0.39500734,0.40088105,0.2914831,0.4581498,0.50220263,0.52349484,,0.5462555,0.5330396,0.6240822,0.5565345,0.6343612,0.5609398,0.5712188,0.71145374,0.7004405,0.7180617,0.9030837
+iLTM (default),0.051395006,0.045521293,0.04992658,0.06167401,0.057268724,0.057268724,0.04992658,0.06020558,0.080763586,0.101321585,0.105726875,0.12481645,0.09985316,0.091042586,0.11306901,0.14096916,0.14096916,0.14977974,0.09397944,0.105726875,0.12701909,0.1380323,0.13950074,0.20998532,0.16740088,0.106461085,0.17033774,0.15418503,0.14977974,0.101321585,0.17327459,0.22320117,0.17327459,0.16593245,0.19456682,0.17327459,0.19456682,0.19970632,0.19970632,0.20851688,0.21145375,0.22320117,0.28340676,0.20558003,0.25256976,0.28928047,0.25697502,0.25697502,0.5095448,0.5051395,0.37444934,0.28781205,0.27312776,0.39647576,0.42584434,0.43171805,0.5859031,0.4684288,0.4537445,,0.47577092,0.6182085,0.5550661,0.6328928,0.67107195,0.5462555,0.5638766,0.7283407,0.7812041,0.8854626
+ExtraTrees (default),0.014684288,0.029368576,0.04992658,0.04992658,0.051395006,0.04698972,0.06754772,0.07048458,0.04698972,0.05506608,0.057268724,0.08370044,0.08223201,0.12922174,0.12481645,0.16299559,0.14243759,0.09397944,0.0969163,0.18355359,0.116005875,0.101321585,0.13069016,0.20558003,0.07488987,0.12628488,0.14390603,0.14684288,0.17180617,0.15565345,0.14684288,0.18502203,0.23348017,0.2349486,0.26284876,0.20190896,0.20998532,0.2393539,0.24669604,0.2290749,0.25697502,0.25110132,0.09397944,0.26284876,0.3010279,0.17180617,0.2922173,0.36123347,0.36417034,0.36270192,0.23348017,0.3113069,0.27312776,0.3171806,0.4331865,0.38179147,0.52716595,0.52716595,0.46696034,0.52422905,,0.5873715,0.5565345,0.59911895,0.64170337,0.5712188,0.5462555,0.7048458,0.76798826,0.96622616
+Linear (tuned + ensembled),0.04698972,0.07048458,0.086637296,0.06020558,0.07488987,0.08516887,0.07342144,0.08516887,0.11306901,0.1174743,0.1585903,0.16666667,0.116740085,0.1071953,0.1585903,0.12481645,0.17180617,0.22613803,0.13950074,0.1277533,0.16740088,0.22026432,0.17180617,0.18502203,0.15565345,0.1483113,0.17914832,0.1585903,0.19530103,0.19089574,0.18942732,0.2657856,0.18061674,0.15271659,0.21365638,0.20264317,0.27239352,0.25697502,0.17914832,0.22320117,0.19236417,0.2716593,0.27239352,0.19089574,0.2922173,0.28781205,0.23788546,0.2804699,0.34654918,0.35535976,0.31424376,0.3215859,0.3171806,0.3274596,0.35976505,0.35829663,0.4640235,0.39207047,0.37591776,0.38179147,0.4126285,,0.38032305,0.633627,0.47723934,0.41409692,0.45080763,0.78414094,0.5756241,0.77679884
+FastaiMLP (default),0.02496329,0.030837005,0.041116007,0.04992658,0.04405286,0.04845815,0.041116007,0.051395006,0.052863438,0.057268724,0.07342144,0.08370044,0.06167401,0.08516887,0.07709251,0.14096916,0.08370044,0.09985316,0.12334802,0.13509545,0.14390603,0.12187959,0.15418503,0.17621145,0.09544787,0.13362701,0.17180617,0.11894273,0.15051395,0.14243759,0.17767988,0.20998532,0.18502203,0.1585903,0.14757709,0.16740088,0.17033774,0.19089574,0.19089574,0.13215859,0.054331865,0.21585903,0.25256976,0.19676946,0.21732746,0.26431718,0.15565345,0.2716593,0.33773863,0.34067547,0.3171806,0.28487518,0.3069016,0.32011747,0.39941263,0.35095447,0.5110132,0.45521292,0.4434655,0.44493392,0.4434655,0.61967695,,0.6328928,0.5814978,0.489721,0.5829662,0.7400881,0.7591777,0.92217326
+Linear (tuned),0.041116007,0.06461087,0.06754772,0.055800293,0.06314244,0.06754772,0.06754772,0.07342144,0.111600585,0.11013216,0.14243759,0.15565345,0.096182086,0.09471366,0.15198238,0.10866373,0.15271659,0.20264317,0.12334802,0.11894273,0.15418503,0.19089574,0.16079295,0.18208517,0.14390603,0.13142438,0.16886932,0.14243759,0.17327459,0.17327459,0.17033774,0.25403818,0.16519824,0.13950074,0.18355359,0.18428782,0.24082232,0.2393539,0.16446403,0.19236417,0.16886932,0.26431718,0.26211452,0.16519824,0.2709251,0.2804699,0.2195301,0.26872247,0.3215859,0.32452276,0.30983847,0.30543318,0.3010279,0.3215859,0.34654918,0.33773863,0.45227605,0.39207047,0.36563876,0.36710718,0.40088105,0.36637297,0.36710718,,0.45154184,0.39353892,0.44640234,0.71659327,0.5609398,0.7694567
+KNN (tuned + ensembled),0.030837005,0.052863438,0.057268724,0.04845815,0.07195301,0.054331865,0.0969163,0.08370044,0.08810572,0.07048458,0.07856094,0.080763586,0.11306901,0.12334802,0.1277533,0.1380323,0.111600585,0.09838473,0.11894273,0.15124816,0.11894273,0.14390603,0.12922174,0.22320117,0.10792951,0.13656388,0.1585903,0.12481645,0.1585903,0.15124816,0.16226138,0.20998532,0.19676946,0.17180617,0.2041116,0.13950074,0.22026432,0.23788546,0.19530103,0.17767988,0.20117474,0.25697502,0.12334802,0.17474303,0.24596182,0.15271659,0.23054332,0.3061674,0.28781205,0.29809105,0.22320117,0.2657856,0.2613803,0.25624081,0.40014684,0.3171806,0.4185022,0.40675476,0.4390602,0.32892805,0.35829663,0.5227606,0.4185022,0.54845816,,0.3979442,0.51395005,0.67547727,0.8641703,0.969163
+RandomForest (default),0.016152717,0.017621145,0.032305434,0.038179148,0.041116007,0.026431719,0.051395006,0.051395006,0.032305434,0.039647575,0.051395006,0.07048458,0.07048458,0.09397944,0.08370044,0.12187959,0.09985316,0.105726875,0.06167401,0.12922174,0.06754772,0.08370044,0.07782672,0.17914832,0.054331865,0.08810572,0.07782672,0.10425844,0.14243759,0.116005875,0.09544787,0.17327459,0.17474303,0.18942732,0.23201175,0.16152717,0.17914832,0.20117474,0.21292217,0.20264317,0.24375919,0.23641704,0.12187959,0.24229075,0.2760646,0.16226138,0.2804699,0.27753305,0.34948605,0.34654918,0.11894273,0.25770926,0.21732746,0.2547724,0.3979442,0.36710718,0.54478705,0.4581498,0.4287812,0.4537445,0.4287812,0.5859031,0.510279,0.6064611,0.6020558,,0.52863437,0.7004405,0.74449337,0.93538916
+PerpetualBooster (default),0.014684288,0.026431719,0.030837005,0.020558003,0.030837005,0.032305434,0.02496329,0.030837005,0.029368576,0.03671072,0.051395006,0.052863438,0.04845815,0.039647575,0.045521293,0.11306901,0.06754772,0.091042586,0.04845815,0.06167401,0.080763586,0.080763586,0.076358296,0.15418503,0.080763586,0.054331865,0.09251101,0.09544787,0.12481645,0.086637296,0.09397944,0.17033774,0.1277533,0.08957416,0.12628488,0.15271659,0.14977974,0.13215859,0.11013216,0.13509545,0.17914832,0.14170338,0.23641704,0.13950074,0.19236417,0.2496329,0.2393539,0.18942732,0.37004405,0.37151247,0.2966226,0.22026432,0.2701909,0.3171806,0.17767988,0.3524229,0.4684288,0.41409692,0.28854626,0.43612334,0.4537445,0.54919237,0.41703376,0.5535976,0.48604992,0.47136563,,0.6380323,0.6387665,0.8428781
+Linear (default),0.029368576,0.038179148,0.047723934,0.035976507,0.05506608,0.04698972,0.03524229,0.049192365,0.09544787,0.051395006,0.09985316,0.10279001,0.07048458,0.07488987,0.12041116,0.08516887,0.07415565,0.11453745,0.076358296,0.08957416,0.12481645,0.14390603,0.13069016,0.16152717,0.11527166,0.08810572,0.13215859,0.09838473,0.13215859,0.11306901,0.1380323,0.22026432,0.1277533,0.111600585,0.116005875,0.10425844,0.18061674,0.15565345,0.13069016,0.12334802,0.12041116,0.17180617,0.18942732,0.10866373,0.14537445,0.19676946,0.1483113,0.17327459,0.2187959,0.2187959,0.22687225,0.19823788,0.19970632,0.23348017,0.2819383,0.22026432,0.3421439,0.29515418,0.29955947,0.2716593,0.29515418,0.21585903,0.2599119,0.28340676,0.32452276,0.29955947,0.36196768,,0.47577092,0.72246695
+KNN (tuned),0.03377386,0.041116007,0.042584434,0.042584434,0.06167401,0.04698972,0.06314244,0.05873715,0.06461087,0.051395006,0.06754772,0.07488987,0.07488987,0.086637296,0.09985316,0.105726875,0.051395006,0.076358296,0.07856094,0.11013216,0.086637296,0.10792951,0.08370044,0.13362701,0.080763586,0.08810572,0.1020558,0.086637296,0.11306901,0.080763586,0.1071953,0.17767988,0.11894273,0.11967695,0.1277533,0.10866373,0.15198238,0.13656388,0.12041116,0.111600585,0.12628488,0.16299559,0.091042586,0.10866373,0.17180617,0.086637296,0.1585903,0.1585903,0.16593245,0.17180617,0.13656388,0.18061674,0.18649046,0.14537445,0.2599119,0.20558003,0.30396476,0.2819383,0.2819383,0.2187959,0.23201175,0.42437592,0.24082232,0.4390602,0.13582966,0.2555066,0.36123347,0.52422905,,0.9265786
+KNN (default),0.007342144,0.005873715,0.007342144,0.005873715,0.010279002,0.008810572,0.017621145,0.016152717,0.007342144,0.01174743,0.007342144,0.010279002,0.01174743,0.029368576,0.020558003,0.02349486,0.007342144,0.008810572,0.016152717,0.038179148,0.014684288,0.019089574,0.016152717,0.017621145,0.013215859,0.02202643,0.02496329,0.016152717,0.026431719,0.019089574,0.027900146,0.04992658,0.057268724,0.04405286,0.04845815,0.02349486,0.045521293,0.03157122,0.051395006,0.02496329,0.054331865,0.045521293,0.010279002,0.04845815,0.06314244,0.01174743,0.06461087,0.04698972,0.076358296,0.076358296,0.03377386,0.05359765,0.066079296,0.03524229,0.07195301,0.08370044,0.11013216,0.07782672,0.0969163,0.11453745,0.03377386,0.22320117,0.07782672,0.23054332,0.030837005,0.06461087,0.15712188,0.27753305,0.07342144,
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_all/leaderboard_overview_explorer.html
index b2d4a41b6fe033475f7e76022160ec96b7abe460..4c71d48824f206af04dc0e385561565f8ec7bd42 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Binary | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_all/leaderboard_table.html b/data/imputation_no/splits_all/tasks_binary/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..456f9dbbac38089c2e33305de182afa8abb3e21f
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_binary/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1792.0,"score":0.89,"rank":6.26,"hrank":1.67,"imp":4.069,"train_time":36.45,"predict_time":7.109,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-90","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1675.0,"score":0.627,"rank":10.18,"hrank":3.92,"imp":8.845,"train_time":272.41,"predict_time":4.561,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-84","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1671.0,"score":0.663,"rank":10.36,"hrank":4.25,"imp":11.378,"train_time":2.63,"predict_time":0.361,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-68","imputed":false},{"position":3,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1602.0,"score":0.596,"rank":13.62,"hrank":5.06,"imp":11.502,"train_time":1.77,"predict_time":0.158,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+113\/-60","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1596.0,"score":0.544,"rank":13.94,"hrank":7.42,"imp":12.76,"train_time":4.82,"predict_time":0.464,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-60","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1546.0,"score":0.465,"rank":16.73,"hrank":7.97,"imp":13.422,"train_time":1924.5,"predict_time":9.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-54","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1532.0,"score":0.471,"rank":17.63,"hrank":10.1,"imp":13.861,"train_time":4.91,"predict_time":0.694,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-50","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1518.0,"score":0.439,"rank":18.49,"hrank":8.6,"imp":14.0,"train_time":1924.5,"predict_time":1.329,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-52","imputed":false},{"position":8,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1500.0,"score":0.391,"rank":19.73,"hrank":7.56,"imp":14.312,"train_time":1152.97,"predict_time":1.783,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-67","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1481.0,"score":0.348,"rank":20.99,"hrank":13.88,"imp":15.238,"train_time":2775.98,"predict_time":8.849,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-53","imputed":false},{"position":10,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1467.0,"score":0.308,"rank":21.97,"hrank":11.19,"imp":15.685,"train_time":1612.07,"predict_time":1.884,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-49","imputed":false},{"position":11,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1439.0,"score":0.338,"rank":24.02,"hrank":10.05,"imp":15.994,"train_time":1.44,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-58","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1425.0,"score":0.246,"rank":25.14,"hrank":17.43,"imp":16.781,"train_time":328.96,"predict_time":1.101,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-39","imputed":false},{"position":13,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1414.0,"score":0.256,"rank":25.95,"hrank":16.88,"imp":16.494,"train_time":8305.07,"predict_time":336.002,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-50","imputed":false},{"position":14,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1412.0,"score":0.248,"rank":26.14,"hrank":12.91,"imp":16.566,"train_time":1612.07,"predict_time":0.232,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-55","imputed":false},{"position":15,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1409.0,"score":0.269,"rank":26.33,"hrank":12.88,"imp":15.799,"train_time":4982.66,"predict_time":358.753,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-53","imputed":false},{"position":16,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1406.0,"score":0.251,"rank":26.61,"hrank":14.16,"imp":16.519,"train_time":738.26,"predict_time":0.246,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-51","imputed":false},{"position":17,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1404.0,"score":0.266,"rank":26.71,"hrank":13.48,"imp":16.487,"train_time":3.04,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-50","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1398.0,"score":0.245,"rank":27.21,"hrank":14.36,"imp":16.683,"train_time":738.26,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-53","imputed":false},{"position":19,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1391.0,"score":0.222,"rank":27.76,"hrank":15.3,"imp":16.822,"train_time":2775.98,"predict_time":0.415,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-54","imputed":false},{"position":20,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1375.0,"score":0.166,"rank":29.02,"hrank":21.69,"imp":17.581,"train_time":328.96,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-30","imputed":false},{"position":21,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1375.0,"score":0.24,"rank":29.06,"hrank":16.21,"imp":18.468,"train_time":250.39,"predict_time":1.708,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-54","imputed":false},{"position":22,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1374.0,"score":0.226,"rank":29.13,"hrank":16.72,"imp":17.968,"train_time":281.39,"predict_time":0.314,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-60","imputed":false},{"position":23,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1374.0,"score":0.268,"rank":29.14,"hrank":16.56,"imp":17.466,"train_time":1.13,"predict_time":0.059,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-70","imputed":false},{"position":24,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1373.0,"score":0.2,"rank":29.18,"hrank":18.65,"imp":17.582,"train_time":473.04,"predict_time":0.621,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-49","imputed":false},{"position":25,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1371.0,"score":0.297,"rank":29.34,"hrank":11.5,"imp":18.364,"train_time":14.26,"predict_time":4.299,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+94\/-84","imputed":false},{"position":26,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1358.0,"score":0.179,"rank":30.44,"hrank":16.91,"imp":17.917,"train_time":3383.91,"predict_time":0.51,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-57","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1355.0,"score":0.252,"rank":30.68,"hrank":15.22,"imp":18.08,"train_time":3383.91,"predict_time":8.39,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-90","imputed":false},{"position":28,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1347.0,"score":0.201,"rank":31.3,"hrank":14.18,"imp":17.206,"train_time":4982.66,"predict_time":42.199,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-57","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1344.0,"score":0.183,"rank":31.57,"hrank":18.65,"imp":18.436,"train_time":281.39,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-58","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1340.0,"score":0.146,"rank":31.92,"hrank":22.12,"imp":18.071,"train_time":473.04,"predict_time":0.13,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-41","imputed":false},{"position":31,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1318.0,"score":0.16,"rank":33.67,"hrank":21.3,"imp":18.462,"train_time":642.4,"predict_time":2.766,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-56","imputed":false},{"position":32,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1317.0,"score":0.157,"rank":33.81,"hrank":19.13,"imp":19.06,"train_time":1.25,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-58","imputed":false},{"position":33,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1312.0,"score":0.139,"rank":34.24,"hrank":24.68,"imp":18.288,"train_time":2241.92,"predict_time":2.415,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-41","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1308.0,"score":0.153,"rank":34.56,"hrank":21.29,"imp":18.597,"train_time":5.03,"predict_time":0.208,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-56","imputed":false},{"position":35,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1304.0,"score":0.111,"rank":34.86,"hrank":24.91,"imp":19.685,"train_time":936.75,"predict_time":0.134,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-45","imputed":false},{"position":36,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1298.0,"score":0.111,"rank":35.36,"hrank":21.24,"imp":19.213,"train_time":8305.07,"predict_time":52.875,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+45\/-49","imputed":false},{"position":37,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1293.0,"score":0.137,"rank":35.76,"hrank":22.99,"imp":19.557,"train_time":642.4,"predict_time":0.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-63","imputed":false},{"position":38,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1289.0,"score":0.16,"rank":36.11,"hrank":16.96,"imp":20.777,"train_time":12.86,"predict_time":3.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-60","imputed":false},{"position":39,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1276.0,"score":0.159,"rank":37.25,"hrank":20.11,"imp":20.033,"train_time":49.67,"predict_time":43.824,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-69","imputed":false},{"position":40,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1270.0,"score":0.073,"rank":37.71,"hrank":27.4,"imp":20.358,"train_time":936.75,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-43","imputed":false},{"position":41,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1266.0,"score":0.104,"rank":38.06,"hrank":21.09,"imp":20.576,"train_time":13.51,"predict_time":0.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-46","imputed":false},{"position":42,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1262.0,"score":0.078,"rank":38.36,"hrank":27.0,"imp":19.481,"train_time":9.33,"predict_time":1.116,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+41\/-47","imputed":false},{"position":43,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1260.0,"score":0.111,"rank":38.51,"hrank":25.91,"imp":19.796,"train_time":559.38,"predict_time":4.079,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-60","imputed":false},{"position":44,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1246.0,"score":0.069,"rank":39.71,"hrank":26.1,"imp":21.132,"train_time":3.27,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-50","imputed":false},{"position":45,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1229.0,"score":0.073,"rank":41.12,"hrank":28.45,"imp":20.59,"train_time":1.24,"predict_time":0.11,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-54","imputed":false},{"position":46,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1226.0,"score":0.063,"rank":41.34,"hrank":31.17,"imp":20.31,"train_time":2241.92,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-38","imputed":false},{"position":47,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1190.0,"score":0.045,"rank":44.28,"hrank":29.6,"imp":21.714,"train_time":559.38,"predict_time":0.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-56","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1186.0,"score":0.032,"rank":44.64,"hrank":35.83,"imp":21.625,"train_time":182.35,"predict_time":0.686,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-56","imputed":false},{"position":49,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1182.0,"score":0.033,"rank":44.94,"hrank":38.35,"imp":21.314,"train_time":0.81,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-36","imputed":false},{"position":50,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1175.0,"score":0.063,"rank":45.52,"hrank":27.78,"imp":23.15,"train_time":210.38,"predict_time":43.847,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-64","imputed":false},{"position":51,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1167.0,"score":0.164,"rank":46.12,"hrank":12.36,"imp":24.458,"train_time":25373.34,"predict_time":3.667,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-91","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1162.0,"score":0.031,"rank":46.5,"hrank":35.26,"imp":22.437,"train_time":182.35,"predict_time":0.058,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-66","imputed":false},{"position":53,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1160.0,"score":0.157,"rank":46.66,"hrank":12.04,"imp":24.586,"train_time":25373.34,"predict_time":14.078,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-91","imputed":false},{"position":54,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1151.0,"score":0.022,"rank":47.34,"hrank":35.79,"imp":23.018,"train_time":169.75,"predict_time":0.603,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-59","imputed":false},{"position":55,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1144.0,"score":0.096,"rank":47.91,"hrank":31.27,"imp":24.134,"train_time":14.46,"predict_time":2.705,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-75","imputed":false},{"position":56,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1120.0,"score":0.01,"rank":49.73,"hrank":38.72,"imp":23.674,"train_time":169.75,"predict_time":0.059,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-60","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1100.0,"score":0.035,"rank":51.14,"hrank":37.01,"imp":25.524,"train_time":117.85,"predict_time":0.773,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-76","imputed":false},{"position":58,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1098.0,"score":0.014,"rank":51.31,"hrank":38.15,"imp":23.987,"train_time":4.95,"predict_time":0.091,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-47","imputed":false},{"position":59,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1067.0,"score":0.022,"rank":53.46,"hrank":40.08,"imp":26.897,"train_time":117.85,"predict_time":0.29,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-76","imputed":false},{"position":60,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1066.0,"score":0.012,"rank":53.54,"hrank":34.88,"imp":25.31,"train_time":2.56,"predict_time":0.267,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-84","imputed":false},{"position":61,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1065.0,"score":0.107,"rank":53.64,"hrank":19.65,"imp":29.197,"train_time":391.11,"predict_time":4.052,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-107","imputed":false},{"position":62,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1052.0,"score":0.053,"rank":54.45,"hrank":34.88,"imp":29.801,"train_time":162.68,"predict_time":0.405,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-104","imputed":false},{"position":63,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1046.0,"score":0.065,"rank":54.9,"hrank":34.11,"imp":29.078,"train_time":0.85,"predict_time":0.129,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-106","imputed":false},{"position":64,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1028.0,"score":0.037,"rank":56.05,"hrank":36.52,"imp":30.423,"train_time":162.68,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-116","imputed":false},{"position":65,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1026.0,"score":0.012,"rank":56.16,"hrank":41.24,"imp":27.262,"train_time":139.88,"predict_time":3.917,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-98","imputed":false},{"position":66,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1004.0,"score":0.011,"rank":57.51,"hrank":49.84,"imp":27.379,"train_time":1.09,"predict_time":0.191,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-97","imputed":false},{"position":67,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.008,"rank":57.73,"hrank":51.02,"imp":27.718,"train_time":0.25,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-82","imputed":false},{"position":68,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":989.0,"score":0.014,"rank":58.35,"hrank":50.4,"imp":30.53,"train_time":2.8,"predict_time":0.564,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-103","imputed":false},{"position":69,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":970.0,"score":0.023,"rank":59.4,"hrank":40.99,"imp":32.263,"train_time":0.84,"predict_time":0.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-111","imputed":false},{"position":70,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":950.0,"score":0.0,"rank":60.45,"hrank":48.94,"imp":30.959,"train_time":15.8,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-79","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":924.0,"score":0.001,"rank":61.75,"hrank":46.0,"imp":32.066,"train_time":139.88,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-132","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":628.0,"score":0.0,"rank":70.54,"hrank":64.56,"imp":46.001,"train_time":0.15,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-164","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_all/pareto_front_explorer.html
index 477418d2f460416b73c88745f212e2dcd6a2f5a9..1a93716acafbf3d78ac5056f1f71ea81cf4be108 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7076126445,"imp":18.4682967936,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0203827935,"imp":16.4867072742,"elo":1404.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.028876079,"imp":16.6825449585,"elo":1397.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.245609885,"imp":16.5187747585,"elo":1405.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0261083185,"imp":19.0597014368,"elo":1316.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0271269609,"imp":18.4363953201,"elo":1343.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3141545533,"imp":17.9682224004,"elo":1373.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0148652021,"imp":21.1322179835,"elo":1246.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0160758722,"imp":20.3582121197,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1335297064,"imp":19.6851656659,"elo":1304.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1905448885,"imp":27.3787868305,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0583336488,"imp":22.4369311607,"elo":1162.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6857611443,"imp":21.6252080132,"elo":1185.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2670364168,"imp":25.30965008,"elo":1066.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2760208318,"imp":21.7144730552,"elo":1190.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0790081457,"imp":19.7964992774,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0925963767,"imp":21.3138099976,"elo":1181.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1264589403,"imp":17.5811602118,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1012944014,"imp":16.7811162289,"elo":1424.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.2986429509,"imp":18.3641297375,"elo":1371.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.310422718,"imp":20.5760221876,"elo":1265.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5103151766,"imp":17.9165767833,"elo":1357.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3903226253,"imp":18.079660704,"elo":1354.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7050443034,"imp":24.1337917973,"elo":1143.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0291154934,"imp":30.9590846155,"elo":950.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2900648816,"imp":26.8974371631,"elo":1067.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7725893154,"imp":25.523860396,"elo":1100.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.02947866,"imp":27.7175500855,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0593784249,"imp":23.6736513468,"elo":1119.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.6033358381,"imp":23.0176779136,"elo":1151.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.1163701521,"imp":19.4805797336,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.4146988006,"imp":16.8218275005,"elo":1391.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":8.8488311223,"imp":15.2384414918,"elo":1481.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6942704557,"imp":13.8607400852,"elo":1531.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3289068866,"imp":13.9998696519,"elo":1518.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.1029658055,"imp":13.4216293747,"elo":1546.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.166690434,"imp":20.7769968093,"elo":1289.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"imp":20.0327749296,"elo":1275.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.1992341236,"imp":17.2062673063,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":358.7531028481,"imp":15.7989040969,"elo":1409.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"imp":15.9939065516,"elo":1439.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"imp":4.0685886094,"elo":1791.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.128947975,"imp":29.0776309936,"elo":1045.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.158249445,"imp":11.502165757,"elo":1601.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2075493916,"imp":18.5967344155,"elo":1307.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2319353399,"imp":16.5661012602,"elo":1411.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.8843761148,"imp":15.6854948662,"elo":1467.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4644557714,"imp":12.760102874,"elo":1595.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3605764018,"imp":11.3777129759,"elo":1671.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0517146513,"imp":29.1966405427,"elo":1064.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6668093955,"imp":24.4584933887,"elo":1166.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0783983006,"imp":24.5863098775,"elo":1160.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0586280558,"imp":17.4657497811,"elo":1373.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0914684387,"imp":23.9865384514,"elo":1098.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1361580238,"imp":20.3104889118,"elo":1226.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4148361224,"imp":18.2881318738,"elo":1311.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1102748358,"imp":20.5900999078,"elo":1228.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1302082399,"imp":18.0712226216,"elo":1339.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6209770484,"imp":17.5816138221,"elo":1373.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":43.847258449,"imp":23.1496096303,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":52.8754683564,"imp":19.2134493566,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":336.0020690595,"imp":16.493553874,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.5638730428,"imp":30.5297394391,"elo":989.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.146245996,"imp":19.5565603067,"elo":1293.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.7662131303,"imp":18.4621614228,"elo":1318.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7076126445,"x_train":250.3856225541,"imp":18.4682967936,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0203827935,"x_train":3.0406505418,"imp":16.4867072742,"elo":1404.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.028876079,"x_train":738.2612405475,"imp":16.6825449585,"elo":1397.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.245609885,"x_train":738.2612405475,"imp":16.5187747585,"elo":1405.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0261083185,"x_train":1.2471436483,"imp":19.0597014368,"elo":1316.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0271269609,"x_train":281.3883698637,"imp":18.4363953201,"elo":1343.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3141545533,"x_train":281.3883698637,"imp":17.9682224004,"elo":1373.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0148652021,"x_train":3.2701795229,"imp":21.1322179835,"elo":1246.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0160758722,"x_train":936.7523176703,"imp":20.3582121197,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1335297064,"x_train":936.7523176703,"imp":19.6851656659,"elo":1304.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1905448885,"x_train":1.0867260517,"imp":27.3787868305,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0583336488,"x_train":182.354525564,"imp":22.4369311607,"elo":1162.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6857611443,"x_train":182.354525564,"imp":21.6252080132,"elo":1185.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2670364168,"x_train":2.556542435,"imp":25.30965008,"elo":1066.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2760208318,"x_train":559.3818159963,"imp":21.7144730552,"elo":1190.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0790081457,"x_train":559.3818159963,"imp":19.7964992774,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0925963767,"x_train":0.8126978053,"imp":21.3138099976,"elo":1181.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1264589403,"x_train":328.9554302437,"imp":17.5811602118,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1012944014,"x_train":328.9554302437,"imp":16.7811162289,"elo":1424.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.2986429509,"x_train":14.2586021738,"imp":18.3641297375,"elo":1371.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.310422718,"x_train":13.5066187328,"imp":20.5760221876,"elo":1265.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5103151766,"x_train":3383.9090129865,"imp":17.9165767833,"elo":1357.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3903226253,"x_train":3383.9090129865,"imp":18.079660704,"elo":1354.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7050443034,"x_train":14.4628692945,"imp":24.1337917973,"elo":1143.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0291154934,"x_train":15.7950255214,"imp":30.9590846155,"elo":950.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2900648816,"x_train":117.8541822739,"imp":26.8974371631,"elo":1067.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7725893154,"x_train":117.8541822739,"imp":25.523860396,"elo":1100.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.02947866,"x_train":0.2544731196,"imp":27.7175500855,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0593784249,"x_train":169.7548111684,"imp":23.6736513468,"elo":1119.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.6033358381,"x_train":169.7548111684,"imp":23.0176779136,"elo":1151.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.1163701521,"x_train":9.3300735239,"imp":19.4805797336,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.4146988006,"x_train":2775.9833627967,"imp":16.8218275005,"elo":1391.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":8.8488311223,"x_train":2775.9833627967,"imp":15.2384414918,"elo":1481.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6942704557,"x_train":4.9087216863,"imp":13.8607400852,"elo":1531.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3289068866,"x_train":1924.5007092935,"imp":13.9998696519,"elo":1518.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.1029658055,"x_train":1924.5007092935,"imp":13.4216293747,"elo":1546.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.166690434,"x_train":12.8582127474,"imp":20.7769968093,"elo":1289.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"x_train":49.671607539,"imp":20.0327749296,"elo":1275.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.1992341236,"x_train":4982.6619115707,"imp":17.2062673063,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":358.7531028481,"x_train":4982.6619115707,"imp":15.7989040969,"elo":1409.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"x_train":1.4436537137,"imp":15.9939065516,"elo":1439.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"x_train":36.4454770138,"imp":4.0685886094,"elo":1791.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.128947975,"x_train":0.8504331063,"imp":29.0776309936,"elo":1045.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.158249445,"x_train":1.7686337423,"imp":11.502165757,"elo":1601.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2075493916,"x_train":5.0292797696,"imp":18.5967344155,"elo":1307.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2319353399,"x_train":1612.0705710762,"imp":16.5661012602,"elo":1411.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.8843761148,"x_train":1612.0705710762,"imp":15.6854948662,"elo":1467.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4644557714,"x_train":4.8153059659,"imp":12.760102874,"elo":1595.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3605764018,"x_train":2.62612376,"imp":11.3777129759,"elo":1671.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0517146513,"x_train":391.1080876361,"imp":29.1966405427,"elo":1064.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6668093955,"x_train":25373.3350332343,"imp":24.4584933887,"elo":1166.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0783983006,"x_train":25373.3350332343,"imp":24.5863098775,"elo":1160.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0586280558,"x_train":1.1266327909,"imp":17.4657497811,"elo":1373.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0914684387,"x_train":4.9502582412,"imp":23.9865384514,"elo":1098.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1361580238,"x_train":2241.9244610212,"imp":20.3104889118,"elo":1226.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4148361224,"x_train":2241.9244610212,"imp":18.2881318738,"elo":1311.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1102748358,"x_train":1.2379804347,"imp":20.5900999078,"elo":1228.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1302082399,"x_train":473.0394868517,"imp":18.0712226216,"elo":1339.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6209770484,"x_train":473.0394868517,"imp":17.5816138221,"elo":1373.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":43.847258449,"x_train":210.3752328358,"imp":23.1496096303,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":52.8754683564,"x_train":8305.0663055001,"imp":19.2134493566,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":336.0020690595,"x_train":8305.0663055001,"imp":16.493553874,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.5638730428,"x_train":2.8035263333,"imp":30.5297394391,"elo":989.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.146245996,"x_train":642.4042496517,"imp":19.5565603067,"elo":1293.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.7662131303,"x_train":642.4042496517,"imp":18.4621614228,"elo":1318.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_all/tuning_trajectories_explorer.html
index 7a45d4d7eb345d98e0e0881c63a1b5fdfe7ebf02..5bb0b1ad42abbf91a184a5ff4e29fb5e2a9b7261 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":250.3856225541,"imp":18.4933189805,"elo":1397.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.0406505418,"imp":16.5104471714,"elo":1426.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":7.9183650076,"imp":16.4167776727,"elo":1432.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":18.1594886071,"imp":16.3645233167,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":36.3465094416,"imp":16.3892082295,"elo":1438.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":92.6292744451,"imp":16.3943655086,"elo":1438.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":181.7510780205,"imp":16.4086167429,"elo":1437.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":367.6200398568,"imp":16.4710396294,"elo":1432.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":549.0966445079,"imp":16.5097848779,"elo":1429.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":738.2612405475,"imp":16.5372195649,"elo":1427.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.2471436483,"imp":19.0829733294,"elo":1336.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.6592018735,"imp":18.6812379658,"elo":1364.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":6.827854614,"imp":18.4929052129,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":13.7690635449,"imp":18.3889370073,"elo":1380.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":34.8312631967,"imp":18.2379336969,"elo":1388.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":69.9560479767,"imp":18.1477267418,"elo":1393.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":140.0107160319,"imp":18.0536408742,"elo":1398.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":209.935077018,"imp":18.0111503293,"elo":1398.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":281.3883698637,"imp":17.9898475453,"elo":1399.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":3.2701795229,"imp":21.1550409794,"elo":1254.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":8.5418580508,"imp":20.7461328006,"elo":1268.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":21.5154250642,"imp":20.3672145186,"elo":1286.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":44.8133243453,"imp":20.2168351975,"elo":1293.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":115.6024980549,"imp":20.0572180124,"elo":1302.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":235.0233869244,"imp":19.919349256,"elo":1310.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":467.0399506352,"imp":19.7939126807,"elo":1316.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":700.0647480226,"imp":19.7572455867,"elo":1319.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":936.7523176703,"imp":19.7105302771,"elo":1322.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.0867260517,"imp":27.4001748997,"elo":1005.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.9830559575,"imp":25.6735585024,"elo":1046.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.7045699739,"imp":23.2477333054,"elo":1120.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.2122172375,"imp":22.3653758851,"elo":1158.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":22.7973551316,"imp":21.912093139,"elo":1178.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":45.453079637,"imp":21.7483992588,"elo":1188.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":90.7421763607,"imp":21.6454460116,"elo":1194.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":136.0852569857,"imp":21.6176267391,"elo":1198.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":182.354525564,"imp":21.6501981557,"elo":1198.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.556542435,"imp":25.3313477134,"elo":1076.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.2460669855,"imp":23.9098892442,"elo":1117.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":13.8120709411,"imp":22.0691811008,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":28.1488717871,"imp":21.1656381222,"elo":1219.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":68.1608030808,"imp":20.4838923587,"elo":1245.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":137.2584674121,"imp":20.155553068,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":277.1007160478,"imp":19.9634823566,"elo":1268.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":416.6378564081,"imp":19.8436916216,"elo":1275.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":559.3818159963,"imp":19.8128283176,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.8126978053,"imp":21.3359565942,"elo":1186.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":2.859374793,"imp":19.3232189963,"elo":1284.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":7.6235569739,"imp":18.0571797374,"elo":1369.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":15.609269256,"imp":17.4638234753,"elo":1410.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":40.8449535661,"imp":17.0877760749,"elo":1432.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":81.9221899435,"imp":16.9569657046,"elo":1441.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":163.7675316385,"imp":16.8463684769,"elo":1448.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":244.6291470871,"imp":16.806397822,"elo":1449.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":328.9554302437,"imp":16.8068902988,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":14.2586021738,"imp":18.2681832537,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":13.5066187328,"imp":20.5987376578,"elo":1282.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":29.3976468222,"imp":19.3828809105,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":85.706508499,"imp":17.4028059878,"elo":1402.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":168.9559526652,"imp":17.0083157792,"elo":1417.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":408.9287781105,"imp":16.6342936931,"elo":1432.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":844.0960252495,"imp":16.8917745728,"elo":1421.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1679.9964089232,"imp":17.4242403311,"elo":1402.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":2502.3959153065,"imp":17.8701238413,"elo":1387.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":3383.9090129865,"imp":18.1123460904,"elo":1378.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":14.4628692945,"imp":24.138681705,"elo":1158.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":15.7950255214,"imp":31.7917728153,"elo":947.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":37.0036028141,"imp":29.4275925671,"elo":1009.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":98.9757470004,"imp":25.6585264214,"elo":1107.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":117.8541822739,"imp":25.5415688376,"elo":1114.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.2544731196,"imp":27.7387153293,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":1.1022482459,"imp":25.4427953271,"elo":1058.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":3.7036683225,"imp":24.2985007049,"elo":1100.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":7.9903720898,"imp":23.7402292325,"elo":1125.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":20.5657356408,"imp":23.2882960277,"elo":1147.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":41.6049917756,"imp":23.1528100801,"elo":1155.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":83.9433065314,"imp":23.0662800175,"elo":1160.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":126.3785182217,"imp":23.0262199814,"elo":1162.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":169.7548111684,"imp":23.0376221464,"elo":1163.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":9.3300735239,"imp":19.5026013508,"elo":1272.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":23.3699912915,"imp":17.6174972413,"elo":1374.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":64.116353982,"imp":16.3676580292,"elo":1446.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":135.6738391001,"imp":15.9061199923,"elo":1472.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":343.9787328013,"imp":15.5519904832,"elo":1493.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":692.1853967506,"imp":15.4217112457,"elo":1502.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1383.1114451594,"imp":15.325694052,"elo":1507.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2072.6413859802,"imp":15.283515667,"elo":1510.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2775.9833627967,"imp":15.2636046028,"elo":1510.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":4.9087216863,"imp":13.8861384299,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.4525053202,"imp":13.8966382852,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":44.7436120415,"imp":13.7257647431,"elo":1555.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":91.7548974847,"imp":13.661718684,"elo":1556.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":236.7117395728,"imp":13.5127479641,"elo":1565.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":474.1680372395,"imp":13.4830451944,"elo":1568.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":957.1717702231,"imp":13.4318526315,"elo":1571.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1438.8272594182,"imp":13.432059224,"elo":1570.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1924.5007092935,"imp":13.4551080371,"elo":1567.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":12.8582127474,"imp":20.7986247854,"elo":1307.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":49.671607539,"imp":20.0548159955,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":74.371713941,"imp":18.9352451494,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":135.8465940794,"imp":17.8796770802,"elo":1361.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":245.5890330499,"imp":17.5153156897,"elo":1373.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":603.2630243511,"imp":16.6880388295,"elo":1403.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1228.1552324622,"imp":16.2966575311,"elo":1418.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2466.8687838379,"imp":16.0640662704,"elo":1426.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3715.4039160262,"imp":15.8960164204,"elo":1431.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4982.6619115707,"imp":15.8278609368,"elo":1433.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":1.4436537137,"imp":16.0168834038,"elo":1457.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":36.4454770138,"imp":4.0882031143,"elo":1815.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.8504331063,"imp":29.0978789966,"elo":1063.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.7686337423,"imp":11.5280550051,"elo":1625.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":5.0292797696,"imp":18.6200249426,"elo":1324.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":12.3959630427,"imp":18.1075360361,"elo":1351.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":37.7043962113,"imp":17.420210667,"elo":1390.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":76.792938919,"imp":16.82113508,"elo":1429.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":197.2453170668,"imp":16.3082031884,"elo":1455.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":398.0123545957,"imp":16.0339563294,"elo":1471.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":792.8708351593,"imp":15.8413077925,"elo":1482.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1198.5048285012,"imp":15.7531162122,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1612.0705710762,"imp":15.708241152,"elo":1492.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":4.8153059659,"imp":12.7851447325,"elo":1624.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":2.62612376,"imp":11.3977989809,"elo":1698.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":391.1080876361,"imp":29.2171330577,"elo":1076.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":569.6608345335,"imp":29.1020946138,"elo":1082.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":976.9745445733,"imp":28.8625681047,"elo":1089.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1509.8128718211,"imp":28.2069059965,"elo":1104.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3235.1736621912,"imp":27.0760733386,"elo":1126.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":6343.9144626669,"imp":25.9428567016,"elo":1146.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":12590.6623522261,"imp":25.326744153,"elo":1161.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":18827.9881620466,"imp":24.9309639133,"elo":1171.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":25373.3350332343,"imp":24.6087815593,"elo":1181.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1266327909,"imp":17.4898874411,"elo":1395.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":4.9502582412,"imp":23.9989774263,"elo":1109.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.9663495618,"imp":21.9556733634,"elo":1172.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":44.495543604,"imp":20.314403314,"elo":1226.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":110.9528145485,"imp":19.6459835949,"elo":1258.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":276.2432638721,"imp":19.0715601782,"elo":1287.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":564.391803329,"imp":18.7085856134,"elo":1308.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1102.6775922424,"imp":18.5179870159,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1669.2077718784,"imp":18.4068762129,"elo":1329.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2241.9244610212,"imp":18.3030306436,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.2379804347,"imp":20.6126579539,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":4.1675681573,"imp":18.7474169543,"elo":1328.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":11.5413095581,"imp":18.2575995706,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":22.4971330897,"imp":17.9851452587,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":58.5886356918,"imp":17.7740802827,"elo":1385.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":117.0352585562,"imp":17.71457341,"elo":1388.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":234.6013258332,"imp":17.6648119772,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":351.8557074643,"imp":17.6284343924,"elo":1394.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":473.0394868517,"imp":17.6058992627,"elo":1396.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":210.3752328358,"imp":23.1712104705,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":506.0952070543,"imp":21.0877128247,"elo":1239.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":1471.9943120247,"imp":18.7858348625,"elo":1329.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":3204.4293403006,"imp":17.6476410912,"elo":1380.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":7937.3877327789,"imp":16.5400463512,"elo":1435.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":8305.0663055001,"imp":16.5161770818,"elo":1437.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.8035263333,"imp":30.5503235546,"elo":996.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":6.6149535103,"imp":26.2377125098,"elo":1065.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":16.126759698,"imp":21.0518574285,"elo":1213.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":31.9971507614,"imp":19.073006682,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":79.9925409669,"imp":18.4795330163,"elo":1345.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":159.8382634935,"imp":18.3421741651,"elo":1351.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":319.7915016575,"imp":18.3790267956,"elo":1346.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":479.1842606686,"imp":18.4265861599,"elo":1343.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":642.4042496517,"imp":18.4707856948,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7076126445,"x_train":250.3856225541,"imp":18.4933189805,"elo":1397.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0203827935,"x_train":3.0406505418,"imp":16.5104471714,"elo":1426.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0439021895,"x_train":7.9183650076,"imp":16.4167776727,"elo":1432.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0872974817,"x_train":18.1594886071,"imp":16.3645233167,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1177578623,"x_train":36.3465094416,"imp":16.3892082295,"elo":1438.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1576011846,"x_train":92.6292744451,"imp":16.3943655086,"elo":1438.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1905796825,"x_train":181.7510780205,"imp":16.4086167429,"elo":1437.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2216909575,"x_train":367.6200398568,"imp":16.4710396294,"elo":1432.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.233726561,"x_train":549.0966445079,"imp":16.5097848779,"elo":1429.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2505091082,"x_train":738.2612405475,"imp":16.5372195649,"elo":1427.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0261083185,"x_train":1.2471436483,"imp":19.0829733294,"elo":1336.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0519030406,"x_train":2.6592018735,"imp":18.6812379658,"elo":1364.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1047084686,"x_train":6.827854614,"imp":18.4929052129,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1608995496,"x_train":13.7690635449,"imp":18.3889370073,"elo":1380.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2235785875,"x_train":34.8312631967,"imp":18.2379336969,"elo":1388.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.268731958,"x_train":69.9560479767,"imp":18.1477267418,"elo":1393.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2953259759,"x_train":140.0107160319,"imp":18.0536408742,"elo":1398.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3063723891,"x_train":209.935077018,"imp":18.0111503293,"elo":1398.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3135203066,"x_train":281.3883698637,"imp":17.9898475453,"elo":1399.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0148652021,"x_train":3.2701795229,"imp":21.1550409794,"elo":1254.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0298750396,"x_train":8.5418580508,"imp":20.7461328006,"elo":1268.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0523006922,"x_train":21.5154250642,"imp":20.3672145186,"elo":1286.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0707230342,"x_train":44.8133243453,"imp":20.2168351975,"elo":1293.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0955594836,"x_train":115.6024980549,"imp":20.0572180124,"elo":1302.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1108747895,"x_train":235.0233869244,"imp":19.919349256,"elo":1310.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1241247455,"x_train":467.0399506352,"imp":19.7939126807,"elo":1316.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1273889013,"x_train":700.0647480226,"imp":19.7572455867,"elo":1319.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1283935846,"x_train":936.7523176703,"imp":19.7105302771,"elo":1322.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1905448885,"x_train":1.0867260517,"imp":27.4001748997,"elo":1005.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2264116831,"x_train":1.9830559575,"imp":25.6735585024,"elo":1046.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2899539484,"x_train":4.7045699739,"imp":23.2477333054,"elo":1120.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3380187835,"x_train":9.2122172375,"imp":22.3653758851,"elo":1158.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4229346143,"x_train":22.7973551316,"imp":21.912093139,"elo":1178.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5030792517,"x_train":45.453079637,"imp":21.7483992588,"elo":1188.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5833685615,"x_train":90.7421763607,"imp":21.6454460116,"elo":1194.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6430515372,"x_train":136.0852569857,"imp":21.6176267391,"elo":1198.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6917689999,"x_train":182.354525564,"imp":21.6501981557,"elo":1198.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2670364168,"x_train":2.556542435,"imp":25.3313477134,"elo":1076.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4726706959,"x_train":5.2460669855,"imp":23.9098892442,"elo":1117.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9145377265,"x_train":13.8120709411,"imp":22.0691811008,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4648401665,"x_train":28.1488717871,"imp":21.1656381222,"elo":1219.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0889905192,"x_train":68.1608030808,"imp":20.4838923587,"elo":1245.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.6339038812,"x_train":137.2584674121,"imp":20.155553068,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.3191961162,"x_train":277.1007160478,"imp":19.9634823566,"elo":1268.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.7369178253,"x_train":416.6378564081,"imp":19.8436916216,"elo":1275.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0812529125,"x_train":559.3818159963,"imp":19.8128283176,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0925963767,"x_train":0.8126978053,"imp":21.3359565942,"elo":1186.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4060297968,"x_train":2.859374793,"imp":19.3232189963,"elo":1284.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6600545749,"x_train":7.6235569739,"imp":18.0571797374,"elo":1369.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8849225155,"x_train":15.609269256,"imp":17.4638234753,"elo":1410.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1404380746,"x_train":40.8449535661,"imp":17.0877760749,"elo":1432.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2644666214,"x_train":81.9221899435,"imp":16.9569657046,"elo":1441.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3003023736,"x_train":163.7675316385,"imp":16.8463684769,"elo":1448.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2279060483,"x_train":244.6291470871,"imp":16.806397822,"elo":1449.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1375151455,"x_train":328.9554302437,"imp":16.8068902988,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.2986429509,"x_train":14.2586021738,"imp":18.2681832537,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.310422718,"x_train":13.5066187328,"imp":20.5987376578,"elo":1282.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8103899414,"x_train":29.3976468222,"imp":19.3828809105,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8318086975,"x_train":85.706508499,"imp":17.4028059878,"elo":1402.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7115644528,"x_train":168.9559526652,"imp":17.0083157792,"elo":1417.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.0880667933,"x_train":408.9287781105,"imp":16.6342936931,"elo":1432.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.4716607961,"x_train":844.0960252495,"imp":16.8917745728,"elo":1421.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.9921274944,"x_train":1679.9964089232,"imp":17.4242403311,"elo":1402.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8596602241,"x_train":2502.3959153065,"imp":17.8701238413,"elo":1387.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3848981892,"x_train":3383.9090129865,"imp":18.1123460904,"elo":1378.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7050443034,"x_train":14.4628692945,"imp":24.138681705,"elo":1158.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0291154934,"x_train":15.7950255214,"imp":31.7917728153,"elo":947.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2360786,"x_train":37.0036028141,"imp":29.4275925671,"elo":1009.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6732920785,"x_train":98.9757470004,"imp":25.6585264214,"elo":1107.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7725893154,"x_train":117.8541822739,"imp":25.5415688376,"elo":1114.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.02947866,"x_train":0.2544731196,"imp":27.7387153293,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0884016237,"x_train":1.1022482459,"imp":25.4427953271,"elo":1058.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1791844925,"x_train":3.7036683225,"imp":24.2985007049,"elo":1100.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2574024682,"x_train":7.9903720898,"imp":23.7402292325,"elo":1125.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3681557693,"x_train":20.5657356408,"imp":23.2882960277,"elo":1147.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.462216789,"x_train":41.6049917756,"imp":23.1528100801,"elo":1155.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5500877206,"x_train":83.9433065314,"imp":23.0662800175,"elo":1160.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5963022541,"x_train":126.3785182217,"imp":23.0262199814,"elo":1162.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6073439144,"x_train":169.7548111684,"imp":23.0376221464,"elo":1163.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1163701521,"x_train":9.3300735239,"imp":19.5026013508,"elo":1272.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5367213253,"x_train":23.3699912915,"imp":17.6174972413,"elo":1374.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.4240428043,"x_train":64.116353982,"imp":16.3676580292,"elo":1446.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.3826374553,"x_train":135.6738391001,"imp":15.9061199923,"elo":1472.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.7349492529,"x_train":343.9787328013,"imp":15.5519904832,"elo":1493.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.2159141116,"x_train":692.1853967506,"imp":15.4217112457,"elo":1502.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.3769693441,"x_train":1383.1114451594,"imp":15.325694052,"elo":1507.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.3169488161,"x_train":2072.6413859802,"imp":15.283515667,"elo":1510.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.7763731411,"x_train":2775.9833627967,"imp":15.2636046028,"elo":1510.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6942704557,"x_train":4.9087216863,"imp":13.8861384299,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7020387497,"x_train":16.4525053202,"imp":13.8966382852,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.0703931915,"x_train":44.7436120415,"imp":13.7257647431,"elo":1555.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1355887912,"x_train":91.7548974847,"imp":13.661718684,"elo":1556.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7695458497,"x_train":236.7117395728,"imp":13.5127479641,"elo":1565.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.7430322432,"x_train":474.1680372395,"imp":13.4830451944,"elo":1568.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9819714023,"x_train":957.1717702231,"imp":13.4318526315,"elo":1571.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7612085046,"x_train":1438.8272594182,"imp":13.432059224,"elo":1570.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.2115603982,"x_train":1924.5007092935,"imp":13.4551080371,"elo":1567.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.166690434,"x_train":12.8582127474,"imp":20.7986247854,"elo":1307.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"x_train":49.671607539,"imp":20.0548159955,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":59.5820773592,"x_train":74.371713941,"imp":18.9352451494,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":88.9924523268,"x_train":135.8465940794,"imp":17.8796770802,"elo":1361.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":119.9760691608,"x_train":245.5890330499,"imp":17.5153156897,"elo":1373.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":184.7836530334,"x_train":603.2630243511,"imp":16.6880388295,"elo":1403.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":246.0445440803,"x_train":1228.1552324622,"imp":16.2966575311,"elo":1418.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":305.831978607,"x_train":2466.8687838379,"imp":16.0640662704,"elo":1426.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":338.3243282561,"x_train":3715.4039160262,"imp":15.8960164204,"elo":1431.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":357.7828705921,"x_train":4982.6619115707,"imp":15.8278609368,"elo":1433.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"x_train":1.4436537137,"imp":16.0168834038,"elo":1457.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"x_train":36.4454770138,"imp":4.0882031143,"elo":1815.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.128947975,"x_train":0.8504331063,"imp":29.0978789966,"elo":1063.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.158249445,"x_train":1.7686337423,"imp":11.5280550051,"elo":1625.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2075493916,"x_train":5.0292797696,"imp":18.6200249426,"elo":1324.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.3669465558,"x_train":12.3959630427,"imp":18.1075360361,"elo":1351.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.641044718,"x_train":37.7043962113,"imp":17.420210667,"elo":1390.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.8182662875,"x_train":76.792938919,"imp":16.82113508,"elo":1429.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.0673724766,"x_train":197.2453170668,"imp":16.3082031884,"elo":1455.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.2587716295,"x_train":398.0123545957,"imp":16.0339563294,"elo":1471.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.5504237308,"x_train":792.8708351593,"imp":15.8413077925,"elo":1482.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.7273482778,"x_train":1198.5048285012,"imp":15.7531162122,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":1.8903034491,"x_train":1612.0705710762,"imp":15.708241152,"elo":1492.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4644557714,"x_train":4.8153059659,"imp":12.7851447325,"elo":1624.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3605764018,"x_train":2.62612376,"imp":11.3977989809,"elo":1698.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0517146513,"x_train":391.1080876361,"imp":29.2171330577,"elo":1076.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5372670149,"x_train":569.6608345335,"imp":29.1020946138,"elo":1082.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.680399048,"x_train":976.9745445733,"imp":28.8625681047,"elo":1089.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5405477526,"x_train":1509.8128718211,"imp":28.2069059965,"elo":1104.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5104627144,"x_train":3235.1736621912,"imp":27.0760733386,"elo":1126.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.5720442233,"x_train":6343.9144626669,"imp":25.9428567016,"elo":1146.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.437201778,"x_train":12590.6623522261,"imp":25.326744153,"elo":1161.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":13.8939261019,"x_train":18827.9881620466,"imp":24.9309639133,"elo":1171.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0865396805,"x_train":25373.3350332343,"imp":24.6087815593,"elo":1181.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0586280558,"x_train":1.1266327909,"imp":17.4898874411,"elo":1395.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0914684387,"x_train":4.9502582412,"imp":23.9989774263,"elo":1109.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1917076474,"x_train":13.9663495618,"imp":21.9556733634,"elo":1172.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4742459424,"x_train":44.495543604,"imp":20.314403314,"elo":1226.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7489571843,"x_train":110.9528145485,"imp":19.6459835949,"elo":1258.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1980104117,"x_train":276.2432638721,"imp":19.0715601782,"elo":1287.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6220136736,"x_train":564.391803329,"imp":18.7085856134,"elo":1308.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8953030401,"x_train":1102.6775922424,"imp":18.5179870159,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.2031294127,"x_train":1669.2077718784,"imp":18.4068762129,"elo":1329.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4133616403,"x_train":2241.9244610212,"imp":18.3030306436,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1102748358,"x_train":1.2379804347,"imp":20.6126579539,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2662387933,"x_train":4.1675681573,"imp":18.7474169543,"elo":1328.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3317369138,"x_train":11.5413095581,"imp":18.2575995706,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3713912294,"x_train":22.4971330897,"imp":17.9851452587,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4711698488,"x_train":58.5886356918,"imp":17.7740802827,"elo":1385.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5335858253,"x_train":117.0352585562,"imp":17.71457341,"elo":1388.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5867602292,"x_train":234.6013258332,"imp":17.6648119772,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6015304474,"x_train":351.8557074643,"imp":17.6284343924,"elo":1394.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6276540518,"x_train":473.0394868517,"imp":17.6058992627,"elo":1396.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":43.847258449,"x_train":210.3752328358,"imp":23.1712104705,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":92.0045040051,"x_train":506.0952070543,"imp":21.0877128247,"elo":1239.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":160.2377449485,"x_train":1471.9943120247,"imp":18.7858348625,"elo":1329.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":248.0978697609,"x_train":3204.4293403006,"imp":17.6476410912,"elo":1380.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":328.1842391713,"x_train":7937.3877327789,"imp":16.5400463512,"elo":1435.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":336.0020690595,"x_train":8305.0663055001,"imp":16.5161770818,"elo":1437.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.5638730428,"x_train":2.8035263333,"imp":30.5503235546,"elo":996.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.803767519,"x_train":6.6149535103,"imp":26.2377125098,"elo":1065.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.9568557255,"x_train":16.126759698,"imp":21.0518574285,"elo":1213.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.3168671243,"x_train":31.9971507614,"imp":19.073006682,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.7124473326,"x_train":79.9925409669,"imp":18.4795330163,"elo":1345.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2004675665,"x_train":159.8382634935,"imp":18.3421741651,"elo":1351.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.497744625,"x_train":319.7915016575,"imp":18.3790267956,"elo":1346.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.6993276206,"x_train":479.1842606686,"imp":18.4265861599,"elo":1343.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.7587528196,"x_train":642.4042496517,"imp":18.4707856948,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_all/winrate_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..8d36ff76f8f0e4f31193f61755b9f423273083c4
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_binary/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "TabDPT-Turbo (default)", "AutoGluon 1.4 (best, 4h)", "RealMLP (tuned + ensembled)", "TabM (tuned + ensembled)", "TabDPT (tuned + ensembled)", "TabSwift (default)", "LimiX (default)", "LightGBM (tuned + ensembled)", "BetaTabPFN (default)", "TabM (tuned)", "TabDPT (tuned)", "RealMLP (tuned)", "LightGBM (tuned)", "iLTM (tuned + ensembled)", "CatBoost (default)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "ModernNCA (tuned)", "TabDPT (default)", "xRFM (tuned + ensembled)", "xRFM (tuned)", "XGBoost (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "SAP-RPT-OSS (default)", "EBM (tuned + ensembled)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "ModernNCA (default)", "XGBoost (tuned)", "TabM (default)", "ChimeraBoost (tuned)", "RealMLP (default)", "EBM (tuned)", "ModernNCA (tuned + ensembled)", "EBM (default)", "ChimeraBoost (default)", "iLTM (tuned)", "FastaiMLP (tuned)", "TabSTAR (tuned)", "TorchMLP (tuned)", "TabSTAR (tuned + ensembled)", "OrionMSP (default)", "ExtraTrees (tuned + ensembled)", "ExtraTrees (tuned)", "XGBoost (default)", "iLTM (default)", "TabFlex (default)", "LightGBM (default)", "TabSTAR (default)", "Linear (tuned + ensembled)", "TorchMLP (default)", "RandomForest (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "Linear (tuned)", "FastaiMLP (default)", "RandomForest (tuned)", "PerpetualBooster (tuned)", "Linear (default)", "KNN (tuned + ensembled)", "xRFM (default)", "ExtraTrees (default)", "RandomForest (default)", "PerpetualBooster (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7968036532402039, 0.7557077407836914, 0.810502290725708, 0.8196346759796143, 0.8150684833526611, 0.8299086689949036, 0.835616409778595, 0.8378995656967163, 0.8561643958091736, 0.8767123222351074, 0.9063926935195923, 0.8972602486610413, 0.8675799369812012, 0.905251145362854, 0.888127863407135, 0.8664383292198181, 0.9383561611175537, 0.8926940560340881, 0.8721461296081543, 0.9109588861465454, 0.9041095972061157, 0.8949771523475647, 0.8926940560340881, 0.9018265008926392, 0.913241982460022, 0.9018265008926392, 0.9041095972061157, 0.8835616707801819, 0.9360730648040771, 0.9178082346916199, 0.8904109597206116, 0.9086757898330688, 0.8972602486610413, 0.9543378949165344, 0.8972602486610413, 0.9452054500579834, 0.9360730648040771, 0.9292237162590027, 0.913241982460022, 0.9178082346916199, 0.9406392574310303, 0.9109588861465454, 0.9383561611175537, 0.9269406199455261, 0.9269406199455261, 0.8196346759796143, 0.9680365324020386, 0.8219178318977356, 0.9429223537445068, 0.9520547986030579, 0.9429223537445068, 0.9703196287155151, 0.9406392574310303, 0.9269406199455261, 0.9726027250289917, 0.853881299495697, 0.9246575236320496, 0.9771689772605896, 0.9771689772605896, 0.9817351698875427, 0.9337899684906006, 0.956620991230011, 0.9748858213424683, 0.9840182662010193, 0.9543378949165344, 0.9726027250289917, 0.9748858213424683, 0.9840182662010193, 0.9771689772605896, 0.9885844588279724, 0.9726027250289917, 0.9931507110595703], [0.20319634675979614, NaN, 0.5958904027938843, 0.5114155411720276, 0.655251145362854, 0.7031963467597961, 0.6940639019012451, 0.7191780805587769, 0.7351598143577576, 0.7191780805587769, 0.7899543642997742, 0.7922374606132507, 0.8196346759796143, 0.8310502171516418, 0.7739726305007935, 0.7853881120681763, 0.7728310227394104, 0.8378995656967163, 0.8493150472640991, 0.810502290725708, 0.8424657583236694, 0.8493150472640991, 0.8310502171516418, 0.8242009282112122, 0.8333333134651184, 0.8607305884361267, 0.8470319509506226, 0.8424657583236694, 0.8150684833526611, 0.8481734991073608, 0.8744292259216309, 0.8287671208381653, 0.844748854637146, 0.8413242101669312, 0.888127863407135, 0.8504565954208374, 0.8721461296081543, 0.8904109597206116, 0.878995418548584, 0.8812785148620605, 0.8630136847496033, 0.9109588861465454, 0.8641552329063416, 0.8698630332946777, 0.905251145362854, 0.8732876777648926, 0.784246563911438, 0.9337899684906006, 0.7968036532402039, 0.9018265008926392, 0.9383561611175537, 0.9337899684906006, 0.9406392574310303, 0.9292237162590027, 0.9041095972061157, 0.9543378949165344, 0.8310502171516418, 0.878995418548584, 0.9474886059761047, 0.9383561611175537, 0.9360730648040771, 0.8904109597206116, 0.9406392574310303, 0.9611872434616089, 0.9520547986030579, 0.931506872177124, 0.9383561611175537, 0.9543378949165344, 0.9703196287155151, 0.9748858213424683, 0.9703196287155151, 0.9634703397750854, 0.9954338073730469], [0.2442922443151474, 0.4041095972061157, NaN, 0.456620991230011, 0.6484017968177795, 0.6301369667053223, 0.6506849527359009, 0.664383590221405, 0.6073059439659119, 0.77625572681427, 0.8162100315093994, 0.7990867495536804, 0.77625572681427, 0.767123281955719, 0.7465753555297852, 0.8310502171516418, 0.7579908967018127, 0.818493127822876, 0.7945205569267273, 0.8470319509506226, 0.8401826620101929, 0.8744292259216309, 0.8436073064804077, 0.853881299495697, 0.8436073064804077, 0.8698630332946777, 0.77625572681427, 0.8675799369812012, 0.8230593800544739, 0.878995418548584, 0.888127863407135, 0.8036529421806335, 0.8401826620101929, 0.8561643958091736, 0.8995434045791626, 0.835616409778595, 0.8926940560340881, 0.878995418548584, 0.8972602486610413, 0.853881299495697, 0.8470319509506226, 0.9337899684906006, 0.8561643958091736, 0.9018265008926392, 0.8926940560340881, 0.8652967810630798, 0.7465753555297852, 0.9200913310050964, 0.7488584518432617, 0.905251145362854, 0.9246575236320496, 0.9086757898330688, 0.9337899684906006, 0.922374427318573, 0.8812785148620605, 0.9474886059761047, 0.801369845867157, 0.8744292259216309, 0.9474886059761047, 0.9429223537445068, 0.9360730648040771, 0.9018265008926392, 0.9383561611175537, 0.9452054500579834, 0.9509132504463196, 0.9257990717887878, 0.9269406199455261, 0.9429223537445068, 0.9497717022895813, 0.9520547986030579, 0.9634703397750854, 0.9452054500579834, 0.9931507110595703], [0.18949772417545319, 0.4885844886302948, 0.543379008769989, NaN, 0.6004565954208374, 0.6312785148620605, 0.646118700504303, 0.664383590221405, 0.7191780805587769, 0.7191780805587769, 0.7442922592163086, 0.7579908967018127, 0.8139269351959229, 0.8082191944122314, 0.784246563911438, 0.7648401856422424, 0.7739726305007935, 0.7945205569267273, 0.8127853870391846, 0.7716894745826721, 0.8036529421806335, 0.8150684833526611, 0.7968036532402039, 0.7853881120681763, 0.7945205569267273, 0.8333333134651184, 0.8401826620101929, 0.8173515796661377, 0.7910959124565125, 0.810502290725708, 0.8242009282112122, 0.801369845867157, 0.8196346759796143, 0.8401826620101929, 0.878995418548584, 0.8504565954208374, 0.8287671208381653, 0.8607305884361267, 0.8424657583236694, 0.8698630332946777, 0.844748854637146, 0.8698630332946777, 0.8561643958091736, 0.8378995656967163, 0.8584474921226501, 0.8812785148620605, 0.7922374606132507, 0.9178082346916199, 0.7899543642997742, 0.9246575236320496, 0.9121004343032837, 0.9109588861465454, 0.913241982460022, 0.9086757898330688, 0.8972602486610413, 0.9406392574310303, 0.835616409778595, 0.8949771523475647, 0.956620991230011, 0.9383561611175537, 0.9429223537445068, 0.9086757898330688, 0.931506872177124, 0.9360730648040771, 0.9520547986030579, 0.9372146129608154, 0.9474886059761047, 0.9520547986030579, 0.9520547986030579, 0.956620991230011, 0.9794520735740662, 0.9589040875434875, 0.9954338073730469], [0.18036529421806335, 0.344748854637146, 0.35159817337989807, 0.3995433747768402, NaN, 0.5456621050834656, 0.534246563911438, 0.5673515796661377, 0.6050228476524353, 0.6575342416763306, 0.7214611768722534, 0.7214611768722534, 0.7214611768722534, 0.7146118879318237, 0.7054794430732727, 0.7557077407836914, 0.7214611768722534, 0.7785388231277466, 0.7305936217308044, 0.7831050157546997, 0.8127853870391846, 0.7945205569267273, 0.7739726305007935, 0.7808219194412231, 0.767123281955719, 0.8242009282112122, 0.7191780805587769, 0.8036529421806335, 0.7831050157546997, 0.7853881120681763, 0.8493150472640991, 0.8059360980987549, 0.801369845867157, 0.7899543642997742, 0.853881299495697, 0.8401826620101929, 0.8310502171516418, 0.8458904027938843, 0.8630136847496033, 0.8561643958091736, 0.8390411138534546, 0.8812785148620605, 0.8242009282112122, 0.8584474921226501, 0.8584474921226501, 0.8424657583236694, 0.7442922592163086, 0.913241982460022, 0.7557077407836914, 0.8515982031822205, 0.913241982460022, 0.905251145362854, 0.9360730648040771, 0.8995434045791626, 0.878995418548584, 0.9474886059761047, 0.7899543642997742, 0.8675799369812012, 0.9497717022895813, 0.9360730648040771, 0.931506872177124, 0.8926940560340881, 0.9109588861465454, 0.9543378949165344, 0.931506872177124, 0.9200913310050964, 0.9269406199455261, 0.9383561611175537, 0.9452054500579834, 0.9611872434616089, 0.9611872434616089, 0.9474886059761047, 0.990867555141449], [0.18493150174617767, 0.29680365324020386, 0.36986300349235535, 0.36872145533561707, 0.4543378949165344, NaN, 0.4817351698875427, 0.5251141786575317, 0.5388127565383911, 0.6506849527359009, 0.6621004343032837, 0.6780821681022644, 0.6826484203338623, 0.7534246444702148, 0.6849315166473389, 0.7214611768722534, 0.7294520735740662, 0.732876718044281, 0.715753436088562, 0.7442922592163086, 0.7648401856422424, 0.72374427318573, 0.7625570893287659, 0.7351598143577576, 0.7351598143577576, 0.7557077407836914, 0.7397260069847107, 0.7945205569267273, 0.7876712083816528, 0.7602739930152893, 0.7990867495536804, 0.801369845867157, 0.7945205569267273, 0.8059360980987549, 0.8264840245246887, 0.8059360980987549, 0.7990867495536804, 0.8082191944122314, 0.8150684833526611, 0.8264840245246887, 0.835616409778595, 0.7831050157546997, 0.8196346759796143, 0.8333333134651184, 0.8458904027938843, 0.853881299495697, 0.7785388231277466, 0.8972602486610413, 0.7819634675979614, 0.913241982460022, 0.8926940560340881, 0.8972602486610413, 0.888127863407135, 0.8926940560340881, 0.9029680490493774, 0.9406392574310303, 0.8127853870391846, 0.853881299495697, 0.9611872434616089, 0.8972602486610413, 0.931506872177124, 0.8652967810630798, 0.9086757898330688, 0.9337899684906006, 0.9452054500579834, 0.9292237162590027, 0.9086757898330688, 0.9429223537445068, 0.9452054500579834, 0.956620991230011, 0.9703196287155151, 0.9452054500579834, 0.9931507110595703], [0.17009133100509644, 0.3059360682964325, 0.3493150770664215, 0.35388126969337463, 0.465753436088562, 0.5182648301124573, NaN, 0.6255707740783691, 0.5319634675979614, 0.6598173379898071, 0.6712328791618347, 0.6712328791618347, 0.6666666865348816, 0.7009132504463196, 0.7031963467597961, 0.7260273694992065, 0.72374427318573, 0.7283105254173279, 0.7168949842453003, 0.7260273694992065, 0.7659817337989807, 0.7340182662010193, 0.7557077407836914, 0.7420091032981873, 0.7465753555297852, 0.801369845867157, 0.6735159754753113, 0.7785388231277466, 0.7625570893287659, 0.7625570893287659, 0.7865296602249146, 0.7728310227394104, 0.7785388231277466, 0.7773972749710083, 0.8219178318977356, 0.7945205569267273, 0.793379008769989, 0.7990867495536804, 0.8059360980987549, 0.810502290725708, 0.8093607425689697, 0.7990867495536804, 0.8127853870391846, 0.8242009282112122, 0.8196346759796143, 0.8470319509506226, 0.7465753555297852, 0.878995418548584, 0.7534246444702148, 0.8767123222351074, 0.8835616707801819, 0.8767123222351074, 0.888127863407135, 0.8812785148620605, 0.8744292259216309, 0.9337899684906006, 0.7899543642997742, 0.853881299495697, 0.9246575236320496, 0.913241982460022, 0.931506872177124, 0.8744292259216309, 0.913241982460022, 0.9246575236320496, 0.9383561611175537, 0.905251145362854, 0.9109588861465454, 0.9269406199455261, 0.9429223537445068, 0.9452054500579834, 0.9611872434616089, 0.931506872177124, 0.990867555141449], [0.16438356041908264, 0.28082191944122314, 0.33561643958091736, 0.33561643958091736, 0.4326483905315399, 0.47488585114479065, 0.37442922592163086, NaN, 0.5228310227394104, 0.6278538703918457, 0.6324200630187988, 0.6484017968177795, 0.6598173379898071, 0.6860730648040771, 0.6621004343032837, 0.7054794430732727, 0.7031963467597961, 0.7111872434616089, 0.7043378949165344, 0.7077625393867493, 0.7602739930152893, 0.7054794430732727, 0.7283105254173279, 0.7283105254173279, 0.72374427318573, 0.7465753555297852, 0.6757990717887878, 0.7522830963134766, 0.7442922592163086, 0.7442922592163086, 0.759132444858551, 0.7602739930152893, 0.7716894745826721, 0.7579908967018127, 0.8082191944122314, 0.7694063782691956, 0.767123281955719, 0.7853881120681763, 0.7922374606132507, 0.7968036532402039, 0.7876712083816528, 0.7888127565383911, 0.7945205569267273, 0.8059360980987549, 0.8036529421806335, 0.8310502171516418, 0.7374429106712341, 0.8767123222351074, 0.7488584518432617, 0.8721461296081543, 0.8687214851379395, 0.8721461296081543, 0.8812785148620605, 0.8698630332946777, 0.8607305884361267, 0.922374427318573, 0.7922374606132507, 0.8333333134651184, 0.922374427318573, 0.888127863407135, 0.9121004343032837, 0.8584474921226501, 0.8995434045791626, 0.913241982460022, 0.9246575236320496, 0.9098173379898071, 0.9041095972061157, 0.9200913310050964, 0.9337899684906006, 0.9406392574310303, 0.9634703397750854, 0.9337899684906006, 0.990867555141449], [0.16210044920444489, 0.26484018564224243, 0.39269405603408813, 0.28082191944122314, 0.3949771821498871, 0.4611872136592865, 0.4680365324020386, 0.4771689474582672, NaN, 0.534246563911438, 0.5730593800544739, 0.5422374606132507, 0.6506849527359009, 0.6883561611175537, 0.6232876777648926, 0.5890411138534546, 0.5730593800544739, 0.6073059439659119, 0.7077625393867493, 0.6678082346916199, 0.6232876777648926, 0.6529680490493774, 0.646118700504303, 0.6689497828483582, 0.664383590221405, 0.6506849527359009, 0.8093607425689697, 0.6940639019012451, 0.7100456357002258, 0.6803653240203857, 0.6849315166473389, 0.6826484203338623, 0.6940639019012451, 0.7374429106712341, 0.7283105254173279, 0.7260273694992065, 0.698630154132843, 0.698630154132843, 0.7214611768722534, 0.7511415481567383, 0.7203196287155151, 0.6803653240203857, 0.7374429106712341, 0.7351598143577576, 0.759132444858551, 0.7945205569267273, 0.7077625393867493, 0.7945205569267273, 0.7168949842453003, 0.8401826620101929, 0.8458904027938843, 0.878995418548584, 0.8173515796661377, 0.7990867495536804, 0.8721461296081543, 0.8607305884361267, 0.7899543642997742, 0.8082191944122314, 0.878995418548584, 0.8767123222351074, 0.8812785148620605, 0.8287671208381653, 0.9063926935195923, 0.8904109597206116, 0.9041095972061157, 0.9063926935195923, 0.9041095972061157, 0.9292237162590027, 0.9303653240203857, 0.9360730648040771, 0.965753436088562, 0.9520547986030579, 0.9931507110595703], [0.1438356190919876, 0.28082191944122314, 0.22374428808689117, 0.28082191944122314, 0.34246575832366943, 0.3493150770664215, 0.34018266201019287, 0.3721461296081543, 0.465753436088562, NaN, 0.5970319509506226, 0.5593607425689697, 0.543379008769989, 0.517123281955719, 0.4817351698875427, 0.6826484203338623, 0.5844748616218567, 0.6255707740783691, 0.568493127822876, 0.6963470578193665, 0.7214611768722534, 0.7123287916183472, 0.7100456357002258, 0.72374427318573, 0.732876718044281, 0.6575342416763306, 0.6050228476524353, 0.6952054500579834, 0.7077625393867493, 0.7831050157546997, 0.7454338073730469, 0.6164383292198181, 0.7283105254173279, 0.7351598143577576, 0.7831050157546997, 0.7031963467597961, 0.7831050157546997, 0.7442922592163086, 0.7785388231277466, 0.77625572681427, 0.7420091032981873, 0.72374427318573, 0.7648401856422424, 0.7945205569267273, 0.7945205569267273, 0.7785388231277466, 0.7031963467597961, 0.8287671208381653, 0.7123287916183472, 0.6974886059761047, 0.7990867495536804, 0.8082191944122314, 0.8972602486610413, 0.844748854637146, 0.7876712083816528, 0.9246575236320496, 0.7602739930152893, 0.836758017539978, 0.8858447670936584, 0.8926940560340881, 0.9178082346916199, 0.8641552329063416, 0.9086757898330688, 0.9029680490493774, 0.9200913310050964, 0.8926940560340881, 0.913241982460022, 0.9292237162590027, 0.9360730648040771, 0.9360730648040771, 0.9452054500579834, 0.9178082346916199, 0.9954338073730469], [0.12328767031431198, 0.21004566550254822, 0.1837899535894394, 0.2557077705860138, 0.2785388231277466, 0.3378995358943939, 0.3287671208381653, 0.3675799071788788, 0.4269406497478485, 0.40296804904937744, NaN, 0.5045661926269531, 0.5011415481567383, 0.4794520437717438, 0.46347030997276306, 0.6187214851379395, 0.5502282977104187, 0.5799086689949036, 0.5490867495536804, 0.7305936217308044, 0.6860730648040771, 0.6757990717887878, 0.6529680490493774, 0.6803653240203857, 0.6826484203338623, 0.6415525078773499, 0.5936073064804077, 0.6872146129608154, 0.6621004343032837, 0.732876718044281, 0.7054794430732727, 0.6255707740783691, 0.7351598143577576, 0.7477169036865234, 0.7283105254173279, 0.6757990717887878, 0.7739726305007935, 0.7420091032981873, 0.740867555141449, 0.8253424763679504, 0.7557077407836914, 0.6849315166473389, 0.7568492889404297, 0.7420091032981873, 0.7579908967018127, 0.7728310227394104, 0.6940639019012451, 0.7876712083816528, 0.7054794430732727, 0.7100456357002258, 0.8378995656967163, 0.827625572681427, 0.8721461296081543, 0.8561643958091736, 0.7648401856422424, 0.9018265008926392, 0.7488584518432617, 0.8264840245246887, 0.8926940560340881, 0.9109588861465454, 0.913241982460022, 0.8310502171516418, 0.9109588861465454, 0.9292237162590027, 0.9246575236320496, 0.8515982031822205, 0.9041095972061157, 0.9246575236320496, 0.9383561611175537, 0.9611872434616089, 0.965753436088562, 0.9246575236320496, 0.9931507110595703], [0.09360730648040771, 0.20776255428791046, 0.2009132355451584, 0.24200913310050964, 0.2785388231277466, 0.3219178020954132, 0.3287671208381653, 0.35159817337989807, 0.45776256918907166, 0.4406392574310303, 0.4954337775707245, NaN, 0.5376712083816528, 0.5456621050834656, 0.46347030997276306, 0.611872136592865, 0.5913242101669312, 0.6746575236320496, 0.5639269351959229, 0.6347032189369202, 0.6655251383781433, 0.6267123222351074, 0.6666666865348816, 0.6529680490493774, 0.6575342416763306, 0.6609588861465454, 0.5890411138534546, 0.6050228476524353, 0.5821917653083801, 0.6963470578193665, 0.7146118879318237, 0.6301369667053223, 0.72374427318573, 0.6735159754753113, 0.689497709274292, 0.6940639019012451, 0.7557077407836914, 0.8299086689949036, 0.767123281955719, 0.7682648301124573, 0.767123281955719, 0.7077625393867493, 0.7568492889404297, 0.7853881120681763, 0.7442922592163086, 0.7351598143577576, 0.7054794430732727, 0.7865296602249146, 0.7100456357002258, 0.6940639019012451, 0.8173515796661377, 0.7990867495536804, 0.853881299495697, 0.8493150472640991, 0.767123281955719, 0.905251145362854, 0.7534246444702148, 0.8333333134651184, 0.8904109597206116, 0.8652967810630798, 0.888127863407135, 0.8481734991073608, 0.8721461296081543, 0.8858447670936584, 0.9109588861465454, 0.878995418548584, 0.8812785148620605, 0.8858447670936584, 0.922374427318573, 0.922374427318573, 0.965753436088562, 0.9086757898330688, 0.9817351698875427], [0.10273972898721695, 0.18036529421806335, 0.22374428808689117, 0.18607306480407715, 0.2785388231277466, 0.3173516094684601, 0.3333333432674408, 0.34018266201019287, 0.3493150770664215, 0.456620991230011, 0.4988584518432617, 0.4623287618160248, NaN, 0.517123281955719, 0.5057077407836914, 0.5251141786575317, 0.4954337775707245, 0.52625572681427, 0.6392694115638733, 0.5901826620101929, 0.5753424763679504, 0.5981734991073608, 0.5821917653083801, 0.5936073064804077, 0.5913242101669312, 0.6027397513389587, 0.6484017968177795, 0.628995418548584, 0.6175799369812012, 0.6015982031822205, 0.6232876777648926, 0.6415525078773499, 0.6221461296081543, 0.664383590221405, 0.681506872177124, 0.663241982460022, 0.6529680490493774, 0.628995418548584, 0.6575342416763306, 0.6621004343032837, 0.6518265008926392, 0.6849315166473389, 0.646118700504303, 0.689497709274292, 0.72374427318573, 0.7351598143577576, 0.655251145362854, 0.7374429106712341, 0.6575342416763306, 0.7831050157546997, 0.7945205569267273, 0.801369845867157, 0.7808219194412231, 0.801369845867157, 0.7968036532402039, 0.8310502171516418, 0.7602739930152893, 0.7511415481567383, 0.8401826620101929, 0.8561643958091736, 0.8584474921226501, 0.7716894745826721, 0.8721461296081543, 0.8561643958091736, 0.8732876777648926, 0.8378995656967163, 0.8664383292198181, 0.9166666865348816, 0.9269406199455261, 0.9246575236320496, 0.9360730648040771, 0.9086757898330688, 0.9931507110595703], [0.13242009282112122, 0.16894976794719696, 0.232876718044281, 0.19178082048892975, 0.28538814187049866, 0.24657534062862396, 0.2990867495536804, 0.31392693519592285, 0.3116438388824463, 0.482876718044281, 0.5205479264259338, 0.4543378949165344, 0.482876718044281, NaN, 0.5251141786575317, 0.5228310227394104, 0.4954337775707245, 0.509132444858551, 0.543379008769989, 0.5958904027938843, 0.5502282977104187, 0.5981734991073608, 0.5639269351959229, 0.5753424763679504, 0.5901826620101929, 0.621004581451416, 0.6141552329063416, 0.6506849527359009, 0.6164383292198181, 0.611872136592865, 0.6278538703918457, 0.6232876777648926, 0.6324200630187988, 0.7054794430732727, 0.6598173379898071, 0.655251145362854, 0.6484017968177795, 0.6392694115638733, 0.6666666865348816, 0.6575342416763306, 0.655251145362854, 0.6598173379898071, 0.655251145362854, 0.7031963467597961, 0.7283105254173279, 0.7442922592163086, 0.7077625393867493, 0.7191780805587769, 0.7191780805587769, 0.844748854637146, 0.801369845867157, 0.7968036532402039, 0.7602739930152893, 0.7785388231277466, 0.888127863407135, 0.7888127565383911, 0.7876712083816528, 0.7899543642997742, 0.8607305884361267, 0.8059360980987549, 0.8652967810630798, 0.8150684833526611, 0.8470319509506226, 0.8242009282112122, 0.8652967810630798, 0.8573059439659119, 0.8515982031822205, 0.8904109597206116, 0.8858447670936584, 0.8949771523475647, 0.9246575236320496, 0.9383561611175537, 0.9931507110595703], [0.09474886208772659, 0.22602739930152893, 0.25342464447021484, 0.21575342118740082, 0.2945205569267273, 0.31506848335266113, 0.29680365324020386, 0.3378995358943939, 0.3767123222351074, 0.5182648301124573, 0.5365296602249146, 0.5365296602249146, 0.4942922294139862, 0.47488585114479065, NaN, 0.6027397513389587, 0.5182648301124573, 0.5490867495536804, 0.5068492889404297, 0.594748854637146, 0.621004581451416, 0.6952054500579834, 0.6301369667053223, 0.6415525078773499, 0.6392694115638733, 0.6278538703918457, 0.5616438388824463, 0.6347032189369202, 0.6130136847496033, 0.6803653240203857, 0.6917808055877686, 0.6369863152503967, 0.6689497828483582, 0.638127863407135, 0.6826484203338623, 0.646118700504303, 0.7031963467597961, 0.6506849527359009, 0.7123287916183472, 0.6621004343032837, 0.6849315166473389, 0.6917808055877686, 0.6803653240203857, 0.7191780805587769, 0.7009132504463196, 0.7111872434616089, 0.638127863407135, 0.7397260069847107, 0.6369863152503967, 0.7271689772605896, 0.7579908967018127, 0.7557077407836914, 0.77625572681427, 0.7694063782691956, 0.7146118879318237, 0.7968036532402039, 0.7043378949165344, 0.7716894745826721, 0.7808219194412231, 0.8036529421806335, 0.8219178318977356, 0.7968036532402039, 0.8287671208381653, 0.8242009282112122, 0.844748854637146, 0.8036529421806335, 0.8082191944122314, 0.888127863407135, 0.878995418548584, 0.8972602486610413, 0.8949771523475647, 0.8515982031822205, 0.965753436088562], [0.11187214404344559, 0.21461187303066254, 0.16894976794719696, 0.23515981435775757, 0.2442922443151474, 0.2785388231277466, 0.27397260069847107, 0.2945205569267273, 0.4109589159488678, 0.3173516094684601, 0.38127854466438293, 0.388127863407135, 0.47488585114479065, 0.4771689474582672, 0.39726027846336365, NaN, 0.4452054798603058, 0.4714611768722534, 0.5205479264259338, 0.5114155411720276, 0.646118700504303, 0.5490867495536804, 0.6244292259216309, 0.621004581451416, 0.6244292259216309, 0.5525113940238953, 0.5936073064804077, 0.5913242101669312, 0.6095890402793884, 0.7351598143577576, 0.6061643958091736, 0.5890411138534546, 0.6541095972061157, 0.6004565954208374, 0.6073059439659119, 0.6141552329063416, 0.7853881120681763, 0.6952054500579834, 0.6621004343032837, 0.6621004343032837, 0.6963470578193665, 0.621004581451416, 0.7534246444702148, 0.7465753555297852, 0.6940639019012451, 0.6872146129608154, 0.681506872177124, 0.7260273694992065, 0.6849315166473389, 0.6757990717887878, 0.8070776462554932, 0.7899543642997742, 0.8698630332946777, 0.8493150472640991, 0.7077625393867493, 0.9497717022895813, 0.7602739930152893, 0.8173515796661377, 0.810502290725708, 0.8995434045791626, 0.8744292259216309, 0.8310502171516418, 0.7853881120681763, 0.922374427318573, 0.8972602486610413, 0.8812785148620605, 0.8995434045791626, 0.8470319509506226, 0.922374427318573, 0.9520547986030579, 0.9452054500579834, 0.9212328791618347, 0.990867555141449], [0.1335616409778595, 0.2271689474582672, 0.24200913310050964, 0.22602739930152893, 0.2785388231277466, 0.2705479562282562, 0.27625569701194763, 0.29680365324020386, 0.4269406497478485, 0.4155251085758209, 0.4497717022895813, 0.40867578983306885, 0.5045661926269531, 0.5045661926269531, 0.4817351698875427, 0.5547945499420166, NaN, 0.4954337775707245, 0.5593607425689697, 0.5205479264259338, 0.5616438388824463, 0.5479452013969421, 0.6050228476524353, 0.5730593800544739, 0.5970319509506226, 0.5833333134651184, 0.5981734991073608, 0.5844748616218567, 0.5639269351959229, 0.6347032189369202, 0.628995418548584, 0.6027397513389587, 0.655251145362854, 0.6130136847496033, 0.6278538703918457, 0.6392694115638733, 0.655251145362854, 0.6803653240203857, 0.655251145362854, 0.655251145362854, 0.6780821681022644, 0.5958904027938843, 0.7305936217308044, 0.689497709274292, 0.6438356041908264, 0.7031963467597961, 0.7260273694992065, 0.7123287916183472, 0.732876718044281, 0.6940639019012451, 0.7922374606132507, 0.7922374606132507, 0.7648401856422424, 0.8127853870391846, 0.7614155411720276, 0.818493127822876, 0.7888127565383911, 0.8310502171516418, 0.810502290725708, 0.8424657583236694, 0.8287671208381653, 0.8561643958091736, 0.7922374606132507, 0.8424657583236694, 0.844748854637146, 0.9109588861465454, 0.9109588861465454, 0.8150684833526611, 0.8630136847496033, 0.8926940560340881, 0.9200913310050964, 0.9497717022895813, 0.9840182662010193], [0.06164383515715599, 0.16210044920444489, 0.18150684237480164, 0.2054794579744339, 0.2214611917734146, 0.267123281955719, 0.2716895043849945, 0.2888127863407135, 0.39269405603408813, 0.37442922592163086, 0.42009133100509644, 0.32534247636795044, 0.47374430298805237, 0.49086758494377136, 0.4509132504463196, 0.5285388231277466, 0.5045661926269531, NaN, 0.5114155411720276, 0.534246563911438, 0.5707762837409973, 0.5525113940238953, 0.6004565954208374, 0.6050228476524353, 0.6095890402793884, 0.6278538703918457, 0.551369845867157, 0.5319634675979614, 0.5365296602249146, 0.6267123222351074, 0.6666666865348816, 0.5547945499420166, 0.611872136592865, 0.6244292259216309, 0.646118700504303, 0.6506849527359009, 0.655251145362854, 0.7100456357002258, 0.6826484203338623, 0.7146118879318237, 0.6803653240203857, 0.6529680490493774, 0.6963470578193665, 0.715753436088562, 0.6712328791618347, 0.7031963467597961, 0.6872146129608154, 0.7340182662010193, 0.6906392574310303, 0.6666666865348816, 0.7511415481567383, 0.7579908967018127, 0.8082191944122314, 0.8150684833526611, 0.7511415481567383, 0.835616409778595, 0.732876718044281, 0.801369845867157, 0.8744292259216309, 0.8333333134651184, 0.8835616707801819, 0.8173515796661377, 0.8242009282112122, 0.8310502171516418, 0.8972602486610413, 0.8493150472640991, 0.853881299495697, 0.8424657583236694, 0.8835616707801819, 0.8926940560340881, 0.956620991230011, 0.8835616707801819, 0.9726027250289917], [0.10730593651533127, 0.15068493783473969, 0.2054794579744339, 0.18721461296081543, 0.26940637826919556, 0.284246563911438, 0.2831050157546997, 0.2956621050834656, 0.29223743081092834, 0.43150684237480164, 0.4509132504463196, 0.43607306480407715, 0.3607305884361267, 0.456620991230011, 0.4931506812572479, 0.4794520437717438, 0.4406392574310303, 0.4885844886302948, NaN, 0.5136986374855042, 0.5136986374855042, 0.5365296602249146, 0.5479452013969421, 0.5547945499420166, 0.551369845867157, 0.5388127565383911, 0.6004565954208374, 0.586758017539978, 0.5502282977104187, 0.5673515796661377, 0.5844748616218567, 0.5707762837409973, 0.5958904027938843, 0.6141552329063416, 0.6347032189369202, 0.611872136592865, 0.5890411138534546, 0.577625572681427, 0.6050228476524353, 0.6255707740783691, 0.6278538703918457, 0.6267123222351074, 0.6278538703918457, 0.6232876777648926, 0.655251145362854, 0.6940639019012451, 0.603881299495697, 0.6963470578193665, 0.6050228476524353, 0.7785388231277466, 0.7397260069847107, 0.7602739930152893, 0.7123287916183472, 0.7351598143577576, 0.7785388231277466, 0.7511415481567383, 0.7374429106712341, 0.740867555141449, 0.810502290725708, 0.7853881120681763, 0.8378995656967163, 0.7716894745826721, 0.844748854637146, 0.8196346759796143, 0.8470319509506226, 0.8196346759796143, 0.8561643958091736, 0.8812785148620605, 0.8835616707801819, 0.8972602486610413, 0.9337899684906006, 0.8972602486610413, 0.9885844588279724], [0.1278538852930069, 0.18949772417545319, 0.15296803414821625, 0.2283104956150055, 0.2168949842453003, 0.2557077705860138, 0.27397260069847107, 0.29223743081092834, 0.3321917951107025, 0.30365297198295593, 0.26940637826919556, 0.3652968108654022, 0.40981733798980713, 0.4041095972061157, 0.405251145362854, 0.4885844886302948, 0.4794520437717438, 0.465753436088562, 0.48630136251449585, NaN, 0.5388127565383911, 0.4942922294139862, 0.543379008769989, 0.5136986374855042, 0.5228310227394104, 0.5228310227394104, 0.5228310227394104, 0.5388127565383911, 0.5559360980987549, 0.5936073064804077, 0.5570776462554932, 0.5479452013969421, 0.6141552329063416, 0.6324200630187988, 0.6324200630187988, 0.5981734991073608, 0.6495434045791626, 0.6392694115638733, 0.6027397513389587, 0.6940639019012451, 0.6598173379898071, 0.5821917653083801, 0.6278538703918457, 0.6506849527359009, 0.6438356041908264, 0.7100456357002258, 0.6712328791618347, 0.72374427318573, 0.6826484203338623, 0.6335616707801819, 0.7420091032981873, 0.7534246444702148, 0.77625572681427, 0.7922374606132507, 0.740867555141449, 0.810502290725708, 0.7351598143577576, 0.7557077407836914, 0.844748854637146, 0.8219178318977356, 0.8470319509506226, 0.7659817337989807, 0.8755707740783691, 0.8378995656967163, 0.8515982031822205, 0.8150684833526611, 0.844748854637146, 0.8675799369812012, 0.8493150472640991, 0.8812785148620605, 0.9474886059761047, 0.8721461296081543, 0.9794520735740662], [0.0890410989522934, 0.15753424167633057, 0.15981735289096832, 0.19634702801704407, 0.18721461296081543, 0.23515981435775757, 0.2340182662010193, 0.2397260218858719, 0.3767123222351074, 0.2785388231277466, 0.31392693519592285, 0.3344748914241791, 0.42465752363204956, 0.4497717022895813, 0.37899544835090637, 0.35388126969337463, 0.4383561611175537, 0.4292237460613251, 0.48630136251449585, 0.4611872136592865, NaN, 0.4817351698875427, 0.560502290725708, 0.5582191944122314, 0.5479452013969421, 0.4885844886302948, 0.5410959124565125, 0.5559360980987549, 0.5502282977104187, 0.6244292259216309, 0.5502282977104187, 0.5410959124565125, 0.621004581451416, 0.5696346759796143, 0.5570776462554932, 0.586758017539978, 0.6849315166473389, 0.6506849527359009, 0.5844748616218567, 0.6404109597206116, 0.6484017968177795, 0.5844748616218567, 0.6872146129608154, 0.689497709274292, 0.6324200630187988, 0.621004581451416, 0.664383590221405, 0.6655251383781433, 0.6689497828483582, 0.6575342416763306, 0.7568492889404297, 0.7659817337989807, 0.818493127822876, 0.8207762837409973, 0.6917808055877686, 0.914383590221405, 0.7420091032981873, 0.7831050157546997, 0.7922374606132507, 0.8401826620101929, 0.8504565954208374, 0.8070776462554932, 0.7648401856422424, 0.8744292259216309, 0.8721461296081543, 0.8584474921226501, 0.8630136847496033, 0.8253424763679504, 0.9063926935195923, 0.9246575236320496, 0.9429223537445068, 0.9086757898330688, 0.9931507110595703], [0.09589041024446487, 0.15068493783473969, 0.12557077407836914, 0.18493150174617767, 0.2054794579744339, 0.27625569701194763, 0.2659817337989807, 0.2945205569267273, 0.34703195095062256, 0.2876712381839752, 0.32420089840888977, 0.3732876777648926, 0.4018264710903168, 0.4018264710903168, 0.3047945201396942, 0.4509132504463196, 0.45205479860305786, 0.44748857617378235, 0.46347030997276306, 0.5057077407836914, 0.5182648301124573, NaN, 0.49771690368652344, 0.4771689474582672, 0.5068492889404297, 0.534246563911438, 0.5205479264259338, 0.5742009282112122, 0.5993150472640991, 0.5650684833526611, 0.5753424763679504, 0.5502282977104187, 0.5981734991073608, 0.6301369667053223, 0.6141552329063416, 0.6027397513389587, 0.6347032189369202, 0.5821917653083801, 0.5913242101669312, 0.664383590221405, 0.6324200630187988, 0.6050228476524353, 0.6392694115638733, 0.6358447670936584, 0.7111872434616089, 0.6883561611175537, 0.6347032189369202, 0.7054794430732727, 0.638127863407135, 0.6267123222351074, 0.7294520735740662, 0.7465753555297852, 0.810502290725708, 0.7899543642997742, 0.7351598143577576, 0.8584474921226501, 0.7191780805587769, 0.7191780805587769, 0.8173515796661377, 0.8401826620101929, 0.835616409778595, 0.7442922592163086, 0.8150684833526611, 0.8607305884361267, 0.8561643958091736, 0.7808219194412231, 0.8424657583236694, 0.9109588861465454, 0.878995418548584, 0.9041095972061157, 0.9086757898330688, 0.8801369667053223, 0.9748858213424683], [0.10502283275127411, 0.16894976794719696, 0.15639269351959229, 0.20319634675979614, 0.22602739930152893, 0.23744292557239532, 0.2442922443151474, 0.2716895043849945, 0.35388126969337463, 0.2899543344974518, 0.34703195095062256, 0.3333333432674408, 0.4178082048892975, 0.43607306480407715, 0.36986300349235535, 0.37557077407836914, 0.3949771821498871, 0.3995433747768402, 0.45205479860305786, 0.456620991230011, 0.439497709274292, 0.5022830963134766, NaN, 0.4931506812572479, 0.4942922294139862, 0.5068492889404297, 0.5136986374855042, 0.5376712083816528, 0.5547945499420166, 0.5696346759796143, 0.5650684833526611, 0.5662100315093994, 0.5890411138534546, 0.5639269351959229, 0.5936073064804077, 0.5456621050834656, 0.6130136847496033, 0.6187214851379395, 0.6255707740783691, 0.621004581451416, 0.6232876777648926, 0.586758017539978, 0.655251145362854, 0.6712328791618347, 0.664383590221405, 0.6484017968177795, 0.672374427318573, 0.7009132504463196, 0.6780821681022644, 0.6004565954208374, 0.7716894745826721, 0.7648401856422424, 0.8173515796661377, 0.8310502171516418, 0.6575342416763306, 0.8515982031822205, 0.7283105254173279, 0.7557077407836914, 0.7990867495536804, 0.8767123222351074, 0.7853881120681763, 0.77625572681427, 0.7876712083816528, 0.8812785148620605, 0.8242009282112122, 0.8242009282112122, 0.8858447670936584, 0.8310502171516418, 0.8744292259216309, 0.8995434045791626, 0.8926940560340881, 0.913241982460022, 0.9863013625144958], [0.10730593651533127, 0.17579908668994904, 0.14611871540546417, 0.21461187303066254, 0.21917808055877686, 0.26484018564224243, 0.25799086689949036, 0.2716895043849945, 0.33105021715164185, 0.27625569701194763, 0.31963470578193665, 0.34703195095062256, 0.4063926935195923, 0.42465752363204956, 0.35844749212265015, 0.37899544835090637, 0.4269406497478485, 0.3949771821498871, 0.4452054798603058, 0.48630136251449585, 0.44178083539009094, 0.5228310227394104, 0.5068492889404297, NaN, 0.5365296602249146, 0.5102739930152893, 0.5159817337989807, 0.5456621050834656, 0.5479452013969421, 0.5559360980987549, 0.5593607425689697, 0.5844748616218567, 0.5616438388824463, 0.5639269351959229, 0.586758017539978, 0.5410959124565125, 0.6084474921226501, 0.6050228476524353, 0.6095890402793884, 0.5958904027938843, 0.5924657583236694, 0.5821917653083801, 0.621004581451416, 0.6506849527359009, 0.6598173379898071, 0.6301369667053223, 0.6598173379898071, 0.6917808055877686, 0.655251145362854, 0.5958904027938843, 0.7694063782691956, 0.7636986374855042, 0.8219178318977356, 0.7865296602249146, 0.6415525078773499, 0.8630136847496033, 0.7168949842453003, 0.7511415481567383, 0.7831050157546997, 0.8698630332946777, 0.7945205569267273, 0.7716894745826721, 0.7808219194412231, 0.8892694115638733, 0.8196346759796143, 0.8059360980987549, 0.8858447670936584, 0.8287671208381653, 0.8904109597206116, 0.9269406199455261, 0.8858447670936584, 0.8949771523475647, 0.990867555141449], [0.09817351400852203, 0.1666666716337204, 0.15639269351959229, 0.2054794579744339, 0.232876718044281, 0.26484018564224243, 0.25342464447021484, 0.27625569701194763, 0.33561643958091736, 0.267123281955719, 0.3173516094684601, 0.34246575832366943, 0.40867578983306885, 0.40981733798980713, 0.3607305884361267, 0.37557077407836914, 0.40296804904937744, 0.3904109597206116, 0.44863012433052063, 0.4771689474582672, 0.45205479860305786, 0.4931506812572479, 0.5057077407836914, 0.46347030997276306, NaN, 0.48630136251449585, 0.5068492889404297, 0.5422374606132507, 0.534246563911438, 0.5753424763679504, 0.5353881120681763, 0.5707762837409973, 0.5707762837409973, 0.534246563911438, 0.5799086689949036, 0.5273972749710083, 0.6187214851379395, 0.594748854637146, 0.5901826620101929, 0.5936073064804077, 0.5936073064804077, 0.5844748616218567, 0.6347032189369202, 0.6347032189369202, 0.6232876777648926, 0.6312785148620605, 0.6621004343032837, 0.681506872177124, 0.6666666865348816, 0.5913242101669312, 0.7465753555297852, 0.7374429106712341, 0.8333333134651184, 0.7694063782691956, 0.6324200630187988, 0.8401826620101929, 0.72374427318573, 0.7465753555297852, 0.7648401856422424, 0.8698630332946777, 0.7808219194412231, 0.7614155411720276, 0.7648401856422424, 0.8698630332946777, 0.810502290725708, 0.7968036532402039, 0.8812785148620605, 0.8219178318977356, 0.8812785148620605, 0.9178082346916199, 0.8949771523475647, 0.9041095972061157, 0.990867555141449], [0.08675798773765564, 0.1392694115638733, 0.13013698160648346, 0.1666666716337204, 0.17579908668994904, 0.2442922443151474, 0.19863013923168182, 0.25342464447021484, 0.3493150770664215, 0.34246575832366943, 0.35844749212265015, 0.3390410840511322, 0.39726027846336365, 0.37899544835090637, 0.3721461296081543, 0.44748857617378235, 0.4166666567325592, 0.3721461296081543, 0.4611872136592865, 0.4771689474582672, 0.5114155411720276, 0.465753436088562, 0.4931506812572479, 0.4897260367870331, 0.5136986374855042, NaN, 0.49771690368652344, 0.5285388231277466, 0.5525113940238953, 0.5502282977104187, 0.5616438388824463, 0.5182648301124573, 0.5525113940238953, 0.5787671208381653, 0.568493127822876, 0.6232876777648926, 0.5673515796661377, 0.543379008769989, 0.5742009282112122, 0.6301369667053223, 0.5913242101669312, 0.5525113940238953, 0.6027397513389587, 0.6095890402793884, 0.611872136592865, 0.6735159754753113, 0.6369863152503967, 0.6963470578193665, 0.6438356041908264, 0.621004581451416, 0.6849315166473389, 0.6940639019012451, 0.7100456357002258, 0.7465753555297852, 0.7305936217308044, 0.7465753555297852, 0.7283105254173279, 0.7168949842453003, 0.7648401856422424, 0.7694063782691956, 0.7808219194412231, 0.7420091032981873, 0.7956621050834656, 0.7785388231277466, 0.8082191944122314, 0.7945205569267273, 0.810502290725708, 0.8721461296081543, 0.8493150472640991, 0.8470319509506226, 0.8835616707801819, 0.8561643958091736, 0.9771689772605896], [0.09817351400852203, 0.15296803414821625, 0.22374428808689117, 0.15981735289096832, 0.28082191944122314, 0.2602739632129669, 0.3264840245246887, 0.32420089840888977, 0.19063927233219147, 0.3949771821498871, 0.4063926935195923, 0.4109589159488678, 0.35159817337989807, 0.38584473729133606, 0.4383561611175537, 0.4063926935195923, 0.4018264710903168, 0.44863012433052063, 0.3995433747768402, 0.4771689474582672, 0.45890411734580994, 0.4794520437717438, 0.48630136251449585, 0.4840182662010193, 0.4931506812572479, 0.5022830963134766, NaN, 0.5273972749710083, 0.5136986374855042, 0.5136986374855042, 0.5707762837409973, 0.5296803712844849, 0.5182648301124573, 0.543379008769989, 0.5707762837409973, 0.577625572681427, 0.5410959124565125, 0.5662100315093994, 0.5753424763679504, 0.5353881120681763, 0.5559360980987549, 0.5479452013969421, 0.5936073064804077, 0.577625572681427, 0.5799086689949036, 0.6369863152503967, 0.5890411138534546, 0.6415525078773499, 0.5936073064804077, 0.7146118879318237, 0.7260273694992065, 0.732876718044281, 0.6712328791618347, 0.6963470578193665, 0.7876712083816528, 0.6712328791618347, 0.6757990717887878, 0.7283105254173279, 0.7853881120681763, 0.7374429106712341, 0.7694063782691956, 0.7442922592163086, 0.801369845867157, 0.7694063782691956, 0.7990867495536804, 0.7922374606132507, 0.8264840245246887, 0.8264840245246887, 0.835616409778595, 0.8424657583236694, 0.8926940560340881, 0.8721461296081543, 0.990867555141449], [0.09589041024446487, 0.15753424167633057, 0.13242009282112122, 0.1826484054327011, 0.19634702801704407, 0.2054794579744339, 0.2214611917734146, 0.24771688878536224, 0.3059360682964325, 0.3047945201396942, 0.31278538703918457, 0.3949771821498871, 0.37100455164909363, 0.3493150770664215, 0.3652968108654022, 0.40867578983306885, 0.4155251085758209, 0.4680365324020386, 0.41324201226234436, 0.4611872136592865, 0.4440639317035675, 0.42579910159111023, 0.4623287618160248, 0.4543378949165344, 0.45776256918907166, 0.4714611768722534, 0.4726027250289917, NaN, 0.5787671208381653, 0.4988584518432617, 0.509132444858551, 0.4954337775707245, 0.5707762837409973, 0.5810502171516418, 0.5114155411720276, 0.5273972749710083, 0.5319634675979614, 0.5570776462554932, 0.5296803712844849, 0.5913242101669312, 0.6164383292198181, 0.5730593800544739, 0.6095890402793884, 0.5388127565383911, 0.5559360980987549, 0.6575342416763306, 0.5993150472640991, 0.6232876777648926, 0.5981734991073608, 0.646118700504303, 0.7146118879318237, 0.7351598143577576, 0.7009132504463196, 0.7100456357002258, 0.7488584518432617, 0.7374429106712341, 0.689497709274292, 0.7694063782691956, 0.7990867495536804, 0.8036529421806335, 0.7785388231277466, 0.7899543642997742, 0.8493150472640991, 0.8242009282112122, 0.8321917653083801, 0.8230593800544739, 0.871004581451416, 0.9337899684906006, 0.9200913310050964, 0.9337899684906006, 0.913241982460022, 0.8812785148620605, 0.9885844588279724], [0.1164383590221405, 0.18493150174617767, 0.17694063484668732, 0.20890410244464874, 0.2168949842453003, 0.21232876181602478, 0.23744292557239532, 0.2557077705860138, 0.2899543344974518, 0.29223743081092834, 0.3378995358943939, 0.4178082048892975, 0.3824200928211212, 0.3835616409778595, 0.38698631525039673, 0.3904109597206116, 0.43607306480407715, 0.46347030997276306, 0.4497717022895813, 0.4440639317035675, 0.4497717022895813, 0.4006849229335785, 0.4452054798603058, 0.45205479860305786, 0.465753436088562, 0.44748857617378235, 0.48630136251449585, 0.4212328791618347, NaN, 0.4680365324020386, 0.517123281955719, 0.534246563911438, 0.551369845867157, 0.5582191944122314, 0.5251141786575317, 0.5547945499420166, 0.5148401856422424, 0.5502282977104187, 0.5388127565383911, 0.5799086689949036, 0.6073059439659119, 0.509132444858551, 0.6027397513389587, 0.5365296602249146, 0.5753424763679504, 0.6666666865348816, 0.6232876777648926, 0.6324200630187988, 0.6278538703918457, 0.621004581451416, 0.7191780805587769, 0.7488584518432617, 0.6369863152503967, 0.7260273694992065, 0.7648401856422424, 0.7009132504463196, 0.7168949842453003, 0.7945205569267273, 0.8150684833526611, 0.7511415481567383, 0.7568492889404297, 0.8242009282112122, 0.8378995656967163, 0.7899543642997742, 0.7979452013969421, 0.8493150472640991, 0.8698630332946777, 0.8972602486610413, 0.8630136847496033, 0.8652967810630798, 0.8949771523475647, 0.8926940560340881, 0.9885844588279724], [0.06392694264650345, 0.15182648599147797, 0.12100456655025482, 0.18949772417545319, 0.21461187303066254, 0.2397260218858719, 0.23744292557239532, 0.2557077705860138, 0.31963470578193665, 0.2168949842453003, 0.267123281955719, 0.30365297198295593, 0.39840182662010193, 0.388127863407135, 0.31963470578193665, 0.26484018564224243, 0.3652968108654022, 0.3732876777648926, 0.4326483905315399, 0.4063926935195923, 0.37557077407836914, 0.43493151664733887, 0.43036529421806335, 0.4440639317035675, 0.42465752363204956, 0.4497717022895813, 0.48630136251449585, 0.5011415481567383, 0.5319634675979614, NaN, 0.45890411734580994, 0.509132444858551, 0.5045661926269531, 0.5182648301124573, 0.5468036532402039, 0.5273972749710083, 0.6347032189369202, 0.5821917653083801, 0.5182648301124573, 0.5639269351959229, 0.5753424763679504, 0.5228310227394104, 0.5936073064804077, 0.560502290725708, 0.5844748616218567, 0.5753424763679504, 0.6347032189369202, 0.6187214851379395, 0.6369863152503967, 0.594748854637146, 0.7146118879318237, 0.7351598143577576, 0.8652967810630798, 0.7305936217308044, 0.6472602486610413, 0.8926940560340881, 0.698630154132843, 0.7397260069847107, 0.7465753555297852, 0.8801369667053223, 0.7397260069847107, 0.7511415481567383, 0.7214611768722534, 0.8755707740783691, 0.7773972749710083, 0.7945205569267273, 0.8401826620101929, 0.8333333134651184, 0.8744292259216309, 0.9200913310050964, 0.8767123222351074, 0.8824200630187988, 0.9794520735740662], [0.08219178020954132, 0.12557077407836914, 0.11187214404344559, 0.17579908668994904, 0.15068493783473969, 0.2009132355451584, 0.21347032487392426, 0.24086758494377136, 0.31506848335266113, 0.2545662224292755, 0.2945205569267273, 0.28538814187049866, 0.3767123222351074, 0.3721461296081543, 0.30821916460990906, 0.3938356041908264, 0.37100455164909363, 0.3333333432674408, 0.4155251085758209, 0.4429223835468292, 0.4497717022895813, 0.42465752363204956, 0.43493151664733887, 0.4406392574310303, 0.46461185812950134, 0.4383561611175537, 0.4292237460613251, 0.49086758494377136, 0.482876718044281, 0.5410959124565125, NaN, 0.5159817337989807, 0.5547945499420166, 0.5, 0.5205479264259338, 0.4817351698875427, 0.6015982031822205, 0.568493127822876, 0.6187214851379395, 0.5639269351959229, 0.5913242101669312, 0.5456621050834656, 0.5913242101669312, 0.7191780805587769, 0.5936073064804077, 0.560502290725708, 0.6575342416763306, 0.6164383292198181, 0.655251145362854, 0.5913242101669312, 0.6735159754753113, 0.6849315166473389, 0.767123281955719, 0.7374429106712341, 0.664383590221405, 0.8002282977104187, 0.7100456357002258, 0.7146118879318237, 0.7305936217308044, 0.7990867495536804, 0.7557077407836914, 0.7283105254173279, 0.7009132504463196, 0.8082191944122314, 0.7808219194412231, 0.767123281955719, 0.810502290725708, 0.7968036532402039, 0.835616409778595, 0.844748854637146, 0.8584474921226501, 0.8630136847496033, 0.9714611768722534], [0.10958904027938843, 0.17123287916183472, 0.19634702801704407, 0.19863013923168182, 0.1940639317035675, 0.19863013923168182, 0.2271689474582672, 0.2397260218858719, 0.3173516094684601, 0.3835616409778595, 0.37442922592163086, 0.36986300349235535, 0.35844749212265015, 0.3767123222351074, 0.36301368474960327, 0.4109589159488678, 0.39726027846336365, 0.4452054798603058, 0.4292237460613251, 0.45205479860305786, 0.45890411734580994, 0.4497717022895813, 0.4337899684906006, 0.4155251085758209, 0.4292237460613251, 0.4817351698875427, 0.47031962871551514, 0.5045661926269531, 0.465753436088562, 0.49086758494377136, 0.4840182662010193, NaN, 0.47488585114479065, 0.5616438388824463, 0.5570776462554932, 0.5410959124565125, 0.5068492889404297, 0.5228310227394104, 0.5159817337989807, 0.5913242101669312, 0.5114155411720276, 0.5319634675979614, 0.5468036532402039, 0.5319634675979614, 0.5308219194412231, 0.6392694115638733, 0.5639269351959229, 0.6324200630187988, 0.5639269351959229, 0.6301369667053223, 0.655251145362854, 0.689497709274292, 0.6255707740783691, 0.6529680490493774, 0.7465753555297852, 0.6780821681022644, 0.698630154132843, 0.7420091032981873, 0.7374429106712341, 0.706620991230011, 0.77625572681427, 0.7534246444702148, 0.7602739930152893, 0.7203196287155151, 0.7888127565383911, 0.8082191944122314, 0.7968036532402039, 0.8310502171516418, 0.8242009282112122, 0.8310502171516418, 0.8310502171516418, 0.878995418548584, 0.9794520735740662], [0.09132420271635056, 0.155251145362854, 0.15981735289096832, 0.18036529421806335, 0.19863013923168182, 0.2054794579744339, 0.2214611917734146, 0.2283104956150055, 0.3059360682964325, 0.2716895043849945, 0.26484018564224243, 0.27625569701194763, 0.3778538703918457, 0.3675799071788788, 0.33105021715164185, 0.3458904027938843, 0.344748854637146, 0.388127863407135, 0.4041095972061157, 0.38584473729133606, 0.37899544835090637, 0.4018264710903168, 0.4109589159488678, 0.4383561611175537, 0.4292237460613251, 0.44748857617378235, 0.4817351698875427, 0.4292237460613251, 0.44863012433052063, 0.4954337775707245, 0.4452054798603058, 0.5251141786575317, NaN, 0.4942922294139862, 0.4931506812572479, 0.509132444858551, 0.5136986374855042, 0.5799086689949036, 0.4931506812572479, 0.5593607425689697, 0.6666666865348816, 0.4794520437717438, 0.6917808055877686, 0.543379008769989, 0.5365296602249146, 0.5582191944122314, 0.6438356041908264, 0.6152967810630798, 0.6415525078773499, 0.621004581451416, 0.7226027250289917, 0.7168949842453003, 0.7180365324020386, 0.7397260069847107, 0.6963470578193665, 0.7420091032981873, 0.7283105254173279, 0.7694063782691956, 0.7031963467597961, 0.7808219194412231, 0.7557077407836914, 0.7831050157546997, 0.7579908967018127, 0.7945205569267273, 0.7751141786575317, 0.8196346759796143, 0.8698630332946777, 0.7694063782691956, 0.8515982031822205, 0.8561643958091736, 0.8835616707801819, 0.8984017968177795, 0.9771689772605896], [0.10273972898721695, 0.15867580473423004, 0.1438356190919876, 0.15981735289096832, 0.21004566550254822, 0.1940639317035675, 0.2226027399301529, 0.24200913310050964, 0.26255708932876587, 0.26484018564224243, 0.25228309631347656, 0.3264840245246887, 0.33561643958091736, 0.2945205569267273, 0.361872136592865, 0.3995433747768402, 0.38698631525039673, 0.37557077407836914, 0.38584473729133606, 0.3675799071788788, 0.43036529421806335, 0.36986300349235535, 0.43607306480407715, 0.43607306480407715, 0.465753436088562, 0.4212328791618347, 0.456620991230011, 0.41894978284835815, 0.44178083539009094, 0.4817351698875427, 0.5, 0.4383561611175537, 0.5057077407836914, NaN, 0.47031962871551514, 0.5273972749710083, 0.5205479264259338, 0.4954337775707245, 0.5251141786575317, 0.5696346759796143, 0.5570776462554932, 0.5, 0.5662100315093994, 0.5159817337989807, 0.5182648301124573, 0.7054794430732727, 0.6164383292198181, 0.5936073064804077, 0.6232876777648926, 0.5844748616218567, 0.6598173379898071, 0.6598173379898071, 0.6415525078773499, 0.6803653240203857, 0.7305936217308044, 0.6860730648040771, 0.7009132504463196, 0.7534246444702148, 0.8047945499420166, 0.7568492889404297, 0.7876712083816528, 0.77625572681427, 0.9360730648040771, 0.767123281955719, 0.8082191944122314, 0.8082191944122314, 0.8584474921226501, 0.8561643958091736, 0.8378995656967163, 0.853881299495697, 0.9383561611175537, 0.8767123222351074, 0.9748858213424683], [0.04566210135817528, 0.11187214404344559, 0.1004566177725792, 0.12100456655025482, 0.14611871540546417, 0.17351597547531128, 0.1780821979045868, 0.19178082048892975, 0.2716895043849945, 0.2168949842453003, 0.2716895043849945, 0.310502290725708, 0.31849315762519836, 0.34018266201019287, 0.3173516094684601, 0.39269405603408813, 0.3721461296081543, 0.35388126969337463, 0.3652968108654022, 0.3675799071788788, 0.4429223835468292, 0.38584473729133606, 0.4063926935195923, 0.41324201226234436, 0.42009133100509644, 0.43150684237480164, 0.4292237460613251, 0.4885844886302948, 0.47488585114479065, 0.45319634675979614, 0.4794520437717438, 0.4429223835468292, 0.5068492889404297, 0.5296803712844849, NaN, 0.5182648301124573, 0.49771690368652344, 0.4885844886302948, 0.5296803712844849, 0.5593607425689697, 0.5570776462554932, 0.5205479264259338, 0.5730593800544739, 0.5410959124565125, 0.5559360980987549, 0.6449771523475647, 0.6187214851379395, 0.7477169036865234, 0.6267123222351074, 0.5799086689949036, 0.689497709274292, 0.6872146129608154, 0.7077625393867493, 0.7431507110595703, 0.7009132504463196, 0.7031963467597961, 0.7214611768722534, 0.7294520735740662, 0.8652967810630798, 0.7511415481567383, 0.7922374606132507, 0.7602739930152893, 0.8025113940238953, 0.7808219194412231, 0.8264840245246887, 0.8127853870391846, 0.8264840245246887, 0.8242009282112122, 0.8036529421806335, 0.8173515796661377, 0.9063926935195923, 0.8744292259216309, 0.9726027250289917], [0.10273972898721695, 0.1495433747768402, 0.16438356041908264, 0.1495433747768402, 0.15981735289096832, 0.1940639317035675, 0.2054794579744339, 0.23059360682964325, 0.27397260069847107, 0.29680365324020386, 0.32420089840888977, 0.3059360682964325, 0.33675798773765564, 0.344748854637146, 0.35388126969337463, 0.38584473729133606, 0.3607305884361267, 0.3493150770664215, 0.388127863407135, 0.4018264710903168, 0.41324201226234436, 0.39726027846336365, 0.4543378949165344, 0.45890411734580994, 0.4726027250289917, 0.3767123222351074, 0.422374427318573, 0.4726027250289917, 0.4452054798603058, 0.4726027250289917, 0.5182648301124573, 0.45890411734580994, 0.49086758494377136, 0.4726027250289917, 0.4817351698875427, NaN, 0.5159817337989807, 0.4771689474582672, 0.5456621050834656, 0.5068492889404297, 0.5319634675979614, 0.4452054798603058, 0.5547945499420166, 0.5593607425689697, 0.5148401856422424, 0.5547945499420166, 0.6187214851379395, 0.5730593800544739, 0.6278538703918457, 0.6438356041908264, 0.6506849527359009, 0.6689497828483582, 0.6164383292198181, 0.7100456357002258, 0.7146118879318237, 0.6689497828483582, 0.7397260069847107, 0.7214611768722534, 0.7922374606132507, 0.7260273694992065, 0.7488584518432617, 0.7454338073730469, 0.7625570893287659, 0.7602739930152893, 0.7808219194412231, 0.8173515796661377, 0.8515982031822205, 0.7831050157546997, 0.8321917653083801, 0.8287671208381653, 0.853881299495697, 0.8949771523475647, 0.9817351698875427], [0.054794520139694214, 0.1278538852930069, 0.10730593651533127, 0.17123287916183472, 0.16894976794719696, 0.2009132355451584, 0.20662100613117218, 0.232876718044281, 0.30136987566947937, 0.2168949842453003, 0.22602739930152893, 0.2442922443151474, 0.34703195095062256, 0.35159817337989807, 0.29680365324020386, 0.21461187303066254, 0.344748854637146, 0.344748854637146, 0.4109589159488678, 0.3504566252231598, 0.31506848335266113, 0.3652968108654022, 0.38698631525039673, 0.39155250787734985, 0.38127854466438293, 0.4326483905315399, 0.45890411734580994, 0.4680365324020386, 0.48515981435775757, 0.3652968108654022, 0.39840182662010193, 0.4931506812572479, 0.48630136251449585, 0.4794520437717438, 0.5022830963134766, 0.4840182662010193, NaN, 0.5388127565383911, 0.465753436088562, 0.5273972749710083, 0.5456621050834656, 0.5114155411720276, 0.5502282977104187, 0.5045661926269531, 0.5547945499420166, 0.5753424763679504, 0.6187214851379395, 0.5890411138534546, 0.6244292259216309, 0.568493127822876, 0.7089040875434875, 0.7260273694992065, 0.7968036532402039, 0.7191780805587769, 0.6347032189369202, 0.8401826620101929, 0.6940639019012451, 0.72374427318573, 0.7191780805587769, 0.8515982031822205, 0.7054794430732727, 0.7488584518432617, 0.7077625393867493, 0.853881299495697, 0.7648401856422424, 0.7853881120681763, 0.8413242101669312, 0.8173515796661377, 0.8721461296081543, 0.8995434045791626, 0.8721461296081543, 0.878995418548584, 0.9748858213424683], [0.06392694264650345, 0.10958904027938843, 0.12100456655025482, 0.1392694115638733, 0.15410958230495453, 0.19178082048892975, 0.2009132355451584, 0.21461187303066254, 0.30136987566947937, 0.2557077705860138, 0.25799086689949036, 0.17009133100509644, 0.37100455164909363, 0.3607305884361267, 0.3493150770664215, 0.3047945201396942, 0.31963470578193665, 0.2899543344974518, 0.422374427318573, 0.3607305884361267, 0.3493150770664215, 0.4178082048892975, 0.38127854466438293, 0.3949771821498871, 0.405251145362854, 0.456620991230011, 0.4337899684906006, 0.4429223835468292, 0.4497717022895813, 0.4178082048892975, 0.43150684237480164, 0.4771689474582672, 0.42009133100509644, 0.5045661926269531, 0.5114155411720276, 0.5228310227394104, 0.4611872136592865, NaN, 0.46347030997276306, 0.5251141786575317, 0.4954337775707245, 0.5022830963134766, 0.49200913310050964, 0.5228310227394104, 0.5376712083816528, 0.5936073064804077, 0.6506849527359009, 0.5913242101669312, 0.6575342416763306, 0.6004565954208374, 0.6803653240203857, 0.6940639019012451, 0.6255707740783691, 0.6826484203338623, 0.6392694115638733, 0.664383590221405, 0.7420091032981873, 0.7168949842453003, 0.7420091032981873, 0.7785388231277466, 0.7363013625144958, 0.7317351698875427, 0.7283105254173279, 0.77625572681427, 0.7465753555297852, 0.7831050157546997, 0.8242009282112122, 0.7557077407836914, 0.835616409778595, 0.8515982031822205, 0.8630136847496033, 0.8904109597206116, 0.9680365324020386], [0.07077625393867493, 0.12100456655025482, 0.10273972898721695, 0.15753424167633057, 0.13698630034923553, 0.18493150174617767, 0.1940639317035675, 0.20776255428791046, 0.2785388231277466, 0.2214611917734146, 0.25913241505622864, 0.232876718044281, 0.34246575832366943, 0.3333333432674408, 0.2876712381839752, 0.3378995358943939, 0.344748854637146, 0.3173516094684601, 0.3949771821498871, 0.39726027846336365, 0.4155251085758209, 0.40867578983306885, 0.37442922592163086, 0.3904109597206116, 0.40981733798980713, 0.42579910159111023, 0.42465752363204956, 0.47031962871551514, 0.4611872136592865, 0.4817351698875427, 0.38127854466438293, 0.4840182662010193, 0.5068492889404297, 0.47488585114479065, 0.47031962871551514, 0.4543378949165344, 0.534246563911438, 0.5365296602249146, NaN, 0.534246563911438, 0.5468036532402039, 0.5045661926269531, 0.5525113940238953, 0.611872136592865, 0.5376712083816528, 0.5376712083816528, 0.6415525078773499, 0.5753424763679504, 0.6392694115638733, 0.5799086689949036, 0.6415525078773499, 0.6621004343032837, 0.7317351698875427, 0.6917808055877686, 0.6369863152503967, 0.7420091032981873, 0.7031963467597961, 0.689497709274292, 0.7191780805587769, 0.7694063782691956, 0.7043378949165344, 0.6963470578193665, 0.6849315166473389, 0.7716894745826721, 0.7317351698875427, 0.7465753555297852, 0.7831050157546997, 0.7716894745826721, 0.8264840245246887, 0.8150684833526611, 0.8527397513389587, 0.8333333134651184, 0.9589040875434875], [0.08675798773765564, 0.11872146278619766, 0.14611871540546417, 0.13013698160648346, 0.1438356190919876, 0.17351597547531128, 0.18949772417545319, 0.20319634675979614, 0.24885845184326172, 0.22374428808689117, 0.17465753853321075, 0.23173515498638153, 0.3378995358943939, 0.34246575832366943, 0.3378995358943939, 0.3378995358943939, 0.344748854637146, 0.28538814187049866, 0.37442922592163086, 0.3059360682964325, 0.3595890402793884, 0.33561643958091736, 0.37899544835090637, 0.4041095972061157, 0.4063926935195923, 0.36986300349235535, 0.46461185812950134, 0.40867578983306885, 0.42009133100509644, 0.43607306480407715, 0.43607306480407715, 0.40867578983306885, 0.4406392574310303, 0.43036529421806335, 0.4406392574310303, 0.4931506812572479, 0.4726027250289917, 0.47488585114479065, 0.465753436088562, NaN, 0.5205479264259338, 0.4337899684906006, 0.5068492889404297, 0.4931506812572479, 0.5022830963134766, 0.577625572681427, 0.6278538703918457, 0.5936073064804077, 0.646118700504303, 0.586758017539978, 0.6575342416763306, 0.6757990717887878, 0.6598173379898071, 0.6917808055877686, 0.7180365324020386, 0.689497709274292, 0.7031963467597961, 0.7054794430732727, 0.7751141786575317, 0.7305936217308044, 0.7557077407836914, 0.7420091032981873, 0.8150684833526611, 0.7579908967018127, 0.7922374606132507, 0.7922374606132507, 0.844748854637146, 0.7831050157546997, 0.8310502171516418, 0.8333333134651184, 0.9086757898330688, 0.8721461296081543, 0.9748858213424683], [0.08219178020954132, 0.13698630034923553, 0.15296803414821625, 0.155251145362854, 0.1609589010477066, 0.16438356041908264, 0.19063927233219147, 0.21232876181602478, 0.27968037128448486, 0.25799086689949036, 0.2442922443151474, 0.232876718044281, 0.3481735289096832, 0.344748854637146, 0.31506848335266113, 0.30365297198295593, 0.3219178020954132, 0.31963470578193665, 0.3721461296081543, 0.34018266201019287, 0.35159817337989807, 0.3675799071788788, 0.3767123222351074, 0.40753424167633057, 0.4063926935195923, 0.40867578983306885, 0.4440639317035675, 0.3835616409778595, 0.39269405603408813, 0.42465752363204956, 0.40867578983306885, 0.4885844886302948, 0.3333333432674408, 0.4429223835468292, 0.4429223835468292, 0.4680365324020386, 0.4543378949165344, 0.5045661926269531, 0.45319634675979614, 0.4794520437717438, NaN, 0.44863012433052063, 0.6152967810630798, 0.4954337775707245, 0.4954337775707245, 0.5376712083816528, 0.6232876777648926, 0.5616438388824463, 0.628995418548584, 0.5901826620101929, 0.6621004343032837, 0.7009132504463196, 0.6780821681022644, 0.6849315166473389, 0.6678082346916199, 0.7077625393867493, 0.7077625393867493, 0.7374429106712341, 0.6826484203338623, 0.7420091032981873, 0.732876718044281, 0.7557077407836914, 0.7146118879318237, 0.7694063782691956, 0.7511415481567383, 0.7853881120681763, 0.8515982031822205, 0.7534246444702148, 0.8424657583236694, 0.8401826620101929, 0.8675799369812012, 0.8995434045791626, 0.9771689772605896], [0.05936073139309883, 0.0890410989522934, 0.06621004641056061, 0.13013698160648346, 0.11872146278619766, 0.2168949842453003, 0.2009132355451584, 0.2111872136592865, 0.31963470578193665, 0.27625569701194763, 0.31506848335266113, 0.29223743081092834, 0.31506848335266113, 0.34018266201019287, 0.30821916460990906, 0.37899544835090637, 0.4041095972061157, 0.34703195095062256, 0.3732876777648926, 0.4178082048892975, 0.4155251085758209, 0.3949771821498871, 0.41324201226234436, 0.4178082048892975, 0.4155251085758209, 0.44748857617378235, 0.45205479860305786, 0.4269406497478485, 0.49086758494377136, 0.4771689474582672, 0.4543378949165344, 0.4680365324020386, 0.5205479264259338, 0.5, 0.4794520437717438, 0.5547945499420166, 0.4885844886302948, 0.49771690368652344, 0.4954337775707245, 0.5662100315093994, 0.551369845867157, NaN, 0.5388127565383911, 0.49771690368652344, 0.5205479264259338, 0.5593607425689697, 0.543379008769989, 0.5707762837409973, 0.5479452013969421, 0.5, 0.5890411138534546, 0.6095890402793884, 0.6689497828483582, 0.6278538703918457, 0.6004565954208374, 0.6872146129608154, 0.6141552329063416, 0.5981734991073608, 0.655251145362854, 0.6575342416763306, 0.6689497828483582, 0.6027397513389587, 0.6940639019012451, 0.6598173379898071, 0.6735159754753113, 0.655251145362854, 0.72374427318573, 0.8139269351959229, 0.7648401856422424, 0.767123281955719, 0.7739726305007935, 0.7579908967018127, 0.931506872177124], [0.0890410989522934, 0.13584475219249725, 0.1438356190919876, 0.1438356190919876, 0.17579908668994904, 0.18036529421806335, 0.18721461296081543, 0.2054794579744339, 0.26255708932876587, 0.23515981435775757, 0.24315068125724792, 0.24315068125724792, 0.35388126969337463, 0.344748854637146, 0.31963470578193665, 0.24657534062862396, 0.26940637826919556, 0.30365297198295593, 0.3721461296081543, 0.3721461296081543, 0.31278538703918457, 0.3607305884361267, 0.344748854637146, 0.37899544835090637, 0.3652968108654022, 0.39726027846336365, 0.4063926935195923, 0.3904109597206116, 0.39726027846336365, 0.4063926935195923, 0.40867578983306885, 0.45319634675979614, 0.30821916460990906, 0.4337899684906006, 0.4269406497478485, 0.4452054798603058, 0.4497717022895813, 0.5079908967018127, 0.44748857617378235, 0.4931506812572479, 0.3847031891345978, 0.4611872136592865, NaN, 0.45890411734580994, 0.4680365324020386, 0.5365296602249146, 0.6095890402793884, 0.5639269351959229, 0.6050228476524353, 0.5936073064804077, 0.6666666865348816, 0.6598173379898071, 0.6369863152503967, 0.7031963467597961, 0.6780821681022644, 0.664383590221405, 0.7054794430732727, 0.7442922592163086, 0.7077625393867493, 0.7511415481567383, 0.7146118879318237, 0.7659817337989807, 0.7214611768722534, 0.7716894745826721, 0.7283105254173279, 0.8173515796661377, 0.8630136847496033, 0.732876718044281, 0.8310502171516418, 0.8219178318977356, 0.8515982031822205, 0.9086757898330688, 0.9748858213424683], [0.06164383515715599, 0.13013698160648346, 0.09817351400852203, 0.16210044920444489, 0.14155250787734985, 0.1666666716337204, 0.17579908668994904, 0.1940639317035675, 0.26484018564224243, 0.2054794579744339, 0.25799086689949036, 0.21461187303066254, 0.310502290725708, 0.29680365324020386, 0.28082191944122314, 0.25342464447021484, 0.310502290725708, 0.284246563911438, 0.3767123222351074, 0.3493150770664215, 0.310502290725708, 0.36415526270866394, 0.3287671208381653, 0.3493150770664215, 0.3652968108654022, 0.3904109597206116, 0.422374427318573, 0.4611872136592865, 0.46347030997276306, 0.439497709274292, 0.28082191944122314, 0.4680365324020386, 0.456620991230011, 0.4840182662010193, 0.45890411734580994, 0.4406392574310303, 0.4954337775707245, 0.4771689474582672, 0.388127863407135, 0.5068492889404297, 0.5045661926269531, 0.5022830963134766, 0.5410959124565125, NaN, 0.534246563911438, 0.5296803712844849, 0.6347032189369202, 0.5547945499420166, 0.6392694115638733, 0.5753424763679504, 0.6506849527359009, 0.6529680490493774, 0.7134703397750854, 0.698630154132843, 0.6324200630187988, 0.7271689772605896, 0.6940639019012451, 0.7031963467597961, 0.6917808055877686, 0.7534246444702148, 0.6803653240203857, 0.7191780805587769, 0.6735159754753113, 0.7602739930152893, 0.6963470578193665, 0.7625570893287659, 0.7773972749710083, 0.7739726305007935, 0.7990867495536804, 0.8082191944122314, 0.810502290725708, 0.8515982031822205, 0.965753436088562], [0.07305935770273209, 0.09474886208772659, 0.10730593651533127, 0.14155250787734985, 0.14155250787734985, 0.15410958230495453, 0.18036529421806335, 0.19634702801704407, 0.24086758494377136, 0.2054794579744339, 0.24200913310050964, 0.2557077705860138, 0.27625569701194763, 0.2716895043849945, 0.2990867495536804, 0.3059360682964325, 0.3561643958091736, 0.3287671208381653, 0.344748854637146, 0.3561643958091736, 0.3675799071788788, 0.2888127863407135, 0.33561643958091736, 0.34018266201019287, 0.3767123222351074, 0.388127863407135, 0.42009133100509644, 0.4440639317035675, 0.42465752363204956, 0.4155251085758209, 0.4063926935195923, 0.46917808055877686, 0.46347030997276306, 0.4817351698875427, 0.4440639317035675, 0.48515981435775757, 0.4452054798603058, 0.4623287618160248, 0.4623287618160248, 0.49771690368652344, 0.5045661926269531, 0.4794520437717438, 0.5319634675979614, 0.465753436088562, NaN, 0.5525113940238953, 0.6152967810630798, 0.5502282977104187, 0.6073059439659119, 0.5045661926269531, 0.6141552329063416, 0.6347032189369202, 0.6095890402793884, 0.663241982460022, 0.6244292259216309, 0.6506849527359009, 0.6780821681022644, 0.6335616707801819, 0.7557077407836914, 0.7100456357002258, 0.6826484203338623, 0.6689497828483582, 0.7534246444702148, 0.72374427318573, 0.7305936217308044, 0.7283105254173279, 0.7831050157546997, 0.8059360980987549, 0.7694063782691956, 0.7922374606132507, 0.8333333134651184, 0.8299086689949036, 0.9589040875434875], [0.07305935770273209, 0.12671232223510742, 0.13470318913459778, 0.11872146278619766, 0.15753424167633057, 0.14611871540546417, 0.15296803414821625, 0.16894976794719696, 0.2054794579744339, 0.2214611917734146, 0.2271689474582672, 0.26484018564224243, 0.26484018564224243, 0.2557077705860138, 0.2888127863407135, 0.31278538703918457, 0.29680365324020386, 0.29680365324020386, 0.3059360682964325, 0.2899543344974518, 0.37899544835090637, 0.3116438388824463, 0.35159817337989807, 0.36986300349235535, 0.36872145533561707, 0.3264840245246887, 0.36301368474960327, 0.34246575832366943, 0.3333333432674408, 0.42465752363204956, 0.439497709274292, 0.3607305884361267, 0.44178083539009094, 0.2945205569267273, 0.3550228178501129, 0.4452054798603058, 0.42465752363204956, 0.4063926935195923, 0.4623287618160248, 0.422374427318573, 0.4623287618160248, 0.4406392574310303, 0.46347030997276306, 0.47031962871551514, 0.44748857617378235, NaN, 0.5525113940238953, 0.5159817337989807, 0.5456621050834656, 0.5068492889404297, 0.5936073064804077, 0.5958904027938843, 0.5799086689949036, 0.5890411138534546, 0.6666666865348816, 0.6027397513389587, 0.6598173379898071, 0.6689497828483582, 0.6906392574310303, 0.6609588861465454, 0.7054794430732727, 0.6974886059761047, 0.7853881120681763, 0.6780821681022644, 0.7442922592163086, 0.7488584518432617, 0.7831050157546997, 0.7488584518432617, 0.767123281955719, 0.7876712083816528, 0.8607305884361267, 0.8401826620101929, 0.965753436088562], [0.18036529421806335, 0.21575342118740082, 0.25342464447021484, 0.20776255428791046, 0.2557077705860138, 0.2214611917734146, 0.25342464447021484, 0.26255708932876587, 0.29223743081092834, 0.29680365324020386, 0.3059360682964325, 0.2945205569267273, 0.344748854637146, 0.29223743081092834, 0.361872136592865, 0.31849315762519836, 0.27397260069847107, 0.31278538703918457, 0.39611873030662537, 0.3287671208381653, 0.33561643958091736, 0.3652968108654022, 0.327625572681427, 0.34018266201019287, 0.3378995358943939, 0.36301368474960327, 0.4109589159488678, 0.4006849229335785, 0.3767123222351074, 0.3652968108654022, 0.34246575832366943, 0.43607306480407715, 0.3561643958091736, 0.3835616409778595, 0.38127854466438293, 0.38127854466438293, 0.38127854466438293, 0.3493150770664215, 0.35844749212265015, 0.3721461296081543, 0.3767123222351074, 0.456620991230011, 0.3904109597206116, 0.3652968108654022, 0.3847031891345978, 0.44748857617378235, NaN, 0.4041095972061157, 0.6095890402793884, 0.5296803712844849, 0.568493127822876, 0.5673515796661377, 0.4817351698875427, 0.4155251085758209, 0.6164383292198181, 0.4885844886302948, 0.9292237162590027, 0.611872136592865, 0.5159817337989807, 0.6050228476524353, 0.6369863152503967, 0.6347032189369202, 0.6415525078773499, 0.6347032189369202, 0.6438356041908264, 0.6917808055877686, 0.7557077407836914, 0.7100456357002258, 0.7831050157546997, 0.7785388231277466, 0.7031963467597961, 0.8515982031822205, 0.931506872177124], [0.031963471323251724, 0.06621004641056061, 0.07990867644548416, 0.08219178020954132, 0.08675798773765564, 0.10273972898721695, 0.12100456655025482, 0.12328767031431198, 0.2054794579744339, 0.17123287916183472, 0.21232876181602478, 0.21347032487392426, 0.26255708932876587, 0.28082191944122314, 0.2602739632129669, 0.27397260069847107, 0.2876712381839752, 0.2659817337989807, 0.30365297198295593, 0.27625569701194763, 0.3344748914241791, 0.2945205569267273, 0.2990867495536804, 0.30821916460990906, 0.31849315762519836, 0.30365297198295593, 0.35844749212265015, 0.3767123222351074, 0.3675799071788788, 0.38127854466438293, 0.3835616409778595, 0.3675799071788788, 0.3847031891345978, 0.4063926935195923, 0.25228309631347656, 0.4269406497478485, 0.4109589159488678, 0.40867578983306885, 0.42465752363204956, 0.4063926935195923, 0.4383561611175537, 0.4292237460613251, 0.43607306480407715, 0.4452054798603058, 0.4497717022895813, 0.4840182662010193, 0.5958904027938843, NaN, 0.6004565954208374, 0.5228310227394104, 0.6130136847496033, 0.6073059439659119, 0.5662100315093994, 0.6506849527359009, 0.6803653240203857, 0.594748854637146, 0.689497709274292, 0.6598173379898071, 0.7682648301124573, 0.6666666865348816, 0.6963470578193665, 0.6746575236320496, 0.7031963467597961, 0.6917808055877686, 0.7305936217308044, 0.7625570893287659, 0.7728310227394104, 0.7602739930152893, 0.7534246444702148, 0.7648401856422424, 0.8470319509506226, 0.810502290725708, 0.9611872434616089], [0.1780821979045868, 0.20319634675979614, 0.2511415481567383, 0.21004566550254822, 0.2442922443151474, 0.21803653240203857, 0.24657534062862396, 0.2511415481567383, 0.2831050157546997, 0.2876712381839752, 0.2945205569267273, 0.2899543344974518, 0.34246575832366943, 0.28082191944122314, 0.36301368474960327, 0.31506848335266113, 0.267123281955719, 0.3093607425689697, 0.3949771821498871, 0.3173516094684601, 0.33105021715164185, 0.361872136592865, 0.3219178020954132, 0.344748854637146, 0.3333333432674408, 0.3561643958091736, 0.4063926935195923, 0.4018264710903168, 0.3721461296081543, 0.36301368474960327, 0.344748854637146, 0.43607306480407715, 0.35844749212265015, 0.3767123222351074, 0.3732876777648926, 0.3721461296081543, 0.37557077407836914, 0.34246575832366943, 0.3607305884361267, 0.35388126969337463, 0.37100455164909363, 0.45205479860305786, 0.3949771821498871, 0.3607305884361267, 0.39269405603408813, 0.4543378949165344, 0.3904109597206116, 0.3995433747768402, NaN, 0.5205479264259338, 0.5479452013969421, 0.5502282977104187, 0.47031962871551514, 0.4178082048892975, 0.6073059439659119, 0.48630136251449585, 0.9474886059761047, 0.6073059439659119, 0.5136986374855042, 0.5958904027938843, 0.6347032189369202, 0.6347032189369202, 0.6392694115638733, 0.6255707740783691, 0.6392694115638733, 0.689497709274292, 0.7442922592163086, 0.7031963467597961, 0.77625572681427, 0.7739726305007935, 0.7009132504463196, 0.8401826620101929, 0.931506872177124], [0.05707762390375137, 0.09817351400852203, 0.09474886208772659, 0.07534246891736984, 0.14840182662010193, 0.08675798773765564, 0.12328767031431198, 0.1278538852930069, 0.15981735289096832, 0.30251142382621765, 0.2899543344974518, 0.3059360682964325, 0.2168949842453003, 0.155251145362854, 0.2728310525417328, 0.32420089840888977, 0.3059360682964325, 0.3333333432674408, 0.2214611917734146, 0.3664383590221405, 0.34246575832366943, 0.3732876777648926, 0.3995433747768402, 0.4041095972061157, 0.40867578983306885, 0.37899544835090637, 0.28538814187049866, 0.35388126969337463, 0.37899544835090637, 0.405251145362854, 0.40867578983306885, 0.36986300349235535, 0.37899544835090637, 0.4155251085758209, 0.42009133100509644, 0.3561643958091736, 0.43150684237480164, 0.3995433747768402, 0.42009133100509644, 0.41324201226234436, 0.40981733798980713, 0.5, 0.4063926935195923, 0.42465752363204956, 0.4954337775707245, 0.4931506812572479, 0.47031962871551514, 0.4771689474582672, 0.4794520437717438, NaN, 0.49771690368652344, 0.5022830963134766, 0.5205479264259338, 0.5296803712844849, 0.6187214851379395, 0.5639269351959229, 0.6073059439659119, 0.6278538703918457, 0.5821917653083801, 0.5547945499420166, 0.6164383292198181, 0.6563926935195923, 0.655251145362854, 0.5890411138534546, 0.6324200630187988, 0.6917808055877686, 0.6621004343032837, 0.7420091032981873, 0.7260273694992065, 0.698630154132843, 0.7922374606132507, 0.7146118879318237, 0.9611872434616089], [0.04794520512223244, 0.06164383515715599, 0.07534246891736984, 0.08789954334497452, 0.08675798773765564, 0.10730593651533127, 0.1164383590221405, 0.13127854466438293, 0.15410958230495453, 0.2009132355451584, 0.16210044920444489, 0.1826484054327011, 0.2054794579744339, 0.19863013923168182, 0.24200913310050964, 0.19292236864566803, 0.20776255428791046, 0.24885845184326172, 0.2602739632129669, 0.25799086689949036, 0.24315068125724792, 0.2705479562282562, 0.2283104956150055, 0.23059360682964325, 0.25342464447021484, 0.31506848335266113, 0.27397260069847107, 0.28538814187049866, 0.28082191944122314, 0.28538814187049866, 0.3264840245246887, 0.344748854637146, 0.2773972749710083, 0.34018266201019287, 0.310502290725708, 0.3493150770664215, 0.29109588265419006, 0.31963470578193665, 0.35844749212265015, 0.34246575832366943, 0.3378995358943939, 0.4109589159488678, 0.3333333432674408, 0.3493150770664215, 0.38584473729133606, 0.4063926935195923, 0.43150684237480164, 0.38698631525039673, 0.45205479860305786, 0.5022830963134766, NaN, 0.6324200630187988, 0.4406392574310303, 0.4931506812572479, 0.5707762837409973, 0.4885844886302948, 0.6221461296081543, 0.5993150472640991, 0.5593607425689697, 0.6586757898330688, 0.577625572681427, 0.628995418548584, 0.6004565954208374, 0.7009132504463196, 0.5981734991073608, 0.7031963467597961, 0.7853881120681763, 0.7077625393867493, 0.8721461296081543, 0.8378995656967163, 0.7260273694992065, 0.8424657583236694, 0.9885844588279724], [0.05707762390375137, 0.06621004641056061, 0.09132420271635056, 0.0890410989522934, 0.09474886208772659, 0.10273972898721695, 0.12328767031431198, 0.1278538852930069, 0.12100456655025482, 0.19178082048892975, 0.172374427318573, 0.2009132355451584, 0.19863013923168182, 0.20319634675979614, 0.2442922443151474, 0.21004566550254822, 0.20776255428791046, 0.24200913310050964, 0.2397260218858719, 0.24657534062862396, 0.2340182662010193, 0.25342464447021484, 0.23515981435775757, 0.23630136251449585, 0.26255708932876587, 0.3059360682964325, 0.267123281955719, 0.26484018564224243, 0.2511415481567383, 0.26484018564224243, 0.31506848335266113, 0.310502290725708, 0.2831050157546997, 0.34018266201019287, 0.31278538703918457, 0.33105021715164185, 0.27397260069847107, 0.3059360682964325, 0.3378995358943939, 0.32420089840888977, 0.2990867495536804, 0.3904109597206116, 0.34018266201019287, 0.34703195095062256, 0.3652968108654022, 0.4041095972061157, 0.4326483905315399, 0.39269405603408813, 0.4497717022895813, 0.49771690368652344, 0.3675799071788788, NaN, 0.4109589159488678, 0.49086758494377136, 0.5707762837409973, 0.4383561611175537, 0.6164383292198181, 0.5913242101669312, 0.543379008769989, 0.6027397513389587, 0.5456621050834656, 0.6164383292198181, 0.5730593800544739, 0.646118700504303, 0.5844748616218567, 0.6963470578193665, 0.7374429106712341, 0.6803653240203857, 0.8059360980987549, 0.8002282977104187, 0.7054794430732727, 0.8493150472640991, 0.9863013625144958], [0.029680365696549416, 0.05936073139309883, 0.06621004641056061, 0.08675798773765564, 0.06392694264650345, 0.11187214404344559, 0.11187214404344559, 0.11872146278619766, 0.1826484054327011, 0.10273972898721695, 0.1278538852930069, 0.14611871540546417, 0.21917808055877686, 0.2397260218858719, 0.22374428808689117, 0.13013698160648346, 0.23515981435775757, 0.19178082048892975, 0.2876712381839752, 0.22374428808689117, 0.18150684237480164, 0.18949772417545319, 0.1826484054327011, 0.1780821979045868, 0.1666666716337204, 0.2899543344974518, 0.3287671208381653, 0.2990867495536804, 0.36301368474960327, 0.13470318913459778, 0.232876718044281, 0.37442922592163086, 0.2819634675979614, 0.35844749212265015, 0.29223743081092834, 0.3835616409778595, 0.20319634675979614, 0.37442922592163086, 0.2682648301124573, 0.34018266201019287, 0.3219178020954132, 0.33105021715164185, 0.36301368474960327, 0.28652969002723694, 0.3904109597206116, 0.42009133100509644, 0.5182648301124573, 0.4337899684906006, 0.5296803712844849, 0.4794520437717438, 0.5593607425689697, 0.5890411138534546, NaN, 0.5730593800544739, 0.5456621050834656, 0.5011415481567383, 0.6187214851379395, 0.6004565954208374, 0.5981734991073608, 0.6278538703918457, 0.5182648301124573, 0.6141552329063416, 0.586758017539978, 0.6575342416763306, 0.5753424763679504, 0.6757990717887878, 0.7168949842453003, 0.7191780805587769, 0.7557077407836914, 0.7716894745826721, 0.7077625393867493, 0.7853881120681763, 0.9383561611175537], [0.05936073139309883, 0.07077625393867493, 0.077625572681427, 0.09132420271635056, 0.1004566177725792, 0.10730593651533127, 0.11872146278619766, 0.13013698160648346, 0.2009132355451584, 0.155251145362854, 0.1438356190919876, 0.15068493783473969, 0.19863013923168182, 0.2214611917734146, 0.23059360682964325, 0.15068493783473969, 0.18721461296081543, 0.18493150174617767, 0.26484018564224243, 0.20776255428791046, 0.17922374606132507, 0.21004566550254822, 0.16894976794719696, 0.21347032487392426, 0.23059360682964325, 0.25342464447021484, 0.30365297198295593, 0.2899543344974518, 0.27397260069847107, 0.26940637826919556, 0.26255708932876587, 0.34703195095062256, 0.2602739632129669, 0.31963470578193665, 0.2568493187427521, 0.2899543344974518, 0.28082191944122314, 0.3173516094684601, 0.30821916460990906, 0.30821916460990906, 0.31506848335266113, 0.3721461296081543, 0.29680365324020386, 0.30136987566947937, 0.33675798773765564, 0.4109589159488678, 0.5844748616218567, 0.3493150770664215, 0.5821917653083801, 0.47031962871551514, 0.5068492889404297, 0.509132444858551, 0.4269406497478485, NaN, 0.543379008769989, 0.46347030997276306, 0.6529680490493774, 0.5981734991073608, 0.5479452013969421, 0.6141552329063416, 0.5182648301124573, 0.6301369667053223, 0.5844748616218567, 0.6347032189369202, 0.5319634675979614, 0.6940639019012451, 0.7579908967018127, 0.6621004343032837, 0.698630154132843, 0.7305936217308044, 0.6415525078773499, 0.8219178318977356, 0.9360730648040771], [0.07305935770273209, 0.09589041024446487, 0.11872146278619766, 0.10273972898721695, 0.12100456655025482, 0.09703196585178375, 0.12557077407836914, 0.1392694115638733, 0.1278538852930069, 0.21232876181602478, 0.23515981435775757, 0.232876718044281, 0.20319634675979614, 0.11187214404344559, 0.28538814187049866, 0.29223743081092834, 0.2385844737291336, 0.24885845184326172, 0.2214611917734146, 0.25913241505622864, 0.30821916460990906, 0.26484018564224243, 0.34246575832366943, 0.35844749212265015, 0.3675799071788788, 0.26940637826919556, 0.21232876181602478, 0.2511415481567383, 0.23515981435775757, 0.35273972153663635, 0.33561643958091736, 0.25342464447021484, 0.30365297198295593, 0.26940637826919556, 0.2990867495536804, 0.28538814187049866, 0.3652968108654022, 0.3607305884361267, 0.36301368474960327, 0.2819634675979614, 0.3321917951107025, 0.3995433747768402, 0.3219178020954132, 0.3675799071788788, 0.37557077407836914, 0.3333333432674408, 0.3835616409778595, 0.31963470578193665, 0.39269405603408813, 0.38127854466438293, 0.4292237460613251, 0.4292237460613251, 0.4543378949165344, 0.456620991230011, NaN, 0.47031962871551514, 0.5296803712844849, 0.5251141786575317, 0.4783104956150055, 0.5022830963134766, 0.5639269351959229, 0.5650684833526611, 0.5456621050834656, 0.5136986374855042, 0.6027397513389587, 0.6141552329063416, 0.611872136592865, 0.6324200630187988, 0.6232876777648926, 0.6529680490493774, 0.698630154132843, 0.6940639019012451, 0.8926940560340881], [0.027397260069847107, 0.04566210135817528, 0.052511416375637054, 0.05936073139309883, 0.052511416375637054, 0.05936073139309883, 0.06621004641056061, 0.077625572681427, 0.1392694115638733, 0.07534246891736984, 0.09817351400852203, 0.09474886208772659, 0.16894976794719696, 0.2111872136592865, 0.20319634675979614, 0.0502283088862896, 0.18150684237480164, 0.16438356041908264, 0.24885845184326172, 0.18949772417545319, 0.08561643958091736, 0.14155250787734985, 0.14840182662010193, 0.13698630034923553, 0.15981735289096832, 0.25342464447021484, 0.3287671208381653, 0.26255708932876587, 0.2990867495536804, 0.10730593651533127, 0.1997716873884201, 0.3219178020954132, 0.25799086689949036, 0.31392693519592285, 0.29680365324020386, 0.33105021715164185, 0.15981735289096832, 0.33561643958091736, 0.25799086689949036, 0.310502290725708, 0.29223743081092834, 0.31278538703918457, 0.33561643958091736, 0.2728310525417328, 0.3493150770664215, 0.39726027846336365, 0.5114155411720276, 0.405251145362854, 0.5136986374855042, 0.43607306480407715, 0.5114155411720276, 0.5616438388824463, 0.4988584518432617, 0.5365296602249146, 0.5296803712844849, NaN, 0.6073059439659119, 0.577625572681427, 0.586758017539978, 0.5913242101669312, 0.5182648301124573, 0.5890411138534546, 0.5707762837409973, 0.6198630332946777, 0.586758017539978, 0.6621004343032837, 0.6917808055877686, 0.7260273694992065, 0.7511415481567383, 0.7682648301124573, 0.732876718044281, 0.7945205569267273, 0.9577625393867493], [0.14611871540546417, 0.16894976794719696, 0.19863013923168182, 0.16438356041908264, 0.21004566550254822, 0.18721461296081543, 0.21004566550254822, 0.20776255428791046, 0.21004566550254822, 0.2397260218858719, 0.2511415481567383, 0.24657534062862396, 0.2397260218858719, 0.21232876181602478, 0.2956621050834656, 0.2397260218858719, 0.2111872136592865, 0.267123281955719, 0.26255708932876587, 0.26484018564224243, 0.25799086689949036, 0.28082191944122314, 0.2716895043849945, 0.2831050157546997, 0.27625569701194763, 0.2716895043849945, 0.32420089840888977, 0.310502290725708, 0.2831050157546997, 0.30136987566947937, 0.2899543344974518, 0.30136987566947937, 0.2716895043849945, 0.2990867495536804, 0.2785388231277466, 0.2602739632129669, 0.3059360682964325, 0.25799086689949036, 0.29680365324020386, 0.29680365324020386, 0.29223743081092834, 0.38584473729133606, 0.2945205569267273, 0.3059360682964325, 0.3219178020954132, 0.34018266201019287, 0.07077625393867493, 0.310502290725708, 0.052511416375637054, 0.39269405603408813, 0.3778538703918457, 0.3835616409778595, 0.38127854466438293, 0.34703195095062256, 0.47031962871551514, 0.39269405603408813, NaN, 0.4885844886302948, 0.3904109597206116, 0.42579910159111023, 0.5410959124565125, 0.5022830963134766, 0.4543378949165344, 0.4497717022895813, 0.5456621050834656, 0.5616438388824463, 0.6050228476524353, 0.6255707740783691, 0.6187214851379395, 0.5936073064804077, 0.6232876777648926, 0.7260273694992065, 0.922374427318573], [0.07534246891736984, 0.12100456655025482, 0.12557077407836914, 0.10502283275127411, 0.13242009282112122, 0.14611871540546417, 0.14611871540546417, 0.1666666716337204, 0.19178082048892975, 0.16324201226234436, 0.17351597547531128, 0.1666666716337204, 0.24885845184326172, 0.21004566550254822, 0.2283104956150055, 0.1826484054327011, 0.16894976794719696, 0.19863013923168182, 0.25913241505622864, 0.2442922443151474, 0.2168949842453003, 0.28082191944122314, 0.2442922443151474, 0.24885845184326172, 0.25342464447021484, 0.2831050157546997, 0.2716895043849945, 0.23059360682964325, 0.2054794579744339, 0.2602739632129669, 0.28538814187049866, 0.25799086689949036, 0.23059360682964325, 0.24657534062862396, 0.2705479562282562, 0.2785388231277466, 0.27625569701194763, 0.2831050157546997, 0.310502290725708, 0.2945205569267273, 0.26255708932876587, 0.4018264710903168, 0.2557077705860138, 0.29680365324020386, 0.3664383590221405, 0.33105021715164185, 0.388127863407135, 0.34018266201019287, 0.39269405603408813, 0.3721461296081543, 0.4006849229335785, 0.40867578983306885, 0.3995433747768402, 0.4018264710903168, 0.47488585114479065, 0.422374427318573, 0.5114155411720276, NaN, 0.4429223835468292, 0.43150684237480164, 0.4794520437717438, 0.6678082346916199, 0.4931506812572479, 0.45890411734580994, 0.5022830963134766, 0.7579908967018127, 0.5753424763679504, 0.5981734991073608, 0.5936073064804077, 0.5662100315093994, 0.621004581451416, 0.6780821681022644, 0.8926940560340881], [0.02283105067908764, 0.052511416375637054, 0.052511416375637054, 0.04337899386882782, 0.0502283088862896, 0.0388127863407135, 0.07534246891736984, 0.077625572681427, 0.12100456655025482, 0.11415524780750275, 0.10730593651533127, 0.10958904027938843, 0.15981735289096832, 0.1392694115638733, 0.21917808055877686, 0.18949772417545319, 0.18949772417545319, 0.12557077407836914, 0.18949772417545319, 0.155251145362854, 0.20776255428791046, 0.1826484054327011, 0.2009132355451584, 0.2168949842453003, 0.23515981435775757, 0.23515981435775757, 0.21461187303066254, 0.2009132355451584, 0.18493150174617767, 0.25342464447021484, 0.26940637826919556, 0.26255708932876587, 0.29680365324020386, 0.1952054798603058, 0.13470318913459778, 0.20776255428791046, 0.28082191944122314, 0.25799086689949036, 0.28082191944122314, 0.22488585114479065, 0.3173516094684601, 0.344748854637146, 0.29223743081092834, 0.30821916460990906, 0.2442922443151474, 0.3093607425689697, 0.4840182662010193, 0.23173515498638153, 0.48630136251449585, 0.4178082048892975, 0.4406392574310303, 0.456620991230011, 0.4018264710903168, 0.45205479860305786, 0.5216894745826721, 0.41324201226234436, 0.6095890402793884, 0.5570776462554932, NaN, 0.49086758494377136, 0.5456621050834656, 0.5799086689949036, 0.577625572681427, 0.5479452013969421, 0.5844748616218567, 0.6598173379898071, 0.698630154132843, 0.6529680490493774, 0.664383590221405, 0.664383590221405, 0.7260273694992065, 0.7785388231277466, 0.9406392574310303], [0.02283105067908764, 0.06164383515715599, 0.05707762390375137, 0.06164383515715599, 0.06392694264650345, 0.10273972898721695, 0.08675798773765564, 0.11187214404344559, 0.12328767031431198, 0.10730593651533127, 0.0890410989522934, 0.13470318913459778, 0.1438356190919876, 0.1940639317035675, 0.19634702801704407, 0.1004566177725792, 0.15753424167633057, 0.1666666716337204, 0.21461187303066254, 0.1780821979045868, 0.15981735289096832, 0.15981735289096832, 0.12328767031431198, 0.13013698160648346, 0.13013698160648346, 0.23059360682964325, 0.26255708932876587, 0.19634702801704407, 0.24885845184326172, 0.11986301094293594, 0.2009132355451584, 0.293379008769989, 0.21917808055877686, 0.24315068125724792, 0.24885845184326172, 0.27397260069847107, 0.14840182662010193, 0.2214611917734146, 0.23059360682964325, 0.26940637826919556, 0.25799086689949036, 0.34246575832366943, 0.24885845184326172, 0.24657534062862396, 0.2899543344974518, 0.3390410840511322, 0.3949771821498871, 0.3333333432674408, 0.4041095972061157, 0.4452054798603058, 0.34132421016693115, 0.39726027846336365, 0.3721461296081543, 0.38584473729133606, 0.49771690368652344, 0.40867578983306885, 0.5742009282112122, 0.568493127822876, 0.509132444858551, NaN, 0.5159817337989807, 0.5844748616218567, 0.5319634675979614, 0.6324200630187988, 0.5410959124565125, 0.6655251383781433, 0.7009132504463196, 0.6849315166473389, 0.7511415481567383, 0.8173515796661377, 0.6278538703918457, 0.8150684833526611, 0.9634703397750854], [0.01826483942568302, 0.06392694264650345, 0.06392694264650345, 0.05707762390375137, 0.06849315017461777, 0.06849315017461777, 0.06849315017461777, 0.08789954334497452, 0.11872146278619766, 0.08219178020954132, 0.08675798773765564, 0.11187214404344559, 0.14155250787734985, 0.13470318913459778, 0.1780821979045868, 0.12557077407836914, 0.17123287916183472, 0.1164383590221405, 0.16210044920444489, 0.15296803414821625, 0.1495433747768402, 0.16438356041908264, 0.21461187303066254, 0.2054794579744339, 0.21917808055877686, 0.21917808055877686, 0.23059360682964325, 0.2214611917734146, 0.24315068125724792, 0.2602739632129669, 0.2442922443151474, 0.22374428808689117, 0.2442922443151474, 0.21232876181602478, 0.20776255428791046, 0.2511415481567383, 0.2945205569267273, 0.26369863748550415, 0.2956621050834656, 0.2442922443151474, 0.267123281955719, 0.33105021715164185, 0.28538814187049866, 0.31963470578193665, 0.3173516094684601, 0.2945205569267273, 0.36301368474960327, 0.30365297198295593, 0.3652968108654022, 0.3835616409778595, 0.422374427318573, 0.4543378949165344, 0.4817351698875427, 0.4817351698875427, 0.43607306480407715, 0.4817351698875427, 0.45890411734580994, 0.5205479264259338, 0.4543378949165344, 0.4840182662010193, NaN, 0.5319634675979614, 0.44748857617378235, 0.509132444858551, 0.6860730648040771, 0.5890411138534546, 0.6130136847496033, 0.6415525078773499, 0.664383590221405, 0.6826484203338623, 0.7922374606132507, 0.7031963467597961, 0.9429223537445068], [0.06621004641056061, 0.10958904027938843, 0.09817351400852203, 0.09132420271635056, 0.10730593651533127, 0.13470318913459778, 0.12557077407836914, 0.14155250787734985, 0.17123287916183472, 0.13584475219249725, 0.16894976794719696, 0.15182648599147797, 0.2283104956150055, 0.18493150174617767, 0.20319634675979614, 0.16894976794719696, 0.1438356190919876, 0.1826484054327011, 0.2283104956150055, 0.2340182662010193, 0.19292236864566803, 0.2557077705860138, 0.22374428808689117, 0.2283104956150055, 0.2385844737291336, 0.25799086689949036, 0.2557077705860138, 0.21004566550254822, 0.17579908668994904, 0.24885845184326172, 0.2716895043849945, 0.24657534062862396, 0.2168949842453003, 0.22374428808689117, 0.2397260218858719, 0.2545662224292755, 0.2511415481567383, 0.2682648301124573, 0.30365297198295593, 0.25799086689949036, 0.2442922443151474, 0.39726027846336365, 0.2340182662010193, 0.28082191944122314, 0.33105021715164185, 0.30251142382621765, 0.3652968108654022, 0.32534247636795044, 0.3652968108654022, 0.3436073064804077, 0.37100455164909363, 0.3835616409778595, 0.38584473729133606, 0.36986300349235535, 0.43493151664733887, 0.4109589159488678, 0.49771690368652344, 0.3321917951107025, 0.42009133100509644, 0.4155251085758209, 0.4680365324020386, NaN, 0.47031962871551514, 0.4337899684906006, 0.4885844886302948, 0.6757990717887878, 0.52625572681427, 0.5890411138534546, 0.568493127822876, 0.5456621050834656, 0.6187214851379395, 0.6598173379898071, 0.8835616707801819], [0.04337899386882782, 0.05936073139309883, 0.06164383515715599, 0.06849315017461777, 0.0890410989522934, 0.09132420271635056, 0.08675798773765564, 0.1004566177725792, 0.09360730648040771, 0.09132420271635056, 0.0890410989522934, 0.1278538852930069, 0.1278538852930069, 0.15296803414821625, 0.17123287916183472, 0.21461187303066254, 0.20776255428791046, 0.17579908668994904, 0.155251145362854, 0.12442922592163086, 0.23515981435775757, 0.18493150174617767, 0.21232876181602478, 0.21917808055877686, 0.23515981435775757, 0.20433789491653442, 0.19863013923168182, 0.15068493783473969, 0.16210044920444489, 0.2785388231277466, 0.2990867495536804, 0.2397260218858719, 0.24200913310050964, 0.06392694264650345, 0.19748859107494354, 0.23744292557239532, 0.29223743081092834, 0.2716895043849945, 0.31506848335266113, 0.18493150174617767, 0.28538814187049866, 0.3059360682964325, 0.2785388231277466, 0.3264840245246887, 0.24657534062862396, 0.21461187303066254, 0.35844749212265015, 0.29680365324020386, 0.3607305884361267, 0.344748854637146, 0.3995433747768402, 0.4269406497478485, 0.41324201226234436, 0.4155251085758209, 0.4543378949165344, 0.4292237460613251, 0.5456621050834656, 0.5068492889404297, 0.422374427318573, 0.4680365324020386, 0.5525113940238953, 0.5296803712844849, NaN, 0.47488585114479065, 0.5799086689949036, 0.5662100315093994, 0.6415525078773499, 0.6232876777648926, 0.6278538703918457, 0.6130136847496033, 0.7146118879318237, 0.7397260069847107, 0.9429223537445068], [0.0251141544431448, 0.0388127863407135, 0.054794520139694214, 0.06392694264650345, 0.04566210135817528, 0.06621004641056061, 0.07534246891736984, 0.08675798773765564, 0.10958904027938843, 0.09703196585178375, 0.07077625393867493, 0.11415524780750275, 0.1438356190919876, 0.17579908668994904, 0.17579908668994904, 0.077625572681427, 0.15753424167633057, 0.16894976794719696, 0.18036529421806335, 0.16210044920444489, 0.12557077407836914, 0.1392694115638733, 0.11872146278619766, 0.1107305958867073, 0.13013698160648346, 0.2214611917734146, 0.23059360682964325, 0.17579908668994904, 0.21004566550254822, 0.12442922592163086, 0.19178082048892975, 0.27968037128448486, 0.2054794579744339, 0.232876718044281, 0.21917808055877686, 0.2397260218858719, 0.14611871540546417, 0.22374428808689117, 0.2283104956150055, 0.24200913310050964, 0.23059360682964325, 0.34018266201019287, 0.2283104956150055, 0.2397260218858719, 0.27625569701194763, 0.3219178020954132, 0.3652968108654022, 0.30821916460990906, 0.37442922592163086, 0.4109589159488678, 0.2990867495536804, 0.35388126969337463, 0.34246575832366943, 0.3652968108654022, 0.48630136251449585, 0.38013699650764465, 0.5502282977104187, 0.5410959124565125, 0.45205479860305786, 0.3675799071788788, 0.49086758494377136, 0.5662100315093994, 0.5251141786575317, NaN, 0.5273972749710083, 0.6529680490493774, 0.6769406199455261, 0.664383590221405, 0.7146118879318237, 0.7363013625144958, 0.611872136592865, 0.8082191944122314, 0.965753436088562], [0.015981735661625862, 0.04794520512223244, 0.04908675700426102, 0.04794520512223244, 0.06849315017461777, 0.054794520139694214, 0.06164383515715599, 0.07534246891736984, 0.09589041024446487, 0.07990867644548416, 0.07534246891736984, 0.0890410989522934, 0.12671232223510742, 0.13470318913459778, 0.155251145362854, 0.10273972898721695, 0.155251145362854, 0.10273972898721695, 0.15296803414821625, 0.14840182662010193, 0.1278538852930069, 0.1438356190919876, 0.17579908668994904, 0.18036529421806335, 0.18949772417545319, 0.19178082048892975, 0.2009132355451584, 0.16780821979045868, 0.20205479860305786, 0.2226027399301529, 0.21917808055877686, 0.2111872136592865, 0.22488585114479065, 0.19178082048892975, 0.17351597547531128, 0.21917808055877686, 0.23515981435775757, 0.25342464447021484, 0.2682648301124573, 0.20776255428791046, 0.24885845184326172, 0.3264840245246887, 0.2716895043849945, 0.30365297198295593, 0.26940637826919556, 0.2557077705860138, 0.3561643958091736, 0.26940637826919556, 0.3607305884361267, 0.3675799071788788, 0.4018264710903168, 0.4155251085758209, 0.42465752363204956, 0.4680365324020386, 0.39726027846336365, 0.41324201226234436, 0.4543378949165344, 0.49771690368652344, 0.4155251085758209, 0.45890411734580994, 0.31392693519592285, 0.5114155411720276, 0.42009133100509644, 0.4726027250289917, NaN, 0.5662100315093994, 0.5799086689949036, 0.6221461296081543, 0.6415525078773499, 0.6575342416763306, 0.7340182662010193, 0.664383590221405, 0.9155251383781433], [0.04566210135817528, 0.06849315017461777, 0.07420091331005096, 0.06278538703918457, 0.07990867644548416, 0.07077625393867493, 0.09474886208772659, 0.09018264710903168, 0.09360730648040771, 0.10730593651533127, 0.14840182662010193, 0.12100456655025482, 0.16210044920444489, 0.14269407093524933, 0.19634702801704407, 0.11872146278619766, 0.0890410989522934, 0.15068493783473969, 0.18036529421806335, 0.18493150174617767, 0.14155250787734985, 0.21917808055877686, 0.17579908668994904, 0.1940639317035675, 0.20319634675979614, 0.2054794579744339, 0.20776255428791046, 0.17694063484668732, 0.15068493783473969, 0.2054794579744339, 0.232876718044281, 0.19178082048892975, 0.18036529421806335, 0.19178082048892975, 0.18721461296081543, 0.1826484054327011, 0.21461187303066254, 0.2168949842453003, 0.25342464447021484, 0.20776255428791046, 0.21461187303066254, 0.344748854637146, 0.1826484054327011, 0.23744292557239532, 0.2716895043849945, 0.2511415481567383, 0.30821916460990906, 0.23744292557239532, 0.310502290725708, 0.30821916460990906, 0.29680365324020386, 0.30365297198295593, 0.32420089840888977, 0.3059360682964325, 0.38584473729133606, 0.3378995358943939, 0.4383561611175537, 0.24200913310050964, 0.34018266201019287, 0.3344748914241791, 0.4109589159488678, 0.32420089840888977, 0.4337899684906006, 0.34703195095062256, 0.4337899684906006, NaN, 0.4429223835468292, 0.5639269351959229, 0.5068492889404297, 0.5114155411720276, 0.5559360980987549, 0.5936073064804077, 0.8333333134651184], [0.027397260069847107, 0.06164383515715599, 0.07305935770273209, 0.052511416375637054, 0.07305935770273209, 0.09132420271635056, 0.0890410989522934, 0.09589041024446487, 0.09589041024446487, 0.08675798773765564, 0.09589041024446487, 0.11872146278619766, 0.1335616409778595, 0.14840182662010193, 0.19178082048892975, 0.1004566177725792, 0.0890410989522934, 0.14611871540546417, 0.1438356190919876, 0.155251145362854, 0.13698630034923553, 0.15753424167633057, 0.11415524780750275, 0.11415524780750275, 0.11872146278619766, 0.18949772417545319, 0.17351597547531128, 0.12899543344974518, 0.13013698160648346, 0.15981735289096832, 0.18949772417545319, 0.20319634675979614, 0.13013698160648346, 0.14155250787734985, 0.17351597547531128, 0.14840182662010193, 0.15867580473423004, 0.17579908668994904, 0.2168949842453003, 0.155251145362854, 0.14840182662010193, 0.27625569701194763, 0.13698630034923553, 0.2226027399301529, 0.2168949842453003, 0.2168949842453003, 0.2442922443151474, 0.2271689474582672, 0.2557077705860138, 0.3378995358943939, 0.21461187303066254, 0.26255708932876587, 0.2831050157546997, 0.24200913310050964, 0.388127863407135, 0.30821916460990906, 0.3949771821498871, 0.42465752363204956, 0.30136987566947937, 0.2990867495536804, 0.38698631525039673, 0.47374430298805237, 0.35844749212265015, 0.3230593502521515, 0.42009133100509644, 0.5570776462554932, NaN, 0.5844748616218567, 0.49086758494377136, 0.5251141786575317, 0.5479452013969421, 0.8458904027938843, 0.9680365324020386], [0.0251141544431448, 0.04566210135817528, 0.05707762390375137, 0.04794520512223244, 0.06164383515715599, 0.05707762390375137, 0.07305935770273209, 0.07990867644548416, 0.07077625393867493, 0.07077625393867493, 0.07534246891736984, 0.11415524780750275, 0.0833333358168602, 0.10958904027938843, 0.11187214404344559, 0.15296803414821625, 0.18493150174617767, 0.15753424167633057, 0.11872146278619766, 0.13242009282112122, 0.17465753853321075, 0.0890410989522934, 0.16894976794719696, 0.17123287916183472, 0.1780821979045868, 0.1278538852930069, 0.17351597547531128, 0.06621004641056061, 0.10273972898721695, 0.1666666716337204, 0.20319634675979614, 0.16894976794719696, 0.23059360682964325, 0.1438356190919876, 0.17579908668994904, 0.2168949842453003, 0.1826484054327011, 0.2442922443151474, 0.2283104956150055, 0.2168949842453003, 0.24657534062862396, 0.18607306480407715, 0.267123281955719, 0.22602739930152893, 0.1940639317035675, 0.2511415481567383, 0.2899543344974518, 0.2397260218858719, 0.29680365324020386, 0.25799086689949036, 0.29223743081092834, 0.31963470578193665, 0.28082191944122314, 0.3378995358943939, 0.3675799071788788, 0.27397260069847107, 0.37442922592163086, 0.4018264710903168, 0.34703195095062256, 0.31506848335266113, 0.35844749212265015, 0.4109589159488678, 0.3767123222351074, 0.33561643958091736, 0.3778538703918457, 0.43607306480407715, 0.4155251085758209, NaN, 0.4497717022895813, 0.41324201226234436, 0.5136986374855042, 0.5228310227394104, 0.8630136847496033], [0.015981735661625862, 0.029680365696549416, 0.0502283088862896, 0.04794520512223244, 0.054794520139694214, 0.054794520139694214, 0.05707762390375137, 0.06621004641056061, 0.06963470578193665, 0.06392694264650345, 0.06164383515715599, 0.077625572681427, 0.07305935770273209, 0.11415524780750275, 0.12100456655025482, 0.077625572681427, 0.13698630034923553, 0.1164383590221405, 0.1164383590221405, 0.15068493783473969, 0.09360730648040771, 0.12100456655025482, 0.12557077407836914, 0.10958904027938843, 0.11872146278619766, 0.15068493783473969, 0.16438356041908264, 0.07990867644548416, 0.13698630034923553, 0.12557077407836914, 0.16438356041908264, 0.17579908668994904, 0.14840182662010193, 0.16210044920444489, 0.19634702801704407, 0.16780821979045868, 0.1278538852930069, 0.16438356041908264, 0.17351597547531128, 0.16894976794719696, 0.15753424167633057, 0.23515981435775757, 0.16894976794719696, 0.2009132355451584, 0.23059360682964325, 0.232876718044281, 0.2168949842453003, 0.24657534062862396, 0.22374428808689117, 0.27397260069847107, 0.1278538852930069, 0.1940639317035675, 0.2442922443151474, 0.30136987566947937, 0.3767123222351074, 0.24885845184326172, 0.38127854466438293, 0.4063926935195923, 0.33561643958091736, 0.24885845184326172, 0.33561643958091736, 0.43150684237480164, 0.3721461296081543, 0.28538814187049866, 0.35844749212265015, 0.4931506812572479, 0.509132444858551, 0.5502282977104187, NaN, 0.5228310227394104, 0.49771690368652344, 0.6415525078773499, 0.956620991230011], [0.02283105067908764, 0.0251141544431448, 0.04794520512223244, 0.04337899386882782, 0.0388127863407135, 0.04337899386882782, 0.054794520139694214, 0.05936073139309883, 0.06392694264650345, 0.06392694264650345, 0.0388127863407135, 0.077625572681427, 0.07534246891736984, 0.10502283275127411, 0.10273972898721695, 0.04794520512223244, 0.10730593651533127, 0.10730593651533127, 0.10273972898721695, 0.11872146278619766, 0.07534246891736984, 0.09589041024446487, 0.1004566177725792, 0.07305935770273209, 0.08219178020954132, 0.15296803414821625, 0.15753424167633057, 0.06621004641056061, 0.13470318913459778, 0.07990867644548416, 0.155251145362854, 0.16894976794719696, 0.1438356190919876, 0.14611871540546417, 0.1826484054327011, 0.17123287916183472, 0.1004566177725792, 0.14840182662010193, 0.18493150174617767, 0.1666666716337204, 0.15981735289096832, 0.232876718044281, 0.1780821979045868, 0.19178082048892975, 0.20776255428791046, 0.21232876181602478, 0.2214611917734146, 0.23515981435775757, 0.22602739930152893, 0.30136987566947937, 0.16210044920444489, 0.1997716873884201, 0.2283104956150055, 0.26940637826919556, 0.34703195095062256, 0.23173515498638153, 0.4063926935195923, 0.4337899684906006, 0.33561643958091736, 0.1826484054327011, 0.3173516094684601, 0.4543378949165344, 0.38698631525039673, 0.26369863748550415, 0.34246575832366943, 0.4885844886302948, 0.47488585114479065, 0.586758017539978, 0.4771689474582672, NaN, 0.5045661926269531, 0.6347032189369202, 0.9406392574310303], [0.01141552533954382, 0.029680365696549416, 0.03652967885136604, 0.02054794505238533, 0.0388127863407135, 0.029680365696549416, 0.0388127863407135, 0.03652967885136604, 0.034246575087308884, 0.054794520139694214, 0.034246575087308884, 0.034246575087308884, 0.06392694264650345, 0.07534246891736984, 0.10502283275127411, 0.054794520139694214, 0.07990867644548416, 0.04337899386882782, 0.06621004641056061, 0.052511416375637054, 0.05707762390375137, 0.09132420271635056, 0.10730593651533127, 0.11415524780750275, 0.10502283275127411, 0.1164383590221405, 0.10730593651533127, 0.08675798773765564, 0.10502283275127411, 0.12328767031431198, 0.14155250787734985, 0.16894976794719696, 0.1164383590221405, 0.06164383515715599, 0.09360730648040771, 0.14611871540546417, 0.1278538852930069, 0.13698630034923553, 0.14726027846336365, 0.09132420271635056, 0.13242009282112122, 0.22602739930152893, 0.14840182662010193, 0.18949772417545319, 0.1666666716337204, 0.1392694115638733, 0.29680365324020386, 0.15296803414821625, 0.2990867495536804, 0.20776255428791046, 0.27397260069847107, 0.2945205569267273, 0.29223743081092834, 0.35844749212265015, 0.30136987566947937, 0.267123281955719, 0.3767123222351074, 0.37899544835090637, 0.27397260069847107, 0.3721461296081543, 0.20776255428791046, 0.38127854466438293, 0.28538814187049866, 0.388127863407135, 0.2659817337989807, 0.4440639317035675, 0.45205479860305786, 0.48630136251449585, 0.5022830963134766, 0.4954337775707245, NaN, 0.5525113940238953, 0.8926940560340881], [0.027397260069847107, 0.03652967885136604, 0.054794520139694214, 0.04109589010477066, 0.052511416375637054, 0.054794520139694214, 0.06849315017461777, 0.06621004641056061, 0.04794520512223244, 0.08219178020954132, 0.07534246891736984, 0.09132420271635056, 0.09132420271635056, 0.06164383515715599, 0.14840182662010193, 0.07876712083816528, 0.0502283088862896, 0.1164383590221405, 0.10273972898721695, 0.1278538852930069, 0.09132420271635056, 0.11986301094293594, 0.08675798773765564, 0.10502283275127411, 0.09589041024446487, 0.1438356190919876, 0.1278538852930069, 0.11872146278619766, 0.10730593651533127, 0.11757990717887878, 0.13698630034923553, 0.12100456655025482, 0.10159817337989807, 0.12328767031431198, 0.12557077407836914, 0.10502283275127411, 0.12100456655025482, 0.10958904027938843, 0.1666666716337204, 0.1278538852930069, 0.1004566177725792, 0.24200913310050964, 0.09132420271635056, 0.14840182662010193, 0.17009133100509644, 0.15981735289096832, 0.14840182662010193, 0.18949772417545319, 0.15981735289096832, 0.28538814187049866, 0.15753424167633057, 0.15068493783473969, 0.21461187303066254, 0.1780821979045868, 0.3059360682964325, 0.2054794579744339, 0.27397260069847107, 0.3219178020954132, 0.2214611917734146, 0.18493150174617767, 0.29680365324020386, 0.34018266201019287, 0.2602739632129669, 0.19178082048892975, 0.33561643958091736, 0.4063926935195923, 0.15410958230495453, 0.4771689474582672, 0.35844749212265015, 0.3652968108654022, 0.44748857617378235, NaN, 0.9497717022895813], [0.006849315017461777, 0.004566209856420755, 0.006849315017461777, 0.004566209856420755, 0.00913241971284151, 0.006849315017461777, 0.00913241971284151, 0.00913241971284151, 0.006849315017461777, 0.004566209856420755, 0.006849315017461777, 0.01826483942568302, 0.006849315017461777, 0.006849315017461777, 0.034246575087308884, 0.00913241971284151, 0.015981735661625862, 0.027397260069847107, 0.01141552533954382, 0.02054794505238533, 0.006849315017461777, 0.0251141544431448, 0.013698630034923553, 0.00913241971284151, 0.00913241971284151, 0.02283105067908764, 0.00913241971284151, 0.01141552533954382, 0.01141552533954382, 0.02054794505238533, 0.028538811951875687, 0.02054794505238533, 0.02283105067908764, 0.0251141544431448, 0.027397260069847107, 0.01826483942568302, 0.0251141544431448, 0.031963471323251724, 0.04109589010477066, 0.0251141544431448, 0.02283105067908764, 0.06849315017461777, 0.0251141544431448, 0.034246575087308884, 0.04109589010477066, 0.034246575087308884, 0.06849315017461777, 0.0388127863407135, 0.06849315017461777, 0.0388127863407135, 0.01141552533954382, 0.013698630034923553, 0.06164383515715599, 0.06392694264650345, 0.10730593651533127, 0.04223744198679924, 0.077625572681427, 0.10730593651533127, 0.05936073139309883, 0.03652967885136604, 0.05707762390375137, 0.1164383590221405, 0.05707762390375137, 0.034246575087308884, 0.08447488397359848, 0.1666666716337204, 0.031963471323251724, 0.13698630034923553, 0.04337899386882782, 0.05936073139309883, 0.10730593651533127, 0.0502283088862896, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9125285347},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8461440876},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.835109083},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8272926168},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.7962011682},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7788241975},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7717053514},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7529173001},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7142313576},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7090943687},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6898306684},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6829337858},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6470383065},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6454369602},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6451040083},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6408231849},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6370497201},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.628392949},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6048959894},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6007737172},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5966197374},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5926718657},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5870909418},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5836821402},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.576531583},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5660514983},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5629439373},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5617706744},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5590277774},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5470414767},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5404616953},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5384005575},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.535927195},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5318524863},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5268106309},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5223236936},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5156329277},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5073091068},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5063102455},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4997621741},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4983986546},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4886478938},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4870148411},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4847158833},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4782946459},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4549879486},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4471714863},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4409722222},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.4395611381},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.4265284116},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3859240227},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3710996955},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3707350313},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3627917293},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.3571632408},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.344051244},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.3381532219},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.3315258758},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3219019528},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3149257995},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3110889155},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.3060787684},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.3036529667},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.2925069771},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2822647135},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.2521562658},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2370782604},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2291666658},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2144216136},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2051306452},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.190116058},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1711377482},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0360857432}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_binary/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..3648e8374408f6943405fd5493358e119cdac4c3
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_all/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),TabDPT-Turbo (default),"AutoGluon 1.4 (best, 4h)",RealMLP (tuned + ensembled),TabM (tuned + ensembled),TabDPT (tuned + ensembled),TabSwift (default),LimiX (default),LightGBM (tuned + ensembled),BetaTabPFN (default),TabM (tuned),TabDPT (tuned),RealMLP (tuned),LightGBM (tuned),iLTM (tuned + ensembled),CatBoost (default),CatBoost (tuned + ensembled),CatBoost (tuned),ModernNCA (tuned),TabDPT (default),xRFM (tuned + ensembled),xRFM (tuned),XGBoost (tuned + ensembled),ChimeraBoost (tuned + ensembled),SAP-RPT-OSS (default),EBM (tuned + ensembled),FastaiMLP (tuned + ensembled),TorchMLP (tuned + ensembled),ModernNCA (default),XGBoost (tuned),TabM (default),ChimeraBoost (tuned),RealMLP (default),EBM (tuned),ModernNCA (tuned + ensembled),EBM (default),ChimeraBoost (default),iLTM (tuned),FastaiMLP (tuned),TabSTAR (tuned),TorchMLP (tuned),TabSTAR (tuned + ensembled),OrionMSP (default),ExtraTrees (tuned + ensembled),ExtraTrees (tuned),XGBoost (default),iLTM (default),TabFlex (default),LightGBM (default),TabSTAR (default),Linear (tuned + ensembled),TorchMLP (default),RandomForest (tuned + ensembled),PerpetualBooster (tuned + ensembled),Linear (tuned),FastaiMLP (default),RandomForest (tuned),PerpetualBooster (tuned),Linear (default),KNN (tuned + ensembled),xRFM (default),ExtraTrees (default),RandomForest (default),PerpetualBooster (default),KNN (tuned),KNN (default)
+TabFM (default),,0.79680365,0.75570774,0.8105023,0.8196347,0.8150685,0.82990867,0.8356164,0.83789957,0.8561644,0.8767123,0.9063927,0.89726025,0.86757994,0.90525115,0.88812786,0.8664383,0.93835616,0.89269406,0.8721461,0.9109589,0.9041096,0.89497715,0.89269406,0.9018265,0.913242,0.9018265,0.9041096,0.8835617,0.93607306,0.91780823,0.89041096,0.9086758,0.89726025,0.9543379,0.89726025,0.94520545,0.93607306,0.9292237,0.913242,0.91780823,0.94063926,0.9109589,0.93835616,0.9269406,0.9269406,0.8196347,0.96803653,0.82191783,0.94292235,0.9520548,0.94292235,0.9703196,0.94063926,0.9269406,0.9726027,0.8538813,0.9246575,0.977169,0.977169,0.98173517,0.93378997,0.956621,0.9748858,0.98401827,0.9543379,0.9726027,0.9748858,0.98401827,0.977169,0.98858446,0.9726027,0.9931507
+TabPFN-3 (default),0.20319635,,0.5958904,0.51141554,0.65525115,0.70319635,0.6940639,0.7191781,0.7351598,0.7191781,0.78995436,0.79223746,0.8196347,0.8310502,0.77397263,0.7853881,0.772831,0.83789957,0.84931505,0.8105023,0.84246576,0.84931505,0.8310502,0.8242009,0.8333333,0.8607306,0.84703195,0.84246576,0.8150685,0.8481735,0.8744292,0.8287671,0.84474885,0.8413242,0.88812786,0.8504566,0.8721461,0.89041096,0.8789954,0.8812785,0.8630137,0.9109589,0.86415523,0.86986303,0.90525115,0.8732877,0.78424656,0.93378997,0.79680365,0.9018265,0.93835616,0.93378997,0.94063926,0.9292237,0.9041096,0.9543379,0.8310502,0.8789954,0.9474886,0.93835616,0.93607306,0.89041096,0.94063926,0.96118724,0.9520548,0.9315069,0.93835616,0.9543379,0.9703196,0.9748858,0.9703196,0.96347034,0.9954338
+"AutoGluon 1.5 (extreme, 4h)",0.24429224,0.4041096,,0.456621,0.6484018,0.63013697,0.65068495,0.6643836,0.60730594,0.7762557,0.81621003,0.79908675,0.7762557,0.7671233,0.74657536,0.8310502,0.7579909,0.8184931,0.79452056,0.84703195,0.84018266,0.8744292,0.8436073,0.8538813,0.8436073,0.86986303,0.7762557,0.86757994,0.8230594,0.8789954,0.88812786,0.80365294,0.84018266,0.8561644,0.8995434,0.8356164,0.89269406,0.8789954,0.89726025,0.8538813,0.84703195,0.93378997,0.8561644,0.9018265,0.89269406,0.8652968,0.74657536,0.92009133,0.74885845,0.90525115,0.9246575,0.9086758,0.93378997,0.9223744,0.8812785,0.9474886,0.80136985,0.8744292,0.9474886,0.94292235,0.93607306,0.9018265,0.93835616,0.94520545,0.95091325,0.9257991,0.9269406,0.94292235,0.9497717,0.9520548,0.96347034,0.94520545,0.9931507
+TabICLv2 (default),0.18949772,0.4885845,0.543379,,0.6004566,0.6312785,0.6461187,0.6643836,0.7191781,0.7191781,0.74429226,0.7579909,0.81392694,0.8082192,0.78424656,0.7648402,0.77397263,0.79452056,0.8127854,0.7716895,0.80365294,0.8150685,0.79680365,0.7853881,0.79452056,0.8333333,0.84018266,0.8173516,0.7910959,0.8105023,0.8242009,0.80136985,0.8196347,0.84018266,0.8789954,0.8504566,0.8287671,0.8607306,0.84246576,0.86986303,0.84474885,0.86986303,0.8561644,0.83789957,0.8584475,0.8812785,0.79223746,0.91780823,0.78995436,0.9246575,0.91210043,0.9109589,0.913242,0.9086758,0.89726025,0.94063926,0.8356164,0.89497715,0.956621,0.93835616,0.94292235,0.9086758,0.9315069,0.93607306,0.9520548,0.9372146,0.9474886,0.9520548,0.9520548,0.956621,0.9794521,0.9589041,0.9954338
+TabPFN-2.6 (default),0.1803653,0.34474885,0.35159817,0.39954337,,0.5456621,0.53424656,0.5673516,0.60502285,0.65753424,0.7214612,0.7214612,0.7214612,0.7146119,0.70547944,0.75570774,0.7214612,0.7785388,0.7305936,0.783105,0.8127854,0.79452056,0.77397263,0.7808219,0.7671233,0.8242009,0.7191781,0.80365294,0.783105,0.7853881,0.84931505,0.8059361,0.80136985,0.78995436,0.8538813,0.84018266,0.8310502,0.8458904,0.8630137,0.8561644,0.8390411,0.8812785,0.8242009,0.8584475,0.8584475,0.84246576,0.74429226,0.913242,0.75570774,0.8515982,0.913242,0.90525115,0.93607306,0.8995434,0.8789954,0.9474886,0.78995436,0.86757994,0.9497717,0.93607306,0.9315069,0.89269406,0.9109589,0.9543379,0.9315069,0.92009133,0.9269406,0.93835616,0.94520545,0.96118724,0.96118724,0.9474886,0.99086756
+RealTabPFN-2.5 (default),0.1849315,0.29680365,0.369863,0.36872146,0.4543379,,0.48173517,0.5251142,0.53881276,0.65068495,0.66210043,0.67808217,0.6826484,0.75342464,0.6849315,0.7214612,0.7294521,0.7328767,0.71575344,0.74429226,0.7648402,0.7237443,0.7625571,0.7351598,0.7351598,0.75570774,0.739726,0.79452056,0.7876712,0.760274,0.79908675,0.80136985,0.79452056,0.8059361,0.826484,0.8059361,0.79908675,0.8082192,0.8150685,0.826484,0.8356164,0.783105,0.8196347,0.8333333,0.8458904,0.8538813,0.7785388,0.89726025,0.78196347,0.913242,0.89269406,0.89726025,0.88812786,0.89269406,0.90296805,0.94063926,0.8127854,0.8538813,0.96118724,0.89726025,0.9315069,0.8652968,0.9086758,0.93378997,0.94520545,0.9292237,0.9086758,0.94292235,0.94520545,0.956621,0.9703196,0.94520545,0.9931507
+RealTabPFN-2.5 (tuned + ensembled),0.17009133,0.30593607,0.34931508,0.35388127,0.46575344,0.51826483,,0.6255708,0.53196347,0.65981734,0.6712329,0.6712329,0.6666667,0.70091325,0.70319635,0.72602737,0.7237443,0.7283105,0.716895,0.72602737,0.76598173,0.73401827,0.75570774,0.7420091,0.74657536,0.80136985,0.673516,0.7785388,0.7625571,0.7625571,0.78652966,0.772831,0.7785388,0.7773973,0.82191783,0.79452056,0.793379,0.79908675,0.8059361,0.8105023,0.80936074,0.79908675,0.8127854,0.8242009,0.8196347,0.84703195,0.74657536,0.8789954,0.75342464,0.8767123,0.8835617,0.8767123,0.88812786,0.8812785,0.8744292,0.93378997,0.78995436,0.8538813,0.9246575,0.913242,0.9315069,0.8744292,0.913242,0.9246575,0.93835616,0.90525115,0.9109589,0.9269406,0.94292235,0.94520545,0.96118724,0.9315069,0.99086756
+RealTabPFN-2.5 (tuned),0.16438356,0.28082192,0.33561644,0.33561644,0.4326484,0.47488585,0.37442923,,0.522831,0.6278539,0.63242006,0.6484018,0.65981734,0.68607306,0.66210043,0.70547944,0.70319635,0.71118724,0.7043379,0.70776254,0.760274,0.70547944,0.7283105,0.7283105,0.7237443,0.74657536,0.6757991,0.7522831,0.74429226,0.74429226,0.75913244,0.760274,0.7716895,0.7579909,0.8082192,0.7694064,0.7671233,0.7853881,0.79223746,0.79680365,0.7876712,0.78881276,0.79452056,0.8059361,0.80365294,0.8310502,0.7374429,0.8767123,0.74885845,0.8721461,0.8687215,0.8721461,0.8812785,0.86986303,0.8607306,0.9223744,0.79223746,0.8333333,0.9223744,0.88812786,0.91210043,0.8584475,0.8995434,0.913242,0.9246575,0.90981734,0.9041096,0.92009133,0.93378997,0.94063926,0.96347034,0.93378997,0.99086756
+TabDPT-Turbo (default),0.16210045,0.2648402,0.39269406,0.28082192,0.39497718,0.4611872,0.46803653,0.47716895,,0.53424656,0.5730594,0.54223746,0.65068495,0.68835616,0.6232877,0.5890411,0.5730594,0.60730594,0.70776254,0.66780823,0.6232877,0.65296805,0.6461187,0.6689498,0.6643836,0.65068495,0.80936074,0.6940639,0.71004564,0.6803653,0.6849315,0.6826484,0.6940639,0.7374429,0.7283105,0.72602737,0.69863015,0.69863015,0.7214612,0.75114155,0.7203196,0.6803653,0.7374429,0.7351598,0.75913244,0.79452056,0.70776254,0.79452056,0.716895,0.84018266,0.8458904,0.8789954,0.8173516,0.79908675,0.8721461,0.8607306,0.78995436,0.8082192,0.8789954,0.8767123,0.8812785,0.8287671,0.9063927,0.89041096,0.9041096,0.9063927,0.9041096,0.9292237,0.9303653,0.93607306,0.96575344,0.9520548,0.9931507
+"AutoGluon 1.4 (best, 4h)",0.14383562,0.28082192,0.22374429,0.28082192,0.34246576,0.34931508,0.34018266,0.37214613,0.46575344,,0.59703195,0.55936074,0.543379,0.5171233,0.48173517,0.6826484,0.58447486,0.6255708,0.5684931,0.69634706,0.7214612,0.7123288,0.71004564,0.7237443,0.7328767,0.65753424,0.60502285,0.69520545,0.70776254,0.783105,0.7454338,0.6164383,0.7283105,0.7351598,0.783105,0.70319635,0.783105,0.74429226,0.7785388,0.7762557,0.7420091,0.7237443,0.7648402,0.79452056,0.79452056,0.7785388,0.70319635,0.8287671,0.7123288,0.6974886,0.79908675,0.8082192,0.89726025,0.84474885,0.7876712,0.9246575,0.760274,0.836758,0.88584477,0.89269406,0.91780823,0.86415523,0.9086758,0.90296805,0.92009133,0.89269406,0.913242,0.9292237,0.93607306,0.93607306,0.94520545,0.91780823,0.9954338
+RealMLP (tuned + ensembled),0.12328767,0.21004567,0.18378995,0.25570777,0.27853882,0.33789954,0.32876712,0.3675799,0.42694065,0.40296805,,0.5045662,0.50114155,0.47945204,0.4634703,0.6187215,0.5502283,0.57990867,0.54908675,0.7305936,0.68607306,0.6757991,0.65296805,0.6803653,0.6826484,0.6415525,0.5936073,0.6872146,0.66210043,0.7328767,0.70547944,0.6255708,0.7351598,0.7477169,0.7283105,0.6757991,0.77397263,0.7420091,0.74086756,0.8253425,0.75570774,0.6849315,0.7568493,0.7420091,0.7579909,0.772831,0.6940639,0.7876712,0.70547944,0.71004564,0.83789957,0.8276256,0.8721461,0.8561644,0.7648402,0.9018265,0.74885845,0.826484,0.89269406,0.9109589,0.913242,0.8310502,0.9109589,0.9292237,0.9246575,0.8515982,0.9041096,0.9246575,0.93835616,0.96118724,0.96575344,0.9246575,0.9931507
+TabM (tuned + ensembled),0.09360731,0.20776255,0.20091324,0.24200913,0.27853882,0.3219178,0.32876712,0.35159817,0.45776257,0.44063926,0.49543378,,0.5376712,0.5456621,0.4634703,0.61187214,0.5913242,0.6746575,0.56392694,0.6347032,0.66552514,0.6267123,0.6666667,0.65296805,0.65753424,0.6609589,0.5890411,0.60502285,0.58219177,0.69634706,0.7146119,0.63013697,0.7237443,0.673516,0.6894977,0.6940639,0.75570774,0.82990867,0.7671233,0.76826483,0.7671233,0.70776254,0.7568493,0.7853881,0.74429226,0.7351598,0.70547944,0.78652966,0.71004564,0.6940639,0.8173516,0.79908675,0.8538813,0.84931505,0.7671233,0.90525115,0.75342464,0.8333333,0.89041096,0.8652968,0.88812786,0.8481735,0.8721461,0.88584477,0.9109589,0.8789954,0.8812785,0.88584477,0.9223744,0.9223744,0.96575344,0.9086758,0.98173517
+TabDPT (tuned + ensembled),0.10273973,0.1803653,0.22374429,0.18607306,0.27853882,0.3173516,0.33333334,0.34018266,0.34931508,0.456621,0.49885845,0.46232876,,0.5171233,0.50570774,0.5251142,0.49543378,0.5262557,0.6392694,0.59018266,0.5753425,0.5981735,0.58219177,0.5936073,0.5913242,0.60273975,0.6484018,0.6289954,0.61757994,0.6015982,0.6232877,0.6415525,0.6221461,0.6643836,0.6815069,0.663242,0.65296805,0.6289954,0.65753424,0.66210043,0.6518265,0.6849315,0.6461187,0.6894977,0.7237443,0.7351598,0.65525115,0.7374429,0.65753424,0.783105,0.79452056,0.80136985,0.7808219,0.80136985,0.79680365,0.8310502,0.760274,0.75114155,0.84018266,0.8561644,0.8584475,0.7716895,0.8721461,0.8561644,0.8732877,0.83789957,0.8664383,0.9166667,0.9269406,0.9246575,0.93607306,0.9086758,0.9931507
+TabSwift (default),0.1324201,0.16894977,0.23287672,0.19178082,0.28538814,0.24657534,0.29908675,0.31392694,0.31164384,0.48287672,0.5205479,0.4543379,0.48287672,,0.5251142,0.522831,0.49543378,0.50913244,0.543379,0.5958904,0.5502283,0.5981735,0.56392694,0.5753425,0.59018266,0.6210046,0.61415523,0.65068495,0.6164383,0.61187214,0.6278539,0.6232877,0.63242006,0.70547944,0.65981734,0.65525115,0.6484018,0.6392694,0.6666667,0.65753424,0.65525115,0.65981734,0.65525115,0.70319635,0.7283105,0.74429226,0.70776254,0.7191781,0.7191781,0.84474885,0.80136985,0.79680365,0.760274,0.7785388,0.88812786,0.78881276,0.7876712,0.78995436,0.8607306,0.8059361,0.8652968,0.8150685,0.84703195,0.8242009,0.8652968,0.85730594,0.8515982,0.89041096,0.88584477,0.89497715,0.9246575,0.93835616,0.9931507
+LimiX (default),0.09474886,0.2260274,0.25342464,0.21575342,0.29452056,0.31506848,0.29680365,0.33789954,0.37671232,0.51826483,0.53652966,0.53652966,0.49429223,0.47488585,,0.60273975,0.51826483,0.54908675,0.5068493,0.59474885,0.6210046,0.69520545,0.63013697,0.6415525,0.6392694,0.6278539,0.56164384,0.6347032,0.6130137,0.6803653,0.6917808,0.6369863,0.6689498,0.63812786,0.6826484,0.6461187,0.70319635,0.65068495,0.7123288,0.66210043,0.6849315,0.6917808,0.6803653,0.7191781,0.70091325,0.71118724,0.63812786,0.739726,0.6369863,0.727169,0.7579909,0.75570774,0.7762557,0.7694064,0.7146119,0.79680365,0.7043379,0.7716895,0.7808219,0.80365294,0.82191783,0.79680365,0.8287671,0.8242009,0.84474885,0.80365294,0.8082192,0.88812786,0.8789954,0.89726025,0.89497715,0.8515982,0.96575344
+LightGBM (tuned + ensembled),0.111872144,0.21461187,0.16894977,0.23515981,0.24429224,0.27853882,0.2739726,0.29452056,0.41095892,0.3173516,0.38127854,0.38812786,0.47488585,0.47716895,0.39726028,,0.44520548,0.47146118,0.5205479,0.51141554,0.6461187,0.54908675,0.6244292,0.6210046,0.6244292,0.5525114,0.5936073,0.5913242,0.60958904,0.7351598,0.6061644,0.5890411,0.6541096,0.6004566,0.60730594,0.61415523,0.7853881,0.69520545,0.66210043,0.66210043,0.69634706,0.6210046,0.75342464,0.74657536,0.6940639,0.6872146,0.6815069,0.72602737,0.6849315,0.6757991,0.80707765,0.78995436,0.86986303,0.84931505,0.70776254,0.9497717,0.760274,0.8173516,0.8105023,0.8995434,0.8744292,0.8310502,0.7853881,0.9223744,0.89726025,0.8812785,0.8995434,0.84703195,0.9223744,0.9520548,0.94520545,0.9212329,0.99086756
+BetaTabPFN (default),0.13356164,0.22716895,0.24200913,0.2260274,0.27853882,0.27054796,0.2762557,0.29680365,0.42694065,0.4155251,0.4497717,0.4086758,0.5045662,0.5045662,0.48173517,0.55479455,,0.49543378,0.55936074,0.5205479,0.56164384,0.5479452,0.60502285,0.5730594,0.59703195,0.5833333,0.5981735,0.58447486,0.56392694,0.6347032,0.6289954,0.60273975,0.65525115,0.6130137,0.6278539,0.6392694,0.65525115,0.6803653,0.65525115,0.65525115,0.67808217,0.5958904,0.7305936,0.6894977,0.6438356,0.70319635,0.72602737,0.7123288,0.7328767,0.6940639,0.79223746,0.79223746,0.7648402,0.8127854,0.76141554,0.8184931,0.78881276,0.8310502,0.8105023,0.84246576,0.8287671,0.8561644,0.79223746,0.84246576,0.84474885,0.9109589,0.9109589,0.8150685,0.8630137,0.89269406,0.92009133,0.9497717,0.98401827
+TabM (tuned),0.061643835,0.16210045,0.18150684,0.20547946,0.22146119,0.26712328,0.2716895,0.2888128,0.39269406,0.37442923,0.42009133,0.32534248,0.4737443,0.49086758,0.45091325,0.5285388,0.5045662,,0.51141554,0.53424656,0.5707763,0.5525114,0.6004566,0.60502285,0.60958904,0.6278539,0.55136985,0.53196347,0.53652966,0.6267123,0.6666667,0.55479455,0.61187214,0.6244292,0.6461187,0.65068495,0.65525115,0.71004564,0.6826484,0.7146119,0.6803653,0.65296805,0.69634706,0.71575344,0.6712329,0.70319635,0.6872146,0.73401827,0.69063926,0.6666667,0.75114155,0.7579909,0.8082192,0.8150685,0.75114155,0.8356164,0.7328767,0.80136985,0.8744292,0.8333333,0.8835617,0.8173516,0.8242009,0.8310502,0.89726025,0.84931505,0.8538813,0.84246576,0.8835617,0.89269406,0.956621,0.8835617,0.9726027
+TabDPT (tuned),0.10730594,0.15068494,0.20547946,0.18721461,0.26940638,0.28424656,0.28310502,0.2956621,0.29223743,0.43150684,0.45091325,0.43607306,0.3607306,0.456621,0.49315068,0.47945204,0.44063926,0.4885845,,0.51369864,0.51369864,0.53652966,0.5479452,0.55479455,0.55136985,0.53881276,0.6004566,0.586758,0.5502283,0.5673516,0.58447486,0.5707763,0.5958904,0.61415523,0.6347032,0.61187214,0.5890411,0.5776256,0.60502285,0.6255708,0.6278539,0.6267123,0.6278539,0.6232877,0.65525115,0.6940639,0.6038813,0.69634706,0.60502285,0.7785388,0.739726,0.760274,0.7123288,0.7351598,0.7785388,0.75114155,0.7374429,0.74086756,0.8105023,0.7853881,0.83789957,0.7716895,0.84474885,0.8196347,0.84703195,0.8196347,0.8561644,0.8812785,0.8835617,0.89726025,0.93378997,0.89726025,0.98858446
+RealMLP (tuned),0.12785389,0.18949772,0.15296803,0.2283105,0.21689498,0.25570777,0.2739726,0.29223743,0.3321918,0.30365297,0.26940638,0.3652968,0.40981734,0.4041096,0.40525115,0.4885845,0.47945204,0.46575344,0.48630136,,0.53881276,0.49429223,0.543379,0.51369864,0.522831,0.522831,0.522831,0.53881276,0.5559361,0.5936073,0.55707765,0.5479452,0.61415523,0.63242006,0.63242006,0.5981735,0.6495434,0.6392694,0.60273975,0.6940639,0.65981734,0.58219177,0.6278539,0.65068495,0.6438356,0.71004564,0.6712329,0.7237443,0.6826484,0.6335617,0.7420091,0.75342464,0.7762557,0.79223746,0.74086756,0.8105023,0.7351598,0.75570774,0.84474885,0.82191783,0.84703195,0.76598173,0.8755708,0.83789957,0.8515982,0.8150685,0.84474885,0.86757994,0.84931505,0.8812785,0.9474886,0.8721461,0.9794521
+LightGBM (tuned),0.0890411,0.15753424,0.15981735,0.19634703,0.18721461,0.23515981,0.23401827,0.23972602,0.37671232,0.27853882,0.31392694,0.3344749,0.42465752,0.4497717,0.37899545,0.35388127,0.43835616,0.42922375,0.48630136,0.4611872,,0.48173517,0.5605023,0.5582192,0.5479452,0.4885845,0.5410959,0.5559361,0.5502283,0.6244292,0.5502283,0.5410959,0.6210046,0.5696347,0.55707765,0.586758,0.6849315,0.65068495,0.58447486,0.64041096,0.6484018,0.58447486,0.6872146,0.6894977,0.63242006,0.6210046,0.6643836,0.66552514,0.6689498,0.65753424,0.7568493,0.76598173,0.8184931,0.8207763,0.6917808,0.9143836,0.7420091,0.783105,0.79223746,0.84018266,0.8504566,0.80707765,0.7648402,0.8744292,0.8721461,0.8584475,0.8630137,0.8253425,0.9063927,0.9246575,0.94292235,0.9086758,0.9931507
+iLTM (tuned + ensembled),0.09589041,0.15068494,0.12557077,0.1849315,0.20547946,0.2762557,0.26598173,0.29452056,0.34703195,0.28767124,0.3242009,0.37328768,0.40182647,0.40182647,0.30479452,0.45091325,0.4520548,0.44748858,0.4634703,0.50570774,0.51826483,,0.4977169,0.47716895,0.5068493,0.53424656,0.5205479,0.5742009,0.59931505,0.5650685,0.5753425,0.5502283,0.5981735,0.63013697,0.61415523,0.60273975,0.6347032,0.58219177,0.5913242,0.6643836,0.63242006,0.60502285,0.6392694,0.63584477,0.71118724,0.68835616,0.6347032,0.70547944,0.63812786,0.6267123,0.7294521,0.74657536,0.8105023,0.78995436,0.7351598,0.8584475,0.7191781,0.7191781,0.8173516,0.84018266,0.8356164,0.74429226,0.8150685,0.8607306,0.8561644,0.7808219,0.84246576,0.9109589,0.8789954,0.9041096,0.9086758,0.88013697,0.9748858
+CatBoost (default),0.10502283,0.16894977,0.1563927,0.20319635,0.2260274,0.23744293,0.24429224,0.2716895,0.35388127,0.28995433,0.34703195,0.33333334,0.4178082,0.43607306,0.369863,0.37557077,0.39497718,0.39954337,0.4520548,0.456621,0.4394977,0.5022831,,0.49315068,0.49429223,0.5068493,0.51369864,0.5376712,0.55479455,0.5696347,0.5650685,0.56621003,0.5890411,0.56392694,0.5936073,0.5456621,0.6130137,0.6187215,0.6255708,0.6210046,0.6232877,0.586758,0.65525115,0.6712329,0.6643836,0.6484018,0.6723744,0.70091325,0.67808217,0.6004566,0.7716895,0.7648402,0.8173516,0.8310502,0.65753424,0.8515982,0.7283105,0.75570774,0.79908675,0.8767123,0.7853881,0.7762557,0.7876712,0.8812785,0.8242009,0.8242009,0.88584477,0.8310502,0.8744292,0.8995434,0.89269406,0.913242,0.98630136
+CatBoost (tuned + ensembled),0.10730594,0.17579909,0.14611872,0.21461187,0.21917808,0.2648402,0.25799087,0.2716895,0.33105022,0.2762557,0.3196347,0.34703195,0.4063927,0.42465752,0.3584475,0.37899545,0.42694065,0.39497718,0.44520548,0.48630136,0.44178084,0.522831,0.5068493,,0.53652966,0.510274,0.51598173,0.5456621,0.5479452,0.5559361,0.55936074,0.58447486,0.56164384,0.56392694,0.586758,0.5410959,0.6084475,0.60502285,0.60958904,0.5958904,0.59246576,0.58219177,0.6210046,0.65068495,0.65981734,0.63013697,0.65981734,0.6917808,0.65525115,0.5958904,0.7694064,0.76369864,0.82191783,0.78652966,0.6415525,0.8630137,0.716895,0.75114155,0.783105,0.86986303,0.79452056,0.7716895,0.7808219,0.8892694,0.8196347,0.8059361,0.88584477,0.8287671,0.89041096,0.9269406,0.88584477,0.89497715,0.99086756
+CatBoost (tuned),0.098173514,0.16666667,0.1563927,0.20547946,0.23287672,0.2648402,0.25342464,0.2762557,0.33561644,0.26712328,0.3173516,0.34246576,0.4086758,0.40981734,0.3607306,0.37557077,0.40296805,0.39041096,0.44863012,0.47716895,0.4520548,0.49315068,0.50570774,0.4634703,,0.48630136,0.5068493,0.54223746,0.53424656,0.5753425,0.5353881,0.5707763,0.5707763,0.53424656,0.57990867,0.5273973,0.6187215,0.59474885,0.59018266,0.5936073,0.5936073,0.58447486,0.6347032,0.6347032,0.6232877,0.6312785,0.66210043,0.6815069,0.6666667,0.5913242,0.74657536,0.7374429,0.8333333,0.7694064,0.63242006,0.84018266,0.7237443,0.74657536,0.7648402,0.86986303,0.7808219,0.76141554,0.7648402,0.86986303,0.8105023,0.79680365,0.8812785,0.82191783,0.8812785,0.91780823,0.89497715,0.9041096,0.99086756
+ModernNCA (tuned),0.08675799,0.13926941,0.13013698,0.16666667,0.17579909,0.24429224,0.19863014,0.25342464,0.34931508,0.34246576,0.3584475,0.33904108,0.39726028,0.37899545,0.37214613,0.44748858,0.41666666,0.37214613,0.4611872,0.47716895,0.51141554,0.46575344,0.49315068,0.48972604,0.51369864,,0.4977169,0.5285388,0.5525114,0.5502283,0.56164384,0.51826483,0.5525114,0.5787671,0.5684931,0.6232877,0.5673516,0.543379,0.5742009,0.63013697,0.5913242,0.5525114,0.60273975,0.60958904,0.61187214,0.673516,0.6369863,0.69634706,0.6438356,0.6210046,0.6849315,0.6940639,0.71004564,0.74657536,0.7305936,0.74657536,0.7283105,0.716895,0.7648402,0.7694064,0.7808219,0.7420091,0.7956621,0.7785388,0.8082192,0.79452056,0.8105023,0.8721461,0.84931505,0.84703195,0.8835617,0.8561644,0.977169
+TabDPT (default),0.098173514,0.15296803,0.22374429,0.15981735,0.28082192,0.26027396,0.32648402,0.3242009,0.19063927,0.39497718,0.4063927,0.41095892,0.35159817,0.38584474,0.43835616,0.4063927,0.40182647,0.44863012,0.39954337,0.47716895,0.45890412,0.47945204,0.48630136,0.48401827,0.49315068,0.5022831,,0.5273973,0.51369864,0.51369864,0.5707763,0.5296804,0.51826483,0.543379,0.5707763,0.5776256,0.5410959,0.56621003,0.5753425,0.5353881,0.5559361,0.5479452,0.5936073,0.5776256,0.57990867,0.6369863,0.5890411,0.6415525,0.5936073,0.7146119,0.72602737,0.7328767,0.6712329,0.69634706,0.7876712,0.6712329,0.6757991,0.7283105,0.7853881,0.7374429,0.7694064,0.74429226,0.80136985,0.7694064,0.79908675,0.79223746,0.826484,0.826484,0.8356164,0.84246576,0.89269406,0.8721461,0.99086756
+xRFM (tuned + ensembled),0.09589041,0.15753424,0.1324201,0.1826484,0.19634703,0.20547946,0.22146119,0.24771689,0.30593607,0.30479452,0.3127854,0.39497718,0.37100455,0.34931508,0.3652968,0.4086758,0.4155251,0.46803653,0.413242,0.4611872,0.44406393,0.4257991,0.46232876,0.4543379,0.45776257,0.47146118,0.47260273,,0.5787671,0.49885845,0.50913244,0.49543378,0.5707763,0.5810502,0.51141554,0.5273973,0.53196347,0.55707765,0.5296804,0.5913242,0.6164383,0.5730594,0.60958904,0.53881276,0.5559361,0.65753424,0.59931505,0.6232877,0.5981735,0.6461187,0.7146119,0.7351598,0.70091325,0.71004564,0.74885845,0.7374429,0.6894977,0.7694064,0.79908675,0.80365294,0.7785388,0.78995436,0.84931505,0.8242009,0.83219177,0.8230594,0.8710046,0.93378997,0.92009133,0.93378997,0.913242,0.8812785,0.98858446
+xRFM (tuned),0.11643836,0.1849315,0.17694063,0.2089041,0.21689498,0.21232876,0.23744293,0.25570777,0.28995433,0.29223743,0.33789954,0.4178082,0.3824201,0.38356164,0.38698632,0.39041096,0.43607306,0.4634703,0.4497717,0.44406393,0.4497717,0.40068492,0.44520548,0.4520548,0.46575344,0.44748858,0.48630136,0.42123288,,0.46803653,0.5171233,0.53424656,0.55136985,0.5582192,0.5251142,0.55479455,0.5148402,0.5502283,0.53881276,0.57990867,0.60730594,0.50913244,0.60273975,0.53652966,0.5753425,0.6666667,0.6232877,0.63242006,0.6278539,0.6210046,0.7191781,0.74885845,0.6369863,0.72602737,0.7648402,0.70091325,0.716895,0.79452056,0.8150685,0.75114155,0.7568493,0.8242009,0.83789957,0.78995436,0.7979452,0.84931505,0.86986303,0.89726025,0.8630137,0.8652968,0.89497715,0.89269406,0.98858446
+XGBoost (tuned + ensembled),0.06392694,0.15182649,0.12100457,0.18949772,0.21461187,0.23972602,0.23744293,0.25570777,0.3196347,0.21689498,0.26712328,0.30365297,0.39840183,0.38812786,0.3196347,0.2648402,0.3652968,0.37328768,0.4326484,0.4063927,0.37557077,0.43493152,0.4303653,0.44406393,0.42465752,0.4497717,0.48630136,0.50114155,0.53196347,,0.45890412,0.50913244,0.5045662,0.51826483,0.54680365,0.5273973,0.6347032,0.58219177,0.51826483,0.56392694,0.5753425,0.522831,0.5936073,0.5605023,0.58447486,0.5753425,0.6347032,0.6187215,0.6369863,0.59474885,0.7146119,0.7351598,0.8652968,0.7305936,0.64726025,0.89269406,0.69863015,0.739726,0.74657536,0.88013697,0.739726,0.75114155,0.7214612,0.8755708,0.7773973,0.79452056,0.84018266,0.8333333,0.8744292,0.92009133,0.8767123,0.88242006,0.9794521
+ChimeraBoost (tuned + ensembled),0.08219178,0.12557077,0.111872144,0.17579909,0.15068494,0.20091324,0.21347032,0.24086758,0.31506848,0.25456622,0.29452056,0.28538814,0.37671232,0.37214613,0.30821916,0.3938356,0.37100455,0.33333334,0.4155251,0.44292238,0.4497717,0.42465752,0.43493152,0.44063926,0.46461186,0.43835616,0.42922375,0.49086758,0.48287672,0.5410959,,0.51598173,0.55479455,0.5,0.5205479,0.48173517,0.6015982,0.5684931,0.6187215,0.56392694,0.5913242,0.5456621,0.5913242,0.7191781,0.5936073,0.5605023,0.65753424,0.6164383,0.65525115,0.5913242,0.673516,0.6849315,0.7671233,0.7374429,0.6643836,0.8002283,0.71004564,0.7146119,0.7305936,0.79908675,0.75570774,0.7283105,0.70091325,0.8082192,0.7808219,0.7671233,0.8105023,0.79680365,0.8356164,0.84474885,0.8584475,0.8630137,0.9714612
+SAP-RPT-OSS (default),0.10958904,0.17123288,0.19634703,0.19863014,0.19406393,0.19863014,0.22716895,0.23972602,0.3173516,0.38356164,0.37442923,0.369863,0.3584475,0.37671232,0.36301368,0.41095892,0.39726028,0.44520548,0.42922375,0.4520548,0.45890412,0.4497717,0.43378997,0.4155251,0.42922375,0.48173517,0.47031963,0.5045662,0.46575344,0.49086758,0.48401827,,0.47488585,0.56164384,0.55707765,0.5410959,0.5068493,0.522831,0.51598173,0.5913242,0.51141554,0.53196347,0.54680365,0.53196347,0.5308219,0.6392694,0.56392694,0.63242006,0.56392694,0.63013697,0.65525115,0.6894977,0.6255708,0.65296805,0.74657536,0.67808217,0.69863015,0.7420091,0.7374429,0.706621,0.7762557,0.75342464,0.760274,0.7203196,0.78881276,0.8082192,0.79680365,0.8310502,0.8242009,0.8310502,0.8310502,0.8789954,0.9794521
+EBM (tuned + ensembled),0.0913242,0.15525115,0.15981735,0.1803653,0.19863014,0.20547946,0.22146119,0.2283105,0.30593607,0.2716895,0.2648402,0.2762557,0.37785387,0.3675799,0.33105022,0.3458904,0.34474885,0.38812786,0.4041096,0.38584474,0.37899545,0.40182647,0.41095892,0.43835616,0.42922375,0.44748858,0.48173517,0.42922375,0.44863012,0.49543378,0.44520548,0.5251142,,0.49429223,0.49315068,0.50913244,0.51369864,0.57990867,0.49315068,0.55936074,0.6666667,0.47945204,0.6917808,0.543379,0.53652966,0.5582192,0.6438356,0.6152968,0.6415525,0.6210046,0.7226027,0.716895,0.71803653,0.739726,0.69634706,0.7420091,0.7283105,0.7694064,0.70319635,0.7808219,0.75570774,0.783105,0.7579909,0.79452056,0.7751142,0.8196347,0.86986303,0.7694064,0.8515982,0.8561644,0.8835617,0.8984018,0.977169
+FastaiMLP (tuned + ensembled),0.10273973,0.1586758,0.14383562,0.15981735,0.21004567,0.19406393,0.22260274,0.24200913,0.2625571,0.2648402,0.2522831,0.32648402,0.33561644,0.29452056,0.36187214,0.39954337,0.38698632,0.37557077,0.38584474,0.3675799,0.4303653,0.369863,0.43607306,0.43607306,0.46575344,0.42123288,0.456621,0.41894978,0.44178084,0.48173517,0.5,0.43835616,0.50570774,,0.47031963,0.5273973,0.5205479,0.49543378,0.5251142,0.5696347,0.55707765,0.5,0.56621003,0.51598173,0.51826483,0.70547944,0.6164383,0.5936073,0.6232877,0.58447486,0.65981734,0.65981734,0.6415525,0.6803653,0.7305936,0.68607306,0.70091325,0.75342464,0.80479455,0.7568493,0.7876712,0.7762557,0.93607306,0.7671233,0.8082192,0.8082192,0.8584475,0.8561644,0.83789957,0.8538813,0.93835616,0.8767123,0.9748858
+TorchMLP (tuned + ensembled),0.0456621,0.111872144,0.10045662,0.12100457,0.14611872,0.17351598,0.1780822,0.19178082,0.2716895,0.21689498,0.2716895,0.3105023,0.31849316,0.34018266,0.3173516,0.39269406,0.37214613,0.35388127,0.3652968,0.3675799,0.44292238,0.38584474,0.4063927,0.413242,0.42009133,0.43150684,0.42922375,0.4885845,0.47488585,0.45319635,0.47945204,0.44292238,0.5068493,0.5296804,,0.51826483,0.4977169,0.4885845,0.5296804,0.55936074,0.55707765,0.5205479,0.5730594,0.5410959,0.5559361,0.64497715,0.6187215,0.7477169,0.6267123,0.57990867,0.6894977,0.6872146,0.70776254,0.7431507,0.70091325,0.70319635,0.7214612,0.7294521,0.8652968,0.75114155,0.79223746,0.760274,0.8025114,0.7808219,0.826484,0.8127854,0.826484,0.8242009,0.80365294,0.8173516,0.9063927,0.8744292,0.9726027
+ModernNCA (default),0.10273973,0.14954337,0.16438356,0.14954337,0.15981735,0.19406393,0.20547946,0.2305936,0.2739726,0.29680365,0.3242009,0.30593607,0.336758,0.34474885,0.35388127,0.38584474,0.3607306,0.34931508,0.38812786,0.40182647,0.413242,0.39726028,0.4543379,0.45890412,0.47260273,0.37671232,0.42237443,0.47260273,0.44520548,0.47260273,0.51826483,0.45890412,0.49086758,0.47260273,0.48173517,,0.51598173,0.47716895,0.5456621,0.5068493,0.53196347,0.44520548,0.55479455,0.55936074,0.5148402,0.55479455,0.6187215,0.5730594,0.6278539,0.6438356,0.65068495,0.6689498,0.6164383,0.71004564,0.7146119,0.6689498,0.739726,0.7214612,0.79223746,0.72602737,0.74885845,0.7454338,0.7625571,0.760274,0.7808219,0.8173516,0.8515982,0.783105,0.83219177,0.8287671,0.8538813,0.89497715,0.98173517
+XGBoost (tuned),0.05479452,0.12785389,0.10730594,0.17123288,0.16894977,0.20091324,0.206621,0.23287672,0.30136988,0.21689498,0.2260274,0.24429224,0.34703195,0.35159817,0.29680365,0.21461187,0.34474885,0.34474885,0.41095892,0.35045663,0.31506848,0.3652968,0.38698632,0.3915525,0.38127854,0.4326484,0.45890412,0.46803653,0.4851598,0.3652968,0.39840183,0.49315068,0.48630136,0.47945204,0.5022831,0.48401827,,0.53881276,0.46575344,0.5273973,0.5456621,0.51141554,0.5502283,0.5045662,0.55479455,0.5753425,0.6187215,0.5890411,0.6244292,0.5684931,0.7089041,0.72602737,0.79680365,0.7191781,0.6347032,0.84018266,0.6940639,0.7237443,0.7191781,0.8515982,0.70547944,0.74885845,0.70776254,0.8538813,0.7648402,0.7853881,0.8413242,0.8173516,0.8721461,0.8995434,0.8721461,0.8789954,0.9748858
+TabM (default),0.06392694,0.10958904,0.12100457,0.13926941,0.15410958,0.19178082,0.20091324,0.21461187,0.30136988,0.25570777,0.25799087,0.17009133,0.37100455,0.3607306,0.34931508,0.30479452,0.3196347,0.28995433,0.42237443,0.3607306,0.34931508,0.4178082,0.38127854,0.39497718,0.40525115,0.456621,0.43378997,0.44292238,0.4497717,0.4178082,0.43150684,0.47716895,0.42009133,0.5045662,0.51141554,0.522831,0.4611872,,0.4634703,0.5251142,0.49543378,0.5022831,0.49200913,0.522831,0.5376712,0.5936073,0.65068495,0.5913242,0.65753424,0.6004566,0.6803653,0.6940639,0.6255708,0.6826484,0.6392694,0.6643836,0.7420091,0.716895,0.7420091,0.7785388,0.73630136,0.73173517,0.7283105,0.7762557,0.74657536,0.783105,0.8242009,0.75570774,0.8356164,0.8515982,0.8630137,0.89041096,0.96803653
+ChimeraBoost (tuned),0.070776254,0.12100457,0.10273973,0.15753424,0.1369863,0.1849315,0.19406393,0.20776255,0.27853882,0.22146119,0.25913242,0.23287672,0.34246576,0.33333334,0.28767124,0.33789954,0.34474885,0.3173516,0.39497718,0.39726028,0.4155251,0.4086758,0.37442923,0.39041096,0.40981734,0.4257991,0.42465752,0.47031963,0.4611872,0.48173517,0.38127854,0.48401827,0.5068493,0.47488585,0.47031963,0.4543379,0.53424656,0.53652966,,0.53424656,0.54680365,0.5045662,0.5525114,0.61187214,0.5376712,0.5376712,0.6415525,0.5753425,0.6392694,0.57990867,0.6415525,0.66210043,0.73173517,0.6917808,0.6369863,0.7420091,0.70319635,0.6894977,0.7191781,0.7694064,0.7043379,0.69634706,0.6849315,0.7716895,0.73173517,0.74657536,0.783105,0.7716895,0.826484,0.8150685,0.85273975,0.8333333,0.9589041
+RealMLP (default),0.08675799,0.11872146,0.14611872,0.13013698,0.14383562,0.17351598,0.18949772,0.20319635,0.24885845,0.22374429,0.17465754,0.23173515,0.33789954,0.34246576,0.33789954,0.33789954,0.34474885,0.28538814,0.37442923,0.30593607,0.35958904,0.33561644,0.37899545,0.4041096,0.4063927,0.369863,0.46461186,0.4086758,0.42009133,0.43607306,0.43607306,0.4086758,0.44063926,0.4303653,0.44063926,0.49315068,0.47260273,0.47488585,0.46575344,,0.5205479,0.43378997,0.5068493,0.49315068,0.5022831,0.5776256,0.6278539,0.5936073,0.6461187,0.586758,0.65753424,0.6757991,0.65981734,0.6917808,0.71803653,0.6894977,0.70319635,0.70547944,0.7751142,0.7305936,0.75570774,0.7420091,0.8150685,0.7579909,0.79223746,0.79223746,0.84474885,0.783105,0.8310502,0.8333333,0.9086758,0.8721461,0.9748858
+EBM (tuned),0.08219178,0.1369863,0.15296803,0.15525115,0.1609589,0.16438356,0.19063927,0.21232876,0.27968037,0.25799087,0.24429224,0.23287672,0.34817353,0.34474885,0.31506848,0.30365297,0.3219178,0.3196347,0.37214613,0.34018266,0.35159817,0.3675799,0.37671232,0.40753424,0.4063927,0.4086758,0.44406393,0.38356164,0.39269406,0.42465752,0.4086758,0.4885845,0.33333334,0.44292238,0.44292238,0.46803653,0.4543379,0.5045662,0.45319635,0.47945204,,0.44863012,0.6152968,0.49543378,0.49543378,0.5376712,0.6232877,0.56164384,0.6289954,0.59018266,0.66210043,0.70091325,0.67808217,0.6849315,0.66780823,0.70776254,0.70776254,0.7374429,0.6826484,0.7420091,0.7328767,0.75570774,0.7146119,0.7694064,0.75114155,0.7853881,0.8515982,0.75342464,0.84246576,0.84018266,0.86757994,0.8995434,0.977169
+ModernNCA (tuned + ensembled),0.05936073,0.0890411,0.06621005,0.13013698,0.11872146,0.21689498,0.20091324,0.21118721,0.3196347,0.2762557,0.31506848,0.29223743,0.31506848,0.34018266,0.30821916,0.37899545,0.4041096,0.34703195,0.37328768,0.4178082,0.4155251,0.39497718,0.413242,0.4178082,0.4155251,0.44748858,0.4520548,0.42694065,0.49086758,0.47716895,0.4543379,0.46803653,0.5205479,0.5,0.47945204,0.55479455,0.4885845,0.4977169,0.49543378,0.56621003,0.55136985,,0.53881276,0.4977169,0.5205479,0.55936074,0.543379,0.5707763,0.5479452,0.5,0.5890411,0.60958904,0.6689498,0.6278539,0.6004566,0.6872146,0.61415523,0.5981735,0.65525115,0.65753424,0.6689498,0.60273975,0.6940639,0.65981734,0.673516,0.65525115,0.7237443,0.81392694,0.7648402,0.7671233,0.77397263,0.7579909,0.9315069
+EBM (default),0.0890411,0.13584475,0.14383562,0.14383562,0.17579909,0.1803653,0.18721461,0.20547946,0.2625571,0.23515981,0.24315068,0.24315068,0.35388127,0.34474885,0.3196347,0.24657534,0.26940638,0.30365297,0.37214613,0.37214613,0.3127854,0.3607306,0.34474885,0.37899545,0.3652968,0.39726028,0.4063927,0.39041096,0.39726028,0.4063927,0.4086758,0.45319635,0.30821916,0.43378997,0.42694065,0.44520548,0.4497717,0.5079909,0.44748858,0.49315068,0.3847032,0.4611872,,0.45890412,0.46803653,0.53652966,0.60958904,0.56392694,0.60502285,0.5936073,0.6666667,0.65981734,0.6369863,0.70319635,0.67808217,0.6643836,0.70547944,0.74429226,0.70776254,0.75114155,0.7146119,0.76598173,0.7214612,0.7716895,0.7283105,0.8173516,0.8630137,0.7328767,0.8310502,0.82191783,0.8515982,0.9086758,0.9748858
+ChimeraBoost (default),0.061643835,0.13013698,0.098173514,0.16210045,0.14155251,0.16666667,0.17579909,0.19406393,0.2648402,0.20547946,0.25799087,0.21461187,0.3105023,0.29680365,0.28082192,0.25342464,0.3105023,0.28424656,0.37671232,0.34931508,0.3105023,0.36415526,0.32876712,0.34931508,0.3652968,0.39041096,0.42237443,0.4611872,0.4634703,0.4394977,0.28082192,0.46803653,0.456621,0.48401827,0.45890412,0.44063926,0.49543378,0.47716895,0.38812786,0.5068493,0.5045662,0.5022831,0.5410959,,0.53424656,0.5296804,0.6347032,0.55479455,0.6392694,0.5753425,0.65068495,0.65296805,0.71347034,0.69863015,0.63242006,0.727169,0.6940639,0.70319635,0.6917808,0.75342464,0.6803653,0.7191781,0.673516,0.760274,0.69634706,0.7625571,0.7773973,0.77397263,0.79908675,0.8082192,0.8105023,0.8515982,0.96575344
+iLTM (tuned),0.07305936,0.09474886,0.10730594,0.14155251,0.14155251,0.15410958,0.1803653,0.19634703,0.24086758,0.20547946,0.24200913,0.25570777,0.2762557,0.2716895,0.29908675,0.30593607,0.3561644,0.32876712,0.34474885,0.3561644,0.3675799,0.2888128,0.33561644,0.34018266,0.37671232,0.38812786,0.42009133,0.44406393,0.42465752,0.4155251,0.4063927,0.46917808,0.4634703,0.48173517,0.44406393,0.4851598,0.44520548,0.46232876,0.46232876,0.4977169,0.5045662,0.47945204,0.53196347,0.46575344,,0.5525114,0.6152968,0.5502283,0.60730594,0.5045662,0.61415523,0.6347032,0.60958904,0.663242,0.6244292,0.65068495,0.67808217,0.6335617,0.75570774,0.71004564,0.6826484,0.6689498,0.75342464,0.7237443,0.7305936,0.7283105,0.783105,0.8059361,0.7694064,0.79223746,0.8333333,0.82990867,0.9589041
+FastaiMLP (tuned),0.07305936,0.12671232,0.13470319,0.11872146,0.15753424,0.14611872,0.15296803,0.16894977,0.20547946,0.22146119,0.22716895,0.2648402,0.2648402,0.25570777,0.2888128,0.3127854,0.29680365,0.29680365,0.30593607,0.28995433,0.37899545,0.31164384,0.35159817,0.369863,0.36872146,0.32648402,0.36301368,0.34246576,0.33333334,0.42465752,0.4394977,0.3607306,0.44178084,0.29452056,0.35502282,0.44520548,0.42465752,0.4063927,0.46232876,0.42237443,0.46232876,0.44063926,0.4634703,0.47031963,0.44748858,,0.5525114,0.51598173,0.5456621,0.5068493,0.5936073,0.5958904,0.57990867,0.5890411,0.6666667,0.60273975,0.65981734,0.6689498,0.69063926,0.6609589,0.70547944,0.6974886,0.7853881,0.67808217,0.74429226,0.74885845,0.783105,0.74885845,0.7671233,0.7876712,0.8607306,0.84018266,0.96575344
+TabSTAR (tuned),0.1803653,0.21575342,0.25342464,0.20776255,0.25570777,0.22146119,0.25342464,0.2625571,0.29223743,0.29680365,0.30593607,0.29452056,0.34474885,0.29223743,0.36187214,0.31849316,0.2739726,0.3127854,0.39611873,0.32876712,0.33561644,0.3652968,0.32762557,0.34018266,0.33789954,0.36301368,0.41095892,0.40068492,0.37671232,0.3652968,0.34246576,0.43607306,0.3561644,0.38356164,0.38127854,0.38127854,0.38127854,0.34931508,0.3584475,0.37214613,0.37671232,0.456621,0.39041096,0.3652968,0.3847032,0.44748858,,0.4041096,0.60958904,0.5296804,0.5684931,0.5673516,0.48173517,0.4155251,0.6164383,0.4885845,0.9292237,0.61187214,0.51598173,0.60502285,0.6369863,0.6347032,0.6415525,0.6347032,0.6438356,0.6917808,0.75570774,0.71004564,0.783105,0.7785388,0.70319635,0.8515982,0.9315069
+TorchMLP (tuned),0.03196347,0.06621005,0.07990868,0.08219178,0.08675799,0.10273973,0.12100457,0.12328767,0.20547946,0.17123288,0.21232876,0.21347032,0.2625571,0.28082192,0.26027396,0.2739726,0.28767124,0.26598173,0.30365297,0.2762557,0.3344749,0.29452056,0.29908675,0.30821916,0.31849316,0.30365297,0.3584475,0.37671232,0.3675799,0.38127854,0.38356164,0.3675799,0.3847032,0.4063927,0.2522831,0.42694065,0.41095892,0.4086758,0.42465752,0.4063927,0.43835616,0.42922375,0.43607306,0.44520548,0.4497717,0.48401827,0.5958904,,0.6004566,0.522831,0.6130137,0.60730594,0.56621003,0.65068495,0.6803653,0.59474885,0.6894977,0.65981734,0.76826483,0.6666667,0.69634706,0.6746575,0.70319635,0.6917808,0.7305936,0.7625571,0.772831,0.760274,0.75342464,0.7648402,0.84703195,0.8105023,0.96118724
+TabSTAR (tuned + ensembled),0.1780822,0.20319635,0.25114155,0.21004567,0.24429224,0.21803653,0.24657534,0.25114155,0.28310502,0.28767124,0.29452056,0.28995433,0.34246576,0.28082192,0.36301368,0.31506848,0.26712328,0.30936074,0.39497718,0.3173516,0.33105022,0.36187214,0.3219178,0.34474885,0.33333334,0.3561644,0.4063927,0.40182647,0.37214613,0.36301368,0.34474885,0.43607306,0.3584475,0.37671232,0.37328768,0.37214613,0.37557077,0.34246576,0.3607306,0.35388127,0.37100455,0.4520548,0.39497718,0.3607306,0.39269406,0.4543379,0.39041096,0.39954337,,0.5205479,0.5479452,0.5502283,0.47031963,0.4178082,0.60730594,0.48630136,0.9474886,0.60730594,0.51369864,0.5958904,0.6347032,0.6347032,0.6392694,0.6255708,0.6392694,0.6894977,0.74429226,0.70319635,0.7762557,0.77397263,0.70091325,0.84018266,0.9315069
+OrionMSP (default),0.057077624,0.098173514,0.09474886,0.07534247,0.14840183,0.08675799,0.12328767,0.12785389,0.15981735,0.30251142,0.28995433,0.30593607,0.21689498,0.15525115,0.27283105,0.3242009,0.30593607,0.33333334,0.22146119,0.36643836,0.34246576,0.37328768,0.39954337,0.4041096,0.4086758,0.37899545,0.28538814,0.35388127,0.37899545,0.40525115,0.4086758,0.369863,0.37899545,0.4155251,0.42009133,0.3561644,0.43150684,0.39954337,0.42009133,0.413242,0.40981734,0.5,0.4063927,0.42465752,0.49543378,0.49315068,0.47031963,0.47716895,0.47945204,,0.4977169,0.5022831,0.5205479,0.5296804,0.6187215,0.56392694,0.60730594,0.6278539,0.58219177,0.55479455,0.6164383,0.6563927,0.65525115,0.5890411,0.63242006,0.6917808,0.66210043,0.7420091,0.72602737,0.69863015,0.79223746,0.7146119,0.96118724
+ExtraTrees (tuned + ensembled),0.047945205,0.061643835,0.07534247,0.08789954,0.08675799,0.10730594,0.11643836,0.13127854,0.15410958,0.20091324,0.16210045,0.1826484,0.20547946,0.19863014,0.24200913,0.19292237,0.20776255,0.24885845,0.26027396,0.25799087,0.24315068,0.27054796,0.2283105,0.2305936,0.25342464,0.31506848,0.2739726,0.28538814,0.28082192,0.28538814,0.32648402,0.34474885,0.27739727,0.34018266,0.3105023,0.34931508,0.29109588,0.3196347,0.3584475,0.34246576,0.33789954,0.41095892,0.33333334,0.34931508,0.38584474,0.4063927,0.43150684,0.38698632,0.4520548,0.5022831,,0.63242006,0.44063926,0.49315068,0.5707763,0.4885845,0.6221461,0.59931505,0.55936074,0.6586758,0.5776256,0.6289954,0.6004566,0.70091325,0.5981735,0.70319635,0.7853881,0.70776254,0.8721461,0.83789957,0.72602737,0.84246576,0.98858446
+ExtraTrees (tuned),0.057077624,0.06621005,0.0913242,0.0890411,0.09474886,0.10273973,0.12328767,0.12785389,0.12100457,0.19178082,0.17237443,0.20091324,0.19863014,0.20319635,0.24429224,0.21004567,0.20776255,0.24200913,0.23972602,0.24657534,0.23401827,0.25342464,0.23515981,0.23630136,0.2625571,0.30593607,0.26712328,0.2648402,0.25114155,0.2648402,0.31506848,0.3105023,0.28310502,0.34018266,0.3127854,0.33105022,0.2739726,0.30593607,0.33789954,0.3242009,0.29908675,0.39041096,0.34018266,0.34703195,0.3652968,0.4041096,0.4326484,0.39269406,0.4497717,0.4977169,0.3675799,,0.41095892,0.49086758,0.5707763,0.43835616,0.6164383,0.5913242,0.543379,0.60273975,0.5456621,0.6164383,0.5730594,0.6461187,0.58447486,0.69634706,0.7374429,0.6803653,0.8059361,0.8002283,0.70547944,0.84931505,0.98630136
+XGBoost (default),0.029680366,0.05936073,0.06621005,0.08675799,0.06392694,0.111872144,0.111872144,0.11872146,0.1826484,0.10273973,0.12785389,0.14611872,0.21917808,0.23972602,0.22374429,0.13013698,0.23515981,0.19178082,0.28767124,0.22374429,0.18150684,0.18949772,0.1826484,0.1780822,0.16666667,0.28995433,0.32876712,0.29908675,0.36301368,0.13470319,0.23287672,0.37442923,0.28196347,0.3584475,0.29223743,0.38356164,0.20319635,0.37442923,0.26826483,0.34018266,0.3219178,0.33105022,0.36301368,0.2865297,0.39041096,0.42009133,0.51826483,0.43378997,0.5296804,0.47945204,0.55936074,0.5890411,,0.5730594,0.5456621,0.50114155,0.6187215,0.6004566,0.5981735,0.6278539,0.51826483,0.61415523,0.586758,0.65753424,0.5753425,0.6757991,0.716895,0.7191781,0.75570774,0.7716895,0.70776254,0.7853881,0.93835616
+iLTM (default),0.05936073,0.070776254,0.07762557,0.0913242,0.10045662,0.10730594,0.11872146,0.13013698,0.20091324,0.15525115,0.14383562,0.15068494,0.19863014,0.22146119,0.2305936,0.15068494,0.18721461,0.1849315,0.2648402,0.20776255,0.17922375,0.21004567,0.16894977,0.21347032,0.2305936,0.25342464,0.30365297,0.28995433,0.2739726,0.26940638,0.2625571,0.34703195,0.26027396,0.3196347,0.25684932,0.28995433,0.28082192,0.3173516,0.30821916,0.30821916,0.31506848,0.37214613,0.29680365,0.30136988,0.336758,0.41095892,0.58447486,0.34931508,0.58219177,0.47031963,0.5068493,0.50913244,0.42694065,,0.543379,0.4634703,0.65296805,0.5981735,0.5479452,0.61415523,0.51826483,0.63013697,0.58447486,0.6347032,0.53196347,0.6940639,0.7579909,0.66210043,0.69863015,0.7305936,0.6415525,0.82191783,0.93607306
+TabFlex (default),0.07305936,0.09589041,0.11872146,0.10273973,0.12100457,0.097031966,0.12557077,0.13926941,0.12785389,0.21232876,0.23515981,0.23287672,0.20319635,0.111872144,0.28538814,0.29223743,0.23858447,0.24885845,0.22146119,0.25913242,0.30821916,0.2648402,0.34246576,0.3584475,0.3675799,0.26940638,0.21232876,0.25114155,0.23515981,0.35273972,0.33561644,0.25342464,0.30365297,0.26940638,0.29908675,0.28538814,0.3652968,0.3607306,0.36301368,0.28196347,0.3321918,0.39954337,0.3219178,0.3675799,0.37557077,0.33333334,0.38356164,0.3196347,0.39269406,0.38127854,0.42922375,0.42922375,0.4543379,0.456621,,0.47031963,0.5296804,0.5251142,0.4783105,0.5022831,0.56392694,0.5650685,0.5456621,0.51369864,0.60273975,0.61415523,0.61187214,0.63242006,0.6232877,0.65296805,0.69863015,0.6940639,0.89269406
+LightGBM (default),0.02739726,0.0456621,0.052511416,0.05936073,0.052511416,0.05936073,0.06621005,0.07762557,0.13926941,0.07534247,0.098173514,0.09474886,0.16894977,0.21118721,0.20319635,0.05022831,0.18150684,0.16438356,0.24885845,0.18949772,0.08561644,0.14155251,0.14840183,0.1369863,0.15981735,0.25342464,0.32876712,0.2625571,0.29908675,0.10730594,0.19977169,0.3219178,0.25799087,0.31392694,0.29680365,0.33105022,0.15981735,0.33561644,0.25799087,0.3105023,0.29223743,0.3127854,0.33561644,0.27283105,0.34931508,0.39726028,0.51141554,0.40525115,0.51369864,0.43607306,0.51141554,0.56164384,0.49885845,0.53652966,0.5296804,,0.60730594,0.5776256,0.586758,0.5913242,0.51826483,0.5890411,0.5707763,0.61986303,0.586758,0.66210043,0.6917808,0.72602737,0.75114155,0.76826483,0.7328767,0.79452056,0.95776254
+TabSTAR (default),0.14611872,0.16894977,0.19863014,0.16438356,0.21004567,0.18721461,0.21004567,0.20776255,0.21004567,0.23972602,0.25114155,0.24657534,0.23972602,0.21232876,0.2956621,0.23972602,0.21118721,0.26712328,0.2625571,0.2648402,0.25799087,0.28082192,0.2716895,0.28310502,0.2762557,0.2716895,0.3242009,0.3105023,0.28310502,0.30136988,0.28995433,0.30136988,0.2716895,0.29908675,0.27853882,0.26027396,0.30593607,0.25799087,0.29680365,0.29680365,0.29223743,0.38584474,0.29452056,0.30593607,0.3219178,0.34018266,0.070776254,0.3105023,0.052511416,0.39269406,0.37785387,0.38356164,0.38127854,0.34703195,0.47031963,0.39269406,,0.4885845,0.39041096,0.4257991,0.5410959,0.5022831,0.4543379,0.4497717,0.5456621,0.56164384,0.60502285,0.6255708,0.6187215,0.5936073,0.6232877,0.72602737,0.9223744
+Linear (tuned + ensembled),0.07534247,0.12100457,0.12557077,0.10502283,0.1324201,0.14611872,0.14611872,0.16666667,0.19178082,0.16324201,0.17351598,0.16666667,0.24885845,0.21004567,0.2283105,0.1826484,0.16894977,0.19863014,0.25913242,0.24429224,0.21689498,0.28082192,0.24429224,0.24885845,0.25342464,0.28310502,0.2716895,0.2305936,0.20547946,0.26027396,0.28538814,0.25799087,0.2305936,0.24657534,0.27054796,0.27853882,0.2762557,0.28310502,0.3105023,0.29452056,0.2625571,0.40182647,0.25570777,0.29680365,0.36643836,0.33105022,0.38812786,0.34018266,0.39269406,0.37214613,0.40068492,0.4086758,0.39954337,0.40182647,0.47488585,0.42237443,0.51141554,,0.44292238,0.43150684,0.47945204,0.66780823,0.49315068,0.45890412,0.5022831,0.7579909,0.5753425,0.5981735,0.5936073,0.56621003,0.6210046,0.67808217,0.89269406
+TorchMLP (default),0.02283105,0.052511416,0.052511416,0.043378994,0.05022831,0.038812786,0.07534247,0.07762557,0.12100457,0.11415525,0.10730594,0.10958904,0.15981735,0.13926941,0.21917808,0.18949772,0.18949772,0.12557077,0.18949772,0.15525115,0.20776255,0.1826484,0.20091324,0.21689498,0.23515981,0.23515981,0.21461187,0.20091324,0.1849315,0.25342464,0.26940638,0.2625571,0.29680365,0.19520548,0.13470319,0.20776255,0.28082192,0.25799087,0.28082192,0.22488585,0.3173516,0.34474885,0.29223743,0.30821916,0.24429224,0.30936074,0.48401827,0.23173515,0.48630136,0.4178082,0.44063926,0.456621,0.40182647,0.4520548,0.5216895,0.413242,0.60958904,0.55707765,,0.49086758,0.5456621,0.57990867,0.5776256,0.5479452,0.58447486,0.65981734,0.69863015,0.65296805,0.6643836,0.6643836,0.72602737,0.7785388,0.94063926
+RandomForest (tuned + ensembled),0.02283105,0.061643835,0.057077624,0.061643835,0.06392694,0.10273973,0.08675799,0.111872144,0.12328767,0.10730594,0.0890411,0.13470319,0.14383562,0.19406393,0.19634703,0.10045662,0.15753424,0.16666667,0.21461187,0.1780822,0.15981735,0.15981735,0.12328767,0.13013698,0.13013698,0.2305936,0.2625571,0.19634703,0.24885845,0.11986301,0.20091324,0.293379,0.21917808,0.24315068,0.24885845,0.2739726,0.14840183,0.22146119,0.2305936,0.26940638,0.25799087,0.34246576,0.24885845,0.24657534,0.28995433,0.33904108,0.39497718,0.33333334,0.4041096,0.44520548,0.3413242,0.39726028,0.37214613,0.38584474,0.4977169,0.4086758,0.5742009,0.5684931,0.50913244,,0.51598173,0.58447486,0.53196347,0.63242006,0.5410959,0.66552514,0.70091325,0.6849315,0.75114155,0.8173516,0.6278539,0.8150685,0.96347034
+PerpetualBooster (tuned + ensembled),0.01826484,0.06392694,0.06392694,0.057077624,0.06849315,0.06849315,0.06849315,0.08789954,0.11872146,0.08219178,0.08675799,0.111872144,0.14155251,0.13470319,0.1780822,0.12557077,0.17123288,0.11643836,0.16210045,0.15296803,0.14954337,0.16438356,0.21461187,0.20547946,0.21917808,0.21917808,0.2305936,0.22146119,0.24315068,0.26027396,0.24429224,0.22374429,0.24429224,0.21232876,0.20776255,0.25114155,0.29452056,0.26369864,0.2956621,0.24429224,0.26712328,0.33105022,0.28538814,0.3196347,0.3173516,0.29452056,0.36301368,0.30365297,0.3652968,0.38356164,0.42237443,0.4543379,0.48173517,0.48173517,0.43607306,0.48173517,0.45890412,0.5205479,0.4543379,0.48401827,,0.53196347,0.44748858,0.50913244,0.68607306,0.5890411,0.6130137,0.6415525,0.6643836,0.6826484,0.79223746,0.70319635,0.94292235
+Linear (tuned),0.06621005,0.10958904,0.098173514,0.0913242,0.10730594,0.13470319,0.12557077,0.14155251,0.17123288,0.13584475,0.16894977,0.15182649,0.2283105,0.1849315,0.20319635,0.16894977,0.14383562,0.1826484,0.2283105,0.23401827,0.19292237,0.25570777,0.22374429,0.2283105,0.23858447,0.25799087,0.25570777,0.21004567,0.17579909,0.24885845,0.2716895,0.24657534,0.21689498,0.22374429,0.23972602,0.25456622,0.25114155,0.26826483,0.30365297,0.25799087,0.24429224,0.39726028,0.23401827,0.28082192,0.33105022,0.30251142,0.3652968,0.32534248,0.3652968,0.3436073,0.37100455,0.38356164,0.38584474,0.369863,0.43493152,0.41095892,0.4977169,0.3321918,0.42009133,0.4155251,0.46803653,,0.47031963,0.43378997,0.4885845,0.6757991,0.5262557,0.5890411,0.5684931,0.5456621,0.6187215,0.65981734,0.8835617
+FastaiMLP (default),0.043378994,0.05936073,0.061643835,0.06849315,0.0890411,0.0913242,0.08675799,0.10045662,0.09360731,0.0913242,0.0890411,0.12785389,0.12785389,0.15296803,0.17123288,0.21461187,0.20776255,0.17579909,0.15525115,0.124429226,0.23515981,0.1849315,0.21232876,0.21917808,0.23515981,0.2043379,0.19863014,0.15068494,0.16210045,0.27853882,0.29908675,0.23972602,0.24200913,0.06392694,0.19748859,0.23744293,0.29223743,0.2716895,0.31506848,0.1849315,0.28538814,0.30593607,0.27853882,0.32648402,0.24657534,0.21461187,0.3584475,0.29680365,0.3607306,0.34474885,0.39954337,0.42694065,0.413242,0.4155251,0.4543379,0.42922375,0.5456621,0.5068493,0.42237443,0.46803653,0.5525114,0.5296804,,0.47488585,0.57990867,0.56621003,0.6415525,0.6232877,0.6278539,0.6130137,0.7146119,0.739726,0.94292235
+RandomForest (tuned),0.025114154,0.038812786,0.05479452,0.06392694,0.0456621,0.06621005,0.07534247,0.08675799,0.10958904,0.097031966,0.070776254,0.11415525,0.14383562,0.17579909,0.17579909,0.07762557,0.15753424,0.16894977,0.1803653,0.16210045,0.12557077,0.13926941,0.11872146,0.110730596,0.13013698,0.22146119,0.2305936,0.17579909,0.21004567,0.124429226,0.19178082,0.27968037,0.20547946,0.23287672,0.21917808,0.23972602,0.14611872,0.22374429,0.2283105,0.24200913,0.2305936,0.34018266,0.2283105,0.23972602,0.2762557,0.3219178,0.3652968,0.30821916,0.37442923,0.41095892,0.29908675,0.35388127,0.34246576,0.3652968,0.48630136,0.380137,0.5502283,0.5410959,0.4520548,0.3675799,0.49086758,0.56621003,0.5251142,,0.5273973,0.65296805,0.6769406,0.6643836,0.7146119,0.73630136,0.61187214,0.8082192,0.96575344
+PerpetualBooster (tuned),0.015981736,0.047945205,0.049086757,0.047945205,0.06849315,0.05479452,0.061643835,0.07534247,0.09589041,0.07990868,0.07534247,0.0890411,0.12671232,0.13470319,0.15525115,0.10273973,0.15525115,0.10273973,0.15296803,0.14840183,0.12785389,0.14383562,0.17579909,0.1803653,0.18949772,0.19178082,0.20091324,0.16780822,0.2020548,0.22260274,0.21917808,0.21118721,0.22488585,0.19178082,0.17351598,0.21917808,0.23515981,0.25342464,0.26826483,0.20776255,0.24885845,0.32648402,0.2716895,0.30365297,0.26940638,0.25570777,0.3561644,0.26940638,0.3607306,0.3675799,0.40182647,0.4155251,0.42465752,0.46803653,0.39726028,0.413242,0.4543379,0.4977169,0.4155251,0.45890412,0.31392694,0.51141554,0.42009133,0.47260273,,0.56621003,0.57990867,0.6221461,0.6415525,0.65753424,0.73401827,0.6643836,0.91552514
+Linear (default),0.0456621,0.06849315,0.07420091,0.06278539,0.07990868,0.070776254,0.09474886,0.09018265,0.09360731,0.10730594,0.14840183,0.12100457,0.16210045,0.14269407,0.19634703,0.11872146,0.0890411,0.15068494,0.1803653,0.1849315,0.14155251,0.21917808,0.17579909,0.19406393,0.20319635,0.20547946,0.20776255,0.17694063,0.15068494,0.20547946,0.23287672,0.19178082,0.1803653,0.19178082,0.18721461,0.1826484,0.21461187,0.21689498,0.25342464,0.20776255,0.21461187,0.34474885,0.1826484,0.23744293,0.2716895,0.25114155,0.30821916,0.23744293,0.3105023,0.30821916,0.29680365,0.30365297,0.3242009,0.30593607,0.38584474,0.33789954,0.43835616,0.24200913,0.34018266,0.3344749,0.41095892,0.3242009,0.43378997,0.34703195,0.43378997,,0.44292238,0.56392694,0.5068493,0.51141554,0.5559361,0.5936073,0.8333333
+KNN (tuned + ensembled),0.02739726,0.061643835,0.07305936,0.052511416,0.07305936,0.0913242,0.0890411,0.09589041,0.09589041,0.08675799,0.09589041,0.11872146,0.13356164,0.14840183,0.19178082,0.10045662,0.0890411,0.14611872,0.14383562,0.15525115,0.1369863,0.15753424,0.11415525,0.11415525,0.11872146,0.18949772,0.17351598,0.12899543,0.13013698,0.15981735,0.18949772,0.20319635,0.13013698,0.14155251,0.17351598,0.14840183,0.1586758,0.17579909,0.21689498,0.15525115,0.14840183,0.2762557,0.1369863,0.22260274,0.21689498,0.21689498,0.24429224,0.22716895,0.25570777,0.33789954,0.21461187,0.2625571,0.28310502,0.24200913,0.38812786,0.30821916,0.39497718,0.42465752,0.30136988,0.29908675,0.38698632,0.4737443,0.3584475,0.32305935,0.42009133,0.55707765,,0.58447486,0.49086758,0.5251142,0.5479452,0.8458904,0.96803653
+xRFM (default),0.025114154,0.0456621,0.057077624,0.047945205,0.061643835,0.057077624,0.07305936,0.07990868,0.070776254,0.070776254,0.07534247,0.11415525,0.083333336,0.10958904,0.111872144,0.15296803,0.1849315,0.15753424,0.11872146,0.1324201,0.17465754,0.0890411,0.16894977,0.17123288,0.1780822,0.12785389,0.17351598,0.06621005,0.10273973,0.16666667,0.20319635,0.16894977,0.2305936,0.14383562,0.17579909,0.21689498,0.1826484,0.24429224,0.2283105,0.21689498,0.24657534,0.18607306,0.26712328,0.2260274,0.19406393,0.25114155,0.28995433,0.23972602,0.29680365,0.25799087,0.29223743,0.3196347,0.28082192,0.33789954,0.3675799,0.2739726,0.37442923,0.40182647,0.34703195,0.31506848,0.3584475,0.41095892,0.37671232,0.33561644,0.37785387,0.43607306,0.4155251,,0.4497717,0.413242,0.51369864,0.522831,0.8630137
+ExtraTrees (default),0.015981736,0.029680366,0.05022831,0.047945205,0.05479452,0.05479452,0.057077624,0.06621005,0.069634706,0.06392694,0.061643835,0.07762557,0.07305936,0.11415525,0.12100457,0.07762557,0.1369863,0.11643836,0.11643836,0.15068494,0.09360731,0.12100457,0.12557077,0.10958904,0.11872146,0.15068494,0.16438356,0.07990868,0.1369863,0.12557077,0.16438356,0.17579909,0.14840183,0.16210045,0.19634703,0.16780822,0.12785389,0.16438356,0.17351598,0.16894977,0.15753424,0.23515981,0.16894977,0.20091324,0.2305936,0.23287672,0.21689498,0.24657534,0.22374429,0.2739726,0.12785389,0.19406393,0.24429224,0.30136988,0.37671232,0.24885845,0.38127854,0.4063927,0.33561644,0.24885845,0.33561644,0.43150684,0.37214613,0.28538814,0.3584475,0.49315068,0.50913244,0.5502283,,0.522831,0.4977169,0.6415525,0.956621
+RandomForest (default),0.02283105,0.025114154,0.047945205,0.043378994,0.038812786,0.043378994,0.05479452,0.05936073,0.06392694,0.06392694,0.038812786,0.07762557,0.07534247,0.10502283,0.10273973,0.047945205,0.10730594,0.10730594,0.10273973,0.11872146,0.07534247,0.09589041,0.10045662,0.07305936,0.08219178,0.15296803,0.15753424,0.06621005,0.13470319,0.07990868,0.15525115,0.16894977,0.14383562,0.14611872,0.1826484,0.17123288,0.10045662,0.14840183,0.1849315,0.16666667,0.15981735,0.23287672,0.1780822,0.19178082,0.20776255,0.21232876,0.22146119,0.23515981,0.2260274,0.30136988,0.16210045,0.19977169,0.2283105,0.26940638,0.34703195,0.23173515,0.4063927,0.43378997,0.33561644,0.1826484,0.3173516,0.4543379,0.38698632,0.26369864,0.34246576,0.4885845,0.47488585,0.586758,0.47716895,,0.5045662,0.6347032,0.94063926
+PerpetualBooster (default),0.011415525,0.029680366,0.03652968,0.020547945,0.038812786,0.029680366,0.038812786,0.03652968,0.034246575,0.05479452,0.034246575,0.034246575,0.06392694,0.07534247,0.10502283,0.05479452,0.07990868,0.043378994,0.06621005,0.052511416,0.057077624,0.0913242,0.10730594,0.11415525,0.10502283,0.11643836,0.10730594,0.08675799,0.10502283,0.12328767,0.14155251,0.16894977,0.11643836,0.061643835,0.09360731,0.14611872,0.12785389,0.1369863,0.14726028,0.0913242,0.1324201,0.2260274,0.14840183,0.18949772,0.16666667,0.13926941,0.29680365,0.15296803,0.29908675,0.20776255,0.2739726,0.29452056,0.29223743,0.3584475,0.30136988,0.26712328,0.37671232,0.37899545,0.2739726,0.37214613,0.20776255,0.38127854,0.28538814,0.38812786,0.26598173,0.44406393,0.4520548,0.48630136,0.5022831,0.49543378,,0.5525114,0.89269406
+KNN (tuned),0.02739726,0.03652968,0.05479452,0.04109589,0.052511416,0.05479452,0.06849315,0.06621005,0.047945205,0.08219178,0.07534247,0.0913242,0.0913242,0.061643835,0.14840183,0.07876712,0.05022831,0.11643836,0.10273973,0.12785389,0.0913242,0.11986301,0.08675799,0.10502283,0.09589041,0.14383562,0.12785389,0.11872146,0.10730594,0.11757991,0.1369863,0.12100457,0.10159817,0.12328767,0.12557077,0.10502283,0.12100457,0.10958904,0.16666667,0.12785389,0.10045662,0.24200913,0.0913242,0.14840183,0.17009133,0.15981735,0.14840183,0.18949772,0.15981735,0.28538814,0.15753424,0.15068494,0.21461187,0.1780822,0.30593607,0.20547946,0.2739726,0.3219178,0.22146119,0.1849315,0.29680365,0.34018266,0.26027396,0.19178082,0.33561644,0.4063927,0.15410958,0.47716895,0.3584475,0.3652968,0.44748858,,0.9497717
+KNN (default),0.006849315,0.00456621,0.006849315,0.00456621,0.00913242,0.006849315,0.00913242,0.00913242,0.006849315,0.00456621,0.006849315,0.01826484,0.006849315,0.006849315,0.034246575,0.00913242,0.015981736,0.02739726,0.011415525,0.020547945,0.006849315,0.025114154,0.01369863,0.00913242,0.00913242,0.02283105,0.00913242,0.011415525,0.011415525,0.020547945,0.028538812,0.020547945,0.02283105,0.025114154,0.02739726,0.01826484,0.025114154,0.03196347,0.04109589,0.025114154,0.02283105,0.06849315,0.025114154,0.034246575,0.04109589,0.034246575,0.06849315,0.038812786,0.06849315,0.038812786,0.011415525,0.01369863,0.061643835,0.06392694,0.10730594,0.042237442,0.07762557,0.10730594,0.05936073,0.03652968,0.057077624,0.11643836,0.057077624,0.034246575,0.084474884,0.16666667,0.03196347,0.1369863,0.043378994,0.05936073,0.10730594,0.05022831,
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
index 02221d2ee68ae0f310c57d8f18b53a9dced2a0a1..cb6e12417956af7de49739a46354b5b6dfb83d9a 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Binary | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_all/tasks_binary/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..5bbc3021dff87998c8a46b9ee8ee2fe83dd6dedf
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_binary/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2061.0,"score":0.714,"rank":5.23,"hrank":2.76,"imp":5.398,"train_time":128.64,"predict_time":2.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+171\/-60","imputed":false},{"position":1,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2055.0,"score":0.878,"rank":5.36,"hrank":1.55,"imp":3.02,"train_time":42.17,"predict_time":19.818,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+380\/-164","imputed":false},{"position":2,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1904.0,"score":0.616,"rank":9.74,"hrank":3.48,"imp":6.355,"train_time":325.92,"predict_time":2.156,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+184\/-74","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1869.0,"score":0.591,"rank":11.12,"hrank":3.99,"imp":8.448,"train_time":1.18,"predict_time":0.306,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+263\/-167","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1787.0,"score":0.456,"rank":14.86,"hrank":9.5,"imp":9.828,"train_time":2.76,"predict_time":0.886,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-92","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1782.0,"score":0.481,"rank":15.12,"hrank":4.82,"imp":8.27,"train_time":0.65,"predict_time":0.198,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+178\/-112","imputed":false},{"position":6,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1761.0,"score":0.49,"rank":16.19,"hrank":7.23,"imp":8.289,"train_time":1.38,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+140\/-84","imputed":false},{"position":7,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1760.0,"score":0.446,"rank":16.25,"hrank":12.41,"imp":9.784,"train_time":1883.82,"predict_time":80.768,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+114\/-62","imputed":false},{"position":8,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1755.0,"score":0.47,"rank":16.54,"hrank":7.94,"imp":8.625,"train_time":329.05,"predict_time":0.166,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+158\/-91","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1745.0,"score":0.416,"rank":17.05,"hrank":11.77,"imp":9.151,"train_time":1633.8,"predict_time":2.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-49","imputed":false},{"position":10,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1740.0,"score":0.461,"rank":17.33,"hrank":7.91,"imp":8.922,"train_time":329.05,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+173\/-106","imputed":false},{"position":11,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1720.0,"score":0.426,"rank":18.44,"hrank":10.47,"imp":10.523,"train_time":95.92,"predict_time":0.332,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+174\/-133","imputed":false},{"position":12,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1715.0,"score":0.342,"rank":18.73,"hrank":10.42,"imp":10.352,"train_time":583.55,"predict_time":0.579,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+175\/-109","imputed":false},{"position":13,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1712.0,"score":0.381,"rank":18.91,"hrank":12.0,"imp":10.95,"train_time":110.13,"predict_time":0.386,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-88","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1698.0,"score":0.36,"rank":19.77,"hrank":13.22,"imp":11.42,"train_time":86.85,"predict_time":0.668,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+155\/-77","imputed":false},{"position":15,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1675.0,"score":0.351,"rank":21.1,"hrank":13.51,"imp":10.709,"train_time":2139.17,"predict_time":6.41,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-132","imputed":false},{"position":16,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1655.0,"score":0.358,"rank":22.36,"hrank":12.13,"imp":11.186,"train_time":95.92,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+158\/-106","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1651.0,"score":0.305,"rank":22.63,"hrank":13.23,"imp":10.511,"train_time":1633.8,"predict_time":0.088,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-66","imputed":false},{"position":18,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1650.0,"score":0.297,"rank":22.64,"hrank":12.56,"imp":11.894,"train_time":842.25,"predict_time":15.118,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+163\/-100","imputed":false},{"position":19,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1640.0,"score":0.291,"rank":23.3,"hrank":18.18,"imp":11.775,"train_time":110.13,"predict_time":0.061,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-73","imputed":false},{"position":20,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1636.0,"score":0.279,"rank":23.59,"hrank":12.45,"imp":11.248,"train_time":583.55,"predict_time":0.056,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+168\/-112","imputed":false},{"position":21,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1617.0,"score":0.292,"rank":24.78,"hrank":15.21,"imp":11.861,"train_time":0.48,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+163\/-120","imputed":false},{"position":22,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1609.0,"score":0.262,"rank":25.32,"hrank":13.71,"imp":12.444,"train_time":842.25,"predict_time":1.465,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+167\/-106","imputed":false},{"position":23,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1590.0,"score":0.257,"rank":26.57,"hrank":16.26,"imp":12.197,"train_time":2.3,"predict_time":0.896,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+133\/-113","imputed":false},{"position":24,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1583.0,"score":0.206,"rank":26.99,"hrank":20.04,"imp":13.106,"train_time":86.85,"predict_time":0.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+141\/-78","imputed":false},{"position":25,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1572.0,"score":0.246,"rank":27.72,"hrank":15.39,"imp":12.869,"train_time":8.53,"predict_time":3.578,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-147","imputed":false},{"position":26,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1568.0,"score":0.217,"rank":28.01,"hrank":15.13,"imp":12.775,"train_time":1883.82,"predict_time":10.881,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-109","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1557.0,"score":0.171,"rank":28.72,"hrank":21.09,"imp":13.195,"train_time":2139.17,"predict_time":0.349,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-72","imputed":false},{"position":28,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1539.0,"score":0.191,"rank":30.0,"hrank":17.42,"imp":12.773,"train_time":4860.88,"predict_time":475.78,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-141","imputed":false},{"position":29,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1538.0,"score":0.186,"rank":30.05,"hrank":22.88,"imp":12.583,"train_time":1156.69,"predict_time":1.558,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-107","imputed":false},{"position":30,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1530.0,"score":0.226,"rank":30.57,"hrank":13.99,"imp":14.371,"train_time":0.76,"predict_time":0.203,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+196\/-140","imputed":false},{"position":31,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1486.0,"score":0.189,"rank":33.67,"hrank":23.26,"imp":13.749,"train_time":572.82,"predict_time":1.935,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-125","imputed":false},{"position":32,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1482.0,"score":0.146,"rank":33.91,"hrank":27.23,"imp":15.71,"train_time":188.8,"predict_time":0.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+132\/-110","imputed":false},{"position":33,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1473.0,"score":0.154,"rank":34.56,"hrank":19.67,"imp":14.168,"train_time":0.49,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-129","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1470.0,"score":0.141,"rank":34.74,"hrank":26.92,"imp":13.804,"train_time":1.5,"predict_time":0.054,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-146","imputed":false},{"position":35,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1455.0,"score":0.149,"rank":35.79,"hrank":23.08,"imp":14.857,"train_time":0.65,"predict_time":0.154,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+178\/-163","imputed":false},{"position":36,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1454.0,"score":0.154,"rank":35.91,"hrank":28.34,"imp":15.465,"train_time":572.82,"predict_time":0.109,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-124","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1437.0,"score":0.06,"rank":37.1,"hrank":33.97,"imp":14.238,"train_time":6.41,"predict_time":0.247,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-59","imputed":false},{"position":38,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1427.0,"score":0.106,"rank":37.81,"hrank":30.36,"imp":14.727,"train_time":371.59,"predict_time":2.386,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-149","imputed":false},{"position":39,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1426.0,"score":0.077,"rank":37.86,"hrank":30.44,"imp":16.866,"train_time":96.61,"predict_time":0.754,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-135","imputed":false},{"position":40,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1409.0,"score":0.079,"rank":39.05,"hrank":34.48,"imp":16.781,"train_time":188.8,"predict_time":0.008,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-104","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1392.0,"score":0.066,"rank":40.19,"hrank":33.63,"imp":14.805,"train_time":1156.69,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-118","imputed":false},{"position":42,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1375.0,"score":0.055,"rank":41.35,"hrank":27.17,"imp":16.118,"train_time":4860.88,"predict_time":42.641,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+119\/-143","imputed":false},{"position":43,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1369.0,"score":0.05,"rank":41.8,"hrank":37.07,"imp":18.373,"train_time":0.71,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-112","imputed":false},{"position":44,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1359.0,"score":0.037,"rank":42.48,"hrank":39.09,"imp":15.811,"train_time":0.29,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-112","imputed":false},{"position":45,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1338.0,"score":0.054,"rank":43.9,"hrank":38.92,"imp":19.541,"train_time":82.93,"predict_time":8.231,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-117","imputed":false},{"position":46,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1324.0,"score":0.013,"rank":44.81,"hrank":41.52,"imp":17.083,"train_time":8.34,"predict_time":0.25,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-84","imputed":false},{"position":47,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1323.0,"score":0.003,"rank":44.85,"hrank":42.66,"imp":17.01,"train_time":42.95,"predict_time":0.226,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-67","imputed":false},{"position":48,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1311.0,"score":0.012,"rank":45.67,"hrank":41.98,"imp":18.698,"train_time":16.41,"predict_time":16.24,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-114","imputed":false},{"position":49,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1291.0,"score":0.005,"rank":46.93,"hrank":44.82,"imp":16.58,"train_time":59.3,"predict_time":0.156,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-106","imputed":false},{"position":50,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1287.0,"score":0.006,"rank":47.15,"hrank":43.65,"imp":17.883,"train_time":42.95,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-76","imputed":false},{"position":51,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1284.0,"score":0.029,"rank":47.4,"hrank":40.87,"imp":18.045,"train_time":371.59,"predict_time":0.102,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-139","imputed":false},{"position":52,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1283.0,"score":0.019,"rank":47.44,"hrank":37.01,"imp":20.148,"train_time":47.03,"predict_time":45.876,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+160\/-190","imputed":false},{"position":53,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1267.0,"score":0.076,"rank":48.46,"hrank":31.46,"imp":22.562,"train_time":5.11,"predict_time":2.699,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+184\/-316","imputed":false},{"position":54,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.0,"rank":49.85,"hrank":47.99,"imp":18.065,"train_time":1.86,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-85","imputed":false},{"position":55,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1244.0,"score":0.0,"rank":49.85,"hrank":47.62,"imp":17.828,"train_time":59.3,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-103","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1220.0,"score":0.033,"rank":51.28,"hrank":44.39,"imp":20.939,"train_time":1.52,"predict_time":0.115,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-208","imputed":false},{"position":57,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1157.0,"score":0.009,"rank":54.91,"hrank":49.11,"imp":25.91,"train_time":43.64,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+171\/-229","imputed":false},{"position":58,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1141.0,"score":0.01,"rank":55.78,"hrank":46.38,"imp":23.022,"train_time":513.06,"predict_time":78.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-172","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1141.0,"score":0.0,"rank":55.78,"hrank":53.92,"imp":25.608,"train_time":3265.68,"predict_time":1.948,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-117","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1132.0,"score":0.0,"rank":56.28,"hrank":54.48,"imp":25.685,"train_time":3265.68,"predict_time":5.38,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-123","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1116.0,"score":0.005,"rank":57.12,"hrank":52.24,"imp":26.784,"train_time":43.64,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+177\/-221","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1077.0,"score":0.0,"rank":59.01,"hrank":50.8,"imp":25.041,"train_time":20.26,"predict_time":0.641,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+134\/-222","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1042.0,"score":0.0,"rank":60.64,"hrank":55.35,"imp":26.165,"train_time":20.26,"predict_time":0.258,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+133\/-240","imputed":false},{"position":64,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1011.0,"score":0.0,"rank":62.01,"hrank":60.53,"imp":24.431,"train_time":0.34,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-232","imputed":false},{"position":65,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1005.0,"score":0.0,"rank":62.26,"hrank":59.87,"imp":28.451,"train_time":0.22,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-238","imputed":false},{"position":66,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":62.46,"hrank":61.48,"imp":24.92,"train_time":0.12,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-220","imputed":false},{"position":67,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":990.0,"score":0.0,"rank":62.85,"hrank":61.48,"imp":26.242,"train_time":17.63,"predict_time":7.0,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+109\/-205","imputed":false},{"position":68,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":988.0,"score":0.023,"rank":62.95,"hrank":55.55,"imp":30.971,"train_time":2.62,"predict_time":0.186,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+183\/-376","imputed":false},{"position":69,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":967.0,"score":0.0,"rank":63.75,"hrank":62.86,"imp":31.128,"train_time":101.08,"predict_time":2.021,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-120","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":936.0,"score":0.0,"rank":64.89,"hrank":63.78,"imp":32.252,"train_time":0.24,"predict_time":0.036,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-163","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":917.0,"score":0.0,"rank":65.57,"hrank":65.27,"imp":29.572,"train_time":513.06,"predict_time":2.568,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-112","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":856.0,"score":0.0,"rank":67.43,"hrank":66.17,"imp":34.486,"train_time":1.21,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+158\/-266","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":572.0,"score":0.0,"rank":72.49,"hrank":72.46,"imp":46.783,"train_time":0.11,"predict_time":0.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-272","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_medium/pareto_front_explorer.html
index 38429986c5ad78e12aae8481fe627d99c558efb8..99a6dab2e3dfeb730e15ad8b11cd428c54299823 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.754458801,"imp":16.8661638698,"elo":1425.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.01764933,"imp":8.2891264052,"elo":1761.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0196975147,"imp":8.9223628261,"elo":1740.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1663811446,"imp":8.6254985581,"elo":1754.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0233879248,"imp":11.8612633831,"elo":1617.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0265970701,"imp":11.1861216719,"elo":1655.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3315968945,"imp":10.5225022079,"elo":1720.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"imp":18.3734807884,"elo":1368.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0083401831,"imp":16.781436104,"elo":1408.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0872584538,"imp":15.7095805742,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0570892384,"imp":24.4306517864,"elo":1010.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0153132761,"imp":17.8275871335,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1561119171,"imp":16.5804781208,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1146303549,"imp":20.9390549114,"elo":1220.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1021965345,"imp":18.0452245989,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.385543961,"imp":14.7272167972,"elo":1426.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0329178148,"imp":15.8110309462,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0776085786,"imp":13.1055792349,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.6682734383,"imp":11.4198756165,"elo":1697.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.6994157224,"imp":22.5618336989,"elo":1266.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"imp":17.0831560703,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3490410688,"imp":13.1951996828,"elo":1557.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.4096341011,"imp":10.7090216199,"elo":1675.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.9998247538,"imp":26.2416785073,"elo":990.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0296504802,"imp":34.4859755804,"elo":855.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2581384468,"imp":26.1649335607,"elo":1042.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6407534322,"imp":25.0413536419,"elo":1077.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0137007533,"imp":24.9196784107,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0211930497,"imp":17.8834807855,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2259131636,"imp":17.0100195458,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2472765634,"imp":14.2382516089,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0879217199,"imp":10.5105993564,"elo":1650.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.053386569,"imp":9.1511055435,"elo":1745.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8963303312,"imp":12.1973167677,"elo":1589.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4652611394,"imp":12.4444200979,"elo":1608.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":15.1184451567,"imp":11.8937892778,"elo":1650.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.2398511784,"imp":18.6984349634,"elo":1310.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.8758317963,"imp":20.1484844796,"elo":1283.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.6414441586,"imp":16.11836212,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":475.7800751889,"imp":12.7728075069,"elo":1538.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2033080244,"imp":14.3714973635,"elo":1530.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":19.8176830022,"imp":3.020015071,"elo":2054.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0356431156,"imp":32.2515992944,"elo":936.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.5782128678,"imp":12.8687318099,"elo":1572.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1979712895,"imp":8.2697512852,"elo":1781.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0535231299,"imp":13.8042192486,"elo":1470.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0562011083,"imp":11.2477717069,"elo":1635.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5790353938,"imp":10.3518431999,"elo":1715.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8857412969,"imp":9.8284104744,"elo":1787.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3059932611,"imp":8.4484775587,"elo":1869.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0205027027,"imp":31.1278628307,"elo":967.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9483778428,"imp":25.6077503055,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3796059211,"imp":25.685200574,"elo":1131.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.154334046,"imp":14.8573655394,"elo":1455.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0463224066,"imp":18.0651304935,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0539785703,"imp":14.8048858983,"elo":1392.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5584373713,"imp":12.5825653736,"elo":1537.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"imp":14.168487644,"elo":1473.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0608575349,"imp":11.7747720598,"elo":1640.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3857911875,"imp":10.9496331211,"elo":1712.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":8.2311408315,"imp":19.5407183337,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.8813375711,"imp":12.7751500563,"elo":1567.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":80.7682550616,"imp":9.7844509775,"elo":1760.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1859407599,"imp":30.97120717,"elo":988.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1086014839,"imp":15.4647952687,"elo":1453.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.9349365505,"imp":13.7487303387,"elo":1485.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.754458801,"x_train":96.610665791,"imp":16.8661638698,"elo":1425.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.01764933,"x_train":1.3820827827,"imp":8.2891264052,"elo":1761.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0196975147,"x_train":329.0507531312,"imp":8.9223628261,"elo":1740.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1663811446,"x_train":329.0507531312,"imp":8.6254985581,"elo":1754.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0233879248,"x_train":0.4789788316,"imp":11.8612633831,"elo":1617.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0265970701,"x_train":95.9169155889,"imp":11.1861216719,"elo":1655.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3315968945,"x_train":95.9169155889,"imp":10.5225022079,"elo":1720.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"x_train":0.7112587015,"imp":18.3734807884,"elo":1368.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0083401831,"x_train":188.7956405878,"imp":16.781436104,"elo":1408.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0872584538,"x_train":188.7956405878,"imp":15.7095805742,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0570892384,"x_train":0.3400693672,"imp":24.4306517864,"elo":1010.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0153132761,"x_train":59.3011906074,"imp":17.8275871335,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1561119171,"x_train":59.3011906074,"imp":16.5804781208,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1146303549,"x_train":1.5179051995,"imp":20.9390549114,"elo":1220.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1021965345,"x_train":371.5880936781,"imp":18.0452245989,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.385543961,"x_train":371.5880936781,"imp":14.7272167972,"elo":1426.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0329178148,"x_train":0.2873206973,"imp":15.8110309462,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0776085786,"x_train":86.8470118165,"imp":13.1055792349,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.6682734383,"x_train":86.8470118165,"imp":11.4198756165,"elo":1697.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.6994157224,"x_train":5.1116723286,"imp":22.5618336989,"elo":1266.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"x_train":8.3438180493,"imp":17.0831560703,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3490410688,"x_train":2139.1684153663,"imp":13.1951996828,"elo":1557.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.4096341011,"x_train":2139.1684153663,"imp":10.7090216199,"elo":1675.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.9998247538,"x_train":17.6345194999,"imp":26.2416785073,"elo":990.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0296504802,"x_train":1.213148609,"imp":34.4859755804,"elo":855.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2581384468,"x_train":20.2619047675,"imp":26.1649335607,"elo":1042.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6407534322,"x_train":20.2619047675,"imp":25.0413536419,"elo":1077.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0137007533,"x_train":0.1197809845,"imp":24.9196784107,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0211930497,"x_train":42.9504760952,"imp":17.8834807855,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2259131636,"x_train":42.9504760952,"imp":17.0100195458,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2472765634,"x_train":6.412776503,"imp":14.2382516089,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0879217199,"x_train":1633.8040014148,"imp":10.5105993564,"elo":1650.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.053386569,"x_train":1633.8040014148,"imp":9.1511055435,"elo":1745.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8963303312,"x_train":2.2999820948,"imp":12.1973167677,"elo":1589.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4652611394,"x_train":842.247738377,"imp":12.4444200979,"elo":1608.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":15.1184451567,"x_train":842.247738377,"imp":11.8937892778,"elo":1650.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.2398511784,"x_train":16.4124685619,"imp":18.6984349634,"elo":1310.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.8758317963,"x_train":47.0281853011,"imp":20.1484844796,"elo":1283.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.6414441586,"x_train":4860.878886905,"imp":16.11836212,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":475.7800751889,"x_train":4860.878886905,"imp":12.7728075069,"elo":1538.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2033080244,"x_train":0.7584924197,"imp":14.3714973635,"elo":1530.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":19.8176830022,"x_train":42.1723102051,"imp":3.020015071,"elo":2054.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0356431156,"x_train":0.2354314672,"imp":32.2515992944,"elo":936.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.5782128678,"x_train":8.5347633004,"imp":12.8687318099,"elo":1572.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1979712895,"x_train":0.6451456984,"imp":8.2697512852,"elo":1781.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0535231299,"x_train":1.4982282715,"imp":13.8042192486,"elo":1470.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0562011083,"x_train":583.5491237683,"imp":11.2477717069,"elo":1635.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5790353938,"x_train":583.5491237683,"imp":10.3518431999,"elo":1715.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8857412969,"x_train":2.7558592937,"imp":9.8284104744,"elo":1787.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3059932611,"x_train":1.1838833522,"imp":8.4484775587,"elo":1869.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0205027027,"x_train":101.0759279672,"imp":31.1278628307,"elo":967.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9483778428,"x_train":3265.6782173608,"imp":25.6077503055,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3796059211,"x_train":3265.6782173608,"imp":25.685200574,"elo":1131.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.154334046,"x_train":0.6533526697,"imp":14.8573655394,"elo":1455.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0463224066,"x_train":1.8588422186,"imp":18.0651304935,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0539785703,"x_train":1156.6875333303,"imp":14.8048858983,"elo":1392.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5584373713,"x_train":1156.6875333303,"imp":12.5825653736,"elo":1537.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"x_train":0.4869354634,"imp":14.168487644,"elo":1473.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0608575349,"x_train":110.1339438717,"imp":11.7747720598,"elo":1640.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3857911875,"x_train":110.1339438717,"imp":10.9496331211,"elo":1712.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":8.2311408315,"x_train":82.933952473,"imp":19.5407183337,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.8813375711,"x_train":1883.8223518872,"imp":12.7751500563,"elo":1567.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":80.7682550616,"x_train":1883.8223518872,"imp":9.7844509775,"elo":1760.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1859407599,"x_train":2.6170575688,"imp":30.97120717,"elo":988.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1086014839,"x_train":572.8239309061,"imp":15.4647952687,"elo":1453.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.9349365505,"x_train":572.8239309061,"imp":13.7487303387,"elo":1485.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
index db50b300f39846bf507725ec7e6de19d2d3bb1cf..10e5183fd13bda3072a3e0bb44b3690dd3ff1791 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":96.610665791,"imp":16.8869456784,"elo":1462.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.3820827827,"imp":8.3086430794,"elo":1810.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.6835494217,"imp":8.4706547398,"elo":1805.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.7546497969,"imp":8.5841259108,"elo":1813.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":23.6295991191,"imp":8.6424368764,"elo":1812.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":47.3749289529,"imp":8.5691655817,"elo":1821.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":86.7772971055,"imp":8.5420983708,"elo":1824.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":167.2712430131,"imp":8.5965913098,"elo":1816.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":246.0389923704,"imp":8.6173993693,"elo":1812.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":329.0507531312,"imp":8.6369523227,"elo":1808.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4789788316,"imp":11.8817963637,"elo":1661.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.9589536975,"imp":11.2914262643,"elo":1718.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.356217589,"imp":11.0046004786,"elo":1743.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.7221268742,"imp":10.8577960514,"elo":1756.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.9255203586,"imp":10.7741504301,"elo":1764.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":23.8768976028,"imp":10.691764247,"elo":1768.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":47.751833203,"imp":10.5870641658,"elo":1774.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":71.5860615637,"imp":10.546056945,"elo":1776.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":95.9169155889,"imp":10.5395976031,"elo":1776.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":0.7112587015,"imp":18.3887323699,"elo":1389.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":1.8058880969,"imp":17.5681117799,"elo":1423.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":4.573941659,"imp":16.9302588345,"elo":1451.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":9.1250887196,"imp":16.6193459022,"elo":1465.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":23.2616910123,"imp":16.3041625121,"elo":1483.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":47.2291260007,"imp":16.1193353693,"elo":1494.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":94.0295132726,"imp":15.9112062584,"elo":1507.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":141.071175821,"imp":15.8388892899,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":188.7956405878,"imp":15.7173309074,"elo":1519.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3400693672,"imp":24.4462285072,"elo":1015.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.5790262037,"imp":22.0416278917,"elo":1082.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.5414359953,"imp":18.6787831286,"elo":1203.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9353651221,"imp":17.5257408655,"elo":1257.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.3111220646,"imp":16.855997351,"elo":1289.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.5437419558,"imp":16.6465273424,"elo":1305.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":29.1909327794,"imp":16.4874499806,"elo":1317.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.978271297,"imp":16.5043302134,"elo":1319.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":59.3011906074,"imp":16.5978412796,"elo":1319.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.5179051995,"imp":20.9534199059,"elo":1250.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.2504528706,"imp":19.3327356804,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":9.005873763,"imp":17.5571979893,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":18.7663631809,"imp":16.7222137808,"elo":1379.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":44.7672715291,"imp":15.8951142548,"elo":1415.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":90.2243280036,"imp":15.4651613267,"elo":1435.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":183.9710430372,"imp":15.0996327621,"elo":1451.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":276.7808531494,"imp":14.8700614389,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":371.5880936781,"imp":14.7354431653,"elo":1471.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.2873206973,"imp":15.8281297505,"elo":1386.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.7523280043,"imp":14.9384872067,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":1.9490903848,"imp":13.7308752861,"elo":1575.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":4.0385354064,"imp":12.8004998388,"elo":1660.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":10.406685589,"imp":12.0514195339,"elo":1711.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":21.6557265527,"imp":11.735781379,"elo":1736.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":43.0306651896,"imp":11.5390381014,"elo":1746.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":64.0073656446,"imp":11.4725564601,"elo":1748.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":86.8470118165,"imp":11.4489706957,"elo":1750.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":5.1116723286,"imp":22.0964299162,"elo":1313.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":8.3438180493,"imp":17.0999174747,"elo":1354.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.3687146046,"imp":15.0875586421,"elo":1456.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.9177628832,"imp":12.6189562445,"elo":1614.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":112.4005567224,"imp":11.8503204406,"elo":1663.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":260.3290997978,"imp":11.1003331005,"elo":1713.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":532.4813413639,"imp":10.8147952016,"elo":1732.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1051.303381757,"imp":10.7709785782,"elo":1738.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1569.7102742529,"imp":10.7545797716,"elo":1736.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2139.1684153663,"imp":10.7426084688,"elo":1732.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":17.6345194999,"imp":26.2069426341,"elo":1005.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.213148609,"imp":34.9227236672,"elo":870.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.4317773957,"imp":32.6695542912,"elo":935.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":16.9493056602,"imp":25.175368626,"elo":1096.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.2619047675,"imp":25.0463843754,"elo":1104.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1197809845,"imp":24.9344281054,"elo":1000.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3353669751,"imp":20.1928378008,"elo":1158.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.0588012865,"imp":18.3428197494,"elo":1255.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":2.1551843555,"imp":17.7161521778,"elo":1294.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":5.269352713,"imp":17.2837785806,"elo":1326.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":10.5968016873,"imp":17.1321851807,"elo":1338.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":21.2572302017,"imp":17.0271739624,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":32.0070168617,"imp":16.9741463147,"elo":1354.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":42.9504760952,"imp":17.0267485259,"elo":1357.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.412776503,"imp":14.2522040484,"elo":1467.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.4470659835,"imp":12.2716889306,"elo":1611.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.1186736736,"imp":11.2130847519,"elo":1686.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":79.8382984311,"imp":10.5621920308,"elo":1721.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":204.957931587,"imp":9.911741712,"elo":1761.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":413.9931126009,"imp":9.6468682449,"elo":1774.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":813.6028057178,"imp":9.3433666543,"elo":1790.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1220.4699358016,"imp":9.1915960063,"elo":1800.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1633.8040014148,"imp":9.168219635,"elo":1805.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2.2999820948,"imp":12.21913411,"elo":1627.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.6597534236,"imp":12.1879102238,"elo":1635.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":21.5428942424,"imp":12.1199442425,"elo":1646.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":39.0796097222,"imp":12.0869199997,"elo":1653.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":88.9628219424,"imp":11.9639017638,"elo":1677.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":188.8767701417,"imp":11.9693640401,"elo":1690.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":396.9834212212,"imp":11.961761151,"elo":1697.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":604.8684512841,"imp":11.9607451636,"elo":1696.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":842.247738377,"imp":11.9072122952,"elo":1694.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":16.4124685619,"imp":18.7099466002,"elo":1341.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":47.0281853011,"imp":20.1619369601,"elo":1316.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.0348846629,"imp":18.390424005,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":127.5381763694,"imp":16.6621751076,"elo":1425.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":230.7293949461,"imp":16.0702194481,"elo":1447.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":589.1524400719,"imp":14.7273331995,"elo":1510.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1198.3439089076,"imp":13.9901152254,"elo":1545.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2408.9332260936,"imp":13.5007187905,"elo":1565.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3630.8641964758,"imp":13.0556520127,"elo":1583.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4860.878886905,"imp":12.786090517,"elo":1588.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.7584924197,"imp":14.3878981343,"elo":1572.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":42.1723102051,"imp":3.0198899691,"elo":2094.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.2354314672,"imp":32.2629483255,"elo":946.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":8.5347633004,"imp":12.8892694171,"elo":1619.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6451456984,"imp":8.2928759373,"elo":1826.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.4982282715,"imp":13.8226609481,"elo":1506.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":3.9284284946,"imp":13.0585880383,"elo":1550.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":12.4876393146,"imp":12.1100926661,"elo":1616.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":27.3834783781,"imp":11.5365261569,"elo":1666.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":71.1997682157,"imp":10.9618621861,"elo":1705.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":143.8112757102,"imp":10.6986168262,"elo":1730.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":285.6452634754,"imp":10.4860402965,"elo":1747.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":432.2819705079,"imp":10.4082446029,"elo":1755.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":583.5491237683,"imp":10.3634823237,"elo":1762.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.7558592937,"imp":9.8498615705,"elo":1839.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.1838833522,"imp":8.4535084303,"elo":1914.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":101.0759279672,"imp":31.1369080501,"elo":971.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":111.2079544113,"imp":31.05854612,"elo":986.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":141.6266688837,"imp":30.8939189678,"elo":994.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":187.760864208,"imp":30.4005168147,"elo":1015.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":414.9961090703,"imp":29.3533425922,"elo":1046.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":831.557341273,"imp":28.0743782128,"elo":1081.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1652.4911170928,"imp":27.2274709966,"elo":1110.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2441.373538211,"imp":26.4148787287,"elo":1139.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3265.6782173608,"imp":25.6967661148,"elo":1171.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.6533526697,"imp":14.8784132282,"elo":1494.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":1.8588422186,"imp":18.0767520838,"elo":1266.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.7947477355,"imp":16.6067405067,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":26.9428583764,"imp":15.2096234808,"elo":1407.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":62.6027434746,"imp":14.5141871393,"elo":1460.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":150.4831797618,"imp":13.7553693218,"elo":1507.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":283.6255748988,"imp":13.2666108735,"elo":1541.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":555.0904532489,"imp":12.9692715177,"elo":1564.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":859.203668693,"imp":12.761614669,"elo":1578.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1156.6875333303,"imp":12.6027470128,"elo":1586.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4869354634,"imp":14.1881619037,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":1.1317321011,"imp":12.5919116327,"elo":1643.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":2.9964300137,"imp":11.9134244829,"elo":1704.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":5.3381407276,"imp":11.4190066821,"elo":1735.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":13.9223759351,"imp":11.1330668545,"elo":1750.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":27.5993202576,"imp":11.0312057564,"elo":1756.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":55.0233345183,"imp":10.9726161464,"elo":1762.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":81.9474032161,"imp":10.9693188118,"elo":1764.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":110.1339438717,"imp":10.9720087259,"elo":1766.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":82.933952473,"imp":19.5561303695,"elo":1369.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":155.9256875342,"imp":17.2867604146,"elo":1427.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":374.1243102893,"imp":14.4138746108,"elo":1549.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":733.7366585123,"imp":12.2405692825,"elo":1664.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1813.3707967819,"imp":9.8456285395,"elo":1813.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1883.8223518872,"imp":9.8041530963,"elo":1819.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.6170575688,"imp":30.9829138539,"elo":999.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.9970579818,"imp":24.2891420854,"elo":1113.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":14.4751935279,"imp":17.1529800491,"elo":1326.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":28.4433104494,"imp":14.6789332659,"elo":1477.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":71.1413647457,"imp":13.8532051581,"elo":1530.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":142.1475184089,"imp":13.571857429,"elo":1547.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":285.1331865035,"imp":13.5502598296,"elo":1545.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":427.4360460386,"imp":13.6114489793,"elo":1540.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":572.8239309061,"imp":13.7136725988,"elo":1532.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.754458801,"x_train":96.610665791,"imp":16.8869456784,"elo":1462.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.01764933,"x_train":1.3820827827,"imp":8.3086430794,"elo":1810.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0357929968,"x_train":3.6835494217,"imp":8.4706547398,"elo":1805.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0590771616,"x_train":12.7546497969,"imp":8.5841259108,"elo":1813.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0772621764,"x_train":23.6295991191,"imp":8.6424368764,"elo":1812.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0947269938,"x_train":47.3749289529,"imp":8.5691655817,"elo":1821.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1111539594,"x_train":86.7772971055,"imp":8.5420983708,"elo":1824.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1397193986,"x_train":167.2712430131,"imp":8.5965913098,"elo":1816.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1540251012,"x_train":246.0389923704,"imp":8.6173993693,"elo":1812.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1663811446,"x_train":329.0507531312,"imp":8.6369523227,"elo":1808.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0233879248,"x_train":0.4789788316,"imp":11.8817963637,"elo":1661.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0478671165,"x_train":0.9589536975,"imp":11.2914262643,"elo":1718.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1129020181,"x_train":2.356217589,"imp":11.0046004786,"elo":1743.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1771034691,"x_train":4.7221268742,"imp":10.8577960514,"elo":1756.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2454903367,"x_train":11.9255203586,"imp":10.7741504301,"elo":1764.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2926890222,"x_train":23.8768976028,"imp":10.691764247,"elo":1768.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.312025413,"x_train":47.751833203,"imp":10.5870641658,"elo":1774.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3237207716,"x_train":71.5860615637,"imp":10.546056945,"elo":1776.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3206222047,"x_train":95.9169155889,"imp":10.5395976031,"elo":1776.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"x_train":0.7112587015,"imp":18.3887323699,"elo":1389.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0116608277,"x_train":1.8058880969,"imp":17.5681117799,"elo":1423.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0226990233,"x_train":4.573941659,"imp":16.9302588345,"elo":1451.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333135949,"x_train":9.1250887196,"imp":16.6193459022,"elo":1465.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0501304583,"x_train":23.2616910123,"imp":16.3041625121,"elo":1483.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0610135476,"x_train":47.2291260007,"imp":16.1193353693,"elo":1494.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0728842068,"x_train":94.0295132726,"imp":15.9112062584,"elo":1507.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0813406855,"x_train":141.071175821,"imp":15.8388892899,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0885616373,"x_train":188.7956405878,"imp":15.7173309074,"elo":1519.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0570892384,"x_train":0.3400693672,"imp":24.4462285072,"elo":1015.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0703457475,"x_train":0.5790262037,"imp":22.0416278917,"elo":1082.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0928856349,"x_train":1.5414359953,"imp":18.6787831286,"elo":1203.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1010934443,"x_train":2.9353651221,"imp":17.5257408655,"elo":1257.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1082818488,"x_train":7.3111220646,"imp":16.855997351,"elo":1289.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1223033865,"x_train":14.5437419558,"imp":16.6465273424,"elo":1305.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1386914014,"x_train":29.1909327794,"imp":16.4874499806,"elo":1317.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1459534487,"x_train":43.978271297,"imp":16.5043302134,"elo":1319.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1524033833,"x_train":59.3011906074,"imp":16.5978412796,"elo":1319.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1146303549,"x_train":1.5179051995,"imp":20.9534199059,"elo":1250.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.224806949,"x_train":3.2504528706,"imp":19.3327356804,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5565998143,"x_train":9.005873763,"imp":17.5571979893,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9690827943,"x_train":18.7663631809,"imp":16.7222137808,"elo":1379.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3936753554,"x_train":44.7672715291,"imp":15.8951142548,"elo":1415.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6932007533,"x_train":90.2243280036,"imp":15.4651613267,"elo":1435.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0269078994,"x_train":183.9710430372,"imp":15.0996327621,"elo":1451.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.2179323369,"x_train":276.7808531494,"imp":14.8700614389,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.3776360952,"x_train":371.5880936781,"imp":14.7354431653,"elo":1471.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0329178148,"x_train":0.2873206973,"imp":15.8281297505,"elo":1386.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.3022864661,"x_train":0.7523280043,"imp":14.9384872067,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5169466094,"x_train":1.9490903848,"imp":13.7308752861,"elo":1575.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7806727427,"x_train":4.0385354064,"imp":12.8004998388,"elo":1660.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7659565508,"x_train":10.406685589,"imp":12.0514195339,"elo":1711.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7608339553,"x_train":21.6557265527,"imp":11.735781379,"elo":1736.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6794432016,"x_train":43.0306651896,"imp":11.5390381014,"elo":1746.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.642292477,"x_train":64.0073656446,"imp":11.4725564601,"elo":1748.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6494090854,"x_train":86.8470118165,"imp":11.4489706957,"elo":1750.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":2.6994157224,"x_train":5.1116723286,"imp":22.0964299162,"elo":1313.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"x_train":8.3438180493,"imp":17.0999174747,"elo":1354.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5785744608,"x_train":25.3687146046,"imp":15.0875586421,"elo":1456.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.2947480655,"x_train":60.9177628832,"imp":12.6189562445,"elo":1614.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1427076773,"x_train":112.4005567224,"imp":11.8503204406,"elo":1663.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.2223339105,"x_train":260.3290997978,"imp":11.1003331005,"elo":1713.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.4032452473,"x_train":532.4813413639,"imp":10.8147952016,"elo":1732.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5772898048,"x_train":1051.303381757,"imp":10.7709785782,"elo":1738.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.1475244152,"x_train":1569.7102742529,"imp":10.7545797716,"elo":1736.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3256045535,"x_train":2139.1684153663,"imp":10.7426084688,"elo":1732.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.9998247538,"x_train":17.6345194999,"imp":26.2069426341,"elo":1005.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0296504802,"x_train":1.213148609,"imp":34.9227236672,"elo":870.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2293591891,"x_train":6.4317773957,"imp":32.6695542912,"elo":935.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5854122012,"x_train":16.9493056602,"imp":25.175368626,"elo":1096.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6407534322,"x_train":20.2619047675,"imp":25.0463843754,"elo":1104.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0137007533,"x_train":0.1197809845,"imp":24.9344281054,"elo":1000.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0410613688,"x_train":0.3353669751,"imp":20.1928378008,"elo":1158.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0866608789,"x_train":1.0588012865,"imp":18.3428197494,"elo":1255.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1305578206,"x_train":2.1551843555,"imp":17.7161521778,"elo":1294.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1467222069,"x_train":5.269352713,"imp":17.2837785806,"elo":1326.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1724734092,"x_train":10.5968016873,"imp":17.1321851807,"elo":1338.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2013078773,"x_train":21.2572302017,"imp":17.0271739624,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2097607806,"x_train":32.0070168617,"imp":16.9741463147,"elo":1354.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2232886539,"x_train":42.9504760952,"imp":17.0267485259,"elo":1357.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2472765634,"x_train":6.412776503,"imp":14.2522040484,"elo":1467.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4377557757,"x_train":15.4470659835,"imp":12.2716889306,"elo":1611.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.6660949816,"x_train":39.1186736736,"imp":11.2130847519,"elo":1686.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.8989691909,"x_train":79.8382984311,"imp":10.5621920308,"elo":1721.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1399419366,"x_train":204.957931587,"imp":9.911741712,"elo":1761.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.3802752953,"x_train":413.9931126009,"imp":9.6468682449,"elo":1774.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6732415593,"x_train":813.6028057178,"imp":9.3433666543,"elo":1790.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.9229356811,"x_train":1220.4699358016,"imp":9.1915960063,"elo":1800.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.0414922251,"x_train":1633.8040014148,"imp":9.168219635,"elo":1805.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8963303312,"x_train":2.2999820948,"imp":12.21913411,"elo":1627.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.927863305,"x_train":6.6597534236,"imp":12.1879102238,"elo":1635.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.2372972886,"x_train":21.5428942424,"imp":12.1199442425,"elo":1646.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.3430098245,"x_train":39.0796097222,"imp":12.0869199997,"elo":1653.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1180506729,"x_train":88.9628219424,"imp":11.9639017638,"elo":1677.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.5551205244,"x_train":188.8767701417,"imp":11.9693640401,"elo":1690.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.3131673972,"x_train":396.9834212212,"imp":11.961761151,"elo":1697.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":13.5329260857,"x_train":604.8684512841,"imp":11.9607451636,"elo":1696.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":14.6273407585,"x_train":842.247738377,"imp":11.9072122952,"elo":1694.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.2398511784,"x_train":16.4124685619,"imp":18.7099466002,"elo":1341.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":45.8758317963,"x_train":47.0281853011,"imp":20.1619369601,"elo":1316.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":66.3525228473,"x_train":70.0348846629,"imp":18.390424005,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":98.2456271023,"x_train":127.5381763694,"imp":16.6621751076,"elo":1425.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":142.0328330544,"x_train":230.7293949461,"imp":16.0702194481,"elo":1447.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":218.1812948024,"x_train":589.1524400719,"imp":14.7273331995,"elo":1510.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":285.9879390033,"x_train":1198.3439089076,"imp":13.9901152254,"elo":1545.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":390.5997136252,"x_train":2408.9332260936,"imp":13.5007187905,"elo":1565.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":442.8772138735,"x_train":3630.8641964758,"imp":13.0556520127,"elo":1583.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":481.8516025188,"x_train":4860.878886905,"imp":12.786090517,"elo":1588.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2033080244,"x_train":0.7584924197,"imp":14.3878981343,"elo":1572.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":19.8176830022,"x_train":42.1723102051,"imp":3.0198899691,"elo":2094.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0356431156,"x_train":0.2354314672,"imp":32.2629483255,"elo":946.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":3.5782128678,"x_train":8.5347633004,"imp":12.8892694171,"elo":1619.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1979712895,"x_train":0.6451456984,"imp":8.2928759373,"elo":1826.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0535231299,"x_train":1.4982282715,"imp":13.8226609481,"elo":1506.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1154003256,"x_train":3.9284284946,"imp":13.0585880383,"elo":1550.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2000419921,"x_train":12.4876393146,"imp":12.1100926661,"elo":1616.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2767821239,"x_train":27.3834783781,"imp":11.5365261569,"elo":1666.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.36488929,"x_train":71.1997682157,"imp":10.9618621861,"elo":1705.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4641659542,"x_train":143.8112757102,"imp":10.6986168262,"elo":1730.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.5371776272,"x_train":285.6452634754,"imp":10.4860402965,"elo":1747.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.5567799537,"x_train":432.2819705079,"imp":10.4082446029,"elo":1755.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5682121276,"x_train":583.5491237683,"imp":10.3634823237,"elo":1762.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8857412969,"x_train":2.7558592937,"imp":9.8498615705,"elo":1839.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3059932611,"x_train":1.1838833522,"imp":8.4535084303,"elo":1914.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0205027027,"x_train":101.0759279672,"imp":31.1369080501,"elo":971.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0983982624,"x_train":111.2079544113,"imp":31.05854612,"elo":986.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.2224011522,"x_train":141.6266688837,"imp":30.8939189678,"elo":994.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5850970839,"x_train":187.760864208,"imp":30.4005168147,"elo":1015.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.2850393528,"x_train":414.9961090703,"imp":29.3533425922,"elo":1046.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4241099579,"x_train":831.557341273,"imp":28.0743782128,"elo":1081.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.4436810211,"x_train":1652.4911170928,"imp":27.2274709966,"elo":1110.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.0366030881,"x_train":2441.373538211,"imp":26.4148787287,"elo":1139.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.365156164,"x_train":3265.6782173608,"imp":25.6967661148,"elo":1171.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.154334046,"x_train":0.6533526697,"imp":14.8784132282,"elo":1494.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0463224066,"x_train":1.8588422186,"imp":18.0767520838,"elo":1266.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1000380446,"x_train":6.7947477355,"imp":16.6067405067,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.262302914,"x_train":26.9428583764,"imp":15.2096234808,"elo":1407.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4175923212,"x_train":62.6027434746,"imp":14.5141871393,"elo":1460.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6171906817,"x_train":150.4831797618,"imp":13.7553693218,"elo":1507.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9111530149,"x_train":283.6255748988,"imp":13.2666108735,"elo":1541.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1543711789,"x_train":555.0904532489,"imp":12.9692715177,"elo":1564.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3269752577,"x_train":859.203668693,"imp":12.761614669,"elo":1578.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5455206258,"x_train":1156.6875333303,"imp":12.6027470128,"elo":1586.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"x_train":0.4869354634,"imp":14.1881619037,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1297403015,"x_train":1.1317321011,"imp":12.5919116327,"elo":1643.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2494283602,"x_train":2.9964300137,"imp":11.9134244829,"elo":1704.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2835102125,"x_train":5.3381407276,"imp":11.4190066821,"elo":1735.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3287421252,"x_train":13.9223759351,"imp":11.1330668545,"elo":1750.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3407457713,"x_train":27.5993202576,"imp":11.0312057564,"elo":1756.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3388091225,"x_train":55.0233345183,"imp":10.9726161464,"elo":1762.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3715591178,"x_train":81.9474032161,"imp":10.9693188118,"elo":1764.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3857911875,"x_train":110.1339438717,"imp":10.9720087259,"elo":1766.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":8.2311408315,"x_train":82.933952473,"imp":19.5561303695,"elo":1369.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":17.9776071875,"x_train":155.9256875342,"imp":17.2867604146,"elo":1427.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":33.2102432394,"x_train":374.1243102893,"imp":14.4138746108,"elo":1549.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":49.0550574883,"x_train":733.7366585123,"imp":12.2405692825,"elo":1664.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":80.6306891729,"x_train":1813.3707967819,"imp":9.8456285395,"elo":1813.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":81.4550745619,"x_train":1883.8223518872,"imp":9.8041530963,"elo":1819.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1859407599,"x_train":2.6170575688,"imp":30.9829138539,"elo":999.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.292827281,"x_train":5.9970579818,"imp":24.2891420854,"elo":1113.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.4370737653,"x_train":14.4751935279,"imp":17.1529800491,"elo":1326.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.6030397775,"x_train":28.4433104494,"imp":14.6789332659,"elo":1477.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.893684495,"x_train":71.1413647457,"imp":13.8532051581,"elo":1530.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.1268006825,"x_train":142.1475184089,"imp":13.571857429,"elo":1547.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.4622467576,"x_train":285.1331865035,"imp":13.5502598296,"elo":1545.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.7229354568,"x_train":427.4360460386,"imp":13.6114489793,"elo":1540.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9204374353,"x_train":572.8239309061,"imp":13.7136725988,"elo":1532.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..48ee12c35d4c7cb70e32e38ba7ca3d53b93743ad
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_binary/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "TabFM (default)", "AutoGluon 1.4 (best, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "CatBoost (default)", "iLTM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "RealMLP (tuned + ensembled)", "CatBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "TabM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "LightGBM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "ChimeraBoost (tuned)", "RealMLP (tuned)", "RealTabPFN-2.5 (tuned + ensembled)", "XGBoost (tuned)", "TabM (tuned)", "ChimeraBoost (default)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (default)", "LightGBM (tuned)", "TabICL (default)", "iLTM (tuned)", "ModernNCA (tuned)", "TabDPT (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "xRFM (tuned + ensembled)", "EBM (tuned + ensembled)", "XGBoost (default)", "TabM (default)", "TabSwift (default)", "xRFM (tuned)", "RealMLP (default)", "FastaiMLP (tuned + ensembled)", "BetaTabPFN (default)", "EBM (tuned)", "TorchMLP (tuned)", "TabDPT (tuned)", "EBM (default)", "LightGBM (default)", "iLTM (default)", "ModernNCA (default)", "RandomForest (tuned + ensembled)", "SAP-RPT-OSS (default)", "ExtraTrees (tuned + ensembled)", "RandomForest (tuned)", "FastaiMLP (tuned)", "TabDPT (default)", "LimiX (default)", "TorchMLP (default)", "ExtraTrees (tuned)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "KNN (tuned + ensembled)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "PerpetualBooster (tuned)", "ExtraTrees (default)", "Linear (default)", "RandomForest (default)", "OrionMSP (default)", "xRFM (default)", "TabSTAR (default)", "TabFlex (default)", "KNN (tuned)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.3333333432674408, 0.7160493731498718, 0.45679011940956116, 0.8024691343307495, 0.6419752836227417, 0.8888888955116272, 0.9259259104728699, 0.9135802388191223, 0.9753086566925049, 0.9012345671653748, 0.9135802388191223, 0.790123462677002, 0.9506173133850098, 0.9259259104728699, 0.9629629850387573, 0.9259259104728699, 0.9629629850387573, 0.9135802388191223, 0.9876543283462524, 0.8518518805503845, 0.9629629850387573, 0.9135802388191223, 0.9135802388191223, 0.9753086566925049, 0.8888888955116272, 0.8765432238578796, 0.9876543283462524, 0.9135802388191223, 0.9876543283462524, 0.9012345671653748, 0.9753086566925049, 0.9876543283462524, 0.9629629850387573, 1.0, 0.9382715821266174, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9506173133850098, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9876543283462524, 1.0, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.6666666865348816, NaN, 0.6790123581886292, 0.9012345671653748, 0.9506173133850098, 0.9012345671653748, 0.7654321193695068, 0.8271604776382446, 0.7530864477157593, 0.8024691343307495, 0.8024691343307495, 0.8641975522041321, 0.9876543283462524, 0.8641975522041321, 0.8395061492919922, 0.8765432238578796, 0.8641975522041321, 0.8148148059844971, 0.9629629850387573, 0.9012345671653748, 1.0, 0.8888888955116272, 0.9629629850387573, 0.9753086566925049, 0.9629629850387573, 0.9506173133850098, 0.8888888955116272, 0.9382715821266174, 0.9629629850387573, 0.9629629850387573, 0.9629629850387573, 0.8765432238578796, 0.9876543283462524, 0.9753086566925049, 1.0, 0.9506173133850098, 0.9012345671653748, 1.0, 0.9012345671653748, 1.0, 1.0, 1.0, 0.9629629850387573, 1.0, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9753086566925049, 0.9506173133850098, 1.0, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9506173133850098, 1.0, 1.0, 0.9753086566925049, 1.0, 1.0, 0.9876543283462524, 1.0, 0.9876543283462524, 1.0, 0.9506173133850098, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2839506268501282, 0.32098764181137085, NaN, 0.4691357910633087, 0.6790123581886292, 0.6172839403152466, 0.6913580298423767, 0.8271604776382446, 0.7407407164573669, 0.8518518805503845, 0.7530864477157593, 0.7530864477157593, 0.7160493731498718, 0.8518518805503845, 0.8271604776382446, 0.7654321193695068, 0.8518518805503845, 0.9135802388191223, 0.7037037014961243, 0.8765432238578796, 0.7777777910232544, 0.8765432238578796, 0.7530864477157593, 0.7407407164573669, 0.8888888955116272, 0.7407407164573669, 0.8395061492919922, 0.8888888955116272, 0.7777777910232544, 0.9012345671653748, 0.790123462677002, 0.9506173133850098, 0.9506173133850098, 0.9135802388191223, 0.9135802388191223, 0.8271604776382446, 0.9629629850387573, 0.9629629850387573, 0.9629629850387573, 0.9012345671653748, 0.9506173133850098, 0.9753086566925049, 0.9629629850387573, 0.9506173133850098, 0.9506173133850098, 0.9629629850387573, 0.9876543283462524, 0.9876543283462524, 0.9629629850387573, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9506173133850098, 0.9753086566925049, 0.9876543283462524, 1.0, 0.9876543283462524, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9629629850387573, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5432098507881165, 0.09876543283462524, 0.5308641791343689, NaN, 0.7283950448036194, 0.5308641791343689, 0.654321014881134, 0.6790123581886292, 0.6666666865348816, 0.7160493731498718, 0.654321014881134, 0.7530864477157593, 0.7654321193695068, 0.7160493731498718, 0.7037037014961243, 0.8765432238578796, 0.7407407164573669, 0.7283950448036194, 0.8518518805503845, 0.7530864477157593, 0.8395061492919922, 0.7407407164573669, 0.8641975522041321, 0.8641975522041321, 0.8888888955116272, 0.8518518805503845, 0.8518518805503845, 0.8641975522041321, 0.8765432238578796, 0.8148148059844971, 0.9506173133850098, 0.7654321193695068, 0.8518518805503845, 0.9012345671653748, 0.9876543283462524, 0.9629629850387573, 0.7777777910232544, 0.9382715821266174, 0.8888888955116272, 0.9382715821266174, 0.9012345671653748, 0.9629629850387573, 0.9629629850387573, 0.9506173133850098, 0.9876543283462524, 0.9876543283462524, 0.9753086566925049, 0.9382715821266174, 0.9135802388191223, 0.9629629850387573, 0.9629629850387573, 0.9259259104728699, 0.9629629850387573, 0.8888888955116272, 0.9629629850387573, 0.9629629850387573, 0.9259259104728699, 0.9135802388191223, 0.9259259104728699, 1.0, 1.0, 0.9259259104728699, 1.0, 1.0, 1.0, 0.9506173133850098, 1.0, 1.0, 0.9135802388191223, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1975308656692505, 0.04938271641731262, 0.32098764181137085, 0.2716049253940582, NaN, 0.5185185074806213, 0.4691357910633087, 0.5061728358268738, 0.5185185074806213, 0.5802469253540039, 0.5185185074806213, 0.6419752836227417, 0.5802469253540039, 0.5308641791343689, 0.654321014881134, 0.6666666865348816, 0.6790123581886292, 0.6790123581886292, 0.7777777910232544, 0.7037037014961243, 0.7160493731498718, 0.6419752836227417, 0.7777777910232544, 0.8395061492919922, 0.8024691343307495, 0.7530864477157593, 0.7283950448036194, 0.8024691343307495, 0.8395061492919922, 0.8271604776382446, 0.8395061492919922, 0.7407407164573669, 0.8641975522041321, 0.9012345671653748, 0.9135802388191223, 0.8518518805503845, 0.790123462677002, 1.0, 0.8518518805503845, 0.8888888955116272, 0.9629629850387573, 0.9506173133850098, 0.9506173133850098, 0.9629629850387573, 0.9382715821266174, 0.9382715821266174, 0.9753086566925049, 0.9506173133850098, 0.9506173133850098, 0.9753086566925049, 0.9876543283462524, 0.9382715821266174, 0.9382715821266174, 0.8641975522041321, 1.0, 0.9876543283462524, 0.9135802388191223, 0.9629629850387573, 0.9382715821266174, 1.0, 1.0, 0.9629629850387573, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9629629850387573, 1.0, 1.0, 0.9135802388191223, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3580246865749359, 0.09876543283462524, 0.3827160596847534, 0.4691357910633087, 0.48148149251937866, NaN, 0.5679012537002563, 0.5925925970077515, 0.5679012537002563, 0.5308641791343689, 0.5679012537002563, 0.5802469253540039, 0.5679012537002563, 0.5802469253540039, 0.654321014881134, 0.6790123581886292, 0.5802469253540039, 0.5925925970077515, 0.8148148059844971, 0.604938268661499, 0.6913580298423767, 0.604938268661499, 0.8395061492919922, 0.8271604776382446, 0.7530864477157593, 0.8148148059844971, 0.7283950448036194, 0.7777777910232544, 0.8888888955116272, 0.790123462677002, 0.8641975522041321, 0.6666666865348816, 0.7530864477157593, 0.8518518805503845, 0.9382715821266174, 0.9012345671653748, 0.6913580298423767, 0.9012345671653748, 0.8271604776382446, 0.9135802388191223, 0.9012345671653748, 0.9012345671653748, 0.9382715821266174, 0.9259259104728699, 0.9382715821266174, 0.9135802388191223, 0.9506173133850098, 0.9629629850387573, 0.9135802388191223, 0.9876543283462524, 0.9629629850387573, 0.9259259104728699, 0.9506173133850098, 0.9012345671653748, 0.9753086566925049, 0.9876543283462524, 0.9135802388191223, 0.9135802388191223, 0.9506173133850098, 0.9753086566925049, 0.9753086566925049, 0.9382715821266174, 1.0, 1.0, 1.0, 0.9629629850387573, 1.0, 1.0, 0.9135802388191223, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.23456789553165436, 0.3086419701576233, 0.34567901492118835, 0.5308641791343689, 0.43209877610206604, NaN, 0.5308641791343689, 0.45679011940956116, 0.5802469253540039, 0.5308641791343689, 0.5925925970077515, 0.5555555820465088, 0.5925925970077515, 0.604938268661499, 0.5925925970077515, 0.7160493731498718, 0.6419752836227417, 0.5802469253540039, 0.7283950448036194, 0.6172839403152466, 0.790123462677002, 0.604938268661499, 0.7160493731498718, 0.7654321193695068, 0.7037037014961243, 0.7283950448036194, 0.6913580298423767, 0.654321014881134, 0.7654321193695068, 0.654321014881134, 0.9382715821266174, 0.9012345671653748, 0.8765432238578796, 0.7654321193695068, 0.790123462677002, 0.9506173133850098, 0.8024691343307495, 0.8148148059844971, 0.8395061492919922, 0.9012345671653748, 0.9506173133850098, 0.790123462677002, 0.9259259104728699, 0.9259259104728699, 0.9382715821266174, 0.9876543283462524, 0.9753086566925049, 0.9506173133850098, 0.9629629850387573, 0.9629629850387573, 0.9753086566925049, 0.8888888955116272, 0.8395061492919922, 0.9753086566925049, 0.9876543283462524, 0.9506173133850098, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9876543283462524, 0.9876543283462524, 0.9753086566925049, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.07407407462596893, 0.17283950746059418, 0.17283950746059418, 0.32098764181137085, 0.4938271641731262, 0.40740740299224854, 0.4691357910633087, NaN, 0.45679011940956116, 0.4938271641731262, 0.4691357910633087, 0.4938271641731262, 0.5308641791343689, 0.5925925970077515, 0.6296296119689941, 0.5679012537002563, 0.5802469253540039, 0.6172839403152466, 0.6419752836227417, 0.7530864477157593, 0.6172839403152466, 0.604938268661499, 0.6913580298423767, 0.7407407164573669, 0.7530864477157593, 0.7160493731498718, 0.6790123581886292, 0.7654321193695068, 0.7160493731498718, 0.8518518805503845, 0.6790123581886292, 0.8765432238578796, 0.9012345671653748, 0.8765432238578796, 0.8148148059844971, 0.7407407164573669, 0.9506173133850098, 0.9506173133850098, 0.9629629850387573, 0.8518518805503845, 0.9135802388191223, 0.9506173133850098, 0.9012345671653748, 0.9382715821266174, 0.9135802388191223, 1.0, 0.9753086566925049, 0.9753086566925049, 0.9753086566925049, 0.9506173133850098, 0.9876543283462524, 0.9629629850387573, 0.9259259104728699, 0.8395061492919922, 1.0, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9629629850387573, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.08641975373029709, 0.2469135820865631, 0.25925925374031067, 0.3333333432674408, 0.48148149251937866, 0.43209877610206604, 0.5432098507881165, 0.5432098507881165, NaN, 0.5802469253540039, 0.5802469253540039, 0.5432098507881165, 0.5555555820465088, 0.6296296119689941, 0.5679012537002563, 0.5432098507881165, 0.7037037014961243, 0.6790123581886292, 0.5802469253540039, 0.7037037014961243, 0.604938268661499, 0.7654321193695068, 0.5925925970077515, 0.7283950448036194, 0.7037037014961243, 0.654321014881134, 0.7160493731498718, 0.709876537322998, 0.6296296119689941, 0.7654321193695068, 0.6296296119689941, 0.9135802388191223, 0.9259259104728699, 0.8518518805503845, 0.7407407164573669, 0.7530864477157593, 0.9382715821266174, 0.8148148059844971, 0.8518518805503845, 0.8271604776382446, 0.9012345671653748, 0.9506173133850098, 0.790123462677002, 0.9382715821266174, 0.9506173133850098, 0.9135802388191223, 0.9506173133850098, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9629629850387573, 0.9629629850387573, 0.8765432238578796, 0.8024691343307495, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0], [0.02469135820865631, 0.1975308656692505, 0.14814814925193787, 0.2839506268501282, 0.4197530746459961, 0.4691357910633087, 0.4197530746459961, 0.5061728358268738, 0.4197530746459961, NaN, 0.4691357910633087, 0.4197530746459961, 0.4691357910633087, 0.5555555820465088, 0.5802469253540039, 0.5925925970077515, 0.5432098507881165, 0.7530864477157593, 0.6172839403152466, 0.6666666865348816, 0.5308641791343689, 0.5432098507881165, 0.654321014881134, 0.6790123581886292, 0.7530864477157593, 0.6419752836227417, 0.6913580298423767, 0.7654321193695068, 0.6790123581886292, 0.8395061492919922, 0.7407407164573669, 0.8271604776382446, 0.8395061492919922, 0.8271604776382446, 0.8888888955116272, 0.7283950448036194, 0.8765432238578796, 1.0, 0.9753086566925049, 0.8395061492919922, 0.9135802388191223, 0.9259259104728699, 0.9012345671653748, 0.9382715821266174, 0.9506173133850098, 0.9012345671653748, 0.9506173133850098, 0.9753086566925049, 0.9629629850387573, 1.0, 0.9629629850387573, 0.9753086566925049, 0.9012345671653748, 0.9382715821266174, 0.9876543283462524, 0.9753086566925049, 0.9629629850387573, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.09876543283462524, 0.1975308656692505, 0.2469135820865631, 0.34567901492118835, 0.48148149251937866, 0.43209877610206604, 0.4691357910633087, 0.5308641791343689, 0.4197530746459961, 0.5308641791343689, NaN, 0.5802469253540039, 0.5679012537002563, 0.6296296119689941, 0.5185185074806213, 0.5555555820465088, 0.7037037014961243, 0.6666666865348816, 0.5432098507881165, 0.7283950448036194, 0.5802469253540039, 0.7283950448036194, 0.5925925970077515, 0.6790123581886292, 0.7160493731498718, 0.6666666865348816, 0.6790123581886292, 0.654321014881134, 0.6296296119689941, 0.7530864477157593, 0.6172839403152466, 0.9259259104728699, 0.9135802388191223, 0.8395061492919922, 0.7777777910232544, 0.7407407164573669, 0.9259259104728699, 0.8271604776382446, 0.8024691343307495, 0.8395061492919922, 0.9012345671653748, 0.9382715821266174, 0.7777777910232544, 0.9629629850387573, 0.9135802388191223, 0.9259259104728699, 0.9259259104728699, 0.9753086566925049, 0.9629629850387573, 0.9506173133850098, 0.9629629850387573, 0.9506173133850098, 0.8641975522041321, 0.8024691343307495, 0.9753086566925049, 0.9629629850387573, 0.9382715821266174, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9753086566925049, 0.9876543283462524, 0.9753086566925049, 1.0, 0.9876543283462524, 0.9753086566925049, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.08641975373029709, 0.1358024626970291, 0.2469135820865631, 0.2469135820865631, 0.3580246865749359, 0.4197530746459961, 0.40740740299224854, 0.5061728358268738, 0.45679011940956116, 0.5802469253540039, 0.4197530746459961, NaN, 0.5679012537002563, 0.5308641791343689, 0.5802469253540039, 0.5061728358268738, 0.7777777910232544, 0.6913580298423767, 0.5679012537002563, 0.654321014881134, 0.654321014881134, 0.8641975522041321, 0.6296296119689941, 0.6296296119689941, 0.6296296119689941, 0.6172839403152466, 0.7037037014961243, 0.6913580298423767, 0.6666666865348816, 0.6790123581886292, 0.6790123581886292, 0.8888888955116272, 0.9135802388191223, 0.7530864477157593, 0.8024691343307495, 0.6296296119689941, 0.9135802388191223, 0.8395061492919922, 0.8395061492919922, 0.8641975522041321, 0.9506173133850098, 0.8518518805503845, 0.8148148059844971, 0.9753086566925049, 0.8641975522041321, 0.9506173133850098, 0.9012345671653748, 0.9876543283462524, 0.9753086566925049, 0.8765432238578796, 0.9876543283462524, 0.8765432238578796, 0.8148148059844971, 0.8271604776382446, 0.9012345671653748, 0.8765432238578796, 0.8641975522041321, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 0.8641975522041321, 0.8765432238578796, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9135802388191223, 1.0], [0.20987653732299805, 0.012345679104328156, 0.2839506268501282, 0.23456789553165436, 0.4197530746459961, 0.43209877610206604, 0.4444444477558136, 0.4691357910633087, 0.4444444477558136, 0.5308641791343689, 0.43209877610206604, 0.43209877610206604, NaN, 0.4197530746459961, 0.5185185074806213, 0.5802469253540039, 0.5061728358268738, 0.6172839403152466, 0.6296296119689941, 0.5555555820465088, 0.7283950448036194, 0.5802469253540039, 0.654321014881134, 0.6296296119689941, 0.6790123581886292, 0.6419752836227417, 0.7037037014961243, 0.8148148059844971, 0.7283950448036194, 0.6666666865348816, 0.790123462677002, 0.6172839403152466, 0.7407407164573669, 0.7407407164573669, 0.9629629850387573, 0.8395061492919922, 0.6172839403152466, 0.9753086566925049, 0.8024691343307495, 0.9012345671653748, 0.9382715821266174, 0.8641975522041321, 0.8888888955116272, 0.9259259104728699, 1.0, 0.9259259104728699, 0.9629629850387573, 0.9259259104728699, 0.9012345671653748, 0.9259259104728699, 0.9259259104728699, 0.9012345671653748, 0.8765432238578796, 0.8518518805503845, 0.9753086566925049, 0.9506173133850098, 0.8888888955116272, 0.9259259104728699, 0.9382715821266174, 1.0, 1.0, 0.9382715821266174, 0.9753086566925049, 1.0, 1.0, 0.9753086566925049, 1.0, 1.0, 0.9012345671653748, 1.0, 1.0, 1.0, 1.0, 1.0], [0.04938271641731262, 0.1358024626970291, 0.14814814925193787, 0.2839506268501282, 0.4691357910633087, 0.4197530746459961, 0.40740740299224854, 0.40740740299224854, 0.37037035822868347, 0.4444444477558136, 0.37037035822868347, 0.4691357910633087, 0.5802469253540039, NaN, 0.5061728358268738, 0.4691357910633087, 0.5308641791343689, 0.5432098507881165, 0.5555555820465088, 0.8641975522041321, 0.6419752836227417, 0.6296296119689941, 0.5925925970077515, 0.6419752836227417, 0.6666666865348816, 0.5555555820465088, 0.5925925970077515, 0.604938268661499, 0.654321014881134, 0.7777777910232544, 0.654321014881134, 0.8024691343307495, 0.9012345671653748, 0.9382715821266174, 0.790123462677002, 0.654321014881134, 0.9012345671653748, 0.8271604776382446, 0.7777777910232544, 0.8641975522041321, 0.9382715821266174, 0.8271604776382446, 0.8271604776382446, 0.9753086566925049, 0.9629629850387573, 0.9629629850387573, 0.9876543283462524, 1.0, 0.9506173133850098, 0.9876543283462524, 0.9876543283462524, 0.8641975522041321, 0.8765432238578796, 0.7530864477157593, 1.0, 0.9876543283462524, 0.8888888955116272, 1.0, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9135802388191223, 0.9629629850387573, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07407407462596893, 0.16049382090568542, 0.17283950746059418, 0.29629629850387573, 0.34567901492118835, 0.34567901492118835, 0.395061731338501, 0.37037035822868347, 0.43209877610206604, 0.4197530746459961, 0.48148149251937866, 0.4197530746459961, 0.48148149251937866, 0.4938271641731262, NaN, 0.45679011940956116, 0.5679012537002563, 0.5185185074806213, 0.5185185074806213, 0.654321014881134, 0.5925925970077515, 0.5802469253540039, 0.5555555820465088, 0.5802469253540039, 0.8641975522041321, 0.6296296119689941, 0.6666666865348816, 0.6172839403152466, 0.5679012537002563, 0.7037037014961243, 0.6913580298423767, 0.7407407164573669, 0.8148148059844971, 0.8024691343307495, 0.7777777910232544, 0.7160493731498718, 0.8148148059844971, 0.8024691343307495, 0.7407407164573669, 0.8518518805503845, 0.9012345671653748, 0.9012345671653748, 0.7777777910232544, 0.9629629850387573, 0.9753086566925049, 0.9753086566925049, 0.9753086566925049, 0.9629629850387573, 0.9506173133850098, 0.9753086566925049, 0.9506173133850098, 0.9012345671653748, 0.9506173133850098, 0.7654321193695068, 0.9753086566925049, 0.9753086566925049, 0.8518518805503845, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 0.9876543283462524, 0.9506173133850098, 1.0, 1.0, 1.0, 1.0, 1.0], [0.03703703731298447, 0.12345679104328156, 0.23456789553165436, 0.12345679104328156, 0.3333333432674408, 0.32098764181137085, 0.40740740299224854, 0.43209877610206604, 0.45679011940956116, 0.40740740299224854, 0.4444444477558136, 0.4938271641731262, 0.4197530746459961, 0.5308641791343689, 0.5432098507881165, NaN, 0.5925925970077515, 0.5432098507881165, 0.5925925970077515, 0.5555555820465088, 0.5185185074806213, 0.5925925970077515, 0.604938268661499, 0.6172839403152466, 0.5925925970077515, 0.604938268661499, 0.604938268661499, 0.7160493731498718, 0.654321014881134, 0.7037037014961243, 0.6790123581886292, 0.6790123581886292, 0.8148148059844971, 0.7283950448036194, 0.8024691343307495, 0.6666666865348816, 0.7530864477157593, 0.8765432238578796, 0.8518518805503845, 0.8024691343307495, 0.8395061492919922, 0.8271604776382446, 0.8641975522041321, 0.8518518805503845, 0.9012345671653748, 0.8271604776382446, 0.9506173133850098, 0.8765432238578796, 0.8395061492919922, 0.8888888955116272, 0.8641975522041321, 0.9259259104728699, 0.9135802388191223, 0.9506173133850098, 0.8765432238578796, 0.8765432238578796, 0.9629629850387573, 0.8888888955116272, 0.9876543283462524, 0.9135802388191223, 0.9135802388191223, 0.9012345671653748, 0.9876543283462524, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 1.0, 0.9506173133850098, 1.0, 1.0, 0.9259259104728699, 1.0, 1.0, 1.0], [0.07407407462596893, 0.1358024626970291, 0.14814814925193787, 0.25925925374031067, 0.32098764181137085, 0.4197530746459961, 0.2839506268501282, 0.4197530746459961, 0.29629629850387573, 0.45679011940956116, 0.29629629850387573, 0.2222222238779068, 0.4938271641731262, 0.4691357910633087, 0.43209877610206604, 0.40740740299224854, NaN, 0.6419752836227417, 0.5308641791343689, 0.5802469253540039, 0.5555555820465088, 0.6790123581886292, 0.5432098507881165, 0.5679012537002563, 0.5679012537002563, 0.5432098507881165, 0.604938268661499, 0.6419752836227417, 0.6172839403152466, 0.604938268661499, 0.5802469253540039, 0.8518518805503845, 0.8024691343307495, 0.7283950448036194, 0.7654321193695068, 0.5925925970077515, 0.8888888955116272, 0.8271604776382446, 0.790123462677002, 0.8395061492919922, 0.8641975522041321, 0.8024691343307495, 0.790123462677002, 0.9012345671653748, 0.790123462677002, 0.8888888955116272, 0.8395061492919922, 0.9629629850387573, 0.9259259104728699, 0.8518518805503845, 0.9629629850387573, 0.8395061492919922, 0.7530864477157593, 0.8148148059844971, 0.9012345671653748, 0.8518518805503845, 0.8518518805503845, 1.0, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 1.0, 0.8641975522041321, 0.8641975522041321, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 0.9259259104728699, 1.0], [0.03703703731298447, 0.18518517911434174, 0.08641975373029709, 0.2716049253940582, 0.32098764181137085, 0.40740740299224854, 0.3580246865749359, 0.3827160596847534, 0.32098764181137085, 0.2469135820865631, 0.3333333432674408, 0.3086419701576233, 0.3827160596847534, 0.45679011940956116, 0.48148149251937866, 0.45679011940956116, 0.3580246865749359, NaN, 0.4938271641731262, 0.5925925970077515, 0.4444444477558136, 0.48148149251937866, 0.5308641791343689, 0.5802469253540039, 0.6666666865348816, 0.5308641791343689, 0.6419752836227417, 0.6419752836227417, 0.6172839403152466, 0.7530864477157593, 0.5432098507881165, 0.7283950448036194, 0.7160493731498718, 0.7530864477157593, 0.7407407164573669, 0.604938268661499, 0.7777777910232544, 0.8888888955116272, 0.8641975522041321, 0.7530864477157593, 0.790123462677002, 0.9012345671653748, 0.8148148059844971, 0.8148148059844971, 0.8395061492919922, 0.8148148059844971, 0.8395061492919922, 0.9382715821266174, 0.8765432238578796, 0.8888888955116272, 0.9135802388191223, 0.9629629850387573, 0.8271604776382446, 0.8888888955116272, 0.9259259104728699, 0.9506173133850098, 0.9506173133850098, 0.9753086566925049, 0.9629629850387573, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9629629850387573, 0.9506173133850098, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 1.0, 0.9753086566925049, 1.0, 1.0, 1.0], [0.08641975373029709, 0.03703703731298447, 0.29629629850387573, 0.14814814925193787, 0.2222222238779068, 0.18518517911434174, 0.4197530746459961, 0.3580246865749359, 0.4197530746459961, 0.3827160596847534, 0.45679011940956116, 0.43209877610206604, 0.37037035822868347, 0.4444444477558136, 0.48148149251937866, 0.40740740299224854, 0.4691357910633087, 0.5061728358268738, NaN, 0.4938271641731262, 0.5061728358268738, 0.5679012537002563, 0.7777777910232544, 0.6790123581886292, 0.604938268661499, 0.6666666865348816, 0.5802469253540039, 0.654321014881134, 0.654321014881134, 0.6790123581886292, 0.604938268661499, 0.6172839403152466, 0.7037037014961243, 0.7407407164573669, 0.7160493731498718, 0.8395061492919922, 0.6790123581886292, 0.8024691343307495, 0.7037037014961243, 0.7777777910232544, 0.8024691343307495, 0.8024691343307495, 0.8765432238578796, 0.8518518805503845, 0.9135802388191223, 0.8395061492919922, 0.8024691343307495, 0.9012345671653748, 0.8395061492919922, 0.9629629850387573, 0.9506173133850098, 0.8641975522041321, 0.8888888955116272, 0.7407407164573669, 0.9506173133850098, 0.9629629850387573, 0.8888888955116272, 0.8271604776382446, 0.9012345671653748, 1.0, 1.0, 0.8395061492919922, 0.9629629850387573, 0.9753086566925049, 0.9876543283462524, 0.9506173133850098, 1.0, 0.9876543283462524, 0.9135802388191223, 1.0, 1.0, 1.0, 1.0, 1.0], [0.012345679104328156, 0.09876543283462524, 0.12345679104328156, 0.2469135820865631, 0.29629629850387573, 0.395061731338501, 0.2716049253940582, 0.2469135820865631, 0.29629629850387573, 0.3333333432674408, 0.2716049253940582, 0.34567901492118835, 0.4444444477558136, 0.1358024626970291, 0.34567901492118835, 0.4444444477558136, 0.4197530746459961, 0.40740740299224854, 0.5061728358268738, NaN, 0.5679012537002563, 0.5555555820465088, 0.5185185074806213, 0.5555555820465088, 0.5679012537002563, 0.4938271641731262, 0.5185185074806213, 0.5925925970077515, 0.5679012537002563, 0.6419752836227417, 0.604938268661499, 0.7160493731498718, 0.8024691343307495, 0.8148148059844971, 0.7530864477157593, 0.6296296119689941, 0.790123462677002, 0.8024691343307495, 0.7283950448036194, 0.8395061492919922, 0.9012345671653748, 0.8024691343307495, 0.7777777910232544, 0.9135802388191223, 0.9259259104728699, 0.9259259104728699, 0.9629629850387573, 0.9629629850387573, 0.9382715821266174, 0.9506173133850098, 0.9382715821266174, 0.8518518805503845, 0.8395061492919922, 0.7283950448036194, 1.0, 0.9753086566925049, 0.8395061492919922, 1.0, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9012345671653748, 0.9629629850387573, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9506173133850098, 1.0, 1.0, 1.0, 1.0, 1.0], [0.14814814925193787, 0.0, 0.2222222238779068, 0.16049382090568542, 0.2839506268501282, 0.3086419701576233, 0.3827160596847534, 0.3827160596847534, 0.395061731338501, 0.4691357910633087, 0.4197530746459961, 0.34567901492118835, 0.2716049253940582, 0.3580246865749359, 0.40740740299224854, 0.48148149251937866, 0.4444444477558136, 0.5555555820465088, 0.4938271641731262, 0.43209877610206604, NaN, 0.45679011940956116, 0.5185185074806213, 0.5432098507881165, 0.4938271641731262, 0.5432098507881165, 0.654321014881134, 0.7037037014961243, 0.604938268661499, 0.6172839403152466, 0.654321014881134, 0.5679012537002563, 0.6172839403152466, 0.7283950448036194, 0.8271604776382446, 0.7283950448036194, 0.6172839403152466, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8518518805503845, 0.8271604776382446, 0.8024691343307495, 0.8765432238578796, 0.8641975522041321, 0.8395061492919922, 0.9012345671653748, 0.8641975522041321, 0.8518518805503845, 0.8518518805503845, 0.8765432238578796, 0.8641975522041321, 0.8271604776382446, 0.8518518805503845, 0.9753086566925049, 0.9135802388191223, 0.9012345671653748, 0.8765432238578796, 0.8765432238578796, 1.0, 1.0, 0.8888888955116272, 0.9753086566925049, 0.9876543283462524, 0.9753086566925049, 0.9259259104728699, 0.9876543283462524, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.03703703731298447, 0.1111111119389534, 0.12345679104328156, 0.25925925374031067, 0.3580246865749359, 0.395061731338501, 0.20987653732299805, 0.395061731338501, 0.23456789553165436, 0.45679011940956116, 0.2716049253940582, 0.1358024626970291, 0.4197530746459961, 0.37037035822868347, 0.4197530746459961, 0.40740740299224854, 0.32098764181137085, 0.5185185074806213, 0.43209877610206604, 0.4444444477558136, 0.5432098507881165, NaN, 0.48148149251937866, 0.48148149251937866, 0.5061728358268738, 0.5061728358268738, 0.6172839403152466, 0.604938268661499, 0.5555555820465088, 0.604938268661499, 0.5925925970077515, 0.7777777910232544, 0.7283950448036194, 0.6790123581886292, 0.6419752836227417, 0.604938268661499, 0.8641975522041321, 0.7283950448036194, 0.790123462677002, 0.7283950448036194, 0.8024691343307495, 0.7530864477157593, 0.8148148059844971, 0.8518518805503845, 0.7283950448036194, 0.9135802388191223, 0.790123462677002, 0.9382715821266174, 0.9506173133850098, 0.8518518805503845, 0.9012345671653748, 0.8518518805503845, 0.8024691343307495, 0.8148148059844971, 0.8518518805503845, 0.8395061492919922, 0.8518518805503845, 1.0, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 0.8641975522041321, 0.8518518805503845, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0], [0.08641975373029709, 0.03703703731298447, 0.2469135820865631, 0.1358024626970291, 0.2222222238779068, 0.16049382090568542, 0.395061731338501, 0.3086419701576233, 0.40740740299224854, 0.34567901492118835, 0.40740740299224854, 0.37037035822868347, 0.34567901492118835, 0.40740740299224854, 0.4444444477558136, 0.395061731338501, 0.45679011940956116, 0.4691357910633087, 0.2222222238779068, 0.48148149251937866, 0.48148149251937866, 0.5185185074806213, NaN, 0.6296296119689941, 0.604938268661499, 0.6296296119689941, 0.5308641791343689, 0.5679012537002563, 0.654321014881134, 0.6172839403152466, 0.5432098507881165, 0.6172839403152466, 0.6296296119689941, 0.7283950448036194, 0.6419752836227417, 0.8148148059844971, 0.654321014881134, 0.7407407164573669, 0.6666666865348816, 0.7283950448036194, 0.7037037014961243, 0.790123462677002, 0.8024691343307495, 0.8148148059844971, 0.8765432238578796, 0.8024691343307495, 0.7654321193695068, 0.8395061492919922, 0.8148148059844971, 0.9012345671653748, 0.9259259104728699, 0.8641975522041321, 0.8271604776382446, 0.7283950448036194, 0.9012345671653748, 0.9259259104728699, 0.8765432238578796, 0.8024691343307495, 0.8765432238578796, 0.9876543283462524, 1.0, 0.8518518805503845, 0.9629629850387573, 0.9382715821266174, 0.9753086566925049, 0.9259259104728699, 0.9876543283462524, 0.9876543283462524, 0.8888888955116272, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.08641975373029709, 0.02469135820865631, 0.25925925374031067, 0.1358024626970291, 0.16049382090568542, 0.17283950746059418, 0.2839506268501282, 0.25925925374031067, 0.2716049253940582, 0.32098764181137085, 0.32098764181137085, 0.37037035822868347, 0.37037035822868347, 0.3580246865749359, 0.4197530746459961, 0.3827160596847534, 0.43209877610206604, 0.4197530746459961, 0.32098764181137085, 0.4444444477558136, 0.45679011940956116, 0.5185185074806213, 0.37037035822868347, NaN, 0.5555555820465088, 0.5802469253540039, 0.5802469253540039, 0.5061728358268738, 0.5802469253540039, 0.5802469253540039, 0.4938271641731262, 0.6419752836227417, 0.5925925970077515, 0.7283950448036194, 0.5432098507881165, 0.8518518805503845, 0.654321014881134, 0.6790123581886292, 0.6296296119689941, 0.7654321193695068, 0.7037037014961243, 0.7777777910232544, 0.7530864477157593, 0.7530864477157593, 0.8641975522041321, 0.8148148059844971, 0.7777777910232544, 0.8271604776382446, 0.8888888955116272, 0.9135802388191223, 0.8888888955116272, 0.7777777910232544, 0.8395061492919922, 0.7407407164573669, 0.9506173133850098, 0.9382715821266174, 0.8271604776382446, 0.8271604776382446, 0.8888888955116272, 1.0, 1.0, 0.8271604776382446, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 0.9135802388191223, 0.9753086566925049, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.02469135820865631, 0.03703703731298447, 0.1111111119389534, 0.1111111119389534, 0.1975308656692505, 0.2469135820865631, 0.23456789553165436, 0.2469135820865631, 0.29629629850387573, 0.2469135820865631, 0.2839506268501282, 0.37037035822868347, 0.32098764181137085, 0.3333333432674408, 0.1358024626970291, 0.40740740299224854, 0.43209877610206604, 0.3333333432674408, 0.395061731338501, 0.43209877610206604, 0.5061728358268738, 0.4938271641731262, 0.395061731338501, 0.4444444477558136, NaN, 0.4691357910633087, 0.4938271641731262, 0.5061728358268738, 0.5308641791343689, 0.5308641791343689, 0.6172839403152466, 0.6913580298423767, 0.7283950448036194, 0.7160493731498718, 0.7283950448036194, 0.5925925970077515, 0.6913580298423767, 0.7407407164573669, 0.7283950448036194, 0.7654321193695068, 0.7654321193695068, 0.6666666865348816, 0.7160493731498718, 0.8395061492919922, 0.9629629850387573, 0.9259259104728699, 0.9259259104728699, 0.8518518805503845, 0.8024691343307495, 0.8148148059844971, 0.8765432238578796, 0.7654321193695068, 0.8148148059844971, 0.7160493731498718, 0.9259259104728699, 0.9135802388191223, 0.790123462677002, 0.8518518805503845, 0.9012345671653748, 1.0, 1.0, 0.9012345671653748, 0.9382715821266174, 0.9629629850387573, 0.9876543283462524, 0.9629629850387573, 0.9876543283462524, 0.9876543283462524, 0.9012345671653748, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.1111111119389534, 0.04938271641731262, 0.25925925374031067, 0.14814814925193787, 0.2469135820865631, 0.18518517911434174, 0.29629629850387573, 0.2839506268501282, 0.34567901492118835, 0.3580246865749359, 0.3333333432674408, 0.3827160596847534, 0.3580246865749359, 0.4444444477558136, 0.37037035822868347, 0.395061731338501, 0.45679011940956116, 0.4691357910633087, 0.3333333432674408, 0.5061728358268738, 0.45679011940956116, 0.4938271641731262, 0.37037035822868347, 0.4197530746459961, 0.5308641791343689, NaN, 0.48148149251937866, 0.45679011940956116, 0.604938268661499, 0.5925925970077515, 0.5925925970077515, 0.6172839403152466, 0.6913580298423767, 0.6790123581886292, 0.5679012537002563, 0.6172839403152466, 0.6666666865348816, 0.6419752836227417, 0.654321014881134, 0.654321014881134, 0.7037037014961243, 0.6419752836227417, 0.8395061492919922, 0.7283950448036194, 0.8518518805503845, 0.7407407164573669, 0.8518518805503845, 0.790123462677002, 0.7530864477157593, 0.7777777910232544, 0.8024691343307495, 0.6790123581886292, 0.9135802388191223, 0.7654321193695068, 0.7777777910232544, 0.8148148059844971, 0.7654321193695068, 0.7283950448036194, 0.8395061492919922, 0.9259259104728699, 0.9382715821266174, 0.7283950448036194, 0.9753086566925049, 1.0, 1.0, 0.9259259104728699, 1.0, 1.0, 0.9259259104728699, 1.0, 0.9753086566925049, 1.0, 1.0, 1.0], [0.12345679104328156, 0.1111111119389534, 0.16049382090568542, 0.14814814925193787, 0.2716049253940582, 0.2716049253940582, 0.2716049253940582, 0.32098764181137085, 0.2839506268501282, 0.3086419701576233, 0.32098764181137085, 0.29629629850387573, 0.29629629850387573, 0.40740740299224854, 0.3333333432674408, 0.395061731338501, 0.395061731338501, 0.3580246865749359, 0.4197530746459961, 0.48148149251937866, 0.34567901492118835, 0.3827160596847534, 0.4691357910633087, 0.4197530746459961, 0.5061728358268738, 0.5185185074806213, NaN, 0.48148149251937866, 0.5061728358268738, 0.4444444477558136, 0.5555555820465088, 0.6790123581886292, 0.5679012537002563, 0.5555555820465088, 0.5802469253540039, 0.5679012537002563, 0.6790123581886292, 0.6172839403152466, 0.7037037014961243, 0.6172839403152466, 0.6419752836227417, 0.5802469253540039, 0.7654321193695068, 0.654321014881134, 0.7777777910232544, 0.7407407164573669, 0.8271604776382446, 0.8148148059844971, 0.8271604776382446, 0.8148148059844971, 0.8395061492919922, 0.8395061492919922, 0.8395061492919922, 0.8518518805503845, 0.8765432238578796, 0.8518518805503845, 0.8888888955116272, 0.8888888955116272, 0.9876543283462524, 0.9382715821266174, 0.9382715821266174, 0.9259259104728699, 0.9259259104728699, 0.9135802388191223, 0.9629629850387573, 1.0, 0.9629629850387573, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0], [0.012345679104328156, 0.06172839552164078, 0.1111111119389534, 0.1358024626970291, 0.1975308656692505, 0.2222222238779068, 0.3086419701576233, 0.23456789553165436, 0.29012346267700195, 0.23456789553165436, 0.34567901492118835, 0.3086419701576233, 0.18518517911434174, 0.395061731338501, 0.3827160596847534, 0.2839506268501282, 0.3580246865749359, 0.3580246865749359, 0.34567901492118835, 0.40740740299224854, 0.29629629850387573, 0.395061731338501, 0.43209877610206604, 0.4938271641731262, 0.4938271641731262, 0.5432098507881165, 0.5185185074806213, NaN, 0.5308641791343689, 0.5679012537002563, 0.5925925970077515, 0.5061728358268738, 0.5802469253540039, 0.5925925970077515, 0.5802469253540039, 0.604938268661499, 0.5802469253540039, 0.6913580298423767, 0.6296296119689941, 0.654321014881134, 0.654321014881134, 0.6913580298423767, 0.7530864477157593, 0.6913580298423767, 0.7407407164573669, 0.7407407164573669, 0.8888888955116272, 0.8024691343307495, 0.7654321193695068, 0.8271604776382446, 0.790123462677002, 0.9135802388191223, 0.8518518805503845, 0.8024691343307495, 0.8395061492919922, 0.8395061492919922, 0.9012345671653748, 0.8518518805503845, 0.9506173133850098, 0.9506173133850098, 0.9506173133850098, 0.8888888955116272, 0.9382715821266174, 0.9506173133850098, 0.9629629850387573, 0.9876543283462524, 0.9753086566925049, 0.9876543283462524, 0.9506173133850098, 1.0, 0.9753086566925049, 1.0, 1.0, 1.0], [0.08641975373029709, 0.03703703731298447, 0.2222222238779068, 0.12345679104328156, 0.16049382090568542, 0.1111111119389534, 0.34567901492118835, 0.2839506268501282, 0.37037035822868347, 0.32098764181137085, 0.37037035822868347, 0.3333333432674408, 0.2716049253940582, 0.34567901492118835, 0.43209877610206604, 0.34567901492118835, 0.3827160596847534, 0.3827160596847534, 0.34567901492118835, 0.43209877610206604, 0.395061731338501, 0.4444444477558136, 0.34567901492118835, 0.4197530746459961, 0.4691357910633087, 0.395061731338501, 0.4938271641731262, 0.4691357910633087, NaN, 0.5802469253540039, 0.45679011940956116, 0.5679012537002563, 0.5555555820465088, 0.604938268661499, 0.5555555820465088, 0.6419752836227417, 0.6172839403152466, 0.604938268661499, 0.654321014881134, 0.5802469253540039, 0.5925925970077515, 0.7283950448036194, 0.8765432238578796, 0.6296296119689941, 0.7530864477157593, 0.7037037014961243, 0.7654321193695068, 0.7407407164573669, 0.7160493731498718, 0.7530864477157593, 0.7283950448036194, 0.7777777910232544, 0.9382715821266174, 0.7407407164573669, 0.7530864477157593, 0.790123462677002, 0.8148148059844971, 0.7283950448036194, 0.8271604776382446, 0.8395061492919922, 0.8395061492919922, 0.7283950448036194, 0.9876543283462524, 0.9753086566925049, 0.9876543283462524, 0.8765432238578796, 0.9876543283462524, 0.9876543283462524, 0.9135802388191223, 1.0, 0.9629629850387573, 1.0, 1.0, 1.0], [0.012345679104328156, 0.03703703731298447, 0.09876543283462524, 0.18518517911434174, 0.17283950746059418, 0.20987653732299805, 0.23456789553165436, 0.14814814925193787, 0.23456789553165436, 0.16049382090568542, 0.2469135820865631, 0.32098764181137085, 0.3333333432674408, 0.2222222238779068, 0.29629629850387573, 0.29629629850387573, 0.395061731338501, 0.2469135820865631, 0.32098764181137085, 0.3580246865749359, 0.3827160596847534, 0.395061731338501, 0.3827160596847534, 0.4197530746459961, 0.4691357910633087, 0.40740740299224854, 0.5555555820465088, 0.43209877610206604, 0.4197530746459961, NaN, 0.395061731338501, 0.5555555820465088, 0.5185185074806213, 0.6296296119689941, 0.5802469253540039, 0.6419752836227417, 0.6172839403152466, 0.6172839403152466, 0.6172839403152466, 0.5555555820465088, 0.6666666865348816, 0.9259259104728699, 0.6790123581886292, 0.7037037014961243, 0.7283950448036194, 0.8271604776382446, 0.7160493731498718, 0.7777777910232544, 0.8024691343307495, 0.8765432238578796, 0.8395061492919922, 0.8148148059844971, 0.7407407164573669, 0.7407407164573669, 0.9876543283462524, 0.9012345671653748, 0.8148148059844971, 0.9012345671653748, 0.9012345671653748, 0.9135802388191223, 0.9135802388191223, 0.9012345671653748, 0.9753086566925049, 0.9629629850387573, 0.9876543283462524, 0.9506173133850098, 0.9876543283462524, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.09876543283462524, 0.03703703731298447, 0.20987653732299805, 0.04938271641731262, 0.16049382090568542, 0.1358024626970291, 0.34567901492118835, 0.32098764181137085, 0.37037035822868347, 0.25925925374031067, 0.3827160596847534, 0.32098764181137085, 0.20987653732299805, 0.34567901492118835, 0.3086419701576233, 0.32098764181137085, 0.4197530746459961, 0.45679011940956116, 0.395061731338501, 0.395061731338501, 0.34567901492118835, 0.40740740299224854, 0.45679011940956116, 0.5061728358268738, 0.3827160596847534, 0.40740740299224854, 0.4444444477558136, 0.40740740299224854, 0.5432098507881165, 0.604938268661499, NaN, 0.5802469253540039, 0.5308641791343689, 0.5925925970077515, 0.5185185074806213, 0.6790123581886292, 0.6172839403152466, 0.6666666865348816, 0.604938268661499, 0.5679012537002563, 0.5802469253540039, 0.7037037014961243, 0.8148148059844971, 0.6790123581886292, 0.7654321193695068, 0.654321014881134, 0.6666666865348816, 0.7160493731498718, 0.6666666865348816, 0.7407407164573669, 0.7777777910232544, 0.7160493731498718, 0.9012345671653748, 0.8518518805503845, 0.7777777910232544, 0.8024691343307495, 0.8641975522041321, 0.7037037014961243, 0.8395061492919922, 0.8765432238578796, 0.8765432238578796, 0.7283950448036194, 0.9629629850387573, 0.9629629850387573, 0.9135802388191223, 0.8641975522041321, 0.9506173133850098, 0.8888888955116272, 0.9012345671653748, 1.0, 0.8765432238578796, 1.0, 1.0, 1.0], [0.02469135820865631, 0.12345679104328156, 0.04938271641731262, 0.23456789553165436, 0.25925925374031067, 0.3333333432674408, 0.06172839552164078, 0.12345679104328156, 0.08641975373029709, 0.17283950746059418, 0.07407407462596893, 0.1111111119389534, 0.3827160596847534, 0.1975308656692505, 0.25925925374031067, 0.32098764181137085, 0.14814814925193787, 0.2716049253940582, 0.3827160596847534, 0.2839506268501282, 0.43209877610206604, 0.2222222238779068, 0.3827160596847534, 0.3580246865749359, 0.3086419701576233, 0.3827160596847534, 0.32098764181137085, 0.4938271641731262, 0.43209877610206604, 0.4444444477558136, 0.4197530746459961, NaN, 0.604938268661499, 0.40740740299224854, 0.5061728358268738, 0.4197530746459961, 0.6419752836227417, 0.5802469253540039, 0.7037037014961243, 0.5925925970077515, 0.6790123581886292, 0.6419752836227417, 0.5185185074806213, 0.6913580298423767, 0.5185185074806213, 0.604938268661499, 0.6296296119689941, 0.7160493731498718, 0.7654321193695068, 0.8148148059844971, 0.7283950448036194, 0.7777777910232544, 0.654321014881134, 0.7654321193695068, 0.7407407164573669, 0.8024691343307495, 0.8518518805503845, 0.9753086566925049, 0.9629629850387573, 0.8395061492919922, 0.8395061492919922, 1.0, 0.790123462677002, 0.8395061492919922, 0.9753086566925049, 1.0, 0.9753086566925049, 0.9629629850387573, 1.0, 0.8888888955116272, 0.9753086566925049, 0.9629629850387573, 0.8888888955116272, 1.0], [0.012345679104328156, 0.012345679104328156, 0.04938271641731262, 0.14814814925193787, 0.1358024626970291, 0.2469135820865631, 0.09876543283462524, 0.09876543283462524, 0.07407407462596893, 0.16049382090568542, 0.08641975373029709, 0.08641975373029709, 0.25925925374031067, 0.09876543283462524, 0.18518517911434174, 0.18518517911434174, 0.1975308656692505, 0.2839506268501282, 0.29629629850387573, 0.1975308656692505, 0.3827160596847534, 0.2716049253940582, 0.37037035822868347, 0.40740740299224854, 0.2716049253940582, 0.3086419701576233, 0.43209877610206604, 0.4197530746459961, 0.4444444477558136, 0.48148149251937866, 0.4691357910633087, 0.395061731338501, NaN, 0.5679012537002563, 0.5555555820465088, 0.5308641791343689, 0.5432098507881165, 0.6913580298423767, 0.6296296119689941, 0.654321014881134, 0.7530864477157593, 0.7654321193695068, 0.6666666865348816, 0.9135802388191223, 0.6419752836227417, 0.6913580298423767, 0.6913580298423767, 0.7407407164573669, 0.8641975522041321, 0.7777777910232544, 0.7777777910232544, 0.7283950448036194, 0.6419752836227417, 0.6666666865348816, 0.8518518805503845, 0.8271604776382446, 0.7777777910232544, 0.8888888955116272, 0.9012345671653748, 0.9753086566925049, 0.9629629850387573, 0.9135802388191223, 0.8641975522041321, 0.8518518805503845, 0.8765432238578796, 0.9382715821266174, 0.8641975522041321, 0.8641975522041321, 0.8024691343307495, 1.0, 0.9753086566925049, 1.0, 0.8888888955116272, 1.0], [0.03703703731298447, 0.02469135820865631, 0.08641975373029709, 0.09876543283462524, 0.09876543283462524, 0.14814814925193787, 0.12345679104328156, 0.12345679104328156, 0.14814814925193787, 0.17283950746059418, 0.16049382090568542, 0.2469135820865631, 0.25925925374031067, 0.06172839552164078, 0.1975308656692505, 0.2716049253940582, 0.2716049253940582, 0.2469135820865631, 0.25925925374031067, 0.18518517911434174, 0.2716049253940582, 0.32098764181137085, 0.2716049253940582, 0.2716049253940582, 0.2839506268501282, 0.32098764181137085, 0.4444444477558136, 0.40740740299224854, 0.395061731338501, 0.37037035822868347, 0.40740740299224854, 0.5925925970077515, 0.43209877610206604, NaN, 0.48148149251937866, 0.4938271641731262, 0.6419752836227417, 0.4938271641731262, 0.654321014881134, 0.5432098507881165, 0.5432098507881165, 0.6296296119689941, 0.6172839403152466, 0.604938268661499, 0.48148149251937866, 0.790123462677002, 0.6913580298423767, 0.8395061492919922, 0.8148148059844971, 0.8518518805503845, 0.8518518805503845, 0.6790123581886292, 0.6172839403152466, 0.6790123581886292, 0.9135802388191223, 0.8888888955116272, 0.7407407164573669, 0.8641975522041321, 0.8765432238578796, 0.9382715821266174, 0.9506173133850098, 0.8765432238578796, 0.8641975522041321, 0.9259259104728699, 0.9135802388191223, 0.8888888955116272, 0.9135802388191223, 0.9876543283462524, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.08641975373029709, 0.012345679104328156, 0.08641975373029709, 0.06172839552164078, 0.23456789553165436, 0.18518517911434174, 0.25925925374031067, 0.1111111119389534, 0.2222222238779068, 0.1975308656692505, 0.03703703731298447, 0.20987653732299805, 0.2222222238779068, 0.1975308656692505, 0.23456789553165436, 0.25925925374031067, 0.2839506268501282, 0.2469135820865631, 0.17283950746059418, 0.3580246865749359, 0.3580246865749359, 0.45679011940956116, 0.2716049253940582, 0.43209877610206604, 0.4197530746459961, 0.4197530746459961, 0.4444444477558136, 0.4197530746459961, 0.48148149251937866, 0.4938271641731262, 0.4444444477558136, 0.5185185074806213, NaN, 0.5061728358268738, 0.5308641791343689, 0.5679012537002563, 0.6172839403152466, 0.5555555820465088, 0.5925925970077515, 0.5679012537002563, 0.6790123581886292, 0.6419752836227417, 0.6913580298423767, 0.6790123581886292, 0.8024691343307495, 0.7407407164573669, 0.6913580298423767, 0.7160493731498718, 0.7407407164573669, 0.7530864477157593, 0.6913580298423767, 0.6419752836227417, 0.8024691343307495, 0.7530864477157593, 0.790123462677002, 0.7777777910232544, 0.8271604776382446, 0.8888888955116272, 0.8888888955116272, 0.7654321193695068, 0.9753086566925049, 0.9753086566925049, 0.9382715821266174, 0.8888888955116272, 0.9135802388191223, 0.9876543283462524, 0.8888888955116272, 1.0, 0.9506173133850098, 1.0, 1.0, 1.0], [0.06172839552164078, 0.04938271641731262, 0.17283950746059418, 0.03703703731298447, 0.14814814925193787, 0.09876543283462524, 0.20987653732299805, 0.25925925374031067, 0.2469135820865631, 0.2716049253940582, 0.25925925374031067, 0.37037035822868347, 0.16049382090568542, 0.34567901492118835, 0.2839506268501282, 0.3333333432674408, 0.40740740299224854, 0.395061731338501, 0.16049382090568542, 0.37037035822868347, 0.2716049253940582, 0.395061731338501, 0.18518517911434174, 0.14814814925193787, 0.40740740299224854, 0.3827160596847534, 0.43209877610206604, 0.395061731338501, 0.3580246865749359, 0.3580246865749359, 0.32098764181137085, 0.5802469253540039, 0.4691357910633087, 0.5061728358268738, 0.4938271641731262, NaN, 0.5308641791343689, 0.4197530746459961, 0.5802469253540039, 0.5185185074806213, 0.5308641791343689, 0.5185185074806213, 0.5802469253540039, 0.5555555820465088, 0.5802469253540039, 0.6790123581886292, 0.6172839403152466, 0.5925925970077515, 0.5925925970077515, 0.6913580298423767, 0.6172839403152466, 0.6419752836227417, 0.8395061492919922, 0.7530864477157593, 0.654321014881134, 0.6913580298423767, 0.6419752836227417, 0.6913580298423767, 0.7654321193695068, 0.7777777910232544, 0.7777777910232544, 0.7283950448036194, 0.9506173133850098, 0.9753086566925049, 0.9135802388191223, 0.790123462677002, 0.9012345671653748, 1.0, 0.8888888955116272, 0.9135802388191223, 1.0, 0.9629629850387573, 1.0, 1.0], [0.012345679104328156, 0.09876543283462524, 0.03703703731298447, 0.2222222238779068, 0.20987653732299805, 0.3086419701576233, 0.04938271641731262, 0.04938271641731262, 0.06172839552164078, 0.12345679104328156, 0.07407407462596893, 0.08641975373029709, 0.3827160596847534, 0.09876543283462524, 0.18518517911434174, 0.2469135820865631, 0.1111111119389534, 0.2222222238779068, 0.32098764181137085, 0.20987653732299805, 0.3827160596847534, 0.1358024626970291, 0.34567901492118835, 0.34567901492118835, 0.3086419701576233, 0.3333333432674408, 0.32098764181137085, 0.4197530746459961, 0.3827160596847534, 0.3827160596847534, 0.3827160596847534, 0.3580246865749359, 0.45679011940956116, 0.3580246865749359, 0.4691357910633087, 0.4691357910633087, NaN, 0.5555555820465088, 0.6296296119689941, 0.5925925970077515, 0.6172839403152466, 0.604938268661499, 0.5679012537002563, 0.5802469253540039, 0.5061728358268738, 0.5308641791343689, 0.6296296119689941, 0.6913580298423767, 0.8888888955116272, 0.7283950448036194, 0.7283950448036194, 0.8024691343307495, 0.6296296119689941, 0.6913580298423767, 0.7530864477157593, 0.8271604776382446, 0.8518518805503845, 0.9506173133850098, 0.9753086566925049, 0.8148148059844971, 0.8271604776382446, 0.9876543283462524, 0.8395061492919922, 0.8148148059844971, 0.9382715821266174, 1.0, 0.8888888955116272, 0.8641975522041321, 0.9876543283462524, 0.9876543283462524, 0.9506173133850098, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.03703703731298447, 0.06172839552164078, 0.0, 0.09876543283462524, 0.1975308656692505, 0.04938271641731262, 0.18518517911434174, 0.0, 0.17283950746059418, 0.16049382090568542, 0.02469135820865631, 0.17283950746059418, 0.1975308656692505, 0.12345679104328156, 0.17283950746059418, 0.1111111119389534, 0.1975308656692505, 0.1975308656692505, 0.1111111119389534, 0.2716049253940582, 0.25925925374031067, 0.32098764181137085, 0.25925925374031067, 0.3580246865749359, 0.3827160596847534, 0.3086419701576233, 0.395061731338501, 0.3827160596847534, 0.3333333432674408, 0.4197530746459961, 0.3086419701576233, 0.5061728358268738, 0.43209877610206604, 0.5802469253540039, 0.4444444477558136, NaN, 0.4444444477558136, 0.5061728358268738, 0.4444444477558136, 0.6419752836227417, 0.6296296119689941, 0.48148149251937866, 0.6419752836227417, 0.6790123581886292, 0.7037037014961243, 0.6666666865348816, 0.6913580298423767, 0.7654321193695068, 0.7407407164573669, 0.8024691343307495, 0.7037037014961243, 0.6790123581886292, 0.8765432238578796, 0.790123462677002, 0.8271604776382446, 0.8148148059844971, 0.8518518805503845, 0.9629629850387573, 0.9629629850387573, 0.8641975522041321, 0.9506173133850098, 0.9382715821266174, 0.9506173133850098, 0.9012345671653748, 0.9506173133850098, 0.9506173133850098, 0.8641975522041321, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.0, 0.09876543283462524, 0.03703703731298447, 0.1111111119389534, 0.14814814925193787, 0.17283950746059418, 0.18518517911434174, 0.03703703731298447, 0.14814814925193787, 0.02469135820865631, 0.1975308656692505, 0.16049382090568542, 0.1975308656692505, 0.2222222238779068, 0.25925925374031067, 0.14814814925193787, 0.20987653732299805, 0.1358024626970291, 0.29629629850387573, 0.2716049253940582, 0.2222222238779068, 0.20987653732299805, 0.3333333432674408, 0.37037035822868347, 0.2716049253940582, 0.34567901492118835, 0.29629629850387573, 0.37037035822868347, 0.34567901492118835, 0.3827160596847534, 0.395061731338501, 0.29629629850387573, 0.37037035822868347, 0.34567901492118835, 0.3827160596847534, 0.4197530746459961, 0.37037035822868347, 0.5555555820465088, NaN, 0.4444444477558136, 0.4938271641731262, 0.5432098507881165, 0.5802469253540039, 0.5308641791343689, 0.5802469253540039, 0.4938271641731262, 0.6913580298423767, 0.6419752836227417, 0.6790123581886292, 0.654321014881134, 0.6666666865348816, 0.9259259104728699, 0.6913580298423767, 0.6172839403152466, 0.7530864477157593, 0.6913580298423767, 0.9629629850387573, 0.8765432238578796, 0.8641975522041321, 0.7530864477157593, 0.7530864477157593, 0.8888888955116272, 0.8641975522041321, 0.8518518805503845, 0.9259259104728699, 0.9753086566925049, 0.9259259104728699, 0.8148148059844971, 1.0, 0.8888888955116272, 0.9259259104728699, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.09876543283462524, 0.06172839552164078, 0.1111111119389534, 0.08641975373029709, 0.16049382090568542, 0.14814814925193787, 0.17283950746059418, 0.16049382090568542, 0.16049382090568542, 0.1358024626970291, 0.09876543283462524, 0.1358024626970291, 0.14814814925193787, 0.1975308656692505, 0.16049382090568542, 0.2469135820865631, 0.2222222238779068, 0.16049382090568542, 0.2222222238779068, 0.2716049253940582, 0.2716049253940582, 0.23456789553165436, 0.23456789553165436, 0.34567901492118835, 0.3827160596847534, 0.34567901492118835, 0.4197530746459961, 0.4444444477558136, 0.43209877610206604, 0.40740740299224854, 0.34567901492118835, 0.45679011940956116, 0.4444444477558136, 0.48148149251937866, 0.40740740299224854, 0.4938271641731262, 0.5555555820465088, NaN, 0.4197530746459961, 0.5555555820465088, 0.6790123581886292, 0.5802469253540039, 0.654321014881134, 0.604938268661499, 0.6666666865348816, 0.6172839403152466, 0.7160493731498718, 0.6790123581886292, 0.654321014881134, 0.7037037014961243, 0.6419752836227417, 0.7283950448036194, 0.7407407164573669, 0.7037037014961243, 0.7407407164573669, 0.7283950448036194, 0.9012345671653748, 0.9753086566925049, 0.9876543283462524, 0.7407407164573669, 0.8518518805503845, 0.8518518805503845, 0.8395061492919922, 0.9382715821266174, 0.8518518805503845, 0.8765432238578796, 0.790123462677002, 1.0, 0.9629629850387573, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.04938271641731262, 0.09876543283462524, 0.03703703731298447, 0.09876543283462524, 0.09876543283462524, 0.08641975373029709, 0.09876543283462524, 0.08641975373029709, 0.09876543283462524, 0.04938271641731262, 0.06172839552164078, 0.06172839552164078, 0.09876543283462524, 0.16049382090568542, 0.1358024626970291, 0.20987653732299805, 0.1975308656692505, 0.09876543283462524, 0.14814814925193787, 0.1975308656692505, 0.29629629850387573, 0.29629629850387573, 0.23456789553165436, 0.29629629850387573, 0.3580246865749359, 0.34567901492118835, 0.40740740299224854, 0.3333333432674408, 0.4197530746459961, 0.32098764181137085, 0.2469135820865631, 0.45679011940956116, 0.40740740299224854, 0.4691357910633087, 0.3827160596847534, 0.5555555820465088, 0.5061728358268738, 0.5802469253540039, NaN, 0.5802469253540039, 0.5432098507881165, 0.7160493731498718, 0.5432098507881165, 0.5925925970077515, 0.6419752836227417, 0.6666666865348816, 0.8024691343307495, 0.654321014881134, 0.6913580298423767, 0.7037037014961243, 0.5555555820465088, 0.654321014881134, 0.8271604776382446, 0.7283950448036194, 0.7160493731498718, 0.8395061492919922, 0.8765432238578796, 0.9382715821266174, 0.9382715821266174, 0.8765432238578796, 0.8518518805503845, 0.8395061492919922, 0.8148148059844971, 0.9135802388191223, 0.8271604776382446, 0.8641975522041321, 0.790123462677002, 1.0, 0.9876543283462524, 1.0, 0.8888888955116272, 1.0], [0.02469135820865631, 0.0, 0.02469135820865631, 0.03703703731298447, 0.04938271641731262, 0.09876543283462524, 0.04938271641731262, 0.04938271641731262, 0.04938271641731262, 0.07407407462596893, 0.06172839552164078, 0.14814814925193787, 0.1358024626970291, 0.17283950746059418, 0.09876543283462524, 0.17283950746059418, 0.1975308656692505, 0.09876543283462524, 0.1975308656692505, 0.1975308656692505, 0.17283950746059418, 0.2469135820865631, 0.20987653732299805, 0.2222222238779068, 0.3333333432674408, 0.3580246865749359, 0.4197530746459961, 0.3086419701576233, 0.2716049253940582, 0.07407407462596893, 0.29629629850387573, 0.3580246865749359, 0.23456789553165436, 0.37037035822868347, 0.43209877610206604, 0.48148149251937866, 0.395061731338501, 0.3580246865749359, 0.45679011940956116, 0.4444444477558136, 0.4197530746459961, NaN, 0.43209877610206604, 0.4444444477558136, 0.5308641791343689, 0.7037037014961243, 0.6172839403152466, 0.604938268661499, 0.6172839403152466, 0.6790123581886292, 0.6913580298423767, 0.604938268661499, 0.5802469253540039, 0.5802469253540039, 0.8765432238578796, 0.7530864477157593, 0.654321014881134, 0.8641975522041321, 0.8271604776382446, 0.8271604776382446, 0.8271604776382446, 0.8641975522041321, 0.9506173133850098, 0.9506173133850098, 0.8888888955116272, 0.9135802388191223, 0.9135802388191223, 0.9753086566925049, 0.8765432238578796, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 1.0], [0.04938271641731262, 0.03703703731298447, 0.03703703731298447, 0.03703703731298447, 0.04938271641731262, 0.06172839552164078, 0.20987653732299805, 0.09876543283462524, 0.20987653732299805, 0.09876543283462524, 0.2222222238779068, 0.18518517911434174, 0.1111111119389534, 0.17283950746059418, 0.2222222238779068, 0.1358024626970291, 0.20987653732299805, 0.18518517911434174, 0.12345679104328156, 0.2222222238779068, 0.1975308656692505, 0.18518517911434174, 0.1975308656692505, 0.2469135820865631, 0.2839506268501282, 0.16049382090568542, 0.23456789553165436, 0.2469135820865631, 0.12345679104328156, 0.32098764181137085, 0.18518517911434174, 0.48148149251937866, 0.3333333432674408, 0.3827160596847534, 0.32098764181137085, 0.4197530746459961, 0.43209877610206604, 0.37037035822868347, 0.4197530746459961, 0.32098764181137085, 0.45679011940956116, 0.5679012537002563, NaN, 0.4691357910633087, 0.5185185074806213, 0.4938271641731262, 0.5185185074806213, 0.5432098507881165, 0.4938271641731262, 0.6419752836227417, 0.654321014881134, 0.6419752836227417, 0.7592592835426331, 0.6913580298423767, 0.6419752836227417, 0.7160493731498718, 0.7530864477157593, 0.654321014881134, 0.7777777910232544, 0.6296296119689941, 0.6419752836227417, 0.7160493731498718, 0.9629629850387573, 0.9629629850387573, 0.9259259104728699, 0.7777777910232544, 0.9382715821266174, 0.9012345671653748, 0.8888888955116272, 0.9012345671653748, 0.8888888955116272, 0.9753086566925049, 1.0, 1.0], [0.0, 0.0, 0.04938271641731262, 0.04938271641731262, 0.03703703731298447, 0.07407407462596893, 0.07407407462596893, 0.06172839552164078, 0.06172839552164078, 0.06172839552164078, 0.03703703731298447, 0.02469135820865631, 0.07407407462596893, 0.02469135820865631, 0.03703703731298447, 0.14814814925193787, 0.09876543283462524, 0.18518517911434174, 0.14814814925193787, 0.08641975373029709, 0.12345679104328156, 0.14814814925193787, 0.18518517911434174, 0.2469135820865631, 0.16049382090568542, 0.2716049253940582, 0.34567901492118835, 0.3086419701576233, 0.37037035822868347, 0.29629629850387573, 0.32098764181137085, 0.3086419701576233, 0.08641975373029709, 0.395061731338501, 0.3580246865749359, 0.4444444477558136, 0.4197530746459961, 0.5185185074806213, 0.4691357910633087, 0.4197530746459961, 0.2839506268501282, 0.5555555820465088, 0.5308641791343689, NaN, 0.5061728358268738, 0.6172839403152466, 0.604938268661499, 0.6172839403152466, 0.6419752836227417, 0.6419752836227417, 0.6790123581886292, 0.7037037014961243, 0.5555555820465088, 0.654321014881134, 0.8148148059844971, 0.7283950448036194, 0.7037037014961243, 0.790123462677002, 0.8641975522041321, 0.8271604776382446, 0.8271604776382446, 0.8271604776382446, 0.8518518805503845, 0.8395061492919922, 0.8024691343307495, 0.9012345671653748, 0.8024691343307495, 0.8518518805503845, 0.7777777910232544, 1.0, 0.9753086566925049, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.04938271641731262, 0.012345679104328156, 0.06172839552164078, 0.06172839552164078, 0.07407407462596893, 0.08641975373029709, 0.04938271641731262, 0.04938271641731262, 0.08641975373029709, 0.1358024626970291, 0.0, 0.03703703731298447, 0.02469135820865631, 0.09876543283462524, 0.20987653732299805, 0.16049382090568542, 0.08641975373029709, 0.07407407462596893, 0.1358024626970291, 0.2716049253940582, 0.12345679104328156, 0.1358024626970291, 0.03703703731298447, 0.14814814925193787, 0.2222222238779068, 0.25925925374031067, 0.2469135820865631, 0.2716049253940582, 0.23456789553165436, 0.48148149251937866, 0.3580246865749359, 0.5185185074806213, 0.3086419701576233, 0.4197530746459961, 0.4938271641731262, 0.3580246865749359, 0.4197530746459961, 0.34567901492118835, 0.45679011940956116, 0.4691357910633087, 0.48148149251937866, 0.4938271641731262, NaN, 0.5679012537002563, 0.5555555820465088, 0.6172839403152466, 0.5432098507881165, 0.6172839403152466, 0.6419752836227417, 0.5555555820465088, 0.6296296119689941, 0.6666666865348816, 0.7160493731498718, 0.7160493731498718, 0.6666666865348816, 0.654321014881134, 0.7407407164573669, 0.8271604776382446, 0.8395061492919922, 0.6666666865348816, 0.8641975522041321, 0.9506173133850098, 0.9382715821266174, 0.790123462677002, 0.9629629850387573, 0.9382715821266174, 0.8888888955116272, 1.0, 0.9135802388191223, 1.0, 1.0, 1.0], [0.0, 0.0, 0.03703703731298447, 0.012345679104328156, 0.06172839552164078, 0.08641975373029709, 0.06172839552164078, 0.0, 0.08641975373029709, 0.09876543283462524, 0.07407407462596893, 0.04938271641731262, 0.07407407462596893, 0.03703703731298447, 0.02469135820865631, 0.17283950746059418, 0.1111111119389534, 0.18518517911434174, 0.16049382090568542, 0.07407407462596893, 0.16049382090568542, 0.08641975373029709, 0.1975308656692505, 0.18518517911434174, 0.07407407462596893, 0.25925925374031067, 0.25925925374031067, 0.25925925374031067, 0.29629629850387573, 0.17283950746059418, 0.34567901492118835, 0.395061731338501, 0.3086419701576233, 0.20987653732299805, 0.32098764181137085, 0.32098764181137085, 0.4691357910633087, 0.32098764181137085, 0.5061728358268738, 0.395061731338501, 0.40740740299224854, 0.29629629850387573, 0.5061728358268738, 0.3827160596847534, 0.43209877610206604, NaN, 0.5185185074806213, 0.5432098507881165, 0.6419752836227417, 0.5308641791343689, 0.5555555820465088, 0.6913580298423767, 0.5308641791343689, 0.6296296119689941, 0.604938268661499, 0.604938268661499, 0.7407407164573669, 0.7283950448036194, 0.8641975522041321, 0.8765432238578796, 0.8888888955116272, 0.7777777910232544, 0.7283950448036194, 0.7160493731498718, 0.8271604776382446, 0.8765432238578796, 0.8395061492919922, 0.9135802388191223, 0.8765432238578796, 0.9876543283462524, 0.9629629850387573, 0.8888888955116272, 0.7777777910232544, 1.0], [0.0, 0.0, 0.012345679104328156, 0.02469135820865631, 0.02469135820865631, 0.04938271641731262, 0.012345679104328156, 0.02469135820865631, 0.04938271641731262, 0.04938271641731262, 0.07407407462596893, 0.09876543283462524, 0.03703703731298447, 0.012345679104328156, 0.02469135820865631, 0.04938271641731262, 0.16049382090568542, 0.16049382090568542, 0.1975308656692505, 0.03703703731298447, 0.09876543283462524, 0.20987653732299805, 0.23456789553165436, 0.2222222238779068, 0.07407407462596893, 0.14814814925193787, 0.17283950746059418, 0.1111111119389534, 0.23456789553165436, 0.2839506268501282, 0.3333333432674408, 0.37037035822868347, 0.3086419701576233, 0.3086419701576233, 0.1975308656692505, 0.3827160596847534, 0.37037035822868347, 0.29629629850387573, 0.3086419701576233, 0.3333333432674408, 0.3580246865749359, 0.3827160596847534, 0.48148149251937866, 0.395061731338501, 0.4444444477558136, 0.48148149251937866, NaN, 0.5802469253540039, 0.5308641791343689, 0.5185185074806213, 0.654321014881134, 0.4938271641731262, 0.5308641791343689, 0.5925925970077515, 0.7283950448036194, 0.604938268661499, 0.6419752836227417, 0.6666666865348816, 0.8148148059844971, 0.7160493731498718, 0.7283950448036194, 0.7160493731498718, 0.8271604776382446, 0.8518518805503845, 0.9259259104728699, 0.8518518805503845, 0.9506173133850098, 0.9012345671653748, 0.8888888955116272, 1.0, 0.9135802388191223, 1.0, 0.9135802388191223, 1.0], [0.0, 0.02469135820865631, 0.012345679104328156, 0.06172839552164078, 0.04938271641731262, 0.03703703731298447, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.012345679104328156, 0.07407407462596893, 0.0, 0.03703703731298447, 0.12345679104328156, 0.03703703731298447, 0.06172839552164078, 0.09876543283462524, 0.03703703731298447, 0.1358024626970291, 0.06172839552164078, 0.16049382090568542, 0.17283950746059418, 0.14814814925193787, 0.20987653732299805, 0.18518517911434174, 0.1975308656692505, 0.25925925374031067, 0.2222222238779068, 0.2839506268501282, 0.2839506268501282, 0.25925925374031067, 0.16049382090568542, 0.25925925374031067, 0.40740740299224854, 0.3086419701576233, 0.3333333432674408, 0.3580246865749359, 0.3827160596847534, 0.3333333432674408, 0.395061731338501, 0.45679011940956116, 0.3827160596847534, 0.3827160596847534, 0.45679011940956116, 0.4197530746459961, NaN, 0.5679012537002563, 0.5925925970077515, 0.7283950448036194, 0.5925925970077515, 0.5061728358268738, 0.5925925970077515, 0.6666666865348816, 0.654321014881134, 0.6666666865348816, 0.7777777910232544, 0.8765432238578796, 0.7777777910232544, 0.790123462677002, 0.8395061492919922, 0.8395061492919922, 0.8148148059844971, 0.9629629850387573, 0.9506173133850098, 0.9629629850387573, 0.9259259104728699, 0.8765432238578796, 0.9259259104728699, 0.9629629850387573, 1.0, 0.8888888955116272, 1.0], [0.0, 0.02469135820865631, 0.03703703731298447, 0.08641975373029709, 0.04938271641731262, 0.08641975373029709, 0.04938271641731262, 0.02469135820865631, 0.02469135820865631, 0.03703703731298447, 0.03703703731298447, 0.02469135820865631, 0.09876543283462524, 0.04938271641731262, 0.04938271641731262, 0.16049382090568542, 0.07407407462596893, 0.12345679104328156, 0.16049382090568542, 0.06172839552164078, 0.14814814925193787, 0.04938271641731262, 0.18518517911434174, 0.1111111119389534, 0.1975308656692505, 0.2469135820865631, 0.17283950746059418, 0.23456789553165436, 0.2839506268501282, 0.1975308656692505, 0.3333333432674408, 0.23456789553165436, 0.1358024626970291, 0.18518517911434174, 0.3086419701576233, 0.40740740299224854, 0.1111111119389534, 0.3086419701576233, 0.32098764181137085, 0.2839506268501282, 0.1975308656692505, 0.3827160596847534, 0.5061728358268738, 0.3580246865749359, 0.45679011940956116, 0.3580246865749359, 0.4691357910633087, 0.43209877610206604, NaN, 0.5679012537002563, 0.4691357910633087, 0.5679012537002563, 0.5802469253540039, 0.654321014881134, 0.604938268661499, 0.7530864477157593, 0.6790123581886292, 0.790123462677002, 0.8765432238578796, 0.6296296119689941, 0.654321014881134, 0.8271604776382446, 0.8271604776382446, 0.8518518805503845, 0.8395061492919922, 0.9753086566925049, 0.8395061492919922, 0.8518518805503845, 0.8148148059844971, 0.9506173133850098, 0.9629629850387573, 1.0, 0.8888888955116272, 1.0], [0.0, 0.012345679104328156, 0.012345679104328156, 0.03703703731298447, 0.02469135820865631, 0.012345679104328156, 0.03703703731298447, 0.04938271641731262, 0.012345679104328156, 0.0, 0.04938271641731262, 0.12345679104328156, 0.07407407462596893, 0.012345679104328156, 0.02469135820865631, 0.1111111119389534, 0.14814814925193787, 0.1111111119389534, 0.03703703731298447, 0.04938271641731262, 0.14814814925193787, 0.14814814925193787, 0.09876543283462524, 0.08641975373029709, 0.18518517911434174, 0.2222222238779068, 0.18518517911434174, 0.17283950746059418, 0.2469135820865631, 0.12345679104328156, 0.25925925374031067, 0.18518517911434174, 0.2222222238779068, 0.14814814925193787, 0.2839506268501282, 0.3086419701576233, 0.2716049253940582, 0.23456789553165436, 0.34567901492118835, 0.32098764181137085, 0.34567901492118835, 0.32098764181137085, 0.3580246865749359, 0.3580246865749359, 0.3827160596847534, 0.4691357910633087, 0.48148149251937866, 0.40740740299224854, 0.43209877610206604, NaN, 0.5061728358268738, 0.48148149251937866, 0.48148149251937866, 0.5308641791343689, 0.604938268661499, 0.7283950448036194, 0.604938268661499, 0.6913580298423767, 0.7160493731498718, 0.6913580298423767, 0.6913580298423767, 0.7654321193695068, 0.7283950448036194, 0.7777777910232544, 0.9753086566925049, 0.8888888955116272, 0.9876543283462524, 0.9382715821266174, 0.9012345671653748, 0.8395061492919922, 0.9629629850387573, 0.8888888955116272, 1.0, 1.0], [0.012345679104328156, 0.02469135820865631, 0.02469135820865631, 0.03703703731298447, 0.012345679104328156, 0.03703703731298447, 0.03703703731298447, 0.012345679104328156, 0.03703703731298447, 0.03703703731298447, 0.03703703731298447, 0.012345679104328156, 0.07407407462596893, 0.012345679104328156, 0.04938271641731262, 0.1358024626970291, 0.03703703731298447, 0.08641975373029709, 0.04938271641731262, 0.06172839552164078, 0.12345679104328156, 0.09876543283462524, 0.07407407462596893, 0.1111111119389534, 0.12345679104328156, 0.1975308656692505, 0.16049382090568542, 0.20987653732299805, 0.2716049253940582, 0.16049382090568542, 0.2222222238779068, 0.2716049253940582, 0.2222222238779068, 0.14814814925193787, 0.25925925374031067, 0.3827160596847534, 0.2716049253940582, 0.25925925374031067, 0.3333333432674408, 0.34567901492118835, 0.3086419701576233, 0.3086419701576233, 0.34567901492118835, 0.32098764181137085, 0.3580246865749359, 0.4444444477558136, 0.34567901492118835, 0.2716049253940582, 0.5308641791343689, 0.4938271641731262, NaN, 0.5185185074806213, 0.45679011940956116, 0.5802469253540039, 0.5308641791343689, 0.5925925970077515, 0.6296296119689941, 0.6790123581886292, 0.8395061492919922, 0.7654321193695068, 0.7777777910232544, 0.7530864477157593, 0.8148148059844971, 0.8148148059844971, 0.9135802388191223, 0.9012345671653748, 0.9320987462997437, 0.9135802388191223, 0.8395061492919922, 0.9259259104728699, 0.9753086566925049, 1.0, 0.8888888955116272, 1.0], [0.0, 0.02469135820865631, 0.02469135820865631, 0.07407407462596893, 0.06172839552164078, 0.07407407462596893, 0.02469135820865631, 0.03703703731298447, 0.03703703731298447, 0.02469135820865631, 0.04938271641731262, 0.12345679104328156, 0.09876543283462524, 0.1358024626970291, 0.09876543283462524, 0.07407407462596893, 0.16049382090568542, 0.03703703731298447, 0.1358024626970291, 0.14814814925193787, 0.1358024626970291, 0.14814814925193787, 0.1358024626970291, 0.2222222238779068, 0.23456789553165436, 0.32098764181137085, 0.16049382090568542, 0.08641975373029709, 0.2222222238779068, 0.18518517911434174, 0.2839506268501282, 0.2222222238779068, 0.2716049253940582, 0.32098764181137085, 0.2469135820865631, 0.3580246865749359, 0.1975308656692505, 0.1975308656692505, 0.07407407462596893, 0.29629629850387573, 0.29629629850387573, 0.395061731338501, 0.3580246865749359, 0.29629629850387573, 0.4444444477558136, 0.3086419701576233, 0.5061728358268738, 0.40740740299224854, 0.43209877610206604, 0.5185185074806213, 0.48148149251937866, NaN, 0.5432098507881165, 0.4691357910633087, 0.5432098507881165, 0.5679012537002563, 0.6790123581886292, 0.6296296119689941, 0.6790123581886292, 0.6172839403152466, 0.6172839403152466, 0.7037037014961243, 0.8024691343307495, 0.8148148059844971, 0.8024691343307495, 0.8888888955116272, 0.8518518805503845, 0.7530864477157593, 0.8765432238578796, 0.8765432238578796, 0.8518518805503845, 0.8765432238578796, 0.9506173133850098, 1.0], [0.02469135820865631, 0.02469135820865631, 0.04938271641731262, 0.03703703731298447, 0.06172839552164078, 0.04938271641731262, 0.1111111119389534, 0.07407407462596893, 0.12345679104328156, 0.09876543283462524, 0.1358024626970291, 0.18518517911434174, 0.12345679104328156, 0.12345679104328156, 0.04938271641731262, 0.08641975373029709, 0.2469135820865631, 0.17283950746059418, 0.1111111119389534, 0.16049382090568542, 0.17283950746059418, 0.1975308656692505, 0.17283950746059418, 0.16049382090568542, 0.18518517911434174, 0.08641975373029709, 0.16049382090568542, 0.14814814925193787, 0.06172839552164078, 0.25925925374031067, 0.09876543283462524, 0.34567901492118835, 0.3580246865749359, 0.3827160596847534, 0.3086419701576233, 0.16049382090568542, 0.37037035822868347, 0.29629629850387573, 0.3086419701576233, 0.3580246865749359, 0.4444444477558136, 0.4197530746459961, 0.24074074625968933, 0.4444444477558136, 0.37037035822868347, 0.4691357910633087, 0.4691357910633087, 0.4938271641731262, 0.4197530746459961, 0.5185185074806213, 0.5432098507881165, 0.45679011940956116, NaN, 0.5061728358268738, 0.5802469253540039, 0.5555555820465088, 0.5925925970077515, 0.6666666865348816, 0.5679012537002563, 0.6296296119689941, 0.6296296119689941, 0.6790123581886292, 0.7654321193695068, 0.8148148059844971, 0.6419752836227417, 0.7160493731498718, 0.6666666865348816, 0.6913580298423767, 0.7283950448036194, 0.6790123581886292, 0.8395061492919922, 0.7037037014961243, 0.9753086566925049, 1.0], [0.012345679104328156, 0.04938271641731262, 0.02469135820865631, 0.1111111119389534, 0.1358024626970291, 0.09876543283462524, 0.16049382090568542, 0.16049382090568542, 0.1975308656692505, 0.06172839552164078, 0.1975308656692505, 0.17283950746059418, 0.14814814925193787, 0.2469135820865631, 0.23456789553165436, 0.04938271641731262, 0.18518517911434174, 0.1111111119389534, 0.25925925374031067, 0.2716049253940582, 0.14814814925193787, 0.18518517911434174, 0.2716049253940582, 0.25925925374031067, 0.2839506268501282, 0.23456789553165436, 0.14814814925193787, 0.1975308656692505, 0.25925925374031067, 0.25925925374031067, 0.14814814925193787, 0.23456789553165436, 0.3333333432674408, 0.32098764181137085, 0.3580246865749359, 0.2469135820865631, 0.3086419701576233, 0.32098764181137085, 0.3827160596847534, 0.2716049253940582, 0.34567901492118835, 0.4197530746459961, 0.3086419701576233, 0.34567901492118835, 0.3333333432674408, 0.37037035822868347, 0.40740740299224854, 0.40740740299224854, 0.34567901492118835, 0.4691357910633087, 0.4197530746459961, 0.5308641791343689, 0.4938271641731262, NaN, 0.43209877610206604, 0.5061728358268738, 0.5679012537002563, 0.604938268661499, 0.5185185074806213, 0.4691357910633087, 0.48148149251937866, 0.6172839403152466, 0.5925925970077515, 0.6419752836227417, 0.6790123581886292, 0.6172839403152466, 0.7283950448036194, 0.5925925970077515, 0.7530864477157593, 0.5802469253540039, 0.6172839403152466, 0.6419752836227417, 0.7654321193695068, 0.8765432238578796], [0.0, 0.0, 0.012345679104328156, 0.03703703731298447, 0.0, 0.02469135820865631, 0.02469135820865631, 0.0, 0.012345679104328156, 0.012345679104328156, 0.02469135820865631, 0.09876543283462524, 0.02469135820865631, 0.0, 0.02469135820865631, 0.12345679104328156, 0.09876543283462524, 0.07407407462596893, 0.04938271641731262, 0.0, 0.02469135820865631, 0.14814814925193787, 0.09876543283462524, 0.04938271641731262, 0.07407407462596893, 0.2222222238779068, 0.12345679104328156, 0.16049382090568542, 0.2469135820865631, 0.012345679104328156, 0.2222222238779068, 0.25925925374031067, 0.14814814925193787, 0.08641975373029709, 0.1975308656692505, 0.34567901492118835, 0.2469135820865631, 0.12345679104328156, 0.2469135820865631, 0.25925925374031067, 0.17283950746059418, 0.12345679104328156, 0.3580246865749359, 0.18518517911434174, 0.2839506268501282, 0.395061731338501, 0.2716049253940582, 0.3333333432674408, 0.395061731338501, 0.395061731338501, 0.4691357910633087, 0.45679011940956116, 0.4197530746459961, 0.5679012537002563, NaN, 0.4938271641731262, 0.5061728358268738, 0.7037037014961243, 0.7777777910232544, 0.7037037014961243, 0.7037037014961243, 0.7407407164573669, 0.7777777910232544, 0.8024691343307495, 0.8518518805503845, 0.8641975522041321, 0.8765432238578796, 0.9259259104728699, 0.8765432238578796, 0.8888888955116272, 0.9506173133850098, 0.9876543283462524, 0.9506173133850098, 1.0], [0.0, 0.012345679104328156, 0.0, 0.03703703731298447, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.02469135820865631, 0.02469135820865631, 0.03703703731298447, 0.12345679104328156, 0.04938271641731262, 0.012345679104328156, 0.02469135820865631, 0.12345679104328156, 0.14814814925193787, 0.04938271641731262, 0.03703703731298447, 0.02469135820865631, 0.08641975373029709, 0.16049382090568542, 0.07407407462596893, 0.06172839552164078, 0.08641975373029709, 0.18518517911434174, 0.14814814925193787, 0.16049382090568542, 0.20987653732299805, 0.09876543283462524, 0.1975308656692505, 0.1975308656692505, 0.17283950746059418, 0.1111111119389534, 0.2469135820865631, 0.3086419701576233, 0.17283950746059418, 0.20987653732299805, 0.3086419701576233, 0.29629629850387573, 0.2716049253940582, 0.2469135820865631, 0.2839506268501282, 0.2716049253940582, 0.2839506268501282, 0.395061731338501, 0.395061731338501, 0.34567901492118835, 0.2469135820865631, 0.2716049253940582, 0.40740740299224854, 0.43209877610206604, 0.4444444477558136, 0.4938271641731262, 0.5061728358268738, NaN, 0.4938271641731262, 0.6790123581886292, 0.654321014881134, 0.6790123581886292, 0.6666666865348816, 0.7160493731498718, 0.6790123581886292, 0.7530864477157593, 0.9382715821266174, 0.8765432238578796, 0.9506173133850098, 0.9259259104728699, 0.8888888955116272, 0.8271604776382446, 0.9629629850387573, 0.9012345671653748, 0.9876543283462524, 1.0], [0.0, 0.02469135820865631, 0.012345679104328156, 0.07407407462596893, 0.08641975373029709, 0.08641975373029709, 0.04938271641731262, 0.0, 0.02469135820865631, 0.03703703731298447, 0.06172839552164078, 0.1358024626970291, 0.1111111119389534, 0.1111111119389534, 0.14814814925193787, 0.03703703731298447, 0.14814814925193787, 0.04938271641731262, 0.1111111119389534, 0.16049382090568542, 0.09876543283462524, 0.14814814925193787, 0.12345679104328156, 0.17283950746059418, 0.20987653732299805, 0.23456789553165436, 0.1111111119389534, 0.09876543283462524, 0.18518517911434174, 0.18518517911434174, 0.1358024626970291, 0.14814814925193787, 0.2222222238779068, 0.25925925374031067, 0.20987653732299805, 0.3580246865749359, 0.14814814925193787, 0.17283950746059418, 0.03703703731298447, 0.25925925374031067, 0.2839506268501282, 0.34567901492118835, 0.2469135820865631, 0.29629629850387573, 0.3333333432674408, 0.25925925374031067, 0.3580246865749359, 0.3333333432674408, 0.32098764181137085, 0.395061731338501, 0.37037035822868347, 0.32098764181137085, 0.40740740299224854, 0.43209877610206604, 0.4938271641731262, 0.5061728358268738, NaN, 0.5432098507881165, 0.5555555820465088, 0.5308641791343689, 0.5432098507881165, 0.5925925970077515, 0.7654321193695068, 0.7777777910232544, 0.6790123581886292, 0.6419752836227417, 0.6790123581886292, 0.7037037014961243, 0.790123462677002, 0.7777777910232544, 0.7777777910232544, 0.790123462677002, 0.8765432238578796, 1.0], [0.0, 0.02469135820865631, 0.0, 0.08641975373029709, 0.03703703731298447, 0.08641975373029709, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.07407407462596893, 0.0, 0.02469135820865631, 0.1111111119389534, 0.0, 0.02469135820865631, 0.17283950746059418, 0.0, 0.12345679104328156, 0.0, 0.1975308656692505, 0.17283950746059418, 0.14814814925193787, 0.2716049253940582, 0.1111111119389534, 0.14814814925193787, 0.2716049253940582, 0.09876543283462524, 0.29629629850387573, 0.02469135820865631, 0.1111111119389534, 0.1358024626970291, 0.2222222238779068, 0.3086419701576233, 0.04938271641731262, 0.18518517911434174, 0.12345679104328156, 0.2716049253940582, 0.16049382090568542, 0.1358024626970291, 0.34567901492118835, 0.20987653732299805, 0.34567901492118835, 0.2716049253940582, 0.3333333432674408, 0.2222222238779068, 0.20987653732299805, 0.3086419701576233, 0.32098764181137085, 0.37037035822868347, 0.3333333432674408, 0.395061731338501, 0.29629629850387573, 0.32098764181137085, 0.45679011940956116, NaN, 0.3580246865749359, 0.48148149251937866, 0.48148149251937866, 0.7962962985038757, 0.48148149251937866, 0.5061728358268738, 0.604938268661499, 0.9259259104728699, 0.6172839403152466, 0.654321014881134, 0.7654321193695068, 0.5555555820465088, 0.654321014881134, 0.5802469253540039, 0.7407407164573669, 0.8888888955116272], [0.012345679104328156, 0.04938271641731262, 0.012345679104328156, 0.07407407462596893, 0.06172839552164078, 0.04938271641731262, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.06172839552164078, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.02469135820865631, 0.03703703731298447, 0.09876543283462524, 0.012345679104328156, 0.12345679104328156, 0.02469135820865631, 0.12345679104328156, 0.1111111119389534, 0.09876543283462524, 0.16049382090568542, 0.012345679104328156, 0.04938271641731262, 0.17283950746059418, 0.09876543283462524, 0.16049382090568542, 0.03703703731298447, 0.09876543283462524, 0.12345679104328156, 0.17283950746059418, 0.23456789553165436, 0.02469135820865631, 0.14814814925193787, 0.1358024626970291, 0.09876543283462524, 0.12345679104328156, 0.17283950746059418, 0.2222222238779068, 0.1358024626970291, 0.25925925374031067, 0.1358024626970291, 0.18518517911434174, 0.12345679104328156, 0.12345679104328156, 0.2839506268501282, 0.16049382090568542, 0.32098764181137085, 0.43209877610206604, 0.48148149251937866, 0.2222222238779068, 0.34567901492118835, 0.4444444477558136, 0.6419752836227417, NaN, 0.3827160596847534, 0.3827160596847534, 0.7037037014961243, 0.5925925970077515, 0.6419752836227417, 0.6296296119689941, 0.8024691343307495, 0.7654321193695068, 0.6419752836227417, 0.7654321193695068, 0.6666666865348816, 0.7160493731498718, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.08641975373029709, 0.012345679104328156, 0.02469135820865631, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.012345679104328156, 0.0, 0.0, 0.07407407462596893, 0.06172839552164078, 0.04938271641731262, 0.16049382090568542, 0.08641975373029709, 0.12345679104328156, 0.16049382090568542, 0.02469135820865631, 0.06172839552164078, 0.1111111119389534, 0.2222222238779068, 0.18518517911434174, 0.03703703731298447, 0.2469135820865631, 0.02469135820865631, 0.06172839552164078, 0.17283950746059418, 0.37037035822868347, 0.17283950746059418, 0.17283950746059418, 0.12345679104328156, 0.2839506268501282, 0.2222222238779068, 0.37037035822868347, 0.3086419701576233, 0.23456789553165436, 0.3827160596847534, 0.37037035822868347, 0.5308641791343689, 0.29629629850387573, 0.32098764181137085, 0.4691357910633087, 0.5185185074806213, 0.6172839403152466, NaN, 0.6604938507080078, 0.5308641791343689, 0.7037037014961243, 0.7037037014961243, 0.6666666865348816, 0.7160493731498718, 0.6666666865348816, 0.654321014881134, 0.6666666865348816, 1.0, 0.8641975522041321, 0.8641975522041321, 0.7777777910232544, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.08641975373029709, 0.012345679104328156, 0.02469135820865631, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06172839552164078, 0.06172839552164078, 0.04938271641731262, 0.16049382090568542, 0.08641975373029709, 0.12345679104328156, 0.16049382090568542, 0.03703703731298447, 0.04938271641731262, 0.1111111119389534, 0.2222222238779068, 0.17283950746059418, 0.03703703731298447, 0.2469135820865631, 0.012345679104328156, 0.06172839552164078, 0.17283950746059418, 0.3580246865749359, 0.17283950746059418, 0.16049382090568542, 0.1111111119389534, 0.2716049253940582, 0.20987653732299805, 0.34567901492118835, 0.3086419701576233, 0.2222222238779068, 0.3827160596847534, 0.37037035822868347, 0.5185185074806213, 0.29629629850387573, 0.3333333432674408, 0.45679011940956116, 0.5185185074806213, 0.6172839403152466, 0.3395061790943146, NaN, 0.5308641791343689, 0.7037037014961243, 0.7037037014961243, 0.6666666865348816, 0.7160493731498718, 0.6666666865348816, 0.654321014881134, 0.6666666865348816, 1.0, 0.8641975522041321, 0.8641975522041321, 0.7777777910232544, 0.8888888955116272], [0.0, 0.02469135820865631, 0.0, 0.07407407462596893, 0.03703703731298447, 0.06172839552164078, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06172839552164078, 0.0, 0.012345679104328156, 0.09876543283462524, 0.0, 0.012345679104328156, 0.16049382090568542, 0.0, 0.1111111119389534, 0.0, 0.14814814925193787, 0.17283950746059418, 0.09876543283462524, 0.2716049253940582, 0.07407407462596893, 0.1111111119389534, 0.2716049253940582, 0.09876543283462524, 0.2716049253940582, 0.0, 0.08641975373029709, 0.12345679104328156, 0.23456789553165436, 0.2716049253940582, 0.012345679104328156, 0.1358024626970291, 0.1111111119389534, 0.25925925374031067, 0.12345679104328156, 0.1358024626970291, 0.2839506268501282, 0.17283950746059418, 0.3333333432674408, 0.2222222238779068, 0.2839506268501282, 0.16049382090568542, 0.17283950746059418, 0.23456789553165436, 0.2469135820865631, 0.29629629850387573, 0.32098764181137085, 0.3827160596847534, 0.25925925374031067, 0.2839506268501282, 0.40740740299224854, 0.20370370149612427, 0.29629629850387573, 0.4691357910633087, 0.4691357910633087, NaN, 0.48148149251937866, 0.4938271641731262, 0.5679012537002563, 0.8641975522041321, 0.5679012537002563, 0.6296296119689941, 0.7283950448036194, 0.5555555820465088, 0.6172839403152466, 0.5802469253540039, 0.7407407164573669, 0.8888888955116272], [0.012345679104328156, 0.0, 0.02469135820865631, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.1358024626970291, 0.02469135820865631, 0.08641975373029709, 0.02469135820865631, 0.012345679104328156, 0.1358024626970291, 0.03703703731298447, 0.03703703731298447, 0.09876543283462524, 0.02469135820865631, 0.1358024626970291, 0.03703703731298447, 0.02469135820865631, 0.06172839552164078, 0.02469135820865631, 0.07407407462596893, 0.06172839552164078, 0.012345679104328156, 0.02469135820865631, 0.03703703731298447, 0.20987653732299805, 0.1358024626970291, 0.1358024626970291, 0.02469135820865631, 0.04938271641731262, 0.16049382090568542, 0.04938271641731262, 0.1358024626970291, 0.14814814925193787, 0.14814814925193787, 0.04938271641731262, 0.03703703731298447, 0.14814814925193787, 0.1358024626970291, 0.2716049253940582, 0.17283950746059418, 0.16049382090568542, 0.17283950746059418, 0.2716049253940582, 0.18518517911434174, 0.1975308656692505, 0.23456789553165436, 0.40740740299224854, 0.2222222238779068, 0.32098764181137085, 0.23456789553165436, 0.5185185074806213, 0.40740740299224854, 0.29629629850387573, 0.29629629850387573, 0.5185185074806213, NaN, 0.8395061492919922, 0.5802469253540039, 0.5185185074806213, 0.5925925970077515, 0.6666666865348816, 0.6296296119689941, 0.3827160596847534, 0.5679012537002563, 0.6172839403152466, 0.9135802388191223, 0.9506173133850098], [0.0, 0.0, 0.03703703731298447, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.03703703731298447, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.12345679104328156, 0.0, 0.03703703731298447, 0.02469135820865631, 0.012345679104328156, 0.1358024626970291, 0.04938271641731262, 0.02469135820865631, 0.03703703731298447, 0.012345679104328156, 0.14814814925193787, 0.06172839552164078, 0.012345679104328156, 0.03703703731298447, 0.0, 0.08641975373029709, 0.04938271641731262, 0.02469135820865631, 0.03703703731298447, 0.03703703731298447, 0.16049382090568542, 0.14814814925193787, 0.07407407462596893, 0.02469135820865631, 0.02469135820865631, 0.18518517911434174, 0.06172839552164078, 0.14814814925193787, 0.14814814925193787, 0.16049382090568542, 0.04938271641731262, 0.03703703731298447, 0.16049382090568542, 0.04938271641731262, 0.2839506268501282, 0.14814814925193787, 0.18518517911434174, 0.14814814925193787, 0.2222222238779068, 0.18518517911434174, 0.18518517911434174, 0.18518517911434174, 0.3580246865749359, 0.1975308656692505, 0.2469135820865631, 0.2222222238779068, 0.4938271641731262, 0.3580246865749359, 0.29629629850387573, 0.29629629850387573, 0.5061728358268738, 0.16049382090568542, NaN, 0.5555555820465088, 0.4938271641731262, 0.5925925970077515, 0.6666666865348816, 0.6296296119689941, 0.34567901492118835, 0.5061728358268738, 0.5308641791343689, 0.8888888955116272, 0.9506173133850098], [0.0, 0.012345679104328156, 0.012345679104328156, 0.0, 0.0, 0.0, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.0, 0.02469135820865631, 0.012345679104328156, 0.0, 0.012345679104328156, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.0, 0.03703703731298447, 0.03703703731298447, 0.012345679104328156, 0.012345679104328156, 0.08641975373029709, 0.02469135820865631, 0.12345679104328156, 0.08641975373029709, 0.06172839552164078, 0.08641975373029709, 0.06172839552164078, 0.04938271641731262, 0.07407407462596893, 0.16049382090568542, 0.18518517911434174, 0.1111111119389534, 0.07407407462596893, 0.1975308656692505, 0.06172839552164078, 0.17283950746059418, 0.07407407462596893, 0.03703703731298447, 0.16049382090568542, 0.02469135820865631, 0.08641975373029709, 0.1975308656692505, 0.3580246865749359, 0.32098764181137085, 0.14814814925193787, 0.06172839552164078, 0.32098764181137085, 0.395061731338501, 0.37037035822868347, 0.3333333432674408, 0.3333333432674408, 0.43209877610206604, 0.4197530746459961, 0.4444444477558136, NaN, 0.4691357910633087, 0.4444444477558136, 0.43209877610206604, 0.6790123581886292, 0.5061728358268738, 0.5432098507881165, 0.6172839403152466, 0.7654321193695068, 1.0], [0.0, 0.0, 0.0, 0.04938271641731262, 0.03703703731298447, 0.03703703731298447, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.04938271641731262, 0.0, 0.07407407462596893, 0.0, 0.07407407462596893, 0.08641975373029709, 0.03703703731298447, 0.07407407462596893, 0.0, 0.012345679104328156, 0.12345679104328156, 0.04938271641731262, 0.1358024626970291, 0.0, 0.06172839552164078, 0.1111111119389534, 0.1111111119389534, 0.20987653732299805, 0.0, 0.09876543283462524, 0.02469135820865631, 0.06172839552164078, 0.08641975373029709, 0.08641975373029709, 0.2222222238779068, 0.09876543283462524, 0.20987653732299805, 0.12345679104328156, 0.14814814925193787, 0.04938271641731262, 0.02469135820865631, 0.1111111119389534, 0.09876543283462524, 0.1111111119389534, 0.2839506268501282, 0.3827160596847534, 0.1358024626970291, 0.12345679104328156, 0.3580246865749359, 0.07407407462596893, 0.1975308656692505, 0.2839506268501282, 0.2839506268501282, 0.1358024626970291, 0.48148149251937866, 0.5061728358268738, 0.5308641791343689, NaN, 0.5432098507881165, 0.5555555820465088, 0.7160493731498718, 0.5308641791343689, 0.48148149251937866, 0.5802469253540039, 0.7407407164573669, 0.8765432238578796], [0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.0, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.0, 0.03703703731298447, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.04938271641731262, 0.02469135820865631, 0.1358024626970291, 0.08641975373029709, 0.08641975373029709, 0.09876543283462524, 0.1111111119389534, 0.04938271641731262, 0.07407407462596893, 0.14814814925193787, 0.17283950746059418, 0.08641975373029709, 0.06172839552164078, 0.1975308656692505, 0.03703703731298447, 0.16049382090568542, 0.04938271641731262, 0.03703703731298447, 0.16049382090568542, 0.012345679104328156, 0.06790123134851456, 0.14814814925193787, 0.3333333432674408, 0.2716049253940582, 0.12345679104328156, 0.04938271641731262, 0.32098764181137085, 0.3827160596847534, 0.23456789553165436, 0.3333333432674408, 0.3333333432674408, 0.43209877610206604, 0.40740740299224854, 0.40740740299224854, 0.5555555820465088, 0.45679011940956116, NaN, 0.43209877610206604, 0.790123462677002, 0.45679011940956116, 0.5432098507881165, 0.6172839403152466, 0.7654321193695068, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.0, 0.02469135820865631, 0.0, 0.0, 0.0, 0.012345679104328156, 0.04938271641731262, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.0, 0.1111111119389534, 0.03703703731298447, 0.1358024626970291, 0.012345679104328156, 0.012345679104328156, 0.0, 0.1358024626970291, 0.04938271641731262, 0.18518517911434174, 0.12345679104328156, 0.1358024626970291, 0.02469135820865631, 0.09876543283462524, 0.14814814925193787, 0.06172839552164078, 0.08641975373029709, 0.09876543283462524, 0.07407407462596893, 0.14814814925193787, 0.06172839552164078, 0.08641975373029709, 0.2469135820865631, 0.3086419701576233, 0.40740740299224854, 0.07407407462596893, 0.07407407462596893, 0.29629629850387573, 0.34567901492118835, 0.3580246865749359, 0.34567901492118835, 0.34567901492118835, 0.37037035822868347, 0.3333333432674408, 0.3333333432674408, 0.5679012537002563, 0.4444444477558136, 0.5679012537002563, NaN, 0.5925925970077515, 0.4938271641731262, 0.5925925970077515, 0.43209877610206604, 0.604938268661499, 0.9382715821266174], [0.0, 0.04938271641731262, 0.0, 0.08641975373029709, 0.08641975373029709, 0.08641975373029709, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.09876543283462524, 0.012345679104328156, 0.04938271641731262, 0.0, 0.0, 0.02469135820865631, 0.08641975373029709, 0.04938271641731262, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.09876543283462524, 0.07407407462596893, 0.012345679104328156, 0.04938271641731262, 0.08641975373029709, 0.1111111119389534, 0.09876543283462524, 0.0, 0.1975308656692505, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.012345679104328156, 0.1358024626970291, 0.0, 0.20987653732299805, 0.20987653732299805, 0.12345679104328156, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.12345679104328156, 0.1111111119389534, 0.12345679104328156, 0.18518517911434174, 0.09876543283462524, 0.16049382090568542, 0.12345679104328156, 0.2716049253940582, 0.2469135820865631, 0.12345679104328156, 0.1111111119389534, 0.20987653732299805, 0.23456789553165436, 0.23456789553165436, 0.3333333432674408, 0.3333333432674408, 0.2716049253940582, 0.37037035822868347, 0.37037035822868347, 0.32098764181137085, 0.2839506268501282, 0.20987653732299805, 0.40740740299224854, NaN, 0.40740740299224854, 0.40740740299224854, 0.43209877610206604, 0.5308641791343689, 0.8518518805503845], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.08641975373029709, 0.012345679104328156, 0.0, 0.1111111119389534, 0.0, 0.0, 0.02469135820865631, 0.09876543283462524, 0.0, 0.0, 0.012345679104328156, 0.0, 0.07407407462596893, 0.04938271641731262, 0.16049382090568542, 0.07407407462596893, 0.12345679104328156, 0.32098764181137085, 0.4197530746459961, 0.1111111119389534, 0.17283950746059418, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.0, 0.0, 0.4444444477558136, 0.6172839403152466, 0.654321014881134, 0.4938271641731262, 0.4691357910633087, 0.5432098507881165, 0.5061728358268738, 0.5925925970077515, NaN, 0.6419752836227417, 0.6666666865348816, 0.7777777910232544, 0.8641975522041321], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.07407407462596893, 0.0, 0.02469135820865631, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.02469135820865631, 0.03703703731298447, 0.0, 0.12345679104328156, 0.02469135820865631, 0.02469135820865631, 0.0, 0.04938271641731262, 0.0, 0.04938271641731262, 0.012345679104328156, 0.07407407462596893, 0.03703703731298447, 0.012345679104328156, 0.012345679104328156, 0.1111111119389534, 0.02469135820865631, 0.08641975373029709, 0.03703703731298447, 0.08641975373029709, 0.03703703731298447, 0.03703703731298447, 0.03703703731298447, 0.02469135820865631, 0.14814814925193787, 0.16049382090568542, 0.3827160596847534, 0.04938271641731262, 0.03703703731298447, 0.2222222238779068, 0.34567901492118835, 0.2839506268501282, 0.1358024626970291, 0.1358024626970291, 0.3827160596847534, 0.43209877610206604, 0.4938271641731262, 0.45679011940956116, 0.5185185074806213, 0.45679011940956116, 0.40740740299224854, 0.5925925970077515, 0.3580246865749359, NaN, 0.5061728358268738, 0.6666666865348816, 0.7777777910232544], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03703703731298447, 0.0, 0.0, 0.0, 0.03703703731298447, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.12345679104328156, 0.29629629850387573, 0.3580246865749359, 0.012345679104328156, 0.09876543283462524, 0.20987653732299805, 0.4197530746459961, 0.0, 0.1358024626970291, 0.1358024626970291, 0.4197530746459961, 0.3827160596847534, 0.4691357910633087, 0.3827160596847534, 0.4197530746459961, 0.3827160596847534, 0.5679012537002563, 0.5679012537002563, 0.3333333432674408, 0.4938271641731262, NaN, 0.790123462677002, 0.9876543283462524], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08641975373029709, 0.0, 0.0, 0.0, 0.0, 0.07407407462596893, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.08641975373029709, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.04938271641731262, 0.02469135820865631, 0.23456789553165436, 0.04938271641731262, 0.012345679104328156, 0.12345679104328156, 0.25925925374031067, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.25925925374031067, 0.08641975373029709, 0.1111111119389534, 0.23456789553165436, 0.25925925374031067, 0.23456789553165436, 0.395061731338501, 0.4691357910633087, 0.2222222238779068, 0.3333333432674408, 0.20987653732299805, NaN, 0.8641975522041321], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.12345679104328156, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.04938271641731262, 0.04938271641731262, 0.0, 0.12345679104328156, 0.0, 0.06172839552164078, 0.14814814925193787, 0.1358024626970291, 0.2222222238779068, 0.012345679104328156, 0.1358024626970291, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9419922233},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9403010403},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8802638327},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8613225148},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8100794876},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8065279926},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7919837676},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7911381736},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7871638797},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7801454429},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7762557097},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7610350113},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.757145272},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7546084925},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.742939293},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7246744487},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.7074243212},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7037037056},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7035345874},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6945712845},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6905124332},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.674277018},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.6668357841},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6497547779},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6440047355},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6340267208},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6299678678},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6202435348},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6027397267},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6020632498},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.5949602589},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5525114164},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5491290403},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5403348547},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5377980728},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5234229667},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5217317777},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5054963666},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4956874704},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.4950109936},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4787755787},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4632166432},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4472349052},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4410620656},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4317605267},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.4123118533},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3997970572},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3992897028},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3881278559},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3708777267},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3677490257},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3644512097},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.3638592919},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.3499069826},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3307965518},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3307965515},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.3111787586},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2615423643},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2496194822},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2495349237},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2427701684},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2312700826},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2053103317},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1829866398},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1642144424},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1608320649},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.1580416036},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.1527143586},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1513614064},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1403686793},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1248097419},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1155082027},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0899712498},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0206325046}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_binary/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..149c36da91037844344e538b861f66877eda9677
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,"AutoGluon 1.5 (extreme, 4h)",TabFM (default),"AutoGluon 1.4 (best, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),TabICLv2 (default),CatBoost (default),iLTM (tuned + ensembled),CatBoost (tuned + ensembled),RealMLP (tuned + ensembled),CatBoost (tuned),ChimeraBoost (tuned + ensembled),TabM (tuned + ensembled),XGBoost (tuned + ensembled),LightGBM (tuned + ensembled),ModernNCA (tuned + ensembled),ChimeraBoost (tuned),RealMLP (tuned),RealTabPFN-2.5 (tuned + ensembled),XGBoost (tuned),TabM (tuned),ChimeraBoost (default),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (default),LightGBM (tuned),TabICL (default),iLTM (tuned),ModernNCA (tuned),TabDPT (tuned + ensembled),TorchMLP (tuned + ensembled),TabDPT-Turbo (default),xRFM (tuned + ensembled),EBM (tuned + ensembled),XGBoost (default),TabM (default),TabSwift (default),xRFM (tuned),RealMLP (default),FastaiMLP (tuned + ensembled),BetaTabPFN (default),EBM (tuned),TorchMLP (tuned),TabDPT (tuned),EBM (default),LightGBM (default),iLTM (default),ModernNCA (default),RandomForest (tuned + ensembled),SAP-RPT-OSS (default),ExtraTrees (tuned + ensembled),RandomForest (tuned),FastaiMLP (tuned),TabDPT (default),LimiX (default),TorchMLP (default),ExtraTrees (tuned),FastaiMLP (default),Linear (tuned + ensembled),KNN (tuned + ensembled),TabSTAR (tuned),TabSTAR (tuned + ensembled),Linear (tuned),PerpetualBooster (tuned + ensembled),PerpetualBooster (tuned),ExtraTrees (default),Linear (default),RandomForest (default),OrionMSP (default),xRFM (default),TabSTAR (default),TabFlex (default),KNN (tuned),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.33333334,0.7160494,0.45679012,0.80246913,0.6419753,0.8888889,0.9259259,0.91358024,0.97530866,0.90123457,0.91358024,0.79012346,0.9506173,0.9259259,0.962963,0.9259259,0.962963,0.91358024,0.9876543,0.8518519,0.962963,0.91358024,0.91358024,0.97530866,0.8888889,0.8765432,0.9876543,0.91358024,0.9876543,0.90123457,0.97530866,0.9876543,0.962963,1.0,0.9382716,0.9876543,1.0,1.0,1.0,1.0,0.97530866,0.9506173,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9876543,1.0,0.97530866,0.9876543,1.0,1.0,1.0,1.0,0.9876543,1.0,1.0,1.0,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.6666667,,0.67901236,0.90123457,0.9506173,0.90123457,0.7654321,0.8271605,0.75308645,0.80246913,0.80246913,0.86419755,0.9876543,0.86419755,0.83950615,0.8765432,0.86419755,0.8148148,0.962963,0.90123457,1.0,0.8888889,0.962963,0.97530866,0.962963,0.9506173,0.8888889,0.9382716,0.962963,0.962963,0.962963,0.8765432,0.9876543,0.97530866,1.0,0.9506173,0.90123457,1.0,0.90123457,1.0,1.0,1.0,0.962963,1.0,1.0,1.0,1.0,0.97530866,0.97530866,0.9876543,0.97530866,0.97530866,0.97530866,0.9506173,1.0,0.9876543,0.97530866,0.97530866,0.9506173,1.0,1.0,0.97530866,1.0,1.0,0.9876543,1.0,0.9876543,1.0,0.9506173,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.28395063,0.32098764,,0.4691358,0.67901236,0.61728394,0.69135803,0.8271605,0.7407407,0.8518519,0.75308645,0.75308645,0.7160494,0.8518519,0.8271605,0.7654321,0.8518519,0.91358024,0.7037037,0.8765432,0.7777778,0.8765432,0.75308645,0.7407407,0.8888889,0.7407407,0.83950615,0.8888889,0.7777778,0.90123457,0.79012346,0.9506173,0.9506173,0.91358024,0.91358024,0.8271605,0.962963,0.962963,0.962963,0.90123457,0.9506173,0.97530866,0.962963,0.9506173,0.9506173,0.962963,0.9876543,0.9876543,0.962963,0.9876543,0.97530866,0.97530866,0.9506173,0.97530866,0.9876543,1.0,0.9876543,1.0,0.9876543,1.0,1.0,1.0,0.97530866,0.962963,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.54320985,0.09876543,0.5308642,,0.72839504,0.5308642,0.654321,0.67901236,0.6666667,0.7160494,0.654321,0.75308645,0.7654321,0.7160494,0.7037037,0.8765432,0.7407407,0.72839504,0.8518519,0.75308645,0.83950615,0.7407407,0.86419755,0.86419755,0.8888889,0.8518519,0.8518519,0.86419755,0.8765432,0.8148148,0.9506173,0.7654321,0.8518519,0.90123457,0.9876543,0.962963,0.7777778,0.9382716,0.8888889,0.9382716,0.90123457,0.962963,0.962963,0.9506173,0.9876543,0.9876543,0.97530866,0.9382716,0.91358024,0.962963,0.962963,0.9259259,0.962963,0.8888889,0.962963,0.962963,0.9259259,0.91358024,0.9259259,1.0,1.0,0.9259259,1.0,1.0,1.0,0.9506173,1.0,1.0,0.91358024,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.19753087,0.049382716,0.32098764,0.27160493,,0.5185185,0.4691358,0.50617284,0.5185185,0.5802469,0.5185185,0.6419753,0.5802469,0.5308642,0.654321,0.6666667,0.67901236,0.67901236,0.7777778,0.7037037,0.7160494,0.6419753,0.7777778,0.83950615,0.80246913,0.75308645,0.72839504,0.80246913,0.83950615,0.8271605,0.83950615,0.7407407,0.86419755,0.90123457,0.91358024,0.8518519,0.79012346,1.0,0.8518519,0.8888889,0.962963,0.9506173,0.9506173,0.962963,0.9382716,0.9382716,0.97530866,0.9506173,0.9506173,0.97530866,0.9876543,0.9382716,0.9382716,0.86419755,1.0,0.9876543,0.91358024,0.962963,0.9382716,1.0,1.0,0.962963,0.9876543,0.9876543,1.0,0.962963,1.0,1.0,0.91358024,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.3580247,0.09876543,0.38271606,0.4691358,0.4814815,,0.56790125,0.5925926,0.56790125,0.5308642,0.56790125,0.5802469,0.56790125,0.5802469,0.654321,0.67901236,0.5802469,0.5925926,0.8148148,0.60493827,0.69135803,0.60493827,0.83950615,0.8271605,0.75308645,0.8148148,0.72839504,0.7777778,0.8888889,0.79012346,0.86419755,0.6666667,0.75308645,0.8518519,0.9382716,0.90123457,0.69135803,0.90123457,0.8271605,0.91358024,0.90123457,0.90123457,0.9382716,0.9259259,0.9382716,0.91358024,0.9506173,0.962963,0.91358024,0.9876543,0.962963,0.9259259,0.9506173,0.90123457,0.97530866,0.9876543,0.91358024,0.91358024,0.9506173,0.97530866,0.97530866,0.9382716,1.0,1.0,1.0,0.962963,1.0,1.0,0.91358024,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.11111111,0.2345679,0.30864197,0.34567901,0.5308642,0.43209878,,0.5308642,0.45679012,0.5802469,0.5308642,0.5925926,0.5555556,0.5925926,0.60493827,0.5925926,0.7160494,0.6419753,0.5802469,0.72839504,0.61728394,0.79012346,0.60493827,0.7160494,0.7654321,0.7037037,0.72839504,0.69135803,0.654321,0.7654321,0.654321,0.9382716,0.90123457,0.8765432,0.7654321,0.79012346,0.9506173,0.80246913,0.8148148,0.83950615,0.90123457,0.9506173,0.79012346,0.9259259,0.9259259,0.9382716,0.9876543,0.97530866,0.9506173,0.962963,0.962963,0.97530866,0.8888889,0.83950615,0.97530866,0.9876543,0.9506173,0.9876543,0.9876543,1.0,1.0,1.0,0.9876543,0.9876543,0.97530866,1.0,0.9876543,0.9876543,1.0,1.0,0.9876543,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.074074075,0.17283951,0.17283951,0.32098764,0.49382716,0.4074074,0.4691358,,0.45679012,0.49382716,0.4691358,0.49382716,0.5308642,0.5925926,0.6296296,0.56790125,0.5802469,0.61728394,0.6419753,0.75308645,0.61728394,0.60493827,0.69135803,0.7407407,0.75308645,0.7160494,0.67901236,0.7654321,0.7160494,0.8518519,0.67901236,0.8765432,0.90123457,0.8765432,0.8148148,0.7407407,0.9506173,0.9506173,0.962963,0.8518519,0.91358024,0.9506173,0.90123457,0.9382716,0.91358024,1.0,0.97530866,0.97530866,0.97530866,0.9506173,0.9876543,0.962963,0.9259259,0.83950615,1.0,0.9876543,1.0,0.9876543,0.9876543,1.0,1.0,1.0,0.97530866,0.962963,0.9876543,1.0,0.9876543,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.086419754,0.24691358,0.25925925,0.33333334,0.4814815,0.43209878,0.54320985,0.54320985,,0.5802469,0.5802469,0.54320985,0.5555556,0.6296296,0.56790125,0.54320985,0.7037037,0.67901236,0.5802469,0.7037037,0.60493827,0.7654321,0.5925926,0.72839504,0.7037037,0.654321,0.7160494,0.70987654,0.6296296,0.7654321,0.6296296,0.91358024,0.9259259,0.8518519,0.7407407,0.75308645,0.9382716,0.8148148,0.8518519,0.8271605,0.90123457,0.9506173,0.79012346,0.9382716,0.9506173,0.91358024,0.9506173,0.97530866,0.97530866,0.9876543,0.962963,0.962963,0.8765432,0.80246913,0.9876543,0.97530866,0.97530866,0.9876543,0.9876543,1.0,1.0,1.0,0.97530866,0.9876543,0.9876543,1.0,0.9876543,0.9876543,1.0,1.0,0.9876543,0.9876543,1.0,1.0
+RealMLP (tuned + ensembled),0.024691358,0.19753087,0.14814815,0.28395063,0.41975307,0.4691358,0.41975307,0.50617284,0.41975307,,0.4691358,0.41975307,0.4691358,0.5555556,0.5802469,0.5925926,0.54320985,0.75308645,0.61728394,0.6666667,0.5308642,0.54320985,0.654321,0.67901236,0.75308645,0.6419753,0.69135803,0.7654321,0.67901236,0.83950615,0.7407407,0.8271605,0.83950615,0.8271605,0.8888889,0.72839504,0.8765432,1.0,0.97530866,0.83950615,0.91358024,0.9259259,0.90123457,0.9382716,0.9506173,0.90123457,0.9506173,0.97530866,0.962963,1.0,0.962963,0.97530866,0.90123457,0.9382716,0.9876543,0.97530866,0.962963,1.0,0.9876543,1.0,1.0,1.0,0.97530866,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.09876543,0.19753087,0.24691358,0.34567901,0.4814815,0.43209878,0.4691358,0.5308642,0.41975307,0.5308642,,0.5802469,0.56790125,0.6296296,0.5185185,0.5555556,0.7037037,0.6666667,0.54320985,0.72839504,0.5802469,0.72839504,0.5925926,0.67901236,0.7160494,0.6666667,0.67901236,0.654321,0.6296296,0.75308645,0.61728394,0.9259259,0.91358024,0.83950615,0.7777778,0.7407407,0.9259259,0.8271605,0.80246913,0.83950615,0.90123457,0.9382716,0.7777778,0.962963,0.91358024,0.9259259,0.9259259,0.97530866,0.962963,0.9506173,0.962963,0.9506173,0.86419755,0.80246913,0.97530866,0.962963,0.9382716,0.9876543,0.9876543,0.9876543,0.9876543,1.0,0.97530866,0.9876543,0.97530866,1.0,0.9876543,0.97530866,1.0,1.0,0.9876543,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.086419754,0.13580246,0.24691358,0.24691358,0.3580247,0.41975307,0.4074074,0.50617284,0.45679012,0.5802469,0.41975307,,0.56790125,0.5308642,0.5802469,0.50617284,0.7777778,0.69135803,0.56790125,0.654321,0.654321,0.86419755,0.6296296,0.6296296,0.6296296,0.61728394,0.7037037,0.69135803,0.6666667,0.67901236,0.67901236,0.8888889,0.91358024,0.75308645,0.80246913,0.6296296,0.91358024,0.83950615,0.83950615,0.86419755,0.9506173,0.8518519,0.8148148,0.97530866,0.86419755,0.9506173,0.90123457,0.9876543,0.97530866,0.8765432,0.9876543,0.8765432,0.8148148,0.8271605,0.90123457,0.8765432,0.86419755,1.0,0.9876543,1.0,1.0,1.0,0.86419755,0.8765432,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.91358024,1.0
+TabM (tuned + ensembled),0.20987654,0.012345679,0.28395063,0.2345679,0.41975307,0.43209878,0.44444445,0.4691358,0.44444445,0.5308642,0.43209878,0.43209878,,0.41975307,0.5185185,0.5802469,0.50617284,0.61728394,0.6296296,0.5555556,0.72839504,0.5802469,0.654321,0.6296296,0.67901236,0.6419753,0.7037037,0.8148148,0.72839504,0.6666667,0.79012346,0.61728394,0.7407407,0.7407407,0.962963,0.83950615,0.61728394,0.97530866,0.80246913,0.90123457,0.9382716,0.86419755,0.8888889,0.9259259,1.0,0.9259259,0.962963,0.9259259,0.90123457,0.9259259,0.9259259,0.90123457,0.8765432,0.8518519,0.97530866,0.9506173,0.8888889,0.9259259,0.9382716,1.0,1.0,0.9382716,0.97530866,1.0,1.0,0.97530866,1.0,1.0,0.90123457,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.049382716,0.13580246,0.14814815,0.28395063,0.4691358,0.41975307,0.4074074,0.4074074,0.37037036,0.44444445,0.37037036,0.4691358,0.5802469,,0.50617284,0.4691358,0.5308642,0.54320985,0.5555556,0.86419755,0.6419753,0.6296296,0.5925926,0.6419753,0.6666667,0.5555556,0.5925926,0.60493827,0.654321,0.7777778,0.654321,0.80246913,0.90123457,0.9382716,0.79012346,0.654321,0.90123457,0.8271605,0.7777778,0.86419755,0.9382716,0.8271605,0.8271605,0.97530866,0.962963,0.962963,0.9876543,1.0,0.9506173,0.9876543,0.9876543,0.86419755,0.8765432,0.75308645,1.0,0.9876543,0.8888889,1.0,0.9876543,0.9876543,0.9876543,1.0,0.91358024,0.962963,0.9876543,1.0,1.0,1.0,0.9876543,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.074074075,0.16049382,0.17283951,0.2962963,0.34567901,0.34567901,0.39506173,0.37037036,0.43209878,0.41975307,0.4814815,0.41975307,0.4814815,0.49382716,,0.45679012,0.56790125,0.5185185,0.5185185,0.654321,0.5925926,0.5802469,0.5555556,0.5802469,0.86419755,0.6296296,0.6666667,0.61728394,0.56790125,0.7037037,0.69135803,0.7407407,0.8148148,0.80246913,0.7777778,0.7160494,0.8148148,0.80246913,0.7407407,0.8518519,0.90123457,0.90123457,0.7777778,0.962963,0.97530866,0.97530866,0.97530866,0.962963,0.9506173,0.97530866,0.9506173,0.90123457,0.9506173,0.7654321,0.97530866,0.97530866,0.8518519,0.97530866,0.97530866,0.9876543,0.9876543,0.9876543,0.97530866,0.97530866,0.9876543,1.0,1.0,0.9876543,0.9506173,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.037037037,0.12345679,0.2345679,0.12345679,0.33333334,0.32098764,0.4074074,0.43209878,0.45679012,0.4074074,0.44444445,0.49382716,0.41975307,0.5308642,0.54320985,,0.5925926,0.54320985,0.5925926,0.5555556,0.5185185,0.5925926,0.60493827,0.61728394,0.5925926,0.60493827,0.60493827,0.7160494,0.654321,0.7037037,0.67901236,0.67901236,0.8148148,0.72839504,0.80246913,0.6666667,0.75308645,0.8765432,0.8518519,0.80246913,0.83950615,0.8271605,0.86419755,0.8518519,0.90123457,0.8271605,0.9506173,0.8765432,0.83950615,0.8888889,0.86419755,0.9259259,0.91358024,0.9506173,0.8765432,0.8765432,0.962963,0.8888889,0.9876543,0.91358024,0.91358024,0.90123457,0.9876543,0.9876543,0.97530866,0.97530866,1.0,0.9506173,1.0,1.0,0.9259259,1.0,1.0,1.0
+ChimeraBoost (tuned),0.074074075,0.13580246,0.14814815,0.25925925,0.32098764,0.41975307,0.28395063,0.41975307,0.2962963,0.45679012,0.2962963,0.22222222,0.49382716,0.4691358,0.43209878,0.4074074,,0.6419753,0.5308642,0.5802469,0.5555556,0.67901236,0.54320985,0.56790125,0.56790125,0.54320985,0.60493827,0.6419753,0.61728394,0.60493827,0.5802469,0.8518519,0.80246913,0.72839504,0.7654321,0.5925926,0.8888889,0.8271605,0.79012346,0.83950615,0.86419755,0.80246913,0.79012346,0.90123457,0.79012346,0.8888889,0.83950615,0.962963,0.9259259,0.8518519,0.962963,0.83950615,0.75308645,0.8148148,0.90123457,0.8518519,0.8518519,1.0,0.97530866,0.9876543,0.9876543,1.0,0.86419755,0.86419755,0.9876543,1.0,0.9876543,0.9876543,1.0,1.0,1.0,1.0,0.9259259,1.0
+RealMLP (tuned),0.037037037,0.18518518,0.086419754,0.27160493,0.32098764,0.4074074,0.3580247,0.38271606,0.32098764,0.24691358,0.33333334,0.30864197,0.38271606,0.45679012,0.4814815,0.45679012,0.3580247,,0.49382716,0.5925926,0.44444445,0.4814815,0.5308642,0.5802469,0.6666667,0.5308642,0.6419753,0.6419753,0.61728394,0.75308645,0.54320985,0.72839504,0.7160494,0.75308645,0.7407407,0.60493827,0.7777778,0.8888889,0.86419755,0.75308645,0.79012346,0.90123457,0.8148148,0.8148148,0.83950615,0.8148148,0.83950615,0.9382716,0.8765432,0.8888889,0.91358024,0.962963,0.8271605,0.8888889,0.9259259,0.9506173,0.9506173,0.97530866,0.962963,0.97530866,0.97530866,0.9876543,0.962963,0.9506173,0.9876543,1.0,0.9876543,0.97530866,0.97530866,1.0,0.97530866,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.086419754,0.037037037,0.2962963,0.14814815,0.22222222,0.18518518,0.41975307,0.3580247,0.41975307,0.38271606,0.45679012,0.43209878,0.37037036,0.44444445,0.4814815,0.4074074,0.4691358,0.50617284,,0.49382716,0.50617284,0.56790125,0.7777778,0.67901236,0.60493827,0.6666667,0.5802469,0.654321,0.654321,0.67901236,0.60493827,0.61728394,0.7037037,0.7407407,0.7160494,0.83950615,0.67901236,0.80246913,0.7037037,0.7777778,0.80246913,0.80246913,0.8765432,0.8518519,0.91358024,0.83950615,0.80246913,0.90123457,0.83950615,0.962963,0.9506173,0.86419755,0.8888889,0.7407407,0.9506173,0.962963,0.8888889,0.8271605,0.90123457,1.0,1.0,0.83950615,0.962963,0.97530866,0.9876543,0.9506173,1.0,0.9876543,0.91358024,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.012345679,0.09876543,0.12345679,0.24691358,0.2962963,0.39506173,0.27160493,0.24691358,0.2962963,0.33333334,0.27160493,0.34567901,0.44444445,0.13580246,0.34567901,0.44444445,0.41975307,0.4074074,0.50617284,,0.56790125,0.5555556,0.5185185,0.5555556,0.56790125,0.49382716,0.5185185,0.5925926,0.56790125,0.6419753,0.60493827,0.7160494,0.80246913,0.8148148,0.75308645,0.6296296,0.79012346,0.80246913,0.72839504,0.83950615,0.90123457,0.80246913,0.7777778,0.91358024,0.9259259,0.9259259,0.962963,0.962963,0.9382716,0.9506173,0.9382716,0.8518519,0.83950615,0.72839504,1.0,0.97530866,0.83950615,1.0,0.9876543,0.9876543,0.9876543,1.0,0.90123457,0.962963,0.9876543,1.0,1.0,1.0,0.9506173,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.14814815,0.0,0.22222222,0.16049382,0.28395063,0.30864197,0.38271606,0.38271606,0.39506173,0.4691358,0.41975307,0.34567901,0.27160493,0.3580247,0.4074074,0.4814815,0.44444445,0.5555556,0.49382716,0.43209878,,0.45679012,0.5185185,0.54320985,0.49382716,0.54320985,0.654321,0.7037037,0.60493827,0.61728394,0.654321,0.56790125,0.61728394,0.72839504,0.8271605,0.72839504,0.61728394,0.8888889,0.7777778,0.7777778,0.8518519,0.8271605,0.80246913,0.8765432,0.86419755,0.83950615,0.90123457,0.86419755,0.8518519,0.8518519,0.8765432,0.86419755,0.8271605,0.8518519,0.97530866,0.91358024,0.90123457,0.8765432,0.8765432,1.0,1.0,0.8888889,0.97530866,0.9876543,0.97530866,0.9259259,0.9876543,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.037037037,0.11111111,0.12345679,0.25925925,0.3580247,0.39506173,0.20987654,0.39506173,0.2345679,0.45679012,0.27160493,0.13580246,0.41975307,0.37037036,0.41975307,0.4074074,0.32098764,0.5185185,0.43209878,0.44444445,0.54320985,,0.4814815,0.4814815,0.50617284,0.50617284,0.61728394,0.60493827,0.5555556,0.60493827,0.5925926,0.7777778,0.72839504,0.67901236,0.6419753,0.60493827,0.86419755,0.72839504,0.79012346,0.72839504,0.80246913,0.75308645,0.8148148,0.8518519,0.72839504,0.91358024,0.79012346,0.9382716,0.9506173,0.8518519,0.90123457,0.8518519,0.80246913,0.8148148,0.8518519,0.83950615,0.8518519,1.0,0.97530866,0.9876543,1.0,1.0,0.86419755,0.8518519,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0
+RealTabPFN-2.5 (tuned),0.086419754,0.037037037,0.24691358,0.13580246,0.22222222,0.16049382,0.39506173,0.30864197,0.4074074,0.34567901,0.4074074,0.37037036,0.34567901,0.4074074,0.44444445,0.39506173,0.45679012,0.4691358,0.22222222,0.4814815,0.4814815,0.5185185,,0.6296296,0.60493827,0.6296296,0.5308642,0.56790125,0.654321,0.61728394,0.54320985,0.61728394,0.6296296,0.72839504,0.6419753,0.8148148,0.654321,0.7407407,0.6666667,0.72839504,0.7037037,0.79012346,0.80246913,0.8148148,0.8765432,0.80246913,0.7654321,0.83950615,0.8148148,0.90123457,0.9259259,0.86419755,0.8271605,0.72839504,0.90123457,0.9259259,0.8765432,0.80246913,0.8765432,0.9876543,1.0,0.8518519,0.962963,0.9382716,0.97530866,0.9259259,0.9876543,0.9876543,0.8888889,1.0,0.9876543,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.086419754,0.024691358,0.25925925,0.13580246,0.16049382,0.17283951,0.28395063,0.25925925,0.27160493,0.32098764,0.32098764,0.37037036,0.37037036,0.3580247,0.41975307,0.38271606,0.43209878,0.41975307,0.32098764,0.44444445,0.45679012,0.5185185,0.37037036,,0.5555556,0.5802469,0.5802469,0.50617284,0.5802469,0.5802469,0.49382716,0.6419753,0.5925926,0.72839504,0.54320985,0.8518519,0.654321,0.67901236,0.6296296,0.7654321,0.7037037,0.7777778,0.75308645,0.75308645,0.86419755,0.8148148,0.7777778,0.8271605,0.8888889,0.91358024,0.8888889,0.7777778,0.83950615,0.7407407,0.9506173,0.9382716,0.8271605,0.8271605,0.8888889,1.0,1.0,0.8271605,0.97530866,0.9876543,0.9876543,0.91358024,0.97530866,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.024691358,0.037037037,0.11111111,0.11111111,0.19753087,0.24691358,0.2345679,0.24691358,0.2962963,0.24691358,0.28395063,0.37037036,0.32098764,0.33333334,0.13580246,0.4074074,0.43209878,0.33333334,0.39506173,0.43209878,0.50617284,0.49382716,0.39506173,0.44444445,,0.4691358,0.49382716,0.50617284,0.5308642,0.5308642,0.61728394,0.69135803,0.72839504,0.7160494,0.72839504,0.5925926,0.69135803,0.7407407,0.72839504,0.7654321,0.7654321,0.6666667,0.7160494,0.83950615,0.962963,0.9259259,0.9259259,0.8518519,0.80246913,0.8148148,0.8765432,0.7654321,0.8148148,0.7160494,0.9259259,0.91358024,0.79012346,0.8518519,0.90123457,1.0,1.0,0.90123457,0.9382716,0.962963,0.9876543,0.962963,0.9876543,0.9876543,0.90123457,1.0,0.9876543,1.0,1.0,1.0
+TabICL (default),0.11111111,0.049382716,0.25925925,0.14814815,0.24691358,0.18518518,0.2962963,0.28395063,0.34567901,0.3580247,0.33333334,0.38271606,0.3580247,0.44444445,0.37037036,0.39506173,0.45679012,0.4691358,0.33333334,0.50617284,0.45679012,0.49382716,0.37037036,0.41975307,0.5308642,,0.4814815,0.45679012,0.60493827,0.5925926,0.5925926,0.61728394,0.69135803,0.67901236,0.56790125,0.61728394,0.6666667,0.6419753,0.654321,0.654321,0.7037037,0.6419753,0.83950615,0.72839504,0.8518519,0.7407407,0.8518519,0.79012346,0.75308645,0.7777778,0.80246913,0.67901236,0.91358024,0.7654321,0.7777778,0.8148148,0.7654321,0.72839504,0.83950615,0.9259259,0.9382716,0.72839504,0.97530866,1.0,1.0,0.9259259,1.0,1.0,0.9259259,1.0,0.97530866,1.0,1.0,1.0
+iLTM (tuned),0.12345679,0.11111111,0.16049382,0.14814815,0.27160493,0.27160493,0.27160493,0.32098764,0.28395063,0.30864197,0.32098764,0.2962963,0.2962963,0.4074074,0.33333334,0.39506173,0.39506173,0.3580247,0.41975307,0.4814815,0.34567901,0.38271606,0.4691358,0.41975307,0.50617284,0.5185185,,0.4814815,0.50617284,0.44444445,0.5555556,0.67901236,0.56790125,0.5555556,0.5802469,0.56790125,0.67901236,0.61728394,0.7037037,0.61728394,0.6419753,0.5802469,0.7654321,0.654321,0.7777778,0.7407407,0.8271605,0.8148148,0.8271605,0.8148148,0.83950615,0.83950615,0.83950615,0.8518519,0.8765432,0.8518519,0.8888889,0.8888889,0.9876543,0.9382716,0.9382716,0.9259259,0.9259259,0.91358024,0.962963,1.0,0.962963,0.97530866,0.9876543,0.9876543,0.9876543,1.0,1.0,1.0
+ModernNCA (tuned),0.012345679,0.061728396,0.11111111,0.13580246,0.19753087,0.22222222,0.30864197,0.2345679,0.29012346,0.2345679,0.34567901,0.30864197,0.18518518,0.39506173,0.38271606,0.28395063,0.3580247,0.3580247,0.34567901,0.4074074,0.2962963,0.39506173,0.43209878,0.49382716,0.49382716,0.54320985,0.5185185,,0.5308642,0.56790125,0.5925926,0.50617284,0.5802469,0.5925926,0.5802469,0.60493827,0.5802469,0.69135803,0.6296296,0.654321,0.654321,0.69135803,0.75308645,0.69135803,0.7407407,0.7407407,0.8888889,0.80246913,0.7654321,0.8271605,0.79012346,0.91358024,0.8518519,0.80246913,0.83950615,0.83950615,0.90123457,0.8518519,0.9506173,0.9506173,0.9506173,0.8888889,0.9382716,0.9506173,0.962963,0.9876543,0.97530866,0.9876543,0.9506173,1.0,0.97530866,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.086419754,0.037037037,0.22222222,0.12345679,0.16049382,0.11111111,0.34567901,0.28395063,0.37037036,0.32098764,0.37037036,0.33333334,0.27160493,0.34567901,0.43209878,0.34567901,0.38271606,0.38271606,0.34567901,0.43209878,0.39506173,0.44444445,0.34567901,0.41975307,0.4691358,0.39506173,0.49382716,0.4691358,,0.5802469,0.45679012,0.56790125,0.5555556,0.60493827,0.5555556,0.6419753,0.61728394,0.60493827,0.654321,0.5802469,0.5925926,0.72839504,0.8765432,0.6296296,0.75308645,0.7037037,0.7654321,0.7407407,0.7160494,0.75308645,0.72839504,0.7777778,0.9382716,0.7407407,0.75308645,0.79012346,0.8148148,0.72839504,0.8271605,0.83950615,0.83950615,0.72839504,0.9876543,0.97530866,0.9876543,0.8765432,0.9876543,0.9876543,0.91358024,1.0,0.962963,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.012345679,0.037037037,0.09876543,0.18518518,0.17283951,0.20987654,0.2345679,0.14814815,0.2345679,0.16049382,0.24691358,0.32098764,0.33333334,0.22222222,0.2962963,0.2962963,0.39506173,0.24691358,0.32098764,0.3580247,0.38271606,0.39506173,0.38271606,0.41975307,0.4691358,0.4074074,0.5555556,0.43209878,0.41975307,,0.39506173,0.5555556,0.5185185,0.6296296,0.5802469,0.6419753,0.61728394,0.61728394,0.61728394,0.5555556,0.6666667,0.9259259,0.67901236,0.7037037,0.72839504,0.8271605,0.7160494,0.7777778,0.80246913,0.8765432,0.83950615,0.8148148,0.7407407,0.7407407,0.9876543,0.90123457,0.8148148,0.90123457,0.90123457,0.91358024,0.91358024,0.90123457,0.97530866,0.962963,0.9876543,0.9506173,0.9876543,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.09876543,0.037037037,0.20987654,0.049382716,0.16049382,0.13580246,0.34567901,0.32098764,0.37037036,0.25925925,0.38271606,0.32098764,0.20987654,0.34567901,0.30864197,0.32098764,0.41975307,0.45679012,0.39506173,0.39506173,0.34567901,0.4074074,0.45679012,0.50617284,0.38271606,0.4074074,0.44444445,0.4074074,0.54320985,0.60493827,,0.5802469,0.5308642,0.5925926,0.5185185,0.67901236,0.61728394,0.6666667,0.60493827,0.56790125,0.5802469,0.7037037,0.8148148,0.67901236,0.7654321,0.654321,0.6666667,0.7160494,0.6666667,0.7407407,0.7777778,0.7160494,0.90123457,0.8518519,0.7777778,0.80246913,0.86419755,0.7037037,0.83950615,0.8765432,0.8765432,0.72839504,0.962963,0.962963,0.91358024,0.86419755,0.9506173,0.8888889,0.90123457,1.0,0.8765432,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.024691358,0.12345679,0.049382716,0.2345679,0.25925925,0.33333334,0.061728396,0.12345679,0.086419754,0.17283951,0.074074075,0.11111111,0.38271606,0.19753087,0.25925925,0.32098764,0.14814815,0.27160493,0.38271606,0.28395063,0.43209878,0.22222222,0.38271606,0.3580247,0.30864197,0.38271606,0.32098764,0.49382716,0.43209878,0.44444445,0.41975307,,0.60493827,0.4074074,0.50617284,0.41975307,0.6419753,0.5802469,0.7037037,0.5925926,0.67901236,0.6419753,0.5185185,0.69135803,0.5185185,0.60493827,0.6296296,0.7160494,0.7654321,0.8148148,0.72839504,0.7777778,0.654321,0.7654321,0.7407407,0.80246913,0.8518519,0.97530866,0.962963,0.83950615,0.83950615,1.0,0.79012346,0.83950615,0.97530866,1.0,0.97530866,0.962963,1.0,0.8888889,0.97530866,0.962963,0.8888889,1.0
+EBM (tuned + ensembled),0.012345679,0.012345679,0.049382716,0.14814815,0.13580246,0.24691358,0.09876543,0.09876543,0.074074075,0.16049382,0.086419754,0.086419754,0.25925925,0.09876543,0.18518518,0.18518518,0.19753087,0.28395063,0.2962963,0.19753087,0.38271606,0.27160493,0.37037036,0.4074074,0.27160493,0.30864197,0.43209878,0.41975307,0.44444445,0.4814815,0.4691358,0.39506173,,0.56790125,0.5555556,0.5308642,0.54320985,0.69135803,0.6296296,0.654321,0.75308645,0.7654321,0.6666667,0.91358024,0.6419753,0.69135803,0.69135803,0.7407407,0.86419755,0.7777778,0.7777778,0.72839504,0.6419753,0.6666667,0.8518519,0.8271605,0.7777778,0.8888889,0.90123457,0.97530866,0.962963,0.91358024,0.86419755,0.8518519,0.8765432,0.9382716,0.86419755,0.86419755,0.80246913,1.0,0.97530866,1.0,0.8888889,1.0
+XGBoost (default),0.037037037,0.024691358,0.086419754,0.09876543,0.09876543,0.14814815,0.12345679,0.12345679,0.14814815,0.17283951,0.16049382,0.24691358,0.25925925,0.061728396,0.19753087,0.27160493,0.27160493,0.24691358,0.25925925,0.18518518,0.27160493,0.32098764,0.27160493,0.27160493,0.28395063,0.32098764,0.44444445,0.4074074,0.39506173,0.37037036,0.4074074,0.5925926,0.43209878,,0.4814815,0.49382716,0.6419753,0.49382716,0.654321,0.54320985,0.54320985,0.6296296,0.61728394,0.60493827,0.4814815,0.79012346,0.69135803,0.83950615,0.8148148,0.8518519,0.8518519,0.67901236,0.61728394,0.67901236,0.91358024,0.8888889,0.7407407,0.86419755,0.8765432,0.9382716,0.9506173,0.8765432,0.86419755,0.9259259,0.91358024,0.8888889,0.91358024,0.9876543,0.8888889,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.086419754,0.012345679,0.086419754,0.061728396,0.2345679,0.18518518,0.25925925,0.11111111,0.22222222,0.19753087,0.037037037,0.20987654,0.22222222,0.19753087,0.2345679,0.25925925,0.28395063,0.24691358,0.17283951,0.3580247,0.3580247,0.45679012,0.27160493,0.43209878,0.41975307,0.41975307,0.44444445,0.41975307,0.4814815,0.49382716,0.44444445,0.5185185,,0.50617284,0.5308642,0.56790125,0.61728394,0.5555556,0.5925926,0.56790125,0.67901236,0.6419753,0.69135803,0.67901236,0.80246913,0.7407407,0.69135803,0.7160494,0.7407407,0.75308645,0.69135803,0.6419753,0.80246913,0.75308645,0.79012346,0.7777778,0.8271605,0.8888889,0.8888889,0.7654321,0.97530866,0.97530866,0.9382716,0.8888889,0.91358024,0.9876543,0.8888889,1.0,0.9506173,1.0,1.0,1.0
+TabSwift (default),0.061728396,0.049382716,0.17283951,0.037037037,0.14814815,0.09876543,0.20987654,0.25925925,0.24691358,0.27160493,0.25925925,0.37037036,0.16049382,0.34567901,0.28395063,0.33333334,0.4074074,0.39506173,0.16049382,0.37037036,0.27160493,0.39506173,0.18518518,0.14814815,0.4074074,0.38271606,0.43209878,0.39506173,0.3580247,0.3580247,0.32098764,0.5802469,0.4691358,0.50617284,0.49382716,,0.5308642,0.41975307,0.5802469,0.5185185,0.5308642,0.5185185,0.5802469,0.5555556,0.5802469,0.67901236,0.61728394,0.5925926,0.5925926,0.69135803,0.61728394,0.6419753,0.83950615,0.75308645,0.654321,0.69135803,0.6419753,0.69135803,0.7654321,0.7777778,0.7777778,0.72839504,0.9506173,0.97530866,0.91358024,0.79012346,0.90123457,1.0,0.8888889,0.91358024,1.0,0.962963,1.0,1.0
+xRFM (tuned),0.012345679,0.09876543,0.037037037,0.22222222,0.20987654,0.30864197,0.049382716,0.049382716,0.061728396,0.12345679,0.074074075,0.086419754,0.38271606,0.09876543,0.18518518,0.24691358,0.11111111,0.22222222,0.32098764,0.20987654,0.38271606,0.13580246,0.34567901,0.34567901,0.30864197,0.33333334,0.32098764,0.41975307,0.38271606,0.38271606,0.38271606,0.3580247,0.45679012,0.3580247,0.4691358,0.4691358,,0.5555556,0.6296296,0.5925926,0.61728394,0.60493827,0.56790125,0.5802469,0.50617284,0.5308642,0.6296296,0.69135803,0.8888889,0.72839504,0.72839504,0.80246913,0.6296296,0.69135803,0.75308645,0.8271605,0.8518519,0.9506173,0.97530866,0.8148148,0.8271605,0.9876543,0.83950615,0.8148148,0.9382716,1.0,0.8888889,0.86419755,0.9876543,0.9876543,0.9506173,1.0,0.8888889,1.0
+RealMLP (default),0.0,0.0,0.037037037,0.061728396,0.0,0.09876543,0.19753087,0.049382716,0.18518518,0.0,0.17283951,0.16049382,0.024691358,0.17283951,0.19753087,0.12345679,0.17283951,0.11111111,0.19753087,0.19753087,0.11111111,0.27160493,0.25925925,0.32098764,0.25925925,0.3580247,0.38271606,0.30864197,0.39506173,0.38271606,0.33333334,0.41975307,0.30864197,0.50617284,0.43209878,0.5802469,0.44444445,,0.44444445,0.50617284,0.44444445,0.6419753,0.6296296,0.4814815,0.6419753,0.67901236,0.7037037,0.6666667,0.69135803,0.7654321,0.7407407,0.80246913,0.7037037,0.67901236,0.8765432,0.79012346,0.8271605,0.8148148,0.8518519,0.962963,0.962963,0.86419755,0.9506173,0.9382716,0.9506173,0.90123457,0.9506173,0.9506173,0.86419755,1.0,0.9876543,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.09876543,0.037037037,0.11111111,0.14814815,0.17283951,0.18518518,0.037037037,0.14814815,0.024691358,0.19753087,0.16049382,0.19753087,0.22222222,0.25925925,0.14814815,0.20987654,0.13580246,0.2962963,0.27160493,0.22222222,0.20987654,0.33333334,0.37037036,0.27160493,0.34567901,0.2962963,0.37037036,0.34567901,0.38271606,0.39506173,0.2962963,0.37037036,0.34567901,0.38271606,0.41975307,0.37037036,0.5555556,,0.44444445,0.49382716,0.54320985,0.5802469,0.5308642,0.5802469,0.49382716,0.69135803,0.6419753,0.67901236,0.654321,0.6666667,0.9259259,0.69135803,0.61728394,0.75308645,0.69135803,0.962963,0.8765432,0.86419755,0.75308645,0.75308645,0.8888889,0.86419755,0.8518519,0.9259259,0.97530866,0.9259259,0.8148148,1.0,0.8888889,0.9259259,0.8888889,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.09876543,0.061728396,0.11111111,0.086419754,0.16049382,0.14814815,0.17283951,0.16049382,0.16049382,0.13580246,0.09876543,0.13580246,0.14814815,0.19753087,0.16049382,0.24691358,0.22222222,0.16049382,0.22222222,0.27160493,0.27160493,0.2345679,0.2345679,0.34567901,0.38271606,0.34567901,0.41975307,0.44444445,0.43209878,0.4074074,0.34567901,0.45679012,0.44444445,0.4814815,0.4074074,0.49382716,0.5555556,,0.41975307,0.5555556,0.67901236,0.5802469,0.654321,0.60493827,0.6666667,0.61728394,0.7160494,0.67901236,0.654321,0.7037037,0.6419753,0.72839504,0.7407407,0.7037037,0.7407407,0.72839504,0.90123457,0.97530866,0.9876543,0.7407407,0.8518519,0.8518519,0.83950615,0.9382716,0.8518519,0.8765432,0.79012346,1.0,0.962963,1.0,0.8888889,1.0
+EBM (tuned),0.0,0.0,0.049382716,0.09876543,0.037037037,0.09876543,0.09876543,0.086419754,0.09876543,0.086419754,0.09876543,0.049382716,0.061728396,0.061728396,0.09876543,0.16049382,0.13580246,0.20987654,0.19753087,0.09876543,0.14814815,0.19753087,0.2962963,0.2962963,0.2345679,0.2962963,0.3580247,0.34567901,0.4074074,0.33333334,0.41975307,0.32098764,0.24691358,0.45679012,0.4074074,0.4691358,0.38271606,0.5555556,0.50617284,0.5802469,,0.5802469,0.54320985,0.7160494,0.54320985,0.5925926,0.6419753,0.6666667,0.80246913,0.654321,0.69135803,0.7037037,0.5555556,0.654321,0.8271605,0.72839504,0.7160494,0.83950615,0.8765432,0.9382716,0.9382716,0.8765432,0.8518519,0.83950615,0.8148148,0.91358024,0.8271605,0.86419755,0.79012346,1.0,0.9876543,1.0,0.8888889,1.0
+TorchMLP (tuned),0.024691358,0.0,0.024691358,0.037037037,0.049382716,0.09876543,0.049382716,0.049382716,0.049382716,0.074074075,0.061728396,0.14814815,0.13580246,0.17283951,0.09876543,0.17283951,0.19753087,0.09876543,0.19753087,0.19753087,0.17283951,0.24691358,0.20987654,0.22222222,0.33333334,0.3580247,0.41975307,0.30864197,0.27160493,0.074074075,0.2962963,0.3580247,0.2345679,0.37037036,0.43209878,0.4814815,0.39506173,0.3580247,0.45679012,0.44444445,0.41975307,,0.43209878,0.44444445,0.5308642,0.7037037,0.61728394,0.60493827,0.61728394,0.67901236,0.69135803,0.60493827,0.5802469,0.5802469,0.8765432,0.75308645,0.654321,0.86419755,0.8271605,0.8271605,0.8271605,0.86419755,0.9506173,0.9506173,0.8888889,0.91358024,0.91358024,0.97530866,0.8765432,0.97530866,0.9876543,1.0,1.0,1.0
+TabDPT (tuned),0.049382716,0.037037037,0.037037037,0.037037037,0.049382716,0.061728396,0.20987654,0.09876543,0.20987654,0.09876543,0.22222222,0.18518518,0.11111111,0.17283951,0.22222222,0.13580246,0.20987654,0.18518518,0.12345679,0.22222222,0.19753087,0.18518518,0.19753087,0.24691358,0.28395063,0.16049382,0.2345679,0.24691358,0.12345679,0.32098764,0.18518518,0.4814815,0.33333334,0.38271606,0.32098764,0.41975307,0.43209878,0.37037036,0.41975307,0.32098764,0.45679012,0.56790125,,0.4691358,0.5185185,0.49382716,0.5185185,0.54320985,0.49382716,0.6419753,0.654321,0.6419753,0.7592593,0.69135803,0.6419753,0.7160494,0.75308645,0.654321,0.7777778,0.6296296,0.6419753,0.7160494,0.962963,0.962963,0.9259259,0.7777778,0.9382716,0.90123457,0.8888889,0.90123457,0.8888889,0.97530866,1.0,1.0
+EBM (default),0.0,0.0,0.049382716,0.049382716,0.037037037,0.074074075,0.074074075,0.061728396,0.061728396,0.061728396,0.037037037,0.024691358,0.074074075,0.024691358,0.037037037,0.14814815,0.09876543,0.18518518,0.14814815,0.086419754,0.12345679,0.14814815,0.18518518,0.24691358,0.16049382,0.27160493,0.34567901,0.30864197,0.37037036,0.2962963,0.32098764,0.30864197,0.086419754,0.39506173,0.3580247,0.44444445,0.41975307,0.5185185,0.4691358,0.41975307,0.28395063,0.5555556,0.5308642,,0.50617284,0.61728394,0.60493827,0.61728394,0.6419753,0.6419753,0.67901236,0.7037037,0.5555556,0.654321,0.8148148,0.72839504,0.7037037,0.79012346,0.86419755,0.8271605,0.8271605,0.8271605,0.8518519,0.83950615,0.80246913,0.90123457,0.80246913,0.8518519,0.7777778,1.0,0.97530866,1.0,0.8888889,1.0
+LightGBM (default),0.0,0.0,0.049382716,0.012345679,0.061728396,0.061728396,0.074074075,0.086419754,0.049382716,0.049382716,0.086419754,0.13580246,0.0,0.037037037,0.024691358,0.09876543,0.20987654,0.16049382,0.086419754,0.074074075,0.13580246,0.27160493,0.12345679,0.13580246,0.037037037,0.14814815,0.22222222,0.25925925,0.24691358,0.27160493,0.2345679,0.4814815,0.3580247,0.5185185,0.30864197,0.41975307,0.49382716,0.3580247,0.41975307,0.34567901,0.45679012,0.4691358,0.4814815,0.49382716,,0.56790125,0.5555556,0.61728394,0.54320985,0.61728394,0.6419753,0.5555556,0.6296296,0.6666667,0.7160494,0.7160494,0.6666667,0.654321,0.7407407,0.8271605,0.83950615,0.6666667,0.86419755,0.9506173,0.9382716,0.79012346,0.962963,0.9382716,0.8888889,1.0,0.91358024,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.037037037,0.012345679,0.061728396,0.086419754,0.061728396,0.0,0.086419754,0.09876543,0.074074075,0.049382716,0.074074075,0.037037037,0.024691358,0.17283951,0.11111111,0.18518518,0.16049382,0.074074075,0.16049382,0.086419754,0.19753087,0.18518518,0.074074075,0.25925925,0.25925925,0.25925925,0.2962963,0.17283951,0.34567901,0.39506173,0.30864197,0.20987654,0.32098764,0.32098764,0.4691358,0.32098764,0.50617284,0.39506173,0.4074074,0.2962963,0.50617284,0.38271606,0.43209878,,0.5185185,0.54320985,0.6419753,0.5308642,0.5555556,0.69135803,0.5308642,0.6296296,0.60493827,0.60493827,0.7407407,0.72839504,0.86419755,0.8765432,0.8888889,0.7777778,0.72839504,0.7160494,0.8271605,0.8765432,0.83950615,0.91358024,0.8765432,0.9876543,0.962963,0.8888889,0.7777778,1.0
+ModernNCA (default),0.0,0.0,0.012345679,0.024691358,0.024691358,0.049382716,0.012345679,0.024691358,0.049382716,0.049382716,0.074074075,0.09876543,0.037037037,0.012345679,0.024691358,0.049382716,0.16049382,0.16049382,0.19753087,0.037037037,0.09876543,0.20987654,0.2345679,0.22222222,0.074074075,0.14814815,0.17283951,0.11111111,0.2345679,0.28395063,0.33333334,0.37037036,0.30864197,0.30864197,0.19753087,0.38271606,0.37037036,0.2962963,0.30864197,0.33333334,0.3580247,0.38271606,0.4814815,0.39506173,0.44444445,0.4814815,,0.5802469,0.5308642,0.5185185,0.654321,0.49382716,0.5308642,0.5925926,0.72839504,0.60493827,0.6419753,0.6666667,0.8148148,0.7160494,0.72839504,0.7160494,0.8271605,0.8518519,0.9259259,0.8518519,0.9506173,0.90123457,0.8888889,1.0,0.91358024,1.0,0.91358024,1.0
+RandomForest (tuned + ensembled),0.0,0.024691358,0.012345679,0.061728396,0.049382716,0.037037037,0.024691358,0.024691358,0.024691358,0.024691358,0.024691358,0.012345679,0.074074075,0.0,0.037037037,0.12345679,0.037037037,0.061728396,0.09876543,0.037037037,0.13580246,0.061728396,0.16049382,0.17283951,0.14814815,0.20987654,0.18518518,0.19753087,0.25925925,0.22222222,0.28395063,0.28395063,0.25925925,0.16049382,0.25925925,0.4074074,0.30864197,0.33333334,0.3580247,0.38271606,0.33333334,0.39506173,0.45679012,0.38271606,0.38271606,0.45679012,0.41975307,,0.56790125,0.5925926,0.72839504,0.5925926,0.50617284,0.5925926,0.6666667,0.654321,0.6666667,0.7777778,0.8765432,0.7777778,0.79012346,0.83950615,0.83950615,0.8148148,0.962963,0.9506173,0.962963,0.9259259,0.8765432,0.9259259,0.962963,1.0,0.8888889,1.0
+SAP-RPT-OSS (default),0.0,0.024691358,0.037037037,0.086419754,0.049382716,0.086419754,0.049382716,0.024691358,0.024691358,0.037037037,0.037037037,0.024691358,0.09876543,0.049382716,0.049382716,0.16049382,0.074074075,0.12345679,0.16049382,0.061728396,0.14814815,0.049382716,0.18518518,0.11111111,0.19753087,0.24691358,0.17283951,0.2345679,0.28395063,0.19753087,0.33333334,0.2345679,0.13580246,0.18518518,0.30864197,0.4074074,0.11111111,0.30864197,0.32098764,0.28395063,0.19753087,0.38271606,0.50617284,0.3580247,0.45679012,0.3580247,0.4691358,0.43209878,,0.56790125,0.4691358,0.56790125,0.5802469,0.654321,0.60493827,0.75308645,0.67901236,0.79012346,0.8765432,0.6296296,0.654321,0.8271605,0.8271605,0.8518519,0.83950615,0.97530866,0.83950615,0.8518519,0.8148148,0.9506173,0.962963,1.0,0.8888889,1.0
+ExtraTrees (tuned + ensembled),0.0,0.012345679,0.012345679,0.037037037,0.024691358,0.012345679,0.037037037,0.049382716,0.012345679,0.0,0.049382716,0.12345679,0.074074075,0.012345679,0.024691358,0.11111111,0.14814815,0.11111111,0.037037037,0.049382716,0.14814815,0.14814815,0.09876543,0.086419754,0.18518518,0.22222222,0.18518518,0.17283951,0.24691358,0.12345679,0.25925925,0.18518518,0.22222222,0.14814815,0.28395063,0.30864197,0.27160493,0.2345679,0.34567901,0.32098764,0.34567901,0.32098764,0.3580247,0.3580247,0.38271606,0.4691358,0.4814815,0.4074074,0.43209878,,0.50617284,0.4814815,0.4814815,0.5308642,0.60493827,0.72839504,0.60493827,0.69135803,0.7160494,0.69135803,0.69135803,0.7654321,0.72839504,0.7777778,0.97530866,0.8888889,0.9876543,0.9382716,0.90123457,0.83950615,0.962963,0.8888889,1.0,1.0
+RandomForest (tuned),0.012345679,0.024691358,0.024691358,0.037037037,0.012345679,0.037037037,0.037037037,0.012345679,0.037037037,0.037037037,0.037037037,0.012345679,0.074074075,0.012345679,0.049382716,0.13580246,0.037037037,0.086419754,0.049382716,0.061728396,0.12345679,0.09876543,0.074074075,0.11111111,0.12345679,0.19753087,0.16049382,0.20987654,0.27160493,0.16049382,0.22222222,0.27160493,0.22222222,0.14814815,0.25925925,0.38271606,0.27160493,0.25925925,0.33333334,0.34567901,0.30864197,0.30864197,0.34567901,0.32098764,0.3580247,0.44444445,0.34567901,0.27160493,0.5308642,0.49382716,,0.5185185,0.45679012,0.5802469,0.5308642,0.5925926,0.6296296,0.67901236,0.83950615,0.7654321,0.7777778,0.75308645,0.8148148,0.8148148,0.91358024,0.90123457,0.93209875,0.91358024,0.83950615,0.9259259,0.97530866,1.0,0.8888889,1.0
+FastaiMLP (tuned),0.0,0.024691358,0.024691358,0.074074075,0.061728396,0.074074075,0.024691358,0.037037037,0.037037037,0.024691358,0.049382716,0.12345679,0.09876543,0.13580246,0.09876543,0.074074075,0.16049382,0.037037037,0.13580246,0.14814815,0.13580246,0.14814815,0.13580246,0.22222222,0.2345679,0.32098764,0.16049382,0.086419754,0.22222222,0.18518518,0.28395063,0.22222222,0.27160493,0.32098764,0.24691358,0.3580247,0.19753087,0.19753087,0.074074075,0.2962963,0.2962963,0.39506173,0.3580247,0.2962963,0.44444445,0.30864197,0.50617284,0.4074074,0.43209878,0.5185185,0.4814815,,0.54320985,0.4691358,0.54320985,0.56790125,0.67901236,0.6296296,0.67901236,0.61728394,0.61728394,0.7037037,0.80246913,0.8148148,0.80246913,0.8888889,0.8518519,0.75308645,0.8765432,0.8765432,0.8518519,0.8765432,0.9506173,1.0
+TabDPT (default),0.024691358,0.024691358,0.049382716,0.037037037,0.061728396,0.049382716,0.11111111,0.074074075,0.12345679,0.09876543,0.13580246,0.18518518,0.12345679,0.12345679,0.049382716,0.086419754,0.24691358,0.17283951,0.11111111,0.16049382,0.17283951,0.19753087,0.17283951,0.16049382,0.18518518,0.086419754,0.16049382,0.14814815,0.061728396,0.25925925,0.09876543,0.34567901,0.3580247,0.38271606,0.30864197,0.16049382,0.37037036,0.2962963,0.30864197,0.3580247,0.44444445,0.41975307,0.24074075,0.44444445,0.37037036,0.4691358,0.4691358,0.49382716,0.41975307,0.5185185,0.54320985,0.45679012,,0.50617284,0.5802469,0.5555556,0.5925926,0.6666667,0.56790125,0.6296296,0.6296296,0.67901236,0.7654321,0.8148148,0.6419753,0.7160494,0.6666667,0.69135803,0.72839504,0.67901236,0.83950615,0.7037037,0.97530866,1.0
+LimiX (default),0.012345679,0.049382716,0.024691358,0.11111111,0.13580246,0.09876543,0.16049382,0.16049382,0.19753087,0.061728396,0.19753087,0.17283951,0.14814815,0.24691358,0.2345679,0.049382716,0.18518518,0.11111111,0.25925925,0.27160493,0.14814815,0.18518518,0.27160493,0.25925925,0.28395063,0.2345679,0.14814815,0.19753087,0.25925925,0.25925925,0.14814815,0.2345679,0.33333334,0.32098764,0.3580247,0.24691358,0.30864197,0.32098764,0.38271606,0.27160493,0.34567901,0.41975307,0.30864197,0.34567901,0.33333334,0.37037036,0.4074074,0.4074074,0.34567901,0.4691358,0.41975307,0.5308642,0.49382716,,0.43209878,0.50617284,0.56790125,0.60493827,0.5185185,0.4691358,0.4814815,0.61728394,0.5925926,0.6419753,0.67901236,0.61728394,0.72839504,0.5925926,0.75308645,0.5802469,0.61728394,0.6419753,0.7654321,0.8765432
+TorchMLP (default),0.0,0.0,0.012345679,0.037037037,0.0,0.024691358,0.024691358,0.0,0.012345679,0.012345679,0.024691358,0.09876543,0.024691358,0.0,0.024691358,0.12345679,0.09876543,0.074074075,0.049382716,0.0,0.024691358,0.14814815,0.09876543,0.049382716,0.074074075,0.22222222,0.12345679,0.16049382,0.24691358,0.012345679,0.22222222,0.25925925,0.14814815,0.086419754,0.19753087,0.34567901,0.24691358,0.12345679,0.24691358,0.25925925,0.17283951,0.12345679,0.3580247,0.18518518,0.28395063,0.39506173,0.27160493,0.33333334,0.39506173,0.39506173,0.4691358,0.45679012,0.41975307,0.56790125,,0.49382716,0.50617284,0.7037037,0.7777778,0.7037037,0.7037037,0.7407407,0.7777778,0.80246913,0.8518519,0.86419755,0.8765432,0.9259259,0.8765432,0.8888889,0.9506173,0.9876543,0.9506173,1.0
+ExtraTrees (tuned),0.0,0.012345679,0.0,0.037037037,0.012345679,0.012345679,0.012345679,0.012345679,0.024691358,0.024691358,0.037037037,0.12345679,0.049382716,0.012345679,0.024691358,0.12345679,0.14814815,0.049382716,0.037037037,0.024691358,0.086419754,0.16049382,0.074074075,0.061728396,0.086419754,0.18518518,0.14814815,0.16049382,0.20987654,0.09876543,0.19753087,0.19753087,0.17283951,0.11111111,0.24691358,0.30864197,0.17283951,0.20987654,0.30864197,0.2962963,0.27160493,0.24691358,0.28395063,0.27160493,0.28395063,0.39506173,0.39506173,0.34567901,0.24691358,0.27160493,0.4074074,0.43209878,0.44444445,0.49382716,0.50617284,,0.49382716,0.67901236,0.654321,0.67901236,0.6666667,0.7160494,0.67901236,0.75308645,0.9382716,0.8765432,0.9506173,0.9259259,0.8888889,0.8271605,0.962963,0.90123457,0.9876543,1.0
+FastaiMLP (default),0.0,0.024691358,0.012345679,0.074074075,0.086419754,0.086419754,0.049382716,0.0,0.024691358,0.037037037,0.061728396,0.13580246,0.11111111,0.11111111,0.14814815,0.037037037,0.14814815,0.049382716,0.11111111,0.16049382,0.09876543,0.14814815,0.12345679,0.17283951,0.20987654,0.2345679,0.11111111,0.09876543,0.18518518,0.18518518,0.13580246,0.14814815,0.22222222,0.25925925,0.20987654,0.3580247,0.14814815,0.17283951,0.037037037,0.25925925,0.28395063,0.34567901,0.24691358,0.2962963,0.33333334,0.25925925,0.3580247,0.33333334,0.32098764,0.39506173,0.37037036,0.32098764,0.4074074,0.43209878,0.49382716,0.50617284,,0.54320985,0.5555556,0.5308642,0.54320985,0.5925926,0.7654321,0.7777778,0.67901236,0.6419753,0.67901236,0.7037037,0.79012346,0.7777778,0.7777778,0.79012346,0.8765432,1.0
+Linear (tuned + ensembled),0.0,0.024691358,0.0,0.086419754,0.037037037,0.086419754,0.012345679,0.012345679,0.012345679,0.0,0.012345679,0.0,0.074074075,0.0,0.024691358,0.11111111,0.0,0.024691358,0.17283951,0.0,0.12345679,0.0,0.19753087,0.17283951,0.14814815,0.27160493,0.11111111,0.14814815,0.27160493,0.09876543,0.2962963,0.024691358,0.11111111,0.13580246,0.22222222,0.30864197,0.049382716,0.18518518,0.12345679,0.27160493,0.16049382,0.13580246,0.34567901,0.20987654,0.34567901,0.27160493,0.33333334,0.22222222,0.20987654,0.30864197,0.32098764,0.37037036,0.33333334,0.39506173,0.2962963,0.32098764,0.45679012,,0.3580247,0.4814815,0.4814815,0.7962963,0.4814815,0.50617284,0.60493827,0.9259259,0.61728394,0.654321,0.7654321,0.5555556,0.654321,0.5802469,0.7407407,0.8888889
+KNN (tuned + ensembled),0.012345679,0.049382716,0.012345679,0.074074075,0.061728396,0.049382716,0.012345679,0.012345679,0.012345679,0.012345679,0.012345679,0.012345679,0.061728396,0.012345679,0.024691358,0.012345679,0.024691358,0.037037037,0.09876543,0.012345679,0.12345679,0.024691358,0.12345679,0.11111111,0.09876543,0.16049382,0.012345679,0.049382716,0.17283951,0.09876543,0.16049382,0.037037037,0.09876543,0.12345679,0.17283951,0.2345679,0.024691358,0.14814815,0.13580246,0.09876543,0.12345679,0.17283951,0.22222222,0.13580246,0.25925925,0.13580246,0.18518518,0.12345679,0.12345679,0.28395063,0.16049382,0.32098764,0.43209878,0.4814815,0.22222222,0.34567901,0.44444445,0.6419753,,0.38271606,0.38271606,0.7037037,0.5925926,0.6419753,0.6296296,0.80246913,0.7654321,0.6419753,0.7654321,0.6666667,0.7160494,1.0,0.8888889,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.024691358,0.0,0.0,0.0,0.0,0.012345679,0.0,0.0,0.012345679,0.012345679,0.086419754,0.012345679,0.024691358,0.0,0.012345679,0.0,0.012345679,0.012345679,0.0,0.0,0.074074075,0.061728396,0.049382716,0.16049382,0.086419754,0.12345679,0.16049382,0.024691358,0.061728396,0.11111111,0.22222222,0.18518518,0.037037037,0.24691358,0.024691358,0.061728396,0.17283951,0.37037036,0.17283951,0.17283951,0.12345679,0.28395063,0.22222222,0.37037036,0.30864197,0.2345679,0.38271606,0.37037036,0.5308642,0.2962963,0.32098764,0.4691358,0.5185185,0.61728394,,0.66049385,0.5308642,0.7037037,0.7037037,0.6666667,0.7160494,0.6666667,0.654321,0.6666667,1.0,0.86419755,0.86419755,0.7777778,0.8888889
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.024691358,0.0,0.0,0.0,0.0,0.012345679,0.0,0.0,0.012345679,0.012345679,0.086419754,0.012345679,0.024691358,0.0,0.012345679,0.0,0.0,0.0,0.0,0.0,0.061728396,0.061728396,0.049382716,0.16049382,0.086419754,0.12345679,0.16049382,0.037037037,0.049382716,0.11111111,0.22222222,0.17283951,0.037037037,0.24691358,0.012345679,0.061728396,0.17283951,0.3580247,0.17283951,0.16049382,0.11111111,0.27160493,0.20987654,0.34567901,0.30864197,0.22222222,0.38271606,0.37037036,0.5185185,0.2962963,0.33333334,0.45679012,0.5185185,0.61728394,0.33950618,,0.5308642,0.7037037,0.7037037,0.6666667,0.7160494,0.6666667,0.654321,0.6666667,1.0,0.86419755,0.86419755,0.7777778,0.8888889
+Linear (tuned),0.0,0.024691358,0.0,0.074074075,0.037037037,0.061728396,0.0,0.0,0.0,0.0,0.0,0.0,0.061728396,0.0,0.012345679,0.09876543,0.0,0.012345679,0.16049382,0.0,0.11111111,0.0,0.14814815,0.17283951,0.09876543,0.27160493,0.074074075,0.11111111,0.27160493,0.09876543,0.27160493,0.0,0.086419754,0.12345679,0.2345679,0.27160493,0.012345679,0.13580246,0.11111111,0.25925925,0.12345679,0.13580246,0.28395063,0.17283951,0.33333334,0.22222222,0.28395063,0.16049382,0.17283951,0.2345679,0.24691358,0.2962963,0.32098764,0.38271606,0.25925925,0.28395063,0.4074074,0.2037037,0.2962963,0.4691358,0.4691358,,0.4814815,0.49382716,0.56790125,0.86419755,0.56790125,0.6296296,0.72839504,0.5555556,0.61728394,0.5802469,0.7407407,0.8888889
+PerpetualBooster (tuned + ensembled),0.012345679,0.0,0.024691358,0.0,0.012345679,0.0,0.012345679,0.024691358,0.024691358,0.024691358,0.024691358,0.13580246,0.024691358,0.086419754,0.024691358,0.012345679,0.13580246,0.037037037,0.037037037,0.09876543,0.024691358,0.13580246,0.037037037,0.024691358,0.061728396,0.024691358,0.074074075,0.061728396,0.012345679,0.024691358,0.037037037,0.20987654,0.13580246,0.13580246,0.024691358,0.049382716,0.16049382,0.049382716,0.13580246,0.14814815,0.14814815,0.049382716,0.037037037,0.14814815,0.13580246,0.27160493,0.17283951,0.16049382,0.17283951,0.27160493,0.18518518,0.19753087,0.2345679,0.4074074,0.22222222,0.32098764,0.2345679,0.5185185,0.4074074,0.2962963,0.2962963,0.5185185,,0.83950615,0.5802469,0.5185185,0.5925926,0.6666667,0.6296296,0.38271606,0.56790125,0.61728394,0.91358024,0.9506173
+PerpetualBooster (tuned),0.0,0.0,0.037037037,0.0,0.012345679,0.0,0.012345679,0.037037037,0.012345679,0.012345679,0.012345679,0.12345679,0.0,0.037037037,0.024691358,0.012345679,0.13580246,0.049382716,0.024691358,0.037037037,0.012345679,0.14814815,0.061728396,0.012345679,0.037037037,0.0,0.086419754,0.049382716,0.024691358,0.037037037,0.037037037,0.16049382,0.14814815,0.074074075,0.024691358,0.024691358,0.18518518,0.061728396,0.14814815,0.14814815,0.16049382,0.049382716,0.037037037,0.16049382,0.049382716,0.28395063,0.14814815,0.18518518,0.14814815,0.22222222,0.18518518,0.18518518,0.18518518,0.3580247,0.19753087,0.24691358,0.22222222,0.49382716,0.3580247,0.2962963,0.2962963,0.50617284,0.16049382,,0.5555556,0.49382716,0.5925926,0.6666667,0.6296296,0.34567901,0.50617284,0.5308642,0.8888889,0.9506173
+ExtraTrees (default),0.0,0.012345679,0.012345679,0.0,0.0,0.0,0.024691358,0.012345679,0.012345679,0.0,0.024691358,0.012345679,0.0,0.012345679,0.012345679,0.024691358,0.012345679,0.012345679,0.012345679,0.012345679,0.024691358,0.012345679,0.024691358,0.012345679,0.012345679,0.0,0.037037037,0.037037037,0.012345679,0.012345679,0.086419754,0.024691358,0.12345679,0.086419754,0.061728396,0.086419754,0.061728396,0.049382716,0.074074075,0.16049382,0.18518518,0.11111111,0.074074075,0.19753087,0.061728396,0.17283951,0.074074075,0.037037037,0.16049382,0.024691358,0.086419754,0.19753087,0.3580247,0.32098764,0.14814815,0.061728396,0.32098764,0.39506173,0.37037036,0.33333334,0.33333334,0.43209878,0.41975307,0.44444445,,0.4691358,0.44444445,0.43209878,0.67901236,0.50617284,0.54320985,0.61728394,0.7654321,1.0
+Linear (default),0.0,0.0,0.0,0.049382716,0.037037037,0.037037037,0.0,0.0,0.0,0.0,0.0,0.0,0.024691358,0.0,0.0,0.024691358,0.0,0.0,0.049382716,0.0,0.074074075,0.0,0.074074075,0.086419754,0.037037037,0.074074075,0.0,0.012345679,0.12345679,0.049382716,0.13580246,0.0,0.061728396,0.11111111,0.11111111,0.20987654,0.0,0.09876543,0.024691358,0.061728396,0.086419754,0.086419754,0.22222222,0.09876543,0.20987654,0.12345679,0.14814815,0.049382716,0.024691358,0.11111111,0.09876543,0.11111111,0.28395063,0.38271606,0.13580246,0.12345679,0.3580247,0.074074075,0.19753087,0.28395063,0.28395063,0.13580246,0.4814815,0.50617284,0.5308642,,0.54320985,0.5555556,0.7160494,0.5308642,0.4814815,0.5802469,0.7407407,0.8765432
+RandomForest (default),0.0,0.012345679,0.0,0.0,0.0,0.0,0.012345679,0.012345679,0.012345679,0.0,0.012345679,0.0,0.0,0.0,0.0,0.0,0.012345679,0.012345679,0.0,0.0,0.012345679,0.0,0.012345679,0.024691358,0.012345679,0.0,0.037037037,0.024691358,0.012345679,0.012345679,0.049382716,0.024691358,0.13580246,0.086419754,0.086419754,0.09876543,0.11111111,0.049382716,0.074074075,0.14814815,0.17283951,0.086419754,0.061728396,0.19753087,0.037037037,0.16049382,0.049382716,0.037037037,0.16049382,0.012345679,0.06790123,0.14814815,0.33333334,0.27160493,0.12345679,0.049382716,0.32098764,0.38271606,0.2345679,0.33333334,0.33333334,0.43209878,0.4074074,0.4074074,0.5555556,0.45679012,,0.43209878,0.79012346,0.45679012,0.54320985,0.61728394,0.7654321,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.012345679,0.012345679,0.012345679,0.0,0.024691358,0.0,0.0,0.0,0.012345679,0.049382716,0.012345679,0.024691358,0.012345679,0.0,0.0,0.0,0.012345679,0.0,0.012345679,0.0,0.024691358,0.012345679,0.012345679,0.0,0.11111111,0.037037037,0.13580246,0.012345679,0.012345679,0.0,0.13580246,0.049382716,0.18518518,0.12345679,0.13580246,0.024691358,0.09876543,0.14814815,0.061728396,0.086419754,0.09876543,0.074074075,0.14814815,0.061728396,0.086419754,0.24691358,0.30864197,0.4074074,0.074074075,0.074074075,0.2962963,0.34567901,0.3580247,0.34567901,0.34567901,0.37037036,0.33333334,0.33333334,0.56790125,0.44444445,0.56790125,,0.5925926,0.49382716,0.5925926,0.43209878,0.60493827,0.9382716
+xRFM (default),0.0,0.049382716,0.0,0.086419754,0.086419754,0.086419754,0.0,0.0,0.0,0.0,0.0,0.0,0.09876543,0.012345679,0.049382716,0.0,0.0,0.024691358,0.086419754,0.049382716,0.11111111,0.0,0.11111111,0.11111111,0.09876543,0.074074075,0.012345679,0.049382716,0.086419754,0.11111111,0.09876543,0.0,0.19753087,0.11111111,0.11111111,0.11111111,0.012345679,0.13580246,0.0,0.20987654,0.20987654,0.12345679,0.11111111,0.22222222,0.11111111,0.12345679,0.11111111,0.12345679,0.18518518,0.09876543,0.16049382,0.12345679,0.27160493,0.24691358,0.12345679,0.11111111,0.20987654,0.2345679,0.2345679,0.33333334,0.33333334,0.27160493,0.37037036,0.37037036,0.32098764,0.28395063,0.20987654,0.4074074,,0.4074074,0.4074074,0.43209878,0.5308642,0.8518519
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.012345679,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.086419754,0.012345679,0.0,0.11111111,0.0,0.0,0.024691358,0.09876543,0.0,0.0,0.012345679,0.0,0.074074075,0.049382716,0.16049382,0.074074075,0.12345679,0.32098764,0.41975307,0.11111111,0.17283951,0.22222222,0.44444445,0.33333334,0.0,0.0,0.44444445,0.61728394,0.654321,0.49382716,0.4691358,0.54320985,0.50617284,0.5925926,,0.6419753,0.6666667,0.7777778,0.86419755
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.012345679,0.0,0.012345679,0.0,0.012345679,0.0,0.0,0.0,0.0,0.074074075,0.0,0.024691358,0.0,0.0,0.0,0.0,0.012345679,0.0,0.012345679,0.024691358,0.012345679,0.024691358,0.037037037,0.0,0.12345679,0.024691358,0.024691358,0.0,0.049382716,0.0,0.049382716,0.012345679,0.074074075,0.037037037,0.012345679,0.012345679,0.11111111,0.024691358,0.086419754,0.037037037,0.086419754,0.037037037,0.037037037,0.037037037,0.024691358,0.14814815,0.16049382,0.38271606,0.049382716,0.037037037,0.22222222,0.34567901,0.28395063,0.13580246,0.13580246,0.38271606,0.43209878,0.49382716,0.45679012,0.5185185,0.45679012,0.4074074,0.5925926,0.3580247,,0.50617284,0.6666667,0.7777778
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.012345679,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.037037037,0.0,0.0,0.0,0.037037037,0.0,0.0,0.11111111,0.0,0.0,0.0,0.024691358,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.12345679,0.2962963,0.3580247,0.012345679,0.09876543,0.20987654,0.41975307,0.0,0.13580246,0.13580246,0.41975307,0.38271606,0.4691358,0.38271606,0.41975307,0.38271606,0.56790125,0.56790125,0.33333334,0.49382716,,0.79012346,0.9876543
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.086419754,0.0,0.0,0.0,0.0,0.074074075,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.22222222,0.086419754,0.11111111,0.11111111,0.0,0.11111111,0.049382716,0.024691358,0.2345679,0.049382716,0.012345679,0.12345679,0.25925925,0.11111111,0.22222222,0.22222222,0.25925925,0.086419754,0.11111111,0.2345679,0.25925925,0.2345679,0.39506173,0.4691358,0.22222222,0.33333334,0.20987654,,0.86419755
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.12345679,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.049382716,0.049382716,0.0,0.12345679,0.0,0.061728396,0.14814815,0.13580246,0.22222222,0.012345679,0.13580246,
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_small/leaderboard_overview_explorer.html
index 0ce74b71daef45b54a7e53af9939c134ad624e65..d92dad3963f61790b6752f73e37f50e36d15a81b 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Binary | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_small/leaderboard_table.html b/data/imputation_no/splits_all/tasks_binary/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..356d271f5953862f24d0eda27f6fdfd53af32c38
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_binary/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1740.0,"score":0.896,"rank":6.67,"hrank":1.72,"imp":4.518,"train_time":36.24,"predict_time":5.432,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+154\/-109","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1641.0,"score":0.693,"rank":10.09,"hrank":4.39,"imp":12.633,"train_time":3.85,"predict_time":0.374,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-62","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1591.0,"score":0.589,"rank":12.35,"hrank":4.79,"imp":10.323,"train_time":368.19,"predict_time":6.655,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-79","imputed":false},{"position":3,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1576.0,"score":0.644,"rank":13.06,"hrank":5.19,"imp":12.887,"train_time":2.09,"predict_time":0.152,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-85","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1565.0,"score":0.58,"rank":13.64,"hrank":6.81,"imp":14.017,"train_time":5.48,"predict_time":0.433,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-66","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1558.0,"score":0.56,"rank":13.98,"hrank":8.72,"imp":14.574,"train_time":5.96,"predict_time":0.554,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-44","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1552.0,"score":0.536,"rank":14.35,"hrank":6.9,"imp":14.076,"train_time":2052.28,"predict_time":8.252,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-65","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1527.0,"score":0.512,"rank":15.72,"hrank":7.43,"imp":14.666,"train_time":2052.28,"predict_time":1.265,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-56","imputed":false},{"position":8,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1438.0,"score":0.384,"rank":21.47,"hrank":9.01,"imp":16.689,"train_time":2.3,"predict_time":0.114,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-65","imputed":false},{"position":9,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1438.0,"score":0.392,"rank":21.47,"hrank":9.06,"imp":16.565,"train_time":19.82,"predict_time":4.463,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-80","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1420.0,"score":0.317,"rank":22.83,"hrank":15.11,"imp":17.847,"train_time":2950.72,"predict_time":15.589,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-48","imputed":false},{"position":11,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1410.0,"score":0.293,"rank":23.52,"hrank":11.62,"imp":17.971,"train_time":2289.91,"predict_time":2.709,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-53","imputed":false},{"position":12,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1402.0,"score":0.293,"rank":24.12,"hrank":15.47,"imp":17.722,"train_time":1921.63,"predict_time":1.717,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-66","imputed":false},{"position":13,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1390.0,"score":0.3,"rank":25.02,"hrank":11.64,"imp":17.096,"train_time":5351.93,"predict_time":268.579,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-67","imputed":false},{"position":14,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1383.0,"score":0.308,"rank":25.56,"hrank":13.55,"imp":19.155,"train_time":317.92,"predict_time":1.982,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-61","imputed":false},{"position":15,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1371.0,"score":0.317,"rank":26.55,"hrank":14.91,"imp":18.584,"train_time":1.23,"predict_time":0.052,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-68","imputed":false},{"position":16,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1361.0,"score":0.263,"rank":27.35,"hrank":11.83,"imp":17.673,"train_time":5351.93,"predict_time":37.606,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-69","imputed":false},{"position":17,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1360.0,"score":0.234,"rank":27.43,"hrank":13.18,"imp":18.845,"train_time":2289.91,"predict_time":0.266,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-52","imputed":false},{"position":18,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1358.0,"score":0.196,"rank":27.6,"hrank":20.4,"imp":19.079,"train_time":512.67,"predict_time":1.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-38","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1327.0,"score":0.148,"rank":30.12,"hrank":22.67,"imp":19.499,"train_time":512.67,"predict_time":0.243,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-33","imputed":false},{"position":20,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1326.0,"score":0.185,"rank":30.16,"hrank":16.46,"imp":19.527,"train_time":2950.72,"predict_time":1.225,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-56","imputed":false},{"position":21,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1325.0,"score":0.173,"rank":30.23,"hrank":20.12,"imp":19.369,"train_time":15277.76,"predict_time":543.233,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-55","imputed":false},{"position":22,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1315.0,"score":0.155,"rank":31.08,"hrank":21.51,"imp":19.902,"train_time":1260.85,"predict_time":0.312,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-41","imputed":false},{"position":23,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1312.0,"score":0.168,"rank":31.34,"hrank":21.59,"imp":20.0,"train_time":3.83,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-45","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1311.0,"score":0.18,"rank":31.38,"hrank":15.76,"imp":19.94,"train_time":4806.04,"predict_time":0.512,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-60","imputed":false},{"position":25,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1309.0,"score":0.15,"rank":31.59,"hrank":22.16,"imp":20.008,"train_time":1260.85,"predict_time":0.031,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-40","imputed":false},{"position":26,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1300.0,"score":0.223,"rank":32.33,"hrank":13.54,"imp":21.668,"train_time":9.05,"predict_time":1.763,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-74","imputed":false},{"position":27,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1289.0,"score":0.218,"rank":33.27,"hrank":16.89,"imp":19.983,"train_time":51.83,"predict_time":43.611,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-77","imputed":false},{"position":28,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1283.0,"score":0.12,"rank":33.77,"hrank":24.73,"imp":20.424,"train_time":693.49,"predict_time":1.397,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-46","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1282.0,"score":0.139,"rank":33.87,"hrank":22.75,"imp":21.159,"train_time":550.77,"predict_time":0.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-58","imputed":false},{"position":30,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1281.0,"score":0.148,"rank":33.94,"hrank":20.63,"imp":20.482,"train_time":711.98,"predict_time":2.983,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-65","imputed":false},{"position":31,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1272.0,"score":0.156,"rank":34.72,"hrank":19.63,"imp":20.651,"train_time":6.8,"predict_time":0.248,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-60","imputed":false},{"position":32,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1269.0,"score":0.207,"rank":34.95,"hrank":16.26,"imp":21.239,"train_time":4806.04,"predict_time":8.45,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-102","imputed":false},{"position":33,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1262.0,"score":0.141,"rank":35.52,"hrank":17.48,"imp":22.073,"train_time":14.38,"predict_time":0.334,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-64","imputed":false},{"position":34,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1262.0,"score":0.094,"rank":35.57,"hrank":24.17,"imp":21.389,"train_time":1353.2,"predict_time":0.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-49","imputed":false},{"position":35,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1260.0,"score":0.106,"rank":35.71,"hrank":24.4,"imp":21.544,"train_time":550.77,"predict_time":0.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-53","imputed":false},{"position":36,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1259.0,"score":0.081,"rank":35.84,"hrank":24.56,"imp":20.77,"train_time":693.49,"predict_time":0.194,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-36","imputed":false},{"position":37,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1257.0,"score":0.13,"rank":35.97,"hrank":21.37,"imp":21.31,"train_time":711.98,"predict_time":0.162,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-64","imputed":false},{"position":38,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1253.0,"score":0.119,"rank":36.28,"hrank":25.81,"imp":20.733,"train_time":2665.48,"predict_time":2.619,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-52","imputed":false},{"position":39,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1240.0,"score":0.07,"rank":37.43,"hrank":25.33,"imp":21.891,"train_time":1353.2,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-46","imputed":false},{"position":40,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1235.0,"score":0.098,"rank":37.89,"hrank":21.62,"imp":22.145,"train_time":2.58,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-60","imputed":false},{"position":41,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1225.0,"score":0.065,"rank":38.72,"hrank":26.17,"imp":21.973,"train_time":15277.76,"predict_time":119.284,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-48","imputed":false},{"position":42,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1221.0,"score":0.114,"rank":39.09,"hrank":24.49,"imp":21.969,"train_time":644.68,"predict_time":5.974,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-77","imputed":false},{"position":43,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1220.0,"score":0.077,"rank":39.13,"hrank":23.29,"imp":22.315,"train_time":6.67,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-56","imputed":false},{"position":44,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1220.0,"score":0.085,"rank":39.18,"hrank":24.94,"imp":21.727,"train_time":10.9,"predict_time":2.248,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-50","imputed":false},{"position":45,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1187.0,"score":0.137,"rank":41.94,"hrank":25.91,"imp":23.23,"train_time":14.16,"predict_time":2.52,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-73","imputed":false},{"position":46,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1185.0,"score":0.063,"rank":42.12,"hrank":30.44,"imp":22.67,"train_time":2665.48,"predict_time":0.143,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-46","imputed":false},{"position":47,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1182.0,"score":0.234,"rank":42.38,"hrank":9.3,"imp":23.966,"train_time":34059.64,"predict_time":3.908,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+116\/-95","imputed":false},{"position":48,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1175.0,"score":0.225,"rank":42.95,"hrank":9.04,"imp":24.115,"train_time":34059.64,"predict_time":18.811,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+122\/-94","imputed":false},{"position":49,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1172.0,"score":0.051,"rank":43.23,"hrank":26.56,"imp":23.287,"train_time":644.68,"predict_time":0.398,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-77","imputed":false},{"position":50,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1162.0,"score":0.044,"rank":43.99,"hrank":33.18,"imp":23.787,"train_time":346.84,"predict_time":0.936,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-79","imputed":false},{"position":51,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1160.0,"score":0.038,"rank":44.23,"hrank":35.68,"imp":23.342,"train_time":1.94,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-55","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1145.0,"score":0.044,"rank":45.41,"hrank":31.9,"imp":24.412,"train_time":346.84,"predict_time":0.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-82","imputed":false},{"position":53,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1133.0,"score":0.029,"rank":46.36,"hrank":38.41,"imp":23.672,"train_time":2.44,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-55","imputed":false},{"position":54,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1131.0,"score":0.067,"rank":46.53,"hrank":24.86,"imp":24.696,"train_time":335.81,"predict_time":85.05,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-69","imputed":false},{"position":55,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1110.0,"score":0.05,"rank":48.19,"hrank":33.46,"imp":25.731,"train_time":218.34,"predict_time":0.904,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-77","imputed":false},{"position":56,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1103.0,"score":0.03,"rank":48.74,"hrank":33.69,"imp":25.592,"train_time":484.44,"predict_time":1.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-95","imputed":false},{"position":57,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1091.0,"score":0.153,"rank":49.73,"hrank":15.2,"imp":28.369,"train_time":460.12,"predict_time":6.641,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-118","imputed":false},{"position":58,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1076.0,"score":0.032,"rank":50.82,"hrank":36.08,"imp":27.211,"train_time":218.34,"predict_time":0.322,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-79","imputed":false},{"position":59,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1073.0,"score":0.093,"rank":51.03,"hrank":28.52,"imp":27.717,"train_time":1.3,"predict_time":0.15,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-126","imputed":false},{"position":60,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1071.0,"score":0.011,"rank":51.18,"hrank":37.2,"imp":26.155,"train_time":484.44,"predict_time":0.086,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-92","imputed":false},{"position":61,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1056.0,"score":0.02,"rank":52.27,"hrank":35.23,"imp":26.524,"train_time":9.39,"predict_time":0.13,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-85","imputed":false},{"position":62,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1024.0,"score":0.072,"rank":54.57,"hrank":31.12,"imp":31.469,"train_time":278.97,"predict_time":0.873,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-133","imputed":false},{"position":63,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1020.0,"score":0.003,"rank":54.84,"hrank":32.09,"imp":27.183,"train_time":2.97,"predict_time":0.388,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-90","imputed":false},{"position":64,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1003.0,"score":0.051,"rank":55.9,"hrank":32.44,"imp":31.983,"train_time":278.97,"predict_time":0.103,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-131","imputed":false},{"position":65,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1002.0,"score":0.016,"rank":56.02,"hrank":46.6,"imp":28.642,"train_time":2.07,"predict_time":0.314,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-128","imputed":false},{"position":66,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.011,"rank":56.12,"hrank":47.82,"imp":28.917,"train_time":0.42,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-119","imputed":false},{"position":67,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":991.0,"score":0.013,"rank":56.68,"hrank":39.55,"imp":29.078,"train_time":90.74,"predict_time":1.622,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-116","imputed":false},{"position":68,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":990.0,"score":0.011,"rank":56.78,"hrank":48.73,"imp":30.341,"train_time":3.71,"predict_time":0.919,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-110","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":972.0,"score":0.0,"rank":57.89,"hrank":44.22,"imp":29.448,"train_time":30.83,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-66","imputed":false},{"position":70,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":960.0,"score":0.033,"rank":58.58,"hrank":36.25,"imp":33.897,"train_time":1.48,"predict_time":0.139,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-165","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":924.0,"score":0.001,"rank":60.54,"hrank":41.0,"imp":33.136,"train_time":90.74,"predict_time":0.125,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-138","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":634.0,"score":0.0,"rank":70.12,"hrank":62.0,"imp":45.667,"train_time":0.22,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+101\/-192","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_small/pareto_front_explorer.html
index 9e15d92f548afde9dcca644d8f88e9ee1510657e..6f08b30e5f1fd6259d0495dc011e06c667938598 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.9820731122,"imp":19.1549251895,"elo":1383.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"imp":19.9999562181,"elo":1311.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0307451934,"imp":20.008337301,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3119174081,"imp":19.9016074158,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0288287122,"imp":22.144746317,"elo":1234.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0276568517,"imp":21.5436554551,"elo":1260.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2363905932,"imp":21.15924534,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0179394275,"imp":22.3145339242,"elo":1220.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0173877944,"imp":21.8911161264,"elo":1239.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1553298525,"imp":21.3889878481,"elo":1261.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3137781461,"imp":28.642273278,"elo":1001.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0776142438,"imp":24.4123643152,"elo":1145.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9356355775,"imp":23.78723511,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3877560139,"imp":27.1827622952,"elo":1019.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3977238019,"imp":23.2870081079,"elo":1171.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.9735168391,"imp":21.9690489118,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1213716865,"imp":23.6721438768,"elo":1133.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.2429152387,"imp":19.4992663448,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1797709942,"imp":19.078790777,"elo":1357.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.4627845186,"imp":16.5651137541,"elo":1438.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3338546996,"imp":22.0729648093,"elo":1262.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5122402643,"imp":19.9400241121,"elo":1311.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4495512241,"imp":21.2385060258,"elo":1269.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"imp":23.2304117787,"elo":1187.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"imp":29.4475599162,"elo":971.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3219913165,"imp":27.2113672784,"elo":1076.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9044251985,"imp":25.7306490048,"elo":1110.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0408261303,"imp":28.9166379461,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0857054373,"imp":26.155153016,"elo":1071.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.0254370493,"imp":25.5923886427,"elo":1103.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":2.2477938016,"imp":21.727291787,"elo":1219.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.2247970899,"imp":19.5266395622,"elo":1326.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":15.5885181427,"imp":17.8472997553,"elo":1419.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5540470709,"imp":14.5736357928,"elo":1558.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2654117448,"imp":14.6664908893,"elo":1527.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.2520254135,"imp":14.0764179877,"elo":1551.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7629525797,"imp":21.6678090289,"elo":1300.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.6109974271,"imp":19.9831851224,"elo":1288.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":37.6060491721,"imp":17.6725123861,"elo":1360.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":268.5791952447,"imp":17.0958026355,"elo":1390.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1141716321,"imp":16.6892247751,"elo":1438.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.4319322507,"imp":4.5179772688,"elo":1740.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1499608083,"imp":27.7173588647,"elo":1073.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1522267247,"imp":12.887486245,"elo":1576.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2482801599,"imp":20.650669487,"elo":1271.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2655194186,"imp":18.8453853545,"elo":1359.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.7085526712,"imp":17.9713455804,"elo":1410.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4326735827,"imp":14.0165424738,"elo":1564.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.374483993,"imp":12.6330995832,"elo":1641.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.6409697056,"imp":28.3689738479,"elo":1090.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.9075519343,"imp":23.9659547102,"elo":1181.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"imp":24.1153567219,"elo":1174.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0519986869,"imp":18.5836287419,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1303337018,"imp":26.524284719,"elo":1056.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1432880461,"imp":22.6700330604,"elo":1184.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.6194496632,"imp":20.7333746595,"elo":1253.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209029857,"imp":23.3422194494,"elo":1159.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1936851087,"imp":20.7697014338,"elo":1258.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.3974981596,"imp":20.4238912654,"elo":1282.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":85.0504523595,"imp":24.6962773288,"elo":1131.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":119.2837367952,"imp":21.9727204853,"elo":1224.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":543.2330416203,"imp":19.3688836867,"elo":1325.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"imp":30.3405389829,"elo":989.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1622176419,"imp":21.3101738943,"elo":1257.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9831667257,"imp":20.482203316,"elo":1280.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.9820731122,"x_train":317.9218245308,"imp":19.1549251895,"elo":1383.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"x_train":3.8348511861,"imp":19.9999562181,"elo":1311.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0307451934,"x_train":1260.8478030837,"imp":20.008337301,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3119174081,"x_train":1260.8478030837,"imp":19.9016074158,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0288287122,"x_train":2.5798987548,"imp":22.144746317,"elo":1234.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0276568517,"x_train":550.7724074523,"imp":21.5436554551,"elo":1260.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2363905932,"x_train":550.7724074523,"imp":21.15924534,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0179394275,"x_train":6.6731468509,"imp":22.3145339242,"elo":1220.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0173877944,"x_train":1353.2049878904,"imp":21.8911161264,"elo":1239.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1553298525,"x_train":1353.2049878904,"imp":21.3889878481,"elo":1261.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3137781461,"x_train":2.0680378675,"imp":28.642273278,"elo":1001.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0776142438,"x_train":346.8355167309,"imp":24.4123643152,"elo":1145.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9356355775,"x_train":346.8355167309,"imp":23.78723511,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3877560139,"x_train":2.9685174704,"imp":27.1827622952,"elo":1019.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3977238019,"x_train":644.6819040616,"imp":23.2870081079,"elo":1171.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.9735168391,"x_train":644.6819040616,"imp":21.9690489118,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1213716865,"x_train":2.4426615648,"imp":23.6721438768,"elo":1133.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.2429152387,"x_train":512.6694116522,"imp":19.4992663448,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1797709942,"x_train":512.6694116522,"imp":19.078790777,"elo":1357.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.4627845186,"x_train":19.824824253,"imp":16.5651137541,"elo":1438.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3338546996,"x_train":14.3767255664,"imp":22.0729648093,"elo":1262.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5122402643,"x_train":4806.0384822712,"imp":19.9400241121,"elo":1311.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4495512241,"x_train":4806.0384822712,"imp":21.2385060258,"elo":1269.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"x_train":14.1568412145,"imp":23.2304117787,"elo":1187.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"x_train":30.8283410858,"imp":29.4475599162,"elo":971.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3219913165,"x_train":218.3366110643,"imp":27.2113672784,"elo":1076.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9044251985,"x_train":218.3366110643,"imp":25.7306490048,"elo":1110.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0408261303,"x_train":0.4188243081,"imp":28.9166379461,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0857054373,"x_train":484.4354636329,"imp":26.155153016,"elo":1071.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.0254370493,"x_train":484.4354636329,"imp":25.5923886427,"elo":1103.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":2.2477938016,"x_train":10.8961223125,"imp":21.727291787,"elo":1219.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.2247970899,"x_train":2950.7244476948,"imp":19.5266395622,"elo":1326.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":15.5885181427,"x_train":2950.7244476948,"imp":17.8472997553,"elo":1419.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5540470709,"x_train":5.9621453989,"imp":14.5736357928,"elo":1558.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2654117448,"x_train":2052.2840837184,"imp":14.6664908893,"elo":1527.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.2520254135,"x_train":2052.2840837184,"imp":14.0764179877,"elo":1551.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7629525797,"x_train":9.0524779449,"imp":21.6678090289,"elo":1300.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.6109974271,"x_train":51.8297758697,"imp":19.9831851224,"elo":1288.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":37.6060491721,"x_train":5351.9343480662,"imp":17.6725123861,"elo":1360.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":268.5791952447,"x_train":5351.9343480662,"imp":17.0958026355,"elo":1390.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1141716321,"x_train":2.2957537095,"imp":16.6892247751,"elo":1438.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.4319322507,"x_train":36.2449142975,"imp":4.5179772688,"elo":1740.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1499608083,"x_train":1.2985567013,"imp":27.7173588647,"elo":1073.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1522267247,"x_train":2.0883166208,"imp":12.887486245,"elo":1576.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2482801599,"x_train":6.8016802788,"imp":20.650669487,"elo":1271.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2655194186,"x_train":2289.910634932,"imp":18.8453853545,"elo":1359.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.7085526712,"x_train":2289.910634932,"imp":17.9713455804,"elo":1410.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4326735827,"x_train":5.4825798778,"imp":14.0165424738,"elo":1564.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.374483993,"x_train":3.8458762973,"imp":12.6330995832,"elo":1641.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.6409697056,"x_train":460.1168119192,"imp":28.3689738479,"elo":1090.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.9075519343,"x_train":34059.6410419582,"imp":23.9659547102,"elo":1181.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"x_train":34059.6410419582,"imp":24.1153567219,"elo":1174.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0519986869,"x_train":1.2303610879,"imp":18.5836287419,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1303337018,"x_train":9.3924084625,"imp":26.524284719,"elo":1056.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1432880461,"x_train":2665.4817950505,"imp":22.6700330604,"elo":1184.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.6194496632,"x_train":2665.4817950505,"imp":20.7333746595,"elo":1253.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209029857,"x_train":1.9409389396,"imp":23.3422194494,"elo":1159.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1936851087,"x_train":693.4907982506,"imp":20.7697014338,"elo":1258.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.3974981596,"x_train":693.4907982506,"imp":20.4238912654,"elo":1282.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":85.0504523595,"x_train":335.8142132441,"imp":24.6962773288,"elo":1131.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":119.2837367952,"x_train":15277.755542326,"imp":21.9727204853,"elo":1224.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":543.2330416203,"x_train":15277.755542326,"imp":19.3688836867,"elo":1325.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"x_train":3.7137863239,"imp":30.3405389829,"elo":989.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1622176419,"x_train":711.9845683972,"imp":21.3101738943,"elo":1257.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9831667257,"x_train":711.9845683972,"imp":20.482203316,"elo":1280.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_small/tuning_trajectories_explorer.html
index 42bf6aa8838e0ab5b30bc268923ec3164bfc87b1..13f875f665c14682837e051ff8e2630373fc774b 100644
--- a/data/imputation_no/splits_all/tasks_binary/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":317.9218245308,"imp":19.1817646815,"elo":1402.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.8348511861,"imp":20.0255060679,"elo":1329.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":10.7408674121,"imp":19.8222589297,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":27.0793552473,"imp":19.6989793478,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":55.6913166543,"imp":19.7092530951,"elo":1344.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":153.0098769834,"imp":19.7480226201,"elo":1342.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":311.0672039337,"imp":19.7799817595,"elo":1340.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":627.0754171345,"imp":19.845803195,"elo":1335.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":940.3702560612,"imp":19.8922358101,"elo":1332.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1260.8478030837,"imp":19.923048383,"elo":1330.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.5798987548,"imp":22.169192029,"elo":1247.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.3455889861,"imp":21.8483001236,"elo":1266.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":13.3365552104,"imp":21.7021786704,"elo":1274.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":26.9265316391,"imp":21.6165688455,"elo":1278.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":68.2541543508,"imp":21.436697954,"elo":1287.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":136.870208803,"imp":21.3431392395,"elo":1293.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":274.0188126874,"imp":21.2536023207,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":410.8168496029,"imp":21.2104760654,"elo":1298.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":550.7724074523,"imp":21.1828118062,"elo":1299.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":6.6731468509,"imp":22.3406018121,"elo":1229.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":13.6833132635,"imp":22.1081418095,"elo":1237.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":33.7629667518,"imp":21.8401955261,"elo":1252.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":67.3034244419,"imp":21.758616324,"elo":1257.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":168.595185677,"imp":21.6656703697,"elo":1262.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":338.3045915746,"imp":21.5479266361,"elo":1269.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":673.6824905284,"imp":21.4579297188,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1009.6693952358,"imp":21.4365411425,"elo":1276.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1353.2049878904,"imp":21.4219014356,"elo":1277.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.0680378675,"imp":28.666151925,"elo":1002.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.8007843383,"imp":27.2301001927,"elo":1038.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.0291964114,"imp":25.2058548097,"elo":1101.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":17.6458570814,"imp":24.4395051793,"elo":1137.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.5325190163,"imp":24.0789913339,"elo":1155.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":86.5677692966,"imp":23.9349157944,"elo":1163.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":172.7997856037,"imp":23.8560157392,"elo":1169.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":258.9344533948,"imp":23.8090395359,"elo":1174.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":346.8355167309,"imp":23.8154939597,"elo":1174.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.9685174704,"imp":27.207602488,"elo":1026.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.1141929845,"imp":25.8715264859,"elo":1073.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":16.0049294587,"imp":24.0028881486,"elo":1147.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":32.4300042816,"imp":23.06996284,"elo":1184.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":78.9900203661,"imp":22.450511546,"elo":1208.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":158.4909250486,"imp":22.1657209571,"elo":1222.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":319.6026300848,"imp":22.0479893256,"elo":1228.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":480.1317591961,"imp":21.9752474141,"elo":1232.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":644.6819040616,"imp":21.9888505257,"elo":1234.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.4426615648,"imp":23.696453813,"elo":1136.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":5.5591739929,"imp":21.2023897633,"elo":1245.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":13.0753440884,"imp":19.9113102165,"elo":1323.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":25.7142984261,"imp":19.462390748,"elo":1352.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":63.8902151248,"imp":19.2462145925,"elo":1367.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":127.6621046473,"imp":19.1946161298,"elo":1371.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":255.1492455474,"imp":19.1209386379,"elo":1377.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":382.2249600708,"imp":19.0923298342,"elo":1378.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":512.6694116522,"imp":19.1031415573,"elo":1378.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":19.824824253,"imp":16.6275061127,"elo":1456.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.3767255664,"imp":22.098232022,"elo":1279.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":49.2187024257,"imp":21.2237333112,"elo":1290.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":133.7824058533,"imp":19.4530273064,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":249.6696643307,"imp":19.21888521,"elo":1361.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":597.3147049605,"imp":19.00599109,"elo":1366.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1201.702225649,"imp":19.4961943033,"elo":1344.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2376.3135821935,"imp":20.2756382252,"elo":1315.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3550.4691289242,"imp":20.9196427284,"elo":1295.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4806.0384822712,"imp":21.2708050711,"elo":1282.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":14.1568412145,"imp":23.252284164,"elo":1197.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":30.8283410858,"imp":30.449936736,"elo":962.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":72.2336761586,"imp":28.0381803997,"elo":1026.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":181.7482066596,"imp":25.8655940481,"elo":1113.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":218.3366110643,"imp":25.75379075,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4188243081,"imp":28.940552711,"elo":999.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.8140094679,"imp":27.6927771241,"elo":1032.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":10.1111434028,"imp":26.8509354002,"elo":1059.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":22.2070189832,"imp":26.3219765417,"elo":1081.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":58.3686895535,"imp":25.8616606479,"elo":1101.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":118.7702866394,"imp":25.7330778941,"elo":1108.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":239.7007801605,"imp":25.6544683268,"elo":1111.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":360.8189481305,"imp":25.6199658385,"elo":1113.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":484.4354636329,"imp":25.613710841,"elo":1113.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.8961223125,"imp":21.7527716232,"elo":1227.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.6705594408,"imp":19.9085579458,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":69.643527334,"imp":18.5767608623,"elo":1394.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":141.1529977589,"imp":18.196374833,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":364.6011469598,"imp":17.9692399565,"elo":1435.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":736.9754853516,"imp":17.8966439603,"elo":1442.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1469.2653973047,"imp":17.889548651,"elo":1444.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2197.5320041687,"imp":17.8943383788,"elo":1444.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2950.7244476948,"imp":17.8759124462,"elo":1443.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.9621453989,"imp":14.6005688527,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.7696901807,"imp":14.6289503115,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":46.2247717016,"imp":14.413973529,"elo":1580.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":96.4776470867,"imp":14.3366324058,"elo":1578.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":247.5589870969,"imp":14.1765391927,"elo":1579.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":503.7582274337,"imp":14.1317656891,"elo":1578.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1020.2013695698,"imp":14.0618918374,"elo":1578.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1526.9732493853,"imp":14.0626223927,"elo":1577.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2052.2840837184,"imp":14.1184919265,"elo":1574.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":9.0524779449,"imp":21.6937725791,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.8297758697,"imp":20.0089070107,"elo":1304.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":77.5384458131,"imp":19.1687399255,"elo":1332.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":145.4125168687,"imp":18.4014636399,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":257.902985689,"imp":18.1346426504,"elo":1373.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":651.1867262454,"imp":17.5283412424,"elo":1393.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1324.8715392081,"imp":17.2851756621,"elo":1402.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2657.8350777127,"imp":17.1626437619,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3995.7346684077,"imp":17.1133154523,"elo":1409.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":5351.9343480662,"imp":17.131476831,"elo":1410.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.2957537095,"imp":16.7150199479,"elo":1453.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":36.2449142975,"imp":4.5460516051,"elo":1762.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.2985567013,"imp":27.7414207128,"elo":1089.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":2.0883166208,"imp":12.9145603199,"elo":1598.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.8016802788,"imp":20.6760380831,"elo":1285.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":15.442951022,"imp":20.2713708923,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":49.8362170361,"imp":19.6959755245,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":110.6090004093,"imp":19.0859674756,"elo":1376.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":279.2571002214,"imp":18.5994921895,"elo":1401.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":563.1030178809,"imp":18.3205304022,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1110.4089598597,"imp":18.1364224337,"elo":1424.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1689.4279279682,"imp":18.0437754732,"elo":1429.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2289.910634932,"imp":17.9988520784,"elo":1433.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.4825798778,"imp":14.0431232305,"elo":1590.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.8458762973,"imp":12.6596377883,"elo":1669.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":460.1168119192,"imp":28.3943723467,"elo":1099.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":734.7515368462,"imp":28.2636153969,"elo":1105.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1321.1026922005,"imp":27.9919891633,"elo":1111.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1962.2511632163,"imp":27.2667870744,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4372.5625009241,"imp":26.1001008014,"elo":1147.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":8707.0506318291,"imp":25.0293474825,"elo":1166.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":17339.6234835031,"imp":24.5121469343,"elo":1180.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":25624.7257379004,"imp":24.295000421,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34059.6410419582,"imp":24.1425024641,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.2303610879,"imp":18.6090906752,"elo":1388.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.3924084625,"imp":26.5370740017,"elo":1067.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":19.9499332005,"imp":24.2480731591,"elo":1133.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":60.5341760166,"imp":22.5021660997,"elo":1183.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":136.8122971072,"imp":21.845324933,"elo":1210.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":328.250602853,"imp":21.349927688,"elo":1234.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":672.0587261642,"imp":21.0408605019,"elo":1250.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1298.4117955909,"imp":20.8960079437,"elo":1259.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1997.6434256187,"imp":20.8262740175,"elo":1266.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2665.4817950505,"imp":20.7460093425,"elo":1271.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.9409389396,"imp":23.3660134041,"elo":1163.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.5392416815,"imp":21.3854906635,"elo":1241.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":16.764226514,"imp":20.9765317511,"elo":1263.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":31.9140091666,"imp":20.7992046487,"elo":1277.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":85.9052711013,"imp":20.6202288947,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":172.6999358448,"imp":20.578873833,"elo":1289.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":344.2808053206,"imp":20.5328959048,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":515.1408947248,"imp":20.4823410698,"elo":1295.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":693.4907982506,"imp":20.448995207,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":335.8142132441,"imp":24.7205305138,"elo":1141.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":882.5688189042,"imp":22.7166924291,"elo":1196.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2676.7112743199,"imp":20.6595321132,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":5942.9139958052,"imp":19.9649575806,"elo":1309.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":14596.1261611529,"imp":19.4090825562,"elo":1340.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":15277.755542326,"imp":19.3927587899,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.7137863239,"imp":30.364927712,"elo":996.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":7.2328490388,"imp":27.0728141202,"elo":1055.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.7783258681,"imp":22.7228048768,"elo":1193.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":35.5509910734,"imp":20.9561810032,"elo":1279.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.843717188,"imp":20.4622449555,"elo":1306.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":177.5290085781,"imp":20.3865956235,"elo":1308.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":354.4498168115,"imp":20.4484983525,"elo":1302.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":530.9324752986,"imp":20.4902163801,"elo":1299.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":711.9845683972,"imp":20.5095484503,"elo":1296.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.9820731122,"x_train":317.9218245308,"imp":19.1817646815,"elo":1402.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"x_train":3.8348511861,"imp":20.0255060679,"elo":1329.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0505639318,"x_train":10.7408674121,"imp":19.8222589297,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1055185035,"x_train":27.0793552473,"imp":19.6989793478,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1439083199,"x_train":55.6913166543,"imp":19.7092530951,"elo":1344.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1870134039,"x_train":153.0098769834,"imp":19.7480226201,"elo":1342.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2323100656,"x_train":311.0672039337,"imp":19.7799817595,"elo":1340.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2769297222,"x_train":627.0754171345,"imp":19.845803195,"elo":1335.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3021056593,"x_train":940.3702560612,"imp":19.8922358101,"elo":1332.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3157761587,"x_train":1260.8478030837,"imp":19.923048383,"elo":1330.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0288287122,"x_train":2.5798987548,"imp":22.169192029,"elo":1247.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0559389647,"x_train":5.3455889861,"imp":21.8483001236,"elo":1266.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.095923574,"x_train":13.3365552104,"imp":21.7021786704,"elo":1274.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1330256679,"x_train":26.9265316391,"imp":21.6165688455,"elo":1278.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1721658573,"x_train":68.2541543508,"imp":21.436697954,"elo":1287.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1894366664,"x_train":136.870208803,"imp":21.3431392395,"elo":1293.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2208324522,"x_train":274.0188126874,"imp":21.2536023207,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.226853687,"x_train":410.8168496029,"imp":21.2104760654,"elo":1298.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2454393854,"x_train":550.7724074523,"imp":21.1828118062,"elo":1299.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0179394275,"x_train":6.6731468509,"imp":22.3406018121,"elo":1229.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.034525777,"x_train":13.6833132635,"imp":22.1081418095,"elo":1237.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.058506366,"x_train":33.7629667518,"imp":21.8401955261,"elo":1252.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0797047657,"x_train":67.3034244419,"imp":21.758616324,"elo":1257.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1112174743,"x_train":168.595185677,"imp":21.6656703697,"elo":1262.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.128617728,"x_train":338.3045915746,"imp":21.5479266361,"elo":1269.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1479697023,"x_train":673.6824905284,"imp":21.4579297188,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1471018087,"x_train":1009.6693952358,"imp":21.4365411425,"elo":1276.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1455019881,"x_train":1353.2049878904,"imp":21.4219014356,"elo":1277.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3137781461,"x_train":2.0680378675,"imp":28.666151925,"elo":1002.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3830515747,"x_train":3.8007843383,"imp":27.2301001927,"elo":1038.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4375192148,"x_train":9.0291964114,"imp":25.2058548097,"elo":1101.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4630669777,"x_train":17.6458570814,"imp":24.4395051793,"elo":1137.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5956078235,"x_train":43.5325190163,"imp":24.0789913339,"elo":1155.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7076605511,"x_train":86.5677692966,"imp":23.9349157944,"elo":1163.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7982051709,"x_train":172.7997856037,"imp":23.8560157392,"elo":1169.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8552846086,"x_train":258.9344533948,"imp":23.8090395359,"elo":1174.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9356355775,"x_train":346.8355167309,"imp":23.8154939597,"elo":1174.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3877560139,"x_train":2.9685174704,"imp":27.207602488,"elo":1026.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6804478637,"x_train":6.1141929845,"imp":25.8715264859,"elo":1073.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3000093411,"x_train":16.0049294587,"imp":24.0028881486,"elo":1147.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0596118244,"x_train":32.4300042816,"imp":23.06996284,"elo":1184.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1608857062,"x_train":78.9900203661,"imp":22.450511546,"elo":1208.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.8825453215,"x_train":158.4909250486,"imp":22.1657209571,"elo":1222.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.8073759391,"x_train":319.6026300848,"imp":22.0479893256,"elo":1228.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6151177765,"x_train":480.1317591961,"imp":21.9752474141,"elo":1232.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.0018475561,"x_train":644.6819040616,"imp":21.9888505257,"elo":1234.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1213716865,"x_train":2.4426615648,"imp":23.696453813,"elo":1136.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4646453215,"x_train":5.5591739929,"imp":21.2023897633,"elo":1245.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7052079417,"x_train":13.0753440884,"imp":19.9113102165,"elo":1323.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9222412801,"x_train":25.7142984261,"imp":19.462390748,"elo":1352.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1944952069,"x_train":63.8902151248,"imp":19.2462145925,"elo":1367.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3691324541,"x_train":127.6621046473,"imp":19.1946161298,"elo":1371.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3439559328,"x_train":255.1492455474,"imp":19.1209386379,"elo":1377.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3558664616,"x_train":382.2249600708,"imp":19.0923298342,"elo":1378.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.190246053,"x_train":512.6694116522,"imp":19.1031415573,"elo":1378.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.4627845186,"x_train":19.824824253,"imp":16.6275061127,"elo":1456.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3338546996,"x_train":14.3767255664,"imp":22.098232022,"elo":1279.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8515440488,"x_train":49.2187024257,"imp":21.2237333112,"elo":1290.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9302543292,"x_train":133.7824058533,"imp":19.4530273064,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.8814779166,"x_train":249.6696643307,"imp":19.21888521,"elo":1361.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1438371055,"x_train":597.3147049605,"imp":19.00599109,"elo":1366.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.494799053,"x_train":1201.702225649,"imp":19.4961943033,"elo":1344.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.0528486148,"x_train":2376.3135821935,"imp":20.2756382252,"elo":1315.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.9073730127,"x_train":3550.4691289242,"imp":20.9196427284,"elo":1295.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4341438441,"x_train":4806.0384822712,"imp":21.2708050711,"elo":1282.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"x_train":14.1568412145,"imp":23.252284164,"elo":1197.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"x_train":30.8283410858,"imp":30.449936736,"elo":962.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2383939645,"x_train":72.2336761586,"imp":28.0381803997,"elo":1026.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7611719559,"x_train":181.7482066596,"imp":25.8655940481,"elo":1113.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9044251985,"x_train":218.3366110643,"imp":25.75379075,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0408261303,"x_train":0.4188243081,"imp":28.940552711,"elo":999.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1187728047,"x_train":2.8140094679,"imp":27.6927771241,"elo":1032.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2923837664,"x_train":10.1111434028,"imp":26.8509354002,"elo":1059.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4489618597,"x_train":22.2070189832,"imp":26.3219765417,"elo":1081.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.643034933,"x_train":58.3686895535,"imp":25.8616606479,"elo":1101.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7794781581,"x_train":118.7702866394,"imp":25.7330778941,"elo":1108.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9035741347,"x_train":239.7007801605,"imp":25.6544683268,"elo":1111.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9768570151,"x_train":360.8189481305,"imp":25.6199658385,"elo":1113.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0127633212,"x_train":484.4354636329,"imp":25.613710841,"elo":1113.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.2477938016,"x_train":10.8961223125,"imp":21.7527716232,"elo":1227.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.7014714901,"x_train":24.6705594408,"imp":19.9085579458,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.7878975485,"x_train":69.643527334,"imp":18.5767608623,"elo":1394.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.822444877,"x_train":141.1529977589,"imp":18.196374833,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.0337121308,"x_train":364.6011469598,"imp":17.9692399565,"elo":1435.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.0024192762,"x_train":736.9754853516,"imp":17.8966439603,"elo":1442.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.6952744743,"x_train":1469.2653973047,"imp":17.889548651,"elo":1444.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.7348675092,"x_train":2197.5320041687,"imp":17.8943383788,"elo":1444.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.535629646,"x_train":2950.7244476948,"imp":17.8759124462,"elo":1443.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5540470709,"x_train":5.9621453989,"imp":14.6005688527,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3798151402,"x_train":16.7696901807,"imp":14.6289503115,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.6778202239,"x_train":46.2247717016,"imp":14.413973529,"elo":1580.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.9104411493,"x_train":96.4776470867,"imp":14.3366324058,"elo":1578.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.6891164694,"x_train":247.5589870969,"imp":14.1765391927,"elo":1579.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8180134178,"x_train":503.7582274337,"imp":14.1317656891,"elo":1578.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6847264435,"x_train":1020.2013695698,"imp":14.0618918374,"elo":1578.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.2438200441,"x_train":1526.9732493853,"imp":14.0626223927,"elo":1577.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.660934628,"x_train":2052.2840837184,"imp":14.1184919265,"elo":1574.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7629525797,"x_train":9.0524779449,"imp":21.6937725791,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.6109974271,"x_train":51.8297758697,"imp":20.0089070107,"elo":1304.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.9837454782,"x_train":77.5384458131,"imp":19.1687399255,"elo":1332.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":67.6378270251,"x_train":145.4125168687,"imp":18.4014636399,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":86.9150727201,"x_train":257.902985689,"imp":18.1346426504,"elo":1373.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":128.1526540176,"x_train":651.1867262454,"imp":17.5283412424,"elo":1393.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":168.6065426596,"x_train":1324.8715392081,"imp":17.2851756621,"elo":1402.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":206.1759779511,"x_train":2657.8350777127,"imp":17.1626437619,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":240.0720860454,"x_train":3995.7346684077,"imp":17.1133154523,"elo":1409.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":264.8900452851,"x_train":5351.9343480662,"imp":17.131476831,"elo":1410.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1141716321,"x_train":2.2957537095,"imp":16.7150199479,"elo":1453.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":5.4319322507,"x_train":36.2449142975,"imp":4.5460516051,"elo":1762.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1499608083,"x_train":1.2985567013,"imp":27.7414207128,"elo":1089.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1522267247,"x_train":2.0883166208,"imp":12.9145603199,"elo":1598.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2482801599,"x_train":6.8016802788,"imp":20.6760380831,"elo":1285.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.5017988445,"x_train":15.442951022,"imp":20.2713708923,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9596655794,"x_train":49.8362170361,"imp":19.6959755245,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.3053184018,"x_train":110.6090004093,"imp":19.0859674756,"elo":1376.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.7312799575,"x_train":279.2571002214,"imp":18.5994921895,"elo":1401.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":2.0837991009,"x_train":563.1030178809,"imp":18.3205304022,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":2.3607400079,"x_train":1110.4089598597,"imp":18.1364224337,"elo":1424.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.5984736372,"x_train":1689.4279279682,"imp":18.0437754732,"elo":1429.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.955519638,"x_train":2289.910634932,"imp":17.9988520784,"elo":1433.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4326735827,"x_train":5.4825798778,"imp":14.0431232305,"elo":1590.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.374483993,"x_train":3.8458762973,"imp":12.6596377883,"elo":1669.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.6409697056,"x_train":460.1168119192,"imp":28.3943723467,"elo":1099.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.3541500481,"x_train":734.7515368462,"imp":28.2636153969,"elo":1105.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1404765981,"x_train":1321.1026922005,"imp":27.9919891633,"elo":1111.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.0682608485,"x_train":1962.2511632163,"imp":27.2667870744,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.4116860286,"x_train":4372.5625009241,"imp":26.1001008014,"elo":1147.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.581168526,"x_train":8707.0506318291,"imp":25.0293474825,"elo":1166.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.0586264094,"x_train":17339.6234835031,"imp":24.5121469343,"elo":1180.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8750179092,"x_train":25624.7257379004,"imp":24.295000421,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"x_train":34059.6410419582,"imp":24.1425024641,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0519986869,"x_train":1.2303610879,"imp":18.6090906752,"elo":1388.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1303337018,"x_train":9.3924084625,"imp":26.5370740017,"elo":1067.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2624076267,"x_train":19.9499332005,"imp":24.2480731591,"elo":1133.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5692050445,"x_train":60.5341760166,"imp":22.5021660997,"elo":1183.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8653863265,"x_train":136.8122971072,"imp":21.845324933,"elo":1210.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2483118673,"x_train":328.250602853,"imp":21.349927688,"elo":1234.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6672661515,"x_train":672.0587261642,"imp":21.0408605019,"elo":1250.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1766444304,"x_train":1298.4117955909,"imp":20.8960079437,"elo":1259.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4248878932,"x_train":1997.6434256187,"imp":20.8262740175,"elo":1266.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.6309014567,"x_train":2665.4817950505,"imp":20.7460093425,"elo":1271.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209029857,"x_train":1.9409389396,"imp":23.3660134041,"elo":1163.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4595679196,"x_train":5.5392416815,"imp":21.3854906635,"elo":1241.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6794374605,"x_train":16.764226514,"imp":20.9765317511,"elo":1263.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7612793276,"x_train":31.9140091666,"imp":20.7992046487,"elo":1277.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9549360649,"x_train":85.9052711013,"imp":20.6202288947,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0795695279,"x_train":172.6999358448,"imp":20.578873833,"elo":1289.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1949903748,"x_train":344.2808053206,"imp":20.5328959048,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.3595882138,"x_train":515.1408947248,"imp":20.4823410698,"elo":1295.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.3974981596,"x_train":693.4907982506,"imp":20.448995207,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":85.0504523595,"x_train":335.8142132441,"imp":24.7205305138,"elo":1141.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":166.0881040231,"x_train":882.5688189042,"imp":22.7166924291,"elo":1196.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":280.1665667632,"x_train":2676.7112743199,"imp":20.6595321132,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":413.7384149023,"x_train":5942.9139958052,"imp":19.9649575806,"elo":1309.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":528.3767220648,"x_train":14596.1261611529,"imp":19.4090825562,"elo":1340.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":543.2330416203,"x_train":15277.755542326,"imp":19.3927587899,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"x_train":3.7137863239,"imp":30.364927712,"elo":996.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.0338429173,"x_train":7.2328490388,"imp":27.0728141202,"elo":1055.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1335512608,"x_train":17.7783258681,"imp":22.7228048768,"elo":1193.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.4441868365,"x_train":35.5509910734,"imp":20.9561810032,"elo":1279.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.7785012767,"x_train":88.843717188,"imp":20.4622449555,"elo":1306.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2922510872,"x_train":177.5290085781,"imp":20.3865956235,"elo":1308.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.6799279262,"x_train":354.4498168115,"imp":20.4484983525,"elo":1302.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.8648219542,"x_train":530.9324752986,"imp":20.4902163801,"elo":1299.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.9514581229,"x_train":711.9845683972,"imp":20.5095484503,"elo":1296.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_small/winrate_explorer.html b/data/imputation_no/splits_all/tasks_binary/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..573b1f01b887156e5e18ec5cb67bce3cfb467ca5
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_binary/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "TabICLv2 (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "TabDPT-Turbo (default)", "LimiX (default)", "TabSwift (default)", "AutoGluon 1.4 (best, 4h)", "RealMLP (tuned + ensembled)", "BetaTabPFN (default)", "TabM (tuned + ensembled)", "TabDPT (tuned + ensembled)", "TabDPT (tuned)", "LightGBM (tuned + ensembled)", "TabM (tuned)", "TabDPT (default)", "LightGBM (tuned)", "RealMLP (tuned)", "SAP-RPT-OSS (default)", "xRFM (tuned)", "xRFM (tuned + ensembled)", "ModernNCA (tuned)", "ModernNCA (default)", "iLTM (tuned + ensembled)", "CatBoost (default)", "FastaiMLP (tuned + ensembled)", "CatBoost (tuned + ensembled)", "EBM (tuned + ensembled)", "CatBoost (tuned)", "TorchMLP (tuned + ensembled)", "EBM (tuned)", "TabM (default)", "XGBoost (tuned + ensembled)", "RealMLP (default)", "EBM (default)", "TabSTAR (tuned)", "ChimeraBoost (tuned + ensembled)", "OrionMSP (default)", "TabSTAR (tuned + ensembled)", "FastaiMLP (tuned)", "XGBoost (tuned)", "ChimeraBoost (tuned)", "iLTM (tuned)", "ChimeraBoost (default)", "TorchMLP (tuned)", "ModernNCA (tuned + ensembled)", "TabFlex (default)", "ExtraTrees (tuned + ensembled)", "TabSTAR (default)", "ExtraTrees (tuned)", "iLTM (default)", "Linear (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "XGBoost (default)", "LightGBM (default)", "Linear (tuned)", "TorchMLP (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "RandomForest (tuned + ensembled)", "RandomForest (tuned)", "Linear (default)", "xRFM (default)", "KNN (tuned + ensembled)", "ExtraTrees (default)", "RandomForest (default)", "PerpetualBooster (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7731092572212219, 0.7899159789085388, 0.7759103775024414, 0.7787114977836609, 0.7899159789085388, 0.7997198700904846, 0.8067227005958557, 0.8095238208770752, 0.894957959651947, 0.848739504814148, 0.8963585495948792, 0.8935574293136597, 0.8361344337463379, 0.8879551887512207, 0.8823529481887817, 0.8767507076263428, 0.8991596698760986, 0.924369752407074, 0.8851540684700012, 0.8991596698760986, 0.8851540684700012, 0.8711484670639038, 0.8795518279075623, 0.9103641510009766, 0.9075630307197571, 0.8739495873451233, 0.9215686321258545, 0.924369752407074, 0.8963585495948792, 0.924369752407074, 0.8907563090324402, 0.924369752407074, 0.9523809552192688, 0.8991596698760986, 0.9215686321258545, 0.9523809552192688, 0.8935574293136597, 0.8907563090324402, 0.7787114977836609, 0.9299719929695129, 0.9299719929695129, 0.7815126180648804, 0.9159663915634155, 0.9551820755004883, 0.9439775943756104, 0.9355742335319519, 0.9495798349380493, 0.9607843160629272, 0.9551820755004883, 0.9103641510009766, 0.9439775943756104, 0.8207283020019531, 0.9327731132507324, 0.9271708726882935, 0.913165271282196, 0.9775910377502441, 0.9691876769065857, 0.9663865566253662, 0.924369752407074, 0.9719887971878052, 0.9803921580314636, 0.9523809552192688, 0.9775910377502441, 0.9747899174690247, 0.9439775943756104, 0.9803921580314636, 0.9775910377502441, 0.9831932783126831, 0.9747899174690247, 0.9859943985939026, 0.9663865566253662, 0.9915966391563416], [0.22689075767993927, NaN, 0.5070028305053711, 0.6078431606292725, 0.6666666865348816, 0.6386554837226868, 0.6582633256912231, 0.686274528503418, 0.686274528503418, 0.7478991746902466, 0.8011204600334167, 0.761904776096344, 0.8067227005958557, 0.7352941036224365, 0.7983193397521973, 0.8067227005958557, 0.8235294222831726, 0.8039215803146362, 0.83753502368927, 0.8207283020019531, 0.831932783126831, 0.8291316628456116, 0.8095238208770752, 0.8235294222831726, 0.8599439859390259, 0.8599439859390259, 0.8221288323402405, 0.8879551887512207, 0.8711484670639038, 0.8305321931838989, 0.8599439859390259, 0.843137264251709, 0.8739495873451233, 0.9047619104385376, 0.8543417453765869, 0.8683473467826843, 0.8781512379646301, 0.8683473467826843, 0.8445377945899963, 0.7352941036224365, 0.9019607901573181, 0.8795518279075623, 0.7507002949714661, 0.8613445162773132, 0.8991596698760986, 0.9103641510009766, 0.9173669219017029, 0.8991596698760986, 0.9271708726882935, 0.918767511844635, 0.8823529481887817, 0.9327731132507324, 0.7927170991897583, 0.9271708726882935, 0.9159663915634155, 0.8711484670639038, 0.9215686321258545, 0.9495798349380493, 0.9467787146568298, 0.8823529481887817, 0.9439775943756104, 0.9411764740943909, 0.9439775943756104, 0.9383753538131714, 0.9607843160629272, 0.9271708726882935, 0.9635854363441467, 0.9411764740943909, 0.9635854363441467, 0.9691876769065857, 0.9635854363441467, 0.9551820755004883, 0.994397759437561], [0.21008403599262238, 0.4929971992969513, NaN, 0.5854341983795166, 0.586834728717804, 0.6274510025978088, 0.6078431606292725, 0.6246498823165894, 0.686274528503418, 0.7577030658721924, 0.7871148586273193, 0.7955182194709778, 0.7927170991897583, 0.7422969341278076, 0.8011204600334167, 0.7969187498092651, 0.7843137383460999, 0.7899159789085388, 0.8179271817207336, 0.8151260614395142, 0.8151260614395142, 0.8123249411582947, 0.7759103775024414, 0.813725471496582, 0.8515406250953674, 0.8459383845329285, 0.8277310729026794, 0.8655462265014648, 0.848739504814148, 0.843137264251709, 0.8347339034080505, 0.8347339034080505, 0.8459383845329285, 0.8991596698760986, 0.831932783126831, 0.843137264251709, 0.8627451062202454, 0.8627451062202454, 0.8403361439704895, 0.7507002949714661, 0.8795518279075623, 0.9075630307197571, 0.7478991746902466, 0.8711484670639038, 0.8795518279075623, 0.9019607901573181, 0.8879551887512207, 0.8907563090324402, 0.9215686321258545, 0.913165271282196, 0.8739495873451233, 0.894957959651947, 0.7983193397521973, 0.8935574293136597, 0.9075630307197571, 0.8907563090324402, 0.9299719929695129, 0.9271708726882935, 0.9411764740943909, 0.9019607901573181, 0.9523809552192688, 0.9411764740943909, 0.9355742335319519, 0.9327731132507324, 0.9299719929695129, 0.9313725233078003, 0.9607843160629272, 0.9467787146568298, 0.9411764740943909, 0.9467787146568298, 0.9747899174690247, 0.9495798349380493, 0.994397759437561], [0.2240896373987198, 0.3921568691730499, 0.4145658314228058, NaN, 0.5658263564109802, 0.6134454011917114, 0.5910364389419556, 0.6078431606292725, 0.5406162738800049, 0.6918767690658569, 0.7282913327217102, 0.7899159789085388, 0.7801120281219482, 0.7030812501907349, 0.8011204600334167, 0.7450980544090271, 0.7591036558151245, 0.8095238208770752, 0.8109243512153625, 0.7310924530029297, 0.8095238208770752, 0.8207283020019531, 0.7591036558151245, 0.7857142686843872, 0.843137264251709, 0.843137264251709, 0.7983193397521973, 0.8627451062202454, 0.8333333134651184, 0.8235294222831726, 0.8403361439704895, 0.8067227005958557, 0.8305321931838989, 0.8795518279075623, 0.8123249411582947, 0.8515406250953674, 0.8627451062202454, 0.8207283020019531, 0.8235294222831726, 0.6890756487846375, 0.8823529481887817, 0.8837534785270691, 0.6918767690658569, 0.8347339034080505, 0.8711484670639038, 0.8907563090324402, 0.8963585495948792, 0.8879551887512207, 0.9075630307197571, 0.9271708726882935, 0.8543417453765869, 0.9075630307197571, 0.756302535533905, 0.8879551887512207, 0.9047619104385376, 0.8459383845329285, 0.924369752407074, 0.9271708726882935, 0.9355742335319519, 0.8795518279075623, 0.9355742335319519, 0.9397758841514587, 0.924369752407074, 0.9299719929695129, 0.9355742335319519, 0.9089635610580444, 0.9299719929695129, 0.913165271282196, 0.9383753538131714, 0.9411764740943909, 0.9551820755004883, 0.9327731132507324, 0.9915966391563416], [0.2212885171175003, 0.3333333432674408, 0.41316527128219604, 0.43417367339134216, NaN, 0.5210084319114685, 0.518207311630249, 0.5602241158485413, 0.5490196347236633, 0.6722689270973206, 0.7310924530029297, 0.7394958138465881, 0.7394958138465881, 0.7212885022163391, 0.7478991746902466, 0.7058823704719543, 0.7072829008102417, 0.7899159789085388, 0.7955182194709778, 0.7170868515968323, 0.8123249411582947, 0.8179271817207336, 0.7815126180648804, 0.8179271817207336, 0.8291316628456116, 0.8123249411582947, 0.8123249411582947, 0.8291316628456116, 0.8711484670639038, 0.8459383845329285, 0.8403361439704895, 0.8403361439704895, 0.8291316628456116, 0.8823529481887817, 0.8655462265014648, 0.8683473467826843, 0.8515406250953674, 0.8599439859390259, 0.8347339034080505, 0.7282913327217102, 0.8963585495948792, 0.8935574293136597, 0.732492983341217, 0.8711484670639038, 0.8795518279075623, 0.9019607901573181, 0.906162440776825, 0.9047619104385376, 0.924369752407074, 0.8739495873451233, 0.8809523582458496, 0.8879551887512207, 0.7703081369400024, 0.8879551887512207, 0.9103641510009766, 0.8599439859390259, 0.9215686321258545, 0.924369752407074, 0.9579831957817078, 0.8739495873451233, 0.9635854363441467, 0.9355742335319519, 0.9271708726882935, 0.913165271282196, 0.9439775943756104, 0.9327731132507324, 0.9551820755004883, 0.913165271282196, 0.9355742335319519, 0.9523809552192688, 0.9635854363441467, 0.9327731132507324, 0.9915966391563416], [0.21008403599262238, 0.3613445460796356, 0.37254902720451355, 0.38655462861061096, 0.4789915978908539, NaN, 0.4789915978908539, 0.5196078419685364, 0.5518207550048828, 0.6694678068161011, 0.6834734082221985, 0.7338935732841492, 0.7535014152526855, 0.6834734082221985, 0.7535014152526855, 0.6946778893470764, 0.680672287940979, 0.7787114977836609, 0.7927170991897583, 0.6694678068161011, 0.8151260614395142, 0.8067227005958557, 0.7731092572212219, 0.7815126180648804, 0.8179271817207336, 0.8291316628456116, 0.8095238208770752, 0.8599439859390259, 0.843137264251709, 0.7759103775024414, 0.8403361439704895, 0.7871148586273193, 0.8235294222831726, 0.8599439859390259, 0.8109243512153625, 0.8305321931838989, 0.843137264251709, 0.8235294222831726, 0.7927170991897583, 0.686274528503418, 0.8963585495948792, 0.8179271817207336, 0.7002801299095154, 0.8207283020019531, 0.8599439859390259, 0.9047619104385376, 0.8879551887512207, 0.9075630307197571, 0.9047619104385376, 0.9299719929695129, 0.8515406250953674, 0.8991596698760986, 0.7422969341278076, 0.8865545988082886, 0.8907563090324402, 0.8459383845329285, 0.918767511844635, 0.9439775943756104, 0.9495798349380493, 0.8767507076263428, 0.9383753538131714, 0.918767511844635, 0.9103641510009766, 0.9327731132507324, 0.9467787146568298, 0.9103641510009766, 0.9439775943756104, 0.924369752407074, 0.9327731132507324, 0.9523809552192688, 0.9523809552192688, 0.9355742335319519, 0.9887955188751221], [0.2002801150083542, 0.34173670411109924, 0.3921568691730499, 0.4089635908603668, 0.48179271817207336, 0.5210084319114685, NaN, 0.5910364389419556, 0.5154061913490295, 0.6946778893470764, 0.6694678068161011, 0.7422969341278076, 0.7366946935653687, 0.7114846110343933, 0.7394958138465881, 0.6694678068161011, 0.680672287940979, 0.7815126180648804, 0.7787114977836609, 0.6246498823165894, 0.8025209903717041, 0.7759103775024414, 0.7577030658721924, 0.7815126180648804, 0.8151260614395142, 0.8347339034080505, 0.7927170991897583, 0.819327712059021, 0.831932783126831, 0.7941176295280457, 0.8151260614395142, 0.7955182194709778, 0.8123249411582947, 0.8543417453765869, 0.8109243512153625, 0.8179271817207336, 0.8347339034080505, 0.8123249411582947, 0.8039215803146362, 0.6890756487846375, 0.8669467568397522, 0.8515406250953674, 0.6974790096282959, 0.843137264251709, 0.8613445162773132, 0.8823529481887817, 0.8739495873451233, 0.8823529481887817, 0.8963585495948792, 0.8879551887512207, 0.8459383845329285, 0.8655462265014648, 0.7422969341278076, 0.8571428656578064, 0.8907563090324402, 0.8599439859390259, 0.924369752407074, 0.9215686321258545, 0.9383753538131714, 0.8823529481887817, 0.918767511844635, 0.9299719929695129, 0.918767511844635, 0.9159663915634155, 0.918767511844635, 0.894957959651947, 0.9299719929695129, 0.913165271282196, 0.9327731132507324, 0.9327731132507324, 0.9523809552192688, 0.9159663915634155, 0.9887955188751221], [0.19327731430530548, 0.3137255012989044, 0.37535014748573303, 0.3921568691730499, 0.43977591395378113, 0.4803921580314636, 0.4089635908603668, NaN, 0.518207311630249, 0.6470588445663452, 0.656862735748291, 0.7142857313156128, 0.6974790096282959, 0.6974790096282959, 0.7170868515968323, 0.6610644459724426, 0.6820728182792664, 0.7647058963775635, 0.7633053064346313, 0.6414566040039062, 0.7955182194709778, 0.761904776096344, 0.7478991746902466, 0.7647058963775635, 0.7829131484031677, 0.7871148586273193, 0.7703081369400024, 0.7955182194709778, 0.8039215803146362, 0.7787114977836609, 0.8011204600334167, 0.8039215803146362, 0.7955182194709778, 0.8515406250953674, 0.8067227005958557, 0.8179271817207336, 0.8207283020019531, 0.8095238208770752, 0.7899159789085388, 0.680672287940979, 0.8473389148712158, 0.8459383845329285, 0.6918767690658569, 0.8235294222831726, 0.831932783126831, 0.8683473467826843, 0.8655462265014648, 0.8711484670639038, 0.8963585495948792, 0.8781512379646301, 0.831932783126831, 0.8613445162773132, 0.7450980544090271, 0.8599439859390259, 0.8851540684700012, 0.8403361439704895, 0.900560200214386, 0.9159663915634155, 0.9327731132507324, 0.8599439859390259, 0.9271708726882935, 0.9215686321258545, 0.9047619104385376, 0.8991596698760986, 0.9103641510009766, 0.906162440776825, 0.9271708726882935, 0.9103641510009766, 0.924369752407074, 0.9299719929695129, 0.9551820755004883, 0.918767511844635, 0.9887955188751221], [0.190476194024086, 0.3137255012989044, 0.3137255012989044, 0.4593837559223175, 0.45098039507865906, 0.4481792747974396, 0.48459383845329285, 0.48179271817207336, NaN, 0.5714285969734192, 0.6904761791229248, 0.6078431606292725, 0.6442577242851257, 0.5742297172546387, 0.6176470518112183, 0.67507004737854, 0.6834734082221985, 0.6526610851287842, 0.6666666865348816, 0.7885153889656067, 0.6778711676597595, 0.7156862616539001, 0.686274528503418, 0.7310924530029297, 0.7198879718780518, 0.7058823704719543, 0.7394958138465881, 0.7282913327217102, 0.7142857313156128, 0.767507016658783, 0.7366946935653687, 0.7310924530029297, 0.7282913327217102, 0.756302535533905, 0.7521008253097534, 0.7394958138465881, 0.756302535533905, 0.7703081369400024, 0.7507002949714661, 0.6694678068161011, 0.767507016658783, 0.8291316628456116, 0.680672287940979, 0.8123249411582947, 0.767507016658783, 0.7899159789085388, 0.8305321931838989, 0.8095238208770752, 0.8151260614395142, 0.761904776096344, 0.8711484670639038, 0.8697478771209717, 0.7422969341278076, 0.8963585495948792, 0.831932783126831, 0.831932783126831, 0.8627451062202454, 0.8683473467826843, 0.8823529481887817, 0.8515406250953674, 0.9019607901573181, 0.8907563090324402, 0.9159663915634155, 0.913165271282196, 0.9159663915634155, 0.9159663915634155, 0.9355742335319519, 0.918767511844635, 0.9341736435890198, 0.9327731132507324, 0.9579831957817078, 0.9411764740943909, 0.9915966391563416], [0.10504201799631119, 0.2521008551120758, 0.24229691922664642, 0.30812326073646545, 0.32773110270500183, 0.3305322229862213, 0.30532214045524597, 0.3529411852359772, 0.4285714328289032, NaN, 0.5266106724739075, 0.6302521228790283, 0.6442577242851257, 0.5742297172546387, 0.6246498823165894, 0.5476190447807312, 0.5518207550048828, 0.686274528503418, 0.6400560140609741, 0.5770308375358582, 0.6974790096282959, 0.7044817805290222, 0.7030812501907349, 0.6820728182792664, 0.7254902124404907, 0.7254902124404907, 0.7002801299095154, 0.8165265917778015, 0.7366946935653687, 0.6960784196853638, 0.7422969341278076, 0.7450980544090271, 0.7394958138465881, 0.7787114977836609, 0.761904776096344, 0.7170868515968323, 0.7787114977836609, 0.7394958138465881, 0.756302535533905, 0.6764705777168274, 0.8095238208770752, 0.7577030658721924, 0.6722689270973206, 0.7521008253097534, 0.8011204600334167, 0.831932783126831, 0.8263305425643921, 0.8403361439704895, 0.8123249411582947, 0.83753502368927, 0.7366946935653687, 0.8235294222831726, 0.732492983341217, 0.8123249411582947, 0.8599439859390259, 0.8095238208770752, 0.8739495873451233, 0.8795518279075623, 0.9019607901573181, 0.83753502368927, 0.8599439859390259, 0.8907563090324402, 0.8879551887512207, 0.8935574293136597, 0.9159663915634155, 0.8459383845329285, 0.918767511844635, 0.8739495873451233, 0.924369752407074, 0.9355742335319519, 0.924369752407074, 0.8991596698760986, 0.9859943985939026], [0.15126051008701324, 0.19887955486774445, 0.21288515627384186, 0.2717086970806122, 0.2689075767993927, 0.3165266215801239, 0.3305322229862213, 0.343137264251709, 0.3095238208770752, 0.4733893573284149, NaN, 0.5532212853431702, 0.5770308375358582, 0.4901960790157318, 0.5210084319114685, 0.5112044811248779, 0.5350140333175659, 0.5770308375358582, 0.5630252361297607, 0.5630252361297607, 0.5826330780982971, 0.6414566040039062, 0.6302521228790283, 0.6358543634414673, 0.6666666865348816, 0.6722689270973206, 0.6638655662536621, 0.67507004737854, 0.6442577242851257, 0.7338935732841492, 0.6498599648475647, 0.6694678068161011, 0.6652660965919495, 0.7282913327217102, 0.6834734082221985, 0.6722689270973206, 0.6722689270973206, 0.7114846110343933, 0.6778711676597595, 0.6918767690658569, 0.686274528503418, 0.8095238208770752, 0.7058823704719543, 0.767507016658783, 0.7114846110343933, 0.7254902124404907, 0.7955182194709778, 0.7731092572212219, 0.7647058963775635, 0.7338935732841492, 0.8627451062202454, 0.8263305425643921, 0.7591036558151245, 0.8207283020019531, 0.8011204600334167, 0.8123249411582947, 0.8459383845329285, 0.8179271817207336, 0.8361344337463379, 0.8347339034080505, 0.9075630307197571, 0.8403361439704895, 0.8935574293136597, 0.8543417453765869, 0.8711484670639038, 0.8725489974021912, 0.8907563090324402, 0.8711484670639038, 0.8795518279075623, 0.8935574293136597, 0.9075630307197571, 0.9327731132507324, 0.9915966391563416], [0.10364145785570145, 0.2380952388048172, 0.2044817954301834, 0.21008403599262238, 0.26050421595573425, 0.2661064565181732, 0.25770309567451477, 0.2857142984867096, 0.3921568691730499, 0.36974790692329407, 0.44677871465682983, NaN, 0.5392156839370728, 0.5126050710678101, 0.523809552192688, 0.4901960790157318, 0.4789915978908539, 0.6498599648475647, 0.5910364389419556, 0.5266106724739075, 0.6834734082221985, 0.6470588445663452, 0.5378151535987854, 0.6498599648475647, 0.6372548937797546, 0.605042040348053, 0.6386554837226868, 0.686274528503418, 0.7142857313156128, 0.6834734082221985, 0.7198879718780518, 0.6778711676597595, 0.7282913327217102, 0.756302535533905, 0.6946778893470764, 0.7058823704719543, 0.767507016658783, 0.7338935732841492, 0.7226890921592712, 0.6358543634414673, 0.743697464466095, 0.6288515329360962, 0.6470588445663452, 0.7338935732841492, 0.761904776096344, 0.761904776096344, 0.7843137383460999, 0.7759103775024414, 0.7955182194709778, 0.7142857313156128, 0.7394958138465881, 0.756302535533905, 0.7058823704719543, 0.7647058963775635, 0.8179271817207336, 0.7997198700904846, 0.9047619104385376, 0.8935574293136597, 0.918767511844635, 0.8333333134651184, 0.8627451062202454, 0.9103641510009766, 0.8907563090324402, 0.8711484670639038, 0.8865545988082886, 0.8683473467826843, 0.913165271282196, 0.8963585495948792, 0.924369752407074, 0.9215686321258545, 0.9327731132507324, 0.8991596698760986, 0.994397759437561], [0.10644257813692093, 0.19327731430530548, 0.2072829157114029, 0.21988795697689056, 0.26050421595573425, 0.24649859964847565, 0.26330533623695374, 0.3025210201740265, 0.35574230551719666, 0.35574230551719666, 0.42296919226646423, 0.46078431606292725, NaN, 0.48459383845329285, 0.5126050710678101, 0.46078431606292725, 0.4691876769065857, 0.6274510025978088, 0.5910364389419556, 0.523809552192688, 0.6708683371543884, 0.7254902124404907, 0.5490196347236633, 0.6134454011917114, 0.6554622054100037, 0.6134454011917114, 0.6134454011917114, 0.7142857313156128, 0.7058823704719543, 0.6960784196853638, 0.7394958138465881, 0.7114846110343933, 0.7310924530029297, 0.7030812501907349, 0.7198879718780518, 0.7086834907531738, 0.7731092572212219, 0.7857142686843872, 0.7156862616539001, 0.6246498823165894, 0.7703081369400024, 0.6442577242851257, 0.6386554837226868, 0.7268907427787781, 0.7983193397521973, 0.7857142686843872, 0.7731092572212219, 0.7871148586273193, 0.756302535533905, 0.7058823704719543, 0.7114846110343933, 0.8011204600334167, 0.6918767690658569, 0.7941176295280457, 0.8459383845329285, 0.7871148586273193, 0.8991596698760986, 0.8823529481887817, 0.8907563090324402, 0.7927170991897583, 0.8711484670639038, 0.9103641510009766, 0.8991596698760986, 0.8963585495948792, 0.9215686321258545, 0.8179271817207336, 0.9075630307197571, 0.8851540684700012, 0.924369752407074, 0.9523809552192688, 0.9579831957817078, 0.9075630307197571, 0.9915966391563416], [0.16386555135250092, 0.2647058963775635, 0.25770309567451477, 0.2969187796115875, 0.2787114977836609, 0.3165266215801239, 0.2885154187679291, 0.3025210201740265, 0.4257703125476837, 0.4257703125476837, 0.5098039507865906, 0.48739495873451233, 0.5154061913490295, NaN, 0.4789915978908539, 0.523809552192688, 0.5322129130363464, 0.6470588445663452, 0.5574229955673218, 0.5882353186607361, 0.6358543634414673, 0.5826330780982971, 0.5770308375358582, 0.599439799785614, 0.6246498823165894, 0.6372548937797546, 0.6330532431602478, 0.6386554837226868, 0.7058823704719543, 0.6260504126548767, 0.6638655662536621, 0.7254902124404907, 0.6960784196853638, 0.6694678068161011, 0.7366946935653687, 0.7338935732841492, 0.7478991746902466, 0.6918767690658569, 0.7647058963775635, 0.6694678068161011, 0.7408963441848755, 0.6526610851287842, 0.67507004737854, 0.7030812501907349, 0.767507016658783, 0.767507016658783, 0.7030812501907349, 0.7843137383460999, 0.7478991746902466, 0.686274528503418, 0.7156862616539001, 0.8179271817207336, 0.7408963441848755, 0.8123249411582947, 0.8599439859390259, 0.8543417453765869, 0.8235294222831726, 0.8347339034080505, 0.8557422757148743, 0.8823529481887817, 0.8263305425643921, 0.843137264251709, 0.8039215803146362, 0.8935574293136597, 0.8851540684700012, 0.9047619104385376, 0.8207283020019531, 0.913165271282196, 0.8683473467826843, 0.9019607901573181, 0.9271708726882935, 0.9383753538131714, 0.9803921580314636], [0.1120448186993599, 0.20168067514896393, 0.19887955486774445, 0.19887955486774445, 0.2521008551120758, 0.24649859964847565, 0.26050421595573425, 0.2829131782054901, 0.38235294818878174, 0.37535014748573303, 0.4789915978908539, 0.4761904776096344, 0.48739495873451233, 0.5210084319114685, NaN, 0.49439775943756104, 0.4901960790157318, 0.6330532431602478, 0.66246497631073, 0.523809552192688, 0.66246497631073, 0.6386554837226868, 0.5686274766921997, 0.5742297172546387, 0.6022409200668335, 0.6260504126548767, 0.6330532431602478, 0.66246497631073, 0.7170868515968323, 0.6442577242851257, 0.7002801299095154, 0.7198879718780518, 0.7086834907531738, 0.6946778893470764, 0.7282913327217102, 0.7997198700904846, 0.7591036558151245, 0.7212885022163391, 0.7184873819351196, 0.6386554837226868, 0.7787114977836609, 0.6246498823165894, 0.6442577242851257, 0.6974790096282959, 0.8011204600334167, 0.8263305425643921, 0.7535014152526855, 0.831932783126831, 0.7689075469970703, 0.7366946935653687, 0.7142857313156128, 0.7927170991897583, 0.6974790096282959, 0.7647058963775635, 0.831932783126831, 0.8123249411582947, 0.8683473467826843, 0.8795518279075623, 0.8837534785270691, 0.8277310729026794, 0.8711484670639038, 0.8907563090324402, 0.8683473467826843, 0.8515406250953674, 0.8767507076263428, 0.8571428656578064, 0.8823529481887817, 0.8683473467826843, 0.9047619104385376, 0.9047619104385376, 0.9579831957817078, 0.8879551887512207, 0.9775910377502441], [0.11764705926179886, 0.19327731430530548, 0.20308123528957367, 0.2549019753932953, 0.29411765933036804, 0.30532214045524597, 0.3305322229862213, 0.33893558382987976, 0.32492998242378235, 0.4523809552192688, 0.48879551887512207, 0.5098039507865906, 0.5392156839370728, 0.4761904776096344, 0.505602240562439, NaN, 0.5854341983795166, 0.5462185144424438, 0.5560224056243896, 0.5826330780982971, 0.599439799785614, 0.6372548937797546, 0.6246498823165894, 0.6176470518112183, 0.6428571343421936, 0.6330532431602478, 0.6400560140609741, 0.6694678068161011, 0.6358543634414673, 0.6666666865348816, 0.6442577242851257, 0.6372548937797546, 0.6414566040039062, 0.7044817805290222, 0.6652660965919495, 0.6456582546234131, 0.6596638560295105, 0.67507004737854, 0.6498599648475647, 0.6134454011917114, 0.6890756487846375, 0.7366946935653687, 0.6162465214729309, 0.7254902124404907, 0.7030812501907349, 0.7198879718780518, 0.7759103775024414, 0.7450980544090271, 0.7394958138465881, 0.761904776096344, 0.7591036558151245, 0.8039215803146362, 0.7058823704719543, 0.8039215803146362, 0.8235294222831726, 0.756302535533905, 0.8291316628456116, 0.8207283020019531, 0.848739504814148, 0.7815126180648804, 0.8599439859390259, 0.8501400351524353, 0.8851540684700012, 0.8823529481887817, 0.8851540684700012, 0.8291316628456116, 0.9173669219017029, 0.8753501176834106, 0.913165271282196, 0.9103641510009766, 0.9215686321258545, 0.8879551887512207, 0.9915966391563416], [0.12324929982423782, 0.1764705926179886, 0.21568627655506134, 0.24089635908603668, 0.2927170991897583, 0.3193277418613434, 0.3193277418613434, 0.31792718172073364, 0.3165266215801239, 0.4481792747974396, 0.46498599648475647, 0.5210084319114685, 0.5308123230934143, 0.46778711676597595, 0.5098039507865906, 0.4145658314228058, NaN, 0.5378151535987854, 0.5546218752861023, 0.5644257664680481, 0.5658263564109802, 0.5882353186607361, 0.5882353186607361, 0.5770308375358582, 0.6106442809104919, 0.605042040348053, 0.6330532431602478, 0.6358543634414673, 0.6246498823165894, 0.6582633256912231, 0.6330532431602478, 0.6554622054100037, 0.6260504126548767, 0.7058823704719543, 0.6666666865348816, 0.6358543634414673, 0.656862735748291, 0.6834734082221985, 0.6638655662536621, 0.5980392098426819, 0.67507004737854, 0.7507002949714661, 0.5966386795043945, 0.7058823704719543, 0.6722689270973206, 0.6946778893470764, 0.7507002949714661, 0.7226890921592712, 0.7254902124404907, 0.738095223903656, 0.7535014152526855, 0.761904776096344, 0.7002801299095154, 0.7703081369400024, 0.7899159789085388, 0.7605041861534119, 0.8095238208770752, 0.7871148586273193, 0.8039215803146362, 0.7843137383460999, 0.848739504814148, 0.8207283020019531, 0.8655462265014648, 0.8403361439704895, 0.8571428656578064, 0.8291316628456116, 0.8795518279075623, 0.8739495873451233, 0.8739495873451233, 0.8879551887512207, 0.918767511844635, 0.8795518279075623, 0.9859943985939026], [0.10084033757448196, 0.19607843458652496, 0.21008403599262238, 0.190476194024086, 0.21008403599262238, 0.2212885171175003, 0.21848739683628082, 0.23529411852359772, 0.3473389446735382, 0.3137255012989044, 0.42296919226646423, 0.3501400649547577, 0.37254902720451355, 0.3529411852359772, 0.3669467866420746, 0.45378151535987854, 0.462184876203537, NaN, 0.44397759437561035, 0.5126050710678101, 0.5966386795043945, 0.5098039507865906, 0.5070028305053711, 0.5630252361297607, 0.5574229955673218, 0.5378151535987854, 0.5322129130363464, 0.5896358489990234, 0.6764705777168274, 0.5686274766921997, 0.6638655662536621, 0.6176470518112183, 0.656862735748291, 0.5854341983795166, 0.6498599648475647, 0.6764705777168274, 0.7899159789085388, 0.6302521228790283, 0.7058823704719543, 0.6120448112487793, 0.6484593749046326, 0.605042040348053, 0.6162465214729309, 0.6386554837226868, 0.8151260614395142, 0.6834734082221985, 0.7002801299095154, 0.7843137383460999, 0.686274528503418, 0.6582633256912231, 0.6414566040039062, 0.7689075469970703, 0.7058823704719543, 0.7478991746902466, 0.8207283020019531, 0.7815126180648804, 0.8515406250953674, 0.8851540684700012, 0.9439775943756104, 0.7955182194709778, 0.7731092572212219, 0.8795518279075623, 0.7703081369400024, 0.8851540684700012, 0.9159663915634155, 0.8543417453765869, 0.8235294222831726, 0.8823529481887817, 0.9075630307197571, 0.9411764740943909, 0.9327731132507324, 0.9033613204956055, 0.9887955188751221], [0.07563025504350662, 0.16246499121189117, 0.18207283318042755, 0.18907563388347626, 0.2044817954301834, 0.2072829157114029, 0.2212885171175003, 0.23669467866420746, 0.3333333432674408, 0.3599439859390259, 0.43697479367256165, 0.4089635908603668, 0.4089635908603668, 0.4425770342350006, 0.33753502368927, 0.44397759437561035, 0.4453781545162201, 0.5560224056243896, NaN, 0.48879551887512207, 0.5882353186607361, 0.529411792755127, 0.48739495873451233, 0.518207311630249, 0.523809552192688, 0.6106442809104919, 0.593837559223175, 0.5910364389419556, 0.6498599648475647, 0.5896358489990234, 0.6526610851287842, 0.6106442809104919, 0.6526610851287842, 0.6526610851287842, 0.6414566040039062, 0.6834734082221985, 0.6876750588417053, 0.67507004737854, 0.6554622054100037, 0.6162465214729309, 0.7394958138465881, 0.5910364389419556, 0.6204481720924377, 0.6666666865348816, 0.7058823704719543, 0.7366946935653687, 0.67507004737854, 0.7745097875595093, 0.7128851413726807, 0.6918767690658569, 0.6946778893470764, 0.7282913327217102, 0.6722689270973206, 0.7226890921592712, 0.8095238208770752, 0.7843137383460999, 0.8627451062202454, 0.8263305425643921, 0.8291316628456116, 0.8011204600334167, 0.8515406250953674, 0.8767507076263428, 0.8067227005958557, 0.8263305425643921, 0.8207283020019531, 0.831932783126831, 0.831932783126831, 0.848739504814148, 0.8627451062202454, 0.8711484670639038, 0.9467787146568298, 0.8571428656578064, 0.9663865566253662], [0.11484593898057938, 0.17927171289920807, 0.18487395346164703, 0.2689075767993927, 0.2829131782054901, 0.3305322229862213, 0.37535014748573303, 0.35854342579841614, 0.21148459613323212, 0.42296919226646423, 0.43697479367256165, 0.4733893573284149, 0.4761904776096344, 0.4117647111415863, 0.4761904776096344, 0.41736695170402527, 0.4355742335319519, 0.48739495873451233, 0.5112044811248779, NaN, 0.5210084319114685, 0.5462185144424438, 0.5546218752861023, 0.5462185144424438, 0.5686274766921997, 0.5826330780982971, 0.6022409200668335, 0.5714285969734192, 0.5714285969734192, 0.5966386795043945, 0.5658263564109802, 0.5546218752861023, 0.5742297172546387, 0.6414566040039062, 0.581232488155365, 0.6246498823165894, 0.6022409200668335, 0.5896358489990234, 0.6274510025978088, 0.5798319578170776, 0.6582633256912231, 0.7198879718780518, 0.5854341983795166, 0.6778711676597595, 0.6274510025978088, 0.6498599648475647, 0.67507004737854, 0.6638655662536621, 0.6918767690658569, 0.6526610851287842, 0.7759103775024414, 0.7731092572212219, 0.67507004737854, 0.7731092572212219, 0.7478991746902466, 0.7422969341278076, 0.7703081369400024, 0.7366946935653687, 0.7394958138465881, 0.7591036558151245, 0.831932783126831, 0.7955182194709778, 0.848739504814148, 0.7927170991897583, 0.8207283020019531, 0.8095238208770752, 0.848739504814148, 0.8851540684700012, 0.8795518279075623, 0.8823529481887817, 0.8739495873451233, 0.9103641510009766, 0.9887955188751221], [0.10084033757448196, 0.16806723177433014, 0.18487395346164703, 0.190476194024086, 0.18767507374286652, 0.18487395346164703, 0.1974789947271347, 0.2044817954301834, 0.32212886214256287, 0.3025210201740265, 0.41736695170402527, 0.3165266215801239, 0.3291316628456116, 0.3641456663608551, 0.33753502368927, 0.4005602300167084, 0.43417367339134216, 0.40336135029792786, 0.4117647111415863, 0.4789915978908539, NaN, 0.4901960790157318, 0.48179271817207336, 0.518207311630249, 0.5252100825309753, 0.48459383845329285, 0.5098039507865906, 0.5350140333175659, 0.6344537734985352, 0.5336134433746338, 0.6176470518112183, 0.5966386795043945, 0.6078431606292725, 0.5630252361297607, 0.6218487620353699, 0.6330532431602478, 0.6904761791229248, 0.6176470518112183, 0.6526610851287842, 0.5882353186607361, 0.5910364389419556, 0.5826330780982971, 0.593837559223175, 0.5882353186607361, 0.7422969341278076, 0.6190476417541504, 0.6638655662536621, 0.7338935732841492, 0.6652660965919495, 0.6246498823165894, 0.6246498823165894, 0.743697464466095, 0.6834734082221985, 0.732492983341217, 0.7969187498092651, 0.767507016658783, 0.8305321931838989, 0.8417366743087769, 0.9033613204956055, 0.7857142686843872, 0.761904776096344, 0.8515406250953674, 0.7591036558151245, 0.83753502368927, 0.8739495873451233, 0.8347339034080505, 0.8081232309341431, 0.8543417453765869, 0.8879551887512207, 0.9103641510009766, 0.9299719929695129, 0.8879551887512207, 0.9915966391563416], [0.11484593898057938, 0.17086835205554962, 0.18767507374286652, 0.17927171289920807, 0.18207283318042755, 0.19327731430530548, 0.2240896373987198, 0.2380952388048172, 0.28431373834609985, 0.2955182194709778, 0.35854342579841614, 0.3529411852359772, 0.27450981736183167, 0.41736695170402527, 0.3613445460796356, 0.36274510622024536, 0.4117647111415863, 0.4901960790157318, 0.47058823704719543, 0.45378151535987854, 0.5098039507865906, NaN, 0.4733893573284149, 0.505602240562439, 0.4957983195781708, 0.4957983195781708, 0.5434173941612244, 0.5196078419685364, 0.5854341983795166, 0.5798319578170776, 0.5574229955673218, 0.5910364389419556, 0.5658263564109802, 0.605042040348053, 0.6302521228790283, 0.6162465214729309, 0.6246498823165894, 0.6498599648475647, 0.5854341983795166, 0.6022409200668335, 0.6134454011917114, 0.5560224056243896, 0.6162465214729309, 0.6526610851287842, 0.66246497631073, 0.6582633256912231, 0.6442577242851257, 0.6890756487846375, 0.6834734082221985, 0.6106442809104919, 0.6876750588417053, 0.7086834907531738, 0.67507004737854, 0.7086834907531738, 0.7871148586273193, 0.7058823704719543, 0.8207283020019531, 0.7815126180648804, 0.8039215803146362, 0.7156862616539001, 0.8263305425643921, 0.8291316628456116, 0.8585433959960938, 0.7955182194709778, 0.8207283020019531, 0.7731092572212219, 0.843137264251709, 0.8179271817207336, 0.8179271817207336, 0.8571428656578064, 0.9355742335319519, 0.843137264251709, 0.9747899174690247], [0.12885154783725739, 0.190476194024086, 0.2240896373987198, 0.24089635908603668, 0.21848739683628082, 0.22689075767993927, 0.24229691922664642, 0.2521008551120758, 0.3137255012989044, 0.2969187796115875, 0.36974790692329407, 0.462184876203537, 0.45098039507865906, 0.42296919226646423, 0.4313725531101227, 0.37535014748573303, 0.4117647111415863, 0.4929971992969513, 0.5126050710678101, 0.4453781545162201, 0.518207311630249, 0.5266106724739075, NaN, 0.5462185144424438, 0.5658263564109802, 0.5378151535987854, 0.5574229955673218, 0.5462185144424438, 0.5210084319114685, 0.6162465214729309, 0.5042017102241516, 0.5518207550048828, 0.518207311630249, 0.6386554837226868, 0.5826330780982971, 0.5714285969734192, 0.5910364389419556, 0.6554622054100037, 0.5896358489990234, 0.5490196347236633, 0.5882353186607361, 0.5798319578170776, 0.5434173941612244, 0.6554622054100037, 0.6078431606292725, 0.6162465214729309, 0.6120448112487793, 0.6414566040039062, 0.6890756487846375, 0.6162465214729309, 0.6974790096282959, 0.67507004737854, 0.6414566040039062, 0.67507004737854, 0.7198879718780518, 0.7310924530029297, 0.7647058963775635, 0.7254902124404907, 0.7282913327217102, 0.7366946935653687, 0.767507016658783, 0.7745097875595093, 0.7787114977836609, 0.7689075469970703, 0.7773109078407288, 0.7703081369400024, 0.8347339034080505, 0.7787114977836609, 0.8207283020019531, 0.8291316628456116, 0.8179271817207336, 0.8515406250953674, 0.9747899174690247], [0.12044817954301834, 0.1764705926179886, 0.18627451360225677, 0.2142857164144516, 0.18207283318042755, 0.21848739683628082, 0.21848739683628082, 0.23529411852359772, 0.2689075767993927, 0.31792718172073364, 0.3641456663608551, 0.3501400649547577, 0.38655462861061096, 0.4005602300167084, 0.4257703125476837, 0.38235294818878174, 0.42296919226646423, 0.43697479367256165, 0.48179271817207336, 0.45378151535987854, 0.48179271817207336, 0.49439775943756104, 0.45378151535987854, NaN, 0.4355742335319519, 0.45378151535987854, 0.5378151535987854, 0.4803921580314636, 0.5350140333175659, 0.5420168042182922, 0.5406162738800049, 0.5728291273117065, 0.5546218752861023, 0.5574229955673218, 0.605042040348053, 0.5686274766921997, 0.5518207550048828, 0.5854341983795166, 0.6078431606292725, 0.5798319578170776, 0.6148459315299988, 0.5658263564109802, 0.5826330780982971, 0.6358543634414673, 0.5840336084365845, 0.6358543634414673, 0.6330532431602478, 0.6274510025978088, 0.6386554837226868, 0.5686274766921997, 0.7226890921592712, 0.7170868515968323, 0.6554622054100037, 0.7310924530029297, 0.7703081369400024, 0.7591036558151245, 0.738095223903656, 0.7002801299095154, 0.7450980544090271, 0.7871148586273193, 0.8291316628456116, 0.7941176295280457, 0.8347339034080505, 0.7647058963775635, 0.8039215803146362, 0.8151260614395142, 0.8767507076263428, 0.8459383845329285, 0.8459383845329285, 0.8599439859390259, 0.8963585495948792, 0.8683473467826843, 0.9859943985939026], [0.08963585644960403, 0.14005602896213531, 0.14845938980579376, 0.1568627506494522, 0.17086835205554962, 0.18207283318042755, 0.18487395346164703, 0.21708683669567108, 0.28011205792427063, 0.27450981736183167, 0.3333333432674408, 0.36274510622024536, 0.3445378243923187, 0.37535014748573303, 0.3977591097354889, 0.3571428656578064, 0.38935574889183044, 0.4425770342350006, 0.4761904776096344, 0.4313725531101227, 0.47478991746902466, 0.5042017102241516, 0.43417367339134216, 0.5644257664680481, NaN, 0.4663865566253662, 0.5042017102241516, 0.49439775943756104, 0.5532212853431702, 0.5532212853431702, 0.5378151535987854, 0.5630252361297607, 0.5448179244995117, 0.5266106724739075, 0.6022409200668335, 0.5686274766921997, 0.5672268867492676, 0.593837559223175, 0.5910364389419556, 0.5448179244995117, 0.599439799785614, 0.5742297172546387, 0.5434173941612244, 0.6302521228790283, 0.5882353186607361, 0.6162465214729309, 0.6092436909675598, 0.6106442809104919, 0.6190476417541504, 0.6302521228790283, 0.6974790096282959, 0.6918767690658569, 0.6442577242851257, 0.7198879718780518, 0.7338935732841492, 0.7226890921592712, 0.7759103775024414, 0.767507016658783, 0.7871148586273193, 0.7422969341278076, 0.8123249411582947, 0.8305321931838989, 0.848739504814148, 0.8235294222831726, 0.8459383845329285, 0.7829131484031677, 0.918767511844635, 0.8501400351524353, 0.9075630307197571, 0.924369752407074, 0.918767511844635, 0.8627451062202454, 0.9859943985939026], [0.09243697673082352, 0.14005602896213531, 0.15406163036823273, 0.1568627506494522, 0.18767507374286652, 0.17086835205554962, 0.16526611149311066, 0.21288515627384186, 0.29411765933036804, 0.27450981736183167, 0.32773110270500183, 0.3949579894542694, 0.38655462861061096, 0.36274510622024536, 0.3739495873451233, 0.3669467866420746, 0.3949579894542694, 0.462184876203537, 0.38935574889183044, 0.41736695170402527, 0.5154061913490295, 0.5042017102241516, 0.462184876203537, 0.5462185144424438, 0.5336134433746338, NaN, 0.5630252361297607, 0.518207311630249, 0.5350140333175659, 0.5672268867492676, 0.5350140333175659, 0.5462185144424438, 0.5518207550048828, 0.5686274766921997, 0.5770308375358582, 0.5350140333175659, 0.5854341983795166, 0.6162465214729309, 0.5826330780982971, 0.5658263564109802, 0.6190476417541504, 0.5378151535987854, 0.5742297172546387, 0.6190476417541504, 0.6036414504051208, 0.6232492923736572, 0.6330532431602478, 0.6582633256912231, 0.6974790096282959, 0.6134454011917114, 0.67507004737854, 0.6526610851287842, 0.6666666865348816, 0.6610644459724426, 0.7478991746902466, 0.686274528503418, 0.7450980544090271, 0.7366946935653687, 0.7478991746902466, 0.7086834907531738, 0.7478991746902466, 0.7759103775024414, 0.7717086672782898, 0.761904776096344, 0.7759103775024414, 0.7507002949714661, 0.8543417453765869, 0.7787114977836609, 0.8235294222831726, 0.8179271817207336, 0.8571428656578064, 0.8235294222831726, 0.9719887971878052], [0.1260504275560379, 0.17787115275859833, 0.17226891219615936, 0.20168067514896393, 0.18767507374286652, 0.190476194024086, 0.2072829157114029, 0.22969187796115875, 0.26050421595573425, 0.299719899892807, 0.3361344635486603, 0.3613445460796356, 0.38655462861061096, 0.3669467866420746, 0.3669467866420746, 0.3599439859390259, 0.3669467866420746, 0.46778711676597595, 0.40616247057914734, 0.3977591097354889, 0.4901960790157318, 0.456582635641098, 0.4425770342350006, 0.462184876203537, 0.4957983195781708, 0.43697479367256165, NaN, 0.48179271817207336, 0.5546218752861023, 0.5098039507865906, 0.5518207550048828, 0.5322129130363464, 0.5630252361297607, 0.5266106724739075, 0.5714285969734192, 0.5406162738800049, 0.5770308375358582, 0.5546218752861023, 0.5910364389419556, 0.5966386795043945, 0.6134454011917114, 0.5854341983795166, 0.605042040348053, 0.5686274766921997, 0.6246498823165894, 0.6330532431602478, 0.5924369692802429, 0.6386554837226868, 0.6162465214729309, 0.5350140333175659, 0.6694678068161011, 0.680672287940979, 0.680672287940979, 0.6834734082221985, 0.761904776096344, 0.7338935732841492, 0.7310924530029297, 0.686274528503418, 0.7198879718780518, 0.7521008253097534, 0.8067227005958557, 0.7647058963775635, 0.7899159789085388, 0.7591036558151245, 0.7843137383460999, 0.8095238208770752, 0.7591036558151245, 0.8599439859390259, 0.8109243512153625, 0.8011204600334167, 0.8403361439704895, 0.8711484670639038, 0.9775910377502441], [0.0784313753247261, 0.1120448186993599, 0.13445378839969635, 0.13725490868091583, 0.17086835205554962, 0.14005602896213531, 0.1806722730398178, 0.2044817954301834, 0.2717086970806122, 0.1834733933210373, 0.32492998242378235, 0.3137255012989044, 0.2857142984867096, 0.3613445460796356, 0.33753502368927, 0.3305322229862213, 0.3641456663608551, 0.41036415100097656, 0.4089635908603668, 0.4285714328289032, 0.46498599648475647, 0.4803921580314636, 0.45378151535987854, 0.5196078419685364, 0.505602240562439, 0.48179271817207336, 0.518207311630249, NaN, 0.5042017102241516, 0.5546218752861023, 0.48179271817207336, 0.529411792755127, 0.5154061913490295, 0.5602241158485413, 0.5686274766921997, 0.529411792755127, 0.5588235259056091, 0.599439799785614, 0.5714285969734192, 0.5518207550048828, 0.593837559223175, 0.5448179244995117, 0.5560224056243896, 0.6260504126548767, 0.6078431606292725, 0.593837559223175, 0.7184873819351196, 0.6428571343421936, 0.6498599648475647, 0.6134454011917114, 0.67507004737854, 0.6792716979980469, 0.6554622054100037, 0.6918767690658569, 0.7422969341278076, 0.6582633256912231, 0.8039215803146362, 0.7955182194709778, 0.8459383845329285, 0.686274528503418, 0.7759103775024414, 0.831932783126831, 0.7731092572212219, 0.8095238208770752, 0.831932783126831, 0.7310924530029297, 0.8907563090324402, 0.8095238208770752, 0.8543417453765869, 0.8851540684700012, 0.8879551887512207, 0.8529411554336548, 0.9691876769065857], [0.07563025504350662, 0.12885154783725739, 0.15126051008701324, 0.1666666716337204, 0.12885154783725739, 0.1568627506494522, 0.16806723177433014, 0.19607843458652496, 0.2857142984867096, 0.26330533623695374, 0.35574230551719666, 0.2857142984867096, 0.29411765933036804, 0.29411765933036804, 0.2829131782054901, 0.3641456663608551, 0.37535014748573303, 0.3235294222831726, 0.3501400649547577, 0.4285714328289032, 0.36554622650146484, 0.4145658314228058, 0.4789915978908539, 0.46498599648475647, 0.44677871465682983, 0.46498599648475647, 0.4453781545162201, 0.4957983195781708, NaN, 0.5070028305053711, 0.5014005899429321, 0.518207311630249, 0.4859943985939026, 0.5546218752861023, 0.5602241158485413, 0.5854341983795166, 0.5644257664680481, 0.5798319578170776, 0.593837559223175, 0.5980392098426819, 0.5588235259056091, 0.5126050710678101, 0.605042040348053, 0.5742297172546387, 0.586834728717804, 0.605042040348053, 0.6498599648475647, 0.6442577242851257, 0.6442577242851257, 0.5854341983795166, 0.5826330780982971, 0.7282913327217102, 0.6666666865348816, 0.7142857313156128, 0.8067227005958557, 0.7030812501907349, 0.7394958138465881, 0.8039215803146362, 0.8347339034080505, 0.7254902124404907, 0.7591036558151245, 0.7871148586273193, 0.7507002949714661, 0.8543417453765869, 0.8627451062202454, 0.7843137383460999, 0.7927170991897583, 0.8627451062202454, 0.8515406250953674, 0.8795518279075623, 0.8683473467826843, 0.8935574293136597, 0.9831932783126831], [0.10364145785570145, 0.16946779191493988, 0.1568627506494522, 0.1764705926179886, 0.15406163036823273, 0.2240896373987198, 0.20588235557079315, 0.2212885171175003, 0.23249299824237823, 0.30392158031463623, 0.2661064565181732, 0.3165266215801239, 0.30392158031463623, 0.3739495873451233, 0.35574230551719666, 0.3333333432674408, 0.34173670411109924, 0.4313725531101227, 0.41036415100097656, 0.40336135029792786, 0.4663865566253662, 0.42016807198524475, 0.3837535083293915, 0.45798319578170776, 0.44677871465682983, 0.4327731132507324, 0.4901960790157318, 0.4453781545162201, 0.4929971992969513, NaN, 0.5014005899429321, 0.5364145636558533, 0.5266106724739075, 0.4901960790157318, 0.5714285969734192, 0.5210084319114685, 0.5406162738800049, 0.5728291273117065, 0.5742297172546387, 0.5854341983795166, 0.5770308375358582, 0.5322129130363464, 0.593837559223175, 0.6554622054100037, 0.5770308375358582, 0.5966386795043945, 0.5686274766921997, 0.5854341983795166, 0.605042040348053, 0.5798319578170776, 0.686274528503418, 0.6610644459724426, 0.6582633256912231, 0.6526610851287842, 0.7226890921592712, 0.7254902124404907, 0.7703081369400024, 0.7086834907531738, 0.7100840210914612, 0.7507002949714661, 0.8165265917778015, 0.7983193397521973, 0.9299719929695129, 0.7829131484031677, 0.7899159789085388, 0.7703081369400024, 0.8235294222831726, 0.8571428656578064, 0.8179271817207336, 0.83753502368927, 0.924369752407074, 0.8739495873451233, 0.9691876769065857], [0.07563025504350662, 0.14005602896213531, 0.16526611149311066, 0.1596638709306717, 0.1596638709306717, 0.1596638709306717, 0.18487395346164703, 0.19887955486774445, 0.26330533623695374, 0.25770309567451477, 0.3501400649547577, 0.28011205792427063, 0.26050421595573425, 0.3361344635486603, 0.299719899892807, 0.35574230551719666, 0.3669467866420746, 0.3361344635486603, 0.3473389446735382, 0.43417367339134216, 0.38235294818878174, 0.4425770342350006, 0.4957983195781708, 0.4593837559223175, 0.462184876203537, 0.46498599648475647, 0.4481792747974396, 0.518207311630249, 0.49859943985939026, 0.49859943985939026, NaN, 0.4789915978908539, 0.5266106724739075, 0.5462185144424438, 0.5224089622497559, 0.5742297172546387, 0.5392156839370728, 0.5462185144424438, 0.5490196347236633, 0.5826330780982971, 0.5630252361297607, 0.5070028305053711, 0.5770308375358582, 0.5546218752861023, 0.586834728717804, 0.5882353186607361, 0.6470588445663452, 0.6246498823165894, 0.6330532431602478, 0.5910364389419556, 0.5630252361297607, 0.7198879718780518, 0.6526610851287842, 0.7156862616539001, 0.7577030658721924, 0.6974790096282959, 0.7535014152526855, 0.8151260614395142, 0.843137264251709, 0.7198879718780518, 0.7366946935653687, 0.7815126180648804, 0.7366946935653687, 0.8459383845329285, 0.8725489974021912, 0.761904776096344, 0.7899159789085388, 0.8627451062202454, 0.8683473467826843, 0.913165271282196, 0.8599439859390259, 0.8739495873451233, 0.9887955188751221], [0.10924369841814041, 0.1568627506494522, 0.16526611149311066, 0.19327731430530548, 0.1596638709306717, 0.21288515627384186, 0.2044817954301834, 0.19607843458652496, 0.2689075767993927, 0.2549019753932953, 0.3305322229862213, 0.32212886214256287, 0.2885154187679291, 0.27450981736183167, 0.28011205792427063, 0.36274510622024536, 0.3445378243923187, 0.38235294818878174, 0.38935574889183044, 0.4453781545162201, 0.40336135029792786, 0.4089635908603668, 0.4481792747974396, 0.42717087268829346, 0.43697479367256165, 0.45378151535987854, 0.46778711676597595, 0.47058823704719543, 0.48179271817207336, 0.46358543634414673, 0.5210084319114685, NaN, 0.5070028305053711, 0.4957983195781708, 0.6470588445663452, 0.5854341983795166, 0.5854341983795166, 0.529411792755127, 0.6414566040039062, 0.5686274766921997, 0.5266106724739075, 0.5658263564109802, 0.5686274766921997, 0.5196078419685364, 0.5854341983795166, 0.5602241158485413, 0.5602241158485413, 0.605042040348053, 0.581232488155365, 0.5462185144424438, 0.6330532431602478, 0.7100840210914612, 0.6666666865348816, 0.6918767690658569, 0.7507002949714661, 0.7422969341278076, 0.7310924530029297, 0.7521008253097534, 0.7647058963775635, 0.7535014152526855, 0.6694678068161011, 0.7577030658721924, 0.7535014152526855, 0.7899159789085388, 0.7983193397521973, 0.7927170991897583, 0.761904776096344, 0.8627451062202454, 0.8459383845329285, 0.8543417453765869, 0.8823529481887817, 0.8753501176834106, 0.9719887971878052], [0.07563025504350662, 0.1260504275560379, 0.15406163036823273, 0.16946779191493988, 0.17086835205554962, 0.1764705926179886, 0.18767507374286652, 0.2044817954301834, 0.2717086970806122, 0.26050421595573425, 0.33473390340805054, 0.2717086970806122, 0.2689075767993927, 0.30392158031463623, 0.29131653904914856, 0.35854342579841614, 0.3739495873451233, 0.343137264251709, 0.3473389446735382, 0.4257703125476837, 0.3921568691730499, 0.43417367339134216, 0.48179271817207336, 0.4453781545162201, 0.4551820755004883, 0.4481792747974396, 0.43697479367256165, 0.48459383845329285, 0.5140056014060974, 0.4733893573284149, 0.4733893573284149, 0.4929971992969513, NaN, 0.5406162738800049, 0.523809552192688, 0.5532212853431702, 0.5630252361297607, 0.5406162738800049, 0.5602241158485413, 0.5882353186607361, 0.5252100825309753, 0.5042017102241516, 0.593837559223175, 0.5588235259056091, 0.593837559223175, 0.5644257664680481, 0.6106442809104919, 0.6134454011917114, 0.6232492923736572, 0.5910364389419556, 0.5518207550048828, 0.7002801299095154, 0.6610644459724426, 0.686274528503418, 0.7338935732841492, 0.6918767690658569, 0.7366946935653687, 0.831932783126831, 0.8235294222831726, 0.7072829008102417, 0.7170868515968323, 0.7703081369400024, 0.7254902124404907, 0.8459383845329285, 0.848739504814148, 0.7507002949714661, 0.7815126180648804, 0.8571428656578064, 0.8599439859390259, 0.9019607901573181, 0.8711484670639038, 0.8823529481887817, 0.9887955188751221], [0.0476190485060215, 0.095238097012043, 0.10084033757448196, 0.12044817954301834, 0.11764705926179886, 0.14005602896213531, 0.14565826952457428, 0.14845938980579376, 0.24369747936725616, 0.2212885171175003, 0.2717086970806122, 0.24369747936725616, 0.2969187796115875, 0.3305322229862213, 0.30532214045524597, 0.2955182194709778, 0.29411765933036804, 0.4145658314228058, 0.3473389446735382, 0.35854342579841614, 0.43697479367256165, 0.3949579894542694, 0.3613445460796356, 0.4425770342350006, 0.4733893573284149, 0.4313725531101227, 0.4733893573284149, 0.43977591395378113, 0.4453781545162201, 0.5098039507865906, 0.45378151535987854, 0.5042017102241516, 0.4593837559223175, NaN, 0.5322129130363464, 0.46778711676597595, 0.505602240562439, 0.5462185144424438, 0.5434173941612244, 0.5518207550048828, 0.5154061913490295, 0.48459383845329285, 0.5616246461868286, 0.6064425706863403, 0.529411792755127, 0.5602241158485413, 0.5560224056243896, 0.5742297172546387, 0.7072829008102417, 0.5714285969734192, 0.6330532431602478, 0.6470588445663452, 0.6582633256912231, 0.6386554837226868, 0.7240896224975586, 0.6904761791229248, 0.7507002949714661, 0.7254902124404907, 0.6974790096282959, 0.7282913327217102, 0.83753502368927, 0.7955182194709778, 0.7997198700904846, 0.7450980544090271, 0.767507016658783, 0.7815126180648804, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.7787114977836609, 0.8851540684700012, 0.8459383845329285, 0.9663865566253662], [0.10084033757448196, 0.14565826952457428, 0.16806723177433014, 0.18767507374286652, 0.13445378839969635, 0.18907563388347626, 0.18907563388347626, 0.19327731430530548, 0.2478991597890854, 0.2380952388048172, 0.3165266215801239, 0.30532214045524597, 0.28011205792427063, 0.26330533623695374, 0.2717086970806122, 0.33473390340805054, 0.3333333432674408, 0.3501400649547577, 0.35854342579841614, 0.418767511844635, 0.3781512677669525, 0.36974790692329407, 0.41736695170402527, 0.3949579894542694, 0.3977591097354889, 0.42296919226646423, 0.4285714328289032, 0.4313725531101227, 0.43977591395378113, 0.4285714328289032, 0.47759103775024414, 0.3529411852359772, 0.4761904776096344, 0.46778711676597595, NaN, 0.5266106724739075, 0.5070028305053711, 0.462184876203537, 0.5924369692802429, 0.5518207550048828, 0.4901960790157318, 0.5280112028121948, 0.5588235259056091, 0.5, 0.5350140333175659, 0.5252100825309753, 0.5266106724739075, 0.5630252361297607, 0.5574229955673218, 0.5140056014060974, 0.5952380895614624, 0.6638655662536621, 0.6414566040039062, 0.6946778893470764, 0.7058823704719543, 0.7142857313156128, 0.7058823704719543, 0.7282913327217102, 0.7450980544090271, 0.7282913327217102, 0.6498599648475647, 0.7310924530029297, 0.7142857313156128, 0.7591036558151245, 0.7871148586273193, 0.756302535533905, 0.7450980544090271, 0.8459383845329285, 0.848739504814148, 0.843137264251709, 0.8627451062202454, 0.8767507076263428, 0.9719887971878052], [0.0784313753247261, 0.13165266811847687, 0.1568627506494522, 0.14845938980579376, 0.13165266811847687, 0.16946779191493988, 0.18207283318042755, 0.18207283318042755, 0.26050421595573425, 0.2829131782054901, 0.32773110270500183, 0.29411765933036804, 0.29131653904914856, 0.2661064565181732, 0.2002801150083542, 0.3543417453765869, 0.3641456663608551, 0.3235294222831726, 0.3165266215801239, 0.37535014748573303, 0.3669467866420746, 0.3837535083293915, 0.4285714328289032, 0.4313725531101227, 0.4313725531101227, 0.46498599648475647, 0.4593837559223175, 0.47058823704719543, 0.4145658314228058, 0.4789915978908539, 0.4257703125476837, 0.4145658314228058, 0.44677871465682983, 0.5322129130363464, 0.4733893573284149, NaN, 0.46498599648475647, 0.5154061913490295, 0.45798319578170776, 0.5966386795043945, 0.48459383845329285, 0.5126050710678101, 0.605042040348053, 0.5574229955673218, 0.5098039507865906, 0.5154061913490295, 0.5644257664680481, 0.5602241158485413, 0.5966386795043945, 0.5714285969734192, 0.5686274766921997, 0.6722689270973206, 0.6834734082221985, 0.680672287940979, 0.6834734082221985, 0.7030812501907349, 0.6820728182792664, 0.6498599648475647, 0.6582633256912231, 0.7240896224975586, 0.7282913327217102, 0.6946778893470764, 0.7142857313156128, 0.7871148586273193, 0.7843137383460999, 0.7591036558151245, 0.7254902124404907, 0.8235294222831726, 0.8123249411582947, 0.83753502368927, 0.831932783126831, 0.8655462265014648, 0.9607843160629272], [0.0476190485060215, 0.12184873968362808, 0.13725490868091583, 0.13725490868091583, 0.14845938980579376, 0.1568627506494522, 0.16526611149311066, 0.17927171289920807, 0.24369747936725616, 0.2212885171175003, 0.32773110270500183, 0.23249299824237823, 0.22689075767993927, 0.2521008551120758, 0.24089635908603668, 0.3403361439704895, 0.343137264251709, 0.21008403599262238, 0.3123249411582947, 0.3977591097354889, 0.3095238208770752, 0.37535014748573303, 0.4089635908603668, 0.4481792747974396, 0.4327731132507324, 0.4145658314228058, 0.42296919226646423, 0.44117647409439087, 0.4355742335319519, 0.4593837559223175, 0.46078431606292725, 0.4145658314228058, 0.43697479367256165, 0.49439775943756104, 0.4929971992969513, 0.5350140333175659, NaN, 0.5042017102241516, 0.5070028305053711, 0.5546218752861023, 0.456582635641098, 0.5028011202812195, 0.5574229955673218, 0.5098039507865906, 0.5826330780982971, 0.5154061913490295, 0.5826330780982971, 0.5448179244995117, 0.5714285969734192, 0.5350140333175659, 0.5672268867492676, 0.6526610851287842, 0.6302521228790283, 0.6778711676597595, 0.6778711676597595, 0.680672287940979, 0.7002801299095154, 0.848739504814148, 0.8767507076263428, 0.6946778893470764, 0.6890756487846375, 0.7352941036224365, 0.6834734082221985, 0.8529411554336548, 0.8501400351524353, 0.7478991746902466, 0.7983193397521973, 0.8067227005958557, 0.848739504814148, 0.9019607901573181, 0.848739504814148, 0.8557422757148743, 0.9747899174690247], [0.10644257813692093, 0.13165266811847687, 0.13725490868091583, 0.17927171289920807, 0.14005602896213531, 0.1764705926179886, 0.18767507374286652, 0.190476194024086, 0.22969187796115875, 0.26050421595573425, 0.2885154187679291, 0.2661064565181732, 0.2142857164144516, 0.30812326073646545, 0.2787114977836609, 0.32492998242378235, 0.3165266215801239, 0.36974790692329407, 0.32492998242378235, 0.41036415100097656, 0.38235294818878174, 0.3501400649547577, 0.3445378243923187, 0.4145658314228058, 0.40616247057914734, 0.3837535083293915, 0.4453781545162201, 0.4005602300167084, 0.42016807198524475, 0.42717087268829346, 0.45378151535987854, 0.47058823704719543, 0.4593837559223175, 0.45378151535987854, 0.5378151535987854, 0.48459383845329285, 0.4957983195781708, NaN, 0.5126050710678101, 0.5518207550048828, 0.49859943985939026, 0.5042017102241516, 0.5742297172546387, 0.5266106724739075, 0.5350140333175659, 0.5322129130363464, 0.529411792755127, 0.5434173941612244, 0.5826330780982971, 0.5042017102241516, 0.656862735748291, 0.6330532431602478, 0.6358543634414673, 0.6498599648475647, 0.6946778893470764, 0.680672287940979, 0.7114846110343933, 0.6946778893470764, 0.7002801299095154, 0.7142857313156128, 0.7521008253097534, 0.7591036558151245, 0.8123249411582947, 0.7450980544090271, 0.761904776096344, 0.767507016658783, 0.7647058963775635, 0.843137264251709, 0.8039215803146362, 0.8067227005958557, 0.8879551887512207, 0.843137264251709, 0.9691876769065857], [0.10924369841814041, 0.15546219050884247, 0.1596638709306717, 0.1764705926179886, 0.16526611149311066, 0.2072829157114029, 0.19607843458652496, 0.21008403599262238, 0.24929971992969513, 0.24369747936725616, 0.32212886214256287, 0.27731093764305115, 0.28431373834609985, 0.23529411852359772, 0.28151261806488037, 0.3501400649547577, 0.3361344635486603, 0.29411765933036804, 0.3445378243923187, 0.37254902720451355, 0.3473389446735382, 0.4145658314228058, 0.41036415100097656, 0.3921568691730499, 0.4089635908603668, 0.41736695170402527, 0.4089635908603668, 0.4285714328289032, 0.40616247057914734, 0.4257703125476837, 0.45098039507865906, 0.35854342579841614, 0.43977591395378113, 0.456582635641098, 0.4075630307197571, 0.5420168042182922, 0.4929971992969513, 0.48739495873451233, NaN, 0.5602241158485413, 0.4957983195781708, 0.5350140333175659, 0.5546218752861023, 0.49859943985939026, 0.5322129130363464, 0.5266106724739075, 0.4957983195781708, 0.529411792755127, 0.5658263564109802, 0.5322129130363464, 0.6106442809104919, 0.6722689270973206, 0.6386554837226868, 0.6442577242851257, 0.7226890921592712, 0.7338935732841492, 0.6834734082221985, 0.6918767690658569, 0.7002801299095154, 0.7521008253097534, 0.6834734082221985, 0.7030812501907349, 0.7254902124404907, 0.7815126180648804, 0.7927170991897583, 0.7983193397521973, 0.7226890921592712, 0.8627451062202454, 0.83753502368927, 0.8263305425643921, 0.843137264251709, 0.8879551887512207, 0.9691876769065857], [0.2212885171175003, 0.2647058963775635, 0.24929971992969513, 0.31092438101768494, 0.2717086970806122, 0.3137255012989044, 0.31092438101768494, 0.3193277418613434, 0.3305322229862213, 0.3235294222831726, 0.30812326073646545, 0.3641456663608551, 0.37535014748573303, 0.3305322229862213, 0.3613445460796356, 0.38655462861061096, 0.4019607901573181, 0.3879551887512207, 0.3837535083293915, 0.42016807198524475, 0.4117647111415863, 0.3977591097354889, 0.45098039507865906, 0.42016807198524475, 0.4551820755004883, 0.43417367339134216, 0.40336135029792786, 0.4481792747974396, 0.4019607901573181, 0.4145658314228058, 0.41736695170402527, 0.4313725531101227, 0.4117647111415863, 0.4481792747974396, 0.4481792747974396, 0.40336135029792786, 0.4453781545162201, 0.4481792747974396, 0.43977591395378113, NaN, 0.42016807198524475, 0.5014005899429321, 0.5980392098426819, 0.462184876203537, 0.46498599648475647, 0.43697479367256165, 0.45798319578170776, 0.4453781545162201, 0.456582635641098, 0.5406162738800049, 0.5602241158485413, 0.6274510025978088, 0.913165271282196, 0.6232492923736572, 0.48179271817207336, 0.6330532431602478, 0.6218487620353699, 0.5770308375358582, 0.5602241158485413, 0.6582633256912231, 0.5658263564109802, 0.6302521228790283, 0.680672287940979, 0.6918767690658569, 0.7254902124404907, 0.686274528503418, 0.7198879718780518, 0.7871148586273193, 0.8095238208770752, 0.8039215803146362, 0.686274528503418, 0.848739504814148, 0.9411764740943909], [0.07002801448106766, 0.09803921729326248, 0.12044817954301834, 0.11764705926179886, 0.10364145785570145, 0.10364145785570145, 0.1330532282590866, 0.15266107022762299, 0.23249299824237823, 0.190476194024086, 0.3137255012989044, 0.25630253553390503, 0.22969187796115875, 0.2591036558151245, 0.2212885171175003, 0.31092438101768494, 0.32492998242378235, 0.35154062509536743, 0.26050421595573425, 0.34173670411109924, 0.4089635908603668, 0.38655462861061096, 0.4117647111415863, 0.3851540684700012, 0.4005602300167084, 0.380952388048172, 0.38655462861061096, 0.40616247057914734, 0.44117647409439087, 0.42296919226646423, 0.43697479367256165, 0.4733893573284149, 0.47478991746902466, 0.48459383845329285, 0.5098039507865906, 0.5154061913490295, 0.5434173941612244, 0.5014005899429321, 0.5042017102241516, 0.5798319578170776, NaN, 0.49859943985939026, 0.5770308375358582, 0.48879551887512207, 0.5896358489990234, 0.5826330780982971, 0.5686274766921997, 0.686274528503418, 0.5630252361297607, 0.5546218752861023, 0.5882353186607361, 0.6274510025978088, 0.6442577242851257, 0.6414566040039062, 0.6890756487846375, 0.6498599648475647, 0.7310924530029297, 0.7703081369400024, 0.7857142686843872, 0.6666666865348816, 0.6918767690658569, 0.7591036558151245, 0.6638655662536621, 0.756302535533905, 0.767507016658783, 0.7142857313156128, 0.7507002949714661, 0.7703081369400024, 0.8011204600334167, 0.8095238208770752, 0.8459383845329285, 0.831932783126831, 0.9649859666824341], [0.07002801448106766, 0.12044817954301834, 0.09243697673082352, 0.11624649912118912, 0.10644257813692093, 0.18207283318042755, 0.14845938980579376, 0.15406163036823273, 0.17086835205554962, 0.24229691922664642, 0.190476194024086, 0.3711484670639038, 0.35574230551719666, 0.3473389446735382, 0.37535014748573303, 0.26330533623695374, 0.24929971992969513, 0.3949579894542694, 0.4089635908603668, 0.28011205792427063, 0.41736695170402527, 0.44397759437561035, 0.42016807198524475, 0.43417367339134216, 0.4257703125476837, 0.462184876203537, 0.4145658314228058, 0.4551820755004883, 0.48739495873451233, 0.46778711676597595, 0.4929971992969513, 0.43417367339134216, 0.4957983195781708, 0.5154061913490295, 0.4719887971878052, 0.48739495873451233, 0.4971988797187805, 0.4957983195781708, 0.46498599648475647, 0.49859943985939026, 0.5014005899429321, NaN, 0.5098039507865906, 0.5490196347236633, 0.529411792755127, 0.5126050710678101, 0.6022409200668335, 0.5210084319114685, 0.5798319578170776, 0.6022409200668335, 0.6246498823165894, 0.5966386795043945, 0.6330532431602478, 0.599439799785614, 0.6302521228790283, 0.6918767690658569, 0.680672287940979, 0.6358543634414673, 0.6778711676597595, 0.7212885022163391, 0.6974790096282959, 0.7002801299095154, 0.7366946935653687, 0.6638655662536621, 0.7030812501907349, 0.7478991746902466, 0.7759103775024414, 0.7310924530029297, 0.761904776096344, 0.7282913327217102, 0.8347339034080505, 0.7787114977836609, 0.9663865566253662], [0.21848739683628082, 0.24929971992969513, 0.2521008551120758, 0.30812326073646545, 0.26750701665878296, 0.299719899892807, 0.3025210201740265, 0.30812326073646545, 0.3193277418613434, 0.32773110270500183, 0.29411765933036804, 0.3529411852359772, 0.3613445460796356, 0.32492998242378235, 0.35574230551719666, 0.3837535083293915, 0.40336135029792786, 0.3837535083293915, 0.37955182790756226, 0.4145658314228058, 0.40616247057914734, 0.3837535083293915, 0.456582635641098, 0.41736695170402527, 0.456582635641098, 0.4257703125476837, 0.3949579894542694, 0.44397759437561035, 0.3949579894542694, 0.40616247057914734, 0.42296919226646423, 0.4313725531101227, 0.40616247057914734, 0.4383753538131714, 0.44117647409439087, 0.3949579894542694, 0.4425770342350006, 0.4257703125476837, 0.4453781545162201, 0.4019607901573181, 0.42296919226646423, 0.4901960790157318, NaN, 0.47058823704719543, 0.45798319578170776, 0.43977591395378113, 0.46778711676597595, 0.4425770342350006, 0.45098039507865906, 0.5350140333175659, 0.5490196347236633, 0.6022409200668335, 0.9355742335319519, 0.599439799785614, 0.48739495873451233, 0.6274510025978088, 0.6190476417541504, 0.5658263564109802, 0.5602241158485413, 0.6582633256912231, 0.5630252361297607, 0.6246498823165894, 0.680672287940979, 0.6834734082221985, 0.7170868515968323, 0.6834734082221985, 0.7114846110343933, 0.7731092572212219, 0.8011204600334167, 0.7983193397521973, 0.6834734082221985, 0.8347339034080505, 0.9411764740943909], [0.08403361588716507, 0.13865546882152557, 0.12885154783725739, 0.16526611149311066, 0.12885154783725739, 0.17927171289920807, 0.1568627506494522, 0.1764705926179886, 0.18767507374286652, 0.2478991597890854, 0.23249299824237823, 0.2661064565181732, 0.2731092572212219, 0.2969187796115875, 0.3025210201740265, 0.27450981736183167, 0.29411765933036804, 0.3613445460796356, 0.3333333432674408, 0.32212886214256287, 0.4117647111415863, 0.3473389446735382, 0.3445378243923187, 0.3641456663608551, 0.36974790692329407, 0.380952388048172, 0.4313725531101227, 0.3739495873451233, 0.4257703125476837, 0.3445378243923187, 0.4453781545162201, 0.4803921580314636, 0.44117647409439087, 0.39355742931365967, 0.5, 0.4425770342350006, 0.4901960790157318, 0.4733893573284149, 0.5014005899429321, 0.5378151535987854, 0.5112044811248779, 0.45098039507865906, 0.529411792755127, NaN, 0.48739495873451233, 0.5308123230934143, 0.5126050710678101, 0.5434173941612244, 0.5434173941612244, 0.523809552192688, 0.6246498823165894, 0.6106442809104919, 0.6106442809104919, 0.6022409200668335, 0.6526610851287842, 0.6778711676597595, 0.6834734082221985, 0.6386554837226868, 0.6386554837226868, 0.6960784196853638, 0.7240896224975586, 0.7282913327217102, 0.8095238208770752, 0.7184873819351196, 0.7226890921592712, 0.7170868515968323, 0.7198879718780518, 0.8067227005958557, 0.7591036558151245, 0.7731092572212219, 0.8403361439704895, 0.831932783126831, 0.9579831957817078], [0.04481792822480202, 0.10084033757448196, 0.12044817954301834, 0.12885154783725739, 0.12044817954301834, 0.14005602896213531, 0.13865546882152557, 0.16806723177433014, 0.23249299824237823, 0.19887955486774445, 0.2885154187679291, 0.2380952388048172, 0.20168067514896393, 0.23249299824237823, 0.19887955486774445, 0.2969187796115875, 0.32773110270500183, 0.18487395346164703, 0.29411765933036804, 0.37254902720451355, 0.25770309567451477, 0.33753502368927, 0.3921568691730499, 0.4159663915634155, 0.4117647111415863, 0.39635854959487915, 0.37535014748573303, 0.3921568691730499, 0.41316527128219604, 0.42296919226646423, 0.41316527128219604, 0.4145658314228058, 0.40616247057914734, 0.47058823704719543, 0.46498599648475647, 0.4901960790157318, 0.41736695170402527, 0.46498599648475647, 0.46778711676597595, 0.5350140333175659, 0.41036415100097656, 0.47058823704719543, 0.5420168042182922, 0.5126050710678101, NaN, 0.4761904776096344, 0.5630252361297607, 0.4929971992969513, 0.5406162738800049, 0.5266106724739075, 0.5518207550048828, 0.6540616154670715, 0.6246498823165894, 0.6694678068161011, 0.6722689270973206, 0.6610644459724426, 0.6610644459724426, 0.7927170991897583, 0.8207283020019531, 0.6918767690658569, 0.6554622054100037, 0.7198879718780518, 0.6778711676597595, 0.8263305425643921, 0.8347339034080505, 0.7366946935653687, 0.7871148586273193, 0.8081232309341431, 0.8459383845329285, 0.8767507076263428, 0.843137264251709, 0.8515406250953674, 0.9691876769065857], [0.05602240934967995, 0.08963585644960403, 0.09803921729326248, 0.10924369841814041, 0.09803921729326248, 0.095238097012043, 0.11764705926179886, 0.13165266811847687, 0.21008403599262238, 0.16806723177433014, 0.27450981736183167, 0.2380952388048172, 0.2142857164144516, 0.23249299824237823, 0.1736694723367691, 0.28011205792427063, 0.30532214045524597, 0.3165266215801239, 0.26330533623695374, 0.3501400649547577, 0.380952388048172, 0.34173670411109924, 0.3837535083293915, 0.3641456663608551, 0.3837535083293915, 0.3767507076263428, 0.3669467866420746, 0.40616247057914734, 0.3949579894542694, 0.40336135029792786, 0.4117647111415863, 0.43977591395378113, 0.4355742335319519, 0.43977591395378113, 0.47478991746902466, 0.48459383845329285, 0.48459383845329285, 0.46778711676597595, 0.4733893573284149, 0.5630252361297607, 0.41736695170402527, 0.48739495873451233, 0.5602241158485413, 0.4691876769065857, 0.523809552192688, NaN, 0.5224089622497559, 0.5966386795043945, 0.523809552192688, 0.5266106724739075, 0.5546218752861023, 0.593837559223175, 0.6358543634414673, 0.6190476417541504, 0.6470588445663452, 0.6190476417541504, 0.668067216873169, 0.732492983341217, 0.7310924530029297, 0.6274510025978088, 0.6778711676597595, 0.7016806602478027, 0.6470588445663452, 0.7254902124404907, 0.7282913327217102, 0.6890756487846375, 0.7198879718780518, 0.7394958138465881, 0.7899159789085388, 0.7759103775024414, 0.8361344337463379, 0.7955182194709778, 0.9495798349380493], [0.06442577391862869, 0.08263305574655533, 0.1120448186993599, 0.10364145785570145, 0.09383753687143326, 0.1120448186993599, 0.1260504275560379, 0.13445378839969635, 0.16946779191493988, 0.1736694723367691, 0.2044817954301834, 0.21568627655506134, 0.22689075767993927, 0.2969187796115875, 0.24649859964847565, 0.2240896373987198, 0.24929971992969513, 0.299719899892807, 0.32492998242378235, 0.32492998242378235, 0.3361344635486603, 0.35574230551719666, 0.3879551887512207, 0.3669467866420746, 0.3907563090324402, 0.3669467866420746, 0.4075630307197571, 0.28151261806488037, 0.3501400649547577, 0.4313725531101227, 0.3529411852359772, 0.43977591395378113, 0.38935574889183044, 0.44397759437561035, 0.4733893573284149, 0.4355742335319519, 0.41736695170402527, 0.47058823704719543, 0.5042017102241516, 0.5420168042182922, 0.4313725531101227, 0.3977591097354889, 0.5322129130363464, 0.48739495873451233, 0.43697479367256165, 0.47759103775024414, NaN, 0.48459383845329285, 0.5434173941612244, 0.49859943985939026, 0.5420168042182922, 0.5686274766921997, 0.6078431606292725, 0.5854341983795166, 0.6456582546234131, 0.575630247592926, 0.6274510025978088, 0.6218487620353699, 0.6218487620353699, 0.6106442809104919, 0.7282913327217102, 0.6890756487846375, 0.7226890921592712, 0.686274528503418, 0.6974790096282959, 0.6666666865348816, 0.7647058963775635, 0.7366946935653687, 0.7254902124404907, 0.7535014152526855, 0.7955182194709778, 0.7913165092468262, 0.9495798349380493], [0.05042016878724098, 0.10084033757448196, 0.10924369841814041, 0.1120448186993599, 0.095238097012043, 0.09243697673082352, 0.11764705926179886, 0.12885154783725739, 0.190476194024086, 0.1596638709306717, 0.22689075767993927, 0.2240896373987198, 0.21288515627384186, 0.21568627655506134, 0.16806723177433014, 0.2549019753932953, 0.27731093764305115, 0.21568627655506134, 0.22549019753932953, 0.3361344635486603, 0.2661064565181732, 0.31092438101768494, 0.35854342579841614, 0.37254902720451355, 0.38935574889183044, 0.34173670411109924, 0.3613445460796356, 0.3571428656578064, 0.35574230551719666, 0.4145658314228058, 0.37535014748573303, 0.3949579894542694, 0.38655462861061096, 0.4257703125476837, 0.43697479367256165, 0.43977591395378113, 0.4551820755004883, 0.456582635641098, 0.47058823704719543, 0.5546218752861023, 0.3137255012989044, 0.4789915978908539, 0.5574229955673218, 0.456582635641098, 0.5070028305053711, 0.40336135029792786, 0.5154061913490295, NaN, 0.5098039507865906, 0.523809552192688, 0.5490196347236633, 0.605042040348053, 0.6246498823165894, 0.6106442809104919, 0.6498599648475647, 0.6358543634414673, 0.6386554837226868, 0.7212885022163391, 0.7268907427787781, 0.6554622054100037, 0.6554622054100037, 0.6610644459724426, 0.6330532431602478, 0.7114846110343933, 0.7282913327217102, 0.7086834907531738, 0.7226890921592712, 0.732492983341217, 0.756302535533905, 0.7647058963775635, 0.7927170991897583, 0.8179271817207336, 0.9579831957817078], [0.03921568766236305, 0.07282913476228714, 0.0784313753247261, 0.09243697673082352, 0.07563025504350662, 0.095238097012043, 0.10364145785570145, 0.10364145785570145, 0.18487395346164703, 0.18767507374286652, 0.23529411852359772, 0.2044817954301834, 0.24369747936725616, 0.2521008551120758, 0.2310924381017685, 0.26050421595573425, 0.27450981736183167, 0.3137255012989044, 0.28711485862731934, 0.30812326073646545, 0.33473390340805054, 0.3165266215801239, 0.31092438101768494, 0.3613445460796356, 0.380952388048172, 0.3025210201740265, 0.3837535083293915, 0.3501400649547577, 0.35574230551719666, 0.3949579894542694, 0.3669467866420746, 0.418767511844635, 0.3767507076263428, 0.2927170991897583, 0.4425770342350006, 0.40336135029792786, 0.4285714328289032, 0.41736695170402527, 0.43417367339134216, 0.5434173941612244, 0.43697479367256165, 0.42016807198524475, 0.5490196347236633, 0.456582635641098, 0.4593837559223175, 0.4761904776096344, 0.456582635641098, 0.4901960790157318, NaN, 0.48739495873451233, 0.6106442809104919, 0.5980392098426819, 0.6246498823165894, 0.5742297172546387, 0.6386554837226868, 0.6134454011917114, 0.6386554837226868, 0.6106442809104919, 0.6092436909675598, 0.6316526532173157, 0.743697464466095, 0.680672287940979, 0.7142857313156128, 0.680672287940979, 0.6918767690658569, 0.7282913327217102, 0.7338935732841492, 0.7605041861534119, 0.7226890921592712, 0.7310924530029297, 0.8123249411582947, 0.767507016658783, 0.9523809552192688], [0.04481792822480202, 0.08123249560594559, 0.08683473616838455, 0.07282913476228714, 0.1260504275560379, 0.07002801448106766, 0.1120448186993599, 0.12184873968362808, 0.2380952388048172, 0.16246499121189117, 0.2661064565181732, 0.2857142984867096, 0.29411765933036804, 0.3137255012989044, 0.26330533623695374, 0.2380952388048172, 0.261904776096344, 0.34173670411109924, 0.30812326073646545, 0.3473389446735382, 0.37535014748573303, 0.38935574889183044, 0.3837535083293915, 0.4313725531101227, 0.36974790692329407, 0.38655462861061096, 0.46498599648475647, 0.38655462861061096, 0.4145658314228058, 0.42016807198524475, 0.4089635908603668, 0.45378151535987854, 0.4089635908603668, 0.4285714328289032, 0.4859943985939026, 0.4285714328289032, 0.46498599648475647, 0.4957983195781708, 0.46778711676597595, 0.4593837559223175, 0.4453781545162201, 0.3977591097354889, 0.46498599648475647, 0.4761904776096344, 0.4733893573284149, 0.4733893573284149, 0.5014005899429321, 0.4761904776096344, 0.5126050710678101, NaN, 0.5266106724739075, 0.5210084319114685, 0.5266106724739075, 0.5490196347236633, 0.5826330780982971, 0.5322129130363464, 0.5966386795043945, 0.6554622054100037, 0.6386554837226868, 0.5350140333175659, 0.605042040348053, 0.6022409200668335, 0.6330532431602478, 0.6078431606292725, 0.6134454011917114, 0.5826330780982971, 0.7717086672782898, 0.6638655662536621, 0.7170868515968323, 0.7142857313156128, 0.7226890921592712, 0.7030812501907349, 0.9159663915634155], [0.08963585644960403, 0.11764705926179886, 0.1260504275560379, 0.14565826952457428, 0.1190476194024086, 0.14845938980579376, 0.15406163036823273, 0.16806723177433014, 0.12885154783725739, 0.26330533623695374, 0.13725490868091583, 0.26050421595573425, 0.2885154187679291, 0.28431373834609985, 0.2857142984867096, 0.24089635908603668, 0.24649859964847565, 0.35854342579841614, 0.30532214045524597, 0.2240896373987198, 0.37535014748573303, 0.3123249411582947, 0.3025210201740265, 0.27731093764305115, 0.3025210201740265, 0.32492998242378235, 0.3305322229862213, 0.32492998242378235, 0.41736695170402527, 0.3137255012989044, 0.43697479367256165, 0.3669467866420746, 0.4481792747974396, 0.3669467866420746, 0.4047619104385376, 0.4313725531101227, 0.4327731132507324, 0.343137264251709, 0.38935574889183044, 0.43977591395378113, 0.4117647111415863, 0.37535014748573303, 0.45098039507865906, 0.37535014748573303, 0.4481792747974396, 0.4453781545162201, 0.45798319578170776, 0.45098039507865906, 0.38935574889183044, 0.4733893573284149, NaN, 0.518207311630249, 0.5686274766921997, 0.518207311630249, 0.5518207550048828, 0.5658263564109802, 0.593837559223175, 0.5574229955673218, 0.5574229955673218, 0.6064425706863403, 0.575630247592926, 0.6274510025978088, 0.6190476417541504, 0.6078431606292725, 0.6246498823165894, 0.6358543634414673, 0.6414566040039062, 0.686274528503418, 0.6610644459724426, 0.6974790096282959, 0.7058823704719543, 0.7366946935653687, 0.918767511844635], [0.05602240934967995, 0.06722689419984818, 0.10504201799631119, 0.09243697673082352, 0.1120448186993599, 0.10084033757448196, 0.13445378839969635, 0.13865546882152557, 0.13025210797786713, 0.1764705926179886, 0.1736694723367691, 0.24369747936725616, 0.19887955486774445, 0.18207283318042755, 0.2072829157114029, 0.19607843458652496, 0.2380952388048172, 0.2310924381017685, 0.2717086970806122, 0.22689075767993927, 0.25630253553390503, 0.29131653904914856, 0.32492998242378235, 0.2829131782054901, 0.30812326073646545, 0.3473389446735382, 0.3193277418613434, 0.3207283020019531, 0.2717086970806122, 0.33893558382987976, 0.28011205792427063, 0.2899159789085388, 0.299719899892807, 0.3529411852359772, 0.3361344635486603, 0.32773110270500183, 0.3473389446735382, 0.3669467866420746, 0.32773110270500183, 0.37254902720451355, 0.37254902720451355, 0.40336135029792786, 0.3977591097354889, 0.38935574889183044, 0.34593838453292847, 0.40616247057914734, 0.4313725531101227, 0.3949579894542694, 0.4019607901573181, 0.4789915978908539, 0.48179271817207336, NaN, 0.5728291273117065, 0.6106442809104919, 0.49859943985939026, 0.5784313678741455, 0.5434173941612244, 0.5070028305053711, 0.5126050710678101, 0.5980392098426819, 0.5490196347236633, 0.5574229955673218, 0.599439799785614, 0.7156862616539001, 0.7450980544090271, 0.6610644459724426, 0.6638655662536621, 0.8011204600334167, 0.848739504814148, 0.8039215803146362, 0.6638655662536621, 0.831932783126831, 0.9859943985939026], [0.17927171289920807, 0.2072829157114029, 0.20168067514896393, 0.24369747936725616, 0.22969187796115875, 0.25770309567451477, 0.25770309567451477, 0.2549019753932953, 0.25770309567451477, 0.26750701665878296, 0.24089635908603668, 0.29411765933036804, 0.30812326073646545, 0.2591036558151245, 0.3025210201740265, 0.29411765933036804, 0.299719899892807, 0.29411765933036804, 0.32773110270500183, 0.32492998242378235, 0.3165266215801239, 0.32492998242378235, 0.35854342579841614, 0.3445378243923187, 0.35574230551719666, 0.3333333432674408, 0.3193277418613434, 0.3445378243923187, 0.3333333432674408, 0.34173670411109924, 0.3473389446735382, 0.3333333432674408, 0.33893558382987976, 0.34173670411109924, 0.35854342579841614, 0.3165266215801239, 0.36974790692329407, 0.3641456663608551, 0.3613445460796356, 0.08683473616838455, 0.35574230551719666, 0.3669467866420746, 0.06442577391862869, 0.38935574889183044, 0.37535014748573303, 0.3641456663608551, 0.3921568691730499, 0.37535014748573303, 0.37535014748573303, 0.4733893573284149, 0.4313725531101227, 0.42717087268829346, NaN, 0.4313725531101227, 0.42296919226646423, 0.49859943985939026, 0.523809552192688, 0.46778711676597595, 0.48179271817207336, 0.5154061913490295, 0.45378151535987854, 0.5210084319114685, 0.5070028305053711, 0.505602240562439, 0.5350140333175659, 0.5826330780982971, 0.6330532431602478, 0.6666666865348816, 0.6470588445663452, 0.605042040348053, 0.5882353186607361, 0.7394958138465881, 0.9355742335319519], [0.06722689419984818, 0.07282913476228714, 0.10644257813692093, 0.1120448186993599, 0.1120448186993599, 0.11344537883996964, 0.1428571492433548, 0.14005602896213531, 0.10364145785570145, 0.18767507374286652, 0.17927171289920807, 0.23529411852359772, 0.20588235557079315, 0.18767507374286652, 0.23529411852359772, 0.19607843458652496, 0.22969187796115875, 0.2521008551120758, 0.27731093764305115, 0.22689075767993927, 0.26750701665878296, 0.29131653904914856, 0.32492998242378235, 0.2689075767993927, 0.28011205792427063, 0.33893558382987976, 0.3165266215801239, 0.30812326073646545, 0.2857142984867096, 0.3473389446735382, 0.28431373834609985, 0.30812326073646545, 0.3137255012989044, 0.3613445460796356, 0.30532214045524597, 0.3193277418613434, 0.32212886214256287, 0.3501400649547577, 0.35574230551719666, 0.3767507076263428, 0.35854342579841614, 0.4005602300167084, 0.4005602300167084, 0.3977591097354889, 0.3305322229862213, 0.380952388048172, 0.4145658314228058, 0.38935574889183044, 0.4257703125476837, 0.45098039507865906, 0.48179271817207336, 0.38935574889183044, 0.5686274766921997, NaN, 0.5126050710678101, 0.5714285969734192, 0.5154061913490295, 0.4789915978908539, 0.4733893573284149, 0.593837559223175, 0.5518207550048828, 0.5462185144424438, 0.5910364389419556, 0.6610644459724426, 0.7002801299095154, 0.6554622054100037, 0.6330532431602478, 0.756302535533905, 0.7759103775024414, 0.7661064267158508, 0.6414566040039062, 0.83753502368927, 0.9831932783126831], [0.07282913476228714, 0.08403361588716507, 0.09243697673082352, 0.095238097012043, 0.08963585644960403, 0.10924369841814041, 0.10924369841814041, 0.11484593898057938, 0.16806723177433014, 0.14005602896213531, 0.19887955486774445, 0.18207283318042755, 0.15406163036823273, 0.14005602896213531, 0.16806723177433014, 0.1764705926179886, 0.21008403599262238, 0.17927171289920807, 0.190476194024086, 0.2521008551120758, 0.20308123528957367, 0.21288515627384186, 0.28011205792427063, 0.22969187796115875, 0.2661064565181732, 0.2521008551120758, 0.2380952388048172, 0.25770309567451477, 0.19327731430530548, 0.27731093764305115, 0.24229691922664642, 0.24929971992969513, 0.2661064565181732, 0.2759103775024414, 0.29411765933036804, 0.3165266215801239, 0.32212886214256287, 0.30532214045524597, 0.27731093764305115, 0.518207311630249, 0.31092438101768494, 0.36974790692329407, 0.5126050710678101, 0.3473389446735382, 0.32773110270500183, 0.3529411852359772, 0.3543417453765869, 0.3501400649547577, 0.3613445460796356, 0.41736695170402527, 0.4481792747974396, 0.5014005899429321, 0.5770308375358582, 0.48739495873451233, NaN, 0.5686274766921997, 0.47058823704719543, 0.4761904776096344, 0.47058823704719543, 0.5966386795043945, 0.5350140333175659, 0.4901960790157318, 0.5490196347236633, 0.6302521228790283, 0.6526610851287842, 0.6526610851287842, 0.6134454011917114, 0.7338935732841492, 0.6694678068161011, 0.7058823704719543, 0.6106442809104919, 0.8067227005958557, 0.9215686321258545], [0.08683473616838455, 0.12885154783725739, 0.10924369841814041, 0.15406163036823273, 0.14005602896213531, 0.15406163036823273, 0.14005602896213531, 0.1596638709306717, 0.16806723177433014, 0.190476194024086, 0.18767507374286652, 0.2002801150083542, 0.21288515627384186, 0.14565826952457428, 0.18767507374286652, 0.24369747936725616, 0.23949579894542694, 0.21848739683628082, 0.21568627655506134, 0.25770309567451477, 0.23249299824237823, 0.29411765933036804, 0.2689075767993927, 0.24089635908603668, 0.27731093764305115, 0.3137255012989044, 0.2661064565181732, 0.34173670411109924, 0.2969187796115875, 0.27450981736183167, 0.3025210201740265, 0.25770309567451477, 0.30812326073646545, 0.3095238208770752, 0.2857142984867096, 0.2969187796115875, 0.3193277418613434, 0.3193277418613434, 0.2661064565181732, 0.3669467866420746, 0.3501400649547577, 0.30812326073646545, 0.37254902720451355, 0.32212886214256287, 0.33893558382987976, 0.380952388048172, 0.424369752407074, 0.3641456663608551, 0.38655462861061096, 0.46778711676597595, 0.43417367339134216, 0.4215686321258545, 0.5014005899429321, 0.4285714328289032, 0.4313725531101227, NaN, 0.4789915978908539, 0.4593837559223175, 0.43977591395378113, 0.6386554837226868, 0.4761904776096344, 0.5014005899429321, 0.5014005899429321, 0.4789915978908539, 0.4901960790157318, 0.7198879718780518, 0.5602241158485413, 0.6246498823165894, 0.5910364389419556, 0.5546218752861023, 0.593837559223175, 0.7002801299095154, 0.8935574293136597], [0.02240896411240101, 0.0784313753247261, 0.07002801448106766, 0.07563025504350662, 0.0784313753247261, 0.08123249560594559, 0.07563025504350662, 0.09943977743387222, 0.13725490868091583, 0.1260504275560379, 0.15406163036823273, 0.095238097012043, 0.10084033757448196, 0.1764705926179886, 0.13165266811847687, 0.17086835205554962, 0.190476194024086, 0.14845938980579376, 0.13725490868091583, 0.22969187796115875, 0.16946779191493988, 0.17927171289920807, 0.23529411852359772, 0.261904776096344, 0.2240896373987198, 0.2549019753932953, 0.2689075767993927, 0.19607843458652496, 0.26050421595573425, 0.22969187796115875, 0.24649859964847565, 0.2689075767993927, 0.26330533623695374, 0.24929971992969513, 0.29411765933036804, 0.31792718172073364, 0.299719899892807, 0.2885154187679291, 0.3165266215801239, 0.3781512677669525, 0.2689075767993927, 0.3193277418613434, 0.380952388048172, 0.3165266215801239, 0.33893558382987976, 0.33193278312683105, 0.37254902720451355, 0.3613445460796356, 0.3613445460796356, 0.40336135029792786, 0.40616247057914734, 0.456582635641098, 0.4761904776096344, 0.48459383845329285, 0.529411792755127, 0.5210084319114685, NaN, 0.5602241158485413, 0.5602241158485413, 0.5350140333175659, 0.5070028305053711, 0.651260495185852, 0.4957983195781708, 0.5574229955673218, 0.5826330780982971, 0.605042040348053, 0.6442577242851257, 0.6596638560295105, 0.6834734082221985, 0.7030812501907349, 0.7647058963775635, 0.7226890921592712, 0.9411764740943909], [0.030812324956059456, 0.05042016878724098, 0.07282913476228714, 0.07282913476228714, 0.07563025504350662, 0.05602240934967995, 0.0784313753247261, 0.08403361588716507, 0.13165266811847687, 0.12044817954301834, 0.18207283318042755, 0.10644257813692093, 0.11764705926179886, 0.16526611149311066, 0.12044817954301834, 0.17927171289920807, 0.21288515627384186, 0.11484593898057938, 0.1736694723367691, 0.26330533623695374, 0.15826331079006195, 0.21848739683628082, 0.27450981736183167, 0.299719899892807, 0.23249299824237823, 0.26330533623695374, 0.3137255012989044, 0.2044817954301834, 0.19607843458652496, 0.29131653904914856, 0.18487395346164703, 0.2478991597890854, 0.16806723177433014, 0.27450981736183167, 0.2717086970806122, 0.3501400649547577, 0.15126051008701324, 0.30532214045524597, 0.30812326073646545, 0.42296919226646423, 0.22969187796115875, 0.3641456663608551, 0.43417367339134216, 0.3613445460796356, 0.2072829157114029, 0.26750701665878296, 0.3781512677669525, 0.2787114977836609, 0.38935574889183044, 0.3445378243923187, 0.4425770342350006, 0.4929971992969513, 0.5322129130363464, 0.5210084319114685, 0.523809552192688, 0.5406162738800049, 0.43977591395378113, NaN, 0.505602240562439, 0.5546218752861023, 0.5266106724739075, 0.4957983195781708, 0.5518207550048828, 0.5798319578170776, 0.6134454011917114, 0.6274510025978088, 0.680672287940979, 0.680672287940979, 0.7198879718780518, 0.7394958138465881, 0.6414566040039062, 0.7366946935653687, 0.924369752407074], [0.03361344709992409, 0.053221289068460464, 0.05882352963089943, 0.06442577391862869, 0.042016807943582535, 0.05042016878724098, 0.06162464991211891, 0.06722689419984818, 0.11764705926179886, 0.09803921729326248, 0.16386555135250092, 0.08123249560594559, 0.10924369841814041, 0.14425770938396454, 0.11624649912118912, 0.15126051008701324, 0.19607843458652496, 0.05602240934967995, 0.17086835205554962, 0.26050421595573425, 0.09663865715265274, 0.19607843458652496, 0.2717086970806122, 0.2549019753932953, 0.21288515627384186, 0.2521008551120758, 0.28011205792427063, 0.15406163036823273, 0.16526611149311066, 0.2899159789085388, 0.1568627506494522, 0.23529411852359772, 0.1764705926179886, 0.3025210201740265, 0.2549019753932953, 0.34173670411109924, 0.12324929982423782, 0.299719899892807, 0.299719899892807, 0.43977591395378113, 0.2142857164144516, 0.32212886214256287, 0.43977591395378113, 0.3613445460796356, 0.17927171289920807, 0.2689075767993927, 0.3781512677669525, 0.2731092572212219, 0.3907563090324402, 0.3613445460796356, 0.4425770342350006, 0.48739495873451233, 0.518207311630249, 0.5266106724739075, 0.529411792755127, 0.5602241158485413, 0.43977591395378113, 0.49439775943756104, NaN, 0.5714285969734192, 0.5574229955673218, 0.5042017102241516, 0.5490196347236633, 0.5854341983795166, 0.6148459315299988, 0.6330532431602478, 0.6890756487846375, 0.680672287940979, 0.7086834907531738, 0.7240896224975586, 0.6722689270973206, 0.7478991746902466, 0.9481792449951172], [0.07563025504350662, 0.11764705926179886, 0.09803921729326248, 0.12044817954301834, 0.1260504275560379, 0.12324929982423782, 0.11764705926179886, 0.14005602896213531, 0.14845938980579376, 0.16246499121189117, 0.16526611149311066, 0.1666666716337204, 0.2072829157114029, 0.11764705926179886, 0.17226891219615936, 0.21848739683628082, 0.21568627655506134, 0.2044817954301834, 0.19887955486774445, 0.24089635908603668, 0.2142857164144516, 0.28431373834609985, 0.26330533623695374, 0.21288515627384186, 0.25770309567451477, 0.29131653904914856, 0.2478991597890854, 0.3137255012989044, 0.27450981736183167, 0.24929971992969513, 0.28011205792427063, 0.24649859964847565, 0.2927170991897583, 0.2717086970806122, 0.2717086970806122, 0.2759103775024414, 0.30532214045524597, 0.2857142984867096, 0.2478991597890854, 0.34173670411109924, 0.3333333432674408, 0.2787114977836609, 0.34173670411109924, 0.30392158031463623, 0.30812326073646545, 0.37254902720451355, 0.38935574889183044, 0.3445378243923187, 0.3683473467826843, 0.46498599648475647, 0.39355742931365967, 0.4019607901573181, 0.48459383845329285, 0.40616247057914734, 0.40336135029792786, 0.3613445460796356, 0.46498599648475647, 0.4453781545162201, 0.4285714328289032, NaN, 0.456582635641098, 0.48739495873451233, 0.48459383845329285, 0.4733893573284149, 0.4761904776096344, 0.6330532431602478, 0.5574229955673218, 0.5784313678741455, 0.5686274766921997, 0.5406162738800049, 0.5910364389419556, 0.6778711676597595, 0.8823529481887817], [0.028011204674839973, 0.05602240934967995, 0.0476190485060215, 0.06442577391862869, 0.03641456738114357, 0.06162464991211891, 0.08123249560594559, 0.07282913476228714, 0.09803921729326248, 0.14005602896213531, 0.09243697673082352, 0.13725490868091583, 0.12885154783725739, 0.1736694723367691, 0.12885154783725739, 0.14005602896213531, 0.15126051008701324, 0.22689075767993927, 0.14845938980579376, 0.16806723177433014, 0.2380952388048172, 0.1736694723367691, 0.23249299824237823, 0.17086835205554962, 0.18767507374286652, 0.2521008551120758, 0.19327731430530548, 0.2240896373987198, 0.24089635908603668, 0.1834733933210373, 0.26330533623695374, 0.3305322229862213, 0.2829131782054901, 0.16246499121189117, 0.3501400649547577, 0.2717086970806122, 0.31092438101768494, 0.2478991597890854, 0.3165266215801239, 0.43417367339134216, 0.30812326073646545, 0.3025210201740265, 0.43697479367256165, 0.2759103775024414, 0.3445378243923187, 0.32212886214256287, 0.2717086970806122, 0.3445378243923187, 0.25630253553390503, 0.3949579894542694, 0.424369752407074, 0.45098039507865906, 0.5462185144424438, 0.4481792747974396, 0.46498599648475647, 0.523809552192688, 0.4929971992969513, 0.4733893573284149, 0.4425770342350006, 0.5434173941612244, NaN, 0.5350140333175659, 0.593837559223175, 0.5266106724739075, 0.5658263564109802, 0.6134454011917114, 0.6022409200668335, 0.680672287940979, 0.6218487620353699, 0.6162465214729309, 0.67507004737854, 0.7310924530029297, 0.9271708726882935], [0.019607843831181526, 0.05882352963089943, 0.05882352963089943, 0.06022408977150917, 0.06442577391862869, 0.08123249560594559, 0.07002801448106766, 0.0784313753247261, 0.10924369841814041, 0.10924369841814041, 0.1596638709306717, 0.08963585644960403, 0.08963585644960403, 0.1568627506494522, 0.10924369841814041, 0.1498599499464035, 0.17927171289920807, 0.12044817954301834, 0.12324929982423782, 0.2044817954301834, 0.14845938980579376, 0.17086835205554962, 0.22549019753932953, 0.20588235557079315, 0.16946779191493988, 0.2240896373987198, 0.23529411852359772, 0.16806723177433014, 0.21288515627384186, 0.20168067514896393, 0.21848739683628082, 0.24229691922664642, 0.22969187796115875, 0.2044817954301834, 0.2689075767993927, 0.30532214045524597, 0.2647058963775635, 0.24089635908603668, 0.2969187796115875, 0.36974790692329407, 0.24089635908603668, 0.299719899892807, 0.37535014748573303, 0.2717086970806122, 0.28011205792427063, 0.29831933975219727, 0.31092438101768494, 0.33893558382987976, 0.3193277418613434, 0.3977591097354889, 0.37254902720451355, 0.4425770342350006, 0.4789915978908539, 0.45378151535987854, 0.5098039507865906, 0.49859943985939026, 0.34873950481414795, 0.5042017102241516, 0.4957983195781708, 0.5126050710678101, 0.46498599648475647, NaN, 0.46498599648475647, 0.5210084319114685, 0.5378151535987854, 0.5826330780982971, 0.6204481720924377, 0.6302521228790283, 0.6610644459724426, 0.6722689270973206, 0.6988795399665833, 0.6946778893470764, 0.9075630307197571], [0.0476190485060215, 0.05602240934967995, 0.06442577391862869, 0.07563025504350662, 0.07282913476228714, 0.08963585644960403, 0.08123249560594559, 0.095238097012043, 0.08403361588716507, 0.1120448186993599, 0.10644257813692093, 0.10924369841814041, 0.10084033757448196, 0.19607843458652496, 0.13165266811847687, 0.11484593898057938, 0.13445378839969635, 0.22969187796115875, 0.19327731430530548, 0.15126051008701324, 0.24089635908603668, 0.14145658910274506, 0.2212885171175003, 0.16526611149311066, 0.15126051008701324, 0.228291317820549, 0.21008403599262238, 0.22689075767993927, 0.24929971992969513, 0.07002801448106766, 0.26330533623695374, 0.24649859964847565, 0.27450981736183167, 0.2002801150083542, 0.2857142984867096, 0.2857142984867096, 0.3165266215801239, 0.18767507374286652, 0.27450981736183167, 0.3193277418613434, 0.3361344635486603, 0.26330533623695374, 0.3193277418613434, 0.190476194024086, 0.32212886214256287, 0.3529411852359772, 0.27731093764305115, 0.3669467866420746, 0.2857142984867096, 0.3669467866420746, 0.380952388048172, 0.4005602300167084, 0.4929971992969513, 0.4089635908603668, 0.45098039507865906, 0.49859943985939026, 0.5042017102241516, 0.4481792747974396, 0.45098039507865906, 0.5154061913490295, 0.40616247057914734, 0.5350140333175659, NaN, 0.49859943985939026, 0.49859943985939026, 0.5490196347236633, 0.5854341983795166, 0.6610644459724426, 0.6162465214729309, 0.5980392098426819, 0.6778711676597595, 0.7282913327217102, 0.9299719929695129], [0.02240896411240101, 0.06162464991211891, 0.06722689419984818, 0.07002801448106766, 0.08683473616838455, 0.06722689419984818, 0.08403361588716507, 0.10084033757448196, 0.08683473616838455, 0.10644257813692093, 0.14565826952457428, 0.12885154783725739, 0.10364145785570145, 0.10644257813692093, 0.14845938980579376, 0.11764705926179886, 0.1596638709306717, 0.11484593898057938, 0.1736694723367691, 0.2072829157114029, 0.16246499121189117, 0.2044817954301834, 0.2310924381017685, 0.23529411852359772, 0.1764705926179886, 0.2380952388048172, 0.24089635908603668, 0.190476194024086, 0.14565826952457428, 0.21708683669567108, 0.15406163036823273, 0.21008403599262238, 0.15406163036823273, 0.2549019753932953, 0.24089635908603668, 0.21288515627384186, 0.14705882966518402, 0.2549019753932953, 0.21848739683628082, 0.30812326073646545, 0.24369747936725616, 0.3361344635486603, 0.3165266215801239, 0.28151261806488037, 0.1736694723367691, 0.27450981736183167, 0.3137255012989044, 0.2885154187679291, 0.3193277418613434, 0.3921568691730499, 0.3921568691730499, 0.28431373834609985, 0.49439775943756104, 0.33893558382987976, 0.36974790692329407, 0.5210084319114685, 0.4425770342350006, 0.42016807198524475, 0.4145658314228058, 0.5266106724739075, 0.4733893573284149, 0.4789915978908539, 0.5014005899429321, NaN, 0.6106442809104919, 0.6008403301239014, 0.6414566040039062, 0.6610644459724426, 0.7030812501907349, 0.7843137383460999, 0.5686274766921997, 0.7731092572212219, 0.9551820755004883], [0.02521008439362049, 0.03921568766236305, 0.07002801448106766, 0.06442577391862869, 0.05602240934967995, 0.053221289068460464, 0.08123249560594559, 0.08963585644960403, 0.08403361588716507, 0.08403361588716507, 0.12885154783725739, 0.11344537883996964, 0.0784313753247261, 0.11484593898057938, 0.12324929982423782, 0.11484593898057938, 0.1428571492433548, 0.08403361588716507, 0.17927171289920807, 0.17927171289920807, 0.1260504275560379, 0.17927171289920807, 0.22268907725811005, 0.19607843458652496, 0.15406163036823273, 0.2240896373987198, 0.21568627655506134, 0.16806723177433014, 0.13725490868091583, 0.21008403599262238, 0.12745098769664764, 0.20168067514896393, 0.15126051008701324, 0.23249299824237823, 0.21288515627384186, 0.21568627655506134, 0.1498599499464035, 0.2380952388048172, 0.2072829157114029, 0.27450981736183167, 0.23249299824237823, 0.2969187796115875, 0.2829131782054901, 0.27731093764305115, 0.16526611149311066, 0.2717086970806122, 0.3025210201740265, 0.2717086970806122, 0.30812326073646545, 0.38655462861061096, 0.37535014748573303, 0.2549019753932953, 0.46498599648475647, 0.299719899892807, 0.3473389446735382, 0.5098039507865906, 0.41736695170402527, 0.38655462861061096, 0.3851540684700012, 0.523809552192688, 0.43417367339134216, 0.462184876203537, 0.5014005899429321, 0.38935574889183044, NaN, 0.5966386795043945, 0.6246498823165894, 0.6400560140609741, 0.6694678068161011, 0.6918767690658569, 0.5490196347236633, 0.7647058963775635, 0.9579831957817078], [0.05602240934967995, 0.07282913476228714, 0.06862745434045792, 0.09103641659021378, 0.06722689419984818, 0.08963585644960403, 0.10504201799631119, 0.09383753687143326, 0.08403361588716507, 0.15406163036823273, 0.12745098769664764, 0.13165266811847687, 0.18207283318042755, 0.095238097012043, 0.1428571492433548, 0.17086835205554962, 0.17086835205554962, 0.14565826952457428, 0.16806723177433014, 0.190476194024086, 0.16526611149311066, 0.22689075767993927, 0.22969187796115875, 0.18487395346164703, 0.21708683669567108, 0.24929971992969513, 0.190476194024086, 0.2689075767993927, 0.21568627655506134, 0.22969187796115875, 0.2380952388048172, 0.2072829157114029, 0.24929971992969513, 0.21848739683628082, 0.24369747936725616, 0.24089635908603668, 0.2521008551120758, 0.23249299824237823, 0.20168067514896393, 0.3137255012989044, 0.2857142984867096, 0.2521008551120758, 0.3165266215801239, 0.2829131782054901, 0.26330533623695374, 0.31092438101768494, 0.3333333432674408, 0.29131653904914856, 0.2717086970806122, 0.41736695170402527, 0.3641456663608551, 0.33893558382987976, 0.41736695170402527, 0.3445378243923187, 0.3473389446735382, 0.28011205792427063, 0.3949579894542694, 0.37254902720451355, 0.3669467866420746, 0.3669467866420746, 0.38655462861061096, 0.41736695170402527, 0.45098039507865906, 0.39915966987609863, 0.40336135029792786, NaN, 0.529411792755127, 0.49859943985939026, 0.5014005899429321, 0.5042017102241516, 0.5140056014060974, 0.5966386795043945, 0.8235294222831726], [0.019607843831181526, 0.03641456738114357, 0.03921568766236305, 0.07002801448106766, 0.04481792822480202, 0.05602240934967995, 0.07002801448106766, 0.07282913476228714, 0.06442577391862869, 0.08123249560594559, 0.10924369841814041, 0.08683473616838455, 0.09243697673082352, 0.17927171289920807, 0.11764705926179886, 0.08263305574655533, 0.12044817954301834, 0.1764705926179886, 0.16806723177433014, 0.15126051008701324, 0.19187675416469574, 0.1568627506494522, 0.16526611149311066, 0.12324929982423782, 0.08123249560594559, 0.14565826952457428, 0.24089635908603668, 0.10924369841814041, 0.2072829157114029, 0.1764705926179886, 0.21008403599262238, 0.2380952388048172, 0.21848739683628082, 0.190476194024086, 0.2549019753932953, 0.27450981736183167, 0.20168067514896393, 0.23529411852359772, 0.27731093764305115, 0.28011205792427063, 0.24929971992969513, 0.2240896373987198, 0.2885154187679291, 0.28011205792427063, 0.21288515627384186, 0.28011205792427063, 0.23529411852359772, 0.27731093764305115, 0.2661064565181732, 0.228291317820549, 0.35854342579841614, 0.3361344635486603, 0.3669467866420746, 0.3669467866420746, 0.38655462861061096, 0.43977591395378113, 0.35574230551719666, 0.3193277418613434, 0.31092438101768494, 0.4425770342350006, 0.3977591097354889, 0.37955182790756226, 0.4145658314228058, 0.35854342579841614, 0.37535014748573303, 0.47058823704719543, NaN, 0.456582635641098, 0.4789915978908539, 0.4593837559223175, 0.5098039507865906, 0.5434173941612244, 0.8655462265014648], [0.02240896411240101, 0.05882352963089943, 0.053221289068460464, 0.08683473616838455, 0.08683473616838455, 0.07563025504350662, 0.08683473616838455, 0.08963585644960403, 0.08123249560594559, 0.1260504275560379, 0.12885154783725739, 0.10364145785570145, 0.11484593898057938, 0.08683473616838455, 0.13165266811847687, 0.12464985996484756, 0.1260504275560379, 0.11764705926179886, 0.15126051008701324, 0.11484593898057938, 0.14565826952457428, 0.18207283318042755, 0.2212885171175003, 0.15406163036823273, 0.1498599499464035, 0.2212885171175003, 0.14005602896213531, 0.190476194024086, 0.13725490868091583, 0.1428571492433548, 0.13725490868091583, 0.13725490868091583, 0.1428571492433548, 0.190476194024086, 0.15406163036823273, 0.1764705926179886, 0.19327731430530548, 0.1568627506494522, 0.13725490868091583, 0.21288515627384186, 0.22969187796115875, 0.2689075767993927, 0.22689075767993927, 0.19327731430530548, 0.19187675416469574, 0.26050421595573425, 0.26330533623695374, 0.26750701665878296, 0.23949579894542694, 0.3361344635486603, 0.3137255012989044, 0.19887955486774445, 0.3333333432674408, 0.24369747936725616, 0.2661064565181732, 0.37535014748573303, 0.3403361439704895, 0.3193277418613434, 0.3193277418613434, 0.4215686321258545, 0.3193277418613434, 0.36974790692329407, 0.33893558382987976, 0.33893558382987976, 0.3599439859390259, 0.5014005899429321, 0.5434173941612244, NaN, 0.4593837559223175, 0.47058823704719543, 0.47058823704719543, 0.8109243512153625, 0.9607843160629272], [0.016806723549962044, 0.03641456738114357, 0.05882352963089943, 0.06162464991211891, 0.06442577391862869, 0.06722689419984818, 0.06722689419984818, 0.07563025504350662, 0.06582633405923843, 0.07563025504350662, 0.12044817954301834, 0.07563025504350662, 0.07563025504350662, 0.13165266811847687, 0.095238097012043, 0.08683473616838455, 0.1260504275560379, 0.09243697673082352, 0.13725490868091583, 0.12044817954301834, 0.1120448186993599, 0.18207283318042755, 0.17927171289920807, 0.15406163036823273, 0.09243697673082352, 0.1764705926179886, 0.18907563388347626, 0.14565826952457428, 0.14845938980579376, 0.18207283318042755, 0.13165266811847687, 0.15406163036823273, 0.14005602896213531, 0.2380952388048172, 0.15126051008701324, 0.18767507374286652, 0.15126051008701324, 0.19607843458652496, 0.16246499121189117, 0.190476194024086, 0.19887955486774445, 0.2380952388048172, 0.19887955486774445, 0.24089635908603668, 0.15406163036823273, 0.21008403599262238, 0.27450981736183167, 0.24369747936725616, 0.27731093764305115, 0.2829131782054901, 0.33893558382987976, 0.15126051008701324, 0.3529411852359772, 0.2240896373987198, 0.3305322229862213, 0.4089635908603668, 0.3165266215801239, 0.28011205792427063, 0.29131653904914856, 0.4313725531101227, 0.3781512677669525, 0.33893558382987976, 0.3837535083293915, 0.2969187796115875, 0.3305322229862213, 0.49859943985939026, 0.5210084319114685, 0.5406162738800049, NaN, 0.5406162738800049, 0.43697479367256165, 0.6470588445663452, 0.9467787146568298], [0.02521008439362049, 0.030812324956059456, 0.053221289068460464, 0.05882352963089943, 0.0476190485060215, 0.0476190485060215, 0.06722689419984818, 0.07002801448106766, 0.06722689419984818, 0.06442577391862869, 0.10644257813692093, 0.0784313753247261, 0.0476190485060215, 0.09803921729326248, 0.095238097012043, 0.08963585644960403, 0.1120448186993599, 0.05882352963089943, 0.12885154783725739, 0.11764705926179886, 0.08963585644960403, 0.1428571492433548, 0.17086835205554962, 0.14005602896213531, 0.07563025504350662, 0.18207283318042755, 0.19887955486774445, 0.11484593898057938, 0.12044817954301834, 0.16246499121189117, 0.08683473616838455, 0.14565826952457428, 0.09803921729326248, 0.2212885171175003, 0.1568627506494522, 0.16246499121189117, 0.09803921729326248, 0.19327731430530548, 0.1736694723367691, 0.19607843458652496, 0.190476194024086, 0.2717086970806122, 0.20168067514896393, 0.22689075767993927, 0.12324929982423782, 0.2240896373987198, 0.24649859964847565, 0.23529411852359772, 0.2689075767993927, 0.2857142984867096, 0.3025210201740265, 0.19607843458652496, 0.3949579894542694, 0.23389355838298798, 0.29411765933036804, 0.4453781545162201, 0.2969187796115875, 0.26050421595573425, 0.2759103775024414, 0.4593837559223175, 0.3837535083293915, 0.32773110270500183, 0.4019607901573181, 0.21568627655506134, 0.30812326073646545, 0.4957983195781708, 0.5406162738800049, 0.529411792755127, 0.4593837559223175, NaN, 0.4453781545162201, 0.6386554837226868, 0.9271708726882935], [0.014005602337419987, 0.03641456738114357, 0.02521008439362049, 0.04481792822480202, 0.03641456738114357, 0.0476190485060215, 0.0476190485060215, 0.04481792822480202, 0.042016807943582535, 0.07563025504350662, 0.09243697673082352, 0.06722689419984818, 0.042016807943582535, 0.07282913476228714, 0.042016807943582535, 0.0784313753247261, 0.08123249560594559, 0.06722689419984818, 0.053221289068460464, 0.1260504275560379, 0.07002801448106766, 0.06442577391862869, 0.18207283318042755, 0.10364145785570145, 0.08123249560594559, 0.1428571492433548, 0.1596638709306717, 0.1120448186993599, 0.13165266811847687, 0.07563025504350662, 0.14005602896213531, 0.11764705926179886, 0.12885154783725739, 0.11484593898057938, 0.13725490868091583, 0.16806723177433014, 0.15126051008701324, 0.1120448186993599, 0.1568627506494522, 0.3137255012989044, 0.15406163036823273, 0.16526611149311066, 0.3165266215801239, 0.1596638709306717, 0.1568627506494522, 0.16386555135250092, 0.2044817954301834, 0.2072829157114029, 0.18767507374286652, 0.27731093764305115, 0.29411765933036804, 0.3361344635486603, 0.4117647111415863, 0.35854342579841614, 0.38935574889183044, 0.40616247057914734, 0.23529411852359772, 0.35854342579841614, 0.32773110270500183, 0.4089635908603668, 0.32492998242378235, 0.30112046003341675, 0.32212886214256287, 0.4313725531101227, 0.45098039507865906, 0.4859943985939026, 0.4901960790157318, 0.529411792755127, 0.5630252361297607, 0.5546218752861023, NaN, 0.6302521228790283, 0.8991596698760986], [0.03361344709992409, 0.04481792822480202, 0.05042016878724098, 0.06722689419984818, 0.06722689419984818, 0.06442577391862869, 0.08403361588716507, 0.08123249560594559, 0.05882352963089943, 0.10084033757448196, 0.06722689419984818, 0.10084033757448196, 0.09243697673082352, 0.06162464991211891, 0.1120448186993599, 0.1120448186993599, 0.12044817954301834, 0.09663865715265274, 0.1428571492433548, 0.08963585644960403, 0.1120448186993599, 0.1568627506494522, 0.14845938980579376, 0.13165266811847687, 0.13725490868091583, 0.1764705926179886, 0.12885154783725739, 0.14705882966518402, 0.10644257813692093, 0.1260504275560379, 0.1260504275560379, 0.12464985996484756, 0.11764705926179886, 0.15406163036823273, 0.12324929982423782, 0.13445378839969635, 0.14425770938396454, 0.1568627506494522, 0.1120448186993599, 0.15126051008701324, 0.16806723177433014, 0.2212885171175003, 0.16526611149311066, 0.16806723177433014, 0.14845938980579376, 0.2044817954301834, 0.20868347585201263, 0.18207283318042755, 0.23249299824237823, 0.2969187796115875, 0.26330533623695374, 0.16806723177433014, 0.26050421595573425, 0.16246499121189117, 0.19327731430530548, 0.299719899892807, 0.27731093764305115, 0.26330533623695374, 0.2521008551120758, 0.32212886214256287, 0.2689075767993927, 0.30532214045524597, 0.2717086970806122, 0.22689075767993927, 0.23529411852359772, 0.40336135029792786, 0.456582635641098, 0.18907563388347626, 0.3529411852359772, 0.3613445460796356, 0.36974790692329407, NaN, 0.9411764740943909], [0.008403361774981022, 0.005602241028100252, 0.005602241028100252, 0.008403361774981022, 0.008403361774981022, 0.011204482056200504, 0.011204482056200504, 0.011204482056200504, 0.008403361774981022, 0.014005602337419987, 0.008403361774981022, 0.005602241028100252, 0.008403361774981022, 0.019607843831181526, 0.02240896411240101, 0.008403361774981022, 0.014005602337419987, 0.011204482056200504, 0.03361344709992409, 0.011204482056200504, 0.008403361774981022, 0.02521008439362049, 0.02521008439362049, 0.014005602337419987, 0.014005602337419987, 0.028011204674839973, 0.02240896411240101, 0.030812324956059456, 0.016806723549962044, 0.030812324956059456, 0.011204482056200504, 0.028011204674839973, 0.011204482056200504, 0.03361344709992409, 0.028011204674839973, 0.03921568766236305, 0.02521008439362049, 0.030812324956059456, 0.030812324956059456, 0.05882352963089943, 0.03501400724053383, 0.03361344709992409, 0.05882352963089943, 0.042016807943582535, 0.030812324956059456, 0.05042016878724098, 0.05042016878724098, 0.042016807943582535, 0.0476190485060215, 0.08403361588716507, 0.08123249560594559, 0.014005602337419987, 0.06442577391862869, 0.016806723549962044, 0.0784313753247261, 0.10644257813692093, 0.05882352963089943, 0.07563025504350662, 0.05182072892785072, 0.11764705926179886, 0.07282913476228714, 0.09243697673082352, 0.07002801448106766, 0.04481792822480202, 0.042016807943582535, 0.1764705926179886, 0.13445378839969635, 0.03921568766236305, 0.053221289068460464, 0.07282913476228714, 0.10084033757448196, 0.05882352963089943, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9062597305},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8426704069},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8320300403},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8106909496},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8078898301},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.7928727126},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7870564964},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7723311614},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7407018449},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7117180293},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6726774166},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6698179376},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6689036821},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6688064231},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6657329683},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6564348047},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6397642513},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6172969286},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6138344336},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.6072401321},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5853174675},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.5768752056},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5720510568},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5668962137},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5633364559},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.553513086},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5493308562},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5474050836},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5403244751},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5395852895},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5370954046},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5321739915},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5308706921},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5091231031},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5022759201},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.500058368},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.499319181},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4979964324},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4969071064},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4914604827},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4899626629},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.4881730579},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.4836406886},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4753929452},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.4744008802},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4601618516},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4435301984},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4411764806},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4355936903},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4347183425},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.4095666151},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3888694453},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.3825863784},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3797852572},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.351073773},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.3474362062},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3345199282},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3315631902},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3232570891},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.323179279},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3195417148},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.304213359},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.3017040222},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2951875266},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.2748988556},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.2726034927},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.246576414},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2339519197},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2252956799},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2153166872},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2125544719},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1833955855},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0395269229}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_binary/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_binary/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..61fcc7e32a8f6d42090ec97af05f196c58ec6180
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_binary/datasets_small/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),TabPFN-3 (default),TabICLv2 (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),TabDPT-Turbo (default),LimiX (default),TabSwift (default),"AutoGluon 1.4 (best, 4h)",RealMLP (tuned + ensembled),BetaTabPFN (default),TabM (tuned + ensembled),TabDPT (tuned + ensembled),TabDPT (tuned),LightGBM (tuned + ensembled),TabM (tuned),TabDPT (default),LightGBM (tuned),RealMLP (tuned),SAP-RPT-OSS (default),xRFM (tuned),xRFM (tuned + ensembled),ModernNCA (tuned),ModernNCA (default),iLTM (tuned + ensembled),CatBoost (default),FastaiMLP (tuned + ensembled),CatBoost (tuned + ensembled),EBM (tuned + ensembled),CatBoost (tuned),TorchMLP (tuned + ensembled),EBM (tuned),TabM (default),XGBoost (tuned + ensembled),RealMLP (default),EBM (default),TabSTAR (tuned),ChimeraBoost (tuned + ensembled),OrionMSP (default),TabSTAR (tuned + ensembled),FastaiMLP (tuned),XGBoost (tuned),ChimeraBoost (tuned),iLTM (tuned),ChimeraBoost (default),TorchMLP (tuned),ModernNCA (tuned + ensembled),TabFlex (default),ExtraTrees (tuned + ensembled),TabSTAR (default),ExtraTrees (tuned),iLTM (default),Linear (tuned + ensembled),PerpetualBooster (tuned + ensembled),XGBoost (default),LightGBM (default),Linear (tuned),TorchMLP (default),PerpetualBooster (tuned),FastaiMLP (default),RandomForest (tuned + ensembled),RandomForest (tuned),Linear (default),xRFM (default),KNN (tuned + ensembled),ExtraTrees (default),RandomForest (default),PerpetualBooster (default),KNN (tuned),KNN (default)
+TabFM (default),,0.77310926,0.789916,0.7759104,0.7787115,0.789916,0.79971987,0.8067227,0.8095238,0.89495796,0.8487395,0.89635855,0.8935574,0.83613443,0.8879552,0.88235295,0.8767507,0.89915967,0.92436975,0.88515407,0.89915967,0.88515407,0.87114847,0.8795518,0.91036415,0.90756303,0.8739496,0.92156863,0.92436975,0.89635855,0.92436975,0.8907563,0.92436975,0.95238096,0.89915967,0.92156863,0.95238096,0.8935574,0.8907563,0.7787115,0.929972,0.929972,0.7815126,0.9159664,0.9551821,0.9439776,0.93557423,0.94957983,0.9607843,0.9551821,0.91036415,0.9439776,0.8207283,0.9327731,0.9271709,0.9131653,0.97759104,0.9691877,0.96638656,0.92436975,0.9719888,0.98039216,0.95238096,0.97759104,0.9747899,0.9439776,0.98039216,0.97759104,0.9831933,0.9747899,0.9859944,0.96638656,0.99159664
+TabPFN-3 (default),0.22689076,,0.50700283,0.60784316,0.6666667,0.6386555,0.6582633,0.6862745,0.6862745,0.7478992,0.80112046,0.7619048,0.8067227,0.7352941,0.79831934,0.8067227,0.8235294,0.8039216,0.837535,0.8207283,0.8319328,0.82913166,0.8095238,0.8235294,0.859944,0.859944,0.82212883,0.8879552,0.87114847,0.8305322,0.859944,0.84313726,0.8739496,0.9047619,0.85434175,0.86834735,0.87815124,0.86834735,0.8445378,0.7352941,0.9019608,0.8795518,0.7507003,0.8613445,0.89915967,0.91036415,0.9173669,0.89915967,0.9271709,0.9187675,0.88235295,0.9327731,0.7927171,0.9271709,0.9159664,0.87114847,0.92156863,0.94957983,0.9467787,0.88235295,0.9439776,0.9411765,0.9439776,0.93837535,0.9607843,0.9271709,0.96358544,0.9411765,0.96358544,0.9691877,0.96358544,0.9551821,0.99439776
+TabICLv2 (default),0.21008404,0.4929972,,0.5854342,0.5868347,0.627451,0.60784316,0.6246499,0.6862745,0.75770307,0.78711486,0.7955182,0.7927171,0.74229693,0.80112046,0.79691875,0.78431374,0.789916,0.8179272,0.81512606,0.81512606,0.81232494,0.7759104,0.8137255,0.8515406,0.8459384,0.8277311,0.8655462,0.8487395,0.84313726,0.8347339,0.8347339,0.8459384,0.89915967,0.8319328,0.84313726,0.8627451,0.8627451,0.84033614,0.7507003,0.8795518,0.90756303,0.7478992,0.87114847,0.8795518,0.9019608,0.8879552,0.8907563,0.92156863,0.9131653,0.8739496,0.89495796,0.79831934,0.8935574,0.90756303,0.8907563,0.929972,0.9271709,0.9411765,0.9019608,0.95238096,0.9411765,0.93557423,0.9327731,0.929972,0.9313725,0.9607843,0.9467787,0.9411765,0.9467787,0.9747899,0.94957983,0.99439776
+"AutoGluon 1.5 (extreme, 4h)",0.22408964,0.39215687,0.41456583,,0.56582636,0.6134454,0.59103644,0.60784316,0.5406163,0.69187677,0.72829133,0.789916,0.780112,0.70308125,0.80112046,0.74509805,0.75910366,0.8095238,0.81092435,0.73109245,0.8095238,0.8207283,0.75910366,0.78571427,0.84313726,0.84313726,0.79831934,0.8627451,0.8333333,0.8235294,0.84033614,0.8067227,0.8305322,0.8795518,0.81232494,0.8515406,0.8627451,0.8207283,0.8235294,0.68907565,0.88235295,0.8837535,0.69187677,0.8347339,0.87114847,0.8907563,0.89635855,0.8879552,0.90756303,0.9271709,0.85434175,0.90756303,0.75630254,0.8879552,0.9047619,0.8459384,0.92436975,0.9271709,0.93557423,0.8795518,0.93557423,0.9397759,0.92436975,0.929972,0.93557423,0.90896356,0.929972,0.9131653,0.93837535,0.9411765,0.9551821,0.9327731,0.99159664
+RealTabPFN-2.5 (default),0.22128852,0.33333334,0.41316527,0.43417367,,0.52100843,0.5182073,0.5602241,0.54901963,0.6722689,0.73109245,0.7394958,0.7394958,0.7212885,0.7478992,0.7058824,0.7072829,0.789916,0.7955182,0.71708685,0.81232494,0.8179272,0.7815126,0.8179272,0.82913166,0.81232494,0.81232494,0.82913166,0.87114847,0.8459384,0.84033614,0.84033614,0.82913166,0.88235295,0.8655462,0.86834735,0.8515406,0.859944,0.8347339,0.72829133,0.89635855,0.8935574,0.732493,0.87114847,0.8795518,0.9019608,0.90616244,0.9047619,0.92436975,0.8739496,0.88095236,0.8879552,0.77030814,0.8879552,0.91036415,0.859944,0.92156863,0.92436975,0.9579832,0.8739496,0.96358544,0.93557423,0.9271709,0.9131653,0.9439776,0.9327731,0.9551821,0.9131653,0.93557423,0.95238096,0.96358544,0.9327731,0.99159664
+TabPFN-2.6 (default),0.21008404,0.36134455,0.37254903,0.38655463,0.4789916,,0.4789916,0.51960784,0.55182076,0.6694678,0.6834734,0.7338936,0.7535014,0.6834734,0.7535014,0.6946779,0.6806723,0.7787115,0.7927171,0.6694678,0.81512606,0.8067227,0.77310926,0.7815126,0.8179272,0.82913166,0.8095238,0.859944,0.84313726,0.7759104,0.84033614,0.78711486,0.8235294,0.859944,0.81092435,0.8305322,0.84313726,0.8235294,0.7927171,0.6862745,0.89635855,0.8179272,0.70028013,0.8207283,0.859944,0.9047619,0.8879552,0.90756303,0.9047619,0.929972,0.8515406,0.89915967,0.74229693,0.8865546,0.8907563,0.8459384,0.9187675,0.9439776,0.94957983,0.8767507,0.93837535,0.9187675,0.91036415,0.9327731,0.9467787,0.91036415,0.9439776,0.92436975,0.9327731,0.95238096,0.95238096,0.93557423,0.9887955
+RealTabPFN-2.5 (tuned + ensembled),0.20028012,0.3417367,0.39215687,0.4089636,0.48179272,0.52100843,,0.59103644,0.5154062,0.6946779,0.6694678,0.74229693,0.7366947,0.7114846,0.7394958,0.6694678,0.6806723,0.7815126,0.7787115,0.6246499,0.802521,0.7759104,0.75770307,0.7815126,0.81512606,0.8347339,0.7927171,0.8193277,0.8319328,0.7941176,0.81512606,0.7955182,0.81232494,0.85434175,0.81092435,0.8179272,0.8347339,0.81232494,0.8039216,0.68907565,0.86694676,0.8515406,0.697479,0.84313726,0.8613445,0.88235295,0.8739496,0.88235295,0.89635855,0.8879552,0.8459384,0.8655462,0.74229693,0.85714287,0.8907563,0.859944,0.92436975,0.92156863,0.93837535,0.88235295,0.9187675,0.929972,0.9187675,0.9159664,0.9187675,0.89495796,0.929972,0.9131653,0.9327731,0.9327731,0.95238096,0.9159664,0.9887955
+RealTabPFN-2.5 (tuned),0.19327731,0.3137255,0.37535015,0.39215687,0.4397759,0.48039216,0.4089636,,0.5182073,0.64705884,0.65686274,0.71428573,0.697479,0.697479,0.71708685,0.66106445,0.6820728,0.7647059,0.7633053,0.6414566,0.7955182,0.7619048,0.7478992,0.7647059,0.78291315,0.78711486,0.77030814,0.7955182,0.8039216,0.7787115,0.80112046,0.8039216,0.7955182,0.8515406,0.8067227,0.8179272,0.8207283,0.8095238,0.789916,0.6806723,0.8473389,0.8459384,0.69187677,0.8235294,0.8319328,0.86834735,0.8655462,0.87114847,0.89635855,0.87815124,0.8319328,0.8613445,0.74509805,0.859944,0.88515407,0.84033614,0.9005602,0.9159664,0.9327731,0.859944,0.9271709,0.92156863,0.9047619,0.89915967,0.91036415,0.90616244,0.9271709,0.91036415,0.92436975,0.929972,0.9551821,0.9187675,0.9887955
+TabDPT-Turbo (default),0.1904762,0.3137255,0.3137255,0.45938376,0.4509804,0.44817927,0.48459384,0.48179272,,0.5714286,0.6904762,0.60784316,0.6442577,0.5742297,0.61764705,0.67507005,0.6834734,0.6526611,0.6666667,0.7885154,0.67787117,0.71568626,0.6862745,0.73109245,0.719888,0.7058824,0.7394958,0.72829133,0.71428573,0.767507,0.7366947,0.73109245,0.72829133,0.75630254,0.7521008,0.7394958,0.75630254,0.77030814,0.7507003,0.6694678,0.767507,0.82913166,0.6806723,0.81232494,0.767507,0.789916,0.8305322,0.8095238,0.81512606,0.7619048,0.87114847,0.8697479,0.74229693,0.89635855,0.8319328,0.8319328,0.8627451,0.86834735,0.88235295,0.8515406,0.9019608,0.8907563,0.9159664,0.9131653,0.9159664,0.9159664,0.93557423,0.9187675,0.93417364,0.9327731,0.9579832,0.9411765,0.99159664
+LimiX (default),0.10504202,0.25210086,0.24229692,0.30812326,0.3277311,0.33053222,0.30532214,0.3529412,0.42857143,,0.5266107,0.6302521,0.6442577,0.5742297,0.6246499,0.54761904,0.55182076,0.6862745,0.640056,0.57703084,0.697479,0.7044818,0.70308125,0.6820728,0.7254902,0.7254902,0.70028013,0.8165266,0.7366947,0.6960784,0.74229693,0.74509805,0.7394958,0.7787115,0.7619048,0.71708685,0.7787115,0.7394958,0.75630254,0.6764706,0.8095238,0.75770307,0.6722689,0.7521008,0.80112046,0.8319328,0.82633054,0.84033614,0.81232494,0.837535,0.7366947,0.8235294,0.732493,0.81232494,0.859944,0.8095238,0.8739496,0.8795518,0.9019608,0.837535,0.859944,0.8907563,0.8879552,0.8935574,0.9159664,0.8459384,0.9187675,0.8739496,0.92436975,0.93557423,0.92436975,0.89915967,0.9859944
+TabSwift (default),0.15126051,0.19887955,0.21288516,0.2717087,0.26890758,0.31652662,0.33053222,0.34313726,0.30952382,0.47338936,,0.5532213,0.57703084,0.49019608,0.52100843,0.5112045,0.53501403,0.57703084,0.56302524,0.56302524,0.5826331,0.6414566,0.6302521,0.63585436,0.6666667,0.6722689,0.66386557,0.67507005,0.6442577,0.7338936,0.64985996,0.6694678,0.6652661,0.72829133,0.6834734,0.6722689,0.6722689,0.7114846,0.67787117,0.69187677,0.6862745,0.8095238,0.7058824,0.767507,0.7114846,0.7254902,0.7955182,0.77310926,0.7647059,0.7338936,0.8627451,0.82633054,0.75910366,0.8207283,0.80112046,0.81232494,0.8459384,0.8179272,0.83613443,0.8347339,0.90756303,0.84033614,0.8935574,0.85434175,0.87114847,0.872549,0.8907563,0.87114847,0.8795518,0.8935574,0.90756303,0.9327731,0.99159664
+"AutoGluon 1.4 (best, 4h)",0.10364146,0.23809524,0.2044818,0.21008404,0.26050422,0.26610646,0.2577031,0.2857143,0.39215687,0.3697479,0.4467787,,0.5392157,0.5126051,0.52380955,0.49019608,0.4789916,0.64985996,0.59103644,0.5266107,0.6834734,0.64705884,0.53781515,0.64985996,0.6372549,0.60504204,0.6386555,0.6862745,0.71428573,0.6834734,0.719888,0.67787117,0.72829133,0.75630254,0.6946779,0.7058824,0.767507,0.7338936,0.7226891,0.63585436,0.74369746,0.62885153,0.64705884,0.7338936,0.7619048,0.7619048,0.78431374,0.7759104,0.7955182,0.71428573,0.7394958,0.75630254,0.7058824,0.7647059,0.8179272,0.79971987,0.9047619,0.8935574,0.9187675,0.8333333,0.8627451,0.91036415,0.8907563,0.87114847,0.8865546,0.86834735,0.9131653,0.89635855,0.92436975,0.92156863,0.9327731,0.89915967,0.99439776
+RealMLP (tuned + ensembled),0.10644258,0.19327731,0.20728292,0.21988796,0.26050422,0.2464986,0.26330534,0.30252102,0.3557423,0.3557423,0.4229692,0.46078432,,0.48459384,0.5126051,0.46078432,0.46918768,0.627451,0.59103644,0.52380955,0.67086834,0.7254902,0.54901963,0.6134454,0.6554622,0.6134454,0.6134454,0.71428573,0.7058824,0.6960784,0.7394958,0.7114846,0.73109245,0.70308125,0.719888,0.7086835,0.77310926,0.78571427,0.71568626,0.6246499,0.77030814,0.6442577,0.6386555,0.72689074,0.79831934,0.78571427,0.77310926,0.78711486,0.75630254,0.7058824,0.7114846,0.80112046,0.69187677,0.7941176,0.8459384,0.78711486,0.89915967,0.88235295,0.8907563,0.7927171,0.87114847,0.91036415,0.89915967,0.89635855,0.92156863,0.8179272,0.90756303,0.88515407,0.92436975,0.95238096,0.9579832,0.90756303,0.99159664
+BetaTabPFN (default),0.16386555,0.2647059,0.2577031,0.29691878,0.2787115,0.31652662,0.28851542,0.30252102,0.4257703,0.4257703,0.50980395,0.48739496,0.5154062,,0.4789916,0.52380955,0.5322129,0.64705884,0.557423,0.5882353,0.63585436,0.5826331,0.57703084,0.5994398,0.6246499,0.6372549,0.63305324,0.6386555,0.7058824,0.6260504,0.66386557,0.7254902,0.6960784,0.6694678,0.7366947,0.7338936,0.7478992,0.69187677,0.7647059,0.6694678,0.74089634,0.6526611,0.67507005,0.70308125,0.767507,0.767507,0.70308125,0.78431374,0.7478992,0.6862745,0.71568626,0.8179272,0.74089634,0.81232494,0.859944,0.85434175,0.8235294,0.8347339,0.8557423,0.88235295,0.82633054,0.84313726,0.8039216,0.8935574,0.88515407,0.9047619,0.8207283,0.9131653,0.86834735,0.9019608,0.9271709,0.93837535,0.98039216
+TabM (tuned + ensembled),0.11204482,0.20168068,0.19887955,0.19887955,0.25210086,0.2464986,0.26050422,0.28291318,0.38235295,0.37535015,0.4789916,0.47619048,0.48739496,0.52100843,,0.49439776,0.49019608,0.63305324,0.662465,0.52380955,0.662465,0.6386555,0.5686275,0.5742297,0.6022409,0.6260504,0.63305324,0.662465,0.71708685,0.6442577,0.70028013,0.719888,0.7086835,0.6946779,0.72829133,0.79971987,0.75910366,0.7212885,0.7184874,0.6386555,0.7787115,0.6246499,0.6442577,0.697479,0.80112046,0.82633054,0.7535014,0.8319328,0.76890755,0.7366947,0.71428573,0.7927171,0.697479,0.7647059,0.8319328,0.81232494,0.86834735,0.8795518,0.8837535,0.8277311,0.87114847,0.8907563,0.86834735,0.8515406,0.8767507,0.85714287,0.88235295,0.86834735,0.9047619,0.9047619,0.9579832,0.8879552,0.97759104
+TabDPT (tuned + ensembled),0.11764706,0.19327731,0.20308124,0.25490198,0.29411766,0.30532214,0.33053222,0.33893558,0.32492998,0.45238096,0.48879552,0.50980395,0.5392157,0.47619048,0.50560224,,0.5854342,0.5462185,0.5560224,0.5826331,0.5994398,0.6372549,0.6246499,0.61764705,0.64285713,0.63305324,0.640056,0.6694678,0.63585436,0.6666667,0.6442577,0.6372549,0.6414566,0.7044818,0.6652661,0.64565825,0.65966386,0.67507005,0.64985996,0.6134454,0.68907565,0.7366947,0.6162465,0.7254902,0.70308125,0.719888,0.7759104,0.74509805,0.7394958,0.7619048,0.75910366,0.8039216,0.7058824,0.8039216,0.8235294,0.75630254,0.82913166,0.8207283,0.8487395,0.7815126,0.859944,0.85014004,0.88515407,0.88235295,0.88515407,0.82913166,0.9173669,0.8753501,0.9131653,0.91036415,0.92156863,0.8879552,0.99159664
+TabDPT (tuned),0.1232493,0.1764706,0.21568628,0.24089636,0.2927171,0.31932774,0.31932774,0.31792718,0.31652662,0.44817927,0.464986,0.52100843,0.5308123,0.46778712,0.50980395,0.41456583,,0.53781515,0.5546219,0.56442577,0.56582636,0.5882353,0.5882353,0.57703084,0.6106443,0.60504204,0.63305324,0.63585436,0.6246499,0.6582633,0.63305324,0.6554622,0.6260504,0.7058824,0.6666667,0.63585436,0.65686274,0.6834734,0.66386557,0.5980392,0.67507005,0.7507003,0.5966387,0.7058824,0.6722689,0.6946779,0.7507003,0.7226891,0.7254902,0.7380952,0.7535014,0.7619048,0.70028013,0.77030814,0.789916,0.7605042,0.8095238,0.78711486,0.8039216,0.78431374,0.8487395,0.8207283,0.8655462,0.84033614,0.85714287,0.82913166,0.8795518,0.8739496,0.8739496,0.8879552,0.9187675,0.8795518,0.9859944
+LightGBM (tuned + ensembled),0.10084034,0.19607843,0.21008404,0.1904762,0.21008404,0.22128852,0.2184874,0.23529412,0.34733894,0.3137255,0.4229692,0.35014006,0.37254903,0.3529412,0.3669468,0.45378152,0.46218488,,0.4439776,0.5126051,0.5966387,0.50980395,0.50700283,0.56302524,0.557423,0.53781515,0.5322129,0.58963585,0.6764706,0.5686275,0.66386557,0.61764705,0.65686274,0.5854342,0.64985996,0.6764706,0.789916,0.6302521,0.7058824,0.6120448,0.6484594,0.60504204,0.6162465,0.6386555,0.81512606,0.6834734,0.70028013,0.78431374,0.6862745,0.6582633,0.6414566,0.76890755,0.7058824,0.7478992,0.8207283,0.7815126,0.8515406,0.88515407,0.9439776,0.7955182,0.77310926,0.8795518,0.77030814,0.88515407,0.9159664,0.85434175,0.8235294,0.88235295,0.90756303,0.9411765,0.9327731,0.9033613,0.9887955
+TabM (tuned),0.075630255,0.16246499,0.18207283,0.18907563,0.2044818,0.20728292,0.22128852,0.23669468,0.33333334,0.359944,0.4369748,0.4089636,0.4089636,0.44257703,0.33753502,0.4439776,0.44537815,0.5560224,,0.48879552,0.5882353,0.5294118,0.48739496,0.5182073,0.52380955,0.6106443,0.59383756,0.59103644,0.64985996,0.58963585,0.6526611,0.6106443,0.6526611,0.6526611,0.6414566,0.6834734,0.68767506,0.67507005,0.6554622,0.6162465,0.7394958,0.59103644,0.6204482,0.6666667,0.7058824,0.7366947,0.67507005,0.7745098,0.71288514,0.69187677,0.6946779,0.72829133,0.6722689,0.7226891,0.8095238,0.78431374,0.8627451,0.82633054,0.82913166,0.80112046,0.8515406,0.8767507,0.8067227,0.82633054,0.8207283,0.8319328,0.8319328,0.8487395,0.8627451,0.87114847,0.9467787,0.85714287,0.96638656
+TabDPT (default),0.11484594,0.17927171,0.18487395,0.26890758,0.28291318,0.33053222,0.37535015,0.35854343,0.2114846,0.4229692,0.4369748,0.47338936,0.47619048,0.4117647,0.47619048,0.41736695,0.43557423,0.48739496,0.5112045,,0.52100843,0.5462185,0.5546219,0.5462185,0.5686275,0.5826331,0.6022409,0.5714286,0.5714286,0.5966387,0.56582636,0.5546219,0.5742297,0.6414566,0.5812325,0.6246499,0.6022409,0.58963585,0.627451,0.57983196,0.6582633,0.719888,0.5854342,0.67787117,0.627451,0.64985996,0.67507005,0.66386557,0.69187677,0.6526611,0.7759104,0.77310926,0.67507005,0.77310926,0.7478992,0.74229693,0.77030814,0.7366947,0.7394958,0.75910366,0.8319328,0.7955182,0.8487395,0.7927171,0.8207283,0.8095238,0.8487395,0.88515407,0.8795518,0.88235295,0.8739496,0.91036415,0.9887955
+LightGBM (tuned),0.10084034,0.16806723,0.18487395,0.1904762,0.18767507,0.18487395,0.197479,0.2044818,0.32212886,0.30252102,0.41736695,0.31652662,0.32913166,0.36414567,0.33753502,0.40056023,0.43417367,0.40336135,0.4117647,0.4789916,,0.49019608,0.48179272,0.5182073,0.5252101,0.48459384,0.50980395,0.53501403,0.6344538,0.53361344,0.61764705,0.5966387,0.60784316,0.56302524,0.62184876,0.63305324,0.6904762,0.61764705,0.6526611,0.5882353,0.59103644,0.5826331,0.59383756,0.5882353,0.74229693,0.61904764,0.66386557,0.7338936,0.6652661,0.6246499,0.6246499,0.74369746,0.6834734,0.732493,0.79691875,0.767507,0.8305322,0.8417367,0.9033613,0.78571427,0.7619048,0.8515406,0.75910366,0.837535,0.8739496,0.8347339,0.80812323,0.85434175,0.8879552,0.91036415,0.929972,0.8879552,0.99159664
+RealMLP (tuned),0.11484594,0.17086835,0.18767507,0.17927171,0.18207283,0.19327731,0.22408964,0.23809524,0.28431374,0.29551822,0.35854343,0.3529412,0.27450982,0.41736695,0.36134455,0.3627451,0.4117647,0.49019608,0.47058824,0.45378152,0.50980395,,0.47338936,0.50560224,0.49579832,0.49579832,0.5434174,0.51960784,0.5854342,0.57983196,0.557423,0.59103644,0.56582636,0.60504204,0.6302521,0.6162465,0.6246499,0.64985996,0.5854342,0.6022409,0.6134454,0.5560224,0.6162465,0.6526611,0.662465,0.6582633,0.6442577,0.68907565,0.6834734,0.6106443,0.68767506,0.7086835,0.67507005,0.7086835,0.78711486,0.7058824,0.8207283,0.7815126,0.8039216,0.71568626,0.82633054,0.82913166,0.8585434,0.7955182,0.8207283,0.77310926,0.84313726,0.8179272,0.8179272,0.85714287,0.93557423,0.84313726,0.9747899
+SAP-RPT-OSS (default),0.12885155,0.1904762,0.22408964,0.24089636,0.2184874,0.22689076,0.24229692,0.25210086,0.3137255,0.29691878,0.3697479,0.46218488,0.4509804,0.4229692,0.43137255,0.37535015,0.4117647,0.4929972,0.5126051,0.44537815,0.5182073,0.5266107,,0.5462185,0.56582636,0.53781515,0.557423,0.5462185,0.52100843,0.6162465,0.5042017,0.55182076,0.5182073,0.6386555,0.5826331,0.5714286,0.59103644,0.6554622,0.58963585,0.54901963,0.5882353,0.57983196,0.5434174,0.6554622,0.60784316,0.6162465,0.6120448,0.6414566,0.68907565,0.6162465,0.697479,0.67507005,0.6414566,0.67507005,0.719888,0.73109245,0.7647059,0.7254902,0.72829133,0.7366947,0.767507,0.7745098,0.7787115,0.76890755,0.7773109,0.77030814,0.8347339,0.7787115,0.8207283,0.82913166,0.8179272,0.8515406,0.9747899
+xRFM (tuned),0.12044818,0.1764706,0.18627451,0.21428572,0.18207283,0.2184874,0.2184874,0.23529412,0.26890758,0.31792718,0.36414567,0.35014006,0.38655463,0.40056023,0.4257703,0.38235295,0.4229692,0.4369748,0.48179272,0.45378152,0.48179272,0.49439776,0.45378152,,0.43557423,0.45378152,0.53781515,0.48039216,0.53501403,0.5420168,0.5406163,0.5728291,0.5546219,0.557423,0.60504204,0.5686275,0.55182076,0.5854342,0.60784316,0.57983196,0.61484593,0.56582636,0.5826331,0.63585436,0.5840336,0.63585436,0.63305324,0.627451,0.6386555,0.5686275,0.7226891,0.71708685,0.6554622,0.73109245,0.77030814,0.75910366,0.7380952,0.70028013,0.74509805,0.78711486,0.82913166,0.7941176,0.8347339,0.7647059,0.8039216,0.81512606,0.8767507,0.8459384,0.8459384,0.859944,0.89635855,0.86834735,0.9859944
+xRFM (tuned + ensembled),0.08963586,0.14005603,0.14845939,0.15686275,0.17086835,0.18207283,0.18487395,0.21708684,0.28011206,0.27450982,0.33333334,0.3627451,0.34453782,0.37535015,0.3977591,0.35714287,0.38935575,0.44257703,0.47619048,0.43137255,0.47478992,0.5042017,0.43417367,0.56442577,,0.46638656,0.5042017,0.49439776,0.5532213,0.5532213,0.53781515,0.56302524,0.5448179,0.5266107,0.6022409,0.5686275,0.5672269,0.59383756,0.59103644,0.5448179,0.5994398,0.5742297,0.5434174,0.6302521,0.5882353,0.6162465,0.6092437,0.6106443,0.61904764,0.6302521,0.697479,0.69187677,0.6442577,0.719888,0.7338936,0.7226891,0.7759104,0.767507,0.78711486,0.74229693,0.81232494,0.8305322,0.8487395,0.8235294,0.8459384,0.78291315,0.9187675,0.85014004,0.90756303,0.92436975,0.9187675,0.8627451,0.9859944
+ModernNCA (tuned),0.09243698,0.14005603,0.15406163,0.15686275,0.18767507,0.17086835,0.16526611,0.21288516,0.29411766,0.27450982,0.3277311,0.394958,0.38655463,0.3627451,0.3739496,0.3669468,0.394958,0.46218488,0.38935575,0.41736695,0.5154062,0.5042017,0.46218488,0.5462185,0.53361344,,0.56302524,0.5182073,0.53501403,0.5672269,0.53501403,0.5462185,0.55182076,0.5686275,0.57703084,0.53501403,0.5854342,0.6162465,0.5826331,0.56582636,0.61904764,0.53781515,0.5742297,0.61904764,0.60364145,0.6232493,0.63305324,0.6582633,0.697479,0.6134454,0.67507005,0.6526611,0.6666667,0.66106445,0.7478992,0.6862745,0.74509805,0.7366947,0.7478992,0.7086835,0.7478992,0.7759104,0.77170867,0.7619048,0.7759104,0.7507003,0.85434175,0.7787115,0.8235294,0.8179272,0.85714287,0.8235294,0.9719888
+ModernNCA (default),0.12605043,0.17787115,0.17226891,0.20168068,0.18767507,0.1904762,0.20728292,0.22969188,0.26050422,0.2997199,0.33613446,0.36134455,0.38655463,0.3669468,0.3669468,0.359944,0.3669468,0.46778712,0.40616247,0.3977591,0.49019608,0.45658264,0.44257703,0.46218488,0.49579832,0.4369748,,0.48179272,0.5546219,0.50980395,0.55182076,0.5322129,0.56302524,0.5266107,0.5714286,0.5406163,0.57703084,0.5546219,0.59103644,0.5966387,0.6134454,0.5854342,0.60504204,0.5686275,0.6246499,0.63305324,0.59243697,0.6386555,0.6162465,0.53501403,0.6694678,0.6806723,0.6806723,0.6834734,0.7619048,0.7338936,0.73109245,0.6862745,0.719888,0.7521008,0.8067227,0.7647059,0.789916,0.75910366,0.78431374,0.8095238,0.75910366,0.859944,0.81092435,0.80112046,0.84033614,0.87114847,0.97759104
+iLTM (tuned + ensembled),0.078431375,0.11204482,0.13445379,0.13725491,0.17086835,0.14005603,0.18067227,0.2044818,0.2717087,0.1834734,0.32492998,0.3137255,0.2857143,0.36134455,0.33753502,0.33053222,0.36414567,0.41036415,0.4089636,0.42857143,0.464986,0.48039216,0.45378152,0.51960784,0.50560224,0.48179272,0.5182073,,0.5042017,0.5546219,0.48179272,0.5294118,0.5154062,0.5602241,0.5686275,0.5294118,0.5588235,0.5994398,0.5714286,0.55182076,0.59383756,0.5448179,0.5560224,0.6260504,0.60784316,0.59383756,0.7184874,0.64285713,0.64985996,0.6134454,0.67507005,0.6792717,0.6554622,0.69187677,0.74229693,0.6582633,0.8039216,0.7955182,0.8459384,0.6862745,0.7759104,0.8319328,0.77310926,0.8095238,0.8319328,0.73109245,0.8907563,0.8095238,0.85434175,0.88515407,0.8879552,0.85294116,0.9691877
+CatBoost (default),0.075630255,0.12885155,0.15126051,0.16666667,0.12885155,0.15686275,0.16806723,0.19607843,0.2857143,0.26330534,0.3557423,0.2857143,0.29411766,0.29411766,0.28291318,0.36414567,0.37535015,0.32352942,0.35014006,0.42857143,0.36554623,0.41456583,0.4789916,0.464986,0.4467787,0.464986,0.44537815,0.49579832,,0.50700283,0.5014006,0.5182073,0.4859944,0.5546219,0.5602241,0.5854342,0.56442577,0.57983196,0.59383756,0.5980392,0.5588235,0.5126051,0.60504204,0.5742297,0.5868347,0.60504204,0.64985996,0.6442577,0.6442577,0.5854342,0.5826331,0.72829133,0.6666667,0.71428573,0.8067227,0.70308125,0.7394958,0.8039216,0.8347339,0.7254902,0.75910366,0.78711486,0.7507003,0.85434175,0.8627451,0.78431374,0.7927171,0.8627451,0.8515406,0.8795518,0.86834735,0.8935574,0.9831933
+FastaiMLP (tuned + ensembled),0.10364146,0.16946779,0.15686275,0.1764706,0.15406163,0.22408964,0.20588236,0.22128852,0.232493,0.30392158,0.26610646,0.31652662,0.30392158,0.3739496,0.3557423,0.33333334,0.3417367,0.43137255,0.41036415,0.40336135,0.46638656,0.42016807,0.3837535,0.4579832,0.4467787,0.4327731,0.49019608,0.44537815,0.4929972,,0.5014006,0.53641456,0.5266107,0.49019608,0.5714286,0.52100843,0.5406163,0.5728291,0.5742297,0.5854342,0.57703084,0.5322129,0.59383756,0.6554622,0.57703084,0.5966387,0.5686275,0.5854342,0.60504204,0.57983196,0.6862745,0.66106445,0.6582633,0.6526611,0.7226891,0.7254902,0.77030814,0.7086835,0.710084,0.7507003,0.8165266,0.79831934,0.929972,0.78291315,0.789916,0.77030814,0.8235294,0.85714287,0.8179272,0.837535,0.92436975,0.8739496,0.9691877
+CatBoost (tuned + ensembled),0.075630255,0.14005603,0.16526611,0.15966387,0.15966387,0.15966387,0.18487395,0.19887955,0.26330534,0.2577031,0.35014006,0.28011206,0.26050422,0.33613446,0.2997199,0.3557423,0.3669468,0.33613446,0.34733894,0.43417367,0.38235295,0.44257703,0.49579832,0.45938376,0.46218488,0.464986,0.44817927,0.5182073,0.49859944,0.49859944,,0.4789916,0.5266107,0.5462185,0.52240896,0.5742297,0.5392157,0.5462185,0.54901963,0.5826331,0.56302524,0.50700283,0.57703084,0.5546219,0.5868347,0.5882353,0.64705884,0.6246499,0.63305324,0.59103644,0.56302524,0.719888,0.6526611,0.71568626,0.75770307,0.697479,0.7535014,0.81512606,0.84313726,0.719888,0.7366947,0.7815126,0.7366947,0.8459384,0.872549,0.7619048,0.789916,0.8627451,0.86834735,0.9131653,0.859944,0.8739496,0.9887955
+EBM (tuned + ensembled),0.1092437,0.15686275,0.16526611,0.19327731,0.15966387,0.21288516,0.2044818,0.19607843,0.26890758,0.25490198,0.33053222,0.32212886,0.28851542,0.27450982,0.28011206,0.3627451,0.34453782,0.38235295,0.38935575,0.44537815,0.40336135,0.4089636,0.44817927,0.42717087,0.4369748,0.45378152,0.46778712,0.47058824,0.48179272,0.46358544,0.52100843,,0.50700283,0.49579832,0.64705884,0.5854342,0.5854342,0.5294118,0.6414566,0.5686275,0.5266107,0.56582636,0.5686275,0.51960784,0.5854342,0.5602241,0.5602241,0.60504204,0.5812325,0.5462185,0.63305324,0.710084,0.6666667,0.69187677,0.7507003,0.74229693,0.73109245,0.7521008,0.7647059,0.7535014,0.6694678,0.75770307,0.7535014,0.789916,0.79831934,0.7927171,0.7619048,0.8627451,0.8459384,0.85434175,0.88235295,0.8753501,0.9719888
+CatBoost (tuned),0.075630255,0.12605043,0.15406163,0.16946779,0.17086835,0.1764706,0.18767507,0.2044818,0.2717087,0.26050422,0.3347339,0.2717087,0.26890758,0.30392158,0.29131654,0.35854343,0.3739496,0.34313726,0.34733894,0.4257703,0.39215687,0.43417367,0.48179272,0.44537815,0.45518208,0.44817927,0.4369748,0.48459384,0.5140056,0.47338936,0.47338936,0.4929972,,0.5406163,0.52380955,0.5532213,0.56302524,0.5406163,0.5602241,0.5882353,0.5252101,0.5042017,0.59383756,0.5588235,0.59383756,0.56442577,0.6106443,0.6134454,0.6232493,0.59103644,0.55182076,0.70028013,0.66106445,0.6862745,0.7338936,0.69187677,0.7366947,0.8319328,0.8235294,0.7072829,0.71708685,0.77030814,0.7254902,0.8459384,0.8487395,0.7507003,0.7815126,0.85714287,0.859944,0.9019608,0.87114847,0.88235295,0.9887955
+TorchMLP (tuned + ensembled),0.04761905,0.0952381,0.10084034,0.12044818,0.11764706,0.14005603,0.14565827,0.14845939,0.24369748,0.22128852,0.2717087,0.24369748,0.29691878,0.33053222,0.30532214,0.29551822,0.29411766,0.41456583,0.34733894,0.35854343,0.4369748,0.394958,0.36134455,0.44257703,0.47338936,0.43137255,0.47338936,0.4397759,0.44537815,0.50980395,0.45378152,0.5042017,0.45938376,,0.5322129,0.46778712,0.50560224,0.5462185,0.5434174,0.55182076,0.5154062,0.48459384,0.56162465,0.6064426,0.5294118,0.5602241,0.5560224,0.5742297,0.7072829,0.5714286,0.63305324,0.64705884,0.6582633,0.6386555,0.7240896,0.6904762,0.7507003,0.7254902,0.697479,0.72829133,0.837535,0.7955182,0.79971987,0.74509805,0.767507,0.7815126,0.8095238,0.8095238,0.7619048,0.7787115,0.88515407,0.8459384,0.96638656
+EBM (tuned),0.10084034,0.14565827,0.16806723,0.18767507,0.13445379,0.18907563,0.18907563,0.19327731,0.24789916,0.23809524,0.31652662,0.30532214,0.28011206,0.26330534,0.2717087,0.3347339,0.33333334,0.35014006,0.35854343,0.4187675,0.37815127,0.3697479,0.41736695,0.394958,0.3977591,0.4229692,0.42857143,0.43137255,0.4397759,0.42857143,0.47759104,0.3529412,0.47619048,0.46778712,,0.5266107,0.50700283,0.46218488,0.59243697,0.55182076,0.49019608,0.5280112,0.5588235,0.5,0.53501403,0.5252101,0.5266107,0.56302524,0.557423,0.5140056,0.5952381,0.66386557,0.6414566,0.6946779,0.7058824,0.71428573,0.7058824,0.72829133,0.74509805,0.72829133,0.64985996,0.73109245,0.71428573,0.75910366,0.78711486,0.75630254,0.74509805,0.8459384,0.8487395,0.84313726,0.8627451,0.8767507,0.9719888
+TabM (default),0.078431375,0.13165267,0.15686275,0.14845939,0.13165267,0.16946779,0.18207283,0.18207283,0.26050422,0.28291318,0.3277311,0.29411766,0.29131654,0.26610646,0.20028012,0.35434175,0.36414567,0.32352942,0.31652662,0.37535015,0.3669468,0.3837535,0.42857143,0.43137255,0.43137255,0.464986,0.45938376,0.47058824,0.41456583,0.4789916,0.4257703,0.41456583,0.4467787,0.5322129,0.47338936,,0.464986,0.5154062,0.4579832,0.5966387,0.48459384,0.5126051,0.60504204,0.557423,0.50980395,0.5154062,0.56442577,0.5602241,0.5966387,0.5714286,0.5686275,0.6722689,0.6834734,0.6806723,0.6834734,0.70308125,0.6820728,0.64985996,0.6582633,0.7240896,0.72829133,0.6946779,0.71428573,0.78711486,0.78431374,0.75910366,0.7254902,0.8235294,0.81232494,0.837535,0.8319328,0.8655462,0.9607843
+XGBoost (tuned + ensembled),0.04761905,0.12184874,0.13725491,0.13725491,0.14845939,0.15686275,0.16526611,0.17927171,0.24369748,0.22128852,0.3277311,0.232493,0.22689076,0.25210086,0.24089636,0.34033614,0.34313726,0.21008404,0.31232494,0.3977591,0.30952382,0.37535015,0.4089636,0.44817927,0.4327731,0.41456583,0.4229692,0.44117647,0.43557423,0.45938376,0.46078432,0.41456583,0.4369748,0.49439776,0.4929972,0.53501403,,0.5042017,0.50700283,0.5546219,0.45658264,0.5028011,0.557423,0.50980395,0.5826331,0.5154062,0.5826331,0.5448179,0.5714286,0.53501403,0.5672269,0.6526611,0.6302521,0.67787117,0.67787117,0.6806723,0.70028013,0.8487395,0.8767507,0.6946779,0.68907565,0.7352941,0.6834734,0.85294116,0.85014004,0.7478992,0.79831934,0.8067227,0.8487395,0.9019608,0.8487395,0.8557423,0.9747899
+RealMLP (default),0.10644258,0.13165267,0.13725491,0.17927171,0.14005603,0.1764706,0.18767507,0.1904762,0.22969188,0.26050422,0.28851542,0.26610646,0.21428572,0.30812326,0.2787115,0.32492998,0.31652662,0.3697479,0.32492998,0.41036415,0.38235295,0.35014006,0.34453782,0.41456583,0.40616247,0.3837535,0.44537815,0.40056023,0.42016807,0.42717087,0.45378152,0.47058824,0.45938376,0.45378152,0.53781515,0.48459384,0.49579832,,0.5126051,0.55182076,0.49859944,0.5042017,0.5742297,0.5266107,0.53501403,0.5322129,0.5294118,0.5434174,0.5826331,0.5042017,0.65686274,0.63305324,0.63585436,0.64985996,0.6946779,0.6806723,0.7114846,0.6946779,0.70028013,0.71428573,0.7521008,0.75910366,0.81232494,0.74509805,0.7619048,0.767507,0.7647059,0.84313726,0.8039216,0.8067227,0.8879552,0.84313726,0.9691877
+EBM (default),0.1092437,0.15546219,0.15966387,0.1764706,0.16526611,0.20728292,0.19607843,0.21008404,0.24929972,0.24369748,0.32212886,0.27731094,0.28431374,0.23529412,0.28151262,0.35014006,0.33613446,0.29411766,0.34453782,0.37254903,0.34733894,0.41456583,0.41036415,0.39215687,0.4089636,0.41736695,0.4089636,0.42857143,0.40616247,0.4257703,0.4509804,0.35854343,0.4397759,0.45658264,0.40756303,0.5420168,0.4929972,0.48739496,,0.5602241,0.49579832,0.53501403,0.5546219,0.49859944,0.5322129,0.5266107,0.49579832,0.5294118,0.56582636,0.5322129,0.6106443,0.6722689,0.6386555,0.6442577,0.7226891,0.7338936,0.6834734,0.69187677,0.70028013,0.7521008,0.6834734,0.70308125,0.7254902,0.7815126,0.7927171,0.79831934,0.7226891,0.8627451,0.837535,0.82633054,0.84313726,0.8879552,0.9691877
+TabSTAR (tuned),0.22128852,0.2647059,0.24929972,0.31092438,0.2717087,0.3137255,0.31092438,0.31932774,0.33053222,0.32352942,0.30812326,0.36414567,0.37535015,0.33053222,0.36134455,0.38655463,0.4019608,0.3879552,0.3837535,0.42016807,0.4117647,0.3977591,0.4509804,0.42016807,0.45518208,0.43417367,0.40336135,0.44817927,0.4019608,0.41456583,0.41736695,0.43137255,0.4117647,0.44817927,0.44817927,0.40336135,0.44537815,0.44817927,0.4397759,,0.42016807,0.5014006,0.5980392,0.46218488,0.464986,0.4369748,0.4579832,0.44537815,0.45658264,0.5406163,0.5602241,0.627451,0.9131653,0.6232493,0.48179272,0.63305324,0.62184876,0.57703084,0.5602241,0.6582633,0.56582636,0.6302521,0.6806723,0.69187677,0.7254902,0.6862745,0.719888,0.78711486,0.8095238,0.8039216,0.6862745,0.8487395,0.9411765
+ChimeraBoost (tuned + ensembled),0.070028014,0.09803922,0.12044818,0.11764706,0.10364146,0.10364146,0.13305323,0.15266107,0.232493,0.1904762,0.3137255,0.25630254,0.22969188,0.25910366,0.22128852,0.31092438,0.32492998,0.35154063,0.26050422,0.3417367,0.4089636,0.38655463,0.4117647,0.38515407,0.40056023,0.3809524,0.38655463,0.40616247,0.44117647,0.4229692,0.4369748,0.47338936,0.47478992,0.48459384,0.50980395,0.5154062,0.5434174,0.5014006,0.5042017,0.57983196,,0.49859944,0.57703084,0.48879552,0.58963585,0.5826331,0.5686275,0.6862745,0.56302524,0.5546219,0.5882353,0.627451,0.6442577,0.6414566,0.68907565,0.64985996,0.73109245,0.77030814,0.78571427,0.6666667,0.69187677,0.75910366,0.66386557,0.75630254,0.767507,0.71428573,0.7507003,0.77030814,0.80112046,0.8095238,0.8459384,0.8319328,0.96498597
+OrionMSP (default),0.070028014,0.12044818,0.09243698,0.1162465,0.10644258,0.18207283,0.14845939,0.15406163,0.17086835,0.24229692,0.1904762,0.37114847,0.3557423,0.34733894,0.37535015,0.26330534,0.24929972,0.394958,0.4089636,0.28011206,0.41736695,0.4439776,0.42016807,0.43417367,0.4257703,0.46218488,0.41456583,0.45518208,0.48739496,0.46778712,0.4929972,0.43417367,0.49579832,0.5154062,0.4719888,0.48739496,0.49719888,0.49579832,0.464986,0.49859944,0.5014006,,0.50980395,0.54901963,0.5294118,0.5126051,0.6022409,0.52100843,0.57983196,0.6022409,0.6246499,0.5966387,0.63305324,0.5994398,0.6302521,0.69187677,0.6806723,0.63585436,0.67787117,0.7212885,0.697479,0.70028013,0.7366947,0.66386557,0.70308125,0.7478992,0.7759104,0.73109245,0.7619048,0.72829133,0.8347339,0.7787115,0.96638656
+TabSTAR (tuned + ensembled),0.2184874,0.24929972,0.25210086,0.30812326,0.26750702,0.2997199,0.30252102,0.30812326,0.31932774,0.3277311,0.29411766,0.3529412,0.36134455,0.32492998,0.3557423,0.3837535,0.40336135,0.3837535,0.37955183,0.41456583,0.40616247,0.3837535,0.45658264,0.41736695,0.45658264,0.4257703,0.394958,0.4439776,0.394958,0.40616247,0.4229692,0.43137255,0.40616247,0.43837535,0.44117647,0.394958,0.44257703,0.4257703,0.44537815,0.4019608,0.4229692,0.49019608,,0.47058824,0.4579832,0.4397759,0.46778712,0.44257703,0.4509804,0.53501403,0.54901963,0.6022409,0.93557423,0.5994398,0.48739496,0.627451,0.61904764,0.56582636,0.5602241,0.6582633,0.56302524,0.6246499,0.6806723,0.6834734,0.71708685,0.6834734,0.7114846,0.77310926,0.80112046,0.79831934,0.6834734,0.8347339,0.9411765
+FastaiMLP (tuned),0.084033616,0.13865547,0.12885155,0.16526611,0.12885155,0.17927171,0.15686275,0.1764706,0.18767507,0.24789916,0.232493,0.26610646,0.27310926,0.29691878,0.30252102,0.27450982,0.29411766,0.36134455,0.33333334,0.32212886,0.4117647,0.34733894,0.34453782,0.36414567,0.3697479,0.3809524,0.43137255,0.3739496,0.4257703,0.34453782,0.44537815,0.48039216,0.44117647,0.39355743,0.5,0.44257703,0.49019608,0.47338936,0.5014006,0.53781515,0.5112045,0.4509804,0.5294118,,0.48739496,0.5308123,0.5126051,0.5434174,0.5434174,0.52380955,0.6246499,0.6106443,0.6106443,0.6022409,0.6526611,0.67787117,0.6834734,0.6386555,0.6386555,0.6960784,0.7240896,0.72829133,0.8095238,0.7184874,0.7226891,0.71708685,0.719888,0.8067227,0.75910366,0.77310926,0.84033614,0.8319328,0.9579832
+XGBoost (tuned),0.04481793,0.10084034,0.12044818,0.12885155,0.12044818,0.14005603,0.13865547,0.16806723,0.232493,0.19887955,0.28851542,0.23809524,0.20168068,0.232493,0.19887955,0.29691878,0.3277311,0.18487395,0.29411766,0.37254903,0.2577031,0.33753502,0.39215687,0.4159664,0.4117647,0.39635855,0.37535015,0.39215687,0.41316527,0.4229692,0.41316527,0.41456583,0.40616247,0.47058824,0.464986,0.49019608,0.41736695,0.464986,0.46778712,0.53501403,0.41036415,0.47058824,0.5420168,0.5126051,,0.47619048,0.56302524,0.4929972,0.5406163,0.5266107,0.55182076,0.6540616,0.6246499,0.6694678,0.6722689,0.66106445,0.66106445,0.7927171,0.8207283,0.69187677,0.6554622,0.719888,0.67787117,0.82633054,0.8347339,0.7366947,0.78711486,0.80812323,0.8459384,0.8767507,0.84313726,0.8515406,0.9691877
+ChimeraBoost (tuned),0.05602241,0.08963586,0.09803922,0.1092437,0.09803922,0.0952381,0.11764706,0.13165267,0.21008404,0.16806723,0.27450982,0.23809524,0.21428572,0.232493,0.17366947,0.28011206,0.30532214,0.31652662,0.26330534,0.35014006,0.3809524,0.3417367,0.3837535,0.36414567,0.3837535,0.3767507,0.3669468,0.40616247,0.394958,0.40336135,0.4117647,0.4397759,0.43557423,0.4397759,0.47478992,0.48459384,0.48459384,0.46778712,0.47338936,0.56302524,0.41736695,0.48739496,0.5602241,0.46918768,0.52380955,,0.52240896,0.5966387,0.52380955,0.5266107,0.5546219,0.59383756,0.63585436,0.61904764,0.64705884,0.61904764,0.6680672,0.732493,0.73109245,0.627451,0.67787117,0.70168066,0.64705884,0.7254902,0.72829133,0.68907565,0.719888,0.7394958,0.789916,0.7759104,0.83613443,0.7955182,0.94957983
+iLTM (tuned),0.064425774,0.082633056,0.11204482,0.10364146,0.09383754,0.11204482,0.12605043,0.13445379,0.16946779,0.17366947,0.2044818,0.21568628,0.22689076,0.29691878,0.2464986,0.22408964,0.24929972,0.2997199,0.32492998,0.32492998,0.33613446,0.3557423,0.3879552,0.3669468,0.3907563,0.3669468,0.40756303,0.28151262,0.35014006,0.43137255,0.3529412,0.4397759,0.38935575,0.4439776,0.47338936,0.43557423,0.41736695,0.47058824,0.5042017,0.5420168,0.43137255,0.3977591,0.5322129,0.48739496,0.4369748,0.47759104,,0.48459384,0.5434174,0.49859944,0.5420168,0.5686275,0.60784316,0.5854342,0.64565825,0.57563025,0.627451,0.62184876,0.62184876,0.6106443,0.72829133,0.68907565,0.7226891,0.6862745,0.697479,0.6666667,0.7647059,0.7366947,0.7254902,0.7535014,0.7955182,0.7913165,0.94957983
+ChimeraBoost (default),0.05042017,0.10084034,0.1092437,0.11204482,0.0952381,0.09243698,0.11764706,0.12885155,0.1904762,0.15966387,0.22689076,0.22408964,0.21288516,0.21568628,0.16806723,0.25490198,0.27731094,0.21568628,0.2254902,0.33613446,0.26610646,0.31092438,0.35854343,0.37254903,0.38935575,0.3417367,0.36134455,0.35714287,0.3557423,0.41456583,0.37535015,0.394958,0.38655463,0.4257703,0.4369748,0.4397759,0.45518208,0.45658264,0.47058824,0.5546219,0.3137255,0.4789916,0.557423,0.45658264,0.50700283,0.40336135,0.5154062,,0.50980395,0.52380955,0.54901963,0.60504204,0.6246499,0.6106443,0.64985996,0.63585436,0.6386555,0.7212885,0.72689074,0.6554622,0.6554622,0.66106445,0.63305324,0.7114846,0.72829133,0.7086835,0.7226891,0.732493,0.75630254,0.7647059,0.7927171,0.8179272,0.9579832
+TorchMLP (tuned),0.039215688,0.072829135,0.078431375,0.09243698,0.075630255,0.0952381,0.10364146,0.10364146,0.18487395,0.18767507,0.23529412,0.2044818,0.24369748,0.25210086,0.23109244,0.26050422,0.27450982,0.3137255,0.28711486,0.30812326,0.3347339,0.31652662,0.31092438,0.36134455,0.3809524,0.30252102,0.3837535,0.35014006,0.3557423,0.394958,0.3669468,0.4187675,0.3767507,0.2927171,0.44257703,0.40336135,0.42857143,0.41736695,0.43417367,0.5434174,0.4369748,0.42016807,0.54901963,0.45658264,0.45938376,0.47619048,0.45658264,0.49019608,,0.48739496,0.6106443,0.5980392,0.6246499,0.5742297,0.6386555,0.6134454,0.6386555,0.6106443,0.6092437,0.63165265,0.74369746,0.6806723,0.71428573,0.6806723,0.69187677,0.72829133,0.7338936,0.7605042,0.7226891,0.73109245,0.81232494,0.767507,0.95238096
+ModernNCA (tuned + ensembled),0.04481793,0.081232496,0.086834736,0.072829135,0.12605043,0.070028014,0.11204482,0.12184874,0.23809524,0.16246499,0.26610646,0.2857143,0.29411766,0.3137255,0.26330534,0.23809524,0.26190478,0.3417367,0.30812326,0.34733894,0.37535015,0.38935575,0.3837535,0.43137255,0.3697479,0.38655463,0.464986,0.38655463,0.41456583,0.42016807,0.4089636,0.45378152,0.4089636,0.42857143,0.4859944,0.42857143,0.464986,0.49579832,0.46778712,0.45938376,0.44537815,0.3977591,0.464986,0.47619048,0.47338936,0.47338936,0.5014006,0.47619048,0.5126051,,0.5266107,0.52100843,0.5266107,0.54901963,0.5826331,0.5322129,0.5966387,0.6554622,0.6386555,0.53501403,0.60504204,0.6022409,0.63305324,0.60784316,0.6134454,0.5826331,0.77170867,0.66386557,0.71708685,0.71428573,0.7226891,0.70308125,0.9159664
+TabFlex (default),0.08963586,0.11764706,0.12605043,0.14565827,0.11904762,0.14845939,0.15406163,0.16806723,0.12885155,0.26330534,0.13725491,0.26050422,0.28851542,0.28431374,0.2857143,0.24089636,0.2464986,0.35854343,0.30532214,0.22408964,0.37535015,0.31232494,0.30252102,0.27731094,0.30252102,0.32492998,0.33053222,0.32492998,0.41736695,0.3137255,0.4369748,0.3669468,0.44817927,0.3669468,0.4047619,0.43137255,0.4327731,0.34313726,0.38935575,0.4397759,0.4117647,0.37535015,0.4509804,0.37535015,0.44817927,0.44537815,0.4579832,0.4509804,0.38935575,0.47338936,,0.5182073,0.5686275,0.5182073,0.55182076,0.56582636,0.59383756,0.557423,0.557423,0.6064426,0.57563025,0.627451,0.61904764,0.60784316,0.6246499,0.63585436,0.6414566,0.6862745,0.66106445,0.697479,0.7058824,0.7366947,0.9187675
+ExtraTrees (tuned + ensembled),0.05602241,0.067226894,0.10504202,0.09243698,0.11204482,0.10084034,0.13445379,0.13865547,0.13025211,0.1764706,0.17366947,0.24369748,0.19887955,0.18207283,0.20728292,0.19607843,0.23809524,0.23109244,0.2717087,0.22689076,0.25630254,0.29131654,0.32492998,0.28291318,0.30812326,0.34733894,0.31932774,0.3207283,0.2717087,0.33893558,0.28011206,0.28991598,0.2997199,0.3529412,0.33613446,0.3277311,0.34733894,0.3669468,0.3277311,0.37254903,0.37254903,0.40336135,0.3977591,0.38935575,0.34593838,0.40616247,0.43137255,0.394958,0.4019608,0.4789916,0.48179272,,0.5728291,0.6106443,0.49859944,0.57843137,0.5434174,0.50700283,0.5126051,0.5980392,0.54901963,0.557423,0.5994398,0.71568626,0.74509805,0.66106445,0.66386557,0.80112046,0.8487395,0.8039216,0.66386557,0.8319328,0.9859944
+TabSTAR (default),0.17927171,0.20728292,0.20168068,0.24369748,0.22969188,0.2577031,0.2577031,0.25490198,0.2577031,0.26750702,0.24089636,0.29411766,0.30812326,0.25910366,0.30252102,0.29411766,0.2997199,0.29411766,0.3277311,0.32492998,0.31652662,0.32492998,0.35854343,0.34453782,0.3557423,0.33333334,0.31932774,0.34453782,0.33333334,0.3417367,0.34733894,0.33333334,0.33893558,0.3417367,0.35854343,0.31652662,0.3697479,0.36414567,0.36134455,0.086834736,0.3557423,0.3669468,0.064425774,0.38935575,0.37535015,0.36414567,0.39215687,0.37535015,0.37535015,0.47338936,0.43137255,0.42717087,,0.43137255,0.4229692,0.49859944,0.52380955,0.46778712,0.48179272,0.5154062,0.45378152,0.52100843,0.50700283,0.50560224,0.53501403,0.5826331,0.63305324,0.6666667,0.64705884,0.60504204,0.5882353,0.7394958,0.93557423
+ExtraTrees (tuned),0.067226894,0.072829135,0.10644258,0.11204482,0.11204482,0.11344538,0.14285715,0.14005603,0.10364146,0.18767507,0.17927171,0.23529412,0.20588236,0.18767507,0.23529412,0.19607843,0.22969188,0.25210086,0.27731094,0.22689076,0.26750702,0.29131654,0.32492998,0.26890758,0.28011206,0.33893558,0.31652662,0.30812326,0.2857143,0.34733894,0.28431374,0.30812326,0.3137255,0.36134455,0.30532214,0.31932774,0.32212886,0.35014006,0.3557423,0.3767507,0.35854343,0.40056023,0.40056023,0.3977591,0.33053222,0.3809524,0.41456583,0.38935575,0.4257703,0.4509804,0.48179272,0.38935575,0.5686275,,0.5126051,0.5714286,0.5154062,0.4789916,0.47338936,0.59383756,0.55182076,0.5462185,0.59103644,0.66106445,0.70028013,0.6554622,0.63305324,0.75630254,0.7759104,0.7661064,0.6414566,0.837535,0.9831933
+iLTM (default),0.072829135,0.084033616,0.09243698,0.0952381,0.08963586,0.1092437,0.1092437,0.11484594,0.16806723,0.14005603,0.19887955,0.18207283,0.15406163,0.14005603,0.16806723,0.1764706,0.21008404,0.17927171,0.1904762,0.25210086,0.20308124,0.21288516,0.28011206,0.22969188,0.26610646,0.25210086,0.23809524,0.2577031,0.19327731,0.27731094,0.24229692,0.24929972,0.26610646,0.27591038,0.29411766,0.31652662,0.32212886,0.30532214,0.27731094,0.5182073,0.31092438,0.3697479,0.5126051,0.34733894,0.3277311,0.3529412,0.35434175,0.35014006,0.36134455,0.41736695,0.44817927,0.5014006,0.57703084,0.48739496,,0.5686275,0.47058824,0.47619048,0.47058824,0.5966387,0.53501403,0.49019608,0.54901963,0.6302521,0.6526611,0.6526611,0.6134454,0.7338936,0.6694678,0.7058824,0.6106443,0.8067227,0.92156863
+Linear (tuned + ensembled),0.086834736,0.12885155,0.1092437,0.15406163,0.14005603,0.15406163,0.14005603,0.15966387,0.16806723,0.1904762,0.18767507,0.20028012,0.21288516,0.14565827,0.18767507,0.24369748,0.2394958,0.2184874,0.21568628,0.2577031,0.232493,0.29411766,0.26890758,0.24089636,0.27731094,0.3137255,0.26610646,0.3417367,0.29691878,0.27450982,0.30252102,0.2577031,0.30812326,0.30952382,0.2857143,0.29691878,0.31932774,0.31932774,0.26610646,0.3669468,0.35014006,0.30812326,0.37254903,0.32212886,0.33893558,0.3809524,0.42436975,0.36414567,0.38655463,0.46778712,0.43417367,0.42156863,0.5014006,0.42857143,0.43137255,,0.4789916,0.45938376,0.4397759,0.6386555,0.47619048,0.5014006,0.5014006,0.4789916,0.49019608,0.719888,0.5602241,0.6246499,0.59103644,0.5546219,0.59383756,0.70028013,0.8935574
+PerpetualBooster (tuned + ensembled),0.022408964,0.078431375,0.070028014,0.075630255,0.078431375,0.081232496,0.075630255,0.09943978,0.13725491,0.12605043,0.15406163,0.0952381,0.10084034,0.1764706,0.13165267,0.17086835,0.1904762,0.14845939,0.13725491,0.22969188,0.16946779,0.17927171,0.23529412,0.26190478,0.22408964,0.25490198,0.26890758,0.19607843,0.26050422,0.22969188,0.2464986,0.26890758,0.26330534,0.24929972,0.29411766,0.31792718,0.2997199,0.28851542,0.31652662,0.37815127,0.26890758,0.31932774,0.3809524,0.31652662,0.33893558,0.33193278,0.37254903,0.36134455,0.36134455,0.40336135,0.40616247,0.45658264,0.47619048,0.48459384,0.5294118,0.52100843,,0.5602241,0.5602241,0.53501403,0.50700283,0.6512605,0.49579832,0.557423,0.5826331,0.60504204,0.6442577,0.65966386,0.6834734,0.70308125,0.7647059,0.7226891,0.9411765
+XGBoost (default),0.030812325,0.05042017,0.072829135,0.072829135,0.075630255,0.05602241,0.078431375,0.084033616,0.13165267,0.12044818,0.18207283,0.10644258,0.11764706,0.16526611,0.12044818,0.17927171,0.21288516,0.11484594,0.17366947,0.26330534,0.15826331,0.2184874,0.27450982,0.2997199,0.232493,0.26330534,0.3137255,0.2044818,0.19607843,0.29131654,0.18487395,0.24789916,0.16806723,0.27450982,0.2717087,0.35014006,0.15126051,0.30532214,0.30812326,0.4229692,0.22969188,0.36414567,0.43417367,0.36134455,0.20728292,0.26750702,0.37815127,0.2787115,0.38935575,0.34453782,0.44257703,0.4929972,0.5322129,0.52100843,0.52380955,0.5406163,0.4397759,,0.50560224,0.5546219,0.5266107,0.49579832,0.55182076,0.57983196,0.6134454,0.627451,0.6806723,0.6806723,0.719888,0.7394958,0.6414566,0.7366947,0.92436975
+LightGBM (default),0.033613447,0.05322129,0.05882353,0.064425774,0.042016808,0.05042017,0.06162465,0.067226894,0.11764706,0.09803922,0.16386555,0.081232496,0.1092437,0.14425771,0.1162465,0.15126051,0.19607843,0.05602241,0.17086835,0.26050422,0.09663866,0.19607843,0.2717087,0.25490198,0.21288516,0.25210086,0.28011206,0.15406163,0.16526611,0.28991598,0.15686275,0.23529412,0.1764706,0.30252102,0.25490198,0.3417367,0.1232493,0.2997199,0.2997199,0.4397759,0.21428572,0.32212886,0.4397759,0.36134455,0.17927171,0.26890758,0.37815127,0.27310926,0.3907563,0.36134455,0.44257703,0.48739496,0.5182073,0.5266107,0.5294118,0.5602241,0.4397759,0.49439776,,0.5714286,0.557423,0.5042017,0.54901963,0.5854342,0.61484593,0.63305324,0.68907565,0.6806723,0.7086835,0.7240896,0.6722689,0.7478992,0.94817924
+Linear (tuned),0.075630255,0.11764706,0.09803922,0.12044818,0.12605043,0.1232493,0.11764706,0.14005603,0.14845939,0.16246499,0.16526611,0.16666667,0.20728292,0.11764706,0.17226891,0.2184874,0.21568628,0.2044818,0.19887955,0.24089636,0.21428572,0.28431374,0.26330534,0.21288516,0.2577031,0.29131654,0.24789916,0.3137255,0.27450982,0.24929972,0.28011206,0.2464986,0.2927171,0.2717087,0.2717087,0.27591038,0.30532214,0.2857143,0.24789916,0.3417367,0.33333334,0.2787115,0.3417367,0.30392158,0.30812326,0.37254903,0.38935575,0.34453782,0.36834735,0.464986,0.39355743,0.4019608,0.48459384,0.40616247,0.40336135,0.36134455,0.464986,0.44537815,0.42857143,,0.45658264,0.48739496,0.48459384,0.47338936,0.47619048,0.63305324,0.557423,0.57843137,0.5686275,0.5406163,0.59103644,0.67787117,0.88235295
+TorchMLP (default),0.028011205,0.05602241,0.04761905,0.064425774,0.036414567,0.06162465,0.081232496,0.072829135,0.09803922,0.14005603,0.09243698,0.13725491,0.12885155,0.17366947,0.12885155,0.14005603,0.15126051,0.22689076,0.14845939,0.16806723,0.23809524,0.17366947,0.232493,0.17086835,0.18767507,0.25210086,0.19327731,0.22408964,0.24089636,0.1834734,0.26330534,0.33053222,0.28291318,0.16246499,0.35014006,0.2717087,0.31092438,0.24789916,0.31652662,0.43417367,0.30812326,0.30252102,0.4369748,0.27591038,0.34453782,0.32212886,0.2717087,0.34453782,0.25630254,0.394958,0.42436975,0.4509804,0.5462185,0.44817927,0.464986,0.52380955,0.4929972,0.47338936,0.44257703,0.5434174,,0.53501403,0.59383756,0.5266107,0.56582636,0.6134454,0.6022409,0.6806723,0.62184876,0.6162465,0.67507005,0.73109245,0.9271709
+PerpetualBooster (tuned),0.019607844,0.05882353,0.05882353,0.06022409,0.064425774,0.081232496,0.070028014,0.078431375,0.1092437,0.1092437,0.15966387,0.08963586,0.08963586,0.15686275,0.1092437,0.14985995,0.17927171,0.12044818,0.1232493,0.2044818,0.14845939,0.17086835,0.2254902,0.20588236,0.16946779,0.22408964,0.23529412,0.16806723,0.21288516,0.20168068,0.2184874,0.24229692,0.22969188,0.2044818,0.26890758,0.30532214,0.2647059,0.24089636,0.29691878,0.3697479,0.24089636,0.2997199,0.37535015,0.2717087,0.28011206,0.29831934,0.31092438,0.33893558,0.31932774,0.3977591,0.37254903,0.44257703,0.4789916,0.45378152,0.50980395,0.49859944,0.3487395,0.5042017,0.49579832,0.5126051,0.464986,,0.464986,0.52100843,0.53781515,0.5826331,0.6204482,0.6302521,0.66106445,0.6722689,0.69887954,0.6946779,0.90756303
+FastaiMLP (default),0.04761905,0.05602241,0.064425774,0.075630255,0.072829135,0.08963586,0.081232496,0.0952381,0.084033616,0.11204482,0.10644258,0.1092437,0.10084034,0.19607843,0.13165267,0.11484594,0.13445379,0.22969188,0.19327731,0.15126051,0.24089636,0.14145659,0.22128852,0.16526611,0.15126051,0.22829132,0.21008404,0.22689076,0.24929972,0.070028014,0.26330534,0.2464986,0.27450982,0.20028012,0.2857143,0.2857143,0.31652662,0.18767507,0.27450982,0.31932774,0.33613446,0.26330534,0.31932774,0.1904762,0.32212886,0.3529412,0.27731094,0.3669468,0.2857143,0.3669468,0.3809524,0.40056023,0.4929972,0.4089636,0.4509804,0.49859944,0.5042017,0.44817927,0.4509804,0.5154062,0.40616247,0.53501403,,0.49859944,0.49859944,0.54901963,0.5854342,0.66106445,0.6162465,0.5980392,0.67787117,0.72829133,0.929972
+RandomForest (tuned + ensembled),0.022408964,0.06162465,0.067226894,0.070028014,0.086834736,0.067226894,0.084033616,0.10084034,0.086834736,0.10644258,0.14565827,0.12885155,0.10364146,0.10644258,0.14845939,0.11764706,0.15966387,0.11484594,0.17366947,0.20728292,0.16246499,0.2044818,0.23109244,0.23529412,0.1764706,0.23809524,0.24089636,0.1904762,0.14565827,0.21708684,0.15406163,0.21008404,0.15406163,0.25490198,0.24089636,0.21288516,0.14705883,0.25490198,0.2184874,0.30812326,0.24369748,0.33613446,0.31652662,0.28151262,0.17366947,0.27450982,0.3137255,0.28851542,0.31932774,0.39215687,0.39215687,0.28431374,0.49439776,0.33893558,0.3697479,0.52100843,0.44257703,0.42016807,0.41456583,0.5266107,0.47338936,0.4789916,0.5014006,,0.6106443,0.60084033,0.6414566,0.66106445,0.70308125,0.78431374,0.5686275,0.77310926,0.9551821
+RandomForest (tuned),0.025210084,0.039215688,0.070028014,0.064425774,0.05602241,0.05322129,0.081232496,0.08963586,0.084033616,0.084033616,0.12885155,0.11344538,0.078431375,0.11484594,0.1232493,0.11484594,0.14285715,0.084033616,0.17927171,0.17927171,0.12605043,0.17927171,0.22268908,0.19607843,0.15406163,0.22408964,0.21568628,0.16806723,0.13725491,0.21008404,0.12745099,0.20168068,0.15126051,0.232493,0.21288516,0.21568628,0.14985995,0.23809524,0.20728292,0.27450982,0.232493,0.29691878,0.28291318,0.27731094,0.16526611,0.2717087,0.30252102,0.2717087,0.30812326,0.38655463,0.37535015,0.25490198,0.464986,0.2997199,0.34733894,0.50980395,0.41736695,0.38655463,0.38515407,0.52380955,0.43417367,0.46218488,0.5014006,0.38935575,,0.5966387,0.6246499,0.640056,0.6694678,0.69187677,0.54901963,0.7647059,0.9579832
+Linear (default),0.05602241,0.072829135,0.068627454,0.09103642,0.067226894,0.08963586,0.10504202,0.09383754,0.084033616,0.15406163,0.12745099,0.13165267,0.18207283,0.0952381,0.14285715,0.17086835,0.17086835,0.14565827,0.16806723,0.1904762,0.16526611,0.22689076,0.22969188,0.18487395,0.21708684,0.24929972,0.1904762,0.26890758,0.21568628,0.22969188,0.23809524,0.20728292,0.24929972,0.2184874,0.24369748,0.24089636,0.25210086,0.232493,0.20168068,0.3137255,0.2857143,0.25210086,0.31652662,0.28291318,0.26330534,0.31092438,0.33333334,0.29131654,0.2717087,0.41736695,0.36414567,0.33893558,0.41736695,0.34453782,0.34733894,0.28011206,0.394958,0.37254903,0.3669468,0.3669468,0.38655463,0.41736695,0.4509804,0.39915967,0.40336135,,0.5294118,0.49859944,0.5014006,0.5042017,0.5140056,0.5966387,0.8235294
+xRFM (default),0.019607844,0.036414567,0.039215688,0.070028014,0.04481793,0.05602241,0.070028014,0.072829135,0.064425774,0.081232496,0.1092437,0.086834736,0.09243698,0.17927171,0.11764706,0.082633056,0.12044818,0.1764706,0.16806723,0.15126051,0.19187675,0.15686275,0.16526611,0.1232493,0.081232496,0.14565827,0.24089636,0.1092437,0.20728292,0.1764706,0.21008404,0.23809524,0.2184874,0.1904762,0.25490198,0.27450982,0.20168068,0.23529412,0.27731094,0.28011206,0.24929972,0.22408964,0.28851542,0.28011206,0.21288516,0.28011206,0.23529412,0.27731094,0.26610646,0.22829132,0.35854343,0.33613446,0.3669468,0.3669468,0.38655463,0.4397759,0.3557423,0.31932774,0.31092438,0.44257703,0.3977591,0.37955183,0.41456583,0.35854343,0.37535015,0.47058824,,0.45658264,0.4789916,0.45938376,0.50980395,0.5434174,0.8655462
+KNN (tuned + ensembled),0.022408964,0.05882353,0.05322129,0.086834736,0.086834736,0.075630255,0.086834736,0.08963586,0.081232496,0.12605043,0.12885155,0.10364146,0.11484594,0.086834736,0.13165267,0.12464986,0.12605043,0.11764706,0.15126051,0.11484594,0.14565827,0.18207283,0.22128852,0.15406163,0.14985995,0.22128852,0.14005603,0.1904762,0.13725491,0.14285715,0.13725491,0.13725491,0.14285715,0.1904762,0.15406163,0.1764706,0.19327731,0.15686275,0.13725491,0.21288516,0.22969188,0.26890758,0.22689076,0.19327731,0.19187675,0.26050422,0.26330534,0.26750702,0.2394958,0.33613446,0.3137255,0.19887955,0.33333334,0.24369748,0.26610646,0.37535015,0.34033614,0.31932774,0.31932774,0.42156863,0.31932774,0.3697479,0.33893558,0.33893558,0.359944,0.5014006,0.5434174,,0.45938376,0.47058824,0.47058824,0.81092435,0.9607843
+ExtraTrees (default),0.016806724,0.036414567,0.05882353,0.06162465,0.064425774,0.067226894,0.067226894,0.075630255,0.065826334,0.075630255,0.12044818,0.075630255,0.075630255,0.13165267,0.0952381,0.086834736,0.12605043,0.09243698,0.13725491,0.12044818,0.11204482,0.18207283,0.17927171,0.15406163,0.09243698,0.1764706,0.18907563,0.14565827,0.14845939,0.18207283,0.13165267,0.15406163,0.14005603,0.23809524,0.15126051,0.18767507,0.15126051,0.19607843,0.16246499,0.1904762,0.19887955,0.23809524,0.19887955,0.24089636,0.15406163,0.21008404,0.27450982,0.24369748,0.27731094,0.28291318,0.33893558,0.15126051,0.3529412,0.22408964,0.33053222,0.4089636,0.31652662,0.28011206,0.29131654,0.43137255,0.37815127,0.33893558,0.3837535,0.29691878,0.33053222,0.49859944,0.52100843,0.5406163,,0.5406163,0.4369748,0.64705884,0.9467787
+RandomForest (default),0.025210084,0.030812325,0.05322129,0.05882353,0.04761905,0.04761905,0.067226894,0.070028014,0.067226894,0.064425774,0.10644258,0.078431375,0.04761905,0.09803922,0.0952381,0.08963586,0.11204482,0.05882353,0.12885155,0.11764706,0.08963586,0.14285715,0.17086835,0.14005603,0.075630255,0.18207283,0.19887955,0.11484594,0.12044818,0.16246499,0.086834736,0.14565827,0.09803922,0.22128852,0.15686275,0.16246499,0.09803922,0.19327731,0.17366947,0.19607843,0.1904762,0.2717087,0.20168068,0.22689076,0.1232493,0.22408964,0.2464986,0.23529412,0.26890758,0.2857143,0.30252102,0.19607843,0.394958,0.23389356,0.29411766,0.44537815,0.29691878,0.26050422,0.27591038,0.45938376,0.3837535,0.3277311,0.4019608,0.21568628,0.30812326,0.49579832,0.5406163,0.5294118,0.45938376,,0.44537815,0.6386555,0.9271709
+PerpetualBooster (default),0.014005602,0.036414567,0.025210084,0.04481793,0.036414567,0.04761905,0.04761905,0.04481793,0.042016808,0.075630255,0.09243698,0.067226894,0.042016808,0.072829135,0.042016808,0.078431375,0.081232496,0.067226894,0.05322129,0.12605043,0.070028014,0.064425774,0.18207283,0.10364146,0.081232496,0.14285715,0.15966387,0.11204482,0.13165267,0.075630255,0.14005603,0.11764706,0.12885155,0.11484594,0.13725491,0.16806723,0.15126051,0.11204482,0.15686275,0.3137255,0.15406163,0.16526611,0.31652662,0.15966387,0.15686275,0.16386555,0.2044818,0.20728292,0.18767507,0.27731094,0.29411766,0.33613446,0.4117647,0.35854343,0.38935575,0.40616247,0.23529412,0.35854343,0.3277311,0.4089636,0.32492998,0.30112046,0.32212886,0.43137255,0.4509804,0.4859944,0.49019608,0.5294118,0.56302524,0.5546219,,0.6302521,0.89915967
+KNN (tuned),0.033613447,0.04481793,0.05042017,0.067226894,0.067226894,0.064425774,0.084033616,0.081232496,0.05882353,0.10084034,0.067226894,0.10084034,0.09243698,0.06162465,0.11204482,0.11204482,0.12044818,0.09663866,0.14285715,0.08963586,0.11204482,0.15686275,0.14845939,0.13165267,0.13725491,0.1764706,0.12885155,0.14705883,0.10644258,0.12605043,0.12605043,0.12464986,0.11764706,0.15406163,0.1232493,0.13445379,0.14425771,0.15686275,0.11204482,0.15126051,0.16806723,0.22128852,0.16526611,0.16806723,0.14845939,0.2044818,0.20868348,0.18207283,0.232493,0.29691878,0.26330534,0.16806723,0.26050422,0.16246499,0.19327731,0.2997199,0.27731094,0.26330534,0.25210086,0.32212886,0.26890758,0.30532214,0.2717087,0.22689076,0.23529412,0.40336135,0.45658264,0.18907563,0.3529412,0.36134455,0.3697479,,0.9411765
+KNN (default),0.008403362,0.005602241,0.005602241,0.008403362,0.008403362,0.011204482,0.011204482,0.011204482,0.008403362,0.014005602,0.008403362,0.005602241,0.008403362,0.019607844,0.022408964,0.008403362,0.014005602,0.011204482,0.033613447,0.011204482,0.008403362,0.025210084,0.025210084,0.014005602,0.014005602,0.028011205,0.022408964,0.030812325,0.016806724,0.030812325,0.011204482,0.028011205,0.011204482,0.033613447,0.028011205,0.039215688,0.025210084,0.030812325,0.030812325,0.05882353,0.035014007,0.033613447,0.05882353,0.042016808,0.030812325,0.05042017,0.05042017,0.042016808,0.04761905,0.084033616,0.081232496,0.014005602,0.064425774,0.016806724,0.078431375,0.10644258,0.05882353,0.075630255,0.05182073,0.11764706,0.072829135,0.09243698,0.070028014,0.04481793,0.042016808,0.1764706,0.13445379,0.039215688,0.05322129,0.072829135,0.10084034,0.05882353,
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_all/leaderboard_overview_explorer.html
index 855148d86e8a57b26e9c60dbdfa909c193163458..d81b359deb953079f897fc72ea74e5fee5b6beae 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Classification | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_all/leaderboard_table.html b/data/imputation_no/splits_all/tasks_classification/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..3fce8dae23c77f10d6a8a3e053b233173969889d
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_classification/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1753.0,"score":0.873,"rank":7.32,"hrank":1.72,"imp":5.013,"train_time":38.57,"predict_time":7.109,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-119","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1681.0,"score":0.646,"rank":9.86,"hrank":4.11,"imp":9.466,"train_time":267.31,"predict_time":3.981,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-78","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1646.0,"score":0.643,"rank":11.38,"hrank":4.53,"imp":11.799,"train_time":3.64,"predict_time":0.391,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-70","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1599.0,"score":0.56,"rank":13.67,"hrank":7.52,"imp":13.155,"train_time":5.17,"predict_time":0.539,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-63","imputed":false},{"position":4,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1585.0,"score":0.585,"rank":14.4,"hrank":5.33,"imp":12.406,"train_time":2.07,"predict_time":0.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-73","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1570.0,"score":0.517,"rank":15.23,"hrank":6.68,"imp":12.939,"train_time":2046.25,"predict_time":8.969,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-70","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1548.0,"score":0.496,"rank":16.55,"hrank":7.3,"imp":13.609,"train_time":2046.25,"predict_time":1.329,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-64","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1537.0,"score":0.488,"rank":17.22,"hrank":9.0,"imp":13.834,"train_time":5.76,"predict_time":0.794,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-60","imputed":false},{"position":8,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1501.0,"score":0.391,"rank":19.53,"hrank":7.91,"imp":15.058,"train_time":1684.65,"predict_time":1.969,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-54","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1482.0,"score":0.367,"rank":20.81,"hrank":12.62,"imp":15.66,"train_time":2879.46,"predict_time":12.484,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-42","imputed":false},{"position":10,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1462.0,"score":0.325,"rank":22.21,"hrank":11.46,"imp":16.637,"train_time":2168.86,"predict_time":2.347,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-47","imputed":false},{"position":11,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1434.0,"score":0.337,"rank":24.3,"hrank":10.19,"imp":17.107,"train_time":1.94,"predict_time":0.19,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-60","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1429.0,"score":0.266,"rank":24.65,"hrank":17.46,"imp":17.384,"train_time":382.09,"predict_time":1.487,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-34","imputed":false},{"position":13,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1411.0,"score":0.266,"rank":25.98,"hrank":15.09,"imp":17.034,"train_time":1107.0,"predict_time":0.329,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-47","imputed":false},{"position":14,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1410.0,"score":0.267,"rank":26.08,"hrank":13.24,"imp":17.493,"train_time":2168.86,"predict_time":0.258,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-50","imputed":false},{"position":15,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1408.0,"score":0.287,"rank":26.28,"hrank":13.0,"imp":17.152,"train_time":4941.98,"predict_time":307.735,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-50","imputed":false},{"position":16,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1405.0,"score":0.242,"rank":26.46,"hrank":17.88,"imp":17.435,"train_time":12597.42,"predict_time":502.448,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-43","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1405.0,"score":0.264,"rank":26.47,"hrank":14.15,"imp":17.101,"train_time":2879.46,"predict_time":0.598,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-44","imputed":false},{"position":18,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1403.0,"score":0.348,"rank":26.65,"hrank":8.76,"imp":17.383,"train_time":18.41,"predict_time":4.667,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-80","imputed":false},{"position":19,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1401.0,"score":0.252,"rank":26.81,"hrank":15.37,"imp":17.295,"train_time":1107.0,"predict_time":0.038,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-42","imputed":false},{"position":20,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1387.0,"score":0.248,"rank":27.85,"hrank":15.11,"imp":17.642,"train_time":5.41,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-42","imputed":false},{"position":21,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1387.0,"score":0.199,"rank":27.89,"hrank":21.05,"imp":18.063,"train_time":382.09,"predict_time":0.254,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-33","imputed":false},{"position":22,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1376.0,"score":0.211,"rank":28.76,"hrank":19.11,"imp":18.172,"train_time":685.87,"predict_time":1.454,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-46","imputed":false},{"position":23,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1371.0,"score":0.254,"rank":29.12,"hrank":16.93,"imp":19.46,"train_time":282.82,"predict_time":1.886,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-48","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1361.0,"score":0.187,"rank":29.96,"hrank":16.35,"imp":18.456,"train_time":4880.28,"predict_time":0.525,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-39","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1359.0,"score":0.243,"rank":30.08,"hrank":16.1,"imp":18.528,"train_time":4880.28,"predict_time":8.736,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-78","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1350.0,"score":0.169,"rank":30.84,"hrank":21.68,"imp":18.539,"train_time":685.87,"predict_time":0.205,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-41","imputed":false},{"position":27,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1350.0,"score":0.24,"rank":30.87,"hrank":18.38,"imp":19.196,"train_time":1.2,"predict_time":0.067,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-56","imputed":false},{"position":28,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1349.0,"score":0.221,"rank":30.89,"hrank":14.32,"imp":18.541,"train_time":4941.98,"predict_time":41.595,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-51","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1345.0,"score":0.209,"rank":31.26,"hrank":18.69,"imp":19.325,"train_time":470.37,"predict_time":0.436,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-61","imputed":false},{"position":30,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1326.0,"score":0.184,"rank":32.8,"hrank":20.49,"imp":19.387,"train_time":934.27,"predict_time":3.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-46","imputed":false},{"position":31,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1323.0,"score":0.177,"rank":33.06,"hrank":20.29,"imp":19.681,"train_time":470.37,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-59","imputed":false},{"position":32,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1313.0,"score":0.135,"rank":33.89,"hrank":24.65,"imp":20.189,"train_time":1499.05,"predict_time":0.15,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-37","imputed":false},{"position":33,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1313.0,"score":0.176,"rank":33.89,"hrank":20.68,"imp":19.684,"train_time":6.85,"predict_time":0.246,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-47","imputed":false},{"position":34,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1308.0,"score":0.137,"rank":34.29,"hrank":25.11,"imp":19.373,"train_time":2389.47,"predict_time":2.157,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-38","imputed":false},{"position":35,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1295.0,"score":0.202,"rank":35.32,"hrank":13.16,"imp":20.915,"train_time":13.7,"predict_time":2.033,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-71","imputed":false},{"position":36,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1294.0,"score":0.11,"rank":35.43,"hrank":22.76,"imp":19.927,"train_time":12597.42,"predict_time":70.851,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+41\/-39","imputed":false},{"position":37,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1288.0,"score":0.145,"rank":35.94,"hrank":22.14,"imp":20.725,"train_time":934.27,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-52","imputed":false},{"position":38,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1284.0,"score":0.148,"rank":36.25,"hrank":21.21,"imp":20.831,"train_time":1.86,"predict_time":0.035,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-64","imputed":false},{"position":39,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1277.0,"score":0.093,"rank":36.83,"hrank":27.19,"imp":20.869,"train_time":1499.05,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-40","imputed":false},{"position":40,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1270.0,"score":0.16,"rank":37.43,"hrank":19.08,"imp":21.298,"train_time":49.21,"predict_time":43.824,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-57","imputed":false},{"position":41,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1258.0,"score":0.132,"rank":38.36,"hrank":22.41,"imp":21.246,"train_time":618.96,"predict_time":4.766,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-62","imputed":false},{"position":42,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1252.0,"score":0.084,"rank":38.94,"hrank":26.39,"imp":21.818,"train_time":5.72,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-46","imputed":false},{"position":43,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1250.0,"score":0.087,"rank":39.06,"hrank":23.05,"imp":22.165,"train_time":14.78,"predict_time":0.346,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+42\/-41","imputed":false},{"position":44,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1239.0,"score":0.061,"rank":39.93,"hrank":29.19,"imp":21.418,"train_time":10.48,"predict_time":1.707,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+35\/-32","imputed":false},{"position":45,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1232.0,"score":0.067,"rank":40.56,"hrank":29.57,"imp":21.187,"train_time":1.77,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-50","imputed":false},{"position":46,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1228.0,"score":0.073,"rank":40.89,"hrank":30.62,"imp":21.409,"train_time":2389.47,"predict_time":0.152,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-36","imputed":false},{"position":47,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1213.0,"score":0.078,"rank":42.1,"hrank":28.16,"imp":22.199,"train_time":286.48,"predict_time":0.764,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-52","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1192.0,"score":0.058,"rank":43.73,"hrank":28.29,"imp":22.987,"train_time":618.96,"predict_time":0.298,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-61","imputed":false},{"position":49,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1190.0,"score":0.056,"rank":43.89,"hrank":36.19,"imp":22.153,"train_time":1.79,"predict_time":0.12,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-42","imputed":false},{"position":50,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1184.0,"score":0.074,"rank":44.38,"hrank":26.81,"imp":23.255,"train_time":323.8,"predict_time":0.743,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-60","imputed":false},{"position":51,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1184.0,"score":0.067,"rank":44.43,"hrank":26.51,"imp":23.178,"train_time":286.48,"predict_time":0.073,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-54","imputed":false},{"position":52,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1151.0,"score":0.053,"rank":46.98,"hrank":30.64,"imp":24.496,"train_time":262.93,"predict_time":64.506,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-58","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1149.0,"score":0.053,"rank":47.08,"hrank":26.01,"imp":24.038,"train_time":323.8,"predict_time":0.076,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-60","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1136.0,"score":0.129,"rank":48.12,"hrank":14.72,"imp":26.312,"train_time":29479.47,"predict_time":4.088,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-87","imputed":false},{"position":55,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1131.0,"score":0.124,"rank":48.48,"hrank":14.32,"imp":26.389,"train_time":29479.47,"predict_time":18.315,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-86","imputed":false},{"position":56,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1119.0,"score":0.076,"rank":49.4,"hrank":34.11,"imp":26.361,"train_time":16.32,"predict_time":2.859,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-64","imputed":false},{"position":57,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1091.0,"score":0.011,"rank":51.38,"hrank":38.45,"imp":25.539,"train_time":6.83,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-50","imputed":false},{"position":58,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1066.0,"score":0.028,"rank":53.15,"hrank":39.56,"imp":31.055,"train_time":187.03,"predict_time":0.955,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-60","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1044.0,"score":0.01,"rank":54.62,"hrank":37.85,"imp":27.71,"train_time":2.91,"predict_time":0.368,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-68","imputed":false},{"position":60,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1040.0,"score":0.042,"rank":54.83,"hrank":37.56,"imp":32.425,"train_time":252.6,"predict_time":0.528,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-86","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1031.0,"score":0.018,"rank":55.42,"hrank":39.86,"imp":32.509,"train_time":187.03,"predict_time":0.338,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-60","imputed":false},{"position":62,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1028.0,"score":0.084,"rank":55.58,"hrank":22.92,"imp":31.845,"train_time":439.45,"predict_time":5.883,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-106","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1023.0,"score":0.023,"rank":55.92,"hrank":37.04,"imp":30.242,"train_time":139.88,"predict_time":3.615,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-88","imputed":false},{"position":64,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1014.0,"score":0.051,"rank":56.44,"hrank":37.65,"imp":31.69,"train_time":1.11,"predict_time":0.189,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-91","imputed":false},{"position":65,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1014.0,"score":0.03,"rank":56.45,"hrank":39.24,"imp":33.096,"train_time":252.6,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-86","imputed":false},{"position":66,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1012.0,"score":0.023,"rank":56.61,"hrank":48.69,"imp":30.871,"train_time":3.61,"predict_time":0.938,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-86","imputed":false},{"position":67,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.006,"rank":57.3,"hrank":51.1,"imp":29.873,"train_time":0.38,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-62","imputed":false},{"position":68,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":994.0,"score":0.009,"rank":57.67,"hrank":50.94,"imp":29.874,"train_time":1.68,"predict_time":0.295,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-70","imputed":false},{"position":69,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":936.0,"score":0.018,"rank":60.76,"hrank":44.47,"imp":35.779,"train_time":1.29,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-98","imputed":false},{"position":70,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":928.0,"score":0.0,"rank":61.16,"hrank":44.45,"imp":36.617,"train_time":26.07,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-77","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":910.0,"score":0.011,"rank":62.0,"hrank":45.96,"imp":36.085,"train_time":139.88,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-108","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":591.0,"score":0.0,"rank":70.89,"hrank":66.03,"imp":51.385,"train_time":0.18,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-107","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_all/pareto_front_explorer.html
index 3a7dd616af49dbab0ca68e59266c29fc1c3c21f9..934173bda8564ecf805a4e27fc574cd9e31235b6 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8858780824,"imp":19.4603326324,"elo":1371.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0255239609,"imp":17.6418831081,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0377219356,"imp":17.2949223221,"elo":1400.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3286022328,"imp":17.0340780938,"elo":1411.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0349046894,"imp":20.8308869747,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0333867141,"imp":19.6814490305,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4361928292,"imp":19.3252875469,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0170265849,"imp":21.8177278863,"elo":1251.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0172938338,"imp":20.8692370358,"elo":1277.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1503678683,"imp":20.1892723153,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2946660708,"imp":29.8740838341,"elo":993.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0734711664,"imp":23.1777175457,"elo":1183.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7641423489,"imp":22.1993688405,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3681069516,"imp":27.7099980235,"elo":1043.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2978802219,"imp":22.9871709369,"elo":1192.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.7655686359,"imp":21.2460816448,"elo":1258.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1204998116,"imp":22.1526904401,"elo":1190.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.253630478,"imp":18.0633828427,"elo":1386.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.4869408199,"imp":17.3844330214,"elo":1428.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.6674634265,"imp":17.3834790986,"elo":1402.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"imp":22.1647731729,"elo":1249.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5246167968,"imp":18.4556324732,"elo":1360.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7364459708,"imp":18.5280660881,"elo":1359.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.8592422653,"imp":26.3607466183,"elo":1118.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0410553489,"imp":36.6174806611,"elo":927.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3381841475,"imp":32.5094386414,"elo":1030.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9547740596,"imp":31.0546372303,"elo":1065.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0372103395,"imp":29.8733488873,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0764179217,"imp":24.0381975853,"elo":1149.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7428875097,"imp":23.254536984,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7071575933,"imp":21.4181030792,"elo":1239.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5977456464,"imp":17.1008499815,"elo":1405.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":12.4836618956,"imp":15.6602492434,"elo":1481.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7935168864,"imp":13.8336294446,"elo":1536.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3289068866,"imp":13.6090643932,"elo":1547.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9691710732,"imp":12.939386077,"elo":1570.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0331410808,"imp":20.9149979551,"elo":1295.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"imp":21.2977853057,"elo":1269.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":41.5948089023,"imp":18.5407412411,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":307.7353108496,"imp":17.1517351572,"elo":1407.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1897902902,"imp":17.1066400134,"elo":1433.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"imp":5.0125041353,"elo":1753.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1888600498,"imp":31.6899617016,"elo":1014.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"imp":12.4055400385,"elo":1585.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2459026963,"imp":19.6843599668,"elo":1312.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2584066824,"imp":17.4933162976,"elo":1410.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.3471511219,"imp":16.6368059624,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"imp":13.1554131904,"elo":1598.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3910008352,"imp":11.7991829167,"elo":1645.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.8830380236,"imp":31.845298093,"elo":1028.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0878026987,"imp":26.3117913949,"elo":1135.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3145858894,"imp":26.3893153598,"elo":1130.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0665397304,"imp":19.1957679596,"elo":1349.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1470302095,"imp":25.5387551645,"elo":1091.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1517792148,"imp":21.4086879296,"elo":1227.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1570271396,"imp":19.373068261,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.117054282,"imp":21.1866289175,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2050537617,"imp":18.5394348142,"elo":1349.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.4544829944,"imp":18.1717495859,"elo":1375.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":64.5055925989,"imp":24.4956519569,"elo":1150.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":70.8512833361,"imp":19.9273628334,"elo":1294.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":502.4479719176,"imp":17.4353708182,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9382527357,"imp":30.8711599893,"elo":1011.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1687223142,"imp":20.7246325333,"elo":1287.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.0619945914,"imp":19.3874085581,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8858780824,"x_train":282.8166166302,"imp":19.4603326324,"elo":1371.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0255239609,"x_train":5.4086527682,"imp":17.6418831081,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0377219356,"x_train":1106.9971386445,"imp":17.2949223221,"elo":1400.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3286022328,"x_train":1106.9971386445,"imp":17.0340780938,"elo":1411.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0349046894,"x_train":1.857962315,"imp":20.8308869747,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0333867141,"x_train":470.3718666937,"imp":19.6814490305,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4361928292,"x_train":470.3718666937,"imp":19.3252875469,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0170265849,"x_train":5.7213239382,"imp":21.8177278863,"elo":1251.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0172938338,"x_train":1499.0498932557,"imp":20.8692370358,"elo":1277.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1503678683,"x_train":1499.0498932557,"imp":20.1892723153,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2946660708,"x_train":1.6841584145,"imp":29.8740838341,"elo":993.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0734711664,"x_train":286.4783954015,"imp":23.1777175457,"elo":1183.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7641423489,"x_train":286.4783954015,"imp":22.1993688405,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3681069516,"x_train":2.9123333331,"imp":27.7099980235,"elo":1043.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2978802219,"x_train":618.9598462323,"imp":22.9871709369,"elo":1192.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.7655686359,"x_train":618.9598462323,"imp":21.2460816448,"elo":1258.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1204998116,"x_train":1.7915100586,"imp":22.1526904401,"elo":1190.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.253630478,"x_train":382.0935074289,"imp":18.0633828427,"elo":1386.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.4869408199,"x_train":382.0935074289,"imp":17.3844330214,"elo":1428.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.6674634265,"x_train":18.4149364165,"imp":17.3834790986,"elo":1402.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"x_train":14.7795222568,"imp":22.1647731729,"elo":1249.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5246167968,"x_train":4880.2786288265,"imp":18.4556324732,"elo":1360.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7364459708,"x_train":4880.2786288265,"imp":18.5280660881,"elo":1359.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.8592422653,"x_train":16.3156181032,"imp":26.3607466183,"elo":1118.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0410553489,"x_train":26.0689936509,"imp":36.6174806611,"elo":927.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3381841475,"x_train":187.0264910689,"imp":32.5094386414,"elo":1030.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9547740596,"x_train":187.0264910689,"imp":31.0546372303,"elo":1065.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0372103395,"x_train":0.3814409026,"imp":29.8733488873,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0764179217,"x_train":323.8039447074,"imp":24.0381975853,"elo":1149.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7428875097,"x_train":323.8039447074,"imp":23.254536984,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7071575933,"x_train":10.4758857753,"imp":21.4181030792,"elo":1239.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5977456464,"x_train":2879.4580429663,"imp":17.1008499815,"elo":1405.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":12.4836618956,"x_train":2879.4580429663,"imp":15.6602492434,"elo":1481.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7935168864,"x_train":5.7589146636,"imp":13.8336294446,"elo":1536.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3289068866,"x_train":2046.2534629078,"imp":13.6090643932,"elo":1547.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9691710732,"x_train":2046.2534629078,"imp":12.939386077,"elo":1570.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0331410808,"x_train":13.7043041236,"imp":20.9149979551,"elo":1295.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"x_train":49.2127895844,"imp":21.2977853057,"elo":1269.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":41.5948089023,"x_train":4941.9777879175,"imp":18.5407412411,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":307.7353108496,"x_train":4941.9777879175,"imp":17.1517351572,"elo":1407.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1897902902,"x_train":1.9424208612,"imp":17.1066400134,"elo":1433.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"x_train":38.5654426559,"imp":5.0125041353,"elo":1753.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1888600498,"x_train":1.1127929815,"imp":31.6899617016,"elo":1014.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"x_train":2.070384719,"imp":12.4055400385,"elo":1585.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2459026963,"x_train":6.8460080835,"imp":19.6843599668,"elo":1312.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2584066824,"x_train":2168.8617901524,"imp":17.4933162976,"elo":1410.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.3471511219,"x_train":2168.8617901524,"imp":16.6368059624,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"x_train":5.1697320864,"imp":13.1554131904,"elo":1598.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3910008352,"x_train":3.6410702833,"imp":11.7991829167,"elo":1645.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.8830380236,"x_train":439.4466040087,"imp":31.845298093,"elo":1028.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0878026987,"x_train":29479.468606441,"imp":26.3117913949,"elo":1135.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3145858894,"x_train":29479.468606441,"imp":26.3893153598,"elo":1130.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0665397304,"x_train":1.2028502699,"imp":19.1957679596,"elo":1349.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1470302095,"x_train":6.834853082,"imp":25.5387551645,"elo":1091.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1517792148,"x_train":2389.4741835732,"imp":21.4086879296,"elo":1227.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1570271396,"x_train":2389.4741835732,"imp":19.373068261,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.117054282,"x_train":1.7712088618,"imp":21.1866289175,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2050537617,"x_train":685.8651054054,"imp":18.5394348142,"elo":1349.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.4544829944,"x_train":685.8651054054,"imp":18.1717495859,"elo":1375.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":64.5055925989,"x_train":262.9267967989,"imp":24.4956519569,"elo":1150.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":70.8512833361,"x_train":12597.4213063423,"imp":19.9273628334,"elo":1294.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":502.4479719176,"x_train":12597.4213063423,"imp":17.4353708182,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9382527357,"x_train":3.607105251,"imp":30.8711599893,"elo":1011.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1687223142,"x_train":934.2696678038,"imp":20.7246325333,"elo":1287.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.0619945914,"x_train":934.2696678038,"imp":19.3874085581,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_all/tuning_trajectories_explorer.html
index 724cac50f524bf05c8c884e60c272e0059f5adc6..96459b52f334be45652168612d951856599a72fd 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":282.8166166302,"imp":19.4866106899,"elo":1390.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":5.4086527682,"imp":17.666981162,"elo":1405.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":10.084780712,"imp":17.2608793082,"elo":1422.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":24.9344211319,"imp":17.0788703133,"elo":1433.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":49.1678167274,"imp":17.0282981898,"elo":1435.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":137.1191573934,"imp":16.96826421,"elo":1438.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":271.2351529076,"imp":16.9752776429,"elo":1437.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":546.8626199056,"imp":17.0110430549,"elo":1434.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":821.0408680788,"imp":17.0404098845,"elo":1432.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1106.9971386445,"imp":17.0551103834,"elo":1430.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.857962315,"imp":20.8556790977,"elo":1296.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.2122341586,"imp":20.4381051772,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.9434770322,"imp":20.2069721077,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.49430198,"imp":20.0276209902,"elo":1340.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":57.9467632245,"imp":19.7507894054,"elo":1350.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":116.8428883394,"imp":19.6103736056,"elo":1356.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":234.1123581247,"imp":19.4500895732,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":351.0170231939,"imp":19.3838495716,"elo":1364.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":470.3718666937,"imp":19.3489634939,"elo":1365.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":5.7213239382,"imp":21.8421506432,"elo":1257.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":13.9986026157,"imp":21.4353836476,"elo":1270.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":37.2821101451,"imp":20.9881102923,"elo":1290.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":74.1266951378,"imp":20.8189639604,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":187.3186015742,"imp":20.6112059692,"elo":1307.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":376.118404862,"imp":20.4510224786,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":747.8066316207,"imp":20.3116979146,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1119.3701634608,"imp":20.2646509543,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1499.0498932557,"imp":20.2160062698,"elo":1328.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.6841584145,"imp":29.8955515471,"elo":992.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1012271428,"imp":27.6105441527,"elo":1041.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.347596203,"imp":24.319219429,"elo":1123.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.4431649541,"imp":23.221982236,"elo":1170.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":35.7497456308,"imp":22.5560332812,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":71.3524964885,"imp":22.3816570621,"elo":1211.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":142.6147774706,"imp":22.2593854874,"elo":1218.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":213.7904556538,"imp":22.210114809,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":286.4783954015,"imp":22.2239881392,"elo":1223.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.9123333331,"imp":27.7332911484,"elo":1050.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.8242862424,"imp":26.2147818646,"elo":1094.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":15.2779819431,"imp":24.0251575326,"elo":1168.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":31.0766737178,"imp":22.9752310431,"elo":1206.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":75.7672592237,"imp":22.0976024937,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":152.1226481198,"imp":21.6845207101,"elo":1255.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":306.7750654563,"imp":21.4531714276,"elo":1264.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":460.9921495532,"imp":21.31750511,"elo":1271.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":618.9598462323,"imp":21.2649963784,"elo":1274.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":1.7915100586,"imp":22.17664284,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":3.9851860484,"imp":20.3443928335,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":9.4702073578,"imp":19.1046560579,"elo":1360.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":18.3526064623,"imp":18.4658773655,"elo":1398.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":46.5150486041,"imp":17.9003691912,"elo":1425.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":94.2919114747,"imp":17.6699601562,"elo":1438.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":189.6982542539,"imp":17.5020138475,"elo":1447.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":283.010042869,"imp":17.442411899,"elo":1449.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":382.0935074289,"imp":17.4110234301,"elo":1450.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":18.4149364165,"imp":17.3233288219,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.7795222568,"imp":22.1886117827,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":44.6910607231,"imp":20.7050815343,"elo":1295.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":123.8073281883,"imp":18.5338673434,"elo":1383.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":240.4444302535,"imp":18.0550755477,"elo":1402.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":579.7896207628,"imp":17.5720933352,"elo":1421.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1211.3449017346,"imp":17.7010932215,"elo":1413.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2400.9271967121,"imp":18.1069307467,"elo":1398.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3593.5581996112,"imp":18.4011722052,"elo":1387.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4880.2786288265,"imp":18.5604462976,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":16.3156181032,"imp":26.3940957227,"elo":1129.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":26.0689936509,"imp":37.2896660748,"elo":923.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":61.4802624456,"imp":35.0948793172,"elo":980.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":157.8952583504,"imp":31.2561591327,"elo":1069.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":187.0264910689,"imp":31.0711974922,"elo":1076.2,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3814409026,"imp":29.90818391,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.0007658996,"imp":26.5555898652,"elo":1067.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":6.8718138048,"imp":24.9279138756,"elo":1113.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":14.972210111,"imp":24.1428004801,"elo":1143.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":39.102050789,"imp":23.5708415756,"elo":1176.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":79.423821505,"imp":23.4225500964,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":160.2640051076,"imp":23.3238837728,"elo":1190.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":241.0813778596,"imp":23.278774322,"elo":1192.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":323.8039447074,"imp":23.2753167759,"elo":1194.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.4758857753,"imp":21.4421562303,"elo":1244.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.912673739,"imp":18.9123026351,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":67.3088373681,"imp":17.3193831435,"elo":1435.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":140.0742273376,"imp":16.7175077653,"elo":1463.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":358.3731533565,"imp":16.247194002,"elo":1486.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":721.2122449282,"imp":16.0371916549,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1434.2417543333,"imp":15.8487486354,"elo":1502.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2147.1635778763,"imp":15.7559563965,"elo":1506.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2879.4580429663,"imp":15.6869734974,"elo":1507.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.7589146636,"imp":13.8609515983,"elo":1552.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.7323897082,"imp":13.9051821489,"elo":1553.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":46.0214163694,"imp":13.7132293833,"elo":1562.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":96.2671462093,"imp":13.601411572,"elo":1572.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":246.3500390069,"imp":13.401664045,"elo":1584.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":501.1624695611,"imp":13.2179676654,"elo":1589.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1017.6750647229,"imp":13.0946905309,"elo":1591.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1526.8431451139,"imp":13.0315422268,"elo":1591.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2046.2534629078,"imp":12.9736951058,"elo":1588.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.7043041236,"imp":20.9385519844,"elo":1310.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":49.2127895844,"imp":21.3196765384,"elo":1281.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":71.8416880018,"imp":20.3603425203,"elo":1309.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":133.3966980893,"imp":19.3844833097,"elo":1349.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":245.8810811274,"imp":18.9922815096,"elo":1365.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":600.6176271602,"imp":18.1646241206,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1226.0480499304,"imp":17.7397225139,"elo":1409.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2452.678200464,"imp":17.4612911896,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3677.3248587818,"imp":17.2749504085,"elo":1426.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4941.9777879175,"imp":17.1799473592,"elo":1428.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":1.9424208612,"imp":17.1299485628,"elo":1449.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.5654426559,"imp":5.0349777723,"elo":1771.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.1127929815,"imp":31.7087515483,"elo":1026.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":2.070384719,"imp":12.4331942634,"elo":1606.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.8460080835,"imp":19.7096432898,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":15.9731517787,"imp":19.1437730748,"elo":1352.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":48.3100942842,"imp":18.3947839817,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":104.5645791303,"imp":17.7988731991,"elo":1424.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":264.7537616657,"imp":17.2963337301,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":533.594254715,"imp":17.0081271393,"elo":1463.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1058.8890383576,"imp":16.8061805354,"elo":1474.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1605.9821816147,"imp":16.7167699732,"elo":1479.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2168.8617901524,"imp":16.6615491875,"elo":1483.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.1697320864,"imp":13.1824317544,"elo":1624.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.6410702833,"imp":11.8218667003,"elo":1668.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":439.4466040087,"imp":31.8672351585,"elo":1035.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":670.2438710225,"imp":31.7285615606,"elo":1042.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1014.6754881516,"imp":31.4413340163,"elo":1049.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1638.3361981714,"imp":30.7160356751,"elo":1066.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3535.8153662444,"imp":29.3604271249,"elo":1090.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":7084.5550409751,"imp":27.9885464942,"elo":1111.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":14613.5276339936,"imp":27.2362361968,"elo":1127.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":21800.1101862642,"imp":26.7767552698,"elo":1138.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":29479.468606441,"imp":26.4132619841,"elo":1149.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.2028502699,"imp":19.2200244632,"elo":1367.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.834853082,"imp":25.5545350359,"elo":1099.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":16.5429640262,"imp":23.4622161402,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":54.7346146645,"imp":21.7621864364,"elo":1215.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":118.7324150095,"imp":20.968876405,"elo":1251.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":292.558905056,"imp":20.3014901583,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":595.2571207683,"imp":19.8991595219,"elo":1300.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1187.6701815634,"imp":19.6630299605,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1789.3376646728,"imp":19.5247634204,"elo":1321.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2389.4741835732,"imp":19.3918326612,"elo":1327.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.7712088618,"imp":21.210682286,"elo":1236.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.6564696531,"imp":19.4940713306,"elo":1321.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":16.9650142951,"imp":19.0167248254,"elo":1350.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":32.997686743,"imp":18.7140127135,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.160387483,"imp":18.4790700959,"elo":1379.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":172.4620957623,"imp":18.3846352787,"elo":1384.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":343.1961190306,"imp":18.3069368401,"elo":1389.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":511.4517561336,"imp":18.2496166324,"elo":1393.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":685.8651054054,"imp":18.1974575044,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":262.9267967989,"imp":24.5185711598,"elo":1158.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":698.1262023376,"imp":22.3158899069,"elo":1214.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2185.405093286,"imp":19.934293905,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4868.6453358671,"imp":18.755990989,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12038.5904051722,"imp":17.4870693316,"elo":1423.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12597.4213063423,"imp":17.4600133996,"elo":1425.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.607105251,"imp":30.8927551306,"elo":1016.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":9.0537118684,"imp":27.0028382135,"elo":1085.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":23.1440570954,"imp":22.2905863942,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":46.3767226799,"imp":20.3963838454,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":116.4989862715,"imp":19.6649008319,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":232.7740378227,"imp":19.440521796,"elo":1350.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":465.4926636368,"imp":19.3864051899,"elo":1349.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":697.1621157189,"imp":19.3893201421,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":934.2696678038,"imp":19.4004530955,"elo":1344.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8858780824,"x_train":282.8166166302,"imp":19.4866106899,"elo":1390.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0255239609,"x_train":5.4086527682,"imp":17.666981162,"elo":1405.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0519447497,"x_train":10.084780712,"imp":17.2608793082,"elo":1422.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1089050066,"x_train":24.9344211319,"imp":17.0788703133,"elo":1433.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1494108066,"x_train":49.1678167274,"imp":17.0282981898,"elo":1435.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2091965714,"x_train":137.1191573934,"imp":16.96826421,"elo":1438.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2443429842,"x_train":271.2351529076,"imp":16.9752776429,"elo":1437.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2858799211,"x_train":546.8626199056,"imp":17.0110430549,"elo":1434.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3088118892,"x_train":821.0408680788,"imp":17.0404098845,"elo":1432.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.33304612,"x_train":1106.9971386445,"imp":17.0551103834,"elo":1430.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0349046894,"x_train":1.857962315,"imp":20.8556790977,"elo":1296.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0727101961,"x_train":4.2122341586,"imp":20.4381051772,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1347246305,"x_train":10.9434770322,"imp":20.2069721077,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1933333513,"x_train":22.49430198,"imp":20.0276209902,"elo":1340.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2718991335,"x_train":57.9467632245,"imp":19.7507894054,"elo":1350.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3291196359,"x_train":116.8428883394,"imp":19.6103736056,"elo":1356.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3904301416,"x_train":234.1123581247,"imp":19.4500895732,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4239634557,"x_train":351.0170231939,"imp":19.3838495716,"elo":1364.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4227445213,"x_train":470.3718666937,"imp":19.3489634939,"elo":1365.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0170265849,"x_train":5.7213239382,"imp":21.8421506432,"elo":1257.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333155319,"x_train":13.9986026157,"imp":21.4353836476,"elo":1270.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0583218862,"x_train":37.2821101451,"imp":20.9881102923,"elo":1290.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.078412591,"x_train":74.1266951378,"imp":20.8189639604,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1031236538,"x_train":187.3186015742,"imp":20.6112059692,"elo":1307.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1212226385,"x_train":376.118404862,"imp":20.4510224786,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1403339145,"x_train":747.8066316207,"imp":20.3116979146,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1466477215,"x_train":1119.3701634608,"imp":20.2646509543,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1440681396,"x_train":1499.0498932557,"imp":20.2160062698,"elo":1328.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2946660708,"x_train":1.6841584145,"imp":29.8955515471,"elo":992.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3055075075,"x_train":3.1012271428,"imp":27.6105441527,"elo":1041.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3457056156,"x_train":7.347596203,"imp":24.319219429,"elo":1123.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3944436772,"x_train":14.4431649541,"imp":23.221982236,"elo":1170.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5014192887,"x_train":35.7497456308,"imp":22.5560332812,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5959923774,"x_train":71.3524964885,"imp":22.3816570621,"elo":1211.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6717315754,"x_train":142.6147774706,"imp":22.2593854874,"elo":1218.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7207892432,"x_train":213.7904556538,"imp":22.210114809,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7589964064,"x_train":286.4783954015,"imp":22.2239881392,"elo":1223.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3681069516,"x_train":2.9123333331,"imp":27.7332911484,"elo":1050.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.608277041,"x_train":5.8242862424,"imp":26.2147818646,"elo":1094.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1293887702,"x_train":15.2779819431,"imp":24.0251575326,"elo":1168.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7335087318,"x_train":31.0766737178,"imp":22.9752310431,"elo":1206.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4971865346,"x_train":75.7672592237,"imp":22.0976024937,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.2671043117,"x_train":152.1226481198,"imp":21.6845207101,"elo":1255.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0044565763,"x_train":306.7750654563,"imp":21.4531714276,"elo":1264.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4204326806,"x_train":460.9921495532,"imp":21.31750511,"elo":1271.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.7390931271,"x_train":618.9598462323,"imp":21.2649963784,"elo":1274.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1204998116,"x_train":1.7915100586,"imp":22.17664284,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4762849552,"x_train":3.9851860484,"imp":20.3443928335,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7292175393,"x_train":9.4702073578,"imp":19.1046560579,"elo":1360.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9391818478,"x_train":18.3526064623,"imp":18.4658773655,"elo":1398.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2648300791,"x_train":46.5150486041,"imp":17.9003691912,"elo":1425.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4080976208,"x_train":94.2919114747,"imp":17.6699601562,"elo":1438.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4975096766,"x_train":189.6982542539,"imp":17.5020138475,"elo":1447.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4963379792,"x_train":283.010042869,"imp":17.442411899,"elo":1449.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5345635664,"x_train":382.0935074289,"imp":17.4110234301,"elo":1450.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.6674634265,"x_train":18.4149364165,"imp":17.3233288219,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"x_train":14.7795222568,"imp":22.1886117827,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8804757883,"x_train":44.6910607231,"imp":20.7050815343,"elo":1295.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9515166073,"x_train":123.8073281883,"imp":18.5338673434,"elo":1383.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9139871186,"x_train":240.4444302535,"imp":18.0550755477,"elo":1402.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.2991192179,"x_train":579.7896207628,"imp":17.5720933352,"elo":1421.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.6324379245,"x_train":1211.3449017346,"imp":17.7010932215,"elo":1413.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.2170906616,"x_train":2400.9271967121,"imp":18.1069307467,"elo":1398.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.2545907476,"x_train":3593.5581996112,"imp":18.4011722052,"elo":1387.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7287422808,"x_train":4880.2786288265,"imp":18.5604462976,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.8592422653,"x_train":16.3156181032,"imp":26.3940957227,"elo":1129.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0410553489,"x_train":26.0689936509,"imp":37.2896660748,"elo":923.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3573866152,"x_train":61.4802624456,"imp":35.0948793172,"elo":980.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.8506460976,"x_train":157.8952583504,"imp":31.2561591327,"elo":1069.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9547740596,"x_train":187.0264910689,"imp":31.0711974922,"elo":1076.2,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0372103395,"x_train":0.3814409026,"imp":29.90818391,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1067923159,"x_train":2.0007658996,"imp":26.5555898652,"elo":1067.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2162743727,"x_train":6.8718138048,"imp":24.9279138756,"elo":1113.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3276673648,"x_train":14.972210111,"imp":24.1428004801,"elo":1143.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4915865441,"x_train":39.102050789,"imp":23.5708415756,"elo":1176.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.582457396,"x_train":79.423821505,"imp":23.4225500964,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6587344304,"x_train":160.2640051076,"imp":23.3238837728,"elo":1190.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7122437923,"x_train":241.0813778596,"imp":23.278774322,"elo":1192.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7428776826,"x_train":323.8039447074,"imp":23.2753167759,"elo":1194.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7071575933,"x_train":10.4758857753,"imp":21.4421562303,"elo":1244.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.4603108225,"x_train":24.912673739,"imp":18.9123026351,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.7287224795,"x_train":67.3088373681,"imp":17.3193831435,"elo":1435.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.7139024305,"x_train":140.0742273376,"imp":16.7175077653,"elo":1463.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.8490124424,"x_train":358.3731533565,"imp":16.247194002,"elo":1486.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.7712471898,"x_train":721.2122449282,"imp":16.0371916549,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.4105608019,"x_train":1434.2417543333,"imp":15.8487486354,"elo":1502.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.2019759137,"x_train":2147.1635778763,"imp":15.7559563965,"elo":1506.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.6005448312,"x_train":2879.4580429663,"imp":15.6869734974,"elo":1507.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7935168864,"x_train":5.7589146636,"imp":13.8609515983,"elo":1552.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.5040919517,"x_train":16.7323897082,"imp":13.9051821489,"elo":1553.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.0181434206,"x_train":46.0214163694,"imp":13.7132293833,"elo":1562.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1961228236,"x_train":96.2671462093,"imp":13.601411572,"elo":1572.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7695458497,"x_train":246.3500390069,"imp":13.401664045,"elo":1584.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.0980977495,"x_train":501.1624695611,"imp":13.2179676654,"elo":1589.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3132120072,"x_train":1017.6750647229,"imp":13.0946905309,"elo":1591.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.2061952514,"x_train":1526.8431451139,"imp":13.0315422268,"elo":1591.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.0822605356,"x_train":2046.2534629078,"imp":12.9736951058,"elo":1588.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0331410808,"x_train":13.7043041236,"imp":20.9385519844,"elo":1310.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"x_train":49.2127895844,"imp":21.3196765384,"elo":1281.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":57.3203633186,"x_train":71.8416880018,"imp":20.3603425203,"elo":1309.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":83.2470200097,"x_train":133.3966980893,"imp":19.3844833097,"elo":1349.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":114.641595606,"x_train":245.8810811274,"imp":18.9922815096,"elo":1365.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":172.6551856784,"x_train":600.6176271602,"imp":18.1646241206,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":232.5924596656,"x_train":1226.0480499304,"imp":17.7397225139,"elo":1409.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":270.3155274697,"x_train":2452.678200464,"imp":17.4612911896,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":289.6259857616,"x_train":3677.3248587818,"imp":17.2749504085,"elo":1426.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":311.2606556608,"x_train":4941.9777879175,"imp":17.1799473592,"elo":1428.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1897902902,"x_train":1.9424208612,"imp":17.1299485628,"elo":1449.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"x_train":38.5654426559,"imp":5.0349777723,"elo":1771.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1888600498,"x_train":1.1127929815,"imp":31.7087515483,"elo":1026.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"x_train":2.070384719,"imp":12.4331942634,"elo":1606.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2459026963,"x_train":6.8460080835,"imp":19.7096432898,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4940504378,"x_train":15.9731517787,"imp":19.1437730748,"elo":1352.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9205220996,"x_train":48.3100942842,"imp":18.3947839817,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.1374930629,"x_train":104.5645791303,"imp":17.7988731991,"elo":1424.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.5093622829,"x_train":264.7537616657,"imp":17.2963337301,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.8290582337,"x_train":533.594254715,"imp":17.0081271393,"elo":1463.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":2.1248042643,"x_train":1058.8890383576,"imp":16.8061805354,"elo":1474.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.297931986,"x_train":1605.9821816147,"imp":16.7167699732,"elo":1479.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.3421030774,"x_train":2168.8617901524,"imp":16.6615491875,"elo":1483.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"x_train":5.1697320864,"imp":13.1824317544,"elo":1624.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3910008352,"x_train":3.6410702833,"imp":11.8218667003,"elo":1668.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.8830380236,"x_train":439.4466040087,"imp":31.8672351585,"elo":1035.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.4248868953,"x_train":670.2438710225,"imp":31.7285615606,"elo":1042.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8753585219,"x_train":1014.6754881516,"imp":31.4413340163,"elo":1049.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.2526853475,"x_train":1638.3361981714,"imp":30.7160356751,"elo":1066.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.0721037925,"x_train":3535.8153662444,"imp":29.3604271249,"elo":1090.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":13.3360172525,"x_train":7084.5550409751,"imp":27.9885464942,"elo":1111.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":15.3290278651,"x_train":14613.5276339936,"imp":27.2362361968,"elo":1127.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.4140156092,"x_train":21800.1101862642,"imp":26.7767552698,"elo":1138.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3145858894,"x_train":29479.468606441,"imp":26.4132619841,"elo":1149.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0665397304,"x_train":1.2028502699,"imp":19.2200244632,"elo":1367.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1470302095,"x_train":6.834853082,"imp":25.5545350359,"elo":1099.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2938262905,"x_train":16.5429640262,"imp":23.4622161402,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5650998098,"x_train":54.7346146645,"imp":21.7621864364,"elo":1215.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8070919893,"x_train":118.7324150095,"imp":20.968876405,"elo":1251.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2465421293,"x_train":292.558905056,"imp":20.3014901583,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6220136736,"x_train":595.2571207683,"imp":19.8991595219,"elo":1300.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8705493288,"x_train":1187.6701815634,"imp":19.6630299605,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0769295602,"x_train":1789.3376646728,"imp":19.5247634204,"elo":1321.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1462178945,"x_train":2389.4741835732,"imp":19.3918326612,"elo":1327.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.117054282,"x_train":1.7712088618,"imp":21.210682286,"elo":1236.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5009884397,"x_train":5.6564696531,"imp":19.4940713306,"elo":1321.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7880470326,"x_train":16.9650142951,"imp":19.0167248254,"elo":1350.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8826892134,"x_train":32.997686743,"imp":18.7140127135,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0482676981,"x_train":86.160387483,"imp":18.4790700959,"elo":1379.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1873182694,"x_train":172.4620957623,"imp":18.3846352787,"elo":1384.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2414220916,"x_train":343.1961190306,"imp":18.3069368401,"elo":1389.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4140647498,"x_train":511.4517561336,"imp":18.2496166324,"elo":1393.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4519965713,"x_train":685.8651054054,"imp":18.1974575044,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":64.5055925989,"x_train":262.9267967989,"imp":24.5185711598,"elo":1158.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":130.1486843843,"x_train":698.1262023376,"imp":22.3158899069,"elo":1214.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":260.8212573113,"x_train":2185.405093286,"imp":19.934293905,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":378.933825373,"x_train":4868.6453358671,"imp":18.755990989,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":496.0296677873,"x_train":12038.5904051722,"imp":17.4870693316,"elo":1423.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":502.4479719176,"x_train":12597.4213063423,"imp":17.4600133996,"elo":1425.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9382527357,"x_train":3.607105251,"imp":30.8927551306,"elo":1016.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.9524919805,"x_train":9.0537118684,"imp":27.0028382135,"elo":1085.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1739792718,"x_train":23.1440570954,"imp":22.2905863942,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.4724798591,"x_train":46.3767226799,"imp":20.3963838454,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8253774488,"x_train":116.4989862715,"imp":19.6649008319,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2506435468,"x_train":232.7740378227,"imp":19.440521796,"elo":1350.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.5518696321,"x_train":465.4926636368,"imp":19.3864051899,"elo":1349.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.795048319,"x_train":697.1621157189,"imp":19.3893201421,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.0515616235,"x_train":934.2696678038,"imp":19.4004530955,"elo":1344.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_all/winrate_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..88aeb5ff9952e6579cf39ff4143ffe3bd74b4404
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_classification/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned)", "TabDPT-Turbo (default)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabM (tuned + ensembled)", "LimiX (default)", "TabDPT (tuned + ensembled)", "RealMLP (tuned)", "LightGBM (tuned + ensembled)", "TabM (tuned)", "BetaTabPFN (default)", "TabDPT (tuned)", "CatBoost (tuned + ensembled)", "TabSwift (default)", "LightGBM (tuned)", "iLTM (tuned + ensembled)", "CatBoost (tuned)", "ModernNCA (tuned)", "CatBoost (default)", "xRFM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabDPT (default)", "XGBoost (tuned)", "xRFM (tuned)", "SAP-RPT-OSS (default)", "EBM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "TabM (default)", "FastaiMLP (tuned + ensembled)", "ModernNCA (default)", "ChimeraBoost (tuned + ensembled)", "EBM (tuned)", "iLTM (tuned)", "ChimeraBoost (tuned)", "EBM (default)", "RealMLP (default)", "TorchMLP (tuned)", "FastaiMLP (tuned)", "ExtraTrees (tuned + ensembled)", "ChimeraBoost (default)", "ExtraTrees (tuned)", "XGBoost (default)", "TabSTAR (tuned)", "OrionMSP (default)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned + ensembled)", "LightGBM (default)", "RandomForest (tuned)", "iLTM (default)", "TorchMLP (default)", "Linear (tuned + ensembled)", "TabFlex (default)", "PerpetualBooster (tuned + ensembled)", "Linear (tuned)", "TabSTAR (default)", "FastaiMLP (default)", "xRFM (default)", "PerpetualBooster (tuned)", "KNN (tuned + ensembled)", "ExtraTrees (default)", "RandomForest (default)", "Linear (default)", "PerpetualBooster (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7592592835426331, 0.7979797720909119, 0.8333333134651184, 0.8181818127632141, 0.7853535413742065, 0.7996633052825928, 0.7912458181381226, 0.8518518805503845, 0.8518518805503845, 0.8636363744735718, 0.8838383555412292, 0.8762626051902771, 0.8989899158477783, 0.8484848737716675, 0.8939393758773804, 0.9107744097709656, 0.8796296119689941, 0.8922559022903442, 0.8973063826560974, 0.8821548819541931, 0.9090909361839294, 0.9090909361839294, 0.9090909361839294, 0.9090909361839294, 0.9057239294052124, 0.9090909361839294, 0.9242424368858337, 0.9040403962135315, 0.9292929172515869, 0.8956229090690613, 0.8585858345031738, 0.9090909361839294, 0.936026930809021, 0.9478114247322083, 0.9158248901367188, 0.8989899158477783, 0.9191918969154358, 0.9208754301071167, 0.9208754301071167, 0.9292929172515869, 0.9292929172515869, 0.9158248901367188, 0.9309764504432678, 0.9595959782600403, 0.9309764504432678, 0.9461279511451721, 0.939393937587738, 0.9377104640007019, 0.9612794518470764, 0.8518518805503845, 0.941077470779419, 0.8518518805503845, 0.9595959782600403, 0.9612794518470764, 0.9579124450683594, 0.941077470779419, 0.9663299918174744, 0.9427609443664551, 0.9377104640007019, 0.9848484992980957, 0.9494949579238892, 0.8905723690986633, 0.9680134654045105, 0.9629629850387573, 0.9814814925193787, 0.9629629850387573, 0.9882155060768127, 0.9814814925193787, 0.9663299918174744, 0.9831649661064148, 0.9629629850387573, 0.9949495196342468], [0.24074074625968933, NaN, 0.4444444477558136, 0.4932659864425659, 0.6414141654968262, 0.5892255902290344, 0.5993266105651855, 0.6094276309013367, 0.6447811722755432, 0.7954545617103577, 0.7845118045806885, 0.7946127653121948, 0.7457912564277649, 0.7895622849464417, 0.8215488195419312, 0.8518518805503845, 0.8072390556335449, 0.7777777910232544, 0.8131313323974609, 0.8636363744735718, 0.8249158263206482, 0.8552188277244568, 0.8855218887329102, 0.8594276309013367, 0.875420868396759, 0.8712121248245239, 0.8787878751754761, 0.875420868396759, 0.7946127653121948, 0.8787878751754761, 0.8543770909309387, 0.7626262903213501, 0.8619528412818909, 0.9175084233283997, 0.9023569226264954, 0.8619528412818909, 0.8619528412818909, 0.8703703880310059, 0.9141414165496826, 0.8737373948097229, 0.9057239294052124, 0.9175084233283997, 0.8787878751754761, 0.8872053623199463, 0.9292929172515869, 0.8855218887329102, 0.9191918969154358, 0.9242424368858337, 0.9074074029922485, 0.9494949579238892, 0.81144779920578, 0.9284511804580688, 0.8131313323974609, 0.9343434572219849, 0.9545454382896423, 0.936026930809021, 0.9427609443664551, 0.9579124450683594, 0.9006733894348145, 0.9124578833580017, 0.9511784315109253, 0.9225589036941528, 0.8518518805503845, 0.9528619647026062, 0.9545454382896423, 0.9570707082748413, 0.941077470779419, 0.9612794518470764, 0.9629629850387573, 0.945286214351654, 0.9646464586257935, 0.9545454382896423, 0.9949495196342468], [0.20202019810676575, 0.5555555820465088, NaN, 0.5336700081825256, 0.6329966187477112, 0.6178451180458069, 0.6430976390838623, 0.6397306323051453, 0.7356902360916138, 0.7272727489471436, 0.7171717286109924, 0.7626262903213501, 0.7306397557258606, 0.7996633052825928, 0.7626262903213501, 0.8030303120613098, 0.804713785648346, 0.7685185074806213, 0.8299663066864014, 0.8265992999076843, 0.8619528412818909, 0.8434343338012695, 0.8518518805503845, 0.8333333134651184, 0.8535353541374207, 0.8468013405799866, 0.8434343338012695, 0.8341751098632812, 0.8417508602142334, 0.8501683473587036, 0.8400673270225525, 0.7676767706871033, 0.8434343338012695, 0.8872053623199463, 0.8872053623199463, 0.8451178669929504, 0.8409090638160706, 0.881313145160675, 0.8888888955116272, 0.8619528412818909, 0.9082491397857666, 0.8872053623199463, 0.8796296119689941, 0.9090909361839294, 0.9242424368858337, 0.8914141654968262, 0.9309764504432678, 0.8872053623199463, 0.9292929172515869, 0.9511784315109253, 0.8409090638160706, 0.9259259104728699, 0.8451178669929504, 0.9242424368858337, 0.9478114247322083, 0.9461279511451721, 0.9461279511451721, 0.9528619647026062, 0.9074074029922485, 0.9292929172515869, 0.9461279511451721, 0.9158248901367188, 0.875420868396759, 0.9545454382896423, 0.9612794518470764, 0.9579124450683594, 0.941077470779419, 0.9781144857406616, 0.9814814925193787, 0.9494949579238892, 0.9696969985961914, 0.9595959782600403, 0.996632993221283], [0.1666666716337204, 0.5067340135574341, 0.466329962015152, NaN, 0.5774410963058472, 0.560606062412262, 0.5850168466567993, 0.5757575631141663, 0.7121211886405945, 0.7020202279090881, 0.7070707082748413, 0.7390572428703308, 0.7314814925193787, 0.7971380352973938, 0.7289562225341797, 0.7845118045806885, 0.7676767706871033, 0.7676767706871033, 0.7962962985038757, 0.7861952781677246, 0.8383838534355164, 0.8181818127632141, 0.8316498398780823, 0.7929292917251587, 0.8299663066864014, 0.8215488195419312, 0.81144779920578, 0.8030303120613098, 0.819865345954895, 0.8131313323974609, 0.8005050420761108, 0.7323232293128967, 0.8215488195419312, 0.8400673270225525, 0.8670033812522888, 0.8316498398780823, 0.8367003202438354, 0.877946138381958, 0.8518518805503845, 0.8484848737716675, 0.8653198480606079, 0.8636363744735718, 0.8703703880310059, 0.8855218887329102, 0.9090909361839294, 0.8872053623199463, 0.8897306323051453, 0.8636363744735718, 0.8939393758773804, 0.9175084233283997, 0.8299663066864014, 0.9225589036941528, 0.8282828330993652, 0.9208754301071167, 0.9377104640007019, 0.9208754301071167, 0.9175084233283997, 0.941077470779419, 0.9191918969154358, 0.9107744097709656, 0.9511784315109253, 0.9276094436645508, 0.8771044015884399, 0.9461279511451721, 0.9377104640007019, 0.9562289714813232, 0.9444444179534912, 0.9579124450683594, 0.9663299918174744, 0.9537037014961243, 0.9764309525489807, 0.9545454382896423, 0.996632993221283], [0.1818181872367859, 0.3585858643054962, 0.3670033812522888, 0.4225589334964752, NaN, 0.48653197288513184, 0.5471380352973938, 0.5126262903213501, 0.627946138381958, 0.7037037014961243, 0.6750841736793518, 0.7239057421684265, 0.680134654045105, 0.7020202279090881, 0.7491582632064819, 0.7828282713890076, 0.7744107842445374, 0.7542087435722351, 0.7272727489471436, 0.7895622849464417, 0.7777777910232544, 0.8333333134651184, 0.8299663066864014, 0.7828282713890076, 0.8282828330993652, 0.8215488195419312, 0.8030303120613098, 0.816498339176178, 0.7373737096786499, 0.8451178669929504, 0.8030303120613098, 0.7710437774658203, 0.819865345954895, 0.8771044015884399, 0.872053861618042, 0.8341751098632812, 0.8316498398780823, 0.8653198480606079, 0.875420868396759, 0.8577440977096558, 0.8821548819541931, 0.8821548819541931, 0.8518518805503845, 0.8855218887329102, 0.9158248901367188, 0.8787878751754761, 0.8973063826560974, 0.8821548819541931, 0.8897306323051453, 0.9478114247322083, 0.8097643256187439, 0.8821548819541931, 0.8181818127632141, 0.9090909361839294, 0.9444444179534912, 0.932659924030304, 0.9259259104728699, 0.9494949579238892, 0.8973063826560974, 0.9107744097709656, 0.941077470779419, 0.9175084233283997, 0.8434343338012695, 0.9343434572219849, 0.9377104640007019, 0.9377104640007019, 0.9343434572219849, 0.9595959782600403, 0.9713804721832275, 0.941077470779419, 0.9629629850387573, 0.9511784315109253, 0.9932659864425659], [0.21464645862579346, 0.4107744097709656, 0.3821548819541931, 0.43939393758773804, 0.5134680271148682, NaN, 0.5824915766716003, 0.6094276309013367, 0.5774410963058472, 0.691919207572937, 0.7121211886405945, 0.7070707082748413, 0.7020202279090881, 0.6851851940155029, 0.7356902360916138, 0.7710437774658203, 0.7491582632064819, 0.7744107842445374, 0.7272727489471436, 0.7946127653121948, 0.7727272510528564, 0.8021885752677917, 0.7752525210380554, 0.7946127653121948, 0.8232323527336121, 0.804713785648346, 0.8063973188400269, 0.804713785648346, 0.7037037014961243, 0.8223905563354492, 0.804713785648346, 0.7685185074806213, 0.819865345954895, 0.816498339176178, 0.8535353541374207, 0.8131313323974609, 0.8190235495567322, 0.8316498398780823, 0.8308081030845642, 0.8442760705947876, 0.8552188277244568, 0.8417508602142334, 0.8451178669929504, 0.8569023609161377, 0.8973063826560974, 0.875420868396759, 0.875420868396759, 0.8552188277244568, 0.8771044015884399, 0.9141414165496826, 0.81144779920578, 0.9057239294052124, 0.816498339176178, 0.9040403962135315, 0.9343434572219849, 0.9208754301071167, 0.9124578833580017, 0.941077470779419, 0.8905723690986633, 0.9074074029922485, 0.9461279511451721, 0.9057239294052124, 0.8434343338012695, 0.9343434572219849, 0.9377104640007019, 0.9478114247322083, 0.9158248901367188, 0.9562289714813232, 0.9562289714813232, 0.930134654045105, 0.9646464586257935, 0.9377104640007019, 0.9932659864425659], [0.20033669471740723, 0.40067338943481445, 0.3569023609161377, 0.4149831533432007, 0.4528619647026062, 0.41750842332839966, NaN, 0.4654882252216339, 0.5521885752677917, 0.6616161465644836, 0.6750841736793518, 0.7104377150535583, 0.6851851940155029, 0.6818181872367859, 0.7255892157554626, 0.7424242496490479, 0.755892276763916, 0.7634680271148682, 0.7112794518470764, 0.7592592835426331, 0.8013467788696289, 0.7811447978019714, 0.7356902360916138, 0.7659932374954224, 0.7626262903213501, 0.7929292917251587, 0.8030303120613098, 0.7777777910232544, 0.7340067625045776, 0.808080792427063, 0.81144779920578, 0.7828282713890076, 0.8097643256187439, 0.7659932374954224, 0.8468013405799866, 0.8131313323974609, 0.8367003202438354, 0.8333333134651184, 0.8215488195419312, 0.8417508602142334, 0.8611111044883728, 0.8316498398780823, 0.8400673270225525, 0.8703703880310059, 0.8989899158477783, 0.8771044015884399, 0.872053861618042, 0.8602693676948547, 0.8855218887329102, 0.8939393758773804, 0.8333333134651184, 0.932659924030304, 0.8341751098632812, 0.8653198480606079, 0.9343434572219849, 0.9141414165496826, 0.9175084233283997, 0.9612794518470764, 0.8922559022903442, 0.9267676472663879, 0.9461279511451721, 0.8989899158477783, 0.8619528412818909, 0.9309764504432678, 0.941077470779419, 0.9528619647026062, 0.9023569226264954, 0.9579124450683594, 0.9579124450683594, 0.9478114247322083, 0.9713804721832275, 0.9461279511451721, 0.9949495196342468], [0.20875421166419983, 0.3905723989009857, 0.36026936769485474, 0.42424243688583374, 0.4873737394809723, 0.3905723989009857, 0.5345118045806885, NaN, 0.5723905563354492, 0.6649831533432007, 0.6868686676025391, 0.7020202279090881, 0.6700336933135986, 0.6818181872367859, 0.7272727489471436, 0.7542087435722351, 0.75, 0.755892276763916, 0.7264309525489807, 0.7811447978019714, 0.7617844939231873, 0.8013467788696289, 0.752525269985199, 0.7777777910232544, 0.7794612646102905, 0.7912458181381226, 0.7904040217399597, 0.7878788113594055, 0.7104377150535583, 0.804713785648346, 0.7912458181381226, 0.7609427571296692, 0.819865345954895, 0.8072390556335449, 0.8451178669929504, 0.8063973188400269, 0.8030303120613098, 0.816498339176178, 0.813973069190979, 0.8333333134651184, 0.8434343338012695, 0.8350168466567993, 0.8350168466567993, 0.8484848737716675, 0.8956229090690613, 0.8670033812522888, 0.869528591632843, 0.8468013405799866, 0.8787878751754761, 0.9074074029922485, 0.8030303120613098, 0.9006733894348145, 0.81144779920578, 0.8872053623199463, 0.9292929172515869, 0.9124578833580017, 0.9040403962135315, 0.941077470779419, 0.875420868396759, 0.8973063826560974, 0.9318181872367859, 0.8956229090690613, 0.8434343338012695, 0.9259259104728699, 0.936026930809021, 0.9377104640007019, 0.9124578833580017, 0.9494949579238892, 0.9528619647026062, 0.933501660823822, 0.9646464586257935, 0.9427609443664551, 0.9932659864425659], [0.14814814925193787, 0.3552188575267792, 0.26430976390838623, 0.28787878155708313, 0.372053861618042, 0.4225589334964752, 0.44781145453453064, 0.4276094138622284, NaN, 0.5454545617103577, 0.5286195278167725, 0.5361952781677246, 0.5925925970077515, 0.624579131603241, 0.618686854839325, 0.6094276309013367, 0.5892255902290344, 0.5740740895271301, 0.680134654045105, 0.6531986594200134, 0.7264309525489807, 0.6212121248245239, 0.6936026811599731, 0.6464646458625793, 0.6565656661987305, 0.6750841736793518, 0.6767676472663879, 0.6632996797561646, 0.8072390556335449, 0.6767676472663879, 0.7053872346878052, 0.6414141654968262, 0.6818181872367859, 0.680134654045105, 0.7255892157554626, 0.6683501601219177, 0.7508417367935181, 0.755892276763916, 0.7390572428703308, 0.7196969985961914, 0.7735690474510193, 0.7659932374954224, 0.7491582632064819, 0.7777777910232544, 0.7912458181381226, 0.8148148059844971, 0.8257575631141663, 0.7760942578315735, 0.8535353541374207, 0.8265992999076843, 0.7542087435722351, 0.8670033812522888, 0.7609427571296692, 0.8451178669929504, 0.8602693676948547, 0.8602693676948547, 0.8367003202438354, 0.872053861618042, 0.8249158263206482, 0.9057239294052124, 0.8973063826560974, 0.8367003202438354, 0.8299663066864014, 0.9158248901367188, 0.8922559022903442, 0.9141414165496826, 0.9090909361839294, 0.9486532211303711, 0.9528619647026062, 0.9225589036941528, 0.9612794518470764, 0.9511784315109253, 0.9949495196342468], [0.14814814925193787, 0.20454545319080353, 0.27272728085517883, 0.29797980189323425, 0.29629629850387573, 0.3080808222293854, 0.33838382363319397, 0.3350168466567993, 0.4545454680919647, NaN, 0.4722222089767456, 0.5538720488548279, 0.469696968793869, 0.505892276763916, 0.6776094436645508, 0.6734007000923157, 0.6111111044883728, 0.6329966187477112, 0.5429292917251587, 0.695286214351654, 0.5488215684890747, 0.7230639457702637, 0.6969696879386902, 0.7053872346878052, 0.6666666865348816, 0.7121211886405945, 0.7255892157554626, 0.7693602442741394, 0.5892255902290344, 0.7929292917251587, 0.7138047218322754, 0.6262626051902771, 0.7693602442741394, 0.6851851940155029, 0.7356902360916138, 0.7777777910232544, 0.7904040217399597, 0.7323232293128967, 0.7542087435722351, 0.7861952781677246, 0.7828282713890076, 0.7786195278167725, 0.7887205481529236, 0.869528591632843, 0.7929292917251587, 0.810606062412262, 0.8131313323974609, 0.7794612646102905, 0.8223905563354492, 0.8888888955116272, 0.7676767706871033, 0.7828282713890076, 0.7744107842445374, 0.8855218887329102, 0.9057239294052124, 0.9107744097709656, 0.8787878751754761, 0.9141414165496826, 0.8703703880310059, 0.8265992999076843, 0.9309764504432678, 0.872053861618042, 0.8131313323974609, 0.9343434572219849, 0.9276094436645508, 0.9377104640007019, 0.9090909361839294, 0.9545454382896423, 0.9545454382896423, 0.8905723690986633, 0.9663299918174744, 0.932659924030304, 0.9949495196342468], [0.13636364042758942, 0.21548821032047272, 0.28282827138900757, 0.2929292917251587, 0.3249158263206482, 0.28787878155708313, 0.3249158263206482, 0.31313130259513855, 0.47138047218322754, 0.5277777910232544, NaN, 0.5454545617103577, 0.4444444477558136, 0.5252525210380554, 0.6144781112670898, 0.6969696879386902, 0.6043770909309387, 0.6060606241226196, 0.5521885752677917, 0.691919207572937, 0.5715488195419312, 0.7222222089767456, 0.7053872346878052, 0.7171717286109924, 0.6548821330070496, 0.7407407164573669, 0.691077470779419, 0.747474730014801, 0.6094276309013367, 0.7390572428703308, 0.7171717286109924, 0.5959596037864685, 0.744107723236084, 0.695286214351654, 0.7760942578315735, 0.7289562225341797, 0.7609427571296692, 0.7356902360916138, 0.7836700081825256, 0.7609427571296692, 0.7845118045806885, 0.8063973188400269, 0.7929292917251587, 0.8030303120613098, 0.8316498398780823, 0.81144779920578, 0.7508417367935181, 0.8299663066864014, 0.7794612646102905, 0.8838383555412292, 0.7659932374954224, 0.753367006778717, 0.7727272510528564, 0.8350168466567993, 0.9090909361839294, 0.8678451180458069, 0.880471408367157, 0.8939393758773804, 0.8661616444587708, 0.8383838534355164, 0.9242424368858337, 0.8897306323051453, 0.8148148059844971, 0.9276094436645508, 0.8956229090690613, 0.9276094436645508, 0.8922559022903442, 0.9343434572219849, 0.9276094436645508, 0.9191918969154358, 0.9478114247322083, 0.9276094436645508, 0.996632993221283], [0.11616161465644836, 0.2053872048854828, 0.2373737394809723, 0.2609427571296692, 0.2760942876338959, 0.2929292917251587, 0.28956228494644165, 0.29797980189323425, 0.4638047218322754, 0.4461279511451721, 0.4545454680919647, NaN, 0.43265992403030396, 0.5092592835426331, 0.567340075969696, 0.6144781112670898, 0.6666666865348816, 0.6397306323051453, 0.5488215684890747, 0.6127946376800537, 0.5723905563354492, 0.6557239294052124, 0.6170033812522888, 0.6329966187477112, 0.6388888955116272, 0.6969696879386902, 0.6144781112670898, 0.6902356743812561, 0.5959596037864685, 0.7272727489471436, 0.6178451180458069, 0.6329966187477112, 0.747474730014801, 0.6734007000923157, 0.7020202279090881, 0.8442760705947876, 0.7138047218322754, 0.7171717286109924, 0.7457912564277649, 0.7861952781677246, 0.7323232293128967, 0.7777777910232544, 0.7786195278167725, 0.817340075969696, 0.7954545617103577, 0.7744107842445374, 0.7457912564277649, 0.8148148059844971, 0.747474730014801, 0.8484848737716675, 0.7676767706871033, 0.7592592835426331, 0.7744107842445374, 0.7962962985038757, 0.9015151262283325, 0.8400673270225525, 0.8737373948097229, 0.9057239294052124, 0.8670033812522888, 0.8282828330993652, 0.9057239294052124, 0.8829966187477112, 0.816498339176178, 0.9006733894348145, 0.8670033812522888, 0.9259259104728699, 0.8838383555412292, 0.9208754301071167, 0.9141414165496826, 0.9107744097709656, 0.9629629850387573, 0.9242424368858337, 0.9865319728851318], [0.12373737245798111, 0.2542087435722351, 0.2693602740764618, 0.26851850748062134, 0.31986531615257263, 0.29797980189323425, 0.31481480598449707, 0.32996633648872375, 0.40740740299224854, 0.5303030014038086, 0.5555555820465088, 0.567340075969696, NaN, 0.5075757503509521, 0.5749158263206482, 0.6548821330070496, 0.5782828330993652, 0.5824915766716003, 0.5387205481529236, 0.6683501601219177, 0.5656565427780151, 0.6700336933135986, 0.7129629850387573, 0.6700336933135986, 0.6683501601219177, 0.6969696879386902, 0.6902356743812561, 0.7037037014961243, 0.5791245698928833, 0.7222222089767456, 0.689393937587738, 0.6296296119689941, 0.7121211886405945, 0.6818181872367859, 0.691919207572937, 0.6700336933135986, 0.6759259104728699, 0.7239057421684265, 0.7575757503509521, 0.7373737096786499, 0.7373737096786499, 0.7659932374954224, 0.7390572428703308, 0.7457912564277649, 0.7626262903213501, 0.7584174871444702, 0.7929292917251587, 0.7794612646102905, 0.7962962985038757, 0.8215488195419312, 0.7281144857406616, 0.7920875549316406, 0.7239057421684265, 0.8316498398780823, 0.8350168466567993, 0.8451178669929504, 0.8232323527336121, 0.8282828330993652, 0.8282828330993652, 0.7878788113594055, 0.8636363744735718, 0.8484848737716675, 0.7819865345954895, 0.8737373948097229, 0.9107744097709656, 0.8787878751754761, 0.8451178669929504, 0.9107744097709656, 0.9242424368858337, 0.8552188277244568, 0.9141414165496826, 0.8787878751754761, 0.9747474789619446], [0.10101009905338287, 0.21043771505355835, 0.20033669471740723, 0.202861949801445, 0.29797980189323425, 0.31481480598449707, 0.3181818127632141, 0.3181818127632141, 0.37542086839675903, 0.49410775303840637, 0.4747474789619446, 0.49074074625968933, 0.49242424964904785, NaN, 0.5715488195419312, 0.563973069190979, 0.5395622849464417, 0.5370370149612427, 0.6498316526412964, 0.5808081030845642, 0.6001683473587036, 0.5959596037864685, 0.627946138381958, 0.5875421166419983, 0.6127946376800537, 0.624579131603241, 0.6355218887329102, 0.6052188277244568, 0.6784511804580688, 0.6380471587181091, 0.6540403962135315, 0.6094276309013367, 0.6405723690986633, 0.6666666865348816, 0.6792929172515869, 0.6355218887329102, 0.695286214351654, 0.7146464586257935, 0.6868686676025391, 0.6708754301071167, 0.7289562225341797, 0.7053872346878052, 0.6734007000923157, 0.7154881954193115, 0.7542087435722351, 0.7592592835426331, 0.7676767706871033, 0.7390572428703308, 0.7979797720909119, 0.7979797720909119, 0.7222222089767456, 0.8299663066864014, 0.7239057421684265, 0.8282828330993652, 0.8400673270225525, 0.8383838534355164, 0.8232323527336121, 0.8552188277244568, 0.7811447978019714, 0.8501683473587036, 0.8855218887329102, 0.7996633052825928, 0.8063973188400269, 0.8905723690986633, 0.9015151262283325, 0.8964646458625793, 0.8930976390838623, 0.9461279511451721, 0.939393937587738, 0.868686854839325, 0.941077470779419, 0.9259259104728699, 0.9949495196342468], [0.1515151560306549, 0.17845118045806885, 0.2373737394809723, 0.2710437774658203, 0.25084173679351807, 0.26430976390838623, 0.27441078424453735, 0.27272728085517883, 0.38131314516067505, 0.3223905861377716, 0.38552188873291016, 0.43265992403030396, 0.4250841736793518, 0.42845118045806885, NaN, 0.5555555820465088, 0.5168350338935852, 0.5774410963058472, 0.4932659864425659, 0.563973069190979, 0.4932659864425659, 0.5959596037864685, 0.554713785648346, 0.5723905563354492, 0.5723905563354492, 0.6212121248245239, 0.6010100841522217, 0.6498316526412964, 0.5387205481529236, 0.6843434572219849, 0.630471408367157, 0.5622895359992981, 0.6750841736793518, 0.6111111044883728, 0.6767676472663879, 0.6851851940155029, 0.6969696879386902, 0.6683501601219177, 0.6430976390838623, 0.7188552021980286, 0.6936026811599731, 0.6717171669006348, 0.6902356743812561, 0.7693602442741394, 0.755892276763916, 0.7693602442741394, 0.7407407164573669, 0.7138047218322754, 0.7592592835426331, 0.8097643256187439, 0.7508417367935181, 0.7247474789619446, 0.755892276763916, 0.8131313323974609, 0.8350168466567993, 0.8299663066864014, 0.8299663066864014, 0.8771044015884399, 0.816498339176178, 0.808922529220581, 0.8838383555412292, 0.8257575631141663, 0.8013467788696289, 0.9082491397857666, 0.8787878751754761, 0.8838383555412292, 0.8670033812522888, 0.8838383555412292, 0.8939393758773804, 0.8636363744735718, 0.9511784315109253, 0.8989899158477783, 0.9848484992980957], [0.10606060922145844, 0.14814814925193787, 0.19696970283985138, 0.21548821032047272, 0.21717171370983124, 0.2289562225341797, 0.25757575035095215, 0.2457912415266037, 0.3905723989009857, 0.3265993297100067, 0.3030303120613098, 0.38552188873291016, 0.34511783719062805, 0.436026930809021, 0.4444444477558136, NaN, 0.46717172861099243, 0.469696968793869, 0.48821547627449036, 0.563973069190979, 0.5202020406723022, 0.6085858345031738, 0.5412458181381226, 0.5799663066864014, 0.5269360542297363, 0.6557239294052124, 0.5808081030845642, 0.6818181872367859, 0.5909090638160706, 0.7138047218322754, 0.6346801519393921, 0.5808081030845642, 0.6641414165496826, 0.563973069190979, 0.6127946376800537, 0.6742424368858337, 0.6178451180458069, 0.6548821330070496, 0.7062289714813232, 0.7138047218322754, 0.6734007000923157, 0.7373737096786499, 0.7710437774658203, 0.7087541818618774, 0.7306397557258606, 0.7154881954193115, 0.7331649661064148, 0.808080792427063, 0.7373737096786499, 0.8653198480606079, 0.7516834735870361, 0.7407407164573669, 0.7542087435722351, 0.8131313323974609, 0.9478114247322083, 0.8653198480606079, 0.8888888955116272, 0.8316498398780823, 0.8367003202438354, 0.7845118045806885, 0.8922559022903442, 0.8569023609161377, 0.8215488195419312, 0.8417508602142334, 0.8265992999076843, 0.9124578833580017, 0.8888888955116272, 0.9292929172515869, 0.939393937587738, 0.9124578833580017, 0.9444444179534912, 0.930134654045105, 0.9932659864425659], [0.08922559022903442, 0.19276094436645508, 0.19528619945049286, 0.23232322931289673, 0.22558923065662384, 0.25084173679351807, 0.24410773813724518, 0.25, 0.4107744097709656, 0.3888888955116272, 0.3956229090690613, 0.3333333432674408, 0.42171716690063477, 0.46043771505355835, 0.4831649959087372, 0.5328282713890076, NaN, 0.5538720488548279, 0.49663299322128296, 0.560606062412262, 0.5269360542297363, 0.563973069190979, 0.563973069190979, 0.5791245698928833, 0.5993266105651855, 0.631313145160675, 0.5555555820465088, 0.6136363744735718, 0.5648148059844971, 0.631313145160675, 0.5824915766716003, 0.5555555820465088, 0.6498316526412964, 0.6212121248245239, 0.6632996797561646, 0.7222222089767456, 0.6691918969154358, 0.6750841736793518, 0.6936026811599731, 0.7138047218322754, 0.6632996797561646, 0.7003366947174072, 0.7289562225341797, 0.7676767706871033, 0.7483165264129639, 0.7407407164573669, 0.695286214351654, 0.753367006778717, 0.7121211886405945, 0.804713785648346, 0.7508417367935181, 0.7356902360916138, 0.7601010203361511, 0.7744107842445374, 0.8249158263206482, 0.7912458181381226, 0.8484848737716675, 0.8905723690986633, 0.8468013405799866, 0.816498339176178, 0.9040403962135315, 0.8585858345031738, 0.7996633052825928, 0.8636363744735718, 0.8282828330993652, 0.9124578833580017, 0.8535353541374207, 0.8771044015884399, 0.8888888955116272, 0.8888888955116272, 0.9545454382896423, 0.9057239294052124, 0.9797979593276978], [0.12037037312984467, 0.2222222238779068, 0.23148147761821747, 0.23232322931289673, 0.2457912415266037, 0.22558923065662384, 0.23653198778629303, 0.24410773813724518, 0.42592594027519226, 0.3670033812522888, 0.39393940567970276, 0.36026936769485474, 0.41750842332839966, 0.46296295523643494, 0.4225589334964752, 0.5303030014038086, 0.4461279511451721, NaN, 0.5168350338935852, 0.5218855142593384, 0.5134680271148682, 0.5404040217399597, 0.5336700081825256, 0.5496633052825928, 0.5530303120613098, 0.6127946376800537, 0.560606062412262, 0.5791245698928833, 0.5824915766716003, 0.5942760705947876, 0.5690235495567322, 0.5538720488548279, 0.6565656661987305, 0.5521885752677917, 0.6397306323051453, 0.627946138381958, 0.6439393758773804, 0.6481481194496155, 0.6557239294052124, 0.7003366947174072, 0.6212121248245239, 0.6683501601219177, 0.7424242496490479, 0.7104377150535583, 0.7188552021980286, 0.7255892157554626, 0.7087541818618774, 0.7306397557258606, 0.7255892157554626, 0.747474730014801, 0.7811447978019714, 0.7390572428703308, 0.7861952781677246, 0.7609427571296692, 0.7836700081825256, 0.7744107842445374, 0.8451178669929504, 0.8299663066864014, 0.8602693676948547, 0.8240740895271301, 0.8569023609161377, 0.8787878751754761, 0.8409090638160706, 0.8451178669929504, 0.7929292917251587, 0.875420868396759, 0.8838383555412292, 0.8535353541374207, 0.8619528412818909, 0.9343434572219849, 0.9259259104728699, 0.9562289714813232, 0.9882155060768127], [0.10774410516023636, 0.18686868250370026, 0.17003366351127625, 0.20370370149612427, 0.27272728085517883, 0.27272728085517883, 0.2887205481529236, 0.2735690176486969, 0.31986531615257263, 0.4570707082748413, 0.44781145453453064, 0.4511784613132477, 0.4612794518470764, 0.3501683473587036, 0.5067340135574341, 0.5117844939231873, 0.503367006778717, 0.4831649959087372, NaN, 0.5336700081825256, 0.5404040217399597, 0.5404040217399597, 0.5555555820465088, 0.5412458181381226, 0.5538720488548279, 0.5774410963058472, 0.5858585834503174, 0.5732323527336121, 0.6346801519393921, 0.5824915766716003, 0.5875421166419983, 0.5505050420761108, 0.5942760705947876, 0.6069023609161377, 0.624579131603241, 0.5942760705947876, 0.6397306323051453, 0.6414141654968262, 0.6481481194496155, 0.6346801519393921, 0.6515151262283325, 0.6582491397857666, 0.6430976390838623, 0.6616161465644836, 0.6936026811599731, 0.7188552021980286, 0.7121211886405945, 0.6868686676025391, 0.7390572428703308, 0.7239057421684265, 0.6843434572219849, 0.8265992999076843, 0.6851851940155029, 0.752525269985199, 0.7777777910232544, 0.7895622849464417, 0.7727272510528564, 0.8249158263206482, 0.7617844939231873, 0.8367003202438354, 0.8636363744735718, 0.7828282713890076, 0.7878788113594055, 0.8703703880310059, 0.8670033812522888, 0.8737373948097229, 0.8838383555412292, 0.9090909361839294, 0.9074074029922485, 0.8518518805503845, 0.939393937587738, 0.9141414165496826, 0.9915825128555298], [0.1026936024427414, 0.13636364042758942, 0.1734006702899933, 0.2138047069311142, 0.21043771505355835, 0.2053872048854828, 0.24074074625968933, 0.21885521709918976, 0.3468013405799866, 0.30471381545066833, 0.3080808222293854, 0.3872053921222687, 0.3316498398780823, 0.4191919267177582, 0.436026930809021, 0.436026930809021, 0.43939393758773804, 0.4781144857406616, 0.466329962015152, NaN, 0.5084174871444702, 0.48400673270225525, 0.5269360542297363, 0.5782828330993652, 0.5210437774658203, 0.5925925970077515, 0.560606062412262, 0.5749158263206482, 0.5370370149612427, 0.6102693676948547, 0.6077440977096558, 0.5723905563354492, 0.5993266105651855, 0.5656565427780151, 0.6060606241226196, 0.6212121248245239, 0.6010100841522217, 0.6178451180458069, 0.6531986594200134, 0.6439393758773804, 0.6683501601219177, 0.680134654045105, 0.68855220079422, 0.68855220079422, 0.7205387353897095, 0.683501660823822, 0.7390572428703308, 0.7222222089767456, 0.7567340135574341, 0.8484848737716675, 0.744107723236084, 0.695286214351654, 0.7390572428703308, 0.8148148059844971, 0.8703703880310059, 0.8611111044883728, 0.8425925970077515, 0.8299663066864014, 0.8063973188400269, 0.7356902360916138, 0.8417508602142334, 0.8232323527336121, 0.7895622849464417, 0.8333333134651184, 0.8484848737716675, 0.8602693676948547, 0.8821548819541931, 0.9057239294052124, 0.9259259104728699, 0.8569023609161377, 0.9074074029922485, 0.9107744097709656, 0.9932659864425659], [0.11784511804580688, 0.1750841736793518, 0.13804714381694794, 0.16161616146564484, 0.2222222238779068, 0.22727273404598236, 0.1986531913280487, 0.23821549117565155, 0.2735690176486969, 0.4511784613132477, 0.42845118045806885, 0.4276094138622284, 0.4343434274196625, 0.3998316526412964, 0.5067340135574341, 0.47979798913002014, 0.47306397557258606, 0.48653197288513184, 0.4595959484577179, 0.4915824830532074, NaN, 0.5016834735870361, 0.5353535413742065, 0.5143097639083862, 0.5370370149612427, 0.5420875549316406, 0.5622895359992981, 0.5538720488548279, 0.567340075969696, 0.5841751098632812, 0.5622895359992981, 0.5841751098632812, 0.5858585834503174, 0.5723905563354492, 0.619528591632843, 0.6111111044883728, 0.6936026811599731, 0.627946138381958, 0.6127946376800537, 0.6178451180458069, 0.6616161465644836, 0.6430976390838623, 0.6262626051902771, 0.6632996797561646, 0.691919207572937, 0.7272727489471436, 0.6734007000923157, 0.7188552021980286, 0.695286214351654, 0.695286214351654, 0.752525269985199, 0.8670033812522888, 0.7575757503509521, 0.6936026811599731, 0.7819865345954895, 0.747474730014801, 0.7861952781677246, 0.8181818127632141, 0.7609427571296692, 0.9057239294052124, 0.8838383555412292, 0.7878788113594055, 0.8265992999076843, 0.8552188277244568, 0.819865345954895, 0.8872053623199463, 0.8518518805503845, 0.8569023609161377, 0.8872053623199463, 0.8863636255264282, 0.9309764504432678, 0.9427609443664551, 0.9949495196342468], [0.09090909361839294, 0.14478114247322083, 0.15656565129756927, 0.1818181872367859, 0.1666666716337204, 0.19781145453453064, 0.21885521709918976, 0.1986531913280487, 0.3787878751754761, 0.27693602442741394, 0.2777777910232544, 0.34427610039711, 0.32996633648872375, 0.4040403962135315, 0.4040403962135315, 0.3914141356945038, 0.436026930809021, 0.4595959484577179, 0.4595959484577179, 0.5159932374954224, 0.4983164966106415, NaN, 0.48653197288513184, 0.5252525210380554, 0.47979798913002014, 0.6119528412818909, 0.5496633052825928, 0.5867003202438354, 0.5538720488548279, 0.624579131603241, 0.5841751098632812, 0.5420875549316406, 0.6329966187477112, 0.5336700081825256, 0.5824915766716003, 0.6380471587181091, 0.5917508602142334, 0.6296296119689941, 0.6548821330070496, 0.6784511804580688, 0.624579131603241, 0.6767676472663879, 0.7239057421684265, 0.689393937587738, 0.686026930809021, 0.6616161465644836, 0.691077470779419, 0.7626262903213501, 0.7230639457702637, 0.8291245698928833, 0.7407407164573669, 0.7306397557258606, 0.7457912564277649, 0.7575757503509521, 0.9166666865348816, 0.8249158263206482, 0.8678451180458069, 0.8131313323974609, 0.8131313323974609, 0.7727272510528564, 0.874579131603241, 0.8392255902290344, 0.808080792427063, 0.8265992999076843, 0.813973069190979, 0.8939393758773804, 0.8636363744735718, 0.9175084233283997, 0.9158248901367188, 0.8956229090690613, 0.9427609443664551, 0.9208754301071167, 0.9949495196342468], [0.09090909361839294, 0.11447811126708984, 0.14814814925193787, 0.16835017502307892, 0.17003366351127625, 0.22474747896194458, 0.26430976390838623, 0.24747474491596222, 0.30639731884002686, 0.3030303120613098, 0.2946127951145172, 0.3829966187477112, 0.28703704476356506, 0.372053861618042, 0.44528618454933167, 0.45875421166419983, 0.436026930809021, 0.466329962015152, 0.4444444477558136, 0.47306397557258606, 0.46464645862579346, 0.5134680271148682, NaN, 0.503367006778717, 0.5420875549316406, 0.5404040217399597, 0.5782828330993652, 0.5429292917251587, 0.5303030014038086, 0.5976430773735046, 0.625420868396759, 0.5471380352973938, 0.6010100841522217, 0.5707070827484131, 0.5976430773735046, 0.5656565427780151, 0.6430976390838623, 0.6565656661987305, 0.6565656661987305, 0.6414141654968262, 0.7062289714813232, 0.6666666865348816, 0.6717171669006348, 0.7070707082748413, 0.7037037014961243, 0.7146464586257935, 0.7062289714813232, 0.7146464586257935, 0.7390572428703308, 0.8030303120613098, 0.6969696879386902, 0.691077470779419, 0.6978114247322083, 0.7861952781677246, 0.8535353541374207, 0.8232323527336121, 0.8434343338012695, 0.8299663066864014, 0.747474730014801, 0.804713785648346, 0.8518518805503845, 0.7828282713890076, 0.7861952781677246, 0.8619528412818909, 0.8787878751754761, 0.872053861618042, 0.8468013405799866, 0.9090909361839294, 0.9090909361839294, 0.8367003202438354, 0.9124578833580017, 0.8863636255264282, 0.9814814925193787], [0.09090909361839294, 0.14057238399982452, 0.1666666716337204, 0.2070707082748413, 0.21717171370983124, 0.2053872048854828, 0.23400673270225525, 0.2222222238779068, 0.35353535413742065, 0.2946127951145172, 0.28282827138900757, 0.3670033812522888, 0.32996633648872375, 0.4124579131603241, 0.4276094138622284, 0.42003366351127625, 0.4208754301071167, 0.4503366947174072, 0.45875421166419983, 0.42171716690063477, 0.48569023609161377, 0.4747474789619446, 0.49663299322128296, NaN, 0.5, 0.5782828330993652, 0.5462962985038757, 0.5757575631141663, 0.5269360542297363, 0.6077440977096558, 0.5942760705947876, 0.5572390556335449, 0.6026936173439026, 0.558922529220581, 0.6043770909309387, 0.6102693676948547, 0.5757575631141663, 0.5993266105651855, 0.618686854839325, 0.631313145160675, 0.6346801519393921, 0.6590909361839294, 0.691919207572937, 0.6767676472663879, 0.7079124450683594, 0.6742424368858337, 0.7171717286109924, 0.7104377150535583, 0.7289562225341797, 0.8417508602142334, 0.7424242496490479, 0.691919207572937, 0.744107723236084, 0.8063973188400269, 0.8434343338012695, 0.8400673270225525, 0.8299663066864014, 0.8097643256187439, 0.8013467788696289, 0.7289562225341797, 0.8316498398780823, 0.8156565427780151, 0.7946127653121948, 0.819865345954895, 0.8367003202438354, 0.8535353541374207, 0.8737373948097229, 0.8939393758773804, 0.9175084233283997, 0.8501683473587036, 0.9141414165496826, 0.9175084233283997, 0.9932659864425659], [0.09090909361839294, 0.12457912415266037, 0.14646464586257935, 0.17003366351127625, 0.17171716690063477, 0.17676767706871033, 0.2373737394809723, 0.22053872048854828, 0.34343433380126953, 0.3333333432674408, 0.34511783719062805, 0.3611111044883728, 0.3316498398780823, 0.3872053921222687, 0.4276094138622284, 0.47306397557258606, 0.40067338943481445, 0.4469696879386902, 0.4461279511451721, 0.4789562225341797, 0.46296295523643494, 0.5202020406723022, 0.4579124450683594, 0.5, NaN, 0.5370370149612427, 0.5496633052825928, 0.5370370149612427, 0.5168350338935852, 0.5446127653121948, 0.6010100841522217, 0.5303030014038086, 0.5740740895271301, 0.5008417367935181, 0.5757575631141663, 0.5505050420761108, 0.5951178669929504, 0.6750841736793518, 0.6212121248245239, 0.619528591632843, 0.6144781112670898, 0.6287878751754761, 0.6363636255264282, 0.680134654045105, 0.7154881954193115, 0.6936026811599731, 0.680134654045105, 0.6902356743812561, 0.7037037014961243, 0.7222222089767456, 0.7138047218322754, 0.7104377150535583, 0.7205387353897095, 0.744107723236084, 0.7626262903213501, 0.7643097639083862, 0.8063973188400269, 0.804713785648346, 0.7626262903213501, 0.7979797720909119, 0.8299663066864014, 0.7929292917251587, 0.7912458181381226, 0.8341751098632812, 0.8703703880310059, 0.8501683473587036, 0.8400673270225525, 0.880471408367157, 0.8670033812522888, 0.8468013405799866, 0.9006733894348145, 0.8872053623199463, 0.9831649661064148], [0.09427609294652939, 0.12878787517547607, 0.15319865942001343, 0.17845118045806885, 0.17845118045806885, 0.19528619945049286, 0.2070707082748413, 0.20875421166419983, 0.3249158263206482, 0.28787878155708313, 0.25925925374031067, 0.3030303120613098, 0.3030303120613098, 0.37542086839675903, 0.3787878751754761, 0.34427610039711, 0.36868685483932495, 0.3872053921222687, 0.4225589334964752, 0.40740740299224854, 0.4579124450683594, 0.38804712891578674, 0.4595959484577179, 0.42171716690063477, 0.46296295523643494, NaN, 0.4957912564277649, 0.47727271914482117, 0.5286195278167725, 0.5143097639083862, 0.5622895359992981, 0.5454545617103577, 0.563973069190979, 0.5117844939231873, 0.5723905563354492, 0.5656565427780151, 0.5707070827484131, 0.567340075969696, 0.625420868396759, 0.624579131603241, 0.6178451180458069, 0.6683501601219177, 0.6649831533432007, 0.6481481194496155, 0.68855220079422, 0.6582491397857666, 0.680134654045105, 0.7424242496490479, 0.7003366947174072, 0.7693602442741394, 0.7382155060768127, 0.6784511804580688, 0.7407407164573669, 0.7659932374954224, 0.8232323527336121, 0.8013467788696289, 0.875420868396759, 0.816498339176178, 0.7794612646102905, 0.747474730014801, 0.8215488195419312, 0.8013467788696289, 0.7979797720909119, 0.8350168466567993, 0.8013467788696289, 0.8569023609161377, 0.8653198480606079, 0.8787878751754761, 0.8872053623199463, 0.8703703880310059, 0.9006733894348145, 0.9225589036941528, 0.9898989796638489], [0.09090909361839294, 0.12121212482452393, 0.15656565129756927, 0.18855218589305878, 0.19696970283985138, 0.19360269606113434, 0.19696970283985138, 0.2095959633588791, 0.3232323229312897, 0.27441078424453735, 0.30892255902290344, 0.38552188873291016, 0.3097642958164215, 0.36447811126708984, 0.39898988604545593, 0.4191919267177582, 0.4444444477558136, 0.43939393758773804, 0.4141414165496826, 0.43939393758773804, 0.4377104341983795, 0.4503366947174072, 0.42171716690063477, 0.45370370149612427, 0.4503366947174072, 0.5042087435722351, NaN, 0.48232322931289673, 0.4898989796638489, 0.5050504803657532, 0.6085858345031738, 0.48821547627449036, 0.5471380352973938, 0.5117844939231873, 0.5235690474510193, 0.5454545617103577, 0.5833333134651184, 0.5555555820465088, 0.5942760705947876, 0.6144781112670898, 0.566498339176178, 0.6043770909309387, 0.6363636255264282, 0.6262626051902771, 0.631313145160675, 0.6632996797561646, 0.6565656661987305, 0.6296296119689941, 0.7003366947174072, 0.7020202279090881, 0.6961279511451721, 0.7323232293128967, 0.695286214351654, 0.7407407164573669, 0.7188552021980286, 0.7811447978019714, 0.7457912564277649, 0.8131313323974609, 0.819865345954895, 0.8131313323974609, 0.8148148059844971, 0.8367003202438354, 0.7676767706871033, 0.880471408367157, 0.941077470779419, 0.8493266105651855, 0.8829966187477112, 0.939393937587738, 0.9276094436645508, 0.869528591632843, 0.9090909361839294, 0.9090909361839294, 0.9915825128555298], [0.07575757801532745, 0.12457912415266037, 0.16582491993904114, 0.19696970283985138, 0.1835016906261444, 0.19528619945049286, 0.2222222238779068, 0.21212121844291687, 0.33670035004615784, 0.2306397259235382, 0.2525252401828766, 0.3097642958164215, 0.29629629850387573, 0.3947811424732208, 0.3501683473587036, 0.3181818127632141, 0.3863636255264282, 0.4208754301071167, 0.4267676770687103, 0.4250841736793518, 0.4461279511451721, 0.41329964995384216, 0.4570707082748413, 0.42424243688583374, 0.46296295523643494, 0.5227272510528564, 0.5176767706871033, NaN, 0.5, 0.5917508602142334, 0.5757575631141663, 0.5252525210380554, 0.560606062412262, 0.503367006778717, 0.5782828330993652, 0.5774410963058472, 0.563973069190979, 0.5976430773735046, 0.5740740895271301, 0.624579131603241, 0.6010100841522217, 0.6144781112670898, 0.6616161465644836, 0.6397306323051453, 0.6531986594200134, 0.6228955984115601, 0.6649831533432007, 0.6574074029922485, 0.7087541818618774, 0.8821548819541931, 0.7205387353897095, 0.689393937587738, 0.7205387353897095, 0.7971380352973938, 0.8872053623199463, 0.8223905563354492, 0.7996633052825928, 0.7929292917251587, 0.7946127653121948, 0.7382155060768127, 0.7979797720909119, 0.8063973188400269, 0.7744107842445374, 0.7912458181381226, 0.8232323527336121, 0.8240740895271301, 0.8434343338012695, 0.8855218887329102, 0.9175084233283997, 0.8484848737716675, 0.8989899158477783, 0.8964646458625793, 0.9848484992980957], [0.09595959633588791, 0.2053872048854828, 0.1582491546869278, 0.18013468384742737, 0.2626262605190277, 0.29629629850387573, 0.26599326729774475, 0.28956228494644165, 0.19276094436645508, 0.4107744097709656, 0.3905723989009857, 0.4040403962135315, 0.4208754301071167, 0.32154881954193115, 0.4612794518470764, 0.40909090638160706, 0.43518519401550293, 0.41750842332839966, 0.3653198778629303, 0.46296295523643494, 0.43265992403030396, 0.4461279511451721, 0.469696968793869, 0.47306397557258606, 0.4831649959087372, 0.47138047218322754, 0.5101010203361511, 0.5, NaN, 0.5185185074806213, 0.5168350338935852, 0.5151515007019043, 0.4983164966106415, 0.5151515007019043, 0.5336700081825256, 0.5437710285186768, 0.5538720488548279, 0.5707070827484131, 0.5774410963058472, 0.5328282713890076, 0.5572390556335449, 0.5875421166419983, 0.5622895359992981, 0.5513467788696289, 0.6026936173439026, 0.6363636255264282, 0.6548821330070496, 0.6296296119689941, 0.6818181872367859, 0.6329966187477112, 0.6380471587181091, 0.7424242496490479, 0.6380471587181091, 0.6717171669006348, 0.691919207572937, 0.7222222089767456, 0.7407407164573669, 0.7508417367935181, 0.695286214351654, 0.8434343338012695, 0.7929292917251587, 0.7239057421684265, 0.7340067625045776, 0.8299663066864014, 0.7962962985038757, 0.819865345954895, 0.8350168466567993, 0.8585858345031738, 0.8316498398780823, 0.8299663066864014, 0.8922559022903442, 0.875420868396759, 0.9932659864425659], [0.07070706784725189, 0.12121212482452393, 0.1498316526412964, 0.18686868250370026, 0.15488214790821075, 0.1776094287633896, 0.19191919267177582, 0.19528619945049286, 0.3232323229312897, 0.2070707082748413, 0.2609427571296692, 0.27272728085517883, 0.2777777910232544, 0.36195287108421326, 0.3156565725803375, 0.2861952781677246, 0.36868685483932495, 0.40572389960289, 0.41750842332839966, 0.38973063230514526, 0.41582491993904114, 0.37542086839675903, 0.402356892824173, 0.39225590229034424, 0.4553872048854828, 0.48569023609161377, 0.49494948983192444, 0.408249169588089, 0.48148149251937866, NaN, 0.5412458181381226, 0.5168350338935852, 0.5404040217399597, 0.4983164966106415, 0.5454545617103577, 0.5488215684890747, 0.5336700081825256, 0.5656565427780151, 0.5277777910232544, 0.5959596037864685, 0.5774410963058472, 0.5774410963058472, 0.6296296119689941, 0.6127946376800537, 0.6296296119689941, 0.6212121248245239, 0.6658248901367188, 0.6161616444587708, 0.7037037014961243, 0.8249158263206482, 0.7070707082748413, 0.6717171669006348, 0.7095959782600403, 0.7744107842445374, 0.8518518805503845, 0.8131313323974609, 0.7912458181381226, 0.7727272510528564, 0.7828282713890076, 0.7289562225341797, 0.7727272510528564, 0.804713785648346, 0.7710437774658203, 0.7811447978019714, 0.81144779920578, 0.816498339176178, 0.8442760705947876, 0.8872053623199463, 0.9040403962135315, 0.8417508602142334, 0.8973063826560974, 0.8939393758773804, 0.9814814925193787], [0.10437710583209991, 0.14562289416790009, 0.15993265807628632, 0.19949494302272797, 0.19696970283985138, 0.19528619945049286, 0.18855218589305878, 0.20875421166419983, 0.2946127951145172, 0.2861952781677246, 0.28282827138900757, 0.3821548819541931, 0.31060606241226196, 0.3459596037864685, 0.3695286214351654, 0.3653198778629303, 0.41750842332839966, 0.43097642064094543, 0.4124579131603241, 0.39225590229034424, 0.4377104341983795, 0.41582491993904114, 0.37457913160324097, 0.40572389960289, 0.39898988604545593, 0.4377104341983795, 0.3914141356945038, 0.42424243688583374, 0.4831649959087372, 0.45875421166419983, NaN, 0.5, 0.505892276763916, 0.42592594027519226, 0.5067340135574341, 0.5235690474510193, 0.5496633052825928, 0.5404040217399597, 0.558080792427063, 0.5740740895271301, 0.5404040217399597, 0.5808081030845642, 0.5976430773735046, 0.5808081030845642, 0.6127946376800537, 0.6397306323051453, 0.6397306323051453, 0.6144781112670898, 0.6851851940155029, 0.6094276309013367, 0.7037037014961243, 0.7053872346878052, 0.7087541818618774, 0.6717171669006348, 0.6734007000923157, 0.7272727489471436, 0.747474730014801, 0.8063973188400269, 0.816498339176178, 0.8249158263206482, 0.7937710285186768, 0.8400673270225525, 0.7861952781677246, 0.8619528412818909, 0.9040403962135315, 0.8308081030845642, 0.872053861618042, 0.8855218887329102, 0.8653198480606079, 0.8872053623199463, 0.8989899158477783, 0.9141414165496826, 0.9915825128555298], [0.14141413569450378, 0.2373737394809723, 0.23232322931289673, 0.26767677068710327, 0.2289562225341797, 0.23148147761821747, 0.21717171370983124, 0.2390572428703308, 0.3585858643054962, 0.3737373650074005, 0.4040403962135315, 0.3670033812522888, 0.37037035822868347, 0.3905723989009857, 0.4377104341983795, 0.4191919267177582, 0.4444444477558136, 0.4461279511451721, 0.44949495792388916, 0.4276094138622284, 0.41582491993904114, 0.4579124450683594, 0.4528619647026062, 0.4427609443664551, 0.469696968793869, 0.4545454680919647, 0.5117844939231873, 0.4747474789619446, 0.4848484992980957, 0.4831649959087372, 0.5, NaN, 0.4932659864425659, 0.5084174871444702, 0.5808081030845642, 0.5336700081825256, 0.5909090638160706, 0.5420875549316406, 0.4831649959087372, 0.5269360542297363, 0.5227272510528564, 0.5117844939231873, 0.5597643256187439, 0.6346801519393921, 0.6397306323051453, 0.6649831533432007, 0.6127946376800537, 0.5505050420761108, 0.6380471587181091, 0.5942760705947876, 0.6127946376800537, 0.6414141654968262, 0.6094276309013367, 0.6523569226264954, 0.6616161465644836, 0.6675084233283997, 0.6599326729774475, 0.7356902360916138, 0.7609427571296692, 0.7508417367935181, 0.7744107842445374, 0.7693602442741394, 0.7643097639083862, 0.7710437774658203, 0.7727272510528564, 0.7819865345954895, 0.7424242496490479, 0.7794612646102905, 0.7845118045806885, 0.81144779920578, 0.8131313323974609, 0.8518518805503845, 0.9848484992980957], [0.09090909361839294, 0.13804714381694794, 0.15656565129756927, 0.17845118045806885, 0.18013468384742737, 0.18013468384742737, 0.1902356892824173, 0.18013468384742737, 0.3181818127632141, 0.2306397259235382, 0.25589224696159363, 0.2525252401828766, 0.28787878155708313, 0.3594276010990143, 0.3249158263206482, 0.3358585834503174, 0.3501683473587036, 0.34343433380126953, 0.40572389960289, 0.40067338943481445, 0.4141414165496826, 0.3670033812522888, 0.39898988604545593, 0.3973063826560974, 0.42592594027519226, 0.436026930809021, 0.4528619647026062, 0.43939393758773804, 0.5016834735870361, 0.4595959484577179, 0.49410775303840637, 0.5067340135574341, NaN, 0.4444444477558136, 0.5319865345954895, 0.5303030014038086, 0.5176767706871033, 0.5437710285186768, 0.5185185074806213, 0.689393937587738, 0.5370370149612427, 0.5505050420761108, 0.7340067625045776, 0.6094276309013367, 0.6388888955116272, 0.5934343338012695, 0.6641414165496826, 0.6212121248245239, 0.6700336933135986, 0.6927609443664551, 0.7205387353897095, 0.68855220079422, 0.7205387353897095, 0.7104377150535583, 0.7087541818618774, 0.7323232293128967, 0.7946127653121948, 0.7609427571296692, 0.8148148059844971, 0.7744107842445374, 0.7996633052825928, 0.8333333134651184, 0.7962962985038757, 0.816498339176178, 0.7643097639083862, 0.817340075969696, 0.8417508602142334, 0.8417508602142334, 0.8417508602142334, 0.8653198480606079, 0.8973063826560974, 0.9099326729774475, 0.9831649661064148], [0.06397306174039841, 0.08249158412218094, 0.11279461532831192, 0.15993265807628632, 0.12289562076330185, 0.1835016906261444, 0.23400673270225525, 0.19276094436645508, 0.31986531615257263, 0.31481480598449707, 0.30471381545066833, 0.3265993297100067, 0.3181818127632141, 0.3333333432674408, 0.3888888955116272, 0.436026930809021, 0.3787878751754761, 0.44781145453453064, 0.3930976390838623, 0.4343434274196625, 0.4276094138622284, 0.466329962015152, 0.4292929172515869, 0.44107744097709656, 0.49915826320648193, 0.48821547627449036, 0.48821547627449036, 0.49663299322128296, 0.4848484992980957, 0.5016834735870361, 0.5740740895271301, 0.4915824830532074, 0.5555555820465088, NaN, 0.5151515007019043, 0.5168350338935852, 0.5488215684890747, 0.6481481194496155, 0.5538720488548279, 0.5951178669929504, 0.5656565427780151, 0.5858585834503174, 0.6026936173439026, 0.6515151262283325, 0.624579131603241, 0.6262626051902771, 0.6296296119689941, 0.6010100841522217, 0.6632996797561646, 0.7087541818618774, 0.6414141654968262, 0.619528591632843, 0.6430976390838623, 0.683501660823822, 0.7205387353897095, 0.68855220079422, 0.7205387353897095, 0.7272727489471436, 0.6818181872367859, 0.7037037014961243, 0.7508417367935181, 0.6969696879386902, 0.7070707082748413, 0.7609427571296692, 0.8358585834503174, 0.752525269985199, 0.7744107842445374, 0.8232323527336121, 0.8148148059844971, 0.744107723236084, 0.8215488195419312, 0.804713785648346, 0.9494949579238892], [0.05218855291604996, 0.09764309972524643, 0.11279461532831192, 0.13299663364887238, 0.12794612348079681, 0.14646464586257935, 0.15319865942001343, 0.15488214790821075, 0.27441078424453735, 0.26430976390838623, 0.22390572726726532, 0.29797980189323425, 0.3080808222293854, 0.3207070827484131, 0.3232323229312897, 0.3872053921222687, 0.33670035004615784, 0.36026936769485474, 0.37542086839675903, 0.39393940567970276, 0.3804713785648346, 0.41750842332839966, 0.402356892824173, 0.3956229090690613, 0.42424243688583374, 0.4276094138622284, 0.4764309823513031, 0.42171716690063477, 0.466329962015152, 0.4545454680919647, 0.4932659864425659, 0.4191919267177582, 0.4680134654045105, 0.4848484992980957, NaN, 0.46296295523643494, 0.5774410963058472, 0.5538720488548279, 0.5151515007019043, 0.5303030014038086, 0.554713785648346, 0.5521885752677917, 0.563973069190979, 0.5993266105651855, 0.7592592835426331, 0.6877104640007019, 0.6464646458625793, 0.5723905563354492, 0.6599326729774475, 0.6750841736793518, 0.6936026811599731, 0.631313145160675, 0.7011784315109253, 0.6969696879386902, 0.691919207572937, 0.7222222089767456, 0.7752525210380554, 0.8737373948097229, 0.7685185074806213, 0.7777777910232544, 0.8063973188400269, 0.8030303120613098, 0.7878788113594055, 0.8442760705947876, 0.7845118045806885, 0.8417508602142334, 0.804713785648346, 0.8030303120613098, 0.8013467788696289, 0.8602693676948547, 0.8989899158477783, 0.8922559022903442, 0.9797979593276978], [0.08417508751153946, 0.13804714381694794, 0.15488214790821075, 0.16835017502307892, 0.16582491993904114, 0.18686868250370026, 0.18686868250370026, 0.19360269606113434, 0.3316498398780823, 0.2222222238779068, 0.2710437774658203, 0.15572389960289001, 0.32996633648872375, 0.36447811126708984, 0.31481480598449707, 0.32575756311416626, 0.2777777910232544, 0.372053861618042, 0.40572389960289, 0.3787878751754761, 0.3888888955116272, 0.36195287108421326, 0.4343434274196625, 0.38973063230514526, 0.44949495792388916, 0.4343434274196625, 0.4545454680919647, 0.4225589334964752, 0.45622894167900085, 0.4511784613132477, 0.4764309823513031, 0.466329962015152, 0.469696968793869, 0.4831649959087372, 0.5370370149612427, NaN, 0.5690235495567322, 0.5488215684890747, 0.5, 0.5437710285186768, 0.5361952781677246, 0.5235690474510193, 0.5513467788696289, 0.6111111044883728, 0.627946138381958, 0.6515151262283325, 0.6161616444587708, 0.5875421166419983, 0.6414141654968262, 0.6363636255264282, 0.7239057421684265, 0.6632996797561646, 0.7306397557258606, 0.7037037014961243, 0.6498316526412964, 0.7154881954193115, 0.747474730014801, 0.7811447978019714, 0.7693602442741394, 0.7340067625045776, 0.7904040217399597, 0.7853535413742065, 0.804713785648346, 0.7946127653121948, 0.7424242496490479, 0.7962962985038757, 0.81144779920578, 0.8299663066864014, 0.8282828330993652, 0.8400673270225525, 0.880471408367157, 0.9090909361839294, 0.9764309525489807], [0.10101009905338287, 0.13804714381694794, 0.15909090638160706, 0.16329966485500336, 0.16835017502307892, 0.18097643554210663, 0.16329966485500336, 0.19696970283985138, 0.24915824830532074, 0.2095959633588791, 0.2390572428703308, 0.2861952781677246, 0.32407405972480774, 0.30471381545066833, 0.3030303120613098, 0.3821548819541931, 0.3308080732822418, 0.35606059432029724, 0.36026936769485474, 0.39898988604545593, 0.30639731884002686, 0.408249169588089, 0.3569023609161377, 0.42424243688583374, 0.40488216280937195, 0.4292929172515869, 0.4166666567325592, 0.436026930809021, 0.4461279511451721, 0.466329962015152, 0.4503366947174072, 0.40909090638160706, 0.48232322931289673, 0.4511784613132477, 0.4225589334964752, 0.43097642064094543, NaN, 0.5370370149612427, 0.5353535413742065, 0.5319865345954895, 0.4747474789619446, 0.5572390556335449, 0.5437710285186768, 0.566498339176178, 0.5622895359992981, 0.7087541818618774, 0.6026936173439026, 0.560606062412262, 0.6127946376800537, 0.6111111044883728, 0.6447811722755432, 0.5875421166419983, 0.6498316526412964, 0.6877104640007019, 0.6506733894348145, 0.6986532211303711, 0.6969696879386902, 0.7601010203361511, 0.7710437774658203, 0.7811447978019714, 0.7609427571296692, 0.7996633052825928, 0.7626262903213501, 0.9511784315109253, 0.7845118045806885, 0.7895622849464417, 0.804713785648346, 0.8265992999076843, 0.7895622849464417, 0.8585858345031738, 0.8855218887329102, 0.8821548819541931, 0.9797979593276978], [0.08080808073282242, 0.12962962687015533, 0.11868686974048615, 0.12205386906862259, 0.1346801370382309, 0.16835017502307892, 0.1666666716337204, 0.1835016906261444, 0.24410773813724518, 0.26767677068710327, 0.26430976390838623, 0.28282827138900757, 0.2760942876338959, 0.28535354137420654, 0.3316498398780823, 0.34511783719062805, 0.3249158263206482, 0.35185185074806213, 0.3585858643054962, 0.3821548819541931, 0.372053861618042, 0.37037035822868347, 0.34343433380126953, 0.40067338943481445, 0.3249158263206482, 0.43265992403030396, 0.4444444477558136, 0.402356892824173, 0.4292929172515869, 0.4343434274196625, 0.4595959484577179, 0.4579124450683594, 0.45622894167900085, 0.35185185074806213, 0.4461279511451721, 0.4511784613132477, 0.46296295523643494, NaN, 0.5707070827484131, 0.5117844939231873, 0.46717172861099243, 0.5858585834503174, 0.5505050420761108, 0.5134680271148682, 0.5420875549316406, 0.5437710285186768, 0.5757575631141663, 0.6228955984115601, 0.6228955984115601, 0.5791245698928833, 0.6649831533432007, 0.691919207572937, 0.6717171669006348, 0.6515151262283325, 0.6700336933135986, 0.7171717286109924, 0.7306397557258606, 0.7794612646102905, 0.7373737096786499, 0.7710437774658203, 0.7845118045806885, 0.7651515007019043, 0.7744107842445374, 0.7895622849464417, 0.7575757503509521, 0.8148148059844971, 0.8451178669929504, 0.8358585834503174, 0.8367003202438354, 0.8602693676948547, 0.8670033812522888, 0.8973063826560974, 0.9865319728851318], [0.0791245773434639, 0.08585858345031738, 0.1111111119389534, 0.14814814925193787, 0.12457912415266037, 0.16919191181659698, 0.17845118045806885, 0.186026930809021, 0.2609427571296692, 0.2457912415266037, 0.21632996201515198, 0.2542087435722351, 0.24242424964904785, 0.31313130259513855, 0.3569023609161377, 0.29377105832099915, 0.30639731884002686, 0.34427610039711, 0.35185185074806213, 0.3468013405799866, 0.3872053921222687, 0.34511783719062805, 0.34343433380126953, 0.38131314516067505, 0.3787878751754761, 0.37457913160324097, 0.40572389960289, 0.42592594027519226, 0.4225589334964752, 0.4722222089767456, 0.4419191777706146, 0.5168350338935852, 0.48148149251937866, 0.4461279511451721, 0.4848484992980957, 0.5, 0.46464645862579346, 0.4292929172515869, NaN, 0.5202020406723022, 0.5185185074806213, 0.5951178669929504, 0.5370370149612427, 0.5387205481529236, 0.5791245698928833, 0.5277777910232544, 0.5572390556335449, 0.7744107842445374, 0.5976430773735046, 0.6447811722755432, 0.695286214351654, 0.6363636255264282, 0.695286214351654, 0.6481481194496155, 0.7314814925193787, 0.6936026811599731, 0.7643097639083862, 0.7003366947174072, 0.7070707082748413, 0.744107723236084, 0.7491582632064819, 0.7272727489471436, 0.7710437774658203, 0.7710437774658203, 0.7659932374954224, 0.7878788113594055, 0.7693602442741394, 0.804713785648346, 0.7912458181381226, 0.8232323527336121, 0.8400673270225525, 0.8636363744735718, 0.9789562225341797], [0.0791245773434639, 0.1262626200914383, 0.13804714381694794, 0.1515151560306549, 0.14225588738918304, 0.15572389960289001, 0.1582491546869278, 0.1666666716337204, 0.280303031206131, 0.2138047069311142, 0.2390572428703308, 0.2138047069311142, 0.2626262605190277, 0.3291245698928833, 0.28114476799964905, 0.2861952781677246, 0.2861952781677246, 0.2996633052825928, 0.3653198778629303, 0.35606059432029724, 0.3821548819541931, 0.32154881954193115, 0.3585858643054962, 0.36868685483932495, 0.3804713785648346, 0.37542086839675903, 0.38552188873291016, 0.37542086839675903, 0.46717172861099243, 0.4040403962135315, 0.42592594027519226, 0.47306397557258606, 0.31060606241226196, 0.40488216280937195, 0.469696968793869, 0.45622894167900085, 0.4680134654045105, 0.48821547627449036, 0.47979798913002014, NaN, 0.48148149251937866, 0.505892276763916, 0.6422559022903442, 0.5286195278167725, 0.5858585834503174, 0.566498339176178, 0.6010100841522217, 0.5740740895271301, 0.6397306323051453, 0.6414141654968262, 0.7020202279090881, 0.6523569226264954, 0.7079124450683594, 0.6649831533432007, 0.6599326729774475, 0.7037037014961243, 0.744107723236084, 0.7356902360916138, 0.7794612646102905, 0.753367006778717, 0.7777777910232544, 0.8013467788696289, 0.7794612646102905, 0.7811447978019714, 0.7407407164573669, 0.7946127653121948, 0.8148148059844971, 0.8282828330993652, 0.8215488195419312, 0.8400673270225525, 0.8787878751754761, 0.9107744097709656, 0.9831649661064148], [0.07070706784725189, 0.09427609294652939, 0.0917508453130722, 0.1346801370382309, 0.11784511804580688, 0.14478114247322083, 0.1388888955116272, 0.15656565129756927, 0.2264309823513031, 0.21717171370983124, 0.21548821032047272, 0.26767677068710327, 0.2626262605190277, 0.2710437774658203, 0.30639731884002686, 0.3265993297100067, 0.33670035004615784, 0.3787878751754761, 0.3484848439693451, 0.3316498398780823, 0.33838382363319397, 0.37542086839675903, 0.29377105832099915, 0.3653198778629303, 0.38552188873291016, 0.3821548819541931, 0.4335016906261444, 0.39898988604545593, 0.4427609443664551, 0.4225589334964752, 0.4595959484577179, 0.47727271914482117, 0.46296295523643494, 0.4343434274196625, 0.44528618454933167, 0.4638047218322754, 0.5252525210380554, 0.5328282713890076, 0.48148149251937866, 0.5185185074806213, NaN, 0.5328282713890076, 0.558922529220581, 0.5622895359992981, 0.5707070827484131, 0.5892255902290344, 0.5707070827484131, 0.563973069190979, 0.6296296119689941, 0.6161616444587708, 0.6776094436645508, 0.6010100841522217, 0.6717171669006348, 0.6649831533432007, 0.6700336933135986, 0.7104377150535583, 0.746632993221283, 0.7828282713890076, 0.6725589036941528, 0.7163299918174744, 0.7390572428703308, 0.7138047218322754, 0.7491582632064819, 0.8030303120613098, 0.7895622849464417, 0.7794612646102905, 0.7811447978019714, 0.81144779920578, 0.8131313323974609, 0.7929292917251587, 0.8535353541374207, 0.8459596037864685, 0.9696969985961914], [0.07070706784725189, 0.08249158412218094, 0.11279461532831192, 0.13636364042758942, 0.11784511804580688, 0.1582491546869278, 0.16835017502307892, 0.16498316824436188, 0.23400673270225525, 0.22138047218322754, 0.19360269606113434, 0.2222222238779068, 0.23400673270225525, 0.2946127951145172, 0.32828283309936523, 0.2626262605190277, 0.2996633052825928, 0.3316498398780823, 0.341750830411911, 0.31986531615257263, 0.3569023609161377, 0.3232323229312897, 0.3333333432674408, 0.34090909361839294, 0.3712121248245239, 0.3316498398780823, 0.3956229090690613, 0.38552188873291016, 0.4124579131603241, 0.4225589334964752, 0.4191919267177582, 0.48821547627449036, 0.44949495792388916, 0.4141414165496826, 0.44781145453453064, 0.4764309823513031, 0.4427609443664551, 0.4141414165496826, 0.40488216280937195, 0.49410775303840637, 0.46717172861099243, NaN, 0.5117844939231873, 0.5218855142593384, 0.5488215684890747, 0.5042087435722351, 0.5353535413742065, 0.695286214351654, 0.5824915766716003, 0.6271044015884399, 0.6818181872367859, 0.6262626051902771, 0.6818181872367859, 0.624579131603241, 0.691919207572937, 0.6683501601219177, 0.7289562225341797, 0.6936026811599731, 0.691919207572937, 0.7154881954193115, 0.7112794518470764, 0.7020202279090881, 0.7592592835426331, 0.7542087435722351, 0.7424242496490479, 0.746632993221283, 0.7491582632064819, 0.7946127653121948, 0.7693602442741394, 0.8063973188400269, 0.8324915766716003, 0.8417508602142334, 0.9696969985961914], [0.08417508751153946, 0.12121212482452393, 0.12037037312984467, 0.12962962687015533, 0.14814814925193787, 0.15488214790821075, 0.15993265807628632, 0.16498316824436188, 0.25084173679351807, 0.2112794667482376, 0.2070707082748413, 0.22138047218322754, 0.2609427571296692, 0.3265993297100067, 0.3097642958164215, 0.2289562225341797, 0.2710437774658203, 0.25757575035095215, 0.3569023609161377, 0.31144779920578003, 0.3737373650074005, 0.2760942876338959, 0.32828283309936523, 0.3080808222293854, 0.3636363744735718, 0.3350168466567993, 0.3636363744735718, 0.33838382363319397, 0.4377104341983795, 0.37037035822868347, 0.402356892824173, 0.4402357041835785, 0.26599326729774475, 0.3973063826560974, 0.436026930809021, 0.4486531913280487, 0.45622894167900085, 0.44949495792388916, 0.46296295523643494, 0.35774409770965576, 0.44107744097709656, 0.48821547627449036, NaN, 0.5084174871444702, 0.5841751098632812, 0.5521885752677917, 0.5808081030845642, 0.5387205481529236, 0.5993266105651855, 0.5656565427780151, 0.68855220079422, 0.6481481194496155, 0.6868686676025391, 0.6414141654968262, 0.6228955984115601, 0.6717171669006348, 0.7457912564277649, 0.7272727489471436, 0.7575757503509521, 0.7575757503509521, 0.7592592835426331, 0.7920875549316406, 0.7760942578315735, 0.7861952781677246, 0.7087541818618774, 0.7744107842445374, 0.8383838534355164, 0.8148148059844971, 0.7996633052825928, 0.8653198480606079, 0.8585858345031738, 0.9259259104728699, 0.9814814925193787], [0.06902357190847397, 0.11279461532831192, 0.09090909361839294, 0.11447811126708984, 0.11447811126708984, 0.1430976390838623, 0.12962962687015533, 0.1515151560306549, 0.2222222238779068, 0.1304713785648346, 0.19696970283985138, 0.18265993893146515, 0.2542087435722351, 0.2845117747783661, 0.2306397259235382, 0.29124578833580017, 0.23232322931289673, 0.28956228494644165, 0.33838382363319397, 0.31144779920578003, 0.33670035004615784, 0.31060606241226196, 0.2929292917251587, 0.3232323229312897, 0.31986531615257263, 0.35185185074806213, 0.3737373650074005, 0.36026936769485474, 0.4486531913280487, 0.3872053921222687, 0.4191919267177582, 0.3653198778629303, 0.3905723989009857, 0.3484848439693451, 0.40067338943481445, 0.3888888955116272, 0.4335016906261444, 0.48653197288513184, 0.4612794518470764, 0.47138047218322754, 0.4377104341983795, 0.4781144857406616, 0.4915824830532074, NaN, 0.558922529220581, 0.5690235495567322, 0.567340075969696, 0.5252525210380554, 0.6060606241226196, 0.5909090638160706, 0.6700336933135986, 0.5993266105651855, 0.683501660823822, 0.6397306323051453, 0.6346801519393921, 0.6784511804580688, 0.7154881954193115, 0.7550504803657532, 0.7289562225341797, 0.7735690474510193, 0.7777777910232544, 0.7710437774658203, 0.7659932374954224, 0.8468013405799866, 0.7306397557258606, 0.819865345954895, 0.808080792427063, 0.8131313323974609, 0.7946127653121948, 0.8451178669929504, 0.8939393758773804, 0.8872053623199463, 0.9814814925193787], [0.04040404036641121, 0.07070706784725189, 0.07575757801532745, 0.09090909361839294, 0.08417508751153946, 0.1026936024427414, 0.10101009905338287, 0.10437710583209991, 0.20875421166419983, 0.2070707082748413, 0.16835017502307892, 0.20454545319080353, 0.2373737394809723, 0.2457912415266037, 0.24410773813724518, 0.2693602740764618, 0.2516835033893585, 0.28114476799964905, 0.30639731884002686, 0.2794612646102905, 0.3080808222293854, 0.313973069190979, 0.29629629850387573, 0.2920875549316406, 0.2845117747783661, 0.31144779920578003, 0.36868685483932495, 0.3468013405799866, 0.3973063826560974, 0.37037035822868347, 0.3872053921222687, 0.36026936769485474, 0.3611111044883728, 0.37542086839675903, 0.24074074625968933, 0.372053861618042, 0.4377104341983795, 0.4579124450683594, 0.4208754301071167, 0.4141414165496826, 0.4292929172515869, 0.4511784613132477, 0.41582491993904114, 0.44107744097709656, NaN, 0.5303030014038086, 0.5530303120613098, 0.49663299322128296, 0.5757575631141663, 0.5404040217399597, 0.6683501601219177, 0.5690235495567322, 0.6700336933135986, 0.6077440977096558, 0.5799663066864014, 0.6430976390838623, 0.6936026811599731, 0.7819865345954895, 0.683501660823822, 0.755892276763916, 0.7222222089767456, 0.7079124450683594, 0.7592592835426331, 0.7626262903213501, 0.7239057421684265, 0.7592592835426331, 0.7550504803657532, 0.752525269985199, 0.7542087435722351, 0.8215488195419312, 0.8417508602142334, 0.8451178669929504, 0.9713804721832275], [0.06902357190847397, 0.11447811126708984, 0.10858585685491562, 0.11279461532831192, 0.12121212482452393, 0.12457912415266037, 0.12289562076330185, 0.13299663364887238, 0.18518517911434174, 0.18939393758773804, 0.18855218589305878, 0.22558923065662384, 0.2415824979543686, 0.24074074625968933, 0.2306397259235382, 0.2845117747783661, 0.25925925374031067, 0.27441078424453735, 0.28114476799964905, 0.3164983093738556, 0.27272728085517883, 0.33838382363319397, 0.28535354137420654, 0.32575756311416626, 0.30639731884002686, 0.341750830411911, 0.33670035004615784, 0.37710437178611755, 0.3636363744735718, 0.3787878751754761, 0.36026936769485474, 0.3350168466567993, 0.40656566619873047, 0.3737373650074005, 0.3122895658016205, 0.3484848439693451, 0.29124578833580017, 0.45622894167900085, 0.4722222089767456, 0.4335016906261444, 0.4107744097709656, 0.4957912564277649, 0.44781145453453064, 0.43097642064094543, 0.469696968793869, NaN, 0.5336700081825256, 0.5202020406723022, 0.5538720488548279, 0.5454545617103577, 0.5892255902290344, 0.5269360542297363, 0.5875421166419983, 0.6069023609161377, 0.5757575631141663, 0.6212121248245239, 0.6296296119689941, 0.6691918969154358, 0.6936026811599731, 0.7121211886405945, 0.695286214351654, 0.7264309525489807, 0.7323232293128967, 0.8383838534355164, 0.7070707082748413, 0.7373737096786499, 0.755892276763916, 0.7676767706871033, 0.7373737096786499, 0.8148148059844971, 0.8249158263206482, 0.8552188277244568, 0.9730639457702637], [0.05387205258011818, 0.08080808073282242, 0.06902357190847397, 0.11026936024427414, 0.1026936024427414, 0.12457912415266037, 0.12794612348079681, 0.1304713785648346, 0.17424242198467255, 0.18686868250370026, 0.24915824830532074, 0.2542087435722351, 0.2070707082748413, 0.23232322931289673, 0.25925925374031067, 0.2668350040912628, 0.30471381545066833, 0.29124578833580017, 0.28787878155708313, 0.2609427571296692, 0.3265993297100067, 0.30892255902290344, 0.29377105832099915, 0.28282827138900757, 0.31986531615257263, 0.31986531615257263, 0.34343433380126953, 0.3350168466567993, 0.34511783719062805, 0.33417508006095886, 0.36026936769485474, 0.3872053921222687, 0.3358585834503174, 0.37037035822868347, 0.35353535413742065, 0.38383838534355164, 0.3973063826560974, 0.42424243688583374, 0.4427609443664551, 0.39898988604545593, 0.4292929172515869, 0.46464645862579346, 0.4191919267177582, 0.43265992403030396, 0.4469696879386902, 0.466329962015152, NaN, 0.48821547627449036, 0.6717171669006348, 0.4983164966106415, 0.558922529220581, 0.619528591632843, 0.5740740895271301, 0.6632996797561646, 0.5707070827484131, 0.7272727489471436, 0.6026936173439026, 0.619528591632843, 0.6456229090690613, 0.6784511804580688, 0.6700336933135986, 0.6675084233283997, 0.7045454382896423, 0.680134654045105, 0.7373737096786499, 0.691919207572937, 0.8299663066864014, 0.9057239294052124, 0.872053861618042, 0.7676767706871033, 0.7828282713890076, 0.8838383555412292, 0.9915825128555298], [0.06060606241226196, 0.07575757801532745, 0.11279461532831192, 0.13636364042758942, 0.11784511804580688, 0.14478114247322083, 0.13973063230514526, 0.15319865942001343, 0.22390572726726532, 0.22053872048854828, 0.17003366351127625, 0.18518517911434174, 0.22053872048854828, 0.2609427571296692, 0.2861952781677246, 0.19191919267177582, 0.24663299322128296, 0.2693602740764618, 0.31313130259513855, 0.2777777910232544, 0.28114476799964905, 0.2373737394809723, 0.28535354137420654, 0.28956228494644165, 0.3097642958164215, 0.25757575035095215, 0.37037035822868347, 0.34259259700775146, 0.37037035822868347, 0.38383838534355164, 0.38552188873291016, 0.44949495792388916, 0.3787878751754761, 0.39898988604545593, 0.4276094138622284, 0.4124579131603241, 0.43939393758773804, 0.37710437178611755, 0.22558923065662384, 0.42592594027519226, 0.436026930809021, 0.30471381545066833, 0.4612794518470764, 0.4747474789619446, 0.503367006778717, 0.47979798913002014, 0.5117844939231873, NaN, 0.5319865345954895, 0.5597643256187439, 0.6565656661987305, 0.5707070827484131, 0.6649831533432007, 0.5925925970077515, 0.6102693676948547, 0.6111111044883728, 0.683501660823822, 0.6346801519393921, 0.6784511804580688, 0.6414141654968262, 0.6717171669006348, 0.691919207572937, 0.7272727489471436, 0.6936026811599731, 0.695286214351654, 0.6969696879386902, 0.6944444179534912, 0.6717171669006348, 0.7272727489471436, 0.8013467788696289, 0.7760942578315735, 0.8518518805503845, 0.9747474789619446], [0.062289562076330185, 0.09259258955717087, 0.07070706784725189, 0.10606060922145844, 0.11026936024427414, 0.12289562076330185, 0.11447811126708984, 0.12121212482452393, 0.14646464586257935, 0.1776094287633896, 0.22053872048854828, 0.2525252401828766, 0.20370370149612427, 0.20202019810676575, 0.24074074625968933, 0.2626262605190277, 0.28787878155708313, 0.27441078424453735, 0.2609427571296692, 0.24326598644256592, 0.30471381545066833, 0.27693602442741394, 0.2609427571296692, 0.2710437774658203, 0.29629629850387573, 0.2996633052825928, 0.2996633052825928, 0.29124578833580017, 0.3181818127632141, 0.29629629850387573, 0.31481480598449707, 0.36195287108421326, 0.32996633648872375, 0.33670035004615784, 0.3400673270225525, 0.3585858643054962, 0.3872053921222687, 0.37710437178611755, 0.402356892824173, 0.36026936769485474, 0.37037035822868347, 0.41750842332839966, 0.40067338943481445, 0.39393940567970276, 0.42424243688583374, 0.4461279511451721, 0.32828283309936523, 0.4680134654045105, NaN, 0.43939393758773804, 0.5597643256187439, 0.6077440977096558, 0.5707070827484131, 0.5555555820465088, 0.5202020406723022, 0.6338383555412292, 0.5723905563354492, 0.5892255902290344, 0.6262626051902771, 0.6750841736793518, 0.6430976390838623, 0.6397306323051453, 0.7003366947174072, 0.6565656661987305, 0.6936026811599731, 0.6717171669006348, 0.7895622849464417, 0.8282828330993652, 0.8257575631141663, 0.7575757503509521, 0.7626262903213501, 0.8888888955116272, 0.9898989796638489], [0.03872053697705269, 0.05050504952669144, 0.048821549862623215, 0.08249158412218094, 0.05218855291604996, 0.08585858345031738, 0.10606060922145844, 0.09259258955717087, 0.1734006702899933, 0.1111111119389534, 0.11616161465644836, 0.1515151560306549, 0.17845118045806885, 0.20202019810676575, 0.1902356892824173, 0.1346801370382309, 0.19528619945049286, 0.2525252401828766, 0.2760942876338959, 0.1515151560306549, 0.30471381545066833, 0.1708754152059555, 0.19696970283985138, 0.1582491546869278, 0.2777777910232544, 0.2306397259235382, 0.29797980189323425, 0.11784511804580688, 0.3670033812522888, 0.1750841736793518, 0.3905723989009857, 0.40572389960289, 0.3072390556335449, 0.29124578833580017, 0.3249158263206482, 0.3636363744735718, 0.3888888955116272, 0.4208754301071167, 0.3552188575267792, 0.3585858643054962, 0.38383838534355164, 0.37289562821388245, 0.4343434274196625, 0.40909090638160706, 0.4595959484577179, 0.4545454680919647, 0.5016834735870361, 0.4402357041835785, 0.560606062412262, NaN, 0.6144781112670898, 0.5875421166419983, 0.619528591632843, 0.5538720488548279, 0.5530303120613098, 0.6161616444587708, 0.6750841736793518, 0.6430976390838623, 0.6363636255264282, 0.6616161465644836, 0.6178451180458069, 0.6565656661987305, 0.7154881954193115, 0.6818181872367859, 0.7121211886405945, 0.6683501601219177, 0.7424242496490479, 0.7861952781677246, 0.7979797720909119, 0.7609427571296692, 0.7643097639083862, 0.8232323527336121, 0.9545454382896423], [0.14814814925193787, 0.18855218589305878, 0.15909090638160706, 0.17003366351127625, 0.1902356892824173, 0.18855218589305878, 0.1666666716337204, 0.19696970283985138, 0.2457912415266037, 0.23232322931289673, 0.23400673270225525, 0.23232322931289673, 0.2718855142593384, 0.2777777910232544, 0.24915824830532074, 0.24831649661064148, 0.24915824830532074, 0.21885521709918976, 0.3156565725803375, 0.25589224696159363, 0.24747474491596222, 0.25925925374031067, 0.3030303120613098, 0.25757575035095215, 0.2861952781677246, 0.26178452372550964, 0.3038720488548279, 0.2794612646102905, 0.36195287108421326, 0.2929292917251587, 0.29629629850387573, 0.3872053921222687, 0.2794612646102905, 0.3585858643054962, 0.30639731884002686, 0.2760942876338959, 0.3552188575267792, 0.3350168466567993, 0.30471381545066833, 0.29797980189323425, 0.3223905861377716, 0.3181818127632141, 0.31144779920578003, 0.32996633648872375, 0.3316498398780823, 0.4107744097709656, 0.44107744097709656, 0.34343433380126953, 0.4402357041835785, 0.38552188873291016, NaN, 0.4764309823513031, 0.5917508602142334, 0.46464645862579346, 0.40909090638160706, 0.4983164966106415, 0.38383838534355164, 0.44107744097709656, 0.6010100841522217, 0.5757575631141663, 0.6111111044883728, 0.631313145160675, 0.9478114247322083, 0.6329966187477112, 0.5858585834503174, 0.6262626051902771, 0.7003366947174072, 0.68855220079422, 0.6868686676025391, 0.7491582632064819, 0.6649831533432007, 0.8535353541374207, 0.9461279511451721], [0.05892255902290344, 0.07154881954193115, 0.07407407462596893, 0.07744107395410538, 0.11784511804580688, 0.09427609294652939, 0.06734006851911545, 0.09932659566402435, 0.13299663364887238, 0.21717171370983124, 0.24663299322128296, 0.24074074625968933, 0.20791245996952057, 0.17003366351127625, 0.2752525210380554, 0.25925925374031067, 0.26430976390838623, 0.2609427571296692, 0.1734006702899933, 0.30471381545066833, 0.13299663364887238, 0.2693602740764618, 0.30892255902290344, 0.3080808222293854, 0.28956228494644165, 0.32154881954193115, 0.26767677068710327, 0.31060606241226196, 0.25757575035095215, 0.32828283309936523, 0.2946127951145172, 0.3585858643054962, 0.31144779920578003, 0.3804713785648346, 0.36868685483932495, 0.33670035004615784, 0.4124579131603241, 0.3080808222293854, 0.3636363744735718, 0.347643107175827, 0.39898988604545593, 0.3737373650074005, 0.35185185074806213, 0.40067338943481445, 0.43097642064094543, 0.47306397557258606, 0.3804713785648346, 0.4292929172515869, 0.39225590229034424, 0.4124579131603241, 0.5235690474510193, NaN, 0.5252525210380554, 0.4343434274196625, 0.47979798913002014, 0.469696968793869, 0.5185185074806213, 0.5303030014038086, 0.6127946376800537, 0.6565656661987305, 0.6329966187477112, 0.6388888955116272, 0.683501660823822, 0.680134654045105, 0.6464646458625793, 0.6683501601219177, 0.6380471587181091, 0.6599326729774475, 0.6548821330070496, 0.7457912564277649, 0.7811447978019714, 0.7676767706871033, 0.9713804721832275], [0.14814814925193787, 0.18686868250370026, 0.15488214790821075, 0.17171716690063477, 0.1818181872367859, 0.1835016906261444, 0.16582491993904114, 0.18855218589305878, 0.2390572428703308, 0.22558923065662384, 0.22727273404598236, 0.22558923065662384, 0.2760942876338959, 0.2760942876338959, 0.24410773813724518, 0.2457912415266037, 0.23989899456501007, 0.2138047069311142, 0.31481480598449707, 0.2609427571296692, 0.24242424964904785, 0.2542087435722351, 0.30218854546546936, 0.25589224696159363, 0.2794612646102905, 0.25925925374031067, 0.30471381545066833, 0.2794612646102905, 0.36195287108421326, 0.2904040515422821, 0.29124578833580017, 0.3905723989009857, 0.2794612646102905, 0.3569023609161377, 0.2988215386867523, 0.2693602740764618, 0.3501683473587036, 0.32828283309936523, 0.30471381545066833, 0.2920875549316406, 0.32828283309936523, 0.3181818127632141, 0.31313130259513855, 0.3164983093738556, 0.32996633648872375, 0.4124579131603241, 0.42592594027519226, 0.3350168466567993, 0.4292929172515869, 0.3804713785648346, 0.408249169588089, 0.4747474789619446, NaN, 0.4579124450683594, 0.4141414165496826, 0.4932659864425659, 0.3872053921222687, 0.4377104341983795, 0.5909090638160706, 0.567340075969696, 0.6060606241226196, 0.627946138381958, 0.9612794518470764, 0.627946138381958, 0.5824915766716003, 0.619528591632843, 0.68855220079422, 0.6818181872367859, 0.683501660823822, 0.7424242496490479, 0.6632996797561646, 0.8451178669929504, 0.9461279511451721], [0.04040404036641121, 0.06565656512975693, 0.07575757801532745, 0.0791245773434639, 0.09090909361839294, 0.09595959633588791, 0.1346801370382309, 0.11279461532831192, 0.15488214790821075, 0.11447811126708984, 0.16498316824436188, 0.20370370149612427, 0.16835017502307892, 0.17171716690063477, 0.18686868250370026, 0.18686868250370026, 0.22558923065662384, 0.2390572428703308, 0.24747474491596222, 0.18518517911434174, 0.30639731884002686, 0.24242424964904785, 0.2138047069311142, 0.19360269606113434, 0.25589224696159363, 0.23400673270225525, 0.25925925374031067, 0.202861949801445, 0.32828283309936523, 0.22558923065662384, 0.32828283309936523, 0.347643107175827, 0.28956228494644165, 0.3164983093738556, 0.3030303120613098, 0.29629629850387573, 0.3122895658016205, 0.3484848439693451, 0.35185185074806213, 0.3350168466567993, 0.3350168466567993, 0.37542086839675903, 0.3585858643054962, 0.36026936769485474, 0.39225590229034424, 0.3930976390838623, 0.33670035004615784, 0.40740740299224854, 0.4444444477558136, 0.4461279511451721, 0.5353535413742065, 0.5656565427780151, 0.5420875549316406, NaN, 0.49663299322128296, 0.6776094436645508, 0.5084174871444702, 0.5707070827484131, 0.6094276309013367, 0.6161616444587708, 0.6212121248245239, 0.6228955984115601, 0.6708754301071167, 0.6262626051902771, 0.7053872346878052, 0.6447811722755432, 0.755892276763916, 0.804713785648346, 0.8619528412818909, 0.7331649661064148, 0.7087541818618774, 0.8619528412818909, 0.9730639457702637], [0.03872053697705269, 0.04545454680919647, 0.05218855291604996, 0.062289562076330185, 0.0555555559694767, 0.06565656512975693, 0.06565656512975693, 0.07070706784725189, 0.13973063230514526, 0.09427609294652939, 0.09090909361839294, 0.09848485141992569, 0.16498316824436188, 0.15993265807628632, 0.16498316824436188, 0.05218855291604996, 0.1750841736793518, 0.21632996201515198, 0.2222222238779068, 0.12962962687015533, 0.2180134654045105, 0.0833333358168602, 0.14646464586257935, 0.15656565129756927, 0.2373737394809723, 0.17676767706871033, 0.28114476799964905, 0.11279461532831192, 0.3080808222293854, 0.14814814925193787, 0.3265993297100067, 0.33838382363319397, 0.29124578833580017, 0.2794612646102905, 0.3080808222293854, 0.3501683473587036, 0.34932661056518555, 0.32996633648872375, 0.26851850748062134, 0.3400673270225525, 0.32996633648872375, 0.3080808222293854, 0.37710437178611755, 0.3653198778629303, 0.42003366351127625, 0.42424243688583374, 0.4292929172515869, 0.38973063230514526, 0.47979798913002014, 0.4469696879386902, 0.5909090638160706, 0.5202020406723022, 0.5858585834503174, 0.503367006778717, NaN, 0.5446127653121948, 0.5993266105651855, 0.6077440977096558, 0.5892255902290344, 0.624579131603241, 0.6127946376800537, 0.6094276309013367, 0.695286214351654, 0.6548821330070496, 0.6767676472663879, 0.6734007000923157, 0.7138047218322754, 0.7239057421684265, 0.7314814925193787, 0.7457912564277649, 0.7727272510528564, 0.8299663066864014, 0.9688552021980286], [0.04208754375576973, 0.06397306174039841, 0.05387205258011818, 0.0791245773434639, 0.06734006851911545, 0.0791245773434639, 0.08585858345031738, 0.0875420868396759, 0.13973063230514526, 0.08922559022903442, 0.13215488195419312, 0.15993265807628632, 0.15488214790821075, 0.16161616146564484, 0.17003366351127625, 0.1346801370382309, 0.20875421166419983, 0.22558923065662384, 0.21043771505355835, 0.1388888955116272, 0.2525252401828766, 0.1750841736793518, 0.17676767706871033, 0.15993265807628632, 0.23569023609161377, 0.1986531913280487, 0.21885521709918976, 0.1776094287633896, 0.2777777910232544, 0.18686868250370026, 0.27272728085517883, 0.33249157667160034, 0.26767677068710327, 0.31144779920578003, 0.2777777910232544, 0.2845117747783661, 0.3013468086719513, 0.28282827138900757, 0.30639731884002686, 0.29629629850387573, 0.28956228494644165, 0.3316498398780823, 0.32828283309936523, 0.32154881954193115, 0.3569023609161377, 0.3787878751754761, 0.27272728085517883, 0.3888888955116272, 0.36616161465644836, 0.38383838534355164, 0.5016834735870361, 0.5303030014038086, 0.5067340135574341, 0.3223905861377716, 0.4553872048854828, NaN, 0.48148149251937866, 0.5117844939231873, 0.5808081030845642, 0.6026936173439026, 0.5976430773735046, 0.5976430773735046, 0.6481481194496155, 0.6178451180458069, 0.6582491397857666, 0.631313145160675, 0.7230639457702637, 0.7373737096786499, 0.7609427571296692, 0.7188552021980286, 0.691919207572937, 0.8518518805503845, 0.9747474789619446], [0.05892255902290344, 0.05723905563354492, 0.05387205258011818, 0.08249158412218094, 0.07407407462596893, 0.0875420868396759, 0.08249158412218094, 0.09595959633588791, 0.16329966485500336, 0.12121212482452393, 0.1195286214351654, 0.1262626200914383, 0.17676767706871033, 0.17676767706871033, 0.17003366351127625, 0.1111111119389534, 0.1515151560306549, 0.15488214790821075, 0.22727273404598236, 0.15740740299224854, 0.2138047069311142, 0.13215488195419312, 0.15656565129756927, 0.17003366351127625, 0.19360269606113434, 0.12457912415266037, 0.2542087435722351, 0.20033669471740723, 0.25925925374031067, 0.20875421166419983, 0.2525252401828766, 0.3400673270225525, 0.2053872048854828, 0.2794612646102905, 0.22474747896194458, 0.2525252401828766, 0.3030303120613098, 0.2693602740764618, 0.23569023609161377, 0.25589224696159363, 0.25336700677871704, 0.2710437774658203, 0.2542087435722351, 0.2845117747783661, 0.30639731884002686, 0.37037035822868347, 0.3973063826560974, 0.3164983093738556, 0.4276094138622284, 0.3249158263206482, 0.6161616444587708, 0.48148149251937866, 0.6127946376800537, 0.4915824830532074, 0.40067338943481445, 0.5185185074806213, NaN, 0.5370370149612427, 0.5909090638160706, 0.5875421166419983, 0.558922529220581, 0.6161616444587708, 0.68855220079422, 0.6094276309013367, 0.6060606241226196, 0.5757575631141663, 0.7020202279090881, 0.631313145160675, 0.68855220079422, 0.7171717286109924, 0.6565656661987305, 0.808080792427063, 0.936026930809021], [0.033670034259557724, 0.04208754375576973, 0.047138046473264694, 0.05892255902290344, 0.05050504952669144, 0.05892255902290344, 0.03872053697705269, 0.05892255902290344, 0.12794612348079681, 0.08585858345031738, 0.10606060922145844, 0.09427609294652939, 0.17171716690063477, 0.14478114247322083, 0.12289562076330185, 0.16835017502307892, 0.10942760854959488, 0.17003366351127625, 0.1750841736793518, 0.17003366351127625, 0.1818181872367859, 0.18686868250370026, 0.17003366351127625, 0.1902356892824173, 0.19528619945049286, 0.1835016906261444, 0.18686868250370026, 0.2070707082748413, 0.24915824830532074, 0.22727273404598236, 0.19360269606113434, 0.26430976390838623, 0.2390572428703308, 0.27272728085517883, 0.1262626200914383, 0.21885521709918976, 0.23989899456501007, 0.22053872048854828, 0.2996633052825928, 0.26430976390838623, 0.21717171370983124, 0.30639731884002686, 0.27272728085517883, 0.24494948983192444, 0.2180134654045105, 0.3308080732822418, 0.3804713785648346, 0.3653198778629303, 0.4107744097709656, 0.3569023609161377, 0.558922529220581, 0.469696968793869, 0.5622895359992981, 0.4292929172515869, 0.39225590229034424, 0.48821547627449036, 0.46296295523643494, NaN, 0.5909090638160706, 0.5984848737716675, 0.5723905563354492, 0.619528591632843, 0.6851851940155029, 0.6515151262283325, 0.6212121248245239, 0.6178451180458069, 0.683501660823822, 0.6632996797561646, 0.6414141654968262, 0.7340067625045776, 0.7154881954193115, 0.8181818127632141, 0.9562289714813232], [0.05723905563354492, 0.09932659566402435, 0.09259258955717087, 0.08080808073282242, 0.1026936024427414, 0.10942760854959488, 0.10774410516023636, 0.12457912415266037, 0.1750841736793518, 0.12962962687015533, 0.13383838534355164, 0.13299663364887238, 0.17171716690063477, 0.21885521709918976, 0.1835016906261444, 0.16329966485500336, 0.15319865942001343, 0.13973063230514526, 0.23821549117565155, 0.19360269606113434, 0.2390572428703308, 0.18686868250370026, 0.2525252401828766, 0.1986531913280487, 0.2373737394809723, 0.22053872048854828, 0.18013468384742737, 0.2053872048854828, 0.30471381545066833, 0.21717171370983124, 0.1835016906261444, 0.2390572428703308, 0.18518517911434174, 0.3181818127632141, 0.23148147761821747, 0.2306397259235382, 0.2289562225341797, 0.2626262605190277, 0.2929292917251587, 0.22053872048854828, 0.3274410665035248, 0.3080808222293854, 0.24242424964904785, 0.2710437774658203, 0.3164983093738556, 0.30639731884002686, 0.3543770909309387, 0.32154881954193115, 0.3737373650074005, 0.3636363744735718, 0.39898988604545593, 0.3872053921222687, 0.40909090638160706, 0.3905723989009857, 0.4107744097709656, 0.4191919267177582, 0.40909090638160706, 0.40909090638160706, NaN, 0.49663299322128296, 0.47306397557258606, 0.691919207572937, 0.5420875549316406, 0.48148149251937866, 0.5387205481529236, 0.49494948983192444, 0.5656565427780151, 0.5538720488548279, 0.5420875549316406, 0.808080792427063, 0.6010100841522217, 0.7020202279090881, 0.9040403962135315], [0.062289562076330185, 0.0875420868396759, 0.07070706784725189, 0.08922559022903442, 0.08922559022903442, 0.09259258955717087, 0.07323232293128967, 0.1026936024427414, 0.09427609294652939, 0.1734006702899933, 0.16161616146564484, 0.17171716690063477, 0.21212121844291687, 0.1498316526412964, 0.19107744097709656, 0.21548821032047272, 0.1835016906261444, 0.17592592537403107, 0.16329966485500336, 0.26430976390838623, 0.09427609294652939, 0.22727273404598236, 0.19528619945049286, 0.2710437774658203, 0.20202019810676575, 0.2525252401828766, 0.18686868250370026, 0.26178452372550964, 0.15656565129756927, 0.2710437774658203, 0.1750841736793518, 0.24915824830532074, 0.22558923065662384, 0.29629629850387573, 0.2222222238779068, 0.26599326729774475, 0.21885521709918976, 0.2289562225341797, 0.25589224696159363, 0.24663299322128296, 0.283670037984848, 0.2845117747783661, 0.24242424964904785, 0.2264309823513031, 0.24410773813724518, 0.28787878155708313, 0.32154881954193115, 0.3585858643054962, 0.3249158263206482, 0.33838382363319397, 0.42424243688583374, 0.34343433380126953, 0.43265992403030396, 0.38383838534355164, 0.37542086839675903, 0.3973063826560974, 0.4124579131603241, 0.4015151560306549, 0.503367006778717, NaN, 0.5067340135574341, 0.5345118045806885, 0.5808081030845642, 0.563973069190979, 0.5336700081825256, 0.5656565427780151, 0.5656565427780151, 0.5336700081825256, 0.5707070827484131, 0.6565656661987305, 0.6363636255264282, 0.7239057421684265, 0.9208754301071167], [0.01515151560306549, 0.048821549862623215, 0.05387205258011818, 0.048821549862623215, 0.05892255902290344, 0.05387205258011818, 0.05387205258011818, 0.06818182021379471, 0.1026936024427414, 0.06902357190847397, 0.07575757801532745, 0.09427609294652939, 0.13636364042758942, 0.11447811126708984, 0.11616161465644836, 0.10774410516023636, 0.09595959633588791, 0.1430976390838623, 0.13636364042758942, 0.1582491546869278, 0.11616161465644836, 0.12542086839675903, 0.14814814925193787, 0.16835017502307892, 0.17003366351127625, 0.17845118045806885, 0.18518517911434174, 0.20202019810676575, 0.2070707082748413, 0.22727273404598236, 0.20622895658016205, 0.22558923065662384, 0.20033669471740723, 0.24915824830532074, 0.19360269606113434, 0.2095959633588791, 0.2390572428703308, 0.21548821032047272, 0.25084173679351807, 0.2222222238779068, 0.2609427571296692, 0.2887205481529236, 0.24074074625968933, 0.2222222238779068, 0.2777777910232544, 0.30471381545066833, 0.32996633648872375, 0.32828283309936523, 0.3569023609161377, 0.3821548819541931, 0.3888888955116272, 0.3670033812522888, 0.39393940567970276, 0.3787878751754761, 0.3872053921222687, 0.402356892824173, 0.44107744097709656, 0.4276094138622284, 0.5269360542297363, 0.4932659864425659, NaN, 0.5437710285186768, 0.4781144857406616, 0.48821547627449036, 0.5505050420761108, 0.7129629850387573, 0.5446127653121948, 0.5909090638160706, 0.5791245698928833, 0.6127946376800537, 0.7979797720909119, 0.68855220079422, 0.9427609443664551], [0.05050504952669144, 0.07744107395410538, 0.08417508751153946, 0.07239057123661041, 0.08249158412218094, 0.09427609294652939, 0.10101009905338287, 0.10437710583209991, 0.16329966485500336, 0.12794612348079681, 0.11026936024427414, 0.11700336635112762, 0.1515151560306549, 0.20033669471740723, 0.17424242198467255, 0.1430976390838623, 0.14141413569450378, 0.12121212482452393, 0.21717171370983124, 0.17676767706871033, 0.21212121844291687, 0.16077440977096558, 0.21717171370983124, 0.18434342741966248, 0.2070707082748413, 0.1986531913280487, 0.16329966485500336, 0.19360269606113434, 0.2760942876338959, 0.19528619945049286, 0.15993265807628632, 0.2306397259235382, 0.1666666716337204, 0.3030303120613098, 0.19696970283985138, 0.21464645862579346, 0.20033669471740723, 0.2348484843969345, 0.27272728085517883, 0.1986531913280487, 0.2861952781677246, 0.29797980189323425, 0.20791245996952057, 0.2289562225341797, 0.2920875549316406, 0.2735690176486969, 0.33249157667160034, 0.3080808222293854, 0.36026936769485474, 0.34343433380126953, 0.36868685483932495, 0.3611111044883728, 0.372053861618042, 0.37710437178611755, 0.3905723989009857, 0.402356892824173, 0.38383838534355164, 0.3804713785648346, 0.3080808222293854, 0.4654882252216339, 0.45622894167900085, NaN, 0.5286195278167725, 0.4545454680919647, 0.5269360542297363, 0.47979798913002014, 0.5277777910232544, 0.5353535413742065, 0.5117844939231873, 0.7239057421684265, 0.5925925970077515, 0.6851851940155029, 0.8973063826560974], [0.10942760854959488, 0.14814814925193787, 0.12457912415266037, 0.12289562076330185, 0.15656565129756927, 0.15656565129756927, 0.13804714381694794, 0.15656565129756927, 0.17003366351127625, 0.18686868250370026, 0.18518517911434174, 0.1835016906261444, 0.2180134654045105, 0.19360269606113434, 0.1986531913280487, 0.17845118045806885, 0.20033669471740723, 0.15909090638160706, 0.21212121844291687, 0.21043771505355835, 0.1734006702899933, 0.19191919267177582, 0.2138047069311142, 0.2053872048854828, 0.20875421166419983, 0.20202019810676575, 0.23232322931289673, 0.22558923065662384, 0.26599326729774475, 0.2289562225341797, 0.2138047069311142, 0.23569023609161377, 0.20370370149612427, 0.2929292917251587, 0.21212121844291687, 0.19528619945049286, 0.2373737394809723, 0.22558923065662384, 0.2289562225341797, 0.22053872048854828, 0.25084173679351807, 0.24074074625968933, 0.22390572726726532, 0.23400673270225525, 0.24074074625968933, 0.26767677068710327, 0.2954545319080353, 0.27272728085517883, 0.2996633052825928, 0.2845117747783661, 0.05218855291604996, 0.3164983093738556, 0.03872053697705269, 0.3291245698928833, 0.30471381545066833, 0.35185185074806213, 0.31144779920578003, 0.31481480598449707, 0.4579124450683594, 0.4191919267177582, 0.5218855142593384, 0.47138047218322754, NaN, 0.40909090638160706, 0.48148149251937866, 0.5404040217399597, 0.5404040217399597, 0.5269360542297363, 0.48148149251937866, 0.5976430773735046, 0.5892255902290344, 0.7053872346878052, 0.932659924030304], [0.031986530870199203, 0.047138046473264694, 0.04545454680919647, 0.05387205258011818, 0.06565656512975693, 0.06565656512975693, 0.06902357190847397, 0.07407407462596893, 0.08417508751153946, 0.06565656512975693, 0.07239057123661041, 0.09932659566402435, 0.1262626200914383, 0.10942760854959488, 0.0917508453130722, 0.1582491546869278, 0.13636364042758942, 0.15488214790821075, 0.12962962687015533, 0.1666666716337204, 0.14478114247322083, 0.1734006702899933, 0.13804714381694794, 0.18013468384742737, 0.16582491993904114, 0.16498316824436188, 0.1195286214351654, 0.20875421166419983, 0.17003366351127625, 0.21885521709918976, 0.13804714381694794, 0.2289562225341797, 0.1835016906261444, 0.2390572428703308, 0.15572389960289001, 0.2053872048854828, 0.048821549862623215, 0.21043771505355835, 0.2289562225341797, 0.21885521709918976, 0.19696970283985138, 0.2457912415266037, 0.2138047069311142, 0.15319865942001343, 0.2373737394809723, 0.16161616146564484, 0.31986531615257263, 0.30639731884002686, 0.34343433380126953, 0.3181818127632141, 0.3670033812522888, 0.31986531615257263, 0.372053861618042, 0.3737373650074005, 0.34511783719062805, 0.3821548819541931, 0.3905723989009857, 0.3484848439693451, 0.5185185074806213, 0.436026930809021, 0.5117844939231873, 0.5454545617103577, 0.5909090638160706, NaN, 0.5420875549316406, 0.5404040217399597, 0.5791245698928833, 0.5202020406723022, 0.5361952781677246, 0.6548821330070496, 0.6565656661987305, 0.7542087435722351, 0.9562289714813232], [0.03703703731298447, 0.04545454680919647, 0.03872053697705269, 0.062289562076330185, 0.062289562076330185, 0.062289562076330185, 0.05892255902290344, 0.06397306174039841, 0.10774410516023636, 0.07239057123661041, 0.10437710583209991, 0.13299663364887238, 0.08922559022903442, 0.09848485141992569, 0.12121212482452393, 0.1734006702899933, 0.17171716690063477, 0.2070707082748413, 0.13299663364887238, 0.1515151560306549, 0.18013468384742737, 0.186026930809021, 0.12121212482452393, 0.16329966485500336, 0.12962962687015533, 0.1986531913280487, 0.05892255902290344, 0.17676767706871033, 0.20370370149612427, 0.18855218589305878, 0.09595959633588791, 0.22727273404598236, 0.23569023609161377, 0.16414141654968262, 0.21548821032047272, 0.25757575035095215, 0.21548821032047272, 0.24242424964904785, 0.23400673270225525, 0.25925925374031067, 0.21043771505355835, 0.25757575035095215, 0.29124578833580017, 0.2693602740764618, 0.2760942876338959, 0.2929292917251587, 0.2626262605190277, 0.30471381545066833, 0.30639731884002686, 0.28787878155708313, 0.4141414165496826, 0.35353535413742065, 0.41750842332839966, 0.2946127951145172, 0.3232323229312897, 0.341750830411911, 0.39393940567970276, 0.3787878751754761, 0.4612794518470764, 0.466329962015152, 0.44949495792388916, 0.47306397557258606, 0.5185185074806213, 0.4579124450683594, NaN, 0.4739057123661041, 0.47306397557258606, 0.46296295523643494, 0.48148149251937866, 0.5572390556335449, 0.5791245698928833, 0.627946138381958, 0.8989899158477783], [0.018518518656492233, 0.04292929172515869, 0.04208754375576973, 0.04377104341983795, 0.062289562076330185, 0.05218855291604996, 0.047138046473264694, 0.062289562076330185, 0.08585858345031738, 0.062289562076330185, 0.07239057123661041, 0.07407407462596893, 0.12121212482452393, 0.10353535413742065, 0.11616161465644836, 0.0875420868396759, 0.0875420868396759, 0.12457912415266037, 0.1262626200914383, 0.13973063230514526, 0.11279461532831192, 0.10606060922145844, 0.12794612348079681, 0.14646464586257935, 0.1498316526412964, 0.1430976390838623, 0.15067340433597565, 0.17592592537403107, 0.18013468384742737, 0.1835016906261444, 0.16919191181659698, 0.2180134654045105, 0.18265993893146515, 0.24747474491596222, 0.1582491546869278, 0.20370370149612427, 0.21043771505355835, 0.18518517911434174, 0.21212121844291687, 0.2053872048854828, 0.22053872048854828, 0.25336700677871704, 0.22558923065662384, 0.18013468384742737, 0.24074074625968933, 0.2626262605190277, 0.3080808222293854, 0.3030303120613098, 0.32828283309936523, 0.3316498398780823, 0.3737373650074005, 0.3316498398780823, 0.3804713785648346, 0.3552188575267792, 0.3265993297100067, 0.36868685483932495, 0.42424243688583374, 0.3821548819541931, 0.5050504803657532, 0.4343434274196625, 0.28703704476356506, 0.5202020406723022, 0.4595959484577179, 0.4595959484577179, 0.5260942578315735, NaN, 0.5101010203361511, 0.5471380352973938, 0.5521885752677917, 0.5892255902290344, 0.7171717286109924, 0.6582491397857666, 0.9208754301071167], [0.03703703731298447, 0.05892255902290344, 0.05892255902290344, 0.0555555559694767, 0.06565656512975693, 0.08417508751153946, 0.09764309972524643, 0.0875420868396759, 0.09090909361839294, 0.09090909361839294, 0.10774410516023636, 0.11616161465644836, 0.15488214790821075, 0.1069023534655571, 0.13299663364887238, 0.1111111119389534, 0.14646464586257935, 0.11616161465644836, 0.11616161465644836, 0.11784511804580688, 0.14814814925193787, 0.13636364042758942, 0.15319865942001343, 0.1262626200914383, 0.15993265807628632, 0.1346801370382309, 0.11700336635112762, 0.15656565129756927, 0.16498316824436188, 0.15572389960289001, 0.12794612348079681, 0.25757575035095215, 0.1582491546869278, 0.22558923065662384, 0.19528619945049286, 0.18855218589305878, 0.19528619945049286, 0.15488214790821075, 0.2306397259235382, 0.18518517911434174, 0.21885521709918976, 0.25084173679351807, 0.16161616146564484, 0.19191919267177582, 0.24494948983192444, 0.24410773813724518, 0.17003366351127625, 0.3055555522441864, 0.21043771505355835, 0.25757575035095215, 0.2996633052825928, 0.36195287108421326, 0.31144779920578003, 0.24410773813724518, 0.2861952781677246, 0.27693602442741394, 0.29797980189323425, 0.3164983093738556, 0.4343434274196625, 0.4343434274196625, 0.4553872048854828, 0.4722222089767456, 0.4595959484577179, 0.4208754301071167, 0.5269360542297363, 0.4898989796638489, NaN, 0.47138047218322754, 0.4898989796638489, 0.5959596037864685, 0.5875421166419983, 0.869528591632843, 0.9747474789619446], [0.011784511618316174, 0.03872053697705269, 0.021885521709918976, 0.04208754375576973, 0.04040404036641121, 0.04377104341983795, 0.04208754375576973, 0.05050504952669144, 0.0513468012213707, 0.04545454680919647, 0.06565656512975693, 0.0791245773434639, 0.08922559022903442, 0.05387205258011818, 0.11616161465644836, 0.07070706784725189, 0.12289562076330185, 0.14646464586257935, 0.09090909361839294, 0.09427609294652939, 0.1430976390838623, 0.08249158412218094, 0.09090909361839294, 0.10606060922145844, 0.1195286214351654, 0.12121212482452393, 0.06060606241226196, 0.11447811126708984, 0.14141413569450378, 0.11279461532831192, 0.11447811126708984, 0.22053872048854828, 0.1582491546869278, 0.17676767706871033, 0.19696970283985138, 0.17003366351127625, 0.1734006702899933, 0.16414141654968262, 0.19528619945049286, 0.17171716690063477, 0.18855218589305878, 0.2053872048854828, 0.18518517911434174, 0.18686868250370026, 0.24747474491596222, 0.23232322931289673, 0.09427609294652939, 0.32828283309936523, 0.17171716690063477, 0.2138047069311142, 0.31144779920578003, 0.3400673270225525, 0.3181818127632141, 0.19528619945049286, 0.2760942876338959, 0.2626262605190277, 0.36868685483932495, 0.33670035004615784, 0.4461279511451721, 0.466329962015152, 0.40909090638160706, 0.46464645862579346, 0.47306397557258606, 0.47979798913002014, 0.5370370149612427, 0.4528619647026062, 0.5286195278167725, NaN, 0.5101010203361511, 0.5892255902290344, 0.5538720488548279, 0.695286214351654, 0.9680134654045105], [0.018518518656492233, 0.03703703731298447, 0.018518518656492233, 0.033670034259557724, 0.02861952781677246, 0.04377104341983795, 0.04208754375576973, 0.047138046473264694, 0.047138046473264694, 0.04545454680919647, 0.07239057123661041, 0.08585858345031738, 0.07575757801532745, 0.06060606241226196, 0.10606060922145844, 0.06060606241226196, 0.1111111119389534, 0.13804714381694794, 0.09259258955717087, 0.07407407462596893, 0.11279461532831192, 0.08417508751153946, 0.09090909361839294, 0.08249158412218094, 0.13299663364887238, 0.11279461532831192, 0.07239057123661041, 0.08249158412218094, 0.16835017502307892, 0.09595959633588791, 0.1346801370382309, 0.21548821032047272, 0.1582491546869278, 0.18518517911434174, 0.1986531913280487, 0.17171716690063477, 0.21043771505355835, 0.16329966485500336, 0.20875421166419983, 0.17845118045806885, 0.18686868250370026, 0.2306397259235382, 0.20033669471740723, 0.2053872048854828, 0.2457912415266037, 0.2626262605190277, 0.12794612348079681, 0.27272728085517883, 0.17424242198467255, 0.20202019810676575, 0.31313130259513855, 0.34511783719062805, 0.3164983093738556, 0.13804714381694794, 0.26851850748062134, 0.2390572428703308, 0.31144779920578003, 0.3585858643054962, 0.4579124450683594, 0.4292929172515869, 0.4208754301071167, 0.48821547627449036, 0.5185185074806213, 0.4638047218322754, 0.5185185074806213, 0.44781145453453064, 0.5101010203361511, 0.4898989796638489, NaN, 0.5892255902290344, 0.560606062412262, 0.7020202279090881, 0.9562289714813232], [0.033670034259557724, 0.05471380427479744, 0.05050504952669144, 0.046296294778585434, 0.05892255902290344, 0.06986531615257263, 0.05218855291604996, 0.06649831682443619, 0.07744107395410538, 0.10942760854959488, 0.08080808073282242, 0.08922559022903442, 0.14478114247322083, 0.13131313025951385, 0.13636364042758942, 0.0875420868396759, 0.1111111119389534, 0.06565656512975693, 0.14814814925193787, 0.1430976390838623, 0.11363636702299118, 0.10437710583209991, 0.16329966485500336, 0.1498316526412964, 0.15319865942001343, 0.12962962687015533, 0.1304713785648346, 0.1515151560306549, 0.17003366351127625, 0.1582491546869278, 0.11279461532831192, 0.18855218589305878, 0.1346801370382309, 0.25589224696159363, 0.13973063230514526, 0.15993265807628632, 0.14141413569450378, 0.13973063230514526, 0.17676767706871033, 0.15993265807628632, 0.2070707082748413, 0.19360269606113434, 0.1346801370382309, 0.15488214790821075, 0.17845118045806885, 0.18518517911434174, 0.23232322931289673, 0.1986531913280487, 0.24242424964904785, 0.2390572428703308, 0.25084173679351807, 0.2542087435722351, 0.25757575035095215, 0.2668350040912628, 0.2542087435722351, 0.28114476799964905, 0.28282827138900757, 0.26599326729774475, 0.19191919267177582, 0.34343433380126953, 0.3872053921222687, 0.2760942876338959, 0.402356892824173, 0.34511783719062805, 0.4427609443664551, 0.4107744097709656, 0.4040403962135315, 0.4107744097709656, 0.4107744097709656, NaN, 0.5176767706871033, 0.6077440977096558, 0.8552188277244568], [0.016835017129778862, 0.035353533923625946, 0.03030303120613098, 0.023569023236632347, 0.03703703731298447, 0.035353533923625946, 0.02861952781677246, 0.035353533923625946, 0.03872053697705269, 0.033670034259557724, 0.05218855291604996, 0.03703703731298447, 0.08585858345031738, 0.05892255902290344, 0.048821549862623215, 0.0555555559694767, 0.04545454680919647, 0.07407407462596893, 0.06060606241226196, 0.09259258955717087, 0.06902357190847397, 0.05723905563354492, 0.0875420868396759, 0.08585858345031738, 0.09932659566402435, 0.09932659566402435, 0.09090909361839294, 0.10101009905338287, 0.10774410516023636, 0.1026936024427414, 0.10101009905338287, 0.18686868250370026, 0.1026936024427414, 0.17845118045806885, 0.10101009905338287, 0.1195286214351654, 0.11447811126708984, 0.13299663364887238, 0.15993265807628632, 0.12121212482452393, 0.14646464586257935, 0.16750842332839966, 0.14141413569450378, 0.10606060922145844, 0.1582491546869278, 0.1750841736793518, 0.21717171370983124, 0.22390572726726532, 0.2373737394809723, 0.23569023609161377, 0.3350168466567993, 0.21885521709918976, 0.33670035004615784, 0.29124578833580017, 0.22727273404598236, 0.3080808222293854, 0.34343433380126953, 0.2845117747783661, 0.39898988604545593, 0.3636363744735718, 0.20202019810676575, 0.40740740299224854, 0.4107744097709656, 0.34343433380126953, 0.4208754301071167, 0.28282827138900757, 0.4124579131603241, 0.4461279511451721, 0.43939393758773804, 0.48232322931289673, NaN, 0.5690235495567322, 0.8569023609161377], [0.03703703731298447, 0.04545454680919647, 0.04040404036641121, 0.04545454680919647, 0.048821549862623215, 0.062289562076330185, 0.05387205258011818, 0.05723905563354492, 0.048821549862623215, 0.06734006851911545, 0.07239057123661041, 0.07575757801532745, 0.12121212482452393, 0.07407407462596893, 0.10101009905338287, 0.06986531615257263, 0.09427609294652939, 0.04377104341983795, 0.08585858345031738, 0.08922559022903442, 0.05723905563354492, 0.0791245773434639, 0.11363636702299118, 0.08249158412218094, 0.11279461532831192, 0.07744107395410538, 0.09090909361839294, 0.10353535413742065, 0.12457912415266037, 0.10606060922145844, 0.08585858345031738, 0.14814814925193787, 0.09006734192371368, 0.19528619945049286, 0.10774410516023636, 0.09090909361839294, 0.11784511804580688, 0.1026936024427414, 0.13636364042758942, 0.08922559022903442, 0.1540404111146927, 0.1582491546869278, 0.07407407462596893, 0.11279461532831192, 0.15488214790821075, 0.14478114247322083, 0.11616161465644836, 0.14814814925193787, 0.1111111119389534, 0.17676767706871033, 0.14646464586257935, 0.23232322931289673, 0.15488214790821075, 0.13804714381694794, 0.17003366351127625, 0.14814814925193787, 0.19191919267177582, 0.1818181872367859, 0.29797980189323425, 0.2760942876338959, 0.31144779920578003, 0.31481480598449707, 0.2946127951145172, 0.2457912415266037, 0.372053861618042, 0.341750830411911, 0.1304713785648346, 0.30471381545066833, 0.29797980189323425, 0.39225590229034424, 0.43097642064094543, NaN, 0.9444444179534912], [0.005050505045801401, 0.005050505045801401, 0.003367003286257386, 0.003367003286257386, 0.006734006572514772, 0.006734006572514772, 0.005050505045801401, 0.006734006572514772, 0.005050505045801401, 0.005050505045801401, 0.003367003286257386, 0.013468013145029545, 0.02525252476334572, 0.005050505045801401, 0.01515151560306549, 0.006734006572514772, 0.020202020183205605, 0.011784511618316174, 0.008417508564889431, 0.006734006572514772, 0.005050505045801401, 0.005050505045801401, 0.018518518656492233, 0.006734006572514772, 0.016835017129778862, 0.010101010091602802, 0.008417508564889431, 0.01515151560306549, 0.006734006572514772, 0.018518518656492233, 0.008417508564889431, 0.01515151560306549, 0.016835017129778862, 0.05050504952669144, 0.020202020183205605, 0.023569023236632347, 0.020202020183205605, 0.013468013145029545, 0.021043771877884865, 0.016835017129778862, 0.03030303120613098, 0.03030303120613098, 0.018518518656492233, 0.018518518656492233, 0.02861952781677246, 0.02693602629005909, 0.008417508564889431, 0.02525252476334572, 0.010101010091602802, 0.04545454680919647, 0.05387205258011818, 0.02861952781677246, 0.05387205258011818, 0.02693602629005909, 0.031144781038165092, 0.02525252476334572, 0.06397306174039841, 0.04377104341983795, 0.09595959633588791, 0.0791245773434639, 0.05723905563354492, 0.1026936024427414, 0.06734006851911545, 0.04377104341983795, 0.10101009905338287, 0.0791245773434639, 0.02525252476334572, 0.031986530870199203, 0.04377104341983795, 0.14478114247322083, 0.1430976390838623, 0.0555555559694767, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9113472758},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.845819306},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.843738308},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8220515361},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8078937555},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8013818761},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7912808661},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7879606262},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7146932284},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7146464659},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7088711196},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6875584589},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6830106631},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6640946513},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6441147567},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6395085092},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6383861786},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6259586598},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.620183315},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6104447258},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6091119525},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6052539272},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6005541513},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5980873543},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5848648487},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5730101898},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5713383849},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5708707446},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.557683314},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5490436786},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5475004694},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.544098393},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5413276274},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5360666843},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5295548097},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5258020017},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5134095593},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5061962193},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5023265068},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4992985397},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4948442768},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4777520601},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4777286745},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4722339118},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4438131304},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.440644873},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4389029158},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4324728769},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.409780677},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3951903293},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3873807515},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3840955851},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3813715867},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3777473816},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3577908715},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3416573125},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3308080805},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3192924605},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.3091446874},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.3028783196},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2836349608},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2827581374},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2795548063},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2656659175},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2611999626},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2545127201},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2452651502},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2248410021},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2229353718},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.2066264473},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1852085666},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1543911331},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0295314251}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_classification/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..5d4dc37cd6c1604c4979cfd72a9e43838577723e
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_all/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),TabICLv2 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned),TabDPT-Turbo (default),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabM (tuned + ensembled),LimiX (default),TabDPT (tuned + ensembled),RealMLP (tuned),LightGBM (tuned + ensembled),TabM (tuned),BetaTabPFN (default),TabDPT (tuned),CatBoost (tuned + ensembled),TabSwift (default),LightGBM (tuned),iLTM (tuned + ensembled),CatBoost (tuned),ModernNCA (tuned),CatBoost (default),xRFM (tuned + ensembled),XGBoost (tuned + ensembled),TabDPT (default),XGBoost (tuned),xRFM (tuned),SAP-RPT-OSS (default),EBM (tuned + ensembled),ModernNCA (tuned + ensembled),TorchMLP (tuned + ensembled),TabM (default),FastaiMLP (tuned + ensembled),ModernNCA (default),ChimeraBoost (tuned + ensembled),EBM (tuned),iLTM (tuned),ChimeraBoost (tuned),EBM (default),RealMLP (default),TorchMLP (tuned),FastaiMLP (tuned),ExtraTrees (tuned + ensembled),ChimeraBoost (default),ExtraTrees (tuned),XGBoost (default),TabSTAR (tuned),OrionMSP (default),TabSTAR (tuned + ensembled),RandomForest (tuned + ensembled),LightGBM (default),RandomForest (tuned),iLTM (default),TorchMLP (default),Linear (tuned + ensembled),TabFlex (default),PerpetualBooster (tuned + ensembled),Linear (tuned),TabSTAR (default),FastaiMLP (default),xRFM (default),PerpetualBooster (tuned),KNN (tuned + ensembled),ExtraTrees (default),RandomForest (default),Linear (default),PerpetualBooster (default),KNN (tuned),KNN (default)
+TabFM (default),,0.7592593,0.7979798,0.8333333,0.8181818,0.78535354,0.7996633,0.7912458,0.8518519,0.8518519,0.8636364,0.88383836,0.8762626,0.8989899,0.8484849,0.8939394,0.9107744,0.8796296,0.8922559,0.8973064,0.8821549,0.90909094,0.90909094,0.90909094,0.90909094,0.9057239,0.90909094,0.92424244,0.9040404,0.9292929,0.8956229,0.85858583,0.90909094,0.93602693,0.9478114,0.9158249,0.8989899,0.9191919,0.92087543,0.92087543,0.9292929,0.9292929,0.9158249,0.93097645,0.959596,0.93097645,0.94612795,0.93939394,0.93771046,0.96127945,0.8518519,0.9410775,0.8518519,0.959596,0.96127945,0.95791245,0.9410775,0.96633,0.94276094,0.93771046,0.9848485,0.94949496,0.89057237,0.96801347,0.962963,0.9814815,0.962963,0.9882155,0.9814815,0.96633,0.98316497,0.962963,0.9949495
+"AutoGluon 1.5 (extreme, 4h)",0.24074075,,0.44444445,0.493266,0.64141417,0.5892256,0.5993266,0.60942763,0.6447812,0.79545456,0.7845118,0.79461277,0.74579126,0.7895623,0.8215488,0.8518519,0.80723906,0.7777778,0.81313133,0.8636364,0.8249158,0.8552188,0.8855219,0.85942763,0.87542087,0.8712121,0.8787879,0.87542087,0.79461277,0.8787879,0.8543771,0.7626263,0.86195284,0.9175084,0.9023569,0.86195284,0.86195284,0.8703704,0.9141414,0.8737374,0.9057239,0.9175084,0.8787879,0.88720536,0.9292929,0.8855219,0.9191919,0.92424244,0.9074074,0.94949496,0.8114478,0.9284512,0.81313133,0.93434346,0.95454544,0.93602693,0.94276094,0.95791245,0.9006734,0.9124579,0.95117843,0.9225589,0.8518519,0.95286196,0.95454544,0.9570707,0.9410775,0.96127945,0.962963,0.9452862,0.96464646,0.95454544,0.9949495
+TabPFN-3 (default),0.2020202,0.5555556,,0.53367,0.6329966,0.6178451,0.64309764,0.63973063,0.73569024,0.72727275,0.7171717,0.7626263,0.73063976,0.7996633,0.7626263,0.8030303,0.8047138,0.7685185,0.8299663,0.8265993,0.86195284,0.84343433,0.8518519,0.8333333,0.85353535,0.84680134,0.84343433,0.8341751,0.84175086,0.85016835,0.8400673,0.7676768,0.84343433,0.88720536,0.88720536,0.84511787,0.84090906,0.88131315,0.8888889,0.86195284,0.90824914,0.88720536,0.8796296,0.90909094,0.92424244,0.89141417,0.93097645,0.88720536,0.9292929,0.95117843,0.84090906,0.9259259,0.84511787,0.92424244,0.9478114,0.94612795,0.94612795,0.95286196,0.9074074,0.9292929,0.94612795,0.9158249,0.87542087,0.95454544,0.96127945,0.95791245,0.9410775,0.9781145,0.9814815,0.94949496,0.969697,0.959596,0.996633
+TabICLv2 (default),0.16666667,0.506734,0.46632996,,0.5774411,0.56060606,0.58501685,0.57575756,0.7121212,0.7020202,0.7070707,0.73905724,0.7314815,0.79713804,0.7289562,0.7845118,0.7676768,0.7676768,0.7962963,0.7861953,0.83838385,0.8181818,0.83164984,0.7929293,0.8299663,0.8215488,0.8114478,0.8030303,0.81986535,0.81313133,0.80050504,0.7323232,0.8215488,0.8400673,0.8670034,0.83164984,0.8367003,0.87794614,0.8518519,0.8484849,0.86531985,0.8636364,0.8703704,0.8855219,0.90909094,0.88720536,0.88973063,0.8636364,0.8939394,0.9175084,0.8299663,0.9225589,0.82828283,0.92087543,0.93771046,0.92087543,0.9175084,0.9410775,0.9191919,0.9107744,0.95117843,0.92760944,0.8771044,0.94612795,0.93771046,0.956229,0.9444444,0.95791245,0.96633,0.9537037,0.97643095,0.95454544,0.996633
+TabPFN-2.6 (default),0.18181819,0.35858586,0.36700338,0.42255893,,0.48653197,0.54713804,0.5126263,0.62794614,0.7037037,0.6750842,0.72390574,0.68013465,0.7020202,0.74915826,0.7828283,0.7744108,0.75420874,0.72727275,0.7895623,0.7777778,0.8333333,0.8299663,0.7828283,0.82828283,0.8215488,0.8030303,0.81649834,0.7373737,0.84511787,0.8030303,0.7710438,0.81986535,0.8771044,0.87205386,0.8341751,0.83164984,0.86531985,0.87542087,0.8577441,0.8821549,0.8821549,0.8518519,0.8855219,0.9158249,0.8787879,0.8973064,0.8821549,0.88973063,0.9478114,0.8097643,0.8821549,0.8181818,0.90909094,0.9444444,0.9326599,0.9259259,0.94949496,0.8973064,0.9107744,0.9410775,0.9175084,0.84343433,0.93434346,0.93771046,0.93771046,0.93434346,0.959596,0.9713805,0.9410775,0.962963,0.95117843,0.993266
+RealTabPFN-2.5 (tuned + ensembled),0.21464646,0.4107744,0.38215488,0.43939394,0.513468,,0.5824916,0.60942763,0.5774411,0.6919192,0.7121212,0.7070707,0.7020202,0.6851852,0.73569024,0.7710438,0.74915826,0.7744108,0.72727275,0.79461277,0.77272725,0.8021886,0.7752525,0.79461277,0.82323235,0.8047138,0.8063973,0.8047138,0.7037037,0.82239056,0.8047138,0.7685185,0.81986535,0.81649834,0.85353535,0.81313133,0.81902355,0.83164984,0.8308081,0.8442761,0.8552188,0.84175086,0.84511787,0.85690236,0.8973064,0.87542087,0.87542087,0.8552188,0.8771044,0.9141414,0.8114478,0.9057239,0.81649834,0.9040404,0.93434346,0.92087543,0.9124579,0.9410775,0.89057237,0.9074074,0.94612795,0.9057239,0.84343433,0.93434346,0.93771046,0.9478114,0.9158249,0.956229,0.956229,0.93013465,0.96464646,0.93771046,0.993266
+RealTabPFN-2.5 (default),0.2003367,0.4006734,0.35690236,0.41498315,0.45286196,0.41750842,,0.46548823,0.5521886,0.66161615,0.6750842,0.7104377,0.6851852,0.6818182,0.7255892,0.74242425,0.7558923,0.763468,0.71127945,0.7592593,0.8013468,0.7811448,0.73569024,0.76599324,0.7626263,0.7929293,0.8030303,0.7777778,0.73400676,0.8080808,0.8114478,0.7828283,0.8097643,0.76599324,0.84680134,0.81313133,0.8367003,0.8333333,0.8215488,0.84175086,0.8611111,0.83164984,0.8400673,0.8703704,0.8989899,0.8771044,0.87205386,0.86026937,0.8855219,0.8939394,0.8333333,0.9326599,0.8341751,0.86531985,0.93434346,0.9141414,0.9175084,0.96127945,0.8922559,0.92676765,0.94612795,0.8989899,0.86195284,0.93097645,0.9410775,0.95286196,0.9023569,0.95791245,0.95791245,0.9478114,0.9713805,0.94612795,0.9949495
+RealTabPFN-2.5 (tuned),0.20875421,0.3905724,0.36026937,0.42424244,0.48737374,0.3905724,0.5345118,,0.57239056,0.66498315,0.68686867,0.7020202,0.6700337,0.6818182,0.72727275,0.75420874,0.75,0.7558923,0.72643095,0.7811448,0.7617845,0.8013468,0.75252527,0.7777778,0.77946126,0.7912458,0.790404,0.7878788,0.7104377,0.8047138,0.7912458,0.76094276,0.81986535,0.80723906,0.84511787,0.8063973,0.8030303,0.81649834,0.81397307,0.8333333,0.84343433,0.83501685,0.83501685,0.8484849,0.8956229,0.8670034,0.8695286,0.84680134,0.8787879,0.9074074,0.8030303,0.9006734,0.8114478,0.88720536,0.9292929,0.9124579,0.9040404,0.9410775,0.87542087,0.8973064,0.9318182,0.8956229,0.84343433,0.9259259,0.93602693,0.93771046,0.9124579,0.94949496,0.95286196,0.93350166,0.96464646,0.94276094,0.993266
+TabDPT-Turbo (default),0.14814815,0.35521886,0.26430976,0.28787878,0.37205386,0.42255893,0.44781145,0.4276094,,0.54545456,0.5286195,0.5361953,0.5925926,0.62457913,0.61868685,0.60942763,0.5892256,0.5740741,0.68013465,0.65319866,0.72643095,0.6212121,0.6936027,0.64646465,0.65656567,0.6750842,0.67676765,0.6632997,0.80723906,0.67676765,0.70538723,0.64141417,0.6818182,0.68013465,0.7255892,0.66835016,0.75084174,0.7558923,0.73905724,0.719697,0.77356905,0.76599324,0.74915826,0.7777778,0.7912458,0.8148148,0.82575756,0.77609426,0.85353535,0.8265993,0.75420874,0.8670034,0.76094276,0.84511787,0.86026937,0.86026937,0.8367003,0.87205386,0.8249158,0.9057239,0.8973064,0.8367003,0.8299663,0.9158249,0.8922559,0.9141414,0.90909094,0.9486532,0.95286196,0.9225589,0.96127945,0.95117843,0.9949495
+RealMLP (tuned + ensembled),0.14814815,0.20454545,0.27272728,0.2979798,0.2962963,0.30808082,0.33838382,0.33501685,0.45454547,,0.4722222,0.55387205,0.46969697,0.5058923,0.67760944,0.6734007,0.6111111,0.6329966,0.5429293,0.6952862,0.54882157,0.72306395,0.6969697,0.70538723,0.6666667,0.7121212,0.7255892,0.76936024,0.5892256,0.7929293,0.7138047,0.6262626,0.76936024,0.6851852,0.73569024,0.7777778,0.790404,0.7323232,0.75420874,0.7861953,0.7828283,0.7786195,0.78872055,0.8695286,0.7929293,0.81060606,0.81313133,0.77946126,0.82239056,0.8888889,0.7676768,0.7828283,0.7744108,0.8855219,0.9057239,0.9107744,0.8787879,0.9141414,0.8703704,0.8265993,0.93097645,0.87205386,0.81313133,0.93434346,0.92760944,0.93771046,0.90909094,0.95454544,0.95454544,0.89057237,0.96633,0.9326599,0.9949495
+"AutoGluon 1.4 (best, 4h)",0.13636364,0.21548821,0.28282827,0.2929293,0.32491583,0.28787878,0.32491583,0.3131313,0.47138047,0.5277778,,0.54545456,0.44444445,0.5252525,0.6144781,0.6969697,0.6043771,0.6060606,0.5521886,0.6919192,0.5715488,0.7222222,0.70538723,0.7171717,0.65488213,0.7407407,0.6910775,0.74747473,0.60942763,0.73905724,0.7171717,0.5959596,0.7441077,0.6952862,0.77609426,0.7289562,0.76094276,0.73569024,0.78367,0.76094276,0.7845118,0.8063973,0.7929293,0.8030303,0.83164984,0.8114478,0.75084174,0.8299663,0.77946126,0.88383836,0.76599324,0.753367,0.77272725,0.83501685,0.90909094,0.8678451,0.8804714,0.8939394,0.86616164,0.83838385,0.92424244,0.88973063,0.8148148,0.92760944,0.8956229,0.92760944,0.8922559,0.93434346,0.92760944,0.9191919,0.9478114,0.92760944,0.996633
+TabM (tuned + ensembled),0.116161615,0.2053872,0.23737374,0.26094276,0.2760943,0.2929293,0.28956228,0.2979798,0.46380472,0.44612795,0.45454547,,0.43265992,0.5092593,0.5673401,0.6144781,0.6666667,0.63973063,0.54882157,0.61279464,0.57239056,0.6557239,0.6170034,0.6329966,0.6388889,0.6969697,0.6144781,0.6902357,0.5959596,0.72727275,0.6178451,0.6329966,0.74747473,0.6734007,0.7020202,0.8442761,0.7138047,0.7171717,0.74579126,0.7861953,0.7323232,0.7777778,0.7786195,0.8173401,0.79545456,0.7744108,0.74579126,0.8148148,0.74747473,0.8484849,0.7676768,0.7592593,0.7744108,0.7962963,0.9015151,0.8400673,0.8737374,0.9057239,0.8670034,0.82828283,0.9057239,0.8829966,0.81649834,0.9006734,0.8670034,0.9259259,0.88383836,0.92087543,0.9141414,0.9107744,0.962963,0.92424244,0.986532
+LimiX (default),0.12373737,0.25420874,0.26936027,0.2685185,0.31986532,0.2979798,0.3148148,0.32996634,0.4074074,0.530303,0.5555556,0.5673401,,0.50757575,0.5749158,0.65488213,0.57828283,0.5824916,0.53872055,0.66835016,0.56565654,0.6700337,0.712963,0.6700337,0.66835016,0.6969697,0.6902357,0.7037037,0.57912457,0.7222222,0.68939394,0.6296296,0.7121212,0.6818182,0.6919192,0.6700337,0.6759259,0.72390574,0.75757575,0.7373737,0.7373737,0.76599324,0.73905724,0.74579126,0.7626263,0.7584175,0.7929293,0.77946126,0.7962963,0.8215488,0.7281145,0.79208755,0.72390574,0.83164984,0.83501685,0.84511787,0.82323235,0.82828283,0.82828283,0.7878788,0.8636364,0.8484849,0.78198653,0.8737374,0.9107744,0.8787879,0.84511787,0.9107744,0.92424244,0.8552188,0.9141414,0.8787879,0.9747475
+TabDPT (tuned + ensembled),0.1010101,0.21043772,0.2003367,0.20286195,0.2979798,0.3148148,0.3181818,0.3181818,0.37542087,0.49410775,0.47474748,0.49074075,0.49242425,,0.5715488,0.56397307,0.5395623,0.537037,0.64983165,0.5808081,0.60016835,0.5959596,0.62794614,0.5875421,0.61279464,0.62457913,0.6355219,0.6052188,0.6784512,0.63804716,0.6540404,0.60942763,0.64057237,0.6666667,0.6792929,0.6355219,0.6952862,0.71464646,0.68686867,0.67087543,0.7289562,0.70538723,0.6734007,0.7154882,0.75420874,0.7592593,0.7676768,0.73905724,0.7979798,0.7979798,0.7222222,0.8299663,0.72390574,0.82828283,0.8400673,0.83838385,0.82323235,0.8552188,0.7811448,0.85016835,0.8855219,0.7996633,0.8063973,0.89057237,0.9015151,0.89646465,0.89309764,0.94612795,0.93939394,0.86868685,0.9410775,0.9259259,0.9949495
+RealMLP (tuned),0.15151516,0.17845118,0.23737374,0.27104378,0.25084174,0.26430976,0.27441078,0.27272728,0.38131315,0.3223906,0.3855219,0.43265992,0.42508417,0.42845118,,0.5555556,0.51683503,0.5774411,0.493266,0.56397307,0.493266,0.5959596,0.5547138,0.57239056,0.57239056,0.6212121,0.6010101,0.64983165,0.53872055,0.68434346,0.6304714,0.56228954,0.6750842,0.6111111,0.67676765,0.6851852,0.6969697,0.66835016,0.64309764,0.7188552,0.6936027,0.67171717,0.6902357,0.76936024,0.7558923,0.76936024,0.7407407,0.7138047,0.7592593,0.8097643,0.75084174,0.7247475,0.7558923,0.81313133,0.83501685,0.8299663,0.8299663,0.8771044,0.81649834,0.8089225,0.88383836,0.82575756,0.8013468,0.90824914,0.8787879,0.88383836,0.8670034,0.88383836,0.8939394,0.8636364,0.95117843,0.8989899,0.9848485
+LightGBM (tuned + ensembled),0.10606061,0.14814815,0.1969697,0.21548821,0.21717171,0.22895622,0.25757575,0.24579124,0.3905724,0.32659933,0.3030303,0.3855219,0.34511784,0.43602693,0.44444445,,0.46717173,0.46969697,0.48821548,0.56397307,0.52020204,0.60858583,0.5412458,0.5799663,0.52693605,0.6557239,0.5808081,0.6818182,0.59090906,0.7138047,0.63468015,0.5808081,0.6641414,0.56397307,0.61279464,0.67424244,0.6178451,0.65488213,0.706229,0.7138047,0.6734007,0.7373737,0.7710438,0.7087542,0.73063976,0.7154882,0.73316497,0.8080808,0.7373737,0.86531985,0.7516835,0.7407407,0.75420874,0.81313133,0.9478114,0.86531985,0.8888889,0.83164984,0.8367003,0.7845118,0.8922559,0.85690236,0.8215488,0.84175086,0.8265993,0.9124579,0.8888889,0.9292929,0.93939394,0.9124579,0.9444444,0.93013465,0.993266
+TabM (tuned),0.08922559,0.19276094,0.1952862,0.23232323,0.22558923,0.25084174,0.24410774,0.25,0.4107744,0.3888889,0.3956229,0.33333334,0.42171717,0.46043772,0.483165,0.5328283,,0.55387205,0.496633,0.56060606,0.52693605,0.56397307,0.56397307,0.57912457,0.5993266,0.63131315,0.5555556,0.6136364,0.5648148,0.63131315,0.5824916,0.5555556,0.64983165,0.6212121,0.6632997,0.7222222,0.6691919,0.6750842,0.6936027,0.7138047,0.6632997,0.7003367,0.7289562,0.7676768,0.7483165,0.7407407,0.6952862,0.753367,0.7121212,0.8047138,0.75084174,0.73569024,0.760101,0.7744108,0.8249158,0.7912458,0.8484849,0.89057237,0.84680134,0.81649834,0.9040404,0.85858583,0.7996633,0.8636364,0.82828283,0.9124579,0.85353535,0.8771044,0.8888889,0.8888889,0.95454544,0.9057239,0.97979796
+BetaTabPFN (default),0.12037037,0.22222222,0.23148148,0.23232323,0.24579124,0.22558923,0.23653199,0.24410774,0.42592594,0.36700338,0.3939394,0.36026937,0.41750842,0.46296296,0.42255893,0.530303,0.44612795,,0.51683503,0.5218855,0.513468,0.540404,0.53367,0.5496633,0.5530303,0.61279464,0.56060606,0.57912457,0.5824916,0.5942761,0.56902355,0.55387205,0.65656567,0.5521886,0.63973063,0.62794614,0.6439394,0.6481481,0.6557239,0.7003367,0.6212121,0.66835016,0.74242425,0.7104377,0.7188552,0.7255892,0.7087542,0.73063976,0.7255892,0.74747473,0.7811448,0.73905724,0.7861953,0.76094276,0.78367,0.7744108,0.84511787,0.8299663,0.86026937,0.8240741,0.85690236,0.8787879,0.84090906,0.84511787,0.7929293,0.87542087,0.88383836,0.85353535,0.86195284,0.93434346,0.9259259,0.956229,0.9882155
+TabDPT (tuned),0.107744105,0.18686868,0.17003366,0.2037037,0.27272728,0.27272728,0.28872055,0.27356902,0.31986532,0.4570707,0.44781145,0.45117846,0.46127945,0.35016835,0.506734,0.5117845,0.503367,0.483165,,0.53367,0.540404,0.540404,0.5555556,0.5412458,0.55387205,0.5774411,0.5858586,0.57323235,0.63468015,0.5824916,0.5875421,0.55050504,0.5942761,0.60690236,0.62457913,0.5942761,0.63973063,0.64141417,0.6481481,0.63468015,0.6515151,0.65824914,0.64309764,0.66161615,0.6936027,0.7188552,0.7121212,0.68686867,0.73905724,0.72390574,0.68434346,0.8265993,0.6851852,0.75252527,0.7777778,0.7895623,0.77272725,0.8249158,0.7617845,0.8367003,0.8636364,0.7828283,0.7878788,0.8703704,0.8670034,0.8737374,0.88383836,0.90909094,0.9074074,0.8518519,0.93939394,0.9141414,0.9915825
+CatBoost (tuned + ensembled),0.1026936,0.13636364,0.17340067,0.2138047,0.21043772,0.2053872,0.24074075,0.21885522,0.34680134,0.30471382,0.30808082,0.3872054,0.33164984,0.41919193,0.43602693,0.43602693,0.43939394,0.4781145,0.46632996,,0.5084175,0.48400673,0.52693605,0.57828283,0.5210438,0.5925926,0.56060606,0.5749158,0.537037,0.61026937,0.6077441,0.57239056,0.5993266,0.56565654,0.6060606,0.6212121,0.6010101,0.6178451,0.65319866,0.6439394,0.66835016,0.68013465,0.6885522,0.6885522,0.72053874,0.68350166,0.73905724,0.7222222,0.756734,0.8484849,0.7441077,0.6952862,0.73905724,0.8148148,0.8703704,0.8611111,0.8425926,0.8299663,0.8063973,0.73569024,0.84175086,0.82323235,0.7895623,0.8333333,0.8484849,0.86026937,0.8821549,0.9057239,0.9259259,0.85690236,0.9074074,0.9107744,0.993266
+TabSwift (default),0.11784512,0.17508417,0.13804714,0.16161616,0.22222222,0.22727273,0.19865319,0.23821549,0.27356902,0.45117846,0.42845118,0.4276094,0.43434343,0.39983165,0.506734,0.479798,0.47306398,0.48653197,0.45959595,0.49158248,,0.5016835,0.53535354,0.51430976,0.537037,0.54208755,0.56228954,0.55387205,0.5673401,0.5841751,0.56228954,0.5841751,0.5858586,0.57239056,0.6195286,0.6111111,0.6936027,0.62794614,0.61279464,0.6178451,0.66161615,0.64309764,0.6262626,0.6632997,0.6919192,0.72727275,0.6734007,0.7188552,0.6952862,0.6952862,0.75252527,0.8670034,0.75757575,0.6936027,0.78198653,0.74747473,0.7861953,0.8181818,0.76094276,0.9057239,0.88383836,0.7878788,0.8265993,0.8552188,0.81986535,0.88720536,0.8518519,0.85690236,0.88720536,0.8863636,0.93097645,0.94276094,0.9949495
+LightGBM (tuned),0.09090909,0.14478114,0.15656565,0.18181819,0.16666667,0.19781145,0.21885522,0.19865319,0.37878788,0.27693602,0.2777778,0.3442761,0.32996634,0.4040404,0.4040404,0.39141414,0.43602693,0.45959595,0.45959595,0.51599324,0.4983165,,0.48653197,0.5252525,0.479798,0.61195284,0.5496633,0.5867003,0.55387205,0.62457913,0.5841751,0.54208755,0.6329966,0.53367,0.5824916,0.63804716,0.59175086,0.6296296,0.65488213,0.6784512,0.62457913,0.67676765,0.72390574,0.68939394,0.68602693,0.66161615,0.6910775,0.7626263,0.72306395,0.82912457,0.7407407,0.73063976,0.74579126,0.75757575,0.9166667,0.8249158,0.8678451,0.81313133,0.81313133,0.77272725,0.87457913,0.8392256,0.8080808,0.8265993,0.81397307,0.8939394,0.8636364,0.9175084,0.9158249,0.8956229,0.94276094,0.92087543,0.9949495
+iLTM (tuned + ensembled),0.09090909,0.11447811,0.14814815,0.16835018,0.17003366,0.22474748,0.26430976,0.24747474,0.30639732,0.3030303,0.2946128,0.38299662,0.28703704,0.37205386,0.44528618,0.4587542,0.43602693,0.46632996,0.44444445,0.47306398,0.46464646,0.513468,,0.503367,0.54208755,0.540404,0.57828283,0.5429293,0.530303,0.5976431,0.62542087,0.54713804,0.6010101,0.5707071,0.5976431,0.56565654,0.64309764,0.65656567,0.65656567,0.64141417,0.706229,0.6666667,0.67171717,0.7070707,0.7037037,0.71464646,0.706229,0.71464646,0.73905724,0.8030303,0.6969697,0.6910775,0.6978114,0.7861953,0.85353535,0.82323235,0.84343433,0.8299663,0.74747473,0.8047138,0.8518519,0.7828283,0.7861953,0.86195284,0.8787879,0.87205386,0.84680134,0.90909094,0.90909094,0.8367003,0.9124579,0.8863636,0.9814815
+CatBoost (tuned),0.09090909,0.14057238,0.16666667,0.20707071,0.21717171,0.2053872,0.23400673,0.22222222,0.35353535,0.2946128,0.28282827,0.36700338,0.32996634,0.4124579,0.4276094,0.42003366,0.42087543,0.4503367,0.4587542,0.42171717,0.48569024,0.47474748,0.496633,,0.5,0.57828283,0.5462963,0.57575756,0.52693605,0.6077441,0.5942761,0.55723906,0.6026936,0.5589225,0.6043771,0.61026937,0.57575756,0.5993266,0.61868685,0.63131315,0.63468015,0.65909094,0.6919192,0.67676765,0.70791245,0.67424244,0.7171717,0.7104377,0.7289562,0.84175086,0.74242425,0.6919192,0.7441077,0.8063973,0.84343433,0.8400673,0.8299663,0.8097643,0.8013468,0.7289562,0.83164984,0.81565654,0.79461277,0.81986535,0.8367003,0.85353535,0.8737374,0.8939394,0.9175084,0.85016835,0.9141414,0.9175084,0.993266
+ModernNCA (tuned),0.09090909,0.124579124,0.14646465,0.17003366,0.17171717,0.17676768,0.23737374,0.22053872,0.34343433,0.33333334,0.34511784,0.3611111,0.33164984,0.3872054,0.4276094,0.47306398,0.4006734,0.4469697,0.44612795,0.47895622,0.46296296,0.52020204,0.45791245,0.5,,0.537037,0.5496633,0.537037,0.51683503,0.54461277,0.6010101,0.530303,0.5740741,0.50084174,0.57575756,0.55050504,0.59511787,0.6750842,0.6212121,0.6195286,0.6144781,0.6287879,0.6363636,0.68013465,0.7154882,0.6936027,0.68013465,0.6902357,0.7037037,0.7222222,0.7138047,0.7104377,0.72053874,0.7441077,0.7626263,0.76430976,0.8063973,0.8047138,0.7626263,0.7979798,0.8299663,0.7929293,0.7912458,0.8341751,0.8703704,0.85016835,0.8400673,0.8804714,0.8670034,0.84680134,0.9006734,0.88720536,0.98316497
+CatBoost (default),0.09427609,0.12878788,0.15319866,0.17845118,0.17845118,0.1952862,0.20707071,0.20875421,0.32491583,0.28787878,0.25925925,0.3030303,0.3030303,0.37542087,0.37878788,0.3442761,0.36868685,0.3872054,0.42255893,0.4074074,0.45791245,0.38804713,0.45959595,0.42171717,0.46296296,,0.49579126,0.47727272,0.5286195,0.51430976,0.56228954,0.54545456,0.56397307,0.5117845,0.57239056,0.56565654,0.5707071,0.5673401,0.62542087,0.62457913,0.6178451,0.66835016,0.66498315,0.6481481,0.6885522,0.65824914,0.68013465,0.74242425,0.7003367,0.76936024,0.7382155,0.6784512,0.7407407,0.76599324,0.82323235,0.8013468,0.87542087,0.81649834,0.77946126,0.74747473,0.8215488,0.8013468,0.7979798,0.83501685,0.8013468,0.85690236,0.86531985,0.8787879,0.88720536,0.8703704,0.9006734,0.9225589,0.989899
+xRFM (tuned + ensembled),0.09090909,0.121212125,0.15656565,0.18855219,0.1969697,0.1936027,0.1969697,0.20959596,0.32323232,0.27441078,0.30892256,0.3855219,0.3097643,0.3644781,0.3989899,0.41919193,0.44444445,0.43939394,0.41414142,0.43939394,0.43771043,0.4503367,0.42171717,0.4537037,0.4503367,0.50420874,,0.48232323,0.48989898,0.5050505,0.60858583,0.48821548,0.54713804,0.5117845,0.52356905,0.54545456,0.5833333,0.5555556,0.5942761,0.6144781,0.56649834,0.6043771,0.6363636,0.6262626,0.63131315,0.6632997,0.65656567,0.6296296,0.7003367,0.7020202,0.69612795,0.7323232,0.6952862,0.7407407,0.7188552,0.7811448,0.74579126,0.81313133,0.81986535,0.81313133,0.8148148,0.8367003,0.7676768,0.8804714,0.9410775,0.8493266,0.8829966,0.93939394,0.92760944,0.8695286,0.90909094,0.90909094,0.9915825
+XGBoost (tuned + ensembled),0.07575758,0.124579124,0.16582492,0.1969697,0.18350169,0.1952862,0.22222222,0.21212122,0.33670035,0.23063973,0.25252524,0.3097643,0.2962963,0.39478114,0.35016835,0.3181818,0.38636363,0.42087543,0.42676768,0.42508417,0.44612795,0.41329965,0.4570707,0.42424244,0.46296296,0.52272725,0.5176768,,0.5,0.59175086,0.57575756,0.5252525,0.56060606,0.503367,0.57828283,0.5774411,0.56397307,0.5976431,0.5740741,0.62457913,0.6010101,0.6144781,0.66161615,0.63973063,0.65319866,0.6228956,0.66498315,0.6574074,0.7087542,0.8821549,0.72053874,0.68939394,0.72053874,0.79713804,0.88720536,0.82239056,0.7996633,0.7929293,0.79461277,0.7382155,0.7979798,0.8063973,0.7744108,0.7912458,0.82323235,0.8240741,0.84343433,0.8855219,0.9175084,0.8484849,0.8989899,0.89646465,0.9848485
+TabDPT (default),0.0959596,0.2053872,0.15824915,0.18013468,0.26262626,0.2962963,0.26599327,0.28956228,0.19276094,0.4107744,0.3905724,0.4040404,0.42087543,0.32154882,0.46127945,0.4090909,0.4351852,0.41750842,0.36531988,0.46296296,0.43265992,0.44612795,0.46969697,0.47306398,0.483165,0.47138047,0.510101,0.5,,0.5185185,0.51683503,0.5151515,0.4983165,0.5151515,0.53367,0.543771,0.55387205,0.5707071,0.5774411,0.5328283,0.55723906,0.5875421,0.56228954,0.5513468,0.6026936,0.6363636,0.65488213,0.6296296,0.6818182,0.6329966,0.63804716,0.74242425,0.63804716,0.67171717,0.6919192,0.7222222,0.7407407,0.75084174,0.6952862,0.84343433,0.7929293,0.72390574,0.73400676,0.8299663,0.7962963,0.81986535,0.83501685,0.85858583,0.83164984,0.8299663,0.8922559,0.87542087,0.993266
+XGBoost (tuned),0.07070707,0.121212125,0.14983165,0.18686868,0.15488215,0.17760943,0.1919192,0.1952862,0.32323232,0.20707071,0.26094276,0.27272728,0.2777778,0.36195287,0.31565657,0.28619528,0.36868685,0.4057239,0.41750842,0.38973063,0.41582492,0.37542087,0.4023569,0.3922559,0.4553872,0.48569024,0.4949495,0.40824917,0.4814815,,0.5412458,0.51683503,0.540404,0.4983165,0.54545456,0.54882157,0.53367,0.56565654,0.5277778,0.5959596,0.5774411,0.5774411,0.6296296,0.61279464,0.6296296,0.6212121,0.6658249,0.61616164,0.7037037,0.8249158,0.7070707,0.67171717,0.709596,0.7744108,0.8518519,0.81313133,0.7912458,0.77272725,0.7828283,0.7289562,0.77272725,0.8047138,0.7710438,0.7811448,0.8114478,0.81649834,0.8442761,0.88720536,0.9040404,0.84175086,0.8973064,0.8939394,0.9814815
+xRFM (tuned),0.104377106,0.1456229,0.15993266,0.19949494,0.1969697,0.1952862,0.18855219,0.20875421,0.2946128,0.28619528,0.28282827,0.38215488,0.31060606,0.3459596,0.36952862,0.36531988,0.41750842,0.43097642,0.4124579,0.3922559,0.43771043,0.41582492,0.37457913,0.4057239,0.3989899,0.43771043,0.39141414,0.42424244,0.483165,0.4587542,,0.5,0.5058923,0.42592594,0.506734,0.52356905,0.5496633,0.540404,0.5580808,0.5740741,0.540404,0.5808081,0.5976431,0.5808081,0.61279464,0.63973063,0.63973063,0.6144781,0.6851852,0.60942763,0.7037037,0.70538723,0.7087542,0.67171717,0.6734007,0.72727275,0.74747473,0.8063973,0.81649834,0.8249158,0.793771,0.8400673,0.7861953,0.86195284,0.9040404,0.8308081,0.87205386,0.8855219,0.86531985,0.88720536,0.8989899,0.9141414,0.9915825
+SAP-RPT-OSS (default),0.14141414,0.23737374,0.23232323,0.26767677,0.22895622,0.23148148,0.21717171,0.23905724,0.35858586,0.37373737,0.4040404,0.36700338,0.37037036,0.3905724,0.43771043,0.41919193,0.44444445,0.44612795,0.44949496,0.4276094,0.41582492,0.45791245,0.45286196,0.44276094,0.46969697,0.45454547,0.5117845,0.47474748,0.4848485,0.483165,0.5,,0.493266,0.5084175,0.5808081,0.53367,0.59090906,0.54208755,0.483165,0.52693605,0.52272725,0.5117845,0.5597643,0.63468015,0.63973063,0.66498315,0.61279464,0.55050504,0.63804716,0.5942761,0.61279464,0.64141417,0.60942763,0.6523569,0.66161615,0.6675084,0.6599327,0.73569024,0.76094276,0.75084174,0.7744108,0.76936024,0.76430976,0.7710438,0.77272725,0.78198653,0.74242425,0.77946126,0.7845118,0.8114478,0.81313133,0.8518519,0.9848485
+EBM (tuned + ensembled),0.09090909,0.13804714,0.15656565,0.17845118,0.18013468,0.18013468,0.19023569,0.18013468,0.3181818,0.23063973,0.25589225,0.25252524,0.28787878,0.3594276,0.32491583,0.33585858,0.35016835,0.34343433,0.4057239,0.4006734,0.41414142,0.36700338,0.3989899,0.39730638,0.42592594,0.43602693,0.45286196,0.43939394,0.5016835,0.45959595,0.49410775,0.506734,,0.44444445,0.53198653,0.530303,0.5176768,0.543771,0.5185185,0.68939394,0.537037,0.55050504,0.73400676,0.60942763,0.6388889,0.59343433,0.6641414,0.6212121,0.6700337,0.69276094,0.72053874,0.6885522,0.72053874,0.7104377,0.7087542,0.7323232,0.79461277,0.76094276,0.8148148,0.7744108,0.7996633,0.8333333,0.7962963,0.81649834,0.76430976,0.8173401,0.84175086,0.84175086,0.84175086,0.86531985,0.8973064,0.9099327,0.98316497
+ModernNCA (tuned + ensembled),0.06397306,0.082491584,0.112794615,0.15993266,0.12289562,0.18350169,0.23400673,0.19276094,0.31986532,0.3148148,0.30471382,0.32659933,0.3181818,0.33333334,0.3888889,0.43602693,0.37878788,0.44781145,0.39309764,0.43434343,0.4276094,0.46632996,0.42929292,0.44107744,0.49915826,0.48821548,0.48821548,0.496633,0.4848485,0.5016835,0.5740741,0.49158248,0.5555556,,0.5151515,0.51683503,0.54882157,0.6481481,0.55387205,0.59511787,0.56565654,0.5858586,0.6026936,0.6515151,0.62457913,0.6262626,0.6296296,0.6010101,0.6632997,0.7087542,0.64141417,0.6195286,0.64309764,0.68350166,0.72053874,0.6885522,0.72053874,0.72727275,0.6818182,0.7037037,0.75084174,0.6969697,0.7070707,0.76094276,0.8358586,0.75252527,0.7744108,0.82323235,0.8148148,0.7441077,0.8215488,0.8047138,0.94949496
+TorchMLP (tuned + ensembled),0.052188553,0.0976431,0.112794615,0.13299663,0.12794612,0.14646465,0.15319866,0.15488215,0.27441078,0.26430976,0.22390573,0.2979798,0.30808082,0.32070708,0.32323232,0.3872054,0.33670035,0.36026937,0.37542087,0.3939394,0.38047138,0.41750842,0.4023569,0.3956229,0.42424244,0.4276094,0.47643098,0.42171717,0.46632996,0.45454547,0.493266,0.41919193,0.46801347,0.4848485,,0.46296296,0.5774411,0.55387205,0.5151515,0.530303,0.5547138,0.5521886,0.56397307,0.5993266,0.7592593,0.68771046,0.64646465,0.57239056,0.6599327,0.6750842,0.6936027,0.63131315,0.70117843,0.6969697,0.6919192,0.7222222,0.7752525,0.8737374,0.7685185,0.7777778,0.8063973,0.8030303,0.7878788,0.8442761,0.7845118,0.84175086,0.8047138,0.8030303,0.8013468,0.86026937,0.8989899,0.8922559,0.97979796
+TabM (default),0.08417509,0.13804714,0.15488215,0.16835018,0.16582492,0.18686868,0.18686868,0.1936027,0.33164984,0.22222222,0.27104378,0.1557239,0.32996634,0.3644781,0.3148148,0.32575756,0.2777778,0.37205386,0.4057239,0.37878788,0.3888889,0.36195287,0.43434343,0.38973063,0.44949496,0.43434343,0.45454547,0.42255893,0.45622894,0.45117846,0.47643098,0.46632996,0.46969697,0.483165,0.537037,,0.56902355,0.54882157,0.5,0.543771,0.5361953,0.52356905,0.5513468,0.6111111,0.62794614,0.6515151,0.61616164,0.5875421,0.64141417,0.6363636,0.72390574,0.6632997,0.73063976,0.7037037,0.64983165,0.7154882,0.74747473,0.7811448,0.76936024,0.73400676,0.790404,0.78535354,0.8047138,0.79461277,0.74242425,0.7962963,0.8114478,0.8299663,0.82828283,0.8400673,0.8804714,0.90909094,0.97643095
+FastaiMLP (tuned + ensembled),0.1010101,0.13804714,0.1590909,0.16329966,0.16835018,0.18097644,0.16329966,0.1969697,0.24915825,0.20959596,0.23905724,0.28619528,0.32407406,0.30471382,0.3030303,0.38215488,0.33080807,0.3560606,0.36026937,0.3989899,0.30639732,0.40824917,0.35690236,0.42424244,0.40488216,0.42929292,0.41666666,0.43602693,0.44612795,0.46632996,0.4503367,0.4090909,0.48232323,0.45117846,0.42255893,0.43097642,,0.537037,0.53535354,0.53198653,0.47474748,0.55723906,0.543771,0.56649834,0.56228954,0.7087542,0.6026936,0.56060606,0.61279464,0.6111111,0.6447812,0.5875421,0.64983165,0.68771046,0.6506734,0.6986532,0.6969697,0.760101,0.7710438,0.7811448,0.76094276,0.7996633,0.7626263,0.95117843,0.7845118,0.7895623,0.8047138,0.8265993,0.7895623,0.85858583,0.8855219,0.8821549,0.97979796
+ModernNCA (default),0.08080808,0.12962963,0.11868687,0.12205387,0.13468014,0.16835018,0.16666667,0.18350169,0.24410774,0.26767677,0.26430976,0.28282827,0.2760943,0.28535354,0.33164984,0.34511784,0.32491583,0.35185185,0.35858586,0.38215488,0.37205386,0.37037036,0.34343433,0.4006734,0.32491583,0.43265992,0.44444445,0.4023569,0.42929292,0.43434343,0.45959595,0.45791245,0.45622894,0.35185185,0.44612795,0.45117846,0.46296296,,0.5707071,0.5117845,0.46717173,0.5858586,0.55050504,0.513468,0.54208755,0.543771,0.57575756,0.6228956,0.6228956,0.57912457,0.66498315,0.6919192,0.67171717,0.6515151,0.6700337,0.7171717,0.73063976,0.77946126,0.7373737,0.7710438,0.7845118,0.7651515,0.7744108,0.7895623,0.75757575,0.8148148,0.84511787,0.8358586,0.8367003,0.86026937,0.8670034,0.8973064,0.986532
+ChimeraBoost (tuned + ensembled),0.07912458,0.08585858,0.11111111,0.14814815,0.124579124,0.16919191,0.17845118,0.18602693,0.26094276,0.24579124,0.21632996,0.25420874,0.24242425,0.3131313,0.35690236,0.29377106,0.30639732,0.3442761,0.35185185,0.34680134,0.3872054,0.34511784,0.34343433,0.38131315,0.37878788,0.37457913,0.4057239,0.42592594,0.42255893,0.4722222,0.44191918,0.51683503,0.4814815,0.44612795,0.4848485,0.5,0.46464646,0.42929292,,0.52020204,0.5185185,0.59511787,0.537037,0.53872055,0.57912457,0.5277778,0.55723906,0.7744108,0.5976431,0.6447812,0.6952862,0.6363636,0.6952862,0.6481481,0.7314815,0.6936027,0.76430976,0.7003367,0.7070707,0.7441077,0.74915826,0.72727275,0.7710438,0.7710438,0.76599324,0.7878788,0.76936024,0.8047138,0.7912458,0.82323235,0.8400673,0.8636364,0.9789562
+EBM (tuned),0.07912458,0.12626262,0.13804714,0.15151516,0.14225589,0.1557239,0.15824915,0.16666667,0.28030303,0.2138047,0.23905724,0.2138047,0.26262626,0.32912457,0.28114477,0.28619528,0.28619528,0.2996633,0.36531988,0.3560606,0.38215488,0.32154882,0.35858586,0.36868685,0.38047138,0.37542087,0.3855219,0.37542087,0.46717173,0.4040404,0.42592594,0.47306398,0.31060606,0.40488216,0.46969697,0.45622894,0.46801347,0.48821548,0.479798,,0.4814815,0.5058923,0.6422559,0.5286195,0.5858586,0.56649834,0.6010101,0.5740741,0.63973063,0.64141417,0.7020202,0.6523569,0.70791245,0.66498315,0.6599327,0.7037037,0.7441077,0.73569024,0.77946126,0.753367,0.7777778,0.8013468,0.77946126,0.7811448,0.7407407,0.79461277,0.8148148,0.82828283,0.8215488,0.8400673,0.8787879,0.9107744,0.98316497
+iLTM (tuned),0.07070707,0.09427609,0.091750845,0.13468014,0.11784512,0.14478114,0.1388889,0.15656565,0.22643098,0.21717171,0.21548821,0.26767677,0.26262626,0.27104378,0.30639732,0.32659933,0.33670035,0.37878788,0.34848484,0.33164984,0.33838382,0.37542087,0.29377106,0.36531988,0.3855219,0.38215488,0.4335017,0.3989899,0.44276094,0.42255893,0.45959595,0.47727272,0.46296296,0.43434343,0.44528618,0.46380472,0.5252525,0.5328283,0.4814815,0.5185185,,0.5328283,0.5589225,0.56228954,0.5707071,0.5892256,0.5707071,0.56397307,0.6296296,0.61616164,0.67760944,0.6010101,0.67171717,0.66498315,0.6700337,0.7104377,0.746633,0.7828283,0.6725589,0.71633,0.73905724,0.7138047,0.74915826,0.8030303,0.7895623,0.77946126,0.7811448,0.8114478,0.81313133,0.7929293,0.85353535,0.8459596,0.969697
+ChimeraBoost (tuned),0.07070707,0.082491584,0.112794615,0.13636364,0.11784512,0.15824915,0.16835018,0.16498317,0.23400673,0.22138047,0.1936027,0.22222222,0.23400673,0.2946128,0.32828283,0.26262626,0.2996633,0.33164984,0.34175083,0.31986532,0.35690236,0.32323232,0.33333334,0.3409091,0.37121212,0.33164984,0.3956229,0.3855219,0.4124579,0.42255893,0.41919193,0.48821548,0.44949496,0.41414142,0.44781145,0.47643098,0.44276094,0.41414142,0.40488216,0.49410775,0.46717173,,0.5117845,0.5218855,0.54882157,0.50420874,0.53535354,0.6952862,0.5824916,0.6271044,0.6818182,0.6262626,0.6818182,0.62457913,0.6919192,0.66835016,0.7289562,0.6936027,0.6919192,0.7154882,0.71127945,0.7020202,0.7592593,0.75420874,0.74242425,0.746633,0.74915826,0.79461277,0.76936024,0.8063973,0.8324916,0.84175086,0.969697
+EBM (default),0.08417509,0.121212125,0.12037037,0.12962963,0.14814815,0.15488215,0.15993266,0.16498317,0.25084174,0.21127947,0.20707071,0.22138047,0.26094276,0.32659933,0.3097643,0.22895622,0.27104378,0.25757575,0.35690236,0.3114478,0.37373737,0.2760943,0.32828283,0.30808082,0.36363637,0.33501685,0.36363637,0.33838382,0.43771043,0.37037036,0.4023569,0.4402357,0.26599327,0.39730638,0.43602693,0.4486532,0.45622894,0.44949496,0.46296296,0.3577441,0.44107744,0.48821548,,0.5084175,0.5841751,0.5521886,0.5808081,0.53872055,0.5993266,0.56565654,0.6885522,0.6481481,0.68686867,0.64141417,0.6228956,0.67171717,0.74579126,0.72727275,0.75757575,0.75757575,0.7592593,0.79208755,0.77609426,0.7861953,0.7087542,0.7744108,0.83838385,0.8148148,0.7996633,0.86531985,0.85858583,0.9259259,0.9814815
+RealMLP (default),0.06902357,0.112794615,0.09090909,0.11447811,0.11447811,0.14309764,0.12962963,0.15151516,0.22222222,0.13047138,0.1969697,0.18265994,0.25420874,0.28451177,0.23063973,0.2912458,0.23232323,0.28956228,0.33838382,0.3114478,0.33670035,0.31060606,0.2929293,0.32323232,0.31986532,0.35185185,0.37373737,0.36026937,0.4486532,0.3872054,0.41919193,0.36531988,0.3905724,0.34848484,0.4006734,0.3888889,0.4335017,0.48653197,0.46127945,0.47138047,0.43771043,0.4781145,0.49158248,,0.5589225,0.56902355,0.5673401,0.5252525,0.6060606,0.59090906,0.6700337,0.5993266,0.68350166,0.63973063,0.63468015,0.6784512,0.7154882,0.7550505,0.7289562,0.77356905,0.7777778,0.7710438,0.76599324,0.84680134,0.73063976,0.81986535,0.8080808,0.81313133,0.79461277,0.84511787,0.8939394,0.88720536,0.9814815
+TorchMLP (tuned),0.04040404,0.07070707,0.07575758,0.09090909,0.08417509,0.1026936,0.1010101,0.104377106,0.20875421,0.20707071,0.16835018,0.20454545,0.23737374,0.24579124,0.24410774,0.26936027,0.2516835,0.28114477,0.30639732,0.27946126,0.30808082,0.31397307,0.2962963,0.29208755,0.28451177,0.3114478,0.36868685,0.34680134,0.39730638,0.37037036,0.3872054,0.36026937,0.3611111,0.37542087,0.24074075,0.37205386,0.43771043,0.45791245,0.42087543,0.41414142,0.42929292,0.45117846,0.41582492,0.44107744,,0.530303,0.5530303,0.496633,0.57575756,0.540404,0.66835016,0.56902355,0.6700337,0.6077441,0.5799663,0.64309764,0.6936027,0.78198653,0.68350166,0.7558923,0.7222222,0.70791245,0.7592593,0.7626263,0.72390574,0.7592593,0.7550505,0.75252527,0.75420874,0.8215488,0.84175086,0.84511787,0.9713805
+FastaiMLP (tuned),0.06902357,0.11447811,0.10858586,0.112794615,0.121212125,0.124579124,0.12289562,0.13299663,0.18518518,0.18939394,0.18855219,0.22558923,0.2415825,0.24074075,0.23063973,0.28451177,0.25925925,0.27441078,0.28114477,0.3164983,0.27272728,0.33838382,0.28535354,0.32575756,0.30639732,0.34175083,0.33670035,0.37710437,0.36363637,0.37878788,0.36026937,0.33501685,0.40656567,0.37373737,0.31228957,0.34848484,0.2912458,0.45622894,0.4722222,0.4335017,0.4107744,0.49579126,0.44781145,0.43097642,0.46969697,,0.53367,0.52020204,0.55387205,0.54545456,0.5892256,0.52693605,0.5875421,0.60690236,0.57575756,0.6212121,0.6296296,0.6691919,0.6936027,0.7121212,0.6952862,0.72643095,0.7323232,0.83838385,0.7070707,0.7373737,0.7558923,0.7676768,0.7373737,0.8148148,0.8249158,0.8552188,0.97306395
+ExtraTrees (tuned + ensembled),0.053872053,0.08080808,0.06902357,0.11026936,0.1026936,0.124579124,0.12794612,0.13047138,0.17424242,0.18686868,0.24915825,0.25420874,0.20707071,0.23232323,0.25925925,0.266835,0.30471382,0.2912458,0.28787878,0.26094276,0.32659933,0.30892256,0.29377106,0.28282827,0.31986532,0.31986532,0.34343433,0.33501685,0.34511784,0.33417508,0.36026937,0.3872054,0.33585858,0.37037036,0.35353535,0.3838384,0.39730638,0.42424244,0.44276094,0.3989899,0.42929292,0.46464646,0.41919193,0.43265992,0.4469697,0.46632996,,0.48821548,0.67171717,0.4983165,0.5589225,0.6195286,0.5740741,0.6632997,0.5707071,0.72727275,0.6026936,0.6195286,0.6456229,0.6784512,0.6700337,0.6675084,0.70454544,0.68013465,0.7373737,0.6919192,0.8299663,0.9057239,0.87205386,0.7676768,0.7828283,0.88383836,0.9915825
+ChimeraBoost (default),0.060606062,0.07575758,0.112794615,0.13636364,0.11784512,0.14478114,0.13973063,0.15319866,0.22390573,0.22053872,0.17003366,0.18518518,0.22053872,0.26094276,0.28619528,0.1919192,0.246633,0.26936027,0.3131313,0.2777778,0.28114477,0.23737374,0.28535354,0.28956228,0.3097643,0.25757575,0.37037036,0.3425926,0.37037036,0.3838384,0.3855219,0.44949496,0.37878788,0.3989899,0.4276094,0.4124579,0.43939394,0.37710437,0.22558923,0.42592594,0.43602693,0.30471382,0.46127945,0.47474748,0.503367,0.479798,0.5117845,,0.53198653,0.5597643,0.65656567,0.5707071,0.66498315,0.5925926,0.61026937,0.6111111,0.68350166,0.63468015,0.6784512,0.64141417,0.67171717,0.6919192,0.72727275,0.6936027,0.6952862,0.6969697,0.6944444,0.67171717,0.72727275,0.8013468,0.77609426,0.8518519,0.9747475
+ExtraTrees (tuned),0.062289562,0.09259259,0.07070707,0.10606061,0.11026936,0.12289562,0.11447811,0.121212125,0.14646465,0.17760943,0.22053872,0.25252524,0.2037037,0.2020202,0.24074075,0.26262626,0.28787878,0.27441078,0.26094276,0.24326599,0.30471382,0.27693602,0.26094276,0.27104378,0.2962963,0.2996633,0.2996633,0.2912458,0.3181818,0.2962963,0.3148148,0.36195287,0.32996634,0.33670035,0.34006733,0.35858586,0.3872054,0.37710437,0.4023569,0.36026937,0.37037036,0.41750842,0.4006734,0.3939394,0.42424244,0.44612795,0.32828283,0.46801347,,0.43939394,0.5597643,0.6077441,0.5707071,0.5555556,0.52020204,0.63383836,0.57239056,0.5892256,0.6262626,0.6750842,0.64309764,0.63973063,0.7003367,0.65656567,0.6936027,0.67171717,0.7895623,0.82828283,0.82575756,0.75757575,0.7626263,0.8888889,0.989899
+XGBoost (default),0.038720537,0.05050505,0.04882155,0.082491584,0.052188553,0.08585858,0.10606061,0.09259259,0.17340067,0.11111111,0.116161615,0.15151516,0.17845118,0.2020202,0.19023569,0.13468014,0.1952862,0.25252524,0.2760943,0.15151516,0.30471382,0.17087542,0.1969697,0.15824915,0.2777778,0.23063973,0.2979798,0.11784512,0.36700338,0.17508417,0.3905724,0.4057239,0.30723906,0.2912458,0.32491583,0.36363637,0.3888889,0.42087543,0.35521886,0.35858586,0.3838384,0.37289563,0.43434343,0.4090909,0.45959595,0.45454547,0.5016835,0.4402357,0.56060606,,0.6144781,0.5875421,0.6195286,0.55387205,0.5530303,0.61616164,0.6750842,0.64309764,0.6363636,0.66161615,0.6178451,0.65656567,0.7154882,0.6818182,0.7121212,0.66835016,0.74242425,0.7861953,0.7979798,0.76094276,0.76430976,0.82323235,0.95454544
+TabSTAR (tuned),0.14814815,0.18855219,0.1590909,0.17003366,0.19023569,0.18855219,0.16666667,0.1969697,0.24579124,0.23232323,0.23400673,0.23232323,0.2718855,0.2777778,0.24915825,0.2483165,0.24915825,0.21885522,0.31565657,0.25589225,0.24747474,0.25925925,0.3030303,0.25757575,0.28619528,0.26178452,0.30387205,0.27946126,0.36195287,0.2929293,0.2962963,0.3872054,0.27946126,0.35858586,0.30639732,0.2760943,0.35521886,0.33501685,0.30471382,0.2979798,0.3223906,0.3181818,0.3114478,0.32996634,0.33164984,0.4107744,0.44107744,0.34343433,0.4402357,0.3855219,,0.47643098,0.59175086,0.46464646,0.4090909,0.4983165,0.3838384,0.44107744,0.6010101,0.57575756,0.6111111,0.63131315,0.9478114,0.6329966,0.5858586,0.6262626,0.7003367,0.6885522,0.68686867,0.74915826,0.66498315,0.85353535,0.94612795
+OrionMSP (default),0.05892256,0.07154882,0.074074075,0.077441074,0.11784512,0.09427609,0.06734007,0.099326596,0.13299663,0.21717171,0.246633,0.24074075,0.20791246,0.17003366,0.27525252,0.25925925,0.26430976,0.26094276,0.17340067,0.30471382,0.13299663,0.26936027,0.30892256,0.30808082,0.28956228,0.32154882,0.26767677,0.31060606,0.25757575,0.32828283,0.2946128,0.35858586,0.3114478,0.38047138,0.36868685,0.33670035,0.4124579,0.30808082,0.36363637,0.3476431,0.3989899,0.37373737,0.35185185,0.4006734,0.43097642,0.47306398,0.38047138,0.42929292,0.3922559,0.4124579,0.52356905,,0.5252525,0.43434343,0.479798,0.46969697,0.5185185,0.530303,0.61279464,0.65656567,0.6329966,0.6388889,0.68350166,0.68013465,0.64646465,0.66835016,0.63804716,0.6599327,0.65488213,0.74579126,0.7811448,0.7676768,0.9713805
+TabSTAR (tuned + ensembled),0.14814815,0.18686868,0.15488215,0.17171717,0.18181819,0.18350169,0.16582492,0.18855219,0.23905724,0.22558923,0.22727273,0.22558923,0.2760943,0.2760943,0.24410774,0.24579124,0.239899,0.2138047,0.3148148,0.26094276,0.24242425,0.25420874,0.30218855,0.25589225,0.27946126,0.25925925,0.30471382,0.27946126,0.36195287,0.29040405,0.2912458,0.3905724,0.27946126,0.35690236,0.29882154,0.26936027,0.35016835,0.32828283,0.30471382,0.29208755,0.32828283,0.3181818,0.3131313,0.3164983,0.32996634,0.4124579,0.42592594,0.33501685,0.42929292,0.38047138,0.40824917,0.47474748,,0.45791245,0.41414142,0.493266,0.3872054,0.43771043,0.59090906,0.5673401,0.6060606,0.62794614,0.96127945,0.62794614,0.5824916,0.6195286,0.6885522,0.6818182,0.68350166,0.74242425,0.6632997,0.84511787,0.94612795
+RandomForest (tuned + ensembled),0.04040404,0.065656565,0.07575758,0.07912458,0.09090909,0.0959596,0.13468014,0.112794615,0.15488215,0.11447811,0.16498317,0.2037037,0.16835018,0.17171717,0.18686868,0.18686868,0.22558923,0.23905724,0.24747474,0.18518518,0.30639732,0.24242425,0.2138047,0.1936027,0.25589225,0.23400673,0.25925925,0.20286195,0.32828283,0.22558923,0.32828283,0.3476431,0.28956228,0.3164983,0.3030303,0.2962963,0.31228957,0.34848484,0.35185185,0.33501685,0.33501685,0.37542087,0.35858586,0.36026937,0.3922559,0.39309764,0.33670035,0.4074074,0.44444445,0.44612795,0.53535354,0.56565654,0.54208755,,0.496633,0.67760944,0.5084175,0.5707071,0.60942763,0.61616164,0.6212121,0.6228956,0.67087543,0.6262626,0.70538723,0.6447812,0.7558923,0.8047138,0.86195284,0.73316497,0.7087542,0.86195284,0.97306395
+LightGBM (default),0.038720537,0.045454547,0.052188553,0.062289562,0.055555556,0.065656565,0.065656565,0.07070707,0.13973063,0.09427609,0.09090909,0.09848485,0.16498317,0.15993266,0.16498317,0.052188553,0.17508417,0.21632996,0.22222222,0.12962963,0.21801347,0.083333336,0.14646465,0.15656565,0.23737374,0.17676768,0.28114477,0.112794615,0.30808082,0.14814815,0.32659933,0.33838382,0.2912458,0.27946126,0.30808082,0.35016835,0.3493266,0.32996634,0.2685185,0.34006733,0.32996634,0.30808082,0.37710437,0.36531988,0.42003366,0.42424244,0.42929292,0.38973063,0.479798,0.4469697,0.59090906,0.52020204,0.5858586,0.503367,,0.54461277,0.5993266,0.6077441,0.5892256,0.62457913,0.61279464,0.60942763,0.6952862,0.65488213,0.67676765,0.6734007,0.7138047,0.72390574,0.7314815,0.74579126,0.77272725,0.8299663,0.9688552
+RandomForest (tuned),0.042087544,0.06397306,0.053872053,0.07912458,0.06734007,0.07912458,0.08585858,0.08754209,0.13973063,0.08922559,0.13215488,0.15993266,0.15488215,0.16161616,0.17003366,0.13468014,0.20875421,0.22558923,0.21043772,0.1388889,0.25252524,0.17508417,0.17676768,0.15993266,0.23569024,0.19865319,0.21885522,0.17760943,0.2777778,0.18686868,0.27272728,0.33249158,0.26767677,0.3114478,0.2777778,0.28451177,0.3013468,0.28282827,0.30639732,0.2962963,0.28956228,0.33164984,0.32828283,0.32154882,0.35690236,0.37878788,0.27272728,0.3888889,0.3661616,0.3838384,0.5016835,0.530303,0.506734,0.3223906,0.4553872,,0.4814815,0.5117845,0.5808081,0.6026936,0.5976431,0.5976431,0.6481481,0.6178451,0.65824914,0.63131315,0.72306395,0.7373737,0.76094276,0.7188552,0.6919192,0.8518519,0.9747475
+iLTM (default),0.05892256,0.057239056,0.053872053,0.082491584,0.074074075,0.08754209,0.082491584,0.0959596,0.16329966,0.121212125,0.11952862,0.12626262,0.17676768,0.17676768,0.17003366,0.11111111,0.15151516,0.15488215,0.22727273,0.1574074,0.2138047,0.13215488,0.15656565,0.17003366,0.1936027,0.124579124,0.25420874,0.2003367,0.25925925,0.20875421,0.25252524,0.34006733,0.2053872,0.27946126,0.22474748,0.25252524,0.3030303,0.26936027,0.23569024,0.25589225,0.253367,0.27104378,0.25420874,0.28451177,0.30639732,0.37037036,0.39730638,0.3164983,0.4276094,0.32491583,0.61616164,0.4814815,0.61279464,0.49158248,0.4006734,0.5185185,,0.537037,0.59090906,0.5875421,0.5589225,0.61616164,0.6885522,0.60942763,0.6060606,0.57575756,0.7020202,0.63131315,0.6885522,0.7171717,0.65656567,0.8080808,0.93602693
+TorchMLP (default),0.033670034,0.042087544,0.047138046,0.05892256,0.05050505,0.05892256,0.038720537,0.05892256,0.12794612,0.08585858,0.10606061,0.09427609,0.17171717,0.14478114,0.12289562,0.16835018,0.10942761,0.17003366,0.17508417,0.17003366,0.18181819,0.18686868,0.17003366,0.19023569,0.1952862,0.18350169,0.18686868,0.20707071,0.24915825,0.22727273,0.1936027,0.26430976,0.23905724,0.27272728,0.12626262,0.21885522,0.239899,0.22053872,0.2996633,0.26430976,0.21717171,0.30639732,0.27272728,0.24494949,0.21801347,0.33080807,0.38047138,0.36531988,0.4107744,0.35690236,0.5589225,0.46969697,0.56228954,0.42929292,0.3922559,0.48821548,0.46296296,,0.59090906,0.5984849,0.57239056,0.6195286,0.6851852,0.6515151,0.6212121,0.6178451,0.68350166,0.6632997,0.64141417,0.73400676,0.7154882,0.8181818,0.956229
+Linear (tuned + ensembled),0.057239056,0.099326596,0.09259259,0.08080808,0.1026936,0.10942761,0.107744105,0.124579124,0.17508417,0.12962963,0.13383839,0.13299663,0.17171717,0.21885522,0.18350169,0.16329966,0.15319866,0.13973063,0.23821549,0.1936027,0.23905724,0.18686868,0.25252524,0.19865319,0.23737374,0.22053872,0.18013468,0.2053872,0.30471382,0.21717171,0.18350169,0.23905724,0.18518518,0.3181818,0.23148148,0.23063973,0.22895622,0.26262626,0.2929293,0.22053872,0.32744107,0.30808082,0.24242425,0.27104378,0.3164983,0.30639732,0.3543771,0.32154882,0.37373737,0.36363637,0.3989899,0.3872054,0.4090909,0.3905724,0.4107744,0.41919193,0.4090909,0.4090909,,0.496633,0.47306398,0.6919192,0.54208755,0.4814815,0.53872055,0.4949495,0.56565654,0.55387205,0.54208755,0.8080808,0.6010101,0.7020202,0.9040404
+TabFlex (default),0.062289562,0.08754209,0.07070707,0.08922559,0.08922559,0.09259259,0.07323232,0.1026936,0.09427609,0.17340067,0.16161616,0.17171717,0.21212122,0.14983165,0.19107744,0.21548821,0.18350169,0.17592593,0.16329966,0.26430976,0.09427609,0.22727273,0.1952862,0.27104378,0.2020202,0.25252524,0.18686868,0.26178452,0.15656565,0.27104378,0.17508417,0.24915825,0.22558923,0.2962963,0.22222222,0.26599327,0.21885522,0.22895622,0.25589225,0.246633,0.28367004,0.28451177,0.24242425,0.22643098,0.24410774,0.28787878,0.32154882,0.35858586,0.32491583,0.33838382,0.42424244,0.34343433,0.43265992,0.3838384,0.37542087,0.39730638,0.4124579,0.40151516,0.503367,,0.506734,0.5345118,0.5808081,0.56397307,0.53367,0.56565654,0.56565654,0.53367,0.5707071,0.65656567,0.6363636,0.72390574,0.92087543
+PerpetualBooster (tuned + ensembled),0.015151516,0.04882155,0.053872053,0.04882155,0.05892256,0.053872053,0.053872053,0.06818182,0.1026936,0.06902357,0.07575758,0.09427609,0.13636364,0.11447811,0.116161615,0.107744105,0.0959596,0.14309764,0.13636364,0.15824915,0.116161615,0.12542087,0.14814815,0.16835018,0.17003366,0.17845118,0.18518518,0.2020202,0.20707071,0.22727273,0.20622896,0.22558923,0.2003367,0.24915825,0.1936027,0.20959596,0.23905724,0.21548821,0.25084174,0.22222222,0.26094276,0.28872055,0.24074075,0.22222222,0.2777778,0.30471382,0.32996634,0.32828283,0.35690236,0.38215488,0.3888889,0.36700338,0.3939394,0.37878788,0.3872054,0.4023569,0.44107744,0.4276094,0.52693605,0.493266,,0.543771,0.4781145,0.48821548,0.55050504,0.712963,0.54461277,0.59090906,0.57912457,0.61279464,0.7979798,0.6885522,0.94276094
+Linear (tuned),0.05050505,0.077441074,0.08417509,0.07239057,0.082491584,0.09427609,0.1010101,0.104377106,0.16329966,0.12794612,0.11026936,0.11700337,0.15151516,0.2003367,0.17424242,0.14309764,0.14141414,0.121212125,0.21717171,0.17676768,0.21212122,0.16077441,0.21717171,0.18434343,0.20707071,0.19865319,0.16329966,0.1936027,0.2760943,0.1952862,0.15993266,0.23063973,0.16666667,0.3030303,0.1969697,0.21464646,0.2003367,0.23484848,0.27272728,0.19865319,0.28619528,0.2979798,0.20791246,0.22895622,0.29208755,0.27356902,0.33249158,0.30808082,0.36026937,0.34343433,0.36868685,0.3611111,0.37205386,0.37710437,0.3905724,0.4023569,0.3838384,0.38047138,0.30808082,0.46548823,0.45622894,,0.5286195,0.45454547,0.52693605,0.479798,0.5277778,0.53535354,0.5117845,0.72390574,0.5925926,0.6851852,0.8973064
+TabSTAR (default),0.10942761,0.14814815,0.124579124,0.12289562,0.15656565,0.15656565,0.13804714,0.15656565,0.17003366,0.18686868,0.18518518,0.18350169,0.21801347,0.1936027,0.19865319,0.17845118,0.2003367,0.1590909,0.21212122,0.21043772,0.17340067,0.1919192,0.2138047,0.2053872,0.20875421,0.2020202,0.23232323,0.22558923,0.26599327,0.22895622,0.2138047,0.23569024,0.2037037,0.2929293,0.21212122,0.1952862,0.23737374,0.22558923,0.22895622,0.22053872,0.25084174,0.24074075,0.22390573,0.23400673,0.24074075,0.26767677,0.29545453,0.27272728,0.2996633,0.28451177,0.052188553,0.3164983,0.038720537,0.32912457,0.30471382,0.35185185,0.3114478,0.3148148,0.45791245,0.41919193,0.5218855,0.47138047,,0.4090909,0.4814815,0.540404,0.540404,0.52693605,0.4814815,0.5976431,0.5892256,0.70538723,0.9326599
+FastaiMLP (default),0.03198653,0.047138046,0.045454547,0.053872053,0.065656565,0.065656565,0.06902357,0.074074075,0.08417509,0.065656565,0.07239057,0.099326596,0.12626262,0.10942761,0.091750845,0.15824915,0.13636364,0.15488215,0.12962963,0.16666667,0.14478114,0.17340067,0.13804714,0.18013468,0.16582492,0.16498317,0.11952862,0.20875421,0.17003366,0.21885522,0.13804714,0.22895622,0.18350169,0.23905724,0.1557239,0.2053872,0.04882155,0.21043772,0.22895622,0.21885522,0.1969697,0.24579124,0.2138047,0.15319866,0.23737374,0.16161616,0.31986532,0.30639732,0.34343433,0.3181818,0.36700338,0.31986532,0.37205386,0.37373737,0.34511784,0.38215488,0.3905724,0.34848484,0.5185185,0.43602693,0.5117845,0.54545456,0.59090906,,0.54208755,0.540404,0.57912457,0.52020204,0.5361953,0.65488213,0.65656567,0.75420874,0.956229
+xRFM (default),0.037037037,0.045454547,0.038720537,0.062289562,0.062289562,0.062289562,0.05892256,0.06397306,0.107744105,0.07239057,0.104377106,0.13299663,0.08922559,0.09848485,0.121212125,0.17340067,0.17171717,0.20707071,0.13299663,0.15151516,0.18013468,0.18602693,0.121212125,0.16329966,0.12962963,0.19865319,0.05892256,0.17676768,0.2037037,0.18855219,0.0959596,0.22727273,0.23569024,0.16414142,0.21548821,0.25757575,0.21548821,0.24242425,0.23400673,0.25925925,0.21043772,0.25757575,0.2912458,0.26936027,0.2760943,0.2929293,0.26262626,0.30471382,0.30639732,0.28787878,0.41414142,0.35353535,0.41750842,0.2946128,0.32323232,0.34175083,0.3939394,0.37878788,0.46127945,0.46632996,0.44949496,0.47306398,0.5185185,0.45791245,,0.4739057,0.47306398,0.46296296,0.4814815,0.55723906,0.57912457,0.62794614,0.8989899
+PerpetualBooster (tuned),0.018518519,0.04292929,0.042087544,0.043771043,0.062289562,0.052188553,0.047138046,0.062289562,0.08585858,0.062289562,0.07239057,0.074074075,0.121212125,0.103535354,0.116161615,0.08754209,0.08754209,0.124579124,0.12626262,0.13973063,0.112794615,0.10606061,0.12794612,0.14646465,0.14983165,0.14309764,0.1506734,0.17592593,0.18013468,0.18350169,0.16919191,0.21801347,0.18265994,0.24747474,0.15824915,0.2037037,0.21043772,0.18518518,0.21212122,0.2053872,0.22053872,0.253367,0.22558923,0.18013468,0.24074075,0.26262626,0.30808082,0.3030303,0.32828283,0.33164984,0.37373737,0.33164984,0.38047138,0.35521886,0.32659933,0.36868685,0.42424244,0.38215488,0.5050505,0.43434343,0.28703704,0.52020204,0.45959595,0.45959595,0.52609426,,0.510101,0.54713804,0.5521886,0.5892256,0.7171717,0.65824914,0.92087543
+KNN (tuned + ensembled),0.037037037,0.05892256,0.05892256,0.055555556,0.065656565,0.08417509,0.0976431,0.08754209,0.09090909,0.09090909,0.107744105,0.116161615,0.15488215,0.10690235,0.13299663,0.11111111,0.14646465,0.116161615,0.116161615,0.11784512,0.14814815,0.13636364,0.15319866,0.12626262,0.15993266,0.13468014,0.11700337,0.15656565,0.16498317,0.1557239,0.12794612,0.25757575,0.15824915,0.22558923,0.1952862,0.18855219,0.1952862,0.15488215,0.23063973,0.18518518,0.21885522,0.25084174,0.16161616,0.1919192,0.24494949,0.24410774,0.17003366,0.30555555,0.21043772,0.25757575,0.2996633,0.36195287,0.3114478,0.24410774,0.28619528,0.27693602,0.2979798,0.3164983,0.43434343,0.43434343,0.4553872,0.4722222,0.45959595,0.42087543,0.52693605,0.48989898,,0.47138047,0.48989898,0.5959596,0.5875421,0.8695286,0.9747475
+ExtraTrees (default),0.011784512,0.038720537,0.021885522,0.042087544,0.04040404,0.043771043,0.042087544,0.05050505,0.0513468,0.045454547,0.065656565,0.07912458,0.08922559,0.053872053,0.116161615,0.07070707,0.12289562,0.14646465,0.09090909,0.09427609,0.14309764,0.082491584,0.09090909,0.10606061,0.11952862,0.121212125,0.060606062,0.11447811,0.14141414,0.112794615,0.11447811,0.22053872,0.15824915,0.17676768,0.1969697,0.17003366,0.17340067,0.16414142,0.1952862,0.17171717,0.18855219,0.2053872,0.18518518,0.18686868,0.24747474,0.23232323,0.09427609,0.32828283,0.17171717,0.2138047,0.3114478,0.34006733,0.3181818,0.1952862,0.2760943,0.26262626,0.36868685,0.33670035,0.44612795,0.46632996,0.4090909,0.46464646,0.47306398,0.479798,0.537037,0.45286196,0.5286195,,0.510101,0.5892256,0.55387205,0.6952862,0.96801347
+RandomForest (default),0.018518519,0.037037037,0.018518519,0.033670034,0.028619528,0.043771043,0.042087544,0.047138046,0.047138046,0.045454547,0.07239057,0.08585858,0.07575758,0.060606062,0.10606061,0.060606062,0.11111111,0.13804714,0.09259259,0.074074075,0.112794615,0.08417509,0.09090909,0.082491584,0.13299663,0.112794615,0.07239057,0.082491584,0.16835018,0.0959596,0.13468014,0.21548821,0.15824915,0.18518518,0.19865319,0.17171717,0.21043772,0.16329966,0.20875421,0.17845118,0.18686868,0.23063973,0.2003367,0.2053872,0.24579124,0.26262626,0.12794612,0.27272728,0.17424242,0.2020202,0.3131313,0.34511784,0.3164983,0.13804714,0.2685185,0.23905724,0.3114478,0.35858586,0.45791245,0.42929292,0.42087543,0.48821548,0.5185185,0.46380472,0.5185185,0.44781145,0.510101,0.48989898,,0.5892256,0.56060606,0.7020202,0.956229
+Linear (default),0.033670034,0.054713804,0.05050505,0.046296295,0.05892256,0.069865316,0.052188553,0.06649832,0.077441074,0.10942761,0.08080808,0.08922559,0.14478114,0.13131313,0.13636364,0.08754209,0.11111111,0.065656565,0.14814815,0.14309764,0.11363637,0.104377106,0.16329966,0.14983165,0.15319866,0.12962963,0.13047138,0.15151516,0.17003366,0.15824915,0.112794615,0.18855219,0.13468014,0.25589225,0.13973063,0.15993266,0.14141414,0.13973063,0.17676768,0.15993266,0.20707071,0.1936027,0.13468014,0.15488215,0.17845118,0.18518518,0.23232323,0.19865319,0.24242425,0.23905724,0.25084174,0.25420874,0.25757575,0.266835,0.25420874,0.28114477,0.28282827,0.26599327,0.1919192,0.34343433,0.3872054,0.2760943,0.4023569,0.34511784,0.44276094,0.4107744,0.4040404,0.4107744,0.4107744,,0.5176768,0.6077441,0.8552188
+PerpetualBooster (default),0.016835017,0.035353534,0.030303031,0.023569023,0.037037037,0.035353534,0.028619528,0.035353534,0.038720537,0.033670034,0.052188553,0.037037037,0.08585858,0.05892256,0.04882155,0.055555556,0.045454547,0.074074075,0.060606062,0.09259259,0.06902357,0.057239056,0.08754209,0.08585858,0.099326596,0.099326596,0.09090909,0.1010101,0.107744105,0.1026936,0.1010101,0.18686868,0.1026936,0.17845118,0.1010101,0.11952862,0.11447811,0.13299663,0.15993266,0.121212125,0.14646465,0.16750842,0.14141414,0.10606061,0.15824915,0.17508417,0.21717171,0.22390573,0.23737374,0.23569024,0.33501685,0.21885522,0.33670035,0.2912458,0.22727273,0.30808082,0.34343433,0.28451177,0.3989899,0.36363637,0.2020202,0.4074074,0.4107744,0.34343433,0.42087543,0.28282827,0.4124579,0.44612795,0.43939394,0.48232323,,0.56902355,0.85690236
+KNN (tuned),0.037037037,0.045454547,0.04040404,0.045454547,0.04882155,0.062289562,0.053872053,0.057239056,0.04882155,0.06734007,0.07239057,0.07575758,0.121212125,0.074074075,0.1010101,0.069865316,0.09427609,0.043771043,0.08585858,0.08922559,0.057239056,0.07912458,0.11363637,0.082491584,0.112794615,0.077441074,0.09090909,0.103535354,0.124579124,0.10606061,0.08585858,0.14814815,0.09006734,0.1952862,0.107744105,0.09090909,0.11784512,0.1026936,0.13636364,0.08922559,0.15404041,0.15824915,0.074074075,0.112794615,0.15488215,0.14478114,0.116161615,0.14814815,0.11111111,0.17676768,0.14646465,0.23232323,0.15488215,0.13804714,0.17003366,0.14814815,0.1919192,0.18181819,0.2979798,0.2760943,0.3114478,0.3148148,0.2946128,0.24579124,0.37205386,0.34175083,0.13047138,0.30471382,0.2979798,0.3922559,0.43097642,,0.9444444
+KNN (default),0.005050505,0.005050505,0.0033670033,0.0033670033,0.0067340066,0.0067340066,0.005050505,0.0067340066,0.005050505,0.005050505,0.0033670033,0.013468013,0.025252525,0.005050505,0.015151516,0.0067340066,0.02020202,0.011784512,0.008417509,0.0067340066,0.005050505,0.005050505,0.018518519,0.0067340066,0.016835017,0.01010101,0.008417509,0.015151516,0.0067340066,0.018518519,0.008417509,0.015151516,0.016835017,0.05050505,0.02020202,0.023569023,0.02020202,0.013468013,0.021043772,0.016835017,0.030303031,0.030303031,0.018518519,0.018518519,0.028619528,0.026936026,0.008417509,0.025252525,0.01010101,0.045454547,0.053872053,0.028619528,0.053872053,0.026936026,0.031144781,0.025252525,0.06397306,0.043771043,0.0959596,0.07912458,0.057239056,0.1026936,0.06734007,0.043771043,0.1010101,0.07912458,0.025252525,0.03198653,0.043771043,0.14478114,0.14309764,0.055555556,
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
index acdc169382735a445e9b87fa5ae8e52cbb5ab1d0..3ecb0cca03735d02eae1940862450a4eb541a3e2 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Classification | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_all/tasks_classification/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..2289f5d9be8792c0f31bea1108236245650375a3
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_classification/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2031.0,"score":0.728,"rank":5.0,"hrank":2.76,"imp":5.25,"train_time":147.57,"predict_time":1.973,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-51","imputed":false},{"position":1,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2006.0,"score":0.852,"rank":5.54,"hrank":1.68,"imp":3.722,"train_time":38.86,"predict_time":18.759,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+293\/-161","imputed":false},{"position":2,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1875.0,"score":0.632,"rank":9.31,"hrank":3.6,"imp":6.35,"train_time":314.18,"predict_time":2.249,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+170\/-80","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1850.0,"score":0.622,"rank":10.26,"hrank":3.71,"imp":7.876,"train_time":1.1,"predict_time":0.289,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+280\/-175","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1761.0,"score":0.477,"rank":14.12,"hrank":9.24,"imp":9.735,"train_time":2.41,"predict_time":0.701,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-87","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1751.0,"score":0.483,"rank":14.6,"hrank":5.07,"imp":8.717,"train_time":0.63,"predict_time":0.17,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+190\/-110","imputed":false},{"position":6,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1711.0,"score":0.46,"rank":16.71,"hrank":8.41,"imp":9.316,"train_time":378.5,"predict_time":0.168,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+142\/-85","imputed":false},{"position":7,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1705.0,"score":0.47,"rank":17.02,"hrank":7.77,"imp":9.178,"train_time":1.49,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-76","imputed":false},{"position":8,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1704.0,"score":0.43,"rank":17.04,"hrank":13.03,"imp":10.529,"train_time":1607.74,"predict_time":72.982,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-68","imputed":false},{"position":9,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1697.0,"score":0.452,"rank":17.43,"hrank":8.37,"imp":9.588,"train_time":378.5,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+155\/-95","imputed":false},{"position":10,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1673.0,"score":0.381,"rank":18.8,"hrank":12.4,"imp":11.387,"train_time":207.65,"predict_time":0.398,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-92","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1669.0,"score":0.367,"rank":19.06,"hrank":13.13,"imp":11.7,"train_time":93.21,"predict_time":0.758,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-76","imputed":false},{"position":12,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1668.0,"score":0.411,"rank":19.11,"hrank":11.12,"imp":11.214,"train_time":90.17,"predict_time":0.314,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-131","imputed":false},{"position":13,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1652.0,"score":0.374,"rank":20.02,"hrank":12.72,"imp":10.95,"train_time":1624.55,"predict_time":1.763,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+122\/-90","imputed":false},{"position":14,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1640.0,"score":0.35,"rank":20.76,"hrank":10.11,"imp":11.175,"train_time":783.29,"predict_time":13.79,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+190\/-114","imputed":false},{"position":15,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1636.0,"score":0.308,"rank":21.0,"hrank":11.26,"imp":11.793,"train_time":860.28,"predict_time":0.554,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+159\/-104","imputed":false},{"position":16,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1628.0,"score":0.344,"rank":21.53,"hrank":14.17,"imp":11.383,"train_time":2338.19,"predict_time":14.248,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-124","imputed":false},{"position":17,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1610.0,"score":0.35,"rank":22.66,"hrank":12.79,"imp":11.808,"train_time":90.17,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+144\/-110","imputed":false},{"position":18,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1606.0,"score":0.299,"rank":22.92,"hrank":18.3,"imp":12.162,"train_time":207.65,"predict_time":0.056,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-73","imputed":false},{"position":19,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1599.0,"score":0.31,"rank":23.37,"hrank":12.04,"imp":11.899,"train_time":783.29,"predict_time":1.924,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+180\/-114","imputed":false},{"position":20,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1571.0,"score":0.275,"rank":25.16,"hrank":14.26,"imp":12.216,"train_time":1624.55,"predict_time":0.083,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-94","imputed":false},{"position":21,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1565.0,"score":0.251,"rank":25.57,"hrank":13.41,"imp":12.672,"train_time":860.28,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-125","imputed":false},{"position":22,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1562.0,"score":0.227,"rank":25.74,"hrank":19.22,"imp":13.259,"train_time":93.21,"predict_time":0.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-73","imputed":false},{"position":23,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1561.0,"score":0.268,"rank":25.82,"hrank":16.13,"imp":12.843,"train_time":0.38,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-116","imputed":false},{"position":24,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1552.0,"score":0.258,"rank":26.4,"hrank":16.78,"imp":12.737,"train_time":2.3,"predict_time":0.82,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+133\/-99","imputed":false},{"position":25,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1531.0,"score":0.244,"rank":27.8,"hrank":16.13,"imp":13.434,"train_time":7.51,"predict_time":3.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-135","imputed":false},{"position":26,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1522.0,"score":0.205,"rank":28.46,"hrank":15.98,"imp":13.587,"train_time":1607.74,"predict_time":10.661,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-91","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1518.0,"score":0.177,"rank":28.68,"hrank":21.64,"imp":13.717,"train_time":2338.19,"predict_time":0.54,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-76","imputed":false},{"position":28,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1514.0,"score":0.247,"rank":28.99,"hrank":13.95,"imp":14.355,"train_time":0.79,"predict_time":0.23,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+182\/-170","imputed":false},{"position":29,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1490.0,"score":0.174,"rank":30.63,"hrank":18.37,"imp":13.722,"train_time":4805.68,"predict_time":461.79,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-129","imputed":false},{"position":30,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1488.0,"score":0.168,"rank":30.77,"hrank":23.79,"imp":13.578,"train_time":1169.32,"predict_time":1.442,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-77","imputed":false},{"position":31,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1474.0,"score":0.176,"rank":31.72,"hrank":23.99,"imp":15.517,"train_time":442.2,"predict_time":0.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+136\/-112","imputed":false},{"position":32,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1456.0,"score":0.175,"rank":33.02,"hrank":19.6,"imp":14.317,"train_time":0.47,"predict_time":0.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-108","imputed":false},{"position":33,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1424.0,"score":0.17,"rank":35.31,"hrank":24.58,"imp":15.536,"train_time":709.86,"predict_time":2.929,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-144","imputed":false},{"position":34,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1418.0,"score":0.141,"rank":35.67,"hrank":23.87,"imp":15.521,"train_time":0.62,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+151\/-150","imputed":false},{"position":35,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1407.0,"score":0.115,"rank":36.5,"hrank":29.68,"imp":16.508,"train_time":442.2,"predict_time":0.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+139\/-105","imputed":false},{"position":36,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1407.0,"score":0.127,"rank":36.51,"hrank":28.3,"imp":15.92,"train_time":1.56,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-128","imputed":false},{"position":37,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1393.0,"score":0.139,"rank":37.49,"hrank":29.68,"imp":17.32,"train_time":709.86,"predict_time":0.19,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+142\/-146","imputed":false},{"position":38,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1369.0,"score":0.054,"rank":39.18,"hrank":35.44,"imp":16.81,"train_time":6.98,"predict_time":0.233,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-89","imputed":false},{"position":39,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1363.0,"score":0.069,"rank":39.58,"hrank":31.87,"imp":18.84,"train_time":84.36,"predict_time":0.919,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-125","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1359.0,"score":0.06,"rank":39.86,"hrank":33.92,"imp":15.588,"train_time":1169.32,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-79","imputed":false},{"position":41,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1356.0,"score":0.095,"rank":40.08,"hrank":31.95,"imp":17.692,"train_time":389.62,"predict_time":2.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-164","imputed":false},{"position":42,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1354.0,"score":0.071,"rank":40.21,"hrank":35.46,"imp":18.314,"train_time":1.44,"predict_time":0.007,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-108","imputed":false},{"position":43,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1335.0,"score":0.049,"rank":41.54,"hrank":28.22,"imp":17.081,"train_time":4805.68,"predict_time":43.211,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-117","imputed":false},{"position":44,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1334.0,"score":0.048,"rank":41.57,"hrank":33.67,"imp":16.69,"train_time":52.45,"predict_time":0.219,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-104","imputed":false},{"position":45,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1315.0,"score":0.058,"rank":42.86,"hrank":38.29,"imp":19.676,"train_time":68.97,"predict_time":8.043,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-105","imputed":false},{"position":46,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1312.0,"score":0.033,"rank":43.04,"hrank":39.84,"imp":17.073,"train_time":0.28,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-99","imputed":false},{"position":47,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1302.0,"score":0.168,"rank":43.71,"hrank":7.78,"imp":20.306,"train_time":5.06,"predict_time":2.932,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+231\/-279","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1300.0,"score":0.049,"rank":43.87,"hrank":35.85,"imp":17.509,"train_time":52.45,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-111","imputed":false},{"position":49,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1292.0,"score":0.012,"rank":44.38,"hrank":41.41,"imp":17.813,"train_time":10.5,"predict_time":0.606,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-75","imputed":false},{"position":50,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1277.0,"score":0.018,"rank":45.38,"hrank":42.99,"imp":16.935,"train_time":50.19,"predict_time":0.154,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-94","imputed":false},{"position":51,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1276.0,"score":0.011,"rank":45.48,"hrank":42.15,"imp":19.434,"train_time":14.85,"predict_time":13.554,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-105","imputed":false},{"position":52,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1251.0,"score":0.017,"rank":47.08,"hrank":37.59,"imp":20.743,"train_time":46.9,"predict_time":45.623,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-146","imputed":false},{"position":53,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1227.0,"score":0.026,"rank":48.6,"hrank":42.19,"imp":20.621,"train_time":389.62,"predict_time":0.112,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-148","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1226.0,"score":0.0,"rank":48.66,"hrank":46.4,"imp":18.338,"train_time":50.19,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-94","imputed":false},{"position":55,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1203.0,"score":0.0,"rank":50.08,"hrank":48.37,"imp":19.659,"train_time":1.87,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-73","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1159.0,"score":0.029,"rank":52.64,"hrank":45.84,"imp":24.202,"train_time":1.61,"predict_time":0.112,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-214","imputed":false},{"position":57,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1103.0,"score":0.0,"rank":55.76,"hrank":54.08,"imp":26.789,"train_time":2951.8,"predict_time":1.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-104","imputed":false},{"position":58,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1099.0,"score":0.008,"rank":55.96,"hrank":50.37,"imp":28.713,"train_time":43.46,"predict_time":0.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-242","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1094.0,"score":0.0,"rank":56.22,"hrank":54.6,"imp":26.863,"train_time":2951.8,"predict_time":4.991,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-104","imputed":false},{"position":60,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1073.0,"score":0.009,"rank":57.31,"hrank":48.05,"imp":26.652,"train_time":331.68,"predict_time":44.947,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+101\/-218","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1056.0,"score":0.004,"rank":58.16,"hrank":53.45,"imp":29.813,"train_time":43.46,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+157\/-250","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1032.0,"score":0.0,"rank":59.3,"hrank":51.72,"imp":27.448,"train_time":22.82,"predict_time":0.606,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+125\/-201","imputed":false},{"position":63,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":60.76,"hrank":59.36,"imp":25.121,"train_time":0.12,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-180","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":994.0,"score":0.0,"rank":61.03,"hrank":56.14,"imp":28.773,"train_time":22.82,"predict_time":0.253,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-228","imputed":false},{"position":65,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":968.0,"score":0.0,"rank":62.12,"hrank":60.83,"imp":27.361,"train_time":17.17,"predict_time":6.883,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+99\/-178","imputed":false},{"position":66,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":968.0,"score":0.0,"rank":62.14,"hrank":60.79,"imp":27.159,"train_time":0.34,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-217","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":950.0,"score":0.0,"rank":62.88,"hrank":60.63,"imp":31.351,"train_time":0.2,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-229","imputed":false},{"position":68,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":935.0,"score":0.0,"rank":63.46,"hrank":62.64,"imp":32.487,"train_time":86.28,"predict_time":1.898,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-106","imputed":false},{"position":69,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":932.0,"score":0.021,"rank":63.54,"hrank":56.64,"imp":33.606,"train_time":2.8,"predict_time":0.224,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+189\/-424","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":880.0,"score":0.0,"rank":65.42,"hrank":64.37,"imp":34.878,"train_time":0.2,"predict_time":0.033,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-164","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":852.0,"score":0.0,"rank":66.31,"hrank":65.97,"imp":33.83,"train_time":331.68,"predict_time":1.626,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-124","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":804.0,"score":0.0,"rank":67.69,"hrank":66.53,"imp":36.946,"train_time":1.64,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-282","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":512.0,"score":0.0,"rank":72.64,"hrank":72.61,"imp":50.452,"train_time":0.12,"predict_time":0.101,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-308","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_medium/pareto_front_explorer.html
index e3fa8097feb5778d8bc801cfe57e1b9277f31e3c..887e971c4f5fe583c4792268ca42ad821b522149 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.9189171706,"imp":18.840007406,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0190099757,"imp":9.1783534259,"elo":1704.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0199434097,"imp":9.5878875099,"elo":1697.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1678973097,"imp":9.315864629,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0229625096,"imp":12.843489519,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0260723508,"imp":11.8083619918,"elo":1609.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3141545533,"imp":11.2140890816,"elo":1667.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0069074234,"imp":18.313628056,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0091418408,"imp":16.5078148968,"elo":1406.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0944568378,"imp":15.5170264457,"elo":1474.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0569778809,"imp":27.1590909048,"elo":967.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0150080481,"imp":18.3381917932,"elo":1225.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1535912971,"imp":16.9352042076,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1123933797,"imp":24.2020361148,"elo":1159.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1117599745,"imp":20.620988534,"elo":1226.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.1135199956,"imp":17.6917823414,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.033160957,"imp":17.0730560129,"elo":1312.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0872426084,"imp":13.2588415734,"elo":1562.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7582948004,"imp":11.7004556275,"elo":1668.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.9321947194,"imp":20.305650329,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.6055455764,"imp":17.8133281766,"elo":1292.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5401190643,"imp":13.7174094198,"elo":1518.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":14.2476482136,"imp":11.3829611751,"elo":1627.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8827204019,"imp":27.3609275393,"elo":968.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0226836019,"imp":36.946006529,"elo":804.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2526043838,"imp":28.7726864862,"elo":993.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6062271231,"imp":27.4482786558,"elo":1032.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0133914852,"imp":25.1205416265,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0207808341,"imp":17.5089228838,"elo":1300.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2185753382,"imp":16.6904413475,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.232511973,"imp":16.8103385739,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0834429515,"imp":12.215693912,"elo":1571.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.762731528,"imp":10.9499682998,"elo":1652.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8199508222,"imp":12.7365093479,"elo":1552.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.9240530403,"imp":11.8993811355,"elo":1598.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":13.789522936,"imp":11.1745533202,"elo":1640.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.5537416022,"imp":19.4340799677,"elo":1275.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.622831341,"imp":20.7430151958,"elo":1250.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.2108859387,"imp":17.0805751385,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":461.7902853066,"imp":13.722448173,"elo":1489.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2302397985,"imp":14.3547062199,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":18.7591592077,"imp":3.7223594114,"elo":2005.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0330391224,"imp":34.8784648588,"elo":879.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.1457473701,"imp":13.434078273,"elo":1531.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1703078893,"imp":8.7168375107,"elo":1751.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0548347304,"imp":15.9198153896,"elo":1406.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0550494549,"imp":12.6721520809,"elo":1564.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.554086605,"imp":11.7928607556,"elo":1636.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7011909333,"imp":9.7349142793,"elo":1760.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2888892556,"imp":7.8760094387,"elo":1849.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.898348797,"imp":32.4874771321,"elo":934.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6889181125,"imp":26.7888300543,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9908559667,"imp":26.8629927897,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1693477495,"imp":15.5212999312,"elo":1418.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0462763773,"imp":19.6594845905,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0534965201,"imp":15.5879677352,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4424042458,"imp":13.577761493,"elo":1488.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.052450313,"imp":14.3165551195,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0561708961,"imp":12.1622111016,"elo":1605.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3981946384,"imp":11.387418587,"elo":1673.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":8.0432165428,"imp":19.6755609009,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.6611582164,"imp":13.586548919,"elo":1521.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":72.9817040915,"imp":10.5294798486,"elo":1704.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.2236113698,"imp":33.606488769,"elo":932.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1897669991,"imp":17.3202372496,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9293738561,"imp":15.5356888427,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.9189171706,"x_train":84.3646061894,"imp":18.840007406,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0190099757,"x_train":1.4934673843,"imp":9.1783534259,"elo":1704.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0199434097,"x_train":378.4994428039,"imp":9.5878875099,"elo":1697.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1678973097,"x_train":378.4994428039,"imp":9.315864629,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0229625096,"x_train":0.3843823491,"imp":12.843489519,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0260723508,"x_train":90.1654472213,"imp":11.8083619918,"elo":1609.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3141545533,"x_train":90.1654472213,"imp":11.2140890816,"elo":1667.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0069074234,"x_train":1.4432254313,"imp":18.313628056,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0091418408,"x_train":442.2024676551,"imp":16.5078148968,"elo":1406.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0944568378,"x_train":442.2024676551,"imp":15.5170264457,"elo":1474.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0569778809,"x_train":0.3394264702,"imp":27.1590909048,"elo":967.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0150080481,"x_train":50.1933853639,"imp":18.3381917932,"elo":1225.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1535912971,"x_train":50.1933853639,"imp":16.9352042076,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1123933797,"x_train":1.6121148448,"imp":24.2020361148,"elo":1159.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1117599745,"x_train":389.6192568267,"imp":20.620988534,"elo":1226.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.1135199956,"x_train":389.6192568267,"imp":17.6917823414,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.033160957,"x_train":0.2814937163,"imp":17.0730560129,"elo":1312.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0872426084,"x_train":93.2136338393,"imp":13.2588415734,"elo":1562.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7582948004,"x_train":93.2136338393,"imp":11.7004556275,"elo":1668.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.9321947194,"x_train":5.058611156,"imp":20.305650329,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.6055455764,"x_train":10.5029291491,"imp":17.8133281766,"elo":1292.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5401190643,"x_train":2338.1908652586,"imp":13.7174094198,"elo":1518.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":14.2476482136,"x_train":2338.1908652586,"imp":11.3829611751,"elo":1627.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8827204019,"x_train":17.1713211146,"imp":27.3609275393,"elo":968.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0226836019,"x_train":1.6442071407,"imp":36.946006529,"elo":804.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2526043838,"x_train":22.8161312471,"imp":28.7726864862,"elo":993.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6062271231,"x_train":22.8161312471,"imp":27.4482786558,"elo":1032.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0133914852,"x_train":0.1211565882,"imp":25.1205416265,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0207808341,"x_train":52.4497680024,"imp":17.5089228838,"elo":1300.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2185753382,"x_train":52.4497680024,"imp":16.6904413475,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.232511973,"x_train":6.9801966051,"imp":16.8103385739,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0834429515,"x_train":1624.5504043652,"imp":12.215693912,"elo":1571.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.762731528,"x_train":1624.5504043652,"imp":10.9499682998,"elo":1652.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8199508222,"x_train":2.2973997315,"imp":12.7365093479,"elo":1552.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.9240530403,"x_train":783.2895068512,"imp":11.8993811355,"elo":1598.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":13.789522936,"x_train":783.2895068512,"imp":11.1745533202,"elo":1640.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.5537416022,"x_train":14.8539049818,"imp":19.4340799677,"elo":1275.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.622831341,"x_train":46.8983759891,"imp":20.7430151958,"elo":1250.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.2108859387,"x_train":4805.6782161313,"imp":17.0805751385,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":461.7902853066,"x_train":4805.6782161313,"imp":13.722448173,"elo":1489.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2302397985,"x_train":0.7869111961,"imp":14.3547062199,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":18.7591592077,"x_train":38.8556954397,"imp":3.7223594114,"elo":2005.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0330391224,"x_train":0.1957996411,"imp":34.8784648588,"elo":879.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.1457473701,"x_train":7.5078983053,"imp":13.434078273,"elo":1531.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1703078893,"x_train":0.6258098026,"imp":8.7168375107,"elo":1751.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0548347304,"x_train":1.5586488241,"imp":15.9198153896,"elo":1406.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0550494549,"x_train":860.2847801702,"imp":12.6721520809,"elo":1564.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.554086605,"x_train":860.2847801702,"imp":11.7928607556,"elo":1636.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7011909333,"x_train":2.4084538179,"imp":9.7349142793,"elo":1760.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2888892556,"x_train":1.0968929682,"imp":7.8760094387,"elo":1849.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.898348797,"x_train":86.2757906839,"imp":32.4874771321,"elo":934.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6889181125,"x_train":2951.7955583902,"imp":26.7888300543,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9908559667,"x_train":2951.7955583902,"imp":26.8629927897,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1693477495,"x_train":0.6181258773,"imp":15.5212999312,"elo":1418.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0462763773,"x_train":1.8659541839,"imp":19.6594845905,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0534965201,"x_train":1169.3150011596,"imp":15.5879677352,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4424042458,"x_train":1169.3150011596,"imp":13.577761493,"elo":1488.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.052450313,"x_train":0.4700674902,"imp":14.3165551195,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0561708961,"x_train":207.6491745508,"imp":12.1622111016,"elo":1605.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3981946384,"x_train":207.6491745508,"imp":11.387418587,"elo":1673.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":8.0432165428,"x_train":68.9669811874,"imp":19.6755609009,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.6611582164,"x_train":1607.7353976095,"imp":13.586548919,"elo":1521.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":72.9817040915,"x_train":1607.7353976095,"imp":10.5294798486,"elo":1704.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.2236113698,"x_train":2.8035263333,"imp":33.606488769,"elo":932.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1897669991,"x_train":709.8550005699,"imp":17.3202372496,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9293738561,"x_train":709.8550005699,"imp":15.5356888427,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
index d1d5d217dfe7dc824471cbec21e6c6cbd5badb79..21cf4c179a770d5f5962e245d9869038e3c30ef2 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":84.3646061894,"imp":18.8599955331,"elo":1386.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.4934673843,"imp":9.1976006657,"elo":1740.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":4.0498969951,"imp":9.3187381366,"elo":1738.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.8156310096,"imp":9.3725589058,"elo":1748.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":23.9442866701,"imp":9.3921018347,"elo":1750.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":52.2059341627,"imp":9.2916172381,"elo":1760.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":97.6448146128,"imp":9.2566381211,"elo":1764.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":190.15396393,"imp":9.2967413641,"elo":1758.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":281.6941410441,"imp":9.3113324842,"elo":1756.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":378.4994428039,"imp":9.3273441341,"elo":1753.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.3843823491,"imp":12.8635534115,"elo":1590.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.8231266994,"imp":12.2327635752,"elo":1643.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.0711841529,"imp":11.8377408404,"elo":1670.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.198580454,"imp":11.6213761336,"elo":1685.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.0736254563,"imp":11.5012152174,"elo":1694.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.535781385,"imp":11.4021767141,"elo":1699.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":44.9181007897,"imp":11.2853772779,"elo":1707.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":67.208700505,"imp":11.2408277916,"elo":1709.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":90.1654472213,"imp":11.2307808149,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":1.4432254313,"imp":18.3289887858,"elo":1367.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":4.2347903447,"imp":17.4720804235,"elo":1401.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":10.3173350008,"imp":16.7542431694,"elo":1434.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":21.8514222981,"imp":16.4427611222,"elo":1448.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":55.5931037878,"imp":16.1033145209,"elo":1468.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":111.097193586,"imp":15.9162222115,"elo":1478.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":219.5483355792,"imp":15.709674653,"elo":1491.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":330.9116540058,"imp":15.6383543743,"elo":1495.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":442.2024676551,"imp":15.52589414,"elo":1503.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3394264702,"imp":27.1741880295,"elo":963.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.561280236,"imp":24.2835226498,"elo":1033.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.3509100065,"imp":19.8198892022,"elo":1165.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.5531799736,"imp":18.2046872025,"elo":1230.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.2677435481,"imp":17.2453969381,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.4252027007,"imp":17.0238417989,"elo":1284.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.8317026638,"imp":16.8635575395,"elo":1295.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.332646039,"imp":16.8714868407,"elo":1296.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":50.1933853639,"imp":16.9523835395,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.6121148448,"imp":24.2159078135,"elo":1177.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.4214998387,"imp":22.7745276103,"elo":1209.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":9.5320935926,"imp":20.9902035951,"elo":1271.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":19.7206332887,"imp":19.9998660631,"elo":1301.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":47.120631421,"imp":19.1321650838,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":95.0331726255,"imp":18.6307722458,"elo":1352.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":192.9295597629,"imp":18.1898274964,"elo":1366.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":290.1521255355,"imp":17.891970254,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":389.6192568267,"imp":17.7004215583,"elo":1387.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.2814937163,"imp":17.0899008251,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.8419891145,"imp":16.2689403096,"elo":1399.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":2.1642156548,"imp":14.9266707853,"elo":1504.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":4.3828192424,"imp":13.8517848473,"elo":1586.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":11.3294435796,"imp":12.6195613378,"elo":1649.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":23.1362359393,"imp":12.0578006895,"elo":1690.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":46.1918681929,"imp":11.8306177522,"elo":1704.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":68.9587529747,"imp":11.7603011738,"elo":1706.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":93.2136338393,"imp":11.7281550364,"elo":1710.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":5.058611156,"imp":19.8907338623,"elo":1337.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":10.5029291491,"imp":17.8299401868,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.3830456847,"imp":15.8929011855,"elo":1406.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":65.7796741249,"imp":13.5293238,"elo":1549.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":124.0730628064,"imp":12.7321692308,"elo":1595.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":283.3505795288,"imp":11.9230807327,"elo":1644.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":583.6843020141,"imp":11.5856976387,"elo":1664.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1153.7001044299,"imp":11.4878041124,"elo":1671.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1719.5734803795,"imp":11.4420520926,"elo":1671.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2338.1908652586,"imp":11.4152796867,"elo":1670.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":17.1713211146,"imp":27.4282235907,"elo":974.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.6442071407,"imp":37.3438261467,"elo":805.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.5769737451,"imp":34.813557491,"elo":882.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":19.1886688806,"imp":27.6055384276,"elo":1038.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":22.8161312471,"imp":27.4540417274,"elo":1047.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1211565882,"imp":25.1437303937,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.39306177,"imp":20.0876024453,"elo":1163.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.261863005,"imp":18.1118560726,"elo":1264.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":2.6044216897,"imp":17.448305273,"elo":1301.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":6.4126782162,"imp":16.9889325644,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":12.9417195679,"imp":16.8332836861,"elo":1344.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":25.9261667663,"imp":16.7203484605,"elo":1354.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":39.086650906,"imp":16.6653861195,"elo":1359.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":52.4497680024,"imp":16.7072446847,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.9801966051,"imp":16.8241005098,"elo":1387.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.4217554801,"imp":14.7107487455,"elo":1518.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.0914867245,"imp":13.5019228572,"elo":1585.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":79.6483382915,"imp":12.714245834,"elo":1619.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":203.6966152959,"imp":11.9655799475,"elo":1654.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":410.5193453014,"imp":11.6538509223,"elo":1666.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":808.8154712049,"imp":11.2152910283,"elo":1681.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1213.5228253628,"imp":11.0331396261,"elo":1691.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1624.5504043652,"imp":10.9669183657,"elo":1697.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2.2973997315,"imp":12.757790933,"elo":1576.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.7558525217,"imp":12.7040141437,"elo":1585.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":18.4525573231,"imp":12.5678686593,"elo":1599.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":36.2094345784,"imp":12.5003093797,"elo":1607.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":88.4354783372,"imp":12.1553628698,"elo":1640.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":183.3566697473,"imp":11.7002365934,"elo":1661.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":377.7921394838,"imp":11.4270632498,"elo":1669.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":572.0185730315,"imp":11.2797380522,"elo":1669.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":783.2895068512,"imp":11.188360367,"elo":1668.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":14.8539049818,"imp":19.4459273431,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.8983759891,"imp":20.7566063243,"elo":1275.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":69.8822600854,"imp":19.1332286663,"elo":1314.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":127.4067152197,"imp":17.5591056151,"elo":1372.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":229.7746727932,"imp":16.995992756,"elo":1393.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":589.1070615002,"imp":15.6827832788,"elo":1452.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1193.0508698741,"imp":14.9633943872,"elo":1484.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2388.3823364922,"imp":14.4617778749,"elo":1503.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3593.4863060271,"imp":14.0010964181,"elo":1520.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4805.6782161313,"imp":13.7360803063,"elo":1526.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.7869111961,"imp":14.371186543,"elo":1545.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.8556954397,"imp":3.7240503537,"elo":2039.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1957996411,"imp":34.8895165819,"elo":878.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":7.5078983053,"imp":13.454211295,"elo":1564.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6258098026,"imp":8.7394017522,"elo":1785.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.5586488241,"imp":15.9377258692,"elo":1428.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":5.192079831,"imp":14.9598444176,"elo":1471.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":19.2797854853,"imp":13.7607602335,"elo":1533.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":41.1266447543,"imp":13.0968145146,"elo":1579.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":106.1594477641,"imp":12.4863504012,"elo":1615.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":213.4318571252,"imp":12.1925610385,"elo":1638.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":422.7331976766,"imp":11.9493041925,"elo":1654.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":638.9519356459,"imp":11.8594998702,"elo":1662.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":860.2847801702,"imp":11.8049633358,"elo":1670.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.4084538179,"imp":9.7560536872,"elo":1801.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.0968929682,"imp":7.8824971032,"elo":1878.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":86.2757906839,"imp":32.4967224715,"elo":932.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":96.7574708184,"imp":32.41479258,"elo":947.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":132.620297349,"imp":32.2594861504,"elo":954.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":182.3615021409,"imp":31.786012596,"elo":975.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":379.7267185426,"imp":30.7058200242,"elo":1005.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":737.6739956698,"imp":29.3722505198,"elo":1037.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1465.1527881613,"imp":28.4744557852,"elo":1065.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2179.2889039095,"imp":27.6256143537,"elo":1092.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2951.7955583902,"imp":26.8746550395,"elo":1123.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.6181258773,"imp":15.5418068987,"elo":1445.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":1.8659541839,"imp":19.6714412852,"elo":1215.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.8422781275,"imp":18.2754375515,"elo":1279.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":28.1225860036,"imp":16.8936521045,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":62.6333146165,"imp":16.0329912533,"elo":1394.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":153.2940248569,"imp":15.1322927453,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":287.4960670101,"imp":14.5504300077,"elo":1473.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":565.1252028376,"imp":14.1383335222,"elo":1498.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":871.368400213,"imp":13.8720944137,"elo":1512.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1169.3150011596,"imp":13.597536455,"elo":1523.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4700674902,"imp":14.3359708531,"elo":1484.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.0561352068,"imp":12.8895219142,"elo":1600.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":5.7242090848,"imp":12.2710877404,"elo":1655.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":9.9945517672,"imp":11.8250868461,"elo":1683.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":26.428780062,"imp":11.5620514204,"elo":1698.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":52.4146823069,"imp":11.4682434169,"elo":1705.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":103.8233704142,"imp":11.4119948439,"elo":1710.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":154.5284843647,"imp":11.4080903645,"elo":1713.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":207.6491745508,"imp":11.4089279264,"elo":1716.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":68.9669811874,"imp":19.6910335019,"elo":1337.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":130.6527630186,"imp":17.5680857216,"elo":1390.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":316.404564055,"imp":14.9245895934,"elo":1499.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":624.6899718112,"imp":12.8882163313,"elo":1603.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1547.1736962358,"imp":10.591051261,"elo":1743.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1607.7353976095,"imp":10.5488846537,"elo":1749.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.8035263333,"imp":33.6178632238,"elo":931.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":6.8864266433,"imp":27.0655054285,"elo":1050.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.5478119749,"imp":19.7955614958,"elo":1258.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":35.2197842651,"imp":16.9393451699,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.4231720748,"imp":15.9543591258,"elo":1449.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":176.2334816176,"imp":15.5960824642,"elo":1465.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":353.6101338707,"imp":15.4855978336,"elo":1464.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":529.672353661,"imp":15.4670973384,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":709.8550005699,"imp":15.5056797319,"elo":1455.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.9189171706,"x_train":84.3646061894,"imp":18.8599955331,"elo":1386.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0190099757,"x_train":1.4934673843,"imp":9.1976006657,"elo":1740.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.038142376,"x_train":4.0498969951,"imp":9.3187381366,"elo":1738.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0612481453,"x_train":12.8156310096,"imp":9.3725589058,"elo":1748.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0782544909,"x_train":23.9442866701,"imp":9.3921018347,"elo":1750.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.095374063,"x_train":52.2059341627,"imp":9.2916172381,"elo":1760.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1151706511,"x_train":97.6448146128,"imp":9.2566381211,"elo":1764.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1432773038,"x_train":190.15396393,"imp":9.2967413641,"elo":1758.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1564459455,"x_train":281.6941410441,"imp":9.3113324842,"elo":1756.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1678973097,"x_train":378.4994428039,"imp":9.3273441341,"elo":1753.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0229625096,"x_train":0.3843823491,"imp":12.8635534115,"elo":1590.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0474733333,"x_train":0.8231266994,"imp":12.2327635752,"elo":1643.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1047084686,"x_train":2.0711841529,"imp":11.8377408404,"elo":1670.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1608995496,"x_train":4.198580454,"imp":11.6213761336,"elo":1685.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2235785875,"x_train":11.0736254563,"imp":11.5012152174,"elo":1694.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.268731958,"x_train":22.535781385,"imp":11.4021767141,"elo":1699.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2953259759,"x_train":44.9181007897,"imp":11.2853772779,"elo":1707.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3063723891,"x_train":67.208700505,"imp":11.2408277916,"elo":1709.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3135203066,"x_train":90.1654472213,"imp":11.2307808149,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0069074234,"x_train":1.4432254313,"imp":18.3289887858,"elo":1367.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0179091186,"x_train":4.2347903447,"imp":17.4720804235,"elo":1401.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0310818868,"x_train":10.3173350008,"imp":16.7542431694,"elo":1434.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0444548419,"x_train":21.8514222981,"imp":16.4427611222,"elo":1448.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0642533403,"x_train":55.5931037878,"imp":16.1033145209,"elo":1468.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0756378371,"x_train":111.097193586,"imp":15.9162222115,"elo":1478.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.08260308,"x_train":219.5483355792,"imp":15.709674653,"elo":1491.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.08968946,"x_train":330.9116540058,"imp":15.6383543743,"elo":1495.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0948963283,"x_train":442.2024676551,"imp":15.52589414,"elo":1503.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0569778809,"x_train":0.3394264702,"imp":27.1741880295,"elo":963.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0649760559,"x_train":0.561280236,"imp":24.2835226498,"elo":1033.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0823683729,"x_train":1.3509100065,"imp":19.8198892022,"elo":1165.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0925044765,"x_train":2.5531799736,"imp":18.2046872025,"elo":1230.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1048375565,"x_train":6.2677435481,"imp":17.2453969381,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1198241829,"x_train":12.4252027007,"imp":17.0238417989,"elo":1284.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1381552995,"x_train":24.8317026638,"imp":16.8635575395,"elo":1295.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1436546335,"x_train":37.332646039,"imp":16.8714868407,"elo":1296.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1517370302,"x_train":50.1933853639,"imp":16.9523835395,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1123933797,"x_train":1.6121148448,"imp":24.2159078135,"elo":1177.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2234013195,"x_train":3.4214998387,"imp":22.7745276103,"elo":1209.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5317849911,"x_train":9.5320935926,"imp":20.9902035951,"elo":1271.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8880009049,"x_train":19.7206332887,"imp":19.9998660631,"elo":1301.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2374143618,"x_train":47.120631421,"imp":19.1321650838,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4963820182,"x_train":95.0331726255,"imp":18.6307722458,"elo":1352.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.871926073,"x_train":192.9295597629,"imp":18.1898274964,"elo":1366.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.9971010285,"x_train":290.1521255355,"imp":17.891970254,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0921384936,"x_train":389.6192568267,"imp":17.7004215583,"elo":1387.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.033160957,"x_train":0.2814937163,"imp":17.0899008251,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.3076291887,"x_train":0.8419891145,"imp":16.2689403096,"elo":1399.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5603369831,"x_train":2.1642156548,"imp":14.9266707853,"elo":1504.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8536127137,"x_train":4.3828192424,"imp":13.8517848473,"elo":1586.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.935960933,"x_train":11.3294435796,"imp":12.6195613378,"elo":1649.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9476843576,"x_train":23.1362359393,"imp":12.0578006895,"elo":1690.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8576552911,"x_train":46.1918681929,"imp":11.8306177522,"elo":1704.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7970201041,"x_train":68.9587529747,"imp":11.7603011738,"elo":1706.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.756028165,"x_train":93.2136338393,"imp":11.7281550364,"elo":1710.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":2.9321947194,"x_train":5.058611156,"imp":19.8907338623,"elo":1337.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.6055455764,"x_train":10.5029291491,"imp":17.8299401868,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.1645670464,"x_train":25.3830456847,"imp":15.8929011855,"elo":1406.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.5989848543,"x_train":65.7796741249,"imp":13.5293238,"elo":1549.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1554075804,"x_train":124.0730628064,"imp":12.7321692308,"elo":1595.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.8948147244,"x_train":283.3505795288,"imp":11.9230807327,"elo":1644.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.4555428919,"x_train":583.6843020141,"imp":11.5856976387,"elo":1664.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":11.9967506884,"x_train":1153.7001044299,"imp":11.4878041124,"elo":1671.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":13.3726529631,"x_train":1719.5734803795,"imp":11.4420520926,"elo":1671.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":14.1602029953,"x_train":2338.1908652586,"imp":11.4152796867,"elo":1670.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8827204019,"x_train":17.1713211146,"imp":27.4282235907,"elo":974.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0226836019,"x_train":1.6442071407,"imp":37.3438261467,"elo":805.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1912576397,"x_train":6.5769737451,"imp":34.813557491,"elo":882.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5361050995,"x_train":19.1886688806,"imp":27.6055384276,"elo":1038.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6062271231,"x_train":22.8161312471,"imp":27.4540417274,"elo":1047.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0133914852,"x_train":0.1211565882,"imp":25.1437303937,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.040619544,"x_train":0.39306177,"imp":20.0876024453,"elo":1163.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0822556302,"x_train":1.261863005,"imp":18.1118560726,"elo":1264.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1187605168,"x_train":2.6044216897,"imp":17.448305273,"elo":1301.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1439903713,"x_train":6.4126782162,"imp":16.9889325644,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1661672354,"x_train":12.9417195679,"imp":16.8332836861,"elo":1344.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1936671381,"x_train":25.9261667663,"imp":16.7203484605,"elo":1354.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2030888401,"x_train":39.086650906,"imp":16.6653861195,"elo":1359.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.215523944,"x_train":52.4497680024,"imp":16.7072446847,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.232511973,"x_train":6.9801966051,"imp":16.8241005098,"elo":1387.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.366818368,"x_train":15.4217554801,"imp":14.7107487455,"elo":1518.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5883942127,"x_train":39.0914867245,"imp":13.5019228572,"elo":1585.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.8292026744,"x_train":79.6483382915,"imp":12.714245834,"elo":1619.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.0621411706,"x_train":203.6966152959,"imp":11.9655799475,"elo":1654.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.2831539961,"x_train":410.5193453014,"imp":11.6538509223,"elo":1666.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5130210219,"x_train":808.8154712049,"imp":11.2152910283,"elo":1681.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6874000841,"x_train":1213.5228253628,"imp":11.0331396261,"elo":1691.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7535445217,"x_train":1624.5504043652,"imp":10.9669183657,"elo":1697.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8199508222,"x_train":2.2973997315,"imp":12.757790933,"elo":1576.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7781160341,"x_train":5.7558525217,"imp":12.7040141437,"elo":1585.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.528542093,"x_train":18.4525573231,"imp":12.5678686593,"elo":1599.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.5801414596,"x_train":36.2094345784,"imp":12.5003093797,"elo":1607.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.0669182071,"x_train":88.4354783372,"imp":12.1553628698,"elo":1640.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.3558901149,"x_train":183.3566697473,"imp":11.7002365934,"elo":1661.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.0333907004,"x_train":377.7921394838,"imp":11.4270632498,"elo":1669.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.9376757094,"x_train":572.0185730315,"imp":11.2797380522,"elo":1669.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":13.6073656299,"x_train":783.2895068512,"imp":11.188360367,"elo":1668.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.5537416022,"x_train":14.8539049818,"imp":19.4459273431,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":45.622831341,"x_train":46.8983759891,"imp":20.7566063243,"elo":1275.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":66.3156018301,"x_train":69.8822600854,"imp":19.1332286663,"elo":1314.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":98.021147049,"x_train":127.4067152197,"imp":17.5591056151,"elo":1372.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":134.082270619,"x_train":229.7746727932,"imp":16.995992756,"elo":1393.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":215.0366892607,"x_train":589.1070615002,"imp":15.6827832788,"elo":1452.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":284.8856925256,"x_train":1193.0508698741,"imp":14.9633943872,"elo":1484.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":380.3864389015,"x_train":2388.3823364922,"imp":14.4617778749,"elo":1503.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":441.6863563071,"x_train":3593.4863060271,"imp":14.0010964181,"elo":1520.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":464.2805045187,"x_train":4805.6782161313,"imp":13.7360803063,"elo":1526.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2302397985,"x_train":0.7869111961,"imp":14.371186543,"elo":1545.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":18.7591592077,"x_train":38.8556954397,"imp":3.7240503537,"elo":2039.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0330391224,"x_train":0.1957996411,"imp":34.8895165819,"elo":878.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":3.1457473701,"x_train":7.5078983053,"imp":13.454211295,"elo":1564.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1703078893,"x_train":0.6258098026,"imp":8.7394017522,"elo":1785.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0548347304,"x_train":1.5586488241,"imp":15.9377258692,"elo":1428.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1165748723,"x_train":5.192079831,"imp":14.9598444176,"elo":1471.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2008241117,"x_train":19.2797854853,"imp":13.7607602335,"elo":1533.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2729097923,"x_train":41.1266447543,"imp":13.0968145146,"elo":1579.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3551141856,"x_train":106.1594477641,"imp":12.4863504012,"elo":1615.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4346764878,"x_train":213.4318571252,"imp":12.1925610385,"elo":1638.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4982388933,"x_train":422.7331976766,"imp":11.9493041925,"elo":1654.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.5348355254,"x_train":638.9519356459,"imp":11.8594998702,"elo":1662.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5548771281,"x_train":860.2847801702,"imp":11.8049633358,"elo":1670.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7011909333,"x_train":2.4084538179,"imp":9.7560536872,"elo":1801.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2888892556,"x_train":1.0968929682,"imp":7.8824971032,"elo":1878.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.898348797,"x_train":86.2757906839,"imp":32.4967224715,"elo":932.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.938220505,"x_train":96.7574708184,"imp":32.41479258,"elo":947.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0360121046,"x_train":132.620297349,"imp":32.2594861504,"elo":954.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.3732929292,"x_train":182.3615021409,"imp":31.786012596,"elo":975.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9911165015,"x_train":379.7267185426,"imp":30.7058200242,"elo":1005.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4182120307,"x_train":737.6739956698,"imp":29.3722505198,"elo":1037.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2639960398,"x_train":1465.1527881613,"imp":28.4744557852,"elo":1065.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9050217868,"x_train":2179.2889039095,"imp":27.6256143537,"elo":1092.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9836310881,"x_train":2951.7955583902,"imp":26.8746550395,"elo":1123.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1693477495,"x_train":0.6181258773,"imp":15.5418068987,"elo":1445.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0462763773,"x_train":1.8659541839,"imp":19.6714412852,"elo":1215.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0986462001,"x_train":6.8422781275,"imp":18.2754375515,"elo":1279.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2529547278,"x_train":28.1225860036,"imp":16.8936521045,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4092168594,"x_train":62.6333146165,"imp":16.0329912533,"elo":1394.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6013598742,"x_train":153.2940248569,"imp":15.1322927453,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8316842129,"x_train":287.4960670101,"imp":14.5504300077,"elo":1473.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.09015958,"x_train":565.1252028376,"imp":14.1383335222,"elo":1498.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2700433543,"x_train":871.368400213,"imp":13.8720944137,"elo":1512.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4406412634,"x_train":1169.3150011596,"imp":13.597536455,"elo":1523.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.052450313,"x_train":0.4700674902,"imp":14.3359708531,"elo":1484.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1462358449,"x_train":2.0561352068,"imp":12.8895219142,"elo":1600.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2538924294,"x_train":5.7242090848,"imp":12.2710877404,"elo":1655.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2894544358,"x_train":9.9945517672,"imp":11.8250868461,"elo":1683.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3450335972,"x_train":26.428780062,"imp":11.5620514204,"elo":1698.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3624432873,"x_train":52.4146823069,"imp":11.4682434169,"elo":1705.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3639681589,"x_train":103.8233704142,"imp":11.4119948439,"elo":1710.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3851024669,"x_train":154.5284843647,"imp":11.4080903645,"elo":1713.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.398537392,"x_train":207.6491745508,"imp":11.4089279264,"elo":1716.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":8.0432165428,"x_train":68.9669811874,"imp":19.6910335019,"elo":1337.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":15.6237886223,"x_train":130.6527630186,"imp":17.5680857216,"elo":1390.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":29.3242995716,"x_train":316.404564055,"imp":14.9245895934,"elo":1499.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":45.5811393322,"x_train":624.6899718112,"imp":12.8882163313,"elo":1603.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":71.7349528943,"x_train":1547.1736962358,"imp":10.591051261,"elo":1743.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":73.3251138416,"x_train":1607.7353976095,"imp":10.5488846537,"elo":1749.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.2236113698,"x_train":2.8035263333,"imp":33.6178632238,"elo":931.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.390344083,"x_train":6.8864266433,"imp":27.0655054285,"elo":1050.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.6801837529,"x_train":17.5478119749,"imp":19.7955614958,"elo":1258.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.9680498121,"x_train":35.2197842651,"imp":16.9393451699,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.4009052838,"x_train":88.4231720748,"imp":15.9543591258,"elo":1449.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.8435618428,"x_train":176.2334816176,"imp":15.5960824642,"elo":1465.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.3657351437,"x_train":353.6101338707,"imp":15.4855978336,"elo":1464.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.6852473948,"x_train":529.672353661,"imp":15.4670973384,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.884834451,"x_train":709.8550005699,"imp":15.5056797319,"elo":1455.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..3f1c8ae7f5add45cbeadd00662cb7f11c0d67b9c
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_classification/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "TabFM (default)", "AutoGluon 1.4 (best, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "CatBoost (tuned + ensembled)", "CatBoost (default)", "iLTM (tuned + ensembled)", "CatBoost (tuned)", "XGBoost (tuned + ensembled)", "LightGBM (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "RealMLP (tuned + ensembled)", "RealTabPFN-2.5 (tuned + ensembled)", "TabM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "ChimeraBoost (tuned)", "XGBoost (tuned)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned)", "TabM (tuned)", "LightGBM (tuned)", "ChimeraBoost (default)", "RealTabPFN-2.5 (default)", "TabICL (default)", "iLTM (tuned)", "ModernNCA (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "EBM (tuned + ensembled)", "XGBoost (default)", "xRFM (tuned + ensembled)", "TabSwift (default)", "EBM (tuned)", "TabM (default)", "xRFM (tuned)", "RealMLP (default)", "BetaTabPFN (default)", "TorchMLP (tuned)", "FastaiMLP (tuned + ensembled)", "EBM (default)", "TabDPT (tuned)", "RandomForest (tuned + ensembled)", "iLTM (default)", "LightGBM (default)", "LimiX (default)", "RandomForest (tuned)", "ModernNCA (default)", "ExtraTrees (tuned + ensembled)", "SAP-RPT-OSS (default)", "TabDPT (default)", "FastaiMLP (tuned)", "ExtraTrees (tuned)", "TorchMLP (default)", "FastaiMLP (default)", "TabSTAR (tuned)", "Linear (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "KNN (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "RandomForest (default)", "PerpetualBooster (tuned)", "OrionMSP (default)", "ExtraTrees (default)", "Linear (default)", "TabSTAR (default)", "xRFM (default)", "TabFlex (default)", "KNN (tuned)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.4000000059604645, 0.7444444298744202, 0.4444444477558136, 0.8222222328186035, 0.6777777671813965, 0.9222221970558167, 0.8999999761581421, 0.9333333373069763, 0.9111111164093018, 0.9555555582046509, 0.9333333373069763, 0.9222221970558167, 0.9777777791023254, 0.8999999761581421, 0.8111110925674438, 0.9666666388511658, 0.9333333373069763, 0.9888888597488403, 0.9222221970558167, 0.9666666388511658, 0.8666666746139526, 0.9777777791023254, 0.9666666388511658, 0.9222221970558167, 0.8999999761581421, 0.8888888955116272, 0.9888888597488403, 0.9111111164093018, 0.9222221970558167, 0.9888888597488403, 0.9888888597488403, 0.9666666388511658, 0.9777777791023254, 0.9444444179534912, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 0.8888888955116272, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.6000000238418579, NaN, 0.6333333253860474, 0.8111110925674438, 0.8999999761581421, 0.9111111164093018, 0.7777777910232544, 0.7888888716697693, 0.8444444537162781, 0.8222222328186035, 0.8777777552604675, 0.855555534362793, 0.8777777552604675, 0.8222222328186035, 0.8666666746139526, 0.9888888597488403, 0.8888888955116272, 0.8777777552604675, 0.9111111164093018, 0.8777777552604675, 0.8333333134651184, 1.0, 0.9666666388511658, 0.8999999761581421, 0.9777777791023254, 0.9555555582046509, 0.8999999761581421, 0.9444444179534912, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9888888597488403, 0.9777777791023254, 0.8888888955116272, 0.9555555582046509, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 0.9111111164093018, 1.0, 0.9666666388511658, 0.9777777791023254, 1.0, 1.0, 0.855555534362793, 0.9777777791023254, 1.0, 0.9888888597488403, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9888888597488403, 1.0, 0.9777777791023254, 1.0, 0.9777777791023254, 1.0, 0.9555555582046509, 0.9777777791023254, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0], [0.25555557012557983, 0.36666667461395264, NaN, 0.42222222685813904, 0.7111111283302307, 0.6555555462837219, 0.7666666507720947, 0.7222222089767456, 0.8444444537162781, 0.7777777910232544, 0.8666666746139526, 0.8444444537162781, 0.7777777910232544, 0.8666666746139526, 0.644444465637207, 0.7444444298744202, 0.7888888716697693, 0.8666666746139526, 0.8888888955116272, 0.7444444298744202, 0.9222221970558167, 0.800000011920929, 0.8999999761581421, 0.8888888955116272, 0.7666666507720947, 0.7666666507720947, 0.855555534362793, 0.8999999761581421, 0.8111110925674438, 0.800000011920929, 0.9111111164093018, 0.9555555582046509, 0.9222221970558167, 0.9555555582046509, 0.8444444537162781, 0.9555555582046509, 0.9222221970558167, 0.9666666388511658, 0.9666666388511658, 0.9111111164093018, 0.9777777791023254, 0.9666666388511658, 0.9555555582046509, 0.9666666388511658, 0.9888888597488403, 0.9666666388511658, 0.9555555582046509, 0.8777777552604675, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.9666666388511658, 0.9555555582046509, 0.9777777791023254, 1.0, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.9777777791023254, 1.0, 0.9666666388511658, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5555555820465088, 0.18888889253139496, 0.5777778029441833, NaN, 0.7555555701255798, 0.5777778029441833, 0.699999988079071, 0.6888889074325562, 0.7111111283302307, 0.6888889074325562, 0.7444444298744202, 0.7333333492279053, 0.7777777910232544, 0.7444444298744202, 0.855555534362793, 0.7888888716697693, 0.8888888955116272, 0.7666666507720947, 0.7777777910232544, 0.8777777552604675, 0.7555555701255798, 0.855555534362793, 0.8999999761581421, 0.7666666507720947, 0.8777777552604675, 0.8666666746139526, 0.8666666746139526, 0.8777777552604675, 0.9555555582046509, 0.8888888955116272, 0.8333333134651184, 0.8666666746139526, 0.9111111164093018, 0.7888888716697693, 0.9666666388511658, 0.9111111164093018, 0.9888888597488403, 0.800000011920929, 0.9444444179534912, 0.9444444179534912, 0.9666666388511658, 0.8999999761581421, 0.9555555582046509, 0.9666666388511658, 0.9444444179534912, 0.9888888597488403, 0.9888888597488403, 0.800000011920929, 0.9666666388511658, 0.9777777791023254, 0.9666666388511658, 0.9222221970558167, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 1.0, 0.9222221970558167, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 0.9222221970558167, 1.0, 1.0, 1.0, 1.0], [0.17777778208255768, 0.10000000149011612, 0.2888889014720917, 0.24444444477558136, NaN, 0.5666666626930237, 0.5666666626930237, 0.5222222208976746, 0.5555555820465088, 0.5666666626930237, 0.5777778029441833, 0.6888889074325562, 0.6777777671813965, 0.6222222447395325, 0.699999988079071, 0.6222222447395325, 0.699999988079071, 0.7111111283302307, 0.7333333492279053, 0.699999988079071, 0.7111111283302307, 0.7444444298744202, 0.8222222328186035, 0.6777777671813965, 0.855555534362793, 0.7777777910232544, 0.7555555701255798, 0.8222222328186035, 0.855555534362793, 0.855555534362793, 0.8444444537162781, 0.8777777552604675, 0.9111111164093018, 0.7666666507720947, 0.8666666746139526, 0.9666666388511658, 0.9222221970558167, 0.8111110925674438, 1.0, 0.8999999761581421, 0.9555555582046509, 0.8666666746139526, 0.9666666388511658, 0.9555555582046509, 0.9555555582046509, 0.9444444179534912, 0.9444444179534912, 0.7777777910232544, 0.9888888597488403, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 0.9444444179534912, 0.9444444179534912, 0.9888888597488403, 1.0, 0.9222221970558167, 1.0, 0.9666666388511658, 1.0, 0.9444444179534912, 0.9666666388511658, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9666666388511658, 1.0, 0.9222221970558167, 1.0, 1.0, 1.0, 1.0], [0.3222222328186035, 0.08888889104127884, 0.3444444537162781, 0.42222222685813904, 0.4333333373069763, NaN, 0.6111111044883728, 0.6111111044883728, 0.6333333253860474, 0.6000000238418579, 0.6222222447395325, 0.6666666865348816, 0.6222222447395325, 0.5777778029441833, 0.7333333492279053, 0.6111111044883728, 0.7111111283302307, 0.6222222447395325, 0.644444465637207, 0.7555555701255798, 0.6333333253860474, 0.7222222089767456, 0.7666666507720947, 0.644444465637207, 0.8444444537162781, 0.8333333134651184, 0.7555555701255798, 0.800000011920929, 0.8777777552604675, 0.8999999761581421, 0.8111110925674438, 0.7555555701255798, 0.8666666746139526, 0.699999988079071, 0.9111111164093018, 0.8999999761581421, 0.9444444179534912, 0.7222222089767456, 0.9111111164093018, 0.9222221970558167, 0.9111111164093018, 0.8444444537162781, 0.9333333373069763, 0.9444444179534912, 0.9555555582046509, 0.9222221970558167, 0.9444444179534912, 0.8111110925674438, 0.9666666388511658, 0.9555555582046509, 0.9888888597488403, 0.9222221970558167, 0.9555555582046509, 0.9333333373069763, 0.9888888597488403, 0.9777777791023254, 0.9222221970558167, 0.9777777791023254, 0.9222221970558167, 0.9777777791023254, 0.9555555582046509, 0.9444444179534912, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9666666388511658, 1.0, 0.9222221970558167, 1.0, 1.0, 1.0, 1.0], [0.07777778059244156, 0.2222222238779068, 0.23333333432674408, 0.30000001192092896, 0.4333333373069763, 0.3888888955116272, NaN, 0.5888888835906982, 0.5777778029441833, 0.5888888835906982, 0.6111111044883728, 0.5111111402511597, 0.5888888835906982, 0.6222222447395325, 0.5222222208976746, 0.6000000238418579, 0.5888888835906982, 0.7333333492279053, 0.6888889074325562, 0.5333333611488342, 0.7111111283302307, 0.644444465637207, 0.644444465637207, 0.7888888716697693, 0.7333333492279053, 0.6888889074325562, 0.7444444298744202, 0.7388888597488403, 0.6000000238418579, 0.6666666865348816, 0.7888888716697693, 0.8444444537162781, 0.8222222328186035, 0.9222221970558167, 0.7777777910232544, 0.8222222328186035, 0.7666666507720947, 0.9444444179534912, 0.8333333134651184, 0.8444444537162781, 0.9555555582046509, 0.8666666746139526, 0.9444444179534912, 0.8111110925674438, 0.8888888955116272, 0.9222221970558167, 0.9555555582046509, 0.7222222089767456, 0.8777777552604675, 0.9555555582046509, 0.9888888597488403, 0.9777777791023254, 0.8888888955116272, 0.9666666388511658, 0.9777777791023254, 0.9888888597488403, 0.9777777791023254, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0], [0.10000000149011612, 0.21111111342906952, 0.2777777910232544, 0.31111112236976624, 0.47777777910232544, 0.3888888955116272, 0.41111111640930176, NaN, 0.5222222208976746, 0.47777777910232544, 0.5333333611488342, 0.5444444417953491, 0.6000000238418579, 0.6222222447395325, 0.5222222208976746, 0.6000000238418579, 0.5888888835906982, 0.7222222089767456, 0.6555555462837219, 0.5444444417953491, 0.6777777671813965, 0.6555555462837219, 0.6888889074325562, 0.8111110925674438, 0.7111111283302307, 0.7111111283302307, 0.7555555701255798, 0.7111111283302307, 0.5888888835906982, 0.6888889074325562, 0.7888888716697693, 0.8111110925674438, 0.7888888716697693, 0.9444444179534912, 0.8111110925674438, 0.8111110925674438, 0.7888888716697693, 0.9555555582046509, 0.8222222328186035, 0.855555534362793, 0.9555555582046509, 0.8333333134651184, 0.8888888955116272, 0.8111110925674438, 0.8777777552604675, 0.9444444179534912, 0.9333333373069763, 0.7555555701255798, 0.8777777552604675, 0.9888888597488403, 0.9666666388511658, 0.9555555582046509, 0.8999999761581421, 0.9777777791023254, 0.9888888597488403, 0.9777777791023254, 0.9555555582046509, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0], [0.06666667014360428, 0.15555556118488312, 0.15555556118488312, 0.2888889014720917, 0.4444444477558136, 0.36666667461395264, 0.42222222685813904, 0.47777777910232544, NaN, 0.4444444477558136, 0.5333333611488342, 0.5666666626930237, 0.5, 0.5444444417953491, 0.5777778029441833, 0.5777778029441833, 0.5666666626930237, 0.5888888835906982, 0.6777777671813965, 0.6222222447395325, 0.6555555462837219, 0.6555555462837219, 0.6777777671813965, 0.644444465637207, 0.7333333492279053, 0.7333333492279053, 0.7111111283302307, 0.7777777910232544, 0.6111111044883728, 0.7444444298744202, 0.8666666746139526, 0.8111110925674438, 0.7888888716697693, 0.8888888955116272, 0.7666666507720947, 0.8222222328186035, 0.8333333134651184, 0.9555555582046509, 0.9555555582046509, 0.8666666746139526, 0.9555555582046509, 0.9666666388511658, 0.9111111164093018, 0.9111111164093018, 0.8777777552604675, 1.0, 0.9222221970558167, 0.7555555701255798, 0.8888888955116272, 0.9777777791023254, 0.9555555582046509, 0.9777777791023254, 0.9333333373069763, 0.9666666388511658, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9666666388511658, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.08888889104127884, 0.17777778208255768, 0.2222222238779068, 0.31111112236976624, 0.4333333373069763, 0.4000000059604645, 0.41111111640930176, 0.5222222208976746, 0.5555555820465088, NaN, 0.6222222447395325, 0.46666666865348816, 0.6222222447395325, 0.5777778029441833, 0.4888888895511627, 0.6111111044883728, 0.6000000238418579, 0.7333333492279053, 0.7111111283302307, 0.5333333611488342, 0.699999988079071, 0.6222222447395325, 0.6666666865348816, 0.7555555701255798, 0.6888889074325562, 0.699999988079071, 0.7111111283302307, 0.6888889074325562, 0.5888888835906982, 0.6666666865348816, 0.7777777910232544, 0.8333333134651184, 0.8111110925674438, 0.9333333373069763, 0.7666666507720947, 0.8222222328186035, 0.800000011920929, 0.9333333373069763, 0.8444444537162781, 0.855555534362793, 0.9444444179534912, 0.8222222328186035, 0.9666666388511658, 0.800000011920929, 0.8888888955116272, 0.9333333373069763, 0.9222221970558167, 0.7222222089767456, 0.8777777552604675, 0.9333333373069763, 0.9555555582046509, 0.9666666388511658, 0.8777777552604675, 0.9555555582046509, 0.9666666388511658, 0.9777777791023254, 0.9444444179534912, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0], [0.04444444552063942, 0.12222222238779068, 0.13333334028720856, 0.25555557012557983, 0.42222222685813904, 0.3777777850627899, 0.3888888955116272, 0.46666666865348816, 0.46666666865348816, 0.3777777850627899, NaN, 0.4555555582046509, 0.5222222208976746, 0.5, 0.5, 0.6222222447395325, 0.5222222208976746, 0.5777778029441833, 0.8666666746139526, 0.5333333611488342, 0.5888888835906982, 0.6777777671813965, 0.6222222447395325, 0.6666666865348816, 0.6555555462837219, 0.6000000238418579, 0.6333333253860474, 0.644444465637207, 0.6222222447395325, 0.6888889074325562, 0.800000011920929, 0.8111110925674438, 0.9333333373069763, 0.8222222328186035, 0.6888889074325562, 0.8444444537162781, 0.8111110925674438, 0.9111111164093018, 0.8444444537162781, 0.8777777552604675, 0.8444444537162781, 0.800000011920929, 0.9777777791023254, 0.8444444537162781, 0.8999999761581421, 0.9666666388511658, 0.9666666388511658, 0.6777777671813965, 0.8999999761581421, 0.9888888597488403, 0.9888888597488403, 0.9555555582046509, 0.8888888955116272, 0.8777777552604675, 0.9888888597488403, 1.0, 0.8999999761581421, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9888888597488403, 1.0, 0.9222221970558167, 1.0, 0.9666666388511658, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.14444445073604584, 0.15555556118488312, 0.2666666805744171, 0.31111112236976624, 0.3333333432674408, 0.4888888895511627, 0.4555555582046509, 0.4333333373069763, 0.5333333611488342, 0.5444444417953491, NaN, 0.47777777910232544, 0.47777777910232544, 0.46666666865348816, 0.5333333611488342, 0.5111111402511597, 0.6111111044883728, 0.6888889074325562, 0.5, 0.5666666626930237, 0.6333333253860474, 0.8666666746139526, 0.6222222447395325, 0.6222222447395325, 0.6666666865348816, 0.699999988079071, 0.6555555462837219, 0.6666666865348816, 0.6111111044883728, 0.7333333492279053, 0.7666666507720947, 0.8222222328186035, 0.7666666507720947, 0.7444444298744202, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 0.8222222328186035, 0.8666666746139526, 0.9111111164093018, 0.7666666507720947, 0.9666666388511658, 0.800000011920929, 0.8999999761581421, 0.9777777791023254, 0.9777777791023254, 0.6888889074325562, 0.9111111164093018, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 0.9555555582046509, 0.9111111164093018, 0.9777777791023254, 0.9777777791023254, 0.8666666746139526, 0.9888888597488403, 0.9777777791023254, 0.9888888597488403, 0.9777777791023254, 0.9888888597488403, 0.9777777791023254, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0], [0.07777778059244156, 0.12222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3222222328186035, 0.3777777850627899, 0.41111111640930176, 0.4000000059604645, 0.5, 0.3777777850627899, 0.47777777910232544, 0.5222222208976746, NaN, 0.6222222447395325, 0.5111111402511597, 0.6111111044883728, 0.5222222208976746, 0.7555555701255798, 0.5888888835906982, 0.5666666626930237, 0.7222222089767456, 0.6888889074325562, 0.5666666626930237, 0.8777777552604675, 0.6333333253860474, 0.6333333253860474, 0.7333333492279053, 0.7111111283302307, 0.6111111044883728, 0.699999988079071, 0.7111111283302307, 0.8222222328186035, 0.6777777671813965, 0.8999999761581421, 0.6666666865348816, 0.855555534362793, 0.8222222328186035, 0.9222221970558167, 0.855555534362793, 0.8777777552604675, 0.8666666746139526, 0.855555534362793, 0.9333333373069763, 0.8333333134651184, 0.8888888955116272, 0.9555555582046509, 0.8777777552604675, 0.7444444298744202, 0.8888888955116272, 0.9111111164093018, 0.8888888955116272, 0.9777777791023254, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.9111111164093018, 0.8777777552604675, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.8777777552604675, 1.0, 0.8888888955116272, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9222221970558167, 1.0], [0.02222222276031971, 0.17777778208255768, 0.13333334028720856, 0.25555557012557983, 0.3777777850627899, 0.42222222685813904, 0.3777777850627899, 0.3777777850627899, 0.4555555582046509, 0.42222222685813904, 0.5, 0.5222222208976746, 0.3777777850627899, NaN, 0.5555555820465088, 0.42222222685813904, 0.5333333611488342, 0.4888888895511627, 0.6000000238418579, 0.5888888835906982, 0.7555555701255798, 0.47777777910232544, 0.6777777671813965, 0.4888888895511627, 0.6111111044883728, 0.5777778029441833, 0.6222222447395325, 0.6888889074325562, 0.6666666865348816, 0.6111111044883728, 0.7555555701255798, 0.7555555701255798, 0.7444444298744202, 0.8444444537162781, 0.6555555462837219, 0.8222222328186035, 0.8999999761581421, 0.8888888955116272, 1.0, 0.855555534362793, 0.8333333134651184, 0.9777777791023254, 0.8444444537162781, 0.8111110925674438, 0.8777777552604675, 0.8111110925674438, 0.9333333373069763, 0.8444444537162781, 0.8666666746139526, 0.855555534362793, 0.8999999761581421, 0.8888888955116272, 0.8222222328186035, 0.9777777791023254, 0.8777777552604675, 0.9888888597488403, 0.9666666388511658, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9333333373069763, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.13333334028720856, 0.35555556416511536, 0.14444445073604584, 0.30000001192092896, 0.2666666805744171, 0.47777777910232544, 0.47777777910232544, 0.42222222685813904, 0.5111111402511597, 0.5, 0.5333333611488342, 0.4888888895511627, 0.4444444477558136, NaN, 0.4333333373069763, 0.46666666865348816, 0.5222222208976746, 0.5444444417953491, 0.800000011920929, 0.5555555820465088, 0.5555555820465088, 0.644444465637207, 0.6111111044883728, 0.7111111283302307, 0.699999988079071, 0.6222222447395325, 0.6888889074325562, 0.644444465637207, 0.6888889074325562, 0.7111111283302307, 0.7333333492279053, 0.7666666507720947, 0.6555555462837219, 0.855555534362793, 0.8222222328186035, 0.7444444298744202, 0.7111111283302307, 0.8222222328186035, 0.800000011920929, 0.8222222328186035, 0.7333333492279053, 0.8666666746139526, 0.8888888955116272, 0.9111111164093018, 0.855555534362793, 0.9222221970558167, 0.6666666865348816, 0.9555555582046509, 0.8222222328186035, 0.9666666388511658, 0.855555534362793, 0.8999999761581421, 0.8777777552604675, 0.9666666388511658, 0.9555555582046509, 0.8999999761581421, 1.0, 0.8444444537162781, 1.0, 0.9111111164093018, 0.855555534362793, 0.9666666388511658, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.9555555582046509, 1.0, 0.9222221970558167, 1.0, 1.0, 1.0, 1.0], [0.18888889253139496, 0.011111111380159855, 0.25555557012557983, 0.21111111342906952, 0.3777777850627899, 0.3888888955116272, 0.4000000059604645, 0.4000000059604645, 0.42222222685813904, 0.3888888955116272, 0.3777777850627899, 0.46666666865348816, 0.3888888955116272, 0.5777778029441833, 0.5666666626930237, NaN, 0.5222222208976746, 0.4555555582046509, 0.5, 0.5888888835906982, 0.6555555462837219, 0.7333333492279053, 0.6111111044883728, 0.5222222208976746, 0.5666666626930237, 0.5777778029441833, 0.6333333253860474, 0.7333333492279053, 0.7111111283302307, 0.6555555462837219, 0.6111111044883728, 0.6666666865348816, 0.6666666865348816, 0.6555555462837219, 0.7555555701255798, 0.8444444537162781, 0.9666666388511658, 0.6555555462837219, 0.9777777791023254, 0.9111111164093018, 0.7777777910232544, 0.8222222328186035, 0.8333333134651184, 0.8777777552604675, 0.8333333134651184, 0.8333333134651184, 1.0, 0.7666666507720947, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.8777777552604675, 0.8777777552604675, 0.9111111164093018, 0.8666666746139526, 0.9777777791023254, 0.8999999761581421, 1.0, 0.9333333373069763, 1.0, 0.9444444179534912, 0.9444444179534912, 0.9777777791023254, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0], [0.03333333507180214, 0.1111111119389534, 0.21111111342906952, 0.1111111119389534, 0.30000001192092896, 0.2888889014720917, 0.41111111640930176, 0.41111111640930176, 0.4333333373069763, 0.4000000059604645, 0.47777777910232544, 0.4888888895511627, 0.47777777910232544, 0.46666666865348816, 0.5333333611488342, 0.47777777910232544, NaN, 0.5555555820465088, 0.5, 0.5444444417953491, 0.5888888835906982, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.6222222447395325, 0.6333333253860474, 0.644444465637207, 0.7444444298744202, 0.6111111044883728, 0.6888889074325562, 0.7333333492279053, 0.7333333492279053, 0.6555555462837219, 0.7111111283302307, 0.699999988079071, 0.7555555701255798, 0.8222222328186035, 0.7777777910232544, 0.8888888955116272, 0.8222222328186035, 0.8444444537162781, 0.8666666746139526, 0.8222222328186035, 0.8777777552604675, 0.7888888716697693, 0.8444444537162781, 0.9111111164093018, 0.855555534362793, 0.7777777910232544, 0.9555555582046509, 0.8999999761581421, 0.855555534362793, 0.9222221970558167, 0.9333333373069763, 0.8888888955116272, 0.8888888955116272, 0.9666666388511658, 0.9222221970558167, 0.8999999761581421, 0.9222221970558167, 0.9888888597488403, 0.9111111164093018, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.06666667014360428, 0.12222222238779068, 0.13333334028720856, 0.23333333432674408, 0.2888889014720917, 0.3777777850627899, 0.2666666805744171, 0.2777777910232544, 0.41111111640930176, 0.2666666805744171, 0.42222222685813904, 0.3888888955116272, 0.24444444477558136, 0.5111111402511597, 0.47777777910232544, 0.5444444417953491, 0.4444444477558136, NaN, 0.5222222208976746, 0.4888888895511627, 0.6777777671813965, 0.6000000238418579, 0.5111111402511597, 0.7111111283302307, 0.5777778029441833, 0.5777778029441833, 0.644444465637207, 0.6666666865348816, 0.5222222208976746, 0.6555555462837219, 0.644444465637207, 0.7222222089767456, 0.6555555462837219, 0.8666666746139526, 0.6333333253860474, 0.7777777910232544, 0.7888888716697693, 0.8999999761581421, 0.8444444537162781, 0.855555534362793, 0.8222222328186035, 0.8111110925674438, 0.855555534362793, 0.8111110925674438, 0.8666666746139526, 0.8999999761581421, 0.8111110925674438, 0.7333333492279053, 0.8666666746139526, 0.855555534362793, 0.8666666746139526, 0.9333333373069763, 0.7777777910232544, 0.855555534362793, 0.8666666746139526, 0.9111111164093018, 0.8666666746139526, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9777777791023254, 1.0, 0.8777777552604675, 0.9888888597488403, 0.8777777552604675, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0], [0.011111111380159855, 0.08888889104127884, 0.1111111119389534, 0.2222222238779068, 0.2666666805744171, 0.35555556416511536, 0.31111112236976624, 0.3444444537162781, 0.3222222328186035, 0.2888889014720917, 0.13333334028720856, 0.31111112236976624, 0.41111111640930176, 0.4000000059604645, 0.4555555582046509, 0.5, 0.5, 0.47777777910232544, NaN, 0.46666666865348816, 0.46666666865348816, 0.6111111044883728, 0.5111111402511597, 0.6000000238418579, 0.5666666626930237, 0.5444444417953491, 0.5666666626930237, 0.6333333253860474, 0.5777778029441833, 0.6111111044883728, 0.6777777671813965, 0.7222222089767456, 0.7666666507720947, 0.7444444298744202, 0.6666666865348816, 0.8111110925674438, 0.7777777910232544, 0.8111110925674438, 0.8222222328186035, 0.855555534362793, 0.8222222328186035, 0.7555555701255798, 0.9222221970558167, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.6555555462837219, 0.855555534362793, 0.9666666388511658, 0.9555555582046509, 0.9444444179534912, 0.855555534362793, 0.8666666746139526, 0.9777777791023254, 1.0, 0.855555534362793, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9888888597488403, 1.0, 0.9111111164093018, 1.0, 0.9666666388511658, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0], [0.07777778059244156, 0.12222222238779068, 0.25555557012557983, 0.12222222238779068, 0.30000001192092896, 0.24444444477558136, 0.46666666865348816, 0.4555555582046509, 0.3777777850627899, 0.46666666865348816, 0.46666666865348816, 0.5, 0.4333333373069763, 0.41111111640930176, 0.20000000298023224, 0.41111111640930176, 0.4555555582046509, 0.5111111402511597, 0.5333333611488342, NaN, 0.5222222208976746, 0.5333333611488342, 0.644444465637207, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.5777778029441833, 0.6111111044883728, 0.5888888835906982, 0.6888889074325562, 0.6555555462837219, 0.6666666865348816, 0.7555555701255798, 0.6555555462837219, 0.8333333134651184, 0.7333333492279053, 0.6777777671813965, 0.6888889074325562, 0.7666666507720947, 0.7555555701255798, 0.8111110925674438, 0.699999988079071, 0.8333333134651184, 0.8222222328186035, 0.855555534362793, 0.8222222328186035, 0.8888888955116272, 0.6555555462837219, 0.9333333373069763, 0.7888888716697693, 0.9111111164093018, 0.8333333134651184, 0.8444444537162781, 0.8777777552604675, 0.9333333373069763, 0.9111111164093018, 0.8888888955116272, 0.9888888597488403, 0.8222222328186035, 1.0, 0.8888888955116272, 0.8666666746139526, 0.9666666388511658, 0.9888888597488403, 0.9444444179534912, 0.9888888597488403, 0.9777777791023254, 0.9333333373069763, 1.0, 0.8999999761581421, 0.9888888597488403, 1.0, 1.0, 1.0], [0.03333333507180214, 0.1666666716337204, 0.07777778059244156, 0.24444444477558136, 0.2888889014720917, 0.36666667461395264, 0.2888889014720917, 0.3222222328186035, 0.3444444537162781, 0.30000001192092896, 0.41111111640930176, 0.4333333373069763, 0.2777777910232544, 0.24444444477558136, 0.4444444477558136, 0.3444444537162781, 0.41111111640930176, 0.3222222328186035, 0.5333333611488342, 0.47777777910232544, NaN, 0.4000000059604645, 0.6000000238418579, 0.4333333373069763, 0.5222222208976746, 0.47777777910232544, 0.5777778029441833, 0.5777778029441833, 0.4888888895511627, 0.5555555820465088, 0.6777777671813965, 0.644444465637207, 0.6777777671813965, 0.7555555701255798, 0.5444444417953491, 0.7111111283302307, 0.7666666507720947, 0.800000011920929, 0.8999999761581421, 0.7777777910232544, 0.8111110925674438, 0.8777777552604675, 0.7333333492279053, 0.7333333492279053, 0.8444444537162781, 0.7333333492279053, 0.8222222328186035, 0.800000011920929, 0.8222222328186035, 0.7555555701255798, 0.800000011920929, 0.7888888716697693, 0.7444444298744202, 0.9666666388511658, 0.855555534362793, 0.9333333373069763, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9666666388511658, 0.9888888597488403, 0.9666666388511658, 0.9111111164093018, 0.9555555582046509, 0.9777777791023254, 0.9888888597488403, 1.0, 1.0, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.20000000298023224, 0.14444445073604584, 0.25555557012557983, 0.2777777910232544, 0.35555556416511536, 0.3444444537162781, 0.3444444537162781, 0.3777777850627899, 0.3222222328186035, 0.36666667461395264, 0.31111112236976624, 0.5222222208976746, 0.4444444477558136, 0.2666666805744171, 0.4333333373069763, 0.4000000059604645, 0.3888888955116272, 0.46666666865348816, 0.6000000238418579, NaN, 0.4444444477558136, 0.41111111640930176, 0.4888888895511627, 0.4888888895511627, 0.5888888835906982, 0.6333333253860474, 0.5888888835906982, 0.5444444417953491, 0.5666666626930237, 0.5555555820465088, 0.6555555462837219, 0.6111111044883728, 0.6555555462837219, 0.7666666507720947, 0.8444444537162781, 0.6555555462837219, 0.8999999761581421, 0.800000011920929, 0.7444444298744202, 0.800000011920929, 0.7888888716697693, 0.7666666507720947, 0.7777777910232544, 0.7555555701255798, 0.8666666746139526, 0.7666666507720947, 0.7888888716697693, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8777777552604675, 0.8333333134651184, 0.9777777791023254, 0.9111111164093018, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8999999761581421, 0.9777777791023254, 0.9666666388511658, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9333333373069763, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.02222222276031971, 0.03333333507180214, 0.10000000149011612, 0.10000000149011612, 0.17777778208255768, 0.23333333432674408, 0.35555556416511536, 0.31111112236976624, 0.3222222328186035, 0.3333333432674408, 0.3777777850627899, 0.13333334028720856, 0.4333333373069763, 0.3222222328186035, 0.35555556416511536, 0.3888888955116272, 0.46666666865348816, 0.4888888895511627, 0.4888888895511627, 0.35555556416511536, 0.4000000059604645, 0.5555555820465088, NaN, 0.5444444417953491, 0.4888888895511627, 0.5222222208976746, 0.5444444417953491, 0.5555555820465088, 0.6000000238418579, 0.5777778029441833, 0.5777778029441833, 0.6777777671813965, 0.7444444298744202, 0.7222222089767456, 0.6333333253860474, 0.7333333492279053, 0.7555555701255798, 0.7222222089767456, 0.7666666507720947, 0.7888888716697693, 0.699999988079071, 0.7555555701255798, 0.855555534362793, 0.7444444298744202, 0.7888888716697693, 0.9333333373069763, 0.9666666388511658, 0.644444465637207, 0.8444444537162781, 0.9333333373069763, 0.8333333134651184, 0.8222222328186035, 0.8333333134651184, 0.7888888716697693, 0.9222221970558167, 0.9333333373069763, 0.8111110925674438, 1.0, 0.8666666746139526, 1.0, 0.9111111164093018, 0.9111111164093018, 0.9444444179534912, 0.9888888597488403, 0.9666666388511658, 0.9888888597488403, 0.9888888597488403, 0.9666666388511658, 1.0, 0.9111111164093018, 0.9888888597488403, 1.0, 1.0, 1.0], [0.03333333507180214, 0.10000000149011612, 0.1111111119389534, 0.23333333432674408, 0.3222222328186035, 0.35555556416511536, 0.21111111342906952, 0.18888889253139496, 0.35555556416511536, 0.24444444477558136, 0.3333333432674408, 0.3777777850627899, 0.12222222238779068, 0.5111111402511597, 0.3888888955116272, 0.47777777910232544, 0.36666667461395264, 0.2888889014720917, 0.4000000059604645, 0.4333333373069763, 0.5666666626930237, 0.5888888835906982, 0.4555555582046509, NaN, 0.4333333373069763, 0.4555555582046509, 0.5666666626930237, 0.5444444417953491, 0.5333333611488342, 0.5666666626930237, 0.6222222447395325, 0.6555555462837219, 0.6111111044883728, 0.800000011920929, 0.6000000238418579, 0.7222222089767456, 0.6777777671813965, 0.8777777552604675, 0.7555555701255798, 0.7555555701255798, 0.7444444298744202, 0.8111110925674438, 0.7666666507720947, 0.8333333134651184, 0.8444444537162781, 0.8333333134651184, 0.7555555701255798, 0.7333333492279053, 0.8111110925674438, 0.8111110925674438, 0.7666666507720947, 0.9555555582046509, 0.8222222328186035, 0.8666666746139526, 0.8444444537162781, 0.8666666746139526, 0.8666666746139526, 0.9888888597488403, 1.0, 1.0, 0.9777777791023254, 1.0, 0.8777777552604675, 1.0, 0.8666666746139526, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0], [0.07777778059244156, 0.02222222276031971, 0.23333333432674408, 0.12222222238779068, 0.14444445073604584, 0.15555556118488312, 0.2666666805744171, 0.2888889014720917, 0.2666666805744171, 0.31111112236976624, 0.3444444537162781, 0.3777777850627899, 0.36666667461395264, 0.3888888955116272, 0.2888889014720917, 0.4333333373069763, 0.3777777850627899, 0.42222222685813904, 0.4333333373069763, 0.3333333432674408, 0.47777777910232544, 0.5111111402511597, 0.5111111402511597, 0.5666666626930237, NaN, 0.6111111044883728, 0.6222222447395325, 0.5222222208976746, 0.4444444477558136, 0.6222222447395325, 0.6222222447395325, 0.5333333611488342, 0.6888889074325562, 0.6777777671813965, 0.8666666746139526, 0.6333333253860474, 0.5888888835906982, 0.6888889074325562, 0.7111111283302307, 0.7888888716697693, 0.800000011920929, 0.6666666865348816, 0.7222222089767456, 0.7777777910232544, 0.7444444298744202, 0.8333333134651184, 0.8777777552604675, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.9222221970558167, 0.8999999761581421, 0.855555534362793, 0.800000011920929, 0.9444444179534912, 0.9555555582046509, 0.8444444537162781, 1.0, 0.8444444537162781, 1.0, 0.8999999761581421, 0.8444444537162781, 0.9777777791023254, 0.9777777791023254, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9222221970558167, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.04444444552063942, 0.23333333432674408, 0.13333334028720856, 0.2222222238779068, 0.1666666716337204, 0.31111112236976624, 0.2888889014720917, 0.2666666805744171, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.36666667461395264, 0.42222222685813904, 0.30000001192092896, 0.42222222685813904, 0.36666667461395264, 0.42222222685813904, 0.4555555582046509, 0.3333333432674408, 0.5222222208976746, 0.5111111402511597, 0.47777777910232544, 0.5444444417953491, 0.3888888955116272, NaN, 0.5333333611488342, 0.4444444477558136, 0.5333333611488342, 0.644444465637207, 0.6333333253860474, 0.6222222447395325, 0.6111111044883728, 0.6555555462837219, 0.6555555462837219, 0.6333333253860474, 0.6111111044883728, 0.699999988079071, 0.6777777671813965, 0.6888889074325562, 0.6777777671813965, 0.6888889074325562, 0.6777777671813965, 0.855555534362793, 0.7111111283302307, 0.7666666507720947, 0.8666666746139526, 0.6888889074325562, 0.7222222089767456, 0.8666666746139526, 0.800000011920929, 0.7777777910232544, 0.9222221970558167, 0.7111111283302307, 0.8333333134651184, 0.800000011920929, 0.7888888716697693, 0.9333333373069763, 0.7555555701255798, 0.9444444179534912, 0.855555534362793, 0.7555555701255798, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9777777791023254, 1.0, 1.0, 1.0], [0.1111111119389534, 0.10000000149011612, 0.14444445073604584, 0.13333334028720856, 0.24444444477558136, 0.24444444477558136, 0.25555557012557983, 0.24444444477558136, 0.2888889014720917, 0.2888889014720917, 0.36666667461395264, 0.30000001192092896, 0.2666666805744171, 0.3777777850627899, 0.3777777850627899, 0.36666667461395264, 0.35555556416511536, 0.35555556416511536, 0.4333333373069763, 0.42222222685813904, 0.42222222685813904, 0.41111111640930176, 0.4555555582046509, 0.4333333373069763, 0.3777777850627899, 0.46666666865348816, NaN, 0.4333333373069763, 0.5, 0.5555555820465088, 0.4888888895511627, 0.5111111402511597, 0.5, 0.7111111283302307, 0.5777778029441833, 0.5777778029441833, 0.6222222447395325, 0.7111111283302307, 0.6555555462837219, 0.6555555462837219, 0.6111111044883728, 0.7333333492279053, 0.6000000238418579, 0.7888888716697693, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.7555555701255798, 0.8444444537162781, 0.7444444298744202, 0.8444444537162781, 0.855555534362793, 0.855555534362793, 0.8666666746139526, 0.8888888955116272, 0.8999999761581421, 0.9444444179534912, 0.8999999761581421, 0.9444444179534912, 0.9888888597488403, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9222221970558167, 0.9777777791023254, 0.9666666388511658, 1.0, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0], [0.011111111380159855, 0.0555555559694767, 0.10000000149011612, 0.12222222238779068, 0.17777778208255768, 0.20000000298023224, 0.2611111104488373, 0.2888889014720917, 0.2222222238779068, 0.31111112236976624, 0.35555556416511536, 0.3444444537162781, 0.2888889014720917, 0.31111112236976624, 0.31111112236976624, 0.2666666805744171, 0.25555557012557983, 0.3333333432674408, 0.36666667461395264, 0.3888888955116272, 0.42222222685813904, 0.36666667461395264, 0.4444444477558136, 0.4555555582046509, 0.47777777910232544, 0.5555555820465088, 0.5666666626930237, NaN, 0.5333333611488342, 0.5777778029441833, 0.6111111044883728, 0.5222222208976746, 0.5333333611488342, 0.5555555820465088, 0.644444465637207, 0.5888888835906982, 0.6222222447395325, 0.6222222447395325, 0.7222222089767456, 0.6888889074325562, 0.7222222089767456, 0.6666666865348816, 0.644444465637207, 0.7777777910232544, 0.7222222089767456, 0.7666666507720947, 0.7666666507720947, 0.7222222089767456, 0.7111111283302307, 0.8999999761581421, 0.8444444537162781, 0.7888888716697693, 0.8666666746139526, 0.9222221970558167, 0.855555534362793, 0.855555534362793, 0.9111111164093018, 0.9555555582046509, 0.8666666746139526, 0.9555555582046509, 0.9555555582046509, 0.8999999761581421, 0.9444444179534912, 0.9777777791023254, 0.9555555582046509, 0.9888888597488403, 0.9666666388511658, 0.9888888597488403, 1.0, 0.9555555582046509, 0.9777777791023254, 1.0, 1.0, 1.0], [0.08888889104127884, 0.03333333507180214, 0.18888889253139496, 0.04444444552063942, 0.14444445073604584, 0.12222222238779068, 0.4000000059604645, 0.41111111640930176, 0.3888888955116272, 0.41111111640930176, 0.3777777850627899, 0.3333333432674408, 0.3888888955116272, 0.3333333432674408, 0.35555556416511536, 0.2888889014720917, 0.3888888955116272, 0.47777777910232544, 0.42222222685813904, 0.41111111640930176, 0.5111111402511597, 0.41111111640930176, 0.4000000059604645, 0.46666666865348816, 0.5555555820465088, 0.46666666865348816, 0.5, 0.46666666865348816, NaN, 0.5888888835906982, 0.644444465637207, 0.5111111402511597, 0.6000000238418579, 0.6222222447395325, 0.7111111283302307, 0.5777778029441833, 0.5666666626930237, 0.6555555462837219, 0.699999988079071, 0.6111111044883728, 0.7333333492279053, 0.644444465637207, 0.7111111283302307, 0.8333333134651184, 0.6777777671813965, 0.6888889074325562, 0.7888888716697693, 0.7666666507720947, 0.7555555701255798, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.9111111164093018, 0.7444444298744202, 0.8222222328186035, 0.800000011920929, 0.8777777552604675, 0.8888888955116272, 0.7333333492279053, 0.8888888955116272, 0.855555534362793, 0.7555555701255798, 0.9666666388511658, 0.9555555582046509, 0.9666666388511658, 0.8999999761581421, 0.9222221970558167, 0.8777777552604675, 1.0, 0.9111111164093018, 0.8888888955116272, 1.0, 1.0, 1.0], [0.07777778059244156, 0.03333333507180214, 0.20000000298023224, 0.1111111119389534, 0.14444445073604584, 0.10000000149011612, 0.3333333432674408, 0.31111112236976624, 0.25555557012557983, 0.3333333432674408, 0.31111112236976624, 0.3888888955116272, 0.30000001192092896, 0.3888888955116272, 0.31111112236976624, 0.3444444537162781, 0.31111112236976624, 0.3444444537162781, 0.3888888955116272, 0.31111112236976624, 0.4444444477558136, 0.4555555582046509, 0.42222222685813904, 0.4333333373069763, 0.3777777850627899, 0.35555556416511536, 0.4444444477558136, 0.42222222685813904, 0.41111111640930176, NaN, 0.5777778029441833, 0.5, 0.5444444417953491, 0.6111111044883728, 0.6111111044883728, 0.5333333611488342, 0.6000000238418579, 0.6555555462837219, 0.644444465637207, 0.6222222447395325, 0.7222222089767456, 0.6888889074325562, 0.5666666626930237, 0.8888888955116272, 0.6666666865348816, 0.6333333253860474, 0.7777777910232544, 0.6666666865348816, 0.6555555462837219, 0.7888888716697693, 0.6777777671813965, 0.7444444298744202, 0.9444444179534912, 0.800000011920929, 0.7888888716697693, 0.7777777910232544, 0.8333333134651184, 0.855555534362793, 0.7555555701255798, 0.855555534362793, 0.8444444537162781, 0.7555555701255798, 0.9888888597488403, 0.9888888597488403, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.8888888955116272, 1.0, 0.9222221970558167, 0.9666666388511658, 1.0, 1.0, 1.0], [0.011111111380159855, 0.03333333507180214, 0.08888889104127884, 0.1666666716337204, 0.15555556118488312, 0.18888889253139496, 0.21111111342906952, 0.21111111342906952, 0.13333334028720856, 0.2222222238779068, 0.20000000298023224, 0.2666666805744171, 0.2888889014720917, 0.24444444477558136, 0.2888889014720917, 0.3888888955116272, 0.2666666805744171, 0.35555556416511536, 0.3222222328186035, 0.3444444537162781, 0.3222222328186035, 0.4333333373069763, 0.42222222685813904, 0.3777777850627899, 0.3777777850627899, 0.36666667461395264, 0.5111111402511597, 0.3888888955116272, 0.35555556416511536, 0.42222222685813904, NaN, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.5888888835906982, 0.6000000238418579, 0.6222222447395325, 0.6555555462837219, 0.6555555462837219, 0.6000000238418579, 0.8777777552604675, 0.6555555462837219, 0.6333333253860474, 0.7111111283302307, 0.699999988079071, 0.7555555701255798, 0.7555555701255798, 0.6666666865348816, 0.7555555701255798, 0.7333333492279053, 0.800000011920929, 0.8222222328186035, 0.7666666507720947, 0.8333333134651184, 0.8666666746139526, 0.9888888597488403, 0.8333333134651184, 0.9222221970558167, 0.9111111164093018, 0.9222221970558167, 0.9111111164093018, 0.9111111164093018, 0.9777777791023254, 0.9888888597488403, 0.9666666388511658, 1.0, 0.9888888597488403, 0.9555555582046509, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.011111111380159855, 0.011111111380159855, 0.04444444552063942, 0.13333334028720856, 0.12222222238779068, 0.24444444477558136, 0.15555556118488312, 0.18888889253139496, 0.18888889253139496, 0.1666666716337204, 0.18888889253139496, 0.23333333432674408, 0.17777778208255768, 0.24444444477558136, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.35555556416511536, 0.4444444477558136, 0.3222222328186035, 0.3444444537162781, 0.46666666865348816, 0.3777777850627899, 0.4888888895511627, 0.47777777910232544, 0.4888888895511627, 0.5, 0.5333333611488342, NaN, 0.6111111044883728, 0.4555555582046509, 0.5777778029441833, 0.7444444298744202, 0.6000000238418579, 0.5888888835906982, 0.7222222089767456, 0.6888889074325562, 0.7888888716697693, 0.6666666865348816, 0.9222221970558167, 0.699999988079071, 0.7222222089767456, 0.7222222089767456, 0.6777777671813965, 0.6000000238418579, 0.7666666507720947, 0.7222222089767456, 0.800000011920929, 0.8777777552604675, 0.6777777671813965, 0.7555555701255798, 0.8444444537162781, 0.8666666746139526, 0.800000011920929, 0.9777777791023254, 0.8999999761581421, 0.9666666388511658, 0.9111111164093018, 0.9222221970558167, 0.8777777552604675, 0.8777777552604675, 0.8666666746139526, 0.8777777552604675, 0.8888888955116272, 0.9444444179534912, 1.0, 0.8222222328186035, 0.9777777791023254, 1.0, 0.8999999761581421, 1.0], [0.03333333507180214, 0.02222222276031971, 0.07777778059244156, 0.08888889104127884, 0.08888889104127884, 0.13333334028720856, 0.17777778208255768, 0.21111111342906952, 0.21111111342906952, 0.18888889253139496, 0.06666667014360428, 0.17777778208255768, 0.3222222328186035, 0.25555557012557983, 0.23333333432674408, 0.3333333432674408, 0.3444444537162781, 0.3444444537162781, 0.23333333432674408, 0.24444444477558136, 0.3222222328186035, 0.3444444537162781, 0.25555557012557983, 0.3888888955116272, 0.31111112236976624, 0.3888888955116272, 0.5, 0.46666666865348816, 0.4000000059604645, 0.4555555582046509, 0.4333333373069763, 0.3888888955116272, NaN, 0.6333333253860474, 0.5444444417953491, 0.4888888895511627, 0.5333333611488342, 0.6777777671813965, 0.5444444417953491, 0.5888888835906982, 0.6666666865348816, 0.6888889074325562, 0.644444465637207, 0.6555555462837219, 0.7555555701255798, 0.8111110925674438, 0.5333333611488342, 0.6111111044883728, 0.7777777910232544, 0.7222222089767456, 0.8666666746139526, 0.8333333134651184, 0.6555555462837219, 0.7111111283302307, 0.8999999761581421, 0.9222221970558167, 0.7666666507720947, 0.9444444179534912, 0.8777777552604675, 0.9555555582046509, 0.8888888955116272, 0.8888888955116272, 0.8777777552604675, 0.9222221970558167, 0.9333333373069763, 0.9888888597488403, 0.9222221970558167, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.02222222276031971, 0.1111111119389534, 0.04444444552063942, 0.21111111342906952, 0.23333333432674408, 0.30000001192092896, 0.07777778059244156, 0.0555555559694767, 0.1111111119389534, 0.06666667014360428, 0.17777778208255768, 0.23333333432674408, 0.10000000149011612, 0.15555556118488312, 0.3444444537162781, 0.3444444537162781, 0.2888889014720917, 0.13333334028720856, 0.25555557012557983, 0.3444444537162781, 0.24444444477558136, 0.3888888955116272, 0.2777777910232544, 0.20000000298023224, 0.3222222328186035, 0.3444444537162781, 0.2888889014720917, 0.4444444477558136, 0.3777777850627899, 0.3888888955116272, 0.4000000059604645, 0.5444444417953491, 0.36666667461395264, NaN, 0.3777777850627899, 0.6111111044883728, 0.5555555820465088, 0.6777777671813965, 0.6222222447395325, 0.6333333253860474, 0.5777778029441833, 0.7333333492279053, 0.6222222447395325, 0.46666666865348816, 0.644444465637207, 0.5444444417953491, 0.46666666865348816, 0.6888889074325562, 0.6555555462837219, 0.5666666626930237, 0.7333333492279053, 0.6888889074325562, 0.5888888835906982, 0.800000011920929, 0.7222222089767456, 0.7555555701255798, 0.8666666746139526, 0.855555534362793, 0.9777777791023254, 0.855555534362793, 0.9666666388511658, 1.0, 0.8111110925674438, 0.8777777552604675, 0.855555534362793, 0.9666666388511658, 0.9777777791023254, 1.0, 0.8999999761581421, 1.0, 0.9777777791023254, 0.9666666388511658, 0.8999999761581421, 1.0], [0.0555555559694767, 0.04444444552063942, 0.15555556118488312, 0.03333333507180214, 0.13333334028720856, 0.08888889104127884, 0.2222222238779068, 0.18888889253139496, 0.23333333432674408, 0.23333333432674408, 0.31111112236976624, 0.25555557012557983, 0.3333333432674408, 0.3444444537162781, 0.14444445073604584, 0.24444444477558136, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.1666666716337204, 0.4555555582046509, 0.3444444537162781, 0.36666667461395264, 0.4000000059604645, 0.13333334028720856, 0.3444444537162781, 0.42222222685813904, 0.35555556416511536, 0.2888889014720917, 0.3888888955116272, 0.41111111640930176, 0.42222222685813904, 0.4555555582046509, 0.6222222447395325, NaN, 0.47777777910232544, 0.5444444417953491, 0.5777778029441833, 0.47777777910232544, 0.5666666626930237, 0.5555555820465088, 0.6222222447395325, 0.5, 0.6222222447395325, 0.5333333611488342, 0.644444465637207, 0.6222222447395325, 0.6777777671813965, 0.5555555820465088, 0.6555555462837219, 0.6333333253860474, 0.6333333253860474, 0.855555534362793, 0.6777777671813965, 0.699999988079071, 0.6888889074325562, 0.6777777671813965, 0.800000011920929, 0.7222222089767456, 0.800000011920929, 0.7888888716697693, 0.7555555701255798, 0.9555555582046509, 0.9111111164093018, 0.9777777791023254, 1.0, 0.9222221970558167, 0.8111110925674438, 0.9222221970558167, 0.8999999761581421, 1.0, 0.9666666388511658, 1.0, 1.0], [0.0, 0.0, 0.04444444552063942, 0.08888889104127884, 0.03333333507180214, 0.10000000149011612, 0.17777778208255768, 0.18888889253139496, 0.17777778208255768, 0.17777778208255768, 0.15555556118488312, 0.13333334028720856, 0.14444445073604584, 0.17777778208255768, 0.17777778208255768, 0.15555556118488312, 0.24444444477558136, 0.2222222238779068, 0.18888889253139496, 0.2666666805744171, 0.2888889014720917, 0.23333333432674408, 0.2666666805744171, 0.2777777910232544, 0.36666667461395264, 0.36666667461395264, 0.42222222685813904, 0.41111111640930176, 0.42222222685813904, 0.46666666865348816, 0.4000000059604645, 0.25555557012557983, 0.5111111402511597, 0.3888888955116272, 0.5222222208976746, NaN, 0.46666666865348816, 0.4444444477558136, 0.6000000238418579, 0.6222222447395325, 0.6222222447395325, 0.5555555820465088, 0.7444444298744202, 0.5888888835906982, 0.6333333253860474, 0.6333333253860474, 0.5888888835906982, 0.5888888835906982, 0.6666666865348816, 0.6777777671813965, 0.6888889074325562, 0.8222222328186035, 0.6000000238418579, 0.7333333492279053, 0.7555555701255798, 0.8444444537162781, 0.7444444298744202, 0.9444444179534912, 0.855555534362793, 0.9444444179534912, 0.8888888955116272, 0.8888888955116272, 0.8666666746139526, 0.8444444537162781, 0.855555534362793, 0.8777777552604675, 0.8333333134651184, 0.9222221970558167, 1.0, 0.8111110925674438, 0.9888888597488403, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.07777778059244156, 0.011111111380159855, 0.07777778059244156, 0.0555555559694767, 0.23333333432674408, 0.21111111342906952, 0.1666666716337204, 0.20000000298023224, 0.18888889253139496, 0.20000000298023224, 0.17777778208255768, 0.10000000149011612, 0.25555557012557983, 0.03333333507180214, 0.17777778208255768, 0.21111111342906952, 0.2222222238779068, 0.3222222328186035, 0.23333333432674408, 0.15555556118488312, 0.24444444477558136, 0.3222222328186035, 0.41111111640930176, 0.3888888955116272, 0.3777777850627899, 0.3777777850627899, 0.4333333373069763, 0.4000000059604645, 0.3777777850627899, 0.4000000059604645, 0.46666666865348816, 0.4444444477558136, 0.4555555582046509, 0.5333333611488342, NaN, 0.4888888895511627, 0.6111111044883728, 0.6000000238418579, 0.5111111402511597, 0.6555555462837219, 0.5777778029441833, 0.6111111044883728, 0.6666666865348816, 0.6111111044883728, 0.6222222447395325, 0.5777778029441833, 0.6666666865348816, 0.7222222089767456, 0.644444465637207, 0.6222222447395325, 0.6222222447395325, 0.7777777910232544, 0.6777777671813965, 0.7777777910232544, 0.8111110925674438, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8444444537162781, 0.7888888716697693, 0.9777777791023254, 0.8222222328186035, 0.9777777791023254, 0.9777777791023254, 0.9444444179534912, 0.8999999761581421, 1.0, 0.8999999761581421, 0.9555555582046509, 1.0, 1.0, 1.0], [0.011111111380159855, 0.08888889104127884, 0.03333333507180214, 0.20000000298023224, 0.18888889253139496, 0.2777777910232544, 0.0555555559694767, 0.04444444552063942, 0.04444444552063942, 0.06666667014360428, 0.08888889104127884, 0.1666666716337204, 0.07777778059244156, 0.1111111119389534, 0.2888889014720917, 0.3444444537162781, 0.2222222238779068, 0.10000000149011612, 0.18888889253139496, 0.31111112236976624, 0.20000000298023224, 0.3444444537162781, 0.2777777910232544, 0.12222222238779068, 0.31111112236976624, 0.30000001192092896, 0.2888889014720917, 0.3777777850627899, 0.3444444537162781, 0.3444444537162781, 0.3444444537162781, 0.41111111640930176, 0.3222222328186035, 0.3222222328186035, 0.42222222685813904, 0.5555555820465088, 0.5111111402511597, NaN, 0.6000000238418579, 0.6333333253860474, 0.5444444417953491, 0.6666666865348816, 0.5222222208976746, 0.5111111402511597, 0.6222222447395325, 0.47777777910232544, 0.4555555582046509, 0.6222222447395325, 0.6555555462837219, 0.5666666626930237, 0.6555555462837219, 0.800000011920929, 0.5666666626930237, 0.8222222328186035, 0.7444444298744202, 0.7333333492279053, 0.8666666746139526, 0.8333333134651184, 0.9555555582046509, 0.8444444537162781, 0.9777777791023254, 0.9888888597488403, 0.855555534362793, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.9444444179534912, 1.0, 0.9888888597488403, 0.9888888597488403, 0.9555555582046509, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.03333333507180214, 0.0555555559694767, 0.0, 0.08888889104127884, 0.1666666716337204, 0.17777778208255768, 0.04444444552063942, 0.15555556118488312, 0.15555556118488312, 0.17777778208255768, 0.14444445073604584, 0.0, 0.17777778208255768, 0.02222222276031971, 0.1111111119389534, 0.15555556118488312, 0.17777778208255768, 0.23333333432674408, 0.10000000149011612, 0.10000000149011612, 0.23333333432674408, 0.24444444477558136, 0.2888889014720917, 0.3222222328186035, 0.3444444537162781, 0.2777777910232544, 0.30000001192092896, 0.35555556416511536, 0.3444444537162781, 0.2777777910232544, 0.4555555582046509, 0.3777777850627899, 0.5222222208976746, 0.4000000059604645, 0.3888888955116272, 0.4000000059604645, NaN, 0.4555555582046509, 0.5777778029441833, 0.5, 0.4333333373069763, 0.5666666626930237, 0.6000000238418579, 0.6111111044883728, 0.5777778029441833, 0.6111111044883728, 0.6666666865348816, 0.6333333253860474, 0.6888889074325562, 0.6222222447395325, 0.6333333253860474, 0.8222222328186035, 0.7111111283302307, 0.7888888716697693, 0.8444444537162781, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8777777552604675, 0.9555555582046509, 0.855555534362793, 0.9444444179534912, 0.8666666746139526, 0.9555555582046509, 0.9111111164093018, 1.0, 0.8777777552604675, 0.9888888597488403, 1.0, 1.0, 1.0], [0.0, 0.0, 0.08888889104127884, 0.0555555559694767, 0.10000000149011612, 0.07777778059244156, 0.15555556118488312, 0.14444445073604584, 0.13333334028720856, 0.14444445073604584, 0.12222222238779068, 0.13333334028720856, 0.12222222238779068, 0.14444445073604584, 0.20000000298023224, 0.08888889104127884, 0.17777778208255768, 0.14444445073604584, 0.14444445073604584, 0.24444444477558136, 0.2222222238779068, 0.20000000298023224, 0.21111111342906952, 0.24444444477558136, 0.21111111342906952, 0.31111112236976624, 0.3444444537162781, 0.31111112236976624, 0.3888888955116272, 0.3777777850627899, 0.4000000059604645, 0.31111112236976624, 0.41111111640930176, 0.36666667461395264, 0.4333333373069763, 0.3777777850627899, 0.4000000059604645, 0.36666667461395264, 0.5444444417953491, NaN, 0.5, 0.6000000238418579, 0.5222222208976746, 0.6111111044883728, 0.5555555820465088, 0.5444444417953491, 0.5888888835906982, 0.6555555462837219, 0.5888888835906982, 0.6000000238418579, 0.6111111044883728, 0.644444465637207, 0.5777778029441833, 0.7333333492279053, 0.6333333253860474, 0.6666666865348816, 0.7666666507720947, 0.9444444179534912, 0.7555555701255798, 0.9555555582046509, 0.9111111164093018, 0.7666666507720947, 0.8666666746139526, 0.7666666507720947, 0.8666666746139526, 0.855555534362793, 0.855555534362793, 0.9444444179534912, 1.0, 0.8111110925674438, 0.9666666388511658, 1.0, 0.8999999761581421, 1.0], [0.02222222276031971, 0.0, 0.02222222276031971, 0.03333333507180214, 0.04444444552063942, 0.08888889104127884, 0.04444444552063942, 0.04444444552063942, 0.04444444552063942, 0.0555555559694767, 0.15555556118488312, 0.08888889104127884, 0.13333334028720856, 0.1666666716337204, 0.17777778208255768, 0.2222222238779068, 0.15555556118488312, 0.17777778208255768, 0.17777778208255768, 0.18888889253139496, 0.18888889253139496, 0.25555557012557983, 0.30000001192092896, 0.25555557012557983, 0.20000000298023224, 0.3222222328186035, 0.3888888955116272, 0.2777777910232544, 0.2666666805744171, 0.2777777910232544, 0.12222222238779068, 0.21111111342906952, 0.3333333432674408, 0.42222222685813904, 0.4444444477558136, 0.3777777850627899, 0.4888888895511627, 0.4555555582046509, 0.42222222685813904, 0.5, NaN, 0.5111111402511597, 0.4000000059604645, 0.4888888895511627, 0.5444444417953491, 0.644444465637207, 0.5777778029441833, 0.5222222208976746, 0.6222222447395325, 0.6555555462837219, 0.6111111044883728, 0.6555555462837219, 0.6222222447395325, 0.644444465637207, 0.7333333492279053, 0.8888888955116272, 0.6888889074325562, 0.8444444537162781, 0.8777777552604675, 0.8444444537162781, 0.8444444537162781, 0.8777777552604675, 0.9555555582046509, 0.9222221970558167, 0.9555555582046509, 0.9777777791023254, 0.8999999761581421, 0.9222221970558167, 0.9777777791023254, 0.8888888955116272, 0.9888888597488403, 1.0, 1.0, 1.0], [0.0, 0.08888889104127884, 0.03333333507180214, 0.10000000149011612, 0.13333334028720856, 0.15555556118488312, 0.13333334028720856, 0.1666666716337204, 0.03333333507180214, 0.17777778208255768, 0.20000000298023224, 0.23333333432674408, 0.14444445073604584, 0.02222222276031971, 0.2666666805744171, 0.17777778208255768, 0.13333334028720856, 0.18888889253139496, 0.24444444477558136, 0.30000001192092896, 0.12222222238779068, 0.20000000298023224, 0.24444444477558136, 0.18888889253139496, 0.3333333432674408, 0.31111112236976624, 0.2666666805744171, 0.3333333432674408, 0.35555556416511536, 0.31111112236976624, 0.3444444537162781, 0.3333333432674408, 0.31111112236976624, 0.2666666805744171, 0.3777777850627899, 0.4444444477558136, 0.3444444537162781, 0.3333333432674408, 0.5, 0.4000000059604645, 0.4888888895511627, NaN, 0.47777777910232544, 0.5222222208976746, 0.5777778029441833, 0.4444444477558136, 0.5222222208976746, 0.5555555820465088, 0.6000000238418579, 0.6222222447395325, 0.5888888835906982, 0.6111111044883728, 0.6222222447395325, 0.8444444537162781, 0.6222222447395325, 0.6777777671813965, 0.9666666388511658, 0.6777777671813965, 0.8888888955116272, 0.6777777671813965, 0.8777777552604675, 0.8999999761581421, 0.855555534362793, 0.8333333134651184, 0.8666666746139526, 0.7333333492279053, 0.9333333373069763, 0.9777777791023254, 0.8666666746139526, 1.0, 0.9333333373069763, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.04444444552063942, 0.04444444552063942, 0.03333333507180214, 0.06666667014360428, 0.0555555559694767, 0.1111111119389534, 0.08888889104127884, 0.03333333507180214, 0.02222222276031971, 0.03333333507180214, 0.06666667014360428, 0.15555556118488312, 0.13333334028720856, 0.1666666716337204, 0.17777778208255768, 0.14444445073604584, 0.07777778059244156, 0.1666666716337204, 0.2666666805744171, 0.21111111342906952, 0.14444445073604584, 0.23333333432674408, 0.2777777910232544, 0.3222222328186035, 0.4000000059604645, 0.35555556416511536, 0.2888889014720917, 0.4333333373069763, 0.36666667461395264, 0.07777778059244156, 0.35555556416511536, 0.3777777850627899, 0.5, 0.25555557012557983, 0.42222222685813904, 0.47777777910232544, 0.5666666626930237, 0.47777777910232544, 0.6000000238418579, 0.5222222208976746, NaN, 0.5777778029441833, 0.5555555820465088, 0.644444465637207, 0.5555555820465088, 0.5888888835906982, 0.6111111044883728, 0.644444465637207, 0.6777777671813965, 0.6777777671813965, 0.6000000238418579, 0.7333333492279053, 0.7555555701255798, 0.8333333134651184, 0.7333333492279053, 0.8444444537162781, 0.8111110925674438, 0.8444444537162781, 0.8777777552604675, 0.8444444537162781, 0.8666666746139526, 0.8222222328186035, 0.855555534362793, 0.8666666746139526, 0.8222222328186035, 0.9111111164093018, 1.0, 0.800000011920929, 0.9777777791023254, 1.0, 0.8999999761581421, 1.0], [0.04444444552063942, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.04444444552063942, 0.0555555559694767, 0.18888889253139496, 0.18888889253139496, 0.08888889104127884, 0.20000000298023224, 0.15555556118488312, 0.20000000298023224, 0.1666666716337204, 0.18888889253139496, 0.1111111119389534, 0.12222222238779068, 0.12222222238779068, 0.18888889253139496, 0.20000000298023224, 0.17777778208255768, 0.2666666805744171, 0.23333333432674408, 0.25555557012557983, 0.1666666716337204, 0.2222222238779068, 0.14444445073604584, 0.21111111342906952, 0.2222222238779068, 0.1666666716337204, 0.1111111119389534, 0.2888889014720917, 0.30000001192092896, 0.3444444537162781, 0.5333333611488342, 0.3777777850627899, 0.41111111640930176, 0.3888888955116272, 0.4888888895511627, 0.4333333373069763, 0.3888888955116272, 0.5111111402511597, 0.47777777910232544, 0.42222222685813904, NaN, 0.4888888895511627, 0.4444444477558136, 0.5666666626930237, 0.6222222447395325, 0.5888888835906982, 0.46666666865348816, 0.5777778029441833, 0.5222222208976746, 0.75, 0.6777777671813965, 0.644444465637207, 0.6777777671813965, 0.7777777910232544, 0.6666666865348816, 0.6888889074325562, 0.6777777671813965, 0.800000011920929, 0.7444444298744202, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 0.9111111164093018, 0.9333333373069763, 0.800000011920929, 0.9111111164093018, 0.8999999761581421, 0.8999999761581421, 0.9777777791023254, 1.0, 1.0], [0.0, 0.02222222276031971, 0.011111111380159855, 0.0555555559694767, 0.04444444552063942, 0.04444444552063942, 0.1111111119389534, 0.12222222238779068, 0.12222222238779068, 0.1111111119389534, 0.10000000149011612, 0.10000000149011612, 0.1111111119389534, 0.12222222238779068, 0.08888889104127884, 0.1666666716337204, 0.21111111342906952, 0.13333334028720856, 0.13333334028720856, 0.14444445073604584, 0.15555556118488312, 0.2222222238779068, 0.21111111342906952, 0.15555556118488312, 0.25555557012557983, 0.2888889014720917, 0.2666666805744171, 0.2777777910232544, 0.3222222328186035, 0.3333333432674408, 0.30000001192092896, 0.2777777910232544, 0.24444444477558136, 0.35555556416511536, 0.46666666865348816, 0.36666667461395264, 0.3333333432674408, 0.3777777850627899, 0.4000000059604645, 0.4444444477558136, 0.4555555582046509, 0.42222222685813904, 0.4444444477558136, 0.5111111402511597, NaN, 0.5111111402511597, 0.4444444477558136, 0.5333333611488342, 0.7444444298744202, 0.47777777910232544, 0.6333333253860474, 0.6111111044883728, 0.5555555820465088, 0.6333333253860474, 0.6888889074325562, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8111110925674438, 0.8888888955116272, 0.855555534362793, 0.855555534362793, 0.9666666388511658, 0.8333333134651184, 0.9333333373069763, 0.9666666388511658, 0.9555555582046509, 0.9333333373069763, 0.8888888955116272, 0.9666666388511658, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.03333333507180214, 0.011111111380159855, 0.0555555559694767, 0.07777778059244156, 0.07777778059244156, 0.0555555559694767, 0.0, 0.06666667014360428, 0.03333333507180214, 0.02222222276031971, 0.04444444552063942, 0.18888889253139496, 0.14444445073604584, 0.1666666716337204, 0.15555556118488312, 0.10000000149011612, 0.06666667014360428, 0.17777778208255768, 0.2666666805744171, 0.24444444477558136, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.31111112236976624, 0.36666667461395264, 0.24444444477558136, 0.2777777910232544, 0.18888889253139496, 0.4555555582046509, 0.35555556416511536, 0.36666667461395264, 0.3888888955116272, 0.5222222208976746, 0.3888888955116272, 0.4555555582046509, 0.35555556416511536, 0.5555555820465088, 0.35555556416511536, 0.5555555820465088, 0.4888888895511627, NaN, 0.4888888895511627, 0.5666666626930237, 0.5, 0.5666666626930237, 0.4888888895511627, 0.6777777671813965, 0.5777778029441833, 0.7222222089767456, 0.644444465637207, 0.644444465637207, 0.7666666507720947, 0.8888888955116272, 0.7555555701255798, 0.8999999761581421, 0.8777777552604675, 0.800000011920929, 0.7555555701255798, 0.855555534362793, 0.7444444298744202, 0.9222221970558167, 0.8444444537162781, 0.8888888955116272, 0.9888888597488403, 0.8888888955116272, 0.9666666388511658, 0.8999999761581421, 0.800000011920929, 1.0], [0.0, 0.0, 0.04444444552063942, 0.011111111380159855, 0.0555555559694767, 0.0555555559694767, 0.04444444552063942, 0.06666667014360428, 0.07777778059244156, 0.07777778059244156, 0.03333333507180214, 0.02222222276031971, 0.12222222238779068, 0.06666667014360428, 0.07777778059244156, 0.0, 0.08888889104127884, 0.18888889253139496, 0.06666667014360428, 0.1111111119389534, 0.17777778208255768, 0.13333334028720856, 0.03333333507180214, 0.24444444477558136, 0.12222222238779068, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.21111111342906952, 0.2222222238779068, 0.24444444477558136, 0.3222222328186035, 0.46666666865348816, 0.5333333611488342, 0.3777777850627899, 0.41111111640930176, 0.3777777850627899, 0.5444444417953491, 0.42222222685813904, 0.41111111640930176, 0.42222222685813904, 0.47777777910232544, 0.4444444477558136, 0.4333333373069763, 0.5555555820465088, 0.5111111402511597, NaN, 0.6000000238418579, 0.5777778029441833, 0.5, 0.5555555820465088, 0.4888888895511627, 0.5666666626930237, 0.6000000238418579, 0.644444465637207, 0.7444444298744202, 0.699999988079071, 0.8444444537162781, 0.6888889074325562, 0.855555534362793, 0.7666666507720947, 0.699999988079071, 0.8777777552604675, 0.8888888955116272, 0.9555555582046509, 0.9444444179534912, 0.9444444179534912, 0.8111110925674438, 1.0, 0.8999999761581421, 0.9222221970558167, 1.0, 1.0, 1.0], [0.1111111119389534, 0.14444445073604584, 0.12222222238779068, 0.20000000298023224, 0.2222222238779068, 0.18888889253139496, 0.2777777910232544, 0.24444444477558136, 0.24444444477558136, 0.2777777910232544, 0.3222222328186035, 0.31111112236976624, 0.25555557012557983, 0.15555556118488312, 0.3333333432674408, 0.23333333432674408, 0.14444445073604584, 0.2666666805744171, 0.3444444537162781, 0.3444444537162781, 0.20000000298023224, 0.23333333432674408, 0.35555556416511536, 0.2666666805744171, 0.3333333432674408, 0.31111112236976624, 0.23333333432674408, 0.2777777910232544, 0.23333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3888888955116272, 0.31111112236976624, 0.3222222328186035, 0.41111111640930176, 0.42222222685813904, 0.3777777850627899, 0.3888888955116272, 0.3444444537162781, 0.47777777910232544, 0.4444444477558136, 0.41111111640930176, 0.3777777850627899, 0.46666666865348816, 0.4333333373069763, 0.4000000059604645, NaN, 0.47777777910232544, 0.46666666865348816, 0.5222222208976746, 0.41111111640930176, 0.5444444417953491, 0.5777778029441833, 0.5555555820465088, 0.4888888895511627, 0.6111111044883728, 0.5222222208976746, 0.644444465637207, 0.5333333611488342, 0.5666666626930237, 0.6555555462837219, 0.6333333253860474, 0.7555555701255798, 0.6777777671813965, 0.6333333253860474, 0.7111111283302307, 0.6555555462837219, 0.6222222447395325, 0.7777777910232544, 0.6555555462837219, 0.6777777671813965, 0.7888888716697693, 0.8888888955116272], [0.011111111380159855, 0.02222222276031971, 0.02222222276031971, 0.03333333507180214, 0.011111111380159855, 0.03333333507180214, 0.12222222238779068, 0.12222222238779068, 0.1111111119389534, 0.12222222238779068, 0.10000000149011612, 0.08888889104127884, 0.1111111119389534, 0.13333334028720856, 0.04444444552063942, 0.1666666716337204, 0.2222222238779068, 0.13333334028720856, 0.14444445073604584, 0.06666667014360428, 0.17777778208255768, 0.21111111342906952, 0.15555556118488312, 0.18888889253139496, 0.20000000298023224, 0.2777777910232544, 0.24444444477558136, 0.2888889014720917, 0.24444444477558136, 0.3444444537162781, 0.24444444477558136, 0.23333333432674408, 0.2222222238779068, 0.3444444537162781, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3444444537162781, 0.3333333432674408, 0.41111111640930176, 0.3777777850627899, 0.4000000059604645, 0.3888888955116272, 0.41111111640930176, 0.25555557012557983, 0.5, 0.42222222685813904, 0.5222222208976746, NaN, 0.41111111640930176, 0.5444444417953491, 0.5777778029441833, 0.5111111402511597, 0.5666666626930237, 0.6333333253860474, 0.5777778029441833, 0.6666666865348816, 0.7888888716697693, 0.7111111283302307, 0.800000011920929, 0.855555534362793, 0.7777777910232544, 0.8333333134651184, 0.9388889074325562, 0.8333333134651184, 0.9222221970558167, 0.9222221970558167, 0.9111111164093018, 0.9333333373069763, 0.855555534362793, 0.9777777791023254, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.011111111380159855, 0.02222222276031971, 0.02222222276031971, 0.04444444552063942, 0.04444444552063942, 0.011111111380159855, 0.02222222276031971, 0.06666667014360428, 0.011111111380159855, 0.02222222276031971, 0.08888889104127884, 0.14444445073604584, 0.17777778208255768, 0.10000000149011612, 0.04444444552063942, 0.14444445073604584, 0.03333333507180214, 0.21111111342906952, 0.24444444477558136, 0.1666666716337204, 0.06666667014360428, 0.18888889253139496, 0.20000000298023224, 0.13333334028720856, 0.15555556118488312, 0.10000000149011612, 0.30000001192092896, 0.21111111342906952, 0.2666666805744171, 0.2777777910232544, 0.2777777910232544, 0.4333333373069763, 0.3444444537162781, 0.3222222328186035, 0.2777777910232544, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.3444444537162781, 0.3777777850627899, 0.35555556416511536, 0.5333333611488342, 0.5222222208976746, 0.4333333373069763, 0.5, 0.5333333611488342, 0.5888888835906982, NaN, 0.46666666865348816, 0.5777778029441833, 0.5666666626930237, 0.5444444417953491, 0.5555555820465088, 0.7555555701255798, 0.6777777671813965, 0.7444444298744202, 0.699999988079071, 0.7555555701255798, 0.8333333134651184, 0.7444444298744202, 0.8444444537162781, 0.9555555582046509, 0.8666666746139526, 0.9111111164093018, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8999999761581421, 0.9222221970558167, 1.0, 0.9222221970558167, 1.0], [0.0, 0.011111111380159855, 0.011111111380159855, 0.03333333507180214, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.03333333507180214, 0.04444444552063942, 0.04444444552063942, 0.011111111380159855, 0.02222222276031971, 0.1111111119389534, 0.10000000149011612, 0.03333333507180214, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.04444444552063942, 0.08888889104127884, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.23333333432674408, 0.07777778059244156, 0.20000000298023224, 0.25555557012557983, 0.15555556118488312, 0.23333333432674408, 0.3222222328186035, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.36666667461395264, 0.31111112236976624, 0.35555556416511536, 0.3444444537162781, 0.31111112236976624, 0.3888888955116272, 0.3888888955116272, 0.41111111640930176, 0.3222222328186035, 0.42222222685813904, 0.36666667461395264, 0.5111111402511597, 0.4444444477558136, 0.47777777910232544, 0.4555555582046509, 0.5333333611488342, NaN, 0.4888888895511627, 0.5333333611488342, 0.5333333611488342, 0.7555555701255798, 0.644444465637207, 0.644444465637207, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.7444444298744202, 0.7888888716697693, 0.7555555701255798, 0.9888888597488403, 0.800000011920929, 0.9444444179534912, 0.9777777791023254, 0.8999999761581421, 0.855555534362793, 0.9111111164093018, 0.9666666388511658, 0.8999999761581421, 1.0, 1.0], [0.0, 0.02222222276031971, 0.03333333507180214, 0.07777778059244156, 0.04444444552063942, 0.07777778059244156, 0.02222222276031971, 0.04444444552063942, 0.02222222276031971, 0.03333333507180214, 0.04444444552063942, 0.04444444552063942, 0.02222222276031971, 0.1111111119389534, 0.14444445073604584, 0.12222222238779068, 0.14444445073604584, 0.06666667014360428, 0.0555555559694767, 0.1666666716337204, 0.21111111342906952, 0.20000000298023224, 0.17777778208255768, 0.04444444552063942, 0.10000000149011612, 0.2222222238779068, 0.15555556118488312, 0.21111111342906952, 0.30000001192092896, 0.25555557012557983, 0.17777778208255768, 0.12222222238779068, 0.1666666716337204, 0.31111112236976624, 0.36666667461395264, 0.17777778208255768, 0.3777777850627899, 0.20000000298023224, 0.3777777850627899, 0.35555556416511536, 0.3444444537162781, 0.3888888955116272, 0.3222222328186035, 0.47777777910232544, 0.3888888955116272, 0.3222222328186035, 0.5111111402511597, 0.5888888835906982, 0.42222222685813904, 0.42222222685813904, 0.5111111402511597, NaN, 0.5777778029441833, 0.6111111044883728, 0.6777777671813965, 0.644444465637207, 0.7111111283302307, 0.6666666865348816, 0.8111110925674438, 0.6888889074325562, 0.8888888955116272, 0.8444444537162781, 0.8444444537162781, 0.8222222328186035, 0.8666666746139526, 0.8666666746139526, 0.855555534362793, 0.9777777791023254, 0.9555555582046509, 0.8333333134651184, 0.9666666388511658, 1.0, 0.8999999761581421, 1.0], [0.02222222276031971, 0.02222222276031971, 0.04444444552063942, 0.03333333507180214, 0.0555555559694767, 0.04444444552063942, 0.1111111119389534, 0.10000000149011612, 0.06666667014360428, 0.12222222238779068, 0.1111111119389534, 0.04444444552063942, 0.1666666716337204, 0.17777778208255768, 0.10000000149011612, 0.12222222238779068, 0.07777778059244156, 0.2222222238779068, 0.14444445073604584, 0.15555556118488312, 0.25555557012557983, 0.20000000298023224, 0.1666666716337204, 0.17777778208255768, 0.14444445073604584, 0.07777778059244156, 0.14444445073604584, 0.13333334028720856, 0.08888889104127884, 0.0555555559694767, 0.23333333432674408, 0.3222222328186035, 0.3444444537162781, 0.41111111640930176, 0.14444445073604584, 0.4000000059604645, 0.3777777850627899, 0.4333333373069763, 0.36666667461395264, 0.42222222685813904, 0.3777777850627899, 0.3777777850627899, 0.4000000059604645, 0.25, 0.4444444477558136, 0.42222222685813904, 0.4333333373069763, 0.4555555582046509, 0.4888888895511627, 0.4333333373069763, 0.46666666865348816, 0.42222222685813904, NaN, 0.5111111402511597, 0.5, 0.6222222447395325, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.6111111044883728, 0.7111111283302307, 0.7888888716697693, 0.6777777671813965, 0.8333333134651184, 0.7222222089767456, 0.6777777671813965, 0.7444444298744202, 0.7111111283302307, 0.7555555701255798, 0.855555534362793, 0.7333333492279053, 0.9777777791023254, 1.0], [0.0, 0.02222222276031971, 0.02222222276031971, 0.06666667014360428, 0.0555555559694767, 0.06666667014360428, 0.03333333507180214, 0.02222222276031971, 0.03333333507180214, 0.04444444552063942, 0.12222222238779068, 0.08888889104127884, 0.1111111119389534, 0.02222222276031971, 0.12222222238779068, 0.08888889104127884, 0.06666667014360428, 0.14444445073604584, 0.13333334028720856, 0.12222222238779068, 0.03333333507180214, 0.12222222238779068, 0.21111111342906952, 0.13333334028720856, 0.20000000298023224, 0.2888889014720917, 0.14444445073604584, 0.07777778059244156, 0.25555557012557983, 0.20000000298023224, 0.1666666716337204, 0.24444444477558136, 0.2888889014720917, 0.20000000298023224, 0.3222222328186035, 0.2666666805744171, 0.2222222238779068, 0.17777778208255768, 0.17777778208255768, 0.2666666805744171, 0.35555556416511536, 0.15555556118488312, 0.2666666805744171, 0.3222222328186035, 0.36666667461395264, 0.2777777910232544, 0.4000000059604645, 0.42222222685813904, 0.4333333373069763, 0.4555555582046509, 0.46666666865348816, 0.3888888955116272, 0.4888888895511627, NaN, 0.5111111402511597, 0.4888888895511627, 0.7111111283302307, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.7111111283302307, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.6777777671813965, 0.8222222328186035, 0.8999999761581421, 0.8777777552604675, 0.8888888955116272, 0.8666666746139526, 0.8888888955116272, 0.9555555582046509, 1.0], [0.0, 0.011111111380159855, 0.0, 0.03333333507180214, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.03333333507180214, 0.011111111380159855, 0.02222222276031971, 0.1111111119389534, 0.12222222238779068, 0.03333333507180214, 0.13333334028720856, 0.1111111119389534, 0.13333334028720856, 0.02222222276031971, 0.06666667014360428, 0.14444445073604584, 0.1666666716337204, 0.07777778059244156, 0.15555556118488312, 0.0555555559694767, 0.1666666716337204, 0.13333334028720856, 0.14444445073604584, 0.17777778208255768, 0.21111111342906952, 0.13333334028720856, 0.15555556118488312, 0.10000000149011612, 0.2777777910232544, 0.30000001192092896, 0.24444444477558136, 0.3222222328186035, 0.25555557012557983, 0.2888889014720917, 0.36666667461395264, 0.2666666805744171, 0.3777777850627899, 0.24444444477558136, 0.35555556416511536, 0.31111112236976624, 0.35555556416511536, 0.35555556416511536, 0.4444444477558136, 0.36666667461395264, 0.4444444477558136, 0.24444444477558136, 0.3222222328186035, 0.5, 0.4888888895511627, NaN, 0.5555555820465088, 0.5444444417953491, 0.7111111283302307, 0.7111111283302307, 0.699999988079071, 0.6888889074325562, 0.7444444298744202, 0.7111111283302307, 0.9555555582046509, 0.7777777910232544, 0.9333333373069763, 0.9444444179534912, 0.8888888955116272, 0.8444444537162781, 0.8999999761581421, 0.9666666388511658, 0.9111111164093018, 0.9888888597488403, 1.0], [0.0, 0.0, 0.011111111380159855, 0.03333333507180214, 0.0, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.08888889104127884, 0.011111111380159855, 0.04444444552063942, 0.02222222276031971, 0.1111111119389534, 0.08888889104127884, 0.0, 0.08888889104127884, 0.06666667014360428, 0.02222222276031971, 0.06666667014360428, 0.13333334028720856, 0.04444444552063942, 0.20000000298023224, 0.1111111119389534, 0.14444445073604584, 0.20000000298023224, 0.2222222238779068, 0.011111111380159855, 0.13333334028720856, 0.07777778059244156, 0.24444444477558136, 0.31111112236976624, 0.15555556118488312, 0.2222222238779068, 0.2666666805744171, 0.21111111342906952, 0.3333333432674408, 0.1111111119389534, 0.3222222328186035, 0.1666666716337204, 0.3222222328186035, 0.30000001192092896, 0.35555556416511536, 0.25555557012557983, 0.5111111402511597, 0.42222222685813904, 0.24444444477558136, 0.35555556416511536, 0.35555556416511536, 0.3777777850627899, 0.5111111402511597, 0.4444444477558136, NaN, 0.5555555820465088, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.7888888716697693, 0.8222222328186035, 0.9222221970558167, 0.8666666746139526, 0.8777777552604675, 0.8999999761581421, 0.8888888955116272, 0.9555555582046509, 0.9888888597488403, 0.9555555582046509, 1.0], [0.0, 0.02222222276031971, 0.011111111380159855, 0.06666667014360428, 0.07777778059244156, 0.07777778059244156, 0.02222222276031971, 0.04444444552063942, 0.0, 0.0555555559694767, 0.10000000149011612, 0.13333334028720856, 0.12222222238779068, 0.03333333507180214, 0.10000000149011612, 0.10000000149011612, 0.03333333507180214, 0.13333334028720856, 0.14444445073604584, 0.1111111119389534, 0.04444444552063942, 0.08888889104127884, 0.18888889253139496, 0.13333334028720856, 0.15555556118488312, 0.21111111342906952, 0.10000000149011612, 0.08888889104127884, 0.12222222238779068, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.13333334028720856, 0.3222222328186035, 0.25555557012557983, 0.18888889253139496, 0.13333334028720856, 0.15555556118488312, 0.23333333432674408, 0.31111112236976624, 0.03333333507180214, 0.2666666805744171, 0.2222222238779068, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.3888888955116272, 0.3333333432674408, 0.3222222328186035, 0.35555556416511536, 0.2888889014720917, 0.36666667461395264, 0.2888889014720917, 0.4555555582046509, 0.4444444477558136, NaN, 0.47777777910232544, 0.5555555820465088, 0.4888888895511627, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.6111111044883728, 0.7222222089767456, 0.6333333253860474, 0.6222222447395325, 0.6777777671813965, 0.699999988079071, 0.8111110925674438, 0.800000011920929, 0.8111110925674438, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.07777778059244156, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.0, 0.0, 0.011111111380159855, 0.0, 0.06666667014360428, 0.0555555559694767, 0.04444444552063942, 0.1111111119389534, 0.14444445073604584, 0.07777778059244156, 0.02222222276031971, 0.0555555559694767, 0.14444445073604584, 0.20000000298023224, 0.0555555559694767, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.0555555559694767, 0.15555556118488312, 0.3222222328186035, 0.15555556118488312, 0.3333333432674408, 0.20000000298023224, 0.1111111119389534, 0.15555556118488312, 0.47777777910232544, 0.21111111342906952, 0.25555557012557983, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2888889014720917, 0.2666666805744171, 0.5222222208976746, NaN, 0.5666666626930237, 0.6555555462837219, 0.6555555462837219, 0.5777778029441833, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.644444465637207, 0.699999988079071, 0.7444444298744202, 1.0, 0.699999988079071, 0.8777777552604675, 0.8777777552604675, 0.800000011920929, 0.8999999761581421], [0.0, 0.02222222276031971, 0.0, 0.07777778059244156, 0.03333333507180214, 0.07777778059244156, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.02222222276031971, 0.0, 0.0, 0.15555556118488312, 0.06666667014360428, 0.10000000149011612, 0.0, 0.0, 0.17777778208255768, 0.02222222276031971, 0.1111111119389534, 0.13333334028720856, 0.0, 0.15555556118488312, 0.24444444477558136, 0.10000000149011612, 0.13333334028720856, 0.2666666805744171, 0.24444444477558136, 0.08888889104127884, 0.10000000149011612, 0.12222222238779068, 0.02222222276031971, 0.2777777910232544, 0.14444445073604584, 0.20000000298023224, 0.04444444552063942, 0.1666666716337204, 0.24444444477558136, 0.12222222238779068, 0.1111111119389534, 0.18888889253139496, 0.31111112236976624, 0.20000000298023224, 0.24444444477558136, 0.31111112236976624, 0.35555556416511536, 0.2888889014720917, 0.30000001192092896, 0.2777777910232544, 0.18888889253139496, 0.30000001192092896, 0.3333333432674408, 0.2888889014720917, 0.2666666805744171, 0.4444444477558136, 0.4333333373069763, NaN, 0.4333333373069763, 0.42222222685813904, 0.8166666626930237, 0.4444444477558136, 0.5555555820465088, 0.47777777910232544, 0.5888888835906982, 0.5444444417953491, 0.9333333373069763, 0.5, 0.7777777910232544, 0.6888889074325562, 0.6222222447395325, 0.7666666507720947, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.07777778059244156, 0.011111111380159855, 0.011111111380159855, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0555555559694767, 0.0555555559694767, 0.04444444552063942, 0.1111111119389534, 0.14444445073604584, 0.07777778059244156, 0.03333333507180214, 0.04444444552063942, 0.14444445073604584, 0.20000000298023224, 0.0555555559694767, 0.10000000149011612, 0.15555556118488312, 0.13333334028720856, 0.04444444552063942, 0.15555556118488312, 0.3222222328186035, 0.15555556118488312, 0.3222222328186035, 0.18888889253139496, 0.10000000149011612, 0.14444445073604584, 0.46666666865348816, 0.20000000298023224, 0.24444444477558136, 0.2777777910232544, 0.31111112236976624, 0.3333333432674408, 0.4444444477558136, 0.30000001192092896, 0.2666666805744171, 0.5111111402511597, 0.3444444537162781, 0.5666666626930237, NaN, 0.6555555462837219, 0.5777778029441833, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.644444465637207, 0.699999988079071, 0.7444444298744202, 1.0, 0.699999988079071, 0.8777777552604675, 0.8777777552604675, 0.800000011920929, 0.8999999761581421], [0.011111111380159855, 0.04444444552063942, 0.011111111380159855, 0.06666667014360428, 0.0555555559694767, 0.04444444552063942, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.08888889104127884, 0.0555555559694767, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.1111111119389534, 0.03333333507180214, 0.1111111119389534, 0.08888889104127884, 0.02222222276031971, 0.10000000149011612, 0.14444445073604584, 0.011111111380159855, 0.04444444552063942, 0.14444445073604584, 0.15555556118488312, 0.08888889104127884, 0.08888889104127884, 0.1111111119389534, 0.03333333507180214, 0.21111111342906952, 0.1111111119389534, 0.15555556118488312, 0.02222222276031971, 0.13333334028720856, 0.08888889104127884, 0.15555556118488312, 0.12222222238779068, 0.12222222238779068, 0.20000000298023224, 0.1111111119389534, 0.12222222238779068, 0.23333333432674408, 0.4333333373069763, 0.14444445073604584, 0.1666666716337204, 0.25555557012557983, 0.1111111119389534, 0.3888888955116272, 0.2888889014720917, 0.31111112236976624, 0.20000000298023224, 0.4000000059604645, 0.3444444537162781, 0.5777778029441833, 0.3444444537162781, NaN, 0.6333333253860474, 0.5333333611488342, 0.6888889074325562, 0.5777778029441833, 0.5777778029441833, 0.5666666626930237, 0.7222222089767456, 0.6000000238418579, 0.7111111283302307, 0.6666666865348816, 1.0, 0.8444444537162781, 1.0], [0.0, 0.02222222276031971, 0.0, 0.06666667014360428, 0.03333333507180214, 0.0555555559694767, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.0, 0.0, 0.14444445073604584, 0.0555555559694767, 0.08888889104127884, 0.0, 0.0, 0.13333334028720856, 0.011111111380159855, 0.10000000149011612, 0.08888889104127884, 0.0, 0.15555556118488312, 0.24444444477558136, 0.06666667014360428, 0.10000000149011612, 0.24444444477558136, 0.24444444477558136, 0.08888889104127884, 0.07777778059244156, 0.1111111119389534, 0.0, 0.24444444477558136, 0.1111111119389534, 0.21111111342906952, 0.011111111380159855, 0.12222222238779068, 0.23333333432674408, 0.12222222238779068, 0.10000000149011612, 0.15555556118488312, 0.25555557012557983, 0.14444445073604584, 0.20000000298023224, 0.30000001192092896, 0.3444444537162781, 0.2222222238779068, 0.25555557012557983, 0.21111111342906952, 0.15555556118488312, 0.2888889014720917, 0.2666666805744171, 0.25555557012557983, 0.23333333432674408, 0.36666667461395264, 0.42222222685813904, 0.18333333730697632, 0.42222222685813904, 0.36666667461395264, NaN, 0.4333333373069763, 0.5111111402511597, 0.4555555582046509, 0.5666666626930237, 0.5111111402511597, 0.8666666746139526, 0.5, 0.7222222089767456, 0.6555555462837219, 0.6222222447395325, 0.7444444298744202, 0.8999999761581421], [0.011111111380159855, 0.0, 0.02222222276031971, 0.0, 0.011111111380159855, 0.0, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.02222222276031971, 0.07777778059244156, 0.02222222276031971, 0.12222222238779068, 0.02222222276031971, 0.03333333507180214, 0.02222222276031971, 0.011111111380159855, 0.12222222238779068, 0.08888889104127884, 0.03333333507180214, 0.03333333507180214, 0.02222222276031971, 0.0555555559694767, 0.12222222238779068, 0.02222222276031971, 0.02222222276031971, 0.06666667014360428, 0.0555555559694767, 0.03333333507180214, 0.011111111380159855, 0.02222222276031971, 0.12222222238779068, 0.12222222238779068, 0.18888889253139496, 0.04444444552063942, 0.13333334028720856, 0.02222222276031971, 0.14444445073604584, 0.04444444552063942, 0.13333334028720856, 0.04444444552063942, 0.14444445073604584, 0.13333334028720856, 0.03333333507180214, 0.14444445073604584, 0.24444444477558136, 0.12222222238779068, 0.36666667461395264, 0.1666666716337204, 0.15555556118488312, 0.24444444477558136, 0.15555556118488312, 0.21111111342906952, 0.20000000298023224, 0.2888889014720917, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.5555555820465088, 0.2666666805744171, 0.46666666865348816, 0.5666666626930237, NaN, 0.5333333611488342, 0.855555534362793, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.36666667461395264, 0.6555555462837219, 0.6111111044883728, 0.6555555462837219, 0.9222221970558167, 0.9555555582046509], [0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.011111111380159855, 0.0, 0.011111111380159855, 0.0, 0.011111111380159855, 0.08888889104127884, 0.03333333507180214, 0.011111111380159855, 0.0, 0.02222222276031971, 0.0, 0.03333333507180214, 0.02222222276031971, 0.04444444552063942, 0.011111111380159855, 0.011111111380159855, 0.12222222238779068, 0.07777778059244156, 0.12222222238779068, 0.08888889104127884, 0.15555556118488312, 0.17777778208255768, 0.20000000298023224, 0.14444445073604584, 0.23333333432674408, 0.07777778059244156, 0.1666666716337204, 0.17777778208255768, 0.10000000149011612, 0.03333333507180214, 0.14444445073604584, 0.1111111119389534, 0.24444444477558136, 0.06111111119389534, 0.04444444552063942, 0.011111111380159855, 0.17777778208255768, 0.3222222328186035, 0.23333333432674408, 0.04444444552063942, 0.21111111342906952, 0.3888888955116272, 0.4000000059604645, 0.4444444477558136, 0.4000000059604645, 0.31111112236976624, 0.4888888895511627, 0.46666666865348816, NaN, 0.46666666865348816, 0.4888888895511627, 0.6000000238418579, 0.5111111402511597, 0.5111111402511597, 0.8111110925674438, 0.5888888835906982, 0.6555555462837219, 0.7888888716697693, 1.0], [0.0, 0.0, 0.03333333507180214, 0.0, 0.011111111380159855, 0.0, 0.011111111380159855, 0.011111111380159855, 0.03333333507180214, 0.011111111380159855, 0.03333333507180214, 0.02222222276031971, 0.1111111119389534, 0.011111111380159855, 0.02222222276031971, 0.0, 0.011111111380159855, 0.12222222238779068, 0.03333333507180214, 0.0555555559694767, 0.04444444552063942, 0.011111111380159855, 0.03333333507180214, 0.13333334028720856, 0.011111111380159855, 0.0, 0.07777778059244156, 0.04444444552063942, 0.03333333507180214, 0.02222222276031971, 0.03333333507180214, 0.13333334028720856, 0.06666667014360428, 0.14444445073604584, 0.02222222276031971, 0.14444445073604584, 0.02222222276031971, 0.1666666716337204, 0.0555555559694767, 0.13333334028720856, 0.04444444552063942, 0.13333334028720856, 0.14444445073604584, 0.03333333507180214, 0.1666666716337204, 0.25555557012557983, 0.04444444552063942, 0.3222222328186035, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.2222222238779068, 0.17777778208255768, 0.2777777910232544, 0.2666666805744171, 0.5222222208976746, 0.2666666805744171, 0.42222222685813904, 0.5444444417953491, 0.14444445073604584, 0.5333333611488342, NaN, 0.6000000238418579, 0.5444444417953491, 0.5333333611488342, 0.31111112236976624, 0.644444465637207, 0.5444444417953491, 0.5777778029441833, 0.8999999761581421, 0.9555555582046509], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.0, 0.011111111380159855, 0.0, 0.0, 0.011111111380159855, 0.0, 0.04444444552063942, 0.011111111380159855, 0.0, 0.011111111380159855, 0.02222222276031971, 0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.02222222276031971, 0.011111111380159855, 0.10000000149011612, 0.011111111380159855, 0.0, 0.12222222238779068, 0.011111111380159855, 0.03333333507180214, 0.0, 0.12222222238779068, 0.02222222276031971, 0.13333334028720856, 0.13333334028720856, 0.14444445073604584, 0.02222222276031971, 0.2666666805744171, 0.13333334028720856, 0.08888889104127884, 0.06666667014360428, 0.07777778059244156, 0.0555555559694767, 0.36666667461395264, 0.07777778059244156, 0.08888889104127884, 0.0555555559694767, 0.13333334028720856, 0.2777777910232544, 0.3222222328186035, 0.06666667014360428, 0.07777778059244156, 0.36666667461395264, 0.35555556416511536, 0.41111111640930176, 0.35555556416511536, 0.42222222685813904, 0.4333333373069763, 0.4000000059604645, 0.5111111402511597, 0.4000000059604645, NaN, 0.6111111044883728, 0.5, 0.5444444417953491, 0.6333333253860474, 0.6333333253860474, 0.4888888895511627, 0.644444465637207, 0.9444444179534912], [0.0, 0.011111111380159855, 0.011111111380159855, 0.0, 0.0, 0.0, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.011111111380159855, 0.0, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.03333333507180214, 0.03333333507180214, 0.07777778059244156, 0.011111111380159855, 0.011111111380159855, 0.1111111119389534, 0.07777778059244156, 0.02222222276031971, 0.07777778059244156, 0.1666666716337204, 0.0555555559694767, 0.0555555559694767, 0.04444444552063942, 0.14444445073604584, 0.10000000149011612, 0.06666667014360428, 0.17777778208255768, 0.06666667014360428, 0.03333333507180214, 0.15555556118488312, 0.0555555559694767, 0.2888889014720917, 0.07777778059244156, 0.06666667014360428, 0.02222222276031971, 0.14444445073604584, 0.3222222328186035, 0.17777778208255768, 0.0555555559694767, 0.13333334028720856, 0.3777777850627899, 0.30000001192092896, 0.4555555582046509, 0.30000001192092896, 0.4333333373069763, 0.4888888895511627, 0.4000000059604645, 0.4000000059604645, 0.4555555582046509, 0.3888888955116272, NaN, 0.5222222208976746, 0.4555555582046509, 0.7111111283302307, 0.5888888835906982, 0.6555555462837219, 0.7888888716697693, 1.0], [0.0, 0.0, 0.0, 0.04444444552063942, 0.03333333507180214, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.02222222276031971, 0.02222222276031971, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.03333333507180214, 0.0, 0.07777778059244156, 0.06666667014360428, 0.0, 0.011111111380159855, 0.12222222238779068, 0.1111111119389534, 0.04444444552063942, 0.0555555559694767, 0.10000000149011612, 0.0, 0.18888889253139496, 0.07777778059244156, 0.10000000149011612, 0.0, 0.08888889104127884, 0.0555555559694767, 0.07777778059244156, 0.02222222276031971, 0.08888889104127884, 0.20000000298023224, 0.04444444552063942, 0.1111111119389534, 0.18888889253139496, 0.3444444537162781, 0.08888889104127884, 0.13333334028720856, 0.10000000149011612, 0.02222222276031971, 0.25555557012557983, 0.10000000149011612, 0.1111111119389534, 0.12222222238779068, 0.3222222328186035, 0.25555557012557983, 0.06666667014360428, 0.25555557012557983, 0.2777777910232544, 0.13333334028720856, 0.4333333373069763, 0.4888888895511627, 0.46666666865348816, 0.5, 0.47777777910232544, NaN, 0.47777777910232544, 0.7111111283302307, 0.5222222208976746, 0.6222222447395325, 0.7444444298744202, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.07777778059244156, 0.0, 0.0, 0.011111111380159855, 0.0, 0.0, 0.02222222276031971, 0.13333334028720856, 0.0, 0.08888889104127884, 0.06666667014360428, 0.011111111380159855, 0.0, 0.3777777850627899, 0.06666667014360428, 0.0, 0.14444445073604584, 0.04444444552063942, 0.2888889014720917, 0.12222222238779068, 0.15555556118488312, 0.10000000149011612, 0.30000001192092896, 0.0, 0.5, 0.0, 0.4000000059604645, 0.5, 0.6333333253860474, 0.4888888895511627, 0.6888889074325562, 0.4555555582046509, 0.5444444417953491, 0.5222222208976746, NaN, 0.6333333253860474, 0.6777777671813965, 0.699999988079071, 0.800000011920929, 0.8777777552604675], [0.0, 0.04444444552063942, 0.0, 0.07777778059244156, 0.07777778059244156, 0.07777778059244156, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.04444444552063942, 0.0, 0.0, 0.07777778059244156, 0.08888889104127884, 0.0, 0.0, 0.04444444552063942, 0.10000000149011612, 0.02222222276031971, 0.10000000149011612, 0.08888889104127884, 0.0, 0.10000000149011612, 0.06666667014360428, 0.011111111380159855, 0.04444444552063942, 0.08888889104127884, 0.07777778059244156, 0.10000000149011612, 0.17777778208255768, 0.10000000149011612, 0.0, 0.10000000149011612, 0.18888889253139496, 0.10000000149011612, 0.011111111380159855, 0.12222222238779068, 0.18888889253139496, 0.1111111119389534, 0.0, 0.20000000298023224, 0.10000000149011612, 0.1111111119389534, 0.1111111119389534, 0.10000000149011612, 0.2222222238779068, 0.14444445073604584, 0.10000000149011612, 0.08888889104127884, 0.1666666716337204, 0.24444444477558136, 0.1111111119389534, 0.10000000149011612, 0.1111111119389534, 0.18888889253139496, 0.30000001192092896, 0.2222222238779068, 0.30000001192092896, 0.2888889014720917, 0.2777777910232544, 0.3444444537162781, 0.18888889253139496, 0.35555556416511536, 0.36666667461395264, 0.2888889014720917, 0.2888889014720917, 0.36666667461395264, NaN, 0.4333333373069763, 0.4888888895511627, 0.5333333611488342, 0.8666666746139526], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.011111111380159855, 0.02222222276031971, 0.0, 0.011111111380159855, 0.0, 0.0, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.1111111119389534, 0.03333333507180214, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.0, 0.011111111380159855, 0.04444444552063942, 0.04444444552063942, 0.011111111380159855, 0.03333333507180214, 0.011111111380159855, 0.06666667014360428, 0.02222222276031971, 0.10000000149011612, 0.03333333507180214, 0.03333333507180214, 0.07777778059244156, 0.3444444537162781, 0.02222222276031971, 0.07777778059244156, 0.03333333507180214, 0.03333333507180214, 0.14444445073604584, 0.13333334028720856, 0.03333333507180214, 0.04444444552063942, 0.20000000298023224, 0.12222222238779068, 0.31111112236976624, 0.12222222238779068, 0.3333333432674408, 0.3444444537162781, 0.3888888955116272, 0.41111111640930176, 0.4555555582046509, 0.36666667461395264, 0.41111111640930176, 0.47777777910232544, 0.3222222328186035, 0.5666666626930237, NaN, 0.5555555820465088, 0.644444465637207, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.02222222276031971, 0.0, 0.10000000149011612, 0.0, 0.3222222328186035, 0.0, 0.0, 0.10000000149011612, 0.0, 0.2666666805744171, 0.1111111119389534, 0.08888889104127884, 0.011111111380159855, 0.18888889253139496, 0.12222222238779068, 0.3777777850627899, 0.12222222238779068, 0.0, 0.3777777850627899, 0.3444444537162781, 0.3444444537162781, 0.42222222685813904, 0.5111111402511597, 0.3444444537162781, 0.3777777850627899, 0.30000001192092896, 0.5111111402511597, 0.4444444477558136, NaN, 0.7111111283302307, 0.9888888597488403], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07777778059244156, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.0, 0.21111111342906952, 0.10000000149011612, 0.07777778059244156, 0.0, 0.10000000149011612, 0.02222222276031971, 0.04444444552063942, 0.011111111380159855, 0.04444444552063942, 0.1111111119389534, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.15555556118488312, 0.25555557012557983, 0.07777778059244156, 0.21111111342906952, 0.10000000149011612, 0.35555556416511536, 0.21111111342906952, 0.25555557012557983, 0.20000000298023224, 0.46666666865348816, 0.35555556416511536, 0.2888889014720917, NaN, 0.8777777552604675], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.04444444552063942, 0.0, 0.04444444552063942, 0.0555555559694767, 0.0, 0.1111111119389534, 0.12222222238779068, 0.13333334028720856, 0.20000000298023224, 0.011111111380159855, 0.12222222238779068, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9452054721},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9377473299},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8861491562},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8732115616},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8202435257},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8136986271},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7848554006},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7805174988},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7802130879},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7748858436},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7561643813},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7526636235},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7519025854},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7394216134},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7293759524},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7260273956},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7187214613},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.7033485538},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6996955841},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.6936073061},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6691019812},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6634703191},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6610350058},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.6599695588},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6520547965},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6328767152},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6238964996},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6207762578},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6165905655},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5940639282},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5922374447},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5791476407},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5613394215},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.529984781},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5251141585},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5136986321},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5135464266},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.50015221},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4770167455},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.4715372923},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4677321191},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4646879792},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4628614962},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4446727578},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4442922394},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.4266362271},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4240487081},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4149162916},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4127092872},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4057838691},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3920852384},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3907153768},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.3687214643},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3479452103},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3471841738},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3277016777},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2925418603},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2499238986},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2471080708},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.243531205},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2286149208},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2169710838},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2013698657},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.1813546444},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.177625575},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.1627092868},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1624048727},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1523592111},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.144444445},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1432267917},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1175038078},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1053272472},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0864535782},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0185692545}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_classification/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..a4112e79d76d37df84069fdf77068b27756b6027
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,"AutoGluon 1.5 (extreme, 4h)",TabFM (default),"AutoGluon 1.4 (best, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),TabICLv2 (default),CatBoost (tuned + ensembled),CatBoost (default),iLTM (tuned + ensembled),CatBoost (tuned),XGBoost (tuned + ensembled),LightGBM (tuned + ensembled),ChimeraBoost (tuned + ensembled),RealMLP (tuned + ensembled),RealTabPFN-2.5 (tuned + ensembled),TabM (tuned + ensembled),ModernNCA (tuned + ensembled),ChimeraBoost (tuned),XGBoost (tuned),RealTabPFN-2.5 (tuned),RealMLP (tuned),TabM (tuned),LightGBM (tuned),ChimeraBoost (default),RealTabPFN-2.5 (default),TabICL (default),iLTM (tuned),ModernNCA (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),TorchMLP (tuned + ensembled),EBM (tuned + ensembled),XGBoost (default),xRFM (tuned + ensembled),TabSwift (default),EBM (tuned),TabM (default),xRFM (tuned),RealMLP (default),BetaTabPFN (default),TorchMLP (tuned),FastaiMLP (tuned + ensembled),EBM (default),TabDPT (tuned),RandomForest (tuned + ensembled),iLTM (default),LightGBM (default),LimiX (default),RandomForest (tuned),ModernNCA (default),ExtraTrees (tuned + ensembled),SAP-RPT-OSS (default),TabDPT (default),FastaiMLP (tuned),ExtraTrees (tuned),TorchMLP (default),FastaiMLP (default),TabSTAR (tuned),Linear (tuned + ensembled),TabSTAR (tuned + ensembled),KNN (tuned + ensembled),Linear (tuned),PerpetualBooster (tuned + ensembled),RandomForest (default),PerpetualBooster (tuned),OrionMSP (default),ExtraTrees (default),Linear (default),TabSTAR (default),xRFM (default),TabFlex (default),KNN (tuned),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.4,0.74444443,0.44444445,0.82222223,0.67777777,0.9222222,0.9,0.93333334,0.9111111,0.95555556,0.93333334,0.9222222,0.9777778,0.9,0.8111111,0.96666664,0.93333334,0.98888886,0.9222222,0.96666664,0.8666667,0.9777778,0.96666664,0.9222222,0.9,0.8888889,0.98888886,0.9111111,0.9222222,0.98888886,0.98888886,0.96666664,0.9777778,0.9444444,1.0,1.0,0.98888886,1.0,1.0,0.9777778,1.0,1.0,0.95555556,1.0,1.0,1.0,0.8888889,0.98888886,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.98888886,1.0,0.98888886,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.6,,0.6333333,0.8111111,0.9,0.9111111,0.7777778,0.7888889,0.84444445,0.82222223,0.87777776,0.85555553,0.87777776,0.82222223,0.8666667,0.98888886,0.8888889,0.87777776,0.9111111,0.87777776,0.8333333,1.0,0.96666664,0.9,0.9777778,0.95555556,0.9,0.9444444,0.96666664,0.96666664,0.96666664,0.98888886,0.9777778,0.8888889,0.95555556,1.0,1.0,0.9111111,1.0,1.0,1.0,0.9111111,1.0,0.96666664,0.9777778,1.0,1.0,0.85555553,0.9777778,1.0,0.98888886,0.9777778,0.9777778,0.9777778,0.98888886,1.0,0.9777778,1.0,0.9777778,1.0,0.95555556,0.9777778,1.0,0.98888886,1.0,1.0,0.98888886,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.25555557,0.36666667,,0.42222223,0.7111111,0.65555555,0.76666665,0.7222222,0.84444445,0.7777778,0.8666667,0.84444445,0.7777778,0.8666667,0.64444447,0.74444443,0.7888889,0.8666667,0.8888889,0.74444443,0.9222222,0.8,0.9,0.8888889,0.76666665,0.76666665,0.85555553,0.9,0.8111111,0.8,0.9111111,0.95555556,0.9222222,0.95555556,0.84444445,0.95555556,0.9222222,0.96666664,0.96666664,0.9111111,0.9777778,0.96666664,0.95555556,0.96666664,0.98888886,0.96666664,0.95555556,0.87777776,0.9777778,0.98888886,0.98888886,0.96666664,0.95555556,0.9777778,1.0,0.98888886,0.98888886,1.0,1.0,1.0,0.98888886,1.0,0.9777778,1.0,0.96666664,1.0,0.98888886,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.5555556,0.18888889,0.5777778,,0.75555557,0.5777778,0.7,0.6888889,0.7111111,0.6888889,0.74444443,0.73333335,0.7777778,0.74444443,0.85555553,0.7888889,0.8888889,0.76666665,0.7777778,0.87777776,0.75555557,0.85555553,0.9,0.76666665,0.87777776,0.8666667,0.8666667,0.87777776,0.95555556,0.8888889,0.8333333,0.8666667,0.9111111,0.7888889,0.96666664,0.9111111,0.98888886,0.8,0.9444444,0.9444444,0.96666664,0.9,0.95555556,0.96666664,0.9444444,0.98888886,0.98888886,0.8,0.96666664,0.9777778,0.96666664,0.9222222,0.96666664,0.93333334,0.96666664,0.96666664,0.93333334,1.0,0.9222222,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,0.9222222,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.17777778,0.1,0.2888889,0.24444444,,0.56666666,0.56666666,0.5222222,0.5555556,0.56666666,0.5777778,0.6888889,0.67777777,0.62222224,0.7,0.62222224,0.7,0.7111111,0.73333335,0.7,0.7111111,0.74444443,0.82222223,0.67777777,0.85555553,0.7777778,0.75555557,0.82222223,0.85555553,0.85555553,0.84444445,0.87777776,0.9111111,0.76666665,0.8666667,0.96666664,0.9222222,0.8111111,1.0,0.9,0.95555556,0.8666667,0.96666664,0.95555556,0.95555556,0.9444444,0.9444444,0.7777778,0.98888886,0.9777778,0.9777778,0.95555556,0.9444444,0.9444444,0.98888886,1.0,0.9222222,1.0,0.96666664,1.0,0.9444444,0.96666664,0.98888886,1.0,0.98888886,1.0,1.0,0.96666664,1.0,0.9222222,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.32222223,0.08888889,0.34444445,0.42222223,0.43333334,,0.6111111,0.6111111,0.6333333,0.6,0.62222224,0.6666667,0.62222224,0.5777778,0.73333335,0.6111111,0.7111111,0.62222224,0.64444447,0.75555557,0.6333333,0.7222222,0.76666665,0.64444447,0.84444445,0.8333333,0.75555557,0.8,0.87777776,0.9,0.8111111,0.75555557,0.8666667,0.7,0.9111111,0.9,0.9444444,0.7222222,0.9111111,0.9222222,0.9111111,0.84444445,0.93333334,0.9444444,0.95555556,0.9222222,0.9444444,0.8111111,0.96666664,0.95555556,0.98888886,0.9222222,0.95555556,0.93333334,0.98888886,0.9777778,0.9222222,0.9777778,0.9222222,0.9777778,0.95555556,0.9444444,1.0,1.0,1.0,1.0,1.0,0.96666664,1.0,0.9222222,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.07777778,0.22222222,0.23333333,0.3,0.43333334,0.3888889,,0.5888889,0.5777778,0.5888889,0.6111111,0.51111114,0.5888889,0.62222224,0.5222222,0.6,0.5888889,0.73333335,0.6888889,0.53333336,0.7111111,0.64444447,0.64444447,0.7888889,0.73333335,0.6888889,0.74444443,0.73888886,0.6,0.6666667,0.7888889,0.84444445,0.82222223,0.9222222,0.7777778,0.82222223,0.76666665,0.9444444,0.8333333,0.84444445,0.95555556,0.8666667,0.9444444,0.8111111,0.8888889,0.9222222,0.95555556,0.7222222,0.87777776,0.95555556,0.98888886,0.9777778,0.8888889,0.96666664,0.9777778,0.98888886,0.9777778,1.0,0.98888886,1.0,0.98888886,1.0,0.9777778,0.98888886,0.98888886,0.98888886,0.98888886,1.0,1.0,1.0,0.98888886,0.98888886,1.0,1.0
+CatBoost (default),0.1,0.21111111,0.2777778,0.31111112,0.47777778,0.3888889,0.41111112,,0.5222222,0.47777778,0.53333336,0.54444444,0.6,0.62222224,0.5222222,0.6,0.5888889,0.7222222,0.65555555,0.54444444,0.67777777,0.65555555,0.6888889,0.8111111,0.7111111,0.7111111,0.75555557,0.7111111,0.5888889,0.6888889,0.7888889,0.8111111,0.7888889,0.9444444,0.8111111,0.8111111,0.7888889,0.95555556,0.82222223,0.85555553,0.95555556,0.8333333,0.8888889,0.8111111,0.87777776,0.9444444,0.93333334,0.75555557,0.87777776,0.98888886,0.96666664,0.95555556,0.9,0.9777778,0.98888886,0.9777778,0.95555556,1.0,0.98888886,1.0,0.98888886,1.0,0.98888886,0.98888886,0.98888886,0.98888886,0.9777778,1.0,1.0,1.0,0.98888886,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.06666667,0.15555556,0.15555556,0.2888889,0.44444445,0.36666667,0.42222223,0.47777778,,0.44444445,0.53333336,0.56666666,0.5,0.54444444,0.5777778,0.5777778,0.56666666,0.5888889,0.67777777,0.62222224,0.65555555,0.65555555,0.67777777,0.64444447,0.73333335,0.73333335,0.7111111,0.7777778,0.6111111,0.74444443,0.8666667,0.8111111,0.7888889,0.8888889,0.76666665,0.82222223,0.8333333,0.95555556,0.95555556,0.8666667,0.95555556,0.96666664,0.9111111,0.9111111,0.87777776,1.0,0.9222222,0.75555557,0.8888889,0.9777778,0.95555556,0.9777778,0.93333334,0.96666664,0.98888886,1.0,1.0,1.0,0.98888886,1.0,0.98888886,1.0,0.9777778,0.98888886,0.96666664,0.98888886,0.98888886,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.08888889,0.17777778,0.22222222,0.31111112,0.43333334,0.4,0.41111112,0.5222222,0.5555556,,0.62222224,0.46666667,0.62222224,0.5777778,0.4888889,0.6111111,0.6,0.73333335,0.7111111,0.53333336,0.7,0.62222224,0.6666667,0.75555557,0.6888889,0.7,0.7111111,0.6888889,0.5888889,0.6666667,0.7777778,0.8333333,0.8111111,0.93333334,0.76666665,0.82222223,0.8,0.93333334,0.84444445,0.85555553,0.9444444,0.82222223,0.96666664,0.8,0.8888889,0.93333334,0.9222222,0.7222222,0.87777776,0.93333334,0.95555556,0.96666664,0.87777776,0.95555556,0.96666664,0.9777778,0.9444444,0.98888886,0.98888886,0.98888886,0.98888886,1.0,0.9777778,0.98888886,0.98888886,0.9777778,0.9777778,1.0,1.0,1.0,0.98888886,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.044444446,0.12222222,0.13333334,0.25555557,0.42222223,0.37777779,0.3888889,0.46666667,0.46666667,0.37777779,,0.45555556,0.5222222,0.5,0.5,0.62222224,0.5222222,0.5777778,0.8666667,0.53333336,0.5888889,0.67777777,0.62222224,0.6666667,0.65555555,0.6,0.6333333,0.64444447,0.62222224,0.6888889,0.8,0.8111111,0.93333334,0.82222223,0.6888889,0.84444445,0.8111111,0.9111111,0.84444445,0.87777776,0.84444445,0.8,0.9777778,0.84444445,0.9,0.96666664,0.96666664,0.67777777,0.9,0.98888886,0.98888886,0.95555556,0.8888889,0.87777776,0.98888886,1.0,0.9,0.98888886,1.0,0.98888886,0.98888886,1.0,0.9222222,1.0,0.96666664,1.0,0.98888886,1.0,1.0,0.98888886,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.06666667,0.14444445,0.15555556,0.26666668,0.31111112,0.33333334,0.4888889,0.45555556,0.43333334,0.53333336,0.54444444,,0.47777778,0.47777778,0.46666667,0.53333336,0.51111114,0.6111111,0.6888889,0.5,0.56666666,0.6333333,0.8666667,0.62222224,0.62222224,0.6666667,0.7,0.65555555,0.6666667,0.6111111,0.73333335,0.76666665,0.82222223,0.76666665,0.74444443,0.8666667,0.8,0.8333333,0.82222223,0.8666667,0.9111111,0.76666665,0.96666664,0.8,0.9,0.9777778,0.9777778,0.6888889,0.9111111,0.9777778,0.9777778,0.95555556,0.95555556,0.9111111,0.9777778,0.9777778,0.8666667,0.98888886,0.9777778,0.98888886,0.9777778,0.98888886,0.9777778,1.0,0.9777778,0.98888886,0.98888886,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.07777778,0.12222222,0.22222222,0.22222222,0.32222223,0.37777779,0.41111112,0.4,0.5,0.37777779,0.47777778,0.5222222,,0.62222224,0.51111114,0.6111111,0.5222222,0.75555557,0.5888889,0.56666666,0.7222222,0.6888889,0.56666666,0.87777776,0.6333333,0.6333333,0.73333335,0.7111111,0.6111111,0.7,0.7111111,0.82222223,0.67777777,0.9,0.6666667,0.85555553,0.82222223,0.9222222,0.85555553,0.87777776,0.8666667,0.85555553,0.93333334,0.8333333,0.8888889,0.95555556,0.87777776,0.74444443,0.8888889,0.9111111,0.8888889,0.9777778,0.8333333,0.8888889,0.8888889,0.9111111,0.87777776,1.0,1.0,1.0,0.98888886,1.0,0.87777776,1.0,0.8888889,1.0,0.98888886,1.0,1.0,1.0,1.0,1.0,0.9222222,1.0
+RealMLP (tuned + ensembled),0.022222223,0.17777778,0.13333334,0.25555557,0.37777779,0.42222223,0.37777779,0.37777779,0.45555556,0.42222223,0.5,0.5222222,0.37777779,,0.5555556,0.42222223,0.53333336,0.4888889,0.6,0.5888889,0.75555557,0.47777778,0.67777777,0.4888889,0.6111111,0.5777778,0.62222224,0.6888889,0.6666667,0.6111111,0.75555557,0.75555557,0.74444443,0.84444445,0.65555555,0.82222223,0.9,0.8888889,1.0,0.85555553,0.8333333,0.9777778,0.84444445,0.8111111,0.87777776,0.8111111,0.93333334,0.84444445,0.8666667,0.85555553,0.9,0.8888889,0.82222223,0.9777778,0.87777776,0.98888886,0.96666664,1.0,1.0,1.0,0.98888886,1.0,0.9777778,0.93333334,0.98888886,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.1,0.13333334,0.35555556,0.14444445,0.3,0.26666668,0.47777778,0.47777778,0.42222223,0.51111114,0.5,0.53333336,0.4888889,0.44444445,,0.43333334,0.46666667,0.5222222,0.54444444,0.8,0.5555556,0.5555556,0.64444447,0.6111111,0.7111111,0.7,0.62222224,0.6888889,0.64444447,0.6888889,0.7111111,0.73333335,0.76666665,0.65555555,0.85555553,0.82222223,0.74444443,0.7111111,0.82222223,0.8,0.82222223,0.73333335,0.8666667,0.8888889,0.9111111,0.85555553,0.9222222,0.6666667,0.95555556,0.82222223,0.96666664,0.85555553,0.9,0.87777776,0.96666664,0.95555556,0.9,1.0,0.84444445,1.0,0.9111111,0.85555553,0.96666664,1.0,0.9777778,0.98888886,0.98888886,0.95555556,1.0,0.9222222,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.18888889,0.011111111,0.25555557,0.21111111,0.37777779,0.3888889,0.4,0.4,0.42222223,0.3888889,0.37777779,0.46666667,0.3888889,0.5777778,0.56666666,,0.5222222,0.45555556,0.5,0.5888889,0.65555555,0.73333335,0.6111111,0.5222222,0.56666666,0.5777778,0.6333333,0.73333335,0.7111111,0.65555555,0.6111111,0.6666667,0.6666667,0.65555555,0.75555557,0.84444445,0.96666664,0.65555555,0.9777778,0.9111111,0.7777778,0.82222223,0.8333333,0.87777776,0.8333333,0.8333333,1.0,0.76666665,0.8333333,0.9,0.8333333,0.87777776,0.87777776,0.9111111,0.8666667,0.9777778,0.9,1.0,0.93333334,1.0,0.9444444,0.9444444,0.9777778,0.98888886,1.0,1.0,1.0,0.9777778,1.0,0.9111111,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.033333335,0.11111111,0.21111111,0.11111111,0.3,0.2888889,0.41111112,0.41111112,0.43333334,0.4,0.47777778,0.4888889,0.47777778,0.46666667,0.53333336,0.47777778,,0.5555556,0.5,0.54444444,0.5888889,0.56666666,0.53333336,0.6333333,0.62222224,0.6333333,0.64444447,0.74444443,0.6111111,0.6888889,0.73333335,0.73333335,0.65555555,0.7111111,0.7,0.75555557,0.82222223,0.7777778,0.8888889,0.82222223,0.84444445,0.8666667,0.82222223,0.87777776,0.7888889,0.84444445,0.9111111,0.85555553,0.7777778,0.95555556,0.9,0.85555553,0.9222222,0.93333334,0.8888889,0.8888889,0.96666664,0.9222222,0.9,0.9222222,0.98888886,0.9111111,0.98888886,1.0,0.98888886,0.95555556,0.9777778,0.9777778,1.0,1.0,0.93333334,1.0,1.0,1.0
+ChimeraBoost (tuned),0.06666667,0.12222222,0.13333334,0.23333333,0.2888889,0.37777779,0.26666668,0.2777778,0.41111112,0.26666668,0.42222223,0.3888889,0.24444444,0.51111114,0.47777778,0.54444444,0.44444445,,0.5222222,0.4888889,0.67777777,0.6,0.51111114,0.7111111,0.5777778,0.5777778,0.64444447,0.6666667,0.5222222,0.65555555,0.64444447,0.7222222,0.65555555,0.8666667,0.6333333,0.7777778,0.7888889,0.9,0.84444445,0.85555553,0.82222223,0.8111111,0.85555553,0.8111111,0.8666667,0.9,0.8111111,0.73333335,0.8666667,0.85555553,0.8666667,0.93333334,0.7777778,0.85555553,0.8666667,0.9111111,0.8666667,0.98888886,1.0,0.98888886,0.9777778,1.0,0.87777776,0.98888886,0.87777776,0.98888886,0.98888886,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0
+XGBoost (tuned),0.011111111,0.08888889,0.11111111,0.22222222,0.26666668,0.35555556,0.31111112,0.34444445,0.32222223,0.2888889,0.13333334,0.31111112,0.41111112,0.4,0.45555556,0.5,0.5,0.47777778,,0.46666667,0.46666667,0.6111111,0.51111114,0.6,0.56666666,0.54444444,0.56666666,0.6333333,0.5777778,0.6111111,0.67777777,0.7222222,0.76666665,0.74444443,0.6666667,0.8111111,0.7777778,0.8111111,0.82222223,0.85555553,0.82222223,0.75555557,0.9222222,0.8,0.8666667,0.93333334,0.93333334,0.65555555,0.85555553,0.96666664,0.95555556,0.9444444,0.85555553,0.8666667,0.9777778,1.0,0.85555553,0.98888886,1.0,0.98888886,0.98888886,1.0,0.9111111,1.0,0.96666664,1.0,0.98888886,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.07777778,0.12222222,0.25555557,0.12222222,0.3,0.24444444,0.46666667,0.45555556,0.37777779,0.46666667,0.46666667,0.5,0.43333334,0.41111112,0.2,0.41111112,0.45555556,0.51111114,0.53333336,,0.5222222,0.53333336,0.64444447,0.56666666,0.6666667,0.6666667,0.5777778,0.6111111,0.5888889,0.6888889,0.65555555,0.6666667,0.75555557,0.65555555,0.8333333,0.73333335,0.67777777,0.6888889,0.76666665,0.75555557,0.8111111,0.7,0.8333333,0.82222223,0.85555553,0.82222223,0.8888889,0.65555555,0.93333334,0.7888889,0.9111111,0.8333333,0.84444445,0.87777776,0.93333334,0.9111111,0.8888889,0.98888886,0.82222223,1.0,0.8888889,0.8666667,0.96666664,0.98888886,0.9444444,0.98888886,0.9777778,0.93333334,1.0,0.9,0.98888886,1.0,1.0,1.0
+RealMLP (tuned),0.033333335,0.16666667,0.07777778,0.24444444,0.2888889,0.36666667,0.2888889,0.32222223,0.34444445,0.3,0.41111112,0.43333334,0.2777778,0.24444444,0.44444445,0.34444445,0.41111112,0.32222223,0.53333336,0.47777778,,0.4,0.6,0.43333334,0.5222222,0.47777778,0.5777778,0.5777778,0.4888889,0.5555556,0.67777777,0.64444447,0.67777777,0.75555557,0.54444444,0.7111111,0.76666665,0.8,0.9,0.7777778,0.8111111,0.87777776,0.73333335,0.73333335,0.84444445,0.73333335,0.82222223,0.8,0.82222223,0.75555557,0.8,0.7888889,0.74444443,0.96666664,0.85555553,0.93333334,0.95555556,0.9777778,0.9777778,0.9777778,0.96666664,0.98888886,0.96666664,0.9111111,0.95555556,0.9777778,0.98888886,1.0,1.0,0.9777778,0.9777778,1.0,1.0,1.0
+TabM (tuned),0.13333334,0.0,0.2,0.14444445,0.25555557,0.2777778,0.35555556,0.34444445,0.34444445,0.37777779,0.32222223,0.36666667,0.31111112,0.5222222,0.44444445,0.26666668,0.43333334,0.4,0.3888889,0.46666667,0.6,,0.44444445,0.41111112,0.4888889,0.4888889,0.5888889,0.6333333,0.5888889,0.54444444,0.56666666,0.5555556,0.65555555,0.6111111,0.65555555,0.76666665,0.84444445,0.65555555,0.9,0.8,0.74444443,0.8,0.7888889,0.76666665,0.7777778,0.75555557,0.8666667,0.76666665,0.7888889,0.8333333,0.76666665,0.8,0.8,0.87777776,0.8333333,0.9777778,0.9111111,1.0,0.8888889,1.0,0.8888889,0.9,0.9777778,0.96666664,0.98888886,1.0,0.9777778,0.93333334,1.0,0.9,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.022222223,0.033333335,0.1,0.1,0.17777778,0.23333333,0.35555556,0.31111112,0.32222223,0.33333334,0.37777779,0.13333334,0.43333334,0.32222223,0.35555556,0.3888889,0.46666667,0.4888889,0.4888889,0.35555556,0.4,0.5555556,,0.54444444,0.4888889,0.5222222,0.54444444,0.5555556,0.6,0.5777778,0.5777778,0.67777777,0.74444443,0.7222222,0.6333333,0.73333335,0.75555557,0.7222222,0.76666665,0.7888889,0.7,0.75555557,0.85555553,0.74444443,0.7888889,0.93333334,0.96666664,0.64444447,0.84444445,0.93333334,0.8333333,0.82222223,0.8333333,0.7888889,0.9222222,0.93333334,0.8111111,1.0,0.8666667,1.0,0.9111111,0.9111111,0.9444444,0.98888886,0.96666664,0.98888886,0.98888886,0.96666664,1.0,0.9111111,0.98888886,1.0,1.0,1.0
+ChimeraBoost (default),0.033333335,0.1,0.11111111,0.23333333,0.32222223,0.35555556,0.21111111,0.18888889,0.35555556,0.24444444,0.33333334,0.37777779,0.12222222,0.51111114,0.3888889,0.47777778,0.36666667,0.2888889,0.4,0.43333334,0.56666666,0.5888889,0.45555556,,0.43333334,0.45555556,0.56666666,0.54444444,0.53333336,0.56666666,0.62222224,0.65555555,0.6111111,0.8,0.6,0.7222222,0.67777777,0.87777776,0.75555557,0.75555557,0.74444443,0.8111111,0.76666665,0.8333333,0.84444445,0.8333333,0.75555557,0.73333335,0.8111111,0.8111111,0.76666665,0.95555556,0.82222223,0.8666667,0.84444445,0.8666667,0.8666667,0.98888886,1.0,1.0,0.9777778,1.0,0.87777776,1.0,0.8666667,1.0,0.98888886,1.0,1.0,1.0,1.0,1.0,0.9,1.0
+RealTabPFN-2.5 (default),0.07777778,0.022222223,0.23333333,0.12222222,0.14444445,0.15555556,0.26666668,0.2888889,0.26666668,0.31111112,0.34444445,0.37777779,0.36666667,0.3888889,0.2888889,0.43333334,0.37777779,0.42222223,0.43333334,0.33333334,0.47777778,0.51111114,0.51111114,0.56666666,,0.6111111,0.62222224,0.5222222,0.44444445,0.62222224,0.62222224,0.53333336,0.6888889,0.67777777,0.8666667,0.6333333,0.5888889,0.6888889,0.7111111,0.7888889,0.8,0.6666667,0.7222222,0.7777778,0.74444443,0.8333333,0.87777776,0.6666667,0.8,0.8,0.9222222,0.9,0.85555553,0.8,0.9444444,0.95555556,0.84444445,1.0,0.84444445,1.0,0.9,0.84444445,0.9777778,0.9777778,0.98888886,1.0,0.98888886,0.9222222,1.0,0.9,1.0,1.0,1.0,1.0
+TabICL (default),0.1,0.044444446,0.23333333,0.13333334,0.22222222,0.16666667,0.31111112,0.2888889,0.26666668,0.3,0.4,0.33333334,0.36666667,0.42222223,0.3,0.42222223,0.36666667,0.42222223,0.45555556,0.33333334,0.5222222,0.51111114,0.47777778,0.54444444,0.3888889,,0.53333336,0.44444445,0.53333336,0.64444447,0.6333333,0.62222224,0.6111111,0.65555555,0.65555555,0.6333333,0.6111111,0.7,0.67777777,0.6888889,0.67777777,0.6888889,0.67777777,0.85555553,0.7111111,0.76666665,0.8666667,0.6888889,0.7222222,0.8666667,0.8,0.7777778,0.9222222,0.7111111,0.8333333,0.8,0.7888889,0.93333334,0.75555557,0.9444444,0.85555553,0.75555557,0.9777778,1.0,1.0,1.0,1.0,0.93333334,1.0,0.93333334,0.9777778,1.0,1.0,1.0
+iLTM (tuned),0.11111111,0.1,0.14444445,0.13333334,0.24444444,0.24444444,0.25555557,0.24444444,0.2888889,0.2888889,0.36666667,0.3,0.26666668,0.37777779,0.37777779,0.36666667,0.35555556,0.35555556,0.43333334,0.42222223,0.42222223,0.41111112,0.45555556,0.43333334,0.37777779,0.46666667,,0.43333334,0.5,0.5555556,0.4888889,0.51111114,0.5,0.7111111,0.5777778,0.5777778,0.62222224,0.7111111,0.65555555,0.65555555,0.6111111,0.73333335,0.6,0.7888889,0.73333335,0.76666665,0.8,0.76666665,0.75555557,0.84444445,0.74444443,0.84444445,0.85555553,0.85555553,0.8666667,0.8888889,0.9,0.9444444,0.9,0.9444444,0.98888886,0.93333334,0.93333334,0.96666664,0.9222222,0.9777778,0.96666664,1.0,0.98888886,0.98888886,0.98888886,1.0,1.0,1.0
+ModernNCA (tuned),0.011111111,0.055555556,0.1,0.12222222,0.17777778,0.2,0.2611111,0.2888889,0.22222222,0.31111112,0.35555556,0.34444445,0.2888889,0.31111112,0.31111112,0.26666668,0.25555557,0.33333334,0.36666667,0.3888889,0.42222223,0.36666667,0.44444445,0.45555556,0.47777778,0.5555556,0.56666666,,0.53333336,0.5777778,0.6111111,0.5222222,0.53333336,0.5555556,0.64444447,0.5888889,0.62222224,0.62222224,0.7222222,0.6888889,0.7222222,0.6666667,0.64444447,0.7777778,0.7222222,0.76666665,0.76666665,0.7222222,0.7111111,0.9,0.84444445,0.7888889,0.8666667,0.9222222,0.85555553,0.85555553,0.9111111,0.95555556,0.8666667,0.95555556,0.95555556,0.9,0.9444444,0.9777778,0.95555556,0.98888886,0.96666664,0.98888886,1.0,0.95555556,0.9777778,1.0,1.0,1.0
+TabDPT-Turbo (default),0.08888889,0.033333335,0.18888889,0.044444446,0.14444445,0.12222222,0.4,0.41111112,0.3888889,0.41111112,0.37777779,0.33333334,0.3888889,0.33333334,0.35555556,0.2888889,0.3888889,0.47777778,0.42222223,0.41111112,0.51111114,0.41111112,0.4,0.46666667,0.5555556,0.46666667,0.5,0.46666667,,0.5888889,0.64444447,0.51111114,0.6,0.62222224,0.7111111,0.5777778,0.56666666,0.65555555,0.7,0.6111111,0.73333335,0.64444447,0.7111111,0.8333333,0.67777777,0.6888889,0.7888889,0.76666665,0.75555557,0.7,0.76666665,0.7,0.9111111,0.74444443,0.82222223,0.8,0.87777776,0.8888889,0.73333335,0.8888889,0.85555553,0.75555557,0.96666664,0.95555556,0.96666664,0.9,0.9222222,0.87777776,1.0,0.9111111,0.8888889,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.07777778,0.033333335,0.2,0.11111111,0.14444445,0.1,0.33333334,0.31111112,0.25555557,0.33333334,0.31111112,0.3888889,0.3,0.3888889,0.31111112,0.34444445,0.31111112,0.34444445,0.3888889,0.31111112,0.44444445,0.45555556,0.42222223,0.43333334,0.37777779,0.35555556,0.44444445,0.42222223,0.41111112,,0.5777778,0.5,0.54444444,0.6111111,0.6111111,0.53333336,0.6,0.65555555,0.64444447,0.62222224,0.7222222,0.6888889,0.56666666,0.8888889,0.6666667,0.6333333,0.7777778,0.6666667,0.65555555,0.7888889,0.67777777,0.74444443,0.9444444,0.8,0.7888889,0.7777778,0.8333333,0.85555553,0.75555557,0.85555553,0.84444445,0.75555557,0.98888886,0.98888886,0.9777778,0.98888886,0.98888886,0.8888889,1.0,0.9222222,0.96666664,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.011111111,0.033333335,0.08888889,0.16666667,0.15555556,0.18888889,0.21111111,0.21111111,0.13333334,0.22222222,0.2,0.26666668,0.2888889,0.24444444,0.2888889,0.3888889,0.26666668,0.35555556,0.32222223,0.34444445,0.32222223,0.43333334,0.42222223,0.37777779,0.37777779,0.36666667,0.51111114,0.3888889,0.35555556,0.42222223,,0.46666667,0.56666666,0.6,0.5888889,0.6,0.62222224,0.65555555,0.65555555,0.6,0.87777776,0.65555555,0.6333333,0.7111111,0.7,0.75555557,0.75555557,0.6666667,0.75555557,0.73333335,0.8,0.82222223,0.76666665,0.8333333,0.8666667,0.98888886,0.8333333,0.9222222,0.9111111,0.9222222,0.9111111,0.9111111,0.9777778,0.98888886,0.96666664,1.0,0.98888886,0.95555556,1.0,0.9,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.011111111,0.011111111,0.044444446,0.13333334,0.12222222,0.24444444,0.15555556,0.18888889,0.18888889,0.16666667,0.18888889,0.23333333,0.17777778,0.24444444,0.26666668,0.33333334,0.26666668,0.2777778,0.2777778,0.33333334,0.35555556,0.44444445,0.32222223,0.34444445,0.46666667,0.37777779,0.4888889,0.47777778,0.4888889,0.5,0.53333336,,0.6111111,0.45555556,0.5777778,0.74444443,0.6,0.5888889,0.7222222,0.6888889,0.7888889,0.6666667,0.9222222,0.7,0.7222222,0.7222222,0.67777777,0.6,0.76666665,0.7222222,0.8,0.87777776,0.67777777,0.75555557,0.84444445,0.8666667,0.8,0.9777778,0.9,0.96666664,0.9111111,0.9222222,0.87777776,0.87777776,0.8666667,0.87777776,0.8888889,0.9444444,1.0,0.82222223,0.9777778,1.0,0.9,1.0
+XGBoost (default),0.033333335,0.022222223,0.07777778,0.08888889,0.08888889,0.13333334,0.17777778,0.21111111,0.21111111,0.18888889,0.06666667,0.17777778,0.32222223,0.25555557,0.23333333,0.33333334,0.34444445,0.34444445,0.23333333,0.24444444,0.32222223,0.34444445,0.25555557,0.3888889,0.31111112,0.3888889,0.5,0.46666667,0.4,0.45555556,0.43333334,0.3888889,,0.6333333,0.54444444,0.4888889,0.53333336,0.67777777,0.54444444,0.5888889,0.6666667,0.6888889,0.64444447,0.65555555,0.75555557,0.8111111,0.53333336,0.6111111,0.7777778,0.7222222,0.8666667,0.8333333,0.65555555,0.7111111,0.9,0.9222222,0.76666665,0.9444444,0.87777776,0.95555556,0.8888889,0.8888889,0.87777776,0.9222222,0.93333334,0.98888886,0.9222222,0.9,1.0,0.9,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.022222223,0.11111111,0.044444446,0.21111111,0.23333333,0.3,0.07777778,0.055555556,0.11111111,0.06666667,0.17777778,0.23333333,0.1,0.15555556,0.34444445,0.34444445,0.2888889,0.13333334,0.25555557,0.34444445,0.24444444,0.3888889,0.2777778,0.2,0.32222223,0.34444445,0.2888889,0.44444445,0.37777779,0.3888889,0.4,0.54444444,0.36666667,,0.37777779,0.6111111,0.5555556,0.67777777,0.62222224,0.6333333,0.5777778,0.73333335,0.62222224,0.46666667,0.64444447,0.54444444,0.46666667,0.6888889,0.65555555,0.56666666,0.73333335,0.6888889,0.5888889,0.8,0.7222222,0.75555557,0.8666667,0.85555553,0.9777778,0.85555553,0.96666664,1.0,0.8111111,0.87777776,0.85555553,0.96666664,0.9777778,1.0,0.9,1.0,0.9777778,0.96666664,0.9,1.0
+TabSwift (default),0.055555556,0.044444446,0.15555556,0.033333335,0.13333334,0.08888889,0.22222222,0.18888889,0.23333333,0.23333333,0.31111112,0.25555557,0.33333334,0.34444445,0.14444445,0.24444444,0.3,0.36666667,0.33333334,0.16666667,0.45555556,0.34444445,0.36666667,0.4,0.13333334,0.34444445,0.42222223,0.35555556,0.2888889,0.3888889,0.41111112,0.42222223,0.45555556,0.62222224,,0.47777778,0.54444444,0.5777778,0.47777778,0.56666666,0.5555556,0.62222224,0.5,0.62222224,0.53333336,0.64444447,0.62222224,0.67777777,0.5555556,0.65555555,0.6333333,0.6333333,0.85555553,0.67777777,0.7,0.6888889,0.67777777,0.8,0.7222222,0.8,0.7888889,0.75555557,0.95555556,0.9111111,0.9777778,1.0,0.9222222,0.8111111,0.9222222,0.9,1.0,0.96666664,1.0,1.0
+EBM (tuned),0.0,0.0,0.044444446,0.08888889,0.033333335,0.1,0.17777778,0.18888889,0.17777778,0.17777778,0.15555556,0.13333334,0.14444445,0.17777778,0.17777778,0.15555556,0.24444444,0.22222222,0.18888889,0.26666668,0.2888889,0.23333333,0.26666668,0.2777778,0.36666667,0.36666667,0.42222223,0.41111112,0.42222223,0.46666667,0.4,0.25555557,0.51111114,0.3888889,0.5222222,,0.46666667,0.44444445,0.6,0.62222224,0.62222224,0.5555556,0.74444443,0.5888889,0.6333333,0.6333333,0.5888889,0.5888889,0.6666667,0.67777777,0.6888889,0.82222223,0.6,0.73333335,0.75555557,0.84444445,0.74444443,0.9444444,0.85555553,0.9444444,0.8888889,0.8888889,0.8666667,0.84444445,0.85555553,0.87777776,0.8333333,0.9222222,1.0,0.8111111,0.98888886,1.0,0.9,1.0
+TabM (default),0.0,0.0,0.07777778,0.011111111,0.07777778,0.055555556,0.23333333,0.21111111,0.16666667,0.2,0.18888889,0.2,0.17777778,0.1,0.25555557,0.033333335,0.17777778,0.21111111,0.22222222,0.32222223,0.23333333,0.15555556,0.24444444,0.32222223,0.41111112,0.3888889,0.37777779,0.37777779,0.43333334,0.4,0.37777779,0.4,0.46666667,0.44444445,0.45555556,0.53333336,,0.4888889,0.6111111,0.6,0.51111114,0.65555555,0.5777778,0.6111111,0.6666667,0.6111111,0.62222224,0.5777778,0.6666667,0.7222222,0.64444447,0.62222224,0.62222224,0.7777778,0.67777777,0.7777778,0.8111111,0.9,0.8,0.9,0.84444445,0.7888889,0.9777778,0.82222223,0.9777778,0.9777778,0.9444444,0.9,1.0,0.9,0.95555556,1.0,1.0,1.0
+xRFM (tuned),0.011111111,0.08888889,0.033333335,0.2,0.18888889,0.2777778,0.055555556,0.044444446,0.044444446,0.06666667,0.08888889,0.16666667,0.07777778,0.11111111,0.2888889,0.34444445,0.22222222,0.1,0.18888889,0.31111112,0.2,0.34444445,0.2777778,0.12222222,0.31111112,0.3,0.2888889,0.37777779,0.34444445,0.34444445,0.34444445,0.41111112,0.32222223,0.32222223,0.42222223,0.5555556,0.51111114,,0.6,0.6333333,0.54444444,0.6666667,0.5222222,0.51111114,0.62222224,0.47777778,0.45555556,0.62222224,0.65555555,0.56666666,0.65555555,0.8,0.56666666,0.82222223,0.74444443,0.73333335,0.8666667,0.8333333,0.95555556,0.84444445,0.9777778,0.98888886,0.85555553,0.8,0.8333333,0.8666667,0.9444444,1.0,0.98888886,0.98888886,0.95555556,1.0,0.9,1.0
+RealMLP (default),0.0,0.0,0.033333335,0.055555556,0.0,0.08888889,0.16666667,0.17777778,0.044444446,0.15555556,0.15555556,0.17777778,0.14444445,0.0,0.17777778,0.022222223,0.11111111,0.15555556,0.17777778,0.23333333,0.1,0.1,0.23333333,0.24444444,0.2888889,0.32222223,0.34444445,0.2777778,0.3,0.35555556,0.34444445,0.2777778,0.45555556,0.37777779,0.5222222,0.4,0.3888889,0.4,,0.45555556,0.5777778,0.5,0.43333334,0.56666666,0.6,0.6111111,0.5777778,0.6111111,0.6666667,0.6333333,0.6888889,0.62222224,0.6333333,0.82222223,0.7111111,0.7888889,0.84444445,0.8666667,0.8333333,0.8666667,0.8666667,0.87777776,0.95555556,0.85555553,0.9444444,0.8666667,0.95555556,0.9111111,1.0,0.87777776,0.98888886,1.0,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.08888889,0.055555556,0.1,0.07777778,0.15555556,0.14444445,0.13333334,0.14444445,0.12222222,0.13333334,0.12222222,0.14444445,0.2,0.08888889,0.17777778,0.14444445,0.14444445,0.24444444,0.22222222,0.2,0.21111111,0.24444444,0.21111111,0.31111112,0.34444445,0.31111112,0.3888889,0.37777779,0.4,0.31111112,0.41111112,0.36666667,0.43333334,0.37777779,0.4,0.36666667,0.54444444,,0.5,0.6,0.5222222,0.6111111,0.5555556,0.54444444,0.5888889,0.65555555,0.5888889,0.6,0.6111111,0.64444447,0.5777778,0.73333335,0.6333333,0.6666667,0.76666665,0.9444444,0.75555557,0.95555556,0.9111111,0.76666665,0.8666667,0.76666665,0.8666667,0.85555553,0.85555553,0.9444444,1.0,0.8111111,0.96666664,1.0,0.9,1.0
+TorchMLP (tuned),0.022222223,0.0,0.022222223,0.033333335,0.044444446,0.08888889,0.044444446,0.044444446,0.044444446,0.055555556,0.15555556,0.08888889,0.13333334,0.16666667,0.17777778,0.22222222,0.15555556,0.17777778,0.17777778,0.18888889,0.18888889,0.25555557,0.3,0.25555557,0.2,0.32222223,0.3888889,0.2777778,0.26666668,0.2777778,0.12222222,0.21111111,0.33333334,0.42222223,0.44444445,0.37777779,0.4888889,0.45555556,0.42222223,0.5,,0.51111114,0.4,0.4888889,0.54444444,0.64444447,0.5777778,0.5222222,0.62222224,0.65555555,0.6111111,0.65555555,0.62222224,0.64444447,0.73333335,0.8888889,0.6888889,0.84444445,0.87777776,0.84444445,0.84444445,0.87777776,0.95555556,0.9222222,0.95555556,0.9777778,0.9,0.9222222,0.9777778,0.8888889,0.98888886,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.08888889,0.033333335,0.1,0.13333334,0.15555556,0.13333334,0.16666667,0.033333335,0.17777778,0.2,0.23333333,0.14444445,0.022222223,0.26666668,0.17777778,0.13333334,0.18888889,0.24444444,0.3,0.12222222,0.2,0.24444444,0.18888889,0.33333334,0.31111112,0.26666668,0.33333334,0.35555556,0.31111112,0.34444445,0.33333334,0.31111112,0.26666668,0.37777779,0.44444445,0.34444445,0.33333334,0.5,0.4,0.4888889,,0.47777778,0.5222222,0.5777778,0.44444445,0.5222222,0.5555556,0.6,0.62222224,0.5888889,0.6111111,0.62222224,0.84444445,0.62222224,0.67777777,0.96666664,0.67777777,0.8888889,0.67777777,0.87777776,0.9,0.85555553,0.8333333,0.8666667,0.73333335,0.93333334,0.9777778,0.8666667,1.0,0.93333334,0.9,1.0,1.0
+EBM (default),0.0,0.0,0.044444446,0.044444446,0.033333335,0.06666667,0.055555556,0.11111111,0.08888889,0.033333335,0.022222223,0.033333335,0.06666667,0.15555556,0.13333334,0.16666667,0.17777778,0.14444445,0.07777778,0.16666667,0.26666668,0.21111111,0.14444445,0.23333333,0.2777778,0.32222223,0.4,0.35555556,0.2888889,0.43333334,0.36666667,0.07777778,0.35555556,0.37777779,0.5,0.25555557,0.42222223,0.47777778,0.56666666,0.47777778,0.6,0.5222222,,0.5777778,0.5555556,0.64444447,0.5555556,0.5888889,0.6111111,0.64444447,0.67777777,0.67777777,0.6,0.73333335,0.75555557,0.8333333,0.73333335,0.84444445,0.8111111,0.84444445,0.87777776,0.84444445,0.8666667,0.82222223,0.85555553,0.8666667,0.82222223,0.9111111,1.0,0.8,0.9777778,1.0,0.9,1.0
+TabDPT (tuned),0.044444446,0.033333335,0.033333335,0.033333335,0.044444446,0.055555556,0.18888889,0.18888889,0.08888889,0.2,0.15555556,0.2,0.16666667,0.18888889,0.11111111,0.12222222,0.12222222,0.18888889,0.2,0.17777778,0.26666668,0.23333333,0.25555557,0.16666667,0.22222222,0.14444445,0.21111111,0.22222222,0.16666667,0.11111111,0.2888889,0.3,0.34444445,0.53333336,0.37777779,0.41111112,0.3888889,0.4888889,0.43333334,0.3888889,0.51111114,0.47777778,0.42222223,,0.4888889,0.44444445,0.56666666,0.62222224,0.5888889,0.46666667,0.5777778,0.5222222,0.75,0.67777777,0.64444447,0.67777777,0.7777778,0.6666667,0.6888889,0.67777777,0.8,0.74444443,0.96666664,0.9,0.96666664,0.9111111,0.93333334,0.8,0.9111111,0.9,0.9,0.9777778,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.022222223,0.011111111,0.055555556,0.044444446,0.044444446,0.11111111,0.12222222,0.12222222,0.11111111,0.1,0.1,0.11111111,0.12222222,0.08888889,0.16666667,0.21111111,0.13333334,0.13333334,0.14444445,0.15555556,0.22222222,0.21111111,0.15555556,0.25555557,0.2888889,0.26666668,0.2777778,0.32222223,0.33333334,0.3,0.2777778,0.24444444,0.35555556,0.46666667,0.36666667,0.33333334,0.37777779,0.4,0.44444445,0.45555556,0.42222223,0.44444445,0.51111114,,0.51111114,0.44444445,0.53333336,0.74444443,0.47777778,0.6333333,0.6111111,0.5555556,0.6333333,0.6888889,0.7,0.7,0.8,0.8,0.8111111,0.8888889,0.85555553,0.85555553,0.96666664,0.8333333,0.93333334,0.96666664,0.95555556,0.93333334,0.8888889,0.96666664,1.0,0.9,1.0
+iLTM (default),0.0,0.0,0.033333335,0.011111111,0.055555556,0.07777778,0.07777778,0.055555556,0.0,0.06666667,0.033333335,0.022222223,0.044444446,0.18888889,0.14444445,0.16666667,0.15555556,0.1,0.06666667,0.17777778,0.26666668,0.24444444,0.06666667,0.16666667,0.16666667,0.23333333,0.23333333,0.23333333,0.31111112,0.36666667,0.24444444,0.2777778,0.18888889,0.45555556,0.35555556,0.36666667,0.3888889,0.5222222,0.3888889,0.45555556,0.35555556,0.5555556,0.35555556,0.5555556,0.4888889,,0.4888889,0.56666666,0.5,0.56666666,0.4888889,0.67777777,0.5777778,0.7222222,0.64444447,0.64444447,0.76666665,0.8888889,0.75555557,0.9,0.87777776,0.8,0.75555557,0.85555553,0.74444443,0.9222222,0.84444445,0.8888889,0.98888886,0.8888889,0.96666664,0.9,0.8,1.0
+LightGBM (default),0.0,0.0,0.044444446,0.011111111,0.055555556,0.055555556,0.044444446,0.06666667,0.07777778,0.07777778,0.033333335,0.022222223,0.12222222,0.06666667,0.07777778,0.0,0.08888889,0.18888889,0.06666667,0.11111111,0.17777778,0.13333334,0.033333335,0.24444444,0.12222222,0.13333334,0.2,0.23333333,0.21111111,0.22222222,0.24444444,0.32222223,0.46666667,0.53333336,0.37777779,0.41111112,0.37777779,0.54444444,0.42222223,0.41111112,0.42222223,0.47777778,0.44444445,0.43333334,0.5555556,0.51111114,,0.6,0.5777778,0.5,0.5555556,0.4888889,0.56666666,0.6,0.64444447,0.74444443,0.7,0.84444445,0.6888889,0.85555553,0.76666665,0.7,0.87777776,0.8888889,0.95555556,0.9444444,0.9444444,0.8111111,1.0,0.9,0.9222222,1.0,1.0,1.0
+LimiX (default),0.11111111,0.14444445,0.12222222,0.2,0.22222222,0.18888889,0.2777778,0.24444444,0.24444444,0.2777778,0.32222223,0.31111112,0.25555557,0.15555556,0.33333334,0.23333333,0.14444445,0.26666668,0.34444445,0.34444445,0.2,0.23333333,0.35555556,0.26666668,0.33333334,0.31111112,0.23333333,0.2777778,0.23333333,0.33333334,0.33333334,0.4,0.3888889,0.31111112,0.32222223,0.41111112,0.42222223,0.37777779,0.3888889,0.34444445,0.47777778,0.44444445,0.41111112,0.37777779,0.46666667,0.43333334,0.4,,0.47777778,0.46666667,0.5222222,0.41111112,0.54444444,0.5777778,0.5555556,0.4888889,0.6111111,0.5222222,0.64444447,0.53333336,0.56666666,0.65555555,0.6333333,0.75555557,0.67777777,0.6333333,0.7111111,0.65555555,0.62222224,0.7777778,0.65555555,0.67777777,0.7888889,0.8888889
+RandomForest (tuned),0.011111111,0.022222223,0.022222223,0.033333335,0.011111111,0.033333335,0.12222222,0.12222222,0.11111111,0.12222222,0.1,0.08888889,0.11111111,0.13333334,0.044444446,0.16666667,0.22222222,0.13333334,0.14444445,0.06666667,0.17777778,0.21111111,0.15555556,0.18888889,0.2,0.2777778,0.24444444,0.2888889,0.24444444,0.34444445,0.24444444,0.23333333,0.22222222,0.34444445,0.44444445,0.33333334,0.33333334,0.34444445,0.33333334,0.41111112,0.37777779,0.4,0.3888889,0.41111112,0.25555557,0.5,0.42222223,0.5222222,,0.41111112,0.54444444,0.5777778,0.51111114,0.56666666,0.6333333,0.5777778,0.6666667,0.7888889,0.7111111,0.8,0.85555553,0.7777778,0.8333333,0.9388889,0.8333333,0.9222222,0.9222222,0.9111111,0.93333334,0.85555553,0.9777778,1.0,0.9,1.0
+ModernNCA (default),0.0,0.0,0.011111111,0.022222223,0.022222223,0.044444446,0.044444446,0.011111111,0.022222223,0.06666667,0.011111111,0.022222223,0.08888889,0.14444445,0.17777778,0.1,0.044444446,0.14444445,0.033333335,0.21111111,0.24444444,0.16666667,0.06666667,0.18888889,0.2,0.13333334,0.15555556,0.1,0.3,0.21111111,0.26666668,0.2777778,0.2777778,0.43333334,0.34444445,0.32222223,0.2777778,0.43333334,0.36666667,0.4,0.34444445,0.37777779,0.35555556,0.53333336,0.5222222,0.43333334,0.5,0.53333336,0.5888889,,0.46666667,0.5777778,0.56666666,0.54444444,0.5555556,0.75555557,0.67777777,0.74444443,0.7,0.75555557,0.8333333,0.74444443,0.84444445,0.95555556,0.8666667,0.9111111,0.93333334,0.8666667,1.0,0.9,0.9222222,1.0,0.9222222,1.0
+ExtraTrees (tuned + ensembled),0.0,0.011111111,0.011111111,0.033333335,0.022222223,0.011111111,0.011111111,0.033333335,0.044444446,0.044444446,0.011111111,0.022222223,0.11111111,0.1,0.033333335,0.16666667,0.1,0.13333334,0.044444446,0.08888889,0.2,0.23333333,0.16666667,0.23333333,0.07777778,0.2,0.25555557,0.15555556,0.23333333,0.32222223,0.2,0.2,0.13333334,0.26666668,0.36666667,0.31111112,0.35555556,0.34444445,0.31111112,0.3888889,0.3888889,0.41111112,0.32222223,0.42222223,0.36666667,0.51111114,0.44444445,0.47777778,0.45555556,0.53333336,,0.4888889,0.53333336,0.53333336,0.75555557,0.64444447,0.64444447,0.7222222,0.7222222,0.7222222,0.74444443,0.7888889,0.75555557,0.98888886,0.8,0.9444444,0.9777778,0.9,0.85555553,0.9111111,0.96666664,0.9,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.022222223,0.033333335,0.07777778,0.044444446,0.07777778,0.022222223,0.044444446,0.022222223,0.033333335,0.044444446,0.044444446,0.022222223,0.11111111,0.14444445,0.12222222,0.14444445,0.06666667,0.055555556,0.16666667,0.21111111,0.2,0.17777778,0.044444446,0.1,0.22222222,0.15555556,0.21111111,0.3,0.25555557,0.17777778,0.12222222,0.16666667,0.31111112,0.36666667,0.17777778,0.37777779,0.2,0.37777779,0.35555556,0.34444445,0.3888889,0.32222223,0.47777778,0.3888889,0.32222223,0.51111114,0.5888889,0.42222223,0.42222223,0.51111114,,0.5777778,0.6111111,0.67777777,0.64444447,0.7111111,0.6666667,0.8111111,0.6888889,0.8888889,0.84444445,0.84444445,0.82222223,0.8666667,0.8666667,0.85555553,0.9777778,0.95555556,0.8333333,0.96666664,1.0,0.9,1.0
+TabDPT (default),0.022222223,0.022222223,0.044444446,0.033333335,0.055555556,0.044444446,0.11111111,0.1,0.06666667,0.12222222,0.11111111,0.044444446,0.16666667,0.17777778,0.1,0.12222222,0.07777778,0.22222222,0.14444445,0.15555556,0.25555557,0.2,0.16666667,0.17777778,0.14444445,0.07777778,0.14444445,0.13333334,0.08888889,0.055555556,0.23333333,0.32222223,0.34444445,0.41111112,0.14444445,0.4,0.37777779,0.43333334,0.36666667,0.42222223,0.37777779,0.37777779,0.4,0.25,0.44444445,0.42222223,0.43333334,0.45555556,0.4888889,0.43333334,0.46666667,0.42222223,,0.51111114,0.5,0.62222224,0.6333333,0.6666667,0.7,0.6666667,0.6111111,0.7111111,0.7888889,0.67777777,0.8333333,0.7222222,0.67777777,0.74444443,0.7111111,0.75555557,0.85555553,0.73333335,0.9777778,1.0
+FastaiMLP (tuned),0.0,0.022222223,0.022222223,0.06666667,0.055555556,0.06666667,0.033333335,0.022222223,0.033333335,0.044444446,0.12222222,0.08888889,0.11111111,0.022222223,0.12222222,0.08888889,0.06666667,0.14444445,0.13333334,0.12222222,0.033333335,0.12222222,0.21111111,0.13333334,0.2,0.2888889,0.14444445,0.07777778,0.25555557,0.2,0.16666667,0.24444444,0.2888889,0.2,0.32222223,0.26666668,0.22222222,0.17777778,0.17777778,0.26666668,0.35555556,0.15555556,0.26666668,0.32222223,0.36666667,0.2777778,0.4,0.42222223,0.43333334,0.45555556,0.46666667,0.3888889,0.4888889,,0.51111114,0.4888889,0.7111111,0.5555556,0.6666667,0.5555556,0.7111111,0.73333335,0.8,0.76666665,0.8333333,0.67777777,0.82222223,0.9,0.87777776,0.8888889,0.8666667,0.8888889,0.95555556,1.0
+ExtraTrees (tuned),0.0,0.011111111,0.0,0.033333335,0.011111111,0.011111111,0.022222223,0.011111111,0.011111111,0.033333335,0.011111111,0.022222223,0.11111111,0.12222222,0.033333335,0.13333334,0.11111111,0.13333334,0.022222223,0.06666667,0.14444445,0.16666667,0.07777778,0.15555556,0.055555556,0.16666667,0.13333334,0.14444445,0.17777778,0.21111111,0.13333334,0.15555556,0.1,0.2777778,0.3,0.24444444,0.32222223,0.25555557,0.2888889,0.36666667,0.26666668,0.37777779,0.24444444,0.35555556,0.31111112,0.35555556,0.35555556,0.44444445,0.36666667,0.44444445,0.24444444,0.32222223,0.5,0.4888889,,0.5555556,0.54444444,0.7111111,0.7111111,0.7,0.6888889,0.74444443,0.7111111,0.95555556,0.7777778,0.93333334,0.9444444,0.8888889,0.84444445,0.9,0.96666664,0.9111111,0.98888886,1.0
+TorchMLP (default),0.0,0.0,0.011111111,0.033333335,0.0,0.022222223,0.011111111,0.022222223,0.0,0.022222223,0.0,0.022222223,0.08888889,0.011111111,0.044444446,0.022222223,0.11111111,0.08888889,0.0,0.08888889,0.06666667,0.022222223,0.06666667,0.13333334,0.044444446,0.2,0.11111111,0.14444445,0.2,0.22222222,0.011111111,0.13333334,0.07777778,0.24444444,0.31111112,0.15555556,0.22222222,0.26666668,0.21111111,0.33333334,0.11111111,0.32222223,0.16666667,0.32222223,0.3,0.35555556,0.25555557,0.51111114,0.42222223,0.24444444,0.35555556,0.35555556,0.37777779,0.51111114,0.44444445,,0.5555556,0.73333335,0.73333335,0.73333335,0.8,0.76666665,0.8,0.7888889,0.82222223,0.9222222,0.8666667,0.87777776,0.9,0.8888889,0.95555556,0.98888886,0.95555556,1.0
+FastaiMLP (default),0.0,0.022222223,0.011111111,0.06666667,0.07777778,0.07777778,0.022222223,0.044444446,0.0,0.055555556,0.1,0.13333334,0.12222222,0.033333335,0.1,0.1,0.033333335,0.13333334,0.14444445,0.11111111,0.044444446,0.08888889,0.18888889,0.13333334,0.15555556,0.21111111,0.1,0.08888889,0.12222222,0.16666667,0.16666667,0.2,0.23333333,0.13333334,0.32222223,0.25555557,0.18888889,0.13333334,0.15555556,0.23333333,0.31111112,0.033333335,0.26666668,0.22222222,0.3,0.23333333,0.3,0.3888889,0.33333334,0.32222223,0.35555556,0.2888889,0.36666667,0.2888889,0.45555556,0.44444445,,0.47777778,0.5555556,0.4888889,0.6,0.6333333,0.7,0.6111111,0.7222222,0.6333333,0.62222224,0.67777777,0.7,0.8111111,0.8,0.8111111,0.8888889,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.011111111,0.011111111,0.011111111,0.0,0.0,0.0,0.0,0.07777778,0.011111111,0.011111111,0.011111111,0.022222223,0.0,0.0,0.011111111,0.0,0.06666667,0.055555556,0.044444446,0.11111111,0.14444445,0.07777778,0.022222223,0.055555556,0.14444445,0.2,0.055555556,0.1,0.16666667,0.13333334,0.055555556,0.15555556,0.32222223,0.15555556,0.33333334,0.2,0.11111111,0.15555556,0.47777778,0.21111111,0.25555557,0.2777778,0.33333334,0.33333334,0.44444445,0.2888889,0.26666668,0.5222222,,0.56666666,0.65555555,0.65555555,0.5777778,0.73333335,0.6,0.73333335,0.64444447,0.7,0.74444443,1.0,0.7,0.87777776,0.87777776,0.8,0.9
+Linear (tuned + ensembled),0.0,0.022222223,0.0,0.07777778,0.033333335,0.07777778,0.011111111,0.011111111,0.011111111,0.011111111,0.0,0.022222223,0.0,0.0,0.15555556,0.06666667,0.1,0.0,0.0,0.17777778,0.022222223,0.11111111,0.13333334,0.0,0.15555556,0.24444444,0.1,0.13333334,0.26666668,0.24444444,0.08888889,0.1,0.12222222,0.022222223,0.2777778,0.14444445,0.2,0.044444446,0.16666667,0.24444444,0.12222222,0.11111111,0.18888889,0.31111112,0.2,0.24444444,0.31111112,0.35555556,0.2888889,0.3,0.2777778,0.18888889,0.3,0.33333334,0.2888889,0.26666668,0.44444445,0.43333334,,0.43333334,0.42222223,0.81666666,0.44444445,0.5555556,0.47777778,0.5888889,0.54444444,0.93333334,0.5,0.7777778,0.6888889,0.62222224,0.76666665,0.9
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.011111111,0.011111111,0.011111111,0.0,0.0,0.0,0.0,0.07777778,0.011111111,0.011111111,0.0,0.022222223,0.0,0.0,0.0,0.0,0.055555556,0.055555556,0.044444446,0.11111111,0.14444445,0.07777778,0.033333335,0.044444446,0.14444445,0.2,0.055555556,0.1,0.15555556,0.13333334,0.044444446,0.15555556,0.32222223,0.15555556,0.32222223,0.18888889,0.1,0.14444445,0.46666667,0.2,0.24444444,0.2777778,0.31111112,0.33333334,0.44444445,0.3,0.26666668,0.51111114,0.34444445,0.56666666,,0.65555555,0.5777778,0.73333335,0.6,0.73333335,0.64444447,0.7,0.74444443,1.0,0.7,0.87777776,0.87777776,0.8,0.9
+KNN (tuned + ensembled),0.011111111,0.044444446,0.011111111,0.06666667,0.055555556,0.044444446,0.011111111,0.011111111,0.011111111,0.011111111,0.011111111,0.022222223,0.011111111,0.011111111,0.08888889,0.055555556,0.011111111,0.022222223,0.011111111,0.11111111,0.033333335,0.11111111,0.08888889,0.022222223,0.1,0.14444445,0.011111111,0.044444446,0.14444445,0.15555556,0.08888889,0.08888889,0.11111111,0.033333335,0.21111111,0.11111111,0.15555556,0.022222223,0.13333334,0.08888889,0.15555556,0.12222222,0.12222222,0.2,0.11111111,0.12222222,0.23333333,0.43333334,0.14444445,0.16666667,0.25555557,0.11111111,0.3888889,0.2888889,0.31111112,0.2,0.4,0.34444445,0.5777778,0.34444445,,0.6333333,0.53333336,0.6888889,0.5777778,0.5777778,0.56666666,0.7222222,0.6,0.7111111,0.6666667,1.0,0.84444445,1.0
+Linear (tuned),0.0,0.022222223,0.0,0.06666667,0.033333335,0.055555556,0.0,0.0,0.0,0.0,0.0,0.011111111,0.0,0.0,0.14444445,0.055555556,0.08888889,0.0,0.0,0.13333334,0.011111111,0.1,0.08888889,0.0,0.15555556,0.24444444,0.06666667,0.1,0.24444444,0.24444444,0.08888889,0.07777778,0.11111111,0.0,0.24444444,0.11111111,0.21111111,0.011111111,0.12222222,0.23333333,0.12222222,0.1,0.15555556,0.25555557,0.14444445,0.2,0.3,0.34444445,0.22222222,0.25555557,0.21111111,0.15555556,0.2888889,0.26666668,0.25555557,0.23333333,0.36666667,0.42222223,0.18333334,0.42222223,0.36666667,,0.43333334,0.51111114,0.45555556,0.56666666,0.51111114,0.8666667,0.5,0.7222222,0.65555555,0.62222224,0.74444443,0.9
+PerpetualBooster (tuned + ensembled),0.011111111,0.0,0.022222223,0.0,0.011111111,0.0,0.022222223,0.011111111,0.022222223,0.022222223,0.07777778,0.022222223,0.12222222,0.022222223,0.033333335,0.022222223,0.011111111,0.12222222,0.08888889,0.033333335,0.033333335,0.022222223,0.055555556,0.12222222,0.022222223,0.022222223,0.06666667,0.055555556,0.033333335,0.011111111,0.022222223,0.12222222,0.12222222,0.18888889,0.044444446,0.13333334,0.022222223,0.14444445,0.044444446,0.13333334,0.044444446,0.14444445,0.13333334,0.033333335,0.14444445,0.24444444,0.12222222,0.36666667,0.16666667,0.15555556,0.24444444,0.15555556,0.21111111,0.2,0.2888889,0.2,0.3,0.26666668,0.5555556,0.26666668,0.46666667,0.56666666,,0.53333336,0.85555553,0.6,0.6,0.56666666,0.36666667,0.65555555,0.6111111,0.65555555,0.9222222,0.95555556
+RandomForest (default),0.0,0.011111111,0.0,0.0,0.0,0.0,0.011111111,0.011111111,0.011111111,0.011111111,0.0,0.0,0.0,0.06666667,0.0,0.011111111,0.0,0.011111111,0.0,0.011111111,0.08888889,0.033333335,0.011111111,0.0,0.022222223,0.0,0.033333335,0.022222223,0.044444446,0.011111111,0.011111111,0.12222222,0.07777778,0.12222222,0.08888889,0.15555556,0.17777778,0.2,0.14444445,0.23333333,0.07777778,0.16666667,0.17777778,0.1,0.033333335,0.14444445,0.11111111,0.24444444,0.06111111,0.044444446,0.011111111,0.17777778,0.32222223,0.23333333,0.044444446,0.21111111,0.3888889,0.4,0.44444445,0.4,0.31111112,0.4888889,0.46666667,,0.46666667,0.4888889,0.6,0.51111114,0.51111114,0.8111111,0.5888889,0.65555555,0.7888889,1.0
+PerpetualBooster (tuned),0.0,0.0,0.033333335,0.0,0.011111111,0.0,0.011111111,0.011111111,0.033333335,0.011111111,0.033333335,0.022222223,0.11111111,0.011111111,0.022222223,0.0,0.011111111,0.12222222,0.033333335,0.055555556,0.044444446,0.011111111,0.033333335,0.13333334,0.011111111,0.0,0.07777778,0.044444446,0.033333335,0.022222223,0.033333335,0.13333334,0.06666667,0.14444445,0.022222223,0.14444445,0.022222223,0.16666667,0.055555556,0.13333334,0.044444446,0.13333334,0.14444445,0.033333335,0.16666667,0.25555557,0.044444446,0.32222223,0.16666667,0.13333334,0.2,0.13333334,0.16666667,0.16666667,0.22222222,0.17777778,0.2777778,0.26666668,0.5222222,0.26666668,0.42222223,0.54444444,0.14444445,0.53333336,,0.6,0.54444444,0.53333336,0.31111112,0.64444447,0.54444444,0.5777778,0.9,0.95555556
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.011111111,0.011111111,0.011111111,0.022222223,0.0,0.011111111,0.0,0.0,0.011111111,0.0,0.044444446,0.011111111,0.0,0.011111111,0.022222223,0.0,0.011111111,0.0,0.0,0.0,0.022222223,0.011111111,0.1,0.011111111,0.0,0.12222222,0.011111111,0.033333335,0.0,0.12222222,0.022222223,0.13333334,0.13333334,0.14444445,0.022222223,0.26666668,0.13333334,0.08888889,0.06666667,0.07777778,0.055555556,0.36666667,0.07777778,0.08888889,0.055555556,0.13333334,0.2777778,0.32222223,0.06666667,0.07777778,0.36666667,0.35555556,0.41111112,0.35555556,0.42222223,0.43333334,0.4,0.51111114,0.4,,0.6111111,0.5,0.54444444,0.6333333,0.6333333,0.4888889,0.64444447,0.9444444
+ExtraTrees (default),0.0,0.011111111,0.011111111,0.0,0.0,0.0,0.011111111,0.022222223,0.011111111,0.022222223,0.011111111,0.011111111,0.011111111,0.0,0.011111111,0.0,0.022222223,0.011111111,0.011111111,0.022222223,0.011111111,0.022222223,0.011111111,0.011111111,0.011111111,0.0,0.033333335,0.033333335,0.07777778,0.011111111,0.011111111,0.11111111,0.07777778,0.022222223,0.07777778,0.16666667,0.055555556,0.055555556,0.044444446,0.14444445,0.1,0.06666667,0.17777778,0.06666667,0.033333335,0.15555556,0.055555556,0.2888889,0.07777778,0.06666667,0.022222223,0.14444445,0.32222223,0.17777778,0.055555556,0.13333334,0.37777779,0.3,0.45555556,0.3,0.43333334,0.4888889,0.4,0.4,0.45555556,0.3888889,,0.5222222,0.45555556,0.7111111,0.5888889,0.65555555,0.7888889,1.0
+Linear (default),0.0,0.0,0.0,0.044444446,0.033333335,0.033333335,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.022222223,0.022222223,0.0,0.0,0.06666667,0.0,0.06666667,0.033333335,0.0,0.07777778,0.06666667,0.0,0.011111111,0.12222222,0.11111111,0.044444446,0.055555556,0.1,0.0,0.18888889,0.07777778,0.1,0.0,0.08888889,0.055555556,0.07777778,0.022222223,0.08888889,0.2,0.044444446,0.11111111,0.18888889,0.34444445,0.08888889,0.13333334,0.1,0.022222223,0.25555557,0.1,0.11111111,0.12222222,0.32222223,0.25555557,0.06666667,0.25555557,0.2777778,0.13333334,0.43333334,0.4888889,0.46666667,0.5,0.47777778,,0.47777778,0.7111111,0.5222222,0.62222224,0.74444443,0.8888889
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.011111111,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.07777778,0.0,0.0,0.011111111,0.0,0.0,0.022222223,0.13333334,0.0,0.08888889,0.06666667,0.011111111,0.0,0.37777779,0.06666667,0.0,0.14444445,0.044444446,0.2888889,0.12222222,0.15555556,0.1,0.3,0.0,0.5,0.0,0.4,0.5,0.6333333,0.4888889,0.6888889,0.45555556,0.54444444,0.5222222,,0.6333333,0.67777777,0.7,0.8,0.87777776
+xRFM (default),0.0,0.044444446,0.0,0.07777778,0.07777778,0.07777778,0.0,0.0,0.0,0.0,0.011111111,0.044444446,0.0,0.0,0.07777778,0.08888889,0.0,0.0,0.044444446,0.1,0.022222223,0.1,0.08888889,0.0,0.1,0.06666667,0.011111111,0.044444446,0.08888889,0.07777778,0.1,0.17777778,0.1,0.0,0.1,0.18888889,0.1,0.011111111,0.12222222,0.18888889,0.11111111,0.0,0.2,0.1,0.11111111,0.11111111,0.1,0.22222222,0.14444445,0.1,0.08888889,0.16666667,0.24444444,0.11111111,0.1,0.11111111,0.18888889,0.3,0.22222222,0.3,0.2888889,0.2777778,0.34444445,0.18888889,0.35555556,0.36666667,0.2888889,0.2888889,0.36666667,,0.43333334,0.4888889,0.53333336,0.8666667
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.011111111,0.011111111,0.0,0.011111111,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.011111111,0.022222223,0.0,0.011111111,0.0,0.0,0.022222223,0.011111111,0.022222223,0.11111111,0.033333335,0.0,0.022222223,0.0,0.022222223,0.0,0.011111111,0.044444446,0.044444446,0.011111111,0.033333335,0.011111111,0.06666667,0.022222223,0.1,0.033333335,0.033333335,0.07777778,0.34444445,0.022222223,0.07777778,0.033333335,0.033333335,0.14444445,0.13333334,0.033333335,0.044444446,0.2,0.12222222,0.31111112,0.12222222,0.33333334,0.34444445,0.3888889,0.41111112,0.45555556,0.36666667,0.41111112,0.47777778,0.32222223,0.56666666,,0.5555556,0.64444447,0.8
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.011111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.033333335,0.033333335,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.022222223,0.0,0.1,0.0,0.32222223,0.0,0.0,0.1,0.0,0.26666668,0.11111111,0.08888889,0.011111111,0.18888889,0.12222222,0.37777779,0.12222222,0.0,0.37777779,0.34444445,0.34444445,0.42222223,0.51111114,0.34444445,0.37777779,0.3,0.51111114,0.44444445,,0.7111111,0.98888886
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07777778,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.1,0.0,0.1,0.0,0.1,0.0,0.0,0.1,0.0,0.1,0.2,0.0,0.21111111,0.1,0.07777778,0.0,0.1,0.022222223,0.044444446,0.011111111,0.044444446,0.11111111,0.2,0.23333333,0.2,0.15555556,0.25555557,0.07777778,0.21111111,0.1,0.35555556,0.21111111,0.25555557,0.2,0.46666667,0.35555556,0.2888889,,0.87777776
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.0,0.1,0.044444446,0.0,0.044444446,0.055555556,0.0,0.11111111,0.12222222,0.13333334,0.2,0.011111111,0.12222222,
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_small/leaderboard_overview_explorer.html
index 5196e552d10da5d74af1816748314b6f9ca30aa4..bcc5d3ddf0196d38fcde5e4710eb9a7dfcb58efe 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Classification | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_small/leaderboard_table.html b/data/imputation_no/splits_all/tasks_classification/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..41c4872a0040c1ec336c17a28b6d755a991c3724
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_classification/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1708.0,"score":0.881,"rank":7.97,"hrank":1.74,"imp":5.473,"train_time":38.57,"predict_time":6.317,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+168\/-140","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1616.0,"score":0.615,"rank":11.64,"hrank":5.0,"imp":10.971,"train_time":363.16,"predict_time":6.598,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-81","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1612.0,"score":0.651,"rank":11.83,"hrank":4.92,"imp":13.2,"train_time":7.17,"predict_time":0.407,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-80","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1581.0,"score":0.575,"rank":13.36,"hrank":5.96,"imp":13.57,"train_time":2289.39,"predict_time":8.114,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-76","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1576.0,"score":0.589,"rank":13.59,"hrank":7.07,"imp":14.377,"train_time":6.26,"predict_time":0.539,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-68","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1567.0,"score":0.568,"rank":14.08,"hrank":7.75,"imp":14.225,"train_time":6.08,"predict_time":0.762,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-71","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1564.0,"score":0.561,"rank":14.23,"hrank":6.41,"imp":14.22,"train_time":2289.39,"predict_time":1.253,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-66","imputed":false},{"position":7,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1562.0,"score":0.621,"rank":14.38,"hrank":5.45,"imp":13.723,"train_time":3.22,"predict_time":0.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-84","imputed":false},{"position":8,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1458.0,"score":0.412,"rank":20.81,"hrank":9.18,"imp":16.34,"train_time":27.77,"predict_time":5.044,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-77","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1452.0,"score":0.363,"rank":21.24,"hrank":12.63,"imp":17.342,"train_time":3423.61,"predict_time":18.933,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-52","imputed":false},{"position":10,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1430.0,"score":0.33,"rank":22.79,"hrank":11.58,"imp":18.367,"train_time":2469.48,"predict_time":3.418,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-55","imputed":false},{"position":11,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1430.0,"score":0.368,"rank":22.81,"hrank":9.34,"imp":18.089,"train_time":4.35,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-63","imputed":false},{"position":12,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1424.0,"score":0.303,"rank":23.27,"hrank":13.98,"imp":18.167,"train_time":3144.62,"predict_time":1.882,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-55","imputed":false},{"position":13,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1401.0,"score":0.326,"rank":24.95,"hrank":11.81,"imp":18.376,"train_time":5271.47,"predict_time":231.832,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-60","imputed":false},{"position":14,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1393.0,"score":0.319,"rank":25.62,"hrank":14.55,"imp":19.682,"train_time":331.0,"predict_time":2.359,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-50","imputed":false},{"position":15,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1382.0,"score":0.273,"rank":26.45,"hrank":13.23,"imp":19.215,"train_time":2469.48,"predict_time":0.408,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-60","imputed":false},{"position":16,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1378.0,"score":0.228,"rank":26.77,"hrank":19.94,"imp":19.414,"train_time":735.37,"predict_time":1.9,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-34","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1374.0,"score":0.259,"rank":27.12,"hrank":14.15,"imp":18.846,"train_time":3423.61,"predict_time":1.533,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-54","imputed":false},{"position":18,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1370.0,"score":0.283,"rank":27.39,"hrank":12.23,"imp":19.062,"train_time":5271.47,"predict_time":33.978,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-57","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1352.0,"score":0.188,"rank":28.82,"hrank":21.92,"imp":19.779,"train_time":735.37,"predict_time":0.311,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-34","imputed":false},{"position":20,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1345.0,"score":0.274,"rank":29.39,"hrank":17.13,"imp":20.508,"train_time":2.24,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-58","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1345.0,"score":0.195,"rank":29.4,"hrank":21.21,"imp":19.791,"train_time":1542.95,"predict_time":0.396,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-35","imputed":false},{"position":22,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1339.0,"score":0.174,"rank":29.91,"hrank":20.74,"imp":19.902,"train_time":21290.02,"predict_time":663.263,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-49","imputed":false},{"position":23,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1334.0,"score":0.179,"rank":30.27,"hrank":21.99,"imp":20.047,"train_time":1542.95,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-31","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1330.0,"score":0.189,"rank":30.58,"hrank":15.16,"imp":20.148,"train_time":6049.28,"predict_time":0.525,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-49","imputed":false},{"position":25,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1316.0,"score":0.167,"rank":31.81,"hrank":22.92,"imp":20.665,"train_time":7.18,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-32","imputed":false},{"position":26,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1314.0,"score":0.271,"rank":31.97,"hrank":10.58,"imp":21.444,"train_time":13.27,"predict_time":1.624,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-73","imputed":false},{"position":27,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1312.0,"score":0.188,"rank":32.14,"hrank":19.39,"imp":20.763,"train_time":1053.94,"predict_time":3.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-62","imputed":false},{"position":28,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1308.0,"score":0.149,"rank":32.45,"hrank":23.87,"imp":20.595,"train_time":798.08,"predict_time":2.227,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-48","imputed":false},{"position":29,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1299.0,"score":0.193,"rank":33.17,"hrank":18.93,"imp":21.029,"train_time":8.79,"predict_time":0.411,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-61","imputed":false},{"position":30,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1298.0,"score":0.205,"rank":33.26,"hrank":17.06,"imp":21.08,"train_time":6049.28,"predict_time":8.736,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-78","imputed":false},{"position":31,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1291.0,"score":0.121,"rank":33.83,"hrank":23.33,"imp":20.817,"train_time":798.08,"predict_time":0.353,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-41","imputed":false},{"position":32,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1286.0,"score":0.211,"rank":34.31,"hrank":16.28,"imp":21.496,"train_time":51.45,"predict_time":43.707,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-78","imputed":false},{"position":33,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1279.0,"score":0.119,"rank":34.88,"hrank":25.0,"imp":21.858,"train_time":2499.4,"predict_time":0.174,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-37","imputed":false},{"position":34,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1270.0,"score":0.148,"rank":35.64,"hrank":20.38,"imp":21.94,"train_time":1053.94,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-52","imputed":false},{"position":35,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1269.0,"score":0.136,"rank":35.73,"hrank":24.97,"imp":22.222,"train_time":1142.82,"predict_time":0.536,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-60","imputed":false},{"position":36,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1268.0,"score":0.126,"rank":35.78,"hrank":25.84,"imp":21.443,"train_time":2977.91,"predict_time":2.415,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-49","imputed":false},{"position":37,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1254.0,"score":0.114,"rank":36.92,"hrank":25.81,"imp":22.493,"train_time":1142.82,"predict_time":0.043,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-47","imputed":false},{"position":38,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1252.0,"score":0.085,"rank":37.18,"hrank":26.52,"imp":22.427,"train_time":2499.4,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-42","imputed":false},{"position":39,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1248.0,"score":0.112,"rank":37.48,"hrank":19.98,"imp":23.719,"train_time":15.55,"predict_time":0.346,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-60","imputed":false},{"position":40,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1242.0,"score":0.145,"rank":37.99,"hrank":20.32,"imp":22.515,"train_time":1424.22,"predict_time":9.206,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-60","imputed":false},{"position":41,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1240.0,"score":0.076,"rank":38.11,"hrank":27.23,"imp":22.192,"train_time":21290.02,"predict_time":161.221,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+44\/-39","imputed":false},{"position":42,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1233.0,"score":0.089,"rank":38.72,"hrank":24.29,"imp":23.069,"train_time":7.09,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-48","imputed":false},{"position":43,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1216.0,"score":0.104,"rank":40.16,"hrank":24.02,"imp":23.684,"train_time":4.96,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-61","imputed":false},{"position":44,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1213.0,"score":0.064,"rank":40.44,"hrank":27.58,"imp":23.064,"train_time":14.89,"predict_time":4.432,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+42\/-40","imputed":false},{"position":45,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1203.0,"score":0.099,"rank":41.22,"hrank":25.17,"imp":24.079,"train_time":703.03,"predict_time":1.431,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-72","imputed":false},{"position":46,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1200.0,"score":0.078,"rank":41.51,"hrank":29.77,"imp":23.488,"train_time":2977.91,"predict_time":0.156,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-58","imputed":false},{"position":47,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1191.0,"score":0.07,"rank":42.24,"hrank":25.38,"imp":23.832,"train_time":1424.22,"predict_time":0.749,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-54","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1179.0,"score":0.091,"rank":43.22,"hrank":23.06,"imp":24.906,"train_time":703.03,"predict_time":0.184,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-72","imputed":false},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1176.0,"score":0.029,"rank":43.46,"hrank":36.52,"imp":23.64,"train_time":2.8,"predict_time":0.238,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-46","imputed":false},{"position":50,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1163.0,"score":0.062,"rank":44.5,"hrank":35.31,"imp":23.967,"train_time":3.15,"predict_time":0.141,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-51","imputed":false},{"position":51,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1155.0,"score":0.103,"rank":45.21,"hrank":29.57,"imp":26.004,"train_time":15.45,"predict_time":2.611,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-74","imputed":false},{"position":52,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1149.0,"score":0.083,"rank":45.64,"hrank":25.08,"imp":25.599,"train_time":571.73,"predict_time":1.389,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-87","imputed":false},{"position":53,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1148.0,"score":0.175,"rank":45.73,"hrank":11.7,"imp":26.141,"train_time":40588.15,"predict_time":5.606,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+94\/-104","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1144.0,"score":0.169,"rank":46.05,"hrank":11.35,"imp":26.22,"train_time":40588.15,"predict_time":22.575,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-104","imputed":false},{"position":55,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1113.0,"score":0.055,"rank":48.48,"hrank":23.77,"imp":26.37,"train_time":571.73,"predict_time":0.129,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-78","imputed":false},{"position":56,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1110.0,"score":0.05,"rank":48.72,"hrank":28.74,"imp":26.217,"train_time":463.05,"predict_time":147.937,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-70","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1075.0,"score":0.037,"rank":51.3,"hrank":36.79,"imp":32.343,"train_time":391.09,"predict_time":1.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-76","imputed":false},{"position":58,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1064.0,"score":0.015,"rank":52.13,"hrank":35.97,"imp":27.638,"train_time":11.29,"predict_time":0.204,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-62","imputed":false},{"position":59,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1050.0,"score":0.115,"rank":53.13,"hrank":18.72,"imp":31.616,"train_time":502.34,"predict_time":7.328,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-139","imputed":false},{"position":60,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1043.0,"score":0.07,"rank":53.58,"hrank":32.88,"imp":30.551,"train_time":2.59,"predict_time":0.533,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-94","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1040.0,"score":0.024,"rank":53.77,"hrank":36.31,"imp":33.844,"train_time":391.09,"predict_time":0.404,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-80","imputed":false},{"position":62,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1030.0,"score":0.024,"rank":54.47,"hrank":46.55,"imp":29.894,"train_time":4.95,"predict_time":1.012,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-104","imputed":false},{"position":63,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1027.0,"score":0.054,"rank":54.69,"hrank":34.52,"imp":33.75,"train_time":483.07,"predict_time":0.874,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-92","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1013.0,"score":0.004,"rank":55.6,"hrank":35.77,"imp":28.963,"train_time":5.04,"predict_time":0.645,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-74","imputed":false},{"position":65,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1011.0,"score":0.028,"rank":55.72,"hrank":34.36,"imp":31.523,"train_time":121.17,"predict_time":1.63,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-101","imputed":false},{"position":66,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1005.0,"score":0.039,"rank":56.11,"hrank":35.92,"imp":34.268,"train_time":483.07,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-94","imputed":false},{"position":67,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.009,"rank":56.42,"hrank":48.94,"imp":31.571,"train_time":0.46,"predict_time":0.068,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-79","imputed":false},{"position":68,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.012,"rank":56.43,"hrank":48.39,"imp":30.844,"train_time":3.61,"predict_time":0.552,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-97","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":953.0,"score":0.0,"rank":59.19,"hrank":39.87,"imp":36.5,"train_time":56.11,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-77","imputed":false},{"position":70,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":932.0,"score":0.024,"rank":60.34,"hrank":40.75,"imp":37.36,"train_time":2.39,"predict_time":0.143,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-124","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":922.0,"score":0.016,"rank":60.82,"hrank":41.61,"imp":36.89,"train_time":121.17,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+101\/-103","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":602.0,"score":0.0,"rank":70.62,"hrank":64.24,"imp":51.718,"train_time":0.28,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-138","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_small/pareto_front_explorer.html
index f6332b92a1f2b26090159422496da3f2bbd70ddd..e911846d3e6e30980b36c31856e240cfb088667f 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3591113273,"imp":19.6818773562,"elo":1392.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404555565,"imp":20.6645722803,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.044036937,"imp":20.047434755,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"imp":19.7905829027,"elo":1345.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.044213771,"imp":23.6835289232,"elo":1216.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0431349373,"imp":22.4932658301,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5357489286,"imp":22.2221441416,"elo":1268.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0222777214,"imp":23.0691921114,"elo":1233.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0223863596,"imp":22.4268877997,"elo":1251.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1740181026,"imp":21.8579315544,"elo":1278.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5520616377,"imp":30.8437241659,"elo":999.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1838318082,"imp":24.9061196002,"elo":1179.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4313148571,"imp":24.079427638,"elo":1203.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6452165158,"imp":28.9628415623,"elo":1013.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7488523806,"imp":23.8322360808,"elo":1191.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.2058236264,"imp":22.5154742531,"elo":1241.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.140680637,"imp":23.9668455927,"elo":1163.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3111174713,"imp":19.7792904389,"elo":1352.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.8999116636,"imp":19.4144249479,"elo":1377.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.0441668928,"imp":16.3398465162,"elo":1458.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"imp":23.7188606715,"elo":1248.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5246167968,"imp":20.1478549923,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7364459708,"imp":21.0798892713,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6112675715,"imp":26.0035391466,"elo":1154.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0479211725,"imp":36.500149994,"elo":953.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4043784213,"imp":33.8439929825,"elo":1040.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0518021797,"imp":32.3426224355,"elo":1075.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0681202832,"imp":31.5707800518,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1287782959,"imp":26.3700814072,"elo":1113.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.3894088878,"imp":25.5988568541,"elo":1149.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.4321040652,"imp":23.0637332597,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.5333309154,"imp":18.8455485777,"elo":1373.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":18.9330661418,"imp":17.3424924375,"elo":1451.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7616019805,"imp":14.2254580506,"elo":1567.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2533832566,"imp":14.2196655566,"elo":1564.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1139940547,"imp":13.5696834901,"elo":1580.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6235508255,"imp":21.4438972364,"elo":1313.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"imp":21.4959174878,"elo":1285.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":33.9783644772,"imp":19.0622291349,"elo":1370.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":231.8320376913,"imp":18.3764805087,"elo":1401.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"imp":18.0894735111,"elo":1430.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.3168064912,"imp":5.4732701081,"elo":1708.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5331077226,"imp":30.5512105741,"elo":1043.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"imp":13.7229337984,"elo":1561.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4108430406,"imp":21.028840173,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4078633109,"imp":19.2151606606,"elo":1382.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.4179947198,"imp":18.3667863933,"elo":1430.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"imp":14.3770199443,"elo":1576.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4074238608,"imp":13.2003163017,"elo":1611.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.3276990403,"imp":31.6159484362,"elo":1049.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.6062080426,"imp":26.1414204452,"elo":1148.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.5750090197,"imp":26.220144849,"elo":1144.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0545537099,"imp":20.5080779697,"elo":1345.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.203890602,"imp":27.6384946553,"elo":1064.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.155776825,"imp":23.4875165705,"elo":1199.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4148361224,"imp":21.4428206781,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2382075678,"imp":23.6402267025,"elo":1176.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3532173564,"imp":20.8170147115,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.2269115228,"imp":20.5947249427,"elo":1307.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":147.9369405767,"imp":26.2171130483,"elo":1109.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":161.2206474003,"imp":22.1919392314,"elo":1240.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":663.2627455245,"imp":19.9017604501,"elo":1338.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.0121457497,"imp":29.8942568536,"elo":1030.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1687223142,"imp":21.9404879917,"elo":1269.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.0619945914,"imp":20.7630227422,"elo":1311.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3591113273,"x_train":331.0009616013,"imp":19.6818773562,"elo":1392.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404555565,"x_train":7.1761452725,"imp":20.6645722803,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.044036937,"x_train":1542.9492660346,"imp":20.047434755,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"x_train":1542.9492660346,"imp":19.7905829027,"elo":1345.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.044213771,"x_train":4.9608581751,"imp":23.6835289232,"elo":1216.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0431349373,"x_train":1142.8199158523,"imp":22.4932658301,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5357489286,"x_train":1142.8199158523,"imp":22.2221441416,"elo":1268.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0222777214,"x_train":7.0915913087,"imp":23.0691921114,"elo":1233.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0223863596,"x_train":2499.3997394449,"imp":22.4268877997,"elo":1251.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1740181026,"x_train":2499.3997394449,"imp":21.8579315544,"elo":1278.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5520616377,"x_train":3.6063372166,"imp":30.8437241659,"elo":999.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1838318082,"x_train":703.0330785914,"imp":24.9061196002,"elo":1179.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4313148571,"x_train":703.0330785914,"imp":24.079427638,"elo":1203.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6452165158,"x_train":5.0370258118,"imp":28.9628415623,"elo":1013.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7488523806,"x_train":1424.216740031,"imp":23.8322360808,"elo":1191.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.2058236264,"x_train":1424.216740031,"imp":22.5154742531,"elo":1241.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.140680637,"x_train":3.149413842,"imp":23.9668455927,"elo":1163.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3111174713,"x_train":735.3700351286,"imp":19.7792904389,"elo":1352.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.8999116636,"x_train":735.3700351286,"imp":19.4144249479,"elo":1377.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.0441668928,"x_train":27.7685088788,"imp":16.3398465162,"elo":1458.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"x_train":15.5476029367,"imp":23.7188606715,"elo":1248.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5246167968,"x_train":6049.2787189098,"imp":20.1478549923,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7364459708,"x_train":6049.2787189098,"imp":21.0798892713,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6112675715,"x_train":15.4475945518,"imp":26.0035391466,"elo":1154.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0479211725,"x_train":56.1148745356,"imp":36.500149994,"elo":953.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4043784213,"x_train":391.0862000393,"imp":33.8439929825,"elo":1040.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0518021797,"x_train":391.0862000393,"imp":32.3426224355,"elo":1075.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0681202832,"x_train":0.4628071975,"imp":31.5707800518,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1287782959,"x_train":571.7337775834,"imp":26.3700814072,"elo":1113.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.3894088878,"x_train":571.7337775834,"imp":25.5988568541,"elo":1149.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.4321040652,"x_train":14.8938989136,"imp":23.0637332597,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.5333309154,"x_train":3423.6100841542,"imp":18.8455485777,"elo":1373.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":18.9330661418,"x_train":3423.6100841542,"imp":17.3424924375,"elo":1451.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7616019805,"x_train":6.0754426474,"imp":14.2254580506,"elo":1567.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2533832566,"x_train":2289.3887157667,"imp":14.2196655566,"elo":1564.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1139940547,"x_train":2289.3887157667,"imp":13.5696834901,"elo":1580.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6235508255,"x_train":13.2671754693,"imp":21.4438972364,"elo":1313.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"x_train":51.4453139381,"imp":21.4959174878,"elo":1285.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":33.9783644772,"x_train":5271.4650532739,"imp":19.0622291349,"elo":1370.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":231.8320376913,"x_train":5271.4650532739,"imp":18.3764805087,"elo":1401.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"x_train":4.3495748865,"imp":18.0894735111,"elo":1430.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.3168064912,"x_train":38.5654426559,"imp":5.4732701081,"elo":1708.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5331077226,"x_train":2.5922354756,"imp":30.5512105741,"elo":1043.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"x_train":3.2221111875,"imp":13.7229337984,"elo":1561.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4108430406,"x_train":8.785094234,"imp":21.028840173,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4078633109,"x_train":2469.4800149006,"imp":19.2151606606,"elo":1382.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.4179947198,"x_train":2469.4800149006,"imp":18.3667863933,"elo":1430.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"x_train":6.2559884135,"imp":14.3770199443,"elo":1576.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4074238608,"x_train":7.1656661471,"imp":13.2003163017,"elo":1611.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.3276990403,"x_train":502.344886049,"imp":31.6159484362,"elo":1049.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.6062080426,"x_train":40588.1540174129,"imp":26.1414204452,"elo":1148.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.5750090197,"x_train":40588.1540174129,"imp":26.220144849,"elo":1144.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0545537099,"x_train":2.2399715166,"imp":20.5080779697,"elo":1345.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.203890602,"x_train":11.2922428764,"imp":27.6384946553,"elo":1064.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.155776825,"x_train":2977.9095379662,"imp":23.4875165705,"elo":1199.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4148361224,"x_train":2977.9095379662,"imp":21.4428206781,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2382075678,"x_train":2.7957309689,"imp":23.6402267025,"elo":1176.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3532173564,"x_train":798.0801278695,"imp":20.8170147115,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.2269115228,"x_train":798.0801278695,"imp":20.5947249427,"elo":1307.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":147.9369405767,"x_train":463.0485235354,"imp":26.2171130483,"elo":1109.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":161.2206474003,"x_train":21290.0189805033,"imp":22.1919392314,"elo":1240.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":663.2627455245,"x_train":21290.0189805033,"imp":19.9017604501,"elo":1338.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.0121457497,"x_train":4.9505740724,"imp":29.8942568536,"elo":1030.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1687223142,"x_train":1053.935912149,"imp":21.9404879917,"elo":1269.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.0619945914,"x_train":1053.935912149,"imp":20.7630227422,"elo":1311.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_small/tuning_trajectories_explorer.html
index d8f06605d0a31d488696a63d67df25331048670b..4d701648e7112e9d358fd6c8387598335215f0ac 100644
--- a/data/imputation_no/splits_all/tasks_classification/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":331.0009616013,"imp":19.7104018173,"elo":1409.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":7.1761452725,"imp":20.6917599107,"elo":1330.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":13.6550050397,"imp":20.097358298,"elo":1353.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":34.7005566859,"imp":19.8311243875,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":69.9246996151,"imp":19.7555111737,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":185.5351574323,"imp":19.7099238429,"elo":1369.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":379.2610359226,"imp":19.731934615,"elo":1367.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":771.7715864818,"imp":19.7661508016,"elo":1364.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1152.9335256343,"imp":19.8007946703,"elo":1361.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1542.9492660346,"imp":19.8150269011,"elo":1359.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.9608581751,"imp":23.7100097,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.8616521277,"imp":23.3685843207,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":27.0121608835,"imp":23.1959832746,"elo":1248.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":54.6281407695,"imp":23.0298512961,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":140.6370736031,"imp":22.6970659011,"elo":1265.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":283.4636990992,"imp":22.5418724954,"elo":1271.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":569.3917360284,"imp":22.3660582501,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":853.7391955308,"imp":22.2920716358,"elo":1280.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1142.8199158523,"imp":22.2483144507,"elo":1282.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.0915913087,"imp":23.0968513065,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":16.5198837522,"imp":22.8508490847,"elo":1247.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":59.6568017021,"imp":22.5002056933,"elo":1263.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":123.907431834,"imp":22.3818935455,"elo":1269.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":318.4600106407,"imp":22.2211672007,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":623.1113495163,"imp":22.0705940025,"elo":1283.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1256.5403166692,"imp":21.9552776509,"elo":1288.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1885.0271107114,"imp":21.9168997328,"elo":1291.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2499.3997394449,"imp":21.8910463161,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.6063372166,"imp":30.8674670891,"elo":999.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.3528355841,"imp":28.798766118,"elo":1044.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":18.0710159165,"imp":25.9261230814,"elo":1117.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":35.5062409712,"imp":25.0138733194,"elo":1162.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":87.9242300599,"imp":24.452689118,"elo":1193.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":175.4018189887,"imp":24.2951625133,"elo":1202.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":350.0379937538,"imp":24.1864668973,"elo":1209.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":524.8382968951,"imp":24.1167676549,"elo":1214.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":703.0330785914,"imp":24.1067040677,"elo":1214.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.0370258118,"imp":28.9894994822,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":12.7722389468,"imp":27.4434440982,"elo":1068.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":35.5543374641,"imp":25.1090696531,"elo":1149.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":67.9976554587,"imp":24.0378613931,"elo":1191.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":172.4339504664,"imp":23.156687283,"elo":1224.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":350.2703386124,"imp":22.7751451617,"elo":1242.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":703.7735059836,"imp":22.618651403,"elo":1250.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1059.2142650988,"imp":22.5409104157,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1424.216740031,"imp":22.5380588141,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":3.149413842,"imp":23.9933364168,"elo":1165.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":7.1207733159,"imp":21.799911592,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":17.9053372504,"imp":20.5967936552,"elo":1334.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":35.9737069678,"imp":20.1137675506,"elo":1361.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":91.1663943012,"imp":19.786371996,"elo":1379.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":182.6006011031,"imp":19.6743028228,"elo":1386.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":365.5882548593,"imp":19.5275124529,"elo":1394.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":548.151079493,"imp":19.471737158,"elo":1396.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":735.3700351286,"imp":19.440619285,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":27.7685088788,"imp":16.4063984503,"elo":1475.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":15.5476029367,"imp":23.7452802097,"elo":1259.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":52.4814325925,"imp":22.4237173732,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":150.689630028,"imp":20.3212043231,"elo":1352.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":293.311977236,"imp":19.956113518,"elo":1363.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":719.031159336,"imp":19.5895978361,"elo":1375.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1498.353139385,"imp":19.8851630726,"elo":1360.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3004.008741271,"imp":20.4709045447,"elo":1338.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4473.3946349608,"imp":20.8865722454,"elo":1322.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6049.2787189098,"imp":21.1122915159,"elo":1311.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":15.4475945518,"imp":26.0247643412,"elo":1162.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":56.1148745356,"imp":37.2703231919,"elo":945.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":121.64643757,"imp":35.195351398,"elo":1000.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":327.8283737012,"imp":32.5599522417,"elo":1077.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":391.0862000393,"imp":32.3630388368,"elo":1084.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4628071975,"imp":31.6097744515,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.2094586922,"imp":28.8655853723,"elo":1044.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":11.9623791751,"imp":27.3622202338,"elo":1078.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":26.2110927853,"imp":26.5336916255,"elo":1107.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":68.7900402557,"imp":25.9215233653,"elo":1143.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":140.1670296597,"imp":25.7758595287,"elo":1151.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":282.7281586439,"imp":25.6822892415,"elo":1155.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":425.7671432604,"imp":25.64069868,"elo":1156.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":571.7337775834,"imp":25.6210568085,"elo":1157.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":14.8938989136,"imp":23.0914618447,"elo":1217.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":30.7592659452,"imp":20.4128575956,"elo":1327.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":78.4494180107,"imp":18.6827618171,"elo":1412.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":164.8792540854,"imp":18.1472441693,"elo":1440.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":421.7704708255,"imp":17.7763418786,"elo":1460.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":853.4354612721,"imp":17.602670488,"elo":1470.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1707.2754627676,"imp":17.5035549237,"elo":1474.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2551.8451865301,"imp":17.4426766716,"elo":1476.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3423.6100841542,"imp":17.3727074731,"elo":1475.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.0754426474,"imp":14.2549375502,"elo":1588.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":20.0220773915,"imp":14.3341707223,"elo":1586.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":54.3500923902,"imp":14.1222867847,"elo":1594.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":108.9572713635,"imp":13.9946623549,"elo":1606.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":283.232361067,"imp":13.8467716076,"elo":1608.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":566.4138152119,"imp":13.7600144768,"elo":1607.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1137.2039338041,"imp":13.6902717027,"elo":1606.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1710.4116369596,"imp":13.6571865748,"elo":1605.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2289.3887157667,"imp":13.6113146554,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.2671754693,"imp":21.4716322134,"elo":1329.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.4453139381,"imp":21.5207730434,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":75.9360943292,"imp":20.7985974681,"elo":1323.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":142.0530215894,"imp":20.0364039149,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":257.4275777592,"imp":19.7052417787,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":641.1662937545,"imp":19.0509958498,"elo":1395.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1303.1613548368,"imp":18.7312682734,"elo":1407.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2614.4217732811,"imp":18.5325459449,"elo":1415.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3931.8869009419,"imp":18.4441839765,"elo":1418.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":5271.4650532739,"imp":18.4098998781,"elo":1420.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":4.3495748865,"imp":18.1152207127,"elo":1443.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.5654426559,"imp":5.5031661361,"elo":1724.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":2.5922354756,"imp":30.5727640363,"elo":1056.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":3.2221111875,"imp":13.7524058745,"elo":1581.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":8.785094234,"imp":21.0567566542,"elo":1311.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":20.4659364224,"imp":20.6380333095,"elo":1334.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":57.4905112627,"imp":20.0497924632,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":119.8455247692,"imp":19.4781798721,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":304.1351923333,"imp":19.014184919,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":611.19598978,"imp":18.7279721753,"elo":1434.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1222.0069753852,"imp":18.5407792293,"elo":1443.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1840.5789024835,"imp":18.4515092957,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2469.4800149006,"imp":18.3960441345,"elo":1452.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":6.2559884135,"imp":14.406138207,"elo":1600.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":7.1656661471,"imp":13.2287844135,"elo":1635.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":502.344886049,"imp":31.642418261,"elo":1057.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":797.2338293006,"imp":31.4834790537,"elo":1063.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1342.9528314957,"imp":31.1491368256,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2054.553219683,"imp":30.3339010605,"elo":1086.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4867.9697698991,"imp":28.8799296609,"elo":1109.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":9985.461750605,"imp":27.494366485,"elo":1129.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":20581.7115313909,"imp":26.7940149153,"elo":1144.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":30569.2328830115,"imp":26.4735913113,"elo":1152.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":40588.1540174129,"imp":26.24847875,"elo":1159.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":2.2399715166,"imp":20.5336735934,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":11.2922428764,"imp":27.6556399469,"elo":1072.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":26.0516500508,"imp":25.3146370647,"elo":1137.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":70.9730313123,"imp":23.5009486978,"elo":1191.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":152.8912289988,"imp":22.7316925306,"elo":1224.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":376.394024437,"imp":22.1476320915,"elo":1249.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":738.844201808,"imp":21.8094200626,"elo":1264.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1456.0991336408,"imp":21.6361358313,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2205.8455340006,"imp":21.5435737799,"elo":1279.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2977.9095379662,"imp":21.4612241634,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.7957309689,"imp":23.6659363692,"elo":1177.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":7.086534818,"imp":21.8528389792,"elo":1252.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":19.7630506746,"imp":21.4258809272,"elo":1276.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":38.4377569942,"imp":21.1743433804,"elo":1292.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":99.4892942614,"imp":20.9494339085,"elo":1304.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":198.7048872189,"imp":20.8547752294,"elo":1309.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":397.0783434511,"imp":20.7694161244,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":594.6588102741,"imp":20.6930188709,"elo":1319.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":798.0801278695,"imp":20.6219323537,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":463.0485235354,"imp":26.242691752,"elo":1117.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1277.7801478637,"imp":24.0115342588,"elo":1175.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":3759.8173217043,"imp":21.7234740162,"elo":1267.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":8270.172943981,"imp":20.8516247953,"elo":1309.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":20344.7602093409,"imp":19.9499329283,"elo":1352.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":21290.0189805033,"imp":19.9282736659,"elo":1355.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":4.9505740724,"imp":29.9195022402,"elo":1035.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":10.1548540276,"imp":26.9804570653,"elo":1095.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":26.178393176,"imp":23.1816667151,"elo":1222.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":52.4324179391,"imp":21.6310405152,"elo":1296.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":131.4097749593,"imp":20.9900942983,"elo":1326.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":262.5999052849,"imp":20.813535843,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":524.6656622172,"imp":20.7795506743,"elo":1331.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":786.5490444842,"imp":20.7901140006,"elo":1329.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1053.935912149,"imp":20.7914435825,"elo":1328.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3591113273,"x_train":331.0009616013,"imp":19.7104018173,"elo":1409.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404555565,"x_train":7.1761452725,"imp":20.6917599107,"elo":1330.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0841906456,"x_train":13.6550050397,"imp":20.097358298,"elo":1353.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.158469367,"x_train":34.7005566859,"imp":19.8311243875,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2195178645,"x_train":69.9246996151,"imp":19.7555111737,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2815893316,"x_train":185.5351574323,"imp":19.7099238429,"elo":1369.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3179145648,"x_train":379.2610359226,"imp":19.731934615,"elo":1367.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806002107,"x_train":771.7715864818,"imp":19.7661508016,"elo":1364.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3873615123,"x_train":1152.9335256343,"imp":19.8007946703,"elo":1361.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"x_train":1542.9492660346,"imp":19.8150269011,"elo":1359.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.044213771,"x_train":4.9608581751,"imp":23.7100097,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0842869021,"x_train":10.8616521277,"imp":23.3685843207,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1733534673,"x_train":27.0121608835,"imp":23.1959832746,"elo":1248.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2454079283,"x_train":54.6281407695,"imp":23.0298512961,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3360644036,"x_train":140.6370736031,"imp":22.6970659011,"elo":1265.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3991994324,"x_train":283.4636990992,"imp":22.5418724954,"elo":1271.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4584564684,"x_train":569.3917360284,"imp":22.3660582501,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5006577968,"x_train":853.7391955308,"imp":22.2920716358,"elo":1280.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5312563547,"x_train":1142.8199158523,"imp":22.2483144507,"elo":1282.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0222777214,"x_train":7.0915913087,"imp":23.0968513065,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.051273724,"x_train":16.5198837522,"imp":22.8508490847,"elo":1247.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0788065997,"x_train":59.6568017021,"imp":22.5002056933,"elo":1263.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.097778986,"x_train":123.907431834,"imp":22.3818935455,"elo":1269.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1309347475,"x_train":318.4600106407,"imp":22.2211672007,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1473544657,"x_train":623.1113495163,"imp":22.0705940025,"elo":1283.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1683065578,"x_train":1256.5403166692,"imp":21.9552776509,"elo":1288.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1758856014,"x_train":1885.0271107114,"imp":21.9168997328,"elo":1291.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.181036219,"x_train":2499.3997394449,"imp":21.8910463161,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5520616377,"x_train":3.6063372166,"imp":30.8674670891,"elo":999.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6246844205,"x_train":7.3528355841,"imp":28.798766118,"elo":1044.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6183400546,"x_train":18.0710159165,"imp":25.9261230814,"elo":1117.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7242510207,"x_train":35.5062409712,"imp":25.0138733194,"elo":1162.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9536012796,"x_train":87.9242300599,"imp":24.452689118,"elo":1193.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1056205394,"x_train":175.4018189887,"imp":24.2951625133,"elo":1202.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2895955274,"x_train":350.0379937538,"imp":24.1864668973,"elo":1209.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3699672148,"x_train":524.8382968951,"imp":24.1167676549,"elo":1214.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4314907954,"x_train":703.0330785914,"imp":24.1067040677,"elo":1214.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6452165158,"x_train":5.0370258118,"imp":28.9894994822,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3091545347,"x_train":12.7722389468,"imp":27.4434440982,"elo":1068.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7086381558,"x_train":35.5543374641,"imp":25.1090696531,"elo":1149.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.7948756042,"x_train":67.9976554587,"imp":24.0378613931,"elo":1191.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.5085868791,"x_train":172.4339504664,"imp":23.156687283,"elo":1224.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.5711464723,"x_train":350.2703386124,"imp":22.7751451617,"elo":1242.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7445847733,"x_train":703.7735059836,"imp":22.618651403,"elo":1250.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.6616951336,"x_train":1059.2142650988,"imp":22.5409104157,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.1991520011,"x_train":1424.216740031,"imp":22.5380588141,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.140680637,"x_train":3.149413842,"imp":23.9933364168,"elo":1165.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.557564348,"x_train":7.1207733159,"imp":21.799911592,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8204403589,"x_train":17.9053372504,"imp":20.5967936552,"elo":1334.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0270516849,"x_train":35.9737069678,"imp":20.1137675506,"elo":1361.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3215357247,"x_train":91.1663943012,"imp":19.786371996,"elo":1379.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5580626458,"x_train":182.6006011031,"imp":19.6743028228,"elo":1386.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6848179893,"x_train":365.5882548593,"imp":19.5275124529,"elo":1394.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6963950347,"x_train":548.151079493,"imp":19.471737158,"elo":1396.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.8237866308,"x_train":735.3700351286,"imp":19.440619285,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":5.0441668928,"x_train":27.7685088788,"imp":16.4063984503,"elo":1475.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"x_train":15.5476029367,"imp":23.7452802097,"elo":1259.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8804757883,"x_train":52.4814325925,"imp":22.4237173732,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9515166073,"x_train":150.689630028,"imp":20.3212043231,"elo":1352.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9139871186,"x_train":293.311977236,"imp":19.956113518,"elo":1363.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.2991192179,"x_train":719.031159336,"imp":19.5895978361,"elo":1375.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.6324379245,"x_train":1498.353139385,"imp":19.8851630726,"elo":1360.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.2170906616,"x_train":3004.008741271,"imp":20.4709045447,"elo":1338.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.2545907476,"x_train":4473.3946349608,"imp":20.8865722454,"elo":1322.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7287422808,"x_train":6049.2787189098,"imp":21.1122915159,"elo":1311.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6112675715,"x_train":15.4475945518,"imp":26.0247643412,"elo":1162.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0479211725,"x_train":56.1148745356,"imp":37.2703231919,"elo":945.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4213204184,"x_train":121.64643757,"imp":35.195351398,"elo":1000.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9452456188,"x_train":327.8283737012,"imp":32.5599522417,"elo":1077.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0563537894,"x_train":391.0862000393,"imp":32.3630388368,"elo":1084.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0681202832,"x_train":0.4628071975,"imp":31.6097744515,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2374252662,"x_train":3.2094586922,"imp":28.8655853723,"elo":1044.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.441254012,"x_train":11.9623791751,"imp":27.3622202338,"elo":1078.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.648284888,"x_train":26.2110927853,"imp":26.5336916255,"elo":1107.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8046320893,"x_train":68.7900402557,"imp":25.9215233653,"elo":1143.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9831168585,"x_train":140.1670296597,"imp":25.7758595287,"elo":1151.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2033557443,"x_train":282.7281586439,"imp":25.6822892415,"elo":1155.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.305785724,"x_train":425.7671432604,"imp":25.64069868,"elo":1156.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.388919598,"x_train":571.7337775834,"imp":25.6210568085,"elo":1157.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.4321040652,"x_train":14.8938989136,"imp":23.0914618447,"elo":1217.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.5278018741,"x_train":30.7592659452,"imp":20.4128575956,"elo":1327.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.9302452502,"x_train":78.4494180107,"imp":18.6827618171,"elo":1412.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.8524000954,"x_train":164.8792540854,"imp":18.1472441693,"elo":1440.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.294458449,"x_train":421.7704708255,"imp":17.7763418786,"elo":1460.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4003571581,"x_train":853.4354612721,"imp":17.602670488,"elo":1470.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.0745491458,"x_train":1707.2754627676,"imp":17.5035549237,"elo":1474.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.8336064118,"x_train":2551.8451865301,"imp":17.4426766716,"elo":1476.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.0794572094,"x_train":3423.6100841542,"imp":17.3727074731,"elo":1475.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7616019805,"x_train":6.0754426474,"imp":14.2549375502,"elo":1588.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3608475094,"x_train":20.0220773915,"imp":14.3341707223,"elo":1586.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.9479323005,"x_train":54.3500923902,"imp":14.1222867847,"elo":1594.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1961228236,"x_train":108.9572713635,"imp":13.9946623549,"elo":1606.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7212619669,"x_train":283.232361067,"imp":13.8467716076,"elo":1608.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8314060794,"x_train":566.4138152119,"imp":13.7600144768,"elo":1607.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5661153905,"x_train":1137.2039338041,"imp":13.6902717027,"elo":1606.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0685248517,"x_train":1710.4116369596,"imp":13.6571865748,"elo":1605.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.5533679569,"x_train":2289.3887157667,"imp":13.6113146554,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6235508255,"x_train":13.2671754693,"imp":21.4716322134,"elo":1329.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"x_train":51.4453139381,"imp":21.5207730434,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.7846214148,"x_train":75.9360943292,"imp":20.7985974681,"elo":1323.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":68.4502119688,"x_train":142.0530215894,"imp":20.0364039149,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":88.2706830363,"x_train":257.4275777592,"imp":19.7052417787,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":120.5529066959,"x_train":641.1662937545,"imp":19.0509958498,"elo":1395.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":148.5135318639,"x_train":1303.1613548368,"imp":18.7312682734,"elo":1407.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":181.9231911908,"x_train":2614.4217732811,"imp":18.5325459449,"elo":1415.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":200.1306803402,"x_train":3931.8869009419,"imp":18.4441839765,"elo":1418.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":228.0648501265,"x_train":5271.4650532739,"imp":18.4098998781,"elo":1420.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"x_train":4.3495748865,"imp":18.1152207127,"elo":1443.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.3168064912,"x_train":38.5654426559,"imp":5.5031661361,"elo":1724.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5331077226,"x_train":2.5922354756,"imp":30.5727640363,"elo":1056.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"x_train":3.2221111875,"imp":13.7524058745,"elo":1581.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4108430406,"x_train":8.785094234,"imp":21.0567566542,"elo":1311.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.7817725977,"x_train":20.4659364224,"imp":20.6380333095,"elo":1334.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.405202684,"x_train":57.4905112627,"imp":20.0497924632,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.9090624077,"x_train":119.8455247692,"imp":19.4781798721,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.5663201327,"x_train":304.1351923333,"imp":19.014184919,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":2.956697172,"x_train":611.19598978,"imp":18.7279721753,"elo":1434.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.3235732502,"x_train":1222.0069753852,"imp":18.5407792293,"elo":1443.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":3.405584308,"x_train":1840.5789024835,"imp":18.4515092957,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":3.4143908721,"x_train":2469.4800149006,"imp":18.3960441345,"elo":1452.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"x_train":6.2559884135,"imp":14.406138207,"elo":1600.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4074238608,"x_train":7.1656661471,"imp":13.2287844135,"elo":1635.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.3276990403,"x_train":502.344886049,"imp":31.642418261,"elo":1057.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.2461770633,"x_train":797.2338293006,"imp":31.4834790537,"elo":1063.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.851933236,"x_train":1342.9528314957,"imp":31.1491368256,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.6518831225,"x_train":2054.553219683,"imp":30.3339010605,"elo":1086.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0496115636,"x_train":4867.9697698991,"imp":28.8799296609,"elo":1109.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.7051263351,"x_train":9985.461750605,"imp":27.494366485,"elo":1129.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.2001866591,"x_train":20581.7115313909,"imp":26.7940149153,"elo":1144.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.3628807226,"x_train":30569.2328830115,"imp":26.4735913113,"elo":1152.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.2209155837,"x_train":40588.1540174129,"imp":26.24847875,"elo":1159.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0545537099,"x_train":2.2399715166,"imp":20.5336735934,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.203890602,"x_train":11.2922428764,"imp":27.6556399469,"elo":1072.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3965671437,"x_train":26.0516500508,"imp":25.3146370647,"elo":1137.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7770360585,"x_train":70.9730313123,"imp":23.5009486978,"elo":1191.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0521362816,"x_train":152.8912289988,"imp":22.7316925306,"elo":1224.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.364215168,"x_train":376.394024437,"imp":22.1476320915,"elo":1249.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6688707494,"x_train":738.844201808,"imp":21.8094200626,"elo":1264.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0488352505,"x_train":1456.0991336408,"imp":21.6361358313,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.2724182677,"x_train":2205.8455340006,"imp":21.5435737799,"elo":1279.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4133616403,"x_train":2977.9095379662,"imp":21.4612241634,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2382075678,"x_train":2.7957309689,"imp":23.6659363692,"elo":1177.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6139127796,"x_train":7.086534818,"imp":21.8528389792,"elo":1252.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2441928973,"x_train":19.7630506746,"imp":21.4258809272,"elo":1276.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.450807434,"x_train":38.4377569942,"imp":21.1743433804,"elo":1292.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.643707399,"x_train":99.4892942614,"imp":20.9494339085,"elo":1304.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8330479874,"x_train":198.7048872189,"imp":20.8547752294,"elo":1309.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0008744691,"x_train":397.0783434511,"imp":20.7694161244,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0954188348,"x_train":594.6588102741,"imp":20.6930188709,"elo":1319.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.1816683976,"x_train":798.0801278695,"imp":20.6219323537,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":147.9369405767,"x_train":463.0485235354,"imp":26.242691752,"elo":1117.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":301.0343409385,"x_train":1277.7801478637,"imp":24.0115342588,"elo":1175.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":433.1102919089,"x_train":3759.8173217043,"imp":21.7234740162,"elo":1267.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":548.6068914868,"x_train":8270.172943981,"imp":20.8516247953,"elo":1309.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":636.7319509309,"x_train":20344.7602093409,"imp":19.9499329283,"elo":1352.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":654.0373562851,"x_train":21290.0189805033,"imp":19.9282736659,"elo":1355.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.0121457497,"x_train":4.9505740724,"imp":29.9195022402,"elo":1035.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.0736404567,"x_train":10.1548540276,"imp":26.9804570653,"elo":1095.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.2590656982,"x_train":26.178393176,"imp":23.1816667151,"elo":1222.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.6225412385,"x_train":52.4324179391,"imp":21.6310405152,"elo":1296.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8253774488,"x_train":131.4097749593,"imp":20.9900942983,"elo":1326.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2506435468,"x_train":262.5999052849,"imp":20.813535843,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.5518696321,"x_train":524.6656622172,"imp":20.7795506743,"elo":1331.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.795048319,"x_train":786.5490444842,"imp":20.7901140006,"elo":1329.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.0515616235,"x_train":1053.935912149,"imp":20.7914435825,"elo":1328.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_small/winrate_explorer.html b/data/imputation_no/splits_all/tasks_classification/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..46a980f2e53bb4511a08ccbc9068586048a5ed37
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_classification/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned)", "TabDPT-Turbo (default)", "LimiX (default)", "RealMLP (tuned + ensembled)", "TabM (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabDPT (tuned + ensembled)", "BetaTabPFN (default)", "TabDPT (tuned)", "RealMLP (tuned)", "TabM (tuned)", "TabSwift (default)", "LightGBM (tuned + ensembled)", "LightGBM (tuned)", "TabDPT (default)", "CatBoost (tuned + ensembled)", "ModernNCA (tuned)", "xRFM (tuned + ensembled)", "SAP-RPT-OSS (default)", "iLTM (tuned + ensembled)", "CatBoost (tuned)", "xRFM (tuned)", "XGBoost (tuned + ensembled)", "CatBoost (default)", "EBM (tuned + ensembled)", "TabM (default)", "ModernNCA (default)", "XGBoost (tuned)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "EBM (tuned)", "EBM (default)", "iLTM (tuned)", "RealMLP (default)", "ChimeraBoost (tuned + ensembled)", "FastaiMLP (tuned)", "ExtraTrees (tuned + ensembled)", "TorchMLP (tuned)", "ChimeraBoost (tuned)", "OrionMSP (default)", "ExtraTrees (tuned)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "ChimeraBoost (default)", "RandomForest (tuned + ensembled)", "XGBoost (default)", "LightGBM (default)", "TabFlex (default)", "RandomForest (tuned)", "Linear (tuned + ensembled)", "TorchMLP (default)", "iLTM (default)", "TabSTAR (default)", "PerpetualBooster (tuned + ensembled)", "Linear (tuned)", "xRFM (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "ExtraTrees (default)", "RandomForest (default)", "Linear (default)", "PerpetualBooster (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.795634925365448, 0.7876983880996704, 0.8194444179534912, 0.7678571343421936, 0.7708333134651184, 0.8035714030265808, 0.7757936716079712, 0.8313491940498352, 0.879960298538208, 0.8571428656578064, 0.8650793433189392, 0.9047619104385376, 0.886904776096344, 0.858134925365448, 0.8789682388305664, 0.851190447807312, 0.8948412537574768, 0.8690476417541504, 0.9007936716079712, 0.898809552192688, 0.8908730149269104, 0.9186508059501648, 0.9027777910232544, 0.9126983880996704, 0.8373016119003296, 0.920634925365448, 0.9246031641960144, 0.8928571343421936, 0.932539701461792, 0.9265872836112976, 0.8948412537574768, 0.9007936716079712, 0.9047619104385376, 0.932539701461792, 0.89682537317276, 0.9444444179534912, 0.9444444179534912, 0.9067460298538208, 0.9007936716079712, 0.9345238208770752, 0.9186508059501648, 0.9285714030265808, 0.9226190447807312, 0.9384920597076416, 0.9523809552192688, 0.9384920597076416, 0.9305555820465088, 0.9285714030265808, 0.8253968358039856, 0.8253968358039856, 0.9464285969734192, 0.9563491940498352, 0.9583333134651184, 0.954365074634552, 0.9265872836112976, 0.954365074634552, 0.9365079402923584, 0.9603174328804016, 0.9305555820465088, 0.8710317611694336, 0.9821428656578064, 0.9444444179534912, 0.9642857313156128, 0.97817462682724, 0.966269850730896, 0.9642857313156128, 0.988095223903656, 0.9801587462425232, 0.9603174328804016, 0.9801587462425232, 0.9563491940498352, 0.9940476417541504], [0.204365074634552, NaN, 0.5555555820465088, 0.5257936716079712, 0.601190447807312, 0.5753968358039856, 0.6111111044883728, 0.5972222089767456, 0.6964285969734192, 0.7182539701461792, 0.7242063283920288, 0.7579365372657776, 0.7420634627342224, 0.783730149269104, 0.7371031641960144, 0.8055555820465088, 0.7638888955116272, 0.795634925365448, 0.8432539701461792, 0.8154761791229248, 0.8333333134651184, 0.8194444179534912, 0.8492063283920288, 0.8492063283920288, 0.85317462682724, 0.7400793433189392, 0.8769841194152832, 0.8591269850730896, 0.8472222089767456, 0.8501983880996704, 0.875, 0.8392857313156128, 0.8194444179534912, 0.8640872836112976, 0.863095223903656, 0.8303571343421936, 0.89682537317276, 0.886904776096344, 0.85317462682724, 0.8660714030265808, 0.91567462682724, 0.9027777910232544, 0.908730149269104, 0.8839285969734192, 0.9246031641960144, 0.9166666865348816, 0.908730149269104, 0.9126983880996704, 0.9226190447807312, 0.8125, 0.817460298538208, 0.908730149269104, 0.920634925365448, 0.9583333134651184, 0.9404761791229248, 0.9166666865348816, 0.942460298538208, 0.9047619104385376, 0.9503968358039856, 0.9384920597076416, 0.85317462682724, 0.9365079402923584, 0.9126983880996704, 0.9682539701461792, 0.9503968358039856, 0.9583333134651184, 0.942460298538208, 0.9742063283920288, 0.97817462682724, 0.9484127163887024, 0.9642857313156128, 0.9523809552192688, 0.9960317611694336], [0.2123015820980072, 0.4444444477558136, NaN, 0.460317462682724, 0.5416666865348816, 0.533730149269104, 0.6091269850730896, 0.5535714030265808, 0.5972222089767456, 0.7202380895614624, 0.7628968358039856, 0.7916666865348816, 0.7916666865348816, 0.7658730149269104, 0.738095223903656, 0.7876983880996704, 0.795634925365448, 0.7966269850730896, 0.8035714030265808, 0.8373016119003296, 0.8333333134651184, 0.761904776096344, 0.85317462682724, 0.8551587462425232, 0.8611111044883728, 0.7202380895614624, 0.8769841194152832, 0.8501983880996704, 0.8303571343421936, 0.8611111044883728, 0.8660714030265808, 0.8392857313156128, 0.8373016119003296, 0.8472222089767456, 0.8591269850730896, 0.8373016119003296, 0.886904776096344, 0.908730149269104, 0.851190447807312, 0.8571428656578064, 0.908730149269104, 0.8670634627342224, 0.9126983880996704, 0.8650793433189392, 0.9047619104385376, 0.920634925365448, 0.9146825671195984, 0.91567462682724, 0.8908730149269104, 0.7777777910232544, 0.7797619104385376, 0.9166666865348816, 0.9226190447807312, 0.9464285969734192, 0.9464285969734192, 0.89682537317276, 0.9265872836112976, 0.8829365372657776, 0.9503968358039856, 0.932539701461792, 0.8253968358039856, 0.9444444179534912, 0.908730149269104, 0.9464285969734192, 0.949404776096344, 0.9444444179534912, 0.932539701461792, 0.954365074634552, 0.9563491940498352, 0.9355158805847168, 0.9583333134651184, 0.9464285969734192, 0.9940476417541504], [0.1805555522441864, 0.4742063581943512, 0.5396825671195984, NaN, 0.538690447807312, 0.5297619104385376, 0.60317462682724, 0.5436508059501648, 0.682539701461792, 0.7172619104385376, 0.7242063283920288, 0.761904776096344, 0.77182537317276, 0.778769850730896, 0.7400793433189392, 0.7698412537574768, 0.7460317611694336, 0.7757936716079712, 0.8253968358039856, 0.8055555820465088, 0.8273809552192688, 0.795634925365448, 0.817460298538208, 0.8353174328804016, 0.8313491940498352, 0.6984127163887024, 0.8670634627342224, 0.8273809552192688, 0.8144841194152832, 0.8353174328804016, 0.8591269850730896, 0.8333333134651184, 0.8115079402923584, 0.8640872836112976, 0.8432539701461792, 0.8353174328804016, 0.8769841194152832, 0.863095223903656, 0.8392857313156128, 0.8591269850730896, 0.8849206566810608, 0.8809523582458496, 0.8928571343421936, 0.8789682388305664, 0.8720238208770752, 0.908730149269104, 0.9067460298538208, 0.908730149269104, 0.8769841194152832, 0.8035714030265808, 0.8015872836112976, 0.9027777910232544, 0.9146825671195984, 0.9265872836112976, 0.9365079402923584, 0.8948412537574768, 0.9126983880996704, 0.9186508059501648, 0.9345238208770752, 0.9166666865348816, 0.8551587462425232, 0.942460298538208, 0.9246031641960144, 0.9404761791229248, 0.9484127163887024, 0.9503968358039856, 0.942460298538208, 0.9503968358039856, 0.9603174328804016, 0.9513888955116272, 0.9722222089767456, 0.9464285969734192, 0.9960317611694336], [0.2321428507566452, 0.3988095223903656, 0.4583333432674408, 0.4613095223903656, NaN, 0.4404761791229248, 0.5079365372657776, 0.48908731341362, 0.5714285969734192, 0.6884920597076416, 0.7103174328804016, 0.7599206566810608, 0.7539682388305664, 0.692460298538208, 0.7589285969734192, 0.699404776096344, 0.7698412537574768, 0.7996031641960144, 0.7896825671195984, 0.807539701461792, 0.829365074634552, 0.7123016119003296, 0.8472222089767456, 0.8055555820465088, 0.8253968358039856, 0.761904776096344, 0.8194444179534912, 0.8472222089767456, 0.8333333134651184, 0.8551587462425232, 0.8829365372657776, 0.8591269850730896, 0.85317462682724, 0.8392857313156128, 0.875, 0.8670634627342224, 0.886904776096344, 0.8353174328804016, 0.8789682388305664, 0.8611111044883728, 0.903769850730896, 0.898809552192688, 0.9027777910232544, 0.8908730149269104, 0.863095223903656, 0.9166666865348816, 0.9047619104385376, 0.920634925365448, 0.875, 0.8035714030265808, 0.8045634627342224, 0.9126983880996704, 0.886904776096344, 0.9305555820465088, 0.9444444179534912, 0.913690447807312, 0.9345238208770752, 0.9007936716079712, 0.9623016119003296, 0.932539701461792, 0.8373016119003296, 0.9404761791229248, 0.908730149269104, 0.9484127163887024, 0.9464285969734192, 0.9464285969734192, 0.9027777910232544, 0.9523809552192688, 0.954365074634552, 0.9523809552192688, 0.966269850730896, 0.9365079402923584, 0.9940476417541504], [0.2291666716337204, 0.4246031641960144, 0.466269850730896, 0.4702380895614624, 0.5595238208770752, NaN, 0.5515872836112976, 0.5753968358039856, 0.5654761791229248, 0.7083333134651184, 0.7361111044883728, 0.7559523582458496, 0.7757936716079712, 0.6845238208770752, 0.7698412537574768, 0.6984127163887024, 0.7678571343421936, 0.783730149269104, 0.7579365372657776, 0.8134920597076416, 0.8303571343421936, 0.6686508059501648, 0.851190447807312, 0.8472222089767456, 0.8333333134651184, 0.7529761791229248, 0.8382936716079712, 0.8452380895614624, 0.8214285969734192, 0.8591269850730896, 0.863095223903656, 0.8353174328804016, 0.8253968358039856, 0.8333333134651184, 0.8720238208770752, 0.83432537317276, 0.8789682388305664, 0.8789682388305664, 0.8482142686843872, 0.841269850730896, 0.89682537317276, 0.863095223903656, 0.891865074634552, 0.875, 0.8591269850730896, 0.9107142686843872, 0.898809552192688, 0.8908730149269104, 0.8611111044883728, 0.7777777910232544, 0.783730149269104, 0.898809552192688, 0.9027777910232544, 0.9404761791229248, 0.9365079402923584, 0.8908730149269104, 0.9146825671195984, 0.898809552192688, 0.9384920597076416, 0.9226190447807312, 0.8154761791229248, 0.942460298538208, 0.9146825671195984, 0.9404761791229248, 0.942460298538208, 0.9404761791229248, 0.9166666865348816, 0.9503968358039856, 0.9484127163887024, 0.925595223903656, 0.9583333134651184, 0.9265872836112976, 0.9920634627342224], [0.1964285671710968, 0.3888888955116272, 0.3908730149269104, 0.3968254029750824, 0.4920634925365448, 0.44841268658638, NaN, 0.4791666567325592, 0.5873016119003296, 0.6626983880996704, 0.7182539701461792, 0.7420634627342224, 0.7440476417541504, 0.6746031641960144, 0.72817462682724, 0.6865079402923584, 0.7559523582458496, 0.7797619104385376, 0.761904776096344, 0.7996031641960144, 0.8353174328804016, 0.7003968358039856, 0.829365074634552, 0.829365074634552, 0.8095238208770752, 0.738095223903656, 0.8789682388305664, 0.8214285969734192, 0.8015872836112976, 0.8591269850730896, 0.875, 0.8095238208770752, 0.8184523582458496, 0.8452380895614624, 0.8650793433189392, 0.8253968358039856, 0.8769841194152832, 0.908730149269104, 0.8382936716079712, 0.8313491940498352, 0.9047619104385376, 0.8650793433189392, 0.9107142686843872, 0.8670634627342224, 0.8829365372657776, 0.908730149269104, 0.9126983880996704, 0.8611111044883728, 0.8720238208770752, 0.7757936716079712, 0.7857142686843872, 0.9186508059501648, 0.9007936716079712, 0.954365074634552, 0.9444444179534912, 0.8948412537574768, 0.9226190447807312, 0.8849206566810608, 0.9404761791229248, 0.9226190447807312, 0.8154761791229248, 0.932539701461792, 0.908730149269104, 0.9404761791229248, 0.9285714030265808, 0.9365079402923584, 0.932539701461792, 0.9523809552192688, 0.966269850730896, 0.9365079402923584, 0.9563491940498352, 0.942460298538208, 0.9920634627342224], [0.22420634329319, 0.4027777910232544, 0.4464285671710968, 0.4563491940498352, 0.5109127163887024, 0.4246031641960144, 0.5208333134651184, NaN, 0.5694444179534912, 0.6726190447807312, 0.7103174328804016, 0.7539682388305664, 0.7638888955116272, 0.6805555820465088, 0.7559523582458496, 0.70932537317276, 0.7638888955116272, 0.788690447807312, 0.7490079402923584, 0.7996031641960144, 0.829365074634552, 0.6865079402923584, 0.8373016119003296, 0.8095238208770752, 0.8144841194152832, 0.7480158805847168, 0.8194444179534912, 0.8333333134651184, 0.8095238208770752, 0.8452380895614624, 0.851190447807312, 0.8472222089767456, 0.829365074634552, 0.8214285969734192, 0.85317462682724, 0.8214285969734192, 0.8789682388305664, 0.870039701461792, 0.851190447807312, 0.8353174328804016, 0.8908730149269104, 0.863095223903656, 0.8819444179534912, 0.8650793433189392, 0.8621031641960144, 0.9107142686843872, 0.8928571343421936, 0.8849206566810608, 0.8690476417541504, 0.7698412537574768, 0.7777777910232544, 0.89682537317276, 0.8928571343421936, 0.9345238208770752, 0.9365079402923584, 0.8809523582458496, 0.908730149269104, 0.8849206566810608, 0.9464285969734192, 0.9186508059501648, 0.8154761791229248, 0.925595223903656, 0.9007936716079712, 0.942460298538208, 0.9365079402923584, 0.932539701461792, 0.9166666865348816, 0.9444444179534912, 0.9464285969734192, 0.9335317611694336, 0.9583333134651184, 0.932539701461792, 0.9920634627342224], [0.1686507910490036, 0.3035714328289032, 0.4027777910232544, 0.3174603283405304, 0.4285714328289032, 0.4345238208770752, 0.4126984179019928, 0.4305555522441864, NaN, 0.5615079402923584, 0.5833333134651184, 0.5803571343421936, 0.5892857313156128, 0.6309523582458496, 0.567460298538208, 0.6527777910232544, 0.6378968358039856, 0.6210317611694336, 0.7291666865348816, 0.658730149269104, 0.6607142686843872, 0.788690447807312, 0.6984127163887024, 0.6904761791229248, 0.6865079402923584, 0.6309523582458496, 0.7480158805847168, 0.6884920597076416, 0.7142857313156128, 0.7142857313156128, 0.7222222089767456, 0.7123016119003296, 0.6865079402923584, 0.7658730149269104, 0.7222222089767456, 0.7698412537574768, 0.7400793433189392, 0.7321428656578064, 0.745039701461792, 0.7559523582458496, 0.8224206566810608, 0.7916666865348816, 0.8015872836112976, 0.8273809552192688, 0.836309552192688, 0.8015872836112976, 0.817460298538208, 0.8611111044883728, 0.8591269850730896, 0.7301587462425232, 0.738095223903656, 0.8313491940498352, 0.875, 0.8670634627342224, 0.8730158805847168, 0.908730149269104, 0.8789682388305664, 0.841269850730896, 0.8849206566810608, 0.863095223903656, 0.7996031641960144, 0.8849206566810608, 0.851190447807312, 0.8888888955116272, 0.9047619104385376, 0.9226190447807312, 0.9186508059501648, 0.9533730149269104, 0.9523809552192688, 0.9305555820465088, 0.954365074634552, 0.942460298538208, 0.9940476417541504], [0.1200396791100502, 0.2817460298538208, 0.2797619104385376, 0.2827380895614624, 0.3115079402923584, 0.2916666567325592, 0.3373015820980072, 0.3273809552192688, 0.4384920597076416, NaN, 0.5972222089767456, 0.6269841194152832, 0.6329365372657776, 0.538690447807312, 0.625, 0.567460298538208, 0.641865074634552, 0.6398809552192688, 0.6091269850730896, 0.716269850730896, 0.726190447807312, 0.5853174328804016, 0.738095223903656, 0.738095223903656, 0.7579365372657776, 0.6686508059501648, 0.7966269850730896, 0.7400793433189392, 0.745039701461792, 0.77182537317276, 0.7777777910232544, 0.7678571343421936, 0.7142857313156128, 0.7698412537574768, 0.7896825671195984, 0.7172619104385376, 0.7559523582458496, 0.7777777910232544, 0.795634925365448, 0.7976190447807312, 0.8273809552192688, 0.8095238208770752, 0.8472222089767456, 0.79067462682724, 0.841269850730896, 0.8134920597076416, 0.8551587462425232, 0.8204365372657776, 0.8392857313156128, 0.7648809552192688, 0.7579365372657776, 0.8710317611694336, 0.89682537317276, 0.898809552192688, 0.9126983880996704, 0.8115079402923584, 0.9107142686843872, 0.8611111044883728, 0.8888888955116272, 0.8928571343421936, 0.8105158805847168, 0.9047619104385376, 0.8829365372657776, 0.9345238208770752, 0.9146825671195984, 0.920634925365448, 0.8948412537574768, 0.9464285969734192, 0.954365074634552, 0.8908730149269104, 0.9365079402923584, 0.9146825671195984, 0.9900793433189392], [0.1428571492433548, 0.2757936418056488, 0.2371031790971756, 0.2757936418056488, 0.289682537317276, 0.2638888955116272, 0.2817460298538208, 0.289682537317276, 0.4166666567325592, 0.4027777910232544, NaN, 0.5773809552192688, 0.5327380895614624, 0.4871031641960144, 0.5932539701461792, 0.4950396716594696, 0.663690447807312, 0.6349206566810608, 0.5297619104385376, 0.7003968358039856, 0.7311508059501648, 0.5476190447807312, 0.7519841194152832, 0.6626983880996704, 0.704365074634552, 0.579365074634552, 0.7400793433189392, 0.7559523582458496, 0.682539701461792, 0.817460298538208, 0.77182537317276, 0.77182537317276, 0.7559523582458496, 0.7103174328804016, 0.8273809552192688, 0.7569444179534912, 0.7321428656578064, 0.7123016119003296, 0.7797619104385376, 0.778769850730896, 0.8115079402923584, 0.846230149269104, 0.8214285969734192, 0.7807539701461792, 0.7976190447807312, 0.7857142686843872, 0.8303571343421936, 0.7440476417541504, 0.8125, 0.726190447807312, 0.7341269850730896, 0.8313491940498352, 0.886904776096344, 0.9146825671195984, 0.9007936716079712, 0.795634925365448, 0.9186508059501648, 0.8472222089767456, 0.9007936716079712, 0.8908730149269104, 0.7797619104385376, 0.9226190447807312, 0.8492063283920288, 0.9146825671195984, 0.9285714030265808, 0.9285714030265808, 0.8948412537574768, 0.9464285969734192, 0.9583333134651184, 0.8710317611694336, 0.9603174328804016, 0.920634925365448, 0.9940476417541504], [0.1349206417798996, 0.2420634925365448, 0.2083333283662796, 0.2380952388048172, 0.2400793582201004, 0.244047611951828, 0.2579365074634552, 0.2460317462682724, 0.4196428656578064, 0.3730158805847168, 0.4226190447807312, NaN, 0.4900793731212616, 0.483134925365448, 0.591269850730896, 0.4900793731212616, 0.5515872836112976, 0.6547619104385376, 0.5396825671195984, 0.6408730149269104, 0.663690447807312, 0.545634925365448, 0.6507936716079712, 0.6220238208770752, 0.6071428656578064, 0.5892857313156128, 0.6517857313156128, 0.6765872836112976, 0.6111111044883728, 0.7460317611694336, 0.75, 0.761904776096344, 0.8224206566810608, 0.6845238208770752, 0.7678571343421936, 0.6944444179534912, 0.7182539701461792, 0.7003968358039856, 0.7757936716079712, 0.7688491940498352, 0.75, 0.788690447807312, 0.8095238208770752, 0.75, 0.7301587462425232, 0.7986111044883728, 0.8353174328804016, 0.716269850730896, 0.726190447807312, 0.726190447807312, 0.7341269850730896, 0.8670634627342224, 0.7896825671195984, 0.8809523582458496, 0.8839285969734192, 0.7976190447807312, 0.841269850730896, 0.8551587462425232, 0.8928571343421936, 0.8809523582458496, 0.783730149269104, 0.8928571343421936, 0.8720238208770752, 0.8591269850730896, 0.9126983880996704, 0.9007936716079712, 0.8730158805847168, 0.9067460298538208, 0.9007936716079712, 0.898809552192688, 0.9563491940498352, 0.9107142686843872, 0.9841269850730896], [0.095238097012043, 0.2579365074634552, 0.2083333283662796, 0.2281745970249176, 0.2460317462682724, 0.22420634329319, 0.255952388048172, 0.2361111044883728, 0.4107142984867096, 0.3670634925365448, 0.4672619104385376, 0.5099206566810608, NaN, 0.4761904776096344, 0.5515872836112976, 0.4781745970249176, 0.5595238208770752, 0.5694444179534912, 0.5228174328804016, 0.670634925365448, 0.6904761791229248, 0.5476190447807312, 0.6785714030265808, 0.6111111044883728, 0.6438491940498352, 0.5297619104385376, 0.6805555820465088, 0.7063491940498352, 0.6726190447807312, 0.726190447807312, 0.7440476417541504, 0.7063491940498352, 0.6944444179534912, 0.6904761791229248, 0.7123016119003296, 0.7242063283920288, 0.7519841194152832, 0.6785714030265808, 0.726190447807312, 0.7638888955116272, 0.77182537317276, 0.773809552192688, 0.7847222089767456, 0.7817460298538208, 0.7083333134651184, 0.8055555820465088, 0.795634925365448, 0.70932537317276, 0.7400793433189392, 0.7242063283920288, 0.7321428656578064, 0.8194444179534912, 0.807539701461792, 0.8769841194152832, 0.9007936716079712, 0.8095238208770752, 0.8482142686843872, 0.8422619104385376, 0.8769841194152832, 0.8650793433189392, 0.7817460298538208, 0.9146825671195984, 0.870039701461792, 0.8769841194152832, 0.920634925365448, 0.9166666865348816, 0.875, 0.9246031641960144, 0.9146825671195984, 0.9047619104385376, 0.9384920597076416, 0.9146825671195984, 0.9960317611694336], [0.1130952388048172, 0.2162698358297348, 0.2341269850730896, 0.2212301641702652, 0.3075396716594696, 0.3154761791229248, 0.3253968358039856, 0.3194444477558136, 0.369047611951828, 0.4613095223903656, 0.5128968358039856, 0.516865074634552, 0.523809552192688, NaN, 0.52182537317276, 0.6071428656578064, 0.5942460298538208, 0.5545634627342224, 0.5982142686843872, 0.5952380895614624, 0.6269841194152832, 0.6309523582458496, 0.625, 0.64682537317276, 0.6398809552192688, 0.5853174328804016, 0.6944444179534912, 0.6329365372657776, 0.653769850730896, 0.6577380895614624, 0.6805555820465088, 0.66567462682724, 0.641865074634552, 0.7013888955116272, 0.682539701461792, 0.6964285969734192, 0.6974206566810608, 0.7301587462425232, 0.6954365372657776, 0.692460298538208, 0.7797619104385376, 0.72817462682724, 0.7559523582458496, 0.7519841194152832, 0.783730149269104, 0.7599206566810608, 0.7698412537574768, 0.8015872836112976, 0.7996031641960144, 0.6984127163887024, 0.7003968358039856, 0.7936508059501648, 0.8571428656578064, 0.8432539701461792, 0.851190447807312, 0.829365074634552, 0.8710317611694336, 0.7857142686843872, 0.8690476417541504, 0.8571428656578064, 0.77182537317276, 0.8670634627342224, 0.807539701461792, 0.8978174328804016, 0.8819444179534912, 0.9007936716079712, 0.9017857313156128, 0.9384920597076416, 0.9305555820465088, 0.8650793433189392, 0.9305555820465088, 0.9126983880996704, 0.9940476417541504], [0.141865074634552, 0.2628968358039856, 0.261904776096344, 0.2599206268787384, 0.2410714328289032, 0.230158731341362, 0.2718254029750824, 0.244047611951828, 0.4325396716594696, 0.375, 0.4067460298538208, 0.408730149269104, 0.44841268658638, 0.4781745970249176, NaN, 0.5, 0.4583333432674408, 0.4900793731212616, 0.5277777910232544, 0.601190447807312, 0.5992063283920288, 0.5833333134651184, 0.5873016119003296, 0.596230149269104, 0.5952380895614624, 0.5376983880996704, 0.6051587462425232, 0.6220238208770752, 0.6051587462425232, 0.6607142686843872, 0.6964285969734192, 0.7182539701461792, 0.6686508059501648, 0.6567460298538208, 0.6746031641960144, 0.6517857313156128, 0.682539701461792, 0.6190476417541504, 0.7579365372657776, 0.7817460298538208, 0.670634925365448, 0.7400793433189392, 0.7509920597076416, 0.7242063283920288, 0.726190447807312, 0.7579365372657776, 0.761904776096344, 0.7182539701461792, 0.7420634627342224, 0.7519841194152832, 0.7559523582458496, 0.817460298538208, 0.7976190447807312, 0.807539701461792, 0.8184523582458496, 0.7986111044883728, 0.807539701461792, 0.8789682388305664, 0.8591269850730896, 0.898809552192688, 0.8125, 0.8551587462425232, 0.898809552192688, 0.7896825671195984, 0.8769841194152832, 0.8591269850730896, 0.8789682388305664, 0.85317462682724, 0.8789682388305664, 0.932539701461792, 0.9305555820465088, 0.9484127163887024, 0.9861111044883728], [0.1210317462682724, 0.1944444477558136, 0.2123015820980072, 0.230158731341362, 0.300595223903656, 0.30158731341362, 0.3134920597076416, 0.2906745970249176, 0.3472222089767456, 0.4325396716594696, 0.504960298538208, 0.5099206566810608, 0.52182537317276, 0.3928571343421936, 0.5, NaN, 0.5496031641960144, 0.5515872836112976, 0.5694444179534912, 0.567460298538208, 0.591269850730896, 0.6140872836112976, 0.5952380895614624, 0.613095223903656, 0.5952380895614624, 0.5555555820465088, 0.6388888955116272, 0.6021825671195984, 0.6051587462425232, 0.6478174328804016, 0.64682537317276, 0.64682537317276, 0.6309523582458496, 0.6726190447807312, 0.6507936716079712, 0.6686508059501648, 0.6845238208770752, 0.6934523582458496, 0.6746031641960144, 0.682539701461792, 0.7301587462425232, 0.7023809552192688, 0.7341269850730896, 0.726190447807312, 0.7361111044883728, 0.726190447807312, 0.7420634627342224, 0.8115079402923584, 0.7559523582458496, 0.6875, 0.6865079402923584, 0.7797619104385376, 0.7996031641960144, 0.7916666865348816, 0.8154761791229248, 0.8253968358039856, 0.8253968358039856, 0.7748016119003296, 0.851190447807312, 0.8313491940498352, 0.7658730149269104, 0.8452380895614624, 0.7896825671195984, 0.8611111044883728, 0.8571428656578064, 0.886904776096344, 0.898809552192688, 0.9047619104385376, 0.908730149269104, 0.8611111044883728, 0.9285714030265808, 0.9027777910232544, 0.9900793433189392], [0.1488095223903656, 0.2361111044883728, 0.204365074634552, 0.2539682686328888, 0.230158731341362, 0.2321428507566452, 0.244047611951828, 0.2361111044883728, 0.3621031641960144, 0.358134925365448, 0.3363095223903656, 0.44841268658638, 0.4404761791229248, 0.4057539701461792, 0.5416666865348816, 0.4503968358039856, NaN, 0.5376983880996704, 0.4841269850730896, 0.5773809552192688, 0.5952380895614624, 0.5019841194152832, 0.613095223903656, 0.5714285969734192, 0.5734127163887024, 0.52182537317276, 0.5922619104385376, 0.6210317611694336, 0.6001983880996704, 0.692460298538208, 0.6746031641960144, 0.6805555820465088, 0.670634925365448, 0.6527777910232544, 0.711309552192688, 0.6646825671195984, 0.6765872836112976, 0.64682537317276, 0.7202380895614624, 0.682539701461792, 0.7142857313156128, 0.7460317611694336, 0.7083333134651184, 0.7341269850730896, 0.7301587462425232, 0.7460317611694336, 0.7341269850730896, 0.6795634627342224, 0.7420634627342224, 0.7103174328804016, 0.716269850730896, 0.7638888955116272, 0.807539701461792, 0.8333333134651184, 0.8373016119003296, 0.778769850730896, 0.8313491940498352, 0.7876983880996704, 0.8670634627342224, 0.8472222089767456, 0.7658730149269104, 0.8690476417541504, 0.7966269850730896, 0.8611111044883728, 0.8710317611694336, 0.8998016119003296, 0.8492063283920288, 0.8650793433189392, 0.8908730149269104, 0.8392857313156128, 0.942460298538208, 0.8809523582458496, 0.9821428656578064], [0.105158731341362, 0.204365074634552, 0.2033730149269104, 0.22420634329319, 0.2003968209028244, 0.2162698358297348, 0.2202380895614624, 0.2113095223903656, 0.3789682686328888, 0.3601190447807312, 0.3650793731212616, 0.3452380895614624, 0.4305555522441864, 0.4454365074634552, 0.5099206566810608, 0.44841268658638, 0.4623015820980072, NaN, 0.5039682388305664, 0.5625, 0.5853174328804016, 0.5228174328804016, 0.5972222089767456, 0.5932539701461792, 0.545634925365448, 0.511904776096344, 0.60317462682724, 0.6150793433189392, 0.5694444179534912, 0.66567462682724, 0.682539701461792, 0.6666666865348816, 0.7003968358039856, 0.64682537317276, 0.6746031641960144, 0.6458333134651184, 0.6805555820465088, 0.6547619104385376, 0.704365074634552, 0.7182539701461792, 0.6765872836112976, 0.7440476417541504, 0.761904776096344, 0.716269850730896, 0.682539701461792, 0.7490079402923584, 0.7539682388305664, 0.6884920597076416, 0.6904761791229248, 0.7063491940498352, 0.7172619104385376, 0.8144841194152832, 0.773809552192688, 0.8313491940498352, 0.817460298538208, 0.783730149269104, 0.7916666865348816, 0.8392857313156128, 0.875, 0.8650793433189392, 0.7638888955116272, 0.8908730149269104, 0.851190447807312, 0.8154761791229248, 0.898809552192688, 0.8551587462425232, 0.8472222089767456, 0.8591269850730896, 0.875, 0.8809523582458496, 0.9464285969734192, 0.8888888955116272, 0.976190447807312], [0.130952388048172, 0.1567460298538208, 0.1964285671710968, 0.1746031790971756, 0.210317462682724, 0.2420634925365448, 0.2380952388048172, 0.2509920597076416, 0.2708333432674408, 0.3908730149269104, 0.4702380895614624, 0.460317462682724, 0.477182537317276, 0.4017857015132904, 0.4722222089767456, 0.4305555522441864, 0.5158730149269104, 0.4960317313671112, NaN, 0.5198412537574768, 0.5257936716079712, 0.5158730149269104, 0.5396825671195984, 0.5694444179534912, 0.5515872836112976, 0.5753968358039856, 0.5892857313156128, 0.5644841194152832, 0.5595238208770752, 0.5972222089767456, 0.6051587462425232, 0.6150793433189392, 0.6230158805847168, 0.6230158805847168, 0.6289682388305664, 0.7063491940498352, 0.6567460298538208, 0.6210317611694336, 0.6428571343421936, 0.648809552192688, 0.704365074634552, 0.6964285969734192, 0.6626983880996704, 0.7361111044883728, 0.6805555820465088, 0.716269850730896, 0.692460298538208, 0.8432539701461792, 0.6944444179534912, 0.7440476417541504, 0.75, 0.7757936716079712, 0.7222222089767456, 0.738095223903656, 0.8105158805847168, 0.8888888955116272, 0.7817460298538208, 0.7678571343421936, 0.841269850730896, 0.8115079402923584, 0.8095238208770752, 0.8710317611694336, 0.7936508059501648, 0.8055555820465088, 0.8710317611694336, 0.886904776096344, 0.863095223903656, 0.8452380895614624, 0.8829365372657776, 0.8998016119003296, 0.9186508059501648, 0.9384920597076416, 0.9940476417541504], [0.0992063507437706, 0.184523805975914, 0.1626984179019928, 0.1944444477558136, 0.1924603134393692, 0.1865079402923584, 0.2003968209028244, 0.2003968209028244, 0.341269850730896, 0.283730149269104, 0.2996031641960144, 0.3591269850730896, 0.329365074634552, 0.4047619104385376, 0.3988095223903656, 0.4325396716594696, 0.4226190447807312, 0.4375, 0.4801587164402008, NaN, 0.5625, 0.5257936716079712, 0.5773809552192688, 0.5039682388305664, 0.5476190447807312, 0.5138888955116272, 0.5605158805847168, 0.5882936716079712, 0.5992063283920288, 0.7063491940498352, 0.6914682388305664, 0.6458333134651184, 0.6517857313156128, 0.5972222089767456, 0.7182539701461792, 0.591269850730896, 0.591269850730896, 0.5734127163887024, 0.6865079402923584, 0.7361111044883728, 0.6686508059501648, 0.6884920597076416, 0.7470238208770752, 0.6805555820465088, 0.6894841194152832, 0.6984127163887024, 0.7599206566810608, 0.6964285969734192, 0.6944444179534912, 0.70932537317276, 0.7123016119003296, 0.841269850730896, 0.7976190447807312, 0.8730158805847168, 0.942460298538208, 0.7460317611694336, 0.8571428656578064, 0.8115079402923584, 0.8055555820465088, 0.8730158805847168, 0.7896825671195984, 0.8769841194152832, 0.8333333134651184, 0.8035714030265808, 0.9007936716079712, 0.8373016119003296, 0.8730158805847168, 0.9186508059501648, 0.9285714030265808, 0.89682537317276, 0.9345238208770752, 0.9176587462425232, 0.9920634627342224], [0.1011904776096344, 0.1666666716337204, 0.1666666716337204, 0.1726190447807312, 0.170634925365448, 0.1696428507566452, 0.164682537317276, 0.170634925365448, 0.3392857015132904, 0.2738095223903656, 0.2688491940498352, 0.3363095223903656, 0.3095238208770752, 0.3730158805847168, 0.4007936418056488, 0.408730149269104, 0.4047619104385376, 0.414682537317276, 0.4742063581943512, 0.4375, NaN, 0.5039682388305664, 0.5446428656578064, 0.466269850730896, 0.5188491940498352, 0.4920634925365448, 0.5158730149269104, 0.5595238208770752, 0.5595238208770752, 0.6240079402923584, 0.66567462682724, 0.625, 0.6170634627342224, 0.5753968358039856, 0.648809552192688, 0.5625, 0.5833333134651184, 0.545634925365448, 0.6686508059501648, 0.7003968358039856, 0.6388888955116272, 0.675595223903656, 0.6944444179534912, 0.6388888955116272, 0.66567462682724, 0.6835317611694336, 0.7103174328804016, 0.6845238208770752, 0.6875, 0.6944444179534912, 0.7003968358039856, 0.8015872836112976, 0.7519841194152832, 0.8442460298538208, 0.9077380895614624, 0.7341269850730896, 0.8214285969734192, 0.8035714030265808, 0.7916666865348816, 0.8561508059501648, 0.773809552192688, 0.8621031641960144, 0.8263888955116272, 0.7966269850730896, 0.8809523582458496, 0.829365074634552, 0.8551587462425232, 0.9047619104385376, 0.9027777910232544, 0.8829365372657776, 0.932539701461792, 0.9067460298538208, 0.9940476417541504], [0.1091269850730896, 0.1805555522441864, 0.2380952388048172, 0.204365074634552, 0.2876984179019928, 0.3313491940498352, 0.2996031641960144, 0.3134920597076416, 0.2113095223903656, 0.414682537317276, 0.4523809552192688, 0.454365074634552, 0.4523809552192688, 0.369047611951828, 0.4166666567325592, 0.38591268658638, 0.4980158805847168, 0.477182537317276, 0.4841269850730896, 0.4742063581943512, 0.4960317313671112, NaN, 0.5257936716079712, 0.545634925365448, 0.5277777910232544, 0.5317460298538208, 0.5416666865348816, 0.5357142686843872, 0.5317460298538208, 0.5694444179534912, 0.5376983880996704, 0.5297619104385376, 0.5734127163887024, 0.5952380895614624, 0.5853174328804016, 0.5853174328804016, 0.5873016119003296, 0.5932539701461792, 0.5565476417541504, 0.591269850730896, 0.6309523582458496, 0.58432537317276, 0.6507936716079712, 0.658730149269104, 0.6884920597076416, 0.6428571343421936, 0.6527777910232544, 0.7460317611694336, 0.7142857313156128, 0.6329365372657776, 0.6329365372657776, 0.7103174328804016, 0.7123016119003296, 0.6845238208770752, 0.738095223903656, 0.841269850730896, 0.7638888955116272, 0.6944444179534912, 0.773809552192688, 0.7976190447807312, 0.738095223903656, 0.7936508059501648, 0.726190447807312, 0.8035714030265808, 0.817460298538208, 0.8650793433189392, 0.875, 0.8908730149269104, 0.8591269850730896, 0.8452380895614624, 0.8769841194152832, 0.9007936716079712, 0.9920634627342224], [0.0813492089509964, 0.15079365670681, 0.1468254029750824, 0.1825396865606308, 0.1527777761220932, 0.1488095223903656, 0.170634925365448, 0.1626984179019928, 0.30158731341362, 0.261904776096344, 0.2480158656835556, 0.3492063581943512, 0.3214285671710968, 0.375, 0.4126984179019928, 0.4047619104385376, 0.386904776096344, 0.4027777910232544, 0.460317462682724, 0.4226190447807312, 0.4553571343421936, 0.4742063581943512, NaN, 0.4821428656578064, 0.4960317313671112, 0.5, 0.5178571343421936, 0.5763888955116272, 0.5476190447807312, 0.5684523582458496, 0.5932539701461792, 0.5555555820465088, 0.5952380895614624, 0.557539701461792, 0.596230149269104, 0.5535714030265808, 0.5734127163887024, 0.5615079402923584, 0.6121031641960144, 0.6428571343421936, 0.6547619104385376, 0.6626983880996704, 0.6646825671195984, 0.6329365372657776, 0.6944444179534912, 0.6785714030265808, 0.670634925365448, 0.6428571343421936, 0.7172619104385376, 0.6984127163887024, 0.692460298538208, 0.7103174328804016, 0.8015872836112976, 0.85317462682724, 0.8551587462425232, 0.6904761791229248, 0.858134925365448, 0.773809552192688, 0.8015872836112976, 0.8283730149269104, 0.7519841194152832, 0.817460298538208, 0.7916666865348816, 0.8214285969734192, 0.8373016119003296, 0.807539701461792, 0.863095223903656, 0.8908730149269104, 0.9146825671195984, 0.8313491940498352, 0.8908730149269104, 0.89682537317276, 0.9920634627342224], [0.0972222238779068, 0.15079365670681, 0.144841268658638, 0.164682537317276, 0.1944444477558136, 0.1527777761220932, 0.170634925365448, 0.190476194024086, 0.3095238208770752, 0.261904776096344, 0.3373015820980072, 0.3779761791229248, 0.3888888955116272, 0.3531745970249176, 0.403769850730896, 0.386904776096344, 0.4285714328289032, 0.4067460298538208, 0.4305555522441864, 0.4960317313671112, 0.533730149269104, 0.454365074634552, 0.5178571343421936, NaN, 0.5486111044883728, 0.4841269850730896, 0.5, 0.533730149269104, 0.5972222089767456, 0.5694444179534912, 0.5813491940498352, 0.5833333134651184, 0.5376983880996704, 0.6349206566810608, 0.5763888955116272, 0.5823412537574768, 0.5694444179534912, 0.5446428656578064, 0.625, 0.6349206566810608, 0.6230158805847168, 0.6726190447807312, 0.6805555820465088, 0.6527777910232544, 0.6507936716079712, 0.7142857313156128, 0.6815476417541504, 0.6607142686843872, 0.6765872836112976, 0.670634925365448, 0.6785714030265808, 0.7321428656578064, 0.7480158805847168, 0.7559523582458496, 0.761904776096344, 0.7658730149269104, 0.773809552192688, 0.7440476417541504, 0.795634925365448, 0.8134920597076416, 0.7539682388305664, 0.8095238208770752, 0.773809552192688, 0.8551587462425232, 0.8313491940498352, 0.8204365372657776, 0.8194444179534912, 0.8650793433189392, 0.8472222089767456, 0.8214285969734192, 0.8829365372657776, 0.8670634627342224, 0.9801587462425232], [0.0873015895485878, 0.1468254029750824, 0.1388888955116272, 0.1686507910490036, 0.1746031790971756, 0.1666666716337204, 0.190476194024086, 0.1855158656835556, 0.3134920597076416, 0.2420634925365448, 0.295634925365448, 0.3928571343421936, 0.3561508059501648, 0.3601190447807312, 0.4047619104385376, 0.4047619104385376, 0.42658731341362, 0.454365074634552, 0.44841268658638, 0.4523809552192688, 0.4811508059501648, 0.4722222089767456, 0.5039682388305664, 0.4513888955116272, NaN, 0.4523809552192688, 0.477182537317276, 0.5228174328804016, 0.596230149269104, 0.5367063283920288, 0.58432537317276, 0.5476190447807312, 0.5436508059501648, 0.5535714030265808, 0.5496031641960144, 0.5565476417541504, 0.545634925365448, 0.5515872836112976, 0.6150793433189392, 0.6388888955116272, 0.6160714030265808, 0.6269841194152832, 0.682539701461792, 0.6388888955116272, 0.6428571343421936, 0.6408730149269104, 0.6884920597076416, 0.6904761791229248, 0.6964285969734192, 0.6676587462425232, 0.6666666865348816, 0.7063491940498352, 0.7579365372657776, 0.761904776096344, 0.7638888955116272, 0.783730149269104, 0.8035714030265808, 0.7916666865348816, 0.8234127163887024, 0.7817460298538208, 0.7440476417541504, 0.8154761791229248, 0.807539701461792, 0.9305555820465088, 0.8482142686843872, 0.8829365372657776, 0.8680555820465088, 0.932539701461792, 0.9365079402923584, 0.846230149269104, 0.9107142686843872, 0.898809552192688, 0.9900793433189392], [0.1626984179019928, 0.2599206268787384, 0.2797619104385376, 0.30158731341362, 0.2380952388048172, 0.247023805975914, 0.261904776096344, 0.2519841194152832, 0.369047611951828, 0.3313491940498352, 0.420634925365448, 0.4107142984867096, 0.4702380895614624, 0.414682537317276, 0.4623015820980072, 0.4444444477558136, 0.4781745970249176, 0.488095223903656, 0.4246031641960144, 0.4861111044883728, 0.5079365372657776, 0.4682539701461792, 0.5, 0.5158730149269104, 0.5476190447807312, NaN, 0.5297619104385376, 0.5158730149269104, 0.5535714030265808, 0.5515872836112976, 0.5277777910232544, 0.5595238208770752, 0.5615079402923584, 0.5634920597076416, 0.5595238208770752, 0.6269841194152832, 0.6527777910232544, 0.5734127163887024, 0.5892857313156128, 0.6021825671195984, 0.5882936716079712, 0.6805555820465088, 0.5654761791229248, 0.6746031641960144, 0.6309523582458496, 0.692460298538208, 0.591269850730896, 0.601190447807312, 0.6309523582458496, 0.60317462682724, 0.5952380895614624, 0.6408730149269104, 0.699404776096344, 0.670634925365448, 0.6884920597076416, 0.7123016119003296, 0.711309552192688, 0.7519841194152832, 0.7519841194152832, 0.7202380895614624, 0.7301587462425232, 0.761904776096344, 0.7559523582458496, 0.761904776096344, 0.766865074634552, 0.7817460298538208, 0.716269850730896, 0.7658730149269104, 0.7777777910232544, 0.7817460298538208, 0.7976190447807312, 0.8253968358039856, 0.9821428656578064], [0.0793650820851326, 0.1230158731341362, 0.1230158731341362, 0.1329365074634552, 0.1805555522441864, 0.16170634329319, 0.1210317462682724, 0.1805555522441864, 0.2519841194152832, 0.2033730149269104, 0.2599206268787384, 0.3482142984867096, 0.3194444477558136, 0.3055555522441864, 0.3948412835597992, 0.3611111044883728, 0.4077380895614624, 0.3968254029750824, 0.4107142984867096, 0.4394841194152832, 0.4841269850730896, 0.4583333432674408, 0.4821428656578064, 0.5, 0.5228174328804016, 0.4702380895614624, NaN, 0.5138888955116272, 0.5664682388305664, 0.5446428656578064, 0.5515872836112976, 0.5634920597076416, 0.5178571343421936, 0.5992063283920288, 0.5833333134651184, 0.5853174328804016, 0.5496031641960144, 0.5714285969734192, 0.6091269850730896, 0.6289682388305664, 0.7053571343421936, 0.6626983880996704, 0.6845238208770752, 0.6696428656578064, 0.6617063283920288, 0.658730149269104, 0.6805555820465088, 0.6378968358039856, 0.6944444179534912, 0.6428571343421936, 0.6438491940498352, 0.7271825671195984, 0.7698412537574768, 0.8055555820465088, 0.841269850730896, 0.7698412537574768, 0.8115079402923584, 0.704365074634552, 0.7996031641960144, 0.8154761791229248, 0.7480158805847168, 0.829365074634552, 0.7440476417541504, 0.8571428656578064, 0.8551587462425232, 0.8373016119003296, 0.8214285969734192, 0.8948412537574768, 0.8948412537574768, 0.807539701461792, 0.89682537317276, 0.8660714030265808, 0.97817462682724], [0.075396828353405, 0.1408730149269104, 0.1498015820980072, 0.1726190447807312, 0.1527777761220932, 0.1547619104385376, 0.1785714328289032, 0.1666666716337204, 0.3115079402923584, 0.2599206268787384, 0.244047611951828, 0.32341268658638, 0.2936508059501648, 0.3670634925365448, 0.3779761791229248, 0.397817462682724, 0.3789682686328888, 0.3849206268787384, 0.4355158805847168, 0.4117063581943512, 0.4404761791229248, 0.4642857015132904, 0.4236111044883728, 0.466269850730896, 0.477182537317276, 0.4841269850730896, 0.4861111044883728, NaN, 0.533730149269104, 0.567460298538208, 0.5882936716079712, 0.5615079402923584, 0.5763888955116272, 0.5396825671195984, 0.5892857313156128, 0.5317460298538208, 0.5734127163887024, 0.5515872836112976, 0.5972222089767456, 0.6428571343421936, 0.6210317611694336, 0.64682537317276, 0.6180555820465088, 0.6240079402923584, 0.6746031641960144, 0.66567462682724, 0.6458333134651184, 0.6408730149269104, 0.6865079402923584, 0.6984127163887024, 0.7003968358039856, 0.7023809552192688, 0.7916666865348816, 0.8472222089767456, 0.829365074634552, 0.682539701461792, 0.8333333134651184, 0.7678571343421936, 0.7797619104385376, 0.8115079402923584, 0.7579365372657776, 0.8055555820465088, 0.7827380895614624, 0.807539701461792, 0.829365074634552, 0.7976190447807312, 0.85317462682724, 0.8789682388305664, 0.9047619104385376, 0.8234127163887024, 0.898809552192688, 0.9027777910232544, 0.9920634627342224], [0.1071428582072258, 0.1527777761220932, 0.1696428507566452, 0.1855158656835556, 0.1666666716337204, 0.1785714328289032, 0.1984127014875412, 0.190476194024086, 0.2857142984867096, 0.2549603283405304, 0.3174603283405304, 0.3888888955116272, 0.3273809552192688, 0.346230149269104, 0.3948412835597992, 0.3948412835597992, 0.3998015820980072, 0.4305555522441864, 0.4404761791229248, 0.4007936418056488, 0.4404761791229248, 0.4682539701461792, 0.4523809552192688, 0.4027777910232544, 0.403769850730896, 0.4464285671710968, 0.4335317313671112, 0.466269850730896, NaN, 0.4841269850730896, 0.5079365372657776, 0.5228174328804016, 0.5257936716079712, 0.5357142686843872, 0.5069444179534912, 0.528769850730896, 0.5357142686843872, 0.4623015820980072, 0.5773809552192688, 0.6111111044883728, 0.5853174328804016, 0.5773809552192688, 0.6438491940498352, 0.6071428656578064, 0.636904776096344, 0.625, 0.6666666865348816, 0.6765872836112976, 0.6746031641960144, 0.6805555820465088, 0.6845238208770752, 0.7023809552192688, 0.6805555820465088, 0.6607142686843872, 0.7123016119003296, 0.8015872836112976, 0.7400793433189392, 0.7916666865348816, 0.8194444179534912, 0.795634925365448, 0.75, 0.7827380895614624, 0.8134920597076416, 0.8888888955116272, 0.8303571343421936, 0.8611111044883728, 0.85317462682724, 0.875, 0.8769841194152832, 0.8670634627342224, 0.898809552192688, 0.898809552192688, 0.9900793433189392], [0.0674603208899498, 0.1498015820980072, 0.1388888955116272, 0.164682537317276, 0.144841268658638, 0.1408730149269104, 0.1408730149269104, 0.1547619104385376, 0.2857142984867096, 0.2281745970249176, 0.1825396865606308, 0.2539682686328888, 0.2738095223903656, 0.3422619104385376, 0.3392857015132904, 0.352182537317276, 0.3075396716594696, 0.3343254029750824, 0.4027777910232544, 0.2936508059501648, 0.3759920597076416, 0.4305555522441864, 0.431547611951828, 0.4305555522441864, 0.4632936418056488, 0.44841268658638, 0.4553571343421936, 0.4325396716594696, 0.5158730149269104, NaN, 0.5327380895614624, 0.5158730149269104, 0.5357142686843872, 0.5277777910232544, 0.5426587462425232, 0.52182537317276, 0.538690447807312, 0.5, 0.5853174328804016, 0.6051587462425232, 0.5952380895614624, 0.60317462682724, 0.5833333134651184, 0.5773809552192688, 0.6071428656578064, 0.6190476417541504, 0.6210317611694336, 0.6339285969734192, 0.658730149269104, 0.6726190447807312, 0.6726190447807312, 0.6557539701461792, 0.778769850730896, 0.8730158805847168, 0.8730158805847168, 0.6914682388305664, 0.8085317611694336, 0.7579365372657776, 0.7559523582458496, 0.7698412537574768, 0.7341269850730896, 0.7757936716079712, 0.77182537317276, 0.7936508059501648, 0.7986111044883728, 0.77182537317276, 0.817460298538208, 0.8670634627342224, 0.9027777910232544, 0.8214285969734192, 0.8809523582458496, 0.8779761791229248, 0.9821428656578064], [0.0734127014875412, 0.125, 0.1339285671710968, 0.1408730149269104, 0.1170634925365448, 0.1369047611951828, 0.125, 0.1488095223903656, 0.2777777910232544, 0.2222222238779068, 0.2281745970249176, 0.25, 0.255952388048172, 0.3194444477558136, 0.3035714328289032, 0.3531745970249176, 0.3253968358039856, 0.3174603283405304, 0.3948412835597992, 0.3085317313671112, 0.3343254029750824, 0.4623015820980072, 0.4067460298538208, 0.4186508059501648, 0.4156745970249176, 0.4722222089767456, 0.44841268658638, 0.4117063581943512, 0.4920634925365448, 0.4672619104385376, NaN, 0.5198412537574768, 0.5257936716079712, 0.4920634925365448, 0.48908731341362, 0.523809552192688, 0.533730149269104, 0.4980158805847168, 0.591269850730896, 0.625, 0.5932539701461792, 0.6170634627342224, 0.629960298538208, 0.601190447807312, 0.6289682388305664, 0.6408730149269104, 0.658730149269104, 0.6230158805847168, 0.648809552192688, 0.6914682388305664, 0.6944444179534912, 0.7301587462425232, 0.7460317611694336, 0.7658730149269104, 0.8035714030265808, 0.704365074634552, 0.7876983880996704, 0.7420634627342224, 0.7876983880996704, 0.863095223903656, 0.761904776096344, 0.7916666865348816, 0.7658730149269104, 0.7658730149269104, 0.8333333134651184, 0.8134920597076416, 0.8432539701461792, 0.8611111044883728, 0.8690476417541504, 0.8472222089767456, 0.8829365372657776, 0.908730149269104, 0.988095223903656], [0.105158731341362, 0.1607142835855484, 0.1607142835855484, 0.1666666716337204, 0.1408730149269104, 0.164682537317276, 0.190476194024086, 0.1527777761220932, 0.2876984179019928, 0.2321428507566452, 0.2281745970249176, 0.2380952388048172, 0.2936508059501648, 0.3343254029750824, 0.2817460298538208, 0.3531745970249176, 0.3194444477558136, 0.3333333432674408, 0.3849206268787384, 0.3541666567325592, 0.375, 0.4702380895614624, 0.4444444477558136, 0.4166666567325592, 0.4523809552192688, 0.4404761791229248, 0.4365079402923584, 0.4384920597076416, 0.477182537317276, 0.4841269850730896, 0.4801587164402008, NaN, 0.5178571343421936, 0.511904776096344, 0.4920634925365448, 0.4910714328289032, 0.5317460298538208, 0.4761904776096344, 0.6795634627342224, 0.7003968358039856, 0.545634925365448, 0.5892857313156128, 0.579365074634552, 0.5644841194152832, 0.6398809552192688, 0.6121031641960144, 0.5992063283920288, 0.6547619104385376, 0.6388888955116272, 0.6746031641960144, 0.6765872836112976, 0.670634925365448, 0.7083333134651184, 0.7073412537574768, 0.7142857313156128, 0.738095223903656, 0.726190447807312, 0.7996031641960144, 0.7420634627342224, 0.807539701461792, 0.7599206566810608, 0.7857142686843872, 0.817460298538208, 0.7539682388305664, 0.8085317611694336, 0.8194444179534912, 0.829365074634552, 0.8333333134651184, 0.8353174328804016, 0.851190447807312, 0.89682537317276, 0.8938491940498352, 0.9801587462425232], [0.0992063507437706, 0.1805555522441864, 0.1626984179019928, 0.1884920597076416, 0.1468254029750824, 0.1746031790971756, 0.181547611951828, 0.170634925365448, 0.3134920597076416, 0.2857142984867096, 0.244047611951828, 0.1775793582201004, 0.3055555522441864, 0.358134925365448, 0.3313491940498352, 0.369047611951828, 0.329365074634552, 0.2996031641960144, 0.3769841194152832, 0.3482142984867096, 0.3829365074634552, 0.42658731341362, 0.4047619104385376, 0.4623015820980072, 0.4563491940498352, 0.4384920597076416, 0.4821428656578064, 0.4236111044883728, 0.4742063581943512, 0.4642857015132904, 0.4742063581943512, 0.4821428656578064, NaN, 0.5178571343421936, 0.4920634925365448, 0.5535714030265808, 0.5654761791229248, 0.5376983880996704, 0.545634925365448, 0.5466269850730896, 0.5644841194152832, 0.6111111044883728, 0.557539701461792, 0.6289682388305664, 0.6111111044883728, 0.648809552192688, 0.579365074634552, 0.6071428656578064, 0.6349206566810608, 0.692460298538208, 0.7003968358039856, 0.6349206566810608, 0.7103174328804016, 0.6666666865348816, 0.6547619104385376, 0.6944444179534912, 0.7242063283920288, 0.7638888955116272, 0.7817460298538208, 0.77182537317276, 0.7698412537574768, 0.7569444179534912, 0.7847222089767456, 0.7142857313156128, 0.7638888955116272, 0.7916666865348816, 0.8055555820465088, 0.8095238208770752, 0.829365074634552, 0.829365074634552, 0.8591269850730896, 0.8928571343421936, 0.9722222089767456], [0.095238097012043, 0.1359127014875412, 0.1527777761220932, 0.1359127014875412, 0.1607142835855484, 0.1666666716337204, 0.1547619104385376, 0.1785714328289032, 0.2341269850730896, 0.230158731341362, 0.289682537317276, 0.3154761791229248, 0.3095238208770752, 0.2986111044883728, 0.3432539701461792, 0.3273809552192688, 0.3472222089767456, 0.3531745970249176, 0.3769841194152832, 0.4027777910232544, 0.4246031641960144, 0.4047619104385376, 0.4424603283405304, 0.3650793731212616, 0.4464285671710968, 0.4365079402923584, 0.4007936418056488, 0.460317462682724, 0.4642857015132904, 0.4722222089767456, 0.5079365372657776, 0.488095223903656, 0.4821428656578064, NaN, 0.5059523582458496, 0.4781745970249176, 0.4781745970249176, 0.4067460298538208, 0.545634925365448, 0.5853174328804016, 0.5228174328804016, 0.5396825671195984, 0.6567460298538208, 0.5436508059501648, 0.5952380895614624, 0.5773809552192688, 0.6646825671195984, 0.6527777910232544, 0.6349206566810608, 0.6507936716079712, 0.6567460298538208, 0.7003968358039856, 0.6746031641960144, 0.6329365372657776, 0.7003968358039856, 0.7440476417541504, 0.7400793433189392, 0.7440476417541504, 0.783730149269104, 0.783730149269104, 0.7341269850730896, 0.773809552192688, 0.7688491940498352, 0.7321428656578064, 0.8055555820465088, 0.8095238208770752, 0.8472222089767456, 0.8184523582458496, 0.8154761791229248, 0.8591269850730896, 0.8571428656578064, 0.8789682388305664, 0.9841269850730896], [0.0674603208899498, 0.1369047611951828, 0.1408730149269104, 0.1567460298538208, 0.125, 0.127976194024086, 0.1349206417798996, 0.1468254029750824, 0.2777777910232544, 0.210317462682724, 0.1726190447807312, 0.2321428507566452, 0.2876984179019928, 0.3174603283405304, 0.3253968358039856, 0.3492063581943512, 0.2886904776096344, 0.3253968358039856, 0.3710317313671112, 0.2817460298538208, 0.3511904776096344, 0.414682537317276, 0.403769850730896, 0.4236111044883728, 0.4503968358039856, 0.4404761791229248, 0.4166666567325592, 0.4107142984867096, 0.4930555522441864, 0.4573412835597992, 0.5109127163887024, 0.5079365372657776, 0.5079365372657776, 0.494047611951828, NaN, 0.494047611951828, 0.52182537317276, 0.4980158805847168, 0.557539701461792, 0.5773809552192688, 0.579365074634552, 0.5753968358039856, 0.5486111044883728, 0.5773809552192688, 0.6140872836112976, 0.5952380895614624, 0.5952380895614624, 0.613095223903656, 0.6547619104385376, 0.6567460298538208, 0.6597222089767456, 0.6190476417541504, 0.7579365372657776, 0.8353174328804016, 0.8373016119003296, 0.6805555820465088, 0.8055555820465088, 0.7440476417541504, 0.7321428656578064, 0.7658730149269104, 0.7301587462425232, 0.7480158805847168, 0.7698412537574768, 0.7857142686843872, 0.7896825671195984, 0.7678571343421936, 0.8184523582458496, 0.8690476417541504, 0.886904776096344, 0.8134920597076416, 0.8789682388305664, 0.875, 0.97817462682724], [0.1031746044754982, 0.1696428507566452, 0.1626984179019928, 0.164682537317276, 0.1329365074634552, 0.1656745970249176, 0.1746031790971756, 0.1785714328289032, 0.230158731341362, 0.2827380895614624, 0.2430555522441864, 0.3055555522441864, 0.2757936418056488, 0.3035714328289032, 0.3482142984867096, 0.3313491940498352, 0.335317462682724, 0.3541666567325592, 0.2936508059501648, 0.408730149269104, 0.4375, 0.414682537317276, 0.4464285671710968, 0.4176587164402008, 0.443452388048172, 0.3730158805847168, 0.414682537317276, 0.4682539701461792, 0.471230149269104, 0.4781745970249176, 0.4761904776096344, 0.5089285969734192, 0.4464285671710968, 0.52182537317276, 0.5059523582458496, NaN, 0.4365079402923584, 0.5079365372657776, 0.5476190447807312, 0.5555555820465088, 0.511904776096344, 0.5783730149269104, 0.6051587462425232, 0.6845238208770752, 0.6051587462425232, 0.5753968358039856, 0.6230158805847168, 0.5615079402923584, 0.6111111044883728, 0.6388888955116272, 0.6448412537574768, 0.6269841194152832, 0.7073412537574768, 0.6646825671195984, 0.6736111044883728, 0.7539682388305664, 0.716269850730896, 0.75, 0.7748016119003296, 0.7420634627342224, 0.7440476417541504, 0.7440476417541504, 0.7817460298538208, 0.7460317611694336, 0.7757936716079712, 0.9484127163887024, 0.7916666865348816, 0.807539701461792, 0.7817460298538208, 0.8373016119003296, 0.8650793433189392, 0.8789682388305664, 0.976190447807312], [0.0555555559694767, 0.1031746044754982, 0.1130952388048172, 0.1230158731341362, 0.1130952388048172, 0.1210317462682724, 0.1230158731341362, 0.1210317462682724, 0.2599206268787384, 0.244047611951828, 0.2678571343421936, 0.2817460298538208, 0.2480158656835556, 0.3025793731212616, 0.3174603283405304, 0.3154761791229248, 0.32341268658638, 0.3194444477558136, 0.3432539701461792, 0.408730149269104, 0.4166666567325592, 0.4126984179019928, 0.42658731341362, 0.4305555522441864, 0.454365074634552, 0.3472222089767456, 0.4503968358039856, 0.42658731341362, 0.4642857015132904, 0.4613095223903656, 0.466269850730896, 0.4682539701461792, 0.4345238208770752, 0.52182537317276, 0.4781745970249176, 0.5634920597076416, NaN, 0.523809552192688, 0.5178571343421936, 0.5515872836112976, 0.5625, 0.5892857313156128, 0.5555555820465088, 0.6617063283920288, 0.6190476417541504, 0.738095223903656, 0.5873016119003296, 0.5654761791229248, 0.6230158805847168, 0.6527777910232544, 0.6617063283920288, 0.6071428656578064, 0.6964285969734192, 0.6944444179534912, 0.6805555820465088, 0.738095223903656, 0.716269850730896, 0.7430555820465088, 0.85317462682724, 0.778769850730896, 0.75, 0.7757936716079712, 0.783730149269104, 0.7638888955116272, 0.8194444179534912, 0.846230149269104, 0.7857142686843872, 0.7698412537574768, 0.7678571343421936, 0.8432539701461792, 0.8809523582458496, 0.8730158805847168, 0.976190447807312], [0.0555555559694767, 0.1130952388048172, 0.0912698432803154, 0.1369047611951828, 0.164682537317276, 0.1210317462682724, 0.0912698432803154, 0.1299603134393692, 0.2678571343421936, 0.2222222238779068, 0.2876984179019928, 0.2996031641960144, 0.3214285671710968, 0.2698412835597992, 0.380952388048172, 0.306547611951828, 0.3531745970249176, 0.3452380895614624, 0.3789682686328888, 0.42658731341362, 0.454365074634552, 0.4067460298538208, 0.4384920597076416, 0.4553571343421936, 0.44841268658638, 0.42658731341362, 0.4285714328289032, 0.44841268658638, 0.5376983880996704, 0.5, 0.5019841194152832, 0.523809552192688, 0.4623015820980072, 0.5932539701461792, 0.5019841194152832, 0.4920634925365448, 0.4761904776096344, NaN, 0.5664682388305664, 0.5634920597076416, 0.5515872836112976, 0.6091269850730896, 0.567460298538208, 0.5714285969734192, 0.5813491940498352, 0.5853174328804016, 0.591269850730896, 0.5595238208770752, 0.6230158805847168, 0.591269850730896, 0.5932539701461792, 0.5952380895614624, 0.6646825671195984, 0.7182539701461792, 0.6865079402923584, 0.6626983880996704, 0.6726190447807312, 0.6428571343421936, 0.6984127163887024, 0.6984127163887024, 0.6547619104385376, 0.7083333134651184, 0.658730149269104, 0.8065476417541504, 0.7103174328804016, 0.7242063283920288, 0.7361111044883728, 0.795634925365448, 0.7817460298538208, 0.7023809552192688, 0.7896825671195984, 0.7698412537574768, 0.9404761791229248], [0.0932539701461792, 0.1468254029750824, 0.1488095223903656, 0.1607142835855484, 0.1210317462682724, 0.1517857164144516, 0.16170634329319, 0.1488095223903656, 0.2549603283405304, 0.204365074634552, 0.2202380895614624, 0.22420634329319, 0.2738095223903656, 0.3045634925365448, 0.2420634925365448, 0.3253968358039856, 0.2797619104385376, 0.295634925365448, 0.3571428656578064, 0.3134920597076416, 0.3313491940498352, 0.443452388048172, 0.3878968358039856, 0.375, 0.3849206268787384, 0.4107142984867096, 0.3908730149269104, 0.4027777910232544, 0.4226190447807312, 0.414682537317276, 0.408730149269104, 0.3204365074634552, 0.454365074634552, 0.454365074634552, 0.4424603283405304, 0.4523809552192688, 0.4821428656578064, 0.4335317313671112, NaN, 0.6240079402923584, 0.4920634925365448, 0.5158730149269104, 0.5396825671195984, 0.5367063283920288, 0.5853174328804016, 0.579365074634552, 0.5565476417541504, 0.6121031641960144, 0.6190476417541504, 0.658730149269104, 0.66567462682724, 0.6269841194152832, 0.670634925365448, 0.6646825671195984, 0.6726190447807312, 0.711309552192688, 0.7103174328804016, 0.7658730149269104, 0.716269850730896, 0.7638888955116272, 0.7400793433189392, 0.761904776096344, 0.7857142686843872, 0.72817462682724, 0.783730149269104, 0.7876983880996704, 0.8015872836112976, 0.8273809552192688, 0.817460298538208, 0.8253968358039856, 0.875, 0.8948412537574768, 0.9801587462425232], [0.0992063507437706, 0.1339285671710968, 0.1428571492433548, 0.1408730149269104, 0.1388888955116272, 0.1587301641702652, 0.1686507910490036, 0.164682537317276, 0.244047611951828, 0.2023809552192688, 0.2212301641702652, 0.2311507910490036, 0.2361111044883728, 0.3075396716594696, 0.2182539701461792, 0.3174603283405304, 0.3174603283405304, 0.2817460298538208, 0.3511904776096344, 0.2638888955116272, 0.2996031641960144, 0.408730149269104, 0.3571428656578064, 0.3650793731212616, 0.3611111044883728, 0.397817462682724, 0.3710317313671112, 0.3571428656578064, 0.3888888955116272, 0.3948412835597992, 0.375, 0.2996031641960144, 0.4533730149269104, 0.414682537317276, 0.4226190447807312, 0.4444444477558136, 0.44841268658638, 0.4365079402923584, 0.3759920597076416, NaN, 0.44841268658638, 0.4980158805847168, 0.533730149269104, 0.5198412537574768, 0.5634920597076416, 0.5813491940498352, 0.5496031641960144, 0.6091269850730896, 0.5714285969734192, 0.6607142686843872, 0.658730149269104, 0.5932539701461792, 0.6567460298538208, 0.60317462682724, 0.6349206566810608, 0.7182539701461792, 0.682539701461792, 0.7480158805847168, 0.7083333134651184, 0.7638888955116272, 0.7361111044883728, 0.7400793433189392, 0.7827380895614624, 0.692460298538208, 0.7599206566810608, 0.795634925365448, 0.8313491940498352, 0.8134920597076416, 0.795634925365448, 0.8571428656578064, 0.851190447807312, 0.9126983880996704, 0.97817462682724], [0.065476194024086, 0.0843253955245018, 0.0912698432803154, 0.115079365670681, 0.0962301567196846, 0.1031746044754982, 0.095238097012043, 0.1091269850730896, 0.1775793582201004, 0.1726190447807312, 0.1884920597076416, 0.25, 0.2281745970249176, 0.2202380895614624, 0.329365074634552, 0.2698412835597992, 0.2857142984867096, 0.32341268658638, 0.295634925365448, 0.3313491940498352, 0.3611111044883728, 0.369047611951828, 0.3452380895614624, 0.3769841194152832, 0.3839285671710968, 0.4117063581943512, 0.2946428656578064, 0.3789682686328888, 0.414682537317276, 0.4047619104385376, 0.4067460298538208, 0.454365074634552, 0.4355158805847168, 0.477182537317276, 0.420634925365448, 0.488095223903656, 0.4375, 0.44841268658638, 0.5079365372657776, 0.5515872836112976, NaN, 0.545634925365448, 0.5198412537574768, 0.5416666865348816, 0.5396825671195984, 0.5634920597076416, 0.5644841194152832, 0.533730149269104, 0.5873016119003296, 0.629960298538208, 0.6230158805847168, 0.5873016119003296, 0.6527777910232544, 0.636904776096344, 0.64682537317276, 0.6676587462425232, 0.7023809552192688, 0.6319444179534912, 0.7638888955116272, 0.7430555820465088, 0.7063491940498352, 0.704365074634552, 0.6746031641960144, 0.7539682388305664, 0.7539682388305664, 0.7857142686843872, 0.7440476417541504, 0.783730149269104, 0.7857142686843872, 0.7559523582458496, 0.8273809552192688, 0.8184523582458496, 0.9642857313156128], [0.0813492089509964, 0.0972222238779068, 0.1329365074634552, 0.1190476194024086, 0.1011904776096344, 0.1369047611951828, 0.1349206417798996, 0.1369047611951828, 0.2083333283662796, 0.190476194024086, 0.1537698358297348, 0.2113095223903656, 0.2261904776096344, 0.2718254029750824, 0.2599206268787384, 0.2976190447807312, 0.2539682686328888, 0.255952388048172, 0.3035714328289032, 0.3115079402923584, 0.324404776096344, 0.4156745970249176, 0.3373015820980072, 0.3273809552192688, 0.3730158805847168, 0.3194444477558136, 0.3373015820980072, 0.3531745970249176, 0.4226190447807312, 0.3968254029750824, 0.3829365074634552, 0.4107142984867096, 0.3888888955116272, 0.460317462682724, 0.4246031641960144, 0.4216269850730896, 0.4107142984867096, 0.3908730149269104, 0.4841269850730896, 0.5019841194152832, 0.454365074634552, NaN, 0.5178571343421936, 0.523809552192688, 0.545634925365448, 0.5555555820465088, 0.5357142686843872, 0.5515872836112976, 0.5873016119003296, 0.6349206566810608, 0.6507936716079712, 0.5753968358039856, 0.64682537317276, 0.6150793433189392, 0.6448412537574768, 0.7351190447807312, 0.6805555820465088, 0.7103174328804016, 0.7490079402923584, 0.7341269850730896, 0.7242063283920288, 0.7460317611694336, 0.7519841194152832, 0.704365074634552, 0.7976190447807312, 0.8472222089767456, 0.7976190447807312, 0.7876983880996704, 0.783730149269104, 0.8333333134651184, 0.875, 0.8670634627342224, 0.97817462682724], [0.0714285746216774, 0.0912698432803154, 0.0873015895485878, 0.1071428582072258, 0.0972222238779068, 0.1081349179148674, 0.0892857164144516, 0.1180555522441864, 0.1984127014875412, 0.1527777761220932, 0.1785714328289032, 0.190476194024086, 0.2152777761220932, 0.244047611951828, 0.2490079402923584, 0.2658730149269104, 0.2916666567325592, 0.2380952388048172, 0.3373015820980072, 0.2529761791229248, 0.3055555522441864, 0.3492063581943512, 0.335317462682724, 0.3194444477558136, 0.3174603283405304, 0.4345238208770752, 0.3154761791229248, 0.3819444477558136, 0.3561508059501648, 0.4166666567325592, 0.3700396716594696, 0.420634925365448, 0.4424603283405304, 0.3432539701461792, 0.4513888955116272, 0.3948412835597992, 0.4444444477558136, 0.4325396716594696, 0.460317462682724, 0.466269850730896, 0.4801587164402008, 0.4821428656578064, NaN, 0.4632936418056488, 0.4980158805847168, 0.5277777910232544, 0.5664682388305664, 0.5714285969734192, 0.545634925365448, 0.6408730149269104, 0.6408730149269104, 0.7559523582458496, 0.6051587462425232, 0.6388888955116272, 0.7053571343421936, 0.6984127163887024, 0.658730149269104, 0.6547619104385376, 0.6626983880996704, 0.7301587462425232, 0.7301587462425232, 0.726190447807312, 0.6785714030265808, 0.7242063283920288, 0.7698412537574768, 0.7519841194152832, 0.7301587462425232, 0.77182537317276, 0.7539682388305664, 0.7916666865348816, 0.8253968358039856, 0.8392857313156128, 0.9751983880996704], [0.0773809552192688, 0.1160714253783226, 0.1349206417798996, 0.1210317462682724, 0.1091269850730896, 0.125, 0.1329365074634552, 0.1349206417798996, 0.1726190447807312, 0.2093254029750824, 0.2192460298538208, 0.25, 0.2182539701461792, 0.2480158656835556, 0.2757936418056488, 0.2738095223903656, 0.2658730149269104, 0.283730149269104, 0.2638888955116272, 0.3194444477558136, 0.3611111044883728, 0.341269850730896, 0.3670634925365448, 0.3472222089767456, 0.3611111044883728, 0.3253968358039856, 0.3303571343421936, 0.3759920597076416, 0.3928571343421936, 0.4226190447807312, 0.3988095223903656, 0.4355158805847168, 0.3710317313671112, 0.4563491940498352, 0.4226190447807312, 0.3154761791229248, 0.3382936418056488, 0.4285714328289032, 0.4632936418056488, 0.4801587164402008, 0.4583333432674408, 0.4761904776096344, 0.5367063283920288, NaN, 0.545634925365448, 0.4900793731212616, 0.5585317611694336, 0.5, 0.5615079402923584, 0.5952380895614624, 0.5932539701461792, 0.5892857313156128, 0.6498016119003296, 0.591269850730896, 0.6071428656578064, 0.6845238208770752, 0.6547619104385376, 0.6984127163887024, 0.7013888955116272, 0.692460298538208, 0.7063491940498352, 0.6765872836112976, 0.7251983880996704, 0.6746031641960144, 0.7202380895614624, 0.8611111044883728, 0.7638888955116272, 0.7579365372657776, 0.7321428656578064, 0.7996031641960144, 0.8015872836112976, 0.8492063283920288, 0.9682539701461792], [0.0615079365670681, 0.075396828353405, 0.095238097012043, 0.127976194024086, 0.1369047611951828, 0.1408730149269104, 0.1170634925365448, 0.1378968209028244, 0.1636904776096344, 0.1587301641702652, 0.2023809552192688, 0.2698412835597992, 0.2916666567325592, 0.2162698358297348, 0.2738095223903656, 0.2638888955116272, 0.2698412835597992, 0.3174603283405304, 0.3194444477558136, 0.3105158805847168, 0.3343254029750824, 0.3115079402923584, 0.3055555522441864, 0.3492063581943512, 0.3571428656578064, 0.369047611951828, 0.3382936418056488, 0.3253968358039856, 0.363095223903656, 0.3928571343421936, 0.3710317313671112, 0.3601190447807312, 0.3888888955116272, 0.4047619104385376, 0.38591268658638, 0.3948412835597992, 0.380952388048172, 0.4186508059501648, 0.414682537317276, 0.4365079402923584, 0.460317462682724, 0.454365074634552, 0.5019841194152832, 0.454365074634552, NaN, 0.4573412835597992, 0.523809552192688, 0.5615079402923584, 0.6567460298538208, 0.5297619104385376, 0.5476190447807312, 0.533730149269104, 0.716269850730896, 0.5634920597076416, 0.5932539701461792, 0.6269841194152832, 0.7757936716079712, 0.6319444179534912, 0.6150793433189392, 0.6190476417541504, 0.6775793433189392, 0.6547619104385376, 0.6458333134651184, 0.7063491940498352, 0.6726190447807312, 0.6865079402923584, 0.8452380895614624, 0.8928571343421936, 0.851190447807312, 0.7440476417541504, 0.7440476417541504, 0.8809523582458496, 0.9900793433189392], [0.0476190485060215, 0.0833333358168602, 0.0793650820851326, 0.0912698432803154, 0.0833333358168602, 0.0892857164144516, 0.0912698432803154, 0.0892857164144516, 0.1984127014875412, 0.1865079402923584, 0.2142857164144516, 0.2013888955116272, 0.1944444477558136, 0.2400793582201004, 0.2420634925365448, 0.2738095223903656, 0.2539682686328888, 0.2509920597076416, 0.283730149269104, 0.30158731341362, 0.3164682686328888, 0.3571428656578064, 0.3214285671710968, 0.2857142984867096, 0.3591269850730896, 0.3075396716594696, 0.341269850730896, 0.3343254029750824, 0.375, 0.380952388048172, 0.3591269850730896, 0.3878968358039856, 0.3511904776096344, 0.4226190447807312, 0.4047619104385376, 0.4246031641960144, 0.261904776096344, 0.414682537317276, 0.420634925365448, 0.4186508059501648, 0.4365079402923584, 0.4444444477558136, 0.4722222089767456, 0.5099206566810608, 0.5426587462425232, NaN, 0.5, 0.4960317313671112, 0.5476190447807312, 0.636904776096344, 0.6388888955116272, 0.5396825671195984, 0.6190476417541504, 0.5773809552192688, 0.5803571343421936, 0.7142857313156128, 0.64682537317276, 0.648809552192688, 0.7628968358039856, 0.7023809552192688, 0.7202380895614624, 0.6805555820465088, 0.6775793433189392, 0.6944444179534912, 0.7242063283920288, 0.7757936716079712, 0.7390872836112976, 0.726190447807312, 0.7242063283920288, 0.8035714030265808, 0.8134920597076416, 0.817460298538208, 0.966269850730896], [0.0615079365670681, 0.0912698432803154, 0.085317462682724, 0.0932539701461792, 0.095238097012043, 0.1011904776096344, 0.0873015895485878, 0.1071428582072258, 0.1825396865606308, 0.144841268658638, 0.1696428507566452, 0.164682537317276, 0.204365074634552, 0.230158731341362, 0.2380952388048172, 0.2579365074634552, 0.2658730149269104, 0.2460317462682724, 0.3075396716594696, 0.2400793582201004, 0.289682537317276, 0.3472222089767456, 0.329365074634552, 0.318452388048172, 0.3115079402923584, 0.408730149269104, 0.3194444477558136, 0.3541666567325592, 0.3333333432674408, 0.3789682686328888, 0.341269850730896, 0.4007936418056488, 0.420634925365448, 0.335317462682724, 0.4047619104385376, 0.3769841194152832, 0.4126984179019928, 0.408730149269104, 0.443452388048172, 0.4503968358039856, 0.4355158805847168, 0.4642857015132904, 0.4335317313671112, 0.4414682686328888, 0.4761904776096344, 0.5, NaN, 0.5615079402923584, 0.5317460298538208, 0.6269841194152832, 0.6269841194152832, 0.692460298538208, 0.5813491940498352, 0.6220238208770752, 0.670634925365448, 0.6646825671195984, 0.6329365372657776, 0.636904776096344, 0.6547619104385376, 0.6984127163887024, 0.716269850730896, 0.6815476417541504, 0.648809552192688, 0.6964285969734192, 0.7232142686843872, 0.7341269850730896, 0.7083333134651184, 0.7599206566810608, 0.7301587462425232, 0.77182537317276, 0.8144841194152832, 0.8134920597076416, 0.9642857313156128], [0.0694444477558136, 0.0873015895485878, 0.0843253955245018, 0.0912698432803154, 0.0793650820851326, 0.1091269850730896, 0.1388888955116272, 0.115079365670681, 0.1388888955116272, 0.1795634925365448, 0.255952388048172, 0.283730149269104, 0.2906745970249176, 0.1984127014875412, 0.2817460298538208, 0.1884920597076416, 0.3204365074634552, 0.3115079402923584, 0.1567460298538208, 0.3035714328289032, 0.3154761791229248, 0.2539682686328888, 0.3571428656578064, 0.3392857015132904, 0.3095238208770752, 0.3988095223903656, 0.3621031641960144, 0.3591269850730896, 0.32341268658638, 0.3660714328289032, 0.3769841194152832, 0.3452380895614624, 0.3928571343421936, 0.3472222089767456, 0.386904776096344, 0.4384920597076416, 0.4345238208770752, 0.4404761791229248, 0.3878968358039856, 0.3908730149269104, 0.466269850730896, 0.44841268658638, 0.4285714328289032, 0.5, 0.4384920597076416, 0.5039682388305664, 0.4384920597076416, NaN, 0.4503968358039856, 0.5535714030265808, 0.5555555820465088, 0.5059523582458496, 0.5, 0.4841269850730896, 0.5555555820465088, 0.6607142686843872, 0.5396825671195984, 0.648809552192688, 0.6111111044883728, 0.5972222089767456, 0.7083333134651184, 0.6746031641960144, 0.675595223903656, 0.648809552192688, 0.716269850730896, 0.7361111044883728, 0.6765872836112976, 0.6686508059501648, 0.6805555820465088, 0.7896825671195984, 0.8055555820465088, 0.817460298538208, 0.976190447807312], [0.0714285746216774, 0.0773809552192688, 0.1091269850730896, 0.1230158731341362, 0.125, 0.1388888955116272, 0.127976194024086, 0.130952388048172, 0.1408730149269104, 0.1607142835855484, 0.1875, 0.2738095223903656, 0.2599206268787384, 0.2003968209028244, 0.2579365074634552, 0.244047611951828, 0.2579365074634552, 0.3095238208770752, 0.3055555522441864, 0.3055555522441864, 0.3125, 0.2857142984867096, 0.2827380895614624, 0.32341268658638, 0.3035714328289032, 0.369047611951828, 0.3055555522441864, 0.3134920597076416, 0.3253968358039856, 0.341269850730896, 0.3511904776096344, 0.3611111044883728, 0.3650793731212616, 0.3650793731212616, 0.3452380895614624, 0.3888888955116272, 0.3769841194152832, 0.3769841194152832, 0.380952388048172, 0.4285714328289032, 0.4126984179019928, 0.4126984179019928, 0.454365074634552, 0.4384920597076416, 0.3432539701461792, 0.4523809552192688, 0.4682539701461792, 0.5496031641960144, NaN, 0.5327380895614624, 0.5476190447807312, 0.523809552192688, 0.5992063283920288, 0.5, 0.5496031641960144, 0.6230158805847168, 0.6815476417541504, 0.6111111044883728, 0.5952380895614624, 0.6111111044883728, 0.6746031641960144, 0.6309523582458496, 0.6210317611694336, 0.6567460298538208, 0.6527777910232544, 0.6765872836112976, 0.807539701461792, 0.807539701461792, 0.8025793433189392, 0.7341269850730896, 0.7222222089767456, 0.8849206566810608, 0.988095223903656], [0.1746031790971756, 0.1875, 0.2222222238779068, 0.1964285671710968, 0.1964285671710968, 0.2222222238779068, 0.22420634329319, 0.230158731341362, 0.2698412835597992, 0.2351190447807312, 0.2738095223903656, 0.2738095223903656, 0.2757936418056488, 0.30158731341362, 0.2480158656835556, 0.3125, 0.289682537317276, 0.2936508059501648, 0.255952388048172, 0.2906745970249176, 0.3055555522441864, 0.3670634925365448, 0.30158731341362, 0.329365074634552, 0.3323412835597992, 0.3968254029750824, 0.3571428656578064, 0.30158731341362, 0.3194444477558136, 0.3273809552192688, 0.3085317313671112, 0.3253968358039856, 0.3075396716594696, 0.3492063581943512, 0.3432539701461792, 0.3611111044883728, 0.3472222089767456, 0.408730149269104, 0.341269850730896, 0.3392857015132904, 0.3700396716594696, 0.3650793731212616, 0.3591269850730896, 0.4047619104385376, 0.4702380895614624, 0.363095223903656, 0.3730158805847168, 0.4464285671710968, 0.4672619104385376, NaN, 0.5803571343421936, 0.4027777910232544, 0.511904776096344, 0.4444444477558136, 0.454365074634552, 0.52182537317276, 0.5496031641960144, 0.6071428656578064, 0.4722222089767456, 0.4325396716594696, 0.9384920597076416, 0.5892857313156128, 0.6408730149269104, 0.5654761791229248, 0.6071428656578064, 0.6527777910232544, 0.7083333134651184, 0.6865079402923584, 0.7023809552192688, 0.75, 0.6408730149269104, 0.8492063283920288, 0.954365074634552], [0.1746031790971756, 0.1825396865606308, 0.2202380895614624, 0.1984127014875412, 0.1954365074634552, 0.2162698358297348, 0.2142857164144516, 0.2222222238779068, 0.261904776096344, 0.2420634925365448, 0.2658730149269104, 0.2658730149269104, 0.2678571343421936, 0.2996031641960144, 0.244047611951828, 0.3134920597076416, 0.283730149269104, 0.2827380895614624, 0.25, 0.2876984179019928, 0.2996031641960144, 0.3670634925365448, 0.3075396716594696, 0.3214285671710968, 0.3333333432674408, 0.4047619104385376, 0.3561508059501648, 0.2996031641960144, 0.3154761791229248, 0.3273809552192688, 0.3055555522441864, 0.32341268658638, 0.2996031641960144, 0.3432539701461792, 0.3402777910232544, 0.3551587164402008, 0.3382936418056488, 0.4067460298538208, 0.3343254029750824, 0.341269850730896, 0.3769841194152832, 0.3492063581943512, 0.3591269850730896, 0.4067460298538208, 0.4523809552192688, 0.3611111044883728, 0.3730158805847168, 0.4444444477558136, 0.4523809552192688, 0.4196428656578064, NaN, 0.3948412835597992, 0.5059523582458496, 0.4404761791229248, 0.4623015820980072, 0.511904776096344, 0.545634925365448, 0.5952380895614624, 0.4682539701461792, 0.4384920597076416, 0.954365074634552, 0.5833333134651184, 0.636904776096344, 0.5615079402923584, 0.5992063283920288, 0.648809552192688, 0.6944444179534912, 0.6785714030265808, 0.6984127163887024, 0.7420634627342224, 0.6388888955116272, 0.8392857313156128, 0.954365074634552], [0.0535714291036129, 0.0912698432803154, 0.0833333358168602, 0.0972222238779068, 0.0873015895485878, 0.1011904776096344, 0.0813492089509964, 0.1031746044754982, 0.1686507910490036, 0.1289682537317276, 0.1686507910490036, 0.1329365074634552, 0.1805555522441864, 0.2063492089509964, 0.1825396865606308, 0.2202380895614624, 0.2361111044883728, 0.1855158656835556, 0.22420634329319, 0.1587301641702652, 0.1984127014875412, 0.289682537317276, 0.289682537317276, 0.2678571343421936, 0.2936508059501648, 0.3591269850730896, 0.272817462682724, 0.2976190447807312, 0.2976190447807312, 0.3442460298538208, 0.2698412835597992, 0.329365074634552, 0.3650793731212616, 0.2996031641960144, 0.380952388048172, 0.3730158805847168, 0.3928571343421936, 0.4047619104385376, 0.3730158805847168, 0.4067460298538208, 0.4126984179019928, 0.4246031641960144, 0.244047611951828, 0.4107142984867096, 0.466269850730896, 0.460317462682724, 0.3075396716594696, 0.494047611951828, 0.4761904776096344, 0.5972222089767456, 0.6051587462425232, NaN, 0.5476190447807312, 0.550595223903656, 0.58432537317276, 0.5773809552192688, 0.5753968358039856, 0.6210317611694336, 0.5932539701461792, 0.6567460298538208, 0.6785714030265808, 0.6349206566810608, 0.636904776096344, 0.6408730149269104, 0.6666666865348816, 0.6626983880996704, 0.6438491940498352, 0.6150793433189392, 0.6785714030265808, 0.7658730149269104, 0.7539682388305664, 0.8253968358039856, 0.9702380895614624], [0.0436507947742939, 0.0793650820851326, 0.0773809552192688, 0.085317462682724, 0.1130952388048172, 0.0972222238779068, 0.0992063507437706, 0.1071428582072258, 0.125, 0.1031746044754982, 0.1130952388048172, 0.210317462682724, 0.1924603134393692, 0.1428571492433548, 0.2023809552192688, 0.2003968209028244, 0.1924603134393692, 0.2261904776096344, 0.2777777910232544, 0.2023809552192688, 0.2480158656835556, 0.2876984179019928, 0.1984127014875412, 0.2519841194152832, 0.2420634925365448, 0.300595223903656, 0.230158731341362, 0.2083333283662796, 0.3194444477558136, 0.2212301641702652, 0.2539682686328888, 0.2916666567325592, 0.289682537317276, 0.3253968358039856, 0.2420634925365448, 0.2926587164402008, 0.3035714328289032, 0.335317462682724, 0.329365074634552, 0.3432539701461792, 0.3472222089767456, 0.3531745970249176, 0.3948412835597992, 0.3501984179019928, 0.283730149269104, 0.380952388048172, 0.4186508059501648, 0.5, 0.4007936418056488, 0.488095223903656, 0.494047611951828, 0.4523809552192688, NaN, 0.4821428656578064, 0.5059523582458496, 0.5535714030265808, 0.66567462682724, 0.5753968358039856, 0.5476190447807312, 0.5079365372657776, 0.6240079402923584, 0.579365074634552, 0.5813491940498352, 0.6726190447807312, 0.6111111044883728, 0.613095223903656, 0.7321428656578064, 0.7757936716079712, 0.8432539701461792, 0.6934523582458496, 0.6746031641960144, 0.8373016119003296, 0.9682539701461792], [0.0416666679084301, 0.0416666679084301, 0.0535714291036129, 0.0734127014875412, 0.0694444477558136, 0.0595238097012043, 0.0456349216401577, 0.065476194024086, 0.1329365074634552, 0.1011904776096344, 0.085317462682724, 0.1190476194024086, 0.1230158731341362, 0.1567460298538208, 0.1924603134393692, 0.2083333283662796, 0.1666666716337204, 0.1686507910490036, 0.261904776096344, 0.1269841343164444, 0.1557539701461792, 0.3154761791229248, 0.1468254029750824, 0.244047611951828, 0.2380952388048172, 0.329365074634552, 0.1944444477558136, 0.1527777761220932, 0.3392857015132904, 0.1269841343164444, 0.2341269850730896, 0.2926587164402008, 0.3333333432674408, 0.3670634925365448, 0.164682537317276, 0.335317462682724, 0.3055555522441864, 0.2817460298538208, 0.335317462682724, 0.3968254029750824, 0.363095223903656, 0.3849206268787384, 0.3611111044883728, 0.408730149269104, 0.4365079402923584, 0.4226190447807312, 0.3779761791229248, 0.5158730149269104, 0.5, 0.5555555820465088, 0.5595238208770752, 0.449404776096344, 0.5178571343421936, NaN, 0.5565476417541504, 0.601190447807312, 0.5873016119003296, 0.5932539701461792, 0.5932539701461792, 0.6507936716079712, 0.6646825671195984, 0.5714285969734192, 0.6150793433189392, 0.6785714030265808, 0.6210317611694336, 0.6666666865348816, 0.716269850730896, 0.761904776096344, 0.7757936716079712, 0.7361111044883728, 0.7222222089767456, 0.7916666865348816, 0.9464285969734192], [0.0456349216401577, 0.0595238097012043, 0.0535714291036129, 0.0634920671582222, 0.0555555559694767, 0.0634920671582222, 0.0555555559694767, 0.0634920671582222, 0.1269841343164444, 0.0873015895485878, 0.0992063507437706, 0.1160714253783226, 0.0992063507437706, 0.1488095223903656, 0.181547611951828, 0.184523805975914, 0.1626984179019928, 0.1825396865606308, 0.1894841343164444, 0.0575396828353405, 0.092261902987957, 0.261904776096344, 0.144841268658638, 0.2380952388048172, 0.2361111044883728, 0.3115079402923584, 0.1587301641702652, 0.170634925365448, 0.2876984179019928, 0.1269841343164444, 0.1964285671710968, 0.2857142984867096, 0.3452380895614624, 0.2996031641960144, 0.1626984179019928, 0.3263888955116272, 0.3194444477558136, 0.3134920597076416, 0.3273809552192688, 0.3650793731212616, 0.3531745970249176, 0.3551587164402008, 0.2946428656578064, 0.3928571343421936, 0.4067460298538208, 0.4196428656578064, 0.329365074634552, 0.4444444477558136, 0.4503968358039856, 0.545634925365448, 0.5376983880996704, 0.4156745970249176, 0.494047611951828, 0.443452388048172, NaN, 0.5714285969734192, 0.538690447807312, 0.5714285969734192, 0.5833333134651184, 0.6150793433189392, 0.6408730149269104, 0.5654761791229248, 0.5932539701461792, 0.636904776096344, 0.6230158805847168, 0.64682537317276, 0.704365074634552, 0.6845238208770752, 0.7033730149269104, 0.7341269850730896, 0.7321428656578064, 0.7996031641960144, 0.9632936716079712], [0.0734127014875412, 0.0833333358168602, 0.1031746044754982, 0.105158731341362, 0.0863095223903656, 0.1091269850730896, 0.105158731341362, 0.1190476194024086, 0.0912698432803154, 0.1884920597076416, 0.204365074634552, 0.2023809552192688, 0.190476194024086, 0.170634925365448, 0.2013888955116272, 0.1746031790971756, 0.2212301641702652, 0.2162698358297348, 0.1111111119389534, 0.2539682686328888, 0.2658730149269104, 0.1587301641702652, 0.3095238208770752, 0.2341269850730896, 0.2162698358297348, 0.2876984179019928, 0.230158731341362, 0.3174603283405304, 0.1984127014875412, 0.3085317313671112, 0.295634925365448, 0.261904776096344, 0.3055555522441864, 0.255952388048172, 0.3194444477558136, 0.2460317462682724, 0.261904776096344, 0.3373015820980072, 0.2886904776096344, 0.2817460298538208, 0.3323412835597992, 0.2648809552192688, 0.30158731341362, 0.3154761791229248, 0.3730158805847168, 0.2857142984867096, 0.335317462682724, 0.3392857015132904, 0.3769841194152832, 0.4781745970249176, 0.488095223903656, 0.4226190447807312, 0.4464285671710968, 0.3988095223903656, 0.4285714328289032, NaN, 0.4642857015132904, 0.5376983880996704, 0.4652777910232544, 0.4801587164402008, 0.6269841194152832, 0.5277777910232544, 0.5684523582458496, 0.5277777910232544, 0.5853174328804016, 0.6289682388305664, 0.6071428656578064, 0.5555555820465088, 0.5992063283920288, 0.6884920597076416, 0.6349206566810608, 0.7539682388305664, 0.942460298538208], [0.0456349216401577, 0.0575396828353405, 0.0734127014875412, 0.0873015895485878, 0.065476194024086, 0.085317462682724, 0.0773809552192688, 0.0912698432803154, 0.1210317462682724, 0.0892857164144516, 0.0813492089509964, 0.1587301641702652, 0.1517857164144516, 0.1289682537317276, 0.1924603134393692, 0.1746031790971756, 0.1686507910490036, 0.2083333283662796, 0.2182539701461792, 0.1428571492433548, 0.1785714328289032, 0.2361111044883728, 0.141865074634552, 0.2261904776096344, 0.1964285671710968, 0.2886904776096344, 0.1884920597076416, 0.1666666716337204, 0.2599206268787384, 0.1914682537317276, 0.2123015820980072, 0.2738095223903656, 0.2757936418056488, 0.2599206268787384, 0.1944444477558136, 0.283730149269104, 0.283730149269104, 0.3273809552192688, 0.289682537317276, 0.3174603283405304, 0.2976190447807312, 0.3194444477558136, 0.341269850730896, 0.3452380895614624, 0.22420634329319, 0.3531745970249176, 0.3670634925365448, 0.460317462682724, 0.318452388048172, 0.4503968358039856, 0.454365074634552, 0.4246031641960144, 0.3343254029750824, 0.4126984179019928, 0.4613095223903656, 0.5357142686843872, NaN, 0.557539701461792, 0.5, 0.4781745970249176, 0.5972222089767456, 0.5555555820465088, 0.5654761791229248, 0.6230158805847168, 0.5952380895614624, 0.6091269850730896, 0.699404776096344, 0.704365074634552, 0.7291666865348816, 0.6845238208770752, 0.6547619104385376, 0.8253968358039856, 0.9702380895614624], [0.0634920671582222, 0.095238097012043, 0.1170634925365448, 0.0813492089509964, 0.0992063507437706, 0.1011904776096344, 0.115079365670681, 0.115079365670681, 0.1587301641702652, 0.1388888955116272, 0.1527777761220932, 0.144841268658638, 0.1577380895614624, 0.2142857164144516, 0.1210317462682724, 0.2251984179019928, 0.2123015820980072, 0.1607142835855484, 0.2321428507566452, 0.1884920597076416, 0.1964285671710968, 0.3055555522441864, 0.2261904776096344, 0.255952388048172, 0.2083333283662796, 0.2480158656835556, 0.295634925365448, 0.2321428507566452, 0.2083333283662796, 0.2420634925365448, 0.2579365074634552, 0.2003968209028244, 0.2361111044883728, 0.255952388048172, 0.255952388048172, 0.25, 0.2569444477558136, 0.3571428656578064, 0.2341269850730896, 0.2519841194152832, 0.3680555522441864, 0.289682537317276, 0.3452380895614624, 0.30158731341362, 0.3680555522441864, 0.3511904776096344, 0.363095223903656, 0.3511904776096344, 0.3888888955116272, 0.3928571343421936, 0.4047619104385376, 0.3789682686328888, 0.4246031641960144, 0.4067460298538208, 0.4285714328289032, 0.4623015820980072, 0.4424603283405304, NaN, 0.4345238208770752, 0.4384920597076416, 0.5496031641960144, 0.4781745970249176, 0.6696428656578064, 0.4960317313671112, 0.4980158805847168, 0.488095223903656, 0.591269850730896, 0.5555555820465088, 0.5396825671195984, 0.7857142686843872, 0.5714285969734192, 0.716269850730896, 0.9047619104385376], [0.0396825410425663, 0.0496031753718853, 0.0496031753718853, 0.065476194024086, 0.0376984141767025, 0.0615079365670681, 0.0595238097012043, 0.0535714291036129, 0.115079365670681, 0.1111111119389534, 0.0992063507437706, 0.1071428582072258, 0.1230158731341362, 0.130952388048172, 0.1408730149269104, 0.1488095223903656, 0.1329365074634552, 0.125, 0.1587301641702652, 0.1944444477558136, 0.2083333283662796, 0.2261904776096344, 0.1984127014875412, 0.204365074634552, 0.1765872985124588, 0.2480158656835556, 0.2003968209028244, 0.2202380895614624, 0.1805555522441864, 0.244047611951828, 0.2123015820980072, 0.2579365074634552, 0.2182539701461792, 0.2162698358297348, 0.2678571343421936, 0.2251984179019928, 0.1468254029750824, 0.30158731341362, 0.283730149269104, 0.2916666567325592, 0.2361111044883728, 0.2509920597076416, 0.3373015820980072, 0.2986111044883728, 0.3849206268787384, 0.2371031790971756, 0.3452380895614624, 0.3888888955116272, 0.4047619104385376, 0.5277777910232544, 0.5317460298538208, 0.4067460298538208, 0.4523809552192688, 0.4067460298538208, 0.4166666567325592, 0.5347222089767456, 0.5, 0.5654761791229248, NaN, 0.4821428656578064, 0.64682537317276, 0.5317460298538208, 0.5932539701461792, 0.5734127163887024, 0.5813491940498352, 0.6686508059501648, 0.6626983880996704, 0.6269841194152832, 0.6150793433189392, 0.7083333134651184, 0.6726190447807312, 0.7876983880996704, 0.9484127163887024], [0.0694444477558136, 0.0615079365670681, 0.0674603208899498, 0.0833333358168602, 0.0674603208899498, 0.0773809552192688, 0.0773809552192688, 0.0813492089509964, 0.1369047611951828, 0.1071428582072258, 0.1091269850730896, 0.1190476194024086, 0.1349206417798996, 0.1428571492433548, 0.1011904776096344, 0.1686507910490036, 0.1527777761220932, 0.1349206417798996, 0.1884920597076416, 0.1269841343164444, 0.1438492089509964, 0.2023809552192688, 0.1716269850730896, 0.1865079402923584, 0.2182539701461792, 0.2797619104385376, 0.184523805975914, 0.1884920597076416, 0.204365074634552, 0.230158731341362, 0.1369047611951828, 0.1924603134393692, 0.2281745970249176, 0.2162698358297348, 0.2341269850730896, 0.2579365074634552, 0.2212301641702652, 0.30158731341362, 0.2361111044883728, 0.2361111044883728, 0.2569444477558136, 0.2658730149269104, 0.2698412835597992, 0.3075396716594696, 0.380952388048172, 0.2976190447807312, 0.30158731341362, 0.4027777910232544, 0.3888888955116272, 0.567460298538208, 0.5615079402923584, 0.3432539701461792, 0.4920634925365448, 0.3492063581943512, 0.3849206268787384, 0.5198412537574768, 0.52182537317276, 0.5615079402923584, 0.5178571343421936, NaN, 0.6349206566810608, 0.523809552192688, 0.5833333134651184, 0.5555555820465088, 0.545634925365448, 0.5813491940498352, 0.670634925365448, 0.5932539701461792, 0.658730149269104, 0.6865079402923584, 0.6309523582458496, 0.7916666865348816, 0.9246031641960144], [0.1289682537317276, 0.1468254029750824, 0.1746031790971756, 0.144841268658638, 0.1626984179019928, 0.184523805975914, 0.184523805975914, 0.184523805975914, 0.2003968209028244, 0.1894841343164444, 0.2202380895614624, 0.2162698358297348, 0.2182539701461792, 0.2281745970249176, 0.1875, 0.2341269850730896, 0.2341269850730896, 0.2361111044883728, 0.190476194024086, 0.210317462682724, 0.2261904776096344, 0.261904776096344, 0.2480158656835556, 0.2460317462682724, 0.255952388048172, 0.2698412835597992, 0.2519841194152832, 0.2420634925365448, 0.25, 0.2658730149269104, 0.2380952388048172, 0.2400793582201004, 0.230158731341362, 0.2658730149269104, 0.2698412835597992, 0.255952388048172, 0.25, 0.3452380895614624, 0.2599206268787384, 0.2638888955116272, 0.2936508059501648, 0.2757936418056488, 0.2698412835597992, 0.2936508059501648, 0.3224206268787384, 0.2797619104385376, 0.283730149269104, 0.2916666567325592, 0.3253968358039856, 0.0615079365670681, 0.0456349216401577, 0.3214285671710968, 0.3759920597076416, 0.335317462682724, 0.3591269850730896, 0.3730158805847168, 0.4027777910232544, 0.4503968358039856, 0.3531745970249176, 0.3650793731212616, NaN, 0.5019841194152832, 0.466269850730896, 0.454365074634552, 0.5138888955116272, 0.4285714328289032, 0.5654761791229248, 0.523809552192688, 0.4801587164402008, 0.6111111044883728, 0.5515872836112976, 0.7063491940498352, 0.942460298538208], [0.01785714365541935, 0.0634920671582222, 0.0555555559694767, 0.0575396828353405, 0.0595238097012043, 0.0575396828353405, 0.0674603208899498, 0.0744047611951828, 0.115079365670681, 0.095238097012043, 0.0773809552192688, 0.1071428582072258, 0.085317462682724, 0.1329365074634552, 0.144841268658638, 0.1547619104385376, 0.130952388048172, 0.1091269850730896, 0.1289682537317276, 0.1230158731341362, 0.1378968209028244, 0.2063492089509964, 0.1825396865606308, 0.190476194024086, 0.184523805975914, 0.2380952388048172, 0.170634925365448, 0.1944444477558136, 0.2172619104385376, 0.22420634329319, 0.2083333283662796, 0.2142857164144516, 0.2430555522441864, 0.2261904776096344, 0.2519841194152832, 0.255952388048172, 0.22420634329319, 0.2916666567325592, 0.2380952388048172, 0.2599206268787384, 0.295634925365448, 0.2539682686328888, 0.2738095223903656, 0.32341268658638, 0.3452380895614624, 0.3194444477558136, 0.318452388048172, 0.3253968358039856, 0.369047611951828, 0.4107142984867096, 0.4166666567325592, 0.3650793731212616, 0.420634925365448, 0.4285714328289032, 0.4345238208770752, 0.4722222089767456, 0.4444444477558136, 0.52182537317276, 0.4682539701461792, 0.4761904776096344, 0.4980158805847168, NaN, 0.5396825671195984, 0.5317460298538208, 0.6875, 0.52182537317276, 0.5585317611694336, 0.5892857313156128, 0.5873016119003296, 0.6210317611694336, 0.7757936716079712, 0.6944444179534912, 0.9404761791229248], [0.0555555559694767, 0.0873015895485878, 0.0912698432803154, 0.075396828353405, 0.0912698432803154, 0.085317462682724, 0.0912698432803154, 0.0992063507437706, 0.1488095223903656, 0.1170634925365448, 0.15079365670681, 0.127976194024086, 0.1299603134393692, 0.1924603134393692, 0.1011904776096344, 0.210317462682724, 0.2033730149269104, 0.1488095223903656, 0.2063492089509964, 0.1666666716337204, 0.1736111044883728, 0.2738095223903656, 0.2083333283662796, 0.2261904776096344, 0.1924603134393692, 0.244047611951828, 0.255952388048172, 0.2172619104385376, 0.1865079402923584, 0.2281745970249176, 0.2341269850730896, 0.1825396865606308, 0.2152777761220932, 0.2311507910490036, 0.230158731341362, 0.2182539701461792, 0.2162698358297348, 0.341269850730896, 0.2142857164144516, 0.2172619104385376, 0.3253968358039856, 0.2480158656835556, 0.3214285671710968, 0.2748015820980072, 0.3541666567325592, 0.3224206268787384, 0.3511904776096344, 0.324404776096344, 0.3789682686328888, 0.3591269850730896, 0.363095223903656, 0.363095223903656, 0.4186508059501648, 0.3849206268787384, 0.4067460298538208, 0.431547611951828, 0.4345238208770752, 0.3303571343421936, 0.4067460298538208, 0.4166666567325592, 0.533730149269104, 0.460317462682724, NaN, 0.4920634925365448, 0.4841269850730896, 0.4702380895614624, 0.5565476417541504, 0.5396825671195984, 0.511904776096344, 0.6984127163887024, 0.5654761791229248, 0.6964285969734192, 0.89682537317276], [0.0357142873108387, 0.0317460335791111, 0.0535714291036129, 0.0595238097012043, 0.0515873022377491, 0.0595238097012043, 0.0595238097012043, 0.0575396828353405, 0.1111111119389534, 0.065476194024086, 0.085317462682724, 0.1408730149269104, 0.1230158731341362, 0.102182537317276, 0.210317462682724, 0.1388888955116272, 0.1388888955116272, 0.184523805975914, 0.1944444477558136, 0.1964285671710968, 0.2033730149269104, 0.1964285671710968, 0.1785714328289032, 0.144841268658638, 0.0694444477558136, 0.2380952388048172, 0.1428571492433548, 0.1924603134393692, 0.1111111119389534, 0.2063492089509964, 0.2341269850730896, 0.2460317462682724, 0.2857142984867096, 0.2678571343421936, 0.2142857164144516, 0.2539682686328888, 0.2361111044883728, 0.193452388048172, 0.2718254029750824, 0.3075396716594696, 0.2460317462682724, 0.295634925365448, 0.2757936418056488, 0.3253968358039856, 0.2936508059501648, 0.3055555522441864, 0.3035714328289032, 0.3511904776096344, 0.3432539701461792, 0.4345238208770752, 0.4384920597076416, 0.3591269850730896, 0.3273809552192688, 0.3214285671710968, 0.363095223903656, 0.4722222089767456, 0.3769841194152832, 0.5039682388305664, 0.42658731341362, 0.4444444477558136, 0.545634925365448, 0.4682539701461792, 0.5079365372657776, NaN, 0.4950396716594696, 0.5059523582458496, 0.5059523582458496, 0.494047611951828, 0.533730149269104, 0.6051587462425232, 0.5873016119003296, 0.6527777910232544, 0.9047619104385376], [0.02182539738714695, 0.0496031753718853, 0.0505952388048172, 0.0515873022377491, 0.0535714291036129, 0.0575396828353405, 0.0714285746216774, 0.0634920671582222, 0.095238097012043, 0.085317462682724, 0.0714285746216774, 0.0873015895485878, 0.0793650820851326, 0.1180555522441864, 0.1230158731341362, 0.1428571492433548, 0.1289682537317276, 0.1011904776096344, 0.1289682537317276, 0.0992063507437706, 0.1190476194024086, 0.1825396865606308, 0.1626984179019928, 0.1686507910490036, 0.1517857164144516, 0.233134925365448, 0.144841268658638, 0.170634925365448, 0.1696428507566452, 0.2013888955116272, 0.1666666716337204, 0.1914682537317276, 0.2361111044883728, 0.1944444477558136, 0.210317462682724, 0.22420634329319, 0.1805555522441864, 0.289682537317276, 0.2162698358297348, 0.2400793582201004, 0.2460317462682724, 0.2023809552192688, 0.230158731341362, 0.2797619104385376, 0.3273809552192688, 0.2757936418056488, 0.2767857015132904, 0.283730149269104, 0.3472222089767456, 0.3928571343421936, 0.4007936418056488, 0.3333333432674408, 0.3888888955116272, 0.3789682686328888, 0.3769841194152832, 0.414682537317276, 0.4047619104385376, 0.5019841194152832, 0.4186508059501648, 0.454365074634552, 0.4861111044883728, 0.3125, 0.5158730149269104, 0.504960298538208, NaN, 0.4920634925365448, 0.5257936716079712, 0.5476190447807312, 0.5555555820465088, 0.5992063283920288, 0.6845238208770752, 0.6726190447807312, 0.9146825671195984], [0.0337301604449749, 0.0416666679084301, 0.0555555559694767, 0.0496031753718853, 0.0535714291036129, 0.0595238097012043, 0.0634920671582222, 0.0674603208899498, 0.0773809552192688, 0.0793650820851326, 0.0714285746216774, 0.0992063507437706, 0.0833333358168602, 0.0992063507437706, 0.1408730149269104, 0.1130952388048172, 0.1001984104514122, 0.144841268658638, 0.1130952388048172, 0.1626984179019928, 0.170634925365448, 0.1349206417798996, 0.1924603134393692, 0.1795634925365448, 0.1170634925365448, 0.2182539701461792, 0.1626984179019928, 0.2023809552192688, 0.1388888955116272, 0.2281745970249176, 0.1865079402923584, 0.1805555522441864, 0.2083333283662796, 0.190476194024086, 0.2321428507566452, 0.0515873022377491, 0.1537698358297348, 0.2757936418056488, 0.2123015820980072, 0.204365074634552, 0.2142857164144516, 0.1527777761220932, 0.2480158656835556, 0.1388888955116272, 0.3134920597076416, 0.22420634329319, 0.2658730149269104, 0.2638888955116272, 0.32341268658638, 0.3472222089767456, 0.3511904776096344, 0.3373015820980072, 0.386904776096344, 0.3333333432674408, 0.3531745970249176, 0.3710317313671112, 0.3908730149269104, 0.511904776096344, 0.3313491940498352, 0.4186508059501648, 0.5714285969734192, 0.4781745970249176, 0.5297619104385376, 0.494047611951828, 0.5079365372657776, NaN, 0.5753968358039856, 0.5019841194152832, 0.5228174328804016, 0.6507936716079712, 0.6150793433189392, 0.7440476417541504, 0.9484127163887024], [0.0357142873108387, 0.0575396828353405, 0.0674603208899498, 0.0575396828353405, 0.0972222238779068, 0.0833333358168602, 0.0674603208899498, 0.0833333358168602, 0.0813492089509964, 0.105158731341362, 0.105158731341362, 0.1269841343164444, 0.125, 0.0982142835855484, 0.1210317462682724, 0.1011904776096344, 0.15079365670681, 0.1527777761220932, 0.1369047611951828, 0.1269841343164444, 0.144841268658638, 0.125, 0.1369047611951828, 0.1805555522441864, 0.1319444477558136, 0.283730149269104, 0.1785714328289032, 0.1468254029750824, 0.1468254029750824, 0.1825396865606308, 0.1567460298538208, 0.170634925365448, 0.1944444477558136, 0.1527777761220932, 0.181547611951828, 0.2083333283662796, 0.2142857164144516, 0.2638888955116272, 0.1984127014875412, 0.1686507910490036, 0.255952388048172, 0.2023809552192688, 0.2698412835597992, 0.2361111044883728, 0.1547619104385376, 0.26091268658638, 0.2916666567325592, 0.32341268658638, 0.1924603134393692, 0.2916666567325592, 0.3055555522441864, 0.3561508059501648, 0.2678571343421936, 0.283730149269104, 0.295634925365448, 0.3928571343421936, 0.300595223903656, 0.408730149269104, 0.3373015820980072, 0.329365074634552, 0.4345238208770752, 0.4414682686328888, 0.443452388048172, 0.494047611951828, 0.4742063581943512, 0.4246031641960144, NaN, 0.454365074634552, 0.454365074634552, 0.5734127163887024, 0.5416666865348816, 0.846230149269104, 0.9702380895614624], [0.011904762126505375, 0.02579365111887455, 0.0456349216401577, 0.0496031753718853, 0.0476190485060215, 0.0496031753718853, 0.0476190485060215, 0.0555555559694767, 0.0466269850730896, 0.0535714291036129, 0.0535714291036129, 0.0932539701461792, 0.075396828353405, 0.0615079365670681, 0.1468254029750824, 0.095238097012043, 0.1349206417798996, 0.1408730149269104, 0.1547619104385376, 0.0813492089509964, 0.095238097012043, 0.1091269850730896, 0.1091269850730896, 0.1349206417798996, 0.0674603208899498, 0.2341269850730896, 0.105158731341362, 0.1210317462682724, 0.125, 0.1329365074634552, 0.1388888955116272, 0.1666666716337204, 0.190476194024086, 0.181547611951828, 0.130952388048172, 0.1924603134393692, 0.230158731341362, 0.204365074634552, 0.1726190447807312, 0.1865079402923584, 0.2162698358297348, 0.2123015820980072, 0.2281745970249176, 0.2420634925365448, 0.1071428582072258, 0.2738095223903656, 0.2400793582201004, 0.3313491940498352, 0.1924603134393692, 0.3134920597076416, 0.3214285671710968, 0.3849206268787384, 0.22420634329319, 0.2380952388048172, 0.3154761791229248, 0.4444444477558136, 0.295634925365448, 0.4444444477558136, 0.3730158805847168, 0.4067460298538208, 0.4761904776096344, 0.4107142984867096, 0.460317462682724, 0.5059523582458496, 0.4523809552192688, 0.4980158805847168, 0.545634925365448, NaN, 0.5297619104385376, 0.601190447807312, 0.511904776096344, 0.7023809552192688, 0.9623016119003296], [0.01984127052128315, 0.02182539738714695, 0.0436507947742939, 0.0396825410425663, 0.0456349216401577, 0.0515873022377491, 0.0337301604449749, 0.0535714291036129, 0.0476190485060215, 0.0456349216401577, 0.0416666679084301, 0.0992063507437706, 0.085317462682724, 0.0694444477558136, 0.1210317462682724, 0.0912698432803154, 0.1091269850730896, 0.125, 0.1170634925365448, 0.0714285746216774, 0.0972222238779068, 0.1408730149269104, 0.085317462682724, 0.1527777761220932, 0.0634920671582222, 0.2222222238779068, 0.105158731341362, 0.095238097012043, 0.1230158731341362, 0.0972222238779068, 0.130952388048172, 0.164682537317276, 0.170634925365448, 0.184523805975914, 0.1130952388048172, 0.2182539701461792, 0.2321428507566452, 0.2182539701461792, 0.1825396865606308, 0.204365074634552, 0.2142857164144516, 0.2162698358297348, 0.2460317462682724, 0.2678571343421936, 0.1488095223903656, 0.2757936418056488, 0.2698412835597992, 0.3194444477558136, 0.1974206417798996, 0.2976190447807312, 0.30158731341362, 0.3214285671710968, 0.1567460298538208, 0.22420634329319, 0.2966269850730896, 0.4007936418056488, 0.2708333432674408, 0.460317462682724, 0.3849206268787384, 0.341269850730896, 0.5198412537574768, 0.4126984179019928, 0.488095223903656, 0.466269850730896, 0.4444444477558136, 0.477182537317276, 0.545634925365448, 0.4702380895614624, NaN, 0.60317462682724, 0.5198412537574768, 0.7103174328804016, 0.9484127163887024], [0.0396825410425663, 0.0515873022377491, 0.0644841268658638, 0.0486111119389534, 0.0476190485060215, 0.0744047611951828, 0.0634920671582222, 0.0664682537317276, 0.0694444477558136, 0.1091269850730896, 0.1289682537317276, 0.1011904776096344, 0.095238097012043, 0.1349206417798996, 0.0674603208899498, 0.1388888955116272, 0.1607142835855484, 0.1190476194024086, 0.1001984104514122, 0.1031746044754982, 0.1170634925365448, 0.1547619104385376, 0.1686507910490036, 0.1785714328289032, 0.1537698358297348, 0.2182539701461792, 0.1924603134393692, 0.1765872985124588, 0.1329365074634552, 0.1785714328289032, 0.1527777761220932, 0.1488095223903656, 0.170634925365448, 0.1408730149269104, 0.1865079402923584, 0.1626984179019928, 0.1567460298538208, 0.2976190447807312, 0.1746031790971756, 0.1428571492433548, 0.244047611951828, 0.1666666716337204, 0.2083333283662796, 0.2003968209028244, 0.255952388048172, 0.1964285671710968, 0.2281745970249176, 0.210317462682724, 0.2658730149269104, 0.25, 0.2579365074634552, 0.2341269850730896, 0.306547611951828, 0.2638888955116272, 0.2658730149269104, 0.3115079402923584, 0.3154761791229248, 0.2142857164144516, 0.2916666567325592, 0.3134920597076416, 0.3888888955116272, 0.3789682686328888, 0.30158731341362, 0.3948412835597992, 0.4007936418056488, 0.3492063581943512, 0.42658731341362, 0.3988095223903656, 0.3968254029750824, NaN, 0.477182537317276, 0.6051587462425232, 0.8492063283920288], [0.01984127052128315, 0.0357142873108387, 0.0416666679084301, 0.02777777798473835, 0.0337301604449749, 0.0416666679084301, 0.0436507947742939, 0.0416666679084301, 0.0456349216401577, 0.0634920671582222, 0.0396825410425663, 0.0436507947742939, 0.0615079365670681, 0.0694444477558136, 0.0694444477558136, 0.0714285746216774, 0.0575396828353405, 0.0535714291036129, 0.0813492089509964, 0.065476194024086, 0.0674603208899498, 0.1230158731341362, 0.1091269850730896, 0.1170634925365448, 0.0892857164144516, 0.2023809552192688, 0.1031746044754982, 0.1011904776096344, 0.1011904776096344, 0.1190476194024086, 0.1170634925365448, 0.1031746044754982, 0.1408730149269104, 0.1428571492433548, 0.1210317462682724, 0.1349206417798996, 0.1190476194024086, 0.210317462682724, 0.125, 0.1488095223903656, 0.1726190447807312, 0.125, 0.1746031790971756, 0.1984127014875412, 0.255952388048172, 0.1865079402923584, 0.1855158656835556, 0.1944444477558136, 0.2777777910232544, 0.3591269850730896, 0.3611111044883728, 0.2460317462682724, 0.3253968358039856, 0.2777777910232544, 0.2678571343421936, 0.3650793731212616, 0.3452380895614624, 0.4285714328289032, 0.3273809552192688, 0.369047611951828, 0.44841268658638, 0.22420634329319, 0.4345238208770752, 0.4126984179019928, 0.3154761791229248, 0.3849206268787384, 0.4583333432674408, 0.488095223903656, 0.4801587164402008, 0.5228174328804016, NaN, 0.6190476417541504, 0.85317462682724], [0.0436507947742939, 0.0476190485060215, 0.0535714291036129, 0.0535714291036129, 0.0634920671582222, 0.0734127014875412, 0.0575396828353405, 0.0674603208899498, 0.0575396828353405, 0.085317462682724, 0.0793650820851326, 0.0892857164144516, 0.085317462682724, 0.0873015895485878, 0.0515873022377491, 0.0972222238779068, 0.1190476194024086, 0.1111111119389534, 0.0615079365670681, 0.082341268658638, 0.0932539701461792, 0.0992063507437706, 0.1031746044754982, 0.1329365074634552, 0.1011904776096344, 0.1746031790971756, 0.1339285671710968, 0.0972222238779068, 0.1011904776096344, 0.122023805975914, 0.0912698432803154, 0.1061507910490036, 0.1071428582072258, 0.1210317462682724, 0.125, 0.1210317462682724, 0.1269841343164444, 0.230158731341362, 0.105158731341362, 0.0873015895485878, 0.181547611951828, 0.1329365074634552, 0.1607142835855484, 0.15079365670681, 0.1190476194024086, 0.1825396865606308, 0.1865079402923584, 0.1825396865606308, 0.115079365670681, 0.15079365670681, 0.1607142835855484, 0.1746031790971756, 0.1626984179019928, 0.2083333283662796, 0.2003968209028244, 0.2460317462682724, 0.1746031790971756, 0.283730149269104, 0.2123015820980072, 0.2083333283662796, 0.2936508059501648, 0.3055555522441864, 0.3035714328289032, 0.3472222089767456, 0.3273809552192688, 0.255952388048172, 0.1537698358297348, 0.2976190447807312, 0.289682537317276, 0.3948412835597992, 0.380952388048172, NaN, 0.9365079402923584], [0.0059523810632526875, 0.003968254197388887, 0.0059523810632526875, 0.003968254197388887, 0.0059523810632526875, 0.007936508394777775, 0.007936508394777775, 0.007936508394777775, 0.0059523810632526875, 0.009920635260641575, 0.0059523810632526875, 0.01587301678955555, 0.003968254197388887, 0.0059523810632526875, 0.013888888992369175, 0.009920635260641575, 0.01785714365541935, 0.02380952425301075, 0.0059523810632526875, 0.007936508394777775, 0.0059523810632526875, 0.007936508394777775, 0.007936508394777775, 0.01984127052128315, 0.009920635260641575, 0.01785714365541935, 0.02182539738714695, 0.007936508394777775, 0.009920635260641575, 0.01785714365541935, 0.011904762126505375, 0.01984127052128315, 0.02777777798473835, 0.01587301678955555, 0.02182539738714695, 0.02380952425301075, 0.02380952425301075, 0.0595238097012043, 0.01984127052128315, 0.02182539738714695, 0.0357142873108387, 0.02182539738714695, 0.02480158768594265, 0.0317460335791111, 0.009920635260641575, 0.0337301604449749, 0.0357142873108387, 0.02380952425301075, 0.011904762126505375, 0.0456349216401577, 0.0456349216401577, 0.02976190485060215, 0.0317460335791111, 0.0535714291036129, 0.0367063507437706, 0.0575396828353405, 0.02976190485060215, 0.095238097012043, 0.0515873022377491, 0.075396828353405, 0.0575396828353405, 0.0595238097012043, 0.1031746044754982, 0.095238097012043, 0.085317462682724, 0.0515873022377491, 0.02976190485060215, 0.0376984141767025, 0.0515873022377491, 0.15079365670681, 0.1468254029750824, 0.0634920671582222, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9066771376},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8384589925},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8279596571},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8235918192},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8160411192},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8141672154},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8055831099},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8047426117},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7318397266},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7306272062},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7098214266},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.680321319},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6769179875},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6760085981},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6531360243},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.650779869},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6391782396},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6334738748},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6236634722},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6190889564},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5949487452},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5907049148},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5790619496},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5782903447},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5782627876},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5710703279},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.568355929},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5663304696},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5554591052},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5373952814},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5357831777},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5340746211},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5276813258},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.52345128},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5217565069},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5217289481},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5178020279},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5032379839},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4963624343},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4800347215},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4714092807},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4709959213},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4574652775},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.457051917},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4467868165},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4391809979},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4371555354},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.423225309},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4205109127},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4114721115},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.4055197297},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.3913414894},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3654789464},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.365093146},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3452380958},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.3357445988},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.328634811},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.3202160503},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3174740941},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3132164911},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.303323413},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2978808434},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2945739646},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2820767205},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2678020287},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2606646834},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2480296519},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2355875225},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2299107147},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.2161044983},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2026289694},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1628913149},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0314429021}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_classification/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_classification/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ef7da0d29f7779583b98e70e8ac665c1d49342a8
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_classification/datasets_small/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned + ensembled),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned),TabDPT-Turbo (default),LimiX (default),RealMLP (tuned + ensembled),TabM (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabDPT (tuned + ensembled),BetaTabPFN (default),TabDPT (tuned),RealMLP (tuned),TabM (tuned),TabSwift (default),LightGBM (tuned + ensembled),LightGBM (tuned),TabDPT (default),CatBoost (tuned + ensembled),ModernNCA (tuned),xRFM (tuned + ensembled),SAP-RPT-OSS (default),iLTM (tuned + ensembled),CatBoost (tuned),xRFM (tuned),XGBoost (tuned + ensembled),CatBoost (default),EBM (tuned + ensembled),TabM (default),ModernNCA (default),XGBoost (tuned),FastaiMLP (tuned + ensembled),TorchMLP (tuned + ensembled),ModernNCA (tuned + ensembled),EBM (tuned),EBM (default),iLTM (tuned),RealMLP (default),ChimeraBoost (tuned + ensembled),FastaiMLP (tuned),ExtraTrees (tuned + ensembled),TorchMLP (tuned),ChimeraBoost (tuned),OrionMSP (default),ExtraTrees (tuned),TabSTAR (tuned),TabSTAR (tuned + ensembled),ChimeraBoost (default),RandomForest (tuned + ensembled),XGBoost (default),LightGBM (default),TabFlex (default),RandomForest (tuned),Linear (tuned + ensembled),TorchMLP (default),iLTM (default),TabSTAR (default),PerpetualBooster (tuned + ensembled),Linear (tuned),xRFM (default),PerpetualBooster (tuned),FastaiMLP (default),KNN (tuned + ensembled),ExtraTrees (default),RandomForest (default),Linear (default),PerpetualBooster (default),KNN (tuned),KNN (default)
+TabFM (default),,0.7956349,0.7876984,0.8194444,0.76785713,0.7708333,0.8035714,0.7757937,0.8313492,0.8799603,0.85714287,0.86507934,0.9047619,0.8869048,0.8581349,0.87896824,0.85119045,0.89484125,0.86904764,0.9007937,0.89880955,0.890873,0.9186508,0.9027778,0.9126984,0.8373016,0.9206349,0.92460316,0.89285713,0.9325397,0.9265873,0.89484125,0.9007937,0.9047619,0.9325397,0.8968254,0.9444444,0.9444444,0.90674603,0.9007937,0.9345238,0.9186508,0.9285714,0.92261904,0.93849206,0.95238096,0.93849206,0.9305556,0.9285714,0.82539684,0.82539684,0.9464286,0.9563492,0.9583333,0.9543651,0.9265873,0.9543651,0.93650794,0.96031743,0.9305556,0.87103176,0.98214287,0.9444444,0.96428573,0.9781746,0.96626985,0.96428573,0.9880952,0.98015875,0.96031743,0.98015875,0.9563492,0.99404764
+TabPFN-3 (default),0.20436507,,0.5555556,0.5257937,0.60119045,0.57539684,0.6111111,0.5972222,0.6964286,0.71825397,0.7242063,0.75793654,0.74206346,0.78373015,0.73710316,0.8055556,0.7638889,0.7956349,0.84325397,0.8154762,0.8333333,0.8194444,0.8492063,0.8492063,0.8531746,0.74007934,0.8769841,0.859127,0.8472222,0.8501984,0.875,0.83928573,0.8194444,0.8640873,0.8630952,0.83035713,0.8968254,0.8869048,0.8531746,0.8660714,0.9156746,0.9027778,0.90873015,0.8839286,0.92460316,0.9166667,0.90873015,0.9126984,0.92261904,0.8125,0.8174603,0.90873015,0.9206349,0.9583333,0.9404762,0.9166667,0.9424603,0.9047619,0.95039684,0.93849206,0.8531746,0.93650794,0.9126984,0.96825397,0.95039684,0.9583333,0.9424603,0.9742063,0.9781746,0.9484127,0.96428573,0.95238096,0.99603176
+"AutoGluon 1.5 (extreme, 4h)",0.21230158,0.44444445,,0.46031746,0.5416667,0.53373015,0.609127,0.5535714,0.5972222,0.7202381,0.76289684,0.7916667,0.7916667,0.765873,0.7380952,0.7876984,0.7956349,0.796627,0.8035714,0.8373016,0.8333333,0.7619048,0.8531746,0.85515875,0.8611111,0.7202381,0.8769841,0.8501984,0.83035713,0.8611111,0.8660714,0.83928573,0.8373016,0.8472222,0.859127,0.8373016,0.8869048,0.90873015,0.85119045,0.85714287,0.90873015,0.86706346,0.9126984,0.86507934,0.9047619,0.9206349,0.91468257,0.9156746,0.890873,0.7777778,0.7797619,0.9166667,0.92261904,0.9464286,0.9464286,0.8968254,0.9265873,0.88293654,0.95039684,0.9325397,0.82539684,0.9444444,0.90873015,0.9464286,0.9494048,0.9444444,0.9325397,0.9543651,0.9563492,0.9355159,0.9583333,0.9464286,0.99404764
+TabICLv2 (default),0.18055555,0.47420636,0.53968257,,0.53869045,0.5297619,0.6031746,0.5436508,0.6825397,0.7172619,0.7242063,0.7619048,0.7718254,0.77876985,0.74007934,0.76984125,0.74603176,0.7757937,0.82539684,0.8055556,0.82738096,0.7956349,0.8174603,0.83531743,0.8313492,0.6984127,0.86706346,0.82738096,0.8144841,0.83531743,0.859127,0.8333333,0.81150794,0.8640873,0.84325397,0.83531743,0.8769841,0.8630952,0.83928573,0.859127,0.88492066,0.88095236,0.89285713,0.87896824,0.8720238,0.90873015,0.90674603,0.90873015,0.8769841,0.8035714,0.8015873,0.9027778,0.91468257,0.9265873,0.93650794,0.89484125,0.9126984,0.9186508,0.9345238,0.9166667,0.85515875,0.9424603,0.92460316,0.9404762,0.9484127,0.95039684,0.9424603,0.95039684,0.96031743,0.9513889,0.9722222,0.9464286,0.99603176
+RealTabPFN-2.5 (default),0.23214285,0.39880952,0.45833334,0.46130952,,0.44047618,0.50793654,0.4890873,0.5714286,0.68849206,0.71031743,0.75992066,0.75396824,0.6924603,0.7589286,0.6994048,0.76984125,0.79960316,0.78968257,0.8075397,0.8293651,0.7123016,0.8472222,0.8055556,0.82539684,0.7619048,0.8194444,0.8472222,0.8333333,0.85515875,0.88293654,0.859127,0.8531746,0.83928573,0.875,0.86706346,0.8869048,0.83531743,0.87896824,0.8611111,0.90376985,0.89880955,0.9027778,0.890873,0.8630952,0.9166667,0.9047619,0.9206349,0.875,0.8035714,0.80456346,0.9126984,0.8869048,0.9305556,0.9444444,0.91369045,0.9345238,0.9007937,0.9623016,0.9325397,0.8373016,0.9404762,0.90873015,0.9484127,0.9464286,0.9464286,0.9027778,0.95238096,0.9543651,0.95238096,0.96626985,0.93650794,0.99404764
+RealTabPFN-2.5 (tuned + ensembled),0.22916667,0.42460316,0.46626985,0.4702381,0.5595238,,0.5515873,0.57539684,0.5654762,0.7083333,0.7361111,0.75595236,0.7757937,0.6845238,0.76984125,0.6984127,0.76785713,0.78373015,0.75793654,0.81349206,0.83035713,0.6686508,0.85119045,0.8472222,0.8333333,0.7529762,0.8382937,0.8452381,0.8214286,0.859127,0.8630952,0.83531743,0.82539684,0.8333333,0.8720238,0.8343254,0.87896824,0.87896824,0.84821427,0.84126985,0.8968254,0.8630952,0.8918651,0.875,0.859127,0.91071427,0.89880955,0.890873,0.8611111,0.7777778,0.78373015,0.89880955,0.9027778,0.9404762,0.93650794,0.890873,0.91468257,0.89880955,0.93849206,0.92261904,0.8154762,0.9424603,0.91468257,0.9404762,0.9424603,0.9404762,0.9166667,0.95039684,0.9484127,0.9255952,0.9583333,0.9265873,0.99206346
+TabPFN-2.6 (default),0.19642857,0.3888889,0.39087301,0.3968254,0.4920635,0.4484127,,0.47916666,0.5873016,0.6626984,0.71825397,0.74206346,0.74404764,0.67460316,0.7281746,0.68650794,0.75595236,0.7797619,0.7619048,0.79960316,0.83531743,0.70039684,0.8293651,0.8293651,0.8095238,0.7380952,0.87896824,0.8214286,0.8015873,0.859127,0.875,0.8095238,0.81845236,0.8452381,0.86507934,0.82539684,0.8769841,0.90873015,0.8382937,0.8313492,0.9047619,0.86507934,0.91071427,0.86706346,0.88293654,0.90873015,0.9126984,0.8611111,0.8720238,0.7757937,0.78571427,0.9186508,0.9007937,0.9543651,0.9444444,0.89484125,0.92261904,0.88492066,0.9404762,0.92261904,0.8154762,0.9325397,0.90873015,0.9404762,0.9285714,0.93650794,0.9325397,0.95238096,0.96626985,0.93650794,0.9563492,0.9424603,0.99206346
+RealTabPFN-2.5 (tuned),0.22420634,0.4027778,0.44642857,0.4563492,0.5109127,0.42460316,0.5208333,,0.5694444,0.67261904,0.71031743,0.75396824,0.7638889,0.6805556,0.75595236,0.7093254,0.7638889,0.78869045,0.74900794,0.79960316,0.8293651,0.68650794,0.8373016,0.8095238,0.8144841,0.7480159,0.8194444,0.8333333,0.8095238,0.8452381,0.85119045,0.8472222,0.8293651,0.8214286,0.8531746,0.8214286,0.87896824,0.8700397,0.85119045,0.83531743,0.890873,0.8630952,0.8819444,0.86507934,0.86210316,0.91071427,0.89285713,0.88492066,0.86904764,0.76984125,0.7777778,0.8968254,0.89285713,0.9345238,0.93650794,0.88095236,0.90873015,0.88492066,0.9464286,0.9186508,0.8154762,0.9255952,0.9007937,0.9424603,0.93650794,0.9325397,0.9166667,0.9444444,0.9464286,0.93353176,0.9583333,0.9325397,0.99206346
+TabDPT-Turbo (default),0.16865079,0.30357143,0.4027778,0.31746033,0.42857143,0.43452382,0.41269842,0.43055555,,0.56150794,0.5833333,0.58035713,0.58928573,0.63095236,0.5674603,0.6527778,0.63789684,0.62103176,0.7291667,0.65873015,0.66071427,0.78869045,0.6984127,0.6904762,0.68650794,0.63095236,0.7480159,0.68849206,0.71428573,0.71428573,0.7222222,0.7123016,0.68650794,0.765873,0.7222222,0.76984125,0.74007934,0.73214287,0.7450397,0.75595236,0.82242066,0.7916667,0.8015873,0.82738096,0.83630955,0.8015873,0.8174603,0.8611111,0.859127,0.73015875,0.7380952,0.8313492,0.875,0.86706346,0.8730159,0.90873015,0.87896824,0.84126985,0.88492066,0.8630952,0.79960316,0.88492066,0.85119045,0.8888889,0.9047619,0.92261904,0.9186508,0.953373,0.95238096,0.9305556,0.9543651,0.9424603,0.99404764
+LimiX (default),0.12003968,0.28174603,0.2797619,0.2827381,0.31150794,0.29166666,0.33730158,0.32738096,0.43849206,,0.5972222,0.6269841,0.63293654,0.53869045,0.625,0.5674603,0.6418651,0.63988096,0.609127,0.71626985,0.72619045,0.58531743,0.7380952,0.7380952,0.75793654,0.6686508,0.796627,0.74007934,0.7450397,0.7718254,0.7777778,0.76785713,0.71428573,0.76984125,0.78968257,0.7172619,0.75595236,0.7777778,0.7956349,0.79761904,0.82738096,0.8095238,0.8472222,0.7906746,0.84126985,0.81349206,0.85515875,0.82043654,0.83928573,0.76488096,0.75793654,0.87103176,0.8968254,0.89880955,0.9126984,0.81150794,0.91071427,0.8611111,0.8888889,0.89285713,0.8105159,0.9047619,0.88293654,0.9345238,0.91468257,0.9206349,0.89484125,0.9464286,0.9543651,0.890873,0.93650794,0.91468257,0.99007934
+RealMLP (tuned + ensembled),0.14285715,0.27579364,0.23710318,0.27579364,0.28968254,0.2638889,0.28174603,0.28968254,0.41666666,0.4027778,,0.57738096,0.5327381,0.48710316,0.59325397,0.49503967,0.66369045,0.63492066,0.5297619,0.70039684,0.7311508,0.54761904,0.7519841,0.6626984,0.7043651,0.5793651,0.74007934,0.75595236,0.6825397,0.8174603,0.7718254,0.7718254,0.75595236,0.71031743,0.82738096,0.7569444,0.73214287,0.7123016,0.7797619,0.77876985,0.81150794,0.84623015,0.8214286,0.78075397,0.79761904,0.78571427,0.83035713,0.74404764,0.8125,0.72619045,0.734127,0.8313492,0.8869048,0.91468257,0.9007937,0.7956349,0.9186508,0.8472222,0.9007937,0.890873,0.7797619,0.92261904,0.8492063,0.91468257,0.9285714,0.9285714,0.89484125,0.9464286,0.9583333,0.87103176,0.96031743,0.9206349,0.99404764
+TabM (tuned + ensembled),0.13492064,0.24206349,0.20833333,0.23809524,0.24007936,0.24404761,0.2579365,0.24603175,0.41964287,0.37301588,0.42261904,,0.49007937,0.48313493,0.59126985,0.49007937,0.5515873,0.6547619,0.53968257,0.640873,0.66369045,0.5456349,0.6507937,0.6220238,0.60714287,0.58928573,0.65178573,0.6765873,0.6111111,0.74603176,0.75,0.7619048,0.82242066,0.6845238,0.76785713,0.6944444,0.71825397,0.70039684,0.7757937,0.7688492,0.75,0.78869045,0.8095238,0.75,0.73015875,0.7986111,0.83531743,0.71626985,0.72619045,0.72619045,0.734127,0.86706346,0.78968257,0.88095236,0.8839286,0.79761904,0.84126985,0.85515875,0.89285713,0.88095236,0.78373015,0.89285713,0.8720238,0.859127,0.9126984,0.9007937,0.8730159,0.90674603,0.9007937,0.89880955,0.9563492,0.91071427,0.984127
+"AutoGluon 1.4 (best, 4h)",0.0952381,0.2579365,0.20833333,0.2281746,0.24603175,0.22420634,0.2559524,0.2361111,0.4107143,0.3670635,0.4672619,0.50992066,,0.47619048,0.5515873,0.4781746,0.5595238,0.5694444,0.52281743,0.6706349,0.6904762,0.54761904,0.6785714,0.6111111,0.6438492,0.5297619,0.6805556,0.7063492,0.67261904,0.72619045,0.74404764,0.7063492,0.6944444,0.6904762,0.7123016,0.7242063,0.7519841,0.6785714,0.72619045,0.7638889,0.7718254,0.77380955,0.7847222,0.78174603,0.7083333,0.8055556,0.7956349,0.7093254,0.74007934,0.7242063,0.73214287,0.8194444,0.8075397,0.8769841,0.9007937,0.8095238,0.84821427,0.8422619,0.8769841,0.86507934,0.78174603,0.91468257,0.8700397,0.8769841,0.9206349,0.9166667,0.875,0.92460316,0.91468257,0.9047619,0.93849206,0.91468257,0.99603176
+TabDPT (tuned + ensembled),0.11309524,0.21626984,0.23412699,0.22123016,0.30753967,0.31547618,0.32539684,0.31944445,0.3690476,0.46130952,0.51289684,0.5168651,0.52380955,,0.5218254,0.60714287,0.59424603,0.55456346,0.59821427,0.5952381,0.6269841,0.63095236,0.625,0.6468254,0.63988096,0.58531743,0.6944444,0.63293654,0.65376985,0.6577381,0.6805556,0.6656746,0.6418651,0.7013889,0.6825397,0.6964286,0.69742066,0.73015875,0.69543654,0.6924603,0.7797619,0.7281746,0.75595236,0.7519841,0.78373015,0.75992066,0.76984125,0.8015873,0.79960316,0.6984127,0.70039684,0.7936508,0.85714287,0.84325397,0.85119045,0.8293651,0.87103176,0.78571427,0.86904764,0.85714287,0.7718254,0.86706346,0.8075397,0.89781743,0.8819444,0.9007937,0.90178573,0.93849206,0.9305556,0.86507934,0.9305556,0.9126984,0.99404764
+BetaTabPFN (default),0.14186507,0.26289684,0.26190478,0.25992063,0.24107143,0.23015873,0.2718254,0.24404761,0.43253967,0.375,0.40674603,0.40873015,0.4484127,0.4781746,,0.5,0.45833334,0.49007937,0.5277778,0.60119045,0.5992063,0.5833333,0.5873016,0.59623015,0.5952381,0.5376984,0.60515875,0.6220238,0.60515875,0.66071427,0.6964286,0.71825397,0.6686508,0.65674603,0.67460316,0.65178573,0.6825397,0.61904764,0.75793654,0.78174603,0.6706349,0.74007934,0.75099206,0.7242063,0.72619045,0.75793654,0.7619048,0.71825397,0.74206346,0.7519841,0.75595236,0.8174603,0.79761904,0.8075397,0.81845236,0.7986111,0.8075397,0.87896824,0.859127,0.89880955,0.8125,0.85515875,0.89880955,0.78968257,0.8769841,0.859127,0.87896824,0.8531746,0.87896824,0.9325397,0.9305556,0.9484127,0.9861111
+TabDPT (tuned),0.121031746,0.19444445,0.21230158,0.23015873,0.30059522,0.3015873,0.31349206,0.2906746,0.3472222,0.43253967,0.5049603,0.50992066,0.5218254,0.39285713,0.5,,0.54960316,0.5515873,0.5694444,0.5674603,0.59126985,0.6140873,0.5952381,0.6130952,0.5952381,0.5555556,0.6388889,0.60218257,0.60515875,0.64781743,0.6468254,0.6468254,0.63095236,0.67261904,0.6507937,0.6686508,0.6845238,0.69345236,0.67460316,0.6825397,0.73015875,0.70238096,0.734127,0.72619045,0.7361111,0.72619045,0.74206346,0.81150794,0.75595236,0.6875,0.68650794,0.7797619,0.79960316,0.7916667,0.8154762,0.82539684,0.82539684,0.7748016,0.85119045,0.8313492,0.765873,0.8452381,0.78968257,0.8611111,0.85714287,0.8869048,0.89880955,0.9047619,0.90873015,0.8611111,0.9285714,0.9027778,0.99007934
+RealMLP (tuned),0.14880952,0.2361111,0.20436507,0.25396827,0.23015873,0.23214285,0.24404761,0.2361111,0.36210316,0.35813493,0.33630952,0.4484127,0.44047618,0.40575397,0.5416667,0.45039684,,0.5376984,0.48412699,0.57738096,0.5952381,0.5019841,0.6130952,0.5714286,0.5734127,0.5218254,0.5922619,0.62103176,0.6001984,0.6924603,0.67460316,0.6805556,0.6706349,0.6527778,0.71130955,0.66468257,0.6765873,0.6468254,0.7202381,0.6825397,0.71428573,0.74603176,0.7083333,0.734127,0.73015875,0.74603176,0.734127,0.67956346,0.74206346,0.71031743,0.71626985,0.7638889,0.8075397,0.8333333,0.8373016,0.77876985,0.8313492,0.7876984,0.86706346,0.8472222,0.765873,0.86904764,0.796627,0.8611111,0.87103176,0.8998016,0.8492063,0.86507934,0.890873,0.83928573,0.9424603,0.88095236,0.98214287
+TabM (tuned),0.10515873,0.20436507,0.20337301,0.22420634,0.20039682,0.21626984,0.22023809,0.21130952,0.37896827,0.36011904,0.36507937,0.3452381,0.43055555,0.4454365,0.50992066,0.4484127,0.46230158,,0.50396824,0.5625,0.58531743,0.52281743,0.5972222,0.59325397,0.5456349,0.5119048,0.6031746,0.61507934,0.5694444,0.6656746,0.6825397,0.6666667,0.70039684,0.6468254,0.67460316,0.6458333,0.6805556,0.6547619,0.7043651,0.71825397,0.6765873,0.74404764,0.7619048,0.71626985,0.6825397,0.74900794,0.75396824,0.68849206,0.6904762,0.7063492,0.7172619,0.8144841,0.77380955,0.8313492,0.8174603,0.78373015,0.7916667,0.83928573,0.875,0.86507934,0.7638889,0.890873,0.85119045,0.8154762,0.89880955,0.85515875,0.8472222,0.859127,0.875,0.88095236,0.9464286,0.8888889,0.97619045
+TabSwift (default),0.13095239,0.15674603,0.19642857,0.17460318,0.21031746,0.24206349,0.23809524,0.25099206,0.27083334,0.39087301,0.4702381,0.46031746,0.47718254,0.4017857,0.4722222,0.43055555,0.515873,0.49603173,,0.51984125,0.5257937,0.515873,0.53968257,0.5694444,0.5515873,0.57539684,0.58928573,0.5644841,0.5595238,0.5972222,0.60515875,0.61507934,0.6230159,0.6230159,0.62896824,0.7063492,0.65674603,0.62103176,0.64285713,0.64880955,0.7043651,0.6964286,0.6626984,0.7361111,0.6805556,0.71626985,0.6924603,0.84325397,0.6944444,0.74404764,0.75,0.7757937,0.7222222,0.7380952,0.8105159,0.8888889,0.78174603,0.76785713,0.84126985,0.81150794,0.8095238,0.87103176,0.7936508,0.8055556,0.87103176,0.8869048,0.8630952,0.8452381,0.88293654,0.8998016,0.9186508,0.93849206,0.99404764
+LightGBM (tuned + ensembled),0.09920635,0.1845238,0.16269842,0.19444445,0.19246031,0.18650794,0.20039682,0.20039682,0.34126985,0.28373015,0.29960316,0.35912699,0.32936507,0.4047619,0.39880952,0.43253967,0.42261904,0.4375,0.48015872,,0.5625,0.5257937,0.57738096,0.50396824,0.54761904,0.5138889,0.5605159,0.5882937,0.5992063,0.7063492,0.69146824,0.6458333,0.65178573,0.5972222,0.71825397,0.59126985,0.59126985,0.5734127,0.68650794,0.7361111,0.6686508,0.68849206,0.7470238,0.6805556,0.6894841,0.6984127,0.75992066,0.6964286,0.6944444,0.7093254,0.7123016,0.84126985,0.79761904,0.8730159,0.9424603,0.74603176,0.85714287,0.81150794,0.8055556,0.8730159,0.78968257,0.8769841,0.8333333,0.8035714,0.9007937,0.8373016,0.8730159,0.9186508,0.9285714,0.8968254,0.9345238,0.91765875,0.99206346
+LightGBM (tuned),0.10119048,0.16666667,0.16666667,0.17261904,0.17063493,0.16964285,0.16468254,0.17063493,0.3392857,0.27380952,0.2688492,0.33630952,0.30952382,0.37301588,0.40079364,0.40873015,0.4047619,0.41468254,0.47420636,0.4375,,0.50396824,0.54464287,0.46626985,0.5188492,0.4920635,0.515873,0.5595238,0.5595238,0.62400794,0.6656746,0.625,0.61706346,0.57539684,0.64880955,0.5625,0.5833333,0.5456349,0.6686508,0.70039684,0.6388889,0.6755952,0.6944444,0.6388889,0.6656746,0.68353176,0.71031743,0.6845238,0.6875,0.6944444,0.70039684,0.8015873,0.7519841,0.84424603,0.9077381,0.734127,0.8214286,0.8035714,0.7916667,0.8561508,0.77380955,0.86210316,0.8263889,0.796627,0.88095236,0.8293651,0.85515875,0.9047619,0.9027778,0.88293654,0.9325397,0.90674603,0.99404764
+TabDPT (default),0.109126985,0.18055555,0.23809524,0.20436507,0.28769842,0.3313492,0.29960316,0.31349206,0.21130952,0.41468254,0.45238096,0.45436507,0.45238096,0.3690476,0.41666666,0.3859127,0.49801588,0.47718254,0.48412699,0.47420636,0.49603173,,0.5257937,0.5456349,0.5277778,0.53174603,0.5416667,0.53571427,0.53174603,0.5694444,0.5376984,0.5297619,0.5734127,0.5952381,0.58531743,0.58531743,0.5873016,0.59325397,0.55654764,0.59126985,0.63095236,0.5843254,0.6507937,0.65873015,0.68849206,0.64285713,0.6527778,0.74603176,0.71428573,0.63293654,0.63293654,0.71031743,0.7123016,0.6845238,0.7380952,0.84126985,0.7638889,0.6944444,0.77380955,0.79761904,0.7380952,0.7936508,0.72619045,0.8035714,0.8174603,0.86507934,0.875,0.890873,0.859127,0.8452381,0.8769841,0.9007937,0.99206346
+CatBoost (tuned + ensembled),0.08134921,0.15079366,0.1468254,0.18253969,0.15277778,0.14880952,0.17063493,0.16269842,0.3015873,0.26190478,0.24801587,0.34920636,0.32142857,0.375,0.41269842,0.4047619,0.38690478,0.4027778,0.46031746,0.42261904,0.45535713,0.47420636,,0.48214287,0.49603173,0.5,0.51785713,0.5763889,0.54761904,0.56845236,0.59325397,0.5555556,0.5952381,0.5575397,0.59623015,0.5535714,0.5734127,0.56150794,0.61210316,0.64285713,0.6547619,0.6626984,0.66468257,0.63293654,0.6944444,0.6785714,0.6706349,0.64285713,0.7172619,0.6984127,0.6924603,0.71031743,0.8015873,0.8531746,0.85515875,0.6904762,0.8581349,0.77380955,0.8015873,0.828373,0.7519841,0.8174603,0.7916667,0.8214286,0.8373016,0.8075397,0.8630952,0.890873,0.91468257,0.8313492,0.890873,0.8968254,0.99206346
+ModernNCA (tuned),0.097222224,0.15079366,0.14484127,0.16468254,0.19444445,0.15277778,0.17063493,0.1904762,0.30952382,0.26190478,0.33730158,0.37797618,0.3888889,0.3531746,0.40376985,0.38690478,0.42857143,0.40674603,0.43055555,0.49603173,0.53373015,0.45436507,0.51785713,,0.5486111,0.48412699,0.5,0.53373015,0.5972222,0.5694444,0.5813492,0.5833333,0.5376984,0.63492066,0.5763889,0.58234125,0.5694444,0.54464287,0.625,0.63492066,0.6230159,0.67261904,0.6805556,0.6527778,0.6507937,0.71428573,0.68154764,0.66071427,0.6765873,0.6706349,0.6785714,0.73214287,0.7480159,0.75595236,0.7619048,0.765873,0.77380955,0.74404764,0.7956349,0.81349206,0.75396824,0.8095238,0.77380955,0.85515875,0.8313492,0.82043654,0.8194444,0.86507934,0.8472222,0.8214286,0.88293654,0.86706346,0.98015875
+xRFM (tuned + ensembled),0.08730159,0.1468254,0.1388889,0.16865079,0.17460318,0.16666667,0.1904762,0.18551587,0.31349206,0.24206349,0.29563493,0.39285713,0.3561508,0.36011904,0.4047619,0.4047619,0.4265873,0.45436507,0.4484127,0.45238096,0.4811508,0.4722222,0.50396824,0.4513889,,0.45238096,0.47718254,0.52281743,0.59623015,0.5367063,0.5843254,0.54761904,0.5436508,0.5535714,0.54960316,0.55654764,0.5456349,0.5515873,0.61507934,0.6388889,0.6160714,0.6269841,0.6825397,0.6388889,0.64285713,0.640873,0.68849206,0.6904762,0.6964286,0.66765875,0.6666667,0.7063492,0.75793654,0.7619048,0.7638889,0.78373015,0.8035714,0.7916667,0.8234127,0.78174603,0.74404764,0.8154762,0.8075397,0.9305556,0.84821427,0.88293654,0.8680556,0.9325397,0.93650794,0.84623015,0.91071427,0.89880955,0.99007934
+SAP-RPT-OSS (default),0.16269842,0.25992063,0.2797619,0.3015873,0.23809524,0.2470238,0.26190478,0.25198412,0.3690476,0.3313492,0.42063493,0.4107143,0.4702381,0.41468254,0.46230158,0.44444445,0.4781746,0.48809522,0.42460316,0.4861111,0.50793654,0.46825397,0.5,0.515873,0.54761904,,0.5297619,0.515873,0.5535714,0.5515873,0.5277778,0.5595238,0.56150794,0.56349206,0.5595238,0.6269841,0.6527778,0.5734127,0.58928573,0.60218257,0.5882937,0.6805556,0.5654762,0.67460316,0.63095236,0.6924603,0.59126985,0.60119045,0.63095236,0.6031746,0.5952381,0.640873,0.6994048,0.6706349,0.68849206,0.7123016,0.71130955,0.7519841,0.7519841,0.7202381,0.73015875,0.7619048,0.75595236,0.7619048,0.7668651,0.78174603,0.71626985,0.765873,0.7777778,0.78174603,0.79761904,0.82539684,0.98214287
+iLTM (tuned + ensembled),0.07936508,0.12301587,0.12301587,0.13293651,0.18055555,0.16170634,0.121031746,0.18055555,0.25198412,0.20337301,0.25992063,0.3482143,0.31944445,0.30555555,0.39484128,0.3611111,0.4077381,0.3968254,0.4107143,0.43948412,0.48412699,0.45833334,0.48214287,0.5,0.52281743,0.4702381,,0.5138889,0.56646824,0.54464287,0.5515873,0.56349206,0.51785713,0.5992063,0.5833333,0.58531743,0.54960316,0.5714286,0.609127,0.62896824,0.70535713,0.6626984,0.6845238,0.66964287,0.6617063,0.65873015,0.6805556,0.63789684,0.6944444,0.64285713,0.6438492,0.72718257,0.76984125,0.8055556,0.84126985,0.76984125,0.81150794,0.7043651,0.79960316,0.8154762,0.7480159,0.8293651,0.74404764,0.85714287,0.85515875,0.8373016,0.8214286,0.89484125,0.89484125,0.8075397,0.8968254,0.8660714,0.9781746
+CatBoost (tuned),0.07539683,0.14087301,0.14980158,0.17261904,0.15277778,0.15476191,0.17857143,0.16666667,0.31150794,0.25992063,0.24404761,0.3234127,0.2936508,0.3670635,0.37797618,0.39781746,0.37896827,0.38492063,0.43551588,0.41170636,0.44047618,0.4642857,0.4236111,0.46626985,0.47718254,0.48412699,0.4861111,,0.53373015,0.5674603,0.5882937,0.56150794,0.5763889,0.53968257,0.58928573,0.53174603,0.5734127,0.5515873,0.5972222,0.64285713,0.62103176,0.6468254,0.6180556,0.62400794,0.67460316,0.6656746,0.6458333,0.640873,0.68650794,0.6984127,0.70039684,0.70238096,0.7916667,0.8472222,0.8293651,0.6825397,0.8333333,0.76785713,0.7797619,0.81150794,0.75793654,0.8055556,0.7827381,0.8075397,0.8293651,0.79761904,0.8531746,0.87896824,0.9047619,0.8234127,0.89880955,0.9027778,0.99206346
+xRFM (tuned),0.10714286,0.15277778,0.16964285,0.18551587,0.16666667,0.17857143,0.1984127,0.1904762,0.2857143,0.25496033,0.31746033,0.3888889,0.32738096,0.34623015,0.39484128,0.39484128,0.39980158,0.43055555,0.44047618,0.40079364,0.44047618,0.46825397,0.45238096,0.4027778,0.40376985,0.44642857,0.43353173,0.46626985,,0.48412699,0.50793654,0.52281743,0.5257937,0.53571427,0.5069444,0.52876985,0.53571427,0.46230158,0.57738096,0.6111111,0.58531743,0.57738096,0.6438492,0.60714287,0.6369048,0.625,0.6666667,0.6765873,0.67460316,0.6805556,0.6845238,0.70238096,0.6805556,0.66071427,0.7123016,0.8015873,0.74007934,0.7916667,0.8194444,0.7956349,0.75,0.7827381,0.81349206,0.8888889,0.83035713,0.8611111,0.8531746,0.875,0.8769841,0.86706346,0.89880955,0.89880955,0.99007934
+XGBoost (tuned + ensembled),0.06746032,0.14980158,0.1388889,0.16468254,0.14484127,0.14087301,0.14087301,0.15476191,0.2857143,0.2281746,0.18253969,0.25396827,0.27380952,0.3422619,0.3392857,0.35218254,0.30753967,0.3343254,0.4027778,0.2936508,0.37599206,0.43055555,0.4315476,0.43055555,0.46329364,0.4484127,0.45535713,0.43253967,0.515873,,0.5327381,0.515873,0.53571427,0.5277778,0.54265875,0.5218254,0.53869045,0.5,0.58531743,0.60515875,0.5952381,0.6031746,0.5833333,0.57738096,0.60714287,0.61904764,0.62103176,0.6339286,0.65873015,0.67261904,0.67261904,0.65575397,0.77876985,0.8730159,0.8730159,0.69146824,0.80853176,0.75793654,0.75595236,0.76984125,0.734127,0.7757937,0.7718254,0.7936508,0.7986111,0.7718254,0.8174603,0.86706346,0.9027778,0.8214286,0.88095236,0.8779762,0.98214287
+CatBoost (default),0.0734127,0.125,0.13392857,0.14087301,0.11706349,0.13690476,0.125,0.14880952,0.2777778,0.22222222,0.2281746,0.25,0.2559524,0.31944445,0.30357143,0.3531746,0.32539684,0.31746033,0.39484128,0.30853173,0.3343254,0.46230158,0.40674603,0.4186508,0.4156746,0.4722222,0.4484127,0.41170636,0.4920635,0.4672619,,0.51984125,0.5257937,0.4920635,0.4890873,0.52380955,0.53373015,0.49801588,0.59126985,0.625,0.59325397,0.61706346,0.6299603,0.60119045,0.62896824,0.640873,0.65873015,0.6230159,0.64880955,0.69146824,0.6944444,0.73015875,0.74603176,0.765873,0.8035714,0.7043651,0.7876984,0.74206346,0.7876984,0.8630952,0.7619048,0.7916667,0.765873,0.765873,0.8333333,0.81349206,0.84325397,0.8611111,0.86904764,0.8472222,0.88293654,0.90873015,0.9880952
+EBM (tuned + ensembled),0.10515873,0.16071428,0.16071428,0.16666667,0.14087301,0.16468254,0.1904762,0.15277778,0.28769842,0.23214285,0.2281746,0.23809524,0.2936508,0.3343254,0.28174603,0.3531746,0.31944445,0.33333334,0.38492063,0.35416666,0.375,0.4702381,0.44444445,0.41666666,0.45238096,0.44047618,0.43650794,0.43849206,0.47718254,0.48412699,0.48015872,,0.51785713,0.5119048,0.4920635,0.49107143,0.53174603,0.47619048,0.67956346,0.70039684,0.5456349,0.58928573,0.5793651,0.5644841,0.63988096,0.61210316,0.5992063,0.6547619,0.6388889,0.67460316,0.6765873,0.6706349,0.7083333,0.70734125,0.71428573,0.7380952,0.72619045,0.79960316,0.74206346,0.8075397,0.75992066,0.78571427,0.8174603,0.75396824,0.80853176,0.8194444,0.8293651,0.8333333,0.83531743,0.85119045,0.8968254,0.8938492,0.98015875
+TabM (default),0.09920635,0.18055555,0.16269842,0.18849206,0.1468254,0.17460318,0.18154761,0.17063493,0.31349206,0.2857143,0.24404761,0.17757936,0.30555555,0.35813493,0.3313492,0.3690476,0.32936507,0.29960316,0.37698412,0.3482143,0.3829365,0.4265873,0.4047619,0.46230158,0.4563492,0.43849206,0.48214287,0.4236111,0.47420636,0.4642857,0.47420636,0.48214287,,0.51785713,0.4920635,0.5535714,0.5654762,0.5376984,0.5456349,0.546627,0.5644841,0.6111111,0.5575397,0.62896824,0.6111111,0.64880955,0.5793651,0.60714287,0.63492066,0.6924603,0.70039684,0.63492066,0.71031743,0.6666667,0.6547619,0.6944444,0.7242063,0.7638889,0.78174603,0.7718254,0.76984125,0.7569444,0.7847222,0.71428573,0.7638889,0.7916667,0.8055556,0.8095238,0.8293651,0.8293651,0.859127,0.89285713,0.9722222
+ModernNCA (default),0.0952381,0.1359127,0.15277778,0.1359127,0.16071428,0.16666667,0.15476191,0.17857143,0.23412699,0.23015873,0.28968254,0.31547618,0.30952382,0.2986111,0.34325397,0.32738096,0.3472222,0.3531746,0.37698412,0.4027778,0.42460316,0.4047619,0.44246033,0.36507937,0.44642857,0.43650794,0.40079364,0.46031746,0.4642857,0.4722222,0.50793654,0.48809522,0.48214287,,0.50595236,0.4781746,0.4781746,0.40674603,0.5456349,0.58531743,0.52281743,0.53968257,0.65674603,0.5436508,0.5952381,0.57738096,0.66468257,0.6527778,0.63492066,0.6507937,0.65674603,0.70039684,0.67460316,0.63293654,0.70039684,0.74404764,0.74007934,0.74404764,0.78373015,0.78373015,0.734127,0.77380955,0.7688492,0.73214287,0.8055556,0.8095238,0.8472222,0.81845236,0.8154762,0.859127,0.85714287,0.87896824,0.984127
+XGBoost (tuned),0.06746032,0.13690476,0.14087301,0.15674603,0.125,0.1279762,0.13492064,0.1468254,0.2777778,0.21031746,0.17261904,0.23214285,0.28769842,0.31746033,0.32539684,0.34920636,0.28869048,0.32539684,0.37103173,0.28174603,0.35119048,0.41468254,0.40376985,0.4236111,0.45039684,0.44047618,0.41666666,0.4107143,0.49305555,0.45734128,0.5109127,0.50793654,0.50793654,0.4940476,,0.4940476,0.5218254,0.49801588,0.5575397,0.57738096,0.5793651,0.57539684,0.5486111,0.57738096,0.6140873,0.5952381,0.5952381,0.6130952,0.6547619,0.65674603,0.6597222,0.61904764,0.75793654,0.83531743,0.8373016,0.6805556,0.8055556,0.74404764,0.73214287,0.765873,0.73015875,0.7480159,0.76984125,0.78571427,0.78968257,0.76785713,0.81845236,0.86904764,0.8869048,0.81349206,0.87896824,0.875,0.9781746
+FastaiMLP (tuned + ensembled),0.103174604,0.16964285,0.16269842,0.16468254,0.13293651,0.1656746,0.17460318,0.17857143,0.23015873,0.2827381,0.24305555,0.30555555,0.27579364,0.30357143,0.3482143,0.3313492,0.33531746,0.35416666,0.2936508,0.40873015,0.4375,0.41468254,0.44642857,0.41765872,0.4434524,0.37301588,0.41468254,0.46825397,0.47123015,0.4781746,0.47619048,0.5089286,0.44642857,0.5218254,0.50595236,,0.43650794,0.50793654,0.54761904,0.5555556,0.5119048,0.578373,0.60515875,0.6845238,0.60515875,0.57539684,0.6230159,0.56150794,0.6111111,0.6388889,0.64484125,0.6269841,0.70734125,0.66468257,0.6736111,0.75396824,0.71626985,0.75,0.7748016,0.74206346,0.74404764,0.74404764,0.78174603,0.74603176,0.7757937,0.9484127,0.7916667,0.8075397,0.78174603,0.8373016,0.86507934,0.87896824,0.97619045
+TorchMLP (tuned + ensembled),0.055555556,0.103174604,0.11309524,0.12301587,0.11309524,0.121031746,0.12301587,0.121031746,0.25992063,0.24404761,0.26785713,0.28174603,0.24801587,0.30257937,0.31746033,0.31547618,0.3234127,0.31944445,0.34325397,0.40873015,0.41666666,0.41269842,0.4265873,0.43055555,0.45436507,0.3472222,0.45039684,0.4265873,0.4642857,0.46130952,0.46626985,0.46825397,0.43452382,0.5218254,0.4781746,0.56349206,,0.52380955,0.51785713,0.5515873,0.5625,0.58928573,0.5555556,0.6617063,0.61904764,0.7380952,0.5873016,0.5654762,0.6230159,0.6527778,0.6617063,0.60714287,0.6964286,0.6944444,0.6805556,0.7380952,0.71626985,0.7430556,0.8531746,0.77876985,0.75,0.7757937,0.78373015,0.7638889,0.8194444,0.84623015,0.78571427,0.76984125,0.76785713,0.84325397,0.88095236,0.8730159,0.97619045
+ModernNCA (tuned + ensembled),0.055555556,0.11309524,0.09126984,0.13690476,0.16468254,0.121031746,0.09126984,0.12996031,0.26785713,0.22222222,0.28769842,0.29960316,0.32142857,0.26984128,0.3809524,0.3065476,0.3531746,0.3452381,0.37896827,0.4265873,0.45436507,0.40674603,0.43849206,0.45535713,0.4484127,0.4265873,0.42857143,0.4484127,0.5376984,0.5,0.5019841,0.52380955,0.46230158,0.59325397,0.5019841,0.4920635,0.47619048,,0.56646824,0.56349206,0.5515873,0.609127,0.5674603,0.5714286,0.5813492,0.58531743,0.59126985,0.5595238,0.6230159,0.59126985,0.59325397,0.5952381,0.66468257,0.71825397,0.68650794,0.6626984,0.67261904,0.64285713,0.6984127,0.6984127,0.6547619,0.7083333,0.65873015,0.80654764,0.71031743,0.7242063,0.7361111,0.7956349,0.78174603,0.70238096,0.78968257,0.76984125,0.9404762
+EBM (tuned),0.09325397,0.1468254,0.14880952,0.16071428,0.121031746,0.15178572,0.16170634,0.14880952,0.25496033,0.20436507,0.22023809,0.22420634,0.27380952,0.3045635,0.24206349,0.32539684,0.2797619,0.29563493,0.35714287,0.31349206,0.3313492,0.4434524,0.38789684,0.375,0.38492063,0.4107143,0.39087301,0.4027778,0.42261904,0.41468254,0.40873015,0.3204365,0.45436507,0.45436507,0.44246033,0.45238096,0.48214287,0.43353173,,0.62400794,0.4920635,0.515873,0.53968257,0.5367063,0.58531743,0.5793651,0.55654764,0.61210316,0.61904764,0.65873015,0.6656746,0.6269841,0.6706349,0.66468257,0.67261904,0.71130955,0.71031743,0.765873,0.71626985,0.7638889,0.74007934,0.7619048,0.78571427,0.7281746,0.78373015,0.7876984,0.8015873,0.82738096,0.8174603,0.82539684,0.875,0.89484125,0.98015875
+EBM (default),0.09920635,0.13392857,0.14285715,0.14087301,0.1388889,0.15873016,0.16865079,0.16468254,0.24404761,0.20238096,0.22123016,0.23115079,0.2361111,0.30753967,0.21825397,0.31746033,0.31746033,0.28174603,0.35119048,0.2638889,0.29960316,0.40873015,0.35714287,0.36507937,0.3611111,0.39781746,0.37103173,0.35714287,0.3888889,0.39484128,0.375,0.29960316,0.45337301,0.41468254,0.42261904,0.44444445,0.4484127,0.43650794,0.37599206,,0.4484127,0.49801588,0.53373015,0.51984125,0.56349206,0.5813492,0.54960316,0.609127,0.5714286,0.66071427,0.65873015,0.59325397,0.65674603,0.6031746,0.63492066,0.71825397,0.6825397,0.7480159,0.7083333,0.7638889,0.7361111,0.74007934,0.7827381,0.6924603,0.75992066,0.7956349,0.8313492,0.81349206,0.7956349,0.85714287,0.85119045,0.9126984,0.9781746
+iLTM (tuned),0.065476194,0.084325396,0.09126984,0.115079366,0.09623016,0.103174604,0.0952381,0.109126985,0.17757936,0.17261904,0.18849206,0.25,0.2281746,0.22023809,0.32936507,0.26984128,0.2857143,0.3234127,0.29563493,0.3313492,0.3611111,0.3690476,0.3452381,0.37698412,0.38392857,0.41170636,0.29464287,0.37896827,0.41468254,0.4047619,0.40674603,0.45436507,0.43551588,0.47718254,0.42063493,0.48809522,0.4375,0.4484127,0.50793654,0.5515873,,0.5456349,0.51984125,0.5416667,0.53968257,0.56349206,0.5644841,0.53373015,0.5873016,0.6299603,0.6230159,0.5873016,0.6527778,0.6369048,0.6468254,0.66765875,0.70238096,0.6319444,0.7638889,0.7430556,0.7063492,0.7043651,0.67460316,0.75396824,0.75396824,0.78571427,0.74404764,0.78373015,0.78571427,0.75595236,0.82738096,0.81845236,0.96428573
+RealMLP (default),0.08134921,0.097222224,0.13293651,0.11904762,0.10119048,0.13690476,0.13492064,0.13690476,0.20833333,0.1904762,0.15376984,0.21130952,0.22619048,0.2718254,0.25992063,0.29761904,0.25396827,0.2559524,0.30357143,0.31150794,0.32440478,0.4156746,0.33730158,0.32738096,0.37301588,0.31944445,0.33730158,0.3531746,0.42261904,0.3968254,0.3829365,0.4107143,0.3888889,0.46031746,0.42460316,0.42162699,0.4107143,0.39087301,0.48412699,0.5019841,0.45436507,,0.51785713,0.52380955,0.5456349,0.5555556,0.53571427,0.5515873,0.5873016,0.63492066,0.6507937,0.57539684,0.6468254,0.61507934,0.64484125,0.73511904,0.6805556,0.71031743,0.74900794,0.734127,0.7242063,0.74603176,0.7519841,0.7043651,0.79761904,0.8472222,0.79761904,0.7876984,0.78373015,0.8333333,0.875,0.86706346,0.9781746
+ChimeraBoost (tuned + ensembled),0.071428575,0.09126984,0.08730159,0.10714286,0.097222224,0.10813492,0.08928572,0.11805555,0.1984127,0.15277778,0.17857143,0.1904762,0.21527778,0.24404761,0.24900794,0.26587301,0.29166666,0.23809524,0.33730158,0.25297618,0.30555555,0.34920636,0.33531746,0.31944445,0.31746033,0.43452382,0.31547618,0.38194445,0.3561508,0.41666666,0.37003967,0.42063493,0.44246033,0.34325397,0.4513889,0.39484128,0.44444445,0.43253967,0.46031746,0.46626985,0.48015872,0.48214287,,0.46329364,0.49801588,0.5277778,0.56646824,0.5714286,0.5456349,0.640873,0.640873,0.75595236,0.60515875,0.6388889,0.70535713,0.6984127,0.65873015,0.6547619,0.6626984,0.73015875,0.73015875,0.72619045,0.6785714,0.7242063,0.76984125,0.7519841,0.73015875,0.7718254,0.75396824,0.7916667,0.82539684,0.83928573,0.9751984
+FastaiMLP (tuned),0.077380955,0.116071425,0.13492064,0.121031746,0.109126985,0.125,0.13293651,0.13492064,0.17261904,0.2093254,0.21924603,0.25,0.21825397,0.24801587,0.27579364,0.27380952,0.26587301,0.28373015,0.2638889,0.31944445,0.3611111,0.34126985,0.3670635,0.3472222,0.3611111,0.32539684,0.33035713,0.37599206,0.39285713,0.42261904,0.39880952,0.43551588,0.37103173,0.4563492,0.42261904,0.31547618,0.33829364,0.42857143,0.46329364,0.48015872,0.45833334,0.47619048,0.5367063,,0.5456349,0.49007937,0.55853176,0.5,0.56150794,0.5952381,0.59325397,0.58928573,0.6498016,0.59126985,0.60714287,0.6845238,0.6547619,0.6984127,0.7013889,0.6924603,0.7063492,0.6765873,0.7251984,0.67460316,0.7202381,0.8611111,0.7638889,0.75793654,0.73214287,0.79960316,0.8015873,0.8492063,0.96825397
+ExtraTrees (tuned + ensembled),0.061507937,0.07539683,0.0952381,0.1279762,0.13690476,0.14087301,0.11706349,0.13789682,0.16369048,0.15873016,0.20238096,0.26984128,0.29166666,0.21626984,0.27380952,0.2638889,0.26984128,0.31746033,0.31944445,0.31051588,0.3343254,0.31150794,0.30555555,0.34920636,0.35714287,0.3690476,0.33829364,0.32539684,0.36309522,0.39285713,0.37103173,0.36011904,0.3888889,0.4047619,0.3859127,0.39484128,0.3809524,0.4186508,0.41468254,0.43650794,0.46031746,0.45436507,0.5019841,0.45436507,,0.45734128,0.52380955,0.56150794,0.65674603,0.5297619,0.54761904,0.53373015,0.71626985,0.56349206,0.59325397,0.6269841,0.7757937,0.6319444,0.61507934,0.61904764,0.67757934,0.6547619,0.6458333,0.7063492,0.67261904,0.68650794,0.8452381,0.89285713,0.85119045,0.74404764,0.74404764,0.88095236,0.99007934
+TorchMLP (tuned),0.04761905,0.083333336,0.07936508,0.09126984,0.083333336,0.08928572,0.09126984,0.08928572,0.1984127,0.18650794,0.21428572,0.2013889,0.19444445,0.24007936,0.24206349,0.27380952,0.25396827,0.25099206,0.28373015,0.3015873,0.31646827,0.35714287,0.32142857,0.2857143,0.35912699,0.30753967,0.34126985,0.3343254,0.375,0.3809524,0.35912699,0.38789684,0.35119048,0.42261904,0.4047619,0.42460316,0.26190478,0.41468254,0.42063493,0.4186508,0.43650794,0.44444445,0.4722222,0.50992066,0.54265875,,0.5,0.49603173,0.54761904,0.6369048,0.6388889,0.53968257,0.61904764,0.57738096,0.58035713,0.71428573,0.6468254,0.64880955,0.76289684,0.70238096,0.7202381,0.6805556,0.67757934,0.6944444,0.7242063,0.7757937,0.7390873,0.72619045,0.7242063,0.8035714,0.81349206,0.8174603,0.96626985
+ChimeraBoost (tuned),0.061507937,0.09126984,0.08531746,0.09325397,0.0952381,0.10119048,0.08730159,0.10714286,0.18253969,0.14484127,0.16964285,0.16468254,0.20436507,0.23015873,0.23809524,0.2579365,0.26587301,0.24603175,0.30753967,0.24007936,0.28968254,0.3472222,0.32936507,0.3184524,0.31150794,0.40873015,0.31944445,0.35416666,0.33333334,0.37896827,0.34126985,0.40079364,0.42063493,0.33531746,0.4047619,0.37698412,0.41269842,0.40873015,0.4434524,0.45039684,0.43551588,0.4642857,0.43353173,0.44146827,0.47619048,0.5,,0.56150794,0.53174603,0.6269841,0.6269841,0.6924603,0.5813492,0.6220238,0.6706349,0.66468257,0.63293654,0.6369048,0.6547619,0.6984127,0.71626985,0.68154764,0.64880955,0.6964286,0.72321427,0.734127,0.7083333,0.75992066,0.73015875,0.7718254,0.8144841,0.81349206,0.96428573
+OrionMSP (default),0.06944445,0.08730159,0.084325396,0.09126984,0.07936508,0.109126985,0.1388889,0.115079366,0.1388889,0.17956349,0.2559524,0.28373015,0.2906746,0.1984127,0.28174603,0.18849206,0.3204365,0.31150794,0.15674603,0.30357143,0.31547618,0.25396827,0.35714287,0.3392857,0.30952382,0.39880952,0.36210316,0.35912699,0.3234127,0.36607143,0.37698412,0.3452381,0.39285713,0.3472222,0.38690478,0.43849206,0.43452382,0.44047618,0.38789684,0.39087301,0.46626985,0.4484127,0.42857143,0.5,0.43849206,0.50396824,0.43849206,,0.45039684,0.5535714,0.5555556,0.50595236,0.5,0.48412699,0.5555556,0.66071427,0.53968257,0.64880955,0.6111111,0.5972222,0.7083333,0.67460316,0.6755952,0.64880955,0.71626985,0.7361111,0.6765873,0.6686508,0.6805556,0.78968257,0.8055556,0.8174603,0.97619045
+ExtraTrees (tuned),0.071428575,0.077380955,0.109126985,0.12301587,0.125,0.1388889,0.1279762,0.13095239,0.14087301,0.16071428,0.1875,0.27380952,0.25992063,0.20039682,0.2579365,0.24404761,0.2579365,0.30952382,0.30555555,0.30555555,0.3125,0.2857143,0.2827381,0.3234127,0.30357143,0.3690476,0.30555555,0.31349206,0.32539684,0.34126985,0.35119048,0.3611111,0.36507937,0.36507937,0.3452381,0.3888889,0.37698412,0.37698412,0.3809524,0.42857143,0.41269842,0.41269842,0.45436507,0.43849206,0.34325397,0.45238096,0.46825397,0.54960316,,0.5327381,0.54761904,0.52380955,0.5992063,0.5,0.54960316,0.6230159,0.68154764,0.6111111,0.5952381,0.6111111,0.67460316,0.63095236,0.62103176,0.65674603,0.6527778,0.6765873,0.8075397,0.8075397,0.80257934,0.734127,0.7222222,0.88492066,0.9880952
+TabSTAR (tuned),0.17460318,0.1875,0.22222222,0.19642857,0.19642857,0.22222222,0.22420634,0.23015873,0.26984128,0.23511904,0.27380952,0.27380952,0.27579364,0.3015873,0.24801587,0.3125,0.28968254,0.2936508,0.2559524,0.2906746,0.30555555,0.3670635,0.3015873,0.32936507,0.33234128,0.3968254,0.35714287,0.3015873,0.31944445,0.32738096,0.30853173,0.32539684,0.30753967,0.34920636,0.34325397,0.3611111,0.3472222,0.40873015,0.34126985,0.3392857,0.37003967,0.36507937,0.35912699,0.4047619,0.4702381,0.36309522,0.37301588,0.44642857,0.4672619,,0.58035713,0.4027778,0.5119048,0.44444445,0.45436507,0.5218254,0.54960316,0.60714287,0.4722222,0.43253967,0.93849206,0.58928573,0.640873,0.5654762,0.60714287,0.6527778,0.7083333,0.68650794,0.70238096,0.75,0.640873,0.8492063,0.9543651
+TabSTAR (tuned + ensembled),0.17460318,0.18253969,0.22023809,0.1984127,0.19543651,0.21626984,0.21428572,0.22222222,0.26190478,0.24206349,0.26587301,0.26587301,0.26785713,0.29960316,0.24404761,0.31349206,0.28373015,0.2827381,0.25,0.28769842,0.29960316,0.3670635,0.30753967,0.32142857,0.33333334,0.4047619,0.3561508,0.29960316,0.31547618,0.32738096,0.30555555,0.3234127,0.29960316,0.34325397,0.3402778,0.35515872,0.33829364,0.40674603,0.3343254,0.34126985,0.37698412,0.34920636,0.35912699,0.40674603,0.45238096,0.3611111,0.37301588,0.44444445,0.45238096,0.41964287,,0.39484128,0.50595236,0.44047618,0.46230158,0.5119048,0.5456349,0.5952381,0.46825397,0.43849206,0.9543651,0.5833333,0.6369048,0.56150794,0.5992063,0.64880955,0.6944444,0.6785714,0.6984127,0.74206346,0.6388889,0.83928573,0.9543651
+ChimeraBoost (default),0.05357143,0.09126984,0.083333336,0.097222224,0.08730159,0.10119048,0.08134921,0.103174604,0.16865079,0.12896825,0.16865079,0.13293651,0.18055555,0.20634921,0.18253969,0.22023809,0.2361111,0.18551587,0.22420634,0.15873016,0.1984127,0.28968254,0.28968254,0.26785713,0.2936508,0.35912699,0.27281746,0.29761904,0.29761904,0.34424603,0.26984128,0.32936507,0.36507937,0.29960316,0.3809524,0.37301588,0.39285713,0.4047619,0.37301588,0.40674603,0.41269842,0.42460316,0.24404761,0.4107143,0.46626985,0.46031746,0.30753967,0.4940476,0.47619048,0.5972222,0.60515875,,0.54761904,0.5505952,0.5843254,0.57738096,0.57539684,0.62103176,0.59325397,0.65674603,0.6785714,0.63492066,0.6369048,0.640873,0.6666667,0.6626984,0.6438492,0.61507934,0.6785714,0.765873,0.75396824,0.82539684,0.9702381
+RandomForest (tuned + ensembled),0.043650795,0.07936508,0.077380955,0.08531746,0.11309524,0.097222224,0.09920635,0.10714286,0.125,0.103174604,0.11309524,0.21031746,0.19246031,0.14285715,0.20238096,0.20039682,0.19246031,0.22619048,0.2777778,0.20238096,0.24801587,0.28769842,0.1984127,0.25198412,0.24206349,0.30059522,0.23015873,0.20833333,0.31944445,0.22123016,0.25396827,0.29166666,0.28968254,0.32539684,0.24206349,0.29265872,0.30357143,0.33531746,0.32936507,0.34325397,0.3472222,0.3531746,0.39484128,0.35019842,0.28373015,0.3809524,0.4186508,0.5,0.40079364,0.48809522,0.4940476,0.45238096,,0.48214287,0.50595236,0.5535714,0.6656746,0.57539684,0.54761904,0.50793654,0.62400794,0.5793651,0.5813492,0.67261904,0.6111111,0.6130952,0.73214287,0.7757937,0.84325397,0.69345236,0.67460316,0.8373016,0.96825397
+XGBoost (default),0.041666668,0.041666668,0.05357143,0.0734127,0.06944445,0.05952381,0.04563492,0.065476194,0.13293651,0.10119048,0.08531746,0.11904762,0.12301587,0.15674603,0.19246031,0.20833333,0.16666667,0.16865079,0.26190478,0.12698413,0.15575397,0.31547618,0.1468254,0.24404761,0.23809524,0.32936507,0.19444445,0.15277778,0.3392857,0.12698413,0.23412699,0.29265872,0.33333334,0.3670635,0.16468254,0.33531746,0.30555555,0.28174603,0.33531746,0.3968254,0.36309522,0.38492063,0.3611111,0.40873015,0.43650794,0.42261904,0.37797618,0.515873,0.5,0.5555556,0.5595238,0.44940478,0.51785713,,0.55654764,0.60119045,0.5873016,0.59325397,0.59325397,0.6507937,0.66468257,0.5714286,0.61507934,0.6785714,0.62103176,0.6666667,0.71626985,0.7619048,0.7757937,0.7361111,0.7222222,0.7916667,0.9464286
+LightGBM (default),0.04563492,0.05952381,0.05357143,0.06349207,0.055555556,0.06349207,0.055555556,0.06349207,0.12698413,0.08730159,0.09920635,0.116071425,0.09920635,0.14880952,0.18154761,0.1845238,0.16269842,0.18253969,0.18948413,0.057539683,0.0922619,0.26190478,0.14484127,0.23809524,0.2361111,0.31150794,0.15873016,0.17063493,0.28769842,0.12698413,0.19642857,0.2857143,0.3452381,0.29960316,0.16269842,0.3263889,0.31944445,0.31349206,0.32738096,0.36507937,0.3531746,0.35515872,0.29464287,0.39285713,0.40674603,0.41964287,0.32936507,0.44444445,0.45039684,0.5456349,0.5376984,0.4156746,0.4940476,0.4434524,,0.5714286,0.53869045,0.5714286,0.5833333,0.61507934,0.640873,0.5654762,0.59325397,0.6369048,0.6230159,0.6468254,0.7043651,0.6845238,0.703373,0.734127,0.73214287,0.79960316,0.9632937
+TabFlex (default),0.0734127,0.083333336,0.103174604,0.10515873,0.08630952,0.109126985,0.10515873,0.11904762,0.09126984,0.18849206,0.20436507,0.20238096,0.1904762,0.17063493,0.2013889,0.17460318,0.22123016,0.21626984,0.11111111,0.25396827,0.26587301,0.15873016,0.30952382,0.23412699,0.21626984,0.28769842,0.23015873,0.31746033,0.1984127,0.30853173,0.29563493,0.26190478,0.30555555,0.2559524,0.31944445,0.24603175,0.26190478,0.33730158,0.28869048,0.28174603,0.33234128,0.26488096,0.3015873,0.31547618,0.37301588,0.2857143,0.33531746,0.3392857,0.37698412,0.4781746,0.48809522,0.42261904,0.44642857,0.39880952,0.42857143,,0.4642857,0.5376984,0.4652778,0.48015872,0.6269841,0.5277778,0.56845236,0.5277778,0.58531743,0.62896824,0.60714287,0.5555556,0.5992063,0.68849206,0.63492066,0.75396824,0.9424603
+RandomForest (tuned),0.04563492,0.057539683,0.0734127,0.08730159,0.065476194,0.08531746,0.077380955,0.09126984,0.121031746,0.08928572,0.08134921,0.15873016,0.15178572,0.12896825,0.19246031,0.17460318,0.16865079,0.20833333,0.21825397,0.14285715,0.17857143,0.2361111,0.14186507,0.22619048,0.19642857,0.28869048,0.18849206,0.16666667,0.25992063,0.19146825,0.21230158,0.27380952,0.27579364,0.25992063,0.19444445,0.28373015,0.28373015,0.32738096,0.28968254,0.31746033,0.29761904,0.31944445,0.34126985,0.3452381,0.22420634,0.3531746,0.3670635,0.46031746,0.3184524,0.45039684,0.45436507,0.42460316,0.3343254,0.41269842,0.46130952,0.53571427,,0.5575397,0.5,0.4781746,0.5972222,0.5555556,0.5654762,0.6230159,0.5952381,0.609127,0.6994048,0.7043651,0.7291667,0.6845238,0.6547619,0.82539684,0.9702381
+Linear (tuned + ensembled),0.06349207,0.0952381,0.11706349,0.08134921,0.09920635,0.10119048,0.115079366,0.115079366,0.15873016,0.1388889,0.15277778,0.14484127,0.15773809,0.21428572,0.121031746,0.22519842,0.21230158,0.16071428,0.23214285,0.18849206,0.19642857,0.30555555,0.22619048,0.2559524,0.20833333,0.24801587,0.29563493,0.23214285,0.20833333,0.24206349,0.2579365,0.20039682,0.2361111,0.2559524,0.2559524,0.25,0.25694445,0.35714287,0.23412699,0.25198412,0.36805555,0.28968254,0.3452381,0.3015873,0.36805555,0.35119048,0.36309522,0.35119048,0.3888889,0.39285713,0.4047619,0.37896827,0.42460316,0.40674603,0.42857143,0.46230158,0.44246033,,0.43452382,0.43849206,0.54960316,0.4781746,0.66964287,0.49603173,0.49801588,0.48809522,0.59126985,0.5555556,0.53968257,0.78571427,0.5714286,0.71626985,0.9047619
+TorchMLP (default),0.03968254,0.049603175,0.049603175,0.065476194,0.037698414,0.061507937,0.05952381,0.05357143,0.115079366,0.11111111,0.09920635,0.10714286,0.12301587,0.13095239,0.14087301,0.14880952,0.13293651,0.125,0.15873016,0.19444445,0.20833333,0.22619048,0.1984127,0.20436507,0.1765873,0.24801587,0.20039682,0.22023809,0.18055555,0.24404761,0.21230158,0.2579365,0.21825397,0.21626984,0.26785713,0.22519842,0.1468254,0.3015873,0.28373015,0.29166666,0.2361111,0.25099206,0.33730158,0.2986111,0.38492063,0.23710318,0.3452381,0.3888889,0.4047619,0.5277778,0.53174603,0.40674603,0.45238096,0.40674603,0.41666666,0.5347222,0.5,0.5654762,,0.48214287,0.6468254,0.53174603,0.59325397,0.5734127,0.5813492,0.6686508,0.6626984,0.6269841,0.61507934,0.7083333,0.67261904,0.7876984,0.9484127
+iLTM (default),0.06944445,0.061507937,0.06746032,0.083333336,0.06746032,0.077380955,0.077380955,0.08134921,0.13690476,0.10714286,0.109126985,0.11904762,0.13492064,0.14285715,0.10119048,0.16865079,0.15277778,0.13492064,0.18849206,0.12698413,0.14384921,0.20238096,0.17162699,0.18650794,0.21825397,0.2797619,0.1845238,0.18849206,0.20436507,0.23015873,0.13690476,0.19246031,0.2281746,0.21626984,0.23412699,0.2579365,0.22123016,0.3015873,0.2361111,0.2361111,0.25694445,0.26587301,0.26984128,0.30753967,0.3809524,0.29761904,0.3015873,0.4027778,0.3888889,0.5674603,0.56150794,0.34325397,0.4920635,0.34920636,0.38492063,0.51984125,0.5218254,0.56150794,0.51785713,,0.63492066,0.52380955,0.5833333,0.5555556,0.5456349,0.5813492,0.6706349,0.59325397,0.65873015,0.68650794,0.63095236,0.7916667,0.92460316
+TabSTAR (default),0.12896825,0.1468254,0.17460318,0.14484127,0.16269842,0.1845238,0.1845238,0.1845238,0.20039682,0.18948413,0.22023809,0.21626984,0.21825397,0.2281746,0.1875,0.23412699,0.23412699,0.2361111,0.1904762,0.21031746,0.22619048,0.26190478,0.24801587,0.24603175,0.2559524,0.26984128,0.25198412,0.24206349,0.25,0.26587301,0.23809524,0.24007936,0.23015873,0.26587301,0.26984128,0.2559524,0.25,0.3452381,0.25992063,0.2638889,0.2936508,0.27579364,0.26984128,0.2936508,0.32242063,0.2797619,0.28373015,0.29166666,0.32539684,0.061507937,0.04563492,0.32142857,0.37599206,0.33531746,0.35912699,0.37301588,0.4027778,0.45039684,0.3531746,0.36507937,,0.5019841,0.46626985,0.45436507,0.5138889,0.42857143,0.5654762,0.52380955,0.48015872,0.6111111,0.5515873,0.7063492,0.9424603
+PerpetualBooster (tuned + ensembled),0.017857144,0.06349207,0.055555556,0.057539683,0.05952381,0.057539683,0.06746032,0.07440476,0.115079366,0.0952381,0.077380955,0.10714286,0.08531746,0.13293651,0.14484127,0.15476191,0.13095239,0.109126985,0.12896825,0.12301587,0.13789682,0.20634921,0.18253969,0.1904762,0.1845238,0.23809524,0.17063493,0.19444445,0.21726191,0.22420634,0.20833333,0.21428572,0.24305555,0.22619048,0.25198412,0.2559524,0.22420634,0.29166666,0.23809524,0.25992063,0.29563493,0.25396827,0.27380952,0.3234127,0.3452381,0.31944445,0.3184524,0.32539684,0.3690476,0.4107143,0.41666666,0.36507937,0.42063493,0.42857143,0.43452382,0.4722222,0.44444445,0.5218254,0.46825397,0.47619048,0.49801588,,0.53968257,0.53174603,0.6875,0.5218254,0.55853176,0.58928573,0.5873016,0.62103176,0.7757937,0.6944444,0.9404762
+Linear (tuned),0.055555556,0.08730159,0.09126984,0.07539683,0.09126984,0.08531746,0.09126984,0.09920635,0.14880952,0.11706349,0.15079366,0.1279762,0.12996031,0.19246031,0.10119048,0.21031746,0.20337301,0.14880952,0.20634921,0.16666667,0.1736111,0.27380952,0.20833333,0.22619048,0.19246031,0.24404761,0.2559524,0.21726191,0.18650794,0.2281746,0.23412699,0.18253969,0.21527778,0.23115079,0.23015873,0.21825397,0.21626984,0.34126985,0.21428572,0.21726191,0.32539684,0.24801587,0.32142857,0.27480158,0.35416666,0.32242063,0.35119048,0.32440478,0.37896827,0.35912699,0.36309522,0.36309522,0.4186508,0.38492063,0.40674603,0.4315476,0.43452382,0.33035713,0.40674603,0.41666666,0.53373015,0.46031746,,0.4920635,0.48412699,0.4702381,0.55654764,0.53968257,0.5119048,0.6984127,0.5654762,0.6964286,0.8968254
+xRFM (default),0.035714287,0.031746034,0.05357143,0.05952381,0.051587302,0.05952381,0.05952381,0.057539683,0.11111111,0.065476194,0.08531746,0.14087301,0.12301587,0.10218254,0.21031746,0.1388889,0.1388889,0.1845238,0.19444445,0.19642857,0.20337301,0.19642857,0.17857143,0.14484127,0.06944445,0.23809524,0.14285715,0.19246031,0.11111111,0.20634921,0.23412699,0.24603175,0.2857143,0.26785713,0.21428572,0.25396827,0.2361111,0.19345239,0.2718254,0.30753967,0.24603175,0.29563493,0.27579364,0.32539684,0.2936508,0.30555555,0.30357143,0.35119048,0.34325397,0.43452382,0.43849206,0.35912699,0.32738096,0.32142857,0.36309522,0.4722222,0.37698412,0.50396824,0.4265873,0.44444445,0.5456349,0.46825397,0.50793654,,0.49503967,0.50595236,0.50595236,0.4940476,0.53373015,0.60515875,0.5873016,0.6527778,0.9047619
+PerpetualBooster (tuned),0.021825397,0.049603175,0.05059524,0.051587302,0.05357143,0.057539683,0.071428575,0.06349207,0.0952381,0.08531746,0.071428575,0.08730159,0.07936508,0.11805555,0.12301587,0.14285715,0.12896825,0.10119048,0.12896825,0.09920635,0.11904762,0.18253969,0.16269842,0.16865079,0.15178572,0.23313493,0.14484127,0.17063493,0.16964285,0.2013889,0.16666667,0.19146825,0.2361111,0.19444445,0.21031746,0.22420634,0.18055555,0.28968254,0.21626984,0.24007936,0.24603175,0.20238096,0.23015873,0.2797619,0.32738096,0.27579364,0.2767857,0.28373015,0.3472222,0.39285713,0.40079364,0.33333334,0.3888889,0.37896827,0.37698412,0.41468254,0.4047619,0.5019841,0.4186508,0.45436507,0.4861111,0.3125,0.515873,0.5049603,,0.4920635,0.5257937,0.54761904,0.5555556,0.5992063,0.6845238,0.67261904,0.91468257
+FastaiMLP (default),0.03373016,0.041666668,0.055555556,0.049603175,0.05357143,0.05952381,0.06349207,0.06746032,0.077380955,0.07936508,0.071428575,0.09920635,0.083333336,0.09920635,0.14087301,0.11309524,0.10019841,0.14484127,0.11309524,0.16269842,0.17063493,0.13492064,0.19246031,0.17956349,0.11706349,0.21825397,0.16269842,0.20238096,0.1388889,0.2281746,0.18650794,0.18055555,0.20833333,0.1904762,0.23214285,0.051587302,0.15376984,0.27579364,0.21230158,0.20436507,0.21428572,0.15277778,0.24801587,0.1388889,0.31349206,0.22420634,0.26587301,0.2638889,0.3234127,0.3472222,0.35119048,0.33730158,0.38690478,0.33333334,0.3531746,0.37103173,0.39087301,0.5119048,0.3313492,0.4186508,0.5714286,0.4781746,0.5297619,0.4940476,0.50793654,,0.57539684,0.5019841,0.52281743,0.6507937,0.61507934,0.74404764,0.9484127
+KNN (tuned + ensembled),0.035714287,0.057539683,0.06746032,0.057539683,0.097222224,0.083333336,0.06746032,0.083333336,0.08134921,0.10515873,0.10515873,0.12698413,0.125,0.09821428,0.121031746,0.10119048,0.15079366,0.15277778,0.13690476,0.12698413,0.14484127,0.125,0.13690476,0.18055555,0.13194445,0.28373015,0.17857143,0.1468254,0.1468254,0.18253969,0.15674603,0.17063493,0.19444445,0.15277778,0.18154761,0.20833333,0.21428572,0.2638889,0.1984127,0.16865079,0.2559524,0.20238096,0.26984128,0.2361111,0.15476191,0.2609127,0.29166666,0.3234127,0.19246031,0.29166666,0.30555555,0.3561508,0.26785713,0.28373015,0.29563493,0.39285713,0.30059522,0.40873015,0.33730158,0.32936507,0.43452382,0.44146827,0.4434524,0.4940476,0.47420636,0.42460316,,0.45436507,0.45436507,0.5734127,0.5416667,0.84623015,0.9702381
+ExtraTrees (default),0.011904762,0.025793651,0.04563492,0.049603175,0.04761905,0.049603175,0.04761905,0.055555556,0.046626985,0.05357143,0.05357143,0.09325397,0.07539683,0.061507937,0.1468254,0.0952381,0.13492064,0.14087301,0.15476191,0.08134921,0.0952381,0.109126985,0.109126985,0.13492064,0.06746032,0.23412699,0.10515873,0.121031746,0.125,0.13293651,0.1388889,0.16666667,0.1904762,0.18154761,0.13095239,0.19246031,0.23015873,0.20436507,0.17261904,0.18650794,0.21626984,0.21230158,0.2281746,0.24206349,0.10714286,0.27380952,0.24007936,0.3313492,0.19246031,0.31349206,0.32142857,0.38492063,0.22420634,0.23809524,0.31547618,0.44444445,0.29563493,0.44444445,0.37301588,0.40674603,0.47619048,0.4107143,0.46031746,0.50595236,0.45238096,0.49801588,0.5456349,,0.5297619,0.60119045,0.5119048,0.70238096,0.9623016
+RandomForest (default),0.01984127,0.021825397,0.043650795,0.03968254,0.04563492,0.051587302,0.03373016,0.05357143,0.04761905,0.04563492,0.041666668,0.09920635,0.08531746,0.06944445,0.121031746,0.09126984,0.109126985,0.125,0.11706349,0.071428575,0.097222224,0.14087301,0.08531746,0.15277778,0.06349207,0.22222222,0.10515873,0.0952381,0.12301587,0.097222224,0.13095239,0.16468254,0.17063493,0.1845238,0.11309524,0.21825397,0.23214285,0.21825397,0.18253969,0.20436507,0.21428572,0.21626984,0.24603175,0.26785713,0.14880952,0.27579364,0.26984128,0.31944445,0.19742064,0.29761904,0.3015873,0.32142857,0.15674603,0.22420634,0.29662699,0.40079364,0.27083334,0.46031746,0.38492063,0.34126985,0.51984125,0.41269842,0.48809522,0.46626985,0.44444445,0.47718254,0.5456349,0.4702381,,0.6031746,0.51984125,0.71031743,0.9484127
+Linear (default),0.03968254,0.051587302,0.06448413,0.048611112,0.04761905,0.07440476,0.06349207,0.066468254,0.06944445,0.109126985,0.12896825,0.10119048,0.0952381,0.13492064,0.06746032,0.1388889,0.16071428,0.11904762,0.10019841,0.103174604,0.11706349,0.15476191,0.16865079,0.17857143,0.15376984,0.21825397,0.19246031,0.1765873,0.13293651,0.17857143,0.15277778,0.14880952,0.17063493,0.14087301,0.18650794,0.16269842,0.15674603,0.29761904,0.17460318,0.14285715,0.24404761,0.16666667,0.20833333,0.20039682,0.2559524,0.19642857,0.2281746,0.21031746,0.26587301,0.25,0.2579365,0.23412699,0.3065476,0.2638889,0.26587301,0.31150794,0.31547618,0.21428572,0.29166666,0.31349206,0.3888889,0.37896827,0.3015873,0.39484128,0.40079364,0.34920636,0.4265873,0.39880952,0.3968254,,0.47718254,0.60515875,0.8492063
+PerpetualBooster (default),0.01984127,0.035714287,0.041666668,0.027777778,0.03373016,0.041666668,0.043650795,0.041666668,0.04563492,0.06349207,0.03968254,0.043650795,0.061507937,0.06944445,0.06944445,0.071428575,0.057539683,0.05357143,0.08134921,0.065476194,0.06746032,0.12301587,0.109126985,0.11706349,0.08928572,0.20238096,0.103174604,0.10119048,0.10119048,0.11904762,0.11706349,0.103174604,0.14087301,0.14285715,0.121031746,0.13492064,0.11904762,0.21031746,0.125,0.14880952,0.17261904,0.125,0.17460318,0.1984127,0.2559524,0.18650794,0.18551587,0.19444445,0.2777778,0.35912699,0.3611111,0.24603175,0.32539684,0.2777778,0.26785713,0.36507937,0.3452381,0.42857143,0.32738096,0.3690476,0.4484127,0.22420634,0.43452382,0.41269842,0.31547618,0.38492063,0.45833334,0.48809522,0.48015872,0.52281743,,0.61904764,0.8531746
+KNN (tuned),0.043650795,0.04761905,0.05357143,0.05357143,0.06349207,0.0734127,0.057539683,0.06746032,0.057539683,0.08531746,0.07936508,0.08928572,0.08531746,0.08730159,0.051587302,0.097222224,0.11904762,0.11111111,0.061507937,0.08234127,0.09325397,0.09920635,0.103174604,0.13293651,0.10119048,0.17460318,0.13392857,0.097222224,0.10119048,0.122023806,0.09126984,0.10615079,0.10714286,0.121031746,0.125,0.121031746,0.12698413,0.23015873,0.10515873,0.08730159,0.18154761,0.13293651,0.16071428,0.15079366,0.11904762,0.18253969,0.18650794,0.18253969,0.115079366,0.15079366,0.16071428,0.17460318,0.16269842,0.20833333,0.20039682,0.24603175,0.17460318,0.28373015,0.21230158,0.20833333,0.2936508,0.30555555,0.30357143,0.3472222,0.32738096,0.2559524,0.15376984,0.29761904,0.28968254,0.39484128,0.3809524,,0.93650794
+KNN (default),0.005952381,0.003968254,0.005952381,0.003968254,0.005952381,0.007936508,0.007936508,0.007936508,0.005952381,0.009920635,0.005952381,0.015873017,0.003968254,0.005952381,0.013888889,0.009920635,0.017857144,0.023809524,0.005952381,0.007936508,0.005952381,0.007936508,0.007936508,0.01984127,0.009920635,0.017857144,0.021825397,0.007936508,0.009920635,0.017857144,0.011904762,0.01984127,0.027777778,0.015873017,0.021825397,0.023809524,0.023809524,0.05952381,0.01984127,0.021825397,0.035714287,0.021825397,0.024801588,0.031746034,0.009920635,0.03373016,0.035714287,0.023809524,0.011904762,0.04563492,0.04563492,0.029761905,0.031746034,0.05357143,0.03670635,0.057539683,0.029761905,0.0952381,0.051587302,0.07539683,0.057539683,0.05952381,0.103174604,0.0952381,0.08531746,0.051587302,0.029761905,0.037698414,0.051587302,0.15079366,0.1468254,0.06349207,
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
index 6a90c06c3325bfcc03b22f967f3dcbab82d40575..2591fdf9a7bf2e9108c0ff8f423dbd197f07a6a7 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Multiclass | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/leaderboard_table.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..48c785bbf1e8aea358bc05bbfb3fc54288c48855
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_multiclass/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1740.0,"score":0.715,"rank":8.68,"hrank":5.03,"imp":11.792,"train_time":251.72,"predict_time":2.753,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+203\/-81","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1724.0,"score":0.712,"rank":9.26,"hrank":4.66,"imp":12.144,"train_time":2710.49,"predict_time":1.346,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+192\/-90","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1716.0,"score":0.71,"rank":9.58,"hrank":4.15,"imp":11.131,"train_time":2710.49,"predict_time":8.095,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+249\/-127","imputed":false},{"position":3,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1675.0,"score":0.809,"rank":11.28,"hrank":1.96,"imp":8.552,"train_time":55.13,"predict_time":7.685,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+384\/-230","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1645.0,"score":0.62,"rank":12.69,"hrank":7.92,"imp":14.638,"train_time":9.94,"predict_time":0.613,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+222\/-115","imputed":false},{"position":5,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1596.0,"score":0.57,"rank":15.24,"hrank":5.96,"imp":13.38,"train_time":11.66,"predict_time":0.489,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+232\/-146","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1588.0,"score":0.553,"rank":15.67,"hrank":6.4,"imp":13.732,"train_time":10.23,"predict_time":0.844,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+214\/-140","imputed":false},{"position":7,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1572.0,"score":0.539,"rank":16.58,"hrank":4.62,"imp":13.706,"train_time":44.33,"predict_time":7.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+232\/-144","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1560.0,"score":0.543,"rank":17.29,"hrank":6.71,"imp":15.793,"train_time":5.64,"predict_time":0.197,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+269\/-191","imputed":false},{"position":9,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1535.0,"score":0.388,"rank":18.82,"hrank":9.59,"imp":17.855,"train_time":6219.75,"predict_time":3.78,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-101","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1514.0,"score":0.44,"rank":20.15,"hrank":9.42,"imp":17.242,"train_time":4150.82,"predict_time":19.128,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+174\/-124","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1491.0,"score":0.421,"rank":21.61,"hrank":11.05,"imp":18.147,"train_time":4150.82,"predict_time":3.42,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+139\/-93","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1473.0,"score":0.339,"rank":22.83,"hrank":17.54,"imp":19.647,"train_time":1377.48,"predict_time":5.201,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-70","imputed":false},{"position":13,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1470.0,"score":0.388,"rank":23.07,"hrank":12.59,"imp":20.204,"train_time":3881.83,"predict_time":4.213,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+147\/-110","imputed":false},{"position":14,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1462.0,"score":0.324,"rank":23.61,"hrank":20.02,"imp":18.966,"train_time":2635.98,"predict_time":1.181,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-43","imputed":false},{"position":15,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1462.0,"score":0.323,"rank":23.64,"hrank":18.95,"imp":19.872,"train_time":1377.48,"predict_time":1.259,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-66","imputed":false},{"position":16,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1438.0,"score":0.279,"rank":25.3,"hrank":20.89,"imp":19.591,"train_time":2635.98,"predict_time":0.173,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-45","imputed":false},{"position":17,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1437.0,"score":0.334,"rank":25.34,"hrank":10.78,"imp":21.279,"train_time":5.46,"predict_time":0.271,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+221\/-134","imputed":false},{"position":18,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1430.0,"score":0.34,"rank":25.86,"hrank":14.64,"imp":20.97,"train_time":3881.83,"predict_time":0.698,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-97","imputed":false},{"position":19,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1427.0,"score":0.354,"rank":26.08,"hrank":13.44,"imp":22.225,"train_time":4635.17,"predict_time":212.235,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+176\/-132","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1417.0,"score":0.255,"rank":26.78,"hrank":20.19,"imp":20.295,"train_time":2210.17,"predict_time":0.974,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-73","imputed":false},{"position":21,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1412.0,"score":0.252,"rank":27.15,"hrank":21.04,"imp":20.385,"train_time":2210.17,"predict_time":3.984,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-75","imputed":false},{"position":22,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1402.0,"score":0.209,"rank":27.85,"hrank":20.55,"imp":20.21,"train_time":7128.67,"predict_time":9.211,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-76","imputed":false},{"position":23,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1398.0,"score":0.217,"rank":28.15,"hrank":14.54,"imp":20.477,"train_time":7128.67,"predict_time":0.776,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-56","imputed":false},{"position":24,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1396.0,"score":0.191,"rank":28.35,"hrank":22.99,"imp":20.967,"train_time":28234.45,"predict_time":841.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-79","imputed":false},{"position":25,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1382.0,"score":0.307,"rank":29.33,"hrank":20.31,"imp":23.18,"train_time":507.78,"predict_time":4.427,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+154\/-122","imputed":false},{"position":26,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1382.0,"score":0.299,"rank":29.36,"hrank":14.84,"imp":23.545,"train_time":4635.17,"predict_time":34.761,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+197\/-139","imputed":false},{"position":27,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1380.0,"score":0.272,"rank":29.53,"hrank":17.91,"imp":22.857,"train_time":1980.61,"predict_time":3.604,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+149\/-143","imputed":false},{"position":28,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1370.0,"score":0.226,"rank":30.22,"hrank":23.72,"imp":22.08,"train_time":27107.4,"predict_time":0.271,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-91","imputed":false},{"position":29,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1355.0,"score":0.264,"rank":31.38,"hrank":18.65,"imp":23.763,"train_time":12.83,"predict_time":0.673,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-132","imputed":false},{"position":30,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1345.0,"score":0.179,"rank":32.12,"hrank":27.56,"imp":21.974,"train_time":26.14,"predict_time":0.202,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-69","imputed":false},{"position":31,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1342.0,"score":0.362,"rank":32.39,"hrank":7.15,"imp":21.433,"train_time":22.52,"predict_time":1.681,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+198\/-272","imputed":false},{"position":32,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1339.0,"score":0.247,"rank":32.59,"hrank":15.61,"imp":24.352,"train_time":984.66,"predict_time":1.487,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+186\/-143","imputed":false},{"position":33,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1330.0,"score":0.268,"rank":33.3,"hrank":13.81,"imp":24.143,"train_time":738.02,"predict_time":1.862,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+178\/-174","imputed":false},{"position":34,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1327.0,"score":0.171,"rank":33.54,"hrank":26.41,"imp":22.786,"train_time":27107.4,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-92","imputed":false},{"position":35,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1314.0,"score":0.127,"rank":34.52,"hrank":26.86,"imp":23.442,"train_time":3902.08,"predict_time":2.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+111\/-116","imputed":false},{"position":36,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1299.0,"score":0.107,"rank":35.68,"hrank":31.05,"imp":22.605,"train_time":28234.45,"predict_time":276.786,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-82","imputed":false},{"position":37,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1294.0,"score":0.142,"rank":36.03,"hrank":27.56,"imp":24.388,"train_time":13.33,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-117","imputed":false},{"position":38,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1287.0,"score":0.175,"rank":36.63,"hrank":19.46,"imp":25.105,"train_time":1980.61,"predict_time":0.61,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-119","imputed":false},{"position":39,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1286.0,"score":0.202,"rank":36.69,"hrank":13.73,"imp":25.956,"train_time":984.66,"predict_time":0.279,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-142","imputed":false},{"position":40,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1280.0,"score":0.217,"rank":37.15,"hrank":11.66,"imp":25.405,"train_time":738.02,"predict_time":0.316,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+192\/-191","imputed":false},{"position":41,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1277.0,"score":0.134,"rank":37.36,"hrank":31.28,"imp":25.683,"train_time":4.32,"predict_time":0.084,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-112","imputed":false},{"position":42,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1271.0,"score":0.208,"rank":37.79,"hrank":14.88,"imp":26.682,"train_time":2644.99,"predict_time":12.291,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+174\/-181","imputed":false},{"position":43,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1267.0,"score":0.166,"rank":38.12,"hrank":16.0,"imp":26.042,"train_time":43.39,"predict_time":42.555,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+214\/-134","imputed":false},{"position":44,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1263.0,"score":0.047,"rank":38.42,"hrank":34.69,"imp":23.424,"train_time":4.72,"predict_time":0.542,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-65","imputed":false},{"position":45,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1260.0,"score":0.156,"rank":38.64,"hrank":30.29,"imp":24.35,"train_time":1599.92,"predict_time":0.205,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-133","imputed":false},{"position":46,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1253.0,"score":0.108,"rank":39.21,"hrank":28.7,"imp":25.527,"train_time":3902.08,"predict_time":0.223,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+107\/-138","imputed":false},{"position":47,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1252.0,"score":0.146,"rank":39.24,"hrank":33.53,"imp":24.414,"train_time":1599.92,"predict_time":1.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-125","imputed":false},{"position":48,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1243.0,"score":0.142,"rank":39.95,"hrank":29.9,"imp":25.298,"train_time":5.04,"predict_time":0.429,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+151\/-128","imputed":false},{"position":49,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1219.0,"score":0.11,"rank":41.66,"hrank":24.26,"imp":27.76,"train_time":2644.99,"predict_time":1.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-168","imputed":false},{"position":50,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1203.0,"score":0.022,"rank":42.84,"hrank":35.45,"imp":28.123,"train_time":17.95,"predict_time":0.691,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-115","imputed":false},{"position":51,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1167.0,"score":0.113,"rank":45.39,"hrank":35.75,"imp":27.473,"train_time":7.2,"predict_time":0.19,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-172","imputed":false},{"position":52,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1161.0,"score":0.0,"rank":45.79,"hrank":41.97,"imp":28.684,"train_time":19.2,"predict_time":7.62,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-105","imputed":false},{"position":53,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1096.0,"score":0.057,"rank":50.08,"hrank":43.18,"imp":32.151,"train_time":9.43,"predict_time":1.675,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-202","imputed":false},{"position":54,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1071.0,"score":0.0,"rank":51.61,"hrank":39.61,"imp":31.36,"train_time":13.69,"predict_time":0.314,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-130","imputed":false},{"position":55,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1057.0,"score":0.012,"rank":52.46,"hrank":49.93,"imp":29.543,"train_time":514.72,"predict_time":211.976,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-137","imputed":false},{"position":56,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1014.0,"score":0.0,"rank":54.96,"hrank":51.69,"imp":34.712,"train_time":21.06,"predict_time":3.367,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-132","imputed":false},{"position":57,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1013.0,"score":0.063,"rank":55.01,"hrank":26.82,"imp":41.417,"train_time":151.75,"predict_time":2.345,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+225\/-315","imputed":false},{"position":58,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1008.0,"score":0.0,"rank":55.3,"hrank":49.73,"imp":33.151,"train_time":40588.15,"predict_time":31.228,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-189","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1002.0,"score":0.0,"rank":55.62,"hrank":51.66,"imp":33.262,"train_time":40588.15,"predict_time":8.466,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-184","imputed":false},{"position":60,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":55.71,"hrank":51.41,"imp":37.958,"train_time":0.74,"predict_time":0.137,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-161","imputed":false},{"position":61,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":990.0,"score":0.0,"rank":56.23,"hrank":52.7,"imp":42.263,"train_time":893.09,"predict_time":0.818,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-260","imputed":false},{"position":62,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":957.0,"score":0.0,"rank":57.97,"hrank":54.38,"imp":43.118,"train_time":893.09,"predict_time":0.146,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-258","imputed":false},{"position":63,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":951.0,"score":0.0,"rank":58.25,"hrank":55.57,"imp":39.231,"train_time":4.85,"predict_time":0.786,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-160","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":943.0,"score":0.004,"rank":58.65,"hrank":55.55,"imp":36.711,"train_time":8.52,"predict_time":0.734,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-169","imputed":false},{"position":65,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":899.0,"score":0.0,"rank":60.68,"hrank":53.35,"imp":51.795,"train_time":1409.51,"predict_time":10.372,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-206","imputed":false},{"position":66,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":862.0,"score":0.0,"rank":62.22,"hrank":61.7,"imp":41.486,"train_time":5.06,"predict_time":1.013,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-155","imputed":false},{"position":67,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":849.0,"score":0.0,"rank":62.76,"hrank":39.05,"imp":53.554,"train_time":1409.51,"predict_time":3.696,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-221","imputed":false},{"position":68,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":845.0,"score":0.0,"rank":62.89,"hrank":61.07,"imp":41.778,"train_time":724.31,"predict_time":8.133,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-212","imputed":false},{"position":69,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":843.0,"score":0.052,"rank":62.97,"hrank":45.8,"imp":51.156,"train_time":151.75,"predict_time":0.258,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+269\/-418","imputed":false},{"position":70,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":820.0,"score":0.0,"rank":63.83,"hrank":33.08,"imp":57.836,"train_time":228.95,"predict_time":0.465,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+142\/-295","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":757.0,"score":0.0,"rank":65.86,"hrank":65.18,"imp":48.963,"train_time":4.53,"predict_time":0.164,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-240","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":359.0,"score":0.0,"rank":72.21,"hrank":72.18,"imp":71.574,"train_time":0.52,"predict_time":0.076,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-311","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/pareto_front_explorer.html
index 6ef56fb52b152110ff53cd718fedc501914a7a95..0cdd9c43050d2d3602abacf8f0a055469c627b03 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.4269406194,"imp":23.1804670281,"elo":1382.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2017044414,"imp":21.9737924852,"elo":1345.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1731293093,"imp":19.5913374354,"elo":1437.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.1808204254,"imp":18.9664656013,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1904913614,"imp":27.4728327418,"elo":1166.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2050159281,"imp":24.3504004446,"elo":1260.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2759753363,"imp":24.414281846,"elo":1252.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0400823469,"imp":24.3883900218,"elo":1294.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0510244193,"imp":22.7855804711,"elo":1326.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.2710569482,"imp":22.0796722505,"elo":1370.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7862858499,"imp":39.2314475974,"elo":951.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2793367209,"imp":25.9556664895,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4872412963,"imp":24.3524719429,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.733891056,"imp":36.7113028113,"elo":942.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1478371705,"imp":27.7597879934,"elo":1219.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.2912040793,"imp":26.6820155225,"elo":1271.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.4290808604,"imp":25.2984920996,"elo":1242.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.2591317348,"imp":19.8717177086,"elo":1461.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":5.2006257965,"imp":19.6468709936,"elo":1473.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.0619559967,"imp":13.7060392024,"elo":1572.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.690602132,"imp":28.1225893675,"elo":1203.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7756352585,"imp":20.4770913101,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2105111376,"imp":20.2095862786,"elo":1402.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3671495151,"imp":34.7118271972,"elo":1013.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4648882862,"imp":57.8364658322,"elo":819.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":3.6964435846,"imp":53.5544441849,"elo":848.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":10.3718427289,"imp":51.795050359,"elo":898.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1372541163,"imp":37.9575943939,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3155208059,"imp":25.4052459794,"elo":1279.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.8624917599,"imp":24.1427584977,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.6199152504,"imp":28.6838156253,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.4203022504,"imp":18.1471842853,"elo":1491.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":19.128322436,"imp":17.2420283118,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8441387842,"imp":13.7319645423,"elo":1588.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3463854051,"imp":12.1435446729,"elo":1724.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0946622325,"imp":11.1309737105,"elo":1716.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.681335577,"imp":21.432502252,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.5545582952,"imp":26.0415742159,"elo":1267.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":34.7609767403,"imp":23.5450184967,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":212.2351742215,"imp":22.224851633,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2707653455,"imp":21.2793904952,"elo":1437.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.6845942368,"imp":8.5521873572,"elo":1675.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":1.0132382115,"imp":41.4862018567,"elo":862.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1967767849,"imp":15.7931935941,"elo":1560.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.6734553632,"imp":23.7629557845,"elo":1355.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.6978379022,"imp":20.9703726876,"elo":1430.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.2126633842,"imp":20.2042225729,"elo":1469.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6126587114,"imp":14.6378268767,"elo":1645.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4885714957,"imp":13.3796951947,"elo":1595.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1330110917,"imp":41.7777639065,"elo":845.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.4659125188,"imp":33.2616589182,"elo":1001.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.2276662759,"imp":33.1505859182,"elo":1007.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0836134922,"imp":25.6833361288,"elo":1277.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.314314201,"imp":31.3595678387,"elo":1071.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.22258711,"imp":25.5269342463,"elo":1252.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0277285638,"imp":23.441579713,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.5423707001,"imp":23.4236127042,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.9738781112,"imp":20.2952305363,"elo":1417.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.9840757918,"imp":20.3847587002,"elo":1412.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":211.9764977363,"imp":29.5433106817,"elo":1057.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":276.7864889801,"imp":22.6045383713,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"imp":20.9671843589,"elo":1395.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.6749433874,"imp":32.1514870525,"elo":1095.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.6100366829,"imp":25.1049033831,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.603859777,"imp":22.8570853158,"elo":1379.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.4269406194,"x_train":507.7823899342,"imp":23.1804670281,"elo":1382.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2017044414,"x_train":26.1374278937,"imp":21.9737924852,"elo":1345.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1731293093,"x_train":2635.9752136464,"imp":19.5913374354,"elo":1437.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.1808204254,"x_train":2635.9752136464,"imp":18.9664656013,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1904913614,"x_train":7.2011684966,"imp":27.4728327418,"elo":1166.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2050159281,"x_train":1599.9245048672,"imp":24.3504004446,"elo":1260.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2759753363,"x_train":1599.9245048672,"imp":24.414281846,"elo":1252.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0400823469,"x_train":13.333580501,"imp":24.3883900218,"elo":1294.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0510244193,"x_train":27107.3972485932,"imp":22.7855804711,"elo":1326.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.2710569482,"x_train":27107.3972485932,"imp":22.0796722505,"elo":1370.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7862858499,"x_train":4.8501982351,"imp":39.2314475974,"elo":951.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2793367209,"x_train":984.6610737813,"imp":25.9556664895,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4872412963,"x_train":984.6610737813,"imp":24.3524719429,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.733891056,"x_train":8.5155691544,"imp":36.7113028113,"elo":942.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1478371705,"x_train":2644.9909431875,"imp":27.7597879934,"elo":1219.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.2912040793,"x_train":2644.9909431875,"imp":26.6820155225,"elo":1271.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.4290808604,"x_train":5.0434629122,"imp":25.2984920996,"elo":1242.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.2591317348,"x_train":1377.4770257446,"imp":19.8717177086,"elo":1461.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":5.2006257965,"x_train":1377.4770257446,"imp":19.6468709936,"elo":1473.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.0619559967,"x_train":44.325578779,"imp":13.7060392024,"elo":1572.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.690602132,"x_train":17.9457268623,"imp":28.1225893675,"elo":1203.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7756352585,"x_train":7128.6715875046,"imp":20.4770913101,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2105111376,"x_train":7128.6715875046,"imp":20.2095862786,"elo":1402.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3671495151,"x_train":21.0623624205,"imp":34.7118271972,"elo":1013.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4648882862,"x_train":228.9454185462,"imp":57.8364658322,"elo":819.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":3.6964435846,"x_train":1409.5097915247,"imp":53.5544441849,"elo":848.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":10.3718427289,"x_train":1409.5097915247,"imp":51.795050359,"elo":898.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1372541163,"x_train":0.7350357633,"imp":37.9575943939,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3155208059,"x_train":738.0209327471,"imp":25.4052459794,"elo":1279.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.8624917599,"x_train":738.0209327471,"imp":24.1427584977,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.6199152504,"x_train":19.2044919241,"imp":28.6838156253,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.4203022504,"x_train":4150.8158923996,"imp":18.1471842853,"elo":1491.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":19.128322436,"x_train":4150.8158923996,"imp":17.2420283118,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8441387842,"x_train":10.2256680645,"imp":13.7319645423,"elo":1588.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3463854051,"x_train":2710.4880101354,"imp":12.1435446729,"elo":1724.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0946622325,"x_train":2710.4880101354,"imp":11.1309737105,"elo":1716.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.681335577,"x_train":22.5239093941,"imp":21.432502252,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.5545582952,"x_train":43.394404174,"imp":26.0415742159,"elo":1267.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":34.7609767403,"x_train":4635.1714883924,"imp":23.5450184967,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":212.2351742215,"x_train":4635.1714883924,"imp":22.224851633,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2707653455,"x_train":5.4556230963,"imp":21.2793904952,"elo":1437.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.6845942368,"x_train":55.129309117,"imp":8.5521873572,"elo":1675.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":1.0132382115,"x_train":5.0558157998,"imp":41.4862018567,"elo":862.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1967767849,"x_train":5.6364435241,"imp":15.7931935941,"elo":1560.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.6734553632,"x_train":12.8284122123,"imp":23.7629557845,"elo":1355.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.6978379022,"x_train":3881.8318340105,"imp":20.9703726876,"elo":1430.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.2126633842,"x_train":3881.8318340105,"imp":20.2042225729,"elo":1469.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6126587114,"x_train":9.9408095472,"imp":14.6378268767,"elo":1645.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4885714957,"x_train":11.6635229296,"imp":13.3796951947,"elo":1595.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1330110917,"x_train":724.3138183772,"imp":41.7777639065,"elo":845.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.4659125188,"x_train":40588.1540174129,"imp":33.2616589182,"elo":1001.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.2276662759,"x_train":40588.1540174129,"imp":33.1505859182,"elo":1007.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0836134922,"x_train":4.3206513889,"imp":25.6833361288,"elo":1277.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.314314201,"x_train":13.688894957,"imp":31.3595678387,"elo":1071.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.22258711,"x_train":3902.0763863206,"imp":25.5269342463,"elo":1252.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0277285638,"x_train":3902.0763863206,"imp":23.441579713,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.5423707001,"x_train":4.7244136606,"imp":23.4236127042,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.9738781112,"x_train":2210.16539451,"imp":20.2952305363,"elo":1417.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.9840757918,"x_train":2210.16539451,"imp":20.3847587002,"elo":1412.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":211.9764977363,"x_train":514.7174339635,"imp":29.5433106817,"elo":1057.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":276.7864889801,"x_train":28234.4532410041,"imp":22.6045383713,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"x_train":28234.4532410041,"imp":20.9671843589,"elo":1395.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.6749433874,"x_train":9.4251823487,"imp":32.1514870525,"elo":1095.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.6100366829,"x_train":1980.6125379489,"imp":25.1049033831,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.603859777,"x_train":1980.6125379489,"imp":22.8570853158,"elo":1379.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
index 9638baba3615e2e413d9e51792e45b3546757c02..470320591c30315b5e863b381246254fbeb749cc 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":507.7823899342,"imp":23.2114545998,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":26.1374278937,"imp":22.0039836269,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":49.1954460239,"imp":20.4262604412,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":117.0085419585,"imp":19.7576715507,"elo":1438.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":233.7248660534,"imp":19.424885541,"elo":1452.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":454.7581834642,"imp":19.1203843406,"elo":1467.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":753.831728936,"imp":19.100256018,"elo":1467.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1385.7187287572,"imp":19.0360559005,"elo":1470.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2005.179917238,"imp":19.0302536594,"elo":1471.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":2635.9752136464,"imp":18.9972009529,"elo":1471.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":7.2011684966,"imp":27.5033257289,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":15.101549461,"imp":27.0263572198,"elo":1177.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":38.0760065843,"imp":26.6347229635,"elo":1189.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":76.9297326944,"imp":26.1726859259,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":196.6496559406,"imp":25.4239983125,"elo":1220.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":396.6608033968,"imp":25.0952993451,"elo":1230.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":796.7427370653,"imp":24.6867721945,"elo":1241.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1194.3788869413,"imp":24.53147173,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1599.9245048672,"imp":24.4456483013,"elo":1251.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":13.333580501,"imp":24.4188118821,"elo":1290.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":155.2301034791,"imp":24.0200743235,"elo":1301.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":423.1216867301,"imp":23.3164694437,"elo":1327.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":1126.0038118286,"imp":23.0769468215,"elo":1335.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":3121.6567594042,"imp":22.6886608071,"elo":1348.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":6636.6428563278,"imp":22.4447970632,"elo":1358.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":13596.8410765133,"imp":22.2533925416,"elo":1366.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":20505.2305784284,"imp":22.1674210825,"elo":1371.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":27107.3972485932,"imp":22.1115412421,"elo":1373.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.8501982351,"imp":39.253213975,"elo":941.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.7103499513,"imp":34.8742403411,"elo":1025.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.7095068985,"imp":28.3372923926,"elo":1153.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":49.0400181801,"imp":26.4342560518,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":122.0237750052,"imp":24.9708088144,"elo":1318.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":244.0803421306,"imp":24.7563738246,"elo":1327.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":488.408774127,"imp":24.5616585216,"elo":1335.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":733.7086039727,"imp":24.4319450711,"elo":1342.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":984.6610737813,"imp":24.3757005775,"elo":1344.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.5155691544,"imp":36.7405790294,"elo":938.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":21.559937581,"imp":34.8581291908,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":62.1360327566,"imp":31.3600691517,"elo":1114.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":125.9632889407,"imp":29.7612044964,"elo":1174.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":322.1066821739,"imp":28.1490155003,"elo":1233.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":651.6805621192,"imp":27.4181493683,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1305.7711888267,"imp":27.0395054438,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1966.2811643084,"imp":26.8443056915,"elo":1277.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2644.9909431875,"imp":26.7106266064,"elo":1280.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":5.0434629122,"imp":25.3292162618,"elo":1242.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":12.7823939958,"imp":24.1737947226,"elo":1296.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":32.375532788,"imp":23.0326922597,"elo":1344.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":66.0618660449,"imp":22.2235794539,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":169.9915970428,"imp":20.9475933773,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":340.9497698538,"imp":20.3436893496,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":684.7234530331,"imp":19.960683987,"elo":1472.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":1025.1619507673,"imp":19.8274646878,"elo":1478.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1377.4770257446,"imp":19.6765226725,"elo":1486.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":44.325578779,"imp":13.7801247025,"elo":1577.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":17.9457268623,"imp":28.1506397508,"elo":1200.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":57.1222738852,"imp":25.6633338737,"elo":1243.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":171.10400172,"imp":22.7753474268,"elo":1331.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":346.9635750688,"imp":21.9804246796,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":858.4265706116,"imp":21.088841993,"elo":1400.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1767.2813809485,"imp":20.7360381544,"elo":1408.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3533.1581339321,"imp":20.6670198053,"elo":1408.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":5301.5364312197,"imp":20.3926035696,"elo":1408.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":7128.6715875046,"imp":20.2408220747,"elo":1406.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":21.0623624205,"imp":34.8518982888,"elo":1005.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":228.9454185462,"imp":57.9067657976,"elo":816.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":466.150787739,"imp":56.34720463,"elo":845.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1174.4301363506,"imp":52.2472818,"elo":885.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1409.5097915247,"imp":51.807304947,"elo":894.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.7350357633,"imp":38.0436910874,"elo":996.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.5934411859,"imp":30.728569383,"elo":1105.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":15.6940857589,"imp":27.2882132656,"elo":1176.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":34.1407666441,"imp":25.6524426584,"elo":1229.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":89.3035790361,"imp":24.6303873802,"elo":1312.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":181.3479112981,"imp":24.4340751577,"elo":1323.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":365.5123989512,"imp":24.2898978553,"elo":1331.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":549.883689003,"imp":24.2258530991,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":738.0209327471,"imp":24.1666716364,"elo":1336.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":19.2044919241,"imp":28.7154870283,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":41.019845821,"imp":23.7678228618,"elo":1318.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":103.1070784858,"imp":20.8883523221,"elo":1419.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":206.5111037718,"imp":19.7602119139,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":515.981423509,"imp":18.8542071974,"elo":1488.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1035.4018073353,"imp":18.3452431894,"elo":1508.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2068.879284709,"imp":17.8102033233,"elo":1518.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3101.33807397,"imp":17.5276091319,"elo":1525.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4150.8158923996,"imp":17.2746068522,"elo":1528.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":10.2256680645,"imp":13.7665009799,"elo":1589.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":22.8990925067,"imp":13.9372216378,"elo":1598.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":62.9203931706,"imp":13.6662217842,"elo":1628.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":129.5212073934,"imp":13.3752599018,"elo":1680.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":331.7365286813,"imp":12.9850993487,"elo":1711.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":668.706695891,"imp":12.2239269317,"elo":1729.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1349.1531554668,"imp":11.8303326538,"elo":1727.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2024.0527372009,"imp":11.5296034873,"elo":1726.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2710.4880101354,"imp":11.1683966135,"elo":1723.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":22.5239093941,"imp":21.4632789806,"elo":1341.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":43.394404174,"imp":26.0629035743,"elo":1264.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.8442293296,"imp":25.7044576611,"elo":1283.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":122.0779658592,"imp":25.0275066703,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":246.0408977706,"imp":24.530903334,"elo":1351.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":571.7774981312,"imp":23.701818962,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1153.6863993273,"imp":23.1512161992,"elo":1401.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2305.5325709939,"imp":22.7008846365,"elo":1421.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3450.9693523062,"imp":22.4459528636,"elo":1429.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4635.1714883924,"imp":22.2502714432,"elo":1434.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":5.4556230963,"imp":21.3039429088,"elo":1444.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":55.129309117,"imp":8.5853827398,"elo":1675.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":5.0558157998,"imp":41.4995236172,"elo":853.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":5.6364435241,"imp":15.8274664818,"elo":1568.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":12.8284122123,"imp":23.7957120917,"elo":1351.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":30.6749168115,"imp":23.0296619698,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":92.4143274115,"imp":22.0494339117,"elo":1410.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":190.6444632314,"imp":21.4653911458,"elo":1431.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":476.8006390989,"imp":21.0018232613,"elo":1448.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":956.8892667394,"imp":20.6612676764,"elo":1464.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1912.1428059667,"imp":20.4244533212,"elo":1472.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2886.0992294917,"imp":20.3304715772,"elo":1475.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3881.8318340105,"imp":20.2364543206,"elo":1480.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":9.9408095472,"imp":14.6722580868,"elo":1661.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":11.6635229296,"imp":13.4121206479,"elo":1600.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":724.3138183772,"imp":41.8051180363,"elo":834.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":912.4748239301,"imp":41.5778126111,"elo":847.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1342.7568425834,"imp":41.1117061849,"elo":859.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2054.553219683,"imp":40.1252719699,"elo":884.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4867.9697698991,"imp":37.9267538233,"elo":921.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":9985.461750605,"imp":35.6598832163,"elo":953.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":20581.7115313909,"imp":34.3968313611,"elo":979.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":30569.2328830115,"imp":33.6984728566,"elo":998.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":40588.1540174129,"imp":33.180063577,"elo":1014.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":4.3206513889,"imp":25.7080382962,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.688894957,"imp":31.3878760718,"elo":1065.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":28.9516297239,"imp":29.111751553,"elo":1132.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":95.1897765055,"imp":27.1913731454,"elo":1197.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":204.1284404869,"imp":25.9297244428,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":488.13995387,"imp":24.9137275839,"elo":1276.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":949.9701252728,"imp":24.3638116785,"elo":1293.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1876.0678595078,"imp":23.9569410027,"elo":1303.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2872.3862481998,"imp":23.7168404483,"elo":1312.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3902.0763863206,"imp":23.4748402271,"elo":1318.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":4.7244136606,"imp":23.4532735312,"elo":1256.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":15.9463545731,"imp":22.2940252417,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":50.0714874559,"imp":21.863444531,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":100.9432671475,"imp":21.4472656691,"elo":1360.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":268.6777580855,"imp":21.1227818953,"elo":1380.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":544.7091931075,"imp":20.8973672866,"elo":1393.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":1096.7820562853,"imp":20.7149050758,"elo":1404.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":1644.1757072757,"imp":20.5790500324,"elo":1412.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":2210.16539451,"imp":20.415800911,"elo":1423.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":514.7174339635,"imp":29.571173745,"elo":1049.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1597.2950547029,"imp":26.921553965,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":4783.0314636057,"imp":24.2410153143,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":10691.0627275524,"imp":22.9123031059,"elo":1317.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":26917.4383401199,"imp":21.0384055081,"elo":1399.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":28234.4532410041,"imp":20.9993995911,"elo":1403.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":9.4251823487,"imp":32.1768735408,"elo":1088.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":19.1707317375,"imp":29.8720596025,"elo":1167.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":47.812888724,"imp":26.9358200158,"elo":1272.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":97.2172107229,"imp":25.359048208,"elo":1309.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":246.1477821059,"imp":24.1100301403,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":491.5550544333,"imp":23.5593254116,"elo":1366.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":987.2594279642,"imp":23.1640741686,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1477.605852992,"imp":22.9995725755,"elo":1382.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1980.6125379489,"imp":22.8867058478,"elo":1386.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.4269406194,"x_train":507.7823899342,"imp":23.2114545998,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2017044414,"x_train":26.1374278937,"imp":22.0039836269,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3433286258,"x_train":49.1954460239,"imp":20.4262604412,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.520712217,"x_train":117.0085419585,"imp":19.7576715507,"elo":1438.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.6542443518,"x_train":233.7248660534,"imp":19.424885541,"elo":1452.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.8258809333,"x_train":454.7581834642,"imp":19.1203843406,"elo":1467.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.9640813791,"x_train":753.831728936,"imp":19.100256018,"elo":1467.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.0748368064,"x_train":1385.7187287572,"imp":19.0360559005,"elo":1470.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1178203711,"x_train":2005.179917238,"imp":19.0302536594,"elo":1471.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1808204254,"x_train":2635.9752136464,"imp":18.9972009529,"elo":1471.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1904913614,"x_train":7.2011684966,"imp":27.5033257289,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3136490334,"x_train":15.101549461,"imp":27.0263572198,"elo":1177.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4619268878,"x_train":38.0760065843,"imp":26.6347229635,"elo":1189.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.599982434,"x_train":76.9297326944,"imp":26.1726859259,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.7770406751,"x_train":196.6496559406,"imp":25.4239983125,"elo":1220.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.924512349,"x_train":396.6608033968,"imp":25.0952993451,"elo":1230.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.0877966046,"x_train":796.7427370653,"imp":24.6867721945,"elo":1241.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1969191452,"x_train":1194.3788869413,"imp":24.53147173,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2759753363,"x_train":1599.9245048672,"imp":24.4456483013,"elo":1251.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0400823469,"x_train":13.333580501,"imp":24.4188118821,"elo":1290.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0986631642,"x_train":155.2301034791,"imp":24.0200743235,"elo":1301.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.1470099052,"x_train":423.1216867301,"imp":23.3164694437,"elo":1327.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.1694069784,"x_train":1126.0038118286,"imp":23.0769468215,"elo":1335.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.2138690235,"x_train":3121.6567594042,"imp":22.6886608071,"elo":1348.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.2330722424,"x_train":6636.6428563278,"imp":22.4447970632,"elo":1358.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.251670447,"x_train":13596.8410765133,"imp":22.2533925416,"elo":1366.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.2739203723,"x_train":20505.2305784284,"imp":22.1674210825,"elo":1371.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.2710569482,"x_train":27107.3972485932,"imp":22.1115412421,"elo":1373.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7862858499,"x_train":4.8501982351,"imp":39.253213975,"elo":941.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9602823864,"x_train":9.7103499513,"imp":34.8742403411,"elo":1025.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0018135834,"x_train":24.7095068985,"imp":28.3372923926,"elo":1153.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0557265009,"x_train":49.0400181801,"imp":26.4342560518,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2034774794,"x_train":122.0237750052,"imp":24.9708088144,"elo":1318.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2879477145,"x_train":244.0803421306,"imp":24.7563738246,"elo":1327.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4038021731,"x_train":488.408774127,"imp":24.5616585216,"elo":1335.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4626957587,"x_train":733.7086039727,"imp":24.4319450711,"elo":1342.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4872412963,"x_train":984.6610737813,"imp":24.3757005775,"elo":1344.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.733891056,"x_train":8.5155691544,"imp":36.7405790294,"elo":938.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7139282714,"x_train":21.559937581,"imp":34.8581291908,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.4892327247,"x_train":62.1360327566,"imp":31.3600691517,"elo":1114.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0003746992,"x_train":125.9632889407,"imp":29.7612044964,"elo":1174.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.3351232966,"x_train":322.1066821739,"imp":28.1490155003,"elo":1233.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7914380124,"x_train":651.6805621192,"imp":27.4181493683,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.8304499119,"x_train":1305.7711888267,"imp":27.0395054438,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":11.5305755623,"x_train":1966.2811643084,"imp":26.8443056915,"elo":1277.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.2697346691,"x_train":2644.9909431875,"imp":26.7106266064,"elo":1280.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4290808604,"x_train":5.0434629122,"imp":25.3292162618,"elo":1242.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6502979595,"x_train":12.7823939958,"imp":24.1737947226,"elo":1296.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.4043257832,"x_train":32.375532788,"imp":23.0326922597,"elo":1344.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.1150961465,"x_train":66.0618660449,"imp":22.2235794539,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.9668372585,"x_train":169.9915970428,"imp":20.9475933773,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.5089803244,"x_train":340.9497698538,"imp":20.3436893496,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.9926620726,"x_train":684.7234530331,"imp":19.960683987,"elo":1472.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.0269000283,"x_train":1025.1619507673,"imp":19.8274646878,"elo":1478.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.1757953638,"x_train":1377.4770257446,"imp":19.6765226725,"elo":1486.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":7.0619559967,"x_train":44.325578779,"imp":13.7801247025,"elo":1577.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.690602132,"x_train":17.9457268623,"imp":28.1506397508,"elo":1200.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.2899674847,"x_train":57.1222738852,"imp":25.6633338737,"elo":1243.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.0939148169,"x_train":171.10400172,"imp":22.7753474268,"elo":1331.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.0865111274,"x_train":346.9635750688,"imp":21.9804246796,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.7950901544,"x_train":858.4265706116,"imp":21.088841993,"elo":1400.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.224511556,"x_train":1767.2813809485,"imp":20.7360381544,"elo":1408.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8020549495,"x_train":3533.1581339321,"imp":20.6670198053,"elo":1408.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.6927772436,"x_train":5301.5364312197,"imp":20.3926035696,"elo":1408.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2105111376,"x_train":7128.6715875046,"imp":20.2408220747,"elo":1406.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3671495151,"x_train":21.0623624205,"imp":34.8518982888,"elo":1005.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4648882862,"x_train":228.9454185462,"imp":57.9067657976,"elo":816.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":4.1872771761,"x_train":466.150787739,"imp":56.34720463,"elo":845.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":9.9244331069,"x_train":1174.4301363506,"imp":52.2472818,"elo":885.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":10.3718427289,"x_train":1409.5097915247,"imp":51.807304947,"elo":894.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1372541163,"x_train":0.7350357633,"imp":38.0436910874,"elo":996.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4864845935,"x_train":4.5934411859,"imp":30.728569383,"elo":1105.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8733492574,"x_train":15.6940857589,"imp":27.2882132656,"elo":1176.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.1516799057,"x_train":34.1407666441,"imp":25.6524426584,"elo":1229.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4009230269,"x_train":89.3035790361,"imp":24.6303873802,"elo":1312.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.5852633134,"x_train":181.3479112981,"imp":24.4340751577,"elo":1323.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7100231398,"x_train":365.5123989512,"imp":24.2898978553,"elo":1331.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.824130377,"x_train":549.883689003,"imp":24.2258530991,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.8696383358,"x_train":738.0209327471,"imp":24.1666716364,"elo":1336.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.6199152504,"x_train":19.2044919241,"imp":28.7154870283,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.7479796856,"x_train":41.019845821,"imp":23.7678228618,"elo":1318.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.2922134144,"x_train":103.1070784858,"imp":20.8883523221,"elo":1419.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.8666522912,"x_train":206.5111037718,"imp":19.7602119139,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.1922827589,"x_train":515.981423509,"imp":18.8542071974,"elo":1488.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4563530196,"x_train":1035.4018073353,"imp":18.3452431894,"elo":1508.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.0745491458,"x_train":2068.879284709,"imp":17.8102033233,"elo":1518.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.8336064118,"x_train":3101.33807397,"imp":17.5276091319,"elo":1525.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.128322436,"x_train":4150.8158923996,"imp":17.2746068522,"elo":1528.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8441387842,"x_train":10.2256680645,"imp":13.7665009799,"elo":1589.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.321516012,"x_train":22.8990925067,"imp":13.9372216378,"elo":1598.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.9479323005,"x_train":62.9203931706,"imp":13.6662217842,"elo":1628.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1961228236,"x_train":129.5212073934,"imp":13.3752599018,"elo":1680.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7707065307,"x_train":331.7365286813,"imp":12.9850993487,"elo":1711.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4473075753,"x_train":668.706695891,"imp":12.2239269317,"elo":1729.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4240520347,"x_train":1349.1531554668,"imp":11.8303326538,"elo":1727.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6603879275,"x_train":2024.0527372009,"imp":11.5296034873,"elo":1726.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0946622325,"x_train":2710.4880101354,"imp":11.1683966135,"elo":1723.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.681335577,"x_train":22.5239093941,"imp":21.4632789806,"elo":1341.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":42.5545582952,"x_train":43.394404174,"imp":26.0629035743,"elo":1264.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":51.1925760104,"x_train":63.8442293296,"imp":25.7044576611,"elo":1283.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":76.2941155935,"x_train":122.0779658592,"imp":25.0275066703,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":103.6508586102,"x_train":246.0408977706,"imp":24.530903334,"elo":1351.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":130.9325740919,"x_train":571.7774981312,"imp":23.701818962,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":148.5135318639,"x_train":1153.6863993273,"imp":23.1512161992,"elo":1401.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":181.9231911908,"x_train":2305.5325709939,"imp":22.7008846365,"elo":1421.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":200.1306803402,"x_train":3450.9693523062,"imp":22.4459528636,"elo":1429.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":212.2117190867,"x_train":4635.1714883924,"imp":22.2502714432,"elo":1434.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2707653455,"x_train":5.4556230963,"imp":21.3039429088,"elo":1444.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.6845942368,"x_train":55.129309117,"imp":8.5853827398,"elo":1675.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":1.0132382115,"x_train":5.0558157998,"imp":41.4995236172,"elo":853.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1967767849,"x_train":5.6364435241,"imp":15.8274664818,"elo":1568.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.6734553632,"x_train":12.8284122123,"imp":23.7957120917,"elo":1351.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":1.2218434244,"x_train":30.6749168115,"imp":23.0296619698,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.9529233417,"x_train":92.4143274115,"imp":22.0494339117,"elo":1410.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.354244056,"x_train":190.6444632314,"imp":21.4653911458,"elo":1431.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.8659233046,"x_train":476.8006390989,"imp":21.0018232613,"elo":1448.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.2670450085,"x_train":956.8892667394,"imp":20.6612676764,"elo":1464.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.7302509696,"x_train":1912.1428059667,"imp":20.4244533212,"elo":1472.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":3.9794024446,"x_train":2886.0992294917,"imp":20.3304715772,"elo":1475.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.2126633842,"x_train":3881.8318340105,"imp":20.2364543206,"elo":1480.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6126587114,"x_train":9.9408095472,"imp":14.6722580868,"elo":1661.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4885714957,"x_train":11.6635229296,"imp":13.4121206479,"elo":1600.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1330110917,"x_train":724.3138183772,"imp":41.8051180363,"elo":834.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.7458480887,"x_train":912.4748239301,"imp":41.5778126111,"elo":847.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.2200939284,"x_train":1342.7568425834,"imp":41.1117061849,"elo":859.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.5493158009,"x_train":2054.553219683,"imp":40.1252719699,"elo":884.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.646418544,"x_train":4867.9697698991,"imp":37.9267538233,"elo":921.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.164374429,"x_train":9985.461750605,"imp":35.6598832163,"elo":953.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9920396608,"x_train":20581.7115313909,"imp":34.3968313611,"elo":979.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":28.1227862886,"x_train":30569.2328830115,"imp":33.6984728566,"elo":998.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.2276662759,"x_train":40588.1540174129,"imp":33.180063577,"elo":1014.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0836134922,"x_train":4.3206513889,"imp":25.7080382962,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.314314201,"x_train":13.688894957,"imp":31.3878760718,"elo":1065.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5651033733,"x_train":28.9516297239,"imp":29.111751553,"elo":1132.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9390551091,"x_train":95.1897765055,"imp":27.1913731454,"elo":1197.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1240569211,"x_train":204.1284404869,"imp":25.9297244428,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.364215168,"x_train":488.13995387,"imp":24.9137275839,"elo":1276.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5390396052,"x_train":949.9701252728,"imp":24.3638116785,"elo":1293.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.725093994,"x_train":1876.0678595078,"imp":23.9569410027,"elo":1303.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8858986032,"x_train":2872.3862481998,"imp":23.7168404483,"elo":1312.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0241196975,"x_train":3902.0763863206,"imp":23.4748402271,"elo":1318.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5423707001,"x_train":4.7244136606,"imp":23.4532735312,"elo":1256.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.1711649272,"x_train":15.9463545731,"imp":22.2940252417,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.7546089687,"x_train":50.0714874559,"imp":21.863444531,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.6509055616,"x_train":100.9432671475,"imp":21.4472656691,"elo":1360.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.4198280312,"x_train":268.6777580855,"imp":21.1227818953,"elo":1380.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.6537513112,"x_train":544.7091931075,"imp":20.8973672866,"elo":1393.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.8978167069,"x_train":1096.7820562853,"imp":20.7149050758,"elo":1404.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.9192814316,"x_train":1644.1757072757,"imp":20.5790500324,"elo":1412.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.9829120694,"x_train":2210.16539451,"imp":20.415800911,"elo":1423.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":211.9764977363,"x_train":514.7174339635,"imp":29.571173745,"elo":1049.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":443.2333462935,"x_train":1597.2950547029,"imp":26.921553965,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":708.6888571295,"x_train":4783.0314636057,"imp":24.2410153143,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":775.3272509789,"x_train":10691.0627275524,"imp":22.9123031059,"elo":1317.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":828.5067468066,"x_train":26917.4383401199,"imp":21.0384055081,"elo":1399.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"x_train":28234.4532410041,"imp":20.9993995911,"elo":1403.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.6749433874,"x_train":9.4251823487,"imp":32.1768735408,"elo":1088.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":2.2180011085,"x_train":19.1707317375,"imp":29.8720596025,"elo":1167.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.5522410409,"x_train":47.812888724,"imp":26.9358200158,"elo":1272.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":2.8792809641,"x_train":97.2172107229,"imp":25.359048208,"elo":1309.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.1924978095,"x_train":246.1477821059,"imp":24.1100301403,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2945397285,"x_train":491.5550544333,"imp":23.5593254116,"elo":1366.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.8546394203,"x_train":987.2594279642,"imp":23.1640741686,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":3.2333264707,"x_train":1477.605852992,"imp":22.9995725755,"elo":1382.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.6075158555,"x_train":1980.6125379489,"imp":22.8867058478,"elo":1386.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/winrate_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..0f1fed6e67affb3c90f5397f8a4fe6f39966e47f
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_multiclass/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "TabPFN-3 (default)", "RealTabPFN-2.5 (default)", "TabICLv2 (default)", "LimiX (default)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "TabM (tuned)", "TabDPT (tuned)", "CatBoost (tuned)", "XGBoost (tuned)", "XGBoost (tuned + ensembled)", "ModernNCA (tuned)", "LightGBM (tuned + ensembled)", "LightGBM (tuned)", "iLTM (tuned + ensembled)", "xRFM (tuned + ensembled)", "BetaTabPFN (default)", "ExtraTrees (tuned + ensembled)", "TabM (default)", "SAP-RPT-OSS (default)", "EBM (tuned + ensembled)", "RandomForest (tuned + ensembled)", "TabDPT (default)", "iLTM (tuned)", "TorchMLP (tuned + ensembled)", "CatBoost (default)", "ExtraTrees (tuned)", "xRFM (tuned)", "TabSwift (default)", "EBM (tuned)", "RandomForest (tuned)", "XGBoost (default)", "FastaiMLP (tuned + ensembled)", "ModernNCA (default)", "TorchMLP (tuned)", "EBM (default)", "FastaiMLP (tuned)", "ChimeraBoost (tuned)", "LightGBM (default)", "ChimeraBoost (tuned + ensembled)", "RealMLP (default)", "xRFM (default)", "TorchMLP (default)", "ChimeraBoost (default)", "RandomForest (default)", "KNN (tuned + ensembled)", "OrionMSP (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "iLTM (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "PerpetualBooster (tuned)", "PerpetualBooster (default)", "FastaiMLP (default)", "TabFlex (default)", "TabSTAR (default)", "KNN (tuned)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.6666666865348816, 0.6602563858032227, 0.7692307829856873, 0.8141025900840759, 0.8012820482254028, 0.7564102411270142, 0.8974359035491943, 0.7948718070983887, 0.7820512652397156, 0.7820512652397156, 0.8910256624221802, 0.9038461446762085, 0.8846153616905212, 0.8205128312110901, 0.9102563858032227, 0.9230769276618958, 0.8333333134651184, 0.8910256624221802, 0.9294871687889099, 0.8846153616905212, 0.8910256624221802, 0.8974359035491943, 0.9102563858032227, 0.9038461446762085, 0.9230769276618958, 0.9230769276618958, 0.9166666865348816, 0.9294871687889099, 0.8589743375778198, 0.7692307829856873, 0.9102563858032227, 0.9102563858032227, 0.9102563858032227, 0.9358974099159241, 0.9294871687889099, 0.9358974099159241, 0.9230769276618958, 0.9294871687889099, 0.9230769276618958, 0.9294871687889099, 0.9102563858032227, 0.9358974099159241, 0.9038461446762085, 0.9807692170143127, 0.9358974099159241, 0.9294871687889099, 0.942307710647583, 0.9294871687889099, 0.9294871687889099, 0.9294871687889099, 0.9807692170143127, 0.9294871687889099, 0.9358974099159241, 0.942307710647583, 0.9935897588729858, 0.9358974099159241, 0.9358974099159241, 1.0, 0.9935897588729858, 0.942307710647583, 0.942307710647583, 0.9358974099159241, 0.9935897588729858, 0.9935897588729858, 0.9743589758872986, 0.9679487347602844, 1.0, 0.9679487347602844, 0.9935897588729858, 0.9358974099159241, 1.0, 1.0], [0.3333333432674408, NaN, 0.43589743971824646, 0.5448718070983887, 0.6410256624221802, 0.5833333134651184, 0.7019230723381042, 0.6730769276618958, 0.692307710647583, 0.7564102411270142, 0.7820512652397156, 0.7115384340286255, 0.7435897588729858, 0.8525640964508057, 0.8525640964508057, 0.9294871687889099, 0.8589743375778198, 0.8589743375778198, 0.7884615659713745, 0.9294871687889099, 0.9102563858032227, 0.9102563858032227, 0.8717948794364929, 0.8910256624221802, 0.9166666865348816, 0.8846153616905212, 0.8974359035491943, 0.9038461446762085, 0.8717948794364929, 0.8653846383094788, 0.7628205418586731, 0.9551281929016113, 0.8846153616905212, 0.807692289352417, 0.9551281929016113, 0.9487179517745972, 0.9679487347602844, 0.8974359035491943, 0.9230769276618958, 0.9743589758872986, 0.9615384340286255, 0.9102563858032227, 0.9807692170143127, 0.9294871687889099, 0.9487179517745972, 0.9487179517745972, 0.9487179517745972, 0.9679487347602844, 0.9551281929016113, 0.9487179517745972, 0.9679487347602844, 0.9935897588729858, 0.9807692170143127, 0.9935897588729858, 0.9615384340286255, 0.9871794581413269, 0.9358974099159241, 0.9807692170143127, 0.9935897588729858, 0.9935897588729858, 1.0, 0.9871794581413269, 0.9871794581413269, 1.0, 0.9871794581413269, 0.9743589758872986, 0.9679487347602844, 1.0, 1.0, 0.9871794581413269, 0.9679487347602844, 1.0, 1.0], [0.33974358439445496, 0.5641025900840759, NaN, 0.5833333134651184, 0.6474359035491943, 0.5961538553237915, 0.7628205418586731, 0.6794871687889099, 0.6987179517745972, 0.75, 0.7628205418586731, 0.7051281929016113, 0.7371794581413269, 0.8589743375778198, 0.807692289352417, 0.942307710647583, 0.8653846383094788, 0.807692289352417, 0.7564102411270142, 0.9294871687889099, 0.9038461446762085, 0.9230769276618958, 0.8846153616905212, 0.8974359035491943, 0.9038461446762085, 0.8910256624221802, 0.8846153616905212, 0.9166666865348816, 0.8525640964508057, 0.8525640964508057, 0.7564102411270142, 0.9358974099159241, 0.8782051205635071, 0.7884615659713745, 0.9551281929016113, 0.942307710647583, 0.942307710647583, 0.8782051205635071, 0.9230769276618958, 0.9743589758872986, 0.942307710647583, 0.9102563858032227, 0.9871794581413269, 0.9358974099159241, 0.9358974099159241, 0.9487179517745972, 0.9358974099159241, 0.9551281929016113, 0.942307710647583, 0.9358974099159241, 0.9551281929016113, 0.9871794581413269, 0.9679487347602844, 0.9871794581413269, 0.942307710647583, 0.9871794581413269, 0.9294871687889099, 0.9871794581413269, 0.9935897588729858, 0.9935897588729858, 1.0, 0.9935897588729858, 0.9935897588729858, 0.9935897588729858, 0.9871794581413269, 0.9743589758872986, 0.9743589758872986, 0.9935897588729858, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.23076923191547394, 0.45512819290161133, 0.4166666567325592, NaN, 0.6217948794364929, 0.557692289352417, 0.5128205418586731, 0.5961538553237915, 0.7435897588729858, 0.7371794581413269, 0.75, 0.75, 0.8269230723381042, 0.807692289352417, 0.7820512652397156, 0.8910256624221802, 0.8717948794364929, 0.7756410241127014, 0.8653846383094788, 0.9038461446762085, 0.8397436141967773, 0.8653846383094788, 0.8910256624221802, 0.9102563858032227, 0.8974359035491943, 0.9166666865348816, 0.9102563858032227, 0.8333333134651184, 0.9038461446762085, 0.8141025900840759, 0.6474359035491943, 0.9230769276618958, 0.9102563858032227, 0.8461538553237915, 0.942307710647583, 0.9102563858032227, 0.9487179517745972, 0.9038461446762085, 0.942307710647583, 0.9871794581413269, 0.9487179517745972, 0.9102563858032227, 0.9935897588729858, 0.8782051205635071, 0.9679487347602844, 0.9551281929016113, 0.942307710647583, 0.942307710647583, 0.9743589758872986, 0.9743589758872986, 0.9871794581413269, 0.9807692170143127, 0.9871794581413269, 0.9871794581413269, 0.9871794581413269, 0.9935897588729858, 0.9807692170143127, 0.9935897588729858, 0.9935897588729858, 0.9743589758872986, 1.0, 0.9935897588729858, 0.9935897588729858, 0.9807692170143127, 0.9935897588729858, 0.9743589758872986, 0.9679487347602844, 0.9935897588729858, 1.0, 0.9935897588729858, 0.9807692170143127, 1.0, 1.0], [0.18589743971824646, 0.3589743673801422, 0.35256409645080566, 0.3782051205635071, NaN, 0.4294871687889099, 0.5512820482254028, 0.4871794879436493, 0.6089743375778198, 0.6538461446762085, 0.6538461446762085, 0.692307710647583, 0.6474359035491943, 0.7243589758872986, 0.7307692170143127, 0.8141025900840759, 0.8653846383094788, 0.7628205418586731, 0.7179487347602844, 0.8269230723381042, 0.8846153616905212, 0.9038461446762085, 0.8397436141967773, 0.8589743375778198, 0.8910256624221802, 0.9294871687889099, 0.8012820482254028, 0.8461538553237915, 0.8525640964508057, 0.8012820482254028, 0.6730769276618958, 0.8717948794364929, 0.8333333134651184, 0.7884615659713745, 0.9487179517745972, 0.9230769276618958, 0.9551281929016113, 0.8461538553237915, 0.8589743375778198, 0.9551281929016113, 0.9102563858032227, 0.8717948794364929, 0.9807692170143127, 0.9487179517745972, 0.9358974099159241, 0.9230769276618958, 0.9294871687889099, 0.9807692170143127, 0.9358974099159241, 0.9358974099159241, 0.9487179517745972, 0.9679487347602844, 0.9358974099159241, 0.9487179517745972, 0.9487179517745972, 1.0, 0.9551281929016113, 0.9679487347602844, 1.0, 0.9807692170143127, 1.0, 0.9935897588729858, 0.9935897588729858, 0.9871794581413269, 0.9679487347602844, 0.9551281929016113, 0.9679487347602844, 1.0, 1.0, 0.9935897588729858, 0.9615384340286255, 1.0, 1.0], [0.19871795177459717, 0.4166666567325592, 0.4038461446762085, 0.4423076808452606, 0.5705128312110901, NaN, 0.4743589758872986, 0.5961538553237915, 0.6089743375778198, 0.5512820482254028, 0.6282051205635071, 0.7371794581413269, 0.7435897588729858, 0.7115384340286255, 0.6794871687889099, 0.8333333134651184, 0.8205128312110901, 0.7115384340286255, 0.7756410241127014, 0.8333333134651184, 0.7884615659713745, 0.7948718070983887, 0.8333333134651184, 0.8525640964508057, 0.8461538553237915, 0.8589743375778198, 0.8461538553237915, 0.7564102411270142, 0.9102563858032227, 0.7179487347602844, 0.5961538553237915, 0.8397436141967773, 0.8846153616905212, 0.8269230723381042, 0.9166666865348816, 0.8846153616905212, 0.8910256624221802, 0.9166666865348816, 0.9102563858032227, 0.9487179517745972, 0.8589743375778198, 0.9038461446762085, 0.9807692170143127, 0.8397436141967773, 0.9679487347602844, 0.8974359035491943, 0.9230769276618958, 0.942307710647583, 0.9102563858032227, 0.9294871687889099, 0.9294871687889099, 0.9871794581413269, 0.9807692170143127, 0.9679487347602844, 0.9358974099159241, 1.0, 0.9487179517745972, 0.9935897588729858, 1.0, 0.9871794581413269, 0.9935897588729858, 1.0, 0.9807692170143127, 0.9871794581413269, 0.9743589758872986, 0.9743589758872986, 0.9679487347602844, 0.9935897588729858, 1.0, 1.0, 0.9487179517745972, 1.0, 1.0], [0.24358974397182465, 0.29807692766189575, 0.2371794879436493, 0.4871794879436493, 0.44871795177459717, 0.5256410241127014, NaN, 0.5448718070983887, 0.6858974099159241, 0.6602563858032227, 0.6730769276618958, 0.5897436141967773, 0.6794871687889099, 0.7435897588729858, 0.8012820482254028, 0.8269230723381042, 0.7179487347602844, 0.8205128312110901, 0.6987179517745972, 0.8525640964508057, 0.8333333134651184, 0.8269230723381042, 0.7820512652397156, 0.8012820482254028, 0.8269230723381042, 0.7692307829856873, 0.8269230723381042, 0.8589743375778198, 0.8141025900840759, 0.8269230723381042, 0.7307692170143127, 0.8525640964508057, 0.7756410241127014, 0.7179487347602844, 0.9038461446762085, 0.9038461446762085, 0.8782051205635071, 0.8525640964508057, 0.8782051205635071, 0.9358974099159241, 0.8589743375778198, 0.8589743375778198, 0.9102563858032227, 0.9230769276618958, 0.9102563858032227, 0.9038461446762085, 0.8974359035491943, 0.942307710647583, 0.8782051205635071, 0.9166666865348816, 0.8846153616905212, 0.9935897588729858, 0.9358974099159241, 0.9615384340286255, 0.9358974099159241, 0.9615384340286255, 0.8846153616905212, 0.9871794581413269, 0.9935897588729858, 1.0, 0.9871794581413269, 0.9871794581413269, 0.9807692170143127, 0.9935897588729858, 0.9871794581413269, 0.9743589758872986, 0.9743589758872986, 0.9935897588729858, 0.9935897588729858, 1.0, 0.9487179517745972, 1.0, 1.0], [0.10256410390138626, 0.32692307233810425, 0.3205128312110901, 0.4038461446762085, 0.5128205418586731, 0.4038461446762085, 0.45512819290161133, NaN, 0.5833333134651184, 0.5833333134651184, 0.6089743375778198, 0.692307710647583, 0.75, 0.6730769276618958, 0.6858974099159241, 0.7884615659713745, 0.7564102411270142, 0.692307710647583, 0.75, 0.7884615659713745, 0.7692307829856873, 0.7820512652397156, 0.8205128312110901, 0.8397436141967773, 0.8589743375778198, 0.8782051205635071, 0.7948718070983887, 0.75, 0.8269230723381042, 0.75, 0.5384615659713745, 0.8269230723381042, 0.8717948794364929, 0.7628205418586731, 0.8846153616905212, 0.8333333134651184, 0.8910256624221802, 0.8461538553237915, 0.8269230723381042, 0.9230769276618958, 0.8589743375778198, 0.8782051205635071, 0.9294871687889099, 0.8269230723381042, 0.9551281929016113, 0.8846153616905212, 0.9102563858032227, 0.9038461446762085, 0.9230769276618958, 0.9294871687889099, 0.9294871687889099, 0.9294871687889099, 0.8974359035491943, 0.8974359035491943, 0.9358974099159241, 0.9935897588729858, 0.9358974099159241, 0.9166666865348816, 0.9743589758872986, 0.9871794581413269, 0.942307710647583, 0.9358974099159241, 0.9358974099159241, 0.9807692170143127, 0.9743589758872986, 0.9679487347602844, 0.9679487347602844, 0.9871794581413269, 0.9487179517745972, 0.9935897588729858, 0.942307710647583, 1.0, 1.0], [0.20512820780277252, 0.3076923191547394, 0.30128204822540283, 0.25641027092933655, 0.3910256326198578, 0.3910256326198578, 0.31410256028175354, 0.4166666567325592, NaN, 0.5128205418586731, 0.5192307829856873, 0.49358972907066345, 0.5448718070983887, 0.6602563858032227, 0.6538461446762085, 0.7435897588729858, 0.6538461446762085, 0.6602563858032227, 0.6282051205635071, 0.7564102411270142, 0.7756410241127014, 0.7692307829856873, 0.7820512652397156, 0.8012820482254028, 0.807692289352417, 0.7628205418586731, 0.8461538553237915, 0.7628205418586731, 0.8910256624221802, 0.7243589758872986, 0.6089743375778198, 0.8333333134651184, 0.9102563858032227, 0.6282051205635071, 0.8397436141967773, 0.7179487347602844, 0.8846153616905212, 0.9102563858032227, 0.9038461446762085, 0.8205128312110901, 0.8846153616905212, 0.9038461446762085, 0.9487179517745972, 0.7820512652397156, 0.942307710647583, 0.8269230723381042, 0.9038461446762085, 0.8910256624221802, 0.9166666865348816, 0.942307710647583, 0.942307710647583, 0.9807692170143127, 0.9743589758872986, 0.9615384340286255, 0.9487179517745972, 1.0, 0.9487179517745972, 0.9743589758872986, 1.0, 0.9871794581413269, 0.9743589758872986, 0.9807692170143127, 0.9679487347602844, 0.9935897588729858, 0.9807692170143127, 0.9743589758872986, 0.9679487347602844, 1.0, 0.9935897588729858, 1.0, 0.9551281929016113, 1.0, 1.0], [0.21794871985912323, 0.24358974397182465, 0.25, 0.2628205120563507, 0.3461538553237915, 0.44871795177459717, 0.33974358439445496, 0.4166666567325592, 0.4871794879436493, NaN, 0.5288461446762085, 0.5320512652397156, 0.5192307829856873, 0.6666666865348816, 0.692307710647583, 0.7371794581413269, 0.6858974099159241, 0.6987179517745972, 0.5256410241127014, 0.7692307829856873, 0.8461538553237915, 0.8717948794364929, 0.7371794581413269, 0.8269230723381042, 0.8269230723381042, 0.7564102411270142, 0.8333333134651184, 0.8653846383094788, 0.7435897588729858, 0.8782051205635071, 0.6282051205635071, 0.8653846383094788, 0.8141025900840759, 0.5769230723381042, 0.8525640964508057, 0.7564102411270142, 0.8782051205635071, 0.807692289352417, 0.8589743375778198, 0.7435897588729858, 0.8717948794364929, 0.8589743375778198, 0.9358974099159241, 0.9102563858032227, 0.8910256624221802, 0.807692289352417, 0.8782051205635071, 0.9166666865348816, 0.8846153616905212, 0.9166666865348816, 0.8910256624221802, 0.9935897588729858, 0.9358974099159241, 0.9743589758872986, 0.8846153616905212, 0.9358974099159241, 0.9230769276618958, 0.9871794581413269, 1.0, 0.9935897588729858, 0.942307710647583, 0.9743589758872986, 0.9679487347602844, 0.9871794581413269, 0.9807692170143127, 0.9743589758872986, 0.9679487347602844, 1.0, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.21794871985912323, 0.21794871985912323, 0.2371794879436493, 0.25, 0.3461538553237915, 0.3717948794364929, 0.32692307233810425, 0.3910256326198578, 0.48076921701431274, 0.4711538553237915, NaN, 0.5192307829856873, 0.48076921701431274, 0.6153846383094788, 0.6217948794364929, 0.7051281929016113, 0.692307710647583, 0.6602563858032227, 0.5128205418586731, 0.7115384340286255, 0.7820512652397156, 0.807692289352417, 0.7115384340286255, 0.7435897588729858, 0.7564102411270142, 0.7243589758872986, 0.7756410241127014, 0.8525640964508057, 0.7371794581413269, 0.8141025900840759, 0.6025640964508057, 0.8461538553237915, 0.7884615659713745, 0.5833333134651184, 0.8333333134651184, 0.8012820482254028, 0.8397436141967773, 0.7756410241127014, 0.8397436141967773, 0.7435897588729858, 0.8846153616905212, 0.807692289352417, 0.9038461446762085, 0.8782051205635071, 0.8653846383094788, 0.8461538553237915, 0.8653846383094788, 0.9358974099159241, 0.8653846383094788, 0.9038461446762085, 0.8846153616905212, 0.9807692170143127, 0.9102563858032227, 0.9679487347602844, 0.8910256624221802, 0.9294871687889099, 0.9294871687889099, 0.9807692170143127, 0.9807692170143127, 0.9871794581413269, 0.9358974099159241, 0.9743589758872986, 0.9615384340286255, 0.9935897588729858, 0.9871794581413269, 0.9743589758872986, 0.9615384340286255, 1.0, 1.0, 0.9871794581413269, 0.9743589758872986, 1.0, 1.0], [0.10897435992956161, 0.2884615361690521, 0.29487180709838867, 0.25, 0.3076923191547394, 0.2628205120563507, 0.41025641560554504, 0.3076923191547394, 0.5064102411270142, 0.46794870495796204, 0.48076921701431274, NaN, 0.5512820482254028, 0.5128205418586731, 0.5192307829856873, 0.6089743375778198, 0.6794871687889099, 0.5384615659713745, 0.6025640964508057, 0.5961538553237915, 0.6153846383094788, 0.6153846383094788, 0.6730769276618958, 0.6666666865348816, 0.6153846383094788, 0.807692289352417, 0.6282051205635071, 0.5769230723381042, 0.7692307829856873, 0.5833333134651184, 0.5256410241127014, 0.6474359035491943, 0.7564102411270142, 0.8012820482254028, 0.8141025900840759, 0.7179487347602844, 0.7564102411270142, 0.7820512652397156, 0.692307710647583, 0.8333333134651184, 0.7179487347602844, 0.7756410241127014, 0.8525640964508057, 0.7884615659713745, 0.8397436141967773, 0.7820512652397156, 0.7820512652397156, 0.8717948794364929, 0.8910256624221802, 0.8589743375778198, 0.8910256624221802, 0.8525640964508057, 0.7884615659713745, 0.8525640964508057, 0.8910256624221802, 1.0, 0.9230769276618958, 0.942307710647583, 1.0, 0.8717948794364929, 0.942307710647583, 0.8846153616905212, 0.8846153616905212, 0.8589743375778198, 0.942307710647583, 0.942307710647583, 0.9487179517745972, 0.942307710647583, 1.0, 0.942307710647583, 0.9487179517745972, 0.9679487347602844, 1.0], [0.09615384787321091, 0.25641027092933655, 0.2628205120563507, 0.17307692766189575, 0.35256409645080566, 0.25641027092933655, 0.3205128312110901, 0.25, 0.45512819290161133, 0.48076921701431274, 0.5192307829856873, 0.44871795177459717, NaN, 0.5256410241127014, 0.5705128312110901, 0.5448718070983887, 0.6153846383094788, 0.5769230723381042, 0.6794871687889099, 0.5769230723381042, 0.5961538553237915, 0.6153846383094788, 0.6410256624221802, 0.6730769276618958, 0.6538461446762085, 0.7115384340286255, 0.6538461446762085, 0.6538461446762085, 0.692307710647583, 0.6538461446762085, 0.5192307829856873, 0.692307710647583, 0.75, 0.7628205418586731, 0.7435897588729858, 0.6730769276618958, 0.7435897588729858, 0.7884615659713745, 0.7564102411270142, 0.8333333134651184, 0.7243589758872986, 0.7884615659713745, 0.8461538553237915, 0.7820512652397156, 0.8589743375778198, 0.8012820482254028, 0.75, 0.8269230723381042, 0.8397436141967773, 0.8653846383094788, 0.8653846383094788, 0.8653846383094788, 0.8589743375778198, 0.8974359035491943, 0.8782051205635071, 0.9807692170143127, 0.9679487347602844, 0.9615384340286255, 1.0, 0.8653846383094788, 0.8846153616905212, 0.9102563858032227, 0.9102563858032227, 0.8782051205635071, 0.9615384340286255, 0.9615384340286255, 0.9551281929016113, 0.942307710647583, 1.0, 0.9358974099159241, 0.9743589758872986, 0.9551281929016113, 1.0], [0.11538461595773697, 0.14743590354919434, 0.14102564752101898, 0.19230769574642181, 0.2756410241127014, 0.2884615361690521, 0.25641027092933655, 0.32692307233810425, 0.33974358439445496, 0.3333333432674408, 0.38461539149284363, 0.4871794879436493, 0.4743589758872986, NaN, 0.5064102411270142, 0.6025640964508057, 0.6153846383094788, 0.5448718070983887, 0.5064102411270142, 0.6730769276618958, 0.6153846383094788, 0.6474359035491943, 0.6474359035491943, 0.7371794581413269, 0.7243589758872986, 0.6858974099159241, 0.6794871687889099, 0.6666666865348816, 0.6153846383094788, 0.6858974099159241, 0.5384615659713745, 0.7884615659713745, 0.6730769276618958, 0.6217948794364929, 0.7564102411270142, 0.7564102411270142, 0.8269230723381042, 0.6987179517745972, 0.7435897588729858, 0.7243589758872986, 0.8141025900840759, 0.7692307829856873, 0.8461538553237915, 0.8333333134651184, 0.8269230723381042, 0.8397436141967773, 0.8717948794364929, 0.9038461446762085, 0.8846153616905212, 0.8653846383094788, 0.8910256624221802, 0.8782051205635071, 0.8012820482254028, 0.9166666865348816, 0.9294871687889099, 0.9038461446762085, 0.8333333134651184, 0.9102563858032227, 0.9294871687889099, 0.9487179517745972, 0.9807692170143127, 0.942307710647583, 0.942307710647583, 0.9615384340286255, 0.942307710647583, 0.9487179517745972, 0.9551281929016113, 0.9807692170143127, 0.9807692170143127, 0.9679487347602844, 0.9551281929016113, 0.9935897588729858, 1.0], [0.1794871836900711, 0.14743590354919434, 0.19230769574642181, 0.21794871985912323, 0.26923078298568726, 0.3205128312110901, 0.19871795177459717, 0.31410256028175354, 0.3461538553237915, 0.3076923191547394, 0.3782051205635071, 0.48076921701431274, 0.4294871687889099, 0.49358972907066345, NaN, 0.5, 0.5769230723381042, 0.6442307829856873, 0.5064102411270142, 0.5641025900840759, 0.6474359035491943, 0.6730769276618958, 0.5769230723381042, 0.6217948794364929, 0.6282051205635071, 0.5897436141967773, 0.6410256624221802, 0.7756410241127014, 0.5448718070983887, 0.8846153616905212, 0.6410256624221802, 0.8141025900840759, 0.6025640964508057, 0.6153846383094788, 0.6987179517745972, 0.7371794581413269, 0.7820512652397156, 0.6025640964508057, 0.7179487347602844, 0.6474359035491943, 0.8397436141967773, 0.7115384340286255, 0.8333333134651184, 0.8269230723381042, 0.7820512652397156, 0.8205128312110901, 0.8397436141967773, 0.8846153616905212, 0.807692289352417, 0.8910256624221802, 0.8333333134651184, 0.9551281929016113, 0.8141025900840759, 0.9487179517745972, 0.8974359035491943, 0.8910256624221802, 0.8910256624221802, 0.942307710647583, 0.9166666865348816, 0.9615384340286255, 0.942307710647583, 0.942307710647583, 0.9551281929016113, 0.9807692170143127, 0.9551281929016113, 0.9679487347602844, 0.9551281929016113, 0.9807692170143127, 1.0, 0.9935897588729858, 0.9679487347602844, 1.0, 1.0], [0.08974359184503555, 0.07051282376050949, 0.057692307978868484, 0.10897435992956161, 0.18589743971824646, 0.1666666716337204, 0.17307692766189575, 0.21153846383094788, 0.25641027092933655, 0.2628205120563507, 0.29487180709838867, 0.3910256326198578, 0.45512819290161133, 0.39743590354919434, 0.5, NaN, 0.5192307829856873, 0.5641025900840759, 0.5256410241127014, 0.6955128312110901, 0.6153846383094788, 0.6282051205635071, 0.5512820482254028, 0.5961538553237915, 0.6025640964508057, 0.5384615659713745, 0.6025640964508057, 0.6217948794364929, 0.6538461446762085, 0.6666666865348816, 0.5384615659713745, 0.7051281929016113, 0.6602563858032227, 0.5961538553237915, 0.692307710647583, 0.6602563858032227, 0.8333333134651184, 0.7371794581413269, 0.7756410241127014, 0.7435897588729858, 0.7884615659713745, 0.7820512652397156, 0.9230769276618958, 0.7051281929016113, 0.8333333134651184, 0.8012820482254028, 0.8782051205635071, 0.8333333134651184, 0.8782051205635071, 0.8910256624221802, 0.9166666865348816, 0.9487179517745972, 0.9038461446762085, 0.9615384340286255, 0.9230769276618958, 0.9230769276618958, 0.8717948794364929, 0.9743589758872986, 0.9487179517745972, 0.9615384340286255, 1.0, 0.9807692170143127, 0.9743589758872986, 0.9679487347602844, 0.9743589758872986, 0.9743589758872986, 0.9679487347602844, 0.9807692170143127, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.07692307978868484, 0.14102564752101898, 0.13461539149284363, 0.12820513546466827, 0.13461539149284363, 0.1794871836900711, 0.28205129504203796, 0.24358974397182465, 0.3461538553237915, 0.31410256028175354, 0.3076923191547394, 0.3205128312110901, 0.38461539149284363, 0.38461539149284363, 0.42307692766189575, 0.48076921701431274, NaN, 0.46794870495796204, 0.44871795177459717, 0.5128205418586731, 0.5384615659713745, 0.5512820482254028, 0.6442307829856873, 0.5961538553237915, 0.6089743375778198, 0.5256410241127014, 0.6602563858032227, 0.5705128312110901, 0.7435897588729858, 0.5705128312110901, 0.557692289352417, 0.6538461446762085, 0.7564102411270142, 0.5769230723381042, 0.692307710647583, 0.6153846383094788, 0.6987179517745972, 0.8141025900840759, 0.807692289352417, 0.6730769276618958, 0.7179487347602844, 0.7692307829856873, 0.8205128312110901, 0.6858974099159241, 0.9102563858032227, 0.7756410241127014, 0.7820512652397156, 0.8141025900840759, 0.8397436141967773, 0.8141025900840759, 0.8333333134651184, 0.8910256624221802, 0.8974359035491943, 0.9294871687889099, 0.8910256624221802, 0.9487179517745972, 0.9166666865348816, 0.9551281929016113, 0.9871794581413269, 0.9166666865348816, 0.9807692170143127, 0.9166666865348816, 0.9102563858032227, 0.9615384340286255, 0.9807692170143127, 0.9743589758872986, 0.9551281929016113, 0.9487179517745972, 0.9935897588729858, 0.9679487347602844, 0.9358974099159241, 0.9935897588729858, 1.0], [0.1666666716337204, 0.14102564752101898, 0.19230769574642181, 0.22435897588729858, 0.2371794879436493, 0.2884615361690521, 0.1794871836900711, 0.3076923191547394, 0.33974358439445496, 0.30128204822540283, 0.33974358439445496, 0.4615384638309479, 0.42307692766189575, 0.45512819290161133, 0.35576921701431274, 0.43589743971824646, 0.5320512652397156, NaN, 0.45512819290161133, 0.49358972907066345, 0.5641025900840759, 0.5769230723381042, 0.5192307829856873, 0.5448718070983887, 0.5448718070983887, 0.5961538553237915, 0.6217948794364929, 0.692307710647583, 0.5384615659713745, 0.7564102411270142, 0.557692289352417, 0.7564102411270142, 0.6089743375778198, 0.6025640964508057, 0.6410256624221802, 0.7115384340286255, 0.7179487347602844, 0.5833333134651184, 0.7115384340286255, 0.6282051205635071, 0.807692289352417, 0.6794871687889099, 0.7948718070983887, 0.7948718070983887, 0.7435897588729858, 0.7884615659713745, 0.8205128312110901, 0.8461538553237915, 0.75, 0.7948718070983887, 0.7692307829856873, 0.9166666865348816, 0.7884615659713745, 0.9358974099159241, 0.8589743375778198, 0.8782051205635071, 0.8525640964508057, 0.9294871687889099, 0.8589743375778198, 0.9743589758872986, 0.942307710647583, 0.9294871687889099, 0.9551281929016113, 0.9743589758872986, 0.9615384340286255, 0.9551281929016113, 0.9487179517745972, 0.9743589758872986, 1.0, 0.9871794581413269, 0.9679487347602844, 1.0, 1.0], [0.10897435992956161, 0.21153846383094788, 0.24358974397182465, 0.13461539149284363, 0.28205129504203796, 0.22435897588729858, 0.30128204822540283, 0.25, 0.3717948794364929, 0.4743589758872986, 0.4871794879436493, 0.39743590354919434, 0.3205128312110901, 0.49358972907066345, 0.49358972907066345, 0.4743589758872986, 0.5512820482254028, 0.5448718070983887, NaN, 0.5128205418586731, 0.5641025900840759, 0.5897436141967773, 0.5961538553237915, 0.6025640964508057, 0.6153846383094788, 0.6089743375778198, 0.5833333134651184, 0.6025640964508057, 0.6346153616905212, 0.6410256624221802, 0.49358972907066345, 0.5897436141967773, 0.6602563858032227, 0.7307692170143127, 0.6410256624221802, 0.5961538553237915, 0.6602563858032227, 0.6794871687889099, 0.692307710647583, 0.7756410241127014, 0.6538461446762085, 0.7051281929016113, 0.7564102411270142, 0.7115384340286255, 0.7243589758872986, 0.6858974099159241, 0.6858974099159241, 0.7884615659713745, 0.807692289352417, 0.8525640964508057, 0.8269230723381042, 0.7628205418586731, 0.8269230723381042, 0.8653846383094788, 0.8653846383094788, 0.9358974099159241, 0.9615384340286255, 0.9615384340286255, 0.9807692170143127, 0.8205128312110901, 0.8782051205635071, 0.9102563858032227, 0.9102563858032227, 0.8141025900840759, 0.9358974099159241, 0.9487179517745972, 0.9551281929016113, 0.942307710647583, 1.0, 0.9294871687889099, 0.9615384340286255, 0.942307710647583, 1.0], [0.07051282376050949, 0.07051282376050949, 0.07051282376050949, 0.09615384787321091, 0.17307692766189575, 0.1666666716337204, 0.14743590354919434, 0.21153846383094788, 0.24358974397182465, 0.23076923191547394, 0.2884615361690521, 0.4038461446762085, 0.42307692766189575, 0.32692307233810425, 0.43589743971824646, 0.3044871687889099, 0.4871794879436493, 0.5064102411270142, 0.4871794879436493, NaN, 0.5769230723381042, 0.5769230723381042, 0.5384615659713745, 0.5448718070983887, 0.5384615659713745, 0.5064102411270142, 0.557692289352417, 0.5833333134651184, 0.6346153616905212, 0.6538461446762085, 0.5192307829856873, 0.692307710647583, 0.6282051205635071, 0.5833333134651184, 0.6666666865348816, 0.6730769276618958, 0.7820512652397156, 0.7051281929016113, 0.7628205418586731, 0.6987179517745972, 0.7371794581413269, 0.7564102411270142, 0.8653846383094788, 0.692307710647583, 0.8012820482254028, 0.7820512652397156, 0.8525640964508057, 0.7948718070983887, 0.8525640964508057, 0.8525640964508057, 0.8525640964508057, 0.9102563858032227, 0.8782051205635071, 0.9358974099159241, 0.9230769276618958, 0.9166666865348816, 0.8525640964508057, 0.9743589758872986, 0.9294871687889099, 0.9551281929016113, 1.0, 0.9679487347602844, 0.9615384340286255, 0.9679487347602844, 0.9743589758872986, 0.9743589758872986, 0.9679487347602844, 0.9743589758872986, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.11538461595773697, 0.08974359184503555, 0.09615384787321091, 0.16025641560554504, 0.11538461595773697, 0.21153846383094788, 0.1666666716337204, 0.23076923191547394, 0.22435897588729858, 0.1538461595773697, 0.21794871985912323, 0.38461539149284363, 0.4038461446762085, 0.38461539149284363, 0.35256409645080566, 0.38461539149284363, 0.4615384638309479, 0.43589743971824646, 0.43589743971824646, 0.42307692766189575, NaN, 0.5288461446762085, 0.5192307829856873, 0.4871794879436493, 0.5448718070983887, 0.5064102411270142, 0.5705128312110901, 0.5769230723381042, 0.5448718070983887, 0.5769230723381042, 0.5833333134651184, 0.692307710647583, 0.557692289352417, 0.5448718070983887, 0.6410256624221802, 0.6666666865348816, 0.7628205418586731, 0.6410256624221802, 0.6987179517745972, 0.5961538553237915, 0.7371794581413269, 0.6987179517745972, 0.9038461446762085, 0.6858974099159241, 0.7948718070983887, 0.7435897588729858, 0.8525640964508057, 0.75, 0.8910256624221802, 0.8846153616905212, 0.8910256624221802, 0.8525640964508057, 0.7948718070983887, 0.9230769276618958, 0.9294871687889099, 0.9166666865348816, 0.8525640964508057, 0.9615384340286255, 0.9294871687889099, 0.9487179517745972, 0.9935897588729858, 0.9551281929016113, 0.9487179517745972, 0.9615384340286255, 0.9615384340286255, 0.9615384340286255, 0.9679487347602844, 0.9871794581413269, 0.9935897588729858, 0.9871794581413269, 0.9358974099159241, 1.0, 1.0], [0.10897435992956161, 0.08974359184503555, 0.07692307978868484, 0.13461539149284363, 0.09615384787321091, 0.20512820780277252, 0.17307692766189575, 0.21794871985912323, 0.23076923191547394, 0.12820513546466827, 0.19230769574642181, 0.38461539149284363, 0.38461539149284363, 0.35256409645080566, 0.32692307233810425, 0.3717948794364929, 0.44871795177459717, 0.42307692766189575, 0.41025641560554504, 0.42307692766189575, 0.4711538553237915, NaN, 0.5, 0.46794870495796204, 0.5192307829856873, 0.5192307829856873, 0.5641025900840759, 0.5769230723381042, 0.5256410241127014, 0.5641025900840759, 0.5705128312110901, 0.7179487347602844, 0.5641025900840759, 0.5384615659713745, 0.6474359035491943, 0.6666666865348816, 0.7820512652397156, 0.6346153616905212, 0.6987179517745972, 0.6089743375778198, 0.7628205418586731, 0.6730769276618958, 0.9294871687889099, 0.692307710647583, 0.7948718070983887, 0.75, 0.8525640964508057, 0.7564102411270142, 0.8846153616905212, 0.8717948794364929, 0.8974359035491943, 0.8525640964508057, 0.7948718070983887, 0.9230769276618958, 0.9294871687889099, 0.9102563858032227, 0.8525640964508057, 0.9551281929016113, 0.9166666865348816, 0.9487179517745972, 0.9935897588729858, 0.9615384340286255, 0.9551281929016113, 0.9615384340286255, 0.9615384340286255, 0.9551281929016113, 0.9615384340286255, 0.9871794581413269, 0.9935897588729858, 0.9871794581413269, 0.9358974099159241, 1.0, 1.0], [0.10256410390138626, 0.12820513546466827, 0.11538461595773697, 0.10897435992956161, 0.16025641560554504, 0.1666666716337204, 0.21794871985912323, 0.1794871836900711, 0.21794871985912323, 0.2628205120563507, 0.2884615361690521, 0.32692307233810425, 0.3589743673801422, 0.35256409645080566, 0.42307692766189575, 0.44871795177459717, 0.35576921701431274, 0.48076921701431274, 0.4038461446762085, 0.4615384638309479, 0.48076921701431274, 0.5, NaN, 0.5448718070983887, 0.5448718070983887, 0.43589743971824646, 0.6089743375778198, 0.5320512652397156, 0.6666666865348816, 0.5705128312110901, 0.5641025900840759, 0.6346153616905212, 0.6730769276618958, 0.5705128312110901, 0.6217948794364929, 0.5961538553237915, 0.6602563858032227, 0.7307692170143127, 0.7371794581413269, 0.6987179517745972, 0.6987179517745972, 0.7243589758872986, 0.7564102411270142, 0.6410256624221802, 0.8205128312110901, 0.7692307829856873, 0.7307692170143127, 0.75, 0.7820512652397156, 0.807692289352417, 0.7884615659713745, 0.8205128312110901, 0.8653846383094788, 0.9166666865348816, 0.9166666865348816, 0.9230769276618958, 0.9230769276618958, 0.9615384340286255, 0.9679487347602844, 0.8910256624221802, 0.9743589758872986, 0.9294871687889099, 0.9358974099159241, 0.9358974099159241, 0.9679487347602844, 0.9679487347602844, 0.9487179517745972, 0.942307710647583, 0.9871794581413269, 0.9679487347602844, 0.9743589758872986, 0.9935897588729858, 1.0], [0.08974359184503555, 0.10897435992956161, 0.10256410390138626, 0.08974359184503555, 0.14102564752101898, 0.14743590354919434, 0.19871795177459717, 0.16025641560554504, 0.19871795177459717, 0.17307692766189575, 0.25641027092933655, 0.3333333432674408, 0.32692307233810425, 0.2628205120563507, 0.3782051205635071, 0.4038461446762085, 0.4038461446762085, 0.45512819290161133, 0.39743590354919434, 0.45512819290161133, 0.5128205418586731, 0.5320512652397156, 0.45512819290161133, NaN, 0.5032051205635071, 0.5192307829856873, 0.5512820482254028, 0.5384615659713745, 0.5256410241127014, 0.6153846383094788, 0.557692289352417, 0.692307710647583, 0.5705128312110901, 0.5833333134651184, 0.6153846383094788, 0.6282051205635071, 0.7435897588729858, 0.5897436141967773, 0.7051281929016113, 0.6410256624221802, 0.7628205418586731, 0.7051281929016113, 0.8525640964508057, 0.6666666865348816, 0.7692307829856873, 0.7435897588729858, 0.8205128312110901, 0.7948718070983887, 0.9487179517745972, 0.942307710647583, 0.9871794581413269, 0.8397436141967773, 0.7692307829856873, 0.8910256624221802, 0.9807692170143127, 0.9038461446762085, 0.8589743375778198, 0.9230769276618958, 0.9487179517745972, 0.8910256624221802, 1.0, 0.9487179517745972, 0.9487179517745972, 0.9294871687889099, 0.942307710647583, 0.9551281929016113, 0.942307710647583, 1.0, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.09615384787321091, 0.0833333358168602, 0.09615384787321091, 0.10256410390138626, 0.10897435992956161, 0.1538461595773697, 0.17307692766189575, 0.14102564752101898, 0.19230769574642181, 0.17307692766189575, 0.24358974397182465, 0.38461539149284363, 0.3461538553237915, 0.2756410241127014, 0.3717948794364929, 0.39743590354919434, 0.3910256326198578, 0.45512819290161133, 0.38461539149284363, 0.4615384638309479, 0.45512819290161133, 0.48076921701431274, 0.45512819290161133, 0.4967948794364929, NaN, 0.5, 0.5320512652397156, 0.5192307829856873, 0.5064102411270142, 0.6025640964508057, 0.5448718070983887, 0.6666666865348816, 0.5256410241127014, 0.5897436141967773, 0.6025640964508057, 0.6538461446762085, 0.7564102411270142, 0.6025640964508057, 0.6794871687889099, 0.6346153616905212, 0.7628205418586731, 0.6858974099159241, 0.8589743375778198, 0.6538461446762085, 0.75, 0.7435897588729858, 0.8269230723381042, 0.7756410241127014, 0.9358974099159241, 0.9230769276618958, 0.9487179517745972, 0.8269230723381042, 0.7820512652397156, 0.8717948794364929, 0.9679487347602844, 0.8910256624221802, 0.8653846383094788, 0.9358974099159241, 0.9487179517745972, 0.8974359035491943, 1.0, 0.9551281929016113, 0.9615384340286255, 0.9294871687889099, 0.942307710647583, 0.9551281929016113, 0.942307710647583, 1.0, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.07692307978868484, 0.11538461595773697, 0.10897435992956161, 0.0833333358168602, 0.07051282376050949, 0.14102564752101898, 0.23076923191547394, 0.12179487198591232, 0.2371794879436493, 0.24358974397182465, 0.2756410241127014, 0.19230769574642181, 0.2884615361690521, 0.31410256028175354, 0.41025641560554504, 0.4615384638309479, 0.4743589758872986, 0.4038461446762085, 0.3910256326198578, 0.49358972907066345, 0.49358972907066345, 0.48076921701431274, 0.5641025900840759, 0.48076921701431274, 0.5, NaN, 0.5897436141967773, 0.5064102411270142, 0.6410256624221802, 0.5192307829856873, 0.5384615659713745, 0.6089743375778198, 0.6346153616905212, 0.557692289352417, 0.692307710647583, 0.5512820482254028, 0.6602563858032227, 0.7179487347602844, 0.6987179517745972, 0.6410256624221802, 0.6666666865348816, 0.7179487347602844, 0.7820512652397156, 0.6794871687889099, 0.807692289352417, 0.6987179517745972, 0.7628205418586731, 0.7884615659713745, 0.8782051205635071, 0.8397436141967773, 0.8846153616905212, 0.8269230723381042, 0.7884615659713745, 0.8653846383094788, 0.9358974099159241, 0.9230769276618958, 0.8589743375778198, 0.8717948794364929, 0.9935897588729858, 0.8269230723381042, 0.9935897588729858, 0.8717948794364929, 0.8653846383094788, 0.8910256624221802, 0.8974359035491943, 0.9166666865348816, 0.9230769276618958, 0.9935897588729858, 1.0, 0.9743589758872986, 0.9038461446762085, 0.9935897588729858, 1.0], [0.07692307978868484, 0.10256410390138626, 0.11538461595773697, 0.08974359184503555, 0.19871795177459717, 0.1538461595773697, 0.17307692766189575, 0.20512820780277252, 0.1538461595773697, 0.1666666716337204, 0.22435897588729858, 0.3717948794364929, 0.3461538553237915, 0.3205128312110901, 0.3589743673801422, 0.39743590354919434, 0.33974358439445496, 0.3782051205635071, 0.4166666567325592, 0.4423076808452606, 0.4294871687889099, 0.43589743971824646, 0.3910256326198578, 0.44871795177459717, 0.46794870495796204, 0.41025641560554504, NaN, 0.5064102411270142, 0.49358972907066345, 0.5128205418586731, 0.46794870495796204, 0.48076921701431274, 0.5641025900840759, 0.5384615659713745, 0.5961538553237915, 0.557692289352417, 0.6217948794364929, 0.6025640964508057, 0.692307710647583, 0.6858974099159241, 0.6089743375778198, 0.6602563858032227, 0.7051281929016113, 0.5897436141967773, 0.6346153616905212, 0.6538461446762085, 0.7115384340286255, 0.6794871687889099, 0.8141025900840759, 0.6666666865348816, 0.8333333134651184, 0.7243589758872986, 0.9615384340286255, 0.8525640964508057, 0.8846153616905212, 0.9102563858032227, 0.9166666865348816, 0.9743589758872986, 0.9935897588729858, 0.9615384340286255, 0.8461538553237915, 0.9679487347602844, 0.9679487347602844, 0.9679487347602844, 0.9166666865348816, 0.8974359035491943, 0.8974359035491943, 0.9679487347602844, 0.9935897588729858, 0.9871794581413269, 0.9871794581413269, 1.0, 1.0], [0.0833333358168602, 0.09615384787321091, 0.0833333358168602, 0.1666666716337204, 0.1538461595773697, 0.24358974397182465, 0.14102564752101898, 0.25, 0.2371794879436493, 0.13461539149284363, 0.14743590354919434, 0.42307692766189575, 0.3461538553237915, 0.3333333432674408, 0.22435897588729858, 0.3782051205635071, 0.4294871687889099, 0.3076923191547394, 0.39743590354919434, 0.4166666567325592, 0.42307692766189575, 0.42307692766189575, 0.46794870495796204, 0.4615384638309479, 0.48076921701431274, 0.49358972907066345, 0.49358972907066345, NaN, 0.4743589758872986, 0.48076921701431274, 0.4166666567325592, 0.6602563858032227, 0.5320512652397156, 0.5384615659713745, 0.557692289352417, 0.6730769276618958, 0.6346153616905212, 0.5384615659713745, 0.5833333134651184, 0.5384615659713745, 0.7628205418586731, 0.5833333134651184, 0.6987179517745972, 0.7307692170143127, 0.6730769276618958, 0.7371794581413269, 0.7756410241127014, 0.7884615659713745, 0.7051281929016113, 0.6858974099159241, 0.7307692170143127, 0.8653846383094788, 0.7307692170143127, 0.8846153616905212, 0.8461538553237915, 0.7756410241127014, 0.807692289352417, 0.8653846383094788, 0.8269230723381042, 0.942307710647583, 0.9358974099159241, 0.9358974099159241, 0.9358974099159241, 0.942307710647583, 0.9358974099159241, 0.9615384340286255, 0.942307710647583, 0.9935897588729858, 1.0, 0.9871794581413269, 0.9743589758872986, 1.0, 1.0], [0.07051282376050949, 0.12820513546466827, 0.14743590354919434, 0.09615384787321091, 0.14743590354919434, 0.08974359184503555, 0.18589743971824646, 0.17307692766189575, 0.10897435992956161, 0.25641027092933655, 0.2628205120563507, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.45512819290161133, 0.3461538553237915, 0.25641027092933655, 0.4615384638309479, 0.36538460850715637, 0.36538460850715637, 0.45512819290161133, 0.4743589758872986, 0.3333333432674408, 0.4743589758872986, 0.49358972907066345, 0.3589743673801422, 0.5064102411270142, 0.5256410241127014, NaN, 0.5641025900840759, 0.5064102411270142, 0.5, 0.6762820482254028, 0.5448718070983887, 0.5512820482254028, 0.4743589758872986, 0.5769230723381042, 0.7820512652397156, 0.5833333134651184, 0.6858974099159241, 0.5705128312110901, 0.8012820482254028, 0.6602563858032227, 0.557692289352417, 0.6346153616905212, 0.6153846383094788, 0.6602563858032227, 0.6346153616905212, 0.7628205418586731, 0.8012820482254028, 0.7692307829856873, 0.6858974099159241, 0.8205128312110901, 0.7884615659713745, 0.8782051205635071, 0.9679487347602844, 0.9551281929016113, 0.9487179517745972, 1.0, 0.7756410241127014, 0.9102563858032227, 0.9166666865348816, 0.9166666865348816, 0.7756410241127014, 0.9294871687889099, 0.9551281929016113, 0.942307710647583, 0.9038461446762085, 0.9807692170143127, 0.9358974099159241, 1.0, 0.9487179517745972, 1.0], [0.14102564752101898, 0.13461539149284363, 0.14743590354919434, 0.18589743971824646, 0.19871795177459717, 0.28205129504203796, 0.17307692766189575, 0.25, 0.2756410241127014, 0.12179487198591232, 0.18589743971824646, 0.4166666567325592, 0.3461538553237915, 0.31410256028175354, 0.11538461595773697, 0.3333333432674408, 0.4294871687889099, 0.24358974397182465, 0.3589743673801422, 0.3461538553237915, 0.42307692766189575, 0.43589743971824646, 0.4294871687889099, 0.38461539149284363, 0.39743590354919434, 0.48076921701431274, 0.4871794879436493, 0.5192307829856873, 0.43589743971824646, NaN, 0.43589743971824646, 0.6089743375778198, 0.49358972907066345, 0.5192307829856873, 0.5320512652397156, 0.6089743375778198, 0.5833333134651184, 0.49358972907066345, 0.5512820482254028, 0.46794870495796204, 0.6794871687889099, 0.5448718070983887, 0.6666666865348816, 0.75, 0.6217948794364929, 0.7307692170143127, 0.7179487347602844, 0.8141025900840759, 0.692307710647583, 0.6089743375778198, 0.692307710647583, 0.8525640964508057, 0.7051281929016113, 0.8910256624221802, 0.7692307829856873, 0.7628205418586731, 0.7756410241127014, 0.8397436141967773, 0.8141025900840759, 0.9166666865348816, 0.9294871687889099, 0.9294871687889099, 0.9358974099159241, 0.9358974099159241, 0.942307710647583, 0.9358974099159241, 0.9294871687889099, 0.9807692170143127, 1.0, 0.9807692170143127, 0.9615384340286255, 1.0, 1.0], [0.23076923191547394, 0.2371794879436493, 0.24358974397182465, 0.35256409645080566, 0.32692307233810425, 0.4038461446762085, 0.26923078298568726, 0.4615384638309479, 0.3910256326198578, 0.3717948794364929, 0.39743590354919434, 0.4743589758872986, 0.48076921701431274, 0.4615384638309479, 0.3589743673801422, 0.4615384638309479, 0.4423076808452606, 0.4423076808452606, 0.5064102411270142, 0.48076921701431274, 0.4166666567325592, 0.4294871687889099, 0.43589743971824646, 0.4423076808452606, 0.45512819290161133, 0.4615384638309479, 0.5320512652397156, 0.5833333134651184, 0.49358972907066345, 0.5641025900840759, NaN, 0.5448718070983887, 0.5, 0.5256410241127014, 0.5, 0.6474359035491943, 0.5128205418586731, 0.49358972907066345, 0.5961538553237915, 0.5256410241127014, 0.5705128312110901, 0.5192307829856873, 0.5064102411270142, 0.6730769276618958, 0.5448718070983887, 0.6602563858032227, 0.5961538553237915, 0.7371794581413269, 0.5, 0.6153846383094788, 0.48076921701431274, 0.7564102411270142, 0.6089743375778198, 0.7307692170143127, 0.6025640964508057, 0.6538461446762085, 0.5897436141967773, 0.6730769276618958, 0.6538461446762085, 0.8141025900840759, 0.6794871687889099, 0.75, 0.7371794581413269, 0.8141025900840759, 0.7692307829856873, 0.7628205418586731, 0.7628205418586731, 0.8012820482254028, 0.7628205418586731, 0.9487179517745972, 0.7756410241127014, 0.8205128312110901, 1.0], [0.08974359184503555, 0.044871795922517776, 0.06410256773233414, 0.07692307978868484, 0.12820513546466827, 0.16025641560554504, 0.14743590354919434, 0.17307692766189575, 0.1666666716337204, 0.13461539149284363, 0.1538461595773697, 0.35256409645080566, 0.3076923191547394, 0.21153846383094788, 0.18589743971824646, 0.29487180709838867, 0.3461538553237915, 0.24358974397182465, 0.41025641560554504, 0.3076923191547394, 0.3076923191547394, 0.28205129504203796, 0.36538460850715637, 0.3076923191547394, 0.3333333432674408, 0.3910256326198578, 0.5192307829856873, 0.33974358439445496, 0.5, 0.3910256326198578, 0.45512819290161133, NaN, 0.5128205418586731, 0.557692289352417, 0.5384615659713745, 0.6410256624221802, 0.5064102411270142, 0.5384615659713745, 0.6217948794364929, 0.5448718070983887, 0.7532051205635071, 0.557692289352417, 0.6217948794364929, 0.5833333134651184, 0.6410256624221802, 0.7051281929016113, 0.8525640964508057, 0.692307710647583, 0.7115384340286255, 0.6153846383094788, 0.7243589758872986, 0.75, 0.75, 0.9230769276618958, 0.8397436141967773, 0.8012820482254028, 0.7628205418586731, 0.8782051205635071, 0.8141025900840759, 0.942307710647583, 0.9487179517745972, 0.9358974099159241, 0.942307710647583, 0.9743589758872986, 0.9230769276618958, 0.9358974099159241, 0.9358974099159241, 0.9807692170143127, 0.9935897588729858, 0.9871794581413269, 0.942307710647583, 0.9935897588729858, 1.0], [0.08974359184503555, 0.11538461595773697, 0.12179487198591232, 0.08974359184503555, 0.1666666716337204, 0.11538461595773697, 0.22435897588729858, 0.12820513546466827, 0.08974359184503555, 0.18589743971824646, 0.21153846383094788, 0.24358974397182465, 0.25, 0.32692307233810425, 0.39743590354919434, 0.33974358439445496, 0.24358974397182465, 0.3910256326198578, 0.33974358439445496, 0.3717948794364929, 0.4423076808452606, 0.43589743971824646, 0.32692307233810425, 0.4294871687889099, 0.4743589758872986, 0.36538460850715637, 0.43589743971824646, 0.46794870495796204, 0.32371795177459717, 0.5064102411270142, 0.5, 0.4871794879436493, NaN, 0.5128205418586731, 0.4615384638309479, 0.45512819290161133, 0.5448718070983887, 0.5769230723381042, 0.5512820482254028, 0.6217948794364929, 0.5512820482254028, 0.8044871687889099, 0.6538461446762085, 0.5064102411270142, 0.557692289352417, 0.557692289352417, 0.6666666865348816, 0.5448718070983887, 0.7820512652397156, 0.7435897588729858, 0.7756410241127014, 0.6153846383094788, 0.7628205418586731, 0.7435897588729858, 0.8589743375778198, 0.9871794581413269, 0.9102563858032227, 0.9038461446762085, 0.9551281929016113, 0.7243589758872986, 0.8525640964508057, 0.9294871687889099, 0.9294871687889099, 0.7307692170143127, 0.9166666865348816, 0.9358974099159241, 0.9358974099159241, 0.8910256624221802, 0.9487179517745972, 0.942307710647583, 0.9935897588729858, 0.9230769276618958, 1.0], [0.08974359184503555, 0.19230769574642181, 0.21153846383094788, 0.1538461595773697, 0.21153846383094788, 0.17307692766189575, 0.28205129504203796, 0.2371794879436493, 0.3717948794364929, 0.42307692766189575, 0.4166666567325592, 0.19871795177459717, 0.2371794879436493, 0.3782051205635071, 0.38461539149284363, 0.4038461446762085, 0.42307692766189575, 0.39743590354919434, 0.26923078298568726, 0.4166666567325592, 0.45512819290161133, 0.4615384638309479, 0.4294871687889099, 0.4166666567325592, 0.41025641560554504, 0.4423076808452606, 0.4615384638309479, 0.4615384638309479, 0.45512819290161133, 0.48076921701431274, 0.4743589758872986, 0.4423076808452606, 0.4871794879436493, NaN, 0.49358972907066345, 0.4294871687889099, 0.4294871687889099, 0.5384615659713745, 0.5256410241127014, 0.5641025900840759, 0.46794870495796204, 0.5897436141967773, 0.5256410241127014, 0.5833333134651184, 0.5512820482254028, 0.49358972907066345, 0.4743589758872986, 0.6346153616905212, 0.6217948794364929, 0.75, 0.5961538553237915, 0.5961538553237915, 0.7115384340286255, 0.6538461446762085, 0.7756410241127014, 0.8012820482254028, 0.8589743375778198, 0.8205128312110901, 0.9230769276618958, 0.6025640964508057, 0.8653846383094788, 0.7756410241127014, 0.7628205418586731, 0.6666666865348816, 0.8589743375778198, 0.8782051205635071, 0.8910256624221802, 0.9102563858032227, 1.0, 0.8974359035491943, 0.8846153616905212, 0.9358974099159241, 1.0], [0.06410256773233414, 0.044871795922517776, 0.044871795922517776, 0.057692307978868484, 0.05128205195069313, 0.0833333358168602, 0.09615384787321091, 0.11538461595773697, 0.16025641560554504, 0.14743590354919434, 0.1666666716337204, 0.18589743971824646, 0.25641027092933655, 0.24358974397182465, 0.30128204822540283, 0.3076923191547394, 0.3076923191547394, 0.3589743673801422, 0.3589743673801422, 0.3333333432674408, 0.3589743673801422, 0.35256409645080566, 0.3782051205635071, 0.38461539149284363, 0.39743590354919434, 0.3076923191547394, 0.4038461446762085, 0.4423076808452606, 0.44871795177459717, 0.46794870495796204, 0.5, 0.4615384638309479, 0.5384615659713745, 0.5064102411270142, NaN, 0.44871795177459717, 0.5128205418586731, 0.6153846383094788, 0.557692289352417, 0.5256410241127014, 0.557692289352417, 0.6730769276618958, 0.6346153616905212, 0.6474359035491943, 0.6666666865348816, 0.6282051205635071, 0.6346153616905212, 0.692307710647583, 0.7307692170143127, 0.7243589758872986, 0.692307710647583, 0.7435897588729858, 0.7435897588729858, 0.8589743375778198, 0.8397436141967773, 0.8717948794364929, 0.7756410241127014, 0.8717948794364929, 0.9294871687889099, 0.7820512652397156, 0.9807692170143127, 0.8525640964508057, 0.8525640964508057, 0.8397436141967773, 0.8974359035491943, 0.9166666865348816, 0.9102563858032227, 0.942307710647583, 0.9743589758872986, 0.9487179517745972, 0.8910256624221802, 0.9743589758872986, 1.0], [0.07051282376050949, 0.05128205195069313, 0.057692307978868484, 0.08974359184503555, 0.07692307978868484, 0.11538461595773697, 0.09615384787321091, 0.1666666716337204, 0.28205129504203796, 0.24358974397182465, 0.19871795177459717, 0.28205129504203796, 0.32692307233810425, 0.24358974397182465, 0.2628205120563507, 0.33974358439445496, 0.38461539149284363, 0.2884615361690521, 0.4038461446762085, 0.32692307233810425, 0.3333333432674408, 0.3333333432674408, 0.4038461446762085, 0.3717948794364929, 0.3461538553237915, 0.44871795177459717, 0.4423076808452606, 0.32692307233810425, 0.5256410241127014, 0.3910256326198578, 0.35256409645080566, 0.3589743673801422, 0.5448718070983887, 0.5705128312110901, 0.5512820482254028, NaN, 0.4871794879436493, 0.5833333134651184, 0.5448718070983887, 0.49358972907066345, 0.45512819290161133, 0.557692289352417, 0.5833333134651184, 0.7115384340286255, 0.6538461446762085, 0.7916666865348816, 0.5384615659713745, 0.807692289352417, 0.6153846383094788, 0.6602563858032227, 0.6153846383094788, 0.7115384340286255, 0.6730769276618958, 0.8974359035491943, 0.6602563858032227, 0.7564102411270142, 0.7435897588729858, 0.7756410241127014, 0.8012820482254028, 0.8782051205635071, 0.8653846383094788, 0.9038461446762085, 0.9102563858032227, 0.9230769276618958, 0.8461538553237915, 0.8846153616905212, 0.8782051205635071, 0.9615384340286255, 0.9935897588729858, 0.9743589758872986, 0.942307710647583, 0.9935897588729858, 1.0], [0.06410256773233414, 0.03205128386616707, 0.057692307978868484, 0.05128205195069313, 0.044871795922517776, 0.10897435992956161, 0.12179487198591232, 0.10897435992956161, 0.11538461595773697, 0.12179487198591232, 0.16025641560554504, 0.24358974397182465, 0.25641027092933655, 0.17307692766189575, 0.21794871985912323, 0.1666666716337204, 0.30128204822540283, 0.28205129504203796, 0.33974358439445496, 0.21794871985912323, 0.2371794879436493, 0.21794871985912323, 0.33974358439445496, 0.25641027092933655, 0.24358974397182465, 0.33974358439445496, 0.3782051205635071, 0.36538460850715637, 0.42307692766189575, 0.4166666567325592, 0.4871794879436493, 0.49358972907066345, 0.45512819290161133, 0.5705128312110901, 0.4871794879436493, 0.5128205418586731, NaN, 0.5192307829856873, 0.5833333134651184, 0.5192307829856873, 0.6282051205635071, 0.5769230723381042, 0.6346153616905212, 0.5897436141967773, 0.6282051205635071, 0.6538461446762085, 0.692307710647583, 0.6858974099159241, 0.7884615659713745, 0.7435897588729858, 0.7948718070983887, 0.7243589758872986, 0.7179487347602844, 0.8653846383094788, 0.942307710647583, 0.8525640964508057, 0.807692289352417, 0.8974359035491943, 0.8910256624221802, 0.8461538553237915, 1.0, 0.9230769276618958, 0.9166666865348816, 0.8717948794364929, 0.9230769276618958, 0.9487179517745972, 0.9230769276618958, 0.9679487347602844, 1.0, 0.9935897588729858, 0.9487179517745972, 1.0, 1.0], [0.07692307978868484, 0.10256410390138626, 0.12179487198591232, 0.09615384787321091, 0.1538461595773697, 0.0833333358168602, 0.14743590354919434, 0.1538461595773697, 0.08974359184503555, 0.19230769574642181, 0.22435897588729858, 0.21794871985912323, 0.21153846383094788, 0.30128204822540283, 0.39743590354919434, 0.2628205120563507, 0.18589743971824646, 0.4166666567325592, 0.3205128312110901, 0.29487180709838867, 0.3589743673801422, 0.36538460850715637, 0.26923078298568726, 0.41025641560554504, 0.39743590354919434, 0.28205129504203796, 0.39743590354919434, 0.4615384638309479, 0.21794871985912323, 0.5064102411270142, 0.5064102411270142, 0.4615384638309479, 0.42307692766189575, 0.4615384638309479, 0.38461539149284363, 0.4166666567325592, 0.48076921701431274, NaN, 0.49358972907066345, 0.5897436141967773, 0.5320512652397156, 0.5993589758872986, 0.5192307829856873, 0.5192307829856873, 0.5064102411270142, 0.5128205418586731, 0.5705128312110901, 0.5641025900840759, 0.6410256624221802, 0.75, 0.6474359035491943, 0.5897436141967773, 0.7307692170143127, 0.7179487347602844, 0.807692289352417, 0.8974359035491943, 0.9358974099159241, 0.9166666865348816, 0.8910256624221802, 0.7243589758872986, 0.8012820482254028, 0.9166666865348816, 0.9102563858032227, 0.7051281929016113, 0.9166666865348816, 0.9166666865348816, 0.9230769276618958, 0.8910256624221802, 0.9679487347602844, 0.9358974099159241, 1.0, 0.9294871687889099, 1.0], [0.07051282376050949, 0.07692307978868484, 0.07692307978868484, 0.057692307978868484, 0.14102564752101898, 0.08974359184503555, 0.12179487198591232, 0.17307692766189575, 0.09615384787321091, 0.14102564752101898, 0.16025641560554504, 0.3076923191547394, 0.24358974397182465, 0.25641027092933655, 0.28205129504203796, 0.22435897588729858, 0.19230769574642181, 0.2884615361690521, 0.3076923191547394, 0.2371794879436493, 0.30128204822540283, 0.30128204822540283, 0.2628205120563507, 0.29487180709838867, 0.3205128312110901, 0.30128204822540283, 0.3076923191547394, 0.4166666567325592, 0.4166666567325592, 0.44871795177459717, 0.4038461446762085, 0.3782051205635071, 0.44871795177459717, 0.4743589758872986, 0.4423076808452606, 0.45512819290161133, 0.4166666567325592, 0.5064102411270142, NaN, 0.5897436141967773, 0.48076921701431274, 0.5512820482254028, 0.5320512652397156, 0.5256410241127014, 0.5, 0.557692289352417, 0.5833333134651184, 0.5641025900840759, 0.6987179517745972, 0.5961538553237915, 0.6730769276618958, 0.5833333134651184, 0.9230769276618958, 0.7820512652397156, 0.8333333134651184, 0.8653846383094788, 0.8782051205635071, 0.942307710647583, 0.9487179517745972, 0.8782051205635071, 0.807692289352417, 0.9294871687889099, 0.9358974099159241, 0.8846153616905212, 0.8974359035491943, 0.9230769276618958, 0.9102563858032227, 0.9294871687889099, 0.9935897588729858, 0.9807692170143127, 0.9743589758872986, 0.9935897588729858, 1.0], [0.07692307978868484, 0.025641025975346565, 0.025641025975346565, 0.012820512987673283, 0.044871795922517776, 0.05128205195069313, 0.06410256773233414, 0.07692307978868484, 0.1794871836900711, 0.25641027092933655, 0.25641027092933655, 0.1666666716337204, 0.1666666716337204, 0.2756410241127014, 0.35256409645080566, 0.25641027092933655, 0.32692307233810425, 0.3717948794364929, 0.22435897588729858, 0.30128204822540283, 0.4038461446762085, 0.3910256326198578, 0.30128204822540283, 0.3589743673801422, 0.36538460850715637, 0.3589743673801422, 0.31410256028175354, 0.4615384638309479, 0.31410256028175354, 0.5320512652397156, 0.4743589758872986, 0.45512819290161133, 0.3782051205635071, 0.43589743971824646, 0.4743589758872986, 0.5064102411270142, 0.48076921701431274, 0.41025641560554504, 0.41025641560554504, NaN, 0.5128205418586731, 0.5320512652397156, 0.5128205418586731, 0.6602563858032227, 0.5512820482254028, 0.6153846383094788, 0.5448718070983887, 0.6794871687889099, 0.5769230723381042, 0.7628205418586731, 0.5705128312110901, 0.6794871687889099, 0.6217948794364929, 0.6987179517745972, 0.7628205418586731, 0.8653846383094788, 0.8525640964508057, 0.9294871687889099, 0.7756410241127014, 0.6794871687889099, 0.807692289352417, 0.8782051205635071, 0.8653846383094788, 0.7115384340286255, 0.9358974099159241, 0.9487179517745972, 0.9487179517745972, 0.8782051205635071, 0.9551281929016113, 0.9358974099159241, 0.9551281929016113, 0.9679487347602844, 1.0], [0.07051282376050949, 0.03846153989434242, 0.057692307978868484, 0.05128205195069313, 0.08974359184503555, 0.14102564752101898, 0.14102564752101898, 0.14102564752101898, 0.11538461595773697, 0.12820513546466827, 0.11538461595773697, 0.28205129504203796, 0.2756410241127014, 0.18589743971824646, 0.16025641560554504, 0.21153846383094788, 0.28205129504203796, 0.19230769574642181, 0.3461538553237915, 0.2628205120563507, 0.2628205120563507, 0.2371794879436493, 0.30128204822540283, 0.2371794879436493, 0.2371794879436493, 0.3333333432674408, 0.3910256326198578, 0.2371794879436493, 0.4294871687889099, 0.3205128312110901, 0.4294871687889099, 0.24679486453533173, 0.44871795177459717, 0.5320512652397156, 0.4423076808452606, 0.5448718070983887, 0.3717948794364929, 0.46794870495796204, 0.5192307829856873, 0.4871794879436493, NaN, 0.5192307829856873, 0.5384615659713745, 0.5384615659713745, 0.5448718070983887, 0.6538461446762085, 0.7179487347602844, 0.6474359035491943, 0.6538461446762085, 0.5256410241127014, 0.6794871687889099, 0.6666666865348816, 0.7051281929016113, 0.8846153616905212, 0.7948718070983887, 0.7692307829856873, 0.7115384340286255, 0.8269230723381042, 0.7884615659713745, 0.8974359035491943, 0.9102563858032227, 0.9230769276618958, 0.9294871687889099, 0.9294871687889099, 0.9038461446762085, 0.9166666865348816, 0.9102563858032227, 0.9679487347602844, 0.9935897588729858, 0.9807692170143127, 0.942307710647583, 0.9935897588729858, 1.0], [0.08974359184503555, 0.08974359184503555, 0.08974359184503555, 0.08974359184503555, 0.12820513546466827, 0.09615384787321091, 0.14102564752101898, 0.12179487198591232, 0.09615384787321091, 0.14102564752101898, 0.19230769574642181, 0.22435897588729858, 0.21153846383094788, 0.23076923191547394, 0.2884615361690521, 0.21794871985912323, 0.23076923191547394, 0.3205128312110901, 0.29487180709838867, 0.24358974397182465, 0.30128204822540283, 0.32692307233810425, 0.2756410241127014, 0.29487180709838867, 0.31410256028175354, 0.28205129504203796, 0.33974358439445496, 0.4166666567325592, 0.19871795177459717, 0.45512819290161133, 0.48076921701431274, 0.4423076808452606, 0.1955128163099289, 0.41025641560554504, 0.32692307233810425, 0.4423076808452606, 0.42307692766189575, 0.4006410241127014, 0.44871795177459717, 0.46794870495796204, 0.48076921701431274, NaN, 0.5, 0.49358972907066345, 0.4038461446762085, 0.49358972907066345, 0.6089743375778198, 0.5384615659713745, 0.6217948794364929, 0.6666666865348816, 0.6282051205635071, 0.5448718070983887, 0.6410256624221802, 0.6794871687889099, 0.807692289352417, 0.8301281929016113, 0.8525640964508057, 0.8653846383094788, 0.8012820482254028, 0.692307710647583, 0.807692289352417, 0.8846153616905212, 0.8782051205635071, 0.6858974099159241, 0.8974359035491943, 0.9230769276618958, 0.9166666865348816, 0.8782051205635071, 0.9294871687889099, 0.9230769276618958, 0.9743589758872986, 0.9038461446762085, 1.0], [0.06410256773233414, 0.01923076994717121, 0.012820512987673283, 0.006410256493836641, 0.01923076994717121, 0.01923076994717121, 0.08974359184503555, 0.07051282376050949, 0.05128205195069313, 0.06410256773233414, 0.09615384787321091, 0.14743590354919434, 0.1538461595773697, 0.1538461595773697, 0.1666666716337204, 0.07692307978868484, 0.1794871836900711, 0.20512820780277252, 0.24358974397182465, 0.13461539149284363, 0.09615384787321091, 0.07051282376050949, 0.24358974397182465, 0.14743590354919434, 0.14102564752101898, 0.21794871985912323, 0.29487180709838867, 0.30128204822540283, 0.33974358439445496, 0.3333333432674408, 0.49358972907066345, 0.3782051205635071, 0.3461538553237915, 0.4743589758872986, 0.36538460850715637, 0.4166666567325592, 0.36538460850715637, 0.48076921701431274, 0.46794870495796204, 0.4871794879436493, 0.4615384638309479, 0.5, NaN, 0.4743589758872986, 0.5256410241127014, 0.5320512652397156, 0.6346153616905212, 0.5512820482254028, 0.6666666865348816, 0.6987179517745972, 0.6987179517745972, 0.6025640964508057, 0.692307710647583, 0.7692307829856873, 0.8717948794364929, 0.8717948794364929, 0.8141025900840759, 0.8910256624221802, 0.8717948794364929, 0.7371794581413269, 0.9615384340286255, 0.8846153616905212, 0.8717948794364929, 0.7756410241127014, 0.8974359035491943, 0.9294871687889099, 0.9230769276618958, 0.9487179517745972, 0.9871794581413269, 0.9871794581413269, 0.9294871687889099, 1.0, 1.0], [0.09615384787321091, 0.07051282376050949, 0.06410256773233414, 0.12179487198591232, 0.05128205195069313, 0.16025641560554504, 0.07692307978868484, 0.17307692766189575, 0.21794871985912323, 0.08974359184503555, 0.12179487198591232, 0.21153846383094788, 0.21794871985912323, 0.1666666716337204, 0.17307692766189575, 0.29487180709838867, 0.31410256028175354, 0.20512820780277252, 0.2884615361690521, 0.3076923191547394, 0.31410256028175354, 0.3076923191547394, 0.3589743673801422, 0.3333333432674408, 0.3461538553237915, 0.3205128312110901, 0.41025641560554504, 0.26923078298568726, 0.4423076808452606, 0.25, 0.32692307233810425, 0.4166666567325592, 0.49358972907066345, 0.4166666567325592, 0.35256409645080566, 0.2884615361690521, 0.41025641560554504, 0.48076921701431274, 0.4743589758872986, 0.33974358439445496, 0.4615384638309479, 0.5064102411270142, 0.5256410241127014, NaN, 0.5641025900840759, 0.4743589758872986, 0.48076921701431274, 0.7179487347602844, 0.6474359035491943, 0.5512820482254028, 0.6346153616905212, 0.557692289352417, 0.5833333134651184, 0.6346153616905212, 0.6858974099159241, 0.6089743375778198, 0.6538461446762085, 0.5961538553237915, 0.7948718070983887, 0.8205128312110901, 0.7435897588729858, 0.7243589758872986, 0.7243589758872986, 0.8653846383094788, 0.6858974099159241, 0.7371794581413269, 0.7371794581413269, 0.9935897588729858, 0.9230769276618958, 0.9358974099159241, 0.8974359035491943, 1.0, 0.9935897588729858], [0.01923076994717121, 0.05128205195069313, 0.06410256773233414, 0.03205128386616707, 0.06410256773233414, 0.03205128386616707, 0.08974359184503555, 0.044871795922517776, 0.057692307978868484, 0.10897435992956161, 0.13461539149284363, 0.16025641560554504, 0.14102564752101898, 0.17307692766189575, 0.21794871985912323, 0.1666666716337204, 0.08974359184503555, 0.25641027092933655, 0.2756410241127014, 0.19871795177459717, 0.20512820780277252, 0.20512820780277252, 0.1794871836900711, 0.23076923191547394, 0.25, 0.19230769574642181, 0.36538460850715637, 0.32692307233810425, 0.36538460850715637, 0.3782051205635071, 0.45512819290161133, 0.3589743673801422, 0.4423076808452606, 0.44871795177459717, 0.3333333432674408, 0.3461538553237915, 0.3717948794364929, 0.49358972907066345, 0.5, 0.44871795177459717, 0.45512819290161133, 0.5961538553237915, 0.4743589758872986, 0.43589743971824646, NaN, 0.45512819290161133, 0.5384615659713745, 0.5128205418586731, 0.6987179517745972, 0.6730769276618958, 0.7179487347602844, 0.5320512652397156, 0.6858974099159241, 0.7435897588729858, 0.8012820482254028, 0.8589743375778198, 0.8269230723381042, 0.8269230723381042, 0.8461538553237915, 0.7820512652397156, 0.7884615659713745, 0.7948718070983887, 0.7948718070983887, 0.8205128312110901, 0.8846153616905212, 0.9102563858032227, 0.9038461446762085, 0.8653846383094788, 0.9294871687889099, 0.8717948794364929, 0.9038461446762085, 0.9807692170143127, 1.0], [0.06410256773233414, 0.05128205195069313, 0.05128205195069313, 0.044871795922517776, 0.07692307978868484, 0.10256410390138626, 0.09615384787321091, 0.11538461595773697, 0.17307692766189575, 0.19230769574642181, 0.1538461595773697, 0.21794871985912323, 0.19871795177459717, 0.16025641560554504, 0.1794871836900711, 0.19871795177459717, 0.22435897588729858, 0.21153846383094788, 0.31410256028175354, 0.21794871985912323, 0.25641027092933655, 0.25, 0.23076923191547394, 0.25641027092933655, 0.25641027092933655, 0.30128204822540283, 0.3461538553237915, 0.2628205120563507, 0.38461539149284363, 0.26923078298568726, 0.33974358439445496, 0.29487180709838867, 0.4423076808452606, 0.5064102411270142, 0.3717948794364929, 0.2083333283662796, 0.3461538553237915, 0.4871794879436493, 0.4423076808452606, 0.38461539149284363, 0.3461538553237915, 0.5064102411270142, 0.46794870495796204, 0.5256410241127014, 0.5448718070983887, NaN, 0.3589743673801422, 0.6602563858032227, 0.5256410241127014, 0.5384615659713745, 0.5256410241127014, 0.5384615659713745, 0.6217948794364929, 0.8205128312110901, 0.6410256624221802, 0.7243589758872986, 0.7051281929016113, 0.6987179517745972, 0.75, 0.75, 0.8141025900840759, 0.8717948794364929, 0.8653846383094788, 0.8012820482254028, 0.7948718070983887, 0.8397436141967773, 0.8269230723381042, 0.9294871687889099, 0.9679487347602844, 0.9551281929016113, 0.942307710647583, 0.9871794581413269, 1.0], [0.07051282376050949, 0.05128205195069313, 0.06410256773233414, 0.057692307978868484, 0.07051282376050949, 0.07692307978868484, 0.10256410390138626, 0.08974359184503555, 0.09615384787321091, 0.12179487198591232, 0.13461539149284363, 0.21794871985912323, 0.25, 0.12820513546466827, 0.16025641560554504, 0.12179487198591232, 0.21794871985912323, 0.1794871836900711, 0.31410256028175354, 0.14743590354919434, 0.14743590354919434, 0.14743590354919434, 0.26923078298568726, 0.1794871836900711, 0.17307692766189575, 0.2371794879436493, 0.2884615361690521, 0.22435897588729858, 0.33974358439445496, 0.28205129504203796, 0.4038461446762085, 0.14743590354919434, 0.3333333432674408, 0.5256410241127014, 0.36538460850715637, 0.4615384638309479, 0.3076923191547394, 0.4294871687889099, 0.4166666567325592, 0.45512819290161133, 0.28205129504203796, 0.3910256326198578, 0.36538460850715637, 0.5192307829856873, 0.4615384638309479, 0.6410256624221802, NaN, 0.5961538553237915, 0.6025640964508057, 0.5064102411270142, 0.6153846383094788, 0.5512820482254028, 0.6410256624221802, 0.7820512652397156, 0.7628205418586731, 0.7371794581413269, 0.7692307829856873, 0.8012820482254028, 0.7692307829856873, 0.7948718070983887, 0.8653846383094788, 0.9102563858032227, 0.9166666865348816, 0.8653846383094788, 0.8846153616905212, 0.9038461446762085, 0.8782051205635071, 0.9679487347602844, 0.9807692170143127, 0.9743589758872986, 0.9743589758872986, 1.0, 1.0], [0.057692307978868484, 0.03205128386616707, 0.044871795922517776, 0.057692307978868484, 0.01923076994717121, 0.057692307978868484, 0.057692307978868484, 0.09615384787321091, 0.10897435992956161, 0.0833333358168602, 0.06410256773233414, 0.12820513546466827, 0.17307692766189575, 0.09615384787321091, 0.11538461595773697, 0.1666666716337204, 0.18589743971824646, 0.1538461595773697, 0.21153846383094788, 0.20512820780277252, 0.25, 0.24358974397182465, 0.25, 0.20512820780277252, 0.22435897588729858, 0.21153846383094788, 0.3205128312110901, 0.21153846383094788, 0.36538460850715637, 0.18589743971824646, 0.2628205120563507, 0.3076923191547394, 0.45512819290161133, 0.36538460850715637, 0.3076923191547394, 0.19230769574642181, 0.31410256028175354, 0.43589743971824646, 0.43589743971824646, 0.3205128312110901, 0.35256409645080566, 0.4615384638309479, 0.44871795177459717, 0.28205129504203796, 0.4871794879436493, 0.33974358439445496, 0.4038461446762085, NaN, 0.5897436141967773, 0.5, 0.5641025900840759, 0.45512819290161133, 0.5897436141967773, 0.6089743375778198, 0.6602563858032227, 0.5961538553237915, 0.6794871687889099, 0.5833333134651184, 0.7692307829856873, 0.7628205418586731, 0.7435897588729858, 0.692307710647583, 0.7051281929016113, 0.807692289352417, 0.6666666865348816, 0.7179487347602844, 0.7243589758872986, 0.9871794581413269, 0.8397436141967773, 0.9358974099159241, 0.8974359035491943, 1.0, 0.9935897588729858], [0.07051282376050949, 0.044871795922517776, 0.057692307978868484, 0.025641025975346565, 0.06410256773233414, 0.08974359184503555, 0.12179487198591232, 0.07692307978868484, 0.0833333358168602, 0.11538461595773697, 0.13461539149284363, 0.10897435992956161, 0.16025641560554504, 0.11538461595773697, 0.19230769574642181, 0.12179487198591232, 0.16025641560554504, 0.25, 0.19230769574642181, 0.14743590354919434, 0.10897435992956161, 0.11538461595773697, 0.21794871985912323, 0.05128205195069313, 0.06410256773233414, 0.12179487198591232, 0.18589743971824646, 0.29487180709838867, 0.2371794879436493, 0.3076923191547394, 0.5, 0.2884615361690521, 0.21794871985912323, 0.3782051205635071, 0.26923078298568726, 0.38461539149284363, 0.21153846383094788, 0.3589743673801422, 0.30128204822540283, 0.42307692766189575, 0.3461538553237915, 0.3782051205635071, 0.3333333432674408, 0.35256409645080566, 0.30128204822540283, 0.4743589758872986, 0.39743590354919434, 0.41025641560554504, NaN, 0.5512820482254028, 0.4711538553237915, 0.4871794879436493, 0.6602563858032227, 0.6217948794364929, 0.9294871687889099, 0.6410256624221802, 0.6538461446762085, 0.7564102411270142, 0.7051281929016113, 0.6987179517745972, 0.8333333134651184, 0.7948718070983887, 0.8012820482254028, 0.7179487347602844, 0.7307692170143127, 0.7884615659713745, 0.7756410241127014, 0.9487179517745972, 0.9358974099159241, 0.9166666865348816, 0.8653846383094788, 0.9743589758872986, 1.0], [0.07051282376050949, 0.05128205195069313, 0.06410256773233414, 0.025641025975346565, 0.06410256773233414, 0.07051282376050949, 0.0833333358168602, 0.07051282376050949, 0.057692307978868484, 0.0833333358168602, 0.09615384787321091, 0.14102564752101898, 0.13461539149284363, 0.13461539149284363, 0.10897435992956161, 0.10897435992956161, 0.18589743971824646, 0.20512820780277252, 0.14743590354919434, 0.14743590354919434, 0.11538461595773697, 0.12820513546466827, 0.19230769574642181, 0.057692307978868484, 0.07692307978868484, 0.16025641560554504, 0.3333333432674408, 0.31410256028175354, 0.19871795177459717, 0.3910256326198578, 0.38461539149284363, 0.38461539149284363, 0.25641027092933655, 0.25, 0.2756410241127014, 0.33974358439445496, 0.25641027092933655, 0.25, 0.4038461446762085, 0.2371794879436493, 0.4743589758872986, 0.3333333432674408, 0.30128204822540283, 0.44871795177459717, 0.32692307233810425, 0.4615384638309479, 0.49358972907066345, 0.5, 0.44871795177459717, NaN, 0.4615384638309479, 0.5192307829856873, 0.5384615659713745, 0.6666666865348816, 0.7179487347602844, 0.6282051205635071, 0.7756410241127014, 0.7564102411270142, 0.6474359035491943, 0.6217948794364929, 0.7756410241127014, 0.8141025900840759, 0.7884615659713745, 0.6666666865348816, 0.8782051205635071, 0.9166666865348816, 0.8846153616905212, 0.8910256624221802, 0.8910256624221802, 0.942307710647583, 0.9294871687889099, 0.9807692170143127, 1.0], [0.07051282376050949, 0.03205128386616707, 0.044871795922517776, 0.012820512987673283, 0.05128205195069313, 0.07051282376050949, 0.11538461595773697, 0.07051282376050949, 0.057692307978868484, 0.10897435992956161, 0.11538461595773697, 0.10897435992956161, 0.13461539149284363, 0.10897435992956161, 0.1666666716337204, 0.0833333358168602, 0.1666666716337204, 0.23076923191547394, 0.17307692766189575, 0.14743590354919434, 0.10897435992956161, 0.10256410390138626, 0.21153846383094788, 0.012820512987673283, 0.05128205195069313, 0.11538461595773697, 0.1666666716337204, 0.26923078298568726, 0.23076923191547394, 0.3076923191547394, 0.5192307829856873, 0.2756410241127014, 0.22435897588729858, 0.4038461446762085, 0.3076923191547394, 0.38461539149284363, 0.20512820780277252, 0.35256409645080566, 0.32692307233810425, 0.4294871687889099, 0.3205128312110901, 0.3717948794364929, 0.30128204822540283, 0.36538460850715637, 0.28205129504203796, 0.4743589758872986, 0.38461539149284363, 0.43589743971824646, 0.5288461446762085, 0.5384615659713745, NaN, 0.46794870495796204, 0.6794871687889099, 0.6153846383094788, 0.9294871687889099, 0.6410256624221802, 0.6538461446762085, 0.7628205418586731, 0.7179487347602844, 0.6858974099159241, 0.8397436141967773, 0.8012820482254028, 0.807692289352417, 0.7243589758872986, 0.7307692170143127, 0.807692289352417, 0.7884615659713745, 0.9679487347602844, 0.9679487347602844, 0.942307710647583, 0.8653846383094788, 0.9807692170143127, 1.0], [0.01923076994717121, 0.006410256493836641, 0.012820512987673283, 0.01923076994717121, 0.03205128386616707, 0.012820512987673283, 0.006410256493836641, 0.07051282376050949, 0.01923076994717121, 0.006410256493836641, 0.01923076994717121, 0.14743590354919434, 0.13461539149284363, 0.12179487198591232, 0.044871795922517776, 0.05128205195069313, 0.10897435992956161, 0.0833333358168602, 0.2371794879436493, 0.08974359184503555, 0.14743590354919434, 0.14743590354919434, 0.1794871836900711, 0.16025641560554504, 0.17307692766189575, 0.17307692766189575, 0.2756410241127014, 0.13461539149284363, 0.31410256028175354, 0.14743590354919434, 0.24358974397182465, 0.25, 0.38461539149284363, 0.4038461446762085, 0.25641027092933655, 0.2884615361690521, 0.2756410241127014, 0.41025641560554504, 0.4166666567325592, 0.3205128312110901, 0.3333333432674408, 0.45512819290161133, 0.39743590354919434, 0.4423076808452606, 0.46794870495796204, 0.4615384638309479, 0.44871795177459717, 0.5448718070983887, 0.5128205418586731, 0.48076921701431274, 0.5320512652397156, NaN, 0.5833333134651184, 0.6987179517745972, 0.6153846383094788, 0.6858974099159241, 0.7051281929016113, 0.6346153616905212, 0.7628205418586731, 0.7948718070983887, 0.7820512652397156, 0.7884615659713745, 0.7884615659713745, 0.8525640964508057, 0.8397436141967773, 0.8974359035491943, 0.8525640964508057, 0.9358974099159241, 0.9294871687889099, 0.942307710647583, 0.9294871687889099, 0.9935897588729858, 1.0], [0.07051282376050949, 0.01923076994717121, 0.03205128386616707, 0.012820512987673283, 0.06410256773233414, 0.01923076994717121, 0.06410256773233414, 0.10256410390138626, 0.025641025975346565, 0.06410256773233414, 0.08974359184503555, 0.21153846383094788, 0.14102564752101898, 0.19871795177459717, 0.18589743971824646, 0.09615384787321091, 0.10256410390138626, 0.21153846383094788, 0.17307692766189575, 0.12179487198591232, 0.20512820780277252, 0.20512820780277252, 0.13461539149284363, 0.23076923191547394, 0.21794871985912323, 0.21153846383094788, 0.03846153989434242, 0.26923078298568726, 0.1794871836900711, 0.29487180709838867, 0.3910256326198578, 0.25, 0.2371794879436493, 0.2884615361690521, 0.25641027092933655, 0.32692307233810425, 0.28205129504203796, 0.26923078298568726, 0.07692307978868484, 0.3782051205635071, 0.29487180709838867, 0.3589743673801422, 0.3076923191547394, 0.4166666567325592, 0.31410256028175354, 0.3782051205635071, 0.3589743673801422, 0.41025641560554504, 0.33974358439445496, 0.4615384638309479, 0.3205128312110901, 0.4166666567325592, NaN, 0.46794870495796204, 0.5256410241127014, 0.6730769276618958, 0.6346153616905212, 0.6217948794364929, 0.5, 0.6282051205635071, 0.5512820482254028, 0.7628205418586731, 0.7564102411270142, 0.6474359035491943, 0.7051281929016113, 0.7435897588729858, 0.7628205418586731, 0.6858974099159241, 0.7435897588729858, 0.9230769276618958, 0.9230769276618958, 0.8974359035491943, 1.0], [0.06410256773233414, 0.006410256493836641, 0.012820512987673283, 0.012820512987673283, 0.05128205195069313, 0.03205128386616707, 0.03846153989434242, 0.10256410390138626, 0.03846153989434242, 0.025641025975346565, 0.03205128386616707, 0.14743590354919434, 0.10256410390138626, 0.0833333358168602, 0.05128205195069313, 0.03846153989434242, 0.07051282376050949, 0.06410256773233414, 0.13461539149284363, 0.06410256773233414, 0.07692307978868484, 0.07692307978868484, 0.0833333358168602, 0.10897435992956161, 0.12820513546466827, 0.13461539149284363, 0.14743590354919434, 0.11538461595773697, 0.21153846383094788, 0.10897435992956161, 0.26923078298568726, 0.07692307978868484, 0.25641027092933655, 0.3461538553237915, 0.14102564752101898, 0.10256410390138626, 0.13461539149284363, 0.28205129504203796, 0.21794871985912323, 0.30128204822540283, 0.11538461595773697, 0.3205128312110901, 0.23076923191547394, 0.36538460850715637, 0.25641027092933655, 0.1794871836900711, 0.21794871985912323, 0.3910256326198578, 0.3782051205635071, 0.3333333432674408, 0.38461539149284363, 0.30128204822540283, 0.5320512652397156, NaN, 0.5256410241127014, 0.5769230723381042, 0.6410256624221802, 0.6153846383094788, 0.6602563858032227, 0.6858974099159241, 0.49358972907066345, 0.7692307829856873, 0.7756410241127014, 0.7307692170143127, 0.6474359035491943, 0.7115384340286255, 0.6858974099159241, 0.8589743375778198, 0.8141025900840759, 0.8974359035491943, 0.9294871687889099, 0.942307710647583, 1.0], [0.057692307978868484, 0.03846153989434242, 0.057692307978868484, 0.012820512987673283, 0.05128205195069313, 0.06410256773233414, 0.06410256773233414, 0.06410256773233414, 0.05128205195069313, 0.11538461595773697, 0.10897435992956161, 0.10897435992956161, 0.12179487198591232, 0.07051282376050949, 0.10256410390138626, 0.07692307978868484, 0.10897435992956161, 0.14102564752101898, 0.13461539149284363, 0.07692307978868484, 0.07051282376050949, 0.07051282376050949, 0.0833333358168602, 0.01923076994717121, 0.03205128386616707, 0.06410256773233414, 0.11538461595773697, 0.1538461595773697, 0.12179487198591232, 0.23076923191547394, 0.39743590354919434, 0.16025641560554504, 0.14102564752101898, 0.22435897588729858, 0.16025641560554504, 0.33974358439445496, 0.057692307978868484, 0.19230769574642181, 0.1666666716337204, 0.2371794879436493, 0.20512820780277252, 0.19230769574642181, 0.12820513546466827, 0.31410256028175354, 0.19871795177459717, 0.3589743673801422, 0.2371794879436493, 0.33974358439445496, 0.07051282376050949, 0.28205129504203796, 0.07051282376050949, 0.38461539149284363, 0.4743589758872986, 0.4743589758872986, NaN, 0.5, 0.4615384638309479, 0.557692289352417, 0.31410256028175354, 0.6089743375778198, 0.6410256624221802, 0.7179487347602844, 0.7371794581413269, 0.6153846383094788, 0.6474359035491943, 0.6987179517745972, 0.6794871687889099, 0.75, 0.6666666865348816, 0.8205128312110901, 0.8525640964508057, 0.9102563858032227, 1.0], [0.006410256493836641, 0.012820512987673283, 0.012820512987673283, 0.006410256493836641, 0.0, 0.0, 0.03846153989434242, 0.006410256493836641, 0.0, 0.06410256773233414, 0.07051282376050949, 0.0, 0.01923076994717121, 0.09615384787321091, 0.10897435992956161, 0.07692307978868484, 0.05128205195069313, 0.12179487198591232, 0.06410256773233414, 0.0833333358168602, 0.0833333358168602, 0.08974359184503555, 0.07692307978868484, 0.09615384787321091, 0.10897435992956161, 0.07692307978868484, 0.08974359184503555, 0.22435897588729858, 0.03205128386616707, 0.2371794879436493, 0.3461538553237915, 0.19871795177459717, 0.012820512987673283, 0.19871795177459717, 0.12820513546466827, 0.24358974397182465, 0.14743590354919434, 0.10256410390138626, 0.13461539149284363, 0.13461539149284363, 0.23076923191547394, 0.16987179219722748, 0.12820513546466827, 0.3910256326198578, 0.14102564752101898, 0.2756410241127014, 0.2628205120563507, 0.4038461446762085, 0.3589743673801422, 0.3717948794364929, 0.3589743673801422, 0.31410256028175354, 0.32692307233810425, 0.42307692766189575, 0.5, NaN, 0.6089743375778198, 0.46794870495796204, 0.5256410241127014, 0.5256410241127014, 0.4294871687889099, 0.5705128312110901, 0.5705128312110901, 0.5833333134651184, 0.7115384340286255, 0.7435897588729858, 0.7179487347602844, 0.6794871687889099, 0.6602563858032227, 0.8333333134651184, 0.8910256624221802, 0.8717948794364929, 1.0], [0.06410256773233414, 0.06410256773233414, 0.07051282376050949, 0.01923076994717121, 0.044871795922517776, 0.05128205195069313, 0.11538461595773697, 0.06410256773233414, 0.05128205195069313, 0.07692307978868484, 0.07051282376050949, 0.07692307978868484, 0.03205128386616707, 0.1666666716337204, 0.10897435992956161, 0.12820513546466827, 0.0833333358168602, 0.14743590354919434, 0.03846153989434242, 0.14743590354919434, 0.14743590354919434, 0.14743590354919434, 0.07692307978868484, 0.14102564752101898, 0.13461539149284363, 0.14102564752101898, 0.0833333358168602, 0.19230769574642181, 0.044871795922517776, 0.22435897588729858, 0.41025641560554504, 0.2371794879436493, 0.08974359184503555, 0.14102564752101898, 0.22435897588729858, 0.25641027092933655, 0.19230769574642181, 0.06410256773233414, 0.12179487198591232, 0.14743590354919434, 0.2884615361690521, 0.14743590354919434, 0.18589743971824646, 0.3461538553237915, 0.17307692766189575, 0.29487180709838867, 0.23076923191547394, 0.3205128312110901, 0.3461538553237915, 0.22435897588729858, 0.3461538553237915, 0.29487180709838867, 0.36538460850715637, 0.3589743673801422, 0.5384615659713745, 0.3910256326198578, NaN, 0.4294871687889099, 0.4166666567325592, 0.4615384638309479, 0.45512819290161133, 0.45512819290161133, 0.46794870495796204, 0.46794870495796204, 0.6474359035491943, 0.6858974099159241, 0.6987179517745972, 0.5961538553237915, 0.5641025900840759, 0.6410256624221802, 0.9358974099159241, 0.7051281929016113, 0.9935897588729858], [0.06410256773233414, 0.01923076994717121, 0.012820512987673283, 0.006410256493836641, 0.03205128386616707, 0.006410256493836641, 0.012820512987673283, 0.0833333358168602, 0.025641025975346565, 0.012820512987673283, 0.01923076994717121, 0.057692307978868484, 0.03846153989434242, 0.08974359184503555, 0.057692307978868484, 0.025641025975346565, 0.044871795922517776, 0.07051282376050949, 0.03846153989434242, 0.025641025975346565, 0.03846153989434242, 0.044871795922517776, 0.03846153989434242, 0.07692307978868484, 0.06410256773233414, 0.12820513546466827, 0.025641025975346565, 0.13461539149284363, 0.05128205195069313, 0.16025641560554504, 0.32692307233810425, 0.12179487198591232, 0.09615384787321091, 0.1794871836900711, 0.12820513546466827, 0.22435897588729858, 0.10256410390138626, 0.0833333358168602, 0.057692307978868484, 0.07051282376050949, 0.17307692766189575, 0.13461539149284363, 0.10897435992956161, 0.4038461446762085, 0.17307692766189575, 0.30128204822540283, 0.19871795177459717, 0.4166666567325592, 0.24358974397182465, 0.24358974397182465, 0.2371794879436493, 0.36538460850715637, 0.3782051205635071, 0.38461539149284363, 0.4423076808452606, 0.5320512652397156, 0.5705128312110901, NaN, 0.4743589758872986, 0.5705128312110901, 0.4871794879436493, 0.6730769276618958, 0.6538461446762085, 0.5897436141967773, 0.6794871687889099, 0.7692307829856873, 0.75, 0.75, 0.7628205418586731, 0.8974359035491943, 0.9166666865348816, 0.8974359035491943, 1.0], [0.0, 0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.0, 0.0, 0.006410256493836641, 0.025641025975346565, 0.0, 0.0, 0.01923076994717121, 0.0, 0.0, 0.07051282376050949, 0.0833333358168602, 0.05128205195069313, 0.012820512987673283, 0.14102564752101898, 0.01923076994717121, 0.07051282376050949, 0.07051282376050949, 0.0833333358168602, 0.03205128386616707, 0.05128205195069313, 0.05128205195069313, 0.006410256493836641, 0.006410256493836641, 0.17307692766189575, 0.0, 0.18589743971824646, 0.3461538553237915, 0.18589743971824646, 0.044871795922517776, 0.07692307978868484, 0.07051282376050949, 0.19871795177459717, 0.10897435992956161, 0.10897435992956161, 0.05128205195069313, 0.22435897588729858, 0.21153846383094788, 0.19871795177459717, 0.12820513546466827, 0.20512820780277252, 0.1538461595773697, 0.25, 0.23076923191547394, 0.23076923191547394, 0.29487180709838867, 0.35256409645080566, 0.28205129504203796, 0.2371794879436493, 0.5, 0.33974358439445496, 0.6858974099159241, 0.4743589758872986, 0.5833333134651184, 0.5256410241127014, NaN, 0.557692289352417, 0.557692289352417, 0.5769230723381042, 0.5833333134651184, 0.557692289352417, 0.6153846383094788, 0.7179487347602844, 0.7115384340286255, 0.7820512652397156, 0.7179487347602844, 0.7307692170143127, 0.8461538553237915, 0.8589743375778198, 1.0], [0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.025641025975346565, 0.01923076994717121, 0.012820512987673283, 0.0, 0.012820512987673283, 0.012820512987673283, 0.006410256493836641, 0.012820512987673283, 0.12820513546466827, 0.13461539149284363, 0.05128205195069313, 0.03846153989434242, 0.03846153989434242, 0.0833333358168602, 0.025641025975346565, 0.1794871836900711, 0.044871795922517776, 0.05128205195069313, 0.05128205195069313, 0.10897435992956161, 0.10897435992956161, 0.10256410390138626, 0.17307692766189575, 0.03846153989434242, 0.057692307978868484, 0.22435897588729858, 0.0833333358168602, 0.18589743971824646, 0.057692307978868484, 0.2756410241127014, 0.39743590354919434, 0.21794871985912323, 0.12179487198591232, 0.1538461595773697, 0.2756410241127014, 0.12179487198591232, 0.3205128312110901, 0.10256410390138626, 0.3076923191547394, 0.2628205120563507, 0.1794871836900711, 0.21794871985912323, 0.25, 0.20512820780277252, 0.2371794879436493, 0.30128204822540283, 0.3782051205635071, 0.31410256028175354, 0.20512820780277252, 0.3717948794364929, 0.31410256028175354, 0.3910256326198578, 0.4743589758872986, 0.5384615659713745, 0.4294871687889099, 0.4423076808452606, NaN, 0.4294871687889099, 0.4294871687889099, 0.45512819290161133, 0.7596153616905212, 0.45512819290161133, 0.4743589758872986, 0.5448718070983887, 0.44871795177459717, 0.557692289352417, 0.6282051205635071, 0.7692307829856873, 0.9487179517745972, 0.9358974099159241], [0.057692307978868484, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.012820512987673283, 0.057692307978868484, 0.025641025975346565, 0.057692307978868484, 0.06410256773233414, 0.057692307978868484, 0.11538461595773697, 0.01923076994717121, 0.057692307978868484, 0.0, 0.01923076994717121, 0.057692307978868484, 0.12179487198591232, 0.0, 0.006410256493836641, 0.006410256493836641, 0.025641025975346565, 0.0, 0.0, 0.006410256493836641, 0.1538461595773697, 0.06410256773233414, 0.08974359184503555, 0.07051282376050949, 0.3205128312110901, 0.05128205195069313, 0.14743590354919434, 0.13461539149284363, 0.01923076994717121, 0.13461539149284363, 0.0, 0.19871795177459717, 0.19230769574642181, 0.19230769574642181, 0.08974359184503555, 0.19230769574642181, 0.03846153989434242, 0.25641027092933655, 0.21153846383094788, 0.18589743971824646, 0.13461539149284363, 0.25641027092933655, 0.1666666716337204, 0.22435897588729858, 0.16025641560554504, 0.21794871985912323, 0.44871795177459717, 0.5064102411270142, 0.3589743673801422, 0.5705128312110901, 0.5448718070983887, 0.5128205418586731, 0.4423076808452606, 0.5705128312110901, NaN, 0.7051281929016113, 0.6987179517745972, 0.5769230723381042, 0.6730769276618958, 0.6987179517745972, 0.6987179517745972, 0.6794871687889099, 0.7115384340286255, 0.7884615659713745, 0.7692307829856873, 0.7820512652397156, 0.9358974099159241], [0.057692307978868484, 0.012820512987673283, 0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.0, 0.012820512987673283, 0.06410256773233414, 0.01923076994717121, 0.025641025975346565, 0.025641025975346565, 0.11538461595773697, 0.08974359184503555, 0.057692307978868484, 0.057692307978868484, 0.01923076994717121, 0.0833333358168602, 0.07051282376050949, 0.08974359184503555, 0.03205128386616707, 0.044871795922517776, 0.03846153989434242, 0.07051282376050949, 0.05128205195069313, 0.044871795922517776, 0.12820513546466827, 0.03205128386616707, 0.06410256773233414, 0.0833333358168602, 0.07051282376050949, 0.25, 0.06410256773233414, 0.07051282376050949, 0.22435897588729858, 0.14743590354919434, 0.09615384787321091, 0.07692307978868484, 0.0833333358168602, 0.07051282376050949, 0.12179487198591232, 0.07692307978868484, 0.11538461595773697, 0.11538461595773697, 0.2756410241127014, 0.20512820780277252, 0.12820513546466827, 0.08974359184503555, 0.3076923191547394, 0.20512820780277252, 0.18589743971824646, 0.19871795177459717, 0.21153846383094788, 0.2371794879436493, 0.23076923191547394, 0.28205129504203796, 0.4294871687889099, 0.5448718070983887, 0.32692307233810425, 0.42307692766189575, 0.5705128312110901, 0.29487180709838867, NaN, 0.5416666865348816, 0.6217948794364929, 0.5384615659713745, 0.5769230723381042, 0.557692289352417, 0.6089743375778198, 0.4615384638309479, 1.0, 0.8589743375778198, 0.9102563858032227, 0.9871794581413269], [0.06410256773233414, 0.012820512987673283, 0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.01923076994717121, 0.01923076994717121, 0.06410256773233414, 0.03205128386616707, 0.03205128386616707, 0.03846153989434242, 0.11538461595773697, 0.08974359184503555, 0.057692307978868484, 0.044871795922517776, 0.025641025975346565, 0.08974359184503555, 0.044871795922517776, 0.08974359184503555, 0.03846153989434242, 0.05128205195069313, 0.044871795922517776, 0.06410256773233414, 0.05128205195069313, 0.03846153989434242, 0.13461539149284363, 0.03205128386616707, 0.06410256773233414, 0.0833333358168602, 0.06410256773233414, 0.2628205120563507, 0.057692307978868484, 0.07051282376050949, 0.2371794879436493, 0.14743590354919434, 0.08974359184503555, 0.0833333358168602, 0.08974359184503555, 0.06410256773233414, 0.13461539149284363, 0.07051282376050949, 0.12179487198591232, 0.12820513546466827, 0.2756410241127014, 0.20512820780277252, 0.13461539149284363, 0.0833333358168602, 0.29487180709838867, 0.19871795177459717, 0.21153846383094788, 0.19230769574642181, 0.21153846383094788, 0.24358974397182465, 0.22435897588729858, 0.2628205120563507, 0.4294871687889099, 0.5320512652397156, 0.3461538553237915, 0.4166666567325592, 0.5448718070983887, 0.30128204822540283, 0.4583333432674408, NaN, 0.6089743375778198, 0.5256410241127014, 0.5641025900840759, 0.557692289352417, 0.5961538553237915, 0.45512819290161133, 1.0, 0.8589743375778198, 0.8910256624221802, 0.9871794581413269], [0.006410256493836641, 0.0, 0.006410256493836641, 0.01923076994717121, 0.012820512987673283, 0.012820512987673283, 0.006410256493836641, 0.01923076994717121, 0.006410256493836641, 0.012820512987673283, 0.006410256493836641, 0.14102564752101898, 0.12179487198591232, 0.03846153989434242, 0.01923076994717121, 0.03205128386616707, 0.03846153989434242, 0.025641025975346565, 0.18589743971824646, 0.03205128386616707, 0.03846153989434242, 0.03846153989434242, 0.06410256773233414, 0.07051282376050949, 0.07051282376050949, 0.10897435992956161, 0.03205128386616707, 0.057692307978868484, 0.22435897588729858, 0.06410256773233414, 0.18589743971824646, 0.025641025975346565, 0.26923078298568726, 0.3333333432674408, 0.16025641560554504, 0.07692307978868484, 0.12820513546466827, 0.29487180709838867, 0.11538461595773697, 0.2884615361690521, 0.07051282376050949, 0.31410256028175354, 0.22435897588729858, 0.13461539149284363, 0.1794871836900711, 0.19871795177459717, 0.13461539149284363, 0.19230769574642181, 0.28205129504203796, 0.3333333432674408, 0.2756410241127014, 0.14743590354919434, 0.35256409645080566, 0.26923078298568726, 0.38461539149284363, 0.4166666567325592, 0.5320512652397156, 0.41025641560554504, 0.4423076808452606, 0.24038460850715637, 0.42307692766189575, 0.3782051205635071, 0.3910256326198578, NaN, 0.42307692766189575, 0.45512819290161133, 0.5192307829856873, 0.41025641560554504, 0.5512820482254028, 0.6153846383094788, 0.7564102411270142, 0.8589743375778198, 0.9358974099159241], [0.006410256493836641, 0.012820512987673283, 0.012820512987673283, 0.006410256493836641, 0.03205128386616707, 0.025641025975346565, 0.012820512987673283, 0.025641025975346565, 0.01923076994717121, 0.01923076994717121, 0.012820512987673283, 0.057692307978868484, 0.03846153989434242, 0.057692307978868484, 0.044871795922517776, 0.025641025975346565, 0.01923076994717121, 0.03846153989434242, 0.06410256773233414, 0.025641025975346565, 0.03846153989434242, 0.03846153989434242, 0.03205128386616707, 0.057692307978868484, 0.057692307978868484, 0.10256410390138626, 0.0833333358168602, 0.06410256773233414, 0.07051282376050949, 0.057692307978868484, 0.23076923191547394, 0.07692307978868484, 0.0833333358168602, 0.14102564752101898, 0.10256410390138626, 0.1538461595773697, 0.07692307978868484, 0.0833333358168602, 0.10256410390138626, 0.06410256773233414, 0.09615384787321091, 0.10256410390138626, 0.10256410390138626, 0.31410256028175354, 0.11538461595773697, 0.20512820780277252, 0.11538461595773697, 0.3333333432674408, 0.26923078298568726, 0.12179487198591232, 0.26923078298568726, 0.16025641560554504, 0.29487180709838867, 0.35256409645080566, 0.35256409645080566, 0.2884615361690521, 0.35256409645080566, 0.3205128312110901, 0.38461539149284363, 0.5448718070983887, 0.32692307233810425, 0.4615384638309479, 0.4743589758872986, 0.5769230723381042, NaN, 0.7884615659713745, 0.8141025900840759, 0.6025640964508057, 0.6538461446762085, 0.5320512652397156, 0.6474359035491943, 0.6794871687889099, 0.942307710647583], [0.025641025975346565, 0.025641025975346565, 0.025641025975346565, 0.025641025975346565, 0.044871795922517776, 0.025641025975346565, 0.025641025975346565, 0.03205128386616707, 0.025641025975346565, 0.025641025975346565, 0.025641025975346565, 0.057692307978868484, 0.03846153989434242, 0.05128205195069313, 0.03205128386616707, 0.025641025975346565, 0.025641025975346565, 0.044871795922517776, 0.05128205195069313, 0.025641025975346565, 0.03846153989434242, 0.044871795922517776, 0.03205128386616707, 0.044871795922517776, 0.044871795922517776, 0.0833333358168602, 0.10256410390138626, 0.03846153989434242, 0.044871795922517776, 0.06410256773233414, 0.2371794879436493, 0.06410256773233414, 0.06410256773233414, 0.12179487198591232, 0.0833333358168602, 0.11538461595773697, 0.05128205195069313, 0.0833333358168602, 0.07692307978868484, 0.05128205195069313, 0.0833333358168602, 0.07692307978868484, 0.07051282376050949, 0.2628205120563507, 0.08974359184503555, 0.16025641560554504, 0.09615384787321091, 0.28205129504203796, 0.21153846383094788, 0.0833333358168602, 0.19230769574642181, 0.10256410390138626, 0.25641027092933655, 0.2884615361690521, 0.30128204822540283, 0.25641027092933655, 0.31410256028175354, 0.23076923191547394, 0.28205129504203796, 0.5256410241127014, 0.30128204822540283, 0.42307692766189575, 0.43589743971824646, 0.5448718070983887, 0.21153846383094788, NaN, 0.6698718070983887, 0.5705128312110901, 0.5384615659713745, 0.4743589758872986, 0.6410256624221802, 0.6538461446762085, 0.9358974099159241], [0.03205128386616707, 0.03205128386616707, 0.025641025975346565, 0.03205128386616707, 0.03205128386616707, 0.03205128386616707, 0.025641025975346565, 0.03205128386616707, 0.03205128386616707, 0.03205128386616707, 0.03846153989434242, 0.05128205195069313, 0.044871795922517776, 0.044871795922517776, 0.044871795922517776, 0.03205128386616707, 0.044871795922517776, 0.05128205195069313, 0.044871795922517776, 0.03205128386616707, 0.03205128386616707, 0.03846153989434242, 0.05128205195069313, 0.057692307978868484, 0.057692307978868484, 0.07692307978868484, 0.10256410390138626, 0.057692307978868484, 0.057692307978868484, 0.07051282376050949, 0.2371794879436493, 0.06410256773233414, 0.06410256773233414, 0.10897435992956161, 0.08974359184503555, 0.12179487198591232, 0.07692307978868484, 0.07692307978868484, 0.08974359184503555, 0.05128205195069313, 0.08974359184503555, 0.0833333358168602, 0.07692307978868484, 0.2628205120563507, 0.09615384787321091, 0.17307692766189575, 0.12179487198591232, 0.2756410241127014, 0.22435897588729858, 0.11538461595773697, 0.21153846383094788, 0.14743590354919434, 0.2371794879436493, 0.31410256028175354, 0.3205128312110901, 0.28205129504203796, 0.30128204822540283, 0.25, 0.2884615361690521, 0.45512819290161133, 0.30128204822540283, 0.4423076808452606, 0.4423076808452606, 0.48076921701431274, 0.18589743971824646, 0.33012819290161133, NaN, 0.5064102411270142, 0.5384615659713745, 0.5064102411270142, 0.6153846383094788, 0.5897436141967773, 0.7564102411270142], [0.0, 0.0, 0.006410256493836641, 0.006410256493836641, 0.0, 0.006410256493836641, 0.006410256493836641, 0.012820512987673283, 0.0, 0.0, 0.0, 0.057692307978868484, 0.057692307978868484, 0.01923076994717121, 0.01923076994717121, 0.01923076994717121, 0.05128205195069313, 0.025641025975346565, 0.057692307978868484, 0.025641025975346565, 0.012820512987673283, 0.012820512987673283, 0.057692307978868484, 0.0, 0.0, 0.006410256493836641, 0.03205128386616707, 0.006410256493836641, 0.09615384787321091, 0.01923076994717121, 0.19871795177459717, 0.01923076994717121, 0.10897435992956161, 0.08974359184503555, 0.057692307978868484, 0.03846153989434242, 0.03205128386616707, 0.10897435992956161, 0.07051282376050949, 0.12179487198591232, 0.03205128386616707, 0.12179487198591232, 0.05128205195069313, 0.006410256493836641, 0.13461539149284363, 0.07051282376050949, 0.03205128386616707, 0.012820512987673283, 0.05128205195069313, 0.10897435992956161, 0.03205128386616707, 0.06410256773233414, 0.31410256028175354, 0.14102564752101898, 0.25, 0.3205128312110901, 0.4038461446762085, 0.25, 0.21794871985912323, 0.5512820482254028, 0.3205128312110901, 0.3910256326198578, 0.4038461446762085, 0.5897436141967773, 0.39743590354919434, 0.4294871687889099, 0.49358972907066345, NaN, 0.38461539149284363, 0.7179487347602844, 0.7948718070983887, 0.9038461446762085, 0.9935897588729858], [0.03205128386616707, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.05128205195069313, 0.006410256493836641, 0.0, 0.0, 0.0, 0.0, 0.01923076994717121, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.0, 0.006410256493836641, 0.006410256493836641, 0.012820512987673283, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.01923076994717121, 0.0, 0.2371794879436493, 0.006410256493836641, 0.05128205195069313, 0.0, 0.025641025975346565, 0.006410256493836641, 0.0, 0.03205128386616707, 0.006410256493836641, 0.044871795922517776, 0.006410256493836641, 0.07051282376050949, 0.012820512987673283, 0.07692307978868484, 0.07051282376050949, 0.03205128386616707, 0.01923076994717121, 0.16025641560554504, 0.06410256773233414, 0.10897435992956161, 0.03205128386616707, 0.07051282376050949, 0.25641027092933655, 0.18589743971824646, 0.3333333432674408, 0.33974358439445496, 0.43589743971824646, 0.2371794879436493, 0.28205129504203796, 0.4423076808452606, 0.2884615361690521, 0.5384615659713745, 0.5448718070983887, 0.44871795177459717, 0.3461538553237915, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, NaN, 0.7243589758872986, 0.807692289352417, 0.7756410241127014, 1.0], [0.006410256493836641, 0.012820512987673283, 0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.0, 0.0, 0.006410256493836641, 0.0, 0.006410256493836641, 0.012820512987673283, 0.057692307978868484, 0.06410256773233414, 0.03205128386616707, 0.006410256493836641, 0.006410256493836641, 0.03205128386616707, 0.012820512987673283, 0.07051282376050949, 0.006410256493836641, 0.012820512987673283, 0.012820512987673283, 0.03205128386616707, 0.006410256493836641, 0.006410256493836641, 0.025641025975346565, 0.012820512987673283, 0.012820512987673283, 0.06410256773233414, 0.01923076994717121, 0.05128205195069313, 0.012820512987673283, 0.057692307978868484, 0.10256410390138626, 0.05128205195069313, 0.025641025975346565, 0.006410256493836641, 0.06410256773233414, 0.01923076994717121, 0.06410256773233414, 0.01923076994717121, 0.07692307978868484, 0.012820512987673283, 0.06410256773233414, 0.12820513546466827, 0.044871795922517776, 0.025641025975346565, 0.06410256773233414, 0.0833333358168602, 0.057692307978868484, 0.057692307978868484, 0.057692307978868484, 0.07692307978868484, 0.10256410390138626, 0.1794871836900711, 0.1666666716337204, 0.3589743673801422, 0.10256410390138626, 0.26923078298568726, 0.3717948794364929, 0.21153846383094788, 0.0, 0.0, 0.38461539149284363, 0.46794870495796204, 0.5256410241127014, 0.49358972907066345, 0.28205129504203796, 0.2756410241127014, NaN, 0.6474359035491943, 0.6987179517745972, 0.9615384340286255], [0.06410256773233414, 0.03205128386616707, 0.044871795922517776, 0.01923076994717121, 0.03846153989434242, 0.05128205195069313, 0.05128205195069313, 0.057692307978868484, 0.044871795922517776, 0.044871795922517776, 0.025641025975346565, 0.05128205195069313, 0.025641025975346565, 0.044871795922517776, 0.03205128386616707, 0.044871795922517776, 0.06410256773233414, 0.03205128386616707, 0.03846153989434242, 0.044871795922517776, 0.06410256773233414, 0.06410256773233414, 0.025641025975346565, 0.044871795922517776, 0.044871795922517776, 0.09615384787321091, 0.012820512987673283, 0.025641025975346565, 0.0, 0.03846153989434242, 0.22435897588729858, 0.057692307978868484, 0.006410256493836641, 0.11538461595773697, 0.10897435992956161, 0.057692307978868484, 0.05128205195069313, 0.0, 0.025641025975346565, 0.044871795922517776, 0.057692307978868484, 0.025641025975346565, 0.07051282376050949, 0.10256410390138626, 0.09615384787321091, 0.057692307978868484, 0.025641025975346565, 0.10256410390138626, 0.13461539149284363, 0.07051282376050949, 0.13461539149284363, 0.07051282376050949, 0.07692307978868484, 0.07051282376050949, 0.14743590354919434, 0.10897435992956161, 0.06410256773233414, 0.0833333358168602, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.14102564752101898, 0.14102564752101898, 0.24358974397182465, 0.35256409645080566, 0.3589743673801422, 0.38461539149284363, 0.20512820780277252, 0.19230769574642181, 0.35256409645080566, NaN, 0.35256409645080566, 0.9294871687889099], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03205128386616707, 0.044871795922517776, 0.006410256493836641, 0.0, 0.0, 0.006410256493836641, 0.0, 0.057692307978868484, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.05128205195069313, 0.0, 0.1794871836900711, 0.006410256493836641, 0.07692307978868484, 0.06410256773233414, 0.025641025975346565, 0.006410256493836641, 0.0, 0.07051282376050949, 0.006410256493836641, 0.03205128386616707, 0.006410256493836641, 0.09615384787321091, 0.0, 0.0, 0.01923076994717121, 0.012820512987673283, 0.0, 0.0, 0.025641025975346565, 0.01923076994717121, 0.01923076994717121, 0.006410256493836641, 0.10256410390138626, 0.057692307978868484, 0.08974359184503555, 0.12820513546466827, 0.29487180709838867, 0.10256410390138626, 0.14102564752101898, 0.05128205195069313, 0.21794871985912323, 0.08974359184503555, 0.10897435992956161, 0.14102564752101898, 0.3205128312110901, 0.3461538553237915, 0.41025641560554504, 0.09615384787321091, 0.22435897588729858, 0.30128204822540283, 0.6474359035491943, NaN, 0.9166666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.06410256773233414, 0.06410256773233414, 0.012820512987673283, 0.012820512987673283, 0.06410256773233414, 0.057692307978868484, 0.06410256773233414, 0.24358974397182465, 0.006410256493836641, 0.0, 0.03846153989434242, 0.07051282376050949, 0.0833333358168602, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9080306225},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8863514927},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8847044131},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8758903135},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.840722933},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8369836153},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.826255337},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8073361807},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7894408825},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7843215793},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7658030594},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7159900321},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7119836189},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7082443031},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7005430934},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.685585827},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6692040622},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6664440876},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6631054114},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6586093296},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6428507833},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6377759971},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6376869669},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6358173113},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6294960819},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.622685188},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5982015663},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5948183736},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5876513523},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5777243591},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5600961538},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5564903879},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.554131053},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5429131042},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5413105438},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5372596145},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5334757867},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.5183849747},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5151353266},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5071225072},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5018251445},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4796563382},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.463853276},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.461627492},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.460915243},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4517895332},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4516559852},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4003739338},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.3975694463},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3963675252},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3952546327},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.3949430214},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3511396034},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3119658132},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.2857906001},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2729255702},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2682514272},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2649572669},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2540954411},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2463051993},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2410078366},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2195067676},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2179932345},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.217280984},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2065527088},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1765936634},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1714298442},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1590099728},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1504629647},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.115028491},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1073717964},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0787927367},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0111289177}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e8deebf93d65aa640ab7dc846abe3851e5677958
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_all/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),TabPFN-3 (default),RealTabPFN-2.5 (default),TabICLv2 (default),LimiX (default),RealMLP (tuned + ensembled),RealMLP (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabM (tuned + ensembled),CatBoost (tuned + ensembled),ModernNCA (tuned + ensembled),TabM (tuned),TabDPT (tuned),CatBoost (tuned),XGBoost (tuned),XGBoost (tuned + ensembled),ModernNCA (tuned),LightGBM (tuned + ensembled),LightGBM (tuned),iLTM (tuned + ensembled),xRFM (tuned + ensembled),BetaTabPFN (default),ExtraTrees (tuned + ensembled),TabM (default),SAP-RPT-OSS (default),EBM (tuned + ensembled),RandomForest (tuned + ensembled),TabDPT (default),iLTM (tuned),TorchMLP (tuned + ensembled),CatBoost (default),ExtraTrees (tuned),xRFM (tuned),TabSwift (default),EBM (tuned),RandomForest (tuned),XGBoost (default),FastaiMLP (tuned + ensembled),ModernNCA (default),TorchMLP (tuned),EBM (default),FastaiMLP (tuned),ChimeraBoost (tuned),LightGBM (default),ChimeraBoost (tuned + ensembled),RealMLP (default),xRFM (default),TorchMLP (default),ChimeraBoost (default),RandomForest (default),KNN (tuned + ensembled),OrionMSP (default),ExtraTrees (default),Linear (tuned + ensembled),iLTM (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),Linear (tuned),PerpetualBooster (tuned + ensembled),PerpetualBooster (tuned),PerpetualBooster (default),FastaiMLP (default),TabFlex (default),TabSTAR (default),KNN (tuned),Linear (default),KNN (default)
+TabFM (default),,0.6666667,0.6602564,0.7692308,0.8141026,0.80128205,0.75641024,0.8974359,0.7948718,0.78205127,0.78205127,0.89102566,0.90384614,0.88461536,0.82051283,0.9102564,0.9230769,0.8333333,0.89102566,0.92948717,0.88461536,0.89102566,0.8974359,0.9102564,0.90384614,0.9230769,0.9230769,0.9166667,0.92948717,0.85897434,0.7692308,0.9102564,0.9102564,0.9102564,0.9358974,0.92948717,0.9358974,0.9230769,0.92948717,0.9230769,0.92948717,0.9102564,0.9358974,0.90384614,0.9807692,0.9358974,0.92948717,0.9423077,0.92948717,0.92948717,0.92948717,0.9807692,0.92948717,0.9358974,0.9423077,0.99358976,0.9358974,0.9358974,1.0,0.99358976,0.9423077,0.9423077,0.9358974,0.99358976,0.99358976,0.974359,0.96794873,1.0,0.96794873,0.99358976,0.9358974,1.0,1.0
+RealTabPFN-2.5 (tuned),0.33333334,,0.43589744,0.5448718,0.64102566,0.5833333,0.7019231,0.6730769,0.6923077,0.75641024,0.78205127,0.71153843,0.74358976,0.8525641,0.8525641,0.92948717,0.85897434,0.85897434,0.78846157,0.92948717,0.9102564,0.9102564,0.8717949,0.89102566,0.9166667,0.88461536,0.8974359,0.90384614,0.8717949,0.86538464,0.76282054,0.9551282,0.88461536,0.8076923,0.9551282,0.94871795,0.96794873,0.8974359,0.9230769,0.974359,0.96153843,0.9102564,0.9807692,0.92948717,0.94871795,0.94871795,0.94871795,0.96794873,0.9551282,0.94871795,0.96794873,0.99358976,0.9807692,0.99358976,0.96153843,0.98717946,0.9358974,0.9807692,0.99358976,0.99358976,1.0,0.98717946,0.98717946,1.0,0.98717946,0.974359,0.96794873,1.0,1.0,0.98717946,0.96794873,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.33974358,0.5641026,,0.5833333,0.6474359,0.59615386,0.76282054,0.67948717,0.69871795,0.75,0.76282054,0.7051282,0.73717946,0.85897434,0.8076923,0.9423077,0.86538464,0.8076923,0.75641024,0.92948717,0.90384614,0.9230769,0.88461536,0.8974359,0.90384614,0.89102566,0.88461536,0.9166667,0.8525641,0.8525641,0.75641024,0.9358974,0.8782051,0.78846157,0.9551282,0.9423077,0.9423077,0.8782051,0.9230769,0.974359,0.9423077,0.9102564,0.98717946,0.9358974,0.9358974,0.94871795,0.9358974,0.9551282,0.9423077,0.9358974,0.9551282,0.98717946,0.96794873,0.98717946,0.9423077,0.98717946,0.92948717,0.98717946,0.99358976,0.99358976,1.0,0.99358976,0.99358976,0.99358976,0.98717946,0.974359,0.974359,0.99358976,1.0,0.99358976,0.9551282,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.23076923,0.4551282,0.41666666,,0.6217949,0.5576923,0.51282054,0.59615386,0.74358976,0.73717946,0.75,0.75,0.8269231,0.8076923,0.78205127,0.89102566,0.8717949,0.775641,0.86538464,0.90384614,0.8397436,0.86538464,0.89102566,0.9102564,0.8974359,0.9166667,0.9102564,0.8333333,0.90384614,0.8141026,0.6474359,0.9230769,0.9102564,0.84615386,0.9423077,0.9102564,0.94871795,0.90384614,0.9423077,0.98717946,0.94871795,0.9102564,0.99358976,0.8782051,0.96794873,0.9551282,0.9423077,0.9423077,0.974359,0.974359,0.98717946,0.9807692,0.98717946,0.98717946,0.98717946,0.99358976,0.9807692,0.99358976,0.99358976,0.974359,1.0,0.99358976,0.99358976,0.9807692,0.99358976,0.974359,0.96794873,0.99358976,1.0,0.99358976,0.9807692,1.0,1.0
+TabPFN-2.6 (default),0.18589744,0.35897437,0.3525641,0.37820512,,0.42948717,0.55128205,0.4871795,0.60897434,0.65384614,0.65384614,0.6923077,0.6474359,0.724359,0.7307692,0.8141026,0.86538464,0.76282054,0.71794873,0.8269231,0.88461536,0.90384614,0.8397436,0.85897434,0.89102566,0.92948717,0.80128205,0.84615386,0.8525641,0.80128205,0.6730769,0.8717949,0.8333333,0.78846157,0.94871795,0.9230769,0.9551282,0.84615386,0.85897434,0.9551282,0.9102564,0.8717949,0.9807692,0.94871795,0.9358974,0.9230769,0.92948717,0.9807692,0.9358974,0.9358974,0.94871795,0.96794873,0.9358974,0.94871795,0.94871795,1.0,0.9551282,0.96794873,1.0,0.9807692,1.0,0.99358976,0.99358976,0.98717946,0.96794873,0.9551282,0.96794873,1.0,1.0,0.99358976,0.96153843,1.0,1.0
+TabPFN-3 (default),0.19871795,0.41666666,0.40384614,0.44230768,0.57051283,,0.47435898,0.59615386,0.60897434,0.55128205,0.6282051,0.73717946,0.74358976,0.71153843,0.67948717,0.8333333,0.82051283,0.71153843,0.775641,0.8333333,0.78846157,0.7948718,0.8333333,0.8525641,0.84615386,0.85897434,0.84615386,0.75641024,0.9102564,0.71794873,0.59615386,0.8397436,0.88461536,0.8269231,0.9166667,0.88461536,0.89102566,0.9166667,0.9102564,0.94871795,0.85897434,0.90384614,0.9807692,0.8397436,0.96794873,0.8974359,0.9230769,0.9423077,0.9102564,0.92948717,0.92948717,0.98717946,0.9807692,0.96794873,0.9358974,1.0,0.94871795,0.99358976,1.0,0.98717946,0.99358976,1.0,0.9807692,0.98717946,0.974359,0.974359,0.96794873,0.99358976,1.0,1.0,0.94871795,1.0,1.0
+RealTabPFN-2.5 (default),0.24358974,0.29807693,0.23717949,0.4871795,0.44871795,0.525641,,0.5448718,0.6858974,0.6602564,0.6730769,0.5897436,0.67948717,0.74358976,0.80128205,0.8269231,0.71794873,0.82051283,0.69871795,0.8525641,0.8333333,0.8269231,0.78205127,0.80128205,0.8269231,0.7692308,0.8269231,0.85897434,0.8141026,0.8269231,0.7307692,0.8525641,0.775641,0.71794873,0.90384614,0.90384614,0.8782051,0.8525641,0.8782051,0.9358974,0.85897434,0.85897434,0.9102564,0.9230769,0.9102564,0.90384614,0.8974359,0.9423077,0.8782051,0.9166667,0.88461536,0.99358976,0.9358974,0.96153843,0.9358974,0.96153843,0.88461536,0.98717946,0.99358976,1.0,0.98717946,0.98717946,0.9807692,0.99358976,0.98717946,0.974359,0.974359,0.99358976,0.99358976,1.0,0.94871795,1.0,1.0
+TabICLv2 (default),0.102564104,0.32692307,0.32051283,0.40384614,0.51282054,0.40384614,0.4551282,,0.5833333,0.5833333,0.60897434,0.6923077,0.75,0.6730769,0.6858974,0.78846157,0.75641024,0.6923077,0.75,0.78846157,0.7692308,0.78205127,0.82051283,0.8397436,0.85897434,0.8782051,0.7948718,0.75,0.8269231,0.75,0.53846157,0.8269231,0.8717949,0.76282054,0.88461536,0.8333333,0.89102566,0.84615386,0.8269231,0.9230769,0.85897434,0.8782051,0.92948717,0.8269231,0.9551282,0.88461536,0.9102564,0.90384614,0.9230769,0.92948717,0.92948717,0.92948717,0.8974359,0.8974359,0.9358974,0.99358976,0.9358974,0.9166667,0.974359,0.98717946,0.9423077,0.9358974,0.9358974,0.9807692,0.974359,0.96794873,0.96794873,0.98717946,0.94871795,0.99358976,0.9423077,1.0,1.0
+LimiX (default),0.20512821,0.30769232,0.30128205,0.25641027,0.39102563,0.39102563,0.31410256,0.41666666,,0.51282054,0.5192308,0.49358973,0.5448718,0.6602564,0.65384614,0.74358976,0.65384614,0.6602564,0.6282051,0.75641024,0.775641,0.7692308,0.78205127,0.80128205,0.8076923,0.76282054,0.84615386,0.76282054,0.89102566,0.724359,0.60897434,0.8333333,0.9102564,0.6282051,0.8397436,0.71794873,0.88461536,0.9102564,0.90384614,0.82051283,0.88461536,0.90384614,0.94871795,0.78205127,0.9423077,0.8269231,0.90384614,0.89102566,0.9166667,0.9423077,0.9423077,0.9807692,0.974359,0.96153843,0.94871795,1.0,0.94871795,0.974359,1.0,0.98717946,0.974359,0.9807692,0.96794873,0.99358976,0.9807692,0.974359,0.96794873,1.0,0.99358976,1.0,0.9551282,1.0,1.0
+RealMLP (tuned + ensembled),0.21794872,0.24358974,0.25,0.2628205,0.34615386,0.44871795,0.33974358,0.41666666,0.4871795,,0.52884614,0.53205127,0.5192308,0.6666667,0.6923077,0.73717946,0.6858974,0.69871795,0.525641,0.7692308,0.84615386,0.8717949,0.73717946,0.8269231,0.8269231,0.75641024,0.8333333,0.86538464,0.74358976,0.8782051,0.6282051,0.86538464,0.8141026,0.5769231,0.8525641,0.75641024,0.8782051,0.8076923,0.85897434,0.74358976,0.8717949,0.85897434,0.9358974,0.9102564,0.89102566,0.8076923,0.8782051,0.9166667,0.88461536,0.9166667,0.89102566,0.99358976,0.9358974,0.974359,0.88461536,0.9358974,0.9230769,0.98717946,1.0,0.99358976,0.9423077,0.974359,0.96794873,0.98717946,0.9807692,0.974359,0.96794873,1.0,1.0,0.99358976,0.9551282,1.0,1.0
+RealMLP (tuned),0.21794872,0.21794872,0.23717949,0.25,0.34615386,0.37179488,0.32692307,0.39102563,0.48076922,0.47115386,,0.5192308,0.48076922,0.61538464,0.6217949,0.7051282,0.6923077,0.6602564,0.51282054,0.71153843,0.78205127,0.8076923,0.71153843,0.74358976,0.75641024,0.724359,0.775641,0.8525641,0.73717946,0.8141026,0.6025641,0.84615386,0.78846157,0.5833333,0.8333333,0.80128205,0.8397436,0.775641,0.8397436,0.74358976,0.88461536,0.8076923,0.90384614,0.8782051,0.86538464,0.84615386,0.86538464,0.9358974,0.86538464,0.90384614,0.88461536,0.9807692,0.9102564,0.96794873,0.89102566,0.92948717,0.92948717,0.9807692,0.9807692,0.98717946,0.9358974,0.974359,0.96153843,0.99358976,0.98717946,0.974359,0.96153843,1.0,1.0,0.98717946,0.974359,1.0,1.0
+TabDPT-Turbo (default),0.10897436,0.28846154,0.2948718,0.25,0.30769232,0.2628205,0.41025642,0.30769232,0.50641024,0.4679487,0.48076922,,0.55128205,0.51282054,0.5192308,0.60897434,0.67948717,0.53846157,0.6025641,0.59615386,0.61538464,0.61538464,0.6730769,0.6666667,0.61538464,0.8076923,0.6282051,0.5769231,0.7692308,0.5833333,0.525641,0.6474359,0.75641024,0.80128205,0.8141026,0.71794873,0.75641024,0.78205127,0.6923077,0.8333333,0.71794873,0.775641,0.8525641,0.78846157,0.8397436,0.78205127,0.78205127,0.8717949,0.89102566,0.85897434,0.89102566,0.8525641,0.78846157,0.8525641,0.89102566,1.0,0.9230769,0.9423077,1.0,0.8717949,0.9423077,0.88461536,0.88461536,0.85897434,0.9423077,0.9423077,0.94871795,0.9423077,1.0,0.9423077,0.94871795,0.96794873,1.0
+TabDPT (tuned + ensembled),0.09615385,0.25641027,0.2628205,0.17307693,0.3525641,0.25641027,0.32051283,0.25,0.4551282,0.48076922,0.5192308,0.44871795,,0.525641,0.57051283,0.5448718,0.61538464,0.5769231,0.67948717,0.5769231,0.59615386,0.61538464,0.64102566,0.6730769,0.65384614,0.71153843,0.65384614,0.65384614,0.6923077,0.65384614,0.5192308,0.6923077,0.75,0.76282054,0.74358976,0.6730769,0.74358976,0.78846157,0.75641024,0.8333333,0.724359,0.78846157,0.84615386,0.78205127,0.85897434,0.80128205,0.75,0.8269231,0.8397436,0.86538464,0.86538464,0.86538464,0.85897434,0.8974359,0.8782051,0.9807692,0.96794873,0.96153843,1.0,0.86538464,0.88461536,0.9102564,0.9102564,0.8782051,0.96153843,0.96153843,0.9551282,0.9423077,1.0,0.9358974,0.974359,0.9551282,1.0
+"AutoGluon 1.4 (best, 4h)",0.115384616,0.1474359,0.14102565,0.1923077,0.27564102,0.28846154,0.25641027,0.32692307,0.33974358,0.33333334,0.3846154,0.4871795,0.47435898,,0.50641024,0.6025641,0.61538464,0.5448718,0.50641024,0.6730769,0.61538464,0.6474359,0.6474359,0.73717946,0.724359,0.6858974,0.67948717,0.6666667,0.61538464,0.6858974,0.53846157,0.78846157,0.6730769,0.6217949,0.75641024,0.75641024,0.8269231,0.69871795,0.74358976,0.724359,0.8141026,0.7692308,0.84615386,0.8333333,0.8269231,0.8397436,0.8717949,0.90384614,0.88461536,0.86538464,0.89102566,0.8782051,0.80128205,0.9166667,0.92948717,0.90384614,0.8333333,0.9102564,0.92948717,0.94871795,0.9807692,0.9423077,0.9423077,0.96153843,0.9423077,0.94871795,0.9551282,0.9807692,0.9807692,0.96794873,0.9551282,0.99358976,1.0
+TabM (tuned + ensembled),0.17948718,0.1474359,0.1923077,0.21794872,0.26923078,0.32051283,0.19871795,0.31410256,0.34615386,0.30769232,0.37820512,0.48076922,0.42948717,0.49358973,,0.5,0.5769231,0.6442308,0.50641024,0.5641026,0.6474359,0.6730769,0.5769231,0.6217949,0.6282051,0.5897436,0.64102566,0.775641,0.5448718,0.88461536,0.64102566,0.8141026,0.6025641,0.61538464,0.69871795,0.73717946,0.78205127,0.6025641,0.71794873,0.6474359,0.8397436,0.71153843,0.8333333,0.8269231,0.78205127,0.82051283,0.8397436,0.88461536,0.8076923,0.89102566,0.8333333,0.9551282,0.8141026,0.94871795,0.8974359,0.89102566,0.89102566,0.9423077,0.9166667,0.96153843,0.9423077,0.9423077,0.9551282,0.9807692,0.9551282,0.96794873,0.9551282,0.9807692,1.0,0.99358976,0.96794873,1.0,1.0
+CatBoost (tuned + ensembled),0.08974359,0.070512824,0.057692308,0.10897436,0.18589744,0.16666667,0.17307693,0.21153846,0.25641027,0.2628205,0.2948718,0.39102563,0.4551282,0.3974359,0.5,,0.5192308,0.5641026,0.525641,0.69551283,0.61538464,0.6282051,0.55128205,0.59615386,0.6025641,0.53846157,0.6025641,0.6217949,0.65384614,0.6666667,0.53846157,0.7051282,0.6602564,0.59615386,0.6923077,0.6602564,0.8333333,0.73717946,0.775641,0.74358976,0.78846157,0.78205127,0.9230769,0.7051282,0.8333333,0.80128205,0.8782051,0.8333333,0.8782051,0.89102566,0.9166667,0.94871795,0.90384614,0.96153843,0.9230769,0.9230769,0.8717949,0.974359,0.94871795,0.96153843,1.0,0.9807692,0.974359,0.96794873,0.974359,0.974359,0.96794873,0.9807692,1.0,0.99358976,0.9551282,1.0,1.0
+ModernNCA (tuned + ensembled),0.07692308,0.14102565,0.13461539,0.12820514,0.13461539,0.17948718,0.2820513,0.24358974,0.34615386,0.31410256,0.30769232,0.32051283,0.3846154,0.3846154,0.42307693,0.48076922,,0.4679487,0.44871795,0.51282054,0.53846157,0.55128205,0.6442308,0.59615386,0.60897434,0.525641,0.6602564,0.57051283,0.74358976,0.57051283,0.5576923,0.65384614,0.75641024,0.5769231,0.6923077,0.61538464,0.69871795,0.8141026,0.8076923,0.6730769,0.71794873,0.7692308,0.82051283,0.6858974,0.9102564,0.775641,0.78205127,0.8141026,0.8397436,0.8141026,0.8333333,0.89102566,0.8974359,0.92948717,0.89102566,0.94871795,0.9166667,0.9551282,0.98717946,0.9166667,0.9807692,0.9166667,0.9102564,0.96153843,0.9807692,0.974359,0.9551282,0.94871795,0.99358976,0.96794873,0.9358974,0.99358976,1.0
+TabM (tuned),0.16666667,0.14102565,0.1923077,0.22435898,0.23717949,0.28846154,0.17948718,0.30769232,0.33974358,0.30128205,0.33974358,0.46153846,0.42307693,0.4551282,0.35576922,0.43589744,0.53205127,,0.4551282,0.49358973,0.5641026,0.5769231,0.5192308,0.5448718,0.5448718,0.59615386,0.6217949,0.6923077,0.53846157,0.75641024,0.5576923,0.75641024,0.60897434,0.6025641,0.64102566,0.71153843,0.71794873,0.5833333,0.71153843,0.6282051,0.8076923,0.67948717,0.7948718,0.7948718,0.74358976,0.78846157,0.82051283,0.84615386,0.75,0.7948718,0.7692308,0.9166667,0.78846157,0.9358974,0.85897434,0.8782051,0.8525641,0.92948717,0.85897434,0.974359,0.9423077,0.92948717,0.9551282,0.974359,0.96153843,0.9551282,0.94871795,0.974359,1.0,0.98717946,0.96794873,1.0,1.0
+TabDPT (tuned),0.10897436,0.21153846,0.24358974,0.13461539,0.2820513,0.22435898,0.30128205,0.25,0.37179488,0.47435898,0.4871795,0.3974359,0.32051283,0.49358973,0.49358973,0.47435898,0.55128205,0.5448718,,0.51282054,0.5641026,0.5897436,0.59615386,0.6025641,0.61538464,0.60897434,0.5833333,0.6025641,0.63461536,0.64102566,0.49358973,0.5897436,0.6602564,0.7307692,0.64102566,0.59615386,0.6602564,0.67948717,0.6923077,0.775641,0.65384614,0.7051282,0.75641024,0.71153843,0.724359,0.6858974,0.6858974,0.78846157,0.8076923,0.8525641,0.8269231,0.76282054,0.8269231,0.86538464,0.86538464,0.9358974,0.96153843,0.96153843,0.9807692,0.82051283,0.8782051,0.9102564,0.9102564,0.8141026,0.9358974,0.94871795,0.9551282,0.9423077,1.0,0.92948717,0.96153843,0.9423077,1.0
+CatBoost (tuned),0.070512824,0.070512824,0.070512824,0.09615385,0.17307693,0.16666667,0.1474359,0.21153846,0.24358974,0.23076923,0.28846154,0.40384614,0.42307693,0.32692307,0.43589744,0.30448717,0.4871795,0.50641024,0.4871795,,0.5769231,0.5769231,0.53846157,0.5448718,0.53846157,0.50641024,0.5576923,0.5833333,0.63461536,0.65384614,0.5192308,0.6923077,0.6282051,0.5833333,0.6666667,0.6730769,0.78205127,0.7051282,0.76282054,0.69871795,0.73717946,0.75641024,0.86538464,0.6923077,0.80128205,0.78205127,0.8525641,0.7948718,0.8525641,0.8525641,0.8525641,0.9102564,0.8782051,0.9358974,0.9230769,0.9166667,0.8525641,0.974359,0.92948717,0.9551282,1.0,0.96794873,0.96153843,0.96794873,0.974359,0.974359,0.96794873,0.974359,1.0,0.99358976,0.9551282,1.0,1.0
+XGBoost (tuned),0.115384616,0.08974359,0.09615385,0.16025642,0.115384616,0.21153846,0.16666667,0.23076923,0.22435898,0.15384616,0.21794872,0.3846154,0.40384614,0.3846154,0.3525641,0.3846154,0.46153846,0.43589744,0.43589744,0.42307693,,0.52884614,0.5192308,0.4871795,0.5448718,0.50641024,0.57051283,0.5769231,0.5448718,0.5769231,0.5833333,0.6923077,0.5576923,0.5448718,0.64102566,0.6666667,0.76282054,0.64102566,0.69871795,0.59615386,0.73717946,0.69871795,0.90384614,0.6858974,0.7948718,0.74358976,0.8525641,0.75,0.89102566,0.88461536,0.89102566,0.8525641,0.7948718,0.9230769,0.92948717,0.9166667,0.8525641,0.96153843,0.92948717,0.94871795,0.99358976,0.9551282,0.94871795,0.96153843,0.96153843,0.96153843,0.96794873,0.98717946,0.99358976,0.98717946,0.9358974,1.0,1.0
+XGBoost (tuned + ensembled),0.10897436,0.08974359,0.07692308,0.13461539,0.09615385,0.20512821,0.17307693,0.21794872,0.23076923,0.12820514,0.1923077,0.3846154,0.3846154,0.3525641,0.32692307,0.37179488,0.44871795,0.42307693,0.41025642,0.42307693,0.47115386,,0.5,0.4679487,0.5192308,0.5192308,0.5641026,0.5769231,0.525641,0.5641026,0.57051283,0.71794873,0.5641026,0.53846157,0.6474359,0.6666667,0.78205127,0.63461536,0.69871795,0.60897434,0.76282054,0.6730769,0.92948717,0.6923077,0.7948718,0.75,0.8525641,0.75641024,0.88461536,0.8717949,0.8974359,0.8525641,0.7948718,0.9230769,0.92948717,0.9102564,0.8525641,0.9551282,0.9166667,0.94871795,0.99358976,0.96153843,0.9551282,0.96153843,0.96153843,0.9551282,0.96153843,0.98717946,0.99358976,0.98717946,0.9358974,1.0,1.0
+ModernNCA (tuned),0.102564104,0.12820514,0.115384616,0.10897436,0.16025642,0.16666667,0.21794872,0.17948718,0.21794872,0.2628205,0.28846154,0.32692307,0.35897437,0.3525641,0.42307693,0.44871795,0.35576922,0.48076922,0.40384614,0.46153846,0.48076922,0.5,,0.5448718,0.5448718,0.43589744,0.60897434,0.53205127,0.6666667,0.57051283,0.5641026,0.63461536,0.6730769,0.57051283,0.6217949,0.59615386,0.6602564,0.7307692,0.73717946,0.69871795,0.69871795,0.724359,0.75641024,0.64102566,0.82051283,0.7692308,0.7307692,0.75,0.78205127,0.8076923,0.78846157,0.82051283,0.86538464,0.9166667,0.9166667,0.9230769,0.9230769,0.96153843,0.96794873,0.89102566,0.974359,0.92948717,0.9358974,0.9358974,0.96794873,0.96794873,0.94871795,0.9423077,0.98717946,0.96794873,0.974359,0.99358976,1.0
+LightGBM (tuned + ensembled),0.08974359,0.10897436,0.102564104,0.08974359,0.14102565,0.1474359,0.19871795,0.16025642,0.19871795,0.17307693,0.25641027,0.33333334,0.32692307,0.2628205,0.37820512,0.40384614,0.40384614,0.4551282,0.3974359,0.4551282,0.51282054,0.53205127,0.4551282,,0.5032051,0.5192308,0.55128205,0.53846157,0.525641,0.61538464,0.5576923,0.6923077,0.57051283,0.5833333,0.61538464,0.6282051,0.74358976,0.5897436,0.7051282,0.64102566,0.76282054,0.7051282,0.8525641,0.6666667,0.7692308,0.74358976,0.82051283,0.7948718,0.94871795,0.9423077,0.98717946,0.8397436,0.7692308,0.89102566,0.9807692,0.90384614,0.85897434,0.9230769,0.94871795,0.89102566,1.0,0.94871795,0.94871795,0.92948717,0.9423077,0.9551282,0.9423077,1.0,1.0,0.99358976,0.9551282,1.0,1.0
+LightGBM (tuned),0.09615385,0.083333336,0.09615385,0.102564104,0.10897436,0.15384616,0.17307693,0.14102565,0.1923077,0.17307693,0.24358974,0.3846154,0.34615386,0.27564102,0.37179488,0.3974359,0.39102563,0.4551282,0.3846154,0.46153846,0.4551282,0.48076922,0.4551282,0.49679488,,0.5,0.53205127,0.5192308,0.50641024,0.6025641,0.5448718,0.6666667,0.525641,0.5897436,0.6025641,0.65384614,0.75641024,0.6025641,0.67948717,0.63461536,0.76282054,0.6858974,0.85897434,0.65384614,0.75,0.74358976,0.8269231,0.775641,0.9358974,0.9230769,0.94871795,0.8269231,0.78205127,0.8717949,0.96794873,0.89102566,0.86538464,0.9358974,0.94871795,0.8974359,1.0,0.9551282,0.96153843,0.92948717,0.9423077,0.9551282,0.9423077,1.0,1.0,0.99358976,0.9551282,1.0,1.0
+iLTM (tuned + ensembled),0.07692308,0.115384616,0.10897436,0.083333336,0.070512824,0.14102565,0.23076923,0.12179487,0.23717949,0.24358974,0.27564102,0.1923077,0.28846154,0.31410256,0.41025642,0.46153846,0.47435898,0.40384614,0.39102563,0.49358973,0.49358973,0.48076922,0.5641026,0.48076922,0.5,,0.5897436,0.50641024,0.64102566,0.5192308,0.53846157,0.60897434,0.63461536,0.5576923,0.6923077,0.55128205,0.6602564,0.71794873,0.69871795,0.64102566,0.6666667,0.71794873,0.78205127,0.67948717,0.8076923,0.69871795,0.76282054,0.78846157,0.8782051,0.8397436,0.88461536,0.8269231,0.78846157,0.86538464,0.9358974,0.9230769,0.85897434,0.8717949,0.99358976,0.8269231,0.99358976,0.8717949,0.86538464,0.89102566,0.8974359,0.9166667,0.9230769,0.99358976,1.0,0.974359,0.90384614,0.99358976,1.0
+xRFM (tuned + ensembled),0.07692308,0.102564104,0.115384616,0.08974359,0.19871795,0.15384616,0.17307693,0.20512821,0.15384616,0.16666667,0.22435898,0.37179488,0.34615386,0.32051283,0.35897437,0.3974359,0.33974358,0.37820512,0.41666666,0.44230768,0.42948717,0.43589744,0.39102563,0.44871795,0.4679487,0.41025642,,0.50641024,0.49358973,0.51282054,0.4679487,0.48076922,0.5641026,0.53846157,0.59615386,0.5576923,0.6217949,0.6025641,0.6923077,0.6858974,0.60897434,0.6602564,0.7051282,0.5897436,0.63461536,0.65384614,0.71153843,0.67948717,0.8141026,0.6666667,0.8333333,0.724359,0.96153843,0.8525641,0.88461536,0.9102564,0.9166667,0.974359,0.99358976,0.96153843,0.84615386,0.96794873,0.96794873,0.96794873,0.9166667,0.8974359,0.8974359,0.96794873,0.99358976,0.98717946,0.98717946,1.0,1.0
+BetaTabPFN (default),0.083333336,0.09615385,0.083333336,0.16666667,0.15384616,0.24358974,0.14102565,0.25,0.23717949,0.13461539,0.1474359,0.42307693,0.34615386,0.33333334,0.22435898,0.37820512,0.42948717,0.30769232,0.3974359,0.41666666,0.42307693,0.42307693,0.4679487,0.46153846,0.48076922,0.49358973,0.49358973,,0.47435898,0.48076922,0.41666666,0.6602564,0.53205127,0.53846157,0.5576923,0.6730769,0.63461536,0.53846157,0.5833333,0.53846157,0.76282054,0.5833333,0.69871795,0.7307692,0.6730769,0.73717946,0.775641,0.78846157,0.7051282,0.6858974,0.7307692,0.86538464,0.7307692,0.88461536,0.84615386,0.775641,0.8076923,0.86538464,0.8269231,0.9423077,0.9358974,0.9358974,0.9358974,0.9423077,0.9358974,0.96153843,0.9423077,0.99358976,1.0,0.98717946,0.974359,1.0,1.0
+ExtraTrees (tuned + ensembled),0.070512824,0.12820514,0.1474359,0.09615385,0.1474359,0.08974359,0.18589744,0.17307693,0.10897436,0.25641027,0.2628205,0.23076923,0.30769232,0.3846154,0.4551282,0.34615386,0.25641027,0.46153846,0.3653846,0.3653846,0.4551282,0.47435898,0.33333334,0.47435898,0.49358973,0.35897437,0.50641024,0.525641,,0.5641026,0.50641024,0.5,0.67628205,0.5448718,0.55128205,0.47435898,0.5769231,0.78205127,0.5833333,0.6858974,0.57051283,0.80128205,0.6602564,0.5576923,0.63461536,0.61538464,0.6602564,0.63461536,0.76282054,0.80128205,0.7692308,0.6858974,0.82051283,0.78846157,0.8782051,0.96794873,0.9551282,0.94871795,1.0,0.775641,0.9102564,0.9166667,0.9166667,0.775641,0.92948717,0.9551282,0.9423077,0.90384614,0.9807692,0.9358974,1.0,0.94871795,1.0
+TabM (default),0.14102565,0.13461539,0.1474359,0.18589744,0.19871795,0.2820513,0.17307693,0.25,0.27564102,0.12179487,0.18589744,0.41666666,0.34615386,0.31410256,0.115384616,0.33333334,0.42948717,0.24358974,0.35897437,0.34615386,0.42307693,0.43589744,0.42948717,0.3846154,0.3974359,0.48076922,0.4871795,0.5192308,0.43589744,,0.43589744,0.60897434,0.49358973,0.5192308,0.53205127,0.60897434,0.5833333,0.49358973,0.55128205,0.4679487,0.67948717,0.5448718,0.6666667,0.75,0.6217949,0.7307692,0.71794873,0.8141026,0.6923077,0.60897434,0.6923077,0.8525641,0.7051282,0.89102566,0.7692308,0.76282054,0.775641,0.8397436,0.8141026,0.9166667,0.92948717,0.92948717,0.9358974,0.9358974,0.9423077,0.9358974,0.92948717,0.9807692,1.0,0.9807692,0.96153843,1.0,1.0
+SAP-RPT-OSS (default),0.23076923,0.23717949,0.24358974,0.3525641,0.32692307,0.40384614,0.26923078,0.46153846,0.39102563,0.37179488,0.3974359,0.47435898,0.48076922,0.46153846,0.35897437,0.46153846,0.44230768,0.44230768,0.50641024,0.48076922,0.41666666,0.42948717,0.43589744,0.44230768,0.4551282,0.46153846,0.53205127,0.5833333,0.49358973,0.5641026,,0.5448718,0.5,0.525641,0.5,0.6474359,0.51282054,0.49358973,0.59615386,0.525641,0.57051283,0.5192308,0.50641024,0.6730769,0.5448718,0.6602564,0.59615386,0.73717946,0.5,0.61538464,0.48076922,0.75641024,0.60897434,0.7307692,0.6025641,0.65384614,0.5897436,0.6730769,0.65384614,0.8141026,0.67948717,0.75,0.73717946,0.8141026,0.7692308,0.76282054,0.76282054,0.80128205,0.76282054,0.94871795,0.775641,0.82051283,1.0
+EBM (tuned + ensembled),0.08974359,0.044871796,0.06410257,0.07692308,0.12820514,0.16025642,0.1474359,0.17307693,0.16666667,0.13461539,0.15384616,0.3525641,0.30769232,0.21153846,0.18589744,0.2948718,0.34615386,0.24358974,0.41025642,0.30769232,0.30769232,0.2820513,0.3653846,0.30769232,0.33333334,0.39102563,0.5192308,0.33974358,0.5,0.39102563,0.4551282,,0.51282054,0.5576923,0.53846157,0.64102566,0.50641024,0.53846157,0.6217949,0.5448718,0.7532051,0.5576923,0.6217949,0.5833333,0.64102566,0.7051282,0.8525641,0.6923077,0.71153843,0.61538464,0.724359,0.75,0.75,0.9230769,0.8397436,0.80128205,0.76282054,0.8782051,0.8141026,0.9423077,0.94871795,0.9358974,0.9423077,0.974359,0.9230769,0.9358974,0.9358974,0.9807692,0.99358976,0.98717946,0.9423077,0.99358976,1.0
+RandomForest (tuned + ensembled),0.08974359,0.115384616,0.12179487,0.08974359,0.16666667,0.115384616,0.22435898,0.12820514,0.08974359,0.18589744,0.21153846,0.24358974,0.25,0.32692307,0.3974359,0.33974358,0.24358974,0.39102563,0.33974358,0.37179488,0.44230768,0.43589744,0.32692307,0.42948717,0.47435898,0.3653846,0.43589744,0.4679487,0.32371795,0.50641024,0.5,0.4871795,,0.51282054,0.46153846,0.4551282,0.5448718,0.5769231,0.55128205,0.6217949,0.55128205,0.80448717,0.65384614,0.50641024,0.5576923,0.5576923,0.6666667,0.5448718,0.78205127,0.74358976,0.775641,0.61538464,0.76282054,0.74358976,0.85897434,0.98717946,0.9102564,0.90384614,0.9551282,0.724359,0.8525641,0.92948717,0.92948717,0.7307692,0.9166667,0.9358974,0.9358974,0.89102566,0.94871795,0.9423077,0.99358976,0.9230769,1.0
+TabDPT (default),0.08974359,0.1923077,0.21153846,0.15384616,0.21153846,0.17307693,0.2820513,0.23717949,0.37179488,0.42307693,0.41666666,0.19871795,0.23717949,0.37820512,0.3846154,0.40384614,0.42307693,0.3974359,0.26923078,0.41666666,0.4551282,0.46153846,0.42948717,0.41666666,0.41025642,0.44230768,0.46153846,0.46153846,0.4551282,0.48076922,0.47435898,0.44230768,0.4871795,,0.49358973,0.42948717,0.42948717,0.53846157,0.525641,0.5641026,0.4679487,0.5897436,0.525641,0.5833333,0.55128205,0.49358973,0.47435898,0.63461536,0.6217949,0.75,0.59615386,0.59615386,0.71153843,0.65384614,0.775641,0.80128205,0.85897434,0.82051283,0.9230769,0.6025641,0.86538464,0.775641,0.76282054,0.6666667,0.85897434,0.8782051,0.89102566,0.9102564,1.0,0.8974359,0.88461536,0.9358974,1.0
+iLTM (tuned),0.06410257,0.044871796,0.044871796,0.057692308,0.051282052,0.083333336,0.09615385,0.115384616,0.16025642,0.1474359,0.16666667,0.18589744,0.25641027,0.24358974,0.30128205,0.30769232,0.30769232,0.35897437,0.35897437,0.33333334,0.35897437,0.3525641,0.37820512,0.3846154,0.3974359,0.30769232,0.40384614,0.44230768,0.44871795,0.4679487,0.5,0.46153846,0.53846157,0.50641024,,0.44871795,0.51282054,0.61538464,0.5576923,0.525641,0.5576923,0.6730769,0.63461536,0.6474359,0.6666667,0.6282051,0.63461536,0.6923077,0.7307692,0.724359,0.6923077,0.74358976,0.74358976,0.85897434,0.8397436,0.8717949,0.775641,0.8717949,0.92948717,0.78205127,0.9807692,0.8525641,0.8525641,0.8397436,0.8974359,0.9166667,0.9102564,0.9423077,0.974359,0.94871795,0.89102566,0.974359,1.0
+TorchMLP (tuned + ensembled),0.070512824,0.051282052,0.057692308,0.08974359,0.07692308,0.115384616,0.09615385,0.16666667,0.2820513,0.24358974,0.19871795,0.2820513,0.32692307,0.24358974,0.2628205,0.33974358,0.3846154,0.28846154,0.40384614,0.32692307,0.33333334,0.33333334,0.40384614,0.37179488,0.34615386,0.44871795,0.44230768,0.32692307,0.525641,0.39102563,0.3525641,0.35897437,0.5448718,0.57051283,0.55128205,,0.4871795,0.5833333,0.5448718,0.49358973,0.4551282,0.5576923,0.5833333,0.71153843,0.65384614,0.7916667,0.53846157,0.8076923,0.61538464,0.6602564,0.61538464,0.71153843,0.6730769,0.8974359,0.6602564,0.75641024,0.74358976,0.775641,0.80128205,0.8782051,0.86538464,0.90384614,0.9102564,0.9230769,0.84615386,0.88461536,0.8782051,0.96153843,0.99358976,0.974359,0.9423077,0.99358976,1.0
+CatBoost (default),0.06410257,0.032051284,0.057692308,0.051282052,0.044871796,0.10897436,0.12179487,0.10897436,0.115384616,0.12179487,0.16025642,0.24358974,0.25641027,0.17307693,0.21794872,0.16666667,0.30128205,0.2820513,0.33974358,0.21794872,0.23717949,0.21794872,0.33974358,0.25641027,0.24358974,0.33974358,0.37820512,0.3653846,0.42307693,0.41666666,0.4871795,0.49358973,0.4551282,0.57051283,0.4871795,0.51282054,,0.5192308,0.5833333,0.5192308,0.6282051,0.5769231,0.63461536,0.5897436,0.6282051,0.65384614,0.6923077,0.6858974,0.78846157,0.74358976,0.7948718,0.724359,0.71794873,0.86538464,0.9423077,0.8525641,0.8076923,0.8974359,0.89102566,0.84615386,1.0,0.9230769,0.9166667,0.8717949,0.9230769,0.94871795,0.9230769,0.96794873,1.0,0.99358976,0.94871795,1.0,1.0
+ExtraTrees (tuned),0.07692308,0.102564104,0.12179487,0.09615385,0.15384616,0.083333336,0.1474359,0.15384616,0.08974359,0.1923077,0.22435898,0.21794872,0.21153846,0.30128205,0.3974359,0.2628205,0.18589744,0.41666666,0.32051283,0.2948718,0.35897437,0.3653846,0.26923078,0.41025642,0.3974359,0.2820513,0.3974359,0.46153846,0.21794872,0.50641024,0.50641024,0.46153846,0.42307693,0.46153846,0.3846154,0.41666666,0.48076922,,0.49358973,0.5897436,0.53205127,0.599359,0.5192308,0.5192308,0.50641024,0.51282054,0.57051283,0.5641026,0.64102566,0.75,0.6474359,0.5897436,0.7307692,0.71794873,0.8076923,0.8974359,0.9358974,0.9166667,0.89102566,0.724359,0.80128205,0.9166667,0.9102564,0.7051282,0.9166667,0.9166667,0.9230769,0.89102566,0.96794873,0.9358974,1.0,0.92948717,1.0
+xRFM (tuned),0.070512824,0.07692308,0.07692308,0.057692308,0.14102565,0.08974359,0.12179487,0.17307693,0.09615385,0.14102565,0.16025642,0.30769232,0.24358974,0.25641027,0.2820513,0.22435898,0.1923077,0.28846154,0.30769232,0.23717949,0.30128205,0.30128205,0.2628205,0.2948718,0.32051283,0.30128205,0.30769232,0.41666666,0.41666666,0.44871795,0.40384614,0.37820512,0.44871795,0.47435898,0.44230768,0.4551282,0.41666666,0.50641024,,0.5897436,0.48076922,0.55128205,0.53205127,0.525641,0.5,0.5576923,0.5833333,0.5641026,0.69871795,0.59615386,0.6730769,0.5833333,0.9230769,0.78205127,0.8333333,0.86538464,0.8782051,0.9423077,0.94871795,0.8782051,0.8076923,0.92948717,0.9358974,0.88461536,0.8974359,0.9230769,0.9102564,0.92948717,0.99358976,0.9807692,0.974359,0.99358976,1.0
+TabSwift (default),0.07692308,0.025641026,0.025641026,0.012820513,0.044871796,0.051282052,0.06410257,0.07692308,0.17948718,0.25641027,0.25641027,0.16666667,0.16666667,0.27564102,0.3525641,0.25641027,0.32692307,0.37179488,0.22435898,0.30128205,0.40384614,0.39102563,0.30128205,0.35897437,0.3653846,0.35897437,0.31410256,0.46153846,0.31410256,0.53205127,0.47435898,0.4551282,0.37820512,0.43589744,0.47435898,0.50641024,0.48076922,0.41025642,0.41025642,,0.51282054,0.53205127,0.51282054,0.6602564,0.55128205,0.61538464,0.5448718,0.67948717,0.5769231,0.76282054,0.57051283,0.67948717,0.6217949,0.69871795,0.76282054,0.86538464,0.8525641,0.92948717,0.775641,0.67948717,0.8076923,0.8782051,0.86538464,0.71153843,0.9358974,0.94871795,0.94871795,0.8782051,0.9551282,0.9358974,0.9551282,0.96794873,1.0
+EBM (tuned),0.070512824,0.03846154,0.057692308,0.051282052,0.08974359,0.14102565,0.14102565,0.14102565,0.115384616,0.12820514,0.115384616,0.2820513,0.27564102,0.18589744,0.16025642,0.21153846,0.2820513,0.1923077,0.34615386,0.2628205,0.2628205,0.23717949,0.30128205,0.23717949,0.23717949,0.33333334,0.39102563,0.23717949,0.42948717,0.32051283,0.42948717,0.24679486,0.44871795,0.53205127,0.44230768,0.5448718,0.37179488,0.4679487,0.5192308,0.4871795,,0.5192308,0.53846157,0.53846157,0.5448718,0.65384614,0.71794873,0.6474359,0.65384614,0.525641,0.67948717,0.6666667,0.7051282,0.88461536,0.7948718,0.7692308,0.71153843,0.8269231,0.78846157,0.8974359,0.9102564,0.9230769,0.92948717,0.92948717,0.90384614,0.9166667,0.9102564,0.96794873,0.99358976,0.9807692,0.9423077,0.99358976,1.0
+RandomForest (tuned),0.08974359,0.08974359,0.08974359,0.08974359,0.12820514,0.09615385,0.14102565,0.12179487,0.09615385,0.14102565,0.1923077,0.22435898,0.21153846,0.23076923,0.28846154,0.21794872,0.23076923,0.32051283,0.2948718,0.24358974,0.30128205,0.32692307,0.27564102,0.2948718,0.31410256,0.2820513,0.33974358,0.41666666,0.19871795,0.4551282,0.48076922,0.44230768,0.19551282,0.41025642,0.32692307,0.44230768,0.42307693,0.40064102,0.44871795,0.4679487,0.48076922,,0.5,0.49358973,0.40384614,0.49358973,0.60897434,0.53846157,0.6217949,0.6666667,0.6282051,0.5448718,0.64102566,0.67948717,0.8076923,0.8301282,0.8525641,0.86538464,0.80128205,0.6923077,0.8076923,0.88461536,0.8782051,0.6858974,0.8974359,0.9230769,0.9166667,0.8782051,0.92948717,0.9230769,0.974359,0.90384614,1.0
+XGBoost (default),0.06410257,0.01923077,0.012820513,0.0064102565,0.01923077,0.01923077,0.08974359,0.070512824,0.051282052,0.06410257,0.09615385,0.1474359,0.15384616,0.15384616,0.16666667,0.07692308,0.17948718,0.20512821,0.24358974,0.13461539,0.09615385,0.070512824,0.24358974,0.1474359,0.14102565,0.21794872,0.2948718,0.30128205,0.33974358,0.33333334,0.49358973,0.37820512,0.34615386,0.47435898,0.3653846,0.41666666,0.3653846,0.48076922,0.4679487,0.4871795,0.46153846,0.5,,0.47435898,0.525641,0.53205127,0.63461536,0.55128205,0.6666667,0.69871795,0.69871795,0.6025641,0.6923077,0.7692308,0.8717949,0.8717949,0.8141026,0.89102566,0.8717949,0.73717946,0.96153843,0.88461536,0.8717949,0.775641,0.8974359,0.92948717,0.9230769,0.94871795,0.98717946,0.98717946,0.92948717,1.0,1.0
+FastaiMLP (tuned + ensembled),0.09615385,0.070512824,0.06410257,0.12179487,0.051282052,0.16025642,0.07692308,0.17307693,0.21794872,0.08974359,0.12179487,0.21153846,0.21794872,0.16666667,0.17307693,0.2948718,0.31410256,0.20512821,0.28846154,0.30769232,0.31410256,0.30769232,0.35897437,0.33333334,0.34615386,0.32051283,0.41025642,0.26923078,0.44230768,0.25,0.32692307,0.41666666,0.49358973,0.41666666,0.3525641,0.28846154,0.41025642,0.48076922,0.47435898,0.33974358,0.46153846,0.50641024,0.525641,,0.5641026,0.47435898,0.48076922,0.71794873,0.6474359,0.55128205,0.63461536,0.5576923,0.5833333,0.63461536,0.6858974,0.60897434,0.65384614,0.59615386,0.7948718,0.82051283,0.74358976,0.724359,0.724359,0.86538464,0.6858974,0.73717946,0.73717946,0.99358976,0.9230769,0.9358974,0.8974359,1.0,0.99358976
+ModernNCA (default),0.01923077,0.051282052,0.06410257,0.032051284,0.06410257,0.032051284,0.08974359,0.044871796,0.057692308,0.10897436,0.13461539,0.16025642,0.14102565,0.17307693,0.21794872,0.16666667,0.08974359,0.25641027,0.27564102,0.19871795,0.20512821,0.20512821,0.17948718,0.23076923,0.25,0.1923077,0.3653846,0.32692307,0.3653846,0.37820512,0.4551282,0.35897437,0.44230768,0.44871795,0.33333334,0.34615386,0.37179488,0.49358973,0.5,0.44871795,0.4551282,0.59615386,0.47435898,0.43589744,,0.4551282,0.53846157,0.51282054,0.69871795,0.6730769,0.71794873,0.53205127,0.6858974,0.74358976,0.80128205,0.85897434,0.8269231,0.8269231,0.84615386,0.78205127,0.78846157,0.7948718,0.7948718,0.82051283,0.88461536,0.9102564,0.90384614,0.86538464,0.92948717,0.8717949,0.90384614,0.9807692,1.0
+TorchMLP (tuned),0.06410257,0.051282052,0.051282052,0.044871796,0.07692308,0.102564104,0.09615385,0.115384616,0.17307693,0.1923077,0.15384616,0.21794872,0.19871795,0.16025642,0.17948718,0.19871795,0.22435898,0.21153846,0.31410256,0.21794872,0.25641027,0.25,0.23076923,0.25641027,0.25641027,0.30128205,0.34615386,0.2628205,0.3846154,0.26923078,0.33974358,0.2948718,0.44230768,0.50641024,0.37179488,0.20833333,0.34615386,0.4871795,0.44230768,0.3846154,0.34615386,0.50641024,0.4679487,0.525641,0.5448718,,0.35897437,0.6602564,0.525641,0.53846157,0.525641,0.53846157,0.6217949,0.82051283,0.64102566,0.724359,0.7051282,0.69871795,0.75,0.75,0.8141026,0.8717949,0.86538464,0.80128205,0.7948718,0.8397436,0.8269231,0.92948717,0.96794873,0.9551282,0.9423077,0.98717946,1.0
+EBM (default),0.070512824,0.051282052,0.06410257,0.057692308,0.070512824,0.07692308,0.102564104,0.08974359,0.09615385,0.12179487,0.13461539,0.21794872,0.25,0.12820514,0.16025642,0.12179487,0.21794872,0.17948718,0.31410256,0.1474359,0.1474359,0.1474359,0.26923078,0.17948718,0.17307693,0.23717949,0.28846154,0.22435898,0.33974358,0.2820513,0.40384614,0.1474359,0.33333334,0.525641,0.3653846,0.46153846,0.30769232,0.42948717,0.41666666,0.4551282,0.2820513,0.39102563,0.3653846,0.5192308,0.46153846,0.64102566,,0.59615386,0.6025641,0.50641024,0.61538464,0.55128205,0.64102566,0.78205127,0.76282054,0.73717946,0.7692308,0.80128205,0.7692308,0.7948718,0.86538464,0.9102564,0.9166667,0.86538464,0.88461536,0.90384614,0.8782051,0.96794873,0.9807692,0.974359,0.974359,1.0,1.0
+FastaiMLP (tuned),0.057692308,0.032051284,0.044871796,0.057692308,0.01923077,0.057692308,0.057692308,0.09615385,0.10897436,0.083333336,0.06410257,0.12820514,0.17307693,0.09615385,0.115384616,0.16666667,0.18589744,0.15384616,0.21153846,0.20512821,0.25,0.24358974,0.25,0.20512821,0.22435898,0.21153846,0.32051283,0.21153846,0.3653846,0.18589744,0.2628205,0.30769232,0.4551282,0.3653846,0.30769232,0.1923077,0.31410256,0.43589744,0.43589744,0.32051283,0.3525641,0.46153846,0.44871795,0.2820513,0.4871795,0.33974358,0.40384614,,0.5897436,0.5,0.5641026,0.4551282,0.5897436,0.60897434,0.6602564,0.59615386,0.67948717,0.5833333,0.7692308,0.76282054,0.74358976,0.6923077,0.7051282,0.8076923,0.6666667,0.71794873,0.724359,0.98717946,0.8397436,0.9358974,0.8974359,1.0,0.99358976
+ChimeraBoost (tuned),0.070512824,0.044871796,0.057692308,0.025641026,0.06410257,0.08974359,0.12179487,0.07692308,0.083333336,0.115384616,0.13461539,0.10897436,0.16025642,0.115384616,0.1923077,0.12179487,0.16025642,0.25,0.1923077,0.1474359,0.10897436,0.115384616,0.21794872,0.051282052,0.06410257,0.12179487,0.18589744,0.2948718,0.23717949,0.30769232,0.5,0.28846154,0.21794872,0.37820512,0.26923078,0.3846154,0.21153846,0.35897437,0.30128205,0.42307693,0.34615386,0.37820512,0.33333334,0.3525641,0.30128205,0.47435898,0.3974359,0.41025642,,0.55128205,0.47115386,0.4871795,0.6602564,0.6217949,0.92948717,0.64102566,0.65384614,0.75641024,0.7051282,0.69871795,0.8333333,0.7948718,0.80128205,0.71794873,0.7307692,0.78846157,0.775641,0.94871795,0.9358974,0.9166667,0.86538464,0.974359,1.0
+LightGBM (default),0.070512824,0.051282052,0.06410257,0.025641026,0.06410257,0.070512824,0.083333336,0.070512824,0.057692308,0.083333336,0.09615385,0.14102565,0.13461539,0.13461539,0.10897436,0.10897436,0.18589744,0.20512821,0.1474359,0.1474359,0.115384616,0.12820514,0.1923077,0.057692308,0.07692308,0.16025642,0.33333334,0.31410256,0.19871795,0.39102563,0.3846154,0.3846154,0.25641027,0.25,0.27564102,0.33974358,0.25641027,0.25,0.40384614,0.23717949,0.47435898,0.33333334,0.30128205,0.44871795,0.32692307,0.46153846,0.49358973,0.5,0.44871795,,0.46153846,0.5192308,0.53846157,0.6666667,0.71794873,0.6282051,0.775641,0.75641024,0.6474359,0.6217949,0.775641,0.8141026,0.78846157,0.6666667,0.8782051,0.9166667,0.88461536,0.89102566,0.89102566,0.9423077,0.92948717,0.9807692,1.0
+ChimeraBoost (tuned + ensembled),0.070512824,0.032051284,0.044871796,0.012820513,0.051282052,0.070512824,0.115384616,0.070512824,0.057692308,0.10897436,0.115384616,0.10897436,0.13461539,0.10897436,0.16666667,0.083333336,0.16666667,0.23076923,0.17307693,0.1474359,0.10897436,0.102564104,0.21153846,0.012820513,0.051282052,0.115384616,0.16666667,0.26923078,0.23076923,0.30769232,0.5192308,0.27564102,0.22435898,0.40384614,0.30769232,0.3846154,0.20512821,0.3525641,0.32692307,0.42948717,0.32051283,0.37179488,0.30128205,0.3653846,0.2820513,0.47435898,0.3846154,0.43589744,0.52884614,0.53846157,,0.4679487,0.67948717,0.61538464,0.92948717,0.64102566,0.65384614,0.76282054,0.71794873,0.6858974,0.8397436,0.80128205,0.8076923,0.724359,0.7307692,0.8076923,0.78846157,0.96794873,0.96794873,0.9423077,0.86538464,0.9807692,1.0
+RealMLP (default),0.01923077,0.0064102565,0.012820513,0.01923077,0.032051284,0.012820513,0.0064102565,0.070512824,0.01923077,0.0064102565,0.01923077,0.1474359,0.13461539,0.12179487,0.044871796,0.051282052,0.10897436,0.083333336,0.23717949,0.08974359,0.1474359,0.1474359,0.17948718,0.16025642,0.17307693,0.17307693,0.27564102,0.13461539,0.31410256,0.1474359,0.24358974,0.25,0.3846154,0.40384614,0.25641027,0.28846154,0.27564102,0.41025642,0.41666666,0.32051283,0.33333334,0.4551282,0.3974359,0.44230768,0.4679487,0.46153846,0.44871795,0.5448718,0.51282054,0.48076922,0.53205127,,0.5833333,0.69871795,0.61538464,0.6858974,0.7051282,0.63461536,0.76282054,0.7948718,0.78205127,0.78846157,0.78846157,0.8525641,0.8397436,0.8974359,0.8525641,0.9358974,0.92948717,0.9423077,0.92948717,0.99358976,1.0
+xRFM (default),0.070512824,0.01923077,0.032051284,0.012820513,0.06410257,0.01923077,0.06410257,0.102564104,0.025641026,0.06410257,0.08974359,0.21153846,0.14102565,0.19871795,0.18589744,0.09615385,0.102564104,0.21153846,0.17307693,0.12179487,0.20512821,0.20512821,0.13461539,0.23076923,0.21794872,0.21153846,0.03846154,0.26923078,0.17948718,0.2948718,0.39102563,0.25,0.23717949,0.28846154,0.25641027,0.32692307,0.2820513,0.26923078,0.07692308,0.37820512,0.2948718,0.35897437,0.30769232,0.41666666,0.31410256,0.37820512,0.35897437,0.41025642,0.33974358,0.46153846,0.32051283,0.41666666,,0.4679487,0.525641,0.6730769,0.63461536,0.6217949,0.5,0.6282051,0.55128205,0.76282054,0.75641024,0.6474359,0.7051282,0.74358976,0.76282054,0.6858974,0.74358976,0.9230769,0.9230769,0.8974359,1.0
+TorchMLP (default),0.06410257,0.0064102565,0.012820513,0.012820513,0.051282052,0.032051284,0.03846154,0.102564104,0.03846154,0.025641026,0.032051284,0.1474359,0.102564104,0.083333336,0.051282052,0.03846154,0.070512824,0.06410257,0.13461539,0.06410257,0.07692308,0.07692308,0.083333336,0.10897436,0.12820514,0.13461539,0.1474359,0.115384616,0.21153846,0.10897436,0.26923078,0.07692308,0.25641027,0.34615386,0.14102565,0.102564104,0.13461539,0.2820513,0.21794872,0.30128205,0.115384616,0.32051283,0.23076923,0.3653846,0.25641027,0.17948718,0.21794872,0.39102563,0.37820512,0.33333334,0.3846154,0.30128205,0.53205127,,0.525641,0.5769231,0.64102566,0.61538464,0.6602564,0.6858974,0.49358973,0.7692308,0.775641,0.7307692,0.6474359,0.71153843,0.6858974,0.85897434,0.8141026,0.8974359,0.92948717,0.9423077,1.0
+ChimeraBoost (default),0.057692308,0.03846154,0.057692308,0.012820513,0.051282052,0.06410257,0.06410257,0.06410257,0.051282052,0.115384616,0.10897436,0.10897436,0.12179487,0.070512824,0.102564104,0.07692308,0.10897436,0.14102565,0.13461539,0.07692308,0.070512824,0.070512824,0.083333336,0.01923077,0.032051284,0.06410257,0.115384616,0.15384616,0.12179487,0.23076923,0.3974359,0.16025642,0.14102565,0.22435898,0.16025642,0.33974358,0.057692308,0.1923077,0.16666667,0.23717949,0.20512821,0.1923077,0.12820514,0.31410256,0.19871795,0.35897437,0.23717949,0.33974358,0.070512824,0.2820513,0.070512824,0.3846154,0.47435898,0.47435898,,0.5,0.46153846,0.5576923,0.31410256,0.60897434,0.64102566,0.71794873,0.73717946,0.61538464,0.6474359,0.69871795,0.67948717,0.75,0.6666667,0.82051283,0.8525641,0.9102564,1.0
+RandomForest (default),0.0064102565,0.012820513,0.012820513,0.0064102565,0.0,0.0,0.03846154,0.0064102565,0.0,0.06410257,0.070512824,0.0,0.01923077,0.09615385,0.10897436,0.07692308,0.051282052,0.12179487,0.06410257,0.083333336,0.083333336,0.08974359,0.07692308,0.09615385,0.10897436,0.07692308,0.08974359,0.22435898,0.032051284,0.23717949,0.34615386,0.19871795,0.012820513,0.19871795,0.12820514,0.24358974,0.1474359,0.102564104,0.13461539,0.13461539,0.23076923,0.16987179,0.12820514,0.39102563,0.14102565,0.27564102,0.2628205,0.40384614,0.35897437,0.37179488,0.35897437,0.31410256,0.32692307,0.42307693,0.5,,0.60897434,0.4679487,0.525641,0.525641,0.42948717,0.57051283,0.57051283,0.5833333,0.71153843,0.74358976,0.71794873,0.67948717,0.6602564,0.8333333,0.89102566,0.8717949,1.0
+KNN (tuned + ensembled),0.06410257,0.06410257,0.070512824,0.01923077,0.044871796,0.051282052,0.115384616,0.06410257,0.051282052,0.07692308,0.070512824,0.07692308,0.032051284,0.16666667,0.10897436,0.12820514,0.083333336,0.1474359,0.03846154,0.1474359,0.1474359,0.1474359,0.07692308,0.14102565,0.13461539,0.14102565,0.083333336,0.1923077,0.044871796,0.22435898,0.41025642,0.23717949,0.08974359,0.14102565,0.22435898,0.25641027,0.1923077,0.06410257,0.12179487,0.1474359,0.28846154,0.1474359,0.18589744,0.34615386,0.17307693,0.2948718,0.23076923,0.32051283,0.34615386,0.22435898,0.34615386,0.2948718,0.3653846,0.35897437,0.53846157,0.39102563,,0.42948717,0.41666666,0.46153846,0.4551282,0.4551282,0.4679487,0.4679487,0.6474359,0.6858974,0.69871795,0.59615386,0.5641026,0.64102566,0.9358974,0.7051282,0.99358976
+OrionMSP (default),0.06410257,0.01923077,0.012820513,0.0064102565,0.032051284,0.0064102565,0.012820513,0.083333336,0.025641026,0.012820513,0.01923077,0.057692308,0.03846154,0.08974359,0.057692308,0.025641026,0.044871796,0.070512824,0.03846154,0.025641026,0.03846154,0.044871796,0.03846154,0.07692308,0.06410257,0.12820514,0.025641026,0.13461539,0.051282052,0.16025642,0.32692307,0.12179487,0.09615385,0.17948718,0.12820514,0.22435898,0.102564104,0.083333336,0.057692308,0.070512824,0.17307693,0.13461539,0.10897436,0.40384614,0.17307693,0.30128205,0.19871795,0.41666666,0.24358974,0.24358974,0.23717949,0.3653846,0.37820512,0.3846154,0.44230768,0.53205127,0.57051283,,0.47435898,0.57051283,0.4871795,0.6730769,0.65384614,0.5897436,0.67948717,0.7692308,0.75,0.75,0.76282054,0.8974359,0.9166667,0.8974359,1.0
+ExtraTrees (default),0.0,0.0064102565,0.0064102565,0.0064102565,0.0,0.0,0.0064102565,0.025641026,0.0,0.0,0.01923077,0.0,0.0,0.070512824,0.083333336,0.051282052,0.012820513,0.14102565,0.01923077,0.070512824,0.070512824,0.083333336,0.032051284,0.051282052,0.051282052,0.0064102565,0.0064102565,0.17307693,0.0,0.18589744,0.34615386,0.18589744,0.044871796,0.07692308,0.070512824,0.19871795,0.10897436,0.10897436,0.051282052,0.22435898,0.21153846,0.19871795,0.12820514,0.20512821,0.15384616,0.25,0.23076923,0.23076923,0.2948718,0.3525641,0.2820513,0.23717949,0.5,0.33974358,0.6858974,0.47435898,0.5833333,0.525641,,0.5576923,0.5576923,0.5769231,0.5833333,0.5576923,0.61538464,0.71794873,0.71153843,0.78205127,0.71794873,0.7307692,0.84615386,0.85897434,1.0
+Linear (tuned + ensembled),0.0064102565,0.0064102565,0.0064102565,0.025641026,0.01923077,0.012820513,0.0,0.012820513,0.012820513,0.0064102565,0.012820513,0.12820514,0.13461539,0.051282052,0.03846154,0.03846154,0.083333336,0.025641026,0.17948718,0.044871796,0.051282052,0.051282052,0.10897436,0.10897436,0.102564104,0.17307693,0.03846154,0.057692308,0.22435898,0.083333336,0.18589744,0.057692308,0.27564102,0.3974359,0.21794872,0.12179487,0.15384616,0.27564102,0.12179487,0.32051283,0.102564104,0.30769232,0.2628205,0.17948718,0.21794872,0.25,0.20512821,0.23717949,0.30128205,0.37820512,0.31410256,0.20512821,0.37179488,0.31410256,0.39102563,0.47435898,0.53846157,0.42948717,0.44230768,,0.42948717,0.42948717,0.4551282,0.75961536,0.4551282,0.47435898,0.5448718,0.44871795,0.5576923,0.6282051,0.7692308,0.94871795,0.9358974
+iLTM (default),0.057692308,0.0,0.0,0.0,0.0,0.0064102565,0.012820513,0.057692308,0.025641026,0.057692308,0.06410257,0.057692308,0.115384616,0.01923077,0.057692308,0.0,0.01923077,0.057692308,0.12179487,0.0,0.0064102565,0.0064102565,0.025641026,0.0,0.0,0.0064102565,0.15384616,0.06410257,0.08974359,0.070512824,0.32051283,0.051282052,0.1474359,0.13461539,0.01923077,0.13461539,0.0,0.19871795,0.1923077,0.1923077,0.08974359,0.1923077,0.03846154,0.25641027,0.21153846,0.18589744,0.13461539,0.25641027,0.16666667,0.22435898,0.16025642,0.21794872,0.44871795,0.50641024,0.35897437,0.57051283,0.5448718,0.51282054,0.44230768,0.57051283,,0.7051282,0.69871795,0.5769231,0.6730769,0.69871795,0.69871795,0.67948717,0.71153843,0.78846157,0.7692308,0.78205127,0.9358974
+TabSTAR (tuned),0.057692308,0.012820513,0.0064102565,0.0064102565,0.0064102565,0.0,0.012820513,0.06410257,0.01923077,0.025641026,0.025641026,0.115384616,0.08974359,0.057692308,0.057692308,0.01923077,0.083333336,0.070512824,0.08974359,0.032051284,0.044871796,0.03846154,0.070512824,0.051282052,0.044871796,0.12820514,0.032051284,0.06410257,0.083333336,0.070512824,0.25,0.06410257,0.070512824,0.22435898,0.1474359,0.09615385,0.07692308,0.083333336,0.070512824,0.12179487,0.07692308,0.115384616,0.115384616,0.27564102,0.20512821,0.12820514,0.08974359,0.30769232,0.20512821,0.18589744,0.19871795,0.21153846,0.23717949,0.23076923,0.2820513,0.42948717,0.5448718,0.32692307,0.42307693,0.57051283,0.2948718,,0.5416667,0.6217949,0.53846157,0.5769231,0.5576923,0.60897434,0.46153846,1.0,0.85897434,0.9102564,0.98717946
+TabSTAR (tuned + ensembled),0.06410257,0.012820513,0.0064102565,0.0064102565,0.0064102565,0.01923077,0.01923077,0.06410257,0.032051284,0.032051284,0.03846154,0.115384616,0.08974359,0.057692308,0.044871796,0.025641026,0.08974359,0.044871796,0.08974359,0.03846154,0.051282052,0.044871796,0.06410257,0.051282052,0.03846154,0.13461539,0.032051284,0.06410257,0.083333336,0.06410257,0.2628205,0.057692308,0.070512824,0.23717949,0.1474359,0.08974359,0.083333336,0.08974359,0.06410257,0.13461539,0.070512824,0.12179487,0.12820514,0.27564102,0.20512821,0.13461539,0.083333336,0.2948718,0.19871795,0.21153846,0.1923077,0.21153846,0.24358974,0.22435898,0.2628205,0.42948717,0.53205127,0.34615386,0.41666666,0.5448718,0.30128205,0.45833334,,0.60897434,0.525641,0.5641026,0.5576923,0.59615386,0.4551282,1.0,0.85897434,0.89102566,0.98717946
+Linear (tuned),0.0064102565,0.0,0.0064102565,0.01923077,0.012820513,0.012820513,0.0064102565,0.01923077,0.0064102565,0.012820513,0.0064102565,0.14102565,0.12179487,0.03846154,0.01923077,0.032051284,0.03846154,0.025641026,0.18589744,0.032051284,0.03846154,0.03846154,0.06410257,0.070512824,0.070512824,0.10897436,0.032051284,0.057692308,0.22435898,0.06410257,0.18589744,0.025641026,0.26923078,0.33333334,0.16025642,0.07692308,0.12820514,0.2948718,0.115384616,0.28846154,0.070512824,0.31410256,0.22435898,0.13461539,0.17948718,0.19871795,0.13461539,0.1923077,0.2820513,0.33333334,0.27564102,0.1474359,0.3525641,0.26923078,0.3846154,0.41666666,0.53205127,0.41025642,0.44230768,0.24038461,0.42307693,0.37820512,0.39102563,,0.42307693,0.4551282,0.5192308,0.41025642,0.55128205,0.61538464,0.75641024,0.85897434,0.9358974
+PerpetualBooster (tuned + ensembled),0.0064102565,0.012820513,0.012820513,0.0064102565,0.032051284,0.025641026,0.012820513,0.025641026,0.01923077,0.01923077,0.012820513,0.057692308,0.03846154,0.057692308,0.044871796,0.025641026,0.01923077,0.03846154,0.06410257,0.025641026,0.03846154,0.03846154,0.032051284,0.057692308,0.057692308,0.102564104,0.083333336,0.06410257,0.070512824,0.057692308,0.23076923,0.07692308,0.083333336,0.14102565,0.102564104,0.15384616,0.07692308,0.083333336,0.102564104,0.06410257,0.09615385,0.102564104,0.102564104,0.31410256,0.115384616,0.20512821,0.115384616,0.33333334,0.26923078,0.12179487,0.26923078,0.16025642,0.2948718,0.3525641,0.3525641,0.28846154,0.3525641,0.32051283,0.3846154,0.5448718,0.32692307,0.46153846,0.47435898,0.5769231,,0.78846157,0.8141026,0.6025641,0.65384614,0.53205127,0.6474359,0.67948717,0.9423077
+PerpetualBooster (tuned),0.025641026,0.025641026,0.025641026,0.025641026,0.044871796,0.025641026,0.025641026,0.032051284,0.025641026,0.025641026,0.025641026,0.057692308,0.03846154,0.051282052,0.032051284,0.025641026,0.025641026,0.044871796,0.051282052,0.025641026,0.03846154,0.044871796,0.032051284,0.044871796,0.044871796,0.083333336,0.102564104,0.03846154,0.044871796,0.06410257,0.23717949,0.06410257,0.06410257,0.12179487,0.083333336,0.115384616,0.051282052,0.083333336,0.07692308,0.051282052,0.083333336,0.07692308,0.070512824,0.2628205,0.08974359,0.16025642,0.09615385,0.2820513,0.21153846,0.083333336,0.1923077,0.102564104,0.25641027,0.28846154,0.30128205,0.25641027,0.31410256,0.23076923,0.2820513,0.525641,0.30128205,0.42307693,0.43589744,0.5448718,0.21153846,,0.6698718,0.57051283,0.53846157,0.47435898,0.64102566,0.65384614,0.9358974
+PerpetualBooster (default),0.032051284,0.032051284,0.025641026,0.032051284,0.032051284,0.032051284,0.025641026,0.032051284,0.032051284,0.032051284,0.03846154,0.051282052,0.044871796,0.044871796,0.044871796,0.032051284,0.044871796,0.051282052,0.044871796,0.032051284,0.032051284,0.03846154,0.051282052,0.057692308,0.057692308,0.07692308,0.102564104,0.057692308,0.057692308,0.070512824,0.23717949,0.06410257,0.06410257,0.10897436,0.08974359,0.12179487,0.07692308,0.07692308,0.08974359,0.051282052,0.08974359,0.083333336,0.07692308,0.2628205,0.09615385,0.17307693,0.12179487,0.27564102,0.22435898,0.115384616,0.21153846,0.1474359,0.23717949,0.31410256,0.32051283,0.2820513,0.30128205,0.25,0.28846154,0.4551282,0.30128205,0.44230768,0.44230768,0.48076922,0.18589744,0.3301282,,0.50641024,0.53846157,0.50641024,0.61538464,0.5897436,0.75641024
+FastaiMLP (default),0.0,0.0,0.0064102565,0.0064102565,0.0,0.0064102565,0.0064102565,0.012820513,0.0,0.0,0.0,0.057692308,0.057692308,0.01923077,0.01923077,0.01923077,0.051282052,0.025641026,0.057692308,0.025641026,0.012820513,0.012820513,0.057692308,0.0,0.0,0.0064102565,0.032051284,0.0064102565,0.09615385,0.01923077,0.19871795,0.01923077,0.10897436,0.08974359,0.057692308,0.03846154,0.032051284,0.10897436,0.070512824,0.12179487,0.032051284,0.12179487,0.051282052,0.0064102565,0.13461539,0.070512824,0.032051284,0.012820513,0.051282052,0.10897436,0.032051284,0.06410257,0.31410256,0.14102565,0.25,0.32051283,0.40384614,0.25,0.21794872,0.55128205,0.32051283,0.39102563,0.40384614,0.5897436,0.3974359,0.42948717,0.49358973,,0.3846154,0.71794873,0.7948718,0.90384614,0.99358976
+TabFlex (default),0.032051284,0.0,0.0,0.0,0.0,0.0,0.0064102565,0.051282052,0.0064102565,0.0,0.0,0.0,0.0,0.01923077,0.0,0.0,0.0064102565,0.0,0.0,0.0,0.0064102565,0.0064102565,0.012820513,0.0,0.0,0.0,0.0064102565,0.0,0.01923077,0.0,0.23717949,0.0064102565,0.051282052,0.0,0.025641026,0.0064102565,0.0,0.032051284,0.0064102565,0.044871796,0.0064102565,0.070512824,0.012820513,0.07692308,0.070512824,0.032051284,0.01923077,0.16025642,0.06410257,0.10897436,0.032051284,0.070512824,0.25641027,0.18589744,0.33333334,0.33974358,0.43589744,0.23717949,0.2820513,0.44230768,0.28846154,0.53846157,0.5448718,0.44871795,0.34615386,0.46153846,0.46153846,0.61538464,,0.724359,0.8076923,0.775641,1.0
+TabSTAR (default),0.0064102565,0.012820513,0.0064102565,0.0064102565,0.0064102565,0.0,0.0,0.0064102565,0.0,0.0064102565,0.012820513,0.057692308,0.06410257,0.032051284,0.0064102565,0.0064102565,0.032051284,0.012820513,0.070512824,0.0064102565,0.012820513,0.012820513,0.032051284,0.0064102565,0.0064102565,0.025641026,0.012820513,0.012820513,0.06410257,0.01923077,0.051282052,0.012820513,0.057692308,0.102564104,0.051282052,0.025641026,0.0064102565,0.06410257,0.01923077,0.06410257,0.01923077,0.07692308,0.012820513,0.06410257,0.12820514,0.044871796,0.025641026,0.06410257,0.083333336,0.057692308,0.057692308,0.057692308,0.07692308,0.102564104,0.17948718,0.16666667,0.35897437,0.102564104,0.26923078,0.37179488,0.21153846,0.0,0.0,0.3846154,0.4679487,0.525641,0.49358973,0.2820513,0.27564102,,0.6474359,0.69871795,0.96153843
+KNN (tuned),0.06410257,0.032051284,0.044871796,0.01923077,0.03846154,0.051282052,0.051282052,0.057692308,0.044871796,0.044871796,0.025641026,0.051282052,0.025641026,0.044871796,0.032051284,0.044871796,0.06410257,0.032051284,0.03846154,0.044871796,0.06410257,0.06410257,0.025641026,0.044871796,0.044871796,0.09615385,0.012820513,0.025641026,0.0,0.03846154,0.22435898,0.057692308,0.0064102565,0.115384616,0.10897436,0.057692308,0.051282052,0.0,0.025641026,0.044871796,0.057692308,0.025641026,0.070512824,0.102564104,0.09615385,0.057692308,0.025641026,0.102564104,0.13461539,0.070512824,0.13461539,0.070512824,0.07692308,0.070512824,0.1474359,0.10897436,0.06410257,0.083333336,0.15384616,0.23076923,0.23076923,0.14102565,0.14102565,0.24358974,0.3525641,0.35897437,0.3846154,0.20512821,0.1923077,0.3525641,,0.3525641,0.92948717
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.032051284,0.044871796,0.0064102565,0.0,0.0,0.0064102565,0.0,0.057692308,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.0064102565,0.0,0.0,0.051282052,0.0,0.17948718,0.0064102565,0.07692308,0.06410257,0.025641026,0.0064102565,0.0,0.070512824,0.0064102565,0.032051284,0.0064102565,0.09615385,0.0,0.0,0.01923077,0.012820513,0.0,0.0,0.025641026,0.01923077,0.01923077,0.0064102565,0.102564104,0.057692308,0.08974359,0.12820514,0.2948718,0.102564104,0.14102565,0.051282052,0.21794872,0.08974359,0.10897436,0.14102565,0.32051283,0.34615386,0.41025642,0.09615385,0.22435898,0.30128205,0.6474359,,0.9166667
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.06410257,0.06410257,0.012820513,0.012820513,0.06410257,0.057692308,0.06410257,0.24358974,0.0064102565,0.0,0.03846154,0.070512824,0.083333336,
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
index 04935a83181fd8164b85f56a10cd61f125789e5d..6fe1a809799e1511f2f02d30bdc3b0bd95a4ec6e 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Multiclass | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..a76a81eed8463fcabdcca1a299e756505fc547d5
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_multiclass/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":9121.0,"score":1.0,"rank":1.0,"hrank":1.0,"imp":0.0,"train_time":3.93,"predict_time":3.165,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":8156.0,"score":0.905,"rank":2.44,"hrank":2.3,"imp":2.724,"train_time":0.91,"predict_time":0.272,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":8028.0,"score":0.861,"rank":2.89,"hrank":2.7,"imp":3.919,"train_time":257.86,"predict_time":1.392,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":7765.0,"score":0.832,"rank":3.78,"hrank":3.67,"imp":4.701,"train_time":133.98,"predict_time":8.253,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":4,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":7084.0,"score":0.77,"rank":5.44,"hrank":5.29,"imp":6.303,"train_time":276.88,"predict_time":4.662,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":6998.0,"score":0.743,"rank":5.78,"hrank":5.74,"imp":6.994,"train_time":133.98,"predict_time":2.816,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":6,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":6643.0,"score":0.617,"rank":7.22,"hrank":7.05,"imp":10.043,"train_time":35.54,"predict_time":17.701,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":7,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":6586.0,"score":0.665,"rank":7.44,"hrank":7.41,"imp":8.893,"train_time":1.83,"predict_time":0.517,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":5356.0,"score":0.5,"rank":9.89,"hrank":9.76,"imp":12.741,"train_time":0.48,"predict_time":0.143,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":9,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":5104.0,"score":0.452,"rank":12.0,"hrank":11.6,"imp":13.784,"train_time":6148.01,"predict_time":0.15,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":10,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":5104.0,"score":0.452,"rank":12.0,"hrank":11.63,"imp":13.814,"train_time":61.95,"predict_time":0.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":5047.0,"score":0.432,"rank":12.67,"hrank":12.37,"imp":14.226,"train_time":157.72,"predict_time":3.649,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":12,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":4974.0,"score":0.44,"rank":13.56,"hrank":13.17,"imp":14.045,"train_time":6148.01,"predict_time":0.011,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":13,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":4915.0,"score":0.437,"rank":14.33,"hrank":13.74,"imp":14.138,"train_time":61.95,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":4897.0,"score":0.412,"rank":14.56,"hrank":14.09,"imp":14.638,"train_time":157.72,"predict_time":0.374,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":15,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":4879.0,"score":0.433,"rank":14.78,"hrank":13.63,"imp":14.204,"train_time":0.82,"predict_time":0.277,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":16,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":4651.0,"score":0.38,"rank":17.78,"hrank":17.71,"imp":15.327,"train_time":466.47,"predict_time":1.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":4614.0,"score":0.37,"rank":18.22,"hrank":17.84,"imp":15.529,"train_time":1199.98,"predict_time":0.344,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":4605.0,"score":0.367,"rank":18.33,"hrank":17.35,"imp":15.578,"train_time":1199.98,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":19,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":4530.0,"score":0.365,"rank":19.22,"hrank":19.06,"imp":15.649,"train_time":0.39,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":4502.0,"score":0.365,"rank":19.56,"hrank":19.45,"imp":15.649,"train_time":466.47,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":21,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":4102.0,"score":0.287,"rank":24.22,"hrank":23.87,"imp":17.235,"train_time":1222.57,"predict_time":21.653,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":22,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":4076.0,"score":0.289,"rank":24.56,"hrank":24.32,"imp":17.181,"train_time":3.47,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":23,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":4050.0,"score":0.268,"rank":24.89,"hrank":23.57,"imp":17.589,"train_time":1.48,"predict_time":0.53,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":24,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":4032.0,"score":0.277,"rank":25.11,"hrank":24.99,"imp":17.438,"train_time":68.36,"predict_time":0.189,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":25,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":4015.0,"score":0.278,"rank":25.33,"hrank":25.19,"imp":17.409,"train_time":68.36,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":26,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":4006.0,"score":0.276,"rank":25.44,"hrank":25.37,"imp":17.448,"train_time":4423.3,"predict_time":29.199,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":27,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":3971.0,"score":0.259,"rank":25.89,"hrank":25.51,"imp":17.775,"train_time":11.15,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":28,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":3743.0,"score":0.227,"rank":28.33,"hrank":28.28,"imp":18.417,"train_time":4423.3,"predict_time":2.237,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":29,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":3717.0,"score":0.221,"rank":28.56,"hrank":28.46,"imp":18.522,"train_time":2.82,"predict_time":1.073,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":30,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":2844.0,"score":0.137,"rank":31.44,"hrank":31.42,"imp":20.128,"train_time":26.01,"predict_time":0.151,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2640.0,"score":0.095,"rank":32.44,"hrank":32.4,"imp":20.889,"train_time":1222.57,"predict_time":7.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":32,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2486.0,"score":0.095,"rank":33.44,"hrank":33.4,"imp":20.889,"train_time":52.0,"predict_time":7.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":33,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":2344.0,"score":0.061,"rank":34.56,"hrank":34.41,"imp":21.497,"train_time":0.58,"predict_time":0.2,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":2266.0,"score":0.052,"rank":35.22,"hrank":35.16,"imp":21.684,"train_time":0.23,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":35,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2142.0,"score":0.018,"rank":36.33,"hrank":36.29,"imp":22.269,"train_time":4638.97,"predict_time":343.207,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":36,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":2080.0,"score":0.0,"rank":36.89,"hrank":36.76,"imp":22.636,"train_time":2389.34,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":37,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":2042.0,"score":0.003,"rank":37.22,"hrank":37.03,"imp":22.535,"train_time":2389.34,"predict_time":0.267,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":38,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1962.0,"score":0.0,"rank":37.89,"hrank":37.75,"imp":22.934,"train_time":26.01,"predict_time":0.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":39,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1587.0,"score":0.0,"rank":40.44,"hrank":40.44,"imp":24.385,"train_time":14.87,"predict_time":1.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":40,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1449.0,"score":0.0,"rank":41.44,"hrank":41.39,"imp":24.762,"train_time":1137.02,"predict_time":0.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":41,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1244.0,"score":0.0,"rank":43.22,"hrank":43.16,"imp":25.74,"train_time":4638.97,"predict_time":43.78,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":42,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1232.0,"score":0.0,"rank":43.33,"hrank":43.2,"imp":25.492,"train_time":1137.02,"predict_time":0.054,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":43,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1184.0,"score":0.0,"rank":43.78,"hrank":43.69,"imp":26.055,"train_time":9.87,"predict_time":9.237,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":44,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1172.0,"score":0.0,"rank":43.89,"hrank":43.81,"imp":26.094,"train_time":40.54,"predict_time":38.74,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":45,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":45.44,"hrank":45.32,"imp":26.928,"train_time":0.18,"predict_time":0.008,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":46,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":831.0,"score":0.0,"rank":46.78,"hrank":46.76,"imp":27.14,"train_time":1294.95,"predict_time":0.417,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":47,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":662.0,"score":0.0,"rank":47.89,"hrank":47.87,"imp":27.562,"train_time":1294.95,"predict_time":0.06,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":48,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":623.0,"score":0.0,"rank":48.11,"hrank":48.07,"imp":28.431,"train_time":0.18,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":49,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":29.0,"score":0.0,"rank":50.11,"hrank":50.11,"imp":31.618,"train_time":1863.61,"predict_time":6.302,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":50,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-412.0,"score":0.0,"rank":51.67,"hrank":51.66,"imp":34.019,"train_time":1863.61,"predict_time":1.131,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":51,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-502.0,"score":0.0,"rank":52.11,"hrank":52.08,"imp":34.009,"train_time":3.96,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":52,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-568.0,"score":0.0,"rank":52.44,"hrank":52.43,"imp":34.96,"train_time":3.96,"predict_time":0.077,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":53,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-1053.0,"score":0.0,"rank":55.0,"hrank":54.98,"imp":36.605,"train_time":66.07,"predict_time":1.792,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-1131.0,"score":0.0,"rank":55.5,"hrank":55.48,"imp":37.419,"train_time":1904.43,"predict_time":1.396,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":55,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-1140.0,"score":0.0,"rank":55.56,"hrank":55.48,"imp":37.434,"train_time":16.71,"predict_time":6.766,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":56,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-1166.0,"score":0.0,"rank":55.72,"hrank":55.71,"imp":37.463,"train_time":1904.43,"predict_time":2.971,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":57,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-1661.0,"score":0.0,"rank":57.89,"hrank":57.89,"imp":39.959,"train_time":7.55,"predict_time":0.149,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-2186.0,"score":0.0,"rank":59.44,"hrank":59.44,"imp":43.803,"train_time":569.0,"predict_time":0.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-2398.0,"score":0.0,"rank":60.44,"hrank":60.44,"imp":44.373,"train_time":569.0,"predict_time":1.658,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":60,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-2460.0,"score":0.0,"rank":60.78,"hrank":60.76,"imp":44.724,"train_time":65.83,"predict_time":1.583,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-2662.0,"score":0.0,"rank":61.89,"hrank":61.83,"imp":49.111,"train_time":28.2,"predict_time":0.515,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":62,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":-2916.0,"score":0.0,"rank":63.33,"hrank":63.32,"imp":51.715,"train_time":0.21,"predict_time":0.039,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-3106.0,"score":0.0,"rank":64.56,"hrank":64.48,"imp":52.242,"train_time":28.2,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-3155.0,"score":0.0,"rank":64.89,"hrank":64.88,"imp":53.569,"train_time":2.86,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":65,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-3236.0,"score":0.0,"rank":65.44,"hrank":65.43,"imp":53.943,"train_time":30.41,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":66,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-3542.0,"score":0.0,"rank":67.56,"hrank":67.55,"imp":57.076,"train_time":30.41,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-3661.0,"score":0.0,"rank":68.44,"hrank":68.43,"imp":57.449,"train_time":0.13,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":68,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-3719.0,"score":0.0,"rank":68.89,"hrank":68.8,"imp":57.324,"train_time":3.09,"predict_time":0.997,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-3865.0,"score":0.0,"rank":70.0,"hrank":69.96,"imp":59.086,"train_time":2.7,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-3896.0,"score":0.0,"rank":70.22,"hrank":70.21,"imp":58.52,"train_time":0.11,"predict_time":0.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":71,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-4035.0,"score":0.0,"rank":71.11,"hrank":71.1,"imp":59.325,"train_time":76.27,"predict_time":3.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":72,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-4973.0,"score":0.0,"rank":73.0,"hrank":73.0,"imp":72.158,"train_time":76.27,"predict_time":0.37,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-5748.0,"score":0.0,"rank":74.0,"hrank":74.0,"imp":83.474,"train_time":0.14,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/pareto_front_explorer.html
index 5ca97b2d459e916f09cb79eb13660c2ea657424c..9653ccb8521d9fe5d890eabe615f0cec2e8c0609 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7916536996,"imp":36.6045992319,"elo":-1052.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0212757285,"imp":17.181396612,"elo":4075.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0810120956,"imp":15.5776096649,"elo":4604.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3436400049,"imp":15.529159267,"elo":4614.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135873471,"imp":21.6835247423,"elo":2266.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0169735693,"imp":17.4085248711,"elo":4014.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.189181224,"imp":17.4383709448,"elo":4032.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"imp":17.7749534644,"elo":3971.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0112425999,"imp":14.0452240321,"elo":4974.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1501267968,"imp":13.7840392894,"elo":5103.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.039263078,"imp":51.7150429706,"elo":-2916.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0094929976,"imp":22.9336337304,"elo":1961.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1510706771,"imp":20.1277389889,"elo":2843.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0849591873,"imp":53.5688669452,"elo":-3154.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1696430964,"imp":43.8028639504,"elo":-2186.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6581625315,"imp":44.3728722396,"elo":-2397.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0398739264,"imp":28.431281613,"elo":623.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3738761825,"imp":14.6382026202,"elo":4897.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":3.6487831267,"imp":14.225675726,"elo":5046.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":3.1649737164,"imp":0.0,"elo":9121.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8552799147,"imp":24.3848771338,"elo":1586.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2366352943,"imp":18.4172970522,"elo":3743.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":29.1989921988,"imp":17.4484171722,"elo":4006.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7656160501,"imp":37.434168827,"elo":-1139.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0157167235,"imp":59.0862850659,"elo":-3864.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1481536883,"imp":52.242462816,"elo":-3105.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5145997597,"imp":49.1106037807,"elo":-2662.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0076988881,"imp":26.928310568,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0164622104,"imp":14.1379017686,"elo":4914.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1699002263,"imp":13.8142375627,"elo":5103.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.1492926143,"imp":39.9591212587,"elo":-1661.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0604026988,"imp":27.5615449118,"elo":662.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":0.4165686723,"imp":27.1397331062,"elo":831.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5296965383,"imp":17.5892425696,"elo":4049.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8157300323,"imp":6.9940304738,"elo":6997.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.253098869,"imp":4.7014297019,"elo":7765.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2370616309,"imp":26.0548850064,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.7402173897,"imp":26.0937916411,"elo":1172.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7803277189,"imp":25.7404923043,"elo":1243.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":343.2065516381,"imp":22.2692141684,"elo":2142.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2773446206,"imp":14.2035859276,"elo":4879.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":17.7006354133,"imp":10.0434584757,"elo":6643.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0228731968,"imp":58.5202549384,"elo":-3895.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.0732772508,"imp":18.5221964405,"elo":3716.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426444891,"imp":12.7406135408,"elo":5355.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0767204065,"imp":34.9601806584,"elo":-568.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0538978015,"imp":25.4915754476,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.4463981702,"imp":24.7620187568,"elo":1448.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5166405697,"imp":8.8934485237,"elo":6585.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.27178525,"imp":2.7237963588,"elo":8156.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.5831571832,"imp":44.7240058444,"elo":-2459.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.3962170912,"imp":37.4185477938,"elo":-1131.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9708756879,"imp":37.4631227318,"elo":-1166.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.19959583,"imp":21.4967094579,"elo":2343.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0246185533,"imp":34.0086714629,"elo":-502.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.023043642,"imp":22.6357042665,"elo":2080.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2669741002,"imp":22.5345265676,"elo":2042.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"imp":15.6491623994,"elo":4529.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"imp":15.6491624778,"elo":4502.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0358030998,"imp":15.3274877808,"elo":4650.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"imp":20.8891440058,"elo":2486.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"imp":20.8891386829,"elo":2639.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":21.6530909246,"imp":17.2347396887,"elo":4101.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9973054408,"imp":57.3240231598,"elo":-3718.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":1.1308430496,"imp":34.019215077,"elo":-412.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":6.3018889681,"imp":31.6183153791,"elo":29.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7916536996,"x_train":66.0653250838,"imp":36.6045992319,"elo":-1052.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0212757285,"x_train":3.4684114436,"imp":17.181396612,"elo":4075.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0810120956,"x_train":1199.9825110223,"imp":15.5776096649,"elo":4604.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3436400049,"x_train":1199.9825110223,"imp":15.529159267,"elo":4614.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135873471,"x_train":0.2327292548,"imp":21.6835247423,"elo":2266.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0169735693,"x_train":68.3588435593,"imp":17.4085248711,"elo":4014.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.189181224,"x_train":68.3588435593,"imp":17.4383709448,"elo":4032.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"x_train":11.1470283442,"imp":17.7749534644,"elo":3971.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0112425999,"x_train":6148.0105869876,"imp":14.0452240321,"elo":4974.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1501267968,"x_train":6148.0105869876,"imp":13.7840392894,"elo":5103.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.039263078,"x_train":0.210649465,"imp":51.7150429706,"elo":-2916.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0094929976,"x_train":26.0128124598,"imp":22.9336337304,"elo":1961.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1510706771,"x_train":26.0128124598,"imp":20.1277389889,"elo":2843.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0849591873,"x_train":2.8561491959,"imp":53.5688669452,"elo":-3154.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1696430964,"x_train":569.0042450609,"imp":43.8028639504,"elo":-2186.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6581625315,"x_train":569.0042450609,"imp":44.3728722396,"elo":-2397.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0398739264,"x_train":0.177779185,"imp":28.431281613,"elo":623.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3738761825,"x_train":157.7188520507,"imp":14.6382026202,"elo":4897.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":3.6487831267,"x_train":157.7188520507,"imp":14.225675726,"elo":5046.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":3.1649737164,"x_train":3.9298653893,"imp":0.0,"elo":9121.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8552799147,"x_train":14.8696863446,"imp":24.3848771338,"elo":1586.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2366352943,"x_train":4423.2983658455,"imp":18.4172970522,"elo":3743.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":29.1989921988,"x_train":4423.2983658455,"imp":17.4484171722,"elo":4006.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7656160501,"x_train":16.7081227293,"imp":37.434168827,"elo":-1139.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0157167235,"x_train":2.7029391458,"imp":59.0862850659,"elo":-3864.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1481536883,"x_train":28.1974045386,"imp":52.242462816,"elo":-3105.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5145997597,"x_train":28.1974045386,"imp":49.1106037807,"elo":-2662.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0076988881,"x_train":0.1832795453,"imp":26.928310568,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0164622104,"x_train":61.9490599096,"imp":14.1379017686,"elo":4914.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1699002263,"x_train":61.9490599096,"imp":13.8142375627,"elo":5103.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.1492926143,"x_train":7.5476167071,"imp":39.9591212587,"elo":-1661.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0604026988,"x_train":1294.9500816771,"imp":27.5615449118,"elo":662.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":0.4165686723,"x_train":1294.9500816771,"imp":27.1397331062,"elo":831.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5296965383,"x_train":1.4775118336,"imp":17.5892425696,"elo":4049.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8157300323,"x_train":133.9809627068,"imp":6.9940304738,"elo":6997.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.253098869,"x_train":133.9809627068,"imp":4.7014297019,"elo":7765.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2370616309,"x_train":9.8664582493,"imp":26.0548850064,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.7402173897,"x_train":40.5370562747,"imp":26.0937916411,"elo":1172.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7803277189,"x_train":4638.9676267895,"imp":25.7404923043,"elo":1243.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":343.2065516381,"x_train":4638.9676267895,"imp":22.2692141684,"elo":2142.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2773446206,"x_train":0.8153299726,"imp":14.2035859276,"elo":4879.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":17.7006354133,"x_train":35.5390806742,"imp":10.0434584757,"elo":6643.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0228731968,"x_train":0.1112663146,"imp":58.5202549384,"elo":-3895.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.0732772508,"x_train":2.8164229235,"imp":18.5221964405,"elo":3716.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426444891,"x_train":0.4844563774,"imp":12.7406135408,"elo":5355.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0767204065,"x_train":3.9605489519,"imp":34.9601806584,"elo":-568.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0538978015,"x_train":1137.0204365721,"imp":25.4915754476,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.4463981702,"x_train":1137.0204365721,"imp":24.7620187568,"elo":1448.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5166405697,"x_train":1.8291240885,"imp":8.8934485237,"elo":6585.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.27178525,"x_train":0.9127892597,"imp":2.7237963588,"elo":8156.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.5831571832,"x_train":65.832151552,"imp":44.7240058444,"elo":-2459.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.3962170912,"x_train":1904.4279322336,"imp":37.4185477938,"elo":-1131.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9708756879,"x_train":1904.4279322336,"imp":37.4631227318,"elo":-1166.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.19959583,"x_train":0.5828990849,"imp":21.4967094579,"elo":2343.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0246185533,"x_train":3.9645298444,"imp":34.0086714629,"elo":-502.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.023043642,"x_train":2389.3401785968,"imp":22.6357042665,"elo":2080.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2669741002,"x_train":2389.3401785968,"imp":22.5345265676,"elo":2042.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"x_train":0.3879892054,"imp":15.6491623994,"elo":4529.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"x_train":466.4747671387,"imp":15.6491624778,"elo":4502.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0358030998,"x_train":466.4747671387,"imp":15.3274877808,"elo":4650.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"x_train":52.0002317866,"imp":20.8891440058,"elo":2486.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"x_train":1222.5730299669,"imp":20.8891386829,"elo":2639.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":21.6530909246,"x_train":1222.5730299669,"imp":17.2347396887,"elo":4101.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9973054408,"x_train":3.0877205487,"imp":57.3240231598,"elo":-3718.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":1.1308430496,"x_train":1863.6070699824,"imp":34.019215077,"elo":-412.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":6.3018889681,"x_train":1863.6070699824,"imp":31.6183153791,"elo":29.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
index 84fd883e477e835c1dc91e81609e237dd502589d..b08b9a6be95668f322426536a4fab58bb69fac24 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":66.0653250838,"imp":36.6174442254,"elo":-433.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.4684114436,"imp":17.1982189422,"elo":3642.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":9.4286940119,"imp":16.9514887084,"elo":3743.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":26.7881730821,"imp":16.4684558604,"elo":3906.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":53.5391935238,"imp":16.1390864594,"elo":4030.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":152.7281555404,"imp":15.7936821457,"elo":4186.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":297.8045100678,"imp":15.6874958736,"elo":4222.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":592.1492998961,"imp":15.5980918525,"elo":4279.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":887.2283364787,"imp":15.5567305187,"elo":4342.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1199.9825110223,"imp":15.5408704363,"elo":4364.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.2327292548,"imp":21.6993668415,"elo":2177.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.5808796011,"imp":20.704799374,"elo":2386.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.5148941182,"imp":19.3360040969,"elo":2906.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.1362950218,"imp":18.4935968731,"elo":3184.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":8.2937811532,"imp":18.0447983035,"elo":3316.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":16.9925774777,"imp":17.7958889184,"elo":3431.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":34.110705437,"imp":17.5701952871,"elo":3543.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":51.0567278918,"imp":17.4937654108,"elo":3600.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":68.3588435593,"imp":17.451429721,"elo":3645.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":11.1470283442,"imp":17.7912965293,"elo":3509.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":47.4246188425,"imp":16.6077982163,"elo":3875.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":136.0261588349,"imp":15.1701021831,"elo":4406.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":277.1128819248,"imp":14.8534981024,"elo":4487.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":742.3379842722,"imp":14.2956825998,"elo":4697.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":1521.2363697091,"imp":14.0882037921,"elo":4793.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":3056.5759955925,"imp":13.8958902042,"elo":4893.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":4572.6784881051,"imp":13.8335401341,"elo":4942.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":6148.0105869876,"imp":13.8029632333,"elo":4965.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.210649465,"imp":51.7258237302,"elo":-1781.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3400177245,"imp":44.4605754723,"elo":-1218.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.7306475516,"imp":30.0898438644,"elo":451.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.3716768748,"imp":24.3152042357,"elo":1575.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.3038879991,"imp":20.7499932217,"elo":2366.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.5251049409,"imp":20.419671907,"elo":2481.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.9730278534,"imp":20.2485255705,"elo":2554.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":19.420546809,"imp":20.1758964858,"elo":2608.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":26.0128124598,"imp":20.1432638787,"elo":2676.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.8561491959,"imp":53.5782989817,"elo":-1992.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.5274992778,"imp":53.7506549797,"elo":-2052.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.4182504903,"imp":51.8872540473,"elo":-1798.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":29.2437710823,"imp":49.4987366034,"elo":-1663.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":68.0373867319,"imp":48.2656225446,"elo":-1593.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":137.9784811137,"imp":47.1212705176,"elo":-1526.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":282.4399735782,"imp":46.0015801052,"elo":-1467.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":424.0861878909,"imp":45.0891495896,"elo":-1362.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":569.0042450609,"imp":44.3852270948,"elo":-1189.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.177779185,"imp":28.4458404969,"elo":670.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":1.2034968977,"imp":28.243018236,"elo":728.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":3.3940605174,"imp":25.6888302783,"elo":1242.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":7.2354516871,"imp":23.3133499239,"elo":1849.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":19.3256413209,"imp":17.7328375726,"elo":3453.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":38.6974931332,"imp":14.9559744843,"elo":4500.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":77.3490317839,"imp":14.4548346099,"elo":4728.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":116.3662460744,"imp":14.3500035973,"elo":4786.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":157.7188520507,"imp":14.2408141027,"elo":4868.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":3.9298653893,"imp":0.0394693775,"elo":7356.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.8696863446,"imp":24.4001445962,"elo":1564.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":38.3994266998,"imp":23.1409840761,"elo":1870.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":107.9937237268,"imp":21.7226317994,"elo":2123.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":214.6834801346,"imp":20.6688083433,"elo":2423.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":519.7617286711,"imp":19.327809423,"elo":2912.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1089.2077290986,"imp":18.5238195726,"elo":3193.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2164.8113675288,"imp":17.9392339201,"elo":3392.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3248.9873614597,"imp":17.6293029817,"elo":3494.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4423.2983658455,"imp":17.4693206477,"elo":3605.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":16.7081227293,"imp":38.4197521996,"elo":-632.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":2.7029391458,"imp":59.1337484624,"elo":-2915.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":8.4856383099,"imp":54.1095862893,"elo":-2110.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":23.7831181563,"imp":49.4770666427,"elo":-1671.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":28.1974045386,"imp":49.1229578957,"elo":-1582.1,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1832795453,"imp":27.0274509886,"elo":939.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4706668018,"imp":19.1404842461,"elo":2978.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.4894550512,"imp":16.0331829814,"elo":4155.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":3.053659024,"imp":15.0376831298,"elo":4398.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":7.5560037194,"imp":14.3353184179,"elo":4638.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":15.2866374484,"imp":14.1431702342,"elo":4725.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":30.5951033309,"imp":13.9589189437,"elo":4848.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":46.1662849504,"imp":13.8865443621,"elo":4899.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":61.9490599096,"imp":13.8317101134,"elo":4970.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":7.5476167071,"imp":39.9711686616,"elo":-797.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":13.2940077986,"imp":36.6622870804,"elo":-451.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":31.6350352615,"imp":34.1014658056,"elo":-123.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":62.0439269442,"imp":32.0827300636,"elo":188.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":161.8790996214,"imp":30.4501240663,"elo":361.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":328.4372725404,"imp":29.7166950188,"elo":484.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":646.6627533516,"imp":28.0626103943,"elo":679.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":968.2000565056,"imp":27.607032205,"elo":772.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1294.9500816771,"imp":27.1552069413,"elo":888.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.4775118336,"imp":17.6057023404,"elo":3593.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.9115484001,"imp":17.348949422,"elo":3696.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":3.2307266592,"imp":16.5991884106,"elo":3993.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.3145738811,"imp":16.2208138,"elo":4106.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":13.6646267949,"imp":13.878512824,"elo":4728.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":29.585585492,"imp":9.278089573,"elo":5816.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":63.4593657803,"imp":6.6147821389,"elo":6202.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":95.8186666767,"imp":5.1506740501,"elo":6370.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":133.9809627068,"imp":4.7186930132,"elo":6615.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":9.8664582493,"imp":26.0697540295,"elo":1174.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":40.5370562747,"imp":26.1086306019,"elo":1128.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.6468711363,"imp":25.8184706186,"elo":1225.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":120.4467271947,"imp":25.6314801824,"elo":1265.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":216.0600802864,"imp":25.327952527,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":558.8453433529,"imp":24.2818339933,"elo":1606.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1141.3485940959,"imp":23.7229068436,"elo":1730.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2295.535465635,"imp":23.1113096347,"elo":1870.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3458.2236513386,"imp":22.5100960665,"elo":1986.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4638.9676267895,"imp":22.2859884097,"elo":2076.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.8153299726,"imp":14.2207822206,"elo":4633.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":35.5390806742,"imp":10.0614938152,"elo":5853.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1112663146,"imp":58.5286308898,"elo":-2876.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":2.8164229235,"imp":18.5386881963,"elo":3231.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.4844563774,"imp":12.7581340857,"elo":5200.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":3.9605489519,"imp":34.9733101595,"elo":-169.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.0361437663,"imp":32.0711518311,"elo":193.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":26.071931656,"imp":28.6167683404,"elo":619.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":54.8698111305,"imp":27.1394097339,"elo":902.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":141.1191273125,"imp":26.2067443373,"elo":1112.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":283.0524385403,"imp":25.6380589495,"elo":1212.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":559.8211318778,"imp":25.1186792563,"elo":1331.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":845.621900784,"imp":24.9207972757,"elo":1388.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1137.0204365721,"imp":24.7782924451,"elo":1461.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":1.8291240885,"imp":8.9117827376,"elo":5853.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":0.9127892597,"imp":2.7433951595,"elo":6848.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":65.832151552,"imp":44.7350522638,"elo":-1316.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":73.4850259904,"imp":44.6210107202,"elo":-1234.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":96.0490631917,"imp":44.5495907933,"elo":-1230.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":123.4027199075,"imp":44.2554746282,"elo":-1127.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":246.4013576646,"imp":42.8781169115,"elo":-968.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":469.2886989974,"imp":41.053101283,"elo":-862.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":931.600810241,"imp":39.6973188826,"elo":-758.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1393.3493582788,"imp":38.5222349779,"elo":-665.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1904.4279322336,"imp":37.4756553614,"elo":-529.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.5828990849,"imp":21.512349933,"elo":2274.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":3.9645298444,"imp":34.0236440978,"elo":-123.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.6616998352,"imp":33.2937109549,"elo":33.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":43.5410231146,"imp":32.0499097177,"elo":183.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":110.0430207107,"imp":29.7022282795,"elo":484.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":278.6743902359,"imp":27.524603557,"elo":792.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":589.8673134165,"imp":26.1048022154,"elo":1139.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1190.1861421136,"imp":24.659891563,"elo":1479.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1804.3563357667,"imp":23.8664121165,"elo":1685.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2389.3401785968,"imp":22.5506414343,"elo":1990.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.3879892054,"imp":15.6662513974,"elo":4171.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":3.362783389,"imp":15.5680144478,"elo":4194.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":10.9221958575,"imp":15.4900570576,"elo":4217.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":20.5995586615,"imp":15.4798083228,"elo":4233.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":57.8450913396,"imp":15.4229125134,"elo":4286.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":116.3010606073,"imp":15.4015823622,"elo":4317.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":231.8962989377,"imp":15.366403122,"elo":4351.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":345.7526810549,"imp":15.3570343393,"elo":4370.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":466.4747671387,"imp":15.3412007302,"elo":4412.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":52.0002317866,"imp":20.9051616936,"elo":2408.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":96.7947394331,"imp":20.1000134846,"elo":2654.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":239.4028626458,"imp":19.5210244374,"elo":2840.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":475.5145519061,"imp":18.7170397709,"elo":3202.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1177.5846933433,"imp":17.2998557543,"elo":3651.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1222.5730299669,"imp":17.2514686705,"elo":3699.7,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.0877205487,"imp":57.3324075538,"elo":-2708.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":12.6641668459,"imp":52.0527755162,"elo":-1877.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":42.6272411378,"imp":43.5787945161,"elo":-1132.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":88.9979412195,"imp":37.2830523058,"elo":-515.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":230.8719772304,"imp":34.8647448355,"elo":-242.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":461.2572637645,"imp":33.8141077812,"elo":-105.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":931.0175207888,"imp":32.9036398699,"elo":28.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1389.8161177677,"imp":32.1679325704,"elo":145.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1863.6070699824,"imp":31.63374393,"elo":241.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7916536996,"x_train":66.0653250838,"imp":36.6174442254,"elo":-433.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0212757285,"x_train":3.4684114436,"imp":17.1982189422,"elo":3642.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0478586519,"x_train":9.4286940119,"imp":16.9514887084,"elo":3743.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.10020597,"x_train":26.7881730821,"imp":16.4684558604,"elo":3906.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1219205427,"x_train":53.5391935238,"imp":16.1390864594,"elo":4030.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2112305874,"x_train":152.7281555404,"imp":15.7936821457,"elo":4186.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2478990887,"x_train":297.8045100678,"imp":15.6874958736,"elo":4222.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2868644888,"x_train":592.1492998961,"imp":15.5980918525,"elo":4279.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3131577283,"x_train":887.2283364787,"imp":15.5567305187,"elo":4342.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.348787563,"x_train":1199.9825110223,"imp":15.5408704363,"elo":4364.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135873471,"x_train":0.2327292548,"imp":21.6993668415,"elo":2177.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306763366,"x_train":0.5808796011,"imp":20.704799374,"elo":2386.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.058031433,"x_train":1.5148941182,"imp":19.3360040969,"elo":2906.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0799589954,"x_train":3.1362950218,"imp":18.4935968731,"elo":3184.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.131080448,"x_train":8.2937811532,"imp":18.0447983035,"elo":3316.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1586631664,"x_train":16.9925774777,"imp":17.7958889184,"elo":3431.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1774151287,"x_train":34.110705437,"imp":17.5701952871,"elo":3543.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1906943335,"x_train":51.0567278918,"imp":17.4937654108,"elo":3600.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1929433857,"x_train":68.3588435593,"imp":17.451429721,"elo":3645.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"x_train":11.1470283442,"imp":17.7912965293,"elo":3509.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0304247242,"x_train":47.4246188425,"imp":16.6077982163,"elo":3875.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.054592842,"x_train":136.0261588349,"imp":15.1701021831,"elo":4406.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0693827636,"x_train":277.1128819248,"imp":14.8534981024,"elo":4487.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1024467109,"x_train":742.3379842722,"imp":14.2956825998,"elo":4697.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1234968274,"x_train":1521.2363697091,"imp":14.0882037921,"elo":4793.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1405330049,"x_train":3056.5759955925,"imp":13.8958902042,"elo":4893.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1464533011,"x_train":4572.6784881051,"imp":13.8335401341,"elo":4942.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1501267968,"x_train":6148.0105869876,"imp":13.8029632333,"elo":4965.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.039263078,"x_train":0.210649465,"imp":51.7258237302,"elo":-1781.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0420610938,"x_train":0.3400177245,"imp":44.4605754723,"elo":-1218.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.035991066,"x_train":0.7306475516,"imp":30.0898438644,"elo":451.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0285236942,"x_train":1.3716768748,"imp":24.3152042357,"elo":1575.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0404946536,"x_train":3.3038879991,"imp":20.7499932217,"elo":2366.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0642507934,"x_train":6.5251049409,"imp":20.419671907,"elo":2481.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1069767422,"x_train":12.9730278534,"imp":20.2485255705,"elo":2554.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1321511543,"x_train":19.420546809,"imp":20.1758964858,"elo":2608.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1510706771,"x_train":26.0128124598,"imp":20.1432638787,"elo":2676.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0849591873,"x_train":2.8561491959,"imp":53.5782989817,"elo":-1992.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1669188087,"x_train":5.5274992778,"imp":53.7506549797,"elo":-2052.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3945555164,"x_train":14.4182504903,"imp":51.8872540473,"elo":-1798.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6441505685,"x_train":29.2437710823,"imp":49.4987366034,"elo":-1663.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9551913221,"x_train":68.0373867319,"imp":48.2656225446,"elo":-1593.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2793194595,"x_train":137.9784811137,"imp":47.1212705176,"elo":-1526.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7169442466,"x_train":282.4399735782,"imp":46.0015801052,"elo":-1467.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7762697202,"x_train":424.0861878909,"imp":45.0891495896,"elo":-1362.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6589060596,"x_train":569.0042450609,"imp":44.3852270948,"elo":-1189.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0398739264,"x_train":0.177779185,"imp":28.4458404969,"elo":670.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.707636671,"x_train":1.2034968977,"imp":28.243018236,"elo":728.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7934990344,"x_train":3.3940605174,"imp":25.6888302783,"elo":1242.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4799330005,"x_train":7.2354516871,"imp":23.3133499239,"elo":1849.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3202136893,"x_train":19.3256413209,"imp":17.7328375726,"elo":3453.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3734167029,"x_train":38.6974931332,"imp":14.9559744843,"elo":4500.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.0760086735,"x_train":77.3490317839,"imp":14.4548346099,"elo":4728.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.5418049787,"x_train":116.3662460744,"imp":14.3500035973,"elo":4786.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.6487831267,"x_train":157.7188520507,"imp":14.2408141027,"elo":4868.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":3.1649737164,"x_train":3.9298653893,"imp":0.0394693775,"elo":7356.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8552799147,"x_train":14.8696863446,"imp":24.4001445962,"elo":1564.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.8092665406,"x_train":38.3994266998,"imp":23.1409840761,"elo":1870.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.9564161978,"x_train":107.9937237268,"imp":21.7226317994,"elo":2123.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":11.1778615785,"x_train":214.6834801346,"imp":20.6688083433,"elo":2423.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":16.8970014449,"x_train":519.7617286711,"imp":19.327809423,"elo":2912.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":20.2753612803,"x_train":1089.2077290986,"imp":18.5238195726,"elo":3193.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":23.4597615325,"x_train":2164.8113675288,"imp":17.9392339201,"elo":3392.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":26.8100352995,"x_train":3248.9873614597,"imp":17.6293029817,"elo":3494.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":29.2376658843,"x_train":4423.2983658455,"imp":17.4693206477,"elo":3605.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7656160501,"x_train":16.7081227293,"imp":38.4197521996,"elo":-632.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0157167235,"x_train":2.7029391458,"imp":59.1337484624,"elo":-2915.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1460145839,"x_train":8.4856383099,"imp":54.1095862893,"elo":-2110.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4447962725,"x_train":23.7831181563,"imp":49.4770666427,"elo":-1671.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5145997597,"x_train":28.1974045386,"imp":49.1229578957,"elo":-1582.1,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0076988881,"x_train":0.1832795453,"imp":27.0274509886,"elo":939.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0211762886,"x_train":0.4706668018,"imp":19.1404842461,"elo":2978.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0401206899,"x_train":1.4894550512,"imp":16.0331829814,"elo":4155.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0553042309,"x_train":3.053659024,"imp":15.0376831298,"elo":4398.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0810592215,"x_train":7.5560037194,"imp":14.3353184179,"elo":4638.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1070101817,"x_train":15.2866374484,"imp":14.1431702342,"elo":4725.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1387414025,"x_train":30.5951033309,"imp":13.9589189437,"elo":4848.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1516373222,"x_train":46.1662849504,"imp":13.8865443621,"elo":4899.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1706958863,"x_train":61.9490599096,"imp":13.8317101134,"elo":4970.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.1492926143,"x_train":7.5476167071,"imp":39.9711686616,"elo":-797.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.209324614,"x_train":13.2940077986,"imp":36.6622870804,"elo":-451.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2344952565,"x_train":31.6350352615,"imp":34.1014658056,"elo":-123.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2811166627,"x_train":62.0439269442,"imp":32.0827300636,"elo":188.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3605099605,"x_train":161.8790996214,"imp":30.4501240663,"elo":361.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4238472839,"x_train":328.4372725404,"imp":29.7166950188,"elo":484.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3839306352,"x_train":646.6627533516,"imp":28.0626103943,"elo":679.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.406885263,"x_train":968.2000565056,"imp":27.607032205,"elo":772.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4165686723,"x_train":1294.9500816771,"imp":27.1552069413,"elo":888.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5296965383,"x_train":1.4775118336,"imp":17.6057023404,"elo":3593.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6843170741,"x_train":1.9115484001,"imp":17.348949422,"elo":3696.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9983103811,"x_train":3.2307266592,"imp":16.5991884106,"elo":3993.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7351012291,"x_train":5.3145738811,"imp":16.2208138,"elo":4106.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2130894628,"x_train":13.6646267949,"imp":13.878512824,"elo":4728.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.4628158113,"x_train":29.585585492,"imp":9.278089573,"elo":5816.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.400599732,"x_train":63.4593657803,"imp":6.6147821389,"elo":6202.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4275461958,"x_train":95.8186666767,"imp":5.1506740501,"elo":6370.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.253098869,"x_train":133.9809627068,"imp":4.7186930132,"elo":6615.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2370616309,"x_train":9.8664582493,"imp":26.0697540295,"elo":1174.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":38.7402173897,"x_train":40.5370562747,"imp":26.1086306019,"elo":1128.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.185823341,"x_train":63.6468711363,"imp":25.8184706186,"elo":1225.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":81.0004561794,"x_train":120.4467271947,"imp":25.6314801824,"elo":1265.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":114.1751843382,"x_train":216.0600802864,"imp":25.327952527,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":170.4030706025,"x_train":558.8453433529,"imp":24.2818339933,"elo":1606.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":231.5103412116,"x_train":1141.3485940959,"imp":23.7229068436,"elo":1730.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":294.9832486808,"x_train":2295.535465635,"imp":23.1113096347,"elo":1870.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":314.3749405201,"x_train":3458.2236513386,"imp":22.5100960665,"elo":1986.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":343.2065516381,"x_train":4638.9676267895,"imp":22.2859884097,"elo":2076.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2773446206,"x_train":0.8153299726,"imp":14.2207822206,"elo":4633.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":17.7006354133,"x_train":35.5390806742,"imp":10.0614938152,"elo":5853.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0228731968,"x_train":0.1112663146,"imp":58.5286308898,"elo":-2876.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.0732772508,"x_train":2.8164229235,"imp":18.5386881963,"elo":3231.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426444891,"x_train":0.4844563774,"imp":12.7581340857,"elo":5200.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0767204065,"x_train":3.9605489519,"imp":34.9733101595,"elo":-169.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1313391893,"x_train":9.0361437663,"imp":32.0711518311,"elo":193.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2016062313,"x_train":26.071931656,"imp":28.6167683404,"elo":619.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2690374607,"x_train":54.8698111305,"imp":27.1394097339,"elo":902.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3287615415,"x_train":141.1191273125,"imp":26.2067443373,"elo":1112.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.3928311571,"x_train":283.0524385403,"imp":25.6380589495,"elo":1212.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4060980443,"x_train":559.8211318778,"imp":25.1186792563,"elo":1331.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.4244678038,"x_train":845.621900784,"imp":24.9207972757,"elo":1388.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.4463981702,"x_train":1137.0204365721,"imp":24.7782924451,"elo":1461.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5166405697,"x_train":1.8291240885,"imp":8.9117827376,"elo":5853.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.27178525,"x_train":0.9127892597,"imp":2.7433951595,"elo":6848.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.5831571832,"x_train":65.832151552,"imp":44.7350522638,"elo":-1316.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6334117897,"x_train":73.4850259904,"imp":44.6210107202,"elo":-1234.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.7494818728,"x_train":96.0490631917,"imp":44.5495907933,"elo":-1230.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.8701977363,"x_train":123.4027199075,"imp":44.2554746282,"elo":-1127.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.3864186433,"x_train":246.4013576646,"imp":42.8781169115,"elo":-968.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.8424963274,"x_train":469.2886989974,"imp":41.053101283,"elo":-862.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6043703251,"x_train":931.600810241,"imp":39.6973188826,"elo":-758.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.506791696,"x_train":1393.3493582788,"imp":38.5222349779,"elo":-665.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9708756879,"x_train":1904.4279322336,"imp":37.4756553614,"elo":-529.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.19959583,"x_train":0.5828990849,"imp":21.512349933,"elo":2274.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0246185533,"x_train":3.9645298444,"imp":34.0236440978,"elo":-123.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0576358482,"x_train":14.6616998352,"imp":33.2937109549,"elo":33.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1132573234,"x_train":43.5410231146,"imp":32.0499097177,"elo":183.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.152027174,"x_train":110.0430207107,"imp":29.7022282795,"elo":484.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2158473884,"x_train":278.6743902359,"imp":27.524603557,"elo":792.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2573350856,"x_train":589.8673134165,"imp":26.1048022154,"elo":1139.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.277175668,"x_train":1190.1861421136,"imp":24.659891563,"elo":1479.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2880694779,"x_train":1804.3563357667,"imp":23.8664121165,"elo":1685.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2668851689,"x_train":2389.3401785968,"imp":22.5506414343,"elo":1990.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"x_train":0.3879892054,"imp":15.6662513974,"elo":4171.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.366384192,"x_train":3.362783389,"imp":15.5680144478,"elo":4194.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6186026273,"x_train":10.9221958575,"imp":15.4900570576,"elo":4217.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6552489612,"x_train":20.5995586615,"imp":15.4798083228,"elo":4233.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8977020343,"x_train":57.8450913396,"imp":15.4229125134,"elo":4286.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9105374137,"x_train":116.3010606073,"imp":15.4015823622,"elo":4317.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8753171531,"x_train":231.8962989377,"imp":15.366403122,"elo":4351.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9589825129,"x_train":345.7526810549,"imp":15.3570343393,"elo":4370.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0370834578,"x_train":466.4747671387,"imp":15.3412007302,"elo":4412.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"x_train":52.0002317866,"imp":20.9051616936,"elo":2408.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":11.8109670804,"x_train":96.7947394331,"imp":20.1000134846,"elo":2654.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":16.6639574996,"x_train":239.4028626458,"imp":19.5210244374,"elo":2840.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":18.6752813789,"x_train":475.5145519061,"imp":18.7170397709,"elo":3202.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":21.4685568314,"x_train":1177.5846933433,"imp":17.2998557543,"elo":3651.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":21.6530909246,"x_train":1222.5730299669,"imp":17.2514686705,"elo":3699.7,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9973054408,"x_train":3.0877205487,"imp":57.3324075538,"elo":-2708.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.9976912644,"x_train":12.6641668459,"imp":52.0527755162,"elo":-1877.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.9612924274,"x_train":42.6272411378,"imp":43.5787945161,"elo":-1132.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":3.5953994677,"x_train":88.9979412195,"imp":37.2830523058,"elo":-515.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":5.3075739184,"x_train":230.8719772304,"imp":34.8647448355,"elo":-242.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":6.4637514204,"x_train":461.2572637645,"imp":33.8141077812,"elo":-105.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":7.0798232983,"x_train":931.0175207888,"imp":32.9036398699,"elo":28.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":7.046435977,"x_train":1389.8161177677,"imp":32.1679325704,"elo":145.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":6.3018889681,"x_train":1863.6070699824,"imp":31.63374393,"elo":241.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..33c06dbc68389e17b442e7eb555309b0b0f34a2c
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_multiclass/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["LimiX (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (tuned)", "TabFM (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "EBM (tuned + ensembled)", "RandomForest (tuned + ensembled)", "LightGBM (tuned + ensembled)", "EBM (tuned)", "RandomForest (tuned)", "LightGBM (tuned)", "TabDPT-Turbo (default)", "XGBoost (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "XGBoost (default)", "XGBoost (tuned)", "iLTM (tuned + ensembled)", "CatBoost (default)", "RealTabPFN-2.5 (default)", "ChimeraBoost (tuned + ensembled)", "ChimeraBoost (tuned)", "ModernNCA (tuned + ensembled)", "EBM (default)", "ModernNCA (tuned)", "TabICL (default)", "ExtraTrees (tuned + ensembled)", "iLTM (tuned)", "iLTM (default)", "TabSwift (default)", "ChimeraBoost (default)", "TabDPT (tuned + ensembled)", "TorchMLP (tuned)", "TorchMLP (tuned + ensembled)", "ExtraTrees (tuned)", "ModernNCA (default)", "TabM (tuned + ensembled)", "TabDPT (tuned)", "TabM (tuned)", "SAP-RPT-OSS (default)", "TabDPT (default)", "RandomForest (default)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "LightGBM (default)", "xRFM (tuned + ensembled)", "xRFM (tuned)", "TorchMLP (default)", "TabM (default)", "BetaTabPFN (default)", "TabSTAR (tuned)", "OrionMSP (default)", "TabSTAR (tuned + ensembled)", "RealMLP (default)", "FastaiMLP (tuned)", "FastaiMLP (tuned + ensembled)", "TabSTAR (default)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "xRFM (default)", "PerpetualBooster (default)", "TabFlex (default)", "KNN (tuned + ensembled)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, NaN, 0.6666666865348816, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.3333333432674408, NaN, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.2222222238779068, NaN, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, NaN, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.3333333432674408, NaN, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, NaN, 0.4444444477558136, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5555555820465088, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.4444444477558136, NaN, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, NaN, 0.5555555820465088, 0.5555555820465088, 0.8888888955116272, 0.4444444477558136, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.4444444477558136, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, NaN, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, NaN, 0.5555555820465088, 0.4444444477558136, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, NaN, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.8888888955116272, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, NaN, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, NaN, 0.6666666865348816, 1.0, 1.0, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, NaN, 1.0, 1.0, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.0, 0.0, NaN, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4444444477558136, NaN, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, NaN, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.6666666865348816, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, NaN, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.2222222238779068, 0.6666666865348816, 0.4444444477558136, NaN, 0.7777777910232544, 0.4444444477558136, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.3333333432674408, 0.2222222238779068, NaN, 0.5555555820465088, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, NaN, 0.7777777910232544, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, NaN, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, NaN, 1.0, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, NaN, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, NaN, 0.4444444477558136, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.5555555820465088, NaN, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.3333333432674408, NaN, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, NaN, 0.5555555820465088, 0.5555555820465088, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, NaN, 0.5555555820465088, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, NaN, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, NaN, 0.6666666865348816, 1.0, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, NaN, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, NaN, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, NaN, 0.3333333432674408, 0.5555555820465088, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.2222222238779068, 0.6666666865348816, NaN, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, NaN, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, NaN, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, NaN, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, NaN, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.5555555820465088, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.4444444477558136, NaN, 0.4444444477558136, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.5555555820465088, NaN, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, NaN, 0.5555555820465088, 0.6111111044883728, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, NaN, 0.4444444477558136, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.3888888955116272, 0.5555555820465088, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, NaN, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, NaN, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, NaN, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, NaN, 0.5555555820465088, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4444444477558136, NaN, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, NaN, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.3333333432674408, NaN, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, NaN, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, NaN, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.0, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, NaN, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, NaN, 0.7777777910232544, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, NaN, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":1.0},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.9802130902},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9741248101},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9619482497},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9391171999},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.9345509896},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9147640792},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.9117199395},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8782343995},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8493150703},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8493150698},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8401826496},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.8280060892},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.8173515994},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.8143074592},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.8112633212},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7701674286},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7640791491},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7625570794},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.7503805183},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.7458143081},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6818873689},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6773211586},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6727549482},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6697108084},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.6666666676},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6651445976},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.6590563181},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6255707767},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6225266365},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5829528162},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5692541863},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.5555555561},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5403348564},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5312024365},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5159817362},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.5083713861},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5038051758},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4946727554},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4596651452},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4459665154},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4216133954},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.4200913251},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4140030454},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4124809746},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.3911719948},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3729071543},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.3576864539},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3546423137},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.3272450534},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.3059360736},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2998477932},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.2952815834},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.2602739734},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2534246579},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2526636229},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2503805181},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.2207001522},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.1993911724},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.1856925423},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1811263321},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1659056322},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1461187219},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1293759522},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1248097416},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1171993914},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0882800614},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0761035013},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.0700152218},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0547945214},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.0517503809},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.0395738205},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.0136986301},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..c46e757463f155afa7e64c9397d3004d23e910e0
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,LimiX (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (tuned),TabFM (default),TabPFN-2.6 (default),TabICLv2 (default),EBM (tuned + ensembled),RandomForest (tuned + ensembled),LightGBM (tuned + ensembled),EBM (tuned),RandomForest (tuned),LightGBM (tuned),TabDPT-Turbo (default),XGBoost (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),XGBoost (default),XGBoost (tuned),iLTM (tuned + ensembled),CatBoost (default),RealTabPFN-2.5 (default),ChimeraBoost (tuned + ensembled),ChimeraBoost (tuned),ModernNCA (tuned + ensembled),EBM (default),ModernNCA (tuned),TabICL (default),ExtraTrees (tuned + ensembled),iLTM (tuned),iLTM (default),TabSwift (default),ChimeraBoost (default),TabDPT (tuned + ensembled),TorchMLP (tuned),TorchMLP (tuned + ensembled),ExtraTrees (tuned),ModernNCA (default),TabM (tuned + ensembled),TabDPT (tuned),TabM (tuned),SAP-RPT-OSS (default),TabDPT (default),RandomForest (default),RealMLP (tuned + ensembled),RealMLP (tuned),LightGBM (default),xRFM (tuned + ensembled),xRFM (tuned),TorchMLP (default),TabM (default),BetaTabPFN (default),TabSTAR (tuned),OrionMSP (default),TabSTAR (tuned + ensembled),RealMLP (default),FastaiMLP (tuned),FastaiMLP (tuned + ensembled),TabSTAR (default),PerpetualBooster (tuned + ensembled),ExtraTrees (default),PerpetualBooster (tuned),FastaiMLP (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),xRFM (default),PerpetualBooster (default),TabFlex (default),KNN (tuned + ensembled),KNN (tuned),KNN (default)
+LimiX (default),,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.0,,0.6666667,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.0,0.33333334,,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,0.11111111,0.22222222,,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.0,0.0,0.11111111,,0.6666667,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.0,0.0,0.0,0.33333334,,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.0,0.0,0.0,0.0,0.22222222,0.11111111,,0.44444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.5555556,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.7777778,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.44444445,,0.6666667,0.6666667,0.8888889,0.7777778,0.6666667,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.33333334,0.33333334,,0.5555556,0.5555556,0.8888889,0.44444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.33333334,0.33333334,0.44444445,,0.44444445,0.5555556,0.44444445,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.11111111,0.44444445,0.5555556,,0.44444445,0.44444445,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.22222222,0.11111111,0.44444445,0.5555556,,0.44444445,0.7777778,0.8888889,0.7777778,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.33333334,0.5555556,0.5555556,0.5555556,0.5555556,,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.33333334,,0.5555556,0.44444445,0.8888889,0.8888889,1.0,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.44444445,,0.6666667,0.5555556,0.5555556,0.8888889,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.22222222,0.33333334,0.5555556,0.33333334,,0.5555556,0.5555556,0.7777778,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.33333334,0.11111111,0.44444445,0.44444445,,0.6666667,1.0,1.0,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.33333334,0.11111111,0.44444445,0.44444445,0.33333334,,1.0,1.0,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.0,0.0,,0.5555556,0.6666667,0.5555556,0.6666667,0.5555556,0.6666667,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.44444445,,0.6666667,0.6666667,0.7777778,0.5555556,0.5555556,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.33333334,,0.33333334,0.33333334,0.33333334,0.44444445,0.6666667,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.44444445,0.33333334,0.6666667,,0.5555556,0.6666667,0.5555556,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.22222222,0.6666667,0.44444445,,0.7777778,0.44444445,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.44444445,0.44444445,0.6666667,0.33333334,0.22222222,,0.5555556,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.44444445,0.5555556,0.44444445,0.5555556,0.44444445,,0.7777778,0.7777778,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.33333334,0.11111111,0.11111111,0.0,0.22222222,,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICL (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.33333334,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,,1.0,0.6666667,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,,0.6666667,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.33333334,0.33333334,,0.44444445,0.6666667,0.8888889,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.5555556,,0.6666667,0.6666667,0.7777778,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.33333334,,0.6666667,0.5555556,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.33333334,,0.5555556,0.5555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.44444445,0.44444445,,0.5555556,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.11111111,0.22222222,0.44444445,0.44444445,,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,,0.6666667,1.0,0.7777778,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.33333334,,0.7777778,0.7777778,0.6666667,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,,0.5555556,0.7777778,0.6666667,0.5555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.22222222,0.22222222,0.44444445,,0.33333334,0.5555556,0.7777778,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.22222222,0.6666667,,0.5555556,0.6666667,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.33333334,0.44444445,0.44444445,,0.7777778,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.44444445,0.22222222,0.33333334,0.22222222,,0.6666667,0.7777778,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.11111111,0.33333334,,0.7777778,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.22222222,,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.22222222,0.22222222,0.33333334,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.5555556,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.44444445,,0.44444445,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.5555556,,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.6666667,0.6666667,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,,0.5555556,0.6111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.44444445,,0.44444445,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.3888889,0.5555556,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.8888889,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,,0.6666667,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.33333334,,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.22222222,0.22222222,,0.7777778,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,,0.5555556,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.44444445,,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.8888889,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.22222222,,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.22222222,0.33333334,,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,,0.8888889,0.6666667,0.7777778,1.0,1.0,1.0,1.0
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,,0.6666667,0.7777778,0.8888889,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.22222222,0.0,0.11111111,0.33333334,0.33333334,,0.5555556,0.6666667,0.7777778,1.0,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.22222222,0.44444445,,0.5555556,0.5555556,1.0,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.33333334,0.44444445,,0.7777778,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.44444445,0.22222222,,1.0,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
index aa8ccd2a535a8d544906669e32530a9f695b432d..b27079362ee5af5f1a456e3acdc6b83344c74ce7 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Multiclass | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/leaderboard_table.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..379bd8f0a488cc6fe25cf2ade0555715801affde
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_multiclass/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1742.0,"score":0.694,"rank":9.5,"hrank":5.73,"imp":12.917,"train_time":245.58,"predict_time":2.838,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-58","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1735.0,"score":0.708,"rank":9.76,"hrank":4.54,"imp":12.879,"train_time":2742.8,"predict_time":1.218,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+200\/-79","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1719.0,"score":0.693,"rank":10.41,"hrank":4.23,"imp":12.049,"train_time":2742.8,"predict_time":7.936,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+249\/-110","imputed":false},{"position":3,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1687.0,"score":0.836,"rank":11.86,"hrank":1.78,"imp":8.339,"train_time":64.25,"predict_time":6.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+450\/-246","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1654.0,"score":0.613,"rank":13.44,"hrank":7.99,"imp":15.458,"train_time":11.1,"predict_time":0.682,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+218\/-100","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1636.0,"score":0.593,"rank":14.37,"hrank":5.8,"imp":13.181,"train_time":11.21,"predict_time":0.845,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+278\/-155","imputed":false},{"position":6,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1598.0,"score":0.503,"rank":16.48,"hrank":8.46,"imp":15.828,"train_time":4471.59,"predict_time":21.354,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+196\/-120","imputed":false},{"position":7,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1588.0,"score":0.523,"rank":17.07,"hrank":7.72,"imp":14.902,"train_time":12.32,"predict_time":0.524,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+210\/-143","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1573.0,"score":0.481,"rank":18.0,"hrank":9.96,"imp":16.802,"train_time":4471.59,"predict_time":3.755,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-60","imputed":false},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1567.0,"score":0.549,"rank":18.35,"hrank":6.42,"imp":16.229,"train_time":6.2,"predict_time":0.211,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+269\/-186","imputed":false},{"position":10,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1560.0,"score":0.474,"rank":18.81,"hrank":9.58,"imp":15.664,"train_time":46.82,"predict_time":7.43,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+173\/-129","imputed":false},{"position":11,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1532.0,"score":0.443,"rank":20.59,"hrank":11.46,"imp":19.553,"train_time":3919.29,"predict_time":4.506,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-80","imputed":false},{"position":12,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1530.0,"score":0.334,"rank":20.73,"hrank":10.85,"imp":19.505,"train_time":6549.61,"predict_time":2.898,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+198\/-85","imputed":false},{"position":13,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1489.0,"score":0.389,"rank":23.5,"hrank":13.39,"imp":20.324,"train_time":3919.29,"predict_time":0.749,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-76","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1478.0,"score":0.325,"rank":24.28,"hrank":18.65,"imp":20.421,"train_time":1502.45,"predict_time":6.327,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+155\/-88","imputed":false},{"position":15,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1476.0,"score":0.317,"rank":24.38,"hrank":20.37,"imp":19.458,"train_time":3415.61,"predict_time":1.368,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-44","imputed":false},{"position":16,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1471.0,"score":0.402,"rank":24.76,"hrank":12.35,"imp":22.219,"train_time":4631.38,"predict_time":190.548,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+189\/-161","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1469.0,"score":0.309,"rank":24.94,"hrank":19.93,"imp":20.619,"train_time":1502.45,"predict_time":1.367,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-73","imputed":false},{"position":18,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1456.0,"score":0.351,"rank":25.81,"hrank":18.65,"imp":21.263,"train_time":684.78,"predict_time":4.813,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-100","imputed":false},{"position":19,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1450.0,"score":0.266,"rank":26.29,"hrank":21.52,"imp":20.165,"train_time":3415.61,"predict_time":0.217,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-34","imputed":false},{"position":20,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1444.0,"score":0.311,"rank":26.73,"hrank":16.42,"imp":21.605,"train_time":2097.62,"predict_time":3.141,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-142","imputed":false},{"position":21,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1442.0,"score":0.319,"rank":26.84,"hrank":10.46,"imp":22.29,"train_time":6.75,"predict_time":0.264,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+220\/-170","imputed":false},{"position":22,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1433.0,"score":0.341,"rank":27.52,"hrank":13.58,"imp":23.231,"train_time":4631.38,"predict_time":30.351,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+204\/-168","imputed":false},{"position":23,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1429.0,"score":0.239,"rank":27.81,"hrank":20.3,"imp":20.959,"train_time":2551.65,"predict_time":1.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-71","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1424.0,"score":0.215,"rank":28.18,"hrank":13.6,"imp":20.771,"train_time":7359.32,"predict_time":0.596,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-59","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1424.0,"score":0.199,"rank":28.21,"hrank":20.01,"imp":20.604,"train_time":7359.32,"predict_time":9.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-95","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1420.0,"score":0.233,"rank":28.49,"hrank":21.62,"imp":21.107,"train_time":2551.65,"predict_time":4.526,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-69","imputed":false},{"position":27,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1419.0,"score":0.302,"rank":28.51,"hrank":17.09,"imp":22.163,"train_time":13.71,"predict_time":0.733,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+155\/-110","imputed":false},{"position":28,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1413.0,"score":0.177,"rank":28.96,"hrank":22.88,"imp":21.5,"train_time":34415.89,"predict_time":1003.647,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+117\/-78","imputed":false},{"position":29,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1387.0,"score":0.413,"rank":30.9,"hrank":6.39,"imp":20.772,"train_time":22.56,"predict_time":1.484,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+290\/-242","imputed":false},{"position":30,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1362.0,"score":0.193,"rank":32.83,"hrank":27.88,"imp":23.265,"train_time":45754.93,"predict_time":0.315,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-84","imputed":false},{"position":31,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1361.0,"score":0.263,"rank":32.9,"hrank":14.6,"imp":24.956,"train_time":1076.45,"predict_time":1.541,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+208\/-181","imputed":false},{"position":32,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1357.0,"score":0.163,"rank":33.23,"hrank":28.13,"imp":22.658,"train_time":29.23,"predict_time":0.249,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-67","imputed":false},{"position":33,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1343.0,"score":0.145,"rank":34.27,"hrank":25.92,"imp":23.571,"train_time":4142.9,"predict_time":2.104,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+155\/-118","imputed":false},{"position":34,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1339.0,"score":0.2,"rank":34.63,"hrank":17.89,"imp":23.831,"train_time":2097.62,"predict_time":0.514,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-110","imputed":false},{"position":35,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1338.0,"score":0.238,"rank":34.7,"hrank":13.44,"imp":24.155,"train_time":2775.63,"predict_time":12.586,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+225\/-179","imputed":false},{"position":36,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1317.0,"score":0.109,"rank":36.28,"hrank":31.0,"imp":22.85,"train_time":34415.89,"predict_time":299.32,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+94\/-95","imputed":false},{"position":37,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1316.0,"score":0.241,"rank":36.35,"hrank":14.2,"imp":25.618,"train_time":755.27,"predict_time":2.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+200\/-205","imputed":false},{"position":38,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1316.0,"score":0.132,"rank":36.4,"hrank":30.84,"imp":24.034,"train_time":45754.93,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-84","imputed":false},{"position":39,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1312.0,"score":0.231,"rank":36.66,"hrank":12.6,"imp":26.387,"train_time":1076.45,"predict_time":0.31,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+202\/-199","imputed":false},{"position":40,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1302.0,"score":0.19,"rank":37.44,"hrank":14.69,"imp":26.034,"train_time":46.25,"predict_time":45.719,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+229\/-193","imputed":false},{"position":41,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1301.0,"score":0.125,"rank":37.52,"hrank":27.9,"imp":25.333,"train_time":15.52,"predict_time":0.045,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+138\/-122","imputed":false},{"position":42,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1296.0,"score":0.144,"rank":37.9,"hrank":31.0,"imp":26.281,"train_time":5.04,"predict_time":0.071,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+116\/-144","imputed":false},{"position":43,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1283.0,"score":0.162,"rank":38.93,"hrank":28.42,"imp":24.851,"train_time":5.32,"predict_time":0.43,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+168\/-167","imputed":false},{"position":44,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1278.0,"score":0.125,"rank":39.27,"hrank":22.39,"imp":25.468,"train_time":2775.63,"predict_time":1.431,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+195\/-166","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1273.0,"score":0.123,"rank":39.68,"hrank":27.91,"imp":25.94,"train_time":4142.9,"predict_time":0.287,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-148","imputed":false},{"position":46,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1263.0,"score":0.186,"rank":40.4,"hrank":11.41,"imp":27.015,"train_time":755.27,"predict_time":0.317,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+197\/-224","imputed":false},{"position":47,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1261.0,"score":0.138,"rank":40.56,"hrank":31.21,"imp":25.342,"train_time":2032.73,"predict_time":0.268,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-182","imputed":false},{"position":48,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1253.0,"score":0.001,"rank":41.16,"hrank":39.3,"imp":24.534,"train_time":5.48,"predict_time":0.599,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-35","imputed":false},{"position":49,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1251.0,"score":0.127,"rank":41.29,"hrank":35.3,"imp":25.411,"train_time":2032.73,"predict_time":1.367,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+149\/-163","imputed":false},{"position":50,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1223.0,"score":0.025,"rank":43.33,"hrank":34.95,"imp":28.657,"train_time":20.96,"predict_time":0.482,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-126","imputed":false},{"position":51,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1211.0,"score":0.0,"rank":44.2,"hrank":40.45,"imp":27.073,"train_time":19.64,"predict_time":7.682,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-79","imputed":false},{"position":52,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1171.0,"score":0.122,"rank":46.99,"hrank":35.99,"imp":28.3,"train_time":8.45,"predict_time":0.239,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+181\/-233","imputed":false},{"position":53,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1163.0,"score":0.065,"rank":47.54,"hrank":41.05,"imp":28.555,"train_time":10.84,"predict_time":2.189,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-162","imputed":false},{"position":54,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1099.0,"score":0.0,"rank":51.69,"hrank":38.38,"imp":30.981,"train_time":14.75,"predict_time":0.337,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-132","imputed":false},{"position":55,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1080.0,"score":0.072,"rank":52.85,"hrank":24.65,"imp":38.859,"train_time":173.46,"predict_time":1.637,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+213\/-318","imputed":false},{"position":56,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1042.0,"score":0.0,"rank":55.02,"hrank":51.32,"imp":34.323,"train_time":22.95,"predict_time":3.32,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-151","imputed":false},{"position":57,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1042.0,"score":0.0,"rank":55.05,"hrank":51.37,"imp":40.594,"train_time":905.41,"predict_time":0.875,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-217","imputed":false},{"position":58,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1037.0,"score":0.0,"rank":55.31,"hrank":54.11,"imp":30.78,"train_time":581.49,"predict_time":277.193,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-148","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1036.0,"score":0.0,"rank":55.38,"hrank":49.09,"imp":32.535,"train_time":42561.15,"predict_time":35.622,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-190","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1029.0,"score":0.0,"rank":55.77,"hrank":51.28,"imp":32.668,"train_time":42561.15,"predict_time":9.171,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-184","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1011.0,"score":0.0,"rank":56.74,"hrank":53.0,"imp":41.124,"train_time":905.41,"predict_time":0.175,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-233","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":57.32,"hrank":52.61,"imp":39.533,"train_time":0.85,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-229","imputed":false},{"position":63,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":993.0,"score":0.0,"rank":57.67,"hrank":54.72,"imp":37.448,"train_time":5.99,"predict_time":1.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-224","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":989.0,"score":0.005,"rank":57.9,"hrank":54.54,"imp":34.303,"train_time":8.97,"predict_time":0.742,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+154\/-186","imputed":false},{"position":65,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":931.0,"score":0.0,"rank":60.65,"hrank":52.43,"imp":52.179,"train_time":1838.94,"predict_time":13.381,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-220","imputed":false},{"position":66,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":918.0,"score":0.0,"rank":61.23,"hrank":60.76,"imp":39.053,"train_time":5.83,"predict_time":1.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+46\/-101","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":908.0,"score":0.059,"rank":61.68,"hrank":43.54,"imp":48.155,"train_time":173.46,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+262\/-398","imputed":false},{"position":68,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":884.0,"score":0.0,"rank":62.65,"hrank":37.01,"imp":53.742,"train_time":1838.94,"predict_time":5.572,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-245","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":873.0,"score":0.0,"rank":63.09,"hrank":30.79,"imp":57.658,"train_time":295.36,"predict_time":0.573,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-274","imputed":false},{"position":70,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":866.0,"score":0.0,"rank":63.34,"hrank":61.26,"imp":41.357,"train_time":801.38,"predict_time":8.877,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-232","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":799.0,"score":0.0,"rank":65.63,"hrank":64.87,"imp":47.751,"train_time":4.72,"predict_time":0.188,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-257","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":412.0,"score":0.0,"rank":72.09,"hrank":72.06,"imp":69.874,"train_time":0.57,"predict_time":0.067,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-258","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/pareto_front_explorer.html
index 8ba13b411861d1aab3070e478051b58c01c3aa4a..ed91ca24bc9e7ded6a47eb3798bac474e3a88f86 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.8133362736,"imp":21.2627338561,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2489310712,"imp":22.658420467,"elo":1357.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.216888627,"imp":20.1647271168,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.3675697397,"imp":19.4575093633,"elo":1476.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2385689265,"imp":28.2998767417,"elo":1170.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2679377135,"imp":25.3420969551,"elo":1261.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3667157604,"imp":25.4108405462,"elo":1251.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0454217787,"imp":25.3331666729,"elo":1301.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0696988938,"imp":24.0342028196,"elo":1315.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.3149796315,"imp":23.2647626735,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0094189939,"imp":37.4480768298,"elo":993.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3100216521,"imp":26.387385455,"elo":1312.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5409284286,"imp":24.9560052221,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7424241286,"imp":34.3030793636,"elo":988.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4309052415,"imp":25.4679199995,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5856962357,"imp":24.1547502772,"elo":1337.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.430301758,"imp":24.8509507405,"elo":1282.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.3668411392,"imp":20.6193627212,"elo":1468.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":6.3267197862,"imp":20.4213274604,"elo":1477.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.4299917889,"imp":15.6640448028,"elo":1560.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4819080747,"imp":28.656548258,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5959001895,"imp":20.7713476327,"elo":1423.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.0233407175,"imp":20.6040390081,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3196375471,"imp":34.3229212501,"elo":1042.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.573021293,"imp":57.6579202274,"elo":872.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":5.5720509293,"imp":53.7418700948,"elo":884.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":13.3808302986,"imp":52.1785427274,"elo":931.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.14819125,"imp":39.5332063691,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3169111952,"imp":27.014866581,"elo":1263.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":2.1547385269,"imp":25.6182614884,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.6818049083,"imp":27.0730576777,"elo":1211.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.7554737985,"imp":16.8022756243,"elo":1573.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.354012373,"imp":15.828070484,"elo":1598.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8448151087,"imp":13.1809248242,"elo":1636.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2178786994,"imp":12.8791895585,"elo":1735.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.9362255959,"imp":12.0494799975,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.4841490714,"imp":20.7721618586,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.7189326692,"imp":26.0341145838,"elo":1302.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":30.3506797823,"imp":23.2313793813,"elo":1432.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":190.5482422391,"imp":22.218514128,"elo":1471.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2641860705,"imp":22.2902197192,"elo":1442.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4463921345,"imp":8.339148626,"elo":1686.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":1.1673847834,"imp":39.0527657022,"elo":918.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2109379604,"imp":16.2292764588,"elo":1567.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.732957515,"imp":22.1633522311,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7493943858,"imp":20.3244865791,"elo":1489.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.5061538419,"imp":19.5531088323,"elo":1532.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6817692836,"imp":15.4584523557,"elo":1654.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5239137837,"imp":14.901966457,"elo":1588.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.8772316642,"imp":41.3568722011,"elo":866.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.1707473319,"imp":32.6678176503,"elo":1028.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":35.6221458071,"imp":32.5345092305,"elo":1036.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0709157018,"imp":26.2814256533,"elo":1296.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3365764797,"imp":30.9811244639,"elo":1098.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2873908265,"imp":25.9399671005,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1038316976,"imp":23.5711587338,"elo":1343.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.599155634,"imp":24.534248462,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0214870028,"imp":20.9589545446,"elo":1428.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":4.5260005112,"imp":21.1072259744,"elo":1419.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":277.1925819839,"imp":30.7796202068,"elo":1037.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":299.3204233014,"imp":22.8495954696,"elo":1317.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1003.6468494405,"imp":21.5003907404,"elo":1413.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":2.1887529296,"imp":28.5554104657,"elo":1162.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.5143737594,"imp":23.8314302839,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.1408224571,"imp":21.6054810211,"elo":1443.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.8133362736,"x_train":684.7823129486,"imp":21.2627338561,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2489310712,"x_train":29.2262634158,"imp":22.658420467,"elo":1357.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.216888627,"x_train":3415.6125164831,"imp":20.1647271168,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.3675697397,"x_train":3415.6125164831,"imp":19.4575093633,"elo":1476.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2385689265,"x_train":8.4538911064,"imp":28.2998767417,"elo":1170.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2679377135,"x_train":2032.7286825124,"imp":25.3420969551,"elo":1261.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3667157604,"x_train":2032.7286825124,"imp":25.4108405462,"elo":1251.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0454217787,"x_train":15.5201326578,"imp":25.3331666729,"elo":1301.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0696988938,"x_train":45754.9282597275,"imp":24.0342028196,"elo":1315.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.3149796315,"x_train":45754.9282597275,"imp":23.2647626735,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0094189939,"x_train":5.9912431514,"imp":37.4480768298,"elo":993.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3100216521,"x_train":1076.4479804348,"imp":26.387385455,"elo":1312.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5409284286,"x_train":1076.4479804348,"imp":24.9560052221,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7424241286,"x_train":8.9709107352,"imp":34.3030793636,"elo":988.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4309052415,"x_train":2775.6292399688,"imp":25.4679199995,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5856962357,"x_train":2775.6292399688,"imp":24.1547502772,"elo":1337.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.430301758,"x_train":5.3219291517,"imp":24.8509507405,"elo":1282.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.3668411392,"x_train":1502.4464127054,"imp":20.6193627212,"elo":1468.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":6.3267197862,"x_train":1502.4464127054,"imp":20.4213274604,"elo":1477.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.4299917889,"x_train":46.8234528507,"imp":15.6640448028,"elo":1560.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4819080747,"x_train":20.9645091386,"imp":28.656548258,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5959001895,"x_train":7359.3230736405,"imp":20.7713476327,"elo":1423.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.0233407175,"x_train":7359.3230736405,"imp":20.6040390081,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3196375471,"x_train":22.9524183555,"imp":34.3229212501,"elo":1042.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.573021293,"x_train":295.3582726767,"imp":57.6579202274,"elo":872.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":5.5720509293,"x_train":1838.9438886021,"imp":53.7418700948,"elo":884.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":13.3808302986,"x_train":1838.9438886021,"imp":52.1785427274,"elo":931.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.14819125,"x_train":0.8546409522,"imp":39.5332063691,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3169111952,"x_train":755.269328828,"imp":27.014866581,"elo":1263.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":2.1547385269,"x_train":755.269328828,"imp":25.6182614884,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.6818049083,"x_train":19.6357631766,"imp":27.0730576777,"elo":1211.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.7554737985,"x_train":4471.5866070717,"imp":16.8022756243,"elo":1573.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.354012373,"x_train":4471.5866070717,"imp":15.828070484,"elo":1598.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8448151087,"x_train":11.2142729689,"imp":13.1809248242,"elo":1636.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2178786994,"x_train":2742.7953513314,"imp":12.8791895585,"elo":1735.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.9362255959,"x_train":2742.7953513314,"imp":12.0494799975,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.4841490714,"x_train":22.5634771017,"imp":20.7721618586,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.7189326692,"x_train":46.2517520732,"imp":26.0341145838,"elo":1302.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":30.3506797823,"x_train":4631.3753499953,"imp":23.2313793813,"elo":1432.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":190.5482422391,"x_train":4631.3753499953,"imp":22.218514128,"elo":1471.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2641860705,"x_train":6.7467122095,"imp":22.2902197192,"elo":1442.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4463921345,"x_train":64.2501249469,"imp":8.339148626,"elo":1686.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":1.1673847834,"x_train":5.8322784268,"imp":39.0527657022,"elo":918.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2109379604,"x_train":6.2039007574,"imp":16.2292764588,"elo":1567.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.732957515,"x_train":13.7118099506,"imp":22.1633522311,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7493943858,"x_train":3919.285334302,"imp":20.3244865791,"elo":1489.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.5061538419,"x_train":3919.285334302,"imp":19.5531088323,"elo":1532.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6817692836,"x_train":11.1028664446,"imp":15.4584523557,"elo":1654.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5239137837,"x_train":12.3203746583,"imp":14.901966457,"elo":1588.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.8772316642,"x_train":801.375360716,"imp":41.3568722011,"elo":866.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.1707473319,"x_train":42561.154121861,"imp":32.6678176503,"elo":1028.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":35.6221458071,"x_train":42561.154121861,"imp":32.5345092305,"elo":1036.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0709157018,"x_train":5.0404725813,"imp":26.2814256533,"elo":1296.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3365764797,"x_train":14.7477895176,"imp":30.9811244639,"elo":1098.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2873908265,"x_train":4142.8989350983,"imp":25.9399671005,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1038316976,"x_train":4142.8989350983,"imp":23.5711587338,"elo":1343.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.599155634,"x_train":5.4806602189,"imp":24.534248462,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0214870028,"x_train":2551.6500015348,"imp":20.9589545446,"elo":1428.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":4.5260005112,"x_train":2551.6500015348,"imp":21.1072259744,"elo":1419.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":277.1925819839,"x_train":581.4914553919,"imp":30.7796202068,"elo":1037.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":299.3204233014,"x_train":34415.8855588805,"imp":22.8495954696,"elo":1317.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1003.6468494405,"x_train":34415.8855588805,"imp":21.5003907404,"elo":1413.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":2.1887529296,"x_train":10.8402948127,"imp":28.5554104657,"elo":1162.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.5143737594,"x_train":2097.6180059154,"imp":23.8314302839,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.1408224571,"x_train":2097.6180059154,"imp":21.6054810211,"elo":1443.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
index 55dd20aef98637ef2408569614b3ed8c7e02c32b..a6529ac8e17453e18c9f3f459b5b7e2dfe948a5e 100644
--- a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":684.7823129486,"imp":21.2963132247,"elo":1463.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":29.2262634158,"imp":22.690521439,"elo":1360.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":61.1709820217,"imp":20.922656403,"elo":1427.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":144.6176954198,"imp":20.2275595064,"elo":1461.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":300.2328281021,"imp":19.8942854098,"elo":1475.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":478.2713322387,"imp":19.5956275112,"elo":1488.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":841.9181732231,"imp":19.5877931815,"elo":1488.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1705.1687538882,"imp":19.5271936216,"elo":1489.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2553.0942626612,"imp":19.526471251,"elo":1489.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":3415.6125164831,"imp":19.4909624553,"elo":1488.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":8.4538911064,"imp":28.3324627128,"elo":1166.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":18.6841630733,"imp":27.929436912,"elo":1183.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":48.4911153287,"imp":27.6773970873,"elo":1190.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":98.5757062009,"imp":27.2696986477,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":250.870284713,"imp":26.4781697423,"elo":1221.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":503.8763822777,"imp":26.1380722632,"elo":1231.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1010.9316172533,"imp":25.7034260384,"elo":1242.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1516.2919521693,"imp":25.536858347,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2032.7286825124,"imp":25.4448223842,"elo":1252.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":15.5201326578,"imp":25.3655997897,"elo":1299.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":263.0355881156,"imp":25.0789709103,"elo":1305.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":710.2172146253,"imp":24.4802361952,"elo":1324.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":1974.8947417325,"imp":24.2517252099,"elo":1332.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":5239.4557597159,"imp":23.8876576939,"elo":1342.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":11351.8958788799,"imp":23.638596102,"elo":1352.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":22821.1893610533,"imp":23.447321447,"elo":1359.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":34610.401754972,"imp":23.3579755037,"elo":1364.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":45754.9282597275,"imp":23.2984809576,"elo":1366.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":5.9912431514,"imp":37.4714125814,"elo":986.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":11.3518702812,"imp":33.5047638938,"elo":1069.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":27.5426636663,"imp":28.0869278966,"elo":1187.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":54.3053391392,"imp":26.7369777398,"elo":1269.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":134.5396659696,"imp":25.5737824705,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":268.4042300737,"imp":25.3759026699,"elo":1355.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":535.5515141355,"imp":25.1778203718,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":803.2668643965,"imp":25.0399520119,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1076.4479804348,"imp":24.9803343916,"elo":1372.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.9709107352,"imp":34.3351904648,"elo":988.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":23.3777616357,"imp":32.1591969352,"elo":1055.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":65.8044502368,"imp":28.4276141666,"elo":1176.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":132.7278242503,"imp":26.9415570526,"elo":1239.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":338.4412410661,"imp":25.275214494,"elo":1303.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":684.2594302606,"imp":24.6034177755,"elo":1333.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1371.7911574391,"imp":24.3306376351,"elo":1346.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2064.6207054809,"imp":24.2378994204,"elo":1350.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2775.6292399688,"imp":24.1856836795,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":5.3219291517,"imp":24.8839842282,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":12.9118030008,"imp":23.5924770778,"elo":1346.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":34.395160853,"imp":22.6532439713,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":71.4857515779,"imp":22.0678979581,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":185.2078470973,"imp":21.4068442066,"elo":1449.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":372.6794951088,"imp":21.1133629018,"elo":1463.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":746.9586248726,"imp":20.747233898,"elo":1481.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":1118.5590552235,"imp":20.6099591293,"elo":1486.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1502.4464127054,"imp":20.4530524682,"elo":1493.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":46.8234528507,"imp":15.7430754632,"elo":1576.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":20.9645091386,"imp":28.6864247729,"elo":1223.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":58.5698168814,"imp":26.0236695591,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":173.2669344789,"imp":22.9257353735,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":350.5420088728,"imp":22.1677984419,"elo":1399.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":860.8181117489,"imp":21.3404180744,"elo":1434.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1818.8588293534,"imp":21.0520693803,"elo":1440.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3621.9510865846,"imp":21.0567035032,"elo":1437.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":5444.0858786703,"imp":20.7873607964,"elo":1435.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":7359.3230736405,"imp":20.63675085,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":22.9524183555,"imp":34.342204873,"elo":1037.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":295.3582726767,"imp":57.7314825598,"elo":873.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":606.9204093001,"imp":56.666864393,"elo":889.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1533.0305097908,"imp":52.6430268225,"elo":922.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1838.9438886021,"imp":52.1907830972,"elo":930.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.8546409522,"imp":39.617439673,"elo":997.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.9583883142,"imp":32.3840101169,"elo":1088.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":16.3951125276,"imp":28.8960747348,"elo":1151.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":35.2860398048,"imp":27.1688368768,"elo":1209.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":91.6231833671,"imp":26.1011115176,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":185.806167626,"imp":25.9042044325,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":374.1216863756,"imp":25.7657519855,"elo":1321.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":562.8425864643,"imp":25.7028972044,"elo":1322.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":755.269328828,"imp":25.6430947111,"elo":1323.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":19.6357631766,"imp":27.1075325092,"elo":1210.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":44.7978169276,"imp":21.9257565449,"elo":1386.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":111.6196798565,"imp":19.0007646816,"elo":1503.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":222.0642460075,"imp":17.9998521783,"elo":1545.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":554.5469212766,"imp":17.1976476447,"elo":1581.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1115.1182819708,"imp":16.720750071,"elo":1604.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2225.9959788897,"imp":16.3455737417,"elo":1614.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3338.8866547043,"imp":16.08769155,"elo":1622.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4471.5866070717,"imp":15.8630925537,"elo":1623.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":11.2142729689,"imp":13.2180436427,"elo":1644.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":25.4079172215,"imp":13.4498319544,"elo":1651.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":64.4678080693,"imp":13.2472265519,"elo":1679.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":130.8710194046,"imp":12.9687522021,"elo":1739.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":334.8480217005,"imp":12.8574688522,"elo":1748.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":675.6180238549,"imp":12.6447608401,"elo":1747.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1365.6295149721,"imp":12.5754112988,"elo":1740.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2047.0350964005,"imp":12.4408791211,"elo":1737.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2742.7953513314,"imp":12.0897828422,"elo":1732.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":22.5634771017,"imp":20.8052111165,"elo":1392.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.2517520732,"imp":26.0563711418,"elo":1303.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":64.0415875229,"imp":25.6881700958,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":123.7092045238,"imp":24.94122474,"elo":1368.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":248.3053487257,"imp":24.4170391636,"elo":1400.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":584.7096529094,"imp":23.6189596719,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1166.0242045586,"imp":23.0695461071,"elo":1452.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2315.5296763529,"imp":22.6422524939,"elo":1473.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3443.7150532739,"imp":22.4367895489,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4631.3753499953,"imp":22.2451690194,"elo":1485.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":6.7467122095,"imp":22.3158230071,"elo":1452.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":64.2501249469,"imp":8.3745097291,"elo":1689.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":5.8322784268,"imp":39.0667940068,"elo":913.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":6.2039007574,"imp":16.2659425384,"elo":1578.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":13.7118099506,"imp":22.1989123677,"elo":1422.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":31.4151026664,"imp":21.738020561,"elo":1450.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":93.9419399758,"imp":21.111243279,"elo":1481.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":192.8564173001,"imp":20.6548170618,"elo":1502.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":479.3804315766,"imp":20.2582631076,"elo":1519.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":961.3391911829,"imp":19.9502974945,"elo":1536.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1926.418861227,"imp":19.7538496162,"elo":1543.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2910.0650077596,"imp":19.6747107631,"elo":1545.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3919.285334302,"imp":19.5876203028,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":11.1028664446,"imp":15.4951831367,"elo":1674.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":12.3203746583,"imp":14.9362242891,"elo":1601.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":801.375360716,"imp":41.3865560038,"elo":855.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1026.0399950622,"imp":41.1430700241,"elo":867.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1364.8029707909,"imp":40.6205798122,"elo":882.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2058.9341404243,"imp":39.5352430188,"elo":908.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5065.9505852434,"imp":37.2194162392,"elo":947.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":10436.7196251803,"imp":34.8894234924,"elo":982.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":21636.9871111389,"imp":33.639618858,"elo":1010.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":32013.360321672,"imp":33.0093639821,"elo":1030.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":42561.154121861,"imp":32.5664076078,"elo":1046.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":5.0404725813,"imp":26.3074223481,"elo":1300.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.7477895176,"imp":31.0113377824,"elo":1096.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":31.2590378107,"imp":28.5143287813,"elo":1170.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":101.8350811731,"imp":26.4972964922,"elo":1241.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":218.6755099117,"imp":25.3907953233,"elo":1295.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":533.9506807148,"imp":24.540745302,"elo":1322.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1017.6084144692,"imp":24.1150987447,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1968.6547247706,"imp":23.8565194941,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3052.495945877,"imp":23.6954730671,"elo":1349.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4142.8989350983,"imp":23.6068686261,"elo":1352.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":5.4806602189,"imp":24.5657052646,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":18.5840857617,"imp":23.2548839265,"elo":1316.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":57.5257824541,"imp":22.7739284558,"elo":1344.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":117.1284034437,"imp":22.2997595757,"elo":1367.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":309.9745400782,"imp":21.9370489498,"elo":1388.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":629.246307752,"imp":21.6824794187,"elo":1401.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":1266.4892033336,"imp":21.4789767835,"elo":1413.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":1899.605994117,"imp":21.3250522743,"elo":1422.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":2551.6500015348,"imp":21.1407437939,"elo":1433.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":581.4914553919,"imp":30.8091754666,"elo":1028.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1918.622222682,"imp":27.8960597479,"elo":1121.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":5770.5281133302,"imp":24.9152997253,"elo":1264.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":12892.6565755789,"imp":23.5116264395,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":32775.6937551607,"imp":21.5724840444,"elo":1422.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":34415.8855588805,"imp":21.534818294,"elo":1426.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":10.8402948127,"imp":28.5832258246,"elo":1159.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":22.047469012,"imp":26.7033859005,"elo":1235.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":52.9985363102,"imp":24.55825223,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":105.4364802264,"imp":23.6556190512,"elo":1377.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":261.4235869814,"imp":22.5736423267,"elo":1418.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":521.852845102,"imp":22.0943565016,"elo":1439.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":1043.5013351397,"imp":21.7727076399,"elo":1452.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1565.3955882162,"imp":21.6898068619,"elo":1455.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2097.6180059154,"imp":21.637128979,"elo":1457.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.8133362736,"x_train":684.7823129486,"imp":21.2963132247,"elo":1463.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2489310712,"x_train":29.2262634158,"imp":22.690521439,"elo":1360.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3979294404,"x_train":61.1709820217,"imp":20.922656403,"elo":1427.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.602962146,"x_train":144.6176954198,"imp":20.2275595064,"elo":1461.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.7875363724,"x_train":300.2328281021,"imp":19.8942854098,"elo":1475.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.0219051485,"x_train":478.2713322387,"imp":19.5956275112,"elo":1488.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1647414856,"x_train":841.9181732231,"imp":19.5877931815,"elo":1488.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2641468505,"x_train":1705.1687538882,"imp":19.5271936216,"elo":1489.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.3046821467,"x_train":2553.0942626612,"imp":19.526471251,"elo":1489.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.3675697397,"x_train":3415.6125164831,"imp":19.4909624553,"elo":1488.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2385689265,"x_train":8.4538911064,"imp":28.3324627128,"elo":1166.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3630820441,"x_train":18.6841630733,"imp":27.929436912,"elo":1183.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4757621068,"x_train":48.4911153287,"imp":27.6773970873,"elo":1190.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.6044354328,"x_train":98.5757062009,"imp":27.2696986477,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.7940641756,"x_train":250.870284713,"imp":26.4781697423,"elo":1221.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.9459377238,"x_train":503.8763822777,"imp":26.1380722632,"elo":1231.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1537015916,"x_train":1010.9316172533,"imp":25.7034260384,"elo":1242.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2872609782,"x_train":1516.2919521693,"imp":25.536858347,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3667157604,"x_train":2032.7286825124,"imp":25.4448223842,"elo":1252.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0454217787,"x_train":15.5201326578,"imp":25.3655997897,"elo":1299.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.1056375523,"x_train":263.0355881156,"imp":25.0789709103,"elo":1305.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.1472799815,"x_train":710.2172146253,"imp":24.4802361952,"elo":1324.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.1808874215,"x_train":1974.8947417325,"imp":24.2517252099,"elo":1332.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.250317583,"x_train":5239.4557597159,"imp":23.8876576939,"elo":1342.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.2758121216,"x_train":11351.8958788799,"imp":23.638596102,"elo":1352.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.3045866294,"x_train":22821.1893610533,"imp":23.447321447,"elo":1359.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.3279012754,"x_train":34610.401754972,"imp":23.3579755037,"elo":1364.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.3149796315,"x_train":45754.9282597275,"imp":23.2984809576,"elo":1366.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0094189939,"x_train":5.9912431514,"imp":37.4714125814,"elo":986.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.315155335,"x_train":11.3518702812,"imp":33.5047638938,"elo":1069.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3571607009,"x_train":27.5426636663,"imp":28.0869278966,"elo":1187.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3746513091,"x_train":54.3053391392,"imp":26.7369777398,"elo":1269.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4170856465,"x_train":134.5396659696,"imp":25.5737824705,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4416512416,"x_train":268.4042300737,"imp":25.3759026699,"elo":1355.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4692731875,"x_train":535.5515141355,"imp":25.1778203718,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5288396415,"x_train":803.2668643965,"imp":25.0399520119,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5409284286,"x_train":1076.4479804348,"imp":24.9803343916,"elo":1372.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7424241286,"x_train":8.9709107352,"imp":34.3351904648,"elo":988.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7463780049,"x_train":23.3777616357,"imp":32.1591969352,"elo":1055.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.6779868038,"x_train":65.8044502368,"imp":28.4276141666,"elo":1176.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0030791892,"x_train":132.7278242503,"imp":26.9415570526,"elo":1239.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.8189974652,"x_train":338.4412410661,"imp":25.275214494,"elo":1303.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.1925971466,"x_train":684.2594302606,"imp":24.6034177755,"elo":1333.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":10.1501412354,"x_train":1371.7911574391,"imp":24.3306376351,"elo":1346.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":11.6323451448,"x_train":2064.6207054809,"imp":24.2378994204,"elo":1350.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5427574154,"x_train":2775.6292399688,"imp":24.1856836795,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.430301758,"x_train":5.3219291517,"imp":24.8839842282,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6903836077,"x_train":12.9118030008,"imp":23.5924770778,"elo":1346.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.5870394193,"x_train":34.395160853,"imp":22.6532439713,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.7196389123,"x_train":71.4857515779,"imp":22.0678979581,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.1691382103,"x_train":185.2078470973,"imp":21.4068442066,"elo":1449.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.8740373686,"x_train":372.6794951088,"imp":21.1133629018,"elo":1463.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.1746943313,"x_train":746.9586248726,"imp":20.747233898,"elo":1481.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.9914919439,"x_train":1118.5590552235,"imp":20.6099591293,"elo":1486.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.2770589207,"x_train":1502.4464127054,"imp":20.4530524682,"elo":1493.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":7.4299917889,"x_train":46.8234528507,"imp":15.7430754632,"elo":1576.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4819080747,"x_train":20.9645091386,"imp":28.6864247729,"elo":1223.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9993489222,"x_train":58.5698168814,"imp":26.0236695591,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.0885736068,"x_train":173.2669344789,"imp":22.9257353735,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9464963206,"x_train":350.5420088728,"imp":22.1677984419,"elo":1399.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.4544013304,"x_train":860.8181117489,"imp":21.3404180744,"elo":1434.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.770076796,"x_train":1818.8588293534,"imp":21.0520693803,"elo":1440.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.3813327083,"x_train":3621.9510865846,"imp":21.0567035032,"elo":1437.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.6018084824,"x_train":5444.0858786703,"imp":20.7873607964,"elo":1435.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.0233407175,"x_train":7359.3230736405,"imp":20.63675085,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3196375471,"x_train":22.9524183555,"imp":34.342204873,"elo":1037.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.573021293,"x_train":295.3582726767,"imp":57.7314825598,"elo":873.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":5.6054982404,"x_train":606.9204093001,"imp":56.666864393,"elo":889.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":12.7962153938,"x_train":1533.0305097908,"imp":52.6430268225,"elo":922.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":13.3808302986,"x_train":1838.9438886021,"imp":52.1907830972,"elo":930.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.14819125,"x_train":0.8546409522,"imp":39.617439673,"elo":997.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5196360618,"x_train":4.9583883142,"imp":32.3840101169,"elo":1088.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9536404312,"x_train":16.3951125276,"imp":28.8960747348,"elo":1151.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2230254852,"x_train":35.2860398048,"imp":27.1688368768,"elo":1209.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4290952433,"x_train":91.6231833671,"imp":26.1011115176,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7349630625,"x_train":185.806167626,"imp":25.9042044325,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.8945842103,"x_train":374.1216863756,"imp":25.7657519855,"elo":1321.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":2.0822348565,"x_train":562.8425864643,"imp":25.7028972044,"elo":1322.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":2.1678728195,"x_train":755.269328828,"imp":25.6430947111,"elo":1323.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.6818049083,"x_train":19.6357631766,"imp":27.1075325092,"elo":1210.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.9471265792,"x_train":44.7978169276,"imp":21.9257565449,"elo":1386.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.2236776102,"x_train":111.6196798565,"imp":19.0007646816,"elo":1503.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.3629273246,"x_train":222.0642460075,"imp":17.9998521783,"elo":1545.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.0306005118,"x_train":554.5469212766,"imp":17.1976476447,"elo":1581.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.9685177259,"x_train":1115.1182819708,"imp":16.720750071,"elo":1604.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.5293488548,"x_train":2225.9959788897,"imp":16.3455737417,"elo":1614.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.4116068115,"x_train":3338.8866547043,"imp":16.08769155,"elo":1622.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.354012373,"x_train":4471.5866070717,"imp":15.8630925537,"elo":1623.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8448151087,"x_train":11.2142729689,"imp":13.2180436427,"elo":1644.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3418798785,"x_train":25.4079172215,"imp":13.4498319544,"elo":1651.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2164999438,"x_train":64.4678080693,"imp":13.2472265519,"elo":1679.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.2967203027,"x_train":130.8710194046,"imp":12.9687522021,"elo":1739.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.2104520673,"x_train":334.8480217005,"imp":12.8574688522,"elo":1748.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8500761173,"x_train":675.6180238549,"imp":12.6447608401,"elo":1747.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4475043375,"x_train":1365.6295149721,"imp":12.5754112988,"elo":1740.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8932296593,"x_train":2047.0350964005,"imp":12.4408791211,"elo":1737.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.9362255959,"x_train":2742.7953513314,"imp":12.0897828422,"elo":1732.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.4841490714,"x_train":22.5634771017,"imp":20.8052111165,"elo":1392.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":45.7189326692,"x_train":46.2517520732,"imp":26.0563711418,"elo":1303.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":50.1993286798,"x_train":64.0415875229,"imp":25.6881700958,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":71.5877750075,"x_train":123.7092045238,"imp":24.94122474,"elo":1368.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":93.1265328822,"x_train":248.3053487257,"imp":24.4170391636,"elo":1400.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":112.9531593741,"x_train":584.7096529094,"imp":23.6189596719,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":141.5252995103,"x_train":1166.0242045586,"imp":23.0695461071,"elo":1452.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":180.2303493058,"x_train":2315.5296763529,"imp":22.6422524939,"elo":1473.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":188.8136842543,"x_train":3443.7150532739,"imp":22.4367895489,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":190.4954431622,"x_train":4631.3753499953,"imp":22.2451690194,"elo":1485.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2641860705,"x_train":6.7467122095,"imp":22.3158230071,"elo":1452.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.4463921345,"x_train":64.2501249469,"imp":8.3745097291,"elo":1689.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":1.1673847834,"x_train":5.8322784268,"imp":39.0667940068,"elo":913.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2109379604,"x_train":6.2039007574,"imp":16.2659425384,"elo":1578.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.732957515,"x_train":13.7118099506,"imp":22.1989123677,"elo":1422.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":1.251031111,"x_train":31.4151026664,"imp":21.738020561,"elo":1450.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":2.1831848653,"x_train":93.9419399758,"imp":21.111243279,"elo":1481.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.5316249023,"x_train":192.8564173001,"imp":20.6548170618,"elo":1502.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.9885413426,"x_train":479.3804315766,"imp":20.2582631076,"elo":1519.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.2794460528,"x_train":961.3391911829,"imp":19.9502974945,"elo":1536.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.8992617594,"x_train":1926.418861227,"imp":19.7538496162,"elo":1543.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.3258184146,"x_train":2910.0650077596,"imp":19.6747107631,"elo":1545.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.5061538419,"x_train":3919.285334302,"imp":19.5876203028,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6817692836,"x_train":11.1028664446,"imp":15.4951831367,"elo":1674.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5239137837,"x_train":12.3203746583,"imp":14.9362242891,"elo":1601.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.8772316642,"x_train":801.375360716,"imp":41.3865560038,"elo":855.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.4381519904,"x_train":1026.0399950622,"imp":41.1430700241,"elo":867.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.1527950165,"x_train":1364.8029707909,"imp":40.6205798122,"elo":882.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.9581917636,"x_train":2058.9341404243,"imp":39.5352430188,"elo":908.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.771475197,"x_train":5065.9505852434,"imp":37.2194162392,"elo":947.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9040613643,"x_train":10436.7196251803,"imp":34.8894234924,"elo":982.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":29.4903892803,"x_train":21636.9871111389,"imp":33.639618858,"elo":1010.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.6709522759,"x_train":32013.360321672,"imp":33.0093639821,"elo":1030.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":35.6221458071,"x_train":42561.154121861,"imp":32.5664076078,"elo":1046.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0709157018,"x_train":5.0404725813,"imp":26.3074223481,"elo":1300.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3365764797,"x_train":14.7477895176,"imp":31.0113377824,"elo":1096.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6499035356,"x_train":31.2590378107,"imp":28.5143287813,"elo":1170.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0476437019,"x_train":101.8350811731,"imp":26.4972964922,"elo":1241.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2034747343,"x_train":218.6755099117,"imp":25.3907953233,"elo":1295.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4189325906,"x_train":533.9506807148,"imp":24.540745302,"elo":1322.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6704753472,"x_train":1017.6084144692,"imp":24.1150987447,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8715186479,"x_train":1968.6547247706,"imp":23.8565194941,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0005251114,"x_train":3052.495945877,"imp":23.6954730671,"elo":1349.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.096613965,"x_train":4142.8989350983,"imp":23.6068686261,"elo":1352.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.599155634,"x_train":5.4806602189,"imp":24.5657052646,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.0378638981,"x_train":18.5840857617,"imp":23.2548839265,"elo":1316.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.8012686319,"x_train":57.5257824541,"imp":22.7739284558,"elo":1344.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.9499124215,"x_train":117.1284034437,"imp":22.2997595757,"elo":1367.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.0226060034,"x_train":309.9745400782,"imp":21.9370489498,"elo":1388.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.442578599,"x_train":629.246307752,"imp":21.6824794187,"elo":1401.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.7654993382,"x_train":1266.4892033336,"imp":21.4789767835,"elo":1413.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.559561221,"x_train":1899.605994117,"imp":21.3250522743,"elo":1422.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.5236730663,"x_train":2551.6500015348,"imp":21.1407437939,"elo":1433.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":277.1925819839,"x_train":581.4914553919,"imp":30.8091754666,"elo":1028.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":597.2305576187,"x_train":1918.622222682,"imp":27.8960597479,"elo":1121.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":953.3210135052,"x_train":5770.5281133302,"imp":24.9152997253,"elo":1264.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":968.9952282534,"x_train":12892.6565755789,"imp":23.5116264395,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":993.7285537193,"x_train":32775.6937551607,"imp":21.5724840444,"elo":1422.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":1003.6468494405,"x_train":34415.8855588805,"imp":21.534818294,"elo":1426.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":2.1887529296,"x_train":10.8402948127,"imp":28.5832258246,"elo":1159.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":2.4383109526,"x_train":22.047469012,"imp":26.7033859005,"elo":1235.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.3346191076,"x_train":52.9985363102,"imp":24.55825223,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":2.1631624605,"x_train":105.4364802264,"imp":23.6556190512,"elo":1377.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.9055625999,"x_train":261.4235869814,"imp":22.5736423267,"elo":1418.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2090360064,"x_train":521.852845102,"imp":22.0943565016,"elo":1439.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.4238113379,"x_train":1043.5013351397,"imp":21.7727076399,"elo":1452.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.7252746839,"x_train":1565.3955882162,"imp":21.6898068619,"elo":1455.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.1516651241,"x_train":2097.6180059154,"imp":21.637128979,"elo":1457.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/winrate_explorer.html b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..ee939418dea9da889252492f662252c219a720ad
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_multiclass/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (default)", "TabPFN-3 (default)", "RealMLP (tuned + ensembled)", "TabICLv2 (default)", "RealMLP (tuned)", "LimiX (default)", "TabDPT (tuned + ensembled)", "TabM (tuned + ensembled)", "TabDPT-Turbo (default)", "AutoGluon 1.4 (best, 4h)", "TabM (tuned)", "CatBoost (tuned + ensembled)", "TabDPT (tuned)", "ModernNCA (tuned + ensembled)", "CatBoost (tuned)", "ModernNCA (tuned)", "XGBoost (tuned)", "XGBoost (tuned + ensembled)", "LightGBM (tuned + ensembled)", "iLTM (tuned + ensembled)", "LightGBM (tuned)", "BetaTabPFN (default)", "xRFM (tuned + ensembled)", "TabM (default)", "ExtraTrees (tuned + ensembled)", "SAP-RPT-OSS (default)", "TabDPT (default)", "iLTM (tuned)", "TorchMLP (tuned + ensembled)", "EBM (tuned + ensembled)", "RandomForest (tuned + ensembled)", "xRFM (tuned)", "CatBoost (default)", "ExtraTrees (tuned)", "TabSwift (default)", "EBM (tuned)", "FastaiMLP (tuned + ensembled)", "ModernNCA (default)", "RandomForest (tuned)", "TorchMLP (tuned)", "XGBoost (default)", "EBM (default)", "FastaiMLP (tuned)", "RealMLP (default)", "LightGBM (default)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "xRFM (default)", "TorchMLP (default)", "KNN (tuned + ensembled)", "ChimeraBoost (default)", "OrionMSP (default)", "RandomForest (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "iLTM (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "PerpetualBooster (tuned)", "PerpetualBooster (default)", "FastaiMLP (default)", "TabFlex (default)", "KNN (tuned)", "TabSTAR (default)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.7006802558898926, 0.7006802558898926, 0.8163265585899353, 0.8367347121238708, 0.7414966225624084, 0.8503401279449463, 0.7687074542045593, 0.8911564350128174, 0.7687074542045593, 0.8435373902320862, 0.8979591727256775, 0.8095238208770752, 0.884353756904602, 0.9251700639724731, 0.8231292366981506, 0.9047619104385376, 0.884353756904602, 0.918367326259613, 0.9251700639724731, 0.8911564350128174, 0.8775510191917419, 0.884353756904602, 0.9047619104385376, 0.918367326259613, 0.8979591727256775, 0.9115646481513977, 0.918367326259613, 0.8503401279449463, 0.9251700639724731, 0.7551020383834839, 0.9047619104385376, 0.9319728016853333, 0.9251700639724731, 0.9047619104385376, 0.9047619104385376, 0.9251700639724731, 0.9319728016853333, 0.918367326259613, 0.918367326259613, 0.9251700639724731, 0.8979591727256775, 0.9795918464660645, 0.9047619104385376, 0.9319728016853333, 0.9319728016853333, 0.9251700639724731, 0.9387755393981934, 0.9795918464660645, 0.9251700639724731, 0.9251700639724731, 0.9251700639724731, 0.9251700639724731, 0.9319728016853333, 0.9319728016853333, 0.9387755393981934, 0.9319728016853333, 0.9931972622871399, 1.0, 0.9931972622871399, 0.9931972622871399, 0.9387755393981934, 0.9387755393981934, 0.9319728016853333, 0.9931972622871399, 0.9727891087532043, 0.9659863710403442, 1.0, 0.9659863710403442, 0.9319728016853333, 0.9931972622871399, 1.0, 1.0], [0.29931971430778503, NaN, 0.4625850319862366, 0.5782312750816345, 0.6190476417541504, 0.6836734414100647, 0.6190476417541504, 0.7414966225624084, 0.6530612111091614, 0.7687074542045593, 0.7346938848495483, 0.7278911471366882, 0.8435373902320862, 0.6938775777816772, 0.884353756904602, 0.8503401279449463, 0.9251700639724731, 0.7755101919174194, 0.8503401279449463, 0.9251700639724731, 0.8639456033706665, 0.9047619104385376, 0.9047619104385376, 0.884353756904602, 0.8775510191917419, 0.9115646481513977, 0.8979591727256775, 0.8911564350128174, 0.8571428656578064, 0.8639456033706665, 0.7482993006706238, 0.795918345451355, 0.9523809552192688, 0.9455782175064087, 0.9523809552192688, 0.8775510191917419, 0.918367326259613, 0.9659863710403442, 0.8911564350128174, 0.9727891087532043, 0.9591836929321289, 0.9251700639724731, 0.9455782175064087, 0.9047619104385376, 0.9455782175064087, 0.9795918464660645, 0.9455782175064087, 0.9659863710403442, 0.9931972622871399, 0.9455782175064087, 0.9523809552192688, 0.9659863710403442, 0.9795918464660645, 0.9931972622871399, 0.9319728016853333, 0.9591836929321289, 0.9795918464660645, 0.9863945841789246, 0.9931972622871399, 0.9931972622871399, 1.0, 1.0, 0.9863945841789246, 0.9863945841789246, 0.9863945841789246, 0.9727891087532043, 0.9659863710403442, 1.0, 1.0, 0.9659863710403442, 0.9863945841789246, 1.0, 1.0], [0.29931971430778503, 0.5374149680137634, NaN, 0.6054421663284302, 0.6258503198623657, 0.7482993006706238, 0.6258503198623657, 0.7346938848495483, 0.6598639488220215, 0.7482993006706238, 0.7414966225624084, 0.7210884094238281, 0.795918345451355, 0.6870748400688171, 0.8571428656578064, 0.795918345451355, 0.9387755393981934, 0.7414966225624084, 0.8571428656578064, 0.9251700639724731, 0.8775510191917419, 0.8979591727256775, 0.918367326259613, 0.8911564350128174, 0.884353756904602, 0.8979591727256775, 0.9115646481513977, 0.8775510191917419, 0.8435373902320862, 0.8435373902320862, 0.7414966225624084, 0.7755101919174194, 0.9523809552192688, 0.9387755393981934, 0.9319728016853333, 0.8707482814788818, 0.918367326259613, 0.9387755393981934, 0.8707482814788818, 0.9727891087532043, 0.9387755393981934, 0.9319728016853333, 0.9319728016853333, 0.9047619104385376, 0.9455782175064087, 0.9863945841789246, 0.9319728016853333, 0.9523809552192688, 0.9863945841789246, 0.9319728016853333, 0.9387755393981934, 0.9523809552192688, 0.9659863710403442, 0.9863945841789246, 0.9251700639724731, 0.9387755393981934, 0.9863945841789246, 0.9863945841789246, 0.9931972622871399, 0.9931972622871399, 0.9931972622871399, 1.0, 0.9931972622871399, 0.9931972622871399, 0.9863945841789246, 0.9727891087532043, 0.9727891087532043, 0.9931972622871399, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.18367347121238708, 0.4217686951160431, 0.3945578336715698, NaN, 0.5986394286155701, 0.4829931855201721, 0.5714285969734192, 0.7210884094238281, 0.5714285969734192, 0.7346938848495483, 0.7891156673431396, 0.8163265585899353, 0.7687074542045593, 0.7346938848495483, 0.795918345451355, 0.761904776096344, 0.884353756904602, 0.8571428656578064, 0.8639456033706665, 0.8979591727256775, 0.884353756904602, 0.8299319744110107, 0.8571428656578064, 0.9047619104385376, 0.9115646481513977, 0.8911564350128174, 0.8231292366981506, 0.9047619104385376, 0.8027210831642151, 0.8979591727256775, 0.6258503198623657, 0.8367347121238708, 0.9387755393981934, 0.9047619104385376, 0.918367326259613, 0.9047619104385376, 0.9387755393981934, 0.9455782175064087, 0.8979591727256775, 0.9863945841789246, 0.9455782175064087, 0.8707482814788818, 0.9659863710403442, 0.9047619104385376, 0.9523809552192688, 0.9931972622871399, 0.9387755393981934, 0.9387755393981934, 0.9795918464660645, 0.9727891087532043, 0.9727891087532043, 0.9863945841789246, 0.9863945841789246, 0.9863945841789246, 0.9795918464660645, 0.9863945841789246, 0.9931972622871399, 0.9931972622871399, 0.9931972622871399, 0.9727891087532043, 0.9795918464660645, 1.0, 0.9931972622871399, 0.9931972622871399, 0.9931972622871399, 0.9727891087532043, 0.9659863710403442, 0.9931972622871399, 1.0, 0.9795918464660645, 0.9931972622871399, 1.0, 1.0], [0.16326530277729034, 0.380952388048172, 0.3741496503353119, 0.40136054158210754, NaN, 0.523809552192688, 0.45578232407569885, 0.6326530575752258, 0.45578232407569885, 0.6326530575752258, 0.646258533000946, 0.6258503198623657, 0.7142857313156128, 0.6734693646430969, 0.7687074542045593, 0.7482993006706238, 0.8027210831642151, 0.7006802558898926, 0.8571428656578064, 0.8163265585899353, 0.8299319744110107, 0.8775510191917419, 0.8979591727256775, 0.8503401279449463, 0.9251700639724731, 0.884353756904602, 0.8367347121238708, 0.7891156673431396, 0.7891156673431396, 0.8435373902320862, 0.6530612111091614, 0.7755101919174194, 0.9455782175064087, 0.918367326259613, 0.8639456033706665, 0.8231292366981506, 0.8503401279449463, 0.9523809552192688, 0.8367347121238708, 0.9523809552192688, 0.9047619104385376, 0.9455782175064087, 0.9319728016853333, 0.8639456033706665, 0.918367326259613, 0.9795918464660645, 0.9251700639724731, 0.9795918464660645, 0.9659863710403442, 0.9319728016853333, 0.9319728016853333, 0.9455782175064087, 0.9319728016853333, 0.9455782175064087, 0.9523809552192688, 0.9455782175064087, 0.9659863710403442, 1.0, 1.0, 0.9795918464660645, 0.9863945841789246, 1.0, 0.9931972622871399, 0.9931972622871399, 0.9659863710403442, 0.9523809552192688, 0.9659863710403442, 1.0, 1.0, 0.9591836929321289, 0.9931972622871399, 1.0, 1.0], [0.25850340723991394, 0.3163265287876129, 0.25170066952705383, 0.5170068144798279, 0.4761904776096344, NaN, 0.557823121547699, 0.6394557952880859, 0.5782312750816345, 0.6530612111091614, 0.7278911471366882, 0.6598639488220215, 0.7891156673431396, 0.6258503198623657, 0.7891156673431396, 0.8095238208770752, 0.8639456033706665, 0.680272102355957, 0.7414966225624084, 0.8911564350128174, 0.7891156673431396, 0.8639456033706665, 0.8639456033706665, 0.8503401279449463, 0.795918345451355, 0.8707482814788818, 0.8503401279449463, 0.8163265585899353, 0.8163265585899353, 0.8027210831642151, 0.7142857313156128, 0.7006802558898926, 0.8979591727256775, 0.8979591727256775, 0.9047619104385376, 0.8231292366981506, 0.8707482814788818, 0.9115646481513977, 0.8435373902320862, 0.9319728016853333, 0.9115646481513977, 0.918367326259613, 0.9047619104385376, 0.9115646481513977, 0.8979591727256775, 0.9455782175064087, 0.9251700639724731, 0.9387755393981934, 0.9931972622871399, 0.9115646481513977, 0.9115646481513977, 0.918367326259613, 0.9319728016853333, 0.9591836929321289, 0.8775510191917419, 0.9319728016853333, 0.9863945841789246, 0.9591836929321289, 0.9931972622871399, 1.0, 0.9931972622871399, 0.9863945841789246, 0.9863945841789246, 0.9795918464660645, 0.9863945841789246, 0.9727891087532043, 0.9727891087532043, 0.9931972622871399, 0.9931972622871399, 0.9455782175064087, 1.0, 1.0, 1.0], [0.14965985715389252, 0.380952388048172, 0.3741496503353119, 0.4285714328289032, 0.5442177057266235, 0.442176878452301, NaN, 0.523809552192688, 0.5714285969734192, 0.6054421663284302, 0.646258533000946, 0.7278911471366882, 0.6598639488220215, 0.7210884094238281, 0.6938775777816772, 0.6938775777816772, 0.8231292366981506, 0.761904776096344, 0.8095238208770752, 0.8231292366981506, 0.8231292366981506, 0.7755101919174194, 0.7823129296302795, 0.8435373902320862, 0.8503401279449463, 0.8367347121238708, 0.7414966225624084, 0.8367347121238708, 0.7006802558898926, 0.9047619104385376, 0.5714285969734192, 0.8163265585899353, 0.9115646481513977, 0.8775510191917419, 0.8299319744110107, 0.8775510191917419, 0.9047619104385376, 0.884353756904602, 0.9115646481513977, 0.9455782175064087, 0.8503401279449463, 0.8299319744110107, 0.9659863710403442, 0.8979591727256775, 0.8911564350128174, 0.9795918464660645, 0.918367326259613, 0.9387755393981934, 0.9863945841789246, 0.9251700639724731, 0.9047619104385376, 0.9251700639724731, 0.9795918464660645, 0.9659863710403442, 0.9455782175064087, 0.9319728016853333, 0.9931972622871399, 1.0, 1.0, 0.9863945841789246, 0.9863945841789246, 0.9931972622871399, 1.0, 0.9795918464660645, 0.9727891087532043, 0.9727891087532043, 0.9659863710403442, 0.9931972622871399, 1.0, 0.9455782175064087, 1.0, 1.0, 1.0], [0.2312925159931183, 0.25850340723991394, 0.26530611515045166, 0.2789115607738495, 0.36734694242477417, 0.36054420471191406, 0.4761904776096344, NaN, 0.442176878452301, 0.5136054158210754, 0.5170068144798279, 0.5510203838348389, 0.7346938848495483, 0.5646258592605591, 0.7074829936027527, 0.7414966225624084, 0.7823129296302795, 0.557823121547699, 0.7278911471366882, 0.8163265585899353, 0.7823129296302795, 0.8979591727256775, 0.9251700639724731, 0.8775510191917419, 0.8027210831642151, 0.8775510191917419, 0.8571428656578064, 0.8231292366981506, 0.8707482814788818, 0.7891156673431396, 0.6530612111091614, 0.6054421663284302, 0.9047619104385376, 0.8027210831642151, 0.918367326259613, 0.8639456033706665, 0.8503401279449463, 0.9319728016853333, 0.8571428656578064, 0.7891156673431396, 0.9251700639724731, 0.9047619104385376, 0.9455782175064087, 0.9115646481513977, 0.8571428656578064, 0.9931972622871399, 0.9319728016853333, 0.9115646481513977, 0.9931972622871399, 0.9251700639724731, 0.9387755393981934, 0.9455782175064087, 0.9319728016853333, 0.9727891087532043, 0.918367326259613, 0.9387755393981934, 0.9863945841789246, 0.9727891087532043, 1.0, 0.9931972622871399, 0.9863945841789246, 1.0, 0.9727891087532043, 0.9659863710403442, 0.9795918464660645, 0.9727891087532043, 0.9659863710403442, 1.0, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.10884353518486023, 0.3469387888908386, 0.3401360511779785, 0.4285714328289032, 0.5442177057266235, 0.4217686951160431, 0.4285714328289032, 0.557823121547699, NaN, 0.5850340127944946, 0.6190476417541504, 0.7346938848495483, 0.6666666865348816, 0.6734693646430969, 0.7142857313156128, 0.6734693646430969, 0.7755101919174194, 0.7346938848495483, 0.7414966225624084, 0.7823129296302795, 0.8095238208770752, 0.7551020383834839, 0.7687074542045593, 0.8435373902320862, 0.8707482814788818, 0.8571428656578064, 0.7346938848495483, 0.7823129296302795, 0.7346938848495483, 0.8163265585899353, 0.5102040767669678, 0.7482993006706238, 0.8775510191917419, 0.8231292366981506, 0.8299319744110107, 0.8707482814788818, 0.8163265585899353, 0.884353756904602, 0.8367347121238708, 0.918367326259613, 0.8571428656578064, 0.8163265585899353, 0.9523809552192688, 0.8707482814788818, 0.8775510191917419, 0.9251700639724731, 0.9047619104385376, 0.8979591727256775, 0.9251700639724731, 0.9251700639724731, 0.918367326259613, 0.9251700639724731, 0.8911564350128174, 0.8911564350128174, 0.9319728016853333, 0.9319728016853333, 0.9115646481513977, 0.9931972622871399, 0.9727891087532043, 0.9863945841789246, 0.9795918464660645, 0.9387755393981934, 0.9319728016853333, 0.9319728016853333, 0.9727891087532043, 0.9659863710403442, 0.9659863710403442, 0.9863945841789246, 0.9455782175064087, 0.9387755393981934, 0.9931972622871399, 1.0, 1.0], [0.2312925159931183, 0.2312925159931183, 0.25170066952705383, 0.26530611515045166, 0.36734694242477417, 0.3469387888908386, 0.3945578336715698, 0.4863945543766022, 0.41496598720550537, NaN, 0.5102040767669678, 0.5102040767669678, 0.6598639488220215, 0.5510203838348389, 0.6530612111091614, 0.7006802558898926, 0.7482993006706238, 0.5442177057266235, 0.7346938848495483, 0.7551020383834839, 0.7551020383834839, 0.8299319744110107, 0.8571428656578064, 0.7891156673431396, 0.7687074542045593, 0.8027210831642151, 0.8435373902320862, 0.761904776096344, 0.8027210831642151, 0.7823129296302795, 0.6394557952880859, 0.6190476417541504, 0.884353756904602, 0.8503401279449463, 0.8979591727256775, 0.8367347121238708, 0.8299319744110107, 0.8911564350128174, 0.8231292366981506, 0.7891156673431396, 0.9387755393981934, 0.8707482814788818, 0.918367326259613, 0.8571428656578064, 0.8979591727256775, 0.9591836929321289, 0.918367326259613, 0.9319728016853333, 0.9795918464660645, 0.918367326259613, 0.918367326259613, 0.9387755393981934, 0.9047619104385376, 0.9659863710403442, 0.9251700639724731, 0.9455782175064087, 0.9795918464660645, 0.9727891087532043, 0.9795918464660645, 0.9863945841789246, 0.9931972622871399, 0.9931972622871399, 0.9727891087532043, 0.9591836929321289, 0.9863945841789246, 0.9727891087532043, 0.9591836929321289, 1.0, 1.0, 0.9727891087532043, 0.9863945841789246, 1.0, 1.0], [0.15646257996559143, 0.26530611515045166, 0.25850340723991394, 0.21088434755802155, 0.35374149680137634, 0.27210885286331177, 0.35374149680137634, 0.4829931855201721, 0.380952388048172, 0.4897959232330322, NaN, 0.5170068144798279, 0.6326530575752258, 0.4625850319862366, 0.6394557952880859, 0.6394557952880859, 0.7278911471366882, 0.6054421663284302, 0.6326530575752258, 0.7414966225624084, 0.7687074542045593, 0.761904776096344, 0.7551020383834839, 0.7891156673431396, 0.7482993006706238, 0.795918345451355, 0.7482993006706238, 0.8367347121238708, 0.7074829936027527, 0.884353756904602, 0.5850340127944946, 0.6054421663284302, 0.8299319744110107, 0.7006802558898926, 0.8231292366981506, 0.9047619104385376, 0.8979591727256775, 0.8775510191917419, 0.9047619104385376, 0.8095238208770752, 0.8775510191917419, 0.7687074542045593, 0.9387755393981934, 0.8979591727256775, 0.8163265585899353, 0.9455782175064087, 0.8979591727256775, 0.884353756904602, 0.9795918464660645, 0.9387755393981934, 0.9115646481513977, 0.9387755393981934, 0.9727891087532043, 0.9591836929321289, 0.9455782175064087, 0.9455782175064087, 0.9727891087532043, 1.0, 1.0, 0.9863945841789246, 0.9931972622871399, 0.9727891087532043, 0.9795918464660645, 0.9659863710403442, 0.9795918464660645, 0.9727891087532043, 0.9659863710403442, 1.0, 0.9931972622871399, 0.9523809552192688, 1.0, 1.0, 1.0], [0.10204081982374191, 0.27210885286331177, 0.2789115607738495, 0.18367347121238708, 0.3741496503353119, 0.3401360511779785, 0.27210885286331177, 0.44897958636283875, 0.26530611515045166, 0.4897959232330322, 0.4829931855201721, NaN, 0.5442177057266235, 0.4761904776096344, 0.557823121547699, 0.5510203838348389, 0.5782312750816345, 0.6598639488220215, 0.6530612111091614, 0.6122449040412903, 0.680272102355957, 0.6326530575752258, 0.6530612111091614, 0.7142857313156128, 0.7551020383834839, 0.6938775777816772, 0.6326530575752258, 0.6326530575752258, 0.6326530575752258, 0.7346938848495483, 0.4897959232330322, 0.7482993006706238, 0.7891156673431396, 0.680272102355957, 0.7346938848495483, 0.795918345451355, 0.7414966225624084, 0.7891156673431396, 0.7891156673431396, 0.8639456033706665, 0.7687074542045593, 0.7687074542045593, 0.8503401279449463, 0.8367347121238708, 0.8095238208770752, 0.8979591727256775, 0.795918345451355, 0.8163265585899353, 0.8571428656578064, 0.8571428656578064, 0.8911564350128174, 0.918367326259613, 0.8503401279449463, 0.8911564350128174, 0.9659863710403442, 0.9115646481513977, 0.9591836929321289, 0.9795918464660645, 1.0, 0.8571428656578064, 0.8707482814788818, 0.9387755393981934, 0.9047619104385376, 0.9047619104385376, 0.9591836929321289, 0.9591836929321289, 0.9523809552192688, 0.9387755393981934, 1.0, 0.9727891087532043, 0.9319728016853333, 0.9523809552192688, 1.0], [0.190476194024086, 0.15646257996559143, 0.20408163964748383, 0.2312925159931183, 0.2857142984867096, 0.21088434755802155, 0.3401360511779785, 0.26530611515045166, 0.3333333432674408, 0.3401360511779785, 0.36734694242477417, 0.45578232407569885, NaN, 0.5102040767669678, 0.523809552192688, 0.6360543966293335, 0.5306122303009033, 0.4897959232330322, 0.6122449040412903, 0.5986394286155701, 0.6122449040412903, 0.6870748400688171, 0.7142857313156128, 0.6598639488220215, 0.6258503198623657, 0.6666666865348816, 0.761904776096344, 0.6190476417541504, 0.8775510191917419, 0.5782312750816345, 0.6394557952880859, 0.5986394286155701, 0.7414966225624084, 0.7755101919174194, 0.8639456033706665, 0.6394557952880859, 0.7006802558898926, 0.8299319744110107, 0.6326530575752258, 0.6870748400688171, 0.8911564350128174, 0.8163265585899353, 0.8095238208770752, 0.7551020383834839, 0.8707482814788818, 0.884353756904602, 0.8911564350128174, 0.8775510191917419, 0.9523809552192688, 0.884353756904602, 0.8571428656578064, 0.884353756904602, 0.8027210831642151, 0.9455782175064087, 0.884353756904602, 0.9523809552192688, 0.9387755393981934, 0.8911564350128174, 0.9115646481513977, 0.9591836929321289, 0.9795918464660645, 1.0, 0.9387755393981934, 0.9523809552192688, 0.9523809552192688, 0.9659863710403442, 0.9523809552192688, 0.9795918464660645, 1.0, 0.9659863710403442, 0.9931972622871399, 1.0, 1.0], [0.11564625799655914, 0.30612245202064514, 0.31292515993118286, 0.26530611515045166, 0.3265306055545807, 0.3741496503353119, 0.2789115607738495, 0.4353741407394409, 0.3265306055545807, 0.44897958636283875, 0.5374149680137634, 0.523809552192688, 0.4897959232330322, NaN, 0.5442177057266235, 0.5102040767669678, 0.6054421663284302, 0.5782312750816345, 0.6598639488220215, 0.5918367505073547, 0.6530612111091614, 0.6122449040412903, 0.6122449040412903, 0.6734693646430969, 0.795918345451355, 0.6190476417541504, 0.5510203838348389, 0.6054421663284302, 0.557823121547699, 0.7551020383834839, 0.49659863114356995, 0.7891156673431396, 0.8027210831642151, 0.7006802558898926, 0.6666666865348816, 0.7823129296302795, 0.6734693646430969, 0.7414966225624084, 0.7687074542045593, 0.8231292366981506, 0.7278911471366882, 0.7755101919174194, 0.8299319744110107, 0.7891156673431396, 0.7687074542045593, 0.8639456033706665, 0.7687074542045593, 0.8639456033706665, 0.8435373902320862, 0.8503401279449463, 0.884353756904602, 0.884353756904602, 0.7755101919174194, 0.8435373902320862, 0.918367326259613, 0.884353756904602, 0.9387755393981934, 1.0, 1.0, 0.8639456033706665, 0.8503401279449463, 0.9387755393981934, 0.8775510191917419, 0.8775510191917419, 0.9387755393981934, 0.9387755393981934, 0.9455782175064087, 0.9387755393981934, 1.0, 0.9455782175064087, 0.9387755393981934, 0.9659863710403442, 1.0], [0.07482992857694626, 0.11564625799655914, 0.1428571492433548, 0.20408163964748383, 0.2312925159931183, 0.21088434755802155, 0.30612245202064514, 0.2925170063972473, 0.2857142984867096, 0.3469387888908386, 0.36054420471191406, 0.442176878452301, 0.4761904776096344, 0.45578232407569885, NaN, 0.5170068144798279, 0.5782312750816345, 0.4761904776096344, 0.5918367505073547, 0.6530612111091614, 0.6258503198623657, 0.5918367505073547, 0.6258503198623657, 0.7210884094238281, 0.6666666865348816, 0.7074829936027527, 0.646258533000946, 0.6598639488220215, 0.6666666865348816, 0.5918367505073547, 0.5102040767669678, 0.5986394286155701, 0.7414966225624084, 0.7414966225624084, 0.7755101919174194, 0.6530612111091614, 0.7278911471366882, 0.8163265585899353, 0.680272102355957, 0.7074829936027527, 0.8027210831642151, 0.8231292366981506, 0.8163265585899353, 0.7551020383834839, 0.8299319744110107, 0.8367347121238708, 0.8639456033706665, 0.8979591727256775, 0.8707482814788818, 0.8571428656578064, 0.8775510191917419, 0.884353756904602, 0.7891156673431396, 0.9115646481513977, 0.8231292366981506, 0.9251700639724731, 0.9047619104385376, 0.8979591727256775, 0.9251700639724731, 0.9455782175064087, 0.9591836929321289, 0.9795918464660645, 0.9387755393981934, 0.9387755393981934, 0.9387755393981934, 0.9455782175064087, 0.9523809552192688, 0.9795918464660645, 0.9795918464660645, 0.9523809552192688, 0.9659863710403442, 0.9931972622871399, 1.0], [0.17687074840068817, 0.14965985715389252, 0.20408163964748383, 0.2380952388048172, 0.25170066952705383, 0.190476194024086, 0.30612245202064514, 0.25850340723991394, 0.3265306055545807, 0.29931971430778503, 0.36054420471191406, 0.44897958636283875, 0.3639455735683441, 0.4897959232330322, 0.4829931855201721, NaN, 0.4625850319862366, 0.45578232407569885, 0.5646258592605591, 0.523809552192688, 0.5510203838348389, 0.5986394286155701, 0.6122449040412903, 0.5782312750816345, 0.6326530575752258, 0.5782312750816345, 0.6734693646430969, 0.5986394286155701, 0.7414966225624084, 0.5714285969734192, 0.5714285969734192, 0.6054421663284302, 0.680272102355957, 0.7482993006706238, 0.8027210831642151, 0.646258533000946, 0.6938775777816772, 0.761904776096344, 0.6122449040412903, 0.6666666865348816, 0.8571428656578064, 0.7823129296302795, 0.7755101919174194, 0.7210884094238281, 0.8367347121238708, 0.8435373902320862, 0.8707482814788818, 0.8367347121238708, 0.9115646481513977, 0.7891156673431396, 0.795918345451355, 0.8163265585899353, 0.7755101919174194, 0.9319728016853333, 0.8435373902320862, 0.9115646481513977, 0.9251700639724731, 0.884353756904602, 0.8503401279449463, 0.9727891087532043, 0.9727891087532043, 1.0, 0.9251700639724731, 0.9523809552192688, 0.9591836929321289, 0.9523809552192688, 0.9455782175064087, 0.9727891087532043, 1.0, 0.9659863710403442, 0.9863945841789246, 1.0, 1.0], [0.095238097012043, 0.07482992857694626, 0.06122449040412903, 0.11564625799655914, 0.1972789168357849, 0.13605442643165588, 0.17687074840068817, 0.21768707036972046, 0.22448979318141937, 0.25170066952705383, 0.27210885286331177, 0.4217686951160431, 0.4693877696990967, 0.3945578336715698, 0.4217686951160431, 0.5374149680137634, NaN, 0.49659863114356995, 0.4897959232330322, 0.6972789168357849, 0.523809552192688, 0.6190476417541504, 0.6394557952880859, 0.6326530575752258, 0.5170068144798279, 0.6326530575752258, 0.5986394286155701, 0.5782312750816345, 0.646258533000946, 0.6326530575752258, 0.5102040767669678, 0.5714285969734192, 0.6734693646430969, 0.6394557952880859, 0.7414966225624084, 0.6938775777816772, 0.761904776096344, 0.8231292366981506, 0.7210884094238281, 0.7278911471366882, 0.8299319744110107, 0.6870748400688171, 0.8231292366981506, 0.8231292366981506, 0.7891156673431396, 0.9455782175064087, 0.8707482814788818, 0.8231292366981506, 0.9455782175064087, 0.884353756904602, 0.8707482814788818, 0.9115646481513977, 0.8979591727256775, 0.9591836929321289, 0.8639456033706665, 0.918367326259613, 0.9727891087532043, 0.918367326259613, 0.9455782175064087, 0.9591836929321289, 0.9659863710403442, 1.0, 0.9795918464660645, 0.9727891087532043, 0.9727891087532043, 0.9727891087532043, 0.9659863710403442, 0.9795918464660645, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.11564625799655914, 0.22448979318141937, 0.25850340723991394, 0.1428571492433548, 0.29931971430778503, 0.31972789764404297, 0.2380952388048172, 0.442176878452301, 0.26530611515045166, 0.45578232407569885, 0.3945578336715698, 0.3401360511779785, 0.5102040767669678, 0.4217686951160431, 0.523809552192688, 0.5442177057266235, 0.5034013390541077, NaN, 0.5850340127944946, 0.5442177057266235, 0.6326530575752258, 0.5986394286155701, 0.6258503198623657, 0.6394557952880859, 0.646258533000946, 0.6530612111091614, 0.5782312750816345, 0.557823121547699, 0.6190476417541504, 0.6734693646430969, 0.4761904776096344, 0.7346938848495483, 0.680272102355957, 0.6326530575752258, 0.6258503198623657, 0.7006802558898926, 0.6734693646430969, 0.7006802558898926, 0.7210884094238281, 0.8231292366981506, 0.6938775777816772, 0.6938775777816772, 0.7687074542045593, 0.7482993006706238, 0.7278911471366882, 0.8027210831642151, 0.7278911471366882, 0.7755101919174194, 0.7482993006706238, 0.8435373902320862, 0.8571428656578064, 0.8775510191917419, 0.8163265585899353, 0.8571428656578064, 0.9591836929321289, 0.918367326259613, 0.9591836929321289, 0.9591836929321289, 0.9795918464660645, 0.8095238208770752, 0.8027210831642151, 0.9319728016853333, 0.9047619104385376, 0.9047619104385376, 0.9319728016853333, 0.9455782175064087, 0.9523809552192688, 0.9387755393981934, 1.0, 0.9591836929321289, 0.9251700639724731, 0.9387755393981934, 1.0], [0.08163265138864517, 0.14965985715389252, 0.1428571492433548, 0.13605442643165588, 0.1428571492433548, 0.25850340723991394, 0.190476194024086, 0.27210885286331177, 0.25850340723991394, 0.26530611515045166, 0.36734694242477417, 0.3469387888908386, 0.3877550959587097, 0.3401360511779785, 0.40816327929496765, 0.4353741407394409, 0.5102040767669678, 0.41496598720550537, NaN, 0.5442177057266235, 0.6224489808082581, 0.5714285969734192, 0.5850340127944946, 0.6326530575752258, 0.5306122303009033, 0.646258533000946, 0.5442177057266235, 0.6394557952880859, 0.5442177057266235, 0.7278911471366882, 0.5306122303009033, 0.5510203838348389, 0.6734693646430969, 0.5918367505073547, 0.6938775777816772, 0.8027210831642151, 0.795918345451355, 0.7142857313156128, 0.8027210831642151, 0.6530612111091614, 0.761904776096344, 0.6666666865348816, 0.9047619104385376, 0.8163265585899353, 0.761904776096344, 0.8707482814788818, 0.795918345451355, 0.8027210831642151, 0.884353756904602, 0.8027210831642151, 0.8775510191917419, 0.8639456033706665, 0.8911564350128174, 0.9251700639724731, 0.9115646481513977, 0.884353756904602, 0.9523809552192688, 0.9455782175064087, 0.9863945841789246, 0.9115646481513977, 0.9591836929321289, 0.9795918464660645, 0.9115646481513977, 0.9047619104385376, 0.9795918464660645, 0.9727891087532043, 0.9523809552192688, 0.9455782175064087, 0.9931972622871399, 0.9319728016853333, 0.9659863710403442, 0.9931972622871399, 1.0], [0.07482992857694626, 0.07482992857694626, 0.07482992857694626, 0.10204081982374191, 0.18367347121238708, 0.10884353518486023, 0.17687074840068817, 0.18367347121238708, 0.21768707036972046, 0.2448979616165161, 0.25850340723991394, 0.3877550959587097, 0.40136054158210754, 0.40816327929496765, 0.3469387888908386, 0.4761904776096344, 0.3027210831642151, 0.45578232407569885, 0.45578232407569885, NaN, 0.5102040767669678, 0.5782312750816345, 0.5782312750816345, 0.5782312750816345, 0.4897959232330322, 0.557823121547699, 0.557823121547699, 0.5306122303009033, 0.6326530575752258, 0.6122449040412903, 0.4897959232330322, 0.557823121547699, 0.646258533000946, 0.6530612111091614, 0.7278911471366882, 0.6598639488220215, 0.7482993006706238, 0.7687074542045593, 0.6870748400688171, 0.680272102355957, 0.7755101919174194, 0.6734693646430969, 0.7891156673431396, 0.795918345451355, 0.7687074542045593, 0.884353756904602, 0.8435373902320862, 0.7823129296302795, 0.9047619104385376, 0.8435373902320862, 0.8435373902320862, 0.8435373902320862, 0.8707482814788818, 0.9319728016853333, 0.8435373902320862, 0.918367326259613, 0.9727891087532043, 0.9115646481513977, 0.9251700639724731, 0.9523809552192688, 0.9659863710403442, 1.0, 0.9659863710403442, 0.9591836929321289, 0.9727891087532043, 0.9727891087532043, 0.9659863710403442, 0.9727891087532043, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.10884353518486023, 0.13605442643165588, 0.12244898080825806, 0.11564625799655914, 0.17006802558898926, 0.21088434755802155, 0.17687074840068817, 0.21768707036972046, 0.190476194024086, 0.2448979616165161, 0.2312925159931183, 0.31972789764404297, 0.3877550959587097, 0.3469387888908386, 0.3741496503353119, 0.44897958636283875, 0.4761904776096344, 0.36734694242477417, 0.37755101919174194, 0.4897959232330322, NaN, 0.5102040767669678, 0.5306122303009033, 0.5782312750816345, 0.45578232407569885, 0.5782312750816345, 0.5034013390541077, 0.5850340127944946, 0.5442177057266235, 0.646258533000946, 0.5374149680137634, 0.5442177057266235, 0.5986394286155701, 0.5714285969734192, 0.6734693646430969, 0.7142857313156128, 0.7210884094238281, 0.6938775777816772, 0.7142857313156128, 0.680272102355957, 0.7414966225624084, 0.6190476417541504, 0.8095238208770752, 0.7687074542045593, 0.7551020383834839, 0.8027210831642151, 0.761904776096344, 0.7346938848495483, 0.8095238208770752, 0.795918345451355, 0.8231292366981506, 0.8299319744110107, 0.8571428656578064, 0.9115646481513977, 0.918367326259613, 0.9115646481513977, 0.9591836929321289, 0.918367326259613, 0.9659863710403442, 0.884353756904602, 0.9319728016853333, 0.9727891087532043, 0.9251700639724731, 0.9319728016853333, 0.9659863710403442, 0.9659863710403442, 0.9455782175064087, 0.9387755393981934, 0.9863945841789246, 0.9727891087532043, 0.9659863710403442, 0.9931972622871399, 1.0], [0.12244898080825806, 0.095238097012043, 0.10204081982374191, 0.17006802558898926, 0.12244898080825806, 0.13605442643165588, 0.22448979318141937, 0.10204081982374191, 0.2448979616165161, 0.17006802558898926, 0.2380952388048172, 0.36734694242477417, 0.31292515993118286, 0.3877550959587097, 0.40816327929496765, 0.40136054158210754, 0.380952388048172, 0.40136054158210754, 0.4285714328289032, 0.4217686951160431, 0.4897959232330322, NaN, 0.5544217824935913, 0.5170068144798279, 0.4761904776096344, 0.5782312750816345, 0.5510203838348389, 0.5442177057266235, 0.5510203838348389, 0.5170068144798279, 0.557823121547699, 0.5170068144798279, 0.6190476417541504, 0.646258533000946, 0.7346938848495483, 0.5918367505073547, 0.680272102355957, 0.7482993006706238, 0.6190476417541504, 0.5714285969734192, 0.7823129296302795, 0.6666666865348816, 0.7823129296302795, 0.7346938848495483, 0.7278911471366882, 0.9387755393981934, 0.8435373902320862, 0.7346938848495483, 0.8435373902320862, 0.8775510191917419, 0.884353756904602, 0.884353756904602, 0.7823129296302795, 0.918367326259613, 0.8435373902320862, 0.9251700639724731, 0.9591836929321289, 0.9115646481513977, 0.9251700639724731, 0.9455782175064087, 0.9591836929321289, 0.9931972622871399, 0.9523809552192688, 0.9455782175064087, 0.9591836929321289, 0.9591836929321289, 0.9659863710403442, 0.9863945841789246, 0.9931972622871399, 0.9319728016853333, 0.9863945841789246, 1.0, 1.0], [0.11564625799655914, 0.095238097012043, 0.08163265138864517, 0.1428571492433548, 0.10204081982374191, 0.13605442643165588, 0.21768707036972046, 0.07482992857694626, 0.2312925159931183, 0.1428571492433548, 0.2448979616165161, 0.3469387888908386, 0.2857142984867096, 0.3877550959587097, 0.3741496503353119, 0.3877550959587097, 0.36054420471191406, 0.3741496503353119, 0.41496598720550537, 0.4217686951160431, 0.4693877696990967, 0.4455782175064087, NaN, 0.49659863114356995, 0.4897959232330322, 0.5374149680137634, 0.5510203838348389, 0.5374149680137634, 0.5374149680137634, 0.49659863114356995, 0.5442177057266235, 0.5102040767669678, 0.6258503198623657, 0.646258533000946, 0.761904776096344, 0.5986394286155701, 0.680272102355957, 0.7687074542045593, 0.6122449040412903, 0.5850340127944946, 0.8095238208770752, 0.6734693646430969, 0.7823129296302795, 0.7074829936027527, 0.7346938848495483, 0.9319728016853333, 0.8435373902320862, 0.7414966225624084, 0.8435373902320862, 0.8639456033706665, 0.8775510191917419, 0.8911564350128174, 0.7823129296302795, 0.918367326259613, 0.8435373902320862, 0.9251700639724731, 0.9523809552192688, 0.9047619104385376, 0.9115646481513977, 0.9455782175064087, 0.9591836929321289, 0.9931972622871399, 0.9591836929321289, 0.9523809552192688, 0.9591836929321289, 0.9523809552192688, 0.9591836929321289, 0.9863945841789246, 0.9931972622871399, 0.9319728016853333, 0.9863945841789246, 1.0, 1.0], [0.095238097012043, 0.11564625799655914, 0.10884353518486023, 0.095238097012043, 0.14965985715389252, 0.14965985715389252, 0.15646257996559143, 0.12244898080825806, 0.15646257996559143, 0.21088434755802155, 0.21088434755802155, 0.2857142984867096, 0.3401360511779785, 0.3265306055545807, 0.2789115607738495, 0.4217686951160431, 0.36734694242477417, 0.36054420471191406, 0.36734694242477417, 0.4217686951160431, 0.4217686951160431, 0.4829931855201721, 0.5034013390541077, NaN, 0.4897959232330322, 0.47959184646606445, 0.5102040767669678, 0.523809552192688, 0.5918367505073547, 0.49659863114356995, 0.5306122303009033, 0.557823121547699, 0.5918367505073547, 0.6054421663284302, 0.7142857313156128, 0.5850340127944946, 0.6870748400688171, 0.7278911471366882, 0.5646258592605591, 0.6190476417541504, 0.7755101919174194, 0.646258533000946, 0.7551020383834839, 0.7142857313156128, 0.7278911471366882, 0.8435373902320862, 0.8095238208770752, 0.7823129296302795, 0.8299319744110107, 0.9387755393981934, 0.9455782175064087, 0.9863945841789246, 0.7551020383834839, 0.884353756904602, 0.8503401279449463, 0.9795918464660645, 0.918367326259613, 0.8979591727256775, 0.9455782175064087, 0.884353756904602, 0.9251700639724731, 1.0, 0.9455782175064087, 0.9455782175064087, 0.9387755393981934, 0.9523809552192688, 0.9387755393981934, 1.0, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.08163265138864517, 0.12244898080825806, 0.11564625799655914, 0.08843537420034409, 0.07482992857694626, 0.20408163964748383, 0.14965985715389252, 0.1972789168357849, 0.12925170361995697, 0.2312925159931183, 0.25170066952705383, 0.2448979616165161, 0.3741496503353119, 0.20408163964748383, 0.3333333432674408, 0.36734694242477417, 0.4829931855201721, 0.35374149680137634, 0.4693877696990967, 0.5102040767669678, 0.5442177057266235, 0.523809552192688, 0.5102040767669678, 0.5102040767669678, NaN, 0.5306122303009033, 0.4761904776096344, 0.5646258592605591, 0.4897959232330322, 0.6190476417541504, 0.5102040767669678, 0.5306122303009033, 0.6734693646430969, 0.523809552192688, 0.646258533000946, 0.6734693646430969, 0.680272102355957, 0.6666666865348816, 0.7006802558898926, 0.6190476417541504, 0.7074829936027527, 0.6598639488220215, 0.795918345451355, 0.761904776096344, 0.680272102355957, 0.8299319744110107, 0.7687074542045593, 0.7755101919174194, 0.8163265585899353, 0.8299319744110107, 0.8911564350128174, 0.9047619104385376, 0.7755101919174194, 0.8571428656578064, 0.8503401279449463, 0.9319728016853333, 0.8639456033706665, 0.918367326259613, 0.9931972622871399, 0.8163265585899353, 0.884353756904602, 0.9931972622871399, 0.8639456033706665, 0.8571428656578064, 0.8911564350128174, 0.9115646481513977, 0.918367326259613, 0.9931972622871399, 1.0, 0.8979591727256775, 0.9727891087532043, 0.9931972622871399, 1.0], [0.10204081982374191, 0.08843537420034409, 0.10204081982374191, 0.10884353518486023, 0.11564625799655914, 0.12925170361995697, 0.16326530277729034, 0.12244898080825806, 0.1428571492433548, 0.1972789168357849, 0.20408163964748383, 0.30612245202064514, 0.3333333432674408, 0.380952388048172, 0.2925170063972473, 0.4217686951160431, 0.36734694242477417, 0.3469387888908386, 0.35374149680137634, 0.442176878452301, 0.4217686951160431, 0.4217686951160431, 0.4625850319862366, 0.5204081535339355, 0.4693877696990967, NaN, 0.4897959232330322, 0.5034013390541077, 0.5782312750816345, 0.4761904776096344, 0.5170068144798279, 0.5646258592605591, 0.5782312750816345, 0.6326530575752258, 0.6938775777816772, 0.5442177057266235, 0.6598639488220215, 0.7414966225624084, 0.5782312750816345, 0.6122449040412903, 0.7823129296302795, 0.6326530575752258, 0.7346938848495483, 0.6938775777816772, 0.7278911471366882, 0.8503401279449463, 0.8163265585899353, 0.761904776096344, 0.8163265585899353, 0.918367326259613, 0.9319728016853333, 0.9455782175064087, 0.7687074542045593, 0.8639456033706665, 0.8571428656578064, 0.9659863710403442, 0.9319728016853333, 0.884353756904602, 0.9455782175064087, 0.8911564350128174, 0.9251700639724731, 1.0, 0.9523809552192688, 0.9591836929321289, 0.9387755393981934, 0.9523809552192688, 0.9387755393981934, 1.0, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.08843537420034409, 0.10204081982374191, 0.08843537420034409, 0.17687074840068817, 0.16326530277729034, 0.14965985715389252, 0.25850340723991394, 0.1428571492433548, 0.26530611515045166, 0.15646257996559143, 0.25170066952705383, 0.36734694242477417, 0.2380952388048172, 0.44897958636283875, 0.35374149680137634, 0.3265306055545807, 0.40136054158210754, 0.4217686951160431, 0.45578232407569885, 0.442176878452301, 0.49659863114356995, 0.44897958636283875, 0.44897958636283875, 0.4897959232330322, 0.523809552192688, 0.5102040767669678, NaN, 0.523809552192688, 0.5102040767669678, 0.5034013390541077, 0.442176878452301, 0.5714285969734192, 0.5918367505073547, 0.7142857313156128, 0.7006802558898926, 0.5646258592605591, 0.6190476417541504, 0.6734693646430969, 0.5714285969734192, 0.5714285969734192, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.7823129296302795, 0.7414966225624084, 0.8231292366981506, 0.7755101919174194, 0.8571428656578064, 0.7278911471366882, 0.7482993006706238, 0.7755101919174194, 0.7142857313156128, 0.9387755393981934, 0.795918345451355, 0.8979591727256775, 0.8775510191917419, 0.8231292366981506, 0.8163265585899353, 0.9387755393981934, 0.9387755393981934, 0.9931972622871399, 0.9523809552192688, 0.9523809552192688, 0.9319728016853333, 0.9591836929321289, 0.9387755393981934, 0.9931972622871399, 1.0, 0.9727891087532043, 0.9863945841789246, 1.0, 1.0], [0.08163265138864517, 0.10884353518486023, 0.12244898080825806, 0.095238097012043, 0.21088434755802155, 0.18367347121238708, 0.16326530277729034, 0.17687074840068817, 0.21768707036972046, 0.2380952388048172, 0.16326530277729034, 0.36734694242477417, 0.380952388048172, 0.3945578336715698, 0.3401360511779785, 0.40136054158210754, 0.4217686951160431, 0.442176878452301, 0.36054420471191406, 0.4693877696990967, 0.41496598720550537, 0.45578232407569885, 0.4625850319862366, 0.4761904776096344, 0.4353741407394409, 0.49659863114356995, 0.4761904776096344, NaN, 0.4829931855201721, 0.523809552192688, 0.49659863114356995, 0.5714285969734192, 0.6326530575752258, 0.5918367505073547, 0.5102040767669678, 0.5986394286155701, 0.6734693646430969, 0.6598639488220215, 0.6394557952880859, 0.7278911471366882, 0.646258533000946, 0.5646258592605591, 0.6734693646430969, 0.7006802558898926, 0.6938775777816772, 0.7482993006706238, 0.7551020383834839, 0.6598639488220215, 0.7074829936027527, 0.7074829936027527, 0.8639456033706665, 0.884353756904602, 0.9591836929321289, 0.8503401279449463, 0.9115646481513977, 0.9387755393981934, 0.9727891087532043, 0.9659863710403442, 0.9931972622871399, 0.9591836929321289, 0.9659863710403442, 0.8979591727256775, 0.9659863710403442, 0.9659863710403442, 0.9115646481513977, 0.8911564350128174, 0.8911564350128174, 0.9659863710403442, 0.9931972622871399, 0.9863945841789246, 0.9863945841789246, 1.0, 1.0], [0.14965985715389252, 0.1428571492433548, 0.15646257996559143, 0.1972789168357849, 0.21088434755802155, 0.18367347121238708, 0.29931971430778503, 0.12925170361995697, 0.26530611515045166, 0.1972789168357849, 0.2925170063972473, 0.36734694242477417, 0.12244898080825806, 0.442176878452301, 0.3333333432674408, 0.25850340723991394, 0.35374149680137634, 0.380952388048172, 0.45578232407569885, 0.36734694242477417, 0.45578232407569885, 0.44897958636283875, 0.4625850319862366, 0.40816327929496765, 0.5102040767669678, 0.4217686951160431, 0.4897959232330322, 0.5170068144798279, NaN, 0.4625850319862366, 0.4625850319862366, 0.5510203838348389, 0.5646258592605591, 0.646258533000946, 0.646258533000946, 0.523809552192688, 0.5782312750816345, 0.6190476417541504, 0.523809552192688, 0.49659863114356995, 0.7210884094238281, 0.7346938848495483, 0.6598639488220215, 0.5782312750816345, 0.7755101919174194, 0.7074829936027527, 0.761904776096344, 0.8027210831642151, 0.8435373902320862, 0.646258533000946, 0.7346938848495483, 0.7346938848495483, 0.6870748400688171, 0.9115646481513977, 0.761904776096344, 0.8163265585899353, 0.8367347121238708, 0.8095238208770752, 0.8027210831642151, 0.9115646481513977, 0.9319728016853333, 0.9863945841789246, 0.9251700639724731, 0.9319728016853333, 0.9387755393981934, 0.9319728016853333, 0.9251700639724731, 0.9795918464660645, 1.0, 0.9591836929321289, 0.9795918464660645, 1.0, 1.0], [0.07482992857694626, 0.13605442643165588, 0.15646257996559143, 0.10204081982374191, 0.15646257996559143, 0.1972789168357849, 0.095238097012043, 0.21088434755802155, 0.18367347121238708, 0.21768707036972046, 0.11564625799655914, 0.26530611515045166, 0.4217686951160431, 0.2448979616165161, 0.40816327929496765, 0.4285714328289032, 0.36734694242477417, 0.3265306055545807, 0.27210885286331177, 0.3877550959587097, 0.35374149680137634, 0.4829931855201721, 0.5034013390541077, 0.5034013390541077, 0.380952388048172, 0.523809552192688, 0.49659863114356995, 0.4761904776096344, 0.5374149680137634, NaN, 0.4761904776096344, 0.5170068144798279, 0.5306122303009033, 0.44897958636283875, 0.5306122303009033, 0.7176870703697205, 0.557823121547699, 0.6122449040412903, 0.7687074542045593, 0.6734693646430969, 0.6054421663284302, 0.5306122303009033, 0.6122449040412903, 0.8503401279449463, 0.5918367505073547, 0.7006802558898926, 0.7006802558898926, 0.6122449040412903, 0.6666666865348816, 0.7891156673431396, 0.8095238208770752, 0.8163265585899353, 0.8095238208770752, 0.7755101919174194, 0.9523809552192688, 0.8707482814788818, 0.9455782175064087, 0.9659863710403442, 1.0, 0.761904776096344, 0.761904776096344, 0.9115646481513977, 0.9115646481513977, 0.9115646481513977, 0.9251700639724731, 0.9523809552192688, 0.9387755393981934, 0.8979591727256775, 0.9795918464660645, 1.0, 0.9319728016853333, 0.9455782175064087, 1.0], [0.2448979616165161, 0.25170066952705383, 0.25850340723991394, 0.3741496503353119, 0.3469387888908386, 0.2857142984867096, 0.4285714328289032, 0.3469387888908386, 0.4897959232330322, 0.36054420471191406, 0.41496598720550537, 0.5102040767669678, 0.36054420471191406, 0.5034013390541077, 0.4897959232330322, 0.4285714328289032, 0.4897959232330322, 0.523809552192688, 0.4693877696990967, 0.5102040767669678, 0.4625850319862366, 0.442176878452301, 0.45578232407569885, 0.4693877696990967, 0.4897959232330322, 0.4829931855201721, 0.557823121547699, 0.5034013390541077, 0.5374149680137634, 0.523809552192688, NaN, 0.523809552192688, 0.5306122303009033, 0.6870748400688171, 0.5782312750816345, 0.5306122303009033, 0.5714285969734192, 0.5442177057266235, 0.523809552192688, 0.557823121547699, 0.6054421663284302, 0.6530612111091614, 0.5782312750816345, 0.5510203838348389, 0.7006802558898926, 0.5374149680137634, 0.6326530575752258, 0.7210884094238281, 0.7414966225624084, 0.5918367505073547, 0.5306122303009033, 0.5102040767669678, 0.5850340127944946, 0.7142857313156128, 0.5646258592605591, 0.6394557952880859, 0.6530612111091614, 0.6530612111091614, 0.6326530575752258, 0.8027210831642151, 0.8027210831642151, 0.7210884094238281, 0.7346938848495483, 0.7210884094238281, 0.7551020383834839, 0.7482993006706238, 0.7482993006706238, 0.7891156673431396, 0.7482993006706238, 0.761904776096344, 0.9455782175064087, 0.8095238208770752, 1.0], [0.095238097012043, 0.20408163964748383, 0.22448979318141937, 0.16326530277729034, 0.22448979318141937, 0.29931971430778503, 0.18367347121238708, 0.3945578336715698, 0.25170066952705383, 0.380952388048172, 0.3945578336715698, 0.25170066952705383, 0.40136054158210754, 0.21088434755802155, 0.40136054158210754, 0.3945578336715698, 0.4285714328289032, 0.26530611515045166, 0.44897958636283875, 0.442176878452301, 0.45578232407569885, 0.4829931855201721, 0.4897959232330322, 0.442176878452301, 0.4693877696990967, 0.4353741407394409, 0.4285714328289032, 0.4285714328289032, 0.44897958636283875, 0.4829931855201721, 0.4761904776096344, NaN, 0.523809552192688, 0.45578232407569885, 0.4693877696990967, 0.5170068144798279, 0.49659863114356995, 0.45578232407569885, 0.5714285969734192, 0.5986394286155701, 0.49659863114356995, 0.557823121547699, 0.5782312750816345, 0.6258503198623657, 0.523809552192688, 0.557823121547699, 0.5034013390541077, 0.6122449040412903, 0.5714285969734192, 0.7346938848495483, 0.6598639488220215, 0.6326530575752258, 0.6938775777816772, 0.6326530575752258, 0.8503401279449463, 0.8231292366981506, 0.8095238208770752, 0.8027210831642151, 0.918367326259613, 0.5782312750816345, 0.646258533000946, 0.918367326259613, 0.761904776096344, 0.7482993006706238, 0.8503401279449463, 0.8707482814788818, 0.884353756904602, 0.9047619104385376, 1.0, 0.8775510191917419, 0.8911564350128174, 0.9319728016853333, 1.0], [0.06802721321582794, 0.0476190485060215, 0.0476190485060215, 0.06122449040412903, 0.054421767592430115, 0.10204081982374191, 0.08843537420034409, 0.095238097012043, 0.12244898080825806, 0.11564625799655914, 0.17006802558898926, 0.21088434755802155, 0.25850340723991394, 0.1972789168357849, 0.25850340723991394, 0.31972789764404297, 0.3265306055545807, 0.31972789764404297, 0.3265306055545807, 0.35374149680137634, 0.40136054158210754, 0.380952388048172, 0.3741496503353119, 0.40816327929496765, 0.3265306055545807, 0.4217686951160431, 0.40816327929496765, 0.36734694242477417, 0.4353741407394409, 0.4693877696990967, 0.4693877696990967, 0.4761904776096344, NaN, 0.4217686951160431, 0.4897959232330322, 0.5714285969734192, 0.5306122303009033, 0.5442177057266235, 0.5918367505073547, 0.5170068144798279, 0.5918367505073547, 0.6258503198623657, 0.646258533000946, 0.7142857313156128, 0.6122449040412903, 0.6734693646430969, 0.6666666865348816, 0.6734693646430969, 0.7278911471366882, 0.7074829936027527, 0.7755101919174194, 0.7346938848495483, 0.7278911471366882, 0.8503401279449463, 0.761904776096344, 0.8367347121238708, 0.8639456033706665, 0.8639456033706665, 0.9251700639724731, 0.7687074542045593, 0.8299319744110107, 0.9795918464660645, 0.8435373902320862, 0.8435373902320862, 0.8911564350128174, 0.9115646481513977, 0.9047619104385376, 0.9387755393981934, 0.9727891087532043, 0.884353756904602, 0.9455782175064087, 0.9727891087532043, 1.0], [0.07482992857694626, 0.054421767592430115, 0.06122449040412903, 0.095238097012043, 0.08163265138864517, 0.10204081982374191, 0.12244898080825806, 0.1972789168357849, 0.17687074840068817, 0.14965985715389252, 0.29931971430778503, 0.31972789764404297, 0.22448979318141937, 0.29931971430778503, 0.25850340723991394, 0.25170066952705383, 0.36054420471191406, 0.36734694242477417, 0.40816327929496765, 0.3469387888908386, 0.4285714328289032, 0.35374149680137634, 0.35374149680137634, 0.3945578336715698, 0.4761904776096344, 0.36734694242477417, 0.2857142984867096, 0.40816327929496765, 0.35374149680137634, 0.5510203838348389, 0.31292515993118286, 0.5442177057266235, 0.5782312750816345, NaN, 0.380952388048172, 0.5782312750816345, 0.5170068144798279, 0.5170068144798279, 0.5850340127944946, 0.5170068144798279, 0.4829931855201721, 0.6938775777816772, 0.6394557952880859, 0.5918367505073547, 0.8129251599311829, 0.6190476417541504, 0.5714285969734192, 0.795918345451355, 0.6938775777816772, 0.6394557952880859, 0.6530612111091614, 0.6530612111091614, 0.6530612111091614, 0.8911564350128174, 0.7278911471366882, 0.6870748400688171, 0.761904776096344, 0.7414966225624084, 0.7891156673431396, 0.8707482814788818, 0.918367326259613, 0.9115646481513977, 0.8979591727256775, 0.9047619104385376, 0.8367347121238708, 0.8775510191917419, 0.8707482814788818, 0.9591836929321289, 0.9931972622871399, 0.9387755393981934, 0.9727891087532043, 0.9931972622871399, 1.0], [0.095238097012043, 0.0476190485060215, 0.06802721321582794, 0.08163265138864517, 0.13605442643165588, 0.095238097012043, 0.17006802558898926, 0.08163265138864517, 0.17006802558898926, 0.10204081982374191, 0.17687074840068817, 0.26530611515045166, 0.13605442643165588, 0.3333333432674408, 0.22448979318141937, 0.1972789168357849, 0.25850340723991394, 0.3741496503353119, 0.30612245202064514, 0.27210885286331177, 0.3265306055545807, 0.26530611515045166, 0.2380952388048172, 0.2857142984867096, 0.35374149680137634, 0.30612245202064514, 0.29931971430778503, 0.4897959232330322, 0.35374149680137634, 0.4693877696990967, 0.4217686951160431, 0.5306122303009033, 0.5102040767669678, 0.6190476417541504, NaN, 0.5102040767669678, 0.5986394286155701, 0.4761904776096344, 0.5102040767669678, 0.5170068144798279, 0.7585033774375916, 0.557823121547699, 0.6190476417541504, 0.5510203838348389, 0.6870748400688171, 0.5986394286155701, 0.8435373902320862, 0.6734693646430969, 0.7346938848495483, 0.5918367505073547, 0.6938775777816772, 0.7074829936027527, 0.7346938848495483, 0.918367326259613, 0.7482993006706238, 0.8299319744110107, 0.8707482814788818, 0.7891156673431396, 0.8027210831642151, 0.9387755393981934, 0.9727891087532043, 0.9455782175064087, 0.9319728016853333, 0.9387755393981934, 0.918367326259613, 0.9319728016853333, 0.9319728016853333, 0.9795918464660645, 0.9931972622871399, 0.9387755393981934, 0.9863945841789246, 0.9931972622871399, 1.0], [0.095238097012043, 0.12244898080825806, 0.12925170361995697, 0.095238097012043, 0.17687074840068817, 0.17687074840068817, 0.12244898080825806, 0.13605442643165588, 0.12925170361995697, 0.16326530277729034, 0.095238097012043, 0.20408163964748383, 0.36054420471191406, 0.21768707036972046, 0.3469387888908386, 0.35374149680137634, 0.30612245202064514, 0.29931971430778503, 0.1972789168357849, 0.3401360511779785, 0.2857142984867096, 0.40816327929496765, 0.40136054158210754, 0.41496598720550537, 0.3265306055545807, 0.45578232407569885, 0.4353741407394409, 0.40136054158210754, 0.4761904776096344, 0.28231292963027954, 0.4693877696990967, 0.4829931855201721, 0.4285714328289032, 0.4217686951160431, 0.4897959232330322, NaN, 0.523809552192688, 0.5170068144798279, 0.5510203838348389, 0.5986394286155701, 0.5442177057266235, 0.4761904776096344, 0.5306122303009033, 0.7993197441101074, 0.5306122303009033, 0.6326530575752258, 0.646258533000946, 0.5170068144798279, 0.5918367505073547, 0.7278911471366882, 0.7687074542045593, 0.761904776096344, 0.7482993006706238, 0.7278911471366882, 0.9047619104385376, 0.8503401279449463, 0.8979591727256775, 0.9863945841789246, 0.9523809552192688, 0.7074829936027527, 0.7142857313156128, 0.8435373902320862, 0.9251700639724731, 0.9251700639724731, 0.9115646481513977, 0.9319728016853333, 0.9319728016853333, 0.884353756904602, 0.9455782175064087, 0.9931972622871399, 0.9387755393981934, 0.918367326259613, 1.0], [0.07482992857694626, 0.08163265138864517, 0.08163265138864517, 0.06122449040412903, 0.14965985715389252, 0.12925170361995697, 0.095238097012043, 0.14965985715389252, 0.18367347121238708, 0.17006802558898926, 0.10204081982374191, 0.25850340723991394, 0.29931971430778503, 0.3265306055545807, 0.27210885286331177, 0.30612245202064514, 0.2380952388048172, 0.3265306055545807, 0.20408163964748383, 0.25170066952705383, 0.2789115607738495, 0.31972789764404297, 0.31972789764404297, 0.31292515993118286, 0.31972789764404297, 0.3401360511779785, 0.380952388048172, 0.3265306055545807, 0.4217686951160431, 0.442176878452301, 0.4285714328289032, 0.5034013390541077, 0.4693877696990967, 0.4829931855201721, 0.40136054158210754, 0.4761904776096344, NaN, 0.442176878452301, 0.5374149680137634, 0.6258503198623657, 0.5102040767669678, 0.49659863114356995, 0.5306122303009033, 0.5850340127944946, 0.5918367505073547, 0.5646258592605591, 0.6190476417541504, 0.5374149680137634, 0.557823121547699, 0.6326530575752258, 0.7414966225624084, 0.7142857313156128, 0.918367326259613, 0.795918345451355, 0.8707482814788818, 0.884353756904602, 0.9455782175064087, 0.918367326259613, 0.9455782175064087, 0.8707482814788818, 0.8775510191917419, 0.8571428656578064, 0.9251700639724731, 0.9319728016853333, 0.8911564350128174, 0.918367326259613, 0.9047619104385376, 0.9251700639724731, 0.9931972622871399, 0.9727891087532043, 0.9795918464660645, 0.9931972622871399, 1.0], [0.06802721321582794, 0.03401360660791397, 0.06122449040412903, 0.054421767592430115, 0.0476190485060215, 0.08843537420034409, 0.11564625799655914, 0.06802721321582794, 0.11564625799655914, 0.10884353518486023, 0.12244898080825806, 0.21088434755802155, 0.17006802558898926, 0.25850340723991394, 0.18367347121238708, 0.2380952388048172, 0.17687074840068817, 0.29931971430778503, 0.2857142984867096, 0.2312925159931183, 0.30612245202064514, 0.25170066952705383, 0.2312925159931183, 0.27210885286331177, 0.3333333432674408, 0.25850340723991394, 0.3265306055545807, 0.3401360511779785, 0.380952388048172, 0.3877550959587097, 0.45578232407569885, 0.5442177057266235, 0.45578232407569885, 0.4829931855201721, 0.523809552192688, 0.4829931855201721, 0.557823121547699, NaN, 0.4897959232330322, 0.4897959232330322, 0.6666666865348816, 0.5646258592605591, 0.6054421663284302, 0.6054421663284302, 0.6326530575752258, 0.6734693646430969, 0.7006802558898926, 0.6666666865348816, 0.7074829936027527, 0.7278911471366882, 0.7891156673431396, 0.8027210831642151, 0.7006802558898926, 0.8571428656578064, 0.795918345451355, 0.9387755393981934, 0.8911564350128174, 0.8435373902320862, 0.884353756904602, 0.8367347121238708, 0.8639456033706665, 1.0, 0.918367326259613, 0.9115646481513977, 0.918367326259613, 0.9455782175064087, 0.918367326259613, 0.9659863710403442, 1.0, 0.9455782175064087, 0.9931972622871399, 1.0, 1.0], [0.08163265138864517, 0.10884353518486023, 0.12925170361995697, 0.10204081982374191, 0.16326530277729034, 0.15646257996559143, 0.08843537420034409, 0.1428571492433548, 0.16326530277729034, 0.17687074840068817, 0.095238097012043, 0.21088434755802155, 0.36734694242477417, 0.2312925159931183, 0.31972789764404297, 0.3877550959587097, 0.2789115607738495, 0.2789115607738495, 0.1972789168357849, 0.31292515993118286, 0.2857142984867096, 0.380952388048172, 0.3877550959587097, 0.4353741407394409, 0.29931971430778503, 0.4217686951160431, 0.4285714328289032, 0.36054420471191406, 0.4761904776096344, 0.2312925159931183, 0.4761904776096344, 0.4285714328289032, 0.40816327929496765, 0.41496598720550537, 0.4897959232330322, 0.44897958636283875, 0.4625850319862366, 0.5102040767669678, NaN, 0.6122449040412903, 0.5646258592605591, 0.4897959232330322, 0.4829931855201721, 0.6360543966293335, 0.5170068144798279, 0.5510203838348389, 0.6054421663284302, 0.5374149680137634, 0.5646258592605591, 0.7346938848495483, 0.680272102355957, 0.6870748400688171, 0.7142857313156128, 0.7006802558898926, 0.9319728016853333, 0.8503401279449463, 0.9115646481513977, 0.8911564350128174, 0.884353756904602, 0.7074829936027527, 0.6870748400688171, 0.8503401279449463, 0.9115646481513977, 0.9047619104385376, 0.9115646481513977, 0.9115646481513977, 0.918367326259613, 0.884353756904602, 0.9659863710403442, 1.0, 0.9319728016853333, 0.9251700639724731, 1.0], [0.08163265138864517, 0.027210883796215057, 0.027210883796215057, 0.013605441898107529, 0.0476190485060215, 0.06802721321582794, 0.054421767592430115, 0.21088434755802155, 0.08163265138864517, 0.21088434755802155, 0.190476194024086, 0.13605442643165588, 0.31292515993118286, 0.17687074840068817, 0.2925170063972473, 0.3333333432674408, 0.27210885286331177, 0.17687074840068817, 0.3469387888908386, 0.31972789764404297, 0.31972789764404297, 0.4285714328289032, 0.41496598720550537, 0.380952388048172, 0.380952388048172, 0.3877550959587097, 0.4285714328289032, 0.27210885286331177, 0.5034013390541077, 0.3265306055545807, 0.442176878452301, 0.40136054158210754, 0.4829931855201721, 0.4829931855201721, 0.4829931855201721, 0.40136054158210754, 0.3741496503353119, 0.5102040767669678, 0.3877550959587097, NaN, 0.5442177057266235, 0.6394557952880859, 0.523809552192688, 0.5646258592605591, 0.5986394286155701, 0.5442177057266235, 0.5782312750816345, 0.6598639488220215, 0.6598639488220215, 0.7482993006706238, 0.6122449040412903, 0.6054421663284302, 0.5986394286155701, 0.680272102355957, 0.8435373902320862, 0.7823129296302795, 0.9251700639724731, 0.8571428656578064, 0.761904776096344, 0.6598639488220215, 0.6938775777816772, 0.8367347121238708, 0.8707482814788818, 0.8571428656578064, 0.9319728016853333, 0.9455782175064087, 0.9455782175064087, 0.8707482814788818, 0.9523809552192688, 0.9523809552192688, 0.9319728016853333, 0.9659863710403442, 1.0], [0.07482992857694626, 0.040816325694322586, 0.06122449040412903, 0.054421767592430115, 0.095238097012043, 0.08843537420034409, 0.14965985715389252, 0.07482992857694626, 0.1428571492433548, 0.06122449040412903, 0.12244898080825806, 0.2312925159931183, 0.10884353518486023, 0.27210885286331177, 0.1972789168357849, 0.1428571492433548, 0.17006802558898926, 0.30612245202064514, 0.2380952388048172, 0.22448979318141937, 0.25850340723991394, 0.21768707036972046, 0.190476194024086, 0.22448979318141937, 0.2925170063972473, 0.21768707036972046, 0.190476194024086, 0.35374149680137634, 0.2789115607738495, 0.3945578336715698, 0.3945578336715698, 0.5034013390541077, 0.40816327929496765, 0.5170068144798279, 0.24149659276008606, 0.45578232407569885, 0.4897959232330322, 0.3333333432674408, 0.4353741407394409, 0.45578232407569885, NaN, 0.5102040767669678, 0.5170068144798279, 0.523809552192688, 0.6326530575752258, 0.5102040767669678, 0.7006802558898926, 0.6258503198623657, 0.646258533000946, 0.49659863114356995, 0.6326530575752258, 0.6598639488220215, 0.6870748400688171, 0.8775510191917419, 0.6938775777816772, 0.7823129296302795, 0.8163265585899353, 0.7551020383834839, 0.7755101919174194, 0.8911564350128174, 0.9251700639724731, 0.9047619104385376, 0.918367326259613, 0.9251700639724731, 0.8979591727256775, 0.9115646481513977, 0.9047619104385376, 0.9659863710403442, 0.9931972622871399, 0.9387755393981934, 0.9795918464660645, 0.9931972622871399, 1.0], [0.10204081982374191, 0.07482992857694626, 0.06802721321582794, 0.12925170361995697, 0.054421767592430115, 0.08163265138864517, 0.17006802558898926, 0.095238097012043, 0.18367347121238708, 0.12925170361995697, 0.2312925159931183, 0.2312925159931183, 0.18367347121238708, 0.22448979318141937, 0.17687074840068817, 0.21768707036972046, 0.31292515993118286, 0.30612245202064514, 0.3333333432674408, 0.3265306055545807, 0.380952388048172, 0.3333333432674408, 0.3265306055545807, 0.35374149680137634, 0.3401360511779785, 0.36734694242477417, 0.2857142984867096, 0.4353741407394409, 0.26530611515045166, 0.4693877696990967, 0.3469387888908386, 0.442176878452301, 0.3741496503353119, 0.30612245202064514, 0.442176878452301, 0.523809552192688, 0.5034013390541077, 0.4353741407394409, 0.5102040767669678, 0.36054420471191406, 0.4897959232330322, NaN, 0.5986394286155701, 0.5374149680137634, 0.5034013390541077, 0.557823121547699, 0.5102040767669678, 0.7551020383834839, 0.5918367505073547, 0.5850340127944946, 0.6870748400688171, 0.6734693646430969, 0.557823121547699, 0.6734693646430969, 0.6326530575752258, 0.7278911471366882, 0.6326530575752258, 0.646258533000946, 0.7823129296302795, 0.8095238208770752, 0.8571428656578064, 0.7891156673431396, 0.7687074542045593, 0.7687074542045593, 0.680272102355957, 0.7210884094238281, 0.7210884094238281, 0.9931972622871399, 0.918367326259613, 0.8911564350128174, 0.9523809552192688, 1.0, 0.9931972622871399], [0.020408162847161293, 0.054421767592430115, 0.06802721321582794, 0.03401360660791397, 0.06802721321582794, 0.095238097012043, 0.03401360660791397, 0.054421767592430115, 0.0476190485060215, 0.08163265138864517, 0.06122449040412903, 0.14965985715389252, 0.190476194024086, 0.17006802558898926, 0.18367347121238708, 0.22448979318141937, 0.17687074840068817, 0.2312925159931183, 0.095238097012043, 0.21088434755802155, 0.190476194024086, 0.21768707036972046, 0.21768707036972046, 0.2448979616165161, 0.20408163964748383, 0.26530611515045166, 0.2857142984867096, 0.3265306055545807, 0.3401360511779785, 0.3877550959587097, 0.4217686951160431, 0.4217686951160431, 0.35374149680137634, 0.36054420471191406, 0.380952388048172, 0.4693877696990967, 0.4693877696990967, 0.3945578336715698, 0.5170068144798279, 0.4761904776096344, 0.4829931855201721, 0.40136054158210754, NaN, 0.6326530575752258, 0.4829931855201721, 0.5034013390541077, 0.5714285969734192, 0.4829931855201721, 0.5034013390541077, 0.6530612111091614, 0.7414966225624084, 0.761904776096344, 0.6666666865348816, 0.7278911471366882, 0.8163265585899353, 0.8503401279449463, 0.8163265585899353, 0.8503401279449463, 0.8367347121238708, 0.7687074542045593, 0.8095238208770752, 0.8367347121238708, 0.7823129296302795, 0.7823129296302795, 0.8775510191917419, 0.9047619104385376, 0.8979591727256775, 0.8571428656578064, 0.9251700639724731, 0.8979591727256775, 0.8639456033706665, 0.9795918464660645, 1.0], [0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.13605442643165588, 0.08843537420034409, 0.10204081982374191, 0.08843537420034409, 0.12925170361995697, 0.1428571492433548, 0.10204081982374191, 0.16326530277729034, 0.2448979616165161, 0.21088434755802155, 0.2448979616165161, 0.2789115607738495, 0.17687074840068817, 0.25170066952705383, 0.18367347121238708, 0.20408163964748383, 0.2312925159931183, 0.26530611515045166, 0.2925170063972473, 0.2857142984867096, 0.2380952388048172, 0.30612245202064514, 0.380952388048172, 0.29931971430778503, 0.4217686951160431, 0.14965985715389252, 0.44897958636283875, 0.3741496503353119, 0.2857142984867096, 0.40816327929496765, 0.44897958636283875, 0.20068027079105377, 0.41496598720550537, 0.3945578336715698, 0.3639455735683441, 0.4353741407394409, 0.4761904776096344, 0.4625850319862366, 0.36734694242477417, NaN, 0.4625850319862366, 0.4761904776096344, 0.5850340127944946, 0.5102040767669678, 0.5170068144798279, 0.646258533000946, 0.5986394286155701, 0.6054421663284302, 0.6190476417541504, 0.6598639488220215, 0.8435373902320862, 0.795918345451355, 0.8571428656578064, 0.819727897644043, 0.7891156673431396, 0.6734693646430969, 0.6666666865348816, 0.795918345451355, 0.8775510191917419, 0.8707482814788818, 0.8911564350128174, 0.918367326259613, 0.9115646481513977, 0.8707482814788818, 0.9251700639724731, 0.9727891087532043, 0.918367326259613, 0.8979591727256775, 1.0], [0.06802721321582794, 0.054421767592430115, 0.054421767592430115, 0.0476190485060215, 0.08163265138864517, 0.10204081982374191, 0.10884353518486023, 0.1428571492433548, 0.12244898080825806, 0.10204081982374191, 0.18367347121238708, 0.190476194024086, 0.12925170361995697, 0.2312925159931183, 0.17006802558898926, 0.16326530277729034, 0.21088434755802155, 0.27210885286331177, 0.2380952388048172, 0.2312925159931183, 0.2448979616165161, 0.27210885286331177, 0.26530611515045166, 0.27210885286331177, 0.31972789764404297, 0.27210885286331177, 0.21768707036972046, 0.30612245202064514, 0.22448979318141937, 0.40816327929496765, 0.29931971430778503, 0.4761904776096344, 0.3877550959587097, 0.18707482516765594, 0.31292515993118286, 0.4693877696990967, 0.40816327929496765, 0.36734694242477417, 0.4829931855201721, 0.40136054158210754, 0.36734694242477417, 0.49659863114356995, 0.5170068144798279, 0.5374149680137634, NaN, 0.49659863114356995, 0.380952388048172, 0.6394557952880859, 0.5102040767669678, 0.5102040767669678, 0.557823121547699, 0.557823121547699, 0.5986394286155701, 0.8095238208770752, 0.6870748400688171, 0.6598639488220215, 0.680272102355957, 0.7074829936027527, 0.7346938848495483, 0.7346938848495483, 0.7891156673431396, 0.8571428656578064, 0.8639456033706665, 0.8571428656578064, 0.7823129296302795, 0.8299319744110107, 0.8163265585899353, 0.9251700639724731, 0.9659863710403442, 0.9387755393981934, 0.9523809552192688, 0.9863945841789246, 1.0], [0.06802721321582794, 0.020408162847161293, 0.013605441898107529, 0.006802720949053764, 0.020408162847161293, 0.054421767592430115, 0.020408162847161293, 0.006802720949053764, 0.07482992857694626, 0.040816325694322586, 0.054421767592430115, 0.10204081982374191, 0.11564625799655914, 0.13605442643165588, 0.16326530277729034, 0.15646257996559143, 0.054421767592430115, 0.1972789168357849, 0.12925170361995697, 0.11564625799655914, 0.1972789168357849, 0.06122449040412903, 0.06802721321582794, 0.15646257996559143, 0.17006802558898926, 0.14965985715389252, 0.25850340723991394, 0.25170066952705383, 0.2925170063972473, 0.29931971430778503, 0.4625850319862366, 0.442176878452301, 0.3265306055545807, 0.380952388048172, 0.40136054158210754, 0.36734694242477417, 0.4353741407394409, 0.3265306055545807, 0.44897958636283875, 0.45578232407569885, 0.4897959232330322, 0.442176878452301, 0.49659863114356995, 0.523809552192688, 0.5034013390541077, NaN, 0.6122449040412903, 0.523809552192688, 0.5782312750816345, 0.680272102355957, 0.646258533000946, 0.680272102355957, 0.6734693646430969, 0.7551020383834839, 0.8027210831642151, 0.8639456033706665, 0.884353756904602, 0.8639456033706665, 0.8639456033706665, 0.7210884094238281, 0.761904776096344, 0.9591836929321289, 0.8775510191917419, 0.8639456033706665, 0.8911564350128174, 0.9251700639724731, 0.918367326259613, 0.9455782175064087, 0.9863945841789246, 0.9251700639724731, 0.9863945841789246, 1.0, 1.0], [0.07482992857694626, 0.054421767592430115, 0.06802721321582794, 0.06122449040412903, 0.07482992857694626, 0.07482992857694626, 0.08163265138864517, 0.06802721321582794, 0.095238097012043, 0.08163265138864517, 0.10204081982374191, 0.20408163964748383, 0.10884353518486023, 0.2312925159931183, 0.13605442643165588, 0.12925170361995697, 0.12925170361995697, 0.27210885286331177, 0.20408163964748383, 0.15646257996559143, 0.2380952388048172, 0.15646257996559143, 0.15646257996559143, 0.190476194024086, 0.2312925159931183, 0.18367347121238708, 0.17687074840068817, 0.2448979616165161, 0.2380952388048172, 0.29931971430778503, 0.36734694242477417, 0.49659863114356995, 0.3333333432674408, 0.4285714328289032, 0.15646257996559143, 0.35374149680137634, 0.380952388048172, 0.29931971430778503, 0.3945578336715698, 0.4217686951160431, 0.29931971430778503, 0.4897959232330322, 0.4285714328289032, 0.41496598720550537, 0.6190476417541504, 0.3877550959587097, NaN, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.6054421663284302, 0.6258503198623657, 0.6190476417541504, 0.7687074542045593, 0.7551020383834839, 0.7482993006706238, 0.7891156673431396, 0.7210884094238281, 0.7551020383834839, 0.7823129296302795, 0.8571428656578064, 0.8639456033706665, 0.9047619104385376, 0.9115646481513977, 0.8775510191917419, 0.8979591727256775, 0.8707482814788818, 0.9659863710403442, 0.9795918464660645, 0.9727891087532043, 0.9727891087532043, 1.0, 1.0], [0.06122449040412903, 0.03401360660791397, 0.0476190485060215, 0.06122449040412903, 0.020408162847161293, 0.06122449040412903, 0.06122449040412903, 0.08843537420034409, 0.10204081982374191, 0.06802721321582794, 0.11564625799655914, 0.18367347121238708, 0.12244898080825806, 0.13605442643165588, 0.10204081982374191, 0.16326530277729034, 0.17687074840068817, 0.22448979318141937, 0.1972789168357849, 0.21768707036972046, 0.26530611515045166, 0.26530611515045166, 0.25850340723991394, 0.21768707036972046, 0.22448979318141937, 0.2380952388048172, 0.22448979318141937, 0.3401360511779785, 0.1972789168357849, 0.3877550959587097, 0.2789115607738495, 0.3877550959587097, 0.3265306055545807, 0.20408163964748383, 0.3265306055545807, 0.4829931855201721, 0.4625850319862366, 0.3333333432674408, 0.4625850319862366, 0.3401360511779785, 0.3741496503353119, 0.2448979616165161, 0.5170068144798279, 0.4897959232330322, 0.36054420471191406, 0.4761904776096344, 0.4285714328289032, NaN, 0.4829931855201721, 0.5306122303009033, 0.6258503198623657, 0.5986394286155701, 0.5646258592605591, 0.646258533000946, 0.6598639488220215, 0.7006802558898926, 0.6190476417541504, 0.6326530575752258, 0.7551020383834839, 0.7482993006706238, 0.795918345451355, 0.7891156673431396, 0.7346938848495483, 0.7482993006706238, 0.6598639488220215, 0.7006802558898926, 0.7074829936027527, 0.9863945841789246, 0.8299319744110107, 0.8911564350128174, 0.9387755393981934, 1.0, 0.9931972622871399], [0.020408162847161293, 0.006802720949053764, 0.013605441898107529, 0.020408162847161293, 0.03401360660791397, 0.006802720949053764, 0.013605441898107529, 0.006802720949053764, 0.07482992857694626, 0.020408162847161293, 0.020408162847161293, 0.1428571492433548, 0.0476190485060215, 0.15646257996559143, 0.12925170361995697, 0.08843537420034409, 0.054421767592430115, 0.25170066952705383, 0.11564625799655914, 0.095238097012043, 0.190476194024086, 0.15646257996559143, 0.15646257996559143, 0.17006802558898926, 0.18367347121238708, 0.18367347121238708, 0.1428571492433548, 0.2925170063972473, 0.15646257996559143, 0.3333333432674408, 0.25850340723991394, 0.4285714328289032, 0.27210885286331177, 0.30612245202064514, 0.26530611515045166, 0.40816327929496765, 0.442176878452301, 0.2925170063972473, 0.4353741407394409, 0.3401360511779785, 0.35374149680137634, 0.40816327929496765, 0.49659863114356995, 0.4829931855201721, 0.4897959232330322, 0.4217686951160431, 0.4761904776096344, 0.5170068144798279, NaN, 0.5102040767669678, 0.5442177057266235, 0.5646258592605591, 0.557823121547699, 0.7414966225624084, 0.6870748400688171, 0.6530612111091614, 0.6666666865348816, 0.7278911471366882, 0.7482993006706238, 0.7823129296302795, 0.8435373902320862, 0.8299319744110107, 0.8367347121238708, 0.8367347121238708, 0.8299319744110107, 0.8911564350128174, 0.8435373902320862, 0.9319728016853333, 0.9251700639724731, 0.9251700639724731, 0.9387755393981934, 0.9931972622871399, 1.0], [0.07482992857694626, 0.054421767592430115, 0.06802721321582794, 0.027210883796215057, 0.06802721321582794, 0.08843537420034409, 0.07482992857694626, 0.07482992857694626, 0.07482992857694626, 0.08163265138864517, 0.06122449040412903, 0.1428571492433548, 0.11564625799655914, 0.14965985715389252, 0.1428571492433548, 0.21088434755802155, 0.11564625799655914, 0.15646257996559143, 0.1972789168357849, 0.15646257996559143, 0.20408163964748383, 0.12244898080825806, 0.13605442643165588, 0.06122449040412903, 0.17006802558898926, 0.08163265138864517, 0.27210885286331177, 0.2925170063972473, 0.35374149680137634, 0.21088434755802155, 0.40816327929496765, 0.26530611515045166, 0.2925170063972473, 0.36054420471191406, 0.40816327929496765, 0.27210885286331177, 0.36734694242477417, 0.27210885286331177, 0.26530611515045166, 0.25170066952705383, 0.5034013390541077, 0.41496598720550537, 0.3469387888908386, 0.35374149680137634, 0.4897959232330322, 0.31972789764404297, 0.523809552192688, 0.4693877696990967, 0.4897959232330322, NaN, 0.4761904776096344, 0.4897959232330322, 0.5102040767669678, 0.646258533000946, 0.761904776096344, 0.761904776096344, 0.7414966225624084, 0.6530612111091614, 0.6258503198623657, 0.5986394286155701, 0.646258533000946, 0.8231292366981506, 0.8027210831642151, 0.7755101919174194, 0.8707482814788818, 0.9115646481513977, 0.8775510191917419, 0.884353756904602, 0.884353756904602, 0.9251700639724731, 0.9387755393981934, 0.9795918464660645, 1.0], [0.07482992857694626, 0.0476190485060215, 0.06122449040412903, 0.027210883796215057, 0.06802721321582794, 0.08843537420034409, 0.095238097012043, 0.06122449040412903, 0.08163265138864517, 0.08163265138864517, 0.08843537420034409, 0.10884353518486023, 0.1428571492433548, 0.11564625799655914, 0.12244898080825806, 0.20408163964748383, 0.12925170361995697, 0.1428571492433548, 0.12244898080825806, 0.15646257996559143, 0.17687074840068817, 0.11564625799655914, 0.12244898080825806, 0.054421767592430115, 0.10884353518486023, 0.06802721321582794, 0.25170066952705383, 0.13605442643165588, 0.26530611515045166, 0.190476194024086, 0.4693877696990967, 0.3401360511779785, 0.22448979318141937, 0.3469387888908386, 0.30612245202064514, 0.2312925159931183, 0.25850340723991394, 0.21088434755802155, 0.31972789764404297, 0.3877550959587097, 0.36734694242477417, 0.31292515993118286, 0.25850340723991394, 0.40136054158210754, 0.442176878452301, 0.35374149680137634, 0.3945578336715698, 0.3741496503353119, 0.45578232407569885, 0.523809552192688, NaN, 0.47278910875320435, 0.6394557952880859, 0.5986394286155701, 0.6326530575752258, 0.9251700639724731, 0.7414966225624084, 0.6190476417541504, 0.6870748400688171, 0.680272102355957, 0.7006802558898926, 0.8231292366981506, 0.7823129296302795, 0.7891156673431396, 0.7142857313156128, 0.7755101919174194, 0.761904776096344, 0.9455782175064087, 0.9319728016853333, 0.8571428656578064, 0.9115646481513977, 0.9727891087532043, 1.0], [0.07482992857694626, 0.03401360660791397, 0.0476190485060215, 0.013605441898107529, 0.054421767592430115, 0.08163265138864517, 0.07482992857694626, 0.054421767592430115, 0.07482992857694626, 0.06122449040412903, 0.06122449040412903, 0.08163265138864517, 0.11564625799655914, 0.11564625799655914, 0.11564625799655914, 0.18367347121238708, 0.08843537420034409, 0.12244898080825806, 0.13605442643165588, 0.15646257996559143, 0.17006802558898926, 0.11564625799655914, 0.10884353518486023, 0.013605441898107529, 0.095238097012043, 0.054421767592430115, 0.22448979318141937, 0.11564625799655914, 0.26530611515045166, 0.18367347121238708, 0.4897959232330322, 0.36734694242477417, 0.26530611515045166, 0.3469387888908386, 0.2925170063972473, 0.2380952388048172, 0.2857142984867096, 0.1972789168357849, 0.31292515993118286, 0.3945578336715698, 0.3401360511779785, 0.3265306055545807, 0.2380952388048172, 0.3945578336715698, 0.442176878452301, 0.31972789764404297, 0.3741496503353119, 0.40136054158210754, 0.4353741407394409, 0.5102040767669678, 0.5272108912467957, NaN, 0.6598639488220215, 0.5918367505073547, 0.6326530575752258, 0.9251700639724731, 0.7482993006706238, 0.6190476417541504, 0.7006802558898926, 0.6666666865348816, 0.7074829936027527, 0.8299319744110107, 0.7891156673431396, 0.795918345451355, 0.7142857313156128, 0.795918345451355, 0.7755101919174194, 0.9659863710403442, 0.9659863710403442, 0.8571428656578064, 0.9387755393981934, 0.9795918464660645, 1.0], [0.07482992857694626, 0.020408162847161293, 0.03401360660791397, 0.013605441898107529, 0.06802721321582794, 0.06802721321582794, 0.020408162847161293, 0.06802721321582794, 0.10884353518486023, 0.095238097012043, 0.027210883796215057, 0.14965985715389252, 0.1972789168357849, 0.22448979318141937, 0.21088434755802155, 0.22448979318141937, 0.10204081982374191, 0.18367347121238708, 0.10884353518486023, 0.12925170361995697, 0.1428571492433548, 0.21768707036972046, 0.21768707036972046, 0.2448979616165161, 0.22448979318141937, 0.2312925159931183, 0.2857142984867096, 0.040816325694322586, 0.31292515993118286, 0.190476194024086, 0.41496598720550537, 0.30612245202064514, 0.27210885286331177, 0.3469387888908386, 0.26530611515045166, 0.25170066952705383, 0.08163265138864517, 0.29931971430778503, 0.2857142984867096, 0.40136054158210754, 0.31292515993118286, 0.442176878452301, 0.3333333432674408, 0.380952388048172, 0.40136054158210754, 0.3265306055545807, 0.380952388048172, 0.4353741407394409, 0.442176878452301, 0.4897959232330322, 0.36054420471191406, 0.3401360511779785, NaN, 0.49659863114356995, 0.6258503198623657, 0.557823121547699, 0.6598639488220215, 0.7142857313156128, 0.5306122303009033, 0.6598639488220215, 0.6666666865348816, 0.5850340127944946, 0.8095238208770752, 0.8027210831642151, 0.7414966225624084, 0.7755101919174194, 0.7755101919174194, 0.7278911471366882, 0.7482993006706238, 0.918367326259613, 0.9795918464660645, 0.9319728016853333, 1.0], [0.06802721321582794, 0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.054421767592430115, 0.040816325694322586, 0.03401360660791397, 0.027210883796215057, 0.10884353518486023, 0.03401360660791397, 0.040816325694322586, 0.10884353518486023, 0.054421767592430115, 0.15646257996559143, 0.08843537420034409, 0.06802721321582794, 0.040816325694322586, 0.1428571492433548, 0.07482992857694626, 0.06802721321582794, 0.08843537420034409, 0.08163265138864517, 0.08163265138864517, 0.11564625799655914, 0.1428571492433548, 0.13605442643165588, 0.06122449040412903, 0.14965985715389252, 0.08843537420034409, 0.22448979318141937, 0.2857142984867096, 0.36734694242477417, 0.14965985715389252, 0.10884353518486023, 0.08163265138864517, 0.27210885286331177, 0.20408163964748383, 0.1428571492433548, 0.29931971430778503, 0.31972789764404297, 0.12244898080825806, 0.3265306055545807, 0.27210885286331177, 0.3401360511779785, 0.190476194024086, 0.2448979616165161, 0.2312925159931183, 0.35374149680137634, 0.25850340723991394, 0.35374149680137634, 0.40136054158210754, 0.40816327929496765, 0.5034013390541077, NaN, 0.6190476417541504, 0.557823121547699, 0.5986394286155701, 0.6122449040412903, 0.6394557952880859, 0.6666666865348816, 0.7142857313156128, 0.523809552192688, 0.7551020383834839, 0.761904776096344, 0.6258503198623657, 0.6938775777816772, 0.6666666865348816, 0.8503401279449463, 0.8027210831642151, 0.9251700639724731, 0.8911564350128174, 0.9387755393981934, 1.0], [0.06802721321582794, 0.06802721321582794, 0.07482992857694626, 0.020408162847161293, 0.0476190485060215, 0.12244898080825806, 0.054421767592430115, 0.08163265138864517, 0.06802721321582794, 0.07482992857694626, 0.054421767592430115, 0.03401360660791397, 0.11564625799655914, 0.08163265138864517, 0.17687074840068817, 0.15646257996559143, 0.13605442643165588, 0.040816325694322586, 0.08843537420034409, 0.15646257996559143, 0.08163265138864517, 0.15646257996559143, 0.15646257996559143, 0.14965985715389252, 0.14965985715389252, 0.1428571492433548, 0.20408163964748383, 0.08843537420034409, 0.2380952388048172, 0.0476190485060215, 0.4353741407394409, 0.14965985715389252, 0.2380952388048172, 0.27210885286331177, 0.25170066952705383, 0.095238097012043, 0.12925170361995697, 0.20408163964748383, 0.06802721321582794, 0.15646257996559143, 0.30612245202064514, 0.36734694242477417, 0.18367347121238708, 0.15646257996559143, 0.31292515993118286, 0.1972789168357849, 0.2448979616165161, 0.3401360511779785, 0.31292515993118286, 0.2380952388048172, 0.36734694242477417, 0.36734694242477417, 0.3741496503353119, 0.380952388048172, NaN, 0.5714285969734192, 0.45578232407569885, 0.41496598720550537, 0.442176878452301, 0.4897959232330322, 0.49659863114356995, 0.4829931855201721, 0.4829931855201721, 0.49659863114356995, 0.6870748400688171, 0.7278911471366882, 0.7142857313156128, 0.6326530575752258, 0.5850340127944946, 0.9319728016853333, 0.680272102355957, 0.7482993006706238, 0.9931972622871399], [0.06122449040412903, 0.040816325694322586, 0.06122449040412903, 0.013605441898107529, 0.054421767592430115, 0.06802721321582794, 0.06802721321582794, 0.06122449040412903, 0.06802721321582794, 0.054421767592430115, 0.054421767592430115, 0.08843537420034409, 0.0476190485060215, 0.11564625799655914, 0.07482992857694626, 0.08843537420034409, 0.08163265138864517, 0.08163265138864517, 0.11564625799655914, 0.08163265138864517, 0.08843537420034409, 0.07482992857694626, 0.07482992857694626, 0.020408162847161293, 0.06802721321582794, 0.03401360660791397, 0.10204081982374191, 0.06122449040412903, 0.18367347121238708, 0.12925170361995697, 0.36054420471191406, 0.17687074840068817, 0.16326530277729034, 0.31292515993118286, 0.17006802558898926, 0.14965985715389252, 0.11564625799655914, 0.06122449040412903, 0.14965985715389252, 0.21768707036972046, 0.21768707036972046, 0.27210885286331177, 0.14965985715389252, 0.20408163964748383, 0.3401360511779785, 0.13605442643165588, 0.25170066952705383, 0.29931971430778503, 0.3469387888908386, 0.2380952388048172, 0.07482992857694626, 0.07482992857694626, 0.442176878452301, 0.442176878452301, 0.4285714328289032, NaN, 0.5306122303009033, 0.4693877696990967, 0.27210885286331177, 0.5850340127944946, 0.5918367505073547, 0.6734693646430969, 0.7006802558898926, 0.7210884094238281, 0.6258503198623657, 0.680272102355957, 0.6598639488220215, 0.7346938848495483, 0.646258533000946, 0.8435373902320862, 0.8095238208770752, 0.9047619104385376, 1.0], [0.06802721321582794, 0.020408162847161293, 0.013605441898107529, 0.006802720949053764, 0.03401360660791397, 0.013605441898107529, 0.006802720949053764, 0.013605441898107529, 0.08843537420034409, 0.020408162847161293, 0.027210883796215057, 0.040816325694322586, 0.06122449040412903, 0.06122449040412903, 0.095238097012043, 0.07482992857694626, 0.027210883796215057, 0.040816325694322586, 0.0476190485060215, 0.027210883796215057, 0.040816325694322586, 0.040816325694322586, 0.0476190485060215, 0.08163265138864517, 0.13605442643165588, 0.06802721321582794, 0.12244898080825806, 0.027210883796215057, 0.16326530277729034, 0.054421767592430115, 0.3469387888908386, 0.190476194024086, 0.13605442643165588, 0.2380952388048172, 0.12925170361995697, 0.10204081982374191, 0.054421767592430115, 0.10884353518486023, 0.08843537420034409, 0.07482992857694626, 0.18367347121238708, 0.36734694242477417, 0.18367347121238708, 0.1428571492433548, 0.31972789764404297, 0.11564625799655914, 0.21088434755802155, 0.380952388048172, 0.3333333432674408, 0.25850340723991394, 0.25850340723991394, 0.25170066952705383, 0.3401360511779785, 0.40136054158210754, 0.5442177057266235, 0.4693877696990967, NaN, 0.5646258592605591, 0.442176878452301, 0.5442177057266235, 0.5646258592605591, 0.5170068144798279, 0.6870748400688171, 0.6666666865348816, 0.6598639488220215, 0.7551020383834839, 0.7346938848495483, 0.7346938848495483, 0.7482993006706238, 0.9115646481513977, 0.8911564350128174, 0.8911564350128174, 1.0], [0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.006802720949053764, 0.0, 0.040816325694322586, 0.0, 0.027210883796215057, 0.006802720949053764, 0.027210883796215057, 0.0, 0.020408162847161293, 0.10884353518486023, 0.0, 0.10204081982374191, 0.11564625799655914, 0.08163265138864517, 0.040816325694322586, 0.054421767592430115, 0.08843537420034409, 0.08163265138864517, 0.08843537420034409, 0.095238097012043, 0.10204081982374191, 0.08163265138864517, 0.11564625799655914, 0.17687074840068817, 0.03401360660791397, 0.190476194024086, 0.03401360660791397, 0.3469387888908386, 0.1972789168357849, 0.13605442643165588, 0.25850340723991394, 0.21088434755802155, 0.013605441898107529, 0.08163265138864517, 0.15646257996559143, 0.10884353518486023, 0.1428571492433548, 0.2448979616165161, 0.35374149680137634, 0.14965985715389252, 0.18027210235595703, 0.2925170063972473, 0.13605442643165588, 0.2789115607738495, 0.36734694242477417, 0.27210885286331177, 0.3469387888908386, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.3877550959587097, 0.5850340127944946, 0.5306122303009033, 0.4353741407394409, NaN, 0.49659863114356995, 0.49659863114356995, 0.557823121547699, 0.45578232407569885, 0.5442177057266235, 0.5442177057266235, 0.6938775777816772, 0.7278911471366882, 0.7006802558898926, 0.6598639488220215, 0.6394557952880859, 0.884353756904602, 0.8231292366981506, 0.8639456033706665, 1.0], [0.0, 0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.0, 0.006802720949053764, 0.0, 0.0, 0.027210883796215057, 0.020408162847161293, 0.0, 0.0, 0.08843537420034409, 0.0, 0.07482992857694626, 0.14965985715389252, 0.054421767592430115, 0.020408162847161293, 0.013605441898107529, 0.07482992857694626, 0.03401360660791397, 0.07482992857694626, 0.08843537420034409, 0.054421767592430115, 0.006802720949053764, 0.054421767592430115, 0.18367347121238708, 0.006802720949053764, 0.1972789168357849, 0.0, 0.36734694242477417, 0.08163265138864517, 0.07482992857694626, 0.21088434755802155, 0.1972789168357849, 0.0476190485060215, 0.054421767592430115, 0.11564625799655914, 0.11564625799655914, 0.2380952388048172, 0.22448979318141937, 0.21768707036972046, 0.16326530277729034, 0.21088434755802155, 0.26530611515045166, 0.13605442643165588, 0.2448979616165161, 0.2448979616165161, 0.25170066952705383, 0.3741496503353119, 0.31292515993118286, 0.29931971430778503, 0.4693877696990967, 0.36054420471191406, 0.557823121547699, 0.7278911471366882, 0.557823121547699, 0.5034013390541077, NaN, 0.5306122303009033, 0.5306122303009033, 0.5918367505073547, 0.6122449040412903, 0.6190476417541504, 0.6394557952880859, 0.7278911471366882, 0.6938775777816772, 0.7755101919174194, 0.7006802558898926, 0.8367347121238708, 0.7755101919174194, 0.8503401279449463, 1.0], [0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.027210883796215057, 0.020408162847161293, 0.0, 0.013605441898107529, 0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.013605441898107529, 0.1428571492433548, 0.040816325694322586, 0.13605442643165588, 0.054421767592430115, 0.027210883796215057, 0.040816325694322586, 0.190476194024086, 0.08843537420034409, 0.0476190485060215, 0.11564625799655914, 0.054421767592430115, 0.054421767592430115, 0.11564625799655914, 0.18367347121238708, 0.10884353518486023, 0.06122449040412903, 0.040816325694322586, 0.08843537420034409, 0.2380952388048172, 0.1972789168357849, 0.4217686951160431, 0.2312925159931183, 0.12925170361995697, 0.06122449040412903, 0.2925170063972473, 0.12925170361995697, 0.16326530277729034, 0.2925170063972473, 0.3401360511779785, 0.10884353518486023, 0.190476194024086, 0.2312925159931183, 0.3265306055545807, 0.26530611515045166, 0.2789115607738495, 0.21768707036972046, 0.25170066952705383, 0.21768707036972046, 0.40136054158210754, 0.31972789764404297, 0.3333333432674408, 0.3401360511779785, 0.3333333432674408, 0.5102040767669678, 0.41496598720550537, 0.45578232407569885, 0.5034013390541077, 0.4693877696990967, NaN, 0.7448979616165161, 0.45578232407569885, 0.45578232407569885, 0.4829931855201721, 0.4761904776096344, 0.4897959232330322, 0.5170068144798279, 0.45578232407569885, 0.5306122303009033, 0.7551020383834839, 0.6666666865348816, 0.9455782175064087, 0.9319728016853333], [0.006802720949053764, 0.0, 0.006802720949053764, 0.020408162847161293, 0.013605441898107529, 0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.020408162847161293, 0.006802720949053764, 0.006802720949053764, 0.12925170361995697, 0.020408162847161293, 0.14965985715389252, 0.040816325694322586, 0.027210883796215057, 0.03401360660791397, 0.1972789168357849, 0.040816325694322586, 0.03401360660791397, 0.06802721321582794, 0.040816325694322586, 0.040816325694322586, 0.07482992857694626, 0.11564625799655914, 0.07482992857694626, 0.06122449040412903, 0.03401360660791397, 0.06802721321582794, 0.2380952388048172, 0.1972789168357849, 0.35374149680137634, 0.17006802558898926, 0.08163265138864517, 0.027210883796215057, 0.2857142984867096, 0.12244898080825806, 0.13605442643165588, 0.31292515993118286, 0.30612245202064514, 0.07482992857694626, 0.1428571492433548, 0.190476194024086, 0.3333333432674408, 0.21088434755802155, 0.2380952388048172, 0.1428571492433548, 0.20408163964748383, 0.15646257996559143, 0.35374149680137634, 0.29931971430778503, 0.2925170063972473, 0.3333333432674408, 0.2857142984867096, 0.5034013390541077, 0.40816327929496765, 0.4353741407394409, 0.442176878452301, 0.4693877696990967, 0.2551020383834839, NaN, 0.44897958636283875, 0.40136054158210754, 0.41496598720550537, 0.44897958636283875, 0.4761904776096344, 0.5034013390541077, 0.4353741407394409, 0.523809552192688, 0.7414966225624084, 0.6530612111091614, 0.8571428656578064, 0.9319728016853333], [0.06122449040412903, 0.0, 0.0, 0.0, 0.0, 0.013605441898107529, 0.006802720949053764, 0.0, 0.06122449040412903, 0.006802720949053764, 0.027210883796215057, 0.06122449040412903, 0.0, 0.06122449040412903, 0.020408162847161293, 0.0, 0.0, 0.06802721321582794, 0.020408162847161293, 0.0, 0.027210883796215057, 0.006802720949053764, 0.006802720949053764, 0.0, 0.006802720949053764, 0.0, 0.006802720949053764, 0.10204081982374191, 0.013605441898107529, 0.08843537420034409, 0.2789115607738495, 0.08163265138864517, 0.020408162847161293, 0.08843537420034409, 0.054421767592430115, 0.15646257996559143, 0.1428571492433548, 0.0, 0.14965985715389252, 0.16326530277729034, 0.095238097012043, 0.21088434755802155, 0.16326530277729034, 0.20408163964748383, 0.1428571492433548, 0.040816325694322586, 0.13605442643165588, 0.21088434755802155, 0.17006802558898926, 0.17687074840068817, 0.17687074840068817, 0.17006802558898926, 0.41496598720550537, 0.4761904776096344, 0.5170068144798279, 0.3265306055545807, 0.4829931855201721, 0.5442177057266235, 0.40816327929496765, 0.5442177057266235, 0.5510203838348389, NaN, 0.6870748400688171, 0.680272102355957, 0.6530612111091614, 0.680272102355957, 0.680272102355957, 0.6598639488220215, 0.6938775777816772, 0.7551020383834839, 0.7755101919174194, 0.7687074542045593, 0.9319728016853333], [0.06122449040412903, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.013605441898107529, 0.0, 0.027210883796215057, 0.06802721321582794, 0.027210883796215057, 0.020408162847161293, 0.095238097012043, 0.06122449040412903, 0.12244898080825806, 0.06122449040412903, 0.07482992857694626, 0.020408162847161293, 0.095238097012043, 0.08843537420034409, 0.03401360660791397, 0.07482992857694626, 0.0476190485060215, 0.040816325694322586, 0.054421767592430115, 0.13605442643165588, 0.0476190485060215, 0.0476190485060215, 0.03401360660791397, 0.07482992857694626, 0.08843537420034409, 0.26530611515045166, 0.2380952388048172, 0.15646257996559143, 0.10204081982374191, 0.06802721321582794, 0.07482992857694626, 0.07482992857694626, 0.08163265138864517, 0.08843537420034409, 0.12925170361995697, 0.08163265138864517, 0.2312925159931183, 0.21768707036972046, 0.12244898080825806, 0.13605442643165588, 0.12244898080825806, 0.095238097012043, 0.26530611515045166, 0.16326530277729034, 0.1972789168357849, 0.21768707036972046, 0.21088434755802155, 0.190476194024086, 0.2448979616165161, 0.5170068144798279, 0.29931971430778503, 0.31292515993118286, 0.45578232407569885, 0.3877550959587097, 0.5442177057266235, 0.5986394286155701, 0.31292515993118286, NaN, 0.5374149680137634, 0.5102040767669678, 0.5510203838348389, 0.5306122303009033, 0.5850340127944946, 0.4285714328289032, 0.8503401279449463, 1.0, 0.9047619104385376, 0.9863945841789246], [0.06802721321582794, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.020408162847161293, 0.020408162847161293, 0.03401360660791397, 0.06802721321582794, 0.040816325694322586, 0.03401360660791397, 0.095238097012043, 0.0476190485060215, 0.12244898080825806, 0.06122449040412903, 0.0476190485060215, 0.027210883796215057, 0.095238097012043, 0.095238097012043, 0.040816325694322586, 0.06802721321582794, 0.054421767592430115, 0.0476190485060215, 0.054421767592430115, 0.1428571492433548, 0.040816325694322586, 0.0476190485060215, 0.03401360660791397, 0.06802721321582794, 0.08843537420034409, 0.2789115607738495, 0.25170066952705383, 0.15646257996559143, 0.095238097012043, 0.06122449040412903, 0.07482992857694626, 0.06802721321582794, 0.08843537420034409, 0.095238097012043, 0.1428571492433548, 0.07482992857694626, 0.2312925159931183, 0.21768707036972046, 0.12925170361995697, 0.1428571492433548, 0.13605442643165588, 0.08843537420034409, 0.25170066952705383, 0.16326530277729034, 0.22448979318141937, 0.21088434755802155, 0.20408163964748383, 0.1972789168357849, 0.2380952388048172, 0.5034013390541077, 0.2789115607738495, 0.3333333432674408, 0.45578232407569885, 0.380952388048172, 0.5170068144798279, 0.5850340127944946, 0.31972789764404297, 0.4625850319862366, NaN, 0.49659863114356995, 0.5374149680137634, 0.5306122303009033, 0.5714285969734192, 0.4217686951160431, 0.8503401279449463, 1.0, 0.884353756904602, 0.9863945841789246], [0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.006802720949053764, 0.03401360660791397, 0.013605441898107529, 0.027210883796215057, 0.020408162847161293, 0.027210883796215057, 0.013605441898107529, 0.020408162847161293, 0.040816325694322586, 0.0476190485060215, 0.06122449040412903, 0.06122449040412903, 0.040816325694322586, 0.027210883796215057, 0.06802721321582794, 0.020408162847161293, 0.027210883796215057, 0.03401360660791397, 0.040816325694322586, 0.040816325694322586, 0.06122449040412903, 0.10884353518486023, 0.06122449040412903, 0.06802721321582794, 0.08843537420034409, 0.06122449040412903, 0.07482992857694626, 0.2448979616165161, 0.14965985715389252, 0.10884353518486023, 0.16326530277729034, 0.08163265138864517, 0.08843537420034409, 0.10884353518486023, 0.08163265138864517, 0.08843537420034409, 0.06802721321582794, 0.10204081982374191, 0.31972789764404297, 0.12244898080825806, 0.10884353518486023, 0.21768707036972046, 0.10884353518486023, 0.12244898080825806, 0.3401360511779785, 0.17006802558898926, 0.12925170361995697, 0.2857142984867096, 0.2857142984867096, 0.25850340723991394, 0.3741496503353119, 0.31292515993118286, 0.3741496503353119, 0.3401360511779785, 0.30612245202064514, 0.36054420471191406, 0.523809552192688, 0.5510203838348389, 0.3469387888908386, 0.4897959232330322, 0.5034013390541077, NaN, 0.7755101919174194, 0.8027210831642151, 0.5850340127944946, 0.6326530575752258, 0.6258503198623657, 0.5510203838348389, 0.6598639488220215, 0.9387755393981934], [0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.0476190485060215, 0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.03401360660791397, 0.027210883796215057, 0.027210883796215057, 0.040816325694322586, 0.03401360660791397, 0.06122449040412903, 0.054421767592430115, 0.0476190485060215, 0.027210883796215057, 0.054421767592430115, 0.027210883796215057, 0.027210883796215057, 0.03401360660791397, 0.040816325694322586, 0.0476190485060215, 0.0476190485060215, 0.08843537420034409, 0.0476190485060215, 0.040816325694322586, 0.10884353518486023, 0.06802721321582794, 0.0476190485060215, 0.25170066952705383, 0.12925170361995697, 0.08843537420034409, 0.12244898080825806, 0.06802721321582794, 0.06802721321582794, 0.08163265138864517, 0.054421767592430115, 0.08843537420034409, 0.054421767592430115, 0.08843537420034409, 0.2789115607738495, 0.095238097012043, 0.08163265138864517, 0.17006802558898926, 0.07482992857694626, 0.10204081982374191, 0.29931971430778503, 0.10884353518486023, 0.08843537420034409, 0.22448979318141937, 0.20408163964748383, 0.22448979318141937, 0.30612245202064514, 0.27210885286331177, 0.31972789764404297, 0.2448979616165161, 0.27210885286331177, 0.27210885286331177, 0.5102040767669678, 0.523809552192688, 0.31972789764404297, 0.44897958636283875, 0.4625850319862366, 0.22448979318141937, NaN, 0.6496598720550537, 0.557823121547699, 0.5170068144798279, 0.6190476417541504, 0.5034013390541077, 0.6394557952880859, 0.9319728016853333], [0.03401360660791397, 0.03401360660791397, 0.027210883796215057, 0.03401360660791397, 0.03401360660791397, 0.027210883796215057, 0.03401360660791397, 0.03401360660791397, 0.03401360660791397, 0.040816325694322586, 0.03401360660791397, 0.0476190485060215, 0.0476190485060215, 0.054421767592430115, 0.0476190485060215, 0.054421767592430115, 0.03401360660791397, 0.0476190485060215, 0.0476190485060215, 0.03401360660791397, 0.054421767592430115, 0.03401360660791397, 0.040816325694322586, 0.06122449040412903, 0.08163265138864517, 0.06122449040412903, 0.06122449040412903, 0.10884353518486023, 0.07482992857694626, 0.06122449040412903, 0.25170066952705383, 0.11564625799655914, 0.095238097012043, 0.12925170361995697, 0.06802721321582794, 0.06802721321582794, 0.095238097012043, 0.08163265138864517, 0.08163265138864517, 0.054421767592430115, 0.095238097012043, 0.2789115607738495, 0.10204081982374191, 0.08843537420034409, 0.18367347121238708, 0.08163265138864517, 0.12925170361995697, 0.2925170063972473, 0.15646257996559143, 0.12244898080825806, 0.2380952388048172, 0.22448979318141937, 0.22448979318141937, 0.3333333432674408, 0.2857142984867096, 0.3401360511779785, 0.26530611515045166, 0.29931971430778503, 0.30612245202064514, 0.4829931855201721, 0.49659863114356995, 0.31972789764404297, 0.4693877696990967, 0.4693877696990967, 0.1972789168357849, 0.3503401279449463, NaN, 0.5374149680137634, 0.5374149680137634, 0.5918367505073547, 0.5374149680137634, 0.6122449040412903, 0.7414966225624084], [0.0, 0.0, 0.006802720949053764, 0.006802720949053764, 0.0, 0.006802720949053764, 0.006802720949053764, 0.0, 0.013605441898107529, 0.0, 0.0, 0.06122449040412903, 0.020408162847161293, 0.06122449040412903, 0.020408162847161293, 0.027210883796215057, 0.020408162847161293, 0.06122449040412903, 0.054421767592430115, 0.027210883796215057, 0.06122449040412903, 0.013605441898107529, 0.013605441898107529, 0.0, 0.006802720949053764, 0.0, 0.006802720949053764, 0.03401360660791397, 0.020408162847161293, 0.10204081982374191, 0.21088434755802155, 0.095238097012043, 0.06122449040412903, 0.040816325694322586, 0.020408162847161293, 0.11564625799655914, 0.07482992857694626, 0.03401360660791397, 0.11564625799655914, 0.12925170361995697, 0.03401360660791397, 0.006802720949053764, 0.1428571492433548, 0.12925170361995697, 0.07482992857694626, 0.054421767592430115, 0.03401360660791397, 0.013605441898107529, 0.06802721321582794, 0.11564625799655914, 0.054421767592430115, 0.03401360660791397, 0.27210885286331177, 0.14965985715389252, 0.36734694242477417, 0.26530611515045166, 0.26530611515045166, 0.3401360511779785, 0.22448979318141937, 0.5442177057266235, 0.5646258592605591, 0.3401360511779785, 0.41496598720550537, 0.4285714328289032, 0.41496598720550537, 0.442176878452301, 0.4625850319862366, NaN, 0.3469387888908386, 0.7823129296302795, 0.761904776096344, 0.8979591727256775, 0.9931972622871399], [0.03401360660791397, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.054421767592430115, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.020408162847161293, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.020408162847161293, 0.25170066952705383, 0.0, 0.027210883796215057, 0.006802720949053764, 0.006802720949053764, 0.054421767592430115, 0.006802720949053764, 0.0, 0.03401360660791397, 0.0476190485060215, 0.006802720949053764, 0.08163265138864517, 0.07482992857694626, 0.07482992857694626, 0.03401360660791397, 0.013605441898107529, 0.020408162847161293, 0.17006802558898926, 0.07482992857694626, 0.11564625799655914, 0.06802721321582794, 0.03401360660791397, 0.25170066952705383, 0.1972789168357849, 0.41496598720550537, 0.35374149680137634, 0.25170066952705383, 0.36054420471191406, 0.29931971430778503, 0.4693877696990967, 0.4761904776096344, 0.30612245202064514, 0.5714285969734192, 0.5782312750816345, 0.36734694242477417, 0.4829931855201721, 0.4625850319862366, 0.6530612111091614, NaN, 0.795918345451355, 0.7687074542045593, 0.8163265585899353, 1.0], [0.06802721321582794, 0.03401360660791397, 0.0476190485060215, 0.020408162847161293, 0.040816325694322586, 0.054421767592430115, 0.054421767592430115, 0.0476190485060215, 0.06122449040412903, 0.027210883796215057, 0.0476190485060215, 0.027210883796215057, 0.03401360660791397, 0.054421767592430115, 0.0476190485060215, 0.03401360660791397, 0.0476190485060215, 0.040816325694322586, 0.06802721321582794, 0.0476190485060215, 0.027210883796215057, 0.06802721321582794, 0.06802721321582794, 0.0476190485060215, 0.10204081982374191, 0.0476190485060215, 0.027210883796215057, 0.013605441898107529, 0.040816325694322586, 0.0, 0.2380952388048172, 0.12244898080825806, 0.11564625799655914, 0.06122449040412903, 0.06122449040412903, 0.006802720949053764, 0.027210883796215057, 0.054421767592430115, 0.0, 0.0476190485060215, 0.06122449040412903, 0.10884353518486023, 0.10204081982374191, 0.027210883796215057, 0.06122449040412903, 0.07482992857694626, 0.027210883796215057, 0.10884353518486023, 0.07482992857694626, 0.07482992857694626, 0.1428571492433548, 0.1428571492433548, 0.08163265138864517, 0.07482992857694626, 0.06802721321582794, 0.15646257996559143, 0.08843537420034409, 0.11564625799655914, 0.16326530277729034, 0.2448979616165161, 0.25850340723991394, 0.2448979616165161, 0.14965985715389252, 0.14965985715389252, 0.3741496503353119, 0.380952388048172, 0.40816327929496765, 0.21768707036972046, 0.20408163964748383, NaN, 0.3741496503353119, 0.3741496503353119, 0.9251700639724731], [0.006802720949053764, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.0, 0.0, 0.006802720949053764, 0.006802720949053764, 0.013605441898107529, 0.0, 0.06802721321582794, 0.006802720949053764, 0.06122449040412903, 0.03401360660791397, 0.013605441898107529, 0.006802720949053764, 0.07482992857694626, 0.03401360660791397, 0.006802720949053764, 0.03401360660791397, 0.013605441898107529, 0.013605441898107529, 0.006802720949053764, 0.027210883796215057, 0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.020408162847161293, 0.06802721321582794, 0.054421767592430115, 0.10884353518486023, 0.054421767592430115, 0.027210883796215057, 0.013605441898107529, 0.06122449040412903, 0.020408162847161293, 0.006802720949053764, 0.06802721321582794, 0.06802721321582794, 0.020408162847161293, 0.0476190485060215, 0.13605442643165588, 0.08163265138864517, 0.0476190485060215, 0.013605441898107529, 0.027210883796215057, 0.06122449040412903, 0.06122449040412903, 0.06122449040412903, 0.08843537420034409, 0.06122449040412903, 0.020408162847161293, 0.10884353518486023, 0.31972789764404297, 0.190476194024086, 0.10884353518486023, 0.17687074840068817, 0.22448979318141937, 0.3333333432674408, 0.3469387888908386, 0.22448979318141937, 0.0, 0.0, 0.44897958636283875, 0.49659863114356995, 0.4625850319862366, 0.2380952388048172, 0.2312925159931183, 0.6258503198623657, NaN, 0.680272102355957, 0.9591836929321289], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.03401360660791397, 0.006802720949053764, 0.0, 0.0, 0.06122449040412903, 0.006802720949053764, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.054421767592430115, 0.190476194024086, 0.06802721321582794, 0.027210883796215057, 0.006802720949053764, 0.006802720949053764, 0.08163265138864517, 0.006802720949053764, 0.0, 0.07482992857694626, 0.03401360660791397, 0.006802720949053764, 0.0, 0.020408162847161293, 0.10204081982374191, 0.013605441898107529, 0.0, 0.0, 0.0, 0.006802720949053764, 0.020408162847161293, 0.027210883796215057, 0.020408162847161293, 0.06802721321582794, 0.06122449040412903, 0.25170066952705383, 0.095238097012043, 0.10884353518486023, 0.13605442643165588, 0.14965985715389252, 0.054421767592430115, 0.1428571492433548, 0.2312925159931183, 0.095238097012043, 0.11564625799655914, 0.3401360511779785, 0.36054420471191406, 0.3877550959587097, 0.10204081982374191, 0.18367347121238708, 0.6258503198623657, 0.31972789764404297, NaN, 0.9115646481513977], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.06802721321582794, 0.06802721321582794, 0.06802721321582794, 0.013605441898107529, 0.013605441898107529, 0.06122449040412903, 0.06802721321582794, 0.25850340723991394, 0.006802720949053764, 0.0, 0.07482992857694626, 0.040816325694322586, 0.08843537420034409, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9076908529},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8834561586},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8800075588},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.86989796},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8364512463},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.835837115},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8282312943},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8091931219},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.803099018},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7904856387},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7765495088},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7235449746},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7157501907},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.71031746},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6941610007},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6811696902},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6809807252},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.677532124},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.66964286},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6524470852},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.638463719},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6367630411},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6298658353},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6234410434},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6192365836},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6183390045},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6150793689},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6145124713},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5947656892},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5874433107},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5686885859},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5505479971},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5391156479},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5388794421},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5386904763},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5361866979},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5276832931},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5247543455},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.5194161},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5046296291},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4820011348},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4783635649},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4606009081},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4591364308},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4478930485},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4465230549},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4390589571},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4125094469},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4054232809},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3986205614},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.3812830706},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.378543084},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3682917607},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3124527604},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2822184424},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.27031368},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2654950887},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.265353365},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2605820091},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2541099777},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2251039313},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2212773997},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2172146629},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2157974308},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2089002259},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1793745286},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1785241883},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1609977329},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1564625839},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.113095238},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1108276645},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.078892668},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0118102798}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_multiclass/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..cd9dc222ea36962518ba89c8e5d9677afe6a02ea
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_multiclass/datasets_small/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),RealTabPFN-2.5 (default),TabPFN-3 (default),RealMLP (tuned + ensembled),TabICLv2 (default),RealMLP (tuned),LimiX (default),TabDPT (tuned + ensembled),TabM (tuned + ensembled),TabDPT-Turbo (default),"AutoGluon 1.4 (best, 4h)",TabM (tuned),CatBoost (tuned + ensembled),TabDPT (tuned),ModernNCA (tuned + ensembled),CatBoost (tuned),ModernNCA (tuned),XGBoost (tuned),XGBoost (tuned + ensembled),LightGBM (tuned + ensembled),iLTM (tuned + ensembled),LightGBM (tuned),BetaTabPFN (default),xRFM (tuned + ensembled),TabM (default),ExtraTrees (tuned + ensembled),SAP-RPT-OSS (default),TabDPT (default),iLTM (tuned),TorchMLP (tuned + ensembled),EBM (tuned + ensembled),RandomForest (tuned + ensembled),xRFM (tuned),CatBoost (default),ExtraTrees (tuned),TabSwift (default),EBM (tuned),FastaiMLP (tuned + ensembled),ModernNCA (default),RandomForest (tuned),TorchMLP (tuned),XGBoost (default),EBM (default),FastaiMLP (tuned),RealMLP (default),LightGBM (default),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),xRFM (default),TorchMLP (default),KNN (tuned + ensembled),ChimeraBoost (default),OrionMSP (default),RandomForest (default),ExtraTrees (default),Linear (tuned + ensembled),Linear (tuned),iLTM (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),PerpetualBooster (tuned + ensembled),PerpetualBooster (tuned),PerpetualBooster (default),FastaiMLP (default),TabFlex (default),KNN (tuned),TabSTAR (default),Linear (default),KNN (default)
+TabFM (default),,0.70068026,0.70068026,0.81632656,0.8367347,0.7414966,0.8503401,0.76870745,0.89115644,0.76870745,0.8435374,0.8979592,0.8095238,0.88435376,0.92517006,0.82312924,0.9047619,0.88435376,0.9183673,0.92517006,0.89115644,0.877551,0.88435376,0.9047619,0.9183673,0.8979592,0.91156465,0.9183673,0.8503401,0.92517006,0.75510204,0.9047619,0.9319728,0.92517006,0.9047619,0.9047619,0.92517006,0.9319728,0.9183673,0.9183673,0.92517006,0.8979592,0.97959185,0.9047619,0.9319728,0.9319728,0.92517006,0.93877554,0.97959185,0.92517006,0.92517006,0.92517006,0.92517006,0.9319728,0.9319728,0.93877554,0.9319728,0.99319726,1.0,0.99319726,0.99319726,0.93877554,0.93877554,0.9319728,0.99319726,0.9727891,0.9659864,1.0,0.9659864,0.9319728,0.99319726,1.0,1.0
+RealTabPFN-2.5 (tuned),0.2993197,,0.46258503,0.5782313,0.61904764,0.68367344,0.61904764,0.7414966,0.6530612,0.76870745,0.7346939,0.72789115,0.8435374,0.6938776,0.88435376,0.8503401,0.92517006,0.7755102,0.8503401,0.92517006,0.8639456,0.9047619,0.9047619,0.88435376,0.877551,0.91156465,0.8979592,0.89115644,0.85714287,0.8639456,0.7482993,0.79591835,0.95238096,0.9455782,0.95238096,0.877551,0.9183673,0.9659864,0.89115644,0.9727891,0.9591837,0.92517006,0.9455782,0.9047619,0.9455782,0.97959185,0.9455782,0.9659864,0.99319726,0.9455782,0.95238096,0.9659864,0.97959185,0.99319726,0.9319728,0.9591837,0.97959185,0.9863946,0.99319726,0.99319726,1.0,1.0,0.9863946,0.9863946,0.9863946,0.9727891,0.9659864,1.0,1.0,0.9659864,0.9863946,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.2993197,0.53741497,,0.60544217,0.6258503,0.7482993,0.6258503,0.7346939,0.65986395,0.7482993,0.7414966,0.7210884,0.79591835,0.68707484,0.85714287,0.79591835,0.93877554,0.7414966,0.85714287,0.92517006,0.877551,0.8979592,0.9183673,0.89115644,0.88435376,0.8979592,0.91156465,0.877551,0.8435374,0.8435374,0.7414966,0.7755102,0.95238096,0.93877554,0.9319728,0.8707483,0.9183673,0.93877554,0.8707483,0.9727891,0.93877554,0.9319728,0.9319728,0.9047619,0.9455782,0.9863946,0.9319728,0.95238096,0.9863946,0.9319728,0.93877554,0.95238096,0.9659864,0.9863946,0.92517006,0.93877554,0.9863946,0.9863946,0.99319726,0.99319726,0.99319726,1.0,0.99319726,0.99319726,0.9863946,0.9727891,0.9727891,0.99319726,1.0,0.95238096,0.99319726,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.18367347,0.4217687,0.39455783,,0.5986394,0.4829932,0.5714286,0.7210884,0.5714286,0.7346939,0.78911567,0.81632656,0.76870745,0.7346939,0.79591835,0.7619048,0.88435376,0.85714287,0.8639456,0.8979592,0.88435376,0.829932,0.85714287,0.9047619,0.91156465,0.89115644,0.82312924,0.9047619,0.8027211,0.8979592,0.6258503,0.8367347,0.93877554,0.9047619,0.9183673,0.9047619,0.93877554,0.9455782,0.8979592,0.9863946,0.9455782,0.8707483,0.9659864,0.9047619,0.95238096,0.99319726,0.93877554,0.93877554,0.97959185,0.9727891,0.9727891,0.9863946,0.9863946,0.9863946,0.97959185,0.9863946,0.99319726,0.99319726,0.99319726,0.9727891,0.97959185,1.0,0.99319726,0.99319726,0.99319726,0.9727891,0.9659864,0.99319726,1.0,0.97959185,0.99319726,1.0,1.0
+TabPFN-2.6 (default),0.1632653,0.3809524,0.37414965,0.40136054,,0.52380955,0.45578232,0.63265306,0.45578232,0.63265306,0.64625853,0.6258503,0.71428573,0.67346936,0.76870745,0.7482993,0.8027211,0.70068026,0.85714287,0.81632656,0.829932,0.877551,0.8979592,0.8503401,0.92517006,0.88435376,0.8367347,0.78911567,0.78911567,0.8435374,0.6530612,0.7755102,0.9455782,0.9183673,0.8639456,0.82312924,0.8503401,0.95238096,0.8367347,0.95238096,0.9047619,0.9455782,0.9319728,0.8639456,0.9183673,0.97959185,0.92517006,0.97959185,0.9659864,0.9319728,0.9319728,0.9455782,0.9319728,0.9455782,0.95238096,0.9455782,0.9659864,1.0,1.0,0.97959185,0.9863946,1.0,0.99319726,0.99319726,0.9659864,0.95238096,0.9659864,1.0,1.0,0.9591837,0.99319726,1.0,1.0
+RealTabPFN-2.5 (default),0.2585034,0.31632653,0.25170067,0.5170068,0.47619048,,0.5578231,0.6394558,0.5782313,0.6530612,0.72789115,0.65986395,0.78911567,0.6258503,0.78911567,0.8095238,0.8639456,0.6802721,0.7414966,0.89115644,0.78911567,0.8639456,0.8639456,0.8503401,0.79591835,0.8707483,0.8503401,0.81632656,0.81632656,0.8027211,0.71428573,0.70068026,0.8979592,0.8979592,0.9047619,0.82312924,0.8707483,0.91156465,0.8435374,0.9319728,0.91156465,0.9183673,0.9047619,0.91156465,0.8979592,0.9455782,0.92517006,0.93877554,0.99319726,0.91156465,0.91156465,0.9183673,0.9319728,0.9591837,0.877551,0.9319728,0.9863946,0.9591837,0.99319726,1.0,0.99319726,0.9863946,0.9863946,0.97959185,0.9863946,0.9727891,0.9727891,0.99319726,0.99319726,0.9455782,1.0,1.0,1.0
+TabPFN-3 (default),0.14965986,0.3809524,0.37414965,0.42857143,0.5442177,0.44217688,,0.52380955,0.5714286,0.60544217,0.64625853,0.72789115,0.65986395,0.7210884,0.6938776,0.6938776,0.82312924,0.7619048,0.8095238,0.82312924,0.82312924,0.7755102,0.7823129,0.8435374,0.8503401,0.8367347,0.7414966,0.8367347,0.70068026,0.9047619,0.5714286,0.81632656,0.91156465,0.877551,0.829932,0.877551,0.9047619,0.88435376,0.91156465,0.9455782,0.8503401,0.829932,0.9659864,0.8979592,0.89115644,0.97959185,0.9183673,0.93877554,0.9863946,0.92517006,0.9047619,0.92517006,0.97959185,0.9659864,0.9455782,0.9319728,0.99319726,1.0,1.0,0.9863946,0.9863946,0.99319726,1.0,0.97959185,0.9727891,0.9727891,0.9659864,0.99319726,1.0,0.9455782,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.23129252,0.2585034,0.26530612,0.27891156,0.36734694,0.3605442,0.47619048,,0.44217688,0.5136054,0.5170068,0.5510204,0.7346939,0.56462586,0.707483,0.7414966,0.7823129,0.5578231,0.72789115,0.81632656,0.7823129,0.8979592,0.92517006,0.877551,0.8027211,0.877551,0.85714287,0.82312924,0.8707483,0.78911567,0.6530612,0.60544217,0.9047619,0.8027211,0.9183673,0.8639456,0.8503401,0.9319728,0.85714287,0.78911567,0.92517006,0.9047619,0.9455782,0.91156465,0.85714287,0.99319726,0.9319728,0.91156465,0.99319726,0.92517006,0.93877554,0.9455782,0.9319728,0.9727891,0.9183673,0.93877554,0.9863946,0.9727891,1.0,0.99319726,0.9863946,1.0,0.9727891,0.9659864,0.97959185,0.9727891,0.9659864,1.0,1.0,0.95238096,0.99319726,1.0,1.0
+TabICLv2 (default),0.108843535,0.3469388,0.34013605,0.42857143,0.5442177,0.4217687,0.42857143,0.5578231,,0.585034,0.61904764,0.7346939,0.6666667,0.67346936,0.71428573,0.67346936,0.7755102,0.7346939,0.7414966,0.7823129,0.8095238,0.75510204,0.76870745,0.8435374,0.8707483,0.85714287,0.7346939,0.7823129,0.7346939,0.81632656,0.5102041,0.7482993,0.877551,0.82312924,0.829932,0.8707483,0.81632656,0.88435376,0.8367347,0.9183673,0.85714287,0.81632656,0.95238096,0.8707483,0.877551,0.92517006,0.9047619,0.8979592,0.92517006,0.92517006,0.9183673,0.92517006,0.89115644,0.89115644,0.9319728,0.9319728,0.91156465,0.99319726,0.9727891,0.9863946,0.97959185,0.93877554,0.9319728,0.9319728,0.9727891,0.9659864,0.9659864,0.9863946,0.9455782,0.93877554,0.99319726,1.0,1.0
+RealMLP (tuned),0.23129252,0.23129252,0.25170067,0.26530612,0.36734694,0.3469388,0.39455783,0.48639455,0.414966,,0.5102041,0.5102041,0.65986395,0.5510204,0.6530612,0.70068026,0.7482993,0.5442177,0.7346939,0.75510204,0.75510204,0.829932,0.85714287,0.78911567,0.76870745,0.8027211,0.8435374,0.7619048,0.8027211,0.7823129,0.6394558,0.61904764,0.88435376,0.8503401,0.8979592,0.8367347,0.829932,0.89115644,0.82312924,0.78911567,0.93877554,0.8707483,0.9183673,0.85714287,0.8979592,0.9591837,0.9183673,0.9319728,0.97959185,0.9183673,0.9183673,0.93877554,0.9047619,0.9659864,0.92517006,0.9455782,0.97959185,0.9727891,0.97959185,0.9863946,0.99319726,0.99319726,0.9727891,0.9591837,0.9863946,0.9727891,0.9591837,1.0,1.0,0.9727891,0.9863946,1.0,1.0
+LimiX (default),0.15646258,0.26530612,0.2585034,0.21088435,0.3537415,0.27210885,0.3537415,0.4829932,0.3809524,0.48979592,,0.5170068,0.63265306,0.46258503,0.6394558,0.6394558,0.72789115,0.60544217,0.63265306,0.7414966,0.76870745,0.7619048,0.75510204,0.78911567,0.7482993,0.79591835,0.7482993,0.8367347,0.707483,0.88435376,0.585034,0.60544217,0.829932,0.70068026,0.82312924,0.9047619,0.8979592,0.877551,0.9047619,0.8095238,0.877551,0.76870745,0.93877554,0.8979592,0.81632656,0.9455782,0.8979592,0.88435376,0.97959185,0.93877554,0.91156465,0.93877554,0.9727891,0.9591837,0.9455782,0.9455782,0.9727891,1.0,1.0,0.9863946,0.99319726,0.9727891,0.97959185,0.9659864,0.97959185,0.9727891,0.9659864,1.0,0.99319726,0.95238096,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.10204082,0.27210885,0.27891156,0.18367347,0.37414965,0.34013605,0.27210885,0.4489796,0.26530612,0.48979592,0.4829932,,0.5442177,0.47619048,0.5578231,0.5510204,0.5782313,0.65986395,0.6530612,0.6122449,0.6802721,0.63265306,0.6530612,0.71428573,0.75510204,0.6938776,0.63265306,0.63265306,0.63265306,0.7346939,0.48979592,0.7482993,0.78911567,0.6802721,0.7346939,0.79591835,0.7414966,0.78911567,0.78911567,0.8639456,0.76870745,0.76870745,0.8503401,0.8367347,0.8095238,0.8979592,0.79591835,0.81632656,0.85714287,0.85714287,0.89115644,0.9183673,0.8503401,0.89115644,0.9659864,0.91156465,0.9591837,0.97959185,1.0,0.85714287,0.8707483,0.93877554,0.9047619,0.9047619,0.9591837,0.9591837,0.95238096,0.93877554,1.0,0.9727891,0.9319728,0.95238096,1.0
+TabM (tuned + ensembled),0.1904762,0.15646258,0.20408164,0.23129252,0.2857143,0.21088435,0.34013605,0.26530612,0.33333334,0.34013605,0.36734694,0.45578232,,0.5102041,0.52380955,0.6360544,0.53061223,0.48979592,0.6122449,0.5986394,0.6122449,0.68707484,0.71428573,0.65986395,0.6258503,0.6666667,0.7619048,0.61904764,0.877551,0.5782313,0.6394558,0.5986394,0.7414966,0.7755102,0.8639456,0.6394558,0.70068026,0.829932,0.63265306,0.68707484,0.89115644,0.81632656,0.8095238,0.75510204,0.8707483,0.88435376,0.89115644,0.877551,0.95238096,0.88435376,0.85714287,0.88435376,0.8027211,0.9455782,0.88435376,0.95238096,0.93877554,0.89115644,0.91156465,0.9591837,0.97959185,1.0,0.93877554,0.95238096,0.95238096,0.9659864,0.95238096,0.97959185,1.0,0.9659864,0.99319726,1.0,1.0
+TabDPT-Turbo (default),0.11564626,0.30612245,0.31292516,0.26530612,0.3265306,0.37414965,0.27891156,0.43537414,0.3265306,0.4489796,0.53741497,0.52380955,0.48979592,,0.5442177,0.5102041,0.60544217,0.5782313,0.65986395,0.59183675,0.6530612,0.6122449,0.6122449,0.67346936,0.79591835,0.61904764,0.5510204,0.60544217,0.5578231,0.75510204,0.49659863,0.78911567,0.8027211,0.70068026,0.6666667,0.7823129,0.67346936,0.7414966,0.76870745,0.82312924,0.72789115,0.7755102,0.829932,0.78911567,0.76870745,0.8639456,0.76870745,0.8639456,0.8435374,0.8503401,0.88435376,0.88435376,0.7755102,0.8435374,0.9183673,0.88435376,0.93877554,1.0,1.0,0.8639456,0.8503401,0.93877554,0.877551,0.877551,0.93877554,0.93877554,0.9455782,0.93877554,1.0,0.9455782,0.93877554,0.9659864,1.0
+"AutoGluon 1.4 (best, 4h)",0.07482993,0.11564626,0.14285715,0.20408164,0.23129252,0.21088435,0.30612245,0.292517,0.2857143,0.3469388,0.3605442,0.44217688,0.47619048,0.45578232,,0.5170068,0.5782313,0.47619048,0.59183675,0.6530612,0.6258503,0.59183675,0.6258503,0.7210884,0.6666667,0.707483,0.64625853,0.65986395,0.6666667,0.59183675,0.5102041,0.5986394,0.7414966,0.7414966,0.7755102,0.6530612,0.72789115,0.81632656,0.6802721,0.707483,0.8027211,0.82312924,0.81632656,0.75510204,0.829932,0.8367347,0.8639456,0.8979592,0.8707483,0.85714287,0.877551,0.88435376,0.78911567,0.91156465,0.82312924,0.92517006,0.9047619,0.8979592,0.92517006,0.9455782,0.9591837,0.97959185,0.93877554,0.93877554,0.93877554,0.9455782,0.95238096,0.97959185,0.97959185,0.95238096,0.9659864,0.99319726,1.0
+TabM (tuned),0.17687075,0.14965986,0.20408164,0.23809524,0.25170067,0.1904762,0.30612245,0.2585034,0.3265306,0.2993197,0.3605442,0.4489796,0.36394557,0.48979592,0.4829932,,0.46258503,0.45578232,0.56462586,0.52380955,0.5510204,0.5986394,0.6122449,0.5782313,0.63265306,0.5782313,0.67346936,0.5986394,0.7414966,0.5714286,0.5714286,0.60544217,0.6802721,0.7482993,0.8027211,0.64625853,0.6938776,0.7619048,0.6122449,0.6666667,0.85714287,0.7823129,0.7755102,0.7210884,0.8367347,0.8435374,0.8707483,0.8367347,0.91156465,0.78911567,0.79591835,0.81632656,0.7755102,0.9319728,0.8435374,0.91156465,0.92517006,0.88435376,0.8503401,0.9727891,0.9727891,1.0,0.92517006,0.95238096,0.9591837,0.95238096,0.9455782,0.9727891,1.0,0.9659864,0.9863946,1.0,1.0
+CatBoost (tuned + ensembled),0.0952381,0.07482993,0.06122449,0.11564626,0.19727892,0.13605443,0.17687075,0.21768707,0.2244898,0.25170067,0.27210885,0.4217687,0.46938777,0.39455783,0.4217687,0.53741497,,0.49659863,0.48979592,0.6972789,0.52380955,0.61904764,0.6394558,0.63265306,0.5170068,0.63265306,0.5986394,0.5782313,0.64625853,0.63265306,0.5102041,0.5714286,0.67346936,0.6394558,0.7414966,0.6938776,0.7619048,0.82312924,0.7210884,0.72789115,0.829932,0.68707484,0.82312924,0.82312924,0.78911567,0.9455782,0.8707483,0.82312924,0.9455782,0.88435376,0.8707483,0.91156465,0.8979592,0.9591837,0.8639456,0.9183673,0.9727891,0.9183673,0.9455782,0.9591837,0.9659864,1.0,0.97959185,0.9727891,0.9727891,0.9727891,0.9659864,0.97959185,1.0,0.95238096,0.99319726,1.0,1.0
+TabDPT (tuned),0.11564626,0.2244898,0.2585034,0.14285715,0.2993197,0.3197279,0.23809524,0.44217688,0.26530612,0.45578232,0.39455783,0.34013605,0.5102041,0.4217687,0.52380955,0.5442177,0.50340134,,0.585034,0.5442177,0.63265306,0.5986394,0.6258503,0.6394558,0.64625853,0.6530612,0.5782313,0.5578231,0.61904764,0.67346936,0.47619048,0.7346939,0.6802721,0.63265306,0.6258503,0.70068026,0.67346936,0.70068026,0.7210884,0.82312924,0.6938776,0.6938776,0.76870745,0.7482993,0.72789115,0.8027211,0.72789115,0.7755102,0.7482993,0.8435374,0.85714287,0.877551,0.81632656,0.85714287,0.9591837,0.9183673,0.9591837,0.9591837,0.97959185,0.8095238,0.8027211,0.9319728,0.9047619,0.9047619,0.9319728,0.9455782,0.95238096,0.93877554,1.0,0.9591837,0.92517006,0.93877554,1.0
+ModernNCA (tuned + ensembled),0.08163265,0.14965986,0.14285715,0.13605443,0.14285715,0.2585034,0.1904762,0.27210885,0.2585034,0.26530612,0.36734694,0.3469388,0.3877551,0.34013605,0.40816328,0.43537414,0.5102041,0.414966,,0.5442177,0.622449,0.5714286,0.585034,0.63265306,0.53061223,0.64625853,0.5442177,0.6394558,0.5442177,0.72789115,0.53061223,0.5510204,0.67346936,0.59183675,0.6938776,0.8027211,0.79591835,0.71428573,0.8027211,0.6530612,0.7619048,0.6666667,0.9047619,0.81632656,0.7619048,0.8707483,0.79591835,0.8027211,0.88435376,0.8027211,0.877551,0.8639456,0.89115644,0.92517006,0.91156465,0.88435376,0.95238096,0.9455782,0.9863946,0.91156465,0.9591837,0.97959185,0.91156465,0.9047619,0.97959185,0.9727891,0.95238096,0.9455782,0.99319726,0.9319728,0.9659864,0.99319726,1.0
+CatBoost (tuned),0.07482993,0.07482993,0.07482993,0.10204082,0.18367347,0.108843535,0.17687075,0.18367347,0.21768707,0.24489796,0.2585034,0.3877551,0.40136054,0.40816328,0.3469388,0.47619048,0.30272108,0.45578232,0.45578232,,0.5102041,0.5782313,0.5782313,0.5782313,0.48979592,0.5578231,0.5578231,0.53061223,0.63265306,0.6122449,0.48979592,0.5578231,0.64625853,0.6530612,0.72789115,0.65986395,0.7482993,0.76870745,0.68707484,0.6802721,0.7755102,0.67346936,0.78911567,0.79591835,0.76870745,0.88435376,0.8435374,0.7823129,0.9047619,0.8435374,0.8435374,0.8435374,0.8707483,0.9319728,0.8435374,0.9183673,0.9727891,0.91156465,0.92517006,0.95238096,0.9659864,1.0,0.9659864,0.9591837,0.9727891,0.9727891,0.9659864,0.9727891,1.0,0.95238096,0.99319726,1.0,1.0
+ModernNCA (tuned),0.108843535,0.13605443,0.12244898,0.11564626,0.17006803,0.21088435,0.17687075,0.21768707,0.1904762,0.24489796,0.23129252,0.3197279,0.3877551,0.3469388,0.37414965,0.4489796,0.47619048,0.36734694,0.37755102,0.48979592,,0.5102041,0.53061223,0.5782313,0.45578232,0.5782313,0.50340134,0.585034,0.5442177,0.64625853,0.53741497,0.5442177,0.5986394,0.5714286,0.67346936,0.71428573,0.7210884,0.6938776,0.71428573,0.6802721,0.7414966,0.61904764,0.8095238,0.76870745,0.75510204,0.8027211,0.7619048,0.7346939,0.8095238,0.79591835,0.82312924,0.829932,0.85714287,0.91156465,0.9183673,0.91156465,0.9591837,0.9183673,0.9659864,0.88435376,0.9319728,0.9727891,0.92517006,0.9319728,0.9659864,0.9659864,0.9455782,0.93877554,0.9863946,0.9727891,0.9659864,0.99319726,1.0
+XGBoost (tuned),0.12244898,0.0952381,0.10204082,0.17006803,0.12244898,0.13605443,0.2244898,0.10204082,0.24489796,0.17006803,0.23809524,0.36734694,0.31292516,0.3877551,0.40816328,0.40136054,0.3809524,0.40136054,0.42857143,0.4217687,0.48979592,,0.5544218,0.5170068,0.47619048,0.5782313,0.5510204,0.5442177,0.5510204,0.5170068,0.5578231,0.5170068,0.61904764,0.64625853,0.7346939,0.59183675,0.6802721,0.7482993,0.61904764,0.5714286,0.7823129,0.6666667,0.7823129,0.7346939,0.72789115,0.93877554,0.8435374,0.7346939,0.8435374,0.877551,0.88435376,0.88435376,0.7823129,0.9183673,0.8435374,0.92517006,0.9591837,0.91156465,0.92517006,0.9455782,0.9591837,0.99319726,0.95238096,0.9455782,0.9591837,0.9591837,0.9659864,0.9863946,0.99319726,0.9319728,0.9863946,1.0,1.0
+XGBoost (tuned + ensembled),0.11564626,0.0952381,0.08163265,0.14285715,0.10204082,0.13605443,0.21768707,0.07482993,0.23129252,0.14285715,0.24489796,0.3469388,0.2857143,0.3877551,0.37414965,0.3877551,0.3605442,0.37414965,0.414966,0.4217687,0.46938777,0.44557822,,0.49659863,0.48979592,0.53741497,0.5510204,0.53741497,0.53741497,0.49659863,0.5442177,0.5102041,0.6258503,0.64625853,0.7619048,0.5986394,0.6802721,0.76870745,0.6122449,0.585034,0.8095238,0.67346936,0.7823129,0.707483,0.7346939,0.9319728,0.8435374,0.7414966,0.8435374,0.8639456,0.877551,0.89115644,0.7823129,0.9183673,0.8435374,0.92517006,0.95238096,0.9047619,0.91156465,0.9455782,0.9591837,0.99319726,0.9591837,0.95238096,0.9591837,0.95238096,0.9591837,0.9863946,0.99319726,0.9319728,0.9863946,1.0,1.0
+LightGBM (tuned + ensembled),0.0952381,0.11564626,0.108843535,0.0952381,0.14965986,0.14965986,0.15646258,0.12244898,0.15646258,0.21088435,0.21088435,0.2857143,0.34013605,0.3265306,0.27891156,0.4217687,0.36734694,0.3605442,0.36734694,0.4217687,0.4217687,0.4829932,0.50340134,,0.48979592,0.47959185,0.5102041,0.52380955,0.59183675,0.49659863,0.53061223,0.5578231,0.59183675,0.60544217,0.71428573,0.585034,0.68707484,0.72789115,0.56462586,0.61904764,0.7755102,0.64625853,0.75510204,0.71428573,0.72789115,0.8435374,0.8095238,0.7823129,0.829932,0.93877554,0.9455782,0.9863946,0.75510204,0.88435376,0.8503401,0.97959185,0.9183673,0.8979592,0.9455782,0.88435376,0.92517006,1.0,0.9455782,0.9455782,0.93877554,0.95238096,0.93877554,1.0,1.0,0.95238096,0.99319726,1.0,1.0
+iLTM (tuned + ensembled),0.08163265,0.12244898,0.11564626,0.088435374,0.07482993,0.20408164,0.14965986,0.19727892,0.1292517,0.23129252,0.25170067,0.24489796,0.37414965,0.20408164,0.33333334,0.36734694,0.4829932,0.3537415,0.46938777,0.5102041,0.5442177,0.52380955,0.5102041,0.5102041,,0.53061223,0.47619048,0.56462586,0.48979592,0.61904764,0.5102041,0.53061223,0.67346936,0.52380955,0.64625853,0.67346936,0.6802721,0.6666667,0.70068026,0.61904764,0.707483,0.65986395,0.79591835,0.7619048,0.6802721,0.829932,0.76870745,0.7755102,0.81632656,0.829932,0.89115644,0.9047619,0.7755102,0.85714287,0.8503401,0.9319728,0.8639456,0.9183673,0.99319726,0.81632656,0.88435376,0.99319726,0.8639456,0.85714287,0.89115644,0.91156465,0.9183673,0.99319726,1.0,0.8979592,0.9727891,0.99319726,1.0
+LightGBM (tuned),0.10204082,0.088435374,0.10204082,0.108843535,0.11564626,0.1292517,0.1632653,0.12244898,0.14285715,0.19727892,0.20408164,0.30612245,0.33333334,0.3809524,0.292517,0.4217687,0.36734694,0.3469388,0.3537415,0.44217688,0.4217687,0.4217687,0.46258503,0.52040815,0.46938777,,0.48979592,0.50340134,0.5782313,0.47619048,0.5170068,0.56462586,0.5782313,0.63265306,0.6938776,0.5442177,0.65986395,0.7414966,0.5782313,0.6122449,0.7823129,0.63265306,0.7346939,0.6938776,0.72789115,0.8503401,0.81632656,0.7619048,0.81632656,0.9183673,0.9319728,0.9455782,0.76870745,0.8639456,0.85714287,0.9659864,0.9319728,0.88435376,0.9455782,0.89115644,0.92517006,1.0,0.95238096,0.9591837,0.93877554,0.95238096,0.93877554,1.0,1.0,0.95238096,0.99319726,1.0,1.0
+BetaTabPFN (default),0.088435374,0.10204082,0.088435374,0.17687075,0.1632653,0.14965986,0.2585034,0.14285715,0.26530612,0.15646258,0.25170067,0.36734694,0.23809524,0.4489796,0.3537415,0.3265306,0.40136054,0.4217687,0.45578232,0.44217688,0.49659863,0.4489796,0.4489796,0.48979592,0.52380955,0.5102041,,0.52380955,0.5102041,0.50340134,0.44217688,0.5714286,0.59183675,0.71428573,0.70068026,0.56462586,0.61904764,0.67346936,0.5714286,0.5714286,0.8095238,0.71428573,0.71428573,0.61904764,0.7823129,0.7414966,0.82312924,0.7755102,0.85714287,0.72789115,0.7482993,0.7755102,0.71428573,0.93877554,0.79591835,0.8979592,0.877551,0.82312924,0.81632656,0.93877554,0.93877554,0.99319726,0.95238096,0.95238096,0.9319728,0.9591837,0.93877554,0.99319726,1.0,0.9727891,0.9863946,1.0,1.0
+xRFM (tuned + ensembled),0.08163265,0.108843535,0.12244898,0.0952381,0.21088435,0.18367347,0.1632653,0.17687075,0.21768707,0.23809524,0.1632653,0.36734694,0.3809524,0.39455783,0.34013605,0.40136054,0.4217687,0.44217688,0.3605442,0.46938777,0.414966,0.45578232,0.46258503,0.47619048,0.43537414,0.49659863,0.47619048,,0.4829932,0.52380955,0.49659863,0.5714286,0.63265306,0.59183675,0.5102041,0.5986394,0.67346936,0.65986395,0.6394558,0.72789115,0.64625853,0.56462586,0.67346936,0.70068026,0.6938776,0.7482993,0.75510204,0.65986395,0.707483,0.707483,0.8639456,0.88435376,0.9591837,0.8503401,0.91156465,0.93877554,0.9727891,0.9659864,0.99319726,0.9591837,0.9659864,0.8979592,0.9659864,0.9659864,0.91156465,0.89115644,0.89115644,0.9659864,0.99319726,0.9863946,0.9863946,1.0,1.0
+TabM (default),0.14965986,0.14285715,0.15646258,0.19727892,0.21088435,0.18367347,0.2993197,0.1292517,0.26530612,0.19727892,0.292517,0.36734694,0.12244898,0.44217688,0.33333334,0.2585034,0.3537415,0.3809524,0.45578232,0.36734694,0.45578232,0.4489796,0.46258503,0.40816328,0.5102041,0.4217687,0.48979592,0.5170068,,0.46258503,0.46258503,0.5510204,0.56462586,0.64625853,0.64625853,0.52380955,0.5782313,0.61904764,0.52380955,0.49659863,0.7210884,0.7346939,0.65986395,0.5782313,0.7755102,0.707483,0.7619048,0.8027211,0.8435374,0.64625853,0.7346939,0.7346939,0.68707484,0.91156465,0.7619048,0.81632656,0.8367347,0.8095238,0.8027211,0.91156465,0.9319728,0.9863946,0.92517006,0.9319728,0.93877554,0.9319728,0.92517006,0.97959185,1.0,0.9591837,0.97959185,1.0,1.0
+ExtraTrees (tuned + ensembled),0.07482993,0.13605443,0.15646258,0.10204082,0.15646258,0.19727892,0.0952381,0.21088435,0.18367347,0.21768707,0.11564626,0.26530612,0.4217687,0.24489796,0.40816328,0.42857143,0.36734694,0.3265306,0.27210885,0.3877551,0.3537415,0.4829932,0.50340134,0.50340134,0.3809524,0.52380955,0.49659863,0.47619048,0.53741497,,0.47619048,0.5170068,0.53061223,0.4489796,0.53061223,0.7176871,0.5578231,0.6122449,0.76870745,0.67346936,0.60544217,0.53061223,0.6122449,0.8503401,0.59183675,0.70068026,0.70068026,0.6122449,0.6666667,0.78911567,0.8095238,0.81632656,0.8095238,0.7755102,0.95238096,0.8707483,0.9455782,0.9659864,1.0,0.7619048,0.7619048,0.91156465,0.91156465,0.91156465,0.92517006,0.95238096,0.93877554,0.8979592,0.97959185,1.0,0.9319728,0.9455782,1.0
+SAP-RPT-OSS (default),0.24489796,0.25170067,0.2585034,0.37414965,0.3469388,0.2857143,0.42857143,0.3469388,0.48979592,0.3605442,0.414966,0.5102041,0.3605442,0.50340134,0.48979592,0.42857143,0.48979592,0.52380955,0.46938777,0.5102041,0.46258503,0.44217688,0.45578232,0.46938777,0.48979592,0.4829932,0.5578231,0.50340134,0.53741497,0.52380955,,0.52380955,0.53061223,0.68707484,0.5782313,0.53061223,0.5714286,0.5442177,0.52380955,0.5578231,0.60544217,0.6530612,0.5782313,0.5510204,0.70068026,0.53741497,0.63265306,0.7210884,0.7414966,0.59183675,0.53061223,0.5102041,0.585034,0.71428573,0.56462586,0.6394558,0.6530612,0.6530612,0.63265306,0.8027211,0.8027211,0.7210884,0.7346939,0.7210884,0.75510204,0.7482993,0.7482993,0.78911567,0.7482993,0.7619048,0.9455782,0.8095238,1.0
+TabDPT (default),0.0952381,0.20408164,0.2244898,0.1632653,0.2244898,0.2993197,0.18367347,0.39455783,0.25170067,0.3809524,0.39455783,0.25170067,0.40136054,0.21088435,0.40136054,0.39455783,0.42857143,0.26530612,0.4489796,0.44217688,0.45578232,0.4829932,0.48979592,0.44217688,0.46938777,0.43537414,0.42857143,0.42857143,0.4489796,0.4829932,0.47619048,,0.52380955,0.45578232,0.46938777,0.5170068,0.49659863,0.45578232,0.5714286,0.5986394,0.49659863,0.5578231,0.5782313,0.6258503,0.52380955,0.5578231,0.50340134,0.6122449,0.5714286,0.7346939,0.65986395,0.63265306,0.6938776,0.63265306,0.8503401,0.82312924,0.8095238,0.8027211,0.9183673,0.5782313,0.64625853,0.9183673,0.7619048,0.7482993,0.8503401,0.8707483,0.88435376,0.9047619,1.0,0.877551,0.89115644,0.9319728,1.0
+iLTM (tuned),0.06802721,0.04761905,0.04761905,0.06122449,0.054421768,0.10204082,0.088435374,0.0952381,0.12244898,0.11564626,0.17006803,0.21088435,0.2585034,0.19727892,0.2585034,0.3197279,0.3265306,0.3197279,0.3265306,0.3537415,0.40136054,0.3809524,0.37414965,0.40816328,0.3265306,0.4217687,0.40816328,0.36734694,0.43537414,0.46938777,0.46938777,0.47619048,,0.4217687,0.48979592,0.5714286,0.53061223,0.5442177,0.59183675,0.5170068,0.59183675,0.6258503,0.64625853,0.71428573,0.6122449,0.67346936,0.6666667,0.67346936,0.72789115,0.707483,0.7755102,0.7346939,0.72789115,0.8503401,0.7619048,0.8367347,0.8639456,0.8639456,0.92517006,0.76870745,0.829932,0.97959185,0.8435374,0.8435374,0.89115644,0.91156465,0.9047619,0.93877554,0.9727891,0.88435376,0.9455782,0.9727891,1.0
+TorchMLP (tuned + ensembled),0.07482993,0.054421768,0.06122449,0.0952381,0.08163265,0.10204082,0.12244898,0.19727892,0.17687075,0.14965986,0.2993197,0.3197279,0.2244898,0.2993197,0.2585034,0.25170067,0.3605442,0.36734694,0.40816328,0.3469388,0.42857143,0.3537415,0.3537415,0.39455783,0.47619048,0.36734694,0.2857143,0.40816328,0.3537415,0.5510204,0.31292516,0.5442177,0.5782313,,0.3809524,0.5782313,0.5170068,0.5170068,0.585034,0.5170068,0.4829932,0.6938776,0.6394558,0.59183675,0.81292516,0.61904764,0.5714286,0.79591835,0.6938776,0.6394558,0.6530612,0.6530612,0.6530612,0.89115644,0.72789115,0.68707484,0.7619048,0.7414966,0.78911567,0.8707483,0.9183673,0.91156465,0.8979592,0.9047619,0.8367347,0.877551,0.8707483,0.9591837,0.99319726,0.93877554,0.9727891,0.99319726,1.0
+EBM (tuned + ensembled),0.0952381,0.04761905,0.06802721,0.08163265,0.13605443,0.0952381,0.17006803,0.08163265,0.17006803,0.10204082,0.17687075,0.26530612,0.13605443,0.33333334,0.2244898,0.19727892,0.2585034,0.37414965,0.30612245,0.27210885,0.3265306,0.26530612,0.23809524,0.2857143,0.3537415,0.30612245,0.2993197,0.48979592,0.3537415,0.46938777,0.4217687,0.53061223,0.5102041,0.61904764,,0.5102041,0.5986394,0.47619048,0.5102041,0.5170068,0.7585034,0.5578231,0.61904764,0.5510204,0.68707484,0.5986394,0.8435374,0.67346936,0.7346939,0.59183675,0.6938776,0.707483,0.7346939,0.9183673,0.7482993,0.829932,0.8707483,0.78911567,0.8027211,0.93877554,0.9727891,0.9455782,0.9319728,0.93877554,0.9183673,0.9319728,0.9319728,0.97959185,0.99319726,0.93877554,0.9863946,0.99319726,1.0
+RandomForest (tuned + ensembled),0.0952381,0.12244898,0.1292517,0.0952381,0.17687075,0.17687075,0.12244898,0.13605443,0.1292517,0.1632653,0.0952381,0.20408164,0.3605442,0.21768707,0.3469388,0.3537415,0.30612245,0.2993197,0.19727892,0.34013605,0.2857143,0.40816328,0.40136054,0.414966,0.3265306,0.45578232,0.43537414,0.40136054,0.47619048,0.28231293,0.46938777,0.4829932,0.42857143,0.4217687,0.48979592,,0.52380955,0.5170068,0.5510204,0.5986394,0.5442177,0.47619048,0.53061223,0.79931974,0.53061223,0.63265306,0.64625853,0.5170068,0.59183675,0.72789115,0.76870745,0.7619048,0.7482993,0.72789115,0.9047619,0.8503401,0.8979592,0.9863946,0.95238096,0.707483,0.71428573,0.8435374,0.92517006,0.92517006,0.91156465,0.9319728,0.9319728,0.88435376,0.9455782,0.99319726,0.93877554,0.9183673,1.0
+xRFM (tuned),0.07482993,0.08163265,0.08163265,0.06122449,0.14965986,0.1292517,0.0952381,0.14965986,0.18367347,0.17006803,0.10204082,0.2585034,0.2993197,0.3265306,0.27210885,0.30612245,0.23809524,0.3265306,0.20408164,0.25170067,0.27891156,0.3197279,0.3197279,0.31292516,0.3197279,0.34013605,0.3809524,0.3265306,0.4217687,0.44217688,0.42857143,0.50340134,0.46938777,0.4829932,0.40136054,0.47619048,,0.44217688,0.53741497,0.6258503,0.5102041,0.49659863,0.53061223,0.585034,0.59183675,0.56462586,0.61904764,0.53741497,0.5578231,0.63265306,0.7414966,0.71428573,0.9183673,0.79591835,0.8707483,0.88435376,0.9455782,0.9183673,0.9455782,0.8707483,0.877551,0.85714287,0.92517006,0.9319728,0.89115644,0.9183673,0.9047619,0.92517006,0.99319726,0.9727891,0.97959185,0.99319726,1.0
+CatBoost (default),0.06802721,0.034013607,0.06122449,0.054421768,0.04761905,0.088435374,0.11564626,0.06802721,0.11564626,0.108843535,0.12244898,0.21088435,0.17006803,0.2585034,0.18367347,0.23809524,0.17687075,0.2993197,0.2857143,0.23129252,0.30612245,0.25170067,0.23129252,0.27210885,0.33333334,0.2585034,0.3265306,0.34013605,0.3809524,0.3877551,0.45578232,0.5442177,0.45578232,0.4829932,0.52380955,0.4829932,0.5578231,,0.48979592,0.48979592,0.6666667,0.56462586,0.60544217,0.60544217,0.63265306,0.67346936,0.70068026,0.6666667,0.707483,0.72789115,0.78911567,0.8027211,0.70068026,0.85714287,0.79591835,0.93877554,0.89115644,0.8435374,0.88435376,0.8367347,0.8639456,1.0,0.9183673,0.91156465,0.9183673,0.9455782,0.9183673,0.9659864,1.0,0.9455782,0.99319726,1.0,1.0
+ExtraTrees (tuned),0.08163265,0.108843535,0.1292517,0.10204082,0.1632653,0.15646258,0.088435374,0.14285715,0.1632653,0.17687075,0.0952381,0.21088435,0.36734694,0.23129252,0.3197279,0.3877551,0.27891156,0.27891156,0.19727892,0.31292516,0.2857143,0.3809524,0.3877551,0.43537414,0.2993197,0.4217687,0.42857143,0.3605442,0.47619048,0.23129252,0.47619048,0.42857143,0.40816328,0.414966,0.48979592,0.4489796,0.46258503,0.5102041,,0.6122449,0.56462586,0.48979592,0.4829932,0.6360544,0.5170068,0.5510204,0.60544217,0.53741497,0.56462586,0.7346939,0.6802721,0.68707484,0.71428573,0.70068026,0.9319728,0.8503401,0.91156465,0.89115644,0.88435376,0.707483,0.68707484,0.8503401,0.91156465,0.9047619,0.91156465,0.91156465,0.9183673,0.88435376,0.9659864,1.0,0.9319728,0.92517006,1.0
+TabSwift (default),0.08163265,0.027210884,0.027210884,0.013605442,0.04761905,0.06802721,0.054421768,0.21088435,0.08163265,0.21088435,0.1904762,0.13605443,0.31292516,0.17687075,0.292517,0.33333334,0.27210885,0.17687075,0.3469388,0.3197279,0.3197279,0.42857143,0.414966,0.3809524,0.3809524,0.3877551,0.42857143,0.27210885,0.50340134,0.3265306,0.44217688,0.40136054,0.4829932,0.4829932,0.4829932,0.40136054,0.37414965,0.5102041,0.3877551,,0.5442177,0.6394558,0.52380955,0.56462586,0.5986394,0.5442177,0.5782313,0.65986395,0.65986395,0.7482993,0.6122449,0.60544217,0.5986394,0.6802721,0.8435374,0.7823129,0.92517006,0.85714287,0.7619048,0.65986395,0.6938776,0.8367347,0.8707483,0.85714287,0.9319728,0.9455782,0.9455782,0.8707483,0.95238096,0.95238096,0.9319728,0.9659864,1.0
+EBM (tuned),0.07482993,0.040816326,0.06122449,0.054421768,0.0952381,0.088435374,0.14965986,0.07482993,0.14285715,0.06122449,0.12244898,0.23129252,0.108843535,0.27210885,0.19727892,0.14285715,0.17006803,0.30612245,0.23809524,0.2244898,0.2585034,0.21768707,0.1904762,0.2244898,0.292517,0.21768707,0.1904762,0.3537415,0.27891156,0.39455783,0.39455783,0.50340134,0.40816328,0.5170068,0.2414966,0.45578232,0.48979592,0.33333334,0.43537414,0.45578232,,0.5102041,0.5170068,0.52380955,0.63265306,0.5102041,0.70068026,0.6258503,0.64625853,0.49659863,0.63265306,0.65986395,0.68707484,0.877551,0.6938776,0.7823129,0.81632656,0.75510204,0.7755102,0.89115644,0.92517006,0.9047619,0.9183673,0.92517006,0.8979592,0.91156465,0.9047619,0.9659864,0.99319726,0.93877554,0.97959185,0.99319726,1.0
+FastaiMLP (tuned + ensembled),0.10204082,0.07482993,0.06802721,0.1292517,0.054421768,0.08163265,0.17006803,0.0952381,0.18367347,0.1292517,0.23129252,0.23129252,0.18367347,0.2244898,0.17687075,0.21768707,0.31292516,0.30612245,0.33333334,0.3265306,0.3809524,0.33333334,0.3265306,0.3537415,0.34013605,0.36734694,0.2857143,0.43537414,0.26530612,0.46938777,0.3469388,0.44217688,0.37414965,0.30612245,0.44217688,0.52380955,0.50340134,0.43537414,0.5102041,0.3605442,0.48979592,,0.5986394,0.53741497,0.50340134,0.5578231,0.5102041,0.75510204,0.59183675,0.585034,0.68707484,0.67346936,0.5578231,0.67346936,0.63265306,0.72789115,0.63265306,0.64625853,0.7823129,0.8095238,0.85714287,0.78911567,0.76870745,0.76870745,0.6802721,0.7210884,0.7210884,0.99319726,0.9183673,0.89115644,0.95238096,1.0,0.99319726
+ModernNCA (default),0.020408163,0.054421768,0.06802721,0.034013607,0.06802721,0.0952381,0.034013607,0.054421768,0.04761905,0.08163265,0.06122449,0.14965986,0.1904762,0.17006803,0.18367347,0.2244898,0.17687075,0.23129252,0.0952381,0.21088435,0.1904762,0.21768707,0.21768707,0.24489796,0.20408164,0.26530612,0.2857143,0.3265306,0.34013605,0.3877551,0.4217687,0.4217687,0.3537415,0.3605442,0.3809524,0.46938777,0.46938777,0.39455783,0.5170068,0.47619048,0.4829932,0.40136054,,0.63265306,0.4829932,0.50340134,0.5714286,0.4829932,0.50340134,0.6530612,0.7414966,0.7619048,0.6666667,0.72789115,0.81632656,0.8503401,0.81632656,0.8503401,0.8367347,0.76870745,0.8095238,0.8367347,0.7823129,0.7823129,0.877551,0.9047619,0.8979592,0.85714287,0.92517006,0.8979592,0.8639456,0.97959185,1.0
+RandomForest (tuned),0.0952381,0.0952381,0.0952381,0.0952381,0.13605443,0.088435374,0.10204082,0.088435374,0.1292517,0.14285715,0.10204082,0.1632653,0.24489796,0.21088435,0.24489796,0.27891156,0.17687075,0.25170067,0.18367347,0.20408164,0.23129252,0.26530612,0.292517,0.2857143,0.23809524,0.30612245,0.3809524,0.2993197,0.4217687,0.14965986,0.4489796,0.37414965,0.2857143,0.40816328,0.4489796,0.20068027,0.414966,0.39455783,0.36394557,0.43537414,0.47619048,0.46258503,0.36734694,,0.46258503,0.47619048,0.585034,0.5102041,0.5170068,0.64625853,0.5986394,0.60544217,0.61904764,0.65986395,0.8435374,0.79591835,0.85714287,0.8197279,0.78911567,0.67346936,0.6666667,0.79591835,0.877551,0.8707483,0.89115644,0.9183673,0.91156465,0.8707483,0.92517006,0.9727891,0.9183673,0.8979592,1.0
+TorchMLP (tuned),0.06802721,0.054421768,0.054421768,0.04761905,0.08163265,0.10204082,0.108843535,0.14285715,0.12244898,0.10204082,0.18367347,0.1904762,0.1292517,0.23129252,0.17006803,0.1632653,0.21088435,0.27210885,0.23809524,0.23129252,0.24489796,0.27210885,0.26530612,0.27210885,0.3197279,0.27210885,0.21768707,0.30612245,0.2244898,0.40816328,0.2993197,0.47619048,0.3877551,0.18707483,0.31292516,0.46938777,0.40816328,0.36734694,0.4829932,0.40136054,0.36734694,0.49659863,0.5170068,0.53741497,,0.49659863,0.3809524,0.6394558,0.5102041,0.5102041,0.5578231,0.5578231,0.5986394,0.8095238,0.68707484,0.65986395,0.6802721,0.707483,0.7346939,0.7346939,0.78911567,0.85714287,0.8639456,0.85714287,0.7823129,0.829932,0.81632656,0.92517006,0.9659864,0.93877554,0.95238096,0.9863946,1.0
+XGBoost (default),0.06802721,0.020408163,0.013605442,0.006802721,0.020408163,0.054421768,0.020408163,0.006802721,0.07482993,0.040816326,0.054421768,0.10204082,0.11564626,0.13605443,0.1632653,0.15646258,0.054421768,0.19727892,0.1292517,0.11564626,0.19727892,0.06122449,0.06802721,0.15646258,0.17006803,0.14965986,0.2585034,0.25170067,0.292517,0.2993197,0.46258503,0.44217688,0.3265306,0.3809524,0.40136054,0.36734694,0.43537414,0.3265306,0.4489796,0.45578232,0.48979592,0.44217688,0.49659863,0.52380955,0.50340134,,0.6122449,0.52380955,0.5782313,0.6802721,0.64625853,0.6802721,0.67346936,0.75510204,0.8027211,0.8639456,0.88435376,0.8639456,0.8639456,0.7210884,0.7619048,0.9591837,0.877551,0.8639456,0.89115644,0.92517006,0.9183673,0.9455782,0.9863946,0.92517006,0.9863946,1.0,1.0
+EBM (default),0.07482993,0.054421768,0.06802721,0.06122449,0.07482993,0.07482993,0.08163265,0.06802721,0.0952381,0.08163265,0.10204082,0.20408164,0.108843535,0.23129252,0.13605443,0.1292517,0.1292517,0.27210885,0.20408164,0.15646258,0.23809524,0.15646258,0.15646258,0.1904762,0.23129252,0.18367347,0.17687075,0.24489796,0.23809524,0.2993197,0.36734694,0.49659863,0.33333334,0.42857143,0.15646258,0.3537415,0.3809524,0.2993197,0.39455783,0.4217687,0.2993197,0.48979592,0.42857143,0.414966,0.61904764,0.3877551,,0.5714286,0.52380955,0.47619048,0.60544217,0.6258503,0.61904764,0.76870745,0.75510204,0.7482993,0.78911567,0.7210884,0.75510204,0.7823129,0.85714287,0.8639456,0.9047619,0.91156465,0.877551,0.8979592,0.8707483,0.9659864,0.97959185,0.9727891,0.9727891,1.0,1.0
+FastaiMLP (tuned),0.06122449,0.034013607,0.04761905,0.06122449,0.020408163,0.06122449,0.06122449,0.088435374,0.10204082,0.06802721,0.11564626,0.18367347,0.12244898,0.13605443,0.10204082,0.1632653,0.17687075,0.2244898,0.19727892,0.21768707,0.26530612,0.26530612,0.2585034,0.21768707,0.2244898,0.23809524,0.2244898,0.34013605,0.19727892,0.3877551,0.27891156,0.3877551,0.3265306,0.20408164,0.3265306,0.4829932,0.46258503,0.33333334,0.46258503,0.34013605,0.37414965,0.24489796,0.5170068,0.48979592,0.3605442,0.47619048,0.42857143,,0.4829932,0.53061223,0.6258503,0.5986394,0.56462586,0.64625853,0.65986395,0.70068026,0.61904764,0.63265306,0.75510204,0.7482993,0.79591835,0.78911567,0.7346939,0.7482993,0.65986395,0.70068026,0.707483,0.9863946,0.829932,0.89115644,0.93877554,1.0,0.99319726
+RealMLP (default),0.020408163,0.006802721,0.013605442,0.020408163,0.034013607,0.006802721,0.013605442,0.006802721,0.07482993,0.020408163,0.020408163,0.14285715,0.04761905,0.15646258,0.1292517,0.088435374,0.054421768,0.25170067,0.11564626,0.0952381,0.1904762,0.15646258,0.15646258,0.17006803,0.18367347,0.18367347,0.14285715,0.292517,0.15646258,0.33333334,0.2585034,0.42857143,0.27210885,0.30612245,0.26530612,0.40816328,0.44217688,0.292517,0.43537414,0.34013605,0.3537415,0.40816328,0.49659863,0.4829932,0.48979592,0.4217687,0.47619048,0.5170068,,0.5102041,0.5442177,0.56462586,0.5578231,0.7414966,0.68707484,0.6530612,0.6666667,0.72789115,0.7482993,0.7823129,0.8435374,0.829932,0.8367347,0.8367347,0.829932,0.89115644,0.8435374,0.9319728,0.92517006,0.92517006,0.93877554,0.99319726,1.0
+LightGBM (default),0.07482993,0.054421768,0.06802721,0.027210884,0.06802721,0.088435374,0.07482993,0.07482993,0.07482993,0.08163265,0.06122449,0.14285715,0.11564626,0.14965986,0.14285715,0.21088435,0.11564626,0.15646258,0.19727892,0.15646258,0.20408164,0.12244898,0.13605443,0.06122449,0.17006803,0.08163265,0.27210885,0.292517,0.3537415,0.21088435,0.40816328,0.26530612,0.292517,0.3605442,0.40816328,0.27210885,0.36734694,0.27210885,0.26530612,0.25170067,0.50340134,0.414966,0.3469388,0.3537415,0.48979592,0.3197279,0.52380955,0.46938777,0.48979592,,0.47619048,0.48979592,0.5102041,0.64625853,0.7619048,0.7619048,0.7414966,0.6530612,0.6258503,0.5986394,0.64625853,0.82312924,0.8027211,0.7755102,0.8707483,0.91156465,0.877551,0.88435376,0.88435376,0.92517006,0.93877554,0.97959185,1.0
+ChimeraBoost (tuned),0.07482993,0.04761905,0.06122449,0.027210884,0.06802721,0.088435374,0.0952381,0.06122449,0.08163265,0.08163265,0.088435374,0.108843535,0.14285715,0.11564626,0.12244898,0.20408164,0.1292517,0.14285715,0.12244898,0.15646258,0.17687075,0.11564626,0.12244898,0.054421768,0.108843535,0.06802721,0.25170067,0.13605443,0.26530612,0.1904762,0.46938777,0.34013605,0.2244898,0.3469388,0.30612245,0.23129252,0.2585034,0.21088435,0.3197279,0.3877551,0.36734694,0.31292516,0.2585034,0.40136054,0.44217688,0.3537415,0.39455783,0.37414965,0.45578232,0.52380955,,0.4727891,0.6394558,0.5986394,0.63265306,0.92517006,0.7414966,0.61904764,0.68707484,0.6802721,0.70068026,0.82312924,0.7823129,0.78911567,0.71428573,0.7755102,0.7619048,0.9455782,0.9319728,0.85714287,0.91156465,0.9727891,1.0
+ChimeraBoost (tuned + ensembled),0.07482993,0.034013607,0.04761905,0.013605442,0.054421768,0.08163265,0.07482993,0.054421768,0.07482993,0.06122449,0.06122449,0.08163265,0.11564626,0.11564626,0.11564626,0.18367347,0.088435374,0.12244898,0.13605443,0.15646258,0.17006803,0.11564626,0.108843535,0.013605442,0.0952381,0.054421768,0.2244898,0.11564626,0.26530612,0.18367347,0.48979592,0.36734694,0.26530612,0.3469388,0.292517,0.23809524,0.2857143,0.19727892,0.31292516,0.39455783,0.34013605,0.3265306,0.23809524,0.39455783,0.44217688,0.3197279,0.37414965,0.40136054,0.43537414,0.5102041,0.5272109,,0.65986395,0.59183675,0.63265306,0.92517006,0.7482993,0.61904764,0.70068026,0.6666667,0.707483,0.829932,0.78911567,0.79591835,0.71428573,0.79591835,0.7755102,0.9659864,0.9659864,0.85714287,0.93877554,0.97959185,1.0
+xRFM (default),0.07482993,0.020408163,0.034013607,0.013605442,0.06802721,0.06802721,0.020408163,0.06802721,0.108843535,0.0952381,0.027210884,0.14965986,0.19727892,0.2244898,0.21088435,0.2244898,0.10204082,0.18367347,0.108843535,0.1292517,0.14285715,0.21768707,0.21768707,0.24489796,0.2244898,0.23129252,0.2857143,0.040816326,0.31292516,0.1904762,0.414966,0.30612245,0.27210885,0.3469388,0.26530612,0.25170067,0.08163265,0.2993197,0.2857143,0.40136054,0.31292516,0.44217688,0.33333334,0.3809524,0.40136054,0.3265306,0.3809524,0.43537414,0.44217688,0.48979592,0.3605442,0.34013605,,0.49659863,0.6258503,0.5578231,0.65986395,0.71428573,0.53061223,0.65986395,0.6666667,0.585034,0.8095238,0.8027211,0.7414966,0.7755102,0.7755102,0.72789115,0.7482993,0.9183673,0.97959185,0.9319728,1.0
+TorchMLP (default),0.06802721,0.006802721,0.013605442,0.013605442,0.054421768,0.040816326,0.034013607,0.027210884,0.108843535,0.034013607,0.040816326,0.108843535,0.054421768,0.15646258,0.088435374,0.06802721,0.040816326,0.14285715,0.07482993,0.06802721,0.088435374,0.08163265,0.08163265,0.11564626,0.14285715,0.13605443,0.06122449,0.14965986,0.088435374,0.2244898,0.2857143,0.36734694,0.14965986,0.108843535,0.08163265,0.27210885,0.20408164,0.14285715,0.2993197,0.3197279,0.12244898,0.3265306,0.27210885,0.34013605,0.1904762,0.24489796,0.23129252,0.3537415,0.2585034,0.3537415,0.40136054,0.40816328,0.50340134,,0.61904764,0.5578231,0.5986394,0.6122449,0.6394558,0.6666667,0.71428573,0.52380955,0.75510204,0.7619048,0.6258503,0.6938776,0.6666667,0.8503401,0.8027211,0.92517006,0.89115644,0.93877554,1.0
+KNN (tuned + ensembled),0.06802721,0.06802721,0.07482993,0.020408163,0.04761905,0.12244898,0.054421768,0.08163265,0.06802721,0.07482993,0.054421768,0.034013607,0.11564626,0.08163265,0.17687075,0.15646258,0.13605443,0.040816326,0.088435374,0.15646258,0.08163265,0.15646258,0.15646258,0.14965986,0.14965986,0.14285715,0.20408164,0.088435374,0.23809524,0.04761905,0.43537414,0.14965986,0.23809524,0.27210885,0.25170067,0.0952381,0.1292517,0.20408164,0.06802721,0.15646258,0.30612245,0.36734694,0.18367347,0.15646258,0.31292516,0.19727892,0.24489796,0.34013605,0.31292516,0.23809524,0.36734694,0.36734694,0.37414965,0.3809524,,0.5714286,0.45578232,0.414966,0.44217688,0.48979592,0.49659863,0.4829932,0.4829932,0.49659863,0.68707484,0.72789115,0.71428573,0.63265306,0.585034,0.9319728,0.6802721,0.7482993,0.99319726
+ChimeraBoost (default),0.06122449,0.040816326,0.06122449,0.013605442,0.054421768,0.06802721,0.06802721,0.06122449,0.06802721,0.054421768,0.054421768,0.088435374,0.04761905,0.11564626,0.07482993,0.088435374,0.08163265,0.08163265,0.11564626,0.08163265,0.088435374,0.07482993,0.07482993,0.020408163,0.06802721,0.034013607,0.10204082,0.06122449,0.18367347,0.1292517,0.3605442,0.17687075,0.1632653,0.31292516,0.17006803,0.14965986,0.11564626,0.06122449,0.14965986,0.21768707,0.21768707,0.27210885,0.14965986,0.20408164,0.34013605,0.13605443,0.25170067,0.2993197,0.3469388,0.23809524,0.07482993,0.07482993,0.44217688,0.44217688,0.42857143,,0.53061223,0.46938777,0.27210885,0.585034,0.59183675,0.67346936,0.70068026,0.7210884,0.6258503,0.6802721,0.65986395,0.7346939,0.64625853,0.8435374,0.8095238,0.9047619,1.0
+OrionMSP (default),0.06802721,0.020408163,0.013605442,0.006802721,0.034013607,0.013605442,0.006802721,0.013605442,0.088435374,0.020408163,0.027210884,0.040816326,0.06122449,0.06122449,0.0952381,0.07482993,0.027210884,0.040816326,0.04761905,0.027210884,0.040816326,0.040816326,0.04761905,0.08163265,0.13605443,0.06802721,0.12244898,0.027210884,0.1632653,0.054421768,0.3469388,0.1904762,0.13605443,0.23809524,0.1292517,0.10204082,0.054421768,0.108843535,0.088435374,0.07482993,0.18367347,0.36734694,0.18367347,0.14285715,0.3197279,0.11564626,0.21088435,0.3809524,0.33333334,0.2585034,0.2585034,0.25170067,0.34013605,0.40136054,0.5442177,0.46938777,,0.56462586,0.44217688,0.5442177,0.56462586,0.5170068,0.68707484,0.6666667,0.65986395,0.75510204,0.7346939,0.7346939,0.7482993,0.91156465,0.89115644,0.89115644,1.0
+RandomForest (default),0.006802721,0.013605442,0.013605442,0.006802721,0.0,0.040816326,0.0,0.027210884,0.006802721,0.027210884,0.0,0.020408163,0.108843535,0.0,0.10204082,0.11564626,0.08163265,0.040816326,0.054421768,0.088435374,0.08163265,0.088435374,0.0952381,0.10204082,0.08163265,0.11564626,0.17687075,0.034013607,0.1904762,0.034013607,0.3469388,0.19727892,0.13605443,0.2585034,0.21088435,0.013605442,0.08163265,0.15646258,0.108843535,0.14285715,0.24489796,0.3537415,0.14965986,0.1802721,0.292517,0.13605443,0.27891156,0.36734694,0.27210885,0.3469388,0.3809524,0.3809524,0.2857143,0.3877551,0.585034,0.53061223,0.43537414,,0.49659863,0.49659863,0.5578231,0.45578232,0.5442177,0.5442177,0.6938776,0.72789115,0.70068026,0.65986395,0.6394558,0.88435376,0.82312924,0.8639456,1.0
+ExtraTrees (default),0.0,0.006802721,0.006802721,0.006802721,0.0,0.006802721,0.0,0.0,0.027210884,0.020408163,0.0,0.0,0.088435374,0.0,0.07482993,0.14965986,0.054421768,0.020408163,0.013605442,0.07482993,0.034013607,0.07482993,0.088435374,0.054421768,0.006802721,0.054421768,0.18367347,0.006802721,0.19727892,0.0,0.36734694,0.08163265,0.07482993,0.21088435,0.19727892,0.04761905,0.054421768,0.11564626,0.11564626,0.23809524,0.2244898,0.21768707,0.1632653,0.21088435,0.26530612,0.13605443,0.24489796,0.24489796,0.25170067,0.37414965,0.31292516,0.2993197,0.46938777,0.3605442,0.5578231,0.72789115,0.5578231,0.50340134,,0.53061223,0.53061223,0.59183675,0.6122449,0.61904764,0.6394558,0.72789115,0.6938776,0.7755102,0.70068026,0.8367347,0.7755102,0.8503401,1.0
+Linear (tuned + ensembled),0.006802721,0.006802721,0.006802721,0.027210884,0.020408163,0.0,0.013605442,0.006802721,0.013605442,0.013605442,0.013605442,0.14285715,0.040816326,0.13605443,0.054421768,0.027210884,0.040816326,0.1904762,0.088435374,0.04761905,0.11564626,0.054421768,0.054421768,0.11564626,0.18367347,0.108843535,0.06122449,0.040816326,0.088435374,0.23809524,0.19727892,0.4217687,0.23129252,0.1292517,0.06122449,0.292517,0.1292517,0.1632653,0.292517,0.34013605,0.108843535,0.1904762,0.23129252,0.3265306,0.26530612,0.27891156,0.21768707,0.25170067,0.21768707,0.40136054,0.3197279,0.33333334,0.34013605,0.33333334,0.5102041,0.414966,0.45578232,0.50340134,0.46938777,,0.74489796,0.45578232,0.45578232,0.4829932,0.47619048,0.48979592,0.5170068,0.45578232,0.53061223,0.75510204,0.6666667,0.9455782,0.9319728
+Linear (tuned),0.006802721,0.0,0.006802721,0.020408163,0.013605442,0.006802721,0.013605442,0.013605442,0.020408163,0.006802721,0.006802721,0.1292517,0.020408163,0.14965986,0.040816326,0.027210884,0.034013607,0.19727892,0.040816326,0.034013607,0.06802721,0.040816326,0.040816326,0.07482993,0.11564626,0.07482993,0.06122449,0.034013607,0.06802721,0.23809524,0.19727892,0.3537415,0.17006803,0.08163265,0.027210884,0.2857143,0.12244898,0.13605443,0.31292516,0.30612245,0.07482993,0.14285715,0.1904762,0.33333334,0.21088435,0.23809524,0.14285715,0.20408164,0.15646258,0.3537415,0.2993197,0.292517,0.33333334,0.2857143,0.50340134,0.40816328,0.43537414,0.44217688,0.46938777,0.25510204,,0.4489796,0.40136054,0.414966,0.4489796,0.47619048,0.50340134,0.43537414,0.52380955,0.7414966,0.6530612,0.85714287,0.9319728
+iLTM (default),0.06122449,0.0,0.0,0.0,0.0,0.013605442,0.006802721,0.0,0.06122449,0.006802721,0.027210884,0.06122449,0.0,0.06122449,0.020408163,0.0,0.0,0.06802721,0.020408163,0.0,0.027210884,0.006802721,0.006802721,0.0,0.006802721,0.0,0.006802721,0.10204082,0.013605442,0.088435374,0.27891156,0.08163265,0.020408163,0.088435374,0.054421768,0.15646258,0.14285715,0.0,0.14965986,0.1632653,0.0952381,0.21088435,0.1632653,0.20408164,0.14285715,0.040816326,0.13605443,0.21088435,0.17006803,0.17687075,0.17687075,0.17006803,0.414966,0.47619048,0.5170068,0.3265306,0.4829932,0.5442177,0.40816328,0.5442177,0.5510204,,0.68707484,0.6802721,0.6530612,0.6802721,0.6802721,0.65986395,0.6938776,0.75510204,0.7755102,0.76870745,0.9319728
+TabSTAR (tuned),0.06122449,0.013605442,0.006802721,0.006802721,0.006802721,0.013605442,0.0,0.027210884,0.06802721,0.027210884,0.020408163,0.0952381,0.06122449,0.12244898,0.06122449,0.07482993,0.020408163,0.0952381,0.088435374,0.034013607,0.07482993,0.04761905,0.040816326,0.054421768,0.13605443,0.04761905,0.04761905,0.034013607,0.07482993,0.088435374,0.26530612,0.23809524,0.15646258,0.10204082,0.06802721,0.07482993,0.07482993,0.08163265,0.088435374,0.1292517,0.08163265,0.23129252,0.21768707,0.12244898,0.13605443,0.12244898,0.0952381,0.26530612,0.1632653,0.19727892,0.21768707,0.21088435,0.1904762,0.24489796,0.5170068,0.2993197,0.31292516,0.45578232,0.3877551,0.5442177,0.5986394,0.31292516,,0.53741497,0.5102041,0.5510204,0.53061223,0.585034,0.42857143,0.8503401,1.0,0.9047619,0.9863946
+TabSTAR (tuned + ensembled),0.06802721,0.013605442,0.006802721,0.006802721,0.006802721,0.020408163,0.020408163,0.034013607,0.06802721,0.040816326,0.034013607,0.0952381,0.04761905,0.12244898,0.06122449,0.04761905,0.027210884,0.0952381,0.0952381,0.040816326,0.06802721,0.054421768,0.04761905,0.054421768,0.14285715,0.040816326,0.04761905,0.034013607,0.06802721,0.088435374,0.27891156,0.25170067,0.15646258,0.0952381,0.06122449,0.07482993,0.06802721,0.088435374,0.0952381,0.14285715,0.07482993,0.23129252,0.21768707,0.1292517,0.14285715,0.13605443,0.088435374,0.25170067,0.1632653,0.2244898,0.21088435,0.20408164,0.19727892,0.23809524,0.50340134,0.27891156,0.33333334,0.45578232,0.3809524,0.5170068,0.585034,0.3197279,0.46258503,,0.49659863,0.53741497,0.53061223,0.5714286,0.4217687,0.8503401,1.0,0.88435376,0.9863946
+PerpetualBooster (tuned + ensembled),0.006802721,0.013605442,0.013605442,0.006802721,0.034013607,0.013605442,0.027210884,0.020408163,0.027210884,0.013605442,0.020408163,0.040816326,0.04761905,0.06122449,0.06122449,0.040816326,0.027210884,0.06802721,0.020408163,0.027210884,0.034013607,0.040816326,0.040816326,0.06122449,0.108843535,0.06122449,0.06802721,0.088435374,0.06122449,0.07482993,0.24489796,0.14965986,0.108843535,0.1632653,0.08163265,0.088435374,0.108843535,0.08163265,0.088435374,0.06802721,0.10204082,0.3197279,0.12244898,0.108843535,0.21768707,0.108843535,0.12244898,0.34013605,0.17006803,0.1292517,0.2857143,0.2857143,0.2585034,0.37414965,0.31292516,0.37414965,0.34013605,0.30612245,0.3605442,0.52380955,0.5510204,0.3469388,0.48979592,0.50340134,,0.7755102,0.8027211,0.585034,0.63265306,0.6258503,0.5510204,0.65986395,0.93877554
+PerpetualBooster (tuned),0.027210884,0.027210884,0.027210884,0.027210884,0.04761905,0.027210884,0.027210884,0.027210884,0.034013607,0.027210884,0.027210884,0.040816326,0.034013607,0.06122449,0.054421768,0.04761905,0.027210884,0.054421768,0.027210884,0.027210884,0.034013607,0.040816326,0.04761905,0.04761905,0.088435374,0.04761905,0.040816326,0.108843535,0.06802721,0.04761905,0.25170067,0.1292517,0.088435374,0.12244898,0.06802721,0.06802721,0.08163265,0.054421768,0.088435374,0.054421768,0.088435374,0.27891156,0.0952381,0.08163265,0.17006803,0.07482993,0.10204082,0.2993197,0.108843535,0.088435374,0.2244898,0.20408164,0.2244898,0.30612245,0.27210885,0.3197279,0.24489796,0.27210885,0.27210885,0.5102041,0.52380955,0.3197279,0.4489796,0.46258503,0.2244898,,0.6496599,0.5578231,0.5170068,0.61904764,0.50340134,0.6394558,0.9319728
+PerpetualBooster (default),0.034013607,0.034013607,0.027210884,0.034013607,0.034013607,0.027210884,0.034013607,0.034013607,0.034013607,0.040816326,0.034013607,0.04761905,0.04761905,0.054421768,0.04761905,0.054421768,0.034013607,0.04761905,0.04761905,0.034013607,0.054421768,0.034013607,0.040816326,0.06122449,0.08163265,0.06122449,0.06122449,0.108843535,0.07482993,0.06122449,0.25170067,0.11564626,0.0952381,0.1292517,0.06802721,0.06802721,0.0952381,0.08163265,0.08163265,0.054421768,0.0952381,0.27891156,0.10204082,0.088435374,0.18367347,0.08163265,0.1292517,0.292517,0.15646258,0.12244898,0.23809524,0.2244898,0.2244898,0.33333334,0.2857143,0.34013605,0.26530612,0.2993197,0.30612245,0.4829932,0.49659863,0.3197279,0.46938777,0.46938777,0.19727892,0.35034013,,0.53741497,0.53741497,0.59183675,0.53741497,0.6122449,0.7414966
+FastaiMLP (default),0.0,0.0,0.006802721,0.006802721,0.0,0.006802721,0.006802721,0.0,0.013605442,0.0,0.0,0.06122449,0.020408163,0.06122449,0.020408163,0.027210884,0.020408163,0.06122449,0.054421768,0.027210884,0.06122449,0.013605442,0.013605442,0.0,0.006802721,0.0,0.006802721,0.034013607,0.020408163,0.10204082,0.21088435,0.0952381,0.06122449,0.040816326,0.020408163,0.11564626,0.07482993,0.034013607,0.11564626,0.1292517,0.034013607,0.006802721,0.14285715,0.1292517,0.07482993,0.054421768,0.034013607,0.013605442,0.06802721,0.11564626,0.054421768,0.034013607,0.27210885,0.14965986,0.36734694,0.26530612,0.26530612,0.34013605,0.2244898,0.5442177,0.56462586,0.34013605,0.414966,0.42857143,0.414966,0.44217688,0.46258503,,0.3469388,0.7823129,0.7619048,0.8979592,0.99319726
+TabFlex (default),0.034013607,0.0,0.0,0.0,0.0,0.006802721,0.0,0.0,0.054421768,0.0,0.006802721,0.0,0.0,0.0,0.020408163,0.0,0.0,0.0,0.006802721,0.0,0.013605442,0.006802721,0.006802721,0.0,0.0,0.0,0.0,0.006802721,0.0,0.020408163,0.25170067,0.0,0.027210884,0.006802721,0.006802721,0.054421768,0.006802721,0.0,0.034013607,0.04761905,0.006802721,0.08163265,0.07482993,0.07482993,0.034013607,0.013605442,0.020408163,0.17006803,0.07482993,0.11564626,0.06802721,0.034013607,0.25170067,0.19727892,0.414966,0.3537415,0.25170067,0.3605442,0.2993197,0.46938777,0.47619048,0.30612245,0.5714286,0.5782313,0.36734694,0.4829932,0.46258503,0.6530612,,0.79591835,0.76870745,0.81632656,1.0
+KNN (tuned),0.06802721,0.034013607,0.04761905,0.020408163,0.040816326,0.054421768,0.054421768,0.04761905,0.06122449,0.027210884,0.04761905,0.027210884,0.034013607,0.054421768,0.04761905,0.034013607,0.04761905,0.040816326,0.06802721,0.04761905,0.027210884,0.06802721,0.06802721,0.04761905,0.10204082,0.04761905,0.027210884,0.013605442,0.040816326,0.0,0.23809524,0.12244898,0.11564626,0.06122449,0.06122449,0.006802721,0.027210884,0.054421768,0.0,0.04761905,0.06122449,0.108843535,0.10204082,0.027210884,0.06122449,0.07482993,0.027210884,0.108843535,0.07482993,0.07482993,0.14285715,0.14285715,0.08163265,0.07482993,0.06802721,0.15646258,0.088435374,0.11564626,0.1632653,0.24489796,0.2585034,0.24489796,0.14965986,0.14965986,0.37414965,0.3809524,0.40816328,0.21768707,0.20408164,,0.37414965,0.37414965,0.92517006
+TabSTAR (default),0.006802721,0.013605442,0.006802721,0.006802721,0.006802721,0.0,0.0,0.006802721,0.006802721,0.013605442,0.0,0.06802721,0.006802721,0.06122449,0.034013607,0.013605442,0.006802721,0.07482993,0.034013607,0.006802721,0.034013607,0.013605442,0.013605442,0.006802721,0.027210884,0.006802721,0.013605442,0.013605442,0.020408163,0.06802721,0.054421768,0.108843535,0.054421768,0.027210884,0.013605442,0.06122449,0.020408163,0.006802721,0.06802721,0.06802721,0.020408163,0.04761905,0.13605443,0.08163265,0.04761905,0.013605442,0.027210884,0.06122449,0.06122449,0.06122449,0.088435374,0.06122449,0.020408163,0.108843535,0.3197279,0.1904762,0.108843535,0.17687075,0.2244898,0.33333334,0.3469388,0.2244898,0.0,0.0,0.4489796,0.49659863,0.46258503,0.23809524,0.23129252,0.6258503,,0.6802721,0.9591837
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.04761905,0.0,0.034013607,0.006802721,0.0,0.0,0.06122449,0.006802721,0.0,0.006802721,0.0,0.0,0.0,0.006802721,0.0,0.0,0.0,0.0,0.054421768,0.1904762,0.06802721,0.027210884,0.006802721,0.006802721,0.08163265,0.006802721,0.0,0.07482993,0.034013607,0.006802721,0.0,0.020408163,0.10204082,0.013605442,0.0,0.0,0.0,0.006802721,0.020408163,0.027210884,0.020408163,0.06802721,0.06122449,0.25170067,0.0952381,0.108843535,0.13605443,0.14965986,0.054421768,0.14285715,0.23129252,0.0952381,0.11564626,0.34013605,0.3605442,0.3877551,0.10204082,0.18367347,0.6258503,0.3197279,,0.91156465
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.006802721,0.0,0.0,0.0,0.0,0.0,0.006802721,0.0,0.0,0.0,0.0,0.0,0.0,0.006802721,0.0,0.0,0.0,0.0,0.06802721,0.06802721,0.06802721,0.013605442,0.013605442,0.06122449,0.06802721,0.2585034,0.006802721,0.0,0.07482993,0.040816326,0.088435374,
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_all/leaderboard_overview_explorer.html
index c42d8f4c7c6d8be036831fdcd6f15543cce09389..b849692acc7cb731581fcec25015af1aa89ade56 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Regression | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_all/leaderboard_table.html b/data/imputation_no/splits_all/tasks_regression/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..17450a48a4d7aeb6a81595e4e411ff2f42b3e972
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_regression/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2031.0,"score":0.915,"rank":3.79,"hrank":1.95,"imp":1.549,"train_time":39.42,"predict_time":6.476,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+151\/-117","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1812.0,"score":0.736,"rank":9.02,"hrank":3.42,"imp":2.318,"train_time":3.66,"predict_time":0.399,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+229\/-132","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1792.0,"score":0.69,"rank":9.68,"hrank":4.32,"imp":3.842,"train_time":335.03,"predict_time":4.334,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-91","imputed":false},{"position":3,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1768.0,"score":0.676,"rank":10.57,"hrank":3.54,"imp":3.341,"train_time":11.58,"predict_time":2.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-76","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1761.0,"score":0.632,"rank":10.84,"hrank":7.8,"imp":3.932,"train_time":8.52,"predict_time":0.696,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+109\/-59","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1759.0,"score":0.623,"rank":10.9,"hrank":6.29,"imp":3.296,"train_time":1709.05,"predict_time":8.119,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+147\/-106","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1740.0,"score":0.622,"rank":11.67,"hrank":4.16,"imp":4.161,"train_time":4786.6,"predict_time":239.304,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-92","imputed":false},{"position":7,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1693.0,"score":0.657,"rank":13.62,"hrank":4.48,"imp":3.867,"train_time":1.32,"predict_time":0.122,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+267\/-150","imputed":false},{"position":8,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1687.0,"score":0.59,"rank":13.89,"hrank":7.47,"imp":4.438,"train_time":6.93,"predict_time":1.389,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-88","imputed":false},{"position":9,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1686.0,"score":0.57,"rank":13.96,"hrank":7.66,"imp":4.603,"train_time":4786.6,"predict_time":38.497,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-73","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1665.0,"score":0.5,"rank":14.91,"hrank":10.32,"imp":4.965,"train_time":3995.01,"predict_time":10.048,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-65","imputed":false},{"position":11,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1646.0,"score":0.516,"rank":15.81,"hrank":8.48,"imp":4.041,"train_time":1709.05,"predict_time":0.811,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-114","imputed":false},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1620.0,"score":0.491,"rank":17.14,"hrank":8.9,"imp":5.12,"train_time":2.08,"predict_time":0.182,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+191\/-101","imputed":false},{"position":13,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1614.0,"score":0.411,"rank":17.43,"hrank":11.74,"imp":6.361,"train_time":1866.35,"predict_time":6.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-88","imputed":false},{"position":14,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1594.0,"score":0.467,"rank":18.49,"hrank":8.98,"imp":5.52,"train_time":46.62,"predict_time":39.212,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-76","imputed":false},{"position":15,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1558.0,"score":0.364,"rank":20.36,"hrank":13.84,"imp":5.929,"train_time":3995.01,"predict_time":0.844,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+109\/-79","imputed":false},{"position":16,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1552.0,"score":0.376,"rank":20.7,"hrank":8.42,"imp":7.226,"train_time":3779.7,"predict_time":7.69,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-121","imputed":false},{"position":17,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1549.0,"score":0.421,"rank":20.89,"hrank":13.95,"imp":5.516,"train_time":7.04,"predict_time":0.508,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+148\/-111","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1498.0,"score":0.239,"rank":23.72,"hrank":16.02,"imp":7.903,"train_time":2229.21,"predict_time":0.38,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+101\/-66","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1490.0,"score":0.239,"rank":24.19,"hrank":18.96,"imp":8.36,"train_time":700.19,"predict_time":9.322,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-80","imputed":false},{"position":20,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1474.0,"score":0.272,"rank":25.18,"hrank":17.4,"imp":7.409,"train_time":714.5,"predict_time":1.376,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-96","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1473.0,"score":0.219,"rank":25.19,"hrank":17.1,"imp":8.04,"train_time":2229.21,"predict_time":0.055,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-74","imputed":false},{"position":22,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1458.0,"score":0.278,"rank":26.15,"hrank":20.24,"imp":6.811,"train_time":3013.86,"predict_time":2.092,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-86","imputed":false},{"position":23,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1445.0,"score":0.212,"rank":26.94,"hrank":20.85,"imp":9.258,"train_time":12685.08,"predict_time":321.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-49","imputed":false},{"position":24,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1424.0,"score":0.172,"rank":28.25,"hrank":22.68,"imp":8.938,"train_time":700.19,"predict_time":0.968,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-80","imputed":false},{"position":25,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1412.0,"score":0.152,"rank":29.02,"hrank":26.23,"imp":8.88,"train_time":834.93,"predict_time":2.614,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-46","imputed":false},{"position":26,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1405.0,"score":0.271,"rank":29.46,"hrank":19.24,"imp":7.178,"train_time":0.91,"predict_time":0.075,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+159\/-107","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1391.0,"score":0.131,"rank":30.41,"hrank":27.07,"imp":9.026,"train_time":834.93,"predict_time":0.388,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-54","imputed":false},{"position":28,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1386.0,"score":0.147,"rank":30.73,"hrank":22.84,"imp":9.53,"train_time":8.73,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-99","imputed":false},{"position":29,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1385.0,"score":0.188,"rank":30.76,"hrank":26.25,"imp":8.072,"train_time":714.5,"predict_time":0.099,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-87","imputed":false},{"position":30,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1382.0,"score":0.194,"rank":31.02,"hrank":20.94,"imp":9.184,"train_time":3779.7,"predict_time":0.395,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+113\/-89","imputed":false},{"position":31,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1381.0,"score":0.191,"rank":31.05,"hrank":23.4,"imp":7.722,"train_time":3013.86,"predict_time":0.181,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-91","imputed":false},{"position":32,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1332.0,"score":0.107,"rank":34.21,"hrank":29.85,"imp":9.214,"train_time":12685.08,"predict_time":59.099,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-56","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1330.0,"score":0.114,"rank":34.31,"hrank":25.74,"imp":11.229,"train_time":420.1,"predict_time":0.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-121","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1293.0,"score":0.09,"rank":36.7,"hrank":29.01,"imp":11.514,"train_time":420.1,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-116","imputed":false},{"position":35,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1278.0,"score":0.108,"rank":37.66,"hrank":30.81,"imp":9.347,"train_time":9.19,"predict_time":0.132,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-109","imputed":false},{"position":36,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1276.0,"score":0.085,"rank":37.8,"hrank":30.54,"imp":10.694,"train_time":15.5,"predict_time":0.298,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-73","imputed":false},{"position":37,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1275.0,"score":0.161,"rank":37.81,"hrank":20.4,"imp":10.758,"train_time":20.24,"predict_time":6.621,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+148\/-130","imputed":false},{"position":38,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1261.0,"score":0.142,"rank":38.75,"hrank":18.05,"imp":10.298,"train_time":74.68,"predict_time":19.758,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+167\/-155","imputed":false},{"position":39,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1237.0,"score":0.094,"rank":40.27,"hrank":29.75,"imp":10.885,"train_time":4608.59,"predict_time":1.233,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-101","imputed":false},{"position":40,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1224.0,"score":0.077,"rank":41.13,"hrank":35.31,"imp":10.435,"train_time":8.9,"predict_time":1.644,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-85","imputed":false},{"position":41,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.097,"rank":41.18,"hrank":32.46,"imp":12.872,"train_time":263.04,"predict_time":0.808,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-98","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1221.0,"score":0.045,"rank":41.32,"hrank":36.88,"imp":12.492,"train_time":2.68,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-129","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1200.0,"score":0.083,"rank":42.68,"hrank":31.66,"imp":13.097,"train_time":263.04,"predict_time":0.066,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-98","imputed":false},{"position":44,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1198.0,"score":0.005,"rank":42.82,"hrank":41.23,"imp":11.319,"train_time":2.11,"predict_time":0.275,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+37\/-38","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1183.0,"score":0.059,"rank":43.74,"hrank":37.21,"imp":11.663,"train_time":4608.59,"predict_time":0.097,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-95","imputed":false},{"position":46,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1181.0,"score":0.026,"rank":43.87,"hrank":38.97,"imp":11.973,"train_time":2.24,"predict_time":0.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-82","imputed":false},{"position":47,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1173.0,"score":0.017,"rank":44.39,"hrank":41.55,"imp":13.153,"train_time":162.38,"predict_time":0.363,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-90","imputed":false},{"position":48,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1154.0,"score":0.059,"rank":45.6,"hrank":36.03,"imp":14.422,"train_time":1865.27,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-162","imputed":false},{"position":49,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1147.0,"score":0.035,"rank":46.03,"hrank":39.41,"imp":13.678,"train_time":2.45,"predict_time":0.742,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-114","imputed":false},{"position":50,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1146.0,"score":0.026,"rank":46.04,"hrank":43.08,"imp":13.906,"train_time":515.75,"predict_time":0.771,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-88","imputed":false},{"position":51,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1117.0,"score":0.0,"rank":47.81,"hrank":45.51,"imp":14.217,"train_time":162.38,"predict_time":0.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-97","imputed":false},{"position":52,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1116.0,"score":0.011,"rank":47.9,"hrank":44.98,"imp":14.386,"train_time":515.75,"predict_time":0.124,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-98","imputed":false},{"position":53,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1110.0,"score":0.039,"rank":48.21,"hrank":38.44,"imp":14.95,"train_time":1865.27,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+125\/-163","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1059.0,"score":0.015,"rank":51.08,"hrank":44.18,"imp":15.532,"train_time":2.54,"predict_time":0.24,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-124","imputed":false},{"position":55,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1049.0,"score":0.021,"rank":51.6,"hrank":44.84,"imp":15.867,"train_time":7.87,"predict_time":0.011,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-168","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1033.0,"score":0.021,"rank":52.49,"hrank":49.26,"imp":15.273,"train_time":540.14,"predict_time":2.672,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-107","imputed":false},{"position":57,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1029.0,"score":0.0,"rank":52.68,"hrank":49.95,"imp":15.03,"train_time":20.49,"predict_time":0.08,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-114","imputed":false},{"position":58,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":54.17,"hrank":52.49,"imp":15.878,"train_time":0.53,"predict_time":0.062,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-79","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":988.0,"score":0.011,"rank":54.79,"hrank":52.33,"imp":15.751,"train_time":540.14,"predict_time":0.325,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-109","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":971.0,"score":0.112,"rank":55.6,"hrank":21.58,"imp":23.445,"train_time":28729.74,"predict_time":19.901,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+221\/-303","imputed":false},{"position":61,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":957.0,"score":0.107,"rank":56.27,"hrank":23.8,"imp":23.609,"train_time":28729.74,"predict_time":5.236,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+222\/-313","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":946.0,"score":0.0,"rank":56.77,"hrank":55.38,"imp":17.681,"train_time":27.32,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-115","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":899.0,"score":0.0,"rank":58.85,"hrank":52.07,"imp":20.905,"train_time":92.55,"predict_time":0.897,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-154","imputed":false},{"position":64,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":873.0,"score":0.0,"rank":59.88,"hrank":59.11,"imp":18.889,"train_time":357.02,"predict_time":81.613,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-110","imputed":false},{"position":65,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":869.0,"score":0.0,"rank":60.04,"hrank":58.66,"imp":19.921,"train_time":2.6,"predict_time":0.392,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-156","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":849.0,"score":0.052,"rank":60.79,"hrank":32.59,"imp":26.539,"train_time":323.52,"predict_time":4.929,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+237\/-346","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":796.0,"score":0.0,"rank":62.63,"hrank":58.81,"imp":23.278,"train_time":92.55,"predict_time":0.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+145\/-167","imputed":false},{"position":68,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":692.0,"score":0.0,"rank":65.51,"hrank":54.25,"imp":30.333,"train_time":0.19,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+173\/-229","imputed":false},{"position":69,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":512.0,"score":0.0,"rank":68.78,"hrank":68.52,"imp":37.415,"train_time":193.98,"predict_time":0.173,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-371","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":478.0,"score":0.0,"rank":69.21,"hrank":68.93,"imp":37.586,"train_time":193.98,"predict_time":0.074,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+144\/-447","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":308.0,"score":0.0,"rank":70.89,"hrank":70.85,"imp":39.978,"train_time":0.95,"predict_time":0.097,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+149\/-405","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_all/pareto_front_explorer.html
index d48cd176664c234ecc0424a2154779b235641b6d..0f0519170c4d6e4aeda1cdccf1cc1d92dc5b7288 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.016935661,"imp":9.5295102763,"elo":1385.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0551291653,"imp":8.0404336056,"elo":1473.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3802116897,"imp":7.9028857742,"elo":1498.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0297627373,"imp":12.4919267039,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0341858448,"imp":11.5140991162,"elo":1292.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"imp":11.2291675003,"elo":1330.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0108193564,"imp":15.8668096844,"elo":1049.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0116341148,"imp":14.9499623717,"elo":1110.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1165303532,"imp":14.4223206891,"elo":1153.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.24026221,"imp":15.5322011749,"elo":1059.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0658904991,"imp":13.0968088515,"elo":1199.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8077239936,"imp":12.8723758679,"elo":1222.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3915661558,"imp":19.9211719655,"elo":869.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.324857112,"imp":15.7507952826,"elo":987.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.6723835679,"imp":15.2727669644,"elo":1032.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2745146333,"imp":11.3185838933,"elo":1197.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9682498776,"imp":8.9378657406,"elo":1424.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":9.3218182128,"imp":8.3595749003,"elo":1490.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":19.7582527317,"imp":10.2976447141,"elo":1260.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2983006451,"imp":10.6939593664,"elo":1275.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3953581979,"imp":9.1839542219,"elo":1381.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.6904227381,"imp":7.226029799,"elo":1552.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.3890423048,"imp":4.4376918389,"elo":1687.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0226031244,"imp":3.3408922469,"elo":1768.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.020234932,"imp":17.6806184935,"elo":946.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1695058591,"imp":14.2167047211,"elo":1117.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3627650406,"imp":13.1525752298,"elo":1172.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0621551845,"imp":15.8784654615,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1235618305,"imp":14.3858342755,"elo":1115.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7709478321,"imp":13.9058764173,"elo":1146.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.644117959,"imp":10.4349964792,"elo":1223.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.8438670931,"imp":5.9294999968,"elo":1558.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":10.0484057771,"imp":4.9646421895,"elo":1665.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"imp":5.5155833924,"elo":1548.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8108737399,"imp":4.0406089457,"elo":1646.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1192270005,"imp":3.2955156064,"elo":1759.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.620625058,"imp":10.758214967,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"imp":5.5201041457,"elo":1593.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.4972730809,"imp":4.6028367382,"elo":1685.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":239.3040135069,"imp":4.1606183521,"elo":1739.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"imp":5.1196075543,"elo":1619.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4760486585,"imp":1.5488984258,"elo":2030.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1222431699,"imp":3.8666164924,"elo":1693.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1324525523,"imp":9.3474029081,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1814320495,"imp":7.7223069039,"elo":1380.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.0923554564,"imp":6.8114636962,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.696318366,"imp":3.9315627109,"elo":1760.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"imp":2.3177658571,"elo":1811.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9291777504,"imp":26.5387622681,"elo":849.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.23600059,"imp":23.6094832182,"elo":957.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.9007182558,"imp":23.4447365419,"elo":971.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0752279828,"imp":7.1778783948,"elo":1405.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0796563802,"imp":15.029521099,"elo":1028.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0968974085,"imp":11.6630279281,"elo":1183.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.232537078,"imp":10.8845502145,"elo":1236.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.242470131,"imp":11.9728342353,"elo":1181.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3882893337,"imp":9.0258712914,"elo":1390.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.6142653492,"imp":8.8803712528,"elo":1412.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":81.6126739219,"imp":18.8885402603,"elo":873.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":59.0992991738,"imp":9.214095793,"elo":1332.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":321.7409586602,"imp":9.258272998,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7424024527,"imp":13.6777604313,"elo":1146.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0990665936,"imp":8.0718064981,"elo":1385.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.3755494169,"imp":7.4092090452,"elo":1473.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.016935661,"x_train":8.7329235239,"imp":9.5295102763,"elo":1385.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0551291653,"x_train":2229.2144624726,"imp":8.0404336056,"elo":1473.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3802116897,"x_train":2229.2144624726,"imp":7.9028857742,"elo":1498.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0297627373,"x_train":2.6789664191,"imp":12.4919267039,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0341858448,"x_train":420.0975247537,"imp":11.5140991162,"elo":1292.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"x_train":420.0975247537,"imp":11.2291675003,"elo":1330.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0108193564,"x_train":7.8692134542,"imp":15.8668096844,"elo":1049.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0116341148,"x_train":1865.2663509929,"imp":14.9499623717,"elo":1110.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1165303532,"x_train":1865.2663509929,"imp":14.4223206891,"elo":1153.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.24026221,"x_train":2.5445212109,"imp":15.5322011749,"elo":1059.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0658904991,"x_train":263.0379305179,"imp":13.0968088515,"elo":1199.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8077239936,"x_train":263.0379305179,"imp":12.8723758679,"elo":1222.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3915661558,"x_train":2.6039043157,"imp":19.9211719655,"elo":869.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.324857112,"x_train":540.1381360311,"imp":15.7507952826,"elo":987.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.6723835679,"x_train":540.1381360311,"imp":15.2727669644,"elo":1032.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2745146333,"x_train":2.1108344816,"imp":11.3185838933,"elo":1197.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9682498776,"x_train":700.1867028829,"imp":8.9378657406,"elo":1424.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":9.3218182128,"x_train":700.1867028829,"imp":8.3595749003,"elo":1490.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":19.7582527317,"x_train":74.6790742113,"imp":10.2976447141,"elo":1260.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2983006451,"x_train":15.5008512826,"imp":10.6939593664,"elo":1275.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3953581979,"x_train":3779.7025911156,"imp":9.1839542219,"elo":1381.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.6904227381,"x_train":3779.7025911156,"imp":7.226029799,"elo":1552.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.3890423048,"x_train":6.9283541331,"imp":4.4376918389,"elo":1687.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0226031244,"x_train":11.5842534555,"imp":3.3408922469,"elo":1768.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.020234932,"x_train":27.322832323,"imp":17.6806184935,"elo":946.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1695058591,"x_train":162.3794942955,"imp":14.2167047211,"elo":1117.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3627650406,"x_train":162.3794942955,"imp":13.1525752298,"elo":1172.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0621551845,"x_train":0.5271703851,"imp":15.8784654615,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1235618305,"x_train":515.7544728433,"imp":14.3858342755,"elo":1115.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7709478321,"x_train":515.7544728433,"imp":13.9058764173,"elo":1146.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.644117959,"x_train":8.9020711389,"imp":10.4349964792,"elo":1223.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.8438670931,"x_train":3995.0078687545,"imp":5.9294999968,"elo":1558.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":10.0484057771,"x_train":3995.0078687545,"imp":4.9646421895,"elo":1665.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"x_train":7.0435968115,"imp":5.5155833924,"elo":1548.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8108737399,"x_train":1709.0533523899,"imp":4.0406089457,"elo":1646.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1192270005,"x_train":1709.0533523899,"imp":3.2955156064,"elo":1759.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.620625058,"x_train":20.2378618082,"imp":10.758214967,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"x_train":46.6209023207,"imp":5.5201041457,"elo":1593.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.4972730809,"x_train":4786.5989331939,"imp":4.6028367382,"elo":1685.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":239.3040135069,"x_train":4786.5989331939,"imp":4.1606183521,"elo":1739.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"x_train":2.0755087942,"imp":5.1196075543,"elo":1619.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4760486585,"x_train":39.4205185389,"imp":1.5488984258,"elo":2030.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1222431699,"x_train":1.3204095155,"imp":3.8666164924,"elo":1693.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1324525523,"x_train":9.1928711765,"imp":9.3474029081,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1814320495,"x_train":3013.8595115999,"imp":7.7223069039,"elo":1380.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.0923554564,"x_train":3013.8595115999,"imp":6.8114636962,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.696318366,"x_train":8.5238425595,"imp":3.9315627109,"elo":1760.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"x_train":3.65846503,"imp":2.3177658571,"elo":1811.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9291777504,"x_train":323.5247663683,"imp":26.5387622681,"elo":849.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.23600059,"x_train":28729.7373655693,"imp":23.6094832182,"elo":957.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.9007182558,"x_train":28729.7373655693,"imp":23.4447365419,"elo":971.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0752279828,"x_train":0.9081877562,"imp":7.1778783948,"elo":1405.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0796563802,"x_train":20.4886632638,"imp":15.029521099,"elo":1028.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0968974085,"x_train":4608.594420269,"imp":11.6630279281,"elo":1183.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.232537078,"x_train":4608.594420269,"imp":10.8845502145,"elo":1236.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.242470131,"x_train":2.2442581404,"imp":11.9728342353,"elo":1181.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3882893337,"x_train":834.9300717203,"imp":9.0258712914,"elo":1390.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.6142653492,"x_train":834.9300717203,"imp":8.8803712528,"elo":1412.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":81.6126739219,"x_train":357.0213835173,"imp":18.8885402603,"elo":873.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":59.0992991738,"x_train":12685.0815501834,"imp":9.214095793,"elo":1332.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":321.7409586602,"x_train":12685.0815501834,"imp":9.258272998,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7424024527,"x_train":2.4487016432,"imp":13.6777604313,"elo":1146.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0990665936,"x_train":714.4999525058,"imp":8.0718064981,"elo":1385.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.3755494169,"x_train":714.4999525058,"imp":7.4092090452,"elo":1473.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_all/tuning_trajectories_explorer.html
index ecc688dfbe40aa0781644a1ebe31c1c887ae01a1..53b0e07ff6c04f1717fa102aa9449e23f03342ba 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":8.7329235239,"imp":9.5353759701,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":21.046557619,"imp":9.0461990243,"elo":1428.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":52.5864800365,"imp":8.5218421343,"elo":1473.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":107.3625813458,"imp":8.2749168681,"elo":1489.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":273.3208727607,"imp":8.0308507273,"elo":1504.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":552.7941001427,"imp":7.9732575207,"elo":1510.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1107.2542254236,"imp":7.9343037544,"elo":1517.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1651.763231487,"imp":7.92496954,"elo":1518.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":2229.2144624726,"imp":7.9091524135,"elo":1520.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.6789664191,"imp":12.4975937692,"elo":1227.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.4383850203,"imp":12.3796259496,"elo":1234.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.0701848574,"imp":12.1936284788,"elo":1255.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":20.464525374,"imp":11.9697365215,"elo":1279.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":51.9588660046,"imp":11.7111835094,"elo":1303.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":104.2658709225,"imp":11.5279126712,"elo":1320.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":208.787997457,"imp":11.3573057852,"elo":1332.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":313.2556024507,"imp":11.2790879483,"elo":1339.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":420.0975247537,"imp":11.2346097142,"elo":1346.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.8692134542,"imp":15.8722220541,"elo":1054.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":17.0945668224,"imp":15.5040713652,"elo":1077.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":52.1405202142,"imp":15.1446322779,"elo":1108.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":97.8428503791,"imp":14.9308940147,"elo":1126.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":240.4577474664,"imp":14.6418568122,"elo":1143.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":484.5851613774,"imp":14.5553732057,"elo":1152.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":933.0688585006,"imp":14.484585249,"elo":1159.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1378.9985712807,"imp":14.4519456161,"elo":1164.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1865.2663509929,"imp":14.4279093563,"elo":1168.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.5445212109,"imp":15.5377117663,"elo":1062.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1259596548,"imp":14.977407387,"elo":1102.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.7721912257,"imp":14.1420248066,"elo":1155.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":13.288092342,"imp":13.7333704871,"elo":1179.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":32.8419339595,"imp":13.1177966015,"elo":1213.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":65.4938634865,"imp":12.9906805715,"elo":1226.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":130.8369715186,"imp":12.9146356805,"elo":1233.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":196.2244804971,"imp":12.8899221211,"elo":1236.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":263.0379305179,"imp":12.8786181764,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.6039043157,"imp":19.92609532,"elo":868.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.1690100082,"imp":19.661131622,"elo":880.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":13.4266251032,"imp":18.8130499001,"elo":915.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":27.320730914,"imp":18.3254561042,"elo":938.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":66.1669393228,"imp":17.3598274361,"elo":974.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":132.6885087308,"imp":16.7866650086,"elo":993.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":267.8675184086,"imp":16.1201830035,"elo":1014.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":402.5490610457,"imp":15.6460946342,"elo":1033.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":540.1381360311,"imp":15.2784360822,"elo":1049.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.1108344816,"imp":11.3245379459,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":5.9546664661,"imp":10.756584727,"elo":1249.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":16.8872918174,"imp":9.8018502175,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":32.7521125678,"imp":9.3117014963,"elo":1393.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":84.7959741874,"imp":8.7853597392,"elo":1461.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":173.2180148137,"imp":8.6267188825,"elo":1486.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":347.0816109705,"imp":8.4879776995,"elo":1505.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":519.953899606,"imp":8.4198835742,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":700.1867028829,"imp":8.3660364871,"elo":1521.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":74.6790742113,"imp":10.3040126165,"elo":1274.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":15.5008512826,"imp":10.6997726176,"elo":1288.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":40.5171687704,"imp":9.9733918546,"elo":1338.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":107.503248256,"imp":8.7985453215,"elo":1427.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":200.5362465049,"imp":8.2725576839,"elo":1476.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":460.9985605524,"imp":7.6898377498,"elo":1529.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":947.159574971,"imp":7.4990468521,"elo":1547.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1857.8046474259,"imp":7.3295844942,"elo":1563.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":2770.8108835678,"imp":7.2508796026,"elo":1570.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":3779.7025911156,"imp":7.2318390451,"elo":1574.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":6.9283541331,"imp":4.4444075824,"elo":1713.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":11.5842534555,"imp":3.3478648597,"elo":1794.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":27.322832323,"imp":17.6856154942,"elo":958.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":54.4506451141,"imp":15.2891793683,"elo":1061.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":135.4944823225,"imp":13.290731348,"elo":1162.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":162.3794942955,"imp":13.1578480182,"elo":1173.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.5271703851,"imp":15.8838229418,"elo":999.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.6581130708,"imp":15.2431015169,"elo":1043.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":13.0531670853,"imp":14.7584691425,"elo":1079.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":26.0784291852,"imp":14.323191656,"elo":1113.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":63.4089966233,"imp":14.0743614667,"elo":1134.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":127.7253484811,"imp":13.9909342199,"elo":1145.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":254.9677280133,"imp":13.9383084007,"elo":1154.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":383.936400338,"imp":13.9204336715,"elo":1158.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":515.7544728433,"imp":13.9114137358,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":8.9020711389,"imp":10.4415159463,"elo":1234.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.3173471418,"imp":8.6083504215,"elo":1346.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":77.6709970732,"imp":6.6932795026,"elo":1504.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":189.6529598329,"imp":6.0296075606,"elo":1569.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":487.3244648114,"imp":5.5393110683,"elo":1628.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":988.6461264381,"imp":5.2903376582,"elo":1654.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1985.9304497408,"imp":5.1260738579,"elo":1674.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2976.2042051285,"imp":5.0345688167,"elo":1689.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3995.0078687545,"imp":4.9715617017,"elo":1702.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":7.0435968115,"imp":5.5222967423,"elo":1564.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.015104703,"imp":5.1210599455,"elo":1587.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":41.0105415705,"imp":4.4217626195,"elo":1630.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":83.3999540139,"imp":4.038294855,"elo":1677.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":210.1679172678,"imp":3.6708426576,"elo":1733.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":423.4200904906,"imp":3.4955200333,"elo":1760.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":852.1950752527,"imp":3.3899731202,"elo":1777.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1277.5788085201,"imp":3.3378559202,"elo":1787.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1709.0533523899,"imp":3.3026131586,"elo":1794.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":20.2378618082,"imp":10.7641870973,"elo":1294.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.6209023207,"imp":5.5268547342,"elo":1605.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.0936255942,"imp":5.1264051337,"elo":1640.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":132.9785409439,"imp":4.7855178058,"elo":1678.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":240.939305723,"imp":4.5422915224,"elo":1707.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":589.3561296929,"imp":4.3320037401,"elo":1735.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1183.6501039466,"imp":4.2352461267,"elo":1749.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2376.2279314343,"imp":4.1821772214,"elo":1760.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3574.1919132683,"imp":4.1701024214,"elo":1762.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4786.5989331939,"imp":4.1669203949,"elo":1763.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0755087942,"imp":5.1261178466,"elo":1635.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":39.4205185389,"imp":1.5556478433,"elo":2064.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.3204095155,"imp":3.8733136542,"elo":1712.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.1928711765,"imp":9.3536327032,"elo":1285.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":22.5165548441,"imp":8.9452683683,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":69.1089159979,"imp":8.6316790067,"elo":1348.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":146.4525878247,"imp":8.1323835934,"elo":1384.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":371.1147116855,"imp":7.6711419131,"elo":1420.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":743.1660518453,"imp":7.3844894287,"elo":1445.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1486.16387375,"imp":7.1771170581,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2240.0758838289,"imp":6.8718997187,"elo":1474.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3013.8595115999,"imp":6.817269664,"elo":1483.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":8.5238425595,"imp":3.9382292956,"elo":1793.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.65846503,"imp":2.3315092965,"elo":1826.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":323.5247663683,"imp":26.5437066212,"elo":850.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":458.087256476,"imp":26.4508381706,"elo":867.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":775.9282960298,"imp":26.2473198846,"elo":882.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1467.4534767341,"imp":25.7801256282,"elo":908.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3523.6508469471,"imp":25.0521742687,"elo":932.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":7117.6518843327,"imp":24.3599867674,"elo":950.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":14227.2413776887,"imp":23.9279306886,"elo":965.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":21385.328840253,"imp":23.6532205307,"elo":977.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":28729.7373655693,"imp":23.4498749414,"elo":992.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.9081877562,"imp":7.1846434449,"elo":1420.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":20.4886632638,"imp":15.0354241324,"elo":1040.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":41.7266327108,"imp":14.4070459058,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":106.9107806776,"imp":13.4001634822,"elo":1117.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":227.2145545781,"imp":12.8730228051,"elo":1148.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":570.0093805222,"imp":12.1517616796,"elo":1186.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1173.9046077839,"imp":11.7009521655,"elo":1216.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2290.0065707371,"imp":11.2373560638,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3456.708959715,"imp":11.0269025431,"elo":1247.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4608.594420269,"imp":10.8904331716,"elo":1257.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.2442581404,"imp":11.9785913399,"elo":1185.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":8.0761527298,"imp":10.5410795078,"elo":1273.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":20.7267313474,"imp":9.84983897,"elo":1333.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":40.9776832255,"imp":9.4901130337,"elo":1367.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":103.9269515729,"imp":9.2122964888,"elo":1395.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":207.504207675,"imp":9.0836511972,"elo":1408.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":414.7237545011,"imp":8.9878774723,"elo":1420.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":622.7467847485,"imp":8.9230143654,"elo":1427.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":834.9300717203,"imp":8.8864513439,"elo":1434.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":357.0213835173,"imp":18.8936466855,"elo":886.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":826.6192885105,"imp":14.7756137651,"elo":1040.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2323.5909795411,"imp":15.7066228744,"elo":1208.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4948.4562156001,"imp":14.1706851906,"elo":1305.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12120.2560802151,"imp":9.2996535471,"elo":1458.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12685.0815501834,"imp":9.264955764,"elo":1465.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.4487016432,"imp":13.6835745524,"elo":1157.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.9610558826,"imp":12.3645381262,"elo":1221.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.0080432107,"imp":10.6344658201,"elo":1298.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":34.8874663118,"imp":9.3164491876,"elo":1369.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.7983697442,"imp":8.1975537975,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":177.9391500768,"imp":7.7973079327,"elo":1460.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":356.5372096884,"imp":7.5534817366,"elo":1478.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":533.621477288,"imp":7.4615924328,"elo":1489.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":714.4999525058,"imp":7.4156273541,"elo":1496.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.016935661,"x_train":8.7329235239,"imp":9.5353759701,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0753245982,"x_train":21.046557619,"imp":9.0461990243,"elo":1428.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1609118157,"x_train":52.5864800365,"imp":8.5218421343,"elo":1473.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2251450506,"x_train":107.3625813458,"imp":8.2749168681,"elo":1489.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2353196357,"x_train":273.3208727607,"imp":8.0308507273,"elo":1504.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2694149278,"x_train":552.7941001427,"imp":7.9732575207,"elo":1510.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3197063318,"x_train":1107.2542254236,"imp":7.9343037544,"elo":1517.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3499660795,"x_train":1651.763231487,"imp":7.92496954,"elo":1518.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3802116897,"x_train":2229.2144624726,"imp":7.9091524135,"elo":1520.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0297627373,"x_train":2.6789664191,"imp":12.4975937692,"elo":1227.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0556929991,"x_train":4.4383850203,"imp":12.3796259496,"elo":1234.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0988043477,"x_train":10.0701848574,"imp":12.1936284788,"elo":1255.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1363240631,"x_train":20.464525374,"imp":11.9697365215,"elo":1279.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1837219602,"x_train":51.9588660046,"imp":11.7111835094,"elo":1303.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2043019279,"x_train":104.2658709225,"imp":11.5279126712,"elo":1320.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2249081343,"x_train":208.787997457,"imp":11.3573057852,"elo":1332.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2311065475,"x_train":313.2556024507,"imp":11.2790879483,"elo":1339.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"x_train":420.0975247537,"imp":11.2346097142,"elo":1346.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0108193564,"x_train":7.8692134542,"imp":15.8722220541,"elo":1054.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0214757223,"x_train":17.0945668224,"imp":15.5040713652,"elo":1077.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0403341023,"x_train":52.1405202142,"imp":15.1446322779,"elo":1108.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0568329264,"x_train":97.8428503791,"imp":14.9308940147,"elo":1126.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0777112708,"x_train":240.4577474664,"imp":14.6418568122,"elo":1143.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0931559756,"x_train":484.5851613774,"imp":14.5553732057,"elo":1152.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1063228048,"x_train":933.0688585006,"imp":14.484585249,"elo":1159.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1134174096,"x_train":1378.9985712807,"imp":14.4519456161,"elo":1164.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.116575014,"x_train":1865.2663509929,"imp":14.4279093563,"elo":1168.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.24026221,"x_train":2.5445212109,"imp":15.5377117663,"elo":1062.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2792114359,"x_train":3.1259596548,"imp":14.977407387,"elo":1102.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3158110254,"x_train":6.7721912257,"imp":14.1420248066,"elo":1155.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3028704861,"x_train":13.288092342,"imp":13.7333704871,"elo":1179.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4105206818,"x_train":32.8419339595,"imp":13.1177966015,"elo":1213.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5221999466,"x_train":65.4938634865,"imp":12.9906805715,"elo":1226.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.660487887,"x_train":130.8369715186,"imp":12.9146356805,"elo":1233.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7291312705,"x_train":196.2244804971,"imp":12.8899221211,"elo":1236.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8077239936,"x_train":263.0379305179,"imp":12.8786181764,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3915661558,"x_train":2.6039043157,"imp":19.92609532,"elo":868.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6643959346,"x_train":5.1690100082,"imp":19.661131622,"elo":880.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1863927783,"x_train":13.4266251032,"imp":18.8130499001,"elo":915.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6398576272,"x_train":27.320730914,"imp":18.3254561042,"elo":938.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0095610562,"x_train":66.1669393228,"imp":17.3598274361,"elo":974.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4439102689,"x_train":132.6885087308,"imp":16.7866650086,"elo":993.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1002617128,"x_train":267.8675184086,"imp":16.1201830035,"elo":1014.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8287572918,"x_train":402.5490610457,"imp":15.6460946342,"elo":1033.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.6723835679,"x_train":540.1381360311,"imp":15.2784360822,"elo":1049.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2745146333,"x_train":2.1108344816,"imp":11.3245379459,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4016545585,"x_train":5.9546664661,"imp":10.756584727,"elo":1249.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.8377517363,"x_train":16.8872918174,"imp":9.8018502175,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.8851780262,"x_train":32.7521125678,"imp":9.3117014963,"elo":1393.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.8791663559,"x_train":84.7959741874,"imp":8.7853597392,"elo":1461.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.5304245512,"x_train":173.2180148137,"imp":8.6267188825,"elo":1486.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.6947623735,"x_train":347.0816109705,"imp":8.4879776995,"elo":1505.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":9.0305331912,"x_train":519.953899606,"imp":8.4198835742,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":9.3218182128,"x_train":700.1867028829,"imp":8.3660364871,"elo":1521.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":19.7582527317,"x_train":74.6790742113,"imp":10.3040126165,"elo":1274.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2983006451,"x_train":15.5008512826,"imp":10.6997726176,"elo":1288.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7075904089,"x_train":40.5171687704,"imp":9.9733918546,"elo":1338.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.5287946275,"x_train":107.503248256,"imp":8.7985453215,"elo":1427.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2651751376,"x_train":200.5362465049,"imp":8.2725576839,"elo":1476.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.3092160521,"x_train":460.9985605524,"imp":7.6898377498,"elo":1529.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.6686809631,"x_train":947.159574971,"imp":7.4990468521,"elo":1547.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2095781123,"x_train":1857.8046474259,"imp":7.3295844942,"elo":1563.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.1496874361,"x_train":2770.8108835678,"imp":7.2508796026,"elo":1570.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.6904227381,"x_train":3779.7025911156,"imp":7.2318390451,"elo":1574.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":1.3890423048,"x_train":6.9283541331,"imp":4.4444075824,"elo":1713.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0226031244,"x_train":11.5842534555,"imp":3.3478648597,"elo":1794.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.020234932,"x_train":27.322832323,"imp":17.6856154942,"elo":958.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1408989609,"x_train":54.4506451141,"imp":15.2891793683,"elo":1061.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3464615835,"x_train":135.4944823225,"imp":13.290731348,"elo":1162.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3627650406,"x_train":162.3794942955,"imp":13.1578480182,"elo":1173.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0621551845,"x_train":0.5271703851,"imp":15.8838229418,"elo":999.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1542795056,"x_train":3.6581130708,"imp":15.2431015169,"elo":1043.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2520439254,"x_train":13.0531670853,"imp":14.7584691425,"elo":1079.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3406296664,"x_train":26.0784291852,"imp":14.323191656,"elo":1113.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5181009628,"x_train":63.4089966233,"imp":14.0743614667,"elo":1134.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5887365969,"x_train":127.7253484811,"imp":13.9909342199,"elo":1145.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7109300332,"x_train":254.9677280133,"imp":13.9383084007,"elo":1154.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7595438385,"x_train":383.936400338,"imp":13.9204336715,"elo":1158.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7784693874,"x_train":515.7544728433,"imp":13.9114137358,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.644117959,"x_train":8.9020711389,"imp":10.4415159463,"elo":1234.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.1298601441,"x_train":24.3173471418,"imp":8.6083504215,"elo":1346.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.3924065732,"x_train":77.6709970732,"imp":6.6932795026,"elo":1504.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.0155793542,"x_train":189.6529598329,"imp":6.0296075606,"elo":1569.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.2080767736,"x_train":487.3244648114,"imp":5.5393110683,"elo":1628.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.640857572,"x_train":988.6461264381,"imp":5.2903376582,"elo":1654.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.6390543881,"x_train":1985.9304497408,"imp":5.1260738579,"elo":1674.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.478029918,"x_train":2976.2042051285,"imp":5.0345688167,"elo":1689.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.0840513338,"x_train":3995.0078687545,"imp":4.9715617017,"elo":1702.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"x_train":7.0435968115,"imp":5.5222967423,"elo":1564.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4031929556,"x_train":16.015104703,"imp":5.1210599455,"elo":1587.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8742272233,"x_train":41.0105415705,"imp":4.4217626195,"elo":1630.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.8222925994,"x_train":83.3999540139,"imp":4.038294855,"elo":1677.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7321639458,"x_train":210.1679172678,"imp":3.6708426576,"elo":1733.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.7691704179,"x_train":423.4200904906,"imp":3.4955200333,"elo":1760.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.1518842984,"x_train":852.1950752527,"imp":3.3899731202,"elo":1777.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8504991295,"x_train":1277.5788085201,"imp":3.3378559202,"elo":1787.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1651715053,"x_train":1709.0533523899,"imp":3.3026131586,"elo":1794.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.620625058,"x_train":20.2378618082,"imp":10.7641870973,"elo":1294.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"x_train":46.6209023207,"imp":5.5268547342,"elo":1605.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":55.2611832093,"x_train":70.0936255942,"imp":5.1264051337,"elo":1640.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":81.5128039491,"x_train":132.9785409439,"imp":4.7855178058,"elo":1678.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":105.6461170065,"x_train":240.939305723,"imp":4.5422915224,"elo":1707.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":122.0577162857,"x_train":589.3561296929,"imp":4.3320037401,"elo":1735.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":163.7123037519,"x_train":1183.6501039466,"imp":4.2352461267,"elo":1749.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":204.6580402482,"x_train":2376.2279314343,"imp":4.1821772214,"elo":1760.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":216.1656039799,"x_train":3574.1919132683,"imp":4.1701024214,"elo":1762.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":239.3040135069,"x_train":4786.5989331939,"imp":4.1669203949,"elo":1763.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"x_train":2.0755087942,"imp":5.1261178466,"elo":1635.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.4760486585,"x_train":39.4205185389,"imp":1.5556478433,"elo":2064.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1222431699,"x_train":1.3204095155,"imp":3.8733136542,"elo":1712.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1324525523,"x_train":9.1928711765,"imp":9.3536327032,"elo":1285.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2760220834,"x_train":22.5165548441,"imp":8.9452683683,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.5604210779,"x_train":69.1089159979,"imp":8.6316790067,"elo":1348.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.7916548768,"x_train":146.4525878247,"imp":8.1323835934,"elo":1384.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.0818885965,"x_train":371.1147116855,"imp":7.6711419131,"elo":1420.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.3749946791,"x_train":743.1660518453,"imp":7.3844894287,"elo":1445.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.6910948789,"x_train":1486.16387375,"imp":7.1771170581,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.9072725195,"x_train":2240.0758838289,"imp":6.8718997187,"elo":1474.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.0923554564,"x_train":3013.8595115999,"imp":6.817269664,"elo":1483.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.696318366,"x_train":8.5238425595,"imp":3.9382292956,"elo":1793.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"x_train":3.65846503,"imp":2.3315092965,"elo":1826.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9291777504,"x_train":323.5247663683,"imp":26.5437066212,"elo":850.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.9691116129,"x_train":458.087256476,"imp":26.4508381706,"elo":867.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.4073606358,"x_train":775.9282960298,"imp":26.2473198846,"elo":882.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.6930338556,"x_train":1467.4534767341,"imp":25.7801256282,"elo":908.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.0890319964,"x_train":3523.6508469471,"imp":25.0521742687,"elo":932.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.4354221138,"x_train":7117.6518843327,"imp":24.3599867674,"elo":950.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.0213237631,"x_train":14227.2413776887,"imp":23.9279306886,"elo":965.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3998340801,"x_train":21385.328840253,"imp":23.6532205307,"elo":977.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.9007182558,"x_train":28729.7373655693,"imp":23.4498749414,"elo":992.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0752279828,"x_train":0.9081877562,"imp":7.1846434449,"elo":1420.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0796563802,"x_train":20.4886632638,"imp":15.0354241324,"elo":1040.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1719706556,"x_train":41.7266327108,"imp":14.4070459058,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.408364249,"x_train":106.9107806776,"imp":13.4001634822,"elo":1117.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5543275658,"x_train":227.2145545781,"imp":12.8730228051,"elo":1148.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7496106434,"x_train":570.0093805222,"imp":12.1517616796,"elo":1186.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9621284433,"x_train":1173.9046077839,"imp":11.7009521655,"elo":1216.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.075322816,"x_train":2290.0065707371,"imp":11.2373560638,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1499754889,"x_train":3456.708959715,"imp":11.0269025431,"elo":1247.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.232537078,"x_train":4608.594420269,"imp":10.8904331716,"elo":1257.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.242470131,"x_train":2.2442581404,"imp":11.9785913399,"elo":1185.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6395763256,"x_train":8.0761527298,"imp":10.5410795078,"elo":1273.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1210539326,"x_train":20.7267313474,"imp":9.84983897,"elo":1333.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2386472307,"x_train":40.9776832255,"imp":9.4901130337,"elo":1367.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8103396721,"x_train":103.9269515729,"imp":9.2122964888,"elo":1395.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.1270291973,"x_train":207.504207675,"imp":9.0836511972,"elo":1408.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3814568805,"x_train":414.7237545011,"imp":8.9878774723,"elo":1420.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.578940555,"x_train":622.7467847485,"imp":8.9230143654,"elo":1427.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.6107066669,"x_train":834.9300717203,"imp":8.8864513439,"elo":1434.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":81.6126739219,"x_train":357.0213835173,"imp":18.8936466855,"elo":886.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":133.585345468,"x_train":826.6192885105,"imp":14.7756137651,"elo":1040.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":216.0194313928,"x_train":2323.5909795411,"imp":15.7066228744,"elo":1208.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":301.6106544607,"x_train":4948.4562156001,"imp":14.1706851906,"elo":1305.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":315.3980497637,"x_train":12120.2560802151,"imp":9.2996535471,"elo":1458.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":321.7409586602,"x_train":12685.0815501834,"imp":9.264955764,"elo":1465.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.7424024527,"x_train":2.4487016432,"imp":13.6835745524,"elo":1157.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8510374477,"x_train":5.9610558826,"imp":12.3645381262,"elo":1221.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.0266701319,"x_train":17.0080432107,"imp":10.6344658201,"elo":1298.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.2520438362,"x_train":34.8874663118,"imp":9.3164491876,"elo":1369.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.5268809255,"x_train":88.7983697442,"imp":8.1975537975,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.802084336,"x_train":177.9391500768,"imp":7.7973079327,"elo":1460.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.4216240621,"x_train":356.5372096884,"imp":7.5534817366,"elo":1478.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.3580011157,"x_train":533.621477288,"imp":7.4615924328,"elo":1489.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.3755494169,"x_train":714.4999525058,"imp":7.4156273541,"elo":1496.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_all/winrate_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..a03a1a40916971071a7f17217992afb6e80a3b91
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_regression/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "Nori-30M (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "Nori (default)", "TabDPT (tuned + ensembled)", "RealMLP (tuned + ensembled)", "TabDPT (tuned)", "RealTabPFN-2.5 (tuned)", "TabDPT (default)", "TabDPT-Turbo (default)", "RealTabPFN-2.5 (default)", "AutoGluon 1.4 (best, 4h)", "RealMLP (tuned)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "LightGBM (tuned + ensembled)", "iLTM (tuned + ensembled)", "TabM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "xRFM (tuned + ensembled)", "TabSwift (default)", "XGBoost (tuned + ensembled)", "CatBoost (default)", "LightGBM (tuned)", "XGBoost (tuned)", "SAP-RPT-OSS (default)", "TabM (tuned)", "xRFM (tuned)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "ChimeraBoost (tuned)", "ModernNCA (tuned)", "TabM (default)", "RealMLP (default)", "ExtraTrees (tuned + ensembled)", "ChimeraBoost (default)", "TorchMLP (tuned + ensembled)", "ModernNCA (default)", "LimiX (default)", "ExtraTrees (tuned)", "EBM (tuned + ensembled)", "LightGBM (default)", "TorchMLP (tuned)", "EBM (tuned)", "RandomForest (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "XGBoost (default)", "RandomForest (tuned)", "EBM (default)", "TabSTAR (tuned + ensembled)", "xRFM (default)", "TabSTAR (tuned)", "PerpetualBooster (tuned)", "ExtraTrees (default)", "FastaiMLP (tuned + ensembled)", "TorchMLP (default)", "FastaiMLP (tuned)", "RandomForest (default)", "TabSTAR (default)", "PerpetualBooster (default)", "KNN (tuned + ensembled)", "FastaiMLP (default)", "iLTM (default)", "KNN (tuned)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 0.6126126050949097, 0.6981981992721558, 0.7297297120094299, 0.815315306186676, 0.7432432174682617, 0.792792797088623, 0.7972972989082336, 0.837837815284729, 0.9009009003639221, 0.9054054021835327, 0.815315306186676, 0.8873873949050903, 0.954954981803894, 0.9324324131011963, 0.9594594836235046, 0.9459459185600281, 0.9639639854431152, 0.977477490901947, 0.9729729890823364, 0.9639639854431152, 0.954954981803894, 0.9594594836235046, 0.9504504799842834, 0.8963963985443115, 0.9639639854431152, 0.9684684872627258, 0.977477490901947, 0.9729729890823364, 0.9324324131011963, 0.9684684872627258, 0.9684684872627258, 0.9729729890823364, 0.977477490901947, 0.9684684872627258, 0.9909909963607788, 0.977477490901947, 0.9909909963607788, 0.9729729890823364, 0.9954954981803894, 0.9909909963607788, 0.9909909963607788, 0.9864864945411682, 0.977477490901947, 0.9729729890823364, 1.0, 0.9954954981803894, 0.9819819927215576, 0.9954954981803894, 0.9909909963607788, 1.0, 0.9954954981803894, 0.9864864945411682, 0.9234234094619751, 0.9909909963607788, 0.9279279112815857, 0.9954954981803894, 0.9819819927215576, 1.0, 1.0, 1.0, 0.9954954981803894, 0.9414414167404175, 1.0, 0.9864864945411682, 1.0, 1.0, 0.9954954981803894, 0.9909909963607788, 1.0, 1.0, 1.0], [0.38738739490509033, NaN, 0.477477490901947, 0.6531531810760498, 0.6801801919937134, 0.545045018196106, 0.6216216087341309, 0.7297297120094299, 0.630630612373352, 0.7207207083702087, 0.7162162065505981, 0.6981981992721558, 0.7387387156486511, 0.7747747898101807, 0.792792797088623, 0.8288288116455078, 0.792792797088623, 0.8918918967247009, 0.9189189076423645, 0.9144144058227539, 0.9009009003639221, 0.9234234094619751, 0.792792797088623, 0.8783783912658691, 0.869369387626648, 0.9279279112815857, 0.9189189076423645, 0.9234234094619751, 0.9369369149208069, 0.8918918967247009, 0.9459459185600281, 0.9279279112815857, 0.9414414167404175, 0.977477490901947, 0.954954981803894, 0.8963963985443115, 0.9684684872627258, 0.9639639854431152, 0.9594594836235046, 0.9729729890823364, 0.9684684872627258, 0.9504504799842834, 0.9054054021835327, 0.9459459185600281, 0.977477490901947, 0.9909909963607788, 0.9819819927215576, 0.9864864945411682, 0.9819819927215576, 0.9864864945411682, 0.9729729890823364, 0.9819819927215576, 0.9909909963607788, 0.9144144058227539, 0.9684684872627258, 0.9234234094619751, 0.9909909963607788, 0.9639639854431152, 0.9954954981803894, 1.0, 1.0, 0.9864864945411682, 0.9279279112815857, 1.0, 0.9819819927215576, 1.0, 0.9954954981803894, 0.9909909963607788, 0.9819819927215576, 1.0, 1.0, 1.0], [0.30180180072784424, 0.522522509098053, NaN, 0.6441441178321838, 0.6396396160125732, 0.6126126050949097, 0.6216216087341309, 0.6396396160125732, 0.6486486196517944, 0.792792797088623, 0.6891891956329346, 0.7522522807121277, 0.7882882952690125, 0.8783783912658691, 0.8783783912658691, 0.8423423171043396, 0.8243243098258972, 0.8558558821678162, 0.8603603839874268, 0.8873873949050903, 0.8963963985443115, 0.8918918967247009, 0.8558558821678162, 0.8783783912658691, 0.9369369149208069, 0.8873873949050903, 0.8738738894462585, 0.9099099040031433, 0.8828828930854797, 0.8738738894462585, 0.9234234094619751, 0.9279279112815857, 0.8783783912658691, 0.9729729890823364, 0.8828828930854797, 0.8963963985443115, 0.9369369149208069, 0.9684684872627258, 0.9459459185600281, 0.9189189076423645, 0.9729729890823364, 0.9684684872627258, 0.9279279112815857, 0.954954981803894, 0.9009009003639221, 0.9909909963607788, 0.977477490901947, 0.9144144058227539, 0.977477490901947, 0.9504504799842834, 0.9504504799842834, 0.977477490901947, 0.9414414167404175, 0.8603603839874268, 0.9639639854431152, 0.8783783912658691, 0.9819819927215576, 0.9684684872627258, 1.0, 1.0, 1.0, 0.9954954981803894, 0.9279279112815857, 1.0, 0.9684684872627258, 1.0, 1.0, 0.9819819927215576, 0.9909909963607788, 1.0, 1.0, 1.0], [0.2702702581882477, 0.3468468487262726, 0.3558558523654938, NaN, 0.5630630850791931, 0.44144144654273987, 0.45945945382118225, 0.522522509098053, 0.5990990996360779, 0.7297297120094299, 0.6216216087341309, 0.6081081032752991, 0.6891891956329346, 0.7432432174682617, 0.7297297120094299, 0.8603603839874268, 0.8108108043670654, 0.8648648858070374, 0.8873873949050903, 0.8873873949050903, 0.9054054021835327, 0.9099099040031433, 0.8018018007278442, 0.8648648858070374, 0.8648648858070374, 0.9234234094619751, 0.9009009003639221, 0.9324324131011963, 0.9369369149208069, 0.7972972989082336, 0.9234234094619751, 0.9099099040031433, 0.9234234094619751, 0.9414414167404175, 0.9414414167404175, 0.954954981803894, 0.9414414167404175, 0.977477490901947, 0.9369369149208069, 0.9459459185600281, 0.9864864945411682, 0.9639639854431152, 0.8603603839874268, 0.9414414167404175, 0.9639639854431152, 0.977477490901947, 0.9864864945411682, 0.9639639854431152, 0.9684684872627258, 0.9909909963607788, 0.9954954981803894, 0.9729729890823364, 0.954954981803894, 0.8963963985443115, 0.9729729890823364, 0.9009009003639221, 0.9954954981803894, 0.9504504799842834, 0.9954954981803894, 1.0, 1.0, 1.0, 0.9324324131011963, 1.0, 0.977477490901947, 1.0, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.18468467891216278, 0.3198198080062866, 0.36036035418510437, 0.4369369447231293, NaN, 0.38738739490509033, 0.47297295928001404, 0.47297295928001404, 0.5945945978164673, 0.6576576828956604, 0.6351351141929626, 0.6261261105537415, 0.6801801919937134, 0.684684693813324, 0.792792797088623, 0.8018018007278442, 0.8018018007278442, 0.8783783912658691, 0.9009009003639221, 0.8738738894462585, 0.8738738894462585, 0.9144144058227539, 0.8108108043670654, 0.8783783912658691, 0.8558558821678162, 0.9369369149208069, 0.9324324131011963, 0.9054054021835327, 0.9414414167404175, 0.8468468189239502, 0.9234234094619751, 0.9504504799842834, 0.9459459185600281, 0.9729729890823364, 0.9594594836235046, 0.9459459185600281, 0.9594594836235046, 0.9729729890823364, 0.9369369149208069, 0.9864864945411682, 0.9594594836235046, 0.9864864945411682, 0.9414414167404175, 0.9459459185600281, 0.9729729890823364, 0.9954954981803894, 0.9864864945411682, 0.9819819927215576, 0.9909909963607788, 0.9954954981803894, 1.0, 0.9909909963607788, 0.9819819927215576, 0.9009009003639221, 0.977477490901947, 0.9099099040031433, 0.9954954981803894, 0.9639639854431152, 0.9819819927215576, 1.0, 0.9909909963607788, 0.9954954981803894, 0.9369369149208069, 1.0, 0.9864864945411682, 0.9954954981803894, 1.0, 0.9864864945411682, 0.9909909963607788, 1.0, 1.0, 1.0], [0.2567567527294159, 0.45495495200157166, 0.38738739490509033, 0.5585585832595825, 0.6126126050949097, NaN, 0.5855855941772461, 0.5675675868988037, 0.684684693813324, 0.7522522807121277, 0.6936936974525452, 0.6981981992721558, 0.7207207083702087, 0.7837837934494019, 0.8333333134651184, 0.8468468189239502, 0.815315306186676, 0.8783783912658691, 0.8738738894462585, 0.8738738894462585, 0.8738738894462585, 0.9144144058227539, 0.8513513803482056, 0.8603603839874268, 0.8333333134651184, 0.8828828930854797, 0.8828828930854797, 0.8918918967247009, 0.8738738894462585, 0.9099099040031433, 0.9189189076423645, 0.8738738894462585, 0.869369387626648, 0.9009009003639221, 0.8738738894462585, 0.8963963985443115, 0.9234234094619751, 0.9459459185600281, 0.9189189076423645, 0.9099099040031433, 0.9144144058227539, 0.9009009003639221, 0.9324324131011963, 0.9144144058227539, 0.9099099040031433, 0.9504504799842834, 0.9054054021835327, 0.9054054021835327, 0.9369369149208069, 0.9324324131011963, 0.9189189076423645, 0.9459459185600281, 0.9414414167404175, 0.869369387626648, 0.9639639854431152, 0.8828828930854797, 0.9324324131011963, 0.9594594836235046, 0.9594594836235046, 0.9594594836235046, 0.9594594836235046, 0.9729729890823364, 0.9324324131011963, 0.9594594836235046, 0.9819819927215576, 0.9594594836235046, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.20720720291137695, 0.37837839126586914, 0.37837839126586914, 0.5405405163764954, 0.5270270109176636, 0.4144144058227539, NaN, 0.5360360145568848, 0.6081081032752991, 0.6981981992721558, 0.6351351141929626, 0.8198198080062866, 0.684684693813324, 0.7477477192878723, 0.8828828930854797, 0.7972972989082336, 0.7792792916297913, 0.8063063025474548, 0.8468468189239502, 0.8513513803482056, 0.837837815284729, 0.9054054021835327, 0.7882882952690125, 0.8468468189239502, 0.8603603839874268, 0.8918918967247009, 0.8918918967247009, 0.869369387626648, 0.8963963985443115, 0.815315306186676, 0.9324324131011963, 0.8873873949050903, 0.8963963985443115, 0.9234234094619751, 0.9324324131011963, 0.9279279112815857, 0.9594594836235046, 0.9504504799842834, 0.8828828930854797, 0.9729729890823364, 0.977477490901947, 0.9459459185600281, 0.9144144058227539, 0.8873873949050903, 0.954954981803894, 0.9864864945411682, 0.9819819927215576, 0.9684684872627258, 0.954954981803894, 0.9864864945411682, 0.9729729890823364, 0.9639639854431152, 0.9729729890823364, 0.8648648858070374, 0.9324324131011963, 0.8783783912658691, 0.9864864945411682, 0.954954981803894, 1.0, 1.0, 1.0, 0.9909909963607788, 0.9189189076423645, 1.0, 0.9684684872627258, 1.0, 1.0, 0.977477490901947, 0.9864864945411682, 1.0, 1.0, 1.0], [0.20270270109176636, 0.2702702581882477, 0.36036035418510437, 0.477477490901947, 0.5270270109176636, 0.4324324429035187, 0.46396395564079285, NaN, 0.5540540814399719, 0.5855855941772461, 0.5900900959968567, 0.545045018196106, 0.6351351141929626, 0.6891891956329346, 0.7657657861709595, 0.7522522807121277, 0.7477477192878723, 0.9189189076423645, 0.9144144058227539, 0.9099099040031433, 0.8513513803482056, 0.8423423171043396, 0.7612612843513489, 0.8423423171043396, 0.7702702879905701, 0.8963963985443115, 0.9144144058227539, 0.9144144058227539, 0.9099099040031433, 0.8468468189239502, 0.8963963985443115, 0.9279279112815857, 0.9279279112815857, 0.9819819927215576, 0.9369369149208069, 0.8288288116455078, 0.9594594836235046, 0.9369369149208069, 0.9504504799842834, 0.9459459185600281, 0.9639639854431152, 0.9144144058227539, 0.9054054021835327, 0.9504504799842834, 0.9414414167404175, 1.0, 0.9864864945411682, 0.9324324131011963, 0.9909909963607788, 0.9504504799842834, 0.954954981803894, 0.9909909963607788, 0.9459459185600281, 0.9009009003639221, 0.9684684872627258, 0.9009009003639221, 0.9729729890823364, 0.977477490901947, 0.9864864945411682, 1.0, 0.9909909963607788, 0.9954954981803894, 0.9234234094619751, 0.9954954981803894, 0.977477490901947, 1.0, 1.0, 0.9864864945411682, 0.9819819927215576, 1.0, 1.0, 1.0], [0.1621621549129486, 0.36936935782432556, 0.3513513505458832, 0.4009009003639221, 0.4054054021835327, 0.315315306186676, 0.3918918967247009, 0.44594594836235046, NaN, 0.5495495200157166, 0.684684693813324, 0.5405405163764954, 0.7747747898101807, 0.7792792916297913, 0.6936936974525452, 0.7747747898101807, 0.6126126050949097, 0.7882882952690125, 0.8063063025474548, 0.815315306186676, 0.8828828930854797, 0.792792797088623, 0.7657657861709595, 0.869369387626648, 0.7612612843513489, 0.8558558821678162, 0.8513513803482056, 0.837837815284729, 0.8783783912658691, 0.7972972989082336, 0.8513513803482056, 0.9009009003639221, 0.8468468189239502, 0.9504504799842834, 0.869369387626648, 0.9189189076423645, 0.9099099040031433, 0.9459459185600281, 0.9504504799842834, 0.9189189076423645, 0.9369369149208069, 0.9504504799842834, 0.8828828930854797, 0.9414414167404175, 0.8873873949050903, 0.977477490901947, 0.9594594836235046, 0.8963963985443115, 0.9819819927215576, 0.9819819927215576, 0.9594594836235046, 0.9864864945411682, 0.9504504799842834, 0.8288288116455078, 0.9594594836235046, 0.8288288116455078, 0.9864864945411682, 0.9639639854431152, 1.0, 1.0, 1.0, 1.0, 0.9144144058227539, 1.0, 0.9819819927215576, 1.0, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.09909909963607788, 0.27927929162979126, 0.20720720291137695, 0.2702702581882477, 0.342342346906662, 0.2477477490901947, 0.30180180072784424, 0.4144144058227539, 0.45045045018196106, NaN, 0.5135135054588318, 0.4189189076423645, 0.5855855941772461, 0.6126126050949097, 0.6036036014556885, 0.6891891956329346, 0.7387387156486511, 0.7837837934494019, 0.7792792916297913, 0.7837837934494019, 0.8558558821678162, 0.7792792916297913, 0.7522522807121277, 0.8783783912658691, 0.7207207083702087, 0.8648648858070374, 0.8288288116455078, 0.8513513803482056, 0.869369387626648, 0.792792797088623, 0.8738738894462585, 0.9279279112815857, 0.8108108043670654, 0.9414414167404175, 0.8243243098258972, 0.8963963985443115, 0.9144144058227539, 0.9864864945411682, 0.9414414167404175, 0.9099099040031433, 0.9054054021835327, 0.9684684872627258, 0.9144144058227539, 0.9504504799842834, 0.8738738894462585, 0.9864864945411682, 0.9369369149208069, 0.9009009003639221, 0.9819819927215576, 0.977477490901947, 0.9684684872627258, 0.9864864945411682, 0.9729729890823364, 0.815315306186676, 0.9684684872627258, 0.8198198080062866, 0.9864864945411682, 0.977477490901947, 1.0, 0.9954954981803894, 0.9954954981803894, 0.9954954981803894, 0.9009009003639221, 1.0, 0.9684684872627258, 1.0, 1.0, 0.9819819927215576, 0.9909909963607788, 1.0, 1.0, 1.0], [0.09459459781646729, 0.28378379344940186, 0.31081080436706543, 0.37837839126586914, 0.36486485600471497, 0.30630630254745483, 0.36486485600471497, 0.4099099040031433, 0.315315306186676, 0.4864864945411682, NaN, 0.5090090036392212, 0.6756756901741028, 0.684684693813324, 0.5945945978164673, 0.7297297120094299, 0.5765765905380249, 0.7612612843513489, 0.7792792916297913, 0.7837837934494019, 0.8333333134651184, 0.7432432174682617, 0.7522522807121277, 0.837837815284729, 0.7162162065505981, 0.8333333134651184, 0.815315306186676, 0.8423423171043396, 0.8468468189239502, 0.7837837934494019, 0.8108108043670654, 0.8873873949050903, 0.8243243098258972, 0.954954981803894, 0.8468468189239502, 0.8963963985443115, 0.8738738894462585, 0.9279279112815857, 0.9369369149208069, 0.9099099040031433, 0.9369369149208069, 0.9459459185600281, 0.8828828930854797, 0.9324324131011963, 0.8603603839874268, 0.977477490901947, 0.9504504799842834, 0.8963963985443115, 0.9684684872627258, 0.9729729890823364, 0.9504504799842834, 0.9729729890823364, 0.9414414167404175, 0.8198198080062866, 0.9639639854431152, 0.8243243098258972, 0.977477490901947, 0.9594594836235046, 0.9954954981803894, 1.0, 1.0, 0.9909909963607788, 0.9189189076423645, 1.0, 0.9819819927215576, 1.0, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.18468467891216278, 0.30180180072784424, 0.2477477490901947, 0.3918918967247009, 0.37387385964393616, 0.30180180072784424, 0.18018017709255219, 0.45495495200157166, 0.45945945382118225, 0.5810810923576355, 0.4909909963607788, NaN, 0.5765765905380249, 0.6351351141929626, 0.6869369149208069, 0.6441441178321838, 0.6531531810760498, 0.6981981992721558, 0.7297297120094299, 0.7207207083702087, 0.7432432174682617, 0.792792797088623, 0.684684693813324, 0.792792797088623, 0.7702702879905701, 0.8108108043670654, 0.7702702879905701, 0.7792792916297913, 0.8108108043670654, 0.7297297120094299, 0.8603603839874268, 0.8603603839874268, 0.7972972989082336, 0.8828828930854797, 0.8333333134651184, 0.8513513803482056, 0.8963963985443115, 0.9144144058227539, 0.8603603839874268, 0.9099099040031433, 0.9144144058227539, 0.9009009003639221, 0.869369387626648, 0.8648648858070374, 0.8828828930854797, 0.9594594836235046, 0.9279279112815857, 0.9054054021835327, 0.8963963985443115, 0.954954981803894, 0.9234234094619751, 0.9099099040031433, 0.9459459185600281, 0.8063063025474548, 0.9009009003639221, 0.8108108043670654, 0.9639639854431152, 0.9099099040031433, 0.977477490901947, 0.9954954981803894, 1.0, 0.9639639854431152, 0.8918918967247009, 1.0, 0.9324324131011963, 0.9954954981803894, 1.0, 0.9729729890823364, 0.9684684872627258, 1.0, 1.0, 1.0], [0.11261261254549026, 0.2612612545490265, 0.21171170473098755, 0.31081080436706543, 0.3198198080062866, 0.27927929162979126, 0.315315306186676, 0.36486485600471497, 0.22522522509098053, 0.4144144058227539, 0.3243243098258972, 0.4234234094619751, NaN, 0.5765765905380249, 0.5045045018196106, 0.5855855941772461, 0.47297295928001404, 0.7027027010917664, 0.7162162065505981, 0.6936936974525452, 0.7792792916297913, 0.6081081032752991, 0.6891891956329346, 0.7297297120094299, 0.6576576828956604, 0.7477477192878723, 0.7522522807121277, 0.7702702879905701, 0.7747747898101807, 0.7297297120094299, 0.6936936974525452, 0.8108108043670654, 0.7702702879905701, 0.8963963985443115, 0.7792792916297913, 0.8063063025474548, 0.8243243098258972, 0.869369387626648, 0.9324324131011963, 0.8198198080062866, 0.9009009003639221, 0.9144144058227539, 0.8558558821678162, 0.9369369149208069, 0.8108108043670654, 0.9414414167404175, 0.9279279112815857, 0.8243243098258972, 0.9729729890823364, 0.9144144058227539, 0.9459459185600281, 0.977477490901947, 0.9099099040031433, 0.8288288116455078, 0.954954981803894, 0.8243243098258972, 0.9729729890823364, 0.954954981803894, 0.9864864945411682, 0.9909909963607788, 0.9864864945411682, 0.9954954981803894, 0.9234234094619751, 1.0, 0.9819819927215576, 1.0, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.045045044273138046, 0.22522522509098053, 0.12162162363529205, 0.2567567527294159, 0.315315306186676, 0.21621622145175934, 0.2522522509098053, 0.31081080436706543, 0.22072072327136993, 0.38738739490509033, 0.315315306186676, 0.36486485600471497, 0.4234234094619751, NaN, 0.45495495200157166, 0.630630612373352, 0.4909909963607788, 0.6981981992721558, 0.7117117047309875, 0.6801801919937134, 0.7252252101898193, 0.6756756901741028, 0.6981981992721558, 0.7387387156486511, 0.6576576828956604, 0.7522522807121277, 0.7432432174682617, 0.7837837934494019, 0.7702702879905701, 0.7207207083702087, 0.7477477192878723, 0.8198198080062866, 0.7612612843513489, 0.8648648858070374, 0.7702702879905701, 0.8333333134651184, 0.8468468189239502, 0.8783783912658691, 0.9099099040031433, 0.8333333134651184, 0.9099099040031433, 0.9144144058227539, 0.837837815284729, 0.9189189076423645, 0.792792797088623, 0.9414414167404175, 0.9279279112815857, 0.8198198080062866, 0.9414414167404175, 0.9144144058227539, 0.9324324131011963, 0.9414414167404175, 0.9009009003639221, 0.7837837934494019, 0.9459459185600281, 0.792792797088623, 0.9639639854431152, 0.9369369149208069, 0.9819819927215576, 0.9909909963607788, 0.9909909963607788, 0.9864864945411682, 0.8873873949050903, 0.9909909963607788, 0.9684684872627258, 1.0, 1.0, 0.9729729890823364, 0.977477490901947, 1.0, 1.0, 1.0], [0.06756756454706192, 0.20720720291137695, 0.12162162363529205, 0.2702702581882477, 0.20720720291137695, 0.1666666716337204, 0.11711711436510086, 0.23423422873020172, 0.30630630254745483, 0.3963963985443115, 0.4054054021835327, 0.3130630552768707, 0.4954954981803894, 0.545045018196106, NaN, 0.6171171069145203, 0.545045018196106, 0.7162162065505981, 0.7612612843513489, 0.6981981992721558, 0.7027027010917664, 0.7207207083702087, 0.6576576828956604, 0.7342342138290405, 0.6891891956329346, 0.7657657861709595, 0.7837837934494019, 0.7972972989082336, 0.792792797088623, 0.7117117047309875, 0.7657657861709595, 0.8198198080062866, 0.8018018007278442, 0.8513513803482056, 0.815315306186676, 0.837837815284729, 0.8783783912658691, 0.8333333134651184, 0.8648648858070374, 0.8828828930854797, 0.9099099040031433, 0.8738738894462585, 0.8423423171043396, 0.8738738894462585, 0.8468468189239502, 0.9324324131011963, 0.9189189076423645, 0.8648648858070374, 0.9054054021835327, 0.9054054021835327, 0.9189189076423645, 0.9054054021835327, 0.9279279112815857, 0.8198198080062866, 0.9144144058227539, 0.8243243098258972, 0.9144144058227539, 0.9009009003639221, 0.9729729890823364, 1.0, 0.9819819927215576, 0.9459459185600281, 0.8963963985443115, 0.9639639854431152, 0.9234234094619751, 0.9954954981803894, 1.0, 0.9504504799842834, 0.9594594836235046, 1.0, 1.0, 1.0], [0.04054053872823715, 0.171171173453331, 0.157657653093338, 0.13963964581489563, 0.19819819927215576, 0.15315315127372742, 0.20270270109176636, 0.2477477490901947, 0.22522522509098053, 0.31081080436706543, 0.2702702581882477, 0.3558558523654938, 0.4144144058227539, 0.36936935782432556, 0.38288289308547974, NaN, 0.5180180072784424, 0.6711711883544922, 0.6711711883544922, 0.7837837934494019, 0.6891891956329346, 0.6261261105537415, 0.6711711883544922, 0.6936936974525452, 0.6081081032752991, 0.8288288116455078, 0.7747747898101807, 0.8603603839874268, 0.8513513803482056, 0.6756756901741028, 0.7252252101898193, 0.792792797088623, 0.8333333134651184, 0.869369387626648, 0.837837815284729, 0.8198198080062866, 0.792792797088623, 0.7837837934494019, 0.869369387626648, 0.9054054021835327, 0.9324324131011963, 0.9189189076423645, 0.815315306186676, 0.8783783912658691, 0.8513513803482056, 0.9639639854431152, 0.9279279112815857, 0.8783783912658691, 0.9144144058227539, 0.9684684872627258, 0.9819819927215576, 0.9189189076423645, 0.9234234094619751, 0.7972972989082336, 0.9279279112815857, 0.8063063025474548, 0.9639639854431152, 0.9189189076423645, 0.9954954981803894, 0.9954954981803894, 1.0, 0.977477490901947, 0.8783783912658691, 0.9909909963607788, 0.9369369149208069, 1.0, 1.0, 0.9639639854431152, 0.9729729890823364, 1.0, 1.0, 1.0], [0.054054055362939835, 0.20720720291137695, 0.1756756752729416, 0.18918919563293457, 0.19819819927215576, 0.18468467891216278, 0.22072072327136993, 0.2522522509098053, 0.38738739490509033, 0.2612612545490265, 0.4234234094619751, 0.3468468487262726, 0.5270270109176636, 0.5090090036392212, 0.45495495200157166, 0.4819819927215576, NaN, 0.6531531810760498, 0.662162184715271, 0.6576576828956604, 0.6981981992721558, 0.6891891956329346, 0.5990990996360779, 0.6711711883544922, 0.5855855941772461, 0.7477477192878723, 0.6981981992721558, 0.7657657861709595, 0.7567567825317383, 0.7297297120094299, 0.7702702879905701, 0.7837837934494019, 0.7297297120094299, 0.8288288116455078, 0.7882882952690125, 0.8288288116455078, 0.8558558821678162, 0.8918918967247009, 0.8198198080062866, 0.8468468189239502, 0.8873873949050903, 0.9369369149208069, 0.815315306186676, 0.8513513803482056, 0.8423423171043396, 0.9144144058227539, 0.9369369149208069, 0.8648648858070374, 0.9279279112815857, 0.9459459185600281, 0.9414414167404175, 0.9504504799842834, 0.9414414167404175, 0.7972972989082336, 0.8873873949050903, 0.8063063025474548, 0.9729729890823364, 0.9054054021835327, 0.977477490901947, 0.9819819927215576, 0.9909909963607788, 0.9909909963607788, 0.8738738894462585, 0.9909909963607788, 0.9504504799842834, 0.9909909963607788, 1.0, 0.9639639854431152, 0.977477490901947, 0.9954954981803894, 0.9954954981803894, 0.9954954981803894], [0.036036036908626556, 0.10810811072587967, 0.14414414763450623, 0.13513512909412384, 0.12162162363529205, 0.12162162363529205, 0.19369369745254517, 0.0810810774564743, 0.21171170473098755, 0.21621622145175934, 0.2387387454509735, 0.30180180072784424, 0.29729729890823364, 0.30180180072784424, 0.28378379344940186, 0.3288288414478302, 0.3468468487262726, NaN, 0.6396396160125732, 0.6036036014556885, 0.5135135054588318, 0.4279279410839081, 0.5990990996360779, 0.4864864945411682, 0.5405405163764954, 0.7657657861709595, 0.7972972989082336, 0.7297297120094299, 0.7837837934494019, 0.6036036014556885, 0.5360360145568848, 0.6531531810760498, 0.7792792916297913, 0.7657657861709595, 0.8423423171043396, 0.7567567825317383, 0.6801801919937134, 0.7567567825317383, 0.8018018007278442, 0.9369369149208069, 0.7882882952690125, 0.8243243098258972, 0.7882882952690125, 0.837837815284729, 0.837837815284729, 0.9504504799842834, 0.8063063025474548, 0.8738738894462585, 0.9189189076423645, 0.9639639854431152, 0.9909909963607788, 0.9234234094619751, 0.9369369149208069, 0.7972972989082336, 0.8738738894462585, 0.7972972989082336, 0.9819819927215576, 0.8918918967247009, 0.9369369149208069, 0.9279279112815857, 0.9684684872627258, 0.9864864945411682, 0.869369387626648, 1.0, 0.9459459185600281, 0.9954954981803894, 1.0, 0.9684684872627258, 0.9729729890823364, 1.0, 1.0, 1.0], [0.022522522136569023, 0.0810810774564743, 0.13963964581489563, 0.11261261254549026, 0.09909909963607788, 0.12612612545490265, 0.15315315127372742, 0.0855855867266655, 0.19369369745254517, 0.22072072327136993, 0.22072072327136993, 0.2702702581882477, 0.28378379344940186, 0.28828829526901245, 0.2387387454509735, 0.3288288414478302, 0.3378378450870514, 0.36036035418510437, NaN, 0.522522509098053, 0.4909909963607788, 0.4189189076423645, 0.5630630850791931, 0.45495495200157166, 0.4954954981803894, 0.6756756901741028, 0.684684693813324, 0.684684693813324, 0.7297297120094299, 0.6036036014556885, 0.5270270109176636, 0.6216216087341309, 0.7342342138290405, 0.7207207083702087, 0.8063063025474548, 0.7342342138290405, 0.6666666865348816, 0.7432432174682617, 0.7882882952690125, 0.9144144058227539, 0.7702702879905701, 0.8108108043670654, 0.7837837934494019, 0.8018018007278442, 0.7972972989082336, 0.9279279112815857, 0.792792797088623, 0.8603603839874268, 0.8873873949050903, 0.954954981803894, 0.9909909963607788, 0.9144144058227539, 0.9234234094619751, 0.792792797088623, 0.8738738894462585, 0.8018018007278442, 0.9729729890823364, 0.8738738894462585, 0.9369369149208069, 0.9234234094619751, 0.9594594836235046, 0.977477490901947, 0.8603603839874268, 0.9954954981803894, 0.9369369149208069, 0.9864864945411682, 1.0, 0.9639639854431152, 0.9684684872627258, 1.0, 1.0, 1.0], [0.027027027681469917, 0.0855855867266655, 0.11261261254549026, 0.11261261254549026, 0.12612612545490265, 0.12612612545490265, 0.14864864945411682, 0.09009008854627609, 0.18468467891216278, 0.21621622145175934, 0.21621622145175934, 0.27927929162979126, 0.30630630254745483, 0.3198198080062866, 0.30180180072784424, 0.21621622145175934, 0.342342346906662, 0.3963963985443115, 0.477477490901947, NaN, 0.4909909963607788, 0.46846845746040344, 0.5, 0.4819819927215576, 0.5405405163764954, 0.6756756901741028, 0.7117117047309875, 0.7792792916297913, 0.7252252101898193, 0.4909909963607788, 0.5675675868988037, 0.5630630850791931, 0.7657657861709595, 0.6666666865348816, 0.8333333134651184, 0.7117117047309875, 0.7252252101898193, 0.7612612843513489, 0.815315306186676, 0.8963963985443115, 0.7117117047309875, 0.792792797088623, 0.7117117047309875, 0.815315306186676, 0.8108108043670654, 0.9819819927215576, 0.7432432174682617, 0.837837815284729, 0.8918918967247009, 0.9729729890823364, 0.977477490901947, 0.8963963985443115, 0.9414414167404175, 0.7747747898101807, 0.8468468189239502, 0.7792792916297913, 0.9729729890823364, 0.8873873949050903, 0.9729729890823364, 0.8963963985443115, 0.9909909963607788, 0.9729729890823364, 0.8468468189239502, 1.0, 0.9234234094619751, 0.9909909963607788, 1.0, 0.9459459185600281, 0.9684684872627258, 1.0, 1.0, 1.0], [0.036036036908626556, 0.09909909963607788, 0.10360360145568848, 0.09459459781646729, 0.12612612545490265, 0.12612612545490265, 0.1621621549129486, 0.14864864945411682, 0.11711711436510086, 0.14414414763450623, 0.1666666716337204, 0.2567567527294159, 0.22072072327136993, 0.2747747600078583, 0.29729729890823364, 0.31081080436706543, 0.30180180072784424, 0.4864864945411682, 0.5090090036392212, 0.5090090036392212, NaN, 0.44144144654273987, 0.5, 0.47297295928001404, 0.522522509098053, 0.6351351141929626, 0.5765765905380249, 0.630630612373352, 0.6666666865348816, 0.5855855941772461, 0.5540540814399719, 0.5990990996360779, 0.6441441178321838, 0.8198198080062866, 0.684684693813324, 0.7162162065505981, 0.5855855941772461, 0.7837837934494019, 0.837837815284729, 0.7297297120094299, 0.8468468189239502, 0.8198198080062866, 0.7882882952690125, 0.8468468189239502, 0.7297297120094299, 0.9144144058227539, 0.8558558821678162, 0.7612612843513489, 0.9369369149208069, 0.8603603839874268, 0.9054054021835327, 0.9414414167404175, 0.815315306186676, 0.7882882952690125, 0.9369369149208069, 0.7882882952690125, 0.9324324131011963, 0.9099099040031433, 0.9819819927215576, 0.9729729890823364, 0.977477490901947, 0.977477490901947, 0.8603603839874268, 0.9819819927215576, 0.9594594836235046, 0.9909909963607788, 0.9864864945411682, 0.9684684872627258, 0.9864864945411682, 0.9909909963607788, 0.9909909963607788, 0.9909909963607788], [0.045045044273138046, 0.07657657563686371, 0.10810811072587967, 0.09009008854627609, 0.0855855867266655, 0.0855855867266655, 0.09459459781646729, 0.157657653093338, 0.20720720291137695, 0.22072072327136993, 0.2567567527294159, 0.20720720291137695, 0.3918918967247009, 0.3243243098258972, 0.27927929162979126, 0.37387385964393616, 0.31081080436706543, 0.5720720887184143, 0.5810810923576355, 0.5315315127372742, 0.5585585832595825, NaN, 0.5090090036392212, 0.5495495200157166, 0.4324324429035187, 0.6261261105537415, 0.6486486196517944, 0.6396396160125732, 0.6711711883544922, 0.5990990996360779, 0.8603603839874268, 0.6441441178321838, 0.6576576828956604, 0.7117117047309875, 0.707207202911377, 0.6441441178321838, 0.9054054021835327, 0.837837815284729, 0.7342342138290405, 0.7882882952690125, 0.8198198080062866, 0.7522522807121277, 0.7162162065505981, 0.7522522807121277, 0.6936936974525452, 0.8333333134651184, 0.8648648858070374, 0.7522522807121277, 0.7612612843513489, 0.8333333134651184, 0.8828828930854797, 0.7702702879905701, 0.8063063025474548, 0.7792792916297913, 0.837837815284729, 0.7882882952690125, 0.8513513803482056, 0.815315306186676, 0.9414414167404175, 0.9684684872627258, 0.9819819927215576, 0.9369369149208069, 0.8648648858070374, 0.9369369149208069, 0.9099099040031433, 0.9639639854431152, 0.9909909963607788, 0.9369369149208069, 0.9459459185600281, 1.0, 1.0, 1.0], [0.04054053872823715, 0.20720720291137695, 0.14414414763450623, 0.19819819927215576, 0.18918919563293457, 0.14864864945411682, 0.21171170473098755, 0.2387387454509735, 0.23423422873020172, 0.2477477490901947, 0.2477477490901947, 0.315315306186676, 0.31081080436706543, 0.30180180072784424, 0.342342346906662, 0.3288288414478302, 0.4009009003639221, 0.4009009003639221, 0.4369369447231293, 0.5, 0.5, 0.4909909963607788, NaN, 0.477477490901947, 0.5270270109176636, 0.5900900959968567, 0.6126126050949097, 0.5810810923576355, 0.5990990996360779, 0.4909909963607788, 0.5765765905380249, 0.5900900959968567, 0.6396396160125732, 0.6216216087341309, 0.6981981992721558, 0.8738738894462585, 0.6126126050949097, 0.7252252101898193, 0.7522522807121277, 0.7252252101898193, 0.7477477192878723, 0.8783783912658691, 0.7297297120094299, 0.7702702879905701, 0.7117117047309875, 0.8018018007278442, 0.7477477192878723, 0.7477477192878723, 0.8558558821678162, 0.869369387626648, 0.9054054021835327, 0.8828828930854797, 0.792792797088623, 0.7657657861709595, 0.9234234094619751, 0.7792792916297913, 0.8963963985443115, 0.8963963985443115, 0.9279279112815857, 0.9099099040031433, 0.9369369149208069, 0.9639639854431152, 0.8198198080062866, 0.954954981803894, 0.954954981803894, 0.9819819927215576, 1.0, 0.977477490901947, 0.9819819927215576, 0.9954954981803894, 0.9954954981803894, 1.0], [0.04954954981803894, 0.12162162363529205, 0.12162162363529205, 0.13513512909412384, 0.12162162363529205, 0.13963964581489563, 0.15315315127372742, 0.157657653093338, 0.13063062727451324, 0.12162162363529205, 0.1621621549129486, 0.20720720291137695, 0.2702702581882477, 0.2612612545490265, 0.2657657563686371, 0.30630630254745483, 0.3288288414478302, 0.5135135054588318, 0.545045018196106, 0.5180180072784424, 0.5270270109176636, 0.45045045018196106, 0.522522509098053, NaN, 0.5090090036392212, 0.6081081032752991, 0.6171171069145203, 0.6351351141929626, 0.6666666865348816, 0.5630630850791931, 0.5495495200157166, 0.7342342138290405, 0.6576576828956604, 0.7657657861709595, 0.6981981992721558, 0.6936936974525452, 0.6396396160125732, 0.6756756901741028, 0.7567567825317383, 0.7387387156486511, 0.7747747898101807, 0.8018018007278442, 0.8423423171043396, 0.7837837934494019, 0.7657657861709595, 0.8783783912658691, 0.7972972989082336, 0.7837837934494019, 0.837837815284729, 0.8198198080062866, 0.9054054021835327, 0.8468468189239502, 0.815315306186676, 0.7657657861709595, 0.9729729890823364, 0.7702702879905701, 0.8513513803482056, 0.9234234094619751, 0.8783783912658691, 0.9369369149208069, 0.9054054021835327, 0.977477490901947, 0.8198198080062866, 0.954954981803894, 0.9684684872627258, 0.9729729890823364, 0.9819819927215576, 0.9819819927215576, 0.9819819927215576, 0.9954954981803894, 0.9954954981803894, 0.9954954981803894], [0.10360360145568848, 0.13063062727451324, 0.06306306272745132, 0.13513512909412384, 0.14414414763450623, 0.1666666716337204, 0.13963964581489563, 0.22972972691059113, 0.2387387454509735, 0.27927929162979126, 0.28378379344940186, 0.22972972691059113, 0.342342346906662, 0.342342346906662, 0.31081080436706543, 0.3918918967247009, 0.4144144058227539, 0.45945945382118225, 0.5045045018196106, 0.45945945382118225, 0.477477490901947, 0.5675675868988037, 0.47297295928001404, 0.4909909963607788, NaN, 0.5180180072784424, 0.522522509098053, 0.5720720887184143, 0.5810810923576355, 0.5765765905380249, 0.6351351141929626, 0.6261261105537415, 0.5720720887184143, 0.6801801919937134, 0.6036036014556885, 0.662162184715271, 0.6351351141929626, 0.8738738894462585, 0.7162162065505981, 0.6666666865348816, 0.792792797088623, 0.8108108043670654, 0.7342342138290405, 0.7387387156486511, 0.7207207083702087, 0.7882882952690125, 0.815315306186676, 0.7657657861709595, 0.7837837934494019, 0.7747747898101807, 0.8558558821678162, 0.7882882952690125, 0.8198198080062866, 0.7792792916297913, 0.8738738894462585, 0.7882882952690125, 0.8333333134651184, 0.8738738894462585, 0.9639639854431152, 0.9459459185600281, 0.9684684872627258, 0.8963963985443115, 0.8558558821678162, 0.9369369149208069, 0.9639639854431152, 1.0, 0.954954981803894, 0.9819819927215576, 0.9909909963607788, 1.0, 1.0, 1.0], [0.036036036908626556, 0.07207207381725311, 0.11261261254549026, 0.07657657563686371, 0.06306306272745132, 0.11711711436510086, 0.10810811072587967, 0.10360360145568848, 0.14414414763450623, 0.13513512909412384, 0.1666666716337204, 0.18918919563293457, 0.2522522509098053, 0.2477477490901947, 0.23423422873020172, 0.171171173453331, 0.2522522509098053, 0.23423422873020172, 0.3243243098258972, 0.3243243098258972, 0.36486485600471497, 0.37387385964393616, 0.4099099040031433, 0.3918918967247009, 0.4819819927215576, NaN, 0.5405405163764954, 0.5090090036392212, 0.6396396160125732, 0.4864864945411682, 0.5090090036392212, 0.5495495200157166, 0.6801801919937134, 0.6351351141929626, 0.707207202911377, 0.6666666865348816, 0.6081081032752991, 0.7297297120094299, 0.7567567825317383, 0.869369387626648, 0.7027027010917664, 0.7522522807121277, 0.7207207083702087, 0.7837837934494019, 0.792792797088623, 0.9729729890823364, 0.7477477192878723, 0.8243243098258972, 0.8738738894462585, 0.9234234094619751, 0.9459459185600281, 0.8918918967247009, 0.9234234094619751, 0.7702702879905701, 0.837837815284729, 0.7837837934494019, 0.9594594836235046, 0.8603603839874268, 0.9459459185600281, 0.8873873949050903, 0.9819819927215576, 0.9819819927215576, 0.837837815284729, 0.9864864945411682, 0.9279279112815857, 0.9909909963607788, 1.0, 0.9639639854431152, 0.9639639854431152, 1.0, 1.0, 1.0], [0.03153153136372566, 0.0810810774564743, 0.12612612545490265, 0.09909909963607788, 0.06756756454706192, 0.11711711436510086, 0.10810811072587967, 0.0855855867266655, 0.14864864945411682, 0.171171173453331, 0.18468467891216278, 0.22972972691059113, 0.2477477490901947, 0.2567567527294159, 0.21621622145175934, 0.22522522509098053, 0.30180180072784424, 0.20270270109176636, 0.315315306186676, 0.28828829526901245, 0.4234234094619751, 0.3513513505458832, 0.38738739490509033, 0.38288289308547974, 0.477477490901947, 0.45945945382118225, NaN, 0.46846845746040344, 0.5090090036392212, 0.4234234094619751, 0.47297295928001404, 0.545045018196106, 0.6351351141929626, 0.5945945978164673, 0.6936936974525452, 0.6126126050949097, 0.5900900959968567, 0.6801801919937134, 0.707207202911377, 0.8648648858070374, 0.6711711883544922, 0.6666666865348816, 0.6261261105537415, 0.7162162065505981, 0.8018018007278442, 0.7972972989082336, 0.7027027010917664, 0.8423423171043396, 0.8063063025474548, 0.9189189076423645, 0.8198198080062866, 0.8198198080062866, 0.9144144058227539, 0.7792792916297913, 0.7747747898101807, 0.7837837934494019, 0.9414414167404175, 0.8333333134651184, 0.9189189076423645, 0.8018018007278442, 0.9324324131011963, 0.9414414167404175, 0.8603603839874268, 1.0, 0.8918918967247009, 0.9864864945411682, 1.0, 0.9594594836235046, 0.9684684872627258, 1.0, 1.0, 1.0], [0.022522522136569023, 0.07657657563686371, 0.09009008854627609, 0.06756756454706192, 0.09459459781646729, 0.10810811072587967, 0.13063062727451324, 0.0855855867266655, 0.1621621549129486, 0.14864864945411682, 0.157657653093338, 0.22072072327136993, 0.22972972691059113, 0.21621622145175934, 0.20270270109176636, 0.13963964581489563, 0.23423422873020172, 0.2702702581882477, 0.315315306186676, 0.22072072327136993, 0.36936935782432556, 0.36036035418510437, 0.4189189076423645, 0.36486485600471497, 0.4279279410839081, 0.4909909963607788, 0.5315315127372742, NaN, 0.5495495200157166, 0.45495495200157166, 0.4864864945411682, 0.45945945382118225, 0.6261261105537415, 0.5585585832595825, 0.6801801919937134, 0.6261261105537415, 0.6441441178321838, 0.707207202911377, 0.7522522807121277, 0.7882882952690125, 0.6711711883544922, 0.707207202911377, 0.6666666865348816, 0.7567567825317383, 0.7477477192878723, 0.9144144058227539, 0.6936936974525452, 0.8108108043670654, 0.8288288116455078, 0.8828828930854797, 0.9459459185600281, 0.837837815284729, 0.8558558821678162, 0.7657657861709595, 0.837837815284729, 0.7747747898101807, 0.9099099040031433, 0.8288288116455078, 0.9189189076423645, 0.869369387626648, 0.954954981803894, 0.9504504799842834, 0.8288288116455078, 0.9864864945411682, 0.9099099040031433, 0.977477490901947, 1.0, 0.9414414167404175, 0.9459459185600281, 0.9909909963607788, 0.9909909963607788, 0.9954954981803894], [0.027027027681469917, 0.06306306272745132, 0.11711711436510086, 0.06306306272745132, 0.05855855718255043, 0.12612612545490265, 0.10360360145568848, 0.09009008854627609, 0.12162162363529205, 0.13063062727451324, 0.15315315127372742, 0.18918919563293457, 0.22522522509098053, 0.22972972691059113, 0.20720720291137695, 0.14864864945411682, 0.2432432472705841, 0.21621622145175934, 0.2702702581882477, 0.2747747600078583, 0.3333333432674408, 0.3288288414478302, 0.4009009003639221, 0.3333333432674408, 0.4189189076423645, 0.36036035418510437, 0.4909909963607788, 0.45045045018196106, NaN, 0.4819819927215576, 0.46846845746040344, 0.5090090036392212, 0.6036036014556885, 0.5675675868988037, 0.6396396160125732, 0.630630612373352, 0.5585585832595825, 0.6981981992721558, 0.7297297120094299, 0.8243243098258972, 0.6936936974525452, 0.7207207083702087, 0.7117117047309875, 0.7522522807121277, 0.7522522807121277, 0.9099099040031433, 0.7387387156486511, 0.8198198080062866, 0.8288288116455078, 0.8918918967247009, 0.9279279112815857, 0.837837815284729, 0.8963963985443115, 0.7702702879905701, 0.8288288116455078, 0.7702702879905701, 0.9234234094619751, 0.8468468189239502, 0.9369369149208069, 0.8963963985443115, 0.9684684872627258, 0.9639639854431152, 0.837837815284729, 0.9864864945411682, 0.9144144058227539, 0.9819819927215576, 1.0, 0.954954981803894, 0.9639639854431152, 1.0, 1.0, 1.0], [0.06756756454706192, 0.10810811072587967, 0.12612612545490265, 0.20270270109176636, 0.15315315127372742, 0.09009008854627609, 0.18468467891216278, 0.15315315127372742, 0.20270270109176636, 0.20720720291137695, 0.21621622145175934, 0.2702702581882477, 0.2702702581882477, 0.27927929162979126, 0.28828829526901245, 0.3243243098258972, 0.2702702581882477, 0.3963963985443115, 0.3963963985443115, 0.5090090036392212, 0.4144144058227539, 0.4009009003639221, 0.5090090036392212, 0.4369369447231293, 0.4234234094619751, 0.5135135054588318, 0.5765765905380249, 0.545045018196106, 0.5180180072784424, NaN, 0.477477490901947, 0.4819819927215576, 0.630630612373352, 0.5495495200157166, 0.630630612373352, 0.6126126050949097, 0.46846845746040344, 0.5900900959968567, 0.6576576828956604, 0.7117117047309875, 0.5990990996360779, 0.6936936974525452, 0.6756756901741028, 0.6576576828956604, 0.7252252101898193, 0.7297297120094299, 0.5900900959968567, 0.7702702879905701, 0.7432432174682617, 0.7522522807121277, 0.7882882952690125, 0.7432432174682617, 0.8288288116455078, 0.7432432174682617, 0.7342342138290405, 0.7522522807121277, 0.7657657861709595, 0.8198198080062866, 0.8828828930854797, 0.8063063025474548, 0.8918918967247009, 0.837837815284729, 0.8783783912658691, 0.9054054021835327, 0.9189189076423645, 0.9369369149208069, 0.9054054021835327, 0.9864864945411682, 0.9864864945411682, 0.9954954981803894, 0.9954954981803894, 1.0], [0.03153153136372566, 0.054054055362939835, 0.07657657563686371, 0.07657657563686371, 0.07657657563686371, 0.0810810774564743, 0.06756756454706192, 0.10360360145568848, 0.14864864945411682, 0.12612612545490265, 0.18918919563293457, 0.13963964581489563, 0.30630630254745483, 0.2522522509098053, 0.23423422873020172, 0.2747747600078583, 0.22972972691059113, 0.46396395564079285, 0.47297295928001404, 0.4324324429035187, 0.44594594836235046, 0.13963964581489563, 0.4234234094619751, 0.45045045018196106, 0.36486485600471497, 0.4909909963607788, 0.5270270109176636, 0.5135135054588318, 0.5315315127372742, 0.522522509098053, NaN, 0.5495495200157166, 0.5675675868988037, 0.6351351141929626, 0.5630630850791931, 0.5405405163764954, 0.7162162065505981, 0.7522522807121277, 0.684684693813324, 0.662162184715271, 0.7027027010917664, 0.6576576828956604, 0.662162184715271, 0.7027027010917664, 0.6261261105537415, 0.7342342138290405, 0.7702702879905701, 0.6531531810760498, 0.6891891956329346, 0.7612612843513489, 0.8063063025474548, 0.7117117047309875, 0.7432432174682617, 0.7612612843513489, 0.792792797088623, 0.7702702879905701, 0.7747747898101807, 0.7567567825317383, 0.869369387626648, 0.9414414167404175, 0.9324324131011963, 0.8873873949050903, 0.837837815284729, 0.8918918967247009, 0.8828828930854797, 0.9144144058227539, 0.9819819927215576, 0.9144144058227539, 0.9369369149208069, 0.9729729890823364, 0.9729729890823364, 0.9954954981803894], [0.03153153136372566, 0.07207207381725311, 0.07207207381725311, 0.09009008854627609, 0.04954954981803894, 0.12612612545490265, 0.11261261254549026, 0.07207207381725311, 0.09909909963607788, 0.07207207381725311, 0.11261261254549026, 0.13963964581489563, 0.18918919563293457, 0.18018017709255219, 0.18018017709255219, 0.20720720291137695, 0.21621622145175934, 0.3468468487262726, 0.37837839126586914, 0.4369369447231293, 0.4009009003639221, 0.3558558523654938, 0.4099099040031433, 0.2657657563686371, 0.37387385964393616, 0.45045045018196106, 0.45495495200157166, 0.5405405163764954, 0.4909909963607788, 0.5180180072784424, 0.45045045018196106, NaN, 0.5090090036392212, 0.6351351141929626, 0.6036036014556885, 0.6171171069145203, 0.5540540814399719, 0.6486486196517944, 0.630630612373352, 0.6486486196517944, 0.7387387156486511, 0.7342342138290405, 0.7657657861709595, 0.684684693813324, 0.7162162065505981, 0.7882882952690125, 0.7342342138290405, 0.7297297120094299, 0.7837837934494019, 0.7432432174682617, 0.8558558821678162, 0.7882882952690125, 0.7522522807121277, 0.7477477192878723, 0.8963963985443115, 0.7522522807121277, 0.792792797088623, 0.8243243098258972, 0.8468468189239502, 0.9234234094619751, 0.8828828930854797, 0.9594594836235046, 0.8018018007278442, 0.9324324131011963, 0.9504504799842834, 0.9504504799842834, 0.9909909963607788, 0.9639639854431152, 0.9729729890823364, 0.9864864945411682, 0.9864864945411682, 1.0], [0.027027027681469917, 0.05855855718255043, 0.12162162363529205, 0.07657657563686371, 0.054054055362939835, 0.13063062727451324, 0.10360360145568848, 0.07207207381725311, 0.15315315127372742, 0.18918919563293457, 0.1756756752729416, 0.20270270109176636, 0.22972972691059113, 0.2387387454509735, 0.19819819927215576, 0.1666666716337204, 0.2702702581882477, 0.22072072327136993, 0.2657657563686371, 0.23423422873020172, 0.3558558523654938, 0.342342346906662, 0.36036035418510437, 0.342342346906662, 0.4279279410839081, 0.3198198080062866, 0.36486485600471497, 0.37387385964393616, 0.3963963985443115, 0.36936935782432556, 0.4324324429035187, 0.4909909963607788, NaN, 0.5090090036392212, 0.6396396160125732, 0.5270270109176636, 0.5810810923576355, 0.6531531810760498, 0.6216216087341309, 0.8558558821678162, 0.6441441178321838, 0.5855855941772461, 0.5495495200157166, 0.6441441178321838, 0.7477477192878723, 0.7342342138290405, 0.6261261105537415, 0.7972972989082336, 0.7567567825317383, 0.8873873949050903, 0.7657657861709595, 0.7747747898101807, 0.8828828930854797, 0.7297297120094299, 0.6801801919937134, 0.7477477192878723, 0.8873873949050903, 0.7432432174682617, 0.8198198080062866, 0.7612612843513489, 0.8243243098258972, 0.9009009003639221, 0.8243243098258972, 0.9504504799842834, 0.8648648858070374, 0.9369369149208069, 0.9594594836235046, 0.9459459185600281, 0.9594594836235046, 0.9954954981803894, 0.9954954981803894, 0.9954954981803894], [0.022522522136569023, 0.022522522136569023, 0.027027027681469917, 0.05855855718255043, 0.027027027681469917, 0.09909909963607788, 0.07657657563686371, 0.018018018454313278, 0.04954954981803894, 0.05855855718255043, 0.045045044273138046, 0.11711711436510086, 0.10360360145568848, 0.13513512909412384, 0.14864864945411682, 0.13063062727451324, 0.171171173453331, 0.23423422873020172, 0.27927929162979126, 0.3333333432674408, 0.18018017709255219, 0.28828829526901245, 0.37837839126586914, 0.23423422873020172, 0.3198198080062866, 0.36486485600471497, 0.4054054021835327, 0.44144144654273987, 0.4324324429035187, 0.45045045018196106, 0.36486485600471497, 0.36486485600471497, 0.4909909963607788, NaN, 0.5405405163764954, 0.5495495200157166, 0.46396395564079285, 0.5945945978164673, 0.6576576828956604, 0.6036036014556885, 0.6216216087341309, 0.6351351141929626, 0.7027027010917664, 0.684684693813324, 0.662162184715271, 0.7522522807121277, 0.6936936974525452, 0.707207202911377, 0.7792792916297913, 0.6801801919937134, 0.8018018007278442, 0.7792792916297913, 0.7252252101898193, 0.7342342138290405, 0.869369387626648, 0.7297297120094299, 0.7477477192878723, 0.8288288116455078, 0.869369387626648, 0.8828828930854797, 0.8873873949050903, 0.9369369149208069, 0.7882882952690125, 0.8963963985443115, 0.9054054021835327, 0.9639639854431152, 0.9819819927215576, 0.9459459185600281, 0.9414414167404175, 1.0, 1.0, 1.0], [0.03153153136372566, 0.045045044273138046, 0.11711711436510086, 0.05855855718255043, 0.04054053872823715, 0.12612612545490265, 0.06756756454706192, 0.06306306272745132, 0.13063062727451324, 0.1756756752729416, 0.15315315127372742, 0.1666666716337204, 0.22072072327136993, 0.22972972691059113, 0.18468467891216278, 0.1621621549129486, 0.21171170473098755, 0.157657653093338, 0.19369369745254517, 0.1666666716337204, 0.315315306186676, 0.29279279708862305, 0.30180180072784424, 0.30180180072784424, 0.3963963985443115, 0.29279279708862305, 0.30630630254745483, 0.3198198080062866, 0.36036035418510437, 0.36936935782432556, 0.4369369447231293, 0.3963963985443115, 0.36036035418510437, 0.45945945382118225, NaN, 0.5, 0.545045018196106, 0.6351351141929626, 0.5990990996360779, 0.7702702879905701, 0.6081081032752991, 0.5675675868988037, 0.5180180072784424, 0.6216216087341309, 0.684684693813324, 0.6711711883544922, 0.5990990996360779, 0.7612612843513489, 0.7162162065505981, 0.8423423171043396, 0.7117117047309875, 0.7477477192878723, 0.8198198080062866, 0.7342342138290405, 0.6666666865348816, 0.7387387156486511, 0.8558558821678162, 0.7297297120094299, 0.792792797088623, 0.7432432174682617, 0.8108108043670654, 0.8468468189239502, 0.8018018007278442, 0.9459459185600281, 0.8513513803482056, 0.9009009003639221, 0.9594594836235046, 0.9234234094619751, 0.9414414167404175, 0.9909909963607788, 0.9864864945411682, 0.9909909963607788], [0.009009009227156639, 0.10360360145568848, 0.10360360145568848, 0.045045044273138046, 0.054054055362939835, 0.10360360145568848, 0.07207207381725311, 0.171171173453331, 0.0810810774564743, 0.10360360145568848, 0.10360360145568848, 0.14864864945411682, 0.19369369745254517, 0.1666666716337204, 0.1621621549129486, 0.18018017709255219, 0.171171173453331, 0.2432432472705841, 0.2657657563686371, 0.28828829526901245, 0.28378379344940186, 0.3558558523654938, 0.12612612545490265, 0.30630630254745483, 0.3378378450870514, 0.3333333432674408, 0.38738739490509033, 0.37387385964393616, 0.36936935782432556, 0.38738739490509033, 0.45945945382118225, 0.38288289308547974, 0.47297295928001404, 0.45045045018196106, 0.5, NaN, 0.5180180072784424, 0.5810810923576355, 0.5630630850791931, 0.5360360145568848, 0.6036036014556885, 0.7387387156486511, 0.5630630850791931, 0.5765765905380249, 0.6036036014556885, 0.6261261105537415, 0.6576576828956604, 0.6486486196517944, 0.7342342138290405, 0.684684693813324, 0.707207202911377, 0.7477477192878723, 0.7027027010917664, 0.7387387156486511, 0.815315306186676, 0.7432432174682617, 0.7612612843513489, 0.7567567825317383, 0.8513513803482056, 0.8108108043670654, 0.8873873949050903, 0.9054054021835327, 0.7792792916297913, 0.8828828930854797, 0.9234234094619751, 0.9459459185600281, 0.9639639854431152, 0.954954981803894, 0.9639639854431152, 0.9819819927215576, 0.9819819927215576, 1.0], [0.022522522136569023, 0.03153153136372566, 0.06306306272745132, 0.05855855718255043, 0.04054053872823715, 0.07657657563686371, 0.04054053872823715, 0.04054053872823715, 0.09009008854627609, 0.0855855867266655, 0.12612612545490265, 0.10360360145568848, 0.1756756752729416, 0.15315315127372742, 0.12162162363529205, 0.20720720291137695, 0.14414414763450623, 0.3198198080062866, 0.3333333432674408, 0.2747747600078583, 0.4144144058227539, 0.09459459781646729, 0.38738739490509033, 0.36036035418510437, 0.36486485600471497, 0.3918918967247009, 0.4099099040031433, 0.3558558523654938, 0.44144144654273987, 0.5315315127372742, 0.28378379344940186, 0.44594594836235046, 0.4189189076423645, 0.5360360145568848, 0.45495495200157166, 0.4819819927215576, NaN, 0.6081081032752991, 0.6261261105537415, 0.545045018196106, 0.6666666865348816, 0.6171171069145203, 0.6351351141929626, 0.6351351141929626, 0.5900900959968567, 0.6666666865348816, 0.7387387156486511, 0.6441441178321838, 0.6441441178321838, 0.6891891956329346, 0.7657657861709595, 0.6531531810760498, 0.7297297120094299, 0.7567567825317383, 0.6936936974525452, 0.7657657861709595, 0.7162162065505981, 0.6891891956329346, 0.8423423171043396, 0.837837815284729, 0.9189189076423645, 0.8423423171043396, 0.8333333134651184, 0.8873873949050903, 0.837837815284729, 0.9054054021835327, 0.9459459185600281, 0.8783783912658691, 0.8873873949050903, 0.9819819927215576, 0.9819819927215576, 0.9954954981803894], [0.009009009227156639, 0.036036036908626556, 0.03153153136372566, 0.022522522136569023, 0.027027027681469917, 0.054054055362939835, 0.04954954981803894, 0.06306306272745132, 0.054054055362939835, 0.013513513840734959, 0.07207207381725311, 0.0855855867266655, 0.13063062727451324, 0.12162162363529205, 0.1666666716337204, 0.21621622145175934, 0.10810811072587967, 0.2432432472705841, 0.2567567527294159, 0.2387387454509735, 0.21621622145175934, 0.1621621549129486, 0.2747747600078583, 0.3243243098258972, 0.12612612545490265, 0.2702702581882477, 0.3198198080062866, 0.29279279708862305, 0.30180180072784424, 0.4099099040031433, 0.2477477490901947, 0.3513513505458832, 0.3468468487262726, 0.4054054021835327, 0.36486485600471497, 0.4189189076423645, 0.3918918967247009, NaN, 0.5900900959968567, 0.47297295928001404, 0.5630630850791931, 0.5630630850791931, 0.5540540814399719, 0.6081081032752991, 0.5630630850791931, 0.545045018196106, 0.6486486196517944, 0.5945945978164673, 0.6891891956329346, 0.5495495200157166, 0.6756756901741028, 0.7117117047309875, 0.6486486196517944, 0.6756756901741028, 0.6936936974525452, 0.684684693813324, 0.6441441178321838, 0.7477477192878723, 0.8333333134651184, 0.8333333134651184, 0.8783783912658691, 0.8198198080062866, 0.8108108043670654, 0.8288288116455078, 0.8828828930854797, 0.9414414167404175, 0.9234234094619751, 0.9594594836235046, 0.9729729890823364, 0.9729729890823364, 0.9729729890823364, 1.0], [0.027027027681469917, 0.04054053872823715, 0.054054055362939835, 0.06306306272745132, 0.06306306272745132, 0.0810810774564743, 0.11711711436510086, 0.04954954981803894, 0.04954954981803894, 0.05855855718255043, 0.06306306272745132, 0.13963964581489563, 0.06756756454706192, 0.09009008854627609, 0.13513512909412384, 0.13063062727451324, 0.18018017709255219, 0.19819819927215576, 0.21171170473098755, 0.18468467891216278, 0.1621621549129486, 0.2657657563686371, 0.2477477490901947, 0.2432432472705841, 0.28378379344940186, 0.2432432472705841, 0.29279279708862305, 0.2477477490901947, 0.2702702581882477, 0.342342346906662, 0.315315306186676, 0.36936935782432556, 0.37837839126586914, 0.342342346906662, 0.4009009003639221, 0.4369369447231293, 0.37387385964393616, 0.4099099040031433, NaN, 0.522522509098053, 0.4054054021835327, 0.45045045018196106, 0.5540540814399719, 0.704954981803894, 0.5630630850791931, 0.45045045018196106, 0.4369369447231293, 0.5945945978164673, 0.7657657861709595, 0.5990990996360779, 0.5765765905380249, 0.7882882952690125, 0.6441441178321838, 0.6891891956329346, 0.6486486196517944, 0.6936936974525452, 0.662162184715271, 0.954954981803894, 0.7432432174682617, 0.5945945978164673, 0.7342342138290405, 0.9234234094619751, 0.8063063025474548, 0.7747747898101807, 0.9729729890823364, 0.9369369149208069, 0.954954981803894, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.0045045046135783195, 0.027027027681469917, 0.0810810774564743, 0.054054055362939835, 0.013513513840734959, 0.09009008854627609, 0.027027027681469917, 0.054054055362939835, 0.0810810774564743, 0.09009008854627609, 0.09009008854627609, 0.09009008854627609, 0.18018017709255219, 0.1666666716337204, 0.11711711436510086, 0.09459459781646729, 0.15315315127372742, 0.06306306272745132, 0.0855855867266655, 0.10360360145568848, 0.2702702581882477, 0.21171170473098755, 0.2747747600078583, 0.2612612545490265, 0.3333333432674408, 0.13063062727451324, 0.13513512909412384, 0.21171170473098755, 0.1756756752729416, 0.28828829526901245, 0.3378378450870514, 0.3513513505458832, 0.14414414763450623, 0.3963963985443115, 0.22972972691059113, 0.46396395564079285, 0.45495495200157166, 0.5270270109176636, 0.477477490901947, NaN, 0.5900900959968567, 0.5135135054588318, 0.5045045018196106, 0.5, 0.6036036014556885, 0.5405405163764954, 0.6081081032752991, 0.684684693813324, 0.6666666865348816, 0.7477477192878723, 0.6576576828956604, 0.6756756901741028, 0.7387387156486511, 0.7117117047309875, 0.630630612373352, 0.7162162065505981, 0.8108108043670654, 0.7207207083702087, 0.7162162065505981, 0.7162162065505981, 0.7477477192878723, 0.837837815284729, 0.7837837934494019, 0.9369369149208069, 0.837837815284729, 0.8918918967247009, 0.9504504799842834, 0.8963963985443115, 0.9234234094619751, 1.0, 1.0, 1.0], [0.009009009227156639, 0.03153153136372566, 0.027027027681469917, 0.013513513840734959, 0.04054053872823715, 0.0855855867266655, 0.022522522136569023, 0.036036036908626556, 0.06306306272745132, 0.09459459781646729, 0.06306306272745132, 0.0855855867266655, 0.09909909963607788, 0.09009008854627609, 0.09009008854627609, 0.06756756454706192, 0.11261261254549026, 0.21171170473098755, 0.22972972691059113, 0.28828829526901245, 0.15315315127372742, 0.18018017709255219, 0.2522522509098053, 0.22522522509098053, 0.20720720291137695, 0.29729729890823364, 0.3288288414478302, 0.3288288414478302, 0.30630630254745483, 0.4009009003639221, 0.29729729890823364, 0.2612612545490265, 0.3558558523654938, 0.37837839126586914, 0.3918918967247009, 0.3963963985443115, 0.3333333432674408, 0.4369369447231293, 0.5945945978164673, 0.4099099040031433, NaN, 0.5180180072784424, 0.5, 0.6126126050949097, 0.4909909963607788, 0.5315315127372742, 0.7207207083702087, 0.5630630850791931, 0.662162184715271, 0.4954954981803894, 0.5945945978164673, 0.684684693813324, 0.6216216087341309, 0.7342342138290405, 0.7252252101898193, 0.7297297120094299, 0.5810810923576355, 0.662162184715271, 0.8018018007278442, 0.9414414167404175, 0.8918918967247009, 0.7837837934494019, 0.8333333134651184, 0.837837815284729, 0.8603603839874268, 0.8963963985443115, 0.8918918967247009, 0.8963963985443115, 0.9054054021835327, 0.9279279112815857, 0.9279279112815857, 1.0], [0.009009009227156639, 0.04954954981803894, 0.03153153136372566, 0.036036036908626556, 0.013513513840734959, 0.09909909963607788, 0.054054055362939835, 0.0855855867266655, 0.04954954981803894, 0.03153153136372566, 0.054054055362939835, 0.09909909963607788, 0.0855855867266655, 0.0855855867266655, 0.12612612545490265, 0.0810810774564743, 0.06306306272745132, 0.1756756752729416, 0.18918919563293457, 0.20720720291137695, 0.18018017709255219, 0.2477477490901947, 0.12162162363529205, 0.19819819927215576, 0.18918919563293457, 0.2477477490901947, 0.3333333432674408, 0.29279279708862305, 0.27927929162979126, 0.30630630254745483, 0.342342346906662, 0.2657657563686371, 0.4144144058227539, 0.36486485600471497, 0.4324324429035187, 0.2612612545490265, 0.38288289308547974, 0.4369369447231293, 0.5495495200157166, 0.4864864945411682, 0.4819819927215576, NaN, 0.4909909963607788, 0.5540540814399719, 0.522522509098053, 0.5, 0.5405405163764954, 0.5405405163764954, 0.6891891956329346, 0.5630630850791931, 0.6126126050949097, 0.7342342138290405, 0.5990990996360779, 0.7162162065505981, 0.7837837934494019, 0.7162162065505981, 0.6756756901741028, 0.7522522807121277, 0.7747747898101807, 0.7567567825317383, 0.8243243098258972, 0.9144144058227539, 0.792792797088623, 0.8558558821678162, 0.9189189076423645, 0.9189189076423645, 0.9594594836235046, 0.9414414167404175, 0.9639639854431152, 0.9594594836235046, 0.9594594836235046, 1.0], [0.013513513840734959, 0.09459459781646729, 0.07207207381725311, 0.13963964581489563, 0.05855855718255043, 0.06756756454706192, 0.0855855867266655, 0.09459459781646729, 0.11711711436510086, 0.0855855867266655, 0.11711711436510086, 0.13063062727451324, 0.14414414763450623, 0.1621621549129486, 0.157657653093338, 0.18468467891216278, 0.18468467891216278, 0.21171170473098755, 0.21621622145175934, 0.28828829526901245, 0.21171170473098755, 0.28378379344940186, 0.2702702581882477, 0.157657653093338, 0.2657657563686371, 0.27927929162979126, 0.37387385964393616, 0.3333333432674408, 0.28828829526901245, 0.3243243098258972, 0.3378378450870514, 0.23423422873020172, 0.45045045018196106, 0.29729729890823364, 0.4819819927215576, 0.4369369447231293, 0.36486485600471497, 0.44594594836235046, 0.44594594836235046, 0.4954954981803894, 0.5, 0.5090090036392212, NaN, 0.45045045018196106, 0.5585585832595825, 0.5, 0.5315315127372742, 0.5945945978164673, 0.5405405163764954, 0.5585585832595825, 0.5720720887184143, 0.5720720887184143, 0.6036036014556885, 0.7297297120094299, 0.6171171069145203, 0.7297297120094299, 0.5720720887184143, 0.5540540814399719, 0.7162162065505981, 0.6576576828956604, 0.7432432174682617, 0.6891891956329346, 0.7657657861709595, 0.7567567825317383, 0.815315306186676, 0.7477477192878723, 0.7522522807121277, 0.869369387626648, 0.9504504799842834, 0.9324324131011963, 0.9324324131011963, 0.9864864945411682], [0.022522522136569023, 0.054054055362939835, 0.045045044273138046, 0.05855855718255043, 0.054054055362939835, 0.0855855867266655, 0.11261261254549026, 0.04954954981803894, 0.05855855718255043, 0.04954954981803894, 0.06756756454706192, 0.13513512909412384, 0.06306306272745132, 0.0810810774564743, 0.12612612545490265, 0.12162162363529205, 0.14864864945411682, 0.1621621549129486, 0.19819819927215576, 0.18468467891216278, 0.15315315127372742, 0.2477477490901947, 0.22972972691059113, 0.21621622145175934, 0.2612612545490265, 0.21621622145175934, 0.28378379344940186, 0.2432432472705841, 0.2477477490901947, 0.342342346906662, 0.29729729890823364, 0.315315306186676, 0.3558558523654938, 0.315315306186676, 0.37837839126586914, 0.4234234094619751, 0.36486485600471497, 0.3918918967247009, 0.29504504799842834, 0.5, 0.38738739490509033, 0.44594594836235046, 0.5495495200157166, NaN, 0.5405405163764954, 0.4279279410839081, 0.4279279410839081, 0.5720720887184143, 0.6936936974525452, 0.5900900959968567, 0.5315315127372742, 0.7342342138290405, 0.6261261105537415, 0.684684693813324, 0.6351351141929626, 0.6936936974525452, 0.6441441178321838, 0.9099099040031433, 0.707207202911377, 0.5765765905380249, 0.7252252101898193, 0.9099099040031433, 0.8018018007278442, 0.7837837934494019, 0.9684684872627258, 0.9234234094619751, 0.954954981803894, 0.9954954981803894, 0.9909909963607788, 1.0, 1.0, 1.0], [0.027027027681469917, 0.022522522136569023, 0.09909909963607788, 0.036036036908626556, 0.027027027681469917, 0.09009008854627609, 0.045045044273138046, 0.05855855718255043, 0.11261261254549026, 0.12612612545490265, 0.13963964581489563, 0.11711711436510086, 0.18918919563293457, 0.20720720291137695, 0.15315315127372742, 0.14864864945411682, 0.157657653093338, 0.1621621549129486, 0.20270270109176636, 0.18918919563293457, 0.2702702581882477, 0.30630630254745483, 0.28828829526901245, 0.23423422873020172, 0.27927929162979126, 0.20720720291137695, 0.19819819927215576, 0.2522522509098053, 0.2477477490901947, 0.2747747600078583, 0.37387385964393616, 0.28378379344940186, 0.2522522509098053, 0.3378378450870514, 0.315315306186676, 0.3963963985443115, 0.4099099040031433, 0.4369369447231293, 0.4369369447231293, 0.3963963985443115, 0.5090090036392212, 0.477477490901947, 0.44144144654273987, 0.45945945382118225, NaN, 0.44144144654273987, 0.5540540814399719, 0.8063063025474548, 0.5180180072784424, 0.6351351141929626, 0.5045045018196106, 0.5540540814399719, 0.8603603839874268, 0.6981981992721558, 0.5585585832595825, 0.7297297120094299, 0.6531531810760498, 0.5360360145568848, 0.6396396160125732, 0.707207202911377, 0.7027027010917664, 0.6711711883544922, 0.8468468189239502, 0.8513513803482056, 0.7972972989082336, 0.8738738894462585, 0.8243243098258972, 0.8288288116455078, 0.869369387626648, 0.9864864945411682, 0.9864864945411682, 0.9954954981803894], [0.0, 0.009009009227156639, 0.009009009227156639, 0.022522522136569023, 0.0045045046135783195, 0.04954954981803894, 0.013513513840734959, 0.0, 0.022522522136569023, 0.013513513840734959, 0.022522522136569023, 0.04054053872823715, 0.05855855718255043, 0.05855855718255043, 0.06756756454706192, 0.036036036908626556, 0.0855855867266655, 0.04954954981803894, 0.07207207381725311, 0.018018018454313278, 0.0855855867266655, 0.1666666716337204, 0.19819819927215576, 0.12162162363529205, 0.21171170473098755, 0.027027027681469917, 0.20270270109176636, 0.0855855867266655, 0.09009008854627609, 0.2702702581882477, 0.2657657563686371, 0.21171170473098755, 0.2657657563686371, 0.2477477490901947, 0.3288288414478302, 0.37387385964393616, 0.3333333432674408, 0.45495495200157166, 0.5495495200157166, 0.45945945382118225, 0.46846845746040344, 0.5, 0.5, 0.5720720887184143, 0.5585585832595825, NaN, 0.5135135054588318, 0.6171171069145203, 0.6081081032752991, 0.5945945978164673, 0.7117117047309875, 0.6531531810760498, 0.6711711883544922, 0.7477477192878723, 0.7117117047309875, 0.7477477192878723, 0.6576576828956604, 0.7477477192878723, 0.7792792916297913, 0.7567567825317383, 0.8333333134651184, 0.8828828930854797, 0.7792792916297913, 0.8918918967247009, 0.8558558821678162, 0.9279279112815857, 0.9909909963607788, 0.9009009003639221, 0.9189189076423645, 0.9864864945411682, 0.9864864945411682, 0.9954954981803894], [0.0045045046135783195, 0.018018018454313278, 0.022522522136569023, 0.013513513840734959, 0.013513513840734959, 0.09459459781646729, 0.018018018454313278, 0.013513513840734959, 0.04054053872823715, 0.06306306272745132, 0.04954954981803894, 0.07207207381725311, 0.07207207381725311, 0.07207207381725311, 0.0810810774564743, 0.07207207381725311, 0.06306306272745132, 0.19369369745254517, 0.20720720291137695, 0.2567567527294159, 0.14414414763450623, 0.13513512909412384, 0.2522522509098053, 0.20270270109176636, 0.18468467891216278, 0.2522522509098053, 0.29729729890823364, 0.30630630254745483, 0.2612612545490265, 0.4099099040031433, 0.22972972691059113, 0.2657657563686371, 0.37387385964393616, 0.30630630254745483, 0.4009009003639221, 0.342342346906662, 0.2612612545490265, 0.3513513505458832, 0.5630630850791931, 0.3918918967247009, 0.27927929162979126, 0.45945945382118225, 0.46846845746040344, 0.5720720887184143, 0.44594594836235046, 0.4864864945411682, NaN, 0.4819819927215576, 0.6081081032752991, 0.46846845746040344, 0.5270270109176636, 0.6216216087341309, 0.5855855941772461, 0.7297297120094299, 0.6666666865348816, 0.7297297120094299, 0.545045018196106, 0.6261261105537415, 0.7027027010917664, 0.8783783912658691, 0.8108108043670654, 0.7207207083702087, 0.8288288116455078, 0.7747747898101807, 0.837837815284729, 0.8333333134651184, 0.8783783912658691, 0.8873873949050903, 0.8828828930854797, 0.9009009003639221, 0.9009009003639221, 1.0], [0.018018018454313278, 0.013513513840734959, 0.0855855867266655, 0.036036036908626556, 0.018018018454313278, 0.09459459781646729, 0.03153153136372566, 0.06756756454706192, 0.10360360145568848, 0.09909909963607788, 0.10360360145568848, 0.09459459781646729, 0.1756756752729416, 0.18018017709255219, 0.13513512909412384, 0.12162162363529205, 0.13513512909412384, 0.12612612545490265, 0.13963964581489563, 0.1621621549129486, 0.2387387454509735, 0.2477477490901947, 0.2522522509098053, 0.21621622145175934, 0.23423422873020172, 0.1756756752729416, 0.157657653093338, 0.18918919563293457, 0.18018017709255219, 0.22972972691059113, 0.3468468487262726, 0.2702702581882477, 0.20270270109176636, 0.29279279708862305, 0.2387387454509735, 0.3513513505458832, 0.3558558523654938, 0.4054054021835327, 0.4054054021835327, 0.315315306186676, 0.4369369447231293, 0.45945945382118225, 0.4054054021835327, 0.4279279410839081, 0.19369369745254517, 0.38288289308547974, 0.5180180072784424, NaN, 0.46846845746040344, 0.5720720887184143, 0.45945945382118225, 0.5135135054588318, 0.7162162065505981, 0.6801801919937134, 0.5405405163764954, 0.707207202911377, 0.6171171069145203, 0.5135135054588318, 0.5945945978164673, 0.6666666865348816, 0.6711711883544922, 0.6261261105537415, 0.837837815284729, 0.8243243098258972, 0.7702702879905701, 0.8468468189239502, 0.8018018007278442, 0.8198198080062866, 0.8468468189239502, 0.9819819927215576, 0.9819819927215576, 0.9954954981803894], [0.0045045046135783195, 0.018018018454313278, 0.022522522136569023, 0.03153153136372566, 0.009009009227156639, 0.06306306272745132, 0.045045044273138046, 0.009009009227156639, 0.018018018454313278, 0.018018018454313278, 0.03153153136372566, 0.10360360145568848, 0.027027027681469917, 0.05855855718255043, 0.09459459781646729, 0.0855855867266655, 0.07207207381725311, 0.0810810774564743, 0.11261261254549026, 0.10810811072587967, 0.06306306272745132, 0.2387387454509735, 0.14414414763450623, 0.1621621549129486, 0.21621622145175934, 0.12612612545490265, 0.19369369745254517, 0.171171173453331, 0.171171173453331, 0.2567567527294159, 0.31081080436706543, 0.21621622145175934, 0.2432432472705841, 0.22072072327136993, 0.28378379344940186, 0.2657657563686371, 0.3558558523654938, 0.31081080436706543, 0.23423422873020172, 0.3333333432674408, 0.3378378450870514, 0.31081080436706543, 0.45945945382118225, 0.30630630254745483, 0.4819819927215576, 0.3918918967247009, 0.3918918967247009, 0.5315315127372742, NaN, 0.5180180072784424, 0.46396395564079285, 0.6824324131011963, 0.5945945978164673, 0.6351351141929626, 0.5405405163764954, 0.6486486196517944, 0.5855855941772461, 0.7207207083702087, 0.6171171069145203, 0.5675675868988037, 0.6666666865348816, 0.9864864945411682, 0.7612612843513489, 0.7747747898101807, 0.9234234094619751, 0.9054054021835327, 0.9459459185600281, 0.9504504799842834, 0.9684684872627258, 1.0, 1.0, 1.0], [0.009009009227156639, 0.013513513840734959, 0.04954954981803894, 0.009009009227156639, 0.0045045046135783195, 0.06756756454706192, 0.013513513840734959, 0.04954954981803894, 0.018018018454313278, 0.022522522136569023, 0.027027027681469917, 0.045045044273138046, 0.0855855867266655, 0.0855855867266655, 0.09459459781646729, 0.03153153136372566, 0.054054055362939835, 0.036036036908626556, 0.045045044273138046, 0.027027027681469917, 0.13963964581489563, 0.1666666716337204, 0.13063062727451324, 0.18018017709255219, 0.22522522509098053, 0.07657657563686371, 0.0810810774564743, 0.11711711436510086, 0.10810811072587967, 0.2477477490901947, 0.2387387454509735, 0.2567567527294159, 0.11261261254549026, 0.3198198080062866, 0.157657653093338, 0.315315306186676, 0.31081080436706543, 0.45045045018196106, 0.4009009003639221, 0.2522522509098053, 0.5045045018196106, 0.4369369447231293, 0.44144144654273987, 0.4099099040031433, 0.36486485600471497, 0.4054054021835327, 0.5315315127372742, 0.4279279410839081, 0.4819819927215576, NaN, 0.545045018196106, 0.5135135054588318, 0.5315315127372742, 0.5945945978164673, 0.5990990996360779, 0.5945945978164673, 0.8108108043670654, 0.5900900959968567, 0.6891891956329346, 0.6486486196517944, 0.7162162065505981, 0.707207202911377, 0.6711711883544922, 0.9639639854431152, 0.7657657861709595, 0.8603603839874268, 0.8828828930854797, 0.8963963985443115, 0.9144144058227539, 0.9819819927215576, 0.9819819927215576, 0.9954954981803894], [0.0, 0.027027027681469917, 0.04954954981803894, 0.0045045046135783195, 0.0, 0.0810810774564743, 0.027027027681469917, 0.045045044273138046, 0.04054053872823715, 0.03153153136372566, 0.04954954981803894, 0.07657657563686371, 0.054054055362939835, 0.06756756454706192, 0.0810810774564743, 0.018018018454313278, 0.05855855718255043, 0.009009009227156639, 0.009009009227156639, 0.022522522136569023, 0.09459459781646729, 0.11711711436510086, 0.09459459781646729, 0.09459459781646729, 0.14414414763450623, 0.054054055362939835, 0.18018017709255219, 0.054054055362939835, 0.07207207381725311, 0.21171170473098755, 0.19369369745254517, 0.14414414763450623, 0.23423422873020172, 0.19819819927215576, 0.28828829526901245, 0.29279279708862305, 0.23423422873020172, 0.3243243098258972, 0.4234234094619751, 0.342342346906662, 0.4054054021835327, 0.38738739490509033, 0.4279279410839081, 0.46846845746040344, 0.4954954981803894, 0.28828829526901245, 0.47297295928001404, 0.5405405163764954, 0.5360360145568848, 0.45495495200157166, NaN, 0.5810810923576355, 0.5720720887184143, 0.707207202911377, 0.684684693813324, 0.707207202911377, 0.6036036014556885, 0.6576576828956604, 0.6486486196517944, 0.6171171069145203, 0.7342342138290405, 0.8423423171043396, 0.7477477192878723, 0.8018018007278442, 0.8423423171043396, 0.815315306186676, 0.9504504799842834, 0.9009009003639221, 0.9189189076423645, 0.9504504799842834, 0.9504504799842834, 0.9864864945411682], [0.0045045046135783195, 0.018018018454313278, 0.022522522136569023, 0.027027027681469917, 0.009009009227156639, 0.054054055362939835, 0.036036036908626556, 0.009009009227156639, 0.013513513840734959, 0.013513513840734959, 0.027027027681469917, 0.09009008854627609, 0.022522522136569023, 0.05855855718255043, 0.09459459781646729, 0.0810810774564743, 0.04954954981803894, 0.07657657563686371, 0.0855855867266655, 0.10360360145568848, 0.05855855718255043, 0.22972972691059113, 0.11711711436510086, 0.15315315127372742, 0.21171170473098755, 0.10810811072587967, 0.18018017709255219, 0.1621621549129486, 0.1621621549129486, 0.2567567527294159, 0.28828829526901245, 0.21171170473098755, 0.22522522509098053, 0.22072072327136993, 0.2522522509098053, 0.2522522509098053, 0.3468468487262726, 0.28828829526901245, 0.21171170473098755, 0.3243243098258972, 0.315315306186676, 0.2657657563686371, 0.4279279410839081, 0.2657657563686371, 0.44594594836235046, 0.3468468487262726, 0.37837839126586914, 0.4864864945411682, 0.3175675570964813, 0.4864864945411682, 0.4189189076423645, NaN, 0.5585585832595825, 0.5810810923576355, 0.5045045018196106, 0.5990990996360779, 0.5360360145568848, 0.6216216087341309, 0.5810810923576355, 0.5495495200157166, 0.6351351141929626, 0.7747747898101807, 0.7567567825317383, 0.7522522807121277, 0.8963963985443115, 0.9009009003639221, 0.9459459185600281, 0.9504504799842834, 0.9594594836235046, 1.0, 1.0, 1.0], [0.013513513840734959, 0.009009009227156639, 0.05855855718255043, 0.045045044273138046, 0.018018018454313278, 0.05855855718255043, 0.027027027681469917, 0.054054055362939835, 0.04954954981803894, 0.027027027681469917, 0.05855855718255043, 0.054054055362939835, 0.09009008854627609, 0.09909909963607788, 0.07207207381725311, 0.07657657563686371, 0.05855855718255043, 0.06306306272745132, 0.07657657563686371, 0.05855855718255043, 0.18468467891216278, 0.19369369745254517, 0.20720720291137695, 0.18468467891216278, 0.18018017709255219, 0.07657657563686371, 0.0855855867266655, 0.14414414763450623, 0.10360360145568848, 0.171171173453331, 0.2567567527294159, 0.2477477490901947, 0.11711711436510086, 0.2747747600078583, 0.18018017709255219, 0.29729729890823364, 0.2702702581882477, 0.3513513505458832, 0.3558558523654938, 0.2612612545490265, 0.37837839126586914, 0.4009009003639221, 0.3963963985443115, 0.37387385964393616, 0.13963964581489563, 0.3288288414478302, 0.4144144058227539, 0.28378379344940186, 0.4054054021835327, 0.46846845746040344, 0.4279279410839081, 0.44144144654273987, NaN, 0.6261261105537415, 0.46846845746040344, 0.630630612373352, 0.5270270109176636, 0.4954954981803894, 0.5540540814399719, 0.6486486196517944, 0.6126126050949097, 0.5855855941772461, 0.7837837934494019, 0.7837837934494019, 0.7657657861709595, 0.792792797088623, 0.7837837934494019, 0.792792797088623, 0.8603603839874268, 0.9864864945411682, 0.9864864945411682, 0.9909909963607788], [0.07657657563686371, 0.0855855867266655, 0.13963964581489563, 0.10360360145568848, 0.09909909963607788, 0.13063062727451324, 0.13513512909412384, 0.09909909963607788, 0.171171173453331, 0.18468467891216278, 0.18018017709255219, 0.19369369745254517, 0.171171173453331, 0.21621622145175934, 0.18018017709255219, 0.20270270109176636, 0.20270270109176636, 0.20270270109176636, 0.20720720291137695, 0.22522522509098053, 0.21171170473098755, 0.22072072327136993, 0.23423422873020172, 0.23423422873020172, 0.22072072327136993, 0.22972972691059113, 0.22072072327136993, 0.23423422873020172, 0.22972972691059113, 0.2567567527294159, 0.2387387454509735, 0.2522522509098053, 0.2702702581882477, 0.2657657563686371, 0.2657657563686371, 0.2612612545490265, 0.2432432472705841, 0.3243243098258972, 0.31081080436706543, 0.28828829526901245, 0.2657657563686371, 0.28378379344940186, 0.2702702581882477, 0.315315306186676, 0.30180180072784424, 0.2522522509098053, 0.2702702581882477, 0.3198198080062866, 0.36486485600471497, 0.4054054021835327, 0.29279279708862305, 0.4189189076423645, 0.37387385964393616, NaN, 0.36486485600471497, 0.630630612373352, 0.44594594836235046, 0.46846845746040344, 0.4054054021835327, 0.3918918967247009, 0.4054054021835327, 0.4234234094619751, 0.9819819927215576, 0.477477490901947, 0.6801801919937134, 0.5990990996360779, 0.5540540814399719, 0.7567567825317383, 0.7882882952690125, 0.8288288116455078, 0.8288288116455078, 0.9144144058227539], [0.009009009227156639, 0.03153153136372566, 0.036036036908626556, 0.027027027681469917, 0.022522522136569023, 0.036036036908626556, 0.06756756454706192, 0.03153153136372566, 0.04054053872823715, 0.03153153136372566, 0.036036036908626556, 0.09909909963607788, 0.045045044273138046, 0.054054055362939835, 0.0855855867266655, 0.07207207381725311, 0.11261261254549026, 0.12612612545490265, 0.12612612545490265, 0.15315315127372742, 0.06306306272745132, 0.1621621549129486, 0.07657657563686371, 0.027027027681469917, 0.12612612545490265, 0.1621621549129486, 0.22522522509098053, 0.1621621549129486, 0.171171173453331, 0.2657657563686371, 0.20720720291137695, 0.10360360145568848, 0.3198198080062866, 0.13063062727451324, 0.3333333432674408, 0.18468467891216278, 0.30630630254745483, 0.30630630254745483, 0.3513513505458832, 0.36936935782432556, 0.2747747600078583, 0.21621622145175934, 0.38288289308547974, 0.36486485600471497, 0.44144144654273987, 0.28828829526901245, 0.3333333432674408, 0.45945945382118225, 0.45945945382118225, 0.4009009003639221, 0.315315306186676, 0.4954954981803894, 0.5315315127372742, 0.6351351141929626, NaN, 0.6351351141929626, 0.44594594836235046, 0.5090090036392212, 0.5135135054588318, 0.5135135054588318, 0.5585585832595825, 0.6441441178321838, 0.7207207083702087, 0.6216216087341309, 0.8738738894462585, 0.7252252101898193, 0.8198198080062866, 0.9234234094619751, 0.977477490901947, 0.9234234094619751, 0.9144144058227539, 0.9909909963607788], [0.07207207381725311, 0.07657657563686371, 0.12162162363529205, 0.09909909963607788, 0.09009008854627609, 0.11711711436510086, 0.12162162363529205, 0.09909909963607788, 0.171171173453331, 0.18018017709255219, 0.1756756752729416, 0.18918919563293457, 0.1756756752729416, 0.20720720291137695, 0.1756756752729416, 0.19369369745254517, 0.19369369745254517, 0.20270270109176636, 0.19819819927215576, 0.22072072327136993, 0.21171170473098755, 0.21171170473098755, 0.22072072327136993, 0.22972972691059113, 0.21171170473098755, 0.21621622145175934, 0.21621622145175934, 0.22522522509098053, 0.22972972691059113, 0.2477477490901947, 0.22972972691059113, 0.2477477490901947, 0.2522522509098053, 0.2702702581882477, 0.2612612545490265, 0.2567567527294159, 0.23423422873020172, 0.315315306186676, 0.30630630254745483, 0.28378379344940186, 0.2702702581882477, 0.28378379344940186, 0.2702702581882477, 0.30630630254745483, 0.2702702581882477, 0.2522522509098053, 0.2702702581882477, 0.29279279708862305, 0.3513513505458832, 0.4054054021835327, 0.29279279708862305, 0.4009009003639221, 0.36936935782432556, 0.36936935782432556, 0.36486485600471497, NaN, 0.44594594836235046, 0.44594594836235046, 0.4054054021835327, 0.38738739490509033, 0.4009009003639221, 0.4054054021835327, 0.9684684872627258, 0.477477490901947, 0.6801801919937134, 0.5990990996360779, 0.5540540814399719, 0.7522522807121277, 0.7882882952690125, 0.8288288116455078, 0.8288288116455078, 0.8963963985443115], [0.0045045046135783195, 0.009009009227156639, 0.018018018454313278, 0.0045045046135783195, 0.0045045046135783195, 0.06756756454706192, 0.013513513840734959, 0.027027027681469917, 0.013513513840734959, 0.013513513840734959, 0.022522522136569023, 0.036036036908626556, 0.027027027681469917, 0.036036036908626556, 0.0855855867266655, 0.036036036908626556, 0.027027027681469917, 0.018018018454313278, 0.027027027681469917, 0.027027027681469917, 0.06756756454706192, 0.14864864945411682, 0.10360360145568848, 0.14864864945411682, 0.1666666716337204, 0.04054053872823715, 0.05855855718255043, 0.09009008854627609, 0.07657657563686371, 0.23423422873020172, 0.22522522509098053, 0.20720720291137695, 0.11261261254549026, 0.2522522509098053, 0.14414414763450623, 0.2387387454509735, 0.28378379344940186, 0.3558558523654938, 0.3378378450870514, 0.18918919563293457, 0.4189189076423645, 0.3243243098258972, 0.4279279410839081, 0.3558558523654938, 0.3468468487262726, 0.342342346906662, 0.45495495200157166, 0.38288289308547974, 0.4144144058227539, 0.18918919563293457, 0.3963963985443115, 0.46396395564079285, 0.47297295928001404, 0.5540540814399719, 0.5540540814399719, 0.5540540814399719, NaN, 0.545045018196106, 0.6396396160125732, 0.6171171069145203, 0.6711711883544922, 0.6711711883544922, 0.6396396160125732, 0.8265765905380249, 0.7207207083702087, 0.792792797088623, 0.8468468189239502, 0.8738738894462585, 0.8963963985443115, 0.977477490901947, 0.977477490901947, 0.9909909963607788], [0.018018018454313278, 0.036036036908626556, 0.03153153136372566, 0.04954954981803894, 0.036036036908626556, 0.04054053872823715, 0.045045044273138046, 0.022522522136569023, 0.036036036908626556, 0.022522522136569023, 0.04054053872823715, 0.09009008854627609, 0.045045044273138046, 0.06306306272745132, 0.09909909963607788, 0.0810810774564743, 0.09459459781646729, 0.10810811072587967, 0.12612612545490265, 0.11261261254549026, 0.09009008854627609, 0.18468467891216278, 0.10360360145568848, 0.07657657563686371, 0.12612612545490265, 0.13963964581489563, 0.1666666716337204, 0.171171173453331, 0.15315315127372742, 0.18018017709255219, 0.2432432472705841, 0.1756756752729416, 0.2567567527294159, 0.171171173453331, 0.2702702581882477, 0.2432432472705841, 0.31081080436706543, 0.2522522509098053, 0.045045044273138046, 0.27927929162979126, 0.3378378450870514, 0.2477477490901947, 0.44594594836235046, 0.09009008854627609, 0.46396395564079285, 0.2522522509098053, 0.37387385964393616, 0.4864864945411682, 0.27927929162979126, 0.4099099040031433, 0.342342346906662, 0.37837839126586914, 0.5045045018196106, 0.5315315127372742, 0.4909909963607788, 0.5540540814399719, 0.45495495200157166, NaN, 0.47297295928001404, 0.4054054021835327, 0.477477490901947, 0.6396396160125732, 0.6981981992721558, 0.6351351141929626, 0.8783783912658691, 0.7657657861709595, 0.8423423171043396, 0.9234234094619751, 0.9819819927215576, 0.9954954981803894, 0.9954954981803894, 1.0], [0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.018018018454313278, 0.04054053872823715, 0.0, 0.013513513840734959, 0.0, 0.0, 0.0045045046135783195, 0.022522522136569023, 0.013513513840734959, 0.018018018454313278, 0.027027027681469917, 0.0045045046135783195, 0.022522522136569023, 0.06306306272745132, 0.06306306272745132, 0.027027027681469917, 0.018018018454313278, 0.05855855718255043, 0.07207207381725311, 0.12162162363529205, 0.036036036908626556, 0.054054055362939835, 0.0810810774564743, 0.0810810774564743, 0.06306306272745132, 0.11711711436510086, 0.13063062727451324, 0.15315315127372742, 0.18018017709255219, 0.13063062727451324, 0.20720720291137695, 0.14864864945411682, 0.157657653093338, 0.1666666716337204, 0.2567567527294159, 0.28378379344940186, 0.19819819927215576, 0.22522522509098053, 0.28378379344940186, 0.29279279708862305, 0.36036035418510437, 0.22072072327136993, 0.29729729890823364, 0.4054054021835327, 0.38288289308547974, 0.31081080436706543, 0.3513513505458832, 0.4189189076423645, 0.44594594836235046, 0.5945945978164673, 0.4864864945411682, 0.5945945978164673, 0.36036035418510437, 0.5270270109176636, NaN, 0.5180180072784424, 0.6216216087341309, 0.5720720887184143, 0.6891891956329346, 0.6396396160125732, 0.7342342138290405, 0.9504504799842834, 0.8288288116455078, 0.8018018007278442, 0.8873873949050903, 0.977477490901947, 0.977477490901947, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.04054053872823715, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.009009009227156639, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.018018018454313278, 0.07207207381725311, 0.07657657563686371, 0.10360360145568848, 0.027027027681469917, 0.03153153136372566, 0.09009008854627609, 0.06306306272745132, 0.054054055362939835, 0.11261261254549026, 0.19819819927215576, 0.13063062727451324, 0.10360360145568848, 0.19369369745254517, 0.05855855718255043, 0.07657657563686371, 0.2387387454509735, 0.11711711436510086, 0.2567567527294159, 0.18918919563293457, 0.1621621549129486, 0.1666666716337204, 0.4054054021835327, 0.28378379344940186, 0.05855855718255043, 0.2432432472705841, 0.342342346906662, 0.4234234094619751, 0.29279279708862305, 0.2432432472705841, 0.12162162363529205, 0.3333333432674408, 0.4324324429035187, 0.3513513505458832, 0.38288289308547974, 0.45045045018196106, 0.3513513505458832, 0.6081081032752991, 0.4864864945411682, 0.6126126050949097, 0.38288289308547974, 0.5945945978164673, 0.4819819927215576, NaN, 0.5405405163764954, 0.6441441178321838, 0.707207202911377, 0.5405405163764954, 0.7747747898101807, 0.6351351141929626, 0.7657657861709595, 0.8513513803482056, 0.8603603839874268, 0.8873873949050903, 0.8873873949050903, 0.9864864945411682], [0.0, 0.0, 0.0, 0.0, 0.009009009227156639, 0.04054053872823715, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.013513513840734959, 0.009009009227156639, 0.018018018454313278, 0.0, 0.009009009227156639, 0.03153153136372566, 0.04054053872823715, 0.009009009227156639, 0.022522522136569023, 0.018018018454313278, 0.06306306272745132, 0.09459459781646729, 0.03153153136372566, 0.018018018454313278, 0.06756756454706192, 0.045045044273138046, 0.03153153136372566, 0.10810811072587967, 0.06756756454706192, 0.11711711436510086, 0.1756756752729416, 0.11261261254549026, 0.18918919563293457, 0.11261261254549026, 0.0810810774564743, 0.12162162363529205, 0.2657657563686371, 0.2522522509098053, 0.10810811072587967, 0.1756756752729416, 0.2567567527294159, 0.2747747600078583, 0.29729729890823364, 0.1666666716337204, 0.18918919563293457, 0.3288288414478302, 0.3333333432674408, 0.28378379344940186, 0.2657657563686371, 0.36486485600471497, 0.38738739490509033, 0.5945945978164673, 0.44144144654273987, 0.5990990996360779, 0.3288288414478302, 0.522522509098053, 0.37837839126586914, 0.45945945382118225, NaN, 0.5585585832595825, 0.6936936974525452, 0.5720720887184143, 0.707207202911377, 0.8198198080062866, 0.792792797088623, 0.7702702879905701, 0.8738738894462585, 0.954954981803894, 0.9504504799842834, 1.0], [0.0045045046135783195, 0.013513513840734959, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.027027027681469917, 0.009009009227156639, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.009009009227156639, 0.036036036908626556, 0.0045045046135783195, 0.013513513840734959, 0.054054055362939835, 0.022522522136569023, 0.009009009227156639, 0.013513513840734959, 0.022522522136569023, 0.027027027681469917, 0.022522522136569023, 0.06306306272745132, 0.036036036908626556, 0.022522522136569023, 0.10360360145568848, 0.018018018454313278, 0.05855855718255043, 0.04954954981803894, 0.036036036908626556, 0.1621621549129486, 0.11261261254549026, 0.04054053872823715, 0.09909909963607788, 0.06306306272745132, 0.15315315127372742, 0.09459459781646729, 0.157657653093338, 0.18018017709255219, 0.07657657563686371, 0.1621621549129486, 0.21621622145175934, 0.0855855867266655, 0.31081080436706543, 0.09009008854627609, 0.3288288414478302, 0.11711711436510086, 0.27927929162979126, 0.37387385964393616, 0.013513513840734959, 0.29279279708862305, 0.157657653093338, 0.22522522509098053, 0.4144144058227539, 0.5765765905380249, 0.3558558523654938, 0.5945945978164673, 0.3288288414478302, 0.36036035418510437, 0.4279279410839081, 0.3558558523654938, 0.44144144654273987, NaN, 0.6936936974525452, 0.5855855941772461, 0.7972972989082336, 0.6801801919937134, 0.7567567825317383, 0.8738738894462585, 0.9189189076423645, 0.954954981803894, 0.954954981803894, 0.9819819927215576], [0.05855855718255043, 0.07207207381725311, 0.07207207381725311, 0.06756756454706192, 0.06306306272745132, 0.06756756454706192, 0.0810810774564743, 0.07657657563686371, 0.0855855867266655, 0.09909909963607788, 0.0810810774564743, 0.10810811072587967, 0.07657657563686371, 0.11261261254549026, 0.10360360145568848, 0.12162162363529205, 0.12612612545490265, 0.13063062727451324, 0.13963964581489563, 0.15315315127372742, 0.13963964581489563, 0.13513512909412384, 0.18018017709255219, 0.18018017709255219, 0.14414414763450623, 0.1621621549129486, 0.13963964581489563, 0.171171173453331, 0.1621621549129486, 0.12162162363529205, 0.1621621549129486, 0.19819819927215576, 0.1756756752729416, 0.21171170473098755, 0.19819819927215576, 0.22072072327136993, 0.1666666716337204, 0.18918919563293457, 0.19369369745254517, 0.21621622145175934, 0.1666666716337204, 0.20720720291137695, 0.23423422873020172, 0.19819819927215576, 0.15315315127372742, 0.22072072327136993, 0.171171173453331, 0.1621621549129486, 0.2387387454509735, 0.3288288414478302, 0.2522522509098053, 0.2432432472705841, 0.21621622145175934, 0.018018018454313278, 0.27927929162979126, 0.03153153136372566, 0.36036035418510437, 0.30180180072784424, 0.31081080436706543, 0.29279279708862305, 0.30630630254745483, 0.30630630254745483, NaN, 0.37837839126586914, 0.5855855941772461, 0.5675675868988037, 0.44144144654273987, 0.6531531810760498, 0.7117117047309875, 0.8243243098258972, 0.8243243098258972, 0.8333333134651184], [0.0, 0.0, 0.0, 0.0, 0.0, 0.04054053872823715, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.009009009227156639, 0.036036036908626556, 0.009009009227156639, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0, 0.018018018454313278, 0.06306306272745132, 0.045045044273138046, 0.045045044273138046, 0.06306306272745132, 0.013513513840734959, 0.0, 0.013513513840734959, 0.013513513840734959, 0.09459459781646729, 0.10810811072587967, 0.06756756454706192, 0.04954954981803894, 0.10360360145568848, 0.054054055362939835, 0.11711711436510086, 0.11261261254549026, 0.171171173453331, 0.22522522509098053, 0.06306306272745132, 0.1621621549129486, 0.14414414763450623, 0.2432432472705841, 0.21621622145175934, 0.14864864945411682, 0.10810811072587967, 0.22522522509098053, 0.1756756752729416, 0.22522522509098053, 0.036036036908626556, 0.19819819927215576, 0.2477477490901947, 0.21621622145175934, 0.522522509098053, 0.37837839126586914, 0.522522509098053, 0.1734234243631363, 0.36486485600471497, 0.36036035418510437, 0.45945945382118225, 0.4279279410839081, 0.4144144058227539, 0.6216216087341309, NaN, 0.5900900959968567, 0.5945945978164673, 0.6486486196517944, 0.7117117047309875, 0.8513513803482056, 0.9144144058227539, 0.9144144058227539, 0.9729729890823364], [0.013513513840734959, 0.018018018454313278, 0.03153153136372566, 0.022522522136569023, 0.013513513840734959, 0.018018018454313278, 0.03153153136372566, 0.022522522136569023, 0.018018018454313278, 0.03153153136372566, 0.018018018454313278, 0.06756756454706192, 0.018018018454313278, 0.03153153136372566, 0.07657657563686371, 0.06306306272745132, 0.04954954981803894, 0.054054055362939835, 0.06306306272745132, 0.07657657563686371, 0.04054053872823715, 0.09009008854627609, 0.045045044273138046, 0.03153153136372566, 0.036036036908626556, 0.07207207381725311, 0.10810811072587967, 0.09009008854627609, 0.0855855867266655, 0.0810810774564743, 0.11711711436510086, 0.04954954981803894, 0.13513512909412384, 0.09459459781646729, 0.14864864945411682, 0.07657657563686371, 0.1621621549129486, 0.11711711436510086, 0.027027027681469917, 0.1621621549129486, 0.13963964581489563, 0.0810810774564743, 0.18468467891216278, 0.03153153136372566, 0.20270270109176636, 0.14414414763450623, 0.1621621549129486, 0.22972972691059113, 0.07657657563686371, 0.23423422873020172, 0.157657653093338, 0.10360360145568848, 0.23423422873020172, 0.3198198080062866, 0.12612612545490265, 0.3198198080062866, 0.27927929162979126, 0.12162162363529205, 0.2657657563686371, 0.22522522509098053, 0.29279279708862305, 0.20270270109176636, 0.4144144058227539, 0.4099099040031433, NaN, 0.44144144654273987, 0.3333333432674408, 0.9324324131011963, 0.9729729890823364, 0.8783783912658691, 0.8783783912658691, 0.9729729890823364], [0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.04054053872823715, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.009009009227156639, 0.0045045046135783195, 0.013513513840734959, 0.009009009227156639, 0.009009009227156639, 0.036036036908626556, 0.018018018454313278, 0.027027027681469917, 0.0, 0.009009009227156639, 0.013513513840734959, 0.022522522136569023, 0.018018018454313278, 0.06306306272745132, 0.0855855867266655, 0.04954954981803894, 0.06306306272745132, 0.036036036908626556, 0.09909909963607788, 0.054054055362939835, 0.09459459781646729, 0.05855855718255043, 0.06306306272745132, 0.10810811072587967, 0.10360360145568848, 0.0810810774564743, 0.2522522509098053, 0.07657657563686371, 0.12612612545490265, 0.07207207381725311, 0.1666666716337204, 0.15315315127372742, 0.09459459781646729, 0.13963964581489563, 0.18468467891216278, 0.09909909963607788, 0.20720720291137695, 0.4009009003639221, 0.2747747600078583, 0.4009009003639221, 0.20720720291137695, 0.23423422873020172, 0.04954954981803894, 0.36486485600471497, 0.18018017709255219, 0.3198198080062866, 0.4324324429035187, 0.4054054021835327, 0.5585585832595825, NaN, 0.45045045018196106, 0.7207207083702087, 0.7972972989082336, 0.9414414167404175, 0.9414414167404175, 0.9954954981803894], [0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.013513513840734959, 0.009009009227156639, 0.0, 0.018018018454313278, 0.045045044273138046, 0.0, 0.0, 0.0, 0.0, 0.09459459781646729, 0.018018018454313278, 0.009009009227156639, 0.04054053872823715, 0.018018018454313278, 0.04054053872823715, 0.036036036908626556, 0.054054055362939835, 0.07657657563686371, 0.045045044273138046, 0.04954954981803894, 0.10810811072587967, 0.04054053872823715, 0.2477477490901947, 0.045045044273138046, 0.1756756752729416, 0.009009009227156639, 0.12162162363529205, 0.19819819927215576, 0.054054055362939835, 0.11711711436510086, 0.04954954981803894, 0.054054055362939835, 0.21621622145175934, 0.44594594836235046, 0.18018017709255219, 0.44594594836235046, 0.15315315127372742, 0.157657653093338, 0.171171173453331, 0.23423422873020172, 0.20720720291137695, 0.2432432472705841, 0.5585585832595825, 0.3513513505458832, 0.6666666865348816, 0.5495495200157166, NaN, 0.7612612843513489, 0.792792797088623, 0.8243243098258972, 0.8198198080062866, 0.8783783912658691], [0.0045045046135783195, 0.009009009227156639, 0.018018018454313278, 0.009009009227156639, 0.013513513840734959, 0.009009009227156639, 0.022522522136569023, 0.013513513840734959, 0.009009009227156639, 0.018018018454313278, 0.009009009227156639, 0.027027027681469917, 0.009009009227156639, 0.027027027681469917, 0.04954954981803894, 0.036036036908626556, 0.036036036908626556, 0.03153153136372566, 0.036036036908626556, 0.054054055362939835, 0.03153153136372566, 0.06306306272745132, 0.022522522136569023, 0.018018018454313278, 0.018018018454313278, 0.036036036908626556, 0.04054053872823715, 0.05855855718255043, 0.045045044273138046, 0.013513513840734959, 0.0855855867266655, 0.036036036908626556, 0.054054055362939835, 0.054054055362939835, 0.07657657563686371, 0.045045044273138046, 0.12162162363529205, 0.04054053872823715, 0.009009009227156639, 0.10360360145568848, 0.10360360145568848, 0.05855855718255043, 0.13063062727451324, 0.0045045046135783195, 0.171171173453331, 0.09909909963607788, 0.11261261254549026, 0.18018017709255219, 0.04954954981803894, 0.10360360145568848, 0.09909909963607788, 0.04954954981803894, 0.20720720291137695, 0.2432432472705841, 0.07657657563686371, 0.2477477490901947, 0.12612612545490265, 0.07657657563686371, 0.19819819927215576, 0.14864864945411682, 0.22972972691059113, 0.12612612545490265, 0.3468468487262726, 0.28828829526901245, 0.06756756454706192, 0.27927929162979126, 0.2387387454509735, NaN, 0.9189189076423645, 0.8603603839874268, 0.8603603839874268, 0.9144144058227539], [0.009009009227156639, 0.018018018454313278, 0.009009009227156639, 0.009009009227156639, 0.009009009227156639, 0.009009009227156639, 0.013513513840734959, 0.018018018454313278, 0.009009009227156639, 0.009009009227156639, 0.009009009227156639, 0.03153153136372566, 0.009009009227156639, 0.022522522136569023, 0.04054053872823715, 0.027027027681469917, 0.022522522136569023, 0.027027027681469917, 0.03153153136372566, 0.03153153136372566, 0.013513513840734959, 0.054054055362939835, 0.018018018454313278, 0.018018018454313278, 0.009009009227156639, 0.036036036908626556, 0.03153153136372566, 0.054054055362939835, 0.036036036908626556, 0.013513513840734959, 0.06306306272745132, 0.027027027681469917, 0.04054053872823715, 0.05855855718255043, 0.05855855718255043, 0.036036036908626556, 0.11261261254549026, 0.027027027681469917, 0.009009009227156639, 0.07657657563686371, 0.09459459781646729, 0.036036036908626556, 0.04954954981803894, 0.009009009227156639, 0.13063062727451324, 0.0810810774564743, 0.11711711436510086, 0.15315315127372742, 0.03153153136372566, 0.0855855867266655, 0.0810810774564743, 0.04054053872823715, 0.13963964581489563, 0.21171170473098755, 0.022522522136569023, 0.21171170473098755, 0.10360360145568848, 0.018018018454313278, 0.11261261254549026, 0.13963964581489563, 0.12612612545490265, 0.0810810774564743, 0.28828829526901245, 0.14864864945411682, 0.027027027681469917, 0.20270270109176636, 0.20720720291137695, 0.0810810774564743, NaN, 0.630630612373352, 0.6351351141929626, 0.6891891956329346], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.027027027681469917, 0.013513513840734959, 0.0045045046135783195, 0.0, 0.009009009227156639, 0.018018018454313278, 0.018018018454313278, 0.027027027681469917, 0.0, 0.0, 0.07207207381725311, 0.04054053872823715, 0.06756756454706192, 0.0, 0.013513513840734959, 0.013513513840734959, 0.09909909963607788, 0.018018018454313278, 0.0, 0.018018018454313278, 0.04954954981803894, 0.0, 0.013513513840734959, 0.171171173453331, 0.07657657563686371, 0.171171173453331, 0.022522522136569023, 0.0045045046135783195, 0.022522522136569023, 0.11261261254549026, 0.045045044273138046, 0.045045044273138046, 0.1756756752729416, 0.0855855867266655, 0.12162162363529205, 0.05855855718255043, 0.1756756752729416, 0.13963964581489563, 0.36936935782432556, NaN, 0.6013513803482056, 0.8243243098258972], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.027027027681469917, 0.013513513840734959, 0.0045045046135783195, 0.0, 0.013513513840734959, 0.018018018454313278, 0.018018018454313278, 0.027027027681469917, 0.0, 0.0, 0.07207207381725311, 0.04054053872823715, 0.06756756454706192, 0.0, 0.013513513840734959, 0.013513513840734959, 0.09909909963607788, 0.018018018454313278, 0.0, 0.018018018454313278, 0.04954954981803894, 0.0, 0.013513513840734959, 0.171171173453331, 0.0855855867266655, 0.171171173453331, 0.022522522136569023, 0.0045045046135783195, 0.022522522136569023, 0.11261261254549026, 0.04954954981803894, 0.045045044273138046, 0.1756756752729416, 0.0855855867266655, 0.12162162363529205, 0.05855855718255043, 0.18018017709255219, 0.13963964581489563, 0.36486485600471497, 0.3986486494541168, NaN, 0.7972972989082336], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.013513513840734959, 0.0, 0.0045045046135783195, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.013513513840734959, 0.0, 0.009009009227156639, 0.0855855867266655, 0.009009009227156639, 0.10360360145568848, 0.009009009227156639, 0.0, 0.0, 0.013513513840734959, 0.0, 0.018018018454313278, 0.1666666716337204, 0.027027027681469917, 0.027027027681469917, 0.0045045046135783195, 0.12162162363529205, 0.0855855867266655, 0.31081080436706543, 0.1756756752729416, 0.20270270109176636, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9459459505},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.8905595751},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8837076543},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8641669847},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8641669849},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8587108263},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8582667196},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.8474812843},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.827559958},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8037051164},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.8012308084},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7832445137},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7531404637},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7396903923},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7392780092},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7294759565},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7286511881},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.673582034},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6504885183},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6494734176},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6404644079},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6396396388},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6371018917},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6327242738},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6284101029},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6029691655},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5811445233},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.580129424},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5762593571},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5761324682},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.5623651825},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5594467701},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5362263659},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5129425191},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5034259583},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5033625178},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5005075466},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4546377348},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4473099868},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.439474683},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4383326983},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4362390577},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.428372035},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4280230925},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4229158745},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.403819312},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4015987802},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.3820581129},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3690204285},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3596624794},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3555386361},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3438967137},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.3288288289},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3281943894},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3268620704},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3182337263},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.314649157},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.3125872349},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.2791523916},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2757264316},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2484456308},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2329653616},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2279533058},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2020365433},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1813856105},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.161273949},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.1471894434},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1272046701},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0865372408},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0532610077},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0502791525},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0205557671}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_regression/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..33332ab4446ab284ad7b1cbb32eb48aa8bad3f40
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_all/winrate_matrix.csv
@@ -0,0 +1,73 @@
+,TabFM (default),Nori-30M (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),Nori (default),TabDPT (tuned + ensembled),RealMLP (tuned + ensembled),TabDPT (tuned),RealTabPFN-2.5 (tuned),TabDPT (default),TabDPT-Turbo (default),RealTabPFN-2.5 (default),"AutoGluon 1.4 (best, 4h)",RealMLP (tuned),CatBoost (tuned + ensembled),CatBoost (tuned),LightGBM (tuned + ensembled),iLTM (tuned + ensembled),TabM (tuned + ensembled),ModernNCA (tuned + ensembled),xRFM (tuned + ensembled),TabSwift (default),XGBoost (tuned + ensembled),CatBoost (default),LightGBM (tuned),XGBoost (tuned),SAP-RPT-OSS (default),TabM (tuned),xRFM (tuned),ChimeraBoost (tuned + ensembled),iLTM (tuned),ChimeraBoost (tuned),ModernNCA (tuned),TabM (default),RealMLP (default),ExtraTrees (tuned + ensembled),ChimeraBoost (default),TorchMLP (tuned + ensembled),ModernNCA (default),LimiX (default),ExtraTrees (tuned),EBM (tuned + ensembled),LightGBM (default),TorchMLP (tuned),EBM (tuned),RandomForest (tuned + ensembled),PerpetualBooster (tuned + ensembled),XGBoost (default),RandomForest (tuned),EBM (default),TabSTAR (tuned + ensembled),xRFM (default),TabSTAR (tuned),PerpetualBooster (tuned),ExtraTrees (default),FastaiMLP (tuned + ensembled),TorchMLP (default),FastaiMLP (tuned),RandomForest (default),TabSTAR (default),PerpetualBooster (default),KNN (tuned + ensembled),FastaiMLP (default),iLTM (default),KNN (tuned),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,0.6126126,0.6981982,0.7297297,0.8153153,0.7432432,0.7927928,0.7972973,0.8378378,0.9009009,0.9054054,0.8153153,0.8873874,0.954955,0.9324324,0.9594595,0.9459459,0.963964,0.9774775,0.972973,0.963964,0.954955,0.9594595,0.9504505,0.8963964,0.963964,0.9684685,0.9774775,0.972973,0.9324324,0.9684685,0.9684685,0.972973,0.9774775,0.9684685,0.990991,0.9774775,0.990991,0.972973,0.9954955,0.990991,0.990991,0.9864865,0.9774775,0.972973,1.0,0.9954955,0.981982,0.9954955,0.990991,1.0,0.9954955,0.9864865,0.9234234,0.990991,0.9279279,0.9954955,0.981982,1.0,1.0,1.0,0.9954955,0.9414414,1.0,0.9864865,1.0,1.0,0.9954955,0.990991,1.0,1.0,1.0
+Nori-30M (default),0.3873874,,0.4774775,0.6531532,0.6801802,0.545045,0.6216216,0.7297297,0.6306306,0.7207207,0.7162162,0.6981982,0.7387387,0.7747748,0.7927928,0.8288288,0.7927928,0.8918919,0.9189189,0.9144144,0.9009009,0.9234234,0.7927928,0.8783784,0.8693694,0.9279279,0.9189189,0.9234234,0.9369369,0.8918919,0.9459459,0.9279279,0.9414414,0.9774775,0.954955,0.8963964,0.9684685,0.963964,0.9594595,0.972973,0.9684685,0.9504505,0.9054054,0.9459459,0.9774775,0.990991,0.981982,0.9864865,0.981982,0.9864865,0.972973,0.981982,0.990991,0.9144144,0.9684685,0.9234234,0.990991,0.963964,0.9954955,1.0,1.0,0.9864865,0.9279279,1.0,0.981982,1.0,0.9954955,0.990991,0.981982,1.0,1.0,1.0
+TabPFN-3 (default),0.3018018,0.5225225,,0.6441441,0.6396396,0.6126126,0.6216216,0.6396396,0.6486486,0.7927928,0.6891892,0.7522523,0.7882883,0.8783784,0.8783784,0.8423423,0.8243243,0.8558559,0.8603604,0.8873874,0.8963964,0.8918919,0.8558559,0.8783784,0.9369369,0.8873874,0.8738739,0.9099099,0.8828829,0.8738739,0.9234234,0.9279279,0.8783784,0.972973,0.8828829,0.8963964,0.9369369,0.9684685,0.9459459,0.9189189,0.972973,0.9684685,0.9279279,0.954955,0.9009009,0.990991,0.9774775,0.9144144,0.9774775,0.9504505,0.9504505,0.9774775,0.9414414,0.8603604,0.963964,0.8783784,0.981982,0.9684685,1.0,1.0,1.0,0.9954955,0.9279279,1.0,0.9684685,1.0,1.0,0.981982,0.990991,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.27027026,0.34684685,0.35585585,,0.5630631,0.44144145,0.45945945,0.5225225,0.5990991,0.7297297,0.6216216,0.6081081,0.6891892,0.7432432,0.7297297,0.8603604,0.8108108,0.8648649,0.8873874,0.8873874,0.9054054,0.9099099,0.8018018,0.8648649,0.8648649,0.9234234,0.9009009,0.9324324,0.9369369,0.7972973,0.9234234,0.9099099,0.9234234,0.9414414,0.9414414,0.954955,0.9414414,0.9774775,0.9369369,0.9459459,0.9864865,0.963964,0.8603604,0.9414414,0.963964,0.9774775,0.9864865,0.963964,0.9684685,0.990991,0.9954955,0.972973,0.954955,0.8963964,0.972973,0.9009009,0.9954955,0.9504505,0.9954955,1.0,1.0,1.0,0.9324324,1.0,0.9774775,1.0,1.0,0.990991,0.990991,1.0,1.0,1.0
+TabPFN-2.6 (default),0.18468468,0.3198198,0.36036035,0.43693694,,0.3873874,0.47297296,0.47297296,0.5945946,0.6576577,0.6351351,0.6261261,0.6801802,0.6846847,0.7927928,0.8018018,0.8018018,0.8783784,0.9009009,0.8738739,0.8738739,0.9144144,0.8108108,0.8783784,0.8558559,0.9369369,0.9324324,0.9054054,0.9414414,0.8468468,0.9234234,0.9504505,0.9459459,0.972973,0.9594595,0.9459459,0.9594595,0.972973,0.9369369,0.9864865,0.9594595,0.9864865,0.9414414,0.9459459,0.972973,0.9954955,0.9864865,0.981982,0.990991,0.9954955,1.0,0.990991,0.981982,0.9009009,0.9774775,0.9099099,0.9954955,0.963964,0.981982,1.0,0.990991,0.9954955,0.9369369,1.0,0.9864865,0.9954955,1.0,0.9864865,0.990991,1.0,1.0,1.0
+TabICLv2 (default),0.25675675,0.45495495,0.3873874,0.5585586,0.6126126,,0.5855856,0.5675676,0.6846847,0.7522523,0.6936937,0.6981982,0.7207207,0.7837838,0.8333333,0.8468468,0.8153153,0.8783784,0.8738739,0.8738739,0.8738739,0.9144144,0.8513514,0.8603604,0.8333333,0.8828829,0.8828829,0.8918919,0.8738739,0.9099099,0.9189189,0.8738739,0.8693694,0.9009009,0.8738739,0.8963964,0.9234234,0.9459459,0.9189189,0.9099099,0.9144144,0.9009009,0.9324324,0.9144144,0.9099099,0.9504505,0.9054054,0.9054054,0.9369369,0.9324324,0.9189189,0.9459459,0.9414414,0.8693694,0.963964,0.8828829,0.9324324,0.9594595,0.9594595,0.9594595,0.9594595,0.972973,0.9324324,0.9594595,0.981982,0.9594595,1.0,0.990991,0.990991,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.2072072,0.3783784,0.3783784,0.5405405,0.527027,0.4144144,,0.536036,0.6081081,0.6981982,0.6351351,0.8198198,0.6846847,0.7477477,0.8828829,0.7972973,0.7792793,0.8063063,0.8468468,0.8513514,0.8378378,0.9054054,0.7882883,0.8468468,0.8603604,0.8918919,0.8918919,0.8693694,0.8963964,0.8153153,0.9324324,0.8873874,0.8963964,0.9234234,0.9324324,0.9279279,0.9594595,0.9504505,0.8828829,0.972973,0.9774775,0.9459459,0.9144144,0.8873874,0.954955,0.9864865,0.981982,0.9684685,0.954955,0.9864865,0.972973,0.963964,0.972973,0.8648649,0.9324324,0.8783784,0.9864865,0.954955,1.0,1.0,1.0,0.990991,0.9189189,1.0,0.9684685,1.0,1.0,0.9774775,0.9864865,1.0,1.0,1.0
+Nori (default),0.2027027,0.27027026,0.36036035,0.4774775,0.527027,0.43243244,0.46396396,,0.5540541,0.5855856,0.5900901,0.545045,0.6351351,0.6891892,0.7657658,0.7522523,0.7477477,0.9189189,0.9144144,0.9099099,0.8513514,0.8423423,0.7612613,0.8423423,0.7702703,0.8963964,0.9144144,0.9144144,0.9099099,0.8468468,0.8963964,0.9279279,0.9279279,0.981982,0.9369369,0.8288288,0.9594595,0.9369369,0.9504505,0.9459459,0.963964,0.9144144,0.9054054,0.9504505,0.9414414,1.0,0.9864865,0.9324324,0.990991,0.9504505,0.954955,0.990991,0.9459459,0.9009009,0.9684685,0.9009009,0.972973,0.9774775,0.9864865,1.0,0.990991,0.9954955,0.9234234,0.9954955,0.9774775,1.0,1.0,0.9864865,0.981982,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.16216215,0.36936936,0.35135135,0.4009009,0.4054054,0.3153153,0.3918919,0.44594595,,0.5495495,0.6846847,0.5405405,0.7747748,0.7792793,0.6936937,0.7747748,0.6126126,0.7882883,0.8063063,0.8153153,0.8828829,0.7927928,0.7657658,0.8693694,0.7612613,0.8558559,0.8513514,0.8378378,0.8783784,0.7972973,0.8513514,0.9009009,0.8468468,0.9504505,0.8693694,0.9189189,0.9099099,0.9459459,0.9504505,0.9189189,0.9369369,0.9504505,0.8828829,0.9414414,0.8873874,0.9774775,0.9594595,0.8963964,0.981982,0.981982,0.9594595,0.9864865,0.9504505,0.8288288,0.9594595,0.8288288,0.9864865,0.963964,1.0,1.0,1.0,1.0,0.9144144,1.0,0.981982,1.0,1.0,0.990991,0.990991,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0990991,0.2792793,0.2072072,0.27027026,0.34234235,0.24774775,0.3018018,0.4144144,0.45045045,,0.5135135,0.4189189,0.5855856,0.6126126,0.6036036,0.6891892,0.7387387,0.7837838,0.7792793,0.7837838,0.8558559,0.7792793,0.7522523,0.8783784,0.7207207,0.8648649,0.8288288,0.8513514,0.8693694,0.7927928,0.8738739,0.9279279,0.8108108,0.9414414,0.8243243,0.8963964,0.9144144,0.9864865,0.9414414,0.9099099,0.9054054,0.9684685,0.9144144,0.9504505,0.8738739,0.9864865,0.9369369,0.9009009,0.981982,0.9774775,0.9684685,0.9864865,0.972973,0.8153153,0.9684685,0.8198198,0.9864865,0.9774775,1.0,0.9954955,0.9954955,0.9954955,0.9009009,1.0,0.9684685,1.0,1.0,0.981982,0.990991,1.0,1.0,1.0
+TabDPT (tuned),0.0945946,0.2837838,0.3108108,0.3783784,0.36486486,0.3063063,0.36486486,0.4099099,0.3153153,0.4864865,,0.509009,0.6756757,0.6846847,0.5945946,0.7297297,0.5765766,0.7612613,0.7792793,0.7837838,0.8333333,0.7432432,0.7522523,0.8378378,0.7162162,0.8333333,0.8153153,0.8423423,0.8468468,0.7837838,0.8108108,0.8873874,0.8243243,0.954955,0.8468468,0.8963964,0.8738739,0.9279279,0.9369369,0.9099099,0.9369369,0.9459459,0.8828829,0.9324324,0.8603604,0.9774775,0.9504505,0.8963964,0.9684685,0.972973,0.9504505,0.972973,0.9414414,0.8198198,0.963964,0.8243243,0.9774775,0.9594595,0.9954955,1.0,1.0,0.990991,0.9189189,1.0,0.981982,1.0,1.0,0.990991,0.990991,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.18468468,0.3018018,0.24774775,0.3918919,0.37387386,0.3018018,0.18018018,0.45495495,0.45945945,0.5810811,0.490991,,0.5765766,0.6351351,0.6869369,0.6441441,0.6531532,0.6981982,0.7297297,0.7207207,0.7432432,0.7927928,0.6846847,0.7927928,0.7702703,0.8108108,0.7702703,0.7792793,0.8108108,0.7297297,0.8603604,0.8603604,0.7972973,0.8828829,0.8333333,0.8513514,0.8963964,0.9144144,0.8603604,0.9099099,0.9144144,0.9009009,0.8693694,0.8648649,0.8828829,0.9594595,0.9279279,0.9054054,0.8963964,0.954955,0.9234234,0.9099099,0.9459459,0.8063063,0.9009009,0.8108108,0.963964,0.9099099,0.9774775,0.9954955,1.0,0.963964,0.8918919,1.0,0.9324324,0.9954955,1.0,0.972973,0.9684685,1.0,1.0,1.0
+TabDPT (default),0.11261261,0.26126125,0.2117117,0.3108108,0.3198198,0.2792793,0.3153153,0.36486486,0.22522523,0.4144144,0.3243243,0.4234234,,0.5765766,0.5045045,0.5855856,0.47297296,0.7027027,0.7162162,0.6936937,0.7792793,0.6081081,0.6891892,0.7297297,0.6576577,0.7477477,0.7522523,0.7702703,0.7747748,0.7297297,0.6936937,0.8108108,0.7702703,0.8963964,0.7792793,0.8063063,0.8243243,0.8693694,0.9324324,0.8198198,0.9009009,0.9144144,0.8558559,0.9369369,0.8108108,0.9414414,0.9279279,0.8243243,0.972973,0.9144144,0.9459459,0.9774775,0.9099099,0.8288288,0.954955,0.8243243,0.972973,0.954955,0.9864865,0.990991,0.9864865,0.9954955,0.9234234,1.0,0.981982,1.0,1.0,0.990991,0.990991,1.0,1.0,1.0
+TabDPT-Turbo (default),0.045045044,0.22522523,0.12162162,0.25675675,0.3153153,0.21621622,0.25225225,0.3108108,0.22072072,0.3873874,0.3153153,0.36486486,0.4234234,,0.45495495,0.6306306,0.490991,0.6981982,0.7117117,0.6801802,0.7252252,0.6756757,0.6981982,0.7387387,0.6576577,0.7522523,0.7432432,0.7837838,0.7702703,0.7207207,0.7477477,0.8198198,0.7612613,0.8648649,0.7702703,0.8333333,0.8468468,0.8783784,0.9099099,0.8333333,0.9099099,0.9144144,0.8378378,0.9189189,0.7927928,0.9414414,0.9279279,0.8198198,0.9414414,0.9144144,0.9324324,0.9414414,0.9009009,0.7837838,0.9459459,0.7927928,0.963964,0.9369369,0.981982,0.990991,0.990991,0.9864865,0.8873874,0.990991,0.9684685,1.0,1.0,0.972973,0.9774775,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.067567565,0.2072072,0.12162162,0.27027026,0.2072072,0.16666667,0.117117114,0.23423423,0.3063063,0.3963964,0.4054054,0.31306306,0.4954955,0.545045,,0.6171171,0.545045,0.7162162,0.7612613,0.6981982,0.7027027,0.7207207,0.6576577,0.7342342,0.6891892,0.7657658,0.7837838,0.7972973,0.7927928,0.7117117,0.7657658,0.8198198,0.8018018,0.8513514,0.8153153,0.8378378,0.8783784,0.8333333,0.8648649,0.8828829,0.9099099,0.8738739,0.8423423,0.8738739,0.8468468,0.9324324,0.9189189,0.8648649,0.9054054,0.9054054,0.9189189,0.9054054,0.9279279,0.8198198,0.9144144,0.8243243,0.9144144,0.9009009,0.972973,1.0,0.981982,0.9459459,0.8963964,0.963964,0.9234234,0.9954955,1.0,0.9504505,0.9594595,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.04054054,0.17117117,0.15765765,0.13963965,0.1981982,0.15315315,0.2027027,0.24774775,0.22522523,0.3108108,0.27027026,0.35585585,0.4144144,0.36936936,0.3828829,,0.518018,0.6711712,0.6711712,0.7837838,0.6891892,0.6261261,0.6711712,0.6936937,0.6081081,0.8288288,0.7747748,0.8603604,0.8513514,0.6756757,0.7252252,0.7927928,0.8333333,0.8693694,0.8378378,0.8198198,0.7927928,0.7837838,0.8693694,0.9054054,0.9324324,0.9189189,0.8153153,0.8783784,0.8513514,0.963964,0.9279279,0.8783784,0.9144144,0.9684685,0.981982,0.9189189,0.9234234,0.7972973,0.9279279,0.8063063,0.963964,0.9189189,0.9954955,0.9954955,1.0,0.9774775,0.8783784,0.990991,0.9369369,1.0,1.0,0.963964,0.972973,1.0,1.0,1.0
+RealMLP (tuned),0.054054055,0.2072072,0.17567568,0.1891892,0.1981982,0.18468468,0.22072072,0.25225225,0.3873874,0.26126125,0.4234234,0.34684685,0.527027,0.509009,0.45495495,0.481982,,0.6531532,0.6621622,0.6576577,0.6981982,0.6891892,0.5990991,0.6711712,0.5855856,0.7477477,0.6981982,0.7657658,0.7567568,0.7297297,0.7702703,0.7837838,0.7297297,0.8288288,0.7882883,0.8288288,0.8558559,0.8918919,0.8198198,0.8468468,0.8873874,0.9369369,0.8153153,0.8513514,0.8423423,0.9144144,0.9369369,0.8648649,0.9279279,0.9459459,0.9414414,0.9504505,0.9414414,0.7972973,0.8873874,0.8063063,0.972973,0.9054054,0.9774775,0.981982,0.990991,0.990991,0.8738739,0.990991,0.9504505,0.990991,1.0,0.963964,0.9774775,0.9954955,0.9954955,0.9954955
+CatBoost (tuned + ensembled),0.036036037,0.10810811,0.14414415,0.13513513,0.12162162,0.12162162,0.1936937,0.08108108,0.2117117,0.21621622,0.23873875,0.3018018,0.2972973,0.3018018,0.2837838,0.32882884,0.34684685,,0.6396396,0.6036036,0.5135135,0.42792794,0.5990991,0.4864865,0.5405405,0.7657658,0.7972973,0.7297297,0.7837838,0.6036036,0.536036,0.6531532,0.7792793,0.7657658,0.8423423,0.7567568,0.6801802,0.7567568,0.8018018,0.9369369,0.7882883,0.8243243,0.7882883,0.8378378,0.8378378,0.9504505,0.8063063,0.8738739,0.9189189,0.963964,0.990991,0.9234234,0.9369369,0.7972973,0.8738739,0.7972973,0.981982,0.8918919,0.9369369,0.9279279,0.9684685,0.9864865,0.8693694,1.0,0.9459459,0.9954955,1.0,0.9684685,0.972973,1.0,1.0,1.0
+CatBoost (tuned),0.022522522,0.08108108,0.13963965,0.11261261,0.0990991,0.12612613,0.15315315,0.08558559,0.1936937,0.22072072,0.22072072,0.27027026,0.2837838,0.2882883,0.23873875,0.32882884,0.33783785,0.36036035,,0.5225225,0.490991,0.4189189,0.5630631,0.45495495,0.4954955,0.6756757,0.6846847,0.6846847,0.7297297,0.6036036,0.527027,0.6216216,0.7342342,0.7207207,0.8063063,0.7342342,0.6666667,0.7432432,0.7882883,0.9144144,0.7702703,0.8108108,0.7837838,0.8018018,0.7972973,0.9279279,0.7927928,0.8603604,0.8873874,0.954955,0.990991,0.9144144,0.9234234,0.7927928,0.8738739,0.8018018,0.972973,0.8738739,0.9369369,0.9234234,0.9594595,0.9774775,0.8603604,0.9954955,0.9369369,0.9864865,1.0,0.963964,0.9684685,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.027027028,0.08558559,0.11261261,0.11261261,0.12612613,0.12612613,0.14864865,0.09009009,0.18468468,0.21621622,0.21621622,0.2792793,0.3063063,0.3198198,0.3018018,0.21621622,0.34234235,0.3963964,0.4774775,,0.490991,0.46846846,0.5,0.481982,0.5405405,0.6756757,0.7117117,0.7792793,0.7252252,0.490991,0.5675676,0.5630631,0.7657658,0.6666667,0.8333333,0.7117117,0.7252252,0.7612613,0.8153153,0.8963964,0.7117117,0.7927928,0.7117117,0.8153153,0.8108108,0.981982,0.7432432,0.8378378,0.8918919,0.972973,0.9774775,0.8963964,0.9414414,0.7747748,0.8468468,0.7792793,0.972973,0.8873874,0.972973,0.8963964,0.990991,0.972973,0.8468468,1.0,0.9234234,0.990991,1.0,0.9459459,0.9684685,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.036036037,0.0990991,0.1036036,0.0945946,0.12612613,0.12612613,0.16216215,0.14864865,0.117117114,0.14414415,0.16666667,0.25675675,0.22072072,0.27477476,0.2972973,0.3108108,0.3018018,0.4864865,0.509009,0.509009,,0.44144145,0.5,0.47297296,0.5225225,0.6351351,0.5765766,0.6306306,0.6666667,0.5855856,0.5540541,0.5990991,0.6441441,0.8198198,0.6846847,0.7162162,0.5855856,0.7837838,0.8378378,0.7297297,0.8468468,0.8198198,0.7882883,0.8468468,0.7297297,0.9144144,0.8558559,0.7612613,0.9369369,0.8603604,0.9054054,0.9414414,0.8153153,0.7882883,0.9369369,0.7882883,0.9324324,0.9099099,0.981982,0.972973,0.9774775,0.9774775,0.8603604,0.981982,0.9594595,0.990991,0.9864865,0.9684685,0.9864865,0.990991,0.990991,0.990991
+TabM (tuned + ensembled),0.045045044,0.076576576,0.10810811,0.09009009,0.08558559,0.08558559,0.0945946,0.15765765,0.2072072,0.22072072,0.25675675,0.2072072,0.3918919,0.3243243,0.2792793,0.37387386,0.3108108,0.5720721,0.5810811,0.5315315,0.5585586,,0.509009,0.5495495,0.43243244,0.6261261,0.6486486,0.6396396,0.6711712,0.5990991,0.8603604,0.6441441,0.6576577,0.7117117,0.7072072,0.6441441,0.9054054,0.8378378,0.7342342,0.7882883,0.8198198,0.7522523,0.7162162,0.7522523,0.6936937,0.8333333,0.8648649,0.7522523,0.7612613,0.8333333,0.8828829,0.7702703,0.8063063,0.7792793,0.8378378,0.7882883,0.8513514,0.8153153,0.9414414,0.9684685,0.981982,0.9369369,0.8648649,0.9369369,0.9099099,0.963964,0.990991,0.9369369,0.9459459,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.04054054,0.2072072,0.14414415,0.1981982,0.1891892,0.14864865,0.2117117,0.23873875,0.23423423,0.24774775,0.24774775,0.3153153,0.3108108,0.3018018,0.34234235,0.32882884,0.4009009,0.4009009,0.43693694,0.5,0.5,0.490991,,0.4774775,0.527027,0.5900901,0.6126126,0.5810811,0.5990991,0.490991,0.5765766,0.5900901,0.6396396,0.6216216,0.6981982,0.8738739,0.6126126,0.7252252,0.7522523,0.7252252,0.7477477,0.8783784,0.7297297,0.7702703,0.7117117,0.8018018,0.7477477,0.7477477,0.8558559,0.8693694,0.9054054,0.8828829,0.7927928,0.7657658,0.9234234,0.7792793,0.8963964,0.8963964,0.9279279,0.9099099,0.9369369,0.963964,0.8198198,0.954955,0.954955,0.981982,1.0,0.9774775,0.981982,0.9954955,0.9954955,1.0
+xRFM (tuned + ensembled),0.04954955,0.12162162,0.12162162,0.13513513,0.12162162,0.13963965,0.15315315,0.15765765,0.13063063,0.12162162,0.16216215,0.2072072,0.27027026,0.26126125,0.26576576,0.3063063,0.32882884,0.5135135,0.545045,0.518018,0.527027,0.45045045,0.5225225,,0.509009,0.6081081,0.6171171,0.6351351,0.6666667,0.5630631,0.5495495,0.7342342,0.6576577,0.7657658,0.6981982,0.6936937,0.6396396,0.6756757,0.7567568,0.7387387,0.7747748,0.8018018,0.8423423,0.7837838,0.7657658,0.8783784,0.7972973,0.7837838,0.8378378,0.8198198,0.9054054,0.8468468,0.8153153,0.7657658,0.972973,0.7702703,0.8513514,0.9234234,0.8783784,0.9369369,0.9054054,0.9774775,0.8198198,0.954955,0.9684685,0.972973,0.981982,0.981982,0.981982,0.9954955,0.9954955,0.9954955
+TabSwift (default),0.1036036,0.13063063,0.06306306,0.13513513,0.14414415,0.16666667,0.13963965,0.22972973,0.23873875,0.2792793,0.2837838,0.22972973,0.34234235,0.34234235,0.3108108,0.3918919,0.4144144,0.45945945,0.5045045,0.45945945,0.4774775,0.5675676,0.47297296,0.490991,,0.518018,0.5225225,0.5720721,0.5810811,0.5765766,0.6351351,0.6261261,0.5720721,0.6801802,0.6036036,0.6621622,0.6351351,0.8738739,0.7162162,0.6666667,0.7927928,0.8108108,0.7342342,0.7387387,0.7207207,0.7882883,0.8153153,0.7657658,0.7837838,0.7747748,0.8558559,0.7882883,0.8198198,0.7792793,0.8738739,0.7882883,0.8333333,0.8738739,0.963964,0.9459459,0.9684685,0.8963964,0.8558559,0.9369369,0.963964,1.0,0.954955,0.981982,0.990991,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.036036037,0.072072074,0.11261261,0.076576576,0.06306306,0.117117114,0.10810811,0.1036036,0.14414415,0.13513513,0.16666667,0.1891892,0.25225225,0.24774775,0.23423423,0.17117117,0.25225225,0.23423423,0.3243243,0.3243243,0.36486486,0.37387386,0.4099099,0.3918919,0.481982,,0.5405405,0.509009,0.6396396,0.4864865,0.509009,0.5495495,0.6801802,0.6351351,0.7072072,0.6666667,0.6081081,0.7297297,0.7567568,0.8693694,0.7027027,0.7522523,0.7207207,0.7837838,0.7927928,0.972973,0.7477477,0.8243243,0.8738739,0.9234234,0.9459459,0.8918919,0.9234234,0.7702703,0.8378378,0.7837838,0.9594595,0.8603604,0.9459459,0.8873874,0.981982,0.981982,0.8378378,0.9864865,0.9279279,0.990991,1.0,0.963964,0.963964,1.0,1.0,1.0
+CatBoost (default),0.03153153,0.08108108,0.12612613,0.0990991,0.067567565,0.117117114,0.10810811,0.08558559,0.14864865,0.17117117,0.18468468,0.22972973,0.24774775,0.25675675,0.21621622,0.22522523,0.3018018,0.2027027,0.3153153,0.2882883,0.4234234,0.35135135,0.3873874,0.3828829,0.4774775,0.45945945,,0.46846846,0.509009,0.4234234,0.47297296,0.545045,0.6351351,0.5945946,0.6936937,0.6126126,0.5900901,0.6801802,0.7072072,0.8648649,0.6711712,0.6666667,0.6261261,0.7162162,0.8018018,0.7972973,0.7027027,0.8423423,0.8063063,0.9189189,0.8198198,0.8198198,0.9144144,0.7792793,0.7747748,0.7837838,0.9414414,0.8333333,0.9189189,0.8018018,0.9324324,0.9414414,0.8603604,1.0,0.8918919,0.9864865,1.0,0.9594595,0.9684685,1.0,1.0,1.0
+LightGBM (tuned),0.022522522,0.076576576,0.09009009,0.067567565,0.0945946,0.10810811,0.13063063,0.08558559,0.16216215,0.14864865,0.15765765,0.22072072,0.22972973,0.21621622,0.2027027,0.13963965,0.23423423,0.27027026,0.3153153,0.22072072,0.36936936,0.36036035,0.4189189,0.36486486,0.42792794,0.490991,0.5315315,,0.5495495,0.45495495,0.4864865,0.45945945,0.6261261,0.5585586,0.6801802,0.6261261,0.6441441,0.7072072,0.7522523,0.7882883,0.6711712,0.7072072,0.6666667,0.7567568,0.7477477,0.9144144,0.6936937,0.8108108,0.8288288,0.8828829,0.9459459,0.8378378,0.8558559,0.7657658,0.8378378,0.7747748,0.9099099,0.8288288,0.9189189,0.8693694,0.954955,0.9504505,0.8288288,0.9864865,0.9099099,0.9774775,1.0,0.9414414,0.9459459,0.990991,0.990991,0.9954955
+XGBoost (tuned),0.027027028,0.06306306,0.117117114,0.06306306,0.058558557,0.12612613,0.1036036,0.09009009,0.12162162,0.13063063,0.15315315,0.1891892,0.22522523,0.22972973,0.2072072,0.14864865,0.24324325,0.21621622,0.27027026,0.27477476,0.33333334,0.32882884,0.4009009,0.33333334,0.4189189,0.36036035,0.490991,0.45045045,,0.481982,0.46846846,0.509009,0.6036036,0.5675676,0.6396396,0.6306306,0.5585586,0.6981982,0.7297297,0.8243243,0.6936937,0.7207207,0.7117117,0.7522523,0.7522523,0.9099099,0.7387387,0.8198198,0.8288288,0.8918919,0.9279279,0.8378378,0.8963964,0.7702703,0.8288288,0.7702703,0.9234234,0.8468468,0.9369369,0.8963964,0.9684685,0.963964,0.8378378,0.9864865,0.9144144,0.981982,1.0,0.954955,0.963964,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.067567565,0.10810811,0.12612613,0.2027027,0.15315315,0.09009009,0.18468468,0.15315315,0.2027027,0.2072072,0.21621622,0.27027026,0.27027026,0.2792793,0.2882883,0.3243243,0.27027026,0.3963964,0.3963964,0.509009,0.4144144,0.4009009,0.509009,0.43693694,0.4234234,0.5135135,0.5765766,0.545045,0.518018,,0.4774775,0.481982,0.6306306,0.5495495,0.6306306,0.6126126,0.46846846,0.5900901,0.6576577,0.7117117,0.5990991,0.6936937,0.6756757,0.6576577,0.7252252,0.7297297,0.5900901,0.7702703,0.7432432,0.7522523,0.7882883,0.7432432,0.8288288,0.7432432,0.7342342,0.7522523,0.7657658,0.8198198,0.8828829,0.8063063,0.8918919,0.8378378,0.8783784,0.9054054,0.9189189,0.9369369,0.9054054,0.9864865,0.9864865,0.9954955,0.9954955,1.0
+TabM (tuned),0.03153153,0.054054055,0.076576576,0.076576576,0.076576576,0.08108108,0.067567565,0.1036036,0.14864865,0.12612613,0.1891892,0.13963965,0.3063063,0.25225225,0.23423423,0.27477476,0.22972973,0.46396396,0.47297296,0.43243244,0.44594595,0.13963965,0.4234234,0.45045045,0.36486486,0.490991,0.527027,0.5135135,0.5315315,0.5225225,,0.5495495,0.5675676,0.6351351,0.5630631,0.5405405,0.7162162,0.7522523,0.6846847,0.6621622,0.7027027,0.6576577,0.6621622,0.7027027,0.6261261,0.7342342,0.7702703,0.6531532,0.6891892,0.7612613,0.8063063,0.7117117,0.7432432,0.7612613,0.7927928,0.7702703,0.7747748,0.7567568,0.8693694,0.9414414,0.9324324,0.8873874,0.8378378,0.8918919,0.8828829,0.9144144,0.981982,0.9144144,0.9369369,0.972973,0.972973,0.9954955
+xRFM (tuned),0.03153153,0.072072074,0.072072074,0.09009009,0.04954955,0.12612613,0.11261261,0.072072074,0.0990991,0.072072074,0.11261261,0.13963965,0.1891892,0.18018018,0.18018018,0.2072072,0.21621622,0.34684685,0.3783784,0.43693694,0.4009009,0.35585585,0.4099099,0.26576576,0.37387386,0.45045045,0.45495495,0.5405405,0.490991,0.518018,0.45045045,,0.509009,0.6351351,0.6036036,0.6171171,0.5540541,0.6486486,0.6306306,0.6486486,0.7387387,0.7342342,0.7657658,0.6846847,0.7162162,0.7882883,0.7342342,0.7297297,0.7837838,0.7432432,0.8558559,0.7882883,0.7522523,0.7477477,0.8963964,0.7522523,0.7927928,0.8243243,0.8468468,0.9234234,0.8828829,0.9594595,0.8018018,0.9324324,0.9504505,0.9504505,0.990991,0.963964,0.972973,0.9864865,0.9864865,1.0
+ChimeraBoost (tuned + ensembled),0.027027028,0.058558557,0.12162162,0.076576576,0.054054055,0.13063063,0.1036036,0.072072074,0.15315315,0.1891892,0.17567568,0.2027027,0.22972973,0.23873875,0.1981982,0.16666667,0.27027026,0.22072072,0.26576576,0.23423423,0.35585585,0.34234235,0.36036035,0.34234235,0.42792794,0.3198198,0.36486486,0.37387386,0.3963964,0.36936936,0.43243244,0.490991,,0.509009,0.6396396,0.527027,0.5810811,0.6531532,0.6216216,0.8558559,0.6441441,0.5855856,0.5495495,0.6441441,0.7477477,0.7342342,0.6261261,0.7972973,0.7567568,0.8873874,0.7657658,0.7747748,0.8828829,0.7297297,0.6801802,0.7477477,0.8873874,0.7432432,0.8198198,0.7612613,0.8243243,0.9009009,0.8243243,0.9504505,0.8648649,0.9369369,0.9594595,0.9459459,0.9594595,0.9954955,0.9954955,0.9954955
+iLTM (tuned),0.022522522,0.022522522,0.027027028,0.058558557,0.027027028,0.0990991,0.076576576,0.018018018,0.04954955,0.058558557,0.045045044,0.117117114,0.1036036,0.13513513,0.14864865,0.13063063,0.17117117,0.23423423,0.2792793,0.33333334,0.18018018,0.2882883,0.3783784,0.23423423,0.3198198,0.36486486,0.4054054,0.44144145,0.43243244,0.45045045,0.36486486,0.36486486,0.490991,,0.5405405,0.5495495,0.46396396,0.5945946,0.6576577,0.6036036,0.6216216,0.6351351,0.7027027,0.6846847,0.6621622,0.7522523,0.6936937,0.7072072,0.7792793,0.6801802,0.8018018,0.7792793,0.7252252,0.7342342,0.8693694,0.7297297,0.7477477,0.8288288,0.8693694,0.8828829,0.8873874,0.9369369,0.7882883,0.8963964,0.9054054,0.963964,0.981982,0.9459459,0.9414414,1.0,1.0,1.0
+ChimeraBoost (tuned),0.03153153,0.045045044,0.117117114,0.058558557,0.04054054,0.12612613,0.067567565,0.06306306,0.13063063,0.17567568,0.15315315,0.16666667,0.22072072,0.22972973,0.18468468,0.16216215,0.2117117,0.15765765,0.1936937,0.16666667,0.3153153,0.2927928,0.3018018,0.3018018,0.3963964,0.2927928,0.3063063,0.3198198,0.36036035,0.36936936,0.43693694,0.3963964,0.36036035,0.45945945,,0.5,0.545045,0.6351351,0.5990991,0.7702703,0.6081081,0.5675676,0.518018,0.6216216,0.6846847,0.6711712,0.5990991,0.7612613,0.7162162,0.8423423,0.7117117,0.7477477,0.8198198,0.7342342,0.6666667,0.7387387,0.8558559,0.7297297,0.7927928,0.7432432,0.8108108,0.8468468,0.8018018,0.9459459,0.8513514,0.9009009,0.9594595,0.9234234,0.9414414,0.990991,0.9864865,0.990991
+ModernNCA (tuned),0.009009009,0.1036036,0.1036036,0.045045044,0.054054055,0.1036036,0.072072074,0.17117117,0.08108108,0.1036036,0.1036036,0.14864865,0.1936937,0.16666667,0.16216215,0.18018018,0.17117117,0.24324325,0.26576576,0.2882883,0.2837838,0.35585585,0.12612613,0.3063063,0.33783785,0.33333334,0.3873874,0.37387386,0.36936936,0.3873874,0.45945945,0.3828829,0.47297296,0.45045045,0.5,,0.518018,0.5810811,0.5630631,0.536036,0.6036036,0.7387387,0.5630631,0.5765766,0.6036036,0.6261261,0.6576577,0.6486486,0.7342342,0.6846847,0.7072072,0.7477477,0.7027027,0.7387387,0.8153153,0.7432432,0.7612613,0.7567568,0.8513514,0.8108108,0.8873874,0.9054054,0.7792793,0.8828829,0.9234234,0.9459459,0.963964,0.954955,0.963964,0.981982,0.981982,1.0
+TabM (default),0.022522522,0.03153153,0.06306306,0.058558557,0.04054054,0.076576576,0.04054054,0.04054054,0.09009009,0.08558559,0.12612613,0.1036036,0.17567568,0.15315315,0.12162162,0.2072072,0.14414415,0.3198198,0.33333334,0.27477476,0.4144144,0.0945946,0.3873874,0.36036035,0.36486486,0.3918919,0.4099099,0.35585585,0.44144145,0.5315315,0.2837838,0.44594595,0.4189189,0.536036,0.45495495,0.481982,,0.6081081,0.6261261,0.545045,0.6666667,0.6171171,0.6351351,0.6351351,0.5900901,0.6666667,0.7387387,0.6441441,0.6441441,0.6891892,0.7657658,0.6531532,0.7297297,0.7567568,0.6936937,0.7657658,0.7162162,0.6891892,0.8423423,0.8378378,0.9189189,0.8423423,0.8333333,0.8873874,0.8378378,0.9054054,0.9459459,0.8783784,0.8873874,0.981982,0.981982,0.9954955
+RealMLP (default),0.009009009,0.036036037,0.03153153,0.022522522,0.027027028,0.054054055,0.04954955,0.06306306,0.054054055,0.013513514,0.072072074,0.08558559,0.13063063,0.12162162,0.16666667,0.21621622,0.10810811,0.24324325,0.25675675,0.23873875,0.21621622,0.16216215,0.27477476,0.3243243,0.12612613,0.27027026,0.3198198,0.2927928,0.3018018,0.4099099,0.24774775,0.35135135,0.34684685,0.4054054,0.36486486,0.4189189,0.3918919,,0.5900901,0.47297296,0.5630631,0.5630631,0.5540541,0.6081081,0.5630631,0.545045,0.6486486,0.5945946,0.6891892,0.5495495,0.6756757,0.7117117,0.6486486,0.6756757,0.6936937,0.6846847,0.6441441,0.7477477,0.8333333,0.8333333,0.8783784,0.8198198,0.8108108,0.8288288,0.8828829,0.9414414,0.9234234,0.9594595,0.972973,0.972973,0.972973,1.0
+ExtraTrees (tuned + ensembled),0.027027028,0.04054054,0.054054055,0.06306306,0.06306306,0.08108108,0.117117114,0.04954955,0.04954955,0.058558557,0.06306306,0.13963965,0.067567565,0.09009009,0.13513513,0.13063063,0.18018018,0.1981982,0.2117117,0.18468468,0.16216215,0.26576576,0.24774775,0.24324325,0.2837838,0.24324325,0.2927928,0.24774775,0.27027026,0.34234235,0.3153153,0.36936936,0.3783784,0.34234235,0.4009009,0.43693694,0.37387386,0.4099099,,0.5225225,0.4054054,0.45045045,0.5540541,0.704955,0.5630631,0.45045045,0.43693694,0.5945946,0.7657658,0.5990991,0.5765766,0.7882883,0.6441441,0.6891892,0.6486486,0.6936937,0.6621622,0.954955,0.7432432,0.5945946,0.7342342,0.9234234,0.8063063,0.7747748,0.972973,0.9369369,0.954955,0.990991,0.990991,1.0,1.0,1.0
+ChimeraBoost (default),0.0045045046,0.027027028,0.08108108,0.054054055,0.013513514,0.09009009,0.027027028,0.054054055,0.08108108,0.09009009,0.09009009,0.09009009,0.18018018,0.16666667,0.117117114,0.0945946,0.15315315,0.06306306,0.08558559,0.1036036,0.27027026,0.2117117,0.27477476,0.26126125,0.33333334,0.13063063,0.13513513,0.2117117,0.17567568,0.2882883,0.33783785,0.35135135,0.14414415,0.3963964,0.22972973,0.46396396,0.45495495,0.527027,0.4774775,,0.5900901,0.5135135,0.5045045,0.5,0.6036036,0.5405405,0.6081081,0.6846847,0.6666667,0.7477477,0.6576577,0.6756757,0.7387387,0.7117117,0.6306306,0.7162162,0.8108108,0.7207207,0.7162162,0.7162162,0.7477477,0.8378378,0.7837838,0.9369369,0.8378378,0.8918919,0.9504505,0.8963964,0.9234234,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.009009009,0.03153153,0.027027028,0.013513514,0.04054054,0.08558559,0.022522522,0.036036037,0.06306306,0.0945946,0.06306306,0.08558559,0.0990991,0.09009009,0.09009009,0.067567565,0.11261261,0.2117117,0.22972973,0.2882883,0.15315315,0.18018018,0.25225225,0.22522523,0.2072072,0.2972973,0.32882884,0.32882884,0.3063063,0.4009009,0.2972973,0.26126125,0.35585585,0.3783784,0.3918919,0.3963964,0.33333334,0.43693694,0.5945946,0.4099099,,0.518018,0.5,0.6126126,0.490991,0.5315315,0.7207207,0.5630631,0.6621622,0.4954955,0.5945946,0.6846847,0.6216216,0.7342342,0.7252252,0.7297297,0.5810811,0.6621622,0.8018018,0.9414414,0.8918919,0.7837838,0.8333333,0.8378378,0.8603604,0.8963964,0.8918919,0.8963964,0.9054054,0.9279279,0.9279279,1.0
+ModernNCA (default),0.009009009,0.04954955,0.03153153,0.036036037,0.013513514,0.0990991,0.054054055,0.08558559,0.04954955,0.03153153,0.054054055,0.0990991,0.08558559,0.08558559,0.12612613,0.08108108,0.06306306,0.17567568,0.1891892,0.2072072,0.18018018,0.24774775,0.12162162,0.1981982,0.1891892,0.24774775,0.33333334,0.2927928,0.2792793,0.3063063,0.34234235,0.26576576,0.4144144,0.36486486,0.43243244,0.26126125,0.3828829,0.43693694,0.5495495,0.4864865,0.481982,,0.490991,0.5540541,0.5225225,0.5,0.5405405,0.5405405,0.6891892,0.5630631,0.6126126,0.7342342,0.5990991,0.7162162,0.7837838,0.7162162,0.6756757,0.7522523,0.7747748,0.7567568,0.8243243,0.9144144,0.7927928,0.8558559,0.9189189,0.9189189,0.9594595,0.9414414,0.963964,0.9594595,0.9594595,1.0
+LimiX (default),0.013513514,0.0945946,0.072072074,0.13963965,0.058558557,0.067567565,0.08558559,0.0945946,0.117117114,0.08558559,0.117117114,0.13063063,0.14414415,0.16216215,0.15765765,0.18468468,0.18468468,0.2117117,0.21621622,0.2882883,0.2117117,0.2837838,0.27027026,0.15765765,0.26576576,0.2792793,0.37387386,0.33333334,0.2882883,0.3243243,0.33783785,0.23423423,0.45045045,0.2972973,0.481982,0.43693694,0.36486486,0.44594595,0.44594595,0.4954955,0.5,0.509009,,0.45045045,0.5585586,0.5,0.5315315,0.5945946,0.5405405,0.5585586,0.5720721,0.5720721,0.6036036,0.7297297,0.6171171,0.7297297,0.5720721,0.5540541,0.7162162,0.6576577,0.7432432,0.6891892,0.7657658,0.7567568,0.8153153,0.7477477,0.7522523,0.8693694,0.9504505,0.9324324,0.9324324,0.9864865
+ExtraTrees (tuned),0.022522522,0.054054055,0.045045044,0.058558557,0.054054055,0.08558559,0.11261261,0.04954955,0.058558557,0.04954955,0.067567565,0.13513513,0.06306306,0.08108108,0.12612613,0.12162162,0.14864865,0.16216215,0.1981982,0.18468468,0.15315315,0.24774775,0.22972973,0.21621622,0.26126125,0.21621622,0.2837838,0.24324325,0.24774775,0.34234235,0.2972973,0.3153153,0.35585585,0.3153153,0.3783784,0.4234234,0.36486486,0.3918919,0.29504505,0.5,0.3873874,0.44594595,0.5495495,,0.5405405,0.42792794,0.42792794,0.5720721,0.6936937,0.5900901,0.5315315,0.7342342,0.6261261,0.6846847,0.6351351,0.6936937,0.6441441,0.9099099,0.7072072,0.5765766,0.7252252,0.9099099,0.8018018,0.7837838,0.9684685,0.9234234,0.954955,0.9954955,0.990991,1.0,1.0,1.0
+EBM (tuned + ensembled),0.027027028,0.022522522,0.0990991,0.036036037,0.027027028,0.09009009,0.045045044,0.058558557,0.11261261,0.12612613,0.13963965,0.117117114,0.1891892,0.2072072,0.15315315,0.14864865,0.15765765,0.16216215,0.2027027,0.1891892,0.27027026,0.3063063,0.2882883,0.23423423,0.2792793,0.2072072,0.1981982,0.25225225,0.24774775,0.27477476,0.37387386,0.2837838,0.25225225,0.33783785,0.3153153,0.3963964,0.4099099,0.43693694,0.43693694,0.3963964,0.509009,0.4774775,0.44144145,0.45945945,,0.44144145,0.5540541,0.8063063,0.518018,0.6351351,0.5045045,0.5540541,0.8603604,0.6981982,0.5585586,0.7297297,0.6531532,0.536036,0.6396396,0.7072072,0.7027027,0.6711712,0.8468468,0.8513514,0.7972973,0.8738739,0.8243243,0.8288288,0.8693694,0.9864865,0.9864865,0.9954955
+LightGBM (default),0.0,0.009009009,0.009009009,0.022522522,0.0045045046,0.04954955,0.013513514,0.0,0.022522522,0.013513514,0.022522522,0.04054054,0.058558557,0.058558557,0.067567565,0.036036037,0.08558559,0.04954955,0.072072074,0.018018018,0.08558559,0.16666667,0.1981982,0.12162162,0.2117117,0.027027028,0.2027027,0.08558559,0.09009009,0.27027026,0.26576576,0.2117117,0.26576576,0.24774775,0.32882884,0.37387386,0.33333334,0.45495495,0.5495495,0.45945945,0.46846846,0.5,0.5,0.5720721,0.5585586,,0.5135135,0.6171171,0.6081081,0.5945946,0.7117117,0.6531532,0.6711712,0.7477477,0.7117117,0.7477477,0.6576577,0.7477477,0.7792793,0.7567568,0.8333333,0.8828829,0.7792793,0.8918919,0.8558559,0.9279279,0.990991,0.9009009,0.9189189,0.9864865,0.9864865,0.9954955
+TorchMLP (tuned),0.0045045046,0.018018018,0.022522522,0.013513514,0.013513514,0.0945946,0.018018018,0.013513514,0.04054054,0.06306306,0.04954955,0.072072074,0.072072074,0.072072074,0.08108108,0.072072074,0.06306306,0.1936937,0.2072072,0.25675675,0.14414415,0.13513513,0.25225225,0.2027027,0.18468468,0.25225225,0.2972973,0.3063063,0.26126125,0.4099099,0.22972973,0.26576576,0.37387386,0.3063063,0.4009009,0.34234235,0.26126125,0.35135135,0.5630631,0.3918919,0.2792793,0.45945945,0.46846846,0.5720721,0.44594595,0.4864865,,0.481982,0.6081081,0.46846846,0.527027,0.6216216,0.5855856,0.7297297,0.6666667,0.7297297,0.545045,0.6261261,0.7027027,0.8783784,0.8108108,0.7207207,0.8288288,0.7747748,0.8378378,0.8333333,0.8783784,0.8873874,0.8828829,0.9009009,0.9009009,1.0
+EBM (tuned),0.018018018,0.013513514,0.08558559,0.036036037,0.018018018,0.0945946,0.03153153,0.067567565,0.1036036,0.0990991,0.1036036,0.0945946,0.17567568,0.18018018,0.13513513,0.12162162,0.13513513,0.12612613,0.13963965,0.16216215,0.23873875,0.24774775,0.25225225,0.21621622,0.23423423,0.17567568,0.15765765,0.1891892,0.18018018,0.22972973,0.34684685,0.27027026,0.2027027,0.2927928,0.23873875,0.35135135,0.35585585,0.4054054,0.4054054,0.3153153,0.43693694,0.45945945,0.4054054,0.42792794,0.1936937,0.3828829,0.518018,,0.46846846,0.5720721,0.45945945,0.5135135,0.7162162,0.6801802,0.5405405,0.7072072,0.6171171,0.5135135,0.5945946,0.6666667,0.6711712,0.6261261,0.8378378,0.8243243,0.7702703,0.8468468,0.8018018,0.8198198,0.8468468,0.981982,0.981982,0.9954955
+RandomForest (tuned + ensembled),0.0045045046,0.018018018,0.022522522,0.03153153,0.009009009,0.06306306,0.045045044,0.009009009,0.018018018,0.018018018,0.03153153,0.1036036,0.027027028,0.058558557,0.0945946,0.08558559,0.072072074,0.08108108,0.11261261,0.10810811,0.06306306,0.23873875,0.14414415,0.16216215,0.21621622,0.12612613,0.1936937,0.17117117,0.17117117,0.25675675,0.3108108,0.21621622,0.24324325,0.22072072,0.2837838,0.26576576,0.35585585,0.3108108,0.23423423,0.33333334,0.33783785,0.3108108,0.45945945,0.3063063,0.481982,0.3918919,0.3918919,0.5315315,,0.518018,0.46396396,0.6824324,0.5945946,0.6351351,0.5405405,0.6486486,0.5855856,0.7207207,0.6171171,0.5675676,0.6666667,0.9864865,0.7612613,0.7747748,0.9234234,0.9054054,0.9459459,0.9504505,0.9684685,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.009009009,0.013513514,0.04954955,0.009009009,0.0045045046,0.067567565,0.013513514,0.04954955,0.018018018,0.022522522,0.027027028,0.045045044,0.08558559,0.08558559,0.0945946,0.03153153,0.054054055,0.036036037,0.045045044,0.027027028,0.13963965,0.16666667,0.13063063,0.18018018,0.22522523,0.076576576,0.08108108,0.117117114,0.10810811,0.24774775,0.23873875,0.25675675,0.11261261,0.3198198,0.15765765,0.3153153,0.3108108,0.45045045,0.4009009,0.25225225,0.5045045,0.43693694,0.44144145,0.4099099,0.36486486,0.4054054,0.5315315,0.42792794,0.481982,,0.545045,0.5135135,0.5315315,0.5945946,0.5990991,0.5945946,0.8108108,0.5900901,0.6891892,0.6486486,0.7162162,0.7072072,0.6711712,0.963964,0.7657658,0.8603604,0.8828829,0.8963964,0.9144144,0.981982,0.981982,0.9954955
+XGBoost (default),0.0,0.027027028,0.04954955,0.0045045046,0.0,0.08108108,0.027027028,0.045045044,0.04054054,0.03153153,0.04954955,0.076576576,0.054054055,0.067567565,0.08108108,0.018018018,0.058558557,0.009009009,0.009009009,0.022522522,0.0945946,0.117117114,0.0945946,0.0945946,0.14414415,0.054054055,0.18018018,0.054054055,0.072072074,0.2117117,0.1936937,0.14414415,0.23423423,0.1981982,0.2882883,0.2927928,0.23423423,0.3243243,0.4234234,0.34234235,0.4054054,0.3873874,0.42792794,0.46846846,0.4954955,0.2882883,0.47297296,0.5405405,0.536036,0.45495495,,0.5810811,0.5720721,0.7072072,0.6846847,0.7072072,0.6036036,0.6576577,0.6486486,0.6171171,0.7342342,0.8423423,0.7477477,0.8018018,0.8423423,0.8153153,0.9504505,0.9009009,0.9189189,0.9504505,0.9504505,0.9864865
+RandomForest (tuned),0.0045045046,0.018018018,0.022522522,0.027027028,0.009009009,0.054054055,0.036036037,0.009009009,0.013513514,0.013513514,0.027027028,0.09009009,0.022522522,0.058558557,0.0945946,0.08108108,0.04954955,0.076576576,0.08558559,0.1036036,0.058558557,0.22972973,0.117117114,0.15315315,0.2117117,0.10810811,0.18018018,0.16216215,0.16216215,0.25675675,0.2882883,0.2117117,0.22522523,0.22072072,0.25225225,0.25225225,0.34684685,0.2882883,0.2117117,0.3243243,0.3153153,0.26576576,0.42792794,0.26576576,0.44594595,0.34684685,0.3783784,0.4864865,0.31756756,0.4864865,0.4189189,,0.5585586,0.5810811,0.5045045,0.5990991,0.536036,0.6216216,0.5810811,0.5495495,0.6351351,0.7747748,0.7567568,0.7522523,0.8963964,0.9009009,0.9459459,0.9504505,0.9594595,1.0,1.0,1.0
+EBM (default),0.013513514,0.009009009,0.058558557,0.045045044,0.018018018,0.058558557,0.027027028,0.054054055,0.04954955,0.027027028,0.058558557,0.054054055,0.09009009,0.0990991,0.072072074,0.076576576,0.058558557,0.06306306,0.076576576,0.058558557,0.18468468,0.1936937,0.2072072,0.18468468,0.18018018,0.076576576,0.08558559,0.14414415,0.1036036,0.17117117,0.25675675,0.24774775,0.117117114,0.27477476,0.18018018,0.2972973,0.27027026,0.35135135,0.35585585,0.26126125,0.3783784,0.4009009,0.3963964,0.37387386,0.13963965,0.32882884,0.4144144,0.2837838,0.4054054,0.46846846,0.42792794,0.44144145,,0.6261261,0.46846846,0.6306306,0.527027,0.4954955,0.5540541,0.6486486,0.6126126,0.5855856,0.7837838,0.7837838,0.7657658,0.7927928,0.7837838,0.7927928,0.8603604,0.9864865,0.9864865,0.990991
+TabSTAR (tuned + ensembled),0.076576576,0.08558559,0.13963965,0.1036036,0.0990991,0.13063063,0.13513513,0.0990991,0.17117117,0.18468468,0.18018018,0.1936937,0.17117117,0.21621622,0.18018018,0.2027027,0.2027027,0.2027027,0.2072072,0.22522523,0.2117117,0.22072072,0.23423423,0.23423423,0.22072072,0.22972973,0.22072072,0.23423423,0.22972973,0.25675675,0.23873875,0.25225225,0.27027026,0.26576576,0.26576576,0.26126125,0.24324325,0.3243243,0.3108108,0.2882883,0.26576576,0.2837838,0.27027026,0.3153153,0.3018018,0.25225225,0.27027026,0.3198198,0.36486486,0.4054054,0.2927928,0.4189189,0.37387386,,0.36486486,0.6306306,0.44594595,0.46846846,0.4054054,0.3918919,0.4054054,0.4234234,0.981982,0.4774775,0.6801802,0.5990991,0.5540541,0.7567568,0.7882883,0.8288288,0.8288288,0.9144144
+xRFM (default),0.009009009,0.03153153,0.036036037,0.027027028,0.022522522,0.036036037,0.067567565,0.03153153,0.04054054,0.03153153,0.036036037,0.0990991,0.045045044,0.054054055,0.08558559,0.072072074,0.11261261,0.12612613,0.12612613,0.15315315,0.06306306,0.16216215,0.076576576,0.027027028,0.12612613,0.16216215,0.22522523,0.16216215,0.17117117,0.26576576,0.2072072,0.1036036,0.3198198,0.13063063,0.33333334,0.18468468,0.3063063,0.3063063,0.35135135,0.36936936,0.27477476,0.21621622,0.3828829,0.36486486,0.44144145,0.2882883,0.33333334,0.45945945,0.45945945,0.4009009,0.3153153,0.4954955,0.5315315,0.6351351,,0.6351351,0.44594595,0.509009,0.5135135,0.5135135,0.5585586,0.6441441,0.7207207,0.6216216,0.8738739,0.7252252,0.8198198,0.9234234,0.9774775,0.9234234,0.9144144,0.990991
+TabSTAR (tuned),0.072072074,0.076576576,0.12162162,0.0990991,0.09009009,0.117117114,0.12162162,0.0990991,0.17117117,0.18018018,0.17567568,0.1891892,0.17567568,0.2072072,0.17567568,0.1936937,0.1936937,0.2027027,0.1981982,0.22072072,0.2117117,0.2117117,0.22072072,0.22972973,0.2117117,0.21621622,0.21621622,0.22522523,0.22972973,0.24774775,0.22972973,0.24774775,0.25225225,0.27027026,0.26126125,0.25675675,0.23423423,0.3153153,0.3063063,0.2837838,0.27027026,0.2837838,0.27027026,0.3063063,0.27027026,0.25225225,0.27027026,0.2927928,0.35135135,0.4054054,0.2927928,0.4009009,0.36936936,0.36936936,0.36486486,,0.44594595,0.44594595,0.4054054,0.3873874,0.4009009,0.4054054,0.9684685,0.4774775,0.6801802,0.5990991,0.5540541,0.7522523,0.7882883,0.8288288,0.8288288,0.8963964
+PerpetualBooster (tuned),0.0045045046,0.009009009,0.018018018,0.0045045046,0.0045045046,0.067567565,0.013513514,0.027027028,0.013513514,0.013513514,0.022522522,0.036036037,0.027027028,0.036036037,0.08558559,0.036036037,0.027027028,0.018018018,0.027027028,0.027027028,0.067567565,0.14864865,0.1036036,0.14864865,0.16666667,0.04054054,0.058558557,0.09009009,0.076576576,0.23423423,0.22522523,0.2072072,0.11261261,0.25225225,0.14414415,0.23873875,0.2837838,0.35585585,0.33783785,0.1891892,0.4189189,0.3243243,0.42792794,0.35585585,0.34684685,0.34234235,0.45495495,0.3828829,0.4144144,0.1891892,0.3963964,0.46396396,0.47297296,0.5540541,0.5540541,0.5540541,,0.545045,0.6396396,0.6171171,0.6711712,0.6711712,0.6396396,0.8265766,0.7207207,0.7927928,0.8468468,0.8738739,0.8963964,0.9774775,0.9774775,0.990991
+ExtraTrees (default),0.018018018,0.036036037,0.03153153,0.04954955,0.036036037,0.04054054,0.045045044,0.022522522,0.036036037,0.022522522,0.04054054,0.09009009,0.045045044,0.06306306,0.0990991,0.08108108,0.0945946,0.10810811,0.12612613,0.11261261,0.09009009,0.18468468,0.1036036,0.076576576,0.12612613,0.13963965,0.16666667,0.17117117,0.15315315,0.18018018,0.24324325,0.17567568,0.25675675,0.17117117,0.27027026,0.24324325,0.3108108,0.25225225,0.045045044,0.2792793,0.33783785,0.24774775,0.44594595,0.09009009,0.46396396,0.25225225,0.37387386,0.4864865,0.2792793,0.4099099,0.34234235,0.3783784,0.5045045,0.5315315,0.490991,0.5540541,0.45495495,,0.47297296,0.4054054,0.4774775,0.6396396,0.6981982,0.6351351,0.8783784,0.7657658,0.8423423,0.9234234,0.981982,0.9954955,0.9954955,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0045045046,0.0,0.0045045046,0.018018018,0.04054054,0.0,0.013513514,0.0,0.0,0.0045045046,0.022522522,0.013513514,0.018018018,0.027027028,0.0045045046,0.022522522,0.06306306,0.06306306,0.027027028,0.018018018,0.058558557,0.072072074,0.12162162,0.036036037,0.054054055,0.08108108,0.08108108,0.06306306,0.117117114,0.13063063,0.15315315,0.18018018,0.13063063,0.2072072,0.14864865,0.15765765,0.16666667,0.25675675,0.2837838,0.1981982,0.22522523,0.2837838,0.2927928,0.36036035,0.22072072,0.2972973,0.4054054,0.3828829,0.3108108,0.35135135,0.4189189,0.44594595,0.5945946,0.4864865,0.5945946,0.36036035,0.527027,,0.518018,0.6216216,0.5720721,0.6891892,0.6396396,0.7342342,0.9504505,0.8288288,0.8018018,0.8873874,0.9774775,0.9774775,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.04054054,0.0,0.0,0.0,0.0045045046,0.0,0.0045045046,0.009009009,0.009009009,0.0,0.0045045046,0.018018018,0.072072074,0.076576576,0.1036036,0.027027028,0.03153153,0.09009009,0.06306306,0.054054055,0.11261261,0.1981982,0.13063063,0.1036036,0.1936937,0.058558557,0.076576576,0.23873875,0.117117114,0.25675675,0.1891892,0.16216215,0.16666667,0.4054054,0.2837838,0.058558557,0.24324325,0.34234235,0.4234234,0.2927928,0.24324325,0.12162162,0.33333334,0.43243244,0.35135135,0.3828829,0.45045045,0.35135135,0.6081081,0.4864865,0.6126126,0.3828829,0.5945946,0.481982,,0.5405405,0.6441441,0.7072072,0.5405405,0.7747748,0.6351351,0.7657658,0.8513514,0.8603604,0.8873874,0.8873874,0.9864865
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.009009009,0.04054054,0.0,0.009009009,0.0,0.0045045046,0.0,0.0,0.013513514,0.009009009,0.018018018,0.0,0.009009009,0.03153153,0.04054054,0.009009009,0.022522522,0.018018018,0.06306306,0.0945946,0.03153153,0.018018018,0.067567565,0.045045044,0.03153153,0.10810811,0.067567565,0.117117114,0.17567568,0.11261261,0.1891892,0.11261261,0.08108108,0.12162162,0.26576576,0.25225225,0.10810811,0.17567568,0.25675675,0.27477476,0.2972973,0.16666667,0.1891892,0.32882884,0.33333334,0.2837838,0.26576576,0.36486486,0.3873874,0.5945946,0.44144145,0.5990991,0.32882884,0.5225225,0.3783784,0.45945945,,0.5585586,0.6936937,0.5720721,0.7072072,0.8198198,0.7927928,0.7702703,0.8738739,0.954955,0.9504505,1.0
+RandomForest (default),0.0045045046,0.013513514,0.0045045046,0.0,0.0045045046,0.027027028,0.009009009,0.0045045046,0.0,0.0045045046,0.009009009,0.036036037,0.0045045046,0.013513514,0.054054055,0.022522522,0.009009009,0.013513514,0.022522522,0.027027028,0.022522522,0.06306306,0.036036037,0.022522522,0.1036036,0.018018018,0.058558557,0.04954955,0.036036037,0.16216215,0.11261261,0.04054054,0.0990991,0.06306306,0.15315315,0.0945946,0.15765765,0.18018018,0.076576576,0.16216215,0.21621622,0.08558559,0.3108108,0.09009009,0.32882884,0.117117114,0.2792793,0.37387386,0.013513514,0.2927928,0.15765765,0.22522523,0.4144144,0.5765766,0.35585585,0.5945946,0.32882884,0.36036035,0.42792794,0.35585585,0.44144145,,0.6936937,0.5855856,0.7972973,0.6801802,0.7567568,0.8738739,0.9189189,0.954955,0.954955,0.981982
+TabSTAR (default),0.058558557,0.072072074,0.072072074,0.067567565,0.06306306,0.067567565,0.08108108,0.076576576,0.08558559,0.0990991,0.08108108,0.10810811,0.076576576,0.11261261,0.1036036,0.12162162,0.12612613,0.13063063,0.13963965,0.15315315,0.13963965,0.13513513,0.18018018,0.18018018,0.14414415,0.16216215,0.13963965,0.17117117,0.16216215,0.12162162,0.16216215,0.1981982,0.17567568,0.2117117,0.1981982,0.22072072,0.16666667,0.1891892,0.1936937,0.21621622,0.16666667,0.2072072,0.23423423,0.1981982,0.15315315,0.22072072,0.17117117,0.16216215,0.23873875,0.32882884,0.25225225,0.24324325,0.21621622,0.018018018,0.2792793,0.03153153,0.36036035,0.3018018,0.3108108,0.2927928,0.3063063,0.3063063,,0.3783784,0.5855856,0.5675676,0.44144145,0.6531532,0.7117117,0.8243243,0.8243243,0.8333333
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.04054054,0.0,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.009009009,0.036036037,0.009009009,0.009009009,0.0,0.0045045046,0.0,0.018018018,0.06306306,0.045045044,0.045045044,0.06306306,0.013513514,0.0,0.013513514,0.013513514,0.0945946,0.10810811,0.067567565,0.04954955,0.1036036,0.054054055,0.117117114,0.11261261,0.17117117,0.22522523,0.06306306,0.16216215,0.14414415,0.24324325,0.21621622,0.14864865,0.10810811,0.22522523,0.17567568,0.22522523,0.036036037,0.1981982,0.24774775,0.21621622,0.5225225,0.3783784,0.5225225,0.17342342,0.36486486,0.36036035,0.45945945,0.42792794,0.4144144,0.6216216,,0.5900901,0.5945946,0.6486486,0.7117117,0.8513514,0.9144144,0.9144144,0.972973
+KNN (tuned + ensembled),0.013513514,0.018018018,0.03153153,0.022522522,0.013513514,0.018018018,0.03153153,0.022522522,0.018018018,0.03153153,0.018018018,0.067567565,0.018018018,0.03153153,0.076576576,0.06306306,0.04954955,0.054054055,0.06306306,0.076576576,0.04054054,0.09009009,0.045045044,0.03153153,0.036036037,0.072072074,0.10810811,0.09009009,0.08558559,0.08108108,0.117117114,0.04954955,0.13513513,0.0945946,0.14864865,0.076576576,0.16216215,0.117117114,0.027027028,0.16216215,0.13963965,0.08108108,0.18468468,0.03153153,0.2027027,0.14414415,0.16216215,0.22972973,0.076576576,0.23423423,0.15765765,0.1036036,0.23423423,0.3198198,0.12612613,0.3198198,0.2792793,0.12162162,0.26576576,0.22522523,0.2927928,0.2027027,0.4144144,0.4099099,,0.44144145,0.33333334,0.9324324,0.972973,0.8783784,0.8783784,0.972973
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0045045046,0.04054054,0.0,0.0,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0045045046,0.0,0.009009009,0.0045045046,0.013513514,0.009009009,0.009009009,0.036036037,0.018018018,0.027027028,0.0,0.009009009,0.013513514,0.022522522,0.018018018,0.06306306,0.08558559,0.04954955,0.06306306,0.036036037,0.0990991,0.054054055,0.0945946,0.058558557,0.06306306,0.10810811,0.1036036,0.08108108,0.25225225,0.076576576,0.12612613,0.072072074,0.16666667,0.15315315,0.0945946,0.13963965,0.18468468,0.0990991,0.2072072,0.4009009,0.27477476,0.4009009,0.2072072,0.23423423,0.04954955,0.36486486,0.18018018,0.3198198,0.43243244,0.4054054,0.5585586,,0.45045045,0.7207207,0.7972973,0.9414414,0.9414414,0.9954955
+iLTM (default),0.0,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.013513514,0.009009009,0.0,0.018018018,0.045045044,0.0,0.0,0.0,0.0,0.0945946,0.018018018,0.009009009,0.04054054,0.018018018,0.04054054,0.036036037,0.054054055,0.076576576,0.045045044,0.04954955,0.10810811,0.04054054,0.24774775,0.045045044,0.17567568,0.009009009,0.12162162,0.1981982,0.054054055,0.117117114,0.04954955,0.054054055,0.21621622,0.44594595,0.18018018,0.44594595,0.15315315,0.15765765,0.17117117,0.23423423,0.2072072,0.24324325,0.5585586,0.35135135,0.6666667,0.5495495,,0.7612613,0.7927928,0.8243243,0.8198198,0.8783784
+KNN (tuned),0.0045045046,0.009009009,0.018018018,0.009009009,0.013513514,0.009009009,0.022522522,0.013513514,0.009009009,0.018018018,0.009009009,0.027027028,0.009009009,0.027027028,0.04954955,0.036036037,0.036036037,0.03153153,0.036036037,0.054054055,0.03153153,0.06306306,0.022522522,0.018018018,0.018018018,0.036036037,0.04054054,0.058558557,0.045045044,0.013513514,0.08558559,0.036036037,0.054054055,0.054054055,0.076576576,0.045045044,0.12162162,0.04054054,0.009009009,0.1036036,0.1036036,0.058558557,0.13063063,0.0045045046,0.17117117,0.0990991,0.11261261,0.18018018,0.04954955,0.1036036,0.0990991,0.04954955,0.2072072,0.24324325,0.076576576,0.24774775,0.12612613,0.076576576,0.1981982,0.14864865,0.22972973,0.12612613,0.34684685,0.2882883,0.067567565,0.2792793,0.23873875,,0.9189189,0.8603604,0.8603604,0.9144144
+KNN (default),0.009009009,0.018018018,0.009009009,0.009009009,0.009009009,0.009009009,0.013513514,0.018018018,0.009009009,0.009009009,0.009009009,0.03153153,0.009009009,0.022522522,0.04054054,0.027027028,0.022522522,0.027027028,0.03153153,0.03153153,0.013513514,0.054054055,0.018018018,0.018018018,0.009009009,0.036036037,0.03153153,0.054054055,0.036036037,0.013513514,0.06306306,0.027027028,0.04054054,0.058558557,0.058558557,0.036036037,0.11261261,0.027027028,0.009009009,0.076576576,0.0945946,0.036036037,0.04954955,0.009009009,0.13063063,0.08108108,0.117117114,0.15315315,0.03153153,0.08558559,0.08108108,0.04054054,0.13963965,0.2117117,0.022522522,0.2117117,0.1036036,0.018018018,0.11261261,0.13963965,0.12612613,0.08108108,0.2882883,0.14864865,0.027027028,0.2027027,0.2072072,0.08108108,,0.6306306,0.6351351,0.6891892
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.009009009,0.0,0.0045045046,0.0045045046,0.0,0.0,0.0,0.009009009,0.0,0.0045045046,0.027027028,0.013513514,0.0045045046,0.0,0.009009009,0.018018018,0.018018018,0.027027028,0.0,0.0,0.072072074,0.04054054,0.067567565,0.0,0.013513514,0.013513514,0.0990991,0.018018018,0.0,0.018018018,0.04954955,0.0,0.013513514,0.17117117,0.076576576,0.17117117,0.022522522,0.0045045046,0.022522522,0.11261261,0.045045044,0.045045044,0.17567568,0.08558559,0.12162162,0.058558557,0.17567568,0.13963965,0.36936936,,0.6013514,0.8243243
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.009009009,0.0,0.0045045046,0.0045045046,0.0,0.0,0.0,0.009009009,0.0,0.0045045046,0.027027028,0.013513514,0.0045045046,0.0,0.013513514,0.018018018,0.018018018,0.027027028,0.0,0.0,0.072072074,0.04054054,0.067567565,0.0,0.013513514,0.013513514,0.0990991,0.018018018,0.0,0.018018018,0.04954955,0.0,0.013513514,0.17117117,0.08558559,0.17117117,0.022522522,0.0045045046,0.022522522,0.11261261,0.04954955,0.045045044,0.17567568,0.08558559,0.12162162,0.058558557,0.18018018,0.13963965,0.36486486,0.39864865,,0.7972973
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.009009009,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0045045046,0.0,0.0045045046,0.0,0.009009009,0.0,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.013513514,0.0,0.0045045046,0.0045045046,0.0,0.0045045046,0.0,0.0045045046,0.013513514,0.0,0.009009009,0.08558559,0.009009009,0.1036036,0.009009009,0.0,0.0,0.013513514,0.0,0.018018018,0.16666667,0.027027028,0.027027028,0.0045045046,0.12162162,0.08558559,0.3108108,0.17567568,0.2027027,
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
index 8ab1db306591eeb42957bb80010adfd26d4a461c..9391bc603796d75fb15895bbd99a008701d7ddc9 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Regression | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_all/tasks_regression/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..7a49097ff46072830aadc2c1bdea572928e5c0ad
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_regression/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2313.0,"score":0.9,"rank":2.22,"hrank":1.72,"imp":0.604,"train_time":25.25,"predict_time":11.645,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+991\/-101","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1956.0,"score":0.634,"rank":7.93,"hrank":6.59,"imp":3.379,"train_time":735.58,"predict_time":9.557,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+686\/-97","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1928.0,"score":0.734,"rank":8.73,"hrank":2.63,"imp":1.883,"train_time":0.91,"predict_time":0.204,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1164\/-274","imputed":false},{"position":3,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1900.0,"score":0.675,"rank":9.62,"hrank":3.63,"imp":4.076,"train_time":366.93,"predict_time":2.982,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+408\/-7","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1886.0,"score":0.561,"rank":10.07,"hrank":8.38,"imp":4.068,"train_time":2.76,"predict_time":0.701,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+580\/-58","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1874.0,"score":0.587,"rank":10.44,"hrank":8.59,"imp":3.855,"train_time":735.58,"predict_time":0.714,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+636\/-92","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1858.0,"score":0.612,"rank":11.0,"hrank":2.78,"imp":3.417,"train_time":4786.6,"predict_time":361.183,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+869\/-152","imputed":false},{"position":7,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1829.0,"score":0.509,"rank":12.07,"hrank":10.29,"imp":4.444,"train_time":2526.31,"predict_time":3.717,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+632\/-37","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1823.0,"score":0.67,"rank":12.27,"hrank":3.56,"imp":3.118,"train_time":0.46,"predict_time":0.104,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1194\/-336","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1795.0,"score":0.488,"rank":13.36,"hrank":11.18,"imp":4.851,"train_time":2028.29,"predict_time":1.675,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+496\/-88","imputed":false},{"position":10,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1793.0,"score":0.469,"rank":13.42,"hrank":9.24,"imp":5.443,"train_time":473.02,"predict_time":8.73,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+352\/-0","imputed":false},{"position":11,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1788.0,"score":0.563,"rank":13.64,"hrank":8.13,"imp":3.716,"train_time":0.69,"predict_time":0.178,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+731\/-186","imputed":false},{"position":12,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1786.0,"score":0.56,"rank":13.73,"hrank":6.09,"imp":3.824,"train_time":4786.6,"predict_time":41.599,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+825\/-140","imputed":false},{"position":13,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1735.0,"score":0.452,"rank":15.82,"hrank":12.31,"imp":5.676,"train_time":4.87,"predict_time":1.868,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+324\/-78","imputed":false},{"position":14,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1700.0,"score":0.461,"rank":17.36,"hrank":12.62,"imp":4.759,"train_time":1.6,"predict_time":0.508,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+699\/-204","imputed":false},{"position":15,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1686.0,"score":0.48,"rank":17.98,"hrank":7.95,"imp":4.446,"train_time":45.77,"predict_time":39.212,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+722\/-147","imputed":false},{"position":16,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1645.0,"score":0.349,"rank":19.89,"hrank":14.6,"imp":5.88,"train_time":2028.29,"predict_time":0.081,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+472\/-154","imputed":false},{"position":17,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1630.0,"score":0.367,"rank":20.62,"hrank":17.01,"imp":6.056,"train_time":1.95,"predict_time":0.641,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+399\/-108","imputed":false},{"position":18,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1605.0,"score":0.333,"rank":21.84,"hrank":18.14,"imp":6.375,"train_time":425.61,"predict_time":0.615,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+319\/-59","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1603.0,"score":0.247,"rank":21.93,"hrank":15.56,"imp":7.997,"train_time":222.05,"predict_time":11.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+318\/-123","imputed":false},{"position":20,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1581.0,"score":0.213,"rank":23.0,"hrank":12.59,"imp":8.164,"train_time":912.76,"predict_time":0.345,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+290\/-92","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1561.0,"score":0.21,"rank":24.02,"hrank":13.2,"imp":8.244,"train_time":912.76,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+306\/-131","imputed":false},{"position":22,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1560.0,"score":0.201,"rank":24.07,"hrank":18.5,"imp":8.292,"train_time":222.05,"predict_time":0.968,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+311\/-127","imputed":false},{"position":23,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1551.0,"score":0.298,"rank":24.53,"hrank":20.99,"imp":6.209,"train_time":2526.31,"predict_time":0.166,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+421\/-174","imputed":false},{"position":24,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1517.0,"score":0.275,"rank":26.31,"hrank":22.6,"imp":7.176,"train_time":2406.36,"predict_time":0.713,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+314\/-182","imputed":false},{"position":25,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1480.0,"score":0.192,"rank":28.24,"hrank":19.56,"imp":8.929,"train_time":2.88,"predict_time":0.01,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+302\/-289","imputed":false},{"position":26,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1479.0,"score":0.199,"rank":28.27,"hrank":26.16,"imp":7.306,"train_time":425.61,"predict_time":0.04,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+242\/-83","imputed":false},{"position":27,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1452.0,"score":0.205,"rank":29.69,"hrank":26.64,"imp":7.924,"train_time":2406.36,"predict_time":0.05,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+233\/-164","imputed":false},{"position":28,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1447.0,"score":0.121,"rank":29.96,"hrank":27.57,"imp":8.863,"train_time":282.13,"predict_time":1.74,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+229\/-141","imputed":false},{"position":29,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1440.0,"score":0.181,"rank":30.36,"hrank":27.44,"imp":7.364,"train_time":6.88,"predict_time":0.094,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+277\/-143","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1434.0,"score":0.107,"rank":30.64,"hrank":27.3,"imp":8.983,"train_time":282.13,"predict_time":0.388,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+254\/-148","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1418.0,"score":0.102,"rank":31.49,"hrank":29.69,"imp":9.603,"train_time":2173.01,"predict_time":65.027,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-29","imputed":false},{"position":32,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1354.0,"score":0.123,"rank":34.96,"hrank":24.59,"imp":12.624,"train_time":123.72,"predict_time":0.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+407\/-458","imputed":false},{"position":33,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1336.0,"score":0.069,"rank":35.96,"hrank":34.77,"imp":9.163,"train_time":2173.01,"predict_time":16.096,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-89","imputed":false},{"position":34,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1333.0,"score":0.176,"rank":36.11,"hrank":23.55,"imp":8.642,"train_time":24.78,"predict_time":8.888,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+490\/-328","imputed":false},{"position":35,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1314.0,"score":0.068,"rank":37.11,"hrank":30.88,"imp":10.089,"train_time":0.6,"predict_time":0.08,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+295\/-282","imputed":false},{"position":36,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1312.0,"score":0.102,"rank":37.22,"hrank":28.94,"imp":12.82,"train_time":123.72,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+360\/-458","imputed":false},{"position":37,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1310.0,"score":0.195,"rank":37.31,"hrank":28.49,"imp":8.967,"train_time":0.48,"predict_time":0.126,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+374\/-338","imputed":false},{"position":38,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1299.0,"score":0.043,"rank":37.93,"hrank":36.19,"imp":10.003,"train_time":17.56,"predict_time":0.39,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+153\/-251","imputed":false},{"position":39,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1264.0,"score":0.04,"rank":39.8,"hrank":37.58,"imp":10.105,"train_time":3.13,"predict_time":0.038,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+159\/-196","imputed":false},{"position":40,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1249.0,"score":0.011,"rank":40.58,"hrank":39.17,"imp":10.726,"train_time":2.28,"predict_time":0.13,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+154\/-149","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1243.0,"score":0.061,"rank":40.87,"hrank":38.93,"imp":10.193,"train_time":4348.61,"predict_time":0.528,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+156\/-178","imputed":false},{"position":42,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1236.0,"score":0.014,"rank":41.24,"hrank":40.31,"imp":10.313,"train_time":0.55,"predict_time":0.233,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-156","imputed":false},{"position":43,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1227.0,"score":0.079,"rank":41.69,"hrank":35.9,"imp":13.84,"train_time":0.73,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+331\/-484","imputed":false},{"position":44,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1202.0,"score":0.0,"rank":42.98,"hrank":41.78,"imp":10.696,"train_time":17.56,"predict_time":0.192,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+139\/-254","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1182.0,"score":0.038,"rank":44.02,"hrank":41.99,"imp":11.165,"train_time":4348.61,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+156\/-219","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1174.0,"score":0.034,"rank":44.42,"hrank":41.51,"imp":12.042,"train_time":49.61,"predict_time":0.231,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+202\/-324","imputed":false},{"position":47,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1146.0,"score":0.001,"rank":45.84,"hrank":43.92,"imp":11.652,"train_time":6.75,"predict_time":0.235,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-249","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1143.0,"score":0.023,"rank":46.0,"hrank":43.56,"imp":12.233,"train_time":49.61,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+172\/-344","imputed":false},{"position":49,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1132.0,"score":0.0,"rank":46.51,"hrank":46.18,"imp":11.918,"train_time":79.08,"predict_time":0.546,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-192","imputed":false},{"position":50,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1119.0,"score":0.067,"rank":47.18,"hrank":37.36,"imp":11.931,"train_time":8.79,"predict_time":7.636,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+361\/-370","imputed":false},{"position":51,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1108.0,"score":0.0,"rank":47.69,"hrank":47.29,"imp":12.214,"train_time":79.08,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-202","imputed":false},{"position":52,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1076.0,"score":0.059,"rank":49.2,"hrank":41.4,"imp":15.75,"train_time":1213.97,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+347\/-554","imputed":false},{"position":53,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1018.0,"score":0.001,"rank":51.71,"hrank":49.72,"imp":15.117,"train_time":15.08,"predict_time":0.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+165\/-330","imputed":false},{"position":54,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1016.0,"score":0.041,"rank":51.82,"hrank":46.32,"imp":16.308,"train_time":1213.97,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+322\/-571","imputed":false},{"position":55,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1002.0,"score":0.0,"rank":52.4,"hrank":51.0,"imp":14.379,"train_time":1.5,"predict_time":0.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+173\/-491","imputed":false},{"position":56,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":52.49,"hrank":52.24,"imp":13.351,"train_time":0.43,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+22\/-248","imputed":false},{"position":57,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":978.0,"score":0.0,"rank":53.38,"hrank":51.19,"imp":14.05,"train_time":0.78,"predict_time":0.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+168\/-520","imputed":false},{"position":58,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":975.0,"score":0.034,"rank":53.51,"hrank":48.32,"imp":16.982,"train_time":2.36,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+323\/-615","imputed":false},{"position":59,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":902.0,"score":0.0,"rank":56.29,"hrank":55.48,"imp":16.112,"train_time":93.33,"predict_time":21.745,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-616","imputed":false},{"position":60,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":842.0,"score":0.0,"rank":58.27,"hrank":57.71,"imp":16.191,"train_time":318.98,"predict_time":0.827,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-506","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":800.0,"score":0.0,"rank":59.56,"hrank":59.13,"imp":16.639,"train_time":318.98,"predict_time":0.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-471","imputed":false},{"position":62,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":763.0,"score":0.0,"rank":60.6,"hrank":60.19,"imp":19.755,"train_time":142.58,"predict_time":1.184,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-387","imputed":false},{"position":63,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":672.0,"score":0.0,"rank":62.84,"hrank":62.57,"imp":23.616,"train_time":4632.6,"predict_time":5.283,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-663","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":662.0,"score":0.0,"rank":63.07,"hrank":62.66,"imp":20.189,"train_time":1.36,"predict_time":0.111,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-719","imputed":false},{"position":65,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":637.0,"score":0.0,"rank":63.6,"hrank":63.39,"imp":23.844,"train_time":4632.6,"predict_time":1.724,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+123\/-664","imputed":false},{"position":66,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":627.0,"score":0.0,"rank":63.8,"hrank":63.55,"imp":22.259,"train_time":142.58,"predict_time":0.064,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-429","imputed":false},{"position":67,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":504.0,"score":0.0,"rank":65.93,"hrank":65.82,"imp":27.081,"train_time":112.83,"predict_time":1.964,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-753","imputed":false},{"position":68,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":408.0,"score":0.0,"rank":67.16,"hrank":66.95,"imp":29.6,"train_time":0.06,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-1120","imputed":false},{"position":69,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":33.0,"score":0.0,"rank":69.9,"hrank":69.9,"imp":41.19,"train_time":44.3,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-1380","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-138.0,"score":0.0,"rank":70.74,"hrank":70.74,"imp":41.565,"train_time":44.3,"predict_time":0.011,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+176\/-1402","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-400.0,"score":0.0,"rank":71.78,"hrank":71.78,"imp":44.679,"train_time":0.13,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+195\/-1626","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_medium/pareto_front_explorer.html
index 20dbf83a4e2552bf3692d34e14f9a31c8a0e2d25..aa23ca566d9e91260f4483111388f87533db3430 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0095047043,"imp":8.9289217905,"elo":1479.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0151739618,"imp":8.2440800428,"elo":1561.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3448570028,"imp":8.1635768776,"elo":1581.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0132050202,"imp":13.8402266011,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0405844599,"imp":12.820361936,"elo":1312.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"imp":12.6244692317,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0053932693,"imp":16.9821360779,"elo":975.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0053282856,"imp":16.307855987,"elo":1015.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0697951177,"imp":15.7503986055,"elo":1075.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0781872102,"imp":14.0502699508,"elo":978.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0196168127,"imp":12.2328759419,"elo":1142.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2313039056,"imp":12.0422079013,"elo":1174.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"imp":20.1888549154,"elo":661.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1141755222,"imp":16.6393490521,"elo":799.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8269429369,"imp":16.1913776362,"elo":842.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2328706766,"imp":10.3133865423,"elo":1235.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9682498776,"imp":8.2922840605,"elo":1560.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":11.2363457974,"imp":7.9966042912,"elo":1602.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.8875279848,"imp":8.6418127076,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.0940273851,"imp":7.3642181904,"elo":1439.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.1656582564,"imp":6.2094250691,"elo":1551.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":3.7166074672,"imp":4.4441658088,"elo":1828.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.6406884541,"imp":6.0560180179,"elo":1629.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8676058389,"imp":5.675536215,"elo":1735.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.009266306,"imp":14.3785932579,"elo":1002.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1920034411,"imp":10.6961018894,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3903416157,"imp":10.0026128041,"elo":1298.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0529855097,"imp":13.3509676333,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0565808637,"imp":12.2137889713,"elo":1108.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5464279632,"imp":11.9179109649,"elo":1132.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"imp":11.6524728164,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0807033173,"imp":5.8804255611,"elo":1645.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.6746674551,"imp":4.851092681,"elo":1795.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"imp":4.7589446613,"elo":1700.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7137361675,"imp":3.8547006757,"elo":1874.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5569470263,"imp":3.3793076451,"elo":1956.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.6364561632,"imp":11.9310816279,"elo":1118.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"imp":4.4457870824,"elo":1686.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":41.5989132411,"imp":3.8237840689,"elo":1785.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":361.1825850771,"imp":3.4173495548,"elo":1858.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1784260866,"imp":3.7162273027,"elo":1787.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":11.6453720048,"imp":0.6036434685,"elo":2313.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1038916814,"imp":3.1179522944,"elo":1823.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0383065968,"imp":10.1054237502,"elo":1263.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0498907511,"imp":7.9241643499,"elo":1452.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.7132695165,"imp":7.1759292759,"elo":1516.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"imp":4.0683965406,"elo":1885.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042898407,"imp":1.8828107499,"elo":1928.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"imp":27.0811357382,"elo":504.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.723789106,"imp":23.8439307085,"elo":636.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"imp":23.6160952188,"elo":671.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1260951636,"imp":8.9674037362,"elo":1310.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0279916773,"imp":15.1168656964,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0335416412,"imp":11.1647618802,"elo":1182.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5275799068,"imp":10.1930533871,"elo":1243.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0799084859,"imp":10.0891414742,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3882893337,"imp":8.9825719868,"elo":1434.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.7397156138,"imp":8.8631737523,"elo":1447.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":21.7449541194,"imp":16.1121022022,"elo":901.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":16.0959669691,"imp":9.1633459226,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.026662372,"imp":9.6026560483,"elo":1418.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1303369377,"imp":10.7259140587,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.040253721,"imp":7.3060870345,"elo":1479.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.6149915399,"imp":6.375444867,"elo":1604.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0095047043,"x_train":2.8843766069,"imp":8.9289217905,"elo":1479.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0151739618,"x_train":912.7561702144,"imp":8.2440800428,"elo":1561.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3448570028,"x_train":912.7561702144,"imp":8.1635768776,"elo":1581.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0132050202,"x_train":0.7327990592,"imp":13.8402266011,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0405844599,"x_train":123.7245358586,"imp":12.820361936,"elo":1312.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"x_train":123.7245358586,"imp":12.6244692317,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0053932693,"x_train":2.363335295,"imp":16.9821360779,"elo":975.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0053282856,"x_train":1213.9718626455,"imp":16.307855987,"elo":1015.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0697951177,"x_train":1213.9718626455,"imp":15.7503986055,"elo":1075.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0781872102,"x_train":0.7788435912,"imp":14.0502699508,"elo":978.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0196168127,"x_train":49.6127891307,"imp":12.2328759419,"elo":1142.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2313039056,"x_train":49.6127891307,"imp":12.0422079013,"elo":1174.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"x_train":1.3590539265,"imp":20.1888549154,"elo":661.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1141755222,"x_train":318.981772136,"imp":16.6393490521,"elo":799.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8269429369,"x_train":318.981772136,"imp":16.1913776362,"elo":842.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2328706766,"x_train":0.5527468904,"imp":10.3133865423,"elo":1235.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9682498776,"x_train":222.0508632426,"imp":8.2922840605,"elo":1560.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":11.2363457974,"x_train":222.0508632426,"imp":7.9966042912,"elo":1602.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.8875279848,"x_train":24.7794415376,"imp":8.6418127076,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.0940273851,"x_train":6.8840236669,"imp":7.3642181904,"elo":1439.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.1656582564,"x_train":2526.3066806002,"imp":6.2094250691,"elo":1551.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":3.7166074672,"x_train":2526.3066806002,"imp":4.4441658088,"elo":1828.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.6406884541,"x_train":1.9532897692,"imp":6.0560180179,"elo":1629.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8676058389,"x_train":4.8736992926,"imp":5.675536215,"elo":1735.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.009266306,"x_train":1.4985629713,"imp":14.3785932579,"elo":1002.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1920034411,"x_train":17.5566235787,"imp":10.6961018894,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3903416157,"x_train":17.5566235787,"imp":10.0026128041,"elo":1298.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0529855097,"x_train":0.4305837825,"imp":13.3509676333,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0565808637,"x_train":79.0783741735,"imp":12.2137889713,"elo":1108.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5464279632,"x_train":79.0783741735,"imp":11.9179109649,"elo":1132.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"x_train":6.7532462417,"imp":11.6524728164,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0807033173,"x_train":2028.289449394,"imp":5.8804255611,"elo":1645.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.6746674551,"x_train":2028.289449394,"imp":4.851092681,"elo":1795.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"x_train":1.6000651298,"imp":4.7589446613,"elo":1700.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7137361675,"x_train":735.5848736649,"imp":3.8547006757,"elo":1874.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5569470263,"x_train":735.5848736649,"imp":3.3793076451,"elo":1956.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.6364561632,"x_train":8.7889189928,"imp":11.9310816279,"elo":1118.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"x_train":45.7738111436,"imp":4.4457870824,"elo":1686.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":41.5989132411,"x_train":4786.5989331939,"imp":3.8237840689,"elo":1785.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":361.1825850771,"x_train":4786.5989331939,"imp":3.4173495548,"elo":1858.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1784260866,"x_train":0.6904298262,"imp":3.7162273027,"elo":1787.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":11.6453720048,"x_train":25.251811811,"imp":0.6036434685,"elo":2313.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1038916814,"x_train":0.4638004915,"imp":3.1179522944,"elo":1823.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0383065968,"x_train":3.1341264164,"imp":10.1054237502,"elo":1263.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0498907511,"x_train":2406.360319586,"imp":7.9241643499,"elo":1452.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.7132695165,"x_train":2406.360319586,"imp":7.1759292759,"elo":1516.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"x_train":2.763002179,"imp":4.0683965406,"elo":1885.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042898407,"x_train":0.9134642862,"imp":1.8828107499,"elo":1928.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"x_train":112.8279660853,"imp":27.0811357382,"elo":504.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.723789106,"x_train":4632.5950923332,"imp":23.8439307085,"elo":636.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"x_train":4632.5950923332,"imp":23.6160952188,"elo":671.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1260951636,"x_train":0.4819098389,"imp":8.9674037362,"elo":1310.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0279916773,"x_train":15.0791065291,"imp":15.1168656964,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0335416412,"x_train":4348.6118964812,"imp":11.1647618802,"elo":1182.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5275799068,"x_train":4348.6118964812,"imp":10.1930533871,"elo":1243.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0799084859,"x_train":0.597628695,"imp":10.0891414742,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3882893337,"x_train":282.1320041804,"imp":8.9825719868,"elo":1434.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.7397156138,"x_train":282.1320041804,"imp":8.8631737523,"elo":1447.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":21.7449541194,"x_train":93.3311418241,"imp":16.1121022022,"elo":901.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":16.0959669691,"x_train":2173.0087869337,"imp":9.1633459226,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.026662372,"x_train":2173.0087869337,"imp":9.6026560483,"elo":1418.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1303369377,"x_train":2.275318476,"imp":10.7259140587,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.040253721,"x_train":425.6136485357,"imp":7.3060870345,"elo":1479.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.6149915399,"x_train":425.6136485357,"imp":6.375444867,"elo":1604.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
index 6d709821f9cebde5d5c0294145c8e3d8585071a5..2f2e1555e735c3cdf802e8106838b4ff564b750a 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":2.8843766069,"imp":8.9299549058,"elo":1489.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":8.2528193911,"imp":8.7105054199,"elo":1511.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":21.6626665741,"imp":8.4887309209,"elo":1549.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":44.1604809025,"imp":8.3923746931,"elo":1560.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":109.0816998293,"imp":8.2570654578,"elo":1579.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":221.6369492763,"imp":8.2079108097,"elo":1589.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":453.4873826918,"imp":8.1769147819,"elo":1600.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":679.1817937831,"imp":8.1709441403,"elo":1600.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":912.7561702144,"imp":8.1646332008,"elo":1601.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.7327990592,"imp":13.8412569008,"elo":1236.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.3862926068,"imp":13.8758796411,"elo":1224.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.0643858395,"imp":13.7477732407,"elo":1243.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.9372791069,"imp":13.4747905089,"elo":1278.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":15.3821406623,"imp":13.1467714941,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":31.1470442744,"imp":12.9349519149,"elo":1343.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":61.7023905687,"imp":12.7559455414,"elo":1360.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":92.1424404788,"imp":12.6633921819,"elo":1372.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":123.7245358586,"imp":12.6257470877,"elo":1381.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.363335295,"imp":16.9830509773,"elo":983.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":10.5715947461,"imp":16.6384137163,"elo":994.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":33.6247878081,"imp":16.3265767301,"elo":1014.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":62.9357438641,"imp":16.1826188225,"elo":1035.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":154.7347742278,"imp":15.94979051,"elo":1057.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":312.6511755596,"imp":15.8658173525,"elo":1076.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":608.0045919755,"imp":15.7966651015,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":896.9079641691,"imp":15.7709114778,"elo":1098.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1213.9718626455,"imp":15.752285937,"elo":1107.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.7788435912,"imp":14.0512924699,"elo":975.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.0217265688,"imp":13.4121426201,"elo":1039.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.7428029518,"imp":12.7467383474,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9291798058,"imp":12.5238198046,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.6147364724,"imp":12.1897531097,"elo":1164.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.7108167547,"imp":12.1313110609,"elo":1174.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.9416823722,"imp":12.0738639893,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.1384318226,"imp":12.0536734972,"elo":1189.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":49.6127891307,"imp":12.043520821,"elo":1195.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.3590539265,"imp":20.1897089635,"elo":664.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.8414619709,"imp":20.0123167582,"elo":687.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":7.8616883329,"imp":19.5912779375,"elo":712.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":16.2112927486,"imp":19.2530853425,"elo":731.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":38.3780166466,"imp":18.1212969579,"elo":765.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":77.3536256265,"imp":17.3890213875,"elo":789.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":157.4362726765,"imp":16.8092188229,"elo":812.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":237.2917035986,"imp":16.4215969874,"elo":839.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":318.981772136,"imp":16.1924281318,"elo":861.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.5527468904,"imp":10.3144188427,"elo":1246.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":1.764654979,"imp":10.1439744184,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":5.1230989868,"imp":9.3779852563,"elo":1380.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":10.4740202796,"imp":8.9461937198,"elo":1445.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":27.1713091927,"imp":8.3796482578,"elo":1546.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":54.7938156246,"imp":8.2125157706,"elo":1587.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":109.9587992843,"imp":8.0892161778,"elo":1612.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":165.4413385162,"imp":8.0466483632,"elo":1625.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":222.0508632426,"imp":7.9979976356,"elo":1642.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":24.7794415376,"imp":8.6432598339,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":6.8840236669,"imp":7.3652532414,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":21.1491685049,"imp":6.8490090606,"elo":1508.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.2869505149,"imp":6.025229913,"elo":1601.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":121.8592858048,"imp":5.5681201505,"elo":1679.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":292.3137710756,"imp":4.9895084388,"elo":1770.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":629.8900400305,"imp":4.7518111757,"elo":1808.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1243.5591974675,"imp":4.5697026849,"elo":1839.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1862.5519447854,"imp":4.4826479243,"elo":1858.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2526.3066806002,"imp":4.4446564679,"elo":1873.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":1.9532897692,"imp":6.0570910871,"elo":1666.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":4.8736992926,"imp":5.6766219205,"elo":1781.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.4985629713,"imp":14.3795592436,"elo":1021.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":5.1141485965,"imp":11.8232633719,"elo":1166.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":14.778241098,"imp":10.0724475671,"elo":1285.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":17.5566235787,"imp":10.003641549,"elo":1300.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4305837825,"imp":13.3519737021,"elo":999.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.815213214,"imp":12.9091076456,"elo":1041.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":2.0588185718,"imp":12.5338637204,"elo":1070.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":4.0449726807,"imp":12.2144596854,"elo":1107.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":9.8632350349,"imp":12.0433557673,"elo":1129.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":19.655091039,"imp":11.9759550183,"elo":1140.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":39.2307694386,"imp":11.9396715877,"elo":1152.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":58.9645108517,"imp":11.9252718852,"elo":1156.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":79.0783741735,"imp":11.9189325871,"elo":1159.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.7532462417,"imp":11.6535004345,"elo":1166.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.9212846988,"imp":9.3001805229,"elo":1330.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":45.3026550705,"imp":7.0868286661,"elo":1549.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":98.5294547231,"imp":6.2737962747,"elo":1632.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":250.4641319867,"imp":5.6133750729,"elo":1735.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":506.2044753501,"imp":5.2815085909,"elo":1771.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1007.9516707303,"imp":5.0497861022,"elo":1808.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1512.7513669099,"imp":4.944648337,"elo":1827.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2028.289449394,"imp":4.8525237555,"elo":1848.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.6000651298,"imp":4.76003587,"elo":1727.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.1278647155,"imp":4.6479612017,"elo":1745.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.2069927888,"imp":4.4088830446,"elo":1768.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":34.2536042268,"imp":4.1041097996,"elo":1821.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":90.6790780885,"imp":3.7645520907,"elo":1914.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":181.3952243454,"imp":3.6005496212,"elo":1955.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":363.9172361826,"imp":3.4795975726,"elo":1986.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":547.7650992279,"imp":3.4134922755,"elo":2007.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":735.5848736649,"imp":3.3801975505,"elo":2017.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":8.7889189928,"imp":11.932087857,"elo":1147.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":45.7738111436,"imp":4.4469104839,"elo":1698.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.0936255942,"imp":4.3204912586,"elo":1722.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":132.9785409439,"imp":4.1649385631,"elo":1745.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":233.9204718624,"imp":4.0433518212,"elo":1765.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":585.2609706035,"imp":3.8074108911,"elo":1801.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1183.6501039466,"imp":3.6469702608,"elo":1827.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2376.2279314343,"imp":3.5169662809,"elo":1855.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3574.1919132683,"imp":3.4472605818,"elo":1868.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4786.5989331939,"imp":3.4184218295,"elo":1877.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.6904298262,"imp":3.7173335385,"elo":1812.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":25.251811811,"imp":0.6047647202,"elo":2372.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.4638004915,"imp":3.1190648248,"elo":1843.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":3.1341264164,"imp":10.1064497634,"elo":1272.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":11.6448003172,"imp":9.496118583,"elo":1343.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":46.7227871137,"imp":8.708612053,"elo":1403.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":111.5296374455,"imp":8.2026296087,"elo":1445.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":297.5742071652,"imp":7.7725618975,"elo":1483.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":594.9119312701,"imp":7.5192422695,"elo":1509.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1160.4787998549,"imp":7.3290046159,"elo":1529.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1766.291540139,"imp":7.2426921608,"elo":1541.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2406.360319586,"imp":7.1769941527,"elo":1555.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.763002179,"imp":4.0694795565,"elo":1928.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":0.9134642862,"imp":1.8854650044,"elo":1927.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":112.8279660853,"imp":27.0818393232,"elo":492.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":136.6220329381,"imp":26.9998231049,"elo":524.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":192.3169267284,"imp":26.8524843999,"elo":535.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":268.6819124436,"imp":26.4817153546,"elo":556.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":587.894611724,"imp":25.7481757445,"elo":583.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1166.9417941917,"imp":24.8517487919,"elo":608.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2300.4888270054,"imp":24.2935720809,"elo":628.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3411.7622050693,"imp":23.902805457,"elo":659.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4632.5950923332,"imp":23.61683989,"elo":695.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.4819098389,"imp":8.9684319131,"elo":1328.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":15.0791065291,"imp":15.1178597201,"elo":1034.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":37.5490244182,"imp":14.1906901618,"elo":1059.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":99.0264645811,"imp":12.8390046101,"elo":1121.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":209.9682919696,"imp":12.1128246668,"elo":1165.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":502.6360049647,"imp":11.2660448555,"elo":1203.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1067.3549756812,"imp":10.8935667918,"elo":1226.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2143.4901830021,"imp":10.5085358073,"elo":1243.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3266.8165720174,"imp":10.3122475245,"elo":1255.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4348.6118964812,"imp":10.1939206839,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.597628695,"imp":10.090168709,"elo":1321.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.4341782987,"imp":9.9737458625,"elo":1331.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":7.341922179,"imp":9.5555612023,"elo":1388.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":13.4253918933,"imp":9.3535205866,"elo":1407.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":35.1370362218,"imp":9.1242230507,"elo":1432.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":69.8388019133,"imp":8.9984966808,"elo":1448.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":138.6843072833,"imp":8.9098779875,"elo":1461.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":209.6003170598,"imp":8.8821110462,"elo":1468.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":282.1320041804,"imp":8.8639971487,"elo":1475.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":93.3311418241,"imp":16.1130673079,"elo":912.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":176.98147887,"imp":14.0309647611,"elo":1004.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":425.9076940912,"imp":18.8066937887,"elo":1116.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":844.3867407395,"imp":14.9020596719,"elo":1229.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":2092.7290649042,"imp":9.6291821749,"elo":1435.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":2173.0087869337,"imp":9.6037114724,"elo":1441.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.275318476,"imp":10.7269469123,"elo":1258.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":4.3886320534,"imp":9.9095442967,"elo":1318.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":10.7096165053,"imp":8.8150655878,"elo":1378.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":21.3971746465,"imp":7.8221470759,"elo":1482.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":53.3414186748,"imp":7.014236119,"elo":1555.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":106.1300468914,"imp":6.6825183636,"elo":1585.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":212.0315758458,"imp":6.4961289435,"elo":1609.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":317.9586579883,"imp":6.4098620166,"elo":1624.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":425.6136485357,"imp":6.3770966809,"elo":1637.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0095047043,"x_train":2.8843766069,"imp":8.9299549058,"elo":1489.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0481663313,"x_train":8.2528193911,"imp":8.7105054199,"elo":1511.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1163670459,"x_train":21.6626665741,"imp":8.4887309209,"elo":1549.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1729198824,"x_train":44.1604809025,"imp":8.3923746931,"elo":1560.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2175739337,"x_train":109.0816998293,"imp":8.2570654578,"elo":1579.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2694149278,"x_train":221.6369492763,"imp":8.2079108097,"elo":1589.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3197063318,"x_train":453.4873826918,"imp":8.1769147819,"elo":1600.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3407350551,"x_train":679.1817937831,"imp":8.1709441403,"elo":1600.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3448570028,"x_train":912.7561702144,"imp":8.1646332008,"elo":1601.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0132050202,"x_train":0.7327990592,"imp":13.8412569008,"elo":1236.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.027612708,"x_train":1.3862926068,"imp":13.8758796411,"elo":1224.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0666759456,"x_train":3.0643858395,"imp":13.7477732407,"elo":1243.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1073929028,"x_train":5.9372791069,"imp":13.4747905089,"elo":1278.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1481068526,"x_train":15.3821406623,"imp":13.1467714941,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1786997869,"x_train":31.1470442744,"imp":12.9349519149,"elo":1343.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2109984061,"x_train":61.7023905687,"imp":12.7559455414,"elo":1360.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2269570142,"x_train":92.1424404788,"imp":12.6633921819,"elo":1372.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"x_train":123.7245358586,"imp":12.6257470877,"elo":1381.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0053932693,"x_train":2.363335295,"imp":16.9830509773,"elo":983.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.010637589,"x_train":10.5715947461,"imp":16.6384137163,"elo":994.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0211515534,"x_train":33.6247878081,"imp":16.3265767301,"elo":1014.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0306648453,"x_train":62.9357438641,"imp":16.1826188225,"elo":1035.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0433861149,"x_train":154.7347742278,"imp":15.94979051,"elo":1057.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0522152152,"x_train":312.6511755596,"imp":15.8658173525,"elo":1076.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0627955752,"x_train":608.0045919755,"imp":15.7966651015,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0676616143,"x_train":896.9079641691,"imp":15.7709114778,"elo":1098.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.070209085,"x_train":1213.9718626455,"imp":15.752285937,"elo":1107.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0781872102,"x_train":0.7788435912,"imp":14.0512924699,"elo":975.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0889630222,"x_train":1.0217265688,"imp":13.4121426201,"elo":1039.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0926460136,"x_train":1.7428029518,"imp":12.7467383474,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1127201263,"x_train":2.9291798058,"imp":12.5238198046,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1338232875,"x_train":6.6147364724,"imp":12.1897531097,"elo":1164.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1601021681,"x_train":12.7108167547,"imp":12.1313110609,"elo":1174.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739802168,"x_train":24.9416823722,"imp":12.0738639893,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1953387622,"x_train":37.1384318226,"imp":12.0536734972,"elo":1189.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2313039056,"x_train":49.6127891307,"imp":12.043520821,"elo":1195.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"x_train":1.3590539265,"imp":20.1897089635,"elo":664.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2081543614,"x_train":2.8414619709,"imp":20.0123167582,"elo":687.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3420605892,"x_train":7.8616883329,"imp":19.5912779375,"elo":712.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4546707256,"x_train":16.2112927486,"imp":19.2530853425,"elo":731.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4701854923,"x_train":38.3780166466,"imp":18.1212969579,"elo":765.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5743045365,"x_train":77.3536256265,"imp":17.3890213875,"elo":789.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7208507445,"x_train":157.4362726765,"imp":16.8092188229,"elo":812.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7541111543,"x_train":237.2917035986,"imp":16.4215969874,"elo":839.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8269429369,"x_train":318.981772136,"imp":16.1924281318,"elo":861.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2328706766,"x_train":0.5527468904,"imp":10.3144188427,"elo":1246.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4016545585,"x_train":1.764654979,"imp":10.1439744184,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2355387213,"x_train":5.1230989868,"imp":9.3779852563,"elo":1380.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.8851780262,"x_train":10.4740202796,"imp":8.9461937198,"elo":1445.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.4690263501,"x_train":27.1713091927,"imp":8.3796482578,"elo":1546.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.1421527385,"x_train":54.7938156246,"imp":8.2125157706,"elo":1587.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.9341743249,"x_train":109.9587992843,"imp":8.0892161778,"elo":1612.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":11.0753701273,"x_train":165.4413385162,"imp":8.0466483632,"elo":1625.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":11.2363457974,"x_train":222.0508632426,"imp":7.9979976356,"elo":1642.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":8.8875279848,"x_train":24.7794415376,"imp":8.6432598339,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.0940273851,"x_train":6.8840236669,"imp":7.3652532414,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2557002328,"x_train":21.1491685049,"imp":6.8490090606,"elo":1508.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.578211498,"x_train":60.2869505149,"imp":6.025229913,"elo":1601.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9818556188,"x_train":121.8592858048,"imp":5.5681201505,"elo":1679.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.6674120378,"x_train":292.3137710756,"imp":4.9895084388,"elo":1770.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2999785133,"x_train":629.8900400305,"imp":4.7518111757,"elo":1808.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.975897112,"x_train":1243.5591974675,"imp":4.5697026849,"elo":1839.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.4550985564,"x_train":1862.5519447854,"imp":4.4826479243,"elo":1858.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.7166074672,"x_train":2526.3066806002,"imp":4.4446564679,"elo":1873.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.6406884541,"x_train":1.9532897692,"imp":6.0570910871,"elo":1666.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8676058389,"x_train":4.8736992926,"imp":5.6766219205,"elo":1781.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.009266306,"x_train":1.4985629713,"imp":14.3795592436,"elo":1021.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1022784942,"x_train":5.1141485965,"imp":11.8232633719,"elo":1166.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3309819924,"x_train":14.778241098,"imp":10.0724475671,"elo":1285.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3903416157,"x_train":17.5566235787,"imp":10.003641549,"elo":1300.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0529855097,"x_train":0.4305837825,"imp":13.3519737021,"elo":999.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1039804779,"x_train":0.815213214,"imp":12.9091076456,"elo":1041.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1684108594,"x_train":2.0588185718,"imp":12.5338637204,"elo":1070.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2358696304,"x_train":4.0449726807,"imp":12.2144596854,"elo":1107.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.356186403,"x_train":9.8632350349,"imp":12.0433557673,"elo":1129.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4195441878,"x_train":19.655091039,"imp":11.9759550183,"elo":1140.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4998627172,"x_train":39.2307694386,"imp":11.9396715877,"elo":1152.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.53892783,"x_train":58.9645108517,"imp":11.9252718852,"elo":1156.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5464279632,"x_train":79.0783741735,"imp":11.9189325871,"elo":1159.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"x_train":6.7532462417,"imp":11.6535004345,"elo":1166.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3080259999,"x_train":15.9212846988,"imp":9.3001805229,"elo":1330.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3619920454,"x_train":45.3026550705,"imp":7.0868286661,"elo":1549.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5857834951,"x_train":98.5294547231,"imp":6.2737962747,"elo":1632.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.897596221,"x_train":250.4641319867,"imp":5.6133750729,"elo":1735.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1621653964,"x_train":506.2044753501,"imp":5.2815085909,"elo":1771.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4580786339,"x_train":1007.9516707303,"imp":5.0497861022,"elo":1808.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5889939537,"x_train":1512.7513669099,"imp":4.944648337,"elo":1827.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6682080328,"x_train":2028.289449394,"imp":4.8525237555,"elo":1848.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"x_train":1.6000651298,"imp":4.76003587,"elo":1727.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4031929556,"x_train":5.1278647155,"imp":4.6479612017,"elo":1745.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2513268047,"x_train":16.2069927888,"imp":4.4088830446,"elo":1768.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.9918356458,"x_train":34.2536042268,"imp":4.1041097996,"elo":1821.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4620925312,"x_train":90.6790780885,"imp":3.7645520907,"elo":1914.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5018360728,"x_train":181.3952243454,"imp":3.6005496212,"elo":1955.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.7259397593,"x_train":363.9172361826,"imp":3.4795975726,"elo":1986.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.0781180135,"x_train":547.7650992279,"imp":3.4134922755,"elo":2007.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.6037629654,"x_train":735.5848736649,"imp":3.3801975505,"elo":2017.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.6364561632,"x_train":8.7889189928,"imp":11.932087857,"elo":1147.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"x_train":45.7738111436,"imp":4.4469104839,"elo":1698.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":59.6608119519,"x_train":70.0936255942,"imp":4.3204912586,"elo":1722.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":83.8163328946,"x_train":132.9785409439,"imp":4.1649385631,"elo":1745.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":117.6114176494,"x_train":233.9204718624,"imp":4.0433518212,"elo":1765.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":205.7656427606,"x_train":585.2609706035,"imp":3.8074108911,"elo":1801.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":266.8254661426,"x_train":1183.6501039466,"imp":3.6469702608,"elo":1827.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":326.9571428165,"x_train":2376.2279314343,"imp":3.5169662809,"elo":1855.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":356.2280399404,"x_train":3574.1919132683,"imp":3.4472605818,"elo":1868.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":361.1825850771,"x_train":4786.5989331939,"imp":3.4184218295,"elo":1877.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1784260866,"x_train":0.6904298262,"imp":3.7173335385,"elo":1812.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":11.6453720048,"x_train":25.251811811,"imp":0.6047647202,"elo":2372.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1038916814,"x_train":0.4638004915,"imp":3.1190648248,"elo":1843.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0383065968,"x_train":3.1341264164,"imp":10.1064497634,"elo":1272.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0804358946,"x_train":11.6448003172,"imp":9.496118583,"elo":1343.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.147897522,"x_train":46.7227871137,"imp":8.708612053,"elo":1403.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2244672749,"x_train":111.5296374455,"imp":8.2026296087,"elo":1445.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3405422347,"x_train":297.5742071652,"imp":7.7725618975,"elo":1483.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4659120865,"x_train":594.9119312701,"imp":7.5192422695,"elo":1509.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.6581520332,"x_train":1160.4787998549,"imp":7.3290046159,"elo":1529.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.7117167229,"x_train":1766.291540139,"imp":7.2426921608,"elo":1541.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.7132695165,"x_train":2406.360319586,"imp":7.1769941527,"elo":1555.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"x_train":2.763002179,"imp":4.0694795565,"elo":1928.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042898407,"x_train":0.9134642862,"imp":1.8854650044,"elo":1927.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"x_train":112.8279660853,"imp":27.0818393232,"elo":492.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9334778523,"x_train":136.6220329381,"imp":26.9998231049,"elo":524.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1406475162,"x_train":192.3169267284,"imp":26.8524843999,"elo":535.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.4886555763,"x_train":268.6819124436,"imp":26.4817153546,"elo":556.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.967901931,"x_train":587.894611724,"imp":25.7481757445,"elo":583.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0057237984,"x_train":1166.9417941917,"imp":24.8517487919,"elo":608.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.7178375205,"x_train":2300.4888270054,"imp":24.2935720809,"elo":628.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9435530344,"x_train":3411.7622050693,"imp":23.902805457,"elo":659.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"x_train":4632.5950923332,"imp":23.61683989,"elo":695.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1260951636,"x_train":0.4819098389,"imp":8.9684319131,"elo":1328.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0279916773,"x_train":15.0791065291,"imp":15.1178597201,"elo":1034.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0762465224,"x_train":37.5490244182,"imp":14.1906901618,"elo":1059.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1408050768,"x_train":99.0264645811,"imp":12.8390046101,"elo":1121.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1800817267,"x_train":209.9682919696,"imp":12.1128246668,"elo":1165.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2447433002,"x_train":502.6360049647,"imp":11.2660448555,"elo":1203.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3486057789,"x_train":1067.3549756812,"imp":10.8935667918,"elo":1226.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4465492625,"x_train":2143.4901830021,"imp":10.5085358073,"elo":1243.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5425320243,"x_train":3266.8165720174,"imp":10.3122475245,"elo":1255.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5275799068,"x_train":4348.6118964812,"imp":10.1939206839,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0799084859,"x_train":0.597628695,"imp":10.090168709,"elo":1321.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3337766093,"x_train":2.4341782987,"imp":9.9737458625,"elo":1331.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8523423396,"x_train":7.341922179,"imp":9.5555612023,"elo":1388.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1380843561,"x_train":13.4253918933,"imp":9.3535205866,"elo":1407.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.5157666044,"x_train":35.1370362218,"imp":9.1242230507,"elo":1432.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7011934543,"x_train":69.8388019133,"imp":8.9984966808,"elo":1448.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.839001453,"x_train":138.6843072833,"imp":8.9098779875,"elo":1461.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8309092575,"x_train":209.6003170598,"imp":8.8821110462,"elo":1468.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7397156138,"x_train":282.1320041804,"imp":8.8639971487,"elo":1475.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":21.7449541194,"x_train":93.3311418241,"imp":16.1130673079,"elo":912.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":27.6110229171,"x_train":176.98147887,"imp":14.0309647611,"elo":1004.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.033754283,"x_train":425.9076940912,"imp":18.8066937887,"elo":1116.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":46.9242331846,"x_train":844.3867407395,"imp":14.9020596719,"elo":1229.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":66.5740410026,"x_train":2092.7290649042,"imp":9.6291821749,"elo":1435.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":65.026662372,"x_train":2173.0087869337,"imp":9.6037114724,"elo":1441.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1303369377,"x_train":2.275318476,"imp":10.7269469123,"elo":1258.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.1626576583,"x_train":4.3886320534,"imp":9.9095442967,"elo":1318.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.175716895,"x_train":10.7096165053,"imp":8.8150655878,"elo":1378.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.2654850598,"x_train":21.3971746465,"imp":7.8221470759,"elo":1482.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.3507721627,"x_train":53.3414186748,"imp":7.014236119,"elo":1555.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.4554854246,"x_train":106.1300468914,"imp":6.6825183636,"elo":1585.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.5400047224,"x_train":212.0315758458,"imp":6.4961289435,"elo":1609.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":0.5704772449,"x_train":317.9586579883,"imp":6.4098620166,"elo":1624.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":0.6149915399,"x_train":425.6136485357,"imp":6.3770966809,"elo":1637.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..da1a3d568732d58feaa433361e1a5f9aa23b76e3
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_regression/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned)", "TabDPT (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "TabICLv2 (default)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabDPT-Turbo (default)", "TabDPT (tuned)", "Nori-30M (default)", "RealTabPFN-2.5 (default)", "TabDPT (default)", "RealMLP (tuned)", "Nori (default)", "xRFM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "LightGBM (tuned)", "ModernNCA (tuned)", "TabM (tuned + ensembled)", "CatBoost (default)", "xRFM (tuned)", "TabM (tuned)", "XGBoost (tuned + ensembled)", "ModernNCA (default)", "XGBoost (tuned)", "iLTM (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "LimiX (default)", "XGBoost (default)", "ChimeraBoost (tuned)", "TabSwift (default)", "PerpetualBooster (tuned + ensembled)", "TabM (default)", "xRFM (default)", "TorchMLP (tuned + ensembled)", "LightGBM (default)", "ChimeraBoost (default)", "PerpetualBooster (tuned)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "RealMLP (default)", "ExtraTrees (tuned)", "RandomForest (tuned + ensembled)", "SAP-RPT-OSS (default)", "RandomForest (tuned)", "EBM (tuned + ensembled)", "TorchMLP (default)", "EBM (tuned)", "PerpetualBooster (default)", "RandomForest (default)", "ExtraTrees (default)", "EBM (default)", "iLTM (default)", "FastaiMLP (tuned + ensembled)", "FastaiMLP (tuned)", "KNN (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "FastaiMLP (default)", "TabSTAR (tuned)", "KNN (tuned)", "TabSTAR (default)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 1.0, 0.7333333492279053, 0.800000011920929, 1.0, 1.0, 0.6888889074325562, 1.0, 0.8888888955116272, 1.0, 1.0, 0.9555555582046509, 0.8444444537162781, 1.0, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, NaN, 0.2222222238779068, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.6000000238418579, 0.800000011920929, 0.2666666805744171, 0.9111111164093018, 0.7777777910232544, 0.6000000238418579, 0.6000000238418579, 0.8222222328186035, 1.0, 0.6000000238418579, 0.9555555582046509, 0.9555555582046509, 0.9777777791023254, 0.8222222328186035, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.9777777791023254, 0.8444444537162781, 1.0, 0.9777777791023254, 0.9777777791023254, 1.0, 0.9555555582046509, 1.0, 0.8444444537162781, 1.0, 1.0, 0.9555555582046509, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2666666805744171, 0.7777777910232544, NaN, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.42222222685813904, 0.800000011920929, 0.8222222328186035, 0.800000011920929, 0.800000011920929, 0.8444444537162781, 0.5333333611488342, 0.800000011920929, 1.0, 0.6888889074325562, 0.800000011920929, 0.9777777791023254, 0.8222222328186035, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.3333333432674408, 0.20000000298023224, NaN, 0.5555555820465088, 0.46666666865348816, 0.35555556416511536, 0.5555555820465088, 0.20000000298023224, 0.7333333492279053, 0.9777777791023254, 0.35555556416511536, 0.42222222685813904, 0.6666666865348816, 0.5111111402511597, 0.6000000238418579, 0.800000011920929, 0.8888888955116272, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2666666805744171, 0.20000000298023224, 0.4444444477558136, NaN, 0.5333333611488342, 0.5111111402511597, 0.6222222447395325, 0.24444444477558136, 0.8222222328186035, 0.644444465637207, 0.4444444477558136, 0.5777778029441833, 0.6666666865348816, 0.6222222447395325, 0.6000000238418579, 0.9555555582046509, 0.8222222328186035, 0.8888888955116272, 0.8222222328186035, 0.8222222328186035, 0.8222222328186035, 0.8444444537162781, 1.0, 0.9555555582046509, 0.9111111164093018, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.13333334028720856, 0.2222222238779068, 0.5333333611488342, 0.46666666865348816, NaN, 0.5333333611488342, 0.6000000238418579, 0.24444444477558136, 0.800000011920929, 0.6888889074325562, 0.5555555820465088, 0.5777778029441833, 0.7555555701255798, 0.800000011920929, 0.5777778029441833, 0.8444444537162781, 0.9111111164093018, 0.9333333373069763, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 0.8222222328186035, 1.0, 1.0, 0.8222222328186035, 1.0, 1.0, 0.9333333373069763, 1.0, 0.8444444537162781, 1.0, 0.8444444537162781, 1.0, 1.0, 0.8444444537162781, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.31111112236976624, 0.4000000059604645, 0.5777778029441833, 0.644444465637207, 0.4888888895511627, 0.46666666865348816, NaN, 0.46666666865348816, 0.4888888895511627, 0.4888888895511627, 0.7111111283302307, 0.7777777910232544, 0.9555555582046509, 0.6666666865348816, 0.7111111283302307, 0.9777777791023254, 0.5111111402511597, 0.8666666746139526, 0.8888888955116272, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.8222222328186035, 0.9111111164093018, 0.7333333492279053, 0.8888888955116272, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.7777777910232544, 0.9555555582046509, 0.800000011920929, 1.0, 0.8444444537162781, 0.8222222328186035, 0.8222222328186035, 0.9111111164093018, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.4444444477558136, 0.3777777850627899, 0.4000000059604645, 0.5333333611488342, NaN, 0.20000000298023224, 0.46666666865348816, 0.5777778029441833, 0.42222222685813904, 0.5333333611488342, 0.644444465637207, 0.6666666865348816, 0.6000000238418579, 0.8444444537162781, 0.7111111283302307, 0.8444444537162781, 0.7555555701255798, 0.7555555701255798, 0.7555555701255798, 0.8222222328186035, 1.0, 0.8444444537162781, 0.8666666746139526, 0.9777777791023254, 0.9555555582046509, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9777777791023254, 0.9111111164093018, 0.9777777791023254, 0.9333333373069763, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.7333333492279053, 0.17777778208255768, 0.800000011920929, 0.7555555701255798, 0.7555555701255798, 0.5111111402511597, 0.800000011920929, NaN, 0.7777777910232544, 0.800000011920929, 0.6000000238418579, 0.5777778029441833, 0.800000011920929, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.9555555582046509, 0.800000011920929, 0.9555555582046509, 0.800000011920929, 0.800000011920929, 0.8222222328186035, 1.0, 1.0, 1.0, 0.8444444537162781, 1.0, 0.8444444537162781, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.08888889104127884, 0.20000000298023224, 0.2666666805744171, 0.17777778208255768, 0.20000000298023224, 0.5111111402511597, 0.5333333611488342, 0.2222222238779068, NaN, 0.6222222447395325, 0.4444444477558136, 0.5333333611488342, 0.5111111402511597, 0.5555555820465088, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.8888888955116272, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.7555555701255798, 0.7777777910232544, 0.9333333373069763, 0.800000011920929, 0.9777777791023254, 0.9777777791023254, 0.9111111164093018, 1.0, 0.8666666746139526, 1.0, 0.800000011920929, 1.0, 0.9777777791023254, 0.8666666746139526, 0.8666666746139526, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2222222238779068, 0.20000000298023224, 0.02222222276031971, 0.35555556416511536, 0.31111112236976624, 0.2888889014720917, 0.42222222685813904, 0.20000000298023224, 0.3777777850627899, NaN, 0.2666666805744171, 0.3777777850627899, 0.4888888895511627, 0.4000000059604645, 0.5777778029441833, 0.7777777910232544, 0.800000011920929, 0.8444444537162781, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.7111111283302307, 0.8888888955116272, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 1.0, 0.9111111164093018, 1.0, 0.9777777791023254, 1.0, 0.9777777791023254, 0.800000011920929, 1.0, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.04444444552063942, 0.4000000059604645, 0.15555556118488312, 0.644444465637207, 0.5555555820465088, 0.4444444477558136, 0.2222222238779068, 0.5777778029441833, 0.4000000059604645, 0.5555555820465088, 0.7333333492279053, NaN, 0.31111112236976624, 0.5777778029441833, 0.7555555701255798, 0.5555555820465088, 0.6000000238418579, 0.8888888955116272, 0.7333333492279053, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.7777777910232544, 0.9777777791023254, 0.800000011920929, 0.9111111164093018, 0.9777777791023254, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.15555556118488312, 0.4000000059604645, 0.46666666865348816, 0.5777778029441833, 0.42222222685813904, 0.42222222685813904, 0.04444444552063942, 0.46666666865348816, 0.42222222685813904, 0.46666666865348816, 0.6222222447395325, 0.6888889074325562, NaN, 0.5777778029441833, 0.644444465637207, 0.8222222328186035, 0.46666666865348816, 0.8444444537162781, 0.8666666746139526, 0.7111111283302307, 0.7111111283302307, 0.7111111283302307, 0.7333333492279053, 0.7555555701255798, 0.9111111164093018, 0.7111111283302307, 0.8888888955116272, 0.9111111164093018, 0.7777777910232544, 0.9111111164093018, 0.7555555701255798, 0.9333333373069763, 0.800000011920929, 1.0, 0.800000011920929, 0.7777777910232544, 0.8222222328186035, 0.8888888955116272, 0.9555555582046509, 0.9111111164093018, 1.0, 1.0, 0.9777777791023254, 0.8888888955116272, 0.9777777791023254, 1.0, 0.9777777791023254, 0.9555555582046509, 1.0, 1.0, 0.9333333373069763, 1.0, 0.8444444537162781, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.17777778208255768, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.24444444477558136, 0.3333333432674408, 0.35555556416511536, 0.20000000298023224, 0.4888888895511627, 0.5111111402511597, 0.42222222685813904, 0.42222222685813904, NaN, 0.4888888895511627, 0.5333333611488342, 0.4888888895511627, 0.7555555701255798, 0.7333333492279053, 0.7555555701255798, 0.7555555701255798, 0.7555555701255798, 0.7555555701255798, 0.6888889074325562, 0.9333333373069763, 0.7555555701255798, 0.8666666746139526, 0.9555555582046509, 0.8888888955116272, 0.8444444537162781, 0.8888888955116272, 0.9555555582046509, 0.800000011920929, 0.9777777791023254, 0.6222222447395325, 0.8888888955116272, 0.8444444537162781, 0.7333333492279053, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 0.9333333373069763, 0.9777777791023254, 0.9555555582046509, 1.0, 1.0, 1.0, 0.9777777791023254, 0.9777777791023254, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.4888888895511627, 0.3777777850627899, 0.20000000298023224, 0.2888889014720917, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.6000000238418579, 0.24444444477558136, 0.35555556416511536, 0.5111111402511597, NaN, 0.3777777850627899, 0.5777778029441833, 0.6000000238418579, 0.6888889074325562, 0.7555555701255798, 0.800000011920929, 0.800000011920929, 0.7555555701255798, 0.800000011920929, 0.9555555582046509, 0.800000011920929, 0.7777777910232544, 0.9555555582046509, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8888888955116272, 0.8444444537162781, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.9777777791023254, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.8222222328186035, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.4000000059604645, 0.31111112236976624, 0.4000000059604645, 0.4000000059604645, 0.42222222685813904, 0.02222222276031971, 0.4000000059604645, 0.4444444477558136, 0.4000000059604645, 0.42222222685813904, 0.4444444477558136, 0.17777778208255768, 0.46666666865348816, 0.6222222447395325, NaN, 0.42222222685813904, 0.7555555701255798, 0.5777778029441833, 0.6666666865348816, 0.7111111283302307, 0.7111111283302307, 0.7333333492279053, 0.5111111402511597, 0.7111111283302307, 0.7111111283302307, 0.8444444537162781, 0.8444444537162781, 0.7333333492279053, 0.7555555701255798, 0.7333333492279053, 0.8888888955116272, 0.800000011920929, 0.9777777791023254, 0.800000011920929, 0.7555555701255798, 0.800000011920929, 0.8444444537162781, 0.7333333492279053, 0.9111111164093018, 0.9333333373069763, 1.0, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.8222222328186035, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.9777777791023254, 0.8222222328186035, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.20000000298023224, 0.20000000298023224, 0.04444444552063942, 0.15555556118488312, 0.4888888895511627, 0.15555556118488312, 0.2222222238779068, 0.20000000298023224, 0.2222222238779068, 0.4000000059604645, 0.5333333611488342, 0.5111111402511597, 0.42222222685813904, 0.5777778029441833, NaN, 0.42222222685813904, 0.5777778029441833, 0.5555555820465088, 0.5777778029441833, 0.5777778029441833, 0.6000000238418579, 0.7111111283302307, 0.7777777910232544, 0.6000000238418579, 0.800000011920929, 0.7777777910232544, 0.7333333492279053, 0.9777777791023254, 0.7333333492279053, 0.8222222328186035, 0.6222222447395325, 0.800000011920929, 0.9111111164093018, 0.8222222328186035, 0.7333333492279053, 0.7333333492279053, 0.9111111164093018, 0.9111111164093018, 0.8888888955116272, 0.9777777791023254, 0.8888888955116272, 0.8222222328186035, 1.0, 1.0, 0.8444444537162781, 0.9555555582046509, 0.8666666746139526, 0.9777777791023254, 1.0, 1.0, 0.9555555582046509, 1.0, 0.9777777791023254, 1.0, 1.0, 0.9111111164093018, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.02222222276031971, 0.1111111119389534, 0.17777778208255768, 0.08888889104127884, 0.13333334028720856, 0.2888889014720917, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.1111111119389534, 0.15555556118488312, 0.24444444477558136, 0.4000000059604645, 0.24444444477558136, 0.5777778029441833, NaN, 0.6000000238418579, 0.7555555701255798, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.3777777850627899, 0.7555555701255798, 0.7777777910232544, 0.8444444537162781, 0.8666666746139526, 0.800000011920929, 0.6888889074325562, 0.800000011920929, 0.8444444537162781, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.9777777791023254, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.02222222276031971, 0.17777778208255768, 0.06666667014360428, 0.1111111119389534, 0.06666667014360428, 0.1111111119389534, 0.15555556118488312, 0.20000000298023224, 0.1111111119389534, 0.15555556118488312, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.31111112236976624, 0.42222222685813904, 0.42222222685813904, 0.4000000059604645, NaN, 0.6000000238418579, 0.6888889074325562, 0.6888889074325562, 0.6222222447395325, 0.4000000059604645, 0.6222222447395325, 0.7111111283302307, 1.0, 0.7555555701255798, 0.7111111283302307, 0.7777777910232544, 0.7111111283302307, 0.8888888955116272, 0.6888889074325562, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7777777910232544, 0.8888888955116272, 0.9777777791023254, 1.0, 1.0, 0.9777777791023254, 0.7777777910232544, 1.0, 1.0, 0.9555555582046509, 1.0, 0.9777777791023254, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.17777778208255768, 0.20000000298023224, 0.0, 0.17777778208255768, 0.20000000298023224, 0.2666666805744171, 0.24444444477558136, 0.20000000298023224, 0.2222222238779068, 0.06666667014360428, 0.2222222238779068, 0.2888889014720917, 0.24444444477558136, 0.24444444477558136, 0.3333333432674408, 0.4444444477558136, 0.24444444477558136, 0.4000000059604645, NaN, 0.46666666865348816, 0.6000000238418579, 0.7333333492279053, 0.5333333611488342, 0.35555556416511536, 0.8222222328186035, 0.4888888895511627, 0.5111111402511597, 0.9777777791023254, 0.5777778029441833, 0.9555555582046509, 0.7555555701255798, 0.9777777791023254, 0.7777777910232544, 0.6000000238418579, 1.0, 1.0, 0.6888889074325562, 1.0, 0.9555555582046509, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.17777778208255768, 0.2222222238779068, 0.2666666805744171, 0.24444444477558136, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.2888889014720917, 0.24444444477558136, 0.20000000298023224, 0.2888889014720917, 0.42222222685813904, 0.2222222238779068, 0.31111112236976624, 0.5333333611488342, NaN, 0.6666666865348816, 0.644444465637207, 0.5333333611488342, 0.2666666805744171, 0.8222222328186035, 0.4888888895511627, 0.3777777850627899, 0.8444444537162781, 0.5555555820465088, 0.9111111164093018, 0.644444465637207, 0.9111111164093018, 0.7555555701255798, 0.6000000238418579, 0.9777777791023254, 0.9555555582046509, 0.6888889074325562, 0.9777777791023254, 0.9333333373069763, 0.7777777910232544, 0.800000011920929, 0.9777777791023254, 1.0, 1.0, 0.800000011920929, 0.9333333373069763, 1.0, 0.9555555582046509, 1.0, 0.800000011920929, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.17777778208255768, 0.20000000298023224, 0.2666666805744171, 0.24444444477558136, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.2888889014720917, 0.24444444477558136, 0.20000000298023224, 0.2888889014720917, 0.42222222685813904, 0.2222222238779068, 0.31111112236976624, 0.4000000059604645, 0.3333333432674408, NaN, 0.5777778029441833, 0.5333333611488342, 0.2666666805744171, 0.7111111283302307, 0.4888888895511627, 0.3777777850627899, 0.7333333492279053, 0.5555555820465088, 0.7555555701255798, 0.644444465637207, 0.9111111164093018, 0.6888889074325562, 0.6000000238418579, 0.9777777791023254, 0.9555555582046509, 0.6888889074325562, 0.9777777791023254, 0.9333333373069763, 0.800000011920929, 0.800000011920929, 0.9777777791023254, 0.9777777791023254, 1.0, 0.800000011920929, 0.9333333373069763, 1.0, 0.9555555582046509, 0.9777777791023254, 0.800000011920929, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.15555556118488312, 0.17777778208255768, 0.2666666805744171, 0.17777778208255768, 0.20000000298023224, 0.24444444477558136, 0.06666667014360428, 0.2222222238779068, 0.2666666805744171, 0.24444444477558136, 0.24444444477558136, 0.2666666805744171, 0.4000000059604645, 0.2222222238779068, 0.3777777850627899, 0.2666666805744171, 0.35555556416511536, 0.42222222685813904, NaN, 0.46666666865348816, 0.24444444477558136, 0.5555555820465088, 0.42222222685813904, 0.3333333432674408, 0.9333333373069763, 0.4888888895511627, 1.0, 0.6222222447395325, 0.9333333373069763, 0.800000011920929, 0.6000000238418579, 1.0, 0.9777777791023254, 0.6000000238418579, 0.9777777791023254, 0.9111111164093018, 0.7777777910232544, 0.800000011920929, 1.0, 0.9777777791023254, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.04444444552063942, 0.0, 0.0, 0.17777778208255768, 0.0, 0.20000000298023224, 0.2222222238779068, 0.2888889014720917, 0.2222222238779068, 0.24444444477558136, 0.31111112236976624, 0.20000000298023224, 0.4888888895511627, 0.2888889014720917, 0.6222222447395325, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, NaN, 0.6222222447395325, 0.46666666865348816, 0.6666666865348816, 0.7111111283302307, 0.6000000238418579, 0.8666666746139526, 0.6000000238418579, 0.7555555701255798, 0.6000000238418579, 0.800000011920929, 0.7555555701255798, 0.6888889074325562, 0.6666666865348816, 0.8222222328186035, 0.8222222328186035, 0.8888888955116272, 0.8444444537162781, 0.9333333373069763, 0.8222222328186035, 0.6666666865348816, 0.9333333373069763, 1.0, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 0.9111111164093018, 0.8888888955116272, 0.9777777791023254, 0.8222222328186035, 1.0, 0.9333333373069763, 1.0, 0.9777777791023254, 0.8888888955116272, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.02222222276031971, 0.0, 0.0, 0.04444444552063942, 0.0, 0.08888889104127884, 0.15555556118488312, 0.0, 0.06666667014360428, 0.1111111119389534, 0.02222222276031971, 0.08888889104127884, 0.06666667014360428, 0.04444444552063942, 0.2888889014720917, 0.2222222238779068, 0.24444444477558136, 0.3777777850627899, 0.644444465637207, 0.7333333492279053, 0.7333333492279053, 0.7555555701255798, 0.3777777850627899, NaN, 0.7333333492279053, 0.4444444477558136, 0.9555555582046509, 0.7777777910232544, 0.5111111402511597, 0.7777777910232544, 0.6888889074325562, 0.7777777910232544, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 1.0, 0.8444444537162781, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 0.8444444537162781, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.15555556118488312, 0.20000000298023224, 0.0, 0.08888889104127884, 0.17777778208255768, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.20000000298023224, 0.2888889014720917, 0.24444444477558136, 0.20000000298023224, 0.2888889014720917, 0.4000000059604645, 0.2222222238779068, 0.2888889014720917, 0.17777778208255768, 0.17777778208255768, 0.2888889014720917, 0.4444444477558136, 0.5333333611488342, 0.2666666805744171, NaN, 0.4888888895511627, 0.4000000059604645, 0.6666666865348816, 0.5555555820465088, 0.7111111283302307, 0.644444465637207, 0.6222222447395325, 0.6888889074325562, 0.6000000238418579, 0.800000011920929, 0.7555555701255798, 0.6888889074325562, 0.8888888955116272, 0.7333333492279053, 0.7777777910232544, 0.800000011920929, 0.7555555701255798, 0.7777777910232544, 0.9333333373069763, 0.800000011920929, 0.7111111283302307, 0.7777777910232544, 0.7111111283302307, 0.800000011920929, 0.6222222447395325, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.1111111119389534, 0.02222222276031971, 0.20000000298023224, 0.02222222276031971, 0.02222222276031971, 0.08888889104127884, 0.1111111119389534, 0.13333334028720856, 0.2222222238779068, 0.15555556118488312, 0.20000000298023224, 0.15555556118488312, 0.0, 0.5111111402511597, 0.5111111402511597, 0.5111111402511597, 0.5777778029441833, 0.3333333432674408, 0.5555555820465088, 0.5111111402511597, NaN, 0.5777778029441833, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7111111283302307, 0.6222222447395325, 0.800000011920929, 0.6000000238418579, 0.7111111283302307, 0.644444465637207, 0.7111111283302307, 0.7333333492279053, 0.9111111164093018, 0.9111111164093018, 0.9555555582046509, 0.8666666746139526, 0.644444465637207, 0.8666666746139526, 0.9777777791023254, 0.800000011920929, 0.9555555582046509, 0.8666666746139526, 0.9555555582046509, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.9777777791023254, 1.0, 0.9555555582046509, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.02222222276031971, 0.0, 0.0, 0.02222222276031971, 0.0, 0.06666667014360428, 0.04444444552063942, 0.0, 0.02222222276031971, 0.04444444552063942, 0.02222222276031971, 0.08888889104127884, 0.04444444552063942, 0.04444444552063942, 0.15555556118488312, 0.2222222238779068, 0.13333334028720856, 0.24444444477558136, 0.4888888895511627, 0.6222222447395325, 0.6222222447395325, 0.6666666865348816, 0.2888889014720917, 0.04444444552063942, 0.6000000238418579, 0.42222222685813904, NaN, 0.7333333492279053, 0.4000000059604645, 0.7333333492279053, 0.46666666865348816, 0.7333333492279053, 0.7777777910232544, 0.6000000238418579, 0.7555555701255798, 0.7333333492279053, 0.6222222447395325, 0.800000011920929, 0.9777777791023254, 0.8444444537162781, 0.6000000238418579, 0.7555555701255798, 0.7555555701255798, 0.800000011920929, 0.7111111283302307, 0.9555555582046509, 1.0, 1.0, 0.800000011920929, 0.8222222328186035, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.08888889104127884, 0.0, 0.20000000298023224, 0.2222222238779068, 0.1111111119389534, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2888889014720917, 0.02222222276031971, 0.15555556118488312, 0.2666666805744171, 0.06666667014360428, 0.4000000059604645, 0.2222222238779068, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, NaN, 0.4444444477558136, 0.5333333611488342, 0.4444444477558136, 0.5555555820465088, 0.7333333492279053, 0.6000000238418579, 0.7777777910232544, 0.5777778029441833, 0.5777778029441833, 0.8666666746139526, 0.6888889074325562, 0.7555555701255798, 0.800000011920929, 0.9777777791023254, 0.8444444537162781, 0.9555555582046509, 0.800000011920929, 0.8666666746139526, 0.9111111164093018, 0.9111111164093018, 1.0, 0.800000011920929, 1.0, 0.9777777791023254, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.06666667014360428, 0.0, 0.20000000298023224, 0.0, 0.08888889104127884, 0.06666667014360428, 0.08888889104127884, 0.15555556118488312, 0.20000000298023224, 0.24444444477558136, 0.02222222276031971, 0.31111112236976624, 0.2222222238779068, 0.42222222685813904, 0.4444444477558136, 0.4444444477558136, 0.5111111402511597, 0.13333334028720856, 0.4888888895511627, 0.4444444477558136, 0.4000000059604645, 0.6000000238418579, 0.5555555820465088, NaN, 0.5777778029441833, 0.5333333611488342, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6222222447395325, 0.6222222447395325, 0.7555555701255798, 0.800000011920929, 0.8888888955116272, 0.7777777910232544, 0.6222222447395325, 0.800000011920929, 0.9777777791023254, 0.800000011920929, 0.8444444537162781, 0.800000011920929, 0.8888888955116272, 0.8666666746139526, 0.9555555582046509, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.9555555582046509, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.20000000298023224, 0.0, 0.0, 0.15555556118488312, 0.2222222238779068, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.0, 0.20000000298023224, 0.24444444477558136, 0.1111111119389534, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2888889014720917, 0.04444444552063942, 0.08888889104127884, 0.24444444477558136, 0.0, 0.4000000059604645, 0.2222222238779068, 0.2888889014720917, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.42222222685813904, NaN, 0.3777777850627899, 0.5333333611488342, 0.644444465637207, 0.6000000238418579, 0.800000011920929, 0.5555555820465088, 0.5555555820465088, 0.7555555701255798, 0.644444465637207, 0.6888889074325562, 0.7777777910232544, 0.8666666746139526, 0.8222222328186035, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.8888888955116272, 0.9111111164093018, 1.0, 0.800000011920929, 1.0, 0.9777777791023254, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.04444444552063942, 0.02222222276031971, 0.20000000298023224, 0.0, 0.02222222276031971, 0.08888889104127884, 0.06666667014360428, 0.04444444552063942, 0.20000000298023224, 0.1111111119389534, 0.17777778208255768, 0.15555556118488312, 0.1111111119389534, 0.24444444477558136, 0.35555556416511536, 0.35555556416511536, 0.3777777850627899, 0.24444444477558136, 0.31111112236976624, 0.35555556416511536, 0.2888889014720917, 0.5333333611488342, 0.5555555820465088, 0.46666666865348816, 0.6222222447395325, NaN, 0.5111111402511597, 0.8444444537162781, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5777778029441833, 0.644444465637207, 0.7111111283302307, 0.9333333373069763, 0.9111111164093018, 0.8888888955116272, 0.5777778029441833, 0.8666666746139526, 0.9555555582046509, 0.8888888955116272, 0.9111111164093018, 0.8888888955116272, 0.9777777791023254, 0.7777777910232544, 0.9777777791023254, 0.7777777910232544, 0.9777777791023254, 0.7777777910232544, 0.9777777791023254, 0.9777777791023254, 0.9111111164093018, 0.7777777910232544, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254], [0.0, 0.15555556118488312, 0.20000000298023224, 0.0, 0.04444444552063942, 0.15555556118488312, 0.20000000298023224, 0.08888889104127884, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.20000000298023224, 0.31111112236976624, 0.02222222276031971, 0.08888889104127884, 0.08888889104127884, 0.06666667014360428, 0.4000000059604645, 0.2222222238779068, 0.3777777850627899, 0.3777777850627899, 0.2666666805744171, 0.4444444477558136, 0.4000000059604645, 0.46666666865348816, 0.4888888895511627, NaN, 0.5111111402511597, 0.4000000059604645, 0.7777777910232544, 0.8222222328186035, 0.6000000238418579, 0.7777777910232544, 0.5777778029441833, 0.5777778029441833, 0.6000000238418579, 0.7777777910232544, 0.9555555582046509, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.7555555701255798, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.644444465637207, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.02222222276031971, 0.0, 0.0, 0.02222222276031971, 0.1111111119389534, 0.02222222276031971, 0.20000000298023224, 0.0, 0.06666667014360428, 0.2222222238779068, 0.24444444477558136, 0.31111112236976624, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.31111112236976624, 0.20000000298023224, 0.2222222238779068, 0.2666666805744171, 0.20000000298023224, 0.35555556416511536, 0.15555556118488312, 0.4888888895511627, NaN, 0.6000000238418579, 0.4888888895511627, 0.4888888895511627, 0.4888888895511627, 0.4444444477558136, 0.6000000238418579, 0.8444444537162781, 0.7111111283302307, 0.7555555701255798, 0.5777778029441833, 0.6222222447395325, 0.8222222328186035, 0.7777777910232544, 0.8666666746139526, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.8222222328186035, 1.0, 0.9111111164093018, 0.800000011920929, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.15555556118488312, 0.06666667014360428, 0.0, 0.02222222276031971, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.15555556118488312, 0.20000000298023224, 0.08888889104127884, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.24444444477558136, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4444444477558136, 0.6000000238418579, 0.4000000059604645, NaN, 0.46666666865348816, 0.6000000238418579, 0.5777778029441833, 0.42222222685813904, 0.4444444477558136, 0.4444444477558136, 0.5333333611488342, 0.42222222685813904, 0.6000000238418579, 0.46666666865348816, 0.644444465637207, 0.4000000059604645, 0.6888889074325562, 0.4000000059604645, 0.6000000238418579, 0.9777777791023254, 0.644444465637207, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.7111111283302307, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.20000000298023224, 0.0, 0.0, 0.15555556118488312, 0.17777778208255768, 0.02222222276031971, 0.20000000298023224, 0.13333334028720856, 0.0, 0.20000000298023224, 0.2222222238779068, 0.1111111119389534, 0.20000000298023224, 0.24444444477558136, 0.17777778208255768, 0.20000000298023224, 0.20000000298023224, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.31111112236976624, 0.20000000298023224, 0.20000000298023224, 0.2888889014720917, 0.24444444477558136, 0.2222222238779068, 0.4000000059604645, 0.20000000298023224, 0.3333333432674408, 0.2222222238779068, 0.5111111402511597, 0.5333333611488342, NaN, 0.3333333432674408, 0.4888888895511627, 0.4888888895511627, 0.35555556416511536, 0.5333333611488342, 0.7111111283302307, 0.46666666865348816, 0.5333333611488342, 0.7777777910232544, 0.7555555701255798, 0.4444444477558136, 0.6000000238418579, 0.46666666865348816, 0.7777777910232544, 0.6222222447395325, 0.800000011920929, 0.9777777791023254, 0.800000011920929, 1.0, 1.0, 1.0, 0.7777777910232544, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.20000000298023224, 0.0, 0.02222222276031971, 0.1111111119389534, 0.17777778208255768, 0.02222222276031971, 0.20000000298023224, 0.13333334028720856, 0.0, 0.20000000298023224, 0.17777778208255768, 0.15555556118488312, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.0, 0.04444444552063942, 0.04444444552063942, 0.02222222276031971, 0.3333333432674408, 0.20000000298023224, 0.24444444477558136, 0.35555556416511536, 0.2666666805744171, 0.42222222685813904, 0.4000000059604645, 0.4444444477558136, 0.4444444477558136, 0.17777778208255768, 0.5111111402511597, 0.4000000059604645, 0.6666666865348816, NaN, 0.6000000238418579, 0.7333333492279053, 0.5333333611488342, 0.5555555820465088, 0.6000000238418579, 0.7333333492279053, 0.8444444537162781, 0.7777777910232544, 0.6000000238418579, 0.6888889074325562, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.7555555701255798, 0.6222222447395325, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8888888955116272, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.04444444552063942, 0.0, 0.04444444552063942, 0.15555556118488312, 0.0, 0.1111111119389534, 0.2666666805744171, 0.0, 0.15555556118488312, 0.2666666805744171, 0.2222222238779068, 0.2222222238779068, 0.31111112236976624, 0.31111112236976624, 0.31111112236976624, 0.4000000059604645, 0.17777778208255768, 0.3333333432674408, 0.31111112236976624, 0.2888889014720917, 0.3777777850627899, 0.42222222685813904, 0.3777777850627899, 0.4444444477558136, 0.42222222685813904, 0.4000000059604645, 0.5111111402511597, 0.42222222685813904, 0.5111111402511597, 0.4000000059604645, NaN, 0.46666666865348816, 0.3777777850627899, 0.6000000238418579, 0.6222222447395325, 0.5333333611488342, 0.42222222685813904, 0.5333333611488342, 0.6222222447395325, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6222222447395325, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.6888889074325562, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.04444444552063942, 0.02222222276031971, 0.20000000298023224, 0.2666666805744171, 0.08888889104127884, 0.20000000298023224, 0.1111111119389534, 0.0, 0.02222222276031971, 0.02222222276031971, 0.02222222276031971, 0.17777778208255768, 0.20000000298023224, 0.1111111119389534, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.3777777850627899, 0.24444444477558136, 0.35555556416511536, 0.2222222238779068, 0.5555555820465088, 0.5777778029441833, 0.5111111402511597, 0.2666666805744171, 0.5333333611488342, NaN, 0.3777777850627899, 0.6222222447395325, 0.7555555701255798, 0.5555555820465088, 0.3777777850627899, 0.9777777791023254, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.7555555701255798, 0.644444465637207, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8444444537162781, 1.0, 0.9555555582046509, 0.9333333373069763, 0.8888888955116272, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.02222222276031971, 0.0, 0.08888889104127884, 0.08888889104127884, 0.0, 0.02222222276031971, 0.04444444552063942, 0.06666667014360428, 0.06666667014360428, 0.08888889104127884, 0.1111111119389534, 0.0, 0.2666666805744171, 0.08888889104127884, 0.02222222276031971, 0.31111112236976624, 0.24444444477558136, 0.35555556416511536, 0.2888889014720917, 0.42222222685813904, 0.4000000059604645, 0.5555555820465088, 0.644444465637207, 0.46666666865348816, 0.6222222447395325, 0.6222222447395325, NaN, 0.42222222685813904, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.6888889074325562, 0.6000000238418579, 0.6222222447395325, 0.8222222328186035, 0.6222222447395325, 0.7333333492279053, 0.800000011920929, 0.7555555701255798, 0.800000011920929, 0.6222222447395325, 0.800000011920929, 0.800000011920929, 0.9555555582046509, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.06666667014360428, 0.1111111119389534, 0.0, 0.0, 0.20000000298023224, 0.2222222238779068, 0.20000000298023224, 0.2222222238779068, 0.15555556118488312, 0.15555556118488312, 0.2222222238779068, 0.08888889104127884, 0.15555556118488312, 0.24444444477558136, 0.20000000298023224, 0.31111112236976624, 0.06666667014360428, 0.42222222685813904, 0.15555556118488312, 0.5555555820465088, 0.46666666865348816, 0.4444444477558136, 0.4000000059604645, 0.3777777850627899, 0.5777778029441833, NaN, 0.3333333432674408, 0.42222222685813904, 0.4444444477558136, 0.5333333611488342, 0.46666666865348816, 0.7777777910232544, 0.6888889074325562, 0.8222222328186035, 0.7333333492279053, 0.800000011920929, 0.7555555701255798, 0.6222222447395325, 0.9555555582046509, 0.644444465637207, 0.7777777910232544, 0.9555555582046509, 0.8888888955116272, 0.7333333492279053, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.02222222276031971, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.2222222238779068, 0.20000000298023224, 0.04444444552063942, 0.4000000059604645, 0.20000000298023224, 0.1111111119389534, 0.2222222238779068, 0.08888889104127884, 0.4000000059604645, 0.2888889014720917, 0.46666666865348816, 0.2888889014720917, 0.4000000059604645, 0.3777777850627899, 0.24444444477558136, 0.46666666865348816, 0.6666666865348816, NaN, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.9777777791023254, 0.6666666865348816, 0.5777778029441833, 0.6666666865348816, 0.7777777910232544, 0.800000011920929, 0.8444444537162781, 0.7111111283302307, 1.0, 0.7777777910232544, 0.7777777910232544, 0.8444444537162781, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.9111111164093018, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.04444444552063942, 0.0, 0.0, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.06666667014360428, 0.1111111119389534, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.17777778208255768, 0.20000000298023224, 0.24444444477558136, 0.13333334028720856, 0.24444444477558136, 0.02222222276031971, 0.2222222238779068, 0.13333334028720856, 0.1111111119389534, 0.2222222238779068, 0.24444444477558136, 0.5777778029441833, 0.5333333611488342, 0.2666666805744171, 0.46666666865348816, 0.4444444477558136, 0.4000000059604645, 0.5777778029441833, 0.6000000238418579, NaN, 0.5111111402511597, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.7111111283302307, 0.6222222447395325, 0.6666666865348816, 0.7555555701255798, 0.644444465637207, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.06666667014360428, 0.02222222276031971, 0.20000000298023224, 0.04444444552063942, 0.0, 0.20000000298023224, 0.1111111119389534, 0.04444444552063942, 0.20000000298023224, 0.20000000298023224, 0.17777778208255768, 0.20000000298023224, 0.2222222238779068, 0.0, 0.0, 0.02222222276031971, 0.02222222276031971, 0.3333333432674408, 0.20000000298023224, 0.2222222238779068, 0.35555556416511536, 0.24444444477558136, 0.15555556118488312, 0.3777777850627899, 0.17777778208255768, 0.42222222685813904, 0.04444444552063942, 0.42222222685813904, 0.4000000059604645, 0.46666666865348816, 0.15555556118488312, 0.5777778029441833, 0.6222222447395325, 0.46666666865348816, 0.5555555820465088, 0.6000000238418579, 0.4888888895511627, NaN, 0.7555555701255798, 0.6000000238418579, 0.4444444477558136, 0.5777778029441833, 0.46666666865348816, 0.5555555820465088, 0.42222222685813904, 0.6000000238418579, 0.6888889074325562, 0.6222222447395325, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8444444537162781, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.04444444552063942, 0.02222222276031971, 0.02222222276031971, 0.20000000298023224, 0.06666667014360428, 0.0, 0.08888889104127884, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.04444444552063942, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.3777777850627899, 0.5333333611488342, 0.2222222238779068, 0.2222222238779068, 0.46666666865348816, 0.02222222276031971, 0.31111112236976624, 0.46666666865348816, 0.6666666865348816, 0.46666666865348816, 0.24444444477558136, NaN, 0.7111111283302307, 0.5333333611488342, 0.5111111402511597, 0.5555555820465088, 0.6222222447395325, 0.6222222447395325, 0.7555555701255798, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9777777791023254, 0.8888888955116272, 0.7555555701255798, 0.800000011920929, 1.0, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.17777778208255768, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.2888889014720917, 0.20000000298023224, 0.02222222276031971, 0.20000000298023224, 0.04444444552063942, 0.4000000059604645, 0.17777778208255768, 0.35555556416511536, 0.24444444477558136, 0.4000000059604645, 0.3777777850627899, 0.20000000298023224, 0.4000000059604645, 0.5333333611488342, 0.02222222276031971, 0.4000000059604645, 0.4000000059604645, 0.2888889014720917, NaN, 0.6222222447395325, 0.46666666865348816, 0.6222222447395325, 0.6222222447395325, 0.800000011920929, 0.644444465637207, 0.644444465637207, 1.0, 0.6666666865348816, 0.7111111283302307, 0.7111111283302307, 0.644444465637207, 0.7555555701255798, 0.8888888955116272, 0.8888888955116272, 0.9555555582046509, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.06666667014360428, 0.15555556118488312, 0.0, 0.04444444552063942, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.20000000298023224, 0.0, 0.2888889014720917, 0.20000000298023224, 0.04444444552063942, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.1111111119389534, 0.2666666805744171, 0.2222222238779068, 0.6000000238418579, 0.5555555820465088, 0.31111112236976624, 0.4000000059604645, 0.4000000059604645, 0.3777777850627899, 0.2222222238779068, 0.3333333432674408, 0.4000000059604645, 0.5555555820465088, 0.46666666865348816, 0.3777777850627899, NaN, 0.5111111402511597, 0.9111111164093018, 0.6000000238418579, 0.800000011920929, 0.4888888895511627, 0.6000000238418579, 0.42222222685813904, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.04444444552063942, 0.0, 0.06666667014360428, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.0, 0.2222222238779068, 0.04444444552063942, 0.0, 0.08888889104127884, 0.15555556118488312, 0.1111111119389534, 0.08888889104127884, 0.2666666805744171, 0.13333334028720856, 0.31111112236976624, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.17777778208255768, 0.31111112236976624, 0.42222222685813904, 0.2888889014720917, 0.42222222685813904, 0.4888888895511627, 0.5333333611488342, 0.4888888895511627, NaN, 0.5333333611488342, 0.644444465637207, 0.644444465637207, 0.6888889074325562, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.7555555701255798, 0.8444444537162781, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.8444444537162781, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.02222222276031971, 0.0, 0.04444444552063942, 0.04444444552063942, 0.0, 0.20000000298023224, 0.0, 0.2888889014720917, 0.13333334028720856, 0.0, 0.08888889104127884, 0.20000000298023224, 0.08888889104127884, 0.1111111119389534, 0.24444444477558136, 0.20000000298023224, 0.6000000238418579, 0.5333333611488342, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.3777777850627899, 0.17777778208255768, 0.3333333432674408, 0.3777777850627899, 0.5333333611488342, 0.4444444477558136, 0.3777777850627899, 0.08888889104127884, 0.46666666865348816, NaN, 0.46666666865348816, 0.800000011920929, 0.4444444477558136, 0.6000000238418579, 0.42222222685813904, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.08888889104127884, 0.20000000298023224, 0.20000000298023224, 0.04444444552063942, 0.20000000298023224, 0.0, 0.1111111119389534, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.4000000059604645, 0.2222222238779068, 0.20000000298023224, 0.4000000059604645, 0.24444444477558136, 0.2666666805744171, 0.2666666805744171, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3777777850627899, 0.3777777850627899, 0.4000000059604645, 0.35555556416511536, 0.5333333611488342, NaN, 0.7777777910232544, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.9777777791023254, 0.644444465637207, 0.6666666865348816, 0.8888888955116272, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.08888889104127884, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.04444444552063942, 0.06666667014360428, 0.06666667014360428, 0.0, 0.17777778208255768, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1111111119389534, 0.15555556118488312, 0.3777777850627899, 0.20000000298023224, 0.17777778208255768, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.2222222238779068, 0.4000000059604645, 0.20000000298023224, 0.02222222276031971, 0.3777777850627899, 0.4000000059604645, 0.2666666805744171, 0.35555556416511536, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.24444444477558136, 0.5777778029441833, 0.3777777850627899, 0.20000000298023224, 0.20000000298023224, 0.35555556416511536, 0.20000000298023224, 0.2222222238779068, NaN, 0.2222222238779068, 0.7555555701255798, 0.5111111402511597, 0.800000011920929, 0.6000000238418579, 0.42222222685813904, 0.6000000238418579, 0.800000011920929, 0.5777778029441833, 0.8222222328186035, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.04444444552063942, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.35555556416511536, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.20000000298023224, 0.24444444477558136, 0.24444444477558136, 0.15555556118488312, 0.35555556416511536, 0.4000000059604645, 0.24444444477558136, 0.35555556416511536, 0.5111111402511597, 0.31111112236976624, 0.5555555820465088, 0.4000000059604645, 0.7777777910232544, NaN, 0.6000000238418579, 0.7111111283302307, 0.6666666865348816, 0.7111111283302307, 0.7111111283302307, 0.6222222447395325, 0.6666666865348816, 0.8888888955116272, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.20000000298023224, 0.04444444552063942, 0.0, 0.20000000298023224, 0.15555556118488312, 0.04444444552063942, 0.20000000298023224, 0.20000000298023224, 0.04444444552063942, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.17777778208255768, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.20000000298023224, 0.02222222276031971, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.24444444477558136, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.2888889014720917, 0.20000000298023224, 0.31111112236976624, 0.20000000298023224, 0.35555556416511536, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.24444444477558136, 0.4000000059604645, NaN, 0.6666666865348816, 1.0, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 1.0, 0.4888888895511627, 0.6000000238418579, 0.6000000238418579, 0.6888889074325562, 0.8888888955116272, 0.7555555701255798, 0.9777777791023254, 0.7777777910232544, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.08888889104127884, 0.08888889104127884, 0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.08888889104127884, 0.0, 0.08888889104127884, 0.02222222276031971, 0.35555556416511536, 0.0, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.2222222238779068, 0.20000000298023224, 0.3777777850627899, 0.04444444552063942, 0.0, 0.20000000298023224, 0.3777777850627899, 0.20000000298023224, 0.0, 0.5777778029441833, 0.4000000059604645, 0.5777778029441833, 0.2666666805744171, 0.4888888895511627, 0.2888889014720917, 0.3333333432674408, NaN, 0.46666666865348816, 0.4000000059604645, 0.4888888895511627, 0.6000000238418579, 0.46666666865348816, 0.5777778029441833, 0.7555555701255798, 0.7777777910232544, 0.9111111164093018, 0.8222222328186035, 0.800000011920929, 0.8444444537162781, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.02222222276031971, 0.0, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.20000000298023224, 0.15555556118488312, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.15555556118488312, 0.20000000298023224, 0.35555556416511536, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.0, 0.5333333611488342, NaN, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.9333333373069763, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.8666666746139526, 0.7555555701255798, 0.9555555582046509, 0.7555555701255798, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.17777778208255768, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.17777778208255768, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.2222222238779068, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.2888889014720917, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.4000000059604645, 0.2888889014720917, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, NaN, 0.2888889014720917, 0.4000000059604645, 0.6222222447395325, 0.7333333492279053, 0.5777778029441833, 0.7111111283302307, 0.6888889074325562, 1.0, 0.9333333373069763, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.31111112236976624, 0.04444444552063942, 0.04444444552063942, 0.04444444552063942, 0.15555556118488312, 0.0, 0.20000000298023224, 0.1111111119389534, 0.2888889014720917, 0.20000000298023224, 0.24444444477558136, 0.20000000298023224, 0.02222222276031971, 0.5777778029441833, 0.2888889014720917, 0.6000000238418579, 0.5111111402511597, 0.6666666865348816, 0.7111111283302307, NaN, 0.5555555820465088, 0.6666666865348816, 0.6888889074325562, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.08888889104127884, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.02222222276031971, 0.04444444552063942, 0.0, 0.0, 0.20000000298023224, 0.0, 0.08888889104127884, 0.20000000298023224, 0.08888889104127884, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.1111111119389534, 0.2666666805744171, 0.06666667014360428, 0.2222222238779068, 0.24444444477558136, 0.35555556416511536, 0.0, 0.15555556118488312, 0.0, 0.35555556416511536, 0.4000000059604645, 0.3777777850627899, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4444444477558136, NaN, 0.6000000238418579, 0.7777777910232544, 0.6000000238418579, 0.644444465637207, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.15555556118488312, 0.02222222276031971, 0.0, 0.20000000298023224, 0.17777778208255768, 0.02222222276031971, 0.20000000298023224, 0.08888889104127884, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.1111111119389534, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.24444444477558136, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.0, 0.5333333611488342, 0.06666667014360428, 0.3777777850627899, 0.3333333432674408, 0.4000000059604645, NaN, 0.4000000059604645, 0.5555555820465088, 0.6000000238418579, 0.6000000238418579, 0.8444444537162781, 0.6666666865348816, 0.8888888955116272, 0.644444465637207, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.20000000298023224, 0.02222222276031971, 0.2888889014720917, 0.02222222276031971, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.02222222276031971, 0.06666667014360428, 0.0, 0.20000000298023224, 0.0, 0.1111111119389534, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1111111119389534, 0.42222222685813904, 0.1111111119389534, 0.5111111402511597, 0.42222222685813904, 0.6000000238418579, 0.2666666805744171, 0.31111112236976624, 0.2222222238779068, 0.6000000238418579, NaN, 0.5333333611488342, 0.5777778029441833, 0.644444465637207, 0.8444444537162781, 0.8222222328186035, 0.8444444537162781, 0.800000011920929, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.0, 0.20000000298023224, 0.02222222276031971, 0.1111111119389534, 0.20000000298023224, 0.0, 0.20000000298023224, 0.06666667014360428, 0.17777778208255768, 0.13333334028720856, 0.4000000059604645, 0.24444444477558136, 0.4000000059604645, 0.42222222685813904, 0.13333334028720856, 0.4000000059604645, 0.4444444477558136, 0.46666666865348816, NaN, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 1.0, 0.800000011920929, 0.5777778029441833, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.02222222276031971, 0.04444444552063942, 0.20000000298023224, 0.0, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.4000000059604645, 0.2222222238779068, 0.4000000059604645, 0.2888889014720917, 0.06666667014360428, 0.35555556416511536, 0.4000000059604645, 0.42222222685813904, 0.20000000298023224, NaN, 0.4000000059604645, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 0.5777778029441833, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.08888889104127884, 0.0, 0.0, 0.0, 0.1111111119389534, 0.02222222276031971, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.15555556118488312, 0.04444444552063942, 0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.20000000298023224, 0.0, 0.31111112236976624, 0.08888889104127884, 0.4000000059604645, 0.31111112236976624, 0.06666667014360428, 0.2666666805744171, 0.4000000059604645, 0.35555556416511536, 0.6000000238418579, 0.6000000238418579, NaN, 0.4888888895511627, 0.6000000238418579, 0.4888888895511627, 1.0, 0.5777778029441833, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.1111111119389534, 0.17777778208255768, 0.13333334028720856, 0.0, 0.02222222276031971, 0.2666666805744171, 0.15555556118488312, 0.15555556118488312, 0.20000000298023224, 0.20000000298023224, 0.5111111402511597, NaN, 0.24444444477558136, 0.7555555701255798, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.02222222276031971, 0.24444444477558136, 0.20000000298023224, 0.24444444477558136, 0.06666667014360428, 0.02222222276031971, 0.06666667014360428, 0.3333333432674408, 0.17777778208255768, 0.0, 0.08888889104127884, 0.4000000059604645, 0.7555555701255798, NaN, 0.7555555701255798, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.02222222276031971, 0.15555556118488312, 0.04444444552063942, 0.0, 0.02222222276031971, 0.2666666805744171, 0.1111111119389534, 0.15555556118488312, 0.20000000298023224, 0.20000000298023224, 0.5111111402511597, 0.24444444477558136, 0.24444444477558136, NaN, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.0, 0.24444444477558136, 0.2222222238779068, 0.0, 0.20000000298023224, 0.35555556416511536, 0.20000000298023224, 0.42222222685813904, 0.42222222685813904, 0.0, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, NaN, 0.4000000059604645, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.15555556118488312, 0.0, 0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.42222222685813904, 0.0, 0.20000000298023224, 0.0, 0.6000000238418579, NaN, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.13333334028720856, 0.0, 0.0, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, NaN, 0.800000011920929, 0.800000011920929, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, NaN, 0.8777777552604675, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.12222222238779068, NaN, 0.9111111164093018], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.08888889104127884, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9827856035},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9023474218},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8910798179},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8785602536},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8723004737},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8669796605},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.859154935},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8441314599},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8413145621},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8259780962},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8250391267},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.8219092401},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.8206572829},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.791236312},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.769640069},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7608763765},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7339593188},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.7236306787},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.7064162815},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7051643241},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6901408508},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6757433548},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6751173753},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6685446079},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6435054839},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6162754383},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.6159624495},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.595931149},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5921752797},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5865414786},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5824726192},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5705790367},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5217527473},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5076682363},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.5054773151},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4913928066},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4898278652},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.4885759082},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4798122131},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4535211337},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.4425665151},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4384976571},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4331768448},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4269170652},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.4087636984},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3940532134},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3884194113},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.3683881119},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3661971882},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3589984394},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3496087691},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3424100201},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3211267646},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2857590025},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.2841940576},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2760563425},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2748043856},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2622848248},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.2604068904},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2212832591},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.1934272321},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.1752738676},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1605633828},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.1289514888},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1258215979},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.1183098609},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1154929588},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.0854460106},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0682316129},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0295774645},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0176838812},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0031298905}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_regression/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..1edad74d2600c0362051b2a0519b19b9032d8962
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,73 @@
+,TabFM (default),RealTabPFN-2.5 (tuned + ensembled),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),RealTabPFN-2.5 (tuned),TabDPT (tuned + ensembled),ModernNCA (tuned + ensembled),TabICLv2 (default),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabDPT-Turbo (default),TabDPT (tuned),Nori-30M (default),RealTabPFN-2.5 (default),TabDPT (default),RealMLP (tuned),Nori (default),xRFM (tuned + ensembled),LightGBM (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),LightGBM (tuned),ModernNCA (tuned),TabM (tuned + ensembled),CatBoost (default),xRFM (tuned),TabM (tuned),XGBoost (tuned + ensembled),ModernNCA (default),XGBoost (tuned),iLTM (tuned + ensembled),ChimeraBoost (tuned + ensembled),iLTM (tuned),LimiX (default),XGBoost (default),ChimeraBoost (tuned),TabSwift (default),PerpetualBooster (tuned + ensembled),TabM (default),xRFM (default),TorchMLP (tuned + ensembled),LightGBM (default),ChimeraBoost (default),PerpetualBooster (tuned),TorchMLP (tuned),ExtraTrees (tuned + ensembled),RealMLP (default),ExtraTrees (tuned),RandomForest (tuned + ensembled),SAP-RPT-OSS (default),RandomForest (tuned),EBM (tuned + ensembled),TorchMLP (default),EBM (tuned),PerpetualBooster (default),RandomForest (default),ExtraTrees (default),EBM (default),iLTM (default),FastaiMLP (tuned + ensembled),FastaiMLP (tuned),KNN (tuned + ensembled),TabSTAR (tuned + ensembled),FastaiMLP (default),TabSTAR (tuned),KNN (tuned),TabSTAR (default),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,1.0,0.73333335,0.8,1.0,1.0,0.6888889,1.0,0.8888889,1.0,1.0,0.95555556,0.84444445,1.0,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,,0.22222222,0.6666667,0.73333335,0.8666667,0.6,0.8,0.26666668,0.9111111,0.7777778,0.6,0.6,0.82222223,1.0,0.6,0.95555556,0.95555556,0.9777778,0.82222223,0.8,0.8,0.8,1.0,0.9777778,0.84444445,1.0,0.9777778,0.9777778,1.0,0.95555556,1.0,0.84444445,1.0,1.0,0.95555556,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.26666668,0.7777778,,0.8,0.8,0.7777778,0.42222223,0.8,0.82222223,0.8,0.8,0.84444445,0.53333336,0.8,1.0,0.6888889,0.8,0.9777778,0.82222223,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,0.8,1.0,0.8,0.93333334,0.8,1.0,1.0,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,0.8,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,0.84444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2,0.33333334,0.2,,0.5555556,0.46666667,0.35555556,0.5555556,0.2,0.73333335,0.9777778,0.35555556,0.42222223,0.6666667,0.51111114,0.6,0.8,0.8888889,0.93333334,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.0,0.26666668,0.2,0.44444445,,0.53333336,0.51111114,0.62222224,0.24444444,0.82222223,0.64444447,0.44444445,0.5777778,0.6666667,0.62222224,0.6,0.95555556,0.82222223,0.8888889,0.82222223,0.82222223,0.82222223,0.84444445,1.0,0.95555556,0.9111111,0.9777778,0.9777778,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.13333334,0.22222222,0.53333336,0.46666667,,0.53333336,0.6,0.24444444,0.8,0.6888889,0.5555556,0.5777778,0.75555557,0.8,0.5777778,0.84444445,0.9111111,0.93333334,0.8,0.7777778,0.8,0.82222223,1.0,1.0,0.82222223,1.0,1.0,0.93333334,1.0,0.84444445,1.0,0.84444445,1.0,1.0,0.84444445,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.31111112,0.4,0.5777778,0.64444447,0.4888889,0.46666667,,0.46666667,0.4888889,0.4888889,0.7111111,0.7777778,0.95555556,0.6666667,0.7111111,0.9777778,0.51111114,0.8666667,0.8888889,0.73333335,0.73333335,0.73333335,0.73333335,0.82222223,0.9111111,0.73333335,0.8888889,0.93333334,0.8,0.93333334,0.7777778,0.95555556,0.8,1.0,0.84444445,0.82222223,0.82222223,0.9111111,0.9777778,0.9777778,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,0.8888889,1.0,0.9777778,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.2,0.2,0.44444445,0.37777779,0.4,0.53333336,,0.2,0.46666667,0.5777778,0.42222223,0.53333336,0.64444447,0.6666667,0.6,0.84444445,0.7111111,0.84444445,0.75555557,0.75555557,0.75555557,0.82222223,1.0,0.84444445,0.8666667,0.9777778,0.95555556,0.93333334,1.0,0.93333334,0.9777778,0.9111111,0.9777778,0.93333334,0.9777778,0.9777778,0.95555556,1.0,1.0,1.0,1.0,0.9777778,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.11111111,0.73333335,0.17777778,0.8,0.75555557,0.75555557,0.51111114,0.8,,0.7777778,0.8,0.6,0.5777778,0.8,0.7777778,0.5555556,0.7777778,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,1.0,0.95555556,0.8,0.95555556,0.8,0.8,0.82222223,1.0,1.0,1.0,0.84444445,1.0,0.84444445,0.8,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.08888889,0.2,0.26666668,0.17777778,0.2,0.51111114,0.53333336,0.22222222,,0.62222224,0.44444445,0.53333336,0.51111114,0.5555556,0.6,0.8,0.73333335,0.8888889,0.7777778,0.8,0.8,0.75555557,0.7777778,0.93333334,0.8,0.9777778,0.9777778,0.9111111,1.0,0.8666667,1.0,0.8,1.0,0.9777778,0.8666667,0.8666667,0.95555556,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.22222222,0.2,0.022222223,0.35555556,0.31111112,0.2888889,0.42222223,0.2,0.37777779,,0.26666668,0.37777779,0.4888889,0.4,0.5777778,0.7777778,0.8,0.84444445,0.93333334,0.93333334,0.93333334,0.93333334,0.7111111,0.8888889,0.9777778,0.9777778,0.95555556,1.0,0.9111111,1.0,0.9777778,1.0,0.9777778,0.8,1.0,1.0,0.84444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.044444446,0.4,0.15555556,0.64444447,0.5555556,0.44444445,0.22222222,0.5777778,0.4,0.5555556,0.73333335,,0.31111112,0.5777778,0.75555557,0.5555556,0.6,0.8888889,0.73333335,0.7777778,0.8,0.8,0.7777778,0.7777778,0.9777778,0.8,0.9111111,0.9777778,0.8,0.93333334,0.8,0.9111111,0.8,1.0,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,0.8,0.95555556,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,0.8,1.0,0.8,1.0,1.0,1.0,0.84444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.15555556,0.4,0.46666667,0.5777778,0.42222223,0.42222223,0.044444446,0.46666667,0.42222223,0.46666667,0.62222224,0.6888889,,0.5777778,0.64444447,0.82222223,0.46666667,0.84444445,0.8666667,0.7111111,0.7111111,0.7111111,0.73333335,0.75555557,0.9111111,0.7111111,0.8888889,0.9111111,0.7777778,0.9111111,0.75555557,0.93333334,0.8,1.0,0.8,0.7777778,0.82222223,0.8888889,0.95555556,0.9111111,1.0,1.0,0.9777778,0.8888889,0.9777778,1.0,0.9777778,0.95555556,1.0,1.0,0.93333334,1.0,0.84444445,1.0,0.9777778,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori-30M (default),0.0,0.17777778,0.2,0.33333334,0.33333334,0.24444444,0.33333334,0.35555556,0.2,0.4888889,0.51111114,0.42222223,0.42222223,,0.4888889,0.53333336,0.4888889,0.75555557,0.73333335,0.75555557,0.75555557,0.75555557,0.75555557,0.6888889,0.93333334,0.75555557,0.8666667,0.95555556,0.8888889,0.84444445,0.8888889,0.95555556,0.8,0.9777778,0.62222224,0.8888889,0.84444445,0.73333335,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.95555556,0.9777778,0.9777778,0.9777778,0.95555556,0.9777778,0.9777778,0.93333334,0.9777778,0.95555556,1.0,1.0,1.0,0.9777778,0.9777778,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.0,0.0,0.4888889,0.37777779,0.2,0.2888889,0.33333334,0.22222222,0.44444445,0.6,0.24444444,0.35555556,0.51111114,,0.37777779,0.5777778,0.6,0.6888889,0.75555557,0.8,0.8,0.75555557,0.8,0.95555556,0.8,0.7777778,0.95555556,0.8,0.8,0.8,0.8,0.8,0.8888889,0.84444445,0.8,0.8,1.0,0.8,1.0,1.0,0.9777778,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,0.8,0.82222223,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.13333334,0.4,0.31111112,0.4,0.4,0.42222223,0.022222223,0.4,0.44444445,0.4,0.42222223,0.44444445,0.17777778,0.46666667,0.62222224,,0.42222223,0.75555557,0.5777778,0.6666667,0.7111111,0.7111111,0.73333335,0.51111114,0.7111111,0.7111111,0.84444445,0.84444445,0.73333335,0.75555557,0.73333335,0.8888889,0.8,0.9777778,0.8,0.75555557,0.8,0.84444445,0.73333335,0.9111111,0.93333334,1.0,0.93333334,0.8,0.93333334,1.0,0.93333334,0.93333334,0.93333334,1.0,0.82222223,1.0,0.8,1.0,0.8,1.0,1.0,0.9777778,0.82222223,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.044444446,0.2,0.2,0.044444446,0.15555556,0.4888889,0.15555556,0.22222222,0.2,0.22222222,0.4,0.53333336,0.51111114,0.42222223,0.5777778,,0.42222223,0.5777778,0.5555556,0.5777778,0.5777778,0.6,0.7111111,0.7777778,0.6,0.8,0.7777778,0.73333335,0.9777778,0.73333335,0.82222223,0.62222224,0.8,0.9111111,0.82222223,0.73333335,0.73333335,0.9111111,0.9111111,0.8888889,0.9777778,0.8888889,0.82222223,1.0,1.0,0.84444445,0.95555556,0.8666667,0.9777778,1.0,1.0,0.95555556,1.0,0.9777778,1.0,1.0,0.9111111,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori (default),0.0,0.044444446,0.022222223,0.11111111,0.17777778,0.08888889,0.13333334,0.2888889,0.2,0.26666668,0.2,0.11111111,0.15555556,0.24444444,0.4,0.24444444,0.5777778,,0.6,0.75555557,0.7777778,0.7777778,0.7777778,0.37777779,0.75555557,0.7777778,0.84444445,0.8666667,0.8,0.6888889,0.8,0.84444445,0.8,1.0,0.8,0.8,0.8,0.7777778,0.8,1.0,1.0,1.0,1.0,0.8,0.9111111,1.0,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,0.8,0.9777778,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.022222223,0.17777778,0.06666667,0.11111111,0.06666667,0.11111111,0.15555556,0.2,0.11111111,0.15555556,0.26666668,0.13333334,0.26666668,0.31111112,0.42222223,0.42222223,0.4,,0.6,0.6888889,0.6888889,0.62222224,0.4,0.62222224,0.7111111,1.0,0.75555557,0.7111111,0.7777778,0.7111111,0.8888889,0.6888889,0.93333334,0.73333335,0.8,0.73333335,0.7777778,0.8888889,0.9777778,1.0,1.0,0.9777778,0.7777778,1.0,1.0,0.95555556,1.0,0.9777778,1.0,0.8,1.0,0.8,1.0,0.84444445,1.0,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.17777778,0.2,0.0,0.17777778,0.2,0.26666668,0.24444444,0.2,0.22222222,0.06666667,0.22222222,0.2888889,0.24444444,0.24444444,0.33333334,0.44444445,0.24444444,0.4,,0.46666667,0.6,0.73333335,0.53333336,0.35555556,0.82222223,0.4888889,0.51111114,0.9777778,0.5777778,0.95555556,0.75555557,0.9777778,0.7777778,0.6,1.0,1.0,0.6888889,1.0,0.95555556,0.8,0.8,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.2,0.2,0.0,0.17777778,0.22222222,0.26666668,0.24444444,0.2,0.2,0.06666667,0.2,0.2888889,0.24444444,0.2,0.2888889,0.42222223,0.22222222,0.31111112,0.53333336,,0.6666667,0.64444447,0.53333336,0.26666668,0.82222223,0.4888889,0.37777779,0.84444445,0.5555556,0.9111111,0.64444447,0.9111111,0.75555557,0.6,0.9777778,0.95555556,0.6888889,0.9777778,0.93333334,0.7777778,0.8,0.9777778,1.0,1.0,0.8,0.93333334,1.0,0.95555556,1.0,0.8,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.2,0.2,0.0,0.17777778,0.2,0.26666668,0.24444444,0.2,0.2,0.06666667,0.2,0.2888889,0.24444444,0.2,0.2888889,0.42222223,0.22222222,0.31111112,0.4,0.33333334,,0.5777778,0.53333336,0.26666668,0.7111111,0.4888889,0.37777779,0.73333335,0.5555556,0.75555557,0.64444447,0.9111111,0.6888889,0.6,0.9777778,0.95555556,0.6888889,0.9777778,0.93333334,0.8,0.8,0.9777778,0.9777778,1.0,0.8,0.93333334,1.0,0.95555556,0.9777778,0.8,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.2,0.2,0.0,0.15555556,0.17777778,0.26666668,0.17777778,0.2,0.24444444,0.06666667,0.22222222,0.26666668,0.24444444,0.24444444,0.26666668,0.4,0.22222222,0.37777779,0.26666668,0.35555556,0.42222223,,0.46666667,0.24444444,0.5555556,0.42222223,0.33333334,0.93333334,0.4888889,1.0,0.62222224,0.93333334,0.8,0.6,1.0,0.9777778,0.6,0.9777778,0.9111111,0.7777778,0.8,1.0,0.9777778,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.2,0.044444446,0.0,0.0,0.17777778,0.0,0.2,0.22222222,0.2888889,0.22222222,0.24444444,0.31111112,0.2,0.4888889,0.2888889,0.62222224,0.6,0.46666667,0.46666667,0.46666667,0.53333336,,0.62222224,0.46666667,0.6666667,0.7111111,0.6,0.8666667,0.6,0.75555557,0.6,0.8,0.75555557,0.6888889,0.6666667,0.82222223,0.82222223,0.8888889,0.84444445,0.93333334,0.82222223,0.6666667,0.93333334,1.0,0.8,0.9111111,0.8,0.9111111,0.8888889,0.9777778,0.82222223,1.0,0.93333334,1.0,0.9777778,0.8888889,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.022222223,0.0,0.0,0.044444446,0.0,0.08888889,0.15555556,0.0,0.06666667,0.11111111,0.022222223,0.08888889,0.06666667,0.044444446,0.2888889,0.22222222,0.24444444,0.37777779,0.64444447,0.73333335,0.73333335,0.75555557,0.37777779,,0.73333335,0.44444445,0.95555556,0.7777778,0.51111114,0.7777778,0.6888889,0.7777778,0.8,0.6,0.8,0.8,0.6666667,0.8,1.0,0.84444445,0.7777778,0.8,0.8,0.8,0.8,1.0,1.0,1.0,0.8,0.84444445,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.15555556,0.2,0.0,0.08888889,0.17777778,0.26666668,0.13333334,0.2,0.2,0.022222223,0.2,0.2888889,0.24444444,0.2,0.2888889,0.4,0.22222222,0.2888889,0.17777778,0.17777778,0.2888889,0.44444445,0.53333336,0.26666668,,0.4888889,0.4,0.6666667,0.5555556,0.7111111,0.64444447,0.62222224,0.6888889,0.6,0.8,0.75555557,0.6888889,0.8888889,0.73333335,0.7777778,0.8,0.75555557,0.7777778,0.93333334,0.8,0.7111111,0.7777778,0.7111111,0.8,0.62222224,0.8,1.0,0.8,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.022222223,0.0,0.11111111,0.022222223,0.2,0.022222223,0.022222223,0.08888889,0.11111111,0.13333334,0.22222222,0.15555556,0.2,0.15555556,0.0,0.51111114,0.51111114,0.51111114,0.5777778,0.33333334,0.5555556,0.51111114,,0.5777778,0.6,0.6,0.6,0.7111111,0.62222224,0.8,0.6,0.7111111,0.64444447,0.7111111,0.73333335,0.9111111,0.9111111,0.95555556,0.8666667,0.64444447,0.8666667,0.9777778,0.8,0.95555556,0.8666667,0.95555556,0.8,1.0,0.8,1.0,0.8,0.9777778,1.0,0.95555556,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.022222223,0.0,0.0,0.022222223,0.0,0.06666667,0.044444446,0.0,0.022222223,0.044444446,0.022222223,0.08888889,0.044444446,0.044444446,0.15555556,0.22222222,0.13333334,0.24444444,0.4888889,0.62222224,0.62222224,0.6666667,0.2888889,0.044444446,0.6,0.42222223,,0.73333335,0.4,0.73333335,0.46666667,0.73333335,0.7777778,0.6,0.75555557,0.73333335,0.62222224,0.8,0.9777778,0.84444445,0.6,0.75555557,0.75555557,0.8,0.7111111,0.95555556,1.0,1.0,0.8,0.82222223,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.022222223,0.2,0.0,0.0,0.06666667,0.2,0.06666667,0.2,0.08888889,0.0,0.2,0.22222222,0.11111111,0.2,0.26666668,0.26666668,0.2,0.2888889,0.022222223,0.15555556,0.26666668,0.06666667,0.4,0.22222222,0.33333334,0.4,0.26666668,,0.44444445,0.53333336,0.44444445,0.5555556,0.73333335,0.6,0.7777778,0.5777778,0.5777778,0.8666667,0.6888889,0.75555557,0.8,0.9777778,0.84444445,0.95555556,0.8,0.8666667,0.9111111,0.9111111,1.0,0.8,1.0,0.9777778,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.044444446,0.0,0.0,0.06666667,0.0,0.2,0.0,0.08888889,0.06666667,0.08888889,0.15555556,0.2,0.24444444,0.022222223,0.31111112,0.22222222,0.42222223,0.44444445,0.44444445,0.51111114,0.13333334,0.4888889,0.44444445,0.4,0.6,0.5555556,,0.5777778,0.53333336,0.6,0.8,0.6,0.6,0.6,0.62222224,0.62222224,0.75555557,0.8,0.8888889,0.7777778,0.62222224,0.8,0.9777778,0.8,0.84444445,0.8,0.8888889,0.8666667,0.95555556,0.8,1.0,0.8,1.0,0.95555556,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.044444446,0.2,0.0,0.0,0.15555556,0.22222222,0.06666667,0.2,0.13333334,0.0,0.2,0.24444444,0.11111111,0.2,0.26666668,0.26666668,0.2,0.2888889,0.044444446,0.08888889,0.24444444,0.0,0.4,0.22222222,0.2888889,0.4,0.26666668,0.46666667,0.42222223,,0.37777779,0.53333336,0.64444447,0.6,0.8,0.5555556,0.5555556,0.75555557,0.64444447,0.6888889,0.7777778,0.8666667,0.82222223,0.93333334,0.8,0.8666667,0.8888889,0.9111111,1.0,0.8,1.0,0.9777778,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.06666667,0.0,0.0,0.0,0.044444446,0.022222223,0.2,0.0,0.022222223,0.08888889,0.06666667,0.044444446,0.2,0.11111111,0.17777778,0.15555556,0.11111111,0.24444444,0.35555556,0.35555556,0.37777779,0.24444444,0.31111112,0.35555556,0.2888889,0.53333336,0.5555556,0.46666667,0.62222224,,0.51111114,0.84444445,0.5555556,0.6666667,0.5555556,0.5777778,0.64444447,0.7111111,0.93333334,0.9111111,0.8888889,0.5777778,0.8666667,0.95555556,0.8888889,0.9111111,0.8888889,0.9777778,0.7777778,0.9777778,0.7777778,0.9777778,0.7777778,0.9777778,0.9777778,0.9111111,0.7777778,0.95555556,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778
+ChimeraBoost (tuned + ensembled),0.0,0.15555556,0.2,0.0,0.044444446,0.15555556,0.2,0.08888889,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.37777779,0.2,0.31111112,0.022222223,0.08888889,0.08888889,0.06666667,0.4,0.22222222,0.37777779,0.37777779,0.26666668,0.44444445,0.4,0.46666667,0.4888889,,0.51111114,0.4,0.7777778,0.82222223,0.6,0.7777778,0.5777778,0.5777778,0.6,0.7777778,0.95555556,0.8,0.6,0.73333335,0.73333335,0.75555557,0.8,0.6,0.8,0.8,0.64444447,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.9111111,0.8,0.8,0.8,1.0,0.84444445,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.022222223,0.0,0.0,0.022222223,0.11111111,0.022222223,0.2,0.0,0.06666667,0.22222222,0.24444444,0.31111112,0.2,0.2,0.2,0.31111112,0.2,0.22222222,0.26666668,0.2,0.35555556,0.15555556,0.4888889,,0.6,0.4888889,0.4888889,0.4888889,0.44444445,0.6,0.84444445,0.7111111,0.75555557,0.5777778,0.62222224,0.82222223,0.7777778,0.8666667,0.8,1.0,0.8,1.0,0.8,1.0,0.8,0.82222223,1.0,0.9111111,0.8,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.0,0.0,0.2,0.0,0.0,0.15555556,0.06666667,0.0,0.022222223,0.2,0.2,0.2,0.37777779,0.15555556,0.2,0.08888889,0.2,0.26666668,0.4,0.4,0.4,0.4,0.24444444,0.4,0.4,0.4,0.4,0.4,0.4,0.4,0.44444445,0.6,0.4,,0.46666667,0.6,0.5777778,0.42222223,0.44444445,0.44444445,0.53333336,0.42222223,0.6,0.46666667,0.64444447,0.4,0.6888889,0.4,0.6,0.9777778,0.64444447,0.8,0.8,0.8,0.8,0.8,0.6666667,0.8,0.7111111,0.9777778,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.0,0.044444446,0.2,0.0,0.0,0.15555556,0.17777778,0.022222223,0.2,0.13333334,0.0,0.2,0.22222222,0.11111111,0.2,0.24444444,0.17777778,0.2,0.2,0.0,0.022222223,0.022222223,0.0,0.31111112,0.2,0.2,0.2888889,0.24444444,0.22222222,0.4,0.2,0.33333334,0.22222222,0.51111114,0.53333336,,0.33333334,0.4888889,0.4888889,0.35555556,0.53333336,0.7111111,0.46666667,0.53333336,0.7777778,0.75555557,0.44444445,0.6,0.46666667,0.7777778,0.62222224,0.8,0.9777778,0.8,1.0,1.0,1.0,0.7777778,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.044444446,0.2,0.0,0.022222223,0.11111111,0.17777778,0.022222223,0.2,0.13333334,0.0,0.2,0.17777778,0.15555556,0.2,0.2,0.26666668,0.2,0.26666668,0.0,0.044444446,0.044444446,0.022222223,0.33333334,0.2,0.24444444,0.35555556,0.26666668,0.42222223,0.4,0.44444445,0.44444445,0.17777778,0.51111114,0.4,0.6666667,,0.6,0.73333335,0.53333336,0.5555556,0.6,0.73333335,0.84444445,0.7777778,0.6,0.6888889,0.73333335,0.73333335,0.8,0.6,0.8,0.75555557,0.62222224,0.8,0.8,0.8,0.7777778,0.8,0.8,0.8,0.8,0.8888889,0.8,0.8,0.8,1.0,0.84444445,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.0,0.0,0.0,0.0,0.08888889,0.044444446,0.0,0.044444446,0.15555556,0.0,0.11111111,0.26666668,0.0,0.15555556,0.26666668,0.22222222,0.22222222,0.31111112,0.31111112,0.31111112,0.4,0.17777778,0.33333334,0.31111112,0.2888889,0.37777779,0.42222223,0.37777779,0.44444445,0.42222223,0.4,0.51111114,0.42222223,0.51111114,0.4,,0.46666667,0.37777779,0.6,0.62222224,0.53333336,0.42222223,0.53333336,0.62222224,0.6,0.8,0.6,0.6,0.73333335,0.62222224,0.8,0.7777778,0.8,0.8,0.6888889,0.8,0.8,0.8,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.2,0.0,0.0,0.0,0.022222223,0.0,0.2,0.0,0.0,0.2,0.044444446,0.022222223,0.2,0.26666668,0.08888889,0.2,0.11111111,0.0,0.022222223,0.022222223,0.022222223,0.17777778,0.2,0.11111111,0.26666668,0.2,0.13333334,0.37777779,0.24444444,0.35555556,0.22222222,0.5555556,0.5777778,0.51111114,0.26666668,0.53333336,,0.37777779,0.62222224,0.75555557,0.5555556,0.37777779,0.9777778,0.8,0.6,0.6666667,0.6,0.75555557,0.64444447,0.8,0.8,0.8,0.84444445,1.0,0.95555556,0.93333334,0.8888889,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.08888889,0.022222223,0.0,0.08888889,0.08888889,0.0,0.022222223,0.044444446,0.06666667,0.06666667,0.08888889,0.11111111,0.0,0.26666668,0.08888889,0.022222223,0.31111112,0.24444444,0.35555556,0.2888889,0.42222223,0.4,0.5555556,0.64444447,0.46666667,0.62222224,0.62222224,,0.42222223,0.53333336,0.6,0.53333336,0.6888889,0.6,0.62222224,0.82222223,0.62222224,0.73333335,0.8,0.75555557,0.8,0.62222224,0.8,0.8,0.95555556,0.8666667,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.0,0.0,0.0,0.022222223,0.0,0.06666667,0.11111111,0.0,0.0,0.2,0.22222222,0.2,0.22222222,0.15555556,0.15555556,0.22222222,0.08888889,0.15555556,0.24444444,0.2,0.31111112,0.06666667,0.42222223,0.15555556,0.5555556,0.46666667,0.44444445,0.4,0.37777779,0.5777778,,0.33333334,0.42222223,0.44444445,0.53333336,0.46666667,0.7777778,0.6888889,0.82222223,0.73333335,0.8,0.75555557,0.62222224,0.95555556,0.64444447,0.7777778,0.95555556,0.8888889,0.73333335,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.022222223,0.022222223,0.0,0.022222223,0.0,0.0,0.2,0.2,0.2,0.2,0.06666667,0.22222222,0.2,0.044444446,0.4,0.2,0.11111111,0.22222222,0.08888889,0.4,0.2888889,0.46666667,0.2888889,0.4,0.37777779,0.24444444,0.46666667,0.6666667,,0.4,0.4,0.33333334,0.9777778,0.6666667,0.5777778,0.6666667,0.7777778,0.8,0.84444445,0.7111111,1.0,0.7777778,0.7777778,0.84444445,0.73333335,0.8,0.93333334,0.9111111,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.044444446,0.0,0.0,0.0,0.022222223,0.022222223,0.0,0.06666667,0.11111111,0.0,0.022222223,0.0,0.022222223,0.022222223,0.0,0.17777778,0.2,0.24444444,0.13333334,0.24444444,0.022222223,0.22222222,0.13333334,0.11111111,0.22222222,0.24444444,0.5777778,0.53333336,0.26666668,0.46666667,0.44444445,0.4,0.5777778,0.6,,0.51111114,0.53333336,0.6,0.6,0.7111111,0.62222224,0.6666667,0.75555557,0.64444447,0.8,0.8,0.8,0.8,1.0,0.93333334,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.2,0.0,0.0,0.0,0.06666667,0.022222223,0.2,0.044444446,0.0,0.2,0.11111111,0.044444446,0.2,0.2,0.17777778,0.2,0.22222222,0.0,0.0,0.022222223,0.022222223,0.33333334,0.2,0.22222222,0.35555556,0.24444444,0.15555556,0.37777779,0.17777778,0.42222223,0.044444446,0.42222223,0.4,0.46666667,0.15555556,0.5777778,0.62222224,0.46666667,0.5555556,0.6,0.4888889,,0.75555557,0.6,0.44444445,0.5777778,0.46666667,0.5555556,0.42222223,0.6,0.6888889,0.62222224,0.8,0.8,0.8,0.7777778,0.8,0.8,0.8,0.8,0.84444445,0.8,0.8,0.8,0.93333334,0.8,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.044444446,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.044444446,0.022222223,0.022222223,0.2,0.06666667,0.0,0.08888889,0.0,0.0,0.0,0.0,0.0,0.06666667,0.2,0.06666667,0.13333334,0.2,0.044444446,0.2,0.06666667,0.13333334,0.2,0.37777779,0.53333336,0.22222222,0.22222222,0.46666667,0.022222223,0.31111112,0.46666667,0.6666667,0.46666667,0.24444444,,0.7111111,0.53333336,0.51111114,0.5555556,0.62222224,0.62222224,0.75555557,0.8,0.8,0.8,0.9777778,0.8888889,0.75555557,0.8,1.0,0.9777778,0.9777778,0.95555556,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.17777778,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.022222223,0.2888889,0.2,0.022222223,0.2,0.044444446,0.4,0.17777778,0.35555556,0.24444444,0.4,0.37777779,0.2,0.4,0.53333336,0.022222223,0.4,0.4,0.2888889,,0.62222224,0.46666667,0.62222224,0.62222224,0.8,0.64444447,0.64444447,1.0,0.6666667,0.7111111,0.7111111,0.64444447,0.75555557,0.8888889,0.8888889,0.95555556,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.022222223,0.0,0.06666667,0.15555556,0.0,0.044444446,0.0,0.06666667,0.06666667,0.0,0.2,0.0,0.2888889,0.2,0.044444446,0.13333334,0.2,0.13333334,0.11111111,0.26666668,0.22222222,0.6,0.5555556,0.31111112,0.4,0.4,0.37777779,0.22222222,0.33333334,0.4,0.5555556,0.46666667,0.37777779,,0.51111114,0.9111111,0.6,0.8,0.4888889,0.6,0.42222223,0.6,0.6,0.8,1.0,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.044444446,0.044444446,0.0,0.06666667,0.044444446,0.0,0.0,0.0,0.0,0.0,0.0,0.08888889,0.0,0.22222222,0.044444446,0.0,0.08888889,0.15555556,0.11111111,0.08888889,0.26666668,0.13333334,0.31111112,0.4,0.26666668,0.2,0.33333334,0.17777778,0.31111112,0.42222223,0.2888889,0.42222223,0.4888889,0.53333336,0.4888889,,0.53333336,0.64444447,0.64444447,0.6888889,0.8,0.6,0.8,0.8,0.75555557,0.84444445,0.8,0.8,1.0,1.0,0.84444445,0.8,1.0,0.8,1.0,0.93333334,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.06666667,0.13333334,0.0,0.022222223,0.0,0.044444446,0.044444446,0.0,0.2,0.0,0.2888889,0.13333334,0.0,0.08888889,0.2,0.08888889,0.11111111,0.24444444,0.2,0.6,0.53333336,0.26666668,0.4,0.4,0.37777779,0.17777778,0.33333334,0.37777779,0.53333336,0.44444445,0.37777779,0.08888889,0.46666667,,0.46666667,0.8,0.44444445,0.6,0.42222223,0.6,0.6,0.8,1.0,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15555556,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.022222223,0.0,0.08888889,0.2,0.2,0.044444446,0.2,0.0,0.11111111,0.0,0.022222223,0.2,0.0,0.4,0.22222222,0.2,0.4,0.24444444,0.26666668,0.26666668,0.22222222,0.33333334,0.44444445,0.37777779,0.37777779,0.4,0.35555556,0.53333336,,0.7777778,0.6,0.6,0.73333335,0.6666667,0.73333335,0.9777778,0.64444447,0.6666667,0.8888889,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.08888889,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.044444446,0.06666667,0.06666667,0.0,0.17777778,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.11111111,0.15555556,0.37777779,0.2,0.17777778,0.2,0.13333334,0.2,0.22222222,0.4,0.2,0.022222223,0.37777779,0.4,0.26666668,0.35555556,0.2,0.2,0.2,0.24444444,0.5777778,0.37777779,0.2,0.2,0.35555556,0.2,0.22222222,,0.22222222,0.75555557,0.51111114,0.8,0.6,0.42222223,0.6,0.8,0.5777778,0.82222223,0.8666667,0.8,0.8,0.9111111,0.8,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15555556,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.2,0.0,0.2,0.0,0.044444446,0.0,0.022222223,0.2,0.0,0.35555556,0.2,0.2,0.37777779,0.2,0.24444444,0.24444444,0.15555556,0.35555556,0.4,0.24444444,0.35555556,0.51111114,0.31111112,0.5555556,0.4,0.7777778,,0.6,0.7111111,0.6666667,0.7111111,0.7111111,0.62222224,0.6666667,0.8888889,0.8666667,0.93333334,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.2,0.0,0.0,0.0,0.11111111,0.0,0.2,0.044444446,0.0,0.2,0.15555556,0.044444446,0.2,0.2,0.044444446,0.2,0.2,0.0,0.0,0.0,0.0,0.17777778,0.2,0.0,0.2,0.2,0.022222223,0.2,0.022222223,0.22222222,0.2,0.2,0.2,0.022222223,0.24444444,0.2,0.2,0.2,0.37777779,0.2888889,0.2,0.31111112,0.2,0.35555556,0.4,0.2,0.4,0.4,0.24444444,0.4,,0.6666667,1.0,0.46666667,0.4,0.4,1.0,0.4888889,0.6,0.6,0.6888889,0.8888889,0.75555557,0.9777778,0.7777778,1.0,0.8,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.08888889,0.08888889,0.08888889,0.06666667,0.0,0.0,0.2,0.0,0.0,0.08888889,0.0,0.08888889,0.022222223,0.35555556,0.0,0.2,0.2,0.37777779,0.22222222,0.2,0.37777779,0.044444446,0.0,0.2,0.37777779,0.2,0.0,0.5777778,0.4,0.5777778,0.26666668,0.4888889,0.2888889,0.33333334,,0.46666667,0.4,0.4888889,0.6,0.46666667,0.5777778,0.75555557,0.7777778,0.9111111,0.82222223,0.8,0.84444445,1.0,0.95555556,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.2,0.0,0.0,0.0,0.022222223,0.0,0.2,0.0,0.0,0.2,0.022222223,0.0,0.2,0.2,0.022222223,0.2,0.15555556,0.0,0.0,0.0,0.0,0.06666667,0.2,0.0,0.2,0.2,0.0,0.2,0.0,0.22222222,0.2,0.2,0.2,0.0,0.2,0.2,0.15555556,0.2,0.35555556,0.22222222,0.2,0.2,0.2,0.33333334,0.4,0.2,0.4,0.33333334,0.2,0.33333334,0.0,0.53333336,,0.4,0.33333334,0.4,0.93333334,0.4,0.6,0.6,0.6,0.8666667,0.75555557,0.95555556,0.75555557,1.0,0.8,1.0,1.0,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.17777778,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.022222223,0.2,0.0,0.0,0.0,0.022222223,0.2,0.17777778,0.2,0.0,0.2,0.2,0.0,0.2,0.22222222,0.22222222,0.2,0.2,0.022222223,0.2888889,0.4,0.2,0.4,0.26666668,0.4,0.2888889,0.53333336,0.6,0.6,,0.2888889,0.4,0.62222224,0.73333335,0.5777778,0.7111111,0.6888889,1.0,0.93333334,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.022222223,0.2,0.0,0.2,0.0,0.2,0.31111112,0.044444446,0.044444446,0.044444446,0.15555556,0.0,0.2,0.11111111,0.2888889,0.2,0.24444444,0.2,0.022222223,0.5777778,0.2888889,0.6,0.51111114,0.6666667,0.7111111,,0.5555556,0.6666667,0.6888889,0.8666667,0.93333334,0.93333334,0.9777778,0.9777778,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.022222223,0.08888889,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.022222223,0.044444446,0.0,0.0,0.2,0.0,0.08888889,0.2,0.08888889,0.33333334,0.22222222,0.22222222,0.2,0.06666667,0.13333334,0.11111111,0.26666668,0.06666667,0.22222222,0.24444444,0.35555556,0.0,0.15555556,0.0,0.35555556,0.4,0.37777779,0.6,0.4,0.6,0.6,0.44444445,,0.6,0.7777778,0.6,0.64444447,0.73333335,0.73333335,0.93333334,0.73333335,0.8,0.8,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.15555556,0.0,0.0,0.0,0.022222223,0.0,0.2,0.0,0.0,0.15555556,0.022222223,0.0,0.2,0.17777778,0.022222223,0.2,0.08888889,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.2,0.2,0.0,0.2,0.0,0.22222222,0.2,0.2,0.2,0.0,0.2,0.2,0.11111111,0.2,0.26666668,0.2,0.2,0.2,0.2,0.24444444,0.4,0.2,0.4,0.33333334,0.2,0.33333334,0.0,0.53333336,0.06666667,0.37777779,0.33333334,0.4,,0.4,0.5555556,0.6,0.6,0.84444445,0.6666667,0.8888889,0.64444447,1.0,0.8,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.2,0.022222223,0.2888889,0.022222223,0.2,0.2,0.0,0.2,0.022222223,0.06666667,0.0,0.2,0.0,0.11111111,0.2,0.2,0.2,0.11111111,0.42222223,0.11111111,0.51111114,0.42222223,0.6,0.26666668,0.31111112,0.22222222,0.6,,0.53333336,0.5777778,0.64444447,0.84444445,0.82222223,0.84444445,0.8,1.0,0.8666667,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.022222223,0.0,0.2,0.0,0.0,0.0,0.0,0.08888889,0.0,0.2,0.022222223,0.11111111,0.2,0.0,0.2,0.06666667,0.17777778,0.13333334,0.4,0.24444444,0.4,0.42222223,0.13333334,0.4,0.44444445,0.46666667,,0.8,0.4,0.8,1.0,0.8,0.5777778,0.8,1.0,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.022222223,0.0,0.2,0.0,0.0,0.0,0.0,0.022222223,0.0,0.2,0.022222223,0.044444446,0.2,0.0,0.2,0.06666667,0.13333334,0.06666667,0.4,0.22222222,0.4,0.2888889,0.06666667,0.35555556,0.4,0.42222223,0.2,,0.4,0.8,0.9111111,0.8,0.5777778,0.8,1.0,1.0,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.08888889,0.0,0.0,0.0,0.11111111,0.022222223,0.044444446,0.0,0.0,0.0,0.0,0.15555556,0.044444446,0.0,0.0,0.15555556,0.0,0.0,0.2,0.0,0.31111112,0.08888889,0.4,0.31111112,0.06666667,0.26666668,0.4,0.35555556,0.6,0.6,,0.4888889,0.6,0.4888889,1.0,0.5777778,1.0,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.11111111,0.17777778,0.13333334,0.0,0.022222223,0.26666668,0.15555556,0.15555556,0.2,0.2,0.51111114,,0.24444444,0.75555557,0.6,1.0,0.6,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.044444446,0.0,0.0,0.0,0.0,0.08888889,0.022222223,0.24444444,0.2,0.24444444,0.06666667,0.022222223,0.06666667,0.33333334,0.17777778,0.0,0.08888889,0.4,0.75555557,,0.75555557,0.4,0.8,0.6,1.0,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.022222223,0.15555556,0.044444446,0.0,0.022222223,0.26666668,0.11111111,0.15555556,0.2,0.2,0.51111114,0.24444444,0.24444444,,0.6,1.0,0.6,1.0,1.0,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.0,0.24444444,0.22222222,0.0,0.2,0.35555556,0.2,0.42222223,0.42222223,0.0,0.4,0.6,0.4,,0.4,1.0,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.15555556,0.0,0.0,0.0,0.15555556,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.2,0.42222223,0.0,0.2,0.0,0.6,,0.6,1.0,1.0,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.2,0.13333334,0.0,0.0,0.0,0.4,0.4,0.4,0.0,0.4,,0.8,0.8,0.8888889
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,,0.87777776,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.12222222,,0.9111111
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.08888889,
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_small/leaderboard_overview_explorer.html
index 7d0e3e0e8f22e3df95ccea251142c20501a2d4d6..d02f533428b72f8869676aaf7697d274952c06d1 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Regression | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_all/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_small/leaderboard_table.html b/data/imputation_no/splits_all/tasks_regression/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..241b7920502685e5e22c19c6b620df5108bc7f4a
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_all/tasks_regression/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1966.0,"score":0.923,"rank":4.77,"hrank":2.13,"imp":2.14,"train_time":49.29,"predict_time":4.094,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+182\/-110","imputed":false},{"position":1,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1859.0,"score":0.816,"rank":7.28,"hrank":2.45,"imp":1.882,"train_time":12.09,"predict_time":2.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+269\/-125","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1798.0,"score":0.737,"rank":9.2,"hrank":4.22,"imp":2.59,"train_time":14.09,"predict_time":0.439,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+194\/-112","imputed":false},{"position":3,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1784.0,"score":0.73,"rank":9.68,"hrank":5.53,"imp":3.426,"train_time":12.37,"predict_time":1.566,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+205\/-111","imputed":false},{"position":4,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1782.0,"score":0.7,"rank":9.72,"hrank":4.9,"imp":3.697,"train_time":322.87,"predict_time":5.571,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+205\/-99","imputed":false},{"position":5,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1740.0,"score":0.677,"rank":11.32,"hrank":7.48,"imp":3.846,"train_time":9.2,"predict_time":0.585,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-77","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1721.0,"score":0.628,"rank":12.09,"hrank":6.03,"imp":4.625,"train_time":4823.37,"predict_time":140.737,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+171\/-76","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1705.0,"score":0.616,"rank":12.75,"hrank":6.11,"imp":3.243,"train_time":2307.57,"predict_time":7.226,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+252\/-129","imputed":false},{"position":8,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1675.0,"score":0.576,"rank":14.1,"hrank":9.13,"imp":5.09,"train_time":4823.37,"predict_time":22.312,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-52","imputed":false},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1667.0,"score":0.649,"rank":14.47,"hrank":5.35,"imp":4.335,"train_time":5.15,"predict_time":0.14,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+251\/-203","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1637.0,"score":0.508,"rank":15.88,"hrank":9.85,"imp":5.036,"train_time":4560.01,"predict_time":33.831,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+151\/-65","imputed":false},{"position":11,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1580.0,"score":0.46,"rank":18.81,"hrank":9.78,"imp":6.192,"train_time":48.14,"predict_time":42.001,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-63","imputed":false},{"position":12,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1574.0,"score":0.471,"rank":19.16,"hrank":8.41,"imp":4.157,"train_time":2307.57,"predict_time":0.902,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+250\/-142","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1571.0,"score":0.446,"rank":19.32,"hrank":9.45,"imp":5.997,"train_time":7.93,"predict_time":0.264,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-81","imputed":false},{"position":14,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1560.0,"score":0.374,"rank":19.93,"hrank":14.12,"imp":6.935,"train_time":5807.07,"predict_time":4.31,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-75","imputed":false},{"position":15,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1547.0,"score":0.373,"rank":20.66,"hrank":13.4,"imp":5.96,"train_time":4560.01,"predict_time":2.233,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-74","imputed":false},{"position":16,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1506.0,"score":0.396,"rank":23.1,"hrank":14.93,"imp":5.988,"train_time":7.67,"predict_time":0.547,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+195\/-133","imputed":false},{"position":17,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1489.0,"score":0.281,"rank":24.1,"hrank":17.57,"imp":9.043,"train_time":39519.07,"predict_time":1519.479,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+122\/-62","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1488.0,"score":0.254,"rank":24.18,"hrank":19.3,"imp":7.74,"train_time":4283.1,"predict_time":0.444,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+139\/-78","imputed":false},{"position":19,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1482.0,"score":0.319,"rank":24.56,"hrank":15.99,"imp":6.059,"train_time":3.53,"predict_time":0.064,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+157\/-90","imputed":false},{"position":20,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1465.0,"score":0.234,"rank":25.6,"hrank":21.96,"imp":8.586,"train_time":1203.24,"predict_time":7.106,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-76","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1460.0,"score":0.225,"rank":25.93,"hrank":20.97,"imp":7.913,"train_time":4283.1,"predict_time":0.063,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-78","imputed":false},{"position":22,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1458.0,"score":0.28,"rank":26.05,"hrank":18.99,"imp":6.584,"train_time":3660.26,"predict_time":4.755,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+186\/-101","imputed":false},{"position":23,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1457.0,"score":0.293,"rank":26.1,"hrank":7.56,"imp":8.965,"train_time":8982.74,"predict_time":8.012,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+184\/-110","imputed":false},{"position":24,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1439.0,"score":0.234,"rank":27.27,"hrank":16.97,"imp":8.055,"train_time":2039.7,"predict_time":2.142,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+139\/-112","imputed":false},{"position":25,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1420.0,"score":0.171,"rank":28.44,"hrank":25.46,"imp":8.891,"train_time":1266.57,"predict_time":3.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-40","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1393.0,"score":0.145,"rank":30.27,"hrank":26.93,"imp":9.053,"train_time":1266.57,"predict_time":0.475,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-42","imputed":false},{"position":27,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1384.0,"score":0.155,"rank":30.86,"hrank":26.42,"imp":9.341,"train_time":1203.24,"predict_time":0.956,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-70","imputed":false},{"position":28,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1368.0,"score":0.183,"rank":31.91,"hrank":21.75,"imp":7.596,"train_time":3660.26,"predict_time":0.492,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+182\/-112","imputed":false},{"position":29,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1367.0,"score":0.22,"rank":31.96,"hrank":15.89,"imp":10.025,"train_time":22.46,"predict_time":1.038,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-163","imputed":false},{"position":30,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1363.0,"score":0.119,"rank":32.28,"hrank":25.51,"imp":9.905,"train_time":13.63,"predict_time":0.049,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-94","imputed":false},{"position":31,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1362.0,"score":0.181,"rank":32.32,"hrank":26.3,"imp":8.55,"train_time":2039.7,"predict_time":0.109,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+141\/-87","imputed":false},{"position":32,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1350.0,"score":0.13,"rank":33.11,"hrank":27.43,"imp":9.246,"train_time":39519.07,"predict_time":167.037,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-73","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1339.0,"score":0.108,"rank":33.91,"hrank":26.51,"imp":10.357,"train_time":1846.43,"predict_time":0.349,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+136\/-96","imputed":false},{"position":34,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1322.0,"score":0.129,"rank":35.07,"hrank":20.91,"imp":11.043,"train_time":8982.74,"predict_time":0.421,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+165\/-97","imputed":false},{"position":35,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1303.0,"score":0.15,"rank":36.33,"hrank":27.7,"imp":8.874,"train_time":13.53,"predict_time":0.467,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+170\/-139","imputed":false},{"position":36,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1302.0,"score":0.082,"rank":36.38,"hrank":29.06,"imp":10.698,"train_time":1846.43,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+134\/-97","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1276.0,"score":0.125,"rank":38.18,"hrank":31.46,"imp":9.674,"train_time":21.52,"predict_time":8.35,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-106","imputed":false},{"position":38,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1262.0,"score":0.137,"rank":39.16,"hrank":28.57,"imp":13.391,"train_time":1163.69,"predict_time":2.098,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-160","imputed":false},{"position":39,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1251.0,"score":0.114,"rank":39.9,"hrank":25.93,"imp":11.317,"train_time":5151.88,"predict_time":1.591,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+140\/-160","imputed":false},{"position":40,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1244.0,"score":0.12,"rank":40.4,"hrank":15.75,"imp":11.333,"train_time":121.77,"predict_time":21.803,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+212\/-212","imputed":false},{"position":41,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1241.0,"score":0.121,"rank":40.61,"hrank":27.05,"imp":13.637,"train_time":1163.69,"predict_time":0.288,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-162","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1233.0,"score":0.023,"rank":41.09,"hrank":37.52,"imp":11.649,"train_time":8.29,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-95","imputed":false},{"position":43,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1213.0,"score":0.025,"rank":42.45,"hrank":32.85,"imp":12.775,"train_time":26.38,"predict_time":0.302,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-83","imputed":false},{"position":44,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1200.0,"score":0.06,"rank":43.35,"hrank":33.33,"imp":13.592,"train_time":3084.49,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-173","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1196.0,"score":0.072,"rank":43.56,"hrank":34.74,"imp":11.974,"train_time":5151.88,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-151","imputed":false},{"position":46,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1193.0,"score":0.0,"rank":43.8,"hrank":41.82,"imp":11.947,"train_time":4.3,"predict_time":0.317,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-44","imputed":false},{"position":47,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1163.0,"score":0.042,"rank":45.75,"hrank":41.34,"imp":15.148,"train_time":675.17,"predict_time":1.54,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-144","imputed":false},{"position":48,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1159.0,"score":0.037,"rank":45.96,"hrank":34.74,"imp":14.101,"train_time":3084.49,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-173","imputed":false},{"position":49,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1126.0,"score":0.018,"rank":48.03,"hrank":43.65,"imp":15.743,"train_time":675.17,"predict_time":0.206,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-148","imputed":false},{"position":50,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1125.0,"score":0.0,"rank":48.09,"hrank":46.6,"imp":13.15,"train_time":4.06,"predict_time":0.299,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-50","imputed":false},{"position":51,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1120.0,"score":0.0,"rank":48.42,"hrank":45.78,"imp":15.121,"train_time":195.75,"predict_time":0.36,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-127","imputed":false},{"position":52,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1113.0,"score":0.035,"rank":48.87,"hrank":45.13,"imp":14.699,"train_time":2215.61,"predict_time":7.791,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-123","imputed":false},{"position":53,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1103.0,"score":0.05,"rank":49.43,"hrank":39.56,"imp":15.523,"train_time":10.2,"predict_time":2.716,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-204","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1100.0,"score":0.024,"rank":49.64,"hrank":40.69,"imp":16.458,"train_time":7.47,"predict_time":1.008,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-196","imputed":false},{"position":55,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1087.0,"score":0.013,"rank":50.4,"hrank":42.91,"imp":15.17,"train_time":12.26,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-152","imputed":false},{"position":56,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1080.0,"score":0.0,"rank":50.82,"hrank":48.2,"imp":16.417,"train_time":195.75,"predict_time":0.165,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-129","imputed":false},{"position":57,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1076.0,"score":0.182,"rank":51.07,"hrank":15.31,"imp":23.338,"train_time":52983.71,"predict_time":36.311,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+225\/-418","imputed":false},{"position":58,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1065.0,"score":0.173,"rank":51.69,"hrank":17.12,"imp":23.463,"train_time":52983.71,"predict_time":5.585,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+224\/-428","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1063.0,"score":0.019,"rank":51.8,"hrank":48.82,"imp":15.195,"train_time":2215.61,"predict_time":0.838,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-131","imputed":false},{"position":60,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1036.0,"score":0.0,"rank":53.29,"hrank":50.09,"imp":14.975,"train_time":22.4,"predict_time":0.134,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-160","imputed":false},{"position":61,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":55.22,"hrank":52.65,"imp":17.458,"train_time":0.63,"predict_time":0.12,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-139","imputed":false},{"position":62,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":952.0,"score":0.084,"rank":57.58,"hrank":24.77,"imp":26.2,"train_time":394.68,"predict_time":6.52,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+237\/-480","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":948.0,"score":0.0,"rank":57.75,"hrank":48.03,"imp":21.623,"train_time":88.43,"predict_time":0.512,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+158\/-331","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":940.0,"score":0.0,"rank":58.14,"hrank":56.41,"imp":19.754,"train_time":6.97,"predict_time":0.818,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-178","imputed":false},{"position":65,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":909.0,"score":0.0,"rank":59.5,"hrank":58.52,"imp":19.744,"train_time":32.6,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-147","imputed":false},{"position":66,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":848.0,"score":0.0,"rank":61.9,"hrank":56.19,"imp":23.914,"train_time":88.43,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+153\/-326","imputed":false},{"position":67,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":842.0,"score":0.0,"rank":62.12,"hrank":61.62,"imp":20.624,"train_time":725.45,"predict_time":268.761,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+43\/-122","imputed":false},{"position":68,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":768.0,"score":0.0,"rank":64.48,"hrank":48.5,"imp":30.791,"train_time":0.42,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+172\/-428","imputed":false},{"position":69,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":610.0,"score":0.0,"rank":68.08,"hrank":67.69,"imp":35.056,"train_time":777.82,"predict_time":0.312,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-386","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":600.0,"score":0.0,"rank":68.26,"hrank":67.84,"imp":35.099,"train_time":777.82,"predict_time":0.108,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+109\/-435","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":450.0,"score":0.0,"rank":70.33,"hrank":70.28,"imp":37.04,"train_time":3.91,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-404","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_small/pareto_front_explorer.html
index d1ffa038c212f9014b821553572f6d2522accd41..7d7f0938e69b4e28f78fddda88b8a321f30f591d 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0489086629,"imp":9.9048780799,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0627456348,"imp":7.9131545823,"elo":1459.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4436888454,"imp":7.7399538345,"elo":1487.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0368319183,"imp":11.6492392681,"elo":1233.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0290955235,"imp":10.6976848538,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.348849021,"imp":10.3571039182,"elo":1338.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0130181103,"imp":15.1697306884,"elo":1087.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0122824328,"imp":14.1012788622,"elo":1159.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1263863053,"imp":13.5922719913,"elo":1199.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0076902476,"imp":16.45840819,"elo":1099.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2883404835,"imp":13.63676692,"elo":1240.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":2.0978537644,"imp":13.391230847,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8180850143,"imp":19.7538701218,"elo":939.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.837804851,"imp":15.1954491766,"elo":1063.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7913188557,"imp":14.6986352945,"elo":1112.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.3174685594,"imp":11.9468322377,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9564936689,"imp":9.3413542906,"elo":1383.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":7.1056715587,"imp":8.586431531,"elo":1465.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":21.8030250918,"imp":11.3325397182,"elo":1243.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3020207606,"imp":12.7750476014,"elo":1213.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.420913606,"imp":11.0430349425,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0116272048,"imp":8.9646947928,"elo":1457.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.5662902939,"imp":3.4262379771,"elo":1783.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0528542696,"imp":1.8817397668,"elo":1859.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.021441561,"imp":19.7443842658,"elo":908.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1652630526,"imp":16.4170814909,"elo":1080.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3602092205,"imp":15.1213017458,"elo":1120.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1203913506,"imp":17.4581516042,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2060693731,"imp":15.7433625906,"elo":1126.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.5403575271,"imp":15.1483548251,"elo":1162.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":8.3504595229,"imp":9.6740737685,"elo":1276.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":2.2331214774,"imp":5.9601715191,"elo":1547.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":33.8307336671,"imp":5.0356106323,"elo":1637.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5470075607,"imp":5.9884825994,"elo":1505.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9019759916,"imp":4.1568016144,"elo":1574.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.2264004569,"imp":3.2431455822,"elo":1705.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.0380986551,"imp":10.0251733039,"elo":1367.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.001452678,"imp":6.1915523102,"elo":1580.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":22.311628541,"imp":5.0897446565,"elo":1674.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":140.7371116271,"imp":4.6251613505,"elo":1720.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2636901296,"imp":5.9967202116,"elo":1571.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":4.0942168713,"imp":2.1396827742,"elo":1966.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1402308582,"imp":4.3345316161,"elo":1666.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4666328992,"imp":8.8736398817,"elo":1303.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4918252514,"imp":7.5961460002,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.7546477722,"imp":6.5836727089,"elo":1457.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5852250627,"imp":3.8460415674,"elo":1739.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4388252423,"imp":2.5896127992,"elo":1797.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5202853799,"imp":26.1997788493,"elo":951.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5850454428,"imp":23.4629535368,"elo":1064.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.3108208411,"imp":23.3376373688,"elo":1075.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0639725065,"imp":6.0594250563,"elo":1481.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1340751587,"imp":14.9749307256,"elo":1036.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1363485289,"imp":11.9744442081,"elo":1196.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5906165471,"imp":11.3167357316,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2989233835,"imp":13.150142211,"elo":1125.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4754931932,"imp":9.0529333567,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.1465188849,"imp":8.8911196906,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":268.7612898715,"imp":20.6238140465,"elo":841.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":167.0374934194,"imp":9.245814462,"elo":1350.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1519.4791766157,"imp":9.0430335915,"elo":1489.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":2.7158548006,"imp":15.5226644141,"elo":1103.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1085198178,"imp":8.5503811628,"elo":1362.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.1419404045,"imp":8.0553116565,"elo":1438.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0489086629,"x_train":13.6316706327,"imp":9.9048780799,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0627456348,"x_train":4283.1043461266,"imp":7.9131545823,"elo":1459.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4436888454,"x_train":4283.1043461266,"imp":7.7399538345,"elo":1487.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0368319183,"x_train":8.2896427598,"imp":11.6492392681,"elo":1233.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0290955235,"x_train":1846.4337432747,"imp":10.6976848538,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.348849021,"x_train":1846.4337432747,"imp":10.3571039182,"elo":1338.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0130181103,"x_train":12.2643063605,"imp":15.1697306884,"elo":1087.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0122824328,"x_train":3084.4850399813,"imp":14.1012788622,"elo":1159.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1263863053,"x_train":3084.4850399813,"imp":13.5922719913,"elo":1199.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0076902476,"x_train":7.4740218671,"imp":16.45840819,"elo":1099.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2883404835,"x_train":1163.6859454819,"imp":13.63676692,"elo":1240.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":2.0978537644,"x_train":1163.6859454819,"imp":13.391230847,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8180850143,"x_train":6.9730562426,"imp":19.7538701218,"elo":939.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.837804851,"x_train":2215.6113217439,"imp":15.1954491766,"elo":1063.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7913188557,"x_train":2215.6113217439,"imp":14.6986352945,"elo":1112.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.3174685594,"x_train":4.301698659,"imp":11.9468322377,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9564936689,"x_train":1203.2369028783,"imp":9.3413542906,"elo":1383.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":7.1056715587,"x_train":1203.2369028783,"imp":8.586431531,"elo":1465.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":21.8030250918,"x_train":121.7676060282,"imp":11.3325397182,"elo":1243.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3020207606,"x_train":26.3794580707,"imp":12.7750476014,"elo":1213.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.420913606,"x_train":8982.7355846534,"imp":11.0430349425,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0116272048,"x_train":8982.7355846534,"imp":8.9646947928,"elo":1457.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.5662902939,"x_train":12.3703933426,"imp":3.4262379771,"elo":1783.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0528542696,"x_train":12.0914930116,"imp":1.8817397668,"elo":1859.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.021441561,"x_train":32.5994632136,"imp":19.7443842658,"elo":908.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1652630526,"x_train":195.7517175714,"imp":16.4170814909,"elo":1080.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3602092205,"x_train":195.7517175714,"imp":15.1213017458,"elo":1120.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1203913506,"x_train":0.6268352454,"imp":17.4581516042,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2060693731,"x_train":675.1680482075,"imp":15.7433625906,"elo":1126.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.5403575271,"x_train":675.1680482075,"imp":15.1483548251,"elo":1162.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":8.3504595229,"x_train":21.5151248338,"imp":9.6740737685,"elo":1276.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":2.2331214774,"x_train":4560.0060476784,"imp":5.9601715191,"elo":1547.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":33.8307336671,"x_train":4560.0060476784,"imp":5.0356106323,"elo":1637.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5470075607,"x_train":7.6690681838,"imp":5.9884825994,"elo":1505.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9019759916,"x_train":2307.574171405,"imp":4.1568016144,"elo":1574.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.2264004569,"x_train":2307.574171405,"imp":3.2431455822,"elo":1705.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.0380986551,"x_train":22.4635991568,"imp":10.0251733039,"elo":1367.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.001452678,"x_train":48.1376226066,"imp":6.1915523102,"elo":1580.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":22.311628541,"x_train":4823.3684232511,"imp":5.0897446565,"elo":1674.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":140.7371116271,"x_train":4823.3684232511,"imp":4.6251613505,"elo":1720.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2636901296,"x_train":7.9333050678,"imp":5.9967202116,"elo":1571.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":4.0942168713,"x_train":49.2939082066,"imp":2.1396827742,"elo":1966.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1402308582,"x_train":5.1543015637,"imp":4.3345316161,"elo":1666.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4666328992,"x_train":13.5301358652,"imp":8.8736398817,"elo":1303.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4918252514,"x_train":3660.2557617825,"imp":7.5961460002,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.7546477722,"x_train":3660.2557617825,"imp":6.5836727089,"elo":1457.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5852250627,"x_train":9.2031778761,"imp":3.8460415674,"elo":1739.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4388252423,"x_train":14.0862798048,"imp":2.5896127992,"elo":1797.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5202853799,"x_train":394.6813525384,"imp":26.1997788493,"elo":951.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5850454428,"x_train":52983.7100836136,"imp":23.4629535368,"elo":1064.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.3108208411,"x_train":52983.7100836136,"imp":23.3376373688,"elo":1075.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0639725065,"x_train":3.5256005858,"imp":6.0594250563,"elo":1481.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1340751587,"x_train":22.3971569167,"imp":14.9749307256,"elo":1036.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1363485289,"x_train":5151.8777257105,"imp":11.9744442081,"elo":1196.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5906165471,"x_train":5151.8777257105,"imp":11.3167357316,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2989233835,"x_train":4.0580642834,"imp":13.150142211,"elo":1125.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4754931932,"x_train":1266.5674390493,"imp":9.0529333567,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.1465188849,"x_train":1266.5674390493,"imp":8.8911196906,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":268.7612898715,"x_train":725.4510762602,"imp":20.6238140465,"elo":841.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":167.0374934194,"x_train":39519.0698998597,"imp":9.245814462,"elo":1350.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1519.4791766157,"x_train":39519.0698998597,"imp":9.0430335915,"elo":1489.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":2.7158548006,"x_train":10.1998611956,"imp":15.5226644141,"elo":1103.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1085198178,"x_train":2039.6992814536,"imp":8.5503811628,"elo":1362.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.1419404045,"x_train":2039.6992814536,"imp":8.0553116565,"elo":1438.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_small/tuning_trajectories_explorer.html
index 8112247e46f20cd0793a7a604f41116491257b35..7111a2df22ea23a9a38cb7abd40e2b0a4a5d9501 100644
--- a/data/imputation_no/splits_all/tasks_regression/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_all/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":13.6316706327,"imp":9.9137641353,"elo":1369.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":36.2728602647,"imp":9.256007527,"elo":1413.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":97.6908201342,"imp":8.5425366427,"elo":1463.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":201.9357530623,"imp":8.2015057275,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":526.2527558948,"imp":7.8894665208,"elo":1496.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":1057.486905704,"imp":7.8265992151,"elo":1500.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":2119.3859898824,"imp":7.7826718622,"elo":1505.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":3172.1186602631,"imp":7.7712354147,"elo":1506.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":4283.1043461266,"imp":7.7494769214,"elo":1510.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":8.2896427598,"imp":11.6578043119,"elo":1239.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":17.6287133887,"imp":11.4444673924,"elo":1255.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":44.8911284914,"imp":11.2222880026,"elo":1279.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":90.9274231145,"imp":11.0290777794,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":228.5740056066,"imp":10.8139410189,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":458.5111602981,"imp":10.6485131439,"elo":1329.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":918.1582493662,"imp":10.4831559376,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1376.6746665085,"imp":10.4138978023,"elo":1343.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1846.4337432747,"imp":10.3651488558,"elo":1350.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":12.2643063605,"imp":15.177953977,"elo":1089.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":29.8736292865,"imp":14.7951073958,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":80.8890311763,"imp":14.4059169952,"elo":1156.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":157.9967250155,"imp":14.1485660099,"elo":1175.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":390.307349424,"imp":13.8243982511,"elo":1192.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":781.3223148424,"imp":13.736345614,"elo":1198.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1537.3695296789,"imp":13.6645353412,"elo":1203.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":2294.6159909916,"imp":13.6275919525,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":3084.4850399813,"imp":13.6001739934,"elo":1210.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.4740218671,"imp":16.4667238266,"elo":1104.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.043565096,"imp":15.9556978664,"elo":1139.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":33.9390995628,"imp":15.0140788435,"elo":1191.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":59.6650516119,"imp":14.4893396637,"elo":1218.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":145.2899663332,"imp":13.6978237839,"elo":1253.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":289.9500533509,"imp":13.5277865157,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":579.1158947796,"imp":13.4401179876,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":868.4436513583,"imp":13.4125775111,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1163.6859454819,"imp":13.4005540235,"elo":1278.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.9730562426,"imp":19.7613367928,"elo":937.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":17.4702969526,"imp":19.4416409119,"elo":947.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":51.586202932,"imp":18.3266573768,"elo":988.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":104.4045259859,"imp":17.7456878302,"elo":1013.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":268.392593962,"imp":16.883908985,"elo":1054.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":544.3008720644,"imp":16.4101922717,"elo":1072.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1091.6618186196,"imp":15.6895356163,"elo":1093.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1644.5558204956,"imp":15.1614056635,"elo":1111.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2215.6113217439,"imp":14.7071910512,"elo":1128.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":4.301698659,"imp":11.9558623854,"elo":1196.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":10.0806694886,"imp":11.1394661698,"elo":1257.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":27.1783376768,"imp":10.0667658183,"elo":1342.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":56.2958523673,"imp":9.5401438565,"elo":1392.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":147.4003920877,"imp":9.0389294151,"elo":1445.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":297.8210784396,"imp":8.8855958273,"elo":1464.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":595.7506287926,"imp":8.7372036506,"elo":1480.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":891.9636616374,"imp":8.6531555811,"elo":1487.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1203.2369028783,"imp":8.5960607692,"elo":1491.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":121.7676060282,"imp":11.3419831057,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":26.3794580707,"imp":12.7838472277,"elo":1222.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":74.5462848654,"imp":11.9261311008,"elo":1277.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":211.6307745864,"imp":10.5318674518,"elo":1368.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":427.0209555269,"imp":9.9628311423,"elo":1410.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":1036.8544848991,"imp":9.3775435691,"elo":1451.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":2221.9585559949,"imp":9.2160691498,"elo":1462.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":4422.440532253,"imp":9.054510625,"elo":1472.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":6612.381411833,"imp":8.9810244016,"elo":1474.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":8982.7355846534,"imp":8.9738281558,"elo":1475.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":12.3703933426,"imp":3.4364803919,"elo":1807.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":12.0914930116,"imp":1.8923916968,"elo":1875.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":32.5994632136,"imp":19.7519006507,"elo":920.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":65.3760935075,"imp":17.4553768661,"elo":1015.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":163.2396739605,"imp":15.302158711,"elo":1111.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":195.7517175714,"imp":15.1292270615,"elo":1121.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.6268352454,"imp":17.4662287165,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.8932399565,"imp":16.7018476864,"elo":1049.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":13.9716136863,"imp":16.1488475314,"elo":1091.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":30.8300578438,"imp":15.6411491376,"elo":1126.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":81.3676986662,"imp":15.3437400288,"elo":1150.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":165.6231513559,"imp":15.2502962209,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":334.2669121033,"imp":15.1874564088,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":502.9575285794,"imp":15.167409788,"elo":1173.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":675.1680482075,"imp":15.1567144537,"elo":1176.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":21.5151248338,"imp":9.6840256411,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":44.1500541097,"imp":8.1759566082,"elo":1375.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":110.9865188464,"imp":6.4473112755,"elo":1511.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":225.999805441,"imp":5.8769896143,"elo":1571.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":564.497352775,"imp":5.4930210655,"elo":1610.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1133.451985238,"imp":5.2958558253,"elo":1633.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2267.2333864899,"imp":5.1737537053,"elo":1644.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3401.9727496793,"imp":5.0907691165,"elo":1657.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4560.0060476784,"imp":5.0459604181,"elo":1666.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":7.6690681838,"imp":5.9987097875,"elo":1519.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":19.8560696815,"imp":5.4167466604,"elo":1544.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":55.2048577583,"imp":4.4298123538,"elo":1598.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":111.3159815209,"imp":3.9971605146,"elo":1642.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":289.4641949987,"imp":3.612274262,"elo":1683.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":572.0729326007,"imp":3.4298765409,"elo":1704.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1145.5080925273,"imp":3.3339578375,"elo":1716.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1725.0191110982,"imp":3.2905831981,"elo":1722.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2307.574171405,"imp":3.2541229137,"elo":1728.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":22.4635991568,"imp":10.0342491225,"elo":1385.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":48.1376226066,"imp":6.2018198907,"elo":1593.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":69.1348663267,"imp":5.6301013057,"elo":1636.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":132.3528960267,"imp":5.1733798325,"elo":1685.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":250.1973072075,"imp":4.8541288356,"elo":1722.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":607.0781754894,"imp":4.6598742707,"elo":1745.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1214.0044600928,"imp":4.6029185428,"elo":1753.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2403.8929291273,"imp":4.5979340591,"elo":1753.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3586.3826643416,"imp":4.6218785712,"elo":1749.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4823.3684232511,"imp":4.6347319983,"elo":1745.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":7.9333050678,"imp":6.0066080393,"elo":1585.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":49.2939082066,"imp":2.1499497953,"elo":1991.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":5.1543015637,"imp":4.3447191725,"elo":1684.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":13.5301358652,"imp":8.8831220405,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":30.4154141023,"imp":8.6009869841,"elo":1320.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":85.0731093685,"imp":8.5835958528,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":178.9262812137,"imp":8.0884798338,"elo":1378.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":451.7833076823,"imp":7.6077544228,"elo":1415.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":903.7701842028,"imp":7.3002689033,"elo":1439.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1809.2442565983,"imp":7.0821873345,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2725.2737601417,"imp":6.6401544424,"elo":1468.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3660.2557617825,"imp":6.5924418585,"elo":1475.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":9.2031778761,"imp":3.8561978825,"elo":1765.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":14.0862798048,"imp":2.6102869791,"elo":1818.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":394.6813525384,"imp":26.2073736825,"elo":958.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":831.1606530777,"imp":26.1077225867,"elo":973.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1632.8165586183,"imp":25.8690920625,"elo":989.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2696.2641572437,"imp":25.3416320493,"elo":1019.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":6500.6961192202,"imp":24.6171733463,"elo":1044.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":13195.5862950654,"imp":24.0526355022,"elo":1062.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":26552.7976576922,"imp":23.6994048184,"elo":1076.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":39668.7876205984,"imp":23.4972299517,"elo":1086.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":52983.7100836136,"imp":23.3455218484,"elo":1096.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":3.5256005858,"imp":6.0697756523,"elo":1493.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":22.3971569167,"imp":14.9839018901,"elo":1048.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":48.9969305386,"imp":14.5422682458,"elo":1082.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":122.7996474091,"imp":13.7508877773,"elo":1127.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":260.4618741099,"imp":13.3481466416,"elo":1153.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":644.8021976959,"imp":12.7053346948,"elo":1193.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1311.6428242239,"imp":12.205568024,"elo":1227.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2562.5226347774,"imp":11.6928687242,"elo":1251.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3859.2194758298,"imp":11.4735619297,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":5151.8777257105,"imp":11.3257534765,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":4.0580642834,"imp":13.1588554842,"elo":1128.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":10.6096996819,"imp":10.8956630362,"elo":1262.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":30.177444805,"imp":10.0337625748,"elo":1327.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":60.5203899854,"imp":9.5754833132,"elo":1371.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":156.9516526756,"imp":9.2673423875,"elo":1402.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":314.3467912249,"imp":9.1368727699,"elo":1415.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":629.7467608773,"imp":9.0366271503,"elo":1427.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":943.5999846765,"imp":8.9485789399,"elo":1434.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":1266.5674390493,"imp":8.9004852158,"elo":1441.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":725.4510762602,"imp":20.6315087964,"elo":856.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":2170.0049801645,"imp":15.2410193927,"elo":1059.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":6794.2091341246,"imp":13.7690785529,"elo":1266.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":15326.7493172037,"imp":13.7135761398,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":37614.2771727845,"imp":9.0936981548,"elo":1500.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":39519.0698998597,"imp":9.0532334462,"elo":1508.1,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":10.1998611956,"imp":15.5314668275,"elo":1117.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":20.3286221704,"imp":13.8989092697,"elo":1187.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":50.8497537215,"imp":11.7715909652,"elo":1279.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":101.4913596974,"imp":10.2503880074,"elo":1336.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":253.6820748947,"imp":8.9371273466,"elo":1395.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":507.7670435057,"imp":8.4940514134,"elo":1426.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":1015.1060714549,"imp":8.2143272322,"elo":1442.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1522.261390547,"imp":8.1189239429,"elo":1451.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2039.6992814536,"imp":8.0647090249,"elo":1456.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0489086629,"x_train":13.6316706327,"imp":9.9137641353,"elo":1369.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1144842333,"x_train":36.2728602647,"imp":9.256007527,"elo":1413.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2272990226,"x_train":97.6908201342,"imp":8.5425366427,"elo":1463.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2840630283,"x_train":201.9357530623,"imp":8.2015057275,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3167883742,"x_train":526.2527558948,"imp":7.8894665208,"elo":1496.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3510472721,"x_train":1057.486905704,"imp":7.8265992151,"elo":1500.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3908307212,"x_train":2119.3859898824,"imp":7.7826718622,"elo":1505.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4190199844,"x_train":3172.1186602631,"imp":7.7712354147,"elo":1506.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4436888454,"x_train":4283.1043461266,"imp":7.7494769214,"elo":1510.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0368319183,"x_train":8.2896427598,"imp":11.6578043119,"elo":1239.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0688276247,"x_train":17.6287133887,"imp":11.4444673924,"elo":1255.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1219067073,"x_train":44.8911284914,"imp":11.2222880026,"elo":1279.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1714117234,"x_train":90.9274231145,"imp":11.0290777794,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2379383628,"x_train":228.5740056066,"imp":10.8139410189,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2766475086,"x_train":458.5111602981,"imp":10.6485131439,"elo":1329.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3155848723,"x_train":918.1582493662,"imp":10.4831559376,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3354039341,"x_train":1376.6746665085,"imp":10.4138978023,"elo":1343.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.348849021,"x_train":1846.4337432747,"imp":10.3651488558,"elo":1350.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0130181103,"x_train":12.2643063605,"imp":15.177953977,"elo":1089.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0248159619,"x_train":29.8736292865,"imp":14.7951073958,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0457946914,"x_train":80.8890311763,"imp":14.4059169952,"elo":1156.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0624903018,"x_train":157.9967250155,"imp":14.1485660099,"elo":1175.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0826354068,"x_train":390.307349424,"imp":13.8243982511,"elo":1192.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0975020907,"x_train":781.3223148424,"imp":13.736345614,"elo":1198.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1099002096,"x_train":1537.3695296789,"imp":13.6645353412,"elo":1203.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1183419318,"x_train":2294.6159909916,"imp":13.6275919525,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1269231171,"x_train":3084.4850399813,"imp":13.6001739934,"elo":1210.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0076902476,"x_train":7.4740218671,"imp":16.4667238266,"elo":1104.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1498964698,"x_train":14.043565096,"imp":15.9556978664,"elo":1139.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1014576612,"x_train":33.9390995628,"imp":15.0140788435,"elo":1191.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0082008847,"x_train":59.6650516119,"imp":14.4893396637,"elo":1218.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0626226355,"x_train":145.2899663332,"imp":13.6978237839,"elo":1253.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3379772641,"x_train":289.9500533509,"imp":13.5277865157,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.6772475427,"x_train":579.1158947796,"imp":13.4401179876,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.953467119,"x_train":868.4436513583,"imp":13.4125775111,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":2.0978537644,"x_train":1163.6859454819,"imp":13.4005540235,"elo":1278.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8180850143,"x_train":6.9730562426,"imp":19.7613367928,"elo":937.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.579872527,"x_train":17.4702969526,"imp":19.4416409119,"elo":947.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8040014596,"x_train":51.586202932,"imp":18.3266573768,"elo":988.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.9064906779,"x_train":104.4045259859,"imp":17.7456878302,"elo":1013.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6070665305,"x_train":268.392593962,"imp":16.883908985,"elo":1054.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.9054304219,"x_train":544.3008720644,"imp":16.4101922717,"elo":1072.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.5132807751,"x_train":1091.6618186196,"imp":15.6895356163,"elo":1093.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7654538542,"x_train":1644.5558204956,"imp":15.1614056635,"elo":1111.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7913188557,"x_train":2215.6113217439,"imp":14.7071910512,"elo":1128.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.3174685594,"x_train":4.301698659,"imp":11.9558623854,"elo":1196.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5015390954,"x_train":10.0806694886,"imp":11.1394661698,"elo":1257.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.8415814505,"x_train":27.1783376768,"imp":10.0667658183,"elo":1342.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.7464507056,"x_train":56.2958523673,"imp":9.5401438565,"elo":1392.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.9217323461,"x_train":147.4003920877,"imp":9.0389294151,"elo":1445.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.166466432,"x_train":297.8210784396,"imp":8.8855958273,"elo":1464.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.9020363863,"x_train":595.7506287926,"imp":8.7372036506,"elo":1480.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.08197779,"x_train":891.9636616374,"imp":8.6531555811,"elo":1487.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.1056715587,"x_train":1203.2369028783,"imp":8.5960607692,"elo":1491.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":21.8030250918,"x_train":121.7676060282,"imp":11.3419831057,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3020207606,"x_train":26.3794580707,"imp":12.7838472277,"elo":1222.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7507962656,"x_train":74.5462848654,"imp":11.9261311008,"elo":1277.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.6316502282,"x_train":211.6307745864,"imp":10.5318674518,"elo":1368.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.4921562567,"x_train":427.0209555269,"imp":9.9628311423,"elo":1410.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.017217154,"x_train":1036.8544848991,"imp":9.3775435691,"elo":1451.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.2418672637,"x_train":2221.9585559949,"imp":9.2160691498,"elo":1462.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.6456169825,"x_train":4422.440532253,"imp":9.054510625,"elo":1472.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5039594902,"x_train":6612.381411833,"imp":8.9810244016,"elo":1474.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0116272048,"x_train":8982.7355846534,"imp":8.9738281558,"elo":1475.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":1.5662902939,"x_train":12.3703933426,"imp":3.4364803919,"elo":1807.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0528542696,"x_train":12.0914930116,"imp":1.8923916968,"elo":1875.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.021441561,"x_train":32.5994632136,"imp":19.7519006507,"elo":920.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1447058964,"x_train":65.3760935075,"imp":17.4553768661,"elo":1015.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3551781139,"x_train":163.2396739605,"imp":15.302158711,"elo":1111.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3602092205,"x_train":195.7517175714,"imp":15.1292270615,"elo":1121.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1203913506,"x_train":0.6268352454,"imp":17.4662287165,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3503086463,"x_train":3.8932399565,"imp":16.7018476864,"elo":1049.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5821895355,"x_train":13.9716136863,"imp":16.1488475314,"elo":1091.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.806509101,"x_train":30.8300578438,"imp":15.6411491376,"elo":1126.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0727860502,"x_train":81.3676986662,"imp":15.3437400288,"elo":1150.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2227674298,"x_train":165.6231513559,"imp":15.2502962209,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4040008059,"x_train":334.2669121033,"imp":15.1874564088,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4924073803,"x_train":502.9575285794,"imp":15.167409788,"elo":1173.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.5448750192,"x_train":675.1680482075,"imp":15.1567144537,"elo":1176.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.3504595229,"x_train":21.5151248338,"imp":9.6840256411,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.3689378053,"x_train":44.1500541097,"imp":8.1759566082,"elo":1375.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.2357930967,"x_train":110.9865188464,"imp":6.4473112755,"elo":1511.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":18.2146470478,"x_train":225.999805441,"imp":5.8769896143,"elo":1571.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.1100006203,"x_train":564.497352775,"imp":5.4930210655,"elo":1610.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":24.4652884983,"x_train":1133.451985238,"imp":5.2958558253,"elo":1633.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":28.5926981827,"x_train":2267.2333864899,"imp":5.1737537053,"elo":1644.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":31.4173043227,"x_train":3401.9727496793,"imp":5.0907691165,"elo":1657.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":33.8307336671,"x_train":4560.0060476784,"imp":5.0459604181,"elo":1666.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5470075607,"x_train":7.6690681838,"imp":5.9987097875,"elo":1519.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4422544088,"x_train":19.8560696815,"imp":5.4167466604,"elo":1544.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.7604907751,"x_train":55.2048577583,"imp":4.4298123538,"elo":1598.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.6627807362,"x_train":111.3159815209,"imp":3.9971605146,"elo":1642.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.2418923926,"x_train":289.4641949987,"imp":3.612274262,"elo":1683.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.0570210776,"x_train":572.0729326007,"imp":3.4298765409,"elo":1704.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4887817681,"x_train":1145.5080925273,"imp":3.3339578375,"elo":1716.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.9791618808,"x_train":1725.0191110982,"imp":3.2905831981,"elo":1722.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.2648971866,"x_train":2307.574171405,"imp":3.2541229137,"elo":1728.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.0380986551,"x_train":22.4635991568,"imp":10.0342491225,"elo":1385.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":42.001452678,"x_train":48.1376226066,"imp":6.2018198907,"elo":1593.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":48.0919926375,"x_train":69.1348663267,"imp":5.6301013057,"elo":1636.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":55.5080414247,"x_train":132.3528960267,"imp":5.1733798325,"elo":1685.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":69.0757561523,"x_train":250.1973072075,"imp":4.8541288356,"elo":1722.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":89.4729687406,"x_train":607.0781754894,"imp":4.6598742707,"elo":1745.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":114.5105718552,"x_train":1214.0044600928,"imp":4.6029185428,"elo":1753.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":141.2613310565,"x_train":2403.8929291273,"imp":4.5979340591,"elo":1753.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":143.7240278852,"x_train":3586.3826643416,"imp":4.6218785712,"elo":1749.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":140.7371116271,"x_train":4823.3684232511,"imp":4.6347319983,"elo":1745.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2636901296,"x_train":7.9333050678,"imp":6.0066080393,"elo":1585.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":4.0942168713,"x_train":49.2939082066,"imp":2.1499497953,"elo":1991.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1402308582,"x_train":5.1543015637,"imp":4.3447191725,"elo":1684.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4666328992,"x_train":13.5301358652,"imp":8.8831220405,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.9441490344,"x_train":30.4154141023,"imp":8.6009869841,"elo":1320.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.8546554659,"x_train":85.0731093685,"imp":8.5835958528,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.4234961622,"x_train":178.9262812137,"imp":8.0884798338,"elo":1378.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":3.1588401041,"x_train":451.7833076823,"imp":7.6077544228,"elo":1415.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.744707909,"x_train":903.7701842028,"imp":7.3002689033,"elo":1439.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":4.3265393551,"x_train":1809.2442565983,"imp":7.0821873345,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.5964870629,"x_train":2725.2737601417,"imp":6.6401544424,"elo":1468.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.7546477722,"x_train":3660.2557617825,"imp":6.5924418585,"elo":1475.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5852250627,"x_train":9.2031778761,"imp":3.8561978825,"elo":1765.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4388252423,"x_train":14.0862798048,"imp":2.6102869791,"elo":1818.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5202853799,"x_train":394.6813525384,"imp":26.2073736825,"elo":958.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3982818311,"x_train":831.1606530777,"imp":26.1077225867,"elo":973.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0787209898,"x_train":1632.8165586183,"imp":25.8690920625,"elo":989.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3600489151,"x_train":2696.2641572437,"imp":25.3416320493,"elo":1019.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":23.7183592261,"x_train":6500.6961192202,"imp":24.6171733463,"elo":1044.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":28.12157704,"x_train":13195.5862950654,"imp":24.0526355022,"elo":1062.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":32.840614162,"x_train":26552.7976576922,"imp":23.6994048184,"elo":1076.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":34.8841403507,"x_train":39668.7876205984,"imp":23.4972299517,"elo":1086.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.3108208411,"x_train":52983.7100836136,"imp":23.3455218484,"elo":1096.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0639725065,"x_train":3.5256005858,"imp":6.0697756523,"elo":1493.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1340751587,"x_train":22.3971569167,"imp":14.9839018901,"elo":1048.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2671714395,"x_train":48.9969305386,"imp":14.5422682458,"elo":1082.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5043831924,"x_train":122.7996474091,"imp":13.7508877773,"elo":1127.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7072384242,"x_train":260.4618741099,"imp":13.3481466416,"elo":1153.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.979629298,"x_train":644.8021976959,"imp":12.7053346948,"elo":1193.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2101455751,"x_train":1311.6428242239,"imp":12.205568024,"elo":1227.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3941877423,"x_train":2562.5226347774,"imp":11.6928687242,"elo":1251.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5127765948,"x_train":3859.2194758298,"imp":11.4735619297,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5906165471,"x_train":5151.8777257105,"imp":11.3257534765,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2989233835,"x_train":4.0580642834,"imp":13.1588554842,"elo":1128.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8790913968,"x_train":10.6096996819,"imp":10.8956630362,"elo":1262.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.6245949041,"x_train":30.177444805,"imp":10.0337625748,"elo":1327.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0221234831,"x_train":60.5203899854,"imp":9.5754833132,"elo":1371.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.4398238016,"x_train":156.9516526756,"imp":9.2673423875,"elo":1402.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.655128881,"x_train":314.3467912249,"imp":9.1368727699,"elo":1415.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.9193143966,"x_train":629.7467608773,"imp":9.0366271503,"elo":1427.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.0384704369,"x_train":943.5999846765,"imp":8.9485789399,"elo":1434.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.1465188849,"x_train":1266.5674390493,"imp":8.9004852158,"elo":1441.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":268.7612898715,"x_train":725.4510762602,"imp":20.6315087964,"elo":856.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":565.8509098848,"x_train":2170.0049801645,"imp":15.2410193927,"elo":1059.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":1025.1468177864,"x_train":6794.2091341246,"imp":13.7690785529,"elo":1266.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":1403.2757745602,"x_train":15326.7493172037,"imp":13.7135761398,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":1492.6468598574,"x_train":37614.2771727845,"imp":9.0936981548,"elo":1500.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":1519.4791766157,"x_train":39519.0698998597,"imp":9.0532334462,"elo":1508.1,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":2.7158548006,"x_train":10.1998611956,"imp":15.5314668275,"elo":1117.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":2.8570323643,"x_train":20.3286221704,"imp":13.8989092697,"elo":1187.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.0885835567,"x_train":50.8497537215,"imp":11.7715909652,"elo":1279.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":2.1746682726,"x_train":101.4913596974,"imp":10.2503880074,"elo":1336.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.1322181978,"x_train":253.6820748947,"imp":8.9371273466,"elo":1395.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.9099808138,"x_train":507.7670435057,"imp":8.4940514134,"elo":1426.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.6891571227,"x_train":1015.1060714549,"imp":8.2143272322,"elo":1442.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.443541988,"x_train":1522.261390547,"imp":8.1189239429,"elo":1451.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.1468082511,"x_train":2039.6992814536,"imp":8.0647090249,"elo":1456.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_small/winrate_explorer.html b/data/imputation_no/splits_all/tasks_regression/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..e9231ac27d81bdff1217104dcd0868b48875e68c
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_all/tasks_regression/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "Nori-30M (default)", "TabPFN-3 (default)", "Nori (default)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "TabDPT (tuned + ensembled)", "RealMLP (tuned + ensembled)", "TabDPT (tuned)", "RealTabPFN-2.5 (tuned)", "TabDPT (default)", "RealTabPFN-2.5 (default)", "RealMLP (tuned)", "TabDPT-Turbo (default)", "AutoGluon 1.4 (best, 4h)", "CatBoost (tuned + ensembled)", "TabSwift (default)", "iLTM (tuned + ensembled)", "CatBoost (tuned)", "TabM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "SAP-RPT-OSS (default)", "xRFM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "XGBoost (tuned)", "CatBoost (default)", "LightGBM (tuned)", "TabM (tuned)", "xRFM (tuned)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "TabM (default)", "ChimeraBoost (tuned)", "RealMLP (default)", "ExtraTrees (tuned + ensembled)", "ModernNCA (tuned)", "EBM (tuned + ensembled)", "ExtraTrees (tuned)", "ChimeraBoost (default)", "TorchMLP (tuned + ensembled)", "LimiX (default)", "EBM (tuned)", "TorchMLP (tuned)", "ModernNCA (default)", "LightGBM (default)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned + ensembled)", "TabSTAR (tuned)", "EBM (default)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "XGBoost (default)", "FastaiMLP (tuned + ensembled)", "xRFM (default)", "PerpetualBooster (tuned)", "TorchMLP (default)", "FastaiMLP (tuned)", "TabSTAR (default)", "RandomForest (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "FastaiMLP (default)", "KNN (tuned)", "iLTM (default)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 0.5141242742538452, 0.6892655491828918, 0.7457627058029175, 0.7062146663665771, 0.7683615684509277, 0.7118644118309021, 0.7401130199432373, 0.8757061958312988, 0.8757061958312988, 0.9209039807319641, 0.7683615684509277, 0.8926553726196289, 0.9152542352676392, 0.9322034120559692, 0.9548022747039795, 0.9491525292396545, 0.9548022747039795, 0.8700565099716187, 0.9548022747039795, 0.9717513918876648, 0.9435028433799744, 0.9661017060279846, 0.9152542352676392, 0.9378530979156494, 0.9548022747039795, 0.9491525292396545, 0.9661017060279846, 0.9604519605636597, 0.9717513918876648, 0.9604519605636597, 0.9604519605636597, 0.9661017060279846, 0.9717513918876648, 0.9717513918876648, 0.9604519605636597, 0.9887005686759949, 0.9661017060279846, 0.9887005686759949, 0.9661017060279846, 0.9717513918876648, 0.994350254535675, 0.9887005686759949, 0.9830508232116699, 0.9774011373519897, 0.994350254535675, 0.9887005686759949, 1.0, 0.903954803943634, 0.994350254535675, 0.909604549407959, 0.9830508232116699, 0.994350254535675, 0.9887005686759949, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 0.994350254535675, 1.0, 1.0, 0.9265536665916443, 0.994350254535675, 0.9830508232116699, 1.0, 1.0, 0.994350254535675, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.4858756959438324, NaN, 0.5480226278305054, 0.7231638431549072, 0.6327683329582214, 0.7683615684509277, 0.7344632744789124, 0.7344632744789124, 0.7062146663665771, 0.7796609997749329, 0.790960431098938, 0.8135592937469482, 0.790960431098938, 0.8700565099716187, 0.8700565099716187, 0.8644067645072937, 0.909604549407959, 0.9265536665916443, 0.903954803943634, 0.887005627155304, 0.9604519605636597, 0.9209039807319641, 0.9548022747039795, 0.8813559412956238, 0.9152542352676392, 0.9378530979156494, 0.903954803943634, 0.9491525292396545, 0.9604519605636597, 0.9661017060279846, 0.9435028433799744, 0.9435028433799744, 0.9774011373519897, 0.9774011373519897, 0.9661017060279846, 0.9830508232116699, 0.9661017060279846, 0.9548022747039795, 0.9491525292396545, 0.9830508232116699, 0.9378530979156494, 0.9774011373519897, 0.9661017060279846, 0.9774011373519897, 0.9830508232116699, 0.9830508232116699, 0.9774011373519897, 0.994350254535675, 0.8926553726196289, 0.9830508232116699, 0.903954803943634, 0.9887005686759949, 0.9830508232116699, 0.9887005686759949, 0.9604519605636597, 0.994350254535675, 0.994350254535675, 0.9661017060279846, 0.994350254535675, 1.0, 1.0, 0.909604549407959, 0.9887005686759949, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9774011373519897, 1.0, 1.0, 1.0], [0.31073445081710815, 0.451977401971817, NaN, 0.5536723136901855, 0.5593220591545105, 0.598870038986206, 0.604519784450531, 0.5819209218025208, 0.7062146663665771, 0.790960431098938, 0.7288135886192322, 0.7457627058029175, 0.8135592937469482, 0.8474576473236084, 0.8305084705352783, 0.887005627155304, 0.8531073331832886, 0.8700565099716187, 0.9209039807319641, 0.887005627155304, 0.8757061958312988, 0.8644067645072937, 0.909604549407959, 0.8418079018592834, 0.8926553726196289, 0.909604549407959, 0.8700565099716187, 0.903954803943634, 0.8926553726196289, 0.9378530979156494, 0.903954803943634, 0.909604549407959, 0.8983050584793091, 0.9661017060279846, 0.9209039807319641, 0.903954803943634, 0.9604519605636597, 0.9322034120559692, 0.9209039807319641, 0.9265536665916443, 0.9435028433799744, 0.9491525292396545, 0.9661017060279846, 0.909604549407959, 0.9435028433799744, 0.9717513918876648, 0.9604519605636597, 0.9887005686759949, 0.8248587846755981, 0.9717513918876648, 0.8474576473236084, 0.9661017060279846, 0.9717513918876648, 0.9887005686759949, 0.9604519605636597, 0.9887005686759949, 1.0, 0.9548022747039795, 0.9887005686759949, 1.0, 1.0, 0.909604549407959, 0.994350254535675, 0.9604519605636597, 1.0, 1.0, 0.9774011373519897, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.2542372941970825, 0.2768361568450928, 0.44632768630981445, NaN, 0.49152541160583496, 0.6158192157745361, 0.5706214904785156, 0.5706214904785156, 0.6610169410705566, 0.6666666865348816, 0.700564980506897, 0.6610169410705566, 0.7344632744789124, 0.8587570786476135, 0.790960431098938, 0.8361582159996033, 0.8926553726196289, 0.9548022747039795, 0.7683615684509277, 0.8531073331832886, 0.9491525292396545, 0.8644067645072937, 0.9491525292396545, 0.8587570786476135, 0.903954803943634, 0.9209039807319641, 0.8813559412956238, 0.9378530979156494, 0.9491525292396545, 0.9491525292396545, 0.903954803943634, 0.9491525292396545, 0.9604519605636597, 0.9774011373519897, 0.9491525292396545, 0.9717513918876648, 0.9209039807319641, 0.9378530979156494, 0.9435028433799744, 0.9774011373519897, 0.9378530979156494, 0.9830508232116699, 0.9548022747039795, 0.9322034120559692, 0.9661017060279846, 0.9830508232116699, 0.9717513918876648, 1.0, 0.8757061958312988, 0.9887005686759949, 0.8757061958312988, 0.9830508232116699, 0.9887005686759949, 0.9887005686759949, 0.9717513918876648, 0.994350254535675, 0.9830508232116699, 0.9604519605636597, 0.9887005686759949, 1.0, 0.9887005686759949, 0.903954803943634, 0.994350254535675, 0.9717513918876648, 1.0, 1.0, 0.9830508232116699, 1.0, 0.9774011373519897, 1.0, 1.0, 1.0], [0.29378530383110046, 0.3672316372394562, 0.4406779706478119, 0.508474588394165, NaN, 0.5762711763381958, 0.4971751272678375, 0.5480226278305054, 0.7288135886192322, 0.7457627058029175, 0.7231638431549072, 0.6836158037185669, 0.7627118825912476, 0.8474576473236084, 0.8248587846755981, 0.8305084705352783, 0.8587570786476135, 0.8983050584793091, 0.790960431098938, 0.8926553726196289, 0.8926553726196289, 0.8926553726196289, 0.8926553726196289, 0.887005627155304, 0.8757061958312988, 0.903954803943634, 0.8644067645072937, 0.8926553726196289, 0.903954803943634, 0.9152542352676392, 0.8983050584793091, 0.8926553726196289, 0.887005627155304, 0.9265536665916443, 0.903954803943634, 0.8926553726196289, 0.9322034120559692, 0.8983050584793091, 0.9209039807319641, 0.9378530979156494, 0.8926553726196289, 0.9378530979156494, 0.9435028433799744, 0.9152542352676392, 0.9322034120559692, 0.9265536665916443, 0.9265536665916443, 0.9491525292396545, 0.8361582159996033, 0.9604519605636597, 0.8531073331832886, 0.9774011373519897, 0.9717513918876648, 0.9661017060279846, 0.9491525292396545, 0.9491525292396545, 0.9491525292396545, 0.9661017060279846, 0.9661017060279846, 0.9491525292396545, 0.9491525292396545, 0.9152542352676392, 0.9661017060279846, 0.9774011373519897, 1.0, 0.9491525292396545, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.23163841664791107, 0.23163841664791107, 0.40112993121147156, 0.38418078422546387, 0.4237288236618042, NaN, 0.4350282549858093, 0.5254237055778503, 0.6158192157745361, 0.6158192157745361, 0.6497175097465515, 0.6497175097465515, 0.700564980506897, 0.8361582159996033, 0.7627118825912476, 0.7457627058029175, 0.8418079018592834, 0.8926553726196289, 0.8192090392112732, 0.8418079018592834, 0.9209039807319641, 0.903954803943634, 0.887005627155304, 0.8079096078872681, 0.8757061958312988, 0.9209039807319641, 0.8587570786476135, 0.9265536665916443, 0.9378530979156494, 0.9209039807319641, 0.909604549407959, 0.9435028433799744, 0.9435028433799744, 0.9661017060279846, 0.9491525292396545, 0.9548022747039795, 0.9661017060279846, 0.9209039807319641, 0.9322034120559692, 0.9661017060279846, 0.9322034120559692, 0.9830508232116699, 0.9491525292396545, 0.9265536665916443, 0.9774011373519897, 0.9830508232116699, 0.9830508232116699, 0.994350254535675, 0.8757061958312988, 0.9887005686759949, 0.887005627155304, 0.9774011373519897, 0.9887005686759949, 0.994350254535675, 0.9548022747039795, 1.0, 0.9774011373519897, 0.9717513918876648, 0.994350254535675, 1.0, 0.9887005686759949, 0.9209039807319641, 0.994350254535675, 0.9830508232116699, 1.0, 0.994350254535675, 0.9830508232116699, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.2881355881690979, 0.26553672552108765, 0.395480215549469, 0.42937853932380676, 0.5028248429298401, 0.5649717450141907, NaN, 0.49152541160583496, 0.6610169410705566, 0.7288135886192322, 0.6723163723945618, 0.6440678238868713, 0.7118644118309021, 0.7853107452392578, 0.8135592937469482, 0.8418079018592834, 0.8305084705352783, 0.8305084705352783, 0.8305084705352783, 0.8813559412956238, 0.8587570786476135, 0.887005627155304, 0.8587570786476135, 0.7683615684509277, 0.8474576473236084, 0.903954803943634, 0.8644067645072937, 0.9209039807319641, 0.8757061958312988, 0.9152542352676392, 0.903954803943634, 0.887005627155304, 0.903954803943634, 0.9265536665916443, 0.9265536665916443, 0.9265536665916443, 0.9717513918876648, 0.9209039807319641, 0.9548022747039795, 0.9548022747039795, 0.9265536665916443, 0.9322034120559692, 0.9830508232116699, 0.8757061958312988, 0.9548022747039795, 0.9830508232116699, 0.9661017060279846, 0.9717513918876648, 0.8700565099716187, 0.9604519605636597, 0.8757061958312988, 0.9435028433799744, 0.9661017060279846, 0.9887005686759949, 0.9378530979156494, 0.994350254535675, 0.994350254535675, 0.9661017060279846, 0.994350254535675, 1.0, 1.0, 0.9152542352676392, 1.0, 0.9717513918876648, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.2598870098590851, 0.26553672552108765, 0.41807910799980164, 0.42937853932380676, 0.451977401971817, 0.47457626461982727, 0.508474588394165, NaN, 0.6101694703102112, 0.6440678238868713, 0.6440678238868713, 0.8079096078872681, 0.7062146663665771, 0.8531073331832886, 0.7344632744789124, 0.7853107452392578, 0.8022598624229431, 0.8079096078872681, 0.8248587846755981, 0.7966101765632629, 0.8587570786476135, 0.887005627155304, 0.8587570786476135, 0.7683615684509277, 0.8135592937469482, 0.8700565099716187, 0.7853107452392578, 0.8813559412956238, 0.903954803943634, 0.887005627155304, 0.9209039807319641, 0.8587570786476135, 0.909604549407959, 0.903954803943634, 0.9491525292396545, 0.9265536665916443, 0.9378530979156494, 0.8531073331832886, 0.909604549407959, 0.9435028433799744, 0.8587570786476135, 0.9661017060279846, 0.9717513918876648, 0.8926553726196289, 0.9604519605636597, 0.9774011373519897, 0.9322034120559692, 0.9830508232116699, 0.8305084705352783, 0.9435028433799744, 0.8474576473236084, 0.9661017060279846, 0.9548022747039795, 0.9830508232116699, 0.9435028433799744, 0.9774011373519897, 1.0, 0.9152542352676392, 0.9830508232116699, 1.0, 1.0, 0.8983050584793091, 0.9887005686759949, 0.9604519605636597, 1.0, 1.0, 0.9717513918876648, 1.0, 0.9830508232116699, 1.0, 1.0, 1.0], [0.12429378181695938, 0.29378530383110046, 0.29378530383110046, 0.33898305892944336, 0.2711864411830902, 0.38418078422546387, 0.33898305892944336, 0.38983049988746643, NaN, 0.5649717450141907, 0.6158192157745361, 0.5593220591545105, 0.7231638431549072, 0.6892655491828918, 0.6384180784225464, 0.7796609997749329, 0.790960431098938, 0.8022598624229431, 0.7231638431549072, 0.8644067645072937, 0.8248587846755981, 0.7627118825912476, 0.8361582159996033, 0.7457627058029175, 0.8644067645072937, 0.8700565099716187, 0.8418079018592834, 0.903954803943634, 0.8813559412956238, 0.8644067645072937, 0.8305084705352783, 0.903954803943634, 0.8587570786476135, 0.9378530979156494, 0.8926553726196289, 0.8813559412956238, 0.9378530979156494, 0.9378530979156494, 0.9435028433799744, 0.887005627155304, 0.9265536665916443, 0.9152542352676392, 0.9209039807319641, 0.8926553726196289, 0.8757061958312988, 0.9491525292396545, 0.9548022747039795, 0.9717513918876648, 0.7853107452392578, 0.9774011373519897, 0.7853107452392578, 0.9435028433799744, 0.9830508232116699, 0.9830508232116699, 0.9548022747039795, 0.994350254535675, 1.0, 0.9491525292396545, 0.9830508232116699, 1.0, 1.0, 0.8926553726196289, 1.0, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.12429378181695938, 0.22033898532390594, 0.20903955399990082, 0.3333333432674408, 0.2542372941970825, 0.38418078422546387, 0.2711864411830902, 0.35593220591545105, 0.4350282549858093, NaN, 0.508474588394165, 0.47457626461982727, 0.5819209218025208, 0.6158192157745361, 0.7231638431549072, 0.6553672552108765, 0.7062146663665771, 0.7796609997749329, 0.6610169410705566, 0.8192090392112732, 0.7740113139152527, 0.7401130199432373, 0.7853107452392578, 0.7401130199432373, 0.8757061958312988, 0.8531073331832886, 0.8079096078872681, 0.8700565099716187, 0.8361582159996033, 0.8757061958312988, 0.8474576473236084, 0.9152542352676392, 0.8135592937469482, 0.9265536665916443, 0.8926553726196289, 0.8135592937469482, 0.9830508232116699, 0.9265536665916443, 0.9265536665916443, 0.8531073331832886, 0.9378530979156494, 0.8983050584793091, 0.8813559412956238, 0.8983050584793091, 0.8757061958312988, 0.9209039807319641, 0.9604519605636597, 0.9830508232116699, 0.7683615684509277, 0.9774011373519897, 0.7740113139152527, 0.9661017060279846, 0.9830508232116699, 0.9717513918876648, 0.9717513918876648, 0.994350254535675, 1.0, 0.9604519605636597, 0.9830508232116699, 0.994350254535675, 0.994350254535675, 0.8757061958312988, 0.994350254535675, 0.9604519605636597, 1.0, 1.0, 0.9774011373519897, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.07909604161977768, 0.20903955399990082, 0.2711864411830902, 0.299435019493103, 0.2768361568450928, 0.3502824902534485, 0.32768362760543823, 0.35593220591545105, 0.38418078422546387, 0.49152541160583496, NaN, 0.5310734510421753, 0.6384180784225464, 0.5819209218025208, 0.604519784450531, 0.6836158037185669, 0.7570621371269226, 0.7740113139152527, 0.6723163723945618, 0.8079096078872681, 0.7966101765632629, 0.700564980506897, 0.8022598624229431, 0.7457627058029175, 0.8305084705352783, 0.8474576473236084, 0.8248587846755981, 0.8700565099716187, 0.8418079018592834, 0.8700565099716187, 0.7853107452392578, 0.887005627155304, 0.8305084705352783, 0.9435028433799744, 0.8644067645072937, 0.8531073331832886, 0.9209039807319641, 0.9265536665916443, 0.9322034120559692, 0.8644067645072937, 0.9152542352676392, 0.9152542352676392, 0.9209039807319641, 0.903954803943634, 0.8757061958312988, 0.9378530979156494, 0.9548022747039795, 0.9774011373519897, 0.7740113139152527, 0.9604519605636597, 0.7796609997749329, 0.9322034120559692, 0.9661017060279846, 0.9774011373519897, 0.9491525292396545, 0.994350254535675, 0.994350254535675, 0.9548022747039795, 0.9774011373519897, 1.0, 1.0, 0.8983050584793091, 0.9887005686759949, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.23163841664791107, 0.18644067645072937, 0.2542372941970825, 0.33898305892944336, 0.3163841664791107, 0.3502824902534485, 0.35593220591545105, 0.19209039211273193, 0.4406779706478119, 0.5254237055778503, 0.4689265489578247, NaN, 0.5762711763381958, 0.6581920981407166, 0.604519784450531, 0.6553672552108765, 0.6327683329582214, 0.6779661178588867, 0.7118644118309021, 0.6779661178588867, 0.7118644118309021, 0.7401130199432373, 0.700564980506897, 0.6610169410705566, 0.7570621371269226, 0.7796609997749329, 0.7062146663665771, 0.8022598624229431, 0.7570621371269226, 0.7683615684509277, 0.8248587846755981, 0.8248587846755981, 0.7853107452392578, 0.8531073331832886, 0.8700565099716187, 0.8192090392112732, 0.8926553726196289, 0.8248587846755981, 0.8135592937469482, 0.8531073331832886, 0.8305084705352783, 0.887005627155304, 0.8926553726196289, 0.8361582159996033, 0.8813559412956238, 0.909604549407959, 0.8757061958312988, 0.9491525292396545, 0.7570621371269226, 0.8700565099716187, 0.7627118825912476, 0.9322034120559692, 0.887005627155304, 0.9435028433799744, 0.887005627155304, 0.9435028433799744, 0.9717513918876648, 0.8757061958312988, 0.9548022747039795, 0.994350254535675, 1.0, 0.8644067645072937, 0.9548022747039795, 0.9152542352676392, 1.0, 0.994350254535675, 0.9661017060279846, 1.0, 0.9604519605636597, 1.0, 1.0, 1.0], [0.10734463483095169, 0.20903955399990082, 0.18644067645072937, 0.26553672552108765, 0.23728813230991364, 0.299435019493103, 0.2881355881690979, 0.29378530383110046, 0.2768361568450928, 0.41807910799980164, 0.3615819215774536, 0.4237288236618042, NaN, 0.47457626461982727, 0.4858756959438324, 0.6101694703102112, 0.6271186470985413, 0.700564980506897, 0.6101694703102112, 0.7514124512672424, 0.7175140976905823, 0.5819209218025208, 0.700564980506897, 0.7062146663665771, 0.7683615684509277, 0.7514124512672424, 0.7627118825912476, 0.7853107452392578, 0.7627118825912476, 0.7796609997749329, 0.6553672552108765, 0.8022598624229431, 0.7627118825912476, 0.8757061958312988, 0.8022598624229431, 0.7740113139152527, 0.8531073331832886, 0.9322034120559692, 0.8813559412956238, 0.8135592937469482, 0.9378530979156494, 0.8248587846755981, 0.8757061958312988, 0.8700565099716187, 0.8305084705352783, 0.909604549407959, 0.9548022747039795, 0.9435028433799744, 0.7853107452392578, 0.9661017060279846, 0.7796609997749329, 0.9322034120559692, 0.9717513918876648, 0.9604519605636597, 0.9491525292396545, 0.994350254535675, 0.9830508232116699, 0.9604519605636597, 0.9830508232116699, 0.9887005686759949, 0.9830508232116699, 0.903954803943634, 0.994350254535675, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.08474576473236084, 0.12994350492954254, 0.1525423675775528, 0.14124293625354767, 0.1525423675775528, 0.16384181380271912, 0.21468926966190338, 0.14689265191555023, 0.31073445081710815, 0.38418078422546387, 0.41807910799980164, 0.34180790185928345, 0.5254237055778503, NaN, 0.5367231369018555, 0.6214689016342163, 0.6214689016342163, 0.694915235042572, 0.6101694703102112, 0.6779661178588867, 0.7514124512672424, 0.6610169410705566, 0.6836158037185669, 0.6384180784225464, 0.7457627058029175, 0.7570621371269226, 0.7401130199432373, 0.790960431098938, 0.7796609997749329, 0.8079096078872681, 0.7175140976905823, 0.8305084705352783, 0.8022598624229431, 0.8418079018592834, 0.8474576473236084, 0.8192090392112732, 0.790960431098938, 0.8305084705352783, 0.8474576473236084, 0.8587570786476135, 0.8418079018592834, 0.903954803943634, 0.8926553726196289, 0.8418079018592834, 0.8813559412956238, 0.8983050584793091, 0.8926553726196289, 0.9152542352676392, 0.7740113139152527, 0.8813559412956238, 0.7796609997749329, 0.9604519605636597, 0.8813559412956238, 0.9322034120559692, 0.8757061958312988, 0.9491525292396545, 0.9661017060279846, 0.8926553726196289, 0.9435028433799744, 1.0, 0.9774011373519897, 0.8700565099716187, 0.9322034120559692, 0.903954803943634, 1.0, 0.994350254535675, 0.9378530979156494, 1.0, 0.9491525292396545, 1.0, 1.0, 1.0], [0.06779661029577255, 0.12994350492954254, 0.16949152946472168, 0.20903955399990082, 0.17514124512672424, 0.23728813230991364, 0.18644067645072937, 0.26553672552108765, 0.3615819215774536, 0.2768361568450928, 0.395480215549469, 0.395480215549469, 0.5141242742538452, 0.46327683329582214, NaN, 0.5367231369018555, 0.5480226278305054, 0.6723163723945618, 0.5480226278305054, 0.6666666865348816, 0.6836158037185669, 0.6666666865348816, 0.6836158037185669, 0.6610169410705566, 0.694915235042572, 0.7514124512672424, 0.7118644118309021, 0.7627118825912476, 0.7231638431549072, 0.8079096078872681, 0.7683615684509277, 0.7796609997749329, 0.7570621371269226, 0.8361582159996033, 0.8418079018592834, 0.8022598624229431, 0.8757061958312988, 0.8135592937469482, 0.8587570786476135, 0.8135592937469482, 0.8474576473236084, 0.8531073331832886, 0.8644067645072937, 0.790960431098938, 0.8361582159996033, 0.9209039807319641, 0.9265536665916443, 0.9209039807319641, 0.7457627058029175, 0.9152542352676392, 0.7570621371269226, 0.9322034120559692, 0.9378530979156494, 0.9548022747039795, 0.903954803943634, 0.9717513918876648, 0.9717513918876648, 0.887005627155304, 0.9661017060279846, 0.9774011373519897, 0.9887005686759949, 0.8418079018592834, 0.9887005686759949, 0.9378530979156494, 0.9887005686759949, 0.9887005686759949, 0.9548022747039795, 1.0, 0.9717513918876648, 0.994350254535675, 0.994350254535675, 0.994350254535675], [0.0451977401971817, 0.1355932205915451, 0.11299435049295425, 0.16384181380271912, 0.16949152946472168, 0.2542372941970825, 0.15819208323955536, 0.21468926966190338, 0.22033898532390594, 0.3446327745914459, 0.3163841664791107, 0.3446327745914459, 0.38983049988746643, 0.3785310685634613, 0.46327683329582214, NaN, 0.604519784450531, 0.6723163723945618, 0.5706214904785156, 0.6779661178588867, 0.6892655491828918, 0.598870038986206, 0.6553672552108765, 0.6610169410705566, 0.7401130199432373, 0.7401130199432373, 0.7288135886192322, 0.7627118825912476, 0.7288135886192322, 0.7853107452392578, 0.6892655491828918, 0.7966101765632629, 0.7514124512672424, 0.8305084705352783, 0.8079096078872681, 0.7627118825912476, 0.8474576473236084, 0.887005627155304, 0.8474576473236084, 0.790960431098938, 0.8983050584793091, 0.8418079018592834, 0.887005627155304, 0.8474576473236084, 0.8248587846755981, 0.909604549407959, 0.909604549407959, 0.9265536665916443, 0.7288135886192322, 0.9265536665916443, 0.7401130199432373, 0.9152542352676392, 0.9265536665916443, 0.9435028433799744, 0.9209039807319641, 0.9661017060279846, 0.9774011373519897, 0.9322034120559692, 0.9661017060279846, 0.9887005686759949, 0.9887005686759949, 0.8587570786476135, 0.9830508232116699, 0.9604519605636597, 0.9887005686759949, 1.0, 0.9661017060279846, 1.0, 0.9717513918876648, 1.0, 1.0, 1.0], [0.050847455859184265, 0.0903954803943634, 0.14689265191555023, 0.10734463483095169, 0.14124293625354767, 0.15819208323955536, 0.16949152946472168, 0.1977401077747345, 0.20903955399990082, 0.29378530383110046, 0.2429378479719162, 0.3672316372394562, 0.37288135290145874, 0.3785310685634613, 0.451977401971817, 0.395480215549469, NaN, 0.604519784450531, 0.5480226278305054, 0.6158192157745361, 0.604519784450531, 0.5593220591545105, 0.7457627058029175, 0.6214689016342163, 0.6553672552108765, 0.7853107452392578, 0.7344632744789124, 0.8135592937469482, 0.7231638431549072, 0.8418079018592834, 0.6666666865348816, 0.7457627058029175, 0.790960431098938, 0.8418079018592834, 0.7401130199432373, 0.7966101765632629, 0.7288135886192322, 0.8361582159996033, 0.8474576473236084, 0.8135592937469482, 0.8474576473236084, 0.8813559412956238, 0.9152542352676392, 0.8192090392112732, 0.8474576473236084, 0.909604549407959, 0.9209039807319641, 0.9548022747039795, 0.7457627058029175, 0.8926553726196289, 0.7570621371269226, 0.903954803943634, 0.8983050584793091, 0.9604519605636597, 0.8983050584793091, 0.9774011373519897, 0.994350254535675, 0.909604549407959, 0.9548022747039795, 0.994350254535675, 1.0, 0.8474576473236084, 0.9717513918876648, 0.9209039807319641, 0.9887005686759949, 1.0, 0.9548022747039795, 1.0, 0.9661017060279846, 1.0, 1.0, 1.0], [0.0451977401971817, 0.07344632595777512, 0.12994350492954254, 0.0451977401971817, 0.10169491171836853, 0.10734463483095169, 0.16949152946472168, 0.19209039211273193, 0.1977401077747345, 0.22033898532390594, 0.2259887009859085, 0.32203391194343567, 0.299435019493103, 0.3050847351551056, 0.32768362760543823, 0.32768362760543823, 0.395480215549469, NaN, 0.5028248429298401, 0.48022598028182983, 0.6327683329582214, 0.4689265489578247, 0.6214689016342163, 0.5536723136901855, 0.5310734510421753, 0.7457627058029175, 0.6892655491828918, 0.7514124512672424, 0.790960431098938, 0.7514124512672424, 0.5762711763381958, 0.694915235042572, 0.7457627058029175, 0.7683615684509277, 0.6158192157745361, 0.8135592937469482, 0.694915235042572, 0.7683615684509277, 0.8135592937469482, 0.7966101765632629, 0.8079096078872681, 0.9209039807319641, 0.7853107452392578, 0.8361582159996033, 0.8418079018592834, 0.8079096078872681, 0.8926553726196289, 0.9435028433799744, 0.7457627058029175, 0.8983050584793091, 0.7457627058029175, 0.9209039807319641, 0.903954803943634, 0.9604519605636597, 0.8644067645072937, 0.994350254535675, 0.9209039807319641, 0.8983050584793091, 0.9774011373519897, 0.9322034120559692, 0.9604519605636597, 0.8361582159996033, 0.9830508232116699, 0.9322034120559692, 1.0, 0.994350254535675, 0.9604519605636597, 1.0, 0.9661017060279846, 1.0, 1.0, 1.0], [0.12994350492954254, 0.09604519605636597, 0.07909604161977768, 0.23163841664791107, 0.20903955399990082, 0.1807909607887268, 0.16949152946472168, 0.17514124512672424, 0.2768361568450928, 0.33898305892944336, 0.32768362760543823, 0.2881355881690979, 0.38983049988746643, 0.38983049988746643, 0.451977401971817, 0.42937853932380676, 0.451977401971817, 0.4971751272678375, NaN, 0.49152541160583496, 0.5536723136901855, 0.6271186470985413, 0.4971751272678375, 0.5367231369018555, 0.5593220591545105, 0.5423728823661804, 0.5819209218025208, 0.6158192157745361, 0.5762711763381958, 0.6158192157745361, 0.700564980506897, 0.7118644118309021, 0.6158192157745361, 0.7231638431549072, 0.700564980506897, 0.6553672552108765, 0.8926553726196289, 0.7457627058029175, 0.7853107452392578, 0.700564980506897, 0.7740113139152527, 0.7288135886192322, 0.8361582159996033, 0.8135592937469482, 0.7570621371269226, 0.8644067645072937, 0.9209039807319641, 0.8531073331832886, 0.7231638431549072, 0.8305084705352783, 0.7344632744789124, 0.8248587846755981, 0.8305084705352783, 0.8531073331832886, 0.8926553726196289, 0.9435028433799744, 0.9548022747039795, 0.9435028433799744, 0.909604549407959, 0.9887005686759949, 0.9604519605636597, 0.8192090392112732, 0.9491525292396545, 0.9604519605636597, 0.9717513918876648, 1.0, 0.9774011373519897, 0.994350254535675, 0.9887005686759949, 1.0, 1.0, 1.0], [0.0451977401971817, 0.11299435049295425, 0.11299435049295425, 0.14689265191555023, 0.10734463483095169, 0.15819208323955536, 0.11864406615495682, 0.20338982343673706, 0.1355932205915451, 0.1807909607887268, 0.19209039211273193, 0.32203391194343567, 0.24858756363391876, 0.32203391194343567, 0.3333333432674408, 0.32203391194343567, 0.38418078422546387, 0.5197740197181702, 0.508474588394165, NaN, 0.5480226278305054, 0.47457626461982727, 0.5762711763381958, 0.5367231369018555, 0.5649717450141907, 0.6553672552108765, 0.6214689016342163, 0.6779661178588867, 0.6327683329582214, 0.694915235042572, 0.5593220591545105, 0.6779661178588867, 0.6779661178588867, 0.8135592937469482, 0.5536723136901855, 0.7175140976905823, 0.7514124512672424, 0.8248587846755981, 0.8361582159996033, 0.7175140976905823, 0.8361582159996033, 0.7683615684509277, 0.8305084705352783, 0.8474576473236084, 0.7570621371269226, 0.8305084705352783, 0.909604549407959, 0.9209039807319641, 0.7401130199432373, 0.9265536665916443, 0.7401130199432373, 0.8248587846755981, 0.9322034120559692, 0.9152542352676392, 0.909604549407959, 0.9661017060279846, 0.9830508232116699, 0.9378530979156494, 0.9491525292396545, 0.9717513918876648, 0.9774011373519897, 0.8305084705352783, 0.9774011373519897, 0.9548022747039795, 0.9830508232116699, 0.994350254535675, 0.9661017060279846, 0.994350254535675, 0.9887005686759949, 0.994350254535675, 0.994350254535675, 0.994350254535675], [0.028248587623238564, 0.03954802080988884, 0.12429378181695938, 0.050847455859184265, 0.10734463483095169, 0.07909604161977768, 0.14124293625354767, 0.14124293625354767, 0.17514124512672424, 0.2259887009859085, 0.20338982343673706, 0.2881355881690979, 0.28248587250709534, 0.24858756363391876, 0.3163841664791107, 0.31073445081710815, 0.395480215549469, 0.3672316372394562, 0.44632768630981445, 0.451977401971817, NaN, 0.4576271176338196, 0.5536723136901855, 0.5536723136901855, 0.49152541160583496, 0.6610169410705566, 0.6440678238868713, 0.7231638431549072, 0.6779661178588867, 0.7118644118309021, 0.5649717450141907, 0.6553672552108765, 0.6892655491828918, 0.7288135886192322, 0.598870038986206, 0.7683615684509277, 0.6779661178588867, 0.7514124512672424, 0.7853107452392578, 0.7457627058029175, 0.7627118825912476, 0.8983050584793091, 0.7627118825912476, 0.8305084705352783, 0.8248587846755981, 0.790960431098938, 0.8757061958312988, 0.9152542352676392, 0.7401130199432373, 0.8644067645072937, 0.7514124512672424, 0.903954803943634, 0.8926553726196289, 0.9491525292396545, 0.8418079018592834, 0.994350254535675, 0.9209039807319641, 0.8926553726196289, 0.9661017060279846, 0.9265536665916443, 0.9491525292396545, 0.8248587846755981, 0.9717513918876648, 0.9209039807319641, 0.994350254535675, 0.9830508232116699, 0.9548022747039795, 1.0, 0.9604519605636597, 1.0, 1.0, 1.0], [0.05649717524647713, 0.07909604161977768, 0.1355932205915451, 0.1355932205915451, 0.10734463483095169, 0.09604519605636597, 0.11299435049295425, 0.11299435049295425, 0.23728813230991364, 0.2598870098590851, 0.299435019493103, 0.2598870098590851, 0.41807910799980164, 0.33898305892944336, 0.3333333432674408, 0.40112993121147156, 0.4406779706478119, 0.5310734510421753, 0.37288135290145874, 0.5254237055778503, 0.5423728823661804, NaN, 0.5028248429298401, 0.5367231369018555, 0.5932203531265259, 0.5875706076622009, 0.598870038986206, 0.6440678238868713, 0.6271186470985413, 0.6101694703102112, 0.8361582159996033, 0.694915235042572, 0.6271186470985413, 0.6892655491828918, 0.8813559412956238, 0.6836158037185669, 0.7966101765632629, 0.6666666865348816, 0.7118644118309021, 0.6666666865348816, 0.6892655491828918, 0.7853107452392578, 0.8305084705352783, 0.7457627058029175, 0.7401130199432373, 0.8813559412956238, 0.8135592937469482, 0.8418079018592834, 0.7231638431549072, 0.7514124512672424, 0.7344632744789124, 0.8079096078872681, 0.7627118825912476, 0.8418079018592834, 0.7683615684509277, 0.903954803943634, 0.9265536665916443, 0.8361582159996033, 0.8644067645072937, 0.9604519605636597, 0.9774011373519897, 0.8305084705352783, 0.9209039807319641, 0.887005627155304, 0.9717513918876648, 0.9548022747039795, 0.9209039807319641, 0.9887005686759949, 0.9322034120559692, 1.0, 1.0, 1.0], [0.033898305147886276, 0.0451977401971817, 0.0903954803943634, 0.050847455859184265, 0.10734463483095169, 0.11299435049295425, 0.14124293625354767, 0.14124293625354767, 0.16384181380271912, 0.21468926966190338, 0.1977401077747345, 0.299435019493103, 0.299435019493103, 0.3163841664791107, 0.3163841664791107, 0.3446327745914459, 0.2542372941970825, 0.3785310685634613, 0.5028248429298401, 0.4237288236618042, 0.44632768630981445, 0.4971751272678375, NaN, 0.4124293923377991, 0.5028248429298401, 0.598870038986206, 0.5649717450141907, 0.6666666865348816, 0.6836158037185669, 0.790960431098938, 0.5819209218025208, 0.5819209218025208, 0.7118644118309021, 0.6384180784225464, 0.6666666865348816, 0.790960431098938, 0.700564980506897, 0.7683615684509277, 0.7570621371269226, 0.7627118825912476, 0.7683615684509277, 0.8700565099716187, 0.6892655491828918, 0.7401130199432373, 0.7966101765632629, 0.7288135886192322, 0.8474576473236084, 0.9774011373519897, 0.7175140976905823, 0.8644067645072937, 0.7231638431549072, 0.9265536665916443, 0.8700565099716187, 0.9661017060279846, 0.8587570786476135, 0.9717513918876648, 0.9661017060279846, 0.8587570786476135, 0.9661017060279846, 0.8926553726196289, 0.9887005686759949, 0.8079096078872681, 0.9661017060279846, 0.903954803943634, 1.0, 0.9887005686759949, 0.9322034120559692, 1.0, 0.9604519605636597, 1.0, 1.0, 1.0], [0.08474576473236084, 0.11864406615495682, 0.15819208323955536, 0.14124293625354767, 0.11299435049295425, 0.19209039211273193, 0.23163841664791107, 0.23163841664791107, 0.2542372941970825, 0.2598870098590851, 0.2542372941970825, 0.33898305892944336, 0.29378530383110046, 0.3615819215774536, 0.33898305892944336, 0.33898305892944336, 0.3785310685634613, 0.44632768630981445, 0.46327683329582214, 0.46327683329582214, 0.44632768630981445, 0.46327683329582214, 0.5875706076622009, NaN, 0.4971751272678375, 0.5932203531265259, 0.6384180784225464, 0.598870038986206, 0.6271186470985413, 0.6327683329582214, 0.5536723136901855, 0.5536723136901855, 0.6892655491828918, 0.6384180784225464, 0.5367231369018555, 0.6892655491828918, 0.6497175097465515, 0.7740113139152527, 0.7401130199432373, 0.7175140976905823, 0.7740113139152527, 0.7457627058029175, 0.700564980506897, 0.8418079018592834, 0.7627118825912476, 0.6892655491828918, 0.8361582159996033, 0.8531073331832886, 0.7288135886192322, 0.8757061958312988, 0.7401130199432373, 0.8361582159996033, 0.8757061958312988, 0.8531073331832886, 0.8757061958312988, 0.8926553726196289, 0.8983050584793091, 0.8700565099716187, 0.8644067645072937, 0.8813559412956238, 0.8983050584793091, 0.8474576473236084, 0.9435028433799744, 0.9491525292396545, 0.9830508232116699, 0.9435028433799744, 0.9830508232116699, 0.9887005686759949, 0.9830508232116699, 0.994350254535675, 0.994350254535675, 1.0], [0.06214689090847969, 0.08474576473236084, 0.10734463483095169, 0.09604519605636597, 0.12429378181695938, 0.12429378181695938, 0.1525423675775528, 0.18644067645072937, 0.1355932205915451, 0.12429378181695938, 0.16949152946472168, 0.2429378479719162, 0.23163841664791107, 0.2542372941970825, 0.3050847351551056, 0.2598870098590851, 0.3446327745914459, 0.4689265489578247, 0.4406779706478119, 0.4350282549858093, 0.508474588394165, 0.4067796468734741, 0.4971751272678375, 0.5028248429298401, NaN, 0.5819209218025208, 0.6158192157745361, 0.6553672552108765, 0.5932203531265259, 0.6384180784225464, 0.4971751272678375, 0.6666666865348816, 0.6497175097465515, 0.7231638431549072, 0.5536723136901855, 0.6892655491828918, 0.5932203531265259, 0.7062146663665771, 0.7683615684509277, 0.7570621371269226, 0.7344632744789124, 0.7288135886192322, 0.7175140976905823, 0.8700565099716187, 0.7683615684509277, 0.7457627058029175, 0.8079096078872681, 0.8531073331832886, 0.7062146663665771, 0.7966101765632629, 0.7118644118309021, 0.790960431098938, 0.8079096078872681, 0.8022598624229431, 0.903954803943634, 0.9322034120559692, 0.8474576473236084, 0.9661017060279846, 0.8135592937469482, 0.9209039807319641, 0.8813559412956238, 0.7740113139152527, 0.9717513918876648, 0.9604519605636597, 0.9435028433799744, 0.9661017060279846, 0.9774011373519897, 0.9774011373519897, 0.9774011373519897, 0.994350254535675, 0.994350254535675, 0.994350254535675], [0.0451977401971817, 0.06214689090847969, 0.0903954803943634, 0.07909604161977768, 0.09604519605636597, 0.07909604161977768, 0.09604519605636597, 0.12994350492954254, 0.12994350492954254, 0.14689265191555023, 0.1525423675775528, 0.22033898532390594, 0.24858756363391876, 0.2429378479719162, 0.24858756363391876, 0.2598870098590851, 0.21468926966190338, 0.2542372941970825, 0.4576271176338196, 0.3446327745914459, 0.33898305892944336, 0.4124293923377991, 0.40112993121147156, 0.4067796468734741, 0.41807910799980164, NaN, 0.4971751272678375, 0.6666666865348816, 0.5932203531265259, 0.6214689016342163, 0.5706214904785156, 0.5875706076622009, 0.7118644118309021, 0.6101694703102112, 0.5875706076622009, 0.7401130199432373, 0.6836158037185669, 0.7288135886192322, 0.7344632744789124, 0.7457627058029175, 0.7514124512672424, 0.8757061958312988, 0.6779661178588867, 0.7514124512672424, 0.7796609997749329, 0.7344632744789124, 0.8305084705352783, 0.9717513918876648, 0.7118644118309021, 0.8418079018592834, 0.7288135886192322, 0.903954803943634, 0.8644067645072937, 0.9378530979156494, 0.8248587846755981, 0.9887005686759949, 0.9322034120559692, 0.8587570786476135, 0.9604519605636597, 0.8813559412956238, 0.9774011373519897, 0.7966101765632629, 0.9774011373519897, 0.909604549407959, 0.9830508232116699, 0.9887005686759949, 0.9548022747039795, 1.0, 0.9548022747039795, 1.0, 1.0, 1.0], [0.050847455859184265, 0.09604519605636597, 0.12994350492954254, 0.11864406615495682, 0.1355932205915451, 0.14124293625354767, 0.1355932205915451, 0.21468926966190338, 0.15819208323955536, 0.19209039211273193, 0.17514124512672424, 0.29378530383110046, 0.23728813230991364, 0.2598870098590851, 0.2881355881690979, 0.2711864411830902, 0.26553672552108765, 0.31073445081710815, 0.41807910799980164, 0.3785310685634613, 0.35593220591545105, 0.40112993121147156, 0.4350282549858093, 0.3615819215774536, 0.38418078422546387, 0.5028248429298401, NaN, 0.5141242742538452, 0.5480226278305054, 0.5197740197181702, 0.48022598028182983, 0.49152541160583496, 0.5706214904785156, 0.5310734510421753, 0.5141242742538452, 0.6271186470985413, 0.6553672552108765, 0.6892655491828918, 0.8418079018592834, 0.6384180784225464, 0.7118644118309021, 0.6610169410705566, 0.6836158037185669, 0.6779661178588867, 0.6836158037185669, 0.6836158037185669, 0.8474576473236084, 0.7570621371269226, 0.7062146663665771, 0.8192090392112732, 0.7231638431549072, 0.7401130199432373, 0.8531073331832886, 0.8361582159996033, 0.8700565099716187, 0.887005627155304, 0.909604549407959, 0.903954803943634, 0.8700565099716187, 0.887005627155304, 0.9209039807319641, 0.7740113139152527, 0.9548022747039795, 0.9435028433799744, 0.9435028433799744, 0.9774011373519897, 0.9717513918876648, 1.0, 0.9774011373519897, 0.994350254535675, 0.994350254535675, 1.0], [0.033898305147886276, 0.050847455859184265, 0.09604519605636597, 0.06214689090847969, 0.10734463483095169, 0.07344632595777512, 0.07909604161977768, 0.11864406615495682, 0.09604519605636597, 0.12994350492954254, 0.12994350492954254, 0.1977401077747345, 0.21468926966190338, 0.20903955399990082, 0.23728813230991364, 0.23728813230991364, 0.18644067645072937, 0.24858756363391876, 0.38418078422546387, 0.32203391194343567, 0.2768361568450928, 0.35593220591545105, 0.3333333432674408, 0.40112993121147156, 0.3446327745914459, 0.3333333432674408, 0.4858756959438324, NaN, 0.5423728823661804, 0.5649717450141907, 0.5197740197181702, 0.5367231369018555, 0.6214689016342163, 0.5480226278305054, 0.5367231369018555, 0.6610169410705566, 0.6497175097465515, 0.694915235042572, 0.6892655491828918, 0.694915235042572, 0.7118644118309021, 0.8248587846755981, 0.6723163723945618, 0.7401130199432373, 0.7740113139152527, 0.7231638431549072, 0.7966101765632629, 0.9209039807319641, 0.7118644118309021, 0.7853107452392578, 0.7118644118309021, 0.8700565099716187, 0.7966101765632629, 0.9265536665916443, 0.8079096078872681, 0.9604519605636597, 0.9209039807319641, 0.8644067645072937, 0.9209039807319641, 0.8926553726196289, 0.9604519605636597, 0.7966101765632629, 0.9548022747039795, 0.8926553726196289, 0.9830508232116699, 0.9774011373519897, 0.9435028433799744, 1.0, 0.9548022747039795, 1.0, 1.0, 1.0], [0.03954802080988884, 0.03954802080988884, 0.10734463483095169, 0.050847455859184265, 0.09604519605636597, 0.06214689090847969, 0.12429378181695938, 0.09604519605636597, 0.11864406615495682, 0.16384181380271912, 0.15819208323955536, 0.2429378479719162, 0.23728813230991364, 0.22033898532390594, 0.2768361568450928, 0.2711864411830902, 0.2768361568450928, 0.20903955399990082, 0.4237288236618042, 0.3672316372394562, 0.32203391194343567, 0.37288135290145874, 0.3163841664791107, 0.37288135290145874, 0.4067796468734741, 0.4067796468734741, 0.451977401971817, 0.4576271176338196, NaN, 0.47457626461982727, 0.49152541160583496, 0.5593220591545105, 0.6384180784225464, 0.5706214904785156, 0.5536723136901855, 0.6779661178588867, 0.6553672552108765, 0.7062146663665771, 0.6327683329582214, 0.7514124512672424, 0.7175140976905823, 0.887005627155304, 0.6384180784225464, 0.6327683329582214, 0.8022598624229431, 0.6779661178588867, 0.694915235042572, 0.8079096078872681, 0.7231638431549072, 0.8079096078872681, 0.7288135886192322, 0.8926553726196289, 0.8248587846755981, 0.9265536665916443, 0.7966101765632629, 0.8248587846755981, 0.8983050584793091, 0.7740113139152527, 0.9435028433799744, 0.8022598624229431, 0.9152542352676392, 0.8248587846755981, 0.9265536665916443, 0.8644067645072937, 1.0, 0.9830508232116699, 0.9491525292396545, 1.0, 0.9604519605636597, 1.0, 1.0, 1.0], [0.028248587623238564, 0.033898305147886276, 0.06214689090847969, 0.050847455859184265, 0.08474576473236084, 0.07909604161977768, 0.08474576473236084, 0.11299435049295425, 0.1355932205915451, 0.12429378181695938, 0.12994350492954254, 0.23163841664791107, 0.22033898532390594, 0.19209039211273193, 0.19209039211273193, 0.21468926966190338, 0.15819208323955536, 0.24858756363391876, 0.38418078422546387, 0.3050847351551056, 0.2881355881690979, 0.38983049988746643, 0.20903955399990082, 0.3672316372394562, 0.3615819215774536, 0.3785310685634613, 0.48022598028182983, 0.4350282549858093, 0.5254237055778503, NaN, 0.5254237055778503, 0.4689265489578247, 0.5480226278305054, 0.4971751272678375, 0.5762711763381958, 0.604519784450531, 0.6327683329582214, 0.6892655491828918, 0.6666666865348816, 0.6836158037185669, 0.694915235042572, 0.7401130199432373, 0.6384180784225464, 0.6836158037185669, 0.7627118825912476, 0.6666666865348816, 0.7627118825912476, 0.8926553726196289, 0.7062146663665771, 0.7853107452392578, 0.7175140976905823, 0.8192090392112732, 0.7966101765632629, 0.8587570786476135, 0.7853107452392578, 0.9322034120559692, 0.8983050584793091, 0.8531073331832886, 0.887005627155304, 0.8531073331832886, 0.9435028433799744, 0.7853107452392578, 0.9378530979156494, 0.887005627155304, 0.9830508232116699, 0.9717513918876648, 0.9265536665916443, 1.0, 0.9322034120559692, 0.9887005686759949, 0.9887005686759949, 0.994350254535675], [0.03954802080988884, 0.05649717524647713, 0.09604519605636597, 0.09604519605636597, 0.10169491171836853, 0.0903954803943634, 0.09604519605636597, 0.07909604161977768, 0.16949152946472168, 0.1525423675775528, 0.21468926966190338, 0.17514124512672424, 0.3446327745914459, 0.28248587250709534, 0.23163841664791107, 0.31073445081710815, 0.3333333432674408, 0.4237288236618042, 0.299435019493103, 0.4406779706478119, 0.4350282549858093, 0.16384181380271912, 0.41807910799980164, 0.44632768630981445, 0.5028248429298401, 0.42937853932380676, 0.5197740197181702, 0.48022598028182983, 0.508474588394165, 0.47457626461982727, NaN, 0.5819209218025208, 0.5254237055778503, 0.598870038986206, 0.6497175097465515, 0.5197740197181702, 0.6892655491828918, 0.6158192157745361, 0.604519784450531, 0.5819209218025208, 0.6271186470985413, 0.6384180784225464, 0.7288135886192322, 0.6779661178588867, 0.6158192157745361, 0.7853107452392578, 0.7231638431549072, 0.7288135886192322, 0.700564980506897, 0.6610169410705566, 0.7118644118309021, 0.7288135886192322, 0.6892655491828918, 0.7514124512672424, 0.694915235042572, 0.8192090392112732, 0.8361582159996033, 0.7796609997749329, 0.7683615684509277, 0.9265536665916443, 0.9152542352676392, 0.7966101765632629, 0.8587570786476135, 0.8531073331832886, 0.9152542352676392, 0.8926553726196289, 0.8926553726196289, 0.9774011373519897, 0.9209039807319641, 0.9661017060279846, 0.9661017060279846, 0.994350254535675], [0.03954802080988884, 0.05649717524647713, 0.0903954803943634, 0.050847455859184265, 0.10734463483095169, 0.05649717524647713, 0.11299435049295425, 0.14124293625354767, 0.09604519605636597, 0.08474576473236084, 0.11299435049295425, 0.17514124512672424, 0.1977401077747345, 0.16949152946472168, 0.22033898532390594, 0.20338982343673706, 0.2542372941970825, 0.3050847351551056, 0.2881355881690979, 0.32203391194343567, 0.3446327745914459, 0.3050847351551056, 0.41807910799980164, 0.44632768630981445, 0.3333333432674408, 0.4124293923377991, 0.508474588394165, 0.46327683329582214, 0.4406779706478119, 0.5310734510421753, 0.41807910799980164, NaN, 0.48022598028182983, 0.5932203531265259, 0.46327683329582214, 0.5932203531265259, 0.5706214904785156, 0.5875706076622009, 0.6892655491828918, 0.694915235042572, 0.6384180784225464, 0.6497175097465515, 0.6836158037185669, 0.8079096078872681, 0.7118644118309021, 0.6723163723945618, 0.7683615684509277, 0.7683615684509277, 0.6836158037185669, 0.7401130199432373, 0.6892655491828918, 0.7401130199432373, 0.7344632744789124, 0.7457627058029175, 0.790960431098938, 0.8926553726196289, 0.8079096078872681, 0.8926553726196289, 0.7740113139152527, 0.903954803943634, 0.8531073331832886, 0.7514124512672424, 0.9491525292396545, 0.9378530979156494, 0.9209039807319641, 0.9378530979156494, 0.9548022747039795, 0.9887005686759949, 0.9661017060279846, 0.9830508232116699, 0.9830508232116699, 1.0], [0.033898305147886276, 0.02259887009859085, 0.10169491171836853, 0.03954802080988884, 0.11299435049295425, 0.05649717524647713, 0.09604519605636597, 0.0903954803943634, 0.14124293625354767, 0.18644067645072937, 0.16949152946472168, 0.21468926966190338, 0.23728813230991364, 0.1977401077747345, 0.2429378479719162, 0.24858756363391876, 0.20903955399990082, 0.2542372941970825, 0.38418078422546387, 0.32203391194343567, 0.31073445081710815, 0.37288135290145874, 0.2881355881690979, 0.31073445081710815, 0.3502824902534485, 0.2881355881690979, 0.42937853932380676, 0.3785310685634613, 0.3615819215774536, 0.451977401971817, 0.47457626461982727, 0.5197740197181702, NaN, 0.508474588394165, 0.5819209218025208, 0.5932203531265259, 0.6327683329582214, 0.5932203531265259, 0.5593220591545105, 0.7344632744789124, 0.6158192157745361, 0.8305084705352783, 0.6553672552108765, 0.5875706076622009, 0.7966101765632629, 0.6327683329582214, 0.6327683329582214, 0.7231638431549072, 0.7118644118309021, 0.7457627058029175, 0.7344632744789124, 0.903954803943634, 0.7683615684509277, 0.9152542352676392, 0.7288135886192322, 0.7627118825912476, 0.8248587846755981, 0.7062146663665771, 0.909604549407959, 0.790960431098938, 0.8305084705352783, 0.8192090392112732, 0.9265536665916443, 0.8531073331832886, 0.9887005686759949, 0.9717513918876648, 0.9322034120559692, 1.0, 0.9491525292396545, 0.994350254535675, 0.994350254535675, 0.994350254535675], [0.028248587623238564, 0.02259887009859085, 0.033898305147886276, 0.02259887009859085, 0.07344632595777512, 0.033898305147886276, 0.07344632595777512, 0.09604519605636597, 0.06214689090847969, 0.07344632595777512, 0.05649717524647713, 0.14689265191555023, 0.12429378181695938, 0.15819208323955536, 0.16384181380271912, 0.16949152946472168, 0.15819208323955536, 0.23163841664791107, 0.2768361568450928, 0.18644067645072937, 0.2711864411830902, 0.31073445081710815, 0.3615819215774536, 0.3615819215774536, 0.2768361568450928, 0.38983049988746643, 0.4689265489578247, 0.451977401971817, 0.42937853932380676, 0.5028248429298401, 0.40112993121147156, 0.4067796468734741, 0.49152541160583496, NaN, 0.42937853932380676, 0.5536723136901855, 0.5254237055778503, 0.6271186470985413, 0.6384180784225464, 0.6271186470985413, 0.6553672552108765, 0.6101694703102112, 0.598870038986206, 0.7288135886192322, 0.6836158037185669, 0.6610169410705566, 0.7457627058029175, 0.7514124512672424, 0.6666666865348816, 0.7231638431549072, 0.6610169410705566, 0.7062146663665771, 0.7231638431549072, 0.7401130199432373, 0.8079096078872681, 0.8813559412956238, 0.8361582159996033, 0.8757061958312988, 0.7796609997749329, 0.8531073331832886, 0.8587570786476135, 0.7344632744789124, 0.9209039807319641, 0.8813559412956238, 0.9152542352676392, 0.9548022747039795, 0.9322034120559692, 0.9830508232116699, 0.9265536665916443, 1.0, 1.0, 1.0], [0.028248587623238564, 0.033898305147886276, 0.07909604161977768, 0.050847455859184265, 0.09604519605636597, 0.050847455859184265, 0.07344632595777512, 0.050847455859184265, 0.10734463483095169, 0.10734463483095169, 0.1355932205915451, 0.12994350492954254, 0.1977401077747345, 0.1525423675775528, 0.15819208323955536, 0.19209039211273193, 0.2598870098590851, 0.38418078422546387, 0.299435019493103, 0.44632768630981445, 0.40112993121147156, 0.11864406615495682, 0.3333333432674408, 0.46327683329582214, 0.44632768630981445, 0.4124293923377991, 0.4858756959438324, 0.46327683329582214, 0.44632768630981445, 0.4237288236618042, 0.3502824902534485, 0.5367231369018555, 0.41807910799980164, 0.5706214904785156, NaN, 0.451977401971817, 0.5536723136901855, 0.6271186470985413, 0.5762711763381958, 0.5367231369018555, 0.6384180784225464, 0.5480226278305054, 0.700564980506897, 0.6553672552108765, 0.604519784450531, 0.7740113139152527, 0.7118644118309021, 0.6836158037185669, 0.694915235042572, 0.6214689016342163, 0.7062146663665771, 0.7118644118309021, 0.6271186470985413, 0.7062146663665771, 0.6440678238868713, 0.7966101765632629, 0.8022598624229431, 0.7627118825912476, 0.7231638431549072, 0.8926553726196289, 0.8983050584793091, 0.790960431098938, 0.8135592937469482, 0.7966101765632629, 0.909604549407959, 0.8813559412956238, 0.8474576473236084, 0.9830508232116699, 0.8587570786476135, 0.9774011373519897, 0.9774011373519897, 0.994350254535675], [0.03954802080988884, 0.016949152573943138, 0.09604519605636597, 0.028248587623238564, 0.10734463483095169, 0.0451977401971817, 0.07344632595777512, 0.07344632595777512, 0.11864406615495682, 0.18644067645072937, 0.14689265191555023, 0.1807909607887268, 0.2259887009859085, 0.1807909607887268, 0.1977401077747345, 0.23728813230991364, 0.20338982343673706, 0.18644067645072937, 0.3446327745914459, 0.28248587250709534, 0.23163841664791107, 0.3163841664791107, 0.20903955399990082, 0.31073445081710815, 0.31073445081710815, 0.2598870098590851, 0.37288135290145874, 0.33898305892944336, 0.32203391194343567, 0.395480215549469, 0.48022598028182983, 0.4067796468734741, 0.4067796468734741, 0.44632768630981445, 0.5480226278305054, NaN, 0.6101694703102112, 0.5762711763381958, 0.5423728823661804, 0.6666666865348816, 0.5932203531265259, 0.7514124512672424, 0.6101694703102112, 0.5480226278305054, 0.7514124512672424, 0.598870038986206, 0.6101694703102112, 0.6553672552108765, 0.7175140976905823, 0.694915235042572, 0.7231638431549072, 0.8248587846755981, 0.7344632744789124, 0.8700565099716187, 0.7175140976905823, 0.7231638431549072, 0.790960431098938, 0.694915235042572, 0.8757061958312988, 0.7740113139152527, 0.8135592937469482, 0.790960431098938, 0.8587570786476135, 0.8418079018592834, 0.9830508232116699, 0.9265536665916443, 0.903954803943634, 1.0, 0.9265536665916443, 0.9887005686759949, 0.9830508232116699, 0.9887005686759949], [0.011299435049295425, 0.033898305147886276, 0.03954802080988884, 0.07909604161977768, 0.06779661029577255, 0.033898305147886276, 0.028248587623238564, 0.06214689090847969, 0.06214689090847969, 0.016949152573943138, 0.07909604161977768, 0.10734463483095169, 0.14689265191555023, 0.20903955399990082, 0.12429378181695938, 0.1525423675775528, 0.2711864411830902, 0.3050847351551056, 0.10734463483095169, 0.24858756363391876, 0.32203391194343567, 0.20338982343673706, 0.299435019493103, 0.3502824902534485, 0.4067796468734741, 0.3163841664791107, 0.3446327745914459, 0.3502824902534485, 0.3446327745914459, 0.3672316372394562, 0.31073445081710815, 0.42937853932380676, 0.3672316372394562, 0.47457626461982727, 0.44632768630981445, 0.38983049988746643, NaN, 0.6158192157745361, 0.5028248429298401, 0.5028248429298401, 0.6271186470985413, 0.4858756959438324, 0.598870038986206, 0.6158192157745361, 0.5423728823661804, 0.6779661178588867, 0.6666666865348816, 0.6101694703102112, 0.6440678238868713, 0.700564980506897, 0.6553672552108765, 0.6101694703102112, 0.7175140976905823, 0.604519784450531, 0.7231638431549072, 0.7457627058029175, 0.790960431098938, 0.790960431098938, 0.6836158037185669, 0.8926553726196289, 0.8474576473236084, 0.7796609997749329, 0.8361582159996033, 0.8926553726196289, 0.8361582159996033, 0.9265536665916443, 0.9491525292396545, 0.9548022747039795, 0.9661017060279846, 0.9661017060279846, 0.9661017060279846, 1.0], [0.033898305147886276, 0.0451977401971817, 0.06779661029577255, 0.06214689090847969, 0.10169491171836853, 0.07909604161977768, 0.07909604161977768, 0.14689265191555023, 0.06214689090847969, 0.07344632595777512, 0.07344632595777512, 0.17514124512672424, 0.06779661029577255, 0.16949152946472168, 0.18644067645072937, 0.11299435049295425, 0.16384181380271912, 0.23163841664791107, 0.2542372941970825, 0.17514124512672424, 0.24858756363391876, 0.3333333432674408, 0.23163841664791107, 0.2259887009859085, 0.29378530383110046, 0.2711864411830902, 0.31073445081710815, 0.3050847351551056, 0.29378530383110046, 0.31073445081710815, 0.38418078422546387, 0.4124293923377991, 0.4067796468734741, 0.37288135290145874, 0.37288135290145874, 0.4237288236618042, 0.38418078422546387, NaN, 0.4971751272678375, 0.5536723136901855, 0.6525423526763916, 0.5141242742538452, 0.4237288236618042, 0.5423728823661804, 0.5932203531265259, 0.451977401971817, 0.5141242742538452, 0.46327683329582214, 0.6101694703102112, 0.8079096078872681, 0.6158192157745361, 0.6553672552108765, 0.8644067645072937, 0.6497175097465515, 0.9435028433799744, 0.5819209218025208, 0.7288135886192322, 0.7570621371269226, 0.7118644118309021, 0.6384180784225464, 0.7175140976905823, 0.7570621371269226, 0.9548022747039795, 0.9661017060279846, 0.8192090392112732, 0.9209039807319641, 0.9887005686759949, 0.994350254535675, 0.9887005686759949, 1.0, 1.0, 1.0], [0.011299435049295425, 0.050847455859184265, 0.07909604161977768, 0.05649717524647713, 0.07909604161977768, 0.06779661029577255, 0.0451977401971817, 0.0903954803943634, 0.05649717524647713, 0.07344632595777512, 0.06779661029577255, 0.18644067645072937, 0.11864406615495682, 0.1525423675775528, 0.14124293625354767, 0.1525423675775528, 0.1525423675775528, 0.18644067645072937, 0.21468926966190338, 0.16384181380271912, 0.21468926966190338, 0.2881355881690979, 0.2429378479719162, 0.2598870098590851, 0.23163841664791107, 0.26553672552108765, 0.15819208323955536, 0.31073445081710815, 0.3672316372394562, 0.3333333432674408, 0.395480215549469, 0.31073445081710815, 0.4406779706478119, 0.3615819215774536, 0.4237288236618042, 0.4576271176338196, 0.4971751272678375, 0.5028248429298401, NaN, 0.5480226278305054, 0.5197740197181702, 0.5028248429298401, 0.5197740197181702, 0.5141242742538452, 0.5762711763381958, 0.5706214904785156, 0.7062146663665771, 0.5762711763381958, 0.6723163723945618, 0.6892655491828918, 0.6779661178588867, 0.6327683329582214, 0.6892655491828918, 0.6497175097465515, 0.7231638431549072, 0.7118644118309021, 0.8135592937469482, 0.8079096078872681, 0.7175140976905823, 0.7627118825912476, 0.8587570786476135, 0.7231638431549072, 0.887005627155304, 0.903954803943634, 0.8531073331832886, 0.9322034120559692, 0.9435028433799744, 0.9548022747039795, 0.9548022747039795, 0.9774011373519897, 0.9774011373519897, 1.0], [0.033898305147886276, 0.016949152573943138, 0.07344632595777512, 0.02259887009859085, 0.06214689090847969, 0.033898305147886276, 0.0451977401971817, 0.05649717524647713, 0.11299435049295425, 0.14689265191555023, 0.1355932205915451, 0.14689265191555023, 0.18644067645072937, 0.14124293625354767, 0.18644067645072937, 0.20903955399990082, 0.18644067645072937, 0.20338982343673706, 0.299435019493103, 0.28248587250709534, 0.2542372941970825, 0.3333333432674408, 0.23728813230991364, 0.28248587250709534, 0.2429378479719162, 0.2542372941970825, 0.3615819215774536, 0.3050847351551056, 0.24858756363391876, 0.3163841664791107, 0.41807910799980164, 0.3050847351551056, 0.26553672552108765, 0.37288135290145874, 0.46327683329582214, 0.3333333432674408, 0.4971751272678375, 0.44632768630981445, 0.451977401971817, NaN, 0.47457626461982727, 0.41807910799980164, 0.5649717450141907, 0.5028248429298401, 0.7570621371269226, 0.604519784450531, 0.5480226278305054, 0.5028248429298401, 0.6497175097465515, 0.5480226278305054, 0.6666666865348816, 0.8248587846755981, 0.5932203531265259, 0.7457627058029175, 0.5706214904785156, 0.6271186470985413, 0.6497175097465515, 0.604519784450531, 0.7683615684509277, 0.7175140976905823, 0.7288135886192322, 0.8079096078872681, 0.7401130199432373, 0.8248587846755981, 0.9491525292396545, 0.903954803943634, 0.8418079018592834, 0.909604549407959, 0.887005627155304, 0.9830508232116699, 0.9830508232116699, 0.994350254535675], [0.028248587623238564, 0.06214689090847969, 0.05649717524647713, 0.06214689090847969, 0.10734463483095169, 0.06779661029577255, 0.07344632595777512, 0.14124293625354767, 0.07344632595777512, 0.06214689090847969, 0.08474576473236084, 0.16949152946472168, 0.06214689090847969, 0.15819208323955536, 0.1525423675775528, 0.10169491171836853, 0.1525423675775528, 0.19209039211273193, 0.2259887009859085, 0.16384181380271912, 0.23728813230991364, 0.31073445081710815, 0.23163841664791107, 0.2259887009859085, 0.26553672552108765, 0.24858756363391876, 0.2881355881690979, 0.2881355881690979, 0.28248587250709534, 0.3050847351551056, 0.37288135290145874, 0.3615819215774536, 0.38418078422546387, 0.3446327745914459, 0.3615819215774536, 0.4067796468734741, 0.37288135290145874, 0.347457617521286, 0.48022598028182983, 0.5254237055778503, NaN, 0.49152541160583496, 0.40112993121147156, 0.5367231369018555, 0.5649717450141907, 0.4406779706478119, 0.508474588394165, 0.4406779706478119, 0.604519784450531, 0.7514124512672424, 0.6158192157745361, 0.6327683329582214, 0.8079096078872681, 0.6384180784225464, 0.887005627155304, 0.5310734510421753, 0.6836158037185669, 0.7514124512672424, 0.694915235042572, 0.6158192157745361, 0.7062146663665771, 0.7514124512672424, 0.9378530979156494, 0.9604519605636597, 0.8305084705352783, 0.903954803943634, 0.994350254535675, 0.994350254535675, 0.9887005686759949, 1.0, 1.0, 1.0], [0.005649717524647713, 0.02259887009859085, 0.050847455859184265, 0.016949152573943138, 0.06214689090847969, 0.016949152573943138, 0.06779661029577255, 0.033898305147886276, 0.08474576473236084, 0.10169491171836853, 0.08474576473236084, 0.11299435049295425, 0.17514124512672424, 0.09604519605636597, 0.14689265191555023, 0.15819208323955536, 0.11864406615495682, 0.07909604161977768, 0.2711864411830902, 0.23163841664791107, 0.10169491171836853, 0.21468926966190338, 0.12994350492954254, 0.2542372941970825, 0.2711864411830902, 0.12429378181695938, 0.33898305892944336, 0.17514124512672424, 0.11299435049295425, 0.2598870098590851, 0.3615819215774536, 0.3502824902534485, 0.16949152946472168, 0.38983049988746643, 0.451977401971817, 0.24858756363391876, 0.5141242742538452, 0.4858756959438324, 0.4971751272678375, 0.5819209218025208, 0.508474588394165, NaN, 0.5875706076622009, 0.5310734510421753, 0.6553672552108765, 0.6101694703102112, 0.5480226278305054, 0.5536723136901855, 0.6892655491828918, 0.694915235042572, 0.694915235042572, 0.7231638431549072, 0.694915235042572, 0.7796609997749329, 0.7062146663665771, 0.7062146663665771, 0.694915235042572, 0.6497175097465515, 0.8248587846755981, 0.7401130199432373, 0.7344632744789124, 0.7796609997749329, 0.8474576473236084, 0.8361582159996033, 0.9717513918876648, 0.9152542352676392, 0.887005627155304, 0.9887005686759949, 0.903954803943634, 1.0, 1.0, 1.0], [0.011299435049295425, 0.033898305147886276, 0.033898305147886276, 0.0451977401971817, 0.05649717524647713, 0.050847455859184265, 0.016949152573943138, 0.028248587623238564, 0.07909604161977768, 0.11864406615495682, 0.07909604161977768, 0.10734463483095169, 0.12429378181695938, 0.10734463483095169, 0.1355932205915451, 0.11299435049295425, 0.08474576473236084, 0.21468926966190338, 0.16384181380271912, 0.16949152946472168, 0.23728813230991364, 0.16949152946472168, 0.31073445081710815, 0.299435019493103, 0.28248587250709534, 0.32203391194343567, 0.3163841664791107, 0.32768362760543823, 0.3615819215774536, 0.3615819215774536, 0.2711864411830902, 0.3163841664791107, 0.3446327745914459, 0.40112993121147156, 0.299435019493103, 0.38983049988746643, 0.40112993121147156, 0.5762711763381958, 0.48022598028182983, 0.4350282549858093, 0.598870038986206, 0.4124293923377991, NaN, 0.508474588394165, 0.508474588394165, 0.6553672552108765, 0.6214689016342163, 0.5649717450141907, 0.6666666865348816, 0.6327683329582214, 0.6610169410705566, 0.5762711763381958, 0.6440678238868713, 0.5593220591545105, 0.6440678238868713, 0.6723163723945618, 0.7740113139152527, 0.7401130199432373, 0.6440678238868713, 0.9265536665916443, 0.8700565099716187, 0.790960431098938, 0.7683615684509277, 0.8248587846755981, 0.8531073331832886, 0.8700565099716187, 0.8700565099716187, 0.8813559412956238, 0.8813559412956238, 0.909604549407959, 0.909604549407959, 1.0], [0.016949152573943138, 0.02259887009859085, 0.0903954803943634, 0.06779661029577255, 0.08474576473236084, 0.07344632595777512, 0.12429378181695938, 0.10734463483095169, 0.10734463483095169, 0.10169491171836853, 0.09604519605636597, 0.16384181380271912, 0.12994350492954254, 0.15819208323955536, 0.20903955399990082, 0.1525423675775528, 0.1807909607887268, 0.16384181380271912, 0.18644067645072937, 0.1525423675775528, 0.16949152946472168, 0.2542372941970825, 0.2598870098590851, 0.15819208323955536, 0.12994350492954254, 0.24858756363391876, 0.32203391194343567, 0.2598870098590851, 0.3672316372394562, 0.3163841664791107, 0.32203391194343567, 0.19209039211273193, 0.4124293923377991, 0.2711864411830902, 0.3446327745914459, 0.451977401971817, 0.38418078422546387, 0.4576271176338196, 0.4858756959438324, 0.4971751272678375, 0.46327683329582214, 0.4689265489578247, 0.49152541160583496, NaN, 0.5423728823661804, 0.5028248429298401, 0.5367231369018555, 0.5197740197181702, 0.6610169410705566, 0.5254237055778503, 0.6610169410705566, 0.5536723136901855, 0.5536723136901855, 0.5932203531265259, 0.5254237055778503, 0.598870038986206, 0.6497175097465515, 0.6610169410705566, 0.598870038986206, 0.6214689016342163, 0.6836158037185669, 0.7062146663665771, 0.6610169410705566, 0.7683615684509277, 0.7457627058029175, 0.6836158037185669, 0.8361582159996033, 0.7627118825912476, 0.9378530979156494, 0.9152542352676392, 0.9152542352676392, 0.9830508232116699], [0.02259887009859085, 0.016949152573943138, 0.05649717524647713, 0.033898305147886276, 0.06779661029577255, 0.02259887009859085, 0.0451977401971817, 0.03954802080988884, 0.12429378181695938, 0.12429378181695938, 0.12429378181695938, 0.11864406615495682, 0.16949152946472168, 0.11864406615495682, 0.16384181380271912, 0.17514124512672424, 0.1525423675775528, 0.15819208323955536, 0.2429378479719162, 0.2429378479719162, 0.17514124512672424, 0.2598870098590851, 0.20338982343673706, 0.23728813230991364, 0.23163841664791107, 0.22033898532390594, 0.3163841664791107, 0.2259887009859085, 0.1977401077747345, 0.23728813230991364, 0.38418078422546387, 0.2881355881690979, 0.20338982343673706, 0.3163841664791107, 0.395480215549469, 0.24858756363391876, 0.4576271176338196, 0.4067796468734741, 0.4237288236618042, 0.2429378479719162, 0.4350282549858093, 0.3446327745914459, 0.49152541160583496, 0.4576271176338196, NaN, 0.5649717450141907, 0.5254237055778503, 0.42937853932380676, 0.6327683329582214, 0.5028248429298401, 0.6440678238868713, 0.6610169410705566, 0.5593220591545105, 0.6779661178588867, 0.5423728823661804, 0.5762711763381958, 0.5932203531265259, 0.5875706076622009, 0.7231638431549072, 0.700564980506897, 0.6892655491828918, 0.7966101765632629, 0.700564980506897, 0.8135592937469482, 0.9322034120559692, 0.8700565099716187, 0.8361582159996033, 0.903954803943634, 0.8587570786476135, 0.9774011373519897, 0.9774011373519897, 0.994350254535675], [0.005649717524647713, 0.016949152573943138, 0.028248587623238564, 0.016949152573943138, 0.07344632595777512, 0.016949152573943138, 0.016949152573943138, 0.02259887009859085, 0.050847455859184265, 0.07909604161977768, 0.06214689090847969, 0.0903954803943634, 0.0903954803943634, 0.10169491171836853, 0.07909604161977768, 0.0903954803943634, 0.0903954803943634, 0.19209039211273193, 0.1355932205915451, 0.16949152946472168, 0.20903955399990082, 0.11864406615495682, 0.2711864411830902, 0.31073445081710815, 0.2542372941970825, 0.26553672552108765, 0.3163841664791107, 0.2768361568450928, 0.32203391194343567, 0.3333333432674408, 0.21468926966190338, 0.32768362760543823, 0.3672316372394562, 0.33898305892944336, 0.2259887009859085, 0.40112993121147156, 0.32203391194343567, 0.5480226278305054, 0.42937853932380676, 0.395480215549469, 0.5593220591545105, 0.38983049988746643, 0.3446327745914459, 0.4971751272678375, 0.4350282549858093, NaN, 0.5706214904785156, 0.508474588394165, 0.6610169410705566, 0.604519784450531, 0.6610169410705566, 0.5423728823661804, 0.6158192157745361, 0.5367231369018555, 0.6214689016342163, 0.598870038986206, 0.6553672552108765, 0.700564980506897, 0.6101694703102112, 0.8474576473236084, 0.7740113139152527, 0.7853107452392578, 0.7231638431549072, 0.7966101765632629, 0.790960431098938, 0.8022598624229431, 0.8587570786476135, 0.8757061958312988, 0.8531073331832886, 0.8757061958312988, 0.8757061958312988, 1.0], [0.011299435049295425, 0.02259887009859085, 0.03954802080988884, 0.028248587623238564, 0.07344632595777512, 0.016949152573943138, 0.033898305147886276, 0.06779661029577255, 0.0451977401971817, 0.03954802080988884, 0.0451977401971817, 0.12429378181695938, 0.0451977401971817, 0.10734463483095169, 0.07344632595777512, 0.0903954803943634, 0.07909604161977768, 0.10734463483095169, 0.07909604161977768, 0.0903954803943634, 0.12429378181695938, 0.18644067645072937, 0.1525423675775528, 0.16384181380271912, 0.19209039211273193, 0.16949152946472168, 0.1525423675775528, 0.20338982343673706, 0.3050847351551056, 0.23728813230991364, 0.2768361568450928, 0.23163841664791107, 0.3672316372394562, 0.2542372941970825, 0.2881355881690979, 0.38983049988746643, 0.3333333432674408, 0.4858756959438324, 0.29378530383110046, 0.451977401971817, 0.49152541160583496, 0.451977401971817, 0.3785310685634613, 0.46327683329582214, 0.47457626461982727, 0.42937853932380676, NaN, 0.42937853932380676, 0.6440678238868713, 0.6384180784225464, 0.6440678238868713, 0.5480226278305054, 0.6779661178588867, 0.5480226278305054, 0.7401130199432373, 0.6158192157745361, 0.7175140976905823, 0.7796609997749329, 0.6440678238868713, 0.694915235042572, 0.7796609997749329, 0.7401130199432373, 0.903954803943634, 0.8983050584793091, 0.8192090392112732, 0.8983050584793091, 0.9265536665916443, 0.9491525292396545, 0.9548022747039795, 0.9491525292396545, 0.9491525292396545, 1.0], [0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.050847455859184265, 0.005649717524647713, 0.028248587623238564, 0.016949152573943138, 0.028248587623238564, 0.016949152573943138, 0.02259887009859085, 0.050847455859184265, 0.05649717524647713, 0.08474576473236084, 0.07909604161977768, 0.07344632595777512, 0.0451977401971817, 0.05649717524647713, 0.14689265191555023, 0.07909604161977768, 0.08474576473236084, 0.15819208323955536, 0.02259887009859085, 0.14689265191555023, 0.14689265191555023, 0.028248587623238564, 0.2429378479719162, 0.07909604161977768, 0.19209039211273193, 0.10734463483095169, 0.2711864411830902, 0.23163841664791107, 0.2768361568450928, 0.24858756363391876, 0.3163841664791107, 0.3446327745914459, 0.38983049988746643, 0.5367231369018555, 0.4237288236618042, 0.4971751272678375, 0.5593220591545105, 0.44632768630981445, 0.4350282549858093, 0.48022598028182983, 0.5706214904785156, 0.49152541160583496, 0.5706214904785156, NaN, 0.6836158037185669, 0.5932203531265259, 0.6836158037185669, 0.6384180784225464, 0.6553672552108765, 0.6327683329582214, 0.700564980506897, 0.7570621371269226, 0.7231638431549072, 0.7457627058029175, 0.6892655491828918, 0.7457627058029175, 0.790960431098938, 0.7231638431549072, 0.8531073331832886, 0.8192090392112732, 0.9152542352676392, 0.909604549407959, 0.8757061958312988, 0.9887005686759949, 0.8983050584793091, 0.9830508232116699, 0.9830508232116699, 0.994350254535675], [0.09604519605636597, 0.10734463483095169, 0.17514124512672424, 0.12429378181695938, 0.16384181380271912, 0.12429378181695938, 0.12994350492954254, 0.16949152946472168, 0.21468926966190338, 0.23163841664791107, 0.2259887009859085, 0.2429378479719162, 0.21468926966190338, 0.2259887009859085, 0.2542372941970825, 0.2711864411830902, 0.2542372941970825, 0.2542372941970825, 0.2768361568450928, 0.2598870098590851, 0.2598870098590851, 0.2768361568450928, 0.28248587250709534, 0.2711864411830902, 0.29378530383110046, 0.2881355881690979, 0.29378530383110046, 0.2881355881690979, 0.2768361568450928, 0.29378530383110046, 0.299435019493103, 0.3163841664791107, 0.2881355881690979, 0.3333333432674408, 0.3050847351551056, 0.28248587250709534, 0.35593220591545105, 0.38983049988746643, 0.32768362760543823, 0.3502824902534485, 0.395480215549469, 0.31073445081710815, 0.3333333432674408, 0.33898305892944336, 0.3672316372394562, 0.33898305892944336, 0.35593220591545105, 0.3163841664791107, NaN, 0.4576271176338196, 0.598870038986206, 0.42937853932380676, 0.5254237055778503, 0.508474588394165, 0.5197740197181702, 0.3672316372394562, 0.4576271176338196, 0.4576271176338196, 0.5593220591545105, 0.44632768630981445, 0.4576271176338196, 0.9774011373519897, 0.5254237055778503, 0.7231638431549072, 0.598870038986206, 0.6892655491828918, 0.7966101765632629, 0.6553672552108765, 0.8361582159996033, 0.7853107452392578, 0.7853107452392578, 0.8926553726196289], [0.005649717524647713, 0.016949152573943138, 0.028248587623238564, 0.011299435049295425, 0.03954802080988884, 0.011299435049295425, 0.03954802080988884, 0.05649717524647713, 0.02259887009859085, 0.02259887009859085, 0.03954802080988884, 0.12994350492954254, 0.033898305147886276, 0.11864406615495682, 0.08474576473236084, 0.07344632595777512, 0.10734463483095169, 0.10169491171836853, 0.16949152946472168, 0.07344632595777512, 0.1355932205915451, 0.24858756363391876, 0.1355932205915451, 0.12429378181695938, 0.20338982343673706, 0.15819208323955536, 0.1807909607887268, 0.21468926966190338, 0.19209039211273193, 0.21468926966190338, 0.33898305892944336, 0.2598870098590851, 0.2542372941970825, 0.2768361568450928, 0.3785310685634613, 0.3050847351551056, 0.299435019493103, 0.19209039211273193, 0.31073445081710815, 0.451977401971817, 0.24858756363391876, 0.3050847351551056, 0.3672316372394562, 0.47457626461982727, 0.4971751272678375, 0.395480215549469, 0.3615819215774536, 0.4067796468734741, 0.5423728823661804, NaN, 0.5593220591545105, 0.5762711763381958, 0.7033898234367371, 0.5875706076622009, 0.7401130199432373, 0.5254237055778503, 0.5367231369018555, 0.6101694703102112, 0.6384180784225464, 0.5254237055778503, 0.598870038986206, 0.700564980506897, 0.9887005686759949, 0.903954803943634, 0.7853107452392578, 0.8813559412956238, 0.9378530979156494, 0.9604519605636597, 0.9604519605636597, 1.0, 1.0, 1.0], [0.0903954803943634, 0.09604519605636597, 0.1525423675775528, 0.12429378181695938, 0.14689265191555023, 0.11299435049295425, 0.12429378181695938, 0.1525423675775528, 0.21468926966190338, 0.2259887009859085, 0.22033898532390594, 0.23728813230991364, 0.22033898532390594, 0.22033898532390594, 0.2429378479719162, 0.2598870098590851, 0.2429378479719162, 0.2542372941970825, 0.26553672552108765, 0.2598870098590851, 0.24858756363391876, 0.26553672552108765, 0.2768361568450928, 0.2598870098590851, 0.2881355881690979, 0.2711864411830902, 0.2768361568450928, 0.2881355881690979, 0.2711864411830902, 0.28248587250709534, 0.2881355881690979, 0.31073445081710815, 0.26553672552108765, 0.33898305892944336, 0.29378530383110046, 0.2768361568450928, 0.3446327745914459, 0.38418078422546387, 0.32203391194343567, 0.3333333432674408, 0.38418078422546387, 0.3050847351551056, 0.33898305892944336, 0.33898305892944336, 0.35593220591545105, 0.33898305892944336, 0.35593220591545105, 0.3163841664791107, 0.40112993121147156, 0.4406779706478119, NaN, 0.4350282549858093, 0.5028248429298401, 0.508474588394165, 0.49152541160583496, 0.3672316372394562, 0.4576271176338196, 0.4576271176338196, 0.5593220591545105, 0.44632768630981445, 0.451977401971817, 0.9604519605636597, 0.5028248429298401, 0.7231638431549072, 0.598870038986206, 0.6892655491828918, 0.790960431098938, 0.6553672552108765, 0.8361582159996033, 0.7853107452392578, 0.7853107452392578, 0.8700565099716187], [0.016949152573943138, 0.011299435049295425, 0.033898305147886276, 0.016949152573943138, 0.02259887009859085, 0.02259887009859085, 0.05649717524647713, 0.033898305147886276, 0.05649717524647713, 0.033898305147886276, 0.06779661029577255, 0.06779661029577255, 0.06779661029577255, 0.03954802080988884, 0.06779661029577255, 0.08474576473236084, 0.09604519605636597, 0.07909604161977768, 0.17514124512672424, 0.17514124512672424, 0.09604519605636597, 0.19209039211273193, 0.07344632595777512, 0.16384181380271912, 0.20903955399990082, 0.09604519605636597, 0.2598870098590851, 0.12994350492954254, 0.10734463483095169, 0.1807909607887268, 0.2711864411830902, 0.2598870098590851, 0.09604519605636597, 0.29378530383110046, 0.2881355881690979, 0.17514124512672424, 0.38983049988746643, 0.3446327745914459, 0.3672316372394562, 0.17514124512672424, 0.3672316372394562, 0.2768361568450928, 0.4237288236618042, 0.44632768630981445, 0.33898305892944336, 0.4576271176338196, 0.451977401971817, 0.3615819215774536, 0.5706214904785156, 0.4237288236618042, 0.5649717450141907, NaN, 0.4689265489578247, 0.5593220591545105, 0.5197740197181702, 0.5367231369018555, 0.5536723136901855, 0.5197740197181702, 0.6101694703102112, 0.6779661178588867, 0.6158192157745361, 0.7288135886192322, 0.6497175097465515, 0.8079096078872681, 0.887005627155304, 0.8248587846755981, 0.8305084705352783, 0.8813559412956238, 0.8757061958312988, 0.9830508232116699, 0.9830508232116699, 0.9887005686759949], [0.005649717524647713, 0.016949152573943138, 0.028248587623238564, 0.011299435049295425, 0.028248587623238564, 0.011299435049295425, 0.033898305147886276, 0.0451977401971817, 0.016949152573943138, 0.016949152573943138, 0.033898305147886276, 0.11299435049295425, 0.028248587623238564, 0.11864406615495682, 0.06214689090847969, 0.07344632595777512, 0.10169491171836853, 0.09604519605636597, 0.16949152946472168, 0.06779661029577255, 0.10734463483095169, 0.23728813230991364, 0.12994350492954254, 0.12429378181695938, 0.19209039211273193, 0.1355932205915451, 0.14689265191555023, 0.20338982343673706, 0.17514124512672424, 0.20338982343673706, 0.31073445081710815, 0.26553672552108765, 0.23163841664791107, 0.2768361568450928, 0.37288135290145874, 0.26553672552108765, 0.28248587250709534, 0.1355932205915451, 0.31073445081710815, 0.4067796468734741, 0.19209039211273193, 0.3050847351551056, 0.35593220591545105, 0.44632768630981445, 0.4406779706478119, 0.38418078422546387, 0.32203391194343567, 0.3446327745914459, 0.47457626461982727, 0.29661017656326294, 0.4971751272678375, 0.5310734510421753, NaN, 0.5593220591545105, 0.6214689016342163, 0.47457626461982727, 0.508474588394165, 0.5706214904785156, 0.6101694703102112, 0.508474588394165, 0.5593220591545105, 0.694915235042572, 0.790960431098938, 0.8700565099716187, 0.7627118825912476, 0.8813559412956238, 0.9378530979156494, 0.9604519605636597, 0.9491525292396545, 1.0, 1.0, 1.0], [0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.033898305147886276, 0.005649717524647713, 0.011299435049295425, 0.016949152573943138, 0.016949152573943138, 0.028248587623238564, 0.02259887009859085, 0.05649717524647713, 0.03954802080988884, 0.06779661029577255, 0.0451977401971817, 0.05649717524647713, 0.03954802080988884, 0.03954802080988884, 0.14689265191555023, 0.08474576473236084, 0.050847455859184265, 0.15819208323955536, 0.033898305147886276, 0.14689265191555023, 0.1977401077747345, 0.06214689090847969, 0.16384181380271912, 0.07344632595777512, 0.07344632595777512, 0.14124293625354767, 0.24858756363391876, 0.2542372941970825, 0.08474576473236084, 0.2598870098590851, 0.29378530383110046, 0.12994350492954254, 0.395480215549469, 0.3502824902534485, 0.3502824902534485, 0.2542372941970825, 0.3615819215774536, 0.22033898532390594, 0.4406779706478119, 0.4067796468734741, 0.32203391194343567, 0.46327683329582214, 0.451977401971817, 0.3672316372394562, 0.49152541160583496, 0.4124293923377991, 0.49152541160583496, 0.4406779706478119, 0.4406779706478119, NaN, 0.5028248429298401, 0.5536723136901855, 0.6101694703102112, 0.5932203531265259, 0.7683615684509277, 0.6101694703102112, 0.6440678238868713, 0.5875706076622009, 0.6440678238868713, 0.7175140976905823, 0.9548022747039795, 0.8248587846755981, 0.8700565099716187, 0.8531073331832886, 0.8926553726196289, 0.9774011373519897, 0.9774011373519897, 0.994350254535675], [0.02259887009859085, 0.03954802080988884, 0.03954802080988884, 0.028248587623238564, 0.050847455859184265, 0.0451977401971817, 0.06214689090847969, 0.05649717524647713, 0.0451977401971817, 0.028248587623238564, 0.050847455859184265, 0.11299435049295425, 0.050847455859184265, 0.12429378181695938, 0.09604519605636597, 0.07909604161977768, 0.10169491171836853, 0.1355932205915451, 0.10734463483095169, 0.0903954803943634, 0.15819208323955536, 0.23163841664791107, 0.14124293625354767, 0.12429378181695938, 0.09604519605636597, 0.17514124512672424, 0.12994350492954254, 0.19209039211273193, 0.20338982343673706, 0.21468926966190338, 0.3050847351551056, 0.20903955399990082, 0.2711864411830902, 0.19209039211273193, 0.35593220591545105, 0.28248587250709534, 0.2768361568450928, 0.05649717524647713, 0.2768361568450928, 0.42937853932380676, 0.11299435049295425, 0.29378530383110046, 0.35593220591545105, 0.47457626461982727, 0.4576271176338196, 0.3785310685634613, 0.2598870098590851, 0.299435019493103, 0.48022598028182983, 0.2598870098590851, 0.508474588394165, 0.48022598028182983, 0.3785310685634613, 0.4971751272678375, NaN, 0.37288135290145874, 0.4406779706478119, 0.5875706076622009, 0.508474588394165, 0.4067796468734741, 0.4350282549858093, 0.6723163723945618, 0.6892655491828918, 0.9152542352676392, 0.6440678238868713, 0.7231638431549072, 0.9548022747039795, 0.8587570786476135, 0.9774011373519897, 0.994350254535675, 0.994350254535675, 1.0], [0.0, 0.005649717524647713, 0.011299435049295425, 0.005649717524647713, 0.050847455859184265, 0.0, 0.005649717524647713, 0.02259887009859085, 0.005649717524647713, 0.005649717524647713, 0.005649717524647713, 0.05649717524647713, 0.005649717524647713, 0.050847455859184265, 0.028248587623238564, 0.033898305147886276, 0.02259887009859085, 0.005649717524647713, 0.05649717524647713, 0.033898305147886276, 0.005649717524647713, 0.09604519605636597, 0.028248587623238564, 0.10734463483095169, 0.06779661029577255, 0.011299435049295425, 0.11299435049295425, 0.03954802080988884, 0.17514124512672424, 0.06779661029577255, 0.1807909607887268, 0.10734463483095169, 0.23728813230991364, 0.11864406615495682, 0.20338982343673706, 0.2768361568450928, 0.2542372941970825, 0.41807910799980164, 0.2881355881690979, 0.37288135290145874, 0.4689265489578247, 0.29378530383110046, 0.32768362760543823, 0.40112993121147156, 0.4237288236618042, 0.40112993121147156, 0.38418078422546387, 0.2429378479719162, 0.6327683329582214, 0.47457626461982727, 0.6327683329582214, 0.46327683329582214, 0.5254237055778503, 0.44632768630981445, 0.6271186470985413, NaN, 0.5593220591545105, 0.7231638431549072, 0.5593220591545105, 0.5706214904785156, 0.6666666865348816, 0.6836158037185669, 0.8022598624229431, 0.8022598624229431, 0.7514124512672424, 0.7683615684509277, 0.8757061958312988, 0.9435028433799744, 0.8983050584793091, 0.9378530979156494, 0.9378530979156494, 0.9830508232116699], [0.0, 0.005649717524647713, 0.0, 0.016949152573943138, 0.050847455859184265, 0.02259887009859085, 0.005649717524647713, 0.0, 0.0, 0.0, 0.005649717524647713, 0.028248587623238564, 0.016949152573943138, 0.033898305147886276, 0.028248587623238564, 0.02259887009859085, 0.005649717524647713, 0.07909604161977768, 0.0451977401971817, 0.016949152573943138, 0.07909604161977768, 0.07344632595777512, 0.033898305147886276, 0.10169491171836853, 0.1525423675775528, 0.06779661029577255, 0.0903954803943634, 0.07909604161977768, 0.10169491171836853, 0.10169491171836853, 0.16384181380271912, 0.19209039211273193, 0.17514124512672424, 0.16384181380271912, 0.1977401077747345, 0.20903955399990082, 0.20903955399990082, 0.2711864411830902, 0.18644067645072937, 0.3502824902534485, 0.3163841664791107, 0.3050847351551056, 0.2259887009859085, 0.3502824902534485, 0.4067796468734741, 0.3446327745914459, 0.28248587250709534, 0.2768361568450928, 0.5423728823661804, 0.46327683329582214, 0.5423728823661804, 0.44632768630981445, 0.49152541160583496, 0.38983049988746643, 0.5593220591545105, 0.4406779706478119, NaN, 0.6101694703102112, 0.44632768630981445, 0.5875706076622009, 0.5762711763381958, 0.6610169410705566, 0.6836158037185669, 0.8192090392112732, 0.694915235042572, 0.9378530979156494, 0.8587570786476135, 0.9209039807319641, 0.8587570786476135, 0.9717513918876648, 0.9717513918876648, 1.0], [0.011299435049295425, 0.033898305147886276, 0.0451977401971817, 0.03954802080988884, 0.033898305147886276, 0.028248587623238564, 0.033898305147886276, 0.08474576473236084, 0.050847455859184265, 0.03954802080988884, 0.0451977401971817, 0.12429378181695938, 0.03954802080988884, 0.10734463483095169, 0.11299435049295425, 0.06779661029577255, 0.0903954803943634, 0.10169491171836853, 0.05649717524647713, 0.06214689090847969, 0.10734463483095169, 0.16384181380271912, 0.14124293625354767, 0.12994350492954254, 0.033898305147886276, 0.14124293625354767, 0.09604519605636597, 0.1355932205915451, 0.2259887009859085, 0.14689265191555023, 0.22033898532390594, 0.10734463483095169, 0.29378530383110046, 0.12429378181695938, 0.23728813230991364, 0.3050847351551056, 0.20903955399990082, 0.2429378479719162, 0.19209039211273193, 0.395480215549469, 0.24858756363391876, 0.3502824902534485, 0.2598870098590851, 0.33898305892944336, 0.4124293923377991, 0.299435019493103, 0.22033898532390594, 0.2542372941970825, 0.5423728823661804, 0.38983049988746643, 0.5423728823661804, 0.48022598028182983, 0.42937853932380676, 0.4067796468734741, 0.4124293923377991, 0.2768361568450928, 0.38983049988746643, NaN, 0.4237288236618042, 0.40112993121147156, 0.44632768630981445, 0.6497175097465515, 0.5649717450141907, 0.8418079018592834, 0.5819209218025208, 0.6553672552108765, 0.903954803943634, 0.7796609997749329, 0.9717513918876648, 0.903954803943634, 0.8926553726196289, 0.9887005686759949], [0.005649717524647713, 0.005649717524647713, 0.011299435049295425, 0.011299435049295425, 0.033898305147886276, 0.005649717524647713, 0.005649717524647713, 0.016949152573943138, 0.016949152573943138, 0.016949152573943138, 0.02259887009859085, 0.0451977401971817, 0.016949152573943138, 0.05649717524647713, 0.033898305147886276, 0.033898305147886276, 0.0451977401971817, 0.02259887009859085, 0.0903954803943634, 0.050847455859184265, 0.033898305147886276, 0.1355932205915451, 0.033898305147886276, 0.1355932205915451, 0.18644067645072937, 0.03954802080988884, 0.12994350492954254, 0.07909604161977768, 0.05649717524647713, 0.11299435049295425, 0.23163841664791107, 0.2259887009859085, 0.0903954803943634, 0.22033898532390594, 0.2768361568450928, 0.12429378181695938, 0.3163841664791107, 0.2881355881690979, 0.28248587250709534, 0.23163841664791107, 0.3050847351551056, 0.17514124512672424, 0.35593220591545105, 0.40112993121147156, 0.2768361568450928, 0.38983049988746643, 0.35593220591545105, 0.31073445081710815, 0.4406779706478119, 0.3615819215774536, 0.4406779706478119, 0.38983049988746643, 0.38983049988746643, 0.23163841664791107, 0.49152541160583496, 0.4406779706478119, 0.5536723136901855, 0.5762711763381958, NaN, 0.5706214904785156, 0.5932203531265259, 0.5480226278305054, 0.6158192157745361, 0.6610169410705566, 0.7881355881690979, 0.7401130199432373, 0.8474576473236084, 0.8079096078872681, 0.8700565099716187, 0.9717513918876648, 0.9717513918876648, 0.9887005686759949], [0.0, 0.0, 0.0, 0.0, 0.050847455859184265, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.02259887009859085, 0.011299435049295425, 0.005649717524647713, 0.06779661029577255, 0.011299435049295425, 0.028248587623238564, 0.07344632595777512, 0.03954802080988884, 0.10734463483095169, 0.11864406615495682, 0.07909604161977768, 0.11864406615495682, 0.11299435049295425, 0.10734463483095169, 0.1977401077747345, 0.14689265191555023, 0.07344632595777512, 0.09604519605636597, 0.20903955399990082, 0.14689265191555023, 0.10734463483095169, 0.2259887009859085, 0.10734463483095169, 0.3615819215774536, 0.23728813230991364, 0.28248587250709534, 0.38418078422546387, 0.2598870098590851, 0.07344632595777512, 0.3785310685634613, 0.299435019493103, 0.1525423675775528, 0.3050847351551056, 0.2542372941970825, 0.5536723136901855, 0.47457626461982727, 0.5536723136901855, 0.32203391194343567, 0.49152541160583496, 0.38983049988746643, 0.5932203531265259, 0.42937853932380676, 0.4124293923377991, 0.598870038986206, 0.42937853932380676, NaN, 0.48022598028182983, 0.6440678238868713, 0.6836158037185669, 0.7401130199432373, 0.5762711763381958, 0.5932203531265259, 0.8135592937469482, 0.8135592937469482, 0.8248587846755981, 0.8587570786476135, 0.8587570786476135, 0.9830508232116699], [0.0, 0.0, 0.0, 0.011299435049295425, 0.050847455859184265, 0.011299435049295425, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.016949152573943138, 0.02259887009859085, 0.011299435049295425, 0.011299435049295425, 0.0, 0.03954802080988884, 0.03954802080988884, 0.02259887009859085, 0.050847455859184265, 0.02259887009859085, 0.011299435049295425, 0.10169491171836853, 0.11864406615495682, 0.02259887009859085, 0.07909604161977768, 0.03954802080988884, 0.08474576473236084, 0.05649717524647713, 0.08474576473236084, 0.14689265191555023, 0.16949152946472168, 0.14124293625354767, 0.10169491171836853, 0.18644067645072937, 0.1525423675775528, 0.28248587250709534, 0.14124293625354767, 0.2711864411830902, 0.29378530383110046, 0.26553672552108765, 0.12994350492954254, 0.3163841664791107, 0.31073445081710815, 0.2259887009859085, 0.22033898532390594, 0.20903955399990082, 0.5423728823661804, 0.40112993121147156, 0.5480226278305054, 0.38418078422546387, 0.4406779706478119, 0.35593220591545105, 0.5649717450141907, 0.3333333432674408, 0.4237288236618042, 0.5536723136901855, 0.4067796468734741, 0.5197740197181702, NaN, 0.6666666865348816, 0.6836158037185669, 0.7853107452392578, 0.6440678238868713, 0.7966101765632629, 0.8192090392112732, 0.887005627155304, 0.8418079018592834, 0.9435028433799744, 0.9378530979156494, 1.0], [0.07344632595777512, 0.0903954803943634, 0.0903954803943634, 0.09604519605636597, 0.08474576473236084, 0.07909604161977768, 0.08474576473236084, 0.10169491171836853, 0.10734463483095169, 0.12429378181695938, 0.10169491171836853, 0.1355932205915451, 0.09604519605636597, 0.12994350492954254, 0.15819208323955536, 0.14124293625354767, 0.1525423675775528, 0.16384181380271912, 0.1807909607887268, 0.16949152946472168, 0.17514124512672424, 0.16949152946472168, 0.19209039211273193, 0.1525423675775528, 0.2259887009859085, 0.20338982343673706, 0.2259887009859085, 0.20338982343673706, 0.17514124512672424, 0.21468926966190338, 0.20338982343673706, 0.24858756363391876, 0.1807909607887268, 0.26553672552108765, 0.20903955399990082, 0.20903955399990082, 0.22033898532390594, 0.2429378479719162, 0.2768361568450928, 0.19209039211273193, 0.24858756363391876, 0.22033898532390594, 0.20903955399990082, 0.29378530383110046, 0.20338982343673706, 0.21468926966190338, 0.2598870098590851, 0.2768361568450928, 0.02259887009859085, 0.299435019493103, 0.03954802080988884, 0.2711864411830902, 0.3050847351551056, 0.4124293923377991, 0.32768362760543823, 0.3163841664791107, 0.33898305892944336, 0.3502824902534485, 0.451977401971817, 0.35593220591545105, 0.3333333432674408, NaN, 0.38418078422546387, 0.6271186470985413, 0.47457626461982727, 0.6610169410705566, 0.6666666865348816, 0.5536723136901855, 0.7401130199432373, 0.7796609997749329, 0.7796609997749329, 0.790960431098938], [0.005649717524647713, 0.011299435049295425, 0.005649717524647713, 0.005649717524647713, 0.033898305147886276, 0.005649717524647713, 0.0, 0.011299435049295425, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0451977401971817, 0.005649717524647713, 0.06779661029577255, 0.011299435049295425, 0.016949152573943138, 0.028248587623238564, 0.016949152573943138, 0.050847455859184265, 0.02259887009859085, 0.028248587623238564, 0.07909604161977768, 0.033898305147886276, 0.05649717524647713, 0.028248587623238564, 0.02259887009859085, 0.0451977401971817, 0.0451977401971817, 0.07344632595777512, 0.06214689090847969, 0.14124293625354767, 0.050847455859184265, 0.07344632595777512, 0.07909604161977768, 0.18644067645072937, 0.14124293625354767, 0.16384181380271912, 0.0451977401971817, 0.11299435049295425, 0.2598870098590851, 0.06214689090847969, 0.1525423675775528, 0.23163841664791107, 0.33898305892944336, 0.299435019493103, 0.2768361568450928, 0.09604519605636597, 0.14689265191555023, 0.47457626461982727, 0.011299435049295425, 0.4971751272678375, 0.3502824902534485, 0.20903955399990082, 0.35593220591545105, 0.31073445081710815, 0.1977401077747345, 0.3163841664791107, 0.4350282549858093, 0.38418078422546387, 0.3163841664791107, 0.3163841664791107, 0.6158192157745361, NaN, 0.7627118825912476, 0.5536723136901855, 0.604519784450531, 0.8418079018592834, 0.7740113139152527, 0.8983050584793091, 0.9435028433799744, 0.9435028433799744, 0.9774011373519897], [0.016949152573943138, 0.02259887009859085, 0.03954802080988884, 0.028248587623238564, 0.02259887009859085, 0.016949152573943138, 0.028248587623238564, 0.03954802080988884, 0.02259887009859085, 0.03954802080988884, 0.02259887009859085, 0.08474576473236084, 0.02259887009859085, 0.09604519605636597, 0.06214689090847969, 0.03954802080988884, 0.07909604161977768, 0.06779661029577255, 0.03954802080988884, 0.0451977401971817, 0.07909604161977768, 0.11299435049295425, 0.09604519605636597, 0.050847455859184265, 0.03954802080988884, 0.0903954803943634, 0.05649717524647713, 0.10734463483095169, 0.1355932205915451, 0.11299435049295425, 0.14689265191555023, 0.06214689090847969, 0.14689265191555023, 0.11864406615495682, 0.20338982343673706, 0.15819208323955536, 0.10734463483095169, 0.033898305147886276, 0.09604519605636597, 0.17514124512672424, 0.03954802080988884, 0.16384181380271912, 0.17514124512672424, 0.23163841664791107, 0.18644067645072937, 0.20338982343673706, 0.10169491171836853, 0.1807909607887268, 0.2768361568450928, 0.09604519605636597, 0.2768361568450928, 0.19209039211273193, 0.12994350492954254, 0.28248587250709534, 0.08474576473236084, 0.1977401077747345, 0.1807909607887268, 0.15819208323955536, 0.33898305892944336, 0.2598870098590851, 0.21468926966190338, 0.37288135290145874, 0.23728813230991364, NaN, 0.4350282549858093, 0.40112993121147156, 0.9152542352676392, 0.32768362760543823, 0.9661017060279846, 0.8474576473236084, 0.8474576473236084, 0.9661017060279846], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.0, 0.028248587623238564, 0.016949152573943138, 0.005649717524647713, 0.028248587623238564, 0.0, 0.016949152573943138, 0.05649717524647713, 0.016949152573943138, 0.05649717524647713, 0.016949152573943138, 0.0, 0.016949152573943138, 0.08474576473236084, 0.07909604161977768, 0.011299435049295425, 0.08474576473236084, 0.0903954803943634, 0.016949152573943138, 0.16384181380271912, 0.1807909607887268, 0.14689265191555023, 0.050847455859184265, 0.16949152946472168, 0.028248587623238564, 0.14689265191555023, 0.2542372941970825, 0.06779661029577255, 0.20903955399990082, 0.1807909607887268, 0.08474576473236084, 0.40112993121147156, 0.21468926966190338, 0.40112993121147156, 0.11299435049295425, 0.23728813230991364, 0.0451977401971817, 0.35593220591545105, 0.24858756363391876, 0.3050847351551056, 0.41807910799980164, 0.2118644118309021, 0.4237288236618042, 0.35593220591545105, 0.5254237055778503, 0.44632768630981445, 0.5649717450141907, NaN, 0.508474588394165, 0.694915235042572, 0.6271186470985413, 0.8135592937469482, 0.8926553726196289, 0.8926553726196289, 0.9661017060279846], [0.0, 0.0, 0.0, 0.0, 0.050847455859184265, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.0, 0.005649717524647713, 0.0, 0.005649717524647713, 0.016949152573943138, 0.0451977401971817, 0.011299435049295425, 0.05649717524647713, 0.033898305147886276, 0.011299435049295425, 0.02259887009859085, 0.02259887009859085, 0.016949152573943138, 0.028248587623238564, 0.10734463483095169, 0.06214689090847969, 0.028248587623238564, 0.0451977401971817, 0.11864406615495682, 0.07344632595777512, 0.07344632595777512, 0.07909604161977768, 0.06779661029577255, 0.09604519605636597, 0.09604519605636597, 0.08474576473236084, 0.12994350492954254, 0.3163841664791107, 0.12994350492954254, 0.1977401077747345, 0.10169491171836853, 0.0903954803943634, 0.31073445081710815, 0.11864406615495682, 0.31073445081710815, 0.17514124512672424, 0.11864406615495682, 0.17514124512672424, 0.2768361568450928, 0.23163841664791107, 0.06214689090847969, 0.3446327745914459, 0.2598870098590851, 0.4067796468734741, 0.20338982343673706, 0.33898305892944336, 0.395480215549469, 0.598870038986206, 0.49152541160583496, NaN, 0.8022598624229431, 0.5197740197181702, 0.8474576473236084, 0.9265536665916443, 0.9265536665916443, 0.994350254535675], [0.005649717524647713, 0.011299435049295425, 0.02259887009859085, 0.016949152573943138, 0.011299435049295425, 0.016949152573943138, 0.011299435049295425, 0.028248587623238564, 0.011299435049295425, 0.02259887009859085, 0.011299435049295425, 0.033898305147886276, 0.011299435049295425, 0.06214689090847969, 0.0451977401971817, 0.033898305147886276, 0.0451977401971817, 0.03954802080988884, 0.02259887009859085, 0.033898305147886276, 0.0451977401971817, 0.07909604161977768, 0.06779661029577255, 0.016949152573943138, 0.02259887009859085, 0.0451977401971817, 0.028248587623238564, 0.05649717524647713, 0.050847455859184265, 0.07344632595777512, 0.10734463483095169, 0.0451977401971817, 0.06779661029577255, 0.06779661029577255, 0.1525423675775528, 0.09604519605636597, 0.050847455859184265, 0.011299435049295425, 0.05649717524647713, 0.15819208323955536, 0.005649717524647713, 0.11299435049295425, 0.12994350492954254, 0.16384181380271912, 0.16384181380271912, 0.14124293625354767, 0.07344632595777512, 0.12429378181695938, 0.20338982343673706, 0.06214689090847969, 0.20903955399990082, 0.16949152946472168, 0.06214689090847969, 0.12994350492954254, 0.0451977401971817, 0.12429378181695938, 0.14124293625354767, 0.09604519605636597, 0.1525423675775528, 0.18644067645072937, 0.1807909607887268, 0.3333333432674408, 0.15819208323955536, 0.08474576473236084, 0.3050847351551056, 0.1977401077747345, NaN, 0.24858756363391876, 0.8983050584793091, 0.8248587846755981, 0.8248587846755981, 0.8926553726196289], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.005649717524647713, 0.0, 0.011299435049295425, 0.0, 0.011299435049295425, 0.02259887009859085, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02259887009859085, 0.011299435049295425, 0.0, 0.016949152573943138, 0.016949152573943138, 0.0, 0.0451977401971817, 0.005649717524647713, 0.0451977401971817, 0.0903954803943634, 0.005649717524647713, 0.011299435049295425, 0.11864406615495682, 0.23728813230991364, 0.09604519605636597, 0.12429378181695938, 0.050847455859184265, 0.011299435049295425, 0.3446327745914459, 0.03954802080988884, 0.3446327745914459, 0.11864406615495682, 0.03954802080988884, 0.14689265191555023, 0.14124293625354767, 0.05649717524647713, 0.07909604161977768, 0.22033898532390594, 0.19209039211273193, 0.18644067645072937, 0.11299435049295425, 0.44632768630981445, 0.2259887009859085, 0.6723163723945618, 0.37288135290145874, 0.48022598028182983, 0.7514124512672424, NaN, 0.7740113139152527, 0.7796609997749329, 0.7740113139152527, 0.8474576473236084], [0.011299435049295425, 0.02259887009859085, 0.011299435049295425, 0.02259887009859085, 0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.016949152573943138, 0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.03954802080988884, 0.011299435049295425, 0.050847455859184265, 0.028248587623238564, 0.028248587623238564, 0.033898305147886276, 0.033898305147886276, 0.011299435049295425, 0.011299435049295425, 0.03954802080988884, 0.06779661029577255, 0.03954802080988884, 0.016949152573943138, 0.02259887009859085, 0.0451977401971817, 0.02259887009859085, 0.0451977401971817, 0.03954802080988884, 0.06779661029577255, 0.07909604161977768, 0.033898305147886276, 0.050847455859184265, 0.07344632595777512, 0.14124293625354767, 0.07344632595777512, 0.033898305147886276, 0.011299435049295425, 0.0451977401971817, 0.11299435049295425, 0.011299435049295425, 0.09604519605636597, 0.11864406615495682, 0.06214689090847969, 0.14124293625354767, 0.14689265191555023, 0.0451977401971817, 0.10169491171836853, 0.16384181380271912, 0.03954802080988884, 0.16384181380271912, 0.12429378181695938, 0.050847455859184265, 0.10734463483095169, 0.02259887009859085, 0.10169491171836853, 0.14124293625354767, 0.028248587623238564, 0.12994350492954254, 0.17514124512672424, 0.15819208323955536, 0.2598870098590851, 0.10169491171836853, 0.033898305147886276, 0.18644067645072937, 0.1525423675775528, 0.10169491171836853, 0.2259887009859085, NaN, 0.5875706076622009, 0.5932203531265259, 0.6384180784225464], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.005649717524647713, 0.005649717524647713, 0.0, 0.005649717524647713, 0.0, 0.0, 0.011299435049295425, 0.033898305147886276, 0.016949152573943138, 0.005649717524647713, 0.0, 0.02259887009859085, 0.011299435049295425, 0.033898305147886276, 0.0, 0.02259887009859085, 0.016949152573943138, 0.0, 0.0, 0.0903954803943634, 0.08474576473236084, 0.02259887009859085, 0.12429378181695938, 0.050847455859184265, 0.016949152573943138, 0.21468926966190338, 0.0, 0.21468926966190338, 0.016949152573943138, 0.0, 0.02259887009859085, 0.005649717524647713, 0.06214689090847969, 0.028248587623238564, 0.09604519605636597, 0.028248587623238564, 0.14124293625354767, 0.05649717524647713, 0.22033898532390594, 0.05649717524647713, 0.1525423675775528, 0.10734463483095169, 0.07344632595777512, 0.17514124512672424, 0.22033898532390594, 0.4124293923377991, NaN, 0.5310734510421753, 0.7796609997749329], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.005649717524647713, 0.005649717524647713, 0.0, 0.005649717524647713, 0.0, 0.0, 0.011299435049295425, 0.033898305147886276, 0.016949152573943138, 0.005649717524647713, 0.0, 0.02259887009859085, 0.016949152573943138, 0.033898305147886276, 0.0, 0.02259887009859085, 0.016949152573943138, 0.0, 0.0, 0.0903954803943634, 0.08474576473236084, 0.02259887009859085, 0.12429378181695938, 0.050847455859184265, 0.016949152573943138, 0.21468926966190338, 0.0, 0.21468926966190338, 0.016949152573943138, 0.0, 0.02259887009859085, 0.005649717524647713, 0.06214689090847969, 0.028248587623238564, 0.10734463483095169, 0.028248587623238564, 0.14124293625354767, 0.06214689090847969, 0.22033898532390594, 0.05649717524647713, 0.1525423675775528, 0.10734463483095169, 0.07344632595777512, 0.17514124512672424, 0.2259887009859085, 0.4067796468734741, 0.4689265489578247, NaN, 0.7683615684509277], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.005649717524647713, 0.0, 0.005649717524647713, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.016949152573943138, 0.005649717524647713, 0.0, 0.0, 0.005649717524647713, 0.10734463483095169, 0.0, 0.12994350492954254, 0.011299435049295425, 0.0, 0.005649717524647713, 0.0, 0.016949152573943138, 0.0, 0.011299435049295425, 0.011299435049295425, 0.016949152573943138, 0.0, 0.20903955399990082, 0.02259887009859085, 0.033898305147886276, 0.033898305147886276, 0.005649717524647713, 0.10734463483095169, 0.1525423675775528, 0.3615819215774536, 0.22033898532390594, 0.23163841664791107, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9365799276},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.9158112486},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8818333749},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.8789687249},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8631336034},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.862099148},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8605076767},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8470597607},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8195273327},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.798042488},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.7962918761},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7619559158},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7511737075},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7315588417},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7273016607},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7187873057},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7051802363},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.669372163},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6639611697},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6582318777},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6440677968},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6386568007},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6353147145},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.633723242},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6139890181},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.605713378},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5844672558},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5746797181},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5722129375},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5559799453},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.5538314666},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5450783808},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5399061021},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5142834402},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5124532498},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.506883104},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4765656078},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4622821667},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.4613670719},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4487944619},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4437415417},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.442667302},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4382907636},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4087689946},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4069388073},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.403517148},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3980265773},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3963555318},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.378849368},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3715683905},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3690618288},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.3462242385},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3442746868},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3291159382},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.3253759836},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3209994409},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.300946923},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2974456906},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2907217313},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2731757775},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2670486194},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2641839732},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2223283197},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1866793986},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1832179514},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1702872584},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1301822231},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.1283520334},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0911912145},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0592822472},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0585660854},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0249860748}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_all/tasks_regression/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_all/tasks_regression/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..3ac6dfa37630c95ee3a39cf33d973a101db9b238
--- /dev/null
+++ b/data/imputation_no/splits_all/tasks_regression/datasets_small/winrate_matrix.csv
@@ -0,0 +1,73 @@
+,TabFM (default),Nori-30M (default),TabPFN-3 (default),Nori (default),TabICLv2 (default),TabPFN-2.6 (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),TabDPT (tuned + ensembled),RealMLP (tuned + ensembled),TabDPT (tuned),RealTabPFN-2.5 (tuned),TabDPT (default),RealTabPFN-2.5 (default),RealMLP (tuned),TabDPT-Turbo (default),"AutoGluon 1.4 (best, 4h)",CatBoost (tuned + ensembled),TabSwift (default),iLTM (tuned + ensembled),CatBoost (tuned),TabM (tuned + ensembled),LightGBM (tuned + ensembled),SAP-RPT-OSS (default),xRFM (tuned + ensembled),XGBoost (tuned + ensembled),ModernNCA (tuned + ensembled),XGBoost (tuned),CatBoost (default),LightGBM (tuned),TabM (tuned),xRFM (tuned),ChimeraBoost (tuned + ensembled),iLTM (tuned),TabM (default),ChimeraBoost (tuned),RealMLP (default),ExtraTrees (tuned + ensembled),ModernNCA (tuned),EBM (tuned + ensembled),ExtraTrees (tuned),ChimeraBoost (default),TorchMLP (tuned + ensembled),LimiX (default),EBM (tuned),TorchMLP (tuned),ModernNCA (default),LightGBM (default),TabSTAR (tuned + ensembled),RandomForest (tuned + ensembled),TabSTAR (tuned),EBM (default),RandomForest (tuned),PerpetualBooster (tuned + ensembled),ExtraTrees (default),XGBoost (default),FastaiMLP (tuned + ensembled),xRFM (default),PerpetualBooster (tuned),TorchMLP (default),FastaiMLP (tuned),TabSTAR (default),RandomForest (default),KNN (tuned + ensembled),PerpetualBooster (default),FastaiMLP (default),KNN (tuned),iLTM (default),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,0.5141243,0.68926555,0.7457627,0.70621467,0.76836157,0.7118644,0.740113,0.8757062,0.8757062,0.920904,0.76836157,0.8926554,0.91525424,0.9322034,0.9548023,0.9491525,0.9548023,0.8700565,0.9548023,0.9717514,0.94350284,0.9661017,0.91525424,0.9378531,0.9548023,0.9491525,0.9661017,0.96045196,0.9717514,0.96045196,0.96045196,0.9661017,0.9717514,0.9717514,0.96045196,0.98870057,0.9661017,0.98870057,0.9661017,0.9717514,0.99435025,0.98870057,0.9830508,0.97740114,0.99435025,0.98870057,1.0,0.9039548,0.99435025,0.90960455,0.9830508,0.99435025,0.98870057,0.97740114,1.0,1.0,0.98870057,0.99435025,1.0,1.0,0.92655367,0.99435025,0.9830508,1.0,1.0,0.99435025,1.0,0.98870057,1.0,1.0,1.0
+Nori-30M (default),0.4858757,,0.5480226,0.72316384,0.63276833,0.76836157,0.7344633,0.7344633,0.70621467,0.779661,0.79096043,0.8135593,0.79096043,0.8700565,0.8700565,0.86440676,0.90960455,0.92655367,0.9039548,0.8870056,0.96045196,0.920904,0.9548023,0.88135594,0.91525424,0.9378531,0.9039548,0.9491525,0.96045196,0.9661017,0.94350284,0.94350284,0.97740114,0.97740114,0.9661017,0.9830508,0.9661017,0.9548023,0.9491525,0.9830508,0.9378531,0.97740114,0.9661017,0.97740114,0.9830508,0.9830508,0.97740114,0.99435025,0.8926554,0.9830508,0.9039548,0.98870057,0.9830508,0.98870057,0.96045196,0.99435025,0.99435025,0.9661017,0.99435025,1.0,1.0,0.90960455,0.98870057,0.97740114,1.0,1.0,0.98870057,1.0,0.97740114,1.0,1.0,1.0
+TabPFN-3 (default),0.31073445,0.4519774,,0.5536723,0.55932206,0.59887004,0.6045198,0.5819209,0.70621467,0.79096043,0.7288136,0.7457627,0.8135593,0.84745765,0.8305085,0.8870056,0.85310733,0.8700565,0.920904,0.8870056,0.8757062,0.86440676,0.90960455,0.8418079,0.8926554,0.90960455,0.8700565,0.9039548,0.8926554,0.9378531,0.9039548,0.90960455,0.89830506,0.9661017,0.920904,0.9039548,0.96045196,0.9322034,0.920904,0.92655367,0.94350284,0.9491525,0.9661017,0.90960455,0.94350284,0.9717514,0.96045196,0.98870057,0.8248588,0.9717514,0.84745765,0.9661017,0.9717514,0.98870057,0.96045196,0.98870057,1.0,0.9548023,0.98870057,1.0,1.0,0.90960455,0.99435025,0.96045196,1.0,1.0,0.97740114,1.0,0.98870057,1.0,1.0,1.0
+Nori (default),0.2542373,0.27683616,0.4463277,,0.4915254,0.6158192,0.5706215,0.5706215,0.66101694,0.6666667,0.700565,0.66101694,0.7344633,0.8587571,0.79096043,0.8361582,0.8926554,0.9548023,0.76836157,0.85310733,0.9491525,0.86440676,0.9491525,0.8587571,0.9039548,0.920904,0.88135594,0.9378531,0.9491525,0.9491525,0.9039548,0.9491525,0.96045196,0.97740114,0.9491525,0.9717514,0.920904,0.9378531,0.94350284,0.97740114,0.9378531,0.9830508,0.9548023,0.9322034,0.9661017,0.9830508,0.9717514,1.0,0.8757062,0.98870057,0.8757062,0.9830508,0.98870057,0.98870057,0.9717514,0.99435025,0.9830508,0.96045196,0.98870057,1.0,0.98870057,0.9039548,0.99435025,0.9717514,1.0,1.0,0.9830508,1.0,0.97740114,1.0,1.0,1.0
+TabICLv2 (default),0.2937853,0.36723164,0.44067797,0.5084746,,0.5762712,0.49717513,0.5480226,0.7288136,0.7457627,0.72316384,0.6836158,0.7627119,0.84745765,0.8248588,0.8305085,0.8587571,0.89830506,0.79096043,0.8926554,0.8926554,0.8926554,0.8926554,0.8870056,0.8757062,0.9039548,0.86440676,0.8926554,0.9039548,0.91525424,0.89830506,0.8926554,0.8870056,0.92655367,0.9039548,0.8926554,0.9322034,0.89830506,0.920904,0.9378531,0.8926554,0.9378531,0.94350284,0.91525424,0.9322034,0.92655367,0.92655367,0.9491525,0.8361582,0.96045196,0.85310733,0.97740114,0.9717514,0.9661017,0.9491525,0.9491525,0.9491525,0.9661017,0.9661017,0.9491525,0.9491525,0.91525424,0.9661017,0.97740114,1.0,0.9491525,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+TabPFN-2.6 (default),0.23163842,0.23163842,0.40112993,0.38418078,0.42372882,,0.43502825,0.5254237,0.6158192,0.6158192,0.6497175,0.6497175,0.700565,0.8361582,0.7627119,0.7457627,0.8418079,0.8926554,0.81920904,0.8418079,0.920904,0.9039548,0.8870056,0.8079096,0.8757062,0.920904,0.8587571,0.92655367,0.9378531,0.920904,0.90960455,0.94350284,0.94350284,0.9661017,0.9491525,0.9548023,0.9661017,0.920904,0.9322034,0.9661017,0.9322034,0.9830508,0.9491525,0.92655367,0.97740114,0.9830508,0.9830508,0.99435025,0.8757062,0.98870057,0.8870056,0.97740114,0.98870057,0.99435025,0.9548023,1.0,0.97740114,0.9717514,0.99435025,1.0,0.98870057,0.920904,0.99435025,0.9830508,1.0,0.99435025,0.9830508,1.0,0.98870057,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2881356,0.26553673,0.39548022,0.42937854,0.50282484,0.56497175,,0.4915254,0.66101694,0.7288136,0.6723164,0.6440678,0.7118644,0.78531075,0.8135593,0.8418079,0.8305085,0.8305085,0.8305085,0.88135594,0.8587571,0.8870056,0.8587571,0.76836157,0.84745765,0.9039548,0.86440676,0.920904,0.8757062,0.91525424,0.9039548,0.8870056,0.9039548,0.92655367,0.92655367,0.92655367,0.9717514,0.920904,0.9548023,0.9548023,0.92655367,0.9322034,0.9830508,0.8757062,0.9548023,0.9830508,0.9661017,0.9717514,0.8700565,0.96045196,0.8757062,0.94350284,0.9661017,0.98870057,0.9378531,0.99435025,0.99435025,0.9661017,0.99435025,1.0,1.0,0.91525424,1.0,0.9717514,1.0,1.0,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.259887,0.26553673,0.4180791,0.42937854,0.4519774,0.47457626,0.5084746,,0.6101695,0.6440678,0.6440678,0.8079096,0.70621467,0.85310733,0.7344633,0.78531075,0.80225986,0.8079096,0.8248588,0.7966102,0.8587571,0.8870056,0.8587571,0.76836157,0.8135593,0.8700565,0.78531075,0.88135594,0.9039548,0.8870056,0.920904,0.8587571,0.90960455,0.9039548,0.9491525,0.92655367,0.9378531,0.85310733,0.90960455,0.94350284,0.8587571,0.9661017,0.9717514,0.8926554,0.96045196,0.97740114,0.9322034,0.9830508,0.8305085,0.94350284,0.84745765,0.9661017,0.9548023,0.9830508,0.94350284,0.97740114,1.0,0.91525424,0.9830508,1.0,1.0,0.89830506,0.98870057,0.96045196,1.0,1.0,0.9717514,1.0,0.9830508,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.12429378,0.2937853,0.2937853,0.33898306,0.27118644,0.38418078,0.33898306,0.3898305,,0.56497175,0.6158192,0.55932206,0.72316384,0.68926555,0.6384181,0.779661,0.79096043,0.80225986,0.72316384,0.86440676,0.8248588,0.7627119,0.8361582,0.7457627,0.86440676,0.8700565,0.8418079,0.9039548,0.88135594,0.86440676,0.8305085,0.9039548,0.8587571,0.9378531,0.8926554,0.88135594,0.9378531,0.9378531,0.94350284,0.8870056,0.92655367,0.91525424,0.920904,0.8926554,0.8757062,0.9491525,0.9548023,0.9717514,0.78531075,0.97740114,0.78531075,0.94350284,0.9830508,0.9830508,0.9548023,0.99435025,1.0,0.9491525,0.9830508,1.0,1.0,0.8926554,1.0,0.97740114,1.0,1.0,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.12429378,0.22033899,0.20903955,0.33333334,0.2542373,0.38418078,0.27118644,0.3559322,0.43502825,,0.5084746,0.47457626,0.5819209,0.6158192,0.72316384,0.65536726,0.70621467,0.779661,0.66101694,0.81920904,0.7740113,0.740113,0.78531075,0.740113,0.8757062,0.85310733,0.8079096,0.8700565,0.8361582,0.8757062,0.84745765,0.91525424,0.8135593,0.92655367,0.8926554,0.8135593,0.9830508,0.92655367,0.92655367,0.85310733,0.9378531,0.89830506,0.88135594,0.89830506,0.8757062,0.920904,0.96045196,0.9830508,0.76836157,0.97740114,0.7740113,0.9661017,0.9830508,0.9717514,0.9717514,0.99435025,1.0,0.96045196,0.9830508,0.99435025,0.99435025,0.8757062,0.99435025,0.96045196,1.0,1.0,0.97740114,1.0,0.98870057,1.0,1.0,1.0
+TabDPT (tuned),0.07909604,0.20903955,0.27118644,0.29943502,0.27683616,0.3502825,0.32768363,0.3559322,0.38418078,0.4915254,,0.53107345,0.6384181,0.5819209,0.6045198,0.6836158,0.75706214,0.7740113,0.6723164,0.8079096,0.7966102,0.700565,0.80225986,0.7457627,0.8305085,0.84745765,0.8248588,0.8700565,0.8418079,0.8700565,0.78531075,0.8870056,0.8305085,0.94350284,0.86440676,0.85310733,0.920904,0.92655367,0.9322034,0.86440676,0.91525424,0.91525424,0.920904,0.9039548,0.8757062,0.9378531,0.9548023,0.97740114,0.7740113,0.96045196,0.779661,0.9322034,0.9661017,0.97740114,0.9491525,0.99435025,0.99435025,0.9548023,0.97740114,1.0,1.0,0.89830506,0.98870057,0.97740114,1.0,1.0,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.23163842,0.18644068,0.2542373,0.33898306,0.31638417,0.3502825,0.3559322,0.19209039,0.44067797,0.5254237,0.46892655,,0.5762712,0.6581921,0.6045198,0.65536726,0.63276833,0.6779661,0.7118644,0.6779661,0.7118644,0.740113,0.700565,0.66101694,0.75706214,0.779661,0.70621467,0.80225986,0.75706214,0.76836157,0.8248588,0.8248588,0.78531075,0.85310733,0.8700565,0.81920904,0.8926554,0.8248588,0.8135593,0.85310733,0.8305085,0.8870056,0.8926554,0.8361582,0.88135594,0.90960455,0.8757062,0.9491525,0.75706214,0.8700565,0.7627119,0.9322034,0.8870056,0.94350284,0.8870056,0.94350284,0.9717514,0.8757062,0.9548023,0.99435025,1.0,0.86440676,0.9548023,0.91525424,1.0,0.99435025,0.9661017,1.0,0.96045196,1.0,1.0,1.0
+TabDPT (default),0.107344635,0.20903955,0.18644068,0.26553673,0.23728813,0.29943502,0.2881356,0.2937853,0.27683616,0.4180791,0.36158192,0.42372882,,0.47457626,0.4858757,0.6101695,0.62711865,0.700565,0.6101695,0.75141245,0.7175141,0.5819209,0.700565,0.70621467,0.76836157,0.75141245,0.7627119,0.78531075,0.7627119,0.779661,0.65536726,0.80225986,0.7627119,0.8757062,0.80225986,0.7740113,0.85310733,0.9322034,0.88135594,0.8135593,0.9378531,0.8248588,0.8757062,0.8700565,0.8305085,0.90960455,0.9548023,0.94350284,0.78531075,0.9661017,0.779661,0.9322034,0.9717514,0.96045196,0.9491525,0.99435025,0.9830508,0.96045196,0.9830508,0.98870057,0.9830508,0.9039548,0.99435025,0.97740114,1.0,1.0,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.084745765,0.1299435,0.15254237,0.14124294,0.15254237,0.16384181,0.21468927,0.14689265,0.31073445,0.38418078,0.4180791,0.3418079,0.5254237,,0.53672314,0.6214689,0.6214689,0.69491524,0.6101695,0.6779661,0.75141245,0.66101694,0.6836158,0.6384181,0.7457627,0.75706214,0.740113,0.79096043,0.779661,0.8079096,0.7175141,0.8305085,0.80225986,0.8418079,0.84745765,0.81920904,0.79096043,0.8305085,0.84745765,0.8587571,0.8418079,0.9039548,0.8926554,0.8418079,0.88135594,0.89830506,0.8926554,0.91525424,0.7740113,0.88135594,0.779661,0.96045196,0.88135594,0.9322034,0.8757062,0.9491525,0.9661017,0.8926554,0.94350284,1.0,0.97740114,0.8700565,0.9322034,0.9039548,1.0,0.99435025,0.9378531,1.0,0.9491525,1.0,1.0,1.0
+RealMLP (tuned),0.06779661,0.1299435,0.16949153,0.20903955,0.17514125,0.23728813,0.18644068,0.26553673,0.36158192,0.27683616,0.39548022,0.39548022,0.5141243,0.46327683,,0.53672314,0.5480226,0.6723164,0.5480226,0.6666667,0.6836158,0.6666667,0.6836158,0.66101694,0.69491524,0.75141245,0.7118644,0.7627119,0.72316384,0.8079096,0.76836157,0.779661,0.75706214,0.8361582,0.8418079,0.80225986,0.8757062,0.8135593,0.8587571,0.8135593,0.84745765,0.85310733,0.86440676,0.79096043,0.8361582,0.920904,0.92655367,0.920904,0.7457627,0.91525424,0.75706214,0.9322034,0.9378531,0.9548023,0.9039548,0.9717514,0.9717514,0.8870056,0.9661017,0.97740114,0.98870057,0.8418079,0.98870057,0.9378531,0.98870057,0.98870057,0.9548023,1.0,0.9717514,0.99435025,0.99435025,0.99435025
+TabDPT-Turbo (default),0.04519774,0.13559322,0.11299435,0.16384181,0.16949153,0.2542373,0.15819208,0.21468927,0.22033899,0.34463277,0.31638417,0.34463277,0.3898305,0.37853107,0.46327683,,0.6045198,0.6723164,0.5706215,0.6779661,0.68926555,0.59887004,0.65536726,0.66101694,0.740113,0.740113,0.7288136,0.7627119,0.7288136,0.78531075,0.68926555,0.7966102,0.75141245,0.8305085,0.8079096,0.7627119,0.84745765,0.8870056,0.84745765,0.79096043,0.89830506,0.8418079,0.8870056,0.84745765,0.8248588,0.90960455,0.90960455,0.92655367,0.7288136,0.92655367,0.740113,0.91525424,0.92655367,0.94350284,0.920904,0.9661017,0.97740114,0.9322034,0.9661017,0.98870057,0.98870057,0.8587571,0.9830508,0.96045196,0.98870057,1.0,0.9661017,1.0,0.9717514,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.050847456,0.09039548,0.14689265,0.107344635,0.14124294,0.15819208,0.16949153,0.19774011,0.20903955,0.2937853,0.24293785,0.36723164,0.37288135,0.37853107,0.4519774,0.39548022,,0.6045198,0.5480226,0.6158192,0.6045198,0.55932206,0.7457627,0.6214689,0.65536726,0.78531075,0.7344633,0.8135593,0.72316384,0.8418079,0.6666667,0.7457627,0.79096043,0.8418079,0.740113,0.7966102,0.7288136,0.8361582,0.84745765,0.8135593,0.84745765,0.88135594,0.91525424,0.81920904,0.84745765,0.90960455,0.920904,0.9548023,0.7457627,0.8926554,0.75706214,0.9039548,0.89830506,0.96045196,0.89830506,0.97740114,0.99435025,0.90960455,0.9548023,0.99435025,1.0,0.84745765,0.9717514,0.920904,0.98870057,1.0,0.9548023,1.0,0.9661017,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.04519774,0.073446326,0.1299435,0.04519774,0.10169491,0.107344635,0.16949153,0.19209039,0.19774011,0.22033899,0.2259887,0.3220339,0.29943502,0.30508474,0.32768363,0.32768363,0.39548022,,0.50282484,0.48022598,0.63276833,0.46892655,0.6214689,0.5536723,0.53107345,0.7457627,0.68926555,0.75141245,0.79096043,0.75141245,0.5762712,0.69491524,0.7457627,0.76836157,0.6158192,0.8135593,0.69491524,0.76836157,0.8135593,0.7966102,0.8079096,0.920904,0.78531075,0.8361582,0.8418079,0.8079096,0.8926554,0.94350284,0.7457627,0.89830506,0.7457627,0.920904,0.9039548,0.96045196,0.86440676,0.99435025,0.920904,0.89830506,0.97740114,0.9322034,0.96045196,0.8361582,0.9830508,0.9322034,1.0,0.99435025,0.96045196,1.0,0.9661017,1.0,1.0,1.0
+TabSwift (default),0.1299435,0.096045196,0.07909604,0.23163842,0.20903955,0.18079096,0.16949153,0.17514125,0.27683616,0.33898306,0.32768363,0.2881356,0.3898305,0.3898305,0.4519774,0.42937854,0.4519774,0.49717513,,0.4915254,0.5536723,0.62711865,0.49717513,0.53672314,0.55932206,0.5423729,0.5819209,0.6158192,0.5762712,0.6158192,0.700565,0.7118644,0.6158192,0.72316384,0.700565,0.65536726,0.8926554,0.7457627,0.78531075,0.700565,0.7740113,0.7288136,0.8361582,0.8135593,0.75706214,0.86440676,0.920904,0.85310733,0.72316384,0.8305085,0.7344633,0.8248588,0.8305085,0.85310733,0.8926554,0.94350284,0.9548023,0.94350284,0.90960455,0.98870057,0.96045196,0.81920904,0.9491525,0.96045196,0.9717514,1.0,0.97740114,0.99435025,0.98870057,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.04519774,0.11299435,0.11299435,0.14689265,0.107344635,0.15819208,0.118644066,0.20338982,0.13559322,0.18079096,0.19209039,0.3220339,0.24858756,0.3220339,0.33333334,0.3220339,0.38418078,0.519774,0.5084746,,0.5480226,0.47457626,0.5762712,0.53672314,0.56497175,0.65536726,0.6214689,0.6779661,0.63276833,0.69491524,0.55932206,0.6779661,0.6779661,0.8135593,0.5536723,0.7175141,0.75141245,0.8248588,0.8361582,0.7175141,0.8361582,0.76836157,0.8305085,0.84745765,0.75706214,0.8305085,0.90960455,0.920904,0.740113,0.92655367,0.740113,0.8248588,0.9322034,0.91525424,0.90960455,0.9661017,0.9830508,0.9378531,0.9491525,0.9717514,0.97740114,0.8305085,0.97740114,0.9548023,0.9830508,0.99435025,0.9661017,0.99435025,0.98870057,0.99435025,0.99435025,0.99435025
+CatBoost (tuned),0.028248588,0.03954802,0.12429378,0.050847456,0.107344635,0.07909604,0.14124294,0.14124294,0.17514125,0.2259887,0.20338982,0.2881356,0.28248587,0.24858756,0.31638417,0.31073445,0.39548022,0.36723164,0.4463277,0.4519774,,0.45762712,0.5536723,0.5536723,0.4915254,0.66101694,0.6440678,0.72316384,0.6779661,0.7118644,0.56497175,0.65536726,0.68926555,0.7288136,0.59887004,0.76836157,0.6779661,0.75141245,0.78531075,0.7457627,0.7627119,0.89830506,0.7627119,0.8305085,0.8248588,0.79096043,0.8757062,0.91525424,0.740113,0.86440676,0.75141245,0.9039548,0.8926554,0.9491525,0.8418079,0.99435025,0.920904,0.8926554,0.9661017,0.92655367,0.9491525,0.8248588,0.9717514,0.920904,0.99435025,0.9830508,0.9548023,1.0,0.96045196,1.0,1.0,1.0
+TabM (tuned + ensembled),0.056497175,0.07909604,0.13559322,0.13559322,0.107344635,0.096045196,0.11299435,0.11299435,0.23728813,0.259887,0.29943502,0.259887,0.4180791,0.33898306,0.33333334,0.40112993,0.44067797,0.53107345,0.37288135,0.5254237,0.5423729,,0.50282484,0.53672314,0.59322035,0.5875706,0.59887004,0.6440678,0.62711865,0.6101695,0.8361582,0.69491524,0.62711865,0.68926555,0.88135594,0.6836158,0.7966102,0.6666667,0.7118644,0.6666667,0.68926555,0.78531075,0.8305085,0.7457627,0.740113,0.88135594,0.8135593,0.8418079,0.72316384,0.75141245,0.7344633,0.8079096,0.7627119,0.8418079,0.76836157,0.9039548,0.92655367,0.8361582,0.86440676,0.96045196,0.97740114,0.8305085,0.920904,0.8870056,0.9717514,0.9548023,0.920904,0.98870057,0.9322034,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.033898305,0.04519774,0.09039548,0.050847456,0.107344635,0.11299435,0.14124294,0.14124294,0.16384181,0.21468927,0.19774011,0.29943502,0.29943502,0.31638417,0.31638417,0.34463277,0.2542373,0.37853107,0.50282484,0.42372882,0.4463277,0.49717513,,0.4124294,0.50282484,0.59887004,0.56497175,0.6666667,0.6836158,0.79096043,0.5819209,0.5819209,0.7118644,0.6384181,0.6666667,0.79096043,0.700565,0.76836157,0.75706214,0.7627119,0.76836157,0.8700565,0.68926555,0.740113,0.7966102,0.7288136,0.84745765,0.97740114,0.7175141,0.86440676,0.72316384,0.92655367,0.8700565,0.9661017,0.8587571,0.9717514,0.9661017,0.8587571,0.9661017,0.8926554,0.98870057,0.8079096,0.9661017,0.9039548,1.0,0.98870057,0.9322034,1.0,0.96045196,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.084745765,0.118644066,0.15819208,0.14124294,0.11299435,0.19209039,0.23163842,0.23163842,0.2542373,0.259887,0.2542373,0.33898306,0.2937853,0.36158192,0.33898306,0.33898306,0.37853107,0.4463277,0.46327683,0.46327683,0.4463277,0.46327683,0.5875706,,0.49717513,0.59322035,0.6384181,0.59887004,0.62711865,0.63276833,0.5536723,0.5536723,0.68926555,0.6384181,0.53672314,0.68926555,0.6497175,0.7740113,0.740113,0.7175141,0.7740113,0.7457627,0.700565,0.8418079,0.7627119,0.68926555,0.8361582,0.85310733,0.7288136,0.8757062,0.740113,0.8361582,0.8757062,0.85310733,0.8757062,0.8926554,0.89830506,0.8700565,0.86440676,0.88135594,0.89830506,0.84745765,0.94350284,0.9491525,0.9830508,0.94350284,0.9830508,0.98870057,0.9830508,0.99435025,0.99435025,1.0
+xRFM (tuned + ensembled),0.06214689,0.084745765,0.107344635,0.096045196,0.12429378,0.12429378,0.15254237,0.18644068,0.13559322,0.12429378,0.16949153,0.24293785,0.23163842,0.2542373,0.30508474,0.259887,0.34463277,0.46892655,0.44067797,0.43502825,0.5084746,0.40677965,0.49717513,0.50282484,,0.5819209,0.6158192,0.65536726,0.59322035,0.6384181,0.49717513,0.6666667,0.6497175,0.72316384,0.5536723,0.68926555,0.59322035,0.70621467,0.76836157,0.75706214,0.7344633,0.7288136,0.7175141,0.8700565,0.76836157,0.7457627,0.8079096,0.85310733,0.70621467,0.7966102,0.7118644,0.79096043,0.8079096,0.80225986,0.9039548,0.9322034,0.84745765,0.9661017,0.8135593,0.920904,0.88135594,0.7740113,0.9717514,0.96045196,0.94350284,0.9661017,0.97740114,0.97740114,0.97740114,0.99435025,0.99435025,0.99435025
+XGBoost (tuned + ensembled),0.04519774,0.06214689,0.09039548,0.07909604,0.096045196,0.07909604,0.096045196,0.1299435,0.1299435,0.14689265,0.15254237,0.22033899,0.24858756,0.24293785,0.24858756,0.259887,0.21468927,0.2542373,0.45762712,0.34463277,0.33898306,0.4124294,0.40112993,0.40677965,0.4180791,,0.49717513,0.6666667,0.59322035,0.6214689,0.5706215,0.5875706,0.7118644,0.6101695,0.5875706,0.740113,0.6836158,0.7288136,0.7344633,0.7457627,0.75141245,0.8757062,0.6779661,0.75141245,0.779661,0.7344633,0.8305085,0.9717514,0.7118644,0.8418079,0.7288136,0.9039548,0.86440676,0.9378531,0.8248588,0.98870057,0.9322034,0.8587571,0.96045196,0.88135594,0.97740114,0.7966102,0.97740114,0.90960455,0.9830508,0.98870057,0.9548023,1.0,0.9548023,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.050847456,0.096045196,0.1299435,0.118644066,0.13559322,0.14124294,0.13559322,0.21468927,0.15819208,0.19209039,0.17514125,0.2937853,0.23728813,0.259887,0.2881356,0.27118644,0.26553673,0.31073445,0.4180791,0.37853107,0.3559322,0.40112993,0.43502825,0.36158192,0.38418078,0.50282484,,0.5141243,0.5480226,0.519774,0.48022598,0.4915254,0.5706215,0.53107345,0.5141243,0.62711865,0.65536726,0.68926555,0.8418079,0.6384181,0.7118644,0.66101694,0.6836158,0.6779661,0.6836158,0.6836158,0.84745765,0.75706214,0.70621467,0.81920904,0.72316384,0.740113,0.85310733,0.8361582,0.8700565,0.8870056,0.90960455,0.9039548,0.8700565,0.8870056,0.920904,0.7740113,0.9548023,0.94350284,0.94350284,0.97740114,0.9717514,1.0,0.97740114,0.99435025,0.99435025,1.0
+XGBoost (tuned),0.033898305,0.050847456,0.096045196,0.06214689,0.107344635,0.073446326,0.07909604,0.118644066,0.096045196,0.1299435,0.1299435,0.19774011,0.21468927,0.20903955,0.23728813,0.23728813,0.18644068,0.24858756,0.38418078,0.3220339,0.27683616,0.3559322,0.33333334,0.40112993,0.34463277,0.33333334,0.4858757,,0.5423729,0.56497175,0.519774,0.53672314,0.6214689,0.5480226,0.53672314,0.66101694,0.6497175,0.69491524,0.68926555,0.69491524,0.7118644,0.8248588,0.6723164,0.740113,0.7740113,0.72316384,0.7966102,0.920904,0.7118644,0.78531075,0.7118644,0.8700565,0.7966102,0.92655367,0.8079096,0.96045196,0.920904,0.86440676,0.920904,0.8926554,0.96045196,0.7966102,0.9548023,0.8926554,0.9830508,0.97740114,0.94350284,1.0,0.9548023,1.0,1.0,1.0
+CatBoost (default),0.03954802,0.03954802,0.107344635,0.050847456,0.096045196,0.06214689,0.12429378,0.096045196,0.118644066,0.16384181,0.15819208,0.24293785,0.23728813,0.22033899,0.27683616,0.27118644,0.27683616,0.20903955,0.42372882,0.36723164,0.3220339,0.37288135,0.31638417,0.37288135,0.40677965,0.40677965,0.4519774,0.45762712,,0.47457626,0.4915254,0.55932206,0.6384181,0.5706215,0.5536723,0.6779661,0.65536726,0.70621467,0.63276833,0.75141245,0.7175141,0.8870056,0.6384181,0.63276833,0.80225986,0.6779661,0.69491524,0.8079096,0.72316384,0.8079096,0.7288136,0.8926554,0.8248588,0.92655367,0.7966102,0.8248588,0.89830506,0.7740113,0.94350284,0.80225986,0.91525424,0.8248588,0.92655367,0.86440676,1.0,0.9830508,0.9491525,1.0,0.96045196,1.0,1.0,1.0
+LightGBM (tuned),0.028248588,0.033898305,0.06214689,0.050847456,0.084745765,0.07909604,0.084745765,0.11299435,0.13559322,0.12429378,0.1299435,0.23163842,0.22033899,0.19209039,0.19209039,0.21468927,0.15819208,0.24858756,0.38418078,0.30508474,0.2881356,0.3898305,0.20903955,0.36723164,0.36158192,0.37853107,0.48022598,0.43502825,0.5254237,,0.5254237,0.46892655,0.5480226,0.49717513,0.5762712,0.6045198,0.63276833,0.68926555,0.6666667,0.6836158,0.69491524,0.740113,0.6384181,0.6836158,0.7627119,0.6666667,0.7627119,0.8926554,0.70621467,0.78531075,0.7175141,0.81920904,0.7966102,0.8587571,0.78531075,0.9322034,0.89830506,0.85310733,0.8870056,0.85310733,0.94350284,0.78531075,0.9378531,0.8870056,0.9830508,0.9717514,0.92655367,1.0,0.9322034,0.98870057,0.98870057,0.99435025
+TabM (tuned),0.03954802,0.056497175,0.096045196,0.096045196,0.10169491,0.09039548,0.096045196,0.07909604,0.16949153,0.15254237,0.21468927,0.17514125,0.34463277,0.28248587,0.23163842,0.31073445,0.33333334,0.42372882,0.29943502,0.44067797,0.43502825,0.16384181,0.4180791,0.4463277,0.50282484,0.42937854,0.519774,0.48022598,0.5084746,0.47457626,,0.5819209,0.5254237,0.59887004,0.6497175,0.519774,0.68926555,0.6158192,0.6045198,0.5819209,0.62711865,0.6384181,0.7288136,0.6779661,0.6158192,0.78531075,0.72316384,0.7288136,0.700565,0.66101694,0.7118644,0.7288136,0.68926555,0.75141245,0.69491524,0.81920904,0.8361582,0.779661,0.76836157,0.92655367,0.91525424,0.7966102,0.8587571,0.85310733,0.91525424,0.8926554,0.8926554,0.97740114,0.920904,0.9661017,0.9661017,0.99435025
+xRFM (tuned),0.03954802,0.056497175,0.09039548,0.050847456,0.107344635,0.056497175,0.11299435,0.14124294,0.096045196,0.084745765,0.11299435,0.17514125,0.19774011,0.16949153,0.22033899,0.20338982,0.2542373,0.30508474,0.2881356,0.3220339,0.34463277,0.30508474,0.4180791,0.4463277,0.33333334,0.4124294,0.5084746,0.46327683,0.44067797,0.53107345,0.4180791,,0.48022598,0.59322035,0.46327683,0.59322035,0.5706215,0.5875706,0.68926555,0.69491524,0.6384181,0.6497175,0.6836158,0.8079096,0.7118644,0.6723164,0.76836157,0.76836157,0.6836158,0.740113,0.68926555,0.740113,0.7344633,0.7457627,0.79096043,0.8926554,0.8079096,0.8926554,0.7740113,0.9039548,0.85310733,0.75141245,0.9491525,0.9378531,0.920904,0.9378531,0.9548023,0.98870057,0.9661017,0.9830508,0.9830508,1.0
+ChimeraBoost (tuned + ensembled),0.033898305,0.02259887,0.10169491,0.03954802,0.11299435,0.056497175,0.096045196,0.09039548,0.14124294,0.18644068,0.16949153,0.21468927,0.23728813,0.19774011,0.24293785,0.24858756,0.20903955,0.2542373,0.38418078,0.3220339,0.31073445,0.37288135,0.2881356,0.31073445,0.3502825,0.2881356,0.42937854,0.37853107,0.36158192,0.4519774,0.47457626,0.519774,,0.5084746,0.5819209,0.59322035,0.63276833,0.59322035,0.55932206,0.7344633,0.6158192,0.8305085,0.65536726,0.5875706,0.7966102,0.63276833,0.63276833,0.72316384,0.7118644,0.7457627,0.7344633,0.9039548,0.76836157,0.91525424,0.7288136,0.7627119,0.8248588,0.70621467,0.90960455,0.79096043,0.8305085,0.81920904,0.92655367,0.85310733,0.98870057,0.9717514,0.9322034,1.0,0.9491525,0.99435025,0.99435025,0.99435025
+iLTM (tuned),0.028248588,0.02259887,0.033898305,0.02259887,0.073446326,0.033898305,0.073446326,0.096045196,0.06214689,0.073446326,0.056497175,0.14689265,0.12429378,0.15819208,0.16384181,0.16949153,0.15819208,0.23163842,0.27683616,0.18644068,0.27118644,0.31073445,0.36158192,0.36158192,0.27683616,0.3898305,0.46892655,0.4519774,0.42937854,0.50282484,0.40112993,0.40677965,0.4915254,,0.42937854,0.5536723,0.5254237,0.62711865,0.6384181,0.62711865,0.65536726,0.6101695,0.59887004,0.7288136,0.6836158,0.66101694,0.7457627,0.75141245,0.6666667,0.72316384,0.66101694,0.70621467,0.72316384,0.740113,0.8079096,0.88135594,0.8361582,0.8757062,0.779661,0.85310733,0.8587571,0.7344633,0.920904,0.88135594,0.91525424,0.9548023,0.9322034,0.9830508,0.92655367,1.0,1.0,1.0
+TabM (default),0.028248588,0.033898305,0.07909604,0.050847456,0.096045196,0.050847456,0.073446326,0.050847456,0.107344635,0.107344635,0.13559322,0.1299435,0.19774011,0.15254237,0.15819208,0.19209039,0.259887,0.38418078,0.29943502,0.4463277,0.40112993,0.118644066,0.33333334,0.46327683,0.4463277,0.4124294,0.4858757,0.46327683,0.4463277,0.42372882,0.3502825,0.53672314,0.4180791,0.5706215,,0.4519774,0.5536723,0.62711865,0.5762712,0.53672314,0.6384181,0.5480226,0.700565,0.65536726,0.6045198,0.7740113,0.7118644,0.6836158,0.69491524,0.6214689,0.70621467,0.7118644,0.62711865,0.70621467,0.6440678,0.7966102,0.80225986,0.7627119,0.72316384,0.8926554,0.89830506,0.79096043,0.8135593,0.7966102,0.90960455,0.88135594,0.84745765,0.9830508,0.8587571,0.97740114,0.97740114,0.99435025
+ChimeraBoost (tuned),0.03954802,0.016949153,0.096045196,0.028248588,0.107344635,0.04519774,0.073446326,0.073446326,0.118644066,0.18644068,0.14689265,0.18079096,0.2259887,0.18079096,0.19774011,0.23728813,0.20338982,0.18644068,0.34463277,0.28248587,0.23163842,0.31638417,0.20903955,0.31073445,0.31073445,0.259887,0.37288135,0.33898306,0.3220339,0.39548022,0.48022598,0.40677965,0.40677965,0.4463277,0.5480226,,0.6101695,0.5762712,0.5423729,0.6666667,0.59322035,0.75141245,0.6101695,0.5480226,0.75141245,0.59887004,0.6101695,0.65536726,0.7175141,0.69491524,0.72316384,0.8248588,0.7344633,0.8700565,0.7175141,0.72316384,0.79096043,0.69491524,0.8757062,0.7740113,0.8135593,0.79096043,0.8587571,0.8418079,0.9830508,0.92655367,0.9039548,1.0,0.92655367,0.98870057,0.9830508,0.98870057
+RealMLP (default),0.011299435,0.033898305,0.03954802,0.07909604,0.06779661,0.033898305,0.028248588,0.06214689,0.06214689,0.016949153,0.07909604,0.107344635,0.14689265,0.20903955,0.12429378,0.15254237,0.27118644,0.30508474,0.107344635,0.24858756,0.3220339,0.20338982,0.29943502,0.3502825,0.40677965,0.31638417,0.34463277,0.3502825,0.34463277,0.36723164,0.31073445,0.42937854,0.36723164,0.47457626,0.4463277,0.3898305,,0.6158192,0.50282484,0.50282484,0.62711865,0.4858757,0.59887004,0.6158192,0.5423729,0.6779661,0.6666667,0.6101695,0.6440678,0.700565,0.65536726,0.6101695,0.7175141,0.6045198,0.72316384,0.7457627,0.79096043,0.79096043,0.6836158,0.8926554,0.84745765,0.779661,0.8361582,0.8926554,0.8361582,0.92655367,0.9491525,0.9548023,0.9661017,0.9661017,0.9661017,1.0
+ExtraTrees (tuned + ensembled),0.033898305,0.04519774,0.06779661,0.06214689,0.10169491,0.07909604,0.07909604,0.14689265,0.06214689,0.073446326,0.073446326,0.17514125,0.06779661,0.16949153,0.18644068,0.11299435,0.16384181,0.23163842,0.2542373,0.17514125,0.24858756,0.33333334,0.23163842,0.2259887,0.2937853,0.27118644,0.31073445,0.30508474,0.2937853,0.31073445,0.38418078,0.4124294,0.40677965,0.37288135,0.37288135,0.42372882,0.38418078,,0.49717513,0.5536723,0.65254235,0.5141243,0.42372882,0.5423729,0.59322035,0.4519774,0.5141243,0.46327683,0.6101695,0.8079096,0.6158192,0.65536726,0.86440676,0.6497175,0.94350284,0.5819209,0.7288136,0.75706214,0.7118644,0.6384181,0.7175141,0.75706214,0.9548023,0.9661017,0.81920904,0.920904,0.98870057,0.99435025,0.98870057,1.0,1.0,1.0
+ModernNCA (tuned),0.011299435,0.050847456,0.07909604,0.056497175,0.07909604,0.06779661,0.04519774,0.09039548,0.056497175,0.073446326,0.06779661,0.18644068,0.118644066,0.15254237,0.14124294,0.15254237,0.15254237,0.18644068,0.21468927,0.16384181,0.21468927,0.2881356,0.24293785,0.259887,0.23163842,0.26553673,0.15819208,0.31073445,0.36723164,0.33333334,0.39548022,0.31073445,0.44067797,0.36158192,0.42372882,0.45762712,0.49717513,0.50282484,,0.5480226,0.519774,0.50282484,0.519774,0.5141243,0.5762712,0.5706215,0.70621467,0.5762712,0.6723164,0.68926555,0.6779661,0.63276833,0.68926555,0.6497175,0.72316384,0.7118644,0.8135593,0.8079096,0.7175141,0.7627119,0.8587571,0.72316384,0.8870056,0.9039548,0.85310733,0.9322034,0.94350284,0.9548023,0.9548023,0.97740114,0.97740114,1.0
+EBM (tuned + ensembled),0.033898305,0.016949153,0.073446326,0.02259887,0.06214689,0.033898305,0.04519774,0.056497175,0.11299435,0.14689265,0.13559322,0.14689265,0.18644068,0.14124294,0.18644068,0.20903955,0.18644068,0.20338982,0.29943502,0.28248587,0.2542373,0.33333334,0.23728813,0.28248587,0.24293785,0.2542373,0.36158192,0.30508474,0.24858756,0.31638417,0.4180791,0.30508474,0.26553673,0.37288135,0.46327683,0.33333334,0.49717513,0.4463277,0.4519774,,0.47457626,0.4180791,0.56497175,0.50282484,0.75706214,0.6045198,0.5480226,0.50282484,0.6497175,0.5480226,0.6666667,0.8248588,0.59322035,0.7457627,0.5706215,0.62711865,0.6497175,0.6045198,0.76836157,0.7175141,0.7288136,0.8079096,0.740113,0.8248588,0.9491525,0.9039548,0.8418079,0.90960455,0.8870056,0.9830508,0.9830508,0.99435025
+ExtraTrees (tuned),0.028248588,0.06214689,0.056497175,0.06214689,0.107344635,0.06779661,0.073446326,0.14124294,0.073446326,0.06214689,0.084745765,0.16949153,0.06214689,0.15819208,0.15254237,0.10169491,0.15254237,0.19209039,0.2259887,0.16384181,0.23728813,0.31073445,0.23163842,0.2259887,0.26553673,0.24858756,0.2881356,0.2881356,0.28248587,0.30508474,0.37288135,0.36158192,0.38418078,0.34463277,0.36158192,0.40677965,0.37288135,0.34745762,0.48022598,0.5254237,,0.4915254,0.40112993,0.53672314,0.56497175,0.44067797,0.5084746,0.44067797,0.6045198,0.75141245,0.6158192,0.63276833,0.8079096,0.6384181,0.8870056,0.53107345,0.6836158,0.75141245,0.69491524,0.6158192,0.70621467,0.75141245,0.9378531,0.96045196,0.8305085,0.9039548,0.99435025,0.99435025,0.98870057,1.0,1.0,1.0
+ChimeraBoost (default),0.0056497175,0.02259887,0.050847456,0.016949153,0.06214689,0.016949153,0.06779661,0.033898305,0.084745765,0.10169491,0.084745765,0.11299435,0.17514125,0.096045196,0.14689265,0.15819208,0.118644066,0.07909604,0.27118644,0.23163842,0.10169491,0.21468927,0.1299435,0.2542373,0.27118644,0.12429378,0.33898306,0.17514125,0.11299435,0.259887,0.36158192,0.3502825,0.16949153,0.3898305,0.4519774,0.24858756,0.5141243,0.4858757,0.49717513,0.5819209,0.5084746,,0.5875706,0.53107345,0.65536726,0.6101695,0.5480226,0.5536723,0.68926555,0.69491524,0.69491524,0.72316384,0.69491524,0.779661,0.70621467,0.70621467,0.69491524,0.6497175,0.8248588,0.740113,0.7344633,0.779661,0.84745765,0.8361582,0.9717514,0.91525424,0.8870056,0.98870057,0.9039548,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.011299435,0.033898305,0.033898305,0.04519774,0.056497175,0.050847456,0.016949153,0.028248588,0.07909604,0.118644066,0.07909604,0.107344635,0.12429378,0.107344635,0.13559322,0.11299435,0.084745765,0.21468927,0.16384181,0.16949153,0.23728813,0.16949153,0.31073445,0.29943502,0.28248587,0.3220339,0.31638417,0.32768363,0.36158192,0.36158192,0.27118644,0.31638417,0.34463277,0.40112993,0.29943502,0.3898305,0.40112993,0.5762712,0.48022598,0.43502825,0.59887004,0.4124294,,0.5084746,0.5084746,0.65536726,0.6214689,0.56497175,0.6666667,0.63276833,0.66101694,0.5762712,0.6440678,0.55932206,0.6440678,0.6723164,0.7740113,0.740113,0.6440678,0.92655367,0.8700565,0.79096043,0.76836157,0.8248588,0.85310733,0.8700565,0.8700565,0.88135594,0.88135594,0.90960455,0.90960455,1.0
+LimiX (default),0.016949153,0.02259887,0.09039548,0.06779661,0.084745765,0.073446326,0.12429378,0.107344635,0.107344635,0.10169491,0.096045196,0.16384181,0.1299435,0.15819208,0.20903955,0.15254237,0.18079096,0.16384181,0.18644068,0.15254237,0.16949153,0.2542373,0.259887,0.15819208,0.1299435,0.24858756,0.3220339,0.259887,0.36723164,0.31638417,0.3220339,0.19209039,0.4124294,0.27118644,0.34463277,0.4519774,0.38418078,0.45762712,0.4858757,0.49717513,0.46327683,0.46892655,0.4915254,,0.5423729,0.50282484,0.53672314,0.519774,0.66101694,0.5254237,0.66101694,0.5536723,0.5536723,0.59322035,0.5254237,0.59887004,0.6497175,0.66101694,0.59887004,0.6214689,0.6836158,0.70621467,0.66101694,0.76836157,0.7457627,0.6836158,0.8361582,0.7627119,0.9378531,0.91525424,0.91525424,0.9830508
+EBM (tuned),0.02259887,0.016949153,0.056497175,0.033898305,0.06779661,0.02259887,0.04519774,0.03954802,0.12429378,0.12429378,0.12429378,0.118644066,0.16949153,0.118644066,0.16384181,0.17514125,0.15254237,0.15819208,0.24293785,0.24293785,0.17514125,0.259887,0.20338982,0.23728813,0.23163842,0.22033899,0.31638417,0.2259887,0.19774011,0.23728813,0.38418078,0.2881356,0.20338982,0.31638417,0.39548022,0.24858756,0.45762712,0.40677965,0.42372882,0.24293785,0.43502825,0.34463277,0.4915254,0.45762712,,0.56497175,0.5254237,0.42937854,0.63276833,0.50282484,0.6440678,0.66101694,0.55932206,0.6779661,0.5423729,0.5762712,0.59322035,0.5875706,0.72316384,0.700565,0.68926555,0.7966102,0.700565,0.8135593,0.9322034,0.8700565,0.8361582,0.9039548,0.8587571,0.97740114,0.97740114,0.99435025
+TorchMLP (tuned),0.0056497175,0.016949153,0.028248588,0.016949153,0.073446326,0.016949153,0.016949153,0.02259887,0.050847456,0.07909604,0.06214689,0.09039548,0.09039548,0.10169491,0.07909604,0.09039548,0.09039548,0.19209039,0.13559322,0.16949153,0.20903955,0.118644066,0.27118644,0.31073445,0.2542373,0.26553673,0.31638417,0.27683616,0.3220339,0.33333334,0.21468927,0.32768363,0.36723164,0.33898306,0.2259887,0.40112993,0.3220339,0.5480226,0.42937854,0.39548022,0.55932206,0.3898305,0.34463277,0.49717513,0.43502825,,0.5706215,0.5084746,0.66101694,0.6045198,0.66101694,0.5423729,0.6158192,0.53672314,0.6214689,0.59887004,0.65536726,0.700565,0.6101695,0.84745765,0.7740113,0.78531075,0.72316384,0.7966102,0.79096043,0.80225986,0.8587571,0.8757062,0.85310733,0.8757062,0.8757062,1.0
+ModernNCA (default),0.011299435,0.02259887,0.03954802,0.028248588,0.073446326,0.016949153,0.033898305,0.06779661,0.04519774,0.03954802,0.04519774,0.12429378,0.04519774,0.107344635,0.073446326,0.09039548,0.07909604,0.107344635,0.07909604,0.09039548,0.12429378,0.18644068,0.15254237,0.16384181,0.19209039,0.16949153,0.15254237,0.20338982,0.30508474,0.23728813,0.27683616,0.23163842,0.36723164,0.2542373,0.2881356,0.3898305,0.33333334,0.4858757,0.2937853,0.4519774,0.4915254,0.4519774,0.37853107,0.46327683,0.47457626,0.42937854,,0.42937854,0.6440678,0.6384181,0.6440678,0.5480226,0.6779661,0.5480226,0.740113,0.6158192,0.7175141,0.779661,0.6440678,0.69491524,0.779661,0.740113,0.9039548,0.89830506,0.81920904,0.89830506,0.92655367,0.9491525,0.9548023,0.9491525,0.9491525,1.0
+LightGBM (default),0.0,0.0056497175,0.011299435,0.0,0.050847456,0.0056497175,0.028248588,0.016949153,0.028248588,0.016949153,0.02259887,0.050847456,0.056497175,0.084745765,0.07909604,0.073446326,0.04519774,0.056497175,0.14689265,0.07909604,0.084745765,0.15819208,0.02259887,0.14689265,0.14689265,0.028248588,0.24293785,0.07909604,0.19209039,0.107344635,0.27118644,0.23163842,0.27683616,0.24858756,0.31638417,0.34463277,0.3898305,0.53672314,0.42372882,0.49717513,0.55932206,0.4463277,0.43502825,0.48022598,0.5706215,0.4915254,0.5706215,,0.6836158,0.59322035,0.6836158,0.6384181,0.65536726,0.63276833,0.700565,0.75706214,0.72316384,0.7457627,0.68926555,0.7457627,0.79096043,0.72316384,0.85310733,0.81920904,0.91525424,0.90960455,0.8757062,0.98870057,0.89830506,0.9830508,0.9830508,0.99435025
+TabSTAR (tuned + ensembled),0.096045196,0.107344635,0.17514125,0.12429378,0.16384181,0.12429378,0.1299435,0.16949153,0.21468927,0.23163842,0.2259887,0.24293785,0.21468927,0.2259887,0.2542373,0.27118644,0.2542373,0.2542373,0.27683616,0.259887,0.259887,0.27683616,0.28248587,0.27118644,0.2937853,0.2881356,0.2937853,0.2881356,0.27683616,0.2937853,0.29943502,0.31638417,0.2881356,0.33333334,0.30508474,0.28248587,0.3559322,0.3898305,0.32768363,0.3502825,0.39548022,0.31073445,0.33333334,0.33898306,0.36723164,0.33898306,0.3559322,0.31638417,,0.45762712,0.59887004,0.42937854,0.5254237,0.5084746,0.519774,0.36723164,0.45762712,0.45762712,0.55932206,0.4463277,0.45762712,0.97740114,0.5254237,0.72316384,0.59887004,0.68926555,0.7966102,0.65536726,0.8361582,0.78531075,0.78531075,0.8926554
+RandomForest (tuned + ensembled),0.0056497175,0.016949153,0.028248588,0.011299435,0.03954802,0.011299435,0.03954802,0.056497175,0.02259887,0.02259887,0.03954802,0.1299435,0.033898305,0.118644066,0.084745765,0.073446326,0.107344635,0.10169491,0.16949153,0.073446326,0.13559322,0.24858756,0.13559322,0.12429378,0.20338982,0.15819208,0.18079096,0.21468927,0.19209039,0.21468927,0.33898306,0.259887,0.2542373,0.27683616,0.37853107,0.30508474,0.29943502,0.19209039,0.31073445,0.4519774,0.24858756,0.30508474,0.36723164,0.47457626,0.49717513,0.39548022,0.36158192,0.40677965,0.5423729,,0.55932206,0.5762712,0.7033898,0.5875706,0.740113,0.5254237,0.53672314,0.6101695,0.6384181,0.5254237,0.59887004,0.700565,0.98870057,0.9039548,0.78531075,0.88135594,0.9378531,0.96045196,0.96045196,1.0,1.0,1.0
+TabSTAR (tuned),0.09039548,0.096045196,0.15254237,0.12429378,0.14689265,0.11299435,0.12429378,0.15254237,0.21468927,0.2259887,0.22033899,0.23728813,0.22033899,0.22033899,0.24293785,0.259887,0.24293785,0.2542373,0.26553673,0.259887,0.24858756,0.26553673,0.27683616,0.259887,0.2881356,0.27118644,0.27683616,0.2881356,0.27118644,0.28248587,0.2881356,0.31073445,0.26553673,0.33898306,0.2937853,0.27683616,0.34463277,0.38418078,0.3220339,0.33333334,0.38418078,0.30508474,0.33898306,0.33898306,0.3559322,0.33898306,0.3559322,0.31638417,0.40112993,0.44067797,,0.43502825,0.50282484,0.5084746,0.4915254,0.36723164,0.45762712,0.45762712,0.55932206,0.4463277,0.4519774,0.96045196,0.50282484,0.72316384,0.59887004,0.68926555,0.79096043,0.65536726,0.8361582,0.78531075,0.78531075,0.8700565
+EBM (default),0.016949153,0.011299435,0.033898305,0.016949153,0.02259887,0.02259887,0.056497175,0.033898305,0.056497175,0.033898305,0.06779661,0.06779661,0.06779661,0.03954802,0.06779661,0.084745765,0.096045196,0.07909604,0.17514125,0.17514125,0.096045196,0.19209039,0.073446326,0.16384181,0.20903955,0.096045196,0.259887,0.1299435,0.107344635,0.18079096,0.27118644,0.259887,0.096045196,0.2937853,0.2881356,0.17514125,0.3898305,0.34463277,0.36723164,0.17514125,0.36723164,0.27683616,0.42372882,0.4463277,0.33898306,0.45762712,0.4519774,0.36158192,0.5706215,0.42372882,0.56497175,,0.46892655,0.55932206,0.519774,0.53672314,0.5536723,0.519774,0.6101695,0.6779661,0.6158192,0.7288136,0.6497175,0.8079096,0.8870056,0.8248588,0.8305085,0.88135594,0.8757062,0.9830508,0.9830508,0.98870057
+RandomForest (tuned),0.0056497175,0.016949153,0.028248588,0.011299435,0.028248588,0.011299435,0.033898305,0.04519774,0.016949153,0.016949153,0.033898305,0.11299435,0.028248588,0.118644066,0.06214689,0.073446326,0.10169491,0.096045196,0.16949153,0.06779661,0.107344635,0.23728813,0.1299435,0.12429378,0.19209039,0.13559322,0.14689265,0.20338982,0.17514125,0.20338982,0.31073445,0.26553673,0.23163842,0.27683616,0.37288135,0.26553673,0.28248587,0.13559322,0.31073445,0.40677965,0.19209039,0.30508474,0.3559322,0.4463277,0.44067797,0.38418078,0.3220339,0.34463277,0.47457626,0.29661018,0.49717513,0.53107345,,0.55932206,0.6214689,0.47457626,0.5084746,0.5706215,0.6101695,0.5084746,0.55932206,0.69491524,0.79096043,0.8700565,0.7627119,0.88135594,0.9378531,0.96045196,0.9491525,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.011299435,0.011299435,0.011299435,0.011299435,0.033898305,0.0056497175,0.011299435,0.016949153,0.016949153,0.028248588,0.02259887,0.056497175,0.03954802,0.06779661,0.04519774,0.056497175,0.03954802,0.03954802,0.14689265,0.084745765,0.050847456,0.15819208,0.033898305,0.14689265,0.19774011,0.06214689,0.16384181,0.073446326,0.073446326,0.14124294,0.24858756,0.2542373,0.084745765,0.259887,0.2937853,0.1299435,0.39548022,0.3502825,0.3502825,0.2542373,0.36158192,0.22033899,0.44067797,0.40677965,0.3220339,0.46327683,0.4519774,0.36723164,0.4915254,0.4124294,0.4915254,0.44067797,0.44067797,,0.50282484,0.5536723,0.6101695,0.59322035,0.76836157,0.6101695,0.6440678,0.5875706,0.6440678,0.7175141,0.9548023,0.8248588,0.8700565,0.85310733,0.8926554,0.97740114,0.97740114,0.99435025
+ExtraTrees (default),0.02259887,0.03954802,0.03954802,0.028248588,0.050847456,0.04519774,0.06214689,0.056497175,0.04519774,0.028248588,0.050847456,0.11299435,0.050847456,0.12429378,0.096045196,0.07909604,0.10169491,0.13559322,0.107344635,0.09039548,0.15819208,0.23163842,0.14124294,0.12429378,0.096045196,0.17514125,0.1299435,0.19209039,0.20338982,0.21468927,0.30508474,0.20903955,0.27118644,0.19209039,0.3559322,0.28248587,0.27683616,0.056497175,0.27683616,0.42937854,0.11299435,0.2937853,0.3559322,0.47457626,0.45762712,0.37853107,0.259887,0.29943502,0.48022598,0.259887,0.5084746,0.48022598,0.37853107,0.49717513,,0.37288135,0.44067797,0.5875706,0.5084746,0.40677965,0.43502825,0.6723164,0.68926555,0.91525424,0.6440678,0.72316384,0.9548023,0.8587571,0.97740114,0.99435025,0.99435025,1.0
+XGBoost (default),0.0,0.0056497175,0.011299435,0.0056497175,0.050847456,0.0,0.0056497175,0.02259887,0.0056497175,0.0056497175,0.0056497175,0.056497175,0.0056497175,0.050847456,0.028248588,0.033898305,0.02259887,0.0056497175,0.056497175,0.033898305,0.0056497175,0.096045196,0.028248588,0.107344635,0.06779661,0.011299435,0.11299435,0.03954802,0.17514125,0.06779661,0.18079096,0.107344635,0.23728813,0.118644066,0.20338982,0.27683616,0.2542373,0.4180791,0.2881356,0.37288135,0.46892655,0.2937853,0.32768363,0.40112993,0.42372882,0.40112993,0.38418078,0.24293785,0.63276833,0.47457626,0.63276833,0.46327683,0.5254237,0.4463277,0.62711865,,0.55932206,0.72316384,0.55932206,0.5706215,0.6666667,0.6836158,0.80225986,0.80225986,0.75141245,0.76836157,0.8757062,0.94350284,0.89830506,0.9378531,0.9378531,0.9830508
+FastaiMLP (tuned + ensembled),0.0,0.0056497175,0.0,0.016949153,0.050847456,0.02259887,0.0056497175,0.0,0.0,0.0,0.0056497175,0.028248588,0.016949153,0.033898305,0.028248588,0.02259887,0.0056497175,0.07909604,0.04519774,0.016949153,0.07909604,0.073446326,0.033898305,0.10169491,0.15254237,0.06779661,0.09039548,0.07909604,0.10169491,0.10169491,0.16384181,0.19209039,0.17514125,0.16384181,0.19774011,0.20903955,0.20903955,0.27118644,0.18644068,0.3502825,0.31638417,0.30508474,0.2259887,0.3502825,0.40677965,0.34463277,0.28248587,0.27683616,0.5423729,0.46327683,0.5423729,0.4463277,0.4915254,0.3898305,0.55932206,0.44067797,,0.6101695,0.4463277,0.5875706,0.5762712,0.66101694,0.6836158,0.81920904,0.69491524,0.9378531,0.8587571,0.920904,0.8587571,0.9717514,0.9717514,1.0
+xRFM (default),0.011299435,0.033898305,0.04519774,0.03954802,0.033898305,0.028248588,0.033898305,0.084745765,0.050847456,0.03954802,0.04519774,0.12429378,0.03954802,0.107344635,0.11299435,0.06779661,0.09039548,0.10169491,0.056497175,0.06214689,0.107344635,0.16384181,0.14124294,0.1299435,0.033898305,0.14124294,0.096045196,0.13559322,0.2259887,0.14689265,0.22033899,0.107344635,0.2937853,0.12429378,0.23728813,0.30508474,0.20903955,0.24293785,0.19209039,0.39548022,0.24858756,0.3502825,0.259887,0.33898306,0.4124294,0.29943502,0.22033899,0.2542373,0.5423729,0.3898305,0.5423729,0.48022598,0.42937854,0.40677965,0.4124294,0.27683616,0.3898305,,0.42372882,0.40112993,0.4463277,0.6497175,0.56497175,0.8418079,0.5819209,0.65536726,0.9039548,0.779661,0.9717514,0.9039548,0.8926554,0.98870057
+PerpetualBooster (tuned),0.0056497175,0.0056497175,0.011299435,0.011299435,0.033898305,0.0056497175,0.0056497175,0.016949153,0.016949153,0.016949153,0.02259887,0.04519774,0.016949153,0.056497175,0.033898305,0.033898305,0.04519774,0.02259887,0.09039548,0.050847456,0.033898305,0.13559322,0.033898305,0.13559322,0.18644068,0.03954802,0.1299435,0.07909604,0.056497175,0.11299435,0.23163842,0.2259887,0.09039548,0.22033899,0.27683616,0.12429378,0.31638417,0.2881356,0.28248587,0.23163842,0.30508474,0.17514125,0.3559322,0.40112993,0.27683616,0.3898305,0.3559322,0.31073445,0.44067797,0.36158192,0.44067797,0.3898305,0.3898305,0.23163842,0.4915254,0.44067797,0.5536723,0.5762712,,0.5706215,0.59322035,0.5480226,0.6158192,0.66101694,0.7881356,0.740113,0.84745765,0.8079096,0.8700565,0.9717514,0.9717514,0.98870057
+TorchMLP (default),0.0,0.0,0.0,0.0,0.050847456,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0056497175,0.011299435,0.0,0.02259887,0.011299435,0.0056497175,0.06779661,0.011299435,0.028248588,0.073446326,0.03954802,0.107344635,0.118644066,0.07909604,0.118644066,0.11299435,0.107344635,0.19774011,0.14689265,0.073446326,0.096045196,0.20903955,0.14689265,0.107344635,0.2259887,0.107344635,0.36158192,0.23728813,0.28248587,0.38418078,0.259887,0.073446326,0.37853107,0.29943502,0.15254237,0.30508474,0.2542373,0.5536723,0.47457626,0.5536723,0.3220339,0.4915254,0.3898305,0.59322035,0.42937854,0.4124294,0.59887004,0.42937854,,0.48022598,0.6440678,0.6836158,0.740113,0.5762712,0.59322035,0.8135593,0.8135593,0.8248588,0.8587571,0.8587571,0.9830508
+FastaiMLP (tuned),0.0,0.0,0.0,0.011299435,0.050847456,0.011299435,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.016949153,0.02259887,0.011299435,0.011299435,0.0,0.03954802,0.03954802,0.02259887,0.050847456,0.02259887,0.011299435,0.10169491,0.118644066,0.02259887,0.07909604,0.03954802,0.084745765,0.056497175,0.084745765,0.14689265,0.16949153,0.14124294,0.10169491,0.18644068,0.15254237,0.28248587,0.14124294,0.27118644,0.2937853,0.26553673,0.1299435,0.31638417,0.31073445,0.2259887,0.22033899,0.20903955,0.5423729,0.40112993,0.5480226,0.38418078,0.44067797,0.3559322,0.56497175,0.33333334,0.42372882,0.5536723,0.40677965,0.519774,,0.6666667,0.6836158,0.78531075,0.6440678,0.7966102,0.81920904,0.8870056,0.8418079,0.94350284,0.9378531,1.0
+TabSTAR (default),0.073446326,0.09039548,0.09039548,0.096045196,0.084745765,0.07909604,0.084745765,0.10169491,0.107344635,0.12429378,0.10169491,0.13559322,0.096045196,0.1299435,0.15819208,0.14124294,0.15254237,0.16384181,0.18079096,0.16949153,0.17514125,0.16949153,0.19209039,0.15254237,0.2259887,0.20338982,0.2259887,0.20338982,0.17514125,0.21468927,0.20338982,0.24858756,0.18079096,0.26553673,0.20903955,0.20903955,0.22033899,0.24293785,0.27683616,0.19209039,0.24858756,0.22033899,0.20903955,0.2937853,0.20338982,0.21468927,0.259887,0.27683616,0.02259887,0.29943502,0.03954802,0.27118644,0.30508474,0.4124294,0.32768363,0.31638417,0.33898306,0.3502825,0.4519774,0.3559322,0.33333334,,0.38418078,0.62711865,0.47457626,0.66101694,0.6666667,0.5536723,0.740113,0.779661,0.779661,0.79096043
+RandomForest (default),0.0056497175,0.011299435,0.0056497175,0.0056497175,0.033898305,0.0056497175,0.0,0.011299435,0.0,0.0056497175,0.011299435,0.04519774,0.0056497175,0.06779661,0.011299435,0.016949153,0.028248588,0.016949153,0.050847456,0.02259887,0.028248588,0.07909604,0.033898305,0.056497175,0.028248588,0.02259887,0.04519774,0.04519774,0.073446326,0.06214689,0.14124294,0.050847456,0.073446326,0.07909604,0.18644068,0.14124294,0.16384181,0.04519774,0.11299435,0.259887,0.06214689,0.15254237,0.23163842,0.33898306,0.29943502,0.27683616,0.096045196,0.14689265,0.47457626,0.011299435,0.49717513,0.3502825,0.20903955,0.3559322,0.31073445,0.19774011,0.31638417,0.43502825,0.38418078,0.31638417,0.31638417,0.6158192,,0.7627119,0.5536723,0.6045198,0.8418079,0.7740113,0.89830506,0.94350284,0.94350284,0.97740114
+KNN (tuned + ensembled),0.016949153,0.02259887,0.03954802,0.028248588,0.02259887,0.016949153,0.028248588,0.03954802,0.02259887,0.03954802,0.02259887,0.084745765,0.02259887,0.096045196,0.06214689,0.03954802,0.07909604,0.06779661,0.03954802,0.04519774,0.07909604,0.11299435,0.096045196,0.050847456,0.03954802,0.09039548,0.056497175,0.107344635,0.13559322,0.11299435,0.14689265,0.06214689,0.14689265,0.118644066,0.20338982,0.15819208,0.107344635,0.033898305,0.096045196,0.17514125,0.03954802,0.16384181,0.17514125,0.23163842,0.18644068,0.20338982,0.10169491,0.18079096,0.27683616,0.096045196,0.27683616,0.19209039,0.1299435,0.28248587,0.084745765,0.19774011,0.18079096,0.15819208,0.33898306,0.259887,0.21468927,0.37288135,0.23728813,,0.43502825,0.40112993,0.91525424,0.32768363,0.9661017,0.84745765,0.84745765,0.9661017
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.011299435,0.011299435,0.011299435,0.0,0.028248588,0.016949153,0.0056497175,0.028248588,0.0,0.016949153,0.056497175,0.016949153,0.056497175,0.016949153,0.0,0.016949153,0.084745765,0.07909604,0.011299435,0.084745765,0.09039548,0.016949153,0.16384181,0.18079096,0.14689265,0.050847456,0.16949153,0.028248588,0.14689265,0.2542373,0.06779661,0.20903955,0.18079096,0.084745765,0.40112993,0.21468927,0.40112993,0.11299435,0.23728813,0.04519774,0.3559322,0.24858756,0.30508474,0.4180791,0.21186441,0.42372882,0.3559322,0.5254237,0.4463277,0.56497175,,0.5084746,0.69491524,0.62711865,0.8135593,0.8926554,0.8926554,0.9661017
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.050847456,0.0056497175,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0056497175,0.011299435,0.0,0.0,0.0056497175,0.0,0.0056497175,0.016949153,0.04519774,0.011299435,0.056497175,0.033898305,0.011299435,0.02259887,0.02259887,0.016949153,0.028248588,0.107344635,0.06214689,0.028248588,0.04519774,0.118644066,0.073446326,0.073446326,0.07909604,0.06779661,0.096045196,0.096045196,0.084745765,0.1299435,0.31638417,0.1299435,0.19774011,0.10169491,0.09039548,0.31073445,0.118644066,0.31073445,0.17514125,0.118644066,0.17514125,0.27683616,0.23163842,0.06214689,0.34463277,0.259887,0.40677965,0.20338982,0.33898306,0.39548022,0.59887004,0.4915254,,0.80225986,0.519774,0.84745765,0.92655367,0.92655367,0.99435025
+KNN (tuned),0.0056497175,0.011299435,0.02259887,0.016949153,0.011299435,0.016949153,0.011299435,0.028248588,0.011299435,0.02259887,0.011299435,0.033898305,0.011299435,0.06214689,0.04519774,0.033898305,0.04519774,0.03954802,0.02259887,0.033898305,0.04519774,0.07909604,0.06779661,0.016949153,0.02259887,0.04519774,0.028248588,0.056497175,0.050847456,0.073446326,0.107344635,0.04519774,0.06779661,0.06779661,0.15254237,0.096045196,0.050847456,0.011299435,0.056497175,0.15819208,0.0056497175,0.11299435,0.1299435,0.16384181,0.16384181,0.14124294,0.073446326,0.12429378,0.20338982,0.06214689,0.20903955,0.16949153,0.06214689,0.1299435,0.04519774,0.12429378,0.14124294,0.096045196,0.15254237,0.18644068,0.18079096,0.33333334,0.15819208,0.084745765,0.30508474,0.19774011,,0.24858756,0.89830506,0.8248588,0.8248588,0.8926554
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0056497175,0.0,0.011299435,0.0,0.011299435,0.02259887,0.0,0.0,0.0,0.0,0.0,0.02259887,0.011299435,0.0,0.016949153,0.016949153,0.0,0.04519774,0.0056497175,0.04519774,0.09039548,0.0056497175,0.011299435,0.118644066,0.23728813,0.096045196,0.12429378,0.050847456,0.011299435,0.34463277,0.03954802,0.34463277,0.118644066,0.03954802,0.14689265,0.14124294,0.056497175,0.07909604,0.22033899,0.19209039,0.18644068,0.11299435,0.4463277,0.2259887,0.6723164,0.37288135,0.48022598,0.75141245,,0.7740113,0.779661,0.7740113,0.84745765
+KNN (default),0.011299435,0.02259887,0.011299435,0.02259887,0.011299435,0.011299435,0.011299435,0.016949153,0.011299435,0.011299435,0.011299435,0.03954802,0.011299435,0.050847456,0.028248588,0.028248588,0.033898305,0.033898305,0.011299435,0.011299435,0.03954802,0.06779661,0.03954802,0.016949153,0.02259887,0.04519774,0.02259887,0.04519774,0.03954802,0.06779661,0.07909604,0.033898305,0.050847456,0.073446326,0.14124294,0.073446326,0.033898305,0.011299435,0.04519774,0.11299435,0.011299435,0.096045196,0.118644066,0.06214689,0.14124294,0.14689265,0.04519774,0.10169491,0.16384181,0.03954802,0.16384181,0.12429378,0.050847456,0.107344635,0.02259887,0.10169491,0.14124294,0.028248588,0.1299435,0.17514125,0.15819208,0.259887,0.10169491,0.033898305,0.18644068,0.15254237,0.10169491,0.2259887,,0.5875706,0.59322035,0.6384181
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0056497175,0.0056497175,0.0,0.0056497175,0.0,0.0,0.011299435,0.033898305,0.016949153,0.0056497175,0.0,0.02259887,0.011299435,0.033898305,0.0,0.02259887,0.016949153,0.0,0.0,0.09039548,0.084745765,0.02259887,0.12429378,0.050847456,0.016949153,0.21468927,0.0,0.21468927,0.016949153,0.0,0.02259887,0.0056497175,0.06214689,0.028248588,0.096045196,0.028248588,0.14124294,0.056497175,0.22033899,0.056497175,0.15254237,0.107344635,0.073446326,0.17514125,0.22033899,0.4124294,,0.53107345,0.779661
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0056497175,0.0056497175,0.0,0.0056497175,0.0,0.0,0.011299435,0.033898305,0.016949153,0.0056497175,0.0,0.02259887,0.016949153,0.033898305,0.0,0.02259887,0.016949153,0.0,0.0,0.09039548,0.084745765,0.02259887,0.12429378,0.050847456,0.016949153,0.21468927,0.0,0.21468927,0.016949153,0.0,0.02259887,0.0056497175,0.06214689,0.028248588,0.107344635,0.028248588,0.14124294,0.06214689,0.22033899,0.056497175,0.15254237,0.107344635,0.073446326,0.17514125,0.2259887,0.40677965,0.46892655,,0.76836157
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0056497175,0.0056497175,0.0,0.0056497175,0.0,0.0056497175,0.011299435,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.016949153,0.0056497175,0.0,0.0,0.0056497175,0.107344635,0.0,0.1299435,0.011299435,0.0,0.0056497175,0.0,0.016949153,0.0,0.011299435,0.011299435,0.016949153,0.0,0.20903955,0.02259887,0.033898305,0.033898305,0.0056497175,0.107344635,0.15254237,0.36158192,0.22033899,0.23163842,
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_all/leaderboard_overview_explorer.html
index f4d53d853edcef7fb02fd1f6415c36836496f1af..40295eabd198b78bcf953d2a98022054329173ef 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | All Tasks | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_all/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_all/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..a70593af7c12e8bd1aba2621671e7cd252d4b629
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_all/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1808.0,"score":0.828,"rank":5.69,"hrank":1.58,"imp":4.4,"train_time":40.79,"predict_time":6.991,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-128","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1679.0,"score":0.64,"rank":9.71,"hrank":3.86,"imp":7.143,"train_time":293.65,"predict_time":4.358,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-90","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1663.0,"score":0.62,"rank":10.34,"hrank":4.05,"imp":8.328,"train_time":3.59,"predict_time":0.395,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-71","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1635.0,"score":0.563,"rank":11.55,"hrank":6.57,"imp":9.842,"train_time":5.75,"predict_time":0.6,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-62","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1620.0,"score":0.55,"rank":12.21,"hrank":5.5,"imp":8.722,"train_time":2059.94,"predict_time":9.785,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-67","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1583.0,"score":0.533,"rank":14.0,"hrank":6.0,"imp":10.282,"train_time":1.96,"predict_time":0.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-88","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1565.0,"score":0.495,"rank":14.93,"hrank":8.34,"imp":10.65,"train_time":2059.94,"predict_time":1.03,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-64","imputed":false},{"position":7,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1556.0,"score":0.458,"rank":15.43,"hrank":7.34,"imp":11.472,"train_time":1754.94,"predict_time":1.766,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-64","imputed":false},{"position":8,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1542.0,"score":0.45,"rank":16.18,"hrank":9.41,"imp":10.828,"train_time":5.72,"predict_time":0.611,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-48","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1514.0,"score":0.408,"rank":17.78,"hrank":11.91,"imp":12.372,"train_time":2791.97,"predict_time":13.886,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-45","imputed":false},{"position":10,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1439.0,"score":0.365,"rank":22.57,"hrank":7.9,"imp":13.352,"train_time":6155.04,"predict_time":386.158,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-59","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1436.0,"score":0.337,"rank":22.76,"hrank":13.38,"imp":13.721,"train_time":2791.97,"predict_time":0.373,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-61","imputed":false},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1433.0,"score":0.34,"rank":22.97,"hrank":11.94,"imp":14.031,"train_time":2.02,"predict_time":0.183,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-48","imputed":false},{"position":13,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1424.0,"score":0.319,"rank":23.63,"hrank":12.72,"imp":14.029,"train_time":2462.02,"predict_time":1.988,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-48","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1418.0,"score":0.253,"rank":24.02,"hrank":18.81,"imp":14.899,"train_time":416.63,"predict_time":2.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-43","imputed":false},{"position":15,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1412.0,"score":0.272,"rank":24.45,"hrank":13.94,"imp":13.908,"train_time":1347.32,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-42","imputed":false},{"position":16,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1410.0,"score":0.278,"rank":24.54,"hrank":15.85,"imp":14.446,"train_time":12741.09,"predict_time":397.882,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-52","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1410.0,"score":0.258,"rank":24.6,"hrank":16.5,"imp":14.172,"train_time":1347.32,"predict_time":0.364,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-41","imputed":false},{"position":18,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1397.0,"score":0.322,"rank":25.48,"hrank":11.97,"imp":14.767,"train_time":6155.04,"predict_time":39.452,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-58","imputed":false},{"position":19,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1387.0,"score":0.302,"rank":26.2,"hrank":12.42,"imp":15.164,"train_time":4618.5,"predict_time":7.735,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-61","imputed":false},{"position":20,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1384.0,"score":0.276,"rank":26.44,"hrank":15.09,"imp":14.72,"train_time":2462.02,"predict_time":0.231,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-56","imputed":false},{"position":21,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1373.0,"score":0.215,"rank":27.21,"hrank":20.18,"imp":15.465,"train_time":700.96,"predict_time":1.438,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-42","imputed":false},{"position":22,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1372.0,"score":0.318,"rank":27.32,"hrank":8.22,"imp":14.549,"train_time":27.33,"predict_time":6.11,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-70","imputed":false},{"position":23,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1357.0,"score":0.2,"rank":28.35,"hrank":22.79,"imp":15.776,"train_time":416.63,"predict_time":0.381,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-48","imputed":false},{"position":24,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1356.0,"score":0.204,"rank":28.45,"hrank":19.86,"imp":15.538,"train_time":5.81,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-45","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1352.0,"score":0.238,"rank":28.75,"hrank":16.56,"imp":15.762,"train_time":4618.5,"predict_time":0.47,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-58","imputed":false},{"position":26,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1349.0,"score":0.239,"rank":28.95,"hrank":17.38,"imp":15.724,"train_time":1.19,"predict_time":0.07,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-58","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1349.0,"score":0.197,"rank":28.96,"hrank":16.5,"imp":15.62,"train_time":700.96,"predict_time":0.213,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-47","imputed":false},{"position":28,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1342.0,"score":0.213,"rank":29.5,"hrank":21.03,"imp":16.932,"train_time":463.3,"predict_time":0.366,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-66","imputed":false},{"position":29,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1339.0,"score":0.236,"rank":29.69,"hrank":18.68,"imp":15.709,"train_time":866.12,"predict_time":2.007,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-51","imputed":false},{"position":30,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1309.0,"score":0.19,"rank":31.98,"hrank":20.49,"imp":17.324,"train_time":463.3,"predict_time":0.031,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-64","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1306.0,"score":0.164,"rank":32.16,"hrank":23.99,"imp":16.078,"train_time":12741.09,"predict_time":60.101,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+51\/-44","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1293.0,"score":0.173,"rank":33.16,"hrank":17.76,"imp":17.003,"train_time":866.12,"predict_time":0.097,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-43","imputed":false},{"position":33,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1290.0,"score":0.223,"rank":33.4,"hrank":17.58,"imp":17.286,"train_time":45.42,"predict_time":39.405,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-71","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1287.0,"score":0.162,"rank":33.62,"hrank":22.74,"imp":17.005,"train_time":7.81,"predict_time":0.237,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-54","imputed":false},{"position":35,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1275.0,"score":0.136,"rank":34.53,"hrank":23.81,"imp":18.211,"train_time":1877.68,"predict_time":0.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-52","imputed":false},{"position":36,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1260.0,"score":0.129,"rank":35.65,"hrank":29.12,"imp":17.11,"train_time":2832.93,"predict_time":1.801,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-55","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1254.0,"score":0.112,"rank":36.12,"hrank":27.59,"imp":17.792,"train_time":10.44,"predict_time":1.714,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-48","imputed":false},{"position":38,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1252.0,"score":0.183,"rank":36.25,"hrank":18.87,"imp":18.78,"train_time":13.96,"predict_time":2.081,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-64","imputed":false},{"position":39,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1249.0,"score":0.129,"rank":36.53,"hrank":28.01,"imp":18.774,"train_time":2.07,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-57","imputed":false},{"position":40,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1221.0,"score":0.1,"rank":38.63,"hrank":27.21,"imp":19.062,"train_time":1877.68,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-56","imputed":false},{"position":41,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1220.0,"score":0.109,"rank":38.65,"hrank":18.78,"imp":19.778,"train_time":13.74,"predict_time":0.316,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-51","imputed":false},{"position":42,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1199.0,"score":0.103,"rank":40.26,"hrank":31.31,"imp":18.901,"train_time":2832.93,"predict_time":0.112,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-64","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1198.0,"score":0.091,"rank":40.29,"hrank":29.52,"imp":19.546,"train_time":257.64,"predict_time":0.846,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-52","imputed":false},{"position":44,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1198.0,"score":0.095,"rank":40.32,"hrank":20.95,"imp":19.773,"train_time":6.9,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-62","imputed":false},{"position":45,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1191.0,"score":0.087,"rank":40.84,"hrank":21.34,"imp":18.816,"train_time":2.06,"predict_time":0.122,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-55","imputed":false},{"position":46,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1174.0,"score":0.09,"rank":42.11,"hrank":32.27,"imp":20.445,"train_time":594.95,"predict_time":4.649,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-66","imputed":false},{"position":47,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1167.0,"score":0.074,"rank":42.58,"hrank":30.29,"imp":20.755,"train_time":257.64,"predict_time":0.068,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-68","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1160.0,"score":0.073,"rank":43.1,"hrank":35.43,"imp":20.666,"train_time":377.14,"predict_time":0.747,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-58","imputed":false},{"position":49,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1158.0,"score":0.061,"rank":43.26,"hrank":37.53,"imp":19.493,"train_time":2.2,"predict_time":0.171,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-50","imputed":false},{"position":50,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1120.0,"score":0.043,"rank":45.9,"hrank":40.52,"imp":21.423,"train_time":377.14,"predict_time":0.091,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-46","imputed":false},{"position":51,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1106.0,"score":0.058,"rank":46.86,"hrank":35.16,"imp":21.987,"train_time":594.95,"predict_time":0.336,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-76","imputed":false},{"position":52,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1087.0,"score":0.061,"rank":48.15,"hrank":30.66,"imp":23.02,"train_time":301.0,"predict_time":65.827,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-71","imputed":false},{"position":53,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1081.0,"score":0.058,"rank":48.58,"hrank":40.24,"imp":26.836,"train_time":176.26,"predict_time":0.499,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-65","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1080.0,"score":0.105,"rank":48.63,"hrank":20.8,"imp":25.535,"train_time":39935.25,"predict_time":4.288,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-94","imputed":false},{"position":55,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1078.0,"score":0.101,"rank":48.74,"hrank":21.68,"imp":25.649,"train_time":39935.25,"predict_time":20.166,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-102","imputed":false},{"position":56,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1034.0,"score":0.019,"rank":51.53,"hrank":47.51,"imp":23.511,"train_time":8.96,"predict_time":0.129,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-60","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1032.0,"score":0.029,"rank":51.67,"hrank":47.56,"imp":28.16,"train_time":176.26,"predict_time":0.191,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-62","imputed":false},{"position":58,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1030.0,"score":0.048,"rank":51.78,"hrank":42.7,"imp":26.433,"train_time":3.14,"predict_time":0.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-77","imputed":false},{"position":59,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1008.0,"score":0.026,"rank":53.12,"hrank":46.17,"imp":25.95,"train_time":1.97,"predict_time":0.251,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-75","imputed":false},{"position":60,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.012,"rank":53.55,"hrank":46.43,"imp":26.122,"train_time":0.43,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-61","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":978.0,"score":0.022,"rank":54.78,"hrank":51.23,"imp":25.675,"train_time":3.12,"predict_time":0.312,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-73","imputed":false},{"position":62,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":977.0,"score":0.073,"rank":54.84,"hrank":21.08,"imp":31.12,"train_time":398.79,"predict_time":4.645,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-116","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":971.0,"score":0.024,"rank":55.17,"hrank":48.82,"imp":27.652,"train_time":129.17,"predict_time":1.627,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-86","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":941.0,"score":0.024,"rank":56.7,"hrank":36.09,"imp":33.201,"train_time":22.76,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-87","imputed":false},{"position":65,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":906.0,"score":0.021,"rank":58.35,"hrank":27.21,"imp":34.226,"train_time":240.78,"predict_time":0.308,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-97","imputed":false},{"position":66,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":880.0,"score":0.017,"rank":59.54,"hrank":37.78,"imp":34.775,"train_time":240.78,"predict_time":0.068,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-110","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":818.0,"score":0.013,"rank":61.92,"hrank":59.17,"imp":33.525,"train_time":129.17,"predict_time":0.103,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-87","imputed":false},{"position":68,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":816.0,"score":0.007,"rank":62.02,"hrank":58.7,"imp":37.288,"train_time":1.23,"predict_time":0.115,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-110","imputed":false},{"position":69,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":620.0,"score":0.0,"rank":67.08,"hrank":66.69,"imp":45.839,"train_time":0.19,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-106","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_all/pareto_front_explorer.html
index 60a62437d9ccf20ef1b7e3bad17c4236ac40a926..c185bc3aa4b27b2446354ebecb5fed8a13986531 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"imp":15.5377012589,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0367019752,"imp":13.9084245042,"elo":1411.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3641930223,"imp":14.1723917299,"elo":1409.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0331429177,"imp":18.7744264176,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306783583,"imp":17.3240680451,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"imp":16.9316157032,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0148501396,"imp":19.7725991452,"elo":1198.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0139622688,"imp":19.061650044,"elo":1220.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1396506436,"imp":18.2107516553,"elo":1275.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2510258327,"imp":25.9497839559,"elo":1007.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0677914174,"imp":20.7552758517,"elo":1167.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8460764958,"imp":19.5455585948,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.311955429,"imp":25.6750552369,"elo":977.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3360688368,"imp":21.9873269621,"elo":1106.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.6489525795,"imp":20.4449871535,"elo":1173.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1711427769,"imp":19.4926263774,"elo":1157.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3811195827,"imp":15.7757103664,"elo":1357.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.2357814014,"imp":14.8989098987,"elo":1417.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.109529595,"imp":14.5485717985,"elo":1371.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3157602832,"imp":19.778071023,"elo":1220.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4695293307,"imp":15.7616286056,"elo":1352.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.7345001255,"imp":15.1639322919,"elo":1387.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0248147768,"imp":33.2013828039,"elo":941.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.191170079,"imp":28.160386648,"elo":1032.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.499326448,"imp":26.8359698793,"elo":1080.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"imp":26.1217262587,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0914129487,"imp":21.423141497,"elo":1120.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7469116355,"imp":20.6660494806,"elo":1160.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7143460095,"imp":17.7924592082,"elo":1254.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3727607921,"imp":13.7212993074,"elo":1436.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":13.8860562126,"imp":12.371801927,"elo":1514.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6105561478,"imp":10.8282818461,"elo":1542.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0296972468,"imp":10.6498498761,"elo":1564.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7853653133,"imp":8.7223345797,"elo":1620.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0805747376,"imp":18.7797835089,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"imp":17.2857772382,"elo":1289.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"imp":14.7667014028,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":386.1584680055,"imp":13.3522283587,"elo":1439.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"imp":14.0312308484,"elo":1433.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.9905568694,"imp":4.3999503576,"elo":1808.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1463864614,"imp":10.2824141521,"elo":1582.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2369687803,"imp":17.0049647562,"elo":1287.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2313083312,"imp":14.7197053031,"elo":1383.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.9881175398,"imp":14.0286995527,"elo":1423.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6004352741,"imp":9.8417888743,"elo":1634.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.394910574,"imp":8.327885548,"elo":1663.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.6449420594,"imp":31.1202640391,"elo":976.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2876390828,"imp":25.5352295001,"elo":1080.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.1657011965,"imp":25.6488080938,"elo":1078.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0695707729,"imp":15.7236417241,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1288563013,"imp":23.5113040512,"elo":1034.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1115292311,"imp":18.9014701036,"elo":1198.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8008738756,"imp":17.1101612807,"elo":1260.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"imp":18.8157636036,"elo":1191.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2127694574,"imp":15.6200706477,"elo":1349.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.4381002715,"imp":15.4650822834,"elo":1373.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":65.8265809722,"imp":23.0198856416,"elo":1087.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":60.1008510867,"imp":16.0778051677,"elo":1306.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":397.8816919076,"imp":14.4458480166,"elo":1410.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7405088776,"imp":26.4328013858,"elo":1030.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0970074274,"imp":17.0029448128,"elo":1293.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0069414377,"imp":15.7088409907,"elo":1339.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"x_train":5.8052200156,"imp":15.5377012589,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0367019752,"x_train":1347.3189819557,"imp":13.9084245042,"elo":1411.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3641930223,"x_train":1347.3189819557,"imp":14.1723917299,"elo":1409.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0331429177,"x_train":2.0744471986,"imp":18.7744264176,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306783583,"x_train":463.3015284407,"imp":17.3240680451,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"x_train":463.3015284407,"imp":16.9316157032,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0148501396,"x_train":6.9024058688,"imp":19.7725991452,"elo":1198.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0139622688,"x_train":1877.6824729228,"imp":19.061650044,"elo":1220.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1396506436,"x_train":1877.6824729228,"imp":18.2107516553,"elo":1275.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2510258327,"x_train":1.9744557778,"imp":25.9497839559,"elo":1007.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0677914174,"x_train":257.6385593165,"imp":20.7552758517,"elo":1167.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8460764958,"x_train":257.6385593165,"imp":19.5455585948,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.311955429,"x_train":3.1153293529,"imp":25.6750552369,"elo":977.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3360688368,"x_train":594.9528585231,"imp":21.9873269621,"elo":1106.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.6489525795,"x_train":594.9528585231,"imp":20.4449871535,"elo":1173.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1711427769,"x_train":2.2021467311,"imp":19.4926263774,"elo":1157.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3811195827,"x_train":416.6281847875,"imp":15.7757103664,"elo":1357.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.2357814014,"x_train":416.6281847875,"imp":14.8989098987,"elo":1417.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.109529595,"x_train":27.3254634149,"imp":14.5485717985,"elo":1371.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3157602832,"x_train":13.7418320826,"imp":19.778071023,"elo":1220.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4695293307,"x_train":4618.5023417328,"imp":15.7616286056,"elo":1352.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.7345001255,"x_train":4618.5023417328,"imp":15.1639322919,"elo":1387.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0248147768,"x_train":22.7552921112,"imp":33.2013828039,"elo":941.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.191170079,"x_train":176.2617571354,"imp":28.160386648,"elo":1032.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.499326448,"x_train":176.2617571354,"imp":26.8359698793,"elo":1080.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":0.431661138,"imp":26.1217262587,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0914129487,"x_train":377.1421152596,"imp":21.423141497,"elo":1120.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7469116355,"x_train":377.1421152596,"imp":20.6660494806,"elo":1160.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7143460095,"x_train":10.4435312748,"imp":17.7924592082,"elo":1254.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3727607921,"x_train":2791.9683399256,"imp":13.7212993074,"elo":1436.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":13.8860562126,"x_train":2791.9683399256,"imp":12.371801927,"elo":1514.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6105561478,"x_train":5.7153047813,"imp":10.8282818461,"elo":1542.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0296972468,"x_train":2059.9443792676,"imp":10.6498498761,"elo":1564.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7853653133,"x_train":2059.9443792676,"imp":8.7223345797,"elo":1620.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0805747376,"x_train":13.9576562813,"imp":18.7797835089,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"x_train":45.4161088134,"imp":17.2857772382,"elo":1289.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"x_train":6155.0377202868,"imp":14.7667014028,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":386.1584680055,"x_train":6155.0377202868,"imp":13.3522283587,"elo":1439.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"x_train":2.0183513985,"imp":14.0312308484,"elo":1433.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.9905568694,"x_train":40.7926562599,"imp":4.3999503576,"elo":1808.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1463864614,"x_train":1.9632436902,"imp":10.2824141521,"elo":1582.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2369687803,"x_train":7.8050600165,"imp":17.0049647562,"elo":1287.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2313083312,"x_train":2462.0152230263,"imp":14.7197053031,"elo":1383.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.9881175398,"x_train":2462.0152230263,"imp":14.0286995527,"elo":1423.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6004352741,"x_train":5.7520413762,"imp":9.8417888743,"elo":1634.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.394910574,"x_train":3.5877237797,"imp":8.327885548,"elo":1663.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.6449420594,"x_train":398.7869028722,"imp":31.1202640391,"elo":976.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2876390828,"x_train":39935.2451367739,"imp":25.5352295001,"elo":1080.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.1657011965,"x_train":39935.2451367739,"imp":25.6488080938,"elo":1078.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0695707729,"x_train":1.1859471423,"imp":15.7236417241,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1288563013,"x_train":8.9576391549,"imp":23.5113040512,"elo":1034.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1115292311,"x_train":2832.9262089828,"imp":18.9014701036,"elo":1198.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8008738756,"x_train":2832.9262089828,"imp":17.1101612807,"elo":1260.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"x_train":2.0592641478,"imp":18.8157636036,"elo":1191.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2127694574,"x_train":700.9607998747,"imp":15.6200706477,"elo":1349.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.4381002715,"x_train":700.9607998747,"imp":15.4650822834,"elo":1373.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":65.8265809722,"x_train":300.998886807,"imp":23.0198856416,"elo":1087.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":60.1008510867,"x_train":12741.0939157936,"imp":16.0778051677,"elo":1306.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":397.8816919076,"x_train":12741.0939157936,"imp":14.4458480166,"elo":1410.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7405088776,"x_train":3.1442091702,"imp":26.4328013858,"elo":1030.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0970074274,"x_train":866.1169883185,"imp":17.0029448128,"elo":1293.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0069414377,"x_train":866.1169883185,"imp":15.7088409907,"elo":1339.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_all/tuning_trajectories_explorer.html
index acb297aa4f3720e86eb7a3680667460f6e1e1bc2..5a2e911dae3d9ee350bdd4fa621e9eb8fd1c2cb2 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":5.8052200156,"imp":15.5648407163,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":11.6250818675,"imp":15.0429267109,"elo":1394.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":30.7197244584,"imp":14.6761246189,"elo":1415.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":60.1926505566,"imp":14.4920220186,"elo":1423.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":159.5146818194,"imp":14.3682252845,"elo":1429.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":318.9223225532,"imp":14.3588958572,"elo":1427.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":662.5706287402,"imp":14.2535496982,"elo":1425.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":998.7349671254,"imp":14.254965554,"elo":1421.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1347.3189819557,"imp":14.1894664163,"elo":1425.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.0744471986,"imp":18.803962251,"elo":1258.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.099236637,"imp":18.2224270029,"elo":1294.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.8289007778,"imp":17.8181484295,"elo":1319.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.1099632833,"imp":17.5743915231,"elo":1331.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":57.3124085491,"imp":17.2606557315,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":115.9554885729,"imp":17.0795268516,"elo":1355.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":231.6358590441,"imp":16.9842884207,"elo":1359.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":346.5027373053,"imp":16.9400894252,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":463.3015284407,"imp":16.9556854491,"elo":1364.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":6.9024058688,"imp":19.797825202,"elo":1203.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":16.7298961864,"imp":19.3101728135,"elo":1225.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":47.9305716086,"imp":18.9171721825,"elo":1250.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":98.1075414046,"imp":18.769810966,"elo":1258.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":242.3014674264,"imp":18.5920759261,"elo":1271.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":485.7841387908,"imp":18.4535638914,"elo":1278.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":938.1843487506,"imp":18.3091182738,"elo":1287.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1390.5258657356,"imp":18.2699631411,"elo":1289.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1877.6824729228,"imp":18.2305129315,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.9744557778,"imp":25.9719850481,"elo":1012.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1741914103,"imp":24.2066963836,"elo":1046.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.7533015627,"imp":21.5446302695,"elo":1112.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":13.1877719851,"imp":20.6514295653,"elo":1147.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":32.4660623442,"imp":19.9524201526,"elo":1177.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":64.6497454444,"imp":19.7272965469,"elo":1189.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":128.5728546231,"imp":19.5537818338,"elo":1199.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":192.4203093077,"imp":19.4579418721,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":257.6385593165,"imp":19.5972358994,"elo":1205.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.1153293529,"imp":25.6948562863,"elo":981.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.1374994683,"imp":24.7941756965,"elo":1028.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":15.2752739191,"imp":23.062273263,"elo":1093.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":30.4682284412,"imp":22.3087196548,"elo":1121.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":73.2357604794,"imp":21.373740229,"elo":1146.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":146.7179006005,"imp":20.9428559211,"elo":1161.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":295.2364780788,"imp":20.6577215773,"elo":1170.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":443.050159936,"imp":20.4959961399,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":594.9528585231,"imp":20.4760080266,"elo":1183.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.2021467311,"imp":19.5198980548,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":4.7004585926,"imp":17.8130691799,"elo":1253.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":10.0203601171,"imp":16.5239322538,"elo":1339.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":19.9125130873,"imp":15.9054526636,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":50.9228061571,"imp":15.3365722443,"elo":1409.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":102.592415423,"imp":15.1501318634,"elo":1424.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":206.3698310558,"imp":15.0196601548,"elo":1433.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":309.4092129514,"imp":14.9371107834,"elo":1438.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":416.6281847875,"imp":14.92582676,"elo":1442.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":27.3254634149,"imp":14.6950238353,"elo":1387.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":13.7418320826,"imp":19.8039194607,"elo":1230.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":40.2965934723,"imp":18.4068609997,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":115.9419475491,"imp":16.3136411365,"elo":1357.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":237.5304229618,"imp":15.8025406631,"elo":1385.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":577.047955689,"imp":15.2967384188,"elo":1411.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1164.1941811379,"imp":15.255157946,"elo":1408.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2285.7499058023,"imp":15.3996032217,"elo":1402.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3416.2270282822,"imp":15.3011569249,"elo":1406.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4618.5023417328,"imp":15.2072981582,"elo":1407.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":22.7552921112,"imp":33.2311888654,"elo":951.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":51.1368372531,"imp":30.2122732637,"elo":1014.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":147.0385313693,"imp":27.0027389411,"elo":1082.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":176.2617571354,"imp":26.8396071498,"elo":1087.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.431661138,"imp":26.1557333168,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.2883199842,"imp":23.4654815334,"elo":1055.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":7.9454413682,"imp":22.0549131531,"elo":1099.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":17.3814840956,"imp":21.3445395345,"elo":1121.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":45.2922595201,"imp":20.8171603696,"elo":1153.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":91.6104975483,"imp":20.6992929272,"elo":1161.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":185.2817035188,"imp":20.6901750904,"elo":1165.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":279.7219448516,"imp":20.7023060389,"elo":1166.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":377.1421152596,"imp":20.6894644237,"elo":1167.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.4435312748,"imp":17.8143936393,"elo":1259.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.7804646251,"imp":15.5381130037,"elo":1363.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":65.9805881846,"imp":13.8954587129,"elo":1451.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":135.870476352,"imp":13.2533408321,"elo":1477.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":350.3376696757,"imp":12.706297322,"elo":1509.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":701.5287510764,"imp":12.4381635538,"elo":1527.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1392.1340578984,"imp":12.3427151902,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2087.8411572447,"imp":12.3840026294,"elo":1539.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2791.9683399256,"imp":12.4037173503,"elo":1542.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.7153047813,"imp":10.8605774455,"elo":1554.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":17.0263004767,"imp":10.754922001,"elo":1558.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":47.4652638957,"imp":10.4909400184,"elo":1568.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":98.3276706028,"imp":10.3982563804,"elo":1586.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":249.0037143861,"imp":10.1750350699,"elo":1607.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":505.0675823061,"imp":10.0315820872,"elo":1611.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1022.195993405,"imp":9.5738934932,"elo":1628.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1532.1033434601,"imp":9.3044372938,"elo":1635.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2059.9443792676,"imp":8.7718751361,"elo":1639.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.9576562813,"imp":18.8030496799,"elo":1264.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":45.4161088134,"imp":17.3071055838,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":79.4617813141,"imp":16.3353524595,"elo":1338.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":160.0616335809,"imp":15.4327638331,"elo":1381.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":301.1906399516,"imp":15.0349456363,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":750.5793401849,"imp":14.3399842573,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1518.8392119437,"imp":13.871556175,"elo":1435.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3050.4213050061,"imp":13.6470637463,"elo":1446.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":4599.9264282578,"imp":13.5027665629,"elo":1450.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":6155.0377202868,"imp":13.3908451647,"elo":1456.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0183513985,"imp":14.056234431,"elo":1449.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":40.7926562599,"imp":4.4299677362,"elo":1825.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.9632436902,"imp":10.3139360053,"elo":1602.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":7.8050600165,"imp":17.0317131179,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":18.1240740517,"imp":16.5600957411,"elo":1324.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":58.9610972546,"imp":15.9011903106,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":118.8818809271,"imp":15.3155851225,"elo":1390.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":303.352703028,"imp":14.8545181325,"elo":1412.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":608.2074565649,"imp":14.5470290664,"elo":1429.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1220.3134501457,"imp":14.2553055009,"elo":1440.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1834.4356979489,"imp":14.1532931618,"elo":1447.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2462.0152230263,"imp":14.0602958732,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.7520413762,"imp":9.8721723247,"elo":1663.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.5877237797,"imp":8.322422076,"elo":1678.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":398.7869028722,"imp":31.1348528181,"elo":982.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":634.250223124,"imp":30.9873744671,"elo":992.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1221.7141616423,"imp":30.5856612352,"elo":1002.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2110.4203991936,"imp":29.7722523023,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5146.9875310979,"imp":28.255093574,"elo":1043.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":10543.907094521,"imp":26.9507222458,"elo":1064.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":19963.1285723042,"imp":26.4078834154,"elo":1075.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":29831.9301413401,"imp":26.0380251999,"elo":1086.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":39935.2451367739,"imp":25.6716945777,"elo":1099.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1859471423,"imp":15.7543926205,"elo":1369.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":8.9576391549,"imp":23.5277410036,"elo":1042.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":22.6590123875,"imp":21.3977230887,"elo":1105.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":65.5045736145,"imp":19.5483021144,"elo":1164.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":143.026937766,"imp":18.7796041734,"elo":1204.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":340.3179128497,"imp":18.0793699099,"elo":1235.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":710.8916252159,"imp":17.6974248663,"elo":1253.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1385.1316080566,"imp":17.4487178376,"elo":1259.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2123.8601198211,"imp":17.2515174336,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2832.9262089828,"imp":17.1343116063,"elo":1276.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.0592641478,"imp":18.84416001,"elo":1196.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.635768988,"imp":17.0330234601,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":17.5495862649,"imp":16.4524635473,"elo":1335.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":33.8580856565,"imp":16.0885473009,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.7445970232,"imp":15.8017615919,"elo":1374.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":173.5566472908,"imp":15.6890194985,"elo":1383.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":346.7594994462,"imp":15.5938521761,"elo":1391.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":520.510860917,"imp":15.5152373823,"elo":1396.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":700.9607998747,"imp":15.488393282,"elo":1396.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":300.998886807,"imp":23.0427043276,"elo":1099.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":757.2419712235,"imp":20.0204594559,"elo":1180.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2344.8516752833,"imp":18.2137985632,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4981.7847745692,"imp":16.8421397984,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12170.8681485011,"imp":14.4555741462,"elo":1428.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12741.0939157936,"imp":14.4817244065,"elo":1428.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.1442091702,"imp":26.4503678235,"elo":1042.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":7.5774846269,"imp":23.1343988247,"elo":1107.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":20.6232394062,"imp":19.0975920618,"elo":1224.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":42.5478154764,"imp":17.3181130029,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":107.8487563604,"imp":16.4810742828,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":214.7680889693,"imp":16.1138559721,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":431.3100440381,"imp":15.8803008463,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":645.974194479,"imp":15.7774000327,"elo":1357.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":866.1169883185,"imp":15.6723469022,"elo":1359.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"x_train":5.8052200156,"imp":15.5648407163,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0601795968,"x_train":11.6250818675,"imp":15.0429267109,"elo":1394.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1181383737,"x_train":30.7197244584,"imp":14.6761246189,"elo":1415.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1643410418,"x_train":60.1926505566,"imp":14.4920220186,"elo":1423.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2285526113,"x_train":159.5146818194,"imp":14.3682252845,"elo":1429.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2603250146,"x_train":318.9223225532,"imp":14.3588958572,"elo":1427.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3044187588,"x_train":662.5706287402,"imp":14.2535496982,"elo":1425.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3313311934,"x_train":998.7349671254,"imp":14.254965554,"elo":1421.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3641930223,"x_train":1347.3189819557,"imp":14.1894664163,"elo":1425.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0331429177,"x_train":2.0744471986,"imp":18.803962251,"elo":1258.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0646948782,"x_train":4.099236637,"imp":18.2224270029,"elo":1294.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1097672871,"x_train":10.8289007778,"imp":17.8181484295,"elo":1319.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1712030423,"x_train":22.1099632833,"imp":17.5743915231,"elo":1331.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2532193674,"x_train":57.3124085491,"imp":17.2606557315,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3030166676,"x_train":115.9554885729,"imp":17.0795268516,"elo":1355.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3373890654,"x_train":231.6358590441,"imp":16.9842884207,"elo":1359.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3569505185,"x_train":346.5027373053,"imp":16.9400894252,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"x_train":463.3015284407,"imp":16.9556854491,"elo":1364.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0148501396,"x_train":6.9024058688,"imp":19.797825202,"elo":1203.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0303468055,"x_train":16.7298961864,"imp":19.3101728135,"elo":1225.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0551923837,"x_train":47.9305716086,"imp":18.9171721825,"elo":1250.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0698156568,"x_train":98.1075414046,"imp":18.769810966,"elo":1258.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0981563113,"x_train":242.3014674264,"imp":18.5920759261,"elo":1271.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1132271626,"x_train":485.7841387908,"imp":18.4535638914,"elo":1278.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1244815695,"x_train":938.1843487506,"imp":18.3091182738,"elo":1287.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1306053559,"x_train":1390.5258657356,"imp":18.2699631411,"elo":1289.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1402144831,"x_train":1877.6824729228,"imp":18.2305129315,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2510258327,"x_train":1.9744557778,"imp":25.9719850481,"elo":1012.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2910788221,"x_train":3.1741914103,"imp":24.2066963836,"elo":1046.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3725395347,"x_train":6.7533015627,"imp":21.5446302695,"elo":1112.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4132860205,"x_train":13.1877719851,"imp":20.6514295653,"elo":1147.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5777882664,"x_train":32.4660623442,"imp":19.9524201526,"elo":1177.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7270465219,"x_train":64.6497454444,"imp":19.7272965469,"elo":1189.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.83253734,"x_train":128.5728546231,"imp":19.5537818338,"elo":1199.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8229876046,"x_train":192.4203093077,"imp":19.4579418721,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8460764958,"x_train":257.6385593165,"imp":19.5972358994,"elo":1205.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.311955429,"x_train":3.1153293529,"imp":25.6948562863,"elo":981.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.599233021,"x_train":6.1374994683,"imp":24.7941756965,"elo":1028.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1109333389,"x_train":15.2752739191,"imp":23.062273263,"elo":1093.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7115436462,"x_train":30.4682284412,"imp":22.3087196548,"elo":1121.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4877540535,"x_train":73.2357604794,"imp":21.373740229,"elo":1146.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.2290986651,"x_train":146.7179006005,"imp":20.9428559211,"elo":1161.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.9684851932,"x_train":295.2364780788,"imp":20.6577215773,"elo":1170.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.1837657267,"x_train":443.050159936,"imp":20.4959961399,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.6489525795,"x_train":594.9528585231,"imp":20.4760080266,"elo":1183.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1711427769,"x_train":2.2021467311,"imp":19.5198980548,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6899284291,"x_train":4.7004585926,"imp":17.8130691799,"elo":1253.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9547904471,"x_train":10.0203601171,"imp":16.5239322538,"elo":1339.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.415717788,"x_train":19.9125130873,"imp":15.9054526636,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5465698671,"x_train":50.9228061571,"imp":15.3365722443,"elo":1409.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6454746717,"x_train":102.592415423,"imp":15.1501318634,"elo":1424.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.9962776452,"x_train":206.3698310558,"imp":15.0196601548,"elo":1433.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1744811378,"x_train":309.4092129514,"imp":14.9371107834,"elo":1438.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2357814014,"x_train":416.6281847875,"imp":14.92582676,"elo":1442.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":6.109529595,"x_train":27.3254634149,"imp":14.6950238353,"elo":1387.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3157602832,"x_train":13.7418320826,"imp":19.8039194607,"elo":1230.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8658037861,"x_train":40.2965934723,"imp":18.4068609997,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7616896461,"x_train":115.9419475491,"imp":16.3136411365,"elo":1357.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7495673171,"x_train":237.5304229618,"imp":15.8025406631,"elo":1385.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.9629243753,"x_train":577.047955689,"imp":15.2967384188,"elo":1411.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1800785058,"x_train":1164.1941811379,"imp":15.255157946,"elo":1408.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.7233625262,"x_train":2285.7499058023,"imp":15.3996032217,"elo":1402.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.6697936774,"x_train":3416.2270282822,"imp":15.3011569249,"elo":1406.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.7345001255,"x_train":4618.5023417328,"imp":15.2072981582,"elo":1407.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0248147768,"x_train":22.7552921112,"imp":33.2311888654,"elo":951.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1956514835,"x_train":51.1368372531,"imp":30.2122732637,"elo":1014.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4510510662,"x_train":147.0385313693,"imp":27.0027389411,"elo":1082.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.499326448,"x_train":176.2617571354,"imp":26.8396071498,"elo":1087.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":0.431661138,"imp":26.1557333168,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1195246673,"x_train":2.2883199842,"imp":23.4654815334,"elo":1055.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2293352158,"x_train":7.9454413682,"imp":22.0549131531,"elo":1099.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3451275073,"x_train":17.3814840956,"imp":21.3445395345,"elo":1121.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4910539573,"x_train":45.2922595201,"imp":20.8171603696,"elo":1153.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5816243107,"x_train":91.6104975483,"imp":20.6992929272,"elo":1161.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.664675779,"x_train":185.2817035188,"imp":20.6901750904,"elo":1165.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7215840165,"x_train":279.7219448516,"imp":20.7023060389,"elo":1166.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7462800839,"x_train":377.1421152596,"imp":20.6894644237,"elo":1167.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7143460095,"x_train":10.4435312748,"imp":17.8143936393,"elo":1259.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.7127272606,"x_train":24.7804646251,"imp":15.5381130037,"elo":1363.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.9943069673,"x_train":65.9805881846,"imp":13.8954587129,"elo":1451.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.4327546965,"x_train":135.870476352,"imp":13.2533408321,"elo":1477.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.3277160444,"x_train":350.3376696757,"imp":12.706297322,"elo":1509.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.3527013663,"x_train":701.5287510764,"imp":12.4381635538,"elo":1527.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.7064755441,"x_train":1392.1340578984,"imp":12.3427151902,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.274656038,"x_train":2087.8411572447,"imp":12.3840026294,"elo":1539.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.8860562126,"x_train":2791.9683399256,"imp":12.4037173503,"elo":1542.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6105561478,"x_train":5.7153047813,"imp":10.8605774455,"elo":1554.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3877879479,"x_train":17.0263004767,"imp":10.754922001,"elo":1558.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8738328207,"x_train":47.4652638957,"imp":10.4909400184,"elo":1568.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1588746477,"x_train":98.3276706028,"imp":10.3982563804,"elo":1586.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.8281839931,"x_train":249.0037143861,"imp":10.1750350699,"elo":1607.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4738495327,"x_train":505.0675823061,"imp":10.0315820872,"elo":1611.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6074456326,"x_train":1022.195993405,"imp":9.5738934932,"elo":1628.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.8784737047,"x_train":1532.1033434601,"imp":9.3044372938,"elo":1635.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7107599974,"x_train":2059.9443792676,"imp":8.7718751361,"elo":1639.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0805747376,"x_train":13.9576562813,"imp":18.8030496799,"elo":1264.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"x_train":45.4161088134,"imp":17.3071055838,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":56.7537152129,"x_train":79.4617813141,"imp":16.3353524595,"elo":1338.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":89.1044883635,"x_train":160.0616335809,"imp":15.4327638331,"elo":1381.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":119.8919914946,"x_train":301.1906399516,"imp":15.0349456363,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":187.2342742423,"x_train":750.5793401849,"imp":14.3399842573,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":245.0072603318,"x_train":1518.8392119437,"imp":13.871556175,"elo":1435.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":313.8851452232,"x_train":3050.4213050061,"imp":13.6470637463,"elo":1446.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":354.6149052654,"x_train":4599.9264282578,"imp":13.5027665629,"elo":1450.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":387.8903208002,"x_train":6155.0377202868,"imp":13.3908451647,"elo":1456.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"x_train":2.0183513985,"imp":14.056234431,"elo":1449.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.9905568694,"x_train":40.7926562599,"imp":4.4299677362,"elo":1825.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1463864614,"x_train":1.9632436902,"imp":10.3139360053,"elo":1602.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2369687803,"x_train":7.8050600165,"imp":17.0317131179,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.439953166,"x_train":18.1240740517,"imp":16.5600957411,"elo":1324.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9304964542,"x_train":58.9610972546,"imp":15.9011903106,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.898947224,"x_train":118.8818809271,"imp":15.3155851225,"elo":1390.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.1135689245,"x_train":303.352703028,"imp":14.8545181325,"elo":1412.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.5576850953,"x_train":608.2074565649,"imp":14.5470290664,"elo":1429.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.7400707788,"x_train":1220.3134501457,"imp":14.2553055009,"elo":1440.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.9026564546,"x_train":1834.4356979489,"imp":14.1532931618,"elo":1447.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":1.9881175398,"x_train":2462.0152230263,"imp":14.0602958732,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6004352741,"x_train":5.7520413762,"imp":9.8721723247,"elo":1663.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.394910574,"x_train":3.5877237797,"imp":8.322422076,"elo":1678.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.6449420594,"x_train":398.7869028722,"imp":31.1348528181,"elo":982.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.0436884998,"x_train":634.250223124,"imp":30.9873744671,"elo":992.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.2115756111,"x_train":1221.7141616423,"imp":30.5856612352,"elo":1002.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.7979635353,"x_train":2110.4203991936,"imp":29.7722523023,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.002547622,"x_train":5146.9875310979,"imp":28.255093574,"elo":1043.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.7064222686,"x_train":10543.907094521,"imp":26.9507222458,"elo":1064.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.6908292471,"x_train":19963.1285723042,"imp":26.4078834154,"elo":1075.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.7650487513,"x_train":29831.9301413401,"imp":26.0380251999,"elo":1086.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.1657011965,"x_train":39935.2451367739,"imp":25.6716945777,"elo":1099.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0695707729,"x_train":1.1859471423,"imp":15.7543926205,"elo":1369.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1288563013,"x_train":8.9576391549,"imp":23.5277410036,"elo":1042.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2470592409,"x_train":22.6590123875,"imp":21.3977230887,"elo":1105.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4546102434,"x_train":65.5045736145,"imp":19.5483021144,"elo":1164.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7104475927,"x_train":143.026937766,"imp":18.7796041734,"elo":1204.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0905794242,"x_train":340.3179128497,"imp":18.0793699099,"elo":1235.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3321372143,"x_train":710.8916252159,"imp":17.6974248663,"elo":1253.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6375611361,"x_train":1385.1316080566,"imp":17.4487178376,"elo":1259.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8070112948,"x_train":2123.8601198211,"imp":17.2515174336,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.9084608704,"x_train":2832.9262089828,"imp":17.1343116063,"elo":1276.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"x_train":2.0592641478,"imp":18.84416001,"elo":1196.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.560565097,"x_train":5.635768988,"imp":17.0330234601,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0675882643,"x_train":17.5495862649,"imp":16.4524635473,"elo":1335.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1769638293,"x_train":33.8580856565,"imp":16.0885473009,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4785625458,"x_train":86.7445970232,"imp":15.8017615919,"elo":1374.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.6035368187,"x_train":173.5566472908,"imp":15.6890194985,"elo":1383.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7255766262,"x_train":346.7594994462,"imp":15.5938521761,"elo":1391.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.5600405676,"x_train":520.510860917,"imp":15.5152373823,"elo":1396.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4381002715,"x_train":700.9607998747,"imp":15.488393282,"elo":1396.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":65.8265809722,"x_train":300.998886807,"imp":23.0427043276,"elo":1099.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":133.265500669,"x_train":757.2419712235,"imp":20.0204594559,"elo":1180.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":232.2933397079,"x_train":2344.8516752833,"imp":18.2137985632,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":345.85504867,"x_train":4981.7847745692,"imp":16.8421397984,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":412.1802530188,"x_train":12170.8681485011,"imp":14.4555741462,"elo":1428.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":397.8816919076,"x_train":12741.0939157936,"imp":14.4817244065,"elo":1428.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.7405088776,"x_train":3.1442091702,"imp":26.4503678235,"elo":1042.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8845279741,"x_train":7.5774846269,"imp":23.1343988247,"elo":1107.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1069120467,"x_train":20.6232394062,"imp":19.0975920618,"elo":1224.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.5245468853,"x_train":42.5478154764,"imp":17.3181130029,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.6856549267,"x_train":107.8487563604,"imp":16.4810742828,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.157086811,"x_train":214.7680889693,"imp":16.1138559721,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.8867738381,"x_train":431.3100440381,"imp":15.8803008463,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.1655042904,"x_train":645.974194479,"imp":15.7774000327,"elo":1357.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9734199806,"x_train":866.1169883185,"imp":15.6723469022,"elo":1359.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_all/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..212ff1b060503a8846b7709a4f08cb02cbb3001a
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_all/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (default)", "RealMLP (tuned + ensembled)", "TabDPT (tuned + ensembled)", "RealMLP (tuned)", "TabDPT-Turbo (default)", "TabM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "CatBoost (tuned)", "iLTM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "TabDPT (tuned)", "ModernNCA (tuned + ensembled)", "TabM (tuned)", "XGBoost (tuned + ensembled)", "LimiX (default)", "LightGBM (tuned)", "CatBoost (default)", "ModernNCA (tuned)", "TabSwift (default)", "XGBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "xRFM (tuned + ensembled)", "ChimeraBoost (tuned)", "iLTM (tuned)", "xRFM (tuned)", "TabDPT (default)", "TabM (default)", "EBM (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "RealMLP (default)", "SAP-RPT-OSS (default)", "ChimeraBoost (default)", "EBM (tuned)", "ModernNCA (default)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "EBM (default)", "XGBoost (default)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (tuned)", "RandomForest (tuned + ensembled)", "LightGBM (default)", "RandomForest (tuned)", "FastaiMLP (tuned)", "iLTM (default)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "TorchMLP (default)", "PerpetualBooster (tuned)", "xRFM (default)", "ExtraTrees (default)", "RandomForest (default)", "FastaiMLP (default)", "TabSTAR (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "KNN (tuned)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.7254902124404907, 0.8039215803146362, 0.843137264251709, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.9019607901573181, 0.8823529481887817, 0.9411764740943909, 0.9411764740943909, 0.9411764740943909, 0.9215686321258545, 0.8627451062202454, 0.9411764740943909, 0.9607843160629272, 0.8823529481887817, 0.9607843160629272, 0.9215686321258545, 0.9019607901573181, 0.9607843160629272, 0.9019607901573181, 0.9215686321258545, 0.8823529481887817, 0.8823529481887817, 0.9215686321258545, 0.9411764740943909, 0.9019607901573181, 0.9411764740943909, 0.9411764740943909, 0.9803921580314636, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272, 0.9215686321258545, 0.9215686321258545, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272, 0.9803921580314636, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9411764740943909, 1.0, 0.9411764740943909, 0.9215686321258545, 0.9803921580314636, 1.0, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 0.9607843160629272, 0.9411764740943909, 1.0, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0], [0.27450981736183167, NaN, 0.45098039507865906, 0.6078431606292725, 0.5686274766921997, 0.5098039507865906, 0.6666666865348816, 0.7450980544090271, 0.6078431606292725, 0.8235294222831726, 0.7450980544090271, 0.843137264251709, 0.7450980544090271, 0.8627451062202454, 0.9215686321258545, 0.8823529481887817, 0.9019607901573181, 0.9019607901573181, 0.7647058963775635, 0.8627451062202454, 0.843137264251709, 0.9411764740943909, 0.7647058963775635, 0.9411764740943909, 0.9411764740943909, 0.9215686321258545, 0.843137264251709, 0.9411764740943909, 0.9411764740943909, 0.9215686321258545, 0.9411764740943909, 0.9215686321258545, 0.9019607901573181, 0.8235294222831726, 0.9019607901573181, 0.8823529481887817, 0.9411764740943909, 0.9019607901573181, 0.8627451062202454, 0.9607843160629272, 0.9019607901573181, 0.9607843160629272, 0.9215686321258545, 0.9411764740943909, 0.9215686321258545, 0.9411764740943909, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9803921580314636, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9019607901573181, 0.8823529481887817, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9019607901573181, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 0.9803921580314636, 1.0], [0.19607843458652496, 0.5490196347236633, NaN, 0.5686274766921997, 0.5588235259056091, 0.6274510025978088, 0.686274528503418, 0.6274510025978088, 0.6470588445663452, 0.7450980544090271, 0.7058823704719543, 0.7647058963775635, 0.843137264251709, 0.843137264251709, 0.8823529481887817, 0.7647058963775635, 0.843137264251709, 0.8039215803146362, 0.8235294222831726, 0.9019607901573181, 0.9019607901573181, 0.8627451062202454, 0.7647058963775635, 0.9411764740943909, 0.8627451062202454, 0.843137264251709, 0.8823529481887817, 0.843137264251709, 0.8823529481887817, 0.8627451062202454, 0.8627451062202454, 0.9215686321258545, 0.9215686321258545, 0.9215686321258545, 0.8823529481887817, 0.8235294222831726, 0.9019607901573181, 0.9411764740943909, 0.843137264251709, 0.9019607901573181, 0.8823529481887817, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.9019607901573181, 0.9411764740943909, 0.9019607901573181, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9607843160629272, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 1.0, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 0.9019607901573181, 0.9607843160629272, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0], [0.1568627506494522, 0.3921568691730499, 0.4313725531101227, NaN, 0.47058823704719543, 0.5098039507865906, 0.5490196347236633, 0.5490196347236633, 0.686274528503418, 0.7254902124404907, 0.7058823704719543, 0.8039215803146362, 0.7647058963775635, 0.7843137383460999, 0.8823529481887817, 0.8039215803146362, 0.843137264251709, 0.843137264251709, 0.7254902124404907, 0.8627451062202454, 0.8235294222831726, 0.8235294222831726, 0.8039215803146362, 0.9411764740943909, 0.9019607901573181, 0.843137264251709, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.8627451062202454, 0.7843137383460999, 0.8823529481887817, 0.9019607901573181, 0.9215686321258545, 0.9215686321258545, 0.8627451062202454, 0.8823529481887817, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9019607901573181, 0.9607843160629272, 0.9215686321258545, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9215686321258545, 0.9607843160629272, 0.9803921580314636, 0.8823529481887817, 0.8823529481887817, 0.9803921580314636, 1.0, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.8823529481887817, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0], [0.13725490868091583, 0.4313725531101227, 0.44117647409439087, 0.529411792755127, NaN, 0.5490196347236633, 0.7254902124404907, 0.6078431606292725, 0.7058823704719543, 0.7450980544090271, 0.6666666865348816, 0.7647058963775635, 0.7058823704719543, 0.7254902124404907, 0.843137264251709, 0.8235294222831726, 0.7843137383460999, 0.8627451062202454, 0.7254902124404907, 0.7647058963775635, 0.8039215803146362, 0.8627451062202454, 0.7450980544090271, 0.8823529481887817, 0.9019607901573181, 0.8627451062202454, 0.8235294222831726, 0.8627451062202454, 0.843137264251709, 0.843137264251709, 0.8823529481887817, 0.843137264251709, 0.8823529481887817, 0.7843137383460999, 0.8823529481887817, 0.8823529481887817, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.9019607901573181, 0.9411764740943909, 0.9019607901573181, 0.9607843160629272, 0.8823529481887817, 0.9411764740943909, 0.9215686321258545, 0.9803921580314636, 0.9411764740943909, 0.9411764740943909, 0.9215686321258545, 0.9607843160629272, 0.8823529481887817, 0.8823529481887817, 0.9607843160629272, 0.9803921580314636, 0.9411764740943909, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 1.0], [0.11764705926179886, 0.4901960790157318, 0.37254902720451355, 0.4901960790157318, 0.45098039507865906, NaN, 0.6274510025978088, 0.6666666865348816, 0.5882353186607361, 0.6666666865348816, 0.7254902124404907, 0.686274528503418, 0.7647058963775635, 0.7254902124404907, 0.8235294222831726, 0.7647058963775635, 0.7843137383460999, 0.7647058963775635, 0.7450980544090271, 0.8235294222831726, 0.7450980544090271, 0.7843137383460999, 0.7450980544090271, 0.843137264251709, 0.8039215803146362, 0.8039215803146362, 0.8235294222831726, 0.7843137383460999, 0.8235294222831726, 0.7843137383460999, 0.8039215803146362, 0.7843137383460999, 0.8235294222831726, 0.7843137383460999, 0.8627451062202454, 0.7843137383460999, 0.8627451062202454, 0.843137264251709, 0.8235294222831726, 0.8235294222831726, 0.843137264251709, 0.9019607901573181, 0.8823529481887817, 0.8823529481887817, 0.9019607901573181, 0.8627451062202454, 0.9411764740943909, 0.9215686321258545, 0.8823529481887817, 0.9215686321258545, 0.8823529481887817, 0.9215686321258545, 0.8823529481887817, 0.9215686321258545, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.9411764740943909, 0.9803921580314636, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0], [0.11764705926179886, 0.3333333432674408, 0.3137255012989044, 0.45098039507865906, 0.27450981736183167, 0.37254902720451355, NaN, 0.5490196347236633, 0.6176470518112183, 0.6470588445663452, 0.6274510025978088, 0.7450980544090271, 0.686274528503418, 0.7647058963775635, 0.7450980544090271, 0.7450980544090271, 0.686274528503418, 0.7647058963775635, 0.686274528503418, 0.7058823704719543, 0.8235294222831726, 0.8039215803146362, 0.6470588445663452, 0.8627451062202454, 0.8627451062202454, 0.8235294222831726, 0.7450980544090271, 0.8235294222831726, 0.7450980544090271, 0.7843137383460999, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.7647058963775635, 0.8627451062202454, 0.8627451062202454, 0.8627451062202454, 0.8627451062202454, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.9215686321258545, 0.9607843160629272, 0.8823529481887817, 0.9411764740943909, 0.8627451062202454, 0.9019607901573181, 0.9215686321258545, 0.9607843160629272, 0.9215686321258545, 0.9215686321258545, 0.9019607901573181, 0.9607843160629272, 0.8627451062202454, 0.843137264251709, 0.9607843160629272, 0.9803921580314636, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.9019607901573181, 0.9215686321258545, 0.9607843160629272, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 1.0], [0.1764705926179886, 0.2549019753932953, 0.37254902720451355, 0.45098039507865906, 0.3921568691730499, 0.3333333432674408, 0.45098039507865906, NaN, 0.45098039507865906, 0.5490196347236633, 0.529411792755127, 0.7058823704719543, 0.6078431606292725, 0.6666666865348816, 0.8235294222831726, 0.7647058963775635, 0.7843137383460999, 0.8039215803146362, 0.5882353186607361, 0.6470588445663452, 0.7450980544090271, 0.8627451062202454, 0.6470588445663452, 0.8627451062202454, 0.8235294222831726, 0.7058823704719543, 0.7058823704719543, 0.8235294222831726, 0.843137264251709, 0.8039215803146362, 0.8627451062202454, 0.8235294222831726, 0.8235294222831726, 0.7450980544090271, 0.8235294222831726, 0.8039215803146362, 0.9215686321258545, 0.8235294222831726, 0.7647058963775635, 0.8823529481887817, 0.9019607901573181, 0.9215686321258545, 0.9411764740943909, 0.8627451062202454, 0.9019607901573181, 0.9215686321258545, 0.9215686321258545, 0.8627451062202454, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272, 0.8235294222831726, 0.8235294222831726, 0.9803921580314636, 0.9803921580314636, 0.9215686321258545, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.8823529481887817, 0.9411764740943909, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0, 1.0], [0.13725490868091583, 0.3921568691730499, 0.3529411852359772, 0.3137255012989044, 0.29411765933036804, 0.4117647111415863, 0.38235294818878174, 0.5490196347236633, NaN, 0.5686274766921997, 0.6470588445663452, 0.6666666865348816, 0.6470588445663452, 0.7058823704719543, 0.7450980544090271, 0.7058823704719543, 0.6274510025978088, 0.7058823704719543, 0.6666666865348816, 0.6470588445663452, 0.7450980544090271, 0.7450980544090271, 0.7058823704719543, 0.7647058963775635, 0.7450980544090271, 0.7058823704719543, 0.843137264251709, 0.7843137383460999, 0.7450980544090271, 0.8039215803146362, 0.7450980544090271, 0.8529411554336548, 0.843137264251709, 0.7647058963775635, 0.843137264251709, 0.8039215803146362, 0.843137264251709, 0.8235294222831726, 0.843137264251709, 0.843137264251709, 0.8823529481887817, 0.8823529481887817, 0.8823529481887817, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.9215686321258545, 0.8627451062202454, 0.8627451062202454, 0.9607843160629272, 0.9215686321258545, 0.9803921580314636, 0.9215686321258545, 0.9411764740943909, 0.9019607901573181, 0.9019607901573181, 0.9803921580314636, 0.9803921580314636, 0.9215686321258545, 0.9803921580314636, 0.9411764740943909, 0.9607843160629272, 0.9215686321258545, 0.9215686321258545, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636], [0.13725490868091583, 0.1764705926179886, 0.2549019753932953, 0.27450981736183167, 0.2549019753932953, 0.3333333432674408, 0.3529411852359772, 0.45098039507865906, 0.4313725531101227, NaN, 0.5098039507865906, 0.686274528503418, 0.5882353186607361, 0.6274510025978088, 0.7058823704719543, 0.7254902124404907, 0.7450980544090271, 0.7843137383460999, 0.5490196347236633, 0.6470588445663452, 0.7058823704719543, 0.7647058963775635, 0.5686274766921997, 0.7647058963775635, 0.7843137383460999, 0.686274528503418, 0.6666666865348816, 0.8039215803146362, 0.7058823704719543, 0.7450980544090271, 0.7647058963775635, 0.8235294222831726, 0.8039215803146362, 0.7058823704719543, 0.8235294222831726, 0.8039215803146362, 0.843137264251709, 0.9215686321258545, 0.8039215803146362, 0.8039215803146362, 0.843137264251709, 0.9019607901573181, 0.843137264251709, 0.8627451062202454, 0.8627451062202454, 0.9019607901573181, 0.9215686321258545, 0.8627451062202454, 0.843137264251709, 0.9411764740943909, 0.9019607901573181, 0.9019607901573181, 0.9215686321258545, 0.9607843160629272, 0.843137264251709, 0.843137264251709, 0.9411764740943909, 0.9803921580314636, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.8823529481887817, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0], [0.09803921729326248, 0.2549019753932953, 0.29411765933036804, 0.29411765933036804, 0.3333333432674408, 0.27450981736183167, 0.37254902720451355, 0.47058823704719543, 0.3529411852359772, 0.4901960790157318, NaN, 0.5686274766921997, 0.4901960790157318, 0.529411792755127, 0.5882353186607361, 0.5490196347236633, 0.6470588445663452, 0.5882353186607361, 0.7450980544090271, 0.529411792755127, 0.5490196347236633, 0.6274510025978088, 0.6274510025978088, 0.6078431606292725, 0.6078431606292725, 0.5686274766921997, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.6666666865348816, 0.6274510025978088, 0.7450980544090271, 0.7254902124404907, 0.8039215803146362, 0.686274528503418, 0.6470588445663452, 0.7450980544090271, 0.7058823704719543, 0.7254902124404907, 0.6470588445663452, 0.7254902124404907, 0.8235294222831726, 0.7647058963775635, 0.7843137383460999, 0.7843137383460999, 0.7450980544090271, 0.7450980544090271, 0.8627451062202454, 0.7843137383460999, 0.8235294222831726, 0.8039215803146362, 0.843137264251709, 0.8039215803146362, 0.843137264251709, 0.8235294222831726, 0.7843137383460999, 0.8627451062202454, 0.9019607901573181, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.9607843160629272, 0.8823529481887817, 0.9215686321258545, 0.9215686321258545, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 0.9215686321258545, 1.0], [0.11764705926179886, 0.1568627506494522, 0.23529411852359772, 0.19607843458652496, 0.23529411852359772, 0.3137255012989044, 0.2549019753932953, 0.29411765933036804, 0.3333333432674408, 0.3137255012989044, 0.4313725531101227, NaN, 0.4117647111415863, 0.5098039507865906, 0.5882353186607361, 0.5490196347236633, 0.529411792755127, 0.5490196347236633, 0.5098039507865906, 0.529411792755127, 0.6078431606292725, 0.6666666865348816, 0.5, 0.6666666865348816, 0.6274510025978088, 0.686274528503418, 0.5686274766921997, 0.7058823704719543, 0.5686274766921997, 0.6470588445663452, 0.6274510025978088, 0.7254902124404907, 0.7058823704719543, 0.6274510025978088, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.8235294222831726, 0.686274528503418, 0.686274528503418, 0.7843137383460999, 0.7647058963775635, 0.8627451062202454, 0.7647058963775635, 0.8039215803146362, 0.843137264251709, 0.843137264251709, 0.8235294222831726, 0.843137264251709, 0.843137264251709, 0.8823529481887817, 0.8823529481887817, 0.9019607901573181, 0.9019607901573181, 0.843137264251709, 0.8627451062202454, 0.9607843160629272, 0.9215686321258545, 0.9215686321258545, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.8627451062202454, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.970588207244873, 0.9803921580314636, 0.9607843160629272, 1.0], [0.05882352963089943, 0.2549019753932953, 0.1568627506494522, 0.23529411852359772, 0.29411765933036804, 0.23529411852359772, 0.3137255012989044, 0.3921568691730499, 0.3529411852359772, 0.4117647111415863, 0.5098039507865906, 0.5882353186607361, NaN, 0.47058823704719543, 0.5490196347236633, 0.5490196347236633, 0.6078431606292725, 0.5882353186607361, 0.5490196347236633, 0.5098039507865906, 0.529411792755127, 0.6666666865348816, 0.6078431606292725, 0.5882353186607361, 0.6470588445663452, 0.5882353186607361, 0.6960784196853638, 0.686274528503418, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7450980544090271, 0.7450980544090271, 0.7254902124404907, 0.7058823704719543, 0.6666666865348816, 0.7450980544090271, 0.7058823704719543, 0.7058823704719543, 0.686274528503418, 0.7058823704719543, 0.7254902124404907, 0.8235294222831726, 0.8235294222831726, 0.7647058963775635, 0.7843137383460999, 0.7843137383460999, 0.843137264251709, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.9019607901573181, 0.843137264251709, 0.843137264251709, 0.8039215803146362, 0.7843137383460999, 0.8823529481887817, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.8627451062202454, 0.9411764740943909, 0.9607843160629272, 0.8823529481887817, 0.8823529481887817, 0.9803921580314636, 0.9215686321258545, 1.0], [0.05882352963089943, 0.13725490868091583, 0.1568627506494522, 0.21568627655506134, 0.27450981736183167, 0.27450981736183167, 0.23529411852359772, 0.3333333432674408, 0.29411765933036804, 0.37254902720451355, 0.47058823704719543, 0.4901960790157318, 0.529411792755127, NaN, 0.5882353186607361, 0.4901960790157318, 0.529411792755127, 0.5490196347236633, 0.5098039507865906, 0.5490196347236633, 0.6274510025978088, 0.5882353186607361, 0.4901960790157318, 0.6274510025978088, 0.6470588445663452, 0.6078431606292725, 0.5882353186607361, 0.686274528503418, 0.6078431606292725, 0.5686274766921997, 0.7254902124404907, 0.6470588445663452, 0.5686274766921997, 0.6470588445663452, 0.8823529481887817, 0.7843137383460999, 0.7450980544090271, 0.8235294222831726, 0.7058823704719543, 0.7450980544090271, 0.7647058963775635, 0.7450980544090271, 0.8039215803146362, 0.7450980544090271, 0.8039215803146362, 0.7843137383460999, 0.8235294222831726, 0.7450980544090271, 0.8039215803146362, 0.8627451062202454, 0.8039215803146362, 0.8823529481887817, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.8627451062202454, 0.9803921580314636, 0.9019607901573181, 0.843137264251709, 0.8627451062202454, 0.8823529481887817, 0.9411764740943909, 0.8823529481887817, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 1.0, 0.9607843160629272, 1.0], [0.05882352963089943, 0.0784313753247261, 0.11764705926179886, 0.11764705926179886, 0.1568627506494522, 0.1764705926179886, 0.2549019753932953, 0.1764705926179886, 0.2549019753932953, 0.29411765933036804, 0.4117647111415863, 0.4117647111415863, 0.45098039507865906, 0.4117647111415863, NaN, 0.5490196347236633, 0.5098039507865906, 0.5686274766921997, 0.45098039507865906, 0.47058823704719543, 0.4901960790157318, 0.5686274766921997, 0.45098039507865906, 0.7450980544090271, 0.7254902124404907, 0.5490196347236633, 0.5490196347236633, 0.6666666865348816, 0.6470588445663452, 0.5882353186607361, 0.7450980544090271, 0.6666666865348816, 0.686274528503418, 0.6078431606292725, 0.7254902124404907, 0.7450980544090271, 0.7647058963775635, 0.7254902124404907, 0.6666666865348816, 0.8039215803146362, 0.8235294222831726, 0.8039215803146362, 0.8039215803146362, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.8235294222831726, 0.8039215803146362, 0.8627451062202454, 0.9607843160629272, 0.9411764740943909, 0.8627451062202454, 0.9215686321258545, 0.9019607901573181, 0.843137264251709, 0.8235294222831726, 0.9215686321258545, 0.9215686321258545, 0.843137264251709, 0.9215686321258545, 0.9803921580314636, 0.9215686321258545, 0.8627451062202454, 0.9411764740943909, 0.9215686321258545, 0.9803921580314636, 0.9803921580314636, 1.0, 0.9607843160629272, 1.0], [0.0784313753247261, 0.11764705926179886, 0.23529411852359772, 0.19607843458652496, 0.1764705926179886, 0.23529411852359772, 0.2549019753932953, 0.23529411852359772, 0.29411765933036804, 0.27450981736183167, 0.45098039507865906, 0.45098039507865906, 0.45098039507865906, 0.5098039507865906, 0.45098039507865906, NaN, 0.4901960790157318, 0.4117647111415863, 0.47058823704719543, 0.5490196347236633, 0.529411792755127, 0.6078431606292725, 0.529411792755127, 0.5490196347236633, 0.686274528503418, 0.5686274766921997, 0.529411792755127, 0.6666666865348816, 0.5686274766921997, 0.5882353186607361, 0.6470588445663452, 0.7450980544090271, 0.686274528503418, 0.5882353186607361, 0.686274528503418, 0.7254902124404907, 0.7647058963775635, 0.7647058963775635, 0.7058823704719543, 0.7843137383460999, 0.7647058963775635, 0.7254902124404907, 0.8235294222831726, 0.7843137383460999, 0.8235294222831726, 0.8627451062202454, 0.7843137383460999, 0.7647058963775635, 0.843137264251709, 0.843137264251709, 0.9019607901573181, 0.843137264251709, 0.8823529481887817, 0.8823529481887817, 0.8235294222831726, 0.8235294222831726, 0.9411764740943909, 0.9411764740943909, 0.843137264251709, 0.8823529481887817, 0.9411764740943909, 0.9215686321258545, 0.843137264251709, 0.9215686321258545, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0], [0.13725490868091583, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.21568627655506134, 0.21568627655506134, 0.3137255012989044, 0.21568627655506134, 0.37254902720451355, 0.2549019753932953, 0.3529411852359772, 0.47058823704719543, 0.3921568691730499, 0.47058823704719543, 0.4901960790157318, 0.5098039507865906, NaN, 0.47058823704719543, 0.4117647111415863, 0.47058823704719543, 0.529411792755127, 0.5882353186607361, 0.4803921580314636, 0.5686274766921997, 0.529411792755127, 0.6078431606292725, 0.529411792755127, 0.6666666865348816, 0.6470588445663452, 0.5686274766921997, 0.6666666865348816, 0.7058823704719543, 0.6666666865348816, 0.5882353186607361, 0.6470588445663452, 0.6666666865348816, 0.7647058963775635, 0.7254902124404907, 0.7254902124404907, 0.7450980544090271, 0.686274528503418, 0.7647058963775635, 0.843137264251709, 0.843137264251709, 0.7647058963775635, 0.8235294222831726, 0.8627451062202454, 0.8235294222831726, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.8039215803146362, 0.8039215803146362, 0.9607843160629272, 0.9411764740943909, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.8627451062202454, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 1.0], [0.05882352963089943, 0.09803921729326248, 0.19607843458652496, 0.1568627506494522, 0.13725490868091583, 0.23529411852359772, 0.23529411852359772, 0.19607843458652496, 0.29411765933036804, 0.21568627655506134, 0.4117647111415863, 0.45098039507865906, 0.4117647111415863, 0.45098039507865906, 0.4313725531101227, 0.5882353186607361, 0.529411792755127, NaN, 0.45098039507865906, 0.5490196347236633, 0.47058823704719543, 0.5882353186607361, 0.4901960790157318, 0.5686274766921997, 0.7058823704719543, 0.5588235259056091, 0.5098039507865906, 0.6470588445663452, 0.5882353186607361, 0.5490196347236633, 0.686274528503418, 0.7450980544090271, 0.686274528503418, 0.5686274766921997, 0.6274510025978088, 0.686274528503418, 0.7450980544090271, 0.7647058963775635, 0.7450980544090271, 0.8627451062202454, 0.7450980544090271, 0.7058823704719543, 0.8235294222831726, 0.8235294222831726, 0.8235294222831726, 0.8627451062202454, 0.7843137383460999, 0.8235294222831726, 0.8823529481887817, 0.843137264251709, 0.9215686321258545, 0.8823529481887817, 0.8823529481887817, 0.9215686321258545, 0.8235294222831726, 0.8235294222831726, 0.9411764740943909, 0.9215686321258545, 0.8823529481887817, 0.9019607901573181, 0.9607843160629272, 0.9215686321258545, 0.843137264251709, 0.9019607901573181, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 1.0], [0.03921568766236305, 0.23529411852359772, 0.1764705926179886, 0.27450981736183167, 0.27450981736183167, 0.2549019753932953, 0.3137255012989044, 0.4117647111415863, 0.3333333432674408, 0.45098039507865906, 0.2549019753932953, 0.4901960790157318, 0.45098039507865906, 0.4901960790157318, 0.5490196347236633, 0.529411792755127, 0.5882353186607361, 0.5490196347236633, NaN, 0.5098039507865906, 0.5098039507865906, 0.6078431606292725, 0.5490196347236633, 0.5882353186607361, 0.5882353186607361, 0.5098039507865906, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.6470588445663452, 0.7058823704719543, 0.6666666865348816, 0.813725471496582, 0.6274510025978088, 0.6274510025978088, 0.6470588445663452, 0.6470588445663452, 0.6470588445663452, 0.686274528503418, 0.6470588445663452, 0.7647058963775635, 0.7450980544090271, 0.7254902124404907, 0.7254902124404907, 0.686274528503418, 0.6666666865348816, 0.8235294222831726, 0.686274528503418, 0.7254902124404907, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.843137264251709, 0.8039215803146362, 0.7843137383460999, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.9411764740943909, 0.8823529481887817, 0.9607843160629272, 0.8627451062202454, 0.8823529481887817, 0.9411764740943909, 0.8627451062202454, 0.8627451062202454, 0.9607843160629272, 0.9019607901573181, 0.9803921580314636], [0.11764705926179886, 0.13725490868091583, 0.09803921729326248, 0.13725490868091583, 0.23529411852359772, 0.1764705926179886, 0.29411765933036804, 0.3529411852359772, 0.3529411852359772, 0.3529411852359772, 0.47058823704719543, 0.47058823704719543, 0.4901960790157318, 0.45098039507865906, 0.529411792755127, 0.45098039507865906, 0.529411792755127, 0.45098039507865906, 0.4901960790157318, NaN, 0.4901960790157318, 0.5882353186607361, 0.47058823704719543, 0.6666666865348816, 0.5882353186607361, 0.6764705777168274, 0.5686274766921997, 0.5882353186607361, 0.6078431606292725, 0.5882353186607361, 0.6274510025978088, 0.6078431606292725, 0.6666666865348816, 0.6078431606292725, 0.6078431606292725, 0.6666666865348816, 0.6470588445663452, 0.7254902124404907, 0.6274510025978088, 0.6666666865348816, 0.6666666865348816, 0.8039215803146362, 0.7843137383460999, 0.7647058963775635, 0.7254902124404907, 0.7843137383460999, 0.8039215803146362, 0.7647058963775635, 0.7843137383460999, 0.843137264251709, 0.7647058963775635, 0.8627451062202454, 0.8235294222831726, 0.843137264251709, 0.7450980544090271, 0.7647058963775635, 0.843137264251709, 0.843137264251709, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.8823529481887817, 0.8627451062202454, 0.8921568393707275, 0.9019607901573181, 0.8627451062202454, 0.8823529481887817, 0.9019607901573181, 0.8823529481887817, 0.9803921580314636], [0.03921568766236305, 0.1568627506494522, 0.09803921729326248, 0.1764705926179886, 0.19607843458652496, 0.2549019753932953, 0.1764705926179886, 0.2549019753932953, 0.2549019753932953, 0.29411765933036804, 0.45098039507865906, 0.3921568691730499, 0.47058823704719543, 0.37254902720451355, 0.5098039507865906, 0.47058823704719543, 0.47058823704719543, 0.529411792755127, 0.4901960790157318, 0.5098039507865906, NaN, 0.5490196347236633, 0.45098039507865906, 0.5490196347236633, 0.6274510025978088, 0.5686274766921997, 0.5490196347236633, 0.6078431606292725, 0.6764705777168274, 0.5098039507865906, 0.6666666865348816, 0.6274510025978088, 0.5490196347236633, 0.6078431606292725, 0.7647058963775635, 0.686274528503418, 0.6666666865348816, 0.7058823704719543, 0.6470588445663452, 0.6666666865348816, 0.7254902124404907, 0.7254902124404907, 0.8039215803146362, 0.7254902124404907, 0.7254902124404907, 0.7843137383460999, 0.7450980544090271, 0.7254902124404907, 0.7450980544090271, 0.8235294222831726, 0.7843137383460999, 0.8627451062202454, 0.7843137383460999, 0.8823529481887817, 0.8235294222831726, 0.8235294222831726, 0.9803921580314636, 0.9019607901573181, 0.8235294222831726, 0.843137264251709, 0.9019607901573181, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636], [0.0784313753247261, 0.05882352963089943, 0.13725490868091583, 0.1764705926179886, 0.13725490868091583, 0.21568627655506134, 0.19607843458652496, 0.13725490868091583, 0.2549019753932953, 0.23529411852359772, 0.37254902720451355, 0.3333333432674408, 0.3333333432674408, 0.4117647111415863, 0.4313725531101227, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.3921568691730499, 0.4117647111415863, 0.45098039507865906, NaN, 0.4313725531101227, 0.5098039507865906, 0.529411792755127, 0.4901960790157318, 0.47058823704719543, 0.6274510025978088, 0.529411792755127, 0.529411792755127, 0.6078431606292725, 0.6274510025978088, 0.6274510025978088, 0.5686274766921997, 0.6470588445663452, 0.686274528503418, 0.7254902124404907, 0.7254902124404907, 0.6078431606292725, 0.7450980544090271, 0.7450980544090271, 0.7058823704719543, 0.7352941036224365, 0.8235294222831726, 0.7843137383460999, 0.8823529481887817, 0.7843137383460999, 0.8039215803146362, 0.8627451062202454, 0.9803921580314636, 0.9019607901573181, 0.8235294222831726, 0.8823529481887817, 0.843137264251709, 0.843137264251709, 0.843137264251709, 0.9215686321258545, 0.8823529481887817, 0.8235294222831726, 0.9215686321258545, 0.9607843160629272, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 1.0], [0.09803921729326248, 0.23529411852359772, 0.23529411852359772, 0.19607843458652496, 0.2549019753932953, 0.2549019753932953, 0.3529411852359772, 0.3529411852359772, 0.29411765933036804, 0.4313725531101227, 0.37254902720451355, 0.5, 0.3921568691730499, 0.5098039507865906, 0.5490196347236633, 0.47058823704719543, 0.5196078419685364, 0.5098039507865906, 0.45098039507865906, 0.529411792755127, 0.5490196347236633, 0.5686274766921997, NaN, 0.5686274766921997, 0.5686274766921997, 0.5686274766921997, 0.4901960790157318, 0.5882353186607361, 0.6078431606292725, 0.529411792755127, 0.6470588445663452, 0.6078431606292725, 0.6470588445663452, 0.5490196347236633, 0.6078431606292725, 0.686274528503418, 0.6666666865348816, 0.7058823704719543, 0.6274510025978088, 0.6666666865348816, 0.7254902124404907, 0.686274528503418, 0.6666666865348816, 0.6666666865348816, 0.7450980544090271, 0.7058823704719543, 0.686274528503418, 0.6666666865348816, 0.7058823704719543, 0.686274528503418, 0.7450980544090271, 0.7843137383460999, 0.7647058963775635, 0.7450980544090271, 0.7745097875595093, 0.7647058963775635, 0.8039215803146362, 0.7647058963775635, 0.8235294222831726, 0.8823529481887817, 0.8823529481887817, 0.8627451062202454, 0.843137264251709, 0.8823529481887817, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.8823529481887817, 0.9215686321258545, 0.9803921580314636], [0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.23529411852359772, 0.23529411852359772, 0.3921568691730499, 0.3333333432674408, 0.4117647111415863, 0.37254902720451355, 0.2549019753932953, 0.45098039507865906, 0.4313725531101227, 0.4313725531101227, 0.4117647111415863, 0.3333333432674408, 0.45098039507865906, 0.4901960790157318, 0.4313725531101227, NaN, 0.5490196347236633, 0.5098039507865906, 0.47058823704719543, 0.529411792755127, 0.5490196347236633, 0.5490196347236633, 0.6470588445663452, 0.5686274766921997, 0.6078431606292725, 0.5686274766921997, 0.7254902124404907, 0.7254902124404907, 0.6666666865348816, 0.686274528503418, 0.6274510025978088, 0.7058823704719543, 0.7450980544090271, 0.6470588445663452, 0.7647058963775635, 0.7254902124404907, 0.7843137383460999, 0.8235294222831726, 0.7843137383460999, 0.7647058963775635, 0.7843137383460999, 0.8627451062202454, 0.8823529481887817, 0.843137264251709, 0.8627451062202454, 0.8823529481887817, 0.8039215803146362, 0.8235294222831726, 0.8823529481887817, 0.9019607901573181, 0.843137264251709, 0.843137264251709, 0.9411764740943909, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9215686321258545, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 1.0], [0.09803921729326248, 0.05882352963089943, 0.13725490868091583, 0.09803921729326248, 0.09803921729326248, 0.19607843458652496, 0.13725490868091583, 0.1764705926179886, 0.2549019753932953, 0.21568627655506134, 0.3921568691730499, 0.37254902720451355, 0.3529411852359772, 0.3529411852359772, 0.27450981736183167, 0.3137255012989044, 0.47058823704719543, 0.29411765933036804, 0.4117647111415863, 0.4117647111415863, 0.37254902720451355, 0.47058823704719543, 0.4313725531101227, 0.45098039507865906, NaN, 0.45098039507865906, 0.5098039507865906, 0.4901960790157318, 0.47058823704719543, 0.5686274766921997, 0.6078431606292725, 0.6470588445663452, 0.6274510025978088, 0.5882353186607361, 0.5882353186607361, 0.6470588445663452, 0.7058823704719543, 0.7450980544090271, 0.6274510025978088, 0.7843137383460999, 0.686274528503418, 0.6274510025978088, 0.7647058963775635, 0.7254902124404907, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.7254902124404907, 0.8039215803146362, 0.7843137383460999, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.843137264251709, 0.8235294222831726, 0.9019607901573181, 0.9411764740943909, 0.8627451062202454, 0.8627451062202454, 0.9215686321258545, 0.9411764740943909, 0.8627451062202454, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0], [0.0784313753247261, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.13725490868091583, 0.19607843458652496, 0.1764705926179886, 0.29411765933036804, 0.29411765933036804, 0.3137255012989044, 0.4313725531101227, 0.3137255012989044, 0.4117647111415863, 0.3921568691730499, 0.45098039507865906, 0.4313725531101227, 0.3921568691730499, 0.44117647409439087, 0.4901960790157318, 0.3235294222831726, 0.4313725531101227, 0.5098039507865906, 0.4313725531101227, 0.4901960790157318, 0.5490196347236633, NaN, 0.5098039507865906, 0.4901960790157318, 0.5098039507865906, 0.529411792755127, 0.5490196347236633, 0.5686274766921997, 0.6274510025978088, 0.5882353186607361, 0.6078431606292725, 0.5686274766921997, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6470588445663452, 0.686274528503418, 0.7647058963775635, 0.7647058963775635, 0.6666666865348816, 0.7058823704719543, 0.7450980544090271, 0.7156862616539001, 0.686274528503418, 0.7254902124404907, 0.7843137383460999, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.8235294222831726, 0.7843137383460999, 0.7843137383460999, 0.8627451062202454, 0.843137264251709, 0.843137264251709, 0.8235294222831726, 0.9019607901573181, 0.9019607901573181, 0.8823529481887817, 0.9215686321258545, 0.8823529481887817, 0.9215686321258545, 0.9411764740943909, 0.9411764740943909, 0.9411764740943909, 1.0], [0.11764705926179886, 0.1568627506494522, 0.11764705926179886, 0.1764705926179886, 0.1764705926179886, 0.1764705926179886, 0.2549019753932953, 0.29411765933036804, 0.1568627506494522, 0.3333333432674408, 0.37254902720451355, 0.4313725531101227, 0.30392158031463623, 0.4117647111415863, 0.45098039507865906, 0.47058823704719543, 0.47058823704719543, 0.4901960790157318, 0.37254902720451355, 0.4313725531101227, 0.45098039507865906, 0.529411792755127, 0.5098039507865906, 0.529411792755127, 0.4901960790157318, 0.4901960790157318, NaN, 0.5490196347236633, 0.4901960790157318, 0.5686274766921997, 0.529411792755127, 0.5686274766921997, 0.6274510025978088, 0.5686274766921997, 0.5882353186607361, 0.6470588445663452, 0.5686274766921997, 0.6470588445663452, 0.7058823704719543, 0.7058823704719543, 0.6470588445663452, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7254902124404907, 0.6666666865348816, 0.8235294222831726, 0.7058823704719543, 0.686274528503418, 0.7647058963775635, 0.7058823704719543, 0.8039215803146362, 0.7843137383460999, 0.8039215803146362, 0.7843137383460999, 0.7843137383460999, 0.8235294222831726, 0.843137264251709, 0.8235294222831726, 0.8823529481887817, 0.8235294222831726, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.8823529481887817, 0.9019607901573181, 0.9803921580314636, 0.9019607901573181, 1.0], [0.11764705926179886, 0.05882352963089943, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.21568627655506134, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.19607843458652496, 0.37254902720451355, 0.29411765933036804, 0.3137255012989044, 0.3137255012989044, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3529411852359772, 0.37254902720451355, 0.4117647111415863, 0.3921568691730499, 0.37254902720451355, 0.4117647111415863, 0.47058823704719543, 0.5098039507865906, 0.5098039507865906, 0.45098039507865906, NaN, 0.529411792755127, 0.5686274766921997, 0.5490196347236633, 0.5882353186607361, 0.5882353186607361, 0.529411792755127, 0.5882353186607361, 0.6274510025978088, 0.6470588445663452, 0.6666666865348816, 0.6078431606292725, 0.686274528503418, 0.7450980544090271, 0.686274528503418, 0.7254902124404907, 0.8235294222831726, 0.7647058963775635, 0.8627451062202454, 0.7647058963775635, 0.8039215803146362, 0.8235294222831726, 0.9019607901573181, 0.8627451062202454, 0.843137264251709, 0.8627451062202454, 0.8039215803146362, 0.843137264251709, 0.843137264251709, 0.9019607901573181, 0.8823529481887817, 0.7843137383460999, 0.9019607901573181, 0.9411764740943909, 0.9019607901573181, 0.8627451062202454, 0.9215686321258545, 0.9019607901573181, 0.9215686321258545, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272, 1.0], [0.0784313753247261, 0.05882352963089943, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.2549019753932953, 0.1568627506494522, 0.2549019753932953, 0.29411765933036804, 0.37254902720451355, 0.4313725531101227, 0.3333333432674408, 0.3921568691730499, 0.3529411852359772, 0.4313725531101227, 0.3529411852359772, 0.4117647111415863, 0.37254902720451355, 0.3921568691730499, 0.3235294222831726, 0.47058823704719543, 0.3921568691730499, 0.45098039507865906, 0.529411792755127, 0.4901960790157318, 0.5098039507865906, 0.47058823704719543, NaN, 0.5490196347236633, 0.6078431606292725, 0.5882353186607361, 0.6470588445663452, 0.5490196347236633, 0.6078431606292725, 0.686274528503418, 0.6274510025978088, 0.7058823704719543, 0.6078431606292725, 0.843137264251709, 0.7254902124404907, 0.6078431606292725, 0.7058823704719543, 0.6666666865348816, 0.7254902124404907, 0.7254902124404907, 0.7647058963775635, 0.686274528503418, 0.7647058963775635, 0.7450980544090271, 0.8039215803146362, 0.7450980544090271, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.8039215803146362, 0.8235294222831726, 0.8627451062202454, 0.8039215803146362, 0.7843137383460999, 0.8627451062202454, 0.8627451062202454, 0.843137264251709, 0.8627451062202454, 0.8627451062202454, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 1.0], [0.05882352963089943, 0.0784313753247261, 0.13725490868091583, 0.1568627506494522, 0.1568627506494522, 0.21568627655506134, 0.21568627655506134, 0.19607843458652496, 0.19607843458652496, 0.2549019753932953, 0.3333333432674408, 0.3529411852359772, 0.3333333432674408, 0.4313725531101227, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.45098039507865906, 0.37254902720451355, 0.4117647111415863, 0.4901960790157318, 0.47058823704719543, 0.47058823704719543, 0.45098039507865906, 0.4313725531101227, 0.47058823704719543, 0.4313725531101227, 0.4313725531101227, 0.45098039507865906, NaN, 0.529411792755127, 0.529411792755127, 0.686274528503418, 0.47058823704719543, 0.6078431606292725, 0.5882353186607361, 0.6078431606292725, 0.6470588445663452, 0.5882353186607361, 0.529411792755127, 0.6274510025978088, 0.7450980544090271, 0.6666666865348816, 0.7450980544090271, 0.6470588445663452, 0.686274528503418, 0.7254902124404907, 0.7843137383460999, 0.8039215803146362, 0.7450980544090271, 0.8235294222831726, 0.7843137383460999, 0.7647058963775635, 0.8039215803146362, 0.7647058963775635, 0.7647058963775635, 0.8823529481887817, 0.843137264251709, 0.9215686321258545, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.8235294222831726, 0.9411764740943909, 0.8823529481887817, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636], [0.09803921729326248, 0.05882352963089943, 0.13725490868091583, 0.13725490868091583, 0.11764705926179886, 0.19607843458652496, 0.1764705926179886, 0.13725490868091583, 0.2549019753932953, 0.23529411852359772, 0.37254902720451355, 0.37254902720451355, 0.3333333432674408, 0.27450981736183167, 0.2549019753932953, 0.3529411852359772, 0.3333333432674408, 0.3137255012989044, 0.3529411852359772, 0.37254902720451355, 0.3333333432674408, 0.3921568691730499, 0.3529411852359772, 0.3529411852359772, 0.3921568691730499, 0.45098039507865906, 0.47058823704719543, 0.45098039507865906, 0.3921568691730499, 0.47058823704719543, NaN, 0.47058823704719543, 0.5882353186607361, 0.5098039507865906, 0.529411792755127, 0.6274510025978088, 0.5882353186607361, 0.7058823704719543, 0.5882353186607361, 0.7254902124404907, 0.6666666865348816, 0.5686274766921997, 0.6274510025978088, 0.6274510025978088, 0.7450980544090271, 0.7450980544090271, 0.7254902124404907, 0.6666666865348816, 0.7254902124404907, 0.7647058963775635, 0.8039215803146362, 0.7647058963775635, 0.7843137383460999, 0.8039215803146362, 0.7843137383460999, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.8039215803146362, 0.8039215803146362, 0.8039215803146362, 0.843137264251709, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 1.0], [0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.09803921729326248, 0.1568627506494522, 0.21568627655506134, 0.13725490868091583, 0.1764705926179886, 0.14705882966518402, 0.1764705926179886, 0.2549019753932953, 0.27450981736183167, 0.2549019753932953, 0.3529411852359772, 0.3333333432674408, 0.2549019753932953, 0.29411765933036804, 0.2549019753932953, 0.29411765933036804, 0.3921568691730499, 0.37254902720451355, 0.37254902720451355, 0.3921568691730499, 0.4313725531101227, 0.3529411852359772, 0.4313725531101227, 0.4313725531101227, 0.4117647111415863, 0.4117647111415863, 0.47058823704719543, 0.529411792755127, NaN, 0.529411792755127, 0.4901960790157318, 0.529411792755127, 0.5490196347236633, 0.5882353186607361, 0.6078431606292725, 0.6470588445663452, 0.529411792755127, 0.6470588445663452, 0.6274510025978088, 0.6470588445663452, 0.686274528503418, 0.7450980544090271, 0.6078431606292725, 0.7843137383460999, 0.7450980544090271, 0.8039215803146362, 0.7450980544090271, 0.8039215803146362, 0.8039215803146362, 0.7745097875595093, 0.7647058963775635, 0.7843137383460999, 0.7843137383460999, 0.9215686321258545, 0.843137264251709, 0.8823529481887817, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.8627451062202454, 0.9411764740943909, 0.9019607901573181, 0.8823529481887817, 0.9215686321258545, 0.9803921580314636, 0.9215686321258545, 1.0], [0.05882352963089943, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.11764705926179886, 0.1764705926179886, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.19607843458652496, 0.27450981736183167, 0.29411765933036804, 0.2549019753932953, 0.4313725531101227, 0.3137255012989044, 0.3137255012989044, 0.3333333432674408, 0.3137255012989044, 0.3333333432674408, 0.3333333432674408, 0.45098039507865906, 0.37254902720451355, 0.3529411852359772, 0.3921568691730499, 0.37254902720451355, 0.37254902720451355, 0.37254902720451355, 0.4117647111415863, 0.3529411852359772, 0.3137255012989044, 0.4117647111415863, 0.47058823704719543, NaN, 0.4313725531101227, 0.5098039507865906, 0.5098039507865906, 0.5490196347236633, 0.5882353186607361, 0.6078431606292725, 0.5098039507865906, 0.5882353186607361, 0.6274510025978088, 0.6666666865348816, 0.7254902124404907, 0.5686274766921997, 0.6078431606292725, 0.686274528503418, 0.7450980544090271, 0.7450980544090271, 0.686274528503418, 0.7843137383460999, 0.7450980544090271, 0.7647058963775635, 0.7450980544090271, 0.7450980544090271, 0.7647058963775635, 0.9019607901573181, 0.8235294222831726, 0.9607843160629272, 0.9215686321258545, 0.8627451062202454, 0.8823529481887817, 0.8235294222831726, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.9019607901573181, 0.9803921580314636, 0.9215686321258545, 1.0], [0.019607843831181526, 0.1764705926179886, 0.0784313753247261, 0.21568627655506134, 0.21568627655506134, 0.21568627655506134, 0.23529411852359772, 0.2549019753932953, 0.23529411852359772, 0.29411765933036804, 0.19607843458652496, 0.37254902720451355, 0.27450981736183167, 0.3529411852359772, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.18627451360225677, 0.3921568691730499, 0.3921568691730499, 0.4313725531101227, 0.45098039507865906, 0.4313725531101227, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.47058823704719543, 0.45098039507865906, 0.529411792755127, 0.4901960790157318, 0.5098039507865906, 0.5686274766921997, NaN, 0.5098039507865906, 0.45098039507865906, 0.5098039507865906, 0.4901960790157318, 0.529411792755127, 0.5686274766921997, 0.5098039507865906, 0.6470588445663452, 0.5686274766921997, 0.6078431606292725, 0.5686274766921997, 0.5882353186607361, 0.6470588445663452, 0.6666666865348816, 0.6666666865348816, 0.686274528503418, 0.7058823704719543, 0.7843137383460999, 0.686274528503418, 0.7450980544090271, 0.7058823704719543, 0.7254902124404907, 0.7843137383460999, 0.8235294222831726, 0.8039215803146362, 0.843137264251709, 0.7843137383460999, 0.8627451062202454, 0.7647058963775635, 0.8235294222831726, 0.8627451062202454, 0.7843137383460999, 0.8039215803146362, 0.9019607901573181, 0.8627451062202454, 0.9803921580314636], [0.05882352963089943, 0.09803921729326248, 0.11764705926179886, 0.11764705926179886, 0.11764705926179886, 0.13725490868091583, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.1764705926179886, 0.3137255012989044, 0.23529411852359772, 0.29411765933036804, 0.11764705926179886, 0.27450981736183167, 0.3137255012989044, 0.3529411852359772, 0.37254902720451355, 0.37254902720451355, 0.3921568691730499, 0.23529411852359772, 0.3529411852359772, 0.3921568691730499, 0.27450981736183167, 0.4117647111415863, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.3921568691730499, 0.3921568691730499, 0.47058823704719543, 0.47058823704719543, 0.4901960790157318, 0.4901960790157318, NaN, 0.5490196347236633, 0.5686274766921997, 0.5882353186607361, 0.5686274766921997, 0.5686274766921997, 0.6078431606292725, 0.6078431606292725, 0.686274528503418, 0.6274510025978088, 0.5980392098426819, 0.7254902124404907, 0.7450980544090271, 0.6666666865348816, 0.7058823704719543, 0.7254902124404907, 0.7254902124404907, 0.843137264251709, 0.7254902124404907, 0.843137264251709, 0.7647058963775635, 0.7843137383460999, 0.8823529481887817, 0.843137264251709, 0.7450980544090271, 0.8039215803146362, 0.8235294222831726, 0.843137264251709, 0.8627451062202454, 0.843137264251709, 0.9019607901573181, 0.9215686321258545, 0.9019607901573181, 0.9803921580314636, 0.9215686321258545, 0.9607843160629272], [0.05882352963089943, 0.11764705926179886, 0.1764705926179886, 0.09803921729326248, 0.11764705926179886, 0.21568627655506134, 0.13725490868091583, 0.19607843458652496, 0.19607843458652496, 0.19607843458652496, 0.3529411852359772, 0.23529411852359772, 0.3333333432674408, 0.21568627655506134, 0.2549019753932953, 0.27450981736183167, 0.3333333432674408, 0.3137255012989044, 0.37254902720451355, 0.3333333432674408, 0.3137255012989044, 0.3137255012989044, 0.3137255012989044, 0.27450981736183167, 0.3529411852359772, 0.4313725531101227, 0.3529411852359772, 0.37254902720451355, 0.3137255012989044, 0.4117647111415863, 0.37254902720451355, 0.45098039507865906, 0.4901960790157318, 0.5490196347236633, 0.45098039507865906, NaN, 0.5098039507865906, 0.5490196347236633, 0.5882353186607361, 0.45098039507865906, 0.7843137383460999, 0.6666666865348816, 0.6078431606292725, 0.6470588445663452, 0.843137264251709, 0.6078431606292725, 0.686274528503418, 0.6078431606292725, 0.6274510025978088, 0.5882353186607361, 0.6470588445663452, 0.7450980544090271, 0.7058823704719543, 0.6666666865348816, 0.8039215803146362, 0.8039215803146362, 0.843137264251709, 0.7254902124404907, 0.7647058963775635, 0.7254902124404907, 0.7450980544090271, 0.8627451062202454, 0.8627451062202454, 0.8823529481887817, 0.843137264251709, 0.9607843160629272, 0.9607843160629272, 0.9215686321258545, 0.9803921580314636, 0.9607843160629272], [0.03921568766236305, 0.05882352963089943, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.2549019753932953, 0.19607843458652496, 0.2549019753932953, 0.2549019753932953, 0.23529411852359772, 0.23529411852359772, 0.23529411852359772, 0.2549019753932953, 0.3529411852359772, 0.3529411852359772, 0.3333333432674408, 0.27450981736183167, 0.3333333432674408, 0.3333333432674408, 0.29411765933036804, 0.3333333432674408, 0.4313725531101227, 0.3529411852359772, 0.37254902720451355, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.45098039507865906, 0.4901960790157318, 0.4313725531101227, 0.4901960790157318, NaN, 0.47058823704719543, 0.5098039507865906, 0.45098039507865906, 0.529411792755127, 0.5098039507865906, 0.7450980544090271, 0.6274510025978088, 0.5490196347236633, 0.6470588445663452, 0.6666666865348816, 0.6470588445663452, 0.686274528503418, 0.686274528503418, 0.7450980544090271, 0.7843137383460999, 0.7647058963775635, 0.7058823704719543, 0.7450980544090271, 0.7647058963775635, 0.9607843160629272, 0.7647058963775635, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.8823529481887817, 0.8039215803146362, 0.8823529481887817, 0.9019607901573181, 0.9215686321258545, 0.9411764740943909, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272], [0.0784313753247261, 0.09803921729326248, 0.05882352963089943, 0.0784313753247261, 0.11764705926179886, 0.1568627506494522, 0.13725490868091583, 0.1764705926179886, 0.1764705926179886, 0.0784313753247261, 0.29411765933036804, 0.1764705926179886, 0.29411765933036804, 0.1764705926179886, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.23529411852359772, 0.3529411852359772, 0.27450981736183167, 0.29411765933036804, 0.27450981736183167, 0.29411765933036804, 0.3137255012989044, 0.2549019753932953, 0.3333333432674408, 0.3529411852359772, 0.3333333432674408, 0.29411765933036804, 0.3529411852359772, 0.29411765933036804, 0.3921568691730499, 0.4117647111415863, 0.5098039507865906, 0.4117647111415863, 0.45098039507865906, 0.529411792755127, NaN, 0.4901960790157318, 0.47058823704719543, 0.5882353186607361, 0.5882353186607361, 0.6078431606292725, 0.6078431606292725, 0.5882353186607361, 0.6666666865348816, 0.6470588445663452, 0.6666666865348816, 0.686274528503418, 0.6078431606292725, 0.7058823704719543, 0.7450980544090271, 0.7254902124404907, 0.7058823704719543, 0.7254902124404907, 0.7647058963775635, 0.8627451062202454, 0.7647058963775635, 0.7647058963775635, 0.8235294222831726, 0.8039215803146362, 0.9411764740943909, 0.8823529481887817, 0.8823529481887817, 0.9215686321258545, 0.9215686321258545, 0.9411764740943909, 0.9607843160629272, 0.9803921580314636, 1.0], [0.0784313753247261, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.11764705926179886, 0.1764705926179886, 0.1568627506494522, 0.23529411852359772, 0.1568627506494522, 0.19607843458652496, 0.27450981736183167, 0.3137255012989044, 0.29411765933036804, 0.29411765933036804, 0.3333333432674408, 0.29411765933036804, 0.27450981736183167, 0.2549019753932953, 0.3529411852359772, 0.37254902720451355, 0.3529411852359772, 0.3921568691730499, 0.37254902720451355, 0.37254902720451355, 0.37254902720451355, 0.3333333432674408, 0.29411765933036804, 0.3921568691730499, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.3529411852359772, 0.3921568691730499, 0.47058823704719543, 0.4313725531101227, 0.4117647111415863, 0.4901960790157318, 0.5098039507865906, NaN, 0.47058823704719543, 0.5098039507865906, 0.5490196347236633, 0.5490196347236633, 0.5490196347236633, 0.5686274766921997, 0.529411792755127, 0.6470588445663452, 0.6274510025978088, 0.5686274766921997, 0.6078431606292725, 0.5882353186607361, 0.7254902124404907, 0.6666666865348816, 0.7450980544090271, 0.686274528503418, 0.686274528503418, 0.7450980544090271, 0.7450980544090271, 0.7254902124404907, 0.7843137383460999, 0.7450980544090271, 0.843137264251709, 0.8823529481887817, 0.8039215803146362, 0.8235294222831726, 0.9019607901573181, 0.8823529481887817, 0.9215686321258545, 0.9019607901573181, 1.0], [0.05882352963089943, 0.03921568766236305, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.11764705926179886, 0.1568627506494522, 0.19607843458652496, 0.3529411852359772, 0.3137255012989044, 0.3137255012989044, 0.2549019753932953, 0.19607843458652496, 0.21568627655506134, 0.2549019753932953, 0.13725490868091583, 0.3137255012989044, 0.3333333432674408, 0.3333333432674408, 0.2549019753932953, 0.3333333432674408, 0.29411765933036804, 0.21568627655506134, 0.3529411852359772, 0.29411765933036804, 0.3137255012989044, 0.1568627506494522, 0.47058823704719543, 0.27450981736183167, 0.47058823704719543, 0.4901960790157318, 0.4313725531101227, 0.4313725531101227, 0.5490196347236633, 0.5490196347236633, 0.529411792755127, 0.529411792755127, NaN, 0.6274510025978088, 0.5490196347236633, 0.5490196347236633, 0.5098039507865906, 0.6078431606292725, 0.6078431606292725, 0.6666666865348816, 0.5490196347236633, 0.6666666865348816, 0.6078431606292725, 0.7254902124404907, 0.686274528503418, 0.7450980544090271, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.7450980544090271, 0.7843137383460999, 0.7450980544090271, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.843137264251709, 0.8235294222831726, 0.8235294222831726, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636], [0.05882352963089943, 0.09803921729326248, 0.11764705926179886, 0.05882352963089943, 0.11764705926179886, 0.1568627506494522, 0.13725490868091583, 0.09803921729326248, 0.11764705926179886, 0.1568627506494522, 0.27450981736183167, 0.21568627655506134, 0.29411765933036804, 0.23529411852359772, 0.1764705926179886, 0.23529411852359772, 0.3137255012989044, 0.2549019753932953, 0.3529411852359772, 0.3333333432674408, 0.27450981736183167, 0.2549019753932953, 0.27450981736183167, 0.2549019753932953, 0.3137255012989044, 0.3137255012989044, 0.3529411852359772, 0.2549019753932953, 0.27450981736183167, 0.37254902720451355, 0.3333333432674408, 0.3529411852359772, 0.4117647111415863, 0.4901960790157318, 0.3921568691730499, 0.21568627655506134, 0.47058823704719543, 0.4117647111415863, 0.4901960790157318, 0.37254902720451355, NaN, 0.5686274766921997, 0.5098039507865906, 0.5490196347236633, 0.5882353186607361, 0.4901960790157318, 0.5490196347236633, 0.5882353186607361, 0.5686274766921997, 0.529411792755127, 0.5882353186607361, 0.686274528503418, 0.686274528503418, 0.6666666865348816, 0.7647058963775635, 0.7647058963775635, 0.7647058963775635, 0.6666666865348816, 0.7647058963775635, 0.7058823704719543, 0.7254902124404907, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.8039215803146362, 0.8823529481887817, 0.9215686321258545, 0.9215686321258545, 0.9215686321258545, 0.9607843160629272], [0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.1764705926179886, 0.23529411852359772, 0.27450981736183167, 0.2549019753932953, 0.19607843458652496, 0.27450981736183167, 0.23529411852359772, 0.29411765933036804, 0.23529411852359772, 0.19607843458652496, 0.27450981736183167, 0.29411765933036804, 0.3137255012989044, 0.3529411852359772, 0.37254902720451355, 0.23529411852359772, 0.3333333432674408, 0.3137255012989044, 0.3921568691730499, 0.2549019753932953, 0.4313725531101227, 0.37254902720451355, 0.37254902720451355, 0.3529411852359772, 0.3921568691730499, 0.3333333432674408, 0.4901960790157318, 0.4117647111415863, 0.45098039507865906, 0.45098039507865906, 0.4313725531101227, NaN, 0.47058823704719543, 0.5686274766921997, 0.45098039507865906, 0.47058823704719543, 0.529411792755127, 0.5490196347236633, 0.6666666865348816, 0.5882353186607361, 0.7058823704719543, 0.6470588445663452, 0.686274528503418, 0.7058823704719543, 0.7843137383460999, 0.7647058963775635, 0.843137264251709, 0.7843137383460999, 0.7254902124404907, 0.8235294222831726, 0.8627451062202454, 0.843137264251709, 0.8235294222831726, 0.8627451062202454, 0.8823529481887817, 0.8627451062202454, 0.8627451062202454, 0.9411764740943909, 0.9019607901573181, 0.9803921580314636], [0.019607843831181526, 0.0784313753247261, 0.03921568766236305, 0.03921568766236305, 0.09803921729326248, 0.11764705926179886, 0.0784313753247261, 0.05882352963089943, 0.11764705926179886, 0.1568627506494522, 0.23529411852359772, 0.13725490868091583, 0.1764705926179886, 0.19607843458652496, 0.19607843458652496, 0.1764705926179886, 0.1568627506494522, 0.1764705926179886, 0.2549019753932953, 0.21568627655506134, 0.19607843458652496, 0.2647058963775635, 0.3333333432674408, 0.23529411852359772, 0.23529411852359772, 0.23529411852359772, 0.3333333432674408, 0.27450981736183167, 0.29411765933036804, 0.3333333432674408, 0.37254902720451355, 0.3529411852359772, 0.3333333432674408, 0.4313725531101227, 0.3137255012989044, 0.3921568691730499, 0.2549019753932953, 0.3921568691730499, 0.45098039507865906, 0.45098039507865906, 0.4901960790157318, 0.529411792755127, NaN, 0.5392156839370728, 0.47058823704719543, 0.4901960790157318, 0.5098039507865906, 0.5882353186607361, 0.5882353186607361, 0.5588235259056091, 0.5882353186607361, 0.6274510025978088, 0.6666666865348816, 0.7058823704719543, 0.7647058963775635, 0.7450980544090271, 0.9215686321258545, 0.7254902124404907, 0.7450980544090271, 0.7254902124404907, 0.6666666865348816, 0.8039215803146362, 0.843137264251709, 0.843137264251709, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272], [0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.03921568766236305, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.21568627655506134, 0.23529411852359772, 0.1764705926179886, 0.2549019753932953, 0.1568627506494522, 0.21568627655506134, 0.1568627506494522, 0.1764705926179886, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.1764705926179886, 0.3333333432674408, 0.27450981736183167, 0.27450981736183167, 0.3333333432674408, 0.3333333432674408, 0.1764705926179886, 0.3333333432674408, 0.2549019753932953, 0.37254902720451355, 0.3137255012989044, 0.27450981736183167, 0.3921568691730499, 0.37254902720451355, 0.3529411852359772, 0.37254902720451355, 0.3921568691730499, 0.45098039507865906, 0.4901960790157318, 0.45098039507865906, 0.4313725531101227, 0.46078431606292725, NaN, 0.4313725531101227, 0.4313725531101227, 0.5098039507865906, 0.7156862616539001, 0.6470588445663452, 0.5098039507865906, 0.7058823704719543, 0.5490196347236633, 0.6470588445663452, 0.6470588445663452, 0.7058823704719543, 0.7254902124404907, 0.6078431606292725, 0.7254902124404907, 0.7254902124404907, 0.9411764740943909, 0.8823529481887817, 0.7647058963775635, 0.7843137383460999, 0.9215686321258545, 0.7843137383460999, 0.8235294222831726, 0.8823529481887817, 0.9411764740943909, 0.8823529481887817, 1.0], [0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.09803921729326248, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.21568627655506134, 0.19607843458652496, 0.23529411852359772, 0.19607843458652496, 0.13725490868091583, 0.1764705926179886, 0.23529411852359772, 0.1764705926179886, 0.27450981736183167, 0.27450981736183167, 0.27450981736183167, 0.21568627655506134, 0.2549019753932953, 0.21568627655506134, 0.23529411852359772, 0.29411765933036804, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.3529411852359772, 0.2549019753932953, 0.2549019753932953, 0.4313725531101227, 0.4313725531101227, 0.4019607901573181, 0.1568627506494522, 0.45098039507865906, 0.4117647111415863, 0.4313725531101227, 0.3921568691730499, 0.4117647111415863, 0.5490196347236633, 0.529411792755127, 0.5686274766921997, NaN, 0.4117647111415863, 0.5686274766921997, 0.5882353186607361, 0.5882353186607361, 0.5098039507865906, 0.6274510025978088, 0.6666666865348816, 0.686274528503418, 0.686274528503418, 0.7254902124404907, 0.7254902124404907, 0.7843137383460999, 0.686274528503418, 0.6666666865348816, 0.7058823704719543, 0.6666666865348816, 0.7843137383460999, 0.843137264251709, 0.843137264251709, 0.8235294222831726, 0.9019607901573181, 0.9215686321258545, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272], [0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.03921568766236305, 0.03921568766236305, 0.13725490868091583, 0.05882352963089943, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.2549019753932953, 0.1568627506494522, 0.21568627655506134, 0.21568627655506134, 0.09803921729326248, 0.13725490868091583, 0.1764705926179886, 0.13725490868091583, 0.3137255012989044, 0.21568627655506134, 0.21568627655506134, 0.11764705926179886, 0.29411765933036804, 0.1764705926179886, 0.23529411852359772, 0.2549019753932953, 0.3333333432674408, 0.13725490868091583, 0.27450981736183167, 0.3137255012989044, 0.2549019753932953, 0.3921568691730499, 0.3921568691730499, 0.4117647111415863, 0.27450981736183167, 0.3921568691730499, 0.3529411852359772, 0.3333333432674408, 0.47058823704719543, 0.3921568691730499, 0.5098039507865906, 0.529411792755127, 0.5098039507865906, 0.5686274766921997, 0.5882353186607361, NaN, 0.5686274766921997, 0.5882353186607361, 0.5882353186607361, 0.529411792755127, 0.6470588445663452, 0.6274510025978088, 0.7843137383460999, 0.6274510025978088, 0.7254902124404907, 0.7254902124404907, 0.7450980544090271, 0.7254902124404907, 0.7450980544090271, 0.7254902124404907, 0.7843137383460999, 0.7647058963775635, 0.7843137383460999, 0.7843137383460999, 0.8627451062202454, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.8823529481887817, 0.9607843160629272], [0.0784313753247261, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.11764705926179886, 0.05882352963089943, 0.13725490868091583, 0.0784313753247261, 0.0784313753247261, 0.0784313753247261, 0.2549019753932953, 0.1568627506494522, 0.21568627655506134, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.13725490868091583, 0.21568627655506134, 0.3333333432674408, 0.19607843458652496, 0.2549019753932953, 0.21568627655506134, 0.3137255012989044, 0.21568627655506134, 0.19607843458652496, 0.28431373834609985, 0.1764705926179886, 0.23529411852359772, 0.23529411852359772, 0.27450981736183167, 0.27450981736183167, 0.21568627655506134, 0.3137255012989044, 0.3529411852359772, 0.2549019753932953, 0.3137255012989044, 0.3333333432674408, 0.3529411852359772, 0.3529411852359772, 0.3333333432674408, 0.45098039507865906, 0.47058823704719543, 0.4901960790157318, 0.4901960790157318, 0.4313725531101227, 0.4313725531101227, NaN, 0.4901960790157318, 0.529411792755127, 0.45098039507865906, 0.5686274766921997, 0.7058823704719543, 0.6470588445663452, 0.6078431606292725, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6274510025978088, 0.7254902124404907, 0.7254902124404907, 0.7058823704719543, 0.9803921580314636, 0.7647058963775635, 0.7843137383460999, 0.843137264251709, 0.8627451062202454, 0.8823529481887817, 0.8627451062202454, 0.9215686321258545, 0.9607843160629272], [0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.2549019753932953, 0.19607843458652496, 0.23529411852359772, 0.1764705926179886, 0.1764705926179886, 0.1764705926179886, 0.23529411852359772, 0.27450981736183167, 0.19607843458652496, 0.3333333432674408, 0.23529411852359772, 0.27450981736183167, 0.3137255012989044, 0.29411765933036804, 0.19607843458652496, 0.3137255012989044, 0.21568627655506134, 0.3333333432674408, 0.2549019753932953, 0.2549019753932953, 0.3333333432674408, 0.3333333432674408, 0.3921568691730499, 0.3529411852359772, 0.3333333432674408, 0.37254902720451355, 0.45098039507865906, 0.4117647111415863, 0.45098039507865906, 0.4117647111415863, 0.28431373834609985, 0.4117647111415863, 0.4117647111415863, 0.5098039507865906, NaN, 0.47058823704719543, 0.4313725531101227, 0.6078431606292725, 0.529411792755127, 0.5882353186607361, 0.6470588445663452, 0.686274528503418, 0.686274528503418, 0.5490196347236633, 0.7058823704719543, 0.6666666865348816, 0.8235294222831726, 0.843137264251709, 0.686274528503418, 0.8039215803146362, 0.843137264251709, 0.7647058963775635, 0.7647058963775635, 0.7647058963775635, 0.9803921580314636, 0.8823529481887817, 0.9803921580314636], [0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.0784313753247261, 0.11764705926179886, 0.0784313753247261, 0.09803921729326248, 0.13725490868091583, 0.1568627506494522, 0.21568627655506134, 0.1568627506494522, 0.1764705926179886, 0.19607843458652496, 0.13725490868091583, 0.1568627506494522, 0.09803921729326248, 0.11764705926179886, 0.3137255012989044, 0.21568627655506134, 0.2549019753932953, 0.13725490868091583, 0.29411765933036804, 0.21568627655506134, 0.19607843458652496, 0.27450981736183167, 0.3137255012989044, 0.1764705926179886, 0.23529411852359772, 0.19607843458652496, 0.27450981736183167, 0.19607843458652496, 0.2549019753932953, 0.3333333432674408, 0.29411765933036804, 0.37254902720451355, 0.3137255012989044, 0.3137255012989044, 0.4313725531101227, 0.3333333432674408, 0.4313725531101227, 0.3333333432674408, 0.4117647111415863, 0.3529411852359772, 0.4117647111415863, 0.4117647111415863, 0.47058823704719543, 0.529411792755127, NaN, 0.4901960790157318, 0.7058823704719543, 0.5490196347236633, 0.6274510025978088, 0.5686274766921997, 0.686274528503418, 0.7058823704719543, 0.686274528503418, 0.6666666865348816, 0.686274528503418, 0.8039215803146362, 0.8823529481887817, 0.7450980544090271, 0.8039215803146362, 0.843137264251709, 0.686274528503418, 0.8039215803146362, 0.7843137383460999, 0.9411764740943909, 0.8627451062202454, 0.9803921580314636], [0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.0784313753247261, 0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.1764705926179886, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.03921568766236305, 0.1568627506494522, 0.13725490868091583, 0.1568627506494522, 0.27450981736183167, 0.1568627506494522, 0.1764705926179886, 0.019607843831181526, 0.3137255012989044, 0.13725490868091583, 0.21568627655506134, 0.21568627655506134, 0.23529411852359772, 0.09803921729326248, 0.2549019753932953, 0.2549019753932953, 0.23529411852359772, 0.2549019753932953, 0.3137255012989044, 0.3137255012989044, 0.27450981736183167, 0.4117647111415863, 0.3137255012989044, 0.3921568691730499, 0.3921568691730499, 0.3921568691730499, 0.47058823704719543, 0.4117647111415863, 0.44117647409439087, 0.4901960790157318, 0.4901960790157318, 0.47058823704719543, 0.5490196347236633, 0.5686274766921997, 0.5098039507865906, NaN, 0.5490196347236633, 0.6078431606292725, 0.7254902124404907, 0.6666666865348816, 0.7058823704719543, 0.7254902124404907, 0.7058823704719543, 0.7058823704719543, 0.7254902124404907, 0.7647058963775635, 0.7647058963775635, 0.7843137383460999, 0.8039215803146362, 0.7647058963775635, 0.8039215803146362, 0.7843137383460999, 0.7843137383460999, 0.9411764740943909, 0.9019607901573181, 0.9411764740943909], [0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.09803921729326248, 0.19607843458652496, 0.11764705926179886, 0.13725490868091583, 0.19607843458652496, 0.05882352963089943, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.21568627655506134, 0.23529411852359772, 0.21568627655506134, 0.09803921729326248, 0.2549019753932953, 0.11764705926179886, 0.1568627506494522, 0.21568627655506134, 0.29411765933036804, 0.13725490868091583, 0.19607843458652496, 0.1764705926179886, 0.19607843458652496, 0.19607843458652496, 0.21568627655506134, 0.29411765933036804, 0.27450981736183167, 0.3529411852359772, 0.2549019753932953, 0.29411765933036804, 0.4117647111415863, 0.27450981736183167, 0.4117647111415863, 0.29411765933036804, 0.4117647111415863, 0.29411765933036804, 0.37254902720451355, 0.3529411852359772, 0.4313725531101227, 0.3921568691730499, 0.29411765933036804, 0.45098039507865906, NaN, 0.5098039507865906, 0.6274510025978088, 0.5490196347236633, 0.6666666865348816, 0.6666666865348816, 0.5882353186607361, 0.5882353186607361, 0.6666666865348816, 0.7254902124404907, 0.7843137383460999, 0.7647058963775635, 0.7647058963775635, 0.843137264251709, 0.6666666865348816, 0.7843137383460999, 0.7647058963775635, 0.9215686321258545, 0.843137264251709, 0.9803921580314636], [0.05882352963089943, 0.05882352963089943, 0.03921568766236305, 0.0784313753247261, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.05882352963089943, 0.019607843831181526, 0.09803921729326248, 0.1568627506494522, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.11764705926179886, 0.1764705926179886, 0.13725490868091583, 0.13725490868091583, 0.1764705926179886, 0.21568627655506134, 0.1568627506494522, 0.1568627506494522, 0.1764705926179886, 0.19607843458652496, 0.1568627506494522, 0.2549019753932953, 0.21568627655506134, 0.23529411852359772, 0.19607843458652496, 0.2549019753932953, 0.21568627655506134, 0.1568627506494522, 0.2549019753932953, 0.21568627655506134, 0.2549019753932953, 0.27450981736183167, 0.3137255012989044, 0.3137255012989044, 0.3529411852359772, 0.37254902720451355, 0.45098039507865906, 0.3333333432674408, 0.37254902720451355, 0.29411765933036804, 0.47058823704719543, 0.45098039507865906, 0.3921568691730499, 0.4901960790157318, NaN, 0.5686274766921997, 0.529411792755127, 0.6274510025978088, 0.6274510025978088, 0.5882353186607361, 0.5686274766921997, 0.6470588445663452, 0.686274528503418, 0.6274510025978088, 0.7843137383460999, 0.7450980544090271, 0.686274528503418, 0.7254902124404907, 0.7843137383460999, 0.7843137383460999, 0.843137264251709, 0.8823529481887817, 0.9411764740943909], [0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.19607843458652496, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.11764705926179886, 0.23529411852359772, 0.13725490868091583, 0.13725490868091583, 0.1764705926179886, 0.21568627655506134, 0.13725490868091583, 0.21568627655506134, 0.23529411852359772, 0.21568627655506134, 0.22549019753932953, 0.23529411852359772, 0.3137255012989044, 0.27450981736183167, 0.29411765933036804, 0.23529411852359772, 0.27450981736183167, 0.3333333432674408, 0.2549019753932953, 0.3137255012989044, 0.3137255012989044, 0.3333333432674408, 0.3529411852359772, 0.3137255012989044, 0.21568627655506134, 0.3529411852359772, 0.4117647111415863, 0.37254902720451355, 0.27450981736183167, 0.37254902720451355, 0.4313725531101227, NaN, 0.47058823704719543, 0.6666666865348816, 0.6470588445663452, 0.5686274766921997, 0.45098039507865906, 0.529411792755127, 0.529411792755127, 0.5490196347236633, 0.6274510025978088, 0.7647058963775635, 0.7647058963775635, 0.5882353186607361, 0.7647058963775635, 0.7647058963775635, 0.8235294222831726, 0.843137264251709, 0.9215686321258545], [0.0, 0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.0784313753247261, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.1568627506494522, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.11764705926179886, 0.1568627506494522, 0.2549019753932953, 0.11764705926179886, 0.09803921729326248, 0.1764705926179886, 0.19607843458652496, 0.19607843458652496, 0.1764705926179886, 0.19607843458652496, 0.19607843458652496, 0.23529411852359772, 0.2549019753932953, 0.2549019753932953, 0.1568627506494522, 0.3333333432674408, 0.29411765933036804, 0.29411765933036804, 0.2549019753932953, 0.23529411852359772, 0.3333333432674408, 0.29411765933036804, 0.29411765933036804, 0.3529411852359772, 0.3137255012989044, 0.37254902720451355, 0.3921568691730499, 0.3529411852359772, 0.4313725531101227, 0.3333333432674408, 0.45098039507865906, 0.47058823704719543, 0.529411792755127, NaN, 0.47058823704719543, 0.47058823704719543, 0.4901960790157318, 0.813725471496582, 0.6078431606292725, 0.5882353186607361, 0.6470588445663452, 0.6078431606292725, 0.4901960790157318, 0.6078431606292725, 0.9019607901573181, 0.7450980544090271, 0.7450980544090271, 0.7450980544090271, 0.7647058963775635, 0.9215686321258545], [0.05882352963089943, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.11764705926179886, 0.09803921729326248, 0.13725490868091583, 0.1764705926179886, 0.09803921729326248, 0.1568627506494522, 0.1764705926179886, 0.1568627506494522, 0.19607843458652496, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.19607843458652496, 0.1764705926179886, 0.19607843458652496, 0.2549019753932953, 0.1764705926179886, 0.1568627506494522, 0.22549019753932953, 0.19607843458652496, 0.1568627506494522, 0.21568627655506134, 0.21568627655506134, 0.1568627506494522, 0.1764705926179886, 0.23529411852359772, 0.21568627655506134, 0.21568627655506134, 0.2549019753932953, 0.29411765933036804, 0.23529411852359772, 0.19607843458652496, 0.2549019753932953, 0.27450981736183167, 0.3137255012989044, 0.23529411852359772, 0.23529411852359772, 0.21568627655506134, 0.23529411852359772, 0.29411765933036804, 0.27450981736183167, 0.27450981736183167, 0.3333333432674408, 0.3137255012989044, 0.3137255012989044, 0.29411765933036804, 0.3333333432674408, 0.37254902720451355, 0.3333333432674408, 0.529411792755127, NaN, 0.5196078419685364, 0.4313725531101227, 0.529411792755127, 0.529411792755127, 0.5686274766921997, 0.5490196347236633, 0.5490196347236633, 0.9607843160629272, 0.5882353186607361, 0.5490196347236633, 0.7058823704719543, 0.7254902124404907, 0.7843137383460999, 0.8039215803146362, 0.843137264251709], [0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.11764705926179886, 0.09803921729326248, 0.1568627506494522, 0.1764705926179886, 0.09803921729326248, 0.1568627506494522, 0.21568627655506134, 0.13725490868091583, 0.21568627655506134, 0.13725490868091583, 0.1764705926179886, 0.1764705926179886, 0.19607843458652496, 0.1764705926179886, 0.21568627655506134, 0.23529411852359772, 0.1764705926179886, 0.1568627506494522, 0.23529411852359772, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.21568627655506134, 0.1568627506494522, 0.19607843458652496, 0.23529411852359772, 0.21568627655506134, 0.21568627655506134, 0.23529411852359772, 0.27450981736183167, 0.21568627655506134, 0.19607843458652496, 0.23529411852359772, 0.23529411852359772, 0.3137255012989044, 0.19607843458652496, 0.23529411852359772, 0.23529411852359772, 0.2549019753932953, 0.27450981736183167, 0.27450981736183167, 0.27450981736183167, 0.3333333432674408, 0.3137255012989044, 0.29411765933036804, 0.27450981736183167, 0.3333333432674408, 0.37254902720451355, 0.3529411852359772, 0.529411792755127, 0.4803921580314636, NaN, 0.45098039507865906, 0.529411792755127, 0.529411792755127, 0.5490196347236633, 0.529411792755127, 0.529411792755127, 0.9803921580314636, 0.5686274766921997, 0.5490196347236633, 0.7058823704719543, 0.7058823704719543, 0.7843137383460999, 0.8039215803146362, 0.843137264251709], [0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.05882352963089943, 0.13725490868091583, 0.03921568766236305, 0.11764705926179886, 0.019607843831181526, 0.0784313753247261, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.13725490868091583, 0.1568627506494522, 0.019607843831181526, 0.0784313753247261, 0.19607843458652496, 0.11764705926179886, 0.09803921729326248, 0.13725490868091583, 0.1764705926179886, 0.09803921729326248, 0.1764705926179886, 0.11764705926179886, 0.1764705926179886, 0.0784313753247261, 0.09803921729326248, 0.21568627655506134, 0.11764705926179886, 0.1568627506494522, 0.03921568766236305, 0.13725490868091583, 0.2549019753932953, 0.2549019753932953, 0.23529411852359772, 0.1568627506494522, 0.0784313753247261, 0.3921568691730499, 0.21568627655506134, 0.2549019753932953, 0.3333333432674408, 0.45098039507865906, 0.3137255012989044, 0.29411765933036804, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.5098039507865906, 0.5686274766921997, 0.5490196347236633, NaN, 0.529411792755127, 0.5490196347236633, 0.6470588445663452, 0.5882353186607361, 0.5490196347236633, 0.7058823704719543, 0.6666666865348816, 0.6274510025978088, 0.7058823704719543, 0.7647058963775635, 0.843137264251709, 0.843137264251709, 0.9215686321258545], [0.0, 0.03921568766236305, 0.0, 0.0, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.09803921729326248, 0.0784313753247261, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.05882352963089943, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.1568627506494522, 0.09803921729326248, 0.11764705926179886, 0.23529411852359772, 0.09803921729326248, 0.05882352963089943, 0.1568627506494522, 0.1568627506494522, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.1568627506494522, 0.1764705926179886, 0.1764705926179886, 0.1568627506494522, 0.27450981736183167, 0.23529411852359772, 0.23529411852359772, 0.2549019753932953, 0.21568627655506134, 0.3333333432674408, 0.21568627655506134, 0.27450981736183167, 0.27450981736183167, 0.3137255012989044, 0.27450981736183167, 0.37254902720451355, 0.29411765933036804, 0.3333333432674408, 0.29411765933036804, 0.4117647111415863, 0.4313725531101227, 0.5490196347236633, 0.18627451360225677, 0.47058823704719543, 0.47058823704719543, 0.47058823704719543, NaN, 0.5686274766921997, 0.5098039507865906, 0.5882353186607361, 0.5882353186607361, 0.4901960790157318, 0.5490196347236633, 0.813725471496582, 0.7254902124404907, 0.7058823704719543, 0.6666666865348816, 0.7450980544090271, 0.9215686321258545], [0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.0784313753247261, 0.0784313753247261, 0.05882352963089943, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.1568627506494522, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.1764705926179886, 0.1764705926179886, 0.1764705926179886, 0.1568627506494522, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.21568627655506134, 0.19607843458652496, 0.0784313753247261, 0.19607843458652496, 0.11764705926179886, 0.03921568766236305, 0.19607843458652496, 0.2549019753932953, 0.23529411852359772, 0.23529411852359772, 0.23529411852359772, 0.27450981736183167, 0.2549019753932953, 0.23529411852359772, 0.27450981736183167, 0.2549019753932953, 0.27450981736183167, 0.3333333432674408, 0.2549019753932953, 0.27450981736183167, 0.3333333432674408, 0.3137255012989044, 0.27450981736183167, 0.3333333432674408, 0.3529411852359772, 0.47058823704719543, 0.3921568691730499, 0.47058823704719543, 0.47058823704719543, 0.45098039507865906, 0.4313725531101227, NaN, 0.4901960790157318, 0.529411792755127, 0.4901960790157318, 0.6078431606292725, 0.529411792755127, 0.5490196347236633, 0.6078431606292725, 0.6078431606292725, 0.7254902124404907, 0.6470588445663452, 0.8823529481887817], [0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.0784313753247261, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.09803921729326248, 0.03921568766236305, 0.13725490868091583, 0.0784313753247261, 0.11764705926179886, 0.03921568766236305, 0.09803921729326248, 0.05882352963089943, 0.13725490868091583, 0.1568627506494522, 0.0784313753247261, 0.11764705926179886, 0.1568627506494522, 0.13725490868091583, 0.1764705926179886, 0.11764705926179886, 0.09803921729326248, 0.21568627655506134, 0.05882352963089943, 0.19607843458652496, 0.13725490868091583, 0.0784313753247261, 0.1568627506494522, 0.19607843458652496, 0.27450981736183167, 0.23529411852359772, 0.1764705926179886, 0.21568627655506134, 0.23529411852359772, 0.29411765933036804, 0.1764705926179886, 0.27450981736183167, 0.05882352963089943, 0.29411765933036804, 0.27450981736183167, 0.27450981736183167, 0.1764705926179886, 0.19607843458652496, 0.23529411852359772, 0.27450981736183167, 0.3137255012989044, 0.47058823704719543, 0.4117647111415863, 0.4313725531101227, 0.45098039507865906, 0.3529411852359772, 0.4901960790157318, 0.5098039507865906, NaN, 0.5098039507865906, 0.5490196347236633, 0.5882353186607361, 0.5686274766921997, 0.6078431606292725, 0.6666666865348816, 0.6666666865348816, 0.7647058963775635, 0.7058823704719543, 0.9607843160629272], [0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.03921568766236305, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.03921568766236305, 0.11764705926179886, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.1764705926179886, 0.05882352963089943, 0.13725490868091583, 0.0784313753247261, 0.19607843458652496, 0.09803921729326248, 0.13725490868091583, 0.21568627655506134, 0.1764705926179886, 0.2549019753932953, 0.19607843458652496, 0.19607843458652496, 0.2549019753932953, 0.23529411852359772, 0.27450981736183167, 0.13725490868091583, 0.3333333432674408, 0.11764705926179886, 0.3333333432674408, 0.21568627655506134, 0.29411765933036804, 0.1568627506494522, 0.11764705926179886, 0.23529411852359772, 0.21568627655506134, 0.37254902720451355, 0.45098039507865906, 0.3529411852359772, 0.45098039507865906, 0.47058823704719543, 0.4117647111415863, 0.4117647111415863, 0.47058823704719543, 0.4901960790157318, NaN, 0.5490196347236633, 0.5882353186607361, 0.6078431606292725, 0.5882353186607361, 0.7058823704719543, 0.7058823704719543, 0.7843137383460999, 0.7450980544090271, 0.9803921580314636], [0.0, 0.019607843831181526, 0.0, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.03921568766236305, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.13725490868091583, 0.09803921729326248, 0.05882352963089943, 0.09803921729326248, 0.09803921729326248, 0.09803921729326248, 0.13725490868091583, 0.0784313753247261, 0.1568627506494522, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.11764705926179886, 0.05882352963089943, 0.1568627506494522, 0.19607843458652496, 0.1764705926179886, 0.1568627506494522, 0.19607843458652496, 0.23529411852359772, 0.21568627655506134, 0.23529411852359772, 0.019607843831181526, 0.3137255012989044, 0.2549019753932953, 0.21568627655506134, 0.23529411852359772, 0.21568627655506134, 0.37254902720451355, 0.3921568691730499, 0.45098039507865906, 0.47058823704719543, 0.45098039507865906, 0.4117647111415863, 0.5098039507865906, 0.45098039507865906, 0.45098039507865906, NaN, 0.6666666865348816, 0.529411792755127, 0.6666666865348816, 0.686274528503418, 0.7058823704719543, 0.7450980544090271, 0.7450980544090271, 0.9019607901573181], [0.03921568766236305, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.0784313753247261, 0.09803921729326248, 0.11764705926179886, 0.0784313753247261, 0.11764705926179886, 0.11764705926179886, 0.13725490868091583, 0.13725490868091583, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.11764705926179886, 0.13725490868091583, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.1568627506494522, 0.13725490868091583, 0.1764705926179886, 0.23529411852359772, 0.13725490868091583, 0.13725490868091583, 0.19607843458652496, 0.11764705926179886, 0.11764705926179886, 0.1568627506494522, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.21568627655506134, 0.1568627506494522, 0.21568627655506134, 0.23529411852359772, 0.19607843458652496, 0.19607843458652496, 0.19607843458652496, 0.23529411852359772, 0.2549019753932953, 0.23529411852359772, 0.5098039507865906, 0.03921568766236305, 0.019607843831181526, 0.29411765933036804, 0.5098039507865906, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.3333333432674408, NaN, 0.5098039507865906, 0.529411792755127, 0.5882353186607361, 0.5882353186607361, 0.6470588445663452, 0.6470588445663452, 0.8039215803146362], [0.05882352963089943, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.0784313753247261, 0.05882352963089943, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.05882352963089943, 0.05882352963089943, 0.09803921729326248, 0.05882352963089943, 0.0784313753247261, 0.03921568766236305, 0.09803921729326248, 0.11764705926179886, 0.10784313827753067, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.0784313753247261, 0.0784313753247261, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.05882352963089943, 0.1568627506494522, 0.05882352963089943, 0.05882352963089943, 0.1764705926179886, 0.1568627506494522, 0.11764705926179886, 0.11764705926179886, 0.11764705926179886, 0.19607843458652496, 0.1764705926179886, 0.13725490868091583, 0.13725490868091583, 0.1568627506494522, 0.0784313753247261, 0.1568627506494522, 0.21568627655506134, 0.21568627655506134, 0.1568627506494522, 0.1568627506494522, 0.23529411852359772, 0.1568627506494522, 0.3137255012989044, 0.23529411852359772, 0.3921568691730499, 0.4117647111415863, 0.4313725531101227, 0.3333333432674408, 0.45098039507865906, 0.47058823704719543, 0.4313725531101227, 0.3921568691730499, 0.47058823704719543, 0.4901960790157318, NaN, 0.5490196347236633, 0.686274528503418, 0.686274528503418, 0.9803921580314636, 0.7647058963775635, 0.9803921580314636], [0.0, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.0784313753247261, 0.019607843831181526, 0.03921568766236305, 0.05882352963089943, 0.0784313753247261, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.09803921729326248, 0.05882352963089943, 0.09803921729326248, 0.13725490868091583, 0.0784313753247261, 0.03921568766236305, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.13725490868091583, 0.11764705926179886, 0.13725490868091583, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.09803921729326248, 0.1568627506494522, 0.09803921729326248, 0.0784313753247261, 0.1764705926179886, 0.1764705926179886, 0.19607843458652496, 0.11764705926179886, 0.1568627506494522, 0.21568627655506134, 0.1764705926179886, 0.13725490868091583, 0.1568627506494522, 0.23529411852359772, 0.3137255012989044, 0.19607843458652496, 0.3333333432674408, 0.27450981736183167, 0.4117647111415863, 0.09803921729326248, 0.45098039507865906, 0.45098039507865906, 0.37254902720451355, 0.18627451360225677, 0.45098039507865906, 0.3921568691730499, 0.4117647111415863, 0.3333333432674408, 0.47058823704719543, 0.45098039507865906, NaN, 0.6078431606292725, 0.6078431606292725, 0.5490196347236633, 0.6470588445663452, 0.843137264251709], [0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.03921568766236305, 0.11764705926179886, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.13725490868091583, 0.13725490868091583, 0.0784313753247261, 0.05882352963089943, 0.09803921729326248, 0.09803921729326248, 0.03921568766236305, 0.0784313753247261, 0.11764705926179886, 0.0784313753247261, 0.05882352963089943, 0.03921568766236305, 0.09803921729326248, 0.11764705926179886, 0.0784313753247261, 0.21568627655506134, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.0784313753247261, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.09803921729326248, 0.1568627506494522, 0.13725490868091583, 0.23529411852359772, 0.19607843458652496, 0.21568627655506134, 0.21568627655506134, 0.21568627655506134, 0.23529411852359772, 0.2549019753932953, 0.29411765933036804, 0.29411765933036804, 0.29411765933036804, 0.27450981736183167, 0.3921568691730499, 0.3333333432674408, 0.29411765933036804, 0.3137255012989044, 0.4117647111415863, 0.3137255012989044, 0.3921568691730499, NaN, 0.7647058963775635, 0.4901960790157318, 0.7647058963775635, 0.7254902124404907], [0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.029411764815449715, 0.11764705926179886, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.13725490868091583, 0.11764705926179886, 0.0784313753247261, 0.03921568766236305, 0.09803921729326248, 0.05882352963089943, 0.019607843831181526, 0.05882352963089943, 0.09803921729326248, 0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.0784313753247261, 0.09803921729326248, 0.19607843458652496, 0.09803921729326248, 0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.13725490868091583, 0.11764705926179886, 0.0784313753247261, 0.13725490868091583, 0.11764705926179886, 0.23529411852359772, 0.21568627655506134, 0.21568627655506134, 0.23529411852359772, 0.21568627655506134, 0.23529411852359772, 0.2549019753932953, 0.27450981736183167, 0.29411765933036804, 0.23529411852359772, 0.29411765933036804, 0.3921568691730499, 0.3333333432674408, 0.29411765933036804, 0.29411765933036804, 0.4117647111415863, 0.3137255012989044, 0.3921568691730499, 0.23529411852359772, NaN, 0.47058823704719543, 0.686274528503418, 0.7058823704719543], [0.019607843831181526, 0.0, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.0, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.0, 0.0, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.09803921729326248, 0.019607843831181526, 0.05882352963089943, 0.11764705926179886, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.019607843831181526, 0.09803921729326248, 0.019607843831181526, 0.0784313753247261, 0.05882352963089943, 0.03921568766236305, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.05882352963089943, 0.05882352963089943, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.13725490868091583, 0.019607843831181526, 0.05882352963089943, 0.05882352963089943, 0.0784313753247261, 0.1568627506494522, 0.1764705926179886, 0.2549019753932953, 0.21568627655506134, 0.21568627655506134, 0.1568627506494522, 0.3333333432674408, 0.27450981736183167, 0.23529411852359772, 0.21568627655506134, 0.2549019753932953, 0.3529411852359772, 0.019607843831181526, 0.45098039507865906, 0.5098039507865906, 0.529411792755127, NaN, 0.5490196347236633, 0.9215686321258545], [0.0, 0.019607843831181526, 0.0, 0.019607843831181526, 0.03921568766236305, 0.0, 0.03921568766236305, 0.0, 0.019607843831181526, 0.019607843831181526, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.11764705926179886, 0.03921568766236305, 0.03921568766236305, 0.0784313753247261, 0.05882352963089943, 0.019607843831181526, 0.05882352963089943, 0.09803921729326248, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.0784313753247261, 0.0784313753247261, 0.13725490868091583, 0.0784313753247261, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.09803921729326248, 0.019607843831181526, 0.0784313753247261, 0.09803921729326248, 0.05882352963089943, 0.11764705926179886, 0.03921568766236305, 0.11764705926179886, 0.0784313753247261, 0.11764705926179886, 0.13725490868091583, 0.09803921729326248, 0.1568627506494522, 0.11764705926179886, 0.1568627506494522, 0.23529411852359772, 0.19607843458652496, 0.19607843458652496, 0.1568627506494522, 0.2549019753932953, 0.3529411852359772, 0.29411765933036804, 0.2549019753932953, 0.2549019753932953, 0.3529411852359772, 0.23529411852359772, 0.3529411852359772, 0.23529411852359772, 0.3137255012989044, 0.45098039507865906, NaN, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.019607843831181526, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.0, 0.019607843831181526, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.019607843831181526, 0.0, 0.0, 0.0, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.0, 0.0, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.0, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.0784313753247261, 0.0784313753247261, 0.11764705926179886, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.19607843458652496, 0.019607843831181526, 0.1568627506494522, 0.27450981736183167, 0.29411765933036804, 0.0784313753247261, 0.3333333432674408, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9320829822},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8738277986},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8645922207},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.847115665},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8375959161},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8115942116},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7980960587},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.790849682},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7800511603},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7567490866},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6874112099},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6845694915},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6815856905},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6720659401},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6663825051},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6601307301},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6588519571},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6579994416},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6452117217},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6348394542},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6312873097},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6202046132},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6184995877},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.603580573},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6021597141},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5977550555},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5949133397},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5947712528},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5869565332},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5842569008},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5510088204},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.548451275},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5339585218},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5304063772},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.527280487},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5140665069},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4978687225},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.491048604},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4890594035},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4850810006},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4546746342},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4543904604},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4309462999},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4305200428},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4300937865},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4225632389},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4042341658},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3974140473},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3898834984},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3874680394},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3492469534},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3353225432},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3167093002},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3104575232},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3097470958},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3081841513},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2676896922},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2657004896},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2639954604},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2446717885},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2384200109},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2205171972},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2196646844},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2149758491},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.192810462},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.168797958},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1516055732},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1170787189},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1156578604},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0423415757}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_all/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9781b2342e27a3ccf6b522fc1c63504f92eba9e2
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_all/winrate_matrix.csv
@@ -0,0 +1,71 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),TabICLv2 (default),RealTabPFN-2.5 (tuned),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (default),RealMLP (tuned + ensembled),TabDPT (tuned + ensembled),RealMLP (tuned),TabDPT-Turbo (default),TabM (tuned + ensembled),LightGBM (tuned + ensembled),CatBoost (tuned),iLTM (tuned + ensembled),CatBoost (tuned + ensembled),TabDPT (tuned),ModernNCA (tuned + ensembled),TabM (tuned),XGBoost (tuned + ensembled),LimiX (default),LightGBM (tuned),CatBoost (default),ModernNCA (tuned),TabSwift (default),XGBoost (tuned),ChimeraBoost (tuned + ensembled),xRFM (tuned + ensembled),ChimeraBoost (tuned),iLTM (tuned),xRFM (tuned),TabDPT (default),TabM (default),EBM (tuned + ensembled),TorchMLP (tuned + ensembled),RealMLP (default),SAP-RPT-OSS (default),ChimeraBoost (default),EBM (tuned),ModernNCA (default),TorchMLP (tuned),ExtraTrees (tuned + ensembled),EBM (default),XGBoost (default),FastaiMLP (tuned + ensembled),ExtraTrees (tuned),RandomForest (tuned + ensembled),LightGBM (default),RandomForest (tuned),FastaiMLP (tuned),iLTM (default),PerpetualBooster (tuned + ensembled),TabSTAR (tuned),TabSTAR (tuned + ensembled),TorchMLP (default),PerpetualBooster (tuned),xRFM (default),ExtraTrees (default),RandomForest (default),FastaiMLP (default),TabSTAR (default),KNN (tuned + ensembled),PerpetualBooster (default),Linear (tuned + ensembled),Linear (tuned),KNN (tuned),Linear (default),KNN (default)
+TabFM (default),,0.7254902,0.8039216,0.84313726,0.8627451,0.88235295,0.88235295,0.8235294,0.8627451,0.8627451,0.9019608,0.88235295,0.9411765,0.9411765,0.9411765,0.92156863,0.8627451,0.9411765,0.9607843,0.88235295,0.9607843,0.92156863,0.9019608,0.9607843,0.9019608,0.92156863,0.88235295,0.88235295,0.92156863,0.9411765,0.9019608,0.9411765,0.9411765,0.98039216,0.9411765,0.9411765,0.9607843,0.92156863,0.92156863,0.9411765,0.9411765,0.9607843,0.98039216,0.9607843,0.9411765,0.9607843,0.92156863,0.9607843,0.9607843,0.9607843,0.9607843,0.9411765,0.9411765,1.0,0.9411765,0.92156863,0.98039216,1.0,0.9607843,0.98039216,0.98039216,1.0,0.9607843,0.9411765,1.0,0.98039216,0.98039216,0.98039216,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.27450982,,0.4509804,0.60784316,0.5686275,0.50980395,0.6666667,0.74509805,0.60784316,0.8235294,0.74509805,0.84313726,0.74509805,0.8627451,0.92156863,0.88235295,0.9019608,0.9019608,0.7647059,0.8627451,0.84313726,0.9411765,0.7647059,0.9411765,0.9411765,0.92156863,0.84313726,0.9411765,0.9411765,0.92156863,0.9411765,0.92156863,0.9019608,0.8235294,0.9019608,0.88235295,0.9411765,0.9019608,0.8627451,0.9607843,0.9019608,0.9607843,0.92156863,0.9411765,0.92156863,0.9411765,0.9019608,0.9411765,0.9607843,0.98039216,0.9607843,0.9411765,0.9607843,0.9411765,0.9019608,0.88235295,0.9607843,0.9607843,0.98039216,0.98039216,0.9607843,0.98039216,0.9019608,0.98039216,0.9607843,0.98039216,0.98039216,1.0,0.98039216,1.0
+TabPFN-3 (default),0.19607843,0.54901963,,0.5686275,0.5588235,0.627451,0.6862745,0.627451,0.64705884,0.74509805,0.7058824,0.7647059,0.84313726,0.84313726,0.88235295,0.7647059,0.84313726,0.8039216,0.8235294,0.9019608,0.9019608,0.8627451,0.7647059,0.9411765,0.8627451,0.84313726,0.88235295,0.84313726,0.88235295,0.8627451,0.8627451,0.92156863,0.92156863,0.92156863,0.88235295,0.8235294,0.9019608,0.9411765,0.84313726,0.9019608,0.88235295,0.9607843,0.9607843,0.9607843,0.9019608,0.9411765,0.9019608,0.9607843,0.9411765,0.9607843,0.98039216,0.9607843,0.98039216,0.9607843,0.9019608,0.9019608,0.9607843,1.0,0.9607843,0.98039216,0.98039216,1.0,0.9019608,0.9607843,0.98039216,0.9607843,0.98039216,0.98039216,1.0,1.0
+TabPFN-2.6 (default),0.15686275,0.39215687,0.43137255,,0.47058824,0.50980395,0.54901963,0.54901963,0.6862745,0.7254902,0.7058824,0.8039216,0.7647059,0.78431374,0.88235295,0.8039216,0.84313726,0.84313726,0.7254902,0.8627451,0.8235294,0.8235294,0.8039216,0.9411765,0.9019608,0.84313726,0.8235294,0.8627451,0.8627451,0.84313726,0.8627451,0.9019608,0.8627451,0.78431374,0.88235295,0.9019608,0.92156863,0.92156863,0.8627451,0.88235295,0.9411765,0.9607843,0.9607843,0.9411765,0.9019608,0.9607843,0.92156863,0.9411765,0.9607843,0.9607843,0.9411765,0.92156863,0.9607843,0.98039216,0.88235295,0.88235295,0.98039216,1.0,0.9607843,0.9607843,0.98039216,0.98039216,0.88235295,0.98039216,0.98039216,0.98039216,0.98039216,0.98039216,0.98039216,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.13725491,0.43137255,0.44117647,0.5294118,,0.54901963,0.7254902,0.60784316,0.7058824,0.74509805,0.6666667,0.7647059,0.7058824,0.7254902,0.84313726,0.8235294,0.78431374,0.8627451,0.7254902,0.7647059,0.8039216,0.8627451,0.74509805,0.88235295,0.9019608,0.8627451,0.8235294,0.8627451,0.84313726,0.84313726,0.88235295,0.84313726,0.88235295,0.78431374,0.88235295,0.88235295,0.8627451,0.88235295,0.88235295,0.8627451,0.88235295,0.88235295,0.9019608,0.9411765,0.9019608,0.9607843,0.88235295,0.9411765,0.92156863,0.98039216,0.9411765,0.9411765,0.92156863,0.9607843,0.88235295,0.88235295,0.9607843,0.98039216,0.9411765,0.98039216,0.98039216,0.98039216,0.9019608,0.9411765,0.9607843,0.9411765,0.9607843,0.9607843,0.9607843,1.0
+TabICLv2 (default),0.11764706,0.49019608,0.37254903,0.49019608,0.4509804,,0.627451,0.6666667,0.5882353,0.6666667,0.7254902,0.6862745,0.7647059,0.7254902,0.8235294,0.7647059,0.78431374,0.7647059,0.74509805,0.8235294,0.74509805,0.78431374,0.74509805,0.84313726,0.8039216,0.8039216,0.8235294,0.78431374,0.8235294,0.78431374,0.8039216,0.78431374,0.8235294,0.78431374,0.8627451,0.78431374,0.8627451,0.84313726,0.8235294,0.8235294,0.84313726,0.9019608,0.88235295,0.88235295,0.9019608,0.8627451,0.9411765,0.92156863,0.88235295,0.92156863,0.88235295,0.92156863,0.88235295,0.92156863,0.9019608,0.9019608,0.9607843,0.9411765,0.92156863,0.92156863,0.9411765,0.98039216,0.92156863,0.9607843,0.9607843,0.9607843,0.98039216,0.98039216,1.0,1.0
+RealTabPFN-2.5 (tuned),0.11764706,0.33333334,0.3137255,0.4509804,0.27450982,0.37254903,,0.54901963,0.61764705,0.64705884,0.627451,0.74509805,0.6862745,0.7647059,0.74509805,0.74509805,0.6862745,0.7647059,0.6862745,0.7058824,0.8235294,0.8039216,0.64705884,0.8627451,0.8627451,0.8235294,0.74509805,0.8235294,0.74509805,0.78431374,0.8235294,0.8627451,0.8627451,0.7647059,0.8627451,0.8627451,0.8627451,0.8627451,0.84313726,0.84313726,0.8627451,0.9019608,0.92156863,0.9607843,0.88235295,0.9411765,0.8627451,0.9019608,0.92156863,0.9607843,0.92156863,0.92156863,0.9019608,0.9607843,0.8627451,0.84313726,0.9607843,0.98039216,0.92156863,0.9607843,0.9607843,0.9607843,0.9019608,0.92156863,0.9607843,0.9411765,0.9411765,0.9607843,0.9607843,1.0
+"AutoGluon 1.4 (best, 4h)",0.1764706,0.25490198,0.37254903,0.4509804,0.39215687,0.33333334,0.4509804,,0.4509804,0.54901963,0.5294118,0.7058824,0.60784316,0.6666667,0.8235294,0.7647059,0.78431374,0.8039216,0.5882353,0.64705884,0.74509805,0.8627451,0.64705884,0.8627451,0.8235294,0.7058824,0.7058824,0.8235294,0.84313726,0.8039216,0.8627451,0.8235294,0.8235294,0.74509805,0.8235294,0.8039216,0.92156863,0.8235294,0.7647059,0.88235295,0.9019608,0.92156863,0.9411765,0.8627451,0.9019608,0.92156863,0.92156863,0.8627451,0.9019608,0.9411765,0.9607843,0.9411765,0.9411765,0.9607843,0.8235294,0.8235294,0.98039216,0.98039216,0.92156863,0.98039216,0.98039216,0.98039216,0.88235295,0.9411765,0.98039216,0.98039216,0.98039216,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.13725491,0.39215687,0.3529412,0.3137255,0.29411766,0.4117647,0.38235295,0.54901963,,0.5686275,0.64705884,0.6666667,0.64705884,0.7058824,0.74509805,0.7058824,0.627451,0.7058824,0.6666667,0.64705884,0.74509805,0.74509805,0.7058824,0.7647059,0.74509805,0.7058824,0.84313726,0.78431374,0.74509805,0.8039216,0.74509805,0.85294116,0.84313726,0.7647059,0.84313726,0.8039216,0.84313726,0.8235294,0.84313726,0.84313726,0.88235295,0.88235295,0.88235295,0.8627451,0.88235295,0.88235295,0.92156863,0.8627451,0.8627451,0.9607843,0.92156863,0.98039216,0.92156863,0.9411765,0.9019608,0.9019608,0.98039216,0.98039216,0.92156863,0.98039216,0.9411765,0.9607843,0.92156863,0.92156863,0.9411765,0.9607843,0.9607843,0.98039216,0.98039216,0.98039216
+RealMLP (tuned + ensembled),0.13725491,0.1764706,0.25490198,0.27450982,0.25490198,0.33333334,0.3529412,0.4509804,0.43137255,,0.50980395,0.6862745,0.5882353,0.627451,0.7058824,0.7254902,0.74509805,0.78431374,0.54901963,0.64705884,0.7058824,0.7647059,0.5686275,0.7647059,0.78431374,0.6862745,0.6666667,0.8039216,0.7058824,0.74509805,0.7647059,0.8235294,0.8039216,0.7058824,0.8235294,0.8039216,0.84313726,0.92156863,0.8039216,0.8039216,0.84313726,0.9019608,0.84313726,0.8627451,0.8627451,0.9019608,0.92156863,0.8627451,0.84313726,0.9411765,0.9019608,0.9019608,0.92156863,0.9607843,0.84313726,0.84313726,0.9411765,0.98039216,0.9411765,0.98039216,0.9607843,0.98039216,0.88235295,0.9607843,0.98039216,0.98039216,0.98039216,0.98039216,0.98039216,1.0
+TabDPT (tuned + ensembled),0.09803922,0.25490198,0.29411766,0.29411766,0.33333334,0.27450982,0.37254903,0.47058824,0.3529412,0.49019608,,0.5686275,0.49019608,0.5294118,0.5882353,0.54901963,0.64705884,0.5882353,0.74509805,0.5294118,0.54901963,0.627451,0.627451,0.60784316,0.60784316,0.5686275,0.627451,0.627451,0.627451,0.6666667,0.627451,0.74509805,0.7254902,0.8039216,0.6862745,0.64705884,0.74509805,0.7058824,0.7254902,0.64705884,0.7254902,0.8235294,0.7647059,0.78431374,0.78431374,0.74509805,0.74509805,0.8627451,0.78431374,0.8235294,0.8039216,0.84313726,0.8039216,0.84313726,0.8235294,0.78431374,0.8627451,0.9019608,0.9411765,0.98039216,0.9607843,0.9607843,0.88235295,0.92156863,0.92156863,0.9019608,0.9019608,0.9607843,0.92156863,1.0
+RealMLP (tuned),0.11764706,0.15686275,0.23529412,0.19607843,0.23529412,0.3137255,0.25490198,0.29411766,0.33333334,0.3137255,0.43137255,,0.4117647,0.50980395,0.5882353,0.54901963,0.5294118,0.54901963,0.50980395,0.5294118,0.60784316,0.6666667,0.5,0.6666667,0.627451,0.6862745,0.5686275,0.7058824,0.5686275,0.64705884,0.627451,0.7254902,0.7058824,0.627451,0.7647059,0.7647059,0.8039216,0.8235294,0.6862745,0.6862745,0.78431374,0.7647059,0.8627451,0.7647059,0.8039216,0.84313726,0.84313726,0.8235294,0.84313726,0.84313726,0.88235295,0.88235295,0.9019608,0.9019608,0.84313726,0.8627451,0.9607843,0.92156863,0.92156863,0.9019608,0.9411765,0.9607843,0.8627451,0.9411765,0.98039216,0.9607843,0.9705882,0.98039216,0.9607843,1.0
+TabDPT-Turbo (default),0.05882353,0.25490198,0.15686275,0.23529412,0.29411766,0.23529412,0.3137255,0.39215687,0.3529412,0.4117647,0.50980395,0.5882353,,0.47058824,0.54901963,0.54901963,0.60784316,0.5882353,0.54901963,0.50980395,0.5294118,0.6666667,0.60784316,0.5882353,0.64705884,0.5882353,0.6960784,0.6862745,0.6666667,0.6666667,0.6666667,0.74509805,0.74509805,0.7254902,0.7058824,0.6666667,0.74509805,0.7058824,0.7058824,0.6862745,0.7058824,0.7254902,0.8235294,0.8235294,0.7647059,0.78431374,0.78431374,0.84313726,0.8235294,0.8627451,0.8627451,0.9019608,0.84313726,0.84313726,0.8039216,0.78431374,0.88235295,0.9019608,0.9019608,0.9607843,0.9411765,0.9607843,0.8627451,0.9411765,0.9607843,0.88235295,0.88235295,0.98039216,0.92156863,1.0
+TabM (tuned + ensembled),0.05882353,0.13725491,0.15686275,0.21568628,0.27450982,0.27450982,0.23529412,0.33333334,0.29411766,0.37254903,0.47058824,0.49019608,0.5294118,,0.5882353,0.49019608,0.5294118,0.54901963,0.50980395,0.54901963,0.627451,0.5882353,0.49019608,0.627451,0.64705884,0.60784316,0.5882353,0.6862745,0.60784316,0.5686275,0.7254902,0.64705884,0.5686275,0.64705884,0.88235295,0.78431374,0.74509805,0.8235294,0.7058824,0.74509805,0.7647059,0.74509805,0.8039216,0.74509805,0.8039216,0.78431374,0.8235294,0.74509805,0.8039216,0.8627451,0.8039216,0.88235295,0.84313726,0.84313726,0.8627451,0.8627451,0.98039216,0.9019608,0.84313726,0.8627451,0.88235295,0.9411765,0.88235295,0.9019608,0.9411765,0.9607843,0.9607843,1.0,0.9607843,1.0
+LightGBM (tuned + ensembled),0.05882353,0.078431375,0.11764706,0.11764706,0.15686275,0.1764706,0.25490198,0.1764706,0.25490198,0.29411766,0.4117647,0.4117647,0.4509804,0.4117647,,0.54901963,0.50980395,0.5686275,0.4509804,0.47058824,0.49019608,0.5686275,0.4509804,0.74509805,0.7254902,0.54901963,0.54901963,0.6666667,0.64705884,0.5882353,0.74509805,0.6666667,0.6862745,0.60784316,0.7254902,0.74509805,0.7647059,0.7254902,0.6666667,0.8039216,0.8235294,0.8039216,0.8039216,0.84313726,0.8627451,0.9019608,0.8235294,0.8039216,0.8627451,0.9607843,0.9411765,0.8627451,0.92156863,0.9019608,0.84313726,0.8235294,0.92156863,0.92156863,0.84313726,0.92156863,0.98039216,0.92156863,0.8627451,0.9411765,0.92156863,0.98039216,0.98039216,1.0,0.9607843,1.0
+CatBoost (tuned),0.078431375,0.11764706,0.23529412,0.19607843,0.1764706,0.23529412,0.25490198,0.23529412,0.29411766,0.27450982,0.4509804,0.4509804,0.4509804,0.50980395,0.4509804,,0.49019608,0.4117647,0.47058824,0.54901963,0.5294118,0.60784316,0.5294118,0.54901963,0.6862745,0.5686275,0.5294118,0.6666667,0.5686275,0.5882353,0.64705884,0.74509805,0.6862745,0.5882353,0.6862745,0.7254902,0.7647059,0.7647059,0.7058824,0.78431374,0.7647059,0.7254902,0.8235294,0.78431374,0.8235294,0.8627451,0.78431374,0.7647059,0.84313726,0.84313726,0.9019608,0.84313726,0.88235295,0.88235295,0.8235294,0.8235294,0.9411765,0.9411765,0.84313726,0.88235295,0.9411765,0.92156863,0.84313726,0.92156863,0.9607843,0.98039216,0.98039216,0.98039216,0.98039216,1.0
+iLTM (tuned + ensembled),0.13725491,0.09803922,0.15686275,0.15686275,0.21568628,0.21568628,0.3137255,0.21568628,0.37254903,0.25490198,0.3529412,0.47058824,0.39215687,0.47058824,0.49019608,0.50980395,,0.47058824,0.4117647,0.47058824,0.5294118,0.5882353,0.48039216,0.5686275,0.5294118,0.60784316,0.5294118,0.6666667,0.64705884,0.5686275,0.6666667,0.7058824,0.6666667,0.5882353,0.64705884,0.6666667,0.7647059,0.7254902,0.7254902,0.74509805,0.6862745,0.7647059,0.84313726,0.84313726,0.7647059,0.8235294,0.8627451,0.8235294,0.9019608,0.8627451,0.9019608,0.8627451,0.9019608,0.9019608,0.8039216,0.8039216,0.9607843,0.9411765,0.92156863,0.9607843,0.9607843,0.9607843,0.8627451,0.9607843,0.9607843,0.9411765,0.9607843,0.9607843,0.9607843,1.0
+CatBoost (tuned + ensembled),0.05882353,0.09803922,0.19607843,0.15686275,0.13725491,0.23529412,0.23529412,0.19607843,0.29411766,0.21568628,0.4117647,0.4509804,0.4117647,0.4509804,0.43137255,0.5882353,0.5294118,,0.4509804,0.54901963,0.47058824,0.5882353,0.49019608,0.5686275,0.7058824,0.5588235,0.50980395,0.64705884,0.5882353,0.54901963,0.6862745,0.74509805,0.6862745,0.5686275,0.627451,0.6862745,0.74509805,0.7647059,0.74509805,0.8627451,0.74509805,0.7058824,0.8235294,0.8235294,0.8235294,0.8627451,0.78431374,0.8235294,0.88235295,0.84313726,0.92156863,0.88235295,0.88235295,0.92156863,0.8235294,0.8235294,0.9411765,0.92156863,0.88235295,0.9019608,0.9607843,0.92156863,0.84313726,0.9019608,0.9607843,0.98039216,0.98039216,0.9607843,0.98039216,1.0
+TabDPT (tuned),0.039215688,0.23529412,0.1764706,0.27450982,0.27450982,0.25490198,0.3137255,0.4117647,0.33333334,0.4509804,0.25490198,0.49019608,0.4509804,0.49019608,0.54901963,0.5294118,0.5882353,0.54901963,,0.50980395,0.50980395,0.60784316,0.54901963,0.5882353,0.5882353,0.50980395,0.627451,0.627451,0.627451,0.627451,0.64705884,0.7058824,0.6666667,0.8137255,0.627451,0.627451,0.64705884,0.64705884,0.64705884,0.6862745,0.64705884,0.7647059,0.74509805,0.7254902,0.7254902,0.6862745,0.6666667,0.8235294,0.6862745,0.7254902,0.78431374,0.8235294,0.8235294,0.84313726,0.8039216,0.78431374,0.8627451,0.9019608,0.9019608,0.9411765,0.88235295,0.9607843,0.8627451,0.88235295,0.9411765,0.8627451,0.8627451,0.9607843,0.9019608,0.98039216
+ModernNCA (tuned + ensembled),0.11764706,0.13725491,0.09803922,0.13725491,0.23529412,0.1764706,0.29411766,0.3529412,0.3529412,0.3529412,0.47058824,0.47058824,0.49019608,0.4509804,0.5294118,0.4509804,0.5294118,0.4509804,0.49019608,,0.49019608,0.5882353,0.47058824,0.6666667,0.5882353,0.6764706,0.5686275,0.5882353,0.60784316,0.5882353,0.627451,0.60784316,0.6666667,0.60784316,0.60784316,0.6666667,0.64705884,0.7254902,0.627451,0.6666667,0.6666667,0.8039216,0.78431374,0.7647059,0.7254902,0.78431374,0.8039216,0.7647059,0.78431374,0.84313726,0.7647059,0.8627451,0.8235294,0.84313726,0.74509805,0.7647059,0.84313726,0.84313726,0.9019608,0.8627451,0.9019608,0.88235295,0.8627451,0.89215684,0.9019608,0.8627451,0.88235295,0.9019608,0.88235295,0.98039216
+TabM (tuned),0.039215688,0.15686275,0.09803922,0.1764706,0.19607843,0.25490198,0.1764706,0.25490198,0.25490198,0.29411766,0.4509804,0.39215687,0.47058824,0.37254903,0.50980395,0.47058824,0.47058824,0.5294118,0.49019608,0.50980395,,0.54901963,0.4509804,0.54901963,0.627451,0.5686275,0.54901963,0.60784316,0.6764706,0.50980395,0.6666667,0.627451,0.54901963,0.60784316,0.7647059,0.6862745,0.6666667,0.7058824,0.64705884,0.6666667,0.7254902,0.7254902,0.8039216,0.7254902,0.7254902,0.78431374,0.74509805,0.7254902,0.74509805,0.8235294,0.78431374,0.8627451,0.78431374,0.88235295,0.8235294,0.8235294,0.98039216,0.9019608,0.8235294,0.84313726,0.9019608,0.9019608,0.8627451,0.9019608,0.9411765,0.92156863,0.92156863,0.98039216,0.9607843,0.98039216
+XGBoost (tuned + ensembled),0.078431375,0.05882353,0.13725491,0.1764706,0.13725491,0.21568628,0.19607843,0.13725491,0.25490198,0.23529412,0.37254903,0.33333334,0.33333334,0.4117647,0.43137255,0.39215687,0.4117647,0.4117647,0.39215687,0.4117647,0.4509804,,0.43137255,0.50980395,0.5294118,0.49019608,0.47058824,0.627451,0.5294118,0.5294118,0.60784316,0.627451,0.627451,0.5686275,0.64705884,0.6862745,0.7254902,0.7254902,0.60784316,0.74509805,0.74509805,0.7058824,0.7352941,0.8235294,0.78431374,0.88235295,0.78431374,0.8039216,0.8627451,0.98039216,0.9019608,0.8235294,0.88235295,0.84313726,0.84313726,0.84313726,0.92156863,0.88235295,0.8235294,0.92156863,0.9607843,0.9019608,0.8627451,0.9019608,0.9019608,0.9411765,0.9607843,0.9411765,0.9607843,1.0
+LimiX (default),0.09803922,0.23529412,0.23529412,0.19607843,0.25490198,0.25490198,0.3529412,0.3529412,0.29411766,0.43137255,0.37254903,0.5,0.39215687,0.50980395,0.54901963,0.47058824,0.51960784,0.50980395,0.4509804,0.5294118,0.54901963,0.5686275,,0.5686275,0.5686275,0.5686275,0.49019608,0.5882353,0.60784316,0.5294118,0.64705884,0.60784316,0.64705884,0.54901963,0.60784316,0.6862745,0.6666667,0.7058824,0.627451,0.6666667,0.7254902,0.6862745,0.6666667,0.6666667,0.74509805,0.7058824,0.6862745,0.6666667,0.7058824,0.6862745,0.74509805,0.78431374,0.7647059,0.74509805,0.7745098,0.7647059,0.8039216,0.7647059,0.8235294,0.88235295,0.88235295,0.8627451,0.84313726,0.88235295,0.8627451,0.9019608,0.9019608,0.88235295,0.92156863,0.98039216
+LightGBM (tuned),0.039215688,0.05882353,0.05882353,0.05882353,0.11764706,0.15686275,0.13725491,0.13725491,0.23529412,0.23529412,0.39215687,0.33333334,0.4117647,0.37254903,0.25490198,0.4509804,0.43137255,0.43137255,0.4117647,0.33333334,0.4509804,0.49019608,0.43137255,,0.54901963,0.50980395,0.47058824,0.5294118,0.54901963,0.54901963,0.64705884,0.5686275,0.60784316,0.5686275,0.7254902,0.7254902,0.6666667,0.6862745,0.627451,0.7058824,0.74509805,0.64705884,0.7647059,0.7254902,0.78431374,0.8235294,0.78431374,0.7647059,0.78431374,0.8627451,0.88235295,0.84313726,0.8627451,0.88235295,0.8039216,0.8235294,0.88235295,0.9019608,0.84313726,0.84313726,0.9411765,0.9019608,0.8627451,0.9019608,0.92156863,0.9019608,0.9411765,0.9607843,0.9411765,1.0
+CatBoost (default),0.09803922,0.05882353,0.13725491,0.09803922,0.09803922,0.19607843,0.13725491,0.1764706,0.25490198,0.21568628,0.39215687,0.37254903,0.3529412,0.3529412,0.27450982,0.3137255,0.47058824,0.29411766,0.4117647,0.4117647,0.37254903,0.47058824,0.43137255,0.4509804,,0.4509804,0.50980395,0.49019608,0.47058824,0.5686275,0.60784316,0.64705884,0.627451,0.5882353,0.5882353,0.64705884,0.7058824,0.74509805,0.627451,0.78431374,0.6862745,0.627451,0.7647059,0.7254902,0.7647059,0.7647059,0.8039216,0.7254902,0.8039216,0.78431374,0.84313726,0.84313726,0.8627451,0.9019608,0.84313726,0.8235294,0.9019608,0.9411765,0.8627451,0.8627451,0.92156863,0.9411765,0.8627451,0.92156863,0.9607843,0.9607843,0.98039216,0.98039216,0.98039216,1.0
+ModernNCA (tuned),0.078431375,0.078431375,0.15686275,0.15686275,0.13725491,0.19607843,0.1764706,0.29411766,0.29411766,0.3137255,0.43137255,0.3137255,0.4117647,0.39215687,0.4509804,0.43137255,0.39215687,0.44117647,0.49019608,0.32352942,0.43137255,0.50980395,0.43137255,0.49019608,0.54901963,,0.50980395,0.49019608,0.50980395,0.5294118,0.54901963,0.5686275,0.627451,0.5882353,0.60784316,0.5686275,0.6666667,0.6666667,0.6666667,0.64705884,0.6862745,0.7647059,0.7647059,0.6666667,0.7058824,0.74509805,0.71568626,0.6862745,0.7254902,0.78431374,0.78431374,0.8235294,0.8235294,0.8235294,0.78431374,0.78431374,0.8627451,0.84313726,0.84313726,0.8235294,0.9019608,0.9019608,0.88235295,0.92156863,0.88235295,0.92156863,0.9411765,0.9411765,0.9411765,1.0
+TabSwift (default),0.11764706,0.15686275,0.11764706,0.1764706,0.1764706,0.1764706,0.25490198,0.29411766,0.15686275,0.33333334,0.37254903,0.43137255,0.30392158,0.4117647,0.4509804,0.47058824,0.47058824,0.49019608,0.37254903,0.43137255,0.4509804,0.5294118,0.50980395,0.5294118,0.49019608,0.49019608,,0.54901963,0.49019608,0.5686275,0.5294118,0.5686275,0.627451,0.5686275,0.5882353,0.64705884,0.5686275,0.64705884,0.7058824,0.7058824,0.64705884,0.6666667,0.6666667,0.6666667,0.7254902,0.6666667,0.8235294,0.7058824,0.6862745,0.7647059,0.7058824,0.8039216,0.78431374,0.8039216,0.78431374,0.78431374,0.8235294,0.84313726,0.8235294,0.88235295,0.8235294,0.9019608,0.8627451,0.9019608,0.9019608,0.88235295,0.9019608,0.98039216,0.9019608,1.0
+XGBoost (tuned),0.11764706,0.05882353,0.15686275,0.13725491,0.13725491,0.21568628,0.1764706,0.1764706,0.21568628,0.19607843,0.37254903,0.29411766,0.3137255,0.3137255,0.33333334,0.33333334,0.33333334,0.3529412,0.37254903,0.4117647,0.39215687,0.37254903,0.4117647,0.47058824,0.50980395,0.50980395,0.4509804,,0.5294118,0.5686275,0.54901963,0.5882353,0.5882353,0.5294118,0.5882353,0.627451,0.64705884,0.6666667,0.60784316,0.6862745,0.74509805,0.6862745,0.7254902,0.8235294,0.7647059,0.8627451,0.7647059,0.8039216,0.8235294,0.9019608,0.8627451,0.84313726,0.8627451,0.8039216,0.84313726,0.84313726,0.9019608,0.88235295,0.78431374,0.9019608,0.9411765,0.9019608,0.8627451,0.92156863,0.9019608,0.92156863,0.9411765,0.9411765,0.9607843,1.0
+ChimeraBoost (tuned + ensembled),0.078431375,0.05882353,0.11764706,0.13725491,0.15686275,0.1764706,0.25490198,0.15686275,0.25490198,0.29411766,0.37254903,0.43137255,0.33333334,0.39215687,0.3529412,0.43137255,0.3529412,0.4117647,0.37254903,0.39215687,0.32352942,0.47058824,0.39215687,0.4509804,0.5294118,0.49019608,0.50980395,0.47058824,,0.54901963,0.60784316,0.5882353,0.64705884,0.54901963,0.60784316,0.6862745,0.627451,0.7058824,0.60784316,0.84313726,0.7254902,0.60784316,0.7058824,0.6666667,0.7254902,0.7254902,0.7647059,0.6862745,0.7647059,0.74509805,0.8039216,0.74509805,0.78431374,0.8235294,0.8235294,0.8039216,0.8235294,0.8627451,0.8039216,0.78431374,0.8627451,0.8627451,0.84313726,0.8627451,0.8627451,0.9411765,0.9607843,0.9607843,0.98039216,1.0
+xRFM (tuned + ensembled),0.05882353,0.078431375,0.13725491,0.15686275,0.15686275,0.21568628,0.21568628,0.19607843,0.19607843,0.25490198,0.33333334,0.3529412,0.33333334,0.43137255,0.4117647,0.4117647,0.43137255,0.4509804,0.37254903,0.4117647,0.49019608,0.47058824,0.47058824,0.4509804,0.43137255,0.47058824,0.43137255,0.43137255,0.4509804,,0.5294118,0.5294118,0.6862745,0.47058824,0.60784316,0.5882353,0.60784316,0.64705884,0.5882353,0.5294118,0.627451,0.74509805,0.6666667,0.74509805,0.64705884,0.6862745,0.7254902,0.78431374,0.8039216,0.74509805,0.8235294,0.78431374,0.7647059,0.8039216,0.7647059,0.7647059,0.88235295,0.84313726,0.92156863,0.9411765,0.92156863,0.92156863,0.8235294,0.9411765,0.88235295,0.9607843,0.98039216,0.98039216,0.9607843,0.98039216
+ChimeraBoost (tuned),0.09803922,0.05882353,0.13725491,0.13725491,0.11764706,0.19607843,0.1764706,0.13725491,0.25490198,0.23529412,0.37254903,0.37254903,0.33333334,0.27450982,0.25490198,0.3529412,0.33333334,0.3137255,0.3529412,0.37254903,0.33333334,0.39215687,0.3529412,0.3529412,0.39215687,0.4509804,0.47058824,0.4509804,0.39215687,0.47058824,,0.47058824,0.5882353,0.50980395,0.5294118,0.627451,0.5882353,0.7058824,0.5882353,0.7254902,0.6666667,0.5686275,0.627451,0.627451,0.74509805,0.74509805,0.7254902,0.6666667,0.7254902,0.7647059,0.8039216,0.7647059,0.78431374,0.8039216,0.78431374,0.78431374,0.8235294,0.8235294,0.8039216,0.8039216,0.8039216,0.84313726,0.84313726,0.84313726,0.8627451,0.9019608,0.9019608,0.9411765,0.9607843,1.0
+iLTM (tuned),0.05882353,0.078431375,0.078431375,0.09803922,0.15686275,0.21568628,0.13725491,0.1764706,0.14705883,0.1764706,0.25490198,0.27450982,0.25490198,0.3529412,0.33333334,0.25490198,0.29411766,0.25490198,0.29411766,0.39215687,0.37254903,0.37254903,0.39215687,0.43137255,0.3529412,0.43137255,0.43137255,0.4117647,0.4117647,0.47058824,0.5294118,,0.5294118,0.49019608,0.5294118,0.54901963,0.5882353,0.60784316,0.64705884,0.5294118,0.64705884,0.627451,0.64705884,0.6862745,0.74509805,0.60784316,0.78431374,0.74509805,0.8039216,0.74509805,0.8039216,0.8039216,0.7745098,0.7647059,0.78431374,0.78431374,0.92156863,0.84313726,0.88235295,0.8627451,0.9019608,0.9019608,0.8627451,0.9411765,0.9019608,0.88235295,0.92156863,0.98039216,0.92156863,1.0
+xRFM (tuned),0.05882353,0.09803922,0.078431375,0.13725491,0.11764706,0.1764706,0.13725491,0.1764706,0.15686275,0.19607843,0.27450982,0.29411766,0.25490198,0.43137255,0.3137255,0.3137255,0.33333334,0.3137255,0.33333334,0.33333334,0.4509804,0.37254903,0.3529412,0.39215687,0.37254903,0.37254903,0.37254903,0.4117647,0.3529412,0.3137255,0.4117647,0.47058824,,0.43137255,0.50980395,0.50980395,0.54901963,0.5882353,0.60784316,0.50980395,0.5882353,0.627451,0.6666667,0.7254902,0.5686275,0.60784316,0.6862745,0.74509805,0.74509805,0.6862745,0.78431374,0.74509805,0.7647059,0.74509805,0.74509805,0.7647059,0.9019608,0.8235294,0.9607843,0.92156863,0.8627451,0.88235295,0.8235294,0.9411765,0.92156863,0.92156863,0.9019608,0.98039216,0.92156863,1.0
+TabDPT (default),0.019607844,0.1764706,0.078431375,0.21568628,0.21568628,0.21568628,0.23529412,0.25490198,0.23529412,0.29411766,0.19607843,0.37254903,0.27450982,0.3529412,0.39215687,0.4117647,0.4117647,0.43137255,0.18627451,0.39215687,0.39215687,0.43137255,0.4509804,0.43137255,0.4117647,0.4117647,0.43137255,0.47058824,0.4509804,0.5294118,0.49019608,0.50980395,0.5686275,,0.50980395,0.4509804,0.50980395,0.49019608,0.5294118,0.5686275,0.50980395,0.64705884,0.5686275,0.60784316,0.5686275,0.5882353,0.64705884,0.6666667,0.6666667,0.6862745,0.7058824,0.78431374,0.6862745,0.74509805,0.7058824,0.7254902,0.78431374,0.8235294,0.8039216,0.84313726,0.78431374,0.8627451,0.7647059,0.8235294,0.8627451,0.78431374,0.8039216,0.9019608,0.8627451,0.98039216
+TabM (default),0.05882353,0.09803922,0.11764706,0.11764706,0.11764706,0.13725491,0.13725491,0.1764706,0.15686275,0.1764706,0.3137255,0.23529412,0.29411766,0.11764706,0.27450982,0.3137255,0.3529412,0.37254903,0.37254903,0.39215687,0.23529412,0.3529412,0.39215687,0.27450982,0.4117647,0.39215687,0.4117647,0.4117647,0.39215687,0.39215687,0.47058824,0.47058824,0.49019608,0.49019608,,0.54901963,0.5686275,0.5882353,0.5686275,0.5686275,0.60784316,0.60784316,0.6862745,0.627451,0.5980392,0.7254902,0.74509805,0.6666667,0.7058824,0.7254902,0.7254902,0.84313726,0.7254902,0.84313726,0.7647059,0.78431374,0.88235295,0.84313726,0.74509805,0.8039216,0.8235294,0.84313726,0.8627451,0.84313726,0.9019608,0.92156863,0.9019608,0.98039216,0.92156863,0.9607843
+EBM (tuned + ensembled),0.05882353,0.11764706,0.1764706,0.09803922,0.11764706,0.21568628,0.13725491,0.19607843,0.19607843,0.19607843,0.3529412,0.23529412,0.33333334,0.21568628,0.25490198,0.27450982,0.33333334,0.3137255,0.37254903,0.33333334,0.3137255,0.3137255,0.3137255,0.27450982,0.3529412,0.43137255,0.3529412,0.37254903,0.3137255,0.4117647,0.37254903,0.4509804,0.49019608,0.54901963,0.4509804,,0.50980395,0.54901963,0.5882353,0.4509804,0.78431374,0.6666667,0.60784316,0.64705884,0.84313726,0.60784316,0.6862745,0.60784316,0.627451,0.5882353,0.64705884,0.74509805,0.7058824,0.6666667,0.8039216,0.8039216,0.84313726,0.7254902,0.7647059,0.7254902,0.74509805,0.8627451,0.8627451,0.88235295,0.84313726,0.9607843,0.9607843,0.92156863,0.98039216,0.9607843
+TorchMLP (tuned + ensembled),0.039215688,0.05882353,0.09803922,0.078431375,0.13725491,0.13725491,0.13725491,0.078431375,0.15686275,0.15686275,0.25490198,0.19607843,0.25490198,0.25490198,0.23529412,0.23529412,0.23529412,0.25490198,0.3529412,0.3529412,0.33333334,0.27450982,0.33333334,0.33333334,0.29411766,0.33333334,0.43137255,0.3529412,0.37254903,0.39215687,0.4117647,0.4117647,0.4509804,0.49019608,0.43137255,0.49019608,,0.47058824,0.50980395,0.4509804,0.5294118,0.50980395,0.74509805,0.627451,0.54901963,0.64705884,0.6666667,0.64705884,0.6862745,0.6862745,0.74509805,0.78431374,0.7647059,0.7058824,0.74509805,0.7647059,0.9607843,0.7647059,0.7647059,0.7647059,0.8039216,0.88235295,0.8039216,0.88235295,0.9019608,0.92156863,0.9411765,0.9411765,0.98039216,0.9607843
+RealMLP (default),0.078431375,0.09803922,0.05882353,0.078431375,0.11764706,0.15686275,0.13725491,0.1764706,0.1764706,0.078431375,0.29411766,0.1764706,0.29411766,0.1764706,0.27450982,0.23529412,0.27450982,0.23529412,0.3529412,0.27450982,0.29411766,0.27450982,0.29411766,0.3137255,0.25490198,0.33333334,0.3529412,0.33333334,0.29411766,0.3529412,0.29411766,0.39215687,0.4117647,0.50980395,0.4117647,0.4509804,0.5294118,,0.49019608,0.47058824,0.5882353,0.5882353,0.60784316,0.60784316,0.5882353,0.6666667,0.64705884,0.6666667,0.6862745,0.60784316,0.7058824,0.74509805,0.7254902,0.7058824,0.7254902,0.7647059,0.8627451,0.7647059,0.7647059,0.8235294,0.8039216,0.9411765,0.88235295,0.88235295,0.92156863,0.92156863,0.9411765,0.9607843,0.98039216,1.0
+SAP-RPT-OSS (default),0.078431375,0.13725491,0.15686275,0.13725491,0.11764706,0.1764706,0.15686275,0.23529412,0.15686275,0.19607843,0.27450982,0.3137255,0.29411766,0.29411766,0.33333334,0.29411766,0.27450982,0.25490198,0.3529412,0.37254903,0.3529412,0.39215687,0.37254903,0.37254903,0.37254903,0.33333334,0.29411766,0.39215687,0.39215687,0.4117647,0.4117647,0.3529412,0.39215687,0.47058824,0.43137255,0.4117647,0.49019608,0.50980395,,0.47058824,0.50980395,0.54901963,0.54901963,0.54901963,0.5686275,0.5294118,0.64705884,0.627451,0.5686275,0.60784316,0.5882353,0.7254902,0.6666667,0.74509805,0.6862745,0.6862745,0.74509805,0.74509805,0.7254902,0.78431374,0.74509805,0.84313726,0.88235295,0.8039216,0.8235294,0.9019608,0.88235295,0.92156863,0.9019608,1.0
+ChimeraBoost (default),0.05882353,0.039215688,0.09803922,0.11764706,0.13725491,0.1764706,0.15686275,0.11764706,0.15686275,0.19607843,0.3529412,0.3137255,0.3137255,0.25490198,0.19607843,0.21568628,0.25490198,0.13725491,0.3137255,0.33333334,0.33333334,0.25490198,0.33333334,0.29411766,0.21568628,0.3529412,0.29411766,0.3137255,0.15686275,0.47058824,0.27450982,0.47058824,0.49019608,0.43137255,0.43137255,0.54901963,0.54901963,0.5294118,0.5294118,,0.627451,0.54901963,0.54901963,0.50980395,0.60784316,0.60784316,0.6666667,0.54901963,0.6666667,0.60784316,0.7254902,0.6862745,0.74509805,0.7647059,0.7647059,0.8039216,0.74509805,0.78431374,0.74509805,0.7647059,0.7647059,0.8039216,0.84313726,0.8235294,0.8235294,0.9019608,0.9019608,0.9607843,0.98039216,0.98039216
+EBM (tuned),0.05882353,0.09803922,0.11764706,0.05882353,0.11764706,0.15686275,0.13725491,0.09803922,0.11764706,0.15686275,0.27450982,0.21568628,0.29411766,0.23529412,0.1764706,0.23529412,0.3137255,0.25490198,0.3529412,0.33333334,0.27450982,0.25490198,0.27450982,0.25490198,0.3137255,0.3137255,0.3529412,0.25490198,0.27450982,0.37254903,0.33333334,0.3529412,0.4117647,0.49019608,0.39215687,0.21568628,0.47058824,0.4117647,0.49019608,0.37254903,,0.5686275,0.50980395,0.54901963,0.5882353,0.49019608,0.54901963,0.5882353,0.5686275,0.5294118,0.5882353,0.6862745,0.6862745,0.6666667,0.7647059,0.7647059,0.7647059,0.6666667,0.7647059,0.7058824,0.7254902,0.8235294,0.8627451,0.8627451,0.8039216,0.88235295,0.92156863,0.92156863,0.92156863,0.9607843
+ModernNCA (default),0.039215688,0.039215688,0.039215688,0.039215688,0.11764706,0.09803922,0.09803922,0.078431375,0.11764706,0.09803922,0.1764706,0.23529412,0.27450982,0.25490198,0.19607843,0.27450982,0.23529412,0.29411766,0.23529412,0.19607843,0.27450982,0.29411766,0.3137255,0.3529412,0.37254903,0.23529412,0.33333334,0.3137255,0.39215687,0.25490198,0.43137255,0.37254903,0.37254903,0.3529412,0.39215687,0.33333334,0.49019608,0.4117647,0.4509804,0.4509804,0.43137255,,0.47058824,0.5686275,0.4509804,0.47058824,0.5294118,0.54901963,0.6666667,0.5882353,0.7058824,0.64705884,0.6862745,0.7058824,0.78431374,0.7647059,0.84313726,0.78431374,0.7254902,0.8235294,0.8627451,0.84313726,0.8235294,0.8627451,0.88235295,0.8627451,0.8627451,0.9411765,0.9019608,0.98039216
+TorchMLP (tuned),0.019607844,0.078431375,0.039215688,0.039215688,0.09803922,0.11764706,0.078431375,0.05882353,0.11764706,0.15686275,0.23529412,0.13725491,0.1764706,0.19607843,0.19607843,0.1764706,0.15686275,0.1764706,0.25490198,0.21568628,0.19607843,0.2647059,0.33333334,0.23529412,0.23529412,0.23529412,0.33333334,0.27450982,0.29411766,0.33333334,0.37254903,0.3529412,0.33333334,0.43137255,0.3137255,0.39215687,0.25490198,0.39215687,0.4509804,0.4509804,0.49019608,0.5294118,,0.5392157,0.47058824,0.49019608,0.50980395,0.5882353,0.5882353,0.5588235,0.5882353,0.627451,0.6666667,0.7058824,0.7647059,0.74509805,0.92156863,0.7254902,0.74509805,0.7254902,0.6666667,0.8039216,0.84313726,0.84313726,0.84313726,0.84313726,0.8627451,0.9411765,0.9411765,0.9607843
+ExtraTrees (tuned + ensembled),0.039215688,0.05882353,0.039215688,0.05882353,0.05882353,0.11764706,0.039215688,0.13725491,0.13725491,0.13725491,0.21568628,0.23529412,0.1764706,0.25490198,0.15686275,0.21568628,0.15686275,0.1764706,0.27450982,0.23529412,0.27450982,0.1764706,0.33333334,0.27450982,0.27450982,0.33333334,0.33333334,0.1764706,0.33333334,0.25490198,0.37254903,0.3137255,0.27450982,0.39215687,0.37254903,0.3529412,0.37254903,0.39215687,0.4509804,0.49019608,0.4509804,0.43137255,0.46078432,,0.43137255,0.43137255,0.50980395,0.71568626,0.64705884,0.50980395,0.7058824,0.54901963,0.64705884,0.64705884,0.7058824,0.7254902,0.60784316,0.7254902,0.7254902,0.9411765,0.88235295,0.7647059,0.78431374,0.92156863,0.78431374,0.8235294,0.88235295,0.9411765,0.88235295,1.0
+EBM (default),0.05882353,0.078431375,0.09803922,0.09803922,0.09803922,0.09803922,0.11764706,0.09803922,0.11764706,0.13725491,0.21568628,0.19607843,0.23529412,0.19607843,0.13725491,0.1764706,0.23529412,0.1764706,0.27450982,0.27450982,0.27450982,0.21568628,0.25490198,0.21568628,0.23529412,0.29411766,0.27450982,0.23529412,0.27450982,0.3529412,0.25490198,0.25490198,0.43137255,0.43137255,0.4019608,0.15686275,0.4509804,0.4117647,0.43137255,0.39215687,0.4117647,0.54901963,0.5294118,0.5686275,,0.4117647,0.5686275,0.5882353,0.5882353,0.50980395,0.627451,0.6666667,0.6862745,0.6862745,0.7254902,0.7254902,0.78431374,0.6862745,0.6666667,0.7058824,0.6666667,0.78431374,0.84313726,0.84313726,0.8235294,0.9019608,0.92156863,0.92156863,0.9607843,0.9607843
+XGBoost (default),0.039215688,0.05882353,0.05882353,0.039215688,0.039215688,0.13725491,0.05882353,0.078431375,0.11764706,0.09803922,0.25490198,0.15686275,0.21568628,0.21568628,0.09803922,0.13725491,0.1764706,0.13725491,0.3137255,0.21568628,0.21568628,0.11764706,0.29411766,0.1764706,0.23529412,0.25490198,0.33333334,0.13725491,0.27450982,0.3137255,0.25490198,0.39215687,0.39215687,0.4117647,0.27450982,0.39215687,0.3529412,0.33333334,0.47058824,0.39215687,0.50980395,0.5294118,0.50980395,0.5686275,0.5882353,,0.5686275,0.5882353,0.5882353,0.5294118,0.64705884,0.627451,0.78431374,0.627451,0.7254902,0.7254902,0.74509805,0.7254902,0.74509805,0.7254902,0.78431374,0.7647059,0.78431374,0.78431374,0.8627451,0.84313726,0.8627451,0.9019608,0.88235295,0.9607843
+FastaiMLP (tuned + ensembled),0.078431375,0.09803922,0.09803922,0.078431375,0.11764706,0.05882353,0.13725491,0.078431375,0.078431375,0.078431375,0.25490198,0.15686275,0.21568628,0.1764706,0.1764706,0.21568628,0.13725491,0.21568628,0.33333334,0.19607843,0.25490198,0.21568628,0.3137255,0.21568628,0.19607843,0.28431374,0.1764706,0.23529412,0.23529412,0.27450982,0.27450982,0.21568628,0.3137255,0.3529412,0.25490198,0.3137255,0.33333334,0.3529412,0.3529412,0.33333334,0.4509804,0.47058824,0.49019608,0.49019608,0.43137255,0.43137255,,0.49019608,0.5294118,0.4509804,0.5686275,0.7058824,0.64705884,0.60784316,0.6666667,0.6666667,0.6666667,0.627451,0.7254902,0.7254902,0.7058824,0.98039216,0.7647059,0.78431374,0.84313726,0.8627451,0.88235295,0.8627451,0.92156863,0.9607843
+ExtraTrees (tuned),0.039215688,0.05882353,0.039215688,0.05882353,0.05882353,0.078431375,0.09803922,0.13725491,0.13725491,0.13725491,0.13725491,0.1764706,0.15686275,0.25490198,0.19607843,0.23529412,0.1764706,0.1764706,0.1764706,0.23529412,0.27450982,0.19607843,0.33333334,0.23529412,0.27450982,0.3137255,0.29411766,0.19607843,0.3137255,0.21568628,0.33333334,0.25490198,0.25490198,0.33333334,0.33333334,0.39215687,0.3529412,0.33333334,0.37254903,0.4509804,0.4117647,0.4509804,0.4117647,0.28431374,0.4117647,0.4117647,0.50980395,,0.47058824,0.43137255,0.60784316,0.5294118,0.5882353,0.64705884,0.6862745,0.6862745,0.54901963,0.7058824,0.6666667,0.8235294,0.84313726,0.6862745,0.8039216,0.84313726,0.7647059,0.7647059,0.7647059,0.98039216,0.88235295,0.98039216
+RandomForest (tuned + ensembled),0.039215688,0.039215688,0.05882353,0.039215688,0.078431375,0.11764706,0.078431375,0.09803922,0.13725491,0.15686275,0.21568628,0.15686275,0.1764706,0.19607843,0.13725491,0.15686275,0.09803922,0.11764706,0.3137255,0.21568628,0.25490198,0.13725491,0.29411766,0.21568628,0.19607843,0.27450982,0.3137255,0.1764706,0.23529412,0.19607843,0.27450982,0.19607843,0.25490198,0.33333334,0.29411766,0.37254903,0.3137255,0.3137255,0.43137255,0.33333334,0.43137255,0.33333334,0.4117647,0.3529412,0.4117647,0.4117647,0.47058824,0.5294118,,0.49019608,0.7058824,0.54901963,0.627451,0.5686275,0.6862745,0.7058824,0.6862745,0.6666667,0.6862745,0.8039216,0.88235295,0.74509805,0.8039216,0.84313726,0.6862745,0.8039216,0.78431374,0.9411765,0.8627451,0.98039216
+LightGBM (default),0.039215688,0.019607844,0.039215688,0.039215688,0.019607844,0.078431375,0.039215688,0.05882353,0.039215688,0.05882353,0.1764706,0.15686275,0.13725491,0.13725491,0.039215688,0.15686275,0.13725491,0.15686275,0.27450982,0.15686275,0.1764706,0.019607844,0.3137255,0.13725491,0.21568628,0.21568628,0.23529412,0.09803922,0.25490198,0.25490198,0.23529412,0.25490198,0.3137255,0.3137255,0.27450982,0.4117647,0.3137255,0.39215687,0.39215687,0.39215687,0.47058824,0.4117647,0.44117647,0.49019608,0.49019608,0.47058824,0.54901963,0.5686275,0.50980395,,0.54901963,0.60784316,0.7254902,0.6666667,0.7058824,0.7254902,0.7058824,0.7058824,0.7254902,0.7647059,0.7647059,0.78431374,0.8039216,0.7647059,0.8039216,0.78431374,0.78431374,0.9411765,0.9019608,0.9411765
+RandomForest (tuned),0.039215688,0.039215688,0.019607844,0.05882353,0.05882353,0.11764706,0.078431375,0.039215688,0.078431375,0.09803922,0.19607843,0.11764706,0.13725491,0.19607843,0.05882353,0.09803922,0.09803922,0.078431375,0.21568628,0.23529412,0.21568628,0.09803922,0.25490198,0.11764706,0.15686275,0.21568628,0.29411766,0.13725491,0.19607843,0.1764706,0.19607843,0.19607843,0.21568628,0.29411766,0.27450982,0.3529412,0.25490198,0.29411766,0.4117647,0.27450982,0.4117647,0.29411766,0.4117647,0.29411766,0.37254903,0.3529412,0.43137255,0.39215687,0.29411766,0.4509804,,0.50980395,0.627451,0.54901963,0.6666667,0.6666667,0.5882353,0.5882353,0.6666667,0.7254902,0.78431374,0.7647059,0.7647059,0.84313726,0.6666667,0.78431374,0.7647059,0.92156863,0.84313726,0.98039216
+FastaiMLP (tuned),0.05882353,0.05882353,0.039215688,0.078431375,0.05882353,0.078431375,0.078431375,0.05882353,0.019607844,0.09803922,0.15686275,0.11764706,0.09803922,0.11764706,0.13725491,0.15686275,0.13725491,0.11764706,0.1764706,0.13725491,0.13725491,0.1764706,0.21568628,0.15686275,0.15686275,0.1764706,0.19607843,0.15686275,0.25490198,0.21568628,0.23529412,0.19607843,0.25490198,0.21568628,0.15686275,0.25490198,0.21568628,0.25490198,0.27450982,0.3137255,0.3137255,0.3529412,0.37254903,0.4509804,0.33333334,0.37254903,0.29411766,0.47058824,0.4509804,0.39215687,0.49019608,,0.5686275,0.5294118,0.627451,0.627451,0.5882353,0.5686275,0.64705884,0.6862745,0.627451,0.78431374,0.74509805,0.6862745,0.7254902,0.78431374,0.78431374,0.84313726,0.88235295,0.9411765
+iLTM (default),0.05882353,0.039215688,0.019607844,0.039215688,0.078431375,0.11764706,0.09803922,0.05882353,0.078431375,0.078431375,0.19607843,0.09803922,0.15686275,0.15686275,0.078431375,0.11764706,0.09803922,0.11764706,0.1764706,0.1764706,0.21568628,0.11764706,0.23529412,0.13725491,0.13725491,0.1764706,0.21568628,0.13725491,0.21568628,0.23529412,0.21568628,0.2254902,0.23529412,0.3137255,0.27450982,0.29411766,0.23529412,0.27450982,0.33333334,0.25490198,0.3137255,0.3137255,0.33333334,0.3529412,0.3137255,0.21568628,0.3529412,0.4117647,0.37254903,0.27450982,0.37254903,0.43137255,,0.47058824,0.6666667,0.64705884,0.5686275,0.4509804,0.5294118,0.5294118,0.54901963,0.627451,0.7647059,0.7647059,0.5882353,0.7647059,0.7647059,0.8235294,0.84313726,0.92156863
+PerpetualBooster (tuned + ensembled),0.0,0.05882353,0.039215688,0.019607844,0.039215688,0.078431375,0.039215688,0.039215688,0.05882353,0.039215688,0.15686275,0.09803922,0.15686275,0.15686275,0.09803922,0.11764706,0.09803922,0.078431375,0.15686275,0.15686275,0.11764706,0.15686275,0.25490198,0.11764706,0.09803922,0.1764706,0.19607843,0.19607843,0.1764706,0.19607843,0.19607843,0.23529412,0.25490198,0.25490198,0.15686275,0.33333334,0.29411766,0.29411766,0.25490198,0.23529412,0.33333334,0.29411766,0.29411766,0.3529412,0.3137255,0.37254903,0.39215687,0.3529412,0.43137255,0.33333334,0.4509804,0.47058824,0.5294118,,0.47058824,0.47058824,0.49019608,0.8137255,0.60784316,0.5882353,0.64705884,0.60784316,0.49019608,0.60784316,0.9019608,0.74509805,0.74509805,0.74509805,0.7647059,0.92156863
+TabSTAR (tuned),0.05882353,0.09803922,0.09803922,0.11764706,0.11764706,0.09803922,0.13725491,0.1764706,0.09803922,0.15686275,0.1764706,0.15686275,0.19607843,0.13725491,0.15686275,0.1764706,0.19607843,0.1764706,0.19607843,0.25490198,0.1764706,0.15686275,0.2254902,0.19607843,0.15686275,0.21568628,0.21568628,0.15686275,0.1764706,0.23529412,0.21568628,0.21568628,0.25490198,0.29411766,0.23529412,0.19607843,0.25490198,0.27450982,0.3137255,0.23529412,0.23529412,0.21568628,0.23529412,0.29411766,0.27450982,0.27450982,0.33333334,0.3137255,0.3137255,0.29411766,0.33333334,0.37254903,0.33333334,0.5294118,,0.51960784,0.43137255,0.5294118,0.5294118,0.5686275,0.54901963,0.54901963,0.9607843,0.5882353,0.54901963,0.7058824,0.7254902,0.78431374,0.8039216,0.84313726
+TabSTAR (tuned + ensembled),0.078431375,0.11764706,0.09803922,0.11764706,0.11764706,0.09803922,0.15686275,0.1764706,0.09803922,0.15686275,0.21568628,0.13725491,0.21568628,0.13725491,0.1764706,0.1764706,0.19607843,0.1764706,0.21568628,0.23529412,0.1764706,0.15686275,0.23529412,0.1764706,0.1764706,0.21568628,0.21568628,0.15686275,0.19607843,0.23529412,0.21568628,0.21568628,0.23529412,0.27450982,0.21568628,0.19607843,0.23529412,0.23529412,0.3137255,0.19607843,0.23529412,0.23529412,0.25490198,0.27450982,0.27450982,0.27450982,0.33333334,0.3137255,0.29411766,0.27450982,0.33333334,0.37254903,0.3529412,0.5294118,0.48039216,,0.4509804,0.5294118,0.5294118,0.54901963,0.5294118,0.5294118,0.98039216,0.5686275,0.54901963,0.7058824,0.7058824,0.78431374,0.8039216,0.84313726
+TorchMLP (default),0.019607844,0.039215688,0.039215688,0.019607844,0.039215688,0.039215688,0.039215688,0.019607844,0.019607844,0.05882353,0.13725491,0.039215688,0.11764706,0.019607844,0.078431375,0.05882353,0.039215688,0.05882353,0.13725491,0.15686275,0.019607844,0.078431375,0.19607843,0.11764706,0.09803922,0.13725491,0.1764706,0.09803922,0.1764706,0.11764706,0.1764706,0.078431375,0.09803922,0.21568628,0.11764706,0.15686275,0.039215688,0.13725491,0.25490198,0.25490198,0.23529412,0.15686275,0.078431375,0.39215687,0.21568628,0.25490198,0.33333334,0.4509804,0.3137255,0.29411766,0.4117647,0.4117647,0.43137255,0.50980395,0.5686275,0.54901963,,0.5294118,0.54901963,0.64705884,0.5882353,0.54901963,0.7058824,0.6666667,0.627451,0.7058824,0.7647059,0.84313726,0.84313726,0.92156863
+PerpetualBooster (tuned),0.0,0.039215688,0.0,0.0,0.019607844,0.05882353,0.019607844,0.019607844,0.019607844,0.019607844,0.09803922,0.078431375,0.09803922,0.09803922,0.078431375,0.05882353,0.05882353,0.078431375,0.09803922,0.15686275,0.09803922,0.11764706,0.23529412,0.09803922,0.05882353,0.15686275,0.15686275,0.11764706,0.13725491,0.15686275,0.1764706,0.15686275,0.1764706,0.1764706,0.15686275,0.27450982,0.23529412,0.23529412,0.25490198,0.21568628,0.33333334,0.21568628,0.27450982,0.27450982,0.3137255,0.27450982,0.37254903,0.29411766,0.33333334,0.29411766,0.4117647,0.43137255,0.54901963,0.18627451,0.47058824,0.47058824,0.47058824,,0.5686275,0.50980395,0.5882353,0.5882353,0.49019608,0.54901963,0.8137255,0.7254902,0.7058824,0.6666667,0.74509805,0.92156863
+xRFM (default),0.039215688,0.019607844,0.039215688,0.039215688,0.05882353,0.078431375,0.078431375,0.078431375,0.078431375,0.05882353,0.05882353,0.078431375,0.09803922,0.15686275,0.15686275,0.15686275,0.078431375,0.11764706,0.09803922,0.09803922,0.1764706,0.1764706,0.1764706,0.15686275,0.13725491,0.15686275,0.1764706,0.21568628,0.19607843,0.078431375,0.19607843,0.11764706,0.039215688,0.19607843,0.25490198,0.23529412,0.23529412,0.23529412,0.27450982,0.25490198,0.23529412,0.27450982,0.25490198,0.27450982,0.33333334,0.25490198,0.27450982,0.33333334,0.3137255,0.27450982,0.33333334,0.3529412,0.47058824,0.39215687,0.47058824,0.47058824,0.4509804,0.43137255,,0.49019608,0.5294118,0.49019608,0.60784316,0.5294118,0.54901963,0.60784316,0.60784316,0.7254902,0.64705884,0.88235295
+ExtraTrees (default),0.019607844,0.019607844,0.019607844,0.039215688,0.019607844,0.078431375,0.039215688,0.019607844,0.019607844,0.019607844,0.019607844,0.09803922,0.039215688,0.13725491,0.078431375,0.11764706,0.039215688,0.09803922,0.05882353,0.13725491,0.15686275,0.078431375,0.11764706,0.15686275,0.13725491,0.1764706,0.11764706,0.09803922,0.21568628,0.05882353,0.19607843,0.13725491,0.078431375,0.15686275,0.19607843,0.27450982,0.23529412,0.1764706,0.21568628,0.23529412,0.29411766,0.1764706,0.27450982,0.05882353,0.29411766,0.27450982,0.27450982,0.1764706,0.19607843,0.23529412,0.27450982,0.3137255,0.47058824,0.4117647,0.43137255,0.4509804,0.3529412,0.49019608,0.50980395,,0.50980395,0.54901963,0.5882353,0.5686275,0.60784316,0.6666667,0.6666667,0.7647059,0.7058824,0.9607843
+RandomForest (default),0.019607844,0.039215688,0.019607844,0.019607844,0.019607844,0.05882353,0.039215688,0.019607844,0.05882353,0.039215688,0.039215688,0.05882353,0.05882353,0.11764706,0.019607844,0.05882353,0.039215688,0.039215688,0.11764706,0.09803922,0.09803922,0.039215688,0.11764706,0.05882353,0.078431375,0.09803922,0.1764706,0.05882353,0.13725491,0.078431375,0.19607843,0.09803922,0.13725491,0.21568628,0.1764706,0.25490198,0.19607843,0.19607843,0.25490198,0.23529412,0.27450982,0.13725491,0.33333334,0.11764706,0.33333334,0.21568628,0.29411766,0.15686275,0.11764706,0.23529412,0.21568628,0.37254903,0.4509804,0.3529412,0.4509804,0.47058824,0.4117647,0.4117647,0.47058824,0.49019608,,0.54901963,0.5882353,0.60784316,0.5882353,0.7058824,0.7058824,0.78431374,0.74509805,0.98039216
+FastaiMLP (default),0.0,0.019607844,0.0,0.019607844,0.019607844,0.019607844,0.039215688,0.019607844,0.039215688,0.019607844,0.039215688,0.039215688,0.039215688,0.05882353,0.078431375,0.078431375,0.039215688,0.078431375,0.039215688,0.11764706,0.09803922,0.09803922,0.13725491,0.09803922,0.05882353,0.09803922,0.09803922,0.09803922,0.13725491,0.078431375,0.15686275,0.09803922,0.11764706,0.13725491,0.15686275,0.13725491,0.11764706,0.05882353,0.15686275,0.19607843,0.1764706,0.15686275,0.19607843,0.23529412,0.21568628,0.23529412,0.019607844,0.3137255,0.25490198,0.21568628,0.23529412,0.21568628,0.37254903,0.39215687,0.4509804,0.47058824,0.4509804,0.4117647,0.50980395,0.4509804,0.4509804,,0.6666667,0.5294118,0.6666667,0.6862745,0.7058824,0.74509805,0.74509805,0.9019608
+TabSTAR (default),0.039215688,0.09803922,0.09803922,0.11764706,0.09803922,0.078431375,0.09803922,0.11764706,0.078431375,0.11764706,0.11764706,0.13725491,0.13725491,0.11764706,0.13725491,0.15686275,0.13725491,0.15686275,0.13725491,0.13725491,0.13725491,0.13725491,0.15686275,0.13725491,0.13725491,0.11764706,0.13725491,0.13725491,0.15686275,0.1764706,0.15686275,0.13725491,0.1764706,0.23529412,0.13725491,0.13725491,0.19607843,0.11764706,0.11764706,0.15686275,0.13725491,0.1764706,0.15686275,0.21568628,0.15686275,0.21568628,0.23529412,0.19607843,0.19607843,0.19607843,0.23529412,0.25490198,0.23529412,0.50980395,0.039215688,0.019607844,0.29411766,0.50980395,0.39215687,0.4117647,0.4117647,0.33333334,,0.50980395,0.5294118,0.5882353,0.5882353,0.64705884,0.64705884,0.8039216
+KNN (tuned + ensembled),0.05882353,0.019607844,0.039215688,0.019607844,0.05882353,0.039215688,0.078431375,0.05882353,0.078431375,0.039215688,0.078431375,0.05882353,0.05882353,0.09803922,0.05882353,0.078431375,0.039215688,0.09803922,0.11764706,0.10784314,0.09803922,0.09803922,0.11764706,0.09803922,0.078431375,0.078431375,0.09803922,0.078431375,0.13725491,0.05882353,0.15686275,0.05882353,0.05882353,0.1764706,0.15686275,0.11764706,0.11764706,0.11764706,0.19607843,0.1764706,0.13725491,0.13725491,0.15686275,0.078431375,0.15686275,0.21568628,0.21568628,0.15686275,0.15686275,0.23529412,0.15686275,0.3137255,0.23529412,0.39215687,0.4117647,0.43137255,0.33333334,0.4509804,0.47058824,0.43137255,0.39215687,0.47058824,0.49019608,,0.54901963,0.6862745,0.6862745,0.98039216,0.7647059,0.98039216
+PerpetualBooster (default),0.0,0.039215688,0.019607844,0.019607844,0.039215688,0.039215688,0.039215688,0.019607844,0.05882353,0.019607844,0.078431375,0.019607844,0.039215688,0.05882353,0.078431375,0.039215688,0.039215688,0.039215688,0.05882353,0.09803922,0.05882353,0.09803922,0.13725491,0.078431375,0.039215688,0.11764706,0.09803922,0.09803922,0.13725491,0.11764706,0.13725491,0.09803922,0.078431375,0.13725491,0.09803922,0.15686275,0.09803922,0.078431375,0.1764706,0.1764706,0.19607843,0.11764706,0.15686275,0.21568628,0.1764706,0.13725491,0.15686275,0.23529412,0.3137255,0.19607843,0.33333334,0.27450982,0.4117647,0.09803922,0.4509804,0.4509804,0.37254903,0.18627451,0.4509804,0.39215687,0.4117647,0.33333334,0.47058824,0.4509804,,0.60784316,0.60784316,0.54901963,0.64705884,0.84313726
+Linear (tuned + ensembled),0.019607844,0.019607844,0.039215688,0.019607844,0.05882353,0.039215688,0.05882353,0.019607844,0.039215688,0.019607844,0.09803922,0.039215688,0.11764706,0.039215688,0.019607844,0.019607844,0.05882353,0.019607844,0.13725491,0.13725491,0.078431375,0.05882353,0.09803922,0.09803922,0.039215688,0.078431375,0.11764706,0.078431375,0.05882353,0.039215688,0.09803922,0.11764706,0.078431375,0.21568628,0.078431375,0.039215688,0.078431375,0.078431375,0.09803922,0.09803922,0.11764706,0.13725491,0.15686275,0.1764706,0.09803922,0.15686275,0.13725491,0.23529412,0.19607843,0.21568628,0.21568628,0.21568628,0.23529412,0.25490198,0.29411766,0.29411766,0.29411766,0.27450982,0.39215687,0.33333334,0.29411766,0.3137255,0.4117647,0.3137255,0.39215687,,0.7647059,0.49019608,0.7647059,0.7254902
+Linear (tuned),0.019607844,0.019607844,0.019607844,0.019607844,0.039215688,0.019607844,0.05882353,0.019607844,0.039215688,0.019607844,0.09803922,0.029411765,0.11764706,0.039215688,0.019607844,0.019607844,0.039215688,0.019607844,0.13725491,0.11764706,0.078431375,0.039215688,0.09803922,0.05882353,0.019607844,0.05882353,0.09803922,0.05882353,0.039215688,0.019607844,0.09803922,0.078431375,0.09803922,0.19607843,0.09803922,0.039215688,0.05882353,0.05882353,0.11764706,0.09803922,0.078431375,0.13725491,0.13725491,0.11764706,0.078431375,0.13725491,0.11764706,0.23529412,0.21568628,0.21568628,0.23529412,0.21568628,0.23529412,0.25490198,0.27450982,0.29411766,0.23529412,0.29411766,0.39215687,0.33333334,0.29411766,0.29411766,0.4117647,0.3137255,0.39215687,0.23529412,,0.47058824,0.6862745,0.7058824
+KNN (tuned),0.019607844,0.0,0.019607844,0.019607844,0.039215688,0.019607844,0.039215688,0.0,0.019607844,0.019607844,0.039215688,0.019607844,0.019607844,0.0,0.0,0.019607844,0.039215688,0.039215688,0.039215688,0.09803922,0.019607844,0.05882353,0.11764706,0.039215688,0.019607844,0.05882353,0.019607844,0.05882353,0.039215688,0.019607844,0.05882353,0.019607844,0.019607844,0.09803922,0.019607844,0.078431375,0.05882353,0.039215688,0.078431375,0.039215688,0.078431375,0.05882353,0.05882353,0.05882353,0.078431375,0.09803922,0.13725491,0.019607844,0.05882353,0.05882353,0.078431375,0.15686275,0.1764706,0.25490198,0.21568628,0.21568628,0.15686275,0.33333334,0.27450982,0.23529412,0.21568628,0.25490198,0.3529412,0.019607844,0.4509804,0.50980395,0.5294118,,0.54901963,0.92156863
+Linear (default),0.0,0.019607844,0.0,0.019607844,0.039215688,0.0,0.039215688,0.0,0.019607844,0.019607844,0.078431375,0.039215688,0.078431375,0.039215688,0.039215688,0.019607844,0.039215688,0.019607844,0.09803922,0.11764706,0.039215688,0.039215688,0.078431375,0.05882353,0.019607844,0.05882353,0.09803922,0.039215688,0.019607844,0.039215688,0.039215688,0.078431375,0.078431375,0.13725491,0.078431375,0.019607844,0.019607844,0.019607844,0.09803922,0.019607844,0.078431375,0.09803922,0.05882353,0.11764706,0.039215688,0.11764706,0.078431375,0.11764706,0.13725491,0.09803922,0.15686275,0.11764706,0.15686275,0.23529412,0.19607843,0.19607843,0.15686275,0.25490198,0.3529412,0.29411766,0.25490198,0.25490198,0.3529412,0.23529412,0.3529412,0.23529412,0.3137255,0.4509804,,0.6666667
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.019607844,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.019607844,0.019607844,0.019607844,0.0,0.019607844,0.0,0.0,0.0,0.0,0.0,0.0,0.019607844,0.0,0.0,0.0,0.019607844,0.039215688,0.039215688,0.039215688,0.0,0.0,0.019607844,0.039215688,0.019607844,0.039215688,0.0,0.039215688,0.039215688,0.039215688,0.019607844,0.019607844,0.05882353,0.019607844,0.05882353,0.078431375,0.078431375,0.15686275,0.15686275,0.078431375,0.078431375,0.11764706,0.039215688,0.019607844,0.09803922,0.19607843,0.019607844,0.15686275,0.27450982,0.29411766,0.078431375,0.33333334,
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_medium/leaderboard_overview_explorer.html
index 1d906a1ba2c41b294c74b8a9e66d31242b697bb0..1abaf65db16709b076012f8189ea2d1528f71974 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | All Tasks | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_all/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..67548ab5be7433d17e498bac6fdf4f6341ae6dd2
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_all/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1941.0,"score":0.774,"rank":5.53,"hrank":1.96,"imp":4.458,"train_time":27.18,"predict_time":13.118,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+240\/-134","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1938.0,"score":0.729,"rank":5.6,"hrank":2.55,"imp":3.418,"train_time":182.61,"predict_time":2.229,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+123\/-57","imputed":false},{"position":2,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1870.0,"score":0.651,"rank":7.33,"hrank":3.53,"imp":4.717,"train_time":432.15,"predict_time":4.214,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+136\/-67","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1801.0,"score":0.67,"rank":9.53,"hrank":2.7,"imp":4.59,"train_time":1.02,"predict_time":0.214,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+236\/-138","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1765.0,"score":0.526,"rank":10.87,"hrank":8.26,"imp":7.097,"train_time":2.64,"predict_time":0.702,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-65","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1713.0,"score":0.496,"rank":13.0,"hrank":7.2,"imp":6.49,"train_time":739.1,"predict_time":11.65,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+163\/-106","imputed":false},{"position":6,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1710.0,"score":0.52,"rank":13.13,"hrank":5.58,"imp":7.508,"train_time":0.56,"predict_time":0.118,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+206\/-125","imputed":false},{"position":7,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1668.0,"score":0.481,"rank":15.07,"hrank":9.87,"imp":6.901,"train_time":2522.59,"predict_time":6.234,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-119","imputed":false},{"position":8,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1662.0,"score":0.456,"rank":15.4,"hrank":9.51,"imp":7.814,"train_time":739.1,"predict_time":1.134,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-99","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1619.0,"score":0.426,"rank":17.6,"hrank":12.43,"imp":8.359,"train_time":1714.29,"predict_time":1.743,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-79","imputed":false},{"position":10,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1583.0,"score":0.345,"rank":19.57,"hrank":11.1,"imp":9.211,"train_time":477.59,"predict_time":0.189,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-74","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1582.0,"score":0.328,"rank":19.6,"hrank":15.31,"imp":10.072,"train_time":129.25,"predict_time":2.621,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-58","imputed":false},{"position":12,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1570.0,"score":0.331,"rank":20.27,"hrank":11.76,"imp":9.407,"train_time":477.59,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-74","imputed":false},{"position":13,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1554.0,"score":0.325,"rank":21.2,"hrank":14.19,"imp":9.628,"train_time":1176.25,"predict_time":0.556,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+144\/-91","imputed":false},{"position":14,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1551.0,"score":0.298,"rank":21.4,"hrank":16.05,"imp":9.808,"train_time":289.35,"predict_time":0.706,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-67","imputed":false},{"position":15,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1546.0,"score":0.333,"rank":21.7,"hrank":14.14,"imp":10.198,"train_time":2522.59,"predict_time":0.318,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-109","imputed":false},{"position":16,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1535.0,"score":0.314,"rank":22.33,"hrank":15.14,"imp":9.974,"train_time":2045.59,"predict_time":69.586,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-87","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1533.0,"score":0.273,"rank":22.47,"hrank":18.5,"imp":10.68,"train_time":129.25,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-76","imputed":false},{"position":18,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1532.0,"score":0.333,"rank":22.53,"hrank":12.4,"imp":9.245,"train_time":1.88,"predict_time":0.011,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-92","imputed":false},{"position":19,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1520.0,"score":0.319,"rank":23.2,"hrank":13.85,"imp":10.256,"train_time":1714.29,"predict_time":0.079,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+111\/-105","imputed":false},{"position":20,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1519.0,"score":0.324,"rank":23.27,"hrank":13.93,"imp":8.951,"train_time":1.91,"predict_time":0.668,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+142\/-110","imputed":false},{"position":21,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1518.0,"score":0.327,"rank":23.33,"hrank":15.12,"imp":11.301,"train_time":92.46,"predict_time":0.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-118","imputed":false},{"position":22,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1499.0,"score":0.261,"rank":24.53,"hrank":19.21,"imp":10.049,"train_time":289.35,"predict_time":0.08,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-86","imputed":false},{"position":23,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1496.0,"score":0.306,"rank":24.73,"hrank":17.71,"imp":11.401,"train_time":92.46,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-104","imputed":false},{"position":24,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1490.0,"score":0.276,"rank":25.07,"hrank":14.45,"imp":10.481,"train_time":1176.25,"predict_time":0.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+171\/-87","imputed":false},{"position":25,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1477.0,"score":0.288,"rank":25.87,"hrank":16.46,"imp":10.969,"train_time":0.76,"predict_time":0.178,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-143","imputed":false},{"position":26,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1470.0,"score":0.269,"rank":26.33,"hrank":8.07,"imp":10.481,"train_time":4532.19,"predict_time":387.89,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-117","imputed":false},{"position":27,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1418.0,"score":0.238,"rank":29.7,"hrank":13.21,"imp":12.916,"train_time":4532.19,"predict_time":39.452,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-138","imputed":false},{"position":28,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1417.0,"score":0.242,"rank":29.73,"hrank":21.97,"imp":11.377,"train_time":559.64,"predict_time":0.88,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-124","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1408.0,"score":0.225,"rank":30.33,"hrank":20.72,"imp":12.413,"train_time":0.48,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+147\/-123","imputed":false},{"position":30,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1392.0,"score":0.152,"rank":31.33,"hrank":21.3,"imp":11.952,"train_time":2045.59,"predict_time":6.231,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-87","imputed":false},{"position":31,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1386.0,"score":0.16,"rank":31.73,"hrank":26.23,"imp":13.54,"train_time":559.64,"predict_time":0.051,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-105","imputed":false},{"position":32,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1380.0,"score":0.197,"rank":32.13,"hrank":10.23,"imp":11.836,"train_time":0.48,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-127","imputed":false},{"position":33,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1379.0,"score":0.194,"rank":32.2,"hrank":17.86,"imp":12.151,"train_time":0.51,"predict_time":0.126,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+151\/-129","imputed":false},{"position":34,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1349.0,"score":0.123,"rank":34.13,"hrank":28.25,"imp":12.29,"train_time":1803.23,"predict_time":1.199,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-68","imputed":false},{"position":35,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1342.0,"score":0.109,"rank":34.6,"hrank":31.16,"imp":13.285,"train_time":1.94,"predict_time":0.046,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-80","imputed":false},{"position":36,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1325.0,"score":0.153,"rank":35.73,"hrank":26.34,"imp":15.154,"train_time":662.24,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-139","imputed":false},{"position":37,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1298.0,"score":0.202,"rank":37.47,"hrank":9.57,"imp":15.204,"train_time":6.51,"predict_time":5.361,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+169\/-159","imputed":false},{"position":38,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1284.0,"score":0.155,"rank":38.37,"hrank":22.45,"imp":15.733,"train_time":41.51,"predict_time":39.094,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-139","imputed":false},{"position":39,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1279.0,"score":0.089,"rank":38.67,"hrank":34.51,"imp":13.775,"train_time":9.07,"predict_time":0.224,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-96","imputed":false},{"position":40,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1279.0,"score":0.061,"rank":38.67,"hrank":35.7,"imp":15.022,"train_time":6.54,"predict_time":0.294,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-82","imputed":false},{"position":41,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1242.0,"score":0.039,"rank":41.0,"hrank":35.55,"imp":14.895,"train_time":65.73,"predict_time":0.23,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-92","imputed":false},{"position":42,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1238.0,"score":0.056,"rank":41.27,"hrank":38.35,"imp":14.649,"train_time":1803.23,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-83","imputed":false},{"position":43,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1229.0,"score":0.043,"rank":41.87,"hrank":39.77,"imp":14.26,"train_time":0.34,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-86","imputed":false},{"position":44,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1224.0,"score":0.033,"rank":42.13,"hrank":39.71,"imp":15.006,"train_time":50.36,"predict_time":0.183,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-116","imputed":false},{"position":45,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.087,"rank":42.2,"hrank":34.87,"imp":16.389,"train_time":662.24,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-140","imputed":false},{"position":46,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1210.0,"score":0.029,"rank":43.03,"hrank":39.34,"imp":15.23,"train_time":65.73,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-80","imputed":false},{"position":47,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1206.0,"score":0.075,"rank":43.27,"hrank":38.59,"imp":17.381,"train_time":2.13,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-132","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1184.0,"score":0.056,"rank":44.6,"hrank":37.94,"imp":17.18,"train_time":354.91,"predict_time":1.636,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-132","imputed":false},{"position":49,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1180.0,"score":0.036,"rank":44.87,"hrank":40.94,"imp":16.592,"train_time":50.36,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+109\/-132","imputed":false},{"position":50,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1174.0,"score":0.039,"rank":45.2,"hrank":39.13,"imp":17.152,"train_time":11.42,"predict_time":10.384,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-110","imputed":false},{"position":51,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1169.0,"score":0.075,"rank":45.47,"hrank":39.79,"imp":18.421,"train_time":86.71,"predict_time":9.515,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+100\/-131","imputed":false},{"position":52,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1087.0,"score":0.005,"rank":50.13,"hrank":48.61,"imp":18.293,"train_time":2.58,"predict_time":0.039,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-90","imputed":false},{"position":53,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1058.0,"score":0.008,"rank":51.67,"hrank":49.21,"imp":19.635,"train_time":354.91,"predict_time":0.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-104","imputed":false},{"position":54,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1057.0,"score":0.016,"rank":51.73,"hrank":47.78,"imp":22.822,"train_time":20.27,"predict_time":0.443,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-156","imputed":false},{"position":55,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1010.0,"score":0.018,"rank":54.07,"hrank":47.37,"imp":26.948,"train_time":2.55,"predict_time":0.157,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+160\/-219","imputed":false},{"position":56,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":54.57,"hrank":53.29,"imp":20.772,"train_time":0.24,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-111","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":994.0,"score":0.003,"rank":54.87,"hrank":52.1,"imp":23.969,"train_time":20.27,"predict_time":0.225,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-158","imputed":false},{"position":58,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":974.0,"score":0.01,"rank":55.8,"hrank":53.4,"imp":24.143,"train_time":150.58,"predict_time":4.952,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-131","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":965.0,"score":0.01,"rank":56.2,"hrank":53.07,"imp":23.089,"train_time":1.49,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-163","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":956.0,"score":0.0,"rank":56.6,"hrank":55.55,"imp":25.585,"train_time":5745.33,"predict_time":2.139,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-98","imputed":false},{"position":61,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":948.0,"score":0.0,"rank":56.93,"hrank":56.01,"imp":25.57,"train_time":5745.33,"predict_time":8.567,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-97","imputed":false},{"position":62,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":937.0,"score":0.0,"rank":57.4,"hrank":55.91,"imp":23.258,"train_time":0.47,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-133","imputed":false},{"position":63,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":901.0,"score":0.006,"rank":58.9,"hrank":54.8,"imp":33.239,"train_time":43.96,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-228","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":884.0,"score":0.0,"rank":59.53,"hrank":57.49,"imp":30.681,"train_time":1.53,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-186","imputed":false},{"position":65,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":858.0,"score":0.0,"rank":60.5,"hrank":57.42,"imp":33.973,"train_time":43.96,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+158\/-234","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":824.0,"score":0.0,"rank":61.67,"hrank":61.18,"imp":30.996,"train_time":113.17,"predict_time":2.156,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+41\/-90","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":797.0,"score":0.0,"rank":62.53,"hrank":62.32,"imp":30.322,"train_time":150.58,"predict_time":0.546,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-120","imputed":false},{"position":68,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":740.0,"score":0.0,"rank":64.13,"hrank":63.51,"imp":36.011,"train_time":0.2,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-206","imputed":false},{"position":69,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":569.0,"score":0.0,"rank":67.6,"hrank":67.48,"imp":43.744,"train_time":0.1,"predict_time":0.068,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-155","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_medium/pareto_front_explorer.html
index e309f771c3def20a7ed86c463b212d32f4fdc9ac..a43bc15408ebfea8b4a0c6417ad79bdd6e056462 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.010890969,"imp":9.2446264705,"elo":1531.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0161908394,"imp":9.4068447854,"elo":1570.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1892457458,"imp":9.2108328349,"elo":1582.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0232162798,"imp":12.4133118638,"elo":1407.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.025881505,"imp":11.4009229884,"elo":1495.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"imp":11.3011884406,"elo":1518.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"imp":17.3805259993,"elo":1205.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0061305091,"imp":16.389359794,"elo":1223.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0928598604,"imp":15.154145312,"elo":1324.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721186209,"imp":23.2582690565,"elo":937.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0167167359,"imp":16.5918385982,"elo":1179.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1833493946,"imp":15.0064476194,"elo":1224.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"imp":23.0885250093,"elo":964.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0941673994,"imp":19.6350538202,"elo":1058.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"imp":17.180063053,"elo":1183.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.06997296,"imp":14.2603049425,"elo":1228.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1174373907,"imp":10.6800957475,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.6207744311,"imp":10.0718129419,"elo":1582.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.3613072104,"imp":15.2036623307,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"imp":13.7754313429,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3182658815,"imp":10.1975534028,"elo":1545.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2338623423,"imp":6.9011250205,"elo":1668.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"imp":30.6806116306,"elo":884.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2251402597,"imp":23.9690418598,"elo":993.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.442693885,"imp":22.8219634277,"elo":1056.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0219328376,"imp":20.7718668,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0250207663,"imp":15.2303427145,"elo":1209.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2298050149,"imp":14.8953173136,"elo":1242.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2940400342,"imp":15.0224816412,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0785368436,"imp":10.2564835391,"elo":1520.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.7432247239,"imp":8.3592702667,"elo":1618.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6684256529,"imp":8.9508162247,"elo":1519.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1340282338,"imp":7.813683812,"elo":1661.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"imp":6.4896903028,"elo":1713.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.3844403101,"imp":17.1519949807,"elo":1173.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.0937641424,"imp":15.7325841932,"elo":1284.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"imp":12.9156193656,"elo":1417.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":387.8903208002,"imp":10.4809941856,"elo":1470.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1777276152,"imp":10.9694457456,"elo":1477.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":13.118037366,"imp":4.457579922,"elo":1941.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1177498102,"imp":7.5082256688,"elo":1710.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.045727832,"imp":13.2846352058,"elo":1342.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0566464663,"imp":10.4806951601,"elo":1490.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5563493456,"imp":9.6276945538,"elo":1554.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"imp":7.0972177851,"elo":1764.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2136890429,"imp":4.5903736993,"elo":1800.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1560825621,"imp":30.9959928616,"elo":824.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.139339258,"imp":25.5852650516,"elo":955.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"imp":25.570357389,"elo":948.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1258468951,"imp":12.1514539075,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0390374572,"imp":18.2928174392,"elo":1087.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0507240116,"imp":14.6486117194,"elo":1238.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1994296217,"imp":12.2903298919,"elo":1349.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"imp":11.8360684126,"elo":1380.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0795849342,"imp":10.0490986518,"elo":1498.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.7060451863,"imp":9.8078613915,"elo":1550.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":9.5146310091,"imp":18.4211458676,"elo":1169.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":6.2309459511,"imp":11.9518940132,"elo":1392.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":69.5857864072,"imp":9.9738007966,"elo":1535.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"imp":26.9481191606,"elo":1010.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0505891669,"imp":13.5401190315,"elo":1386.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.8800015259,"imp":11.3769395133,"elo":1416.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.010890969,"x_train":1.8802002354,"imp":9.2446264705,"elo":1531.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0161908394,"x_train":477.5939557065,"imp":9.4068447854,"elo":1570.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1892457458,"x_train":477.5939557065,"imp":9.2108328349,"elo":1582.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0232162798,"x_train":0.4836414313,"imp":12.4133118638,"elo":1407.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.025881505,"x_train":92.4632546012,"imp":11.4009229884,"elo":1495.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":92.4632546012,"imp":11.3011884406,"elo":1518.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"x_train":2.1343897895,"imp":17.3805259993,"elo":1205.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0061305091,"x_train":662.2364442662,"imp":16.389359794,"elo":1223.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0928598604,"x_train":662.2364442662,"imp":15.154145312,"elo":1324.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721186209,"x_train":0.4655147195,"imp":23.2582690565,"elo":937.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0167167359,"x_train":50.361853205,"imp":16.5918385982,"elo":1179.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1833493946,"x_train":50.361853205,"imp":15.0064476194,"elo":1224.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"x_train":1.4936466441,"imp":23.0885250093,"elo":964.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0941673994,"x_train":354.9067354826,"imp":19.6350538202,"elo":1058.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"x_train":354.9067354826,"imp":17.180063053,"elo":1183.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.06997296,"x_train":0.3380289469,"imp":14.2603049425,"elo":1228.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1174373907,"x_train":129.246647918,"imp":10.6800957475,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.6207744311,"x_train":129.246647918,"imp":10.0718129419,"elo":1582.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.3613072104,"x_train":6.5096502566,"imp":15.2036623307,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"x_train":9.0683107381,"imp":13.7754313429,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3182658815,"x_train":2522.590218997,"imp":10.1975534028,"elo":1545.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2338623423,"x_train":2522.590218997,"imp":6.9011250205,"elo":1668.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"x_train":1.5311965098,"imp":30.6806116306,"elo":884.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2251402597,"x_train":20.2694783344,"imp":23.9690418598,"elo":993.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.442693885,"x_train":20.2694783344,"imp":22.8219634277,"elo":1056.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0219328376,"x_train":0.239812243,"imp":20.7718668,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0250207663,"x_train":65.72538338,"imp":15.2303427145,"elo":1209.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2298050149,"x_train":65.72538338,"imp":14.8953173136,"elo":1242.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2940400342,"x_train":6.5430745829,"imp":15.0224816412,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0785368436,"x_train":1714.2893053174,"imp":10.2564835391,"elo":1520.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.7432247239,"x_train":1714.2893053174,"imp":8.3592702667,"elo":1618.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6684256529,"x_train":1.9087879061,"imp":8.9508162247,"elo":1519.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1340282338,"x_train":739.1047147274,"imp":7.813683812,"elo":1661.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"x_train":739.1047147274,"imp":6.4896903028,"elo":1713.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.3844403101,"x_train":11.4229329799,"imp":17.1519949807,"elo":1173.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.0937641424,"x_train":41.5053698771,"imp":15.7325841932,"elo":1284.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"x_train":4532.1904663458,"imp":12.9156193656,"elo":1417.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":387.8903208002,"x_train":4532.1904663458,"imp":10.4809941856,"elo":1470.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1777276152,"x_train":0.755625227,"imp":10.9694457456,"elo":1477.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":13.118037366,"x_train":27.1820477768,"imp":4.457579922,"elo":1941.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1177498102,"x_train":0.5630642002,"imp":7.5082256688,"elo":1710.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.045727832,"x_train":1.9359607519,"imp":13.2846352058,"elo":1342.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0566464663,"x_train":1176.2542230269,"imp":10.4806951601,"elo":1490.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5563493456,"x_train":1176.2542230269,"imp":9.6276945538,"elo":1554.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"x_train":2.6386678711,"imp":7.0972177851,"elo":1764.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2136890429,"x_train":1.0196284569,"imp":4.5903736993,"elo":1800.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1560825621,"x_train":113.1657286335,"imp":30.9959928616,"elo":824.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.139339258,"x_train":5745.3315346214,"imp":25.5852650516,"elo":955.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"x_train":5745.3315346214,"imp":25.570357389,"elo":948.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1258468951,"x_train":0.5121427198,"imp":12.1514539075,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0390374572,"x_train":2.5795466495,"imp":18.2928174392,"elo":1087.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0507240116,"x_train":1803.2313354459,"imp":14.6486117194,"elo":1238.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1994296217,"x_train":1803.2313354459,"imp":12.2903298919,"elo":1349.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"x_train":0.4833722601,"imp":11.8360684126,"elo":1380.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0795849342,"x_train":289.3497533723,"imp":10.0490986518,"elo":1498.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.7060451863,"x_train":289.3497533723,"imp":9.8078613915,"elo":1550.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":9.5146310091,"x_train":86.7109854477,"imp":18.4211458676,"elo":1169.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":6.2309459511,"x_train":2045.5905262122,"imp":11.9518940132,"elo":1392.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":69.5857864072,"x_train":2045.5905262122,"imp":9.9738007966,"elo":1535.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"x_train":2.5457229657,"imp":26.9481191606,"elo":1010.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0505891669,"x_train":559.6449092182,"imp":13.5401190315,"elo":1386.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.8800015259,"x_train":559.6449092182,"imp":11.3769395133,"elo":1416.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_medium/tuning_trajectories_explorer.html
index c1e4d40b1e5b1a1fb33edc5763596b5878987b66..a0091da01defe6967e2083c8a633c338a7989ba8 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":1.8802002354,"imp":9.3101786703,"elo":1552.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":4.3390813562,"imp":9.3668455506,"elo":1559.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.9107908196,"imp":9.267595022,"elo":1587.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":24.8738378919,"imp":9.2402319461,"elo":1595.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":59.791099159,"imp":9.2572564585,"elo":1609.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":118.7049807087,"imp":9.2515510859,"elo":1612.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":236.2186933654,"imp":9.2783419621,"elo":1610.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":351.7154510353,"imp":9.3010200485,"elo":1608.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":477.5939557065,"imp":9.2664909215,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4836414313,"imp":12.4888788997,"elo":1427.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.0174591425,"imp":12.1684586985,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.3395204881,"imp":11.9575651564,"elo":1487.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.5724929214,"imp":11.6909990873,"elo":1502.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.475979466,"imp":11.5322907251,"elo":1518.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.9675728072,"imp":11.3632338481,"elo":1531.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":46.0351331804,"imp":11.2982349858,"elo":1542.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":69.0484132695,"imp":11.3142188082,"elo":1548.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":92.4632546012,"imp":11.3673653853,"elo":1553.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.1343897895,"imp":17.4415958555,"elo":1219.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":6.5030474549,"imp":16.5576069156,"elo":1262.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":15.551340727,"imp":16.0469153438,"elo":1286.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":32.7988382693,"imp":15.8315388028,"elo":1300.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":82.4434921155,"imp":15.5556478456,"elo":1322.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":165.9098913248,"imp":15.449401335,"elo":1333.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":325.9150394035,"imp":15.3061276456,"elo":1342.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":495.2952240262,"imp":15.261890739,"elo":1342.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":662.2364442662,"imp":15.2135933208,"elo":1346.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.4655147195,"imp":23.3078515058,"elo":932.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.8039431531,"imp":21.0918921598,"elo":992.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.5507866238,"imp":17.5283944157,"elo":1095.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9159741389,"imp":16.0430645763,"elo":1150.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.8290142883,"imp":15.3266460542,"elo":1183.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.9805618038,"imp":14.9844657736,"elo":1208.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":25.3410754681,"imp":14.6296352436,"elo":1235.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.7076806608,"imp":14.6000430678,"elo":1238.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":50.361853205,"imp":15.0595931989,"elo":1234.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.4936466441,"imp":23.1338588854,"elo":973.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.1783708385,"imp":21.9290550691,"elo":1036.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.5233821255,"imp":20.266979712,"elo":1107.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":17.7581257529,"imp":19.5010284137,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":42.7590536162,"imp":18.5238258057,"elo":1160.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":86.3142265197,"imp":18.0598399025,"elo":1176.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":175.409142407,"imp":17.551164854,"elo":1187.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":264.0183986663,"imp":17.3246074232,"elo":1199.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":354.9067354826,"imp":17.2365017671,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.3380289469,"imp":14.3277897919,"elo":1245.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.9616222835,"imp":13.6117994537,"elo":1310.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":2.6147261705,"imp":12.4484880806,"elo":1422.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":6.009789211,"imp":11.6154619585,"elo":1493.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":16.2074876554,"imp":10.6946997732,"elo":1557.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":31.9858466992,"imp":10.3197118265,"elo":1590.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":64.004281301,"imp":10.1824969771,"elo":1604.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":96.5406969888,"imp":10.1481925679,"elo":1610.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":129.246647918,"imp":10.144476849,"elo":1614.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":6.5096502566,"imp":15.2673017486,"elo":1317.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":9.0683107381,"imp":13.8345397258,"elo":1291.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":22.7816772308,"imp":12.3574421813,"elo":1382.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.9509960123,"imp":10.5279683206,"elo":1503.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":121.2878502068,"imp":10.0115247351,"elo":1542.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":291.6044227147,"imp":9.4501715832,"elo":1596.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":628.8473422899,"imp":8.7744306685,"elo":1629.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1243.1111273612,"imp":8.1696685386,"elo":1662.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1857.216089218,"imp":7.4629553504,"elo":1688.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2522.590218997,"imp":6.9899541425,"elo":1703.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.5311965098,"imp":30.6847464252,"elo":906.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.1703965898,"imp":27.5932505743,"elo":982.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":17.0551085935,"imp":23.1659318506,"elo":1055.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.2694783344,"imp":22.8406144223,"elo":1067.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.239812243,"imp":20.8316944461,"elo":1000.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4901066084,"imp":17.3008310184,"elo":1109.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.4763660732,"imp":15.9727945524,"elo":1170.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":3.0456620821,"imp":15.4757041149,"elo":1199.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":7.9306942372,"imp":15.1219825125,"elo":1242.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":16.1953879054,"imp":15.0062206669,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":32.6380042897,"imp":14.9252458929,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":48.9781692218,"imp":14.9213810044,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":65.72538338,"imp":14.9488517824,"elo":1259.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.5430745829,"imp":15.0709468125,"elo":1283.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.4482837345,"imp":12.7764398981,"elo":1409.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.4902474287,"imp":11.0401568285,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":81.1768457292,"imp":10.0828756939,"elo":1551.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":211.6867497619,"imp":9.3638496992,"elo":1592.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":425.9393992898,"imp":8.9914600467,"elo":1614.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":844.9984797236,"imp":8.6564499876,"elo":1626.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1273.8908426812,"imp":8.5179970894,"elo":1642.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1714.2893053174,"imp":8.4329287591,"elo":1651.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.9087879061,"imp":9.0318979829,"elo":1528.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.1930136195,"imp":8.9282202861,"elo":1547.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.3030487246,"imp":8.6970393181,"elo":1566.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":34.3657315438,"imp":8.4975753927,"elo":1592.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":89.1338776753,"imp":8.0943433787,"elo":1653.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":182.7153391231,"imp":7.6911791455,"elo":1676.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":366.5547593939,"imp":7.2073441384,"elo":1708.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":550.802457324,"imp":6.9240564882,"elo":1722.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":739.1047147274,"imp":6.5642947033,"elo":1737.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":11.4229329799,"imp":17.2052766922,"elo":1186.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.5053698771,"imp":15.77888031,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.019350737,"imp":14.4694915963,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":119.7137023169,"imp":13.3885047556,"elo":1383.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":214.9883934306,"imp":12.9181225664,"elo":1404.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":547.8027265576,"imp":11.9376853893,"elo":1443.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1117.1957402472,"imp":11.324402862,"elo":1465.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2246.8238340239,"imp":10.9679290078,"elo":1484.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3382.9355985564,"imp":10.6711842762,"elo":1483.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4532.1904663458,"imp":10.5200659712,"elo":1488.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.755625227,"imp":11.0261391187,"elo":1497.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":27.1820477768,"imp":4.5302588187,"elo":1964.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.5630642002,"imp":7.5867160473,"elo":1730.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.9359607519,"imp":13.348608439,"elo":1361.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.4292557844,"imp":12.665274073,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":25.9857811464,"imp":11.7175055653,"elo":1466.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":55.2138352308,"imp":10.9937623926,"elo":1510.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":144.8913305375,"imp":10.4219450187,"elo":1536.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":291.446310356,"imp":10.1316770987,"elo":1550.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":569.7562297642,"imp":9.8613946529,"elo":1563.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":866.467617543,"imp":9.7713848819,"elo":1576.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1176.2542230269,"imp":9.7068285077,"elo":1588.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.6386678711,"imp":7.170909029,"elo":1797.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.0196284569,"imp":4.53762243,"elo":1797.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":113.1657286335,"imp":31.0261692308,"elo":820.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":141.3356612104,"imp":30.904567666,"elo":835.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":220.2044612008,"imp":30.5734501802,"elo":844.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":327.1108834516,"imp":29.8700168851,"elo":863.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":732.5344880833,"imp":28.5319891784,"elo":886.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1475.7671634624,"imp":27.2140949709,"elo":917.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2923.1088462252,"imp":26.660294051,"elo":933.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4314.2921057007,"imp":26.0667075541,"elo":957.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5745.3315346214,"imp":25.6051802421,"elo":980.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.5121427198,"imp":12.2299705968,"elo":1401.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":2.5795466495,"imp":18.3372127798,"elo":1101.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.5303184982,"imp":16.8976741311,"elo":1160.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":42.5186103306,"imp":15.3546222457,"elo":1221.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":94.7530519512,"imp":14.4400149526,"elo":1268.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":222.7613331852,"imp":13.4934825661,"elo":1304.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":460.6794919634,"imp":13.0332265716,"elo":1330.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":882.9304835882,"imp":12.6771146149,"elo":1354.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1357.9255836308,"imp":12.4785421595,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1803.2313354459,"imp":12.3521741362,"elo":1377.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4833722601,"imp":11.9078835843,"elo":1402.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.4574329869,"imp":11.1335409236,"elo":1469.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":7.3610615249,"imp":10.6634417872,"elo":1516.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":13.6386574902,"imp":10.3447754875,"elo":1539.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":35.9388283515,"imp":10.069810101,"elo":1556.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":71.5452674162,"imp":9.9458136244,"elo":1569.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":142.0559469286,"imp":9.8960743136,"elo":1578.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":215.0273705137,"imp":9.8849729562,"elo":1580.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":289.3497533723,"imp":9.9010563696,"elo":1580.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":86.7109854477,"imp":18.473075499,"elo":1191.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":165.5352279836,"imp":15.9866627301,"elo":1257.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":405.7546652544,"imp":17.7353404452,"elo":1344.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":797.4916792797,"imp":15.6812231729,"elo":1437.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1969.6298609549,"imp":9.9324356592,"elo":1562.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":2045.5905262122,"imp":10.0290284619,"elo":1563.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.5457229657,"imp":26.9867576335,"elo":1026.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.2856913484,"imp":21.7555602931,"elo":1111.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":13.6127137293,"imp":15.9862923705,"elo":1249.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":27.6956293748,"imp":13.5693447461,"elo":1355.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":69.488323154,"imp":12.6572114896,"elo":1400.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":138.8404994793,"imp":12.1949160482,"elo":1416.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":278.0036441284,"imp":11.8122767858,"elo":1426.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":417.6192129824,"imp":11.5271614578,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":559.6449092182,"imp":11.2085821154,"elo":1442.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.010890969,"x_train":1.8802002354,"imp":9.3101786703,"elo":1552.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0416544912,"x_train":4.3390813562,"imp":9.3668455506,"elo":1559.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0663288817,"x_train":12.9107908196,"imp":9.267595022,"elo":1587.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0905500613,"x_train":24.8738378919,"imp":9.2402319461,"elo":1595.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1417099551,"x_train":59.791099159,"imp":9.2572564585,"elo":1609.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1656029978,"x_train":118.7049807087,"imp":9.2515510859,"elo":1612.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1778183572,"x_train":236.2186933654,"imp":9.2783419621,"elo":1610.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1723703847,"x_train":351.7154510353,"imp":9.3010200485,"elo":1608.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1892457458,"x_train":477.5939557065,"imp":9.2664909215,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0232162798,"x_train":0.4836414313,"imp":12.4888788997,"elo":1427.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0460251725,"x_train":1.0174591425,"imp":12.1684586985,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1045606687,"x_train":2.3395204881,"imp":11.9575651564,"elo":1487.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1437162313,"x_train":4.5724929214,"imp":11.6909990873,"elo":1502.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2215730015,"x_train":11.475979466,"imp":11.5322907251,"elo":1518.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2317567073,"x_train":22.9675728072,"imp":11.3632338481,"elo":1531.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2465009094,"x_train":46.0351331804,"imp":11.2982349858,"elo":1542.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2772856864,"x_train":69.0484132695,"imp":11.3142188082,"elo":1548.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":92.4632546012,"imp":11.3673653853,"elo":1553.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"x_train":2.1343897895,"imp":17.4415958555,"elo":1219.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0116188579,"x_train":6.5030474549,"imp":16.5576069156,"elo":1262.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0227492458,"x_train":15.551340727,"imp":16.0469153438,"elo":1286.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333486726,"x_train":32.7988382693,"imp":15.8315388028,"elo":1300.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0515992981,"x_train":82.4434921155,"imp":15.5556478456,"elo":1322.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0585840954,"x_train":165.9098913248,"imp":15.449401335,"elo":1333.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0740425479,"x_train":325.9150394035,"imp":15.3061276456,"elo":1342.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0820593882,"x_train":495.2952240262,"imp":15.261890739,"elo":1342.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0923119309,"x_train":662.2364442662,"imp":15.2135933208,"elo":1346.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721186209,"x_train":0.4655147195,"imp":23.3078515058,"elo":932.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.079982782,"x_train":0.8039431531,"imp":21.0918921598,"elo":992.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.084787187,"x_train":1.5507866238,"imp":17.5283944157,"elo":1095.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.078755296,"x_train":2.9159741389,"imp":16.0430645763,"elo":1150.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1006473264,"x_train":6.8290142883,"imp":15.3266460542,"elo":1183.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1257749856,"x_train":12.9805618038,"imp":14.9844657736,"elo":1208.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1512704682,"x_train":25.3410754681,"imp":14.6296352436,"elo":1235.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1728496468,"x_train":37.7076806608,"imp":14.6000430678,"elo":1238.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739987392,"x_train":50.361853205,"imp":15.0595931989,"elo":1234.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"x_train":1.4936466441,"imp":23.1338588854,"elo":973.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2247585805,"x_train":3.1783708385,"imp":21.9290550691,"elo":1036.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4166305562,"x_train":8.5233821255,"imp":20.266979712,"elo":1107.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6283891459,"x_train":17.7581257529,"imp":19.5010284137,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9164506571,"x_train":42.7590536162,"imp":18.5238258057,"elo":1160.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2468060948,"x_train":86.3142265197,"imp":18.0598399025,"elo":1176.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6512385514,"x_train":175.409142407,"imp":17.551164854,"elo":1187.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7760120815,"x_train":264.0183986663,"imp":17.3246074232,"elo":1199.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"x_train":354.9067354826,"imp":17.2365017671,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.06997296,"x_train":0.3380289469,"imp":14.3277897919,"elo":1245.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4182187449,"x_train":0.9616222835,"imp":13.6117994537,"elo":1310.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7644003426,"x_train":2.6147261705,"imp":12.4484880806,"elo":1422.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0019478197,"x_train":6.009789211,"imp":11.6154619585,"elo":1493.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1773854817,"x_train":16.2074876554,"imp":10.6946997732,"elo":1557.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4649015268,"x_train":31.9858466992,"imp":10.3197118265,"elo":1590.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1045193014,"x_train":64.004281301,"imp":10.1824969771,"elo":1604.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1744811378,"x_train":96.5406969888,"imp":10.1481925679,"elo":1610.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.6207744311,"x_train":129.246647918,"imp":10.144476849,"elo":1614.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":5.3613072104,"x_train":6.5096502566,"imp":15.2673017486,"elo":1317.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"x_train":9.0683107381,"imp":13.8345397258,"elo":1291.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5307544296,"x_train":22.7816772308,"imp":12.3574421813,"elo":1382.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.3312012868,"x_train":60.9509960123,"imp":10.5279683206,"elo":1503.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1424478891,"x_train":121.2878502068,"imp":10.0115247351,"elo":1542.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7588304305,"x_train":291.6044227147,"imp":9.4501715832,"elo":1596.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.9692086024,"x_train":628.8473422899,"imp":8.7744306685,"elo":1629.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.3479391024,"x_train":1243.1111273612,"imp":8.1696685386,"elo":1662.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2360012984,"x_train":1857.216089218,"imp":7.4629553504,"elo":1688.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2338623423,"x_train":2522.590218997,"imp":6.9899541425,"elo":1703.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"x_train":1.5311965098,"imp":30.6847464252,"elo":906.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1165097969,"x_train":6.1703965898,"imp":27.5932505743,"elo":982.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3758127641,"x_train":17.0551085935,"imp":23.1659318506,"elo":1055.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.442693885,"x_train":20.2694783344,"imp":22.8406144223,"elo":1067.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0219328376,"x_train":0.239812243,"imp":20.8316944461,"elo":1000.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0440982753,"x_train":0.4901066084,"imp":17.3008310184,"elo":1109.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1060575328,"x_train":1.4763660732,"imp":15.9727945524,"elo":1170.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1361499657,"x_train":3.0456620821,"imp":15.4757041149,"elo":1199.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2062124655,"x_train":7.9306942372,"imp":15.1219825125,"elo":1242.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2414176328,"x_train":16.1953879054,"imp":15.0062206669,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2481870329,"x_train":32.6380042897,"imp":14.9252458929,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2529555406,"x_train":48.9781692218,"imp":14.9213810044,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2298050149,"x_train":65.72538338,"imp":14.9488517824,"elo":1259.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2940400342,"x_train":6.5430745829,"imp":15.0709468125,"elo":1283.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4024728498,"x_train":15.4482837345,"imp":12.7764398981,"elo":1409.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.6227888714,"x_train":39.4902474287,"imp":11.0401568285,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.8897271358,"x_train":81.1768457292,"imp":10.0828756939,"elo":1551.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.2749551545,"x_train":211.6867497619,"imp":9.3638496992,"elo":1592.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4994174796,"x_train":425.9393992898,"imp":8.9914600467,"elo":1614.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7662329972,"x_train":844.9984797236,"imp":8.6564499876,"elo":1626.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5220276492,"x_train":1273.8908426812,"imp":8.5179970894,"elo":1642.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7432247239,"x_train":1714.2893053174,"imp":8.4329287591,"elo":1651.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6684256529,"x_train":1.9087879061,"imp":9.0318979829,"elo":1528.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4990238167,"x_train":5.1930136195,"imp":8.9282202861,"elo":1547.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.3467414164,"x_train":16.3030487246,"imp":8.6970393181,"elo":1566.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.8781041597,"x_train":34.3657315438,"imp":8.4975753927,"elo":1592.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8044658046,"x_train":89.1338776753,"imp":8.0943433787,"elo":1653.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.8326114376,"x_train":182.7153391231,"imp":7.6911791455,"elo":1676.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.1219292846,"x_train":366.5547593939,"imp":7.2073441384,"elo":1708.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.4798418546,"x_train":550.802457324,"imp":6.9240564882,"elo":1722.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"x_train":739.1047147274,"imp":6.5642947033,"elo":1737.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.3844403101,"x_train":11.4229329799,"imp":17.2052766922,"elo":1186.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.0937641424,"x_train":41.5053698771,"imp":15.77888031,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":55.1055828632,"x_train":63.019350737,"imp":14.4694915963,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":84.5945717618,"x_train":119.7137023169,"imp":13.3885047556,"elo":1383.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":115.9268466871,"x_train":214.9883934306,"imp":12.9181225664,"elo":1404.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":182.0790373328,"x_train":547.8027265576,"imp":11.9376853893,"elo":1443.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":248.1191218836,"x_train":1117.1957402472,"imp":11.324402862,"elo":1465.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":319.6845628759,"x_train":2246.8238340239,"imp":10.9679290078,"elo":1484.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":359.660599797,"x_train":3382.9355985564,"imp":10.6711842762,"elo":1483.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":390.2528105033,"x_train":4532.1904663458,"imp":10.5200659712,"elo":1488.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1777276152,"x_train":0.755625227,"imp":11.0261391187,"elo":1497.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":13.118037366,"x_train":27.1820477768,"imp":4.5302588187,"elo":1964.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1177498102,"x_train":0.5630642002,"imp":7.5867160473,"elo":1730.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.045727832,"x_train":1.9359607519,"imp":13.348608439,"elo":1361.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1008784954,"x_train":6.4292557844,"imp":12.665274073,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.1981532601,"x_train":25.9857811464,"imp":11.7175055653,"elo":1466.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2598475914,"x_train":55.2138352308,"imp":10.9937623926,"elo":1510.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3596014701,"x_train":144.8913305375,"imp":10.4219450187,"elo":1536.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4349645047,"x_train":291.446310356,"imp":10.1316770987,"elo":1550.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4765747411,"x_train":569.7562297642,"imp":9.8613946529,"elo":1563.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.4821521088,"x_train":866.467617543,"imp":9.7713848819,"elo":1576.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5419846432,"x_train":1176.2542230269,"imp":9.7068285077,"elo":1588.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"x_train":2.6386678711,"imp":7.170909029,"elo":1797.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2136890429,"x_train":1.0196284569,"imp":4.53762243,"elo":1797.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1560825621,"x_train":113.1657286335,"imp":31.0261692308,"elo":820.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.2636388537,"x_train":141.3356612104,"imp":30.904567666,"elo":835.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.3912942358,"x_train":220.2044612008,"imp":30.5734501802,"elo":844.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.8322499442,"x_train":327.1108834516,"imp":29.8700168851,"elo":863.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.9995876557,"x_train":732.5344880833,"imp":28.5319891784,"elo":886.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5174296389,"x_train":1475.7671634624,"imp":27.2140949709,"elo":917.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.8892404898,"x_train":2923.1088462252,"imp":26.660294051,"elo":933.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.0618917211,"x_train":4314.2921057007,"imp":26.0667075541,"elo":957.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"x_train":5745.3315346214,"imp":25.6051802421,"elo":980.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1258468951,"x_train":0.5121427198,"imp":12.2299705968,"elo":1401.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0390374572,"x_train":2.5795466495,"imp":18.3372127798,"elo":1101.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0911383405,"x_train":9.5303184982,"imp":16.8976741311,"elo":1160.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2406739756,"x_train":42.5186103306,"imp":15.3546222457,"elo":1221.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3974574041,"x_train":94.7530519512,"imp":14.4400149526,"elo":1268.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5700088429,"x_train":222.7613331852,"imp":13.4934825661,"elo":1304.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7281380887,"x_train":460.6794919634,"imp":13.0332265716,"elo":1330.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9742904808,"x_train":882.9304835882,"imp":12.6771146149,"elo":1354.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9743613689,"x_train":1357.9255836308,"imp":12.4785421595,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2024249833,"x_train":1803.2313354459,"imp":12.3521741362,"elo":1377.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"x_train":0.4833722601,"imp":11.9078835843,"elo":1402.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.176191836,"x_train":2.4574329869,"imp":11.1335409236,"elo":1469.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2515770525,"x_train":7.3610615249,"imp":10.6634417872,"elo":1516.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3262966161,"x_train":13.6386574902,"imp":10.3447754875,"elo":1539.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3865979705,"x_train":35.9388283515,"imp":10.069810101,"elo":1556.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4060971572,"x_train":71.5452674162,"imp":9.9458136244,"elo":1569.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4323510597,"x_train":142.0559469286,"imp":9.8960743136,"elo":1578.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4840201507,"x_train":215.0273705137,"imp":9.8849729562,"elo":1580.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7060451863,"x_train":289.3497533723,"imp":9.9010563696,"elo":1580.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":9.5146310091,"x_train":86.7109854477,"imp":18.473075499,"elo":1191.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":18.8009106596,"x_train":165.5352279836,"imp":15.9866627301,"elo":1257.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":28.6343734044,"x_train":405.7546652544,"imp":17.7353404452,"elo":1344.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.779029417,"x_train":797.4916792797,"imp":15.6812231729,"elo":1437.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":67.7213907223,"x_train":1969.6298609549,"imp":9.9324356592,"elo":1562.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":69.5857864072,"x_train":2045.5905262122,"imp":10.0290284619,"elo":1563.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"x_train":2.5457229657,"imp":26.9867576335,"elo":1026.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.2277922302,"x_train":5.2856913484,"imp":21.7555602931,"elo":1111.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.2497110384,"x_train":13.6127137293,"imp":15.9862923705,"elo":1249.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.3680809292,"x_train":27.6956293748,"imp":13.5693447461,"elo":1355.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.5701916292,"x_train":69.488323154,"imp":12.6572114896,"elo":1400.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.6682251127,"x_train":138.8404994793,"imp":12.1949160482,"elo":1416.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.7937638305,"x_train":278.0036441284,"imp":11.8122767858,"elo":1426.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":0.8299964002,"x_train":417.6192129824,"imp":11.5271614578,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":0.8800015259,"x_train":559.6449092182,"imp":11.2085821154,"elo":1442.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..9c6543b495e0965d01f6f18222f1941385f7b438
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_all/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "AutoGluon 1.4 (best, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabICLv2 (default)", "ModernNCA (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "CatBoost (tuned + ensembled)", "LightGBM (tuned + ensembled)", "CatBoost (tuned)", "TabM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "ModernNCA (tuned)", "iLTM (tuned + ensembled)", "LightGBM (tuned)", "CatBoost (default)", "RealMLP (tuned)", "RealTabPFN-2.5 (default)", "ChimeraBoost (tuned + ensembled)", "XGBoost (tuned)", "ChimeraBoost (tuned)", "TabM (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "TabDPT (tuned)", "xRFM (tuned + ensembled)", "ChimeraBoost (default)", "iLTM (tuned)", "xRFM (tuned)", "XGBoost (default)", "TabSwift (default)", "TorchMLP (tuned + ensembled)", "TabM (default)", "EBM (tuned + ensembled)", "LimiX (default)", "TabDPT (default)", "ModernNCA (default)", "RealMLP (default)", "RandomForest (tuned + ensembled)", "TorchMLP (tuned)", "LightGBM (default)", "ExtraTrees (tuned + ensembled)", "EBM (tuned)", "RandomForest (tuned)", "EBM (default)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (tuned)", "SAP-RPT-OSS (default)", "iLTM (default)", "TorchMLP (default)", "FastaiMLP (tuned)", "PerpetualBooster (tuned + ensembled)", "xRFM (default)", "RandomForest (default)", "PerpetualBooster (tuned)", "KNN (tuned + ensembled)", "FastaiMLP (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned)", "TabSTAR (default)", "KNN (tuned)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 1.0, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 1.0, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4000000059604645, NaN, 0.6666666865348816, 0.4000000059604645, 0.6666666865348816, 0.800000011920929, 0.5333333611488342, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4000000059604645, 0.3333333432674408, NaN, 0.4000000059604645, 0.7333333492279053, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 1.0, 0.9333333373069763, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.6666666865348816, 0.9333333373069763, 1.0, 0.9333333373069763, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2666666805744171, 0.6000000238418579, 0.6000000238418579, NaN, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 1.0, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, NaN, 0.46666666865348816, 0.5333333611488342, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.20000000298023224, 0.46666666865348816, 0.3333333432674408, 0.5333333611488342, NaN, 0.4000000059604645, 0.5333333611488342, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.46666666865348816, 0.46666666865348816, 0.2666666805744171, 0.46666666865348816, 0.6000000238418579, NaN, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 1.0, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.20000000298023224, 0.4000000059604645, 0.13333334028720856, 0.2666666805744171, 0.46666666865348816, 0.2666666805744171, NaN, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 1.0, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.46666666865348816, 0.20000000298023224, 0.2666666805744171, 0.46666666865348816, NaN, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, NaN, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6666666865348816, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.0, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, NaN, 0.4000000059604645, 0.6666666865348816, 0.4000000059604645, 0.5333333611488342, 0.4333333373069763, 0.6000000238418579, 0.46666666865348816, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.6000000238418579, NaN, 0.6666666865348816, 0.3333333432674408, 0.6000000238418579, 0.46666666865348816, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.6000000238418579, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 1.0, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.0, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, NaN, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, NaN, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 1.0, 0.8666666746139526, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 1.0, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.0, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, NaN, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.7333333492279053, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.5333333611488342, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.46666666865348816, 0.5666666626930237, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, NaN, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, NaN, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 0.8666666746139526, 0.6666666865348816, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.4000000059604645, 0.5333333611488342, 0.2666666805744171, 0.5333333611488342, 0.3333333432674408, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, NaN, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.46666666865348816, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 1.0, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, NaN, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.4000000059604645, NaN, 0.5333333611488342, 0.3333333432674408, 0.6000000238418579, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, NaN, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 1.0, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0], [0.13333334028720856, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.5333333611488342, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, NaN, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.20000000298023224, 0.0, 0.0, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, NaN, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.8666666746139526, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.2666666805744171, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, NaN, 0.46666666865348816, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, NaN, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 1.0, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.5333333611488342, 0.3333333432674408, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.3333333432674408, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, NaN, 0.3333333432674408, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5333333611488342, 0.6666666865348816, NaN, 0.800000011920929, 0.6666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.800000011920929, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.20000000298023224, NaN, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.6666666865348816, 0.8999999761581421, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.06666667014360428, 0.06666667014360428, 0.0, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.06666667014360428, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.5333333611488342, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, NaN, 0.3333333432674408, 0.5333333611488342, 0.6666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.46666666865348816, 0.3333333432674408, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, NaN, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.46666666865348816, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, NaN, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.8666666746139526, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.0, 0.0, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.46666666865348816, 0.3333333432674408, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.46666666865348816, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5333333611488342, NaN, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.6666666865348816, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.13333334028720856, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.6000000238418579, 0.5333333611488342, NaN, 0.46666666865348816, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.5333333611488342, 0.800000011920929, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0], [0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.13333334028720856, 0.4000000059604645, 0.5333333611488342, 0.3333333432674408, 0.4000000059604645, 0.13333334028720856, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, NaN, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0], [0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.46666666865348816, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, NaN, 0.4000000059604645, 0.6000000238418579, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.5333333611488342, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 1.0, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.7333333492279053, 1.0, 0.9333333373069763, 1.0, 1.0, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.0, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.0, 0.0, 0.0, 0.13333334028720856, 0.0, 0.06666667014360428, 0.2666666805744171, 0.13333334028720856, 0.06666667014360428, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, NaN, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.9333333373069763, 0.6666666865348816, 1.0, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.8666666746139526, 1.0, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763], [0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, NaN, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.800000011920929, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9333333373069763], [0.0, 0.13333334028720856, 0.13333334028720856, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, NaN, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.5333333611488342, 0.4000000059604645, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, NaN, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.4000000059604645, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.0, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.46666666865348816, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, NaN, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 1.0, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, NaN, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.0, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, NaN, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.9333333373069763, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.0, 0.13333334028720856, 0.13333334028720856, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.46666666865348816, 0.2666666805744171, 0.46666666865348816, 0.20000000298023224, 0.46666666865348816, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, NaN, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, NaN, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.9333333373069763, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 1.0, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.0, 0.2666666805744171, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.3333333432674408, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.06666667014360428, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, NaN, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.46666666865348816, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, NaN, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.8333333134651184, 0.6666666865348816, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.0, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763], [0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.13333334028720856, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, NaN, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.7333333492279053, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 1.0, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 1.0, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, NaN, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 1.0, 1.0, 1.0], [0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, NaN, 0.46666666865348816, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.2666666805744171, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, NaN, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.6666666865348816, 0.8666666746139526, 0.6000000238418579, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.0, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.2666666805744171, 0.06666667014360428, 0.20000000298023224, 0.46666666865348816, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.46666666865348816, 0.5333333611488342, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.0, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.8666666746139526, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.06666667014360428, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, NaN, 0.6666666865348816, 0.6666666865348816, 1.0, 0.6000000238418579, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.7333333492279053, 0.6666666865348816, 1.0, 0.6666666865348816, 0.46666666865348816, 0.7333333492279053, 0.6666666865348816, 1.0], [0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, NaN, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.06666667014360428, 0.3333333432674408, 0.1666666716337204, 0.3333333432674408, 0.4000000059604645, 0.13333334028720856, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.3333333432674408, 0.6000000238418579, NaN, 0.4000000059604645, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.13333334028720856, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.06666667014360428, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.46666666865348816, 0.3333333432674408, 0.4000000059604645, 0.0, 0.5333333611488342, 0.6000000238418579, NaN, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 0.9333333373069763, 0.6666666865348816, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 1.0], [0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.3333333432674408, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, NaN, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 1.0, 0.8666666746139526, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.2666666805744171, 0.0, 0.3333333432674408, 0.13333334028720856, 0.20000000298023224, 0.4000000059604645, 0.13333334028720856, 0.46666666865348816, 0.5333333611488342, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, NaN, 0.6000000238418579, 0.6000000238418579, 0.46666666865348816, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.8666666746139526], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.0, 0.0, 0.3333333432674408, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, NaN, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 1.0, 0.800000011920929, 0.8666666746139526, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.0, 0.06666667014360428, 0.3333333432674408, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.3333333432674408, NaN, 0.5333333611488342, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 1.0, 0.800000011920929, 0.8666666746139526, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.0, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.0, 0.3333333432674408, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.6000000238418579, 0.2666666805744171, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, NaN, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.46666666865348816, 0.6666666865348816, 0.6666666865348816, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.20000000298023224, 0.0, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.5333333611488342, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, NaN, 0.46666666865348816, 0.8999999761581421, 0.3333333432674408, 0.4000000059604645, 0.8666666746139526, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.0, 0.46666666865348816, 0.2666666805744171, 0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.2666666805744171, 0.5333333611488342, NaN, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.8666666746139526], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.5333333611488342, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.10000000149011612, 0.46666666865348816, NaN, 0.3333333432674408, 0.4000000059604645, 0.800000011920929, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.13333334028720856, 0.0, 0.0, 0.0, 0.13333334028720856, 0.13333334028720856, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.5333333611488342, 0.46666666865348816, 0.3333333432674408, 0.5333333611488342, 0.4000000059604645, 0.2666666805744171, 0.0, 0.0, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, NaN, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.20000000298023224, 0.0, 0.0, 0.13333334028720856, 0.0, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.0, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.3333333432674408, NaN, 0.6000000238418579, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.0, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.3333333432674408, 0.5333333611488342, 0.2666666805744171, 0.3333333432674408, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.13333334028720856, 0.46666666865348816, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, NaN, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.0, 0.3333333432674408, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.0, 0.3333333432674408, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9342995208},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9333333364},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9082125646},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8763285094},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8570048371},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8260869643},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8241545975},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7961352742},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7913043558},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7594202992},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7309178824},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7304347918},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7207729551},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7072463877},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7043478344},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.70000001},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6908212654},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.688888898},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6879227137},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6782608786},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6772946949},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.676328513},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6589372061},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.6560386569},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6512077399},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6396135385},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6328502513},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5840579801},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5835748888},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5748792382},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5603864822},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.554589381},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.548792281},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5478260989},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5198067715},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5130434869},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4966183678},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4714975973},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.458454117},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4541062891},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4541062888},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4202898637},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4164251286},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4077294754},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4038647416},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4028985602},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3908212635},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.3874396235},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3681159495},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.364251215},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3594202979},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3555555642},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2879227117},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.265700491},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2647343059},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2309178809},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2236715042},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2193236783},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2057971065},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2000000054},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.1942029045},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.1893719864},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1826087009},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1608695692},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1516908259},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1376811637},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1207729511},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1082125632},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0850241575},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0347826099}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_all/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..049f534de55b0347b36285de3e07e55cbfafd6e2
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,71 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)","AutoGluon 1.4 (best, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),TabICLv2 (default),ModernNCA (tuned + ensembled),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),CatBoost (tuned + ensembled),LightGBM (tuned + ensembled),CatBoost (tuned),TabM (tuned + ensembled),XGBoost (tuned + ensembled),ModernNCA (tuned),iLTM (tuned + ensembled),LightGBM (tuned),CatBoost (default),RealMLP (tuned),RealTabPFN-2.5 (default),ChimeraBoost (tuned + ensembled),XGBoost (tuned),ChimeraBoost (tuned),TabM (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),TabDPT (tuned),xRFM (tuned + ensembled),ChimeraBoost (default),iLTM (tuned),xRFM (tuned),XGBoost (default),TabSwift (default),TorchMLP (tuned + ensembled),TabM (default),EBM (tuned + ensembled),LimiX (default),TabDPT (default),ModernNCA (default),RealMLP (default),RandomForest (tuned + ensembled),TorchMLP (tuned),LightGBM (default),ExtraTrees (tuned + ensembled),EBM (tuned),RandomForest (tuned),EBM (default),FastaiMLP (tuned + ensembled),ExtraTrees (tuned),SAP-RPT-OSS (default),iLTM (default),TorchMLP (default),FastaiMLP (tuned),PerpetualBooster (tuned + ensembled),xRFM (default),RandomForest (default),PerpetualBooster (tuned),KNN (tuned + ensembled),FastaiMLP (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),ExtraTrees (default),Linear (tuned + ensembled),PerpetualBooster (default),Linear (tuned),TabSTAR (default),KNN (tuned),Linear (default),KNN (default)
+TabFM (default),,0.6,0.6,0.73333335,0.8666667,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.93333334,0.8666667,1.0,0.8666667,0.8666667,0.8,1.0,0.8666667,0.8666667,0.93333334,0.8666667,0.8,0.8666667,1.0,1.0,0.8666667,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.8666667,0.93333334,1.0,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.4,,0.6666667,0.4,0.6666667,0.8,0.53333336,0.8,0.8,0.8666667,1.0,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,0.93333334,0.8,1.0,1.0,1.0,0.93333334,0.93333334,0.8,0.73333335,0.93333334,1.0,0.93333334,1.0,0.93333334,0.93333334,1.0,1.0,1.0,0.8666667,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.4,0.33333334,,0.4,0.73333335,0.53333336,0.53333336,0.6,0.6666667,0.8,1.0,0.93333334,1.0,0.8,1.0,0.8,1.0,0.93333334,0.93333334,0.93333334,0.6666667,0.93333334,1.0,0.93333334,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,0.8666667,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.26666668,0.6,0.6,,0.6666667,0.6666667,0.73333335,0.8666667,0.8,0.73333335,0.73333335,0.8,0.73333335,0.8666667,0.73333335,0.8,0.8,0.93333334,0.73333335,0.73333335,0.93333334,0.8,0.73333335,0.8,0.93333334,0.93333334,0.73333335,0.8,0.8,0.8,0.93333334,0.8666667,0.8666667,0.93333334,0.8,1.0,0.8666667,0.8666667,1.0,1.0,1.0,0.93333334,1.0,1.0,0.93333334,0.93333334,1.0,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.13333334,0.33333334,0.26666668,0.33333334,,0.46666667,0.53333336,0.73333335,0.53333336,0.73333335,0.73333335,0.8666667,0.73333335,0.8,0.6666667,0.8,0.73333335,0.93333334,0.8,0.8,0.8,0.8666667,0.8,0.8666667,0.8666667,0.8666667,0.8666667,0.8,0.73333335,0.8,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,1.0,1.0,0.8666667,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.13333334,0.2,0.46666667,0.33333334,0.53333336,,0.4,0.53333336,0.8,0.73333335,0.73333335,0.73333335,0.6666667,0.6,0.73333335,0.8,0.6666667,0.8,0.8,0.73333335,0.8666667,0.73333335,0.8,0.8,0.73333335,0.8,0.73333335,0.8,0.8,0.8,0.8,0.8,0.8666667,0.93333334,0.8,0.93333334,0.93333334,0.73333335,0.8666667,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,0.93333334,0.8666667,1.0,0.93333334,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.13333334,0.46666667,0.46666667,0.26666668,0.46666667,0.6,,0.73333335,0.73333335,0.6666667,0.73333335,0.8,0.73333335,0.73333335,0.73333335,0.8,0.73333335,0.8666667,0.73333335,0.73333335,0.73333335,0.73333335,0.73333335,0.73333335,0.73333335,0.8666667,0.8,0.8,0.73333335,0.73333335,0.8,0.73333335,0.73333335,0.8666667,0.8,1.0,0.73333335,0.8666667,0.8666667,0.93333334,0.93333334,0.93333334,0.8,0.93333334,1.0,0.93333334,0.93333334,0.93333334,0.93333334,1.0,0.93333334,0.93333334,1.0,1.0,0.93333334,0.8666667,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.13333334,0.2,0.4,0.13333334,0.26666668,0.46666667,0.26666668,,0.53333336,0.6,0.6,0.6,0.6,0.6666667,0.73333335,0.8,0.6666667,0.6666667,0.73333335,0.8,0.73333335,0.73333335,0.73333335,0.73333335,0.73333335,0.6666667,0.8666667,0.8,0.93333334,0.8666667,0.8,0.8666667,0.73333335,0.8,0.8666667,0.8666667,0.8666667,0.8,0.8666667,1.0,0.93333334,0.8666667,0.93333334,1.0,0.93333334,0.8666667,0.8666667,0.93333334,1.0,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.93333334,0.93333334,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.06666667,0.2,0.33333334,0.2,0.46666667,0.2,0.26666668,0.46666667,,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.6,0.73333335,0.6,0.8,0.73333335,0.73333335,0.73333335,0.6,0.73333335,0.73333335,0.8,0.8,0.73333335,0.8,0.73333335,0.73333335,0.8,0.8,0.8666667,0.8666667,0.8,0.93333334,0.93333334,0.73333335,0.8666667,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,0.93333334,0.8666667,0.8666667,0.93333334,0.8666667,1.0,1.0,1.0,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.13333334,0.13333334,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.4,0.33333334,,0.6,0.53333336,0.6,0.53333336,0.53333336,0.53333336,0.6666667,0.6,0.6,0.8,0.6666667,0.46666667,0.6666667,0.6,0.6666667,0.6666667,0.6666667,0.6666667,0.8666667,0.6666667,0.8,0.93333334,0.8,0.73333335,0.8666667,0.93333334,0.73333335,0.8,0.8,0.93333334,1.0,0.93333334,0.93333334,1.0,0.93333334,0.8666667,0.93333334,0.8666667,1.0,0.8666667,1.0,0.8666667,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.13333334,0.0,0.0,0.26666668,0.26666668,0.26666668,0.26666668,0.4,0.33333334,0.4,,0.4,0.6666667,0.4,0.53333336,0.43333334,0.6,0.46666667,0.8,0.6,0.6,0.6666667,0.6666667,0.8,0.46666667,0.46666667,0.6,0.6666667,0.73333335,0.8666667,0.8666667,0.8666667,0.8666667,0.6666667,0.8,0.8,0.73333335,0.6666667,0.73333335,0.73333335,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.73333335,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.06666667,0.0,0.06666667,0.2,0.13333334,0.26666668,0.2,0.4,0.33333334,0.46666667,0.6,,0.6666667,0.33333334,0.6,0.46666667,0.6666667,0.73333335,0.73333335,0.53333336,0.53333336,0.6,0.6,0.73333335,0.53333336,0.6666667,0.53333336,0.6,0.6,0.73333335,0.8,0.73333335,0.8666667,0.6,0.73333335,0.8666667,0.8,0.6666667,0.8,0.8666667,0.8666667,0.93333334,0.8666667,1.0,1.0,1.0,1.0,1.0,0.8666667,1.0,0.8666667,1.0,1.0,0.93333334,1.0,0.8666667,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.13333334,0.0,0.0,0.26666668,0.26666668,0.33333334,0.26666668,0.4,0.33333334,0.4,0.33333334,0.33333334,,0.46666667,0.53333336,0.46666667,0.6,0.46666667,0.6666667,0.6,0.53333336,0.6666667,0.6666667,0.6666667,0.46666667,0.46666667,0.6,0.6,0.73333335,0.8666667,0.8666667,0.8,0.8666667,0.6666667,0.8,0.8666667,0.8,0.6666667,0.73333335,0.73333335,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.73333335,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.06666667,0.2,0.13333334,0.2,0.4,0.26666668,0.33333334,0.33333334,0.46666667,0.6,0.6666667,0.53333336,,0.46666667,0.53333336,0.53333336,0.6666667,0.6,0.53333336,0.46666667,0.53333336,0.6666667,0.53333336,0.6666667,0.6,0.53333336,0.6666667,0.53333336,0.6,0.6666667,0.53333336,0.6666667,0.6666667,0.73333335,1.0,0.8666667,0.6666667,0.73333335,0.8,1.0,0.8666667,0.8,0.93333334,0.8666667,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.8,1.0,1.0,0.93333334,0.8666667,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.13333334,0.0,0.0,0.26666668,0.33333334,0.26666668,0.26666668,0.26666668,0.4,0.46666667,0.46666667,0.4,0.46666667,0.53333336,,0.4,0.4,0.4,0.53333336,0.46666667,0.53333336,0.46666667,0.73333335,0.53333336,0.53333336,0.46666667,0.6,0.53333336,0.6,0.73333335,0.73333335,0.6666667,0.8666667,0.53333336,0.8666667,0.8,0.8,0.6,0.8,0.8,0.8666667,0.93333334,0.8666667,1.0,1.0,0.8666667,0.93333334,1.0,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.13333334,0.06666667,0.2,0.2,0.2,0.2,0.2,0.2,0.26666668,0.46666667,0.56666666,0.53333336,0.53333336,0.46666667,0.6,,0.53333336,0.53333336,0.53333336,0.53333336,0.53333336,0.46666667,0.6,0.46666667,0.53333336,0.53333336,0.6666667,0.8,0.6,0.6666667,0.6666667,0.8,0.6,0.8666667,0.8666667,0.73333335,0.6666667,0.73333335,0.8666667,0.93333334,0.8,0.8,0.8666667,0.8666667,0.8666667,0.8,0.93333334,0.8666667,0.8666667,0.8666667,0.8666667,0.93333334,0.93333334,1.0,0.93333334,0.93333334,1.0,0.93333334,1.0,1.0,0.93333334,0.93333334,0.93333334,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.2,0.0,0.0,0.2,0.26666668,0.33333334,0.26666668,0.33333334,0.4,0.33333334,0.4,0.33333334,0.4,0.46666667,0.6,0.46666667,,0.4,0.46666667,0.53333336,0.6,0.53333336,0.6666667,0.53333336,0.53333336,0.4,0.53333336,0.6,0.6,0.6,0.73333335,0.6,0.73333335,0.6,0.8666667,0.73333335,0.73333335,0.73333335,0.73333335,0.6666667,0.8666667,0.93333334,1.0,0.8666667,0.8666667,0.6666667,0.93333334,0.8,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.06666667,0.06666667,0.06666667,0.2,0.13333334,0.33333334,0.2,0.4,0.53333336,0.26666668,0.53333336,0.33333334,0.6,0.46666667,0.6,,0.53333336,0.46666667,0.53333336,0.6,0.6,0.73333335,0.46666667,0.6666667,0.6,0.6,0.6,0.73333335,0.6666667,0.6666667,0.8,0.46666667,0.6666667,0.8666667,0.8,0.6666667,0.8,0.8,0.8666667,0.8666667,0.8,1.0,0.8666667,0.93333334,1.0,0.93333334,0.8666667,0.93333334,0.8666667,1.0,0.93333334,0.93333334,1.0,0.8666667,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.13333334,0.0,0.06666667,0.26666668,0.2,0.2,0.26666668,0.26666668,0.26666668,0.4,0.2,0.26666668,0.33333334,0.4,0.46666667,0.46666667,0.53333336,0.46666667,,0.6,0.6,0.4,0.53333336,0.53333336,0.46666667,0.46666667,0.6,0.6,0.73333335,0.73333335,0.8,0.73333335,0.73333335,0.6666667,0.8666667,0.73333335,0.8,0.6666667,0.73333335,0.8,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.8,0.8666667,0.93333334,0.93333334,0.8666667,0.8666667,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.13333334,0.06666667,0.06666667,0.26666668,0.2,0.26666668,0.26666668,0.2,0.26666668,0.2,0.4,0.46666667,0.4,0.46666667,0.53333336,0.46666667,0.46666667,0.53333336,0.4,,0.53333336,0.33333334,0.6,0.4,0.53333336,0.46666667,0.6666667,0.6666667,0.8,0.53333336,0.73333335,0.73333335,0.73333335,0.6,0.8666667,0.73333335,0.6666667,0.73333335,0.73333335,0.8,0.93333334,0.8666667,0.93333334,0.8666667,0.73333335,0.8,0.8666667,0.8,0.93333334,0.8,0.8666667,0.8,1.0,1.0,0.93333334,0.93333334,0.8666667,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.06666667,0.2,0.33333334,0.06666667,0.2,0.13333334,0.26666668,0.26666668,0.26666668,0.33333334,0.4,0.46666667,0.46666667,0.53333336,0.46666667,0.46666667,0.4,0.46666667,0.4,0.46666667,,0.46666667,0.6,0.46666667,0.6,0.53333336,0.6666667,0.6,0.73333335,0.6666667,0.73333335,0.6666667,0.73333335,0.8666667,0.6666667,0.8,0.6666667,0.6666667,0.6666667,0.8666667,0.8,0.8,0.8666667,1.0,0.8666667,0.8,0.8666667,0.8,0.8,0.8666667,0.93333334,0.8,1.0,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.8666667,1.0,1.0,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0
+ChimeraBoost (tuned + ensembled),0.13333334,0.0,0.06666667,0.2,0.13333334,0.26666668,0.26666668,0.26666668,0.4,0.53333336,0.33333334,0.4,0.33333334,0.46666667,0.53333336,0.53333336,0.46666667,0.4,0.6,0.6666667,0.53333336,,0.53333336,0.6,0.46666667,0.46666667,0.6666667,0.6,0.73333335,0.8666667,0.73333335,0.8,0.73333335,0.53333336,0.73333335,0.73333335,0.8,0.6,0.6666667,0.73333335,0.8666667,0.8666667,0.8,0.8666667,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.8666667,0.8666667,0.8,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,1.0,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0
+XGBoost (tuned),0.2,0.0,0.0,0.26666668,0.2,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.33333334,0.4,0.33333334,0.33333334,0.26666668,0.4,0.33333334,0.4,0.46666667,0.4,0.4,0.46666667,,0.46666667,0.46666667,0.46666667,0.6666667,0.53333336,0.6666667,0.6666667,0.6666667,0.6,0.8666667,0.53333336,0.73333335,0.73333335,0.6666667,0.6,0.73333335,0.8,0.8,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.93333334,0.8,0.93333334,0.8666667,0.93333334,1.0,0.93333334,0.8666667,0.8,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.13333334,0.0,0.06666667,0.2,0.13333334,0.2,0.26666668,0.26666668,0.26666668,0.4,0.2,0.26666668,0.33333334,0.46666667,0.46666667,0.53333336,0.46666667,0.26666668,0.46666667,0.6,0.53333336,0.4,0.53333336,,0.46666667,0.46666667,0.6666667,0.6,0.73333335,0.8,0.73333335,0.8,0.73333335,0.6,0.73333335,0.73333335,0.8,0.6,0.6,0.73333335,0.8666667,0.8666667,0.8,0.8666667,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.8,0.8666667,0.8,0.8,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,1.0,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0
+TabM (tuned),0.0,0.06666667,0.2,0.06666667,0.13333334,0.26666668,0.26666668,0.26666668,0.2,0.33333334,0.53333336,0.46666667,0.53333336,0.33333334,0.46666667,0.46666667,0.46666667,0.53333336,0.53333336,0.46666667,0.4,0.53333336,0.53333336,0.53333336,,0.53333336,0.46666667,0.6,0.46666667,0.46666667,0.6,0.53333336,0.6666667,0.6,0.6,0.8666667,0.73333335,0.6666667,0.6,0.8,0.8666667,0.73333335,0.8,0.8,0.8,0.8666667,0.8,0.8,0.8666667,0.8,0.8666667,0.73333335,1.0,0.93333334,0.93333334,0.8666667,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.06666667,0.2,0.06666667,0.13333334,0.2,0.13333334,0.33333334,0.2,0.33333334,0.53333336,0.33333334,0.53333336,0.4,0.53333336,0.46666667,0.6,0.33333334,0.53333336,0.53333336,0.46666667,0.53333336,0.53333336,0.53333336,0.46666667,,0.33333334,0.53333336,0.6,0.53333336,0.73333335,0.73333335,0.6,0.73333335,0.73333335,0.6666667,0.6,0.8,0.8666667,0.73333335,0.73333335,0.73333335,0.8666667,0.8,0.8,0.6666667,0.8666667,0.6666667,0.73333335,0.8,0.73333335,0.8,0.8666667,0.93333334,0.93333334,0.93333334,1.0,1.0,0.8666667,0.93333334,0.8666667,0.8666667,1.0,0.8,1.0,0.8,1.0,1.0,0.93333334,1.0
+TabDPT (tuned + ensembled),0.13333334,0.2,0.2,0.26666668,0.13333334,0.26666668,0.2,0.13333334,0.26666668,0.33333334,0.4,0.46666667,0.4,0.46666667,0.4,0.33333334,0.46666667,0.4,0.4,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.53333336,0.6666667,,0.8,0.6666667,0.46666667,0.6,0.6666667,0.53333336,0.53333336,0.8,0.6666667,0.53333336,0.6666667,0.93333334,0.8666667,0.73333335,0.73333335,0.8666667,0.8666667,0.73333335,0.73333335,0.73333335,0.73333335,0.73333335,0.8,0.8,0.6666667,0.8666667,0.93333334,1.0,0.93333334,1.0,1.0,0.93333334,0.93333334,0.8666667,0.8666667,1.0,0.8666667,1.0,0.8666667,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.06666667,0.26666668,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.33333334,0.33333334,0.4,0.4,0.33333334,0.46666667,0.2,0.4,0.4,0.4,0.33333334,0.4,0.4,0.46666667,0.4,0.4,0.46666667,0.2,,0.6,0.46666667,0.6,0.6666667,0.46666667,0.53333336,0.53333336,0.53333336,0.46666667,0.6666667,0.9,0.6,0.53333336,0.6,0.8,0.73333335,0.6666667,0.6666667,0.73333335,0.6,0.6,0.8,0.73333335,0.73333335,0.8666667,0.8666667,1.0,0.93333334,0.8666667,1.0,0.8666667,0.93333334,0.8666667,0.8666667,1.0,0.8,1.0,0.8,1.0,1.0,0.93333334,1.0
+xRFM (tuned + ensembled),0.06666667,0.06666667,0.0,0.2,0.26666668,0.2,0.26666668,0.06666667,0.26666668,0.13333334,0.26666668,0.4,0.26666668,0.46666667,0.4,0.4,0.4,0.4,0.26666668,0.2,0.26666668,0.26666668,0.33333334,0.26666668,0.53333336,0.4,0.33333334,0.4,,0.33333334,0.53333336,0.6666667,0.46666667,0.46666667,0.6,0.6,0.6,0.73333335,0.53333336,0.73333335,0.73333335,0.8,0.6666667,0.6666667,0.8,0.8,0.8,0.6666667,0.8,0.8,0.73333335,0.73333335,0.8666667,0.8666667,0.93333334,1.0,0.93333334,0.93333334,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0
+ChimeraBoost (default),0.06666667,0.0,0.0,0.2,0.2,0.2,0.26666668,0.13333334,0.26666668,0.33333334,0.13333334,0.26666668,0.13333334,0.4,0.26666668,0.33333334,0.4,0.26666668,0.26666668,0.46666667,0.33333334,0.13333334,0.33333334,0.2,0.53333336,0.46666667,0.53333336,0.53333336,0.6666667,,0.53333336,0.6666667,0.6,0.46666667,0.6666667,0.53333336,0.6666667,0.6,0.6,0.6666667,0.6666667,0.73333335,0.8,0.6666667,0.6,0.8,0.73333335,0.8,0.8,0.6,0.8,0.73333335,0.8,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.93333334,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.8666667,1.0,0.93333334,1.0,1.0,1.0
+iLTM (tuned),0.06666667,0.06666667,0.0,0.06666667,0.13333334,0.2,0.2,0.2,0.2,0.2,0.13333334,0.2,0.13333334,0.33333334,0.26666668,0.33333334,0.26666668,0.33333334,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.26666668,0.4,0.26666668,0.4,0.4,0.46666667,0.46666667,,0.46666667,0.4,0.53333336,0.6666667,0.53333336,0.6,0.6666667,0.6,0.6,0.6666667,0.8,0.73333335,0.8,0.6,0.6666667,0.8666667,0.6666667,0.8666667,0.8,0.8666667,0.8,0.93333334,0.8666667,0.8666667,0.93333334,0.93333334,0.8666667,1.0,0.93333334,1.0,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.06666667,0.0,0.0,0.13333334,0.13333334,0.2,0.26666668,0.13333334,0.2,0.06666667,0.13333334,0.26666668,0.2,0.46666667,0.33333334,0.2,0.4,0.33333334,0.26666668,0.26666668,0.33333334,0.2,0.4,0.2,0.46666667,0.26666668,0.33333334,0.33333334,0.33333334,0.33333334,0.53333336,,0.46666667,0.46666667,0.6,0.6,0.6,0.6,0.53333336,0.6666667,0.73333335,0.73333335,0.8,0.6666667,0.73333335,0.6666667,0.8,0.6,0.73333335,0.73333335,0.8666667,0.6666667,0.93333334,0.8666667,0.8666667,1.0,0.8666667,0.93333334,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.13333334,0.26666668,0.26666668,0.13333334,0.2,0.13333334,0.13333334,0.13333334,0.33333334,0.13333334,0.4,0.26666668,0.2,0.26666668,0.26666668,0.26666668,0.26666668,0.13333334,0.26666668,0.33333334,0.4,0.46666667,0.53333336,0.53333336,0.4,0.6,0.53333336,,0.46666667,0.6,0.4,0.6,0.6,0.6666667,0.6666667,0.53333336,0.8,0.6666667,0.53333336,0.73333335,0.73333335,0.8,0.8,0.8,0.73333335,0.8,0.93333334,0.8666667,0.8,0.8,0.8,0.93333334,0.93333334,0.93333334,0.8666667,1.0,1.0,0.8666667,0.93333334,1.0,0.93333334,1.0,1.0,0.93333334,1.0
+TabSwift (default),0.13333334,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.2,0.13333334,0.26666668,0.33333334,0.4,0.33333334,0.33333334,0.46666667,0.13333334,0.4,0.53333336,0.33333334,0.4,0.13333334,0.46666667,0.46666667,0.4,0.4,0.26666668,0.46666667,0.46666667,0.53333336,0.53333336,0.46666667,0.53333336,0.53333336,,0.53333336,0.6,0.6,0.6,0.6666667,0.6,0.6,0.6,0.6,0.73333335,0.6,0.6666667,0.6,0.73333335,0.8,0.6666667,0.73333335,0.73333335,0.73333335,0.8,0.8666667,0.8,0.8,0.8666667,0.8666667,0.8,0.8666667,0.8666667,0.93333334,0.8,0.93333334,0.8666667,0.93333334,1.0,0.8666667,1.0
+TorchMLP (tuned + ensembled),0.06666667,0.0,0.0,0.2,0.13333334,0.2,0.2,0.13333334,0.2,0.13333334,0.2,0.26666668,0.2,0.26666668,0.13333334,0.13333334,0.13333334,0.33333334,0.13333334,0.13333334,0.33333334,0.26666668,0.26666668,0.26666668,0.4,0.26666668,0.2,0.46666667,0.4,0.33333334,0.33333334,0.4,0.4,0.46666667,,0.4,0.6,0.46666667,0.53333336,0.46666667,0.6,0.73333335,0.8,0.53333336,0.73333335,0.6666667,0.8,0.6,0.6666667,0.73333335,0.8666667,0.8,1.0,0.93333334,0.73333335,0.8,0.93333334,0.73333335,1.0,0.93333334,1.0,1.0,0.8666667,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.13333334,0.06666667,0.06666667,0.2,0.13333334,0.13333334,0.0,0.2,0.26666668,0.26666668,0.13333334,0.26666668,0.26666668,0.2,0.26666668,0.26666668,0.26666668,0.13333334,0.33333334,0.33333334,0.46666667,0.4,0.46666667,0.46666667,0.4,0.6,0.4,0.6,,0.46666667,0.6,0.53333336,0.6,0.73333335,0.73333335,0.6,0.8,0.6666667,0.6666667,0.73333335,0.73333335,0.8666667,0.73333335,0.73333335,0.73333335,0.8,0.93333334,0.93333334,0.8,0.8666667,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,0.8666667,0.93333334,0.8,1.0,1.0,0.93333334,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.13333334,0.0,0.06666667,0.26666668,0.13333334,0.06666667,0.26666668,0.26666668,0.2,0.2,0.13333334,0.2,0.33333334,0.26666668,0.2,0.2,0.33333334,0.33333334,0.2,0.33333334,0.2,0.26666668,0.4,0.46666667,0.53333336,0.4,0.33333334,0.4,0.4,0.4,0.4,0.4,0.53333336,,0.46666667,0.53333336,0.6666667,0.53333336,0.6,0.6,0.53333336,0.6,0.93333334,0.6666667,1.0,0.6666667,0.6,0.73333335,0.6,0.8666667,0.73333335,0.6666667,0.73333335,0.73333335,0.6666667,0.8666667,0.8666667,1.0,0.93333334,0.73333335,0.93333334,0.73333335,0.93333334,1.0,0.93333334,1.0,0.93333334
+LimiX (default),0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.26666668,0.13333334,0.2,0.26666668,0.2,0.33333334,0.33333334,0.33333334,0.33333334,0.4,0.26666668,0.26666668,0.33333334,0.33333334,0.26666668,0.33333334,0.4,0.4,0.4,0.33333334,0.2,0.33333334,0.33333334,0.26666668,0.4,0.33333334,0.4,0.4,0.4,0.53333336,0.4,0.53333336,,0.53333336,0.46666667,0.6,0.53333336,0.6,0.4,0.4,0.53333336,0.53333336,0.53333336,0.6,0.46666667,0.6,0.53333336,0.6666667,0.73333335,0.6,0.8,0.8666667,0.6666667,0.8,0.8,0.6666667,0.6666667,0.8666667,0.8,0.8,0.8,0.8,0.8,0.8,0.93333334
+TabDPT (default),0.0,0.13333334,0.13333334,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.26666668,0.2,0.26666668,0.26666668,0.2,0.13333334,0.26666668,0.2,0.26666668,0.26666668,0.33333334,0.33333334,0.26666668,0.4,0.4,0.13333334,0.06666667,0.1,0.46666667,0.4,0.4,0.46666667,0.33333334,0.33333334,0.46666667,0.46666667,0.46666667,0.46666667,,0.53333336,0.46666667,0.6,0.6,0.6,0.53333336,0.53333336,0.6666667,0.46666667,0.6,0.53333336,0.6,0.53333336,0.8,0.73333335,0.73333335,0.8,0.73333335,0.8666667,0.6666667,0.8666667,0.8,0.8,0.73333335,0.8,1.0,0.8,0.8,0.8,0.8666667,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.06666667,0.26666668,0.13333334,0.26666668,0.2,0.2,0.06666667,0.33333334,0.2,0.2,0.2,0.13333334,0.26666668,0.2,0.26666668,0.2,0.26666668,0.13333334,0.4,0.26666668,0.33333334,0.4,0.33333334,0.33333334,0.4,0.53333336,0.4,0.33333334,0.53333336,0.46666667,,0.46666667,0.6,0.46666667,0.6,0.53333336,0.53333336,0.6,0.4,0.53333336,0.6,0.6,0.6,0.93333334,0.73333335,0.8,0.8666667,0.93333334,0.93333334,0.93333334,0.93333334,0.8666667,0.8666667,0.93333334,0.8,1.0,0.8,1.0,1.0,0.93333334,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.0,0.13333334,0.13333334,0.06666667,0.0,0.13333334,0.2,0.13333334,0.13333334,0.13333334,0.06666667,0.2,0.13333334,0.2,0.13333334,0.13333334,0.26666668,0.26666668,0.46666667,0.26666668,0.33333334,0.33333334,0.26666668,0.46666667,0.4,0.4,0.26666668,0.46666667,0.4,0.53333336,0.53333336,,0.6666667,0.6,0.53333336,0.6,0.6666667,0.73333335,0.6666667,0.6666667,0.6666667,0.6666667,0.6,0.8,1.0,0.8,0.73333335,0.8,0.8666667,0.8666667,1.0,0.8666667,0.8666667,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.13333334,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.2,0.06666667,0.13333334,0.13333334,0.13333334,0.2,0.13333334,0.13333334,0.13333334,0.26666668,0.26666668,0.26666668,0.4,0.2,0.26666668,0.2,0.26666668,0.2,0.4,0.26666668,0.26666668,0.4,0.46666667,0.4,0.4,0.33333334,,0.53333336,0.53333336,0.53333336,0.53333336,0.53333336,0.46666667,0.53333336,0.6666667,0.73333335,0.8,0.73333335,0.8,0.6666667,0.6666667,0.8666667,0.73333335,0.93333334,0.93333334,1.0,1.0,0.8666667,0.8666667,0.8,0.8666667,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.06666667,0.2,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.2,0.13333334,0.13333334,0.0,0.2,0.06666667,0.06666667,0.13333334,0.2,0.13333334,0.2,0.2,0.13333334,0.13333334,0.2,0.33333334,0.2,0.26666668,0.2,0.33333334,0.4,0.2,0.4,0.4,0.4,0.4,0.53333336,0.4,0.46666667,,0.53333336,0.6,0.6,0.46666667,0.53333336,0.53333336,0.6666667,0.73333335,0.6666667,0.93333334,0.6,0.73333335,0.8,0.73333335,0.73333335,0.8666667,0.8,0.93333334,0.93333334,0.8,0.93333334,0.8666667,0.93333334,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.13333334,0.13333334,0.0,0.13333334,0.13333334,0.0,0.13333334,0.13333334,0.13333334,0.2,0.2,0.13333334,0.26666668,0.33333334,0.33333334,0.2,0.33333334,0.46666667,0.26666668,0.46666667,0.2,0.46666667,0.6,0.4,0.4,0.46666667,0.46666667,0.46666667,,0.53333336,0.53333336,0.46666667,0.46666667,0.6,0.6,0.6,0.6666667,0.8,0.73333335,0.8,0.8,0.8666667,0.93333334,0.8666667,0.8,0.8666667,0.93333334,0.93333334,0.8,0.93333334,0.8,1.0,1.0,0.93333334,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.06666667,0.13333334,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.26666668,0.13333334,0.13333334,0.06666667,0.13333334,0.2,0.2,0.26666668,0.33333334,0.2,0.4,0.4,0.26666668,0.26666668,0.4,0.26666668,0.33333334,0.4,0.6,0.46666667,0.46666667,0.4,0.46666667,0.4,0.46666667,,0.46666667,0.46666667,0.46666667,0.53333336,0.8,0.6,0.6666667,0.53333336,0.6666667,0.6666667,0.6666667,0.93333334,0.73333335,0.8666667,0.8,0.8666667,0.8666667,1.0,0.8,0.8666667,0.93333334,0.8666667,0.93333334,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.13333334,0.06666667,0.13333334,0.26666668,0.0,0.26666668,0.13333334,0.13333334,0.2,0.33333334,0.06666667,0.2,0.2,0.2,0.13333334,0.13333334,0.13333334,0.13333334,0.33333334,0.26666668,0.33333334,0.2,0.2,0.33333334,0.33333334,0.26666668,0.33333334,0.33333334,0.33333334,0.06666667,0.46666667,0.46666667,0.46666667,0.33333334,0.46666667,0.4,0.46666667,0.53333336,,0.53333336,0.6,0.53333336,0.6,0.6666667,0.53333336,0.6666667,0.73333335,0.6666667,0.6666667,0.6666667,0.6666667,0.8,0.8,0.93333334,0.93333334,0.6666667,0.8,0.6666667,0.8666667,1.0,0.93333334,0.93333334,0.93333334
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.13333334,0.0,0.06666667,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.06666667,0.06666667,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.13333334,0.26666668,0.26666668,0.2,0.26666668,0.13333334,0.2,0.2,0.4,0.2,0.26666668,0.33333334,0.46666667,0.33333334,0.4,0.26666668,0.46666667,0.53333336,0.53333336,0.53333336,0.46666667,,0.46666667,0.53333336,0.6,0.73333335,0.73333335,0.73333335,0.73333335,0.6666667,0.6666667,0.8333333,0.6666667,0.93333334,0.93333334,0.93333334,0.93333334,0.8666667,0.8,0.8,0.8,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.0,0.06666667,0.13333334,0.0,0.13333334,0.2,0.06666667,0.06666667,0.2,0.2,0.06666667,0.06666667,0.06666667,0.2,0.33333334,0.26666668,0.4,0.33333334,0.2,0.33333334,0.4,0.2,0.26666668,0.4,0.26666668,0.0,0.46666667,0.53333336,0.6,0.33333334,0.53333336,0.46666667,0.53333336,0.53333336,0.4,0.53333336,,0.6,0.6,0.6,0.53333336,0.6666667,0.73333335,0.6666667,0.6,0.6666667,0.6666667,0.8,0.73333335,0.8666667,0.8666667,0.6666667,0.8,0.6666667,0.8666667,1.0,0.93333334,0.93333334,0.93333334
+FastaiMLP (tuned + ensembled),0.06666667,0.0,0.0,0.06666667,0.06666667,0.13333334,0.06666667,0.0,0.13333334,0.0,0.06666667,0.13333334,0.06666667,0.06666667,0.13333334,0.13333334,0.06666667,0.13333334,0.06666667,0.06666667,0.2,0.13333334,0.2,0.13333334,0.13333334,0.26666668,0.26666668,0.4,0.2,0.2,0.13333334,0.26666668,0.2,0.2,0.33333334,0.13333334,0.33333334,0.4,0.4,0.46666667,0.33333334,0.46666667,0.46666667,0.4,0.46666667,0.46666667,0.46666667,0.4,,0.46666667,0.53333336,0.46666667,0.53333336,0.73333335,0.6,0.6666667,0.6,0.6,0.73333335,1.0,0.73333335,0.73333335,0.73333335,0.93333334,0.8,0.93333334,0.8666667,0.8,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.13333334,0.13333334,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.06666667,0.06666667,0.13333334,0.2,0.13333334,0.13333334,0.06666667,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.26666668,0.26666668,0.33333334,0.26666668,0.26666668,0.4,0.53333336,0.46666667,0.4,0.33333334,0.33333334,0.33333334,0.4,0.2,0.4,0.4,0.4,0.53333336,,0.46666667,0.46666667,0.46666667,0.6,0.6666667,0.6666667,0.8666667,0.8,0.73333335,0.6666667,0.73333335,0.73333335,0.93333334,0.8,0.8666667,0.8,0.8666667,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.06666667,0.13333334,0.06666667,0.13333334,0.13333334,0.13333334,0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.26666668,0.2,0.26666668,0.26666668,0.2,0.13333334,0.13333334,0.2,0.26666668,0.13333334,0.26666668,0.26666668,0.4,0.4,0.4,0.33333334,0.26666668,0.26666668,0.4,0.4,0.33333334,0.26666668,0.4,0.46666667,0.53333336,,0.46666667,0.6,0.73333335,0.73333335,0.6666667,0.6,0.73333335,0.8666667,0.8666667,0.8,0.8,0.8,0.8666667,0.73333335,0.8666667,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.06666667,0.13333334,0.13333334,0.06666667,0.0,0.06666667,0.2,0.06666667,0.06666667,0.0,0.0,0.06666667,0.2,0.2,0.13333334,0.06666667,0.2,0.26666668,0.2,0.33333334,0.26666668,0.26666668,0.26666668,0.2,0.33333334,0.06666667,0.26666668,0.2,0.26666668,0.4,0.46666667,0.46666667,0.4,0.4,0.2,0.33333334,0.33333334,0.33333334,0.46666667,0.26666668,0.46666667,0.53333336,0.53333336,0.53333336,,0.6,0.6666667,0.53333336,0.6,0.6,0.53333336,0.8,0.8,0.8666667,0.8666667,0.6666667,0.8666667,0.6,0.8666667,1.0,0.8666667,0.93333334,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.2,0.0,0.2,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.06666667,0.06666667,0.13333334,0.26666668,0.0,0.2,0.13333334,0.33333334,0.2,0.06666667,0.2,0.26666668,0.06666667,0.2,0.46666667,0.33333334,0.26666668,0.33333334,0.46666667,0.53333336,0.4,0.4,,0.6,0.6,0.6666667,0.6666667,0.6666667,0.8,0.6,0.73333335,0.73333335,0.73333335,0.8,0.73333335,0.8666667,0.93333334,1.0,0.93333334,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.06666667,0.06666667,0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.2,0.06666667,0.06666667,0.26666668,0.26666668,0.26666668,0.26666668,0.0,0.2,0.4,0.26666668,0.33333334,0.26666668,0.26666668,0.26666668,0.26666668,0.4,0.26666668,0.33333334,0.4,,0.6,0.6,0.53333336,0.6,0.53333336,0.8666667,0.6666667,0.6666667,0.6666667,0.73333335,0.6666667,0.73333335,0.8666667,0.8,0.93333334,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.06666667,0.06666667,0.13333334,0.13333334,0.13333334,0.06666667,0.06666667,0.0,0.0,0.06666667,0.13333334,0.13333334,0.13333334,0.2,0.13333334,0.26666668,0.06666667,0.33333334,0.4,0.26666668,0.2,0.2,0.33333334,0.26666668,0.2,0.33333334,0.33333334,0.33333334,0.33333334,0.4,0.33333334,0.26666668,0.46666667,0.4,0.4,,0.6666667,0.6666667,1.0,0.6,0.53333336,0.46666667,0.46666667,0.73333335,0.6666667,1.0,0.6666667,0.46666667,0.73333335,0.6666667,1.0
+xRFM (default),0.06666667,0.0,0.0,0.06666667,0.06666667,0.06666667,0.13333334,0.0,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.06666667,0.06666667,0.0,0.13333334,0.06666667,0.06666667,0.06666667,0.13333334,0.2,0.13333334,0.13333334,0.06666667,0.06666667,0.06666667,0.0,0.13333334,0.06666667,0.0,0.2,0.2,0.2,0.2,0.26666668,0.2,0.2,0.13333334,0.26666668,0.33333334,0.2,0.2,0.33333334,0.33333334,0.33333334,0.4,0.33333334,0.33333334,0.33333334,0.4,0.33333334,0.4,0.33333334,,0.4,0.46666667,0.46666667,0.46666667,0.53333336,0.53333336,0.4,0.46666667,0.53333336,0.46666667,0.53333336,0.6,0.46666667,0.8
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.13333334,0.06666667,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.13333334,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.2,0.06666667,0.13333334,0.26666668,0.13333334,0.26666668,0.06666667,0.2,0.13333334,0.26666668,0.13333334,0.06666667,0.33333334,0.16666667,0.33333334,0.4,0.13333334,0.4,0.4,0.33333334,0.46666667,0.33333334,0.6,,0.4,0.53333336,0.6666667,0.6,0.6,0.73333335,0.6666667,0.73333335,0.6666667,0.6666667,0.6666667,0.73333335,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.0,0.06666667,0.13333334,0.0,0.13333334,0.06666667,0.0,0.0,0.0,0.06666667,0.13333334,0.13333334,0.06666667,0.06666667,0.13333334,0.26666668,0.06666667,0.33333334,0.33333334,0.13333334,0.06666667,0.13333334,0.26666668,0.26666668,0.06666667,0.26666668,0.33333334,0.33333334,0.33333334,0.4,0.2,0.26666668,0.46666667,0.33333334,0.4,0.0,0.53333336,0.6,,0.53333336,0.53333336,0.46666667,0.46666667,0.53333336,0.6666667,0.93333334,0.6666667,0.46666667,0.53333336,0.6666667,1.0
+KNN (tuned + ensembled),0.06666667,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.0,0.06666667,0.0,0.0,0.06666667,0.13333334,0.0,0.06666667,0.13333334,0.2,0.33333334,0.06666667,0.13333334,0.06666667,0.13333334,0.13333334,0.13333334,0.2,0.06666667,0.2,0.26666668,0.26666668,0.13333334,0.2,0.2,0.46666667,0.4,0.53333336,0.46666667,0.46666667,,0.53333336,0.46666667,0.46666667,0.53333336,0.8,0.6666667,0.8,0.6,1.0,0.8666667,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.0,0.13333334,0.13333334,0.06666667,0.13333334,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.06666667,0.13333334,0.2,0.06666667,0.06666667,0.13333334,0.2,0.13333334,0.06666667,0.0,0.06666667,0.2,0.2,0.2,0.2,0.06666667,0.26666668,0.0,0.33333334,0.13333334,0.2,0.4,0.13333334,0.46666667,0.53333336,0.33333334,0.46666667,0.46666667,,0.6,0.6,0.46666667,0.73333335,0.6666667,0.73333335,0.73333335,0.6666667,0.8,0.8666667
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.13333334,0.13333334,0.13333334,0.06666667,0.06666667,0.0,0.06666667,0.0,0.13333334,0.0,0.0,0.0,0.33333334,0.2,0.13333334,0.13333334,0.0,0.06666667,0.13333334,0.13333334,0.06666667,0.06666667,0.13333334,0.26666668,0.26666668,0.2,0.13333334,0.26666668,0.33333334,0.53333336,0.46666667,0.4,0.53333336,0.53333336,0.4,,0.6666667,0.53333336,0.73333335,0.53333336,0.73333335,1.0,0.8,0.8666667,0.8
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.13333334,0.13333334,0.13333334,0.06666667,0.06666667,0.0,0.06666667,0.0,0.13333334,0.0,0.0,0.06666667,0.33333334,0.2,0.13333334,0.13333334,0.0,0.06666667,0.06666667,0.13333334,0.06666667,0.06666667,0.13333334,0.26666668,0.26666668,0.2,0.13333334,0.26666668,0.33333334,0.53333336,0.46666667,0.4,0.53333336,0.53333336,0.4,0.33333334,,0.53333336,0.73333335,0.53333336,0.73333335,1.0,0.8,0.8666667,0.8
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.13333334,0.06666667,0.13333334,0.0,0.26666668,0.13333334,0.26666668,0.06666667,0.06666667,0.13333334,0.2,0.06666667,0.0,0.33333334,0.13333334,0.33333334,0.26666668,0.06666667,0.2,0.33333334,0.26666668,0.33333334,0.26666668,0.6,0.26666668,0.46666667,0.46666667,0.53333336,0.46666667,0.46666667,,0.6666667,0.73333335,0.6666667,0.46666667,0.6666667,0.6666667,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.13333334,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.2,0.13333334,0.2,0.0,0.0,0.06666667,0.0,0.06666667,0.2,0.0,0.13333334,0.06666667,0.2,0.2,0.2,0.06666667,0.13333334,0.06666667,0.2,0.2,0.2,0.2,0.2,0.06666667,0.2,0.13333334,0.13333334,0.2,0.26666668,0.33333334,0.53333336,0.33333334,0.33333334,0.2,0.26666668,0.26666668,0.26666668,0.33333334,,0.46666667,0.9,0.33333334,0.4,0.8666667,0.6666667
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.0,0.06666667,0.13333334,0.06666667,0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.26666668,0.2,0.0,0.0,0.06666667,0.2,0.13333334,0.06666667,0.13333334,0.33333334,0.2,0.33333334,0.2,0.13333334,0.26666668,0.4,0.26666668,0.33333334,0.0,0.46666667,0.26666668,0.06666667,0.33333334,0.33333334,0.46666667,0.46666667,0.26666668,0.53333336,,0.53333336,0.46666667,0.4,0.53333336,0.8666667
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.2,0.13333334,0.2,0.0,0.0,0.0,0.0,0.06666667,0.13333334,0.0,0.2,0.06666667,0.2,0.2,0.2,0.0,0.13333334,0.06666667,0.2,0.06666667,0.13333334,0.2,0.13333334,0.06666667,0.2,0.13333334,0.13333334,0.13333334,0.26666668,0.33333334,0.53333336,0.33333334,0.33333334,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.1,0.46666667,,0.33333334,0.4,0.8,0.6666667
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.0,0.13333334,0.0,0.0,0.0,0.13333334,0.13333334,0.0,0.0,0.06666667,0.13333334,0.53333336,0.46666667,0.33333334,0.53333336,0.4,0.26666668,0.0,0.0,0.53333336,0.6666667,0.53333336,0.6666667,,0.6666667,0.6666667,0.73333335
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.2,0.2,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.2,0.0,0.0,0.13333334,0.0,0.2,0.26666668,0.4,0.33333334,0.46666667,0.0,0.33333334,0.2,0.2,0.33333334,0.6,0.6,0.6,0.33333334,,0.6,1.0
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.13333334,0.0,0.06666667,0.0,0.2,0.13333334,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.06666667,0.06666667,0.33333334,0.53333336,0.26666668,0.33333334,0.13333334,0.2,0.13333334,0.13333334,0.33333334,0.13333334,0.46666667,0.2,0.33333334,0.4,,0.6666667
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.13333334,0.2,0.2,0.0,0.33333334,0.13333334,0.33333334,0.26666668,0.0,0.33333334,
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_small/leaderboard_overview_explorer.html
index 2e2475cf0ffb31b2d4f88353ddea74cf5e508196..ec7bbc5dc2f67293f4cb89081f3eda0f9f891c77 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | All Tasks | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_small/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_all/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..5e5e92122101407b57de1c9f302449cc8f7825e7
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_all/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1784.0,"score":0.85,"rank":5.75,"hrank":1.46,"imp":4.376,"train_time":45.35,"predict_time":5.343,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+243\/-123","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1634.0,"score":0.599,"rank":10.68,"hrank":5.11,"imp":9.885,"train_time":8.57,"predict_time":0.411,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-56","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1617.0,"score":0.602,"rank":11.42,"hrank":4.92,"imp":8.696,"train_time":345.57,"predict_time":5.168,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-78","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1607.0,"score":0.578,"rank":11.83,"hrank":6.06,"imp":10.985,"train_time":7.07,"predict_time":0.555,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-69","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1606.0,"score":0.572,"rank":11.88,"hrank":5.01,"imp":9.653,"train_time":2320.56,"predict_time":8.956,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-92","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1578.0,"score":0.502,"rank":13.22,"hrank":8.29,"imp":11.611,"train_time":6.58,"predict_time":0.586,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-64","imputed":false},{"position":6,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1555.0,"score":0.538,"rank":14.36,"hrank":6.2,"imp":11.438,"train_time":3.89,"predict_time":0.15,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-88","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1548.0,"score":0.511,"rank":14.74,"hrank":7.93,"imp":11.832,"train_time":2320.56,"predict_time":0.941,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-69","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1493.0,"score":0.4,"rank":17.86,"hrank":11.71,"imp":14.044,"train_time":3855.77,"predict_time":24.914,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-55","imputed":false},{"position":9,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1478.0,"score":0.378,"rank":18.81,"hrank":13.31,"imp":14.287,"train_time":3351.52,"predict_time":1.558,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-53","imputed":false},{"position":10,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1444.0,"score":0.405,"rank":21.0,"hrank":7.83,"imp":14.549,"train_time":8061.74,"predict_time":364.097,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+94\/-72","imputed":false},{"position":11,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1432.0,"score":0.362,"rank":21.76,"hrank":10.71,"imp":15.307,"train_time":5.1,"predict_time":0.187,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-66","imputed":false},{"position":12,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1420.0,"score":0.344,"rank":22.58,"hrank":13.19,"imp":15.165,"train_time":3855.77,"predict_time":0.954,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-77","imputed":false},{"position":13,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1413.0,"score":0.366,"rank":23.1,"hrank":7.76,"imp":14.276,"train_time":34.03,"predict_time":7.205,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-84","imputed":false},{"position":14,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1404.0,"score":0.358,"rank":23.72,"hrank":11.52,"imp":15.538,"train_time":8061.74,"predict_time":39.456,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-72","imputed":false},{"position":15,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1391.0,"score":0.317,"rank":24.64,"hrank":12.19,"imp":15.862,"train_time":2564.23,"predict_time":3.453,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-58","imputed":false},{"position":16,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1380.0,"score":0.263,"rank":25.46,"hrank":16.17,"imp":16.309,"train_time":22680.19,"predict_time":529.126,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-49","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1374.0,"score":0.222,"rank":25.86,"hrank":20.8,"imp":16.91,"train_time":833.09,"predict_time":2.182,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-44","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1369.0,"score":0.248,"rank":26.19,"hrank":15.11,"imp":15.784,"train_time":1827.95,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-50","imputed":false},{"position":19,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1363.0,"score":0.222,"rank":26.69,"hrank":20.69,"imp":16.24,"train_time":1827.95,"predict_time":0.391,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-56","imputed":false},{"position":20,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1358.0,"score":0.276,"rank":27.01,"hrank":15.37,"imp":16.486,"train_time":2564.23,"predict_time":0.453,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-64","imputed":false},{"position":21,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1350.0,"score":0.257,"rank":27.6,"hrank":17.19,"imp":17.212,"train_time":2.54,"predict_time":0.058,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-55","imputed":false},{"position":22,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1324.0,"score":0.181,"rank":29.62,"hrank":22.59,"imp":17.822,"train_time":886.81,"predict_time":1.98,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-44","imputed":false},{"position":23,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1323.0,"score":0.234,"rank":29.67,"hrank":17.59,"imp":17.514,"train_time":1346.82,"predict_time":2.149,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-59","imputed":false},{"position":24,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1308.0,"score":0.169,"rank":30.81,"hrank":25.23,"imp":17.899,"train_time":833.09,"predict_time":0.435,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-50","imputed":false},{"position":25,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1308.0,"score":0.17,"rank":30.81,"hrank":15.58,"imp":17.941,"train_time":886.81,"predict_time":0.318,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-54","imputed":false},{"position":26,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1308.0,"score":0.227,"rank":30.83,"hrank":13.91,"imp":18.607,"train_time":6343.61,"predict_time":8.545,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-70","imputed":false},{"position":27,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1307.0,"score":0.15,"rank":30.92,"hrank":26.5,"imp":18.16,"train_time":9.04,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-40","imputed":false},{"position":28,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1301.0,"score":0.251,"rank":31.33,"hrank":16.12,"imp":17.933,"train_time":51.79,"predict_time":39.921,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-82","imputed":false},{"position":29,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1296.0,"score":0.199,"rank":31.69,"hrank":17.84,"imp":18.08,"train_time":6343.61,"predict_time":0.501,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-62","imputed":false},{"position":30,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1292.0,"score":0.166,"rank":32.07,"hrank":25.12,"imp":19.278,"train_time":1180.0,"predict_time":0.457,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-58","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1286.0,"score":0.17,"rank":32.5,"hrank":25.31,"imp":17.797,"train_time":22680.19,"predict_time":126.301,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-59","imputed":false},{"position":32,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1286.0,"score":0.243,"rank":32.53,"hrank":15.52,"imp":19.458,"train_time":16.54,"predict_time":1.254,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-74","imputed":false},{"position":33,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1277.0,"score":0.184,"rank":33.21,"hrank":20.44,"imp":18.555,"train_time":11.22,"predict_time":0.393,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-62","imputed":false},{"position":34,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1270.0,"score":0.178,"rank":33.75,"hrank":15.66,"imp":18.446,"train_time":1346.82,"predict_time":0.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-60","imputed":false},{"position":35,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1266.0,"score":0.128,"rank":34.03,"hrank":22.89,"imp":19.484,"train_time":2241.86,"predict_time":0.152,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-58","imputed":false},{"position":36,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1254.0,"score":0.141,"rank":35.0,"hrank":21.92,"imp":19.792,"train_time":1180.0,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-73","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1253.0,"score":0.133,"rank":35.06,"hrank":25.2,"imp":18.947,"train_time":16.61,"predict_time":4.299,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-66","imputed":false},{"position":38,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1237.0,"score":0.132,"rank":36.28,"hrank":29.5,"imp":19.118,"train_time":3694.49,"predict_time":2.137,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-59","imputed":false},{"position":39,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1226.0,"score":0.106,"rank":37.14,"hrank":24.93,"imp":20.175,"train_time":2241.86,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-58","imputed":false},{"position":40,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1207.0,"score":0.117,"rank":38.64,"hrank":15.78,"imp":22.279,"train_time":14.91,"predict_time":0.323,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-79","imputed":false},{"position":41,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1201.0,"score":0.104,"rank":39.1,"hrank":17.6,"imp":20.769,"train_time":7.6,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-65","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1201.0,"score":0.089,"rank":39.11,"hrank":32.82,"imp":21.425,"train_time":5.52,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-70","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1195.0,"score":0.116,"rank":39.53,"hrank":26.67,"imp":21.437,"train_time":841.6,"predict_time":1.393,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-76","imputed":false},{"position":44,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1191.0,"score":0.122,"rank":39.85,"hrank":29.09,"imp":20.673,"train_time":3694.49,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-66","imputed":false},{"position":45,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1175.0,"score":0.105,"rank":41.07,"hrank":30.38,"imp":21.805,"train_time":1666.93,"predict_time":7.822,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-88","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1168.0,"score":0.09,"rank":41.62,"hrank":27.33,"imp":22.49,"train_time":841.6,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-83","imputed":false},{"position":47,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1138.0,"score":0.069,"rank":43.85,"hrank":36.67,"imp":21.673,"train_time":3.63,"predict_time":0.218,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-63","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1136.0,"score":0.087,"rank":43.97,"hrank":35.38,"imp":23.071,"train_time":575.62,"predict_time":1.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-80","imputed":false},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1129.0,"score":0.042,"rank":44.47,"hrank":38.95,"imp":21.724,"train_time":3.1,"predict_time":0.207,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-61","imputed":false},{"position":50,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1124.0,"score":0.078,"rank":44.86,"hrank":31.42,"imp":22.967,"train_time":1666.93,"predict_time":0.561,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-84","imputed":false},{"position":51,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1118.0,"score":0.149,"rank":45.31,"hrank":16.5,"imp":25.514,"train_time":46972.72,"predict_time":5.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-102","imputed":false},{"position":52,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1118.0,"score":0.143,"rank":45.32,"hrank":17.27,"imp":25.681,"train_time":46972.72,"predict_time":22.64,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-103","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1093.0,"score":0.048,"rank":47.1,"hrank":41.03,"imp":24.003,"train_time":575.62,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-69","imputed":false},{"position":54,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1090.0,"score":0.076,"rank":47.26,"hrank":37.76,"imp":28.508,"train_time":282.83,"predict_time":0.509,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-84","imputed":false},{"position":55,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1061.0,"score":0.055,"rank":49.26,"hrank":27.98,"imp":24.936,"train_time":537.96,"predict_time":157.807,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-93","imputed":false},{"position":56,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1045.0,"score":0.04,"rank":50.33,"hrank":45.89,"imp":29.907,"train_time":282.83,"predict_time":0.162,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-76","imputed":false},{"position":57,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1037.0,"score":0.061,"rank":50.83,"hrank":41.02,"imp":26.218,"train_time":6.09,"predict_time":0.951,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-92","imputed":false},{"position":58,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1029.0,"score":0.037,"rank":51.33,"hrank":43.05,"imp":27.071,"train_time":4.37,"predict_time":0.629,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-94","imputed":false},{"position":59,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1018.0,"score":0.103,"rank":52.0,"hrank":16.56,"imp":31.172,"train_time":459.76,"predict_time":5.723,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-154","imputed":false},{"position":60,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1017.0,"score":0.024,"rank":52.11,"hrank":47.07,"imp":25.686,"train_time":12.68,"predict_time":0.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-95","imputed":false},{"position":61,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.017,"rank":53.12,"hrank":44.07,"imp":28.351,"train_time":0.51,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-87","imputed":false},{"position":62,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":982.0,"score":0.026,"rank":54.19,"hrank":50.5,"imp":26.753,"train_time":5.83,"predict_time":0.571,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-88","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":969.0,"score":0.031,"rank":54.9,"hrank":47.13,"imp":29.114,"train_time":116.33,"predict_time":1.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-94","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":958.0,"score":0.034,"rank":55.51,"hrank":31.25,"imp":34.252,"train_time":38.53,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-96","imputed":false},{"position":65,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":907.0,"score":0.028,"rank":58.12,"hrank":22.49,"imp":34.637,"train_time":520.74,"predict_time":0.491,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-100","imputed":false},{"position":66,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":885.0,"score":0.024,"rank":59.14,"hrank":33.07,"imp":35.11,"train_time":520.74,"predict_time":0.097,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-104","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":837.0,"score":0.01,"rank":61.14,"hrank":56.91,"imp":37.819,"train_time":2.59,"predict_time":0.129,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-136","imputed":false},{"position":68,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":823.0,"score":0.018,"rank":61.67,"hrank":57.95,"imp":34.859,"train_time":116.33,"predict_time":0.075,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-110","imputed":false},{"position":69,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":632.0,"score":0.0,"rank":66.86,"hrank":66.36,"imp":46.713,"train_time":0.37,"predict_time":0.035,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-168","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_small/pareto_front_explorer.html
index 7a0735d13fdfef44d16617bdd564a2dc1d5d8a5a..7229ddf69d63f45f9392d62f38a8694e2a2e7d21 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0408851968,"imp":18.1598157541,"elo":1306.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0482592414,"imp":15.7840827203,"elo":1369.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3907846602,"imp":16.2397079362,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0367584655,"imp":21.424890815,"elo":1200.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0344803412,"imp":19.792045152,"elo":1253.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4568895686,"imp":19.2776270627,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0170112085,"imp":20.7692962893,"elo":1201.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0168225275,"imp":20.1751043148,"elo":1226.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1523084536,"imp":19.4843376316,"elo":1266.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6288369946,"imp":27.0712484973,"elo":1029.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1925559683,"imp":22.490041374,"elo":1167.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3932122077,"imp":21.4368548346,"elo":1195.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5709841633,"imp":26.752776165,"elo":981.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5609692074,"imp":22.9674407712,"elo":1124.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.821672488,"imp":21.8053721954,"elo":1175.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2175696189,"imp":21.6727603085,"elo":1137.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.4350380217,"imp":17.8988831243,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.1817652315,"imp":16.9102002974,"elo":1374.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.2046047302,"imp":14.2756174101,"elo":1412.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3225041576,"imp":22.2791708897,"elo":1206.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5011277894,"imp":18.0799932734,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.5454388038,"imp":18.6067686551,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0412162752,"imp":34.251704126,"elo":958.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1615795606,"imp":29.9067803097,"elo":1044.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5087521152,"imp":28.5084725674,"elo":1090.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852899147,"imp":28.3508343664,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468843405,"imp":24.0034743231,"elo":1092.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"imp":23.0705212169,"elo":1136.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.2987548984,"imp":18.9466165278,"elo":1253.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.954021458,"imp":15.1649725442,"elo":1420.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":24.9138452893,"imp":14.0436901188,"elo":1493.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5863334021,"imp":11.6105591883,"elo":1577.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9412054434,"imp":11.8315857362,"elo":1548.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9558346011,"imp":9.6526030284,"elo":1606.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2543929634,"imp":19.458028729,"elo":1285.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"imp":17.9329410069,"elo":1301.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.456011971,"imp":15.537985585,"elo":1403.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":364.0974768275,"imp":14.5485759308,"elo":1443.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1868636356,"imp":15.3069746412,"elo":1432.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.3427919745,"imp":4.375938039,"elo":1783.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1502327458,"imp":11.4383260201,"elo":1555.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.3927969172,"imp":18.5551020688,"elo":1276.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4533957018,"imp":16.4859595294,"elo":1358.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.4526368273,"imp":15.8624516356,"elo":1390.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5551544404,"imp":10.9853601614,"elo":1607.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4108441184,"imp":9.8851821516,"elo":1633.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"imp":31.1720436965,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3103758735,"imp":25.5143813536,"elo":1117.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.63963438,"imp":25.6814958875,"elo":1117.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0581109952,"imp":17.2120533144,"elo":1350.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1547654046,"imp":25.6856734728,"elo":1016.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305659975,"imp":20.6734944304,"elo":1191.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1373326275,"imp":19.1184243594,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2070918304,"imp":21.7239699331,"elo":1129.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3177561745,"imp":17.9413089793,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.980191462,"imp":17.822257655,"elo":1323.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":157.807153647,"imp":24.9360272142,"elo":1061.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":126.3012190283,"imp":17.7969348154,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":529.125864783,"imp":16.3092010249,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9508145682,"imp":26.2180856463,"elo":1037.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1034549609,"imp":18.4457888884,"elo":1269.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.1486246717,"imp":17.5137999397,"elo":1322.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0408851968,"x_train":9.0359136176,"imp":18.1598157541,"elo":1306.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0482592414,"x_train":1827.947236699,"imp":15.7840827203,"elo":1369.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3907846602,"x_train":1827.947236699,"imp":16.2397079362,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0367584655,"x_train":5.5160770154,"imp":21.424890815,"elo":1200.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0344803412,"x_train":1179.9999181447,"imp":19.792045152,"elo":1253.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4568895686,"x_train":1179.9999181447,"imp":19.2776270627,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0170112085,"x_train":7.5965245684,"imp":20.7692962893,"elo":1201.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0168225275,"x_train":2241.8559420912,"imp":20.1751043148,"elo":1226.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1523084536,"x_train":2241.8559420912,"imp":19.4843376316,"elo":1266.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6288369946,"x_train":4.3684404027,"imp":27.0712484973,"elo":1029.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1925559683,"x_train":841.5972483589,"imp":22.490041374,"elo":1167.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3932122077,"x_train":841.5972483589,"imp":21.4368548346,"elo":1195.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5709841633,"x_train":5.8349009345,"imp":26.752776165,"elo":981.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5609692074,"x_train":1666.9308833948,"imp":22.9674407712,"elo":1124.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.821672488,"x_train":1666.9308833948,"imp":21.8053721954,"elo":1175.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2175696189,"x_train":3.6257108294,"imp":21.6727603085,"elo":1137.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.4350380217,"x_train":833.0924159486,"imp":17.8988831243,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.1817652315,"x_train":833.0924159486,"imp":16.9102002974,"elo":1374.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.2046047302,"x_train":34.0310873482,"imp":14.2756174101,"elo":1412.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3225041576,"x_train":14.9096333145,"imp":22.2791708897,"elo":1206.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5011277894,"x_train":6343.6148094754,"imp":18.0799932734,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.5454388038,"x_train":6343.6148094754,"imp":18.6067686551,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0412162752,"x_train":38.5275966029,"imp":34.251704126,"elo":958.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1615795606,"x_train":282.8267356163,"imp":29.9067803097,"elo":1044.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5087521152,"x_train":282.8267356163,"imp":28.5084725674,"elo":1090.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852899147,"x_train":0.5076178694,"imp":28.3508343664,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468843405,"x_train":575.6224152834,"imp":24.0034743231,"elo":1092.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"x_train":575.6224152834,"imp":23.0705212169,"elo":1136.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.2987548984,"x_train":16.6080899164,"imp":18.9466165278,"elo":1253.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.954021458,"x_train":3855.7666751446,"imp":15.1649725442,"elo":1420.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":24.9138452893,"x_train":3855.7666751446,"imp":14.0436901188,"elo":1493.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5863334021,"x_train":6.5779030898,"imp":11.6105591883,"elo":1577.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9412054434,"x_train":2320.5586596122,"imp":11.8315857362,"elo":1548.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9558346011,"x_train":2320.5586596122,"imp":9.6526030284,"elo":1606.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2543929634,"x_train":16.5391218381,"imp":19.458028729,"elo":1285.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"x_train":51.7864862661,"imp":17.9329410069,"elo":1301.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.456011971,"x_train":8061.743825954,"imp":15.537985585,"elo":1403.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":364.0974768275,"x_train":8061.743825954,"imp":14.5485759308,"elo":1443.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1868636356,"x_train":5.1020093375,"imp":15.3069746412,"elo":1432.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.3427919745,"x_train":45.3524743275,"imp":4.375938039,"elo":1783.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1502327458,"x_train":3.8948553652,"imp":11.4383260201,"elo":1555.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.3927969172,"x_train":11.2200428783,"imp":18.5551020688,"elo":1276.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4533957018,"x_train":2564.2291668214,"imp":16.4859595294,"elo":1358.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.4526368273,"x_train":2564.2291668214,"imp":15.8624516356,"elo":1390.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5551544404,"x_train":7.0738556742,"imp":10.9853601614,"elo":1607.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4108441184,"x_train":8.5714229456,"imp":9.8851821516,"elo":1633.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"x_train":459.757659874,"imp":31.1720436965,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3103758735,"x_train":46972.7202570992,"imp":25.5143813536,"elo":1117.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.63963438,"x_train":46972.7202570992,"imp":25.6814958875,"elo":1117.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0581109952,"x_train":2.5372143019,"imp":17.2120533144,"elo":1350.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1547654046,"x_train":12.6783036886,"imp":25.6856734728,"elo":1016.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305659975,"x_train":3694.4938905556,"imp":20.6734944304,"elo":1191.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1373326275,"x_train":3694.4938905556,"imp":19.1184243594,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2070918304,"x_train":3.0974213034,"imp":21.7239699331,"elo":1129.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3177561745,"x_train":886.8089747599,"imp":17.9413089793,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.980191462,"x_train":886.8089747599,"imp":17.822257655,"elo":1323.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":157.807153647,"x_train":537.959331743,"imp":24.9360272142,"elo":1061.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":126.3012190283,"x_train":22680.1927917466,"imp":17.7969348154,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":529.125864783,"x_train":22680.1927917466,"imp":16.3092010249,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9508145682,"x_train":6.0878006956,"imp":26.2180856463,"elo":1037.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1034549609,"x_train":1346.8150776785,"imp":18.4457888884,"elo":1269.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.1486246717,"x_train":1346.8150776785,"imp":17.5137999397,"elo":1322.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_small/tuning_trajectories_explorer.html
index 097a1520a2f6c03a0ba3508c1073736f537d689d..655a548b31711176bbece11d01dcfe0d9db77802 100644
--- a/data/imputation_no/splits_lite/tasks_all/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":9.0359136176,"imp":18.1709499021,"elo":1317.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":21.9992777967,"imp":17.4079605277,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":53.8141065802,"imp":16.9296786176,"elo":1367.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":96.9989733937,"imp":16.6802678821,"elo":1376.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":231.1646909722,"imp":16.4977956286,"elo":1380.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":468.6200877614,"imp":16.4869561786,"elo":1376.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":950.6643741645,"imp":16.3265529215,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1384.6574015383,"imp":16.3191095146,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1827.947236699,"imp":16.2407062058,"elo":1374.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.5160770154,"imp":21.4352469807,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.4571846403,"imp":20.7449137964,"elo":1241.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":28.0250922704,"imp":20.2600581266,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":57.1334404694,"imp":20.0258050381,"elo":1279.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":145.0536005577,"imp":19.6474744842,"elo":1292.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":292.2184372302,"imp":19.461315603,"elo":1301.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":586.8028092523,"imp":19.3534773518,"elo":1304.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":881.2145061251,"imp":19.2842021823,"elo":1307.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1179.9999181447,"imp":19.2841521423,"elo":1306.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.5965245684,"imp":20.7795874297,"elo":1205.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":20.5778802427,"imp":20.457075271,"elo":1221.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":58.0053529444,"imp":20.113112532,"elo":1247.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":114.2580939201,"imp":19.994091034,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":281.7059626239,"imp":19.857254293,"elo":1262.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":567.1802440047,"imp":19.7052982899,"elo":1268.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1120.6313986655,"imp":19.5603643689,"elo":1276.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1665.6646846813,"imp":19.5233266419,"elo":1279.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2241.8559420912,"imp":19.4875627693,"elo":1280.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.3684404027,"imp":27.0820406907,"elo":1036.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":8.974442795,"imp":25.5045314769,"elo":1063.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":23.2753987371,"imp":23.2180618753,"elo":1120.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":45.9919048085,"imp":22.571581644,"elo":1150.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":110.3208397341,"imp":21.879826027,"elo":1181.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":212.5071471057,"imp":21.7034760359,"elo":1189.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":419.021686722,"imp":21.6055095798,"elo":1194.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":627.2476408835,"imp":21.4820663738,"elo":1203.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":841.5972483589,"imp":21.4879203579,"elo":1203.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.8349009345,"imp":26.7619385366,"elo":983.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.4449316294,"imp":25.9879759579,"elo":1026.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":42.2187698316,"imp":24.2269789092,"elo":1092.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":78.730916991,"imp":23.4785910053,"elo":1121.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":201.7091267248,"imp":22.561204572,"elo":1146.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":410.0601179712,"imp":22.1441125956,"elo":1162.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":823.1441900238,"imp":21.9521202119,"elo":1170.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1238.4424229224,"imp":21.8174081051,"elo":1174.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1666.9308833948,"imp":21.8258023014,"elo":1183.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":3.6257108294,"imp":21.6832764976,"elo":1137.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":8.2503296642,"imp":19.5635982326,"elo":1241.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":20.5596206802,"imp":18.2220339927,"elo":1320.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":40.5867329142,"imp":17.6929487908,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":103.1262709089,"imp":17.2706857739,"elo":1368.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":206.3615235739,"imp":17.1628068788,"elo":1378.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":413.6924902571,"imp":17.0351448122,"elo":1387.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":620.0408191726,"imp":16.9324933732,"elo":1390.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":833.0924159486,"imp":16.9180558896,"elo":1395.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":34.0310873482,"imp":14.4565747048,"elo":1424.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.9096333145,"imp":22.2911610169,"elo":1217.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":51.4995677932,"imp":20.927452174,"elo":1237.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":153.1634884827,"imp":18.7243381432,"elo":1315.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":306.0472940151,"imp":18.2154639665,"elo":1340.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":758.7881779295,"imp":17.7328079336,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1565.6822074431,"imp":17.9554609783,"elo":1344.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3151.1944700481,"imp":18.4120760063,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4699.0006066566,"imp":18.5670742477,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6343.6148094754,"imp":18.6311914981,"elo":1322.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":38.5275966029,"imp":34.2922065488,"elo":964.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":93.0126953024,"imp":31.3035327176,"elo":1025.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":235.686098342,"imp":28.6014085621,"elo":1091.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":282.8267356163,"imp":28.5058541196,"elo":1095.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.5076178694,"imp":28.3740828462,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.3146920986,"imp":26.0340859147,"elo":1039.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":12.5906114909,"imp":24.5891292368,"elo":1078.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":27.1833109568,"imp":23.7898876261,"elo":1098.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":68.9897633789,"imp":23.1901511434,"elo":1129.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":140.3776559984,"imp":23.0714063689,"elo":1138.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":283.2223248418,"imp":23.0922289227,"elo":1141.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":428.3020784862,"imp":23.1110248033,"elo":1141.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":575.6224152834,"imp":23.0813863575,"elo":1141.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":16.6080899164,"imp":18.9574964838,"elo":1259.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":35.8566583253,"imp":16.688810131,"elo":1357.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":91.663010689,"imp":15.0851678314,"elo":1444.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":189.0044514953,"imp":14.5743679731,"elo":1467.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":476.9907049476,"imp":14.0989838315,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":961.4088522866,"imp":13.8742900151,"elo":1512.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1920.3182480183,"imp":13.8786590246,"elo":1516.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2879.4073323842,"imp":13.9948382711,"elo":1519.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3855.7666751446,"imp":14.0582125967,"elo":1519.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.5779030898,"imp":11.6225272216,"elo":1596.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":19.6925140518,"imp":11.5160477155,"elo":1593.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":56.5542449974,"imp":11.2383986435,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":113.5361347503,"imp":11.190206792,"elo":1613.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":293.4788372615,"imp":11.0419899412,"elo":1613.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":577.5893123253,"imp":11.0067499796,"elo":1608.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1152.3312972563,"imp":10.5599557244,"elo":1619.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1735.6880999961,"imp":10.2962626295,"elo":1623.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2320.5586596122,"imp":9.6917003164,"elo":1623.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":16.5391218381,"imp":19.4687884247,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.7864862661,"imp":17.9438661146,"elo":1313.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":90.9755007226,"imp":17.1127944859,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":208.5282027548,"imp":16.2845384487,"elo":1391.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":397.9314678674,"imp":15.9169552487,"elo":1408.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":999.4790170647,"imp":15.3409421189,"elo":1424.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":2004.3956749143,"imp":14.9328700554,"elo":1440.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":4010.9044638578,"imp":14.7633698873,"elo":1449.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":6016.93821747,"imp":14.6825925157,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":8061.743825954,"imp":14.5870031619,"elo":1462.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":5.1020093375,"imp":15.3187741445,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":45.3524743275,"imp":4.3881797852,"elo":1792.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":3.8948553652,"imp":11.4502776545,"elo":1572.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":11.2200428783,"imp":18.5663400674,"elo":1288.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":23.7397758074,"imp":18.1829381029,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":61.4524391665,"imp":17.6443922878,"elo":1326.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":127.9575000974,"imp":17.1163445933,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":319.7265066945,"imp":16.7014235965,"elo":1380.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":634.7918157013,"imp":16.386759053,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1265.5050030954,"imp":16.0861016875,"elo":1409.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1910.2960689809,"imp":15.9790882784,"elo":1414.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2564.2291668214,"imp":15.8742406089,"elo":1412.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":7.0738556742,"imp":10.9976986979,"elo":1632.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":8.5714229456,"imp":9.8994219284,"elo":1653.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":459.757659874,"imp":31.1801376462,"elo":1025.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":859.3423899592,"imp":31.0218773009,"elo":1033.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1648.7767518666,"imp":30.5907491748,"elo":1044.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2593.696023839,"imp":29.7315170595,"elo":1059.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5812.689260478,"imp":28.1397204055,"elo":1085.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11696.3454938888,"imp":26.8409836103,"elo":1103.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23869.0757052958,"imp":26.3027123172,"elo":1114.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":35350.6242030506,"imp":26.026074219,"elo":1123.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":46972.7202570992,"imp":25.6994088842,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":2.5372143019,"imp":17.2229017971,"elo":1368.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":12.6783036886,"imp":25.6904610968,"elo":1022.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":27.894998506,"imp":23.2727434877,"elo":1091.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":88.2967448357,"imp":21.2956687264,"elo":1151.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":185.1308765061,"imp":20.5877663488,"elo":1189.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":457.9055405764,"imp":19.9901563031,"elo":1219.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":916.161299444,"imp":19.6408408224,"elo":1234.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1819.6211273997,"imp":19.4368858471,"elo":1234.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2759.4404936339,"imp":19.2402571312,"elo":1245.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3694.4938905556,"imp":19.1268688855,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":3.0974213034,"imp":21.7342751874,"elo":1131.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":8.3851286101,"imp":19.4911411837,"elo":1240.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":21.5559379962,"imp":18.8645559473,"elo":1280.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":42.3770236352,"imp":18.4817855565,"elo":1304.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":109.5235780735,"imp":18.1900747132,"elo":1319.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":220.0167184889,"imp":18.082021946,"elo":1328.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":441.8368524844,"imp":17.9679262855,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":661.3695455545,"imp":17.8611808931,"elo":1342.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":886.8089747599,"imp":17.8164503288,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":537.959331743,"imp":24.9467163395,"elo":1071.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1311.0677158659,"imp":21.7012080917,"elo":1161.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":3930.4572134166,"imp":18.4131561124,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":8786.6645332738,"imp":17.3258550591,"elo":1332.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":21604.4108540479,"imp":16.3402151825,"elo":1394.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":22680.1927917466,"imp":16.3370143835,"elo":1393.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":6.0878006956,"imp":26.2268720693,"elo":1047.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":12.2392299129,"imp":23.7089148796,"elo":1109.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":32.0134798368,"imp":20.3939669332,"elo":1224.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":65.1287013348,"imp":18.8800997766,"elo":1286.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":166.8282289738,"imp":18.0743504466,"elo":1319.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":335.4693508184,"imp":17.746747607,"elo":1336.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":669.3015790718,"imp":17.5753108714,"elo":1338.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1004.6327529567,"imp":17.5483327722,"elo":1340.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1346.8150776785,"imp":17.5322488966,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0408851968,"x_train":9.0359136176,"imp":18.1709499021,"elo":1317.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0905042501,"x_train":21.9992777967,"imp":17.4079605277,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.172973571,"x_train":53.8141065802,"imp":16.9296786176,"elo":1367.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2079530855,"x_train":96.9989733937,"imp":16.6802678821,"elo":1376.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2459997699,"x_train":231.1646909722,"imp":16.4977956286,"elo":1380.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.291674196,"x_train":468.6200877614,"imp":16.4869561786,"elo":1376.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.343571202,"x_train":950.6643741645,"imp":16.3265529215,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3883571805,"x_train":1384.6574015383,"imp":16.3191095146,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3907846602,"x_train":1827.947236699,"imp":16.2407062058,"elo":1374.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0367584655,"x_train":5.5160770154,"imp":21.4352469807,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0787805569,"x_train":11.4571846403,"imp":20.7449137964,"elo":1241.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.152959706,"x_train":28.0250922704,"imp":20.2600581266,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.221696575,"x_train":57.1334404694,"imp":20.0258050381,"elo":1279.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2819310273,"x_train":145.0536005577,"imp":19.6474744842,"elo":1292.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.339421727,"x_train":292.2184372302,"imp":19.461315603,"elo":1301.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3845866458,"x_train":586.8028092523,"imp":19.3534773518,"elo":1304.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3951493501,"x_train":881.2145061251,"imp":19.2842021823,"elo":1307.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4568895686,"x_train":1179.9999181447,"imp":19.2841521423,"elo":1306.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0170112085,"x_train":7.5965245684,"imp":20.7795874297,"elo":1205.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0331881177,"x_train":20.5778802427,"imp":20.457075271,"elo":1221.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0590854757,"x_train":58.0053529444,"imp":20.113112532,"elo":1247.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0783042685,"x_train":114.2580939201,"imp":19.994091034,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1039353704,"x_train":281.7059626239,"imp":19.857254293,"elo":1262.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1218997247,"x_train":567.1802440047,"imp":19.7052982899,"elo":1268.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1354779968,"x_train":1120.6313986655,"imp":19.5603643689,"elo":1276.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1434728711,"x_train":1665.6646846813,"imp":19.5233266419,"elo":1279.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1523084536,"x_train":2241.8559420912,"imp":19.4875627693,"elo":1280.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6288369946,"x_train":4.3684404027,"imp":27.0820406907,"elo":1036.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7303730135,"x_train":8.974442795,"imp":25.5045314769,"elo":1063.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6701745763,"x_train":23.2753987371,"imp":23.2180618753,"elo":1120.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7102008111,"x_train":45.9919048085,"imp":22.571581644,"elo":1150.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9526662594,"x_train":110.3208397341,"imp":21.879826027,"elo":1181.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2214235524,"x_train":212.5071471057,"imp":21.7034760359,"elo":1189.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5521012524,"x_train":419.021686722,"imp":21.6055095798,"elo":1194.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7473298834,"x_train":627.2476408835,"imp":21.4820663738,"elo":1203.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3932122077,"x_train":841.5972483589,"imp":21.4879203579,"elo":1203.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5709841633,"x_train":5.8349009345,"imp":26.7619385366,"elo":983.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4163112584,"x_train":14.4449316294,"imp":25.9879759579,"elo":1026.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7078527827,"x_train":42.2187698316,"imp":24.2269789092,"elo":1092.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.5424617722,"x_train":78.730916991,"imp":23.4785910053,"elo":1121.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.7771560016,"x_train":201.7091267248,"imp":22.561204572,"elo":1146.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.2313050635,"x_train":410.0601179712,"imp":22.1441125956,"elo":1162.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.1854207053,"x_train":823.1441900238,"imp":21.9521202119,"elo":1170.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.6836795815,"x_train":1238.4424229224,"imp":21.8174081051,"elo":1174.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.821672488,"x_train":1666.9308833948,"imp":21.8258023014,"elo":1183.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2175696189,"x_train":3.6257108294,"imp":21.6832764976,"elo":1137.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7244393627,"x_train":8.2503296642,"imp":19.5635982326,"elo":1241.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2312451116,"x_train":20.5596206802,"imp":18.2220339927,"elo":1320.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5807687941,"x_train":40.5867329142,"imp":17.6929487908,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.7582613878,"x_train":103.1262709089,"imp":17.2706857739,"elo":1368.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.9395731779,"x_train":206.3615235739,"imp":17.1628068788,"elo":1378.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.9939153323,"x_train":413.6924902571,"imp":17.0351448122,"elo":1387.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2268088052,"x_train":620.0408191726,"imp":16.9324933732,"elo":1390.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1817652315,"x_train":833.0924159486,"imp":16.9180558896,"elo":1395.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":7.2046047302,"x_train":34.0310873482,"imp":14.4565747048,"elo":1424.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3225041576,"x_train":14.9096333145,"imp":22.2911610169,"elo":1217.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8785929508,"x_train":51.4995677932,"imp":20.927452174,"elo":1237.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7997608403,"x_train":153.1634884827,"imp":18.7243381432,"elo":1315.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.832599148,"x_train":306.0472940151,"imp":18.2154639665,"elo":1340.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1547858756,"x_train":758.7881779295,"imp":17.7328079336,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.4651963907,"x_train":1565.6822074431,"imp":17.9554609783,"elo":1344.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.2034777482,"x_train":3151.1944700481,"imp":18.4120760063,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.1674130588,"x_train":4699.0006066566,"imp":18.5670742477,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.5454388038,"x_train":6343.6148094754,"imp":18.6311914981,"elo":1322.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0412162752,"x_train":38.5275966029,"imp":34.2922065488,"elo":964.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2867661168,"x_train":93.0126953024,"imp":31.3035327176,"elo":1025.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5061660782,"x_train":235.686098342,"imp":28.6014085621,"elo":1091.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5087521152,"x_train":282.8267356163,"imp":28.5058541196,"elo":1095.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852899147,"x_train":0.5076178694,"imp":28.3740828462,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2679638116,"x_train":3.3146920986,"imp":26.0340859147,"elo":1039.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4603970571,"x_train":12.5906114909,"imp":24.5891292368,"elo":1078.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6007649726,"x_train":27.1833109568,"imp":23.7898876261,"elo":1098.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7995923156,"x_train":68.9897633789,"imp":23.1901511434,"elo":1129.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9217934517,"x_train":140.3776559984,"imp":23.0714063689,"elo":1138.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0817312972,"x_train":283.2223248418,"imp":23.0922289227,"elo":1141.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2223316361,"x_train":428.3020784862,"imp":23.1110248033,"elo":1141.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"x_train":575.6224152834,"imp":23.0813863575,"elo":1141.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.2987548984,"x_train":16.6080899164,"imp":18.9574964838,"elo":1259.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.8477674709,"x_train":35.8566583253,"imp":16.688810131,"elo":1357.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.0672087974,"x_train":91.663010689,"imp":15.0851678314,"elo":1444.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.5254352945,"x_train":189.0044514953,"imp":14.5743679731,"elo":1467.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4341466661,"x_train":476.9907049476,"imp":14.0989838315,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":18.2190658863,"x_train":961.4088522866,"imp":13.8742900151,"elo":1512.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":20.0311215643,"x_train":1920.3182480183,"imp":13.8786590246,"elo":1516.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":22.0848749753,"x_train":2879.4073323842,"imp":13.9948382711,"elo":1519.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":24.9138452893,"x_train":3855.7666751446,"imp":14.0582125967,"elo":1519.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5863334021,"x_train":6.5779030898,"imp":11.6225272216,"elo":1596.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.32293962,"x_train":19.6925140518,"imp":11.5160477155,"elo":1593.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8419737629,"x_train":56.5542449974,"imp":11.2383986435,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.9894747323,"x_train":113.5361347503,"imp":11.190206792,"elo":1613.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.5217042751,"x_train":293.4788372615,"imp":11.0419899412,"elo":1613.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.2594554521,"x_train":577.5893123253,"imp":11.0067499796,"elo":1608.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.0497467794,"x_train":1152.3312972563,"imp":10.5599557244,"elo":1619.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8968974009,"x_train":1735.6880999961,"imp":10.2962626295,"elo":1623.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9558346011,"x_train":2320.5586596122,"imp":9.6917003164,"elo":1623.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2543929634,"x_train":16.5391218381,"imp":19.4687884247,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"x_train":51.7864862661,"imp":17.9438661146,"elo":1313.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":59.7687648763,"x_train":90.9755007226,"imp":17.1127944859,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":91.9802288244,"x_train":208.5282027548,"imp":16.2845384487,"elo":1391.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":132.6956612183,"x_train":397.9314678674,"imp":15.9169552487,"elo":1408.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":197.2683733908,"x_train":999.4790170647,"imp":15.3409421189,"elo":1424.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":241.2137556465,"x_train":2004.3956749143,"imp":14.9328700554,"elo":1440.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":310.7845512859,"x_train":4010.9044638578,"imp":14.7633698873,"elo":1449.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":340.4331355157,"x_train":6016.93821747,"imp":14.6825925157,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":347.4963020094,"x_train":8061.743825954,"imp":14.5870031619,"elo":1462.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1868636356,"x_train":5.1020093375,"imp":15.3187741445,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":5.3427919745,"x_train":45.3524743275,"imp":4.3881797852,"elo":1792.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1502327458,"x_train":3.8948553652,"imp":11.4502776545,"elo":1572.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.3927969172,"x_train":11.2200428783,"imp":18.5663400674,"elo":1288.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.8073226159,"x_train":23.7397758074,"imp":18.1829381029,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.4253994355,"x_train":61.4524391665,"imp":17.6443922878,"elo":1326.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.9037000676,"x_train":127.9575000974,"imp":17.1163445933,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.480365301,"x_train":319.7265066945,"imp":16.7014235965,"elo":1380.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":2.9323777701,"x_train":634.7918157013,"imp":16.386759053,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.1758660276,"x_train":1265.5050030954,"imp":16.0861016875,"elo":1409.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":3.2134702655,"x_train":1910.2960689809,"imp":15.9790882784,"elo":1414.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":3.4526368273,"x_train":2564.2291668214,"imp":15.8742406089,"elo":1412.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5551544404,"x_train":7.0738556742,"imp":10.9976986979,"elo":1632.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4108441184,"x_train":8.5714229456,"imp":9.8994219284,"elo":1653.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"x_train":459.757659874,"imp":31.1801376462,"elo":1025.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.7716558163,"x_train":859.3423899592,"imp":31.0218773009,"elo":1033.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.5636156792,"x_train":1648.7767518666,"imp":30.5907491748,"elo":1044.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.7112841968,"x_train":2593.696023839,"imp":29.7315170595,"elo":1059.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8983325102,"x_train":5812.689260478,"imp":28.1397204055,"elo":1085.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.5003951009,"x_train":11696.3454938888,"imp":26.8409836103,"elo":1103.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":24.9793991039,"x_train":23869.0757052958,"imp":26.3027123172,"elo":1114.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.6066199148,"x_train":35350.6242030506,"imp":26.026074219,"elo":1123.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.63963438,"x_train":46972.7202570992,"imp":25.6994088842,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0581109952,"x_train":2.5372143019,"imp":17.2229017971,"elo":1368.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1547654046,"x_train":12.6783036886,"imp":25.6904610968,"elo":1022.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3184069177,"x_train":27.894998506,"imp":23.2727434877,"elo":1091.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6772463403,"x_train":88.2967448357,"imp":21.2956687264,"elo":1151.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8527607509,"x_train":185.1308765061,"imp":20.5877663488,"elo":1189.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2375187431,"x_train":457.9055405764,"imp":19.9901563031,"elo":1219.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5433408305,"x_train":916.161299444,"imp":19.6408408224,"elo":1234.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.800349055,"x_train":1819.6211273997,"imp":19.4368858471,"elo":1234.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0481960908,"x_train":2759.4404936339,"imp":19.2402571312,"elo":1245.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1373326275,"x_train":3694.4938905556,"imp":19.1268688855,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2070918304,"x_train":3.0974213034,"imp":21.7342751874,"elo":1131.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7336842368,"x_train":8.3851286101,"imp":19.4911411837,"elo":1240.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.3748576592,"x_train":21.5559379962,"imp":18.8645559473,"elo":1280.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.5232150506,"x_train":42.3770236352,"imp":18.4817855565,"elo":1304.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8858828782,"x_train":109.5235780735,"imp":18.1900747132,"elo":1319.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.079369511,"x_train":220.0167184889,"imp":18.082021946,"elo":1328.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3062309572,"x_train":441.8368524844,"imp":17.9679262855,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0953020923,"x_train":661.3695455545,"imp":17.8611808931,"elo":1342.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.854250542,"x_train":886.8089747599,"imp":17.8164503288,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":157.807153647,"x_train":537.959331743,"imp":24.9467163395,"elo":1071.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":288.0546364493,"x_train":1311.0677158659,"imp":21.7012080917,"elo":1161.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":473.6713928263,"x_train":3930.4572134166,"imp":18.4131561124,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":620.719533977,"x_train":8786.6645332738,"imp":17.3258550591,"elo":1332.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":506.099831026,"x_train":21604.4108540479,"imp":16.3402151825,"elo":1394.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":529.125864783,"x_train":22680.1927917466,"imp":16.3370143835,"elo":1393.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9508145682,"x_train":6.0878006956,"imp":26.2268720693,"elo":1047.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.3492504189,"x_train":12.2392299129,"imp":23.7089148796,"elo":1109.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.629173245,"x_train":32.0134798368,"imp":20.3939669332,"elo":1224.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.6347035029,"x_train":65.1287013348,"imp":18.8800997766,"elo":1286.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.9845916387,"x_train":166.8282289738,"imp":18.0743504466,"elo":1319.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2321888977,"x_train":335.4693508184,"imp":17.746747607,"elo":1336.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.3191528609,"x_train":669.3015790718,"imp":17.5753108714,"elo":1338.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.2084021337,"x_train":1004.6327529567,"imp":17.5483327722,"elo":1340.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.1486246717,"x_train":1346.8150776785,"imp":17.5322488966,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_small/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_all/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..1b4ffdc6754f97a96516ee62109637b75ce73d51
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_all/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabDPT (tuned + ensembled)", "TabDPT-Turbo (default)", "RealMLP (tuned)", "LimiX (default)", "TabDPT (tuned)", "TabM (tuned + ensembled)", "iLTM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "CatBoost (tuned)", "CatBoost (tuned + ensembled)", "TabM (tuned)", "TabSwift (default)", "XGBoost (tuned + ensembled)", "xRFM (tuned + ensembled)", "LightGBM (tuned)", "XGBoost (tuned)", "ModernNCA (tuned + ensembled)", "CatBoost (default)", "TabDPT (default)", "ModernNCA (tuned)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "SAP-RPT-OSS (default)", "TabM (default)", "xRFM (tuned)", "EBM (tuned + ensembled)", "ChimeraBoost (tuned)", "RealMLP (default)", "TorchMLP (tuned + ensembled)", "EBM (tuned)", "ModernNCA (default)", "EBM (default)", "ChimeraBoost (default)", "ExtraTrees (tuned + ensembled)", "TorchMLP (tuned)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (tuned)", "LightGBM (default)", "RandomForest (tuned + ensembled)", "XGBoost (default)", "FastaiMLP (tuned)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "iLTM (default)", "PerpetualBooster (tuned)", "xRFM (default)", "ExtraTrees (default)", "TabSTAR (default)", "TorchMLP (default)", "RandomForest (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.8333333134651184, 0.7777777910232544, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 1.0, 0.9166666865348816, 1.0, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9444444179534912, 1.0, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 1.0], [0.1666666716337204, NaN, 0.5277777910232544, 0.5277777910232544, 0.5138888955116272, 0.5277777910232544, 0.5833333134651184, 0.6388888955116272, 0.75, 0.6388888955116272, 0.6944444179534912, 0.8055555820465088, 0.7777777910232544, 0.7222222089767456, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.9166666865348816, 0.7777777910232544, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8055555820465088, 0.8333333134651184, 0.9444444179534912, 0.8055555820465088, 0.8888888955116272, 0.9166666865348816, 0.9444444179534912, 0.8611111044883728, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.8888888955116272, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.8611111044883728, 0.8611111044883728, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 0.9722222089767456, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 1.0, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 1.0], [0.2222222238779068, 0.4722222089767456, NaN, 0.5833333134651184, 0.4722222089767456, 0.5277777910232544, 0.5, 0.6111111044883728, 0.8055555820465088, 0.7777777910232544, 0.7222222089767456, 0.6666666865348816, 0.8055555820465088, 0.7222222089767456, 0.7777777910232544, 0.8333333134651184, 0.8611111044883728, 0.8888888955116272, 0.8333333134651184, 0.8611111044883728, 0.8055555820465088, 0.8055555820465088, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.8055555820465088, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.9166666865348816, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.8611111044883728, 0.8333333134651184, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 1.0], [0.1666666716337204, 0.4722222089767456, 0.4166666567325592, NaN, 0.4722222089767456, 0.6388888955116272, 0.5, 0.5555555820465088, 0.7222222089767456, 0.6666666865348816, 0.6388888955116272, 0.7222222089767456, 0.8055555820465088, 0.7777777910232544, 0.6944444179534912, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8333333134651184, 0.8888888955116272, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272, 0.8888888955116272, 0.9444444179534912, 0.8888888955116272, 0.9166666865348816, 0.9444444179534912, 0.75, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.8888888955116272, 0.8333333134651184, 0.8333333134651184, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 1.0, 0.9722222089767456, 0.9444444179534912, 0.8333333134651184, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.1388888955116272, 0.4861111044883728, 0.5277777910232544, 0.5277777910232544, NaN, 0.6388888955116272, 0.6111111044883728, 0.6944444179534912, 0.75, 0.6666666865348816, 0.6388888955116272, 0.6666666865348816, 0.7777777910232544, 0.75, 0.6944444179534912, 0.7777777910232544, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.8333333134651184, 0.7777777910232544, 0.9166666865348816, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.9444444179534912, 0.75, 0.8888888955116272, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 1.0, 0.9166666865348816, 0.8333333134651184, 0.8333333134651184, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 1.0], [0.1666666716337204, 0.4722222089767456, 0.4722222089767456, 0.3611111044883728, 0.3611111044883728, NaN, 0.4722222089767456, 0.4305555522441864, 0.6666666865348816, 0.6388888955116272, 0.6388888955116272, 0.6944444179534912, 0.75, 0.7222222089767456, 0.6944444179534912, 0.7777777910232544, 0.7222222089767456, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8055555820465088, 0.8888888955116272, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.9027777910232544, 0.8055555820465088, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.8611111044883728, 0.8888888955116272, 0.9722222089767456, 0.8611111044883728, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 1.0, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 1.0, 0.9166666865348816, 0.9722222089767456, 0.8888888955116272, 0.9722222089767456, 0.9444444179534912, 1.0, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 0.9722222089767456], [0.1111111119389534, 0.4166666567325592, 0.5, 0.5, 0.3888888955116272, 0.5277777910232544, NaN, 0.5833333134651184, 0.6666666865348816, 0.75, 0.6944444179534912, 0.7222222089767456, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.75, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.75, 0.8055555820465088, 0.8611111044883728, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8888888955116272, 0.8055555820465088, 0.8888888955116272, 0.8888888955116272, 0.8611111044883728, 0.8333333134651184, 0.9166666865348816, 0.9444444179534912, 0.8888888955116272, 0.9166666865348816, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.8888888955116272, 0.8888888955116272, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 1.0], [0.1388888955116272, 0.3611111044883728, 0.3888888955116272, 0.4444444477558136, 0.3055555522441864, 0.5694444179534912, 0.4166666567325592, NaN, 0.6388888955116272, 0.6388888955116272, 0.5833333134651184, 0.6388888955116272, 0.75, 0.6111111044883728, 0.6388888955116272, 0.8055555820465088, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8333333134651184, 0.6944444179534912, 0.8888888955116272, 0.8055555820465088, 0.8888888955116272, 0.8611111044883728, 0.8055555820465088, 0.9166666865348816, 0.7222222089767456, 0.8611111044883728, 0.8055555820465088, 0.8888888955116272, 0.8055555820465088, 0.8333333134651184, 0.8888888955116272, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.8888888955116272, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.9444444179534912, 0.9166666865348816, 0.8611111044883728, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.8888888955116272, 0.8055555820465088, 0.7777777910232544, 0.8888888955116272, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 1.0], [0.1388888955116272, 0.25, 0.1944444477558136, 0.2777777910232544, 0.25, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, NaN, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.6388888955116272, 0.4722222089767456, 0.5, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728, 0.7222222089767456, 0.6388888955116272, 0.8611111044883728, 0.6944444179534912, 0.8333333134651184, 0.8611111044883728, 0.75, 0.8611111044883728, 0.6666666865348816, 0.75, 0.8055555820465088, 0.8333333134651184, 0.7222222089767456, 0.7777777910232544, 0.75, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.8055555820465088, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.8055555820465088, 0.9444444179534912, 0.8611111044883728, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456, 0.8333333134651184, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.0833333358168602, 0.3611111044883728, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, 0.25, 0.3611111044883728, 0.4444444477558136, NaN, 0.4166666567325592, 0.5277777910232544, 0.6111111044883728, 0.5555555820465088, 0.5, 0.6111111044883728, 0.6944444179534912, 0.7777777910232544, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.5833333134651184, 0.8055555820465088, 0.7222222089767456, 0.8333333134651184, 0.75, 0.6666666865348816, 0.7777777910232544, 0.6944444179534912, 0.6666666865348816, 0.8055555820465088, 0.75, 0.6944444179534912, 0.75, 0.75, 0.7222222089767456, 0.8333333134651184, 0.75, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.8611111044883728, 0.8333333134651184, 0.8055555820465088, 0.9166666865348816, 0.8888888955116272, 0.8055555820465088, 0.9166666865348816, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.75, 0.75, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.8888888955116272, 0.9722222089767456, 0.8333333134651184, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 1.0, 1.0], [0.0833333358168602, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3611111044883728, 0.3611111044883728, 0.3055555522441864, 0.4166666567325592, 0.5555555820465088, 0.5833333134651184, NaN, 0.4166666567325592, 0.6666666865348816, 0.6111111044883728, 0.7222222089767456, 0.5555555820465088, 0.7222222089767456, 0.6388888955116272, 0.6111111044883728, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.7222222089767456, 0.6666666865348816, 0.6944444179534912, 0.75, 0.6944444179534912, 0.6944444179534912, 0.75, 0.6666666865348816, 0.75, 0.8055555820465088, 0.6944444179534912, 0.6944444179534912, 0.75, 0.6944444179534912, 0.75, 0.6944444179534912, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.7222222089767456, 0.8055555820465088, 0.7222222089767456, 0.75, 0.8888888955116272, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.75, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.8333333134651184, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 1.0], [0.0833333358168602, 0.1944444477558136, 0.3333333432674408, 0.2777777910232544, 0.3333333432674408, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.4444444477558136, 0.4722222089767456, 0.5833333134651184, NaN, 0.6111111044883728, 0.5277777910232544, 0.5555555820465088, 0.5, 0.6111111044883728, 0.6388888955116272, 0.5555555820465088, 0.6111111044883728, 0.5555555820465088, 0.6805555820465088, 0.7222222089767456, 0.6944444179534912, 0.6944444179534912, 0.75, 0.5833333134651184, 0.6944444179534912, 0.6666666865348816, 0.6388888955116272, 0.7222222089767456, 0.75, 0.6944444179534912, 0.7222222089767456, 0.75, 0.6944444179534912, 0.7222222089767456, 0.6944444179534912, 0.75, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.75, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.7777777910232544, 0.75, 0.8611111044883728, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.9444444179534912, 0.8055555820465088, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456, 1.0], [0.1111111119389534, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.25, 0.3333333432674408, 0.25, 0.3611111044883728, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, NaN, 0.4027777910232544, 0.4444444477558136, 0.5277777910232544, 0.5555555820465088, 0.6388888955116272, 0.6111111044883728, 0.6111111044883728, 0.6388888955116272, 0.5555555820465088, 0.7222222089767456, 0.5833333134651184, 0.7222222089767456, 0.75, 0.6666666865348816, 0.7222222089767456, 0.5833333134651184, 0.7777777910232544, 0.6666666865348816, 0.7222222089767456, 0.6111111044883728, 0.7777777910232544, 0.6944444179534912, 0.8055555820465088, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.75, 0.8055555820465088, 0.75, 0.7777777910232544, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8333333134651184, 0.7777777910232544, 0.8055555820465088, 0.8888888955116272, 0.8888888955116272, 0.9444444179534912, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.8055555820465088, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9583333134651184, 0.9444444179534912, 0.9722222089767456, 1.0], [0.0833333358168602, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.25, 0.2777777910232544, 0.3055555522441864, 0.3888888955116272, 0.5277777910232544, 0.4444444477558136, 0.3888888955116272, 0.4722222089767456, 0.5972222089767456, NaN, 0.5, 0.5833333134651184, 0.625, 0.6388888955116272, 0.5277777910232544, 0.5833333134651184, 0.6388888955116272, 0.5277777910232544, 0.6388888955116272, 0.6388888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.6388888955116272, 0.6944444179534912, 0.75, 0.75, 0.75, 0.75, 0.7222222089767456, 0.8055555820465088, 0.7777777910232544, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.6944444179534912, 0.7222222089767456, 0.75, 0.8055555820465088, 0.7777777910232544, 0.8333333134651184, 0.8055555820465088, 0.8194444179534912, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 1.0], [0.02777777798473835, 0.1666666716337204, 0.2222222238779068, 0.3055555522441864, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.5, 0.5, 0.2777777910232544, 0.4444444477558136, 0.5555555820465088, 0.5, NaN, 0.5555555820465088, 0.6666666865348816, 0.6111111044883728, 0.5833333134651184, 0.6388888955116272, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6388888955116272, 0.6666666865348816, 0.6944444179534912, 0.6388888955116272, 0.6666666865348816, 0.7777777910232544, 0.6388888955116272, 0.7222222089767456, 0.75, 0.6111111044883728, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.75, 0.6944444179534912, 0.6944444179534912, 0.6388888955116272, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.75, 0.7222222089767456, 0.6944444179534912, 0.8333333134651184, 0.7222222089767456, 0.7222222089767456, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.75, 0.8055555820465088, 0.7777777910232544, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.9166666865348816, 0.8055555820465088, 0.8611111044883728, 0.8888888955116272, 0.9722222089767456, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.9444444179534912, 0.9722222089767456], [0.0833333358168602, 0.1666666716337204, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2777777910232544, 0.1944444477558136, 0.3333333432674408, 0.3888888955116272, 0.4444444477558136, 0.5, 0.4722222089767456, 0.4166666567325592, 0.4444444477558136, NaN, 0.5277777910232544, 0.5555555820465088, 0.4722222089767456, 0.5277777910232544, 0.6111111044883728, 0.5555555820465088, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.6944444179534912, 0.6388888955116272, 0.6666666865348816, 0.6111111044883728, 0.6388888955116272, 0.6388888955116272, 0.6388888955116272, 0.6388888955116272, 0.8333333134651184, 0.5833333134651184, 0.75, 0.8055555820465088, 0.75, 0.75, 0.7222222089767456, 0.7222222089767456, 0.7777777910232544, 0.8055555820465088, 0.6944444179534912, 0.8055555820465088, 0.7777777910232544, 0.6944444179534912, 0.8333333134651184, 0.7777777910232544, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.8888888955116272, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.9722222089767456, 0.8333333134651184, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 1.0, 1.0], [0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.2777777910232544, 0.1944444477558136, 0.2777777910232544, 0.2222222238779068, 0.3055555522441864, 0.2777777910232544, 0.3888888955116272, 0.4444444477558136, 0.375, 0.3333333432674408, 0.4722222089767456, NaN, 0.5555555820465088, 0.5555555820465088, 0.5, 0.5277777910232544, 0.5, 0.5833333134651184, 0.5555555820465088, 0.6388888955116272, 0.6666666865348816, 0.5277777910232544, 0.5555555820465088, 0.5277777910232544, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.6388888955116272, 0.6111111044883728, 0.6944444179534912, 0.6388888955116272, 0.7222222089767456, 0.6666666865348816, 0.7222222089767456, 0.6944444179534912, 0.8055555820465088, 0.75, 0.8055555820465088, 0.8333333134651184, 0.7777777910232544, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.8888888955116272, 0.8611111044883728, 0.8333333134651184, 0.7222222089767456, 0.7222222089767456, 0.8888888955116272, 0.8888888955116272, 0.8611111044883728, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.8055555820465088, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 1.0], [0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3611111044883728, 0.3611111044883728, 0.3611111044883728, 0.3611111044883728, 0.3888888955116272, 0.4444444477558136, 0.4444444477558136, NaN, 0.5, 0.5555555820465088, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.5833333134651184, 0.75, 0.6944444179534912, 0.5, 0.7222222089767456, 0.5277777910232544, 0.5833333134651184, 0.6666666865348816, 0.6111111044883728, 0.5833333134651184, 0.6666666865348816, 0.6666666865348816, 0.7222222089767456, 0.75, 0.6666666865348816, 0.7777777910232544, 0.75, 0.7777777910232544, 0.8055555820465088, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.7222222089767456, 0.9444444179534912, 0.8333333134651184, 0.9166666865348816, 0.8333333134651184, 0.7777777910232544, 0.75, 0.9166666865348816, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.8333333134651184, 0.8888888955116272, 0.8055555820465088, 0.8888888955116272, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912, 0.8888888955116272, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 1.0, 1.0], [0.0555555559694767, 0.2222222238779068, 0.1666666716337204, 0.1666666716337204, 0.1111111119389534, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3888888955116272, 0.4444444477558136, 0.3888888955116272, 0.4722222089767456, 0.4166666567325592, 0.5277777910232544, 0.4444444477558136, 0.5, NaN, 0.4444444477558136, 0.5555555820465088, 0.4722222089767456, 0.6388888955116272, 0.5277777910232544, 0.5833333134651184, 0.6666666865348816, 0.6111111044883728, 0.6944444179534912, 0.5277777910232544, 0.6111111044883728, 0.5277777910232544, 0.6944444179534912, 0.6111111044883728, 0.6111111044883728, 0.6388888955116272, 0.6944444179534912, 0.6388888955116272, 0.6944444179534912, 0.75, 0.7777777910232544, 0.7222222089767456, 0.7777777910232544, 0.75, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.6944444179534912, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8055555820465088, 0.75, 0.75, 0.8888888955116272, 0.8333333134651184, 0.8611111044883728, 0.9166666865348816, 0.8055555820465088, 0.8333333134651184, 0.7777777910232544, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.02777777798473835, 0.1666666716337204, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.1666666716337204, 0.2222222238779068, 0.1944444477558136, 0.1388888955116272, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.3888888955116272, 0.4166666567325592, 0.3611111044883728, 0.4722222089767456, 0.5, 0.4444444477558136, 0.5555555820465088, NaN, 0.4722222089767456, 0.4444444477558136, 0.6111111044883728, 0.4722222089767456, 0.6111111044883728, 0.6388888955116272, 0.6111111044883728, 0.6666666865348816, 0.5, 0.6111111044883728, 0.5555555820465088, 0.6944444179534912, 0.6666666865348816, 0.5555555820465088, 0.6111111044883728, 0.6666666865348816, 0.6388888955116272, 0.7222222089767456, 0.7222222089767456, 0.75, 0.6944444179534912, 0.7777777910232544, 0.8611111044883728, 0.75, 0.7777777910232544, 0.7222222089767456, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.75, 0.75, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.7777777910232544, 0.9166666865348816, 0.9444444179534912, 0.8888888955116272, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 1.0], [0.0555555559694767, 0.1111111119389534, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.25, 0.1666666716337204, 0.2777777910232544, 0.2777777910232544, 0.4444444477558136, 0.4444444477558136, 0.3611111044883728, 0.3611111044883728, 0.4444444477558136, 0.3888888955116272, 0.4722222089767456, 0.5277777910232544, 0.4444444477558136, 0.5277777910232544, NaN, 0.5277777910232544, 0.5833333134651184, 0.5277777910232544, 0.5555555820465088, 0.6388888955116272, 0.6111111044883728, 0.6666666865348816, 0.6111111044883728, 0.6111111044883728, 0.7361111044883728, 0.6388888955116272, 0.5555555820465088, 0.7222222089767456, 0.5555555820465088, 0.6666666865348816, 0.7222222089767456, 0.6388888955116272, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.75, 0.6944444179534912, 0.8055555820465088, 0.6944444179534912, 0.6944444179534912, 0.8333333134651184, 0.75, 0.8333333134651184, 0.8333333134651184, 0.75, 0.75, 0.7777777910232544, 0.8611111044883728, 0.8055555820465088, 0.8888888955116272, 0.8055555820465088, 0.7777777910232544, 0.8055555820465088, 0.9722222089767456, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456], [0.1111111119389534, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1666666716337204, 0.1944444477558136, 0.3055555522441864, 0.3611111044883728, 0.4166666567325592, 0.3333333432674408, 0.3194444477558136, 0.4444444477558136, 0.4722222089767456, 0.3333333432674408, 0.4444444477558136, 0.5, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.4722222089767456, NaN, 0.5555555820465088, 0.5833333134651184, 0.5277777910232544, 0.5833333134651184, 0.5277777910232544, 0.5555555820465088, 0.5277777910232544, 0.6388888955116272, 0.5, 0.6111111044883728, 0.6944444179534912, 0.5833333134651184, 0.6666666865348816, 0.6666666865348816, 0.5833333134651184, 0.6666666865348816, 0.5833333134651184, 0.6388888955116272, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.6944444179534912, 0.6944444179534912, 0.8333333134651184, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.75, 0.75, 0.75, 0.7777777910232544, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.9444444179534912, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456, 1.0], [0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.1944444477558136, 0.1111111119389534, 0.1388888955116272, 0.1944444477558136, 0.2777777910232544, 0.2777777910232544, 0.2777777910232544, 0.3611111044883728, 0.3333333432674408, 0.3611111044883728, 0.4166666567325592, 0.4444444477558136, 0.3611111044883728, 0.3888888955116272, 0.4166666567325592, 0.4444444477558136, NaN, 0.5, 0.5555555820465088, 0.5833333134651184, 0.4722222089767456, 0.5277777910232544, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.5833333134651184, 0.4722222089767456, 0.5833333134651184, 0.6111111044883728, 0.6388888955116272, 0.6388888955116272, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.75, 0.75, 0.6805555820465088, 0.75, 0.75, 0.9722222089767456, 0.8333333134651184, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8611111044883728, 0.8333333134651184, 0.7777777910232544, 0.8888888955116272, 0.8055555820465088, 0.8888888955116272, 0.9444444179534912, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 1.0], [0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.3055555522441864, 0.2777777910232544, 0.3333333432674408, 0.3055555522441864, 0.4166666567325592, 0.3611111044883728, 0.3611111044883728, 0.4166666567325592, 0.4444444477558136, 0.4166666567325592, 0.4722222089767456, 0.5277777910232544, 0.4722222089767456, 0.4166666567325592, 0.5, NaN, 0.4722222089767456, 0.4722222089767456, 0.5555555820465088, 0.5, 0.4444444477558136, 0.5, 0.5277777910232544, 0.5277777910232544, 0.5277777910232544, 0.6111111044883728, 0.6944444179534912, 0.5833333134651184, 0.6388888955116272, 0.6111111044883728, 0.6111111044883728, 0.5555555820465088, 0.75, 0.6388888955116272, 0.6111111044883728, 0.7222222089767456, 0.6666666865348816, 0.6944444179534912, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.75, 0.6944444179534912, 0.6944444179534912, 0.8333333134651184, 0.75, 0.7777777910232544, 0.8055555820465088, 0.8888888955116272, 0.9166666865348816, 0.7777777910232544, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456], [0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.1666666716337204, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.3888888955116272, 0.3611111044883728, 0.25, 0.4166666567325592, 0.3888888955116272, 0.4444444477558136, 0.4722222089767456, 0.4444444477558136, 0.5277777910232544, NaN, 0.5, 0.3333333432674408, 0.5555555820465088, 0.4722222089767456, 0.5277777910232544, 0.5277777910232544, 0.5277777910232544, 0.5277777910232544, 0.6666666865348816, 0.5833333134651184, 0.6944444179534912, 0.6111111044883728, 0.6111111044883728, 0.6666666865348816, 0.6666666865348816, 0.5833333134651184, 0.7222222089767456, 0.6944444179534912, 0.6666666865348816, 0.75, 0.75, 0.6944444179534912, 0.8055555820465088, 0.75, 0.8333333134651184, 0.8055555820465088, 0.7222222089767456, 0.75, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912, 1.0], [0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.1944444477558136, 0.1388888955116272, 0.1388888955116272, 0.25, 0.25, 0.25, 0.25, 0.3333333432674408, 0.3055555522441864, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.3611111044883728, 0.4166666567325592, 0.4166666567325592, 0.5277777910232544, 0.5, NaN, 0.4722222089767456, 0.5277777910232544, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5, 0.5277777910232544, 0.5833333134651184, 0.6111111044883728, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.6388888955116272, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.6666666865348816, 0.75, 0.75, 0.9166666865348816, 0.8055555820465088, 0.8611111044883728, 0.8055555820465088, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728, 0.7777777910232544, 0.8333333134651184, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.8055555820465088, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 1.0], [0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.1944444477558136, 0.1388888955116272, 0.1944444477558136, 0.25, 0.3333333432674408, 0.3055555522441864, 0.4166666567325592, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, 0.3611111044883728, 0.4722222089767456, 0.5, 0.3888888955116272, 0.3888888955116272, 0.3888888955116272, 0.4722222089767456, 0.5277777910232544, 0.4444444477558136, 0.6666666865348816, 0.5277777910232544, NaN, 0.5277777910232544, 0.5, 0.625, 0.5555555820465088, 0.5277777910232544, 0.5, 0.5, 0.5833333134651184, 0.5833333134651184, 0.5833333134651184, 0.6388888955116272, 0.5555555820465088, 0.5833333134651184, 0.7222222089767456, 0.6388888955116272, 0.5833333134651184, 0.6944444179534912, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.7777777910232544, 0.75, 0.8055555820465088, 0.8055555820465088, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.8472222089767456, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.9722222089767456], [0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.1666666716337204, 0.0833333358168602, 0.1388888955116272, 0.2222222238779068, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4722222089767456, 0.5, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, NaN, 0.5277777910232544, 0.4444444477558136, 0.5, 0.5833333134651184, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.5833333134651184, 0.6388888955116272, 0.6944444179534912, 0.6388888955116272, 0.6388888955116272, 0.5555555820465088, 0.6944444179534912, 0.8055555820465088, 0.6666666865348816, 0.6944444179534912, 0.75, 0.6666666865348816, 0.75, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.75, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.9166666865348816, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.02777777798473835, 0.1111111119389534, 0.1944444477558136, 0.25, 0.25, 0.1944444477558136, 0.25, 0.2777777910232544, 0.3333333432674408, 0.3055555522441864, 0.25, 0.3333333432674408, 0.4166666567325592, 0.4444444477558136, 0.2222222238779068, 0.3888888955116272, 0.4722222089767456, 0.4722222089767456, 0.4722222089767456, 0.5, 0.3888888955116272, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.5277777910232544, 0.5555555820465088, 0.5, 0.4722222089767456, NaN, 0.5277777910232544, 0.5, 0.5555555820465088, 0.5, 0.5277777910232544, 0.6111111044883728, 0.4444444477558136, 0.5277777910232544, 0.5, 0.5277777910232544, 0.5, 0.6944444179534912, 0.6111111044883728, 0.6388888955116272, 0.6388888955116272, 0.5555555820465088, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.6944444179534912, 0.8055555820465088, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.75, 0.75, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272, 0.75, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.8888888955116272, 0.8055555820465088, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456], [0.0555555559694767, 0.1388888955116272, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.25, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, 0.2222222238779068, 0.3055555522441864, 0.3611111044883728, 0.3611111044883728, 0.3333333432674408, 0.4166666567325592, 0.3888888955116272, 0.3888888955116272, 0.3888888955116272, 0.3611111044883728, 0.4722222089767456, 0.5, 0.4722222089767456, 0.4444444477558136, 0.375, 0.5555555820465088, 0.4722222089767456, NaN, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.5555555820465088, 0.5555555820465088, 0.5277777910232544, 0.5833333134651184, 0.6111111044883728, 0.5833333134651184, 0.6388888955116272, 0.6944444179534912, 0.6388888955116272, 0.6388888955116272, 0.5833333134651184, 0.7222222089767456, 0.6527777910232544, 0.6111111044883728, 0.75, 0.6944444179534912, 0.8055555820465088, 0.75, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8055555820465088, 0.7777777910232544, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.8333333134651184, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 1.0], [0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.25, 0.2777777910232544, 0.3333333432674408, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3055555522441864, 0.3333333432674408, 0.4722222089767456, 0.4444444477558136, 0.2638888955116272, 0.5, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4444444477558136, 0.5, 0.5, 0.4722222089767456, NaN, 0.5277777910232544, 0.5, 0.5555555820465088, 0.5833333134651184, 0.6388888955116272, 0.6111111044883728, 0.6388888955116272, 0.5833333134651184, 0.6666666865348816, 0.5555555820465088, 0.6388888955116272, 0.8333333134651184, 0.5833333134651184, 0.6666666865348816, 0.7222222089767456, 0.6111111044883728, 0.6944444179534912, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.7777777910232544, 0.75, 0.7777777910232544, 0.8055555820465088, 0.75, 0.8611111044883728, 0.7777777910232544, 0.7222222089767456, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 1.0], [0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.0972222238779068, 0.2222222238779068, 0.1111111119389534, 0.1666666716337204, 0.25, 0.1944444477558136, 0.25, 0.2777777910232544, 0.2777777910232544, 0.25, 0.3611111044883728, 0.3055555522441864, 0.3888888955116272, 0.3055555522441864, 0.3055555522441864, 0.3611111044883728, 0.3888888955116272, 0.4166666567325592, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.4166666567325592, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, NaN, 0.5555555820465088, 0.5277777910232544, 0.5555555820465088, 0.5277777910232544, 0.6388888955116272, 0.5833333134651184, 0.5555555820465088, 0.6388888955116272, 0.6388888955116272, 0.7777777910232544, 0.5555555820465088, 0.7222222089767456, 0.6111111044883728, 0.75, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.6944444179534912, 0.7777777910232544, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.7222222089767456, 0.7638888955116272, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.8055555820465088, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9722222089767456, 1.0], [0.0833333358168602, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.1388888955116272, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.3055555522441864, 0.3888888955116272, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.3611111044883728, 0.4166666567325592, 0.3888888955116272, 0.3333333432674408, 0.4444444477558136, 0.3055555522441864, 0.5277777910232544, 0.4722222089767456, 0.4722222089767456, 0.5, 0.5, 0.4722222089767456, 0.5, 0.4166666567325592, 0.5, 0.4444444477558136, NaN, 0.5, 0.5, 0.4722222089767456, 0.5277777910232544, 0.5833333134651184, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.6666666865348816, 0.7222222089767456, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.6666666865348816, 0.7222222089767456, 0.6388888955116272, 0.6388888955116272, 0.7222222089767456, 0.75, 0.75, 0.75, 0.75, 0.7777777910232544, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 1.0], [0.0833333358168602, 0.1666666716337204, 0.1388888955116272, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.25, 0.3055555522441864, 0.2777777910232544, 0.2222222238779068, 0.3055555522441864, 0.3333333432674408, 0.1666666716337204, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, 0.4444444477558136, 0.2777777910232544, 0.4166666567325592, 0.4166666567325592, 0.3888888955116272, 0.3333333432674408, 0.4722222089767456, 0.5, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4444444477558136, 0.4722222089767456, 0.5, NaN, 0.5277777910232544, 0.5833333134651184, 0.5555555820465088, 0.5277777910232544, 0.5555555820465088, 0.5833333134651184, 0.6111111044883728, 0.5416666865348816, 0.6111111044883728, 0.6111111044883728, 0.7222222089767456, 0.6944444179534912, 0.6388888955116272, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.8055555820465088, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.8055555820465088, 0.7222222089767456, 0.8055555820465088, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.9166666865348816, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912], [0.0555555559694767, 0.0555555559694767, 0.1388888955116272, 0.1388888955116272, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.25, 0.25, 0.25, 0.25, 0.3055555522441864, 0.25, 0.3333333432674408, 0.4166666567325592, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.3888888955116272, 0.4444444477558136, 0.3333333432674408, 0.3888888955116272, 0.3055555522441864, 0.4166666567325592, 0.4166666567325592, 0.4166666567325592, 0.4166666567325592, 0.3888888955116272, 0.4444444477558136, 0.4166666567325592, 0.4444444477558136, 0.5, 0.4722222089767456, NaN, 0.4722222089767456, 0.5, 0.5277777910232544, 0.5277777910232544, 0.5555555820465088, 0.6111111044883728, 0.5555555820465088, 0.5833333134651184, 0.7222222089767456, 0.6111111044883728, 0.6666666865348816, 0.75, 0.6944444179534912, 0.75, 0.6666666865348816, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.7777777910232544, 0.6944444179534912, 0.8055555820465088, 0.7777777910232544, 0.9444444179534912, 0.8888888955116272, 0.75, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.9722222089767456, 1.0], [0.0833333358168602, 0.1944444477558136, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.1944444477558136, 0.25, 0.3055555522441864, 0.25, 0.3611111044883728, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, 0.4166666567325592, 0.3055555522441864, 0.3888888955116272, 0.4166666567325592, 0.4166666567325592, 0.5555555820465088, 0.4722222089767456, 0.3611111044883728, 0.4722222089767456, 0.5277777910232544, 0.4166666567325592, 0.5277777910232544, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.7222222089767456, 0.6666666865348816, 0.7777777910232544, 0.5, 0.6666666865348816, 0.6111111044883728, 0.6944444179534912, 0.6111111044883728, 0.6111111044883728, 0.6388888955116272, 0.6944444179534912, 0.75, 0.7222222089767456, 0.75, 0.6388888955116272, 0.6666666865348816, 0.75, 0.75, 0.7777777910232544, 0.7222222089767456, 0.8055555820465088, 0.8333333134651184, 0.75, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456], [0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.0833333358168602, 0.1388888955116272, 0.1666666716337204, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.25, 0.2777777910232544, 0.2777777910232544, 0.25, 0.25, 0.1944444477558136, 0.2777777910232544, 0.25, 0.3611111044883728, 0.3611111044883728, 0.2777777910232544, 0.4166666567325592, 0.3611111044883728, 0.3611111044883728, 0.3888888955116272, 0.4166666567325592, 0.4166666567325592, 0.3611111044883728, 0.4722222089767456, 0.4166666567325592, 0.3888888955116272, 0.3611111044883728, 0.4722222089767456, 0.4444444477558136, 0.5, 0.5555555820465088, NaN, 0.6388888955116272, 0.5277777910232544, 0.5833333134651184, 0.5, 0.6666666865348816, 0.6944444179534912, 0.5277777910232544, 0.5555555820465088, 0.6666666865348816, 0.6111111044883728, 0.7222222089767456, 0.6666666865348816, 0.75, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.75, 0.8055555820465088, 0.8333333134651184, 0.75, 0.8333333134651184, 0.7777777910232544, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9166666865348816, 1.0], [0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.1666666716337204, 0.1111111119389534, 0.25, 0.3055555522441864, 0.3055555522441864, 0.2222222238779068, 0.25, 0.3055555522441864, 0.25, 0.3333333432674408, 0.3333333432674408, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3333333432674408, 0.3333333432674408, 0.3888888955116272, 0.3888888955116272, 0.3888888955116272, 0.3611111044883728, 0.3055555522441864, 0.5, 0.3888888955116272, 0.3611111044883728, 0.4166666567325592, 0.4166666567325592, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.3611111044883728, NaN, 0.5833333134651184, 0.5555555820465088, 0.6111111044883728, 0.5555555820465088, 0.5277777910232544, 0.6111111044883728, 0.6111111044883728, 0.6388888955116272, 0.6666666865348816, 0.6388888955116272, 0.6944444179534912, 0.75, 0.6388888955116272, 0.6666666865348816, 0.7222222089767456, 0.6944444179534912, 0.6666666865348816, 0.7777777910232544, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.8333333134651184, 0.8888888955116272, 0.8055555820465088, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 1.0], [0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.2777777910232544, 0.25, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.25, 0.2777777910232544, 0.2222222238779068, 0.25, 0.2777777910232544, 0.3055555522441864, 0.4166666567325592, 0.3333333432674408, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, 0.4444444477558136, 0.3611111044883728, 0.4722222089767456, 0.4166666567325592, 0.4166666567325592, 0.4444444477558136, 0.3611111044883728, 0.4444444477558136, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.4166666567325592, NaN, 0.4722222089767456, 0.5277777910232544, 0.5277777910232544, 0.5, 0.5833333134651184, 0.7222222089767456, 0.6666666865348816, 0.6111111044883728, 0.75, 0.6666666865348816, 0.75, 0.7222222089767456, 0.6388888955116272, 0.6666666865348816, 0.7222222089767456, 0.6944444179534912, 0.75, 0.7777777910232544, 0.75, 0.7222222089767456, 0.7222222089767456, 0.9444444179534912, 0.75, 0.8611111044883728, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912], [0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.0833333358168602, 0.1388888955116272, 0.0833333358168602, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.1388888955116272, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.1944444477558136, 0.3611111044883728, 0.2777777910232544, 0.3055555522441864, 0.25, 0.2222222238779068, 0.25, 0.3333333432674408, 0.3611111044883728, 0.3055555522441864, 0.4444444477558136, 0.3333333432674408, 0.3055555522441864, 0.4166666567325592, 0.3611111044883728, 0.5, 0.3611111044883728, 0.3333333432674408, 0.3611111044883728, 0.4166666567325592, 0.4166666567325592, 0.4444444477558136, 0.2777777910232544, 0.4166666567325592, 0.4444444477558136, 0.5277777910232544, NaN, 0.6111111044883728, 0.5833333134651184, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5833333134651184, 0.5555555820465088, 0.6111111044883728, 0.5833333134651184, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.6111111044883728, 0.6666666865348816, 0.75, 0.6666666865348816, 0.8055555820465088, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.75, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456], [0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1944444477558136, 0.2777777910232544, 0.25, 0.2222222238779068, 0.1666666716337204, 0.2777777910232544, 0.1944444477558136, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.3055555522441864, 0.3333333432674408, 0.25, 0.4166666567325592, 0.3611111044883728, 0.2777777910232544, 0.4444444477558136, 0.3055555522441864, 0.3055555522441864, 0.4444444477558136, 0.3611111044883728, 0.3888888955116272, 0.3888888955116272, 0.3888888955116272, 0.3333333432674408, 0.5, 0.3888888955116272, 0.4722222089767456, 0.3888888955116272, NaN, 0.4722222089767456, 0.5, 0.5833333134651184, 0.4722222089767456, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.6944444179534912, 0.5277777910232544, 0.6111111044883728, 0.75, 0.7222222089767456, 0.75, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.6666666865348816, 0.7777777910232544, 0.75, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456], [0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.25, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.3055555522441864, 0.2777777910232544, 0.3055555522441864, 0.3611111044883728, 0.2777777910232544, 0.3055555522441864, 0.3611111044883728, 0.3055555522441864, 0.3888888955116272, 0.3611111044883728, 0.3611111044883728, 0.2222222238779068, 0.3611111044883728, 0.4583333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4722222089767456, 0.4166666567325592, 0.5277777910232544, NaN, 0.4722222089767456, 0.5833333134651184, 0.5555555820465088, 0.5555555820465088, 0.5833333134651184, 0.5, 0.6111111044883728, 0.5, 0.6388888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.75, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.8333333134651184, 0.6666666865348816, 0.8055555820465088, 0.8611111044883728, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456], [0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.2777777910232544, 0.25, 0.25, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.25, 0.1388888955116272, 0.25, 0.2222222238779068, 0.25, 0.3888888955116272, 0.3055555522441864, 0.3055555522441864, 0.4166666567325592, 0.1944444477558136, 0.3611111044883728, 0.3611111044883728, 0.1666666716337204, 0.4444444477558136, 0.4166666567325592, 0.3888888955116272, 0.4166666567325592, 0.5, 0.3055555522441864, 0.4722222089767456, 0.5, 0.5555555820465088, 0.5, 0.5277777910232544, NaN, 0.4722222089767456, 0.4444444477558136, 0.6111111044883728, 0.5277777910232544, 0.5833333134651184, 0.6388888955116272, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.75, 0.7222222089767456, 0.7222222089767456, 0.75, 0.75, 0.6944444179534912, 0.6944444179534912, 0.8055555820465088, 0.7222222089767456, 0.6944444179534912, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456], [0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1666666716337204, 0.0555555559694767, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.25, 0.3055555522441864, 0.1666666716337204, 0.2222222238779068, 0.2777777910232544, 0.25, 0.3055555522441864, 0.3055555522441864, 0.25, 0.2777777910232544, 0.3333333432674408, 0.2222222238779068, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.4166666567325592, 0.4166666567325592, 0.2777777910232544, 0.3888888955116272, 0.3888888955116272, 0.2777777910232544, 0.3333333432674408, 0.4722222089767456, 0.3888888955116272, 0.4166666567325592, 0.4444444477558136, 0.4166666567325592, 0.4166666567325592, 0.5277777910232544, NaN, 0.4861111044883728, 0.5, 0.6805555820465088, 0.5277777910232544, 0.7222222089767456, 0.5, 0.5, 0.6388888955116272, 0.6666666865348816, 0.8055555820465088, 0.6388888955116272, 0.6388888955116272, 0.7222222089767456, 0.75, 0.9166666865348816, 0.75, 0.6388888955116272, 0.8611111044883728, 0.75, 0.9444444179534912, 0.75, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.9444444179534912, 1.0], [0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.1111111119389534, 0.1111111119389534, 0.0833333358168602, 0.0833333358168602, 0.1944444477558136, 0.0833333358168602, 0.2777777910232544, 0.1944444477558136, 0.1666666716337204, 0.3055555522441864, 0.2777777910232544, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.3055555522441864, 0.3194444477558136, 0.3333333432674408, 0.25, 0.3333333432674408, 0.2777777910232544, 0.3055555522441864, 0.4444444477558136, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.3333333432674408, 0.2777777910232544, 0.3888888955116272, 0.3888888955116272, 0.4444444477558136, 0.3888888955116272, 0.2777777910232544, 0.4444444477558136, 0.5277777910232544, 0.4444444477558136, 0.5555555820465088, 0.5138888955116272, NaN, 0.5, 0.5555555820465088, 0.5694444179534912, 0.6388888955116272, 0.5555555820465088, 0.6388888955116272, 0.6944444179534912, 0.6666666865348816, 0.6388888955116272, 0.6944444179534912, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.7777777910232544, 0.9166666865348816, 0.6388888955116272, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912], [0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.0833333358168602, 0.1111111119389534, 0.02777777798473835, 0.0555555559694767, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.25, 0.1944444477558136, 0.1944444477558136, 0.2777777910232544, 0.3055555522441864, 0.2222222238779068, 0.1666666716337204, 0.1944444477558136, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.1666666716337204, 0.25, 0.3055555522441864, 0.25, 0.25, 0.2777777910232544, 0.25, 0.3333333432674408, 0.3472222089767456, 0.2777777910232544, 0.25, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.3333333432674408, 0.4444444477558136, 0.4722222089767456, 0.4444444477558136, 0.3888888955116272, 0.5, 0.5, NaN, 0.5, 0.4722222089767456, 0.5555555820465088, 0.5277777910232544, 0.6944444179534912, 0.6388888955116272, 0.6388888955116272, 0.6111111044883728, 0.6111111044883728, 0.7222222089767456, 0.6388888955116272, 0.75, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.75, 0.9722222089767456, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9444444179534912], [0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.1944444477558136, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.25, 0.1666666716337204, 0.3055555522441864, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.2222222238779068, 0.3055555522441864, 0.2777777910232544, 0.25, 0.2222222238779068, 0.3055555522441864, 0.25, 0.3055555522441864, 0.3333333432674408, 0.2777777910232544, 0.3888888955116272, 0.3888888955116272, 0.2777777910232544, 0.3333333432674408, 0.3611111044883728, 0.25, 0.3888888955116272, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, 0.4166666567325592, 0.4722222089767456, 0.4166666567325592, 0.4722222089767456, 0.3194444477558136, 0.4444444477558136, 0.5, NaN, 0.4444444477558136, 0.5277777910232544, 0.4722222089767456, 0.5, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.6388888955116272, 0.6388888955116272, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.5833333134651184, 0.8333333134651184, 0.6944444179534912, 0.8888888955116272, 0.7222222089767456, 0.75, 0.75, 0.8333333134651184, 0.9722222089767456, 0.9722222089767456], [0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1944444477558136, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.2777777910232544, 0.1666666716337204, 0.1388888955116272, 0.0555555559694767, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.02777777798473835, 0.2222222238779068, 0.1944444477558136, 0.0833333358168602, 0.2222222238779068, 0.25, 0.2777777910232544, 0.25, 0.3055555522441864, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.3055555522441864, 0.3888888955116272, 0.2777777910232544, 0.3611111044883728, 0.25, 0.4444444477558136, 0.4166666567325592, 0.5, 0.4166666567325592, 0.4722222089767456, 0.4305555522441864, 0.5277777910232544, 0.5555555820465088, NaN, 0.5277777910232544, 0.4722222089767456, 0.5555555820465088, 0.6388888955116272, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.75, 0.6111111044883728, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.6666666865348816, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.75, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816], [0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.0833333358168602, 0.1944444477558136, 0.1388888955116272, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.2222222238779068, 0.2777777910232544, 0.2222222238779068, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1388888955116272, 0.25, 0.2777777910232544, 0.1666666716337204, 0.1944444477558136, 0.25, 0.1944444477558136, 0.25, 0.2222222238779068, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.1944444477558136, 0.3055555522441864, 0.3055555522441864, 0.25, 0.3611111044883728, 0.3333333432674408, 0.3055555522441864, 0.3333333432674408, 0.3888888955116272, 0.3055555522441864, 0.3888888955116272, 0.3611111044883728, 0.2777777910232544, 0.3611111044883728, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, NaN, 0.5, 0.4444444477558136, 0.5555555820465088, 0.5833333134651184, 0.7777777910232544, 0.5277777910232544, 0.5555555820465088, 0.6388888955116272, 0.6944444179534912, 0.7777777910232544, 0.7222222089767456, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8055555820465088, 0.6388888955116272, 0.7777777910232544, 0.75, 0.8055555820465088, 0.9166666865348816, 0.9722222089767456], [0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.1666666716337204, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.2222222238779068, 0.1666666716337204, 0.1388888955116272, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.2777777910232544, 0.1111111119389534, 0.2222222238779068, 0.1666666716337204, 0.1388888955116272, 0.1944444477558136, 0.2222222238779068, 0.3055555522441864, 0.1944444477558136, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3055555522441864, 0.25, 0.25, 0.25, 0.4166666567325592, 0.4722222089767456, 0.5, 0.3888888955116272, 0.5, 0.4444444477558136, 0.4722222089767456, 0.5277777910232544, 0.5277777910232544, 0.5, NaN, 0.5555555820465088, 0.6111111044883728, 0.6111111044883728, 0.5833333134651184, 0.5555555820465088, 0.7222222089767456, 0.6388888955116272, 0.7222222089767456, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912], [0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.0833333358168602, 0.1944444477558136, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.25, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.25, 0.3055555522441864, 0.2222222238779068, 0.2777777910232544, 0.1944444477558136, 0.3055555522441864, 0.25, 0.2777777910232544, 0.3611111044883728, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.3611111044883728, 0.3888888955116272, 0.5, 0.3611111044883728, 0.3055555522441864, 0.5, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, NaN, 0.6111111044883728, 0.6111111044883728, 0.5833333134651184, 0.5, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.6666666865348816, 0.6666666865348816, 0.75, 0.75, 0.75, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816], [0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.2222222238779068, 0.25, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.1805555522441864, 0.2222222238779068, 0.1944444477558136, 0.2777777910232544, 0.2222222238779068, 0.25, 0.25, 0.25, 0.25, 0.2222222238779068, 0.3055555522441864, 0.2777777910232544, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.2777777910232544, 0.2222222238779068, 0.3055555522441864, 0.3611111044883728, 0.3333333432674408, 0.3333333432674408, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3611111044883728, 0.3055555522441864, 0.25, 0.3333333432674408, 0.3055555522441864, 0.3611111044883728, 0.3055555522441864, 0.3611111044883728, 0.3333333432674408, 0.3611111044883728, 0.4444444477558136, 0.3888888955116272, 0.3888888955116272, NaN, 0.4583333432674408, 0.4444444477558136, 0.5277777910232544, 0.4166666567325592, 0.5277777910232544, 0.5555555820465088, 0.5833333134651184, 0.9444444179534912, 0.5, 0.6111111044883728, 0.6111111044883728, 0.6111111044883728, 0.5555555820465088, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728], [0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.2222222238779068, 0.2222222238779068, 0.25, 0.25, 0.25, 0.1944444477558136, 0.1944444477558136, 0.25, 0.1944444477558136, 0.2777777910232544, 0.25, 0.25, 0.25, 0.25, 0.25, 0.2222222238779068, 0.3055555522441864, 0.25, 0.2222222238779068, 0.3055555522441864, 0.25, 0.3055555522441864, 0.2777777910232544, 0.25, 0.3055555522441864, 0.3611111044883728, 0.3055555522441864, 0.3055555522441864, 0.25, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3055555522441864, 0.2777777910232544, 0.3333333432674408, 0.25, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, 0.3333333432674408, 0.3611111044883728, 0.4166666567325592, 0.3888888955116272, 0.3888888955116272, 0.5416666865348816, NaN, 0.4444444477558136, 0.5277777910232544, 0.4444444477558136, 0.5277777910232544, 0.5555555820465088, 0.5555555820465088, 0.9722222089767456, 0.5277777910232544, 0.5833333134651184, 0.5833333134651184, 0.5833333134651184, 0.5555555820465088, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728], [0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.0555555559694767, 0.1666666716337204, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.2222222238779068, 0.25, 0.1111111119389534, 0.1666666716337204, 0.1666666716337204, 0.1388888955116272, 0.2777777910232544, 0.1666666716337204, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.2222222238779068, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.3611111044883728, 0.2222222238779068, 0.3055555522441864, 0.2777777910232544, 0.3888888955116272, 0.25, 0.3333333432674408, 0.2777777910232544, 0.1944444477558136, 0.3611111044883728, 0.3888888955116272, 0.3055555522441864, 0.4166666567325592, 0.2222222238779068, 0.4166666567325592, 0.4166666567325592, 0.5555555820465088, 0.5555555820465088, NaN, 0.5, 0.5833333134651184, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5277777910232544, 0.7638888955116272, 0.6944444179534912, 0.8055555820465088, 0.6111111044883728, 0.7777777910232544, 0.75, 0.7777777910232544, 0.8888888955116272, 0.9722222089767456], [0.0, 0.02777777798473835, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.2222222238779068, 0.1944444477558136, 0.1111111119389534, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1111111119389534, 0.1388888955116272, 0.2222222238779068, 0.2222222238779068, 0.25, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.1388888955116272, 0.25, 0.2222222238779068, 0.1944444477558136, 0.2777777910232544, 0.25, 0.1944444477558136, 0.3055555522441864, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3055555522441864, 0.3333333432674408, 0.3333333432674408, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3055555522441864, 0.3888888955116272, 0.3611111044883728, 0.3888888955116272, 0.4722222089767456, 0.4444444477558136, 0.5, 0.4722222089767456, 0.4722222089767456, 0.5, NaN, 0.5555555820465088, 0.7361111044883728, 0.5833333134651184, 0.5277777910232544, 0.5, 0.5277777910232544, 0.6388888955116272, 0.6388888955116272, 0.6111111044883728, 0.8611111044883728, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.75, 0.8888888955116272], [0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.1388888955116272, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.0555555559694767, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.2222222238779068, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.1666666716337204, 0.25, 0.2222222238779068, 0.25, 0.2361111044883728, 0.25, 0.2777777910232544, 0.1944444477558136, 0.25, 0.2222222238779068, 0.2222222238779068, 0.25, 0.25, 0.2777777910232544, 0.25, 0.25, 0.3611111044883728, 0.3333333432674408, 0.2777777910232544, 0.3611111044883728, 0.25, 0.4444444477558136, 0.2777777910232544, 0.3333333432674408, 0.5833333134651184, 0.5555555820465088, 0.4166666567325592, 0.4444444477558136, NaN, 0.4166666567325592, 0.5, 0.4722222089767456, 0.6666666865348816, 0.5555555820465088, 0.5277777910232544, 0.5555555820465088, 0.75, 0.5833333134651184, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272], [0.0, 0.0, 0.0555555559694767, 0.0, 0.02777777798473835, 0.0, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1944444477558136, 0.1388888955116272, 0.1111111119389534, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.2222222238779068, 0.0833333358168602, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.25, 0.1944444477558136, 0.2222222238779068, 0.25, 0.1944444477558136, 0.2777777910232544, 0.2222222238779068, 0.3333333432674408, 0.2777777910232544, 0.3055555522441864, 0.25, 0.2777777910232544, 0.2777777910232544, 0.3611111044883728, 0.3333333432674408, 0.3888888955116272, 0.3611111044883728, 0.3611111044883728, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.2638888955116272, 0.5833333134651184, NaN, 0.5833333134651184, 0.5, 0.5, 0.5277777910232544, 0.5833333134651184, 0.6111111044883728, 0.5555555820465088, 0.7638888955116272, 0.75, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.8888888955116272], [0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1388888955116272, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.1111111119389534, 0.1666666716337204, 0.2222222238779068, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1388888955116272, 0.25, 0.2777777910232544, 0.0555555559694767, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.25, 0.1944444477558136, 0.3333333432674408, 0.3055555522441864, 0.3055555522441864, 0.25, 0.2777777910232544, 0.25, 0.3333333432674408, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4166666567325592, 0.5, 0.4166666567325592, NaN, 0.5277777910232544, 0.6388888955116272, 0.5, 0.5833333134651184, 0.5, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7222222089767456, 0.7777777910232544, 0.9166666865348816], [0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.1111111119389534, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.1111111119389534, 0.0833333358168602, 0.1944444477558136, 0.0555555559694767, 0.1111111119389534, 0.1666666716337204, 0.1388888955116272, 0.2222222238779068, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.2222222238779068, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1111111119389534, 0.2222222238779068, 0.2777777910232544, 0.1666666716337204, 0.2222222238779068, 0.2777777910232544, 0.1111111119389534, 0.2777777910232544, 0.25, 0.2222222238779068, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.0833333358168602, 0.3055555522441864, 0.2777777910232544, 0.2222222238779068, 0.3055555522441864, 0.2222222238779068, 0.3333333432674408, 0.3055555522441864, 0.4166666567325592, 0.4444444477558136, 0.3333333432674408, 0.4722222089767456, 0.5277777910232544, 0.5, 0.4722222089767456, NaN, 0.6388888955116272, 0.3888888955116272, 0.6111111044883728, 0.5555555820465088, 0.6111111044883728, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7222222089767456, 0.8055555820465088, 0.9444444179534912], [0.0555555559694767, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.25, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.25, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.2777777910232544, 0.1944444477558136, 0.25, 0.2222222238779068, 0.1944444477558136, 0.25, 0.2222222238779068, 0.2777777910232544, 0.2222222238779068, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.0555555559694767, 0.02777777798473835, 0.3333333432674408, 0.5, 0.3333333432674408, 0.5, 0.3611111044883728, 0.3611111044883728, NaN, 0.3888888955116272, 0.4444444477558136, 0.3611111044883728, 0.5555555820465088, 0.5277777910232544, 0.5555555820465088, 0.5555555820465088, 0.6388888955116272, 0.6388888955116272, 0.8333333134651184], [0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.1388888955116272, 0.1111111119389534, 0.0555555559694767, 0.1388888955116272, 0.1388888955116272, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.0833333358168602, 0.02777777798473835, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1111111119389534, 0.2222222238779068, 0.1666666716337204, 0.1666666716337204, 0.0833333358168602, 0.1944444477558136, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.1666666716337204, 0.1111111119389534, 0.0555555559694767, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.2777777910232544, 0.3611111044883728, 0.0833333358168602, 0.2777777910232544, 0.4166666567325592, 0.3333333432674408, 0.3333333432674408, 0.3055555522441864, 0.3333333432674408, 0.5, 0.4722222089767456, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.5, 0.6111111044883728, 0.6111111044883728, NaN, 0.5555555820465088, 0.5277777910232544, 0.6111111044883728, 0.5833333134651184, 0.6666666865348816, 0.7222222089767456, 0.8055555820465088, 0.7777777910232544, 0.8888888955116272], [0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.1111111119389534, 0.1111111119389534, 0.1666666716337204, 0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.1388888955116272, 0.1666666716337204, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.1111111119389534, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.25, 0.25, 0.1944444477558136, 0.25, 0.25, 0.1666666716337204, 0.3333333432674408, 0.3055555522441864, 0.1388888955116272, 0.3611111044883728, 0.25, 0.1666666716337204, 0.2777777910232544, 0.1111111119389534, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.4166666567325592, 0.2361111044883728, 0.3611111044883728, 0.4722222089767456, 0.4166666567325592, 0.4166666567325592, 0.3888888955116272, 0.5555555820465088, 0.4444444477558136, NaN, 0.5, 0.6388888955116272, 0.5277777910232544, 0.7222222089767456, 0.7222222089767456, 0.75, 0.8333333134651184, 0.9722222089767456], [0.0, 0.0, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1666666716337204, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.0833333358168602, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.25, 0.1944444477558136, 0.02777777798473835, 0.3055555522441864, 0.2222222238779068, 0.3333333432674408, 0.2777777910232544, 0.25, 0.3888888955116272, 0.4166666567325592, 0.3055555522441864, 0.3611111044883728, 0.4444444477558136, 0.3888888955116272, 0.5, 0.4444444477558136, 0.6388888955116272, 0.4722222089767456, 0.5, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.9166666865348816], [0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1527777761220932, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1944444477558136, 0.0833333358168602, 0.2222222238779068, 0.1944444477558136, 0.0833333358168602, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.1388888955116272, 0.2222222238779068, 0.0555555559694767, 0.1666666716337204, 0.1944444477558136, 0.1111111119389534, 0.2777777910232544, 0.1944444477558136, 0.2777777910232544, 0.25, 0.3888888955116272, 0.4166666567325592, 0.1944444477558136, 0.3888888955116272, 0.25, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.4444444477558136, 0.3888888955116272, 0.3611111044883728, 0.4444444477558136, NaN, 0.5, 0.6388888955116272, 0.6388888955116272, 0.7222222089767456, 0.9722222089767456, 0.9722222089767456], [0.0, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.1111111119389534, 0.0555555559694767, 0.02777777798473835, 0.1111111119389534, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.0555555559694767, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.1111111119389534, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1111111119389534, 0.1944444477558136, 0.25, 0.1666666716337204, 0.1388888955116272, 0.2777777910232544, 0.25, 0.3611111044883728, 0.1944444477558136, 0.25, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.1388888955116272, 0.4166666567325592, 0.2361111044883728, 0.4444444477558136, 0.4444444477558136, 0.4722222089767456, 0.4166666567325592, 0.4722222089767456, 0.3333333432674408, 0.5, NaN, 0.6388888955116272, 0.6388888955116272, 0.6944444179534912, 0.6111111044883728, 0.8333333134651184], [0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.0555555559694767, 0.2222222238779068, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.0833333358168602, 0.0555555559694767, 0.1111111119389534, 0.02777777798473835, 0.1388888955116272, 0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.0555555559694767, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.1666666716337204, 0.25, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.3055555522441864, 0.3055555522441864, 0.2222222238779068, 0.2222222238779068, 0.2777777910232544, 0.25, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.2777777910232544, 0.3333333432674408, 0.3611111044883728, 0.3611111044883728, NaN, 0.7083333134651184, 0.7222222089767456, 0.5277777910232544, 0.75], [0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0416666679084301, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.02777777798473835, 0.1944444477558136, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.1111111119389534, 0.0555555559694767, 0.1388888955116272, 0.02777777798473835, 0.1388888955116272, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1388888955116272, 0.25, 0.2222222238779068, 0.25, 0.1666666716337204, 0.1944444477558136, 0.2777777910232544, 0.3055555522441864, 0.25, 0.2222222238779068, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.3611111044883728, 0.3611111044883728, 0.2916666567325592, NaN, 0.6388888955116272, 0.5, 0.7222222089767456], [0.0, 0.0, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0, 0.0, 0.0555555559694767, 0.02777777798473835, 0.0, 0.1111111119389534, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.1111111119389534, 0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.1666666716337204, 0.02777777798473835, 0.1388888955116272, 0.0833333358168602, 0.02777777798473835, 0.1111111119389534, 0.1388888955116272, 0.0833333358168602, 0.1111111119389534, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.1111111119389534, 0.02777777798473835, 0.02777777798473835, 0.1666666716337204, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.1944444477558136, 0.1388888955116272, 0.1388888955116272, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.2777777910232544, 0.2777777910232544, 0.3611111044883728, 0.1944444477558136, 0.25, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, NaN, 0.4722222089767456, 0.6666666865348816], [0.02777777798473835, 0.02777777798473835, 0.0, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0, 0.0555555559694767, 0.0, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.1388888955116272, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.1111111119389534, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.25, 0.1944444477558136, 0.2777777910232544, 0.2222222238779068, 0.1944444477558136, 0.3611111044883728, 0.2222222238779068, 0.1666666716337204, 0.2222222238779068, 0.02777777798473835, 0.3888888955116272, 0.4722222089767456, 0.5, 0.5277777910232544, NaN, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02777777798473835, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02777777798473835, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02777777798473835, 0.0, 0.0, 0.02777777798473835, 0.0, 0.0, 0.02777777798473835, 0.0, 0.02777777798473835, 0.0, 0.0, 0.0, 0.0, 0.0555555559694767, 0.0, 0.02777777798473835, 0.0, 0.0, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.02777777798473835, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0833333358168602, 0.0555555559694767, 0.1666666716337204, 0.1111111119389534, 0.02777777798473835, 0.0833333358168602, 0.02777777798473835, 0.1666666716337204, 0.25, 0.2777777910232544, 0.3333333432674408, 0.1111111119389534, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9311594151},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8597020893},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8490338155},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8429951653},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8423913026},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8228663424},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8063607092},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8009259239},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7556360692},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7419484695},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.710144927},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6990740739},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.687198068},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6797504042},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6706924342},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6574074074},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6455313997},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6396940427},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6348631245},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6276167472},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6229871181},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6145330107},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5851449288},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5845410629},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5680354289},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.568035429},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5676328505},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.566425122},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5603864764},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5551529797},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5497181959},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5434782618},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5430756841},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5332125619},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5253623193},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5213365552},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5072463781},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.506441226},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4887278576},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4762479897},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4545088576},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4478663448},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4476650549},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4416264101},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4369967832},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4192834145},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4112318852},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3790257667},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3772141746},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3699677945},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3643317264},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3578905015},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3576892134},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3319243199},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3295088592},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.300523353},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2850241561},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2777777815},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2705314043},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2608695695},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2592592614},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2445652185},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2290660251},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2188003238},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2099436401},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1721014511},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1574074095},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1284219023},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1207729484},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0454911443}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_all/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_all/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..c0ad5cc1dd9a8b617cec7903af004bd1eac8e5fa
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_all/datasets_small/winrate_matrix.csv
@@ -0,0 +1,71 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabDPT (tuned + ensembled),TabDPT-Turbo (default),RealMLP (tuned),LimiX (default),TabDPT (tuned),TabM (tuned + ensembled),iLTM (tuned + ensembled),LightGBM (tuned + ensembled),CatBoost (tuned),CatBoost (tuned + ensembled),TabM (tuned),TabSwift (default),XGBoost (tuned + ensembled),xRFM (tuned + ensembled),LightGBM (tuned),XGBoost (tuned),ModernNCA (tuned + ensembled),CatBoost (default),TabDPT (default),ModernNCA (tuned),ChimeraBoost (tuned + ensembled),iLTM (tuned),SAP-RPT-OSS (default),TabM (default),xRFM (tuned),EBM (tuned + ensembled),ChimeraBoost (tuned),RealMLP (default),TorchMLP (tuned + ensembled),EBM (tuned),ModernNCA (default),EBM (default),ChimeraBoost (default),ExtraTrees (tuned + ensembled),TorchMLP (tuned),FastaiMLP (tuned + ensembled),ExtraTrees (tuned),LightGBM (default),RandomForest (tuned + ensembled),XGBoost (default),FastaiMLP (tuned),TabSTAR (tuned),TabSTAR (tuned + ensembled),RandomForest (tuned),PerpetualBooster (tuned + ensembled),iLTM (default),PerpetualBooster (tuned),xRFM (default),ExtraTrees (default),TabSTAR (default),TorchMLP (default),RandomForest (default),FastaiMLP (default),KNN (tuned + ensembled),PerpetualBooster (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.8333333,0.7777778,0.8333333,0.8611111,0.8333333,0.8888889,0.8611111,0.8611111,0.9166667,0.9166667,0.9166667,0.8888889,0.9166667,0.9722222,0.9166667,0.8888889,0.9444444,0.9444444,0.9722222,0.9444444,0.8888889,0.9444444,0.9444444,0.9444444,0.9166667,0.8888889,0.9166667,0.9722222,0.9444444,0.9444444,0.9444444,0.9166667,0.9166667,0.9444444,0.9166667,0.9166667,0.8888889,0.9722222,0.9166667,0.9444444,0.9166667,0.9444444,0.9444444,0.9722222,0.9166667,0.9444444,0.9444444,0.9444444,0.9722222,0.9166667,0.9166667,0.8888889,0.9444444,1.0,0.9166667,1.0,0.9722222,0.9722222,0.9444444,0.9722222,0.9722222,1.0,0.9444444,1.0,0.9722222,0.9722222,1.0,0.9722222,1.0
+TabPFN-3 (default),0.16666667,,0.5277778,0.5277778,0.5138889,0.5277778,0.5833333,0.6388889,0.75,0.6388889,0.6944444,0.8055556,0.7777778,0.7222222,0.8333333,0.8333333,0.8611111,0.9166667,0.7777778,0.8333333,0.8888889,0.8611111,0.9166667,0.8888889,0.9444444,0.8888889,0.9166667,0.9166667,0.8888889,0.8611111,0.9166667,0.9166667,0.8055556,0.8333333,0.9444444,0.8055556,0.8888889,0.9166667,0.9444444,0.8611111,0.9444444,0.8888889,0.9444444,0.9722222,0.9444444,0.8888889,0.9722222,0.9444444,0.9444444,0.9722222,0.9444444,0.8611111,0.8611111,0.9722222,0.9722222,0.9722222,1.0,0.9722222,0.9722222,0.8611111,0.9444444,0.9722222,1.0,0.9722222,0.9722222,0.9722222,0.9722222,1.0,0.9722222,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.22222222,0.4722222,,0.5833333,0.4722222,0.5277778,0.5,0.6111111,0.8055556,0.7777778,0.7222222,0.6666667,0.8055556,0.7222222,0.7777778,0.8333333,0.8611111,0.8888889,0.8333333,0.8611111,0.8055556,0.8055556,0.9166667,0.9166667,0.9166667,0.9166667,0.8888889,0.9166667,0.8055556,0.9166667,0.9166667,0.9166667,0.8055556,0.8611111,0.8611111,0.8333333,0.9166667,0.8611111,0.9166667,0.8611111,0.9444444,0.8888889,0.9444444,0.9166667,0.8888889,0.8611111,0.9166667,0.9722222,0.9444444,0.9444444,0.9166667,0.8611111,0.8333333,0.9444444,0.9166667,0.9444444,0.9444444,0.9722222,0.9722222,0.8611111,0.9444444,0.9444444,0.9722222,0.9722222,0.9444444,0.9722222,0.9722222,0.9722222,1.0,1.0
+TabPFN-2.6 (default),0.16666667,0.4722222,0.41666666,,0.4722222,0.6388889,0.5,0.5555556,0.7222222,0.6666667,0.6388889,0.7222222,0.8055556,0.7777778,0.6944444,0.7777778,0.8888889,0.8888889,0.8333333,0.8888889,0.8055556,0.8055556,0.8888889,0.8888889,0.9444444,0.8888889,0.9166667,0.9444444,0.75,0.8611111,0.8611111,0.9166667,0.8333333,0.8333333,0.8611111,0.8611111,0.8611111,0.8888889,0.9444444,0.9166667,0.9444444,0.8611111,0.9166667,0.9166667,0.9444444,0.9166667,0.9166667,0.9444444,0.9444444,0.9722222,0.8888889,0.8333333,0.8333333,0.9166667,0.9722222,0.9444444,1.0,0.9722222,0.9444444,0.8333333,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.1388889,0.4861111,0.5277778,0.5277778,,0.6388889,0.6111111,0.6944444,0.75,0.6666667,0.6388889,0.6666667,0.7777778,0.75,0.6944444,0.7777778,0.8333333,0.8888889,0.8888889,0.9166667,0.8333333,0.7777778,0.9166667,0.8611111,0.9166667,0.8888889,0.8611111,0.9444444,0.75,0.8888889,0.8888889,0.8611111,0.8611111,0.8611111,0.9166667,0.8611111,0.9166667,0.8611111,0.8888889,0.8611111,0.8611111,0.8888889,0.8888889,0.9166667,0.8888889,0.8888889,0.9166667,0.9722222,0.9166667,1.0,0.9166667,0.8333333,0.8333333,0.9166667,0.9444444,0.9166667,0.9722222,0.9444444,0.9722222,0.8611111,0.9444444,0.9722222,0.9722222,0.9444444,0.9444444,0.9444444,0.9722222,0.9444444,0.9444444,1.0
+RealTabPFN-2.5 (default),0.16666667,0.4722222,0.4722222,0.3611111,0.3611111,,0.4722222,0.43055555,0.6666667,0.6388889,0.6388889,0.6944444,0.75,0.7222222,0.6944444,0.7777778,0.7222222,0.8611111,0.8055556,0.8333333,0.8055556,0.8333333,0.8611111,0.8333333,0.8888889,0.8611111,0.8055556,0.8888889,0.8055556,0.8055556,0.8611111,0.9027778,0.8055556,0.8611111,0.9166667,0.8611111,0.8611111,0.8333333,0.9166667,0.9166667,0.8888889,0.9166667,0.9166667,0.8611111,0.8888889,0.9722222,0.8611111,0.9444444,0.8888889,0.9444444,1.0,0.8611111,0.8611111,0.9444444,0.9444444,0.9722222,1.0,0.9166667,0.9722222,0.8888889,0.9722222,0.9444444,1.0,0.9166667,0.9444444,0.9722222,0.9722222,1.0,0.9722222,0.9722222
+TabICLv2 (default),0.11111111,0.41666666,0.5,0.5,0.3888889,0.5277778,,0.5833333,0.6666667,0.75,0.6944444,0.7222222,0.6666667,0.6944444,0.7222222,0.7222222,0.8055556,0.8333333,0.7777778,0.7777778,0.75,0.8055556,0.8055556,0.8055556,0.8333333,0.8055556,0.8611111,0.8333333,0.75,0.8055556,0.8611111,0.7777778,0.7777778,0.8055556,0.8611111,0.8055556,0.8333333,0.8055556,0.8888889,0.8055556,0.8888889,0.8888889,0.8611111,0.8333333,0.9166667,0.9444444,0.8888889,0.9166667,0.8611111,0.9166667,0.8888889,0.8611111,0.8611111,0.8611111,0.9166667,0.8611111,0.9444444,0.9444444,0.8888889,0.8888889,0.9444444,0.9166667,0.9722222,0.9444444,0.9722222,0.9722222,0.9722222,1.0,0.9722222,1.0
+RealTabPFN-2.5 (tuned),0.1388889,0.3611111,0.3888889,0.44444445,0.30555555,0.5694444,0.41666666,,0.6388889,0.6388889,0.5833333,0.6388889,0.75,0.6111111,0.6388889,0.8055556,0.7222222,0.7777778,0.7777778,0.8055556,0.8333333,0.6944444,0.8888889,0.8055556,0.8888889,0.8611111,0.8055556,0.9166667,0.7222222,0.8611111,0.8055556,0.8888889,0.8055556,0.8333333,0.8888889,0.8333333,0.8611111,0.8333333,0.8888889,0.8333333,0.8888889,0.8611111,0.8888889,0.9444444,0.9166667,0.8611111,0.9166667,0.9444444,0.9166667,0.9722222,0.8888889,0.8055556,0.7777778,0.8888889,0.9444444,0.9166667,0.9722222,0.9166667,0.9444444,0.8611111,0.9444444,0.9444444,0.9444444,0.9166667,0.9444444,0.9444444,0.9444444,0.9444444,0.9444444,1.0
+RealMLP (tuned + ensembled),0.1388889,0.25,0.19444445,0.2777778,0.25,0.33333334,0.33333334,0.3611111,,0.5555556,0.44444445,0.5555556,0.6388889,0.4722222,0.5,0.6666667,0.7777778,0.7777778,0.7777778,0.8611111,0.7222222,0.6388889,0.8611111,0.6944444,0.8333333,0.8611111,0.75,0.8611111,0.6666667,0.75,0.8055556,0.8333333,0.7222222,0.7777778,0.75,0.8333333,0.8333333,0.8888889,0.8333333,0.8333333,0.8888889,0.8611111,0.8611111,0.8333333,0.8055556,0.8888889,0.8611111,0.9166667,0.8055556,0.9444444,0.8611111,0.7777778,0.7777778,0.8888889,0.9444444,0.9444444,0.9722222,0.9166667,0.9722222,0.8333333,0.9166667,0.9722222,0.9722222,0.9444444,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,1.0
+"AutoGluon 1.4 (best, 4h)",0.083333336,0.3611111,0.22222222,0.33333334,0.33333334,0.3611111,0.25,0.3611111,0.44444445,,0.41666666,0.5277778,0.6111111,0.5555556,0.5,0.6111111,0.6944444,0.7777778,0.6666667,0.7222222,0.7222222,0.5833333,0.8055556,0.7222222,0.8333333,0.75,0.6666667,0.7777778,0.6944444,0.6666667,0.8055556,0.75,0.6944444,0.75,0.75,0.7222222,0.8333333,0.75,0.8888889,0.8611111,0.8888889,0.8611111,0.8333333,0.8055556,0.9166667,0.8888889,0.8055556,0.9166667,0.8611111,0.9166667,0.9166667,0.75,0.75,0.9444444,0.9444444,0.9166667,0.9722222,0.8888889,0.9722222,0.8333333,0.9722222,0.9722222,0.9722222,0.9166667,0.9722222,0.9722222,0.9722222,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.083333336,0.30555555,0.2777778,0.3611111,0.3611111,0.3611111,0.30555555,0.41666666,0.5555556,0.5833333,,0.41666666,0.6666667,0.6111111,0.7222222,0.5555556,0.7222222,0.6388889,0.6111111,0.6666667,0.5555556,0.6666667,0.7222222,0.6666667,0.6944444,0.75,0.6944444,0.6944444,0.75,0.6666667,0.75,0.8055556,0.6944444,0.6944444,0.75,0.6944444,0.75,0.6944444,0.7222222,0.7222222,0.8055556,0.8055556,0.7222222,0.8055556,0.7222222,0.75,0.8888889,0.8055556,0.8055556,0.8333333,0.8055556,0.8055556,0.75,0.8333333,0.7777778,0.8611111,0.8611111,0.9444444,0.9722222,0.8333333,0.8611111,0.9444444,0.9722222,0.9166667,0.8888889,0.9166667,0.9166667,0.8888889,0.9444444,1.0
+TabDPT-Turbo (default),0.083333336,0.19444445,0.33333334,0.2777778,0.33333334,0.30555555,0.2777778,0.3611111,0.44444445,0.4722222,0.5833333,,0.6111111,0.5277778,0.5555556,0.5,0.6111111,0.6388889,0.5555556,0.6111111,0.5555556,0.6805556,0.7222222,0.6944444,0.6944444,0.75,0.5833333,0.6944444,0.6666667,0.6388889,0.7222222,0.75,0.6944444,0.7222222,0.75,0.6944444,0.7222222,0.6944444,0.75,0.7222222,0.7222222,0.8055556,0.75,0.8333333,0.8055556,0.8055556,0.8611111,0.8888889,0.8611111,0.8611111,0.8888889,0.7777778,0.75,0.8611111,0.8055556,0.8611111,0.8611111,0.8888889,0.9444444,0.8055556,0.8888889,0.9166667,0.9722222,0.9722222,0.9444444,0.9166667,0.9166667,0.9166667,0.9722222,1.0
+RealMLP (tuned),0.11111111,0.22222222,0.19444445,0.19444445,0.22222222,0.25,0.33333334,0.25,0.3611111,0.3888889,0.33333334,0.3888889,,0.4027778,0.44444445,0.5277778,0.5555556,0.6388889,0.6111111,0.6111111,0.6388889,0.5555556,0.7222222,0.5833333,0.7222222,0.75,0.6666667,0.7222222,0.5833333,0.7777778,0.6666667,0.7222222,0.6111111,0.7777778,0.6944444,0.8055556,0.7222222,0.7777778,0.7777778,0.7777778,0.75,0.8055556,0.75,0.7777778,0.8333333,0.8055556,0.8333333,0.8333333,0.8333333,0.8888889,0.8333333,0.7777778,0.8055556,0.8888889,0.8888889,0.9444444,0.8888889,0.9166667,0.8888889,0.8055556,0.9444444,0.9722222,0.9444444,0.9166667,0.9722222,0.9444444,0.9583333,0.9444444,0.9722222,1.0
+LimiX (default),0.083333336,0.2777778,0.2777778,0.22222222,0.25,0.2777778,0.30555555,0.3888889,0.5277778,0.44444445,0.3888889,0.4722222,0.5972222,,0.5,0.5833333,0.625,0.6388889,0.5277778,0.5833333,0.6388889,0.5277778,0.6388889,0.6388889,0.6666667,0.6666667,0.6666667,0.6666667,0.5555556,0.6944444,0.6944444,0.7222222,0.6388889,0.6944444,0.75,0.75,0.75,0.75,0.7222222,0.8055556,0.7777778,0.8333333,0.7777778,0.7777778,0.6944444,0.7222222,0.75,0.8055556,0.7777778,0.8333333,0.8055556,0.8194444,0.8055556,0.8333333,0.8055556,0.8611111,0.8055556,0.8333333,0.8888889,0.8611111,0.8611111,0.8888889,0.8888889,0.9166667,0.8888889,0.9444444,0.9444444,0.9722222,0.9166667,1.0
+TabDPT (tuned),0.027777778,0.16666667,0.22222222,0.30555555,0.30555555,0.30555555,0.2777778,0.3611111,0.5,0.5,0.2777778,0.44444445,0.5555556,0.5,,0.5555556,0.6666667,0.6111111,0.5833333,0.6388889,0.5555556,0.6666667,0.6666667,0.6388889,0.6666667,0.6944444,0.6388889,0.6666667,0.7777778,0.6388889,0.7222222,0.75,0.6111111,0.6666667,0.6666667,0.6944444,0.75,0.6944444,0.6944444,0.6388889,0.8333333,0.7777778,0.7777778,0.75,0.7222222,0.6944444,0.8333333,0.7222222,0.7222222,0.7777778,0.8055556,0.7777778,0.75,0.8055556,0.7777778,0.8611111,0.8611111,0.8888889,0.9166667,0.8055556,0.8611111,0.8888889,0.9722222,0.8888889,0.9166667,0.8888889,0.8888889,0.8888889,0.9444444,0.9722222
+TabM (tuned + ensembled),0.083333336,0.16666667,0.16666667,0.22222222,0.22222222,0.22222222,0.2777778,0.19444445,0.33333334,0.3888889,0.44444445,0.5,0.4722222,0.41666666,0.44444445,,0.5277778,0.5555556,0.4722222,0.5277778,0.6111111,0.5555556,0.6388889,0.5833333,0.6111111,0.6944444,0.6388889,0.6666667,0.6111111,0.6388889,0.6388889,0.6388889,0.6388889,0.8333333,0.5833333,0.75,0.8055556,0.75,0.75,0.7222222,0.7222222,0.7777778,0.8055556,0.6944444,0.8055556,0.7777778,0.6944444,0.8333333,0.7777778,0.8333333,0.8333333,0.8055556,0.8055556,0.7777778,0.8055556,0.8611111,0.8888889,0.8333333,0.8055556,0.8333333,0.9722222,0.8333333,0.9166667,0.8888889,0.9444444,0.9444444,0.9444444,0.9444444,1.0,1.0
+iLTM (tuned + ensembled),0.11111111,0.1388889,0.1388889,0.11111111,0.16666667,0.2777778,0.19444445,0.2777778,0.22222222,0.30555555,0.2777778,0.3888889,0.44444445,0.375,0.33333334,0.4722222,,0.5555556,0.5555556,0.5,0.5277778,0.5,0.5833333,0.5555556,0.6388889,0.6666667,0.5277778,0.5555556,0.5277778,0.6666667,0.6944444,0.6944444,0.6388889,0.6111111,0.6944444,0.6388889,0.7222222,0.6666667,0.7222222,0.6944444,0.8055556,0.75,0.8055556,0.8333333,0.7777778,0.8333333,0.7777778,0.8611111,0.8888889,0.8611111,0.8333333,0.7222222,0.7222222,0.8888889,0.8888889,0.8611111,0.9444444,0.8888889,0.9444444,0.8055556,0.9444444,0.9444444,0.9444444,0.9444444,0.9444444,0.9166667,0.9444444,0.9444444,0.9444444,1.0
+LightGBM (tuned + ensembled),0.055555556,0.083333336,0.11111111,0.11111111,0.11111111,0.1388889,0.16666667,0.22222222,0.22222222,0.22222222,0.3611111,0.3611111,0.3611111,0.3611111,0.3888889,0.44444445,0.44444445,,0.5,0.5555556,0.4722222,0.5277778,0.5555556,0.5833333,0.75,0.6944444,0.5,0.7222222,0.5277778,0.5833333,0.6666667,0.6111111,0.5833333,0.6666667,0.6666667,0.7222222,0.75,0.6666667,0.7777778,0.75,0.7777778,0.8055556,0.8333333,0.7777778,0.7777778,0.8055556,0.7222222,0.9444444,0.8333333,0.9166667,0.8333333,0.7777778,0.75,0.9166667,0.8611111,0.8888889,0.8888889,0.8333333,0.8888889,0.8055556,0.8888889,0.9722222,0.9166667,0.9444444,0.8888889,0.9722222,0.9722222,0.9444444,1.0,1.0
+CatBoost (tuned),0.055555556,0.22222222,0.16666667,0.16666667,0.11111111,0.19444445,0.22222222,0.22222222,0.22222222,0.33333334,0.3888889,0.44444445,0.3888889,0.4722222,0.41666666,0.5277778,0.44444445,0.5,,0.44444445,0.5555556,0.4722222,0.6388889,0.5277778,0.5833333,0.6666667,0.6111111,0.6944444,0.5277778,0.6111111,0.5277778,0.6944444,0.6111111,0.6111111,0.6388889,0.6944444,0.6388889,0.6944444,0.75,0.7777778,0.7222222,0.7777778,0.75,0.7222222,0.7777778,0.7222222,0.6944444,0.8055556,0.8055556,0.8611111,0.8055556,0.75,0.75,0.8888889,0.8333333,0.8611111,0.9166667,0.8055556,0.8333333,0.7777778,0.9166667,0.9166667,0.9166667,0.8888889,0.9444444,0.9722222,0.9722222,0.9722222,0.9722222,1.0
+CatBoost (tuned + ensembled),0.027777778,0.16666667,0.1388889,0.11111111,0.083333336,0.16666667,0.22222222,0.19444445,0.1388889,0.2777778,0.33333334,0.3888889,0.3888889,0.41666666,0.3611111,0.4722222,0.5,0.44444445,0.5555556,,0.4722222,0.44444445,0.6111111,0.4722222,0.6111111,0.6388889,0.6111111,0.6666667,0.5,0.6111111,0.5555556,0.6944444,0.6666667,0.5555556,0.6111111,0.6666667,0.6388889,0.7222222,0.7222222,0.75,0.6944444,0.7777778,0.8611111,0.75,0.7777778,0.7222222,0.7777778,0.8055556,0.8611111,0.8611111,0.8333333,0.75,0.75,0.9166667,0.8888889,0.8611111,0.8888889,0.8611111,0.8611111,0.7777778,0.9166667,0.9444444,0.8888889,0.8611111,0.9444444,0.9722222,0.9722222,0.9722222,0.9444444,1.0
+TabM (tuned),0.055555556,0.11111111,0.19444445,0.19444445,0.16666667,0.19444445,0.25,0.16666667,0.2777778,0.2777778,0.44444445,0.44444445,0.3611111,0.3611111,0.44444445,0.3888889,0.4722222,0.5277778,0.44444445,0.5277778,,0.5277778,0.5833333,0.5277778,0.5555556,0.6388889,0.6111111,0.6666667,0.6111111,0.6111111,0.7361111,0.6388889,0.5555556,0.7222222,0.5555556,0.6666667,0.7222222,0.6388889,0.6944444,0.6666667,0.6944444,0.6944444,0.75,0.6944444,0.8055556,0.6944444,0.6944444,0.8333333,0.75,0.8333333,0.8333333,0.75,0.75,0.7777778,0.8611111,0.8055556,0.8888889,0.8055556,0.7777778,0.8055556,0.9722222,0.8611111,0.8611111,0.8888889,0.9444444,0.9166667,0.9166667,0.9444444,0.9722222,0.9722222
+TabSwift (default),0.11111111,0.1388889,0.19444445,0.19444445,0.22222222,0.16666667,0.19444445,0.30555555,0.3611111,0.41666666,0.33333334,0.31944445,0.44444445,0.4722222,0.33333334,0.44444445,0.5,0.4722222,0.5277778,0.5555556,0.4722222,,0.5555556,0.5833333,0.5277778,0.5833333,0.5277778,0.5555556,0.5277778,0.6388889,0.5,0.6111111,0.6944444,0.5833333,0.6666667,0.6666667,0.5833333,0.6666667,0.5833333,0.6388889,0.6944444,0.7222222,0.7777778,0.6944444,0.6944444,0.8333333,0.7222222,0.7777778,0.7222222,0.7222222,0.8055556,0.75,0.75,0.75,0.7777778,0.8055556,0.8333333,0.8333333,0.8611111,0.8333333,0.8611111,0.8333333,0.9444444,0.9166667,0.8888889,0.9166667,0.9166667,0.9166667,0.9722222,1.0
+XGBoost (tuned + ensembled),0.055555556,0.083333336,0.083333336,0.11111111,0.083333336,0.1388889,0.19444445,0.11111111,0.1388889,0.19444445,0.2777778,0.2777778,0.2777778,0.3611111,0.33333334,0.3611111,0.41666666,0.44444445,0.3611111,0.3888889,0.41666666,0.44444445,,0.5,0.5555556,0.5833333,0.4722222,0.5277778,0.4722222,0.5277778,0.5555556,0.5833333,0.4722222,0.5833333,0.6111111,0.6388889,0.6388889,0.6666667,0.6666667,0.6944444,0.6666667,0.6944444,0.75,0.75,0.6805556,0.75,0.75,0.9722222,0.8333333,0.8888889,0.7777778,0.7777778,0.7777778,0.8888889,0.7777778,0.8611111,0.8333333,0.7777778,0.8888889,0.8055556,0.8888889,0.9444444,0.8888889,0.8611111,0.8611111,0.9166667,0.9444444,0.9444444,0.9166667,1.0
+xRFM (tuned + ensembled),0.055555556,0.11111111,0.083333336,0.11111111,0.1388889,0.16666667,0.19444445,0.19444445,0.30555555,0.2777778,0.33333334,0.30555555,0.41666666,0.3611111,0.3611111,0.41666666,0.44444445,0.41666666,0.4722222,0.5277778,0.4722222,0.41666666,0.5,,0.4722222,0.4722222,0.5555556,0.5,0.44444445,0.5,0.5277778,0.5277778,0.5277778,0.6111111,0.6944444,0.5833333,0.6388889,0.6111111,0.6111111,0.5555556,0.75,0.6388889,0.6111111,0.7222222,0.6666667,0.6944444,0.7777778,0.7777778,0.8055556,0.7777778,0.75,0.6944444,0.6944444,0.8333333,0.75,0.7777778,0.8055556,0.8888889,0.9166667,0.7777778,0.8888889,0.9166667,0.9166667,0.9166667,0.8611111,0.9444444,0.9722222,0.9444444,0.9722222,0.9722222
+LightGBM (tuned),0.055555556,0.055555556,0.083333336,0.055555556,0.083333336,0.11111111,0.16666667,0.11111111,0.16666667,0.16666667,0.30555555,0.30555555,0.2777778,0.33333334,0.33333334,0.3888889,0.3611111,0.25,0.41666666,0.3888889,0.44444445,0.4722222,0.44444445,0.5277778,,0.5,0.33333334,0.5555556,0.4722222,0.5277778,0.5277778,0.5277778,0.5277778,0.6666667,0.5833333,0.6944444,0.6111111,0.6111111,0.6666667,0.6666667,0.5833333,0.7222222,0.6944444,0.6666667,0.75,0.75,0.6944444,0.8055556,0.75,0.8333333,0.8055556,0.7222222,0.75,0.8333333,0.8333333,0.8055556,0.8611111,0.8333333,0.7777778,0.8055556,0.8611111,0.9166667,0.8888889,0.8888889,0.8888889,0.9166667,0.9166667,0.9166667,0.9444444,1.0
+XGBoost (tuned),0.083333336,0.11111111,0.083333336,0.11111111,0.11111111,0.1388889,0.19444445,0.1388889,0.1388889,0.25,0.25,0.25,0.25,0.33333334,0.30555555,0.30555555,0.33333334,0.30555555,0.33333334,0.3611111,0.3611111,0.41666666,0.41666666,0.5277778,0.5,,0.4722222,0.5277778,0.44444445,0.5555556,0.5555556,0.5555556,0.5,0.5277778,0.5833333,0.6111111,0.5833333,0.6111111,0.6111111,0.6944444,0.6388889,0.6944444,0.6944444,0.7777778,0.6666667,0.75,0.75,0.9166667,0.8055556,0.8611111,0.8055556,0.7777778,0.7777778,0.8611111,0.7777778,0.8333333,0.8333333,0.7777778,0.8611111,0.8055556,0.8611111,0.9166667,0.8888889,0.8888889,0.8611111,0.8888889,0.9166667,0.9444444,0.9166667,1.0
+ModernNCA (tuned + ensembled),0.11111111,0.083333336,0.11111111,0.083333336,0.1388889,0.19444445,0.1388889,0.19444445,0.25,0.33333334,0.30555555,0.41666666,0.33333334,0.33333334,0.3611111,0.3611111,0.4722222,0.5,0.3888889,0.3888889,0.3888889,0.4722222,0.5277778,0.44444445,0.6666667,0.5277778,,0.5277778,0.5,0.625,0.5555556,0.5277778,0.5,0.5,0.5833333,0.5833333,0.5833333,0.6388889,0.5555556,0.5833333,0.7222222,0.6388889,0.5833333,0.6944444,0.7222222,0.7222222,0.6944444,0.7777778,0.75,0.8055556,0.8055556,0.6666667,0.6944444,0.7222222,0.7777778,0.7777778,0.7777778,0.8611111,0.8055556,0.8055556,0.8055556,0.8611111,0.8333333,0.8472222,0.8611111,0.8333333,0.8611111,0.8333333,0.8611111,0.9722222
+CatBoost (default),0.083333336,0.083333336,0.083333336,0.055555556,0.055555556,0.11111111,0.16666667,0.083333336,0.1388889,0.22222222,0.30555555,0.30555555,0.2777778,0.33333334,0.33333334,0.33333334,0.44444445,0.2777778,0.30555555,0.33333334,0.33333334,0.44444445,0.4722222,0.5,0.44444445,0.4722222,0.4722222,,0.5277778,0.44444445,0.5,0.5833333,0.5277778,0.5277778,0.5833333,0.5833333,0.6388889,0.6944444,0.6388889,0.6388889,0.5555556,0.6944444,0.8055556,0.6666667,0.6944444,0.75,0.6666667,0.75,0.7777778,0.7777778,0.7777778,0.7777778,0.75,0.8333333,0.8611111,0.8333333,0.9166667,0.8333333,0.8055556,0.8055556,0.8888889,0.8888889,0.9166667,0.8888889,0.9444444,0.9444444,0.9722222,0.9722222,0.9722222,1.0
+TabDPT (default),0.027777778,0.11111111,0.19444445,0.25,0.25,0.19444445,0.25,0.2777778,0.33333334,0.30555555,0.25,0.33333334,0.41666666,0.44444445,0.22222222,0.3888889,0.4722222,0.4722222,0.4722222,0.5,0.3888889,0.4722222,0.5277778,0.5555556,0.5277778,0.5555556,0.5,0.4722222,,0.5277778,0.5,0.5555556,0.5,0.5277778,0.6111111,0.44444445,0.5277778,0.5,0.5277778,0.5,0.6944444,0.6111111,0.6388889,0.6388889,0.5555556,0.6666667,0.7222222,0.7222222,0.6944444,0.6944444,0.8055556,0.6666667,0.6944444,0.7222222,0.75,0.75,0.8055556,0.8055556,0.8888889,0.75,0.7777778,0.8055556,0.8611111,0.8888889,0.8055556,0.7777778,0.8055556,0.8611111,0.9444444,0.9722222
+ModernNCA (tuned),0.055555556,0.1388889,0.083333336,0.1388889,0.11111111,0.19444445,0.19444445,0.1388889,0.25,0.33333334,0.33333334,0.3611111,0.22222222,0.30555555,0.3611111,0.3611111,0.33333334,0.41666666,0.3888889,0.3888889,0.3888889,0.3611111,0.4722222,0.5,0.4722222,0.44444445,0.375,0.5555556,0.4722222,,0.5277778,0.5277778,0.5833333,0.5555556,0.5555556,0.5277778,0.5833333,0.6111111,0.5833333,0.6388889,0.6944444,0.6388889,0.6388889,0.5833333,0.7222222,0.6527778,0.6111111,0.75,0.6944444,0.8055556,0.75,0.7222222,0.7222222,0.7222222,0.7777778,0.7777778,0.8055556,0.8055556,0.7777778,0.8333333,0.8333333,0.8611111,0.8611111,0.8888889,0.8333333,0.9166667,0.9444444,0.9166667,0.9166667,1.0
+ChimeraBoost (tuned + ensembled),0.055555556,0.083333336,0.083333336,0.1388889,0.11111111,0.1388889,0.1388889,0.19444445,0.19444445,0.19444445,0.25,0.2777778,0.33333334,0.30555555,0.2777778,0.3611111,0.30555555,0.33333334,0.4722222,0.44444445,0.2638889,0.5,0.44444445,0.4722222,0.4722222,0.44444445,0.44444445,0.5,0.5,0.4722222,,0.5277778,0.5,0.5555556,0.5833333,0.6388889,0.6111111,0.6388889,0.5833333,0.6666667,0.5555556,0.6388889,0.8333333,0.5833333,0.6666667,0.7222222,0.6111111,0.6944444,0.7222222,0.7222222,0.6944444,0.7777778,0.75,0.7777778,0.8055556,0.75,0.8611111,0.7777778,0.7222222,0.8055556,0.8333333,0.8333333,0.8611111,0.8055556,0.8333333,0.9166667,0.9444444,0.9722222,0.9444444,1.0
+iLTM (tuned),0.055555556,0.083333336,0.083333336,0.083333336,0.1388889,0.097222224,0.22222222,0.11111111,0.16666667,0.25,0.19444445,0.25,0.2777778,0.2777778,0.25,0.3611111,0.30555555,0.3888889,0.30555555,0.30555555,0.3611111,0.3888889,0.41666666,0.4722222,0.4722222,0.44444445,0.4722222,0.41666666,0.44444445,0.4722222,0.4722222,,0.5555556,0.5277778,0.5555556,0.5277778,0.6388889,0.5833333,0.5555556,0.6388889,0.6388889,0.7777778,0.5555556,0.7222222,0.6111111,0.75,0.7222222,0.7222222,0.8055556,0.6944444,0.7777778,0.6944444,0.6944444,0.7777778,0.7222222,0.7638889,0.8333333,0.8611111,0.8333333,0.8055556,0.9166667,0.8888889,0.8888889,0.9166667,0.8888889,0.8611111,0.8888889,0.8888889,0.9722222,1.0
+SAP-RPT-OSS (default),0.083333336,0.19444445,0.19444445,0.16666667,0.1388889,0.19444445,0.22222222,0.19444445,0.2777778,0.30555555,0.30555555,0.30555555,0.3888889,0.3611111,0.3888889,0.3611111,0.3611111,0.41666666,0.3888889,0.33333334,0.44444445,0.30555555,0.5277778,0.4722222,0.4722222,0.5,0.5,0.4722222,0.5,0.41666666,0.5,0.44444445,,0.5,0.5,0.4722222,0.5277778,0.5833333,0.6388889,0.5833333,0.6111111,0.6388889,0.5833333,0.6111111,0.6666667,0.7222222,0.6666667,0.6944444,0.6944444,0.6666667,0.7222222,0.6388889,0.6388889,0.7222222,0.75,0.75,0.75,0.75,0.7777778,0.8333333,0.8055556,0.8055556,0.8333333,0.7777778,0.8611111,0.9166667,0.8888889,0.8611111,0.8888889,1.0
+TabM (default),0.083333336,0.16666667,0.1388889,0.16666667,0.1388889,0.1388889,0.19444445,0.16666667,0.22222222,0.25,0.30555555,0.2777778,0.22222222,0.30555555,0.33333334,0.16666667,0.3888889,0.33333334,0.3888889,0.44444445,0.2777778,0.41666666,0.41666666,0.3888889,0.33333334,0.4722222,0.5,0.4722222,0.4722222,0.44444445,0.44444445,0.4722222,0.5,,0.5277778,0.5833333,0.5555556,0.5277778,0.5555556,0.5833333,0.6111111,0.5416667,0.6111111,0.6111111,0.7222222,0.6944444,0.6388889,0.6944444,0.6944444,0.7777778,0.8055556,0.6666667,0.6944444,0.7222222,0.8055556,0.7222222,0.8055556,0.7222222,0.7222222,0.8055556,0.9166667,0.8055556,0.8055556,0.8055556,0.8888889,0.9444444,0.9444444,0.9166667,0.9722222,0.9444444
+xRFM (tuned),0.055555556,0.055555556,0.1388889,0.1388889,0.083333336,0.083333336,0.1388889,0.11111111,0.25,0.25,0.25,0.25,0.30555555,0.25,0.33333334,0.41666666,0.30555555,0.33333334,0.3611111,0.3888889,0.44444445,0.33333334,0.3888889,0.30555555,0.41666666,0.41666666,0.41666666,0.41666666,0.3888889,0.44444445,0.41666666,0.44444445,0.5,0.4722222,,0.4722222,0.5,0.5277778,0.5277778,0.5555556,0.6111111,0.5555556,0.5833333,0.7222222,0.6111111,0.6666667,0.75,0.6944444,0.75,0.6666667,0.6944444,0.6666667,0.6944444,0.7777778,0.6944444,0.8055556,0.7777778,0.9444444,0.8888889,0.75,0.8888889,0.8611111,0.8611111,0.9166667,0.9166667,0.8888889,0.8611111,0.8888889,0.9722222,1.0
+EBM (tuned + ensembled),0.083333336,0.19444445,0.16666667,0.1388889,0.1388889,0.1388889,0.19444445,0.16666667,0.16666667,0.2777778,0.30555555,0.30555555,0.19444445,0.25,0.30555555,0.25,0.3611111,0.2777778,0.30555555,0.33333334,0.33333334,0.33333334,0.3611111,0.41666666,0.30555555,0.3888889,0.41666666,0.41666666,0.5555556,0.4722222,0.3611111,0.4722222,0.5277778,0.41666666,0.5277778,,0.44444445,0.5555556,0.5555556,0.7222222,0.6666667,0.7777778,0.5,0.6666667,0.6111111,0.6944444,0.6111111,0.6111111,0.6388889,0.6944444,0.75,0.7222222,0.75,0.6388889,0.6666667,0.75,0.75,0.7777778,0.7222222,0.8055556,0.8333333,0.75,0.8611111,0.8888889,0.8888889,0.9722222,0.9722222,0.9722222,0.9166667,0.9722222
+ChimeraBoost (tuned),0.083333336,0.11111111,0.083333336,0.1388889,0.083333336,0.1388889,0.16666667,0.1388889,0.16666667,0.16666667,0.25,0.2777778,0.2777778,0.25,0.25,0.19444445,0.2777778,0.25,0.3611111,0.3611111,0.2777778,0.41666666,0.3611111,0.3611111,0.3888889,0.41666666,0.41666666,0.3611111,0.4722222,0.41666666,0.3888889,0.3611111,0.4722222,0.44444445,0.5,0.5555556,,0.6388889,0.5277778,0.5833333,0.5,0.6666667,0.6944444,0.5277778,0.5555556,0.6666667,0.6111111,0.7222222,0.6666667,0.75,0.7222222,0.7222222,0.7222222,0.7777778,0.7777778,0.7777778,0.8055556,0.7777778,0.75,0.8055556,0.8333333,0.75,0.8333333,0.7777778,0.8333333,0.8611111,0.8611111,0.9444444,0.9166667,1.0
+RealMLP (default),0.11111111,0.083333336,0.1388889,0.11111111,0.1388889,0.16666667,0.19444445,0.16666667,0.11111111,0.25,0.30555555,0.30555555,0.22222222,0.25,0.30555555,0.25,0.33333334,0.33333334,0.30555555,0.2777778,0.3611111,0.33333334,0.33333334,0.3888889,0.3888889,0.3888889,0.3611111,0.30555555,0.5,0.3888889,0.3611111,0.41666666,0.41666666,0.4722222,0.4722222,0.44444445,0.3611111,,0.5833333,0.5555556,0.6111111,0.5555556,0.5277778,0.6111111,0.6111111,0.6388889,0.6666667,0.6388889,0.6944444,0.75,0.6388889,0.6666667,0.7222222,0.6944444,0.6666667,0.7777778,0.7222222,0.7777778,0.7777778,0.8333333,0.8888889,0.8055556,0.9166667,0.8888889,0.9166667,0.9166667,0.9166667,0.9722222,0.9444444,1.0
+TorchMLP (tuned + ensembled),0.027777778,0.055555556,0.083333336,0.055555556,0.11111111,0.083333336,0.11111111,0.11111111,0.16666667,0.11111111,0.2777778,0.25,0.22222222,0.2777778,0.30555555,0.25,0.2777778,0.22222222,0.25,0.2777778,0.30555555,0.41666666,0.33333334,0.3888889,0.33333334,0.3888889,0.44444445,0.3611111,0.4722222,0.41666666,0.41666666,0.44444445,0.3611111,0.44444445,0.4722222,0.44444445,0.4722222,0.41666666,,0.4722222,0.5277778,0.5277778,0.5,0.5833333,0.7222222,0.6666667,0.6111111,0.75,0.6666667,0.75,0.7222222,0.6388889,0.6666667,0.7222222,0.6944444,0.75,0.7777778,0.75,0.7222222,0.7222222,0.9444444,0.75,0.8611111,0.8333333,0.8888889,0.8888889,0.9166667,0.9722222,0.9166667,0.9444444
+EBM (tuned),0.083333336,0.1388889,0.1388889,0.083333336,0.1388889,0.083333336,0.19444445,0.16666667,0.16666667,0.1388889,0.2777778,0.2777778,0.22222222,0.19444445,0.3611111,0.2777778,0.30555555,0.25,0.22222222,0.25,0.33333334,0.3611111,0.30555555,0.44444445,0.33333334,0.30555555,0.41666666,0.3611111,0.5,0.3611111,0.33333334,0.3611111,0.41666666,0.41666666,0.44444445,0.2777778,0.41666666,0.44444445,0.5277778,,0.6111111,0.5833333,0.44444445,0.5555556,0.5555556,0.5555556,0.5833333,0.5555556,0.6111111,0.5833333,0.6666667,0.6944444,0.6944444,0.6111111,0.6666667,0.75,0.6666667,0.8055556,0.7222222,0.8055556,0.8055556,0.75,0.8333333,0.8888889,0.8611111,0.9166667,0.9444444,0.9166667,0.9166667,0.9722222
+ModernNCA (default),0.055555556,0.055555556,0.055555556,0.055555556,0.1388889,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.19444445,0.2777778,0.25,0.22222222,0.16666667,0.2777778,0.19444445,0.22222222,0.2777778,0.30555555,0.30555555,0.30555555,0.33333334,0.25,0.41666666,0.3611111,0.2777778,0.44444445,0.30555555,0.30555555,0.44444445,0.3611111,0.3888889,0.3888889,0.3888889,0.33333334,0.5,0.3888889,0.4722222,0.3888889,,0.4722222,0.5,0.5833333,0.4722222,0.5277778,0.5277778,0.5833333,0.6944444,0.5277778,0.6111111,0.75,0.7222222,0.75,0.6666667,0.7222222,0.7222222,0.6666667,0.7777778,0.75,0.8055556,0.8333333,0.8055556,0.8333333,0.8333333,0.8888889,0.8888889,0.8888889,0.9166667,0.9722222
+EBM (default),0.083333336,0.11111111,0.11111111,0.1388889,0.11111111,0.083333336,0.11111111,0.1388889,0.1388889,0.1388889,0.19444445,0.19444445,0.19444445,0.16666667,0.22222222,0.22222222,0.25,0.19444445,0.22222222,0.22222222,0.30555555,0.2777778,0.30555555,0.3611111,0.2777778,0.30555555,0.3611111,0.30555555,0.3888889,0.3611111,0.3611111,0.22222222,0.3611111,0.45833334,0.44444445,0.22222222,0.33333334,0.44444445,0.4722222,0.41666666,0.5277778,,0.4722222,0.5833333,0.5555556,0.5555556,0.5833333,0.5,0.6111111,0.5,0.6388889,0.6666667,0.6666667,0.6666667,0.6944444,0.75,0.6944444,0.6944444,0.7222222,0.7777778,0.8333333,0.6666667,0.8055556,0.8611111,0.8888889,0.9444444,0.9444444,0.9722222,0.9166667,0.9722222
+ChimeraBoost (default),0.055555556,0.055555556,0.055555556,0.083333336,0.11111111,0.083333336,0.1388889,0.11111111,0.1388889,0.16666667,0.2777778,0.25,0.25,0.22222222,0.22222222,0.19444445,0.19444445,0.16666667,0.25,0.1388889,0.25,0.22222222,0.25,0.3888889,0.30555555,0.30555555,0.41666666,0.19444445,0.3611111,0.3611111,0.16666667,0.44444445,0.41666666,0.3888889,0.41666666,0.5,0.30555555,0.4722222,0.5,0.5555556,0.5,0.5277778,,0.4722222,0.44444445,0.6111111,0.5277778,0.5833333,0.6388889,0.6111111,0.6111111,0.6944444,0.75,0.7222222,0.7222222,0.75,0.75,0.6944444,0.6944444,0.8055556,0.7222222,0.6944444,0.7777778,0.7777778,0.8055556,0.8611111,0.8611111,0.9722222,0.9444444,0.9722222
+ExtraTrees (tuned + ensembled),0.055555556,0.027777778,0.083333336,0.083333336,0.083333336,0.1388889,0.16666667,0.055555556,0.16666667,0.19444445,0.19444445,0.16666667,0.22222222,0.22222222,0.25,0.30555555,0.16666667,0.22222222,0.2777778,0.25,0.30555555,0.30555555,0.25,0.2777778,0.33333334,0.22222222,0.30555555,0.33333334,0.3611111,0.41666666,0.41666666,0.2777778,0.3888889,0.3888889,0.2777778,0.33333334,0.4722222,0.3888889,0.41666666,0.44444445,0.41666666,0.41666666,0.5277778,,0.4861111,0.5,0.6805556,0.5277778,0.7222222,0.5,0.5,0.6388889,0.6666667,0.8055556,0.6388889,0.6388889,0.7222222,0.75,0.9166667,0.75,0.6388889,0.8611111,0.75,0.9444444,0.75,0.8333333,0.8611111,0.8333333,0.9444444,1.0
+TorchMLP (tuned),0.027777778,0.055555556,0.11111111,0.055555556,0.11111111,0.11111111,0.083333336,0.083333336,0.19444445,0.083333336,0.2777778,0.19444445,0.16666667,0.30555555,0.2777778,0.19444445,0.22222222,0.22222222,0.22222222,0.22222222,0.19444445,0.30555555,0.31944445,0.33333334,0.25,0.33333334,0.2777778,0.30555555,0.44444445,0.2777778,0.33333334,0.3888889,0.33333334,0.2777778,0.3888889,0.3888889,0.44444445,0.3888889,0.2777778,0.44444445,0.5277778,0.44444445,0.5555556,0.5138889,,0.5,0.5555556,0.5694444,0.6388889,0.5555556,0.6388889,0.6944444,0.6666667,0.6388889,0.6944444,0.6666667,0.7222222,0.7222222,0.6944444,0.7777778,0.9166667,0.6388889,0.8055556,0.8333333,0.8333333,0.8055556,0.8333333,0.9166667,0.9166667,0.9444444
+FastaiMLP (tuned + ensembled),0.083333336,0.11111111,0.1388889,0.083333336,0.11111111,0.027777778,0.055555556,0.1388889,0.11111111,0.11111111,0.25,0.19444445,0.19444445,0.2777778,0.30555555,0.22222222,0.16666667,0.19444445,0.2777778,0.2777778,0.30555555,0.16666667,0.25,0.30555555,0.25,0.25,0.2777778,0.25,0.33333334,0.3472222,0.2777778,0.25,0.2777778,0.30555555,0.33333334,0.30555555,0.33333334,0.3611111,0.33333334,0.44444445,0.4722222,0.44444445,0.3888889,0.5,0.5,,0.5,0.4722222,0.5555556,0.5277778,0.6944444,0.6388889,0.6388889,0.6111111,0.6111111,0.7222222,0.6388889,0.75,0.7222222,0.7222222,0.7222222,0.75,0.9722222,0.8055556,0.8611111,0.8333333,0.8611111,0.8888889,0.8888889,0.9444444
+ExtraTrees (tuned),0.055555556,0.027777778,0.083333336,0.083333336,0.083333336,0.1388889,0.11111111,0.083333336,0.1388889,0.19444445,0.11111111,0.1388889,0.16666667,0.25,0.16666667,0.30555555,0.22222222,0.2777778,0.30555555,0.22222222,0.30555555,0.2777778,0.25,0.22222222,0.30555555,0.25,0.30555555,0.33333334,0.2777778,0.3888889,0.3888889,0.2777778,0.33333334,0.3611111,0.25,0.3888889,0.3888889,0.33333334,0.3888889,0.41666666,0.4722222,0.41666666,0.4722222,0.31944445,0.44444445,0.5,,0.44444445,0.5277778,0.4722222,0.5,0.6666667,0.6666667,0.6944444,0.6388889,0.6388889,0.6666667,0.6666667,0.7777778,0.7777778,0.5833333,0.8333333,0.6944444,0.8888889,0.7222222,0.75,0.75,0.8333333,0.9722222,0.9722222
+LightGBM (default),0.055555556,0.055555556,0.027777778,0.055555556,0.027777778,0.055555556,0.083333336,0.055555556,0.083333336,0.083333336,0.19444445,0.11111111,0.16666667,0.19444445,0.2777778,0.16666667,0.1388889,0.055555556,0.19444445,0.19444445,0.16666667,0.22222222,0.027777778,0.22222222,0.19444445,0.083333336,0.22222222,0.25,0.2777778,0.25,0.30555555,0.2777778,0.30555555,0.30555555,0.30555555,0.3888889,0.2777778,0.3611111,0.25,0.44444445,0.41666666,0.5,0.41666666,0.4722222,0.43055555,0.5277778,0.5555556,,0.5277778,0.4722222,0.5555556,0.6388889,0.6388889,0.5833333,0.6111111,0.75,0.6111111,0.6944444,0.6944444,0.7222222,0.6666667,0.7222222,0.7777778,0.7222222,0.75,0.7777778,0.7777778,0.8888889,0.9166667,0.9166667
+RandomForest (tuned + ensembled),0.055555556,0.055555556,0.055555556,0.055555556,0.083333336,0.11111111,0.1388889,0.083333336,0.19444445,0.1388889,0.19444445,0.1388889,0.16666667,0.22222222,0.2777778,0.22222222,0.11111111,0.16666667,0.19444445,0.1388889,0.25,0.2777778,0.16666667,0.19444445,0.25,0.19444445,0.25,0.22222222,0.30555555,0.30555555,0.2777778,0.19444445,0.30555555,0.30555555,0.25,0.3611111,0.33333334,0.30555555,0.33333334,0.3888889,0.30555555,0.3888889,0.3611111,0.2777778,0.3611111,0.44444445,0.4722222,0.4722222,,0.5,0.44444445,0.5555556,0.5833333,0.7777778,0.5277778,0.5555556,0.6388889,0.6944444,0.7777778,0.7222222,0.6666667,0.8888889,0.6666667,0.8055556,0.6388889,0.7777778,0.75,0.8055556,0.9166667,0.9722222
+XGBoost (default),0.027777778,0.027777778,0.055555556,0.027777778,0.0,0.055555556,0.083333336,0.027777778,0.055555556,0.083333336,0.16666667,0.1388889,0.11111111,0.16666667,0.22222222,0.16666667,0.1388889,0.083333336,0.1388889,0.1388889,0.16666667,0.2777778,0.11111111,0.22222222,0.16666667,0.1388889,0.19444445,0.22222222,0.30555555,0.19444445,0.2777778,0.30555555,0.33333334,0.22222222,0.33333334,0.30555555,0.25,0.25,0.25,0.41666666,0.4722222,0.5,0.3888889,0.5,0.44444445,0.4722222,0.5277778,0.5277778,0.5,,0.5555556,0.6111111,0.6111111,0.5833333,0.5555556,0.7222222,0.6388889,0.7222222,0.6666667,0.6944444,0.6944444,0.7222222,0.7222222,0.7222222,0.8055556,0.8055556,0.8333333,0.8611111,0.8611111,0.9444444
+FastaiMLP (tuned),0.083333336,0.055555556,0.083333336,0.11111111,0.083333336,0.0,0.11111111,0.11111111,0.1388889,0.083333336,0.19444445,0.11111111,0.16666667,0.19444445,0.19444445,0.16666667,0.16666667,0.16666667,0.19444445,0.16666667,0.16666667,0.19444445,0.22222222,0.25,0.19444445,0.19444445,0.19444445,0.22222222,0.19444445,0.25,0.30555555,0.22222222,0.2777778,0.19444445,0.30555555,0.25,0.2777778,0.3611111,0.2777778,0.33333334,0.3888889,0.3611111,0.3888889,0.5,0.3611111,0.30555555,0.5,0.44444445,0.5555556,0.44444445,,0.6111111,0.6111111,0.5833333,0.5,0.6666667,0.5555556,0.6666667,0.6944444,0.6944444,0.6666667,0.6666667,0.75,0.75,0.75,0.8055556,0.8055556,0.8611111,0.8611111,0.9166667
+TabSTAR (tuned),0.083333336,0.1388889,0.1388889,0.16666667,0.16666667,0.1388889,0.1388889,0.19444445,0.22222222,0.25,0.19444445,0.22222222,0.22222222,0.18055555,0.22222222,0.19444445,0.2777778,0.22222222,0.25,0.25,0.25,0.25,0.22222222,0.30555555,0.2777778,0.22222222,0.33333334,0.22222222,0.33333334,0.2777778,0.22222222,0.30555555,0.3611111,0.33333334,0.33333334,0.2777778,0.2777778,0.33333334,0.3611111,0.30555555,0.25,0.33333334,0.30555555,0.3611111,0.30555555,0.3611111,0.33333334,0.3611111,0.44444445,0.3888889,0.3888889,,0.45833334,0.44444445,0.5277778,0.41666666,0.5277778,0.5555556,0.5833333,0.9444444,0.5,0.6111111,0.6111111,0.6111111,0.5555556,0.6944444,0.7222222,0.7777778,0.7777778,0.8611111
+TabSTAR (tuned + ensembled),0.11111111,0.1388889,0.16666667,0.16666667,0.16666667,0.1388889,0.1388889,0.22222222,0.22222222,0.25,0.25,0.25,0.19444445,0.19444445,0.25,0.19444445,0.2777778,0.25,0.25,0.25,0.25,0.25,0.22222222,0.30555555,0.25,0.22222222,0.30555555,0.25,0.30555555,0.2777778,0.25,0.30555555,0.3611111,0.30555555,0.30555555,0.25,0.2777778,0.2777778,0.33333334,0.30555555,0.2777778,0.33333334,0.25,0.33333334,0.33333334,0.3611111,0.33333334,0.3611111,0.41666666,0.3888889,0.3888889,0.5416667,,0.44444445,0.5277778,0.44444445,0.5277778,0.5555556,0.5555556,0.9722222,0.5277778,0.5833333,0.5833333,0.5833333,0.5555556,0.6944444,0.6944444,0.7777778,0.7777778,0.8611111
+RandomForest (tuned),0.055555556,0.027777778,0.055555556,0.083333336,0.083333336,0.055555556,0.1388889,0.11111111,0.11111111,0.055555556,0.16666667,0.1388889,0.11111111,0.16666667,0.19444445,0.22222222,0.11111111,0.083333336,0.11111111,0.083333336,0.22222222,0.25,0.11111111,0.16666667,0.16666667,0.1388889,0.2777778,0.16666667,0.2777778,0.2777778,0.22222222,0.22222222,0.2777778,0.2777778,0.22222222,0.3611111,0.22222222,0.30555555,0.2777778,0.3888889,0.25,0.33333334,0.2777778,0.19444445,0.3611111,0.3888889,0.30555555,0.41666666,0.22222222,0.41666666,0.41666666,0.5555556,0.5555556,,0.5,0.5833333,0.5555556,0.6666667,0.6666667,0.6666667,0.5277778,0.7638889,0.6944444,0.8055556,0.6111111,0.7777778,0.75,0.7777778,0.8888889,0.9722222
+PerpetualBooster (tuned + ensembled),0.0,0.027777778,0.083333336,0.027777778,0.055555556,0.055555556,0.083333336,0.055555556,0.055555556,0.055555556,0.22222222,0.19444445,0.11111111,0.19444445,0.22222222,0.19444445,0.11111111,0.1388889,0.16666667,0.11111111,0.1388889,0.22222222,0.22222222,0.25,0.16666667,0.22222222,0.22222222,0.1388889,0.25,0.22222222,0.19444445,0.2777778,0.25,0.19444445,0.30555555,0.33333334,0.22222222,0.33333334,0.30555555,0.33333334,0.33333334,0.30555555,0.2777778,0.3611111,0.30555555,0.3888889,0.3611111,0.3888889,0.4722222,0.44444445,0.5,0.4722222,0.4722222,0.5,,0.5555556,0.7361111,0.5833333,0.5277778,0.5,0.5277778,0.6388889,0.6388889,0.6111111,0.8611111,0.7777778,0.7777778,0.8055556,0.75,0.8888889
+iLTM (default),0.083333336,0.027777778,0.055555556,0.055555556,0.083333336,0.027777778,0.1388889,0.083333336,0.055555556,0.083333336,0.1388889,0.1388889,0.055555556,0.1388889,0.1388889,0.1388889,0.1388889,0.11111111,0.1388889,0.1388889,0.19444445,0.19444445,0.1388889,0.22222222,0.19444445,0.16666667,0.22222222,0.16666667,0.25,0.22222222,0.25,0.2361111,0.25,0.2777778,0.19444445,0.25,0.22222222,0.22222222,0.25,0.25,0.2777778,0.25,0.25,0.3611111,0.33333334,0.2777778,0.3611111,0.25,0.44444445,0.2777778,0.33333334,0.5833333,0.5555556,0.41666666,0.44444445,,0.41666666,0.5,0.4722222,0.6666667,0.5555556,0.5277778,0.5555556,0.75,0.5833333,0.7222222,0.7222222,0.8055556,0.8055556,0.8888889
+PerpetualBooster (tuned),0.0,0.0,0.055555556,0.0,0.027777778,0.0,0.055555556,0.027777778,0.027777778,0.027777778,0.1388889,0.1388889,0.11111111,0.19444445,0.1388889,0.11111111,0.055555556,0.11111111,0.083333336,0.11111111,0.11111111,0.16666667,0.16666667,0.19444445,0.1388889,0.16666667,0.22222222,0.083333336,0.19444445,0.19444445,0.1388889,0.16666667,0.25,0.19444445,0.22222222,0.25,0.19444445,0.2777778,0.22222222,0.33333334,0.2777778,0.30555555,0.25,0.2777778,0.2777778,0.3611111,0.33333334,0.3888889,0.3611111,0.3611111,0.44444445,0.4722222,0.4722222,0.44444445,0.2638889,0.5833333,,0.5833333,0.5,0.5,0.5277778,0.5833333,0.6111111,0.5555556,0.7638889,0.75,0.7222222,0.7777778,0.7222222,0.8888889
+xRFM (default),0.027777778,0.027777778,0.027777778,0.027777778,0.055555556,0.083333336,0.055555556,0.083333336,0.083333336,0.11111111,0.055555556,0.11111111,0.083333336,0.16666667,0.11111111,0.16666667,0.11111111,0.16666667,0.19444445,0.1388889,0.19444445,0.16666667,0.22222222,0.11111111,0.16666667,0.22222222,0.1388889,0.16666667,0.19444445,0.19444445,0.22222222,0.1388889,0.25,0.2777778,0.055555556,0.22222222,0.22222222,0.22222222,0.25,0.19444445,0.33333334,0.30555555,0.30555555,0.25,0.2777778,0.25,0.33333334,0.30555555,0.30555555,0.2777778,0.33333334,0.44444445,0.44444445,0.33333334,0.41666666,0.5,0.41666666,,0.5277778,0.6388889,0.5,0.5833333,0.5,0.5555556,0.5555556,0.6666667,0.6666667,0.7222222,0.7777778,0.9166667
+ExtraTrees (default),0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.11111111,0.055555556,0.027777778,0.027777778,0.027777778,0.055555556,0.11111111,0.11111111,0.083333336,0.19444445,0.055555556,0.11111111,0.16666667,0.1388889,0.22222222,0.1388889,0.11111111,0.083333336,0.22222222,0.1388889,0.19444445,0.19444445,0.11111111,0.22222222,0.2777778,0.16666667,0.22222222,0.2777778,0.11111111,0.2777778,0.25,0.22222222,0.2777778,0.2777778,0.22222222,0.2777778,0.30555555,0.083333336,0.30555555,0.2777778,0.22222222,0.30555555,0.22222222,0.33333334,0.30555555,0.41666666,0.44444445,0.33333334,0.4722222,0.5277778,0.5,0.4722222,,0.6388889,0.3888889,0.6111111,0.5555556,0.6111111,0.5555556,0.6666667,0.6666667,0.7222222,0.8055556,0.9444444
+TabSTAR (default),0.055555556,0.1388889,0.1388889,0.16666667,0.1388889,0.11111111,0.11111111,0.1388889,0.16666667,0.16666667,0.16666667,0.19444445,0.19444445,0.1388889,0.19444445,0.16666667,0.19444445,0.19444445,0.22222222,0.22222222,0.19444445,0.16666667,0.19444445,0.22222222,0.19444445,0.19444445,0.19444445,0.19444445,0.25,0.16666667,0.19444445,0.19444445,0.16666667,0.19444445,0.25,0.19444445,0.19444445,0.16666667,0.2777778,0.19444445,0.25,0.22222222,0.19444445,0.25,0.22222222,0.2777778,0.22222222,0.2777778,0.2777778,0.30555555,0.30555555,0.055555556,0.027777778,0.33333334,0.5,0.33333334,0.5,0.3611111,0.3611111,,0.3888889,0.44444445,0.3611111,0.5555556,0.5277778,0.5555556,0.5555556,0.6388889,0.6388889,0.8333333
+TorchMLP (default),0.027777778,0.055555556,0.055555556,0.027777778,0.055555556,0.027777778,0.055555556,0.055555556,0.083333336,0.027777778,0.1388889,0.11111111,0.055555556,0.1388889,0.1388889,0.027777778,0.055555556,0.11111111,0.083333336,0.083333336,0.027777778,0.1388889,0.11111111,0.11111111,0.1388889,0.1388889,0.19444445,0.11111111,0.22222222,0.16666667,0.16666667,0.083333336,0.19444445,0.083333336,0.11111111,0.16666667,0.16666667,0.11111111,0.055555556,0.19444445,0.19444445,0.16666667,0.2777778,0.3611111,0.083333336,0.2777778,0.41666666,0.33333334,0.33333334,0.30555555,0.33333334,0.5,0.4722222,0.4722222,0.4722222,0.44444445,0.4722222,0.5,0.6111111,0.6111111,,0.5555556,0.5277778,0.6111111,0.5833333,0.6666667,0.7222222,0.8055556,0.7777778,0.8888889
+RandomForest (default),0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.055555556,0.083333336,0.055555556,0.027777778,0.027777778,0.055555556,0.083333336,0.027777778,0.11111111,0.11111111,0.16666667,0.055555556,0.027777778,0.083333336,0.055555556,0.1388889,0.16666667,0.055555556,0.083333336,0.083333336,0.083333336,0.1388889,0.11111111,0.19444445,0.1388889,0.16666667,0.11111111,0.19444445,0.19444445,0.1388889,0.25,0.25,0.19444445,0.25,0.25,0.16666667,0.33333334,0.30555555,0.1388889,0.3611111,0.25,0.16666667,0.2777778,0.11111111,0.2777778,0.33333334,0.3888889,0.41666666,0.2361111,0.3611111,0.4722222,0.41666666,0.41666666,0.3888889,0.5555556,0.44444445,,0.5,0.6388889,0.5277778,0.7222222,0.7222222,0.75,0.8333333,0.9722222
+FastaiMLP (default),0.0,0.0,0.027777778,0.027777778,0.027777778,0.0,0.027777778,0.055555556,0.027777778,0.027777778,0.027777778,0.027777778,0.055555556,0.11111111,0.027777778,0.083333336,0.055555556,0.083333336,0.083333336,0.11111111,0.1388889,0.055555556,0.11111111,0.083333336,0.11111111,0.11111111,0.16666667,0.083333336,0.1388889,0.1388889,0.1388889,0.11111111,0.16666667,0.19444445,0.1388889,0.1388889,0.16666667,0.083333336,0.1388889,0.16666667,0.19444445,0.19444445,0.22222222,0.25,0.19444445,0.027777778,0.30555555,0.22222222,0.33333334,0.2777778,0.25,0.3888889,0.41666666,0.30555555,0.3611111,0.44444445,0.3888889,0.5,0.44444445,0.6388889,0.4722222,0.5,,0.5555556,0.6666667,0.6666667,0.6944444,0.7222222,0.7777778,0.9166667
+KNN (tuned + ensembled),0.055555556,0.027777778,0.027777778,0.027777778,0.055555556,0.083333336,0.055555556,0.083333336,0.055555556,0.083333336,0.083333336,0.027777778,0.083333336,0.083333336,0.11111111,0.11111111,0.055555556,0.055555556,0.11111111,0.1388889,0.11111111,0.083333336,0.1388889,0.083333336,0.11111111,0.11111111,0.15277778,0.11111111,0.11111111,0.11111111,0.19444445,0.083333336,0.22222222,0.19444445,0.083333336,0.11111111,0.22222222,0.11111111,0.16666667,0.11111111,0.16666667,0.1388889,0.22222222,0.055555556,0.16666667,0.19444445,0.11111111,0.2777778,0.19444445,0.2777778,0.25,0.3888889,0.41666666,0.19444445,0.3888889,0.25,0.44444445,0.44444445,0.3888889,0.44444445,0.3888889,0.3611111,0.44444445,,0.5,0.6388889,0.6388889,0.7222222,0.9722222,0.9722222
+PerpetualBooster (default),0.0,0.027777778,0.055555556,0.027777778,0.055555556,0.055555556,0.027777778,0.055555556,0.027777778,0.027777778,0.11111111,0.055555556,0.027777778,0.11111111,0.083333336,0.055555556,0.055555556,0.11111111,0.055555556,0.055555556,0.055555556,0.11111111,0.1388889,0.1388889,0.11111111,0.1388889,0.1388889,0.055555556,0.19444445,0.16666667,0.16666667,0.11111111,0.1388889,0.11111111,0.083333336,0.11111111,0.16666667,0.083333336,0.11111111,0.1388889,0.16666667,0.11111111,0.19444445,0.25,0.16666667,0.1388889,0.2777778,0.25,0.3611111,0.19444445,0.25,0.44444445,0.44444445,0.3888889,0.1388889,0.41666666,0.2361111,0.44444445,0.44444445,0.4722222,0.41666666,0.4722222,0.33333334,0.5,,0.6388889,0.6388889,0.6944444,0.6111111,0.8333333
+Linear (tuned + ensembled),0.027777778,0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.083333336,0.083333336,0.055555556,0.055555556,0.11111111,0.055555556,0.083333336,0.027777778,0.027777778,0.027777778,0.083333336,0.083333336,0.083333336,0.055555556,0.083333336,0.11111111,0.16666667,0.055555556,0.22222222,0.083333336,0.083333336,0.1388889,0.083333336,0.055555556,0.11111111,0.027777778,0.1388889,0.083333336,0.11111111,0.083333336,0.11111111,0.055555556,0.1388889,0.16666667,0.19444445,0.16666667,0.25,0.22222222,0.22222222,0.19444445,0.19444445,0.30555555,0.30555555,0.22222222,0.22222222,0.2777778,0.25,0.33333334,0.33333334,0.44444445,0.33333334,0.2777778,0.33333334,0.3611111,0.3611111,,0.7083333,0.7222222,0.5277778,0.75
+Linear (tuned),0.027777778,0.027777778,0.027777778,0.027777778,0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.083333336,0.083333336,0.041666668,0.055555556,0.11111111,0.055555556,0.055555556,0.027777778,0.027777778,0.027777778,0.083333336,0.083333336,0.055555556,0.027777778,0.083333336,0.083333336,0.1388889,0.027777778,0.19444445,0.055555556,0.055555556,0.11111111,0.11111111,0.055555556,0.1388889,0.027777778,0.1388889,0.083333336,0.083333336,0.055555556,0.11111111,0.055555556,0.1388889,0.1388889,0.16666667,0.1388889,0.25,0.22222222,0.25,0.16666667,0.19444445,0.2777778,0.30555555,0.25,0.22222222,0.2777778,0.2777778,0.33333334,0.33333334,0.44444445,0.2777778,0.2777778,0.30555555,0.3611111,0.3611111,0.29166666,,0.6388889,0.5,0.7222222
+Linear (default),0.0,0.0,0.027777778,0.027777778,0.055555556,0.0,0.0,0.055555556,0.027777778,0.0,0.11111111,0.083333336,0.055555556,0.027777778,0.11111111,0.055555556,0.055555556,0.055555556,0.027777778,0.027777778,0.055555556,0.083333336,0.055555556,0.055555556,0.083333336,0.055555556,0.16666667,0.027777778,0.1388889,0.083333336,0.027777778,0.11111111,0.1388889,0.083333336,0.11111111,0.027777778,0.055555556,0.027777778,0.027777778,0.083333336,0.11111111,0.027777778,0.027777778,0.16666667,0.083333336,0.11111111,0.16666667,0.11111111,0.19444445,0.1388889,0.1388889,0.22222222,0.22222222,0.22222222,0.19444445,0.19444445,0.22222222,0.2777778,0.2777778,0.3611111,0.19444445,0.25,0.2777778,0.2777778,0.30555555,0.2777778,0.3611111,,0.4722222,0.6666667
+KNN (tuned),0.027777778,0.027777778,0.0,0.027777778,0.055555556,0.027777778,0.027777778,0.055555556,0.027777778,0.0,0.055555556,0.027777778,0.027777778,0.083333336,0.055555556,0.0,0.055555556,0.0,0.027777778,0.055555556,0.027777778,0.027777778,0.083333336,0.027777778,0.055555556,0.083333336,0.1388889,0.027777778,0.055555556,0.083333336,0.055555556,0.027777778,0.11111111,0.027777778,0.027777778,0.083333336,0.083333336,0.055555556,0.083333336,0.083333336,0.083333336,0.083333336,0.055555556,0.055555556,0.083333336,0.11111111,0.027777778,0.083333336,0.083333336,0.1388889,0.1388889,0.22222222,0.22222222,0.11111111,0.25,0.19444445,0.2777778,0.22222222,0.19444445,0.3611111,0.22222222,0.16666667,0.22222222,0.027777778,0.3888889,0.4722222,0.5,0.5277778,,0.8888889
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.027777778,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.027777778,0.0,0.0,0.0,0.0,0.0,0.027777778,0.0,0.0,0.027777778,0.0,0.0,0.027777778,0.0,0.027777778,0.0,0.0,0.0,0.0,0.055555556,0.0,0.027777778,0.0,0.0,0.055555556,0.027777778,0.027777778,0.027777778,0.027777778,0.0,0.055555556,0.055555556,0.027777778,0.083333336,0.027777778,0.055555556,0.083333336,0.1388889,0.1388889,0.027777778,0.11111111,0.11111111,0.11111111,0.083333336,0.055555556,0.16666667,0.11111111,0.027777778,0.083333336,0.027777778,0.16666667,0.25,0.2777778,0.33333334,0.11111111,
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_all/leaderboard_overview_explorer.html
index 4b042f36f7fa6ed5a9646cc4ec9bcf68cc99305c..e998078b0fa45ed5ecf1796d4af5d9b38746eeb7 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Binary | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_all/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_binary/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..14b323d468bf1ffbb9c55d2c44157b7c56ea4f71
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_binary/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1780.0,"score":0.8,"rank":6.23,"hrank":1.69,"imp":5.067,"train_time":37.56,"predict_time":7.056,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+162\/-122","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1642.0,"score":0.602,"rank":11.05,"hrank":4.33,"imp":10.219,"train_time":2.9,"predict_time":0.353,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+116\/-77","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1624.0,"score":0.601,"rank":11.9,"hrank":3.88,"imp":7.674,"train_time":270.87,"predict_time":4.72,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+148\/-107","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1592.0,"score":0.525,"rank":13.43,"hrank":7.41,"imp":11.827,"train_time":4.82,"predict_time":0.467,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-75","imputed":false},{"position":4,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1588.0,"score":0.518,"rank":13.63,"hrank":6.99,"imp":11.574,"train_time":1.77,"predict_time":0.16,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-83","imputed":false},{"position":5,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1556.0,"score":0.475,"rank":15.37,"hrank":6.21,"imp":12.16,"train_time":1231.24,"predict_time":1.359,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-76","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1552.0,"score":0.465,"rank":15.58,"hrank":7.55,"imp":12.254,"train_time":1878.36,"predict_time":10.744,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-72","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1547.0,"score":0.456,"rank":15.88,"hrank":9.16,"imp":12.565,"train_time":4.79,"predict_time":0.708,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-54","imputed":false},{"position":8,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1520.0,"score":0.444,"rank":17.53,"hrank":9.33,"imp":13.0,"train_time":1878.36,"predict_time":1.035,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+100\/-59","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1500.0,"score":0.4,"rank":18.77,"hrank":12.56,"imp":13.615,"train_time":2722.13,"predict_time":9.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-57","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1428.0,"score":0.339,"rank":23.78,"hrank":11.84,"imp":15.292,"train_time":2722.13,"predict_time":0.265,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-73","imputed":false},{"position":11,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1420.0,"score":0.326,"rank":24.37,"hrank":10.84,"imp":15.436,"train_time":1641.74,"predict_time":1.602,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-68","imputed":false},{"position":12,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1420.0,"score":0.32,"rank":24.38,"hrank":13.25,"imp":15.416,"train_time":8274.67,"predict_time":361.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-74","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1399.0,"score":0.304,"rank":25.92,"hrank":11.28,"imp":15.58,"train_time":1.48,"predict_time":0.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-74","imputed":false},{"position":14,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1388.0,"score":0.278,"rank":26.73,"hrank":13.32,"imp":15.744,"train_time":735.46,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-62","imputed":false},{"position":15,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1388.0,"score":0.241,"rank":26.77,"hrank":20.31,"imp":16.336,"train_time":328.64,"predict_time":0.773,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-50","imputed":false},{"position":16,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1380.0,"score":0.334,"rank":27.37,"hrank":8.14,"imp":16.736,"train_time":14.27,"predict_time":4.304,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-96","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1379.0,"score":0.26,"rank":27.48,"hrank":16.72,"imp":15.999,"train_time":735.46,"predict_time":0.301,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-73","imputed":false},{"position":18,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1379.0,"score":0.256,"rank":27.5,"hrank":17.13,"imp":15.907,"train_time":3.18,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-46","imputed":false},{"position":19,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1378.0,"score":0.275,"rank":27.58,"hrank":18.2,"imp":17.001,"train_time":275.93,"predict_time":0.306,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-72","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1374.0,"score":0.253,"rank":27.82,"hrank":17.94,"imp":16.831,"train_time":463.01,"predict_time":0.613,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-70","imputed":false},{"position":21,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1374.0,"score":0.27,"rank":27.87,"hrank":11.46,"imp":15.099,"train_time":5891.3,"predict_time":409.932,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-61","imputed":false},{"position":22,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1373.0,"score":0.279,"rank":27.95,"hrank":14.79,"imp":16.433,"train_time":1641.74,"predict_time":0.219,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-82","imputed":false},{"position":23,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1373.0,"score":0.27,"rank":27.97,"hrank":14.11,"imp":16.478,"train_time":3436.74,"predict_time":0.405,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-77","imputed":false},{"position":24,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1363.0,"score":0.249,"rank":28.75,"hrank":16.25,"imp":16.427,"train_time":1.11,"predict_time":0.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-75","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1354.0,"score":0.28,"rank":29.45,"hrank":15.94,"imp":16.635,"train_time":3436.74,"predict_time":8.476,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-94","imputed":false},{"position":26,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1345.0,"score":0.24,"rank":30.13,"hrank":16.74,"imp":17.635,"train_time":275.93,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-78","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1338.0,"score":0.221,"rank":30.67,"hrank":13.28,"imp":17.159,"train_time":463.01,"predict_time":0.108,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-75","imputed":false},{"position":28,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1336.0,"score":0.225,"rank":30.83,"hrank":19.71,"imp":18.153,"train_time":248.52,"predict_time":1.699,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-64","imputed":false},{"position":29,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1334.0,"score":0.229,"rank":30.98,"hrank":16.41,"imp":16.944,"train_time":5891.3,"predict_time":39.564,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-73","imputed":false},{"position":30,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1326.0,"score":0.19,"rank":31.7,"hrank":26.19,"imp":17.465,"train_time":328.64,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-60","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1322.0,"score":0.188,"rank":32.0,"hrank":21.43,"imp":17.186,"train_time":8274.67,"predict_time":40.268,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-54","imputed":false},{"position":32,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1310.0,"score":0.149,"rank":32.97,"hrank":23.11,"imp":18.377,"train_time":976.68,"predict_time":0.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-58","imputed":false},{"position":33,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1301.0,"score":0.137,"rank":33.73,"hrank":24.29,"imp":17.837,"train_time":9.04,"predict_time":1.162,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-51","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1300.0,"score":0.179,"rank":33.78,"hrank":22.21,"imp":17.983,"train_time":5.88,"predict_time":0.202,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-74","imputed":false},{"position":35,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1298.0,"score":0.198,"rank":33.93,"hrank":24.53,"imp":17.359,"train_time":559.77,"predict_time":1.513,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-82","imputed":false},{"position":36,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1295.0,"score":0.174,"rank":34.17,"hrank":25.26,"imp":18.418,"train_time":1.52,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-72","imputed":false},{"position":37,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1282.0,"score":0.154,"rank":35.3,"hrank":26.95,"imp":17.847,"train_time":2206.64,"predict_time":2.301,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-53","imputed":false},{"position":38,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1262.0,"score":0.144,"rank":36.9,"hrank":28.42,"imp":18.927,"train_time":559.77,"predict_time":0.086,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-68","imputed":false},{"position":39,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1255.0,"score":0.12,"rank":37.45,"hrank":15.79,"imp":19.797,"train_time":2.84,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-57","imputed":false},{"position":40,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1253.0,"score":0.089,"rank":37.6,"hrank":27.13,"imp":19.205,"train_time":976.68,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-54","imputed":false},{"position":41,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1245.0,"score":0.16,"rank":38.27,"hrank":23.25,"imp":20.921,"train_time":13.07,"predict_time":3.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-79","imputed":false},{"position":42,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1224.0,"score":0.136,"rank":40.02,"hrank":24.88,"imp":19.883,"train_time":47.89,"predict_time":39.444,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-85","imputed":false},{"position":43,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1218.0,"score":0.112,"rank":40.48,"hrank":30.75,"imp":20.386,"train_time":561.29,"predict_time":4.462,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-92","imputed":false},{"position":44,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1208.0,"score":0.13,"rank":41.3,"hrank":16.42,"imp":20.076,"train_time":1.42,"predict_time":0.118,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-79","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1205.0,"score":0.113,"rank":41.52,"hrank":33.91,"imp":20.164,"train_time":2206.64,"predict_time":0.107,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-66","imputed":false},{"position":46,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1205.0,"score":0.105,"rank":41.57,"hrank":14.81,"imp":20.996,"train_time":11.15,"predict_time":0.312,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-64","imputed":false},{"position":47,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1170.0,"score":0.101,"rank":44.35,"hrank":23.94,"imp":22.77,"train_time":202.68,"predict_time":44.667,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-82","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1168.0,"score":0.064,"rank":44.52,"hrank":37.8,"imp":21.201,"train_time":180.03,"predict_time":0.779,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-62","imputed":false},{"position":49,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1146.0,"score":0.06,"rank":46.2,"hrank":36.48,"imp":22.428,"train_time":180.03,"predict_time":0.056,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-92","imputed":false},{"position":50,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1144.0,"score":0.068,"rank":46.35,"hrank":40.14,"imp":20.886,"train_time":0.93,"predict_time":0.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-69","imputed":false},{"position":51,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1143.0,"score":0.079,"rank":46.4,"hrank":31.27,"imp":22.388,"train_time":561.29,"predict_time":0.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-108","imputed":false},{"position":52,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1143.0,"score":0.145,"rank":46.45,"hrank":15.48,"imp":24.124,"train_time":34925.3,"predict_time":3.663,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-110","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1142.0,"score":0.05,"rank":46.53,"hrank":40.82,"imp":22.608,"train_time":171.61,"predict_time":0.552,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-77","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1133.0,"score":0.138,"rank":47.2,"hrank":17.08,"imp":24.314,"train_time":34925.3,"predict_time":16.932,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-115","imputed":false},{"position":55,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1109.0,"score":0.012,"rank":48.98,"hrank":45.51,"imp":23.183,"train_time":171.61,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-69","imputed":false},{"position":56,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1108.0,"score":0.095,"rank":49.03,"hrank":37.56,"imp":25.257,"train_time":130.48,"predict_time":0.498,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-116","imputed":false},{"position":57,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1090.0,"score":0.05,"rank":50.33,"hrank":42.34,"imp":23.907,"train_time":14.37,"predict_time":2.74,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-79","imputed":false},{"position":58,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1058.0,"score":0.052,"rank":52.6,"hrank":46.98,"imp":26.205,"train_time":130.48,"predict_time":0.143,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-102","imputed":false},{"position":59,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1057.0,"score":0.111,"rank":52.7,"hrank":14.87,"imp":29.295,"train_time":387.37,"predict_time":4.09,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-141","imputed":false},{"position":60,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1049.0,"score":0.025,"rank":53.2,"hrank":47.38,"imp":24.505,"train_time":4.97,"predict_time":0.092,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-71","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1020.0,"score":0.029,"rank":55.1,"hrank":50.72,"imp":25.064,"train_time":2.79,"predict_time":0.259,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-92","imputed":false},{"position":62,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1010.0,"score":0.053,"rank":55.73,"hrank":45.54,"imp":29.485,"train_time":0.81,"predict_time":0.115,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-122","imputed":false},{"position":63,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.02,"rank":56.35,"hrank":44.83,"imp":27.216,"train_time":0.29,"predict_time":0.031,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-104","imputed":false},{"position":64,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":998.0,"score":0.041,"rank":56.47,"hrank":46.38,"imp":27.0,"train_time":1.15,"predict_time":0.22,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-118","imputed":false},{"position":65,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":989.0,"score":0.007,"rank":57.02,"hrank":54.64,"imp":26.968,"train_time":140.69,"predict_time":3.209,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-89","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":982.0,"score":0.041,"rank":57.4,"hrank":29.2,"imp":31.946,"train_time":12.44,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-121","imputed":false},{"position":67,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":980.0,"score":0.038,"rank":57.57,"hrank":19.84,"imp":30.496,"train_time":155.53,"predict_time":0.277,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-114","imputed":false},{"position":68,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":955.0,"score":0.029,"rank":58.95,"hrank":30.18,"imp":31.102,"train_time":155.53,"predict_time":0.05,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-120","imputed":false},{"position":69,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":943.0,"score":0.033,"rank":59.6,"hrank":53.49,"imp":30.394,"train_time":2.71,"predict_time":0.392,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-108","imputed":false},{"position":70,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":914.0,"score":0.01,"rank":61.03,"hrank":56.51,"imp":32.782,"train_time":0.8,"predict_time":0.088,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-106","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":840.0,"score":0.0,"rank":64.3,"hrank":63.59,"imp":32.964,"train_time":140.69,"predict_time":0.188,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-98","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":619.0,"score":0.0,"rank":70.27,"hrank":69.93,"imp":45.705,"train_time":0.15,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-165","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_all/pareto_front_explorer.html
index 530d5eb3a441578623539a6a12ed87874517230f..f643749f6489b7153b63ca98c49114e69caee70a 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.6993924491,"imp":18.1526673656,"elo":1336.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0209339609,"imp":15.9074321593,"elo":1378.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0271393798,"imp":15.7444297863,"elo":1388.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3007363462,"imp":15.9985823163,"elo":1378.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0303915587,"imp":18.4183228609,"elo":1295.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0257224855,"imp":17.6349039127,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3055895504,"imp":17.0012559871,"elo":1377.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0149256706,"imp":19.7969955167,"elo":1255.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0149447918,"imp":19.2050209554,"elo":1253.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1399325633,"imp":18.3772674793,"elo":1310.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2195070167,"imp":27.0004600991,"elo":998.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0555212696,"imp":22.4284851072,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7789141111,"imp":21.2009869851,"elo":1167.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2592580681,"imp":25.0643278631,"elo":1020.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2364230333,"imp":22.3879244871,"elo":1143.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4616886326,"imp":20.3857612497,"elo":1218.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.094055721,"imp":20.886067042,"elo":1143.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0890724278,"imp":17.4649033005,"elo":1325.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7734248689,"imp":16.3364190555,"elo":1388.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.3036225459,"imp":16.7357629181,"elo":1380.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3117928723,"imp":20.9960647526,"elo":1204.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4051957243,"imp":16.4777604777,"elo":1372.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4763406294,"imp":16.6346698606,"elo":1353.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7399372917,"imp":23.9070580073,"elo":1090.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0268138653,"imp":31.94603867,"elo":982.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1426565145,"imp":26.2053049427,"elo":1058.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4978820663,"imp":25.2567439478,"elo":1108.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0312580102,"imp":27.216434118,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0526216987,"imp":23.1834097224,"elo":1109.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5516567504,"imp":22.6078997509,"elo":1141.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.1617619189,"imp":17.8365665548,"elo":1300.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.2645206496,"imp":15.2915746154,"elo":1427.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":9.3100869738,"imp":13.6154985064,"elo":1500.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7081268775,"imp":12.565425149,"elo":1547.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0352207543,"imp":13.0003571839,"elo":1519.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.7444053479,"imp":12.254317466,"elo":1552.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.1690613568,"imp":20.9214883202,"elo":1245.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"imp":19.8826336829,"elo":1223.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.5640473457,"imp":16.9438616487,"elo":1334.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":409.9318208736,"imp":15.0990305135,"elo":1373.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1668648195,"imp":15.5800907751,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"imp":5.0669585426,"elo":1779.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.115458016,"imp":29.4853801163,"elo":1010.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1600398508,"imp":11.573972439,"elo":1588.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2024528313,"imp":17.9828626737,"elo":1300.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.218906887,"imp":16.4329224991,"elo":1372.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.6019358162,"imp":15.4363074544,"elo":1419.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4673275113,"imp":11.8268490906,"elo":1592.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.353062543,"imp":10.2186856195,"elo":1642.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0895012297,"imp":29.295471998,"elo":1056.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6626595929,"imp":24.1243602213,"elo":1142.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.9317836231,"imp":24.3141702212,"elo":1132.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622971538,"imp":16.4265396533,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0923536337,"imp":24.5049712598,"elo":1049.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1067953784,"imp":20.1643112119,"elo":1205.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.3011998096,"imp":17.8468237042,"elo":1281.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1176002559,"imp":20.0764583832,"elo":1208.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1079666958,"imp":17.1587159787,"elo":1338.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6132840031,"imp":16.8310805968,"elo":1374.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":44.6673844457,"imp":22.7695795734,"elo":1169.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":40.2677406188,"imp":17.1855068868,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":361.1026011106,"imp":15.4163566508,"elo":1419.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.3922735327,"imp":30.3938445671,"elo":942.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0860477633,"imp":18.9268719252,"elo":1262.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.5127555025,"imp":17.3587983717,"elo":1298.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.6993924491,"x_train":248.5156197629,"imp":18.1526673656,"elo":1336.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0209339609,"x_train":3.1841367928,"imp":15.9074321593,"elo":1378.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0271393798,"x_train":735.4606570949,"imp":15.7444297863,"elo":1388.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3007363462,"x_train":735.4606570949,"imp":15.9985823163,"elo":1378.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0303915587,"x_train":1.5249213808,"imp":18.4183228609,"elo":1295.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0257224855,"x_train":275.9265613682,"imp":17.6349039127,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3055895504,"x_train":275.9265613682,"imp":17.0012559871,"elo":1377.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0149256706,"x_train":2.8412925593,"imp":19.7969955167,"elo":1255.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0149447918,"x_train":976.6816362244,"imp":19.2050209554,"elo":1253.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1399325633,"x_train":976.6816362244,"imp":18.3772674793,"elo":1310.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2195070167,"x_train":1.1488986663,"imp":27.0004600991,"elo":998.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0555212696,"x_train":180.0329734893,"imp":22.4284851072,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7789141111,"x_train":180.0329734893,"imp":21.2009869851,"elo":1167.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2592580681,"x_train":2.7868984981,"imp":25.0643278631,"elo":1020.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2364230333,"x_train":561.2940883157,"imp":22.3879244871,"elo":1143.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4616886326,"x_train":561.2940883157,"imp":20.3857612497,"elo":1218.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.094055721,"x_train":0.9329196027,"imp":20.886067042,"elo":1143.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0890724278,"x_train":328.641625286,"imp":17.4649033005,"elo":1325.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7734248689,"x_train":328.641625286,"imp":16.3364190555,"elo":1388.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.3036225459,"x_train":14.2669009183,"imp":16.7357629181,"elo":1380.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3117928723,"x_train":11.1478855018,"imp":20.9960647526,"elo":1204.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4051957243,"x_train":3436.7433273226,"imp":16.4777604777,"elo":1372.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4763406294,"x_train":3436.7433273226,"imp":16.6346698606,"elo":1353.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7399372917,"x_train":14.3691534377,"imp":23.9070580073,"elo":1090.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0268138653,"x_train":12.4370351532,"imp":31.94603867,"elo":982.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1426565145,"x_train":130.4829960164,"imp":26.2053049427,"elo":1058.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4978820663,"x_train":130.4829960164,"imp":25.2567439478,"elo":1108.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0312580102,"x_train":0.2940996202,"imp":27.216434118,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0526216987,"x_train":171.6065412889,"imp":23.1834097224,"elo":1109.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5516567504,"x_train":171.6065412889,"imp":22.6078997509,"elo":1141.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.1617619189,"x_train":9.0446607596,"imp":17.8365665548,"elo":1300.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.2645206496,"x_train":2722.1286706272,"imp":15.2915746154,"elo":1427.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":9.3100869738,"x_train":2722.1286706272,"imp":13.6154985064,"elo":1500.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7081268775,"x_train":4.7880152698,"imp":12.565425149,"elo":1547.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0352207543,"x_train":1878.3590992619,"imp":13.0003571839,"elo":1519.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.7444053479,"x_train":1878.3590992619,"imp":12.254317466,"elo":1552.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.1690613568,"x_train":13.0741284928,"imp":20.9214883202,"elo":1245.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"x_train":47.8871994,"imp":19.8826336829,"elo":1223.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.5640473457,"x_train":5891.3014300686,"imp":16.9438616487,"elo":1334.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":409.9318208736,"x_train":5891.3014300686,"imp":15.0990305135,"elo":1373.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1668648195,"x_train":1.4816805767,"imp":15.5800907751,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"x_train":37.5608528654,"imp":5.0669585426,"elo":1779.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.115458016,"x_train":0.8137451985,"imp":29.4853801163,"elo":1010.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1600398508,"x_train":1.7694169027,"imp":11.573972439,"elo":1588.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2024528313,"x_train":5.8795969926,"imp":17.9828626737,"elo":1300.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.218906887,"x_train":1641.7401040643,"imp":16.4329224991,"elo":1372.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.6019358162,"x_train":1641.7401040643,"imp":15.4363074544,"elo":1419.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4673275113,"x_train":4.8183344075,"imp":11.8268490906,"elo":1592.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.353062543,"x_train":2.9028568606,"imp":10.2186856195,"elo":1642.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0895012297,"x_train":387.3713684902,"imp":29.295471998,"elo":1056.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6626595929,"x_train":34925.3007391894,"imp":24.1243602213,"elo":1142.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.9317836231,"x_train":34925.3007391894,"imp":24.3141702212,"elo":1132.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622971538,"x_train":1.1067907259,"imp":16.4265396533,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0923536337,"x_train":4.9663458274,"imp":24.5049712598,"elo":1049.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1067953784,"x_train":2206.6390156442,"imp":20.1643112119,"elo":1205.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.3011998096,"x_train":2206.6390156442,"imp":17.8468237042,"elo":1281.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1176002559,"x_train":1.4234367091,"imp":20.0764583832,"elo":1208.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1079666958,"x_train":463.012192935,"imp":17.1587159787,"elo":1338.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6132840031,"x_train":463.012192935,"imp":16.8310805968,"elo":1374.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":44.6673844457,"x_train":202.6812316751,"imp":22.7695795734,"elo":1169.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":40.2677406188,"x_train":8274.6664503747,"imp":17.1855068868,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":361.1026011106,"x_train":8274.6664503747,"imp":15.4163566508,"elo":1419.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.3922735327,"x_train":2.7144283593,"imp":30.3938445671,"elo":942.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0860477633,"x_train":559.7696517925,"imp":18.9268719252,"elo":1262.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.5127555025,"x_train":559.7696517925,"imp":17.3587983717,"elo":1298.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_all/tuning_trajectories_explorer.html
index 37e48449d79ec06e17bcfaec8013bd9ea650b1a9..330c14ec898c36fa9bd0ad7706f125fd58291e51 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":248.5156197629,"imp":18.1985220089,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.1841367928,"imp":15.9472716281,"elo":1403.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":7.6931563125,"imp":15.9170100606,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":18.2301485936,"imp":15.8820288307,"elo":1407.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":36.373457126,"imp":15.8755588014,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":93.5688559171,"imp":15.9358403304,"elo":1413.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":183.1491958089,"imp":15.9609986119,"elo":1409.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":366.895128245,"imp":15.9883637976,"elo":1404.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":546.7577277937,"imp":16.0568034741,"elo":1398.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":735.4606570949,"imp":16.0209685048,"elo":1402.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.5249213808,"imp":18.4627793194,"elo":1317.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.9231596776,"imp":17.7821610613,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":6.9435649233,"imp":17.3668963083,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":13.7562217845,"imp":17.2036229582,"elo":1394.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":34.4671973421,"imp":17.0401788819,"elo":1406.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":68.8507318629,"imp":16.9204698482,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":137.5904273763,"imp":16.9565809554,"elo":1410.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":206.0245082163,"imp":16.9644016109,"elo":1411.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":275.9265613682,"imp":17.0365172277,"elo":1409.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.8412925593,"imp":19.8340051574,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":8.2691673892,"imp":19.3189290715,"elo":1291.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":21.1083282561,"imp":18.9136277826,"elo":1309.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":45.3683328739,"imp":18.7744647151,"elo":1316.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":119.8161001058,"imp":18.63644326,"elo":1326.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":243.8194356466,"imp":18.4985673732,"elo":1334.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":486.6455896663,"imp":18.411250566,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":730.1621818496,"imp":18.4178680008,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":976.6816362244,"imp":18.4045355559,"elo":1335.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.1488986663,"imp":27.0306011122,"elo":1010.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.0419356023,"imp":25.365172147,"elo":1035.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.7437750172,"imp":22.8965916233,"elo":1097.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.1794740447,"imp":22.0312588622,"elo":1128.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":22.6140856772,"imp":21.6232779832,"elo":1147.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":44.8789786871,"imp":21.3484690839,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":89.703830225,"imp":21.1367244158,"elo":1172.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":134.3807181269,"imp":21.030360209,"elo":1181.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":180.0329734893,"imp":21.2773019509,"elo":1179.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.7868984981,"imp":25.0928001922,"elo":1031.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.47397211,"imp":24.0294000702,"elo":1087.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.1482286155,"imp":22.2738545306,"elo":1155.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":28.5676512334,"imp":21.5347757801,"elo":1181.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":68.1068582413,"imp":20.7714167021,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":137.1828565728,"imp":20.4959094686,"elo":1216.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":277.9218393643,"imp":20.3891344664,"elo":1220.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":418.1539061688,"imp":20.327550808,"elo":1223.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":561.2940883157,"imp":20.4322580444,"elo":1230.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.9329196027,"imp":20.9267620345,"elo":1152.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":2.8974631618,"imp":18.6507312542,"elo":1271.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":7.4728798139,"imp":17.2900290339,"elo":1354.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":15.0938578715,"imp":16.7588564136,"elo":1387.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":39.6915504803,"imp":16.4615228234,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":80.8366710185,"imp":16.3916314277,"elo":1411.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":163.2701936469,"imp":16.3492689805,"elo":1414.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":243.2687421878,"imp":16.2947962091,"elo":1416.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":328.641625286,"imp":16.3758834982,"elo":1416.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":14.2669009183,"imp":16.9328597325,"elo":1398.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":11.1478855018,"imp":21.0320505117,"elo":1215.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":27.3629676777,"imp":19.6248252792,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":84.0873327564,"imp":17.2335333114,"elo":1354.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":169.5069286032,"imp":16.7511552172,"elo":1377.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":413.8881530141,"imp":16.2300532092,"elo":1403.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":856.7378065106,"imp":16.3775739161,"elo":1393.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1708.0482807143,"imp":16.704966384,"elo":1382.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":2544.0473436332,"imp":16.7462245783,"elo":1380.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":3436.7433273226,"imp":16.7020406807,"elo":1381.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":14.3691534377,"imp":23.7395224372,"elo":1111.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":12.4370351532,"imp":31.9817108192,"elo":997.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":39.6579418967,"imp":29.2683791238,"elo":1049.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":109.2939161538,"imp":25.4782347843,"elo":1117.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":130.4829960164,"imp":25.2603152191,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.2940996202,"imp":27.2505839036,"elo":999.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":1.108342455,"imp":24.962617101,"elo":1053.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":3.7342981222,"imp":23.8714038074,"elo":1092.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":8.0549426531,"imp":23.3960853518,"elo":1113.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":20.7189526054,"imp":22.9121609339,"elo":1138.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":41.766842473,"imp":22.8002411401,"elo":1146.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":84.379284379,"imp":22.7152276753,"elo":1151.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":127.399603328,"imp":22.7005586016,"elo":1149.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":171.6065412889,"imp":22.6412060902,"elo":1153.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":9.0446607596,"imp":17.867856331,"elo":1311.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":23.323034735,"imp":16.1509151475,"elo":1401.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":62.8372296721,"imp":14.8626054677,"elo":1471.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":131.5115032503,"imp":14.480253057,"elo":1485.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":336.6014034599,"imp":14.014604163,"elo":1509.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":679.3497815723,"imp":13.8066685751,"elo":1520.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1355.7488156461,"imp":13.7543913886,"elo":1521.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2029.6864482209,"imp":13.7072998548,"elo":1532.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2722.1286706272,"imp":13.6637450519,"elo":1536.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":4.7880152698,"imp":12.6135473201,"elo":1562.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.4527651431,"imp":12.6841519322,"elo":1561.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":44.0076788597,"imp":12.6705328358,"elo":1551.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":89.9272230443,"imp":12.7074779191,"elo":1550.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":233.2858308743,"imp":12.6413300232,"elo":1554.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":464.3616718274,"imp":12.757870963,"elo":1549.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":937.3819334882,"imp":12.470877381,"elo":1563.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1406.5304544301,"imp":12.4124945186,"elo":1571.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1878.3590992619,"imp":12.3310444289,"elo":1574.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.0741284928,"imp":20.9545459287,"elo":1259.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":47.8871994,"imp":19.9125485476,"elo":1236.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":76.8228235044,"imp":18.5578882422,"elo":1281.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":154.1734218688,"imp":17.4088810373,"elo":1332.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":288.3811437278,"imp":16.9835849175,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":734.9046229398,"imp":16.1744857897,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1475.2401534104,"imp":15.59877895,"elo":1381.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2935.4601233793,"imp":15.3293650407,"elo":1391.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":4395.9302046427,"imp":15.2057520494,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":5891.3014300686,"imp":15.1608298999,"elo":1398.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":1.4816805767,"imp":15.6160558805,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":37.5608528654,"imp":5.1109281298,"elo":1804.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.8137451985,"imp":29.5080849433,"elo":1029.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.7694169027,"imp":11.6209007657,"elo":1613.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":5.8795969926,"imp":18.0216258802,"elo":1321.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":13.3369957806,"imp":17.7011061532,"elo":1333.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":38.7568509497,"imp":17.0967272132,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":78.6946304538,"imp":16.5213683433,"elo":1395.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":201.4443768551,"imp":16.0486235788,"elo":1418.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":405.7219066472,"imp":15.7775555347,"elo":1433.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":807.6777264234,"imp":15.5759684234,"elo":1442.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1220.4156808351,"imp":15.5116692814,"elo":1450.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1641.7401040643,"imp":15.4831647745,"elo":1451.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":4.8183344075,"imp":11.8715408954,"elo":1625.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":2.9028568606,"imp":10.1975319938,"elo":1663.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":387.3713684902,"imp":29.3154901806,"elo":1070.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":619.0654872096,"imp":29.1977312744,"elo":1076.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1126.9710593518,"imp":28.8728516091,"elo":1081.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1799.6745277943,"imp":28.1782196434,"elo":1092.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4218.8944407454,"imp":26.8298020948,"elo":1108.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":8667.7234636881,"imp":25.6070686102,"elo":1124.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":17722.5911655635,"imp":25.1176657899,"elo":1132.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":26217.887217876,"imp":24.7184562266,"elo":1144.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34925.3007391894,"imp":24.3478310533,"elo":1157.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1067907259,"imp":16.4725095359,"elo":1388.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":4.9663458274,"imp":24.5208426349,"elo":1061.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.872258751,"imp":22.0196854759,"elo":1135.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":43.9083852063,"imp":19.9665463941,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":107.4432911334,"imp":19.1704504782,"elo":1232.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":267.8896031045,"imp":18.5195106616,"elo":1268.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":552.8424768764,"imp":18.1964781748,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1101.0843510949,"imp":18.1135563983,"elo":1286.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1649.8652006608,"imp":17.9664380859,"elo":1297.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2206.6390156442,"imp":17.8808006095,"elo":1300.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.4234367091,"imp":20.1187535248,"elo":1221.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":4.1993942212,"imp":18.1469125856,"elo":1322.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":11.4672731767,"imp":17.6022222434,"elo":1356.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":22.2498464852,"imp":17.2959121949,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":57.7997562863,"imp":17.0556637181,"elo":1383.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":114.5664011196,"imp":17.0011313401,"elo":1391.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":229.0218261459,"imp":16.9219889297,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":343.7649500386,"imp":16.8413473575,"elo":1404.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":463.012192935,"imp":16.864465112,"elo":1404.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":202.6812316751,"imp":22.8028436384,"elo":1186.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":512.7516159338,"imp":20.2295926298,"elo":1258.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":1466.1655464244,"imp":17.3935983195,"elo":1346.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":3194.8705446357,"imp":16.320058077,"elo":1395.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":7908.2276811319,"imp":15.4224116209,"elo":1445.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":8274.6664503747,"imp":15.4708391742,"elo":1444.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.7144283593,"imp":30.4180479527,"elo":959.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.4835528309,"imp":25.8369407868,"elo":1042.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":13.8167338689,"imp":20.4517745654,"elo":1194.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":27.7848683857,"imp":18.4081761487,"elo":1285.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":69.6362065708,"imp":17.9003166779,"elo":1311.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":138.9025571316,"imp":17.6804750529,"elo":1318.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":278.1866797416,"imp":17.5179980924,"elo":1317.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":417.6393154275,"imp":17.4366108776,"elo":1317.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":559.7696517925,"imp":17.285836362,"elo":1318.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.6993924491,"x_train":248.5156197629,"imp":18.1985220089,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0209339609,"x_train":3.1841367928,"imp":15.9472716281,"elo":1403.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0431790436,"x_train":7.6931563125,"imp":15.9170100606,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0794519096,"x_train":18.2301485936,"imp":15.8820288307,"elo":1407.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1149153864,"x_train":36.373457126,"imp":15.8755588014,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1611888807,"x_train":93.5688559171,"imp":15.9358403304,"elo":1413.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1821892961,"x_train":183.1491958089,"imp":15.9609986119,"elo":1409.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2211955566,"x_train":366.895128245,"imp":15.9883637976,"elo":1404.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2469426441,"x_train":546.7577277937,"imp":16.0568034741,"elo":1398.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3090917862,"x_train":735.4606570949,"imp":16.0209685048,"elo":1402.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0303915587,"x_train":1.5249213808,"imp":18.4627793194,"elo":1317.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0624141331,"x_train":2.9231596776,"imp":17.7821610613,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1045674981,"x_train":6.9435649233,"imp":17.3668963083,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.167681727,"x_train":13.7562217845,"imp":17.2036229582,"elo":1394.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2384325101,"x_train":34.4671973421,"imp":17.0401788819,"elo":1406.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2910163995,"x_train":68.8507318629,"imp":16.9204698482,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3160449363,"x_train":137.5904273763,"imp":16.9565809554,"elo":1410.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3090326441,"x_train":206.0245082163,"imp":16.9644016109,"elo":1411.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3055895504,"x_train":275.9265613682,"imp":17.0365172277,"elo":1409.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0149256706,"x_train":2.8412925593,"imp":19.8340051574,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0294172839,"x_train":8.2691673892,"imp":19.3189290715,"elo":1291.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0524661348,"x_train":21.1083282561,"imp":18.9136277826,"elo":1309.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0721220241,"x_train":45.3683328739,"imp":18.7744647151,"elo":1316.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0973613753,"x_train":119.8161001058,"imp":18.63644326,"elo":1326.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1172786529,"x_train":243.8194356466,"imp":18.4985673732,"elo":1334.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1285427493,"x_train":486.6455896663,"imp":18.411250566,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1300971595,"x_train":730.1621818496,"imp":18.4178680008,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1415209075,"x_train":976.6816362244,"imp":18.4045355559,"elo":1335.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2195070167,"x_train":1.1488986663,"imp":27.0306011122,"elo":1010.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2579008861,"x_train":2.0419356023,"imp":25.365172147,"elo":1035.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3206822077,"x_train":4.7437750172,"imp":22.8965916233,"elo":1097.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3866075277,"x_train":9.1794740447,"imp":22.0312588622,"elo":1128.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.496862742,"x_train":22.6140856772,"imp":21.6232779832,"elo":1147.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6116316629,"x_train":44.8789786871,"imp":21.3484690839,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6264150113,"x_train":89.703830225,"imp":21.1367244158,"elo":1172.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6944098681,"x_train":134.3807181269,"imp":21.030360209,"elo":1181.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7789141111,"x_train":180.0329734893,"imp":21.2773019509,"elo":1179.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2592580681,"x_train":2.7868984981,"imp":25.0928001922,"elo":1031.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.470732767,"x_train":5.47397211,"imp":24.0294000702,"elo":1087.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.0377657426,"x_train":14.1482286155,"imp":22.2738545306,"elo":1155.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.5322826304,"x_train":28.5676512334,"imp":21.5347757801,"elo":1181.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.246588778,"x_train":68.1068582413,"imp":20.7714167021,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.9237399683,"x_train":137.1828565728,"imp":20.4959094686,"elo":1216.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.6333332925,"x_train":277.9218393643,"imp":20.3891344664,"elo":1220.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0459379226,"x_train":418.1539061688,"imp":20.327550808,"elo":1223.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.5532282308,"x_train":561.2940883157,"imp":20.4322580444,"elo":1230.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.094055721,"x_train":0.9329196027,"imp":20.9267620345,"elo":1152.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4273281397,"x_train":2.8974631618,"imp":18.6507312542,"elo":1271.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7023827628,"x_train":7.4728798139,"imp":17.2900290339,"elo":1354.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9043924684,"x_train":15.0938578715,"imp":16.7588564136,"elo":1387.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.098568899,"x_train":39.6915504803,"imp":16.4615228234,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.147173026,"x_train":80.8366710185,"imp":16.3916314277,"elo":1411.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.084673009,"x_train":163.2701936469,"imp":16.3492689805,"elo":1414.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0402490715,"x_train":243.2687421878,"imp":16.2947962091,"elo":1416.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0244823629,"x_train":328.641625286,"imp":16.3758834982,"elo":1416.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.3036225459,"x_train":14.2669009183,"imp":16.9328597325,"elo":1398.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3117928723,"x_train":11.1478855018,"imp":21.0320505117,"elo":1215.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8385654835,"x_train":27.3629676777,"imp":19.6248252792,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7997608403,"x_train":84.0873327564,"imp":17.2335333114,"elo":1354.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9001397479,"x_train":169.5069286032,"imp":16.7511552172,"elo":1377.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.0611635517,"x_train":413.8881530141,"imp":16.2300532092,"elo":1403.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1854995844,"x_train":856.7378065106,"imp":16.3775739161,"elo":1393.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.5251211404,"x_train":1708.0482807143,"imp":16.704966384,"elo":1382.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8121612888,"x_train":2544.0473436332,"imp":16.7462245783,"elo":1380.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4763406294,"x_train":3436.7433273226,"imp":16.7020406807,"elo":1381.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7399372917,"x_train":14.3691534377,"imp":23.7395224372,"elo":1111.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0268138653,"x_train":12.4370351532,"imp":31.9817108192,"elo":997.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2097117411,"x_train":39.6579418967,"imp":29.2683791238,"elo":1049.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4237356511,"x_train":109.2939161538,"imp":25.4782347843,"elo":1117.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4978820663,"x_train":130.4829960164,"imp":25.2603152191,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0312580102,"x_train":0.2940996202,"imp":27.2505839036,"elo":999.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0906042357,"x_train":1.108342455,"imp":24.962617101,"elo":1053.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1840220874,"x_train":3.7342981222,"imp":23.8714038074,"elo":1092.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2757254745,"x_train":8.0549426531,"imp":23.3960853518,"elo":1113.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3625016031,"x_train":20.7189526054,"imp":22.9121609339,"elo":1138.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4725802008,"x_train":41.766842473,"imp":22.8002411401,"elo":1146.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5376208643,"x_train":84.379284379,"imp":22.7152276753,"elo":1151.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5574391287,"x_train":127.399603328,"imp":22.7005586016,"elo":1149.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5606060557,"x_train":171.6065412889,"imp":22.6412060902,"elo":1153.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1617619189,"x_train":9.0446607596,"imp":17.867856331,"elo":1311.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7002960265,"x_train":23.323034735,"imp":16.1509151475,"elo":1401.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.6925437923,"x_train":62.8372296721,"imp":14.8626054677,"elo":1471.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.2033439648,"x_train":131.5115032503,"imp":14.480253057,"elo":1485.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.9399740849,"x_train":336.6014034599,"imp":14.014604163,"elo":1509.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.9896970288,"x_train":679.3497815723,"imp":13.8066685751,"elo":1520.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.8216592508,"x_train":1355.7488156461,"imp":13.7543913886,"elo":1521.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.4687086151,"x_train":2029.6864482209,"imp":13.7072998548,"elo":1532.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.3100869738,"x_train":2722.1286706272,"imp":13.6637450519,"elo":1536.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7081268775,"x_train":4.7880152698,"imp":12.6135473201,"elo":1562.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.6939241991,"x_train":16.4527651431,"imp":12.6841519322,"elo":1561.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.1722189812,"x_train":44.0076788597,"imp":12.6705328358,"elo":1551.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.2123187912,"x_train":89.9272230443,"imp":12.7074779191,"elo":1550.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.1074574709,"x_train":233.2858308743,"imp":12.6413300232,"elo":1554.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3342760367,"x_train":464.3616718274,"imp":12.757870963,"elo":1549.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.530318776,"x_train":937.3819334882,"imp":12.470877381,"elo":1563.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.3639010218,"x_train":1406.5304544301,"imp":12.4124945186,"elo":1571.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.8083515158,"x_train":1878.3590992619,"imp":12.3310444289,"elo":1574.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.1690613568,"x_train":13.0741284928,"imp":20.9545459287,"elo":1259.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"x_train":47.8871994,"imp":19.9125485476,"elo":1236.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":57.0488035322,"x_train":76.8228235044,"imp":18.5578882422,"elo":1281.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":89.3976769749,"x_train":154.1734218688,"imp":17.4088810373,"elo":1332.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":120.7151402618,"x_train":288.3811437278,"imp":16.9835849175,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":191.9678402321,"x_train":734.9046229398,"imp":16.1744857897,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":255.2028837353,"x_train":1475.2401534104,"imp":15.59877895,"elo":1381.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":327.3661339693,"x_train":2935.4601233793,"imp":15.3293650407,"elo":1391.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":361.0134335556,"x_train":4395.9302046427,"imp":15.2057520494,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":410.7225790808,"x_train":5891.3014300686,"imp":15.1608298999,"elo":1398.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1668648195,"x_train":1.4816805767,"imp":15.6160558805,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"x_train":37.5608528654,"imp":5.1109281298,"elo":1804.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.115458016,"x_train":0.8137451985,"imp":29.5080849433,"elo":1029.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1600398508,"x_train":1.7694169027,"imp":11.6209007657,"elo":1613.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2024528313,"x_train":5.8795969926,"imp":18.0216258802,"elo":1321.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.3732754571,"x_train":13.3369957806,"imp":17.7011061532,"elo":1333.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.5461575366,"x_train":38.7568509497,"imp":17.0967272132,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.731945022,"x_train":78.6946304538,"imp":16.5213683433,"elo":1395.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.0226182454,"x_train":201.4443768551,"imp":16.0486235788,"elo":1418.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.1935240832,"x_train":405.7219066472,"imp":15.7775555347,"elo":1433.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.3541917099,"x_train":807.6777264234,"imp":15.5759684234,"elo":1442.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.5669239288,"x_train":1220.4156808351,"imp":15.5116692814,"elo":1450.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":1.5118634561,"x_train":1641.7401040643,"imp":15.4831647745,"elo":1451.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4673275113,"x_train":4.8183344075,"imp":11.8715408954,"elo":1625.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.353062543,"x_train":2.9028568606,"imp":10.1975319938,"elo":1663.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0895012297,"x_train":387.3713684902,"imp":29.3154901806,"elo":1070.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.1246177871,"x_train":619.0654872096,"imp":29.1977312744,"elo":1076.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9933319757,"x_train":1126.9710593518,"imp":28.8728516091,"elo":1081.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.945280813,"x_train":1799.6745277943,"imp":28.1782196434,"elo":1092.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.9921647692,"x_train":4218.8944407454,"imp":26.8298020948,"elo":1108.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.187584759,"x_train":8667.7234636881,"imp":25.6070686102,"elo":1124.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.7769486168,"x_train":17722.5911655635,"imp":25.1176657899,"elo":1132.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.719108285,"x_train":26217.887217876,"imp":24.7184562266,"elo":1144.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.9317836231,"x_train":34925.3007391894,"imp":24.3478310533,"elo":1157.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622971538,"x_train":1.1067907259,"imp":16.4725095359,"elo":1388.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0923536337,"x_train":4.9663458274,"imp":24.5208426349,"elo":1061.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1965690156,"x_train":13.872258751,"imp":22.0196854759,"elo":1135.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4325805708,"x_train":43.9083852063,"imp":19.9665463941,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6854277095,"x_train":107.4432911334,"imp":19.1704504782,"elo":1232.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1471383289,"x_train":267.8896031045,"imp":18.5195106616,"elo":1268.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5876850499,"x_train":552.8424768764,"imp":18.1964781748,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.9574603872,"x_train":1101.0843510949,"imp":18.1135563983,"elo":1286.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1875558499,"x_train":1649.8652006608,"imp":17.9664380859,"elo":1297.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.3011998096,"x_train":2206.6390156442,"imp":17.8808006095,"elo":1300.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1176002559,"x_train":1.4234367091,"imp":20.1187535248,"elo":1221.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.291370012,"x_train":4.1993942212,"imp":18.1469125856,"elo":1322.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4007037631,"x_train":11.4672731767,"imp":17.6022222434,"elo":1356.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4312521607,"x_train":22.2498464852,"imp":17.2959121949,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4858065918,"x_train":57.7997562863,"imp":17.0556637181,"elo":1383.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4981474911,"x_train":114.5664011196,"imp":17.0011313401,"elo":1391.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5038664892,"x_train":229.0218261459,"imp":16.9219889297,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6132491042,"x_train":343.7649500386,"imp":16.8413473575,"elo":1404.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6132840031,"x_train":463.012192935,"imp":16.864465112,"elo":1404.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":44.6673844457,"x_train":202.6812316751,"imp":22.8028436384,"elo":1186.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":94.7353822956,"x_train":512.7516159338,"imp":20.2295926298,"elo":1258.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":154.9782910324,"x_train":1466.1655464244,"imp":17.3935983195,"elo":1346.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":248.7212562072,"x_train":3194.8705446357,"imp":16.320058077,"elo":1395.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":357.2213513445,"x_train":7908.2276811319,"imp":15.4224116209,"elo":1445.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":361.1026011106,"x_train":8274.6664503747,"imp":15.4708391742,"elo":1444.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.3922735327,"x_train":2.7144283593,"imp":30.4180479527,"elo":959.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.5728509945,"x_train":5.4835528309,"imp":25.8369407868,"elo":1042.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.9190160514,"x_train":13.8167338689,"imp":20.4517745654,"elo":1194.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.2127267231,"x_train":27.7848683857,"imp":18.4081761487,"elo":1285.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.4609202066,"x_train":69.6362065708,"imp":17.9003166779,"elo":1311.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.965003678,"x_train":138.9025571316,"imp":17.6804750529,"elo":1318.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.6090145696,"x_train":278.1866797416,"imp":17.5179980924,"elo":1317.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.626504492,"x_train":417.6393154275,"imp":17.4366108776,"elo":1317.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.4524874978,"x_train":559.7696517925,"imp":17.285836362,"elo":1318.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_all/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..2a128cd1ab0c3791464b440f1108724a6be8c665
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_binary/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "TabM (tuned + ensembled)", "iLTM (tuned + ensembled)", "TabDPT-Turbo (default)", "CatBoost (tuned)", "LightGBM (tuned + ensembled)", "LimiX (default)", "CatBoost (tuned + ensembled)", "CatBoost (default)", "ChimeraBoost (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabDPT (tuned + ensembled)", "TabM (tuned)", "ModernNCA (tuned)", "TabSwift (default)", "ModernNCA (tuned + ensembled)", "ChimeraBoost (tuned)", "XGBoost (tuned)", "BetaTabPFN (default)", "TabDPT (tuned)", "LightGBM (tuned)", "iLTM (tuned)", "EBM (tuned + ensembled)", "RealMLP (default)", "TabM (default)", "xRFM (tuned + ensembled)", "ChimeraBoost (default)", "TorchMLP (tuned + ensembled)", "xRFM (tuned)", "EBM (default)", "EBM (tuned)", "SAP-RPT-OSS (default)", "TabDPT (default)", "FastaiMLP (tuned + ensembled)", "XGBoost (default)", "TorchMLP (tuned)", "ModernNCA (default)", "iLTM (default)", "ExtraTrees (tuned + ensembled)", "ExtraTrees (tuned)", "LightGBM (default)", "FastaiMLP (tuned)", "TabSTAR (tuned)", "RandomForest (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "OrionMSP (default)", "PerpetualBooster (tuned)", "TabSTAR (default)", "TorchMLP (default)", "FastaiMLP (default)", "TabFlex (default)", "RandomForest (default)", "ExtraTrees (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "xRFM (default)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.8333333134651184, 0.7333333492279053, 0.8666666746139526, 0.8333333134651184, 0.7666666507720947, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.800000011920929, 0.8333333134651184, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8999999761581421, 1.0, 0.8999999761581421, 0.9666666388511658, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 1.0, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0, 1.0], [0.1666666716337204, NaN, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.6166666746139526, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.699999988079071, 0.8333333134651184, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.8999999761581421, 0.7333333492279053, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.9666666388511658, 0.9333333373069763, 1.0, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.8333333134651184, 1.0, 0.9666666388511658, 1.0, 1.0, 0.8333333134651184, 0.9666666388511658, 1.0, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0, 1.0], [0.2666666805744171, 0.4000000059604645, NaN, 0.6000000238418579, 0.46666666865348816, 0.6333333253860474, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.6333333253860474, 0.8666666746139526, 0.8999999761581421, 0.699999988079071, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.7666666507720947, 0.8999999761581421, 0.7333333492279053, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.7666666507720947, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.8999999761581421, 1.0, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0], [0.13333334028720856, 0.4333333373069763, 0.4000000059604645, NaN, 0.5, 0.5333333611488342, 0.5333333611488342, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.8999999761581421, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.699999988079071, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.7666666507720947, 1.0, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.8666666746139526, 0.800000011920929, 1.0, 0.800000011920929, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 0.800000011920929, 0.9666666388511658, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 0.9333333373069763, 1.0, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 1.0, 1.0], [0.1666666716337204, 0.4333333373069763, 0.5333333611488342, 0.5, NaN, 0.6333333253860474, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.7666666507720947, 0.7666666507720947, 0.9333333373069763, 0.8333333134651184, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.9666666388511658, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 0.8999999761581421, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 1.0, 1.0, 1.0], [0.23333333432674408, 0.4333333373069763, 0.36666667461395264, 0.46666666865348816, 0.36666667461395264, NaN, 0.5, 0.4333333373069763, 0.5333333611488342, 0.6000000238418579, 0.7666666507720947, 0.699999988079071, 0.8333333134651184, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.6333333253860474, 0.8333333134651184, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.8333333134651184, 0.7666666507720947, 0.7333333492279053, 0.8666666746139526, 0.8333333134651184, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.9333333373069763, 0.8999999761581421, 0.7666666507720947, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0, 1.0], [0.06666667014360428, 0.38333332538604736, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.5, NaN, 0.5333333611488342, 0.699999988079071, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.7666666507720947, 0.800000011920929, 0.6666666865348816, 0.8333333134651184, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.6666666865348816, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.699999988079071, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.7666666507720947, 0.8999999761581421, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 1.0, 0.8999999761581421, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.8333333134651184, 0.9333333373069763, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 1.0], [0.1666666716337204, 0.4000000059604645, 0.4000000059604645, 0.36666667461395264, 0.4000000059604645, 0.5666666626930237, 0.46666666865348816, NaN, 0.46666666865348816, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.5333333611488342, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.7333333492279053, 0.699999988079071, 0.800000011920929, 0.6333333253860474, 0.699999988079071, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.8833333253860474, 0.8333333134651184, 0.7666666507720947, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 1.0, 0.9666666388511658, 0.8666666746139526, 0.8999999761581421, 0.8666666746139526, 0.9666666388511658, 0.9333333373069763, 1.0, 1.0, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0], [0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.30000001192092896, 0.5333333611488342, NaN, 0.5333333611488342, 0.6333333253860474, 0.699999988079071, 0.5666666626930237, 0.6333333253860474, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.8333333134651184, 0.6333333253860474, 0.699999988079071, 0.6333333253860474, 0.800000011920929, 0.7666666507720947, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.7666666507720947, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.8333333134651184, 1.0, 0.8999999761581421, 0.9666666388511658, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9666666388511658, 0.8333333134651184, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 1.0], [0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.4333333373069763, 0.46666666865348816, NaN, 0.6666666865348816, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.5333333611488342, 0.800000011920929, 0.7666666507720947, 0.6333333253860474, 0.6666666865348816, 0.6000000238418579, 0.699999988079071, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.6666666865348816, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.7666666507720947, 0.8666666746139526, 0.7666666507720947, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.800000011920929, 0.9333333373069763, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0], [0.1666666716337204, 0.30000001192092896, 0.1666666716337204, 0.30000001192092896, 0.36666667461395264, 0.23333333432674408, 0.36666667461395264, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, NaN, 0.5, 0.46666666865348816, 0.5, 0.5333333611488342, 0.6333333253860474, 0.44999998807907104, 0.5666666626930237, 0.6333333253860474, 0.5, 0.6333333253860474, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.5666666626930237, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.6333333253860474, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.7666666507720947, 0.8999999761581421, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.7833333611488342, 0.8666666746139526, 0.800000011920929, 0.9666666388511658, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.949999988079071, 0.9666666388511658, 0.9333333373069763, 1.0, 1.0], [0.06666667014360428, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.30000001192092896, 0.4000000059604645, 0.5, NaN, 0.5333333611488342, 0.5666666626930237, 0.5, 0.6666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.6333333253860474, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6333333253860474, 0.5666666626930237, 0.6333333253860474, 0.6333333253860474, 0.7666666507720947, 0.7666666507720947, 0.8333333134651184, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.5666666626930237, 0.7666666507720947, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0], [0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.23333333432674408, 0.2666666805744171, 0.1666666716337204, 0.3333333432674408, 0.46666666865348816, 0.4333333373069763, 0.30000001192092896, 0.5333333611488342, 0.46666666865348816, NaN, 0.5, 0.5666666626930237, 0.5666666626930237, 0.44999998807907104, 0.5, 0.5, 0.6333333253860474, 0.6000000238418579, 0.46666666865348816, 0.5, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.699999988079071, 0.5, 0.5, 0.6000000238418579, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.699999988079071, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.7333333492279053, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 1.0], [0.06666667014360428, 0.20000000298023224, 0.36666667461395264, 0.23333333432674408, 0.23333333432674408, 0.3333333432674408, 0.4000000059604645, 0.30000001192092896, 0.36666667461395264, 0.4000000059604645, 0.5, 0.4333333373069763, 0.5, NaN, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.5, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.6333333253860474, 0.4333333373069763, 0.46666666865348816, 0.5833333134651184, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.5333333611488342, 0.699999988079071, 0.5666666626930237, 0.6666666865348816, 0.6333333253860474, 0.6000000238418579, 0.6333333253860474, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.800000011920929, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.6666666865348816, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8333333134651184, 0.8333333134651184, 0.9333333373069763, 0.8999999761581421, 1.0, 1.0], [0.10000000149011612, 0.23333333432674408, 0.13333334028720856, 0.2666666805744171, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.3333333432674408, 0.2666666805744171, 0.30000001192092896, 0.46666666865348816, 0.5, 0.4333333373069763, 0.5333333611488342, NaN, 0.4333333373069763, 0.46666666865348816, 0.46666666865348816, 0.6333333253860474, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.5, 0.46666666865348816, 0.5, 0.5, 0.5666666626930237, 0.6333333253860474, 0.5666666626930237, 0.5, 0.5333333611488342, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.800000011920929, 0.7666666507720947, 0.8666666746139526, 0.7666666507720947, 0.8999999761581421, 0.800000011920929, 0.7666666507720947, 0.8999999761581421, 0.7666666507720947, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 1.0, 1.0], [0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.2666666805744171, 0.36666667461395264, 0.3333333432674408, 0.4333333373069763, 0.4333333373069763, 0.5666666626930237, NaN, 0.36666667461395264, 0.5666666626930237, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5, 0.5, 0.5333333611488342, 0.5666666626930237, 0.6333333253860474, 0.46666666865348816, 0.5333333611488342, 0.7666666507720947, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.7666666507720947, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.9666666388511658, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0], [0.10000000149011612, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.30000001192092896, 0.4000000059604645, 0.46666666865348816, 0.550000011920929, 0.5333333611488342, 0.550000011920929, 0.4000000059604645, 0.5333333611488342, 0.6333333253860474, NaN, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.6333333253860474, 0.4333333373069763, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.6666666865348816, 0.5666666626930237, 0.5333333611488342, 0.6000000238418579, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6000000238418579, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.6833333373069763, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.7666666507720947, 0.7833333611488342, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9666666388511658], [0.06666667014360428, 0.1666666716337204, 0.06666667014360428, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.1666666716337204, 0.3333333432674408, 0.2666666805744171, 0.20000000298023224, 0.4333333373069763, 0.4000000059604645, 0.5, 0.5, 0.5333333611488342, 0.4333333373069763, 0.4333333373069763, NaN, 0.6000000238418579, 0.5, 0.46666666865348816, 0.4333333373069763, 0.4333333373069763, 0.44999998807907104, 0.46666666865348816, 0.5, 0.5666666626930237, 0.5666666626930237, 0.5, 0.5, 0.5666666626930237, 0.699999988079071, 0.6000000238418579, 0.6666666865348816, 0.5, 0.6333333253860474, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.6333333253860474, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.7666666507720947, 0.8999999761581421, 0.7666666507720947, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.8666666746139526, 0.7666666507720947, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 0.9666666388511658, 1.0], [0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.30000001192092896, 0.1666666716337204, 0.23333333432674408, 0.36666667461395264, 0.4000000059604645, 0.5, 0.46666666865348816, 0.36666667461395264, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, NaN, 0.4333333373069763, 0.5, 0.5333333611488342, 0.4333333373069763, 0.4333333373069763, 0.5, 0.5, 0.5666666626930237, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6333333253860474, 0.6666666865348816, 0.7666666507720947, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.9333333373069763, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 1.0, 1.0], [0.10000000149011612, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.5, 0.4333333373069763, 0.36666667461395264, 0.4333333373069763, 0.5333333611488342, 0.5333333611488342, 0.4333333373069763, 0.5, 0.5666666626930237, NaN, 0.5333333611488342, 0.5333333611488342, 0.38333332538604736, 0.5, 0.5333333611488342, 0.5, 0.6333333253860474, 0.5666666626930237, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.6666666865348816, 0.6333333253860474, 0.699999988079071, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.8333333134651184, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 1.0], [0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.30000001192092896, 0.23333333432674408, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.46666666865348816, 0.4000000059604645, 0.4333333373069763, 0.46666666865348816, 0.5333333611488342, 0.36666667461395264, 0.5333333611488342, 0.5, 0.46666666865348816, NaN, 0.5, 0.46666666865348816, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.5666666626930237, 0.699999988079071, 0.4333333373069763, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.5666666626930237, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.8333333134651184, 0.6833333373069763, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 1.0, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 1.0], [0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.3333333432674408, 0.3333333432674408, 0.23333333432674408, 0.36666667461395264, 0.4000000059604645, 0.5333333611488342, 0.4333333373069763, 0.5333333611488342, 0.36666667461395264, 0.5333333611488342, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, 0.46666666865348816, 0.5, NaN, 0.46666666865348816, 0.4333333373069763, 0.5, 0.4333333373069763, 0.5, 0.5, 0.5, 0.6666666865348816, 0.5333333611488342, 0.6333333253860474, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.9333333373069763, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.9666666388511658, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 1.0], [0.03333333507180214, 0.10000000149011612, 0.23333333432674408, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4333333373069763, 0.5, 0.5666666626930237, 0.5, 0.5333333611488342, 0.4333333373069763, 0.5666666626930237, 0.5666666626930237, 0.6166666746139526, 0.5333333611488342, 0.5333333611488342, NaN, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6333333253860474, 0.6333333253860474, 0.5, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.6666666865348816, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.9666666388511658, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658], [0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.23333333432674408, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.23333333432674408, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.4333333373069763, 0.5333333611488342, 0.5333333611488342, 0.5, 0.4333333373069763, 0.550000011920929, 0.5666666626930237, 0.5, 0.5666666626930237, 0.5666666626930237, 0.4333333373069763, NaN, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.6333333253860474, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.6666666865348816, 0.6333333253860474, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.6833333373069763, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.8333333134651184, 0.800000011920929, 0.7666666507720947, 0.7666666507720947, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 1.0], [0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.23333333432674408, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.36666667461395264, 0.4333333373069763, 0.4166666567325592, 0.5, 0.5, 0.5333333611488342, 0.5333333611488342, 0.5, 0.46666666865348816, 0.5333333611488342, 0.5, 0.4333333373069763, 0.46666666865348816, NaN, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.4333333373069763, 0.6000000238418579, 0.5666666626930237, 0.6333333253860474, 0.5333333611488342, 0.5666666626930237, 0.699999988079071, 0.699999988079071, 0.5, 0.6666666865348816, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.9666666388511658, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0], [0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.3333333432674408, 0.4333333373069763, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.5, 0.46666666865348816, 0.4333333373069763, 0.5, 0.5, 0.5, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, NaN, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.6333333253860474, 0.5, 0.6333333253860474, 0.699999988079071, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.8666666746139526, 0.800000011920929, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.8500000238418579, 0.8666666746139526, 0.7666666507720947, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.9666666388511658], [0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.4333333373069763, 0.30000001192092896, 0.36666667461395264, 0.4333333373069763, 0.4333333373069763, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.36666667461395264, 0.4333333373069763, 0.5, 0.36666667461395264, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, NaN, 0.5333333611488342, 0.5, 0.46666666865348816, 0.5, 0.46666666865348816, 0.699999988079071, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.6333333253860474, 0.7666666507720947, 0.7666666507720947, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.8333333134651184, 0.7666666507720947, 0.7666666507720947, 0.8333333134651184, 0.7666666507720947, 0.8666666746139526, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0], [0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.3333333432674408, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.30000001192092896, 0.5, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, NaN, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.5666666626930237, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.6333333253860474, 0.6333333253860474, 0.699999988079071, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8666666746139526, 0.9333333373069763, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0], [0.03333333507180214, 0.03333333507180214, 0.1666666716337204, 0.13333334028720856, 0.06666667014360428, 0.23333333432674408, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.5, 0.46666666865348816, 0.4333333373069763, 0.5333333611488342, 0.4333333373069763, 0.5, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.5, 0.4333333373069763, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.5, 0.5333333611488342, NaN, 0.6000000238418579, 0.5, 0.5, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.6333333253860474, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.7333333492279053, 0.7666666507720947, 0.6333333253860474, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.9333333373069763, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.9333333373069763, 1.0, 1.0], [0.03333333507180214, 0.06666667014360428, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.2666666805744171, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.3333333432674408, 0.4333333373069763, 0.4333333373069763, 0.5, 0.30000001192092896, 0.5, 0.46666666865348816, 0.46666666865348816, 0.5, 0.46666666865348816, 0.5, 0.5, 0.3333333432674408, 0.4333333373069763, 0.36666667461395264, 0.5666666626930237, 0.4333333373069763, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, NaN, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.5, 0.5333333611488342, 0.5666666626930237, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6333333253860474, 0.5333333611488342, 0.5666666626930237, 0.7833333611488342, 0.5333333611488342, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.6333333253860474, 0.7666666507720947, 0.7333333492279053, 0.6333333253860474, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.9666666388511658, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658], [0.03333333507180214, 0.0, 0.06666667014360428, 0.0, 0.1666666716337204, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.46666666865348816, 0.23333333432674408, 0.4000000059604645, 0.4333333373069763, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4333333373069763, 0.4000000059604645, 0.36666667461395264, 0.5, 0.46666666865348816, 0.5, 0.46666666865348816, NaN, 0.5, 0.6333333253860474, 0.6000000238418579, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7666666507720947, 0.6333333253860474, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 1.0], [0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.1666666716337204, 0.23333333432674408, 0.11666666716337204, 0.20000000298023224, 0.23333333432674408, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.36666667461395264, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.5, 0.5333333611488342, 0.46666666865348816, 0.5, 0.4000000059604645, 0.5, NaN, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.6000000238418579, 0.4333333373069763, 0.5666666626930237, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.699999988079071, 0.6333333253860474, 0.7333333492279053, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.6166666746139526, 0.800000011920929, 0.8333333134651184, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.8666666746139526, 0.7666666507720947, 0.8333333134651184, 0.699999988079071, 0.699999988079071, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.9666666388511658, 0.8666666746139526, 1.0, 1.0], [0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.36666667461395264, 0.36666667461395264, 0.30000001192092896, 0.46666666865348816, 0.4333333373069763, 0.46666666865348816, 0.36666667461395264, 0.4333333373069763, NaN, 0.6333333253860474, 0.5666666626930237, 0.5, 0.4333333373069763, 0.6000000238418579, 0.5666666626930237, 0.8333333134651184, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.6000000238418579, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 1.0], [0.10000000149011612, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.13333334028720856, 0.23333333432674408, 0.23333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.4000000059604645, 0.36666667461395264, 0.3333333432674408, 0.46666666865348816, 0.30000001192092896, 0.3333333432674408, 0.4000000059604645, 0.4333333373069763, 0.5, 0.4000000059604645, 0.4333333373069763, 0.36666667461395264, NaN, 0.46666666865348816, 0.4333333373069763, 0.46666666865348816, 0.6333333253860474, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.8333333134651184, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 1.0], [0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.13333334028720856, 0.2666666805744171, 0.30000001192092896, 0.1666666716337204, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.5, 0.4000000059604645, 0.36666667461395264, 0.4333333373069763, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.4333333373069763, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.36666667461395264, 0.46666666865348816, 0.4333333373069763, 0.5333333611488342, NaN, 0.46666666865348816, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.5166666507720947, 0.5333333611488342, 0.6000000238418579, 0.6333333253860474, 0.7333333492279053, 0.6333333253860474, 0.699999988079071, 0.699999988079071, 0.6333333253860474, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.9666666388511658, 0.9333333373069763], [0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.4333333373069763, 0.4000000059604645, 0.5, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.4333333373069763, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.5, 0.5666666626930237, 0.5333333611488342, NaN, 0.36666667461395264, 0.5, 0.6333333253860474, 0.5666666626930237, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.8333333134651184, 0.7666666507720947, 0.7333333492279053, 0.800000011920929, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658], [0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.36666667461395264, 0.3333333432674408, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.5, 0.4000000059604645, 0.36666667461395264, 0.30000001192092896, 0.4333333373069763, 0.3333333432674408, 0.4333333373069763, 0.4000000059604645, 0.4000000059604645, 0.36666667461395264, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.4333333373069763, 0.6333333253860474, NaN, 0.5666666626930237, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.6666666865348816, 0.7666666507720947, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.7666666507720947, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.9666666388511658, 0.9333333373069763, 1.0], [0.03333333507180214, 0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.3333333432674408, 0.23333333432674408, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.4333333373069763, 0.4000000059604645, 0.36666667461395264, 0.4333333373069763, 0.5, 0.4333333373069763, NaN, 0.5333333611488342, 0.4333333373069763, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.8333333134651184, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.699999988079071, 0.9333333373069763, 0.8666666746139526, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658], [0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.3333333432674408, 0.4333333373069763, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.4333333373069763, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.2666666805744171, 0.3333333432674408, 0.4333333373069763, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.4333333373069763, 0.4333333373069763, 0.4333333373069763, 0.36666667461395264, 0.36666667461395264, 0.46666666865348816, NaN, 0.46666666865348816, 0.5666666626930237, 0.6333333253860474, 0.5, 0.6000000238418579, 0.5, 0.6666666865348816, 0.5666666626930237, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 1.0], [0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.36666667461395264, 0.3333333432674408, 0.2666666805744171, 0.1666666716337204, 0.4333333373069763, 0.4833333194255829, 0.4333333373069763, 0.36666667461395264, 0.5666666626930237, 0.5333333611488342, NaN, 0.5, 0.5333333611488342, 0.5333333611488342, 0.6333333253860474, 0.5, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.7666666507720947, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.800000011920929, 0.9333333373069763, 0.9666666388511658, 1.0], [0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.1666666716337204, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.23333333432674408, 0.2666666805744171, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.46666666865348816, 0.36666667461395264, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.3333333432674408, 0.5333333611488342, 0.4333333373069763, 0.5, NaN, 0.6333333253860474, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.6333333253860474, 0.699999988079071, 0.7666666507720947, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9666666388511658, 1.0], [0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.30000001192092896, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.4333333373069763, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.4333333373069763, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.36666667461395264, 0.46666666865348816, 0.36666667461395264, NaN, 0.5666666626930237, 0.6000000238418579, 0.46666666865348816, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.6333333253860474, 0.7333333492279053, 0.6000000238418579, 0.699999988079071, 0.6333333253860474, 0.5666666626930237, 0.6333333253860474, 0.6000000238418579, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 1.0], [0.0, 0.06666667014360428, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.3333333432674408, 0.21666666865348816, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.4333333373069763, 0.4333333373069763, 0.4000000059604645, 0.5, 0.46666666865348816, 0.4000000059604645, 0.4333333373069763, NaN, 0.5, 0.5, 0.46666666865348816, 0.6333333253860474, 0.5333333611488342, 0.5666666626930237, 0.6333333253860474, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.7666666507720947, 0.6666666865348816, 0.7333333492279053, 0.8333333134651184, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9666666388511658], [0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.03333333507180214, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.23333333432674408, 0.36666667461395264, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.36666667461395264, 0.30000001192092896, 0.3166666626930237, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.36666667461395264, 0.46666666865348816, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.36666667461395264, 0.4333333373069763, 0.4000000059604645, 0.5, NaN, 0.4333333373069763, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.7666666507720947, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.6333333253860474, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.9666666388511658, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.9666666388511658], [0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.23333333432674408, 0.1666666716337204, 0.3333333432674408, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.23333333432674408, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.30000001192092896, 0.4333333373069763, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.5, 0.5, 0.4333333373069763, 0.5333333611488342, 0.5, 0.5666666626930237, NaN, 0.5, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.9333333373069763], [0.0, 0.03333333507180214, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.3166666626930237, 0.3333333432674408, 0.23333333432674408, 0.20000000298023224, 0.4000000059604645, 0.23333333432674408, 0.3333333432674408, 0.3333333432674408, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.4333333373069763, 0.4333333373069763, 0.5333333611488342, 0.4333333373069763, 0.5, NaN, 0.6000000238418579, 0.6000000238418579, 0.5833333134651184, 0.6333333253860474, 0.550000011920929, 0.5333333611488342, 0.7333333492279053, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.8999999761581421, 0.7666666507720947, 0.8999999761581421, 0.699999988079071, 0.7666666507720947, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658], [0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.10000000149011612, 0.2666666805744171, 0.1666666716337204, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.3333333432674408, 0.23333333432674408, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.2666666805744171, 0.36666667461395264, 0.3333333432674408, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, NaN, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.7666666507720947, 0.6333333253860474, 0.7666666507720947, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7666666507720947, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658], [0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.23333333432674408, 0.1666666716337204, 0.23333333432674408, 0.20000000298023224, 0.13333334028720856, 0.30000001192092896, 0.20000000298023224, 0.23333333432674408, 0.30000001192092896, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.38333332538604736, 0.4000000059604645, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, NaN, 0.5333333611488342, 0.6000000238418579, 0.4000000059604645, 0.5333333611488342, 0.6666666865348816, 0.5666666626930237, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6333333253860474, 0.5, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.8333333134651184, 0.5666666626930237, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.8666666746139526, 0.9333333373069763], [0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.03333333507180214, 0.13333334028720856, 0.0, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.13333334028720856, 0.30000001192092896, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.13333334028720856, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.4333333373069763, 0.4000000059604645, 0.3333333432674408, 0.4166666567325592, 0.46666666865348816, 0.46666666865348816, NaN, 0.6666666865348816, 0.46666666865348816, 0.4333333373069763, 0.6000000238418579, 0.5333333611488342, 0.6333333253860474, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.6333333253860474, 0.699999988079071, 0.5666666626930237, 0.699999988079071, 0.7666666507720947, 0.8666666746139526, 0.8999999761581421, 0.8666666746139526, 0.7333333492279053, 0.7666666507720947, 0.8666666746139526, 0.7666666507720947, 0.8333333134651184, 0.8999999761581421, 1.0], [0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.03333333507180214, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.1666666716337204, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.36666667461395264, 0.4000000059604645, 0.3333333432674408, 0.36666667461395264, 0.46666666865348816, 0.4000000059604645, 0.3333333432674408, NaN, 0.36666667461395264, 0.4333333373069763, 0.5666666626930237, 0.4333333373069763, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.6333333253860474, 0.6000000238418579, 0.7333333492279053, 0.5, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658], [0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.0, 0.0, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.03333333507180214, 0.30000001192092896, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.0, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.2666666805744171, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.36666667461395264, 0.1666666716337204, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.44999998807907104, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6333333253860474, NaN, 0.5333333611488342, 0.6000000238418579, 0.5, 0.6333333253860474, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.6333333253860474, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.7666666507720947, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.9333333373069763], [0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.03333333507180214, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.23333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, NaN, 0.5666666626930237, 0.46666666865348816, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.6333333253860474, 0.6666666865348816, 0.6333333253860474, 0.7666666507720947, 0.7666666507720947, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.9333333373069763], [0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.23333333432674408, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.3166666626930237, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.3333333432674408, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.3333333432674408, 0.2666666805744171, 0.23333333432674408, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.23333333432674408, 0.3333333432674408, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, NaN, 0.4000000059604645, 0.699999988079071, 0.4333333373069763, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.9333333373069763, 0.5333333611488342, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.8999999761581421], [0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.1666666716337204, 0.13333334028720856, 0.2666666805744171, 0.36666667461395264, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.4333333373069763, 0.46666666865348816, 0.5666666626930237, 0.5, 0.5333333611488342, 0.6000000238418579, NaN, 0.6333333253860474, 0.6666666865348816, 0.5333333611488342, 0.5666666626930237, 0.6333333253860474, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.7666666507720947, 0.6333333253860474, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.9666666388511658], [0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.3333333432674408, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.23333333432674408, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.3333333432674408, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.36666667461395264, 0.4333333373069763, 0.36666667461395264, 0.4333333373069763, 0.30000001192092896, 0.36666667461395264, NaN, 0.4333333373069763, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.9666666388511658, 0.5666666626930237, 0.6000000238418579, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.7333333492279053, 0.7333333492279053, 0.8333333134651184, 0.8999999761581421], [0.03333333507180214, 0.0, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.06666667014360428, 0.23333333432674408, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.03333333507180214, 0.23333333432674408, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.10000000149011612, 0.2666666805744171, 0.2666666805744171, 0.10000000149011612, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.46666666865348816, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.5666666626930237, 0.3333333432674408, 0.5666666626930237, NaN, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.699999988079071, 0.5666666626930237, 0.7666666507720947, 0.7333333492279053, 0.7833333611488342, 0.7666666507720947, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.9666666388511658], [0.0, 0.03333333507180214, 0.10000000149011612, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, NaN, 0.5, 0.800000011920929, 0.4000000059604645, 0.5, 0.5333333611488342, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6333333253860474, 0.8333333134651184, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.9333333373069763], [0.0, 0.0, 0.0, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.0, 0.0, 0.06666667014360428, 0.21666666865348816, 0.1666666716337204, 0.1666666716337204, 0.10000000149011612, 0.23333333432674408, 0.20000000298023224, 0.21666666865348816, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.13333334028720856, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.36666667461395264, 0.4333333373069763, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.46666666865348816, 0.4000000059604645, 0.5, NaN, 0.5, 0.6333333253860474, 0.5, 0.5, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.9666666388511658], [0.0, 0.0, 0.06666667014360428, 0.0, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.0, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.10000000149011612, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.23333333432674408, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.10000000149011612, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.5, 0.36666667461395264, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.20000000298023224, 0.5, NaN, 0.4000000059604645, 0.46666666865348816, 0.5, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.5333333611488342, 0.800000011920929, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.9333333373069763], [0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.23333333432674408, 0.1666666716337204, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.06666667014360428, 0.2666666805744171, 0.03333333507180214, 0.30000001192092896, 0.6000000238418579, 0.36666667461395264, 0.6000000238418579, NaN, 0.4000000059604645, 0.3333333432674408, 0.5666666626930237, 0.5666666626930237, 0.5, 0.5, 0.6333333253860474, 0.5, 0.5, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.8666666746139526], [0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.0, 0.03333333507180214, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.1666666716337204, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.03333333507180214, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.1666666716337204, 0.03333333507180214, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.10000000149011612, 0.1666666716337204, 0.3333333432674408, 0.4333333373069763, 0.5, 0.3333333432674408, 0.36666667461395264, 0.46666666865348816, 0.30000001192092896, 0.4333333373069763, 0.4333333373069763, 0.5, 0.5, 0.5333333611488342, 0.6000000238418579, NaN, 0.5333333611488342, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.8333333134651184, 0.9333333373069763], [0.0, 0.0, 0.03333333507180214, 0.03333333507180214, 0.0, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.03333333507180214, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.1666666716337204, 0.03333333507180214, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.03333333507180214, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.1666666716337204, 0.03333333507180214, 0.2666666805744171, 0.23333333432674408, 0.20000000298023224, 0.3333333432674408, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.23333333432674408, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.23333333432674408, 0.46666666865348816, 0.5, 0.5, 0.6666666865348816, 0.46666666865348816, NaN, 0.5666666626930237, 0.5, 0.5666666626930237, 0.5666666626930237, 0.699999988079071, 0.5666666626930237, 0.5666666626930237, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.9333333373069763], [0.0, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.10000000149011612, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.1666666716337204, 0.03333333507180214, 0.23333333432674408, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.10000000149011612, 0.06666667014360428, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.1666666716337204, 0.30000001192092896, 0.10000000149011612, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.30000001192092896, 0.4333333373069763, 0.4333333373069763, 0.36666667461395264, 0.4333333373069763, NaN, 0.5333333611488342, 0.5, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.6333333253860474, 0.6666666865348816, 0.8333333134651184], [0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.1666666716337204, 0.03333333507180214, 0.10000000149011612, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.30000001192092896, 0.20000000298023224, 0.3333333432674408, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.13333334028720856, 0.30000001192092896, 0.21666666865348816, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4333333373069763, 0.36666667461395264, 0.5, 0.46666666865348816, NaN, 0.5, 0.5, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.9666666388511658], [0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.03333333507180214, 0.0, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.1666666716337204, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.10000000149011612, 0.03333333507180214, 0.20000000298023224, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.1666666716337204, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.20000000298023224, 0.3333333432674408, 0.23333333432674408, 0.4000000059604645, 0.30000001192092896, 0.4333333373069763, 0.5, 0.3333333432674408, 0.4333333373069763, 0.5, 0.5, NaN, 0.4333333373069763, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.6666666865348816, 0.9333333373069763], [0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.0, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.13333334028720856, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.15000000596046448, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.2666666805744171, 0.13333334028720856, 0.23333333432674408, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.13333334028720856, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.36666667461395264, 0.23333333432674408, 0.4000000059604645, 0.20000000298023224, 0.36666667461395264, 0.4000000059604645, 0.46666666865348816, 0.5, 0.3333333432674408, 0.4333333373069763, 0.5333333611488342, 0.5, 0.5666666626930237, NaN, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.9666666388511658, 0.9666666388511658], [0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.13333334028720856, 0.06666667014360428, 0.1666666716337204, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.4333333373069763, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.36666667461395264, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.1666666716337204, 0.3333333432674408, 0.20000000298023224, 0.36666667461395264, 0.3333333432674408, 0.30000001192092896, 0.4333333373069763, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, NaN, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.5, 0.8666666746139526], [0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.10000000149011612, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.23333333432674408, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.13333334028720856, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.36666667461395264, 0.2666666805744171, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.5, 0.36666667461395264, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.3333333432674408, 0.4333333373069763, NaN, 0.7666666507720947, 0.5666666626930237, 0.6666666865348816, 0.5666666626930237, 0.8333333134651184], [0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.05000000074505806, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.10000000149011612, 0.06666667014360428, 0.03333333507180214, 0.1666666716337204, 0.03333333507180214, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.2666666805744171, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.36666667461395264, 0.5, 0.30000001192092896, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.3333333432674408, 0.4333333373069763, 0.23333333432674408, NaN, 0.5666666626930237, 0.6000000238418579, 0.5333333611488342, 0.800000011920929], [0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.03333333507180214, 0.1666666716337204, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.03333333507180214, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.03333333507180214, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.30000001192092896, 0.4333333373069763, 0.4333333373069763, 0.4333333373069763, NaN, 0.4333333373069763, 0.5666666626930237, 0.800000011920929], [0.0, 0.0, 0.03333333507180214, 0.03333333507180214, 0.0, 0.0, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.3333333432674408, 0.46666666865348816, 0.23333333432674408, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.4333333373069763, 0.30000001192092896, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.5666666626930237, NaN, 0.5666666626930237, 0.7333333492279053], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.13333334028720856, 0.03333333507180214, 0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.03333333507180214, 0.06666667014360428, 0.0, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.03333333507180214, 0.13333334028720856, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.1666666716337204, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.03333333507180214, 0.5, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.4333333373069763, NaN, 0.9333333373069763], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.03333333507180214, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.0, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.06666667014360428, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9273148063},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8604166588},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8486111036},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8273148112},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8245370328},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8004629556},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7974537004},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7932870349},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.770370368},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7532407377},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6835648171},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6754629625},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6752314804},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6539351848},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6425925923},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6421296311},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6337963006},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6321759267},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6319444445},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6307870404},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6275462986},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6268518539},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6256944475},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6254629646},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6145833381},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6048611142},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5953703733},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5879629673},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5856481519},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5835648183},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5736111102},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5694444468},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5560185197},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5453703735},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5446759287},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5425925948},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5393518558},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.523611113},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.501388894},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4937500045},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4916666717},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4824074119},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4581018579},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4516203743},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4402777842},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4372685209},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4365740791},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3979166735},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3956018563},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3722222281},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3701388931},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3694444486},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3687500046},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.367592597},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3583333388},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3335648184},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3328703752},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3148148183},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2833333391},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2819444507},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2750000045},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2486111152},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2398148201},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.231250004},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2296296346},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2219907444},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2166666712},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2143518562},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1951388941},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1861111158},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1662037081},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1208333363},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0379629645}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_binary/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..758d57a11a7ace3bdf79da2c777822a5ca2ddef5
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_all/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),TabICLv2 (default),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),RealMLP (tuned),TabM (tuned + ensembled),iLTM (tuned + ensembled),TabDPT-Turbo (default),CatBoost (tuned),LightGBM (tuned + ensembled),LimiX (default),CatBoost (tuned + ensembled),CatBoost (default),ChimeraBoost (tuned + ensembled),XGBoost (tuned + ensembled),TabDPT (tuned + ensembled),TabM (tuned),ModernNCA (tuned),TabSwift (default),ModernNCA (tuned + ensembled),ChimeraBoost (tuned),XGBoost (tuned),BetaTabPFN (default),TabDPT (tuned),LightGBM (tuned),iLTM (tuned),EBM (tuned + ensembled),RealMLP (default),TabM (default),xRFM (tuned + ensembled),ChimeraBoost (default),TorchMLP (tuned + ensembled),xRFM (tuned),EBM (default),EBM (tuned),SAP-RPT-OSS (default),TabDPT (default),FastaiMLP (tuned + ensembled),XGBoost (default),TorchMLP (tuned),ModernNCA (default),iLTM (default),ExtraTrees (tuned + ensembled),ExtraTrees (tuned),LightGBM (default),FastaiMLP (tuned),TabSTAR (tuned),RandomForest (tuned + ensembled),TabSTAR (tuned + ensembled),RandomForest (tuned),PerpetualBooster (tuned + ensembled),OrionMSP (default),PerpetualBooster (tuned),TabSTAR (default),TorchMLP (default),FastaiMLP (default),TabFlex (default),RandomForest (default),ExtraTrees (default),KNN (tuned + ensembled),PerpetualBooster (default),Linear (tuned + ensembled),Linear (tuned),xRFM (default),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.8333333,0.73333335,0.8666667,0.8333333,0.76666665,0.93333334,0.8333333,0.93333334,0.8,0.8333333,0.93333334,0.8,0.93333334,0.9,0.93333334,0.9,0.93333334,0.8666667,0.9,0.9,0.93333334,0.96666664,0.9,0.8666667,0.8666667,0.8666667,0.8333333,0.96666664,0.96666664,0.96666664,0.93333334,0.93333334,0.9,0.93333334,0.93333334,0.93333334,0.96666664,0.93333334,0.93333334,0.93333334,0.9,1.0,0.9,0.96666664,1.0,0.93333334,0.93333334,0.96666664,0.96666664,0.96666664,0.93333334,0.93333334,0.96666664,0.9,0.96666664,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,0.96666664,0.96666664,0.93333334,1.0,0.96666664,0.96666664,0.96666664,1.0,1.0,1.0
+TabPFN-3 (default),0.16666667,,0.6,0.56666666,0.56666666,0.56666666,0.6166667,0.6,0.6666667,0.73333335,0.7,0.8333333,0.73333335,0.8,0.76666665,0.9,0.73333335,0.8333333,0.8666667,0.8666667,0.8333333,0.8,0.9,0.8333333,0.8666667,0.9,0.8333333,0.8,0.96666664,0.93333334,1.0,0.9,0.8333333,0.9,0.8666667,0.8666667,0.9,0.8666667,0.9,0.9,0.93333334,0.8333333,0.93333334,0.8666667,0.93333334,0.96666664,0.93333334,0.96666664,0.96666664,0.96666664,0.96666664,0.93333334,0.8333333,0.93333334,0.8333333,1.0,0.96666664,1.0,1.0,0.8333333,0.96666664,1.0,0.93333334,0.96666664,0.96666664,0.96666664,0.96666664,0.93333334,0.96666664,0.96666664,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.26666668,0.4,,0.6,0.46666667,0.6333333,0.6666667,0.6,0.6666667,0.8,0.8333333,0.8,0.8333333,0.6333333,0.8666667,0.9,0.7,0.93333334,0.9,0.9,0.9,0.8,0.76666665,0.9,0.73333335,0.9,0.9,0.9,0.8333333,0.76666665,0.93333334,0.8666667,0.8333333,0.8333333,0.8666667,0.93333334,0.93333334,0.9,0.8666667,0.8666667,0.8333333,0.8333333,0.8333333,0.8666667,0.9,0.9,0.93333334,0.93333334,0.93333334,0.93333334,0.96666664,0.9,0.8333333,0.93333334,0.8333333,0.93333334,0.9,1.0,0.93333334,0.8333333,0.93333334,0.96666664,0.93333334,0.93333334,0.96666664,0.96666664,0.93333334,0.96666664,0.96666664,0.96666664,0.96666664,1.0,1.0
+TabPFN-2.6 (default),0.13333334,0.43333334,0.4,,0.5,0.53333336,0.53333336,0.6333333,0.6,0.6666667,0.7,0.73333335,0.76666665,0.76666665,0.73333335,0.9,0.76666665,0.8,0.8333333,0.8,0.7,0.73333335,0.8,0.76666665,0.73333335,0.8666667,0.8,0.8,0.8666667,0.76666665,1.0,0.8333333,0.9,0.8666667,0.8666667,0.8333333,0.8333333,0.9,0.8666667,0.8666667,0.93333334,0.8333333,0.8,0.8666667,0.93333334,0.96666664,0.93333334,0.93333334,0.93333334,0.93333334,0.96666664,0.8666667,0.8,1.0,0.8,0.96666664,0.96666664,0.96666664,1.0,0.8,0.96666664,0.96666664,0.9,0.96666664,0.93333334,1.0,0.96666664,0.96666664,0.96666664,0.93333334,0.96666664,1.0,1.0
+TabICLv2 (default),0.16666667,0.43333334,0.53333336,0.5,,0.6333333,0.53333336,0.6,0.6666667,0.6666667,0.6333333,0.7,0.73333335,0.76666665,0.76666665,0.8,0.7,0.76666665,0.8,0.8,0.76666665,0.8333333,0.73333335,0.8,0.8,0.8333333,0.76666665,0.76666665,0.93333334,0.8333333,0.8333333,0.76666665,0.8,0.8333333,0.8666667,0.8333333,0.8,0.8666667,0.8666667,0.9,0.9,0.8333333,0.9,0.96666664,0.8666667,0.9,0.9,0.8333333,0.9,0.93333334,0.96666664,0.93333334,0.8666667,0.93333334,0.8666667,0.93333334,0.93333334,0.93333334,0.96666664,0.8666667,1.0,1.0,1.0,0.93333334,0.9,0.96666664,0.96666664,0.93333334,0.96666664,0.93333334,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.23333333,0.43333334,0.36666667,0.46666667,0.36666667,,0.5,0.43333334,0.53333336,0.6,0.76666665,0.7,0.8333333,0.6666667,0.8,0.8,0.6333333,0.8333333,0.76666665,0.73333335,0.8333333,0.6666667,0.76666665,0.7,0.7,0.6666667,0.76666665,0.8333333,0.76666665,0.73333335,0.8666667,0.8333333,0.76666665,0.76666665,0.8,0.8333333,0.8333333,0.9,0.8666667,0.8333333,0.9,0.8,0.8666667,0.8666667,0.9,0.93333334,0.9,0.9,0.9,0.8666667,0.93333334,0.9,0.76666665,0.93333334,0.8,0.96666664,0.93333334,0.93333334,0.96666664,0.8333333,0.96666664,0.96666664,0.93333334,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.06666667,0.38333333,0.33333334,0.46666667,0.46666667,0.5,,0.53333336,0.7,0.6333333,0.6333333,0.6666667,0.6666667,0.6,0.76666665,0.8,0.6666667,0.8333333,0.8333333,0.76666665,0.8,0.6666667,0.76666665,0.8,0.76666665,0.7,0.8333333,0.8,0.8666667,0.76666665,0.9,0.76666665,0.8333333,0.8,0.8666667,0.8,0.8,0.8,0.8666667,0.8666667,0.8333333,0.8333333,0.8,0.8,0.93333334,0.8666667,0.8333333,0.8666667,0.96666664,0.96666664,1.0,0.9,0.8,0.93333334,0.8,0.96666664,0.93333334,0.96666664,0.96666664,0.8333333,0.93333334,0.96666664,0.8666667,0.96666664,0.96666664,0.93333334,0.93333334,0.9,0.93333334,0.93333334,0.93333334,0.96666664,1.0
+RealTabPFN-2.5 (default),0.16666667,0.4,0.4,0.36666667,0.4,0.56666666,0.46666667,,0.46666667,0.56666666,0.6666667,0.6666667,0.53333336,0.7,0.6666667,0.76666665,0.7,0.6666667,0.7,0.7,0.73333335,0.76666665,0.73333335,0.7,0.8,0.6333333,0.7,0.8,0.8666667,0.73333335,0.8,0.8833333,0.8333333,0.76666665,0.8333333,0.8333333,0.8333333,0.8333333,0.8666667,0.9,0.93333334,0.8666667,0.8333333,0.8666667,0.9,0.8666667,0.9,0.8666667,0.8666667,0.8666667,1.0,0.96666664,0.8666667,0.9,0.8666667,0.96666664,0.93333334,1.0,1.0,0.8666667,1.0,0.93333334,1.0,0.93333334,1.0,0.96666664,0.93333334,0.93333334,0.93333334,0.96666664,0.96666664,1.0,1.0
+RealTabPFN-2.5 (tuned),0.06666667,0.33333334,0.33333334,0.4,0.33333334,0.46666667,0.3,0.53333336,,0.53333336,0.6333333,0.7,0.56666666,0.6333333,0.73333335,0.73333335,0.6,0.73333335,0.8333333,0.6333333,0.7,0.6333333,0.8,0.76666665,0.6666667,0.6666667,0.76666665,0.73333335,0.8333333,0.76666665,0.9,0.8,0.8333333,0.8,0.8666667,0.73333335,0.76666665,0.8,0.8333333,0.8333333,0.8333333,0.8333333,0.8,0.8,0.9,0.9,0.8666667,0.8333333,1.0,0.9,0.96666664,0.8666667,0.8,0.93333334,0.8,0.93333334,0.93333334,1.0,0.96666664,0.8333333,0.93333334,0.93333334,0.8666667,0.96666664,0.96666664,0.93333334,0.93333334,0.9,0.9,0.93333334,0.93333334,0.96666664,1.0
+RealMLP (tuned + ensembled),0.2,0.26666668,0.2,0.33333334,0.33333334,0.4,0.36666667,0.43333334,0.46666667,,0.6666667,0.6,0.7,0.6,0.7,0.73333335,0.53333336,0.8,0.76666665,0.6333333,0.6666667,0.6,0.7,0.6333333,0.6,0.6666667,0.6666667,0.76666665,0.7,0.6666667,0.8,0.76666665,0.8,0.8666667,0.73333335,0.73333335,0.76666665,0.8333333,0.8,0.8333333,0.8333333,0.8333333,0.8333333,0.9,0.8666667,0.8333333,0.9,0.9,0.9,0.8333333,0.93333334,0.8666667,0.76666665,0.8666667,0.76666665,0.93333334,0.93333334,0.93333334,0.96666664,0.8,0.93333334,0.96666664,0.8666667,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,1.0,1.0
+RealMLP (tuned),0.16666667,0.3,0.16666667,0.3,0.36666667,0.23333333,0.36666667,0.33333334,0.36666667,0.33333334,,0.5,0.46666667,0.5,0.53333336,0.6333333,0.45,0.56666666,0.6333333,0.5,0.6333333,0.46666667,0.6,0.6,0.53333336,0.56666666,0.56666666,0.6666667,0.6666667,0.56666666,0.7,0.6666667,0.7,0.76666665,0.7,0.7,0.6333333,0.8,0.6666667,0.73333335,0.73333335,0.7,0.73333335,0.8,0.8,0.8333333,0.73333335,0.8666667,0.8,0.8666667,0.8666667,0.8333333,0.76666665,0.9,0.8,0.93333334,0.8666667,0.78333336,0.8666667,0.8,0.96666664,0.93333334,0.8333333,0.93333334,0.93333334,0.93333334,0.96666664,0.93333334,0.95,0.96666664,0.93333334,1.0,1.0
+TabM (tuned + ensembled),0.06666667,0.16666667,0.2,0.26666668,0.3,0.3,0.33333334,0.33333334,0.3,0.4,0.5,,0.53333336,0.56666666,0.5,0.6666667,0.46666667,0.6,0.6,0.56666666,0.53333336,0.56666666,0.56666666,0.6,0.6333333,0.6,0.7,0.7,0.6333333,0.56666666,0.6333333,0.6333333,0.76666665,0.76666665,0.8333333,0.6,0.6666667,0.6666667,0.56666666,0.76666665,0.7,0.73333335,0.76666665,0.8,0.73333335,0.8,0.8333333,0.76666665,0.76666665,0.73333335,0.8666667,0.8333333,0.8333333,0.8333333,0.8333333,0.8333333,0.8,0.8333333,0.9,0.8333333,0.96666664,0.93333334,0.9,0.8333333,0.8333333,0.8666667,0.93333334,0.93333334,0.93333334,0.8666667,0.93333334,1.0,1.0
+iLTM (tuned + ensembled),0.2,0.26666668,0.16666667,0.23333333,0.26666668,0.16666667,0.33333334,0.46666667,0.43333334,0.3,0.53333336,0.46666667,,0.5,0.56666666,0.56666666,0.45,0.5,0.5,0.6333333,0.6,0.46666667,0.5,0.56666666,0.56666666,0.53333336,0.6333333,0.7,0.5,0.5,0.6,0.7,0.6666667,0.6666667,0.6333333,0.7,0.7,0.76666665,0.73333335,0.7,0.6666667,0.76666665,0.7,0.8333333,0.8,0.8,0.76666665,0.8333333,0.8333333,0.8,0.9,0.8333333,0.73333335,0.93333334,0.73333335,0.93333334,0.8666667,0.8333333,0.93333334,0.8,0.96666664,0.93333334,0.9,0.96666664,0.93333334,0.96666664,0.93333334,0.93333334,0.93333334,0.96666664,0.93333334,0.96666664,1.0
+TabDPT-Turbo (default),0.06666667,0.2,0.36666667,0.23333333,0.23333333,0.33333334,0.4,0.3,0.36666667,0.4,0.5,0.43333334,0.5,,0.46666667,0.56666666,0.6,0.5,0.53333336,0.56666666,0.56666666,0.6333333,0.43333334,0.46666667,0.5833333,0.46666667,0.56666666,0.6,0.53333336,0.7,0.56666666,0.6666667,0.6333333,0.6,0.6333333,0.6666667,0.6,0.6333333,0.7,0.7,0.6666667,0.7,0.8,0.73333335,0.7,0.76666665,0.6666667,0.76666665,0.8,0.8,0.8333333,0.8666667,0.76666665,0.8,0.76666665,0.8333333,0.8,0.9,0.8333333,0.8333333,0.8333333,0.96666664,0.93333334,0.9,0.93333334,0.93333334,0.93333334,0.8333333,0.8333333,0.93333334,0.9,1.0,1.0
+CatBoost (tuned),0.1,0.23333333,0.13333334,0.26666668,0.23333333,0.2,0.23333333,0.33333334,0.26666668,0.3,0.46666667,0.5,0.43333334,0.53333336,,0.43333334,0.46666667,0.46666667,0.6333333,0.46666667,0.53333336,0.46666667,0.5,0.46666667,0.5,0.5,0.56666666,0.6333333,0.56666666,0.5,0.53333336,0.7,0.6666667,0.6666667,0.6,0.6333333,0.6666667,0.7,0.73333335,0.73333335,0.73333335,0.76666665,0.7,0.73333335,0.76666665,0.8,0.7,0.8,0.76666665,0.73333335,0.8333333,0.8,0.76666665,0.8666667,0.76666665,0.9,0.8,0.76666665,0.9,0.76666665,0.93333334,0.8666667,0.8333333,0.9,0.8333333,0.93333334,0.93333334,0.96666664,0.96666664,0.9,0.96666664,1.0,1.0
+LightGBM (tuned + ensembled),0.06666667,0.1,0.1,0.1,0.2,0.2,0.2,0.23333333,0.26666668,0.26666668,0.36666667,0.33333334,0.43333334,0.43333334,0.56666666,,0.36666667,0.56666666,0.6,0.46666667,0.46666667,0.46666667,0.46666667,0.5,0.5,0.53333336,0.56666666,0.6333333,0.46666667,0.53333336,0.76666665,0.6666667,0.6666667,0.6666667,0.6333333,0.6666667,0.7,0.7,0.73333335,0.8,0.76666665,0.6666667,0.7,0.76666665,0.8333333,0.8,0.76666665,0.8666667,0.8666667,0.8,0.96666664,0.76666665,0.8,0.8666667,0.8,0.93333334,0.8333333,0.8,0.8666667,0.8,0.8666667,0.8666667,0.8333333,0.96666664,0.8666667,0.96666664,0.8666667,0.96666664,0.96666664,0.8666667,0.93333334,1.0,1.0
+LimiX (default),0.1,0.26666668,0.3,0.23333333,0.3,0.36666667,0.33333334,0.3,0.4,0.46666667,0.55,0.53333336,0.55,0.4,0.53333336,0.6333333,,0.56666666,0.53333336,0.56666666,0.6333333,0.43333334,0.56666666,0.56666666,0.46666667,0.56666666,0.6,0.6666667,0.56666666,0.53333336,0.6,0.7,0.6666667,0.6666667,0.6333333,0.6,0.6333333,0.6666667,0.7,0.7,0.73333335,0.6666667,0.6666667,0.6333333,0.6666667,0.6666667,0.7,0.7,0.7,0.7,0.7,0.73333335,0.68333334,0.7,0.6666667,0.76666665,0.76666665,0.78333336,0.76666665,0.76666665,0.8,0.8666667,0.8,0.9,0.93333334,0.8666667,0.8333333,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.96666664
+CatBoost (tuned + ensembled),0.06666667,0.16666667,0.06666667,0.2,0.23333333,0.16666667,0.16666667,0.33333334,0.26666668,0.2,0.43333334,0.4,0.5,0.5,0.53333336,0.43333334,0.43333334,,0.6,0.5,0.46666667,0.43333334,0.43333334,0.45,0.46666667,0.5,0.56666666,0.56666666,0.5,0.5,0.56666666,0.7,0.6,0.6666667,0.5,0.6333333,0.8,0.6666667,0.73333335,0.73333335,0.6666667,0.8,0.6333333,0.73333335,0.76666665,0.8,0.73333335,0.8,0.8333333,0.8,0.8333333,0.8333333,0.76666665,0.9,0.76666665,0.93333334,0.8666667,0.73333335,0.8666667,0.76666665,0.93333334,0.8666667,0.8,0.93333334,0.8666667,0.9,0.93333334,0.96666664,0.96666664,0.9,0.96666664,0.96666664,1.0
+CatBoost (default),0.13333334,0.13333334,0.1,0.16666667,0.2,0.23333333,0.16666667,0.3,0.16666667,0.23333333,0.36666667,0.4,0.5,0.46666667,0.36666667,0.4,0.46666667,0.4,,0.43333334,0.5,0.53333336,0.43333334,0.43333334,0.5,0.5,0.56666666,0.53333336,0.46666667,0.53333336,0.53333336,0.6333333,0.6666667,0.76666665,0.6,0.7,0.73333335,0.73333335,0.73333335,0.73333335,0.7,0.7,0.7,0.8,0.8,0.8333333,0.7,0.76666665,0.8,0.73333335,0.8,0.8333333,0.8,0.9,0.8,0.9,0.8333333,0.8,0.9,0.8,0.93333334,0.9,0.8333333,0.9,0.8333333,0.96666664,0.93333334,0.96666664,0.96666664,0.93333334,0.96666664,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.1,0.13333334,0.1,0.2,0.2,0.26666668,0.23333333,0.3,0.36666667,0.36666667,0.5,0.43333334,0.36666667,0.43333334,0.53333336,0.53333336,0.43333334,0.5,0.56666666,,0.53333336,0.53333336,0.38333333,0.5,0.53333336,0.5,0.6333333,0.56666666,0.4,0.5,0.6,0.6333333,0.7,0.6666667,0.6333333,0.7,0.8,0.6666667,0.73333335,0.7,0.73333335,0.6666667,0.6666667,0.76666665,0.76666665,0.73333335,0.7,0.7,0.73333335,0.73333335,0.8333333,0.73333335,0.8,0.8666667,0.8,0.8666667,0.76666665,0.8,0.8333333,0.8,0.8666667,0.8,0.8666667,0.8666667,0.8333333,0.9,0.8333333,0.93333334,0.93333334,0.8666667,0.96666664,0.96666664,1.0
+XGBoost (tuned + ensembled),0.1,0.16666667,0.1,0.3,0.23333333,0.16666667,0.2,0.26666668,0.3,0.33333334,0.36666667,0.46666667,0.4,0.43333334,0.46666667,0.53333336,0.36666667,0.53333336,0.5,0.46666667,,0.5,0.46666667,0.43333334,0.46666667,0.43333334,0.56666666,0.7,0.43333334,0.5,0.6,0.6,0.6333333,0.7,0.56666666,0.56666666,0.6666667,0.6666667,0.7,0.7,0.7,0.6666667,0.7,0.73333335,0.8333333,0.68333334,0.73333335,0.8,0.8666667,0.8,1.0,0.73333335,0.8,0.93333334,0.8,0.96666664,0.73333335,0.8,0.8,0.8,0.9,0.8333333,0.8,0.93333334,0.9,0.9,0.8333333,0.9,0.93333334,0.9,0.93333334,0.93333334,1.0
+TabDPT (tuned + ensembled),0.06666667,0.2,0.2,0.26666668,0.16666667,0.33333334,0.33333334,0.23333333,0.36666667,0.4,0.53333336,0.43333334,0.53333336,0.36666667,0.53333336,0.53333336,0.56666666,0.56666666,0.46666667,0.46666667,0.5,,0.46666667,0.43333334,0.5,0.43333334,0.5,0.5,0.5,0.6666667,0.53333336,0.6333333,0.56666666,0.6,0.6,0.6,0.5,0.6333333,0.6666667,0.7,0.6666667,0.7,0.73333335,0.6333333,0.6666667,0.6666667,0.8,0.73333335,0.76666665,0.8333333,0.76666665,0.8,0.76666665,0.73333335,0.73333335,0.76666665,0.76666665,0.93333334,0.8333333,0.8333333,0.8,0.96666664,0.9,0.93333334,0.96666664,0.9,0.8666667,0.8666667,0.8666667,0.96666664,0.9,0.96666664,1.0
+TabM (tuned),0.033333335,0.1,0.23333333,0.2,0.26666668,0.23333333,0.23333333,0.26666668,0.2,0.3,0.4,0.43333334,0.5,0.56666666,0.5,0.53333336,0.43333334,0.56666666,0.56666666,0.6166667,0.53333336,0.53333336,,0.56666666,0.56666666,0.53333336,0.6333333,0.6,0.56666666,0.56666666,0.53333336,0.6,0.6,0.6333333,0.6333333,0.5,0.6,0.6,0.56666666,0.6666667,0.6666667,0.6666667,0.73333335,0.7,0.73333335,0.76666665,0.8333333,0.6666667,0.73333335,0.7,0.76666665,0.8,0.76666665,0.76666665,0.76666665,0.8,0.8666667,0.8,0.9,0.8,0.96666664,0.8666667,0.9,0.8333333,0.8,0.8666667,0.93333334,0.8666667,0.8666667,0.8333333,0.93333334,0.96666664,0.96666664
+ModernNCA (tuned),0.1,0.16666667,0.1,0.23333333,0.2,0.3,0.2,0.3,0.23333333,0.36666667,0.4,0.4,0.43333334,0.53333336,0.53333336,0.5,0.43333334,0.55,0.56666666,0.5,0.56666666,0.56666666,0.43333334,,0.53333336,0.46666667,0.53333336,0.53333336,0.6,0.6333333,0.56666666,0.53333336,0.56666666,0.6666667,0.6333333,0.6,0.6333333,0.7,0.6666667,0.7,0.6666667,0.7,0.7,0.68333334,0.73333335,0.8,0.8,0.7,0.7,0.73333335,0.8333333,0.8,0.76666665,0.76666665,0.76666665,0.8333333,0.8,0.8,0.8333333,0.8666667,0.8333333,0.8666667,0.8333333,0.9,0.8333333,0.93333334,0.8333333,0.8666667,0.9,0.93333334,0.9,0.93333334,1.0
+TabSwift (default),0.13333334,0.13333334,0.26666668,0.26666668,0.2,0.3,0.23333333,0.2,0.33333334,0.4,0.46666667,0.36666667,0.43333334,0.41666666,0.5,0.5,0.53333336,0.53333336,0.5,0.46666667,0.53333336,0.5,0.43333334,0.46666667,,0.53333336,0.53333336,0.53333336,0.53333336,0.43333334,0.6,0.56666666,0.6333333,0.53333336,0.56666666,0.7,0.7,0.5,0.6666667,0.7,0.6,0.7,0.6666667,0.8,0.6666667,0.6,0.6666667,0.73333335,0.76666665,0.8,0.73333335,0.73333335,0.76666665,0.7,0.76666665,0.73333335,0.76666665,0.9,0.8,0.8333333,0.8,0.8666667,0.96666664,0.8,0.9,0.9,0.8666667,0.8666667,0.9,0.93333334,0.8666667,1.0,1.0
+ModernNCA (tuned + ensembled),0.13333334,0.1,0.1,0.13333334,0.16666667,0.33333334,0.3,0.36666667,0.33333334,0.33333334,0.43333334,0.4,0.46666667,0.53333336,0.5,0.46666667,0.43333334,0.5,0.5,0.5,0.56666666,0.56666666,0.46666667,0.53333336,0.46666667,,0.53333336,0.6,0.56666666,0.56666666,0.6333333,0.5,0.6333333,0.7,0.53333336,0.6,0.56666666,0.6,0.6333333,0.6333333,0.6333333,0.6,0.7,0.73333335,0.73333335,0.76666665,0.76666665,0.73333335,0.73333335,0.7,0.8666667,0.8,0.6666667,0.76666665,0.7,0.73333335,0.76666665,0.76666665,0.76666665,0.8,0.8,0.8333333,0.76666665,0.8333333,0.8,0.85,0.8666667,0.76666665,0.8,0.93333334,0.8,0.8666667,0.96666664
+ChimeraBoost (tuned),0.13333334,0.16666667,0.1,0.2,0.23333333,0.23333333,0.16666667,0.3,0.23333333,0.33333334,0.43333334,0.3,0.36666667,0.43333334,0.43333334,0.43333334,0.4,0.43333334,0.43333334,0.36666667,0.43333334,0.5,0.36666667,0.46666667,0.46666667,0.46666667,,0.53333336,0.5,0.46666667,0.5,0.46666667,0.7,0.6666667,0.53333336,0.6,0.6666667,0.6,0.73333335,0.7,0.7,0.6666667,0.6333333,0.76666665,0.76666665,0.6666667,0.6666667,0.7,0.7,0.73333335,0.8333333,0.76666665,0.76666665,0.8333333,0.76666665,0.8666667,0.73333335,0.76666665,0.8,0.8,0.8666667,0.8,0.8666667,0.8,0.8666667,0.8666667,0.8333333,0.9,0.9,0.8666667,0.93333334,0.93333334,1.0
+XGBoost (tuned),0.16666667,0.2,0.1,0.2,0.23333333,0.16666667,0.2,0.2,0.26666668,0.23333333,0.33333334,0.3,0.3,0.4,0.36666667,0.36666667,0.33333334,0.43333334,0.46666667,0.43333334,0.3,0.5,0.4,0.46666667,0.46666667,0.4,0.46666667,,0.46666667,0.46666667,0.53333336,0.53333336,0.53333336,0.6,0.53333336,0.6666667,0.56666666,0.53333336,0.6666667,0.6666667,0.7,0.6333333,0.6333333,0.7,0.8,0.6666667,0.6666667,0.76666665,0.8666667,0.8,0.9,0.76666665,0.8,0.8666667,0.8,0.9,0.7,0.76666665,0.8,0.8,0.8666667,0.8333333,0.8,0.9,0.8666667,0.93333334,0.8333333,0.8666667,0.9,0.8666667,0.93333334,0.93333334,1.0
+BetaTabPFN (default),0.033333335,0.033333335,0.16666667,0.13333334,0.06666667,0.23333333,0.13333334,0.13333334,0.16666667,0.3,0.33333334,0.36666667,0.5,0.46666667,0.43333334,0.53333336,0.43333334,0.5,0.53333336,0.6,0.56666666,0.5,0.43333334,0.4,0.46666667,0.43333334,0.5,0.53333336,,0.6,0.5,0.5,0.56666666,0.56666666,0.53333336,0.56666666,0.6,0.56666666,0.56666666,0.6333333,0.7,0.6666667,0.6666667,0.6333333,0.73333335,0.76666665,0.6333333,0.73333335,0.73333335,0.73333335,0.8,0.73333335,0.76666665,0.73333335,0.76666665,0.73333335,0.76666665,0.76666665,0.8,0.8333333,0.8333333,0.8333333,0.8666667,0.8,0.73333335,0.93333334,0.8333333,0.8666667,0.8666667,0.8333333,0.93333334,1.0,1.0
+TabDPT (tuned),0.033333335,0.06666667,0.23333333,0.23333333,0.16666667,0.26666668,0.23333333,0.26666668,0.23333333,0.33333334,0.43333334,0.43333334,0.5,0.3,0.5,0.46666667,0.46666667,0.5,0.46666667,0.5,0.5,0.33333334,0.43333334,0.36666667,0.56666666,0.43333334,0.53333336,0.53333336,0.4,,0.53333336,0.6,0.53333336,0.5,0.53333336,0.56666666,0.6,0.53333336,0.6,0.6333333,0.53333336,0.56666666,0.78333336,0.53333336,0.6,0.7,0.73333335,0.76666665,0.7,0.8,0.6333333,0.76666665,0.73333335,0.6333333,0.73333335,0.73333335,0.76666665,0.8666667,0.8333333,0.8,0.8,0.96666664,0.8333333,0.8666667,0.9,0.8333333,0.9,0.8,0.8333333,0.9,0.8666667,0.96666664,0.96666664
+LightGBM (tuned),0.033333335,0.0,0.06666667,0.0,0.16666667,0.13333334,0.1,0.2,0.1,0.2,0.3,0.36666667,0.4,0.43333334,0.46666667,0.23333333,0.4,0.43333334,0.46666667,0.4,0.4,0.46666667,0.46666667,0.43333334,0.4,0.36666667,0.5,0.46666667,0.5,0.46666667,,0.5,0.6333333,0.6,0.6333333,0.6333333,0.6333333,0.56666666,0.6666667,0.6666667,0.6333333,0.6333333,0.6666667,0.7,0.7,0.76666665,0.6333333,0.76666665,0.73333335,0.76666665,0.8333333,0.73333335,0.76666665,0.8,0.8,0.9,0.8,0.76666665,0.8333333,0.8,0.8,0.8333333,0.8,0.9,0.8333333,0.9,0.8666667,0.8333333,0.9,0.8666667,0.9,0.93333334,1.0
+iLTM (tuned),0.06666667,0.1,0.13333334,0.16666667,0.23333333,0.16666667,0.23333333,0.11666667,0.2,0.23333333,0.33333334,0.36666667,0.3,0.33333334,0.3,0.33333334,0.3,0.3,0.36666667,0.36666667,0.4,0.36666667,0.4,0.46666667,0.43333334,0.5,0.53333336,0.46666667,0.5,0.4,0.5,,0.56666666,0.56666666,0.53333336,0.6,0.43333334,0.56666666,0.6666667,0.73333335,0.6666667,0.7,0.6333333,0.73333335,0.56666666,0.6666667,0.6666667,0.6166667,0.8,0.8333333,0.73333335,0.76666665,0.76666665,0.8666667,0.76666665,0.8333333,0.7,0.7,0.8333333,0.8,0.9,0.9,0.8333333,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.9,0.96666664,0.8666667,1.0,1.0
+EBM (tuned + ensembled),0.06666667,0.16666667,0.16666667,0.1,0.2,0.23333333,0.16666667,0.16666667,0.16666667,0.2,0.3,0.23333333,0.33333334,0.36666667,0.33333334,0.33333334,0.33333334,0.4,0.33333334,0.3,0.36666667,0.43333334,0.4,0.43333334,0.36666667,0.36666667,0.3,0.46666667,0.43333334,0.46666667,0.36666667,0.43333334,,0.6333333,0.56666666,0.5,0.43333334,0.6,0.56666666,0.8333333,0.73333335,0.6666667,0.6666667,0.73333335,0.6666667,0.7,0.76666665,0.6,0.76666665,0.7,0.7,0.76666665,0.76666665,0.73333335,0.76666665,0.76666665,0.7,0.76666665,0.73333335,0.8,0.8333333,0.8666667,0.9,0.8,0.8333333,0.93333334,0.8333333,0.93333334,0.93333334,0.8666667,0.96666664,0.96666664,1.0
+RealMLP (default),0.1,0.1,0.16666667,0.13333334,0.16666667,0.23333333,0.2,0.23333333,0.2,0.13333334,0.23333333,0.23333333,0.33333334,0.4,0.33333334,0.33333334,0.33333334,0.33333334,0.23333333,0.33333334,0.3,0.4,0.36666667,0.33333334,0.46666667,0.3,0.33333334,0.4,0.43333334,0.5,0.4,0.43333334,0.36666667,,0.46666667,0.43333334,0.46666667,0.6333333,0.56666666,0.56666666,0.6,0.56666666,0.7,0.6,0.7,0.73333335,0.73333335,0.6666667,0.6666667,0.73333335,0.6666667,0.8,0.73333335,0.73333335,0.76666665,0.76666665,0.73333335,0.73333335,0.76666665,0.8,0.96666664,0.96666664,0.93333334,0.8333333,0.8666667,0.93333334,0.93333334,0.93333334,0.96666664,0.8666667,0.96666664,0.96666664,1.0
+TabM (default),0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.13333334,0.16666667,0.13333334,0.26666668,0.3,0.16666667,0.36666667,0.36666667,0.4,0.36666667,0.36666667,0.5,0.4,0.36666667,0.43333334,0.4,0.36666667,0.36666667,0.43333334,0.46666667,0.46666667,0.46666667,0.46666667,0.46666667,0.36666667,0.46666667,0.43333334,0.53333336,,0.46666667,0.56666666,0.56666666,0.56666666,0.51666665,0.53333336,0.6,0.6333333,0.73333335,0.6333333,0.7,0.7,0.6333333,0.6333333,0.7,0.73333335,0.8,0.73333335,0.76666665,0.73333335,0.76666665,0.8,0.76666665,0.8,0.8,0.9,0.8,0.8333333,0.8,0.8,0.8333333,0.8666667,0.8666667,0.8333333,0.8,0.8666667,0.96666664,0.93333334
+xRFM (tuned + ensembled),0.06666667,0.13333334,0.06666667,0.16666667,0.16666667,0.16666667,0.2,0.16666667,0.26666668,0.26666668,0.3,0.4,0.3,0.33333334,0.36666667,0.33333334,0.4,0.36666667,0.3,0.3,0.43333334,0.4,0.5,0.4,0.3,0.4,0.4,0.33333334,0.43333334,0.43333334,0.36666667,0.4,0.5,0.56666666,0.53333336,,0.36666667,0.5,0.6333333,0.56666666,0.53333336,0.6,0.56666666,0.6666667,0.6,0.6,0.73333335,0.6666667,0.7,0.73333335,0.6666667,0.73333335,0.7,0.76666665,0.7,0.76666665,0.7,0.8333333,0.76666665,0.73333335,0.8,0.9,0.8666667,0.9,0.9,0.9,0.8333333,0.93333334,0.96666664,0.9,0.93333334,0.96666664,0.96666664
+ChimeraBoost (default),0.06666667,0.1,0.06666667,0.16666667,0.2,0.16666667,0.2,0.16666667,0.23333333,0.23333333,0.36666667,0.33333334,0.3,0.4,0.33333334,0.3,0.36666667,0.2,0.26666668,0.2,0.33333334,0.5,0.4,0.36666667,0.3,0.43333334,0.33333334,0.43333334,0.4,0.4,0.36666667,0.56666666,0.56666666,0.53333336,0.43333334,0.6333333,,0.56666666,0.6333333,0.6333333,0.6666667,0.6666667,0.56666666,0.6666667,0.6666667,0.6,0.6333333,0.6666667,0.6666667,0.6666667,0.6666667,0.73333335,0.8,0.76666665,0.8,0.8333333,0.6666667,0.76666665,0.73333335,0.8,0.8333333,0.8,0.8333333,0.8,0.8333333,0.8666667,0.76666665,0.9,0.9,0.8666667,0.96666664,0.93333334,1.0
+TorchMLP (tuned + ensembled),0.033333335,0.13333334,0.1,0.1,0.13333334,0.1,0.2,0.16666667,0.2,0.16666667,0.2,0.33333334,0.23333333,0.36666667,0.3,0.3,0.33333334,0.33333334,0.26666668,0.33333334,0.33333334,0.36666667,0.4,0.3,0.5,0.4,0.4,0.46666667,0.43333334,0.46666667,0.43333334,0.43333334,0.4,0.36666667,0.43333334,0.5,0.43333334,,0.53333336,0.43333334,0.46666667,0.53333336,0.6,0.6,0.6666667,0.73333335,0.6,0.7,0.7,0.7,0.7,0.73333335,0.7,0.76666665,0.7,0.8333333,0.73333335,0.76666665,0.8,0.7,0.93333334,0.8666667,0.9,0.8,0.8,0.9,0.8666667,0.9,0.9,0.8333333,0.96666664,0.96666664,0.96666664
+xRFM (tuned),0.06666667,0.1,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.16666667,0.2,0.33333334,0.43333334,0.26666668,0.3,0.26666668,0.26666668,0.3,0.26666668,0.26666668,0.26666668,0.3,0.33333334,0.43333334,0.33333334,0.33333334,0.36666667,0.26666668,0.33333334,0.43333334,0.4,0.33333334,0.33333334,0.43333334,0.43333334,0.43333334,0.36666667,0.36666667,0.46666667,,0.46666667,0.56666666,0.6333333,0.5,0.6,0.5,0.6666667,0.56666666,0.6666667,0.73333335,0.73333335,0.6666667,0.6666667,0.6666667,0.6666667,0.7,0.73333335,0.6,0.73333335,0.7,0.73333335,0.8333333,0.8666667,0.8,0.8333333,0.9,0.9,0.8666667,0.9,0.9,0.96666664,0.8666667,0.96666664,1.0
+EBM (default),0.06666667,0.1,0.13333334,0.13333334,0.1,0.16666667,0.13333334,0.1,0.16666667,0.16666667,0.26666668,0.23333333,0.3,0.3,0.26666668,0.2,0.3,0.26666668,0.26666668,0.3,0.3,0.3,0.33333334,0.3,0.3,0.36666667,0.3,0.33333334,0.36666667,0.36666667,0.33333334,0.26666668,0.16666667,0.43333334,0.48333332,0.43333334,0.36666667,0.56666666,0.53333336,,0.5,0.53333336,0.53333336,0.6333333,0.5,0.6666667,0.6666667,0.6,0.7,0.7,0.6333333,0.7,0.73333335,0.7,0.73333335,0.73333335,0.76666665,0.76666665,0.8,0.76666665,0.8333333,0.8333333,0.8666667,0.76666665,0.8,0.93333334,0.8666667,0.8666667,0.9,0.8,0.93333334,0.96666664,1.0
+EBM (tuned),0.06666667,0.06666667,0.16666667,0.06666667,0.1,0.1,0.16666667,0.06666667,0.16666667,0.16666667,0.26666668,0.3,0.33333334,0.33333334,0.26666668,0.23333333,0.26666668,0.33333334,0.3,0.26666668,0.3,0.33333334,0.33333334,0.33333334,0.4,0.36666667,0.3,0.3,0.3,0.46666667,0.36666667,0.33333334,0.26666668,0.4,0.46666667,0.46666667,0.33333334,0.53333336,0.43333334,0.5,,0.6333333,0.6,0.56666666,0.56666666,0.56666666,0.6,0.6,0.6333333,0.7,0.6333333,0.7,0.76666665,0.6666667,0.76666665,0.7,0.7,0.73333335,0.7,0.8,0.8,0.8,0.8666667,0.8,0.8,0.93333334,0.8,0.8333333,0.8666667,0.8666667,0.8666667,0.96666664,1.0
+SAP-RPT-OSS (default),0.1,0.16666667,0.16666667,0.16666667,0.16666667,0.2,0.16666667,0.13333334,0.16666667,0.16666667,0.3,0.26666668,0.23333333,0.3,0.23333333,0.33333334,0.33333334,0.2,0.3,0.33333334,0.33333334,0.3,0.33333334,0.3,0.3,0.4,0.33333334,0.36666667,0.33333334,0.43333334,0.36666667,0.3,0.33333334,0.43333334,0.4,0.4,0.33333334,0.46666667,0.36666667,0.46666667,0.36666667,,0.56666666,0.6,0.46666667,0.56666666,0.56666666,0.6,0.6333333,0.73333335,0.6,0.7,0.6333333,0.56666666,0.6333333,0.6,0.76666665,0.7,0.8,0.8,0.73333335,0.8333333,0.8666667,0.8,0.8333333,0.8333333,0.8333333,0.8666667,0.8333333,0.8666667,0.8666667,0.93333334,1.0
+TabDPT (default),0.0,0.06666667,0.16666667,0.2,0.1,0.13333334,0.2,0.16666667,0.2,0.16666667,0.26666668,0.23333333,0.3,0.2,0.3,0.3,0.33333334,0.36666667,0.3,0.33333334,0.3,0.26666668,0.26666668,0.3,0.33333334,0.3,0.36666667,0.36666667,0.33333334,0.21666667,0.33333334,0.36666667,0.33333334,0.3,0.36666667,0.43333334,0.43333334,0.4,0.5,0.46666667,0.4,0.43333334,,0.5,0.5,0.46666667,0.6333333,0.53333336,0.56666666,0.6333333,0.6,0.7,0.6,0.6,0.6333333,0.6666667,0.6666667,0.76666665,0.76666665,0.6666667,0.73333335,0.8333333,0.8,0.73333335,0.73333335,0.73333335,0.8,0.76666665,0.76666665,0.8,0.8,0.8666667,0.96666664
+FastaiMLP (tuned + ensembled),0.1,0.13333334,0.13333334,0.13333334,0.033333335,0.13333334,0.2,0.13333334,0.2,0.1,0.2,0.2,0.16666667,0.26666668,0.26666668,0.23333333,0.36666667,0.26666668,0.2,0.23333333,0.26666668,0.36666667,0.3,0.31666666,0.2,0.26666668,0.23333333,0.3,0.36666667,0.46666667,0.3,0.26666668,0.26666668,0.4,0.26666668,0.33333334,0.33333334,0.4,0.4,0.36666667,0.43333334,0.4,0.5,,0.43333334,0.56666666,0.6,0.56666666,0.6,0.6,0.53333336,0.76666665,0.6,0.56666666,0.6,0.6333333,0.73333335,0.7,0.73333335,0.7,0.76666665,0.96666664,0.8333333,0.76666665,0.8,0.8666667,0.9,0.8,0.8333333,0.8666667,0.8666667,0.9,0.96666664
+XGBoost (default),0.033333335,0.06666667,0.1,0.06666667,0.13333334,0.1,0.06666667,0.1,0.1,0.13333334,0.2,0.26666668,0.2,0.3,0.23333333,0.16666667,0.33333334,0.23333333,0.2,0.23333333,0.16666667,0.33333334,0.26666668,0.26666668,0.33333334,0.26666668,0.23333333,0.2,0.26666668,0.4,0.3,0.43333334,0.33333334,0.3,0.36666667,0.4,0.33333334,0.33333334,0.5,0.5,0.43333334,0.53333336,0.5,0.56666666,,0.5,0.56666666,0.6666667,0.6666667,0.6666667,0.6,0.6,0.6666667,0.6666667,0.6666667,0.6666667,0.6,0.6333333,0.7,0.7,0.7,0.73333335,0.7,0.76666665,0.73333335,0.76666665,0.8,0.8,0.8333333,0.8666667,0.8,0.8666667,0.93333334
+TorchMLP (tuned),0.0,0.033333335,0.1,0.033333335,0.1,0.06666667,0.13333334,0.13333334,0.1,0.16666667,0.16666667,0.2,0.2,0.23333333,0.2,0.2,0.33333334,0.2,0.16666667,0.26666668,0.31666666,0.33333334,0.23333333,0.2,0.4,0.23333333,0.33333334,0.33333334,0.23333333,0.3,0.23333333,0.33333334,0.3,0.26666668,0.3,0.4,0.4,0.26666668,0.33333334,0.33333334,0.43333334,0.43333334,0.53333336,0.43333334,0.5,,0.6,0.6,0.5833333,0.6333333,0.55,0.53333336,0.73333335,0.6,0.7,0.6,0.73333335,0.7,0.73333335,0.76666665,0.9,0.76666665,0.9,0.7,0.76666665,0.8333333,0.8333333,0.8333333,0.8333333,0.8,0.9,0.93333334,0.96666664
+ModernNCA (default),0.06666667,0.06666667,0.06666667,0.06666667,0.1,0.1,0.16666667,0.1,0.13333334,0.1,0.26666668,0.16666667,0.23333333,0.33333334,0.3,0.23333333,0.3,0.26666668,0.3,0.3,0.26666668,0.2,0.16666667,0.2,0.33333334,0.23333333,0.33333334,0.33333334,0.36666667,0.26666668,0.36666667,0.33333334,0.23333333,0.26666668,0.3,0.26666668,0.36666667,0.4,0.43333334,0.33333334,0.4,0.43333334,0.36666667,0.4,0.43333334,0.4,,0.6,0.53333336,0.53333336,0.53333336,0.6,0.76666665,0.6333333,0.76666665,0.6333333,0.6666667,0.7,0.7,0.8333333,0.8333333,0.8,0.73333335,0.8,0.8333333,0.8666667,0.8,0.8,0.8,0.76666665,0.8666667,0.96666664,0.96666664
+iLTM (default),0.06666667,0.033333335,0.06666667,0.06666667,0.16666667,0.1,0.13333334,0.13333334,0.16666667,0.1,0.13333334,0.23333333,0.16666667,0.23333333,0.2,0.13333334,0.3,0.2,0.23333333,0.3,0.2,0.26666668,0.33333334,0.3,0.26666668,0.26666668,0.3,0.23333333,0.26666668,0.23333333,0.23333333,0.38333333,0.4,0.33333334,0.36666667,0.33333334,0.33333334,0.3,0.33333334,0.4,0.4,0.4,0.46666667,0.43333334,0.33333334,0.4,0.4,,0.53333336,0.6,0.4,0.53333336,0.6666667,0.56666666,0.6666667,0.53333336,0.53333336,0.6333333,0.5,0.73333335,0.6666667,0.6666667,0.7,0.6666667,0.7,0.8333333,0.56666666,0.73333335,0.73333335,0.76666665,0.8333333,0.8666667,0.93333334
+ExtraTrees (tuned + ensembled),0.033333335,0.033333335,0.06666667,0.06666667,0.1,0.1,0.033333335,0.13333334,0.0,0.1,0.2,0.23333333,0.16666667,0.2,0.23333333,0.13333334,0.3,0.16666667,0.2,0.26666668,0.13333334,0.23333333,0.26666668,0.3,0.23333333,0.26666668,0.3,0.13333334,0.26666668,0.3,0.26666668,0.2,0.23333333,0.33333334,0.36666667,0.3,0.33333334,0.3,0.26666668,0.3,0.36666667,0.36666667,0.43333334,0.4,0.33333334,0.41666666,0.46666667,0.46666667,,0.6666667,0.46666667,0.43333334,0.6,0.53333336,0.6333333,0.6666667,0.6,0.6666667,0.6333333,0.7,0.56666666,0.7,0.76666665,0.8666667,0.9,0.8666667,0.73333335,0.76666665,0.8666667,0.76666665,0.8333333,0.9,1.0
+ExtraTrees (tuned),0.033333335,0.033333335,0.06666667,0.06666667,0.06666667,0.13333334,0.033333335,0.13333334,0.1,0.16666667,0.13333334,0.26666668,0.2,0.2,0.26666668,0.2,0.3,0.2,0.26666668,0.26666668,0.2,0.16666667,0.3,0.26666668,0.2,0.3,0.26666668,0.2,0.26666668,0.2,0.23333333,0.16666667,0.3,0.26666668,0.3,0.26666668,0.33333334,0.3,0.26666668,0.3,0.3,0.26666668,0.36666667,0.4,0.33333334,0.36666667,0.46666667,0.4,0.33333334,,0.36666667,0.43333334,0.56666666,0.43333334,0.56666666,0.6,0.6,0.6333333,0.6,0.73333335,0.5,0.6,0.73333335,0.8,0.8,0.76666665,0.7,0.7,0.7,0.73333335,0.8333333,0.96666664,0.96666664
+LightGBM (default),0.033333335,0.033333335,0.033333335,0.033333335,0.033333335,0.06666667,0.0,0.0,0.033333335,0.06666667,0.13333334,0.13333334,0.1,0.16666667,0.16666667,0.033333335,0.3,0.16666667,0.2,0.16666667,0.0,0.23333333,0.23333333,0.16666667,0.26666668,0.13333334,0.16666667,0.1,0.2,0.36666667,0.16666667,0.26666668,0.3,0.33333334,0.26666668,0.33333334,0.33333334,0.3,0.33333334,0.36666667,0.36666667,0.4,0.4,0.46666667,0.4,0.45,0.46666667,0.6,0.53333336,0.6333333,,0.53333336,0.6,0.5,0.6333333,0.53333336,0.6,0.56666666,0.6333333,0.7,0.6666667,0.6666667,0.7,0.76666665,0.76666665,0.7,0.73333335,0.7,0.7,0.8,0.8333333,0.9,0.93333334
+FastaiMLP (tuned),0.06666667,0.06666667,0.1,0.13333334,0.06666667,0.1,0.1,0.033333335,0.13333334,0.13333334,0.16666667,0.16666667,0.16666667,0.13333334,0.2,0.23333333,0.26666668,0.16666667,0.16666667,0.26666668,0.26666668,0.2,0.2,0.2,0.26666668,0.2,0.23333333,0.23333333,0.26666668,0.23333333,0.26666668,0.23333333,0.23333333,0.2,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.3,0.3,0.3,0.3,0.23333333,0.4,0.46666667,0.4,0.46666667,0.56666666,0.56666666,0.46666667,,0.56666666,0.46666667,0.56666666,0.53333336,0.56666666,0.56666666,0.6333333,0.6666667,0.6333333,0.76666665,0.76666665,0.7,0.73333335,0.73333335,0.73333335,0.7,0.7,0.76666665,0.8,0.8666667,0.93333334
+TabSTAR (tuned),0.06666667,0.16666667,0.16666667,0.2,0.13333334,0.23333333,0.2,0.13333334,0.2,0.23333333,0.23333333,0.16666667,0.26666668,0.23333333,0.23333333,0.2,0.31666666,0.23333333,0.2,0.2,0.2,0.23333333,0.23333333,0.23333333,0.23333333,0.33333334,0.23333333,0.2,0.23333333,0.26666668,0.23333333,0.23333333,0.23333333,0.26666668,0.26666668,0.3,0.2,0.3,0.33333334,0.26666668,0.23333333,0.36666667,0.4,0.4,0.33333334,0.26666668,0.23333333,0.33333334,0.4,0.43333334,0.4,0.43333334,,0.4,0.7,0.43333334,0.6,0.53333336,0.6,0.93333334,0.53333336,0.6,0.7,0.73333335,0.7,0.6333333,0.6333333,0.6333333,0.6666667,0.76666665,0.73333335,0.8333333,0.9
+RandomForest (tuned + ensembled),0.033333335,0.06666667,0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.1,0.06666667,0.13333334,0.1,0.16666667,0.06666667,0.2,0.13333334,0.13333334,0.3,0.1,0.1,0.13333334,0.06666667,0.26666668,0.23333333,0.23333333,0.3,0.23333333,0.16666667,0.13333334,0.26666668,0.36666667,0.2,0.13333334,0.26666668,0.26666668,0.23333333,0.23333333,0.23333333,0.23333333,0.33333334,0.3,0.33333334,0.43333334,0.4,0.43333334,0.33333334,0.4,0.36666667,0.43333334,0.46666667,0.56666666,0.5,0.53333336,0.6,,0.6333333,0.6666667,0.53333336,0.56666666,0.6333333,0.73333335,0.7,0.7,0.6666667,0.8666667,0.8,0.76666665,0.6333333,0.73333335,0.73333335,0.8,0.8333333,0.9,0.96666664
+TabSTAR (tuned + ensembled),0.1,0.16666667,0.16666667,0.2,0.13333334,0.2,0.2,0.13333334,0.2,0.23333333,0.2,0.16666667,0.26666668,0.23333333,0.23333333,0.2,0.33333334,0.23333333,0.2,0.2,0.2,0.26666668,0.23333333,0.23333333,0.23333333,0.3,0.23333333,0.2,0.23333333,0.26666668,0.2,0.23333333,0.23333333,0.23333333,0.26666668,0.3,0.2,0.3,0.3,0.26666668,0.23333333,0.36666667,0.36666667,0.4,0.33333334,0.3,0.23333333,0.33333334,0.36666667,0.43333334,0.36666667,0.43333334,0.3,0.36666667,,0.43333334,0.6,0.53333336,0.6,0.96666664,0.56666666,0.6,0.6666667,0.7,0.6666667,0.6,0.6333333,0.6333333,0.6333333,0.73333335,0.73333335,0.8333333,0.9
+RandomForest (tuned),0.033333335,0.0,0.06666667,0.033333335,0.06666667,0.033333335,0.033333335,0.033333335,0.06666667,0.06666667,0.06666667,0.16666667,0.06666667,0.16666667,0.1,0.06666667,0.23333333,0.06666667,0.1,0.13333334,0.033333335,0.23333333,0.2,0.16666667,0.26666668,0.26666668,0.13333334,0.1,0.26666668,0.26666668,0.1,0.16666667,0.23333333,0.23333333,0.23333333,0.23333333,0.16666667,0.16666667,0.26666668,0.26666668,0.3,0.4,0.33333334,0.36666667,0.33333334,0.4,0.36666667,0.46666667,0.33333334,0.4,0.46666667,0.46666667,0.56666666,0.33333334,0.56666666,,0.53333336,0.6,0.56666666,0.7,0.56666666,0.76666665,0.73333335,0.78333336,0.76666665,0.8,0.6,0.7,0.7,0.8333333,0.8,0.9,0.96666664
+PerpetualBooster (tuned + ensembled),0.0,0.033333335,0.1,0.033333335,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.2,0.13333334,0.2,0.2,0.16666667,0.23333333,0.13333334,0.16666667,0.23333333,0.26666668,0.23333333,0.13333334,0.2,0.23333333,0.23333333,0.26666668,0.3,0.23333333,0.23333333,0.2,0.3,0.3,0.26666668,0.2,0.3,0.33333334,0.26666668,0.4,0.23333333,0.3,0.23333333,0.33333334,0.26666668,0.4,0.26666668,0.33333334,0.46666667,0.4,0.4,0.4,0.43333334,0.4,0.46666667,0.4,0.46666667,,0.5,0.8,0.4,0.5,0.53333336,0.6,0.7,0.6,0.6333333,0.8333333,0.7,0.7,0.6666667,0.7,0.7,0.93333334
+OrionMSP (default),0.0,0.0,0.0,0.033333335,0.06666667,0.06666667,0.033333335,0.0,0.0,0.06666667,0.21666667,0.16666667,0.16666667,0.1,0.23333333,0.2,0.21666667,0.26666668,0.2,0.2,0.2,0.06666667,0.2,0.2,0.1,0.23333333,0.23333333,0.23333333,0.23333333,0.13333334,0.23333333,0.3,0.23333333,0.26666668,0.23333333,0.16666667,0.23333333,0.23333333,0.26666668,0.23333333,0.26666668,0.3,0.23333333,0.3,0.36666667,0.3,0.3,0.36666667,0.33333334,0.36666667,0.43333334,0.43333334,0.46666667,0.43333334,0.46666667,0.4,0.5,,0.5,0.6333333,0.5,0.5,0.7,0.6,0.7,0.6,0.6666667,0.7,0.7,0.73333335,0.73333335,0.7,0.96666664
+PerpetualBooster (tuned),0.0,0.0,0.06666667,0.0,0.033333335,0.033333335,0.033333335,0.0,0.033333335,0.033333335,0.13333334,0.1,0.06666667,0.16666667,0.1,0.13333334,0.23333333,0.13333334,0.1,0.16666667,0.2,0.16666667,0.1,0.16666667,0.2,0.23333333,0.2,0.2,0.2,0.16666667,0.16666667,0.16666667,0.26666668,0.23333333,0.2,0.23333333,0.26666668,0.2,0.3,0.2,0.3,0.2,0.23333333,0.26666668,0.3,0.26666668,0.3,0.5,0.36666667,0.4,0.36666667,0.36666667,0.4,0.36666667,0.4,0.43333334,0.2,0.5,,0.4,0.46666667,0.5,0.56666666,0.6,0.56666666,0.53333336,0.8,0.6666667,0.6333333,0.6666667,0.6666667,0.6,0.93333334
+TabSTAR (default),0.06666667,0.16666667,0.16666667,0.2,0.13333334,0.16666667,0.16666667,0.13333334,0.16666667,0.2,0.2,0.16666667,0.2,0.16666667,0.23333333,0.2,0.23333333,0.23333333,0.2,0.2,0.2,0.16666667,0.2,0.13333334,0.16666667,0.2,0.2,0.2,0.16666667,0.2,0.2,0.2,0.2,0.2,0.2,0.26666668,0.2,0.3,0.26666668,0.23333333,0.2,0.2,0.33333334,0.3,0.3,0.23333333,0.16666667,0.26666668,0.3,0.26666668,0.3,0.33333334,0.06666667,0.26666668,0.033333335,0.3,0.6,0.36666667,0.6,,0.4,0.33333334,0.56666666,0.56666666,0.5,0.5,0.6333333,0.5,0.5,0.6,0.53333336,0.6666667,0.8666667
+TorchMLP (default),0.0,0.033333335,0.06666667,0.033333335,0.0,0.033333335,0.06666667,0.0,0.06666667,0.06666667,0.033333335,0.033333335,0.033333335,0.16666667,0.06666667,0.13333334,0.2,0.06666667,0.06666667,0.13333334,0.1,0.2,0.033333335,0.16666667,0.2,0.2,0.13333334,0.13333334,0.16666667,0.2,0.2,0.1,0.16666667,0.033333335,0.1,0.2,0.16666667,0.06666667,0.16666667,0.16666667,0.2,0.26666668,0.26666668,0.23333333,0.3,0.1,0.16666667,0.33333334,0.43333334,0.5,0.33333334,0.36666667,0.46666667,0.3,0.43333334,0.43333334,0.5,0.5,0.53333336,0.6,,0.53333336,0.6333333,0.6333333,0.6666667,0.6666667,0.6666667,0.6333333,0.7,0.6666667,0.76666665,0.8333333,0.93333334
+FastaiMLP (default),0.0,0.0,0.033333335,0.033333335,0.0,0.033333335,0.033333335,0.06666667,0.06666667,0.033333335,0.06666667,0.06666667,0.06666667,0.033333335,0.13333334,0.13333334,0.13333334,0.13333334,0.1,0.2,0.16666667,0.033333335,0.13333334,0.13333334,0.13333334,0.16666667,0.2,0.16666667,0.16666667,0.033333335,0.16666667,0.1,0.13333334,0.033333335,0.2,0.1,0.2,0.13333334,0.13333334,0.16666667,0.2,0.16666667,0.16666667,0.033333335,0.26666668,0.23333333,0.2,0.33333334,0.3,0.4,0.33333334,0.23333333,0.4,0.3,0.4,0.23333333,0.46666667,0.5,0.5,0.6666667,0.46666667,,0.56666666,0.5,0.56666666,0.56666666,0.7,0.56666666,0.56666666,0.6666667,0.6,0.8,0.93333334
+TabFlex (default),0.0,0.06666667,0.06666667,0.1,0.0,0.06666667,0.13333334,0.0,0.13333334,0.13333334,0.16666667,0.1,0.1,0.06666667,0.16666667,0.16666667,0.2,0.2,0.16666667,0.13333334,0.2,0.1,0.1,0.16666667,0.033333335,0.23333333,0.13333334,0.2,0.13333334,0.16666667,0.2,0.16666667,0.1,0.06666667,0.16666667,0.13333334,0.16666667,0.1,0.2,0.13333334,0.13333334,0.13333334,0.2,0.16666667,0.3,0.1,0.26666668,0.3,0.23333333,0.26666668,0.3,0.23333333,0.3,0.33333334,0.33333334,0.26666668,0.4,0.3,0.43333334,0.43333334,0.36666667,0.43333334,,0.53333336,0.5,0.46666667,0.56666666,0.6,0.6,0.56666666,0.6333333,0.6666667,0.8333333
+RandomForest (default),0.033333335,0.033333335,0.06666667,0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.16666667,0.033333335,0.1,0.1,0.033333335,0.1,0.06666667,0.1,0.13333334,0.06666667,0.06666667,0.16666667,0.1,0.2,0.16666667,0.2,0.1,0.2,0.13333334,0.1,0.13333334,0.2,0.16666667,0.2,0.1,0.2,0.2,0.16666667,0.23333333,0.2,0.2,0.26666668,0.23333333,0.23333333,0.3,0.2,0.33333334,0.13333334,0.2,0.23333333,0.3,0.26666668,0.13333334,0.3,0.21666667,0.3,0.4,0.4,0.43333334,0.36666667,0.5,0.46666667,,0.5,0.5,0.53333336,0.56666666,0.56666666,0.6666667,0.6333333,0.6666667,0.96666664
+ExtraTrees (default),0.033333335,0.033333335,0.033333335,0.06666667,0.1,0.033333335,0.033333335,0.0,0.033333335,0.033333335,0.06666667,0.16666667,0.06666667,0.06666667,0.16666667,0.13333334,0.06666667,0.13333334,0.16666667,0.16666667,0.1,0.033333335,0.2,0.16666667,0.1,0.2,0.13333334,0.13333334,0.26666668,0.1,0.16666667,0.13333334,0.16666667,0.13333334,0.2,0.1,0.16666667,0.2,0.1,0.2,0.2,0.16666667,0.26666668,0.2,0.26666668,0.23333333,0.16666667,0.3,0.1,0.2,0.23333333,0.26666668,0.3,0.2,0.33333334,0.23333333,0.4,0.3,0.43333334,0.5,0.33333334,0.43333334,0.5,0.5,,0.43333334,0.53333336,0.56666666,0.56666666,0.6,0.56666666,0.6666667,0.93333334
+KNN (tuned + ensembled),0.06666667,0.033333335,0.033333335,0.0,0.033333335,0.033333335,0.06666667,0.033333335,0.06666667,0.033333335,0.06666667,0.13333334,0.033333335,0.06666667,0.06666667,0.033333335,0.13333334,0.1,0.033333335,0.1,0.1,0.1,0.13333334,0.06666667,0.1,0.15,0.13333334,0.06666667,0.06666667,0.16666667,0.1,0.06666667,0.06666667,0.06666667,0.16666667,0.1,0.13333334,0.1,0.1,0.06666667,0.06666667,0.16666667,0.26666668,0.13333334,0.23333333,0.16666667,0.13333334,0.16666667,0.13333334,0.23333333,0.3,0.26666668,0.36666667,0.23333333,0.4,0.2,0.36666667,0.4,0.46666667,0.5,0.33333334,0.43333334,0.53333336,0.5,0.56666666,,0.53333336,0.6666667,0.6666667,0.7,0.7,0.96666664,0.96666664
+PerpetualBooster (default),0.0,0.033333335,0.06666667,0.033333335,0.033333335,0.033333335,0.06666667,0.06666667,0.06666667,0.033333335,0.033333335,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.16666667,0.06666667,0.06666667,0.16666667,0.16666667,0.13333334,0.06666667,0.16666667,0.13333334,0.13333334,0.16666667,0.16666667,0.16666667,0.1,0.13333334,0.13333334,0.16666667,0.06666667,0.13333334,0.16666667,0.23333333,0.13333334,0.13333334,0.13333334,0.2,0.16666667,0.2,0.1,0.2,0.16666667,0.2,0.43333334,0.26666668,0.3,0.26666668,0.26666668,0.36666667,0.36666667,0.36666667,0.4,0.16666667,0.33333334,0.2,0.36666667,0.33333334,0.3,0.43333334,0.46666667,0.46666667,0.46666667,,0.56666666,0.56666666,0.56666666,0.56666666,0.5,0.8666667
+Linear (tuned + ensembled),0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.033333335,0.1,0.06666667,0.1,0.033333335,0.06666667,0.06666667,0.06666667,0.16666667,0.033333335,0.033333335,0.13333334,0.033333335,0.033333335,0.06666667,0.1,0.13333334,0.13333334,0.13333334,0.13333334,0.23333333,0.1,0.13333334,0.13333334,0.2,0.16666667,0.13333334,0.06666667,0.06666667,0.13333334,0.06666667,0.1,0.1,0.1,0.13333334,0.16666667,0.13333334,0.23333333,0.2,0.2,0.16666667,0.2,0.26666668,0.23333333,0.3,0.3,0.3,0.36666667,0.26666668,0.36666667,0.3,0.3,0.3,0.33333334,0.5,0.36666667,0.43333334,0.4,0.43333334,0.43333334,0.33333334,0.43333334,,0.76666665,0.56666666,0.6666667,0.56666666,0.8333333
+Linear (tuned),0.033333335,0.033333335,0.033333335,0.033333335,0.033333335,0.033333335,0.06666667,0.06666667,0.1,0.033333335,0.05,0.06666667,0.06666667,0.16666667,0.033333335,0.033333335,0.13333334,0.033333335,0.033333335,0.06666667,0.06666667,0.13333334,0.13333334,0.1,0.1,0.2,0.1,0.1,0.13333334,0.16666667,0.1,0.1,0.06666667,0.033333335,0.16666667,0.033333335,0.1,0.1,0.1,0.1,0.13333334,0.16666667,0.23333333,0.16666667,0.16666667,0.16666667,0.2,0.26666668,0.13333334,0.3,0.3,0.3,0.33333334,0.26666668,0.36666667,0.3,0.3,0.3,0.36666667,0.5,0.3,0.43333334,0.4,0.43333334,0.43333334,0.33333334,0.43333334,0.23333333,,0.56666666,0.6,0.53333336,0.8
+xRFM (default),0.033333335,0.033333335,0.033333335,0.06666667,0.06666667,0.033333335,0.06666667,0.033333335,0.06666667,0.033333335,0.033333335,0.13333334,0.033333335,0.06666667,0.1,0.13333334,0.06666667,0.1,0.06666667,0.13333334,0.1,0.033333335,0.16666667,0.06666667,0.06666667,0.06666667,0.13333334,0.13333334,0.16666667,0.1,0.13333334,0.033333335,0.13333334,0.13333334,0.2,0.1,0.13333334,0.16666667,0.033333335,0.2,0.13333334,0.13333334,0.2,0.13333334,0.13333334,0.2,0.23333333,0.23333333,0.23333333,0.26666668,0.2,0.23333333,0.23333333,0.2,0.26666668,0.16666667,0.33333334,0.26666668,0.33333334,0.4,0.33333334,0.33333334,0.43333334,0.33333334,0.4,0.3,0.43333334,0.43333334,0.43333334,,0.43333334,0.56666666,0.8
+Linear (default),0.0,0.0,0.033333335,0.033333335,0.0,0.0,0.06666667,0.033333335,0.06666667,0.033333335,0.06666667,0.06666667,0.06666667,0.1,0.033333335,0.06666667,0.13333334,0.033333335,0.033333335,0.033333335,0.06666667,0.1,0.06666667,0.1,0.13333334,0.2,0.06666667,0.06666667,0.06666667,0.13333334,0.1,0.13333334,0.033333335,0.033333335,0.13333334,0.06666667,0.033333335,0.033333335,0.13333334,0.06666667,0.13333334,0.13333334,0.2,0.13333334,0.2,0.1,0.13333334,0.16666667,0.16666667,0.16666667,0.16666667,0.2,0.26666668,0.16666667,0.26666668,0.2,0.3,0.26666668,0.33333334,0.46666667,0.23333333,0.4,0.36666667,0.36666667,0.43333334,0.3,0.43333334,0.33333334,0.4,0.56666666,,0.56666666,0.73333335
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.033333335,0.0,0.033333335,0.0,0.0,0.0,0.033333335,0.0,0.0,0.0,0.13333334,0.033333335,0.0,0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.0,0.13333334,0.06666667,0.06666667,0.0,0.033333335,0.06666667,0.0,0.033333335,0.033333335,0.033333335,0.033333335,0.06666667,0.033333335,0.033333335,0.033333335,0.033333335,0.06666667,0.13333334,0.1,0.13333334,0.06666667,0.033333335,0.13333334,0.1,0.033333335,0.1,0.13333334,0.16666667,0.1,0.16666667,0.1,0.3,0.3,0.4,0.33333334,0.16666667,0.2,0.33333334,0.33333334,0.33333334,0.033333335,0.5,0.43333334,0.46666667,0.43333334,0.43333334,,0.93333334
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.033333335,0.0,0.0,0.0,0.0,0.0,0.033333335,0.0,0.0,0.033333335,0.0,0.0,0.0,0.033333335,0.0,0.0,0.0,0.0,0.06666667,0.033333335,0.0,0.033333335,0.0,0.0,0.0,0.0,0.033333335,0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.0,0.033333335,0.06666667,0.06666667,0.1,0.033333335,0.1,0.033333335,0.06666667,0.033333335,0.06666667,0.13333334,0.06666667,0.06666667,0.16666667,0.033333335,0.06666667,0.033333335,0.13333334,0.16666667,0.2,0.2,0.26666668,0.06666667,
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
index cd6329ff50be2b3370d9aebb4202d81f8453220a..93dd0e68bab04d94a088dca2994fa222c97e3657 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Binary | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..537894af49d08ca34aa5559f2af00d802c9b78b5
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_binary/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2130.0,"score":0.759,"rank":4.11,"hrank":2.26,"imp":2.931,"train_time":124.09,"predict_time":2.117,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+248\/-71","imputed":false},{"position":1,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2070.0,"score":0.726,"rank":5.22,"hrank":2.78,"imp":4.307,"train_time":310.56,"predict_time":1.824,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+216\/-59","imputed":false},{"position":2,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1972.0,"score":0.733,"rank":7.67,"hrank":2.25,"imp":5.662,"train_time":42.22,"predict_time":19.82,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+364\/-147","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1871.0,"score":0.611,"rank":11.11,"hrank":2.76,"imp":6.163,"train_time":1.19,"predict_time":0.309,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+329\/-188","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1858.0,"score":0.516,"rank":11.67,"hrank":8.24,"imp":8.445,"train_time":2.62,"predict_time":0.882,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+177\/-76","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1807.0,"score":0.475,"rank":13.89,"hrank":6.3,"imp":8.984,"train_time":0.64,"predict_time":0.201,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+232\/-143","imputed":false},{"position":6,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1746.0,"score":0.473,"rank":16.89,"hrank":9.66,"imp":7.213,"train_time":2149.06,"predict_time":6.373,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+211\/-151","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1731.0,"score":0.384,"rank":17.67,"hrank":10.07,"imp":8.547,"train_time":841.84,"predict_time":11.882,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+198\/-152","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1727.0,"score":0.449,"rank":17.89,"hrank":12.22,"imp":8.133,"train_time":1655.56,"predict_time":2.483,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-48","imputed":false},{"position":9,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1725.0,"score":0.428,"rank":18.0,"hrank":11.63,"imp":10.268,"train_time":1878.64,"predict_time":70.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+204\/-118","imputed":false},{"position":10,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1725.0,"score":0.392,"rank":18.0,"hrank":13.06,"imp":9.778,"train_time":114.48,"predict_time":0.404,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+200\/-98","imputed":false},{"position":11,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1717.0,"score":0.39,"rank":18.44,"hrank":12.03,"imp":9.401,"train_time":607.67,"predict_time":0.556,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+188\/-100","imputed":false},{"position":12,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1713.0,"score":0.437,"rank":18.67,"hrank":12.25,"imp":9.845,"train_time":92.46,"predict_time":0.366,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+183\/-128","imputed":false},{"position":13,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1700.0,"score":0.425,"rank":19.39,"hrank":9.72,"imp":9.155,"train_time":339.63,"predict_time":0.154,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+178\/-106","imputed":false},{"position":14,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1697.0,"score":0.412,"rank":19.56,"hrank":11.0,"imp":9.377,"train_time":339.63,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+196\/-118","imputed":false},{"position":15,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1691.0,"score":0.408,"rank":19.89,"hrank":14.43,"imp":10.033,"train_time":92.46,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+196\/-134","imputed":false},{"position":16,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1683.0,"score":0.43,"rank":20.33,"hrank":9.75,"imp":8.51,"train_time":1.49,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-82","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1673.0,"score":0.356,"rank":20.89,"hrank":15.69,"imp":10.891,"train_time":87.3,"predict_time":0.533,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-95","imputed":false},{"position":18,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1672.0,"score":0.325,"rank":21.0,"hrank":15.75,"imp":10.379,"train_time":841.84,"predict_time":1.263,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+185\/-109","imputed":false},{"position":19,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1661.0,"score":0.336,"rank":21.61,"hrank":13.77,"imp":11.664,"train_time":2149.06,"predict_time":0.318,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+158\/-112","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1642.0,"score":0.335,"rank":22.78,"hrank":16.49,"imp":10.113,"train_time":114.48,"predict_time":0.058,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+186\/-112","imputed":false},{"position":21,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1636.0,"score":0.346,"rank":23.11,"hrank":12.3,"imp":10.54,"train_time":1655.56,"predict_time":0.086,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+188\/-106","imputed":false},{"position":22,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1620.0,"score":0.331,"rank":24.11,"hrank":11.99,"imp":10.471,"train_time":607.67,"predict_time":0.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+231\/-130","imputed":false},{"position":23,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1602.0,"score":0.274,"rank":25.22,"hrank":21.26,"imp":11.808,"train_time":87.3,"predict_time":0.069,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+173\/-105","imputed":false},{"position":24,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1600.0,"score":0.326,"rank":25.33,"hrank":16.67,"imp":10.54,"train_time":0.48,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+191\/-142","imputed":false},{"position":25,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1579.0,"score":0.258,"rank":26.67,"hrank":16.02,"imp":10.311,"train_time":2.32,"predict_time":0.889,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+175\/-129","imputed":false},{"position":26,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1562.0,"score":0.258,"rank":27.78,"hrank":11.42,"imp":11.884,"train_time":9.96,"predict_time":4.847,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-137","imputed":false},{"position":27,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1527.0,"score":0.213,"rank":30.0,"hrank":17.22,"imp":12.466,"train_time":1878.64,"predict_time":5.934,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+193\/-112","imputed":false},{"position":28,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1510.0,"score":0.173,"rank":31.11,"hrank":24.56,"imp":12.201,"train_time":1128.19,"predict_time":1.674,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-79","imputed":false},{"position":29,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1503.0,"score":0.231,"rank":31.56,"hrank":15.91,"imp":12.578,"train_time":0.68,"predict_time":0.155,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+226\/-208","imputed":false},{"position":30,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1497.0,"score":0.255,"rank":32.0,"hrank":7.16,"imp":12.483,"train_time":0.48,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+175\/-194","imputed":false},{"position":31,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1483.0,"score":0.155,"rank":32.89,"hrank":29.17,"imp":12.872,"train_time":1.55,"predict_time":0.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-124","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1478.0,"score":0.232,"rank":33.22,"hrank":23.61,"imp":11.801,"train_time":559.89,"predict_time":1.156,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+161\/-159","imputed":false},{"position":33,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1468.0,"score":0.173,"rank":33.89,"hrank":29.63,"imp":15.118,"train_time":180.21,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+176\/-127","imputed":false},{"position":34,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1466.0,"score":0.164,"rank":34.0,"hrank":26.25,"imp":14.695,"train_time":559.89,"predict_time":0.056,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-136","imputed":false},{"position":35,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1465.0,"score":0.154,"rank":34.11,"hrank":28.25,"imp":12.591,"train_time":4560.09,"predict_time":421.951,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-130","imputed":false},{"position":36,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1461.0,"score":0.163,"rank":34.33,"hrank":23.17,"imp":14.292,"train_time":0.76,"predict_time":0.204,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+204\/-161","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1438.0,"score":0.092,"rank":35.89,"hrank":33.48,"imp":14.084,"train_time":6.55,"predict_time":0.295,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-78","imputed":false},{"position":38,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1402.0,"score":0.114,"rank":38.28,"hrank":23.51,"imp":16.054,"train_time":4560.09,"predict_time":39.452,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+191\/-177","imputed":false},{"position":39,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1399.0,"score":0.097,"rank":38.44,"hrank":32.05,"imp":14.678,"train_time":368.88,"predict_time":2.507,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-172","imputed":false},{"position":40,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1368.0,"score":0.112,"rank":40.44,"hrank":29.6,"imp":15.503,"train_time":92.9,"predict_time":0.786,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+181\/-176","imputed":false},{"position":41,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1349.0,"score":0.065,"rank":41.67,"hrank":38.09,"imp":17.837,"train_time":0.67,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+166\/-128","imputed":false},{"position":42,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1349.0,"score":0.076,"rank":41.67,"hrank":35.63,"imp":16.766,"train_time":180.21,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+197\/-144","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1339.0,"score":0.08,"rank":42.33,"hrank":38.32,"imp":15.57,"train_time":1128.19,"predict_time":0.055,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-126","imputed":false},{"position":44,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1328.0,"score":0.113,"rank":43.0,"hrank":36.47,"imp":19.25,"train_time":82.69,"predict_time":7.996,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-165","imputed":false},{"position":45,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1327.0,"score":0.015,"rank":43.11,"hrank":42.14,"imp":16.668,"train_time":43.93,"predict_time":0.204,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-74","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1305.0,"score":0.024,"rank":44.44,"hrank":41.96,"imp":16.14,"train_time":58.36,"predict_time":0.123,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-166","imputed":false},{"position":47,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1304.0,"score":0.06,"rank":44.56,"hrank":41.71,"imp":14.86,"train_time":0.34,"predict_time":0.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-141","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1282.0,"score":0.0,"rank":45.89,"hrank":45.4,"imp":17.091,"train_time":43.93,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-60","imputed":false},{"position":49,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1273.0,"score":0.034,"rank":46.44,"hrank":44.46,"imp":16.211,"train_time":9.93,"predict_time":0.224,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-131","imputed":false},{"position":50,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1271.0,"score":0.026,"rank":46.56,"hrank":42.28,"imp":18.665,"train_time":16.58,"predict_time":16.469,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-133","imputed":false},{"position":51,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1267.0,"score":0.124,"rank":46.78,"hrank":30.45,"imp":20.926,"train_time":5.13,"predict_time":2.705,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+218\/-419","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1241.0,"score":0.031,"rank":48.33,"hrank":43.17,"imp":18.59,"train_time":58.36,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+138\/-218","imputed":false},{"position":53,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1223.0,"score":0.023,"rank":49.39,"hrank":44.87,"imp":20.298,"train_time":41.8,"predict_time":39.131,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-246","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1209.0,"score":0.015,"rank":50.22,"hrank":46.93,"imp":18.638,"train_time":368.88,"predict_time":0.084,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-217","imputed":false},{"position":55,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1177.0,"score":0.013,"rank":52.0,"hrank":49.96,"imp":18.151,"train_time":1.81,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-130","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1116.0,"score":0.019,"rank":55.22,"hrank":50.63,"imp":21.386,"train_time":1.61,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-268","imputed":false},{"position":57,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1110.0,"score":0.017,"rank":55.56,"hrank":52.21,"imp":22.435,"train_time":514.21,"predict_time":52.341,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-203","imputed":false},{"position":58,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1106.0,"score":0.016,"rank":55.72,"hrank":50.33,"imp":26.623,"train_time":43.96,"predict_time":0.099,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-246","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1094.0,"score":0.0,"rank":56.33,"hrank":55.06,"imp":25.215,"train_time":3623.0,"predict_time":2.734,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-127","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1073.0,"score":0.0,"rank":57.33,"hrank":56.1,"imp":25.289,"train_time":3623.0,"predict_time":9.843,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-131","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1068.0,"score":0.002,"rank":57.61,"hrank":53.52,"imp":27.327,"train_time":43.96,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+170\/-261","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":60.67,"hrank":59.15,"imp":24.29,"train_time":0.11,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+134\/-288","imputed":false},{"position":63,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":987.0,"score":0.0,"rank":61.22,"hrank":59.95,"imp":24.224,"train_time":17.73,"predict_time":7.022,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+123\/-216","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":976.0,"score":0.0,"rank":61.67,"hrank":59.41,"imp":26.768,"train_time":20.27,"predict_time":0.572,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+149\/-210","imputed":false},{"position":65,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":934.0,"score":0.0,"rank":63.33,"hrank":62.83,"imp":25.19,"train_time":0.33,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-260","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":925.0,"score":0.0,"rank":63.67,"hrank":62.98,"imp":31.34,"train_time":99.35,"predict_time":2.573,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-135","imputed":false},{"position":67,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":922.0,"score":0.024,"rank":63.78,"hrank":54.89,"imp":32.339,"train_time":2.77,"predict_time":0.157,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+225\/-491","imputed":false},{"position":68,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":919.0,"score":0.0,"rank":63.89,"hrank":63.18,"imp":29.063,"train_time":0.22,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-222","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":906.0,"score":0.0,"rank":64.33,"hrank":62.69,"imp":27.991,"train_time":20.27,"predict_time":0.247,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-251","imputed":false},{"position":70,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":890.0,"score":0.0,"rank":64.89,"hrank":64.67,"imp":29.981,"train_time":514.21,"predict_time":2.489,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-136","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":853.0,"score":0.0,"rank":66.11,"hrank":65.39,"imp":34.192,"train_time":0.2,"predict_time":0.035,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-194","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":790.0,"score":0.0,"rank":67.89,"hrank":66.9,"imp":36.787,"train_time":1.32,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+174\/-312","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":541.0,"score":0.0,"rank":72.33,"hrank":72.3,"imp":47.009,"train_time":0.11,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-312","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/pareto_front_explorer.html
index d8811d094d34014605a5e161f3876d65094d6488..95d3818072fd6d32d3d01da80ed2453c7951feca 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.7858204126,"imp":15.5034551235,"elo":1368.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.019012743,"imp":8.5100101051,"elo":1683.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0144557546,"imp":9.3767023248,"elo":1696.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1540111592,"imp":9.1553395514,"elo":1699.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0245734202,"imp":10.5404571105,"elo":1600.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.025881505,"imp":10.0327959053,"elo":1690.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"imp":9.8448535884,"elo":1712.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0055651519,"imp":17.8374386075,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0063239748,"imp":16.7662499865,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0928598604,"imp":15.1179814359,"elo":1468.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0568493697,"imp":25.190452996,"elo":933.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0146798524,"imp":18.5903487894,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1233644034,"imp":16.1401114871,"elo":1305.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"imp":21.3859848317,"elo":1116.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0840284294,"imp":18.6378211225,"elo":1208.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.5074280623,"imp":14.6778588264,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0523242217,"imp":14.8604464106,"elo":1303.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0693419145,"imp":11.8077329112,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.533338997,"imp":10.8910789344,"elo":1673.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.7046252775,"imp":20.9257711492,"elo":1267.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"imp":16.2109312274,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3182658815,"imp":11.6636815593,"elo":1661.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3731434107,"imp":7.2130516693,"elo":1745.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":7.0220909982,"imp":24.2240911066,"elo":986.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0203313112,"imp":36.787229785,"elo":790.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2469408751,"imp":27.9905565021,"elo":906.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.571547743,"imp":26.7680731711,"elo":976.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0124785185,"imp":24.2904330857,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0180054122,"imp":17.0912787012,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2037872055,"imp":16.6677683402,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2945442984,"imp":14.0840458159,"elo":1437.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0857486963,"imp":10.5402822992,"elo":1636.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.4828000197,"imp":8.1328353033,"elo":1727.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8891223145,"imp":10.3108739427,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2634274244,"imp":10.3791191378,"elo":1671.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.8819383621,"imp":8.5465175306,"elo":1731.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.4685760401,"imp":18.6653800737,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.1313967486,"imp":20.2984613004,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"imp":16.0539258759,"elo":1401.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":421.9512729692,"imp":12.5907618928,"elo":1464.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2037859869,"imp":14.2920595104,"elo":1461.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":19.8203012705,"imp":5.6624539657,"elo":1971.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.034500585,"imp":34.1923792442,"elo":852.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":4.8471187111,"imp":11.8836633452,"elo":1561.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2005539263,"imp":8.9841380587,"elo":1806.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0532307625,"imp":12.8722729759,"elo":1483.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0566464663,"imp":10.4705308008,"elo":1619.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5563493456,"imp":9.4012773101,"elo":1716.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8822174168,"imp":8.4446221801,"elo":1857.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3092961409,"imp":6.163322271,"elo":1871.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5734866142,"imp":31.3402831495,"elo":924.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.7336975574,"imp":25.2149811049,"elo":1093.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.8426087184,"imp":25.2886840599,"elo":1073.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1545280266,"imp":12.5776482388,"elo":1503.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.048205775,"imp":18.1506174111,"elo":1176.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0547828197,"imp":15.5701709413,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6737046957,"imp":12.2009644998,"elo":1510.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"imp":12.4832418523,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0584702913,"imp":10.1133985191,"elo":1641.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4036472816,"imp":9.7776638033,"elo":1724.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.9962696925,"imp":19.2502851352,"elo":1328.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":5.9343927935,"imp":12.4657984196,"elo":1527.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":70.7413293281,"imp":10.2681129547,"elo":1724.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"imp":32.3394427831,"elo":921.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0563121316,"imp":14.6952543561,"elo":1466.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.156329922,"imp":11.8014817948,"elo":1478.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.7858204126,"x_train":92.9022516918,"imp":15.5034551235,"elo":1368.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.019012743,"x_train":1.4897650506,"imp":8.5100101051,"elo":1683.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0144557546,"x_train":339.6336294327,"imp":9.3767023248,"elo":1696.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1540111592,"x_train":339.6336294327,"imp":9.1553395514,"elo":1699.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0245734202,"x_train":0.4836414313,"imp":10.5404571105,"elo":1600.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.025881505,"x_train":92.4632546012,"imp":10.0327959053,"elo":1690.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"x_train":92.4632546012,"imp":9.8448535884,"elo":1712.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0055651519,"x_train":0.6682646338,"imp":17.8374386075,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0063239748,"x_train":180.2108414054,"imp":16.7662499865,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0928598604,"x_train":180.2108414054,"imp":15.1179814359,"elo":1468.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0568493697,"x_train":0.3303836838,"imp":25.190452996,"elo":933.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0146798524,"x_train":58.3611515474,"imp":18.5903487894,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1233644034,"x_train":58.3611515474,"imp":16.1401114871,"elo":1305.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"x_train":1.6147778992,"imp":21.3859848317,"elo":1116.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0840284294,"x_train":368.8801690096,"imp":18.6378211225,"elo":1208.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.5074280623,"x_train":368.8801690096,"imp":14.6778588264,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0523242217,"x_train":0.3380289469,"imp":14.8604464106,"elo":1303.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0693419145,"x_train":87.2996876502,"imp":11.8077329112,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.533338997,"x_train":87.2996876502,"imp":10.8910789344,"elo":1673.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.7046252775,"x_train":5.1314871311,"imp":20.9257711492,"elo":1267.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"x_train":9.925993635,"imp":16.2109312274,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3182658815,"x_train":2149.0612249684,"imp":11.6636815593,"elo":1661.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3731434107,"x_train":2149.0612249684,"imp":7.2130516693,"elo":1745.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":7.0220909982,"x_train":17.7296447292,"imp":24.2240911066,"elo":986.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0203313112,"x_train":1.3180150875,"imp":36.787229785,"elo":790.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2469408751,"x_train":20.2694783344,"imp":27.9905565021,"elo":906.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.571547743,"x_train":20.2694783344,"imp":26.7680731711,"elo":976.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0124785185,"x_train":0.112647328,"imp":24.2904330857,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0180054122,"x_train":43.9333129713,"imp":17.0912787012,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2037872055,"x_train":43.9333129713,"imp":16.6677683402,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2945442984,"x_train":6.549410715,"imp":14.0840458159,"elo":1437.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0857486963,"x_train":1655.5627332568,"imp":10.5402822992,"elo":1636.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.4828000197,"x_train":1655.5627332568,"imp":8.1328353033,"elo":1727.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8891223145,"x_train":2.3192637563,"imp":10.3108739427,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2634274244,"x_train":841.8438595298,"imp":10.3791191378,"elo":1671.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.8819383621,"x_train":841.8438595298,"imp":8.5465175306,"elo":1731.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.4685760401,"x_train":16.5759739993,"imp":18.6653800737,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.1313967486,"x_train":41.8036776423,"imp":20.2984613004,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"x_train":4560.0938951404,"imp":16.0539258759,"elo":1401.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":421.9512729692,"x_train":4560.0938951404,"imp":12.5907618928,"elo":1464.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2037859869,"x_train":0.755625227,"imp":14.2920595104,"elo":1461.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":19.8203012705,"x_train":42.2213895265,"imp":5.6624539657,"elo":1971.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.034500585,"x_train":0.1964259267,"imp":34.1923792442,"elo":852.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":4.8471187111,"x_train":9.9621880719,"imp":11.8836633452,"elo":1561.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2005539263,"x_train":0.6380837584,"imp":8.9841380587,"elo":1806.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0532307625,"x_train":1.5496904139,"imp":12.8722729759,"elo":1483.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0566464663,"x_train":607.6744130104,"imp":10.4705308008,"elo":1619.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5563493456,"x_train":607.6744130104,"imp":9.4012773101,"elo":1716.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8822174168,"x_train":2.6215608168,"imp":8.4446221801,"elo":1857.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3092961409,"x_train":1.1913903519,"imp":6.163322271,"elo":1871.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5734866142,"x_train":99.3538077736,"imp":31.3402831495,"elo":924.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.7336975574,"x_train":3623.0018158721,"imp":25.2149811049,"elo":1093.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.8426087184,"x_train":3623.0018158721,"imp":25.2886840599,"elo":1073.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1545280266,"x_train":0.6841801368,"imp":12.5776482388,"elo":1503.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.048205775,"x_train":1.8087875724,"imp":18.1506174111,"elo":1176.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0547828197,"x_train":1128.1946608067,"imp":15.5701709413,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6737046957,"x_train":1128.1946608067,"imp":12.2009644998,"elo":1510.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"x_train":0.4833722601,"imp":12.4832418523,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0584702913,"x_train":114.4782771397,"imp":10.1133985191,"elo":1641.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4036472816,"x_train":114.4782771397,"imp":9.7776638033,"elo":1724.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.9962696925,"x_train":82.6909536839,"imp":19.2502851352,"elo":1328.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":5.9343927935,"x_train":1878.6350381184,"imp":12.4657984196,"elo":1527.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":70.7413293281,"x_train":1878.6350381184,"imp":10.2681129547,"elo":1724.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"x_train":2.766166532,"imp":32.3394427831,"elo":921.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0563121316,"x_train":559.8943943667,"imp":14.6952543561,"elo":1466.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.156329922,"x_train":559.8943943667,"imp":11.8014817948,"elo":1478.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
index 2a2fd870ee2d1afe5cb7262ebefe6763e131d7ea..1138063f9c573ff685720edb83808923f1c4b5f8 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":92.9022516918,"imp":15.6304054648,"elo":1419.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.4897650506,"imp":8.6175174681,"elo":1748.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.6051687222,"imp":8.9090497492,"elo":1741.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.532113363,"imp":8.9584311887,"elo":1759.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":23.4136507344,"imp":9.0073795291,"elo":1766.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":48.9298256458,"imp":9.1405422262,"elo":1779.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":89.6438214128,"imp":9.1678098295,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":172.2716550113,"imp":9.2461851945,"elo":1769.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":253.7610304456,"imp":9.2949339601,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":339.6336294327,"imp":9.2466010383,"elo":1759.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4836414313,"imp":10.6647570108,"elo":1654.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.0174591425,"imp":10.2624047538,"elo":1731.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.3395204881,"imp":10.152082235,"elo":1748.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.5724929214,"imp":9.9439906289,"elo":1762.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.475979466,"imp":9.9069243724,"elo":1768.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.9675728072,"imp":9.7840294945,"elo":1773.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":46.0351331804,"imp":9.7880840369,"elo":1780.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":69.0484132695,"imp":9.8588736329,"elo":1782.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":92.4632546012,"imp":9.9546748098,"elo":1790.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":0.6682646338,"imp":17.9374937959,"elo":1382.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":1.7357034029,"imp":16.7485885801,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":4.534589395,"imp":16.1813561541,"elo":1470.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":8.6492410666,"imp":15.9262941062,"elo":1487.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":22.0622979635,"imp":15.634017391,"elo":1506.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":45.0093118346,"imp":15.5216476661,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":89.7036949426,"imp":15.3378053723,"elo":1517.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":134.7272460967,"imp":15.2818799778,"elo":1512.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":180.2108414054,"imp":15.2148659224,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3303836838,"imp":25.2720625045,"elo":939.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.5821560801,"imp":22.7342112793,"elo":1015.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.5274874744,"imp":18.7683910282,"elo":1165.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9159741389,"imp":17.0732380081,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.250384706,"imp":16.4973246499,"elo":1270.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.4926603773,"imp":16.0164339132,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":29.01329069,"imp":15.4801085211,"elo":1346.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.4906512987,"imp":15.4505921437,"elo":1348.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":58.3611515474,"imp":16.2265867491,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.6147778992,"imp":21.4605604621,"elo":1150.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.3105413306,"imp":19.5493514352,"elo":1238.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.7063156272,"imp":17.2285226049,"elo":1338.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":18.3992783327,"imp":16.3771081549,"elo":1374.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":44.132927448,"imp":15.4921196998,"elo":1405.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":89.5658730903,"imp":15.2294822358,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":182.7601700398,"imp":14.8000127545,"elo":1439.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":274.9982902211,"imp":14.7253165202,"elo":1448.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":368.8801690096,"imp":14.7707568952,"elo":1444.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.3380289469,"imp":14.9714279867,"elo":1354.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.8082590402,"imp":13.8770524654,"elo":1455.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":1.9749981399,"imp":12.6793074379,"elo":1579.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":4.0486763087,"imp":11.8511537137,"elo":1666.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":10.7215076455,"imp":11.3125487279,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":21.3864383794,"imp":11.1109997449,"elo":1733.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":42.7548465236,"imp":11.0041379428,"elo":1742.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":64.660287192,"imp":10.9794984841,"elo":1746.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":87.2996876502,"imp":11.0104346726,"elo":1738.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":5.1314871311,"imp":21.0307677001,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":9.925993635,"imp":16.3078410227,"elo":1299.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.0390242902,"imp":14.2660907298,"elo":1437.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.9509960123,"imp":11.8490302933,"elo":1614.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":113.4600835723,"imp":11.3846883486,"elo":1650.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":261.7792952647,"imp":10.929286301,"elo":1697.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":534.5397963686,"imp":10.0322866144,"elo":1736.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1054.256664045,"imp":9.1909133551,"elo":1774.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1575.3072324116,"imp":8.0985875663,"elo":1803.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2149.0612249684,"imp":7.359343456,"elo":1823.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":17.7296447292,"imp":23.650676519,"elo":1030.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.3180150875,"imp":36.7902034003,"elo":826.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.1703965898,"imp":33.4478287347,"elo":901.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":17.0551085935,"imp":27.259488321,"elo":990.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.2694783344,"imp":26.7981128702,"elo":1005.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.112647328,"imp":24.3796637993,"elo":1000.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3309222318,"imp":19.5589979146,"elo":1164.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.0383939984,"imp":17.8883505387,"elo":1249.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":2.1243719464,"imp":17.3396423226,"elo":1290.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":5.3583380815,"imp":16.924771508,"elo":1347.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":10.7016437794,"imp":16.7951307758,"elo":1352.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":21.6483655058,"imp":16.6996630136,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":32.660208208,"imp":16.7037001297,"elo":1355.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":43.9333129713,"imp":16.7551826166,"elo":1362.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.549410715,"imp":14.1635695209,"elo":1467.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.3225140655,"imp":12.0458058992,"elo":1599.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.3256521543,"imp":10.6591161217,"elo":1686.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":79.4731674816,"imp":9.7218016261,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":207.8738621521,"imp":9.0370268671,"elo":1755.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":419.1813968527,"imp":8.6634115511,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":824.3242193836,"imp":8.4126113257,"elo":1782.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1235.2478685981,"imp":8.288629776,"elo":1797.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1655.5627332568,"imp":8.2540581451,"elo":1805.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2.3192637563,"imp":10.4442783914,"elo":1616.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.6540988882,"imp":10.3587014855,"elo":1642.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":21.4768228985,"imp":10.1758730325,"elo":1664.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":39.3241398729,"imp":10.0612192928,"elo":1683.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":89.1338776753,"imp":9.8567254649,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":188.9339441617,"imp":9.8531462616,"elo":1734.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":396.8907515036,"imp":9.4465158291,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":604.5921677407,"imp":9.186088088,"elo":1776.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":841.8438595298,"imp":8.6686696416,"elo":1790.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":16.5759739993,"imp":18.752656523,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.8036776423,"imp":20.3740740053,"elo":1258.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":64.005785751,"imp":18.2998325203,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":119.9651969318,"imp":16.6080427324,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":214.9883934306,"imp":15.9206822461,"elo":1427.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":546.7601156085,"imp":14.5477934303,"elo":1476.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1120.3502125295,"imp":13.6911932241,"elo":1504.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2258.5773171118,"imp":13.2463437214,"elo":1523.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3411.9202161954,"imp":12.8657222174,"elo":1517.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4560.0938951404,"imp":12.6550523838,"elo":1520.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.755625227,"imp":14.3847318698,"elo":1511.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":42.2213895265,"imp":5.7817499917,"elo":2042.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1964259267,"imp":34.246563302,"elo":876.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":9.9621880719,"imp":12.0028644466,"elo":1617.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6380837584,"imp":9.1131354242,"elo":1863.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.5496904139,"imp":12.9775029644,"elo":1535.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":4.0437474833,"imp":12.4030184403,"elo":1571.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":13.0885329817,"imp":11.5965488724,"elo":1640.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":28.6386949065,"imp":10.8420153944,"elo":1689.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":73.9114278495,"imp":10.2149575839,"elo":1724.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":149.8736667915,"imp":9.9359304829,"elo":1742.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":297.7197032026,"imp":9.6522318264,"elo":1758.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":450.2716217096,"imp":9.5765885634,"elo":1774.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":607.6744130104,"imp":9.5315611642,"elo":1785.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.6215608168,"imp":8.5655234821,"elo":1931.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.1913903519,"imp":6.0659943447,"elo":1911.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":99.3538077736,"imp":31.3894385909,"elo":935.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":109.4145314705,"imp":31.2869146228,"elo":950.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":139.9023655107,"imp":30.974960341,"elo":958.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":205.6659012229,"imp":30.2636319035,"elo":984.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":461.8625479973,"imp":28.8724797652,"elo":1009.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":930.6476255381,"imp":27.4482185724,"elo":1052.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1843.2021584131,"imp":26.7907491687,"elo":1071.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2720.2054158222,"imp":26.0025523356,"elo":1100.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3623.0018158721,"imp":25.3453862524,"elo":1130.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.6841801368,"imp":12.7068829234,"elo":1553.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":1.8087875724,"imp":18.2232203292,"elo":1212.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.3361236019,"imp":16.3802668296,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":27.0568282205,"imp":14.6929540758,"elo":1382.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":59.5222844332,"imp":13.8903221845,"elo":1439.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":141.9573130138,"imp":13.096623762,"elo":1479.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":277.9811204407,"imp":12.6971443003,"elo":1511.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":543.4134565854,"imp":12.4788870399,"elo":1538.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":841.3480660167,"imp":12.3353095196,"elo":1556.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1128.1946608067,"imp":12.3024351581,"elo":1561.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4833722601,"imp":12.6011537279,"elo":1552.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":1.1643639044,"imp":11.4004661213,"elo":1662.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":3.1384428828,"imp":10.8599877747,"elo":1716.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":5.5373139842,"imp":10.434780576,"elo":1752.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":14.4678529679,"imp":10.0903272106,"elo":1771.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":28.6143879135,"imp":9.9494308173,"elo":1786.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":57.062520432,"imp":9.9085698836,"elo":1794.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":85.1163434587,"imp":9.8976928184,"elo":1796.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":114.4782771397,"imp":9.9314116209,"elo":1796.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":82.6909536839,"imp":19.3351619487,"elo":1384.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":155.5869584216,"imp":16.8717041741,"elo":1451.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":374.5843390199,"imp":13.8493671046,"elo":1561.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":733.4208142097,"imp":11.8980579202,"elo":1681.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1807.7783820047,"imp":10.1727090797,"elo":1798.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1878.6350381184,"imp":10.3583880688,"elo":1794.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.766166532,"imp":32.4029572942,"elo":951.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.6121672096,"imp":24.7720666516,"elo":1079.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":14.0207540085,"imp":16.7736184787,"elo":1314.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":27.8741073966,"imp":14.0163454027,"elo":1461.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":69.7840899876,"imp":13.2110038325,"elo":1508.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":138.964614784,"imp":12.7230863393,"elo":1520.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":278.3697153548,"imp":12.2774774892,"elo":1527.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":417.6594178725,"imp":11.9525494205,"elo":1529.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":559.8943943667,"imp":11.5190627788,"elo":1539.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.7858204126,"x_train":92.9022516918,"imp":15.6304054648,"elo":1419.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.019012743,"x_train":1.4897650506,"imp":8.6175174681,"elo":1748.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0371172646,"x_train":3.6051687222,"imp":8.9090497492,"elo":1741.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0610065729,"x_train":12.532113363,"imp":8.9584311887,"elo":1759.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0760820847,"x_train":23.4136507344,"imp":9.0073795291,"elo":1766.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0945525499,"x_train":48.9298256458,"imp":9.1405422262,"elo":1779.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1099725948,"x_train":89.6438214128,"imp":9.1678098295,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1316544584,"x_train":172.2716550113,"imp":9.2461851945,"elo":1769.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1470337609,"x_train":253.7610304456,"imp":9.2949339601,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1540111592,"x_train":339.6336294327,"imp":9.2466010383,"elo":1759.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0245734202,"x_train":0.4836414313,"imp":10.6647570108,"elo":1654.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0497602673,"x_train":1.0174591425,"imp":10.2624047538,"elo":1731.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1045743275,"x_train":2.3395204881,"imp":10.152082235,"elo":1748.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1712030423,"x_train":4.5724929214,"imp":9.9439906289,"elo":1762.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2592282785,"x_train":11.475979466,"imp":9.9069243724,"elo":1768.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3139296897,"x_train":22.9675728072,"imp":9.7840294945,"elo":1773.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3547635864,"x_train":46.0351331804,"imp":9.7880840369,"elo":1780.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3569505185,"x_train":69.0484132695,"imp":9.8588736329,"elo":1782.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"x_train":92.4632546012,"imp":9.9546748098,"elo":1790.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0055651519,"x_train":0.6682646338,"imp":17.9374937959,"elo":1382.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0116188579,"x_train":1.7357034029,"imp":16.7485885801,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0227492458,"x_train":4.534589395,"imp":16.1813561541,"elo":1470.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333486726,"x_train":8.6492410666,"imp":15.9262941062,"elo":1487.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.05377229,"x_train":22.0622979635,"imp":15.634017391,"elo":1506.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0655447367,"x_train":45.0093118346,"imp":15.5216476661,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0784040733,"x_train":89.7036949426,"imp":15.3378053723,"elo":1517.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0899934006,"x_train":134.7272460967,"imp":15.2818799778,"elo":1512.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0923119309,"x_train":180.2108414054,"imp":15.2148659224,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0568493697,"x_train":0.3303836838,"imp":25.2720625045,"elo":939.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0701976016,"x_train":0.5821560801,"imp":22.7342112793,"elo":1015.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.084787187,"x_train":1.5274874744,"imp":18.7683910282,"elo":1165.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0744434252,"x_train":2.9159741389,"imp":17.0732380081,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1006473264,"x_train":7.250384706,"imp":16.4973246499,"elo":1270.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1108145286,"x_train":14.4926603773,"imp":16.0164339132,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.12746224,"x_train":29.01329069,"imp":15.4801085211,"elo":1346.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1444622794,"x_train":43.4906512987,"imp":15.4505921437,"elo":1348.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1233644034,"x_train":58.3611515474,"imp":16.2265867491,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"x_train":1.6147778992,"imp":21.4605604621,"elo":1150.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.229817567,"x_train":3.3105413306,"imp":19.5493514352,"elo":1238.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5679873118,"x_train":8.7063156272,"imp":17.2285226049,"elo":1338.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9361192036,"x_train":18.3992783327,"imp":16.3771081549,"elo":1374.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2238362098,"x_train":44.132927448,"imp":15.4921196998,"elo":1405.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4486181927,"x_train":89.5658730903,"imp":15.2294822358,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.813496387,"x_train":182.7601700398,"imp":14.8000127545,"elo":1439.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.9956219304,"x_train":274.9982902211,"imp":14.7253165202,"elo":1448.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.3842646122,"x_train":368.8801690096,"imp":14.7707568952,"elo":1444.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0523242217,"x_train":0.3380289469,"imp":14.9714279867,"elo":1354.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.314019146,"x_train":0.8082590402,"imp":13.8770524654,"elo":1455.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.507868109,"x_train":1.9749981399,"imp":12.6793074379,"elo":1579.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6446859686,"x_train":4.0486763087,"imp":11.8511537137,"elo":1666.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7299947112,"x_train":10.7215076455,"imp":11.3125487279,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7313868749,"x_train":21.3864383794,"imp":11.1109997449,"elo":1733.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5834680373,"x_train":42.7548465236,"imp":11.0041379428,"elo":1742.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5831249824,"x_train":64.660287192,"imp":10.9794984841,"elo":1746.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.611337261,"x_train":87.2996876502,"imp":11.0104346726,"elo":1738.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":2.7046252775,"x_train":5.1314871311,"imp":21.0307677001,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"x_train":9.925993635,"imp":16.3078410227,"elo":1299.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5307544296,"x_train":25.0390242902,"imp":14.2660907298,"elo":1437.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.3312012868,"x_train":60.9509960123,"imp":11.8490302933,"elo":1614.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1424478891,"x_train":113.4600835723,"imp":11.3846883486,"elo":1650.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.1266315403,"x_train":261.7792952647,"imp":10.929286301,"elo":1697.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1746450553,"x_train":534.5397963686,"imp":10.0322866144,"elo":1736.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5719781161,"x_train":1054.256664045,"imp":9.1909133551,"elo":1774.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3837794763,"x_train":1575.3072324116,"imp":8.0985875663,"elo":1803.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3731434107,"x_train":2149.0612249684,"imp":7.359343456,"elo":1823.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":7.0220909982,"x_train":17.7296447292,"imp":23.650676519,"elo":1030.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0203313112,"x_train":1.3180150875,"imp":36.7902034003,"elo":826.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1531889724,"x_train":6.1703965898,"imp":33.4478287347,"elo":901.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4887140664,"x_train":17.0551085935,"imp":27.259488321,"elo":990.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.571547743,"x_train":20.2694783344,"imp":26.7981128702,"elo":1005.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0124785185,"x_train":0.112647328,"imp":24.3796637993,"elo":1000.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.038943275,"x_train":0.3309222318,"imp":19.5589979146,"elo":1164.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0835714888,"x_train":1.0383939984,"imp":17.8883505387,"elo":1249.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1355199317,"x_train":2.1243719464,"imp":17.3396423226,"elo":1290.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1452933374,"x_train":5.3583380815,"imp":16.924771508,"elo":1347.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1986293863,"x_train":10.7016437794,"imp":16.7951307758,"elo":1352.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2366835006,"x_train":21.6483655058,"imp":16.6996630136,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2524117136,"x_train":32.660208208,"imp":16.7037001297,"elo":1355.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2089988036,"x_train":43.9333129713,"imp":16.7551826166,"elo":1362.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2945442984,"x_train":6.549410715,"imp":14.1635695209,"elo":1467.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5181830946,"x_train":15.3225140655,"imp":12.0458058992,"elo":1599.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.8861217734,"x_train":39.3256521543,"imp":10.6591161217,"elo":1686.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1358686586,"x_train":79.4731674816,"imp":9.7218016261,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.3941285921,"x_train":207.8738621521,"imp":9.0370268671,"elo":1755.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.841674859,"x_train":419.1813968527,"imp":8.6634115511,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.304113458,"x_train":824.3242193836,"imp":8.4126113257,"elo":1782.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.6517056168,"x_train":1235.2478685981,"imp":8.288629776,"elo":1797.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.4828000197,"x_train":1655.5627332568,"imp":8.2540581451,"elo":1805.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8891223145,"x_train":2.3192637563,"imp":10.4442783914,"elo":1616.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7570224004,"x_train":6.6540988882,"imp":10.3587014855,"elo":1642.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.5554651954,"x_train":21.4768228985,"imp":10.1758730325,"elo":1664.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.8781041597,"x_train":39.3241398729,"imp":10.0612192928,"elo":1683.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8489456775,"x_train":89.1338776753,"imp":9.8567254649,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.2088546073,"x_train":188.9339441617,"imp":9.8531462616,"elo":1734.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.4189615616,"x_train":396.8907515036,"imp":9.4465158291,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.8211715812,"x_train":604.5921677407,"imp":9.186088088,"elo":1776.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.8819383621,"x_train":841.8438595298,"imp":8.6686696416,"elo":1790.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.4685760401,"x_train":16.5759739993,"imp":18.752656523,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.1313967486,"x_train":41.8036776423,"imp":20.3740740053,"elo":1258.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":56.7537152129,"x_train":64.005785751,"imp":18.2998325203,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":88.8320015192,"x_train":119.9651969318,"imp":16.6080427324,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":118.3350772121,"x_train":214.9883934306,"imp":15.9206822461,"elo":1427.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":192.322295595,"x_train":546.7601156085,"imp":14.5477934303,"elo":1476.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":288.6025785079,"x_train":1120.3502125295,"imp":13.6911932241,"elo":1504.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":387.0015785179,"x_train":2258.5773171118,"imp":13.2463437214,"elo":1523.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":401.5222034651,"x_train":3411.9202161954,"imp":12.8657222174,"elo":1517.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":415.8360217452,"x_train":4560.0938951404,"imp":12.6550523838,"elo":1520.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2037859869,"x_train":0.755625227,"imp":14.3847318698,"elo":1511.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":19.8203012705,"x_train":42.2213895265,"imp":5.7817499917,"elo":2042.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.034500585,"x_train":0.1964259267,"imp":34.246563302,"elo":876.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":4.8471187111,"x_train":9.9621880719,"imp":12.0028644466,"elo":1617.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2005539263,"x_train":0.6380837584,"imp":9.1131354242,"elo":1863.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0532307625,"x_train":1.5496904139,"imp":12.9775029644,"elo":1535.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1151376498,"x_train":4.0437474833,"imp":12.4030184403,"elo":1571.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2040125941,"x_train":13.0885329817,"imp":11.5965488724,"elo":1640.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2639855015,"x_train":28.6386949065,"imp":10.8420153944,"elo":1689.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3596014701,"x_train":73.9114278495,"imp":10.2149575839,"elo":1724.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4038726776,"x_train":149.8736667915,"imp":9.9359304829,"elo":1742.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4765747411,"x_train":297.7197032026,"imp":9.6522318264,"elo":1758.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.479443369,"x_train":450.2716217096,"imp":9.5765885634,"elo":1774.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.4521463095,"x_train":607.6744130104,"imp":9.5315611642,"elo":1785.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8822174168,"x_train":2.6215608168,"imp":8.5655234821,"elo":1931.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3092961409,"x_train":1.1913903519,"imp":6.0659943447,"elo":1911.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5734866142,"x_train":99.3538077736,"imp":31.3894385909,"elo":935.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5734924126,"x_train":109.4145314705,"imp":31.2869146228,"elo":950.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5646489644,"x_train":139.9023655107,"imp":30.974960341,"elo":958.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.8322499442,"x_train":205.6659012229,"imp":30.2636319035,"elo":984.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.7583190775,"x_train":461.8625479973,"imp":28.8724797652,"elo":1009.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2467243907,"x_train":930.6476255381,"imp":27.4482185724,"elo":1052.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7537273236,"x_train":1843.2021584131,"imp":26.7907491687,"elo":1071.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.0618917211,"x_train":2720.2054158222,"imp":26.0025523356,"elo":1100.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.8426087184,"x_train":3623.0018158721,"imp":25.3453862524,"elo":1130.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1545280266,"x_train":0.6841801368,"imp":12.7068829234,"elo":1553.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.048205775,"x_train":1.8087875724,"imp":18.2232203292,"elo":1212.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1068515144,"x_train":6.3361236019,"imp":16.3802668296,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2859445433,"x_train":27.0568282205,"imp":14.6929540758,"elo":1382.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4165762351,"x_train":59.5222844332,"imp":13.8903221845,"elo":1439.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5967947147,"x_train":141.9573130138,"imp":13.096623762,"elo":1479.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8317571104,"x_train":277.9811204407,"imp":12.6971443003,"elo":1511.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1610815918,"x_train":543.4134565854,"imp":12.4788870399,"elo":1538.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.325976181,"x_train":841.3480660167,"imp":12.3353095196,"elo":1556.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6737046957,"x_train":1128.1946608067,"imp":12.3024351581,"elo":1561.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"x_train":0.4833722601,"imp":12.6011537279,"elo":1552.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1130083333,"x_train":1.1643639044,"imp":11.4004661213,"elo":1662.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2216013349,"x_train":3.1384428828,"imp":10.8599877747,"elo":1716.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2660467618,"x_train":5.5373139842,"imp":10.434780576,"elo":1752.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3235442016,"x_train":14.4678529679,"imp":10.0903272106,"elo":1771.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3097698815,"x_train":28.6143879135,"imp":9.9494308173,"elo":1786.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3428691139,"x_train":57.062520432,"imp":9.9085698836,"elo":1794.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.384116264,"x_train":85.1163434587,"imp":9.8976928184,"elo":1796.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4036472816,"x_train":114.4782771397,"imp":9.9314116209,"elo":1796.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":7.9962696925,"x_train":82.6909536839,"imp":19.3351619487,"elo":1384.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":17.6685334779,"x_train":155.5869584216,"imp":16.8717041741,"elo":1451.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":28.6343734044,"x_train":374.5843390199,"imp":13.8493671046,"elo":1561.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.779029417,"x_train":733.4208142097,"imp":11.8980579202,"elo":1681.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":71.1344199188,"x_train":1807.7783820047,"imp":10.1727090797,"elo":1798.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":70.7413293281,"x_train":1878.6350381184,"imp":10.3583880688,"elo":1794.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"x_train":2.766166532,"imp":32.4029572942,"elo":951.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.2277922302,"x_train":5.6121672096,"imp":24.7720666516,"elo":1079.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.2497110384,"x_train":14.0207540085,"imp":16.7736184787,"elo":1314.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.3680809292,"x_train":27.8741073966,"imp":14.0163454027,"elo":1461.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.5836572817,"x_train":69.7840899876,"imp":13.2110038325,"elo":1508.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.7285042482,"x_train":138.964614784,"imp":12.7230863393,"elo":1520.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.8340835198,"x_train":278.3697153548,"imp":12.2774774892,"elo":1527.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.0359081604,"x_train":417.6594178725,"imp":11.9525494205,"elo":1529.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.1682693315,"x_train":559.8943943667,"imp":11.5190627788,"elo":1539.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..09749679d80eb85dd6d96d691b892082be889f1b
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_binary/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "AutoGluon 1.4 (best, 4h)", "TabFM (default)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "ModernNCA (tuned + ensembled)", "RealTabPFN-2.5 (tuned + ensembled)", "RealMLP (tuned + ensembled)", "iLTM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabM (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "ChimeraBoost (tuned)", "CatBoost (default)", "LightGBM (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "ModernNCA (tuned)", "XGBoost (tuned)", "RealMLP (tuned)", "TabM (tuned)", "LightGBM (tuned)", "ChimeraBoost (default)", "RealTabPFN-2.5 (default)", "TabICL (default)", "iLTM (tuned)", "TorchMLP (tuned + ensembled)", "TabSwift (default)", "XGBoost (default)", "TabM (default)", "xRFM (tuned + ensembled)", "EBM (tuned + ensembled)", "xRFM (tuned)", "TabDPT (tuned + ensembled)", "TabDPT-Turbo (default)", "RealMLP (default)", "TabDPT (tuned)", "FastaiMLP (tuned + ensembled)", "BetaTabPFN (default)", "EBM (default)", "EBM (tuned)", "TorchMLP (tuned)", "iLTM (default)", "RandomForest (tuned + ensembled)", "ExtraTrees (tuned + ensembled)", "LightGBM (default)", "RandomForest (tuned)", "ModernNCA (default)", "SAP-RPT-OSS (default)", "LimiX (default)", "ExtraTrees (tuned)", "TabDPT (default)", "FastaiMLP (tuned)", "TorchMLP (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "Linear (tuned + ensembled)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "Linear (tuned)", "RandomForest (default)", "OrionMSP (default)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "TabSTAR (default)", "xRFM (default)", "Linear (default)", "PerpetualBooster (tuned)", "KNN (tuned)", "TabFlex (default)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.7777777910232544, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5555555820465088, 0.4444444477558136, NaN, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5555555820465088, 0.4444444477558136, 0.2222222238779068, NaN, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.4444444477558136, NaN, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.4444444477558136, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 1.0, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.6666666865348816, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, NaN, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 1.0, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, NaN, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, NaN, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, NaN, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 1.0, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, NaN, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.6666666865348816, 0.3333333432674408, NaN, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.4444444477558136, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.5555555820465088, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, NaN, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.5555555820465088, 1.0, 0.5555555820465088, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, NaN, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.4444444477558136, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 1.0, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, NaN, 0.5555555820465088, 0.6666666865348816, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, NaN, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.5555555820465088, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.2222222238779068, 0.4444444477558136, 0.6111111044883728, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.8888888955116272, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 1.0, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.4444444477558136, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, NaN, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, NaN, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, NaN, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, NaN, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.5555555820465088, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, 0.5555555820465088, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, 0.5555555820465088, NaN, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 0.6666666865348816, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.2222222238779068, 0.1111111119389534, 0.5555555820465088, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.0, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 1.0, 0.8888888955116272, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, NaN, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, NaN, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, NaN, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, NaN, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.0, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, NaN, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, NaN, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.0, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, 0.5555555820465088, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.0, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.0, 0.0, 0.3333333432674408, 0.3333333432674408, 0.0, 0.2222222238779068, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 1.0, 1.0, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, NaN, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.4444444477558136, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.4444444477558136, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.0, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, NaN, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.0, 0.3333333432674408, 0.0, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.0, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, NaN, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, NaN, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.2222222238779068, 0.5555555820465088, NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1666666716337204, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.5555555820465088, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, NaN, 0.5555555820465088, 0.8888888955116272, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.0, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, NaN, 0.4444444477558136, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.5555555820465088, 0.2222222238779068, 0.1111111119389534, 0.5555555820465088, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.2222222238779068, 0.5555555820465088, NaN, 0.7777777910232544, 0.4444444477558136, 0.4444444477558136, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, NaN, 0.4444444477558136, 0.4444444477558136, 0.8333333134651184, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, NaN, 0.8888888955116272, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.1111111119389534, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.3333333432674408, 0.0, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.1666666716337204, 0.4444444477558136, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.4444444477558136, 0.7777777910232544, 0.4444444477558136, 0.8888888955116272, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.7777777910232544, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, NaN, 0.5555555820465088, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, 1.0, 0.5555555820465088, 0.5555555820465088, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, NaN, 0.3333333432674408, 0.7777777910232544, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.0, 0.0, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.7777777910232544, 0.6666666865348816, NaN, 0.6666666865348816, 0.4444444477558136, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, NaN, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.3333333432674408, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.0, 0.4444444477558136, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, NaN, 0.2222222238779068, 0.3333333432674408, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.0, 0.3333333432674408, 0.1111111119389534, 0.0, 0.2222222238779068, 0.0, 0.4444444477558136, 0.1111111119389534, 0.1111111119389534, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, 0.4444444477558136, 0.7777777910232544, NaN, 0.5555555820465088, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.3333333432674408, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.0, 0.0, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, NaN, 0.6666666865348816, 0.7777777910232544], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.0, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, NaN, 0.7777777910232544], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9573820413},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9421613425},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9086758041},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8614916359},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8538812847},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8234398865},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7823439972},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7716895065},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7686453654},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7671232971},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7671232964},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7610350156},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7579908758},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7480974204},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7458143155},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.741248105},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7351598241},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7275494755},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7260274061},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.717656021},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.7016742854},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6971080742},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6834094448},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6681887441},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.6666666746},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6484018351},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6331811346},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6027397347},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5875190335},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.581430757},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5753424738},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5631659135},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5585997035},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5494672823},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.547945213},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5464231438},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.5433790059},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5220700215},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4893455179},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4870624118},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.4596651527},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4429223824},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4429223817},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4337899595},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.4246575406},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4231354711},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4048706305},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4033485605},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3850837193},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3774733677},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3759512989},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.3729071621},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3515981787},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.3371385149},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3257229887},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3013698668},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2572298381},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2526636263},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2503805232},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2420091368},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2283105065},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2245053327},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.1826484048},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.1750380548},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1689497747},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1461187241},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1415525143},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1400304434},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1385083748},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.132420093},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1248097436},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1080669732},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0837138521},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0228310505}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..b99e981fc58a99bbc4a1cf8b80237ea40d95fc5f
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,"AutoGluon 1.5 (extreme, 4h)","AutoGluon 1.4 (best, 4h)",TabFM (default),TabPFN-3 (default),TabPFN-2.6 (default),TabICLv2 (default),ModernNCA (tuned + ensembled),RealTabPFN-2.5 (tuned + ensembled),RealMLP (tuned + ensembled),iLTM (tuned + ensembled),XGBoost (tuned + ensembled),TabM (tuned + ensembled),ChimeraBoost (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),ChimeraBoost (tuned),CatBoost (default),LightGBM (tuned + ensembled),RealTabPFN-2.5 (tuned),ModernNCA (tuned),XGBoost (tuned),RealMLP (tuned),TabM (tuned),LightGBM (tuned),ChimeraBoost (default),RealTabPFN-2.5 (default),TabICL (default),iLTM (tuned),TorchMLP (tuned + ensembled),TabSwift (default),XGBoost (default),TabM (default),xRFM (tuned + ensembled),EBM (tuned + ensembled),xRFM (tuned),TabDPT (tuned + ensembled),TabDPT-Turbo (default),RealMLP (default),TabDPT (tuned),FastaiMLP (tuned + ensembled),BetaTabPFN (default),EBM (default),EBM (tuned),TorchMLP (tuned),iLTM (default),RandomForest (tuned + ensembled),ExtraTrees (tuned + ensembled),LightGBM (default),RandomForest (tuned),ModernNCA (default),SAP-RPT-OSS (default),LimiX (default),ExtraTrees (tuned),TabDPT (default),FastaiMLP (tuned),TorchMLP (default),FastaiMLP (default),KNN (tuned + ensembled),Linear (tuned + ensembled),TabSTAR (tuned),TabSTAR (tuned + ensembled),Linear (tuned),RandomForest (default),OrionMSP (default),PerpetualBooster (tuned + ensembled),ExtraTrees (default),TabSTAR (default),xRFM (default),Linear (default),PerpetualBooster (tuned),KNN (tuned),TabFlex (default),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.5555556,0.44444445,0.44444445,0.6666667,0.6666667,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.44444445,,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.7777778,1.0,1.0,1.0,0.7777778,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,0.7777778,1.0,1.0,0.7777778,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.5555556,0.44444445,,0.7777778,0.8888889,0.7777778,0.7777778,0.8888889,0.7777778,0.6666667,0.7777778,1.0,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.8888889,1.0,0.7777778,0.6666667,0.7777778,1.0,1.0,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.5555556,0.44444445,0.22222222,,0.5555556,0.6666667,0.8888889,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.6666667,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.7777778,0.6666667,0.6666667,0.8888889,1.0,0.7777778,0.8888889,0.7777778,0.8888889,0.6666667,0.8888889,0.8888889,1.0,0.7777778,0.8888889,0.7777778,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.33333334,0.22222222,0.11111111,0.44444445,,0.6666667,0.7777778,0.6666667,0.7777778,0.5555556,0.44444445,0.7777778,0.7777778,0.6666667,0.6666667,0.7777778,0.6666667,0.8888889,0.7777778,0.6666667,0.6666667,0.6666667,0.8888889,1.0,0.6666667,0.8888889,0.8888889,0.7777778,0.7777778,0.7777778,0.8888889,1.0,0.6666667,1.0,0.7777778,1.0,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,,0.6666667,0.7777778,0.5555556,0.6666667,0.6666667,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.7777778,0.6666667,0.6666667,0.5555556,0.8888889,0.6666667,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.6666667,1.0,0.6666667,0.7777778,0.6666667,0.8888889,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,0.7777778,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.33333334,0.22222222,0.11111111,0.22222222,0.33333334,,0.6666667,0.5555556,0.5555556,0.6666667,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.6666667,0.7777778,0.6666667,0.6666667,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,0.22222222,0.11111111,0.33333334,0.33333334,0.22222222,0.33333334,,0.5555556,0.44444445,0.5555556,0.44444445,0.5555556,0.6666667,0.5555556,0.6666667,0.6666667,0.5555556,0.8888889,0.6666667,0.6666667,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.6666667,0.8888889,0.7777778,0.8888889,0.6666667,0.8888889,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,0.6666667,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.22222222,0.33333334,0.22222222,0.44444445,0.44444445,0.44444445,,0.5555556,0.33333334,0.44444445,0.33333334,0.5555556,0.5555556,0.44444445,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.5555556,0.7777778,0.5555556,0.7777778,0.8888889,0.6666667,0.7777778,0.8888889,0.7777778,0.7777778,0.8888889,0.7777778,0.7777778,1.0,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.5555556,0.44444445,,0.6666667,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.44444445,0.6666667,0.5555556,0.7777778,0.6666667,0.5555556,0.5555556,0.6666667,0.7777778,0.6666667,0.6666667,0.8888889,0.6666667,0.7777778,0.7777778,0.8888889,0.7777778,0.7777778,0.6666667,0.6666667,0.8888889,0.7777778,0.8888889,0.7777778,0.7777778,0.6666667,1.0,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.22222222,0.33333334,0.5555556,0.33333334,0.33333334,0.44444445,0.6666667,0.33333334,,0.6666667,0.44444445,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.44444445,0.7777778,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.8888889,0.5555556,0.8888889,0.8888889,0.6666667,0.8888889,0.7777778,0.7777778,0.6666667,0.8888889,0.6666667,0.7777778,0.7777778,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,0.6666667,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.11111111,0.22222222,0.0,0.22222222,0.22222222,0.44444445,0.44444445,0.5555556,0.5555556,0.44444445,0.33333334,,0.44444445,0.5555556,0.44444445,0.44444445,0.5555556,0.7777778,0.5555556,0.7777778,0.6666667,0.6666667,0.5555556,0.6666667,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,1.0,0.5555556,1.0,0.5555556,0.7777778,0.8888889,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.7777778,1.0,0.8888889,1.0,1.0,1.0,0.8888889,0.7777778,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.44444445,0.6666667,0.44444445,0.5555556,0.5555556,,0.44444445,0.44444445,0.6666667,0.5555556,0.5555556,0.6666667,0.5555556,0.5555556,0.7777778,0.5555556,0.5555556,0.7777778,0.6666667,0.7777778,0.7777778,0.7777778,0.44444445,0.7777778,0.7777778,0.8888889,0.8888889,1.0,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,0.7777778,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,0.8888889,1.0,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.44444445,0.5555556,,0.5555556,0.6666667,0.6666667,0.44444445,0.44444445,0.3888889,0.5555556,0.6666667,0.5555556,0.5555556,0.7777778,0.7777778,0.6666667,0.8888889,0.7777778,0.6666667,0.7777778,0.6666667,0.8888889,0.6666667,1.0,0.6666667,0.6666667,0.7777778,0.7777778,0.8888889,0.7777778,0.8888889,0.6666667,1.0,0.8888889,1.0,1.0,0.8888889,1.0,0.8888889,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.44444445,,0.5555556,0.6666667,0.33333334,0.44444445,0.44444445,0.6666667,0.6666667,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,0.7777778,0.8888889,0.6666667,0.6666667,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,0.6666667,1.0,0.8888889,1.0,0.8888889,0.8888889,1.0,0.8888889,1.0,0.7777778,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.5555556,0.44444445,0.44444445,0.5555556,0.33333334,0.33333334,0.44444445,,0.5555556,0.44444445,0.44444445,0.6666667,0.5555556,0.7777778,0.5555556,0.44444445,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.5555556,0.7777778,0.7777778,0.8888889,1.0,1.0,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,0.7777778,1.0,0.8888889,0.8888889,1.0,0.8888889,1.0,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.11111111,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.33333334,0.44444445,,0.44444445,0.33333334,0.44444445,0.44444445,0.6666667,0.5555556,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.6666667,0.7777778,0.6666667,0.8888889,0.7777778,0.8888889,0.6666667,0.6666667,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,0.7777778,1.0,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.44444445,0.44444445,0.44444445,0.5555556,0.33333334,0.22222222,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.7777778,0.5555556,0.5555556,0.5555556,0.7777778,0.6666667,0.5555556,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.5555556,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.7777778,1.0,1.0,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.11111111,0.0,0.22222222,0.22222222,0.11111111,0.33333334,0.11111111,0.44444445,0.33333334,0.33333334,0.44444445,0.33333334,0.5555556,0.5555556,0.5555556,0.6666667,0.5555556,,0.5555556,0.5555556,0.5555556,0.6666667,0.7777778,0.5555556,0.6666667,0.7777778,0.6666667,0.6666667,0.7777778,0.7777778,0.8888889,0.5555556,0.8888889,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.7777778,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,0.6666667,0.7777778,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.11111111,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.44444445,0.44444445,0.5555556,0.22222222,0.44444445,0.6111111,0.5555556,0.33333334,0.5555556,0.5555556,0.44444445,,0.5555556,0.5555556,0.33333334,0.5555556,0.5555556,0.6666667,0.6666667,0.5555556,0.7777778,0.7777778,0.5555556,0.6666667,0.5555556,0.6666667,0.7777778,0.7777778,0.7777778,0.7777778,1.0,0.7777778,0.8888889,0.8888889,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.44444445,0.22222222,0.22222222,0.33333334,0.44444445,0.44444445,0.33333334,0.44444445,0.5555556,0.5555556,0.44444445,0.44444445,,0.44444445,0.5555556,0.6666667,0.5555556,0.44444445,0.5555556,0.5555556,0.6666667,0.5555556,0.8888889,0.7777778,0.7777778,0.6666667,0.6666667,0.8888889,0.6666667,0.7777778,0.6666667,0.6666667,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,0.6666667,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.22222222,0.33333334,0.33333334,0.33333334,0.22222222,0.44444445,0.22222222,0.33333334,0.44444445,0.33333334,0.22222222,0.33333334,0.33333334,0.22222222,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,,0.44444445,0.5555556,0.44444445,0.6666667,0.5555556,0.7777778,0.8888889,0.5555556,0.7777778,0.6666667,0.8888889,0.6666667,0.6666667,0.7777778,0.5555556,1.0,0.7777778,0.8888889,0.7777778,0.7777778,0.7777778,1.0,0.7777778,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.11111111,0.22222222,0.0,0.11111111,0.11111111,0.44444445,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.33333334,0.6666667,0.44444445,0.5555556,,0.44444445,0.33333334,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.7777778,0.44444445,0.7777778,0.5555556,0.6666667,0.7777778,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,1.0,0.8888889,0.6666667,0.7777778,0.7777778,0.7777778,0.8888889,1.0,0.8888889,0.7777778,0.7777778,0.7777778,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.11111111,0.33333334,0.22222222,0.33333334,0.44444445,0.33333334,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,0.44444445,0.22222222,0.22222222,0.44444445,0.33333334,0.44444445,0.5555556,,0.5555556,0.5555556,0.44444445,0.5555556,0.5555556,0.33333334,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.8888889,0.7777778,1.0,1.0,1.0,0.8888889,0.7777778,0.8888889,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.11111111,0.22222222,0.33333334,0.33333334,0.22222222,0.33333334,0.44444445,0.33333334,0.33333334,0.44444445,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,0.5555556,0.6666667,0.44444445,,0.44444445,0.5555556,0.6666667,0.6666667,0.44444445,0.6666667,0.5555556,0.8888889,0.7777778,0.7777778,0.7777778,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,0.8888889,0.8888889,0.8888889,0.7777778,0.8888889,0.7777778,0.6666667,0.8888889,0.7777778,1.0,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+RealTabPFN-2.5 (default),0.0,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.22222222,0.44444445,0.33333334,0.33333334,0.5555556,0.33333334,0.44444445,0.44444445,0.5555556,,0.6666667,0.6666667,0.5555556,0.7777778,0.7777778,0.6666667,0.6666667,0.6666667,0.5555556,0.7777778,0.6666667,0.6666667,0.6666667,0.6666667,1.0,0.8888889,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,1.0,0.8888889,0.8888889,0.8888889,0.6666667,0.7777778,0.7777778,0.8888889,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0
+TabICL (default),0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.22222222,0.44444445,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.22222222,0.33333334,0.44444445,0.22222222,0.33333334,0.44444445,0.44444445,0.33333334,0.5555556,0.44444445,0.33333334,,0.44444445,0.6666667,0.44444445,0.6666667,0.5555556,0.5555556,0.5555556,0.6666667,0.8888889,0.7777778,0.6666667,0.8888889,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,0.7777778,0.7777778,1.0,1.0,0.7777778,0.7777778,0.7777778,0.8888889,0.7777778,0.8888889,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.11111111,0.0,0.11111111,0.11111111,0.22222222,0.22222222,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.44444445,0.44444445,0.22222222,0.44444445,0.44444445,0.33333334,0.33333334,0.5555556,,0.5555556,0.5555556,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,0.5555556,0.44444445,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.7777778,0.8888889,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.11111111,0.33333334,0.22222222,0.22222222,0.22222222,0.33333334,0.11111111,0.11111111,0.11111111,0.33333334,0.22222222,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.22222222,0.33333334,0.11111111,0.44444445,0.44444445,0.33333334,0.44444445,0.33333334,0.44444445,,0.5555556,0.5555556,0.33333334,0.5555556,0.5555556,0.5555556,0.33333334,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,0.5555556,0.6666667,0.8888889,0.7777778,0.8888889,0.8888889,0.5555556,0.8888889,0.7777778,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.11111111,0.0,0.22222222,0.11111111,0.22222222,0.22222222,0.33333334,0.11111111,0.33333334,0.33333334,0.44444445,0.22222222,0.5555556,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.22222222,0.22222222,0.44444445,0.44444445,0.33333334,0.6666667,0.5555556,0.22222222,0.5555556,0.44444445,0.44444445,,0.5555556,0.5555556,0.6666667,0.5555556,0.5555556,0.5555556,0.44444445,0.44444445,0.5555556,0.6666667,0.6666667,0.7777778,0.6666667,0.5555556,0.7777778,0.6666667,0.6666667,0.7777778,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.8888889,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.7777778,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.33333334,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.44444445,0.11111111,0.22222222,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.5555556,0.44444445,0.44444445,,0.44444445,0.6666667,0.5555556,0.6666667,0.5555556,0.5555556,0.44444445,0.6666667,0.6666667,0.5555556,0.6666667,0.6666667,0.5555556,0.8888889,0.8888889,0.8888889,0.5555556,0.8888889,0.7777778,0.8888889,0.7777778,0.8888889,0.7777778,0.6666667,0.8888889,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.11111111,0.33333334,0.22222222,0.33333334,0.22222222,0.22222222,0.44444445,0.33333334,0.44444445,0.44444445,0.6666667,0.44444445,0.5555556,,0.5555556,0.33333334,0.5555556,0.5555556,0.5555556,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.6666667,0.6666667,0.7777778,0.7777778,0.6666667,0.8888889,0.7777778,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.7777778,1.0,1.0,0.6666667,0.8888889,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.11111111,0.22222222,0.33333334,0.33333334,0.11111111,0.33333334,0.22222222,0.11111111,0.33333334,0.44444445,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.44444445,0.44444445,0.22222222,0.11111111,0.5555556,0.33333334,0.11111111,0.33333334,0.44444445,0.33333334,0.44444445,0.33333334,0.33333334,0.44444445,,0.5555556,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.7777778,0.7777778,0.5555556,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.11111111,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.0,0.11111111,0.33333334,0.22222222,0.0,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.33333334,0.22222222,0.22222222,0.22222222,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.6666667,0.44444445,,0.44444445,0.5555556,0.44444445,0.6666667,0.6666667,0.6666667,0.7777778,1.0,0.8888889,0.7777778,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.8888889,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+xRFM (tuned),0.0,0.0,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.11111111,0.22222222,0.22222222,0.44444445,0.0,0.0,0.11111111,0.0,0.11111111,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.44444445,0.22222222,0.22222222,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.33333334,0.44444445,0.5555556,0.5555556,,0.44444445,0.44444445,0.5555556,0.44444445,0.5555556,0.5555556,0.5555556,0.6666667,0.7777778,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.6666667,1.0,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+TabDPT (tuned + ensembled),0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.11111111,0.33333334,0.44444445,0.22222222,0.22222222,0.11111111,0.22222222,0.33333334,0.33333334,0.22222222,0.22222222,0.11111111,0.44444445,0.6666667,0.44444445,0.44444445,0.44444445,0.5555556,0.44444445,0.5555556,,0.6666667,0.6666667,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.6666667,0.7777778,0.6666667,0.8888889,0.6666667,0.6666667,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,0.7777778,0.7777778,0.7777778,0.7777778,1.0,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0
+TabDPT-Turbo (default),0.0,0.11111111,0.0,0.0,0.0,0.0,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.11111111,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.22222222,0.22222222,0.22222222,0.33333334,0.44444445,0.22222222,0.22222222,0.33333334,0.33333334,0.22222222,0.5555556,0.5555556,0.5555556,0.44444445,0.44444445,0.5555556,0.5555556,0.5555556,0.33333334,,0.5555556,0.6666667,0.5555556,0.5555556,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.8888889,0.6666667,0.8888889,0.8888889,0.7777778,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,1.0,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.22222222,0.0,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.44444445,0.5555556,0.5555556,0.33333334,0.44444445,0.33333334,0.44444445,0.33333334,0.44444445,,0.6666667,0.44444445,0.6666667,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.5555556,0.7777778,0.7777778,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.33333334,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.11111111,0.0,0.33333334,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.11111111,0.44444445,0.33333334,0.44444445,0.33333334,0.33333334,0.44444445,0.33333334,0.5555556,0.33333334,0.33333334,0.33333334,,0.33333334,0.33333334,0.44444445,0.5555556,0.7777778,0.6666667,0.44444445,0.5555556,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.7777778,0.8333333,0.7777778,0.7777778,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.8888889,0.7777778,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.0,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.22222222,0.44444445,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,0.6666667,,0.44444445,0.44444445,0.5555556,0.6666667,0.44444445,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.6666667,0.6666667,0.6666667,0.8888889,0.7777778,1.0,0.8888889,0.8888889,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.22222222,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.33333334,0.0,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.22222222,0.33333334,0.22222222,0.44444445,0.44444445,0.44444445,0.33333334,0.6666667,0.5555556,,0.44444445,0.5555556,0.5555556,0.5555556,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,0.7777778,0.6666667,0.5555556,0.5555556,0.7777778,0.7777778,0.7777778,0.8888889,0.6666667,0.8888889,0.8888889,0.6666667,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.33333334,0.44444445,0.22222222,0.33333334,0.22222222,0.33333334,0.0,0.44444445,0.33333334,0.44444445,0.33333334,0.5555556,0.5555556,0.5555556,,0.5555556,0.5555556,0.5555556,0.6666667,0.5555556,0.6666667,0.6666667,0.7777778,0.7777778,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.8888889,0.6666667,0.8888889,0.8888889,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.11111111,0.0,0.0,0.33333334,0.33333334,0.0,0.22222222,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.22222222,0.11111111,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,,0.44444445,0.5555556,0.5555556,0.5555556,0.5555556,0.6666667,0.5555556,0.8888889,0.6666667,0.6666667,0.5555556,0.7777778,0.7777778,0.7777778,0.8888889,0.6666667,1.0,1.0,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.22222222,0.11111111,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.33333334,0.33333334,0.11111111,0.44444445,0.44444445,0.33333334,0.44444445,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.44444445,0.44444445,0.5555556,,0.6666667,0.44444445,0.6666667,0.5555556,0.44444445,0.7777778,0.6666667,0.5555556,0.7777778,0.5555556,0.44444445,0.8888889,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.33333334,0.0,0.22222222,0.33333334,0.22222222,0.33333334,0.22222222,0.22222222,0.11111111,0.22222222,0.33333334,0.44444445,0.44444445,0.44444445,0.33333334,0.44444445,0.33333334,0.5555556,0.44444445,0.44444445,0.44444445,0.33333334,,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.7777778,0.8888889,0.8888889,0.7777778,0.7777778,0.8888889,0.7777778,0.8888889,1.0,0.8888889,0.8888889,0.7777778,0.8888889,1.0,0.7777778,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.5555556,0.33333334,0.5555556,0.33333334,0.44444445,0.5555556,0.6666667,,0.5555556,0.5555556,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.5555556,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,1.0,0.7777778,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.11111111,0.33333334,0.11111111,0.33333334,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.33333334,0.5555556,0.44444445,,0.44444445,0.5555556,0.5555556,0.44444445,0.6666667,0.6666667,0.5555556,0.5555556,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,1.0,1.0,0.7777778,1.0,0.7777778,0.8888889,1.0,0.7777778,0.8888889,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.0,0.33333334,0.0,0.11111111,0.22222222,0.44444445,0.22222222,0.44444445,0.11111111,0.44444445,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.44444445,0.33333334,0.33333334,0.33333334,0.44444445,0.44444445,0.5555556,0.44444445,0.5555556,,0.44444445,0.5555556,0.5555556,0.6666667,0.6666667,0.5555556,0.5555556,0.7777778,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.6666667,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.33333334,0.44444445,0.33333334,0.33333334,0.5555556,0.5555556,0.44444445,0.44444445,0.5555556,,0.5555556,0.6666667,0.6666667,0.5555556,0.44444445,0.6666667,0.7777778,0.8888889,0.8888889,0.6666667,0.8888889,0.8888889,0.6666667,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.11111111,0.22222222,0.11111111,0.33333334,0.11111111,0.44444445,0.22222222,0.44444445,0.22222222,0.44444445,0.22222222,0.44444445,0.22222222,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,,0.44444445,0.6666667,0.5555556,0.44444445,0.5555556,0.8888889,0.8888889,0.8888889,0.6666667,0.7777778,0.7777778,0.6666667,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.11111111,0.0,0.33333334,0.33333334,0.22222222,0.44444445,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.44444445,0.33333334,0.5555556,0.44444445,0.33333334,0.5555556,,0.6666667,0.7777778,0.5555556,0.6666667,0.5555556,0.8888889,1.0,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+LimiX (default),0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.11111111,0.11111111,0.33333334,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.11111111,0.44444445,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,,0.44444445,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.44444445,0.44444445,0.6666667,0.7777778,0.5555556,0.6666667,0.8888889,0.6666667,0.8888889,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889
+ExtraTrees (tuned),0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.22222222,0.0,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.44444445,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.33333334,0.44444445,0.44444445,0.22222222,0.5555556,,0.5555556,0.44444445,0.44444445,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.8888889,0.8888889,0.7777778,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.0,0.0,0.11111111,0.11111111,0.22222222,0.0,0.22222222,0.22222222,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.11111111,0.11111111,0.22222222,0.16666667,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.44444445,0.33333334,0.44444445,,0.5555556,0.6666667,0.7777778,0.44444445,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.7777778,0.7777778,0.8888889,0.6666667,0.8888889,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.22222222,0.22222222,0.22222222,0.5555556,0.33333334,0.22222222,0.44444445,0.44444445,0.33333334,0.44444445,0.33333334,0.44444445,0.5555556,0.44444445,,0.5555556,0.8888889,0.5555556,0.5555556,0.6666667,0.6666667,0.5555556,0.7777778,0.6666667,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.0,0.33333334,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.0,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.22222222,0.44444445,0.22222222,0.22222222,0.11111111,0.44444445,0.44444445,0.5555556,0.33333334,0.44444445,,0.44444445,0.7777778,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.11111111,0.33333334,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.11111111,0.11111111,0.11111111,0.33333334,0.5555556,0.22222222,0.11111111,0.5555556,,0.44444445,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,1.0
+KNN (tuned + ensembled),0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.0,0.22222222,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.33333334,0.44444445,0.5555556,0.44444445,0.22222222,0.5555556,,0.7777778,0.44444445,0.44444445,0.7777778,0.7777778,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.7777778,1.0,0.7777778,0.8888889,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.11111111,0.22222222,0.0,0.11111111,0.22222222,0.11111111,0.0,0.33333334,0.11111111,0.22222222,0.0,0.11111111,0.0,0.22222222,0.33333334,0.11111111,0.33333334,0.11111111,0.33333334,0.33333334,0.33333334,0.11111111,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.22222222,,0.44444445,0.44444445,0.8333333,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.7777778,0.7777778,0.5555556,0.5555556,0.6666667,0.6666667,0.8888889
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.22222222,0.0,0.0,0.11111111,0.0,0.11111111,0.22222222,0.22222222,0.0,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.5555556,0.44444445,0.33333334,0.33333334,0.33333334,0.44444445,0.5555556,0.5555556,,0.8888889,0.5555556,0.6666667,0.6666667,0.7777778,0.6666667,1.0,0.6666667,0.7777778,0.7777778,0.8888889,1.0,0.7777778,0.8888889
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.0,0.0,0.11111111,0.0,0.11111111,0.22222222,0.22222222,0.0,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.5555556,0.44444445,0.33333334,0.33333334,0.33333334,0.44444445,0.5555556,0.5555556,0.11111111,,0.5555556,0.6666667,0.6666667,0.7777778,0.6666667,1.0,0.6666667,0.7777778,0.7777778,0.8888889,1.0,0.7777778,0.8888889
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.22222222,0.0,0.0,0.22222222,0.11111111,0.33333334,0.0,0.11111111,0.0,0.22222222,0.33333334,0.0,0.33333334,0.11111111,0.33333334,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.22222222,0.44444445,0.22222222,0.16666667,0.44444445,0.44444445,,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.7777778,0.7777778,0.5555556,0.5555556,0.6666667,0.6666667,0.8888889
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.33333334,0.22222222,0.22222222,0.44444445,0.22222222,0.44444445,0.33333334,0.33333334,0.44444445,,0.5555556,0.44444445,0.7777778,0.44444445,0.8888889,0.5555556,0.5555556,0.44444445,0.5555556,0.7777778,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.22222222,0.44444445,0.11111111,0.33333334,0.33333334,0.22222222,0.33333334,0.44444445,0.44444445,0.33333334,0.33333334,0.44444445,0.44444445,,0.44444445,0.6666667,0.5555556,0.7777778,0.5555556,0.44444445,0.44444445,0.6666667,0.6666667,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.33333334,0.44444445,0.22222222,0.22222222,0.44444445,0.5555556,0.5555556,,0.5555556,0.22222222,0.5555556,0.44444445,1.0,0.5555556,0.5555556,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.44444445,0.22222222,0.22222222,0.33333334,0.22222222,0.44444445,0.33333334,0.33333334,0.44444445,0.22222222,0.33333334,0.44444445,,0.33333334,0.7777778,0.44444445,0.5555556,0.5555556,0.5555556,0.7777778,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.0,0.0,0.0,0.0,0.33333334,0.22222222,0.33333334,0.11111111,0.11111111,0.22222222,0.33333334,0.44444445,0.0,0.0,0.44444445,0.5555556,0.44444445,0.7777778,0.6666667,,0.6666667,0.44444445,0.7777778,0.6666667,0.6666667,0.7777778,0.7777778
+xRFM (default),0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.22222222,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.22222222,0.11111111,0.22222222,0.44444445,0.22222222,0.33333334,,0.22222222,0.44444445,0.33333334,0.44444445,0.5555556,0.6666667
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.22222222,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.33333334,0.0,0.22222222,0.11111111,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.44444445,0.44444445,0.5555556,0.5555556,0.5555556,0.7777778,,0.5555556,0.5555556,0.5555556,0.6666667,0.8888889
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.22222222,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.44444445,0.22222222,0.22222222,0.44444445,0.44444445,0.5555556,0.0,0.44444445,0.22222222,0.5555556,0.44444445,,0.22222222,0.33333334,0.8888889,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.0,0.0,0.0,0.33333334,0.0,0.33333334,0.11111111,0.0,0.22222222,0.0,0.44444445,0.11111111,0.11111111,0.44444445,0.5555556,0.5555556,0.44444445,0.44444445,0.33333334,0.6666667,0.44444445,0.7777778,,0.5555556,0.8888889,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.33333334,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.33333334,0.0,0.0,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.5555556,0.44444445,0.6666667,0.44444445,,0.6666667,0.7777778
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.22222222,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.0,0.22222222,0.22222222,0.44444445,0.33333334,0.11111111,0.11111111,0.33333334,,0.7777778
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.22222222,0.33333334,0.11111111,0.0,0.0,0.22222222,0.22222222,
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_small/leaderboard_overview_explorer.html
index 345dcfa9ac5f0b78e23682cb48cbb04f11fd5520..2c0043976043e7b284c12f7a89e55808d758c8a7 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Binary | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_small/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_binary/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..69c509c3cecce43d2d5c8eb12e78169438648f5b
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_binary/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1764.0,"score":0.828,"rank":5.71,"hrank":1.53,"imp":4.812,"train_time":37.34,"predict_time":5.423,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+250\/-155","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1605.0,"score":0.599,"rank":11.12,"hrank":5.74,"imp":11.957,"train_time":4.66,"predict_time":0.361,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-68","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1599.0,"score":0.538,"rank":11.4,"hrank":7.75,"imp":13.532,"train_time":5.54,"predict_time":0.55,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-55","imputed":false},{"position":3,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1553.0,"score":0.535,"rank":13.62,"hrank":7.37,"imp":12.684,"train_time":1.96,"predict_time":0.151,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-90","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1542.0,"score":0.527,"rank":14.24,"hrank":7.14,"imp":13.276,"train_time":5.75,"predict_time":0.434,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-84","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1532.0,"score":0.497,"rank":14.79,"hrank":6.82,"imp":13.843,"train_time":2125.17,"predict_time":9.785,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-80","imputed":false},{"position":6,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1523.0,"score":0.531,"rank":15.29,"hrank":5.61,"imp":9.706,"train_time":369.86,"predict_time":6.202,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+147\/-101","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1508.0,"score":0.493,"rank":16.14,"hrank":7.95,"imp":14.124,"train_time":2125.17,"predict_time":0.881,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-95","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1457.0,"score":0.376,"rank":19.33,"hrank":12.77,"imp":15.965,"train_time":2791.97,"predict_time":23.933,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-82","imputed":false},{"position":9,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1456.0,"score":0.424,"rank":19.38,"hrank":6.21,"imp":14.94,"train_time":19.33,"predict_time":4.519,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-108","imputed":false},{"position":10,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1450.0,"score":0.365,"rank":19.76,"hrank":14.72,"imp":15.526,"train_time":1886.83,"predict_time":1.333,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-78","imputed":false},{"position":11,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1409.0,"score":0.36,"rank":22.64,"hrank":9.28,"imp":16.132,"train_time":2.4,"predict_time":0.116,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-82","imputed":false},{"position":12,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1387.0,"score":0.334,"rank":24.26,"hrank":11.67,"imp":17.328,"train_time":2791.97,"predict_time":0.543,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-77","imputed":false},{"position":13,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1370.0,"score":0.315,"rank":25.57,"hrank":9.17,"imp":16.174,"train_time":6619.83,"predict_time":408.772,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-83","imputed":false},{"position":14,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1352.0,"score":0.27,"rank":27.0,"hrank":17.29,"imp":19.288,"train_time":298.47,"predict_time":2.036,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-89","imputed":false},{"position":15,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1351.0,"score":0.298,"rank":27.05,"hrank":10.42,"imp":18.023,"train_time":2289.51,"predict_time":2.042,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-89","imputed":false},{"position":16,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1348.0,"score":0.272,"rank":27.26,"hrank":14.16,"imp":17.623,"train_time":15318.17,"predict_time":500.783,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-78","imputed":false},{"position":17,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1342.0,"score":0.257,"rank":27.74,"hrank":17.07,"imp":18.076,"train_time":1.21,"predict_time":0.052,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-78","imputed":false},{"position":18,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1336.0,"score":0.278,"rank":28.24,"hrank":14.93,"imp":17.325,"train_time":6619.83,"predict_time":39.71,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-84","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1321.0,"score":0.189,"rank":29.48,"hrank":23.54,"imp":18.67,"train_time":512.21,"predict_time":1.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-54","imputed":false},{"position":20,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1318.0,"score":0.256,"rank":29.74,"hrank":16.47,"imp":18.988,"train_time":2289.51,"predict_time":0.259,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-100","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1315.0,"score":0.218,"rank":29.95,"hrank":14.67,"imp":18.473,"train_time":1228.69,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-81","imputed":false},{"position":22,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1306.0,"score":0.178,"rank":30.71,"hrank":25.47,"imp":19.078,"train_time":3.78,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-58","imputed":false},{"position":23,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1304.0,"score":0.238,"rank":30.83,"hrank":15.14,"imp":18.541,"train_time":4618.5,"predict_time":0.426,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+141\/-88","imputed":false},{"position":24,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1301.0,"score":0.187,"rank":31.1,"hrank":24.26,"imp":18.931,"train_time":1228.69,"predict_time":0.333,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-75","imputed":false},{"position":25,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1296.0,"score":0.203,"rank":31.5,"hrank":23.07,"imp":20.068,"train_time":571.47,"predict_time":0.264,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-79","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1288.0,"score":0.191,"rank":32.17,"hrank":21.46,"imp":19.854,"train_time":700.96,"predict_time":0.743,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-73","imputed":false},{"position":27,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1280.0,"score":0.137,"rank":32.81,"hrank":21.19,"imp":19.774,"train_time":1364.2,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-78","imputed":false},{"position":28,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1278.0,"score":0.178,"rank":33.05,"hrank":24.0,"imp":19.208,"train_time":15318.17,"predict_time":75.138,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-72","imputed":false},{"position":29,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1277.0,"score":0.155,"rank":33.1,"hrank":21.82,"imp":19.445,"train_time":11.09,"predict_time":2.46,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-62","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1263.0,"score":0.17,"rank":34.24,"hrank":12.29,"imp":20.178,"train_time":700.96,"predict_time":0.203,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-76","imputed":false},{"position":31,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1261.0,"score":0.186,"rank":34.4,"hrank":20.22,"imp":20.173,"train_time":6.8,"predict_time":0.262,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-96","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1260.0,"score":0.183,"rank":34.48,"hrank":25.02,"imp":19.741,"train_time":536.34,"predict_time":1.738,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-83","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1259.0,"score":0.166,"rank":34.62,"hrank":18.0,"imp":20.893,"train_time":571.47,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-82","imputed":false},{"position":34,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1258.0,"score":0.151,"rank":34.71,"hrank":29.6,"imp":19.889,"train_time":512.21,"predict_time":0.122,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-66","imputed":false},{"position":35,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1255.0,"score":0.196,"rank":34.93,"hrank":22.2,"imp":20.673,"train_time":4618.5,"predict_time":9.218,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-91","imputed":false},{"position":36,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1254.0,"score":0.217,"rank":35.05,"hrank":19.54,"imp":21.888,"train_time":8.99,"predict_time":1.724,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-110","imputed":false},{"position":37,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1243.0,"score":0.142,"rank":35.93,"hrank":12.65,"imp":20.637,"train_time":6.11,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-81","imputed":false},{"position":38,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1241.0,"score":0.094,"rank":36.14,"hrank":24.71,"imp":20.25,"train_time":1364.2,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-66","imputed":false},{"position":39,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1237.0,"score":0.184,"rank":36.43,"hrank":20.99,"imp":19.704,"train_time":52.04,"predict_time":39.81,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-82","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1226.0,"score":0.146,"rank":37.38,"hrank":28.3,"imp":20.266,"train_time":2535.73,"predict_time":2.546,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-82","imputed":false},{"position":41,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1217.0,"score":0.105,"rank":38.14,"hrank":32.62,"imp":21.795,"train_time":3.24,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-67","imputed":false},{"position":42,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1214.0,"score":0.135,"rank":38.43,"hrank":29.62,"imp":20.74,"train_time":536.34,"predict_time":0.097,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-88","imputed":false},{"position":43,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1197.0,"score":0.132,"rank":39.86,"hrank":11.55,"imp":23.047,"train_time":13.4,"predict_time":0.329,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-103","imputed":false},{"position":44,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1178.0,"score":0.127,"rank":41.45,"hrank":32.45,"imp":22.133,"train_time":2535.73,"predict_time":0.112,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-87","imputed":false},{"position":45,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1175.0,"score":0.118,"rank":41.64,"hrank":30.36,"imp":22.832,"train_time":650.76,"predict_time":5.616,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-105","imputed":false},{"position":46,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1164.0,"score":0.207,"rank":42.6,"hrank":11.86,"imp":23.657,"train_time":44524.42,"predict_time":4.557,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-114","imputed":false},{"position":47,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1155.0,"score":0.196,"rank":43.29,"hrank":13.18,"imp":23.897,"train_time":44524.42,"predict_time":20.33,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-124","imputed":false},{"position":48,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1146.0,"score":0.135,"rank":44.05,"hrank":32.58,"imp":24.609,"train_time":196.18,"predict_time":0.35,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-106","imputed":false},{"position":49,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1136.0,"score":0.081,"rank":44.88,"hrank":36.52,"imp":23.37,"train_time":349.07,"predict_time":0.941,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-92","imputed":false},{"position":50,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1133.0,"score":0.106,"rank":45.1,"hrank":27.46,"imp":23.995,"train_time":650.76,"predict_time":0.441,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-141","imputed":false},{"position":51,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1131.0,"score":0.095,"rank":45.26,"hrank":20.97,"imp":24.278,"train_time":335.8,"predict_time":85.392,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-118","imputed":false},{"position":52,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1127.0,"score":0.076,"rank":45.57,"hrank":37.21,"imp":23.331,"train_time":1.98,"predict_time":0.122,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-77","imputed":false},{"position":53,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1125.0,"score":0.072,"rank":45.71,"hrank":34.54,"imp":24.073,"train_time":349.07,"predict_time":0.091,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-98","imputed":false},{"position":54,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1121.0,"score":0.072,"rank":46.1,"hrank":37.78,"imp":23.771,"train_time":13.99,"predict_time":2.578,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-77","imputed":false},{"position":55,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1103.0,"score":0.068,"rank":47.5,"hrank":39.9,"imp":23.468,"train_time":2.47,"predict_time":0.168,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-90","imputed":false},{"position":56,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1096.0,"score":0.075,"rank":48.0,"hrank":42.63,"imp":25.44,"train_time":196.18,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-91","imputed":false},{"position":57,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1092.0,"score":0.065,"rank":48.38,"hrank":40.65,"imp":25.154,"train_time":494.7,"predict_time":0.936,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-96","imputed":false},{"position":58,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1091.0,"score":0.159,"rank":48.43,"hrank":11.21,"imp":28.419,"train_time":439.9,"predict_time":5.445,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-148","imputed":false},{"position":59,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1061.0,"score":0.018,"rank":50.69,"hrank":45.95,"imp":25.794,"train_time":494.7,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-102","imputed":false},{"position":60,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1047.0,"score":0.076,"rank":51.71,"hrank":40.46,"imp":27.468,"train_time":1.29,"predict_time":0.149,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-146","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1024.0,"score":0.058,"rank":53.33,"hrank":23.57,"imp":29.871,"train_time":27.03,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-134","imputed":false},{"position":62,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1015.0,"score":0.058,"rank":53.95,"hrank":41.9,"imp":27.776,"train_time":1.98,"predict_time":0.374,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+121\/-152","imputed":false},{"position":63,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1013.0,"score":0.032,"rank":54.05,"hrank":46.6,"imp":27.228,"train_time":7.76,"predict_time":0.146,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-120","imputed":false},{"position":64,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.029,"rank":54.93,"hrank":40.83,"imp":28.47,"train_time":0.39,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-147","imputed":false},{"position":65,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":993.0,"score":0.033,"rank":55.38,"hrank":51.0,"imp":26.641,"train_time":3.48,"predict_time":0.427,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-112","imputed":false},{"position":66,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":950.0,"score":0.002,"rank":58.02,"hrank":56.26,"imp":28.911,"train_time":89.64,"predict_time":1.627,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-109","imputed":false},{"position":67,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":947.0,"score":0.037,"rank":58.19,"hrank":53.14,"imp":29.56,"train_time":2.66,"predict_time":0.702,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+94\/-126","imputed":false},{"position":68,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":938.0,"score":0.048,"rank":58.69,"hrank":15.78,"imp":32.155,"train_time":280.42,"predict_time":0.382,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-150","imputed":false},{"position":69,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":917.0,"score":0.04,"rank":59.86,"hrank":25.49,"imp":32.72,"train_time":280.42,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-149","imputed":false},{"position":70,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":910.0,"score":0.015,"rank":60.24,"hrank":54.39,"imp":34.375,"train_time":1.48,"predict_time":0.127,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-162","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":816.0,"score":0.0,"rank":64.48,"hrank":63.56,"imp":34.243,"train_time":89.64,"predict_time":0.08,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-133","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":628.0,"score":0.0,"rank":69.81,"hrank":69.36,"imp":45.146,"train_time":0.24,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-168","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_small/pareto_front_explorer.html
index 76b13942806c416615fb9066e7177f54b7141951..7def230510cc2d7d17a69d7db288e668117fa72d 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0361740111,"imp":19.2880440409,"elo":1351.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"imp":19.0777558969,"elo":1305.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0362429023,"imp":18.4734558412,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3334156391,"imp":18.9314006441,"elo":1301.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0364337543,"imp":21.7945510396,"elo":1217.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240617752,"imp":20.8929502016,"elo":1258.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2636490914,"imp":20.0682855865,"elo":1296.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.016650456,"imp":20.6368056206,"elo":1243.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0169525138,"imp":20.2502085135,"elo":1240.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1473350525,"imp":19.774104355,"elo":1280.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3735107819,"imp":27.776177429,"elo":1014.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0910270033,"imp":24.073400672,"elo":1125.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9410076245,"imp":23.3699336271,"elo":1135.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4267781724,"imp":26.6407605908,"elo":993.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4413511276,"imp":23.9951116434,"elo":1133.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6157797575,"imp":22.8320051455,"elo":1175.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1682254962,"imp":23.468475884,"elo":1102.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1220181584,"imp":19.8894048958,"elo":1257.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1696974604,"imp":18.6701362503,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.5187065444,"imp":14.9400451048,"elo":1456.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.329248032,"imp":23.0468362634,"elo":1196.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4255186878,"imp":18.5409371569,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2181811333,"imp":20.6725062284,"elo":1255.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5777056813,"imp":23.771186679,"elo":1120.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0303599261,"imp":29.8712424779,"elo":1023.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1211754978,"imp":25.4401971316,"elo":1096.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3495945281,"imp":24.6090314235,"elo":1146.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0405290578,"imp":28.4704345605,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0927957297,"imp":25.7943230172,"elo":1060.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.9355489781,"imp":25.1536703555,"elo":1091.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":2.4598647569,"imp":19.4447897286,"elo":1277.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5434324627,"imp":17.3278427509,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":23.9328857361,"imp":15.9652113077,"elo":1456.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5499649048,"imp":13.5316613803,"elo":1598.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8814794063,"imp":14.1237449179,"elo":1507.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7853653133,"imp":13.8433745811,"elo":1531.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7241354212,"imp":21.8883918544,"elo":1253.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8096498633,"imp":19.7044218468,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.7095967461,"imp":17.3252626942,"elo":1336.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":408.7718504964,"imp":16.1740027794,"elo":1370.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1160046101,"imp":16.1321041743,"elo":1409.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.4226160049,"imp":4.8117462184,"elo":1763.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1485826053,"imp":27.4680947758,"elo":1046.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1509533846,"imp":12.6839014591,"elo":1553.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2622436544,"imp":20.1731154013,"elo":1261.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.258500598,"imp":18.9882332269,"elo":1317.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.0417782615,"imp":18.0227489449,"elo":1351.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4340112036,"imp":13.2763749094,"elo":1541.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3608180836,"imp":11.9566984831,"elo":1605.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.4449928547,"imp":28.4191243617,"elo":1090.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5569457535,"imp":23.6569512712,"elo":1163.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.3301528826,"imp":23.8965214331,"elo":1155.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0524064596,"imp":18.0760645452,"elo":1342.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1464537212,"imp":27.2282657664,"elo":1013.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1115292311,"imp":22.1332284708,"elo":1177.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.545560207,"imp":20.2664776489,"elo":1226.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"imp":23.3306940393,"elo":1127.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2027324806,"imp":20.1781377471,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.7428396647,"imp":19.8539735083,"elo":1288.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":85.3917069912,"imp":24.2778486183,"elo":1131.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":75.1381347608,"imp":19.208239087,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":500.7827657461,"imp":17.6227468063,"elo":1348.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.701764822,"imp":29.5600167603,"elo":947.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0969305634,"imp":20.740422312,"elo":1213.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.7378187563,"imp":19.740505476,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0361740111,"x_train":298.4715930877,"imp":19.2880440409,"elo":1351.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"x_train":3.776461578,"imp":19.0777558969,"elo":1305.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0362429023,"x_train":1228.6932168834,"imp":18.4734558412,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3334156391,"x_train":1228.6932168834,"imp":18.9314006441,"elo":1301.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0364337543,"x_train":3.2394562483,"imp":21.7945510396,"elo":1217.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240617752,"x_train":571.4736016874,"imp":20.8929502016,"elo":1258.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2636490914,"x_train":571.4736016874,"imp":20.0682855865,"elo":1296.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.016650456,"x_train":6.1072816838,"imp":20.6368056206,"elo":1243.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0169525138,"x_train":1364.2004288288,"imp":20.2502085135,"elo":1240.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1473350525,"x_train":1364.2004288288,"imp":19.774104355,"elo":1280.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3735107819,"x_train":1.983283782,"imp":27.776177429,"elo":1014.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0910270033,"x_train":349.0651424646,"imp":24.073400672,"elo":1125.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9410076245,"x_train":349.0651424646,"imp":23.3699336271,"elo":1135.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4267781724,"x_train":3.4776029853,"imp":26.6407605908,"elo":993.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4413511276,"x_train":650.7592051506,"imp":23.9951116434,"elo":1133.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6157797575,"x_train":650.7592051506,"imp":22.8320051455,"elo":1175.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1682254962,"x_train":2.4736320925,"imp":23.468475884,"elo":1102.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1220181584,"x_train":512.2070722628,"imp":19.8894048958,"elo":1257.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1696974604,"x_train":512.2070722628,"imp":18.6701362503,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.5187065444,"x_train":19.3292393435,"imp":14.9400451048,"elo":1456.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.329248032,"x_train":13.4010979517,"imp":23.0468362634,"elo":1196.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4255186878,"x_train":4618.5023417328,"imp":18.5409371569,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2181811333,"x_train":4618.5023417328,"imp":20.6725062284,"elo":1255.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5777056813,"x_train":13.9928451766,"imp":23.771186679,"elo":1120.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0303599261,"x_train":27.0284046233,"imp":29.8712424779,"elo":1023.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1211754978,"x_train":196.1786690245,"imp":25.4401971316,"elo":1096.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3495945281,"x_train":196.1786690245,"imp":24.6090314235,"elo":1146.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0405290578,"x_train":0.3945714661,"imp":28.4704345605,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0927957297,"x_train":494.6995565402,"imp":25.7943230172,"elo":1060.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.9355489781,"x_train":494.6995565402,"imp":25.1536703555,"elo":1091.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":2.4598647569,"x_train":11.0934085905,"imp":19.4447897286,"elo":1277.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5434324627,"x_train":2791.9683399256,"imp":17.3278427509,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":23.9328857361,"x_train":2791.9683399256,"imp":15.9652113077,"elo":1456.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5499649048,"x_train":5.5361255471,"imp":13.5316613803,"elo":1598.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8814794063,"x_train":2125.1735517383,"imp":14.1237449179,"elo":1507.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7853653133,"x_train":2125.1735517383,"imp":13.8433745811,"elo":1531.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7241354212,"x_train":8.9906883617,"imp":21.8883918544,"elo":1253.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8096498633,"x_train":52.0419884348,"imp":19.7044218468,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.7095967461,"x_train":6619.8287889361,"imp":17.3252626942,"elo":1336.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":408.7718504964,"x_train":6619.8287889361,"imp":16.1740027794,"elo":1370.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1160046101,"x_train":2.397422576,"imp":16.1321041743,"elo":1409.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.4226160049,"x_train":37.343177154,"imp":4.8117462184,"elo":1763.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1485826053,"x_train":1.2915962934,"imp":27.4680947758,"elo":1046.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1509533846,"x_train":1.9632436902,"imp":12.6839014591,"elo":1553.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2622436544,"x_train":6.8040758648,"imp":20.1731154013,"elo":1261.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.258500598,"x_train":2289.5124290727,"imp":18.9882332269,"elo":1317.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.0417782615,"x_train":2289.5124290727,"imp":18.0227489449,"elo":1351.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4340112036,"x_train":5.7520413762,"imp":13.2763749094,"elo":1541.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3608180836,"x_train":4.6649008838,"imp":11.9566984831,"elo":1605.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.4449928547,"x_train":439.9020625591,"imp":28.4191243617,"elo":1090.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5569457535,"x_train":44524.4248051345,"imp":23.6569512712,"elo":1163.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.3301528826,"x_train":44524.4248051345,"imp":23.8965214331,"elo":1155.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0524064596,"x_train":1.2131979048,"imp":18.0760645452,"elo":1342.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1464537212,"x_train":7.7572571145,"imp":27.2282657664,"elo":1013.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1115292311,"x_train":2535.7310668162,"imp":22.1332284708,"elo":1177.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.545560207,"x_train":2535.7310668162,"imp":20.2664776489,"elo":1226.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"x_train":1.9818637578,"imp":23.3306940393,"elo":1127.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2027324806,"x_train":700.9607998747,"imp":20.1781377471,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.7428396647,"x_train":700.9607998747,"imp":19.8539735083,"elo":1288.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":85.3917069912,"x_train":335.8028740883,"imp":24.2778486183,"elo":1131.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":75.1381347608,"x_train":15318.1713297129,"imp":19.208239087,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":500.7827657461,"x_train":15318.1713297129,"imp":17.6227468063,"elo":1348.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.701764822,"x_train":2.6626901865,"imp":29.5600167603,"elo":947.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0969305634,"x_train":536.3421035528,"imp":20.740422312,"elo":1213.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.7378187563,"x_train":536.3421035528,"imp":19.740505476,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_small/tuning_trajectories_explorer.html
index 9fdd5b6685859cad3fb1f55191f8518c6215a848..862737dddb918ad659b605d4a45d7d96e9b20247 100644
--- a/data/imputation_no/splits_lite/tasks_binary/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":298.4715930877,"imp":19.299143385,"elo":1373.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.776461578,"imp":19.0885948395,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":11.3675639805,"imp":18.9204216226,"elo":1324.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":27.7361311763,"imp":18.8492849629,"elo":1323.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":56.7638986603,"imp":18.8190642037,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":152.5302644531,"imp":18.8481109465,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":302.4049671528,"imp":18.8723652329,"elo":1320.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":611.2316361016,"imp":18.8778689133,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":916.6605962686,"imp":18.9547475515,"elo":1308.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1228.6932168834,"imp":18.9242688475,"elo":1315.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.2394562483,"imp":21.8047888802,"elo":1229.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.9235675502,"imp":21.0049137645,"elo":1272.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":13.5672030509,"imp":20.4589594826,"elo":1296.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":26.8798672275,"imp":20.3148939565,"elo":1304.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":69.6756367181,"imp":20.0972879574,"elo":1318.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":141.1714518585,"imp":19.9789442855,"elo":1324.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":284.6537506459,"imp":20.0287939205,"elo":1321.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":426.257320709,"imp":20.0096278872,"elo":1322.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":571.4736016874,"imp":20.0715925497,"elo":1316.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":6.1072816838,"imp":20.6467957409,"elo":1254.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":13.1289398573,"imp":20.4205035678,"elo":1266.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":33.6623952718,"imp":20.0846013378,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":67.1158913983,"imp":19.9951092618,"elo":1285.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":169.6245147956,"imp":19.9231972039,"elo":1293.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":340.6955797003,"imp":19.7743901048,"elo":1301.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":679.8713931322,"imp":19.7284413634,"elo":1302.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1018.1194397335,"imp":19.7618628678,"elo":1303.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1364.2004288288,"imp":19.7715368274,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.983283782,"imp":27.7842605155,"elo":1027.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.8106834519,"imp":26.4927268046,"elo":1041.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.1319737136,"imp":24.6658204497,"elo":1083.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":17.6833597648,"imp":24.1561249426,"elo":1104.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.7139412379,"imp":23.820115126,"elo":1121.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":87.1516779041,"imp":23.6336270142,"elo":1128.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":173.8653607249,"imp":23.5609883706,"elo":1133.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":260.0380199862,"imp":23.4216893798,"elo":1145.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":349.0651424646,"imp":23.4418941802,"elo":1145.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.4776029853,"imp":26.6494743623,"elo":999.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.3198175557,"imp":25.9494209137,"elo":1049.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":16.004059149,"imp":24.4361396416,"elo":1113.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":32.5922652125,"imp":23.7452047624,"elo":1137.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":79.5473090208,"imp":23.0339725602,"elo":1156.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":159.3682421601,"imp":22.7529497112,"elo":1168.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":321.9650866878,"imp":22.7844723429,"elo":1169.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":484.324679178,"imp":22.72850836,"elo":1170.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":650.7592051506,"imp":22.8586156798,"elo":1182.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.4736320925,"imp":23.479048055,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":6.1022691557,"imp":20.6965935923,"elo":1233.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":14.2595602286,"imp":19.2660525751,"elo":1306.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":26.7224200508,"imp":18.8621575707,"elo":1323.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":64.7711179435,"imp":18.6682260072,"elo":1331.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":128.979983649,"imp":18.6547592917,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":256.6446480506,"imp":18.6400394252,"elo":1338.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":382.3781018725,"imp":18.5727809484,"elo":1339.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":512.2070722628,"imp":18.6753615662,"elo":1340.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":19.3292393435,"imp":15.1766134606,"elo":1459.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":13.4010979517,"imp":23.0567117213,"elo":1206.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":46.9223002438,"imp":21.9214258004,"elo":1214.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":130.5934927336,"imp":19.541177462,"elo":1295.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":243.5354599493,"imp":19.0510695895,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":579.0165854136,"imp":18.5018104555,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1164.1941811379,"imp":19.0969827597,"elo":1310.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2285.7499058023,"imp":19.925274825,"elo":1283.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3416.2270282822,"imp":20.4523547263,"elo":1272.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4618.5023417328,"imp":20.706053777,"elo":1266.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":13.9928451766,"imp":23.7775992592,"elo":1132.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":27.0284046233,"imp":29.9209282844,"elo":1031.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":65.0087037442,"imp":27.4771864334,"elo":1080.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":163.7081734538,"imp":24.7148404115,"elo":1148.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":196.1786690245,"imp":24.6012590829,"elo":1153.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3945714661,"imp":28.4809782341,"elo":998.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.8045226314,"imp":27.2784538951,"elo":1024.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":10.2666426077,"imp":26.435569494,"elo":1052.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":22.5942287967,"imp":25.9917037928,"elo":1068.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":59.458872306,"imp":25.4781849735,"elo":1085.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":121.2365649401,"imp":25.3738598676,"elo":1095.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":244.8550522727,"imp":25.293326816,"elo":1099.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":368.7590966134,"imp":25.2706408038,"elo":1098.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":494.6995565402,"imp":25.1637875788,"elo":1101.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":11.0934085905,"imp":19.4554078211,"elo":1285.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.9886873017,"imp":17.9102476825,"elo":1365.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":67.1873109421,"imp":16.6641009016,"elo":1436.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":135.870476352,"imp":16.5195893845,"elo":1443.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":350.3376696757,"imp":16.1478515755,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":701.5287510764,"imp":16.0109215853,"elo":1471.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1392.1340578984,"imp":16.0437257014,"elo":1471.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2087.8411572447,"imp":16.0295870315,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2791.9683399256,"imp":15.9821822977,"elo":1483.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.5361255471,"imp":13.5432340038,"elo":1619.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.6926284807,"imp":13.6807735522,"elo":1601.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":47.4652638957,"imp":13.7396727515,"elo":1570.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":100.4462202862,"imp":13.8415887589,"elo":1559.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":257.9252062887,"imp":13.8347319768,"elo":1548.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":526.5698153079,"imp":14.0027529779,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1055.8495282233,"imp":13.7670323318,"elo":1541.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1596.2092361301,"imp":13.7952401317,"elo":1547.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2125.1735517383,"imp":13.9006336234,"elo":1544.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":8.9906883617,"imp":21.8982128168,"elo":1267.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":52.0419884348,"imp":19.7147519228,"elo":1248.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":82.5176074755,"imp":18.6684835515,"elo":1292.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":165.2724746986,"imp":17.7520974537,"elo":1335.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":324.4987032773,"imp":17.4391146338,"elo":1343.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":823.5187233903,"imp":16.871639658,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1657.5635357281,"imp":16.4163156897,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3300.3444632835,"imp":16.2220884633,"elo":1377.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":4936.7697365795,"imp":16.2086219774,"elo":1383.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":6619.8287889361,"imp":16.2347345496,"elo":1388.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.397422576,"imp":16.1437661708,"elo":1425.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":37.343177154,"imp":4.8234330461,"elo":1773.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.2915962934,"imp":27.4773085038,"elo":1061.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.9632436902,"imp":12.6956573406,"elo":1570.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.8040758648,"imp":20.183392844,"elo":1275.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":15.0911403676,"imp":19.9717151731,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":49.7796487814,"imp":19.453946502,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":110.2249442572,"imp":18.9553767499,"elo":1328.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":278.8447574852,"imp":18.548766148,"elo":1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":562.6102815387,"imp":18.2811091284,"elo":1363.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1109.0426689601,"imp":18.1147126793,"elo":1371.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1688.1872952436,"imp":18.0552753035,"elo":1377.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2289.5124290727,"imp":18.0338520361,"elo":1375.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.7520413762,"imp":13.2884055011,"elo":1563.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":4.6649008838,"imp":11.9681909863,"elo":1623.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":439.9020625591,"imp":28.4266551476,"elo":1100.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":801.2621451867,"imp":28.3023669822,"elo":1104.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1413.7529870408,"imp":27.9719478668,"elo":1109.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2244.4819077532,"imp":27.2844715319,"elo":1118.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5605.4218651295,"imp":25.9543688074,"elo":1133.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11324.5526346468,"imp":24.8180043407,"elo":1145.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":22976.1739807235,"imp":24.4006300562,"elo":1152.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":33851.8786642142,"imp":24.1681293227,"elo":1161.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":44524.4248051345,"imp":23.9203073965,"elo":1172.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.2131979048,"imp":18.0863495127,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":7.7572571145,"imp":27.2198236231,"elo":1021.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":19.6428307806,"imp":24.4365791814,"elo":1094.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":58.6478087255,"imp":22.2266573876,"elo":1150.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":126.8672414768,"imp":21.4333626041,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":319.7555992434,"imp":20.8436050471,"elo":1222.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":628.8366489884,"imp":20.5533355496,"elo":1234.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1222.8237025273,"imp":20.5284146947,"elo":1226.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1880.3608960104,"imp":20.3797789,"elo":1235.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2535.7310668162,"imp":20.2715286601,"elo":1237.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.9818637578,"imp":23.3405820092,"elo":1133.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.635768988,"imp":21.0382467846,"elo":1232.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":16.6218131925,"imp":20.4917513015,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":31.8397504377,"imp":20.2363971744,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.7445970232,"imp":20.0408079356,"elo":1285.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":173.5566472908,"imp":20.0232887071,"elo":1292.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":346.7594994462,"imp":19.9277399495,"elo":1300.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":520.510860917,"imp":19.8171993028,"elo":1307.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":700.9607998747,"imp":19.8357737511,"elo":1307.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":335.8028740883,"imp":24.288992934,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":885.3987291777,"imp":21.6686876822,"elo":1217.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2675.7253682148,"imp":18.9125545545,"elo":1301.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":5956.5440310824,"imp":18.2152010013,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":14642.2890769422,"imp":17.6722841385,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":15318.1713297129,"imp":17.661889648,"elo":1363.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.6626901865,"imp":29.5673725207,"elo":960.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.3549384522,"imp":26.2933154162,"elo":1034.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":13.526226486,"imp":22.028127174,"elo":1174.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":26.9163621783,"imp":20.2903893255,"elo":1251.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":67.3146528947,"imp":19.9100221831,"elo":1270.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":133.4916186547,"imp":19.8050702158,"elo":1277.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":266.8571042097,"imp":19.7639354937,"elo":1273.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":400.0742214596,"imp":19.7869229307,"elo":1272.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":536.3421035528,"imp":19.7573107548,"elo":1270.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0361740111,"x_train":298.4715930877,"imp":19.299143385,"elo":1373.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"x_train":3.776461578,"imp":19.0885948395,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.051816687,"x_train":11.3675639805,"imp":18.9204216226,"elo":1324.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0997115139,"x_train":27.7361311763,"imp":18.8492849629,"elo":1323.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1513344731,"x_train":56.7638986603,"imp":18.8190642037,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1887199469,"x_train":152.5302644531,"imp":18.8481109465,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2103626636,"x_train":302.4049671528,"imp":18.8723652329,"elo":1320.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2622354828,"x_train":611.2316361016,"imp":18.8778689133,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3145487088,"x_train":916.6605962686,"imp":18.9547475515,"elo":1308.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3334156391,"x_train":1228.6932168834,"imp":18.9242688475,"elo":1315.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0364337543,"x_train":3.2394562483,"imp":21.8047888802,"elo":1229.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0646948782,"x_train":5.9235675502,"imp":21.0049137645,"elo":1272.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0881631024,"x_train":13.5672030509,"imp":20.4589594826,"elo":1296.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1641604118,"x_train":26.8798672275,"imp":20.3148939565,"elo":1304.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1371985301,"x_train":69.6756367181,"imp":20.0972879574,"elo":1318.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1571534493,"x_train":141.1714518585,"imp":19.9789442855,"elo":1324.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2082282212,"x_train":284.6537506459,"imp":20.0287939205,"elo":1321.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2292933372,"x_train":426.257320709,"imp":20.0096278872,"elo":1322.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2636490914,"x_train":571.4736016874,"imp":20.0715925497,"elo":1316.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.016650456,"x_train":6.1072816838,"imp":20.6467957409,"elo":1254.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0349337669,"x_train":13.1289398573,"imp":20.4205035678,"elo":1266.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0579169347,"x_train":33.6623952718,"imp":20.0846013378,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0799795713,"x_train":67.1158913983,"imp":19.9951092618,"elo":1285.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1070850335,"x_train":169.6245147956,"imp":19.9231972039,"elo":1293.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1311981574,"x_train":340.6955797003,"imp":19.7743901048,"elo":1301.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1410621911,"x_train":679.8713931322,"imp":19.7284413634,"elo":1302.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1315950546,"x_train":1018.1194397335,"imp":19.7618628678,"elo":1303.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1473350525,"x_train":1364.2004288288,"imp":19.7715368274,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3735107819,"x_train":1.983283782,"imp":27.7842605155,"elo":1027.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4387177066,"x_train":3.8106834519,"imp":26.4927268046,"elo":1041.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4048503292,"x_train":9.1319737136,"imp":24.6658204497,"elo":1083.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4948762569,"x_train":17.6833597648,"imp":24.1561249426,"elo":1104.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6654595574,"x_train":43.7139412379,"imp":23.820115126,"elo":1121.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7853524883,"x_train":87.1516779041,"imp":23.6336270142,"elo":1128.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9217284909,"x_train":173.8653607249,"imp":23.5609883706,"elo":1133.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0260469415,"x_train":260.0380199862,"imp":23.4216893798,"elo":1145.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9968010118,"x_train":349.0651424646,"imp":23.4418941802,"elo":1145.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4267781724,"x_train":3.4776029853,"imp":26.6494743623,"elo":999.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8858037829,"x_train":6.3198175557,"imp":25.9494209137,"elo":1049.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2888728302,"x_train":16.004059149,"imp":24.4361396416,"elo":1113.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0867775237,"x_train":32.5922652125,"imp":23.7452047624,"elo":1137.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.369188339,"x_train":79.5473090208,"imp":23.0339725602,"elo":1156.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.8656965125,"x_train":159.3682421601,"imp":22.7529497112,"elo":1168.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.8893865347,"x_train":321.9650866878,"imp":22.7844723429,"elo":1169.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.8776147813,"x_train":484.324679178,"imp":22.72850836,"elo":1170.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6157797575,"x_train":650.7592051506,"imp":22.8586156798,"elo":1182.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1682254962,"x_train":2.4736320925,"imp":23.479048055,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5110863698,"x_train":6.1022691557,"imp":20.6965935923,"elo":1233.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7399382766,"x_train":14.2595602286,"imp":19.2660525751,"elo":1306.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9104777545,"x_train":26.7224200508,"imp":18.8621575707,"elo":1323.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1034172812,"x_train":64.7711179435,"imp":18.6682260072,"elo":1331.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1824620436,"x_train":128.979983649,"imp":18.6547592917,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3415381857,"x_train":256.6446480506,"imp":18.6400394252,"elo":1338.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3940963769,"x_train":382.3781018725,"imp":18.5727809484,"elo":1339.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2769181111,"x_train":512.2070722628,"imp":18.6753615662,"elo":1340.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.5187065444,"x_train":19.3292393435,"imp":15.1766134606,"elo":1459.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.329248032,"x_train":13.4010979517,"imp":23.0567117213,"elo":1206.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8658037861,"x_train":46.9223002438,"imp":21.9214258004,"elo":1214.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8062580347,"x_train":130.5934927336,"imp":19.541177462,"elo":1295.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9476753212,"x_train":243.5354599493,"imp":19.0510695895,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1594027281,"x_train":579.0165854136,"imp":18.5018104555,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1909206629,"x_train":1164.1941811379,"imp":19.0969827597,"elo":1310.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.7233625262,"x_train":2285.7499058023,"imp":19.925274825,"elo":1283.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3337898401,"x_train":3416.2270282822,"imp":20.4523547263,"elo":1272.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2181811333,"x_train":4618.5023417328,"imp":20.706053777,"elo":1266.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5777056813,"x_train":13.9928451766,"imp":23.7775992592,"elo":1132.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0303599261,"x_train":27.0284046233,"imp":29.9209282844,"elo":1031.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2237719987,"x_train":65.0087037442,"imp":27.4771864334,"elo":1080.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3253204999,"x_train":163.7081734538,"imp":24.7148404115,"elo":1148.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3495945281,"x_train":196.1786690245,"imp":24.6012590829,"elo":1153.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0405290578,"x_train":0.3945714661,"imp":28.4809782341,"elo":998.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1142508754,"x_train":2.8045226314,"imp":27.2784538951,"elo":1024.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2873847788,"x_train":10.2666426077,"imp":26.435569494,"elo":1052.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4467472843,"x_train":22.5942287967,"imp":25.9917037928,"elo":1068.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6179766911,"x_train":59.458872306,"imp":25.4781849735,"elo":1085.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7267800191,"x_train":121.2365649401,"imp":25.3738598676,"elo":1095.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7605122217,"x_train":244.8550522727,"imp":25.293326816,"elo":1099.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8870651701,"x_train":368.7590966134,"imp":25.2706408038,"elo":1098.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9355489781,"x_train":494.6995565402,"imp":25.1637875788,"elo":1101.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.4598647569,"x_train":11.0934085905,"imp":19.4554078211,"elo":1285.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.3602889236,"x_train":24.9886873017,"imp":17.9102476825,"elo":1365.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.2554138788,"x_train":67.1873109421,"imp":16.6641009016,"elo":1436.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.1157993328,"x_train":135.870476352,"imp":16.5195893845,"elo":1443.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.7725018708,"x_train":350.3376696757,"imp":16.1478515755,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.6120198051,"x_train":701.5287510764,"imp":16.0109215853,"elo":1471.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.3758571791,"x_train":1392.1340578984,"imp":16.0437257014,"elo":1471.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.5072646854,"x_train":2087.8411572447,"imp":16.0295870315,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":23.9328857361,"x_train":2791.9683399256,"imp":15.9821822977,"elo":1483.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5499649048,"x_train":5.5361255471,"imp":13.5432340038,"elo":1619.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3877879479,"x_train":16.6926284807,"imp":13.6807735522,"elo":1601.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.0362039712,"x_train":47.4652638957,"imp":13.7396727515,"elo":1570.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1588746477,"x_train":100.4462202862,"imp":13.8415887589,"elo":1559.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.6152050014,"x_train":257.9252062887,"imp":13.8347319768,"elo":1548.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.2170693651,"x_train":526.5698153079,"imp":14.0027529779,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.1535813458,"x_train":1055.8495282233,"imp":13.7670323318,"elo":1541.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0359551071,"x_train":1596.2092361301,"imp":13.7952401317,"elo":1547.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7107599974,"x_train":2125.1735517383,"imp":13.9006336234,"elo":1544.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7241354212,"x_train":8.9906883617,"imp":21.8982128168,"elo":1267.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8096498633,"x_train":52.0419884348,"imp":19.7147519228,"elo":1248.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":58.7248120485,"x_train":82.5176074755,"imp":18.6684835515,"elo":1292.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":90.6781979203,"x_train":165.2724746986,"imp":17.7520974537,"elo":1335.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":125.590395152,"x_train":324.4987032773,"imp":17.4391146338,"elo":1343.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":191.6133848691,"x_train":823.5187233903,"imp":16.871639658,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":254.4244251102,"x_train":1657.5635357281,"imp":16.4163156897,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":323.4247300441,"x_train":3300.3444632835,"imp":16.2220884633,"elo":1377.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":360.6190160112,"x_train":4936.7697365795,"imp":16.2086219774,"elo":1383.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":408.7718504964,"x_train":6619.8287889361,"imp":16.2347345496,"elo":1388.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1160046101,"x_train":2.397422576,"imp":16.1437661708,"elo":1425.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":5.4226160049,"x_train":37.343177154,"imp":4.8234330461,"elo":1773.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1485826053,"x_train":1.2915962934,"imp":27.4773085038,"elo":1061.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1509533846,"x_train":1.9632436902,"imp":12.6956573406,"elo":1570.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2622436544,"x_train":6.8040758648,"imp":20.183392844,"elo":1275.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4930239494,"x_train":15.0911403676,"imp":19.9717151731,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9344890524,"x_train":49.7796487814,"imp":19.453946502,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.1910121728,"x_train":110.2249442572,"imp":18.9553767499,"elo":1328.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.7229810256,"x_train":278.8447574852,"imp":18.548766148,"elo":1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.8482256852,"x_train":562.6102815387,"imp":18.2811091284,"elo":1363.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":2.0517870766,"x_train":1109.0426689601,"imp":18.1147126793,"elo":1371.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.108383211,"x_train":1688.1872952436,"imp":18.0552753035,"elo":1377.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.0417782615,"x_train":2289.5124290727,"imp":18.0338520361,"elo":1375.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4340112036,"x_train":5.7520413762,"imp":13.2884055011,"elo":1563.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3608180836,"x_train":4.6649008838,"imp":11.9681909863,"elo":1623.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.4449928547,"x_train":439.9020625591,"imp":28.4266551476,"elo":1100.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7828189872,"x_train":801.2621451867,"imp":28.3023669822,"elo":1104.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.6843898688,"x_train":1413.7529870408,"imp":27.9719478668,"elo":1109.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5679815934,"x_train":2244.4819077532,"imp":27.2844715319,"elo":1118.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.9965249174,"x_train":5605.4218651295,"imp":25.9543688074,"elo":1133.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.7064222686,"x_train":11324.5526346468,"imp":24.8180043407,"elo":1145.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.8749951615,"x_train":22976.1739807235,"imp":24.4006300562,"elo":1152.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":23.2068554068,"x_train":33851.8786642142,"imp":24.1681293227,"elo":1161.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.3301528826,"x_train":44524.4248051345,"imp":23.9203073965,"elo":1172.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0524064596,"x_train":1.2131979048,"imp":18.0863495127,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1464537212,"x_train":7.7572571145,"imp":27.2198236231,"elo":1021.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2861450331,"x_train":19.6428307806,"imp":24.4365791814,"elo":1094.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7112520661,"x_train":58.6478087255,"imp":22.2266573876,"elo":1150.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0077171512,"x_train":126.8672414768,"imp":21.4333626041,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2404297158,"x_train":319.7555992434,"imp":20.8436050471,"elo":1222.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6141725337,"x_train":628.8366489884,"imp":20.5533355496,"elo":1234.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1193572199,"x_train":1222.8237025273,"imp":20.5284146947,"elo":1226.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.2621370554,"x_train":1880.3608960104,"imp":20.3797789,"elo":1235.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.545560207,"x_train":2535.7310668162,"imp":20.2715286601,"elo":1237.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"x_train":1.9818637578,"imp":23.3405820092,"elo":1133.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.465082175,"x_train":5.635768988,"imp":21.0382467846,"elo":1232.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5859087742,"x_train":16.6218131925,"imp":20.4917513015,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6695677292,"x_train":31.8397504377,"imp":20.2363971744,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7535860957,"x_train":86.7445970232,"imp":20.0408079356,"elo":1285.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8503171682,"x_train":173.5566472908,"imp":20.0232887071,"elo":1292.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9856110848,"x_train":346.7594994462,"imp":19.9277399495,"elo":1300.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.238113448,"x_train":520.510860917,"imp":19.8171993028,"elo":1307.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7548289838,"x_train":700.9607998747,"imp":19.8357737511,"elo":1307.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":85.3917069912,"x_train":335.8028740883,"imp":24.288992934,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":175.3477427959,"x_train":885.3987291777,"imp":21.6686876822,"elo":1217.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":278.6108939781,"x_train":2675.7253682148,"imp":18.9125545545,"elo":1301.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":396.3850032557,"x_train":5956.5440310824,"imp":18.2152010013,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":478.1228243423,"x_train":14642.2890769422,"imp":17.6722841385,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":500.7827657461,"x_train":15318.1713297129,"imp":17.661889648,"elo":1363.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.701764822,"x_train":2.6626901865,"imp":29.5673725207,"elo":960.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8119673878,"x_train":5.3549384522,"imp":26.2933154162,"elo":1034.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.0771486904,"x_train":13.526226486,"imp":22.028127174,"elo":1174.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.5044555898,"x_train":26.9163621783,"imp":20.2903893255,"elo":1251.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.4747611582,"x_train":67.3146528947,"imp":19.9100221831,"elo":1270.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.1791495409,"x_train":133.4916186547,"imp":19.8050702158,"elo":1277.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.7604902387,"x_train":266.8571042097,"imp":19.7639354937,"elo":1273.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.6481693302,"x_train":400.0742214596,"imp":19.7869229307,"elo":1272.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.6495357454,"x_train":536.3421035528,"imp":19.7573107548,"elo":1270.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_small/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_binary/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..d54865996a37b464ae8d6748cd52aaf9755a0e14
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_binary/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "RealTabPFN-2.5 (default)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "LimiX (default)", "AutoGluon 1.4 (best, 4h)", "TabDPT-Turbo (default)", "RealMLP (tuned)", "TabDPT (tuned + ensembled)", "BetaTabPFN (default)", "TabM (tuned + ensembled)", "iLTM (tuned + ensembled)", "TabSwift (default)", "TabDPT (tuned)", "LightGBM (tuned + ensembled)", "TabM (tuned)", "CatBoost (tuned)", "CatBoost (default)", "ModernNCA (tuned)", "CatBoost (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "XGBoost (tuned + ensembled)", "EBM (tuned + ensembled)", "iLTM (tuned)", "RealMLP (default)", "XGBoost (tuned)", "TabM (default)", "xRFM (tuned + ensembled)", "ChimeraBoost (tuned)", "LightGBM (tuned)", "ModernNCA (tuned + ensembled)", "SAP-RPT-OSS (default)", "EBM (default)", "EBM (tuned)", "TabDPT (default)", "TorchMLP (tuned + ensembled)", "ChimeraBoost (default)", "xRFM (tuned)", "ModernNCA (default)", "TorchMLP (tuned)", "FastaiMLP (tuned + ensembled)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (tuned + ensembled)", "FastaiMLP (tuned)", "iLTM (default)", "XGBoost (default)", "ExtraTrees (tuned)", "OrionMSP (default)", "LightGBM (default)", "PerpetualBooster (tuned)", "RandomForest (tuned + ensembled)", "TabSTAR (default)", "RandomForest (tuned)", "TabFlex (default)", "PerpetualBooster (default)", "ExtraTrees (default)", "TorchMLP (default)", "RandomForest (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "xRFM (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 1.0, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 1.0, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 1.0, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.1428571492433548, NaN, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.5952380895614624, 0.6190476417541504, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.9523809552192688, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 1.0, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.9523809552192688, 1.0, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 0.761904776096344, 1.0, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.190476194024086, 0.523809552192688, NaN, 0.4761904776096344, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.976190447807312, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 1.0, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 1.0, 1.0, 1.0, 0.9523809552192688, 0.8095238208770752, 1.0, 1.0, 0.9047619104385376, 1.0, 1.0, 0.9523809552192688, 1.0, 1.0, 1.0, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.1428571492433548, 0.4761904776096344, 0.523809552192688, NaN, 0.5714285969734192, 0.4285714328289032, 0.6190476417541504, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6190476417541504, 0.8095238208770752, 0.9047619104385376, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 1.0, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 1.0, 0.9523809552192688, 0.8571428656578064, 1.0, 0.9047619104385376, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.1428571492433548, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, NaN, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 0.761904776096344, 0.761904776096344, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 1.0, 0.9047619104385376, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.7142857313156128, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.9523809552192688, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 0.7142857313156128, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.0476190485060215, 0.4047619104385376, 0.4285714328289032, 0.5714285969734192, 0.523809552192688, NaN, 0.4761904776096344, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.523809552192688, 0.6666666865348816, 0.6190476417541504, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.8571428656578064, 1.0, 0.9523809552192688, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9523809552192688, 0.761904776096344, 0.9523809552192688, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 1.0], [0.190476194024086, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.5714285969734192, 0.523809552192688, NaN, 0.523809552192688, 0.7142857313156128, 0.5714285969734192, 0.6666666865348816, 0.4761904776096344, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.761904776096344, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.095238097012043, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, NaN, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.8095238208770752, 0.8095238208770752, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 1.0, 0.8095238208770752, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.761904776096344, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 1.0], [0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, NaN, 0.380952388048172, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.5714285969734192, 0.6190476417541504, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.6666666865348816, 0.9047619104385376, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.6666666865348816, 0.6666666865348816, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.095238097012043, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.6190476417541504, NaN, 0.523809552192688, 0.523809552192688, 0.738095223903656, 0.4761904776096344, 0.6666666865348816, 0.6666666865348816, 0.738095223903656, 0.5714285969734192, 0.6190476417541504, 0.8095238208770752, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.9047619104385376, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.6666666865348816, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.7857142686843872, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8809523582458496, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9523809552192688, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.095238097012043, 0.380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, NaN, 0.5714285969734192, 0.6666666865348816, 0.523809552192688, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.5714285969734192, 0.6190476417541504, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.761904776096344, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.095238097012043, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.4761904776096344, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, NaN, 0.523809552192688, 0.761904776096344, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.5952380895614624, 0.7142857313156128, 0.7142857313156128, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0, 1.0], [0.1428571492433548, 0.2857142984867096, 0.190476194024086, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.261904776096344, 0.3333333432674408, 0.4761904776096344, NaN, 0.3333333432674408, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.738095223903656, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.7142857313156128, 0.9523809552192688, 0.761904776096344, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9285714030265808, 0.9047619104385376, 1.0, 1.0], [0.0476190485060215, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.4761904776096344, 0.523809552192688, 0.4761904776096344, 0.2380952388048172, 0.6666666865348816, NaN, 0.4761904776096344, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6666666865348816, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.523809552192688, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.8095238208770752, 0.9047619104385376, 1.0, 0.9047619104385376, 1.0, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 1.0], [0.0476190485060215, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4761904776096344, 0.380952388048172, 0.523809552192688, NaN, 0.4761904776096344, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.8571428656578064, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.7142857313156128, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.523809552192688, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.6190476417541504, 0.8095238208770752, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.5714285969734192, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.761904776096344, 0.9523809552192688, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0, 1.0], [0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.2380952388048172, 0.261904776096344, 0.2380952388048172, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.4285714328289032, NaN, 0.523809552192688, 0.380952388048172, 0.6190476417541504, 0.4761904776096344, 0.5714285969734192, 0.4761904776096344, 0.5714285969734192, 0.4761904776096344, 0.6666666865348816, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.6190476417541504, 0.6190476417541504, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.7142857313156128, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 1.0], [0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4047619104385376, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, NaN, 0.380952388048172, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.4285714328289032, 0.5714285969734192, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.9523809552192688, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 1.0, 1.0], [0.0, 0.0476190485060215, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.523809552192688, 0.3333333432674408, 0.4285714328289032, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, NaN, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.523809552192688, 0.761904776096344, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.8095238208770752, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.9047619104385376, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 1.0, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688], [0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, NaN, 0.4285714328289032, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.761904776096344, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.9523809552192688, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9523809552192688, 0.8571428656578064, 1.0, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 1.0, 1.0], [0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.4761904776096344, 0.3333333432674408, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, NaN, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.6904761791229248, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.523809552192688, 0.7142857313156128, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.7142857313156128, 0.9523809552192688, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688], [0.0476190485060215, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, NaN, 0.6190476417541504, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.6666666865348816, 0.8571428656578064, 0.761904776096344, 0.9047619104385376, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.4761904776096344, 0.3333333432674408, 0.380952388048172, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.380952388048172, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.5714285969734192, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.761904776096344, 0.9047619104385376, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, NaN, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.8095238208770752, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 1.0], [0.0, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.2857142984867096, 0.2380952388048172, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.523809552192688, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, NaN, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.4285714328289032, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.8095238208770752, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.6190476417541504, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.6666666865348816, 0.9047619104385376, 0.7142857313156128, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.2380952388048172, 0.380952388048172, 0.3333333432674408, 0.5714285969734192, 0.380952388048172, 0.523809552192688, 0.3095238208770752, 0.5714285969734192, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, NaN, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.8095238208770752, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 1.0], [0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, NaN, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.5952380895614624, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.8095238208770752, 0.6666666865348816, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 1.0, 0.7142857313156128, 0.9047619104385376, 0.7142857313156128, 0.9523809552192688, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0], [0.095238097012043, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, NaN, 0.4285714328289032, 0.6190476417541504, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.5714285969734192, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.8095238208770752, 0.761904776096344, 0.6190476417541504, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.0476190485060215, 0.095238097012043, 0.02380952425301075, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, NaN, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.5714285969734192, 0.8571428656578064, 0.761904776096344, 0.5952380895614624, 0.6190476417541504, 0.8095238208770752, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.9047619104385376, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.4285714328289032, NaN, 0.4761904776096344, 0.523809552192688, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.523809552192688, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.8095238208770752, 0.9523809552192688, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.4761904776096344, 0.523809552192688, 0.523809552192688, NaN, 0.4285714328289032, 0.6190476417541504, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.523809552192688, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.9047619104385376, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.9047619104385376, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 1.0], [0.095238097012043, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.380952388048172, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.5714285969734192, NaN, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.523809552192688, 0.4047619104385376, 0.4761904776096344, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376], [0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.5714285969734192, 0.380952388048172, 0.5714285969734192, NaN, 0.523809552192688, 0.380952388048172, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4285714328289032, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688], [0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.190476194024086, 0.3333333432674408, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.5714285969734192, 0.3333333432674408, 0.5714285969734192, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, NaN, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 1.0], [0.0476190485060215, 0.0, 0.0476190485060215, 0.190476194024086, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.2380952388048172, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.4761904776096344, NaN, 0.380952388048172, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.6190476417541504, 0.4761904776096344, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 1.0], [0.095238097012043, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.523809552192688, 0.4285714328289032, 0.6190476417541504, 0.5714285969734192, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.6190476417541504, NaN, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.6190476417541504, 0.6666666865348816, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.7857142686843872, 0.9047619104385376, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.9523809552192688], [0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.380952388048172, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, NaN, 0.5714285969734192, 0.4761904776096344, 0.5714285969734192, 0.6190476417541504, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.761904776096344, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 1.0], [0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.2380952388048172, 0.2380952388048172, 0.4761904776096344, 0.4285714328289032, 0.5952380895614624, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, NaN, 0.4761904776096344, 0.4761904776096344, 0.6190476417541504, 0.4761904776096344, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.3333333432674408, 0.4761904776096344, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.523809552192688, 0.5714285969734192, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.523809552192688, NaN, 0.6190476417541504, 0.6190476417541504, 0.4285714328289032, 0.4761904776096344, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 1.0], [0.0, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2380952388048172, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.4761904776096344, 0.3333333432674408, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.523809552192688, 0.380952388048172, NaN, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.761904776096344, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9523809552192688, 0.9523809552192688], [0.0, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.4761904776096344, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.523809552192688, NaN, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.190476194024086, 0.190476194024086, 0.3333333432674408, 0.4761904776096344, 0.523809552192688, 0.4761904776096344, 0.4285714328289032, 0.380952388048172, 0.523809552192688, 0.3333333432674408, 0.3333333432674408, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, NaN, 0.5714285969734192, 0.5714285969734192, 0.4761904776096344, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 1.0], [0.0476190485060215, 0.0476190485060215, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, NaN, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.4761904776096344, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.9523809552192688, 1.0], [0.095238097012043, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.190476194024086, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.4761904776096344, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.523809552192688, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, NaN, 0.4761904776096344, 0.4761904776096344, 0.761904776096344, 0.761904776096344, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688], [0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.4047619104385376, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.523809552192688, 0.3333333432674408, 0.523809552192688, 0.380952388048172, 0.523809552192688, NaN, 0.4761904776096344, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.5476190447807312, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.5476190447807312, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8571428656578064, 0.761904776096344, 0.7142857313156128, 0.9047619104385376, 0.6666666865348816, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688], [0.095238097012043, 0.1428571492433548, 0.0476190485060215, 0.0, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3571428656578064, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.523809552192688, 0.523809552192688, NaN, 0.523809552192688, 0.523809552192688, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.4761904776096344, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.9523809552192688, 0.8571428656578064, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688], [0.095238097012043, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2142857164144516, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2380952388048172, 0.3333333432674408, 0.4761904776096344, NaN, 0.6190476417541504, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.9047619104385376, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.9047619104385376], [0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.380952388048172, 0.4761904776096344, 0.380952388048172, NaN, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.9523809552192688, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.9047619104385376], [0.0, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, NaN, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.7142857313156128, 0.6190476417541504, 0.4761904776096344, 0.6190476417541504, 0.6190476417541504, 0.761904776096344, 0.6190476417541504, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.9047619104385376], [0.0476190485060215, 0.0, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.3333333432674408, 0.1428571492433548, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.4523809552192688, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.523809552192688, NaN, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.6666666865348816, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.8571428656578064, 0.5714285969734192, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.9047619104385376, 1.0], [0.095238097012043, 0.095238097012043, 0.0, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.523809552192688, 0.523809552192688, 0.4285714328289032, 0.6190476417541504, NaN, 0.523809552192688, 0.4285714328289032, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376], [0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.380952388048172, 0.2380952388048172, 0.380952388048172, 0.4047619104385376, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4761904776096344, NaN, 0.4285714328289032, 0.5714285969734192, 0.523809552192688, 0.380952388048172, 0.380952388048172, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.4761904776096344, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376], [0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.380952388048172, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.4761904776096344, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.9047619104385376], [0.0476190485060215, 0.0, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.3333333432674408, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.523809552192688, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.761904776096344, 0.523809552192688, 0.761904776096344, 0.6666666865348816, 0.8571428656578064, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.9523809552192688, 0.9523809552192688], [0.0, 0.0, 0.0, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0, 0.0, 0.095238097012043, 0.1190476194024086, 0.095238097012043, 0.095238097012043, 0.261904776096344, 0.0476190485060215, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.1428571492433548, 0.095238097012043, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.4761904776096344, NaN, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9523809552192688], [0.0476190485060215, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.2857142984867096, 0.3333333432674408, 0.0476190485060215, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.0, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.095238097012043, 0.3333333432674408, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.4523809552192688, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.380952388048172, 0.6190476417541504, 0.4285714328289032, NaN, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376], [0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.190476194024086, 0.0476190485060215, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.1428571492433548, 0.095238097012043, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.2380952388048172, 0.2380952388048172, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, 0.2857142984867096, 0.4285714328289032, 0.4761904776096344, 0.6190476417541504, 0.4285714328289032, 0.523809552192688, 0.4761904776096344, 0.523809552192688, NaN, 0.4761904776096344, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.761904776096344, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.9047619104385376], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.523809552192688, NaN, 0.6190476417541504, 0.7142857313156128, 0.523809552192688, 0.523809552192688, 0.7142857313156128, 0.6666666865348816, 0.8571428656578064, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.9523809552192688], [0.095238097012043, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.095238097012043, 0.0476190485060215, 0.523809552192688, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.523809552192688, 0.380952388048172, NaN, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.380952388048172, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.9047619104385376], [0.0476190485060215, 0.0, 0.0, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.190476194024086, 0.0476190485060215, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, NaN, 0.6190476417541504, 0.4761904776096344, 0.6666666865348816, 0.4761904776096344, 0.738095223903656, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.9523809552192688], [0.0, 0.095238097012043, 0.0, 0.0, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.2380952388048172, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.1428571492433548, 0.190476194024086, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, 0.380952388048172, NaN, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8571428656578064], [0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2380952388048172, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2380952388048172, 0.380952388048172, 0.3333333432674408, 0.523809552192688, 0.2857142984867096, 0.4285714328289032, 0.3333333432674408, 0.380952388048172, 0.2380952388048172, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.4761904776096344, NaN, 0.5714285969734192, 0.4761904776096344, 0.5714285969734192, 0.380952388048172, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.9047619104385376], [0.0476190485060215, 0.0476190485060215, 0.0, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.2857142984867096, 0.2380952388048172, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.1428571492433548, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.5714285969734192, 0.3333333432674408, 0.4761904776096344, 0.4285714328289032, NaN, 0.380952388048172, 0.6190476417541504, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.9047619104385376], [0.0, 0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.190476194024086, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.2380952388048172, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.6190476417541504, NaN, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.9047619104385376], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.2380952388048172, 0.0476190485060215, 0.2380952388048172, 0.1428571492433548, 0.0476190485060215, 0.2380952388048172, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.1428571492433548, 0.4285714328289032, 0.261904776096344, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, NaN, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.9523809552192688], [0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.0, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.0476190485060215, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.6190476417541504, 0.2857142984867096, 0.4285714328289032, 0.6190476417541504, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, NaN, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8095238208770752, 0.9047619104385376], [0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.2142857164144516, 0.2380952388048172, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.3333333432674408, 0.380952388048172, 0.2380952388048172, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.2380952388048172, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.380952388048172, NaN, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.9523809552192688, 0.9523809552192688], [0.0, 0.0, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.0476190485060215, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.4285714328289032, 0.190476194024086, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.3333333432674408, NaN, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.6666666865348816, 0.8571428656578064], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.4761904776096344, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, NaN, 0.738095223903656, 0.6190476417541504, 0.5714285969734192, 0.8095238208770752], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0714285746216774, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.1428571492433548, 0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4761904776096344, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.261904776096344, NaN, 0.523809552192688, 0.523809552192688, 0.761904776096344], [0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.2857142984867096, 0.190476194024086, 0.0476190485060215, 0.1428571492433548, 0.190476194024086, 0.0476190485060215, 0.0476190485060215, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, NaN, 0.5714285969734192, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0, 0.0, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.0476190485060215, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.1428571492433548, 0.3333333432674408, 0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.0476190485060215, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, NaN, 0.9047619104385376], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.095238097012043, 0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.095238097012043, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9345238134},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8594576789},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8554894247},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8247354587},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8161375748},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8085317554},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8015873111},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.789682549},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7453703814},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7447090042},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7394180003},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.699404775},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6769180014},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6587301729},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6388889025},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.638227527},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6352513346},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6286375791},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6216931373},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6044973665},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6008598028},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5978836099},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5873015992},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5856481616},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5820105947},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5763889035},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5671296407},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5582010694},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5548941912},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5542328142},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5383597997},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5360449857},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5350529214},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.533068796},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.531746043},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5287698532},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5271164132},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.5148809616},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5119047728},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5079365185},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4947090054},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4841269964},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.4801587402},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4603174707},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.438161386},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4355158834},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4222883699},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.4126984233},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.402116412},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3905423373},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3875661479},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3852513325},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3809523907},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3789682636},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3736772593},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3541666764},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.3472222303},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3419312251},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.341269852},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.309854503},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2956349278},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2731481545},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2645502716},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2632275201},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2509920697},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2447090011},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2080026503},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2056878362},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1987433909},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1825396878},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1772486814},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1183862455},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0443121702}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_binary/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_binary/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..c6093f52a3ea1fb328830eed8113908c3b51e79c
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_binary/datasets_small/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),TabPFN-3 (default),RealTabPFN-2.5 (default),TabICLv2 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),LimiX (default),"AutoGluon 1.4 (best, 4h)",TabDPT-Turbo (default),RealMLP (tuned),TabDPT (tuned + ensembled),BetaTabPFN (default),TabM (tuned + ensembled),iLTM (tuned + ensembled),TabSwift (default),TabDPT (tuned),LightGBM (tuned + ensembled),TabM (tuned),CatBoost (tuned),CatBoost (default),ModernNCA (tuned),CatBoost (tuned + ensembled),ChimeraBoost (tuned + ensembled),XGBoost (tuned + ensembled),EBM (tuned + ensembled),iLTM (tuned),RealMLP (default),XGBoost (tuned),TabM (default),xRFM (tuned + ensembled),ChimeraBoost (tuned),LightGBM (tuned),ModernNCA (tuned + ensembled),SAP-RPT-OSS (default),EBM (default),EBM (tuned),TabDPT (default),TorchMLP (tuned + ensembled),ChimeraBoost (default),xRFM (tuned),ModernNCA (default),TorchMLP (tuned),FastaiMLP (tuned + ensembled),TabSTAR (tuned),TabSTAR (tuned + ensembled),PerpetualBooster (tuned + ensembled),ExtraTrees (tuned + ensembled),FastaiMLP (tuned),iLTM (default),XGBoost (default),ExtraTrees (tuned),OrionMSP (default),LightGBM (default),PerpetualBooster (tuned),RandomForest (tuned + ensembled),TabSTAR (default),RandomForest (tuned),TabFlex (default),PerpetualBooster (default),ExtraTrees (default),TorchMLP (default),RandomForest (default),FastaiMLP (default),KNN (tuned + ensembled),xRFM (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.85714287,0.8095238,0.85714287,0.85714287,0.95238096,0.8095238,0.9047619,0.8095238,0.9047619,0.9047619,0.9047619,0.85714287,0.95238096,0.95238096,0.9047619,0.85714287,0.9047619,1.0,0.95238096,0.95238096,0.95238096,0.9047619,0.95238096,1.0,0.95238096,0.95238096,0.9047619,0.95238096,0.85714287,0.9047619,0.9047619,0.95238096,0.9047619,0.95238096,0.9047619,0.9047619,0.9047619,0.9047619,1.0,1.0,0.95238096,0.95238096,0.9047619,1.0,0.9047619,0.9047619,0.85714287,1.0,0.95238096,0.9047619,0.9047619,1.0,0.95238096,1.0,0.95238096,1.0,0.95238096,0.9047619,0.95238096,1.0,1.0,0.95238096,1.0,0.95238096,1.0,0.95238096,1.0,0.95238096,0.95238096,1.0,1.0,1.0
+TabPFN-3 (default),0.14285715,,0.47619048,0.52380955,0.5714286,0.5952381,0.61904764,0.61904764,0.7619048,0.6666667,0.61904764,0.71428573,0.71428573,0.7619048,0.95238096,0.85714287,0.7619048,0.85714287,0.95238096,0.95238096,0.9047619,0.8095238,0.95238096,0.85714287,0.9047619,0.9047619,0.9047619,0.8095238,0.9047619,0.85714287,0.85714287,0.8095238,0.9047619,0.85714287,1.0,0.9047619,0.8095238,0.85714287,0.9047619,0.9047619,0.95238096,0.95238096,0.95238096,0.9047619,0.95238096,0.85714287,0.7619048,0.7619048,0.95238096,1.0,0.9047619,0.95238096,0.95238096,1.0,1.0,0.95238096,1.0,0.95238096,0.7619048,1.0,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0,1.0,0.95238096,0.95238096,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.1904762,0.52380955,,0.47619048,0.47619048,0.5714286,0.5714286,0.52380955,0.71428573,0.71428573,0.71428573,0.71428573,0.8095238,0.7619048,0.8095238,0.8095238,0.6666667,0.8095238,0.7619048,0.9047619,0.85714287,0.8095238,0.9047619,0.85714287,0.85714287,0.85714287,0.9047619,0.9047619,0.97619045,0.8095238,0.9047619,0.9047619,0.9047619,0.85714287,0.95238096,0.8095238,0.85714287,0.9047619,0.95238096,0.85714287,0.95238096,0.95238096,1.0,0.9047619,0.9047619,0.95238096,0.8095238,0.8095238,0.9047619,0.9047619,1.0,0.95238096,0.95238096,0.9047619,1.0,1.0,1.0,0.95238096,0.8095238,1.0,1.0,0.9047619,1.0,1.0,0.95238096,1.0,1.0,1.0,0.95238096,0.95238096,1.0,1.0,1.0
+TabICLv2 (default),0.14285715,0.47619048,0.52380955,,0.5714286,0.42857143,0.61904764,0.5714286,0.71428573,0.61904764,0.7619048,0.6666667,0.61904764,0.8095238,0.9047619,0.7619048,0.7619048,0.8095238,0.8095238,0.8095238,0.8095238,0.8095238,0.85714287,0.8095238,0.8095238,0.85714287,0.8095238,0.8095238,0.7619048,0.8095238,0.8095238,0.8095238,0.9047619,0.8095238,0.8095238,0.9047619,0.8095238,0.85714287,0.85714287,0.85714287,0.9047619,0.85714287,0.95238096,0.85714287,0.95238096,1.0,0.8095238,0.8095238,0.9047619,0.85714287,0.9047619,0.8095238,0.95238096,0.9047619,0.9047619,0.95238096,0.95238096,0.9047619,0.8095238,0.9047619,1.0,0.95238096,0.85714287,1.0,0.9047619,1.0,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0,1.0
+TabPFN-2.6 (default),0.14285715,0.42857143,0.52380955,0.42857143,,0.47619048,0.42857143,0.52380955,0.61904764,0.71428573,0.6666667,0.6666667,0.71428573,0.61904764,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.9047619,0.7619048,0.7619048,0.9047619,0.8095238,0.85714287,0.8095238,0.8095238,0.85714287,0.85714287,0.8095238,0.85714287,0.8095238,0.9047619,0.8095238,1.0,0.9047619,0.8095238,0.8095238,0.9047619,0.71428573,0.95238096,0.9047619,0.9047619,0.9047619,0.95238096,0.85714287,0.71428573,0.71428573,0.95238096,0.9047619,0.8095238,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096,1.0,1.0,0.71428573,0.95238096,0.85714287,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,1.0,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.04761905,0.4047619,0.42857143,0.5714286,0.52380955,,0.47619048,0.61904764,0.6666667,0.6666667,0.61904764,0.52380955,0.6666667,0.61904764,0.8095238,0.7619048,0.7619048,0.71428573,0.71428573,0.9047619,0.8095238,0.85714287,0.9047619,0.85714287,0.9047619,0.85714287,0.9047619,0.8095238,0.8095238,0.7619048,0.85714287,0.85714287,0.85714287,0.9047619,0.95238096,0.85714287,0.8095238,0.85714287,0.8095238,0.71428573,0.85714287,0.85714287,0.95238096,0.8095238,0.85714287,0.8095238,0.71428573,0.71428573,0.9047619,0.95238096,0.85714287,0.85714287,1.0,0.95238096,0.95238096,1.0,0.95238096,0.95238096,0.7619048,0.95238096,0.8095238,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.9047619,0.95238096,0.9047619,0.95238096,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.1904762,0.3809524,0.42857143,0.3809524,0.5714286,0.52380955,,0.52380955,0.71428573,0.5714286,0.6666667,0.47619048,0.7619048,0.71428573,0.7619048,0.7619048,0.7619048,0.6666667,0.71428573,0.85714287,0.71428573,0.8095238,0.85714287,0.9047619,0.9047619,0.85714287,0.85714287,0.7619048,0.85714287,0.7619048,0.85714287,0.8095238,0.9047619,0.85714287,0.9047619,0.85714287,0.7619048,0.8095238,0.7619048,0.7619048,0.85714287,0.9047619,0.8095238,0.9047619,0.85714287,0.8095238,0.7619048,0.7619048,0.85714287,0.9047619,0.85714287,0.9047619,0.9047619,0.9047619,1.0,0.95238096,0.9047619,0.9047619,0.7619048,0.9047619,0.9047619,0.9047619,0.95238096,0.9047619,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0952381,0.3809524,0.47619048,0.42857143,0.47619048,0.3809524,0.47619048,,0.5714286,0.5714286,0.61904764,0.5714286,0.6666667,0.5714286,0.8095238,0.8095238,0.6666667,0.61904764,0.71428573,0.8095238,0.85714287,0.8095238,0.9047619,0.85714287,0.8095238,0.7619048,0.85714287,0.8095238,0.85714287,0.7619048,0.8095238,0.85714287,0.8095238,0.85714287,0.95238096,0.8095238,0.8095238,0.8095238,0.8095238,0.71428573,0.85714287,0.85714287,0.9047619,0.85714287,0.9047619,0.8095238,0.71428573,0.71428573,0.9047619,1.0,0.8095238,0.85714287,0.95238096,0.95238096,1.0,0.95238096,0.95238096,0.95238096,0.7619048,0.9047619,0.8095238,0.9047619,0.95238096,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096,0.9047619,0.9047619,0.9047619,0.95238096,1.0
+RealMLP (tuned + ensembled),0.1904762,0.23809524,0.2857143,0.2857143,0.3809524,0.33333334,0.2857143,0.42857143,,0.3809524,0.5714286,0.52380955,0.61904764,0.52380955,0.6666667,0.6666667,0.7619048,0.5714286,0.61904764,0.8095238,0.71428573,0.7619048,0.85714287,0.6666667,0.9047619,0.7619048,0.8095238,0.8095238,0.7619048,0.8095238,0.85714287,0.6666667,0.71428573,0.7619048,0.85714287,0.7619048,0.7619048,0.8095238,0.8095238,0.7619048,0.8095238,0.85714287,0.7619048,0.85714287,0.7619048,0.85714287,0.6666667,0.6666667,0.9047619,0.85714287,0.8095238,0.9047619,0.9047619,0.8095238,0.9047619,0.9047619,0.95238096,0.8095238,0.71428573,0.9047619,0.8095238,0.95238096,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0
+LimiX (default),0.0952381,0.33333334,0.2857143,0.3809524,0.2857143,0.33333334,0.42857143,0.42857143,0.61904764,,0.52380955,0.52380955,0.7380952,0.47619048,0.6666667,0.6666667,0.7380952,0.5714286,0.61904764,0.8095238,0.71428573,0.6666667,0.6666667,0.7619048,0.71428573,0.71428573,0.7619048,0.8095238,0.9047619,0.7619048,0.8095238,0.7619048,0.7619048,0.7619048,0.7619048,0.7619048,0.8095238,0.85714287,0.9047619,0.6666667,0.8095238,0.8095238,0.85714287,0.85714287,0.7619048,0.7619048,0.78571427,0.7619048,0.8095238,0.85714287,0.8095238,0.85714287,0.85714287,0.8095238,0.88095236,0.85714287,0.8095238,0.85714287,0.8095238,0.95238096,0.85714287,0.85714287,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+"AutoGluon 1.4 (best, 4h)",0.0952381,0.3809524,0.2857143,0.23809524,0.33333334,0.3809524,0.33333334,0.3809524,0.42857143,0.47619048,,0.5714286,0.6666667,0.52380955,0.6666667,0.6666667,0.7619048,0.5714286,0.61904764,0.7619048,0.7619048,0.71428573,0.71428573,0.61904764,0.7619048,0.6666667,0.7619048,0.6666667,0.7619048,0.6666667,0.7619048,0.71428573,0.7619048,0.71428573,0.8095238,0.6666667,0.71428573,0.7619048,0.85714287,0.8095238,0.85714287,0.7619048,0.8095238,0.85714287,0.9047619,0.8095238,0.6666667,0.71428573,0.9047619,0.85714287,0.85714287,0.85714287,0.9047619,0.8095238,0.9047619,0.9047619,0.95238096,0.9047619,0.7619048,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0952381,0.2857143,0.2857143,0.33333334,0.33333334,0.47619048,0.52380955,0.42857143,0.47619048,0.47619048,0.42857143,,0.52380955,0.7619048,0.52380955,0.5714286,0.5714286,0.5952381,0.71428573,0.71428573,0.52380955,0.52380955,0.61904764,0.5714286,0.5714286,0.6666667,0.6666667,0.6666667,0.71428573,0.61904764,0.71428573,0.71428573,0.71428573,0.6666667,0.71428573,0.5714286,0.71428573,0.7619048,0.71428573,0.7619048,0.6666667,0.71428573,0.7619048,0.71428573,0.7619048,0.8095238,0.7619048,0.7619048,0.71428573,0.85714287,0.85714287,0.8095238,0.8095238,0.85714287,0.9047619,0.9047619,0.7619048,0.85714287,0.7619048,0.85714287,0.95238096,0.9047619,0.9047619,0.85714287,0.85714287,1.0,1.0,0.95238096,0.9047619,0.9047619,0.9047619,1.0,1.0
+RealMLP (tuned),0.14285715,0.2857143,0.1904762,0.3809524,0.2857143,0.33333334,0.23809524,0.33333334,0.3809524,0.26190478,0.33333334,0.47619048,,0.33333334,0.61904764,0.5714286,0.52380955,0.52380955,0.47619048,0.71428573,0.6666667,0.61904764,0.7619048,0.6666667,0.6666667,0.61904764,0.71428573,0.71428573,0.61904764,0.6666667,0.71428573,0.71428573,0.61904764,0.71428573,0.7619048,0.71428573,0.61904764,0.71428573,0.71428573,0.6666667,0.7619048,0.71428573,0.6666667,0.6666667,0.7619048,0.7619048,0.6666667,0.71428573,0.8095238,0.8095238,0.7619048,0.9047619,0.8095238,0.85714287,0.7380952,0.85714287,0.8095238,0.9047619,0.71428573,0.95238096,0.7619048,0.95238096,0.9047619,0.95238096,0.95238096,0.9047619,0.9047619,0.95238096,0.9047619,0.9285714,0.9047619,1.0,1.0
+TabDPT (tuned + ensembled),0.04761905,0.23809524,0.23809524,0.1904762,0.3809524,0.3809524,0.2857143,0.42857143,0.47619048,0.52380955,0.47619048,0.23809524,0.6666667,,0.47619048,0.52380955,0.61904764,0.52380955,0.6666667,0.5714286,0.52380955,0.61904764,0.52380955,0.52380955,0.6666667,0.61904764,0.61904764,0.61904764,0.71428573,0.5714286,0.6666667,0.6666667,0.61904764,0.6666667,0.61904764,0.61904764,0.71428573,0.71428573,0.6666667,0.6666667,0.61904764,0.61904764,0.71428573,0.7619048,0.61904764,0.6666667,0.7619048,0.71428573,0.6666667,0.8095238,0.7619048,0.8095238,0.7619048,0.85714287,0.95238096,0.7619048,0.7619048,0.7619048,0.7619048,0.8095238,0.9047619,0.8095238,0.95238096,0.8095238,0.9047619,1.0,0.9047619,1.0,0.9047619,0.9047619,0.85714287,0.95238096,1.0
+BetaTabPFN (default),0.04761905,0.04761905,0.1904762,0.0952381,0.14285715,0.1904762,0.23809524,0.1904762,0.33333334,0.33333334,0.33333334,0.47619048,0.3809524,0.52380955,,0.47619048,0.61904764,0.52380955,0.5714286,0.6666667,0.5714286,0.52380955,0.6666667,0.52380955,0.61904764,0.7619048,0.71428573,0.71428573,0.5714286,0.6666667,0.71428573,0.6666667,0.6666667,0.6666667,0.61904764,0.5714286,0.61904764,0.71428573,0.7619048,0.71428573,0.61904764,0.71428573,0.61904764,0.6666667,0.85714287,0.6666667,0.71428573,0.71428573,0.71428573,0.8095238,0.71428573,0.8095238,0.85714287,0.8095238,0.71428573,0.85714287,0.7619048,0.85714287,0.7619048,0.8095238,0.8095238,0.8095238,0.71428573,0.85714287,0.8095238,0.85714287,0.95238096,0.85714287,0.95238096,0.95238096,0.95238096,1.0,1.0
+TabM (tuned + ensembled),0.0952381,0.14285715,0.1904762,0.23809524,0.2857143,0.23809524,0.23809524,0.1904762,0.33333334,0.33333334,0.33333334,0.42857143,0.42857143,0.47619048,0.52380955,,0.5714286,0.5714286,0.42857143,0.61904764,0.5714286,0.52380955,0.61904764,0.52380955,0.61904764,0.61904764,0.61904764,0.6666667,0.61904764,0.6666667,0.71428573,0.7619048,0.61904764,0.8095238,0.61904764,0.6666667,0.6666667,0.6666667,0.5714286,0.71428573,0.6666667,0.71428573,0.5714286,0.7619048,0.7619048,0.7619048,0.7619048,0.7619048,0.71428573,0.71428573,0.7619048,0.7619048,0.7619048,0.6666667,0.7619048,0.8095238,0.85714287,0.7619048,0.7619048,0.7619048,0.85714287,0.9047619,0.7619048,0.95238096,0.7619048,0.9047619,0.85714287,0.85714287,0.9047619,0.9047619,0.9047619,1.0,1.0
+iLTM (tuned + ensembled),0.14285715,0.23809524,0.33333334,0.23809524,0.14285715,0.23809524,0.23809524,0.33333334,0.23809524,0.26190478,0.23809524,0.42857143,0.47619048,0.3809524,0.3809524,0.42857143,,0.52380955,0.3809524,0.61904764,0.47619048,0.5714286,0.47619048,0.5714286,0.47619048,0.6666667,0.5714286,0.61904764,0.71428573,0.5714286,0.6666667,0.5714286,0.61904764,0.6666667,0.61904764,0.5714286,0.6666667,0.6666667,0.6666667,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.8095238,0.61904764,0.61904764,0.8095238,0.8095238,0.8095238,0.7619048,0.8095238,0.71428573,0.7619048,0.9047619,0.9047619,0.9047619,0.71428573,0.9047619,0.85714287,0.9047619,0.9047619,0.95238096,0.95238096,0.9047619,0.95238096,0.95238096,0.9047619,0.9047619,0.9047619,0.95238096,1.0
+TabSwift (default),0.0952381,0.14285715,0.1904762,0.1904762,0.2857143,0.2857143,0.33333334,0.3809524,0.42857143,0.42857143,0.42857143,0.4047619,0.47619048,0.47619048,0.47619048,0.42857143,0.47619048,,0.3809524,0.52380955,0.47619048,0.5714286,0.5714286,0.5714286,0.61904764,0.42857143,0.5714286,0.6666667,0.61904764,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.61904764,0.71428573,0.6666667,0.5714286,0.5714286,0.52380955,0.7619048,0.71428573,0.6666667,0.61904764,0.85714287,0.7619048,0.7619048,0.6666667,0.8095238,0.7619048,0.71428573,0.71428573,0.8095238,0.85714287,0.71428573,0.71428573,0.71428573,0.8095238,0.7619048,0.95238096,0.8095238,0.85714287,0.85714287,0.7619048,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.9047619,1.0,1.0
+TabDPT (tuned),0.0,0.04761905,0.23809524,0.1904762,0.2857143,0.2857143,0.2857143,0.2857143,0.3809524,0.3809524,0.3809524,0.2857143,0.52380955,0.33333334,0.42857143,0.5714286,0.61904764,0.61904764,,0.52380955,0.52380955,0.5714286,0.52380955,0.52380955,0.61904764,0.61904764,0.5714286,0.61904764,0.6666667,0.5714286,0.61904764,0.61904764,0.61904764,0.6666667,0.61904764,0.5714286,0.5714286,0.71428573,0.52380955,0.7619048,0.61904764,0.71428573,0.61904764,0.8095238,0.6666667,0.61904764,0.71428573,0.71428573,0.6666667,0.7619048,0.7619048,0.8095238,0.71428573,0.8095238,0.9047619,0.6666667,0.7619048,0.71428573,0.71428573,0.7619048,0.8095238,0.85714287,0.85714287,0.8095238,0.85714287,1.0,0.85714287,0.9047619,0.85714287,0.9047619,0.85714287,0.95238096,0.95238096
+LightGBM (tuned + ensembled),0.04761905,0.04761905,0.0952381,0.1904762,0.0952381,0.0952381,0.14285715,0.1904762,0.1904762,0.1904762,0.23809524,0.2857143,0.2857143,0.42857143,0.33333334,0.3809524,0.3809524,0.47619048,0.47619048,,0.42857143,0.52380955,0.61904764,0.52380955,0.5714286,0.47619048,0.52380955,0.61904764,0.61904764,0.61904764,0.71428573,0.5714286,0.6666667,0.5714286,0.7619048,0.5714286,0.5714286,0.71428573,0.6666667,0.5714286,0.71428573,0.7619048,0.71428573,0.6666667,0.7619048,0.7619048,0.71428573,0.71428573,0.7619048,0.8095238,0.71428573,0.8095238,0.85714287,0.71428573,0.71428573,0.95238096,0.8095238,0.8095238,0.71428573,0.9047619,0.7619048,0.8095238,0.8095238,0.8095238,0.95238096,0.85714287,1.0,0.85714287,0.95238096,0.95238096,0.9047619,1.0,1.0
+TabM (tuned),0.04761905,0.0952381,0.14285715,0.1904762,0.23809524,0.1904762,0.2857143,0.14285715,0.2857143,0.2857143,0.23809524,0.47619048,0.33333334,0.47619048,0.42857143,0.42857143,0.52380955,0.52380955,0.47619048,0.5714286,,0.52380955,0.61904764,0.52380955,0.61904764,0.6904762,0.61904764,0.52380955,0.61904764,0.5714286,0.6666667,0.5714286,0.52380955,0.71428573,0.5714286,0.61904764,0.5714286,0.5714286,0.52380955,0.71428573,0.61904764,0.71428573,0.5714286,0.7619048,0.71428573,0.6666667,0.6666667,0.6666667,0.8095238,0.71428573,0.7619048,0.6666667,0.7619048,0.6666667,0.71428573,0.7619048,0.85714287,0.7619048,0.71428573,0.7619048,0.85714287,0.9047619,0.71428573,0.95238096,0.7619048,0.8095238,0.85714287,0.8095238,0.85714287,0.85714287,0.9047619,0.95238096,0.95238096
+CatBoost (tuned),0.04761905,0.1904762,0.1904762,0.1904762,0.23809524,0.14285715,0.1904762,0.1904762,0.23809524,0.33333334,0.2857143,0.47619048,0.3809524,0.3809524,0.47619048,0.47619048,0.42857143,0.42857143,0.42857143,0.47619048,0.47619048,,0.61904764,0.47619048,0.47619048,0.42857143,0.5714286,0.61904764,0.61904764,0.5714286,0.61904764,0.52380955,0.52380955,0.5714286,0.52380955,0.52380955,0.6666667,0.6666667,0.7619048,0.61904764,0.6666667,0.61904764,0.6666667,0.61904764,0.71428573,0.6666667,0.6666667,0.6666667,0.71428573,0.71428573,0.7619048,0.7619048,0.7619048,0.6666667,0.6666667,0.8095238,0.85714287,0.8095238,0.6666667,0.85714287,0.7619048,0.9047619,0.7619048,0.9047619,0.85714287,0.85714287,0.9047619,0.85714287,0.95238096,0.95238096,0.95238096,1.0,1.0
+CatBoost (default),0.0952381,0.04761905,0.0952381,0.14285715,0.0952381,0.0952381,0.14285715,0.0952381,0.14285715,0.33333334,0.2857143,0.3809524,0.23809524,0.47619048,0.33333334,0.3809524,0.52380955,0.42857143,0.47619048,0.3809524,0.3809524,0.3809524,,0.42857143,0.42857143,0.42857143,0.5714286,0.61904764,0.5714286,0.71428573,0.5714286,0.5714286,0.61904764,0.61904764,0.52380955,0.52380955,0.5714286,0.6666667,0.6666667,0.61904764,0.6666667,0.7619048,0.6666667,0.5714286,0.7619048,0.7619048,0.71428573,0.71428573,0.7619048,0.7619048,0.7619048,0.71428573,0.8095238,0.6666667,0.71428573,0.7619048,0.85714287,0.85714287,0.71428573,0.85714287,0.7619048,0.9047619,0.7619048,0.9047619,0.85714287,0.85714287,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,1.0,1.0
+ModernNCA (tuned),0.04761905,0.14285715,0.14285715,0.1904762,0.1904762,0.14285715,0.0952381,0.14285715,0.33333334,0.23809524,0.3809524,0.42857143,0.33333334,0.47619048,0.47619048,0.47619048,0.42857143,0.42857143,0.47619048,0.47619048,0.47619048,0.52380955,0.5714286,,0.52380955,0.52380955,0.5714286,0.52380955,0.52380955,0.61904764,0.52380955,0.61904764,0.61904764,0.61904764,0.5714286,0.52380955,0.61904764,0.61904764,0.61904764,0.5714286,0.6666667,0.6666667,0.61904764,0.71428573,0.8095238,0.64285713,0.71428573,0.71428573,0.71428573,0.61904764,0.71428573,0.61904764,0.8095238,0.6666667,0.71428573,0.8095238,0.7619048,0.71428573,0.8095238,0.7619048,0.7619048,0.7619048,0.7619048,0.8095238,0.85714287,0.8095238,0.9047619,0.9047619,0.85714287,0.9047619,0.85714287,0.9047619,1.0
+CatBoost (tuned + ensembled),0.0,0.0952381,0.14285715,0.1904762,0.14285715,0.0952381,0.0952381,0.1904762,0.0952381,0.2857143,0.23809524,0.42857143,0.33333334,0.33333334,0.3809524,0.3809524,0.52380955,0.3809524,0.3809524,0.42857143,0.3809524,0.52380955,0.5714286,0.47619048,,0.47619048,0.52380955,0.5714286,0.61904764,0.61904764,0.5714286,0.42857143,0.52380955,0.52380955,0.5714286,0.52380955,0.71428573,0.6666667,0.6666667,0.52380955,0.61904764,0.8095238,0.61904764,0.6666667,0.71428573,0.6666667,0.6666667,0.6666667,0.8095238,0.7619048,0.7619048,0.7619048,0.7619048,0.7619048,0.61904764,0.8095238,0.8095238,0.85714287,0.6666667,0.9047619,0.71428573,0.9047619,0.8095238,0.9047619,0.9047619,0.8095238,0.85714287,0.85714287,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+ChimeraBoost (tuned + ensembled),0.04761905,0.0952381,0.14285715,0.14285715,0.1904762,0.14285715,0.14285715,0.23809524,0.23809524,0.2857143,0.33333334,0.33333334,0.3809524,0.3809524,0.23809524,0.3809524,0.33333334,0.5714286,0.3809524,0.52380955,0.30952382,0.5714286,0.5714286,0.47619048,0.52380955,,0.52380955,0.61904764,0.5714286,0.5714286,0.5714286,0.5714286,0.61904764,0.61904764,0.61904764,0.5714286,0.52380955,0.5714286,0.61904764,0.5714286,0.61904764,0.8095238,0.61904764,0.61904764,0.6666667,0.71428573,0.71428573,0.71428573,0.71428573,0.6666667,0.6666667,0.61904764,0.7619048,0.6666667,0.71428573,0.8095238,0.8095238,0.8095238,0.71428573,0.8095238,0.8095238,0.7619048,0.7619048,0.85714287,0.8095238,0.7619048,0.85714287,0.8095238,0.9047619,0.9047619,0.95238096,0.95238096,1.0
+XGBoost (tuned + ensembled),0.04761905,0.0952381,0.0952381,0.1904762,0.1904762,0.0952381,0.14285715,0.14285715,0.1904762,0.23809524,0.23809524,0.33333334,0.2857143,0.3809524,0.2857143,0.3809524,0.42857143,0.42857143,0.42857143,0.47619048,0.3809524,0.42857143,0.42857143,0.42857143,0.47619048,0.47619048,,0.52380955,0.5714286,0.61904764,0.6666667,0.42857143,0.52380955,0.5714286,0.5714286,0.47619048,0.52380955,0.5714286,0.61904764,0.5714286,0.5714286,0.6666667,0.6666667,0.61904764,0.5952381,0.71428573,0.71428573,0.71428573,0.61904764,0.8095238,0.6666667,0.7619048,0.8095238,0.7619048,0.71428573,1.0,0.71428573,0.9047619,0.71428573,0.95238096,0.71428573,0.7619048,0.85714287,0.85714287,0.9047619,0.8095238,0.85714287,0.85714287,0.85714287,0.9047619,0.9047619,0.9047619,1.0
+EBM (tuned + ensembled),0.0952381,0.1904762,0.0952381,0.1904762,0.14285715,0.1904762,0.23809524,0.1904762,0.1904762,0.1904762,0.33333334,0.33333334,0.2857143,0.3809524,0.2857143,0.33333334,0.3809524,0.33333334,0.3809524,0.3809524,0.47619048,0.3809524,0.3809524,0.47619048,0.42857143,0.3809524,0.47619048,,0.42857143,0.61904764,0.52380955,0.52380955,0.52380955,0.42857143,0.42857143,0.47619048,0.5714286,0.7619048,0.6666667,0.6666667,0.6666667,0.52380955,0.61904764,0.71428573,0.6666667,0.7619048,0.6666667,0.6666667,0.61904764,0.8095238,0.7619048,0.61904764,0.7619048,0.71428573,0.71428573,0.71428573,0.6666667,0.7619048,0.71428573,0.7619048,0.85714287,0.8095238,0.8095238,0.8095238,0.7619048,0.85714287,0.95238096,0.85714287,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+iLTM (tuned),0.04761905,0.0952381,0.023809524,0.23809524,0.14285715,0.1904762,0.14285715,0.14285715,0.23809524,0.0952381,0.23809524,0.2857143,0.3809524,0.2857143,0.42857143,0.3809524,0.2857143,0.3809524,0.33333334,0.3809524,0.3809524,0.3809524,0.42857143,0.47619048,0.3809524,0.42857143,0.42857143,0.5714286,,0.5714286,0.47619048,0.52380955,0.5714286,0.6666667,0.52380955,0.5714286,0.61904764,0.7619048,0.6666667,0.52380955,0.5714286,0.47619048,0.71428573,0.61904764,0.6666667,0.71428573,0.6666667,0.6666667,0.5714286,0.85714287,0.7619048,0.5952381,0.61904764,0.8095238,0.61904764,0.71428573,0.7619048,0.9047619,0.71428573,0.8095238,0.7619048,0.8095238,0.8095238,0.9047619,0.85714287,0.9047619,0.9047619,0.95238096,0.85714287,0.85714287,0.8095238,1.0,1.0
+RealMLP (default),0.14285715,0.14285715,0.1904762,0.1904762,0.1904762,0.23809524,0.23809524,0.23809524,0.1904762,0.23809524,0.33333334,0.3809524,0.33333334,0.42857143,0.33333334,0.33333334,0.42857143,0.42857143,0.42857143,0.3809524,0.42857143,0.42857143,0.2857143,0.3809524,0.3809524,0.42857143,0.3809524,0.3809524,0.42857143,,0.47619048,0.52380955,0.42857143,0.42857143,0.47619048,0.3809524,0.47619048,0.52380955,0.5714286,0.6666667,0.71428573,0.52380955,0.61904764,0.71428573,0.71428573,0.6666667,0.61904764,0.6666667,0.61904764,0.6666667,0.71428573,0.71428573,0.7619048,0.71428573,0.6666667,0.6666667,0.6666667,0.7619048,0.71428573,0.7619048,0.9047619,0.9047619,0.8095238,0.95238096,0.8095238,0.95238096,0.95238096,0.85714287,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+XGBoost (tuned),0.0952381,0.14285715,0.0952381,0.1904762,0.14285715,0.14285715,0.14285715,0.1904762,0.14285715,0.1904762,0.23809524,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.42857143,0.3809524,0.2857143,0.33333334,0.3809524,0.42857143,0.47619048,0.42857143,0.42857143,0.33333334,0.47619048,0.52380955,0.52380955,,0.42857143,0.61904764,0.47619048,0.47619048,0.42857143,0.52380955,0.5714286,0.61904764,0.52380955,0.47619048,0.5714286,0.6666667,0.52380955,0.5714286,0.71428573,0.71428573,0.71428573,0.61904764,0.85714287,0.71428573,0.71428573,0.7619048,0.7619048,0.6666667,0.9047619,0.71428573,0.85714287,0.71428573,0.9047619,0.71428573,0.7619048,0.8095238,0.8095238,0.85714287,0.8095238,0.9047619,0.85714287,0.8095238,0.85714287,0.9047619,0.9047619,1.0
+TabM (default),0.0952381,0.1904762,0.0952381,0.1904762,0.1904762,0.14285715,0.1904762,0.14285715,0.33333334,0.23809524,0.2857143,0.2857143,0.2857143,0.33333334,0.33333334,0.23809524,0.42857143,0.42857143,0.3809524,0.42857143,0.42857143,0.47619048,0.42857143,0.3809524,0.5714286,0.42857143,0.5714286,0.47619048,0.47619048,0.47619048,0.5714286,,0.42857143,0.5714286,0.42857143,0.5714286,0.52380955,0.4047619,0.47619048,0.5714286,0.52380955,0.61904764,0.5714286,0.61904764,0.71428573,0.71428573,0.61904764,0.61904764,0.71428573,0.61904764,0.7619048,0.5714286,0.6666667,0.6666667,0.6666667,0.6666667,0.71428573,0.7619048,0.71428573,0.7619048,0.7619048,0.8095238,0.71428573,0.9047619,0.7619048,0.7619048,0.8095238,0.7619048,0.9047619,0.9047619,0.85714287,0.95238096,0.9047619
+xRFM (tuned + ensembled),0.04761905,0.0952381,0.0952381,0.0952381,0.0952381,0.14285715,0.0952381,0.1904762,0.2857143,0.23809524,0.23809524,0.2857143,0.3809524,0.3809524,0.33333334,0.3809524,0.3809524,0.2857143,0.3809524,0.33333334,0.47619048,0.47619048,0.3809524,0.3809524,0.47619048,0.3809524,0.47619048,0.47619048,0.42857143,0.5714286,0.3809524,0.5714286,,0.52380955,0.3809524,0.52380955,0.52380955,0.52380955,0.42857143,0.52380955,0.52380955,0.47619048,0.71428573,0.71428573,0.61904764,0.6666667,0.61904764,0.61904764,0.61904764,0.6666667,0.71428573,0.6666667,0.71428573,0.71428573,0.7619048,0.71428573,0.71428573,0.7619048,0.6666667,0.7619048,0.8095238,0.8095238,0.85714287,0.8095238,0.85714287,0.9047619,0.85714287,0.85714287,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096
+ChimeraBoost (tuned),0.0952381,0.14285715,0.14285715,0.1904762,0.1904762,0.0952381,0.14285715,0.14285715,0.23809524,0.23809524,0.2857143,0.33333334,0.2857143,0.33333334,0.33333334,0.1904762,0.33333334,0.42857143,0.33333334,0.42857143,0.2857143,0.42857143,0.3809524,0.3809524,0.47619048,0.3809524,0.42857143,0.5714286,0.33333334,0.5714286,0.52380955,0.42857143,0.47619048,,0.52380955,0.52380955,0.52380955,0.5714286,0.5714286,0.5714286,0.52380955,0.6666667,0.61904764,0.5714286,0.5714286,0.71428573,0.6666667,0.6666667,0.6666667,0.61904764,0.71428573,0.6666667,0.7619048,0.71428573,0.6666667,0.8095238,0.7619048,0.7619048,0.71428573,0.8095238,0.8095238,0.7619048,0.8095238,0.85714287,0.71428573,0.7619048,0.8095238,0.8095238,0.85714287,0.85714287,0.9047619,0.9047619,1.0
+LightGBM (tuned),0.04761905,0.0,0.04761905,0.1904762,0.0,0.04761905,0.0952381,0.04761905,0.14285715,0.23809524,0.1904762,0.2857143,0.23809524,0.3809524,0.3809524,0.3809524,0.3809524,0.42857143,0.3809524,0.23809524,0.42857143,0.47619048,0.47619048,0.42857143,0.42857143,0.3809524,0.42857143,0.5714286,0.47619048,0.52380955,0.52380955,0.5714286,0.61904764,0.47619048,,0.3809524,0.52380955,0.5714286,0.52380955,0.52380955,0.5714286,0.6666667,0.61904764,0.47619048,0.7619048,0.6666667,0.6666667,0.71428573,0.71428573,0.71428573,0.6666667,0.6666667,0.71428573,0.71428573,0.6666667,0.7619048,0.7619048,0.7619048,0.71428573,0.85714287,0.71428573,0.8095238,0.7619048,0.7619048,0.85714287,0.8095238,0.9047619,0.85714287,0.85714287,0.85714287,0.85714287,0.9047619,1.0
+ModernNCA (tuned + ensembled),0.0952381,0.0952381,0.1904762,0.0952381,0.0952381,0.14285715,0.14285715,0.1904762,0.23809524,0.23809524,0.33333334,0.42857143,0.2857143,0.3809524,0.42857143,0.33333334,0.42857143,0.3809524,0.42857143,0.42857143,0.3809524,0.47619048,0.47619048,0.47619048,0.47619048,0.42857143,0.52380955,0.52380955,0.42857143,0.61904764,0.5714286,0.42857143,0.47619048,0.47619048,0.61904764,,0.47619048,0.52380955,0.52380955,0.5714286,0.52380955,0.47619048,0.5714286,0.6666667,0.71428573,0.61904764,0.5714286,0.61904764,0.6666667,0.6666667,0.71428573,0.6666667,0.7619048,0.61904764,0.6666667,0.8095238,0.6666667,0.71428573,0.71428573,0.6666667,0.6666667,0.8095238,0.71428573,0.7619048,0.7619048,0.7619048,0.78571427,0.9047619,0.71428573,0.7619048,0.71428573,0.8095238,0.95238096
+SAP-RPT-OSS (default),0.0952381,0.1904762,0.14285715,0.1904762,0.1904762,0.1904762,0.23809524,0.1904762,0.23809524,0.1904762,0.2857143,0.2857143,0.3809524,0.2857143,0.3809524,0.33333334,0.33333334,0.2857143,0.42857143,0.42857143,0.42857143,0.33333334,0.42857143,0.3809524,0.2857143,0.47619048,0.47619048,0.42857143,0.3809524,0.52380955,0.47619048,0.47619048,0.47619048,0.47619048,0.47619048,0.52380955,,0.5714286,0.47619048,0.5714286,0.61904764,0.47619048,0.52380955,0.5714286,0.6666667,0.7619048,0.5714286,0.5714286,0.71428573,0.6666667,0.71428573,0.6666667,0.61904764,0.71428573,0.6666667,0.6666667,0.7619048,0.6666667,0.71428573,0.71428573,0.8095238,0.8095238,0.8095238,0.8095238,0.8095238,0.8095238,0.7619048,0.85714287,0.9047619,0.85714287,0.8095238,0.9047619,1.0
+EBM (default),0.0952381,0.14285715,0.0952381,0.14285715,0.1904762,0.14285715,0.1904762,0.1904762,0.1904762,0.14285715,0.23809524,0.23809524,0.2857143,0.2857143,0.2857143,0.33333334,0.33333334,0.33333334,0.2857143,0.2857143,0.42857143,0.33333334,0.33333334,0.3809524,0.33333334,0.42857143,0.42857143,0.23809524,0.23809524,0.47619048,0.42857143,0.5952381,0.47619048,0.42857143,0.42857143,0.47619048,0.42857143,,0.47619048,0.47619048,0.61904764,0.47619048,0.5714286,0.61904764,0.71428573,0.6666667,0.6666667,0.6666667,0.71428573,0.7619048,0.6666667,0.61904764,0.5714286,0.71428573,0.71428573,0.61904764,0.7619048,0.71428573,0.6666667,0.7619048,0.8095238,0.85714287,0.8095238,0.85714287,0.7619048,0.85714287,0.95238096,0.8095238,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+EBM (tuned),0.0952381,0.0952381,0.04761905,0.14285715,0.0952381,0.1904762,0.23809524,0.1904762,0.1904762,0.0952381,0.14285715,0.2857143,0.2857143,0.33333334,0.23809524,0.42857143,0.33333334,0.42857143,0.47619048,0.33333334,0.47619048,0.23809524,0.33333334,0.3809524,0.33333334,0.3809524,0.3809524,0.33333334,0.33333334,0.42857143,0.3809524,0.52380955,0.5714286,0.42857143,0.47619048,0.47619048,0.52380955,0.52380955,,0.61904764,0.61904764,0.42857143,0.47619048,0.61904764,0.61904764,0.61904764,0.6666667,0.6666667,0.61904764,0.6666667,0.6666667,0.61904764,0.6666667,0.71428573,0.6666667,0.6666667,0.61904764,0.71428573,0.71428573,0.71428573,0.8095238,0.7619048,0.8095238,0.8095238,0.8095238,0.8095238,0.95238096,0.9047619,0.9047619,0.9047619,0.85714287,0.95238096,1.0
+TabDPT (default),0.0,0.0952381,0.14285715,0.14285715,0.2857143,0.2857143,0.23809524,0.2857143,0.23809524,0.33333334,0.1904762,0.23809524,0.33333334,0.33333334,0.2857143,0.2857143,0.42857143,0.42857143,0.23809524,0.42857143,0.2857143,0.3809524,0.3809524,0.42857143,0.47619048,0.42857143,0.42857143,0.33333334,0.47619048,0.33333334,0.47619048,0.42857143,0.47619048,0.42857143,0.47619048,0.42857143,0.42857143,0.52380955,0.3809524,,0.47619048,0.52380955,0.5714286,0.6666667,0.47619048,0.5714286,0.5714286,0.61904764,0.61904764,0.61904764,0.7619048,0.61904764,0.61904764,0.71428573,0.8095238,0.6666667,0.71428573,0.6666667,0.6666667,0.71428573,0.7619048,0.71428573,0.8095238,0.7619048,0.7619048,0.85714287,0.85714287,0.8095238,0.8095238,0.8095238,0.8095238,0.95238096,0.95238096
+TorchMLP (tuned + ensembled),0.0,0.04761905,0.04761905,0.0952381,0.04761905,0.14285715,0.14285715,0.14285715,0.1904762,0.1904762,0.14285715,0.33333334,0.23809524,0.3809524,0.3809524,0.33333334,0.2857143,0.47619048,0.3809524,0.2857143,0.3809524,0.33333334,0.33333334,0.33333334,0.3809524,0.3809524,0.42857143,0.33333334,0.42857143,0.2857143,0.52380955,0.47619048,0.47619048,0.47619048,0.42857143,0.47619048,0.3809524,0.3809524,0.3809524,0.52380955,,0.47619048,0.52380955,0.52380955,0.6666667,0.61904764,0.5714286,0.5714286,0.61904764,0.61904764,0.6666667,0.6666667,0.71428573,0.61904764,0.6666667,0.7619048,0.71428573,0.71428573,0.5714286,0.8095238,0.85714287,0.8095238,0.71428573,0.9047619,0.71428573,0.85714287,0.85714287,0.8095238,0.85714287,0.85714287,0.95238096,0.95238096,0.95238096
+ChimeraBoost (default),0.04761905,0.04761905,0.04761905,0.14285715,0.0952381,0.14285715,0.0952381,0.14285715,0.14285715,0.1904762,0.23809524,0.2857143,0.2857143,0.3809524,0.2857143,0.2857143,0.2857143,0.23809524,0.2857143,0.23809524,0.2857143,0.3809524,0.23809524,0.33333334,0.1904762,0.1904762,0.33333334,0.47619048,0.52380955,0.47619048,0.42857143,0.3809524,0.52380955,0.33333334,0.33333334,0.52380955,0.52380955,0.52380955,0.5714286,0.47619048,0.52380955,,0.5714286,0.5714286,0.47619048,0.61904764,0.71428573,0.71428573,0.5714286,0.61904764,0.6666667,0.61904764,0.6666667,0.61904764,0.6666667,0.6666667,0.6666667,0.71428573,0.71428573,0.8095238,0.7619048,0.71428573,0.7619048,0.8095238,0.71428573,0.7619048,0.8095238,0.8095238,0.85714287,0.85714287,0.95238096,0.9047619,1.0
+xRFM (tuned),0.04761905,0.04761905,0.0,0.04761905,0.0952381,0.04761905,0.1904762,0.0952381,0.23809524,0.14285715,0.1904762,0.23809524,0.33333334,0.2857143,0.3809524,0.42857143,0.2857143,0.2857143,0.3809524,0.2857143,0.42857143,0.33333334,0.33333334,0.3809524,0.3809524,0.3809524,0.33333334,0.3809524,0.2857143,0.3809524,0.33333334,0.42857143,0.2857143,0.3809524,0.3809524,0.42857143,0.47619048,0.42857143,0.52380955,0.42857143,0.47619048,0.42857143,,0.52380955,0.61904764,0.61904764,0.5714286,0.61904764,0.47619048,0.71428573,0.61904764,0.71428573,0.5714286,0.71428573,0.6666667,0.6666667,0.61904764,0.6666667,0.61904764,0.71428573,0.71428573,0.85714287,0.85714287,0.8095238,0.8095238,0.85714287,0.85714287,0.95238096,0.85714287,0.85714287,0.8095238,0.95238096,1.0
+ModernNCA (default),0.0952381,0.0952381,0.0952381,0.14285715,0.0952381,0.1904762,0.0952381,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.33333334,0.23809524,0.33333334,0.23809524,0.2857143,0.33333334,0.1904762,0.33333334,0.23809524,0.3809524,0.42857143,0.2857143,0.33333334,0.3809524,0.3809524,0.2857143,0.3809524,0.2857143,0.47619048,0.3809524,0.2857143,0.42857143,0.52380955,0.33333334,0.42857143,0.3809524,0.3809524,0.33333334,0.47619048,0.42857143,0.47619048,,0.47619048,0.47619048,0.7619048,0.7619048,0.5714286,0.5714286,0.61904764,0.6666667,0.52380955,0.52380955,0.61904764,0.5714286,0.61904764,0.71428573,0.7619048,0.71428573,0.6666667,0.71428573,0.7619048,0.8095238,0.7619048,0.7619048,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.95238096,0.95238096
+TorchMLP (tuned),0.0,0.04761905,0.0952381,0.04761905,0.04761905,0.14285715,0.14285715,0.0952381,0.23809524,0.23809524,0.0952381,0.23809524,0.23809524,0.3809524,0.14285715,0.23809524,0.2857143,0.3809524,0.33333334,0.23809524,0.2857143,0.2857143,0.23809524,0.1904762,0.2857143,0.33333334,0.4047619,0.33333334,0.33333334,0.2857143,0.42857143,0.2857143,0.3809524,0.42857143,0.23809524,0.2857143,0.33333334,0.2857143,0.3809524,0.52380955,0.33333334,0.52380955,0.3809524,0.52380955,,0.47619048,0.6666667,0.61904764,0.61904764,0.54761904,0.5714286,0.5714286,0.52380955,0.5714286,0.61904764,0.54761904,0.61904764,0.6666667,0.6666667,0.6666667,0.85714287,0.7619048,0.71428573,0.9047619,0.6666667,0.7619048,0.85714287,0.7619048,0.8095238,0.8095238,0.85714287,0.9047619,0.95238096
+FastaiMLP (tuned + ensembled),0.0952381,0.14285715,0.04761905,0.0,0.14285715,0.1904762,0.1904762,0.1904762,0.14285715,0.23809524,0.1904762,0.1904762,0.23809524,0.33333334,0.33333334,0.23809524,0.1904762,0.14285715,0.3809524,0.23809524,0.33333334,0.33333334,0.23809524,0.35714287,0.33333334,0.2857143,0.2857143,0.23809524,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.2857143,0.33333334,0.3809524,0.23809524,0.33333334,0.3809524,0.42857143,0.3809524,0.3809524,0.3809524,0.52380955,0.52380955,,0.52380955,0.52380955,0.6666667,0.5714286,0.71428573,0.61904764,0.47619048,0.5714286,0.6666667,0.47619048,0.6666667,0.52380955,0.61904764,0.61904764,0.8095238,0.85714287,0.7619048,0.7619048,0.71428573,0.95238096,0.85714287,0.8095238,0.7619048,0.8095238,0.8095238,0.9047619,0.95238096
+TabSTAR (tuned),0.0952381,0.23809524,0.1904762,0.1904762,0.2857143,0.2857143,0.23809524,0.2857143,0.33333334,0.21428572,0.33333334,0.23809524,0.33333334,0.23809524,0.2857143,0.23809524,0.3809524,0.23809524,0.2857143,0.2857143,0.33333334,0.33333334,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.33333334,0.3809524,0.2857143,0.3809524,0.3809524,0.33333334,0.33333334,0.42857143,0.42857143,0.33333334,0.33333334,0.42857143,0.42857143,0.2857143,0.42857143,0.23809524,0.33333334,0.47619048,,0.61904764,0.52380955,0.47619048,0.47619048,0.42857143,0.47619048,0.42857143,0.47619048,0.47619048,0.52380955,0.5714286,0.9047619,0.5714286,0.5714286,0.5714286,0.71428573,0.61904764,0.7619048,0.6666667,0.6666667,0.8095238,0.6666667,0.71428573,0.71428573,0.8095238,0.9047619
+TabSTAR (tuned + ensembled),0.14285715,0.23809524,0.1904762,0.1904762,0.2857143,0.2857143,0.23809524,0.2857143,0.33333334,0.23809524,0.2857143,0.23809524,0.2857143,0.2857143,0.2857143,0.23809524,0.3809524,0.23809524,0.2857143,0.2857143,0.33333334,0.33333334,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.33333334,0.33333334,0.2857143,0.3809524,0.3809524,0.33333334,0.2857143,0.3809524,0.42857143,0.33333334,0.33333334,0.3809524,0.42857143,0.2857143,0.3809524,0.23809524,0.3809524,0.47619048,0.3809524,,0.52380955,0.42857143,0.47619048,0.42857143,0.47619048,0.42857143,0.47619048,0.47619048,0.52380955,0.52380955,0.95238096,0.5714286,0.52380955,0.5714286,0.6666667,0.6666667,0.71428573,0.6666667,0.61904764,0.7619048,0.6666667,0.6666667,0.71428573,0.8095238,0.9047619
+PerpetualBooster (tuned + ensembled),0.0,0.04761905,0.0952381,0.0952381,0.04761905,0.0952381,0.14285715,0.0952381,0.0952381,0.1904762,0.0952381,0.2857143,0.1904762,0.33333334,0.2857143,0.2857143,0.1904762,0.33333334,0.33333334,0.23809524,0.1904762,0.2857143,0.23809524,0.2857143,0.1904762,0.2857143,0.3809524,0.3809524,0.42857143,0.3809524,0.3809524,0.2857143,0.3809524,0.33333334,0.2857143,0.33333334,0.2857143,0.2857143,0.3809524,0.3809524,0.3809524,0.42857143,0.52380955,0.42857143,0.3809524,0.33333334,0.47619048,0.47619048,,0.47619048,0.5714286,0.5714286,0.52380955,0.47619048,0.47619048,0.52380955,0.71428573,0.61904764,0.47619048,0.61904764,0.61904764,0.7619048,0.61904764,0.61904764,0.7619048,0.6666667,0.7619048,0.71428573,0.8095238,0.8095238,0.8095238,0.7619048,0.9047619
+ExtraTrees (tuned + ensembled),0.04761905,0.0,0.0952381,0.14285715,0.0952381,0.04761905,0.0952381,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.1904762,0.1904762,0.1904762,0.2857143,0.1904762,0.1904762,0.23809524,0.1904762,0.2857143,0.2857143,0.23809524,0.3809524,0.23809524,0.33333334,0.1904762,0.1904762,0.14285715,0.33333334,0.14285715,0.3809524,0.33333334,0.3809524,0.2857143,0.33333334,0.33333334,0.23809524,0.33333334,0.3809524,0.3809524,0.3809524,0.2857143,0.42857143,0.45238096,0.42857143,0.52380955,0.5714286,0.52380955,,0.3809524,0.42857143,0.42857143,0.6666667,0.52380955,0.47619048,0.5714286,0.5714286,0.6666667,0.71428573,0.6666667,0.61904764,0.85714287,0.5714286,0.8095238,0.71428573,0.9047619,0.71428573,0.8095238,0.85714287,0.7619048,0.9047619,1.0
+FastaiMLP (tuned),0.0952381,0.0952381,0.0,0.0952381,0.1904762,0.14285715,0.14285715,0.1904762,0.1904762,0.1904762,0.14285715,0.14285715,0.23809524,0.23809524,0.2857143,0.23809524,0.1904762,0.23809524,0.23809524,0.2857143,0.23809524,0.23809524,0.23809524,0.2857143,0.23809524,0.33333334,0.33333334,0.23809524,0.23809524,0.2857143,0.2857143,0.23809524,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.33333334,0.23809524,0.33333334,0.33333334,0.3809524,0.3809524,0.42857143,0.2857143,0.52380955,0.52380955,0.42857143,0.61904764,,0.52380955,0.42857143,0.5714286,0.52380955,0.47619048,0.52380955,0.5714286,0.5714286,0.61904764,0.71428573,0.6666667,0.71428573,0.6666667,0.6666667,0.71428573,0.8095238,0.71428573,0.7619048,0.7619048,0.7619048,0.85714287,0.9047619
+iLTM (default),0.0952381,0.04761905,0.04761905,0.1904762,0.0952381,0.14285715,0.0952381,0.14285715,0.0952381,0.14285715,0.14285715,0.1904762,0.0952381,0.1904762,0.1904762,0.23809524,0.23809524,0.2857143,0.1904762,0.1904762,0.33333334,0.23809524,0.2857143,0.3809524,0.23809524,0.3809524,0.23809524,0.3809524,0.4047619,0.2857143,0.2857143,0.42857143,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.3809524,0.3809524,0.3809524,0.33333334,0.3809524,0.2857143,0.33333334,0.42857143,0.3809524,0.5714286,0.5714286,0.42857143,0.5714286,0.47619048,,0.42857143,0.5714286,0.52380955,0.3809524,0.3809524,0.6666667,0.61904764,0.5714286,0.5714286,0.47619048,0.61904764,0.6666667,0.61904764,0.61904764,0.8095238,0.71428573,0.71428573,0.71428573,0.8095238,0.85714287,0.9047619
+XGBoost (default),0.0,0.04761905,0.04761905,0.04761905,0.04761905,0.0,0.0952381,0.04761905,0.0952381,0.14285715,0.0952381,0.1904762,0.1904762,0.23809524,0.14285715,0.23809524,0.1904762,0.2857143,0.2857143,0.14285715,0.23809524,0.23809524,0.1904762,0.1904762,0.23809524,0.23809524,0.1904762,0.23809524,0.3809524,0.23809524,0.23809524,0.33333334,0.2857143,0.23809524,0.2857143,0.23809524,0.3809524,0.42857143,0.33333334,0.3809524,0.2857143,0.33333334,0.42857143,0.47619048,0.47619048,0.52380955,0.52380955,0.52380955,0.47619048,0.5714286,0.5714286,0.5714286,,0.5714286,0.47619048,0.61904764,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.6666667,0.61904764,0.71428573,0.71428573,0.71428573,0.85714287,0.7619048,0.8095238,0.7619048,0.8095238,0.9047619
+ExtraTrees (tuned),0.04761905,0.0,0.0952381,0.0952381,0.0952381,0.04761905,0.0952381,0.04761905,0.1904762,0.1904762,0.1904762,0.14285715,0.14285715,0.14285715,0.1904762,0.33333334,0.2857143,0.1904762,0.1904762,0.2857143,0.33333334,0.33333334,0.33333334,0.33333334,0.23809524,0.33333334,0.23809524,0.2857143,0.1904762,0.2857143,0.23809524,0.33333334,0.2857143,0.2857143,0.2857143,0.3809524,0.2857143,0.2857143,0.2857143,0.2857143,0.3809524,0.3809524,0.2857143,0.47619048,0.42857143,0.42857143,0.5714286,0.5714286,0.52380955,0.33333334,0.42857143,0.42857143,0.42857143,,0.52380955,0.3809524,0.47619048,0.47619048,0.71428573,0.6666667,0.61904764,0.5714286,0.7619048,0.52380955,0.7619048,0.6666667,0.85714287,0.6666667,0.71428573,0.71428573,0.7619048,0.95238096,0.95238096
+OrionMSP (default),0.0,0.0,0.0,0.0952381,0.04761905,0.04761905,0.0,0.0,0.0952381,0.11904762,0.0952381,0.0952381,0.26190478,0.04761905,0.2857143,0.23809524,0.23809524,0.14285715,0.0952381,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.3809524,0.2857143,0.2857143,0.2857143,0.3809524,0.33333334,0.33333334,0.33333334,0.23809524,0.33333334,0.33333334,0.33333334,0.33333334,0.2857143,0.33333334,0.1904762,0.33333334,0.33333334,0.33333334,0.3809524,0.3809524,0.33333334,0.52380955,0.52380955,0.52380955,0.47619048,0.47619048,0.47619048,0.52380955,0.47619048,,0.5714286,0.52380955,0.5714286,0.6666667,0.5714286,0.71428573,0.6666667,0.71428573,0.61904764,0.6666667,0.5714286,0.6666667,0.71428573,0.8095238,0.8095238,0.8095238,0.8095238,0.95238096
+LightGBM (default),0.04761905,0.04761905,0.0,0.04761905,0.04761905,0.0,0.04761905,0.04761905,0.0952381,0.14285715,0.0952381,0.0952381,0.14285715,0.23809524,0.14285715,0.1904762,0.0952381,0.2857143,0.33333334,0.04761905,0.23809524,0.1904762,0.23809524,0.1904762,0.1904762,0.1904762,0.0,0.2857143,0.2857143,0.33333334,0.0952381,0.33333334,0.2857143,0.1904762,0.23809524,0.1904762,0.33333334,0.3809524,0.33333334,0.33333334,0.23809524,0.33333334,0.33333334,0.42857143,0.45238096,0.52380955,0.52380955,0.52380955,0.47619048,0.52380955,0.52380955,0.61904764,0.3809524,0.61904764,0.42857143,,0.47619048,0.52380955,0.5714286,0.5714286,0.61904764,0.61904764,0.71428573,0.61904764,0.71428573,0.6666667,0.6666667,0.7619048,0.71428573,0.71428573,0.8095238,0.85714287,0.9047619
+PerpetualBooster (tuned),0.0,0.0,0.0,0.04761905,0.0,0.04761905,0.0952381,0.04761905,0.04761905,0.1904762,0.04761905,0.23809524,0.1904762,0.23809524,0.23809524,0.14285715,0.0952381,0.2857143,0.23809524,0.1904762,0.14285715,0.14285715,0.14285715,0.23809524,0.1904762,0.1904762,0.2857143,0.33333334,0.23809524,0.33333334,0.2857143,0.2857143,0.2857143,0.23809524,0.23809524,0.33333334,0.23809524,0.23809524,0.3809524,0.2857143,0.2857143,0.33333334,0.3809524,0.3809524,0.3809524,0.33333334,0.47619048,0.47619048,0.2857143,0.42857143,0.47619048,0.61904764,0.42857143,0.52380955,0.47619048,0.52380955,,0.47619048,0.47619048,0.5714286,0.6666667,0.7619048,0.61904764,0.61904764,0.6666667,0.61904764,0.6666667,0.71428573,0.7619048,0.71428573,0.7619048,0.7619048,0.9047619
+RandomForest (tuned + ensembled),0.04761905,0.04761905,0.04761905,0.0952381,0.0,0.04761905,0.0952381,0.04761905,0.1904762,0.14285715,0.0952381,0.14285715,0.0952381,0.23809524,0.14285715,0.23809524,0.0952381,0.2857143,0.2857143,0.1904762,0.23809524,0.1904762,0.14285715,0.2857143,0.14285715,0.1904762,0.0952381,0.23809524,0.0952381,0.23809524,0.14285715,0.23809524,0.23809524,0.23809524,0.23809524,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.2857143,0.33333334,0.47619048,0.42857143,0.47619048,0.3809524,0.42857143,0.42857143,0.33333334,0.42857143,0.52380955,0.42857143,0.47619048,0.52380955,,0.61904764,0.71428573,0.52380955,0.52380955,0.71428573,0.6666667,0.85714287,0.61904764,0.71428573,0.7619048,0.71428573,0.71428573,0.7619048,0.85714287,0.95238096
+TabSTAR (default),0.0952381,0.23809524,0.1904762,0.1904762,0.2857143,0.23809524,0.23809524,0.23809524,0.2857143,0.1904762,0.23809524,0.23809524,0.2857143,0.23809524,0.23809524,0.23809524,0.2857143,0.1904762,0.2857143,0.2857143,0.2857143,0.33333334,0.2857143,0.1904762,0.33333334,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.2857143,0.2857143,0.33333334,0.2857143,0.33333334,0.42857143,0.2857143,0.3809524,0.23809524,0.33333334,0.3809524,0.0952381,0.04761905,0.52380955,0.33333334,0.42857143,0.3809524,0.42857143,0.2857143,0.33333334,0.42857143,0.52380955,0.3809524,,0.42857143,0.52380955,0.5714286,0.42857143,0.52380955,0.5714286,0.3809524,0.5714286,0.5714286,0.52380955,0.52380955,0.5714286,0.6666667,0.9047619
+RandomForest (tuned),0.04761905,0.0,0.0,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.04761905,0.04761905,0.14285715,0.04761905,0.1904762,0.1904762,0.23809524,0.0952381,0.23809524,0.23809524,0.0952381,0.23809524,0.14285715,0.14285715,0.23809524,0.0952381,0.1904762,0.04761905,0.23809524,0.1904762,0.23809524,0.0952381,0.23809524,0.23809524,0.1904762,0.14285715,0.33333334,0.2857143,0.23809524,0.2857143,0.2857143,0.1904762,0.1904762,0.2857143,0.2857143,0.33333334,0.3809524,0.42857143,0.42857143,0.3809524,0.2857143,0.3809524,0.42857143,0.42857143,0.33333334,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,,0.61904764,0.47619048,0.6666667,0.47619048,0.7380952,0.71428573,0.7619048,0.8095238,0.71428573,0.71428573,0.71428573,0.85714287,0.95238096
+TabFlex (default),0.0,0.0952381,0.0,0.0,0.14285715,0.1904762,0.0952381,0.1904762,0.1904762,0.14285715,0.0952381,0.04761905,0.23809524,0.0952381,0.1904762,0.14285715,0.14285715,0.04761905,0.1904762,0.23809524,0.14285715,0.23809524,0.23809524,0.23809524,0.2857143,0.1904762,0.2857143,0.14285715,0.23809524,0.0952381,0.2857143,0.23809524,0.1904762,0.1904762,0.2857143,0.33333334,0.1904762,0.1904762,0.1904762,0.23809524,0.14285715,0.23809524,0.2857143,0.33333334,0.14285715,0.1904762,0.42857143,0.47619048,0.3809524,0.33333334,0.2857143,0.42857143,0.42857143,0.3809524,0.2857143,0.3809524,0.33333334,0.47619048,0.47619048,0.3809524,,0.52380955,0.52380955,0.47619048,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.7619048,0.85714287
+PerpetualBooster (default),0.0,0.04761905,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.04761905,0.14285715,0.04761905,0.0952381,0.04761905,0.1904762,0.1904762,0.0952381,0.0952381,0.1904762,0.14285715,0.1904762,0.0952381,0.0952381,0.0952381,0.23809524,0.0952381,0.23809524,0.23809524,0.1904762,0.1904762,0.0952381,0.23809524,0.1904762,0.1904762,0.23809524,0.1904762,0.1904762,0.1904762,0.14285715,0.23809524,0.2857143,0.1904762,0.2857143,0.14285715,0.2857143,0.23809524,0.14285715,0.42857143,0.42857143,0.23809524,0.3809524,0.33333334,0.52380955,0.2857143,0.42857143,0.33333334,0.3809524,0.23809524,0.47619048,0.42857143,0.52380955,0.47619048,,0.5714286,0.47619048,0.5714286,0.3809524,0.61904764,0.61904764,0.6666667,0.6666667,0.6666667,0.6666667,0.9047619
+ExtraTrees (default),0.04761905,0.04761905,0.0,0.14285715,0.0952381,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.0952381,0.04761905,0.2857143,0.23809524,0.0952381,0.14285715,0.14285715,0.1904762,0.2857143,0.23809524,0.23809524,0.23809524,0.1904762,0.23809524,0.14285715,0.1904762,0.1904762,0.1904762,0.1904762,0.2857143,0.14285715,0.1904762,0.23809524,0.2857143,0.1904762,0.1904762,0.1904762,0.1904762,0.2857143,0.23809524,0.14285715,0.23809524,0.2857143,0.23809524,0.2857143,0.33333334,0.3809524,0.14285715,0.2857143,0.3809524,0.33333334,0.23809524,0.2857143,0.2857143,0.3809524,0.2857143,0.5714286,0.33333334,0.47619048,0.42857143,,0.3809524,0.61904764,0.47619048,0.52380955,0.52380955,0.61904764,0.61904764,0.61904764,0.71428573,0.9047619
+TorchMLP (default),0.0,0.04761905,0.0,0.0,0.04761905,0.0952381,0.0952381,0.0952381,0.0952381,0.0952381,0.04761905,0.14285715,0.04761905,0.1904762,0.14285715,0.04761905,0.04761905,0.14285715,0.1904762,0.1904762,0.04761905,0.0952381,0.0952381,0.1904762,0.0952381,0.14285715,0.14285715,0.1904762,0.0952381,0.04761905,0.1904762,0.0952381,0.1904762,0.14285715,0.23809524,0.23809524,0.1904762,0.14285715,0.1904762,0.23809524,0.0952381,0.1904762,0.1904762,0.1904762,0.0952381,0.23809524,0.3809524,0.33333334,0.3809524,0.42857143,0.33333334,0.33333334,0.3809524,0.47619048,0.3809524,0.3809524,0.3809524,0.33333334,0.47619048,0.52380955,0.52380955,0.52380955,0.61904764,,0.5714286,0.5714286,0.61904764,0.5714286,0.61904764,0.6666667,0.71428573,0.7619048,0.9047619
+RandomForest (default),0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.04761905,0.14285715,0.04761905,0.0952381,0.1904762,0.23809524,0.04761905,0.23809524,0.14285715,0.04761905,0.23809524,0.14285715,0.14285715,0.14285715,0.0952381,0.1904762,0.0952381,0.23809524,0.14285715,0.1904762,0.14285715,0.23809524,0.14285715,0.2857143,0.14285715,0.23809524,0.1904762,0.23809524,0.1904762,0.23809524,0.2857143,0.2857143,0.1904762,0.23809524,0.33333334,0.2857143,0.23809524,0.2857143,0.23809524,0.1904762,0.33333334,0.3809524,0.2857143,0.23809524,0.33333334,0.2857143,0.33333334,0.14285715,0.42857143,0.26190478,0.42857143,0.42857143,0.3809524,0.42857143,,0.52380955,0.61904764,0.5714286,0.61904764,0.61904764,0.6666667,0.7619048,0.95238096
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.0,0.0952381,0.0,0.14285715,0.0952381,0.0952381,0.04761905,0.0,0.14285715,0.1904762,0.14285715,0.14285715,0.1904762,0.1904762,0.23809524,0.1904762,0.14285715,0.0952381,0.04761905,0.1904762,0.23809524,0.0952381,0.23809524,0.1904762,0.23809524,0.1904762,0.14285715,0.1904762,0.14285715,0.14285715,0.23809524,0.14285715,0.23809524,0.23809524,0.04761905,0.33333334,0.33333334,0.33333334,0.2857143,0.2857143,0.3809524,0.2857143,0.33333334,0.42857143,0.33333334,0.3809524,0.3809524,0.61904764,0.2857143,0.42857143,0.61904764,0.52380955,0.42857143,0.47619048,,0.61904764,0.61904764,0.5714286,0.5714286,0.5714286,0.8095238,0.9047619
+KNN (tuned + ensembled),0.04761905,0.0,0.0,0.04761905,0.0,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0,0.0952381,0.0952381,0.04761905,0.14285715,0.04761905,0.04761905,0.14285715,0.0,0.14285715,0.0952381,0.04761905,0.0952381,0.14285715,0.14285715,0.14285715,0.04761905,0.0952381,0.04761905,0.0952381,0.1904762,0.14285715,0.1904762,0.0952381,0.21428572,0.23809524,0.04761905,0.04761905,0.14285715,0.14285715,0.1904762,0.14285715,0.14285715,0.14285715,0.14285715,0.33333334,0.3809524,0.23809524,0.0952381,0.1904762,0.1904762,0.2857143,0.14285715,0.33333334,0.33333334,0.33333334,0.2857143,0.42857143,0.23809524,0.42857143,0.3809524,0.47619048,0.3809524,0.3809524,0.3809524,,0.6666667,0.61904764,0.61904764,0.61904764,0.95238096,0.95238096
+xRFM (default),0.0,0.0,0.0,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0,0.14285715,0.14285715,0.04761905,0.04761905,0.0952381,0.14285715,0.1904762,0.14285715,0.0952381,0.0952381,0.14285715,0.1904762,0.14285715,0.14285715,0.04761905,0.14285715,0.14285715,0.23809524,0.14285715,0.1904762,0.14285715,0.0952381,0.14285715,0.1904762,0.0952381,0.1904762,0.1904762,0.1904762,0.04761905,0.2857143,0.23809524,0.1904762,0.1904762,0.23809524,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.33333334,0.2857143,0.23809524,0.2857143,0.23809524,0.42857143,0.1904762,0.42857143,0.3809524,0.47619048,0.42857143,0.42857143,0.3809524,0.33333334,,0.52380955,0.52380955,0.52380955,0.6666667,0.85714287
+Linear (tuned + ensembled),0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.04761905,0.0952381,0.0952381,0.04761905,0.14285715,0.04761905,0.14285715,0.04761905,0.04761905,0.14285715,0.04761905,0.0952381,0.14285715,0.04761905,0.14285715,0.04761905,0.1904762,0.0952381,0.0952381,0.14285715,0.14285715,0.2857143,0.0952381,0.04761905,0.0952381,0.1904762,0.14285715,0.14285715,0.14285715,0.14285715,0.1904762,0.23809524,0.33333334,0.33333334,0.1904762,0.1904762,0.23809524,0.2857143,0.23809524,0.2857143,0.1904762,0.2857143,0.23809524,0.2857143,0.47619048,0.2857143,0.2857143,0.33333334,0.3809524,0.3809524,0.3809524,0.42857143,0.3809524,0.47619048,,0.7380952,0.61904764,0.5714286,0.8095238
+Linear (tuned),0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.0952381,0.071428575,0.0952381,0.04761905,0.0952381,0.0952381,0.04761905,0.0952381,0.04761905,0.14285715,0.04761905,0.04761905,0.0952381,0.04761905,0.0952381,0.0952381,0.04761905,0.14285715,0.04761905,0.14285715,0.0952381,0.04761905,0.14285715,0.14285715,0.23809524,0.14285715,0.04761905,0.0952381,0.1904762,0.14285715,0.14285715,0.14285715,0.14285715,0.1904762,0.1904762,0.2857143,0.33333334,0.1904762,0.14285715,0.23809524,0.2857143,0.1904762,0.2857143,0.1904762,0.2857143,0.2857143,0.2857143,0.47619048,0.2857143,0.2857143,0.33333334,0.3809524,0.33333334,0.3809524,0.42857143,0.3809524,0.47619048,0.26190478,,0.52380955,0.52380955,0.7619048
+Linear (default),0.0,0.0,0.0,0.0,0.04761905,0.0952381,0.04761905,0.0952381,0.04761905,0.04761905,0.0,0.0952381,0.0952381,0.14285715,0.04761905,0.0952381,0.0952381,0.0952381,0.14285715,0.0952381,0.0952381,0.04761905,0.04761905,0.14285715,0.04761905,0.04761905,0.0952381,0.04761905,0.1904762,0.04761905,0.0952381,0.14285715,0.0952381,0.0952381,0.14285715,0.2857143,0.1904762,0.04761905,0.14285715,0.1904762,0.04761905,0.04761905,0.1904762,0.14285715,0.14285715,0.1904762,0.2857143,0.2857143,0.1904762,0.23809524,0.23809524,0.1904762,0.23809524,0.23809524,0.1904762,0.1904762,0.23809524,0.23809524,0.42857143,0.2857143,0.2857143,0.33333334,0.3809524,0.2857143,0.33333334,0.42857143,0.3809524,0.47619048,0.3809524,0.47619048,,0.5714286,0.6666667
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.04761905,0.0,0.04761905,0.0,0.04761905,0.0,0.0,0.0,0.04761905,0.0,0.0,0.04761905,0.0,0.04761905,0.0,0.04761905,0.0,0.0,0.0952381,0.04761905,0.04761905,0.0952381,0.04761905,0.0,0.04761905,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.1904762,0.0952381,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.1904762,0.1904762,0.23809524,0.0952381,0.14285715,0.14285715,0.1904762,0.04761905,0.1904762,0.14285715,0.23809524,0.14285715,0.33333334,0.14285715,0.23809524,0.33333334,0.2857143,0.23809524,0.23809524,0.1904762,0.04761905,0.33333334,0.42857143,0.47619048,0.42857143,,0.9047619
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.04761905,0.0,0.04761905,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0952381,0.04761905,0.0,0.0,0.04761905,0.0,0.0,0.0,0.04761905,0.04761905,0.0,0.0,0.04761905,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.0,0.0952381,0.0952381,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.04761905,0.0952381,0.04761905,0.14285715,0.0952381,0.0952381,0.0952381,0.04761905,0.0952381,0.04761905,0.14285715,0.1904762,0.23809524,0.33333334,0.0952381,
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_all/leaderboard_overview_explorer.html
index d9b088b1ee86f9429f2dc2e61af31263f1e4d8a4..d124548ddf48e47862b67f5c03a75e59b8c9bd3a 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Classification | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_all/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_classification/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..4e896a2cd3a2fd7d77c32e01903c8addc33ef2c9
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_classification/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1751.0,"score":0.798,"rank":6.87,"hrank":1.64,"imp":5.482,"train_time":40.79,"predict_time":7.056,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+167\/-140","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1638.0,"score":0.62,"rank":10.97,"hrank":3.95,"imp":8.501,"train_time":268.44,"predict_time":4.668,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-106","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1624.0,"score":0.592,"rank":11.57,"hrank":4.56,"imp":10.455,"train_time":3.52,"predict_time":0.385,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+100\/-81","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1595.0,"score":0.539,"rank":12.95,"hrank":7.51,"imp":12.105,"train_time":5.39,"predict_time":0.531,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-71","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1576.0,"score":0.514,"rank":13.93,"hrank":5.78,"imp":10.831,"train_time":2063.62,"predict_time":9.938,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-73","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1567.0,"score":0.515,"rank":14.42,"hrank":6.85,"imp":12.359,"train_time":2.04,"predict_time":0.182,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-92","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1538.0,"score":0.457,"rank":16.0,"hrank":9.12,"imp":12.84,"train_time":5.63,"predict_time":0.771,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-65","imputed":false},{"position":7,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1538.0,"score":0.47,"rank":16.03,"hrank":6.66,"imp":13.323,"train_time":1815.89,"predict_time":1.455,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-60","imputed":false},{"position":8,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1536.0,"score":0.479,"rank":16.12,"hrank":9.0,"imp":13.099,"train_time":2063.62,"predict_time":1.086,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-74","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1485.0,"score":0.395,"rank":19.29,"hrank":11.72,"imp":14.71,"train_time":2774.17,"predict_time":16.363,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-54","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1420.0,"score":0.339,"rank":23.8,"hrank":12.72,"imp":16.129,"train_time":2774.17,"predict_time":0.39,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-53","imputed":false},{"position":11,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1416.0,"score":0.336,"rank":24.11,"hrank":11.52,"imp":16.431,"train_time":2207.17,"predict_time":1.894,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-65","imputed":false},{"position":12,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1400.0,"score":0.299,"rank":25.33,"hrank":14.71,"imp":16.523,"train_time":12550.1,"predict_time":427.59,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-66","imputed":false},{"position":13,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1399.0,"score":0.36,"rank":25.39,"hrank":7.3,"imp":16.03,"train_time":18.2,"predict_time":4.692,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-72","imputed":false},{"position":14,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1398.0,"score":0.296,"rank":25.45,"hrank":13.32,"imp":15.79,"train_time":1087.73,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-53","imputed":false},{"position":15,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1394.0,"score":0.258,"rank":25.76,"hrank":19.48,"imp":17.042,"train_time":391.27,"predict_time":1.243,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-43","imputed":false},{"position":16,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1392.0,"score":0.303,"rank":25.91,"hrank":11.89,"imp":17.028,"train_time":2.1,"predict_time":0.186,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-63","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1390.0,"score":0.28,"rank":26.01,"hrank":16.58,"imp":16.156,"train_time":1087.73,"predict_time":0.348,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-49","imputed":false},{"position":18,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1377.0,"score":0.295,"rank":27.07,"hrank":13.95,"imp":17.228,"train_time":2207.17,"predict_time":0.234,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-62","imputed":false},{"position":19,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1372.0,"score":0.289,"rank":27.42,"hrank":12.42,"imp":16.606,"train_time":6067.42,"predict_time":404.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-59","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1371.0,"score":0.25,"rank":27.49,"hrank":19.02,"imp":17.555,"train_time":682.62,"predict_time":0.971,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-62","imputed":false},{"position":21,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1358.0,"score":0.256,"rank":28.51,"hrank":15.56,"imp":17.561,"train_time":4853.47,"predict_time":0.473,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-72","imputed":false},{"position":22,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1356.0,"score":0.238,"rank":28.66,"hrank":18.83,"imp":17.259,"train_time":4.89,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-54","imputed":false},{"position":23,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1346.0,"score":0.269,"rank":29.47,"hrank":17.32,"imp":17.613,"train_time":4853.47,"predict_time":9.385,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-85","imputed":false},{"position":24,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1343.0,"score":0.247,"rank":29.7,"hrank":20.12,"imp":18.687,"train_time":500.4,"predict_time":0.416,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-70","imputed":false},{"position":25,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1342.0,"score":0.216,"rank":29.74,"hrank":23.42,"imp":17.949,"train_time":391.27,"predict_time":0.186,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-52","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1341.0,"score":0.226,"rank":29.84,"hrank":14.72,"imp":17.769,"train_time":682.62,"predict_time":0.183,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-62","imputed":false},{"position":27,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1327.0,"score":0.215,"rank":30.93,"hrank":18.31,"imp":18.712,"train_time":1.2,"predict_time":0.064,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-63","imputed":false},{"position":28,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1327.0,"score":0.242,"rank":30.99,"hrank":15.77,"imp":18.333,"train_time":6067.42,"predict_time":39.564,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-70","imputed":false},{"position":29,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1322.0,"score":0.22,"rank":31.32,"hrank":20.57,"imp":19.599,"train_time":250.99,"predict_time":2.035,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-57","imputed":false},{"position":30,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1318.0,"score":0.224,"rank":31.68,"hrank":18.57,"imp":19.082,"train_time":500.4,"predict_time":0.031,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-71","imputed":false},{"position":31,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1309.0,"score":0.159,"rank":32.39,"hrank":22.54,"imp":19.413,"train_time":1449.14,"predict_time":0.149,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-54","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1308.0,"score":0.222,"rank":32.47,"hrank":18.83,"imp":18.412,"train_time":1053.8,"predict_time":2.067,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-60","imputed":false},{"position":33,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1307.0,"score":0.186,"rank":32.58,"hrank":22.77,"imp":18.27,"train_time":12550.1,"predict_time":61.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+51\/-54","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1298.0,"score":0.197,"rank":33.28,"hrank":20.74,"imp":19.485,"train_time":7.26,"predict_time":0.25,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-61","imputed":false},{"position":35,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1272.0,"score":0.151,"rank":35.42,"hrank":27.88,"imp":19.304,"train_time":2480.8,"predict_time":2.175,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-52","imputed":false},{"position":36,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1264.0,"score":0.164,"rank":36.08,"hrank":16.55,"imp":19.96,"train_time":1053.8,"predict_time":0.105,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-54","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1263.0,"score":0.127,"rank":36.11,"hrank":26.35,"imp":20.001,"train_time":10.77,"predict_time":1.786,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-46","imputed":false},{"position":38,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1262.0,"score":0.161,"rank":36.18,"hrank":26.35,"imp":20.627,"train_time":1.92,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-79","imputed":false},{"position":39,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1252.0,"score":0.195,"rank":37.03,"hrank":17.44,"imp":21.476,"train_time":13.62,"predict_time":1.984,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-96","imputed":false},{"position":40,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1250.0,"score":0.108,"rank":37.16,"hrank":27.54,"imp":20.365,"train_time":1449.14,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-57","imputed":false},{"position":41,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1248.0,"score":0.126,"rank":37.38,"hrank":17.76,"imp":20.947,"train_time":5.48,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-50","imputed":false},{"position":42,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1215.0,"score":0.143,"rank":40.01,"hrank":22.87,"imp":21.318,"train_time":47.89,"predict_time":39.444,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-67","imputed":false},{"position":43,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1211.0,"score":0.121,"rank":40.36,"hrank":29.36,"imp":22.133,"train_time":622.86,"predict_time":5.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-85","imputed":false},{"position":44,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1206.0,"score":0.115,"rank":40.76,"hrank":18.55,"imp":20.838,"train_time":1.88,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-68","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1204.0,"score":0.117,"rank":40.86,"hrank":30.27,"imp":21.497,"train_time":2480.8,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-62","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1193.0,"score":0.103,"rank":41.79,"hrank":28.76,"imp":21.715,"train_time":282.36,"predict_time":0.86,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-66","imputed":false},{"position":47,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1180.0,"score":0.09,"rank":42.79,"hrank":17.22,"imp":22.905,"train_time":13.35,"predict_time":0.352,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-62","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1168.0,"score":0.096,"rank":43.74,"hrank":34.05,"imp":22.782,"train_time":327.37,"predict_time":0.708,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-72","imputed":false},{"position":49,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1161.0,"score":0.084,"rank":44.28,"hrank":29.42,"imp":23.291,"train_time":282.36,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-71","imputed":false},{"position":50,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1154.0,"score":0.082,"rank":44.83,"hrank":37.31,"imp":21.988,"train_time":2.25,"predict_time":0.159,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-59","imputed":false},{"position":51,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1137.0,"score":0.084,"rank":46.12,"hrank":26.9,"imp":24.356,"train_time":267.89,"predict_time":62.492,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-67","imputed":false},{"position":52,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1133.0,"score":0.076,"rank":46.45,"hrank":32.4,"imp":24.04,"train_time":622.86,"predict_time":0.345,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-76","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1129.0,"score":0.052,"rank":46.75,"hrank":39.97,"imp":23.545,"train_time":327.37,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-62","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1114.0,"score":0.122,"rank":47.89,"hrank":18.05,"imp":25.995,"train_time":38846.7,"predict_time":4.269,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-92","imputed":false},{"position":55,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1103.0,"score":0.111,"rank":48.67,"hrank":19.85,"imp":26.21,"train_time":38846.7,"predict_time":19.475,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-94","imputed":false},{"position":56,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1067.0,"score":0.04,"rank":51.29,"hrank":44.26,"imp":26.3,"train_time":16.07,"predict_time":2.876,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-74","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1060.0,"score":0.075,"rank":51.76,"hrank":40.89,"imp":31.161,"train_time":175.23,"predict_time":0.654,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-86","imputed":false},{"position":58,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1042.0,"score":0.024,"rank":53.0,"hrank":47.99,"imp":26.299,"train_time":5.73,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-74","imputed":false},{"position":59,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1012.0,"score":0.088,"rank":54.89,"hrank":17.71,"imp":32.417,"train_time":403.92,"predict_time":4.972,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-122","imputed":false},{"position":60,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1008.0,"score":0.041,"rank":55.17,"hrank":49.81,"imp":32.688,"train_time":175.23,"predict_time":0.2,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-93","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1001.0,"score":0.027,"rank":55.63,"hrank":51.27,"imp":27.616,"train_time":3.13,"predict_time":0.334,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-62","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.016,"rank":55.67,"hrank":45.9,"imp":29.406,"train_time":0.37,"predict_time":0.038,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-83","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":994.0,"score":0.034,"rank":56.01,"hrank":47.71,"imp":29.787,"train_time":140.69,"predict_time":2.876,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-99","imputed":false},{"position":64,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":985.0,"score":0.034,"rank":56.55,"hrank":47.82,"imp":29.449,"train_time":1.75,"predict_time":0.268,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-95","imputed":false},{"position":65,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":980.0,"score":0.042,"rank":56.87,"hrank":49.84,"imp":30.634,"train_time":3.48,"predict_time":0.677,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-90","imputed":false},{"position":66,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":970.0,"score":0.042,"rank":57.45,"hrank":48.41,"imp":32.357,"train_time":1.04,"predict_time":0.192,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-121","imputed":false},{"position":67,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":969.0,"score":0.03,"rank":57.54,"hrank":22.94,"imp":33.227,"train_time":256.18,"predict_time":0.36,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-100","imputed":false},{"position":68,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":944.0,"score":0.023,"rank":58.89,"hrank":33.47,"imp":33.902,"train_time":256.18,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-108","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":935.0,"score":0.032,"rank":59.38,"hrank":33.08,"imp":38.349,"train_time":20.44,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-110","imputed":false},{"position":70,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":882.0,"score":0.008,"rank":61.97,"hrank":58.09,"imp":36.441,"train_time":1.32,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-111","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":832.0,"score":0.017,"rank":64.08,"hrank":60.23,"imp":36.943,"train_time":140.69,"predict_time":0.188,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-112","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":573.0,"score":0.0,"rank":70.74,"hrank":70.45,"imp":51.213,"train_time":0.18,"predict_time":0.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-130","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_all/pareto_front_explorer.html
index e4294501371e4704bbfa95b27a6e573f0a8d4c7f..a429336f8c6484b76fd325961fc677d756b8e415 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0351934444,"imp":19.5991360091,"elo":1322.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.024809806,"imp":17.2593580753,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0341120659,"imp":15.7899630676,"elo":1397.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3481993779,"imp":16.1559467271,"elo":1390.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0396842484,"imp":20.6266564966,"elo":1262.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306719592,"imp":19.0819305063,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4161492489,"imp":18.6866560916,"elo":1342.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0162731597,"imp":20.9472417246,"elo":1247.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0167928172,"imp":20.365363067,"elo":1250.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1493309259,"imp":19.4134207053,"elo":1309.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2675026243,"imp":29.448605072,"elo":985.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721048132,"imp":23.2906463543,"elo":1161.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8604796191,"imp":21.7153094443,"elo":1192.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3340852103,"imp":27.6157771175,"elo":1000.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3452999313,"imp":24.0396405638,"elo":1132.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0368770459,"imp":22.1334117958,"elo":1210.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1585029211,"imp":21.9878705376,"elo":1154.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1855840981,"imp":17.9491405333,"elo":1342.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.2430091096,"imp":17.041869522,"elo":1393.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.6915866654,"imp":16.0298285664,"elo":1398.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"imp":22.9053008148,"elo":1180.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4731015689,"imp":17.5609330253,"elo":1358.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"imp":17.6129098777,"elo":1345.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.876262961,"imp":26.2998114523,"elo":1066.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0443776494,"imp":38.3494691433,"elo":935.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1999302717,"imp":32.6877193437,"elo":1008.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6542696803,"imp":31.161195656,"elo":1059.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0376682658,"imp":29.4063894551,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0845096703,"imp":23.544975909,"elo":1128.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7082779811,"imp":22.7824294054,"elo":1168.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7864396959,"imp":20.0014973903,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3898971973,"imp":16.1287534896,"elo":1420.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":16.3626514212,"imp":14.709745279,"elo":1485.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7713317233,"imp":12.8403978114,"elo":1538.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0864910732,"imp":13.0990794801,"elo":1536.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.9380601197,"imp":10.8312429122,"elo":1575.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.9837292889,"imp":21.4761627666,"elo":1251.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"imp":21.3177857177,"elo":1215.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.5640473457,"imp":18.3333846884,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":404.0529881985,"imp":16.6059987181,"elo":1372.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1861515215,"imp":17.0275444784,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"imp":5.4819228733,"elo":1751.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1916685529,"imp":32.3572286499,"elo":970.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"imp":12.3593539436,"elo":1566.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2496062173,"imp":19.4848396397,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2337420667,"imp":17.2283717497,"elo":1376.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.8937013268,"imp":16.4308704356,"elo":1416.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"imp":12.1054842961,"elo":1595.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3846051827,"imp":10.4550501811,"elo":1624.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9719997826,"imp":32.4174821311,"elo":1012.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2687972891,"imp":25.9951447671,"elo":1113.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.4748027028,"imp":26.2096152238,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0643226705,"imp":18.7119956902,"elo":1327.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1483357715,"imp":26.2994101272,"elo":1041.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1166104296,"imp":21.4971416867,"elo":1204.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1753468186,"imp":19.3036071183,"elo":1271.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209262941,"imp":20.8376274088,"elo":1205.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1829513734,"imp":17.7691449945,"elo":1341.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.9713359475,"imp":17.5552418689,"elo":1371.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":62.4915524975,"imp":24.3558189487,"elo":1137.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":61.3096906599,"imp":18.2698156508,"elo":1306.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":427.5904998868,"imp":16.5229870014,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.6769281514,"imp":30.6344852973,"elo":980.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1053014255,"imp":19.959547844,"elo":1263.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0671352109,"imp":18.4115272126,"elo":1308.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0351934444,"x_train":250.9922109033,"imp":19.5991360091,"elo":1322.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.024809806,"x_train":4.8861388456,"imp":17.2593580753,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0341120659,"x_train":1087.7269134101,"imp":15.7899630676,"elo":1397.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3481993779,"x_train":1087.7269134101,"imp":16.1559467271,"elo":1390.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0396842484,"x_train":1.9155810093,"imp":20.6266564966,"elo":1262.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306719592,"x_train":500.3975293451,"imp":19.0819305063,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4161492489,"x_train":500.3975293451,"imp":18.6866560916,"elo":1342.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0162731597,"x_train":5.4796099733,"imp":20.9472417246,"elo":1247.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0167928172,"x_train":1449.1413756847,"imp":20.365363067,"elo":1250.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1493309259,"x_train":1449.1413756847,"imp":19.4134207053,"elo":1309.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2675026243,"x_train":1.7477789606,"imp":29.448605072,"elo":985.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721048132,"x_train":282.3556156822,"imp":23.2906463543,"elo":1161.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8604796191,"x_train":282.3556156822,"imp":21.7153094443,"elo":1192.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3340852103,"x_train":3.1251204631,"imp":27.6157771175,"elo":1000.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3452999313,"x_train":622.8560318368,"imp":24.0396405638,"elo":1132.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0368770459,"x_train":622.8560318368,"imp":22.1334117958,"elo":1210.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1585029211,"x_train":2.2519280723,"imp":21.9878705376,"elo":1154.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1855840981,"x_train":391.2703614285,"imp":17.9491405333,"elo":1342.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.2430091096,"x_train":391.2703614285,"imp":17.041869522,"elo":1393.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.6915866654,"x_train":18.1964917538,"imp":16.0298285664,"elo":1398.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"x_train":13.3530116618,"imp":22.9053008148,"elo":1180.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4731015689,"x_train":4853.4730504877,"imp":17.5609330253,"elo":1358.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"x_train":4853.4730504877,"imp":17.6129098777,"elo":1345.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.876262961,"x_train":16.0704711344,"imp":26.2998114523,"elo":1066.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0443776494,"x_train":20.4435114187,"imp":38.3494691433,"elo":935.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1999302717,"x_train":175.2300293672,"imp":32.6877193437,"elo":1008.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6542696803,"x_train":175.2300293672,"imp":31.161195656,"elo":1059.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0376682658,"x_train":0.3738939375,"imp":29.4063894551,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0845096703,"x_train":327.3662383914,"imp":23.544975909,"elo":1128.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7082779811,"x_train":327.3662383914,"imp":22.7824294054,"elo":1168.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7864396959,"x_train":10.7684699326,"imp":20.0014973903,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3898971973,"x_train":2774.1715903227,"imp":16.1287534896,"elo":1420.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":16.3626514212,"x_train":2774.1715903227,"imp":14.709745279,"elo":1485.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7713317233,"x_train":5.6257151642,"imp":12.8403978114,"elo":1538.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0864910732,"x_train":2063.6226141883,"imp":13.0990794801,"elo":1536.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.9380601197,"x_train":2063.6226141883,"imp":10.8312429122,"elo":1575.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.9837292889,"x_train":13.6228654067,"imp":21.4761627666,"elo":1251.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"x_train":47.8871994,"imp":21.3177857177,"elo":1215.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.5640473457,"x_train":6067.4246083224,"imp":18.3333846884,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":404.0529881985,"x_train":6067.4246083224,"imp":16.6059987181,"elo":1372.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1861515215,"x_train":2.0990306801,"imp":17.0275444784,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"x_train":40.7899647401,"imp":5.4819228733,"elo":1751.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1916685529,"x_train":1.0350786389,"imp":32.3572286499,"elo":970.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"x_train":2.0391419632,"imp":12.3593539436,"elo":1566.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2496062173,"x_train":7.2637875629,"imp":19.4848396397,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2337420667,"x_train":2207.1695731088,"imp":17.2283717497,"elo":1376.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.8937013268,"x_train":2207.1695731088,"imp":16.4308704356,"elo":1416.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"x_train":5.3935958382,"imp":12.1054842961,"elo":1595.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3846051827,"x_train":3.5186684216,"imp":10.4550501811,"elo":1624.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9719997826,"x_train":403.917691534,"imp":32.4174821311,"elo":1012.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2687972891,"x_train":38846.7008402763,"imp":25.9951447671,"elo":1113.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.4748027028,"x_train":38846.7008402763,"imp":26.2096152238,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0643226705,"x_train":1.1995725235,"imp":18.7119956902,"elo":1327.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1483357715,"x_train":5.7268864593,"imp":26.2994101272,"elo":1041.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1166104296,"x_train":2480.8031686863,"imp":21.4971416867,"elo":1204.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1753468186,"x_train":2480.8031686863,"imp":19.3036071183,"elo":1271.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209262941,"x_train":1.8759235268,"imp":20.8376274088,"elo":1205.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1829513734,"x_train":682.6224605816,"imp":17.7691449945,"elo":1341.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.9713359475,"x_train":682.6224605816,"imp":17.5552418689,"elo":1371.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":62.4915524975,"x_train":267.891665628,"imp":24.3558189487,"elo":1137.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":61.3096906599,"x_train":12550.0998837068,"imp":18.2698156508,"elo":1306.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":427.5904998868,"x_train":12550.0998837068,"imp":16.5229870014,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.6769281514,"x_train":3.4779953241,"imp":30.6344852973,"elo":980.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1053014255,"x_train":1053.8003845303,"imp":19.959547844,"elo":1263.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0671352109,"x_train":1053.8003845303,"imp":18.4115272126,"elo":1308.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_all/tuning_trajectories_explorer.html
index 88c3c426df3ba1ae35f8530a8dd641db1b8f87ad..182ceae30021a94ebef2ba304666bd28f563c428 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":250.9922109033,"imp":19.6395723532,"elo":1342.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":4.8861388456,"imp":17.295070603,"elo":1375.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":10.6750136791,"imp":16.8081666354,"elo":1393.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":25.2669903942,"imp":16.5553434634,"elo":1403.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":49.0638838885,"imp":16.4131006566,"elo":1410.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":133.206456981,"imp":16.3518455152,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":264.5793957685,"imp":16.3707724611,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":539.0252673695,"imp":16.2473570494,"elo":1408.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":810.5253697201,"imp":16.260193325,"elo":1402.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1087.7269134101,"imp":16.1779645054,"elo":1407.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.9155810093,"imp":20.6656491574,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.4587928667,"imp":20.0079904311,"elo":1316.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.7223762673,"imp":19.5822174931,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":24.3697423433,"imp":19.3248019843,"elo":1345.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":61.9355959829,"imp":18.9868988263,"elo":1357.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":124.7079952924,"imp":18.7984834127,"elo":1365.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":249.4958722159,"imp":18.7259661622,"elo":1366.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":373.7478601635,"imp":18.6892810323,"elo":1368.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":500.3975293451,"imp":18.7185872994,"elo":1367.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":5.4796099733,"imp":20.9805486921,"elo":1253.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":13.9490355599,"imp":20.4975061878,"elo":1273.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":38.1339214312,"imp":20.1244528497,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":74.7850861491,"imp":20.0032103124,"elo":1297.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":183.3974305409,"imp":19.8442079373,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":366.1485755963,"imp":19.6834999267,"elo":1316.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":725.871285674,"imp":19.5170527235,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1084.12220491,"imp":19.4807401023,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1449.1413756847,"imp":19.4397032529,"elo":1326.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.7477789606,"imp":29.4778153146,"elo":992.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1400651467,"imp":27.2957858125,"elo":1028.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.3396935938,"imp":23.9953600978,"elo":1099.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.4314608681,"imp":22.9279464902,"elo":1138.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":35.4438180004,"imp":22.1910011881,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":70.3987810435,"imp":21.9306885793,"elo":1180.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":140.7929692058,"imp":21.7221031901,"elo":1191.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":211.1596497818,"imp":21.5976427028,"elo":1200.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":282.3556156822,"imp":21.7838917473,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.1251204631,"imp":27.6418948918,"elo":1007.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.228658512,"imp":26.5341183951,"elo":1061.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":15.6396665341,"imp":24.5187984291,"elo":1133.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":31.5302468269,"imp":23.6654630089,"elo":1164.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":76.3915347501,"imp":22.7418354905,"elo":1188.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":153.0430713803,"imp":22.3653463542,"elo":1203.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":308.6007823833,"imp":22.1858964506,"elo":1210.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":463.687419557,"imp":22.108048838,"elo":1213.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":622.8560318368,"imp":22.1743713286,"elo":1219.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.2519280723,"imp":22.0237645428,"elo":1157.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":4.5917771836,"imp":20.0235785837,"elo":1259.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":9.9026612043,"imp":18.6688562588,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":18.7813971613,"imp":18.020092398,"elo":1374.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":48.1169020642,"imp":17.4420055099,"elo":1394.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":97.0735394653,"imp":17.2474827759,"elo":1406.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":194.547566338,"imp":17.1370232932,"elo":1413.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":290.5602659885,"imp":17.0606732736,"elo":1415.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":391.2703614285,"imp":17.0772120491,"elo":1416.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":18.1964917538,"imp":16.2254459782,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":13.3530116618,"imp":22.9392691064,"elo":1186.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":43.609446858,"imp":21.2645554845,"elo":1226.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":123.2677201414,"imp":18.7260581766,"elo":1328.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":240.5329414556,"imp":18.1737471821,"elo":1352.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":578.0322705513,"imp":17.622242904,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1205.8399034278,"imp":17.636430837,"elo":1372.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2388.27785775,"imp":17.8823109886,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3575.4198584316,"imp":17.7863190447,"elo":1366.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4853.4730504877,"imp":17.6703087977,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":16.0704711344,"imp":26.1705122737,"elo":1082.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.4435114187,"imp":38.3890458814,"elo":943.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":50.7468311596,"imp":35.0877917631,"elo":999.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":146.088248762,"imp":31.3764424518,"elo":1065.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":175.2300293672,"imp":31.165634527,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3738939375,"imp":29.451436506,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.1861266183,"imp":26.1061150853,"elo":1059.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":7.0625783305,"imp":24.3795036601,"elo":1104.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":15.2157298212,"imp":23.562312552,"elo":1126.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":39.5598749075,"imp":22.9291966748,"elo":1163.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":80.1299502447,"imp":22.8008853546,"elo":1171.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":161.6248068008,"imp":22.8063239404,"elo":1174.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":243.343699307,"imp":22.828636981,"elo":1173.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":327.3662383914,"imp":22.8132527114,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.7684699326,"imp":20.0300075197,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.8845759634,"imp":17.6938413591,"elo":1372.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":65.188545506,"imp":16.1866207402,"elo":1443.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":135.7029218302,"imp":15.5724691733,"elo":1464.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":345.9238137878,"imp":14.9981556978,"elo":1491.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":694.3474135019,"imp":14.726329583,"elo":1503.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1382.0545973419,"imp":14.6397928319,"elo":1506.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2071.1098936624,"imp":14.712127952,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2774.1715903227,"imp":14.7522123019,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.6257151642,"imp":12.882672047,"elo":1549.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.8594644787,"imp":12.9020325142,"elo":1548.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":46.8848802687,"imp":12.813789598,"elo":1546.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":98.2945586224,"imp":12.8400893604,"elo":1555.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":251.1485792262,"imp":12.6962531002,"elo":1566.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":507.8298692924,"imp":12.5633190641,"elo":1567.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1024.2615862152,"imp":11.980726784,"elo":1581.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1537.2603273066,"imp":11.6091395544,"elo":1589.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2063.6226141883,"imp":10.8967314441,"elo":1592.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.6228654067,"imp":21.506525027,"elo":1262.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":47.8871994,"imp":21.3453920972,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":78.2444408904,"imp":20.1832197221,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":157.8326162312,"imp":19.1079008455,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":299.7184949711,"imp":18.6818734271,"elo":1331.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":746.8187102503,"imp":17.8365174611,"elo":1353.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1502.6287683848,"imp":17.243319205,"elo":1372.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3013.6303119073,"imp":16.9756679665,"elo":1382.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":4537.799646108,"imp":16.8000929006,"elo":1385.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":6067.4246083224,"imp":16.6566378644,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0990306801,"imp":17.060090492,"elo":1408.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":40.7899647401,"imp":5.5210822661,"elo":1768.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.0350786389,"imp":32.3787344399,"elo":984.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":2.0391419632,"imp":12.4006296542,"elo":1586.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":7.2637875629,"imp":19.5197783658,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":15.7913063985,"imp":19.076633263,"elo":1330.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":48.8806618329,"imp":18.4122995062,"elo":1357.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":105.9612963967,"imp":17.809015799,"elo":1388.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":268.8269097705,"imp":17.3438398565,"elo":1407.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":542.3321694006,"imp":17.034306941,"elo":1421.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1076.1837280763,"imp":16.7064601338,"elo":1431.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1633.7378956367,"imp":16.5841766915,"elo":1438.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2207.1695731088,"imp":16.4722081329,"elo":1440.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.3935958382,"imp":12.1451333903,"elo":1624.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.5186684216,"imp":10.4426577471,"elo":1640.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":403.917691534,"imp":32.4365504068,"elo":1020.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":655.6361879164,"imp":32.2876141197,"elo":1028.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1251.8865438921,"imp":31.8726351515,"elo":1035.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2072.8542359813,"imp":30.9955503487,"elo":1047.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5070.1005510261,"imp":29.2570137101,"elo":1068.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":10272.9118282161,"imp":27.7314478283,"elo":1086.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":19607.2944133693,"imp":27.1232974198,"elo":1097.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":29100.9871266142,"imp":26.6871190414,"elo":1109.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":38846.7008402763,"imp":26.2398383463,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1995725235,"imp":18.7522014591,"elo":1346.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":5.7268864593,"imp":26.3207574907,"elo":1048.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":16.2195091536,"imp":23.6856472823,"elo":1123.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":55.3202394018,"imp":21.6024291784,"elo":1182.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":125.651096343,"imp":20.7744927873,"elo":1225.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":306.1276418894,"imp":20.1172354644,"elo":1255.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":615.7158666145,"imp":19.7703133892,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1199.8979174815,"imp":19.6156788345,"elo":1273.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1834.8903114269,"imp":19.4329005601,"elo":1284.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2480.8031686863,"imp":19.3351815777,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.8759235268,"imp":20.8750034522,"elo":1213.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.5595037251,"imp":19.053218751,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":17.0856997287,"imp":18.5291491102,"elo":1342.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":32.8489180471,"imp":18.1749355385,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.2976754817,"imp":17.8936188532,"elo":1373.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":171.0781299242,"imp":17.7911654268,"elo":1381.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":341.1269358042,"imp":17.6967888276,"elo":1388.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":508.913221887,"imp":17.6132236191,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":682.6224605816,"imp":17.5856960594,"elo":1394.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":267.891665628,"imp":24.3858972642,"elo":1148.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":693.7919545668,"imp":21.8184507955,"elo":1215.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2176.9330069286,"imp":18.9297052975,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4855.081405543,"imp":17.6799849939,"elo":1362.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":11992.0699103258,"imp":16.5290826329,"elo":1419.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12550.0998837068,"imp":16.5701982089,"elo":1418.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.4779953241,"imp":30.6573879604,"elo":990.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":10.0220449928,"imp":26.6806089053,"elo":1067.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":26.9153595064,"imp":21.8804825534,"elo":1201.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":52.7158440595,"imp":19.9393504966,"elo":1279.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":131.9095207406,"imp":19.1986920554,"elo":1309.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":263.979102644,"imp":18.8213468125,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":526.7568299014,"imp":18.593272316,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":787.1603694099,"imp":18.4935358031,"elo":1324.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1053.8003845303,"imp":18.3615683455,"elo":1326.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0351934444,"x_train":250.9922109033,"imp":19.6395723532,"elo":1342.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.024809806,"x_train":4.8861388456,"imp":17.295070603,"elo":1375.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0523082602,"x_train":10.6750136791,"imp":16.8081666354,"elo":1393.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1051923231,"x_train":25.2669903942,"imp":16.5553434634,"elo":1403.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1507927975,"x_train":49.0638838885,"imp":16.4131006566,"elo":1410.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.198237584,"x_train":133.206456981,"imp":16.3518455152,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2413983074,"x_train":264.5793957685,"imp":16.3707724611,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2948037083,"x_train":539.0252673695,"imp":16.2473570494,"elo":1408.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.317649589,"x_train":810.5253697201,"imp":16.260193325,"elo":1402.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3481993779,"x_train":1087.7269134101,"imp":16.1779645054,"elo":1407.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0396842484,"x_train":1.9155810093,"imp":20.6656491574,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0753273204,"x_train":4.4587928667,"imp":20.0079904311,"elo":1316.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1295300095,"x_train":11.7223762673,"imp":19.5822174931,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1888398904,"x_train":24.3697423433,"imp":19.3248019843,"elo":1345.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2593943388,"x_train":61.9355959829,"imp":18.9868988263,"elo":1357.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3225450949,"x_train":124.7079952924,"imp":18.7984834127,"elo":1365.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3870003342,"x_train":249.4958722159,"imp":18.7259661622,"elo":1366.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4035085653,"x_train":373.7478601635,"imp":18.6892810323,"elo":1368.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4161492489,"x_train":500.3975293451,"imp":18.7185872994,"elo":1367.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0162731597,"x_train":5.4796099733,"imp":20.9805486921,"elo":1253.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0330339112,"x_train":13.9490355599,"imp":20.4975061878,"elo":1273.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0589358123,"x_train":38.1339214312,"imp":20.1244528497,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0759361533,"x_train":74.7850861491,"imp":20.0032103124,"elo":1297.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1039353704,"x_train":183.3974305409,"imp":19.8442079373,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1239174887,"x_train":366.1485755963,"imp":19.6834999267,"elo":1316.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1375888064,"x_train":725.871285674,"imp":19.5170527235,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.135225794,"x_train":1084.12220491,"imp":19.4807401023,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1523084536,"x_train":1449.1413756847,"imp":19.4397032529,"elo":1326.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2675026243,"x_train":1.7477789606,"imp":29.4778153146,"elo":992.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2918606873,"x_train":3.1400651467,"imp":27.2957858125,"elo":1028.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3821908459,"x_train":7.3396935938,"imp":23.9953600978,"elo":1099.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4421545506,"x_train":14.4314608681,"imp":22.9279464902,"elo":1138.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5921825198,"x_train":35.4438180004,"imp":22.1910011881,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.729487764,"x_train":70.3987810435,"imp":21.9306885793,"elo":1180.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8457914876,"x_train":140.7929692058,"imp":21.7221031901,"elo":1191.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8457827933,"x_train":211.1596497818,"imp":21.5976427028,"elo":1200.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8643331318,"x_train":282.3556156822,"imp":21.7838917473,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3340852103,"x_train":3.1251204631,"imp":27.6418948918,"elo":1007.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6312937965,"x_train":6.228658512,"imp":26.5341183951,"elo":1061.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1801924809,"x_train":15.6396665341,"imp":24.5187984291,"elo":1133.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7736202367,"x_train":31.5302468269,"imp":23.6654630089,"elo":1164.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.5336467582,"x_train":76.3915347501,"imp":22.7418354905,"elo":1188.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.419046747,"x_train":153.0430713803,"imp":22.3653463542,"elo":1203.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.1140070815,"x_train":308.6007823833,"imp":22.1858964506,"elo":1210.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.5060075639,"x_train":463.687419557,"imp":22.108048838,"elo":1213.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0469843044,"x_train":622.8560318368,"imp":22.1743713286,"elo":1219.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1585029211,"x_train":2.2519280723,"imp":22.0237645428,"elo":1157.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5387304774,"x_train":4.5917771836,"imp":20.0235785837,"elo":1259.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7747491802,"x_train":9.9026612043,"imp":18.6688562588,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9746279953,"x_train":18.7813971613,"imp":18.020092398,"elo":1374.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1744648066,"x_train":48.1169020642,"imp":17.4420055099,"elo":1394.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2950339303,"x_train":97.0735394653,"imp":17.2474827759,"elo":1406.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4013389398,"x_train":194.547566338,"imp":17.1370232932,"elo":1413.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5863655126,"x_train":290.5602659885,"imp":17.0606732736,"elo":1415.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2823687024,"x_train":391.2703614285,"imp":17.0772120491,"elo":1416.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.6915866654,"x_train":18.1964917538,"imp":16.2254459782,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"x_train":13.3530116618,"imp":22.9392691064,"elo":1186.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9081956315,"x_train":43.609446858,"imp":21.2645554845,"elo":1226.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8094695931,"x_train":123.2677201414,"imp":18.7260581766,"elo":1328.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9001397479,"x_train":240.5329414556,"imp":18.1737471821,"elo":1352.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.3070047996,"x_train":578.0322705513,"imp":17.622242904,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5713626606,"x_train":1205.8399034278,"imp":17.636430837,"elo":1372.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.0463659943,"x_train":2388.27785775,"imp":17.8823109886,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4271738524,"x_train":3575.4198584316,"imp":17.7863190447,"elo":1366.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"x_train":4853.4730504877,"imp":17.6703087977,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.876262961,"x_train":16.0704711344,"imp":26.1705122737,"elo":1082.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0443776494,"x_train":20.4435114187,"imp":38.3890458814,"elo":943.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3417975297,"x_train":50.7468311596,"imp":35.0877917631,"elo":999.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5993854901,"x_train":146.088248762,"imp":31.3764424518,"elo":1065.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6542696803,"x_train":175.2300293672,"imp":31.165634527,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0376682658,"x_train":0.3738939375,"imp":29.451436506,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1044595646,"x_train":2.1861266183,"imp":26.1061150853,"elo":1059.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2235814001,"x_train":7.0625783305,"imp":24.3795036601,"elo":1104.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3083081492,"x_train":15.2157298212,"imp":23.562312552,"elo":1126.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.471766492,"x_train":39.5598749075,"imp":22.9291966748,"elo":1163.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5721305279,"x_train":80.1299502447,"imp":22.8008853546,"elo":1171.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.659224899,"x_train":161.6248068008,"imp":22.8063239404,"elo":1174.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6901438372,"x_train":243.343699307,"imp":22.828636981,"elo":1173.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6730171196,"x_train":327.3662383914,"imp":22.8132527114,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7864396959,"x_train":10.7684699326,"imp":20.0300075197,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.7130532123,"x_train":24.8845759634,"imp":17.6938413591,"elo":1372.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.0407156134,"x_train":65.188545506,"imp":16.1866207402,"elo":1443.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.5693353243,"x_train":135.7029218302,"imp":15.5724691733,"elo":1464.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.5979957571,"x_train":345.9238137878,"imp":14.9981556978,"elo":1491.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.5139570657,"x_train":694.3474135019,"imp":14.726329583,"elo":1503.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.7224848724,"x_train":1382.0545973419,"imp":14.6397928319,"elo":1506.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.7145191322,"x_train":2071.1098936624,"imp":14.712127952,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.3626514212,"x_train":2774.1715903227,"imp":14.7522123019,"elo":1513.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7713317233,"x_train":5.6257151642,"imp":12.882672047,"elo":1549.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3657976323,"x_train":16.8594644787,"imp":12.9020325142,"elo":1548.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8885842429,"x_train":46.8848802687,"imp":12.813789598,"elo":1546.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.9894747323,"x_train":98.2945586224,"imp":12.8400893604,"elo":1555.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.8519951234,"x_train":251.1485792262,"imp":12.6962531002,"elo":1566.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6072809051,"x_train":507.8298692924,"imp":12.5633190641,"elo":1567.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1136602495,"x_train":1024.2615862152,"imp":11.980726784,"elo":1581.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.10648955,"x_train":1537.2603273066,"imp":11.6091395544,"elo":1589.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.9647036297,"x_train":2063.6226141883,"imp":10.8967314441,"elo":1592.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.9837292889,"x_train":13.6228654067,"imp":21.506525027,"elo":1262.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"x_train":47.8871994,"imp":21.3453920972,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":57.0488035322,"x_train":78.2444408904,"imp":20.1832197221,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":89.3976769749,"x_train":157.8326162312,"imp":19.1079008455,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":120.7151402618,"x_train":299.7184949711,"imp":18.6818734271,"elo":1331.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":191.9678402321,"x_train":746.8187102503,"imp":17.8365174611,"elo":1353.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":252.9847402475,"x_train":1502.6287683848,"imp":17.243319205,"elo":1372.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":320.0227255786,"x_train":3013.6303119073,"imp":16.9756679665,"elo":1382.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":359.8655501595,"x_train":4537.799646108,"imp":16.8000929006,"elo":1385.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":405.618626345,"x_train":6067.4246083224,"imp":16.6566378644,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1861515215,"x_train":2.0990306801,"imp":17.060090492,"elo":1408.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"x_train":40.7899647401,"imp":5.5210822661,"elo":1768.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1916685529,"x_train":1.0350786389,"imp":32.3787344399,"elo":984.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"x_train":2.0391419632,"imp":12.4006296542,"elo":1586.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2496062173,"x_train":7.2637875629,"imp":19.5197783658,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4588827195,"x_train":15.7913063985,"imp":19.076633263,"elo":1330.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9324927533,"x_train":48.8806618329,"imp":18.4122995062,"elo":1357.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.0449796984,"x_train":105.9612963967,"imp":17.809015799,"elo":1388.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.418274975,"x_train":268.8269097705,"imp":17.3438398565,"elo":1407.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.7029553903,"x_train":542.3321694006,"imp":17.034306941,"elo":1421.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.8501005988,"x_train":1076.1837280763,"imp":16.7064601338,"elo":1431.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.8904873538,"x_train":1633.7378956367,"imp":16.5841766915,"elo":1438.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":1.8937013268,"x_train":2207.1695731088,"imp":16.4722081329,"elo":1440.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"x_train":5.3935958382,"imp":12.1451333903,"elo":1624.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3846051827,"x_train":3.5186684216,"imp":10.4426577471,"elo":1640.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9719997826,"x_train":403.917691534,"imp":32.4365504068,"elo":1020.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.9132537435,"x_train":655.6361879164,"imp":32.2876141197,"elo":1028.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.44798274,"x_train":1251.8865438921,"imp":31.8726351515,"elo":1035.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.6829725644,"x_train":2072.8542359813,"imp":30.9955503487,"elo":1047.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.9995362697,"x_train":5070.1005510261,"imp":29.2570137101,"elo":1068.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0786013215,"x_train":10272.9118282161,"imp":27.7314478283,"elo":1086.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.8375490304,"x_train":19607.2944133693,"imp":27.1232974198,"elo":1097.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.5577174656,"x_train":29100.9871266142,"imp":26.6871190414,"elo":1109.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.4748027028,"x_train":38846.7008402763,"imp":26.2398383463,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0643226705,"x_train":1.1995725235,"imp":18.7522014591,"elo":1346.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1483357715,"x_train":5.7268864593,"imp":26.3207574907,"elo":1048.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3175868529,"x_train":16.2195091536,"imp":23.6856472823,"elo":1123.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6704028705,"x_train":55.3202394018,"imp":21.6024291784,"elo":1182.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8151812596,"x_train":125.651096343,"imp":20.7744927873,"elo":1225.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.208494529,"x_train":306.1276418894,"imp":20.1172354644,"elo":1255.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5100756369,"x_train":615.7158666145,"imp":19.7703133892,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.9011606861,"x_train":1199.8979174815,"imp":19.6156788345,"elo":1273.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0966186606,"x_train":1834.8903114269,"imp":19.4329005601,"elo":1284.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1753468186,"x_train":2480.8031686863,"imp":19.3351815777,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209262941,"x_train":1.8759235268,"imp":20.8750034522,"elo":1213.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4702992151,"x_train":5.5595037251,"imp":19.053218751,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8389111922,"x_train":17.0856997287,"imp":18.5291491102,"elo":1342.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8064791193,"x_train":32.8489180471,"imp":18.1749355385,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8473214228,"x_train":86.2976754817,"imp":17.8936188532,"elo":1373.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0012724874,"x_train":171.0781299242,"imp":17.7911654268,"elo":1381.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9989995929,"x_train":341.1269358042,"imp":17.6967888276,"elo":1388.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2150238563,"x_train":508.913221887,"imp":17.6132236191,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9713359475,"x_train":682.6224605816,"imp":17.5856960594,"elo":1394.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":62.4915524975,"x_train":267.891665628,"imp":24.3858972642,"elo":1148.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":128.140815719,"x_train":693.7919545668,"imp":21.8184507955,"elo":1215.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":235.7496154858,"x_train":2176.9330069286,"imp":18.9297052975,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":355.4432898564,"x_train":4855.081405543,"imp":17.6799849939,"elo":1362.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":432.5178282259,"x_train":11992.0699103258,"imp":16.5290826329,"elo":1419.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":427.5904998868,"x_train":12550.0998837068,"imp":16.5701982089,"elo":1418.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.6769281514,"x_train":3.4779953241,"imp":30.6573879604,"elo":990.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8523326353,"x_train":10.0220449928,"imp":26.6806089053,"elo":1067.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1689874481,"x_train":26.9153595064,"imp":21.8804825534,"elo":1201.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.525739026,"x_train":52.7158440595,"imp":19.9393504966,"elo":1279.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8428620046,"x_train":131.9095207406,"imp":19.1986920554,"elo":1309.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.1886115821,"x_train":263.979102644,"imp":18.8213468125,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.033782756,"x_train":526.7568299014,"imp":18.593272316,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.9700745263,"x_train":787.1603694099,"imp":18.4935358031,"elo":1324.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9325738702,"x_train":1053.8003845303,"imp":18.3615683455,"elo":1326.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_all/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..bf35d3486dc39a07df0b726a1ffc58b35b19d8ea
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_classification/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabICLv2 (default)", "RealTabPFN-2.5 (default)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "TabM (tuned + ensembled)", "iLTM (tuned + ensembled)", "LimiX (default)", "CatBoost (tuned)", "LightGBM (tuned + ensembled)", "TabDPT-Turbo (default)", "CatBoost (tuned + ensembled)", "TabM (tuned)", "TabDPT (tuned + ensembled)", "XGBoost (tuned + ensembled)", "ModernNCA (tuned)", "CatBoost (default)", "ModernNCA (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "LightGBM (tuned)", "XGBoost (tuned)", "TabSwift (default)", "TabDPT (tuned)", "BetaTabPFN (default)", "ChimeraBoost (tuned)", "EBM (tuned + ensembled)", "xRFM (tuned + ensembled)", "iLTM (tuned)", "TabM (default)", "TorchMLP (tuned + ensembled)", "xRFM (tuned)", "RealMLP (default)", "ChimeraBoost (default)", "SAP-RPT-OSS (default)", "EBM (tuned)", "EBM (default)", "TabDPT (default)", "FastaiMLP (tuned + ensembled)", "XGBoost (default)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "ModernNCA (default)", "RandomForest (tuned + ensembled)", "ExtraTrees (tuned)", "LightGBM (default)", "iLTM (default)", "FastaiMLP (tuned)", "RandomForest (tuned)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "OrionMSP (default)", "PerpetualBooster (tuned + ensembled)", "TorchMLP (default)", "TabSTAR (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "RandomForest (default)", "KNN (tuned + ensembled)", "ExtraTrees (default)", "xRFM (default)", "TabFlex (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (default)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7368420958518982, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.7631579041481018, 0.8947368264198303, 0.8157894611358643, 0.8421052694320679, 0.9210526347160339, 0.8157894611358643, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.8947368264198303, 0.9473684430122375, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9736841917037964, 0.8947368264198303, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9210526347160339, 0.8947368264198303, 0.9736841917037964, 1.0, 0.9736841917037964, 0.9473684430122375, 1.0, 1.0, 0.9736841917037964, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 1.0, 0.9736841917037964, 0.9736841917037964, 1.0, 1.0, 0.9736841917037964, 1.0], [0.2631579041481018, NaN, 0.4736842215061188, 0.6052631735801697, 0.6052631735801697, 0.5263158082962036, 0.5789473652839661, 0.6842105388641357, 0.6842105388641357, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.7368420958518982, 0.8421052694320679, 0.8947368264198303, 0.6842105388641357, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.9210526347160339, 0.8421052694320679, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.9210526347160339, 0.8684210777282715, 0.8684210777282715, 0.9473684430122375, 0.8157894611358643, 0.8684210777282715, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.9210526347160339, 0.9473684430122375, 0.8684210777282715, 0.8684210777282715, 1.0, 0.9210526347160339, 0.9473684430122375, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 1.0, 1.0], [0.18421052396297455, 0.5263158082962036, NaN, 0.5526315569877625, 0.5657894611358643, 0.6052631735801697, 0.5789473652839661, 0.5526315569877625, 0.6578947305679321, 0.7105262875556946, 0.7368420958518982, 0.7894737124443054, 0.7894737124443054, 0.7105262875556946, 0.7368420958518982, 0.8684210777282715, 0.8157894611358643, 0.7894737124443054, 0.8684210777282715, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.8684210777282715, 0.9473684430122375, 0.8157894611358643, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.7894737124443054, 0.8947368264198303, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.8684210777282715, 0.8684210777282715, 1.0, 0.9736841917037964, 0.9473684430122375, 0.8684210777282715, 1.0, 1.0, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0, 0.9736841917037964, 1.0], [0.15789473056793213, 0.3947368562221527, 0.44736841320991516, NaN, 0.4736842215061188, 0.5, 0.6578947305679321, 0.5263158082962036, 0.5526315569877625, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.8157894611358643, 0.7631579041481018, 0.7631579041481018, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.7894737124443054, 0.7368420958518982, 0.7631579041481018, 0.8157894611358643, 0.8684210777282715, 0.8947368264198303, 0.8157894611358643, 0.9736841917037964, 0.8421052694320679, 0.7894737124443054, 0.7631579041481018, 0.8421052694320679, 0.8157894611358643, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8157894611358643, 0.9210526347160339, 0.8684210777282715, 0.8157894611358643, 0.8947368264198303, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.8947368264198303, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.8421052694320679, 1.0, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 1.0], [0.1315789520740509, 0.3947368562221527, 0.43421053886413574, 0.5263158082962036, NaN, 0.5263158082962036, 0.6315789222717285, 0.5526315569877625, 0.7105262875556946, 0.6842105388641357, 0.7105262875556946, 0.6842105388641357, 0.7368420958518982, 0.6842105388641357, 0.8157894611358643, 0.8157894611358643, 0.6578947305679321, 0.8684210777282715, 0.7631579041481018, 0.6842105388641357, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.7631579041481018, 0.7894737124443054, 0.8947368264198303, 0.8421052694320679, 0.8157894611358643, 0.7631579041481018, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8684210777282715, 0.8421052694320679, 0.8157894611358643, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.8157894611358643, 0.8421052694320679, 0.9473684430122375, 0.8684210777282715, 0.9473684430122375, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 0.8421052694320679, 0.8421052694320679, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 1.0], [0.1315789520740509, 0.4736842215061188, 0.3947368562221527, 0.5, 0.4736842215061188, NaN, 0.5789473652839661, 0.6315789222717285, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.6842105388641357, 0.7631579041481018, 0.6842105388641357, 0.7368420958518982, 0.8157894611358643, 0.7894737124443054, 0.7368420958518982, 0.7105262875556946, 0.8157894611358643, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.7631579041481018, 0.8157894611358643, 0.8157894611358643, 0.8947368264198303, 0.7894737124443054, 0.7631579041481018, 0.7894737124443054, 0.7631579041481018, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.9473684430122375, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 0.8421052694320679, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.8947368264198303, 0.9736841917037964, 1.0, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0, 0.9736841917037964, 1.0], [0.15789473056793213, 0.42105263471603394, 0.42105263471603394, 0.34210526943206787, 0.3684210479259491, 0.42105263471603394, NaN, 0.5526315569877625, 0.43421053886413574, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.5789473652839661, 0.7105262875556946, 0.6842105388641357, 0.7631579041481018, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.7368420958518982, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.7105262875556946, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.7368420958518982, 0.8421052694320679, 0.7105262875556946, 0.7894737124443054, 0.8157894611358643, 0.8815789222717285, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.8421052694320679, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 0.9736841917037964, 0.8947368264198303, 0.9736841917037964, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 1.0, 0.9473684430122375, 0.9736841917037964, 0.8947368264198303, 1.0, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 1.0, 0.9473684430122375, 1.0, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0], [0.2368421107530594, 0.31578946113586426, 0.44736841320991516, 0.4736842215061188, 0.44736841320991516, 0.3684210479259491, 0.44736841320991516, NaN, 0.5, 0.5789473652839661, 0.7105262875556946, 0.6578947305679321, 0.8157894611358643, 0.6052631735801697, 0.7631579041481018, 0.7894737124443054, 0.6578947305679321, 0.7894737124443054, 0.7368420958518982, 0.6315789222717285, 0.8157894611358643, 0.6842105388641357, 0.7631579041481018, 0.6578947305679321, 0.7894737124443054, 0.8421052694320679, 0.7894737124443054, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.8157894611358643, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.8421052694320679, 0.7631579041481018, 0.8947368264198303, 0.8684210777282715, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.9210526347160339, 0.8421052694320679, 0.9210526347160339, 0.8684210777282715, 0.8421052694320679, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.7894737124443054, 0.8157894611358643, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.8684210777282715, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9210526347160339, 0.9736841917037964, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 1.0, 1.0, 1.0], [0.10526315867900848, 0.31578946113586426, 0.34210526943206787, 0.44736841320991516, 0.28947368264198303, 0.3947368562221527, 0.5657894611358643, 0.5, NaN, 0.6052631735801697, 0.7105262875556946, 0.7368420958518982, 0.6578947305679321, 0.5789473652839661, 0.7368420958518982, 0.7368420958518982, 0.6578947305679321, 0.7631579041481018, 0.7894737124443054, 0.6578947305679321, 0.7631579041481018, 0.8157894611358643, 0.8684210777282715, 0.7368420958518982, 0.6842105388641357, 0.8684210777282715, 0.7894737124443054, 0.7368420958518982, 0.7631579041481018, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.7631579041481018, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.9736841917037964, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.9473684430122375, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 1.0, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 1.0], [0.18421052396297455, 0.18421052396297455, 0.28947368264198303, 0.28947368264198303, 0.31578946113586426, 0.34210526943206787, 0.3947368562221527, 0.42105263471603394, 0.3947368562221527, NaN, 0.6578947305679321, 0.6052631735801697, 0.7105262875556946, 0.5, 0.6842105388641357, 0.7105262875556946, 0.5789473652839661, 0.7631579041481018, 0.6842105388641357, 0.5789473652839661, 0.7105262875556946, 0.6578947305679321, 0.7631579041481018, 0.6842105388641357, 0.6578947305679321, 0.7631579041481018, 0.7631579041481018, 0.6315789222717285, 0.6315789222717285, 0.7105262875556946, 0.6842105388641357, 0.7894737124443054, 0.7105262875556946, 0.7894737124443054, 0.7631579041481018, 0.8157894611358643, 0.7631579041481018, 0.8947368264198303, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.8421052694320679, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.8947368264198303, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 1.0], [0.15789473056793213, 0.18421052396297455, 0.2631579041481018, 0.2631579041481018, 0.28947368264198303, 0.3684210479259491, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.34210526943206787, NaN, 0.4736842215061188, 0.5, 0.43421053886413574, 0.5, 0.6052631735801697, 0.4736842215061188, 0.5263158082962036, 0.5789473652839661, 0.4736842215061188, 0.6315789222717285, 0.6578947305679321, 0.6315789222717285, 0.6052631735801697, 0.5526315569877625, 0.6578947305679321, 0.6842105388641357, 0.5789473652839661, 0.5789473652839661, 0.6842105388641357, 0.5789473652839661, 0.7105262875556946, 0.6578947305679321, 0.7105262875556946, 0.7105262875556946, 0.7894737124443054, 0.6842105388641357, 0.8157894611358643, 0.6578947305679321, 0.6578947305679321, 0.7368420958518982, 0.7368420958518982, 0.6842105388641357, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.7631579041481018, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.8157894611358643, 0.8421052694320679, 0.8289473652839661, 0.8947368264198303, 0.9473684430122375, 0.8421052694320679, 0.8947368264198303, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9473684430122375, 0.9605262875556946, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 1.0], [0.07894736528396606, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.31578946113586426, 0.31578946113586426, 0.31578946113586426, 0.34210526943206787, 0.2631579041481018, 0.3947368562221527, 0.5263158082962036, NaN, 0.5263158082962036, 0.42105263471603394, 0.4736842215061188, 0.6315789222717285, 0.5789473652839661, 0.5263158082962036, 0.5526315569877625, 0.5789473652839661, 0.5526315569877625, 0.6315789222717285, 0.6052631735801697, 0.6052631735801697, 0.5789473652839661, 0.6052631735801697, 0.6842105388641357, 0.6315789222717285, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.7894737124443054, 0.5789473652839661, 0.6315789222717285, 0.8421052694320679, 0.7105262875556946, 0.5789473652839661, 0.8157894611358643, 0.7105262875556946, 0.7105262875556946, 0.7368420958518982, 0.7894737124443054, 0.7631579041481018, 0.7894737124443054, 0.7368420958518982, 0.7894737124443054, 0.7105262875556946, 0.8421052694320679, 0.7894737124443054, 0.7105262875556946, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.9736841917037964, 0.8684210777282715, 0.9210526347160339, 0.9210526347160339, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 1.0, 1.0], [0.18421052396297455, 0.1315789520740509, 0.21052631735801697, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.42105263471603394, 0.18421052396297455, 0.34210526943206787, 0.28947368264198303, 0.5, 0.4736842215061188, NaN, 0.43421053886413574, 0.5, 0.5263158082962036, 0.44736841320991516, 0.44736841320991516, 0.5, 0.42105263471603394, 0.5263158082962036, 0.6052631735801697, 0.5, 0.5263158082962036, 0.6578947305679321, 0.5526315569877625, 0.6315789222717285, 0.5789473652839661, 0.5, 0.5, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.6842105388641357, 0.6052631735801697, 0.7368420958518982, 0.7368420958518982, 0.7105262875556946, 0.7368420958518982, 0.7105262875556946, 0.6578947305679321, 0.7368420958518982, 0.6842105388641357, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.7894737124443054, 0.8421052694320679, 0.8684210777282715, 0.8157894611358643, 0.8684210777282715, 0.7631579041481018, 0.7631579041481018, 0.8421052694320679, 0.8947368264198303, 0.9473684430122375, 0.8421052694320679, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 1.0], [0.1315789520740509, 0.2631579041481018, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.31578946113586426, 0.28947368264198303, 0.3947368562221527, 0.42105263471603394, 0.5, 0.5657894611358643, 0.5789473652839661, 0.5657894611358643, NaN, 0.5263158082962036, 0.6315789222717285, 0.44736841320991516, 0.5789473652839661, 0.6052631735801697, 0.44736841320991516, 0.6578947305679321, 0.6052631735801697, 0.5789473652839661, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.6842105388641357, 0.5263158082962036, 0.5526315569877625, 0.6052631735801697, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.7368420958518982, 0.6842105388641357, 0.6578947305679321, 0.7368420958518982, 0.7368420958518982, 0.6578947305679321, 0.6578947305679321, 0.7368420958518982, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.75, 0.7368420958518982, 0.8289473652839661, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8947368264198303, 0.9210526347160339, 0.8684210777282715, 0.9473684430122375, 0.9210526347160339, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.9736841917037964], [0.10526315867900848, 0.15789473056793213, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.2631579041481018, 0.31578946113586426, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.5, 0.5263158082962036, 0.5, 0.4736842215061188, NaN, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, 0.5526315569877625, 0.5263158082962036, 0.5789473652839661, 0.5263158082962036, 0.6578947305679321, 0.5789473652839661, 0.5263158082962036, 0.5263158082962036, 0.6578947305679321, 0.5789473652839661, 0.5526315569877625, 0.5789473652839661, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.7368420958518982, 0.6315789222717285, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.7105262875556946, 0.7368420958518982, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.7631579041481018, 0.8157894611358643, 0.7368420958518982, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8684210777282715, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.9210526347160339, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.8684210777282715, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0], [0.07894736528396606, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.21052631735801697, 0.2631579041481018, 0.28947368264198303, 0.3947368562221527, 0.3684210479259491, 0.4736842215061188, 0.3684210479259491, 0.5526315569877625, NaN, 0.5, 0.5789473652839661, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, 0.5, 0.6578947305679321, 0.5263158082962036, 0.5789473652839661, 0.7105262875556946, 0.6315789222717285, 0.5789473652839661, 0.5526315569877625, 0.5526315569877625, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.7368420958518982, 0.7105262875556946, 0.6842105388641357, 0.7368420958518982, 0.6578947305679321, 0.7894737124443054, 0.8157894611358643, 0.7105262875556946, 0.7631579041481018, 0.8684210777282715, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.9473684430122375, 0.8947368264198303, 0.8157894611358643, 0.9210526347160339, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964, 0.9210526347160339, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.9736841917037964, 0.9736841917037964, 0.8947368264198303, 0.9473684430122375, 1.0, 1.0], [0.07894736528396606, 0.31578946113586426, 0.18421052396297455, 0.21052631735801697, 0.34210526943206787, 0.21052631735801697, 0.34210526943206787, 0.34210526943206787, 0.34210526943206787, 0.42105263471603394, 0.5263158082962036, 0.42105263471603394, 0.5526315569877625, 0.5526315569877625, 0.4736842215061188, 0.5, NaN, 0.5, 0.44736841320991516, 0.6315789222717285, 0.5789473652839661, 0.5, 0.5789473652839661, 0.5, 0.6052631735801697, 0.5, 0.6052631735801697, 0.6710526347160339, 0.6842105388641357, 0.5263158082962036, 0.6052631735801697, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.6052631735801697, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.7105262875556946, 0.8157894611358643, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.7631579041481018, 0.6842105388641357, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.7894737124443054, 0.8684210777282715, 0.8157894611358643, 0.7631579041481018, 0.7631579041481018, 0.9210526347160339, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.8947368264198303, 0.9473684430122375, 0.8421052694320679, 0.8421052694320679, 0.9473684430122375, 0.8947368264198303, 0.9736841917037964, 1.0], [0.07894736528396606, 0.10526315867900848, 0.21052631735801697, 0.18421052396297455, 0.1315789520740509, 0.2631579041481018, 0.31578946113586426, 0.21052631735801697, 0.2368421107530594, 0.2368421107530594, 0.4736842215061188, 0.4736842215061188, 0.5526315569877625, 0.42105263471603394, 0.5526315569877625, 0.42105263471603394, 0.5, NaN, 0.5, 0.5, 0.5263158082962036, 0.5131579041481018, 0.6315789222717285, 0.5789473652839661, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.5526315569877625, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.6052631735801697, 0.6315789222717285, 0.7368420958518982, 0.5526315569877625, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.8157894611358643, 0.7368420958518982, 0.6842105388641357, 0.7631579041481018, 0.6842105388641357, 0.7368420958518982, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.8421052694320679, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.9210526347160339, 0.8157894611358643, 0.8947368264198303, 0.8947368264198303, 0.9473684430122375, 0.8684210777282715, 0.8684210777282715, 0.8947368264198303, 0.8421052694320679, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 1.0], [0.05263157933950424, 0.21052631735801697, 0.1315789520740509, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.2631579041481018, 0.2631579041481018, 0.21052631735801697, 0.31578946113586426, 0.42105263471603394, 0.44736841320991516, 0.5, 0.3947368562221527, 0.44736841320991516, 0.5263158082962036, 0.5526315569877625, 0.5, NaN, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.5789473652839661, 0.5526315569877625, 0.6447368264198303, 0.5263158082962036, 0.6052631735801697, 0.6052631735801697, 0.6052631735801697, 0.6315789222717285, 0.6315789222717285, 0.6578947305679321, 0.5263158082962036, 0.6052631735801697, 0.6842105388641357, 0.6315789222717285, 0.5789473652839661, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.7105262875556946, 0.6842105388641357, 0.7368420958518982, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.6842105388641357, 0.8421052694320679, 0.7368420958518982, 0.6842105388641357, 0.7894737124443054, 0.7105262875556946, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.9736841917037964, 0.8421052694320679, 0.9210526347160339, 0.8684210777282715, 0.8684210777282715, 0.8684210777282715, 0.8157894611358643, 0.8421052694320679, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964], [0.07894736528396606, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.31578946113586426, 0.18421052396297455, 0.2368421107530594, 0.3684210479259491, 0.34210526943206787, 0.42105263471603394, 0.5263158082962036, 0.42105263471603394, 0.5789473652839661, 0.5526315569877625, 0.4736842215061188, 0.5263158082962036, 0.3684210479259491, 0.5, 0.44736841320991516, NaN, 0.5263158082962036, 0.4736842215061188, 0.5263158082962036, 0.4736842215061188, 0.5263158082962036, 0.5263158082962036, 0.5263158082962036, 0.5263158082962036, 0.7105262875556946, 0.5526315569877625, 0.5263158082962036, 0.5789473652839661, 0.5789473652839661, 0.6578947305679321, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.6315789222717285, 0.5263158082962036, 0.6578947305679321, 0.6578947305679321, 0.7105262875556946, 0.7631579041481018, 0.6578947305679321, 0.6842105388641357, 0.6842105388641357, 0.7105262875556946, 0.8157894611358643, 0.7105262875556946, 0.8157894611358643, 0.7631579041481018, 0.7368420958518982, 0.7894737124443054, 0.7368420958518982, 0.7894737124443054, 0.7631579041481018, 0.9473684430122375, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375, 0.9473684430122375, 0.8947368264198303, 0.9736841917037964, 0.9473684430122375, 0.9210526347160339, 0.8684210777282715, 0.8684210777282715, 0.8947368264198303, 0.8947368264198303, 0.9473684430122375, 1.0], [0.10526315867900848, 0.07894736528396606, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.2368421107530594, 0.2631579041481018, 0.18421052396297455, 0.2368421107530594, 0.28947368264198303, 0.3684210479259491, 0.44736841320991516, 0.4736842215061188, 0.34210526943206787, 0.42105263471603394, 0.5263158082962036, 0.42105263471603394, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, NaN, 0.4736842215061188, 0.5526315569877625, 0.5, 0.5526315569877625, 0.5789473652839661, 0.6842105388641357, 0.5263158082962036, 0.5, 0.4736842215061188, 0.6315789222717285, 0.6315789222717285, 0.5526315569877625, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.7105262875556946, 0.7105262875556946, 0.6315789222717285, 0.6842105388641357, 0.7105262875556946, 0.6842105388641357, 0.7105262875556946, 0.8684210777282715, 0.6973684430122375, 0.8157894611358643, 0.7894737124443054, 0.8421052694320679, 0.7894737124443054, 0.9736841917037964, 0.8421052694320679, 0.7631579041481018, 0.8947368264198303, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.7894737124443054, 0.8947368264198303, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375, 0.8684210777282715, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.9473684430122375, 0.8684210777282715, 0.9473684430122375, 0.9210526347160339, 1.0], [0.10526315867900848, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.31578946113586426, 0.31578946113586426, 0.18421052396297455, 0.34210526943206787, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.4736842215061188, 0.5, 0.5, 0.4868420958518982, 0.3947368562221527, 0.5263158082962036, 0.5263158082962036, NaN, 0.5789473652839661, 0.43421053886413574, 0.5263158082962036, 0.5526315569877625, 0.5, 0.5526315569877625, 0.5789473652839661, 0.5789473652839661, 0.5526315569877625, 0.5263158082962036, 0.5789473652839661, 0.5789473652839661, 0.6052631735801697, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.6315789222717285, 0.6842105388641357, 0.6710526347160339, 0.7105262875556946, 0.7631579041481018, 0.7105262875556946, 0.8421052694320679, 0.7368420958518982, 0.7368420958518982, 0.7894737124443054, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.9210526347160339, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.8684210777282715, 0.9210526347160339, 0.9210526347160339, 1.0], [0.1315789520740509, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.21052631735801697, 0.28947368264198303, 0.2368421107530594, 0.1315789520740509, 0.2368421107530594, 0.3684210479259491, 0.3947368562221527, 0.5, 0.42105263471603394, 0.34210526943206787, 0.34210526943206787, 0.42105263471603394, 0.3684210479259491, 0.42105263471603394, 0.4736842215061188, 0.44736841320991516, 0.42105263471603394, NaN, 0.4736842215061188, 0.4736842215061188, 0.44736841320991516, 0.4736842215061188, 0.5526315569877625, 0.5, 0.42105263471603394, 0.5789473652839661, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.5526315569877625, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.7368420958518982, 0.6578947305679321, 0.6315789222717285, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.7368420958518982, 0.7894737124443054, 0.7368420958518982, 0.7105262875556946, 0.8157894611358643, 0.6842105388641357, 0.7631579041481018, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.9473684430122375, 0.8421052694320679, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0], [0.1315789520740509, 0.07894736528396606, 0.07894736528396606, 0.10526315867900848, 0.2368421107530594, 0.15789473056793213, 0.34210526943206787, 0.34210526943206787, 0.2631579041481018, 0.31578946113586426, 0.3947368562221527, 0.3947368562221527, 0.4736842215061188, 0.3947368562221527, 0.42105263471603394, 0.4736842215061188, 0.5, 0.42105263471603394, 0.44736841320991516, 0.5263158082962036, 0.5, 0.5657894611358643, 0.5263158082962036, NaN, 0.5263158082962036, 0.6052631735801697, 0.5263158082962036, 0.5, 0.5526315569877625, 0.5789473652839661, 0.5526315569877625, 0.6052631735801697, 0.5526315569877625, 0.5263158082962036, 0.5263158082962036, 0.6052631735801697, 0.6578947305679321, 0.7105262875556946, 0.6052631735801697, 0.5789473652839661, 0.6052631735801697, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.7368420958518982, 0.8157894611358643, 0.7368420958518982, 0.7105262875556946, 0.8421052694320679, 0.7631579041481018, 0.8157894611358643, 0.7105262875556946, 0.7105262875556946, 0.7368420958518982, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.8552631735801697, 0.8421052694320679, 0.8947368264198303, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.9736841917037964], [0.10526315867900848, 0.07894736528396606, 0.1315789520740509, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.28947368264198303, 0.21052631735801697, 0.31578946113586426, 0.34210526943206787, 0.44736841320991516, 0.42105263471603394, 0.34210526943206787, 0.3684210479259491, 0.4736842215061188, 0.42105263471603394, 0.3947368562221527, 0.44736841320991516, 0.3552631437778473, 0.4736842215061188, 0.44736841320991516, 0.4736842215061188, 0.5263158082962036, 0.4736842215061188, NaN, 0.4736842215061188, 0.4736842215061188, 0.5526315569877625, 0.4736842215061188, 0.3947368562221527, 0.5789473652839661, 0.6315789222717285, 0.6052631735801697, 0.6052631735801697, 0.5789473652839661, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.8157894611358643, 0.6315789222717285, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.7105262875556946, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.7631579041481018, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.7105262875556946, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375, 0.8421052694320679, 0.9736841917037964, 0.9473684430122375, 1.0], [0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.10526315867900848, 0.15789473056793213, 0.21052631735801697, 0.15789473056793213, 0.1315789520740509, 0.2368421107530594, 0.34210526943206787, 0.3947368562221527, 0.44736841320991516, 0.3947368562221527, 0.4736842215061188, 0.28947368264198303, 0.5, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, 0.42105263471603394, 0.44736841320991516, 0.5526315569877625, 0.3947368562221527, 0.5263158082962036, NaN, 0.4736842215061188, 0.5, 0.5, 0.5789473652839661, 0.6052631735801697, 0.6578947305679321, 0.6052631735801697, 0.5263158082962036, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.6315789222717285, 0.6842105388641357, 0.6315789222717285, 0.6842105388641357, 0.7105262875556946, 0.6842105388641357, 0.7105262875556946, 0.7631579041481018, 0.7631579041481018, 0.7105262875556946, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.8421052694320679, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 1.0], [0.15789473056793213, 0.07894736528396606, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.21052631735801697, 0.21052631735801697, 0.2368421107530594, 0.31578946113586426, 0.31578946113586426, 0.3684210479259491, 0.31578946113586426, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.4736842215061188, 0.31578946113586426, 0.5, 0.5263158082962036, 0.4736842215061188, 0.5263158082962036, 0.5263158082962036, NaN, 0.5263158082962036, 0.4736842215061188, 0.5, 0.5526315569877625, 0.5526315569877625, 0.6315789222717285, 0.5526315569877625, 0.5526315569877625, 0.5789473652839661, 0.6578947305679321, 0.6315789222717285, 0.6315789222717285, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.6315789222717285, 0.6842105388641357, 0.8421052694320679, 0.6842105388641357, 0.8157894611358643, 0.7368420958518982, 0.7894737124443054, 0.8157894611358643, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.7631579041481018, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.9210526347160339, 0.8684210777282715, 0.9473684430122375, 0.9210526347160339, 1.0], [0.1315789520740509, 0.21052631735801697, 0.1315789520740509, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.15789473056793213, 0.28947368264198303, 0.2631579041481018, 0.3684210479259491, 0.42105263471603394, 0.3684210479259491, 0.42105263471603394, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, 0.32894736528396606, 0.44736841320991516, 0.3947368562221527, 0.4736842215061188, 0.4736842215061188, 0.44736841320991516, 0.44736841320991516, 0.5, 0.44736841320991516, 0.5, 0.4736842215061188, NaN, 0.42105263471603394, 0.5263158082962036, 0.5, 0.5789473652839661, 0.6052631735801697, 0.5263158082962036, 0.5263158082962036, 0.5, 0.6052631735801697, 0.5789473652839661, 0.6842105388641357, 0.6578947305679321, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.7631579041481018, 0.6052631735801697, 0.6052631735801697, 0.6578947305679321, 0.6578947305679321, 0.6315789222717285, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.6578947305679321, 0.7631579041481018, 0.7631579041481018, 0.8947368264198303, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8157894611358643, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.9736841917037964, 0.8421052694320679, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.9736841917037964, 1.0], [0.05263157933950424, 0.18421052396297455, 0.10526315867900848, 0.2368421107530594, 0.2368421107530594, 0.18421052396297455, 0.2631579041481018, 0.28947368264198303, 0.2368421107530594, 0.3684210479259491, 0.42105263471603394, 0.3947368562221527, 0.5, 0.44736841320991516, 0.44736841320991516, 0.44736841320991516, 0.31578946113586426, 0.44736841320991516, 0.3947368562221527, 0.28947368264198303, 0.5, 0.42105263471603394, 0.5, 0.44736841320991516, 0.5263158082962036, 0.5, 0.5263158082962036, 0.5789473652839661, NaN, 0.44736841320991516, 0.5526315569877625, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.5263158082962036, 0.5263158082962036, 0.5789473652839661, 0.5526315569877625, 0.6052631735801697, 0.5526315569877625, 0.5526315569877625, 0.6315789222717285, 0.8026315569877625, 0.5526315569877625, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.7105262875556946, 0.5789473652839661, 0.7631579041481018, 0.6315789222717285, 0.7631579041481018, 0.7631579041481018, 0.7105262875556946, 0.7631579041481018, 0.7631579041481018, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.9473684430122375, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.9210526347160339, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964], [0.05263157933950424, 0.15789473056793213, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.28947368264198303, 0.31578946113586426, 0.31578946113586426, 0.5, 0.3947368562221527, 0.42105263471603394, 0.44736841320991516, 0.4736842215061188, 0.4736842215061188, 0.3684210479259491, 0.44736841320991516, 0.5263158082962036, 0.42105263471603394, 0.5789473652839661, 0.42105263471603394, 0.6052631735801697, 0.42105263471603394, 0.5, 0.4736842215061188, 0.5526315569877625, NaN, 0.5263158082962036, 0.5526315569877625, 0.5263158082962036, 0.5263158082962036, 0.4736842215061188, 0.5789473652839661, 0.5526315569877625, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.7368420958518982, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.7368420958518982, 0.7631579041481018, 0.7368420958518982, 0.7105262875556946, 0.7631579041481018, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.7894737124443054, 0.8947368264198303, 0.7631579041481018, 0.8157894611358643, 0.8947368264198303, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 1.0], [0.1315789520740509, 0.07894736528396606, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.28947368264198303, 0.18421052396297455, 0.21052631735801697, 0.31578946113586426, 0.42105263471603394, 0.31578946113586426, 0.34210526943206787, 0.34210526943206787, 0.3947368562221527, 0.34210526943206787, 0.3947368562221527, 0.3947368562221527, 0.3684210479259491, 0.4736842215061188, 0.3684210479259491, 0.44736841320991516, 0.42105263471603394, 0.44736841320991516, 0.42105263471603394, 0.3947368562221527, 0.44736841320991516, 0.5, 0.44736841320991516, 0.4736842215061188, NaN, 0.6315789222717285, 0.5263158082962036, 0.4736842215061188, 0.5, 0.5789473652839661, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.6315789222717285, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.7105262875556946, 0.7105262875556946, 0.6578947305679321, 0.6315789222717285, 0.6315789222717285, 0.7105262875556946, 0.6842105388641357, 0.7894737124443054, 0.7368420958518982, 0.7368420958518982, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.7631579041481018, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.9473684430122375, 0.9210526347160339, 1.0], [0.07894736528396606, 0.15789473056793213, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.2368421107530594, 0.15789473056793213, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.3684210479259491, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.3947368562221527, 0.34210526943206787, 0.42105263471603394, 0.3684210479259491, 0.4736842215061188, 0.42105263471603394, 0.3947368562221527, 0.3684210479259491, 0.34210526943206787, 0.44736841320991516, 0.42105263471603394, 0.44736841320991516, 0.44736841320991516, 0.3684210479259491, NaN, 0.5, 0.5, 0.5, 0.5789473652839661, 0.5789473652839661, 0.6315789222717285, 0.5, 0.6578947305679321, 0.7631579041481018, 0.7894737124443054, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.7894737124443054, 0.7105262875556946, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.7631579041481018, 0.7894737124443054, 0.7631579041481018, 0.7631579041481018, 0.8684210777282715, 0.8421052694320679, 0.7894737124443054, 0.8684210777282715, 0.8157894611358643, 0.8947368264198303, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9736841917037964, 0.9473684430122375, 1.0], [0.07894736528396606, 0.07894736528396606, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.34210526943206787, 0.42105263471603394, 0.34210526943206787, 0.3684210479259491, 0.3684210479259491, 0.3684210479259491, 0.3947368562221527, 0.3684210479259491, 0.4736842215061188, 0.42105263471603394, 0.44736841320991516, 0.42105263471603394, 0.34210526943206787, 0.44736841320991516, 0.3947368562221527, 0.3947368562221527, 0.3684210479259491, 0.3947368562221527, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, 0.5, NaN, 0.42105263471603394, 0.5526315569877625, 0.5263158082962036, 0.6315789222717285, 0.6052631735801697, 0.44736841320991516, 0.5526315569877625, 0.5263158082962036, 0.5789473652839661, 0.5789473652839661, 0.6578947305679321, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.7368420958518982, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.7368420958518982, 0.7368420958518982, 0.8684210777282715, 0.7631579041481018, 0.8421052694320679, 0.7894737124443054, 0.8157894611358643, 0.8947368264198303, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.8947368264198303, 0.9473684430122375, 0.9736841917037964, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964], [0.07894736528396606, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.18421052396297455, 0.2368421107530594, 0.1184210553765297, 0.21052631735801697, 0.15789473056793213, 0.21052631735801697, 0.28947368264198303, 0.3684210479259491, 0.31578946113586426, 0.31578946113586426, 0.2631579041481018, 0.34210526943206787, 0.31578946113586426, 0.2631579041481018, 0.3947368562221527, 0.34210526943206787, 0.3947368562221527, 0.42105263471603394, 0.34210526943206787, 0.4736842215061188, 0.3947368562221527, 0.4736842215061188, 0.44736841320991516, 0.4736842215061188, 0.3947368562221527, 0.4736842215061188, 0.5263158082962036, 0.5, 0.5789473652839661, NaN, 0.5, 0.5789473652839661, 0.6315789222717285, 0.5789473652839661, 0.4736842215061188, 0.6578947305679321, 0.6052631735801697, 0.7105262875556946, 0.6315789222717285, 0.7368420958518982, 0.5526315569877625, 0.6578947305679321, 0.7368420958518982, 0.7105262875556946, 0.7894737124443054, 0.8157894611358643, 0.7105262875556946, 0.6973684430122375, 0.7631579041481018, 0.7894737124443054, 0.7631579041481018, 0.7631579041481018, 0.7368420958518982, 0.7631579041481018, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964, 1.0], [0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.2368421107530594, 0.28947368264198303, 0.15789473056793213, 0.3947368562221527, 0.34210526943206787, 0.3684210479259491, 0.34210526943206787, 0.3947368562221527, 0.44736841320991516, 0.31578946113586426, 0.42105263471603394, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.4736842215061188, 0.42105263471603394, 0.34210526943206787, 0.44736841320991516, 0.4736842215061188, 0.4736842215061188, 0.5263158082962036, 0.5, 0.5, 0.44736841320991516, 0.5, NaN, 0.5789473652839661, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.5789473652839661, 0.5394737124443054, 0.6315789222717285, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.6052631735801697, 0.7105262875556946, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.8157894611358643, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.7894737124443054, 0.8421052694320679, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964, 0.9473684430122375], [0.05263157933950424, 0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.28947368264198303, 0.2631579041481018, 0.31578946113586426, 0.2631579041481018, 0.2631579041481018, 0.34210526943206787, 0.28947368264198303, 0.3684210479259491, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.3684210479259491, 0.3684210479259491, 0.42105263471603394, 0.5, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, 0.42105263471603394, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, NaN, 0.5526315569877625, 0.44736841320991516, 0.44736841320991516, 0.5, 0.4736842215061188, 0.4736842215061188, 0.6315789222717285, 0.6052631735801697, 0.6578947305679321, 0.7368420958518982, 0.6578947305679321, 0.6052631735801697, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.7631579041481018, 0.9473684430122375, 0.7631579041481018, 0.8421052694320679, 0.8684210777282715, 0.7894737124443054, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964], [0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.2368421107530594, 0.31578946113586426, 0.42105263471603394, 0.2631579041481018, 0.2631579041481018, 0.2631579041481018, 0.28947368264198303, 0.31578946113586426, 0.2631579041481018, 0.42105263471603394, 0.34210526943206787, 0.31578946113586426, 0.31578946113586426, 0.28947368264198303, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.34210526943206787, 0.3947368562221527, 0.42105263471603394, 0.44736841320991516, 0.3684210479259491, 0.42105263471603394, 0.3684210479259491, 0.3684210479259491, 0.44736841320991516, 0.44736841320991516, NaN, 0.5, 0.42105263471603394, 0.5789473652839661, 0.5263158082962036, 0.44736841320991516, 0.5263158082962036, 0.5789473652839661, 0.5, 0.6315789222717285, 0.7105262875556946, 0.5789473652839661, 0.6842105388641357, 0.7368420958518982, 0.6315789222717285, 0.6842105388641357, 0.6578947305679321, 0.7368420958518982, 0.7105262875556946, 0.7368420958518982, 0.7894737124443054, 0.6842105388641357, 0.8684210777282715, 0.7894737124443054, 0.7631579041481018, 0.8421052694320679, 0.8157894611358643, 0.9210526347160339, 0.9210526347160339, 0.9736841917037964, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964, 1.0], [0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.28947368264198303, 0.2631579041481018, 0.2631579041481018, 0.31578946113586426, 0.3684210479259491, 0.2631579041481018, 0.34210526943206787, 0.3684210479259491, 0.28947368264198303, 0.34210526943206787, 0.2368421107530594, 0.28947368264198303, 0.31578946113586426, 0.3684210479259491, 0.3684210479259491, 0.42105263471603394, 0.44736841320991516, 0.3947368562221527, 0.31578946113586426, 0.3684210479259491, 0.3947368562221527, 0.42105263471603394, 0.44736841320991516, 0.5526315569877625, 0.5, NaN, 0.44736841320991516, 0.5, 0.5526315569877625, 0.5, 0.6315789222717285, 0.5789473652839661, 0.6315789222717285, 0.6578947305679321, 0.6052631735801697, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.6052631735801697, 0.6578947305679321, 0.7368420958518982, 0.6842105388641357, 0.7105262875556946, 0.7631579041481018, 0.7105262875556946, 0.7631579041481018, 0.8947368264198303, 0.8421052694320679, 0.7894737124443054, 0.9210526347160339, 0.7894737124443054, 0.8947368264198303, 0.8421052694320679, 0.8157894611358643, 0.9473684430122375, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 1.0], [0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.21052631735801697, 0.2368421107530594, 0.34210526943206787, 0.28947368264198303, 0.2631579041481018, 0.31578946113586426, 0.28947368264198303, 0.2631579041481018, 0.3684210479259491, 0.18421052396297455, 0.3684210479259491, 0.4736842215061188, 0.28947368264198303, 0.34210526943206787, 0.2631579041481018, 0.3947368562221527, 0.18421052396297455, 0.31578946113586426, 0.3684210479259491, 0.31578946113586426, 0.3947368562221527, 0.3684210479259491, 0.31578946113586426, 0.5, 0.5526315569877625, 0.5263158082962036, 0.44736841320991516, 0.5526315569877625, 0.5789473652839661, 0.5526315569877625, NaN, 0.6052631735801697, 0.5789473652839661, 0.5526315569877625, 0.5526315569877625, 0.6315789222717285, 0.5789473652839661, 0.5526315569877625, 0.5789473652839661, 0.6315789222717285, 0.6842105388641357, 0.5789473652839661, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.8157894611358643, 0.7368420958518982, 0.7105262875556946, 0.7894737124443054, 0.8421052694320679, 0.7631579041481018, 0.7894737124443054, 0.7368420958518982, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.7894737124443054, 0.9736841917037964, 0.9473684430122375, 1.0], [0.10526315867900848, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.1315789520740509, 0.2368421107530594, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.21052631735801697, 0.34210526943206787, 0.28947368264198303, 0.28947368264198303, 0.34210526943206787, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.2631579041481018, 0.34210526943206787, 0.34210526943206787, 0.3684210479259491, 0.31578946113586426, 0.34210526943206787, 0.42105263471603394, 0.3684210479259491, 0.3684210479259491, 0.3947368562221527, 0.34210526943206787, 0.44736841320991516, 0.3947368562221527, 0.3684210479259491, 0.34210526943206787, 0.44736841320991516, 0.34210526943206787, 0.44736841320991516, 0.5, 0.42105263471603394, 0.5, 0.3947368562221527, NaN, 0.42105263471603394, 0.5, 0.5526315569877625, 0.6052631735801697, 0.4736842215061188, 0.5789473652839661, 0.6052631735801697, 0.5789473652839661, 0.5526315569877625, 0.6842105388641357, 0.6052631735801697, 0.6052631735801697, 0.7105262875556946, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.6842105388641357, 0.7894737124443054, 0.7368420958518982, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8947368264198303, 1.0], [0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.2631579041481018, 0.2631579041481018, 0.34210526943206787, 0.2631579041481018, 0.28947368264198303, 0.21052631735801697, 0.34210526943206787, 0.31578946113586426, 0.28947368264198303, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.31578946113586426, 0.31578946113586426, 0.31578946113586426, 0.42105263471603394, 0.44736841320991516, 0.2631579041481018, 0.34210526943206787, 0.2368421107530594, 0.4736842215061188, 0.3947368562221527, 0.42105263471603394, 0.5263158082962036, 0.4736842215061188, 0.44736841320991516, 0.42105263471603394, 0.5789473652839661, NaN, 0.5, 0.6052631735801697, 0.5789473652839661, 0.5263158082962036, 0.5526315569877625, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.5789473652839661, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.7631579041481018, 0.7631579041481018, 0.7368420958518982, 0.7631579041481018, 0.8157894611358643, 0.8421052694320679, 0.7631579041481018, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.9473684430122375, 1.0], [0.07894736528396606, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.18421052396297455, 0.2631579041481018, 0.21052631735801697, 0.2631579041481018, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.3684210479259491, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.28947368264198303, 0.31578946113586426, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.31578946113586426, 0.21052631735801697, 0.42105263471603394, 0.28947368264198303, 0.46052631735801697, 0.5263158082962036, 0.5526315569877625, 0.5, 0.44736841320991516, 0.5, 0.5, NaN, 0.5526315569877625, 0.6315789222717285, 0.4736842215061188, 0.6315789222717285, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.5789473652839661, 0.6842105388641357, 0.7105262875556946, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.7368420958518982, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.8947368264198303, 0.9473684430122375, 0.9473684430122375, 1.0], [0.02631578966975212, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.31578946113586426, 0.2368421107530594, 0.31578946113586426, 0.34210526943206787, 0.28947368264198303, 0.28947368264198303, 0.18421052396297455, 0.31578946113586426, 0.2631579041481018, 0.2368421107530594, 0.31578946113586426, 0.31578946113586426, 0.28947368264198303, 0.31578946113586426, 0.31578946113586426, 0.31578946113586426, 0.3684210479259491, 0.34210526943206787, 0.19736842811107635, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.42105263471603394, 0.3684210479259491, 0.3684210479259491, 0.3684210479259491, 0.4736842215061188, 0.3684210479259491, 0.44736841320991516, 0.44736841320991516, 0.3947368562221527, 0.44736841320991516, NaN, 0.5263158082962036, 0.5, 0.44736841320991516, 0.5, 0.6052631735801697, 0.5526315569877625, 0.5789473652839661, 0.6052631735801697, 0.5789473652839661, 0.7105262875556946, 0.6052631735801697, 0.6315789222717285, 0.6578947305679321, 0.7631579041481018, 0.7105262875556946, 0.7105262875556946, 0.7105262875556946, 0.7894737124443054, 0.8157894611358643, 0.7105262875556946, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.7105262875556946, 0.7368420958518982, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.9736841917037964], [0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.05263157933950424, 0.10526315867900848, 0.10526315867900848, 0.18421052396297455, 0.10526315867900848, 0.21052631735801697, 0.21052631735801697, 0.18421052396297455, 0.34210526943206787, 0.2631579041481018, 0.2368421107530594, 0.28947368264198303, 0.2631579041481018, 0.31578946113586426, 0.34210526943206787, 0.28947368264198303, 0.32894736528396606, 0.2368421107530594, 0.2631579041481018, 0.28947368264198303, 0.28947368264198303, 0.31578946113586426, 0.2368421107530594, 0.44736841320991516, 0.3684210479259491, 0.28947368264198303, 0.28947368264198303, 0.34210526943206787, 0.2631579041481018, 0.28947368264198303, 0.3947368562221527, 0.42105263471603394, 0.42105263471603394, 0.3684210479259491, 0.3947368562221527, 0.42105263471603394, 0.3684210479259491, 0.4736842215061188, NaN, 0.44736841320991516, 0.5526315569877625, 0.5789473652839661, 0.6052631735801697, 0.5526315569877625, 0.5789473652839661, 0.5263158082962036, 0.6052631735801697, 0.7368420958518982, 0.6052631735801697, 0.6315789222717285, 0.6315789222717285, 0.6842105388641357, 0.7368420958518982, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.9736841917037964, 0.7368420958518982, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.8684210777282715, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964], [0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.07894736528396606, 0.1315789520740509, 0.18421052396297455, 0.2631579041481018, 0.21052631735801697, 0.2631579041481018, 0.18421052396297455, 0.1315789520740509, 0.2631579041481018, 0.18421052396297455, 0.2631579041481018, 0.31578946113586426, 0.1315789520740509, 0.28947368264198303, 0.2631579041481018, 0.2631579041481018, 0.31578946113586426, 0.2368421107530594, 0.15789473056793213, 0.3947368562221527, 0.3947368562221527, 0.31578946113586426, 0.28947368264198303, 0.34210526943206787, 0.3947368562221527, 0.44736841320991516, 0.34210526943206787, 0.34210526943206787, 0.5, 0.3684210479259491, 0.42105263471603394, 0.5263158082962036, 0.4736842215061188, 0.5263158082962036, 0.5, 0.5526315569877625, NaN, 0.5263158082962036, 0.6315789222717285, 0.6052631735801697, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.7368420958518982, 0.6052631735801697, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.6842105388641357, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.7631579041481018, 0.7631579041481018, 0.7894737124443054, 0.7631579041481018, 0.7368420958518982, 0.7894737124443054, 0.7631579041481018, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375], [0.02631578966975212, 0.10526315867900848, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.28947368264198303, 0.18421052396297455, 0.21052631735801697, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.31578946113586426, 0.30263158679008484, 0.2368421107530594, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.3947368562221527, 0.34210526943206787, 0.28947368264198303, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.34210526943206787, 0.31578946113586426, 0.2631579041481018, 0.3684210479259491, 0.34210526943206787, 0.44736841320991516, 0.42105263471603394, 0.44736841320991516, 0.3684210479259491, 0.5526315569877625, 0.44736841320991516, 0.4736842215061188, NaN, 0.5394737124443054, 0.6315789222717285, 0.5789473652839661, 0.6052631735801697, 0.5657894611358643, 0.6315789222717285, 0.5526315569877625, 0.5789473652839661, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.6842105388641357, 0.8421052694320679, 0.7631579041481018, 0.7894737124443054, 0.9210526347160339, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964], [0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.02631578966975212, 0.18421052396297455, 0.2368421107530594, 0.28947368264198303, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.18421052396297455, 0.31578946113586426, 0.28947368264198303, 0.18421052396297455, 0.28947368264198303, 0.2631579041481018, 0.2631579041481018, 0.34210526943206787, 0.28947368264198303, 0.18421052396297455, 0.34210526943206787, 0.3684210479259491, 0.34210526943206787, 0.3684210479259491, 0.2631579041481018, 0.31578946113586426, 0.2631579041481018, 0.3947368562221527, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.42105263471603394, 0.3947368562221527, 0.3947368562221527, 0.34210526943206787, 0.5, 0.42105263471603394, 0.3684210479259491, 0.46052631735801697, NaN, 0.5, 0.5789473652839661, 0.6973684430122375, 0.5263158082962036, 0.5526315569877625, 0.5, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.6578947305679321, 0.6315789222717285, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.7894737124443054, 0.8157894611358643, 0.7631579041481018, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.9210526347160339, 1.0], [0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.07894736528396606, 0.07894736528396606, 0.07894736528396606, 0.10526315867900848, 0.10526315867900848, 0.2368421107530594, 0.15789473056793213, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.31578946113586426, 0.2368421107530594, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.28947368264198303, 0.18421052396297455, 0.31578946113586426, 0.31578946113586426, 0.2631579041481018, 0.34210526943206787, 0.28947368264198303, 0.34210526943206787, 0.3684210479259491, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.28947368264198303, 0.3947368562221527, 0.42105263471603394, 0.31578946113586426, 0.3684210479259491, 0.42105263471603394, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.3684210479259491, 0.5, NaN, 0.5789473652839661, 0.5, 0.5, 0.5789473652839661, 0.5526315569877625, 0.6315789222717285, 0.7368420958518982, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.8157894611358643, 0.7894737124443054, 0.7368420958518982, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7105262875556946, 0.7631579041481018, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.9736841917037964], [0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.07894736528396606, 0.21052631735801697, 0.15789473056793213, 0.21052631735801697, 0.1315789520740509, 0.2631579041481018, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.2631579041481018, 0.28947368264198303, 0.15789473056793213, 0.2631579041481018, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.2368421107530594, 0.21052631735801697, 0.3684210479259491, 0.42105263471603394, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.2631579041481018, 0.21052631735801697, 0.28947368264198303, 0.28947368264198303, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.44736841320991516, 0.3684210479259491, 0.34210526943206787, 0.44736841320991516, 0.44736841320991516, 0.3947368562221527, 0.42105263471603394, 0.42105263471603394, 0.42105263471603394, NaN, 0.5789473652839661, 0.5, 0.5, 0.5526315569877625, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6052631735801697, 0.7368420958518982, 0.7631579041481018, 0.7105262875556946, 0.7105262875556946, 0.8684210777282715, 0.7894737124443054, 0.8421052694320679, 0.7894737124443054, 0.7105262875556946, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.8157894611358643, 0.9210526347160339, 0.9736841917037964], [0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.15789473056793213, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.15789473056793213, 0.28947368264198303, 0.21052631735801697, 0.2631579041481018, 0.2631579041481018, 0.2368421107530594, 0.21052631735801697, 0.21052631735801697, 0.31578946113586426, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.31578946113586426, 0.28947368264198303, 0.31578946113586426, 0.2631579041481018, 0.18421052396297455, 0.28947368264198303, 0.2368421107530594, 0.34210526943206787, 0.31578946113586426, 0.31578946113586426, 0.2631579041481018, 0.18421052396297455, 0.34210526943206787, 0.34210526943206787, 0.2631579041481018, 0.31578946113586426, 0.42105263471603394, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.42105263471603394, 0.42105263471603394, 0.3684210479259491, 0.3947368562221527, 0.30263158679008484, 0.5, 0.42105263471603394, NaN, 0.44736841320991516, 0.4736842215061188, 0.4736842215061188, 0.5789473652839661, 0.6315789222717285, 0.6315789222717285, 0.7105262875556946, 0.6578947305679321, 0.5789473652839661, 0.7631579041481018, 0.6842105388641357, 0.6315789222717285, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.7105262875556946, 0.7894737124443054, 0.6842105388641357, 0.6842105388641357, 0.7631579041481018, 0.8421052694320679, 0.9736841917037964, 0.9736841917037964], [0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.2631579041481018, 0.18421052396297455, 0.05263157933950424, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.2368421107530594, 0.02631578966975212, 0.21052631735801697, 0.2368421107530594, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.10526315867900848, 0.2631579041481018, 0.3684210479259491, 0.2631579041481018, 0.21052631735801697, 0.34210526943206787, 0.34210526943206787, 0.28947368264198303, 0.31578946113586426, 0.31578946113586426, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.42105263471603394, 0.42105263471603394, 0.3947368562221527, 0.4736842215061188, 0.3947368562221527, 0.43421053886413574, 0.4736842215061188, 0.5, 0.5, 0.5526315569877625, NaN, 0.6315789222717285, 0.5526315569877625, 0.5263158082962036, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.7631579041481018, 0.7105262875556946, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.7631579041481018, 0.7631579041481018, 0.7105262875556946, 0.7105262875556946, 0.7894737124443054, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375], [0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.10526315867900848, 0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.21052631735801697, 0.1315789520740509, 0.2368421107530594, 0.15789473056793213, 0.10526315867900848, 0.21052631735801697, 0.15789473056793213, 0.28947368264198303, 0.2631579041481018, 0.15789473056793213, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.2631579041481018, 0.18421052396297455, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.31578946113586426, 0.30263158679008484, 0.34210526943206787, 0.28947368264198303, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.31578946113586426, 0.31578946113586426, 0.42105263471603394, 0.3947368562221527, 0.2631579041481018, 0.3684210479259491, 0.44736841320991516, 0.42105263471603394, 0.5, 0.5263158082962036, 0.3684210479259491, NaN, 0.4736842215061188, 0.4736842215061188, 0.6842105388641357, 0.6578947305679321, 0.6052631735801697, 0.6052631735801697, 0.6578947305679321, 0.7631579041481018, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.7894737124443054, 0.6842105388641357, 0.6842105388641357, 0.7105262875556946, 0.7105262875556946, 0.7105262875556946, 0.6578947305679321, 0.8157894611358643, 0.8421052694320679, 0.9473684430122375], [0.07894736528396606, 0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.07894736528396606, 0.07894736528396606, 0.02631578966975212, 0.07894736528396606, 0.10526315867900848, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.2368421107530594, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.28947368264198303, 0.21052631735801697, 0.21052631735801697, 0.2631579041481018, 0.2368421107530594, 0.2631579041481018, 0.2631579041481018, 0.2368421107530594, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.2631579041481018, 0.34210526943206787, 0.2631579041481018, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.2631579041481018, 0.3947368562221527, 0.44736841320991516, 0.5, 0.44736841320991516, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, 0.5263158082962036, NaN, 0.5, 0.5789473652839661, 0.5789473652839661, 0.5789473652839661, 0.6315789222717285, 0.6052631735801697, 0.7105262875556946, 0.6842105388641357, 0.8157894611358643, 0.6578947305679321, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7894737124443054, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375], [0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.1315789520740509, 0.1315789520740509, 0.21052631735801697, 0.1315789520740509, 0.21052631735801697, 0.1315789520740509, 0.07894736528396606, 0.18421052396297455, 0.10526315867900848, 0.21052631735801697, 0.2631579041481018, 0.10526315867900848, 0.21052631735801697, 0.15789473056793213, 0.28947368264198303, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.34210526943206787, 0.28947368264198303, 0.28947368264198303, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.21052631735801697, 0.2631579041481018, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.2631579041481018, 0.42105263471603394, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.3947368562221527, 0.3684210479259491, 0.42105263471603394, 0.31578946113586426, 0.3684210479259491, 0.28947368264198303, 0.42105263471603394, 0.4736842215061188, 0.5263158082962036, 0.5, NaN, 0.6315789222717285, 0.6315789222717285, 0.6578947305679321, 0.6052631735801697, 0.6315789222717285, 0.7368420958518982, 0.6578947305679321, 0.7631579041481018, 0.8026315569877625, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.7631579041481018, 0.7105262875556946, 0.6842105388641357, 0.6842105388641357, 0.7894737124443054, 0.8947368264198303, 0.9736841917037964], [0.07894736528396606, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.21052631735801697, 0.15789473056793213, 0.18421052396297455, 0.18421052396297455, 0.15789473056793213, 0.2368421107530594, 0.25, 0.21052631735801697, 0.18421052396297455, 0.2368421107530594, 0.18421052396297455, 0.21052631735801697, 0.21052631735801697, 0.15789473056793213, 0.21052631735801697, 0.18421052396297455, 0.28947368264198303, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.21052631735801697, 0.2368421107530594, 0.2631579041481018, 0.2368421107530594, 0.2631579041481018, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.2368421107530594, 0.34210526943206787, 0.2368421107530594, 0.2631579041481018, 0.3684210479259491, 0.3684210479259491, 0.31578946113586426, 0.2631579041481018, 0.34210526943206787, 0.2631579041481018, 0.34210526943206787, 0.3684210479259491, 0.34210526943206787, 0.31578946113586426, 0.42105263471603394, 0.3684210479259491, NaN, 0.6447368264198303, 0.5, 0.6315789222717285, 0.4736842215061188, 0.9473684430122375, 0.6315789222717285, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.7631579041481018, 0.8421052694320679, 0.9210526347160339], [0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.1315789520740509, 0.2368421107530594, 0.2631579041481018, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.21052631735801697, 0.18421052396297455, 0.2631579041481018, 0.18421052396297455, 0.18421052396297455, 0.15789473056793213, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.21052631735801697, 0.21052631735801697, 0.2631579041481018, 0.2368421107530594, 0.2368421107530594, 0.2631579041481018, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.34210526943206787, 0.2368421107530594, 0.2631579041481018, 0.34210526943206787, 0.3684210479259491, 0.31578946113586426, 0.28947368264198303, 0.31578946113586426, 0.2631579041481018, 0.31578946113586426, 0.3684210479259491, 0.31578946113586426, 0.34210526943206787, 0.42105263471603394, 0.3684210479259491, 0.3552631437778473, NaN, 0.5263158082962036, 0.6315789222717285, 0.5, 0.9736841917037964, 0.6315789222717285, 0.5789473652839661, 0.6315789222717285, 0.6052631735801697, 0.6315789222717285, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.6578947305679321, 0.6578947305679321, 0.7631579041481018, 0.8421052694320679, 0.9210526347160339], [0.02631578966975212, 0.0, 0.0, 0.02631578966975212, 0.02631578966975212, 0.07894736528396606, 0.0, 0.07894736528396606, 0.0, 0.05263157933950424, 0.17105263471603394, 0.1315789520740509, 0.15789473056793213, 0.17105263471603394, 0.18421052396297455, 0.15789473056793213, 0.07894736528396606, 0.21052631735801697, 0.15789473056793213, 0.05263157933950424, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.10526315867900848, 0.10526315867900848, 0.21052631735801697, 0.21052631735801697, 0.2368421107530594, 0.1315789520740509, 0.2631579041481018, 0.21052631735801697, 0.2368421107530594, 0.21052631735801697, 0.28947368264198303, 0.2631579041481018, 0.31578946113586426, 0.2631579041481018, 0.2368421107530594, 0.2368421107530594, 0.31578946113586426, 0.31578946113586426, 0.28947368264198303, 0.2631579041481018, 0.28947368264198303, 0.3684210479259491, 0.28947368264198303, 0.3684210479259491, 0.3947368562221527, 0.42105263471603394, 0.34210526943206787, 0.5, 0.4736842215061188, NaN, 0.5526315569877625, 0.4736842215061188, 0.6842105388641357, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.6315789222717285, 0.7105262875556946, 0.6842105388641357, 0.6842105388641357, 0.6842105388641357, 0.7368420958518982, 0.7368420958518982, 0.9736841917037964], [0.0, 0.07894736528396606, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.10526315867900848, 0.10526315867900848, 0.21052631735801697, 0.21052631735801697, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.21052631735801697, 0.18421052396297455, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.15789473056793213, 0.2368421107530594, 0.31578946113586426, 0.2368421107530594, 0.28947368264198303, 0.21052631735801697, 0.2368421107530594, 0.18421052396297455, 0.28947368264198303, 0.2631579041481018, 0.31578946113586426, 0.2368421107530594, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.3684210479259491, 0.3947368562221527, 0.3684210479259491, 0.3684210479259491, 0.44736841320991516, NaN, 0.42105263471603394, 0.3684210479259491, 0.8157894611358643, 0.5263158082962036, 0.6052631735801697, 0.5526315569877625, 0.5526315569877625, 0.5789473652839661, 0.6052631735801697, 0.6578947305679321, 0.6578947305679321, 0.8684210777282715, 0.6842105388641357, 0.6842105388641357, 0.9210526347160339], [0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.18421052396297455, 0.07894736528396606, 0.10526315867900848, 0.15789473056793213, 0.07894736528396606, 0.02631578966975212, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.05263157933950424, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.21052631735801697, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.07894736528396606, 0.05263157933950424, 0.1315789520740509, 0.10526315867900848, 0.21052631735801697, 0.2631579041481018, 0.18421052396297455, 0.15789473056793213, 0.28947368264198303, 0.28947368264198303, 0.2631579041481018, 0.10526315867900848, 0.3684210479259491, 0.18421052396297455, 0.2631579041481018, 0.42105263471603394, 0.31578946113586426, 0.34210526943206787, 0.3947368562221527, 0.3684210479259491, 0.5263158082962036, 0.5, 0.5263158082962036, 0.5789473652839661, NaN, 0.6578947305679321, 0.6052631735801697, 0.5789473652839661, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.6052631735801697, 0.6842105388641357, 0.6315789222717285, 0.7105262875556946, 0.7105262875556946, 0.7894737124443054, 0.8421052694320679, 0.9473684430122375], [0.05263157933950424, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.18421052396297455, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.21052631735801697, 0.15789473056793213, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.28947368264198303, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.2631579041481018, 0.21052631735801697, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.28947368264198303, 0.2631579041481018, 0.05263157933950424, 0.02631578966975212, 0.31578946113586426, 0.6315789222717285, 0.34210526943206787, NaN, 0.6315789222717285, 0.31578946113586426, 0.5, 0.5263158082962036, 0.4736842215061188, 0.5, 0.5263158082962036, 0.5, 0.5, 0.6578947305679321, 0.5789473652839661, 0.6842105388641357, 0.8947368264198303], [0.0, 0.05263157933950424, 0.0, 0.0, 0.02631578966975212, 0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.10526315867900848, 0.07894736528396606, 0.05263157933950424, 0.18421052396297455, 0.07894736528396606, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.1315789520740509, 0.07894736528396606, 0.18421052396297455, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.21052631735801697, 0.2368421107530594, 0.2368421107530594, 0.21052631735801697, 0.28947368264198303, 0.2631579041481018, 0.28947368264198303, 0.31578946113586426, 0.28947368264198303, 0.42105263471603394, 0.31578946113586426, 0.34210526943206787, 0.3684210479259491, 0.3684210479259491, 0.44736841320991516, 0.18421052396297455, 0.3947368562221527, 0.3684210479259491, NaN, 0.5, 0.5263158082962036, 0.4736842215061188, 0.4736842215061188, 0.5789473652839661, 0.5789473652839661, 0.6315789222717285, 0.6052631735801697, 0.7763158082962036, 0.6578947305679321, 0.5789473652839661, 0.9210526347160339], [0.0, 0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.0, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.05263157933950424, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.1315789520740509, 0.1315789520740509, 0.07894736528396606, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.1315789520740509, 0.05263157933950424, 0.1315789520740509, 0.18421052396297455, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.07894736528396606, 0.21052631735801697, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.02631578966975212, 0.2368421107530594, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.28947368264198303, 0.3684210479259491, 0.2631579041481018, 0.34210526943206787, 0.18421052396297455, 0.2368421107530594, 0.3947368562221527, 0.42105263471603394, 0.4736842215061188, 0.4736842215061188, 0.42105263471603394, 0.6842105388641357, 0.5, NaN, 0.5, 0.5526315569877625, 0.5, 0.6052631735801697, 0.5789473652839661, 0.5789473652839661, 0.6052631735801697, 0.7105262875556946, 0.6578947305679321, 0.7894737124443054, 0.9473684430122375], [0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.07894736528396606, 0.15789473056793213, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.1315789520740509, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.07894736528396606, 0.07894736528396606, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.21052631735801697, 0.18421052396297455, 0.10526315867900848, 0.1315789520740509, 0.21052631735801697, 0.21052631735801697, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.2368421107530594, 0.21052631735801697, 0.2631579041481018, 0.28947368264198303, 0.2631579041481018, 0.21052631735801697, 0.31578946113586426, 0.1315789520740509, 0.18421052396297455, 0.1315789520740509, 0.18421052396297455, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.19736842811107635, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.42105263471603394, 0.5, 0.4736842215061188, 0.5, NaN, 0.5263158082962036, 0.5263158082962036, 0.5789473652839661, 0.5, 0.6052631735801697, 0.6052631735801697, 0.5789473652839661, 0.6578947305679321, 0.7105262875556946, 0.9736841917037964], [0.07894736528396606, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.1315789520740509, 0.05263157933950424, 0.1315789520740509, 0.10526315867900848, 0.07894736528396606, 0.07894736528396606, 0.1315789520740509, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.07894736528396606, 0.14473684132099152, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.10526315867900848, 0.07894736528396606, 0.07894736528396606, 0.18421052396297455, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.18421052396297455, 0.21052631735801697, 0.10526315867900848, 0.10526315867900848, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.21052631735801697, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.34210526943206787, 0.4736842215061188, 0.5263158082962036, 0.44736841320991516, 0.4736842215061188, NaN, 0.5526315569877625, 0.6315789222717285, 0.5526315569877625, 0.6052631735801697, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.9736841917037964, 0.9736841917037964], [0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.0, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.15789473056793213, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.10526315867900848, 0.05263157933950424, 0.1315789520740509, 0.18421052396297455, 0.02631578966975212, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.2368421107530594, 0.15789473056793213, 0.15789473056793213, 0.07894736528396606, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.07894736528396606, 0.15789473056793213, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.21052631735801697, 0.2631579041481018, 0.2368421107530594, 0.07894736528396606, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.2368421107530594, 0.31578946113586426, 0.2631579041481018, 0.21052631735801697, 0.34210526943206787, 0.3684210479259491, 0.31578946113586426, 0.44736841320991516, 0.34210526943206787, 0.5263158082962036, 0.5263158082962036, 0.5, 0.4736842215061188, 0.44736841320991516, NaN, 0.5526315569877625, 0.5263158082962036, 0.5526315569877625, 0.5526315569877625, 0.6052631735801697, 0.6052631735801697, 0.7105262875556946, 0.9473684430122375], [0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.07894736528396606, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.10526315867900848, 0.15789473056793213, 0.05263157933950424, 0.15789473056793213, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.07894736528396606, 0.10526315867900848, 0.21052631735801697, 0.18421052396297455, 0.02631578966975212, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.21052631735801697, 0.18421052396297455, 0.21052631735801697, 0.21052631735801697, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.2631579041481018, 0.21052631735801697, 0.34210526943206787, 0.3684210479259491, 0.3684210479259491, 0.42105263471603394, 0.3947368562221527, 0.5, 0.42105263471603394, 0.3947368562221527, 0.42105263471603394, 0.3684210479259491, 0.44736841320991516, NaN, 0.5, 0.4736842215061188, 0.4736842215061188, 0.5, 0.5263158082962036, 0.6315789222717285, 0.8421052694320679], [0.0, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.10526315867900848, 0.02631578966975212, 0.0, 0.05263157933950424, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.05263157933950424, 0.15789473056793213, 0.07894736528396606, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.02631578966975212, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.07894736528396606, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.07894736528396606, 0.15789473056793213, 0.05263157933950424, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.2368421107530594, 0.07894736528396606, 0.18421052396297455, 0.2368421107530594, 0.28947368264198303, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.21052631735801697, 0.2368421107530594, 0.31578946113586426, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.31578946113586426, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, 0.5, 0.44736841320991516, 0.4736842215061188, 0.5, NaN, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.6315789222717285, 0.6842105388641357, 0.8684210777282715], [0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.10526315867900848, 0.02631578966975212, 0.02631578966975212, 0.15789473056793213, 0.02631578966975212, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.05263157933950424, 0.18421052396297455, 0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.05263157933950424, 0.15789473056793213, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.1315789520740509, 0.28947368264198303, 0.18421052396297455, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.18421052396297455, 0.2631579041481018, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.3684210479259491, 0.5, 0.3684210479259491, 0.42105263471603394, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, NaN, 0.75, 0.5, 0.7368420958518982, 0.6315789222717285, 0.8684210777282715], [0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.02631578966975212, 0.03947368264198303, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.02631578966975212, 0.02631578966975212, 0.15789473056793213, 0.02631578966975212, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.07894736528396606, 0.02631578966975212, 0.15789473056793213, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.1315789520740509, 0.18421052396297455, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.02631578966975212, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.1315789520740509, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.10526315867900848, 0.2631579041481018, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.28947368264198303, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.31578946113586426, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.28947368264198303, 0.5, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, 0.25, NaN, 0.5, 0.6578947305679321, 0.6052631735801697, 0.8421052694320679], [0.0, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.07894736528396606, 0.21052631735801697, 0.15789473056793213, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.21052631735801697, 0.15789473056793213, 0.28947368264198303, 0.2368421107530594, 0.21052631735801697, 0.34210526943206787, 0.2368421107530594, 0.31578946113586426, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.1315789520740509, 0.28947368264198303, 0.34210526943206787, 0.22368420660495758, 0.28947368264198303, 0.42105263471603394, 0.3947368562221527, 0.3947368562221527, 0.5, 0.44736841320991516, 0.5, 0.5, NaN, 0.5263158082962036, 0.5, 0.8421052694320679], [0.0, 0.02631578966975212, 0.0, 0.02631578966975212, 0.05263157933950424, 0.0, 0.02631578966975212, 0.0, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.02631578966975212, 0.05263157933950424, 0.10526315867900848, 0.02631578966975212, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.07894736528396606, 0.02631578966975212, 0.15789473056793213, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.1315789520740509, 0.1315789520740509, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.10526315867900848, 0.10526315867900848, 0.02631578966975212, 0.10526315867900848, 0.02631578966975212, 0.02631578966975212, 0.1315789520740509, 0.10526315867900848, 0.05263157933950424, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.2368421107530594, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.21052631735801697, 0.42105263471603394, 0.34210526943206787, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.4736842215061188, 0.3684210479259491, 0.2631579041481018, 0.34210526943206787, 0.4736842215061188, NaN, 0.6052631735801697, 0.7894737124443054], [0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.0, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.0, 0.05263157933950424, 0.1315789520740509, 0.02631578966975212, 0.0, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.02631578966975212, 0.1315789520740509, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.07894736528396606, 0.02631578966975212, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.2631579041481018, 0.31578946113586426, 0.15789473056793213, 0.31578946113586426, 0.42105263471603394, 0.21052631735801697, 0.28947368264198303, 0.02631578966975212, 0.28947368264198303, 0.3684210479259491, 0.31578946113586426, 0.3684210479259491, 0.3947368562221527, 0.5, 0.3947368562221527, NaN, 0.9473684430122375], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.05263157933950424, 0.02631578966975212, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.07894736528396606, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.21052631735801697, 0.05263157933950424, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.918494155},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.861476613},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8532529281},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8340643276},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8203581919},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8135964945},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7916666671},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7913011708},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.790021933},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7459795314},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.683296785},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6790935714},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.662097957},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6611842093},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.660453214},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6560672491},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6540570199},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6525950279},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6379751469},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6330409383},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6321271962},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6178728102},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6158625715},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6045321636},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6014254402},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6008771955},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5994152052},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5842470787},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5835160811},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5789473689},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.57383041},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5639619912},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5628654984},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5614035093},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5517178356},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5219298246},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5127923967},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5124268995},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5113304095},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4996345054},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4978070178},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4947002913},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4581505819},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4533991217},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4477339221},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4464546783},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4334795335},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4195906412},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4064327465},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3989400573},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3912646201},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.373355264},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3687865497},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3645833345},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3486842107},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3379020455},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3015350878},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2949561412},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2777777778},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.251461987},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2476242691},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2412280716},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2406798256},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2359283625},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2284356721},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2240497075},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2160087714},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2147295322},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1959064329},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1891447376},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1531432763},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1239035088},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0314327484}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_classification/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9463d254dd87a07ef2fcbbe014507e1c658dac31
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_all/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),TabICLv2 (default),RealTabPFN-2.5 (default),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),RealMLP (tuned),TabM (tuned + ensembled),iLTM (tuned + ensembled),LimiX (default),CatBoost (tuned),LightGBM (tuned + ensembled),TabDPT-Turbo (default),CatBoost (tuned + ensembled),TabM (tuned),TabDPT (tuned + ensembled),XGBoost (tuned + ensembled),ModernNCA (tuned),CatBoost (default),ModernNCA (tuned + ensembled),ChimeraBoost (tuned + ensembled),LightGBM (tuned),XGBoost (tuned),TabSwift (default),TabDPT (tuned),BetaTabPFN (default),ChimeraBoost (tuned),EBM (tuned + ensembled),xRFM (tuned + ensembled),iLTM (tuned),TabM (default),TorchMLP (tuned + ensembled),xRFM (tuned),RealMLP (default),ChimeraBoost (default),SAP-RPT-OSS (default),EBM (tuned),EBM (default),TabDPT (default),FastaiMLP (tuned + ensembled),XGBoost (default),TorchMLP (tuned),ExtraTrees (tuned + ensembled),ModernNCA (default),RandomForest (tuned + ensembled),ExtraTrees (tuned),LightGBM (default),iLTM (default),FastaiMLP (tuned),RandomForest (tuned),TabSTAR (tuned),TabSTAR (tuned + ensembled),OrionMSP (default),PerpetualBooster (tuned + ensembled),TorchMLP (default),TabSTAR (default),PerpetualBooster (tuned),FastaiMLP (default),RandomForest (default),KNN (tuned + ensembled),ExtraTrees (default),xRFM (default),TabFlex (default),Linear (tuned + ensembled),Linear (tuned),PerpetualBooster (default),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.7368421,0.81578946,0.84210527,0.8684211,0.8684211,0.84210527,0.7631579,0.8947368,0.81578946,0.84210527,0.92105263,0.81578946,0.8684211,0.8947368,0.92105263,0.92105263,0.92105263,0.94736844,0.92105263,0.8947368,0.8947368,0.8684211,0.8684211,0.8947368,0.94736844,0.84210527,0.8684211,0.94736844,0.94736844,0.8684211,0.92105263,0.92105263,0.92105263,0.92105263,0.94736844,0.92105263,0.8947368,0.92105263,0.8947368,0.92105263,0.92105263,0.9736842,0.8947368,0.94736844,0.9736842,0.94736844,0.94736844,0.94736844,0.94736844,0.94736844,0.92105263,0.92105263,0.94736844,0.92105263,0.8947368,0.9736842,1.0,0.9736842,0.94736844,1.0,1.0,0.9736842,0.92105263,0.9736842,0.94736844,1.0,0.9736842,0.9736842,1.0,1.0,0.9736842,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2631579,,0.47368422,0.6052632,0.6052632,0.5263158,0.57894737,0.68421054,0.68421054,0.81578946,0.81578946,0.81578946,0.8684211,0.7368421,0.84210527,0.8947368,0.68421054,0.8947368,0.7894737,0.81578946,0.92105263,0.8947368,0.92105263,0.92105263,0.92105263,0.92105263,0.92105263,0.7894737,0.81578946,0.84210527,0.92105263,0.84210527,0.92105263,0.8947368,0.8684211,0.92105263,0.8684211,0.8684211,0.94736844,0.81578946,0.8684211,0.8947368,0.84210527,0.8684211,0.92105263,0.8947368,0.92105263,0.94736844,0.94736844,0.92105263,0.9736842,0.94736844,0.92105263,0.94736844,0.8684211,0.8684211,1.0,0.92105263,0.94736844,0.8684211,0.94736844,0.9736842,0.94736844,0.9736842,0.9736842,0.9736842,0.94736844,0.9736842,0.9736842,0.94736844,0.9736842,1.0,1.0
+TabPFN-3 (default),0.18421052,0.5263158,,0.55263156,0.56578946,0.6052632,0.57894737,0.55263156,0.65789473,0.7105263,0.7368421,0.7894737,0.7894737,0.7105263,0.7368421,0.8684211,0.81578946,0.7894737,0.8684211,0.7894737,0.84210527,0.84210527,0.84210527,0.92105263,0.8684211,0.94736844,0.81578946,0.8684211,0.8947368,0.92105263,0.84210527,0.81578946,0.84210527,0.8947368,0.84210527,0.8684211,0.8947368,0.92105263,0.8947368,0.7894737,0.8947368,0.8947368,0.8947368,0.8684211,0.94736844,0.94736844,0.94736844,0.94736844,0.92105263,0.94736844,0.94736844,0.9736842,0.94736844,0.9736842,0.8684211,0.8684211,1.0,0.9736842,0.94736844,0.8684211,1.0,1.0,0.9736842,0.94736844,0.9736842,0.94736844,0.94736844,0.94736844,0.9736842,0.9736842,1.0,0.9736842,1.0
+TabPFN-2.6 (default),0.15789473,0.39473686,0.4473684,,0.47368422,0.5,0.65789473,0.5263158,0.55263156,0.7105263,0.7368421,0.7368421,0.81578946,0.7631579,0.7631579,0.8947368,0.7894737,0.81578946,0.7894737,0.7368421,0.7631579,0.81578946,0.8684211,0.8947368,0.81578946,0.9736842,0.84210527,0.7894737,0.7631579,0.84210527,0.81578946,0.8684211,0.84210527,0.8684211,0.84210527,0.8947368,0.84210527,0.8947368,0.84210527,0.81578946,0.92105263,0.8684211,0.81578946,0.8947368,0.94736844,0.94736844,0.92105263,0.94736844,0.94736844,0.92105263,0.94736844,0.94736844,0.8947368,0.92105263,0.84210527,0.84210527,0.9736842,0.9736842,0.9736842,0.84210527,1.0,0.9736842,0.9736842,0.9736842,0.94736844,0.94736844,0.92105263,0.9736842,0.9736842,0.9736842,0.9736842,0.9736842,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.13157895,0.39473686,0.43421054,0.5263158,,0.5263158,0.6315789,0.55263156,0.7105263,0.68421054,0.7105263,0.68421054,0.7368421,0.68421054,0.81578946,0.81578946,0.65789473,0.8684211,0.7631579,0.68421054,0.84210527,0.84210527,0.8684211,0.7631579,0.7894737,0.8947368,0.84210527,0.81578946,0.7631579,0.8684211,0.84210527,0.84210527,0.81578946,0.81578946,0.84210527,0.81578946,0.8684211,0.84210527,0.81578946,0.8684211,0.84210527,0.8684211,0.81578946,0.84210527,0.94736844,0.8684211,0.94736844,0.8684211,0.92105263,0.94736844,0.9736842,0.8947368,0.92105263,0.94736844,0.84210527,0.84210527,0.9736842,0.94736844,0.94736844,0.8684211,0.9736842,0.9736842,0.9736842,0.92105263,0.9736842,0.94736844,0.8947368,0.92105263,0.94736844,0.94736844,0.94736844,0.94736844,1.0
+TabICLv2 (default),0.13157895,0.47368422,0.39473686,0.5,0.47368422,,0.57894737,0.6315789,0.6052632,0.65789473,0.6315789,0.68421054,0.7631579,0.68421054,0.7368421,0.81578946,0.7894737,0.7368421,0.7105263,0.81578946,0.7631579,0.7894737,0.7894737,0.84210527,0.81578946,0.84210527,0.7631579,0.81578946,0.81578946,0.8947368,0.7894737,0.7631579,0.7894737,0.7631579,0.84210527,0.8684211,0.84210527,0.81578946,0.81578946,0.7631579,0.8684211,0.8947368,0.8684211,0.94736844,0.8684211,0.8947368,0.8684211,0.92105263,0.8947368,0.92105263,0.94736844,0.84210527,0.92105263,0.8947368,0.8684211,0.8684211,0.92105263,0.94736844,0.9736842,0.8947368,0.9736842,1.0,0.94736844,0.94736844,0.92105263,0.92105263,0.9736842,0.94736844,0.9736842,0.9736842,1.0,0.9736842,1.0
+RealTabPFN-2.5 (default),0.15789473,0.42105263,0.42105263,0.34210527,0.36842105,0.42105263,,0.55263156,0.43421054,0.6052632,0.68421054,0.68421054,0.57894737,0.7105263,0.68421054,0.7631579,0.65789473,0.68421054,0.7368421,0.7631579,0.7368421,0.68421054,0.7105263,0.65789473,0.7105263,0.7894737,0.7894737,0.84210527,0.7368421,0.84210527,0.7105263,0.7894737,0.81578946,0.8815789,0.81578946,0.84210527,0.84210527,0.81578946,0.84210527,0.84210527,0.8947368,0.8684211,0.84210527,0.8947368,0.8947368,0.8684211,0.84210527,0.92105263,0.84210527,0.84210527,0.9736842,0.8947368,0.9736842,0.92105263,0.8947368,0.8947368,1.0,0.94736844,0.9736842,0.8947368,1.0,0.94736844,0.94736844,0.92105263,1.0,0.94736844,1.0,0.94736844,0.94736844,0.94736844,0.9736842,0.9736842,1.0
+"AutoGluon 1.4 (best, 4h)",0.23684211,0.31578946,0.4473684,0.47368422,0.4473684,0.36842105,0.4473684,,0.5,0.57894737,0.7105263,0.65789473,0.81578946,0.6052632,0.7631579,0.7894737,0.65789473,0.7894737,0.7368421,0.6315789,0.81578946,0.68421054,0.7631579,0.65789473,0.7894737,0.84210527,0.7894737,0.7105263,0.7105263,0.7631579,0.81578946,0.7631579,0.7894737,0.7894737,0.7894737,0.8947368,0.81578946,0.7894737,0.84210527,0.7631579,0.8947368,0.8684211,0.84210527,0.8947368,0.8947368,0.92105263,0.84210527,0.92105263,0.8684211,0.84210527,0.92105263,0.92105263,0.92105263,0.94736844,0.7894737,0.81578946,0.92105263,0.94736844,0.9736842,0.8684211,0.9736842,0.9736842,0.9736842,0.92105263,0.9736842,0.92105263,0.94736844,0.9736842,0.9736842,0.9736842,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.10526316,0.31578946,0.34210527,0.4473684,0.28947368,0.39473686,0.56578946,0.5,,0.6052632,0.7105263,0.7368421,0.65789473,0.57894737,0.7368421,0.7368421,0.65789473,0.7631579,0.7894737,0.65789473,0.7631579,0.81578946,0.8684211,0.7368421,0.68421054,0.8684211,0.7894737,0.7368421,0.7631579,0.84210527,0.7894737,0.84210527,0.7631579,0.84210527,0.84210527,0.81578946,0.84210527,0.84210527,0.7894737,0.84210527,0.84210527,0.84210527,0.81578946,0.81578946,0.92105263,0.8947368,0.9736842,0.8947368,0.92105263,0.8947368,0.94736844,0.8684211,0.8947368,0.92105263,0.84210527,0.84210527,1.0,0.94736844,0.94736844,0.8684211,0.9736842,0.94736844,0.9736842,0.92105263,0.9736842,0.94736844,0.8947368,0.92105263,0.92105263,0.94736844,0.94736844,0.94736844,1.0
+RealMLP (tuned + ensembled),0.18421052,0.18421052,0.28947368,0.28947368,0.31578946,0.34210527,0.39473686,0.42105263,0.39473686,,0.65789473,0.6052632,0.7105263,0.5,0.68421054,0.7105263,0.57894737,0.7631579,0.68421054,0.57894737,0.7105263,0.65789473,0.7631579,0.68421054,0.65789473,0.7631579,0.7631579,0.6315789,0.6315789,0.7105263,0.68421054,0.7894737,0.7105263,0.7894737,0.7631579,0.81578946,0.7631579,0.8947368,0.7631579,0.7894737,0.81578946,0.81578946,0.7894737,0.8947368,0.8684211,0.81578946,0.81578946,0.8947368,0.7894737,0.81578946,0.92105263,0.8947368,0.8684211,0.8684211,0.81578946,0.81578946,0.94736844,0.94736844,0.92105263,0.84210527,0.9736842,0.9736842,0.94736844,0.94736844,0.9736842,0.94736844,0.8947368,0.9736842,0.9736842,0.9736842,0.9736842,0.9736842,1.0
+RealMLP (tuned),0.15789473,0.18421052,0.2631579,0.2631579,0.28947368,0.36842105,0.31578946,0.28947368,0.28947368,0.34210527,,0.47368422,0.5,0.43421054,0.5,0.6052632,0.47368422,0.5263158,0.57894737,0.47368422,0.6315789,0.65789473,0.6315789,0.6052632,0.55263156,0.65789473,0.68421054,0.57894737,0.57894737,0.68421054,0.57894737,0.7105263,0.65789473,0.7105263,0.7105263,0.7894737,0.68421054,0.81578946,0.65789473,0.65789473,0.7368421,0.7368421,0.68421054,0.7894737,0.81578946,0.81578946,0.7631579,0.7631579,0.84210527,0.84210527,0.84210527,0.8684211,0.84210527,0.8684211,0.81578946,0.84210527,0.82894737,0.8947368,0.94736844,0.84210527,0.8947368,0.94736844,0.92105263,0.92105263,0.94736844,0.94736844,0.8684211,0.94736844,0.9605263,0.9736842,0.94736844,0.9736842,1.0
+TabM (tuned + ensembled),0.078947365,0.18421052,0.21052632,0.2631579,0.31578946,0.31578946,0.31578946,0.34210527,0.2631579,0.39473686,0.5263158,,0.5263158,0.42105263,0.47368422,0.6315789,0.57894737,0.5263158,0.55263156,0.57894737,0.55263156,0.6315789,0.6052632,0.6052632,0.57894737,0.6052632,0.68421054,0.6315789,0.6052632,0.68421054,0.68421054,0.7894737,0.57894737,0.6315789,0.84210527,0.7105263,0.57894737,0.81578946,0.7105263,0.7105263,0.7368421,0.7894737,0.7631579,0.7894737,0.7368421,0.7894737,0.7105263,0.84210527,0.7894737,0.7105263,0.84210527,0.7894737,0.84210527,0.7894737,0.84210527,0.8684211,0.8684211,0.84210527,0.9736842,0.8684211,0.92105263,0.92105263,0.84210527,0.8684211,0.84210527,0.8684211,0.92105263,0.94736844,0.94736844,0.94736844,0.94736844,1.0,1.0
+iLTM (tuned + ensembled),0.18421052,0.13157895,0.21052632,0.18421052,0.2631579,0.23684211,0.42105263,0.18421052,0.34210527,0.28947368,0.5,0.47368422,,0.43421054,0.5,0.5263158,0.4473684,0.4473684,0.5,0.42105263,0.5263158,0.6052632,0.5,0.5263158,0.65789473,0.55263156,0.6315789,0.57894737,0.5,0.5,0.65789473,0.6315789,0.65789473,0.68421054,0.6052632,0.7368421,0.7368421,0.7105263,0.7368421,0.7105263,0.65789473,0.7368421,0.68421054,0.81578946,0.7894737,0.81578946,0.81578946,0.81578946,0.8684211,0.7894737,0.84210527,0.8684211,0.81578946,0.8684211,0.7631579,0.7631579,0.84210527,0.8947368,0.94736844,0.84210527,0.94736844,0.94736844,0.94736844,0.94736844,0.94736844,0.92105263,0.92105263,0.92105263,0.94736844,0.94736844,0.94736844,0.94736844,1.0
+LimiX (default),0.13157895,0.2631579,0.28947368,0.23684211,0.31578946,0.31578946,0.28947368,0.39473686,0.42105263,0.5,0.56578946,0.57894737,0.56578946,,0.5263158,0.6315789,0.4473684,0.57894737,0.6052632,0.4473684,0.65789473,0.6052632,0.57894737,0.6052632,0.6315789,0.6052632,0.68421054,0.5263158,0.55263156,0.6052632,0.65789473,0.68421054,0.6315789,0.68421054,0.65789473,0.68421054,0.7368421,0.7368421,0.68421054,0.65789473,0.7368421,0.7368421,0.65789473,0.65789473,0.7368421,0.7105263,0.7368421,0.7368421,0.7368421,0.7368421,0.7368421,0.7631579,0.7894737,0.7894737,0.75,0.7368421,0.82894737,0.81578946,0.81578946,0.81578946,0.81578946,0.8947368,0.92105263,0.8684211,0.94736844,0.92105263,0.84210527,0.8947368,0.8947368,0.8684211,0.8947368,0.8684211,0.9736842
+CatBoost (tuned),0.10526316,0.15789473,0.2631579,0.23684211,0.18421052,0.2631579,0.31578946,0.23684211,0.2631579,0.31578946,0.5,0.5263158,0.5,0.47368422,,0.4473684,0.5263158,0.4473684,0.55263156,0.5263158,0.57894737,0.5263158,0.65789473,0.57894737,0.5263158,0.5263158,0.65789473,0.57894737,0.55263156,0.57894737,0.6052632,0.65789473,0.6315789,0.7368421,0.6315789,0.7368421,0.7368421,0.7368421,0.7105263,0.7105263,0.7105263,0.7631579,0.7105263,0.7368421,0.81578946,0.81578946,0.7631579,0.7631579,0.81578946,0.7368421,0.81578946,0.84210527,0.81578946,0.8684211,0.7894737,0.81578946,0.81578946,0.84210527,0.92105263,0.81578946,0.92105263,0.8947368,0.92105263,0.8947368,0.8684211,0.8684211,0.8684211,0.9736842,0.9736842,0.94736844,0.9736842,0.9736842,1.0
+LightGBM (tuned + ensembled),0.078947365,0.10526316,0.13157895,0.10526316,0.18421052,0.18421052,0.23684211,0.21052632,0.2631579,0.28947368,0.39473686,0.36842105,0.47368422,0.36842105,0.55263156,,0.5,0.57894737,0.47368422,0.47368422,0.47368422,0.5,0.65789473,0.5263158,0.57894737,0.7105263,0.6315789,0.57894737,0.55263156,0.55263156,0.65789473,0.68421054,0.6315789,0.65789473,0.65789473,0.7368421,0.7105263,0.68421054,0.7368421,0.65789473,0.7894737,0.81578946,0.7105263,0.7631579,0.8684211,0.7894737,0.81578946,0.81578946,0.81578946,0.7631579,0.94736844,0.8947368,0.81578946,0.92105263,0.81578946,0.81578946,0.84210527,0.8684211,0.8947368,0.84210527,0.8947368,0.8947368,0.9736842,0.92105263,0.8947368,0.84210527,0.8684211,0.9736842,0.9736842,0.8947368,0.94736844,1.0,1.0
+TabDPT-Turbo (default),0.078947365,0.31578946,0.18421052,0.21052632,0.34210527,0.21052632,0.34210527,0.34210527,0.34210527,0.42105263,0.5263158,0.42105263,0.55263156,0.55263156,0.47368422,0.5,,0.5,0.4473684,0.6315789,0.57894737,0.5,0.57894737,0.5,0.6052632,0.5,0.6052632,0.67105263,0.68421054,0.5263158,0.6052632,0.6052632,0.6052632,0.68421054,0.6052632,0.65789473,0.68421054,0.6315789,0.6315789,0.65789473,0.65789473,0.7105263,0.81578946,0.7105263,0.7368421,0.7631579,0.7631579,0.68421054,0.7631579,0.7894737,0.81578946,0.7894737,0.8684211,0.81578946,0.7631579,0.7631579,0.92105263,0.81578946,0.84210527,0.84210527,0.8684211,0.94736844,0.92105263,0.92105263,0.94736844,0.8947368,0.94736844,0.84210527,0.84210527,0.94736844,0.8947368,0.9736842,1.0
+CatBoost (tuned + ensembled),0.078947365,0.10526316,0.21052632,0.18421052,0.13157895,0.2631579,0.31578946,0.21052632,0.23684211,0.23684211,0.47368422,0.47368422,0.55263156,0.42105263,0.55263156,0.42105263,0.5,,0.5,0.5,0.5263158,0.5131579,0.6315789,0.57894737,0.55263156,0.5263158,0.6052632,0.55263156,0.55263156,0.5263158,0.6052632,0.6052632,0.6315789,0.7368421,0.55263156,0.7105263,0.7368421,0.7368421,0.81578946,0.7368421,0.68421054,0.7631579,0.68421054,0.7368421,0.81578946,0.81578946,0.81578946,0.7631579,0.84210527,0.7894737,0.81578946,0.84210527,0.84210527,0.8947368,0.81578946,0.81578946,0.7894737,0.8947368,0.92105263,0.81578946,0.8947368,0.8947368,0.94736844,0.8684211,0.8684211,0.8947368,0.84210527,0.9736842,0.9736842,0.94736844,0.9736842,0.94736844,1.0
+TabM (tuned),0.05263158,0.21052632,0.13157895,0.21052632,0.23684211,0.28947368,0.2631579,0.2631579,0.21052632,0.31578946,0.42105263,0.4473684,0.5,0.39473686,0.4473684,0.5263158,0.55263156,0.5,,0.55263156,0.5263158,0.6052632,0.57894737,0.55263156,0.6447368,0.5263158,0.6052632,0.6052632,0.6052632,0.6315789,0.6315789,0.65789473,0.5263158,0.6052632,0.68421054,0.6315789,0.57894737,0.65789473,0.6315789,0.65789473,0.7105263,0.68421054,0.7368421,0.68421054,0.7368421,0.7631579,0.68421054,0.84210527,0.7368421,0.68421054,0.7894737,0.7105263,0.81578946,0.7894737,0.7894737,0.81578946,0.84210527,0.8947368,0.9736842,0.84210527,0.92105263,0.8684211,0.8684211,0.8684211,0.81578946,0.84210527,0.92105263,0.8947368,0.8947368,0.94736844,0.94736844,0.9736842,0.9736842
+TabDPT (tuned + ensembled),0.078947365,0.18421052,0.21052632,0.2631579,0.31578946,0.18421052,0.23684211,0.36842105,0.34210527,0.42105263,0.5263158,0.42105263,0.57894737,0.55263156,0.47368422,0.5263158,0.36842105,0.5,0.4473684,,0.5263158,0.47368422,0.5263158,0.47368422,0.5263158,0.5263158,0.5263158,0.5263158,0.7105263,0.55263156,0.5263158,0.57894737,0.57894737,0.65789473,0.57894737,0.65789473,0.65789473,0.6315789,0.5263158,0.65789473,0.65789473,0.7105263,0.7631579,0.65789473,0.68421054,0.68421054,0.7105263,0.81578946,0.7105263,0.81578946,0.7631579,0.7368421,0.7894737,0.7368421,0.7894737,0.7631579,0.94736844,0.7894737,0.81578946,0.84210527,0.8684211,0.94736844,0.94736844,0.8947368,0.9736842,0.94736844,0.92105263,0.8684211,0.8684211,0.8947368,0.8947368,0.94736844,1.0
+XGBoost (tuned + ensembled),0.10526316,0.078947365,0.15789473,0.23684211,0.15789473,0.23684211,0.2631579,0.18421052,0.23684211,0.28947368,0.36842105,0.4473684,0.47368422,0.34210527,0.42105263,0.5263158,0.42105263,0.47368422,0.47368422,0.47368422,,0.47368422,0.55263156,0.5,0.55263156,0.57894737,0.68421054,0.5263158,0.5,0.47368422,0.6315789,0.6315789,0.55263156,0.6052632,0.6052632,0.68421054,0.68421054,0.7105263,0.7105263,0.6315789,0.68421054,0.7105263,0.68421054,0.7105263,0.8684211,0.69736844,0.81578946,0.7894737,0.84210527,0.7894737,0.9736842,0.84210527,0.7631579,0.8947368,0.84210527,0.84210527,0.84210527,0.7894737,0.8947368,0.84210527,0.84210527,0.8684211,0.94736844,0.8684211,0.92105263,0.84210527,0.84210527,0.92105263,0.94736844,0.8684211,0.94736844,0.92105263,1.0
+ModernNCA (tuned),0.10526316,0.10526316,0.15789473,0.18421052,0.15789473,0.21052632,0.31578946,0.31578946,0.18421052,0.34210527,0.34210527,0.36842105,0.39473686,0.39473686,0.47368422,0.5,0.5,0.4868421,0.39473686,0.5263158,0.5263158,,0.57894737,0.43421054,0.5263158,0.55263156,0.5,0.55263156,0.57894737,0.57894737,0.55263156,0.5263158,0.57894737,0.57894737,0.6052632,0.65789473,0.68421054,0.65789473,0.65789473,0.68421054,0.65789473,0.6315789,0.68421054,0.67105263,0.7105263,0.7631579,0.7105263,0.84210527,0.7368421,0.7368421,0.7894737,0.7631579,0.7894737,0.7894737,0.7894737,0.7894737,0.84210527,0.84210527,0.84210527,0.8947368,0.8684211,0.8684211,0.92105263,0.92105263,0.8684211,0.8947368,0.8684211,0.8947368,0.92105263,0.8684211,0.92105263,0.92105263,1.0
+CatBoost (default),0.13157895,0.078947365,0.15789473,0.13157895,0.13157895,0.21052632,0.28947368,0.23684211,0.13157895,0.23684211,0.36842105,0.39473686,0.5,0.42105263,0.34210527,0.34210527,0.42105263,0.36842105,0.42105263,0.47368422,0.4473684,0.42105263,,0.47368422,0.47368422,0.4473684,0.47368422,0.55263156,0.5,0.42105263,0.57894737,0.57894737,0.65789473,0.65789473,0.55263156,0.7105263,0.7105263,0.7631579,0.7368421,0.65789473,0.6315789,0.7105263,0.7105263,0.7631579,0.7368421,0.7894737,0.7368421,0.7105263,0.81578946,0.68421054,0.7631579,0.81578946,0.81578946,0.84210527,0.81578946,0.81578946,0.81578946,0.8684211,0.94736844,0.84210527,0.92105263,0.92105263,0.92105263,0.92105263,0.84210527,0.8947368,0.8684211,0.94736844,0.9736842,0.94736844,0.9736842,0.9736842,1.0
+ModernNCA (tuned + ensembled),0.13157895,0.078947365,0.078947365,0.10526316,0.23684211,0.15789473,0.34210527,0.34210527,0.2631579,0.31578946,0.39473686,0.39473686,0.47368422,0.39473686,0.42105263,0.47368422,0.5,0.42105263,0.4473684,0.5263158,0.5,0.56578946,0.5263158,,0.5263158,0.6052632,0.5263158,0.5,0.55263156,0.57894737,0.55263156,0.6052632,0.55263156,0.5263158,0.5263158,0.6052632,0.65789473,0.7105263,0.6052632,0.57894737,0.6052632,0.65789473,0.68421054,0.7368421,0.7368421,0.7631579,0.7368421,0.81578946,0.7368421,0.7105263,0.84210527,0.7631579,0.81578946,0.7105263,0.7105263,0.7368421,0.81578946,0.81578946,0.81578946,0.84210527,0.81578946,0.84210527,0.8684211,0.8552632,0.84210527,0.8947368,0.81578946,0.81578946,0.84210527,0.8947368,0.84210527,0.8684211,0.9736842
+ChimeraBoost (tuned + ensembled),0.10526316,0.078947365,0.13157895,0.18421052,0.21052632,0.18421052,0.28947368,0.21052632,0.31578946,0.34210527,0.4473684,0.42105263,0.34210527,0.36842105,0.47368422,0.42105263,0.39473686,0.4473684,0.35526314,0.47368422,0.4473684,0.47368422,0.5263158,0.47368422,,0.47368422,0.47368422,0.55263156,0.47368422,0.39473686,0.57894737,0.6315789,0.6052632,0.6052632,0.57894737,0.6315789,0.68421054,0.68421054,0.81578946,0.6315789,0.68421054,0.65789473,0.68421054,0.7105263,0.68421054,0.7105263,0.65789473,0.68421054,0.7631579,0.68421054,0.7631579,0.7368421,0.7105263,0.81578946,0.81578946,0.81578946,0.7894737,0.7894737,0.84210527,0.84210527,0.84210527,0.84210527,0.84210527,0.84210527,0.81578946,0.84210527,0.8684211,0.92105263,0.94736844,0.84210527,0.9736842,0.94736844,1.0
+LightGBM (tuned),0.05263158,0.078947365,0.05263158,0.02631579,0.10526316,0.15789473,0.21052632,0.15789473,0.13157895,0.23684211,0.34210527,0.39473686,0.4473684,0.39473686,0.47368422,0.28947368,0.5,0.47368422,0.47368422,0.47368422,0.42105263,0.4473684,0.55263156,0.39473686,0.5263158,,0.47368422,0.5,0.5,0.57894737,0.6052632,0.65789473,0.6052632,0.5263158,0.65789473,0.6315789,0.65789473,0.6315789,0.68421054,0.6315789,0.68421054,0.7105263,0.68421054,0.7105263,0.7631579,0.7631579,0.7105263,0.68421054,0.7631579,0.7368421,0.84210527,0.81578946,0.7894737,0.8947368,0.7894737,0.81578946,0.81578946,0.84210527,0.84210527,0.84210527,0.8684211,0.8684211,0.92105263,0.8684211,0.84210527,0.84210527,0.84210527,0.8684211,0.92105263,0.8947368,0.92105263,0.94736844,1.0
+XGBoost (tuned),0.15789473,0.078947365,0.18421052,0.15789473,0.15789473,0.23684211,0.21052632,0.21052632,0.21052632,0.23684211,0.31578946,0.31578946,0.36842105,0.31578946,0.34210527,0.36842105,0.39473686,0.39473686,0.39473686,0.47368422,0.31578946,0.5,0.5263158,0.47368422,0.5263158,0.5263158,,0.5263158,0.47368422,0.5,0.55263156,0.55263156,0.6315789,0.55263156,0.55263156,0.57894737,0.65789473,0.6315789,0.6315789,0.6052632,0.68421054,0.68421054,0.6315789,0.68421054,0.84210527,0.68421054,0.81578946,0.7368421,0.7894737,0.81578946,0.8947368,0.81578946,0.7894737,0.84210527,0.84210527,0.84210527,0.81578946,0.7631579,0.8684211,0.84210527,0.84210527,0.8684211,0.92105263,0.8947368,0.8947368,0.81578946,0.84210527,0.8947368,0.92105263,0.8684211,0.94736844,0.92105263,1.0
+TabSwift (default),0.13157895,0.21052632,0.13157895,0.21052632,0.18421052,0.18421052,0.15789473,0.28947368,0.2631579,0.36842105,0.42105263,0.36842105,0.42105263,0.47368422,0.42105263,0.42105263,0.32894737,0.4473684,0.39473686,0.47368422,0.47368422,0.4473684,0.4473684,0.5,0.4473684,0.5,0.47368422,,0.42105263,0.5263158,0.5,0.57894737,0.6052632,0.5263158,0.5263158,0.5,0.6052632,0.57894737,0.68421054,0.65789473,0.57894737,0.65789473,0.65789473,0.7631579,0.6052632,0.6052632,0.65789473,0.65789473,0.6315789,0.7105263,0.7368421,0.7368421,0.7368421,0.65789473,0.7631579,0.7631579,0.8947368,0.7894737,0.7894737,0.84210527,0.84210527,0.8684211,0.81578946,0.8684211,0.8684211,0.84210527,0.9736842,0.84210527,0.8684211,0.8947368,0.8684211,0.9736842,1.0
+TabDPT (tuned),0.05263158,0.18421052,0.10526316,0.23684211,0.23684211,0.18421052,0.2631579,0.28947368,0.23684211,0.36842105,0.42105263,0.39473686,0.5,0.4473684,0.4473684,0.4473684,0.31578946,0.4473684,0.39473686,0.28947368,0.5,0.42105263,0.5,0.4473684,0.5263158,0.5,0.5263158,0.57894737,,0.4473684,0.55263156,0.55263156,0.5263158,0.6052632,0.5263158,0.5263158,0.57894737,0.55263156,0.6052632,0.55263156,0.55263156,0.6315789,0.80263156,0.55263156,0.6052632,0.65789473,0.6315789,0.7105263,0.57894737,0.7631579,0.6315789,0.7631579,0.7631579,0.7105263,0.7631579,0.7631579,0.8947368,0.7894737,0.81578946,0.81578946,0.8684211,0.94736844,0.84210527,0.84210527,0.92105263,0.8947368,0.8684211,0.81578946,0.81578946,0.92105263,0.8684211,0.94736844,0.9736842
+BetaTabPFN (default),0.05263158,0.15789473,0.078947365,0.15789473,0.13157895,0.10526316,0.15789473,0.23684211,0.15789473,0.28947368,0.31578946,0.31578946,0.5,0.39473686,0.42105263,0.4473684,0.47368422,0.47368422,0.36842105,0.4473684,0.5263158,0.42105263,0.57894737,0.42105263,0.6052632,0.42105263,0.5,0.47368422,0.55263156,,0.5263158,0.55263156,0.5263158,0.5263158,0.47368422,0.57894737,0.55263156,0.6052632,0.6315789,0.6052632,0.7368421,0.6052632,0.65789473,0.6315789,0.68421054,0.7105263,0.65789473,0.65789473,0.68421054,0.65789473,0.7368421,0.7631579,0.7368421,0.7105263,0.7631579,0.7631579,0.7894737,0.81578946,0.84210527,0.8684211,0.84210527,0.8684211,0.7894737,0.8947368,0.7631579,0.81578946,0.8947368,0.8947368,0.8947368,0.8684211,0.94736844,0.9736842,1.0
+ChimeraBoost (tuned),0.13157895,0.078947365,0.15789473,0.18421052,0.15789473,0.21052632,0.28947368,0.18421052,0.21052632,0.31578946,0.42105263,0.31578946,0.34210527,0.34210527,0.39473686,0.34210527,0.39473686,0.39473686,0.36842105,0.47368422,0.36842105,0.4473684,0.42105263,0.4473684,0.42105263,0.39473686,0.4473684,0.5,0.4473684,0.47368422,,0.6315789,0.5263158,0.47368422,0.5,0.57894737,0.6315789,0.68421054,0.68421054,0.6315789,0.65789473,0.68421054,0.65789473,0.7105263,0.7105263,0.65789473,0.6315789,0.6315789,0.7105263,0.68421054,0.7894737,0.7368421,0.7368421,0.81578946,0.7894737,0.7894737,0.7894737,0.7631579,0.84210527,0.84210527,0.81578946,0.81578946,0.7894737,0.81578946,0.84210527,0.84210527,0.8947368,0.8684211,0.8684211,0.84210527,0.94736844,0.92105263,1.0
+EBM (tuned + ensembled),0.078947365,0.15789473,0.18421052,0.13157895,0.15789473,0.23684211,0.21052632,0.23684211,0.15789473,0.21052632,0.28947368,0.21052632,0.36842105,0.31578946,0.34210527,0.31578946,0.39473686,0.39473686,0.34210527,0.42105263,0.36842105,0.47368422,0.42105263,0.39473686,0.36842105,0.34210527,0.4473684,0.42105263,0.4473684,0.4473684,0.36842105,,0.5,0.5,0.5,0.57894737,0.57894737,0.6315789,0.5,0.65789473,0.7631579,0.7894737,0.68421054,0.7105263,0.65789473,0.68421054,0.7368421,0.7894737,0.7105263,0.68421054,0.65789473,0.68421054,0.7631579,0.7368421,0.7631579,0.7894737,0.7631579,0.7631579,0.8684211,0.84210527,0.7894737,0.8684211,0.81578946,0.8947368,0.84210527,0.84210527,0.92105263,0.94736844,0.94736844,0.8684211,0.9736842,0.94736844,1.0
+xRFM (tuned + ensembled),0.078947365,0.078947365,0.15789473,0.15789473,0.18421052,0.21052632,0.18421052,0.21052632,0.23684211,0.28947368,0.34210527,0.42105263,0.34210527,0.36842105,0.36842105,0.36842105,0.39473686,0.36842105,0.47368422,0.42105263,0.4473684,0.42105263,0.34210527,0.4473684,0.39473686,0.39473686,0.36842105,0.39473686,0.47368422,0.47368422,0.47368422,0.5,,0.42105263,0.55263156,0.5263158,0.6315789,0.6052632,0.4473684,0.55263156,0.5263158,0.57894737,0.57894737,0.65789473,0.6052632,0.6052632,0.68421054,0.7631579,0.7368421,0.7368421,0.65789473,0.68421054,0.7368421,0.7631579,0.7368421,0.7368421,0.8684211,0.7631579,0.84210527,0.7894737,0.81578946,0.8947368,0.8947368,0.92105263,0.92105263,0.92105263,0.8947368,0.94736844,0.9736842,0.8684211,0.94736844,0.9736842,0.9736842
+iLTM (tuned),0.078947365,0.10526316,0.10526316,0.13157895,0.18421052,0.23684211,0.118421055,0.21052632,0.15789473,0.21052632,0.28947368,0.36842105,0.31578946,0.31578946,0.2631579,0.34210527,0.31578946,0.2631579,0.39473686,0.34210527,0.39473686,0.42105263,0.34210527,0.47368422,0.39473686,0.47368422,0.4473684,0.47368422,0.39473686,0.47368422,0.5263158,0.5,0.57894737,,0.5,0.57894737,0.6315789,0.57894737,0.47368422,0.65789473,0.6052632,0.7105263,0.6315789,0.7368421,0.55263156,0.65789473,0.7368421,0.7105263,0.7894737,0.81578946,0.7105263,0.69736844,0.7631579,0.7894737,0.7631579,0.7631579,0.7368421,0.7631579,0.8947368,0.84210527,0.8684211,0.8684211,0.8684211,0.92105263,0.8947368,0.8947368,0.8684211,0.84210527,0.8947368,0.8947368,0.8947368,0.9736842,1.0
+TabM (default),0.078947365,0.13157895,0.15789473,0.15789473,0.15789473,0.15789473,0.18421052,0.21052632,0.15789473,0.23684211,0.28947368,0.15789473,0.39473686,0.34210527,0.36842105,0.34210527,0.39473686,0.4473684,0.31578946,0.42105263,0.39473686,0.39473686,0.4473684,0.47368422,0.42105263,0.34210527,0.4473684,0.47368422,0.47368422,0.5263158,0.5,0.5,0.4473684,0.5,,0.57894737,0.55263156,0.55263156,0.55263156,0.55263156,0.57894737,0.5394737,0.6315789,0.7105263,0.65789473,0.68421054,0.6052632,0.7105263,0.7105263,0.65789473,0.68421054,0.65789473,0.81578946,0.7368421,0.7368421,0.7631579,0.7894737,0.84210527,0.92105263,0.84210527,0.84210527,0.81578946,0.7894737,0.81578946,0.81578946,0.7894737,0.8684211,0.8947368,0.8684211,0.8947368,0.8947368,0.9736842,0.94736844
+TorchMLP (tuned + ensembled),0.05263158,0.078947365,0.13157895,0.10526316,0.18421052,0.13157895,0.15789473,0.10526316,0.18421052,0.18421052,0.21052632,0.28947368,0.2631579,0.31578946,0.2631579,0.2631579,0.34210527,0.28947368,0.36842105,0.34210527,0.31578946,0.34210527,0.28947368,0.39473686,0.36842105,0.36842105,0.42105263,0.5,0.47368422,0.42105263,0.42105263,0.42105263,0.47368422,0.42105263,0.42105263,,0.55263156,0.4473684,0.4473684,0.5,0.47368422,0.47368422,0.6315789,0.6052632,0.65789473,0.7368421,0.65789473,0.6052632,0.7105263,0.65789473,0.68421054,0.7105263,0.7368421,0.7631579,0.7105263,0.7368421,0.7631579,0.7631579,0.94736844,0.7631579,0.84210527,0.8684211,0.7894737,0.8684211,0.81578946,0.81578946,0.92105263,0.8947368,0.92105263,0.8947368,0.9736842,0.94736844,0.9736842
+xRFM (tuned),0.078947365,0.13157895,0.10526316,0.15789473,0.13157895,0.15789473,0.15789473,0.18421052,0.15789473,0.23684211,0.31578946,0.42105263,0.2631579,0.2631579,0.2631579,0.28947368,0.31578946,0.2631579,0.42105263,0.34210527,0.31578946,0.31578946,0.28947368,0.34210527,0.31578946,0.34210527,0.34210527,0.39473686,0.42105263,0.4473684,0.36842105,0.42105263,0.36842105,0.36842105,0.4473684,0.4473684,,0.5,0.42105263,0.57894737,0.5263158,0.4473684,0.5263158,0.57894737,0.5,0.6315789,0.7105263,0.57894737,0.68421054,0.7368421,0.6315789,0.68421054,0.65789473,0.7368421,0.7105263,0.7368421,0.7894737,0.68421054,0.8684211,0.7894737,0.7631579,0.84210527,0.81578946,0.92105263,0.92105263,0.9736842,0.84210527,0.8947368,0.8684211,0.8947368,0.8947368,0.9736842,1.0
+RealMLP (default),0.10526316,0.13157895,0.078947365,0.10526316,0.15789473,0.18421052,0.18421052,0.21052632,0.15789473,0.10526316,0.18421052,0.18421052,0.28947368,0.2631579,0.2631579,0.31578946,0.36842105,0.2631579,0.34210527,0.36842105,0.28947368,0.34210527,0.23684211,0.28947368,0.31578946,0.36842105,0.36842105,0.42105263,0.4473684,0.39473686,0.31578946,0.36842105,0.39473686,0.42105263,0.4473684,0.55263156,0.5,,0.4473684,0.5,0.55263156,0.5,0.6315789,0.57894737,0.6315789,0.65789473,0.6052632,0.68421054,0.65789473,0.68421054,0.6052632,0.65789473,0.7368421,0.68421054,0.7105263,0.7631579,0.7105263,0.7631579,0.8947368,0.84210527,0.7894737,0.92105263,0.7894737,0.8947368,0.84210527,0.81578946,0.94736844,0.8947368,0.92105263,0.92105263,0.9736842,0.94736844,1.0
+ChimeraBoost (default),0.078947365,0.05263158,0.10526316,0.15789473,0.18421052,0.18421052,0.15789473,0.15789473,0.21052632,0.23684211,0.34210527,0.28947368,0.2631579,0.31578946,0.28947368,0.2631579,0.36842105,0.18421052,0.36842105,0.47368422,0.28947368,0.34210527,0.2631579,0.39473686,0.18421052,0.31578946,0.36842105,0.31578946,0.39473686,0.36842105,0.31578946,0.5,0.55263156,0.5263158,0.4473684,0.55263156,0.57894737,0.55263156,,0.6052632,0.57894737,0.55263156,0.55263156,0.6315789,0.57894737,0.55263156,0.57894737,0.6315789,0.68421054,0.57894737,0.6052632,0.68421054,0.68421054,0.7368421,0.7631579,0.81578946,0.7368421,0.7105263,0.7894737,0.84210527,0.7631579,0.7894737,0.7368421,0.81578946,0.7894737,0.7894737,0.84210527,0.8684211,0.8684211,0.7894737,0.9736842,0.94736844,1.0
+SAP-RPT-OSS (default),0.10526316,0.18421052,0.21052632,0.18421052,0.13157895,0.23684211,0.15789473,0.23684211,0.15789473,0.21052632,0.34210527,0.28947368,0.28947368,0.34210527,0.28947368,0.34210527,0.34210527,0.2631579,0.34210527,0.34210527,0.36842105,0.31578946,0.34210527,0.42105263,0.36842105,0.36842105,0.39473686,0.34210527,0.4473684,0.39473686,0.36842105,0.34210527,0.4473684,0.34210527,0.4473684,0.5,0.42105263,0.5,0.39473686,,0.42105263,0.5,0.55263156,0.6052632,0.47368422,0.57894737,0.6052632,0.57894737,0.55263156,0.68421054,0.6052632,0.6052632,0.7105263,0.57894737,0.65789473,0.65789473,0.68421054,0.7894737,0.7368421,0.84210527,0.81578946,0.84210527,0.7631579,0.7894737,0.81578946,0.81578946,0.8684211,0.8684211,0.84210527,0.84210527,0.8684211,0.8947368,1.0
+EBM (tuned),0.078947365,0.13157895,0.10526316,0.078947365,0.15789473,0.13157895,0.10526316,0.10526316,0.15789473,0.18421052,0.2631579,0.2631579,0.34210527,0.2631579,0.28947368,0.21052632,0.34210527,0.31578946,0.28947368,0.34210527,0.31578946,0.34210527,0.36842105,0.39473686,0.31578946,0.31578946,0.31578946,0.42105263,0.4473684,0.2631579,0.34210527,0.23684211,0.47368422,0.39473686,0.42105263,0.5263158,0.47368422,0.4473684,0.42105263,0.57894737,,0.5,0.6052632,0.57894737,0.5263158,0.55263156,0.6052632,0.65789473,0.6315789,0.65789473,0.57894737,0.68421054,0.7105263,0.65789473,0.7631579,0.7631579,0.7368421,0.7631579,0.81578946,0.84210527,0.7631579,0.81578946,0.7894737,0.8947368,0.81578946,0.84210527,0.8947368,0.84210527,0.8947368,0.84210527,0.8947368,0.94736844,1.0
+EBM (default),0.078947365,0.10526316,0.10526316,0.13157895,0.13157895,0.10526316,0.13157895,0.13157895,0.15789473,0.18421052,0.2631579,0.21052632,0.2631579,0.2631579,0.23684211,0.18421052,0.28947368,0.23684211,0.31578946,0.28947368,0.28947368,0.36842105,0.28947368,0.34210527,0.34210527,0.28947368,0.31578946,0.34210527,0.36842105,0.39473686,0.31578946,0.21052632,0.42105263,0.28947368,0.46052632,0.5263158,0.55263156,0.5,0.4473684,0.5,0.5,,0.55263156,0.6315789,0.47368422,0.6315789,0.65789473,0.68421054,0.65789473,0.68421054,0.57894737,0.68421054,0.7105263,0.7105263,0.7368421,0.7368421,0.7631579,0.81578946,0.84210527,0.81578946,0.84210527,0.84210527,0.7368421,0.8947368,0.81578946,0.7894737,0.8947368,0.8684211,0.8947368,0.8947368,0.94736844,0.94736844,1.0
+TabDPT (default),0.02631579,0.15789473,0.10526316,0.18421052,0.18421052,0.13157895,0.15789473,0.15789473,0.18421052,0.21052632,0.31578946,0.23684211,0.31578946,0.34210527,0.28947368,0.28947368,0.18421052,0.31578946,0.2631579,0.23684211,0.31578946,0.31578946,0.28947368,0.31578946,0.31578946,0.31578946,0.36842105,0.34210527,0.19736843,0.34210527,0.34210527,0.31578946,0.42105263,0.36842105,0.36842105,0.36842105,0.47368422,0.36842105,0.4473684,0.4473684,0.39473686,0.4473684,,0.5263158,0.5,0.4473684,0.5,0.6052632,0.55263156,0.57894737,0.6052632,0.57894737,0.7105263,0.6052632,0.6315789,0.65789473,0.7631579,0.7105263,0.7105263,0.7105263,0.7894737,0.81578946,0.7105263,0.7631579,0.7894737,0.7894737,0.84210527,0.7105263,0.7368421,0.81578946,0.81578946,0.8684211,0.9736842
+FastaiMLP (tuned + ensembled),0.10526316,0.13157895,0.13157895,0.10526316,0.15789473,0.05263158,0.10526316,0.10526316,0.18421052,0.10526316,0.21052632,0.21052632,0.18421052,0.34210527,0.2631579,0.23684211,0.28947368,0.2631579,0.31578946,0.34210527,0.28947368,0.32894737,0.23684211,0.2631579,0.28947368,0.28947368,0.31578946,0.23684211,0.4473684,0.36842105,0.28947368,0.28947368,0.34210527,0.2631579,0.28947368,0.39473686,0.42105263,0.42105263,0.36842105,0.39473686,0.42105263,0.36842105,0.47368422,,0.4473684,0.55263156,0.57894737,0.6052632,0.55263156,0.57894737,0.5263158,0.6052632,0.7368421,0.6052632,0.6315789,0.6315789,0.68421054,0.7368421,0.7105263,0.7368421,0.7631579,0.9736842,0.7368421,0.81578946,0.7894737,0.81578946,0.8684211,0.81578946,0.84210527,0.8947368,0.8947368,0.8947368,0.9736842
+XGBoost (default),0.05263158,0.078947365,0.05263158,0.05263158,0.05263158,0.13157895,0.10526316,0.10526316,0.078947365,0.13157895,0.18421052,0.2631579,0.21052632,0.2631579,0.18421052,0.13157895,0.2631579,0.18421052,0.2631579,0.31578946,0.13157895,0.28947368,0.2631579,0.2631579,0.31578946,0.23684211,0.15789473,0.39473686,0.39473686,0.31578946,0.28947368,0.34210527,0.39473686,0.4473684,0.34210527,0.34210527,0.5,0.36842105,0.42105263,0.5263158,0.47368422,0.5263158,0.5,0.55263156,,0.5263158,0.6315789,0.6052632,0.6052632,0.6315789,0.6052632,0.7368421,0.6052632,0.6315789,0.68421054,0.68421054,0.68421054,0.68421054,0.7368421,0.7631579,0.7631579,0.7631579,0.7894737,0.7631579,0.7368421,0.7894737,0.7631579,0.81578946,0.84210527,0.84210527,0.84210527,0.8684211,0.94736844
+TorchMLP (tuned),0.02631579,0.10526316,0.05263158,0.05263158,0.13157895,0.10526316,0.13157895,0.078947365,0.10526316,0.18421052,0.18421052,0.21052632,0.18421052,0.28947368,0.18421052,0.21052632,0.23684211,0.18421052,0.23684211,0.31578946,0.3026316,0.23684211,0.21052632,0.23684211,0.28947368,0.23684211,0.31578946,0.39473686,0.34210527,0.28947368,0.34210527,0.31578946,0.39473686,0.34210527,0.31578946,0.2631579,0.36842105,0.34210527,0.4473684,0.42105263,0.4473684,0.36842105,0.55263156,0.4473684,0.47368422,,0.5394737,0.6315789,0.57894737,0.6052632,0.56578946,0.6315789,0.55263156,0.57894737,0.7368421,0.7105263,0.7105263,0.7631579,0.8947368,0.81578946,0.7894737,0.7894737,0.68421054,0.84210527,0.7631579,0.7894737,0.92105263,0.81578946,0.84210527,0.84210527,0.92105263,0.94736844,0.9736842
+ExtraTrees (tuned + ensembled),0.05263158,0.078947365,0.05263158,0.078947365,0.05263158,0.13157895,0.15789473,0.15789473,0.02631579,0.18421052,0.23684211,0.28947368,0.18421052,0.2631579,0.23684211,0.18421052,0.23684211,0.18421052,0.31578946,0.28947368,0.18421052,0.28947368,0.2631579,0.2631579,0.34210527,0.28947368,0.18421052,0.34210527,0.36842105,0.34210527,0.36842105,0.2631579,0.31578946,0.2631579,0.39473686,0.34210527,0.28947368,0.39473686,0.42105263,0.39473686,0.39473686,0.34210527,0.5,0.42105263,0.36842105,0.46052632,,0.5,0.57894737,0.69736844,0.5263158,0.55263156,0.5,0.68421054,0.65789473,0.68421054,0.7368421,0.65789473,0.6315789,0.7368421,0.7105263,0.7105263,0.8684211,0.8947368,0.92105263,0.7894737,0.81578946,0.7631579,0.84210527,0.7894737,0.84210527,0.92105263,1.0
+ModernNCA (default),0.05263158,0.05263158,0.05263158,0.05263158,0.13157895,0.078947365,0.078947365,0.078947365,0.10526316,0.10526316,0.23684211,0.15789473,0.18421052,0.2631579,0.23684211,0.18421052,0.31578946,0.23684211,0.15789473,0.18421052,0.21052632,0.15789473,0.28947368,0.18421052,0.31578946,0.31578946,0.2631579,0.34210527,0.28947368,0.34210527,0.36842105,0.21052632,0.23684211,0.28947368,0.28947368,0.39473686,0.42105263,0.31578946,0.36842105,0.42105263,0.34210527,0.31578946,0.39473686,0.39473686,0.39473686,0.36842105,0.5,,0.57894737,0.5,0.5,0.57894737,0.55263156,0.6315789,0.7368421,0.7368421,0.7105263,0.7105263,0.81578946,0.7894737,0.7368421,0.7894737,0.81578946,0.81578946,0.81578946,0.7105263,0.7631579,0.81578946,0.81578946,0.84210527,0.8684211,0.92105263,0.9736842
+RandomForest (tuned + ensembled),0.05263158,0.05263158,0.078947365,0.05263158,0.078947365,0.10526316,0.15789473,0.13157895,0.078947365,0.21052632,0.15789473,0.21052632,0.13157895,0.2631579,0.18421052,0.18421052,0.23684211,0.15789473,0.2631579,0.28947368,0.15789473,0.2631579,0.18421052,0.2631579,0.23684211,0.23684211,0.21052632,0.36842105,0.42105263,0.31578946,0.28947368,0.28947368,0.2631579,0.21052632,0.28947368,0.28947368,0.31578946,0.34210527,0.31578946,0.4473684,0.36842105,0.34210527,0.4473684,0.4473684,0.39473686,0.42105263,0.42105263,0.42105263,,0.57894737,0.5,0.5,0.55263156,0.7105263,0.65789473,0.68421054,0.6315789,0.6052632,0.7368421,0.7631579,0.7105263,0.7105263,0.8684211,0.7894737,0.84210527,0.7894737,0.7105263,0.7368421,0.7105263,0.7105263,0.81578946,0.92105263,0.9736842
+ExtraTrees (tuned),0.05263158,0.078947365,0.05263158,0.078947365,0.05263158,0.078947365,0.15789473,0.15789473,0.10526316,0.18421052,0.15789473,0.28947368,0.21052632,0.2631579,0.2631579,0.23684211,0.21052632,0.21052632,0.31578946,0.18421052,0.21052632,0.2631579,0.31578946,0.28947368,0.31578946,0.2631579,0.18421052,0.28947368,0.23684211,0.34210527,0.31578946,0.31578946,0.2631579,0.18421052,0.34210527,0.34210527,0.2631579,0.31578946,0.42105263,0.31578946,0.34210527,0.31578946,0.42105263,0.42105263,0.36842105,0.39473686,0.3026316,0.5,0.42105263,,0.4473684,0.47368422,0.47368422,0.57894737,0.6315789,0.6315789,0.7105263,0.65789473,0.57894737,0.7631579,0.68421054,0.6315789,0.81578946,0.7894737,0.81578946,0.7105263,0.7894737,0.68421054,0.68421054,0.7631579,0.84210527,0.9736842,0.9736842
+LightGBM (default),0.05263158,0.02631579,0.05263158,0.05263158,0.02631579,0.05263158,0.02631579,0.078947365,0.05263158,0.078947365,0.15789473,0.15789473,0.15789473,0.2631579,0.18421052,0.05263158,0.18421052,0.18421052,0.21052632,0.23684211,0.02631579,0.21052632,0.23684211,0.15789473,0.23684211,0.15789473,0.10526316,0.2631579,0.36842105,0.2631579,0.21052632,0.34210527,0.34210527,0.28947368,0.31578946,0.31578946,0.36842105,0.39473686,0.39473686,0.39473686,0.42105263,0.42105263,0.39473686,0.47368422,0.39473686,0.43421054,0.47368422,0.5,0.5,0.55263156,,0.6315789,0.55263156,0.5263158,0.65789473,0.68421054,0.6315789,0.68421054,0.68421054,0.7631579,0.7105263,0.7368421,0.7105263,0.7105263,0.7631579,0.7631579,0.7631579,0.7105263,0.7105263,0.7894737,0.8684211,0.92105263,0.94736844
+iLTM (default),0.078947365,0.05263158,0.02631579,0.05263158,0.10526316,0.15789473,0.10526316,0.078947365,0.13157895,0.10526316,0.13157895,0.21052632,0.13157895,0.23684211,0.15789473,0.10526316,0.21052632,0.15789473,0.28947368,0.2631579,0.15789473,0.23684211,0.18421052,0.23684211,0.2631579,0.18421052,0.18421052,0.2631579,0.23684211,0.23684211,0.2631579,0.31578946,0.31578946,0.3026316,0.34210527,0.28947368,0.31578946,0.34210527,0.31578946,0.39473686,0.31578946,0.31578946,0.42105263,0.39473686,0.2631579,0.36842105,0.4473684,0.42105263,0.5,0.5263158,0.36842105,,0.47368422,0.47368422,0.68421054,0.65789473,0.6052632,0.6052632,0.65789473,0.7631579,0.57894737,0.65789473,0.65789473,0.7894737,0.68421054,0.68421054,0.7105263,0.7105263,0.7105263,0.65789473,0.81578946,0.84210527,0.94736844
+FastaiMLP (tuned),0.078947365,0.078947365,0.05263158,0.10526316,0.078947365,0.078947365,0.02631579,0.078947365,0.10526316,0.13157895,0.15789473,0.15789473,0.18421052,0.21052632,0.18421052,0.18421052,0.13157895,0.15789473,0.18421052,0.21052632,0.23684211,0.21052632,0.18421052,0.18421052,0.28947368,0.21052632,0.21052632,0.2631579,0.23684211,0.2631579,0.2631579,0.23684211,0.2631579,0.23684211,0.18421052,0.2631579,0.34210527,0.2631579,0.31578946,0.28947368,0.28947368,0.28947368,0.28947368,0.2631579,0.39473686,0.4473684,0.5,0.4473684,0.4473684,0.5263158,0.4473684,0.5263158,,0.5,0.57894737,0.57894737,0.57894737,0.6315789,0.6052632,0.7105263,0.68421054,0.81578946,0.65789473,0.7105263,0.7368421,0.7368421,0.7894737,0.7105263,0.7105263,0.7631579,0.84210527,0.8684211,0.94736844
+RandomForest (tuned),0.05263158,0.05263158,0.02631579,0.078947365,0.05263158,0.10526316,0.078947365,0.05263158,0.078947365,0.13157895,0.13157895,0.21052632,0.13157895,0.21052632,0.13157895,0.078947365,0.18421052,0.10526316,0.21052632,0.2631579,0.10526316,0.21052632,0.15789473,0.28947368,0.18421052,0.10526316,0.15789473,0.34210527,0.28947368,0.28947368,0.18421052,0.2631579,0.23684211,0.21052632,0.2631579,0.23684211,0.2631579,0.31578946,0.2631579,0.42105263,0.34210527,0.28947368,0.39473686,0.39473686,0.36842105,0.42105263,0.31578946,0.36842105,0.28947368,0.42105263,0.47368422,0.5263158,0.5,,0.6315789,0.6315789,0.65789473,0.6052632,0.6315789,0.7368421,0.65789473,0.7631579,0.80263156,0.7894737,0.7894737,0.7894737,0.7631579,0.7105263,0.68421054,0.68421054,0.7894737,0.8947368,0.9736842
+TabSTAR (tuned),0.078947365,0.13157895,0.13157895,0.15789473,0.15789473,0.13157895,0.10526316,0.21052632,0.15789473,0.18421052,0.18421052,0.15789473,0.23684211,0.25,0.21052632,0.18421052,0.23684211,0.18421052,0.21052632,0.21052632,0.15789473,0.21052632,0.18421052,0.28947368,0.18421052,0.21052632,0.15789473,0.23684211,0.23684211,0.23684211,0.21052632,0.23684211,0.2631579,0.23684211,0.2631579,0.28947368,0.28947368,0.28947368,0.23684211,0.34210527,0.23684211,0.2631579,0.36842105,0.36842105,0.31578946,0.2631579,0.34210527,0.2631579,0.34210527,0.36842105,0.34210527,0.31578946,0.42105263,0.36842105,,0.6447368,0.5,0.6315789,0.47368422,0.94736844,0.6315789,0.6052632,0.65789473,0.6315789,0.65789473,0.65789473,0.68421054,0.65789473,0.68421054,0.65789473,0.7631579,0.84210527,0.92105263
+TabSTAR (tuned + ensembled),0.10526316,0.13157895,0.13157895,0.15789473,0.15789473,0.13157895,0.10526316,0.18421052,0.15789473,0.18421052,0.15789473,0.13157895,0.23684211,0.2631579,0.18421052,0.18421052,0.23684211,0.18421052,0.18421052,0.23684211,0.15789473,0.21052632,0.18421052,0.2631579,0.18421052,0.18421052,0.15789473,0.23684211,0.23684211,0.23684211,0.21052632,0.21052632,0.2631579,0.23684211,0.23684211,0.2631579,0.2631579,0.23684211,0.18421052,0.34210527,0.23684211,0.2631579,0.34210527,0.36842105,0.31578946,0.28947368,0.31578946,0.2631579,0.31578946,0.36842105,0.31578946,0.34210527,0.42105263,0.36842105,0.35526314,,0.5263158,0.6315789,0.5,0.9736842,0.6315789,0.57894737,0.6315789,0.6052632,0.6315789,0.6315789,0.65789473,0.65789473,0.65789473,0.65789473,0.7631579,0.84210527,0.92105263
+OrionMSP (default),0.02631579,0.0,0.0,0.02631579,0.02631579,0.078947365,0.0,0.078947365,0.0,0.05263158,0.17105263,0.13157895,0.15789473,0.17105263,0.18421052,0.15789473,0.078947365,0.21052632,0.15789473,0.05263158,0.15789473,0.15789473,0.18421052,0.18421052,0.21052632,0.18421052,0.18421052,0.10526316,0.10526316,0.21052632,0.21052632,0.23684211,0.13157895,0.2631579,0.21052632,0.23684211,0.21052632,0.28947368,0.2631579,0.31578946,0.2631579,0.23684211,0.23684211,0.31578946,0.31578946,0.28947368,0.2631579,0.28947368,0.36842105,0.28947368,0.36842105,0.39473686,0.42105263,0.34210527,0.5,0.47368422,,0.55263156,0.47368422,0.68421054,0.55263156,0.5263158,0.6052632,0.6052632,0.68421054,0.6315789,0.7105263,0.68421054,0.68421054,0.68421054,0.7368421,0.7368421,0.9736842
+PerpetualBooster (tuned + ensembled),0.0,0.078947365,0.02631579,0.02631579,0.05263158,0.05263158,0.05263158,0.05263158,0.05263158,0.05263158,0.10526316,0.15789473,0.10526316,0.18421052,0.15789473,0.13157895,0.18421052,0.10526316,0.10526316,0.21052632,0.21052632,0.15789473,0.13157895,0.18421052,0.21052632,0.15789473,0.23684211,0.21052632,0.21052632,0.18421052,0.23684211,0.23684211,0.23684211,0.23684211,0.15789473,0.23684211,0.31578946,0.23684211,0.28947368,0.21052632,0.23684211,0.18421052,0.28947368,0.2631579,0.31578946,0.23684211,0.34210527,0.28947368,0.39473686,0.34210527,0.31578946,0.39473686,0.36842105,0.39473686,0.36842105,0.36842105,0.4473684,,0.42105263,0.36842105,0.81578946,0.5263158,0.6052632,0.55263156,0.55263156,0.57894737,0.6052632,0.65789473,0.65789473,0.8684211,0.68421054,0.68421054,0.92105263
+TorchMLP (default),0.02631579,0.05263158,0.05263158,0.02631579,0.05263158,0.02631579,0.02631579,0.02631579,0.05263158,0.078947365,0.05263158,0.02631579,0.05263158,0.18421052,0.078947365,0.10526316,0.15789473,0.078947365,0.02631579,0.18421052,0.10526316,0.15789473,0.05263158,0.18421052,0.15789473,0.15789473,0.13157895,0.21052632,0.18421052,0.15789473,0.15789473,0.13157895,0.15789473,0.10526316,0.078947365,0.05263158,0.13157895,0.10526316,0.21052632,0.2631579,0.18421052,0.15789473,0.28947368,0.28947368,0.2631579,0.10526316,0.36842105,0.18421052,0.2631579,0.42105263,0.31578946,0.34210527,0.39473686,0.36842105,0.5263158,0.5,0.5263158,0.57894737,,0.65789473,0.6052632,0.57894737,0.57894737,0.65789473,0.65789473,0.6052632,0.68421054,0.6315789,0.7105263,0.7105263,0.7894737,0.84210527,0.94736844
+TabSTAR (default),0.05263158,0.13157895,0.13157895,0.15789473,0.13157895,0.10526316,0.10526316,0.13157895,0.13157895,0.15789473,0.15789473,0.13157895,0.15789473,0.18421052,0.18421052,0.15789473,0.15789473,0.18421052,0.15789473,0.15789473,0.15789473,0.10526316,0.15789473,0.15789473,0.15789473,0.15789473,0.15789473,0.15789473,0.18421052,0.13157895,0.15789473,0.15789473,0.21052632,0.15789473,0.15789473,0.23684211,0.21052632,0.15789473,0.15789473,0.15789473,0.15789473,0.18421052,0.28947368,0.2631579,0.23684211,0.18421052,0.2631579,0.21052632,0.23684211,0.23684211,0.23684211,0.23684211,0.28947368,0.2631579,0.05263158,0.02631579,0.31578946,0.6315789,0.34210527,,0.6315789,0.31578946,0.5,0.5263158,0.47368422,0.5,0.5263158,0.5,0.5,0.65789473,0.57894737,0.68421054,0.8947368
+PerpetualBooster (tuned),0.0,0.05263158,0.0,0.0,0.02631579,0.02631579,0.0,0.02631579,0.02631579,0.02631579,0.10526316,0.078947365,0.05263158,0.18421052,0.078947365,0.10526316,0.13157895,0.10526316,0.078947365,0.13157895,0.15789473,0.13157895,0.078947365,0.18421052,0.15789473,0.13157895,0.15789473,0.15789473,0.13157895,0.15789473,0.18421052,0.21052632,0.18421052,0.13157895,0.15789473,0.15789473,0.23684211,0.21052632,0.23684211,0.18421052,0.23684211,0.15789473,0.21052632,0.23684211,0.23684211,0.21052632,0.28947368,0.2631579,0.28947368,0.31578946,0.28947368,0.42105263,0.31578946,0.34210527,0.36842105,0.36842105,0.4473684,0.18421052,0.39473686,0.36842105,,0.5,0.5263158,0.47368422,0.47368422,0.57894737,0.57894737,0.6315789,0.6052632,0.7763158,0.65789473,0.57894737,0.92105263
+FastaiMLP (default),0.0,0.02631579,0.0,0.02631579,0.02631579,0.0,0.05263158,0.02631579,0.05263158,0.02631579,0.05263158,0.078947365,0.05263158,0.10526316,0.10526316,0.10526316,0.05263158,0.10526316,0.13157895,0.05263158,0.13157895,0.13157895,0.078947365,0.15789473,0.15789473,0.13157895,0.13157895,0.13157895,0.05263158,0.13157895,0.18421052,0.13157895,0.10526316,0.13157895,0.18421052,0.13157895,0.15789473,0.078947365,0.21052632,0.15789473,0.18421052,0.15789473,0.18421052,0.02631579,0.23684211,0.21052632,0.28947368,0.21052632,0.28947368,0.36842105,0.2631579,0.34210527,0.18421052,0.23684211,0.39473686,0.42105263,0.47368422,0.47368422,0.42105263,0.68421054,0.5,,0.5,0.55263156,0.5,0.6052632,0.57894737,0.57894737,0.6052632,0.7105263,0.65789473,0.7894737,0.94736844
+RandomForest (default),0.02631579,0.05263158,0.02631579,0.02631579,0.02631579,0.05263158,0.05263158,0.02631579,0.02631579,0.05263158,0.078947365,0.15789473,0.05263158,0.078947365,0.078947365,0.02631579,0.078947365,0.05263158,0.13157895,0.05263158,0.05263158,0.078947365,0.078947365,0.13157895,0.15789473,0.078947365,0.078947365,0.18421052,0.15789473,0.21052632,0.21052632,0.18421052,0.10526316,0.13157895,0.21052632,0.21052632,0.18421052,0.21052632,0.2631579,0.23684211,0.21052632,0.2631579,0.28947368,0.2631579,0.21052632,0.31578946,0.13157895,0.18421052,0.13157895,0.18421052,0.28947368,0.34210527,0.34210527,0.19736843,0.34210527,0.36842105,0.39473686,0.39473686,0.42105263,0.5,0.47368422,0.5,,0.5263158,0.5263158,0.57894737,0.5,0.6052632,0.6052632,0.57894737,0.65789473,0.7105263,0.9736842
+KNN (tuned + ensembled),0.078947365,0.02631579,0.05263158,0.02631579,0.078947365,0.05263158,0.078947365,0.078947365,0.078947365,0.05263158,0.078947365,0.13157895,0.05263158,0.13157895,0.10526316,0.078947365,0.078947365,0.13157895,0.13157895,0.10526316,0.13157895,0.078947365,0.078947365,0.14473684,0.15789473,0.13157895,0.10526316,0.13157895,0.15789473,0.10526316,0.18421052,0.10526316,0.078947365,0.078947365,0.18421052,0.13157895,0.078947365,0.10526316,0.18421052,0.21052632,0.10526316,0.10526316,0.23684211,0.18421052,0.23684211,0.15789473,0.10526316,0.18421052,0.21052632,0.21052632,0.28947368,0.21052632,0.28947368,0.21052632,0.36842105,0.39473686,0.39473686,0.4473684,0.34210527,0.47368422,0.5263158,0.4473684,0.47368422,,0.55263156,0.6315789,0.55263156,0.6052632,0.6052632,0.6052632,0.68421054,0.9736842,0.9736842
+ExtraTrees (default),0.02631579,0.02631579,0.02631579,0.05263158,0.02631579,0.078947365,0.0,0.02631579,0.02631579,0.02631579,0.05263158,0.15789473,0.05263158,0.05263158,0.13157895,0.10526316,0.05263158,0.13157895,0.18421052,0.02631579,0.078947365,0.13157895,0.15789473,0.15789473,0.18421052,0.15789473,0.10526316,0.13157895,0.078947365,0.23684211,0.15789473,0.15789473,0.078947365,0.10526316,0.18421052,0.18421052,0.078947365,0.15789473,0.21052632,0.18421052,0.18421052,0.18421052,0.21052632,0.21052632,0.2631579,0.23684211,0.078947365,0.18421052,0.15789473,0.18421052,0.23684211,0.31578946,0.2631579,0.21052632,0.34210527,0.36842105,0.31578946,0.4473684,0.34210527,0.5263158,0.5263158,0.5,0.47368422,0.4473684,,0.55263156,0.5263158,0.55263156,0.55263156,0.6052632,0.6052632,0.7105263,0.94736844
+xRFM (default),0.05263158,0.02631579,0.05263158,0.05263158,0.05263158,0.078947365,0.05263158,0.078947365,0.05263158,0.05263158,0.05263158,0.13157895,0.078947365,0.078947365,0.13157895,0.15789473,0.10526316,0.10526316,0.15789473,0.05263158,0.15789473,0.10526316,0.10526316,0.10526316,0.15789473,0.15789473,0.18421052,0.15789473,0.10526316,0.18421052,0.15789473,0.15789473,0.078947365,0.10526316,0.21052632,0.18421052,0.02631579,0.18421052,0.21052632,0.18421052,0.15789473,0.21052632,0.21052632,0.18421052,0.21052632,0.21052632,0.21052632,0.28947368,0.21052632,0.28947368,0.23684211,0.31578946,0.2631579,0.21052632,0.34210527,0.36842105,0.36842105,0.42105263,0.39473686,0.5,0.42105263,0.39473686,0.42105263,0.36842105,0.4473684,,0.5,0.47368422,0.47368422,0.5,0.5263158,0.6315789,0.84210527
+TabFlex (default),0.0,0.05263158,0.05263158,0.078947365,0.10526316,0.02631579,0.0,0.05263158,0.10526316,0.10526316,0.13157895,0.078947365,0.078947365,0.15789473,0.13157895,0.13157895,0.05263158,0.15789473,0.078947365,0.078947365,0.15789473,0.13157895,0.13157895,0.18421052,0.13157895,0.15789473,0.15789473,0.02631579,0.13157895,0.10526316,0.10526316,0.078947365,0.10526316,0.13157895,0.13157895,0.078947365,0.15789473,0.05263158,0.15789473,0.13157895,0.10526316,0.10526316,0.15789473,0.13157895,0.23684211,0.078947365,0.18421052,0.23684211,0.28947368,0.21052632,0.23684211,0.28947368,0.21052632,0.23684211,0.31578946,0.34210527,0.28947368,0.39473686,0.31578946,0.47368422,0.42105263,0.42105263,0.5,0.4473684,0.47368422,0.5,,0.55263156,0.55263156,0.55263156,0.6315789,0.68421054,0.8684211
+Linear (tuned + ensembled),0.02631579,0.02631579,0.05263158,0.02631579,0.078947365,0.05263158,0.05263158,0.02631579,0.078947365,0.02631579,0.05263158,0.05263158,0.078947365,0.10526316,0.02631579,0.02631579,0.15789473,0.02631579,0.10526316,0.13157895,0.078947365,0.10526316,0.05263158,0.18421052,0.078947365,0.13157895,0.10526316,0.15789473,0.18421052,0.10526316,0.13157895,0.05263158,0.05263158,0.15789473,0.10526316,0.10526316,0.10526316,0.10526316,0.13157895,0.13157895,0.15789473,0.13157895,0.28947368,0.18421052,0.18421052,0.18421052,0.23684211,0.18421052,0.2631579,0.31578946,0.28947368,0.28947368,0.28947368,0.28947368,0.34210527,0.34210527,0.31578946,0.34210527,0.36842105,0.5,0.36842105,0.42105263,0.39473686,0.39473686,0.4473684,0.5263158,0.4473684,,0.75,0.5,0.7368421,0.6315789,0.8684211
+Linear (tuned),0.02631579,0.02631579,0.02631579,0.02631579,0.05263158,0.02631579,0.05263158,0.02631579,0.078947365,0.02631579,0.039473683,0.05263158,0.05263158,0.10526316,0.02631579,0.02631579,0.15789473,0.02631579,0.10526316,0.13157895,0.05263158,0.078947365,0.02631579,0.15789473,0.05263158,0.078947365,0.078947365,0.13157895,0.18421052,0.10526316,0.13157895,0.05263158,0.02631579,0.10526316,0.13157895,0.078947365,0.13157895,0.078947365,0.13157895,0.15789473,0.10526316,0.10526316,0.2631579,0.15789473,0.15789473,0.15789473,0.15789473,0.18421052,0.28947368,0.31578946,0.28947368,0.28947368,0.28947368,0.31578946,0.31578946,0.34210527,0.31578946,0.34210527,0.28947368,0.5,0.39473686,0.39473686,0.39473686,0.39473686,0.4473684,0.5263158,0.4473684,0.25,,0.5,0.65789473,0.6052632,0.84210527
+PerpetualBooster (default),0.0,0.05263158,0.02631579,0.02631579,0.05263158,0.02631579,0.05263158,0.02631579,0.05263158,0.02631579,0.02631579,0.05263158,0.05263158,0.13157895,0.05263158,0.10526316,0.05263158,0.05263158,0.05263158,0.10526316,0.13157895,0.13157895,0.05263158,0.10526316,0.15789473,0.10526316,0.13157895,0.10526316,0.078947365,0.13157895,0.15789473,0.13157895,0.13157895,0.10526316,0.10526316,0.10526316,0.10526316,0.078947365,0.21052632,0.15789473,0.15789473,0.10526316,0.18421052,0.10526316,0.15789473,0.15789473,0.21052632,0.15789473,0.28947368,0.23684211,0.21052632,0.34210527,0.23684211,0.31578946,0.34210527,0.34210527,0.31578946,0.13157895,0.28947368,0.34210527,0.2236842,0.28947368,0.42105263,0.39473686,0.39473686,0.5,0.4473684,0.5,0.5,,0.5263158,0.5,0.84210527
+Linear (default),0.0,0.02631579,0.0,0.02631579,0.05263158,0.0,0.02631579,0.0,0.05263158,0.02631579,0.05263158,0.05263158,0.05263158,0.10526316,0.02631579,0.05263158,0.10526316,0.02631579,0.05263158,0.10526316,0.05263158,0.078947365,0.02631579,0.15789473,0.02631579,0.078947365,0.05263158,0.13157895,0.13157895,0.05263158,0.05263158,0.02631579,0.05263158,0.10526316,0.10526316,0.02631579,0.10526316,0.02631579,0.02631579,0.13157895,0.10526316,0.05263158,0.18421052,0.10526316,0.15789473,0.078947365,0.15789473,0.13157895,0.18421052,0.15789473,0.13157895,0.18421052,0.15789473,0.21052632,0.23684211,0.23684211,0.2631579,0.31578946,0.21052632,0.42105263,0.34210527,0.34210527,0.34210527,0.31578946,0.39473686,0.47368422,0.36842105,0.2631579,0.34210527,0.47368422,,0.6052632,0.7894737
+KNN (tuned),0.02631579,0.0,0.02631579,0.02631579,0.05263158,0.02631579,0.02631579,0.0,0.05263158,0.02631579,0.02631579,0.0,0.05263158,0.13157895,0.02631579,0.0,0.02631579,0.05263158,0.02631579,0.05263158,0.078947365,0.078947365,0.02631579,0.13157895,0.05263158,0.05263158,0.078947365,0.02631579,0.05263158,0.02631579,0.078947365,0.05263158,0.02631579,0.02631579,0.02631579,0.05263158,0.02631579,0.05263158,0.05263158,0.10526316,0.05263158,0.05263158,0.13157895,0.10526316,0.13157895,0.05263158,0.078947365,0.078947365,0.078947365,0.02631579,0.078947365,0.15789473,0.13157895,0.10526316,0.15789473,0.15789473,0.2631579,0.31578946,0.15789473,0.31578946,0.42105263,0.21052632,0.28947368,0.02631579,0.28947368,0.36842105,0.31578946,0.36842105,0.39473686,0.5,0.39473686,,0.94736844
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.02631579,0.0,0.0,0.0,0.0,0.02631579,0.0,0.0,0.0,0.0,0.02631579,0.0,0.0,0.0,0.0,0.02631579,0.0,0.0,0.0,0.02631579,0.0,0.05263158,0.02631579,0.0,0.0,0.0,0.0,0.0,0.0,0.02631579,0.02631579,0.05263158,0.02631579,0.0,0.02631579,0.02631579,0.02631579,0.05263158,0.05263158,0.05263158,0.02631579,0.078947365,0.078947365,0.02631579,0.078947365,0.05263158,0.10526316,0.078947365,0.05263158,0.02631579,0.02631579,0.05263158,0.15789473,0.13157895,0.13157895,0.15789473,0.15789473,0.21052632,0.05263158,
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
index 0ee4515b019054fc1629916c890a4918c231a35d..9781f3353c131da63179db4f355da741b33b1f21 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Classification | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..d7acd5f2de9c5e57f43588699aaf1632d69b0975
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_classification/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2102.0,"score":0.769,"rank":3.9,"hrank":2.23,"imp":3.082,"train_time":145.29,"predict_time":1.858,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+238\/-74","imputed":false},{"position":1,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2030.0,"score":0.732,"rank":5.2,"hrank":2.91,"imp":4.491,"train_time":307.34,"predict_time":2.186,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-65","imputed":false},{"position":2,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1929.0,"score":0.71,"rank":7.7,"hrank":2.42,"imp":6.449,"train_time":38.94,"predict_time":18.762,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+324\/-148","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1847.0,"score":0.635,"rank":10.4,"hrank":2.84,"imp":6.0,"train_time":1.11,"predict_time":0.291,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+404\/-179","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1827.0,"score":0.532,"rank":11.2,"hrank":8.1,"imp":8.539,"train_time":2.33,"predict_time":0.699,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+162\/-77","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1773.0,"score":0.475,"rank":13.5,"hrank":6.55,"imp":9.506,"train_time":0.62,"predict_time":0.172,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+238\/-136","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1717.0,"score":0.431,"rank":16.2,"hrank":8.15,"imp":8.142,"train_time":780.95,"predict_time":11.64,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+240\/-126","imputed":false},{"position":7,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1688.0,"score":0.459,"rank":17.7,"hrank":10.29,"imp":8.22,"train_time":2335.83,"predict_time":14.145,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+195\/-148","imputed":false},{"position":8,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1680.0,"score":0.395,"rank":18.1,"hrank":13.48,"imp":10.342,"train_time":202.54,"predict_time":0.43,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-101","imputed":false},{"position":9,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1675.0,"score":0.423,"rank":18.4,"hrank":12.2,"imp":10.878,"train_time":1598.12,"predict_time":70.164,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+162\/-118","imputed":false},{"position":10,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1662.0,"score":0.426,"rank":19.15,"hrank":10.16,"imp":9.741,"train_time":384.95,"predict_time":0.163,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+168\/-91","imputed":false},{"position":11,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1661.0,"score":0.428,"rank":19.2,"hrank":12.88,"imp":10.555,"train_time":89.02,"predict_time":0.317,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-127","imputed":false},{"position":12,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1657.0,"score":0.414,"rank":19.4,"hrank":11.45,"imp":9.959,"train_time":384.95,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+178\/-96","imputed":false},{"position":13,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1655.0,"score":0.367,"rank":19.5,"hrank":13.55,"imp":10.08,"train_time":780.95,"predict_time":1.493,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+212\/-105","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1646.0,"score":0.367,"rank":20.0,"hrank":15.22,"imp":11.236,"train_time":91.02,"predict_time":0.568,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+172\/-82","imputed":false},{"position":15,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1643.0,"score":0.404,"rank":20.2,"hrank":14.99,"imp":10.686,"train_time":89.02,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-120","imputed":false},{"position":16,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1638.0,"score":0.351,"rank":20.5,"hrank":12.92,"imp":10.904,"train_time":875.01,"predict_time":0.549,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+200\/-104","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1634.0,"score":0.404,"rank":20.7,"hrank":13.19,"imp":10.118,"train_time":1635.74,"predict_time":2.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-86","imputed":false},{"position":18,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1631.0,"score":0.42,"rank":20.9,"hrank":10.4,"imp":9.393,"train_time":1.53,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-67","imputed":false},{"position":19,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1608.0,"score":0.333,"rank":22.25,"hrank":14.51,"imp":12.289,"train_time":2335.83,"predict_time":0.502,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-106","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1604.0,"score":0.342,"rank":22.5,"hrank":16.78,"imp":10.671,"train_time":202.54,"predict_time":0.055,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+164\/-101","imputed":false},{"position":21,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1580.0,"score":0.293,"rank":24.0,"hrank":19.99,"imp":12.064,"train_time":91.02,"predict_time":0.074,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-99","imputed":false},{"position":22,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1554.0,"score":0.311,"rank":25.6,"hrank":13.29,"imp":12.341,"train_time":1635.74,"predict_time":0.082,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+165\/-115","imputed":false},{"position":23,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1552.0,"score":0.298,"rank":25.7,"hrank":12.89,"imp":11.869,"train_time":875.01,"predict_time":0.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+187\/-138","imputed":false},{"position":24,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1543.0,"score":0.302,"rank":26.3,"hrank":17.59,"imp":11.715,"train_time":0.41,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-119","imputed":false},{"position":25,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1534.0,"score":0.26,"rank":26.9,"hrank":16.77,"imp":11.128,"train_time":2.3,"predict_time":0.818,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+167\/-109","imputed":false},{"position":26,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1517.0,"score":0.254,"rank":28.0,"hrank":12.17,"imp":12.659,"train_time":8.46,"predict_time":3.879,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+178\/-121","imputed":false},{"position":27,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1483.0,"score":0.209,"rank":30.2,"hrank":18.06,"imp":13.283,"train_time":1598.12,"predict_time":6.083,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-105","imputed":false},{"position":28,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1473.0,"score":0.27,"rank":30.9,"hrank":7.66,"imp":12.804,"train_time":0.47,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+233\/-150","imputed":false},{"position":29,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1465.0,"score":0.203,"rank":31.4,"hrank":24.1,"imp":15.025,"train_time":421.22,"predict_time":0.1,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+194\/-123","imputed":false},{"position":30,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1456.0,"score":0.211,"rank":32.0,"hrank":16.85,"imp":13.666,"train_time":0.63,"predict_time":0.171,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+235\/-178","imputed":false},{"position":31,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1453.0,"score":0.155,"rank":32.2,"hrank":25.62,"imp":13.439,"train_time":1128.74,"predict_time":1.573,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-84","imputed":false},{"position":32,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1452.0,"score":0.193,"rank":32.3,"hrank":21.75,"imp":14.305,"train_time":0.79,"predict_time":0.232,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+192\/-163","imputed":false},{"position":33,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1420.0,"score":0.14,"rank":34.4,"hrank":28.93,"imp":13.694,"train_time":4546.14,"predict_time":406.102,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-125","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1415.0,"score":0.139,"rank":34.8,"hrank":30.51,"imp":15.061,"train_time":1.61,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-111","imputed":false},{"position":35,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1413.0,"score":0.208,"rank":34.9,"hrank":24.92,"imp":13.861,"train_time":713.01,"predict_time":2.317,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+148\/-164","imputed":false},{"position":36,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1401.0,"score":0.148,"rank":35.7,"hrank":27.59,"imp":16.621,"train_time":713.01,"predict_time":0.155,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-162","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1365.0,"score":0.082,"rank":38.1,"hrank":34.95,"imp":16.833,"train_time":6.53,"predict_time":0.294,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-114","imputed":false},{"position":38,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1354.0,"score":0.102,"rank":38.85,"hrank":24.66,"imp":17.196,"train_time":4546.14,"predict_time":39.323,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+166\/-155","imputed":false},{"position":39,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1350.0,"score":0.113,"rank":39.1,"hrank":31.74,"imp":16.571,"train_time":421.22,"predict_time":0.007,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+181\/-127","imputed":false},{"position":40,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1338.0,"score":0.061,"rank":39.9,"hrank":32.84,"imp":16.429,"train_time":52.7,"predict_time":0.202,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-107","imputed":false},{"position":41,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1334.0,"score":0.09,"rank":40.2,"hrank":36.59,"imp":17.843,"train_time":1.4,"predict_time":0.007,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-125","imputed":false},{"position":42,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1327.0,"score":0.087,"rank":40.6,"hrank":33.62,"imp":17.764,"train_time":387.73,"predict_time":2.446,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-176","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1307.0,"score":0.072,"rank":41.9,"hrank":38.29,"imp":16.432,"train_time":1128.74,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+111\/-98","imputed":false},{"position":44,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1306.0,"score":0.101,"rank":42.0,"hrank":31.07,"imp":17.663,"train_time":82.52,"predict_time":0.97,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+176\/-165","imputed":false},{"position":45,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1306.0,"score":0.119,"rank":42.0,"hrank":36.09,"imp":19.39,"train_time":68.71,"predict_time":7.89,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+184\/-156","imputed":false},{"position":46,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1303.0,"score":0.212,"rank":42.2,"hrank":7.72,"imp":18.833,"train_time":5.05,"predict_time":2.948,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+252\/-285","imputed":false},{"position":47,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1293.0,"score":0.045,"rank":42.8,"hrank":37.75,"imp":16.851,"train_time":52.7,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-100","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1289.0,"score":0.041,"rank":43.1,"hrank":40.53,"imp":16.551,"train_time":49.21,"predict_time":0.149,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-139","imputed":false},{"position":49,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1258.0,"score":0.054,"rank":45.0,"hrank":42.34,"imp":16.349,"train_time":0.33,"predict_time":0.05,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-132","imputed":false},{"position":50,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.03,"rank":45.9,"hrank":44.09,"imp":17.041,"train_time":11.03,"predict_time":0.489,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-109","imputed":false},{"position":51,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1232.0,"score":0.024,"rank":46.6,"hrank":42.71,"imp":19.606,"train_time":14.93,"predict_time":13.661,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-115","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1227.0,"score":0.042,"rank":46.9,"hrank":42.04,"imp":18.856,"train_time":49.21,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+145\/-194","imputed":false},{"position":53,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1193.0,"score":0.021,"rank":48.95,"hrank":44.89,"imp":21.028,"train_time":41.63,"predict_time":39.259,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-163","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1155.0,"score":0.013,"rank":51.1,"hrank":47.91,"imp":21.17,"train_time":387.73,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-170","imputed":false},{"position":55,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1137.0,"score":0.011,"rank":52.1,"hrank":50.25,"imp":19.824,"train_time":1.84,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-123","imputed":false},{"position":56,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1060.0,"score":0.0,"rank":56.1,"hrank":54.95,"imp":26.384,"train_time":3486.11,"predict_time":2.194,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-103","imputed":false},{"position":57,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1058.0,"score":0.017,"rank":56.2,"hrank":51.77,"imp":24.692,"train_time":1.63,"predict_time":0.11,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+153\/-235","imputed":false},{"position":58,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1046.0,"score":0.015,"rank":56.75,"hrank":51.56,"imp":29.46,"train_time":43.41,"predict_time":0.111,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+185\/-264","imputed":false},{"position":59,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1043.0,"score":0.016,"rank":56.9,"hrank":53.51,"imp":26.105,"train_time":332.4,"predict_time":33.481,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+121\/-226","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1039.0,"score":0.0,"rank":57.1,"hrank":55.99,"imp":26.456,"train_time":3486.11,"predict_time":8.148,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-109","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1008.0,"score":0.002,"rank":58.55,"hrank":54.61,"imp":30.389,"train_time":43.41,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+165\/-280","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":58.9,"hrank":57.01,"imp":24.603,"train_time":0.11,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-242","imputed":false},{"position":63,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":956.0,"score":0.0,"rank":60.8,"hrank":59.64,"imp":25.782,"train_time":17.26,"predict_time":6.893,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-210","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":934.0,"score":0.0,"rank":61.7,"hrank":59.66,"imp":29.251,"train_time":22.23,"predict_time":0.567,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+145\/-196","imputed":false},{"position":65,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":893.0,"score":0.0,"rank":63.3,"hrank":62.85,"imp":27.911,"train_time":0.32,"predict_time":0.056,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-236","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":890.0,"score":0.0,"rank":63.4,"hrank":62.78,"imp":32.855,"train_time":85.23,"predict_time":2.947,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-120","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":865.0,"score":0.0,"rank":64.3,"hrank":63.63,"imp":31.979,"train_time":0.21,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-243","imputed":false},{"position":68,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":865.0,"score":0.0,"rank":64.3,"hrank":62.82,"imp":30.588,"train_time":22.23,"predict_time":0.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-251","imputed":false},{"position":69,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":859.0,"score":0.022,"rank":64.5,"hrank":56.16,"imp":35.072,"train_time":2.96,"predict_time":0.183,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+236\/-497","imputed":false},{"position":70,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":823.0,"score":0.0,"rank":65.7,"hrank":65.42,"imp":34.185,"train_time":332.4,"predict_time":1.631,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-172","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":790.0,"score":0.0,"rank":66.7,"hrank":66.0,"imp":36.745,"train_time":0.18,"predict_time":0.032,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-192","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":737.0,"score":0.0,"rank":68.1,"hrank":67.2,"imp":39.058,"train_time":1.67,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+163\/-360","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":478.0,"score":0.0,"rank":72.5,"hrank":72.47,"imp":50.755,"train_time":0.12,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-349","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/pareto_front_explorer.html
index ed7844a6df3c1f6bb3b526a56934fbcbd5548b4e..d3d31d3d98dcbfccf095316a0402c0a754e1bcd7 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.970420493,"imp":17.6629337184,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0198727775,"imp":9.3929627125,"elo":1630.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0167342488,"imp":9.9587335282,"elo":1657.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1627138271,"imp":9.7410518279,"elo":1661.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.02389485,"imp":11.715466652,"elo":1543.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0257224855,"imp":10.6863195805,"elo":1643.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3166644071,"imp":10.5546263337,"elo":1660.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0068099474,"imp":17.8429705318,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.006754717,"imp":16.5713605893,"elo":1350.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0998897869,"imp":15.0254091034,"elo":1465.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0559129084,"imp":27.9111955894,"elo":892.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0143949688,"imp":18.8561350573,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1486815713,"imp":16.5513343714,"elo":1288.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.109744233,"imp":24.6915792924,"elo":1057.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0890979144,"imp":21.1700554595,"elo":1155.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4458463373,"imp":17.7641123217,"elo":1327.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0504837885,"imp":16.3491381425,"elo":1258.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0742212397,"imp":12.0635574748,"elo":1579.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.5681105528,"imp":11.2357370861,"elo":1646.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.948262606,"imp":18.8331940342,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4891016512,"imp":17.0408331207,"elo":1243.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5024822125,"imp":12.2885548487,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":14.1450303793,"imp":8.2196331689,"elo":1687.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8931236803,"imp":25.7815891216,"elo":956.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0175726795,"imp":39.058107621,"elo":737.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2360405674,"imp":30.5884063132,"elo":865.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5669505531,"imp":29.250530009,"elo":934.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0120761251,"imp":24.6033316585,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0173683474,"imp":16.8508789793,"elo":1293.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2020648674,"imp":16.4285035282,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2942921663,"imp":16.8326922638,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.08214277,"imp":12.3410565047,"elo":1554.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.1463462325,"imp":10.1178278972,"elo":1634.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8184752083,"imp":11.1276358993,"elo":1533.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4934856828,"imp":10.0797482352,"elo":1655.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6399970659,"imp":8.1416301567,"elo":1716.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.6609678586,"imp":19.6059845316,"elo":1232.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2594035993,"imp":21.0282927544,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.3227901343,"imp":17.1961294701,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":406.1020417363,"imp":13.6941328483,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2319547305,"imp":14.3049973813,"elo":1451.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":18.7619482789,"imp":6.4485292983,"elo":1928.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0317264099,"imp":36.7450378085,"elo":789.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.8791795848,"imp":12.6594612614,"elo":1516.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.171595863,"imp":9.5061806177,"elo":1772.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0553381931,"imp":15.0605116327,"elo":1414.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0570275282,"imp":11.8692752948,"elo":1552.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5491669944,"imp":10.9044317748,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6991145285,"imp":8.5386327366,"elo":1826.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2909668033,"imp":5.9995997481,"elo":1847.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9466920315,"imp":32.8545637836,"elo":890.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1940808855,"imp":26.3835533588,"elo":1059.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1480810234,"imp":26.4556542463,"elo":1038.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1710376878,"imp":13.6658034437,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0468974441,"imp":19.8235103881,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0536547712,"imp":16.4323136139,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5732102138,"imp":13.4385258165,"elo":1453.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0471622003,"imp":12.8041808129,"elo":1472.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0546297999,"imp":10.6713217964,"elo":1604.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4302741489,"imp":10.3423013795,"elo":1680.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.8904270487,"imp":19.3895045825,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":6.0826693723,"imp":13.2834660869,"elo":1483.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":70.1635578676,"imp":10.8780238272,"elo":1675.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1830265708,"imp":35.0723495798,"elo":859.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1546216562,"imp":16.6207806431,"elo":1401.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.3169671728,"imp":13.8612441205,"elo":1413.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.970420493,"x_train":82.518930664,"imp":17.6629337184,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0198727775,"x_train":1.5291414753,"imp":9.3929627125,"elo":1630.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0167342488,"x_train":384.9547373775,"imp":9.9587335282,"elo":1657.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1627138271,"x_train":384.9547373775,"imp":9.7410518279,"elo":1661.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.02389485,"x_train":0.4117206383,"imp":11.715466652,"elo":1543.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0257224855,"x_train":89.0178815303,"imp":10.6863195805,"elo":1643.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3166644071,"x_train":89.0178815303,"imp":10.5546263337,"elo":1660.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0068099474,"x_train":1.4013272116,"imp":17.8429705318,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.006754717,"x_train":421.2236428358,"imp":16.5713605893,"elo":1350.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0998897869,"x_train":421.2236428358,"imp":15.0254091034,"elo":1465.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0559129084,"x_train":0.3214016003,"imp":27.9111955894,"elo":892.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0143949688,"x_train":49.2121405424,"imp":18.8561350573,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1486815713,"x_train":49.2121405424,"imp":16.5513343714,"elo":1288.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.109744233,"x_train":1.6301605991,"imp":24.6915792924,"elo":1057.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0890979144,"x_train":387.7306724242,"imp":21.1700554595,"elo":1155.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4458463373,"x_train":387.7306724242,"imp":17.7641123217,"elo":1327.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0504837885,"x_train":0.3317135,"imp":16.3491381425,"elo":1258.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0742212397,"x_train":91.0184357178,"imp":12.0635574748,"elo":1579.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.5681105528,"x_train":91.0184357178,"imp":11.2357370861,"elo":1646.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.948262606,"x_train":5.0473841082,"imp":18.8331940342,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4891016512,"x_train":11.026497565,"imp":17.0408331207,"elo":1243.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5024822125,"x_train":2335.8257219827,"imp":12.2885548487,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":14.1450303793,"x_train":2335.8257219827,"imp":8.2196331689,"elo":1687.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8931236803,"x_train":17.2564525143,"imp":25.7815891216,"elo":956.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0175726795,"x_train":1.6656394605,"imp":39.058107621,"elo":737.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2360405674,"x_train":22.233335416,"imp":30.5884063132,"elo":865.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5669505531,"x_train":22.233335416,"imp":29.250530009,"elo":934.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0120761251,"x_train":0.1136960958,"imp":24.6033316585,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0173683474,"x_train":52.7035618074,"imp":16.8508789793,"elo":1293.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2020648674,"x_train":52.7035618074,"imp":16.4285035282,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2942921663,"x_train":6.5297785644,"imp":16.8326922638,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.08214277,"x_train":1635.7415319042,"imp":12.3410565047,"elo":1554.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.1463462325,"x_train":1635.7415319042,"imp":10.1178278972,"elo":1634.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8184752083,"x_train":2.2966540372,"imp":11.1276358993,"elo":1533.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4934856828,"x_train":780.948577715,"imp":10.0797482352,"elo":1655.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6399970659,"x_train":780.948577715,"imp":8.1416301567,"elo":1716.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.6609678586,"x_train":14.9320242657,"imp":19.6059845316,"elo":1232.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2594035993,"x_train":41.6325123071,"imp":21.0282927544,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.3227901343,"x_train":4546.1421807431,"imp":17.1961294701,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":406.1020417363,"x_train":4546.1421807431,"imp":13.6941328483,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2319547305,"x_train":0.7863945199,"imp":14.3049973813,"elo":1451.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":18.7619482789,"x_train":38.9361051757,"imp":6.4485292983,"elo":1928.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0317264099,"x_train":0.1750892902,"imp":36.7450378085,"elo":789.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.8791795848,"x_train":8.464564148,"imp":12.6594612614,"elo":1516.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.171595863,"x_train":0.6194744057,"imp":9.5061806177,"elo":1772.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0553381931,"x_train":1.6076337396,"imp":15.0605116327,"elo":1414.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0570275282,"x_train":875.0118329531,"imp":11.8692752948,"elo":1552.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5491669944,"x_train":875.0118329531,"imp":10.9044317748,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6991145285,"x_train":2.330497005,"imp":8.5386327366,"elo":1826.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2909668033,"x_train":1.1055094044,"imp":5.9995997481,"elo":1847.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9466920315,"x_train":85.2279597522,"imp":32.8545637836,"elo":890.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1940808855,"x_train":3486.1100818653,"imp":26.3835533588,"elo":1059.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1480810234,"x_train":3486.1100818653,"imp":26.4556542463,"elo":1038.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1710376878,"x_train":0.6314611374,"imp":13.6658034437,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0468974441,"x_train":1.8436799882,"imp":19.8235103881,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0536547712,"x_train":1128.7390334238,"imp":16.4323136139,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5732102138,"x_train":1128.7390334238,"imp":13.4385258165,"elo":1453.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0471622003,"x_train":0.4669717504,"imp":12.8041808129,"elo":1472.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0546297999,"x_train":202.5364616477,"imp":10.6713217964,"elo":1604.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4302741489,"x_train":202.5364616477,"imp":10.3423013795,"elo":1680.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.8904270487,"x_train":68.7091297861,"imp":19.3895045825,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":6.0826693723,"x_train":1598.1220946317,"imp":13.2834660869,"elo":1483.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":70.1635578676,"x_train":1598.1220946317,"imp":10.8780238272,"elo":1675.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1830265708,"x_train":2.9551878511,"imp":35.0723495798,"elo":859.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1546216562,"x_train":713.0056913426,"imp":16.6207806431,"elo":1401.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.3169671728,"x_train":713.0056913426,"imp":13.8612441205,"elo":1413.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
index ee19d2476ef0630f61d7ab72a2c3836f9aacbacd..389ee0b7ef4056d12e0b068c552a30ae88911e1b 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":82.518930664,"imp":17.7783471369,"elo":1348.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.5291414753,"imp":9.491241243,"elo":1684.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.9721250392,"imp":9.6999483994,"elo":1682.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.7214520913,"imp":9.6728929545,"elo":1703.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":24.1437443132,"imp":9.6761930535,"elo":1711.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":53.2602822109,"imp":9.7647789612,"elo":1728.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":99.398536178,"imp":9.7773216807,"elo":1729.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":193.3194459181,"imp":9.8341522645,"elo":1722.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":286.5879298672,"imp":9.8717448659,"elo":1716.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":384.9547373775,"imp":9.824439582,"elo":1713.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4117206383,"imp":11.8287672651,"elo":1586.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.8690160818,"imp":11.3284736317,"elo":1655.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.0802815735,"imp":11.0507653547,"elo":1678.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.1754502095,"imp":10.7601558928,"elo":1693.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.940122673,"imp":10.6749383989,"elo":1701.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.2455560723,"imp":10.5350564433,"elo":1708.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":44.3873766321,"imp":10.5110964613,"elo":1716.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":66.3664018643,"imp":10.5653157475,"elo":1719.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":89.0178815303,"imp":10.6549946425,"elo":1725.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":1.4013272116,"imp":17.9345318934,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":4.1193754289,"imp":16.7471474255,"elo":1421.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":10.042965061,"imp":16.1012884683,"elo":1453.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":20.7240396679,"imp":15.8481919199,"elo":1469.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":52.2528950395,"imp":15.5347747171,"elo":1489.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":105.4596015797,"imp":15.4125820364,"elo":1498.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":207.809367173,"imp":15.232787978,"elo":1505.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":315.0112350614,"imp":15.1784841885,"elo":1502.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":421.2236428358,"imp":15.1141849883,"elo":1503.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3214016003,"imp":27.98552044,"elo":891.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.5493187284,"imp":24.9759763005,"elo":972.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.3228210017,"imp":19.9577975472,"elo":1131.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.5095259777,"imp":17.8375606124,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.1644908467,"imp":16.9416737546,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.2587333157,"imp":16.4669656718,"elo":1288.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.4863048451,"imp":15.9705548137,"elo":1321.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":36.7034709861,"imp":15.9380732553,"elo":1323.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":49.2121405424,"imp":16.6306303792,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.6301605991,"imp":24.7595361474,"elo":1082.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.3926690631,"imp":23.0477806361,"elo":1164.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":9.3189904508,"imp":20.7635168043,"elo":1260.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":19.4603799719,"imp":19.7613730569,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":46.679624122,"imp":18.8448843712,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":94.3885780511,"imp":18.5005763241,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":191.9205953467,"imp":18.0209777098,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":288.7632596624,"imp":17.8684816306,"elo":1365.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":387.7306724242,"imp":17.8487232653,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.3317135,"imp":16.450314925,"elo":1301.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.8849406618,"imp":15.4488333983,"elo":1391.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":2.143134634,"imp":14.0917188516,"elo":1509.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":4.3187074851,"imp":13.0693815002,"elo":1593.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":11.2464284633,"imp":11.9726028643,"elo":1655.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":22.4893847368,"imp":11.4942207921,"elo":1690.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":44.9141349299,"imp":11.3534039222,"elo":1701.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":67.4875589082,"imp":11.3257978613,"elo":1706.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":91.0184357178,"imp":11.344681261,"elo":1702.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":5.0473841082,"imp":18.9276909301,"elo":1351.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":11.026497565,"imp":17.1294418406,"elo":1269.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.1149019147,"imp":15.1706285991,"elo":1394.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":63.584802766,"imp":12.8525382121,"elo":1551.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":122.2265200742,"imp":12.3287993704,"elo":1584.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":281.6186890352,"imp":11.782681671,"elo":1632.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":581.6935693293,"imp":10.889492571,"elo":1667.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1150.8300045772,"imp":10.065987363,"elo":1704.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1716.2616608148,"imp":9.0450371026,"elo":1733.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2335.8257219827,"imp":8.352818795,"elo":1754.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":17.2564525143,"imp":25.2607495901,"elo":992.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.6656394605,"imp":39.0642635035,"elo":761.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.4059982085,"imp":35.591731487,"elo":850.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":18.5028948612,"imp":29.7259121704,"elo":939.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":22.233335416,"imp":29.278456989,"elo":954.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1136960958,"imp":24.6930241703,"elo":999.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3871933784,"imp":19.5833547824,"elo":1167.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.2235890503,"imp":17.7609340623,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":2.5806805069,"imp":17.1613052185,"elo":1295.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":6.4224304798,"imp":16.7142897375,"elo":1351.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":12.9124255888,"imp":16.5753022845,"elo":1357.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":25.9873657979,"imp":16.4674560346,"elo":1366.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":39.2207943229,"imp":16.4652364548,"elo":1365.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":52.7035618074,"imp":16.5087546984,"elo":1371.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.5297785644,"imp":16.9053393771,"elo":1386.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.3163332372,"imp":14.6210264384,"elo":1506.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.3133997603,"imp":13.1111128566,"elo":1584.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":79.4447895462,"imp":12.0724836057,"elo":1617.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":204.9296802966,"imp":11.2915866822,"elo":1649.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":412.61016262,"imp":10.881660853,"elo":1670.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":813.6980979059,"imp":10.4752223495,"elo":1676.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1221.0196908737,"imp":10.3145721226,"elo":1690.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1635.7415319042,"imp":10.2282592278,"elo":1698.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2.2966540372,"imp":11.2492008343,"elo":1562.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.7344117695,"imp":11.1445835898,"elo":1586.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":18.4049918241,"imp":10.9046971829,"elo":1609.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":36.2506481168,"imp":10.7569347224,"elo":1628.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":88.2736709813,"imp":10.3198889772,"elo":1680.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":182.8828378906,"imp":9.8152496316,"elo":1705.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":376.7342623721,"imp":9.1635361732,"elo":1734.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":570.2939458958,"imp":8.7762828203,"elo":1747.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":780.948577715,"imp":8.2533253958,"elo":1762.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":14.9320242657,"imp":19.6858576472,"elo":1259.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.6325123071,"imp":21.0976772389,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.3901218357,"imp":19.2028304181,"elo":1276.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":119.1912414891,"imp":17.6639004573,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":214.6945851817,"imp":17.0159710418,"elo":1374.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":545.8855951454,"imp":15.668596096,"elo":1419.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1118.7729763883,"imp":14.8368783417,"elo":1445.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2249.1619866125,"imp":14.3706389903,"elo":1465.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3393.4090688327,"imp":13.9629897147,"elo":1462.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4546.1421807431,"imp":13.7527405267,"elo":1466.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.7863945199,"imp":14.3899781323,"elo":1495.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.9361051757,"imp":6.557487887,"elo":1992.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1750892902,"imp":36.7945450937,"elo":804.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":8.464564148,"imp":12.7682217685,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6194744057,"imp":9.6238578672,"elo":1823.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.6076337396,"imp":15.1564198567,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":5.2365016338,"imp":14.4005222476,"elo":1491.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":19.5371570641,"imp":13.3518807492,"elo":1554.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":41.9262650686,"imp":12.4994851212,"elo":1602.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":107.9205498469,"imp":11.8438984455,"elo":1635.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":217.2325339637,"imp":11.5183191167,"elo":1651.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":430.1308094699,"imp":11.1900788346,"elo":1667.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":649.8093981365,"imp":11.0914491011,"elo":1682.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":875.0118329531,"imp":11.0230785467,"elo":1696.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.330497005,"imp":8.6491122691,"elo":1890.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.1055094044,"imp":5.9137624298,"elo":1876.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":85.2279597522,"imp":32.899788994,"elo":896.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":99.1248084332,"imp":32.7796115249,"elo":910.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":139.1755162215,"imp":32.4329755298,"elo":919.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":198.0940117736,"imp":31.6399865293,"elo":946.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":444.3646878451,"imp":30.0949731269,"elo":971.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":895.5564058915,"imp":28.5399652438,"elo":1012.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1773.8913331235,"imp":27.8873277974,"elo":1031.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2617.7743320608,"imp":27.1154322071,"elo":1057.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3486.1100818653,"imp":26.5078469087,"elo":1087.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.6314611374,"imp":13.7835229791,"elo":1497.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":1.8436799882,"imp":19.8900519711,"elo":1165.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.4007633821,"imp":18.1795243011,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":27.9733904091,"imp":16.5495848047,"elo":1321.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":59.9007241046,"imp":15.6091450562,"elo":1374.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":143.3365257329,"imp":14.6701208982,"elo":1413.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":278.2745551506,"imp":14.183180672,"elo":1446.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":543.6674489023,"imp":13.8593137304,"elo":1474.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":841.4117357913,"imp":13.6597843306,"elo":1491.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1128.7390334238,"imp":13.5312380654,"elo":1498.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4669717504,"imp":12.9118537073,"elo":1520.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.0321466005,"imp":11.8212367436,"elo":1617.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":5.5778533182,"imp":11.3229193666,"elo":1667.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":9.7317062539,"imp":10.9403239771,"elo":1700.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":25.7016530469,"imp":10.628774182,"elo":1718.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":51.0311253263,"imp":10.5022373116,"elo":1732.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":101.303854312,"imp":10.4647468749,"elo":1740.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":150.6411976385,"imp":10.4529428699,"elo":1742.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":202.5364616477,"imp":10.4820421716,"elo":1743.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":68.7091297861,"imp":19.4673543521,"elo":1353.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":129.9001536616,"imp":17.1822083639,"elo":1413.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":315.8079093334,"imp":14.4126589357,"elo":1512.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":621.987734561,"imp":12.5570854498,"elo":1619.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1537.7988263416,"imp":10.8050873617,"elo":1736.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1598.1220946317,"imp":10.9608112326,"elo":1733.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.9551878511,"imp":35.1302552748,"elo":878.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":6.5948259183,"imp":27.689365732,"elo":1018.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.3219967074,"imp":19.5874877747,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":35.2109614365,"imp":16.4609945688,"elo":1382.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.816423174,"imp":15.4820902855,"elo":1428.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":176.8663518766,"imp":14.9281733183,"elo":1441.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":354.8398796964,"imp":14.4325741349,"elo":1449.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":531.8168061758,"imp":14.0604231438,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":713.0056913426,"imp":13.6083145675,"elo":1461.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.970420493,"x_train":82.518930664,"imp":17.7783471369,"elo":1348.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0198727775,"x_train":1.5291414753,"imp":9.491241243,"elo":1684.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0381500095,"x_train":3.9721250392,"imp":9.6999483994,"elo":1682.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0610258172,"x_train":12.7214520913,"imp":9.6728929545,"elo":1703.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0803301633,"x_train":24.1437443132,"imp":9.6761930535,"elo":1711.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0971069508,"x_train":53.2602822109,"imp":9.7647789612,"elo":1728.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1155626332,"x_train":99.398536178,"imp":9.7773216807,"elo":1729.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1393798126,"x_train":193.3194459181,"imp":9.8341522645,"elo":1722.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1506167429,"x_train":286.5879298672,"imp":9.8717448659,"elo":1716.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1627138271,"x_train":384.9547373775,"imp":9.824439582,"elo":1713.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.02389485,"x_train":0.4117206383,"imp":11.8287672651,"elo":1586.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0478927199,"x_train":0.8690160818,"imp":11.3284736317,"elo":1655.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1045674981,"x_train":2.0802815735,"imp":11.0507653547,"elo":1678.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1647539973,"x_train":4.1754502095,"imp":10.7601558928,"elo":1693.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2414369657,"x_train":10.940122673,"imp":10.6749383989,"elo":1701.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2964729106,"x_train":22.2455560723,"imp":10.5350564433,"elo":1708.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3247321968,"x_train":44.3873766321,"imp":10.5110964613,"elo":1716.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3090326441,"x_train":66.3664018643,"imp":10.5653157475,"elo":1719.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3166644071,"x_train":89.0178815303,"imp":10.6549946425,"elo":1725.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0068099474,"x_train":1.4013272116,"imp":17.9345318934,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0177962725,"x_train":4.1193754289,"imp":16.7471474255,"elo":1421.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0331174137,"x_train":10.042965061,"imp":16.1012884683,"elo":1453.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0439779117,"x_train":20.7240396679,"imp":15.8481919199,"elo":1469.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0691665167,"x_train":52.2528950395,"imp":15.5347747171,"elo":1489.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0836217017,"x_train":105.4596015797,"imp":15.4125820364,"elo":1498.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0948483349,"x_train":207.809367173,"imp":15.232787978,"elo":1505.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1018479444,"x_train":315.0112350614,"imp":15.1784841885,"elo":1502.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0996158221,"x_train":421.2236428358,"imp":15.1141849883,"elo":1503.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0559129084,"x_train":0.3214016003,"imp":27.98552044,"elo":891.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.064531699,"x_train":0.5493187284,"imp":24.9759763005,"elo":972.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0703221149,"x_train":1.3228210017,"imp":19.9577975472,"elo":1131.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.073653813,"x_train":2.5095259777,"imp":17.8375606124,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.086629059,"x_train":6.1644908467,"imp":16.9416737546,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0999308324,"x_train":12.2587333157,"imp":16.4669656718,"elo":1288.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1177984553,"x_train":24.4863048451,"imp":15.9705548137,"elo":1321.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1410295133,"x_train":36.7034709861,"imp":15.9380732553,"elo":1323.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1422071534,"x_train":49.2121405424,"imp":16.6306303792,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.109744233,"x_train":1.6301605991,"imp":24.7595361474,"elo":1082.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2272880738,"x_train":3.3926690631,"imp":23.0477806361,"elo":1164.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5474602559,"x_train":9.3189904508,"imp":20.7635168043,"elo":1260.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8604236954,"x_train":19.4603799719,"imp":19.7613730569,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1405220728,"x_train":46.679624122,"imp":18.8448843712,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3925732088,"x_train":94.3885780511,"imp":18.5005763241,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7973939159,"x_train":191.9205953467,"imp":18.0209777098,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.9804848892,"x_train":288.7632596624,"imp":17.8684816306,"elo":1365.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.3433373669,"x_train":387.7306724242,"imp":17.8487232653,"elo":1362.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0504837885,"x_train":0.3317135,"imp":16.450314925,"elo":1301.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.330362467,"x_train":0.8849406618,"imp":15.4488333983,"elo":1391.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5377965122,"x_train":2.143134634,"imp":14.0917188516,"elo":1509.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7884842205,"x_train":4.3187074851,"imp":13.0693815002,"elo":1593.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7687781828,"x_train":11.2464284633,"imp":11.9726028643,"elo":1655.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.755201967,"x_train":22.4893847368,"imp":11.4942207921,"elo":1690.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7276316945,"x_train":44.9141349299,"imp":11.3534039222,"elo":1701.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6450965711,"x_train":67.4875589082,"imp":11.3257978613,"elo":1706.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6136389605,"x_train":91.0184357178,"imp":11.344681261,"elo":1702.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":2.948262606,"x_train":5.0473841082,"imp":18.9276909301,"elo":1351.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4891016512,"x_train":11.026497565,"imp":17.1294418406,"elo":1269.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.0456997204,"x_train":25.1149019147,"imp":15.1706285991,"elo":1394.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.5533251388,"x_train":63.584802766,"imp":12.8525382121,"elo":1551.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.2314257133,"x_train":122.2265200742,"imp":12.3287993704,"elo":1584.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.994864421,"x_train":281.6186890352,"imp":11.782681671,"elo":1632.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.3596868906,"x_train":581.6935693293,"imp":10.889492571,"elo":1667.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":12.3832071121,"x_train":1150.8300045772,"imp":10.065987363,"elo":1704.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":13.3833030376,"x_train":1716.2616608148,"imp":9.0450371026,"elo":1733.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":14.1450303793,"x_train":2335.8257219827,"imp":8.352818795,"elo":1754.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8931236803,"x_train":17.2564525143,"imp":25.2607495901,"elo":992.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0175726795,"x_train":1.6656394605,"imp":39.0642635035,"elo":761.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1527010552,"x_train":6.4059982085,"imp":35.591731487,"elo":850.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4812091674,"x_train":18.5028948612,"imp":29.7259121704,"elo":939.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5669505531,"x_train":22.233335416,"imp":29.278456989,"elo":954.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0120761251,"x_train":0.1136960958,"imp":24.6930241703,"elo":999.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0375156837,"x_train":0.3871933784,"imp":19.5833547824,"elo":1167.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0764464768,"x_train":1.2235890503,"imp":17.7609340623,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1157837036,"x_train":2.5806805069,"imp":17.1613052185,"elo":1295.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1326680035,"x_train":6.4224304798,"imp":16.7142897375,"elo":1351.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1710393828,"x_train":12.9124255888,"imp":16.5753022845,"elo":1357.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1991880198,"x_train":25.9873657979,"imp":16.4674560346,"elo":1366.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2137395788,"x_train":39.2207943229,"imp":16.4652364548,"elo":1365.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.193317742,"x_train":52.7035618074,"imp":16.5087546984,"elo":1371.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2942921663,"x_train":6.5297785644,"imp":16.9053393771,"elo":1386.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4603279722,"x_train":15.3163332372,"imp":14.6210264384,"elo":1506.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.7544553224,"x_train":39.3133997603,"imp":13.1111128566,"elo":1584.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.0304844995,"x_train":79.4447895462,"imp":12.0724836057,"elo":1617.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.3434994293,"x_train":204.9296802966,"imp":11.2915866822,"elo":1649.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7588377133,"x_train":412.61016262,"imp":10.881660853,"elo":1670.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.0398213365,"x_train":813.6980979059,"imp":10.4752223495,"elo":1676.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.1897696555,"x_train":1221.0196908737,"imp":10.3145721226,"elo":1690.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.1463462325,"x_train":1635.7415319042,"imp":10.2282592278,"elo":1698.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8184752083,"x_train":2.2966540372,"imp":11.2492008343,"elo":1562.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.6939241991,"x_train":5.7344117695,"imp":11.1445835898,"elo":1586.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2294004302,"x_train":18.4049918241,"imp":10.9046971829,"elo":1609.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1368098652,"x_train":36.2506481168,"imp":10.7569347224,"elo":1628.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.826705741,"x_train":88.2736709813,"imp":10.3198889772,"elo":1680.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.0207330225,"x_train":182.8828378906,"imp":9.8152496316,"elo":1705.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.2704454231,"x_train":376.7342623721,"imp":9.1635361732,"elo":1734.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.3919513821,"x_train":570.2939458958,"imp":8.7762828203,"elo":1747.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6399970659,"x_train":780.948577715,"imp":8.2533253958,"elo":1762.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.6609678586,"x_train":14.9320242657,"imp":19.6858576472,"elo":1259.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2594035993,"x_train":41.6325123071,"imp":21.0976772389,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":56.6400616397,"x_train":63.3901218357,"imp":19.2028304181,"elo":1276.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":88.5407748208,"x_train":119.1912414891,"imp":17.6639004573,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":118.0359236254,"x_train":214.6945851817,"imp":17.0159710418,"elo":1374.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":189.7782849186,"x_train":545.8855951454,"imp":15.668596096,"elo":1419.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":271.3697151109,"x_train":1118.7729763883,"imp":14.8368783417,"elo":1445.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":353.6812333996,"x_train":2249.1619866125,"imp":14.3706389903,"elo":1465.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":380.7963519935,"x_train":3393.4090688327,"imp":13.9629897147,"elo":1462.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":414.7537971415,"x_train":4546.1421807431,"imp":13.7527405267,"elo":1466.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2319547305,"x_train":0.7863945199,"imp":14.3899781323,"elo":1495.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":18.7619482789,"x_train":38.9361051757,"imp":6.557487887,"elo":1992.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0317264099,"x_train":0.1750892902,"imp":36.7945450937,"elo":804.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":3.8791795848,"x_train":8.464564148,"imp":12.7682217685,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.171595863,"x_train":0.6194744057,"imp":9.6238578672,"elo":1823.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0553381931,"x_train":1.6076337396,"imp":15.1564198567,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1172004971,"x_train":5.2365016338,"imp":14.4005222476,"elo":1491.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2044016087,"x_train":19.5371570641,"imp":13.3518807492,"elo":1554.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2619165464,"x_train":41.9262650686,"imp":12.4994851212,"elo":1602.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3522960448,"x_train":107.9205498469,"imp":11.8438984455,"elo":1635.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4098866808,"x_train":217.2325339637,"imp":11.5183191167,"elo":1651.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4587122968,"x_train":430.1308094699,"imp":11.1900788346,"elo":1667.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.4807977389,"x_train":649.8093981365,"imp":11.0914491011,"elo":1682.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.4970654764,"x_train":875.0118329531,"imp":11.0230785467,"elo":1696.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6991145285,"x_train":2.330497005,"imp":8.6491122691,"elo":1890.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2909668033,"x_train":1.1055094044,"imp":5.9137624298,"elo":1876.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9466920315,"x_train":85.2279597522,"imp":32.899788994,"elo":896.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9477715354,"x_train":99.1248084332,"imp":32.7796115249,"elo":910.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9426321268,"x_train":139.1755162215,"imp":32.4329755298,"elo":919.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.8633854515,"x_train":198.0940117736,"imp":31.6399865293,"elo":946.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.8789533666,"x_train":444.3646878451,"imp":30.0949731269,"elo":971.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.1163669829,"x_train":895.5564058915,"imp":28.5399652438,"elo":1012.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5458134689,"x_train":1773.8913331235,"imp":27.8873277974,"elo":1031.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.78350456,"x_train":2617.7743320608,"imp":27.1154322071,"elo":1057.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1480810234,"x_train":3486.1100818653,"imp":26.5078469087,"elo":1087.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1710376878,"x_train":0.6314611374,"imp":13.7835229791,"elo":1497.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0468974441,"x_train":1.8436799882,"imp":19.8900519711,"elo":1165.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1038006698,"x_train":6.4007633821,"imp":18.1795243011,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2635867973,"x_train":27.9733904091,"imp":16.5495848047,"elo":1321.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4134180888,"x_train":59.9007241046,"imp":15.6091450562,"elo":1374.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.593126281,"x_train":143.3365257329,"imp":14.6701208982,"elo":1413.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8268268446,"x_train":278.2745551506,"imp":14.183180672,"elo":1446.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0814556315,"x_train":543.6674489023,"imp":13.8593137304,"elo":1474.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3191008047,"x_train":841.4117357913,"imp":13.6597843306,"elo":1491.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5861131346,"x_train":1128.7390334238,"imp":13.5312380654,"elo":1498.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0471622003,"x_train":0.4669717504,"imp":12.9118537073,"elo":1520.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1406300688,"x_train":2.0321466005,"imp":11.8212367436,"elo":1617.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2332870616,"x_train":5.5778533182,"imp":11.3229193666,"elo":1667.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2722786888,"x_train":9.7317062539,"imp":10.9403239771,"elo":1700.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3510039423,"x_train":25.7016530469,"imp":10.628774182,"elo":1718.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3533147712,"x_train":51.0311253263,"imp":10.5022373116,"elo":1732.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3673660669,"x_train":101.303854312,"imp":10.4647468749,"elo":1740.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.417887152,"x_train":150.6411976385,"imp":10.4529428699,"elo":1742.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4302741489,"x_train":202.5364616477,"imp":10.4820421716,"elo":1743.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":7.8904270487,"x_train":68.7091297861,"imp":19.4673543521,"elo":1353.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":15.1962130277,"x_train":129.9001536616,"imp":17.1822083639,"elo":1413.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":26.5841757971,"x_train":315.8079093334,"imp":14.4126589357,"elo":1512.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.6468979155,"x_train":621.987734561,"imp":12.5570854498,"elo":1619.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":69.4279053206,"x_train":1537.7988263416,"imp":10.8050873617,"elo":1736.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":70.1635578676,"x_train":1598.1220946317,"imp":10.9608112326,"elo":1733.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1830265708,"x_train":2.9551878511,"imp":35.1302552748,"elo":878.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.3237118849,"x_train":6.5948259183,"imp":27.689365732,"elo":1018.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.5569193127,"x_train":17.3219967074,"imp":19.5874877747,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.8277646092,"x_train":35.2109614365,"imp":16.4609945688,"elo":1382.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.1468593548,"x_train":88.816423174,"imp":15.4820902855,"elo":1428.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.4856238345,"x_train":176.8663518766,"imp":14.9281733183,"elo":1441.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.8509761365,"x_train":354.8398796964,"imp":14.4325741349,"elo":1449.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.9863302204,"x_train":531.8168061758,"imp":14.0604231438,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.170390818,"x_train":713.0056913426,"imp":13.6083145675,"elo":1461.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..8e19218a6e89d885add6fc9c60f3e6533be9bcb2
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_classification/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "AutoGluon 1.4 (best, 4h)", "TabFM (default)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "XGBoost (tuned + ensembled)", "iLTM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "CatBoost (tuned)", "RealTabPFN-2.5 (tuned)", "LightGBM (tuned + ensembled)", "ChimeraBoost (tuned)", "TabM (tuned + ensembled)", "RealMLP (tuned + ensembled)", "CatBoost (default)", "ModernNCA (tuned)", "XGBoost (tuned)", "LightGBM (tuned)", "RealMLP (tuned)", "TabM (tuned)", "ChimeraBoost (default)", "RealTabPFN-2.5 (default)", "TabICL (default)", "iLTM (tuned)", "XGBoost (default)", "EBM (tuned + ensembled)", "TabSwift (default)", "TorchMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "TabM (default)", "xRFM (tuned + ensembled)", "xRFM (tuned)", "RealMLP (default)", "TabDPT (tuned)", "EBM (tuned)", "RandomForest (tuned + ensembled)", "EBM (default)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned)", "BetaTabPFN (default)", "iLTM (default)", "LimiX (default)", "RandomForest (tuned)", "ExtraTrees (tuned + ensembled)", "LightGBM (default)", "ModernNCA (default)", "SAP-RPT-OSS (default)", "ExtraTrees (tuned)", "TabDPT (default)", "FastaiMLP (tuned)", "TorchMLP (default)", "TabSTAR (tuned)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "KNN (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "Linear (tuned)", "RandomForest (default)", "OrionMSP (default)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "TabSTAR (default)", "PerpetualBooster (tuned)", "Linear (default)", "xRFM (default)", "KNN (tuned)", "TabFlex (default)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.6000000238418579, 0.5, 0.5, 0.699999988079071, 0.699999988079071, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4000000059604645, NaN, 0.6000000238418579, 0.5, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5, 0.4000000059604645, NaN, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 1.0, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.5, 0.5, 0.30000001192092896, NaN, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, NaN, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.5, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 1.0, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.699999988079071, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, NaN, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, NaN, 0.4000000059604645, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.5, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.6000000238418579, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.6000000238418579, NaN, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.5, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.5, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.6000000238418579, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.5, 0.5, 0.6000000238418579, NaN, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.5, 0.4000000059604645, 0.5, NaN, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.699999988079071, 0.5, 0.5, 0.699999988079071, 0.44999998807907104, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.5, 0.5, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.5, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.5, 0.4000000059604645, 0.6000000238418579, NaN, 0.4000000059604645, 0.30000001192092896, 0.6000000238418579, 0.6000000238418579, 0.5, 0.699999988079071, 0.5, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.6000000238418579, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.5, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.5, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.5, 0.699999988079071, 0.4000000059604645, NaN, 0.6000000238418579, 0.30000001192092896, 0.5, 0.6000000238418579, 0.5, 0.5, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 1.0, 0.699999988079071, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.5, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.5, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, 0.5, 0.699999988079071, 0.4000000059604645, NaN, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.6000000238418579, 0.6000000238418579, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.5, 0.5, 0.30000001192092896, 0.5, 0.5, 0.5, 0.4000000059604645, 0.4000000059604645, NaN, 0.5, 0.5, 0.5, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.5, 0.699999988079071, 0.5, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, NaN, 0.5, 0.4000000059604645, 0.5, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.5, 0.4000000059604645, 0.550000011920929, 0.4000000059604645, 0.5, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.5, 0.5, NaN, 0.5, 0.5, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.5, 0.6000000238418579, 0.5, NaN, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, 0.5, 0.5, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, NaN, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.699999988079071, 1.0, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.5, 0.800000011920929, 0.5, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 1.0, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.30000001192092896, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.5, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 1.0, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.5, 0.5, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.699999988079071, NaN, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 1.0, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.5, 0.6000000238418579, NaN, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 1.0, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.5, 0.30000001192092896, NaN, 0.5, 0.6000000238418579, 0.5, 0.5, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, 0.30000001192092896, 0.5, NaN, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.5, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, NaN, 0.5, 0.5, 0.5, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.5, NaN, 0.5, 0.5, 0.5, 0.6000000238418579, 0.699999988079071, 0.5, 0.5, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 1.0, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0], [0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.5, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.5, 0.4000000059604645, 0.5, 0.20000000298023224, 0.5, 0.4000000059604645, 0.5, 0.5, NaN, 0.5, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.5, NaN, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.5, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.5, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.6000000238418579, NaN, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.699999988079071, 0.6000000238418579, NaN, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.0, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, 0.5, 0.5, NaN, 0.5, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.699999988079071, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.10000000149011612, 0.4000000059604645, 0.20000000298023224, 0.10000000149011612, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.5, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.5, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, NaN, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.5, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.0, 0.4000000059604645, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.5, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.0, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.6000000238418579, 0.4000000059604645, NaN, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.699999988079071, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.8500000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.10000000149011612, 0.4000000059604645, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.10000000149011612, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, NaN, 0.5, 0.5, 0.5, 0.5, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.0, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4000000059604645, 0.6000000238418579, 0.5, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.4000000059604645, NaN, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 1.0, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.5, 0.6000000238418579, 0.5, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.5, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.0, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, NaN, 0.5, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.5, 0.5, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.5, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.5, NaN, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 1.0], [0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.5, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.5, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, NaN, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.5, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.5, NaN, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.0, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.6000000238418579, 0.4000000059604645, 0.5, NaN, 0.5, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.5, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.5, NaN, 0.5, 0.5, 0.5, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, 0.30000001192092896, 0.5, 0.5, 0.5, NaN, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.5, 0.30000001192092896, 0.5, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.30000001192092896, NaN, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.15000000596046448, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.5, 0.4000000059604645, 0.5, 0.5, 0.5, 0.4000000059604645, NaN, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.5, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.5, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, NaN, 0.5, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.0, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.10000000149011612, 0.4000000059604645, 0.5, 0.30000001192092896, 0.5, NaN, 0.699999988079071, 0.5, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.5, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, NaN, 0.5, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.5, 0.20000000298023224, 0.10000000149011612, 0.5, 0.5, NaN, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.0, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, NaN, 0.30000001192092896, 0.4000000059604645, 0.8500000238418579, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.0, 0.4000000059604645, 0.5, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.5, 0.699999988079071, NaN, 0.4000000059604645, 0.699999988079071, 0.699999988079071, 0.5, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.5, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.10000000149011612, 0.5, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.0, 0.0, 0.0, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.15000000596046448, 0.30000001192092896, 0.4000000059604645, NaN, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.30000001192092896, 0.4000000059604645, 0.5, NaN, 0.6000000238418579, 0.5, 0.800000011920929, 0.5, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.5, 0.6000000238418579, 0.800000011920929, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.4000000059604645, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.30000001192092896, 0.5, 0.4000000059604645, NaN, 0.5, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.800000011920929, 0.5, 0.699999988079071, 0.699999988079071, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.5, 0.4000000059604645, 0.20000000298023224, 0.5, 0.5, 0.5, NaN, 0.6000000238418579, 0.20000000298023224, 1.0, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.5, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, NaN, 0.30000001192092896, 0.6000000238418579, 0.5, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.30000001192092896, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.0, 0.30000001192092896, 0.5, 0.4000000059604645, 0.0, 0.5, 0.5, 0.4000000059604645, 0.800000011920929, 0.699999988079071, NaN, 0.800000011920929, 0.5, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.5, 0.30000001192092896, 0.20000000298023224, 0.5, 0.4000000059604645, 0.5, 0.0, 0.4000000059604645, 0.20000000298023224, NaN, 0.5, 0.6000000238418579, 0.30000001192092896, 0.4000000059604645, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, 0.5, 0.5, NaN, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.8999999761581421], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, NaN, 0.4000000059604645, 0.5, 0.5, 0.699999988079071], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.30000001192092896, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.0, 0.10000000149011612, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.699999988079071, 0.4000000059604645, 0.6000000238418579, NaN, 0.5, 0.800000011920929, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.30000001192092896, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.0, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.6000000238418579, 0.4000000059604645, 0.5, 0.5, NaN, 0.6000000238418579, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.0, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.5, 0.20000000298023224, 0.4000000059604645, NaN, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.10000000149011612, 0.30000001192092896, 0.0, 0.20000000298023224, 0.20000000298023224, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9602739697},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9424657507},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9082191754},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8712328721},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8602739703},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.828767121},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7917808186},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.771232877},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7657534241},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7616438359},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7513698623},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7506849291},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7479452023},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.746575342},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7397260297},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.7369862994},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7328767118},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7301369856},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7273972572},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.7089041096},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.7054794523},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6849315072},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6630136997},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6616438383},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.6534246581},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6452054803},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6301369878},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6000000027},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5904109585},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5835616412},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5753424685},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.572602744},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.5712328801},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5424657569},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5369863026},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5356164383},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5246575362},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4917808234},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4815068543},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4780821947},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4671232903},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4630137013},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4575342483},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4397260284},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.438356166},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.4383561663},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4356164416},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4273972607},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4232876751},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3972602748},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3849315062},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3753424673},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3712328801},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.3431506868},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3136986311},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3000000003},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2452054812},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2438356182},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2363013735},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2342465766},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2315068513},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2116438384},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2068493182},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.1808219196},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1684931531},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1465753453},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1452054803},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1328767138},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1328767137},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.130136988},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1136986315},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1000000022},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0808219193},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0205479456}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..22da9d5a886c4b239dfb1843d70bd4b90bb8b509
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,"AutoGluon 1.5 (extreme, 4h)","AutoGluon 1.4 (best, 4h)",TabFM (default),TabPFN-3 (default),TabPFN-2.6 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),ModernNCA (tuned + ensembled),XGBoost (tuned + ensembled),iLTM (tuned + ensembled),CatBoost (tuned + ensembled),ChimeraBoost (tuned + ensembled),CatBoost (tuned),RealTabPFN-2.5 (tuned),LightGBM (tuned + ensembled),ChimeraBoost (tuned),TabM (tuned + ensembled),RealMLP (tuned + ensembled),CatBoost (default),ModernNCA (tuned),XGBoost (tuned),LightGBM (tuned),RealMLP (tuned),TabM (tuned),ChimeraBoost (default),RealTabPFN-2.5 (default),TabICL (default),iLTM (tuned),XGBoost (default),EBM (tuned + ensembled),TabSwift (default),TorchMLP (tuned + ensembled),TabDPT-Turbo (default),TabDPT (tuned + ensembled),TabM (default),xRFM (tuned + ensembled),xRFM (tuned),RealMLP (default),TabDPT (tuned),EBM (tuned),RandomForest (tuned + ensembled),EBM (default),FastaiMLP (tuned + ensembled),TorchMLP (tuned),BetaTabPFN (default),iLTM (default),LimiX (default),RandomForest (tuned),ExtraTrees (tuned + ensembled),LightGBM (default),ModernNCA (default),SAP-RPT-OSS (default),ExtraTrees (tuned),TabDPT (default),FastaiMLP (tuned),TorchMLP (default),TabSTAR (tuned),FastaiMLP (default),Linear (tuned + ensembled),KNN (tuned + ensembled),TabSTAR (tuned + ensembled),Linear (tuned),RandomForest (default),OrionMSP (default),PerpetualBooster (tuned + ensembled),ExtraTrees (default),TabSTAR (default),PerpetualBooster (tuned),Linear (default),xRFM (default),KNN (tuned),TabFlex (default),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.6,0.5,0.5,0.7,0.7,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,0.9,1.0,1.0,0.9,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.4,,0.6,0.5,0.8,0.7,0.7,0.7,1.0,1.0,1.0,0.9,1.0,0.9,0.9,0.9,0.8,1.0,0.9,0.9,1.0,1.0,1.0,0.8,1.0,0.8,0.9,1.0,0.9,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.5,0.4,,0.7,0.8,0.8,0.8,0.8,0.8,0.7,0.8,0.8,0.8,0.9,0.9,0.8,1.0,0.8,0.8,0.8,0.7,1.0,0.8,1.0,0.9,0.9,0.8,0.9,0.9,1.0,0.8,0.9,1.0,0.9,1.0,0.9,0.9,1.0,0.9,1.0,1.0,1.0,0.9,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.5,0.5,0.3,,0.6,0.7,0.6,0.9,0.7,0.7,0.7,0.8,0.7,0.8,0.8,0.8,0.8,0.7,0.7,0.8,0.7,1.0,0.7,0.9,0.8,0.9,0.8,0.9,0.9,0.9,0.9,0.7,1.0,0.9,1.0,0.8,0.8,1.0,0.9,1.0,0.9,1.0,0.9,1.0,1.0,1.0,0.8,1.0,0.9,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.3,0.2,0.2,0.4,,0.7,0.6,0.8,0.5,0.6,0.7,0.8,0.7,0.7,0.9,0.8,0.8,0.8,0.7,0.7,0.7,1.0,0.7,0.9,0.7,0.9,0.9,0.8,0.9,1.0,0.8,0.8,1.0,1.0,1.0,0.7,0.8,1.0,0.9,1.0,1.0,1.0,0.9,1.0,0.9,1.0,0.8,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.3,0.3,0.2,0.3,0.3,,0.7,0.7,0.7,0.7,0.7,0.7,0.7,0.8,0.8,0.7,0.6,0.6,0.7,0.8,0.7,0.9,0.7,0.6,0.7,0.8,0.8,0.8,0.7,0.7,0.8,0.8,1.0,0.9,1.0,0.7,0.7,0.9,0.9,1.0,1.0,1.0,0.9,0.8,1.0,0.9,0.8,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,0.3,0.2,0.4,0.4,0.3,,0.4,0.6,0.5,0.7,0.6,0.6,0.9,0.6,0.7,0.5,0.6,0.7,0.7,0.7,0.8,0.6,0.7,0.7,0.8,0.8,0.7,0.8,0.9,0.9,0.7,0.8,0.8,0.9,0.7,0.7,0.9,0.9,0.9,0.9,0.9,0.8,0.9,1.0,0.9,0.6,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.3,0.2,0.1,0.2,0.3,0.6,,0.6,0.5,0.5,0.6,0.5,0.6,0.5,0.6,0.6,0.6,0.6,0.7,0.6,0.6,0.8,0.7,0.8,0.8,0.8,0.7,0.6,0.8,0.7,0.8,0.7,1.0,0.8,0.9,0.8,0.9,0.9,0.8,0.8,0.9,1.0,0.9,0.9,0.9,0.8,0.8,0.9,1.0,1.0,0.9,0.9,1.0,1.0,0.9,0.9,1.0,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.2,0.3,0.5,0.3,0.4,0.4,,0.4,0.6,0.5,0.5,0.6,0.6,0.6,0.7,0.7,0.7,0.5,0.8,0.6,0.6,0.7,0.7,0.7,0.7,0.7,0.9,0.8,0.6,0.9,0.6,0.8,0.9,0.7,0.8,0.9,0.7,0.8,0.9,1.0,0.8,0.9,0.8,0.9,0.6,0.9,1.0,1.0,1.0,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.3,0.3,0.4,0.3,0.5,0.5,0.6,,0.5,0.6,0.5,0.6,0.4,0.6,0.6,0.5,0.6,0.6,0.7,0.5,0.7,0.6,0.7,0.8,0.7,0.7,0.7,0.7,0.7,0.9,0.6,0.7,0.8,0.9,0.8,0.9,0.8,0.6,0.9,0.8,0.9,1.0,0.8,1.0,0.8,0.9,0.9,0.9,0.9,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.2,0.3,0.3,0.3,0.3,0.5,0.4,0.5,,0.6,0.6,0.4,0.4,0.7,0.5,0.5,0.7,0.45,0.6,0.5,0.7,0.6,0.8,0.8,0.7,0.9,0.8,0.6,0.7,0.8,0.6,0.7,0.7,0.9,1.0,0.8,0.8,0.6,0.9,0.9,0.9,1.0,0.8,0.9,0.7,0.9,1.0,0.9,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.1,0.2,0.2,0.2,0.3,0.4,0.4,0.5,0.4,0.4,,0.4,0.6,0.5,0.6,0.6,0.7,0.6,0.6,0.5,0.5,0.8,0.6,0.8,0.7,0.8,0.8,0.7,0.8,0.5,0.8,0.6,0.9,0.8,0.9,1.0,0.9,0.8,0.9,0.9,1.0,0.9,0.9,0.8,0.9,0.7,0.9,0.9,0.9,0.9,1.0,0.9,0.9,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.2,0.3,0.3,0.3,0.4,0.5,0.5,0.5,0.4,0.6,,0.4,0.3,0.6,0.6,0.5,0.7,0.5,0.7,0.5,0.7,0.6,0.8,0.7,0.7,0.9,0.8,0.7,0.7,0.8,0.6,0.7,0.8,0.9,0.9,0.9,0.7,0.6,0.9,0.9,0.9,1.0,0.8,0.9,0.7,0.9,0.9,0.9,0.9,1.0,0.9,0.9,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.1,0.1,0.2,0.3,0.2,0.1,0.4,0.4,0.4,0.6,0.4,0.6,,0.6,0.6,0.5,0.5,0.7,0.6,0.6,0.8,0.6,0.7,0.6,0.7,0.8,0.7,0.8,0.9,0.8,0.7,0.8,0.8,0.9,0.6,0.7,0.9,0.9,0.9,0.9,0.9,0.8,0.9,0.9,0.8,0.6,1.0,1.0,1.0,0.9,0.9,0.8,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.1,0.1,0.2,0.1,0.2,0.4,0.5,0.4,0.6,0.6,0.5,0.7,0.4,,0.6,0.3,0.5,0.6,0.5,0.5,0.8,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.7,0.6,0.7,0.8,0.6,0.8,0.7,0.8,0.8,0.7,1.0,0.9,1.0,0.8,0.9,0.8,1.0,0.7,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.1,0.2,0.2,0.2,0.3,0.3,0.4,0.4,0.4,0.3,0.4,0.4,0.4,0.4,,0.6,0.6,0.6,0.7,0.5,0.4,0.8,0.6,0.7,0.7,0.8,0.8,0.7,0.9,0.6,0.8,0.6,0.9,0.8,0.9,1.0,0.9,0.8,0.9,0.9,1.0,0.9,0.9,0.9,0.8,0.7,0.9,0.9,0.9,0.9,1.0,0.8,0.8,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.1,0.2,0.0,0.2,0.2,0.4,0.5,0.4,0.3,0.4,0.5,0.4,0.4,0.5,0.7,0.4,,0.6,0.5,0.7,0.6,0.6,0.7,0.6,0.5,0.6,0.6,0.6,0.6,0.9,0.7,0.7,0.8,0.7,1.0,0.6,0.6,1.0,0.9,0.9,0.9,0.9,0.9,0.8,0.9,0.7,0.7,0.9,0.8,1.0,1.0,0.9,0.8,0.9,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.2,0.3,0.2,0.4,0.4,0.4,0.3,0.5,0.5,0.3,0.5,0.5,0.5,0.4,0.4,,0.5,0.5,0.5,0.6,0.8,0.6,0.5,0.7,0.5,0.7,0.7,0.7,0.6,0.8,0.7,0.7,0.9,0.8,0.9,1.0,0.7,0.8,0.9,0.8,1.0,0.9,0.8,0.8,0.8,0.9,0.9,1.0,0.9,1.0,0.8,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.1,0.2,0.3,0.3,0.3,0.3,0.4,0.3,0.4,0.3,0.4,0.3,0.3,0.4,0.4,0.5,0.5,,0.5,0.4,0.5,0.7,0.6,0.7,0.8,0.7,0.8,0.7,0.7,0.7,0.9,0.6,0.7,0.7,0.9,0.9,0.9,0.7,0.7,0.9,0.9,0.9,1.0,0.8,0.9,0.7,0.9,0.9,0.9,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.1,0.1,0.2,0.2,0.3,0.2,0.3,0.3,0.5,0.4,0.55,0.4,0.5,0.4,0.5,0.3,0.3,0.5,0.5,,0.5,0.5,0.6,0.4,0.6,0.7,0.7,0.6,0.5,0.6,0.8,0.8,0.7,0.8,0.7,0.6,0.8,0.8,1.0,0.7,0.8,0.8,0.8,0.8,0.9,0.9,0.8,0.9,0.9,0.9,1.0,0.9,0.9,1.0,1.0,0.9,0.9,1.0,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.3,0.3,0.3,0.3,0.3,0.4,0.2,0.3,0.4,0.5,0.3,0.4,0.5,0.5,0.4,0.5,0.6,0.5,,0.6,0.5,0.6,0.6,0.5,0.6,0.6,0.9,0.6,0.6,0.7,0.6,0.9,0.8,0.8,0.7,0.8,0.7,0.8,0.8,0.9,0.7,0.9,0.9,0.9,0.6,0.8,0.9,0.9,1.0,0.9,0.9,0.9,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.1,0.2,0.4,0.4,0.5,0.5,0.5,0.5,0.2,0.2,0.6,0.4,0.4,0.5,0.5,0.4,,0.5,0.6,0.6,0.6,0.5,0.6,0.7,0.7,0.4,0.6,0.8,0.7,0.8,0.7,0.8,0.8,0.7,0.9,0.8,0.9,0.8,0.8,0.8,1.0,0.7,1.0,0.8,1.0,1.0,0.9,0.9,1.0,0.9,0.9,1.0,0.9,0.8,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.2,0.3,0.3,0.3,0.4,0.2,0.4,0.3,0.3,0.2,0.3,0.4,0.4,0.2,0.3,0.2,0.3,0.4,0.5,0.5,,0.4,0.4,0.6,0.5,0.7,0.7,0.6,0.5,0.8,0.5,0.7,0.7,0.9,0.7,1.0,0.7,0.7,0.8,0.7,0.9,0.9,0.8,0.7,0.8,0.8,0.7,0.9,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.1,0.2,0.0,0.1,0.1,0.4,0.3,0.3,0.3,0.4,0.4,0.4,0.4,0.3,0.4,0.4,0.4,0.4,0.4,0.6,0.4,0.4,0.6,,0.3,0.5,0.6,0.5,0.6,0.7,0.6,0.6,0.7,0.6,0.8,0.5,0.6,0.8,0.8,0.9,0.7,0.8,0.8,0.8,0.9,0.6,0.7,0.8,0.7,0.8,1.0,0.9,0.7,0.8,0.9,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.1,0.2,0.3,0.3,0.3,0.2,0.3,0.3,0.2,0.2,0.2,0.4,0.4,0.3,0.5,0.5,0.3,0.4,0.4,0.4,0.6,0.7,,0.4,0.5,0.6,0.6,0.7,0.5,0.7,0.6,0.8,0.6,0.9,0.8,0.7,0.7,0.8,0.8,0.8,0.8,0.9,0.7,0.7,0.7,0.8,0.7,0.7,0.8,1.0,0.7,0.8,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,0.9,1.0
+RealTabPFN-2.5 (default),0.0,0.2,0.1,0.1,0.1,0.2,0.2,0.2,0.3,0.2,0.2,0.3,0.3,0.3,0.4,0.3,0.4,0.3,0.2,0.3,0.5,0.4,0.4,0.5,0.6,,0.7,0.7,0.7,0.6,0.8,0.6,0.6,0.8,0.7,0.7,0.6,0.7,0.7,0.8,0.7,0.8,0.7,0.8,1.0,0.7,0.6,0.8,0.8,1.0,0.9,0.9,0.8,0.8,0.9,1.0,1.0,0.8,0.9,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0
+TabICL (default),0.1,0.1,0.2,0.2,0.1,0.2,0.2,0.2,0.3,0.3,0.3,0.2,0.3,0.2,0.4,0.2,0.4,0.5,0.3,0.3,0.4,0.5,0.5,0.4,0.5,0.3,,0.5,0.6,0.5,0.5,0.7,0.7,0.9,0.6,0.6,0.7,0.7,0.9,0.6,0.8,0.6,0.7,0.7,0.7,0.8,0.7,0.8,0.8,0.9,0.9,0.8,1.0,1.0,0.8,0.8,0.9,0.8,0.8,0.9,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+iLTM (tuned),0.1,0.0,0.1,0.1,0.2,0.2,0.3,0.3,0.3,0.3,0.1,0.2,0.1,0.3,0.2,0.2,0.4,0.3,0.2,0.4,0.4,0.4,0.3,0.5,0.4,0.3,0.5,,0.4,0.5,0.6,0.6,0.4,0.6,0.6,0.7,0.6,0.6,0.6,0.6,0.7,0.6,0.8,0.7,0.7,0.7,0.7,0.8,0.6,0.8,0.8,0.9,0.9,0.9,0.8,0.9,1.0,0.9,0.9,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.1,0.1,0.1,0.1,0.1,0.3,0.2,0.4,0.1,0.3,0.2,0.3,0.2,0.2,0.2,0.3,0.4,0.3,0.3,0.5,0.1,0.3,0.3,0.4,0.4,0.3,0.4,0.6,,0.5,0.5,0.5,0.5,0.6,0.5,0.7,0.7,0.5,0.7,0.6,0.8,0.7,0.7,0.6,0.6,0.9,0.7,0.8,0.9,0.6,0.8,0.9,0.9,0.8,0.7,0.9,1.0,0.8,0.9,0.9,1.0,0.9,0.9,1.0,0.9,0.9,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.1,0.0,0.3,0.1,0.2,0.2,0.3,0.4,0.2,0.3,0.1,0.3,0.1,0.1,0.3,0.3,0.4,0.4,0.3,0.4,0.3,0.3,0.4,0.5,0.5,0.5,,0.5,0.5,0.5,0.6,0.7,0.5,0.5,0.7,0.7,0.9,0.7,1.0,0.7,0.8,0.8,0.6,0.6,0.8,0.7,0.7,0.9,0.9,0.7,0.7,0.8,0.9,1.0,0.9,0.9,0.9,1.0,0.9,0.9,0.9,0.9,0.9,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0
+TabSwift (default),0.1,0.0,0.2,0.1,0.2,0.2,0.1,0.3,0.4,0.3,0.3,0.5,0.3,0.2,0.4,0.4,0.3,0.4,0.3,0.2,0.4,0.6,0.5,0.4,0.5,0.2,0.5,0.4,0.5,0.5,,0.5,0.4,0.6,0.6,0.7,0.6,0.5,0.6,0.6,0.6,0.7,0.7,0.6,0.7,0.7,0.7,0.6,0.6,0.8,0.7,0.7,0.7,0.9,0.7,0.7,0.8,0.7,0.7,0.8,0.8,0.8,0.9,1.0,1.0,1.0,0.9,1.0,0.8,0.9,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.1,0.3,0.2,0.2,0.3,0.2,0.1,0.1,0.2,0.2,0.2,0.3,0.3,0.2,0.3,0.2,0.1,0.2,0.3,0.4,0.2,0.4,0.3,0.4,0.3,0.4,0.5,0.5,0.5,,0.4,0.3,0.4,0.6,0.6,0.6,0.7,0.6,0.8,0.5,0.6,0.8,0.6,0.7,0.6,0.8,0.8,0.6,0.7,0.9,0.8,0.8,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.1,0.0,0.0,0.0,0.0,0.2,0.3,0.4,0.4,0.4,0.4,0.4,0.2,0.2,0.4,0.2,0.3,0.4,0.3,0.4,0.2,0.5,0.3,0.4,0.4,0.3,0.6,0.5,0.5,0.6,0.6,,0.4,0.5,0.6,0.6,0.6,0.7,0.6,0.6,0.6,0.6,0.8,0.6,0.7,0.8,0.8,0.7,0.7,0.6,0.7,0.7,0.9,0.9,0.8,0.8,0.9,0.7,0.8,0.8,0.7,1.0,0.9,1.0,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0
+TabDPT (tuned + ensembled),0.0,0.0,0.1,0.1,0.0,0.1,0.2,0.0,0.2,0.3,0.3,0.1,0.3,0.2,0.4,0.1,0.3,0.3,0.3,0.2,0.1,0.3,0.3,0.4,0.2,0.2,0.1,0.4,0.4,0.4,0.4,0.7,0.6,,0.5,0.6,0.6,0.7,0.7,0.6,0.6,0.6,0.6,0.8,0.6,0.5,0.6,0.6,0.6,0.8,0.9,0.7,0.7,0.9,0.9,0.8,0.8,0.9,0.8,0.9,0.8,0.8,1.0,0.9,1.0,1.0,1.0,1.0,1.0,0.9,1.0,0.9,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.2,0.1,0.2,0.3,0.2,0.2,0.1,0.2,0.2,0.0,0.1,0.3,0.3,0.2,0.2,0.3,0.2,0.4,0.3,0.4,0.4,0.5,0.3,0.4,0.6,0.5,0.5,,0.5,0.5,0.7,0.6,0.6,0.7,0.7,0.8,0.6,0.8,0.7,0.6,0.7,0.6,0.8,0.8,0.7,0.7,0.7,0.9,0.9,1.0,0.9,0.8,0.9,1.0,0.7,0.8,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.1,0.2,0.3,0.3,0.3,0.1,0.3,0.1,0.1,0.1,0.1,0.4,0.3,0.1,0.4,0.2,0.1,0.4,0.2,0.3,0.1,0.5,0.1,0.3,0.4,0.3,0.3,0.5,0.3,0.4,0.4,0.4,0.5,,0.5,0.6,0.5,0.7,0.7,0.6,0.7,0.5,0.7,0.6,0.7,0.7,0.7,0.5,0.7,0.7,0.7,0.6,0.8,0.8,0.9,0.9,1.0,1.0,0.9,1.0,0.9,1.0,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,0.9,1.0
+xRFM (tuned),0.0,0.0,0.1,0.2,0.2,0.3,0.3,0.2,0.2,0.2,0.0,0.0,0.1,0.3,0.2,0.0,0.4,0.1,0.1,0.2,0.3,0.2,0.3,0.4,0.2,0.4,0.3,0.4,0.3,0.5,0.4,0.4,0.4,0.4,0.5,0.5,,0.6,0.4,0.6,0.6,0.5,0.6,0.7,0.6,0.5,0.6,0.7,0.7,0.6,0.6,0.9,0.7,0.6,0.8,0.9,0.9,0.9,1.0,1.0,0.9,1.0,0.8,0.9,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,0.9,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.2,0.1,0.1,0.1,0.2,0.1,0.0,0.0,0.1,0.2,0.2,0.2,0.0,0.2,0.3,0.3,0.3,0.4,0.5,0.3,0.5,0.4,0.4,0.3,0.3,0.4,0.4,,0.6,0.6,0.6,0.6,0.5,0.7,0.6,0.5,0.5,0.7,0.6,0.6,0.7,0.7,0.7,0.7,1.0,0.9,0.9,1.0,0.9,0.9,0.9,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.1,0.0,0.1,0.1,0.1,0.1,0.1,0.1,0.3,0.2,0.2,0.2,0.3,0.1,0.3,0.2,0.1,0.3,0.3,0.0,0.3,0.3,0.3,0.2,0.3,0.3,0.1,0.4,0.3,0.3,0.4,0.3,0.3,0.3,0.4,0.5,0.6,0.4,,0.5,0.4,0.4,0.4,0.7,0.4,0.6,0.6,0.6,0.5,0.6,0.5,0.6,0.7,0.85,0.8,0.8,0.8,0.9,0.7,0.8,0.8,0.7,0.8,0.8,1.0,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.2,0.2,0.4,0.4,0.1,0.4,0.1,0.0,0.1,0.1,0.2,0.3,0.3,0.2,0.1,0.3,0.1,0.2,0.2,0.4,0.4,0.4,0.1,0.4,0.4,0.4,0.4,0.4,0.3,0.4,0.4,0.5,,0.5,0.5,0.5,0.5,0.5,0.6,0.6,0.6,0.6,0.6,0.6,0.9,0.7,0.6,0.8,0.8,1.0,0.8,0.7,0.9,1.0,0.8,0.8,0.9,0.9,0.8,1.0,0.9,0.9,0.8,1.0,1.0,0.9,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.2,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.2,0.2,0.2,0.2,0.3,0.2,0.3,0.2,0.3,0.2,0.3,0.4,0.2,0.4,0.4,0.3,0.3,0.4,0.4,0.6,0.5,,0.4,0.4,0.6,0.6,0.7,0.6,0.6,0.6,0.6,0.6,0.7,0.7,0.6,0.8,0.8,1.0,0.9,0.8,0.9,1.0,0.8,0.9,0.9,0.9,1.0,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.2,0.1,0.0,0.1,0.1,0.0,0.0,0.1,0.2,0.1,0.2,0.1,0.1,0.3,0.2,0.2,0.2,0.4,0.4,0.3,0.0,0.3,0.5,0.4,0.4,0.3,0.4,0.5,0.4,0.6,0.5,0.6,,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.7,0.8,0.8,0.7,0.7,0.8,0.8,0.9,0.8,0.7,0.9,0.9,0.8,0.8,0.9,0.9,0.8,1.0,0.9,0.9,0.8,1.0,1.0,0.9,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.1,0.1,0.1,0.1,0.2,0.0,0.2,0.1,0.1,0.1,0.1,0.2,0.2,0.1,0.1,0.0,0.1,0.2,0.3,0.2,0.1,0.2,0.2,0.3,0.3,0.2,0.3,0.3,0.3,0.4,0.4,0.4,0.2,0.3,0.4,0.5,0.6,0.5,0.6,0.4,,0.6,0.4,0.4,0.6,0.6,0.6,0.6,0.7,0.7,0.6,0.6,0.8,0.7,0.7,1.0,0.9,0.9,0.7,0.9,0.8,0.7,0.9,0.9,0.9,0.9,1.0,1.0,0.9,0.9,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.2,0.1,0.1,0.1,0.0,0.0,0.1,0.0,0.1,0.1,0.1,0.2,0.1,0.0,0.2,0.1,0.2,0.1,0.2,0.1,0.2,0.3,0.3,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.5,0.3,0.3,0.3,0.5,0.4,0.4,0.4,,0.5,0.6,0.5,0.4,0.6,0.6,0.8,0.7,0.7,0.6,0.5,0.9,0.9,0.8,0.9,0.8,0.9,0.9,0.8,0.9,1.0,0.9,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.2,0.2,0.2,0.2,0.2,0.1,0.2,0.1,0.1,0.2,0.2,0.1,0.1,0.2,0.2,0.1,0.3,0.0,0.3,0.3,0.4,0.2,0.3,0.4,0.4,0.4,0.2,0.3,0.4,0.4,0.6,0.5,0.4,0.4,0.6,0.5,,0.5,0.6,0.5,0.5,0.6,0.5,0.7,0.5,0.5,0.8,0.7,0.8,0.8,0.7,0.9,0.8,0.7,0.7,0.9,0.9,0.8,1.0,0.9,0.9,0.8,1.0,1.0,0.9,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.1,0.0,0.1,0.1,0.1,0.2,0.0,0.2,0.3,0.2,0.1,0.1,0.1,0.0,0.3,0.4,0.3,0.3,0.2,0.3,0.1,0.4,0.3,0.3,0.3,0.5,0.3,0.4,0.5,0.5,0.4,0.4,0.3,0.4,0.6,0.4,0.5,,0.6,0.4,0.4,0.5,0.6,0.6,0.7,0.7,0.7,0.7,0.9,0.8,0.8,0.9,0.9,0.8,0.8,0.9,0.8,0.9,1.0,0.8,0.9,0.9,0.9,1.0,0.8,1.0
+LimiX (default),0.1,0.1,0.2,0.2,0.2,0.2,0.4,0.2,0.4,0.2,0.3,0.3,0.3,0.4,0.3,0.3,0.3,0.2,0.3,0.2,0.4,0.3,0.2,0.3,0.3,0.4,0.3,0.3,0.3,0.4,0.3,0.4,0.2,0.4,0.4,0.3,0.4,0.5,0.4,0.4,0.4,0.4,0.4,0.5,0.4,0.4,,0.4,0.4,0.4,0.4,0.4,0.5,0.7,0.6,0.6,0.5,0.7,0.7,0.7,0.5,0.7,0.8,0.6,0.7,0.9,0.7,0.7,0.7,0.9,0.7,0.7,0.8,0.9
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.1,0.1,0.1,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.4,0.3,0.3,0.3,0.3,0.4,0.4,0.4,0.4,0.4,0.6,0.5,0.6,0.6,,0.5,0.6,0.6,0.7,0.6,0.5,0.7,0.8,0.9,0.9,0.7,0.9,0.9,0.7,0.9,1.0,0.9,1.0,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.1,0.1,0.0,0.0,0.1,0.2,0.1,0.1,0.1,0.1,0.2,0.3,0.3,0.3,0.2,0.2,0.4,0.1,0.3,0.4,0.2,0.3,0.4,0.4,0.3,0.3,0.4,0.5,0.4,0.4,0.4,0.4,0.4,0.5,0.6,0.6,0.5,,0.5,0.6,0.5,0.7,0.6,0.6,0.6,0.8,0.7,0.7,0.8,0.8,0.9,1.0,1.0,0.8,1.0,0.8,0.8,1.0,0.9,0.9,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.1,0.0,0.0,0.1,0.0,0.0,0.1,0.1,0.1,0.0,0.1,0.2,0.3,0.0,0.1,0.2,0.4,0.3,0.2,0.4,0.3,0.2,0.2,0.5,0.4,0.4,0.4,0.4,0.4,0.3,0.4,0.4,0.4,0.5,0.6,0.4,0.5,,0.5,0.5,0.6,0.5,0.6,0.8,0.8,0.7,0.7,0.8,0.9,0.7,0.8,0.9,0.9,0.9,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.0,0.1,0.0,0.1,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.1,0.1,0.2,0.2,0.1,0.3,0.3,0.4,0.1,0.2,0.3,0.4,0.3,0.5,0.4,0.4,0.2,0.3,0.2,0.5,0.4,0.6,0.4,0.4,0.5,,0.5,0.5,0.5,0.6,0.9,0.8,0.9,0.7,0.9,0.8,0.7,0.9,0.9,0.9,1.0,1.0,0.9,0.9,0.9,1.0,0.9,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.0,0.1,0.1,0.1,0.2,0.1,0.0,0.1,0.2,0.1,0.1,0.1,0.3,0.1,0.3,0.3,0.3,0.3,0.1,0.3,0.4,0.1,0.3,0.2,0.3,0.3,0.3,0.4,0.6,0.3,0.5,0.5,0.5,,0.7,0.5,0.7,0.6,0.8,0.9,0.8,1.0,0.8,0.8,0.7,0.8,0.9,0.9,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.1,0.1,0.2,0.0,0.2,0.2,0.2,0.1,0.1,0.1,0.1,0.2,0.3,0.3,0.2,0.0,0.1,0.1,0.3,0.3,0.2,0.3,0.3,0.3,0.3,0.3,0.3,0.3,0.3,0.3,0.3,0.4,0.3,0.5,0.3,0.5,0.4,0.3,0.4,0.5,0.3,,0.6,0.5,0.5,0.6,0.5,0.7,0.6,0.6,0.7,0.9,0.9,0.8,0.9,0.8,0.9,1.0,0.9,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.0,0.0,0.2,0.1,0.1,0.1,0.0,0.1,0.0,0.2,0.2,0.2,0.2,0.0,0.1,0.2,0.3,0.1,0.2,0.1,0.1,0.3,0.4,0.4,0.3,0.15,0.4,0.4,0.3,0.4,0.4,0.5,0.3,0.3,0.5,0.4,0.5,0.5,0.5,0.4,,0.6,0.7,0.7,0.8,0.7,0.5,0.7,0.7,0.6,0.7,0.8,0.6,0.7,0.9,0.8,0.8,0.7,0.9,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.1,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.1,0.1,0.2,0.2,0.3,0.2,0.3,0.1,0.1,0.1,0.1,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.5,0.2,0.3,0.4,0.3,0.4,0.4,0.4,0.3,0.5,0.4,,0.5,0.6,0.9,0.6,0.6,0.6,0.6,0.7,0.6,0.9,0.8,0.9,0.9,0.9,0.9,0.9,0.9,0.9,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.0,0.1,0.0,0.2,0.1,0.1,0.1,0.3,0.0,0.2,0.2,0.1,0.2,0.1,0.1,0.2,0.2,0.2,0.2,0.3,0.1,0.3,0.3,0.4,0.2,0.4,0.2,0.1,0.4,0.5,0.3,0.5,,0.7,0.5,0.7,0.8,0.7,0.8,0.7,0.8,0.8,0.8,0.9,0.9,0.9,0.9,1.0,0.9,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.0,0.1,0.1,0.1,0.2,0.0,0.0,0.1,0.3,0.1,0.2,0.1,0.5,0.1,0.2,0.2,0.2,0.2,0.4,0.3,0.4,0.3,,0.5,0.6,0.6,0.9,0.6,0.6,0.7,0.8,0.7,1.0,0.8,0.8,0.7,0.9,1.0,0.8,0.9
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.1,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.2,0.2,0.1,0.2,0.1,0.3,0.1,0.1,0.1,0.1,0.1,0.1,0.0,0.1,0.2,0.1,0.2,0.0,0.2,0.2,0.2,0.3,0.1,0.3,0.3,0.1,0.1,0.5,0.2,0.1,0.5,0.5,,0.6,0.5,0.5,0.6,0.5,0.6,0.7,0.6,0.7,0.7,0.7,0.8,0.8,0.9,0.9,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.1,0.0,0.1,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.2,0.0,0.1,0.0,0.1,0.2,0.1,0.1,0.1,0.3,0.0,0.3,0.2,0.2,0.0,0.0,0.1,0.3,0.3,0.2,0.3,0.1,0.1,0.3,0.2,0.3,0.3,0.3,0.3,0.3,0.2,0.3,0.3,0.4,0.3,0.4,0.4,,0.3,0.4,0.85,0.5,0.5,0.5,0.5,0.5,0.5,0.8,0.8,0.6,0.7,0.7,0.9
+KNN (tuned + ensembled),0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.1,0.1,0.0,0.1,0.1,0.2,0.0,0.2,0.1,0.1,0.0,0.0,0.1,0.2,0.1,0.1,0.1,0.1,0.2,0.1,0.1,0.3,0.1,0.2,0.2,0.1,0.0,0.4,0.5,0.4,0.2,0.4,0.5,0.7,,0.4,0.7,0.7,0.5,0.6,0.7,0.6,0.7,0.8,0.8,1.0,0.8,0.9,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.0,0.1,0.1,0.1,0.2,0.0,0.0,0.1,0.3,0.1,0.2,0.1,0.5,0.1,0.2,0.1,0.2,0.2,0.4,0.3,0.4,0.3,0.1,0.5,0.6,0.6,,0.6,0.6,0.7,0.8,0.7,1.0,0.8,0.8,0.7,0.9,1.0,0.8,0.9
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.1,0.2,0.0,0.1,0.1,0.2,0.0,0.3,0.2,0.3,0.0,0.0,0.0,0.3,0.2,0.2,0.2,0.1,0.1,0.3,0.2,0.3,0.3,0.1,0.3,0.3,0.2,0.3,0.3,0.4,0.2,0.4,0.4,0.15,0.3,0.4,,0.5,0.5,0.5,0.5,0.5,0.5,0.8,0.8,0.6,0.7,0.7,0.9
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.1,0.0,0.1,0.1,0.1,0.1,0.0,0.0,0.0,0.2,0.1,0.2,0.2,0.2,0.2,0.1,0.2,0.2,0.2,0.3,0.2,0.2,0.1,0.0,0.2,0.1,0.3,0.1,0.4,0.3,0.3,0.4,0.5,0.5,0.3,0.4,0.5,,0.6,0.5,0.8,0.5,0.6,0.6,0.9,0.5,0.6,0.8,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.2,0.2,0.1,0.1,0.1,0.3,0.1,0.1,0.1,0.4,0.0,0.0,0.1,0.1,0.2,0.1,0.3,0.4,0.2,0.3,0.4,0.5,0.5,0.3,0.5,0.4,,0.5,0.7,0.6,0.5,0.6,0.8,0.5,0.7,0.7,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.1,0.0,0.1,0.2,0.3,0.1,0.2,0.1,0.1,0.1,0.2,0.2,0.1,0.2,0.2,0.3,0.5,0.4,0.2,0.5,0.5,0.5,,0.6,0.2,1.0,0.5,0.6,0.6,0.6,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.1,0.1,0.2,0.1,0.1,0.0,0.0,0.1,0.0,0.1,0.1,0.4,0.2,0.2,0.3,0.4,0.5,0.3,0.3,0.5,0.2,0.3,0.4,,0.3,0.6,0.5,0.8,0.6,0.6,0.8,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.3,0.0,0.2,0.0,0.0,0.0,0.2,0.3,0.1,0.1,0.0,0.3,0.5,0.4,0.0,0.5,0.5,0.4,0.8,0.7,,0.8,0.5,0.7,0.7,0.7,0.8,0.8
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.1,0.0,0.1,0.2,0.3,0.1,0.2,0.0,0.1,0.1,0.1,0.1,0.1,0.1,0.2,0.3,0.5,0.3,0.2,0.5,0.4,0.5,0.0,0.4,0.2,,0.5,0.6,0.3,0.4,0.9,1.0
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.0,0.2,0.0,0.1,0.0,0.1,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.0,0.1,0.1,0.3,0.0,0.0,0.1,0.1,0.0,0.0,0.2,0.1,0.1,0.2,0.3,0.2,0.2,0.2,0.2,0.4,0.4,0.5,0.5,0.5,0.5,,0.8,0.6,0.6,0.7,0.9
+xRFM (default),0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.1,0.1,0.0,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.0,0.0,0.1,0.1,0.2,0.1,0.2,0.0,0.1,0.2,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.2,0.1,0.1,0.3,0.2,0.2,0.2,0.3,0.2,0.1,0.2,0.4,0.2,0.3,0.4,0.2,,0.4,0.5,0.5,0.7
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.3,0.0,0.1,0.0,0.0,0.0,0.0,0.3,0.1,0.0,0.1,0.2,0.4,0.0,0.1,0.4,0.5,0.5,0.4,0.4,0.3,0.7,0.4,0.6,,0.5,0.8,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.3,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.0,0.1,0.3,0.2,0.0,0.3,0.4,0.3,0.4,0.4,0.3,0.6,0.4,0.5,0.5,,0.6,0.8
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.0,0.0,0.1,0.2,0.2,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.0,0.2,0.1,0.3,0.1,0.2,0.3,0.2,0.3,0.0,0.2,0.2,0.1,0.3,0.5,0.2,0.4,,0.8
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.2,0.0,0.1,0.3,0.0,0.2,0.2,
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_small/leaderboard_overview_explorer.html
index 639939b6a8ff134fcff018ad906f0b71b58f76ea..b8fdcae6b1cac78d234ec274f4a4594012966f90 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Classification | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_small/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_classification/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..075f434e1a846dbde94f03c079a125cb82694e2f
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_classification/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1734.0,"score":0.829,"rank":6.64,"hrank":1.47,"imp":5.137,"train_time":40.79,"predict_time":6.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+284\/-168","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1589.0,"score":0.578,"rank":12.05,"hrank":5.83,"imp":12.046,"train_time":7.56,"predict_time":0.399,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-81","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1585.0,"score":0.525,"rank":12.21,"hrank":7.85,"imp":13.452,"train_time":6.03,"predict_time":0.703,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-66","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1565.0,"score":0.543,"rank":13.2,"hrank":5.24,"imp":11.792,"train_time":2280.19,"predict_time":9.666,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-86","imputed":false},{"position":4,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1559.0,"score":0.566,"rank":13.54,"hrank":5.45,"imp":10.437,"train_time":360.81,"predict_time":5.696,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+111\/-100","imputed":false},{"position":5,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1557.0,"score":0.541,"rank":13.61,"hrank":7.35,"imp":13.379,"train_time":6.28,"predict_time":0.531,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+109\/-79","imputed":false},{"position":6,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1535.0,"score":0.528,"rank":14.82,"hrank":6.99,"imp":13.378,"train_time":3.45,"predict_time":0.182,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+113\/-96","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1532.0,"score":0.518,"rank":14.98,"hrank":8.04,"imp":14.177,"train_time":2280.19,"predict_time":0.968,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+114\/-75","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1466.0,"score":0.39,"rank":18.96,"hrank":11.3,"imp":16.35,"train_time":3487.23,"predict_time":22.857,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-69","imputed":false},{"position":9,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1456.0,"score":0.413,"rank":19.64,"hrank":7.17,"imp":15.029,"train_time":28.08,"predict_time":5.033,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-86","imputed":false},{"position":10,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1451.0,"score":0.374,"rank":19.93,"hrank":13.36,"imp":16.477,"train_time":2993.95,"predict_time":1.4,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-68","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1403.0,"score":0.347,"rank":23.34,"hrank":12.56,"imp":17.482,"train_time":3487.23,"predict_time":1.001,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-73","imputed":false},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1396.0,"score":0.339,"rank":23.88,"hrank":10.26,"imp":18.0,"train_time":4.86,"predict_time":0.17,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-81","imputed":false},{"position":13,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1377.0,"score":0.338,"rank":25.25,"hrank":10.35,"imp":17.646,"train_time":7432.95,"predict_time":404.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-70","imputed":false},{"position":14,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1374.0,"score":0.329,"rank":25.54,"hrank":11.11,"imp":18.405,"train_time":2478.69,"predict_time":3.189,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-76","imputed":false},{"position":15,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1346.0,"score":0.293,"rank":27.7,"hrank":14.39,"imp":19.142,"train_time":2478.69,"predict_time":0.447,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-81","imputed":false},{"position":16,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1346.0,"score":0.251,"rank":27.71,"hrank":14.17,"imp":17.873,"train_time":1580.61,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-51","imputed":false},{"position":17,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1345.0,"score":0.26,"rank":27.75,"hrank":18.4,"imp":20.291,"train_time":303.59,"predict_time":2.323,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-70","imputed":false},{"position":18,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1343.0,"score":0.253,"rank":27.91,"hrank":15.94,"imp":18.539,"train_time":21108.98,"predict_time":510.049,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-66","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1342.0,"score":0.217,"rank":27.96,"hrank":21.83,"imp":19.115,"train_time":774.79,"predict_time":1.455,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-50","imputed":false},{"position":20,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1336.0,"score":0.292,"rank":28.5,"hrank":14.25,"imp":18.74,"train_time":7432.95,"predict_time":39.789,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-85","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1335.0,"score":0.225,"rank":28.57,"hrank":21.46,"imp":18.447,"train_time":1580.61,"predict_time":0.381,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-63","imputed":false},{"position":22,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1308.0,"score":0.216,"rank":30.73,"hrank":19.58,"imp":20.514,"train_time":2.25,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-67","imputed":false},{"position":23,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1306.0,"score":0.225,"rank":30.86,"hrank":16.79,"imp":19.444,"train_time":6288.36,"predict_time":0.473,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-73","imputed":false},{"position":24,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1305.0,"score":0.196,"rank":30.95,"hrank":22.38,"imp":20.131,"train_time":829.5,"predict_time":1.642,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-60","imputed":false},{"position":25,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1298.0,"score":0.17,"rank":31.54,"hrank":26.6,"imp":20.069,"train_time":7.32,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-47","imputed":false},{"position":26,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1294.0,"score":0.226,"rank":31.82,"hrank":17.35,"imp":20.037,"train_time":1165.31,"predict_time":2.067,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-69","imputed":false},{"position":27,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1293.0,"score":0.186,"rank":31.96,"hrank":25.23,"imp":20.051,"train_time":774.79,"predict_time":0.282,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-58","imputed":false},{"position":28,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1285.0,"score":0.182,"rank":32.61,"hrank":14.13,"imp":20.304,"train_time":829.5,"predict_time":0.241,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-61","imputed":false},{"position":29,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1281.0,"score":0.142,"rank":32.93,"hrank":22.09,"imp":20.981,"train_time":2068.2,"predict_time":0.165,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-64","imputed":false},{"position":30,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1281.0,"score":0.215,"rank":32.95,"hrank":18.66,"imp":21.065,"train_time":9.94,"predict_time":0.367,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-83","imputed":false},{"position":31,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1274.0,"score":0.18,"rank":33.52,"hrank":25.26,"imp":21.591,"train_time":1110.83,"predict_time":0.553,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-63","imputed":false},{"position":32,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1273.0,"score":0.178,"rank":33.61,"hrank":25.19,"imp":20.051,"train_time":21108.98,"predict_time":120.297,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-61","imputed":false},{"position":33,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1271.0,"score":0.2,"rank":33.75,"hrank":22.98,"imp":20.968,"train_time":6288.36,"predict_time":9.385,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-82","imputed":false},{"position":34,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1269.0,"score":0.256,"rank":33.89,"hrank":14.42,"imp":22.144,"train_time":13.41,"predict_time":1.64,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-88","imputed":false},{"position":35,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1248.0,"score":0.142,"rank":35.64,"hrank":24.31,"imp":21.133,"train_time":14.59,"predict_time":4.033,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-75","imputed":false},{"position":36,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1246.0,"score":0.158,"rank":35.86,"hrank":20.33,"imp":22.08,"train_time":1110.83,"predict_time":0.043,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-68","imputed":false},{"position":37,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1238.0,"score":0.17,"rank":36.46,"hrank":14.51,"imp":21.152,"train_time":1165.31,"predict_time":0.105,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-84","imputed":false},{"position":38,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1237.0,"score":0.136,"rank":36.59,"hrank":15.04,"imp":22.056,"train_time":7.27,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-66","imputed":false},{"position":39,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1236.0,"score":0.105,"rank":36.68,"hrank":26.39,"imp":21.72,"train_time":2068.2,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-66","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1234.0,"score":0.149,"rank":36.82,"hrank":28.94,"imp":21.398,"train_time":2895.01,"predict_time":2.42,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-77","imputed":false},{"position":41,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1230.0,"score":0.187,"rank":37.18,"hrank":19.54,"imp":21.421,"train_time":51.79,"predict_time":39.921,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-95","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1197.0,"score":0.108,"rank":39.89,"hrank":32.22,"imp":23.809,"train_time":4.88,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-72","imputed":false},{"position":43,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1190.0,"score":0.133,"rank":40.52,"hrank":28.18,"imp":23.694,"train_time":1425.59,"predict_time":7.352,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-80","imputed":false},{"position":44,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1187.0,"score":0.134,"rank":40.73,"hrank":28.25,"imp":23.306,"train_time":2895.01,"predict_time":0.145,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-74","imputed":false},{"position":45,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1177.0,"score":0.125,"rank":41.61,"hrank":26.19,"imp":23.56,"train_time":669.3,"predict_time":1.187,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-85","imputed":false},{"position":46,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1172.0,"score":0.109,"rank":42.0,"hrank":14.18,"imp":25.0,"train_time":14.22,"predict_time":0.352,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-90","imputed":false},{"position":47,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1151.0,"score":0.098,"rank":43.7,"hrank":26.74,"imp":24.874,"train_time":669.3,"predict_time":0.166,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-75","imputed":false},{"position":48,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1141.0,"score":0.059,"rank":44.5,"hrank":37.91,"imp":23.707,"train_time":2.85,"predict_time":0.19,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-65","imputed":false},{"position":49,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1134.0,"score":0.099,"rank":45.07,"hrank":29.14,"imp":25.064,"train_time":1425.59,"predict_time":0.561,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-91","imputed":false},{"position":50,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1134.0,"score":0.089,"rank":45.09,"hrank":36.06,"imp":24.002,"train_time":3.27,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-80","imputed":false},{"position":51,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1131.0,"score":0.165,"rank":45.29,"hrank":14.59,"imp":25.856,"train_time":45012.0,"predict_time":5.273,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-118","imputed":false},{"position":52,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1130.0,"score":0.108,"rank":45.39,"hrank":34.71,"imp":25.052,"train_time":562.08,"predict_time":1.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-92","imputed":false},{"position":53,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1122.0,"score":0.151,"rank":46.02,"hrank":16.16,"imp":26.122,"train_time":45012.0,"predict_time":21.34,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+111\/-121","imputed":false},{"position":54,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1098.0,"score":0.071,"rank":47.88,"hrank":24.79,"imp":26.13,"train_time":454.83,"predict_time":146.788,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-92","imputed":false},{"position":55,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1093.0,"score":0.054,"rank":48.25,"hrank":40.7,"imp":26.485,"train_time":15.58,"predict_time":2.601,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-77","imputed":false},{"position":56,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1090.0,"score":0.053,"rank":48.45,"hrank":41.07,"imp":25.936,"train_time":562.08,"predict_time":0.138,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-86","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1089.0,"score":0.102,"rank":48.57,"hrank":36.9,"imp":31.844,"train_time":399.28,"predict_time":0.696,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-105","imputed":false},{"position":58,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1039.0,"score":0.119,"rank":52.21,"hrank":14.11,"imp":32.261,"train_time":461.86,"predict_time":5.723,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-157","imputed":false},{"position":59,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1038.0,"score":0.056,"rank":52.27,"hrank":46.59,"imp":33.437,"train_time":399.28,"predict_time":0.184,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-97","imputed":false},{"position":60,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1019.0,"score":0.03,"rank":53.61,"hrank":47.45,"imp":28.612,"train_time":9.06,"predict_time":0.195,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-81","imputed":false},{"position":61,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1006.0,"score":0.049,"rank":54.46,"hrank":48.08,"imp":29.05,"train_time":5.43,"predict_time":0.835,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-120","imputed":false},{"position":62,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1006.0,"score":0.046,"rank":54.5,"hrank":44.24,"imp":29.998,"train_time":3.76,"predict_time":0.597,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-116","imputed":false},{"position":63,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1006.0,"score":0.057,"rank":54.5,"hrank":44.37,"imp":30.79,"train_time":2.45,"predict_time":0.519,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-108","imputed":false},{"position":64,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.022,"rank":54.88,"hrank":43.13,"imp":31.122,"train_time":0.46,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-106","imputed":false},{"position":65,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":987.0,"score":0.03,"rank":55.71,"hrank":51.3,"imp":28.66,"train_time":5.03,"predict_time":0.601,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-86","imputed":false},{"position":66,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":982.0,"score":0.041,"rank":56.02,"hrank":46.2,"imp":31.102,"train_time":123.01,"predict_time":1.578,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-114","imputed":false},{"position":67,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":972.0,"score":0.043,"rank":56.62,"hrank":28.07,"imp":38.096,"train_time":48.1,"predict_time":0.05,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-127","imputed":false},{"position":68,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":947.0,"score":0.036,"rank":58.11,"hrank":19.17,"imp":34.573,"train_time":479.33,"predict_time":0.495,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-103","imputed":false},{"position":69,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":926.0,"score":0.03,"rank":59.3,"hrank":29.47,"imp":35.157,"train_time":479.33,"predict_time":0.098,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-110","imputed":false},{"position":70,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":883.0,"score":0.011,"rank":61.5,"hrank":56.63,"imp":38.035,"train_time":2.34,"predict_time":0.132,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-120","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":830.0,"score":0.024,"rank":63.86,"hrank":58.87,"imp":37.928,"train_time":123.01,"predict_time":0.092,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-130","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":586.0,"score":0.0,"rank":70.46,"hrank":70.1,"imp":51.377,"train_time":0.28,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-148","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_small/pareto_front_explorer.html
index d80bead34be40dc4a7c2157daf7cd599436a2edf..5ebe0e6f576c7866be0e6856db0547fa9e60c940 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3228316175,"imp":20.2906368273,"elo":1345.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404950151,"imp":20.0687849906,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0468267403,"imp":17.872545046,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806426698,"imp":18.4469806196,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0464942884,"imp":23.8092242982,"elo":1197.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0428145039,"imp":22.0803629798,"elo":1245.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5527550553,"imp":21.5909524337,"elo":1273.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0218718174,"imp":22.0559100078,"elo":1237.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0200835992,"imp":21.7203639519,"elo":1235.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1652902771,"imp":20.9805677059,"elo":1281.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5968426684,"imp":29.9976798872,"elo":1005.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1664465955,"imp":24.874400389,"elo":1150.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1866680852,"imp":23.559586256,"elo":1176.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6011123119,"imp":28.6601334836,"elo":986.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5609692074,"imp":25.0644923868,"elo":1133.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.3522009809,"imp":23.6938758937,"elo":1189.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1925833078,"imp":24.0017035359,"elo":1133.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.2819017032,"imp":20.0511344827,"elo":1292.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.4553717527,"imp":19.1154882491,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.0331218603,"imp":15.0286266136,"elo":1455.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"imp":24.9997535627,"elo":1171.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4731015689,"imp":19.4439252313,"elo":1306.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"imp":20.9676515594,"elo":1271.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6014690178,"imp":26.4848908562,"elo":1093.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0497782518,"imp":38.0963839727,"elo":972.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1838950332,"imp":33.4374739974,"elo":1038.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6959068701,"imp":31.8435762442,"elo":1088.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0698340289,"imp":31.1217672397,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.138131658,"imp":25.9357248125,"elo":1090.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"imp":25.0516886473,"elo":1129.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.0326517405,"imp":21.133213507,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.0010951073,"imp":17.4815024128,"elo":1402.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":22.857077649,"imp":16.3497157725,"elo":1465.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7026957462,"imp":13.4520984943,"elo":1585.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9678506248,"imp":14.1774120675,"elo":1531.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.6655104794,"imp":11.7918188963,"elo":1565.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6402183396,"imp":22.1440835649,"elo":1269.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"imp":21.4211760617,"elo":1229.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.7894649716,"imp":18.7395472663,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":404.0529881985,"imp":17.6459508145,"elo":1377.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1696165976,"imp":17.9998827274,"elo":1395.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.3100890028,"imp":5.136706293,"elo":1733.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5190992166,"imp":30.7901539504,"elo":1005.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"imp":13.3783444172,"elo":1534.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.3671853745,"imp":21.0649567851,"elo":1280.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4472563167,"imp":19.1423347694,"elo":1345.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.1888600496,"imp":18.4045985287,"elo":1373.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"imp":13.3793598531,"elo":1557.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3985329866,"imp":12.0462824785,"elo":1588.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"imp":32.2613815409,"elo":1039.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.2730552405,"imp":25.8564274129,"elo":1131.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.3395164105,"imp":26.1217441443,"elo":1121.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0547712076,"imp":20.5142072068,"elo":1307.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.195480498,"imp":28.6122314625,"elo":1019.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.144904019,"imp":23.3060088555,"elo":1187.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.419707216,"imp":21.3982790118,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1903629661,"imp":23.7067154788,"elo":1140.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2410402043,"imp":20.3040818509,"elo":1284.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.641714504,"imp":20.1312920436,"elo":1305.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":146.7883490527,"imp":26.1295026509,"elo":1098.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":120.2974149859,"imp":20.0506547807,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":510.0490782849,"imp":18.5390452779,"elo":1343.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.8352682059,"imp":29.0495337678,"elo":1006.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1053014255,"imp":21.1519647015,"elo":1238.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0671352109,"imp":20.036628317,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3228316175,"x_train":303.5874437498,"imp":20.2906368273,"elo":1345.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404950151,"x_train":7.3166111387,"imp":20.0687849906,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0468267403,"x_train":1580.6128837326,"imp":17.872545046,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806426698,"x_train":1580.6128837326,"imp":18.4469806196,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0464942884,"x_train":4.8835669578,"imp":23.8092242982,"elo":1197.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0428145039,"x_train":1110.8290906413,"imp":22.0803629798,"elo":1245.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5527550553,"x_train":1110.8290906413,"imp":21.5909524337,"elo":1273.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0218718174,"x_train":7.2693312731,"imp":22.0559100078,"elo":1237.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0200835992,"x_train":2068.2009167512,"imp":21.7203639519,"elo":1235.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1652902771,"x_train":2068.2009167512,"imp":20.9805677059,"elo":1281.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5968426684,"x_train":3.7608843708,"imp":29.9976798872,"elo":1005.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1664465955,"x_train":669.3008326264,"imp":24.874400389,"elo":1150.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1866680852,"x_train":669.3008326264,"imp":23.559586256,"elo":1176.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6011123119,"x_train":5.0325246315,"imp":28.6601334836,"elo":986.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5609692074,"x_train":1425.5943698968,"imp":25.0644923868,"elo":1133.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.3522009809,"x_train":1425.5943698968,"imp":23.6938758937,"elo":1189.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1925833078,"x_train":3.2714234862,"imp":24.0017035359,"elo":1133.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.2819017032,"x_train":774.7922679504,"imp":20.0511344827,"elo":1292.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.4553717527,"x_train":774.7922679504,"imp":19.1154882491,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.0331218603,"x_train":28.0828822259,"imp":15.0286266136,"elo":1455.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"x_train":14.219089197,"imp":24.9997535627,"elo":1171.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4731015689,"x_train":6288.3552384481,"imp":19.4439252313,"elo":1306.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"x_train":6288.3552384481,"imp":20.9676515594,"elo":1271.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6014690178,"x_train":15.5816092874,"imp":26.4848908562,"elo":1093.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0497782518,"x_train":48.1038174593,"imp":38.0963839727,"elo":972.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1838950332,"x_train":399.2771714041,"imp":33.4374739974,"elo":1038.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6959068701,"x_train":399.2771714041,"imp":31.8435762442,"elo":1088.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0698340289,"x_train":0.4597788004,"imp":31.1217672397,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.138131658,"x_train":562.0823628399,"imp":25.9357248125,"elo":1090.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"x_train":562.0823628399,"imp":25.0516886473,"elo":1129.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.0326517405,"x_train":14.5881086057,"imp":21.133213507,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.0010951073,"x_train":3487.2314440046,"imp":17.4815024128,"elo":1402.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":22.857077649,"x_train":3487.2314440046,"imp":16.3497157725,"elo":1465.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7026957462,"x_train":6.0254254253,"imp":13.4520984943,"elo":1585.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9678506248,"x_train":2280.1896584743,"imp":14.1774120675,"elo":1531.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.6655104794,"x_train":2280.1896584743,"imp":11.7918188963,"elo":1565.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6402183396,"x_train":13.4089193675,"imp":22.1440835649,"elo":1269.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"x_train":51.7864862661,"imp":21.4211760617,"elo":1229.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.7894649716,"x_train":7432.9537119468,"imp":18.7395472663,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":404.0529881985,"x_train":7432.9537119468,"imp":17.6459508145,"elo":1377.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1696165976,"x_train":4.8617027155,"imp":17.9998827274,"elo":1395.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.3100890028,"x_train":40.7899647401,"imp":5.136706293,"elo":1733.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5190992166,"x_train":2.4480622255,"imp":30.7901539504,"elo":1005.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"x_train":3.4521869623,"imp":13.3783444172,"elo":1534.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.3671853745,"x_train":9.93671076,"imp":21.0649567851,"elo":1280.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4472563167,"x_train":2478.69202837,"imp":19.1423347694,"elo":1345.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.1888600496,"x_train":2478.69202837,"imp":18.4045985287,"elo":1373.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"x_train":6.2811404231,"imp":13.3793598531,"elo":1557.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3985329866,"x_train":7.5625787904,"imp":12.0462824785,"elo":1588.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"x_train":461.8626497906,"imp":32.2613815409,"elo":1039.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.2730552405,"x_train":45012.0033435545,"imp":25.8564274129,"elo":1131.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.3395164105,"x_train":45012.0033435545,"imp":26.1217441443,"elo":1121.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0547712076,"x_train":2.2471782685,"imp":20.5142072068,"elo":1307.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.195480498,"x_train":9.0610314235,"imp":28.6122314625,"elo":1019.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.144904019,"x_train":2895.0057162096,"imp":23.3060088555,"elo":1187.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.419707216,"x_train":2895.0057162096,"imp":21.3982790118,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1903629661,"x_train":2.8478128159,"imp":23.7067154788,"elo":1140.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2410402043,"x_train":829.4997787901,"imp":20.3040818509,"elo":1284.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.641714504,"x_train":829.4997787901,"imp":20.1312920436,"elo":1305.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":146.7883490527,"x_train":454.8291985924,"imp":26.1295026509,"elo":1098.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":120.2974149859,"x_train":21108.9759033856,"imp":20.0506547807,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":510.0490782849,"x_train":21108.9759033856,"imp":18.5390452779,"elo":1343.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.8352682059,"x_train":5.4303046313,"imp":29.0495337678,"elo":1006.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1053014255,"x_train":1165.3088146236,"imp":21.1519647015,"elo":1238.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0671352109,"x_train":1165.3088146236,"imp":20.036628317,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_small/tuning_trajectories_explorer.html
index 638519a70ed8eea7bdd23c0c5cb30afe967427e4..7365683bbda6304defb02c9157db7e47dc9a0532 100644
--- a/data/imputation_no/splits_lite/tasks_classification/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":303.5874437498,"imp":20.3042956447,"elo":1359.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":7.3166111387,"imp":20.0821525173,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":13.0021013751,"imp":19.3468160054,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":34.9838507458,"imp":19.0133615022,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":71.4491054101,"imp":18.8191390863,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":197.3421561911,"imp":18.7043692845,"elo":1350.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":391.4036568625,"imp":18.7255763113,"elo":1346.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":788.0795112228,"imp":18.5377873297,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1179.7013801715,"imp":18.5417820604,"elo":1336.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1580.6128837326,"imp":18.4470805495,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.8835669578,"imp":23.8216784046,"elo":1201.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.414093577,"imp":23.1078178595,"elo":1236.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":25.923774854,"imp":22.6291646853,"elo":1256.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":52.8409015956,"imp":22.3836041598,"elo":1265.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":137.1347812527,"imp":21.9554561218,"elo":1280.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":276.1566202862,"imp":21.7497073303,"elo":1288.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":553.6509294091,"imp":21.6598481983,"elo":1288.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":829.2384033907,"imp":21.5906972054,"elo":1290.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1110.8290906413,"imp":21.5984418197,"elo":1287.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.2693312731,"imp":22.0684118345,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":17.3786155791,"imp":21.8369200314,"elo":1249.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":52.8479572348,"imp":21.5612972716,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":105.3309657113,"imp":21.4871454526,"elo":1266.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":259.310462097,"imp":21.3832912303,"elo":1274.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":524.5599999251,"imp":21.2088277446,"elo":1282.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1025.3924289592,"imp":21.0471472755,"elo":1290.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1544.9631429961,"imp":21.0172600715,"elo":1294.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2068.2009167512,"imp":20.9845312045,"elo":1294.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.7608843708,"imp":30.0107777698,"elo":1012.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.2336179369,"imp":28.1242892096,"elo":1040.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":17.356104837,"imp":25.437346723,"elo":1095.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":34.0850647785,"imp":24.7459414465,"elo":1126.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":84.0342963818,"imp":24.0657609858,"elo":1158.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":167.1827755321,"imp":23.8820181891,"elo":1164.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":333.2449706903,"imp":23.7762276103,"elo":1169.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":499.1329867675,"imp":23.6189175055,"elo":1181.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":669.3008326264,"imp":23.6243422358,"elo":1182.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.0325246315,"imp":28.6713087291,"elo":990.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":12.7079451787,"imp":27.7792390233,"elo":1038.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":35.5292055013,"imp":25.859970438,"elo":1109.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":67.8432178902,"imp":25.0597808489,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":172.6394168737,"imp":24.1336037474,"elo":1164.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":350.4018274029,"imp":23.745621365,"elo":1178.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":703.8401639882,"imp":23.6733674294,"elo":1184.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1060.0265172973,"imp":23.6221799835,"elo":1186.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1425.5943698968,"imp":23.7192456368,"elo":1195.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":3.2714234862,"imp":24.0142822634,"elo":1130.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":7.187975863,"imp":21.6574161499,"elo":1239.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":18.491534641,"imp":20.3035481899,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":37.6314766741,"imp":19.7882034329,"elo":1331.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":95.7218520577,"imp":19.3953635976,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":192.3000546929,"imp":19.3022191987,"elo":1348.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":385.4240248592,"imp":19.20260164,"elo":1354.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":577.4328596419,"imp":19.1088430637,"elo":1356.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":774.7922679504,"imp":19.1245444734,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":28.0828822259,"imp":15.2603584954,"elo":1459.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.219089197,"imp":25.0142074156,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":51.0552417446,"imp":23.4409579435,"elo":1194.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":152.0656634461,"imp":20.8237438782,"elo":1282.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":301.2283204956,"imp":20.2612285435,"elo":1305.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":742.3216508123,"imp":19.7078004873,"elo":1327.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1548.7724245944,"imp":20.0460516462,"elo":1309.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3115.4832472775,"imp":20.6738551406,"elo":1287.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4642.1963843119,"imp":20.9082054525,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6288.3552384481,"imp":20.9979837986,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":15.5816092874,"imp":26.4954275178,"elo":1101.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":48.1038174593,"imp":38.1478967306,"elo":975.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":125.2282972555,"imp":34.9078132902,"elo":1027.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":333.0389251457,"imp":31.9659175523,"elo":1090.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":399.2771714041,"imp":31.8396265048,"elo":1094.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4597788004,"imp":31.1508694831,"elo":998.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.1539887452,"imp":28.4356723364,"elo":1034.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":11.7755523341,"imp":26.7432785165,"elo":1072.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":25.7323507846,"imp":25.8483865997,"elo":1090.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":67.6419649919,"imp":25.1488062952,"elo":1124.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":137.8532509555,"imp":25.0243078796,"elo":1132.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":278.1383106963,"imp":25.0702053353,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":418.6187333473,"imp":25.1012800261,"elo":1134.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":562.0823628399,"imp":25.0648591447,"elo":1133.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":14.5881086057,"imp":21.1459604278,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":30.5807961877,"imp":18.7912752594,"elo":1353.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":79.1587952873,"imp":17.2850164129,"elo":1428.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":166.3929204249,"imp":16.8224640189,"elo":1448.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":428.3936464406,"imp":16.3219303462,"elo":1474.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":868.2323938389,"imp":16.099425558,"elo":1484.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1742.3462491724,"imp":16.1271394328,"elo":1485.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2600.7236668564,"imp":16.2826836054,"elo":1488.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3487.2314440046,"imp":16.3679098284,"elo":1488.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.0254254253,"imp":13.4660546229,"elo":1601.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":19.6925140518,"imp":13.5296928443,"elo":1586.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":54.027252147,"imp":13.4956083177,"elo":1573.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":109.1627008491,"imp":13.5840731597,"elo":1577.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":282.2546622885,"imp":13.5449545727,"elo":1572.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":564.5394139879,"imp":13.5447724329,"elo":1563.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1133.3178139172,"imp":12.9868662878,"elo":1571.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1704.1213724065,"imp":12.6208741023,"elo":1577.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2280.1896584743,"imp":11.8408050328,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.4089193675,"imp":22.1567633769,"elo":1280.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.7864862661,"imp":21.4338616894,"elo":1233.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":88.6021525086,"imp":20.5333587593,"elo":1274.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":190.7023074438,"imp":19.6236152699,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":357.7608890189,"imp":19.2768385648,"elo":1337.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":906.5893386893,"imp":18.6107750914,"elo":1353.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1833.1395916312,"imp":18.1027623705,"elo":1370.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3679.7001768605,"imp":17.906035458,"elo":1379.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":5552.0614127314,"imp":17.8133440384,"elo":1384.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":7432.9537119468,"imp":17.6937440564,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":4.8617027155,"imp":18.013702049,"elo":1406.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":40.7899647401,"imp":5.1509374015,"elo":1737.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":2.4480622255,"imp":30.8016592065,"elo":1016.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":3.4521869623,"imp":13.3923338639,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.93671076,"imp":21.0781206905,"elo":1291.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":21.7784528479,"imp":20.7466729114,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":59.121616311,"imp":20.2195919195,"elo":1320.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":120.4004748358,"imp":19.7052767554,"elo":1347.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":303.5800929631,"imp":19.3081046461,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":609.3448213855,"imp":19.0043025925,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1226.0184259582,"imp":18.6765963121,"elo":1387.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1845.3925750472,"imp":18.5458651166,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2478.69202837,"imp":18.4183258422,"elo":1392.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":6.2811404231,"imp":13.3937123621,"elo":1578.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":7.5625787904,"imp":12.0601203605,"elo":1602.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":461.8626497906,"imp":32.2711080542,"elo":1044.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":812.6941385093,"imp":32.1119007608,"elo":1051.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1430.8811369497,"imp":31.6725135878,"elo":1057.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2384.5896366526,"imp":30.7653945699,"elo":1068.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5535.519999844,"imp":28.9577424898,"elo":1089.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11245.833348223,"imp":27.4426916085,"elo":1103.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23102.6961039552,"imp":26.8504294278,"elo":1113.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34138.663197007,"imp":26.5341500537,"elo":1122.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":45012.0033435545,"imp":26.1441210026,"elo":1135.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":2.2471782685,"imp":20.5267294877,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.0610314235,"imp":28.6174380335,"elo":1021.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":23.1218764356,"imp":25.6521197756,"elo":1097.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":66.1742587244,"imp":23.4070164547,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":145.1501764392,"imp":22.6192598341,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":352.024745135,"imp":22.0626335237,"elo":1227.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":719.83310274,"imp":21.7657179311,"elo":1237.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1418.0597495196,"imp":21.6715235145,"elo":1232.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2155.6849806435,"imp":21.4947277849,"elo":1241.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2895.0057162096,"imp":21.4080185463,"elo":1241.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.8478128159,"imp":23.7189855039,"elo":1140.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":7.2765847065,"imp":21.6360694679,"elo":1236.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":20.3573867716,"imp":21.1028025901,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":39.9575239687,"imp":20.7587253819,"elo":1285.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":102.7267296917,"imp":20.4882062357,"elo":1296.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":205.8681515072,"imp":20.3943540394,"elo":1305.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":411.951005552,"imp":20.2796609536,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":617.2039633301,"imp":20.1704667438,"elo":1320.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":829.4997787901,"imp":20.1227153051,"elo":1320.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":454.8291985924,"imp":26.1425197328,"elo":1103.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1266.5525763884,"imp":23.4742516639,"elo":1174.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":3732.2979901517,"imp":20.542936141,"elo":1267.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":8203.1308672787,"imp":19.509591974,"elo":1308.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":20166.4238376739,"imp":18.5733666583,"elo":1354.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":21108.9759033856,"imp":18.5735507004,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":5.4303046313,"imp":29.0599353481,"elo":1012.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":11.776121563,"imp":26.32033861,"elo":1079.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":28.7928271844,"imp":22.6994092601,"elo":1202.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":57.3403812154,"imp":21.1816204709,"elo":1267.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":145.2967129053,"imp":20.5260498304,"elo":1293.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":291.5311600969,"imp":20.2117659176,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":582.7415169755,"imp":20.0792359521,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":872.0774356109,"imp":20.0767903242,"elo":1306.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1165.3088146236,"imp":20.0591589805,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3228316175,"x_train":303.5874437498,"imp":20.3042956447,"elo":1359.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404950151,"x_train":7.3166111387,"imp":20.0821525173,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0764260122,"x_train":13.0021013751,"imp":19.3468160054,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1646517485,"x_train":34.9838507458,"imp":19.0133615022,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2059671983,"x_train":71.4491054101,"imp":18.8191390863,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2459997699,"x_train":197.3421561911,"imp":18.7043692845,"elo":1350.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2734896112,"x_train":391.4036568625,"imp":18.7255763113,"elo":1346.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3310336248,"x_train":788.0795112228,"imp":18.5377873297,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3663053836,"x_train":1179.7013801715,"imp":18.5417820604,"elo":1336.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806426698,"x_train":1580.6128837326,"imp":18.4470805495,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0464942884,"x_train":4.8835669578,"imp":23.8216784046,"elo":1201.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0848784521,"x_train":10.414093577,"imp":23.1078178595,"elo":1236.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1691167697,"x_train":25.923774854,"imp":22.6291646853,"elo":1256.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2269108909,"x_train":52.8409015956,"imp":22.3836041598,"elo":1265.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3076887709,"x_train":137.1347812527,"imp":21.9554561218,"elo":1280.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3654650581,"x_train":276.1566202862,"imp":21.7497073303,"elo":1288.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.462284001,"x_train":553.6509294091,"imp":21.6598481983,"elo":1288.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4730067641,"x_train":829.2384033907,"imp":21.5906972054,"elo":1290.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5527550553,"x_train":1110.8290906413,"imp":21.5984418197,"elo":1287.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0218718174,"x_train":7.2693312731,"imp":22.0684118345,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0508456884,"x_train":17.3786155791,"imp":21.8369200314,"elo":1249.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0739410125,"x_train":52.8479572348,"imp":21.5612972716,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0914252829,"x_train":105.3309657113,"imp":21.4871454526,"elo":1266.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1248677418,"x_train":259.310462097,"imp":21.3832912303,"elo":1274.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1350318723,"x_train":524.5599999251,"imp":21.2088277446,"elo":1282.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1560473551,"x_train":1025.3924289592,"imp":21.0471472755,"elo":1290.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1586418791,"x_train":1544.9631429961,"imp":21.0172600715,"elo":1294.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1652902771,"x_train":2068.2009167512,"imp":20.9845312045,"elo":1294.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5968426684,"x_train":3.7608843708,"imp":30.0107777698,"elo":1012.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6853995207,"x_train":7.2336179369,"imp":28.1242892096,"elo":1040.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6085460645,"x_train":17.356104837,"imp":25.437346723,"elo":1095.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6559709961,"x_train":34.0850647785,"imp":24.7459414465,"elo":1126.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9058839094,"x_train":84.0342963818,"imp":24.0657609858,"elo":1158.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.196621892,"x_train":167.1827755321,"imp":23.8820181891,"elo":1164.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4547771678,"x_train":333.2449706903,"imp":23.7762276103,"elo":1169.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5815845677,"x_train":499.1329867675,"imp":23.6189175055,"elo":1181.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1866680852,"x_train":669.3008326264,"imp":23.6243422358,"elo":1182.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6011123119,"x_train":5.0325246315,"imp":28.6713087291,"elo":990.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4502158609,"x_train":12.7079451787,"imp":27.7792390233,"elo":1038.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7078527827,"x_train":35.5292055013,"imp":25.859970438,"elo":1109.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.3184880832,"x_train":67.8432178902,"imp":25.0597808489,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.5303897769,"x_train":172.6394168737,"imp":24.1336037474,"elo":1164.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.7545080726,"x_train":350.4018274029,"imp":23.745621365,"elo":1178.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.1854207053,"x_train":703.8401639882,"imp":23.6733674294,"elo":1184.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.6864231297,"x_train":1060.0265172973,"imp":23.6221799835,"elo":1186.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.3010312699,"x_train":1425.5943698968,"imp":23.7192456368,"elo":1195.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1925833078,"x_train":3.2714234862,"imp":24.0142822634,"elo":1130.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6370937782,"x_train":7.187975863,"imp":21.6574161499,"elo":1239.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8585219059,"x_train":18.491534641,"imp":20.3035481899,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0687137304,"x_train":37.6314766741,"imp":19.7882034329,"elo":1331.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3923655359,"x_train":95.7218520577,"imp":19.3953635976,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.480046567,"x_train":192.3000546929,"imp":19.3022191987,"elo":1348.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6191358781,"x_train":385.4240248592,"imp":19.20260164,"elo":1354.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.8652145084,"x_train":577.4328596419,"imp":19.1088430637,"elo":1356.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4553717527,"x_train":774.7922679504,"imp":19.1245444734,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":5.0331218603,"x_train":28.0828822259,"imp":15.2603584954,"elo":1459.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"x_train":14.219089197,"imp":25.0142074156,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9081956315,"x_train":51.0552417446,"imp":23.4409579435,"elo":1194.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8094695931,"x_train":152.0656634461,"imp":20.8237438782,"elo":1282.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9001397479,"x_train":301.2283204956,"imp":20.2612285435,"elo":1305.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.3070047996,"x_train":742.3216508123,"imp":19.7078004873,"elo":1327.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5713626606,"x_train":1548.7724245944,"imp":20.0460516462,"elo":1309.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.0463659943,"x_train":3115.4832472775,"imp":20.6738551406,"elo":1287.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4271738524,"x_train":4642.1963843119,"imp":20.9082054525,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"x_train":6288.3552384481,"imp":20.9979837986,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6014690178,"x_train":15.5816092874,"imp":26.4954275178,"elo":1101.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0497782518,"x_train":48.1038174593,"imp":38.1478967306,"elo":975.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4010049235,"x_train":125.2282972555,"imp":34.9078132902,"elo":1027.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7337631815,"x_train":333.0389251457,"imp":31.9659175523,"elo":1090.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6959068701,"x_train":399.2771714041,"imp":31.8396265048,"elo":1094.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0698340289,"x_train":0.4597788004,"imp":31.1508694831,"elo":998.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2614383057,"x_train":3.1539887452,"imp":28.4356723364,"elo":1034.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.45693222,"x_train":11.7755523341,"imp":26.7432785165,"elo":1072.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5950629764,"x_train":25.7323507846,"imp":25.8483865997,"elo":1090.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7642726368,"x_train":67.6419649919,"imp":25.1488062952,"elo":1124.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9217934517,"x_train":137.8532509555,"imp":25.0243078796,"elo":1132.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0817312972,"x_train":278.1383106963,"imp":25.0702053353,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2223316361,"x_train":418.6187333473,"imp":25.1012800261,"elo":1134.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"x_train":562.0823628399,"imp":25.0648591447,"elo":1133.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.0326517405,"x_train":14.5881086057,"imp":21.1459604278,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.5458694726,"x_train":30.5807961877,"imp":18.7912752594,"elo":1353.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.0822974518,"x_train":79.1587952873,"imp":17.2850164129,"elo":1428.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.3031022636,"x_train":166.3929204249,"imp":16.8224640189,"elo":1448.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.3882183239,"x_train":428.3936464406,"imp":16.3219303462,"elo":1474.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4616303712,"x_train":868.2323938389,"imp":16.099425558,"elo":1484.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.6029747084,"x_train":1742.3462491724,"imp":16.1271394328,"elo":1485.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.4107968916,"x_train":2600.7236668564,"imp":16.2826836054,"elo":1488.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":22.857077649,"x_train":3487.2314440046,"imp":16.3679098284,"elo":1488.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7026957462,"x_train":6.0254254253,"imp":13.4660546229,"elo":1601.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.32293962,"x_train":19.6925140518,"imp":13.5296928443,"elo":1586.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8631014568,"x_train":54.027252147,"imp":13.4956083177,"elo":1573.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.9894747323,"x_train":109.1627008491,"imp":13.5840731597,"elo":1577.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.5217042751,"x_train":282.2546622885,"imp":13.5449545727,"elo":1572.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.0109718207,"x_train":564.5394139879,"imp":13.5447724329,"elo":1563.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.1730834001,"x_train":1133.3178139172,"imp":12.9868662878,"elo":1571.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0817714469,"x_train":1704.1213724065,"imp":12.6208741023,"elo":1577.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.6282078215,"x_train":2280.1896584743,"imp":11.8408050328,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6402183396,"x_train":13.4089193675,"imp":22.1567633769,"elo":1280.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"x_train":51.7864862661,"imp":21.4338616894,"elo":1233.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":59.2095298832,"x_train":88.6021525086,"imp":20.5333587593,"elo":1274.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":91.3130194728,"x_train":190.7023074438,"imp":19.6236152699,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":128.7090086601,"x_train":357.7608890189,"imp":19.2768385648,"elo":1337.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":197.3779151734,"x_train":906.5893386893,"imp":18.6107750914,"elo":1353.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":248.1558123661,"x_train":1833.1395916312,"imp":18.1027623705,"elo":1370.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":314.8436401055,"x_train":3679.7001768605,"imp":17.906035458,"elo":1379.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":344.2915338009,"x_train":5552.0614127314,"imp":17.8133440384,"elo":1384.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":400.899764047,"x_train":7432.9537119468,"imp":17.6937440564,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1696165976,"x_train":4.8617027155,"imp":18.013702049,"elo":1406.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.3100890028,"x_train":40.7899647401,"imp":5.1509374015,"elo":1737.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5190992166,"x_train":2.4480622255,"imp":30.8016592065,"elo":1016.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"x_train":3.4521869623,"imp":13.3923338639,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.3671853745,"x_train":9.93671076,"imp":21.0781206905,"elo":1291.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.7792451672,"x_train":21.7784528479,"imp":20.7466729114,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.3835716074,"x_train":59.121616311,"imp":20.2195919195,"elo":1320.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.7666598519,"x_train":120.4004748358,"imp":19.7052767554,"elo":1347.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.2463538995,"x_train":303.5800929631,"imp":19.3081046461,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":2.282918805,"x_train":609.3448213855,"imp":19.0043025925,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":2.6967059891,"x_train":1226.0184259582,"imp":18.6765963121,"elo":1387.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.9081786218,"x_train":1845.3925750472,"imp":18.5458651166,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":3.1888600496,"x_train":2478.69202837,"imp":18.4183258422,"elo":1392.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"x_train":6.2811404231,"imp":13.3937123621,"elo":1578.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3985329866,"x_train":7.5625787904,"imp":12.0601203605,"elo":1602.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"x_train":461.8626497906,"imp":32.2711080542,"elo":1044.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1148611269,"x_train":812.6941385093,"imp":32.1119007608,"elo":1051.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3665562879,"x_train":1430.8811369497,"imp":31.6725135878,"elo":1057.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.8811964581,"x_train":2384.5896366526,"imp":30.7653945699,"elo":1068.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.7428459792,"x_train":5535.519999844,"imp":28.9577424898,"elo":1089.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.4043789983,"x_train":11245.833348223,"imp":27.4426916085,"elo":1103.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":23.5017727083,"x_train":23102.6961039552,"imp":26.8504294278,"elo":1113.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":24.5387683339,"x_train":34138.663197007,"imp":26.5341500537,"elo":1122.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.3395164105,"x_train":45012.0033435545,"imp":26.1441210026,"elo":1135.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0547712076,"x_train":2.2471782685,"imp":20.5267294877,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.195480498,"x_train":9.0610314235,"imp":28.6174380335,"elo":1021.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.385980123,"x_train":23.1218764356,"imp":25.6521197756,"elo":1097.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7451473275,"x_train":66.1742587244,"imp":23.4070164547,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0088740142,"x_train":145.1501764392,"imp":22.6192598341,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2591398849,"x_train":352.024745135,"imp":22.0626335237,"elo":1227.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6135463766,"x_train":719.83310274,"imp":21.7657179311,"elo":1237.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0639002358,"x_train":1418.0597495196,"imp":21.6715235145,"elo":1232.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1878184457,"x_train":2155.6849806435,"imp":21.4947277849,"elo":1241.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.419707216,"x_train":2895.0057162096,"imp":21.4080185463,"elo":1241.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1903629661,"x_train":2.8478128159,"imp":23.7189855039,"elo":1140.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.664697711,"x_train":7.2765847065,"imp":21.6360694679,"elo":1236.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.24520154,"x_train":20.3573867716,"imp":21.1028025901,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4521450072,"x_train":39.9575239687,"imp":20.7587253819,"elo":1285.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8858828782,"x_train":102.7267296917,"imp":20.4882062357,"elo":1296.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.9134288142,"x_train":205.8681515072,"imp":20.3943540394,"elo":1305.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0984566815,"x_train":411.951005552,"imp":20.2796609536,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8473752296,"x_train":617.2039633301,"imp":20.1704667438,"elo":1320.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.503647059,"x_train":829.4997787901,"imp":20.1227153051,"elo":1320.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":146.7883490527,"x_train":454.8291985924,"imp":26.1425197328,"elo":1103.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":268.9452732023,"x_train":1266.5525763884,"imp":23.4742516639,"elo":1174.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":453.5002018107,"x_train":3732.2979901517,"imp":20.542936141,"elo":1267.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":588.0968491642,"x_train":8203.1308672787,"imp":19.509591974,"elo":1308.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":491.6490605571,"x_train":20166.4238376739,"imp":18.5733666583,"elo":1354.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":510.0490782849,"x_train":21108.9759033856,"imp":18.5735507004,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.8352682059,"x_train":5.4303046313,"imp":29.0599353481,"elo":1012.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.0296880071,"x_train":11.776121563,"imp":26.32033861,"elo":1079.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.2425439486,"x_train":28.7928271844,"imp":22.6994092601,"elo":1202.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.5881834196,"x_train":57.3403812154,"imp":21.1816204709,"elo":1267.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.9845916387,"x_train":145.2967129053,"imp":20.5260498304,"elo":1293.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.1886115821,"x_train":291.5311600969,"imp":20.2117659176,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.033782756,"x_train":582.7415169755,"imp":20.0792359521,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.9700745263,"x_train":872.0774356109,"imp":20.0767903242,"elo":1306.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9325738702,"x_train":1165.3088146236,"imp":20.0591589805,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_small/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_classification/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..93881ac4569a10049740a4f55a1266736a1aabf5
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_classification/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "LimiX (default)", "AutoGluon 1.4 (best, 4h)", "RealMLP (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "TabM (tuned + ensembled)", "TabM (tuned)", "CatBoost (tuned)", "BetaTabPFN (default)", "iLTM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "TabDPT (tuned)", "CatBoost (tuned + ensembled)", "TabSwift (default)", "ModernNCA (tuned)", "XGBoost (tuned + ensembled)", "CatBoost (default)", "xRFM (tuned + ensembled)", "LightGBM (tuned)", "XGBoost (tuned)", "EBM (tuned + ensembled)", "TabM (default)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "ModernNCA (tuned + ensembled)", "SAP-RPT-OSS (default)", "RealMLP (default)", "ChimeraBoost (tuned)", "xRFM (tuned)", "EBM (default)", "EBM (tuned)", "TorchMLP (tuned + ensembled)", "TabDPT (default)", "ChimeraBoost (default)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "ModernNCA (default)", "ExtraTrees (tuned)", "XGBoost (default)", "FastaiMLP (tuned)", "LightGBM (default)", "TabSTAR (tuned)", "RandomForest (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "iLTM (default)", "OrionMSP (default)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (default)", "PerpetualBooster (tuned)", "TorchMLP (default)", "xRFM (default)", "ExtraTrees (default)", "TabFlex (default)", "RandomForest (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 1.0, 0.9285714030265808, 1.0, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0, 0.9285714030265808, 1.0, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0], [0.1428571492433548, NaN, 0.4642857015132904, 0.5535714030265808, 0.5357142686843872, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.5714285969734192, 0.75, 0.75, 0.75, 0.7857142686843872, 0.8571428656578064, 0.75, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.75, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.8214285969734192, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 1.0, 0.9642857313156128, 0.9642857313156128, 0.8214285969734192, 1.0, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 1.0, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0], [0.1785714328289032, 0.5357142686843872, NaN, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5, 0.4821428656578064, 0.7142857313156128, 0.75, 0.6785714030265808, 0.7857142686843872, 0.6785714030265808, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.7142857313156128, 0.8928571343421936, 0.75, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9464285969734192, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.9642857313156128, 1.0, 0.9642857313156128, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.9642857313156128, 1.0, 0.9642857313156128, 0.9285714030265808, 0.8571428656578064, 1.0, 0.9642857313156128, 0.9642857313156128, 1.0, 1.0, 0.9642857313156128, 1.0, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0], [0.1071428582072258, 0.4464285671710968, 0.5714285969734192, NaN, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.6428571343421936, 0.75, 0.6071428656578064, 0.6428571343421936, 0.75, 0.7857142686843872, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.7142857313156128, 0.9285714030265808, 0.7857142686843872, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.75, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 1.0, 0.8928571343421936, 0.9642857313156128, 0.7857142686843872, 0.9285714030265808, 0.7857142686843872, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 1.0], [0.1785714328289032, 0.4642857015132904, 0.4285714328289032, 0.4642857015132904, NaN, 0.5714285969734192, 0.4642857015132904, 0.5714285969734192, 0.75, 0.6785714030265808, 0.7142857313156128, 0.75, 0.5714285969734192, 0.75, 0.7857142686843872, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.8571428656578064, 0.75, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.75, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.7857142686843872, 0.9285714030265808, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.8214285969734192, 0.9285714030265808, 0.8214285969734192, 0.9285714030265808, 1.0, 0.9285714030265808, 0.8928571343421936, 0.8214285969734192, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 1.0], [0.1428571492433548, 0.4642857015132904, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.5, 0.6785714030265808, 0.75, 0.6428571343421936, 0.75, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.7142857313156128, 0.8571428656578064, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.8928571343421936, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.7857142686843872, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.75, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.8571428656578064, 0.9285714030265808, 0.7857142686843872, 0.9285714030265808, 0.7857142686843872, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.7857142686843872, 1.0, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.1071428582072258, 0.4285714328289032, 0.5, 0.3928571343421936, 0.5357142686843872, 0.5714285969734192, NaN, 0.5357142686843872, 0.6785714030265808, 0.6428571343421936, 0.75, 0.6071428656578064, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.75, 0.75, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.75, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.75, 0.8928571343421936, 0.7142857313156128, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8571428656578064, 0.9642857313156128, 0.9285714030265808, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 1.0, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0], [0.1071428582072258, 0.3928571343421936, 0.5178571343421936, 0.3571428656578064, 0.4285714328289032, 0.5, 0.4642857015132904, NaN, 0.6428571343421936, 0.5714285969734192, 0.6428571343421936, 0.75, 0.6071428656578064, 0.6071428656578064, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.75, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.7142857313156128, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.75, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.8571428656578064, 0.9285714030265808, 0.7857142686843872, 0.9285714030265808, 0.7857142686843872, 0.8928571343421936, 1.0, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.8571428656578064, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 1.0], [0.1785714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, NaN, 0.3928571343421936, 0.5714285969734192, 0.6071428656578064, 0.5357142686843872, 0.5357142686843872, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6785714030265808, 0.7857142686843872, 0.7857142686843872, 0.6071428656578064, 0.8571428656578064, 0.6428571343421936, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.6785714030265808, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.7142857313156128, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.7142857313156128, 0.8571428656578064, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.75, 0.8571428656578064, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8928571343421936, 0.8214285969734192, 0.9285714030265808, 0.8214285969734192, 0.8928571343421936, 0.75, 0.75, 0.75, 0.9285714030265808, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.7857142686843872, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.1071428582072258, 0.3214285671710968, 0.25, 0.2857142984867096, 0.3214285671710968, 0.25, 0.3571428656578064, 0.4285714328289032, 0.6071428656578064, NaN, 0.5, 0.6964285969734192, 0.5357142686843872, 0.4642857015132904, 0.6785714030265808, 0.7142857313156128, 0.6071428656578064, 0.6785714030265808, 0.6964285969734192, 0.75, 0.6071428656578064, 0.6785714030265808, 0.6071428656578064, 0.75, 0.75, 0.6785714030265808, 0.75, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.75, 0.75, 0.8214285969734192, 0.75, 0.75, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7857142686843872, 0.6428571343421936, 0.8214285969734192, 0.75, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8392857313156128, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.9107142686843872, 0.9285714030265808, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.4285714328289032, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.25, 0.3571428656578064, 0.4285714328289032, 0.5, NaN, 0.6071428656578064, 0.5714285969734192, 0.5, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.6785714030265808, 0.75, 0.75, 0.6071428656578064, 0.7142857313156128, 0.6071428656578064, 0.6071428656578064, 0.75, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.75, 0.7142857313156128, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.7857142686843872, 0.8928571343421936, 0.7857142686843872, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.7142857313156128, 0.8214285969734192, 0.75, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 1.0, 1.0], [0.1428571492433548, 0.25, 0.2142857164144516, 0.25, 0.25, 0.25, 0.3928571343421936, 0.25, 0.3928571343421936, 0.3035714328289032, 0.3928571343421936, NaN, 0.4642857015132904, 0.3928571343421936, 0.5357142686843872, 0.6428571343421936, 0.5714285969734192, 0.6428571343421936, 0.5714285969734192, 0.6785714030265808, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.75, 0.7142857313156128, 0.75, 0.5714285969734192, 0.7142857313156128, 0.75, 0.75, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6071428656578064, 0.75, 0.7142857313156128, 0.6785714030265808, 0.75, 0.75, 0.7857142686843872, 0.6428571343421936, 0.75, 0.75, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8571428656578064, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.75, 0.8571428656578064, 0.7857142686843872, 0.9285714030265808, 0.8035714030265808, 0.8928571343421936, 0.8571428656578064, 0.7857142686843872, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9464285969734192, 0.9285714030265808, 0.9642857313156128, 1.0], [0.1071428582072258, 0.25, 0.3214285671710968, 0.3928571343421936, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.5357142686843872, NaN, 0.7142857313156128, 0.5, 0.5, 0.5, 0.5, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.5357142686843872, 0.6964285969734192, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.75, 0.75, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.75, 0.8214285969734192, 0.75, 0.8214285969734192, 0.9285714030265808, 0.8214285969734192, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 1.0], [0.0714285746216774, 0.25, 0.25, 0.3571428656578064, 0.25, 0.3571428656578064, 0.2142857164144516, 0.3928571343421936, 0.4642857015132904, 0.5357142686843872, 0.5, 0.6071428656578064, 0.2857142984867096, NaN, 0.4642857015132904, 0.4642857015132904, 0.5357142686843872, 0.5357142686843872, 0.6785714030265808, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.75, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.75, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.6785714030265808, 0.6428571343421936, 0.75, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.75, 0.75, 0.7857142686843872, 0.75, 0.75, 0.8214285969734192, 0.9642857313156128, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 1.0], [0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.1785714328289032, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.4642857015132904, 0.5, 0.5357142686843872, NaN, 0.5357142686843872, 0.5, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.5, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.75, 0.7857142686843872, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.6428571343421936, 0.75, 0.7857142686843872, 0.5714285969734192, 0.75, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.9642857313156128, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.7857142686843872, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 1.0, 1.0], [0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.25, 0.25, 0.25, 0.1785714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.5, 0.5357142686843872, 0.4642857015132904, NaN, 0.4642857015132904, 0.5357142686843872, 0.5357142686843872, 0.5714285969734192, 0.5357142686843872, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.5357142686843872, 0.5714285969734192, 0.6785714030265808, 0.6428571343421936, 0.6428571343421936, 0.7321428656578064, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.75, 0.6428571343421936, 0.75, 0.6785714030265808, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7142857313156128, 0.75, 0.75, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.9642857313156128, 0.8214285969734192, 0.75, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128], [0.0714285746216774, 0.25, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.25, 0.3928571343421936, 0.3214285671710968, 0.4285714328289032, 0.5, 0.4642857015132904, 0.5, 0.5357142686843872, NaN, 0.5, 0.5, 0.5, 0.5, 0.4642857015132904, 0.5357142686843872, 0.5357142686843872, 0.6071428656578064, 0.6428571343421936, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.5, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.7142857313156128, 0.75, 0.7142857313156128, 0.6428571343421936, 0.6785714030265808, 0.6785714030265808, 0.75, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.75, 0.8214285969734192, 0.75, 0.8571428656578064, 0.7857142686843872, 0.75, 0.8928571343421936, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.0714285746216774, 0.1071428582072258, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.5, 0.4642857015132904, 0.3928571343421936, 0.4642857015132904, 0.5, NaN, 0.6071428656578064, 0.5357142686843872, 0.5357142686843872, 0.5714285969734192, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.7142857313156128, 0.6071428656578064, 0.5, 0.6428571343421936, 0.6785714030265808, 0.5714285969734192, 0.75, 0.6071428656578064, 0.5357142686843872, 0.5714285969734192, 0.6785714030265808, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.8214285969734192, 0.6428571343421936, 0.7142857313156128, 0.75, 0.6428571343421936, 0.7857142686843872, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.75, 0.8571428656578064, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.75, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.1428571492433548, 0.1785714328289032, 0.2857142984867096, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.25, 0.2142857164144516, 0.3035714328289032, 0.25, 0.4285714328289032, 0.3928571343421936, 0.3214285671710968, 0.4285714328289032, 0.4642857015132904, 0.5, 0.3928571343421936, NaN, 0.5714285969734192, 0.3928571343421936, 0.4285714328289032, 0.5357142686843872, 0.6071428656578064, 0.5, 0.4642857015132904, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.5357142686843872, 0.6785714030265808, 0.6785714030265808, 0.5357142686843872, 0.6071428656578064, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.6785714030265808, 0.5714285969734192, 0.75, 0.7857142686843872, 0.75, 0.7857142686843872, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.6785714030265808, 0.8571428656578064, 0.6785714030265808, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.7857142686843872, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 1.0], [0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.25, 0.25, 0.3214285671710968, 0.3928571343421936, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.5, 0.4642857015132904, 0.4285714328289032, NaN, 0.5, 0.5714285969734192, 0.5714285969734192, 0.5, 0.5, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.5357142686843872, 0.5714285969734192, 0.6428571343421936, 0.6785714030265808, 0.6785714030265808, 0.75, 0.7142857313156128, 0.75, 0.6071428656578064, 0.7857142686843872, 0.75, 0.75, 0.75, 0.75, 0.6785714030265808, 0.8928571343421936, 0.7857142686843872, 0.9285714030265808, 0.75, 0.7857142686843872, 0.75, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 1.0, 1.0], [0.0357142873108387, 0.1071428582072258, 0.25, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.3928571343421936, 0.3928571343421936, 0.3928571343421936, 0.4642857015132904, 0.3214285671710968, 0.2857142984867096, 0.5, 0.4642857015132904, 0.5, 0.4642857015132904, 0.6071428656578064, 0.5, NaN, 0.5357142686843872, 0.6428571343421936, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.5714285969734192, 0.6428571343421936, 0.6785714030265808, 0.5714285969734192, 0.5357142686843872, 0.6071428656578064, 0.6785714030265808, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.6071428656578064, 0.7857142686843872, 0.7142857313156128, 0.6071428656578064, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.7857142686843872, 0.7142857313156128, 0.75, 0.6428571343421936, 0.75, 0.6428571343421936, 0.75, 0.8214285969734192, 0.9285714030265808, 0.75, 0.7142857313156128, 0.75, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.9642857313156128, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128], [0.0357142873108387, 0.1785714328289032, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.1428571492433548, 0.25, 0.1785714328289032, 0.1428571492433548, 0.3214285671710968, 0.2857142984867096, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.5357142686843872, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4642857015132904, NaN, 0.5, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.5357142686843872, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.5, 0.5357142686843872, 0.6785714030265808, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.5714285969734192, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.6071428656578064, 0.8214285969734192, 0.6785714030265808, 0.75, 0.75, 0.7142857313156128, 0.75, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8214285969734192, 0.75, 0.8214285969734192, 0.7142857313156128, 0.8928571343421936, 0.8571428656578064, 0.75, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.7857142686843872, 0.9285714030265808, 0.8571428656578064, 0.8214285969734192, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.25, 0.2142857164144516, 0.1785714328289032, 0.2857142984867096, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.3928571343421936, 0.3035714328289032, 0.4285714328289032, 0.3928571343421936, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.3571428656578064, 0.5, NaN, 0.5357142686843872, 0.5, 0.5, 0.5714285969734192, 0.4642857015132904, 0.5, 0.6071428656578064, 0.5, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.6428571343421936, 0.5714285969734192, 0.5, 0.5714285969734192, 0.75, 0.7857142686843872, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.75, 0.7142857313156128, 0.75, 0.6428571343421936, 0.75, 0.75, 0.8571428656578064, 0.6785714030265808, 0.7142857313156128, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.9642857313156128, 0.7857142686843872, 0.9285714030265808, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 1.0], [0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.2142857164144516, 0.1071428582072258, 0.2857142984867096, 0.25, 0.3928571343421936, 0.25, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.3928571343421936, 0.5, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, NaN, 0.5357142686843872, 0.6071428656578064, 0.5714285969734192, 0.5714285969734192, 0.5, 0.5, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4821428656578064, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6428571343421936, 0.6071428656578064, 0.5714285969734192, 0.6785714030265808, 0.625, 0.75, 0.6428571343421936, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.75, 0.75, 0.7142857313156128, 0.75, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 1.0], [0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.2142857164144516, 0.1071428582072258, 0.1428571492433548, 0.25, 0.25, 0.2857142984867096, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.5, 0.5, 0.4285714328289032, 0.4285714328289032, 0.5, 0.4642857015132904, NaN, 0.5, 0.5, 0.5714285969734192, 0.6428571343421936, 0.5714285969734192, 0.5, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.5, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.6071428656578064, 0.5714285969734192, 0.7142857313156128, 0.6785714030265808, 0.625, 0.75, 0.7142857313156128, 0.75, 0.8571428656578064, 0.7142857313156128, 0.9642857313156128, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.7857142686843872, 0.9285714030265808, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 1.0], [0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.0714285746216774, 0.1428571492433548, 0.3214285671710968, 0.2857142984867096, 0.25, 0.3571428656578064, 0.3928571343421936, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.5357142686843872, 0.3214285671710968, 0.4285714328289032, 0.3928571343421936, 0.5, 0.3928571343421936, 0.5, NaN, 0.5714285969734192, 0.4285714328289032, 0.5, 0.5357142686843872, 0.5, 0.5, 0.6071428656578064, 0.5, 0.5357142686843872, 0.7142857313156128, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.75, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.75, 0.75, 0.7142857313156128, 0.75, 0.7857142686843872, 0.75, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.9285714030265808, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.0714285746216774, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.3214285671710968, 0.25, 0.2857142984867096, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.4285714328289032, 0.5, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.5, 0.5714285969734192, 0.5, 0.4642857015132904, 0.5714285969734192, 0.5, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.5714285969734192, 0.4642857015132904, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.75, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6785714030265808, 0.7142857313156128, 0.8214285969734192, 0.7857142686843872, 0.7142857313156128, 0.75, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128], [0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.3928571343421936, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.5, 0.4285714328289032, 0.3214285671710968, 0.4285714328289032, 0.4642857015132904, 0.5357142686843872, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5, 0.6428571343421936, 0.6071428656578064, 0.5357142686843872, 0.5, 0.3928571343421936, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.5714285969734192, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6785714030265808, 0.5714285969734192, 0.6785714030265808, 0.7857142686843872, 0.75, 0.7857142686843872, 0.7142857313156128, 0.75, 0.75, 0.75, 0.75, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.5, 0.5, 0.3571428656578064, 0.5, 0.5714285969734192, 0.5, NaN, 0.5357142686843872, 0.4642857015132904, 0.5357142686843872, 0.5357142686843872, 0.5, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.6785714030265808, 0.6071428656578064, 0.7857142686843872, 0.6428571343421936, 0.7857142686843872, 0.8214285969734192, 0.75, 0.8928571343421936, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8571428656578064, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 1.0], [0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.3214285671710968, 0.25, 0.3571428656578064, 0.3571428656578064, 0.25, 0.3571428656578064, 0.3571428656578064, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.5, 0.4285714328289032, 0.4642857015132904, 0.5, 0.3571428656578064, 0.4642857015132904, NaN, 0.4285714328289032, 0.4285714328289032, 0.5, 0.4642857015132904, 0.5714285969734192, 0.6071428656578064, 0.4642857015132904, 0.6071428656578064, 0.7142857313156128, 0.7142857313156128, 0.6071428656578064, 0.6785714030265808, 0.5714285969734192, 0.7142857313156128, 0.6428571343421936, 0.75, 0.75, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.75, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.2857142984867096, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.3928571343421936, 0.2142857164144516, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, 0.5, 0.5, 0.4285714328289032, 0.5, 0.5, 0.4285714328289032, 0.3928571343421936, 0.5357142686843872, 0.5714285969734192, NaN, 0.5, 0.5357142686843872, 0.5714285969734192, 0.5, 0.5, 0.6071428656578064, 0.5714285969734192, 0.4821428656578064, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.7142857313156128, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.7857142686843872, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.75, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.9285714030265808, 0.75, 0.75, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808], [0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.25, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.2678571343421936, 0.5, 0.25, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.4642857015132904, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5, 0.5, 0.4642857015132904, 0.4642857015132904, 0.5714285969734192, 0.5, NaN, 0.5357142686843872, 0.5, 0.5, 0.6071428656578064, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.8214285969734192, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.75, 0.7142857313156128, 0.75, 0.6785714030265808, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.75, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 1.0], [0.0714285746216774, 0.1071428582072258, 0.0535714291036129, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.25, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.25, 0.3571428656578064, 0.3571428656578064, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.5357142686843872, 0.5, 0.4642857015132904, 0.5, 0.4642857015132904, 0.4642857015132904, NaN, 0.5357142686843872, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.75, 0.6071428656578064, 0.5714285969734192, 0.5357142686843872, 0.5, 0.7142857313156128, 0.6428571343421936, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.6071428656578064, 0.75, 0.6785714030265808, 0.6785714030265808, 0.8214285969734192, 0.6785714030265808, 0.6964285969734192, 0.6785714030265808, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.9642857313156128, 1.0], [0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.2142857164144516, 0.25, 0.3571428656578064, 0.25, 0.4285714328289032, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.5178571343421936, 0.4642857015132904, 0.5, 0.4285714328289032, 0.6071428656578064, 0.5, 0.5357142686843872, 0.4285714328289032, 0.5, 0.4642857015132904, NaN, 0.4642857015132904, 0.6428571343421936, 0.5357142686843872, 0.6071428656578064, 0.5714285969734192, 0.5357142686843872, 0.5357142686843872, 0.5714285969734192, 0.5357142686843872, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6428571343421936, 0.7857142686843872, 0.75, 0.7857142686843872, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6785714030265808, 0.75, 0.7857142686843872, 0.75, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.75, 0.8214285969734192, 0.7857142686843872, 0.8035714030265808, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.9642857313156128], [0.1071428582072258, 0.25, 0.1785714328289032, 0.1428571492433548, 0.25, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.5, 0.4642857015132904, 0.5, 0.4642857015132904, 0.5, 0.4285714328289032, 0.5, 0.5, 0.4285714328289032, 0.5357142686843872, NaN, 0.5714285969734192, 0.5, 0.5357142686843872, 0.6071428656578064, 0.5357142686843872, 0.6428571343421936, 0.5714285969734192, 0.5357142686843872, 0.7142857313156128, 0.6785714030265808, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.6071428656578064, 0.7142857313156128, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.6785714030265808, 0.75, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.7142857313156128, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 1.0], [0.1428571492433548, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.2857142984867096, 0.25, 0.3571428656578064, 0.3928571343421936, 0.25, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.2857142984867096, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.2857142984867096, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.5, 0.3928571343421936, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, NaN, 0.3928571343421936, 0.5357142686843872, 0.4642857015132904, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.5, 0.6071428656578064, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.8928571343421936, 0.7857142686843872, 0.7857142686843872, 0.9285714030265808, 0.7857142686843872, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.2857142984867096, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.4285714328289032, 0.4642857015132904, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.5357142686843872, 0.3928571343421936, 0.4285714328289032, 0.3571428656578064, 0.4642857015132904, 0.5, 0.6071428656578064, NaN, 0.5, 0.5714285969734192, 0.5714285969734192, 0.5, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.5714285969734192, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.9285714030265808, 0.8928571343421936, 1.0], [0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.2857142984867096, 0.1428571492433548, 0.25, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.4285714328289032, 0.4285714328289032, 0.3214285671710968, 0.3928571343421936, 0.2857142984867096, 0.3214285671710968, 0.4285714328289032, 0.3571428656578064, 0.3928571343421936, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.3571428656578064, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.5, NaN, 0.4285714328289032, 0.5, 0.4642857015132904, 0.5, 0.5, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.5714285969734192, 0.75, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.75, 0.75, 0.6071428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.9642857313156128, 0.8928571343421936, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.9642857313156128, 1.0], [0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.25, 0.25, 0.25, 0.25, 0.3571428656578064, 0.2857142984867096, 0.3214285671710968, 0.2857142984867096, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.3928571343421936, 0.2857142984867096, 0.5178571343421936, 0.4642857015132904, 0.25, 0.4285714328289032, 0.3928571343421936, 0.5357142686843872, 0.4285714328289032, 0.5714285969734192, NaN, 0.5, 0.5357142686843872, 0.5, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.6428571343421936, 0.6785714030265808, 0.5357142686843872, 0.6785714030265808, 0.5357142686843872, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.75, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.1428571492433548, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.25, 0.1785714328289032, 0.3214285671710968, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.5357142686843872, 0.3928571343421936, 0.3571428656578064, 0.2857142984867096, 0.4285714328289032, 0.3928571343421936, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.5, 0.5, NaN, 0.5714285969734192, 0.6071428656578064, 0.5, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.5714285969734192, 0.6785714030265808, 0.5714285969734192, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 1.0], [0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.25, 0.3928571343421936, 0.3214285671710968, 0.5, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4642857015132904, 0.3571428656578064, 0.3928571343421936, 0.5, 0.5357142686843872, 0.4642857015132904, 0.4285714328289032, NaN, 0.5714285969734192, 0.5, 0.6071428656578064, 0.7142857313156128, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.9285714030265808, 0.7857142686843872, 0.75, 0.8928571343421936, 0.7142857313156128, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128], [0.0357142873108387, 0.1428571492433548, 0.1428571492433548, 0.25, 0.2142857164144516, 0.25, 0.1785714328289032, 0.25, 0.25, 0.3571428656578064, 0.2142857164144516, 0.3571428656578064, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.2857142984867096, 0.4285714328289032, 0.3928571343421936, 0.2142857164144516, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4642857015132904, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.3928571343421936, 0.5, 0.5, 0.3928571343421936, 0.4285714328289032, NaN, 0.5357142686843872, 0.5714285969734192, 0.4642857015132904, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.75, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.75, 0.7142857313156128, 0.7857142686843872, 0.8571428656578064, 0.8214285969734192, 0.75, 0.8214285969734192, 0.8571428656578064, 0.75, 0.7142857313156128, 0.75, 0.8214285969734192, 0.9285714030265808, 0.9642857313156128], [0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.25, 0.2857142984867096, 0.3571428656578064, 0.2142857164144516, 0.25, 0.3214285671710968, 0.25, 0.25, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.25, 0.3214285671710968, 0.2857142984867096, 0.25, 0.4285714328289032, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.1785714328289032, 0.5, 0.4642857015132904, 0.4642857015132904, 0.5, 0.3214285671710968, 0.5, 0.4642857015132904, 0.5, 0.5, 0.4642857015132904, NaN, 0.5714285969734192, 0.4285714328289032, 0.5357142686843872, 0.5714285969734192, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.5714285969734192, 0.6785714030265808, 0.6428571343421936, 0.75, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.7857142686843872, 0.7142857313156128, 0.75, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.6428571343421936, 0.75, 0.75, 0.75, 0.8214285969734192, 0.8214285969734192, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.0357142873108387, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.0357142873108387, 0.1785714328289032, 0.1428571492433548, 0.25, 0.1428571492433548, 0.25, 0.25, 0.3214285671710968, 0.25, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.2142857164144516, 0.25, 0.3928571343421936, 0.3214285671710968, 0.2142857164144516, 0.375, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.2857142984867096, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, NaN, 0.5357142686843872, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5, 0.7142857313156128, 0.5, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.75, 0.75, 0.8571428656578064, 0.7142857313156128, 0.9642857313156128, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128], [0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.2142857164144516, 0.25, 0.3571428656578064, 0.2142857164144516, 0.25, 0.25, 0.2142857164144516, 0.25, 0.25, 0.3571428656578064, 0.25, 0.3928571343421936, 0.25, 0.375, 0.2857142984867096, 0.3571428656578064, 0.25, 0.3928571343421936, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.3214285671710968, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.2857142984867096, 0.5357142686843872, 0.5714285969734192, 0.4642857015132904, NaN, 0.5178571343421936, 0.5714285969734192, 0.5714285969734192, 0.5, 0.5714285969734192, 0.5535714030265808, 0.6785714030265808, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.75, 0.7142857313156128, 0.8928571343421936, 0.75, 0.7142857313156128, 0.8928571343421936, 0.6428571343421936, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128], [0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.0357142873108387, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.25, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.25, 0.3214285671710968, 0.25, 0.3214285671710968, 0.3571428656578064, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.25, 0.3928571343421936, 0.4285714328289032, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.4642857015132904, 0.2857142984867096, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.4821428656578064, NaN, 0.4642857015132904, 0.6964285969734192, 0.4642857015132904, 0.4642857015132904, 0.5357142686843872, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.5357142686843872, 0.6428571343421936, 0.75, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.6428571343421936, 0.75, 0.8928571343421936, 0.75, 0.8214285969734192, 0.7142857313156128, 0.9285714030265808, 0.7142857313156128, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.9285714030265808, 1.0], [0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.25, 0.2857142984867096, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2857142984867096, 0.3571428656578064, 0.2142857164144516, 0.2857142984867096, 0.3928571343421936, 0.2142857164144516, 0.4285714328289032, 0.3571428656578064, 0.25, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.25, 0.3928571343421936, 0.3214285671710968, 0.4642857015132904, 0.4285714328289032, 0.3571428656578064, 0.3214285671710968, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5357142686843872, NaN, 0.5, 0.4642857015132904, 0.5357142686843872, 0.5, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.7857142686843872, 0.6428571343421936, 0.75, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128], [0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.2142857164144516, 0.25, 0.2857142984867096, 0.3214285671710968, 0.25, 0.3928571343421936, 0.25, 0.3214285671710968, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.2142857164144516, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.25, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.3571428656578064, 0.4642857015132904, 0.4285714328289032, 0.4285714328289032, 0.3035714328289032, 0.5, NaN, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.6428571343421936, 0.4642857015132904, 0.6428571343421936, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.75, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.6785714030265808, 0.8571428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.9642857313156128, 0.9642857313156128], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.1071428582072258, 0.2857142984867096, 0.1785714328289032, 0.3571428656578064, 0.2142857164144516, 0.1428571492433548, 0.25, 0.2857142984867096, 0.2142857164144516, 0.1785714328289032, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3928571343421936, 0.2142857164144516, 0.3928571343421936, 0.3214285671710968, 0.2857142984867096, 0.4285714328289032, 0.4642857015132904, 0.4285714328289032, 0.2857142984867096, 0.3928571343421936, 0.4285714328289032, 0.5, 0.5, 0.5357142686843872, 0.5357142686843872, 0.5357142686843872, NaN, 0.5714285969734192, 0.6071428656578064, 0.5714285969734192, 0.5357142686843872, 0.5714285969734192, 0.6785714030265808, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.75, 0.6785714030265808, 0.6785714030265808, 0.75, 0.75, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.9285714030265808], [0.1071428582072258, 0.0714285746216774, 0.0, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.25, 0.2857142984867096, 0.2857142984867096, 0.25, 0.2857142984867096, 0.25, 0.25, 0.25, 0.2142857164144516, 0.3571428656578064, 0.25, 0.25, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.25, 0.3928571343421936, 0.2857142984867096, 0.4285714328289032, 0.5357142686843872, 0.4642857015132904, 0.5357142686843872, 0.4285714328289032, NaN, 0.4642857015132904, 0.5714285969734192, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6428571343421936, 0.7857142686843872, 0.75, 0.7142857313156128, 0.75, 0.75, 0.8214285969734192, 0.8571428656578064, 0.9285714030265808], [0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.0714285746216774, 0.3571428656578064, 0.2142857164144516, 0.2857142984867096, 0.25, 0.0357142873108387, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.1071428582072258, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.3214285671710968, 0.2142857164144516, 0.3571428656578064, 0.3928571343421936, 0.25, 0.3571428656578064, 0.4642857015132904, 0.4285714328289032, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.5, 0.4464285671710968, 0.4642857015132904, 0.5, 0.5357142686843872, 0.3928571343421936, 0.5357142686843872, NaN, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.6785714030265808, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808], [0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.25, 0.1607142835855484, 0.2857142984867096, 0.25, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.25, 0.3214285671710968, 0.25, 0.25, 0.25, 0.25, 0.25, 0.2142857164144516, 0.25, 0.3214285671710968, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.25, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, NaN, 0.4642857015132904, 0.5535714030265808, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.5714285969734192, 0.9285714030265808, 0.5714285969734192, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6785714030265808, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.7142857313156128, 0.75, 0.8214285969734192, 0.9285714030265808], [0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.25, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.25, 0.25, 0.25, 0.2142857164144516, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.3571428656578064, 0.1785714328289032, 0.3571428656578064, 0.2857142984867096, 0.1785714328289032, 0.2142857164144516, 0.25, 0.25, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1785714328289032, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.4642857015132904, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.5357142686843872, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.5357142686843872, NaN, 0.5714285969734192, 0.4285714328289032, 0.5357142686843872, 0.75, 0.5, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.75, 0.7857142686843872, 0.6071428656578064, 0.8571428656578064, 0.6428571343421936, 0.75, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.8928571343421936, 0.9642857313156128], [0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.25, 0.1785714328289032, 0.25, 0.2142857164144516, 0.25, 0.25, 0.1785714328289032, 0.25, 0.25, 0.25, 0.3214285671710968, 0.25, 0.25, 0.25, 0.25, 0.25, 0.2142857164144516, 0.25, 0.3214285671710968, 0.25, 0.2142857164144516, 0.2857142984867096, 0.3214285671710968, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.25, 0.3928571343421936, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.4464285671710968, 0.4285714328289032, NaN, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.5714285969734192, 0.9642857313156128, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.5357142686843872, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.75, 0.8214285969734192, 0.9285714030265808], [0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.25, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.25, 0.2857142984867096, 0.1785714328289032, 0.2142857164144516, 0.2857142984867096, 0.25, 0.2142857164144516, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3035714328289032, 0.2857142984867096, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.25, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.4642857015132904, 0.3571428656578064, 0.4642857015132904, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.6071428656578064, 0.5714285969734192, 0.5714285969734192, NaN, 0.5, 0.5, 0.5357142686843872, 0.6785714030265808, 0.5, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.75, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.8214285969734192, 0.9285714030265808], [0.0357142873108387, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0357142873108387, 0.1071428582072258, 0.0, 0.0714285746216774, 0.0892857164144516, 0.1071428582072258, 0.1964285671710968, 0.0714285746216774, 0.0357142873108387, 0.1785714328289032, 0.2142857164144516, 0.25, 0.25, 0.2142857164144516, 0.2142857164144516, 0.0714285746216774, 0.2857142984867096, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.25, 0.1785714328289032, 0.25, 0.25, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.25, 0.3571428656578064, 0.3214285671710968, 0.2857142984867096, 0.25, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4642857015132904, 0.5714285969734192, 0.4642857015132904, 0.5357142686843872, 0.5, NaN, 0.4642857015132904, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.5714285969734192, 0.6428571343421936, 0.6785714030265808, 0.75, 0.75, 0.7857142686843872, 0.8214285969734192, 0.9642857313156128], [0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.25, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.1428571492433548, 0.1071428582072258, 0.25, 0.1071428582072258, 0.3214285671710968, 0.25, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.25, 0.2142857164144516, 0.2142857164144516, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.25, 0.25, 0.3214285671710968, 0.25, 0.3571428656578064, 0.2857142984867096, 0.3928571343421936, 0.3571428656578064, 0.25, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5357142686843872, 0.25, 0.5357142686843872, 0.5, 0.5357142686843872, NaN, 0.5, 0.6428571343421936, 0.5714285969734192, 0.5714285969734192, 0.75, 0.7142857313156128, 0.6785714030265808, 0.7678571343421936, 0.7142857313156128, 0.75, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.8571428656578064, 0.9642857313156128], [0.0, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.25, 0.2857142984867096, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.25, 0.3214285671710968, 0.25, 0.25, 0.3214285671710968, 0.2857142984867096, 0.3928571343421936, 0.2142857164144516, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, 0.5, 0.4285714328289032, 0.4642857015132904, 0.4285714328289032, 0.5, NaN, 0.4285714328289032, 0.75, 0.5, 0.5714285969734192, 0.5357142686843872, 0.6071428656578064, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.8214285969734192, 0.7142857313156128, 0.7142857313156128, 0.75, 0.7142857313156128, 0.8928571343421936], [0.0714285746216774, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.25, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.25, 0.25, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.25, 0.2142857164144516, 0.3214285671710968, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.25, 0.2857142984867096, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3571428656578064, 0.3214285671710968, 0.0714285746216774, 0.3214285671710968, 0.0357142873108387, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.5, 0.3214285671710968, 0.5714285969734192, 0.6071428656578064, 0.5, 0.5, 0.6071428656578064, 0.6785714030265808, 0.9285714030265808], [0.0, 0.0, 0.0, 0.0357142873108387, 0.0714285746216774, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.1428571492433548, 0.0357142873108387, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1071428582072258, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.25, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.25, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.2857142984867096, 0.2142857164144516, 0.25, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.5, 0.4285714328289032, 0.4285714328289032, 0.25, 0.4285714328289032, NaN, 0.5, 0.5714285969734192, 0.5, 0.6428571343421936, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.7321428656578064, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.8928571343421936], [0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.0714285746216774, 0.2857142984867096, 0.25, 0.2857142984867096, 0.1071428582072258, 0.3571428656578064, 0.2142857164144516, 0.3928571343421936, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.4642857015132904, 0.2857142984867096, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.5, 0.5714285969734192, 0.5, NaN, 0.5, 0.6071428656578064, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.75, 0.7857142686843872, 0.9285714030265808], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.25, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.0357142873108387, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.25, 0.25, 0.25, 0.3571428656578064, 0.3571428656578064, 0.25, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.25, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.25, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5, NaN, 0.5357142686843872, 0.5, 0.5357142686843872, 0.4642857015132904, 0.4285714328289032, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.7142857313156128, 0.8928571343421936], [0.0357142873108387, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.2142857164144516, 0.25, 0.1785714328289032, 0.25, 0.0714285746216774, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.25, 0.25, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.25, 0.1428571492433548, 0.2857142984867096, 0.25, 0.2857142984867096, 0.1071428582072258, 0.25, 0.2142857164144516, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.2142857164144516, 0.3928571343421936, 0.3928571343421936, 0.3214285671710968, 0.2857142984867096, 0.4642857015132904, 0.6071428656578064, 0.5, 0.3928571343421936, 0.4642857015132904, NaN, 0.5, 0.5714285969734192, 0.5357142686843872, 0.5, 0.5357142686843872, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.75, 0.9285714030265808], [0.0, 0.0714285746216774, 0.0, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.0357142873108387, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.25, 0.1785714328289032, 0.0714285746216774, 0.1428571492433548, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1071428582072258, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.25, 0.2857142984867096, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.3928571343421936, 0.2857142984867096, 0.3214285671710968, 0.3928571343421936, 0.5357142686843872, 0.3571428656578064, 0.3928571343421936, 0.5, 0.5, NaN, 0.5357142686843872, 0.5357142686843872, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.75, 0.8928571343421936], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.0714285746216774, 0.2142857164144516, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.0714285746216774, 0.2142857164144516, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.25, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.1428571492433548, 0.3571428656578064, 0.3928571343421936, 0.3214285671710968, 0.2321428507566452, 0.3571428656578064, 0.5, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.4642857015132904, NaN, 0.5, 0.6071428656578064, 0.5, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.9642857313156128], [0.0, 0.0, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.25, 0.0357142873108387, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3214285671710968, 0.25, 0.2142857164144516, 0.25, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.2857142984867096, 0.3928571343421936, 0.6785714030265808, 0.4285714328289032, 0.3928571343421936, 0.5357142686843872, 0.4642857015132904, 0.4642857015132904, 0.5, NaN, 0.5714285969734192, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.7857142686843872, 0.9285714030265808], [0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.1964285671710968, 0.2857142984867096, 0.1071428582072258, 0.25, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.25, 0.2142857164144516, 0.1428571492433548, 0.0714285746216774, 0.2142857164144516, 0.1428571492433548, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3571428656578064, 0.25, 0.3928571343421936, 0.25, 0.3571428656578064, 0.25, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.3928571343421936, 0.5714285969734192, 0.5, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, NaN, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.9642857313156128, 0.9642857313156128], [0.0, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.0714285746216774, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.25, 0.25, 0.1428571492433548, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.3928571343421936, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.1785714328289032, 0.3928571343421936, 0.2678571343421936, 0.3928571343421936, 0.5, 0.4642857015132904, 0.4642857015132904, 0.5, 0.3571428656578064, 0.5, NaN, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.8571428656578064], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.2857142984867096, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1428571492433548, 0.3214285671710968, 0.2142857164144516, 0.25, 0.2857142984867096, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.25, 0.2857142984867096, 0.2857142984867096, 0.5, 0.3214285671710968, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.3571428656578064, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.7142857313156128, 0.7142857313156128, 0.6428571343421936, 0.8571428656578064], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0535714291036129, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.0357142873108387, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.25, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.3214285671710968, 0.1785714328289032, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.25, 0.3214285671710968, 0.2857142984867096, 0.5, 0.3571428656578064, 0.3214285671710968, 0.4285714328289032, 0.4285714328289032, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, NaN, 0.6071428656578064, 0.6071428656578064, 0.8214285969734192], [0.0, 0.0, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.0357142873108387, 0.0714285746216774, 0.1428571492433548, 0.0357142873108387, 0.1071428582072258, 0.0357142873108387, 0.1785714328289032, 0.0357142873108387, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.25, 0.25, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3928571343421936, 0.2857142984867096, 0.25, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.2857142984867096, 0.3928571343421936, NaN, 0.6071428656578064, 0.75], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.1785714328289032, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.2857142984867096, 0.25, 0.25, 0.2142857164144516, 0.2142857164144516, 0.0357142873108387, 0.3928571343421936, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, NaN, 0.9285714030265808], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0, 0.0, 0.0714285746216774, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.25, 0.0714285746216774, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9216269793},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8464781748},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8442460336},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8306051579},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8258928561},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8249007937},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8080357159},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.805803572},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.750496034},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7410714298},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7371031765},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.689732141},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.682291667},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6631944431},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6592261875},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6292162688},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6289682531},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6284722233},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6262400747},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6254960306},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6180555578},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6170634937},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5870535729},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5853174627},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5840773823},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5758928575},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5719246042},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5699404762},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5610119031},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.556547622},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5562996029},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5483630967},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.547123016},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5451388866},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5431547615},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5188492028},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5158730165},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5074404799},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.5057043636},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5044642862},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5024801594},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4975198423},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4598214304},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4511408755},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4481646833},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4360119037},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4305555563},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4069940461},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3958333344},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3878968271},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3876488108},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3849206348},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3834325409},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3747519859},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3489583354},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3437500032},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3410218277},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3392857174},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2886904789},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.287946431},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2693452407},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.257440479},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2569444464},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2569444453},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2517361126},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2400793661},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2358630983},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2274305583},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2068452412},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1902281778},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.159722225},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1269841287},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0352182551}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_classification/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_classification/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..dd6a6b503bb335e4e4978fc79b6299b8c19aa884
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_classification/datasets_small/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,TabFM (default),TabPFN-3 (default),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),LimiX (default),"AutoGluon 1.4 (best, 4h)",RealMLP (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),TabM (tuned + ensembled),TabM (tuned),CatBoost (tuned),BetaTabPFN (default),iLTM (tuned + ensembled),LightGBM (tuned + ensembled),TabDPT (tuned),CatBoost (tuned + ensembled),TabSwift (default),ModernNCA (tuned),XGBoost (tuned + ensembled),CatBoost (default),xRFM (tuned + ensembled),LightGBM (tuned),XGBoost (tuned),EBM (tuned + ensembled),TabM (default),ChimeraBoost (tuned + ensembled),iLTM (tuned),ModernNCA (tuned + ensembled),SAP-RPT-OSS (default),RealMLP (default),ChimeraBoost (tuned),xRFM (tuned),EBM (default),EBM (tuned),TorchMLP (tuned + ensembled),TabDPT (default),ChimeraBoost (default),FastaiMLP (tuned + ensembled),TorchMLP (tuned),ExtraTrees (tuned + ensembled),ModernNCA (default),ExtraTrees (tuned),XGBoost (default),FastaiMLP (tuned),LightGBM (default),TabSTAR (tuned),RandomForest (tuned + ensembled),TabSTAR (tuned + ensembled),iLTM (default),OrionMSP (default),RandomForest (tuned),PerpetualBooster (tuned + ensembled),TabSTAR (default),PerpetualBooster (tuned),TorchMLP (default),xRFM (default),ExtraTrees (default),TabFlex (default),RandomForest (default),FastaiMLP (default),KNN (tuned + ensembled),PerpetualBooster (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.85714287,0.8214286,0.89285713,0.8214286,0.85714287,0.89285713,0.89285713,0.8214286,0.89285713,0.89285713,0.85714287,0.89285713,0.9285714,0.89285713,0.9285714,0.9285714,0.9285714,0.85714287,0.9285714,0.96428573,0.96428573,0.89285713,0.9285714,0.9285714,0.89285713,0.9285714,0.9285714,0.89285713,0.89285713,0.89285713,0.9285714,0.9285714,0.89285713,0.89285713,0.85714287,0.89285713,0.9285714,0.89285713,0.89285713,0.96428573,0.96428573,0.9285714,0.89285713,0.96428573,0.9285714,0.9285714,0.9285714,0.96428573,0.89285713,0.9285714,0.89285713,0.9285714,0.85714287,0.89285713,0.96428573,0.9285714,1.0,0.9285714,1.0,0.96428573,0.96428573,0.96428573,1.0,0.96428573,1.0,0.9285714,1.0,0.96428573,0.96428573,1.0,0.96428573,1.0
+TabPFN-3 (default),0.14285715,,0.4642857,0.5535714,0.53571427,0.53571427,0.5714286,0.60714287,0.71428573,0.6785714,0.5714286,0.75,0.75,0.75,0.78571427,0.85714287,0.75,0.89285713,0.8214286,0.89285713,0.89285713,0.8214286,0.85714287,0.85714287,0.89285713,0.89285713,0.85714287,0.9285714,0.85714287,0.78571427,0.78571427,0.89285713,0.89285713,0.9285714,0.75,0.89285713,0.85714287,0.9285714,0.85714287,0.85714287,0.9285714,0.85714287,0.9285714,0.85714287,0.9285714,0.96428573,0.9285714,0.96428573,0.96428573,0.9285714,0.9285714,0.8214286,0.9285714,0.8214286,0.96428573,1.0,0.96428573,0.96428573,0.8214286,1.0,0.9285714,0.96428573,0.96428573,0.9285714,0.96428573,1.0,0.96428573,0.96428573,0.96428573,0.96428573,1.0,0.96428573,1.0
+RealTabPFN-2.5 (default),0.17857143,0.53571427,,0.42857143,0.5714286,0.42857143,0.5,0.48214287,0.71428573,0.75,0.6785714,0.78571427,0.6785714,0.75,0.78571427,0.8214286,0.8214286,0.78571427,0.71428573,0.89285713,0.75,0.85714287,0.85714287,0.8214286,0.89285713,0.89285713,0.85714287,0.9285714,0.89285713,0.85714287,0.85714287,0.85714287,0.9464286,0.8214286,0.8214286,0.85714287,0.85714287,0.9285714,0.89285713,0.9285714,0.9285714,0.85714287,0.9285714,0.96428573,0.89285713,0.85714287,0.9285714,0.85714287,0.96428573,1.0,0.96428573,0.85714287,0.89285713,0.85714287,0.96428573,1.0,0.96428573,0.9285714,0.85714287,1.0,0.96428573,0.96428573,1.0,1.0,0.96428573,1.0,0.9285714,0.9285714,0.96428573,0.96428573,1.0,0.96428573,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.10714286,0.44642857,0.5714286,,0.53571427,0.5714286,0.60714287,0.64285713,0.71428573,0.71428573,0.64285713,0.75,0.60714287,0.64285713,0.75,0.78571427,0.89285713,0.8214286,0.8214286,0.89285713,0.71428573,0.9285714,0.78571427,0.89285713,0.9285714,0.9285714,0.85714287,0.9285714,0.89285713,0.8214286,0.8214286,0.85714287,0.85714287,0.89285713,0.85714287,0.8214286,0.89285713,0.9285714,0.85714287,0.8214286,0.85714287,0.75,0.85714287,0.85714287,0.85714287,0.9285714,0.85714287,0.9285714,1.0,0.89285713,0.96428573,0.78571427,0.9285714,0.78571427,0.89285713,0.96428573,0.9285714,0.9285714,0.8214286,0.96428573,0.9285714,0.96428573,0.96428573,0.85714287,0.96428573,0.96428573,0.9285714,0.9285714,0.9285714,0.96428573,0.9285714,0.9285714,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.17857143,0.4642857,0.42857143,0.4642857,,0.5714286,0.4642857,0.5714286,0.75,0.6785714,0.71428573,0.75,0.5714286,0.75,0.78571427,0.75,0.78571427,0.78571427,0.8214286,0.85714287,0.78571427,0.85714287,0.75,0.89285713,0.89285713,0.89285713,0.89285713,0.89285713,0.89285713,0.78571427,0.8214286,0.89285713,0.89285713,0.89285713,0.75,0.8214286,0.89285713,0.8214286,0.85714287,0.8214286,0.89285713,0.78571427,0.9285714,0.8214286,0.85714287,0.89285713,0.9285714,0.89285713,0.9285714,0.89285713,0.96428573,0.8214286,0.9285714,0.8214286,0.9285714,1.0,0.9285714,0.89285713,0.8214286,0.9285714,0.9285714,0.96428573,0.96428573,0.9285714,0.9285714,0.96428573,0.96428573,0.9285714,0.96428573,0.96428573,0.96428573,1.0,1.0
+TabPFN-2.6 (default),0.14285715,0.4642857,0.5714286,0.42857143,0.42857143,,0.42857143,0.5,0.6785714,0.75,0.64285713,0.75,0.71428573,0.64285713,0.71428573,0.75,0.78571427,0.8214286,0.89285713,0.89285713,0.71428573,0.85714287,0.78571427,0.85714287,0.85714287,0.9285714,0.89285713,0.96428573,0.89285713,0.8214286,0.78571427,0.8214286,0.89285713,0.9285714,0.78571427,0.85714287,0.8214286,0.85714287,0.8214286,0.89285713,0.9285714,0.75,0.89285713,0.89285713,0.9285714,0.89285713,0.9285714,0.89285713,0.96428573,0.85714287,0.9285714,0.78571427,0.9285714,0.78571427,0.9285714,0.96428573,0.89285713,0.96428573,0.78571427,1.0,0.96428573,0.96428573,0.9285714,0.89285713,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+TabICLv2 (default),0.10714286,0.42857143,0.5,0.39285713,0.53571427,0.5714286,,0.53571427,0.6785714,0.64285713,0.75,0.60714287,0.71428573,0.78571427,0.71428573,0.75,0.75,0.85714287,0.78571427,0.8214286,0.78571427,0.75,0.8214286,0.78571427,0.78571427,0.8214286,0.8214286,0.8214286,0.78571427,0.78571427,0.78571427,0.85714287,0.75,0.89285713,0.71428573,0.78571427,0.8214286,0.89285713,0.85714287,0.8214286,0.89285713,0.8214286,0.85714287,0.96428573,0.9285714,0.8214286,0.89285713,0.89285713,0.9285714,0.89285713,0.9285714,0.8214286,0.85714287,0.8214286,0.8214286,0.89285713,0.85714287,0.9285714,0.85714287,0.96428573,0.96428573,0.9285714,0.89285713,0.96428573,0.9285714,1.0,0.9285714,0.96428573,0.96428573,0.96428573,1.0,0.96428573,1.0
+RealTabPFN-2.5 (tuned),0.10714286,0.39285713,0.51785713,0.35714287,0.42857143,0.5,0.4642857,,0.64285713,0.5714286,0.64285713,0.75,0.60714287,0.60714287,0.8214286,0.8214286,0.78571427,0.8214286,0.75,0.78571427,0.71428573,0.8214286,0.71428573,0.89285713,0.89285713,0.9285714,0.8214286,0.89285713,0.85714287,0.8214286,0.8214286,0.78571427,0.89285713,0.85714287,0.8214286,0.8214286,0.85714287,0.89285713,0.8214286,0.8214286,0.85714287,0.75,0.85714287,0.8214286,0.89285713,0.96428573,0.89285713,0.9285714,0.96428573,0.85714287,0.9285714,0.78571427,0.9285714,0.78571427,0.89285713,1.0,0.89285713,0.9285714,0.8214286,0.96428573,0.9285714,0.96428573,0.96428573,0.85714287,0.96428573,0.9285714,0.9285714,0.9285714,0.9285714,0.9285714,0.9285714,0.9285714,1.0
+RealMLP (tuned + ensembled),0.17857143,0.2857143,0.2857143,0.2857143,0.25,0.32142857,0.32142857,0.35714287,,0.39285713,0.5714286,0.60714287,0.53571427,0.53571427,0.6785714,0.71428573,0.75,0.6785714,0.78571427,0.78571427,0.60714287,0.85714287,0.64285713,0.71428573,0.85714287,0.85714287,0.6785714,0.8214286,0.85714287,0.8214286,0.71428573,0.78571427,0.8214286,0.78571427,0.71428573,0.85714287,0.78571427,0.71428573,0.8214286,0.8214286,0.8214286,0.75,0.85714287,0.85714287,0.78571427,0.78571427,0.89285713,0.8214286,0.9285714,0.8214286,0.89285713,0.75,0.75,0.75,0.9285714,0.9285714,0.85714287,0.9285714,0.78571427,0.96428573,0.89285713,0.9285714,0.96428573,0.85714287,0.96428573,0.96428573,0.9285714,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+LimiX (default),0.10714286,0.32142857,0.25,0.2857143,0.32142857,0.25,0.35714287,0.42857143,0.60714287,,0.5,0.6964286,0.53571427,0.4642857,0.6785714,0.71428573,0.60714287,0.6785714,0.6964286,0.75,0.60714287,0.6785714,0.60714287,0.75,0.75,0.6785714,0.75,0.71428573,0.78571427,0.78571427,0.75,0.75,0.8214286,0.75,0.75,0.8214286,0.78571427,0.85714287,0.85714287,0.85714287,0.78571427,0.64285713,0.8214286,0.75,0.78571427,0.85714287,0.85714287,0.8214286,0.89285713,0.85714287,0.85714287,0.83928573,0.85714287,0.8214286,0.89285713,0.91071427,0.9285714,0.85714287,0.85714287,0.85714287,0.89285713,0.9285714,0.96428573,0.89285713,0.96428573,0.96428573,0.9285714,0.89285713,0.96428573,0.96428573,0.96428573,0.9285714,1.0
+"AutoGluon 1.4 (best, 4h)",0.10714286,0.42857143,0.32142857,0.35714287,0.2857143,0.35714287,0.25,0.35714287,0.42857143,0.5,,0.60714287,0.5714286,0.5,0.60714287,0.71428573,0.6785714,0.6785714,0.75,0.75,0.60714287,0.71428573,0.60714287,0.60714287,0.75,0.71428573,0.71428573,0.78571427,0.71428573,0.6785714,0.71428573,0.75,0.71428573,0.64285713,0.6785714,0.71428573,0.78571427,0.75,0.8214286,0.85714287,0.85714287,0.78571427,0.78571427,0.85714287,0.89285713,0.78571427,0.89285713,0.78571427,0.89285713,0.89285713,0.89285713,0.71428573,0.8214286,0.75,0.89285713,0.89285713,0.9285714,0.9285714,0.8214286,0.96428573,0.96428573,0.89285713,0.96428573,0.9285714,0.96428573,0.96428573,0.89285713,0.96428573,0.96428573,0.96428573,1.0,1.0,1.0
+RealMLP (tuned),0.14285715,0.25,0.21428572,0.25,0.25,0.25,0.39285713,0.25,0.39285713,0.30357143,0.39285713,,0.4642857,0.39285713,0.53571427,0.64285713,0.5714286,0.64285713,0.5714286,0.6785714,0.53571427,0.60714287,0.60714287,0.75,0.71428573,0.75,0.5714286,0.71428573,0.75,0.75,0.71428573,0.6785714,0.71428573,0.75,0.60714287,0.75,0.71428573,0.6785714,0.75,0.75,0.78571427,0.64285713,0.75,0.75,0.78571427,0.78571427,0.75,0.85714287,0.85714287,0.78571427,0.8214286,0.75,0.85714287,0.78571427,0.9285714,0.8035714,0.89285713,0.85714287,0.78571427,0.85714287,0.9285714,0.9285714,0.9285714,0.8214286,0.96428573,0.9285714,0.89285713,0.96428573,0.9285714,0.9464286,0.9285714,0.96428573,1.0
+TabDPT-Turbo (default),0.10714286,0.25,0.32142857,0.39285713,0.42857143,0.2857143,0.2857143,0.39285713,0.4642857,0.4642857,0.42857143,0.53571427,,0.71428573,0.5,0.5,0.5,0.5,0.60714287,0.60714287,0.6785714,0.53571427,0.6964286,0.5714286,0.64285713,0.64285713,0.60714287,0.60714287,0.6785714,0.64285713,0.64285713,0.6785714,0.71428573,0.5714286,0.64285713,0.64285713,0.6785714,0.71428573,0.75,0.6785714,0.6785714,0.78571427,0.71428573,0.75,0.75,0.78571427,0.71428573,0.8214286,0.8214286,0.85714287,0.85714287,0.75,0.8214286,0.75,0.8214286,0.9285714,0.8214286,0.75,0.78571427,0.8214286,0.85714287,0.89285713,0.9285714,0.96428573,0.89285713,0.96428573,0.96428573,0.9285714,0.89285713,0.89285713,0.89285713,0.96428573,1.0
+TabDPT (tuned + ensembled),0.071428575,0.25,0.25,0.35714287,0.25,0.35714287,0.21428572,0.39285713,0.4642857,0.53571427,0.5,0.60714287,0.2857143,,0.4642857,0.4642857,0.53571427,0.53571427,0.6785714,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.64285713,0.60714287,0.5714286,0.60714287,0.6785714,0.64285713,0.60714287,0.6785714,0.75,0.64285713,0.64285713,0.60714287,0.6785714,0.6785714,0.75,0.6785714,0.64285713,0.71428573,0.64285713,0.6785714,0.64285713,0.75,0.78571427,0.85714287,0.78571427,0.75,0.75,0.78571427,0.75,0.75,0.8214286,0.96428573,0.78571427,0.71428573,0.78571427,0.8214286,0.8214286,0.96428573,0.96428573,0.9285714,0.9285714,0.96428573,0.89285713,0.85714287,0.89285713,0.89285713,0.85714287,0.9285714,1.0
+TabM (tuned + ensembled),0.10714286,0.21428572,0.21428572,0.25,0.21428572,0.2857143,0.2857143,0.17857143,0.32142857,0.32142857,0.39285713,0.4642857,0.5,0.53571427,,0.53571427,0.5,0.60714287,0.5714286,0.60714287,0.5,0.53571427,0.60714287,0.60714287,0.64285713,0.64285713,0.5714286,0.60714287,0.71428573,0.75,0.78571427,0.64285713,0.64285713,0.6785714,0.64285713,0.75,0.78571427,0.5714286,0.75,0.6785714,0.71428573,0.71428573,0.78571427,0.75,0.78571427,0.6785714,0.78571427,0.6785714,0.78571427,0.78571427,0.78571427,0.78571427,0.75,0.8214286,0.8214286,0.8214286,0.75,0.78571427,0.8214286,0.89285713,0.96428573,0.85714287,0.78571427,0.89285713,0.78571427,0.89285713,0.85714287,0.9285714,0.9285714,0.9285714,0.9285714,1.0,1.0
+TabM (tuned),0.071428575,0.14285715,0.17857143,0.21428572,0.25,0.25,0.25,0.17857143,0.2857143,0.2857143,0.2857143,0.35714287,0.5,0.53571427,0.4642857,,0.4642857,0.53571427,0.53571427,0.5714286,0.53571427,0.53571427,0.60714287,0.60714287,0.60714287,0.64285713,0.53571427,0.5714286,0.6785714,0.64285713,0.64285713,0.73214287,0.64285713,0.64285713,0.5714286,0.60714287,0.71428573,0.5714286,0.64285713,0.64285713,0.64285713,0.71428573,0.75,0.64285713,0.75,0.6785714,0.78571427,0.6785714,0.78571427,0.78571427,0.78571427,0.71428573,0.75,0.75,0.75,0.78571427,0.78571427,0.85714287,0.78571427,0.89285713,0.96428573,0.8214286,0.75,0.89285713,0.8214286,0.8214286,0.85714287,0.9285714,0.89285713,0.89285713,0.9285714,0.96428573,0.96428573
+CatBoost (tuned),0.071428575,0.25,0.17857143,0.10714286,0.21428572,0.21428572,0.25,0.21428572,0.25,0.39285713,0.32142857,0.42857143,0.5,0.4642857,0.5,0.53571427,,0.5,0.5,0.5,0.5,0.4642857,0.53571427,0.53571427,0.60714287,0.64285713,0.53571427,0.53571427,0.64285713,0.64285713,0.5714286,0.5,0.6785714,0.60714287,0.60714287,0.6785714,0.60714287,0.6785714,0.71428573,0.75,0.71428573,0.64285713,0.6785714,0.6785714,0.75,0.71428573,0.71428573,0.6785714,0.8214286,0.78571427,0.78571427,0.71428573,0.78571427,0.75,0.8214286,0.75,0.85714287,0.78571427,0.75,0.89285713,0.89285713,0.8214286,0.8214286,0.8214286,0.89285713,0.89285713,0.85714287,0.9285714,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+BetaTabPFN (default),0.071428575,0.10714286,0.21428572,0.17857143,0.21428572,0.17857143,0.14285715,0.17857143,0.32142857,0.32142857,0.32142857,0.35714287,0.5,0.4642857,0.39285713,0.4642857,0.5,,0.60714287,0.53571427,0.53571427,0.5714286,0.53571427,0.53571427,0.64285713,0.71428573,0.60714287,0.5,0.64285713,0.6785714,0.5714286,0.75,0.60714287,0.53571427,0.5714286,0.6785714,0.6785714,0.60714287,0.6785714,0.8214286,0.64285713,0.71428573,0.75,0.64285713,0.78571427,0.71428573,0.71428573,0.71428573,0.78571427,0.71428573,0.78571427,0.75,0.78571427,0.75,0.85714287,0.75,0.78571427,0.78571427,0.8214286,0.8214286,0.89285713,0.8214286,0.75,0.85714287,0.8214286,0.89285713,0.89285713,0.85714287,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+iLTM (tuned + ensembled),0.14285715,0.17857143,0.2857143,0.17857143,0.17857143,0.10714286,0.21428572,0.25,0.21428572,0.30357143,0.25,0.42857143,0.39285713,0.32142857,0.42857143,0.4642857,0.5,0.39285713,,0.5714286,0.39285713,0.42857143,0.53571427,0.60714287,0.5,0.4642857,0.5714286,0.5714286,0.60714287,0.60714287,0.53571427,0.6785714,0.6785714,0.53571427,0.60714287,0.64285713,0.6785714,0.71428573,0.71428573,0.6785714,0.6785714,0.5714286,0.75,0.78571427,0.75,0.78571427,0.78571427,0.71428573,0.8214286,0.78571427,0.8214286,0.6785714,0.85714287,0.6785714,0.8214286,0.78571427,0.85714287,0.85714287,0.78571427,0.9285714,0.9285714,0.89285713,0.9285714,0.89285713,0.9285714,0.9285714,0.9285714,0.9285714,0.89285713,0.9285714,0.9285714,0.9285714,1.0
+LightGBM (tuned + ensembled),0.071428575,0.10714286,0.10714286,0.10714286,0.14285715,0.10714286,0.17857143,0.21428572,0.21428572,0.25,0.25,0.32142857,0.39285713,0.42857143,0.39285713,0.42857143,0.5,0.4642857,0.42857143,,0.5,0.5714286,0.5714286,0.5,0.5,0.6785714,0.60714287,0.6785714,0.6785714,0.6785714,0.60714287,0.60714287,0.60714287,0.53571427,0.5714286,0.64285713,0.6785714,0.6785714,0.75,0.71428573,0.75,0.60714287,0.78571427,0.75,0.75,0.75,0.75,0.6785714,0.89285713,0.78571427,0.9285714,0.75,0.78571427,0.75,0.85714287,0.78571427,0.89285713,0.8214286,0.78571427,0.85714287,0.85714287,0.8214286,0.85714287,0.8214286,0.96428573,0.89285713,0.9285714,0.85714287,0.96428573,0.96428573,0.9285714,1.0,1.0
+TabDPT (tuned),0.035714287,0.10714286,0.25,0.2857143,0.21428572,0.2857143,0.21428572,0.2857143,0.39285713,0.39285713,0.39285713,0.4642857,0.32142857,0.2857143,0.5,0.4642857,0.5,0.4642857,0.60714287,0.5,,0.53571427,0.64285713,0.5714286,0.5714286,0.5714286,0.53571427,0.5714286,0.60714287,0.64285713,0.5714286,0.64285713,0.6785714,0.5714286,0.53571427,0.60714287,0.6785714,0.5714286,0.71428573,0.5714286,0.60714287,0.78571427,0.71428573,0.60714287,0.64285713,0.6785714,0.78571427,0.78571427,0.71428573,0.75,0.64285713,0.75,0.64285713,0.75,0.8214286,0.9285714,0.75,0.71428573,0.75,0.8214286,0.8214286,0.89285713,0.89285713,0.85714287,0.85714287,0.96428573,0.85714287,0.89285713,0.85714287,0.85714287,0.85714287,0.9285714,0.96428573
+CatBoost (tuned + ensembled),0.035714287,0.17857143,0.14285715,0.071428575,0.14285715,0.14285715,0.25,0.17857143,0.14285715,0.32142857,0.2857143,0.39285713,0.4642857,0.42857143,0.4642857,0.4642857,0.53571427,0.42857143,0.5714286,0.42857143,0.4642857,,0.5,0.53571427,0.5714286,0.60714287,0.53571427,0.53571427,0.60714287,0.60714287,0.5,0.53571427,0.6785714,0.60714287,0.64285713,0.71428573,0.5714286,0.64285713,0.71428573,0.71428573,0.6785714,0.60714287,0.8214286,0.6785714,0.75,0.75,0.71428573,0.75,0.8214286,0.78571427,0.78571427,0.75,0.8214286,0.75,0.8214286,0.71428573,0.89285713,0.85714287,0.75,0.85714287,0.89285713,0.85714287,0.8214286,0.78571427,0.9285714,0.85714287,0.8214286,0.9285714,0.96428573,0.96428573,0.96428573,0.9285714,1.0
+TabSwift (default),0.10714286,0.14285715,0.14285715,0.21428572,0.25,0.21428572,0.17857143,0.2857143,0.35714287,0.39285713,0.39285713,0.39285713,0.30357143,0.42857143,0.39285713,0.39285713,0.4642857,0.4642857,0.4642857,0.42857143,0.35714287,0.5,,0.53571427,0.5,0.5,0.5714286,0.4642857,0.5,0.60714287,0.5,0.42857143,0.5714286,0.5714286,0.64285713,0.60714287,0.53571427,0.60714287,0.64285713,0.5714286,0.5,0.5714286,0.75,0.78571427,0.60714287,0.6785714,0.64285713,0.71428573,0.64285713,0.75,0.71428573,0.75,0.64285713,0.75,0.75,0.85714287,0.6785714,0.71428573,0.8214286,0.78571427,0.8214286,0.8214286,0.8214286,0.96428573,0.78571427,0.9285714,0.89285713,0.85714287,0.89285713,0.89285713,0.89285713,0.96428573,1.0
+ModernNCA (tuned),0.071428575,0.14285715,0.17857143,0.10714286,0.10714286,0.14285715,0.21428572,0.10714286,0.2857143,0.25,0.39285713,0.25,0.42857143,0.42857143,0.39285713,0.39285713,0.4642857,0.4642857,0.39285713,0.5,0.42857143,0.4642857,0.4642857,,0.53571427,0.60714287,0.5714286,0.5714286,0.5,0.5,0.5714286,0.5714286,0.5714286,0.48214287,0.60714287,0.60714287,0.64285713,0.64285713,0.5714286,0.64285713,0.60714287,0.5714286,0.6785714,0.625,0.75,0.64285713,0.78571427,0.6785714,0.78571427,0.71428573,0.75,0.75,0.71428573,0.75,0.71428573,0.78571427,0.75,0.78571427,0.85714287,0.8214286,0.8214286,0.85714287,0.8214286,0.8214286,0.89285713,0.8214286,0.89285713,0.8214286,0.89285713,0.9285714,0.89285713,0.89285713,1.0
+XGBoost (tuned + ensembled),0.071428575,0.10714286,0.10714286,0.071428575,0.10714286,0.14285715,0.21428572,0.10714286,0.14285715,0.25,0.25,0.2857143,0.35714287,0.35714287,0.35714287,0.39285713,0.39285713,0.35714287,0.5,0.5,0.42857143,0.42857143,0.5,0.4642857,,0.5,0.5,0.5714286,0.64285713,0.5714286,0.5,0.5714286,0.5714286,0.53571427,0.5,0.64285713,0.64285713,0.64285713,0.60714287,0.64285713,0.60714287,0.5714286,0.71428573,0.6785714,0.625,0.75,0.71428573,0.75,0.85714287,0.71428573,0.96428573,0.78571427,0.8214286,0.78571427,0.8214286,0.78571427,0.89285713,0.71428573,0.78571427,0.78571427,0.85714287,0.78571427,0.89285713,0.78571427,0.9285714,0.85714287,0.8214286,0.8214286,0.89285713,0.9285714,0.9285714,0.89285713,1.0
+CatBoost (default),0.10714286,0.10714286,0.10714286,0.071428575,0.10714286,0.071428575,0.17857143,0.071428575,0.14285715,0.32142857,0.2857143,0.25,0.35714287,0.39285713,0.35714287,0.35714287,0.35714287,0.2857143,0.53571427,0.32142857,0.42857143,0.39285713,0.5,0.39285713,0.5,,0.5714286,0.42857143,0.5,0.53571427,0.5,0.5,0.60714287,0.5,0.53571427,0.71428573,0.64285713,0.64285713,0.64285713,0.60714287,0.64285713,0.64285713,0.75,0.71428573,0.71428573,0.6785714,0.60714287,0.60714287,0.75,0.75,0.71428573,0.75,0.78571427,0.75,0.78571427,0.75,0.8214286,0.8214286,0.78571427,0.89285713,0.9285714,0.85714287,0.78571427,0.8214286,0.89285713,0.89285713,0.89285713,0.9285714,0.9285714,0.96428573,0.96428573,0.96428573,1.0
+xRFM (tuned + ensembled),0.071428575,0.14285715,0.14285715,0.14285715,0.10714286,0.10714286,0.17857143,0.17857143,0.32142857,0.25,0.2857143,0.42857143,0.39285713,0.42857143,0.42857143,0.4642857,0.4642857,0.39285713,0.42857143,0.39285713,0.4642857,0.4642857,0.42857143,0.42857143,0.5,0.42857143,,0.42857143,0.42857143,0.5,0.5714286,0.5,0.4642857,0.5714286,0.5,0.60714287,0.60714287,0.6785714,0.5714286,0.4642857,0.5714286,0.5714286,0.5714286,0.64285713,0.64285713,0.6785714,0.78571427,0.75,0.71428573,0.71428573,0.71428573,0.6785714,0.75,0.6785714,0.71428573,0.8214286,0.78571427,0.71428573,0.75,0.78571427,0.85714287,0.89285713,0.89285713,0.85714287,0.89285713,0.89285713,0.89285713,0.85714287,0.9285714,0.96428573,0.9285714,0.96428573,0.96428573
+LightGBM (tuned),0.071428575,0.071428575,0.071428575,0.071428575,0.10714286,0.035714287,0.17857143,0.10714286,0.17857143,0.2857143,0.21428572,0.2857143,0.39285713,0.39285713,0.39285713,0.42857143,0.4642857,0.5,0.42857143,0.32142857,0.42857143,0.4642857,0.53571427,0.42857143,0.42857143,0.5714286,0.5714286,,0.5,0.64285713,0.60714287,0.53571427,0.5,0.39285713,0.53571427,0.5714286,0.60714287,0.60714287,0.64285713,0.60714287,0.64285713,0.5714286,0.71428573,0.6785714,0.75,0.6785714,0.5714286,0.6785714,0.78571427,0.75,0.78571427,0.71428573,0.75,0.75,0.75,0.75,0.85714287,0.78571427,0.78571427,0.8214286,0.8214286,0.8214286,0.78571427,0.78571427,0.89285713,0.85714287,0.85714287,0.85714287,0.89285713,0.89285713,0.89285713,0.9285714,1.0
+XGBoost (tuned),0.10714286,0.14285715,0.10714286,0.10714286,0.10714286,0.10714286,0.21428572,0.14285715,0.14285715,0.21428572,0.2857143,0.25,0.32142857,0.32142857,0.2857143,0.32142857,0.35714287,0.35714287,0.39285713,0.32142857,0.39285713,0.39285713,0.5,0.5,0.35714287,0.5,0.5714286,0.5,,0.53571427,0.4642857,0.53571427,0.53571427,0.5,0.5,0.5714286,0.5714286,0.64285713,0.60714287,0.64285713,0.53571427,0.53571427,0.64285713,0.6785714,0.60714287,0.78571427,0.64285713,0.78571427,0.8214286,0.75,0.89285713,0.78571427,0.78571427,0.78571427,0.78571427,0.75,0.85714287,0.71428573,0.78571427,0.78571427,0.8214286,0.78571427,0.85714287,0.78571427,0.89285713,0.85714287,0.85714287,0.8214286,0.85714287,0.89285713,0.9285714,0.89285713,1.0
+EBM (tuned + ensembled),0.10714286,0.21428572,0.14285715,0.17857143,0.21428572,0.17857143,0.21428572,0.17857143,0.17857143,0.21428572,0.32142857,0.25,0.35714287,0.35714287,0.25,0.35714287,0.35714287,0.32142857,0.39285713,0.32142857,0.35714287,0.39285713,0.39285713,0.5,0.42857143,0.4642857,0.5,0.35714287,0.4642857,,0.42857143,0.42857143,0.5,0.4642857,0.5714286,0.60714287,0.4642857,0.60714287,0.71428573,0.71428573,0.60714287,0.6785714,0.5714286,0.71428573,0.64285713,0.75,0.75,0.6785714,0.71428573,0.75,0.64285713,0.6785714,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.78571427,0.75,0.85714287,0.8214286,0.8214286,0.89285713,0.78571427,0.85714287,0.89285713,0.85714287,0.96428573,0.96428573,0.96428573,0.9285714,1.0
+TabM (default),0.10714286,0.21428572,0.14285715,0.17857143,0.17857143,0.21428572,0.21428572,0.17857143,0.2857143,0.25,0.2857143,0.2857143,0.35714287,0.39285713,0.21428572,0.35714287,0.42857143,0.42857143,0.4642857,0.39285713,0.42857143,0.5,0.5,0.42857143,0.5,0.5,0.42857143,0.39285713,0.53571427,0.5714286,,0.5,0.53571427,0.5714286,0.5,0.5,0.60714287,0.5714286,0.48214287,0.5714286,0.5714286,0.60714287,0.60714287,0.6785714,0.71428573,0.60714287,0.6785714,0.64285713,0.71428573,0.78571427,0.64285713,0.64285713,0.71428573,0.6785714,0.64285713,0.71428573,0.75,0.78571427,0.78571427,0.78571427,0.9285714,0.75,0.75,0.8214286,0.78571427,0.78571427,0.78571427,0.85714287,0.9285714,0.9285714,0.89285713,0.96428573,0.9285714
+ChimeraBoost (tuned + ensembled),0.071428575,0.10714286,0.14285715,0.14285715,0.10714286,0.17857143,0.14285715,0.21428572,0.21428572,0.25,0.25,0.32142857,0.32142857,0.32142857,0.35714287,0.26785713,0.5,0.25,0.32142857,0.39285713,0.35714287,0.4642857,0.5714286,0.42857143,0.42857143,0.5,0.5,0.4642857,0.4642857,0.5714286,0.5,,0.53571427,0.5,0.5,0.60714287,0.5714286,0.5714286,0.53571427,0.60714287,0.5714286,0.60714287,0.8214286,0.64285713,0.64285713,0.5714286,0.60714287,0.60714287,0.6785714,0.64285713,0.71428573,0.75,0.71428573,0.75,0.6785714,0.71428573,0.78571427,0.75,0.78571427,0.8214286,0.8214286,0.78571427,0.75,0.8214286,0.78571427,0.8214286,0.78571427,0.78571427,0.89285713,0.9285714,0.96428573,0.9285714,1.0
+iLTM (tuned),0.071428575,0.10714286,0.05357143,0.14285715,0.10714286,0.10714286,0.25,0.10714286,0.17857143,0.17857143,0.2857143,0.2857143,0.2857143,0.25,0.35714287,0.35714287,0.32142857,0.39285713,0.32142857,0.39285713,0.32142857,0.32142857,0.42857143,0.42857143,0.42857143,0.39285713,0.53571427,0.5,0.4642857,0.5,0.4642857,0.4642857,,0.53571427,0.5714286,0.5714286,0.64285713,0.64285713,0.75,0.60714287,0.5714286,0.53571427,0.5,0.71428573,0.64285713,0.78571427,0.6785714,0.78571427,0.60714287,0.75,0.6785714,0.6785714,0.8214286,0.6785714,0.6964286,0.6785714,0.78571427,0.6785714,0.78571427,0.8214286,0.89285713,0.85714287,0.85714287,0.8214286,0.85714287,0.85714287,0.89285713,0.85714287,0.8214286,0.85714287,0.85714287,0.96428573,1.0
+ModernNCA (tuned + ensembled),0.10714286,0.071428575,0.17857143,0.10714286,0.10714286,0.071428575,0.10714286,0.14285715,0.21428572,0.25,0.35714287,0.25,0.42857143,0.35714287,0.32142857,0.35714287,0.39285713,0.4642857,0.4642857,0.4642857,0.42857143,0.39285713,0.42857143,0.51785713,0.4642857,0.5,0.42857143,0.60714287,0.5,0.53571427,0.42857143,0.5,0.4642857,,0.4642857,0.64285713,0.53571427,0.60714287,0.5714286,0.53571427,0.53571427,0.5714286,0.53571427,0.64285713,0.71428573,0.6785714,0.75,0.64285713,0.78571427,0.75,0.78571427,0.64285713,0.71428573,0.6785714,0.71428573,0.75,0.6785714,0.75,0.78571427,0.75,0.78571427,0.85714287,0.78571427,0.75,0.8214286,0.78571427,0.8035714,0.85714287,0.78571427,0.8214286,0.78571427,0.8214286,0.96428573
+SAP-RPT-OSS (default),0.10714286,0.25,0.17857143,0.14285715,0.25,0.21428572,0.2857143,0.17857143,0.2857143,0.25,0.32142857,0.39285713,0.35714287,0.35714287,0.35714287,0.42857143,0.39285713,0.42857143,0.39285713,0.42857143,0.4642857,0.35714287,0.35714287,0.39285713,0.5,0.4642857,0.5,0.4642857,0.5,0.42857143,0.5,0.5,0.42857143,0.53571427,,0.5714286,0.5,0.53571427,0.60714287,0.53571427,0.64285713,0.5714286,0.53571427,0.71428573,0.6785714,0.64285713,0.60714287,0.6785714,0.60714287,0.71428573,0.64285713,0.60714287,0.64285713,0.60714287,0.6785714,0.64285713,0.6785714,0.75,0.78571427,0.78571427,0.78571427,0.78571427,0.78571427,0.8214286,0.78571427,0.8214286,0.71428573,0.8214286,0.89285713,0.85714287,0.8214286,0.85714287,1.0
+RealMLP (default),0.14285715,0.10714286,0.14285715,0.17857143,0.17857143,0.14285715,0.21428572,0.17857143,0.14285715,0.17857143,0.2857143,0.25,0.35714287,0.39285713,0.25,0.39285713,0.32142857,0.32142857,0.35714287,0.35714287,0.39285713,0.2857143,0.39285713,0.39285713,0.35714287,0.2857143,0.39285713,0.42857143,0.42857143,0.39285713,0.5,0.39285713,0.42857143,0.35714287,0.42857143,,0.39285713,0.53571427,0.4642857,0.53571427,0.60714287,0.60714287,0.5,0.60714287,0.64285713,0.60714287,0.6785714,0.6785714,0.6785714,0.64285713,0.60714287,0.64285713,0.6785714,0.71428573,0.71428573,0.6785714,0.6785714,0.6785714,0.78571427,0.71428573,0.89285713,0.78571427,0.78571427,0.9285714,0.78571427,0.89285713,0.89285713,0.89285713,0.89285713,0.89285713,0.96428573,0.9285714,1.0
+ChimeraBoost (tuned),0.10714286,0.14285715,0.14285715,0.10714286,0.10714286,0.17857143,0.17857143,0.14285715,0.21428572,0.21428572,0.21428572,0.2857143,0.32142857,0.32142857,0.21428572,0.2857143,0.39285713,0.32142857,0.32142857,0.32142857,0.32142857,0.42857143,0.4642857,0.35714287,0.35714287,0.35714287,0.39285713,0.39285713,0.42857143,0.53571427,0.39285713,0.42857143,0.35714287,0.4642857,0.5,0.60714287,,0.5,0.5714286,0.5714286,0.5,0.60714287,0.6785714,0.64285713,0.5714286,0.53571427,0.53571427,0.64285713,0.71428573,0.6785714,0.75,0.71428573,0.64285713,0.71428573,0.71428573,0.71428573,0.78571427,0.71428573,0.78571427,0.78571427,0.8214286,0.78571427,0.78571427,0.85714287,0.71428573,0.78571427,0.75,0.78571427,0.8214286,0.8214286,0.9285714,0.89285713,1.0
+xRFM (tuned),0.071428575,0.071428575,0.071428575,0.071428575,0.17857143,0.14285715,0.10714286,0.10714286,0.2857143,0.14285715,0.25,0.32142857,0.2857143,0.32142857,0.42857143,0.42857143,0.32142857,0.39285713,0.2857143,0.32142857,0.42857143,0.35714287,0.39285713,0.35714287,0.35714287,0.35714287,0.32142857,0.39285713,0.35714287,0.39285713,0.42857143,0.42857143,0.35714287,0.39285713,0.4642857,0.4642857,0.5,,0.42857143,0.5,0.4642857,0.5,0.5,0.5714286,0.60714287,0.71428573,0.5714286,0.75,0.5714286,0.60714287,0.64285713,0.64285713,0.71428573,0.6785714,0.75,0.75,0.75,0.60714287,0.71428573,0.71428573,0.85714287,0.96428573,0.89285713,0.78571427,0.8214286,0.8214286,0.89285713,0.89285713,0.85714287,0.8214286,0.85714287,0.96428573,1.0
+EBM (default),0.10714286,0.14285715,0.10714286,0.14285715,0.14285715,0.17857143,0.14285715,0.17857143,0.17857143,0.14285715,0.17857143,0.25,0.25,0.25,0.25,0.35714287,0.2857143,0.32142857,0.2857143,0.25,0.2857143,0.2857143,0.35714287,0.42857143,0.39285713,0.35714287,0.42857143,0.35714287,0.39285713,0.2857143,0.51785713,0.4642857,0.25,0.42857143,0.39285713,0.53571427,0.42857143,0.5714286,,0.5,0.53571427,0.5,0.53571427,0.64285713,0.64285713,0.6785714,0.64285713,0.6785714,0.53571427,0.6785714,0.53571427,0.6785714,0.6785714,0.6785714,0.71428573,0.71428573,0.75,0.78571427,0.75,0.8214286,0.85714287,0.78571427,0.8214286,0.85714287,0.71428573,0.85714287,0.89285713,0.89285713,0.9285714,0.9285714,0.96428573,0.9285714,1.0
+EBM (tuned),0.10714286,0.14285715,0.071428575,0.17857143,0.17857143,0.10714286,0.17857143,0.17857143,0.17857143,0.14285715,0.14285715,0.25,0.32142857,0.32142857,0.32142857,0.35714287,0.25,0.17857143,0.32142857,0.2857143,0.42857143,0.2857143,0.42857143,0.35714287,0.35714287,0.39285713,0.53571427,0.39285713,0.35714287,0.2857143,0.42857143,0.39285713,0.39285713,0.4642857,0.4642857,0.4642857,0.42857143,0.5,0.5,,0.5714286,0.60714287,0.5,0.60714287,0.5714286,0.60714287,0.6785714,0.64285713,0.5714286,0.6785714,0.5714286,0.6785714,0.6785714,0.6785714,0.71428573,0.6785714,0.6785714,0.71428573,0.78571427,0.71428573,0.8214286,0.85714287,0.8214286,0.85714287,0.78571427,0.8214286,0.89285713,0.8214286,0.89285713,0.9285714,0.89285713,0.9285714,1.0
+TorchMLP (tuned + ensembled),0.035714287,0.071428575,0.071428575,0.14285715,0.10714286,0.071428575,0.10714286,0.14285715,0.17857143,0.21428572,0.14285715,0.21428572,0.32142857,0.35714287,0.2857143,0.35714287,0.2857143,0.35714287,0.32142857,0.25,0.39285713,0.32142857,0.5,0.39285713,0.39285713,0.35714287,0.42857143,0.35714287,0.4642857,0.39285713,0.42857143,0.42857143,0.42857143,0.4642857,0.35714287,0.39285713,0.5,0.53571427,0.4642857,0.42857143,,0.5714286,0.5,0.60714287,0.71428573,0.60714287,0.5714286,0.60714287,0.71428573,0.6785714,0.71428573,0.60714287,0.6785714,0.64285713,0.71428573,0.6785714,0.75,0.6785714,0.6785714,0.78571427,0.9285714,0.78571427,0.75,0.89285713,0.71428573,0.85714287,0.8214286,0.85714287,0.85714287,0.89285713,0.96428573,0.9285714,0.96428573
+TabDPT (default),0.035714287,0.14285715,0.14285715,0.25,0.21428572,0.25,0.17857143,0.25,0.25,0.35714287,0.21428572,0.35714287,0.21428572,0.2857143,0.2857143,0.2857143,0.35714287,0.2857143,0.42857143,0.39285713,0.21428572,0.39285713,0.42857143,0.42857143,0.42857143,0.35714287,0.42857143,0.42857143,0.4642857,0.32142857,0.39285713,0.39285713,0.4642857,0.42857143,0.42857143,0.39285713,0.39285713,0.5,0.5,0.39285713,0.42857143,,0.53571427,0.5714286,0.4642857,0.53571427,0.64285713,0.64285713,0.60714287,0.75,0.64285713,0.60714287,0.60714287,0.64285713,0.6785714,0.78571427,0.64285713,0.6785714,0.71428573,0.75,0.71428573,0.78571427,0.85714287,0.8214286,0.75,0.8214286,0.85714287,0.75,0.71428573,0.75,0.8214286,0.9285714,0.96428573
+ChimeraBoost (default),0.071428575,0.071428575,0.071428575,0.14285715,0.071428575,0.10714286,0.14285715,0.14285715,0.14285715,0.17857143,0.21428572,0.25,0.2857143,0.35714287,0.21428572,0.25,0.32142857,0.25,0.25,0.21428572,0.2857143,0.17857143,0.25,0.32142857,0.2857143,0.25,0.42857143,0.2857143,0.35714287,0.42857143,0.39285713,0.17857143,0.5,0.4642857,0.4642857,0.5,0.32142857,0.5,0.4642857,0.5,0.5,0.4642857,,0.5714286,0.42857143,0.53571427,0.5714286,0.53571427,0.5714286,0.60714287,0.5714286,0.6785714,0.64285713,0.75,0.6785714,0.64285713,0.71428573,0.64285713,0.78571427,0.71428573,0.75,0.71428573,0.71428573,0.78571427,0.64285713,0.75,0.75,0.75,0.8214286,0.8214286,0.96428573,0.9285714,1.0
+FastaiMLP (tuned + ensembled),0.10714286,0.14285715,0.035714287,0.14285715,0.17857143,0.10714286,0.035714287,0.17857143,0.14285715,0.25,0.14285715,0.25,0.25,0.32142857,0.25,0.35714287,0.32142857,0.35714287,0.21428572,0.25,0.39285713,0.32142857,0.21428572,0.375,0.32142857,0.2857143,0.35714287,0.32142857,0.32142857,0.2857143,0.32142857,0.35714287,0.2857143,0.35714287,0.2857143,0.39285713,0.35714287,0.42857143,0.35714287,0.39285713,0.39285713,0.42857143,0.42857143,,0.53571427,0.5714286,0.5714286,0.5714286,0.5,0.71428573,0.5,0.60714287,0.53571427,0.60714287,0.6785714,0.6785714,0.60714287,0.6785714,0.6785714,0.71428573,0.71428573,0.75,0.75,0.85714287,0.71428573,0.96428573,0.78571427,0.85714287,0.78571427,0.8214286,0.85714287,0.89285713,0.96428573
+TorchMLP (tuned),0.035714287,0.071428575,0.10714286,0.14285715,0.14285715,0.071428575,0.071428575,0.10714286,0.21428572,0.21428572,0.10714286,0.21428572,0.25,0.35714287,0.21428572,0.25,0.25,0.21428572,0.25,0.25,0.35714287,0.25,0.39285713,0.25,0.375,0.2857143,0.35714287,0.25,0.39285713,0.35714287,0.2857143,0.35714287,0.35714287,0.2857143,0.32142857,0.35714287,0.42857143,0.39285713,0.35714287,0.42857143,0.2857143,0.53571427,0.5714286,0.4642857,,0.51785713,0.5714286,0.5714286,0.5,0.5714286,0.5535714,0.6785714,0.64285713,0.64285713,0.64285713,0.64285713,0.64285713,0.6785714,0.75,0.71428573,0.89285713,0.75,0.71428573,0.89285713,0.64285713,0.78571427,0.85714287,0.78571427,0.78571427,0.8214286,0.89285713,0.9285714,0.96428573
+ExtraTrees (tuned + ensembled),0.071428575,0.035714287,0.14285715,0.071428575,0.10714286,0.10714286,0.17857143,0.035714287,0.21428572,0.14285715,0.21428572,0.21428572,0.21428572,0.25,0.32142857,0.32142857,0.2857143,0.2857143,0.21428572,0.25,0.32142857,0.25,0.32142857,0.35714287,0.25,0.32142857,0.32142857,0.32142857,0.21428572,0.25,0.39285713,0.42857143,0.21428572,0.32142857,0.35714287,0.39285713,0.4642857,0.2857143,0.32142857,0.39285713,0.39285713,0.4642857,0.4642857,0.42857143,0.48214287,,0.4642857,0.6964286,0.4642857,0.4642857,0.53571427,0.60714287,0.64285713,0.64285713,0.53571427,0.64285713,0.75,0.60714287,0.71428573,0.6785714,0.64285713,0.75,0.89285713,0.75,0.8214286,0.71428573,0.9285714,0.71428573,0.78571427,0.8214286,0.78571427,0.9285714,1.0
+ModernNCA (default),0.071428575,0.071428575,0.071428575,0.14285715,0.071428575,0.071428575,0.10714286,0.10714286,0.10714286,0.14285715,0.10714286,0.25,0.2857143,0.21428572,0.21428572,0.21428572,0.2857143,0.2857143,0.21428572,0.25,0.21428572,0.2857143,0.35714287,0.21428572,0.2857143,0.39285713,0.21428572,0.42857143,0.35714287,0.25,0.32142857,0.39285713,0.32142857,0.25,0.39285713,0.32142857,0.4642857,0.42857143,0.35714287,0.32142857,0.42857143,0.35714287,0.42857143,0.42857143,0.42857143,0.53571427,,0.5,0.4642857,0.53571427,0.5,0.71428573,0.64285713,0.71428573,0.64285713,0.64285713,0.71428573,0.64285713,0.71428573,0.6785714,0.78571427,0.64285713,0.75,0.71428573,0.78571427,0.75,0.78571427,0.78571427,0.85714287,0.85714287,0.85714287,0.89285713,0.96428573
+ExtraTrees (tuned),0.071428575,0.035714287,0.14285715,0.071428575,0.10714286,0.10714286,0.10714286,0.071428575,0.17857143,0.17857143,0.21428572,0.14285715,0.17857143,0.14285715,0.32142857,0.32142857,0.32142857,0.2857143,0.2857143,0.32142857,0.21428572,0.25,0.2857143,0.32142857,0.25,0.39285713,0.25,0.32142857,0.21428572,0.32142857,0.35714287,0.39285713,0.21428572,0.35714287,0.32142857,0.32142857,0.35714287,0.25,0.32142857,0.35714287,0.39285713,0.35714287,0.4642857,0.42857143,0.42857143,0.30357143,0.5,,0.4642857,0.4642857,0.4642857,0.64285713,0.4642857,0.64285713,0.53571427,0.64285713,0.64285713,0.60714287,0.75,0.60714287,0.60714287,0.64285713,0.78571427,0.71428573,0.78571427,0.6785714,0.85714287,0.6785714,0.6785714,0.6785714,0.78571427,0.96428573,0.96428573
+XGBoost (default),0.035714287,0.035714287,0.035714287,0.0,0.071428575,0.035714287,0.071428575,0.035714287,0.071428575,0.10714286,0.10714286,0.14285715,0.17857143,0.21428572,0.21428572,0.21428572,0.17857143,0.21428572,0.17857143,0.10714286,0.2857143,0.17857143,0.35714287,0.21428572,0.14285715,0.25,0.2857143,0.21428572,0.17857143,0.2857143,0.2857143,0.32142857,0.39285713,0.21428572,0.39285713,0.32142857,0.2857143,0.42857143,0.4642857,0.42857143,0.2857143,0.39285713,0.42857143,0.5,0.5,0.53571427,0.53571427,0.53571427,,0.5714286,0.60714287,0.5714286,0.53571427,0.5714286,0.6785714,0.5714286,0.5714286,0.60714287,0.6785714,0.6785714,0.6785714,0.75,0.6785714,0.6785714,0.75,0.75,0.71428573,0.78571427,0.78571427,0.8214286,0.8214286,0.8214286,0.9285714
+FastaiMLP (tuned),0.10714286,0.071428575,0.0,0.10714286,0.10714286,0.14285715,0.10714286,0.14285715,0.17857143,0.14285715,0.10714286,0.21428572,0.14285715,0.25,0.21428572,0.21428572,0.21428572,0.2857143,0.21428572,0.21428572,0.25,0.21428572,0.25,0.2857143,0.2857143,0.25,0.2857143,0.25,0.25,0.25,0.21428572,0.35714287,0.25,0.25,0.2857143,0.35714287,0.32142857,0.39285713,0.32142857,0.32142857,0.32142857,0.25,0.39285713,0.2857143,0.42857143,0.53571427,0.4642857,0.53571427,0.42857143,,0.4642857,0.5714286,0.53571427,0.5714286,0.60714287,0.5714286,0.5714286,0.53571427,0.64285713,0.60714287,0.64285713,0.6785714,0.71428573,0.75,0.64285713,0.78571427,0.75,0.71428573,0.75,0.75,0.8214286,0.85714287,0.9285714
+LightGBM (default),0.071428575,0.071428575,0.035714287,0.035714287,0.035714287,0.071428575,0.071428575,0.071428575,0.10714286,0.14285715,0.10714286,0.17857143,0.14285715,0.25,0.21428572,0.21428572,0.21428572,0.21428572,0.17857143,0.071428575,0.35714287,0.21428572,0.2857143,0.25,0.035714287,0.2857143,0.2857143,0.21428572,0.10714286,0.35714287,0.35714287,0.2857143,0.32142857,0.21428572,0.35714287,0.39285713,0.25,0.35714287,0.4642857,0.42857143,0.2857143,0.35714287,0.42857143,0.5,0.44642857,0.4642857,0.5,0.53571427,0.39285713,0.53571427,,0.60714287,0.53571427,0.60714287,0.6785714,0.53571427,0.5714286,0.60714287,0.6785714,0.60714287,0.64285713,0.71428573,0.71428573,0.71428573,0.6785714,0.75,0.6785714,0.71428573,0.71428573,0.71428573,0.85714287,0.89285713,0.9285714
+TabSTAR (tuned),0.10714286,0.17857143,0.14285715,0.21428572,0.17857143,0.21428572,0.17857143,0.21428572,0.25,0.16071428,0.2857143,0.25,0.25,0.21428572,0.21428572,0.2857143,0.2857143,0.25,0.32142857,0.25,0.25,0.25,0.25,0.25,0.21428572,0.25,0.32142857,0.2857143,0.21428572,0.32142857,0.35714287,0.25,0.32142857,0.35714287,0.39285713,0.35714287,0.2857143,0.35714287,0.32142857,0.32142857,0.39285713,0.39285713,0.32142857,0.39285713,0.32142857,0.39285713,0.2857143,0.35714287,0.42857143,0.42857143,0.39285713,,0.4642857,0.5535714,0.39285713,0.42857143,0.4642857,0.5714286,0.9285714,0.5714286,0.53571427,0.64285713,0.64285713,0.5714286,0.6785714,0.64285713,0.64285713,0.60714287,0.6785714,0.71428573,0.75,0.8214286,0.9285714
+RandomForest (tuned + ensembled),0.071428575,0.071428575,0.10714286,0.071428575,0.071428575,0.071428575,0.14285715,0.071428575,0.25,0.14285715,0.17857143,0.14285715,0.17857143,0.25,0.25,0.25,0.21428572,0.21428572,0.14285715,0.21428572,0.35714287,0.17857143,0.35714287,0.2857143,0.17857143,0.21428572,0.25,0.25,0.21428572,0.2857143,0.2857143,0.2857143,0.17857143,0.2857143,0.35714287,0.32142857,0.35714287,0.2857143,0.32142857,0.32142857,0.32142857,0.39285713,0.35714287,0.4642857,0.35714287,0.35714287,0.35714287,0.53571427,0.4642857,0.4642857,0.4642857,0.53571427,,0.5714286,0.42857143,0.53571427,0.75,0.5,0.6785714,0.64285713,0.71428573,0.75,0.78571427,0.60714287,0.85714287,0.64285713,0.75,0.64285713,0.71428573,0.6785714,0.75,0.89285713,0.96428573
+TabSTAR (tuned + ensembled),0.14285715,0.17857143,0.14285715,0.21428572,0.17857143,0.21428572,0.17857143,0.21428572,0.25,0.17857143,0.25,0.21428572,0.25,0.25,0.17857143,0.25,0.25,0.25,0.32142857,0.25,0.25,0.25,0.25,0.25,0.21428572,0.25,0.32142857,0.25,0.21428572,0.2857143,0.32142857,0.25,0.32142857,0.32142857,0.39285713,0.2857143,0.2857143,0.32142857,0.32142857,0.32142857,0.35714287,0.35714287,0.25,0.39285713,0.35714287,0.35714287,0.2857143,0.35714287,0.42857143,0.42857143,0.39285713,0.44642857,0.42857143,,0.42857143,0.4642857,0.4642857,0.5714286,0.96428573,0.5714286,0.5714286,0.60714287,0.60714287,0.53571427,0.64285713,0.60714287,0.60714287,0.60714287,0.6785714,0.6785714,0.75,0.8214286,0.9285714
+iLTM (default),0.10714286,0.035714287,0.035714287,0.10714286,0.071428575,0.071428575,0.17857143,0.10714286,0.071428575,0.10714286,0.10714286,0.071428575,0.17857143,0.17857143,0.17857143,0.25,0.17857143,0.14285715,0.17857143,0.14285715,0.17857143,0.17857143,0.25,0.2857143,0.17857143,0.21428572,0.2857143,0.25,0.21428572,0.2857143,0.35714287,0.32142857,0.30357143,0.2857143,0.32142857,0.2857143,0.2857143,0.25,0.2857143,0.2857143,0.2857143,0.32142857,0.32142857,0.32142857,0.35714287,0.4642857,0.35714287,0.4642857,0.32142857,0.39285713,0.32142857,0.60714287,0.5714286,0.5714286,,0.5,0.5,0.53571427,0.6785714,0.5,0.64285713,0.60714287,0.60714287,0.60714287,0.60714287,0.60714287,0.75,0.60714287,0.6785714,0.6785714,0.78571427,0.8214286,0.9285714
+OrionMSP (default),0.035714287,0.0,0.0,0.035714287,0.0,0.035714287,0.10714286,0.0,0.071428575,0.08928572,0.10714286,0.19642857,0.071428575,0.035714287,0.17857143,0.21428572,0.25,0.25,0.21428572,0.21428572,0.071428575,0.2857143,0.14285715,0.21428572,0.21428572,0.25,0.17857143,0.25,0.25,0.2857143,0.2857143,0.2857143,0.32142857,0.25,0.35714287,0.32142857,0.2857143,0.25,0.2857143,0.32142857,0.32142857,0.21428572,0.35714287,0.32142857,0.35714287,0.35714287,0.35714287,0.35714287,0.42857143,0.42857143,0.4642857,0.5714286,0.4642857,0.53571427,0.5,,0.4642857,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.6785714,0.71428573,0.6785714,0.5714286,0.64285713,0.6785714,0.75,0.75,0.78571427,0.8214286,0.96428573
+RandomForest (tuned),0.071428575,0.035714287,0.035714287,0.071428575,0.071428575,0.10714286,0.14285715,0.10714286,0.14285715,0.071428575,0.071428575,0.10714286,0.17857143,0.21428572,0.25,0.21428572,0.14285715,0.21428572,0.14285715,0.10714286,0.25,0.10714286,0.32142857,0.25,0.10714286,0.17857143,0.21428572,0.14285715,0.14285715,0.2857143,0.25,0.21428572,0.21428572,0.32142857,0.32142857,0.32142857,0.21428572,0.25,0.25,0.32142857,0.25,0.35714287,0.2857143,0.39285713,0.35714287,0.25,0.2857143,0.35714287,0.42857143,0.42857143,0.42857143,0.53571427,0.25,0.53571427,0.5,0.53571427,,0.5,0.64285713,0.5714286,0.5714286,0.75,0.71428573,0.6785714,0.76785713,0.71428573,0.75,0.60714287,0.71428573,0.6785714,0.71428573,0.85714287,0.96428573
+PerpetualBooster (tuned + ensembled),0.0,0.035714287,0.071428575,0.071428575,0.10714286,0.035714287,0.071428575,0.071428575,0.071428575,0.14285715,0.071428575,0.14285715,0.25,0.2857143,0.21428572,0.14285715,0.21428572,0.21428572,0.14285715,0.17857143,0.2857143,0.14285715,0.2857143,0.21428572,0.2857143,0.17857143,0.2857143,0.21428572,0.2857143,0.2857143,0.21428572,0.25,0.32142857,0.25,0.25,0.32142857,0.2857143,0.39285713,0.21428572,0.2857143,0.32142857,0.32142857,0.35714287,0.32142857,0.32142857,0.39285713,0.35714287,0.39285713,0.39285713,0.4642857,0.39285713,0.42857143,0.5,0.42857143,0.4642857,0.42857143,0.5,,0.42857143,0.75,0.5,0.5714286,0.53571427,0.60714287,0.64285713,0.60714287,0.60714287,0.8214286,0.71428573,0.71428573,0.75,0.71428573,0.89285713
+TabSTAR (default),0.071428575,0.17857143,0.14285715,0.17857143,0.17857143,0.21428572,0.14285715,0.17857143,0.21428572,0.14285715,0.17857143,0.21428572,0.21428572,0.21428572,0.17857143,0.21428572,0.25,0.17857143,0.21428572,0.21428572,0.25,0.25,0.17857143,0.14285715,0.21428572,0.21428572,0.25,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.2857143,0.25,0.21428572,0.32142857,0.2857143,0.21428572,0.32142857,0.25,0.2857143,0.2857143,0.25,0.32142857,0.35714287,0.32142857,0.071428575,0.32142857,0.035714287,0.32142857,0.2857143,0.35714287,0.5714286,,0.5714286,0.42857143,0.42857143,0.39285713,0.4642857,0.5,0.32142857,0.5714286,0.60714287,0.5,0.5,0.60714287,0.6785714,0.9285714
+PerpetualBooster (tuned),0.0,0.0,0.0,0.035714287,0.071428575,0.0,0.035714287,0.035714287,0.035714287,0.14285715,0.035714287,0.14285715,0.17857143,0.17857143,0.10714286,0.10714286,0.10714286,0.17857143,0.071428575,0.14285715,0.17857143,0.14285715,0.21428572,0.17857143,0.21428572,0.10714286,0.21428572,0.17857143,0.21428572,0.25,0.21428572,0.17857143,0.17857143,0.25,0.21428572,0.2857143,0.21428572,0.2857143,0.17857143,0.2857143,0.21428572,0.25,0.2857143,0.2857143,0.2857143,0.32142857,0.32142857,0.39285713,0.32142857,0.39285713,0.39285713,0.42857143,0.35714287,0.42857143,0.5,0.42857143,0.42857143,0.25,0.42857143,,0.5,0.5714286,0.5,0.64285713,0.5714286,0.5714286,0.53571427,0.73214287,0.6785714,0.64285713,0.71428573,0.6785714,0.89285713
+TorchMLP (default),0.035714287,0.071428575,0.035714287,0.071428575,0.071428575,0.035714287,0.035714287,0.071428575,0.10714286,0.10714286,0.035714287,0.071428575,0.14285715,0.17857143,0.035714287,0.035714287,0.10714286,0.10714286,0.071428575,0.14285715,0.17857143,0.10714286,0.17857143,0.17857143,0.14285715,0.071428575,0.14285715,0.17857143,0.17857143,0.14285715,0.071428575,0.17857143,0.10714286,0.21428572,0.21428572,0.10714286,0.17857143,0.14285715,0.14285715,0.17857143,0.071428575,0.2857143,0.25,0.2857143,0.10714286,0.35714287,0.21428572,0.39285713,0.32142857,0.35714287,0.35714287,0.4642857,0.2857143,0.42857143,0.35714287,0.42857143,0.42857143,0.5,0.5714286,0.5,,0.5,0.60714287,0.60714287,0.53571427,0.60714287,0.60714287,0.60714287,0.60714287,0.6785714,0.75,0.78571427,0.9285714
+xRFM (default),0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.071428575,0.071428575,0.10714286,0.071428575,0.10714286,0.035714287,0.14285715,0.17857143,0.17857143,0.17857143,0.10714286,0.17857143,0.10714286,0.14285715,0.17857143,0.14285715,0.21428572,0.14285715,0.10714286,0.17857143,0.21428572,0.17857143,0.25,0.21428572,0.14285715,0.14285715,0.21428572,0.21428572,0.21428572,0.035714287,0.21428572,0.14285715,0.21428572,0.21428572,0.2857143,0.25,0.25,0.25,0.35714287,0.35714287,0.25,0.32142857,0.2857143,0.35714287,0.25,0.39285713,0.39285713,0.42857143,0.25,0.42857143,0.5714286,0.42857143,0.5,,0.53571427,0.5,0.53571427,0.4642857,0.42857143,0.5,0.5714286,0.5714286,0.64285713,0.71428573,0.89285713
+ExtraTrees (default),0.035714287,0.035714287,0.0,0.035714287,0.035714287,0.071428575,0.10714286,0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.071428575,0.035714287,0.21428572,0.25,0.17857143,0.25,0.071428575,0.14285715,0.10714286,0.17857143,0.17857143,0.17857143,0.10714286,0.21428572,0.10714286,0.21428572,0.14285715,0.17857143,0.25,0.25,0.14285715,0.21428572,0.21428572,0.21428572,0.21428572,0.10714286,0.17857143,0.17857143,0.25,0.14285715,0.2857143,0.25,0.2857143,0.10714286,0.25,0.21428572,0.32142857,0.2857143,0.2857143,0.35714287,0.21428572,0.39285713,0.39285713,0.32142857,0.2857143,0.4642857,0.60714287,0.5,0.39285713,0.4642857,,0.5,0.5714286,0.53571427,0.5,0.53571427,0.5714286,0.5714286,0.64285713,0.75,0.9285714
+TabFlex (default),0.0,0.071428575,0.0,0.14285715,0.071428575,0.10714286,0.035714287,0.14285715,0.14285715,0.10714286,0.071428575,0.17857143,0.035714287,0.071428575,0.10714286,0.10714286,0.17857143,0.14285715,0.10714286,0.17857143,0.14285715,0.21428572,0.035714287,0.17857143,0.21428572,0.17857143,0.14285715,0.21428572,0.21428572,0.10714286,0.17857143,0.17857143,0.17857143,0.25,0.17857143,0.071428575,0.14285715,0.21428572,0.14285715,0.14285715,0.10714286,0.17857143,0.21428572,0.14285715,0.10714286,0.25,0.2857143,0.2857143,0.32142857,0.25,0.2857143,0.42857143,0.39285713,0.4642857,0.39285713,0.2857143,0.32142857,0.39285713,0.53571427,0.35714287,0.39285713,0.5,0.5,,0.53571427,0.53571427,0.53571427,0.53571427,0.64285713,0.64285713,0.71428573,0.75,0.89285713
+RandomForest (default),0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.071428575,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.10714286,0.071428575,0.21428572,0.17857143,0.10714286,0.17857143,0.071428575,0.035714287,0.14285715,0.071428575,0.21428572,0.10714286,0.071428575,0.10714286,0.10714286,0.10714286,0.10714286,0.21428572,0.21428572,0.21428572,0.14285715,0.17857143,0.21428572,0.21428572,0.2857143,0.17857143,0.2857143,0.21428572,0.2857143,0.25,0.35714287,0.2857143,0.35714287,0.17857143,0.21428572,0.21428572,0.25,0.35714287,0.32142857,0.32142857,0.14285715,0.35714287,0.39285713,0.32142857,0.23214285,0.35714287,0.5,0.42857143,0.4642857,0.4642857,0.42857143,0.4642857,,0.5,0.60714287,0.5,0.64285713,0.64285713,0.6785714,0.78571427,0.96428573
+FastaiMLP (default),0.0,0.0,0.0,0.035714287,0.035714287,0.035714287,0.0,0.071428575,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.035714287,0.10714286,0.17857143,0.10714286,0.10714286,0.071428575,0.10714286,0.035714287,0.14285715,0.071428575,0.17857143,0.14285715,0.10714286,0.10714286,0.14285715,0.14285715,0.14285715,0.21428572,0.17857143,0.14285715,0.21428572,0.17857143,0.10714286,0.21428572,0.17857143,0.14285715,0.17857143,0.14285715,0.17857143,0.25,0.035714287,0.21428572,0.2857143,0.25,0.32142857,0.25,0.21428572,0.25,0.35714287,0.35714287,0.39285713,0.39285713,0.42857143,0.2857143,0.39285713,0.6785714,0.42857143,0.39285713,0.53571427,0.4642857,0.4642857,0.5,,0.5714286,0.64285713,0.5714286,0.60714287,0.64285713,0.78571427,0.9285714
+KNN (tuned + ensembled),0.071428575,0.035714287,0.071428575,0.071428575,0.035714287,0.035714287,0.071428575,0.071428575,0.071428575,0.071428575,0.10714286,0.10714286,0.035714287,0.10714286,0.14285715,0.14285715,0.14285715,0.10714286,0.071428575,0.071428575,0.14285715,0.17857143,0.10714286,0.10714286,0.17857143,0.10714286,0.10714286,0.14285715,0.14285715,0.10714286,0.21428572,0.21428572,0.10714286,0.19642857,0.2857143,0.10714286,0.25,0.10714286,0.10714286,0.10714286,0.17857143,0.14285715,0.25,0.21428572,0.14285715,0.071428575,0.21428572,0.14285715,0.2857143,0.25,0.32142857,0.35714287,0.25,0.39285713,0.25,0.35714287,0.25,0.39285713,0.42857143,0.4642857,0.39285713,0.5714286,0.5,0.4642857,0.39285713,0.42857143,,0.5,0.5714286,0.5714286,0.64285713,0.96428573,0.96428573
+PerpetualBooster (default),0.0,0.035714287,0.071428575,0.071428575,0.071428575,0.035714287,0.035714287,0.071428575,0.035714287,0.10714286,0.035714287,0.035714287,0.071428575,0.14285715,0.071428575,0.071428575,0.071428575,0.14285715,0.071428575,0.14285715,0.10714286,0.071428575,0.14285715,0.17857143,0.17857143,0.071428575,0.14285715,0.14285715,0.17857143,0.14285715,0.14285715,0.21428572,0.14285715,0.14285715,0.17857143,0.10714286,0.21428572,0.10714286,0.10714286,0.17857143,0.14285715,0.25,0.25,0.14285715,0.21428572,0.2857143,0.21428572,0.32142857,0.21428572,0.2857143,0.2857143,0.39285713,0.35714287,0.39285713,0.39285713,0.32142857,0.39285713,0.17857143,0.39285713,0.26785713,0.39285713,0.5,0.4642857,0.4642857,0.5,0.35714287,0.5,,0.5714286,0.5714286,0.60714287,0.60714287,0.85714287
+Linear (tuned + ensembled),0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.035714287,0.035714287,0.071428575,0.10714286,0.10714286,0.071428575,0.10714286,0.035714287,0.035714287,0.10714286,0.035714287,0.14285715,0.035714287,0.10714286,0.10714286,0.10714286,0.071428575,0.071428575,0.10714286,0.14285715,0.035714287,0.071428575,0.10714286,0.17857143,0.21428572,0.10714286,0.10714286,0.17857143,0.14285715,0.071428575,0.10714286,0.14285715,0.2857143,0.17857143,0.21428572,0.21428572,0.21428572,0.14285715,0.32142857,0.21428572,0.25,0.2857143,0.32142857,0.2857143,0.32142857,0.32142857,0.25,0.2857143,0.2857143,0.5,0.32142857,0.39285713,0.42857143,0.42857143,0.35714287,0.35714287,0.42857143,0.42857143,0.42857143,,0.71428573,0.71428573,0.64285713,0.85714287
+Linear (tuned),0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.035714287,0.035714287,0.05357143,0.10714286,0.10714286,0.071428575,0.10714286,0.035714287,0.035714287,0.071428575,0.035714287,0.14285715,0.035714287,0.10714286,0.071428575,0.071428575,0.035714287,0.035714287,0.10714286,0.10714286,0.035714287,0.071428575,0.071428575,0.14285715,0.17857143,0.14285715,0.10714286,0.17857143,0.17857143,0.071428575,0.071428575,0.10714286,0.25,0.17857143,0.17857143,0.17857143,0.17857143,0.14285715,0.32142857,0.17857143,0.25,0.2857143,0.2857143,0.32142857,0.32142857,0.32142857,0.25,0.32142857,0.2857143,0.5,0.35714287,0.32142857,0.42857143,0.42857143,0.35714287,0.35714287,0.39285713,0.42857143,0.42857143,0.2857143,,0.60714287,0.60714287,0.8214286
+Linear (default),0.0,0.0,0.0,0.071428575,0.035714287,0.035714287,0.0,0.071428575,0.035714287,0.035714287,0.0,0.071428575,0.10714286,0.14285715,0.071428575,0.071428575,0.035714287,0.035714287,0.071428575,0.071428575,0.14285715,0.035714287,0.10714286,0.10714286,0.071428575,0.035714287,0.071428575,0.10714286,0.071428575,0.035714287,0.10714286,0.035714287,0.14285715,0.21428572,0.17857143,0.035714287,0.071428575,0.14285715,0.035714287,0.10714286,0.035714287,0.17857143,0.035714287,0.14285715,0.10714286,0.21428572,0.14285715,0.21428572,0.17857143,0.17857143,0.14285715,0.25,0.25,0.25,0.21428572,0.21428572,0.2857143,0.25,0.39285713,0.2857143,0.25,0.35714287,0.35714287,0.2857143,0.32142857,0.35714287,0.35714287,0.39285713,0.2857143,0.39285713,,0.60714287,0.75
+KNN (tuned),0.035714287,0.035714287,0.035714287,0.071428575,0.0,0.035714287,0.035714287,0.071428575,0.035714287,0.071428575,0.0,0.035714287,0.035714287,0.071428575,0.0,0.035714287,0.035714287,0.035714287,0.071428575,0.0,0.071428575,0.071428575,0.035714287,0.10714286,0.10714286,0.035714287,0.035714287,0.071428575,0.10714286,0.071428575,0.035714287,0.071428575,0.035714287,0.17857143,0.14285715,0.071428575,0.10714286,0.035714287,0.071428575,0.071428575,0.071428575,0.071428575,0.071428575,0.10714286,0.071428575,0.071428575,0.10714286,0.035714287,0.17857143,0.14285715,0.10714286,0.17857143,0.10714286,0.17857143,0.17857143,0.17857143,0.14285715,0.2857143,0.32142857,0.32142857,0.21428572,0.2857143,0.25,0.25,0.21428572,0.21428572,0.035714287,0.39285713,0.35714287,0.39285713,0.39285713,,0.9285714
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.035714287,0.0,0.0,0.0,0.0,0.035714287,0.0,0.0,0.0,0.0,0.0,0.035714287,0.0,0.0,0.0,0.071428575,0.0,0.0,0.035714287,0.0,0.0,0.0,0.0,0.0,0.0,0.035714287,0.035714287,0.0,0.035714287,0.035714287,0.0,0.035714287,0.035714287,0.071428575,0.071428575,0.071428575,0.071428575,0.035714287,0.071428575,0.071428575,0.035714287,0.035714287,0.10714286,0.071428575,0.10714286,0.071428575,0.10714286,0.071428575,0.10714286,0.035714287,0.071428575,0.035714287,0.14285715,0.14285715,0.17857143,0.25,0.071428575,
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
index 5067c972c9968d52db51857fbe33a5b31768d4b5..83d18cff403ca45f1931e60efece9e615e00a426 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Multiclass | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..2d876dec2667bb99ac1d090d0a85d5a4d9eb71e7
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_multiclass/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1743.0,"score":0.691,"rank":7.5,"hrank":4.23,"imp":11.605,"train_time":253.44,"predict_time":3.242,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+257\/-72","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1735.0,"score":0.7,"rank":7.75,"hrank":3.08,"imp":5.495,"train_time":2793.87,"predict_time":8.928,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+423\/-159","imputed":false},{"position":2,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1692.0,"score":0.793,"rank":9.25,"hrank":1.46,"imp":7.038,"train_time":49.97,"predict_time":7.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+786\/-263","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1652.0,"score":0.611,"rank":10.81,"hrank":7.92,"imp":13.469,"train_time":2793.87,"predict_time":1.512,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+226\/-124","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1645.0,"score":0.593,"rank":11.12,"hrank":7.91,"imp":13.15,"train_time":9.9,"predict_time":0.591,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+269\/-125","imputed":false},{"position":5,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1593.0,"score":0.557,"rank":13.5,"hrank":5.73,"imp":11.341,"train_time":8.93,"predict_time":0.49,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+353\/-156","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1537.0,"score":0.462,"rank":16.44,"hrank":8.98,"imp":13.872,"train_time":9.88,"predict_time":0.825,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+264\/-111","imputed":false},{"position":7,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1521.0,"score":0.504,"rank":17.38,"hrank":6.36,"imp":15.305,"train_time":4.73,"predict_time":0.225,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+393\/-201","imputed":false},{"position":8,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1510.0,"score":0.459,"rank":18.0,"hrank":5.25,"imp":13.383,"train_time":41.12,"predict_time":8.609,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+313\/-158","imputed":false},{"position":9,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1502.0,"score":0.452,"rank":18.5,"hrank":9.19,"imp":17.683,"train_time":5547.02,"predict_time":3.267,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+185\/-107","imputed":false},{"position":10,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1470.0,"score":0.352,"rank":20.5,"hrank":16.07,"imp":16.746,"train_time":2807.98,"predict_time":1.223,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+196\/-105","imputed":false},{"position":11,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1468.0,"score":0.361,"rank":20.62,"hrank":13.32,"imp":15.961,"train_time":2807.98,"predict_time":0.179,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+227\/-98","imputed":false},{"position":12,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1458.0,"score":0.379,"rank":21.25,"hrank":9.36,"imp":18.813,"train_time":4335.39,"predict_time":21.143,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+174\/-102","imputed":false},{"position":13,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1447.0,"score":0.321,"rank":22.0,"hrank":16.89,"imp":19.687,"train_time":1280.18,"predict_time":4.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-88","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1441.0,"score":0.314,"rank":22.38,"hrank":16.76,"imp":19.765,"train_time":1280.18,"predict_time":1.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-97","imputed":false},{"position":15,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1430.0,"score":0.37,"rank":23.12,"hrank":15.05,"imp":20.16,"train_time":3796.93,"predict_time":4.545,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-121","imputed":false},{"position":16,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1421.0,"score":0.354,"rank":23.75,"hrank":11.48,"imp":20.211,"train_time":3796.93,"predict_time":0.767,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+201\/-125","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1419.0,"score":0.337,"rank":23.88,"hrank":17.69,"imp":19.268,"train_time":4335.39,"predict_time":3.527,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+188\/-119","imputed":false},{"position":18,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1393.0,"score":0.36,"rank":25.75,"hrank":18.08,"imp":22.257,"train_time":9413.13,"predict_time":343.731,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+209\/-153","imputed":false},{"position":19,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1391.0,"score":0.299,"rank":25.88,"hrank":14.92,"imp":22.455,"train_time":5.38,"predict_time":0.274,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+231\/-140","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1386.0,"score":0.238,"rank":26.25,"hrank":24.62,"imp":20.271,"train_time":2029.77,"predict_time":4.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-71","imputed":false},{"position":21,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1378.0,"score":0.244,"rank":26.75,"hrank":24.75,"imp":20.058,"train_time":2029.77,"predict_time":1.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-82","imputed":false},{"position":22,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1375.0,"score":0.311,"rank":27.0,"hrank":10.07,"imp":22.359,"train_time":2079.22,"predict_time":3.347,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+210\/-195","imputed":false},{"position":23,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1349.0,"score":0.222,"rank":28.88,"hrank":25.02,"imp":20.673,"train_time":27893.23,"predict_time":752.506,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-88","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1339.0,"score":0.228,"rank":29.56,"hrank":25.6,"imp":21.281,"train_time":6686.5,"predict_time":9.587,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-92","imputed":false},{"position":25,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1330.0,"score":0.198,"rank":30.25,"hrank":20.65,"imp":23.299,"train_time":17528.83,"predict_time":0.282,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+138\/-147","imputed":false},{"position":26,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1325.0,"score":0.204,"rank":30.56,"hrank":25.28,"imp":21.623,"train_time":6686.5,"predict_time":0.752,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+141\/-112","imputed":false},{"position":27,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1319.0,"score":0.292,"rank":31.0,"hrank":13.76,"imp":23.544,"train_time":9413.13,"predict_time":39.488,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+225\/-161","imputed":false},{"position":28,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1314.0,"score":0.264,"rank":31.38,"hrank":16.61,"imp":25.117,"train_time":14.97,"predict_time":0.624,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+190\/-151","imputed":false},{"position":29,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1312.0,"score":0.249,"rank":31.56,"hrank":15.17,"imp":23.644,"train_time":968.86,"predict_time":1.864,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+206\/-158","imputed":false},{"position":30,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1301.0,"score":0.327,"rank":32.38,"hrank":9.0,"imp":23.556,"train_time":21.85,"predict_time":1.702,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+230\/-296","imputed":false},{"position":31,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1292.0,"score":0.171,"rank":33.0,"hrank":30.01,"imp":22.329,"train_time":26.32,"predict_time":0.254,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-90","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1292.0,"score":0.241,"rank":33.0,"hrank":6.45,"imp":23.832,"train_time":2079.22,"predict_time":0.319,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+249\/-199","imputed":false},{"position":33,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1291.0,"score":0.199,"rank":33.12,"hrank":24.58,"imp":25.023,"train_time":283.65,"predict_time":4.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+153\/-145","imputed":false},{"position":34,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1289.0,"score":0.267,"rank":33.25,"hrank":20.99,"imp":23.437,"train_time":729.27,"predict_time":1.833,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+212\/-213","imputed":false},{"position":35,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1269.0,"score":0.177,"rank":34.75,"hrank":29.77,"imp":22.336,"train_time":27893.23,"predict_time":206.719,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+168\/-116","imputed":false},{"position":36,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1259.0,"score":0.18,"rank":35.5,"hrank":29.21,"imp":24.717,"train_time":17528.83,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-162","imputed":false},{"position":37,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1254.0,"score":0.139,"rank":35.88,"hrank":32.04,"imp":24.767,"train_time":3647.6,"predict_time":2.161,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+107\/-150","imputed":false},{"position":38,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1239.0,"score":0.173,"rank":37.06,"hrank":17.05,"imp":26.524,"train_time":968.86,"predict_time":0.284,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+205\/-172","imputed":false},{"position":39,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1238.0,"score":0.146,"rank":37.12,"hrank":33.46,"imp":25.261,"train_time":12.19,"predict_time":0.042,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-135","imputed":false},{"position":40,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1233.0,"score":0.166,"rank":37.5,"hrank":31.47,"imp":24.508,"train_time":1560.18,"predict_time":0.201,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+129\/-177","imputed":false},{"position":41,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1231.0,"score":0.141,"rank":37.62,"hrank":33.35,"imp":25.007,"train_time":1560.18,"predict_time":1.226,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-170","imputed":false},{"position":42,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1221.0,"score":0.198,"rank":38.38,"hrank":27.45,"imp":24.901,"train_time":729.27,"predict_time":0.325,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+195\/-183","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1221.0,"score":0.135,"rank":38.38,"hrank":21.57,"imp":26.495,"train_time":3647.6,"predict_time":0.194,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+174\/-187","imputed":false},{"position":44,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1216.0,"score":0.058,"rank":38.75,"hrank":36.08,"imp":23.692,"train_time":4.93,"predict_time":0.592,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-88","imputed":false},{"position":45,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1211.0,"score":0.135,"rank":39.12,"hrank":29.51,"imp":26.12,"train_time":5.12,"predict_time":0.444,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+171\/-151","imputed":false},{"position":46,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1211.0,"score":0.088,"rank":39.12,"hrank":34.99,"imp":27.282,"train_time":3.02,"predict_time":0.076,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-120","imputed":false},{"position":47,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1200.0,"score":0.156,"rank":39.88,"hrank":25.11,"imp":28.687,"train_time":2721.87,"predict_time":12.59,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+187\/-203","imputed":false},{"position":48,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1198.0,"score":0.171,"rank":40.0,"hrank":17.56,"imp":26.7,"train_time":46.36,"predict_time":39.803,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+262\/-186","imputed":false},{"position":49,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1145.0,"score":0.114,"rank":43.75,"hrank":31.42,"imp":28.908,"train_time":7.42,"predict_time":0.203,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+181\/-239","imputed":false},{"position":50,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1127.0,"score":0.09,"rank":45.0,"hrank":38.72,"imp":28.12,"train_time":19.13,"predict_time":6.92,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-156","imputed":false},{"position":51,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1103.0,"score":0.074,"rank":46.62,"hrank":39.7,"imp":31.537,"train_time":8.99,"predict_time":1.372,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+166\/-254","imputed":false},{"position":52,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1103.0,"score":0.068,"rank":46.62,"hrank":37.47,"imp":30.234,"train_time":2721.87,"predict_time":1.082,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+136\/-173","imputed":false},{"position":53,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1092.0,"score":0.032,"rank":47.38,"hrank":44.14,"imp":30.065,"train_time":17.24,"predict_time":0.565,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-149","imputed":false},{"position":54,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1015.0,"score":0.137,"rank":52.25,"hrank":32.33,"imp":40.36,"train_time":155.59,"predict_time":2.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+273\/-358","imputed":false},{"position":55,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1015.0,"score":0.021,"rank":52.25,"hrank":50.42,"imp":33.029,"train_time":14.37,"predict_time":0.362,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-156","imputed":false},{"position":56,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1006.0,"score":0.023,"rank":52.75,"hrank":50.1,"imp":30.304,"train_time":495.65,"predict_time":205.513,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+111\/-176","imputed":false},{"position":57,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":53.12,"hrank":50.42,"imp":37.619,"train_time":0.74,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-199","imputed":false},{"position":58,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":997.0,"score":0.035,"rank":53.31,"hrank":47.82,"imp":33.011,"train_time":44782.6,"predict_time":7.958,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-251","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":981.0,"score":0.013,"rank":54.19,"hrank":50.6,"imp":33.318,"train_time":44782.6,"predict_time":36.102,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-232","imputed":false},{"position":60,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":969.0,"score":0.0,"rank":54.88,"hrank":53.32,"imp":35.273,"train_time":20.95,"predict_time":3.326,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-162","imputed":false},{"position":61,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":931.0,"score":0.009,"rank":56.88,"hrank":54.14,"imp":38.629,"train_time":5.16,"predict_time":0.787,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-193","imputed":false},{"position":62,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":919.0,"score":0.0,"rank":57.44,"hrank":55.39,"imp":43.471,"train_time":909.8,"predict_time":0.609,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-267","imputed":false},{"position":63,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":916.0,"score":0.02,"rank":57.62,"hrank":53.45,"imp":37.184,"train_time":8.38,"predict_time":0.659,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-231","imputed":false},{"position":64,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":894.0,"score":0.0,"rank":58.69,"hrank":56.63,"imp":44.403,"train_time":909.8,"predict_time":0.107,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-252","imputed":false},{"position":65,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":817.0,"score":0.0,"rank":62.0,"hrank":61.27,"imp":53.303,"train_time":1308.45,"predict_time":9.593,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-196","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":787.0,"score":0.0,"rank":63.12,"hrank":62.34,"imp":44.125,"train_time":666.72,"predict_time":7.823,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-226","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":784.0,"score":0.083,"rank":63.25,"hrank":50.27,"imp":51.863,"train_time":155.59,"predict_time":0.191,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+298\/-479","imputed":false},{"position":68,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":766.0,"score":0.0,"rank":63.88,"hrank":63.4,"imp":43.127,"train_time":4.67,"predict_time":0.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-218","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":738.0,"score":0.0,"rank":64.81,"hrank":64.35,"imp":56.997,"train_time":1308.45,"predict_time":1.071,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-192","imputed":false},{"position":70,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":716.0,"score":0.0,"rank":65.5,"hrank":64.89,"imp":50.164,"train_time":4.62,"predict_time":0.135,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-269","imputed":false},{"position":71,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":669.0,"score":0.0,"rank":66.81,"hrank":66.12,"imp":62.362,"train_time":218.36,"predict_time":0.46,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-333","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":189.0,"score":0.0,"rank":72.5,"hrank":72.49,"imp":71.869,"train_time":0.52,"predict_time":0.075,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-699","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/pareto_front_explorer.html
index aba1cdf98654831e3c122114b2c3c4b191eef7f5..4411ceeb3d1827ff22911072abd2f7754b706995 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.1027625141,"imp":25.0233934223,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2536073213,"imp":22.3290802602,"elo":1292.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1791910888,"imp":15.9607128727,"elo":1467.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.2234800966,"imp":16.7460632673,"elo":1469.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2033347873,"imp":28.9079076305,"elo":1145.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2010198493,"imp":24.5082802321,"elo":1232.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2260580575,"imp":25.0069064835,"elo":1231.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0424445763,"imp":25.2606650045,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0252302724,"imp":24.7166459855,"elo":1259.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.2817546014,"imp":23.2989953026,"elo":1329.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7866568202,"imp":38.6291487203,"elo":930.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2836586346,"imp":26.5237510309,"elo":1238.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.8636375274,"imp":23.6440186663,"elo":1311.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6590449294,"imp":37.1837118215,"elo":915.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.0818469447,"imp":30.2335758512,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5900916415,"imp":28.6871013436,"elo":1200.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.4437307263,"imp":26.1196336464,"elo":1210.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.0508933307,"imp":19.7650301564,"elo":1441.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.0776987398,"imp":19.6873087712,"elo":1446.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.6091982733,"imp":13.3825747475,"elo":1510.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5648398113,"imp":30.064936048,"elo":1092.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.751659725,"imp":21.6228300791,"elo":1325.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5865347302,"imp":21.2813099415,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3262335461,"imp":35.2726368712,"elo":968.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4602862659,"imp":62.362333418,"elo":668.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0713018657,"imp":56.9967733472,"elo":737.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":9.5934077851,"imp":53.3028895617,"elo":816.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.147930342,"imp":37.6187219692,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3251277774,"imp":24.9008491087,"elo":1220.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.8330056127,"imp":23.4369156098,"elo":1289.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":6.920194808,"imp":28.1199880238,"elo":1127.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.5270444068,"imp":19.2681742676,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.142857233,"imp":18.8131706763,"elo":1458.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8250564093,"imp":13.8715452955,"elo":1537.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.5121585317,"imp":13.4692880907,"elo":1652.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9280086995,"imp":5.4947133355,"elo":1735.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7021367127,"imp":23.5561919407,"elo":1300.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8027161329,"imp":26.6996058482,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4876798649,"imp":23.5440960871,"elo":1319.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":343.7310101457,"imp":22.2571294857,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2735262831,"imp":22.4554958659,"elo":1390.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.6889691672,"imp":7.0380391136,"elo":1691.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.8545350274,"imp":43.1266606509,"elo":765.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2245693736,"imp":15.304534586,"elo":1520.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.624076982,"imp":25.1172532623,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7671600452,"imp":20.2113064398,"elo":1421.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.5447590803,"imp":20.1604816148,"elo":1430.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5909006176,"imp":13.1503663168,"elo":1644.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4902017202,"imp":11.3414172871,"elo":1593.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8225789867,"imp":44.12502013,"elo":787.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.9584829717,"imp":33.0105868135,"elo":996.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.1023197407,"imp":33.3175339836,"elo":981.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0760162666,"imp":27.2824558285,"elo":1210.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3624794907,"imp":33.0285558796,"elo":1015.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.194090358,"imp":26.495255967,"elo":1220.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1614416305,"imp":24.7665449214,"elo":1254.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.5916710541,"imp":23.6920112547,"elo":1215.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0396563215,"imp":20.0582538038,"elo":1378.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":4.1126713529,"imp":20.270846639,"elo":1385.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":205.5125020069,"imp":30.304216606,"elo":1006.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":206.7186712774,"imp":22.3359735156,"elo":1269.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":752.505846815,"imp":20.672850816,"elo":1348.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.371587587,"imp":31.5368880354,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.3186223488,"imp":23.8320825397,"elo":1292.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.3466778136,"imp":22.3592603663,"elo":1375.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.1027625141,"x_train":283.646286296,"imp":25.0233934223,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2536073213,"x_train":26.3194957247,"imp":22.3290802602,"elo":1292.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1791910888,"x_train":2807.9829916593,"imp":15.9607128727,"elo":1467.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.2234800966,"x_train":2807.9829916593,"imp":16.7460632673,"elo":1469.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2033347873,"x_train":7.4221463904,"imp":28.9079076305,"elo":1145.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2010198493,"x_train":1560.1761733922,"imp":24.5082802321,"elo":1232.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2260580575,"x_train":1560.1761733922,"imp":25.0069064835,"elo":1231.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0424445763,"x_train":12.1942564441,"imp":25.2606650045,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0252302724,"x_train":17528.8314911942,"imp":24.7166459855,"elo":1259.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.2817546014,"x_train":17528.8314911942,"imp":23.2989953026,"elo":1329.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7866568202,"x_train":5.1558379712,"imp":38.6291487203,"elo":930.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2836586346,"x_train":968.8626049308,"imp":26.5237510309,"elo":1238.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.8636375274,"x_train":968.8626049308,"imp":23.6440186663,"elo":1311.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6590449294,"x_train":8.3755198931,"imp":37.1837118215,"elo":915.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.0818469447,"x_train":2721.8722746731,"imp":30.2335758512,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5900916415,"x_train":2721.8722746731,"imp":28.6871013436,"elo":1200.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.4437307263,"x_train":5.1200069793,"imp":26.1196336464,"elo":1210.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.0508933307,"x_train":1280.184939635,"imp":19.7650301564,"elo":1441.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.0776987398,"x_train":1280.184939635,"imp":19.6873087712,"elo":1446.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.6091982733,"x_train":41.1234544852,"imp":13.3825747475,"elo":1510.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5648398113,"x_train":17.2441843169,"imp":30.064936048,"elo":1092.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.751659725,"x_train":6686.5004259785,"imp":21.6228300791,"elo":1325.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5865347302,"x_train":6686.5004259785,"imp":21.2813099415,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3262335461,"x_train":20.9501855302,"imp":35.2726368712,"elo":968.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4602862659,"x_train":218.361445605,"imp":62.362333418,"elo":668.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0713018657,"x_train":1308.4507704502,"imp":56.9967733472,"elo":737.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":9.5934077851,"x_train":1308.4507704502,"imp":53.3028895617,"elo":816.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.147930342,"x_train":0.7350368805,"imp":37.6187219692,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3251277774,"x_train":729.2747129799,"imp":24.9008491087,"elo":1220.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.8330056127,"x_train":729.2747129799,"imp":23.4369156098,"elo":1289.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":6.920194808,"x_train":19.1299431712,"imp":28.1199880238,"elo":1127.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.5270444068,"x_train":4335.3941369138,"imp":19.2681742676,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.142857233,"x_train":4335.3941369138,"imp":18.8131706763,"elo":1458.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8250564093,"x_train":9.8834251377,"imp":13.8715452955,"elo":1537.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.5121585317,"x_train":2793.8656494383,"imp":13.4692880907,"elo":1652.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9280086995,"x_train":2793.8656494383,"imp":5.4947133355,"elo":1735.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7021367127,"x_train":21.8451817987,"imp":23.5561919407,"elo":1300.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8027161329,"x_train":46.3634612718,"imp":26.6996058482,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4876798649,"x_train":9413.1269943826,"imp":23.5440960871,"elo":1319.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":343.7310101457,"x_train":9413.1269943826,"imp":22.2571294857,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2735262831,"x_train":5.378689382,"imp":22.4554958659,"elo":1390.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.6889691672,"x_train":49.9745503327,"imp":7.0380391136,"elo":1691.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.8545350274,"x_train":4.6668652982,"imp":43.1266606509,"elo":765.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2245693736,"x_train":4.72524967,"imp":15.304534586,"elo":1520.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.624076982,"x_train":14.968259658,"imp":25.1172532623,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7671600452,"x_train":3796.9310139525,"imp":20.2113064398,"elo":1421.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.5447590803,"x_train":3796.9310139525,"imp":20.1604816148,"elo":1430.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5909006176,"x_train":9.9022651232,"imp":13.1503663168,"elo":1644.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4902017202,"x_train":8.9308378882,"imp":11.3414172871,"elo":1593.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8225789867,"x_train":666.7207045489,"imp":44.12502013,"elo":787.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.9584829717,"x_train":44782.5951977839,"imp":33.0105868135,"elo":996.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.1023197407,"x_train":44782.5951977839,"imp":33.3175339836,"elo":981.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0760162666,"x_train":3.0219759465,"imp":27.2824558285,"elo":1210.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3624794907,"x_train":14.370479118,"imp":33.0285558796,"elo":1015.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.194090358,"x_train":3647.6027397269,"imp":26.495255967,"elo":1220.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1614416305,"x_train":3647.6027397269,"imp":24.7665449214,"elo":1254.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.5916710541,"x_train":4.9305373226,"imp":23.6920112547,"elo":1215.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0396563215,"x_train":2029.7698407236,"imp":20.0582538038,"elo":1378.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":4.1126713529,"x_train":2029.7698407236,"imp":20.270846639,"elo":1385.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":205.5125020069,"x_train":495.6488461748,"imp":30.304216606,"elo":1006.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":206.7186712774,"x_train":27893.2292728524,"imp":22.3359735156,"elo":1269.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":752.505846815,"x_train":27893.2292728524,"imp":20.672850816,"elo":1348.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.371587587,"x_train":8.9914037311,"imp":31.5368880354,"elo":1103.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.3186223488,"x_train":2079.2181272278,"imp":23.8320825397,"elo":1292.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.3466778136,"x_train":2079.2181272278,"imp":22.3592603663,"elo":1375.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
index e196c8b6e2a293640b3d630f76772c8c43e26a65..08d5bdc03d2b1e83eeab65d1d42415151f1b9722 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":283.646286296,"imp":25.0435111439,"elo":1283.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":26.3194957247,"imp":22.3493167591,"elo":1288.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":52.2760454584,"imp":20.1500037908,"elo":1376.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":120.7421051485,"imp":19.080273336,"elo":1419.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":216.6422936563,"imp":18.4288826138,"elo":1444.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":415.1322548777,"imp":17.9118649582,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":727.7924472465,"imp":17.907424396,"elo":1458.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1402.2322064531,"imp":17.2185817434,"elo":1465.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2094.5870262909,"imp":17.022905266,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":2807.9829916593,"imp":16.766699508,"elo":1468.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":7.4221463904,"imp":28.9264110498,"elo":1135.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":15.2214794665,"imp":28.3548505682,"elo":1146.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":38.0924279868,"imp":27.8896719359,"elo":1159.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":76.6924688657,"imp":27.2792233321,"elo":1174.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":193.4343472071,"imp":26.2870986179,"elo":1190.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":388.7016596181,"imp":25.8410342794,"elo":1207.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":776.6628709802,"imp":25.3611606878,"elo":1215.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1163.7786661849,"imp":25.1575788624,"elo":1220.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1560.1761733922,"imp":25.0263500682,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":12.1942564441,"imp":25.280086947,"elo":1219.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":81.2230381354,"imp":24.9171703738,"elo":1228.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":314.4258454848,"imp":24.665046851,"elo":1246.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":700.8132948146,"imp":24.6110063021,"elo":1250.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":1955.0904519336,"imp":24.3733254772,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":4550.043096608,"imp":24.1269970022,"elo":1269.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":8811.0765244269,"imp":23.6638108139,"elo":1300.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":13221.5244172938,"imp":23.4665104829,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":17528.8314911942,"imp":23.3215821166,"elo":1316.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":5.1558379712,"imp":38.6548685735,"elo":917.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.9671223858,"imp":34.5355870581,"elo":1004.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.6192287496,"imp":28.1157418775,"elo":1124.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":48.7817729595,"imp":26.290525095,"elo":1200.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":120.4785637723,"imp":24.3199632065,"elo":1279.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":240.3865094023,"imp":24.1140116869,"elo":1285.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":480.7654959148,"imp":23.917273594,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":722.1627706934,"imp":23.7249520548,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":968.8626049308,"imp":23.6836034837,"elo":1304.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.3755198931,"imp":37.2010000155,"elo":910.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":21.7646234862,"imp":35.9268121135,"elo":955.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":63.386888105,"imp":32.9373380486,"elo":1059.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":129.352553403,"imp":31.6555401166,"elo":1121.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":331.7596087189,"imp":30.130905947,"elo":1157.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":670.337180032,"imp":29.3757346754,"elo":1179.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1343.6187037298,"imp":28.9237538913,"elo":1198.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2023.8043409809,"imp":28.7849164504,"elo":1202.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2721.8722746731,"imp":28.707296144,"elo":1203.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":5.1200069793,"imp":26.1375239488,"elo":1203.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":12.2253394924,"imp":25.1717560691,"elo":1241.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":30.3802629809,"imp":23.8394583519,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":62.1068025438,"imp":22.7497273395,"elo":1347.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":158.8472939274,"imp":21.1188155841,"elo":1381.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":317.66743218,"imp":20.4569253318,"elo":1417.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":636.5825631214,"imp":20.0911019659,"elo":1441.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":952.9418266181,"imp":19.9327122653,"elo":1445.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1280.184939635,"imp":19.7071941149,"elo":1456.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":41.1234544852,"imp":13.5726443998,"elo":1512.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":17.2441843169,"imp":30.0913388364,"elo":1095.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":52.8627222227,"imp":27.4135437542,"elo":1148.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":160.6082009531,"imp":24.3230264211,"elo":1247.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":327.2456661478,"imp":23.5084670506,"elo":1278.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":806.1467207879,"imp":22.8429542598,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1661.0512370873,"imp":22.3571442902,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3324.8557036777,"imp":22.2973532555,"elo":1311.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4975.6687289296,"imp":21.6866732933,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6686.5004259785,"imp":21.3013142362,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":20.9501855302,"imp":35.2867241607,"elo":965.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":218.361445605,"imp":62.4165523647,"elo":639.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":445.0214813344,"imp":56.9105891605,"elo":744.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1099.0649504843,"imp":53.4947212047,"elo":804.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1308.4507704502,"imp":53.3105819315,"elo":804.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.7350368805,"imp":37.704633765,"elo":997.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.4865086941,"imp":30.3942325268,"elo":1090.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":15.4344714131,"imp":26.2848781078,"elo":1168.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":33.707023879,"imp":24.185664553,"elo":1199.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":88.3093327016,"imp":22.9930807033,"elo":1284.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":179.7459885669,"imp":22.8033011591,"elo":1290.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":361.9168504232,"imp":23.1479349346,"elo":1289.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":543.5646193801,"imp":23.3089309041,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":729.2747129799,"imp":23.4584275413,"elo":1282.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":19.1299431712,"imp":28.1380744773,"elo":1123.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":40.534251589,"imp":23.4798146527,"elo":1280.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":105.762419468,"imp":21.1516780122,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":213.4238147529,"imp":19.6682796097,"elo":1420.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":537.3854228471,"imp":18.6864739532,"elo":1460.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1076.4891299391,"imp":18.1750583629,"elo":1476.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2156.0193004102,"imp":17.9600482441,"elo":1483.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3232.7208354751,"imp":18.4802333164,"elo":1471.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4335.3941369138,"imp":18.8339644895,"elo":1464.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":9.8834251377,"imp":13.8918897729,"elo":1537.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":23.9402834756,"imp":13.7190846966,"elo":1535.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":65.6941343642,"imp":13.3510024565,"elo":1570.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":134.7721717608,"imp":13.3373822654,"elo":1619.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":345.3716668375,"imp":12.9022146387,"elo":1672.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":691.4097037424,"imp":11.8337494433,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1395.1578235853,"imp":10.1426620452,"elo":1721.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2094.6976308922,"imp":8.5965584386,"elo":1728.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2793.8656494383,"imp":5.5180577513,"elo":1744.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":21.8451817987,"imp":23.5764466457,"elo":1298.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.3634612718,"imp":26.7185554083,"elo":1184.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":95.1581936188,"imp":26.2782127721,"elo":1213.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":234.8668414501,"imp":25.4792251263,"elo":1271.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":480.2032798425,"imp":25.0504553383,"elo":1300.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":1187.414451069,"imp":24.0691362287,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":2358.7610971892,"imp":23.4103451615,"elo":1361.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":4693.5593417073,"imp":23.149303938,"elo":1373.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":7018.4549761562,"imp":22.7788710925,"elo":1381.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":9413.1269943826,"imp":22.2659177313,"elo":1394.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":5.378689382,"imp":22.475220285,"elo":1389.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":49.9745503327,"imp":7.0591602774,"elo":1684.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":4.6668652982,"imp":43.1436700524,"elo":762.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":4.72524967,"imp":15.3246129861,"elo":1528.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":14.968259658,"imp":25.137850187,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":32.4104572417,"imp":24.2348599246,"elo":1338.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":92.1851660275,"imp":23.3456956053,"elo":1359.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":187.6481584341,"imp":22.6376937582,"elo":1387.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":467.1995115985,"imp":22.2009008978,"elo":1395.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":936.7829918111,"imp":21.7471247143,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1869.6886183547,"imp":20.9458040476,"elo":1420.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2822.7067282211,"imp":20.6060794792,"elo":1426.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3796.9310139525,"imp":20.1811207266,"elo":1432.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":9.9022651232,"imp":13.1711052461,"elo":1662.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":8.9308378882,"imp":11.3618793221,"elo":1594.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":666.7207045489,"imp":44.140526255,"elo":767.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1026.6183216352,"imp":43.8746747897,"elo":790.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1652.5474732053,"imp":43.1218234355,"elo":810.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2473.0165808756,"imp":41.5605404935,"elo":835.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5380.2088519459,"imp":38.3590572676,"elo":890.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11008.2551380325,"imp":35.6978698962,"elo":924.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":22866.390108276,"imp":34.6444160317,"elo":949.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":33810.8321985267,"imp":34.0696045971,"elo":965.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":44782.5951977839,"imp":33.3348656952,"elo":990.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":3.0219759465,"imp":27.3010461712,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.370479118,"imp":33.0704382,"elo":1000.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":28.5804934274,"imp":29.9330040565,"elo":1090.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":89.37412395,"imp":27.7369896196,"elo":1169.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":187.12724454,"imp":26.7896514464,"elo":1222.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":451.1590289841,"imp":26.1087034748,"elo":1230.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":875.4622563657,"imp":25.6721954433,"elo":1236.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1726.0460158647,"imp":25.2486379702,"elo":1244.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2669.9686934389,"imp":24.9321348385,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3647.6027397269,"imp":24.7891102082,"elo":1254.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":4.9305373226,"imp":23.7109406798,"elo":1205.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":15.4118115252,"imp":22.4518668711,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":47.0344042271,"imp":22.0051248606,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":94.1830833987,"imp":21.4712730772,"elo":1332.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":248.7881121837,"imp":21.0359506096,"elo":1357.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":502.7403096728,"imp":20.753793252,"elo":1369.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":1010.0144540287,"imp":20.6022884448,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":1510.9325947178,"imp":20.5077595998,"elo":1384.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":2029.7698407236,"imp":20.2903121122,"elo":1388.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":495.6488461748,"imp":30.3223483612,"elo":998.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1563.1416754751,"imp":27.776668917,"elo":1065.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":4713.7549099867,"imp":24.690106465,"elo":1186.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":10547.3540062779,"imp":22.7797109326,"elo":1255.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":26588.4297619676,"imp":20.6790989279,"elo":1346.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":27893.2292728524,"imp":20.6927945888,"elo":1345.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":8.9914037311,"imp":31.554912989,"elo":1089.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":19.0589824805,"imp":29.8443643494,"elo":1164.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":49.701827647,"imp":27.2381375084,"elo":1254.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":101.4654106495,"imp":25.6812543014,"elo":1277.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":258.6773776274,"imp":24.0675997212,"elo":1324.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":515.1758887421,"imp":23.0996159111,"elo":1356.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":1034.8791027689,"imp":22.6255506547,"elo":1370.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1547.9692908758,"imp":22.4570042735,"elo":1377.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2079.2181272278,"imp":22.3955632839,"elo":1380.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.1027625141,"x_train":283.646286296,"imp":25.0435111439,"elo":1283.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2536073213,"x_train":26.3194957247,"imp":22.3493167591,"elo":1288.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4231601839,"x_train":52.2760454584,"imp":20.1500037908,"elo":1376.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.6051567129,"x_train":120.7421051485,"imp":19.080273336,"elo":1419.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.7512952022,"x_train":216.6422936563,"imp":18.4288826138,"elo":1444.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.8418166173,"x_train":415.1322548777,"imp":17.9118649582,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.007557247,"x_train":727.7924472465,"imp":17.907424396,"elo":1458.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1300458847,"x_train":1402.2322064531,"imp":17.2185817434,"elo":1465.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2201190611,"x_train":2094.5870262909,"imp":17.022905266,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2219998697,"x_train":2807.9829916593,"imp":16.766699508,"elo":1468.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2033347873,"x_train":7.4221463904,"imp":28.9264110498,"elo":1135.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3168559826,"x_train":15.2214794665,"imp":28.3548505682,"elo":1146.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4637258396,"x_train":38.0924279868,"imp":27.8896719359,"elo":1159.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.6360313896,"x_train":76.6924688657,"imp":27.2792233321,"elo":1174.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.8576608402,"x_train":193.4343472071,"imp":26.2870986179,"elo":1190.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.9940308037,"x_train":388.7016596181,"imp":25.8410342794,"elo":1207.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1389353212,"x_train":776.6628709802,"imp":25.3611606878,"elo":1215.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1914455458,"x_train":1163.7786661849,"imp":25.1575788624,"elo":1220.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2260580575,"x_train":1560.1761733922,"imp":25.0263500682,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0424445763,"x_train":12.1942564441,"imp":25.280086947,"elo":1219.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.1131096225,"x_train":81.2230381354,"imp":24.9171703738,"elo":1228.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.1833513515,"x_train":314.4258454848,"imp":24.665046851,"elo":1246.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.1834859617,"x_train":700.8132948146,"imp":24.6110063021,"elo":1250.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1958839273,"x_train":1955.0904519336,"imp":24.3733254772,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1993753888,"x_train":4550.043096608,"imp":24.1269970022,"elo":1269.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.2146537242,"x_train":8811.0765244269,"imp":23.6638108139,"elo":1300.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.2386362077,"x_train":13221.5244172938,"imp":23.4665104829,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.2817546014,"x_train":17528.8314911942,"imp":23.3215821166,"elo":1316.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7866568202,"x_train":5.1558379712,"imp":38.6548685735,"elo":917.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9513567957,"x_train":9.9671223858,"imp":34.5355870581,"elo":1004.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7969717524,"x_train":24.6192287496,"imp":28.1157418775,"elo":1124.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0328003281,"x_train":48.7817729595,"imp":26.290525095,"elo":1200.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4148344838,"x_train":120.4785637723,"imp":24.3199632065,"elo":1279.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4054403817,"x_train":240.3865094023,"imp":24.1140116869,"elo":1285.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.6379815682,"x_train":480.7654959148,"imp":23.917273594,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7473298834,"x_train":722.1627706934,"imp":23.7249520548,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.8636375274,"x_train":968.8626049308,"imp":23.6836034837,"elo":1304.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6590449294,"x_train":8.3755198931,"imp":37.2010000155,"elo":910.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4924031952,"x_train":21.7646234862,"imp":35.9268121135,"elo":955.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1390542652,"x_train":63.386888105,"imp":32.9373380486,"elo":1059.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.194877341,"x_train":129.352553403,"imp":31.6555401166,"elo":1121.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.8927130955,"x_train":331.7596087189,"imp":30.130905947,"elo":1157.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.3270562101,"x_train":670.337180032,"imp":29.3757346754,"elo":1179.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.5813386245,"x_train":1343.6187037298,"imp":28.9237538913,"elo":1198.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":11.4452770419,"x_train":2023.8043409809,"imp":28.7849164504,"elo":1202.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5900916415,"x_train":2721.8722746731,"imp":28.707296144,"elo":1203.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4437307263,"x_train":5.1200069793,"imp":26.1375239488,"elo":1203.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.7310412682,"x_train":12.2253394924,"imp":25.1717560691,"elo":1241.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.3562691131,"x_train":30.3802629809,"imp":23.8394583519,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.953724397,"x_train":62.1068025438,"imp":22.7497273395,"elo":1347.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.4130882094,"x_train":158.8472939274,"imp":21.1188155841,"elo":1381.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.0010341081,"x_train":317.66743218,"imp":20.4569253318,"elo":1417.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.5773453268,"x_train":636.5825631214,"imp":20.0911019659,"elo":1441.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.2557034144,"x_train":952.9418266181,"imp":19.9327122653,"elo":1445.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.0776987398,"x_train":1280.184939635,"imp":19.7071941149,"elo":1456.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":8.6091982733,"x_train":41.1234544852,"imp":13.5726443998,"elo":1512.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5648398113,"x_train":17.2441843169,"imp":30.0913388364,"elo":1095.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.1086986149,"x_train":52.8627222227,"imp":27.4135437542,"elo":1148.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9275358111,"x_train":160.6082009531,"imp":24.3230264211,"elo":1247.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9365839976,"x_train":327.2456661478,"imp":23.5084670506,"elo":1278.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.7350733756,"x_train":806.1467207879,"imp":22.8429542598,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2166850654,"x_train":1661.0512370873,"imp":22.3571442902,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5594018842,"x_train":3324.8557036777,"imp":22.2973532555,"elo":1311.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7492172244,"x_train":4975.6687289296,"imp":21.6866732933,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5865347302,"x_train":6686.5004259785,"imp":21.3013142362,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3262335461,"x_train":20.9501855302,"imp":35.2867241607,"elo":965.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4602862659,"x_train":218.361445605,"imp":62.4165523647,"elo":639.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":5.3441524978,"x_train":445.0214813344,"imp":56.9105891605,"elo":744.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":10.9553867171,"x_train":1099.0649504843,"imp":53.4947212047,"elo":804.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":9.5934077851,"x_train":1308.4507704502,"imp":53.3105819315,"elo":804.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.147930342,"x_train":0.7350368805,"imp":37.704633765,"elo":997.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4717318903,"x_train":4.4865086941,"imp":30.3942325268,"elo":1090.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8456747425,"x_train":15.4344714131,"imp":26.2848781078,"elo":1168.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2491970937,"x_train":33.707023879,"imp":24.185664553,"elo":1199.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.3358363812,"x_train":88.3093327016,"imp":22.9930807033,"elo":1284.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.453941186,"x_train":179.7459885669,"imp":22.8033011591,"elo":1290.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4155319356,"x_train":361.9168504232,"imp":23.1479349346,"elo":1289.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.5473644093,"x_train":543.5646193801,"imp":23.3089309041,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.8330056127,"x_train":729.2747129799,"imp":23.4584275413,"elo":1282.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.920194808,"x_train":19.1299431712,"imp":28.1380744773,"elo":1123.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.2086544334,"x_train":40.534251589,"imp":23.4798146527,"elo":1280.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.1368730282,"x_train":105.762419468,"imp":21.1516780122,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.7650894271,"x_train":213.4238147529,"imp":19.6682796097,"elo":1420.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.4912076922,"x_train":537.3854228471,"imp":18.6864739532,"elo":1460.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4616303712,"x_train":1076.4891299391,"imp":18.1750583629,"elo":1476.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.6029747084,"x_train":2156.0193004102,"imp":17.9600482441,"elo":1483.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.7682775511,"x_train":3232.7208354751,"imp":18.4802333164,"elo":1471.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.142857233,"x_train":4335.3941369138,"imp":18.8339644895,"elo":1464.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8250564093,"x_train":9.8834251377,"imp":13.8918897729,"elo":1537.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2057905305,"x_train":23.9402834756,"imp":13.7190846966,"elo":1535.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.4337944176,"x_train":65.6941343642,"imp":13.3510024565,"elo":1570.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2295930552,"x_train":134.7721717608,"imp":13.3373822654,"elo":1619.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.2854006353,"x_train":345.3716668375,"imp":12.9022146387,"elo":1672.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7906340457,"x_train":691.4097037424,"imp":11.8337494433,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4633585633,"x_train":1395.1578235853,"imp":10.1426620452,"elo":1721.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8898273316,"x_train":2094.6976308922,"imp":8.5965584386,"elo":1728.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9280086995,"x_train":2793.8656494383,"imp":5.5180577513,"elo":1744.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7021367127,"x_train":21.8451817987,"imp":23.5764466457,"elo":1298.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8027161329,"x_train":46.3634612718,"imp":26.7185554083,"elo":1184.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":61.1894558461,"x_train":95.1581936188,"imp":26.2782127721,"elo":1213.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":96.0468231173,"x_train":234.8668414501,"imp":25.4792251263,"elo":1271.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":131.2394515616,"x_train":480.2032798425,"imp":25.0504553383,"elo":1300.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":196.2246230598,"x_train":1187.414451069,"imp":24.0691362287,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":222.0364795428,"x_train":2358.7610971892,"imp":23.4103451615,"elo":1361.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":282.5497243141,"x_train":4693.5593417073,"imp":23.149303938,"elo":1373.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":310.9391931013,"x_train":7018.4549761562,"imp":22.7788710925,"elo":1381.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":343.7310101457,"x_train":9413.1269943826,"imp":22.2659177313,"elo":1394.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2735262831,"x_train":5.378689382,"imp":22.475220285,"elo":1389.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.6889691672,"x_train":49.9745503327,"imp":7.0591602774,"elo":1684.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.8545350274,"x_train":4.6668652982,"imp":43.1436700524,"elo":762.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2245693736,"x_train":4.72524967,"imp":15.3246129861,"elo":1528.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.624076982,"x_train":14.968259658,"imp":25.137850187,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":1.0912900685,"x_train":32.4104572417,"imp":24.2348599246,"elo":1338.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.7054134697,"x_train":92.1851660275,"imp":23.3456956053,"elo":1359.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.3761032339,"x_train":187.6481584341,"imp":22.6376937582,"elo":1387.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.9257260459,"x_train":467.1995115985,"imp":22.2009008978,"elo":1395.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.3279465065,"x_train":936.7829918111,"imp":21.7471247143,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.9820772589,"x_train":1869.6886183547,"imp":20.9458040476,"elo":1420.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.3321042202,"x_train":2822.7067282211,"imp":20.6060794792,"elo":1426.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.5447590803,"x_train":3796.9310139525,"imp":20.1811207266,"elo":1432.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5909006176,"x_train":9.9022651232,"imp":13.1711052461,"elo":1662.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4902017202,"x_train":8.9308378882,"imp":11.3618793221,"elo":1594.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8225789867,"x_train":666.7207045489,"imp":44.140526255,"elo":767.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.7876620486,"x_train":1026.6183216352,"imp":43.8746747897,"elo":790.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":13.2339480691,"x_train":1652.5474732053,"imp":43.1218234355,"elo":810.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":15.9767784961,"x_train":2473.0165808756,"imp":41.5605404935,"elo":835.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8983325102,"x_train":5380.2088519459,"imp":38.3590572676,"elo":890.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.7325471822,"x_train":11008.2551380325,"imp":35.6978698962,"elo":924.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":24.6022552378,"x_train":22866.390108276,"imp":34.6444160317,"elo":949.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":28.6345093899,"x_train":33810.8321985267,"imp":34.0696045971,"elo":965.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.1023197407,"x_train":44782.5951977839,"imp":33.3348656952,"elo":990.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0760162666,"x_train":3.0219759465,"imp":27.3010461712,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3624794907,"x_train":14.370479118,"imp":33.0704382,"elo":1000.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6934737284,"x_train":28.5804934274,"imp":29.9330040565,"elo":1090.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9854209931,"x_train":89.37412395,"imp":27.7369896196,"elo":1169.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0396624232,"x_train":187.12724454,"imp":26.7896514464,"elo":1222.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2562289121,"x_train":451.1590289841,"imp":26.1087034748,"elo":1230.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.445176011,"x_train":875.4622563657,"imp":25.6721954433,"elo":1236.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6802253472,"x_train":1726.0460158647,"imp":25.2486379702,"elo":1244.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.9118329956,"x_train":2669.9686934389,"imp":24.9321348385,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1614416305,"x_train":3647.6027397269,"imp":24.7891102082,"elo":1254.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5916710541,"x_train":4.9305373226,"imp":23.7109406798,"elo":1205.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.9620057603,"x_train":15.4118115252,"imp":22.4518668711,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.5632518606,"x_train":47.0344042271,"imp":22.0051248606,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.7564565665,"x_train":94.1830833987,"imp":21.4712730772,"elo":1332.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.5049134408,"x_train":248.7881121837,"imp":21.0359506096,"elo":1357.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.9073614098,"x_train":502.7403096728,"imp":20.753793252,"elo":1369.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.4289445362,"x_train":1010.0144540287,"imp":20.6022884448,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.0567675025,"x_train":1510.9325947178,"imp":20.5077595998,"elo":1384.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.1126713529,"x_train":2029.7698407236,"imp":20.2903121122,"elo":1388.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":205.5125020069,"x_train":495.6488461748,"imp":30.3223483612,"elo":998.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":447.1825554268,"x_train":1563.1416754751,"imp":27.776668917,"elo":1065.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":729.916383431,"x_train":4713.7549099867,"imp":24.690106465,"elo":1186.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":828.8918571484,"x_train":10547.3540062779,"imp":22.7797109326,"elo":1255.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":777.6640795695,"x_train":26588.4297619676,"imp":20.6790989279,"elo":1346.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":752.505846815,"x_train":27893.2292728524,"imp":20.6927945888,"elo":1345.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.371587587,"x_train":8.9914037311,"imp":31.554912989,"elo":1089.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.7371554717,"x_train":19.0589824805,"imp":29.8443643494,"elo":1164.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.3176035748,"x_train":49.701827647,"imp":27.2381375084,"elo":1254.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":2.7650973713,"x_train":101.4654106495,"imp":25.6812543014,"elo":1277.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.4463880936,"x_train":258.6773776274,"imp":24.0675997212,"elo":1324.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.6826514001,"x_train":515.1758887421,"imp":23.0996159111,"elo":1356.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":3.024314406,"x_train":1034.8791027689,"imp":22.6255506547,"elo":1370.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":3.3820709216,"x_train":1547.9692908758,"imp":22.4570042735,"elo":1377.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.3466778136,"x_train":2079.2181272278,"imp":22.3955632839,"elo":1380.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..8d0410765a8d799c789f0f467a675581a036cbff
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_multiclass/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "TabFM (default)", "RealTabPFN-2.5 (tuned)", "TabPFN-2.6 (default)", "TabPFN-3 (default)", "RealTabPFN-2.5 (default)", "TabICLv2 (default)", "LimiX (default)", "AutoGluon 1.4 (best, 4h)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "RealMLP (tuned + ensembled)", "LightGBM (tuned + ensembled)", "LightGBM (tuned)", "TabM (tuned + ensembled)", "TabM (tuned)", "RealMLP (tuned)", "TabDPT (tuned + ensembled)", "TabDPT-Turbo (default)", "XGBoost (tuned + ensembled)", "XGBoost (tuned)", "xRFM (tuned + ensembled)", "iLTM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "EBM (tuned + ensembled)", "ModernNCA (tuned)", "TabDPT (tuned)", "TabM (default)", "ExtraTrees (tuned + ensembled)", "SAP-RPT-OSS (default)", "CatBoost (default)", "xRFM (tuned)", "BetaTabPFN (default)", "RandomForest (tuned + ensembled)", "iLTM (tuned)", "EBM (tuned)", "TorchMLP (tuned + ensembled)", "ExtraTrees (tuned)", "EBM (default)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "TorchMLP (tuned)", "RandomForest (tuned)", "XGBoost (default)", "TabSwift (default)", "LightGBM (default)", "FastaiMLP (tuned + ensembled)", "TabDPT (default)", "ChimeraBoost (default)", "RealMLP (default)", "FastaiMLP (tuned)", "xRFM (default)", "ModernNCA (default)", "KNN (tuned + ensembled)", "TorchMLP (default)", "iLTM (default)", "RandomForest (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "OrionMSP (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "FastaiMLP (default)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (default)", "KNN (tuned)", "TabFlex (default)", "PerpetualBooster (tuned)", "Linear (default)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.375, 0.25, 0.75, 0.625, 0.75, 0.5, 0.75, 0.875, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.625, NaN, 0.375, 0.75, 0.75, 0.625, 1.0, 0.75, 0.75, 0.75, 1.0, 1.0, 0.875, 0.875, 0.875, 0.75, 0.75, 1.0, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 0.75, 0.75, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.75, 0.625, NaN, 0.75, 0.75, 0.75, 0.875, 1.0, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.25, 0.25, NaN, 0.625, 0.375, 0.6875, 0.625, 0.5, 0.625, 0.875, 0.75, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0, 0.75, 0.75, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 0.75, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.375, 0.25, 0.25, 0.375, NaN, 0.5, 0.75, 0.5, 0.75, 0.5, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 0.75, 1.0, 0.75, 0.75, 0.875, 0.75, 1.0, 0.75, 0.75, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 1.0, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.375, 0.25, 0.625, 0.5, NaN, 0.5, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.75, 1.0, 1.0, 0.75, 0.875, 0.75, 0.75, 0.875, 0.625, 0.75, 0.875, 0.75, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5, 0.0, 0.125, 0.3125, 0.25, 0.5, NaN, 0.5, 0.75, 0.5, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.5, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.875, 0.75, 0.875, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.25, 0.0, 0.375, 0.5, 0.25, 0.5, NaN, 0.625, 0.625, 0.625, 0.625, 0.625, 0.875, 0.875, 0.625, 0.625, 0.625, 0.75, 0.875, 0.75, 0.75, 0.625, 0.875, 0.875, 0.625, 0.75, 0.75, 0.75, 0.75, 0.5, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.25, 0.5, 0.25, 0.375, 0.25, 0.375, NaN, 0.5, 0.625, 0.5, 0.625, 0.625, 0.625, 0.75, 0.75, 0.625, 0.5, 0.625, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.625, 0.75, 0.875, 0.625, 0.75, 0.875, 0.75, 0.875, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 0.875, 0.75, 0.625, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.25, 0.375, 0.5, 0.5, 0.5, 0.375, 0.5, NaN, 0.625, 0.625, 0.5, 0.75, 0.75, 0.5, 0.625, 0.5, 0.5, 0.625, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.625, 0.625, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.625, 0.625, 0.875, 0.875, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.75, 0.875, 0.875, 1.0, 0.75, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.0, 0.125, 0.125, 0.125, 0.375, 0.25, 0.375, 0.375, 0.375, NaN, 0.625, 0.375, 0.375, 0.375, 0.75, 0.75, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 0.75, 0.875, 0.625, 0.75, 0.75, 0.75, 0.75, 0.5, 0.75, 0.75, 0.625, 0.625, 0.875, 0.75, 0.875, 0.75, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0, 0.875, 0.75, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.0, 0.125, 0.25, 0.125, 0.375, 0.25, 0.375, 0.5, 0.375, 0.375, NaN, 0.375, 0.5, 0.5, 0.625, 0.75, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 0.75, 0.875, 0.625, 0.75, 0.75, 0.75, 0.75, 0.5, 0.75, 0.75, 0.625, 0.625, 0.875, 0.625, 0.875, 0.75, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0, 0.875, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.75, 1.0, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.375, 0.25, 0.375, 0.375, 0.5, 0.625, 0.625, NaN, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.5, 0.875, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.5, 0.875, 0.5, 0.625, 0.75, 0.625, 0.75, 0.5, 0.875, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.875, 0.75, 0.875, 0.875, 0.625, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.625, 0.5, 0.375, NaN, 0.5, 0.5, 0.5, 0.5, 0.5, 0.75, 0.5, 0.625, 0.5, 0.625, 0.5, 0.75, 0.5, 0.625, 0.75, 0.625, 0.625, 0.875, 0.625, 0.875, 0.625, 0.625, 0.875, 0.875, 0.625, 0.875, 1.0, 1.0, 0.75, 0.875, 1.0, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0, 0.75, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.625, 0.5, 0.375, 0.5, NaN, 0.5, 0.5, 0.5, 0.5, 0.75, 0.5, 0.5, 0.5, 0.625, 0.5, 0.75, 0.5, 0.625, 0.75, 0.625, 0.625, 0.875, 0.625, 0.875, 0.625, 0.625, 0.875, 0.875, 0.625, 0.875, 1.0, 1.0, 0.75, 0.875, 1.0, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0, 0.75, 0.875, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.125, 0.125, 0.25, 0.375, 0.25, 0.375, 0.25, 0.5, 0.25, 0.375, 0.375, 0.5, 0.5, NaN, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.5, 0.625, 0.875, 0.75, 0.75, 0.875, 0.5, 0.625, 0.625, 0.625, 0.875, 0.625, 0.625, 0.875, 0.875, 0.625, 0.875, 0.625, 0.625, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.375, 0.5, 0.5, 0.5, NaN, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.5, 0.625, 0.875, 0.75, 0.75, 0.875, 0.5, 0.625, 0.625, 0.625, 0.875, 0.625, 0.625, 0.875, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.625, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.0, 0.125, 0.0, 0.125, 0.125, 0.25, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, 0.5, NaN, 0.5, 0.375, 0.625, 0.75, 0.5, 0.625, 0.75, 0.75, 0.875, 0.625, 0.75, 0.625, 0.5, 0.625, 0.75, 0.75, 0.625, 0.875, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.625, 0.875, 0.75, 0.75, 0.75, 0.5, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.5, 0.5, 0.25, 0.25, 0.5, 0.5, 0.5, 0.375, 0.375, 0.5, NaN, 0.375, 0.625, 0.625, 0.5, 0.75, 0.625, 0.625, 0.625, 0.875, 0.5, 0.5, 0.5, 0.75, 0.625, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.875, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0], [0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.5, 0.125, 0.375, 0.375, 0.5, 0.5, 0.5, 0.25, 0.25, 0.375, 0.5, 0.625, 0.625, NaN, 0.625, 0.625, 0.375, 0.75, 0.625, 0.5, 0.625, 0.625, 0.5, 0.625, 0.5, 0.75, 0.625, 0.5, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 1.0, 0.75, 0.625, 0.875, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 0.75, 0.75, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.5, 0.5, 0.375, 0.5, 0.375, 0.375, 0.375, NaN, 0.625, 0.5, 0.75, 0.75, 0.625, 0.625, 0.5, 0.75, 0.625, 0.5, 0.75, 0.625, 0.625, 0.5, 0.625, 0.625, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 0.625, 1.0, 0.75, 0.875, 0.625, 0.625, 0.875, 0.75, 0.875, 0.625, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.5, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, NaN, 0.5, 0.625, 0.75, 0.625, 0.625, 0.5, 0.625, 0.625, 0.5, 0.75, 0.625, 0.625, 0.5, 0.625, 0.625, 0.75, 0.875, 0.75, 0.875, 0.875, 0.75, 0.625, 1.0, 0.75, 0.875, 0.625, 0.625, 0.875, 0.75, 0.875, 0.625, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.5, 0.375, 0.5, 0.5, 0.625, 0.5, 0.5, NaN, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.375, 0.5, 0.625, 0.625, 0.625, 0.5, 0.5, 0.625, 0.75, 0.625, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 1.0, 0.875, 1.0, 1.0, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.125, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.5, 0.5, 0.375, 0.25, 0.25, 0.25, 0.375, 0.5, NaN, 0.5, 0.5, 0.75, 0.5, 0.5, 0.75, 0.5, 0.5, 0.75, 0.5, 0.625, 0.625, 0.625, 0.625, 0.75, 0.875, 0.75, 0.75, 0.875, 0.625, 0.75, 0.625, 0.625, 0.75, 0.625, 0.875, 0.875, 0.75, 0.75, 1.0, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.125, 0.25, 0.375, 0.125, 0.125, 0.25, 0.5, 0.5, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.5, NaN, 0.5, 0.6875, 0.5, 0.5, 0.75, 0.5, 0.625, 0.75, 0.625, 0.625, 0.625, 0.5, 0.625, 0.75, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.375, 0.5, 0.375, 0.375, 0.5, 0.5, 0.5, NaN, 0.625, 0.375, 0.25, 0.625, 0.625, 0.75, 0.625, 0.5, 0.625, 0.75, 0.875, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.875, 0.75, 1.0, 1.0, 0.875, 0.75, 0.875, 0.75, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.375, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.5, 0.5, 0.25, 0.25, 0.125, 0.375, 0.375, 0.375, 0.375, 0.5, 0.25, 0.3125, 0.375, NaN, 0.375, 0.5, 0.75, 0.625, 0.625, 0.75, 0.5, 0.625, 0.75, 0.625, 0.5, 0.75, 0.375, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 0.75, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.5, 0.375, 0.375, 0.25, 0.25, 0.375, 0.125, 0.375, 0.5, 0.5, 0.375, 0.5, 0.5, 0.625, 0.625, NaN, 0.5, 0.375, 0.5, 0.625, 0.5, 0.625, 0.375, 0.625, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.875, 0.625, 0.75, 0.75, 0.875, 0.625, 0.875, 0.875, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0], [0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.5, 0.5, 0.25, 0.375, 0.375, 0.5, 0.5, 0.75, 0.5, 0.5, NaN, 0.5, 0.375, 0.625, 0.5, 0.75, 0.5, 0.625, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.5, 0.625, 0.625, 0.5, 0.625, 0.875, 0.75, 0.75, 0.75, 1.0, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.5, 0.375, 0.375, 0.5, 0.5, 0.375, 0.5, 0.375, 0.375, 0.375, 0.375, 0.25, 0.25, 0.375, 0.25, 0.625, 0.5, NaN, 0.5, 0.5, 0.375, 0.625, 0.75, 0.5, 0.5, 0.5, 0.8125, 0.5, 0.625, 0.625, 0.625, 0.75, 0.5, 0.75, 0.75, 0.5, 0.75, 0.75, 0.625, 0.75, 0.875, 0.625, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.75, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0], [0.25, 0.0, 0.125, 0.125, 0.25, 0.375, 0.25, 0.5, 0.375, 0.375, 0.5, 0.5, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.625, 0.5, 0.5, 0.375, 0.375, 0.5, 0.625, 0.5, NaN, 0.5, 0.625, 0.625, 0.5, 0.5, 0.625, 0.625, 0.5, 0.625, 0.5, 0.5, 0.625, 0.5, 0.5, 0.5, 0.625, 0.625, 0.5, 0.625, 0.75, 0.75, 0.625, 0.625, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.125, 0.375, 0.375, 0.375, 0.25, 0.25, 0.25, 0.25, 0.5, 0.5, 0.375, 0.25, 0.375, 0.375, 0.375, 0.5, 0.5, NaN, 0.625, 0.25, 0.5, 0.75, 0.375, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.75, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.25, 0.375, 0.25, 0.5, 0.5, 0.625, 0.375, 0.375, NaN, 0.5, 0.75, 0.5, 0.375, 0.375, 0.75, 0.375, 0.75, 0.5, 0.5, 0.75, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.75, 0.625, 1.0, 0.625, 1.0, 1.0, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 0.75, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.5, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.5, 0.375, 0.25, 0.375, 0.375, 0.75, 0.5, NaN, 0.5, 0.625, 0.875, 0.625, 0.375, 0.5, 0.625, 0.625, 0.5, 0.625, 0.5, 0.5, 0.5, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.375, 0.25, 0.125, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, 0.375, 0.375, 0.375, 0.375, 0.625, 0.5, 0.25, 0.5, 0.5, 0.25, 0.5, NaN, 0.5, 0.5, 0.5, 0.625, 0.5, 0.75, 0.625, 0.5, 0.875, 0.625, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 0.75, 0.625, 0.875, 0.875, 1.0, 0.875, 1.0, 0.75, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.25, 0.375, 0.375, 0.125, 0.125, 0.125, 0.375, 0.375, 0.375, 0.375, 0.125, 0.25, 0.25, 0.375, 0.375, 0.5, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.5, 0.5, 0.25, 0.5, 0.375, 0.5, NaN, 0.375, 0.625, 0.75, 0.625, 0.5, 0.5, 0.625, 0.625, 0.5, 0.625, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.625, 0.875, 0.875, 0.875, 1.0, 0.875, 0.75, 0.75, 0.875, 1.0, 0.75, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.125, 0.375, 0.375, 0.25, 0.5, 0.375, 0.625, 0.625, 0.125, 0.5, 0.625, NaN, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.375, 0.5, 0.375, 0.625, 0.625, 0.75, 0.625, 0.75, 0.75, 0.875, 0.75, 0.875, 1.0, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.5, 0.5, 0.375, 0.5, 0.375, 0.375, 0.625, 0.375, 0.5, 0.375, 0.5, NaN, 0.5, 0.625, 0.5, 0.5, 0.75, 0.5, 0.625, 0.5, 0.625, 0.625, 0.75, 0.5, 0.75, 0.75, 0.75, 0.625, 0.75, 1.0, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.5, 0.1875, 0.5, 0.5, 0.25, 0.625, 0.375, 0.25, 0.5, 0.5, NaN, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.625, 0.75, 0.5, 0.625, 0.75, 0.5, 0.625, 0.625, 0.625, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 0.625, 0.75, 0.625, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.125, 0.25, 0.375, 0.625, 0.375, 0.375, 0.5, 0.375, 0.375, 0.625, 0.5, 0.5, 0.375, 0.5, 0.375, 0.625, NaN, 0.375, 0.5, 0.5, 0.625, 0.375, 0.5, 0.375, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 1.0, 0.625, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.0, 0.25, 0.25, 0.25, 0.0, 0.0, 0.375, 0.375, 0.375, 0.375, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.25, 0.375, 0.25, 0.5, 0.5, 0.5, 0.5, 0.625, NaN, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.5, 0.75, 0.75, 0.75, 0.625, 0.75, 0.5, 0.625, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.0, 0.25, 0.25, 0.25, 0.0, 0.0, 0.375, 0.25, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.375, NaN, 0.625, 0.625, 0.375, 0.625, 0.5, 0.5, 0.75, 0.875, 0.75, 0.625, 0.75, 0.625, 0.625, 0.75, 0.875, 0.75, 0.875, 0.875, 0.75, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.125, 0.25, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.5, 0.375, 0.5, 0.25, 0.5, 0.5, 0.375, 0.375, NaN, 0.5, 0.375, 0.375, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.875, 0.875, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0], [0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.125, 0.125, 0.375, 0.25, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.5, 0.375, 0.25, 0.375, 0.125, 0.375, 0.5, 0.5, 0.5, 0.375, 0.375, 0.375, 0.5, NaN, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.375, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.75, 0.625, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.25, 0.125, 0.25, 0.125, 0.0, 0.0, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.5, 0.5, 0.5, 0.5, 0.5, 0.375, 0.5, 0.625, 0.375, 0.5, 0.625, 0.5, 0.625, 0.625, 0.5, NaN, 0.625, 0.625, 0.5, 0.5, 0.75, 0.625, 0.625, 0.5, 0.75, 0.75, 0.875, 1.0, 0.875, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.125, 0.375, 0.25, 0.25, 0.375, 0.0, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.5, 0.25, 0.375, 0.5, 0.375, 0.375, 0.5, 0.5, 0.375, 0.5, 0.375, 0.375, 0.625, 0.375, 0.375, NaN, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.5, 0.625, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.125, 0.125, 0.375, 0.375, 0.25, 0.5, 0.375, 0.375, 0.5, 0.25, 0.375, 0.375, 0.5, 0.5, 0.5, 0.375, 0.625, 0.375, 0.25, 0.625, 0.375, 0.5, 0.375, 0.5, 0.375, 0.25, NaN, 0.5, 0.375, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.5, 0.875, 0.875, 0.875, 0.75, 0.75, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.125, 0.25, 0.0, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.5, 0.375, 0.5, 0.25, 0.375, 0.375, 0.5, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.375, 0.5, NaN, 0.375, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.5, 0.75, 0.625, 0.75, 0.75, 0.625, 0.75, 0.875, 1.0, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.375, 0.25, 0.125, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.5, 0.125, 0.125, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.125, 0.375, 0.25, 0.5, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.5, 0.375, 0.625, 0.625, NaN, 0.5, 0.625, 0.75, 0.75, 0.5, 0.875, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 1.0, 0.5, 0.75, 0.625, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.375, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.375, 0.5, 0.25, 0.375, 0.25, 0.375, 0.25, 0.375, 0.375, 0.25, 0.5, 0.25, 0.25, 0.25, 0.125, 0.375, 0.375, 0.25, 0.375, 0.375, 0.5, 0.5, NaN, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.75, 0.5, 0.625, 0.875, 0.625, 0.625, 0.75, 0.75, 0.75, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.25, 0.0, 0.25, 0.0, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.5, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.5, 0.375, 0.375, NaN, 0.5, 0.625, 0.5, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.5, 0.5, NaN, 0.625, 0.625, 0.625, 0.5, 0.75, 0.5, 0.625, 0.625, 0.625, 0.75, 0.75, 1.0, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.375, 0.375, 0.0, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.125, 0.375, 0.25, 0.0, 0.25, 0.25, 0.375, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.5, 0.5, 0.375, 0.375, 0.25, 0.5, 0.375, 0.375, NaN, 0.5, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.375, 0.25, 0.375, 0.375, 0.25, 0.375, 0.25, 0.375, 0.125, 0.125, 0.375, 0.375, 0.25, 0.5, 0.375, 0.25, 0.625, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.5, NaN, 0.625, 0.75, 0.5, 0.875, 0.625, 0.625, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.625, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.0, 0.375, 0.25, 0.0, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.25, 0.375, 0.375, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.125, 0.375, 0.25, 0.375, 0.375, 0.375, NaN, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.75, 0.375, 1.0, 0.625, 0.75, 0.625, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.0, 0.125, 0.25, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.5, 0.375, 0.375, 0.375, 0.5, 0.375, 0.25, 0.625, NaN, 0.375, 0.375, 0.75, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.0, 0.0, 0.25, 0.0, 0.125, 0.0, 0.0, 0.25, 0.25, 0.125, 0.375, 0.0, 0.25, 0.125, 0.25, 0.0, 0.0, 0.25, 0.25, 0.0, 0.125, 0.125, 0.25, 0.25, 0.0, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.5, 0.625, 0.625, NaN, 0.625, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.25, 0.25, 0.125, 0.375, 0.0, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.375, 0.125, 0.125, 0.125, 0.5, 0.375, 0.375, 0.5, 0.375, 0.5, 0.25, 0.125, 0.625, 0.625, 0.375, NaN, 0.625, 0.625, 0.375, 0.625, 0.75, 0.5, 0.75, 0.75, 0.75, 0.875, 0.625, 0.75, 0.75, 0.75, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.0, 0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.625, 0.25, 0.25, 0.375, NaN, 0.4375, 0.375, 0.5, 0.75, 0.625, 0.75, 0.75, 1.0, 0.875, 0.625, 0.75, 0.875, 0.75, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.25, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.375, 0.25, 0.375, 0.625, 0.25, 0.375, 0.375, 0.5625, NaN, 0.5, 0.5, 0.75, 0.5, 0.75, 0.75, 1.0, 0.875, 0.625, 0.75, 0.875, 0.75, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.25, 0.0, 0.0, 0.125, 0.0, 0.375, 0.125, 0.0, 0.125, 0.125, 0.125, 0.25, 0.25, 0.0, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.375, 0.25, 0.375, 0.375, 0.375, 0.25, 0.25, 0.625, 0.375, 0.5, 0.625, 0.625, 0.5, NaN, 0.625, 0.625, 0.625, 0.625, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.25, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.0, 0.375, 0.25, 0.25, 0.375, 0.25, 0.5, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, NaN, 0.5, 0.75, 0.5, 0.625, 0.625, 0.875, 0.625, 0.875, 0.75, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.25, 0.125, 0.125, 0.125, 0.0, 0.25, 0.25, 0.125, 0.125, 0.375, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.125, 0.5, 0.25, 0.25, 0.25, 0.375, 0.125, 0.625, 0.375, 0.375, 0.25, 0.25, 0.25, 0.375, 0.5, NaN, 0.375, 0.6875, 0.5, 0.5, 0.875, 0.625, 0.5, 1.0, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.0, 0.25, 0.0, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.125, 0.0, 0.125, 0.25, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.25, 0.0, 0.375, 0.25, 0.5, 0.25, 0.375, 0.5, 0.375, 0.5, 0.375, 0.25, 0.625, NaN, 0.75, 0.5, 0.75, 0.75, 0.625, 0.5, 0.875, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.25, 0.125, 0.0, 0.25, 0.0, 0.375, 0.125, 0.0, 0.0, 0.375, 0.125, 0.25, 0.0, 0.125, 0.375, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.25, 0.375, 0.125, 0.625, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.5, 0.3125, 0.25, NaN, 0.5, 0.5, 0.875, 0.625, 0.5, 0.875, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.0, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.5, 0.5, 0.5, NaN, 0.25, 0.625, 0.625, 0.875, 0.625, 1.0, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.375, 0.625, 0.125, 0.125, 0.25, 0.0, 0.0, 0.125, 0.375, 0.5, 0.25, 0.5, 0.75, NaN, 0.75, 0.375, 0.75, 0.75, 0.75, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.25, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.375, 0.25, NaN, 0.25, 0.5, 0.25, 0.5, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.125, 0.375, 0.125, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.625, 0.75, NaN, 0.375, 0.625, 0.5, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.5, 0.5, 0.5, 0.125, 0.25, 0.5, 0.625, NaN, 0.625, 0.6875, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.375, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.0, 0.125, 0.125, 0.375, 0.25, 0.75, 0.375, 0.375, NaN, 0.625, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.0, 0.125, 0.125, 0.0, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.0, 0.25, 0.5, 0.5, 0.3125, 0.375, NaN, 0.75], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.25, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9097222222},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.90625},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.8854166667},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8637152778},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.859375},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8263888889},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7855902778},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.7725694444},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7638888889},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7569444444},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7291666667},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7274305556},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.71875},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7083333333},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.703125},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6927083333},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6840277778},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6822916667},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.65625},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6545138889},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6493055556},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6423611111},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6388888889},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6128472222},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6032986111},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.59375},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5894097222},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5833333333},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.578125},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5755208333},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5642361111},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5555555556},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5555555556},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5538194444},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5520833333},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.53125},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5208333333},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.515625},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4991319444},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4982638889},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4930555556},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4913194444},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4809027778},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4809027778},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4756944444},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.4704861111},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4704861111},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4600694444},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4583333333},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.40625},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.3888888889},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3663194444},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3663194444},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3559027778},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2881944444},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2881944444},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.28125},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2760416667},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2734375},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2612847222},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2517361111},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2239583333},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2161458333},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2135416667},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1987847222},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1527777778},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1371527778},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1354166667},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1267361111},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1137152778},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1041666667},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0859375},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0069444444}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..932a01cd5d252d212fbd1cf713aeeb1094da90aa
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_all/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),TabFM (default),RealTabPFN-2.5 (tuned),TabPFN-2.6 (default),TabPFN-3 (default),RealTabPFN-2.5 (default),TabICLv2 (default),LimiX (default),"AutoGluon 1.4 (best, 4h)",CatBoost (tuned + ensembled),CatBoost (tuned),RealMLP (tuned + ensembled),LightGBM (tuned + ensembled),LightGBM (tuned),TabM (tuned + ensembled),TabM (tuned),RealMLP (tuned),TabDPT (tuned + ensembled),TabDPT-Turbo (default),XGBoost (tuned + ensembled),XGBoost (tuned),xRFM (tuned + ensembled),iLTM (tuned + ensembled),ModernNCA (tuned + ensembled),EBM (tuned + ensembled),ModernNCA (tuned),TabDPT (tuned),TabM (default),ExtraTrees (tuned + ensembled),SAP-RPT-OSS (default),CatBoost (default),xRFM (tuned),BetaTabPFN (default),RandomForest (tuned + ensembled),iLTM (tuned),EBM (tuned),TorchMLP (tuned + ensembled),ExtraTrees (tuned),EBM (default),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),TorchMLP (tuned),RandomForest (tuned),XGBoost (default),TabSwift (default),LightGBM (default),FastaiMLP (tuned + ensembled),TabDPT (default),ChimeraBoost (default),RealMLP (default),FastaiMLP (tuned),xRFM (default),ModernNCA (default),KNN (tuned + ensembled),TorchMLP (default),iLTM (default),RandomForest (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),OrionMSP (default),ExtraTrees (default),Linear (tuned + ensembled),FastaiMLP (default),Linear (tuned),PerpetualBooster (tuned + ensembled),TabSTAR (default),KNN (tuned),TabFlex (default),PerpetualBooster (tuned),Linear (default),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.375,0.25,0.75,0.625,0.75,0.5,0.75,0.875,0.875,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,1.0,1.0,0.875,1.0,1.0,0.875,0.875,1.0,0.875,0.875,0.75,1.0,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.625,,0.375,0.75,0.75,0.625,1.0,0.75,0.75,0.75,1.0,1.0,0.875,0.875,0.875,0.75,0.75,1.0,0.75,0.875,1.0,1.0,0.875,1.0,1.0,0.875,1.0,0.75,0.75,0.875,1.0,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.75,0.625,,0.75,0.75,0.75,0.875,1.0,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.25,0.25,0.25,,0.625,0.375,0.6875,0.625,0.5,0.625,0.875,0.75,0.875,0.75,0.75,0.875,0.75,1.0,0.75,0.75,1.0,1.0,0.875,1.0,1.0,0.875,1.0,0.75,0.75,0.875,0.875,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.375,0.25,0.25,0.375,,0.5,0.75,0.5,0.75,0.5,0.875,0.875,0.875,0.875,0.875,0.75,0.75,0.875,0.75,0.875,1.0,1.0,0.875,1.0,1.0,0.75,1.0,0.75,0.75,0.875,0.75,1.0,0.75,0.75,0.75,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.75,1.0,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.25,0.375,0.25,0.625,0.5,,0.5,0.75,0.625,0.5,0.625,0.625,0.625,0.75,0.75,0.625,0.75,0.875,0.75,0.875,0.875,0.875,0.75,1.0,1.0,0.75,0.875,0.75,0.75,0.875,0.625,0.75,0.875,0.75,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.75,0.875,1.0,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.5,0.0,0.125,0.3125,0.25,0.5,,0.5,0.75,0.5,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.5,0.75,0.75,0.75,0.75,0.75,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.625,0.875,0.75,0.875,0.75,0.75,0.75,0.75,0.875,0.75,0.875,1.0,0.875,1.0,0.875,0.875,1.0,1.0,0.875,1.0,0.75,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.25,0.25,0.0,0.375,0.5,0.25,0.5,,0.625,0.625,0.625,0.625,0.625,0.875,0.875,0.625,0.625,0.625,0.75,0.875,0.75,0.75,0.625,0.875,0.875,0.625,0.75,0.75,0.75,0.75,0.5,0.75,0.75,0.75,0.75,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.75,0.875,0.75,0.875,0.875,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0
+LimiX (default),0.125,0.25,0.25,0.5,0.25,0.375,0.25,0.375,,0.5,0.625,0.5,0.625,0.625,0.625,0.75,0.75,0.625,0.5,0.625,0.75,0.75,0.75,0.625,0.75,0.75,0.75,0.625,0.75,0.875,0.625,0.75,0.875,0.75,0.875,0.625,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.75,0.875,0.75,0.625,0.875,1.0,1.0,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.125,0.25,0.25,0.375,0.5,0.5,0.5,0.375,0.5,,0.625,0.625,0.5,0.75,0.75,0.5,0.625,0.5,0.5,0.625,0.75,0.625,0.625,0.75,0.625,0.75,0.625,0.625,0.75,0.625,0.625,0.75,0.625,0.75,0.625,0.625,0.875,0.875,0.75,1.0,1.0,1.0,0.875,0.875,0.875,0.75,0.875,1.0,0.75,0.875,0.875,1.0,0.75,1.0,0.75,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.25,0.0,0.125,0.125,0.125,0.375,0.25,0.375,0.375,0.375,,0.625,0.375,0.375,0.375,0.75,0.75,0.625,0.75,0.5,0.75,0.75,0.625,0.75,0.875,0.625,0.75,0.75,0.75,0.75,0.5,0.75,0.75,0.625,0.625,0.875,0.75,0.875,0.75,0.875,0.75,0.75,0.875,0.75,1.0,0.875,0.75,0.75,0.875,0.875,1.0,0.875,0.875,0.875,0.75,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.25,0.0,0.125,0.25,0.125,0.375,0.25,0.375,0.5,0.375,0.375,,0.375,0.5,0.5,0.625,0.75,0.625,0.75,0.5,0.75,0.75,0.625,0.75,0.875,0.625,0.75,0.75,0.75,0.75,0.5,0.75,0.75,0.625,0.625,0.875,0.625,0.875,0.75,0.875,0.75,0.75,0.875,0.75,1.0,0.875,0.75,0.75,0.75,0.875,1.0,0.875,0.75,1.0,0.75,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.125,0.125,0.125,0.125,0.125,0.375,0.25,0.375,0.375,0.5,0.625,0.625,,0.625,0.625,0.625,0.625,0.625,0.5,0.5,0.875,0.75,0.625,0.75,0.75,0.75,0.75,0.5,0.875,0.5,0.625,0.75,0.625,0.75,0.5,0.875,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.625,0.875,0.75,0.875,0.875,0.625,0.75,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.375,0.25,0.625,0.5,0.375,,0.5,0.5,0.5,0.5,0.5,0.75,0.5,0.625,0.5,0.625,0.5,0.75,0.5,0.625,0.75,0.625,0.625,0.875,0.625,0.875,0.625,0.625,0.875,0.875,0.625,0.875,1.0,1.0,0.75,0.875,1.0,0.875,0.875,0.75,0.75,0.875,0.75,1.0,0.75,1.0,0.75,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.375,0.25,0.625,0.5,0.375,0.5,,0.5,0.5,0.5,0.5,0.75,0.5,0.5,0.5,0.625,0.5,0.75,0.5,0.625,0.75,0.625,0.625,0.875,0.625,0.875,0.625,0.625,0.875,0.875,0.625,0.875,1.0,1.0,0.75,0.875,1.0,0.875,0.875,0.75,0.75,0.875,0.75,1.0,0.75,0.875,0.75,1.0,1.0,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.125,0.25,0.125,0.125,0.25,0.375,0.25,0.375,0.25,0.5,0.25,0.375,0.375,0.5,0.5,,0.5,0.625,0.625,0.625,0.625,0.625,0.5,0.5,0.625,0.875,0.75,0.75,0.875,0.5,0.625,0.625,0.625,0.875,0.625,0.625,0.875,0.875,0.625,0.875,0.625,0.625,0.75,0.625,0.75,0.625,0.75,0.75,0.75,0.875,1.0,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.375,0.25,0.375,0.25,0.25,0.375,0.5,0.5,0.5,,0.5,0.625,0.5,0.5,0.625,0.625,0.5,0.625,0.875,0.75,0.75,0.875,0.5,0.625,0.625,0.625,0.875,0.625,0.625,0.875,0.75,0.625,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.875,0.625,0.75,0.75,0.75,0.875,0.875,0.875,0.875,1.0,0.875,1.0,0.875,1.0,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.25,0.0,0.125,0.0,0.125,0.125,0.25,0.375,0.375,0.5,0.375,0.375,0.375,0.5,0.5,0.375,0.5,,0.5,0.375,0.625,0.75,0.5,0.625,0.75,0.75,0.875,0.625,0.75,0.625,0.5,0.625,0.75,0.75,0.625,0.875,0.75,0.75,0.75,0.75,0.625,0.75,0.75,0.625,0.875,0.75,0.75,0.75,0.5,0.75,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.5,0.5,0.25,0.25,0.5,0.5,0.5,0.375,0.375,0.5,,0.375,0.625,0.625,0.5,0.75,0.625,0.625,0.625,0.875,0.5,0.5,0.5,0.75,0.625,0.75,0.625,0.75,0.625,0.75,0.75,0.75,0.625,0.75,0.75,0.625,0.75,0.625,0.75,0.75,0.875,0.625,0.75,0.75,0.875,0.875,0.875,0.875,0.75,1.0,0.875,0.875,1.0,1.0,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0
+TabDPT-Turbo (default),0.125,0.125,0.125,0.25,0.125,0.125,0.5,0.125,0.375,0.375,0.5,0.5,0.5,0.25,0.25,0.375,0.5,0.625,0.625,,0.625,0.625,0.375,0.75,0.625,0.5,0.625,0.625,0.5,0.625,0.5,0.75,0.625,0.5,0.625,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.875,1.0,0.75,0.625,0.875,0.75,0.75,0.875,0.75,0.75,0.875,0.875,0.875,1.0,0.75,0.75,1.0,1.0,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.125,0.0,0.0,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.125,0.5,0.5,0.375,0.5,0.375,0.375,0.375,,0.625,0.5,0.75,0.75,0.625,0.625,0.5,0.75,0.625,0.5,0.75,0.625,0.625,0.5,0.625,0.625,0.75,0.75,0.75,0.875,0.875,0.75,0.625,1.0,0.75,0.875,0.625,0.625,0.875,0.75,0.875,0.625,1.0,0.75,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.125,0.0,0.0,0.125,0.25,0.25,0.25,0.375,0.25,0.25,0.25,0.375,0.5,0.375,0.375,0.25,0.375,0.375,0.375,,0.5,0.625,0.75,0.625,0.625,0.5,0.625,0.625,0.5,0.75,0.625,0.625,0.5,0.625,0.625,0.75,0.875,0.75,0.875,0.875,0.75,0.625,1.0,0.75,0.875,0.625,0.625,0.875,0.75,0.875,0.625,1.0,0.75,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.375,0.25,0.375,0.375,0.375,0.375,0.5,0.5,0.5,0.375,0.5,0.5,0.625,0.5,0.5,,0.5,0.625,0.5,0.5,0.625,0.625,0.625,0.375,0.5,0.625,0.625,0.625,0.5,0.5,0.625,0.75,0.625,0.75,0.75,0.625,0.75,0.625,0.75,0.625,0.625,0.625,0.75,0.75,0.75,1.0,0.875,1.0,1.0,0.75,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.125,0.375,0.25,0.25,0.25,0.25,0.375,0.375,0.5,0.5,0.375,0.25,0.25,0.25,0.375,0.5,,0.5,0.5,0.75,0.5,0.5,0.75,0.5,0.5,0.75,0.5,0.625,0.625,0.625,0.625,0.75,0.875,0.75,0.75,0.875,0.625,0.75,0.625,0.625,0.75,0.625,0.875,0.875,0.75,0.75,1.0,0.875,0.875,1.0,0.875,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.125,0.25,0.375,0.125,0.125,0.25,0.5,0.5,0.375,0.375,0.25,0.375,0.375,0.25,0.25,0.375,0.5,,0.5,0.6875,0.5,0.5,0.75,0.5,0.625,0.75,0.625,0.625,0.625,0.5,0.625,0.75,0.75,0.625,0.625,0.75,0.625,0.75,0.625,0.75,0.75,0.625,0.75,0.75,0.875,0.75,1.0,0.875,0.875,0.875,1.0,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.125,0.125,0.125,0.125,0.25,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.25,0.25,0.25,0.125,0.125,0.25,0.375,0.5,0.375,0.375,0.5,0.5,0.5,,0.625,0.375,0.25,0.625,0.625,0.75,0.625,0.5,0.625,0.75,0.875,0.5,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.5,0.625,0.75,0.75,0.625,0.75,0.75,0.875,0.75,1.0,1.0,0.875,0.75,0.875,0.75,0.875,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.125,0.0,0.125,0.0,0.0,0.125,0.375,0.25,0.25,0.375,0.25,0.25,0.25,0.5,0.5,0.25,0.25,0.125,0.375,0.375,0.375,0.375,0.5,0.25,0.3125,0.375,,0.375,0.5,0.75,0.625,0.625,0.75,0.5,0.625,0.75,0.625,0.5,0.75,0.375,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.75,0.625,0.75,0.75,1.0,0.875,0.875,1.0,1.0,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.25,0.25,0.5,0.375,0.375,0.25,0.25,0.375,0.125,0.375,0.5,0.5,0.375,0.5,0.5,0.625,0.625,,0.5,0.375,0.5,0.625,0.5,0.625,0.375,0.625,0.625,0.5,0.625,0.625,0.625,0.625,0.5,0.625,0.625,0.625,0.625,0.625,0.875,0.625,0.75,0.75,0.875,0.625,0.875,0.875,0.75,0.75,0.875,0.875,1.0,1.0,0.875,0.875,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0
+TabM (default),0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.125,0.25,0.25,0.125,0.125,0.25,0.5,0.5,0.25,0.375,0.375,0.5,0.5,0.75,0.5,0.5,,0.5,0.375,0.625,0.5,0.75,0.5,0.625,0.75,0.625,0.5,0.625,0.625,0.625,0.625,0.625,0.75,0.625,0.5,0.625,0.625,0.5,0.625,0.875,0.75,0.75,0.75,1.0,0.75,0.75,0.75,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.375,0.25,0.25,0.5,0.375,0.375,0.5,0.5,0.375,0.5,0.375,0.375,0.375,0.375,0.25,0.25,0.375,0.25,0.625,0.5,,0.5,0.5,0.375,0.625,0.75,0.5,0.5,0.5,0.8125,0.5,0.625,0.625,0.625,0.75,0.5,0.75,0.75,0.5,0.75,0.75,0.625,0.75,0.875,0.625,1.0,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.75,0.75,0.75,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0
+SAP-RPT-OSS (default),0.25,0.0,0.125,0.125,0.25,0.375,0.25,0.5,0.375,0.375,0.5,0.5,0.375,0.375,0.375,0.375,0.375,0.5,0.5,0.5,0.5,0.5,0.625,0.5,0.5,0.375,0.375,0.5,0.625,0.5,,0.5,0.625,0.625,0.5,0.5,0.625,0.625,0.5,0.625,0.5,0.5,0.625,0.5,0.5,0.5,0.625,0.625,0.5,0.625,0.75,0.75,0.625,0.625,0.625,0.75,0.625,0.625,0.75,0.75,0.625,0.75,0.875,0.875,0.875,0.875,1.0,0.75,0.875,0.875,0.875,0.875,1.0
+CatBoost (default),0.0,0.0,0.125,0.0,0.0,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.125,0.125,0.375,0.375,0.375,0.25,0.25,0.25,0.25,0.5,0.5,0.375,0.25,0.375,0.375,0.375,0.5,0.5,,0.625,0.25,0.5,0.75,0.375,0.625,0.5,0.625,0.625,0.625,0.625,0.625,0.5,0.75,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.75,1.0,1.0,1.0,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.375,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.25,0.375,0.25,0.5,0.5,0.625,0.375,0.375,,0.5,0.75,0.5,0.375,0.375,0.75,0.375,0.75,0.5,0.5,0.75,0.5,0.625,0.5,0.5,0.625,0.625,0.75,0.625,1.0,0.625,1.0,1.0,0.75,0.75,0.875,0.875,1.0,1.0,0.875,0.75,0.75,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.25,0.125,0.125,0.125,0.125,0.25,0.25,0.5,0.375,0.375,0.375,0.5,0.375,0.5,0.5,0.375,0.25,0.375,0.375,0.75,0.5,,0.5,0.625,0.875,0.625,0.375,0.5,0.625,0.625,0.5,0.625,0.5,0.5,0.5,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.875,0.875,0.75,0.75,0.75,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.125,0.125,0.125,0.25,0.125,0.375,0.25,0.125,0.375,0.375,0.375,0.5,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.5,0.5,0.375,0.375,0.375,0.375,0.375,0.625,0.5,0.25,0.5,0.5,0.25,0.5,,0.5,0.5,0.5,0.625,0.5,0.75,0.625,0.5,0.875,0.625,0.625,0.5,0.5,0.625,0.625,0.625,0.625,0.75,0.625,0.875,0.875,0.75,0.875,0.875,0.875,0.875,1.0,0.75,0.75,0.625,0.875,0.875,1.0,0.875,1.0,0.75,1.0,1.0
+iLTM (tuned),0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.25,0.375,0.375,0.125,0.125,0.125,0.375,0.375,0.375,0.375,0.125,0.25,0.25,0.375,0.375,0.5,0.375,0.375,0.25,0.25,0.375,0.375,0.5,0.5,0.25,0.5,0.375,0.5,,0.375,0.625,0.75,0.625,0.5,0.5,0.625,0.625,0.5,0.625,0.625,0.75,0.625,0.625,0.625,0.75,0.625,0.875,0.875,0.875,1.0,0.875,0.75,0.75,0.875,1.0,0.75,0.75,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.125,0.25,0.375,0.25,0.125,0.125,0.125,0.125,0.25,0.375,0.375,0.375,0.375,0.5,0.375,0.5,0.125,0.375,0.375,0.25,0.5,0.375,0.625,0.625,0.125,0.5,0.625,,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.375,0.5,0.375,0.625,0.625,0.75,0.625,0.75,0.75,0.875,0.75,0.875,1.0,0.75,0.75,0.75,0.75,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.5,0.5,0.5,0.375,0.5,0.375,0.375,0.625,0.375,0.5,0.375,0.5,,0.5,0.625,0.5,0.5,0.75,0.5,0.625,0.5,0.625,0.625,0.75,0.5,0.75,0.75,0.75,0.625,0.75,1.0,0.75,0.75,0.75,0.875,0.75,0.875,0.875,0.875,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.25,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.375,0.25,0.375,0.5,0.1875,0.5,0.5,0.25,0.625,0.375,0.25,0.5,0.5,,0.375,0.5,0.5,0.5,0.5,0.5,0.625,0.75,0.5,0.625,0.75,0.5,0.625,0.625,0.625,0.875,0.875,0.75,0.875,0.875,0.875,1.0,0.875,0.625,0.75,0.625,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0
+EBM (default),0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.0,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.125,0.25,0.375,0.625,0.375,0.375,0.5,0.375,0.375,0.625,0.5,0.5,0.375,0.5,0.375,0.625,,0.375,0.5,0.5,0.625,0.375,0.5,0.375,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.875,1.0,0.625,0.75,0.75,0.75,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.0,0.25,0.25,0.25,0.0,0.0,0.375,0.375,0.375,0.375,0.25,0.125,0.125,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.375,0.5,0.375,0.25,0.375,0.25,0.5,0.5,0.5,0.5,0.625,,0.625,0.625,0.625,0.5,0.625,0.625,0.5,0.75,0.75,0.75,0.625,0.75,0.5,0.625,0.75,0.875,0.75,0.875,0.875,0.875,0.75,0.75,0.875,0.75,0.875,1.0,0.875,1.0,0.875,1.0,0.875,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.0,0.25,0.25,0.25,0.0,0.0,0.375,0.25,0.25,0.25,0.25,0.125,0.125,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.375,0.5,0.5,0.5,0.5,0.5,0.375,,0.625,0.625,0.375,0.625,0.5,0.5,0.75,0.875,0.75,0.625,0.75,0.625,0.625,0.75,0.875,0.75,0.875,0.875,0.75,0.75,0.875,1.0,1.0,0.875,1.0,0.875,0.875,0.875,1.0,0.875,1.0
+TorchMLP (tuned),0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.125,0.25,0.375,0.375,0.5,0.375,0.375,0.375,0.375,0.5,0.5,0.5,0.375,0.5,0.25,0.5,0.5,0.375,0.375,,0.5,0.375,0.375,0.625,0.5,0.625,0.625,0.625,0.625,0.75,0.75,0.875,0.875,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0
+RandomForest (tuned),0.0,0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.125,0.25,0.25,0.375,0.125,0.125,0.375,0.25,0.375,0.375,0.25,0.375,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.375,0.25,0.5,0.375,0.25,0.375,0.125,0.375,0.5,0.5,0.5,0.375,0.375,0.375,0.5,,0.5,0.625,0.5,0.5,0.625,0.625,0.625,0.625,0.625,0.375,0.75,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.75,0.75,0.625,0.875,0.875,0.875,0.875,1.0,0.75,1.0,1.0
+XGBoost (default),0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.0,0.125,0.0,0.0,0.25,0.25,0.125,0.25,0.125,0.0,0.0,0.375,0.25,0.25,0.375,0.375,0.375,0.25,0.5,0.5,0.5,0.5,0.5,0.375,0.5,0.625,0.375,0.5,0.625,0.5,0.625,0.625,0.5,,0.625,0.625,0.5,0.5,0.75,0.625,0.625,0.5,0.75,0.75,0.875,1.0,0.875,0.75,0.75,0.875,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.25,0.25,0.125,0.125,0.25,0.125,0.125,0.375,0.25,0.25,0.375,0.0,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.375,0.25,0.5,0.25,0.375,0.5,0.375,0.375,0.5,0.5,0.375,0.5,0.375,0.375,0.625,0.375,0.375,,0.75,0.625,0.625,0.625,0.75,0.75,0.5,0.625,0.75,0.75,0.75,0.875,0.75,0.75,0.875,0.75,0.75,0.875,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0
+LightGBM (default),0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.25,0.125,0.125,0.375,0.375,0.25,0.5,0.375,0.375,0.5,0.25,0.375,0.375,0.5,0.5,0.5,0.375,0.625,0.375,0.25,0.625,0.375,0.5,0.375,0.5,0.375,0.25,,0.5,0.375,0.625,0.625,0.625,0.625,0.625,0.75,0.75,0.75,0.5,0.875,0.875,0.875,0.75,0.75,1.0,0.75,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.125,0.0,0.125,0.125,0.0,0.125,0.0,0.125,0.25,0.0,0.25,0.25,0.125,0.25,0.25,0.25,0.375,0.25,0.25,0.375,0.375,0.375,0.375,0.25,0.25,0.375,0.375,0.375,0.375,0.5,0.375,0.375,0.5,0.375,0.5,0.25,0.375,0.375,0.5,0.375,0.5,0.5,0.5,0.5,0.5,0.375,0.5,,0.375,0.5,0.5,0.625,0.625,0.625,0.625,0.5,0.75,0.625,0.75,0.75,0.625,0.75,0.875,1.0,0.875,0.75,0.875,0.875,1.0,0.875,1.0,0.875,1.0
+TabDPT (default),0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.375,0.25,0.125,0.25,0.375,0.25,0.25,0.25,0.25,0.5,0.125,0.125,0.375,0.375,0.375,0.375,0.375,0.25,0.375,0.125,0.375,0.25,0.5,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.5,0.375,0.625,0.625,,0.5,0.625,0.75,0.75,0.5,0.875,0.625,0.75,0.625,0.75,0.75,0.75,1.0,0.5,0.75,0.625,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0
+ChimeraBoost (default),0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.25,0.375,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.25,0.375,0.5,0.25,0.375,0.25,0.375,0.25,0.375,0.375,0.25,0.5,0.25,0.25,0.25,0.125,0.375,0.375,0.25,0.375,0.375,0.5,0.5,,0.625,0.5,0.5,0.625,0.625,0.625,0.75,0.5,0.625,0.875,0.625,0.625,0.75,0.75,0.75,0.875,1.0,1.0,0.875,0.875,1.0,0.875,1.0
+RealMLP (default),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.25,0.0,0.125,0.0,0.0,0.0,0.25,0.25,0.0,0.25,0.0,0.25,0.25,0.25,0.25,0.25,0.125,0.25,0.375,0.375,0.25,0.375,0.375,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.25,0.5,0.25,0.25,0.25,0.375,0.375,0.375,0.25,0.375,0.5,0.375,0.375,,0.5,0.625,0.5,0.75,0.625,0.625,0.625,0.625,0.75,0.625,0.75,0.75,0.75,0.75,0.875,1.0,0.875,1.0,0.875,1.0,0.875,1.0
+FastaiMLP (tuned),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.25,0.125,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.375,0.25,0.375,0.25,0.25,0.25,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.375,0.375,0.25,0.5,0.5,,0.625,0.625,0.625,0.5,0.75,0.5,0.625,0.625,0.625,0.75,0.75,1.0,0.75,0.875,0.875,0.875,0.875,0.875,1.0,0.875,1.0
+xRFM (default),0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.125,0.25,0.25,0.125,0.125,0.125,0.125,0.25,0.375,0.375,0.0,0.25,0.25,0.25,0.25,0.125,0.25,0.125,0.375,0.25,0.0,0.25,0.25,0.375,0.25,0.25,0.375,0.25,0.25,0.25,0.25,0.375,0.5,0.5,0.375,0.375,0.25,0.5,0.375,0.375,,0.5,0.625,0.625,0.625,0.75,0.75,0.75,0.75,0.625,0.625,0.625,0.625,0.75,0.875,0.875,0.75,0.75,0.875,0.75,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.375,0.25,0.375,0.375,0.25,0.375,0.25,0.375,0.125,0.125,0.375,0.375,0.25,0.5,0.375,0.25,0.625,0.25,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.5,,0.625,0.75,0.5,0.875,0.625,0.625,0.75,0.75,0.875,0.75,0.875,0.875,0.625,0.75,0.875,0.875,0.875,1.0,1.0
+KNN (tuned + ensembled),0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.25,0.25,0.125,0.25,0.25,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.0,0.125,0.125,0.25,0.125,0.125,0.25,0.0,0.375,0.25,0.0,0.25,0.125,0.125,0.25,0.25,0.125,0.25,0.375,0.375,0.125,0.25,0.25,0.25,0.25,0.375,0.125,0.375,0.25,0.375,0.375,0.375,,0.375,0.375,0.375,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.75,0.375,1.0,0.625,0.75,0.625,0.875,1.0
+TorchMLP (default),0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.0,0.125,0.125,0.0,0.125,0.25,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.25,0.25,0.5,0.375,0.375,0.375,0.5,0.375,0.25,0.625,,0.375,0.375,0.75,0.75,0.625,0.625,0.625,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0
+iLTM (default),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.25,0.125,0.0,0.0,0.25,0.0,0.125,0.0,0.0,0.25,0.25,0.125,0.375,0.0,0.25,0.125,0.25,0.0,0.0,0.25,0.25,0.0,0.125,0.125,0.25,0.25,0.0,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.375,0.5,0.625,0.625,,0.625,0.75,0.625,0.5,0.625,0.625,0.625,0.625,0.875,0.875,0.75,0.75,0.875,0.75,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.25,0.25,0.125,0.375,0.0,0.25,0.25,0.125,0.125,0.25,0.25,0.125,0.375,0.25,0.25,0.375,0.125,0.125,0.125,0.5,0.375,0.375,0.5,0.375,0.5,0.25,0.125,0.625,0.625,0.375,,0.625,0.625,0.375,0.625,0.75,0.5,0.75,0.75,0.75,0.875,0.625,0.75,0.75,0.75,1.0
+TabSTAR (tuned),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.25,0.0,0.0,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.25,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.25,0.25,0.375,0.375,0.375,0.25,0.375,0.625,0.25,0.25,0.375,,0.4375,0.375,0.5,0.75,0.625,0.75,0.75,1.0,0.875,0.625,0.75,0.875,0.75,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.125,0.25,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.25,0.25,0.125,0.25,0.375,0.25,0.375,0.625,0.25,0.375,0.375,0.5625,,0.5,0.5,0.75,0.5,0.75,0.75,1.0,0.875,0.625,0.75,0.875,0.75,1.0
+OrionMSP (default),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.25,0.0,0.0,0.125,0.0,0.375,0.125,0.0,0.125,0.125,0.125,0.25,0.25,0.0,0.25,0.125,0.25,0.25,0.125,0.125,0.125,0.125,0.375,0.25,0.375,0.375,0.375,0.25,0.25,0.625,0.375,0.5,0.625,0.625,0.5,,0.625,0.625,0.625,0.625,0.75,0.875,0.875,0.75,0.75,0.75,0.75,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.25,0.125,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.0,0.375,0.25,0.25,0.375,0.25,0.5,0.375,0.375,0.375,0.5,0.5,0.375,,0.5,0.75,0.5,0.625,0.625,0.875,0.625,0.875,0.75,0.875,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.25,0.125,0.125,0.125,0.0,0.25,0.25,0.125,0.125,0.375,0.125,0.25,0.125,0.25,0.25,0.125,0.25,0.25,0.125,0.5,0.25,0.25,0.25,0.375,0.125,0.625,0.375,0.375,0.25,0.25,0.25,0.375,0.5,,0.375,0.6875,0.5,0.5,0.875,0.625,0.5,1.0,0.75,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.0,0.25,0.0,0.25,0.25,0.125,0.125,0.25,0.125,0.125,0.0,0.125,0.25,0.125,0.125,0.0,0.0,0.25,0.25,0.25,0.0,0.375,0.25,0.5,0.25,0.375,0.5,0.375,0.5,0.375,0.25,0.625,,0.75,0.5,0.75,0.75,0.625,0.5,0.875,0.75,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.25,0.125,0.0,0.25,0.0,0.375,0.125,0.0,0.0,0.375,0.125,0.25,0.0,0.125,0.375,0.125,0.25,0.25,0.125,0.375,0.25,0.25,0.25,0.375,0.125,0.625,0.25,0.375,0.25,0.25,0.25,0.375,0.5,0.3125,0.25,,0.5,0.5,0.875,0.625,0.5,0.875,0.75,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.0,0.25,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.125,0.125,0.25,0.25,0.25,0.25,0.375,0.5,0.5,0.5,,0.25,0.625,0.625,0.875,0.625,1.0,0.875
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.125,0.125,0.375,0.625,0.125,0.125,0.25,0.0,0.0,0.125,0.375,0.5,0.25,0.5,0.75,,0.75,0.375,0.75,0.75,0.75,1.0
+KNN (tuned),0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.0,0.25,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.25,0.0,0.125,0.25,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.375,0.25,,0.25,0.5,0.25,0.5,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.25,0.125,0.375,0.125,0.25,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.625,0.75,,0.375,0.625,0.5,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.125,0.125,0.25,0.25,0.25,0.25,0.125,0.5,0.5,0.5,0.125,0.25,0.5,0.625,,0.625,0.6875,0.875
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.0,0.25,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.25,0.0,0.125,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.375,0.125,0.25,0.25,0.125,0.125,0.25,0.25,0.0,0.125,0.125,0.375,0.25,0.75,0.375,0.375,,0.625,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.0,0.125,0.125,0.0,0.25,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.0,0.25,0.5,0.5,0.3125,0.375,,0.75
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.25,
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
index 68ff00209320c625c53c41749197ccf392dc8865..03242a56c53e5b616e17eed5ee301123fb1e47b8 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Multiclass | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..fec1bc7886a1513d9dcd7e82adf9b6f6bbc6aa77
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_multiclass/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":34554.0,"score":1.0,"rank":1.0,"hrank":1.0,"imp":0.0,"train_time":3.94,"predict_time":3.192,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":33174.0,"score":0.852,"rank":2.0,"hrank":2.0,"imp":4.445,"train_time":259.67,"predict_time":0.61,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":31548.0,"score":0.85,"rank":3.0,"hrank":3.0,"imp":4.498,"train_time":134.59,"predict_time":8.308,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":29818.0,"score":0.849,"rank":4.0,"hrank":4.0,"imp":4.526,"train_time":0.93,"predict_time":0.273,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":4,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":28012.0,"score":0.791,"rank":5.0,"hrank":5.0,"imp":6.149,"train_time":276.98,"predict_time":4.897,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":26991.0,"score":0.746,"rank":6.0,"hrank":6.0,"imp":7.385,"train_time":134.59,"predict_time":2.818,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":6,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":25644.0,"score":0.67,"rank":7.0,"hrank":7.0,"imp":9.385,"train_time":1.84,"predict_time":0.516,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":7,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":24744.0,"score":0.502,"rank":8.0,"hrank":8.0,"imp":13.523,"train_time":35.65,"predict_time":17.704,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":8,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":24030.0,"score":0.474,"rank":9.0,"hrank":9.0,"imp":14.192,"train_time":6601.46,"predict_time":0.107,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":23243.0,"score":0.473,"rank":10.0,"hrank":10.0,"imp":14.205,"train_time":0.48,"predict_time":0.143,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":10,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":22438.0,"score":0.47,"rank":11.0,"hrank":11.0,"imp":14.275,"train_time":61.47,"predict_time":0.154,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":21690.0,"score":0.467,"rank":12.0,"hrank":12.0,"imp":14.338,"train_time":157.4,"predict_time":4.579,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":21043.0,"score":0.466,"rank":13.0,"hrank":13.0,"imp":14.366,"train_time":157.4,"predict_time":0.401,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":20253.0,"score":0.464,"rank":14.0,"hrank":14.0,"imp":14.421,"train_time":0.82,"predict_time":0.278,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":14,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":19480.0,"score":0.452,"rank":15.0,"hrank":15.0,"imp":14.687,"train_time":61.47,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":15,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":18724.0,"score":0.446,"rank":16.0,"hrank":16.0,"imp":14.817,"train_time":6601.46,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":16,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":18005.0,"score":0.438,"rank":17.0,"hrank":17.0,"imp":15.012,"train_time":1261.24,"predict_time":0.234,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":17293.0,"score":0.43,"rank":18.0,"hrank":18.0,"imp":15.197,"train_time":1261.24,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":18,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":16562.0,"score":0.42,"rank":19.0,"hrank":19.0,"imp":15.424,"train_time":456.33,"predict_time":1.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":19,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":15859.0,"score":0.408,"rank":20.0,"hrank":20.0,"imp":15.693,"train_time":456.33,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":20,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":15134.0,"score":0.408,"rank":21.0,"hrank":21.0,"imp":15.693,"train_time":0.32,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":21,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":14414.0,"score":0.377,"rank":22.0,"hrank":22.0,"imp":16.367,"train_time":1229.42,"predict_time":21.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":22,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":13717.0,"score":0.368,"rank":23.0,"hrank":23.0,"imp":16.568,"train_time":66.23,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":23,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":13046.0,"score":0.351,"rank":24.0,"hrank":24.0,"imp":16.943,"train_time":66.23,"predict_time":0.196,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":12359.0,"score":0.335,"rank":25.0,"hrank":25.0,"imp":17.279,"train_time":4378.88,"predict_time":25.595,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":25,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":11659.0,"score":0.332,"rank":26.0,"hrank":26.0,"imp":17.34,"train_time":3.59,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":26,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":10927.0,"score":0.306,"rank":27.0,"hrank":27.0,"imp":17.893,"train_time":13.87,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":10208.0,"score":0.306,"rank":28.0,"hrank":28.0,"imp":17.912,"train_time":4378.88,"predict_time":2.216,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":28,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":9576.0,"score":0.279,"rank":29.0,"hrank":29.0,"imp":18.478,"train_time":1.48,"predict_time":0.535,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":29,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":8935.0,"score":0.222,"rank":30.0,"hrank":30.0,"imp":19.642,"train_time":2.79,"predict_time":1.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":30,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":8262.0,"score":0.192,"rank":31.0,"hrank":31.0,"imp":20.252,"train_time":28.44,"predict_time":0.174,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":7604.0,"score":0.172,"rank":32.0,"hrank":32.0,"imp":20.642,"train_time":1229.42,"predict_time":7.785,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":32,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":7143.0,"score":0.172,"rank":33.0,"hrank":33.0,"imp":20.642,"train_time":52.48,"predict_time":7.785,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":33,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":6522.0,"score":0.141,"rank":34.0,"hrank":34.0,"imp":21.248,"train_time":28.44,"predict_time":0.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":5793.0,"score":0.087,"rank":35.0,"hrank":35.0,"imp":22.291,"train_time":0.23,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":35,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":5284.0,"score":0.025,"rank":36.0,"hrank":36.0,"imp":23.459,"train_time":0.58,"predict_time":0.199,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":36,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":4593.0,"score":0.016,"rank":37.0,"hrank":37.0,"imp":23.624,"train_time":4391.01,"predict_time":275.95,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":37,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":4011.0,"score":0.0,"rank":38.0,"hrank":38.0,"imp":24.192,"train_time":2341.94,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":38,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":3320.0,"score":0.0,"rank":39.0,"hrank":39.0,"imp":24.433,"train_time":1142.35,"predict_time":0.542,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":39,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":2811.0,"score":0.0,"rank":40.0,"hrank":40.0,"imp":24.458,"train_time":1142.35,"predict_time":0.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":40,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":2082.0,"score":0.0,"rank":41.0,"hrank":41.0,"imp":24.51,"train_time":14.53,"predict_time":1.747,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1461.0,"score":0.0,"rank":42.0,"hrank":42.0,"imp":24.577,"train_time":2341.94,"predict_time":0.172,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":42,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":43.0,"hrank":43.0,"imp":27.419,"train_time":0.18,"predict_time":0.007,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":43,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":342.0,"score":0.0,"rank":44.0,"hrank":44.0,"imp":27.476,"train_time":4391.01,"predict_time":38.524,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":44,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-331.0,"score":0.0,"rank":45.0,"hrank":45.0,"imp":27.597,"train_time":41.21,"predict_time":39.405,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":45,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-972.0,"score":0.0,"rank":46.0,"hrank":46.0,"imp":27.983,"train_time":1322.5,"predict_time":0.545,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":46,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-1604.0,"score":0.0,"rank":47.0,"hrank":47.0,"imp":28.071,"train_time":9.89,"predict_time":9.237,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":47,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-2323.0,"score":0.0,"rank":48.0,"hrank":48.0,"imp":28.548,"train_time":1322.5,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":48,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":-3055.0,"score":0.0,"rank":49.0,"hrank":49.0,"imp":29.747,"train_time":0.21,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":49,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-3755.0,"score":0.0,"rank":50.0,"hrank":50.0,"imp":32.399,"train_time":2009.14,"predict_time":3.767,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":50,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-4442.0,"score":0.0,"rank":51.0,"hrank":51.0,"imp":33.951,"train_time":2009.14,"predict_time":0.974,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":51,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-5113.0,"score":0.0,"rank":52.0,"hrank":52.0,"imp":34.755,"train_time":4.05,"predict_time":0.074,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":52,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-5810.0,"score":0.0,"rank":53.0,"hrank":53.0,"imp":34.88,"train_time":3.34,"predict_time":0.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":53,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-6530.0,"score":0.0,"rank":54.0,"hrank":54.0,"imp":36.901,"train_time":3349.22,"predict_time":1.469,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-7255.0,"score":0.0,"rank":55.0,"hrank":55.0,"imp":36.958,"train_time":3349.22,"predict_time":6.037,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":55,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-7958.0,"score":0.0,"rank":56.0,"hrank":56.0,"imp":37.098,"train_time":66.06,"predict_time":1.783,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":56,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-8689.0,"score":0.0,"rank":57.0,"hrank":57.0,"imp":39.799,"train_time":16.78,"predict_time":6.764,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":57,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-9402.0,"score":0.0,"rank":58.0,"hrank":58.0,"imp":41.571,"train_time":6.51,"predict_time":0.162,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-10120.0,"score":0.0,"rank":59.0,"hrank":59.0,"imp":43.96,"train_time":553.49,"predict_time":0.174,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-10876.0,"score":0.0,"rank":60.0,"hrank":60.0,"imp":45.54,"train_time":553.49,"predict_time":1.636,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":60,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-11649.0,"score":0.0,"rank":61.0,"hrank":61.0,"imp":46.483,"train_time":65.84,"predict_time":3.48,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-12439.0,"score":0.0,"rank":62.0,"hrank":62.0,"imp":51.593,"train_time":30.07,"predict_time":0.562,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":62,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":-13086.0,"score":0.0,"rank":63.0,"hrank":63.0,"imp":52.398,"train_time":0.23,"predict_time":0.042,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-13834.0,"score":0.0,"rank":64.0,"hrank":64.0,"imp":53.969,"train_time":30.07,"predict_time":0.124,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-14639.0,"score":0.0,"rank":65.0,"hrank":65.0,"imp":54.442,"train_time":2.73,"predict_time":0.083,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":65,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-15426.0,"score":0.0,"rank":66.0,"hrank":66.0,"imp":54.991,"train_time":30.19,"predict_time":0.134,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":66,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-16140.0,"score":0.0,"rank":67.0,"hrank":67.0,"imp":57.946,"train_time":30.19,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-17040.0,"score":0.0,"rank":68.0,"hrank":68.0,"imp":58.227,"train_time":0.13,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":68,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-18387.0,"score":0.0,"rank":69.0,"hrank":69.0,"imp":59.135,"train_time":76.41,"predict_time":2.853,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-19408.0,"score":0.0,"rank":70.0,"hrank":70.0,"imp":59.496,"train_time":2.73,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":70,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-21214.0,"score":0.0,"rank":71.0,"hrank":71.0,"imp":59.669,"train_time":3.14,"predict_time":0.921,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-22944.0,"score":0.0,"rank":72.0,"hrank":72.0,"imp":59.719,"train_time":0.11,"predict_time":0.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":72,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-24570.0,"score":0.0,"rank":73.0,"hrank":73.0,"imp":72.021,"train_time":76.41,"predict_time":0.258,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-25950.0,"score":0.0,"rank":74.0,"hrank":74.0,"imp":84.461,"train_time":0.14,"predict_time":0.123,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/pareto_front_explorer.html
index 63ec50f5a2aae1b65b307f6295e4909fc760f8f1..46dea7b69f638cd5ea3654e1c915d268342e3719 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7827549071,"imp":37.0982410724,"elo":-7958.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0218721746,"imp":17.3395361789,"elo":11658.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0300453503,"imp":15.1970143595,"elo":17292.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.234055305,"imp":15.012462316,"elo":18005.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0128154423,"imp":22.2905525261,"elo":5792.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0152880152,"imp":16.5680326571,"elo":13716.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1959219282,"imp":16.9425810411,"elo":13045.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.014064448,"imp":17.8927578513,"elo":10927.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0119814523,"imp":14.8173560152,"elo":18724.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1069197134,"imp":14.192258111,"elo":24029.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0418367287,"imp":52.3978789302,"elo":-13086.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0088855815,"imp":21.2482114687,"elo":6522.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739987392,"imp":20.2523403303,"elo":8261.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0832738002,"imp":54.4419294386,"elo":-14639.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1735651487,"imp":43.9601644927,"elo":-10120.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"imp":45.5403937799,"elo":-10875.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0474691168,"imp":29.7473637296,"elo":-3054.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.4009232048,"imp":14.3659785469,"elo":21042.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.5785391893,"imp":14.3376604523,"elo":21690.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":3.1918999344,"imp":0.0,"elo":34554.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7472229818,"imp":24.5099501596,"elo":2081.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2155277801,"imp":17.9124144533,"elo":10208.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":25.5948192523,"imp":17.2788666655,"elo":12359.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7641563624,"imp":39.799071256,"elo":-8688.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"imp":59.4960081453,"elo":-19408.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1240589189,"imp":53.9690546132,"elo":-13833.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5623533631,"imp":51.5926415496,"elo":-12438.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0069604217,"imp":27.4194188138,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0154110106,"imp":14.6872814824,"elo":19479.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1538951715,"imp":14.2751202196,"elo":22437.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.1615432661,"imp":41.5705102944,"elo":-9401.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0552248943,"imp":28.548024354,"elo":-2323.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5454919741,"imp":27.9827612422,"elo":-972.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5348496069,"imp":18.4784935089,"elo":9576.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8184836382,"imp":7.3854101119,"elo":26990.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3079343419,"imp":4.4976437918,"elo":31548.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2367423058,"imp":28.0714246531,"elo":-1604.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"imp":27.5967758404,"elo":-331.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.5244814723,"imp":27.4759618173,"elo":342.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":275.9499442925,"imp":23.6244714479,"elo":4593.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.277558181,"imp":14.4214382191,"elo":20252.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":17.7035952873,"imp":13.5232072922,"elo":24743.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0232842395,"imp":59.7189648865,"elo":-22944.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.056543605,"imp":19.6416425074,"elo":8935.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426377996,"imp":14.2045636484,"elo":23243.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0740791906,"imp":34.7546595439,"elo":-5112.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0574085901,"imp":24.4579757413,"elo":2811.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5419846432,"imp":24.4328219572,"elo":3320.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5160116403,"imp":9.3847277445,"elo":25643.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2726374657,"imp":4.5260970421,"elo":29818.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4795389483,"imp":46.4830894898,"elo":-11648.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.468865642,"imp":36.9007036438,"elo":-6530.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.0373758651,"imp":36.9583859241,"elo":-7254.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1994643012,"imp":23.4592002877,"elo":5283.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0276328914,"imp":34.8795471811,"elo":-5810.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0266472235,"imp":24.1915976671,"elo":4010.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1720376576,"imp":24.5765776663,"elo":1461.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"imp":15.6926314585,"elo":15134.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"imp":15.692631292,"elo":15858.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.1704905161,"imp":15.4240395653,"elo":16562.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"imp":20.6424796077,"elo":7142.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"imp":20.6424750924,"elo":7603.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":21.7408394399,"imp":16.3672216796,"elo":14414.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9209350283,"imp":59.6685107498,"elo":-21214.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.9740680655,"imp":33.9505172259,"elo":-4441.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.7671484184,"imp":32.399105052,"elo":-3755.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7827549071,"x_train":66.0588444215,"imp":37.0982410724,"elo":-7958.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0218721746,"x_train":3.5923477438,"imp":17.3395361789,"elo":11658.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0300453503,"x_train":1261.2440700178,"imp":15.1970143595,"elo":17292.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.234055305,"x_train":1261.2440700178,"imp":15.012462316,"elo":18005.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0128154423,"x_train":0.2281173573,"imp":22.2905525261,"elo":5792.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0152880152,"x_train":66.2292774111,"imp":16.5680326571,"elo":13716.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1959219282,"x_train":66.2292774111,"imp":16.9425810411,"elo":13045.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.014064448,"x_train":13.8733921956,"imp":17.8927578513,"elo":10927.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0119814523,"x_train":6601.4555400381,"imp":14.8173560152,"elo":18724.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1069197134,"x_train":6601.4555400381,"imp":14.192258111,"elo":24029.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0418367287,"x_train":0.2342836282,"imp":52.3978789302,"elo":-13086.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0088855815,"x_train":28.4407435042,"imp":21.2482114687,"elo":6522.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739987392,"x_train":28.4407435042,"imp":20.2523403303,"elo":8261.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0832738002,"x_train":2.7277827402,"imp":54.4419294386,"elo":-14639.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1735651487,"x_train":553.4880496571,"imp":43.9601644927,"elo":-10120.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"x_train":553.4880496571,"imp":45.5403937799,"elo":-10875.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0474691168,"x_train":0.2060542703,"imp":29.7473637296,"elo":-3054.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.4009232048,"x_train":157.4028552082,"imp":14.3659785469,"elo":21042.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.5785391893,"x_train":157.4028552082,"imp":14.3376604523,"elo":21690.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":3.1918999344,"x_train":3.9378111103,"imp":0.0,"elo":34554.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7472229818,"x_train":14.534943689,"imp":24.5099501596,"elo":2081.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2155277801,"x_train":4378.8792899745,"imp":17.9124144533,"elo":10208.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":25.5948192523,"x_train":4378.8792899745,"imp":17.2788666655,"elo":12359.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7641563624,"x_train":16.7832602994,"imp":39.799071256,"elo":-8688.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"x_train":2.7257969394,"imp":59.4960081453,"elo":-19408.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1240589189,"x_train":30.0706477892,"imp":53.9690546132,"elo":-13833.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5623533631,"x_train":30.0706477892,"imp":51.5926415496,"elo":-12438.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0069604217,"x_train":0.1776920981,"imp":27.4194188138,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0154110106,"x_train":61.4738106436,"imp":14.6872814824,"elo":19479.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1538951715,"x_train":61.4738106436,"imp":14.2751202196,"elo":22437.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.1615432661,"x_train":6.5101464138,"imp":41.5705102944,"elo":-9401.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0552248943,"x_train":1322.498714333,"imp":28.548024354,"elo":-2323.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5454919741,"x_train":1322.498714333,"imp":27.9827612422,"elo":-972.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5348496069,"x_train":1.4849621492,"imp":18.4784935089,"elo":9576.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8184836382,"x_train":134.594931747,"imp":7.3854101119,"elo":26990.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3079343419,"x_train":134.594931747,"imp":4.4976437918,"elo":31548.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2367423058,"x_train":9.8924281166,"imp":28.0714246531,"elo":-1604.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"x_train":41.2100429512,"imp":27.5967758404,"elo":-331.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.5244814723,"x_train":4391.0141404006,"imp":27.4759618173,"elo":342.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":275.9499442925,"x_train":4391.0141404006,"imp":23.6244714479,"elo":4593.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.277558181,"x_train":0.8171638129,"imp":14.4214382191,"elo":20252.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":17.7035952873,"x_train":35.6508208249,"imp":13.5232072922,"elo":24743.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0232842395,"x_train":0.1087394813,"imp":59.7189648865,"elo":-22944.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.056543605,"x_train":2.7860859796,"imp":19.6416425074,"elo":8935.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426377996,"x_train":0.4823068139,"imp":14.2045636484,"elo":23243.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0740791906,"x_train":4.047894089,"imp":34.7546595439,"elo":-5112.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0574085901,"x_train":1142.3492528959,"imp":24.4579757413,"elo":2811.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5419846432,"x_train":1142.3492528959,"imp":24.4328219572,"elo":3320.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5160116403,"x_train":1.8403475551,"imp":9.3847277445,"elo":25643.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2726374657,"x_train":0.92595243,"imp":4.5260970421,"elo":29818.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4795389483,"x_train":65.8419489083,"imp":46.4830894898,"elo":-11648.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.468865642,"x_train":3349.2183478585,"imp":36.9007036438,"elo":-6530.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.0373758651,"x_train":3349.2183478585,"imp":36.9583859241,"elo":-7254.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1994643012,"x_train":0.578742138,"imp":23.4592002877,"elo":5283.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0276328914,"x_train":3.3435344852,"imp":34.8795471811,"elo":-5810.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0266472235,"x_train":2341.9377908803,"imp":24.1915976671,"elo":4010.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1720376576,"x_train":2341.9377908803,"imp":24.5765776663,"elo":1461.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"x_train":0.321861902,"imp":15.6926314585,"elo":15134.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"x_train":456.3324763471,"imp":15.692631292,"elo":15858.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.1704905161,"x_train":456.3324763471,"imp":15.4240395653,"elo":16562.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"x_train":52.4793328731,"imp":20.6424796077,"elo":7142.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"x_train":1229.419892487,"imp":20.6424750924,"elo":7603.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":21.7408394399,"x_train":1229.419892487,"imp":16.3672216796,"elo":14414.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9209350283,"x_train":3.1442091702,"imp":59.6685107498,"elo":-21214.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.9740680655,"x_train":2009.1393249946,"imp":33.9505172259,"elo":-4441.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.7671484184,"x_train":2009.1393249946,"imp":32.399105052,"elo":-3755.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
index 6e0beec09485a53526550b608ea9ff2e250d9965..940d6a232d94edeeea4861ed601429f90dd4d9d6 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":66.0588444215,"imp":37.1098221861,"elo":-1672.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.5923477438,"imp":17.3547552173,"elo":4892.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":9.9824633777,"imp":16.8180362511,"elo":5541.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":28.1814882563,"imp":16.1030488468,"elo":6000.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":56.6972466524,"imp":15.6955147734,"elo":6277.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":159.5146818194,"imp":15.3829095761,"elo":6908.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":313.167560406,"imp":15.2629283415,"elo":7042.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":620.4583392781,"imp":15.1258558949,"elo":7277.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":930.7284389971,"imp":15.0630430176,"elo":7351.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1261.2440700178,"imp":15.0249864754,"elo":7483.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.2281173573,"imp":22.3048595539,"elo":2638.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.5627936741,"imp":20.9230935328,"elo":2865.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.4596619621,"imp":19.1389134322,"elo":3972.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.0218870561,"imp":18.1056432687,"elo":4360.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":7.9773474182,"imp":17.5870646377,"elo":4722.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":16.3876808766,"imp":17.2942989824,"elo":5011.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":33.0525808274,"imp":17.0182082807,"elo":5272.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":49.4935904635,"imp":16.9232947793,"elo":5475.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":66.2292774111,"imp":16.9578731367,"elo":5374.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":13.8733921956,"imp":17.9078747707,"elo":4664.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":46.1464646082,"imp":16.7341770342,"elo":5606.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":147.52300074,"imp":15.3806792968,"elo":6971.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":306.6782807185,"imp":15.1452722438,"elo":7199.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":801.6286487563,"imp":14.6415906517,"elo":7770.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":1627.1101123839,"imp":14.430991369,"elo":8296.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":3264.2024061238,"imp":14.2876314294,"elo":8786.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":4901.8335790951,"imp":14.2479220842,"elo":8911.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":6601.4555400381,"imp":14.2080565817,"elo":9473.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.2342836282,"imp":52.4066418594,"elo":-3602.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3800410039,"imp":45.1518614912,"elo":-2648.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.7928114455,"imp":30.662456218,"elo":-235.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.4929370392,"imp":24.7164640518,"elo":1785.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.6241622692,"imp":20.9408156964,"elo":2804.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.1550594653,"imp":20.5217514993,"elo":3200.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.2035240625,"imp":20.3845714473,"elo":3271.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":21.2866256405,"imp":20.325403259,"elo":3343.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":28.4407435042,"imp":20.2670230505,"elo":3488.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.7277827402,"imp":54.4503173153,"elo":-3875.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.2812955867,"imp":54.5336434441,"elo":-4042.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":13.7046141456,"imp":52.5784645996,"elo":-3708.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":28.1489948252,"imp":50.2197571748,"elo":-3358.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":65.9177907885,"imp":49.0197664142,"elo":-3290.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":133.9142240391,"imp":47.9404231189,"elo":-3216.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":275.0443968187,"imp":47.0096623081,"elo":-3137.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":412.7471335694,"imp":46.1569676243,"elo":-2846.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":553.4880496571,"imp":45.5504205965,"elo":-2718.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.2060542703,"imp":29.7602973693,"elo":-106.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":1.2095397384,"imp":29.5948617941,"elo":-41.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":3.4430801699,"imp":26.8034215743,"elo":1316.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":7.3054081422,"imp":24.0334315785,"elo":2224.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":19.34969751,"imp":17.9130900921,"elo":4546.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":38.6369993724,"imp":14.9432102176,"elo":7545.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":77.1677596734,"imp":14.4967977362,"elo":7873.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":116.0761985074,"imp":14.4424922558,"elo":8015.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":157.4028552082,"imp":14.352900556,"elo":8585.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":3.9378111103,"imp":0.0,"elo":12315.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.534943689,"imp":24.5238492021,"elo":1931.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":37.933698739,"imp":23.3114694228,"elo":2537.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":106.8025033221,"imp":21.8841094817,"elo":2690.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":212.1550450122,"imp":20.8257985661,"elo":2928.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":514.1027685745,"imp":19.4632400006,"elo":3906.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1078.615636086,"imp":18.6043461802,"elo":4038.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2143.6938229576,"imp":17.9416534339,"elo":4423.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3217.2348435455,"imp":17.5630829294,"elo":4779.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4378.8792899745,"imp":17.294096846,"elo":5138.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":16.7832602994,"imp":39.7514072301,"elo":-2146.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":2.7257969394,"imp":59.5308044317,"elo":-5961.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":8.7745991633,"imp":54.8868562582,"elo":-4137.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":25.2036626358,"imp":51.9237268152,"elo":-3543.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":30.0706477892,"imp":51.6015540581,"elo":-3452.3,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1776920981,"imp":27.5132675094,"elo":860.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4652499352,"imp":19.8025665932,"elo":3632.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.4763660732,"imp":16.6141857743,"elo":5670.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":3.0369890673,"imp":15.5562712818,"elo":6415.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":7.4865228781,"imp":14.8199538034,"elo":7665.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":15.1232073981,"imp":14.5968458625,"elo":7819.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":30.3263660899,"imp":14.3775932239,"elo":8470.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":45.7813804379,"imp":14.3190633811,"elo":8618.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":61.4738106436,"imp":14.2909034341,"elo":8700.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.5101464138,"imp":41.5812680827,"elo":-2421.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":12.3980382971,"imp":37.7980112911,"elo":-1938.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":30.8974860361,"imp":35.1790834708,"elo":-1346.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":62.2345356879,"imp":33.228621422,"elo":-717.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":164.4196952291,"imp":31.5826250175,"elo":-424.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":334.0371204561,"imp":30.8459045697,"elo":-299.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":660.4294593011,"imp":29.0387215642,"elo":89.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":990.5204223442,"imp":28.5480532422,"elo":290.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1322.498714333,"imp":27.9960689724,"elo":501.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.4849621492,"imp":18.4935028204,"elo":4135.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.9184440201,"imp":18.2175225288,"elo":4296.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":3.2451422052,"imp":17.4641145366,"elo":4835.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.3339567122,"imp":17.0183735883,"elo":5204.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":13.7229026201,"imp":14.4883605878,"elo":7938.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":29.7246860705,"imp":9.4741799614,"elo":10018.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":63.7718881675,"imp":6.6167192702,"elo":10589.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":96.2590274265,"imp":5.0880354109,"elo":10839.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":134.594931747,"imp":4.5152271839,"elo":11356.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":9.8924281166,"imp":28.0846677645,"elo":394.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.2100429512,"imp":27.6101063417,"elo":681.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":62.7744579205,"imp":27.3298114984,"elo":1133.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":116.8910871561,"imp":27.1666199817,"elo":1197.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":207.4972060022,"imp":26.8735702028,"elo":1258.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":532.9957633827,"imp":25.7558200876,"elo":1524.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1085.275659615,"imp":25.1480444001,"elo":1625.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2177.7649560761,"imp":24.4892964104,"elo":2011.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3284.6615266756,"imp":23.8383971903,"elo":2276.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4391.0141404006,"imp":23.6319338123,"elo":2383.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.8171638129,"imp":14.4371944943,"elo":8151.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":35.6508208249,"imp":13.539128945,"elo":9789.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1087394813,"imp":59.7263812191,"elo":-6594.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":2.7860859796,"imp":19.656437666,"elo":3737.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.4823068139,"imp":14.2203598534,"elo":9145.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":4.047894089,"imp":34.7666718876,"elo":-1012.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.1210227808,"imp":32.3780565133,"elo":-485.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":26.2329613573,"imp":29.1498676397,"elo":23.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":55.2138352308,"imp":27.4167126626,"elo":1067.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":141.9296718443,"imp":26.5043661991,"elo":1371.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":284.591401136,"imp":25.7598168207,"elo":1474.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":562.5419157372,"imp":25.0307019084,"elo":1678.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":849.3471745634,"imp":24.7251939402,"elo":1732.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1142.3492528959,"imp":24.446734989,"elo":2130.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":1.8403475551,"imp":9.4014113524,"elo":10168.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":0.92595243,"imp":4.5436751956,"elo":11015.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":65.8419489083,"imp":46.4929426215,"elo":-2981.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":82.3201598909,"imp":46.2138836432,"elo":-2909.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":128.1588279475,"imp":45.555112229,"elo":-2791.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":190.5221223243,"imp":44.0271781614,"elo":-2544.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":426.866827693,"imp":41.0974133818,"elo":-2285.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":860.4651862449,"imp":38.3656852859,"elo":-2007.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1704.5805078339,"imp":37.7565354561,"elo":-1870.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2515.3432482994,"imp":37.1313510511,"elo":-1803.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3349.2183478585,"imp":36.9699928149,"elo":-1576.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.578742138,"imp":23.4732834803,"elo":2487.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":3.3435344852,"imp":34.8915367475,"elo":-1211.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.743652342,"imp":34.3728415445,"elo":-950.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":42.5186103306,"imp":33.259261365,"elo":-774.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":107.3316064659,"imp":31.0785509014,"elo":-362.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":274.494576287,"imp":28.8315951234,"elo":155.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":577.8040181111,"imp":27.5575080172,"elo":789.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1167.0902850561,"imp":26.2831539442,"elo":1423.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1768.1386269443,"imp":25.5800576293,"elo":1574.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2341.9377908803,"imp":24.5904642306,"elo":1837.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.321861902,"imp":15.7081535219,"elo":6172.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":3.1340720026,"imp":15.6081723436,"elo":6346.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":10.4589579493,"imp":15.4893036937,"elo":6545.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":19.8143120293,"imp":15.4902145867,"elo":6482.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":56.2043887857,"imp":15.4747969248,"elo":6659.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":113.5858543435,"imp":15.4774957603,"elo":6605.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":226.5076734886,"imp":15.4703397968,"elo":6709.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":337.6096231295,"imp":15.4501933332,"elo":6756.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":456.3324763471,"imp":15.4377171274,"elo":6852.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":52.4793328731,"imp":20.657085983,"elo":3096.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":96.7397158429,"imp":19.9767460723,"elo":3560.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":241.2633499833,"imp":19.4822854162,"elo":3839.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":479.2228112407,"imp":18.4883332163,"elo":4232.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1184.4533790756,"imp":16.4964918995,"elo":5799.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1229.419892487,"imp":16.3826197074,"elo":5932.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.1442091702,"imp":59.6759371,"elo":-6395.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":13.155466857,"imp":53.945057456,"elo":-3757.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":45.477404513,"imp":44.9123114388,"elo":-2598.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":96.1451044259,"imp":38.4628370628,"elo":-2076.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":249.6329672147,"imp":35.9218683621,"elo":-1413.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":497.3958095536,"imp":34.7739561298,"elo":-1143.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":999.5916054172,"imp":33.8284439456,"elo":-831.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1494.1081451946,"imp":33.031286653,"elo":-660.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2009.1393249946,"imp":32.4115806664,"elo":-603.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7827549071,"x_train":66.0588444215,"imp":37.1098221861,"elo":-1672.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0218721746,"x_train":3.5923477438,"imp":17.3547552173,"elo":4892.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0497134013,"x_train":9.9824633777,"imp":16.8180362511,"elo":5541.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1174311067,"x_train":28.1814882563,"imp":16.1030488468,"elo":6000.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1428266464,"x_train":56.6972466524,"imp":15.6955147734,"elo":6277.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2357748832,"x_train":159.5146818194,"imp":15.3829095761,"elo":6908.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2695920408,"x_train":313.167560406,"imp":15.2629283415,"elo":7042.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3044187588,"x_train":620.4583392781,"imp":15.1258558949,"elo":7277.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2568636437,"x_train":930.7284389971,"imp":15.0630430176,"elo":7351.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.280383328,"x_train":1261.2440700178,"imp":15.0249864754,"elo":7483.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0128154423,"x_train":0.2281173573,"imp":22.3048595539,"elo":2638.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0292978626,"x_train":0.5627936741,"imp":20.9230935328,"elo":2865.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0535690992,"x_train":1.4596619621,"imp":19.1389134322,"elo":3972.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0727289501,"x_train":3.0218870561,"imp":18.1056432687,"elo":4360.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1288669352,"x_train":7.9773474182,"imp":17.5870646377,"elo":4722.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.154154967,"x_train":16.3876808766,"imp":17.2942989824,"elo":5011.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1757045439,"x_train":33.0525808274,"imp":17.0182082807,"elo":5272.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1727469812,"x_train":49.4935904635,"imp":16.9232947793,"elo":5475.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1959219282,"x_train":66.2292774111,"imp":16.9578731367,"elo":5374.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.014064448,"x_train":13.8733921956,"imp":17.9078747707,"elo":4664.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0311340555,"x_train":46.1464646082,"imp":16.7341770342,"elo":5606.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0551923837,"x_train":147.52300074,"imp":15.3806792968,"elo":6971.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0688712999,"x_train":306.6782807185,"imp":15.1452722438,"elo":7199.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0984842812,"x_train":801.6286487563,"imp":14.6415906517,"elo":7770.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1253388792,"x_train":1627.1101123839,"imp":14.430991369,"elo":8296.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1457912936,"x_train":3264.2024061238,"imp":14.2876314294,"elo":8786.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1332857555,"x_train":4901.8335790951,"imp":14.2479220842,"elo":8911.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1069197134,"x_train":6601.4555400381,"imp":14.2080565817,"elo":9473.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0418367287,"x_train":0.2342836282,"imp":52.4066418594,"elo":-3602.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0444077791,"x_train":0.3800410039,"imp":45.1518614912,"elo":-2648.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0356755189,"x_train":0.7928114455,"imp":30.662456218,"elo":-235.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0273270001,"x_train":1.4929370392,"imp":24.7164640518,"elo":1785.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0395302294,"x_train":3.6241622692,"imp":20.9408156964,"elo":2804.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0642314309,"x_train":7.1550594653,"imp":20.5217514993,"elo":3200.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1059977241,"x_train":14.2035240625,"imp":20.3845714473,"elo":3271.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1375967472,"x_train":21.2866256405,"imp":20.325403259,"elo":3343.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739987392,"x_train":28.4407435042,"imp":20.2670230505,"elo":3488.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0832738002,"x_train":2.7277827402,"imp":54.4503173153,"elo":-3875.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1651997039,"x_train":5.2812955867,"imp":54.5336434441,"elo":-4042.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3882144021,"x_train":13.7046141456,"imp":52.5784645996,"elo":-3708.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6283891459,"x_train":28.1489948252,"imp":50.2197571748,"elo":-3358.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9164506571,"x_train":65.9177907885,"imp":49.0197664142,"elo":-3290.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2468060948,"x_train":133.9142240391,"imp":47.9404231189,"elo":-3216.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7812914447,"x_train":275.0443968187,"imp":47.0096623081,"elo":-3137.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7948173827,"x_train":412.7471335694,"imp":46.1569676243,"elo":-2846.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"x_train":553.4880496571,"imp":45.5504205965,"elo":-2718.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0474691168,"x_train":0.2060542703,"imp":29.7602973693,"elo":-106.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7435906933,"x_train":1.2095397384,"imp":29.5948617941,"elo":-41.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8375843537,"x_train":3.4430801699,"imp":26.8034215743,"elo":1316.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6051740249,"x_train":7.3054081422,"imp":24.0334315785,"elo":2224.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3230917617,"x_train":19.34969751,"imp":17.9130900921,"elo":4546.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4649015268,"x_train":38.6369993724,"imp":14.9432102176,"elo":7545.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2768939316,"x_train":77.1677596734,"imp":14.4967977362,"elo":7873.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.0310390923,"x_train":116.0761985074,"imp":14.4424922558,"elo":8015.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.5785391893,"x_train":157.4028552082,"imp":14.352900556,"elo":8585.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":3.1918999344,"x_train":3.9378111103,"imp":0.0,"elo":12315.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7472229818,"x_train":14.534943689,"imp":24.5238492021,"elo":1931.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.6445542539,"x_train":37.933698739,"imp":23.3114694228,"elo":2537.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.9535369347,"x_train":106.8025033221,"imp":21.8841094817,"elo":2690.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":11.2275665264,"x_train":212.1550450122,"imp":20.8257985661,"elo":2928.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":16.3103609399,"x_train":514.1027685745,"imp":19.4632400006,"elo":3906.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":19.9788525201,"x_train":1078.615636086,"imp":18.6043461802,"elo":4038.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":23.1702683559,"x_train":2143.6938229576,"imp":17.9416534339,"elo":4423.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":25.2758129387,"x_train":3217.2348435455,"imp":17.5630829294,"elo":4779.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":25.5948192523,"x_train":4378.8792899745,"imp":17.294096846,"elo":5138.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7641563624,"x_train":16.7832602994,"imp":39.7514072301,"elo":-2146.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"x_train":2.7257969394,"imp":59.5308044317,"elo":-5961.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1522131379,"x_train":8.7745991633,"imp":54.8868562582,"elo":-4137.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4737042685,"x_train":25.2036626358,"imp":51.9237268152,"elo":-3543.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5623533631,"x_train":30.0706477892,"imp":51.6015540581,"elo":-3452.3,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0069604217,"x_train":0.1776920981,"imp":27.5132675094,"elo":860.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0213595479,"x_train":0.4652499352,"imp":19.8025665932,"elo":3632.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0392463929,"x_train":1.4763660732,"imp":16.6141857743,"elo":5670.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0501367115,"x_train":3.0369890673,"imp":15.5562712818,"elo":6415.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0768595462,"x_train":7.4865228781,"imp":14.8199538034,"elo":7665.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.09727726,"x_train":15.1232073981,"imp":14.5968458625,"elo":7819.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1236267341,"x_train":30.3263660899,"imp":14.3775932239,"elo":8470.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1434609645,"x_train":45.7813804379,"imp":14.3190633811,"elo":8618.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1538951715,"x_train":61.4738106436,"imp":14.2909034341,"elo":8700.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.1615432661,"x_train":6.5101464138,"imp":41.5812680827,"elo":-2421.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2453378418,"x_train":12.3980382971,"imp":37.7980112911,"elo":-1938.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3029056071,"x_train":30.8974860361,"imp":35.1790834708,"elo":-1346.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.347018823,"x_train":62.2345356879,"imp":33.228621422,"elo":-717.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4558697672,"x_train":164.4196952291,"imp":31.5826250175,"elo":-424.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5786006034,"x_train":334.0371204561,"imp":30.8459045697,"elo":-299.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.471412833,"x_train":660.4294593011,"imp":29.0387215642,"elo":89.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4979071245,"x_train":990.5204223442,"imp":28.5480532422,"elo":290.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5454919741,"x_train":1322.498714333,"imp":27.9960689724,"elo":501.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5348496069,"x_train":1.4849621492,"imp":18.4935028204,"elo":4135.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6897445483,"x_train":1.9184440201,"imp":18.2175225288,"elo":4296.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0053886682,"x_train":3.2451422052,"imp":17.4641145366,"elo":4835.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7446747876,"x_train":5.3339567122,"imp":17.0183735883,"elo":5204.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.1304312666,"x_train":13.7229026201,"imp":14.4883605878,"elo":7938.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.5017657204,"x_train":29.7246860705,"imp":9.4741799614,"elo":10018.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6074456326,"x_train":63.7718881675,"imp":6.6167192702,"elo":10589.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6520668764,"x_train":96.2590274265,"imp":5.0880354109,"elo":10839.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3079343419,"x_train":134.594931747,"imp":4.5152271839,"elo":11356.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2367423058,"x_train":9.8924281166,"imp":28.0846677645,"elo":394.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"x_train":41.2100429512,"imp":27.6101063417,"elo":681.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":50.8141493102,"x_train":62.7744579205,"imp":27.3298114984,"elo":1133.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":77.2065039497,"x_train":116.8910871561,"imp":27.1666199817,"elo":1197.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":109.5641245161,"x_train":207.4972060022,"imp":26.8735702028,"elo":1258.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":163.2920265134,"x_train":532.9957633827,"imp":25.7558200876,"elo":1524.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":213.967010973,"x_train":1085.275659615,"imp":25.1480444001,"elo":1625.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":269.7422775799,"x_train":2177.7649560761,"imp":24.4892964104,"elo":2011.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":263.3543429854,"x_train":3284.6615266756,"imp":23.8383971903,"elo":2276.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":275.9499442925,"x_train":4391.0141404006,"imp":23.6319338123,"elo":2383.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.277558181,"x_train":0.8171638129,"imp":14.4371944943,"elo":8151.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":17.7035952873,"x_train":35.6508208249,"imp":13.539128945,"elo":9789.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0232842395,"x_train":0.1087394813,"imp":59.7263812191,"elo":-6594.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.056543605,"x_train":2.7860859796,"imp":19.656437666,"elo":3737.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426377996,"x_train":0.4823068139,"imp":14.2203598534,"elo":9145.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0740791906,"x_train":4.047894089,"imp":34.7666718876,"elo":-1012.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.127108128,"x_train":9.1210227808,"imp":32.3780565133,"elo":-485.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2047906233,"x_train":26.2329613573,"imp":29.1498676397,"elo":23.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2449575585,"x_train":55.2138352308,"imp":27.4167126626,"elo":1067.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3319391408,"x_train":141.9296718443,"imp":26.5043661991,"elo":1371.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4159006841,"x_train":284.591401136,"imp":25.7598168207,"elo":1474.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4198951784,"x_train":562.5419157372,"imp":25.0307019084,"elo":1678.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.4821521088,"x_train":849.3471745634,"imp":24.7251939402,"elo":1732.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5419846432,"x_train":1142.3492528959,"imp":24.446734989,"elo":2130.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5160116403,"x_train":1.8403475551,"imp":9.4014113524,"elo":10168.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2726374657,"x_train":0.92595243,"imp":4.5436751956,"elo":11015.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4795389483,"x_train":65.8419489083,"imp":46.4929426215,"elo":-2981.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6272217169,"x_train":82.3201598909,"imp":46.2138836432,"elo":-2909.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0494971432,"x_train":128.1588279475,"imp":45.555112229,"elo":-2791.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.8279085887,"x_train":190.5221223243,"imp":44.0271781614,"elo":-2544.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.9995876557,"x_train":426.866827693,"imp":41.0974133818,"elo":-2285.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6380882465,"x_train":860.4651862449,"imp":38.3656852859,"elo":-2007.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.4734687528,"x_train":1704.5805078339,"imp":37.7565354561,"elo":-1870.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.7675349663,"x_train":2515.3432482994,"imp":37.1313510511,"elo":-1803.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.0373758651,"x_train":3349.2183478585,"imp":36.9699928149,"elo":-1576.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1994643012,"x_train":0.578742138,"imp":23.4732834803,"elo":2487.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0276328914,"x_train":3.3435344852,"imp":34.8915367475,"elo":-1211.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0620535399,"x_train":13.743652342,"imp":34.3728415445,"elo":-950.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1179715238,"x_train":42.5186103306,"imp":33.259261365,"elo":-774.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1651259896,"x_train":107.3316064659,"imp":31.0785509014,"elo":-362.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2389435798,"x_train":274.494576287,"imp":28.8315951234,"elo":155.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2817484073,"x_train":577.8040181111,"imp":27.5575080172,"elo":789.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2701272436,"x_train":1167.0902850561,"imp":26.2831539442,"elo":1423.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.244273469,"x_train":1768.1386269443,"imp":25.5800576293,"elo":1574.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1720376576,"x_train":2341.9377908803,"imp":24.5904642306,"elo":1837.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"x_train":0.321861902,"imp":15.7081535219,"elo":6172.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3465841915,"x_train":3.1340720026,"imp":15.6081723436,"elo":6346.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6102341201,"x_train":10.4589579493,"imp":15.4893036937,"elo":6545.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5888194515,"x_train":19.8143120293,"imp":15.4902145867,"elo":6482.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6938849221,"x_train":56.2043887857,"imp":15.4747969248,"elo":6659.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6851908948,"x_train":113.5858543435,"imp":15.4774957603,"elo":6605.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.74544132,"x_train":226.5076734886,"imp":15.4703397968,"elo":6709.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9582567497,"x_train":337.6096231295,"imp":15.4501933332,"elo":6756.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1704905161,"x_train":456.3324763471,"imp":15.4377171274,"elo":6852.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"x_train":52.4793328731,"imp":20.657085983,"elo":3096.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":11.3040848505,"x_train":96.7397158429,"imp":19.9767460723,"elo":3560.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":18.4470099318,"x_train":241.2633499833,"imp":19.4822854162,"elo":3839.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":20.7391590001,"x_train":479.2228112407,"imp":18.4883332163,"elo":4232.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":23.6267433633,"x_train":1184.4533790756,"imp":16.4964918995,"elo":5799.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":21.7408394399,"x_train":1229.419892487,"imp":16.3826197074,"elo":5932.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9209350283,"x_train":3.1442091702,"imp":59.6759371,"elo":-6395.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.8238649007,"x_train":13.155466857,"imp":53.945057456,"elo":-3757.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.7289794426,"x_train":45.477404513,"imp":44.9123114388,"elo":-2598.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":3.6121868214,"x_train":96.1451044259,"imp":38.4628370628,"elo":-2076.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":4.9673931163,"x_train":249.6329672147,"imp":35.9218683621,"elo":-1413.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":5.8186016306,"x_train":497.3958095536,"imp":34.7739561298,"elo":-1143.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":6.1787775661,"x_train":999.5916054172,"imp":33.8284439456,"elo":-831.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":5.262793854,"x_train":1494.1081451946,"imp":33.031286653,"elo":-660.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.7671484184,"x_train":2009.1393249946,"imp":32.4115806664,"elo":-603.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..ef5a272586be0f36571afb61c5e5104aa0979cda
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_multiclass/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["LimiX (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-3 (default)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (tuned)", "TabPFN-2.6 (default)", "TabFM (default)", "EBM (tuned + ensembled)", "TabICLv2 (default)", "RandomForest (tuned + ensembled)", "LightGBM (tuned + ensembled)", "LightGBM (tuned)", "TabDPT-Turbo (default)", "RandomForest (tuned)", "EBM (tuned)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "XGBoost (tuned + ensembled)", "XGBoost (tuned)", "XGBoost (default)", "iLTM (tuned + ensembled)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "CatBoost (default)", "EBM (default)", "ModernNCA (tuned)", "RealTabPFN-2.5 (default)", "TabICL (default)", "ExtraTrees (tuned + ensembled)", "iLTM (tuned)", "iLTM (default)", "ExtraTrees (tuned)", "ChimeraBoost (default)", "TabSwift (default)", "TabDPT (tuned + ensembled)", "TorchMLP (tuned)", "TabM (tuned + ensembled)", "TabM (tuned)", "ModernNCA (default)", "TorchMLP (tuned + ensembled)", "RandomForest (default)", "TabDPT (tuned)", "TabDPT (default)", "RealMLP (tuned + ensembled)", "SAP-RPT-OSS (default)", "RealMLP (tuned)", "LightGBM (default)", "xRFM (tuned + ensembled)", "xRFM (tuned)", "TabM (default)", "TorchMLP (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "BetaTabPFN (default)", "OrionMSP (default)", "RealMLP (default)", "FastaiMLP (tuned)", "FastaiMLP (tuned + ensembled)", "TabSTAR (default)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "xRFM (default)", "TabFlex (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":1.0},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9863013699},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9726027397},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.9589041096},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9452054795},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.9315068493},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.9178082192},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.904109589},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8904109589},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8767123288},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8630136986},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8493150685},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.8356164384},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.8219178082},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.8082191781},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.7945205479},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7808219178},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7671232877},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7534246575},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.7397260274},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.7260273973},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7123287671},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.698630137},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6849315068},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6712328767},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6575342466},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.6438356164},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6301369863},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6164383562},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.602739726},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5890410959},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5753424658},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.5616438356},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.5479452055},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5342465753},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5205479452},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5068493151},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4931506849},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4794520548},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.4657534247},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4520547945},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4383561644},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.4246575342},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4109589041},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.397260274},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3835616438},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3698630137},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.3561643836},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3424657534},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.3287671233},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.3150684932},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.301369863},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2876712329},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2739726027},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2602739726},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.2465753425},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2328767123},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.2191780822},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2054794521},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.1917808219},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1780821918},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1643835616},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1506849315},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1369863014},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1232876712},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1095890411},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.095890411},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0821917808},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.0684931507},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0547945205},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.0410958904},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.0273972603},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.0136986301},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..034bcd327fe452b3b8dab7198d23566c6f8885fc
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,LimiX (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),TabPFN-3 (default),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (tuned),TabPFN-2.6 (default),TabFM (default),EBM (tuned + ensembled),TabICLv2 (default),RandomForest (tuned + ensembled),LightGBM (tuned + ensembled),LightGBM (tuned),TabDPT-Turbo (default),RandomForest (tuned),EBM (tuned),CatBoost (tuned + ensembled),CatBoost (tuned),XGBoost (tuned + ensembled),XGBoost (tuned),XGBoost (default),iLTM (tuned + ensembled),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),ModernNCA (tuned + ensembled),CatBoost (default),EBM (default),ModernNCA (tuned),RealTabPFN-2.5 (default),TabICL (default),ExtraTrees (tuned + ensembled),iLTM (tuned),iLTM (default),ExtraTrees (tuned),ChimeraBoost (default),TabSwift (default),TabDPT (tuned + ensembled),TorchMLP (tuned),TabM (tuned + ensembled),TabM (tuned),ModernNCA (default),TorchMLP (tuned + ensembled),RandomForest (default),TabDPT (tuned),TabDPT (default),RealMLP (tuned + ensembled),SAP-RPT-OSS (default),RealMLP (tuned),LightGBM (default),xRFM (tuned + ensembled),xRFM (tuned),TabM (default),TorchMLP (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),BetaTabPFN (default),OrionMSP (default),RealMLP (default),FastaiMLP (tuned),FastaiMLP (tuned + ensembled),TabSTAR (default),PerpetualBooster (tuned + ensembled),ExtraTrees (default),PerpetualBooster (tuned),FastaiMLP (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),KNN (tuned + ensembled),PerpetualBooster (default),xRFM (default),TabFlex (default),KNN (tuned),KNN (default)
+LimiX (default),,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICL (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
index 5857f47817ffb84c8894eae032e8d361de62c46b..ee9973c9f2241d28bd5fc4cf998bb3dc082da28f 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Multiclass | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..a40330d90fcac1edad32a08a8314ab832f588410
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_multiclass/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1742.0,"score":0.668,"rank":8.29,"hrank":5.03,"imp":12.628,"train_time":247.2,"predict_time":3.959,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+250\/-62","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1738.0,"score":0.678,"rank":8.43,"hrank":3.09,"imp":5.637,"train_time":2837.96,"predict_time":9.546,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+466\/-162","imputed":false},{"position":2,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1711.0,"score":0.834,"rank":9.43,"hrank":1.31,"imp":6.112,"train_time":52.66,"predict_time":6.434,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1322\/-286","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1660.0,"score":0.591,"rank":11.5,"hrank":8.3,"imp":14.338,"train_time":2837.96,"predict_time":1.132,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+249\/-123","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1656.0,"score":0.581,"rank":11.71,"hrank":8.07,"imp":13.688,"train_time":10.78,"predict_time":0.65,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+258\/-111","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1595.0,"score":0.486,"rank":14.64,"hrank":8.17,"imp":13.213,"train_time":10.97,"predict_time":0.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+323\/-125","imputed":false},{"position":6,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1591.0,"score":0.515,"rank":14.86,"hrank":6.11,"imp":12.315,"train_time":9.45,"predict_time":0.514,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+320\/-149","imputed":false},{"position":7,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1538.0,"score":0.433,"rank":17.86,"hrank":8.41,"imp":17.503,"train_time":4655.15,"predict_time":21.781,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+218\/-98","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1529.0,"score":0.508,"rank":18.43,"hrank":6.04,"imp":15.462,"train_time":5.36,"predict_time":0.234,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+411\/-184","imputed":false},{"position":9,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1497.0,"score":0.382,"rank":20.43,"hrank":13.37,"imp":15.294,"train_time":46.11,"predict_time":9.187,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+248\/-156","imputed":false},{"position":10,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1497.0,"score":0.404,"rank":20.43,"hrank":10.45,"imp":19.331,"train_time":6318.92,"predict_time":3.195,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+228\/-118","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1494.0,"score":0.385,"rank":20.57,"hrank":16.25,"imp":17.942,"train_time":4655.15,"predict_time":4.029,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+185\/-120","imputed":false},{"position":12,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1488.0,"score":0.338,"rank":21.0,"hrank":15.94,"imp":16.994,"train_time":3632.1,"predict_time":1.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+207\/-112","imputed":false},{"position":13,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1488.0,"score":0.422,"rank":21.0,"hrank":13.86,"imp":19.55,"train_time":3829.4,"predict_time":4.64,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+166\/-122","imputed":false},{"position":14,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1488.0,"score":0.35,"rank":21.0,"hrank":12.84,"imp":16.07,"train_time":3632.1,"predict_time":0.238,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+221\/-112","imputed":false},{"position":15,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1479.0,"score":0.404,"rank":21.57,"hrank":10.43,"imp":19.605,"train_time":3829.4,"predict_time":0.879,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+168\/-126","imputed":false},{"position":16,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1450.0,"score":0.299,"rank":23.43,"hrank":17.94,"imp":20.452,"train_time":1394.84,"predict_time":3.577,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-89","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1446.0,"score":0.291,"rank":23.71,"hrank":17.48,"imp":20.536,"train_time":1394.84,"predict_time":1.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+201\/-87","imputed":false},{"position":18,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1444.0,"score":0.355,"rank":23.86,"hrank":9.04,"imp":20.925,"train_time":2149.3,"predict_time":2.926,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+174\/-190","imputed":false},{"position":19,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1438.0,"score":0.407,"rank":24.29,"hrank":16.88,"imp":22.062,"train_time":10182.26,"predict_time":399.334,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+220\/-196","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1396.0,"score":0.211,"rank":27.29,"hrank":25.7,"imp":20.963,"train_time":2342.58,"predict_time":5.129,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-59","imputed":false},{"position":21,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1392.0,"score":0.274,"rank":27.57,"hrank":15.06,"imp":23.603,"train_time":5.65,"predict_time":0.269,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+215\/-169","imputed":false},{"position":22,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1390.0,"score":0.219,"rank":27.71,"hrank":25.61,"imp":20.682,"train_time":2342.58,"predict_time":1.065,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-75","imputed":false},{"position":23,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1378.0,"score":0.302,"rank":28.57,"hrank":15.15,"imp":23.74,"train_time":17.34,"predict_time":0.728,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+161\/-149","imputed":false},{"position":24,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1369.0,"score":0.333,"rank":29.29,"hrank":12.54,"imp":22.982,"train_time":10182.26,"predict_time":40.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+251\/-200","imputed":false},{"position":25,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1361.0,"score":0.198,"rank":29.86,"hrank":25.52,"imp":21.288,"train_time":34015.19,"predict_time":986.352,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-94","imputed":false},{"position":26,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1360.0,"score":0.227,"rank":30.0,"hrank":22.78,"imp":23.298,"train_time":326.32,"predict_time":4.279,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+154\/-133","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1357.0,"score":0.211,"rank":30.21,"hrank":25.69,"imp":21.853,"train_time":6995.99,"predict_time":9.551,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-119","imputed":false},{"position":28,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1354.0,"score":0.374,"rank":30.43,"hrank":8.07,"imp":22.911,"train_time":22.26,"predict_time":1.556,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+309\/-251","imputed":false},{"position":29,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1352.0,"score":0.276,"rank":30.57,"hrank":5.73,"imp":22.387,"train_time":2149.3,"predict_time":0.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+205\/-192","imputed":false},{"position":30,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1347.0,"score":0.187,"rank":30.93,"hrank":24.94,"imp":22.153,"train_time":6995.99,"predict_time":0.583,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+160\/-133","imputed":false},{"position":31,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1336.0,"score":0.255,"rank":31.79,"hrank":14.17,"imp":24.129,"train_time":1042.39,"predict_time":1.873,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+200\/-215","imputed":false},{"position":32,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1316.0,"score":0.157,"rank":33.29,"hrank":25.34,"imp":24.6,"train_time":27727.7,"predict_time":0.31,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-116","imputed":false},{"position":33,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1306.0,"score":0.146,"rank":34.0,"hrank":30.68,"imp":23.042,"train_time":29.58,"predict_time":0.343,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+111\/-92","imputed":false},{"position":34,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1290.0,"score":0.159,"rank":35.14,"hrank":31.07,"imp":24.794,"train_time":3834.57,"predict_time":2.294,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-140","imputed":false},{"position":35,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1288.0,"score":0.176,"rank":35.29,"hrank":29.6,"imp":22.578,"train_time":34015.19,"predict_time":240.683,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-143","imputed":false},{"position":36,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1273.0,"score":0.237,"rank":36.43,"hrank":24.12,"imp":24.746,"train_time":731.23,"predict_time":2.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+198\/-247","imputed":false},{"position":37,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1263.0,"score":0.179,"rank":37.14,"hrank":23.21,"imp":26.279,"train_time":2750.22,"predict_time":13.569,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+279\/-224","imputed":false},{"position":38,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1256.0,"score":0.176,"rank":37.64,"hrank":15.94,"imp":27.277,"train_time":1042.39,"predict_time":0.294,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+222\/-210","imputed":false},{"position":39,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1254.0,"score":0.154,"rank":37.86,"hrank":27.97,"imp":25.601,"train_time":5.15,"predict_time":0.526,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-204","imputed":false},{"position":40,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1248.0,"score":0.14,"rank":38.29,"hrank":33.11,"imp":26.131,"train_time":27727.7,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-119","imputed":false},{"position":41,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1244.0,"score":0.121,"rank":38.57,"hrank":34.65,"imp":26.313,"train_time":10.95,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-141","imputed":false},{"position":42,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.154,"rank":38.57,"hrank":20.36,"imp":26.824,"train_time":3834.57,"predict_time":0.22,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+202\/-184","imputed":false},{"position":43,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1232.0,"score":0.196,"rank":39.43,"hrank":16.17,"imp":26.571,"train_time":51.52,"predict_time":40.201,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+254\/-242","imputed":false},{"position":44,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1230.0,"score":0.11,"rank":39.57,"hrank":35.32,"imp":26.159,"train_time":2001.62,"predict_time":1.335,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-165","imputed":false},{"position":45,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1230.0,"score":0.136,"rank":39.57,"hrank":33.22,"imp":25.643,"train_time":2001.62,"predict_time":0.274,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-195","imputed":false},{"position":46,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1228.0,"score":0.095,"rank":39.71,"hrank":34.99,"imp":27.829,"train_time":3.46,"predict_time":0.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-164","imputed":false},{"position":47,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1208.0,"score":0.007,"rank":41.29,"hrank":40.2,"imp":24.835,"train_time":5.38,"predict_time":0.624,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-73","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1202.0,"score":0.161,"rank":41.71,"hrank":31.15,"imp":26.36,"train_time":731.23,"predict_time":0.334,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+141\/-206","imputed":false},{"position":49,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1180.0,"score":0.085,"rank":43.29,"hrank":37.39,"imp":27.518,"train_time":9.4,"predict_time":1.444,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+141\/-217","imputed":false},{"position":50,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1180.0,"score":0.103,"rank":43.29,"hrank":37.02,"imp":26.198,"train_time":20.64,"predict_time":7.532,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-199","imputed":false},{"position":51,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1156.0,"score":0.078,"rank":45.0,"hrank":35.67,"imp":28.273,"train_time":2750.22,"predict_time":1.238,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-216","imputed":false},{"position":52,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1154.0,"score":0.116,"rank":45.14,"hrank":31.09,"imp":29.853,"train_time":8.73,"predict_time":0.238,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+220\/-289","imputed":false},{"position":53,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1105.0,"score":0.037,"rank":48.43,"hrank":44.8,"imp":30.859,"train_time":19.64,"predict_time":0.478,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-168","imputed":false},{"position":54,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1080.0,"score":0.157,"rank":50.0,"hrank":30.07,"imp":37.677,"train_time":180.48,"predict_time":1.239,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+269\/-350","imputed":false},{"position":55,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1042.0,"score":0.024,"rank":52.29,"hrank":50.21,"imp":32.764,"train_time":14.88,"predict_time":0.391,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+140\/-209","imputed":false},{"position":56,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1024.0,"score":0.04,"rank":53.36,"hrank":47.16,"imp":32.455,"train_time":45499.58,"predict_time":9.024,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+173\/-275","imputed":false},{"position":57,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1009.0,"score":0.015,"rank":54.21,"hrank":50.15,"imp":32.797,"train_time":45499.58,"predict_time":43.922,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-239","imputed":false},{"position":58,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":54.71,"hrank":51.9,"imp":39.076,"train_time":0.78,"predict_time":0.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-286","imputed":false},{"position":59,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1000.0,"score":0.0,"rank":54.71,"hrank":52.96,"imp":34.626,"train_time":23.33,"predict_time":3.304,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-196","imputed":false},{"position":60,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":982.0,"score":0.0,"rank":55.71,"hrank":54.5,"imp":31.684,"train_time":559.69,"predict_time":266.319,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-206","imputed":false},{"position":61,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":974.0,"score":0.011,"rank":56.14,"hrank":53.17,"imp":36.662,"train_time":5.96,"predict_time":0.898,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+139\/-237","imputed":false},{"position":62,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":970.0,"score":0.0,"rank":56.36,"hrank":54.25,"imp":41.826,"train_time":913.37,"predict_time":0.738,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-256","imputed":false},{"position":63,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":963.0,"score":0.022,"rank":56.71,"hrank":52.22,"imp":34.718,"train_time":8.78,"predict_time":0.688,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+175\/-260","imputed":false},{"position":64,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":945.0,"score":0.0,"rank":57.64,"hrank":55.5,"imp":42.468,"train_time":913.37,"predict_time":0.13,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-291","imputed":false},{"position":65,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":848.0,"score":0.095,"rank":62.0,"hrank":48.19,"imp":48.983,"train_time":180.48,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+298\/-483","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":844.0,"score":0.0,"rank":62.14,"hrank":61.31,"imp":53.547,"train_time":1679.96,"predict_time":11.05,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-206","imputed":false},{"position":67,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":826.0,"score":0.0,"rank":62.86,"hrank":62.45,"imp":40.756,"train_time":5.48,"predict_time":0.909,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-170","imputed":false},{"position":68,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":807.0,"score":0.0,"rank":63.57,"hrank":62.69,"imp":43.788,"train_time":866.21,"predict_time":9.353,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-270","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":762.0,"score":0.0,"rank":65.07,"hrank":64.54,"imp":57.429,"train_time":1679.96,"predict_time":1.824,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-265","imputed":false},{"position":70,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":756.0,"score":0.0,"rank":65.29,"hrank":64.6,"imp":49.012,"train_time":4.65,"predict_time":0.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-285","imputed":false},{"position":71,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":714.0,"score":0.0,"rank":66.5,"hrank":65.72,"imp":62.772,"train_time":273.7,"predict_time":0.508,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-339","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":262.0,"score":0.0,"rank":72.43,"hrank":72.41,"imp":70.07,"train_time":0.58,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-488","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/pareto_front_explorer.html
index 3c7f050cdb28ce34b345c9c6b0e29d4552019b10..8304f486aee146e7b09b4793ca10903c27f7f6c0 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.2793557109,"imp":23.2984151865,"elo":1359.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3431890039,"imp":23.0418722718,"elo":1306.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2382022972,"imp":16.0698126603,"elo":1487.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.2361036908,"imp":16.993720546,"elo":1487.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2379844651,"imp":29.853244074,"elo":1154.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2739576431,"imp":25.6426013142,"elo":1230.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3347926666,"imp":26.1589529753,"elo":1230.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0483681042,"imp":26.3132231692,"elo":1243.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0286685262,"imp":26.130830267,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.3103094802,"imp":24.5999577586,"elo":1315.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8984521391,"imp":36.6621872617,"elo":973.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2943369872,"imp":27.2773995398,"elo":1256.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.872894906,"imp":24.1285441429,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6878838583,"imp":34.7182521619,"elo":963.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2378249243,"imp":28.2726346167,"elo":1156.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":13.5689532413,"imp":26.2794881384,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.5257108651,"imp":25.6013864916,"elo":1253.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.0866582746,"imp":20.5363232435,"elo":1446.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":3.5768582903,"imp":20.4515442453,"elo":1450.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":9.1873597509,"imp":15.29437114,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4781119128,"imp":30.8585054606,"elo":1104.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5829764758,"imp":22.1528894542,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5509605348,"imp":21.8530875524,"elo":1356.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3043972003,"imp":34.6260033877,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5079486588,"imp":62.7718084569,"elo":714.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.8243625063,"imp":57.4293045949,"elo":762.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":11.050111703,"imp":53.5472107063,"elo":844.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1545608396,"imp":39.0757652771,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3337366528,"imp":26.3599301981,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":2.0214591481,"imp":24.7457435227,"elo":1272.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.5324903306,"imp":26.1984848423,"elo":1180.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":4.0293360205,"imp":17.9424813982,"elo":1494.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.781269562,"imp":17.5032291669,"elo":1538.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.855277474,"imp":13.2134098365,"elo":1595.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1322378847,"imp":14.3384135162,"elo":1660.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5456556456,"imp":5.6371518418,"elo":1738.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.556301258,"imp":22.9111586961,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":40.2005207582,"imp":26.5714387064,"elo":1232.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":40.1033683003,"imp":22.9824009828,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":399.3341259005,"imp":22.0617949196,"elo":1437.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2694943851,"imp":23.6032183869,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4342474097,"imp":6.1115865167,"elo":1710.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.9093813646,"imp":40.7563314744,"elo":826.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2337945274,"imp":15.4616732914,"elo":1528.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7280240831,"imp":23.7404809364,"elo":1378.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.8794831101,"imp":19.6046393967,"elo":1478.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.6401447093,"imp":19.5501472802,"elo":1487.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.650333935,"imp":13.6883146842,"elo":1655.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5140794313,"imp":12.3150344649,"elo":1591.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3534204381,"imp":43.7881530786,"elo":806.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.0244207573,"imp":32.4548558378,"elo":1023.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":43.9219279194,"imp":32.7974122778,"elo":1008.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622328922,"imp":27.8286351915,"elo":1228.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3908251335,"imp":32.7641285508,"elo":1042.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2202658411,"imp":26.8243500098,"elo":1243.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.293854225,"imp":24.7936831007,"elo":1290.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.6237615495,"imp":24.8347797971,"elo":1207.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0653522593,"imp":20.6819141626,"elo":1389.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":5.1289419395,"imp":20.9632476496,"elo":1395.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":266.3190646898,"imp":31.6844647486,"elo":981.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":240.6831310693,"imp":22.5779018618,"elo":1288.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":986.35207523,"imp":21.2879406926,"elo":1361.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.4435953081,"imp":27.5180847905,"elo":1180.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1464664376,"imp":22.3865918703,"elo":1352.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9262072088,"imp":20.9249968398,"elo":1444.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.2793557109,"x_train":326.3219590044,"imp":23.2984151865,"elo":1359.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3431890039,"x_train":29.5776098075,"imp":23.0418722718,"elo":1306.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2382022972,"x_train":3632.0960320665,"imp":16.0698126603,"elo":1487.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.2361036908,"x_train":3632.0960320665,"imp":16.993720546,"elo":1487.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2379844651,"x_train":8.7255028029,"imp":29.853244074,"elo":1154.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2739576431,"x_train":2001.6183652265,"imp":25.6426013142,"elo":1230.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3347926666,"x_train":2001.6183652265,"imp":26.1589529753,"elo":1230.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0483681042,"x_train":10.9494219165,"imp":26.3132231692,"elo":1243.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0286685262,"x_train":27727.7037054394,"imp":26.130830267,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.3103094802,"x_train":27727.7037054394,"imp":24.5999577586,"elo":1315.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8984521391,"x_train":5.955842539,"imp":36.6621872617,"elo":973.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2943369872,"x_train":1042.3872546633,"imp":27.2773995398,"elo":1256.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.872894906,"x_train":1042.3872546633,"imp":24.1285441429,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6878838583,"x_train":8.7848766331,"imp":34.7182521619,"elo":963.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2378249243,"x_train":2750.2207329287,"imp":28.2726346167,"elo":1156.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":13.5689532413,"x_train":2750.2207329287,"imp":26.2794881384,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.5257108651,"x_train":5.149666856,"imp":25.6013864916,"elo":1253.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.0866582746,"x_train":1394.8404749323,"imp":20.5363232435,"elo":1446.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":3.5768582903,"x_train":1394.8404749323,"imp":20.4515442453,"elo":1450.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":9.1873597509,"x_train":46.1120625115,"imp":15.29437114,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4781119128,"x_train":19.6359284437,"imp":30.8585054606,"elo":1104.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5829764758,"x_train":6995.9928750372,"imp":22.1528894542,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5509605348,"x_train":6995.9928750372,"imp":21.8530875524,"elo":1356.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3043972003,"x_train":23.3325961779,"imp":34.6260033877,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5079486588,"x_train":273.6950495461,"imp":62.7718084569,"elo":714.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.8243625063,"x_train":1679.959262267,"imp":57.4293045949,"elo":762.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":11.050111703,"x_train":1679.959262267,"imp":53.5472107063,"elo":844.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1545608396,"x_train":0.7756363005,"imp":39.0757652771,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3337366528,"x_train":731.2261557986,"imp":26.3599301981,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":2.0214591481,"x_train":731.2261557986,"imp":24.7457435227,"elo":1272.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.5324903306,"x_train":20.6393168714,"imp":26.1984848423,"elo":1180.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":4.0293360205,"x_train":4655.1453813481,"imp":17.9424813982,"elo":1494.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.781269562,"x_train":4655.1453813481,"imp":17.5032291669,"elo":1538.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.855277474,"x_train":10.9716481006,"imp":13.2134098365,"elo":1595.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1322378847,"x_train":2837.9578551597,"imp":14.3384135162,"elo":1660.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5456556456,"x_train":2837.9578551597,"imp":5.6371518418,"elo":1738.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.556301258,"x_train":22.2617203941,"imp":22.9111586961,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":40.2005207582,"x_train":51.5168795924,"imp":26.5714387064,"elo":1232.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":40.1033683003,"x_train":10182.2577368658,"imp":22.9824009828,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":399.3341259005,"x_train":10182.2577368658,"imp":22.0617949196,"elo":1437.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2694943851,"x_train":5.6498915817,"imp":23.6032183869,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4342474097,"x_train":52.6612506816,"imp":6.1115865167,"elo":1710.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.9093813646,"x_train":5.4835182137,"imp":40.7563314744,"elo":826.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2337945274,"x_train":5.3571462631,"imp":15.4616732914,"elo":1528.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7280240831,"x_train":17.3386025246,"imp":23.7404809364,"elo":1378.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.8794831101,"x_train":3829.3977883333,"imp":19.6046393967,"elo":1478.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.6401447093,"x_train":3829.3977883333,"imp":19.5501472802,"elo":1487.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.650333935,"x_train":10.7776505002,"imp":13.6883146842,"elo":1655.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5140794313,"x_train":9.4477958531,"imp":12.3150344649,"elo":1591.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3534204381,"x_train":866.2091617048,"imp":43.7881530786,"elo":806.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.0244207573,"x_train":45499.5818819746,"imp":32.4548558378,"elo":1023.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":43.9219279194,"x_train":45499.5818819746,"imp":32.7974122778,"elo":1008.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622328922,"x_train":3.4571113671,"imp":27.8286351915,"elo":1228.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3908251335,"x_train":14.8754458124,"imp":32.7641285508,"elo":1042.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2202658411,"x_train":3834.5684419838,"imp":26.8243500098,"elo":1243.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.293854225,"x_train":3834.5684419838,"imp":24.7936831007,"elo":1290.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.6237615495,"x_train":5.3772386715,"imp":24.8347797971,"elo":1207.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0653522593,"x_train":2342.5781829117,"imp":20.6819141626,"elo":1389.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":5.1289419395,"x_train":2342.5781829117,"imp":20.9632476496,"elo":1395.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":266.3190646898,"x_train":559.6852708101,"imp":31.6844647486,"elo":981.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":240.6831310693,"x_train":34015.1930441921,"imp":22.5779018618,"elo":1288.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":986.35207523,"x_train":34015.1930441921,"imp":21.2879406926,"elo":1361.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.4435953081,"x_train":9.4039762628,"imp":27.5180847905,"elo":1180.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1464664376,"x_train":2149.2969294609,"imp":22.3865918703,"elo":1352.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9262072088,"x_train":2149.2969294609,"imp":20.9249968398,"elo":1444.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
index 0376e4fab5e29174980fd4a3c312e2ad4fd9c4fa..c6090cd864eb34f8ad2c3c2af24860c737105db7 100644
--- a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":326.3219590044,"imp":23.3197524236,"elo":1348.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":29.5776098075,"imp":23.0628255508,"elo":1299.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":61.6242204097,"imp":20.6259991536,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":146.2103635918,"imp":19.5055911201,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":255.2797833412,"imp":18.8193637339,"elo":1464.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":461.9229712959,"imp":18.2731442985,"elo":1478.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":901.1280977162,"imp":18.2852095466,"elo":1472.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1816.8097050761,"imp":17.5175425789,"elo":1478.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2717.6279686964,"imp":17.3028855872,"elo":1468.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":3632.0960320665,"imp":17.0155156555,"elo":1480.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":8.7255028029,"imp":29.8723469778,"elo":1138.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":18.8295905976,"imp":29.4165301447,"elo":1149.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":48.1908367108,"imp":29.1397802936,"elo":1156.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":97.2610688395,"imp":28.5897347697,"elo":1171.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":247.1661812847,"imp":27.5299606151,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":498.9992240184,"imp":27.0619964647,"elo":1204.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":996.7852549912,"imp":26.5530110316,"elo":1211.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1493.1381281887,"imp":26.33390516,"elo":1215.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2001.6183652265,"imp":26.1789896298,"elo":1224.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":10.9494219165,"imp":26.333260115,"elo":1221.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":116.2996116626,"imp":26.0861694223,"elo":1223.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":481.3286902295,"imp":25.9913850731,"elo":1233.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":1094.9483089107,"imp":25.9632540248,"elo":1236.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":2930.5133530918,"imp":25.7635733094,"elo":1244.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":7066.1129355898,"imp":25.5121406641,"elo":1249.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":13790.4831017652,"imp":25.0032650117,"elo":1281.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":20842.8750821883,"imp":24.7834516827,"elo":1292.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":27727.7037054394,"imp":24.6235143359,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":5.955842539,"imp":36.6903295327,"elo":959.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":11.0977776932,"imp":33.0189764248,"elo":1043.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":26.6390299573,"imp":27.7519255431,"elo":1155.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":52.8400597922,"imp":26.5153909584,"elo":1224.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":130.2628378892,"imp":24.8026985651,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":259.6675727983,"imp":24.6271917138,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":518.6557637866,"imp":24.4219453292,"elo":1313.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":777.9260343247,"imp":24.2106018828,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1042.3872546633,"imp":24.1716864028,"elo":1327.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.7848766331,"imp":34.7368118298,"elo":955.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":23.0929542575,"imp":33.268693352,"elo":1005.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":64.8591819071,"imp":30.131462827,"elo":1115.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":131.6977426498,"imp":29.0035091083,"elo":1179.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":335.6638780889,"imp":27.4324973088,"elo":1218.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":677.7524464582,"imp":26.7236363263,"elo":1242.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1359.9623914477,"imp":26.3400526889,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2046.8674083081,"imp":26.3031948541,"elo":1265.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2750.2207329287,"imp":26.3011355079,"elo":1264.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":5.149666856,"imp":25.6199848887,"elo":1244.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":12.638683475,"imp":24.5398838227,"elo":1286.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":32.1130206474,"imp":23.4160350344,"elo":1346.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":66.7832587484,"imp":22.5663410197,"elo":1389.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":172.6401098684,"imp":21.5767763686,"elo":1409.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":345.9193991848,"imp":21.2445989196,"elo":1422.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":693.5342730413,"imp":20.8902882844,"elo":1446.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":1039.0554015758,"imp":20.7170294095,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1394.8404749323,"imp":20.4720931948,"elo":1457.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":46.1120625115,"imp":15.5115935998,"elo":1502.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":19.6359284437,"imp":30.8866944985,"elo":1100.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":54.5342406386,"imp":27.999554373,"elo":1155.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":167.3188814038,"imp":24.6714431267,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":346.6206170635,"imp":23.8917054055,"elo":1301.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":852.383981166,"imp":23.3257705825,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1740.145368883,"imp":22.8932583059,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3492.7302851634,"imp":22.9195960872,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":5232.8252409525,"imp":22.2757576311,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6995.9928750372,"imp":21.8737738633,"elo":1358.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":23.3325961779,"imp":34.6489122936,"elo":995.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":273.6950495461,"imp":62.8288020694,"elo":684.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":577.7198455713,"imp":57.1996938609,"elo":775.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1420.3840579116,"imp":53.7191489746,"elo":832.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1679.959262267,"imp":53.5547287705,"elo":830.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.7756363005,"imp":39.1605432301,"elo":998.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.64517611,"imp":31.9073276602,"elo":1075.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":15.6783968263,"imp":27.666405584,"elo":1151.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":34.2083641857,"imp":25.4184350204,"elo":1183.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":88.9128109825,"imp":24.1606702604,"elo":1272.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":180.8451833376,"imp":23.9756519157,"elo":1277.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":363.1699124457,"imp":24.4008408933,"elo":1272.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":545.3863630441,"imp":24.593197693,"elo":1273.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":731.2261557986,"imp":24.7680738423,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":20.6393168714,"imp":26.217618248,"elo":1173.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":44.6022278057,"imp":21.4343579901,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":114.6509525781,"imp":19.1477629466,"elo":1444.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":230.7629821174,"imp":17.7310879223,"elo":1506.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":577.5814409002,"imp":16.8441666583,"elo":1553.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1157.7764691244,"imp":16.3649374762,"elo":1573.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2306.7724282356,"imp":16.3773806269,"elo":1578.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3460.7466523769,"imp":17.0419733271,"elo":1559.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4655.1453813481,"imp":17.5250924206,"elo":1548.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":10.9716481006,"imp":13.2345164804,"elo":1596.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":27.742192784,"imp":13.0764507206,"elo":1593.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":70.0993253105,"imp":12.7634150165,"elo":1629.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":140.813217678,"imp":12.8115263622,"elo":1685.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":354.18227158,"imp":12.6756223603,"elo":1704.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":706.681175239,"imp":12.1708307979,"elo":1724.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1422.0173619203,"imp":10.6463681559,"elo":1731.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2134.9546013675,"imp":9.097776014,"elo":1736.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2837.9578551597,"imp":5.6613192609,"elo":1747.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":22.2617203941,"imp":22.9324150573,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.5168795924,"imp":26.5911909892,"elo":1215.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":99.6707673609,"imp":26.1279843826,"elo":1245.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":250.1259988665,"imp":25.2381687184,"elo":1312.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":520.4892551291,"imp":24.7900103576,"elo":1345.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":1282.2422972143,"imp":23.8281813917,"elo":1388.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":2545.8074097094,"imp":23.1621024131,"elo":1412.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":5076.4052782117,"imp":22.9578764419,"elo":1422.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":7591.7531128733,"imp":22.6275102214,"elo":1427.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":10182.2577368658,"imp":22.0707725769,"elo":1442.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":5.6498915817,"imp":23.6235096837,"elo":1386.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":52.6612506816,"imp":6.1334504678,"elo":1698.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":5.4835182137,"imp":40.7747113143,"elo":820.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":5.3571462631,"imp":15.4823634336,"elo":1532.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":17.3386025246,"imp":23.7623042298,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":34.7857185326,"imp":23.0715461263,"elo":1403.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":92.2135159049,"imp":22.5165281719,"elo":1425.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":188.1504220603,"imp":21.9549767719,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":467.3246962797,"imp":21.5861201405,"elo":1455.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":939.5188242924,"imp":21.1738829848,"elo":1470.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1881.8343045706,"imp":20.3622472104,"elo":1482.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2843.5213469349,"imp":20.0176345562,"elo":1488.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3829.3977883333,"imp":19.5717472605,"elo":1490.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":10.7776505002,"imp":13.7096329452,"elo":1672.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":9.4477958531,"imp":12.335908483,"elo":1594.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":866.2091617048,"imp":43.8044667741,"elo":780.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1369.7270777514,"imp":43.5405020964,"elo":805.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1966.5345052574,"imp":42.7742107507,"elo":827.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2789.3976815102,"imp":41.2081636838,"elo":851.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5422.0998389164,"imp":37.967863537,"elo":911.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11167.1140617992,"imp":35.316753412,"elo":946.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23229.2182271869,"imp":34.1998275425,"elo":973.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34425.4477297998,"imp":33.6322122465,"elo":990.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":45499.5818819746,"imp":32.815561821,"elo":1016.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":3.4571113671,"imp":27.8478694128,"elo":1223.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.8754458124,"imp":32.8102812646,"elo":1025.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":30.9336251788,"imp":29.2987415582,"elo":1124.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":94.1324659798,"imp":26.948093656,"elo":1211.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":200.5394916928,"imp":26.1769515242,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":461.9575980829,"imp":25.7197189536,"elo":1271.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":911.8260308778,"imp":25.4028650756,"elo":1274.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1835.2639428382,"imp":25.1008499739,"elo":1277.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2805.3782071109,"imp":24.8395744398,"elo":1290.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3834.5684419838,"imp":24.817488205,"elo":1285.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":5.3772386715,"imp":24.8541959881,"elo":1192.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":17.3939236522,"imp":23.4295375178,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":53.4704779379,"imp":22.9359564559,"elo":1309.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":108.3183028113,"imp":22.3257100045,"elo":1336.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":286.0649937532,"imp":21.830401136,"elo":1363.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":578.8859474941,"imp":21.5075500365,"elo":1377.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":1164.3479303143,"imp":21.3354239659,"elo":1382.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":1743.9656718772,"imp":21.2302690665,"elo":1392.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":2342.5781829117,"imp":20.9835399671,"elo":1395.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":559.6852708101,"imp":31.7031001295,"elo":969.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1879.2300695283,"imp":28.8909436091,"elo":1047.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":5685.7541761762,"imp":25.4340809006,"elo":1189.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":12716.1318239357,"imp":23.392764892,"elo":1267.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":32397.1155311761,"imp":21.2766142176,"elo":1357.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":34015.1930441921,"imp":21.3085338575,"elo":1355.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":9.4039762628,"imp":27.5376238303,"elo":1163.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":19.9957554462,"imp":26.4014081913,"elo":1231.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":52.7123918422,"imp":24.7132555183,"elo":1322.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":106.7857168731,"imp":23.855313907,"elo":1344.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":267.72178804,"imp":22.3741327725,"elo":1395.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":532.9559679306,"imp":21.4318530228,"elo":1431.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":1070.1666001206,"imp":21.0251373274,"elo":1446.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1601.8304365569,"imp":20.9463925049,"elo":1453.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2149.2969294609,"imp":20.9647036579,"elo":1455.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.2793557109,"x_train":326.3219590044,"imp":23.3197524236,"elo":1348.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3431890039,"x_train":29.5776098075,"imp":23.0628255508,"elo":1299.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.5366471716,"x_train":61.6242204097,"imp":20.6259991536,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.7424350969,"x_train":146.2103635918,"imp":19.5055911201,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.9011323043,"x_train":255.2797833412,"imp":18.8193637339,"elo":1464.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.9752719466,"x_train":461.9229712959,"imp":18.2731442985,"elo":1478.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1210385171,"x_train":901.1280977162,"imp":18.2852095466,"elo":1472.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1513120746,"x_train":1816.8097050761,"imp":17.5175425789,"elo":1478.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2237069108,"x_train":2717.6279686964,"imp":17.3028855872,"elo":1468.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2361036908,"x_train":3632.0960320665,"imp":17.0155156555,"elo":1480.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2379844651,"x_train":8.7255028029,"imp":29.8723469778,"elo":1138.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3316836793,"x_train":18.8295905976,"imp":29.4165301447,"elo":1149.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5417567135,"x_train":48.1908367108,"imp":29.1397802936,"elo":1156.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.7237500996,"x_train":97.2610688395,"imp":28.5897347697,"elo":1171.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.891756985,"x_train":247.1661812847,"imp":27.5299606151,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.0241181517,"x_train":498.9992240184,"imp":27.0619964647,"elo":1204.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1473050403,"x_train":996.7852549912,"imp":26.5530110316,"elo":1211.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2443893534,"x_train":1493.1381281887,"imp":26.33390516,"elo":1215.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3347926666,"x_train":2001.6183652265,"imp":26.1789896298,"elo":1224.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0483681042,"x_train":10.9494219165,"imp":26.333260115,"elo":1221.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.1342425713,"x_train":116.2996116626,"imp":26.0861694223,"elo":1223.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.2180498533,"x_train":481.3286902295,"imp":25.9913850731,"elo":1233.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.207248058,"x_train":1094.9483089107,"imp":25.9632540248,"elo":1236.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.2016961442,"x_train":2930.5133530918,"imp":25.7635733094,"elo":1244.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.2168105015,"x_train":7066.1129355898,"imp":25.5121406641,"elo":1249.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.2240727367,"x_train":13790.4831017652,"imp":25.0032650117,"elo":1281.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.2523087712,"x_train":20842.8750821883,"imp":24.7834516827,"elo":1292.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.3103094802,"x_train":27727.7037054394,"imp":24.6235143359,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8984521391,"x_train":5.955842539,"imp":36.6903295327,"elo":959.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1872124249,"x_train":11.0977776932,"imp":33.0189764248,"elo":1043.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9761079981,"x_train":26.6390299573,"imp":27.7519255431,"elo":1155.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2785129959,"x_train":52.8400597922,"imp":26.5153909584,"elo":1224.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.703820181,"x_train":130.2628378892,"imp":24.8026985651,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4670921859,"x_train":259.6675727983,"imp":24.6271917138,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7516614998,"x_train":518.6557637866,"imp":24.4219453292,"elo":1313.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7521164634,"x_train":777.9260343247,"imp":24.2106018828,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.872894906,"x_train":1042.3872546633,"imp":24.1716864028,"elo":1327.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6878838583,"x_train":8.7848766331,"imp":34.7368118298,"elo":955.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.5412796889,"x_train":23.0929542575,"imp":33.268693352,"elo":1005.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1850045276,"x_train":64.8591819071,"imp":30.131462827,"elo":1115.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.2907795257,"x_train":131.6977426498,"imp":29.0035091083,"elo":1179.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.7270220551,"x_train":335.6638780889,"imp":27.4324973088,"elo":1218.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.095832519,"x_train":677.7524464582,"imp":26.7236363263,"elo":1242.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":10.3587654477,"x_train":1359.9623914477,"imp":26.3400526889,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":11.4534017479,"x_train":2046.8674083081,"imp":26.3031948541,"elo":1265.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":13.5689532413,"x_train":2750.2207329287,"imp":26.3011355079,"elo":1264.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5257108651,"x_train":5.149666856,"imp":25.6199848887,"elo":1244.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.7506945395,"x_train":12.638683475,"imp":24.5398838227,"elo":1286.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.7230234059,"x_train":32.1130206474,"imp":23.4160350344,"elo":1346.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.8714779904,"x_train":66.7832587484,"imp":22.5663410197,"elo":1389.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.11547675,"x_train":172.6401098684,"imp":21.5767763686,"elo":1409.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.59344747,"x_train":345.9193991848,"imp":21.2445989196,"elo":1422.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.9034755619,"x_train":693.5342730413,"imp":20.8902882844,"elo":1446.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.7091890428,"x_train":1039.0554015758,"imp":20.7170294095,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.5768582903,"x_train":1394.8404749323,"imp":20.4720931948,"elo":1457.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":9.1873597509,"x_train":46.1120625115,"imp":15.5115935998,"elo":1502.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4781119128,"x_train":19.6359284437,"imp":30.8866944985,"elo":1100.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.0130189541,"x_train":54.5342406386,"imp":27.999554373,"elo":1155.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8126811516,"x_train":167.3188814038,"imp":24.6714431267,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.8125941214,"x_train":346.6206170635,"imp":23.8917054055,"elo":1301.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.4546068712,"x_train":852.383981166,"imp":23.3257705825,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.9518046583,"x_train":1740.145368883,"imp":22.8932583059,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.3693694624,"x_train":3492.7302851634,"imp":22.9195960872,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.5205578648,"x_train":5232.8252409525,"imp":22.2757576311,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5509605348,"x_train":6995.9928750372,"imp":21.8737738633,"elo":1358.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3043972003,"x_train":23.3325961779,"imp":34.6489122936,"elo":995.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5079486588,"x_train":273.6950495461,"imp":62.8288020694,"elo":684.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":7.5114343904,"x_train":577.7198455713,"imp":57.1996938609,"elo":775.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":12.1599901492,"x_train":1420.3840579116,"imp":53.7191489746,"elo":832.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":11.050111703,"x_train":1679.959262267,"imp":53.5547287705,"elo":830.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1545608396,"x_train":0.7756363005,"imp":39.1605432301,"elo":998.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4758379412,"x_train":4.64517611,"imp":31.9073276602,"elo":1075.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8964239223,"x_train":15.6783968263,"imp":27.666405584,"elo":1151.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2954751768,"x_train":34.2083641857,"imp":25.4184350204,"elo":1183.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.5928379977,"x_train":88.9128109825,"imp":24.1606702604,"elo":1272.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7104260673,"x_train":180.8451833376,"imp":23.9756519157,"elo":1277.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7486850861,"x_train":363.1699124457,"imp":24.4008408933,"elo":1272.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7684293692,"x_train":545.3863630441,"imp":24.593197693,"elo":1273.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":2.0214591481,"x_train":731.2261557986,"imp":24.7680738423,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.5324903306,"x_train":20.6393168714,"imp":26.217618248,"elo":1173.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.4467410735,"x_train":44.6022278057,"imp":21.4343579901,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.4636747092,"x_train":114.6509525781,"imp":19.1477629466,"elo":1444.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.7310288169,"x_train":230.7629821174,"imp":17.7310879223,"elo":1506.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.1150100532,"x_train":577.5814409002,"imp":16.8441666583,"elo":1553.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.7078734544,"x_train":1157.7764691244,"imp":16.3649374762,"elo":1573.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.3622951828,"x_train":2306.7724282356,"imp":16.3773806269,"elo":1578.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.0724225096,"x_train":3460.7466523769,"imp":17.0419733271,"elo":1559.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.781269562,"x_train":4655.1453813481,"imp":17.5250924206,"elo":1548.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.855277474,"x_train":10.9716481006,"imp":13.2345164804,"elo":1596.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2634027639,"x_train":27.742192784,"imp":13.0764507206,"elo":1593.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.5574648521,"x_train":70.0993253105,"imp":12.7634150165,"elo":1629.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.4891965121,"x_train":140.813217678,"imp":12.8115263622,"elo":1685.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.4282035488,"x_train":354.18227158,"imp":12.6756223603,"elo":1704.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.8048742763,"x_train":706.681175239,"imp":12.1708307979,"elo":1724.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.3192714939,"x_train":1422.0173619203,"imp":10.6463681559,"elo":1731.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1275877868,"x_train":2134.9546013675,"imp":9.097776014,"elo":1736.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5456556456,"x_train":2837.9578551597,"imp":5.6613192609,"elo":1747.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.556301258,"x_train":22.2617203941,"imp":22.9324150573,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":40.2005207582,"x_train":51.5168795924,"imp":26.5911909892,"elo":1215.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":66.1274474954,"x_train":99.6707673609,"imp":26.1279843826,"elo":1245.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":107.1341355034,"x_train":250.1259988665,"imp":25.2381687184,"elo":1312.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":145.95289892,"x_train":520.4892551291,"imp":24.7900103576,"elo":1345.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":207.3229714868,"x_train":1282.2422972143,"imp":23.8281813917,"elo":1388.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":226.2356139625,"x_train":2545.8074097094,"imp":23.1621024131,"elo":1412.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":283.6685778956,"x_train":5076.4052782117,"imp":22.9578764419,"elo":1422.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":335.2555385893,"x_train":7591.7531128733,"imp":22.6275102214,"elo":1427.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":399.3341259005,"x_train":10182.2577368658,"imp":22.0707725769,"elo":1442.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2694943851,"x_train":5.6498915817,"imp":23.6235096837,"elo":1386.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.4342474097,"x_train":52.6612506816,"imp":6.1334504678,"elo":1698.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.9093813646,"x_train":5.4835182137,"imp":40.7747113143,"elo":820.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2337945274,"x_train":5.3571462631,"imp":15.4823634336,"elo":1532.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.7280240831,"x_train":17.3386025246,"imp":23.7623042298,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":1.1443311646,"x_train":34.7857185326,"imp":23.0715461263,"elo":1403.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.9947973581,"x_train":92.2135159049,"imp":22.5165281719,"elo":1425.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.7539019401,"x_train":188.1504220603,"imp":21.9549767719,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":3.0347115346,"x_train":467.3246962797,"imp":21.5861201405,"elo":1455.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.4643032105,"x_train":939.5188242924,"imp":21.1738829848,"elo":1470.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.9883342671,"x_train":1881.8343045706,"imp":20.3622472104,"elo":1482.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.4674678667,"x_train":2843.5213469349,"imp":20.0176345562,"elo":1488.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.6401447093,"x_train":3829.3977883333,"imp":19.5717472605,"elo":1490.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.650333935,"x_train":10.7776505002,"imp":13.7096329452,"elo":1672.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5140794313,"x_train":9.4477958531,"imp":12.335908483,"elo":1594.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3534204381,"x_train":866.2091617048,"imp":43.8044667741,"elo":780.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":11.7604054174,"x_train":1369.7270777514,"imp":43.5405020964,"elo":805.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":13.9844555509,"x_train":1966.5345052574,"imp":42.7742107507,"elo":827.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.4034660481,"x_train":2789.3976815102,"imp":41.2081636838,"elo":851.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.4052999829,"x_train":5422.0998389164,"imp":37.967863537,"elo":911.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.7782915986,"x_train":11167.1140617992,"imp":35.316753412,"elo":946.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":26.7442737345,"x_train":23229.2182271869,"imp":34.1998275425,"elo":973.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.7906715653,"x_train":34425.4477297998,"imp":33.6322122465,"elo":990.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":43.9219279194,"x_train":45499.5818819746,"imp":32.815561821,"elo":1016.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622328922,"x_train":3.4571113671,"imp":27.8478694128,"elo":1223.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3908251335,"x_train":14.8754458124,"imp":32.8102812646,"elo":1025.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7508941509,"x_train":30.9336251788,"imp":29.2987415582,"elo":1124.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2080831708,"x_train":94.1324659798,"imp":26.948093656,"elo":1211.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2008541117,"x_train":200.5394916928,"imp":26.1769515242,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2778500539,"x_train":461.9575980829,"imp":25.7197189536,"elo":1271.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4463341877,"x_train":911.8260308778,"imp":25.4028650756,"elo":1274.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6817828504,"x_train":1835.2639428382,"imp":25.1008499739,"elo":1277.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0166546965,"x_train":2805.3782071109,"imp":24.8395744398,"elo":1290.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.293854225,"x_train":3834.5684419838,"imp":24.817488205,"elo":1285.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6237615495,"x_train":5.3772386715,"imp":24.8541959881,"elo":1192.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.8908831859,"x_train":17.3939236522,"imp":23.4295375178,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.8059237272,"x_train":53.4704779379,"imp":22.9359564559,"elo":1309.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.2721762964,"x_train":108.3183028113,"imp":22.3257100045,"elo":1336.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.6366016842,"x_train":286.0649937532,"imp":21.830401136,"elo":1363.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":5.2825632775,"x_train":578.8859474941,"imp":21.5075500365,"elo":1377.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":6.035961193,"x_train":1164.3479303143,"imp":21.3354239659,"elo":1382.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":5.2659142989,"x_train":1743.9656718772,"imp":21.2302690665,"elo":1392.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":5.1289419395,"x_train":2342.5781829117,"imp":20.9835399671,"elo":1395.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":266.3190646898,"x_train":559.6852708101,"imp":31.7031001295,"elo":969.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":588.4423040375,"x_train":1879.2300695283,"imp":28.8909436091,"elo":1047.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":968.8511096453,"x_train":5685.7541761762,"imp":25.4340809006,"elo":1189.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":1043.4500106951,"x_train":12716.1318239357,"imp":23.392764892,"elo":1267.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":1036.6685407389,"x_train":32397.1155311761,"imp":21.2766142176,"elo":1357.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":986.35207523,"x_train":34015.1930441921,"imp":21.3085338575,"elo":1355.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.4435953081,"x_train":9.4039762628,"imp":27.5376238303,"elo":1163.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.6504460427,"x_train":19.9957554462,"imp":26.4014081913,"elo":1231.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.9062277071,"x_train":52.7123918422,"imp":24.7132555183,"elo":1322.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.9180079213,"x_train":106.7857168731,"imp":23.855313907,"elo":1344.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.0548434429,"x_train":267.72178804,"imp":22.3741327725,"elo":1395.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.5313546067,"x_train":532.9559679306,"imp":21.4318530228,"elo":1431.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.3070752733,"x_train":1070.1666001206,"imp":21.0251373274,"elo":1446.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.2264927788,"x_train":1601.8304365569,"imp":20.9463925049,"elo":1453.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.9262072088,"x_train":2149.2969294609,"imp":20.9647036579,"elo":1455.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..15c86fd191004dbdc382ebda0498773471bb128a
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_multiclass/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "TabFM (default)", "RealTabPFN-2.5 (tuned)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (default)", "TabPFN-3 (default)", "RealMLP (tuned + ensembled)", "TabICLv2 (default)", "AutoGluon 1.4 (best, 4h)", "LimiX (default)", "RealMLP (tuned)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "TabM (tuned + ensembled)", "TabM (tuned)", "LightGBM (tuned + ensembled)", "LightGBM (tuned)", "xRFM (tuned + ensembled)", "TabDPT (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabDPT-Turbo (default)", "XGBoost (tuned)", "TabM (default)", "TabDPT (tuned)", "iLTM (tuned + ensembled)", "BetaTabPFN (default)", "ModernNCA (tuned + ensembled)", "SAP-RPT-OSS (default)", "xRFM (tuned)", "ModernNCA (tuned)", "ExtraTrees (tuned + ensembled)", "EBM (tuned + ensembled)", "CatBoost (default)", "TorchMLP (tuned + ensembled)", "iLTM (tuned)", "RandomForest (tuned + ensembled)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (tuned)", "LightGBM (default)", "EBM (tuned)", "EBM (default)", "TorchMLP (tuned)", "TabDPT (default)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "TabSwift (default)", "XGBoost (default)", "RandomForest (tuned)", "RealMLP (default)", "xRFM (default)", "FastaiMLP (tuned)", "ChimeraBoost (default)", "ModernNCA (default)", "KNN (tuned + ensembled)", "TorchMLP (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "OrionMSP (default)", "RandomForest (default)", "iLTM (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "FastaiMLP (default)", "Linear (tuned)", "KNN (tuned)", "PerpetualBooster (tuned + ensembled)", "TabFlex (default)", "TabSTAR (default)", "PerpetualBooster (tuned)", "Linear (default)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.2857142984867096, 0.1428571492433548, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.7142857313156128, NaN, 0.2857142984867096, 0.7142857313156128, 0.7142857313156128, 1.0, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.8571428656578064, 0.7142857313156128, NaN, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.2857142984867096, 0.1428571492433548, NaN, 0.5714285969734192, 0.6428571343421936, 0.4285714328289032, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 1.0, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, NaN, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.7142857313156128, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5714285969734192, 0.0, 0.1428571492433548, 0.3571428656578064, 0.2857142984867096, NaN, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 1.0, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, NaN, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 1.0, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.2857142984867096, 0.0, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, NaN, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, NaN, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.7142857313156128, 1.0, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, NaN, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, NaN, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 1.0, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, NaN, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, NaN, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, NaN, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 1.0, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.4285714328289032, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, NaN, 0.2857142984867096, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 0.4285714328289032, 0.8571428656578064, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.0, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.3571428656578064, 0.5714285969734192, 0.7142857313156128, NaN, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, NaN, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.8571428656578064, 0.4285714328289032, 0.7857142686843872, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, NaN, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, NaN, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, NaN, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, NaN, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2142857164144516, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, NaN, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, NaN, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.0, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, NaN, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, NaN, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, NaN, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.0, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.0, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, NaN, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 1.0, 0.8571428656578064, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, NaN, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, NaN, 0.3571428656578064, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.6428571343421936, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.4285714328289032, 0.0, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, NaN, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.5714285969734192, 0.1428571492433548, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, NaN, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, NaN, 0.4285714328289032, 0.6428571343421936, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 1.0, 0.7142857313156128, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, NaN, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.2857142984867096, 0.0, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.0, 0.0, 0.0, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.3571428656578064, 0.2857142984867096, NaN, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, NaN, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, NaN, 0.5714285969734192, 0.2857142984867096, 0.8571428656578064, 0.5714285969734192, 1.0, 0.8571428656578064], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, NaN, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.4285714328289032, 0.5714285969734192, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.2857142984867096, NaN, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, NaN, 0.5714285969734192, 0.6428571343421936, 0.8571428656578064], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.7142857313156128, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, NaN, 0.5714285969734192, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.0, 0.4285714328289032, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, NaN, 0.7142857313156128], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8988095284},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8968254026},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.8829365149},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8541666731},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8511904828},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.810515882},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8075396913},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7658730251},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.7579365191},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7301587413},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7301587404},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7281746117},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7222222317},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7222222328},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7222222324},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.7142857251},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6884920721},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6845238184},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6825396942},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.676587313},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6349206453},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.630952393},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6289682639},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.6170635047},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.607142869},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.599206358},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.597222233},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5942460421},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5912698578},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.589285723},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5843254067},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5724206449},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5515873132},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5416666772},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5257936627},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5238095344},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5079365188},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4980158843},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4910714397},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4880952477},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4821428667},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4781746134},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4781746136},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.466269852},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4642857253},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4642857237},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.4623015968},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4404761982},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.43452382},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4126984214},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.4126984222},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3888888982},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.3869047715},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3412698493},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.3194444527},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.28769842},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2728174685},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2609127052},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2539682618},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2539682616},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2400793721},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2341269902},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2311508006},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2261904834},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2132936573},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1527777838},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1507936555},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1408730189},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1309523845},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1101190508},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1071428607},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0902777809},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0079365083}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..111b3b859a780991f7047a7ffb1c97d106117d75
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_multiclass/datasets_small/winrate_matrix.csv
@@ -0,0 +1,74 @@
+,"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),TabFM (default),RealTabPFN-2.5 (tuned),TabPFN-2.6 (default),RealTabPFN-2.5 (default),TabPFN-3 (default),RealMLP (tuned + ensembled),TabICLv2 (default),"AutoGluon 1.4 (best, 4h)",LimiX (default),RealMLP (tuned),CatBoost (tuned + ensembled),CatBoost (tuned),TabM (tuned + ensembled),TabM (tuned),LightGBM (tuned + ensembled),LightGBM (tuned),xRFM (tuned + ensembled),TabDPT (tuned + ensembled),XGBoost (tuned + ensembled),TabDPT-Turbo (default),XGBoost (tuned),TabM (default),TabDPT (tuned),iLTM (tuned + ensembled),BetaTabPFN (default),ModernNCA (tuned + ensembled),SAP-RPT-OSS (default),xRFM (tuned),ModernNCA (tuned),ExtraTrees (tuned + ensembled),EBM (tuned + ensembled),CatBoost (default),TorchMLP (tuned + ensembled),iLTM (tuned),RandomForest (tuned + ensembled),FastaiMLP (tuned + ensembled),ExtraTrees (tuned),LightGBM (default),EBM (tuned),EBM (default),TorchMLP (tuned),TabDPT (default),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),TabSwift (default),XGBoost (default),RandomForest (tuned),RealMLP (default),xRFM (default),FastaiMLP (tuned),ChimeraBoost (default),ModernNCA (default),KNN (tuned + ensembled),TorchMLP (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),OrionMSP (default),RandomForest (default),iLTM (default),ExtraTrees (default),Linear (tuned + ensembled),FastaiMLP (default),Linear (tuned),KNN (tuned),PerpetualBooster (tuned + ensembled),TabFlex (default),TabSTAR (default),PerpetualBooster (tuned),Linear (default),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.2857143,0.14285715,0.71428573,0.5714286,0.42857143,0.71428573,0.85714287,0.71428573,0.85714287,1.0,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0,0.85714287,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,0.85714287,0.85714287,1.0,1.0,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.71428573,,0.2857143,0.71428573,0.71428573,1.0,0.5714286,0.85714287,0.71428573,0.71428573,0.85714287,1.0,1.0,1.0,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.85714287,1.0,0.71428573,0.71428573,1.0,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.85714287,0.71428573,,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.2857143,0.2857143,0.14285715,,0.5714286,0.64285713,0.42857143,0.85714287,0.5714286,0.71428573,0.5714286,1.0,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.71428573,1.0,0.71428573,0.71428573,1.0,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.42857143,0.2857143,0.14285715,0.42857143,,0.71428573,0.5714286,0.85714287,0.42857143,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.85714287,1.0,0.71428573,0.71428573,1.0,0.71428573,1.0,0.71428573,0.71428573,1.0,0.85714287,0.71428573,1.0,0.85714287,1.0,0.71428573,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.71428573,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.5714286,0.0,0.14285715,0.35714287,0.2857143,,0.5714286,0.71428573,0.5714286,0.5714286,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.5714286,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,0.71428573,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.2857143,0.42857143,0.14285715,0.5714286,0.42857143,0.42857143,,0.5714286,0.71428573,0.42857143,0.71428573,0.85714287,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,1.0,0.71428573,1.0,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.42857143,,0.42857143,0.5714286,0.42857143,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,1.0,0.5714286,0.85714287,0.85714287,0.5714286,0.85714287,0.71428573,0.85714287,0.5714286,0.5714286,0.85714287,0.5714286,0.85714287,0.85714287,0.85714287,1.0,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,1.0,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.2857143,0.2857143,0.0,0.42857143,0.5714286,0.42857143,0.2857143,0.5714286,,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.85714287,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.42857143,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.5714286,0.42857143,0.2857143,,0.5714286,0.42857143,0.5714286,0.5714286,0.42857143,0.5714286,0.71428573,0.71428573,0.5714286,0.42857143,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.5714286,0.5714286,1.0,0.71428573,0.85714287,0.85714287,1.0,0.85714287,0.71428573,1.0,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.85714287,1.0,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.14285715,0.14285715,0.42857143,0.14285715,0.14285715,0.2857143,0.5714286,0.2857143,0.42857143,,0.5714286,0.5714286,0.42857143,0.71428573,0.71428573,0.5714286,0.5714286,0.71428573,0.42857143,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.5714286,0.85714287,0.85714287,0.71428573,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.2857143,0.0,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.42857143,0.42857143,0.5714286,0.42857143,,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,0.42857143,0.5714286,0.71428573,0.42857143,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.5714286,0.71428573,1.0,0.71428573,0.85714287,0.71428573,0.85714287,1.0,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.5714286,0.71428573,0.85714287,0.85714287,1.0,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.2857143,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,,0.5714286,0.71428573,0.71428573,0.42857143,0.42857143,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.42857143,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.2857143,0.0,0.14285715,0.2857143,0.14285715,0.14285715,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.5714286,0.42857143,,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.42857143,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,1.0,0.85714287,1.0,0.71428573,0.85714287,0.85714287,1.0,0.71428573,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.2857143,0.5714286,0.2857143,0.42857143,,0.42857143,0.5714286,0.5714286,0.42857143,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.5714286,0.85714287,0.71428573,0.5714286,0.5714286,0.85714287,0.5714286,1.0,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,1.0,1.0,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.5714286,,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.85714287,0.71428573,0.5714286,0.85714287,0.71428573,0.5714286,0.5714286,0.85714287,0.5714286,1.0,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.85714287,1.0,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,,0.42857143,0.42857143,0.42857143,0.42857143,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.85714287,0.42857143,0.5714286,0.5714286,0.42857143,0.5714286,0.85714287,0.85714287,0.85714287,0.5714286,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,1.0,1.0,0.85714287,1.0,0.85714287,0.71428573,0.71428573,1.0,0.85714287,1.0,0.71428573,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.5714286,,0.42857143,0.42857143,0.42857143,0.71428573,0.42857143,0.71428573,0.5714286,0.5714286,0.85714287,0.42857143,0.5714286,0.5714286,0.42857143,0.5714286,0.85714287,0.85714287,0.85714287,0.5714286,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,1.0,1.0,0.85714287,1.0,0.85714287,0.71428573,0.71428573,1.0,0.85714287,0.85714287,0.71428573,1.0,0.85714287,0.85714287,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,0.42857143,0.42857143,0.5714286,0.42857143,0.5714286,0.5714286,,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.42857143,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.85714287,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,1.0,0.71428573,0.85714287,1.0,1.0,1.0,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.5714286,0.5714286,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.5714286,0.5714286,0.42857143,,0.71428573,0.42857143,0.71428573,0.42857143,0.85714287,0.85714287,0.71428573,0.71428573,0.42857143,0.5714286,0.71428573,0.5714286,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.5714286,0.5714286,0.42857143,0.2857143,,0.42857143,0.5714286,0.71428573,0.42857143,0.71428573,0.5714286,0.71428573,0.42857143,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,1.0,0.71428573,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.14285715,0.42857143,0.14285715,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.5714286,0.5714286,,0.5714286,0.42857143,0.5714286,0.71428573,0.42857143,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,1.0,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,1.0,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.5714286,0.42857143,0.2857143,0.42857143,0.42857143,,0.5714286,0.42857143,0.5714286,0.5714286,0.71428573,0.42857143,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,1.0,0.71428573,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.5714286,0.2857143,0.5714286,0.42857143,,0.5714286,0.5714286,0.71428573,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.5714286,0.71428573,0.85714287,0.5714286,0.85714287,0.71428573,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.14285715,0.5714286,0.42857143,0.5714286,0.42857143,,0.5714286,0.5714286,0.5714286,0.42857143,0.42857143,0.71428573,0.42857143,0.71428573,0.71428573,0.5714286,0.71428573,0.42857143,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,,0.42857143,0.42857143,0.42857143,0.71428573,0.71428573,0.71428573,0.5714286,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.85714287,0.85714287,0.5714286,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.14285715,0.14285715,0.14285715,0.14285715,0.42857143,0.2857143,0.42857143,0.5714286,0.42857143,0.2857143,0.42857143,0.5714286,,0.42857143,0.42857143,0.5714286,0.5714286,0.42857143,0.5714286,0.85714287,0.71428573,0.71428573,0.5714286,0.5714286,0.42857143,0.5714286,1.0,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.14285715,0.42857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.5714286,0.5714286,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.5714286,,0.42857143,0.71428573,0.64285713,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.2857143,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,,0.5714286,0.42857143,0.5714286,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,1.0
+xRFM (tuned),0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,0.42857143,0.2857143,0.42857143,,0.2857143,0.71428573,0.42857143,0.42857143,0.42857143,0.5714286,0.85714287,0.42857143,0.85714287,0.5714286,0.42857143,0.42857143,0.5714286,0.71428573,0.85714287,0.5714286,0.71428573,0.5714286,0.85714287,0.71428573,1.0,0.5714286,0.71428573,0.71428573,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,0.71428573,0.71428573,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.14285715,0.0,0.14285715,0.0,0.0,0.2857143,0.14285715,0.14285715,0.2857143,0.42857143,0.2857143,0.0,0.2857143,0.2857143,0.14285715,0.14285715,0.5714286,0.5714286,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.2857143,0.42857143,0.35714287,0.5714286,0.71428573,,0.71428573,0.42857143,0.71428573,0.42857143,0.71428573,0.71428573,0.5714286,0.71428573,0.5714286,0.71428573,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,0.5714286,0.2857143,0.42857143,0.2857143,0.2857143,,0.42857143,0.5714286,0.42857143,0.42857143,0.85714287,0.42857143,0.78571427,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,0.71428573,0.5714286,1.0,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.71428573,0.71428573,0.71428573,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0
+EBM (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.0,0.0,0.14285715,0.14285715,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,,0.71428573,0.42857143,0.71428573,0.5714286,0.5714286,0.5714286,0.42857143,0.85714287,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.71428573,0.85714287,0.71428573,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.5714286,0.14285715,0.42857143,0.42857143,0.5714286,0.2857143,0.42857143,0.2857143,,0.5714286,0.71428573,0.5714286,0.5714286,0.42857143,0.5714286,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.5714286,0.5714286,0.5714286,0.5714286,0.42857143,,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.85714287,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,1.0,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.2857143,0.42857143,0.42857143,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.14285715,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,0.2857143,0.2857143,0.5714286,,0.5714286,0.71428573,0.71428573,0.5714286,0.42857143,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,1.0,1.0,0.71428573,0.71428573,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.2857143,0.42857143,0.2857143,0.42857143,0.14285715,0.2857143,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.85714287,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.71428573,0.5714286,1.0,0.85714287,0.85714287,0.85714287,1.0,0.71428573,1.0,1.0
+FastaiMLP (tuned + ensembled),0.14285715,0.0,0.14285715,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,,0.5714286,0.5714286,0.42857143,0.42857143,0.5714286,0.42857143,0.5714286,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,1.0,0.85714287,0.85714287,0.71428573,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0
+ExtraTrees (tuned),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.42857143,0.2857143,0.2857143,0.5714286,0.2857143,0.42857143,0.14285715,0.2857143,0.21428572,0.42857143,0.5714286,0.42857143,0.2857143,0.42857143,0.42857143,,0.71428573,0.5714286,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.42857143,0.5714286,0.5714286,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.5714286,0.71428573,0.5714286,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0
+LightGBM (default),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.2857143,,0.71428573,0.71428573,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.5714286,0.85714287,0.71428573,0.71428573,1.0,0.71428573,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.0,0.0,0.14285715,0.14285715,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.2857143,0.0,0.42857143,0.2857143,0.5714286,0.2857143,0.42857143,0.14285715,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.42857143,0.2857143,,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.42857143,0.2857143,0.2857143,0.5714286,0.5714286,0.42857143,0.42857143,0.42857143,0.2857143,0.2857143,0.5714286,0.5714286,0.5714286,0.2857143,0.5714286,,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.42857143,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.14285715,0.42857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.14285715,0.42857143,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,,0.5714286,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,0.85714287,1.0
+TabDPT (default),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.42857143,0.14285715,0.42857143,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,,0.2857143,0.2857143,0.42857143,0.5714286,0.42857143,0.5714286,0.71428573,0.71428573,0.5714286,0.5714286,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,1.0,0.42857143,0.71428573,0.5714286,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0
+ChimeraBoost (tuned),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.0,0.0,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.14285715,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.42857143,0.5714286,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.5714286,0.71428573,0.42857143,,0.5714286,0.42857143,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.5714286,0.5714286,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,1.0,1.0,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0
+TabSwift (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.0,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.42857143,0.42857143,,0.42857143,0.42857143,0.71428573,0.42857143,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0
+XGBoost (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.2857143,0.14285715,0.0,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.42857143,0.14285715,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.42857143,0.42857143,0.5714286,0.5714286,,0.5714286,0.5714286,0.42857143,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.14285715,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.2857143,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.5714286,0.2857143,0.2857143,0.5714286,0.42857143,,0.5714286,0.5714286,0.5714286,0.5714286,0.2857143,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.71428573,1.0,1.0
+RealMLP (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.2857143,0.14285715,0.0,0.0,0.0,0.0,0.0,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,,0.5714286,0.42857143,0.42857143,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,1.0
+xRFM (default),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.0,0.14285715,0.42857143,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.0,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.5714286,0.5714286,0.42857143,0.42857143,,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,1.0,0.85714287,1.0,0.85714287,1.0
+FastaiMLP (tuned),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.0,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.5714286,,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.71428573,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0
+ChimeraBoost (default),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.42857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.5714286,0.42857143,0.42857143,,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.5714286,0.42857143,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.0,0.2857143,0.0,0.14285715,0.2857143,0.0,0.14285715,0.0,0.2857143,0.2857143,0.0,0.42857143,0.14285715,0.2857143,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.5714286,0.42857143,0.42857143,0.2857143,0.71428573,0.42857143,0.42857143,0.2857143,0.42857143,,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.85714287,0.5714286,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,0.5714286,0.85714287,0.85714287,1.0,1.0
+KNN (tuned + ensembled),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.0,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.0,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,1.0,0.85714287,0.5714286,0.42857143,0.85714287,0.71428573,0.85714287,1.0
+TorchMLP (default),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.2857143,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.42857143,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.42857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.5714286,,0.71428573,0.71428573,0.5714286,0.42857143,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0
+TabSTAR (tuned),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.0,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.5714286,0.2857143,,0.35714287,0.2857143,0.42857143,0.2857143,0.42857143,0.71428573,0.5714286,0.71428573,0.85714287,0.71428573,0.5714286,1.0,0.71428573,0.85714287,0.71428573,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.5714286,0.2857143,0.64285713,,0.42857143,0.42857143,0.42857143,0.42857143,0.71428573,0.42857143,0.71428573,0.85714287,0.71428573,0.5714286,1.0,0.71428573,0.85714287,0.71428573,1.0
+OrionMSP (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.14285715,0.0,0.42857143,0.0,0.0,0.0,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.0,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.5714286,0.42857143,0.71428573,0.5714286,,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.14285715,0.0,0.14285715,0.14285715,0.0,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.5714286,0.14285715,0.5714286,0.5714286,0.5714286,0.5714286,0.2857143,,0.42857143,0.5714286,0.71428573,0.42857143,0.71428573,0.85714287,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,1.0
+iLTM (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.14285715,0.0,0.0,0.14285715,0.2857143,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.2857143,0.14285715,0.14285715,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.42857143,0.5714286,0.5714286,0.71428573,0.5714286,0.42857143,0.5714286,,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.2857143,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.0,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.0,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.5714286,0.42857143,0.42857143,0.42857143,,0.42857143,0.71428573,0.42857143,0.85714287,0.71428573,0.5714286,0.71428573,0.85714287,0.71428573,0.85714287,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.2857143,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.42857143,0.2857143,0.14285715,0.14285715,0.2857143,0.5714286,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.5714286,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,,0.42857143,0.64285713,0.85714287,0.5714286,0.5714286,0.5714286,0.5714286,1.0,0.71428573,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.14285715,0.0,0.0,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.0,0.14285715,0.2857143,0.2857143,0.0,0.2857143,0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.0,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.2857143,0.5714286,,0.71428573,0.71428573,0.5714286,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.2857143,0.0,0.0,0.0,0.14285715,0.2857143,0.0,0.2857143,0.0,0.0,0.0,0.14285715,0.42857143,0.14285715,0.42857143,0.2857143,0.0,0.14285715,0.14285715,0.42857143,0.2857143,0.0,0.2857143,0.14285715,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.5714286,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.35714287,0.2857143,,0.85714287,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.71428573,1.0
+KNN (tuned),0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.0,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,,0.42857143,0.2857143,0.2857143,0.5714286,0.2857143,0.5714286,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,,0.5714286,0.2857143,0.85714287,0.5714286,1.0,0.85714287
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.2857143,0.14285715,0.14285715,0.14285715,0.42857143,0.14285715,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.71428573,0.42857143,,0.71428573,0.42857143,0.71428573,0.5714286,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.42857143,0.5714286,0.14285715,0.0,0.0,0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.14285715,0.42857143,0.71428573,0.71428573,0.2857143,,0.71428573,0.71428573,0.71428573,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,0.14285715,0.5714286,0.2857143,,0.5714286,0.64285713,0.85714287
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.2857143,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.2857143,0.0,0.0,0.0,0.0,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.0,0.14285715,0.14285715,0.71428573,0.42857143,0.2857143,0.2857143,0.42857143,,0.5714286,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.0,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.0,0.42857143,0.2857143,0.35714287,0.42857143,,0.71428573
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.2857143,
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_all/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_all/leaderboard_overview_explorer.html
index e7f97397a5beaf98be6cf8cf3588c159a9f7eb56..49ad509680c3422cab20be7945a7f303b96a4875 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Regression | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_all/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_regression/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..14cd2b5a9abbc96fb7196994943262b0a32e261c
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_regression/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2229.0,"score":0.901,"rank":2.69,"hrank":1.54,"imp":1.407,"train_time":40.79,"predict_time":6.492,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+255\/-66","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1945.0,"score":0.687,"rank":7.62,"hrank":4.28,"imp":3.335,"train_time":339.37,"predict_time":3.244,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+126\/-47","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1937.0,"score":0.693,"rank":7.85,"hrank":3.13,"imp":2.269,"train_time":4.02,"predict_time":0.398,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+292\/-150","imputed":false},{"position":3,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1923.0,"score":0.673,"rank":8.23,"hrank":3.94,"imp":3.06,"train_time":11.81,"predict_time":2.007,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+188\/-101","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1902.0,"score":0.639,"rank":8.85,"hrank":5.63,"imp":2.722,"train_time":2043.54,"predict_time":8.366,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+232\/-147","imputed":false},{"position":5,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1887.0,"score":0.623,"rank":9.31,"hrank":5.61,"imp":3.39,"train_time":8.44,"predict_time":0.7,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+183\/-72","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1831.0,"score":0.59,"rank":11.15,"hrank":4.21,"imp":4.004,"train_time":7126.67,"predict_time":336.907,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+255\/-106","imputed":false},{"position":7,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1823.0,"score":0.593,"rank":11.46,"hrank":5.99,"imp":4.114,"train_time":6.86,"predict_time":1.346,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+202\/-112","imputed":false},{"position":8,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1776.0,"score":0.546,"rank":13.23,"hrank":8.66,"imp":4.506,"train_time":7126.67,"predict_time":38.573,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+193\/-89","imputed":false},{"position":9,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1776.0,"score":0.539,"rank":13.23,"hrank":7.5,"imp":3.651,"train_time":2043.54,"predict_time":0.78,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+193\/-145","imputed":false},{"position":10,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1742.0,"score":0.582,"rank":14.62,"hrank":4.7,"imp":4.372,"train_time":1.32,"predict_time":0.122,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+316\/-164","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1707.0,"score":0.443,"rank":16.15,"hrank":14.39,"imp":5.694,"train_time":4057.31,"predict_time":10.474,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-61","imputed":false},{"position":12,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1705.0,"score":0.413,"rank":16.23,"hrank":11.55,"imp":6.221,"train_time":1490.17,"predict_time":5.79,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+109\/-93","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1672.0,"score":0.443,"rank":17.77,"hrank":13.99,"imp":5.429,"train_time":2.02,"predict_time":0.183,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+162\/-94","imputed":false},{"position":14,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1659.0,"score":0.433,"rank":18.38,"hrank":11.31,"imp":5.105,"train_time":6.94,"predict_time":0.519,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+205\/-168","imputed":false},{"position":15,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1654.0,"score":0.432,"rank":18.62,"hrank":11.61,"imp":5.656,"train_time":44.65,"predict_time":38.297,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+207\/-98","imputed":false},{"position":16,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1621.0,"score":0.395,"rank":20.23,"hrank":7.17,"imp":8.163,"train_time":3811.61,"predict_time":7.581,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+185\/-180","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1570.0,"score":0.234,"rank":22.85,"hrank":19.71,"imp":8.792,"train_time":686.5,"predict_time":5.477,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-85","imputed":false},{"position":18,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1564.0,"score":0.332,"rank":23.15,"hrank":18.06,"imp":6.839,"train_time":4057.31,"predict_time":0.337,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-148","imputed":false},{"position":19,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1533.0,"score":0.19,"rank":24.77,"hrank":18.44,"imp":8.535,"train_time":2267.19,"predict_time":0.422,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-83","imputed":false},{"position":20,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1519.0,"score":0.192,"rank":25.54,"hrank":17.84,"imp":8.569,"train_time":2267.19,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-79","imputed":false},{"position":21,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1518.0,"score":0.269,"rank":25.62,"hrank":21.51,"imp":7.963,"train_time":708.95,"predict_time":1.973,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-100","imputed":false},{"position":22,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1518.0,"score":0.265,"rank":25.62,"hrank":21.13,"imp":7.163,"train_time":3448.73,"predict_time":2.111,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-96","imputed":false},{"position":23,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1504.0,"score":0.207,"rank":26.38,"hrank":23.57,"imp":8.531,"train_time":12741.09,"predict_time":387.573,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-55","imputed":false},{"position":24,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1499.0,"score":0.298,"rank":26.69,"hrank":17.87,"imp":7.143,"train_time":0.9,"predict_time":0.072,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+194\/-131","imputed":false},{"position":25,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1471.0,"score":0.217,"rank":28.31,"hrank":23.24,"imp":7.54,"train_time":3448.73,"predict_time":0.194,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+142\/-104","imputed":false},{"position":26,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1471.0,"score":0.161,"rank":28.31,"hrank":23.72,"imp":9.578,"train_time":686.5,"predict_time":0.743,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-98","imputed":false},{"position":27,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1456.0,"score":0.187,"rank":29.15,"hrank":25.84,"imp":8.514,"train_time":708.95,"predict_time":0.069,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-70","imputed":false},{"position":28,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1429.0,"score":0.111,"rank":30.77,"hrank":29.1,"imp":9.492,"train_time":848.99,"predict_time":0.475,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-61","imputed":false},{"position":29,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1429.0,"score":0.109,"rank":30.77,"hrank":29.7,"imp":9.51,"train_time":848.99,"predict_time":2.375,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-55","imputed":false},{"position":30,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1411.0,"score":0.153,"rank":31.85,"hrank":28.8,"imp":10.782,"train_time":16.07,"predict_time":0.289,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-88","imputed":false},{"position":31,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1404.0,"score":0.107,"rank":32.31,"hrank":26.78,"imp":10.663,"train_time":9.02,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-109","imputed":false},{"position":32,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1390.0,"score":0.193,"rank":33.15,"hrank":23.52,"imp":10.655,"train_time":3811.61,"predict_time":0.452,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+141\/-173","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1383.0,"score":0.108,"rank":33.54,"hrank":28.2,"imp":11.959,"train_time":402.93,"predict_time":0.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+139\/-120","imputed":false},{"position":34,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1348.0,"score":0.099,"rank":35.62,"hrank":32.77,"imp":9.828,"train_time":12741.09,"predict_time":60.101,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+116\/-87","imputed":false},{"position":35,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1329.0,"score":0.173,"rank":36.77,"hrank":16.49,"imp":10.388,"train_time":59.38,"predict_time":14.073,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+217\/-201","imputed":false},{"position":36,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1322.0,"score":0.081,"rank":37.15,"hrank":33.26,"imp":12.342,"train_time":402.93,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-125","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1291.0,"score":0.07,"rank":39.0,"hrank":36.27,"imp":9.907,"train_time":9.79,"predict_time":0.134,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-85","imputed":false},{"position":38,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1291.0,"score":0.145,"rank":39.0,"hrank":29.01,"imp":11.044,"train_time":18.77,"predict_time":6.542,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+180\/-156","imputed":false},{"position":39,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1259.0,"score":0.075,"rank":40.92,"hrank":36.62,"imp":11.48,"train_time":9.18,"predict_time":1.415,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-98","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1259.0,"score":0.067,"rank":40.92,"hrank":37.99,"imp":10.851,"train_time":4452.11,"predict_time":0.852,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-100","imputed":false},{"position":41,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1255.0,"score":0.063,"rank":41.15,"hrank":36.65,"imp":13.342,"train_time":257.64,"predict_time":0.813,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-108","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1232.0,"score":0.041,"rank":42.54,"hrank":39.35,"imp":13.517,"train_time":2.46,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-123","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1224.0,"score":0.051,"rank":43.0,"hrank":39.07,"imp":13.483,"train_time":257.64,"predict_time":0.06,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-101","imputed":false},{"position":44,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1212.0,"score":0.057,"rank":43.69,"hrank":39.94,"imp":11.465,"train_time":4452.11,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-126","imputed":false},{"position":45,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1197.0,"score":0.063,"rank":44.62,"hrank":35.02,"imp":14.293,"train_time":2.47,"predict_time":0.755,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-161","imputed":false},{"position":46,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1197.0,"score":0.009,"rank":44.62,"hrank":43.38,"imp":12.343,"train_time":2.03,"predict_time":0.296,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-94","imputed":false},{"position":47,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1188.0,"score":0.075,"rank":45.15,"hrank":32.99,"imp":14.844,"train_time":2006.55,"predict_time":0.103,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+138\/-193","imputed":false},{"position":48,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1167.0,"score":0.014,"rank":46.35,"hrank":44.27,"imp":14.342,"train_time":177.92,"predict_time":0.433,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-114","imputed":false},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1162.0,"score":0.023,"rank":46.62,"hrank":43.09,"imp":13.048,"train_time":2.15,"predict_time":0.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+107\/-123","imputed":false},{"position":50,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1152.0,"score":0.011,"rank":47.15,"hrank":45.31,"imp":14.614,"train_time":526.19,"predict_time":0.77,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-102","imputed":false},{"position":51,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1136.0,"score":0.077,"rank":48.08,"hrank":33.32,"imp":15.396,"train_time":2006.55,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-202","imputed":false},{"position":52,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1124.0,"score":0.0,"rank":48.73,"hrank":47.54,"imp":15.075,"train_time":177.92,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-111","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1115.0,"score":0.012,"rank":49.19,"hrank":47.15,"imp":15.349,"train_time":526.19,"predict_time":0.119,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-129","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1086.0,"score":0.011,"rank":50.69,"hrank":47.18,"imp":15.856,"train_time":2.57,"predict_time":0.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-131","imputed":false},{"position":55,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1052.0,"score":0.0,"rank":52.38,"hrank":51.19,"imp":15.657,"train_time":527.3,"predict_time":2.83,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-104","imputed":false},{"position":56,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1027.0,"score":0.019,"rank":53.62,"hrank":50.26,"imp":16.48,"train_time":7.59,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-160","imputed":false},{"position":57,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1019.0,"score":0.0,"rank":54.0,"hrank":51.96,"imp":15.502,"train_time":20.5,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-127","imputed":false},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1017.0,"score":0.0,"rank":54.08,"hrank":52.67,"imp":16.134,"train_time":527.3,"predict_time":0.311,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-105","imputed":false},{"position":59,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":54.88,"hrank":54.24,"imp":16.648,"train_time":0.63,"predict_time":0.064,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-99","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":986.0,"score":0.067,"rank":55.54,"hrank":35.92,"imp":24.113,"train_time":46109.53,"predict_time":22.149,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+231\/-332","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":953.0,"score":0.0,"rank":57.0,"hrank":55.64,"imp":18.294,"train_time":22.8,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-166","imputed":false},{"position":62,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":944.0,"score":0.054,"rank":57.38,"hrank":43.66,"imp":24.294,"train_time":46109.53,"predict_time":5.06,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+232\/-328","imputed":false},{"position":63,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":887.0,"score":0.0,"rank":59.62,"hrank":58.66,"imp":20.145,"train_time":3.08,"predict_time":0.293,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-147","imputed":false},{"position":64,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":868.0,"score":0.0,"rank":60.31,"hrank":59.77,"imp":19.248,"train_time":373.42,"predict_time":81.49,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-137","imputed":false},{"position":65,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":866.0,"score":0.0,"rank":60.38,"hrank":59.36,"imp":21.535,"train_time":90.87,"predict_time":0.906,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-162","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":797.0,"score":0.027,"rank":62.62,"hrank":55.99,"imp":27.43,"train_time":302.9,"predict_time":4.583,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+254\/-333","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":723.0,"score":0.0,"rank":64.62,"hrank":64.29,"imp":23.655,"train_time":90.87,"predict_time":0.055,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-155","imputed":false},{"position":68,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":626.0,"score":0.0,"rank":66.69,"hrank":66.06,"imp":30.241,"train_time":0.25,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+149\/-251","imputed":false},{"position":69,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":476.0,"score":0.0,"rank":68.96,"hrank":68.81,"imp":37.219,"train_time":177.06,"predict_time":0.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-354","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":416.0,"score":0.0,"rank":69.65,"hrank":69.49,"imp":37.401,"train_time":177.06,"predict_time":0.059,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-486","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":277.0,"score":0.0,"rank":70.92,"hrank":70.9,"imp":39.828,"train_time":0.88,"predict_time":0.109,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+111\/-399","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_all/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_all/pareto_front_explorer.html
index 5cd61d99ee407aa3b460bfd7804b4f173b89d0b7..5803f37b7087c6bdeabee94fe4a55efec4aa5190 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0182719821,"imp":10.6629057031,"elo":1403.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.04636871,"imp":8.5689975774,"elo":1519.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"imp":8.5351365089,"elo":1533.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"imp":13.5165870219,"elo":1231.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0343070535,"imp":12.3418233313,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"imp":11.9586707247,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.01204424,"imp":16.479912159,"elo":1026.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0125056048,"imp":15.3960953256,"elo":1135.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1032251722,"imp":14.8437898138,"elo":1187.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2422366525,"imp":15.8559260551,"elo":1086.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0603961196,"imp":13.4828514032,"elo":1224.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8125804985,"imp":13.3421633346,"elo":1254.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2929314919,"imp":20.1446549368,"elo":887.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3110785788,"imp":16.1337653625,"elo":1017.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8297120981,"imp":15.6574941615,"elo":1052.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2962263441,"imp":12.3430360848,"elo":1196.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7428209859,"imp":9.577727366,"elo":1470.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":5.476648655,"imp":8.7915527894,"elo":1569.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":14.072768612,"imp":10.3876718017,"elo":1328.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2894827463,"imp":10.7820379041,"elo":1411.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4518951639,"imp":10.6548900789,"elo":1389.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5805507968,"imp":8.1631535544,"elo":1621.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.3455286154,"imp":4.1144151877,"elo":1822.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0067377719,"imp":3.0595728499,"elo":1923.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0171543088,"imp":18.2943297341,"elo":952.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1465391338,"imp":15.0751309127,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4333863798,"imp":14.3424516152,"elo":1166.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0638540688,"imp":16.6482084857,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.119307123,"imp":15.3486617095,"elo":1114.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7700427525,"imp":14.6140166256,"elo":1152.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.4152312914,"imp":11.4799030804,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3372766026,"imp":6.8392974584,"elo":1563.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":10.4744362064,"imp":5.6941083754,"elo":1707.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5191857229,"imp":5.1050024316,"elo":1659.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7797947208,"imp":3.6513451358,"elo":1775.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3660135567,"imp":2.7216385458,"elo":1902.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.5421299186,"imp":11.0439680079,"elo":1290.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"imp":5.6563845374,"elo":1654.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.5726172424,"imp":4.5064720405,"elo":1775.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":336.9071136327,"imp":4.0044876883,"elo":1831.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"imp":5.4286497419,"elo":1671.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4924694208,"imp":1.4067478497,"elo":2228.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1218380804,"imp":4.3723770177,"elo":1742.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1337247011,"imp":9.9071412654,"elo":1290.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1936377213,"imp":7.539858265,"elo":1470.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.1105459826,"imp":7.1633433427,"elo":1517.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6996427484,"imp":3.3904763149,"elo":1886.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3981148822,"imp":2.2689880747,"elo":1936.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5829823393,"imp":27.4295870304,"elo":796.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.0596311902,"imp":24.2937076463,"elo":943.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.148843297,"imp":24.1126590799,"elo":985.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719372146,"imp":7.1426209224,"elo":1498.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0813896645,"imp":15.5015821228,"elo":1018.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0885042776,"imp":11.4645270075,"elo":1212.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8515826242,"imp":10.8505791228,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1795378646,"imp":13.048270691,"elo":1162.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4748247737,"imp":9.4916701231,"elo":1429.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.3747946103,"imp":9.5098219062,"elo":1429.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":81.4898053701,"imp":19.2481247323,"elo":868.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":60.1008510867,"imp":9.8281010544,"elo":1348.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":387.572909971,"imp":8.5308610842,"elo":1504.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7548401788,"imp":14.2932711533,"elo":1196.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0686603653,"imp":8.5143347422,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.9734199806,"imp":7.9631079976,"elo":1517.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0182719821,"x_train":9.0226914704,"imp":10.6629057031,"elo":1403.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.04636871,"x_train":2267.1899830578,"imp":8.5689975774,"elo":1519.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"x_train":2267.1899830578,"imp":8.5351365089,"elo":1533.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"x_train":2.4559533883,"imp":13.5165870219,"elo":1231.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0343070535,"x_train":402.9323608119,"imp":12.3418233313,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":402.9323608119,"imp":11.9586707247,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.01204424,"x_train":7.5931963365,"imp":16.479912159,"elo":1026.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0125056048,"x_train":2006.5538981646,"imp":15.3960953256,"elo":1135.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1032251722,"x_train":2006.5538981646,"imp":14.8437898138,"elo":1187.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2422366525,"x_train":2.5746349456,"imp":15.8559260551,"elo":1086.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0603961196,"x_train":257.6385593165,"imp":13.4828514032,"elo":1224.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8125804985,"x_train":257.6385593165,"imp":13.3421633346,"elo":1254.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2929314919,"x_train":3.0840796949,"imp":20.1446549368,"elo":887.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3110785788,"x_train":527.2961316043,"imp":16.1337653625,"elo":1017.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8297120981,"x_train":527.2961316043,"imp":15.6574941615,"elo":1052.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2962263441,"x_train":2.0296176424,"imp":12.3430360848,"elo":1196.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7428209859,"x_train":686.4972156667,"imp":9.577727366,"elo":1470.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":5.476648655,"x_train":686.4972156667,"imp":8.7915527894,"elo":1569.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":14.072768612,"x_train":59.3758334402,"imp":10.3876718017,"elo":1328.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2894827463,"x_train":16.0749964455,"imp":10.7820379041,"elo":1411.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4518951639,"x_train":3811.6104621203,"imp":10.6548900789,"elo":1389.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5805507968,"x_train":3811.6104621203,"imp":8.1631535544,"elo":1621.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.3455286154,"x_train":6.8642025254,"imp":4.1144151877,"elo":1822.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0067377719,"x_train":11.8050270551,"imp":3.0595728499,"elo":1923.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0171543088,"x_train":22.8030707188,"imp":18.2943297341,"elo":952.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1465391338,"x_train":177.9195746923,"imp":15.0751309127,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4333863798,"x_train":177.9195746923,"imp":14.3424516152,"elo":1166.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0638540688,"x_train":0.6331018043,"imp":16.6482084857,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.119307123,"x_train":526.192235626,"imp":15.3486617095,"elo":1114.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7700427525,"x_train":526.192235626,"imp":14.6140166256,"elo":1152.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.4152312914,"x_train":9.1753045614,"imp":11.4799030804,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3372766026,"x_train":4057.3100555326,"imp":6.8392974584,"elo":1563.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":10.4744362064,"x_train":4057.3100555326,"imp":5.6941083754,"elo":1707.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5191857229,"x_train":6.9411546573,"imp":5.1050024316,"elo":1659.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7797947208,"x_train":2043.5362680496,"imp":3.6513451358,"elo":1775.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3660135567,"x_train":2043.5362680496,"imp":2.7216385458,"elo":1902.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.5421299186,"x_train":18.769283676,"imp":11.0439680079,"elo":1290.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"x_train":44.6485648211,"imp":5.6563845374,"elo":1654.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.5726172424,"x_train":7126.6747880976,"imp":4.5064720405,"elo":1775.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":336.9071136327,"x_train":7126.6747880976,"imp":4.0044876883,"elo":1831.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"x_train":2.0183513985,"imp":5.4286497419,"elo":1671.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4924694208,"x_train":40.7926562599,"imp":1.4067478497,"elo":2228.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1218380804,"x_train":1.3151215973,"imp":4.3723770177,"elo":1742.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1337247011,"x_train":9.7873678423,"imp":9.9071412654,"elo":1290.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1936377213,"x_train":3448.7325920652,"imp":7.539858265,"elo":1470.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.1105459826,"x_train":3448.7325920652,"imp":7.1633433427,"elo":1517.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6996427484,"x_train":8.4368371839,"imp":3.3904763149,"elo":1886.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3981148822,"x_train":4.0220384031,"imp":2.2689880747,"elo":1936.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5829823393,"x_train":302.8991048646,"imp":27.4295870304,"elo":796.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.0596311902,"x_train":46109.5334192428,"imp":24.2937076463,"elo":943.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.148843297,"x_train":46109.5334192428,"imp":24.1126590799,"elo":985.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719372146,"x_train":0.8979740772,"imp":7.1426209224,"elo":1498.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0813896645,"x_train":20.5017216002,"imp":15.5015821228,"elo":1018.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0885042776,"x_train":4452.1085706374,"imp":11.4645270075,"elo":1212.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8515826242,"x_train":4452.1085706374,"imp":10.8505791228,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1795378646,"x_train":2.1459956511,"imp":13.048270691,"elo":1162.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4748247737,"x_train":848.9941919866,"imp":9.4916701231,"elo":1429.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.3747946103,"x_train":848.9941919866,"imp":9.5098219062,"elo":1429.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":81.4898053701,"x_train":373.4221052753,"imp":19.2481247323,"elo":868.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":60.1008510867,"x_train":12741.0939157936,"imp":9.8281010544,"elo":1348.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":387.572909971,"x_train":12741.0939157936,"imp":8.5308610842,"elo":1504.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7548401788,"x_train":2.469327894,"imp":14.2932711533,"elo":1196.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0686603653,"x_train":708.9450090283,"imp":8.5143347422,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.9734199806,"x_train":708.9450090283,"imp":7.9631079976,"elo":1517.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_all/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_all/tuning_trajectories_explorer.html
index 508bca494cf66a17333598485bbf8fff5812705c..2ee96026dc3669df2d9628ead183e0cea07718c4 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":9.0226914704,"imp":10.6649854115,"elo":1406.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":22.006737996,"imp":10.0410335189,"elo":1454.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":54.5936853194,"imp":9.3422932439,"elo":1523.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":108.5250024926,"imp":9.0363696279,"elo":1538.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":277.687397929,"imp":8.7304375075,"elo":1551.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":564.9102272658,"imp":8.6385669356,"elo":1552.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1124.7408503677,"imp":8.5860979377,"elo":1552.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1677.2797709559,"imp":8.5542513657,"elo":1557.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":2267.1899830578,"imp":8.537762157,"elo":1558.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.4559533883,"imp":13.5184798184,"elo":1238.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.0653056289,"imp":13.1595745141,"elo":1259.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":9.725445997,"imp":12.8184880099,"elo":1304.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":19.7798589835,"imp":12.6146892497,"elo":1327.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":50.0934628362,"imp":12.3717353184,"elo":1346.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":100.1940386909,"imp":12.2119545046,"elo":1367.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":200.5360400335,"imp":12.0503277586,"elo":1382.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":300.5068144065,"imp":11.984244047,"elo":1393.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":402.9323608119,"imp":11.9597608126,"elo":1401.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.5931963365,"imp":16.4815170931,"elo":1039.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":19.5464706545,"imp":15.9818225032,"elo":1079.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":52.461846709,"imp":15.5322088782,"elo":1138.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":101.1927167213,"imp":15.3095708687,"elo":1154.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":250.7607078715,"imp":15.0786611831,"elo":1175.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":505.4611049176,"imp":15.0054156631,"elo":1183.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1000.4774432421,"imp":14.9259310452,"elo":1194.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1490.4477925659,"imp":14.8788837948,"elo":1198.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2006.5538981646,"imp":14.8444889121,"elo":1204.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.5746349456,"imp":15.8576388612,"elo":1085.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1741914103,"imp":15.312547953,"elo":1122.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.7533015627,"imp":14.5183742413,"elo":1190.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":13.1877719851,"imp":14.1351179951,"elo":1216.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":32.4660623442,"imp":13.547469136,"elo":1247.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":64.6497454444,"imp":13.4253490814,"elo":1260.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":128.5728546231,"imp":13.3545124869,"elo":1269.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":192.4203093077,"imp":13.3423915858,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":257.6385593165,"imp":13.3444260284,"elo":1268.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.0840796949,"imp":20.1459917158,"elo":871.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.6240636486,"imp":19.8504258159,"elo":891.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":13.6732407308,"imp":18.9476612196,"elo":935.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":27.2107956399,"imp":18.4865676439,"elo":950.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":65.2537322804,"imp":17.520082439,"elo":990.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":130.574483839,"imp":16.9310727816,"elo":1014.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":262.2064329926,"imp":16.338156479,"elo":1035.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":393.3022867336,"imp":15.93147183,"elo":1057.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":527.2961316043,"imp":15.6594635682,"elo":1071.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.0296176424,"imp":12.3451040358,"elo":1201.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":5.8086693996,"imp":11.4997719236,"elo":1271.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":16.4346292254,"imp":10.4066250539,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":32.0193358834,"imp":9.8788450456,"elo":1451.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":82.7850514402,"imp":9.3381159903,"elo":1529.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":169.4629431685,"imp":9.1757274338,"elo":1560.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":340.4054795841,"imp":8.9869444807,"elo":1584.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":509.4885196025,"imp":8.8864521179,"elo":1597.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":686.4972156667,"imp":8.7938407814,"elo":1614.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":59.3758334402,"imp":10.3904487974,"elo":1352.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":16.0749964455,"imp":10.7841513848,"elo":1436.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":38.647009759,"imp":10.2023184034,"elo":1470.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":106.244026897,"imp":9.4160045764,"elo":1530.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":199.3707655728,"imp":9.0269539703,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":462.7480963285,"imp":8.6556576511,"elo":1617.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":955.9940479091,"imp":8.4515123264,"elo":1624.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1870.556153527,"imp":8.3000951981,"elo":1630.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":2790.7899911449,"imp":8.1946214692,"elo":1643.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":3811.6104621203,"imp":8.1654997251,"elo":1645.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":6.8642025254,"imp":4.1176969928,"elo":1850.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":11.8050270551,"imp":3.0630923517,"elo":1959.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":22.8030707188,"imp":18.2955753566,"elo":969.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":53.0597722616,"imp":16.107246993,"elo":1070.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":147.0385313693,"imp":14.3675168325,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":177.9195746923,"imp":14.3437457459,"elo":1165.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.6331018043,"imp":16.6499447959,"elo":1001.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.5138045858,"imp":15.8760735961,"elo":1055.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":12.8062748979,"imp":15.3908176771,"elo":1095.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":26.5826232412,"imp":14.9942764132,"elo":1123.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":64.673699967,"imp":14.7770918641,"elo":1142.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":130.0298950791,"imp":14.6901072863,"elo":1152.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":259.9521641476,"imp":14.638686862,"elo":1162.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":391.5234331877,"imp":14.6211649426,"elo":1167.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":526.192235626,"imp":14.6157763538,"elo":1168.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":9.1753045614,"imp":11.4826162385,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":23.6721342187,"imp":9.3869073322,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":78.6133073597,"imp":7.3521875932,"elo":1558.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":191.0765941914,"imp":6.6293265441,"elo":1610.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":503.1010684492,"imp":6.1625582504,"elo":1666.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1009.2362249093,"imp":5.9055348463,"elo":1707.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2015.6576136475,"imp":5.7842906504,"elo":1728.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3026.2086158211,"imp":5.7349512293,"elo":1742.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4057.3100555326,"imp":5.6951806629,"elo":1750.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.9411546573,"imp":5.1081297102,"elo":1679.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":17.2836445024,"imp":4.6374541046,"elo":1706.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":47.5797720049,"imp":3.8614140786,"elo":1759.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":98.3276706028,"imp":3.4224976388,"elo":1822.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":249.0037143861,"imp":2.9685684337,"elo":1889.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":503.1695046416,"imp":2.7947420106,"elo":1913.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1018.7089146459,"imp":2.7018005663,"elo":1946.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1526.668944465,"imp":2.7311655156,"elo":1948.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2043.5362680496,"imp":2.7245619429,"elo":1953.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":18.769283676,"imp":11.0464917637,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":44.6485648211,"imp":5.6593617074,"elo":1685.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":88.2308458205,"imp":5.2454696275,"elo":1715.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":186.0445098453,"imp":4.8490533938,"elo":1740.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":341.1376387803,"imp":4.5346498664,"elo":1777.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":866.4570381261,"imp":4.2812155017,"elo":1811.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1760.5896780945,"imp":4.1783339382,"elo":1826.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3528.4122296034,"imp":4.0803258048,"elo":1844.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":5308.8782679903,"imp":4.0274723245,"elo":1859.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":7126.6747880976,"imp":4.0079622689,"elo":1867.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0183513985,"imp":5.431606219,"elo":1709.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":40.7926562599,"imp":1.4100424178,"elo":2256.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.3151215973,"imp":4.3753875956,"elo":1766.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.7873678423,"imp":9.9099485615,"elo":1299.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":23.4153138909,"imp":9.3565549555,"elo":1355.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":71.5165108752,"imp":8.7147560503,"elo":1401.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":166.1346310591,"imp":8.182159933,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":413.5775428251,"imp":7.7342280619,"elo":1493.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":844.0698066634,"imp":7.4331475795,"elo":1526.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1693.1667185284,"imp":7.2469532672,"elo":1546.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2557.2442472516,"imp":7.2041346491,"elo":1550.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3448.7325920652,"imp":7.1664648695,"elo":1549.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":8.4368371839,"imp":3.393775576,"elo":1930.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":4.0220384031,"imp":2.2837784913,"elo":1953.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":302.8991048646,"imp":27.4310818967,"elo":798.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":632.7525144671,"imp":27.2879140579,"elo":819.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1217.0384701627,"imp":26.9251443516,"elo":853.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2404.0220528997,"imp":26.2981860908,"elo":891.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5871.8477809815,"imp":25.4287396653,"elo":934.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11843.1243713497,"imp":24.7714334055,"elo":969.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23463.5184563875,"imp":24.4196340818,"elo":988.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34621.1415346513,"imp":24.2437727457,"elo":1001.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":46109.5334192428,"imp":24.1141000046,"elo":1011.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.8979740772,"imp":7.1457344991,"elo":1521.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":20.5017216002,"imp":15.5036655668,"elo":1029.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":53.7125230803,"imp":14.8522098568,"elo":1059.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":118.8503302159,"imp":13.6912428585,"elo":1129.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":237.0519073624,"imp":13.0974908775,"elo":1164.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":547.271171747,"imp":12.2733894118,"elo":1202.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1129.5925788802,"imp":11.7896668144,"elo":1233.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2199.7106433499,"imp":11.2664058306,"elo":1253.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3318.9517565565,"imp":11.0270650726,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4452.1085706374,"imp":10.8530281343,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.1459956511,"imp":13.0504189304,"elo":1163.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":8.3667450823,"imp":11.2771958175,"elo":1288.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":20.7207184201,"imp":10.5345260026,"elo":1358.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":41.7239328404,"imp":10.1437778435,"elo":1398.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":105.2854855499,"imp":9.8416341087,"elo":1428.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":210.1253434414,"imp":9.6989215232,"elo":1442.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":421.9300367907,"imp":9.6014900362,"elo":1453.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":635.5350385867,"imp":9.5372274664,"elo":1455.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":848.9941919866,"imp":9.5122528052,"elo":1454.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":373.4221052753,"imp":19.2497229645,"elo":879.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":839.9616197382,"imp":14.9056875458,"elo":1072.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2344.8516752833,"imp":16.2691579543,"elo":1269.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4981.7847745692,"imp":14.5448573032,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12170.8681485011,"imp":8.5512134328,"elo":1528.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12741.0939157936,"imp":8.5336049305,"elo":1531.1,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.469327894,"imp":14.2952393944,"elo":1215.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":6.0503740942,"imp":12.9135512688,"elo":1272.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.3170071504,"imp":11.1117919108,"elo":1350.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":34.6552917016,"imp":9.8076974115,"elo":1416.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.3965998327,"imp":8.6905248694,"elo":1479.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":176.967409496,"imp":8.3533128978,"elo":1515.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":353.1777869696,"imp":8.1041203633,"elo":1529.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":528.7105932474,"imp":7.99224468,"elo":1542.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":708.9450090283,"imp":7.9659724924,"elo":1548.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0182719821,"x_train":9.0226914704,"imp":10.6649854115,"elo":1406.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0761788183,"x_train":22.006737996,"imp":10.0410335189,"elo":1454.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1700551281,"x_train":54.5936853194,"imp":9.3422932439,"elo":1523.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2026453565,"x_train":108.5250024926,"imp":9.0363696279,"elo":1538.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2388904639,"x_train":277.687397929,"imp":8.7304375075,"elo":1551.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3111317954,"x_train":564.9102272658,"imp":8.6385669356,"elo":1552.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.398317095,"x_train":1124.7408503677,"imp":8.5860979377,"elo":1552.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3902829889,"x_train":1677.2797709559,"imp":8.5542513657,"elo":1557.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"x_train":2267.1899830578,"imp":8.537762157,"elo":1558.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"x_train":2.4559533883,"imp":13.5184798184,"elo":1238.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0371787849,"x_train":4.0653056289,"imp":13.1595745141,"elo":1259.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0917331108,"x_train":9.725445997,"imp":12.8184880099,"elo":1304.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1170916349,"x_train":19.7798589835,"imp":12.6146892497,"elo":1327.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1607817441,"x_train":50.0934628362,"imp":12.3717353184,"elo":1346.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1954148357,"x_train":100.1940386909,"imp":12.2119545046,"elo":1367.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.246440357,"x_train":200.5360400335,"imp":12.0503277586,"elo":1382.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2772856864,"x_train":300.5068144065,"imp":11.984244047,"elo":1393.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":402.9323608119,"imp":11.9597608126,"elo":1401.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.01204424,"x_train":7.5931963365,"imp":16.4815170931,"elo":1039.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0228552037,"x_train":19.5464706545,"imp":15.9818225032,"elo":1079.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0418083956,"x_train":52.461846709,"imp":15.5322088782,"elo":1138.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0609044025,"x_train":101.1927167213,"imp":15.3095708687,"elo":1154.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0833477059,"x_train":250.7607078715,"imp":15.0786611831,"elo":1175.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1012762431,"x_train":505.4611049176,"imp":15.0054156631,"elo":1183.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1012126631,"x_train":1000.4774432421,"imp":14.9259310452,"elo":1194.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1049257611,"x_train":1490.4477925659,"imp":14.8788837948,"elo":1198.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1032251722,"x_train":2006.5538981646,"imp":14.8444889121,"elo":1204.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2422366525,"x_train":2.5746349456,"imp":15.8576388612,"elo":1085.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2768373111,"x_train":3.1741914103,"imp":15.312547953,"elo":1122.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3337218245,"x_train":6.7533015627,"imp":14.5183742413,"elo":1190.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2738881738,"x_train":13.1877719851,"imp":14.1351179951,"elo":1216.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.371439796,"x_train":32.4660623442,"imp":13.547469136,"elo":1247.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4677010313,"x_train":64.6497454444,"imp":13.4253490814,"elo":1260.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6995667618,"x_train":128.5728546231,"imp":13.3545124869,"elo":1269.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7460530013,"x_train":192.4203093077,"imp":13.3423915858,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8125804985,"x_train":257.6385593165,"imp":13.3444260284,"elo":1268.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2929314919,"x_train":3.0840796949,"imp":20.1459917158,"elo":871.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4544855614,"x_train":5.6240636486,"imp":19.8504258159,"elo":891.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8902529781,"x_train":13.6732407308,"imp":18.9476612196,"elo":935.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3426618483,"x_train":27.2107956399,"imp":18.4865676439,"elo":950.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7420033115,"x_train":65.2537322804,"imp":17.520082439,"elo":990.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0242567996,"x_train":130.574483839,"imp":16.9310727816,"elo":1014.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.5714421463,"x_train":262.2064329926,"imp":16.338156479,"elo":1035.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7283647341,"x_train":393.3022867336,"imp":15.93147183,"elo":1057.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8297120981,"x_train":527.2961316043,"imp":15.6594635682,"elo":1071.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2962263441,"x_train":2.0296176424,"imp":12.3451040358,"elo":1201.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4089228478,"x_train":5.8086693996,"imp":11.4997719236,"elo":1271.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.8292153136,"x_train":16.4346292254,"imp":10.4066250539,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.8037515446,"x_train":32.0193358834,"imp":9.8788450456,"elo":1451.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.2339770555,"x_train":82.7850514402,"imp":9.3381159903,"elo":1529.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.9246780943,"x_train":169.4629431685,"imp":9.1757274338,"elo":1560.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.5906150954,"x_train":340.4054795841,"imp":8.9869444807,"elo":1584.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.9138833177,"x_train":509.4885196025,"imp":8.8864521179,"elo":1597.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.476648655,"x_train":686.4972156667,"imp":8.7938407814,"elo":1614.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":14.072768612,"x_train":59.3758334402,"imp":10.3904487974,"elo":1352.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2894827463,"x_train":16.0749964455,"imp":10.7841513848,"elo":1436.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.6821414906,"x_train":38.647009759,"imp":10.2023184034,"elo":1470.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.4325592808,"x_train":106.244026897,"imp":9.4160045764,"elo":1530.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1934690352,"x_train":199.3707655728,"imp":9.0269539703,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.1316083376,"x_train":462.7480963285,"imp":8.6556576511,"elo":1617.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.3124525847,"x_train":955.9940479091,"imp":8.4515123264,"elo":1624.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.723377164,"x_train":1870.556153527,"imp":8.3000951981,"elo":1630.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.1097813681,"x_train":2790.7899911449,"imp":8.1946214692,"elo":1643.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5805507968,"x_train":3811.6104621203,"imp":8.1654997251,"elo":1645.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":1.3455286154,"x_train":6.8642025254,"imp":4.1176969928,"elo":1850.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0067377719,"x_train":11.8050270551,"imp":3.0630923517,"elo":1959.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0171543088,"x_train":22.8030707188,"imp":18.2955753566,"elo":969.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1298234739,"x_train":53.0597722616,"imp":16.107246993,"elo":1070.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3758127641,"x_train":147.0385313693,"imp":14.3675168325,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4333863798,"x_train":177.9195746923,"imp":14.3437457459,"elo":1165.7,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0638540688,"x_train":0.6331018043,"imp":16.6499447959,"elo":1001.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468165116,"x_train":3.5138045858,"imp":15.8760735961,"elo":1055.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2412604538,"x_train":12.8062748979,"imp":15.3908176771,"elo":1095.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3470587813,"x_train":26.5826232412,"imp":14.9942764132,"elo":1123.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5360758678,"x_train":64.673699967,"imp":14.7770918641,"elo":1142.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6230115397,"x_train":130.0298950791,"imp":14.6901072863,"elo":1152.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7069692121,"x_train":259.9521641476,"imp":14.638686862,"elo":1162.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7234915951,"x_train":391.5234331877,"imp":14.6211649426,"elo":1167.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7700427525,"x_train":526.192235626,"imp":14.6157763538,"elo":1168.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4152312914,"x_train":9.1753045614,"imp":11.4826162385,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.1252015508,"x_train":23.6721342187,"imp":9.3869073322,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.7976884724,"x_train":78.6133073597,"imp":7.3521875932,"elo":1558.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.7937010153,"x_train":191.0765941914,"imp":6.6293265441,"elo":1610.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.6360113436,"x_train":503.1010684492,"imp":6.1625582504,"elo":1666.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.336614684,"x_train":1009.2362249093,"imp":5.9055348463,"elo":1707.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.6345512743,"x_train":2015.6576136475,"imp":5.7842906504,"elo":1728.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.7498142175,"x_train":3026.2086158211,"imp":5.7349512293,"elo":1742.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.9105234185,"x_train":4057.3100555326,"imp":5.6951806629,"elo":1750.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5191857229,"x_train":6.9411546573,"imp":5.1081297102,"elo":1679.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3928986427,"x_train":17.2836445024,"imp":4.6374541046,"elo":1706.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.831577433,"x_train":47.5797720049,"imp":3.8614140786,"elo":1759.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.3479395149,"x_train":98.3276706028,"imp":3.4224976388,"elo":1822.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.8281839931,"x_train":249.0037143861,"imp":2.9685684337,"elo":1889.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.5954304941,"x_train":503.1695046416,"imp":2.7947420106,"elo":1913.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.945912213,"x_train":1018.7089146459,"imp":2.7018005663,"elo":1946.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.9331824935,"x_train":1526.668944465,"imp":2.7311655156,"elo":1948.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3660135567,"x_train":2043.5362680496,"imp":2.7245619429,"elo":1953.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.5421299186,"x_train":18.769283676,"imp":11.0464917637,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"x_train":44.6485648211,"imp":5.6593617074,"elo":1685.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":54.7143383617,"x_train":88.2308458205,"imp":5.2454696275,"elo":1715.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":82.6659047931,"x_train":186.0445098453,"imp":4.8490533938,"elo":1740.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":109.9030198642,"x_train":341.1376387803,"imp":4.5346498664,"elo":1777.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":170.7606880264,"x_train":866.4570381261,"imp":4.2812155017,"elo":1811.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":225.019656008,"x_train":1760.5896780945,"imp":4.1783339382,"elo":1826.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":283.327169958,"x_train":3528.4122296034,"imp":4.0803258048,"elo":1844.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":320.2042002918,"x_train":5308.8782679903,"imp":4.0274723245,"elo":1859.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":336.9071136327,"x_train":7126.6747880976,"imp":4.0079622689,"elo":1867.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"x_train":2.0183513985,"imp":5.431606219,"elo":1709.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.4924694208,"x_train":40.7926562599,"imp":1.4100424178,"elo":2256.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1218380804,"x_train":1.3151215973,"imp":4.3753875956,"elo":1766.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1337247011,"x_train":9.7873678423,"imp":9.9099485615,"elo":1299.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2761440315,"x_train":23.4153138909,"imp":9.3565549555,"elo":1355.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.551058693,"x_train":71.5165108752,"imp":8.7147560503,"elo":1401.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.760580889,"x_train":166.1346310591,"imp":8.182159933,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.0490418104,"x_train":413.5775428251,"imp":7.7342280619,"elo":1493.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.3239622098,"x_train":844.0698066634,"imp":7.4331475795,"elo":1526.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.6840673106,"x_train":1693.1667185284,"imp":7.2469532672,"elo":1546.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.9026564546,"x_train":2557.2442472516,"imp":7.2041346491,"elo":1550.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.1105459826,"x_train":3448.7325920652,"imp":7.1664648695,"elo":1549.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6996427484,"x_train":8.4368371839,"imp":3.393775576,"elo":1930.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3981148822,"x_train":4.0220384031,"imp":2.2837784913,"elo":1953.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5829823393,"x_train":302.8991048646,"imp":27.4310818967,"elo":798.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.0719468341,"x_train":632.7525144671,"imp":27.2879140579,"elo":819.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.0770172729,"x_train":1217.0384701627,"imp":26.9251443516,"elo":853.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.049836544,"x_train":2404.0220528997,"imp":26.2981860908,"elo":891.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.2612731687,"x_train":5871.8477809815,"imp":25.4287396653,"elo":934.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.7093243804,"x_train":11843.1243713497,"imp":24.7714334055,"elo":969.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.9694047613,"x_train":23463.5184563875,"imp":24.4196340818,"elo":988.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.7650487513,"x_train":34621.1415346513,"imp":24.2437727457,"elo":1001.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.148843297,"x_train":46109.5334192428,"imp":24.1141000046,"elo":1011.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719372146,"x_train":0.8979740772,"imp":7.1457344991,"elo":1521.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0813896645,"x_train":20.5017216002,"imp":15.5036655668,"elo":1029.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1755690741,"x_train":53.7125230803,"imp":14.8522098568,"elo":1059.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3675462419,"x_train":118.8503302159,"imp":13.6912428585,"elo":1129.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5785459292,"x_train":237.0519073624,"imp":13.0974908775,"elo":1164.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7589743999,"x_train":547.271171747,"imp":12.2733894118,"elo":1202.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.982110054,"x_train":1129.5925788802,"imp":11.7896668144,"elo":1233.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1225759388,"x_train":2199.7106433499,"imp":11.2664058306,"elo":1253.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8530829243,"x_train":3318.9517565565,"imp":11.0270650726,"elo":1268.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8515826242,"x_train":4452.1085706374,"imp":10.8530281343,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1795378646,"x_train":2.1459956511,"imp":13.0504189304,"elo":1163.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6639550887,"x_train":8.3667450823,"imp":11.2771958175,"elo":1288.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1108368303,"x_train":20.7207184201,"imp":10.5345260026,"elo":1358.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2649740084,"x_train":41.7239328404,"imp":10.1437778435,"elo":1398.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.6863086909,"x_train":105.2854855499,"imp":9.8416341087,"elo":1428.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0386947605,"x_train":210.1253434414,"imp":9.6989215232,"elo":1442.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0268020363,"x_train":421.9300367907,"imp":9.6014900362,"elo":1453.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.2593267071,"x_train":635.5350385867,"imp":9.5372274664,"elo":1455.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3747946103,"x_train":848.9941919866,"imp":9.5122528052,"elo":1454.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":81.4898053701,"x_train":373.4221052753,"imp":19.2497229645,"elo":879.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":133.265500669,"x_train":839.9616197382,"imp":14.9056875458,"elo":1072.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":212.0443313695,"x_train":2344.8516752833,"imp":16.2691579543,"elo":1269.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":304.6354469252,"x_train":4981.7847745692,"imp":14.5448573032,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":380.3901708253,"x_train":12170.8681485011,"imp":8.5512134328,"elo":1528.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":387.572909971,"x_train":12741.0939157936,"imp":8.5336049305,"elo":1531.1,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.7548401788,"x_train":2.469327894,"imp":14.2952393944,"elo":1215.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8950323165,"x_train":6.0503740942,"imp":12.9135512688,"elo":1272.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.0102057688,"x_train":17.3170071504,"imp":11.1117919108,"elo":1350.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.126261214,"x_train":34.6552917016,"imp":9.8076974115,"elo":1416.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.4472187823,"x_train":88.3965998327,"imp":8.6905248694,"elo":1479.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.7895630258,"x_train":176.967409496,"imp":8.3533128978,"elo":1515.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.8867738381,"x_train":353.1777869696,"imp":8.1041203633,"elo":1529.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.1903114886,"x_train":528.7105932474,"imp":7.99224468,"elo":1542.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9734199806,"x_train":708.9450090283,"imp":7.9659724924,"elo":1548.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_all/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..b4f99a5caac93bdd0de2872c70ae79d6e672109d
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_regression/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "Nori-30M (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-2.6 (default)", "TabDPT (tuned + ensembled)", "Nori (default)", "TabDPT (tuned)", "RealTabPFN-2.5 (tuned)", "TabICLv2 (default)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabDPT-Turbo (default)", "RealTabPFN-2.5 (default)", "TabDPT (default)", "ModernNCA (tuned + ensembled)", "LightGBM (tuned + ensembled)", "RealMLP (tuned)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "xRFM (tuned + ensembled)", "TabM (tuned + ensembled)", "iLTM (tuned + ensembled)", "TabSwift (default)", "LightGBM (tuned)", "TabM (tuned)", "xRFM (tuned)", "XGBoost (tuned)", "XGBoost (tuned + ensembled)", "ModernNCA (default)", "CatBoost (default)", "ModernNCA (tuned)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "LimiX (default)", "ChimeraBoost (tuned)", "SAP-RPT-OSS (default)", "TabM (default)", "TorchMLP (tuned + ensembled)", "RealMLP (default)", "ExtraTrees (tuned + ensembled)", "ChimeraBoost (default)", "ExtraTrees (tuned)", "TorchMLP (tuned)", "xRFM (default)", "LightGBM (default)", "EBM (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "XGBoost (default)", "RandomForest (tuned + ensembled)", "EBM (tuned)", "PerpetualBooster (tuned)", "RandomForest (tuned)", "ExtraTrees (default)", "FastaiMLP (tuned + ensembled)", "EBM (default)", "TorchMLP (default)", "FastaiMLP (tuned)", "RandomForest (default)", "TabSTAR (tuned + ensembled)", "PerpetualBooster (default)", "TabSTAR (tuned)", "FastaiMLP (default)", "iLTM (default)", "KNN (tuned + ensembled)", "TabSTAR (default)", "KNN (tuned)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3076923191547394, NaN, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.23076923191547394, 0.6153846383094788, NaN, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.6153846383094788, 0.38461539149284363, NaN, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.8461538553237915, 0.5384615659713745, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.5384615659713745, 0.4615384638309479, 0.4615384638309479, NaN, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, NaN, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 1.0, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, NaN, 0.4615384638309479, 0.8461538553237915, 0.4615384638309479, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, NaN, 0.692307710647583, 0.4615384638309479, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.38461539149284363, 0.38461539149284363, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.1538461595773697, 0.3076923191547394, NaN, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.5384615659713745, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, 0.23076923191547394, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, NaN, 0.3076923191547394, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.5384615659713745, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.5384615659713745, 0.692307710647583, NaN, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, NaN, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.38461539149284363, 0.23076923191547394, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.4615384638309479, NaN, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.9230769276618958, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.7692307829856873, 1.0, 0.9230769276618958, 0.7692307829856873, 1.0, 0.7692307829856873, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, NaN, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.4615384638309479, 0.4615384638309479, 0.23076923191547394, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.6153846383094788, NaN, 0.5384615659713745, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.23076923191547394, 0.0, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, NaN, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, NaN, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 1.0, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.07692307978868484, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, NaN, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.8461538553237915, 0.5384615659713745, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 1.0, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.0, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.4615384638309479, 0.3076923191547394, 0.5384615659713745, NaN, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 1.0, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.7692307829856873, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, NaN, 0.692307710647583, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.692307710647583, 0.38461539149284363, 0.5384615659713745, 0.7692307829856873, 0.7692307829856873, 0.5384615659713745, 0.9230769276618958, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 1.0, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.3076923191547394, 0.3076923191547394, NaN, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.6153846383094788, 0.4615384638309479, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 1.0, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.5384615659713745, 0.38461539149284363, 0.692307710647583, 0.5384615659713745, NaN, 0.4615384638309479, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.8461538553237915, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.6153846383094788, 0.5384615659713745, 0.5384615659713745, NaN, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.8461538553237915, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.4615384638309479, 0.7692307829856873, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 1.0, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, 0.3076923191547394, 0.4615384638309479, NaN, 0.38461539149284363, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 1.0, 0.7692307829856873, 0.9230769276618958, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.5384615659713745, 0.6153846383094788, NaN, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, NaN, 0.38461539149284363, 0.4615384638309479, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.38461539149284363, 0.4615384638309479, 0.3076923191547394, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.6153846383094788, NaN, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.38461539149284363, 0.7692307829856873, 0.4615384638309479, 0.7692307829856873, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 1.0, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.1538461595773697, 0.4615384638309479, 0.5384615659713745, 0.3076923191547394, 0.5384615659713745, 0.5384615659713745, NaN, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.7692307829856873, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, NaN, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.4615384638309479, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, NaN, 0.4615384638309479, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.5384615659713745, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.3076923191547394, 0.5384615659713745, 0.38461539149284363, 0.3076923191547394, 0.4615384638309479, 0.6153846383094788, 0.23076923191547394, 0.4615384638309479, 0.5384615659713745, NaN, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.692307710647583, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.38461539149284363, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.23076923191547394, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, 0.38461539149284363, NaN, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.4615384638309479, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 1.0, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.0, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.5384615659713745, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, NaN, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 1.0, 1.0, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, NaN, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.9230769276618958, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.692307710647583, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, NaN, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, NaN, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.3076923191547394, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, 0.38461539149284363, NaN, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.8461538553237915, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, NaN, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.692307710647583, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.5384615659713745, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.0, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.0, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, 0.6153846383094788, NaN, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.4615384638309479, NaN, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 1.0, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, NaN, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 1.0, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.6153846383094788, 0.3076923191547394, 0.4615384638309479, 0.38461539149284363, NaN, 0.692307710647583, 0.7692307829856873, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 1.0, 0.7692307829856873, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.23076923191547394, 0.0, 0.0, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.07692307978868484, 0.38461539149284363, 0.07692307978868484, 0.3076923191547394, 0.38461539149284363, 0.1538461595773697, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, 0.4615384638309479, 0.3076923191547394, NaN, 0.4615384638309479, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.9230769276618958, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.5384615659713745, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.5384615659713745, NaN, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.4615384638309479, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.5384615659713745, 0.3076923191547394, 0.23076923191547394, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, NaN, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 1.0, 0.8461538553237915, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.1538461595773697, 0.4615384638309479, 0.3076923191547394, 0.5384615659713745, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, NaN, 0.38461539149284363, 0.4615384638309479, 0.3076923191547394, 0.38461539149284363, 0.6153846383094788, 0.4615384638309479, 0.4615384638309479, 0.692307710647583, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.1538461595773697, 0.4615384638309479, 0.3076923191547394, 0.38461539149284363, 0.1538461595773697, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, NaN, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, NaN, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.8461538553237915, 0.5384615659713745, 0.38461539149284363, 0.38461539149284363, 0.6153846383094788, 1.0, 0.7692307829856873, 0.692307710647583, 0.5384615659713745, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.0, 0.3076923191547394, 0.0, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.4615384638309479, 0.07692307978868484, 0.38461539149284363, 0.1538461595773697, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.07692307978868484, 0.38461539149284363, 0.4615384638309479, 0.692307710647583, 0.38461539149284363, 0.6153846383094788, NaN, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.807692289352417, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 1.0, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.1538461595773697, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.38461539149284363, 0.1538461595773697, 0.3076923191547394, 0.07692307978868484, 0.38461539149284363, 0.23076923191547394, 0.38461539149284363, 0.3076923191547394, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.3076923191547394, 0.5384615659713745, 0.4615384638309479, NaN, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.1538461595773697, 0.0, 0.07692307978868484, 0.0, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.38461539149284363, 0.23076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.6153846383094788, 0.4615384638309479, 0.4615384638309479, NaN, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.5384615659713745, 0.9230769276618958, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, NaN, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.8461538553237915, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.38461539149284363, 0.1538461595773697, 0.4615384638309479, 0.1538461595773697, 0.4615384638309479, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.4615384638309479, 0.5384615659713745, 0.3076923191547394, 0.4615384638309479, 0.19230769574642181, 0.38461539149284363, 0.4615384638309479, 0.6153846383094788, NaN, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.38461539149284363, 0.23076923191547394, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.6153846383094788, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.6153846383094788, 0.38461539149284363, NaN, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.5384615659713745, 0.7307692170143127, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.23076923191547394, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.23076923191547394, 0.0, 0.3076923191547394, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.6153846383094788, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.6153846383094788, 0.38461539149284363, 0.4615384638309479, NaN, 0.4615384638309479, 0.6153846383094788, 0.38461539149284363, 0.4615384638309479, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.4615384638309479, 0.23076923191547394, 0.38461539149284363, 0.23076923191547394, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.23076923191547394, 0.4615384638309479, 0.5384615659713745, NaN, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 1.0, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.0, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.1538461595773697, 0.23076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, NaN, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.0, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.0, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.5384615659713745, 0.0, 0.38461539149284363, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.3076923191547394, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.38461539149284363, NaN, 0.4615384638309479, 0.6153846383094788, 0.692307710647583, 0.38461539149284363, 0.692307710647583, 0.4615384638309479, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.1538461595773697, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.4615384638309479, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, NaN, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.38461539149284363, 0.1538461595773697, 0.07692307978868484, 0.4615384638309479, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.4615384638309479, 0.23076923191547394, 0.26923078298568726, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, 0.4615384638309479, NaN, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, NaN, 0.23076923191547394, 0.8461538553237915, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 1.0, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.23076923191547394, 0.0, 0.07692307978868484, 0.38461539149284363, 0.3076923191547394, 0.07692307978868484, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.6153846383094788, 0.38461539149284363, 0.38461539149284363, 0.7692307829856873, NaN, 0.7692307829856873, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, NaN, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 1.0, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.0, 0.38461539149284363, 0.5384615659713745, 0.3076923191547394, 0.3076923191547394, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, NaN, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.0, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.0, 0.3076923191547394, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.6153846383094788, 0.38461539149284363, 0.6153846383094788, 0.5384615659713745, NaN, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.0, 0.1538461595773697, 0.0, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.0, 0.23076923191547394, 0.23076923191547394, 0.0, 0.07692307978868484, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.1538461595773697, 0.5384615659713745, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, 0.3076923191547394, NaN, 0.5384615659713745, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.0, 0.1538461595773697, 0.0, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, NaN, 0.5384615659713745, 0.5384615659713745, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.0, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.0, 0.4615384638309479, NaN, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.0, 0.38461539149284363, 0.1538461595773697, 0.38461539149284363, 0.23076923191547394, 0.1538461595773697, 0.0, 0.4615384638309479, 0.1538461595773697, NaN, 0.692307710647583, 0.692307710647583, 0.692307710647583], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, NaN, 0.807692289352417, 0.8461538553237915], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, 0.19230769574642181, NaN, 0.7692307829856873], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.1538461595773697, 0.0, 0.3076923191547394, 0.1538461595773697, 0.23076923191547394, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9761646818},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9068255723},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.9035753026},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.8981581849},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.889490796},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8829902539},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8569880877},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.8526543934},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.8277356503},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8277356509},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8082340277},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.786565553},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.785482128},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7638136571},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7551462705},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7518959993},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7291441036},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6923076989},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6879740064},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6652221077},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6543878717},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6533044504},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6533044504},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6424702133},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6381365211},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6153846227},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6153846226},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.6034669634},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5807150661},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5807150649},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5655471369},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5590465943},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.547128937},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5417118184},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5124593793},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4962080287},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4907909067},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4647887406},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4647887398},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4377031498},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4377031492},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4344528782},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4149512532},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4084507117},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3986999005},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3856988147},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3856988152},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3781148501},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3613217827},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3575298009},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3499458349},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.336944752},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.3277356498},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3212351086},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.300108348},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.2762730277},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.2589382501},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2535211311},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2524377081},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2410617588},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.231852659},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2112676097},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2058504917},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1744312066},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.1646803935},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.16359697},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1321776851},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1040086687},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0747562316},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0427952335},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0330444213},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0151679311}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_all/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_regression/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..8cefbe7d26670934c9a2a54764a31eb240f9cc56
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_all/winrate_matrix.csv
@@ -0,0 +1,73 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),Nori-30M (default),RealTabPFN-2.5 (tuned + ensembled),TabPFN-2.6 (default),TabDPT (tuned + ensembled),Nori (default),TabDPT (tuned),RealTabPFN-2.5 (tuned),TabICLv2 (default),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabDPT-Turbo (default),RealTabPFN-2.5 (default),TabDPT (default),ModernNCA (tuned + ensembled),LightGBM (tuned + ensembled),RealMLP (tuned),CatBoost (tuned + ensembled),CatBoost (tuned),xRFM (tuned + ensembled),TabM (tuned + ensembled),iLTM (tuned + ensembled),TabSwift (default),LightGBM (tuned),TabM (tuned),xRFM (tuned),XGBoost (tuned),XGBoost (tuned + ensembled),ModernNCA (default),CatBoost (default),ModernNCA (tuned),ChimeraBoost (tuned + ensembled),iLTM (tuned),LimiX (default),ChimeraBoost (tuned),SAP-RPT-OSS (default),TabM (default),TorchMLP (tuned + ensembled),RealMLP (default),ExtraTrees (tuned + ensembled),ChimeraBoost (default),ExtraTrees (tuned),TorchMLP (tuned),xRFM (default),LightGBM (default),EBM (tuned + ensembled),PerpetualBooster (tuned + ensembled),XGBoost (default),RandomForest (tuned + ensembled),EBM (tuned),PerpetualBooster (tuned),RandomForest (tuned),ExtraTrees (default),FastaiMLP (tuned + ensembled),EBM (default),TorchMLP (default),FastaiMLP (tuned),RandomForest (default),TabSTAR (tuned + ensembled),PerpetualBooster (default),TabSTAR (tuned),FastaiMLP (default),iLTM (default),KNN (tuned + ensembled),TabSTAR (default),KNN (tuned),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,0.6923077,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,1.0,0.84615386,0.9230769,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.30769232,,0.3846154,0.3846154,0.46153846,0.61538464,0.53846157,0.61538464,0.61538464,0.61538464,0.46153846,0.84615386,0.9230769,0.9230769,0.6923077,0.7692308,0.6923077,1.0,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.23076923,0.61538464,,0.61538464,0.53846157,0.61538464,0.46153846,0.6923077,0.61538464,0.7692308,0.6923077,0.84615386,0.84615386,0.9230769,0.84615386,1.0,0.84615386,0.9230769,0.84615386,0.84615386,0.84615386,0.9230769,1.0,1.0,0.9230769,0.9230769,1.0,1.0,0.9230769,0.9230769,1.0,0.9230769,0.84615386,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,0.84615386,0.9230769,0.9230769,1.0,0.84615386,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori-30M (default),0.15384616,0.61538464,0.3846154,,0.53846157,0.61538464,0.53846157,0.6923077,0.84615386,0.53846157,0.61538464,0.7692308,0.84615386,0.9230769,0.6923077,0.7692308,0.6923077,0.9230769,0.7692308,0.9230769,0.9230769,0.9230769,1.0,1.0,0.84615386,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,0.84615386,0.9230769,1.0,0.84615386,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.15384616,0.53846157,0.46153846,0.46153846,,0.53846157,0.61538464,0.61538464,0.61538464,0.7692308,0.61538464,0.9230769,0.7692308,0.84615386,0.9230769,0.6923077,0.7692308,0.9230769,0.9230769,0.84615386,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0,0.9230769,1.0,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.15384616,0.3846154,0.3846154,0.3846154,0.46153846,,0.61538464,0.53846157,0.61538464,0.53846157,0.53846157,0.7692308,0.61538464,0.6923077,0.7692308,0.6923077,0.7692308,0.84615386,1.0,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.15384616,0.46153846,0.53846157,0.46153846,0.3846154,0.3846154,,0.46153846,0.84615386,0.46153846,0.53846157,0.6923077,0.7692308,0.84615386,0.6923077,0.9230769,0.6923077,0.7692308,0.6923077,0.84615386,0.7692308,0.9230769,0.84615386,0.84615386,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.84615386,0.84615386,0.9230769,1.0,0.84615386,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,0.9230769,1.0,0.84615386,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori (default),0.07692308,0.3846154,0.30769232,0.30769232,0.3846154,0.46153846,0.53846157,,0.6923077,0.46153846,0.61538464,0.6923077,0.61538464,0.6923077,0.53846157,0.6923077,0.61538464,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,1.0,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.84615386,0.9230769,0.7692308,0.9230769,1.0,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,0.9230769,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.3846154,0.3846154,0.15384616,0.3846154,0.3846154,0.15384616,0.30769232,,0.53846157,0.46153846,0.6923077,0.7692308,0.84615386,0.53846157,0.84615386,0.6923077,0.84615386,0.6923077,0.84615386,0.7692308,0.9230769,0.7692308,0.84615386,0.7692308,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.84615386,0.7692308,0.9230769,1.0,0.84615386,0.9230769,0.9230769,0.9230769,1.0,0.9230769,1.0,0.9230769,1.0,1.0,0.9230769,1.0,0.84615386,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.15384616,0.3846154,0.23076923,0.46153846,0.23076923,0.46153846,0.53846157,0.53846157,0.46153846,,0.30769232,0.7692308,0.6923077,0.7692308,0.7692308,0.61538464,0.61538464,0.7692308,0.84615386,0.7692308,0.7692308,0.84615386,0.84615386,0.7692308,0.7692308,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,1.0,0.9230769,0.9230769,1.0,0.9230769,1.0,0.84615386,1.0,0.9230769,1.0,1.0,0.9230769,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,1.0,1.0,0.9230769,0.84615386,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.07692308,0.53846157,0.30769232,0.3846154,0.3846154,0.46153846,0.46153846,0.3846154,0.53846157,0.6923077,,0.6923077,0.7692308,0.6923077,0.61538464,0.53846157,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,0.84615386,1.0,0.9230769,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.15384616,0.15384616,0.23076923,0.07692308,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.30769232,,0.53846157,0.61538464,0.53846157,0.46153846,0.53846157,0.6923077,0.7692308,0.84615386,0.84615386,0.84615386,0.6923077,0.84615386,0.7692308,0.7692308,0.7692308,0.9230769,0.9230769,0.9230769,0.9230769,0.84615386,0.7692308,0.84615386,0.9230769,0.7692308,1.0,0.84615386,1.0,0.9230769,1.0,1.0,0.9230769,1.0,0.9230769,0.9230769,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.07692308,0.15384616,0.15384616,0.23076923,0.3846154,0.23076923,0.3846154,0.23076923,0.30769232,0.23076923,0.46153846,,0.46153846,0.46153846,0.46153846,0.61538464,0.9230769,0.6923077,0.84615386,0.7692308,0.84615386,0.6923077,0.6923077,0.6923077,0.9230769,0.7692308,0.84615386,0.9230769,1.0,0.9230769,1.0,0.7692308,1.0,0.9230769,0.7692308,1.0,0.7692308,0.9230769,1.0,0.9230769,0.9230769,1.0,0.9230769,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.30769232,0.15384616,0.30769232,0.15384616,0.23076923,0.30769232,0.3846154,0.53846157,,0.3846154,0.46153846,0.53846157,0.6923077,0.7692308,0.84615386,0.7692308,0.84615386,0.61538464,0.7692308,0.7692308,0.84615386,0.7692308,0.9230769,0.9230769,0.9230769,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,0.7692308,0.84615386,0.84615386,1.0,1.0,0.9230769,1.0,0.84615386,1.0,1.0,0.9230769,1.0,0.84615386,0.9230769,0.9230769,1.0,0.84615386,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.07692308,0.30769232,0.15384616,0.30769232,0.07692308,0.23076923,0.30769232,0.46153846,0.46153846,0.23076923,0.3846154,0.46153846,0.53846157,0.61538464,,0.53846157,0.61538464,0.6923077,0.61538464,0.7692308,0.7692308,0.7692308,0.7692308,0.7692308,0.84615386,0.6923077,0.7692308,0.84615386,0.7692308,0.7692308,0.7692308,0.84615386,0.7692308,0.84615386,0.7692308,0.6923077,0.84615386,0.84615386,0.9230769,0.84615386,0.84615386,0.9230769,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,0.9230769,0.9230769,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0
+TabDPT (default),0.0,0.23076923,0.0,0.23076923,0.30769232,0.30769232,0.07692308,0.30769232,0.15384616,0.3846154,0.46153846,0.53846157,0.53846157,0.53846157,0.46153846,,0.61538464,0.6923077,0.53846157,0.7692308,0.7692308,0.84615386,0.6923077,0.6923077,0.6923077,0.7692308,0.7692308,0.84615386,0.7692308,0.7692308,0.7692308,0.7692308,0.6923077,0.84615386,0.9230769,0.7692308,0.9230769,0.7692308,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.84615386,0.9230769,0.84615386,0.84615386,0.84615386,1.0,0.84615386,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.07692308,0.30769232,0.15384616,0.30769232,0.23076923,0.23076923,0.30769232,0.3846154,0.30769232,0.3846154,0.23076923,0.46153846,0.3846154,0.46153846,0.3846154,0.3846154,,0.6923077,0.6923077,0.53846157,0.53846157,0.6923077,0.61538464,0.6923077,0.7692308,0.84615386,0.61538464,0.6923077,0.7692308,0.84615386,0.7692308,0.7692308,1.0,0.84615386,0.84615386,0.6923077,0.84615386,0.7692308,0.84615386,0.7692308,0.7692308,0.84615386,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,0.84615386,0.9230769,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.15384616,0.30769232,0.07692308,0.30769232,0.30769232,0.30769232,0.30769232,,0.46153846,0.53846157,0.53846157,0.46153846,0.53846157,0.61538464,0.46153846,0.84615386,0.53846157,0.61538464,0.7692308,0.84615386,0.7692308,0.9230769,0.6923077,0.84615386,0.6923077,0.6923077,1.0,0.6923077,0.9230769,0.84615386,0.84615386,0.9230769,1.0,0.9230769,0.84615386,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.07692308,0.15384616,0.23076923,0.07692308,0.0,0.30769232,0.15384616,0.30769232,0.15384616,0.15384616,0.23076923,0.30769232,0.23076923,0.3846154,0.46153846,0.30769232,0.53846157,,0.61538464,0.6923077,0.61538464,0.61538464,0.61538464,0.53846157,0.6923077,0.6923077,0.7692308,0.7692308,0.7692308,0.7692308,0.61538464,0.7692308,0.61538464,0.7692308,0.6923077,0.7692308,0.7692308,0.9230769,0.84615386,0.84615386,0.7692308,0.7692308,0.7692308,1.0,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.9230769,1.0,0.9230769,0.7692308,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.46153846,0.46153846,0.3846154,,0.6923077,0.30769232,0.3846154,0.46153846,0.3846154,0.6923077,0.3846154,0.53846157,0.7692308,0.7692308,0.53846157,0.9230769,0.6923077,0.6923077,0.7692308,0.6923077,0.9230769,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,1.0,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.15384616,0.07692308,0.15384616,0.07692308,0.23076923,0.07692308,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.46153846,0.46153846,0.30769232,0.30769232,,0.46153846,0.46153846,0.46153846,0.3846154,0.61538464,0.46153846,0.53846157,0.6923077,0.6923077,0.61538464,0.7692308,0.6923077,0.6923077,0.7692308,0.6923077,0.7692308,0.6923077,0.84615386,0.84615386,0.84615386,0.84615386,1.0,0.84615386,0.84615386,0.7692308,0.9230769,0.9230769,1.0,1.0,0.9230769,0.9230769,1.0,1.0,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.15384616,0.30769232,0.53846157,0.3846154,0.6923077,0.53846157,,0.46153846,0.6923077,0.53846157,0.61538464,0.53846157,0.84615386,0.61538464,0.53846157,0.6923077,0.6923077,0.61538464,0.61538464,0.84615386,0.7692308,0.6923077,0.6923077,0.7692308,0.84615386,0.7692308,0.9230769,0.7692308,0.9230769,0.84615386,0.9230769,1.0,0.84615386,0.9230769,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,0.9230769,0.84615386,1.0,0.9230769,1.0,0.84615386,0.9230769,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.0,0.0,0.0,0.15384616,0.07692308,0.15384616,0.15384616,0.23076923,0.15384616,0.15384616,0.30769232,0.30769232,0.3846154,0.23076923,0.30769232,0.3846154,0.46153846,0.3846154,0.61538464,0.53846157,0.53846157,,0.53846157,0.46153846,0.6923077,0.84615386,0.53846157,0.6923077,0.6923077,0.46153846,0.7692308,0.53846157,0.6923077,0.6923077,0.6923077,0.84615386,0.6923077,1.0,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.9230769,0.7692308,0.84615386,0.9230769,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,0.9230769,0.84615386,1.0,1.0,1.0,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.23076923,0.15384616,0.15384616,0.30769232,0.23076923,0.23076923,0.30769232,0.30769232,0.3846154,0.3846154,0.53846157,0.53846157,0.30769232,0.46153846,,0.3846154,0.61538464,0.61538464,0.46153846,0.7692308,0.7692308,0.61538464,0.61538464,0.61538464,0.61538464,0.7692308,0.61538464,0.6923077,0.7692308,0.7692308,0.84615386,0.7692308,0.9230769,0.7692308,0.9230769,0.9230769,0.9230769,0.9230769,0.7692308,0.9230769,0.9230769,1.0,0.7692308,0.9230769,1.0,1.0,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.07692308,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.07692308,0.07692308,0.23076923,0.23076923,0.15384616,0.23076923,0.30769232,0.23076923,0.15384616,0.30769232,0.23076923,0.53846157,0.46153846,0.61538464,0.61538464,0.46153846,0.53846157,0.61538464,,0.61538464,0.61538464,0.6923077,0.7692308,0.6923077,0.6923077,0.61538464,0.61538464,0.61538464,0.6923077,0.61538464,0.61538464,0.84615386,0.7692308,0.7692308,0.84615386,0.6923077,0.7692308,0.6923077,0.84615386,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,1.0,0.9230769,0.9230769,1.0,0.84615386,0.84615386,0.9230769,0.84615386,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.07692308,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.07692308,0.15384616,0.30769232,0.23076923,0.15384616,0.15384616,0.30769232,0.30769232,0.3846154,0.3846154,0.30769232,0.3846154,0.3846154,,0.3846154,0.46153846,0.6923077,0.6923077,0.53846157,0.53846157,0.6923077,0.61538464,0.6923077,0.53846157,0.7692308,0.61538464,0.9230769,0.7692308,0.84615386,0.7692308,0.7692308,0.84615386,0.7692308,0.84615386,0.9230769,0.9230769,1.0,1.0,0.84615386,0.9230769,1.0,0.84615386,0.84615386,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.3846154,0.46153846,0.30769232,0.61538464,0.53846157,0.46153846,0.15384616,0.3846154,0.3846154,0.61538464,,0.46153846,0.61538464,0.61538464,0.3846154,0.7692308,0.46153846,0.7692308,0.6923077,0.61538464,0.7692308,0.61538464,1.0,0.7692308,0.84615386,0.84615386,0.7692308,0.84615386,0.9230769,0.7692308,0.9230769,0.7692308,0.84615386,0.9230769,0.7692308,0.7692308,0.84615386,0.7692308,0.9230769,0.9230769,0.84615386,1.0,1.0,1.0,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.15384616,0.07692308,0.15384616,0.15384616,0.30769232,0.3846154,0.23076923,0.46153846,0.46153846,0.15384616,0.46153846,0.53846157,0.30769232,0.53846157,0.53846157,,0.61538464,0.53846157,0.7692308,0.61538464,0.53846157,0.46153846,0.7692308,0.61538464,0.53846157,0.6923077,0.6923077,0.84615386,0.84615386,0.7692308,0.7692308,0.7692308,0.7692308,0.9230769,0.84615386,0.7692308,0.9230769,0.9230769,0.9230769,0.7692308,1.0,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.30769232,0.3846154,0.30769232,0.23076923,0.23076923,0.30769232,0.3846154,0.3846154,,0.53846157,0.53846157,0.46153846,0.53846157,0.53846157,0.6923077,0.6923077,0.53846157,0.61538464,0.6923077,0.84615386,0.7692308,0.84615386,0.84615386,0.7692308,0.84615386,0.6923077,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.0,0.07692308,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.30769232,0.46153846,0.30769232,0.23076923,0.30769232,0.30769232,0.3846154,0.46153846,0.46153846,,0.46153846,0.46153846,0.53846157,0.46153846,0.6923077,0.6923077,0.53846157,0.53846157,0.7692308,0.84615386,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,1.0,0.84615386,1.0,0.9230769,0.9230769,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.07692308,0.0,0.15384616,0.15384616,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.46153846,0.3846154,0.30769232,0.53846157,0.3846154,0.30769232,0.46153846,0.61538464,0.23076923,0.46153846,0.53846157,,0.61538464,0.46153846,0.61538464,0.61538464,0.46153846,0.61538464,0.53846157,0.6923077,0.7692308,0.6923077,0.7692308,0.6923077,0.6923077,0.7692308,0.7692308,0.84615386,0.6923077,0.6923077,0.6923077,0.9230769,0.6923077,0.9230769,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.9230769,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.0,0.15384616,0.15384616,0.23076923,0.23076923,0.07692308,0.3846154,0.07692308,0.23076923,0.30769232,0.23076923,0.3846154,0.3846154,0.46153846,0.23076923,0.3846154,0.53846157,0.53846157,0.3846154,,0.53846157,0.46153846,0.61538464,0.46153846,0.6923077,0.53846157,0.6923077,0.6923077,0.6923077,0.6923077,0.9230769,0.84615386,0.6923077,0.7692308,0.84615386,0.84615386,1.0,0.84615386,0.7692308,0.84615386,1.0,0.84615386,0.9230769,0.9230769,0.9230769,0.7692308,0.9230769,0.9230769,0.84615386,1.0,0.9230769,1.0,1.0,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.15384616,0.15384616,0.07692308,0.07692308,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.23076923,0.30769232,0.0,0.30769232,0.23076923,0.30769232,0.30769232,0.3846154,0.46153846,0.3846154,0.3846154,0.30769232,0.53846157,0.46153846,0.46153846,0.46153846,0.53846157,0.46153846,,0.46153846,0.53846157,0.53846157,0.53846157,0.61538464,0.61538464,0.6923077,0.6923077,0.53846157,0.61538464,0.53846157,0.7692308,0.6923077,0.7692308,0.6923077,0.7692308,0.84615386,0.6923077,0.7692308,0.7692308,0.7692308,0.6923077,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.7692308,0.9230769,0.7692308,1.0,1.0,0.9230769,0.84615386,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.0,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.3846154,0.30769232,0.30769232,0.3846154,0.30769232,0.3846154,0.3846154,0.3846154,0.23076923,0.53846157,0.46153846,0.53846157,0.3846154,0.53846157,0.53846157,,0.53846157,0.46153846,0.6923077,0.53846157,0.6923077,0.61538464,0.7692308,0.6923077,0.9230769,0.6923077,0.6923077,0.6923077,0.6923077,0.84615386,0.9230769,0.84615386,0.7692308,0.84615386,0.9230769,0.7692308,0.6923077,0.9230769,0.9230769,0.7692308,0.84615386,0.9230769,0.7692308,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.15384616,0.30769232,0.23076923,0.23076923,0.23076923,0.15384616,0.30769232,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.30769232,0.30769232,0.3846154,0.3846154,0.46153846,0.46153846,,0.61538464,0.53846157,0.61538464,0.61538464,0.61538464,0.6923077,0.53846157,0.6923077,0.53846157,0.61538464,0.84615386,0.84615386,0.6923077,0.7692308,0.7692308,0.84615386,0.7692308,0.7692308,0.84615386,0.7692308,0.9230769,0.84615386,1.0,0.9230769,1.0,0.84615386,0.9230769,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.15384616,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.23076923,0.23076923,0.23076923,0.30769232,0.23076923,0.30769232,0.30769232,0.30769232,0.30769232,0.30769232,0.23076923,0.30769232,0.3846154,0.3846154,0.46153846,0.3846154,0.3846154,0.30769232,0.30769232,0.53846157,0.53846157,0.46153846,0.53846157,0.3846154,,0.61538464,0.53846157,0.46153846,0.61538464,0.61538464,0.46153846,0.61538464,0.46153846,0.53846157,0.53846157,0.53846157,0.6923077,0.53846157,0.61538464,0.61538464,0.6923077,0.61538464,0.61538464,0.6923077,0.7692308,0.7692308,0.7692308,0.7692308,0.7692308,0.84615386,0.84615386,0.84615386,0.7692308,0.7692308,0.9230769,0.9230769,0.9230769,1.0,0.9230769,0.9230769,1.0
+ChimeraBoost (tuned),0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.0,0.0,0.15384616,0.15384616,0.07692308,0.15384616,0.0,0.23076923,0.07692308,0.23076923,0.30769232,0.15384616,0.30769232,0.3846154,0.23076923,0.23076923,0.46153846,0.46153846,0.46153846,0.3846154,0.30769232,0.46153846,0.30769232,0.46153846,0.3846154,,0.46153846,0.61538464,0.61538464,0.7692308,0.61538464,0.84615386,0.61538464,0.53846157,0.6923077,0.6923077,0.61538464,0.9230769,0.84615386,0.7692308,0.6923077,0.84615386,0.7692308,0.6923077,0.7692308,0.9230769,0.7692308,0.84615386,0.84615386,0.7692308,0.9230769,0.7692308,0.9230769,0.9230769,0.9230769,0.84615386,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.30769232,0.23076923,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.15384616,0.3846154,0.3846154,0.30769232,0.3846154,0.46153846,0.46153846,0.46153846,0.3846154,0.46153846,0.3846154,0.46153846,0.53846157,,0.3846154,0.46153846,0.53846157,0.3846154,0.6923077,0.46153846,0.46153846,0.46153846,0.61538464,0.61538464,0.61538464,0.6923077,0.61538464,0.7692308,0.53846157,0.61538464,0.6923077,0.7692308,0.7692308,0.7692308,0.7692308,0.6923077,0.7692308,0.7692308,0.7692308,0.84615386,0.84615386,0.84615386,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.15384616,0.23076923,0.0,0.23076923,0.23076923,0.07692308,0.0,0.30769232,0.30769232,0.23076923,0.30769232,0.30769232,0.3846154,0.30769232,0.3846154,0.53846157,0.3846154,0.61538464,,0.53846157,0.6923077,0.6923077,0.61538464,0.6923077,0.6923077,0.61538464,0.84615386,0.6923077,0.84615386,0.9230769,0.6923077,0.6923077,0.84615386,0.6923077,0.7692308,0.84615386,0.7692308,0.7692308,0.9230769,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.07692308,0.0,0.0,0.15384616,0.07692308,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.30769232,0.30769232,0.3846154,0.3846154,0.3846154,0.3846154,0.53846157,0.46153846,,0.53846157,0.53846157,0.46153846,0.61538464,0.7692308,0.61538464,0.6923077,0.6923077,0.53846157,0.61538464,0.61538464,0.6923077,0.53846157,0.6923077,0.61538464,0.84615386,0.7692308,1.0,0.9230769,0.84615386,0.84615386,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.30769232,0.3846154,0.23076923,0.46153846,0.30769232,0.46153846,,0.61538464,0.53846157,0.61538464,0.46153846,0.61538464,0.61538464,0.6923077,0.53846157,0.7692308,0.7692308,0.6923077,0.6923077,0.7692308,0.7692308,0.84615386,0.84615386,0.7692308,0.7692308,0.84615386,0.7692308,0.9230769,0.7692308,1.0,0.9230769,0.84615386,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.23076923,0.15384616,0.15384616,0.07692308,0.15384616,0.07692308,0.30769232,0.23076923,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.30769232,0.46153846,0.30769232,0.46153846,0.53846157,0.3846154,0.61538464,0.30769232,0.46153846,0.3846154,,0.6923077,0.7692308,0.46153846,0.53846157,0.46153846,0.61538464,0.61538464,0.61538464,0.84615386,0.61538464,0.7692308,0.7692308,1.0,0.7692308,0.6923077,0.53846157,0.6923077,0.9230769,0.84615386,0.7692308,0.84615386,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.15384616,0.07692308,0.0,0.15384616,0.15384616,0.07692308,0.15384616,0.0,0.23076923,0.0,0.0,0.23076923,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.15384616,0.15384616,0.30769232,0.07692308,0.3846154,0.07692308,0.30769232,0.3846154,0.15384616,0.30769232,0.3846154,0.53846157,0.46153846,0.30769232,,0.46153846,0.53846157,0.61538464,0.61538464,0.6923077,0.9230769,0.6923077,0.61538464,0.7692308,0.84615386,0.6923077,0.6923077,0.7692308,0.7692308,0.61538464,0.6923077,0.84615386,0.7692308,0.9230769,0.7692308,0.84615386,0.9230769,0.84615386,0.84615386,1.0,0.9230769,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.30769232,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.30769232,0.15384616,0.46153846,0.30769232,0.46153846,0.53846157,0.3846154,0.53846157,0.30769232,0.3846154,0.3846154,0.23076923,0.53846157,,0.46153846,0.53846157,0.3846154,0.61538464,0.61538464,0.53846157,0.61538464,0.61538464,0.7692308,0.6923077,0.84615386,0.7692308,0.6923077,0.46153846,0.6923077,0.9230769,0.84615386,0.7692308,0.84615386,0.84615386,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.07692308,0.0,0.0,0.07692308,0.07692308,0.15384616,0.15384616,0.0,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.15384616,0.23076923,0.07692308,0.23076923,0.15384616,0.15384616,0.23076923,0.30769232,0.23076923,0.30769232,0.3846154,0.46153846,0.46153846,0.53846157,0.30769232,0.23076923,0.53846157,0.53846157,0.46153846,0.53846157,,0.61538464,0.53846157,0.61538464,0.53846157,0.53846157,0.61538464,0.61538464,0.53846157,0.61538464,0.61538464,0.6923077,0.7692308,1.0,0.84615386,0.61538464,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.23076923,0.07692308,0.23076923,0.07692308,0.23076923,0.15384616,0.23076923,0.07692308,0.30769232,0.23076923,0.23076923,0.23076923,0.30769232,0.30769232,0.15384616,0.46153846,0.30769232,0.53846157,0.3846154,0.3846154,0.3846154,0.46153846,0.3846154,0.46153846,0.3846154,,0.3846154,0.46153846,0.30769232,0.3846154,0.61538464,0.46153846,0.46153846,0.6923077,0.61538464,0.53846157,0.6923077,0.61538464,0.61538464,0.84615386,0.7692308,0.6923077,0.84615386,0.7692308,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.0,0.0,0.0,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.07692308,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.07692308,0.0,0.15384616,0.15384616,0.23076923,0.30769232,0.15384616,0.46153846,0.30769232,0.3846154,0.15384616,0.30769232,0.3846154,0.53846157,0.3846154,0.61538464,0.46153846,0.61538464,,0.61538464,0.61538464,0.6923077,0.53846157,0.61538464,0.6923077,0.61538464,0.7692308,0.7692308,0.6923077,0.7692308,0.7692308,0.9230769,0.84615386,0.84615386,0.84615386,0.9230769,1.0,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.15384616,0.07692308,0.0,0.0,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.23076923,0.23076923,0.15384616,0.07692308,0.23076923,0.23076923,0.15384616,0.15384616,0.30769232,0.15384616,0.30769232,0.15384616,0.30769232,0.30769232,0.3846154,0.3846154,0.30769232,0.30769232,0.30769232,0.3846154,0.30769232,0.3846154,0.3846154,0.53846157,0.3846154,,0.3846154,0.46153846,0.3846154,0.84615386,0.53846157,0.3846154,0.3846154,0.61538464,1.0,0.7692308,0.6923077,0.53846157,0.84615386,0.7692308,0.9230769,0.84615386,0.7692308,0.84615386,0.9230769,0.84615386,0.84615386,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.15384616,0.0,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.0,0.07692308,0.0,0.0,0.07692308,0.15384616,0.07692308,0.15384616,0.0,0.15384616,0.07692308,0.07692308,0.0,0.30769232,0.0,0.23076923,0.07692308,0.23076923,0.46153846,0.07692308,0.3846154,0.15384616,0.46153846,0.46153846,0.3846154,0.07692308,0.3846154,0.46153846,0.6923077,0.3846154,0.61538464,,0.53846157,0.53846157,0.61538464,0.8076923,0.61538464,0.6923077,0.7692308,0.6923077,0.6923077,0.7692308,0.7692308,0.7692308,1.0,0.7692308,0.84615386,0.9230769,0.7692308,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.07692308,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.0,0.15384616,0.0,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.0,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.30769232,0.15384616,0.15384616,0.15384616,0.23076923,0.3846154,0.15384616,0.30769232,0.07692308,0.3846154,0.23076923,0.3846154,0.30769232,0.46153846,0.46153846,0.61538464,0.30769232,0.53846157,0.46153846,,0.53846157,0.61538464,0.61538464,0.6923077,0.6923077,0.61538464,0.7692308,0.7692308,0.6923077,0.7692308,0.84615386,0.9230769,0.84615386,0.7692308,0.9230769,0.84615386,0.84615386,1.0,1.0,0.9230769,0.9230769,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.15384616,0.0,0.07692308,0.0,0.15384616,0.0,0.15384616,0.15384616,0.23076923,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.30769232,0.23076923,0.15384616,0.3846154,0.23076923,0.3846154,0.30769232,0.3846154,0.23076923,0.15384616,0.3846154,0.3846154,0.3846154,0.3846154,0.46153846,0.61538464,0.46153846,0.46153846,,0.61538464,0.53846157,0.6923077,0.6923077,0.53846157,0.61538464,0.53846157,0.53846157,0.9230769,0.7692308,0.61538464,0.7692308,0.84615386,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.15384616,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.23076923,0.15384616,0.07692308,0.23076923,0.23076923,0.07692308,0.07692308,0.30769232,0.15384616,0.23076923,0.15384616,0.23076923,0.30769232,0.30769232,0.23076923,0.30769232,0.30769232,0.30769232,0.3846154,0.23076923,0.3846154,0.3846154,0.53846157,0.3846154,0.15384616,0.3846154,0.3846154,0.3846154,,0.3846154,0.3846154,0.3846154,0.46153846,0.84615386,0.61538464,0.61538464,0.53846157,0.7692308,0.6923077,0.7692308,0.84615386,0.6923077,0.7692308,0.9230769,0.84615386,0.84615386,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.07692308,0.15384616,0.0,0.15384616,0.0,0.0,0.0,0.07692308,0.0,0.23076923,0.07692308,0.23076923,0.3846154,0.15384616,0.46153846,0.15384616,0.46153846,0.30769232,0.23076923,0.15384616,0.23076923,0.46153846,0.53846157,0.30769232,0.46153846,0.1923077,0.3846154,0.46153846,0.61538464,,0.61538464,0.61538464,0.7692308,0.7692308,0.6923077,0.7692308,0.7692308,0.7692308,0.9230769,0.7692308,0.84615386,0.9230769,0.7692308,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.0,0.0,0.0,0.15384616,0.0,0.15384616,0.15384616,0.23076923,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.23076923,0.23076923,0.15384616,0.3846154,0.23076923,0.3846154,0.30769232,0.30769232,0.23076923,0.23076923,0.30769232,0.30769232,0.3846154,0.30769232,0.3846154,0.61538464,0.3846154,0.30769232,0.30769232,0.61538464,0.3846154,,0.53846157,0.53846157,0.61538464,0.46153846,0.53846157,0.7307692,0.7692308,0.61538464,0.7692308,0.7692308,0.9230769,1.0,0.84615386,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.23076923,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.30769232,0.30769232,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.3846154,0.23076923,0.0,0.30769232,0.15384616,0.3846154,0.3846154,0.23076923,0.61538464,0.30769232,0.30769232,0.30769232,0.61538464,0.3846154,0.46153846,,0.46153846,0.61538464,0.3846154,0.46153846,0.61538464,0.6923077,0.61538464,0.6923077,0.6923077,0.9230769,0.9230769,0.7692308,0.9230769,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.23076923,0.23076923,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.23076923,0.30769232,0.46153846,0.23076923,0.3846154,0.23076923,0.3846154,0.46153846,0.53846157,0.23076923,0.46153846,0.53846157,,0.61538464,0.53846157,0.61538464,0.61538464,0.7692308,0.6923077,0.7692308,1.0,0.7692308,0.6923077,0.84615386,0.7692308,0.9230769,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.0,0.0,0.0,0.0,0.15384616,0.15384616,0.15384616,0.07692308,0.0,0.15384616,0.07692308,0.0,0.0,0.15384616,0.07692308,0.07692308,0.07692308,0.15384616,0.23076923,0.07692308,0.23076923,0.23076923,0.23076923,0.15384616,0.30769232,0.23076923,0.30769232,0.23076923,0.30769232,0.30769232,0.0,0.30769232,0.23076923,0.3846154,0.15384616,0.23076923,0.3846154,0.3846154,0.3846154,,0.61538464,0.53846157,0.46153846,0.6923077,0.61538464,0.6923077,0.61538464,0.6923077,0.6923077,0.9230769,0.84615386,0.84615386,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.07692308,0.23076923,0.07692308,0.23076923,0.0,0.23076923,0.23076923,0.23076923,0.23076923,0.0,0.23076923,0.46153846,0.3846154,0.53846157,0.0,0.3846154,0.23076923,0.23076923,0.30769232,0.23076923,0.46153846,0.3846154,0.30769232,0.53846157,0.61538464,0.46153846,0.3846154,,0.46153846,0.61538464,0.6923077,0.3846154,0.6923077,0.46153846,0.6923077,0.6923077,0.84615386,0.84615386,0.84615386,0.9230769,0.9230769,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.23076923,0.15384616,0.23076923,0.07692308,0.07692308,0.23076923,0.30769232,0.30769232,0.30769232,0.15384616,0.3846154,0.23076923,0.30769232,0.23076923,0.30769232,0.46153846,0.3846154,0.23076923,0.46153846,0.53846157,0.3846154,0.46153846,0.53846157,,0.53846157,0.7692308,0.61538464,0.7692308,0.6923077,0.6923077,0.61538464,0.84615386,0.7692308,0.9230769,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.07692308,0.0,0.23076923,0.15384616,0.30769232,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.07692308,0.3846154,0.15384616,0.07692308,0.46153846,0.23076923,0.23076923,0.07692308,0.46153846,0.23076923,0.26923078,0.3846154,0.3846154,0.53846157,0.3846154,0.46153846,,0.7692308,0.61538464,0.7692308,0.6923077,0.7692308,0.84615386,0.84615386,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.07692308,0.0,0.0,0.0,0.0,0.15384616,0.0,0.15384616,0.15384616,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.30769232,0.23076923,0.30769232,0.30769232,0.23076923,0.23076923,,0.23076923,0.84615386,0.3846154,0.3846154,0.46153846,1.0,0.61538464,0.61538464,0.84615386,0.84615386,0.9230769
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.23076923,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.23076923,0.15384616,0.30769232,0.15384616,0.23076923,0.0,0.07692308,0.3846154,0.30769232,0.07692308,0.3846154,0.3846154,0.30769232,0.3846154,0.61538464,0.3846154,0.3846154,0.7692308,,0.7692308,0.46153846,0.61538464,0.61538464,0.84615386,0.6923077,0.84615386,0.9230769,0.9230769,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.07692308,0.07692308,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.23076923,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.30769232,0.23076923,0.30769232,0.30769232,0.23076923,0.23076923,0.15384616,0.23076923,,0.3846154,0.3846154,0.46153846,1.0,0.61538464,0.61538464,0.84615386,0.84615386,0.9230769
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.23076923,0.07692308,0.15384616,0.07692308,0.07692308,0.0,0.07692308,0.15384616,0.15384616,0.15384616,0.23076923,0.07692308,0.15384616,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.30769232,0.0,0.3846154,0.53846157,0.30769232,0.30769232,0.61538464,0.53846157,0.61538464,,0.46153846,0.46153846,0.61538464,0.61538464,0.7692308,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.23076923,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.0,0.23076923,0.07692308,0.07692308,0.0,0.30769232,0.07692308,0.07692308,0.07692308,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.61538464,0.3846154,0.61538464,0.53846157,,0.6923077,0.7692308,0.7692308,0.84615386,0.9230769,0.9230769,0.9230769
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.0,0.15384616,0.0,0.07692308,0.15384616,0.23076923,0.15384616,0.0,0.23076923,0.23076923,0.0,0.07692308,0.30769232,0.30769232,0.30769232,0.3846154,0.15384616,0.53846157,0.3846154,0.53846157,0.53846157,0.30769232,,0.53846157,1.0,1.0,0.9230769,0.9230769,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.07692308,0.15384616,0.0,0.07692308,0.07692308,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.07692308,0.15384616,0.15384616,0.23076923,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.0,0.15384616,0.0,0.3846154,0.23076923,0.46153846,,0.53846157,0.53846157,0.84615386,0.84615386,0.84615386
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.15384616,0.07692308,0.0,0.0,0.15384616,0.07692308,0.0,0.07692308,0.23076923,0.15384616,0.15384616,0.23076923,0.0,0.3846154,0.30769232,0.3846154,0.3846154,0.23076923,0.0,0.46153846,,0.84615386,0.9230769,0.9230769,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.15384616,0.07692308,0.0,0.0,0.15384616,0.07692308,0.0,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.0,0.3846154,0.15384616,0.3846154,0.23076923,0.15384616,0.0,0.46153846,0.15384616,,0.6923077,0.6923077,0.6923077
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.15384616,0.07692308,0.15384616,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.30769232,,0.8076923,0.84615386
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.15384616,0.07692308,0.15384616,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.30769232,0.1923077,,0.7692308
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.0,0.15384616,0.0,0.30769232,0.15384616,0.23076923,
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
index 6e0e0a7d6dd4944e52e7e5f70e0e706dd14efe7e..324e0a095e01195819399a036228839642ac31cb 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Regression | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..f50b422faac73427b5cf1135cb6cd8ab575cd16c
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_regression/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2478.0,"score":0.907,"rank":1.6,"hrank":1.43,"imp":0.476,"train_time":25.12,"predict_time":11.643,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1402\/-145","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1994.0,"score":0.63,"rank":7.2,"hrank":5.92,"imp":3.186,"train_time":739.1,"predict_time":11.65,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+776\/-70","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1971.0,"score":0.637,"rank":7.8,"hrank":6.01,"imp":3.282,"train_time":739.1,"predict_time":0.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+827\/-105","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1949.0,"score":0.74,"rank":8.4,"hrank":2.46,"imp":1.772,"train_time":0.87,"predict_time":0.204,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1538\/-285","imputed":false},{"position":4,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1902.0,"score":0.65,"rank":9.8,"hrank":3.67,"imp":4.09,"train_time":374.99,"predict_time":3.566,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+624\/-16","imputed":false},{"position":5,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1866.0,"score":0.524,"rank":11.0,"hrank":9.83,"imp":4.264,"train_time":2552.31,"predict_time":4.522,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+607\/-39","imputed":false},{"position":6,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1860.0,"score":0.515,"rank":11.2,"hrank":9.2,"imp":4.214,"train_time":2.78,"predict_time":0.702,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+658\/-45","imputed":false},{"position":7,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1822.0,"score":0.481,"rank":12.6,"hrank":8.19,"imp":5.17,"train_time":474.95,"predict_time":11.837,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+468\/-0","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1806.0,"score":0.624,"rank":13.2,"hrank":4.36,"imp":3.512,"train_time":0.47,"predict_time":0.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1206\/-342","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1801.0,"score":0.468,"rank":13.4,"hrank":12.22,"imp":4.842,"train_time":2068.17,"predict_time":1.355,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+588\/-75","imputed":false},{"position":10,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1790.0,"score":0.528,"rank":13.8,"hrank":3.35,"imp":4.055,"train_time":4529.93,"predict_time":357.201,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+995\/-190","imputed":false},{"position":11,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1780.0,"score":0.468,"rank":14.2,"hrank":11.71,"imp":5.289,"train_time":4.86,"predict_time":1.881,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+581\/-118","imputed":false},{"position":12,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1742.0,"score":0.498,"rank":15.8,"hrank":7.4,"imp":4.355,"train_time":4529.93,"predict_time":39.54,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+911\/-180","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1723.0,"score":0.474,"rank":16.6,"hrank":11.87,"imp":4.298,"train_time":0.66,"predict_time":0.177,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+784\/-112","imputed":false},{"position":14,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1710.0,"score":0.462,"rank":17.2,"hrank":10.54,"imp":4.597,"train_time":1.59,"predict_time":0.506,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+787\/-246","imputed":false},{"position":15,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1635.0,"score":0.335,"rank":20.8,"hrank":16.55,"imp":5.935,"train_time":1.97,"predict_time":0.651,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+498\/-134","imputed":false},{"position":16,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1630.0,"score":0.337,"rank":21.0,"hrank":16.08,"imp":6.087,"train_time":2068.17,"predict_time":0.068,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+549\/-171","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1622.0,"score":0.251,"rank":21.4,"hrank":17.06,"imp":7.744,"train_time":223.05,"predict_time":9.706,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+344\/-106","imputed":false},{"position":18,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1610.0,"score":0.412,"rank":22.0,"hrank":11.6,"imp":5.141,"train_time":41.51,"predict_time":38.297,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+763\/-196","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1606.0,"score":0.227,"rank":22.2,"hrank":17.96,"imp":7.913,"train_time":223.05,"predict_time":0.793,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+376\/-112","imputed":false},{"position":20,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1602.0,"score":0.334,"rank":22.4,"hrank":18.29,"imp":6.016,"train_time":2552.31,"predict_time":0.201,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+550\/-218","imputed":false},{"position":21,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1594.0,"score":0.303,"rank":22.8,"hrank":20.52,"imp":6.408,"train_time":438.56,"predict_time":0.52,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+388\/-58","imputed":false},{"position":22,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1590.0,"score":0.176,"rank":23.0,"hrank":14.3,"imp":8.15,"train_time":917.62,"predict_time":0.422,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+321\/-79","imputed":false},{"position":23,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1557.0,"score":0.161,"rank":24.6,"hrank":12.93,"imp":8.303,"train_time":917.62,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+314\/-94","imputed":false},{"position":24,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1546.0,"score":0.28,"rank":25.2,"hrank":19.69,"imp":7.074,"train_time":2410.31,"predict_time":0.581,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+470\/-184","imputed":false},{"position":25,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1519.0,"score":0.234,"rank":26.6,"hrank":21.16,"imp":7.704,"train_time":2410.31,"predict_time":0.045,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+446\/-178","imputed":false},{"position":26,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1496.0,"score":0.177,"rank":27.8,"hrank":26.39,"imp":7.379,"train_time":438.56,"predict_time":0.035,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+300\/-75","imputed":false},{"position":27,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1486.0,"score":0.155,"rank":28.4,"hrank":21.55,"imp":8.948,"train_time":2.73,"predict_time":0.01,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+334\/-225","imputed":false},{"position":28,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1475.0,"score":0.191,"rank":29.0,"hrank":26.7,"imp":7.245,"train_time":6.87,"predict_time":0.098,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+391\/-136","imputed":false},{"position":29,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1442.0,"score":0.102,"rank":30.8,"hrank":29.57,"imp":8.739,"train_time":289.35,"predict_time":1.798,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+284\/-109","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1431.0,"score":0.097,"rank":31.4,"hrank":29.83,"imp":8.805,"train_time":289.35,"predict_time":0.36,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+312\/-140","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1397.0,"score":0.089,"rank":33.2,"hrank":32.6,"imp":8.165,"train_time":2150.19,"predict_time":65.934,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+202\/-41","imputed":false},{"position":32,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1393.0,"score":0.184,"rank":33.4,"hrank":19.49,"imp":7.945,"train_time":24.62,"predict_time":8.835,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+634\/-299","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1381.0,"score":0.122,"rank":34.0,"hrank":25.43,"imp":12.794,"train_time":125.2,"predict_time":0.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+414\/-474","imputed":false},{"position":34,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1347.0,"score":0.171,"rank":35.8,"hrank":27.75,"imp":9.123,"train_time":0.51,"predict_time":0.126,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+446\/-336","imputed":false},{"position":35,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1343.0,"score":0.108,"rank":36.0,"hrank":30.03,"imp":12.83,"train_time":125.2,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+384\/-461","imputed":false},{"position":36,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1316.0,"score":0.051,"rank":37.4,"hrank":36.66,"imp":9.289,"train_time":2150.19,"predict_time":18.752,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+183\/-127","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1309.0,"score":0.046,"rank":37.8,"hrank":36.05,"imp":9.733,"train_time":3.22,"predict_time":0.039,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+269\/-144","imputed":false},{"position":38,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1298.0,"score":0.059,"rank":38.4,"hrank":34.78,"imp":9.9,"train_time":0.57,"predict_time":0.073,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+238\/-247","imputed":false},{"position":39,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1283.0,"score":0.037,"rank":39.2,"hrank":37.76,"imp":9.965,"train_time":17.81,"predict_time":0.4,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+167\/-245","imputed":false},{"position":40,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1251.0,"score":0.013,"rank":41.0,"hrank":40.24,"imp":10.083,"train_time":0.66,"predict_time":0.213,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+212\/-129","imputed":false},{"position":41,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1247.0,"score":0.006,"rank":41.2,"hrank":39.9,"imp":10.7,"train_time":2.32,"predict_time":0.093,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+195\/-177","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1236.0,"score":0.068,"rank":41.8,"hrank":37.04,"imp":13.809,"train_time":0.74,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+309\/-451","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1228.0,"score":0.066,"rank":42.2,"hrank":39.81,"imp":9.994,"train_time":4108.24,"predict_time":0.479,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+192\/-211","imputed":false},{"position":44,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1193.0,"score":0.001,"rank":44.0,"hrank":43.23,"imp":10.73,"train_time":17.81,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-212","imputed":false},{"position":45,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1185.0,"score":0.026,"rank":44.4,"hrank":41.6,"imp":11.402,"train_time":6.54,"predict_time":0.284,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+296\/-295","imputed":false},{"position":46,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1181.0,"score":0.031,"rank":44.6,"hrank":42.4,"imp":11.081,"train_time":4108.24,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+196\/-277","imputed":false},{"position":47,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1177.0,"score":0.027,"rank":44.8,"hrank":42.56,"imp":11.917,"train_time":50.36,"predict_time":0.232,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+273\/-316","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1152.0,"score":0.026,"rank":46.0,"hrank":43.8,"imp":12.063,"train_time":50.36,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+272\/-335","imputed":false},{"position":49,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1118.0,"score":0.069,"rank":47.6,"hrank":37.0,"imp":12.244,"train_time":8.68,"predict_time":7.525,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+438\/-412","imputed":false},{"position":50,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1118.0,"score":0.063,"rank":47.6,"hrank":34.83,"imp":15.412,"train_time":1069.96,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+394\/-540","imputed":false},{"position":51,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1114.0,"score":0.0,"rank":47.8,"hrank":47.45,"imp":11.829,"train_time":79.96,"predict_time":0.571,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-240","imputed":false},{"position":52,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1108.0,"score":0.0,"rank":48.1,"hrank":47.52,"imp":11.989,"train_time":79.96,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-257","imputed":false},{"position":53,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1049.0,"score":0.0,"rank":50.8,"hrank":49.61,"imp":13.926,"train_time":1.53,"predict_time":0.008,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+168\/-355","imputed":false},{"position":54,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1026.0,"score":0.0,"rank":51.8,"hrank":50.36,"imp":15.231,"train_time":14.05,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-305","imputed":false},{"position":55,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1007.0,"score":0.036,"rank":52.6,"hrank":47.95,"imp":16.025,"train_time":1069.96,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+304\/-570","imputed":false},{"position":56,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":52.9,"hrank":52.59,"imp":13.109,"train_time":0.42,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-226","imputed":false},{"position":57,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":988.0,"score":0.048,"rank":53.4,"hrank":47.5,"imp":16.456,"train_time":2.37,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+336\/-644","imputed":false},{"position":58,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":988.0,"score":0.0,"rank":53.4,"hrank":51.29,"imp":13.952,"train_time":0.86,"predict_time":0.099,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+276\/-538","imputed":false},{"position":59,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":884.0,"score":0.0,"rank":57.2,"hrank":56.6,"imp":16.484,"train_time":93.51,"predict_time":19.996,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-630","imputed":false},{"position":60,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":860.0,"score":0.0,"rank":58.0,"hrank":57.41,"imp":16.012,"train_time":317.48,"predict_time":0.719,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-442","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":828.0,"score":0.0,"rank":59.0,"hrank":58.58,"imp":16.565,"train_time":317.48,"predict_time":0.127,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-470","imputed":false},{"position":62,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":772.0,"score":0.0,"rank":60.6,"hrank":60.22,"imp":20.22,"train_time":142.18,"predict_time":1.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-461","imputed":false},{"position":63,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":701.0,"score":0.0,"rank":62.4,"hrank":61.83,"imp":19.882,"train_time":1.33,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-726","imputed":false},{"position":64,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":683.0,"score":0.0,"rank":62.8,"hrank":62.55,"imp":23.8,"train_time":5782.12,"predict_time":8.567,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-799","imputed":false},{"position":65,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":646.0,"score":0.0,"rank":63.6,"hrank":63.34,"imp":23.989,"train_time":5782.12,"predict_time":2.139,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-861","imputed":false},{"position":66,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":617.0,"score":0.0,"rank":64.2,"hrank":64.01,"imp":22.596,"train_time":142.18,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-514","imputed":false},{"position":67,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":512.0,"score":0.0,"rank":66.0,"hrank":65.89,"imp":27.279,"train_time":113.4,"predict_time":1.615,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-900","imputed":false},{"position":68,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":402.0,"score":0.0,"rank":67.4,"hrank":67.22,"imp":29.723,"train_time":0.06,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-1194","imputed":false},{"position":69,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":65.0,"score":0.0,"rank":69.8,"hrank":69.8,"imp":40.798,"train_time":48.35,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-2248","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-192.0,"score":0.0,"rank":71.0,"hrank":70.99,"imp":41.14,"train_time":48.35,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+218\/-2864","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-347.0,"score":0.0,"rank":71.6,"hrank":71.6,"imp":44.075,"train_time":0.13,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+181\/-2671","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/pareto_front_explorer.html
index e151f986b6cb776adc98d33626e7098c2bc7931e..354f4352e1dd4220e26e32d8e1713522988a1cab 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0103037461,"imp":8.9479539866,"elo":1485.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0161908394,"imp":8.3030672997,"elo":1557.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"imp":8.1503948491,"elo":1589.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"imp":13.8090022874,"elo":1235.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0408241222,"imp":12.8301298042,"elo":1342.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"imp":12.7943126543,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"imp":16.4556369343,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0051447125,"imp":16.0253582032,"elo":1007.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0853557077,"imp":15.4116177291,"elo":1118.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0992770116,"imp":13.9524159906,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0186999882,"imp":12.06324568,"elo":1152.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2319944846,"imp":11.9166741153,"elo":1176.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1256610016,"imp":19.8824164433,"elo":700.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1272848026,"imp":16.5650505415,"elo":828.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7187718544,"imp":16.0119645154,"elo":859.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2133657073,"imp":10.0826385426,"elo":1250.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7933065809,"imp":7.9131722928,"elo":1605.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":9.7062590878,"imp":7.7439646534,"elo":1622.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.8350020269,"imp":7.9445989236,"elo":1392.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.097838007,"imp":7.2446277873,"elo":1474.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2007087367,"imp":6.0155505108,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":4.5218047762,"imp":4.2641087237,"elo":1865.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.6513221504,"imp":5.9350713717,"elo":1634.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8805802898,"imp":5.2894589865,"elo":1780.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0078855342,"imp":13.9256196498,"elo":1049.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1934792264,"imp":10.7303129531,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4003930169,"imp":9.9648302652,"elo":1283.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"imp":13.1089370831,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"imp":11.9892701847,"elo":1107.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5707803637,"imp":11.8289448844,"elo":1114.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2839085103,"imp":11.4020603961,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0682382085,"imp":6.087337608,"elo":1630.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.3549179469,"imp":4.8421550057,"elo":1800.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5059595331,"imp":4.5971768753,"elo":1710.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6894524158,"imp":3.2815549654,"elo":1970.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"imp":3.185810595,"elo":1994.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.5250451841,"imp":12.2440158789,"elo":1118.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"imp":5.1411670708,"elo":1610.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.540415138,"imp":4.3545991568,"elo":1741.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":357.2008760846,"imp":4.0547168602,"elo":1790.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1767387784,"imp":4.2983424742,"elo":1723.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":11.6433178755,"imp":0.4756811693,"elo":2478.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1028500957,"imp":3.5123157709,"elo":1805.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.038732221,"imp":9.732882352,"elo":1309.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0448228661,"imp":7.7035348907,"elo":1518.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5807768467,"imp":7.0742201117,"elo":1545.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"imp":4.214387882,"elo":1860.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042135993,"imp":1.7719216017,"elo":1948.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6147142423,"imp":27.2788510175,"elo":511.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.139339258,"imp":23.9886884373,"elo":646.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"imp":23.7997636743,"elo":683.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1256335989,"imp":9.1227548353,"elo":1346.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0292963248,"imp":15.2314315415,"elo":1026.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0365087137,"imp":11.0812079304,"elo":1180.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4787581258,"imp":9.9939380427,"elo":1228.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.073497228,"imp":9.8998436121,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3603568941,"imp":8.8046523626,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.7976430646,"imp":8.7389814156,"elo":1441.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":19.9963063451,"imp":16.4844284378,"elo":884.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":18.7518939426,"imp":9.2887498658,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.9338217731,"imp":8.1653547355,"elo":1396.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.0932754819,"imp":10.6996583222,"elo":1246.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0348964576,"imp":7.3787958084,"elo":1496.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.5201417769,"imp":6.4083302989,"elo":1593.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0103037461,"x_train":2.7349656008,"imp":8.9479539866,"elo":1485.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0161908394,"x_train":917.6176937357,"imp":8.3030672997,"elo":1557.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"x_train":917.6176937357,"imp":8.1503948491,"elo":1589.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"x_train":0.7437448515,"imp":13.8090022874,"elo":1235.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0408241222,"x_train":125.1955265319,"imp":12.8301298042,"elo":1342.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":125.1955265319,"imp":12.7943126543,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"x_train":2.3730743936,"imp":16.4556369343,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0051447125,"x_train":1069.963809484,"imp":16.0253582032,"elo":1007.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0853557077,"x_train":1069.963809484,"imp":15.4116177291,"elo":1118.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0992770116,"x_train":0.8553953885,"imp":13.9524159906,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0186999882,"x_train":50.361853205,"imp":12.06324568,"elo":1152.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2319944846,"x_train":50.361853205,"imp":11.9166741153,"elo":1176.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1256610016,"x_train":1.3312352951,"imp":19.8824164433,"elo":700.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1272848026,"x_train":317.4820332824,"imp":16.5650505415,"elo":828.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7187718544,"x_train":317.4820332824,"imp":16.0119645154,"elo":859.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2133657073,"x_train":0.6567304564,"imp":10.0826385426,"elo":1250.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7933065809,"x_train":223.0498746731,"imp":7.9131722928,"elo":1605.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":9.7062590878,"x_train":223.0498746731,"imp":7.7439646534,"elo":1622.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.8350020269,"x_train":24.6223741935,"imp":7.9445989236,"elo":1392.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.097838007,"x_train":6.870566258,"imp":7.2446277873,"elo":1474.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2007087367,"x_train":2552.3091253937,"imp":6.0155505108,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":4.5218047762,"x_train":2552.3091253937,"imp":4.2641087237,"elo":1865.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.6513221504,"x_train":1.9712126532,"imp":5.9350713717,"elo":1634.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8805802898,"x_train":4.8565548774,"imp":5.2894589865,"elo":1780.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0078855342,"x_train":1.5311965098,"imp":13.9256196498,"elo":1049.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1934792264,"x_train":17.8122979219,"imp":10.7303129531,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4003930169,"x_train":17.8122979219,"imp":9.9648302652,"elo":1283.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":0.4238705302,"imp":13.1089370831,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":79.9551879424,"imp":11.9892701847,"elo":1107.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5707803637,"x_train":79.9551879424,"imp":11.8289448844,"elo":1114.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2839085103,"x_train":6.5430745829,"imp":11.4020603961,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0682382085,"x_train":2068.1710397435,"imp":6.087337608,"elo":1630.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.3549179469,"x_train":2068.1710397435,"imp":4.8421550057,"elo":1800.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5059595331,"x_train":1.5872114628,"imp":4.5971768753,"elo":1710.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6894524158,"x_train":739.1047147274,"imp":3.2815549654,"elo":1970.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"x_train":739.1047147274,"imp":3.185810595,"elo":1994.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.5250451841,"x_train":8.6828160405,"imp":12.2440158789,"elo":1118.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"x_train":41.5053698771,"imp":5.1411670708,"elo":1610.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.540415138,"x_train":4529.9305925825,"imp":4.3545991568,"elo":1741.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":357.2008760846,"x_train":4529.9305925825,"imp":4.0547168602,"elo":1790.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1767387784,"x_train":0.6611709932,"imp":4.2983424742,"elo":1723.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":11.6433178755,"x_train":25.1196552531,"imp":0.4756811693,"elo":2478.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1028500957,"x_train":0.471867489,"imp":3.5123157709,"elo":1805.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.038732221,"x_train":3.2167024044,"imp":9.732882352,"elo":1309.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0448228661,"x_train":2410.3120769872,"imp":7.7035348907,"elo":1518.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5807768467,"x_train":2410.3120769872,"imp":7.0742201117,"elo":1545.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"x_train":2.7774434997,"imp":4.214387882,"elo":1860.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042135993,"x_train":0.8699671047,"imp":1.7719216017,"elo":1948.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6147142423,"x_train":113.401275248,"imp":27.2788510175,"elo":511.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.139339258,"x_train":5782.1195021638,"imp":23.9886884373,"elo":646.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"x_train":5782.1195021638,"imp":23.7997636743,"elo":683.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1256335989,"x_train":0.5118384438,"imp":9.1227548353,"elo":1346.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0292963248,"x_train":14.0524525282,"imp":15.2314315415,"elo":1026.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0365087137,"x_train":4108.2381109512,"imp":11.0812079304,"elo":1180.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4787581258,"x_train":4108.2381109512,"imp":9.9939380427,"elo":1228.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.073497228,"x_train":0.5708322449,"imp":9.8998436121,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3603568941,"x_train":289.3497533723,"imp":8.8046523626,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.7976430646,"x_train":289.3497533723,"imp":8.7389814156,"elo":1441.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":19.9963063451,"x_train":93.5081427033,"imp":16.4844284378,"elo":884.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":18.7518939426,"x_train":2150.1856290327,"imp":9.2887498658,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.9338217731,"x_train":2150.1856290327,"imp":8.1653547355,"elo":1396.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.0932754819,"x_train":2.3191009865,"imp":10.6996583222,"elo":1246.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0348964576,"x_train":438.5586959956,"imp":7.3787958084,"elo":1496.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.5201417769,"x_train":438.5586959956,"imp":6.4083302989,"elo":1593.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
index 52e40c5d9e0a8e9cdbca7d942eb34af0713e6134..cca1cb894f8ff3dc39f48a03614cb74fa369a21c 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":2.7349656008,"imp":8.948053525,"elo":1477.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":8.045705742,"imp":8.7006398529,"elo":1504.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":21.5859747308,"imp":8.4569991571,"elo":1560.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":44.5235265964,"imp":8.3683097311,"elo":1569.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":109.7648010245,"imp":8.2422114531,"elo":1582.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":222.7546225359,"imp":8.2000098965,"elo":1590.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":456.5709140465,"imp":8.1667213574,"elo":1597.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":683.5034177869,"imp":8.1595704139,"elo":1604.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":917.6176937357,"imp":8.1505936003,"elo":1611.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.7437448515,"imp":13.8091021689,"elo":1228.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.4147629158,"imp":13.8484288322,"elo":1218.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.1548755046,"imp":13.7711647599,"elo":1244.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":6.0549410867,"imp":13.5526854761,"elo":1273.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":15.7085131241,"imp":13.2469953776,"elo":1311.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":31.5530586222,"imp":13.0195886576,"elo":1349.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":62.4108808015,"imp":12.872512035,"elo":1371.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":93.2866322036,"imp":12.8120249297,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":125.1955265319,"imp":12.7921068709,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.3730743936,"imp":16.4557237796,"elo":994.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":9.4785326555,"imp":16.1785258957,"elo":1004.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":29.4810408518,"imp":15.9381690947,"elo":1015.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":54.9981205321,"imp":15.7982325684,"elo":1030.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":136.1756596671,"imp":15.5973941026,"elo":1069.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":274.1891727198,"imp":15.5230399323,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":533.5366593544,"imp":15.452806981,"elo":1112.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":788.165832343,"imp":15.4287038401,"elo":1124.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1069.963809484,"imp":15.4124099857,"elo":1135.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.8553953885,"imp":13.9525136374,"elo":967.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.1442763752,"imp":13.3237238784,"elo":1021.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.877416167,"imp":12.6695881526,"elo":1083.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1018975982,"imp":12.4540725039,"elo":1104.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.8290142883,"imp":12.0965906535,"elo":1142.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.9805618038,"imp":12.0194659771,"elo":1154.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":25.3410754681,"imp":11.9477961033,"elo":1172.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.7076806608,"imp":11.9239826929,"elo":1177.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":50.361853205,"imp":11.9175188382,"elo":1182.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.3312352951,"imp":19.8825043615,"elo":683.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.8523145707,"imp":19.6916039351,"elo":715.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":7.8648273493,"imp":19.2739055273,"elo":748.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":16.1612343105,"imp":18.9803391272,"elo":759.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":38.2613776387,"imp":17.8817086745,"elo":795.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":77.1343579507,"imp":17.1783670591,"elo":808.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":156.8008408086,"imp":16.6115391424,"elo":826.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":236.1441179195,"imp":16.2368590085,"elo":858.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":317.4820332824,"imp":16.0120587707,"elo":883.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.6567304564,"imp":10.0827395258,"elo":1246.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":1.8710214047,"imp":9.9377315646,"elo":1270.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":5.2206722708,"imp":9.1620265388,"elo":1399.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":10.6284575779,"imp":8.7076228752,"elo":1463.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":27.3782330474,"imp":8.1388935909,"elo":1563.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":54.9187089858,"imp":7.9706938952,"elo":1601.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":110.3418351857,"imp":7.8406830869,"elo":1626.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":166.090371503,"imp":7.7929819811,"elo":1639.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":223.0498746731,"imp":7.7440680251,"elo":1664.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":24.6223741935,"imp":7.9465233856,"elo":1388.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":6.870566258,"imp":7.2447354962,"elo":1491.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":22.1686321264,"imp":6.7310693457,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.0453859865,"imp":5.8788285376,"elo":1618.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":121.2878502068,"imp":5.3769754646,"elo":1700.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":291.6044227147,"imp":4.7851514077,"elo":1810.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":631.2850224038,"imp":4.5443068635,"elo":1840.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1243.1111273612,"imp":4.3770308899,"elo":1869.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1862.6240988571,"imp":4.298791846,"elo":1893.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2552.3091253937,"imp":4.2642248374,"elo":1898.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":1.9712126532,"imp":5.9351831745,"elo":1645.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":4.8565548774,"imp":5.2895686255,"elo":1818.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.5311965098,"imp":13.9257122688,"elo":1057.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":5.1565448569,"imp":11.5962887488,"elo":1169.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":14.9849409603,"imp":10.0459712109,"elo":1261.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":17.8122979219,"imp":9.9649292888,"elo":1274.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4238705302,"imp":13.1090349978,"elo":1002.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.8229555909,"imp":12.7357834902,"elo":1040.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":2.0559128426,"imp":12.3965155326,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":4.0681999454,"imp":12.1045019075,"elo":1094.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":9.9071084465,"imp":11.9373680623,"elo":1117.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":19.6617455737,"imp":11.8680574316,"elo":1124.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":39.5175987803,"imp":11.8408256094,"elo":1131.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":59.4870987604,"imp":11.8336701035,"elo":1137.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":79.9551879424,"imp":11.8290459506,"elo":1135.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.5430745829,"imp":11.4021616834,"elo":1188.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":16.387957176,"imp":9.0872668175,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":45.678712897,"imp":6.8982447724,"elo":1563.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":100.5864418887,"imp":6.1036598702,"elo":1638.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":253.2316993705,"imp":5.5083757333,"elo":1724.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":513.1174708444,"imp":5.211058434,"elo":1759.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1026.1644887093,"imp":5.0189052637,"elo":1790.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1544.7351451859,"imp":4.9248470231,"elo":1820.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2068.1710397435,"imp":4.8422678218,"elo":1833.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.5872114628,"imp":4.5972922799,"elo":1720.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.1930136195,"imp":4.4954936788,"elo":1732.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.3030487246,"imp":4.2817235886,"elo":1744.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":34.3657315438,"imp":3.9788567334,"elo":1807.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":91.1547591648,"imp":3.6432521816,"elo":1933.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":182.7153391231,"imp":3.4430381734,"elo":1961.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":366.5547593939,"imp":3.2949600688,"elo":2018.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":550.802457324,"imp":3.2196038239,"elo":2046.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":739.1047147274,"imp":3.1862333181,"elo":2061.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":8.6828160405,"imp":12.2441147822,"elo":1134.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.5053698771,"imp":5.1412864523,"elo":1615.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.019350737,"imp":5.0028139527,"elo":1642.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":119.7137023169,"imp":4.8377133521,"elo":1663.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":218.048215922,"imp":4.7224256158,"elo":1684.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":550.8419093004,"imp":4.4758639758,"elo":1728.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1115.34725308,"imp":4.2994519026,"elo":1756.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2246.8238340239,"imp":4.1625090429,"elo":1782.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3382.9355985564,"imp":4.0875733992,"elo":1791.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4529.9305925825,"imp":4.0547168602,"elo":1800.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.6611709932,"imp":4.2984610916,"elo":1751.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":25.1196552531,"imp":0.4758006822,"elo":2500.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.471867489,"imp":3.5124324077,"elo":1815.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":3.2167024044,"imp":9.7329856034,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":11.7992281299,"imp":9.1947777236,"elo":1418.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":46.940037227,"imp":8.4487551977,"elo":1458.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":112.2128891191,"imp":7.9823169354,"elo":1505.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":298.6395201823,"imp":7.5780381651,"elo":1528.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":596.2505830634,"imp":7.3583930626,"elo":1542.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1163.2433543614,"imp":7.2040262897,"elo":1555.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1770.0748555561,"imp":7.1312564436,"elo":1567.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2410.3120769872,"imp":7.0743284296,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.7774434997,"imp":4.2145025488,"elo":1885.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":0.8699671047,"imp":1.7853424303,"elo":1928.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":113.401275248,"imp":27.2789297045,"elo":500.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":141.7238706247,"imp":27.1544799482,"elo":535.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":221.1962991478,"imp":26.854399481,"elo":551.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":328.7844244998,"imp":26.3300775969,"elo":567.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":736.9725789769,"imp":25.4060212814,"elo":596.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1485.4343663821,"imp":24.562354425,"elo":622.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2941.8656957704,"imp":24.2062265581,"elo":638.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4341.8172896327,"imp":23.9692582481,"elo":671.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5782.1195021638,"imp":23.7998469089,"elo":706.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.5118384438,"imp":9.1228658323,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.0524525282,"imp":15.2315343973,"elo":1023.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":35.7395534737,"imp":14.3339737912,"elo":1043.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":93.9257318856,"imp":12.9646971276,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":201.2335467269,"imp":12.1017547455,"elo":1164.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":481.4580064735,"imp":11.1402059021,"elo":1201.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1015.3213795664,"imp":10.7333183709,"elo":1216.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2031.0046161997,"imp":10.312716384,"elo":1231.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3088.2903635944,"imp":10.1160578174,"elo":1238.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4108.2381109512,"imp":9.9940462779,"elo":1255.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.5708322449,"imp":9.8999433384,"elo":1295.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.4574329869,"imp":9.7581492837,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":7.3610615249,"imp":9.3444866285,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":13.6386574902,"imp":9.1536785083,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":35.9388283515,"imp":8.9518819388,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":71.5452674162,"imp":8.8329662502,"elo":1434.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":142.0559469286,"imp":8.7587291911,"elo":1448.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":215.0273705137,"imp":8.7490331287,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":289.3497533723,"imp":8.7390847655,"elo":1451.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":93.5081427033,"imp":16.4845177929,"elo":871.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":175.8703779457,"imp":13.5955714626,"elo":995.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":423.2568217882,"imp":24.3807034641,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":836.4112409598,"imp":21.9294986191,"elo":1185.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":2070.7507295344,"imp":8.1871322543,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":2150.1856290327,"imp":8.1654629203,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.3191009865,"imp":10.699762351,"elo":1251.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":4.6217004586,"imp":9.8879494154,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":11.2199797524,"imp":8.783901562,"elo":1376.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":22.2331145434,"imp":7.7860451008,"elo":1457.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":55.3170648503,"imp":7.0074538977,"elo":1521.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":109.9515587293,"imp":6.728401508,"elo":1552.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":218.3385295205,"imp":6.5716820878,"elo":1566.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":327.734067188,"imp":6.4606380858,"elo":1582.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":438.5586959956,"imp":6.4091172112,"elo":1604.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0103037461,"x_train":2.7349656008,"imp":8.948053525,"elo":1477.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0484760077,"x_train":8.045705742,"imp":8.7006398529,"elo":1504.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1185677885,"x_train":21.5859747308,"imp":8.4569991571,"elo":1560.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1738042849,"x_train":44.5235265964,"imp":8.3683097311,"elo":1569.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2388904639,"x_train":109.7648010245,"imp":8.2422114531,"elo":1582.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3111317954,"x_train":222.7546225359,"imp":8.2000098965,"elo":1590.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.398317095,"x_train":456.5709140465,"imp":8.1667213574,"elo":1597.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3902829889,"x_train":683.5034177869,"imp":8.1595704139,"elo":1604.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"x_train":917.6176937357,"imp":8.1505936003,"elo":1611.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"x_train":0.7437448515,"imp":13.8091021689,"elo":1228.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0286415584,"x_train":1.4147629158,"imp":13.8484288322,"elo":1218.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0683284014,"x_train":3.1548755046,"imp":13.7711647599,"elo":1244.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1120636664,"x_train":6.0549410867,"imp":13.5526854761,"elo":1273.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1607817441,"x_train":15.7085131241,"imp":13.2469953776,"elo":1311.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1954148357,"x_train":31.5530586222,"imp":13.0195886576,"elo":1349.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.246440357,"x_train":62.4108808015,"imp":12.872512035,"elo":1371.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2772856864,"x_train":93.2866322036,"imp":12.8120249297,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":125.1955265319,"imp":12.7921068709,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"x_train":2.3730743936,"imp":16.4557237796,"elo":994.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0111616749,"x_train":9.4785326555,"imp":16.1785258957,"elo":1004.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0221705298,"x_train":29.4810408518,"imp":15.9381690947,"elo":1015.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0321482347,"x_train":54.9981205321,"imp":15.7982325684,"elo":1030.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0503208887,"x_train":136.1756596671,"imp":15.5973941026,"elo":1069.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0568078567,"x_train":274.1891727198,"imp":15.5230399323,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0668261402,"x_train":533.5366593544,"imp":15.452806981,"elo":1112.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0718867712,"x_train":788.165832343,"imp":15.4287038401,"elo":1124.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0853557077,"x_train":1069.963809484,"imp":15.4124099857,"elo":1135.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0992770116,"x_train":0.8553953885,"imp":13.9525136374,"elo":967.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0906794725,"x_train":1.1442763752,"imp":13.3237238784,"elo":1021.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0932501865,"x_train":1.877416167,"imp":12.6695881526,"elo":1083.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1085131081,"x_train":3.1018975982,"imp":12.4540725039,"elo":1104.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1369942638,"x_train":6.8290142883,"imp":12.0965906535,"elo":1142.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1923137349,"x_train":12.9805618038,"imp":12.0194659771,"elo":1154.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2461819734,"x_train":25.3410754681,"imp":11.9477961033,"elo":1172.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2514474536,"x_train":37.7076806608,"imp":11.9239826929,"elo":1177.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2319944846,"x_train":50.361853205,"imp":11.9175188382,"elo":1182.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1256610016,"x_train":1.3312352951,"imp":19.8825043615,"elo":683.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1869244318,"x_train":2.8523145707,"imp":19.6916039351,"elo":715.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3616276108,"x_train":7.8648273493,"imp":19.2739055273,"elo":748.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4682651613,"x_train":16.1612343105,"imp":18.9803391272,"elo":759.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.474521655,"x_train":38.2613776387,"imp":17.8817086745,"elo":795.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5649569566,"x_train":77.1343579507,"imp":17.1783670591,"elo":808.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6774236469,"x_train":156.8008408086,"imp":16.6115391424,"elo":826.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7132812118,"x_train":236.1441179195,"imp":16.2368590085,"elo":858.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7187718544,"x_train":317.4820332824,"imp":16.0120587707,"elo":883.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2133657073,"x_train":0.6567304564,"imp":10.0827395258,"elo":1246.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4074904298,"x_train":1.8710214047,"imp":9.9377315646,"elo":1270.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1014164121,"x_train":5.2206722708,"imp":9.1620265388,"elo":1399.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.5286110142,"x_train":10.6284575779,"imp":8.7076228752,"elo":1463.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.2339770555,"x_train":27.3782330474,"imp":8.1388935909,"elo":1563.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.9246780943,"x_train":54.9187089858,"imp":7.9706938952,"elo":1601.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.715416536,"x_train":110.3418351857,"imp":7.8406830869,"elo":1626.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.9190819422,"x_train":166.090371503,"imp":7.7929819811,"elo":1639.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":9.7062590878,"x_train":223.0498746731,"imp":7.7440680251,"elo":1664.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":8.8350020269,"x_train":24.6223741935,"imp":7.9465233856,"elo":1388.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.097838007,"x_train":6.870566258,"imp":7.2447354962,"elo":1491.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2613333852,"x_train":22.1686321264,"imp":6.7310693457,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.58187982,"x_train":60.0453859865,"imp":5.8788285376,"elo":1618.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9883252386,"x_train":121.2878502068,"imp":5.3769754646,"elo":1700.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7805982465,"x_train":291.6044227147,"imp":4.7851514077,"elo":1810.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.3680852352,"x_train":631.2850224038,"imp":4.5443068635,"elo":1840.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.0820243271,"x_train":1243.1111273612,"imp":4.3770308899,"elo":1869.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.9287026871,"x_train":1862.6240988571,"imp":4.298791846,"elo":1893.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.5218047762,"x_train":2552.3091253937,"imp":4.2642248374,"elo":1898.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.6513221504,"x_train":1.9712126532,"imp":5.9351831745,"elo":1645.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8805802898,"x_train":4.8565548774,"imp":5.2895686255,"elo":1818.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0078855342,"x_train":1.5311965098,"imp":13.9257122688,"elo":1057.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1032870492,"x_train":5.1565448569,"imp":11.5962887488,"elo":1169.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3397827375,"x_train":14.9849409603,"imp":10.0459712109,"elo":1261.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4003930169,"x_train":17.8122979219,"imp":9.9649292888,"elo":1274.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":0.4238705302,"imp":13.1090349978,"elo":1002.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1053775796,"x_train":0.8229555909,"imp":12.7357834902,"elo":1040.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1860415756,"x_train":2.0559128426,"imp":12.3965155326,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.271253998,"x_train":4.0681999454,"imp":12.1045019075,"elo":1094.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3828553913,"x_train":9.9071084465,"imp":11.9373680623,"elo":1117.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4327898753,"x_train":19.6617455737,"imp":11.8680574316,"elo":1124.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5237437037,"x_train":39.5175987803,"imp":11.8408256094,"elo":1131.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5507249031,"x_train":59.4870987604,"imp":11.8336701035,"elo":1137.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5707803637,"x_train":79.9551879424,"imp":11.8290459506,"elo":1135.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2839085103,"x_train":6.5430745829,"imp":11.4021616834,"elo":1188.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3853454659,"x_train":16.387957176,"imp":9.0872668175,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4768426596,"x_train":45.678712897,"imp":6.8982447724,"elo":1563.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.6715590282,"x_train":100.5864418887,"imp":6.1036598702,"elo":1638.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.9551220317,"x_train":253.2316993705,"imp":5.5083757333,"elo":1724.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.2841445532,"x_train":513.1174708444,"imp":5.211058434,"elo":1759.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5404194607,"x_train":1026.1644887093,"imp":5.0189052637,"elo":1790.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4492208192,"x_train":1544.7351451859,"imp":4.9248470231,"elo":1820.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.3549179469,"x_train":2068.1710397435,"imp":4.8422678218,"elo":1833.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5059595331,"x_train":1.5872114628,"imp":4.5972922799,"elo":1720.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3928986427,"x_train":5.1930136195,"imp":4.4954936788,"elo":1732.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.3467414164,"x_train":16.3030487246,"imp":4.2817235886,"elo":1744.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.2833870702,"x_train":34.3657315438,"imp":3.9788567334,"elo":1807.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.2909325197,"x_train":91.1547591648,"imp":3.6432521816,"elo":1933.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4808001914,"x_train":182.7153391231,"imp":3.4430381734,"elo":1961.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.6468940873,"x_train":366.5547593939,"imp":3.2949600688,"elo":2018.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.637851993,"x_train":550.802457324,"imp":3.2196038239,"elo":2046.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"x_train":739.1047147274,"imp":3.1862333181,"elo":2061.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.5250451841,"x_train":8.6828160405,"imp":12.2441147822,"elo":1134.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"x_train":41.5053698771,"imp":5.1412864523,"elo":1615.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.9928561684,"x_train":63.019350737,"imp":5.0028139527,"elo":1642.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":74.454905827,"x_train":119.7137023169,"imp":4.8377133521,"elo":1663.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":105.0879768186,"x_train":218.048215922,"imp":4.7224256158,"elo":1684.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":170.7606880264,"x_train":550.8419093004,"imp":4.4758639758,"elo":1728.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":225.019656008,"x_train":1115.34725308,"imp":4.2994519026,"elo":1756.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":286.7578860717,"x_train":2246.8238340239,"imp":4.1625090429,"elo":1782.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":338.6200879093,"x_train":3382.9355985564,"imp":4.0875733992,"elo":1791.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":357.2008760846,"x_train":4529.9305925825,"imp":4.0547168602,"elo":1800.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1767387784,"x_train":0.6611709932,"imp":4.2984610916,"elo":1751.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":11.6433178755,"x_train":25.1196552531,"imp":0.4758006822,"elo":2500.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1028500957,"x_train":0.471867489,"imp":3.5124324077,"elo":1815.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.038732221,"x_train":3.2167024044,"imp":9.7329856034,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0791734092,"x_train":11.7992281299,"imp":9.1947777236,"elo":1418.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.1510658648,"x_train":46.940037227,"imp":8.4487551977,"elo":1458.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2267457056,"x_train":112.2128891191,"imp":7.9823169354,"elo":1505.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3639606082,"x_train":298.6395201823,"imp":7.5780381651,"elo":1528.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4652595841,"x_train":596.2505830634,"imp":7.3583930626,"elo":1542.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.6436103929,"x_train":1163.2433543614,"imp":7.2040262897,"elo":1555.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.6500065101,"x_train":1770.0748555561,"imp":7.1312564436,"elo":1567.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5807768467,"x_train":2410.3120769872,"imp":7.0743284296,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"x_train":2.7774434997,"imp":4.2145025488,"elo":1885.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042135993,"x_train":0.8699671047,"imp":1.7853424303,"elo":1928.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6147142423,"x_train":113.401275248,"imp":27.2789297045,"elo":500.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.7368429189,"x_train":141.7238706247,"imp":27.1544799482,"elo":535.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.2568204055,"x_train":221.1962991478,"imp":26.854399481,"elo":551.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.7676176669,"x_train":328.7844244998,"imp":26.3300775969,"elo":567.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.1744464711,"x_train":736.9725789769,"imp":25.4060212814,"elo":596.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5384129217,"x_train":1485.4343663821,"imp":24.562354425,"elo":622.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5382733421,"x_train":2941.8656957704,"imp":24.2062265581,"elo":638.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.2160549793,"x_train":4341.8172896327,"imp":23.9692582481,"elo":671.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"x_train":5782.1195021638,"imp":23.7998469089,"elo":706.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1256335989,"x_train":0.5118384438,"imp":9.1228658323,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0292963248,"x_train":14.0524525282,"imp":15.2315343973,"elo":1023.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0753848671,"x_train":35.7395534737,"imp":14.3339737912,"elo":1043.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1421186744,"x_train":93.9257318856,"imp":12.9646971276,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1772212357,"x_train":201.2335467269,"imp":12.1017547455,"elo":1164.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2513416372,"x_train":481.4580064735,"imp":11.1402059021,"elo":1201.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.39401578,"x_train":1015.3213795664,"imp":10.7333183709,"elo":1216.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5446242225,"x_train":2031.0046161997,"imp":10.312716384,"elo":1231.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5498324906,"x_train":3088.2903635944,"imp":10.1160578174,"elo":1238.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4787581258,"x_train":4108.2381109512,"imp":9.9940462779,"elo":1255.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.073497228,"x_train":0.5708322449,"imp":9.8999433384,"elo":1295.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3133210319,"x_train":2.4574329869,"imp":9.7581492837,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8172787201,"x_train":7.3610615249,"imp":9.3444866285,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0996225013,"x_train":13.6386574902,"imp":9.1536785083,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4572887885,"x_train":35.9388283515,"imp":8.9518819388,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7915150002,"x_train":71.5452674162,"imp":8.8329662502,"elo":1434.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.9530600967,"x_train":142.0559469286,"imp":8.7587291911,"elo":1448.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0684490809,"x_train":215.0273705137,"imp":8.7490331287,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7976430646,"x_train":289.3497533723,"imp":8.7390847655,"elo":1451.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":19.9963063451,"x_train":93.5081427033,"imp":16.4845177929,"elo":871.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":27.6019600638,"x_train":175.8703779457,"imp":13.5955714626,"elo":995.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.3210715012,"x_train":423.2568217882,"imp":24.3807034641,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":45.7301059946,"x_train":836.4112409598,"imp":21.9294986191,"elo":1185.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":67.2056620148,"x_train":2070.7507295344,"imp":8.1871322543,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":65.9338217731,"x_train":2150.1856290327,"imp":8.1654629203,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.0932754819,"x_train":2.3191009865,"imp":10.699762351,"elo":1251.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.122018148,"x_train":4.6217004586,"imp":9.8879494154,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.1678454644,"x_train":11.2199797524,"imp":8.783901562,"elo":1376.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.2461082837,"x_train":22.2331145434,"imp":7.7860451008,"elo":1457.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.3494917743,"x_train":55.3170648503,"imp":7.0074538977,"elo":1521.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.4163168833,"x_train":109.9515587293,"imp":6.728401508,"elo":1552.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.4485143,"x_train":218.3385295205,"imp":6.5716820878,"elo":1566.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":0.4523893299,"x_train":327.734067188,"imp":6.4606380858,"elo":1582.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":0.5201417769,"x_train":438.5586959956,"imp":6.4091172112,"elo":1604.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..aa290efedd857b7057385b3d592e0813721aad7c
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_regression/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "ModernNCA (tuned + ensembled)", "TabPFN-2.6 (default)", "AutoGluon 1.4 (best, 4h)", "TabICLv2 (default)", "RealMLP (tuned + ensembled)", "TabDPT (tuned + ensembled)", "Nori-30M (default)", "TabDPT (tuned)", "TabDPT-Turbo (default)", "RealTabPFN-2.5 (default)", "Nori (default)", "RealMLP (tuned)", "LightGBM (tuned + ensembled)", "TabDPT (default)", "LightGBM (tuned)", "ModernNCA (tuned)", "xRFM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "TabM (tuned + ensembled)", "TabM (tuned)", "xRFM (tuned)", "CatBoost (default)", "ModernNCA (default)", "XGBoost (tuned + ensembled)", "XGBoost (tuned)", "iLTM (tuned + ensembled)", "LimiX (default)", "ChimeraBoost (tuned + ensembled)", "TabSwift (default)", "ChimeraBoost (tuned)", "iLTM (tuned)", "TabM (default)", "XGBoost (default)", "PerpetualBooster (tuned + ensembled)", "LightGBM (default)", "xRFM (default)", "ChimeraBoost (default)", "TorchMLP (tuned + ensembled)", "PerpetualBooster (tuned)", "RealMLP (default)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "ExtraTrees (tuned)", "SAP-RPT-OSS (default)", "EBM (tuned + ensembled)", "RandomForest (tuned + ensembled)", "RandomForest (tuned)", "PerpetualBooster (default)", "TorchMLP (default)", "EBM (tuned)", "RandomForest (default)", "EBM (default)", "ExtraTrees (default)", "iLTM (default)", "FastaiMLP (tuned + ensembled)", "FastaiMLP (tuned)", "KNN (tuned + ensembled)", "FastaiMLP (default)", "TabSTAR (tuned + ensembled)", "TabSTAR (tuned)", "KNN (tuned)", "TabSTAR (default)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, NaN, 0.6000000238418579, 0.20000000298023224, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.4000000059604645, NaN, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 1.0, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.800000011920929, 0.800000011920929, NaN, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, NaN, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 1.0, 0.6000000238418579, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, NaN, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, NaN, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, NaN, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 1.0, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, NaN, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.6000000238418579, 1.0, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, NaN, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.6000000238418579, 1.0, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, NaN, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, NaN, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 1.0, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.800000011920929, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 1.0, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, NaN, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.20000000298023224, 0.4000000059604645, 0.800000011920929, 0.800000011920929, NaN, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.800000011920929, 0.20000000298023224, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.20000000298023224, NaN, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.0, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.20000000298023224, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 0.4000000059604645, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.800000011920929, 0.20000000298023224, 0.800000011920929, 1.0, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.0, 0.4000000059604645, 0.0, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, NaN, 0.20000000298023224, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.800000011920929, NaN, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 1.0, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.0, NaN, 0.800000011920929, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, NaN, 0.6000000238418579, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.800000011920929, 1.0, 0.4000000059604645, 1.0, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, NaN, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.0, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.4000000059604645, 0.0, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.0, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.20000000298023224, 0.30000001192092896, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.0, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.0, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, NaN, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.4000000059604645, 1.0, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.4000000059604645, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, NaN, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, NaN, 0.800000011920929, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, NaN, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.0, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.0, 0.0, 0.6000000238418579, NaN, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, NaN, 0.800000011920929, 0.800000011920929, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, NaN, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, NaN, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9915492963},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9126760601},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.9042253555},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8957746531},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.876056341},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8591549344},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8563380325},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8366197215},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8281690224},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.825352118},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.819718317},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.8140845116},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.7915493027},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7802816959},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.771830993},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.7211267661},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7183098675},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7126760613},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7042253611},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.7014084551},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6985915572},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6929577524},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6901408491},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6676056385},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6591549358},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6394366261},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.6225352178},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6140845124},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.6056338104},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.580281694},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.57183099},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5464788778},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.5436619787},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5352112768},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5098591638},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5070422621},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4873239486},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4816901466},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4732394412},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4619718354},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.436619724},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.4338028215},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4253521203},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4197183146},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.3943662003},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.3887323993},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3859154985},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3830985966},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3661971879},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.343661977},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3436619761},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3408450747},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3366197226},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2985915529},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2845070467},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.2732394405},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2690140878},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.2619718349},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2619718353},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2084507071},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.1971831011},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.1830985936},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1605633829},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1352112692},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.1295774663},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.1183098605},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1098591559},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.0845070433},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0647887334},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0309859155},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0140845073},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0056338029}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_medium/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..8c88bbbb098105d711f86e5809fc5dd9c93c348a
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,73 @@
+,TabFM (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",ModernNCA (tuned + ensembled),TabPFN-2.6 (default),"AutoGluon 1.4 (best, 4h)",TabICLv2 (default),RealMLP (tuned + ensembled),TabDPT (tuned + ensembled),Nori-30M (default),TabDPT (tuned),TabDPT-Turbo (default),RealTabPFN-2.5 (default),Nori (default),RealMLP (tuned),LightGBM (tuned + ensembled),TabDPT (default),LightGBM (tuned),ModernNCA (tuned),xRFM (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),TabM (tuned + ensembled),TabM (tuned),xRFM (tuned),CatBoost (default),ModernNCA (default),XGBoost (tuned + ensembled),XGBoost (tuned),iLTM (tuned + ensembled),LimiX (default),ChimeraBoost (tuned + ensembled),TabSwift (default),ChimeraBoost (tuned),iLTM (tuned),TabM (default),XGBoost (default),PerpetualBooster (tuned + ensembled),LightGBM (default),xRFM (default),ChimeraBoost (default),TorchMLP (tuned + ensembled),PerpetualBooster (tuned),RealMLP (default),TorchMLP (tuned),ExtraTrees (tuned + ensembled),ExtraTrees (tuned),SAP-RPT-OSS (default),EBM (tuned + ensembled),RandomForest (tuned + ensembled),RandomForest (tuned),PerpetualBooster (default),TorchMLP (default),EBM (tuned),RandomForest (default),EBM (default),ExtraTrees (default),iLTM (default),FastaiMLP (tuned + ensembled),FastaiMLP (tuned),KNN (tuned + ensembled),FastaiMLP (default),TabSTAR (tuned + ensembled),TabSTAR (tuned),KNN (tuned),TabSTAR (default),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,,0.6,0.2,0.6,0.8,0.8,0.8,0.6,1.0,0.6,0.8,0.6,0.8,1.0,1.0,1.0,1.0,0.6,0.8,1.0,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.4,,0.2,0.6,0.6,0.8,0.8,0.4,1.0,0.6,1.0,0.6,0.8,0.8,1.0,1.0,0.8,0.6,0.8,1.0,1.0,0.8,0.8,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.2,0.8,0.8,,0.8,0.8,0.8,0.8,0.8,0.8,0.4,0.8,0.6,0.8,1.0,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,0.8,1.0,0.8,0.8,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2,0.4,0.4,0.2,,0.4,0.6,0.8,0.2,0.6,0.4,0.6,0.4,0.8,0.4,0.8,0.8,1.0,0.6,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.2,0.4,0.2,0.6,,0.4,0.6,0.2,0.6,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.6,0.8,1.0,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.0,0.2,0.2,0.2,0.4,0.6,,0.4,0.2,0.6,0.6,0.6,0.6,0.6,0.6,0.8,1.0,0.8,0.6,0.8,1.0,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.2,0.2,0.2,0.2,0.4,0.6,,0.2,0.4,0.4,0.4,0.4,0.6,0.4,0.8,0.8,1.0,0.6,0.8,0.6,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.0,0.4,0.6,0.2,0.8,0.8,0.8,0.8,,0.8,0.6,0.8,0.6,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,1.0,1.0,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.0,0.2,0.4,0.4,0.4,0.6,0.2,,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.6,0.6,0.6,0.6,1.0,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.2,0.4,0.4,0.6,0.6,0.4,0.4,0.6,0.4,0.4,,0.6,1.0,0.8,0.6,0.8,0.4,0.6,1.0,0.6,0.6,0.8,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori-30M (default),0.0,0.2,0.0,0.2,0.4,0.4,0.4,0.6,0.2,0.4,0.4,,0.6,0.8,0.4,0.8,0.4,0.8,0.6,0.8,0.6,0.8,0.8,0.8,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.4,0.4,0.4,0.6,0.4,0.4,0.6,0.4,0.4,0.0,0.4,,0.8,0.6,0.8,0.4,0.6,1.0,0.6,0.6,0.8,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.2,0.2,0.2,0.2,0.4,0.4,0.4,0.4,0.4,0.2,0.2,0.2,,0.6,0.8,0.6,0.6,0.8,0.6,0.8,0.6,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,0.8,1.0,0.8,0.8,1.0,0.8,1.0,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.0,0.2,0.0,0.6,0.4,0.4,0.6,0.4,0.4,0.4,0.6,0.4,0.4,,0.8,0.6,0.6,0.4,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori (default),0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,,0.6,0.6,0.4,0.6,0.4,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.6,0.8,0.8,1.0,0.8,0.8,0.8,0.8,1.0,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.6,0.6,0.6,0.4,0.4,0.4,,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.6,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,1.0,0.8,1.0,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.4,0.4,0.2,0.4,0.4,0.4,0.4,0.4,,0.4,0.6,0.4,0.4,0.6,0.6,0.4,0.4,0.6,1.0,0.6,1.0,0.8,0.8,0.6,0.8,0.6,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.4,0.4,0.0,0.4,0.4,0.4,0.4,0.4,0.4,0.0,0.4,0.0,0.2,0.6,0.6,0.4,0.6,,0.6,0.4,0.6,0.6,0.6,0.6,0.8,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.6,0.6,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.4,,0.4,0.4,0.6,0.6,0.2,0.2,0.4,0.6,0.4,1.0,1.0,0.8,0.6,0.8,0.6,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.4,0.2,0.4,0.4,0.4,0.4,0.2,0.2,0.6,0.4,0.6,0.6,0.6,,0.6,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.6,0.6,1.0,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.2,0.4,0.6,0.4,0.6,0.4,,0.6,0.6,0.6,0.6,1.0,0.6,0.8,0.6,0.6,1.0,0.8,0.6,0.8,0.6,1.0,0.8,0.8,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.4,0.4,0.4,0.4,0.4,,0.8,0.2,0.2,0.6,1.0,0.4,0.8,0.8,0.8,0.6,0.8,0.6,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.2,,0.2,0.2,0.6,0.6,0.4,0.6,0.6,0.8,0.6,0.8,0.6,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.6,0.4,0.8,0.2,0.4,0.8,0.8,,0.8,0.4,0.8,0.4,0.8,0.8,0.8,0.6,0.8,0.6,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.6,0.2,0.8,0.2,0.4,0.8,0.8,0.2,,0.4,0.8,0.4,0.8,0.8,0.6,0.6,0.8,0.6,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,0.6,0.8,1.0,0.8,1.0,1.0,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.4,0.4,0.6,0.2,0.0,0.4,0.4,0.6,0.6,,0.6,0.8,0.6,0.6,0.8,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.8,0.8,1.0,0.6,1.0,1.0,1.0,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.0,0.4,0.4,0.4,0.4,0.0,0.4,0.2,0.2,0.4,,0.4,0.8,0.8,0.8,0.6,0.4,0.6,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.6,1.0,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.2,0.4,0.2,0.4,0.4,0.6,0.2,0.2,0.6,0.6,0.6,0.6,0.2,0.6,,0.6,0.6,0.8,0.4,0.6,0.6,0.6,0.8,0.8,0.6,0.6,0.8,0.8,0.6,1.0,1.0,0.8,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.4,0.0,0.2,0.4,0.2,0.4,0.2,0.2,0.4,0.2,0.4,,0.6,0.4,0.6,0.4,0.4,0.4,0.8,0.6,0.8,1.0,1.0,0.8,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.0,0.2,0.4,0.2,0.4,0.2,0.2,0.4,0.2,0.4,0.4,,0.4,0.6,0.4,0.4,0.4,0.8,0.6,0.8,0.8,0.8,0.6,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.6,0.6,,0.6,0.4,0.4,0.4,0.8,0.6,0.8,0.8,0.8,1.0,0.4,0.8,0.8,0.8,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.0,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.4,0.4,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.6,0.4,0.4,0.4,,0.6,0.6,0.6,0.4,0.4,0.6,0.4,0.4,0.6,0.6,0.8,0.6,0.8,0.8,0.4,0.4,1.0,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.4,0.4,0.2,0.2,0.2,0.2,0.4,0.6,0.4,0.6,0.6,0.6,0.4,,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.6,1.0,0.6,0.8,0.8,0.6,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.2,0.4,0.2,0.4,0.0,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.4,0.6,0.6,0.6,0.4,0.4,,0.4,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.8,0.6,0.6,0.6,0.8,0.8,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.2,0.4,0.0,0.2,0.2,0.2,0.4,0.2,0.4,0.6,0.6,0.6,0.4,0.4,0.6,,0.6,0.6,0.8,0.8,0.8,0.6,1.0,0.6,0.8,0.8,0.6,0.8,0.8,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.6,0.4,0.4,0.4,,0.4,0.4,0.6,0.8,0.8,0.6,0.8,0.6,0.8,0.8,0.6,0.6,0.8,0.8,1.0,1.0,0.8,1.0,0.8,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.4,0.4,0.4,0.6,0.4,0.4,0.4,0.6,,0.8,0.8,0.8,0.6,0.6,0.6,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.6,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.4,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.4,0.2,0.4,0.2,0.6,0.2,,0.6,0.4,0.6,0.4,0.8,0.8,0.6,0.8,0.4,0.4,0.6,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.0,0.4,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.4,0.0,0.2,0.2,0.6,0.2,0.4,0.2,0.4,0.2,0.4,,0.4,0.8,0.2,0.8,1.0,0.6,0.8,0.6,0.6,0.6,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.6,0.2,0.4,0.2,0.2,0.2,0.6,0.6,,0.6,0.4,0.6,0.8,0.6,0.6,0.6,0.6,0.8,0.8,0.6,0.6,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.4,0.0,0.4,0.4,0.4,0.4,0.2,0.4,0.4,0.2,0.4,,0.4,0.4,0.6,0.6,0.4,0.8,0.8,0.8,0.6,0.8,0.8,0.6,0.8,0.6,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.6,0.4,0.0,0.4,0.0,0.4,0.4,0.6,0.8,0.6,0.6,,0.6,0.8,0.6,0.6,0.4,0.4,0.4,0.6,0.6,0.6,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.4,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.4,0.4,0.4,0.2,0.4,0.2,0.2,0.4,0.6,0.4,,0.2,0.6,0.8,0.6,0.6,0.8,0.8,0.6,0.8,0.8,1.0,0.8,0.8,0.8,0.6,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.4,0.2,0.4,0.2,0.4,0.2,0.2,0.0,0.2,0.4,0.2,0.8,,0.4,0.8,0.4,0.4,0.6,0.8,0.6,0.8,1.0,0.8,0.8,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.6,,0.4,0.6,0.6,0.6,0.8,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,1.0,1.0,0.8,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.2,0.2,0.0,0.2,0.4,0.4,0.4,0.2,0.4,0.2,0.2,0.4,0.6,0.4,0.2,0.2,0.6,,0.6,0.6,0.8,0.8,0.6,0.6,0.6,1.0,0.8,0.6,0.8,0.6,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.6,0.2,0.4,0.2,0.4,0.2,0.6,0.4,0.4,0.2,0.6,0.4,0.6,0.4,0.4,,1.0,0.8,0.6,0.6,0.4,0.6,0.4,0.6,0.8,0.6,1.0,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.6,0.2,0.4,0.2,0.4,0.2,0.6,0.4,0.4,0.2,0.6,0.4,0.6,0.4,0.4,0.0,,0.8,0.6,0.4,0.4,0.6,0.4,0.6,0.8,0.6,1.0,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.0,0.4,0.2,0.4,0.2,0.2,0.4,0.4,0.2,0.2,0.6,0.2,0.4,0.4,0.2,0.2,0.2,,0.6,0.2,0.2,0.4,0.6,0.8,0.4,0.8,0.6,0.6,0.8,0.8,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.2,0.2,0.2,0.4,0.4,0.4,,0.4,0.4,0.4,0.8,1.0,0.4,1.0,0.4,0.6,0.6,0.6,0.8,0.8,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.4,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.4,0.4,0.4,0.2,0.4,0.4,0.6,0.8,0.6,,0.4,0.6,0.6,0.6,0.8,0.6,0.6,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.4,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.6,0.6,0.8,0.6,0.6,,0.6,0.6,0.6,0.7,0.6,0.6,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.0,0.2,0.4,0.2,0.2,0.0,0.2,0.4,0.4,0.4,0.6,0.6,0.4,0.4,,0.8,0.8,0.4,0.8,0.4,0.8,0.6,0.8,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.4,0.2,0.4,0.0,0.4,0.2,0.2,0.2,0.2,0.4,0.0,0.2,0.4,0.0,0.6,0.6,0.4,0.2,0.4,0.4,0.2,,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.0,0.4,0.4,0.2,0.4,,0.4,0.8,0.4,0.4,0.6,0.6,0.6,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.4,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.4,0.2,0.2,0.6,0.6,0.2,0.3,0.6,0.4,0.6,,0.6,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.0,0.4,0.4,0.2,0.4,0.2,0.4,,0.4,0.4,0.6,0.6,0.6,0.6,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.4,0.2,0.2,0.4,0.0,0.0,0.4,0.6,0.4,0.4,0.6,0.4,0.6,0.4,0.6,,0.8,0.6,0.6,0.8,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.4,0.4,0.0,0.0,0.2,0.4,0.6,0.2,0.6,0.2,,0.4,0.6,0.6,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.4,0.2,0.4,0.2,0.4,0.4,0.6,,0.6,0.4,1.0,0.8,0.8,0.6,0.8,1.0,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.4,0.2,0.4,0.4,0.4,0.4,,0.4,0.8,0.8,0.8,0.6,0.8,1.0,1.0,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.0,0.0,0.4,0.2,0.0,0.0,0.2,0.2,0.4,0.0,0.4,0.2,0.4,0.6,0.6,,0.6,0.6,0.6,1.0,0.6,1.0,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.2,0.0,0.4,0.2,0.2,0.0,0.2,0.4,,0.8,0.8,0.4,0.8,0.6,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.4,0.2,,0.8,0.6,1.0,0.6,1.0,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,,0.6,1.0,0.6,1.0,1.0,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.4,0.4,0.0,0.6,0.4,0.4,,0.4,1.0,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.4,0.2,0.0,0.0,0.6,,0.6,1.0,1.0,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.4,0.4,0.4,0.0,0.4,,0.8,0.8,0.8
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,,1.0,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,,0.8
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_small/leaderboard_overview_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_small/leaderboard_overview_explorer.html
index d973d45d53059f28fa1e2de6353c18320ebd591d..091e84f346a29ac4318bfb569fe9ad1377848baa 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — no imputation | Lite | Regression | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_no/splits_lite/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_small/leaderboard_table.html b/data/imputation_no/splits_lite/tasks_regression/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..75dc6dbfbf16f1deb4769bcb6b0ce6ebbbeefbef
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_no/splits_lite/tasks_regression/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2248.0,"score":0.897,"rank":3.38,"hrank":1.62,"imp":1.989,"train_time":56.29,"predict_time":4.031,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+315\/-64","imputed":false},{"position":1,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2168.0,"score":0.8,"rank":4.5,"hrank":2.79,"imp":1.666,"train_time":14.41,"predict_time":2.084,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+312\/-75","imputed":false},{"position":2,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2103.0,"score":0.755,"rank":5.62,"hrank":4.28,"imp":2.977,"train_time":14.79,"predict_time":1.521,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+195\/-14","imputed":false},{"position":3,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2071.0,"score":0.71,"rank":6.25,"hrank":4.79,"imp":2.864,"train_time":323.76,"predict_time":2.906,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+194\/-56","imputed":false},{"position":4,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2014.0,"score":0.664,"rank":7.5,"hrank":3.76,"imp":2.58,"train_time":14.15,"predict_time":0.487,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+368\/-113","imputed":false},{"position":5,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1988.0,"score":0.69,"rank":8.12,"hrank":4.51,"imp":2.876,"train_time":10.03,"predict_time":0.63,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+336\/-143","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1936.0,"score":0.628,"rank":9.5,"hrank":5.01,"imp":3.973,"train_time":9830.47,"predict_time":232.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+293\/-131","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1922.0,"score":0.645,"rank":9.88,"hrank":5.47,"imp":2.432,"train_time":2356.36,"predict_time":7.546,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+418\/-203","imputed":false},{"position":8,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1865.0,"score":0.576,"rank":11.62,"hrank":9.69,"imp":4.601,"train_time":9830.47,"predict_time":22.221,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+174\/-71","imputed":false},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1756.0,"score":0.555,"rank":15.5,"hrank":4.93,"imp":4.91,"train_time":6.0,"predict_time":0.139,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+404\/-253","imputed":false},{"position":10,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1732.0,"score":0.444,"rank":16.5,"hrank":11.61,"imp":5.978,"train_time":51.28,"predict_time":43.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+218\/-97","imputed":false},{"position":11,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1729.0,"score":0.478,"rank":16.62,"hrank":8.88,"imp":3.882,"train_time":2356.36,"predict_time":0.878,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+364\/-190","imputed":false},{"position":12,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1700.0,"score":0.427,"rank":17.88,"hrank":16.19,"imp":6.227,"train_time":4918.07,"predict_time":35.587,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-68","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1686.0,"score":0.423,"rank":18.5,"hrank":15.74,"imp":6.135,"train_time":7.43,"predict_time":0.262,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+185\/-85","imputed":false},{"position":14,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1686.0,"score":0.37,"rank":18.5,"hrank":15.54,"imp":6.878,"train_time":4921.21,"predict_time":5.366,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+180\/-102","imputed":false},{"position":15,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1672.0,"score":0.415,"rank":19.12,"hrank":11.85,"imp":5.422,"train_time":8.24,"predict_time":0.532,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+348\/-207","imputed":false},{"position":16,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1632.0,"score":0.378,"rank":21.0,"hrank":14.62,"imp":5.905,"train_time":4.08,"predict_time":0.065,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+208\/-123","imputed":false},{"position":17,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1609.0,"score":0.281,"rank":22.12,"hrank":20.09,"imp":8.759,"train_time":39204.86,"predict_time":1759.104,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-90","imputed":false},{"position":18,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1577.0,"score":0.223,"rank":23.75,"hrank":21.83,"imp":9.446,"train_time":1206.7,"predict_time":4.571,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-92","imputed":false},{"position":19,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1562.0,"score":0.329,"rank":24.5,"hrank":19.56,"imp":7.309,"train_time":4918.07,"predict_time":0.949,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+178\/-136","imputed":false},{"position":20,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1537.0,"score":0.199,"rank":25.88,"hrank":22.51,"imp":8.776,"train_time":3127.07,"predict_time":0.459,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+170\/-88","imputed":false},{"position":21,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1537.0,"score":0.256,"rank":25.88,"hrank":22.14,"imp":7.219,"train_time":3814.24,"predict_time":4.368,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+197\/-118","imputed":false},{"position":22,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1534.0,"score":0.314,"rank":26.0,"hrank":6.13,"imp":10.6,"train_time":9011.78,"predict_time":8.012,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+237\/-199","imputed":false},{"position":23,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1532.0,"score":0.212,"rank":26.12,"hrank":23.41,"imp":8.735,"train_time":3127.07,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+170\/-95","imputed":false},{"position":24,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1509.0,"score":0.247,"rank":27.38,"hrank":22.18,"imp":8.935,"train_time":1673.38,"predict_time":2.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+191\/-138","imputed":false},{"position":25,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1474.0,"score":0.207,"rank":29.38,"hrank":24.77,"imp":7.438,"train_time":3814.24,"predict_time":0.453,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+194\/-134","imputed":false},{"position":26,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1463.0,"score":0.193,"rank":30.0,"hrank":25.51,"imp":9.224,"train_time":1673.38,"predict_time":0.097,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+144\/-98","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1457.0,"score":0.119,"rank":30.38,"hrank":28.66,"imp":9.921,"train_time":1277.57,"predict_time":0.647,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-70","imputed":false},{"position":28,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1451.0,"score":0.114,"rank":30.75,"hrank":29.78,"imp":9.992,"train_time":1277.57,"predict_time":3.744,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-55","imputed":false},{"position":29,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1428.0,"score":0.12,"rank":32.12,"hrank":29.67,"imp":10.618,"train_time":1206.7,"predict_time":0.689,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-106","imputed":false},{"position":30,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1409.0,"score":0.099,"rank":33.25,"hrank":30.26,"imp":11.436,"train_time":1889.62,"predict_time":0.341,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+154\/-115","imputed":false},{"position":31,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1402.0,"score":0.192,"rank":33.62,"hrank":25.56,"imp":10.294,"train_time":25.28,"predict_time":0.842,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+194\/-175","imputed":false},{"position":32,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1402.0,"score":0.129,"rank":33.62,"hrank":30.29,"imp":12.993,"train_time":25.15,"predict_time":0.294,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-129","imputed":false},{"position":33,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1388.0,"score":0.129,"rank":34.5,"hrank":30.73,"imp":10.165,"train_time":39204.86,"predict_time":127.594,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+153\/-133","imputed":false},{"position":34,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1384.0,"score":0.077,"rank":34.75,"hrank":31.56,"imp":11.735,"train_time":10.69,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-111","imputed":false},{"position":35,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1332.0,"score":0.065,"rank":37.88,"hrank":35.65,"imp":12.037,"train_time":1889.62,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-120","imputed":false},{"position":36,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1317.0,"score":0.105,"rank":38.75,"hrank":34.07,"imp":11.529,"train_time":22.42,"predict_time":7.815,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-114","imputed":false},{"position":37,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1315.0,"score":0.167,"rank":38.88,"hrank":15.04,"imp":11.915,"train_time":122.31,"predict_time":20.837,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+276\/-274","imputed":false},{"position":38,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1315.0,"score":0.085,"rank":38.88,"hrank":33.73,"imp":14.233,"train_time":1235.42,"predict_time":1.784,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+145\/-149","imputed":false},{"position":39,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1300.0,"score":0.085,"rank":39.75,"hrank":36.41,"imp":10.016,"train_time":14.25,"predict_time":0.485,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-118","imputed":false},{"position":40,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1298.0,"score":0.105,"rank":39.88,"hrank":28.63,"imp":13.554,"train_time":9011.78,"predict_time":0.526,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+200\/-189","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1294.0,"score":0.067,"rank":40.12,"hrank":36.94,"imp":11.386,"train_time":5196.37,"predict_time":1.749,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-146","imputed":false},{"position":42,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1277.0,"score":0.066,"rank":41.12,"hrank":36.6,"imp":14.37,"train_time":1235.42,"predict_time":0.317,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+134\/-159","imputed":false},{"position":43,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1246.0,"score":0.023,"rank":43.0,"hrank":40.94,"imp":13.334,"train_time":9.1,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-103","imputed":false},{"position":44,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.074,"rank":43.12,"hrank":38.54,"imp":11.704,"train_time":5196.37,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+188\/-187","imputed":false},{"position":45,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1236.0,"score":0.083,"rank":43.62,"hrank":31.94,"imp":14.489,"train_time":2809.99,"predict_time":0.115,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+206\/-206","imputed":false},{"position":46,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1208.0,"score":0.103,"rank":45.25,"hrank":27.99,"imp":15.003,"train_time":2809.99,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+210\/-217","imputed":false},{"position":47,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1182.0,"score":0.019,"rank":46.75,"hrank":44.07,"imp":16.355,"train_time":657.99,"predict_time":1.134,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-187","imputed":false},{"position":48,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1182.0,"score":0.098,"rank":46.75,"hrank":32.53,"imp":16.539,"train_time":7.0,"predict_time":1.362,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+203\/-276","imputed":false},{"position":49,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1180.0,"score":0.007,"rank":46.88,"hrank":45.61,"imp":13.756,"train_time":4.54,"predict_time":0.374,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-107","imputed":false},{"position":50,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1145.0,"score":0.0,"rank":48.88,"hrank":47.94,"imp":15.436,"train_time":2151.53,"predict_time":7.992,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-124","imputed":false},{"position":51,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1142.0,"score":0.018,"rank":49.0,"hrank":44.94,"imp":17.046,"train_time":7.81,"predict_time":0.985,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-224","imputed":false},{"position":52,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1126.0,"score":0.019,"rank":49.88,"hrank":46.92,"imp":17.448,"train_time":657.99,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+111\/-225","imputed":false},{"position":53,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1108.0,"score":0.0,"rank":50.81,"hrank":49.61,"imp":17.078,"train_time":208.16,"predict_time":0.453,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-155","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1105.0,"score":0.109,"rank":51.0,"hrank":28.37,"imp":24.308,"train_time":96418.84,"predict_time":36.577,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+263\/-449","imputed":false},{"position":55,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1105.0,"score":0.0,"rank":51.0,"hrank":49.54,"imp":15.864,"train_time":2151.53,"predict_time":0.566,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-147","imputed":false},{"position":56,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1092.0,"score":0.0,"rank":51.69,"hrank":50.7,"imp":17.791,"train_time":208.16,"predict_time":0.135,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-171","imputed":false},{"position":57,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1090.0,"score":0.0,"rank":51.75,"hrank":50.66,"imp":15.016,"train_time":4.08,"predict_time":0.255,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-114","imputed":false},{"position":58,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1056.0,"score":0.087,"rank":53.5,"hrank":36.56,"imp":24.484,"train_time":96418.84,"predict_time":5.523,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+259\/-448","imputed":false},{"position":59,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1050.0,"score":0.0,"rank":53.75,"hrank":52.16,"imp":16.495,"train_time":11.1,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-155","imputed":false},{"position":60,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1016.0,"score":0.0,"rank":55.38,"hrank":53.01,"imp":15.67,"train_time":22.25,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-196","imputed":false},{"position":61,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":56.12,"hrank":55.32,"imp":18.86,"train_time":0.69,"predict_time":0.125,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-206","imputed":false},{"position":62,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":960.0,"score":0.0,"rank":57.88,"hrank":56.83,"imp":20.309,"train_time":6.77,"predict_time":0.511,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-221","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":899.0,"score":0.0,"rank":60.25,"hrank":58.84,"imp":22.358,"train_time":88.79,"predict_time":0.571,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-297","imputed":false},{"position":64,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":893.0,"score":0.044,"rank":60.5,"hrank":51.18,"imp":27.524,"train_time":455.0,"predict_time":5.747,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+266\/-460","imputed":false},{"position":65,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":882.0,"score":0.0,"rank":60.88,"hrank":60.21,"imp":21.025,"train_time":34.18,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-176","imputed":false},{"position":66,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":841.0,"score":0.0,"rank":62.25,"hrank":61.94,"imp":20.975,"train_time":691.28,"predict_time":270.094,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+39\/-195","imputed":false},{"position":67,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":749.0,"score":0.0,"rank":64.88,"hrank":64.48,"imp":24.318,"train_time":88.79,"predict_time":0.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-291","imputed":false},{"position":68,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":691.0,"score":0.0,"rank":66.25,"hrank":65.36,"imp":30.565,"train_time":0.43,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+164\/-395","imputed":false},{"position":69,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":571.0,"score":0.0,"rank":68.44,"hrank":68.21,"imp":34.982,"train_time":743.23,"predict_time":0.364,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-500","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":546.0,"score":0.0,"rank":68.81,"hrank":68.58,"imp":35.064,"train_time":743.23,"predict_time":0.097,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-506","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":402.0,"score":0.0,"rank":70.5,"hrank":70.48,"imp":37.174,"train_time":3.98,"predict_time":0.12,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-436","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_small/pareto_front_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_small/pareto_front_explorer.html
index 64c31e5cd93ab89e9ccd918a57f8137c891fc8ac..090ab9dabd66d7f3d924ac7b692730a8eadb7e2e 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0440469671,"imp":11.7347505258,"elo":1383.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0570904815,"imp":8.7352040009,"elo":1532.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4587675243,"imp":8.7756000462,"elo":1536.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240241152,"imp":13.333827481,"elo":1246.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0270025453,"imp":12.0366317858,"elo":1331.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3406062626,"imp":11.4363945187,"elo":1408.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0136940061,"imp":16.4950841745,"elo":1050.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0126861072,"imp":15.002806027,"elo":1208.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1149138914,"imp":14.4888973668,"elo":1235.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9846715044,"imp":17.0456198454,"elo":1142.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3172224028,"imp":14.3701049802,"elo":1277.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7842242231,"imp":14.2330940967,"elo":1314.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5109617193,"imp":20.3085539952,"elo":959.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5659855728,"imp":15.8642121256,"elo":1104.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.9916224068,"imp":15.4359501903,"elo":1144.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.3736048612,"imp":13.7557845488,"elo":1180.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.6888310529,"imp":10.6180742868,"elo":1427.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.5709782007,"imp":9.4462953744,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":20.8369258593,"imp":11.9145923506,"elo":1314.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.29377362,"imp":12.992919227,"elo":1402.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5261711988,"imp":13.5544773089,"elo":1298.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0117732516,"imp":10.6000565736,"elo":1534.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.5206234824,"imp":2.9765050728,"elo":2102.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0843698074,"imp":1.6658940146,"elo":2168.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0198449839,"imp":21.0247735369,"elo":882.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1347029659,"imp":17.7906421375,"elo":1091.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4528171238,"imp":17.0784649589,"elo":1108.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1250003559,"imp":18.8602531123,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468843405,"imp":17.4482814124,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1335820465,"imp":16.3546864638,"elo":1182.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.8151845047,"imp":11.528554758,"elo":1316.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.9488198226,"imp":7.3092723649,"elo":1562.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":35.5866553094,"imp":6.2265792315,"elo":1699.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5321004862,"imp":5.4223934042,"elo":1671.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8781008023,"imp":3.8824639923,"elo":1728.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5456187851,"imp":2.4315310151,"elo":1922.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":0.8423700735,"imp":10.2939380886,"elo":1402.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.0616958015,"imp":5.978395454,"elo":1731.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":22.2214037327,"imp":4.6013925928,"elo":1864.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":232.1033993492,"imp":3.9730944558,"elo":1935.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2617923617,"imp":6.1350917842,"elo":1685.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":4.0307404894,"imp":1.9886645249,"elo":2248.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1387839993,"imp":4.909915297,"elo":1756.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4849712473,"imp":10.0160530863,"elo":1300.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4533957018,"imp":7.4375603739,"elo":1474.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.367582289,"imp":7.2190453621,"elo":1536.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6295234248,"imp":2.8755315855,"elo":1988.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4873524134,"imp":2.5796546204,"elo":2013.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7468382623,"imp":27.5237970384,"elo":892.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5231948052,"imp":24.4843446519,"elo":1055.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.5771589943,"imp":24.3082187083,"elo":1104.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0653075084,"imp":5.9050372269,"elo":1632.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305890866,"imp":15.6704262361,"elo":1016.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1207478083,"imp":11.7041014307,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.7488211618,"imp":11.3859797979,"elo":1294.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2554761956,"imp":15.0160376154,"elo":1090.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6471298159,"imp":9.9210562234,"elo":1457.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.7439813565,"imp":9.9915972128,"elo":1450.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":270.0943042458,"imp":20.9754349164,"elo":841.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":127.5938263768,"imp":10.1651955473,"elo":1387.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1759.1043245943,"imp":8.7593025522,"elo":1609.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.3621042316,"imp":16.5392791726,"elo":1182.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0968773188,"imp":9.2240465758,"elo":1463.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.446364541,"imp":8.9348440593,"elo":1509.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0440469671,"x_train":10.6906118793,"imp":11.7347505258,"elo":1383.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0570904815,"x_train":3127.0650185113,"imp":8.7352040009,"elo":1532.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4587675243,"x_train":3127.0650185113,"imp":8.7756000462,"elo":1536.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240241152,"x_train":9.0951433317,"imp":13.333827481,"elo":1246.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0270025453,"x_train":1889.6193046203,"imp":12.0366317858,"elo":1331.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3406062626,"x_train":1889.6193046203,"imp":11.4363945187,"elo":1408.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0136940061,"x_train":11.1028805965,"imp":16.4950841745,"elo":1050.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0126861072,"x_train":2809.98545611,"imp":15.002806027,"elo":1208.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1149138914,"x_train":2809.98545611,"imp":14.4888973668,"elo":1235.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9846715044,"x_train":7.8095418615,"imp":17.0456198454,"elo":1142.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3172224028,"x_train":1235.4175263307,"imp":14.3701049802,"elo":1277.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7842242231,"x_train":1235.4175263307,"imp":14.2330940967,"elo":1314.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5109617193,"x_train":6.7731104239,"imp":20.3085539952,"elo":959.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5659855728,"x_train":2151.5332639388,"imp":15.8642121256,"elo":1104.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.9916224068,"x_train":2151.5332639388,"imp":15.4359501903,"elo":1144.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.3736048612,"x_train":4.544665245,"imp":13.7557845488,"elo":1180.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.6888310529,"x_train":1206.7027545763,"imp":10.6180742868,"elo":1427.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.5709782007,"x_train":1206.7027545763,"imp":9.4462953744,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":20.8369258593,"x_train":122.3127076606,"imp":11.9145923506,"elo":1314.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.29377362,"x_train":25.1467163069,"imp":12.992919227,"elo":1402.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5261711988,"x_train":9011.7804615398,"imp":13.5544773089,"elo":1298.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0117732516,"x_train":9011.7804615398,"imp":10.6000565736,"elo":1534.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.5206234824,"x_train":14.7870051149,"imp":2.9765050728,"elo":2102.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0843698074,"x_train":14.4142036067,"imp":1.6658940146,"elo":2168.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0198449839,"x_train":34.1835619211,"imp":21.0247735369,"elo":882.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1347029659,"x_train":208.1564418277,"imp":17.7906421375,"elo":1091.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4528171238,"x_train":208.1564418277,"imp":17.0784649589,"elo":1108.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1250003559,"x_train":0.6913103896,"imp":18.8602531123,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468843405,"x_train":657.9874551063,"imp":17.4482814124,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1335820465,"x_train":657.9874551063,"imp":16.3546864638,"elo":1182.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.8151845047,"x_train":22.4158785265,"imp":11.528554758,"elo":1316.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.9488198226,"x_train":4918.0723002043,"imp":7.3092723649,"elo":1562.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":35.5866553094,"x_train":4918.0723002043,"imp":6.2265792315,"elo":1699.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5321004862,"x_train":8.242142907,"imp":5.4223934042,"elo":1671.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8781008023,"x_train":2356.3627464618,"imp":3.8824639923,"elo":1728.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5456187851,"x_train":2356.3627464618,"imp":2.4315310151,"elo":1922.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":0.8423700735,"x_train":25.2809733305,"imp":10.2939380886,"elo":1402.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.0616958015,"x_train":51.2843993162,"imp":5.978395454,"elo":1731.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":22.2214037327,"x_train":9830.4728139885,"imp":4.6013925928,"elo":1864.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":232.1033993492,"x_train":9830.4728139885,"imp":3.9730944558,"elo":1935.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2617923617,"x_train":7.4300775989,"imp":6.1350917842,"elo":1685.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":4.0307404894,"x_train":56.2879369055,"imp":1.9886645249,"elo":2248.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1387839993,"x_train":5.9977774395,"imp":4.909915297,"elo":1756.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4849712473,"x_train":14.2468083815,"imp":10.0160530863,"elo":1300.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4533957018,"x_train":3814.2351446507,"imp":7.4375603739,"elo":1474.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.367582289,"x_train":3814.2351446507,"imp":7.2190453621,"elo":1536.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6295234248,"x_train":10.02817758,"imp":2.8755315855,"elo":1988.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4873524134,"x_train":14.1514284946,"imp":2.5796546204,"elo":2013.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7468382623,"x_train":454.9991027301,"imp":27.5237970384,"elo":892.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5231948052,"x_train":96418.8447274511,"imp":24.4843446519,"elo":1055.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.5771589943,"x_train":96418.8447274511,"imp":24.3082187083,"elo":1104.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0653075084,"x_train":4.0841565985,"imp":5.9050372269,"elo":1632.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305890866,"x_train":22.2534811738,"imp":15.6704262361,"elo":1016.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1207478083,"x_train":5196.3714196423,"imp":11.7041014307,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.7488211618,"x_train":5196.3714196423,"imp":11.3859797979,"elo":1294.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2554761956,"x_train":4.0843435263,"imp":15.0160376154,"elo":1090.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6471298159,"x_train":1277.5662482217,"imp":9.9210562234,"elo":1457.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.7439813565,"x_train":1277.5662482217,"imp":9.9915972128,"elo":1450.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":270.0943042458,"x_train":691.2819503211,"imp":20.9754349164,"elo":841.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":127.5938263768,"x_train":39204.8641933754,"imp":10.1651955473,"elo":1387.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1759.1043245943,"x_train":39204.8641933754,"imp":8.7593025522,"elo":1609.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.3621042316,"x_train":7.002585411,"imp":16.5392791726,"elo":1182.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0968773188,"x_train":1673.3760165809,"imp":9.2240465758,"elo":1463.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.446364541,"x_train":1673.3760165809,"imp":8.9348440593,"elo":1509.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_small/tuning_trajectories_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_small/tuning_trajectories_explorer.html
index d06ad82dd2a8cf6210e672d186001dbfe99604ed..7125d2b439a52d4303318aa42b6c3a4d3601a702 100644
--- a/data/imputation_no/splits_lite/tasks_regression/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_no/splits_lite/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":10.6906118793,"imp":11.7380678406,"elo":1396.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":27.5168414441,"imp":10.8787795601,"elo":1457.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":74.1773887355,"imp":9.8956020481,"elo":1536.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":149.7230535562,"imp":9.4539070634,"elo":1555.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":390.8454255101,"imp":9.0355787916,"elo":1569.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":774.7646676657,"imp":8.9126650851,"elo":1567.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1549.9544564854,"imp":8.8482083003,"elo":1563.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2319.2427277232,"imp":8.8009269607,"elo":1567.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":3127.0650185113,"imp":8.779742505,"elo":1565.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":9.0951433317,"imp":13.3368408494,"elo":1259.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":18.0665913735,"imp":12.7290405653,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":47.3379775935,"imp":12.2230650412,"elo":1357.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":95.7569831536,"imp":12.0284416083,"elo":1376.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":237.2122715241,"imp":11.8246977814,"elo":1387.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":473.9846968668,"imp":11.7071831591,"elo":1400.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":942.9066009657,"imp":11.5364625858,"elo":1414.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1412.1394987216,"imp":11.4668809952,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1889.6193046203,"imp":11.4395445262,"elo":1433.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":11.1028805965,"imp":16.4976379141,"elo":1066.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":29.3949504722,"imp":15.8588828829,"elo":1122.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":76.7487383079,"imp":15.2784837429,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":146.1996217178,"imp":15.0041573064,"elo":1225.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":363.6322234704,"imp":14.7544531085,"elo":1242.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":719.7922921179,"imp":14.6819004949,"elo":1241.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1405.778816208,"imp":14.5966335853,"elo":1251.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":2088.5183356085,"imp":14.5352462665,"elo":1251.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2809.98545611,"imp":14.4895382411,"elo":1256.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.8095418615,"imp":17.0483421261,"elo":1148.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.6268296558,"imp":16.5555629996,"elo":1180.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":34.3331356985,"imp":15.6738655468,"elo":1257.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":62.7079095608,"imp":15.1857714271,"elo":1289.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":155.029134045,"imp":14.4542681877,"elo":1319.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":307.6069593637,"imp":14.3040260216,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":613.0063601899,"imp":14.2337102267,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":922.5048472624,"imp":14.2288971439,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1235.4175263307,"imp":14.2362430223,"elo":1333.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.7731104239,"imp":20.3106713122,"elo":951.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":17.0044463073,"imp":19.9496894914,"elo":968.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":50.1505869221,"imp":18.7437585272,"elo":1019.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":101.3841054234,"imp":18.1779604669,"elo":1036.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":260.7094513648,"imp":17.2940660418,"elo":1078.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":528.4734813547,"imp":16.7765138581,"elo":1108.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1059.681319286,"imp":16.1672923144,"elo":1131.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1596.5291960327,"imp":15.7406048434,"elo":1151.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2151.5332639388,"imp":15.4390915666,"elo":1162.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":4.544665245,"imp":13.7590818546,"elo":1188.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":10.636770614,"imp":12.476047148,"elo":1289.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":27.7973506573,"imp":11.1844991259,"elo":1414.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":56.861966814,"imp":10.6108589021,"elo":1476.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":148.4745666134,"imp":10.0876299898,"elo":1544.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":299.3928042834,"imp":9.9288733954,"elo":1573.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":598.0451468439,"imp":9.7033578519,"elo":1598.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":893.892871219,"imp":9.5698709534,"elo":1612.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1206.7027545763,"imp":9.4499487541,"elo":1626.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":122.3127076606,"imp":11.9179021798,"elo":1356.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.1467163069,"imp":12.9962863151,"elo":1436.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":75.362933136,"imp":12.3718490645,"elo":1466.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":210.5456172414,"imp":11.6267396007,"elo":1517.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":424.8014044984,"imp":11.3081905363,"elo":1553.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":1033.2681066658,"imp":11.0747240533,"elo":1560.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":2222.1208631991,"imp":10.8935157408,"elo":1557.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":4429.3203715947,"imp":10.7520103908,"elo":1555.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":6624.9729135871,"imp":10.6295149838,"elo":1565.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":9011.7804615398,"imp":10.6037965299,"elo":1566.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":14.7870051149,"imp":2.9817681293,"elo":2142.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":14.4142036067,"imp":1.6715446805,"elo":2182.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":34.1835619211,"imp":21.0267397864,"elo":909.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":68.7178569705,"imp":18.9265958956,"elo":1019.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":173.2458309649,"imp":17.0684828461,"elo":1110.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":208.1564418277,"imp":17.0805060316,"elo":1110.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.6913103896,"imp":18.8630134196,"elo":1000.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.4537617156,"imp":17.8387549123,"elo":1067.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":14.301129027,"imp":17.2622565173,"elo":1115.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":30.686422522,"imp":16.8003854792,"elo":1148.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":79.9124717904,"imp":16.5519192401,"elo":1167.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":162.1056028409,"imp":16.4538884455,"elo":1178.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":326.1499058044,"imp":16.3873501449,"elo":1190.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":490.3882635115,"imp":16.3633492171,"elo":1195.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":657.9874551063,"imp":16.3574828558,"elo":1198.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":22.4158785265,"imp":11.5329003355,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":44.4251076686,"imp":9.5741826539,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":117.3230828329,"imp":7.6359018563,"elo":1591.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":253.2768449964,"imp":6.9578682152,"elo":1633.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":616.4798700039,"imp":6.5714223235,"elo":1676.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1238.3164021687,"imp":6.339582604,"elo":1721.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2457.5657747181,"imp":6.2626565171,"elo":1737.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3675.6638274823,"imp":6.2412663581,"elo":1742.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4918.0723002043,"imp":6.2282511886,"elo":1747.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":8.242142907,"imp":5.427403104,"elo":1699.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":19.8358235954,"imp":4.7261793707,"elo":1736.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":56.6252728837,"imp":3.5987206349,"elo":1821.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":113.5361347503,"imp":3.0747732047,"elo":1890.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":295.9584460071,"imp":2.5468910913,"elo":1922.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":584.1451401972,"imp":2.3895569087,"elo":1947.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1169.0460603376,"imp":2.3310758771,"elo":1970.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1760.7500065581,"imp":2.4258915729,"elo":1959.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2356.3627464618,"imp":2.4360173334,"elo":1960.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":25.2809733305,"imp":10.2979773772,"elo":1433.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.2843993162,"imp":5.9831587418,"elo":1775.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":95.8813700743,"imp":5.3971294242,"elo":1813.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":254.4393954229,"imp":4.8561409198,"elo":1846.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":502.170242762,"imp":4.4172900231,"elo":1905.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":1229.6581613295,"imp":4.1595602054,"elo":1937.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":2450.1429629138,"imp":4.1026352104,"elo":1942.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":4887.1069852066,"imp":4.028961281,"elo":1955.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":7331.038512164,"imp":3.9899091528,"elo":1978.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":9830.4728139885,"imp":3.9787406493,"elo":1985.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":7.4300775989,"imp":6.1398219236,"elo":1730.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":56.2879369055,"imp":1.9939435026,"elo":2249.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":5.9977774395,"imp":4.914734588,"elo":1785.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":14.2468083815,"imp":10.0205504104,"elo":1312.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":34.9108464136,"imp":9.4576657254,"elo":1345.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":89.6967770714,"imp":8.8810065831,"elo":1397.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":190.6759613979,"imp":8.3070618065,"elo":1458.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":456.3526262514,"imp":7.8318467474,"elo":1506.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":918.9449885462,"imp":7.4798691525,"elo":1551.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1845.4338559674,"imp":7.2737826281,"elo":1576.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2797.5928059,"imp":7.2496835275,"elo":1576.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3814.2351446507,"imp":7.2240501445,"elo":1568.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":10.02817758,"imp":2.8808212181,"elo":2033.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":14.1514284946,"imp":2.5953010295,"elo":2041.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":454.9991027301,"imp":27.5261770168,"elo":908.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":923.2022669306,"imp":27.3713103765,"elo":927.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2264.5016798126,"imp":26.9693598956,"elo":968.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4340.8142220335,"imp":26.2782538996,"elo":1013.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11688.77778151,"imp":25.4429386552,"elo":1062.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23707.0922342587,"imp":24.9021077683,"elo":1102.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":47472.2129899786,"imp":24.5530137841,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":71823.9094609453,"imp":24.4153443067,"elo":1132.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":96418.8447274511,"imp":24.3105081894,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":4.0841565985,"imp":5.9100274158,"elo":1657.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":22.2534811738,"imp":15.6737475478,"elo":1034.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":54.2457268049,"imp":15.1761073979,"elo":1071.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":127.8066374424,"imp":14.1453339402,"elo":1145.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":271.1084504473,"imp":13.71982596,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":648.9159658958,"imp":12.9816291054,"elo":1217.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1311.0552340778,"imp":12.4498845916,"elo":1258.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2566.6969558991,"imp":11.8624617347,"elo":1282.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3875.6410639922,"imp":11.5964446071,"elo":1303.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":5196.3714196423,"imp":11.3898917946,"elo":1322.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":4.0843435263,"imp":15.0194661753,"elo":1094.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":10.4789410396,"imp":12.2265999012,"elo":1294.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":29.8024094892,"imp":11.2783006114,"elo":1372.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":60.219465986,"imp":10.7625899279,"elo":1429.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":157.148713481,"imp":10.397729215,"elo":1464.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":316.488046922,"imp":10.2401435688,"elo":1478.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":634.705809045,"imp":10.1282155644,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":952.8874450538,"imp":10.0298489274,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":1277.5662482217,"imp":9.9954828299,"elo":1487.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":691.2819503211,"imp":20.9779761968,"elo":871.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":2116.0353219536,"imp":15.7245100978,"elo":1115.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":6702.4310187379,"imp":11.1994420107,"elo":1376.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":15189.3088063979,"imp":9.9294564808,"elo":1490.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":37312.4085130914,"imp":8.7787641694,"elo":1645.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":39204.8641933754,"imp":8.7636936868,"elo":1644.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":7.002585411,"imp":16.5424125465,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":15.1443119472,"imp":14.8045524272,"elo":1267.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":39.7770511549,"imp":12.5667233788,"elo":1359.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":81.229812475,"imp":11.0712301058,"elo":1422.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":207.1643683575,"imp":9.7424442267,"elo":1488.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":413.5799198557,"imp":9.3688825164,"elo":1528.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":828.216131447,"imp":9.0618942855,"elo":1542.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1248.051456359,"imp":8.9494988014,"elo":1555.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1673.3760165809,"imp":8.9390070431,"elo":1551.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0440469671,"x_train":10.6906118793,"imp":11.7380678406,"elo":1396.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.094198463,"x_train":27.5168414441,"imp":10.8787795601,"elo":1457.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2094177891,"x_train":74.1773887355,"imp":9.8956020481,"elo":1536.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2708311702,"x_train":149.7230535562,"imp":9.4539070634,"elo":1555.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3286420785,"x_train":390.8454255101,"imp":9.0355787916,"elo":1569.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3693540345,"x_train":774.7646676657,"imp":8.9126650851,"elo":1567.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4160109308,"x_train":1549.9544564854,"imp":8.8482083003,"elo":1563.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4476529792,"x_train":2319.2427277232,"imp":8.8009269607,"elo":1567.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4587675243,"x_train":3127.0650185113,"imp":8.779742505,"elo":1565.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240241152,"x_train":9.0951433317,"imp":13.3368408494,"elo":1259.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0530471399,"x_train":18.0665913735,"imp":12.7290405653,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1229635811,"x_train":47.3379775935,"imp":12.2230650412,"elo":1357.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1719705654,"x_train":95.7569831536,"imp":12.0284416083,"elo":1376.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2033909926,"x_train":237.2122715241,"imp":11.8246977814,"elo":1387.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2377216613,"x_train":473.9846968668,"imp":11.7071831591,"elo":1400.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2621009173,"x_train":942.9066009657,"imp":11.5364625858,"elo":1414.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2795215411,"x_train":1412.1394987216,"imp":11.4668809952,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3406062626,"x_train":1889.6193046203,"imp":11.4395445262,"elo":1433.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0136940061,"x_train":11.1028805965,"imp":16.4976379141,"elo":1066.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0256054927,"x_train":29.3949504722,"imp":15.8588828829,"elo":1122.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0485028448,"x_train":76.7487383079,"imp":15.2784837429,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0670431171,"x_train":146.1996217178,"imp":15.0041573064,"elo":1225.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0946419883,"x_train":363.6322234704,"imp":14.7544531085,"elo":1242.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.10646037,"x_train":719.7922921179,"imp":14.6819004949,"elo":1241.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1141653132,"x_train":1405.778816208,"imp":14.5966335853,"elo":1251.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1073495676,"x_train":2088.5183356085,"imp":14.5352462665,"elo":1251.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1149138914,"x_train":2809.98545611,"imp":14.4895382411,"elo":1256.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9846715044,"x_train":7.8095418615,"imp":17.0483421261,"elo":1148.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1163998088,"x_train":14.6268296558,"imp":16.5555629996,"elo":1180.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0229922831,"x_train":34.3331356985,"imp":15.6738655468,"elo":1257.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8472601246,"x_train":62.7079095608,"imp":15.1857714271,"elo":1289.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0307532617,"x_train":155.029134045,"imp":14.4542681877,"elo":1319.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3416498532,"x_train":307.6069593637,"imp":14.3040260216,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.624415598,"x_train":613.0063601899,"imp":14.2337102267,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.8874966855,"x_train":922.5048472624,"imp":14.2288971439,"elo":1336.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7842242231,"x_train":1235.4175263307,"imp":14.2362430223,"elo":1333.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5109617193,"x_train":6.7731104239,"imp":20.3106713122,"elo":951.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3803532068,"x_train":17.0044463073,"imp":19.9496894914,"elo":968.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8851619804,"x_train":50.1505869221,"imp":18.7437585272,"elo":1019.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.8252238839,"x_train":101.3841054234,"imp":18.1779604669,"elo":1036.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.4333239012,"x_train":260.7094513648,"imp":17.2940660418,"elo":1078.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.6382886698,"x_train":528.4734813547,"imp":16.7765138581,"elo":1108.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.0448649768,"x_train":1059.681319286,"imp":16.1672923144,"elo":1131.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.5681335639,"x_train":1596.5291960327,"imp":15.7406048434,"elo":1151.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.9916224068,"x_train":2151.5332639388,"imp":15.4390915666,"elo":1162.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.3736048612,"x_train":4.544665245,"imp":13.7590818546,"elo":1188.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5668632649,"x_train":10.636770614,"imp":12.476047148,"elo":1289.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.9848948133,"x_train":27.7973506573,"imp":11.1844991259,"elo":1414.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.3643462664,"x_train":56.861966814,"imp":10.6108589021,"elo":1476.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.4293886426,"x_train":148.4745666134,"imp":10.0876299898,"elo":1544.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.7084309099,"x_train":299.3928042834,"imp":9.9288733954,"elo":1573.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.5371993778,"x_train":598.0451468439,"imp":9.7033578519,"elo":1598.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.4245638467,"x_train":893.892871219,"imp":9.5698709534,"elo":1612.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.5709782007,"x_train":1206.7027545763,"imp":9.4499487541,"elo":1626.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":20.8369258593,"x_train":122.3127076606,"imp":11.9179021798,"elo":1356.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.29377362,"x_train":25.1467163069,"imp":12.9962863151,"elo":1436.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7419941645,"x_train":75.362933136,"imp":12.3718490645,"elo":1466.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.5922413434,"x_train":210.5456172414,"imp":11.6267396007,"elo":1517.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.4715181761,"x_train":424.8014044984,"imp":11.3081905363,"elo":1553.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.7597331666,"x_train":1033.2681066658,"imp":11.0747240533,"elo":1560.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1709239995,"x_train":2222.1208631991,"imp":10.8935157408,"elo":1557.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.2034777482,"x_train":4429.3203715947,"imp":10.7520103908,"elo":1555.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8424907827,"x_train":6624.9729135871,"imp":10.6295149838,"elo":1565.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0117732516,"x_train":9011.7804615398,"imp":10.6037965299,"elo":1566.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":1.5206234824,"x_train":14.7870051149,"imp":2.9817681293,"elo":2142.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0843698074,"x_train":14.4142036067,"imp":1.6715446805,"elo":2182.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0198449839,"x_train":34.1835619211,"imp":21.0267397864,"elo":909.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.143539937,"x_train":68.7178569705,"imp":18.9265958956,"elo":1019.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4090035871,"x_train":173.2458309649,"imp":17.0684828461,"elo":1110.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4528171238,"x_train":208.1564418277,"imp":17.0805060316,"elo":1110.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1250003559,"x_train":0.6913103896,"imp":18.8630134196,"elo":1000.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2971430628,"x_train":4.4537617156,"imp":17.8387549123,"elo":1067.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5361761201,"x_train":14.301129027,"imp":17.2622565173,"elo":1115.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7051520621,"x_train":30.686422522,"imp":16.8003854792,"elo":1148.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9048497109,"x_train":79.9124717904,"imp":16.5519192401,"elo":1167.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.962102534,"x_train":162.1056028409,"imp":16.4538884455,"elo":1178.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0937878381,"x_train":326.1499058044,"imp":16.3873501449,"elo":1190.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2475282764,"x_train":490.3882635115,"imp":16.3633492171,"elo":1195.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.1449696518,"x_train":657.9874551063,"imp":16.3574828558,"elo":1198.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.8151845047,"x_train":22.4158785265,"imp":11.5329003355,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.4129743265,"x_train":44.4251076686,"imp":9.5741826539,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.4450757134,"x_train":117.3230828329,"imp":7.6359018563,"elo":1591.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":20.928441831,"x_train":253.2768449964,"imp":6.9578682152,"elo":1633.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":24.2900358461,"x_train":616.4798700039,"imp":6.5714223235,"elo":1676.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":28.5298041944,"x_train":1238.3164021687,"imp":6.339582604,"elo":1721.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":33.0852161799,"x_train":2457.5657747181,"imp":6.2626565171,"elo":1737.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":35.1242790175,"x_train":3675.6638274823,"imp":6.2412663581,"elo":1742.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":35.5866553094,"x_train":4918.0723002043,"imp":6.2282511886,"elo":1747.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5321004862,"x_train":8.242142907,"imp":5.427403104,"elo":1699.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4099952993,"x_train":19.8358235954,"imp":4.7261793707,"elo":1736.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.7879588822,"x_train":56.6252728837,"imp":3.5987206349,"elo":1821.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.0310191127,"x_train":113.5361347503,"imp":3.0747732047,"elo":1890.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.4432569251,"x_train":295.9584460071,"imp":2.5468910913,"elo":1922.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4486360166,"x_train":584.1451401972,"imp":2.3895569087,"elo":1947.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.6972941708,"x_train":1169.0460603376,"imp":2.3310758771,"elo":1970.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.5512603426,"x_train":1760.7500065581,"imp":2.4258915729,"elo":1959.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5456187851,"x_train":2356.3627464618,"imp":2.4360173334,"elo":1960.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":0.8423700735,"x_train":25.2809733305,"imp":10.2979773772,"elo":1433.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.0616958015,"x_train":51.2843993162,"imp":5.9831587418,"elo":1775.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":70.1115350575,"x_train":95.8813700743,"imp":5.3971294242,"elo":1813.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":117.0496168126,"x_train":254.4393954229,"imp":4.8561409198,"elo":1846.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":145.633406717,"x_train":502.170242762,"imp":4.4172900231,"elo":1905.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":177.8219001626,"x_train":1229.6581613295,"imp":4.1595602054,"elo":1937.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":195.075324367,"x_train":2450.1429629138,"imp":4.1026352104,"elo":1942.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":207.9632508624,"x_train":4887.1069852066,"imp":4.028961281,"elo":1955.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":227.9964108122,"x_train":7331.038512164,"imp":3.9899091528,"elo":1978.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":232.1033993492,"x_train":9830.4728139885,"imp":3.9787406493,"elo":1985.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2617923617,"x_train":7.4300775989,"imp":6.1398219236,"elo":1730.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":4.0307404894,"x_train":56.2879369055,"imp":1.9939435026,"elo":2249.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1387839993,"x_train":5.9977774395,"imp":4.914734588,"elo":1785.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4849712473,"x_train":14.2468083815,"imp":10.0205504104,"elo":1312.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.9132928236,"x_train":34.9108464136,"imp":9.4576657254,"elo":1345.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.7529208648,"x_train":89.6967770714,"imp":8.8810065831,"elo":1397.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.4556253929,"x_train":190.6759613979,"imp":8.3070618065,"elo":1458.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":3.1346820285,"x_train":456.3526262514,"imp":7.8318467474,"elo":1506.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.5972380055,"x_train":918.9449885462,"imp":7.4798691525,"elo":1551.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":4.3086986908,"x_train":1845.4338559674,"imp":7.2737826281,"elo":1576.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.6217168271,"x_train":2797.5928059,"imp":7.2496835275,"elo":1576.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.367582289,"x_train":3814.2351446507,"imp":7.2240501445,"elo":1568.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6295234248,"x_train":10.02817758,"imp":2.8808212181,"elo":2033.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4873524134,"x_train":14.1514284946,"imp":2.5953010295,"elo":2041.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7468382623,"x_train":454.9991027301,"imp":27.5261770168,"elo":908.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.6136847143,"x_train":923.2022669306,"imp":27.3713103765,"elo":927.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.3872419149,"x_train":2264.5016798126,"imp":26.9693598956,"elo":968.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.4500316943,"x_train":4340.8142220335,"imp":26.2782538996,"elo":1013.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":26.9593058501,"x_train":11688.77778151,"imp":25.4429386552,"elo":1062.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":33.0819178763,"x_train":23707.0922342587,"imp":24.9021077683,"elo":1102.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":37.9017474296,"x_train":47472.2129899786,"imp":24.5530137841,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":38.4814086805,"x_train":71823.9094609453,"imp":24.4153443067,"elo":1132.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.5771589943,"x_train":96418.8447274511,"imp":24.3105081894,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0653075084,"x_train":4.0841565985,"imp":5.9100274158,"elo":1657.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305890866,"x_train":22.2534811738,"imp":15.6737475478,"elo":1034.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2739656197,"x_train":54.2457268049,"imp":15.1761073979,"elo":1071.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4747909368,"x_train":127.8066374424,"imp":14.1453339402,"elo":1145.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6786610978,"x_train":271.1084504473,"imp":13.71982596,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9735411513,"x_train":648.9159658958,"imp":12.9816291054,"elo":1217.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2437344525,"x_train":1311.0552340778,"imp":12.4498845916,"elo":1258.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4822354588,"x_train":2566.6969558991,"imp":11.8624617347,"elo":1282.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6235068465,"x_train":3875.6410639922,"imp":11.5964446071,"elo":1303.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.7488211618,"x_train":5196.3714196423,"imp":11.3898917946,"elo":1322.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2554761956,"x_train":4.0843435263,"imp":15.0194661753,"elo":1094.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.957900168,"x_train":10.4789410396,"imp":12.2265999012,"elo":1294.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0042194555,"x_train":29.8024094892,"imp":11.2783006114,"elo":1372.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3054096434,"x_train":60.219465986,"imp":10.7625899279,"elo":1429.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.4008496262,"x_train":157.148713481,"imp":10.397729215,"elo":1464.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.6357668007,"x_train":316.488046922,"imp":10.2401435688,"elo":1478.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.6775449023,"x_train":634.705809045,"imp":10.1282155644,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.121628747,"x_train":952.8874450538,"imp":10.0298489274,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.7439813565,"x_train":1277.5662482217,"imp":9.9954828299,"elo":1487.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":270.0943042458,"x_train":691.2819503211,"imp":20.9779761968,"elo":871.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":546.5990400536,"x_train":2116.0353219536,"imp":15.7245100978,"elo":1115.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":1020.4039532515,"x_train":6702.4310187379,"imp":11.1994420107,"elo":1376.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":1274.0935394324,"x_train":15189.3088063979,"imp":9.9294564808,"elo":1490.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":1744.1344030008,"x_train":37312.4085130914,"imp":8.7787641694,"elo":1645.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":1759.1043245943,"x_train":39204.8641933754,"imp":8.7636936868,"elo":1644.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.3621042316,"x_train":7.002585411,"imp":16.5424125465,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.7787898657,"x_train":15.1443119472,"imp":14.8045524272,"elo":1267.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.849660528,"x_train":39.7770511549,"imp":12.5667233788,"elo":1359.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.6661465754,"x_train":81.229812475,"imp":11.0712301058,"elo":1422.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.1507772428,"x_train":207.1643683575,"imp":9.7424442267,"elo":1488.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.6630094827,"x_train":413.5799198557,"imp":9.3688825164,"elo":1528.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":3.2283609867,"x_train":828.216131447,"imp":9.0618942855,"elo":1542.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.4759145961,"x_train":1248.051456359,"imp":8.9494988014,"elo":1555.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.446364541,"x_train":1673.3760165809,"imp":8.9390070431,"elo":1551.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_small/winrate_explorer.html b/data/imputation_no/splits_lite/tasks_regression/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..546bdc2212240cfa71bd8b4fec6fdabbe0ed34bb
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_no/splits_lite/tasks_regression/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "Nori-30M (default)", "Nori (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabDPT (tuned + ensembled)", "RealTabPFN-2.5 (tuned + ensembled)", "TabDPT (tuned)", "TabICLv2 (default)", "TabDPT (default)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabDPT-Turbo (default)", "RealTabPFN-2.5 (default)", "TabSwift (default)", "iLTM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "RealMLP (tuned)", "TabM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "CatBoost (tuned)", "xRFM (tuned + ensembled)", "TabM (tuned)", "xRFM (tuned)", "XGBoost (tuned)", "XGBoost (tuned + ensembled)", "LightGBM (tuned)", "ChimeraBoost (tuned + ensembled)", "ModernNCA (default)", "SAP-RPT-OSS (default)", "iLTM (tuned)", "CatBoost (default)", "ChimeraBoost (tuned)", "RealMLP (default)", "LimiX (default)", "ExtraTrees (tuned + ensembled)", "TabM (default)", "ModernNCA (tuned)", "TorchMLP (tuned + ensembled)", "ExtraTrees (tuned)", "ChimeraBoost (default)", "TorchMLP (tuned)", "EBM (tuned + ensembled)", "EBM (tuned)", "xRFM (default)", "RandomForest (tuned + ensembled)", "LightGBM (default)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (default)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "FastaiMLP (tuned)", "PerpetualBooster (tuned)", "XGBoost (default)", "TabSTAR (tuned)", "EBM (default)", "TorchMLP (default)", "RandomForest (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "TabSTAR (default)", "PerpetualBooster (default)", "iLTM (default)", "KNN (tuned)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 0.75, 0.875, 0.625, 0.75, 0.75, 0.875, 0.75, 1.0, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, NaN, 0.625, 0.75, 0.5, 0.75, 0.625, 0.75, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.375, NaN, 0.5, 0.5, 0.625, 0.75, 0.625, 1.0, 0.875, 0.875, 0.75, 1.0, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.375, 0.25, 0.5, NaN, 0.5, 0.625, 0.625, 0.5, 0.75, 0.625, 0.875, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.5, 0.5, 0.5, NaN, 0.5, 0.5, 0.375, 0.625, 0.625, 1.0, 0.75, 0.875, 0.875, 1.0, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.25, 0.375, 0.375, 0.5, NaN, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.375, 0.25, 0.375, 0.5, 0.375, NaN, 0.375, 0.75, 0.625, 0.875, 0.5, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.25, 0.375, 0.5, 0.625, 0.375, 0.625, NaN, 0.625, 0.625, 0.75, 0.875, 0.875, 0.75, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.25, 0.375, 0.375, 0.25, 0.375, NaN, 0.5, 0.75, 0.625, 0.875, 0.875, 0.875, 0.5, 0.75, 0.875, 1.0, 0.875, 0.75, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.375, 0.375, 0.25, 0.375, 0.375, 0.5, NaN, 0.5, 0.75, 0.625, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.75, 0.875, 0.75, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.125, 0.25, 0.25, 0.5, NaN, 0.375, 0.625, 0.625, 0.75, 0.375, 0.625, 0.625, 0.75, 0.625, 0.75, 0.875, 0.75, 0.875, 1.0, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 0.75, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.5, 0.125, 0.375, 0.25, 0.625, NaN, 0.625, 0.625, 0.75, 0.75, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.75, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.375, 0.375, 0.375, NaN, 0.5, 0.625, 0.5, 0.625, 0.75, 0.75, 0.75, 0.625, 0.875, 0.625, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 1.0, 0.75, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.375, 0.375, 0.5, NaN, 0.375, 0.5, 0.5, 0.5, 0.875, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.375, 0.625, NaN, 0.25, 0.625, 0.625, 0.75, 0.875, 0.5, 0.875, 0.625, 0.75, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.125, 0.5, 0.375, 0.625, 0.25, 0.5, 0.5, 0.75, NaN, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 0.625, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0], [0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.25, NaN, 0.625, 0.625, 0.625, 0.625, 0.75, 0.375, 0.75, 0.625, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 1.0, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.375, 0.375, 0.25, 0.5, 0.375, 0.25, 0.375, NaN, 0.5, 0.5, 0.625, 0.75, 0.5, 0.75, 0.5, 0.75, 0.625, 0.875, 0.875, 0.875, 0.75, 0.875, 0.75, 0.75, 0.875, 0.875, 0.75, 0.625, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 0.75, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.5, NaN, 0.5, 0.625, 0.5, 0.375, 0.5, 0.5, 0.625, 0.625, 0.75, 0.75, 1.0, 0.875, 0.875, 0.625, 0.625, 0.875, 1.0, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.375, 0.25, 0.25, 0.375, 0.125, 0.375, 0.375, 0.5, 0.5, NaN, 0.5, 0.625, 0.375, 0.75, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.625, 0.75, 0.875, 0.75, 0.75, 1.0, 0.875, 0.75, 0.75, 0.75, 1.0, 1.0, 0.875, 0.875, 0.75, 0.875, 1.0, 0.75, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.5, 0.25, 0.375, 0.375, 0.375, 0.5, NaN, 0.5, 0.5, 0.375, 0.625, 0.875, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.75, 0.875, 0.75, 0.75, 0.75, 1.0, 0.75, 0.875, 0.75, 0.875, 0.875, 0.75, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.5, 0.375, 0.5, NaN, 0.625, 0.625, 0.25, 0.5, 0.5, 0.75, 0.75, 0.875, 0.625, 0.625, 0.75, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.375, 0.375, 0.25, 0.375, 0.25, 0.625, 0.5, 0.625, 0.625, 0.5, 0.375, NaN, 0.375, 0.5, 0.5, 0.5, 0.625, 0.75, 0.875, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 1.0, 0.625, 0.875, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.375, 0.25, 0.25, 0.25, 0.25, 0.5, 0.25, 0.625, 0.375, 0.625, NaN, 0.5, 0.625, 0.5, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.875, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.25, 0.0, 0.25, 0.25, 0.25, 0.0, 0.25, 0.375, 0.5, 0.5, 0.375, 0.375, 0.75, 0.5, 0.5, NaN, 0.5, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.625, 0.75, 0.875, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.125, 0.5, 0.5, 0.375, 0.5, NaN, 0.5, 0.5, 0.5, 0.5, 0.75, 0.375, 0.5, 0.625, 0.75, 0.75, 0.75, 0.625, 0.75, 1.0, 0.625, 0.875, 0.75, 0.75, 1.0, 0.75, 0.75, 0.75, 0.75, 1.0, 0.875, 0.875, 0.75, 0.875, 0.75, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.25, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.375, 0.375, 0.25, 0.375, 0.5, 0.5, 0.5, 0.25, 0.5, NaN, 0.625, 0.5, 0.5, 0.375, 0.75, 0.625, 0.75, 0.625, 0.5, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.875, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.25, 0.375, 0.25, 0.375, 0.5, 0.375, NaN, 0.625, 0.5, 0.625, 0.625, 0.5, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.625, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.5, 0.5, 0.5, 0.375, NaN, 0.5, 0.5, 0.5, 0.375, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.125, 0.0, 0.25, 0.375, 0.125, 0.125, 0.25, 0.375, 0.5, 0.5, 0.5, 0.5, NaN, 0.5, 0.625, 0.5, 0.625, 0.5, 0.625, 0.75, 0.5, 0.625, 0.875, 0.875, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.75, 0.75, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.25, 0.125, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.625, 0.375, 0.5, 0.5, NaN, 0.375, 0.5, 0.5, 0.5, 0.75, 0.75, 0.5, 0.625, 0.75, 0.625, 0.625, 0.625, 0.875, 0.75, 0.875, 0.875, 0.75, 0.75, 0.625, 1.0, 0.625, 0.75, 1.0, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.25, 0.5, 0.375, 0.375, 0.25, 0.375, 0.625, 0.25, 0.375, 0.5, 0.375, 0.625, NaN, 0.375, 0.5, 0.625, 0.625, 0.625, 0.5, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.75, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.125, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.625, 0.5, 0.5, 0.625, NaN, 0.5, 0.5, 0.625, 0.625, 0.75, 0.5, 0.5, 0.5, 0.625, 0.625, 0.75, 0.625, 0.625, 0.75, 0.625, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.625, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.375, 0.375, 0.5, 0.5, 0.5, NaN, 0.5, 0.625, 0.625, 0.625, 0.5, 0.75, 0.625, 0.5, 0.5, 0.75, 0.5, 0.625, 0.75, 0.875, 0.75, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.375, 0.25, 0.125, 0.375, 0.125, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, 0.5, 0.5, NaN, 0.625, 0.625, 0.375, 0.625, 0.625, 0.625, 0.625, 0.875, 1.0, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.25, 0.125, 0.0, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.5, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, NaN, 0.75, 0.375, 0.5, 0.625, 0.625, 0.625, 0.5, 0.75, 0.5, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 1.0, 0.75, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.25, NaN, 0.5, 0.625, 0.375, 0.375, 0.5, 0.625, 0.625, 0.5, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 1.0, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.5, 0.5, 0.5, 0.25, 0.375, 0.625, 0.625, 0.5, NaN, 0.5, 0.5, 0.5, 0.5, 0.5, 0.625, 0.375, 0.625, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.5, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.625, 0.625, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.25, 0.0, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.5, 0.5, 0.375, 0.5, 0.375, 0.5, NaN, 0.375, 0.625, 0.5, 0.625, 0.75, 0.5, 0.625, 0.625, 0.75, 1.0, 0.5, 0.75, 1.0, 1.0, 0.75, 0.75, 0.625, 0.875, 0.625, 0.75, 0.75, 0.625, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.25, 0.25, 0.0, 0.375, 0.25, 0.125, 0.125, 0.25, 0.375, 0.5, 0.25, 0.375, 0.375, 0.625, 0.5, 0.625, NaN, 0.5, 0.5, 0.625, 0.625, 0.75, 0.625, 0.625, 0.625, 0.625, 0.875, 0.75, 0.625, 0.625, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.0, 0.125, 0.25, 0.375, 0.25, 0.25, 0.25, 0.125, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.625, 0.5, 0.375, 0.5, NaN, 0.5, 0.375, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.625, 0.625, 0.75, 0.625, 0.875, 0.75, 0.875, 0.625, 0.875, 0.875, 0.75, 1.0, 0.875, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.375, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, NaN, 0.625, 0.5, 0.75, 0.625, 0.625, 0.625, 0.625, 0.875, 0.875, 0.625, 0.625, 0.75, 0.75, 0.875, 0.75, 0.875, 0.75, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.25, 0.0, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.125, 0.5, 0.375, 0.5, 0.375, 0.375, 0.625, 0.375, NaN, 0.5, 0.5, 0.625, 0.625, 0.75, 0.75, 0.375, 0.75, 0.75, 0.875, 0.75, 0.75, 0.625, 0.875, 0.5, 0.75, 0.75, 0.5, 1.0, 0.75, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.125, 0.0, 0.25, 0.0, 0.25, 0.25, 0.125, 0.125, 0.125, 0.375, 0.125, 0.25, 0.25, 0.25, 0.0, 0.25, 0.375, 0.375, 0.25, 0.375, 0.5, 0.5, 0.5, NaN, 0.5, 0.75, 0.75, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 1.0, 0.75, 0.625, 0.875, 0.75, 0.75, 0.75, 0.625, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.25, 0.0, 0.375, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.5, 0.375, 0.5, 0.5, 0.625, 0.5, 0.25, 0.375, 0.25, 0.5, 0.5, NaN, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.75, 1.0, 0.625, 0.875, 0.75, 0.875, 1.0, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.0, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.5, NaN, 0.75, 0.625, 0.375, 0.5, 0.625, 0.375, 0.375, 0.5, 0.875, 0.75, 0.75, 0.625, 0.875, 1.0, 0.75, 0.625, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.375, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.5, 0.25, NaN, 0.625, 0.375, 0.5, 0.375, 0.375, 0.375, 0.5, 0.75, 0.625, 0.5, 0.625, 0.75, 0.875, 0.75, 0.625, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.125, 0.25, 0.25, 0.25, 0.5, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, NaN, 0.5, 0.375, 0.25, 0.5, 0.625, 0.375, 0.625, 0.375, 0.375, 0.375, 0.75, 0.625, 0.5, 0.75, 0.625, 1.0, 0.875, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.125, 0.0, 0.125, 0.125, 0.0, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.5, 0.0, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.625, 0.625, 0.5, NaN, 0.5, 0.375, 0.75, 0.875, 0.625, 0.625, 0.375, 0.625, 0.625, 0.625, 0.625, 0.5, 1.0, 0.75, 1.0, 0.875, 0.625, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.375, 0.125, 0.125, 0.125, 0.125, 0.375, 0.25, 0.375, 0.5, 0.125, 0.25, 0.125, 0.625, 0.375, 0.375, 0.5, 0.5, 0.625, 0.5, NaN, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.375, 0.625, 0.75, 0.625, 0.375, NaN, 0.625, 0.625, 0.375, 0.75, 0.625, 0.375, 0.625, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.125, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.375, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.375, 0.0, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.625, 0.625, 0.5, 0.25, 0.375, 0.375, NaN, 0.5, 0.5, 0.625, 0.375, 0.5, 0.375, 0.625, 0.625, 0.375, 0.75, 0.625, 1.0, 0.75, 0.625, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.5, 0.0, 0.375, 0.375, 0.375, 0.125, 0.25, 0.375, 0.625, 0.625, 0.375, 0.125, 0.375, 0.375, 0.5, NaN, 0.5, 0.625, 0.375, 0.5, 0.375, 0.625, 0.625, 0.375, 0.75, 0.625, 1.0, 0.75, 0.625, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.25, 0.125, 0.0, 0.0, 0.375, 0.375, 0.25, 0.125, 0.25, 0.25, 0.25, 0.0, 0.25, 0.5, 0.5, 0.625, 0.375, 0.375, 0.625, 0.5, 0.5, NaN, 0.625, 0.625, 0.6875, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.0, 0.25, 0.0, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, NaN, 0.25, 0.375, 0.25, 0.875, 0.375, 0.375, 0.375, 0.5, 0.5, 1.0, 0.375, 0.5, 0.625, 0.625, 0.75, 0.75, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.375, 0.375, 0.375, 0.125, 0.125, 0.125, 0.375, 0.375, 0.125, 0.25, 0.375, 0.625, 0.625, 0.375, 0.375, 0.625, 0.625, 0.375, 0.75, NaN, 0.375, 0.5, 0.75, 0.5, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.375, 0.125, 0.0, 0.125, 0.25, 0.375, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.5, 0.625, 0.375, 0.375, 0.625, 0.5, 0.5, 0.3125, 0.625, 0.625, NaN, 0.5, 0.625, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.25, 0.125, 0.0, 0.125, 0.125, 0.0, 0.375, 0.375, 0.0, 0.125, 0.125, 0.5, 0.25, 0.25, 0.375, 0.375, 0.625, 0.375, 0.25, 0.375, 0.625, 0.625, 0.375, 0.75, 0.5, 0.5, NaN, 0.75, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.125, 0.25, 0.375, 0.25, NaN, 0.375, 0.375, 0.375, 0.5, 0.5, 1.0, 0.375, 0.5, 0.625, 0.625, 0.75, 0.75, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.25, 0.125, 0.125, 0.0, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.0, 0.125, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.625, 0.5, 0.25, 0.375, 0.625, NaN, 0.75, 0.5, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.125, 0.125, 0.25, 0.375, 0.125, 0.125, 0.0, 0.5, 0.375, 0.0, 0.25, 0.25, 0.5, 0.5, 0.25, 0.25, 0.625, 0.625, 0.375, 0.625, 0.25, 0.375, 0.25, 0.625, 0.25, NaN, 0.625, 0.25, 0.625, 0.75, 0.5, 0.75, 0.75, 0.75, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.375, 0.0, 0.125, 0.25, 0.125, 0.0, 0.125, 0.375, 0.375, 0.375, 0.25, 0.0, 0.125, 0.125, 0.25, 0.25, 0.375, 0.625, 0.25, 0.375, 0.375, 0.625, 0.5, 0.375, NaN, 0.5, 0.75, 0.75, 0.625, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.375, 0.125, 0.125, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.375, 0.25, 0.125, 0.0, 0.375, 0.375, 0.25, 0.5, 0.375, 0.25, 0.375, 0.5, 0.375, 0.75, 0.5, NaN, 0.5, 0.5, 0.75, 0.625, 0.75, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.375, 0.5, 0.25, 0.375, 0.25, 0.5, 0.25, 0.375, 0.25, 0.5, NaN, 0.5, 0.5, 0.25, 1.0, 1.0, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.0, 0.125, 0.25, 0.25, 0.0, 0.125, 0.25, 0.25, 0.5, 0.5, NaN, 0.25, 0.375, 0.5, 0.5, 0.75, 0.75, 0.75], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.25, 0.375, 0.125, 0.125, 0.375, 0.375, 0.0, 0.625, 0.125, 0.125, 0.125, 0.625, 0.125, 0.5, 0.375, 0.25, 0.5, 0.75, NaN, 0.5, 0.625, 0.75, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.5, 0.125, 0.125, 0.125, 0.5, 0.125, 0.25, 0.25, 0.375, 0.75, 0.625, 0.5, NaN, 0.75, 0.75, 0.875, 0.875, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.375, 0.125, 0.125, 0.0, 0.375, 0.125, 0.25, 0.0, 0.25, 0.0, 0.5, 0.375, 0.25, NaN, 0.75, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.375, 0.125, 0.125, 0.0, 0.375, 0.125, 0.25, 0.0, 0.125, 0.0, 0.5, 0.25, 0.25, 0.25, NaN, 0.625, 0.625, 0.625], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.125, 0.25, 0.0, 0.125, 0.0, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.375, NaN, 0.6875, 0.75], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.125, 0.25, 0.0, 0.125, 0.0, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.375, 0.3125, NaN, 0.75], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.125, 0.0, 0.375, 0.25, 0.25, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9665492958},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.9507042254},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.9348591549},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.926056338},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.9084507042},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8996478873},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8802816901},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.875},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.8503521127},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.7957746479},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7816901408},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7799295775},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7623239437},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7535211268},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7535211268},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7447183099},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.7183098592},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7024647887},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6795774648},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6690140845},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6496478873},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6496478873},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6478873239},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6461267606},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6285211268},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6003521127},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5915492958},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5862676056},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5809859155},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5616197183},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5457746479},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5404929577},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5404929577},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5281690141},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5246478873},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4806338028},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4683098592},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4665492958},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4665492958},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4542253521},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.4524647887},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.448943662},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4348591549},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4084507042},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4066901408},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3996478873},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.3767605634},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3556338028},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3556338028},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3538732394},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3257042254},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.323943662},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3116197183},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.298415493},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2957746479},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2957746479},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2860915493},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.2852112676},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2605633803},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.2570422535},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2341549296},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2235915493},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.198943662},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1654929577},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.161971831},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1566901408},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.1373239437},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1003521127},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0809859155},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0501760563},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0448943662},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0211267606}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_no/splits_lite/tasks_regression/datasets_small/winrate_matrix.csv b/data/imputation_no/splits_lite/tasks_regression/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..4b0555689b4c676c0f79887d8cf8820a49c64a2e
--- /dev/null
+++ b/data/imputation_no/splits_lite/tasks_regression/datasets_small/winrate_matrix.csv
@@ -0,0 +1,73 @@
+,TabFM (default),Nori-30M (default),Nori (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),TabDPT (tuned + ensembled),RealTabPFN-2.5 (tuned + ensembled),TabDPT (tuned),TabICLv2 (default),TabDPT (default),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabDPT-Turbo (default),RealTabPFN-2.5 (default),TabSwift (default),iLTM (tuned + ensembled),LightGBM (tuned + ensembled),RealMLP (tuned),TabM (tuned + ensembled),CatBoost (tuned + ensembled),ModernNCA (tuned + ensembled),CatBoost (tuned),xRFM (tuned + ensembled),TabM (tuned),xRFM (tuned),XGBoost (tuned),XGBoost (tuned + ensembled),LightGBM (tuned),ChimeraBoost (tuned + ensembled),ModernNCA (default),SAP-RPT-OSS (default),iLTM (tuned),CatBoost (default),ChimeraBoost (tuned),RealMLP (default),LimiX (default),ExtraTrees (tuned + ensembled),TabM (default),ModernNCA (tuned),TorchMLP (tuned + ensembled),ExtraTrees (tuned),ChimeraBoost (default),TorchMLP (tuned),EBM (tuned + ensembled),EBM (tuned),xRFM (default),RandomForest (tuned + ensembled),LightGBM (default),FastaiMLP (tuned + ensembled),ExtraTrees (default),RandomForest (tuned),PerpetualBooster (tuned + ensembled),TabSTAR (tuned + ensembled),FastaiMLP (tuned),PerpetualBooster (tuned),XGBoost (default),TabSTAR (tuned),EBM (default),TorchMLP (default),RandomForest (default),FastaiMLP (default),KNN (tuned + ensembled),TabSTAR (default),PerpetualBooster (default),iLTM (default),KNN (tuned),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,0.75,0.875,0.625,0.75,0.75,0.875,0.75,1.0,0.875,1.0,0.75,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori-30M (default),0.25,,0.625,0.75,0.5,0.75,0.625,0.75,1.0,0.875,0.875,0.875,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori (default),0.125,0.375,,0.5,0.5,0.625,0.75,0.625,1.0,0.875,0.875,0.75,1.0,0.875,1.0,0.75,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.375,0.25,0.5,,0.5,0.625,0.625,0.5,0.75,0.625,0.875,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.25,0.5,0.5,0.5,,0.5,0.5,0.375,0.625,0.625,1.0,0.75,0.875,0.875,1.0,0.75,0.875,1.0,1.0,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.25,0.25,0.375,0.375,0.5,,0.625,0.625,0.625,0.75,0.75,0.75,0.875,0.75,0.75,0.875,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,0.875,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.125,0.375,0.25,0.375,0.5,0.375,,0.375,0.75,0.625,0.875,0.5,0.875,0.875,0.875,0.75,1.0,0.875,0.875,0.875,0.875,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.25,0.25,0.375,0.5,0.625,0.375,0.625,,0.625,0.625,0.75,0.875,0.875,0.75,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,1.0,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,0.875,0.875,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.0,0.0,0.25,0.375,0.375,0.25,0.375,,0.5,0.75,0.625,0.875,0.875,0.875,0.5,0.75,0.875,1.0,0.875,0.75,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.125,0.125,0.125,0.375,0.375,0.25,0.375,0.375,0.5,,0.5,0.75,0.625,0.75,0.75,0.625,0.75,0.875,0.875,0.875,0.75,0.875,0.75,0.875,0.75,0.75,0.75,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.75,1.0,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.125,0.125,0.125,0.0,0.25,0.125,0.25,0.25,0.5,,0.375,0.625,0.625,0.75,0.375,0.625,0.625,0.75,0.625,0.75,0.875,0.75,0.875,1.0,0.75,1.0,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,1.0,0.875,0.75,1.0,1.0,0.875,0.875,1.0,1.0,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.25,0.125,0.25,0.25,0.25,0.25,0.5,0.125,0.375,0.25,0.625,,0.625,0.625,0.75,0.75,0.625,0.625,0.75,0.75,0.75,0.75,0.625,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.75,0.875,0.875,0.75,1.0,0.875,1.0,1.0,0.875,0.875,0.75,0.875,1.0,1.0,0.875,0.875,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.375,0.375,0.375,,0.5,0.625,0.5,0.625,0.75,0.75,0.75,0.625,0.875,0.625,0.875,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,1.0,1.0,0.75,1.0,1.0,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.0,0.125,0.0,0.125,0.25,0.125,0.25,0.125,0.25,0.375,0.375,0.5,,0.375,0.5,0.5,0.5,0.875,0.625,0.625,0.75,0.75,0.625,0.75,0.75,0.75,0.875,1.0,1.0,1.0,0.875,0.75,0.875,1.0,1.0,0.875,0.75,0.875,0.875,0.875,1.0,0.875,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.0,0.0,0.0,0.0,0.25,0.125,0.125,0.125,0.25,0.25,0.25,0.375,0.625,,0.25,0.625,0.625,0.75,0.875,0.5,0.875,0.625,0.75,1.0,0.75,0.875,1.0,1.0,1.0,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.75,1.0,1.0,0.875,1.0,1.0,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.125,0.125,0.25,0.125,0.25,0.125,0.25,0.125,0.5,0.375,0.625,0.25,0.5,0.5,0.75,,0.75,0.75,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.875,0.75,0.75,0.75,0.875,0.75,0.75,0.75,0.875,0.875,0.75,0.625,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,1.0,0.875,0.875,1.0,0.875,1.0,1.0,0.875,0.875,1.0,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0
+TabSwift (default),0.125,0.125,0.0,0.0,0.125,0.125,0.0,0.25,0.25,0.25,0.375,0.375,0.375,0.5,0.375,0.25,,0.625,0.625,0.625,0.625,0.75,0.375,0.75,0.625,0.75,0.875,0.875,0.75,0.75,0.75,0.875,0.875,0.75,0.75,0.75,0.875,0.75,0.75,0.875,1.0,0.875,0.75,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.375,0.375,0.25,0.5,0.375,0.25,0.375,,0.5,0.5,0.625,0.75,0.5,0.75,0.5,0.75,0.625,0.875,0.875,0.875,0.75,0.875,0.75,0.75,0.875,0.875,0.75,0.625,1.0,0.875,0.875,0.875,1.0,1.0,0.875,0.75,0.75,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.25,0.25,0.25,0.125,0.25,0.25,0.375,0.5,,0.5,0.625,0.5,0.375,0.5,0.5,0.625,0.625,0.75,0.75,1.0,0.875,0.875,0.625,0.625,0.875,1.0,0.75,0.75,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.375,0.25,0.25,0.375,0.125,0.375,0.375,0.5,0.5,,0.5,0.625,0.375,0.75,0.625,0.625,0.75,0.75,0.75,0.75,0.625,0.75,0.625,0.75,0.625,0.75,0.875,0.75,0.75,1.0,0.875,0.75,0.75,0.75,1.0,1.0,0.875,0.875,0.75,0.875,1.0,0.75,0.875,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.25,0.25,0.25,0.25,0.375,0.375,0.5,0.25,0.375,0.375,0.375,0.5,,0.5,0.5,0.375,0.625,0.875,0.625,0.625,0.625,0.625,0.625,0.5,0.625,0.625,0.75,0.875,0.75,0.75,0.75,1.0,0.75,0.875,0.75,0.875,0.875,0.75,0.875,0.75,0.875,1.0,0.875,0.875,0.875,0.875,0.75,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.125,0.0,0.125,0.125,0.25,0.125,0.25,0.125,0.25,0.25,0.25,0.5,0.375,0.5,,0.625,0.625,0.25,0.5,0.5,0.75,0.75,0.875,0.625,0.625,0.75,0.75,0.875,0.875,0.75,0.75,0.75,0.75,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.875,1.0,0.875,0.875,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.375,0.375,0.25,0.375,0.25,0.625,0.5,0.625,0.625,0.5,0.375,,0.375,0.5,0.5,0.5,0.625,0.75,0.875,0.75,0.625,0.625,0.75,0.625,0.75,0.625,0.625,0.75,0.75,1.0,0.625,0.875,0.75,0.75,0.75,0.75,0.875,0.875,0.75,1.0,0.875,0.875,0.875,0.75,1.0,0.875,0.875,0.75,0.875,0.875,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.375,0.25,0.25,0.25,0.25,0.5,0.25,0.625,0.375,0.625,,0.5,0.625,0.5,0.75,0.75,0.75,0.625,0.75,0.625,0.75,0.875,0.75,0.75,0.75,0.75,0.75,0.875,0.875,0.75,1.0,0.875,0.875,0.875,0.75,0.875,0.875,0.875,0.875,1.0,1.0,0.75,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.25,0.0,0.25,0.25,0.25,0.0,0.25,0.375,0.5,0.5,0.375,0.375,0.75,0.5,0.5,,0.5,0.75,0.625,0.5,0.625,0.625,0.625,0.625,0.75,0.75,0.75,0.625,0.75,0.875,0.75,0.75,0.75,0.875,0.75,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,0.875,0.75,0.875,0.875,1.0,0.75,0.875,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.125,0.5,0.5,0.375,0.5,,0.5,0.5,0.5,0.5,0.75,0.375,0.5,0.625,0.75,0.75,0.75,0.625,0.75,1.0,0.625,0.875,0.75,0.75,1.0,0.75,0.75,0.75,0.75,1.0,0.875,0.875,0.75,0.875,0.75,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.25,0.0,0.125,0.125,0.25,0.125,0.125,0.125,0.375,0.375,0.25,0.375,0.5,0.5,0.5,0.25,0.5,,0.625,0.5,0.5,0.375,0.75,0.625,0.75,0.625,0.5,0.75,0.625,0.75,0.625,0.75,0.75,0.75,0.875,0.625,0.75,0.75,0.875,0.875,0.875,1.0,0.875,0.875,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.25,0.125,0.125,0.25,0.25,0.375,0.25,0.375,0.25,0.375,0.5,0.375,,0.625,0.5,0.625,0.625,0.5,0.625,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.875,0.625,0.875,0.875,0.875,0.875,0.75,0.875,1.0,1.0,0.875,0.875,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.0,0.25,0.25,0.125,0.25,0.25,0.375,0.25,0.25,0.25,0.5,0.5,0.5,0.375,,0.5,0.5,0.5,0.375,0.625,0.625,0.625,0.75,0.75,0.75,0.875,0.75,0.875,0.75,0.875,0.875,0.875,0.875,0.75,0.875,1.0,1.0,0.875,0.875,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.0,0.25,0.25,0.125,0.0,0.25,0.375,0.125,0.125,0.25,0.375,0.5,0.5,0.5,0.5,,0.5,0.625,0.5,0.625,0.5,0.625,0.75,0.5,0.625,0.875,0.875,0.75,0.75,0.625,0.75,0.875,0.875,0.875,0.75,0.875,1.0,0.75,0.75,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.25,0.25,0.125,0.375,0.375,0.375,0.25,0.375,0.375,0.25,0.625,0.375,0.5,0.5,,0.375,0.5,0.5,0.5,0.75,0.75,0.5,0.625,0.75,0.625,0.625,0.625,0.875,0.75,0.875,0.875,0.75,0.75,0.625,1.0,0.625,0.75,1.0,0.75,0.875,1.0,0.875,0.875,1.0,0.875,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.25,0.5,0.375,0.375,0.25,0.375,0.625,0.25,0.375,0.5,0.375,0.625,,0.375,0.5,0.625,0.625,0.625,0.5,0.75,0.625,0.625,0.625,0.625,0.75,0.625,0.625,0.625,0.75,0.875,0.875,0.875,0.875,0.875,0.75,0.75,0.875,0.875,0.75,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.25,0.25,0.25,0.25,0.125,0.25,0.125,0.25,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.5,0.375,0.5,0.625,0.5,0.5,0.625,,0.5,0.5,0.625,0.625,0.75,0.5,0.5,0.5,0.625,0.625,0.75,0.625,0.625,0.75,0.625,0.875,0.875,0.75,0.75,0.875,0.75,0.75,0.75,0.625,0.875,0.75,0.75,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.375,0.25,0.375,0.25,0.25,0.25,0.25,0.375,0.25,0.375,0.375,0.375,0.5,0.5,0.5,,0.5,0.625,0.625,0.625,0.5,0.75,0.625,0.5,0.5,0.75,0.5,0.625,0.75,0.875,0.75,0.875,0.875,0.75,0.75,0.875,0.75,0.875,0.875,1.0,0.75,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.25,0.125,0.125,0.375,0.25,0.125,0.375,0.125,0.25,0.25,0.375,0.375,0.375,0.5,0.5,0.375,0.5,0.5,,0.625,0.625,0.375,0.625,0.625,0.625,0.625,0.875,1.0,0.625,0.75,0.75,0.75,0.75,0.875,0.875,0.875,0.875,1.0,0.75,0.875,1.0,0.875,0.875,0.875,0.75,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.125,0.125,0.25,0.125,0.0,0.25,0.125,0.125,0.25,0.25,0.25,0.25,0.5,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.375,,0.75,0.375,0.5,0.625,0.625,0.625,0.5,0.75,0.5,0.625,0.625,0.75,0.75,0.625,0.75,0.625,0.75,1.0,0.75,0.875,0.875,0.875,0.75,1.0,0.875,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.25,0.25,0.375,0.25,0.375,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.25,,0.5,0.625,0.375,0.375,0.5,0.625,0.625,0.5,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.625,0.75,0.625,0.75,1.0,0.75,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.375,0.25,0.25,0.25,0.25,0.375,0.25,0.25,0.375,0.375,0.25,0.25,0.5,0.5,0.5,0.25,0.375,0.625,0.625,0.5,,0.5,0.5,0.5,0.5,0.5,0.625,0.375,0.625,0.625,0.5,0.5,0.625,0.625,0.625,0.5,0.625,0.75,0.625,0.625,0.625,0.75,0.75,0.75,0.625,0.625,0.875,0.875,0.875,0.75,0.875,1.0,0.875,0.875,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.125,0.25,0.0,0.125,0.25,0.25,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.375,0.375,0.25,0.5,0.5,0.375,0.5,0.375,0.5,,0.375,0.625,0.5,0.625,0.75,0.5,0.625,0.625,0.75,1.0,0.5,0.75,1.0,1.0,0.75,0.75,0.625,0.875,0.625,0.75,0.75,0.625,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.0,0.25,0.25,0.25,0.25,0.0,0.375,0.25,0.125,0.125,0.25,0.375,0.5,0.25,0.375,0.375,0.625,0.5,0.625,,0.5,0.5,0.625,0.625,0.75,0.625,0.625,0.625,0.625,0.875,0.75,0.625,0.625,0.875,0.75,0.875,0.875,1.0,0.75,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.0,0.125,0.125,0.125,0.25,0.125,0.0,0.125,0.25,0.375,0.25,0.25,0.25,0.125,0.375,0.375,0.5,0.375,0.375,0.375,0.625,0.5,0.375,0.5,,0.5,0.375,0.5,0.625,0.625,0.625,0.625,0.625,0.75,0.75,0.625,0.625,0.75,0.625,0.875,0.75,0.875,0.625,0.875,0.875,0.75,1.0,0.875,0.75,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.375,0.125,0.25,0.125,0.25,0.125,0.125,0.25,0.375,0.375,0.375,0.5,0.375,0.375,0.5,0.5,0.5,0.5,0.5,,0.625,0.5,0.75,0.625,0.625,0.625,0.625,0.875,0.875,0.625,0.625,0.75,0.75,0.875,0.75,0.875,0.75,0.75,1.0,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.125,0.25,0.0,0.125,0.25,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.375,0.25,0.25,0.375,0.375,0.375,0.5,0.125,0.5,0.375,0.5,0.375,0.375,0.625,0.375,,0.5,0.5,0.625,0.625,0.75,0.75,0.375,0.75,0.75,0.875,0.75,0.75,0.625,0.875,0.5,0.75,0.75,0.5,1.0,0.75,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.0,0.0,0.25,0.125,0.0,0.25,0.0,0.25,0.25,0.125,0.125,0.125,0.375,0.125,0.25,0.25,0.25,0.0,0.25,0.375,0.375,0.25,0.375,0.5,0.5,0.5,,0.5,0.75,0.75,0.625,0.625,0.625,0.75,0.625,0.75,1.0,0.75,0.625,0.875,0.75,0.75,0.75,0.625,0.875,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.125,0.0,0.125,0.125,0.25,0.125,0.25,0.0,0.375,0.125,0.125,0.25,0.25,0.375,0.375,0.5,0.375,0.5,0.5,0.625,0.5,0.25,0.375,0.25,0.5,0.5,,0.5,0.5,0.625,0.625,0.625,0.625,0.625,0.625,0.75,0.75,0.875,0.75,0.75,0.75,0.75,1.0,0.625,0.875,0.75,0.875,1.0,0.75,0.875,0.875,0.875,0.875,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.0,0.25,0.125,0.25,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.125,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.25,0.5,,0.75,0.625,0.375,0.5,0.625,0.375,0.375,0.5,0.875,0.75,0.75,0.625,0.875,1.0,0.75,0.625,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.25,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.125,0.375,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.25,0.5,0.25,,0.625,0.375,0.5,0.375,0.375,0.375,0.5,0.75,0.625,0.5,0.625,0.75,0.875,0.75,0.625,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.25,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.125,0.25,0.25,0.125,0.25,0.25,0.375,0.125,0.25,0.25,0.25,0.5,0.25,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.375,,0.5,0.375,0.25,0.5,0.625,0.375,0.625,0.375,0.375,0.375,0.75,0.625,0.5,0.75,0.625,1.0,0.875,0.75,0.875,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.125,0.0,0.125,0.125,0.0,0.25,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.25,0.25,0.25,0.25,0.5,0.0,0.375,0.375,0.375,0.25,0.375,0.375,0.625,0.625,0.5,,0.5,0.375,0.75,0.875,0.625,0.625,0.375,0.625,0.625,0.625,0.625,0.5,1.0,0.75,1.0,0.875,0.625,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.125,0.25,0.125,0.0,0.0,0.125,0.0,0.0,0.125,0.375,0.125,0.125,0.125,0.125,0.375,0.25,0.375,0.5,0.125,0.25,0.125,0.625,0.375,0.375,0.5,0.5,0.625,0.5,,0.625,0.625,0.625,0.625,0.75,0.625,0.625,0.75,0.75,0.625,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.25,0.125,0.125,0.25,0.25,0.375,0.25,0.25,0.25,0.125,0.25,0.25,0.375,0.375,0.625,0.75,0.625,0.375,,0.625,0.625,0.375,0.75,0.625,0.375,0.625,0.75,0.75,0.75,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.125,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.25,0.375,0.125,0.25,0.25,0.125,0.375,0.25,0.375,0.0,0.375,0.375,0.375,0.25,0.375,0.375,0.625,0.625,0.5,0.25,0.375,0.375,,0.5,0.5,0.625,0.375,0.5,0.375,0.625,0.625,0.375,0.75,0.625,1.0,0.75,0.625,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.0,0.25,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.5,0.0,0.375,0.375,0.375,0.125,0.25,0.375,0.625,0.625,0.375,0.125,0.375,0.375,0.5,,0.5,0.625,0.375,0.5,0.375,0.625,0.625,0.375,0.75,0.625,1.0,0.75,0.625,0.875,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.25,0.25,0.125,0.0,0.0,0.375,0.375,0.25,0.125,0.25,0.25,0.25,0.0,0.25,0.5,0.5,0.625,0.375,0.375,0.625,0.5,0.5,,0.625,0.625,0.6875,0.625,0.625,0.625,0.625,0.625,0.75,0.625,0.75,1.0,0.875,0.875,0.875,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.125,0.0,0.0,0.25,0.0,0.25,0.0,0.125,0.25,0.125,0.25,0.25,0.125,0.25,0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.25,0.25,0.25,0.125,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.375,,0.25,0.375,0.25,0.875,0.375,0.375,0.375,0.5,0.5,1.0,0.375,0.5,0.625,0.625,0.75,0.75,0.875
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.25,0.125,0.125,0.125,0.375,0.375,0.375,0.125,0.125,0.125,0.375,0.375,0.125,0.25,0.375,0.625,0.625,0.375,0.375,0.625,0.625,0.375,0.75,,0.375,0.5,0.75,0.5,0.75,0.75,0.625,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.375,0.125,0.0,0.125,0.25,0.375,0.125,0.125,0.25,0.25,0.125,0.125,0.25,0.25,0.5,0.625,0.375,0.375,0.625,0.5,0.5,0.3125,0.625,0.625,,0.5,0.625,0.75,0.625,0.625,0.75,0.625,0.75,0.875,0.875,0.875,0.875,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.25,0.125,0.0,0.125,0.125,0.0,0.375,0.375,0.0,0.125,0.125,0.5,0.25,0.25,0.375,0.375,0.625,0.375,0.25,0.375,0.625,0.625,0.375,0.75,0.5,0.5,,0.75,0.625,0.75,0.625,0.625,0.75,0.75,0.875,0.875,1.0,1.0,0.875,0.875,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.125,0.125,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.25,0.25,0.25,0.125,0.25,0.25,0.375,0.25,0.25,0.375,0.375,0.375,0.125,0.25,0.375,0.25,,0.375,0.375,0.375,0.5,0.5,1.0,0.375,0.5,0.625,0.625,0.75,0.75,0.875
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.25,0.125,0.125,0.0,0.125,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.0,0.125,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.625,0.5,0.25,0.375,0.625,,0.75,0.5,0.625,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.25,0.125,0.125,0.25,0.375,0.125,0.125,0.0,0.5,0.375,0.0,0.25,0.25,0.5,0.5,0.25,0.25,0.625,0.625,0.375,0.625,0.25,0.375,0.25,0.625,0.25,,0.625,0.25,0.625,0.75,0.5,0.75,0.75,0.75,0.875,0.875,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.375,0.0,0.125,0.25,0.125,0.0,0.125,0.375,0.375,0.375,0.25,0.0,0.125,0.125,0.25,0.25,0.375,0.625,0.25,0.375,0.375,0.625,0.5,0.375,,0.5,0.75,0.75,0.625,0.75,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.375,0.125,0.125,0.0,0.125,0.25,0.125,0.125,0.125,0.125,0.375,0.25,0.125,0.0,0.375,0.375,0.25,0.5,0.375,0.25,0.375,0.5,0.375,0.75,0.5,,0.5,0.5,0.75,0.625,0.75,0.875,1.0,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.125,0.25,0.125,0.125,0.0,0.0,0.125,0.125,0.0,0.0,0.375,0.5,0.25,0.375,0.25,0.5,0.25,0.375,0.25,0.5,,0.5,0.5,0.25,1.0,1.0,0.875,0.875,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.0,0.125,0.25,0.25,0.0,0.125,0.25,0.25,0.5,0.5,,0.25,0.375,0.5,0.5,0.75,0.75,0.75
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.25,0.375,0.125,0.125,0.375,0.375,0.0,0.625,0.125,0.125,0.125,0.625,0.125,0.5,0.375,0.25,0.5,0.75,,0.5,0.625,0.75,0.875,0.875,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.0,0.0,0.125,0.0,0.0,0.25,0.125,0.125,0.125,0.0,0.0,0.125,0.0,0.125,0.125,0.5,0.125,0.125,0.125,0.5,0.125,0.25,0.25,0.375,0.75,0.625,0.5,,0.75,0.75,0.875,0.875,0.875
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.375,0.125,0.125,0.0,0.375,0.125,0.25,0.0,0.25,0.0,0.5,0.375,0.25,,0.75,0.875,0.875,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.0,0.125,0.125,0.0,0.0,0.125,0.375,0.125,0.125,0.0,0.375,0.125,0.25,0.0,0.125,0.0,0.5,0.25,0.25,0.25,,0.625,0.625,0.625
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.0,0.125,0.25,0.0,0.125,0.0,0.0,0.125,0.25,0.125,0.125,0.125,0.375,,0.6875,0.75
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.0,0.125,0.25,0.0,0.125,0.0,0.0,0.125,0.25,0.125,0.125,0.125,0.375,0.3125,,0.75
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.125,0.0,0.375,0.25,0.25,
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_all/leaderboard_overview_explorer.html
index dfb3180879b69257364188a972e5595e7e038333..3887dc9513828081281d1577ebe60645c495cafa 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — All Tasks | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_all/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..627d994d20d546e7a52f2907949ae97dc9ada0dc
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_all/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1765.0,"score":0.886,"rank":6.81,"hrank":1.79,"imp":4.146,"train_time":38.81,"predict_time":6.985,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-103","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1658.0,"score":0.66,"rank":10.7,"hrank":4.36,"imp":8.048,"train_time":289.07,"predict_time":4.031,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-64","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1641.0,"score":0.672,"rank":11.5,"hrank":4.33,"imp":9.398,"train_time":3.66,"predict_time":0.399,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-54","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1593.0,"score":0.587,"rank":13.92,"hrank":8.14,"imp":10.82,"train_time":5.48,"predict_time":0.555,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-46","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1573.0,"score":0.553,"rank":15.02,"hrank":6.99,"imp":10.497,"train_time":2040.22,"predict_time":8.908,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-56","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1570.0,"score":0.61,"rank":15.21,"hrank":5.33,"imp":10.245,"train_time":2.05,"predict_time":0.151,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-58","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1533.0,"score":0.51,"rank":17.47,"hrank":8.06,"imp":11.186,"train_time":2040.22,"predict_time":1.218,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-48","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1504.0,"score":0.48,"rank":19.4,"hrank":10.78,"imp":11.729,"train_time":5.81,"predict_time":0.645,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-43","imputed":false},{"position":8,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1482.0,"score":0.402,"rank":20.93,"hrank":8.99,"imp":12.856,"train_time":1735.72,"predict_time":2.558,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-46","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1481.0,"score":0.411,"rank":21.07,"hrank":12.7,"imp":12.949,"train_time":2950.72,"predict_time":11.975,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-44","imputed":false},{"position":10,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1437.0,"score":0.387,"rank":24.36,"hrank":10.45,"imp":14.066,"train_time":2.07,"predict_time":0.182,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-46","imputed":false},{"position":11,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1437.0,"score":0.381,"rank":24.37,"hrank":8.74,"imp":13.855,"train_time":4910.38,"predict_time":286.537,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-44","imputed":false},{"position":12,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1425.0,"score":0.319,"rank":25.34,"hrank":13.73,"imp":14.148,"train_time":2450.13,"predict_time":2.247,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-40","imputed":false},{"position":13,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1408.0,"score":0.27,"rank":26.69,"hrank":18.86,"imp":15.099,"train_time":417.05,"predict_time":2.639,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+33\/-29","imputed":false},{"position":14,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1402.0,"score":0.305,"rank":27.17,"hrank":14.98,"imp":14.268,"train_time":2950.72,"predict_time":0.663,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-43","imputed":false},{"position":15,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1397.0,"score":0.273,"rank":27.61,"hrank":16.11,"imp":14.721,"train_time":1346.21,"predict_time":0.344,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-34","imputed":false},{"position":16,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1385.0,"score":0.256,"rank":28.65,"hrank":16.56,"imp":14.95,"train_time":1346.21,"predict_time":0.038,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-38","imputed":false},{"position":17,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1384.0,"score":0.316,"rank":28.69,"hrank":12.4,"imp":15.003,"train_time":4910.38,"predict_time":39.96,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-51","imputed":false},{"position":18,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1383.0,"score":0.248,"rank":28.82,"hrank":19.66,"imp":15.366,"train_time":12685.08,"predict_time":464.37,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+41\/-40","imputed":false},{"position":19,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1372.0,"score":0.254,"rank":29.79,"hrank":15.86,"imp":15.018,"train_time":2450.13,"predict_time":0.251,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+44\/-40","imputed":false},{"position":20,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1368.0,"score":0.284,"rank":30.13,"hrank":13.69,"imp":15.661,"train_time":4621.67,"predict_time":8.138,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-51","imputed":false},{"position":21,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1366.0,"score":0.206,"rank":30.33,"hrank":22.88,"imp":15.752,"train_time":417.05,"predict_time":0.334,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+25\/-27","imputed":false},{"position":22,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1357.0,"score":0.234,"rank":31.1,"hrank":17.32,"imp":15.589,"train_time":5.88,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-39","imputed":false},{"position":23,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1355.0,"score":0.21,"rank":31.2,"hrank":21.76,"imp":15.818,"train_time":693.49,"predict_time":1.689,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-33","imputed":false},{"position":24,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1344.0,"score":0.295,"rank":32.2,"hrank":10.66,"imp":15.593,"train_time":26.46,"predict_time":6.242,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-57","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1337.0,"score":0.202,"rank":32.79,"hrank":18.45,"imp":16.107,"train_time":4621.67,"predict_time":0.475,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+38\/-38","imputed":false},{"position":26,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1334.0,"score":0.255,"rank":33.11,"hrank":20.53,"imp":16.148,"train_time":1.18,"predict_time":0.072,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-50","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1333.0,"score":0.174,"rank":33.17,"hrank":24.16,"imp":16.129,"train_time":693.49,"predict_time":0.308,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+30\/-31","imputed":false},{"position":28,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1331.0,"score":0.213,"rank":33.4,"hrank":21.05,"imp":16.349,"train_time":846.89,"predict_time":2.549,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+45\/-40","imputed":false},{"position":29,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1322.0,"score":0.298,"rank":34.13,"hrank":13.83,"imp":16.718,"train_time":3031.5,"predict_time":21.442,"imputed_pct":35.29,"hardware":"GPU","elo_ci":"+63\/-64","imputed":true},{"position":30,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1314.0,"score":0.196,"rank":34.89,"hrank":21.31,"imp":17.276,"train_time":427.39,"predict_time":0.404,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-57","imputed":false},{"position":31,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1312.0,"score":0.286,"rank":35.12,"hrank":15.07,"imp":17.236,"train_time":87.65,"predict_time":2.505,"imputed_pct":35.29,"hardware":"GPU","elo_ci":"+62\/-64","imputed":true},{"position":32,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1309.0,"score":0.251,"rank":35.31,"hrank":15.68,"imp":17.291,"train_time":47.65,"predict_time":43.738,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-51","imputed":false},{"position":33,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1307.0,"score":0.232,"rank":35.51,"hrank":18.3,"imp":17.137,"train_time":6.63,"predict_time":1.48,"imputed_pct":29.41,"hardware":"GPU","elo_ci":"+51\/-57","imputed":true},{"position":34,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1292.0,"score":0.165,"rank":36.93,"hrank":23.22,"imp":17.614,"train_time":427.39,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-58","imputed":false},{"position":35,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1286.0,"score":0.161,"rank":37.45,"hrank":24.73,"imp":17.514,"train_time":846.89,"predict_time":0.13,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+45\/-38","imputed":false},{"position":36,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1282.0,"score":0.169,"rank":37.82,"hrank":24.04,"imp":17.064,"train_time":7.49,"predict_time":0.244,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+45\/-41","imputed":false},{"position":37,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1281.0,"score":0.123,"rank":37.91,"hrank":25.8,"imp":17.211,"train_time":12685.08,"predict_time":62.13,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+35\/-32","imputed":false},{"position":38,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1270.0,"score":0.14,"rank":38.93,"hrank":28.16,"imp":17.224,"train_time":2875.52,"predict_time":1.952,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-44","imputed":false},{"position":39,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1270.0,"score":0.191,"rank":38.94,"hrank":22.57,"imp":18.563,"train_time":205.88,"predict_time":1.343,"imputed_pct":25.49,"hardware":"GPU","elo_ci":"+58\/-50","imputed":true},{"position":40,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1268.0,"score":0.222,"rank":39.06,"hrank":17.16,"imp":18.206,"train_time":3031.5,"predict_time":0.462,"imputed_pct":35.29,"hardware":"GPU","elo_ci":"+58\/-56","imputed":true},{"position":41,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1268.0,"score":0.189,"rank":39.11,"hrank":15.49,"imp":18.34,"train_time":14.11,"predict_time":2.075,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-50","imputed":false},{"position":42,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1253.0,"score":0.125,"rank":40.46,"hrank":28.89,"imp":18.734,"train_time":1711.25,"predict_time":0.141,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-39","imputed":false},{"position":43,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1250.0,"score":0.135,"rank":40.7,"hrank":25.09,"imp":18.719,"train_time":1.99,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-54","imputed":false},{"position":44,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1241.0,"score":0.201,"rank":41.53,"hrank":20.18,"imp":18.936,"train_time":3.36,"predict_time":0.313,"imputed_pct":35.29,"hardware":"GPU","elo_ci":"+63\/-61","imputed":true},{"position":45,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1239.0,"score":0.096,"rank":41.71,"hrank":26.27,"imp":19.255,"train_time":14.87,"predict_time":0.307,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+38\/-36","imputed":false},{"position":46,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1220.0,"score":0.089,"rank":43.51,"hrank":31.6,"imp":19.375,"train_time":1711.25,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-39","imputed":false},{"position":47,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1218.0,"score":0.074,"rank":43.61,"hrank":32.75,"imp":18.633,"train_time":10.06,"predict_time":1.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+35\/-38","imputed":false},{"position":48,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1208.0,"score":0.066,"rank":44.61,"hrank":33.47,"imp":18.851,"train_time":1.94,"predict_time":0.123,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-37","imputed":false},{"position":49,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1204.0,"score":0.083,"rank":44.92,"hrank":31.04,"imp":19.836,"train_time":263.04,"predict_time":0.766,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-43","imputed":false},{"position":50,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1201.0,"score":0.078,"rank":45.19,"hrank":34.53,"imp":18.939,"train_time":2875.52,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-42","imputed":false},{"position":51,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1193.0,"score":0.106,"rank":45.93,"hrank":27.87,"imp":19.739,"train_time":593.24,"predict_time":4.467,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-54","imputed":false},{"position":52,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1190.0,"score":0.077,"rank":46.15,"hrank":31.54,"imp":20.315,"train_time":6.67,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-48","imputed":false},{"position":53,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1180.0,"score":0.05,"rank":47.05,"hrank":39.95,"imp":19.405,"train_time":1.96,"predict_time":0.142,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+31\/-29","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1179.0,"score":0.073,"rank":47.2,"hrank":29.3,"imp":20.622,"train_time":263.04,"predict_time":0.071,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-46","imputed":false},{"position":55,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1172.0,"score":0.063,"rank":47.79,"hrank":31.07,"imp":20.885,"train_time":373.24,"predict_time":0.771,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-43","imputed":false},{"position":56,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1156.0,"score":0.18,"rank":49.2,"hrank":11.82,"imp":23.123,"train_time":0.53,"predict_time":0.08,"imputed_pct":74.51,"hardware":"GPU","elo_ci":"+72\/-82","imputed":true},{"position":57,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1146.0,"score":0.16,"rank":50.07,"hrank":21.43,"imp":23.402,"train_time":0.53,"predict_time":0.08,"imputed_pct":74.51,"hardware":"GPU","elo_ci":"+72\/-77","imputed":true},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1137.0,"score":0.048,"rank":50.83,"hrank":34.27,"imp":21.158,"train_time":593.24,"predict_time":0.306,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-52","imputed":false},{"position":59,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1136.0,"score":0.043,"rank":50.95,"hrank":30.64,"imp":21.591,"train_time":373.24,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-39","imputed":false},{"position":60,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1089.0,"score":0.043,"rank":54.83,"hrank":37.25,"imp":23.08,"train_time":296.64,"predict_time":68.173,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+46\/-48","imputed":false},{"position":61,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1088.0,"score":0.123,"rank":54.95,"hrank":16.93,"imp":25.638,"train_time":28729.74,"predict_time":4.268,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-78","imputed":false},{"position":62,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1087.0,"score":0.121,"rank":55.04,"hrank":16.21,"imp":25.653,"train_time":28729.74,"predict_time":18.811,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-78","imputed":false},{"position":63,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1086.0,"score":0.054,"rank":55.11,"hrank":40.88,"imp":23.703,"train_time":13.12,"predict_time":2.52,"imputed_pct":25.49,"hardware":"GPU","elo_ci":"+49\/-47","imputed":true},{"position":64,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1085.0,"score":0.03,"rank":55.18,"hrank":42.64,"imp":26.504,"train_time":185.31,"predict_time":0.6,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-45","imputed":false},{"position":65,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1074.0,"score":0.01,"rank":56.06,"hrank":44.27,"imp":22.874,"train_time":9.99,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-45","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1050.0,"score":0.016,"rank":57.93,"hrank":43.69,"imp":27.859,"train_time":185.31,"predict_time":0.258,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-42","imputed":false},{"position":67,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1038.0,"score":0.029,"rank":58.83,"hrank":49.85,"imp":26.502,"train_time":3.23,"predict_time":0.919,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-66","imputed":false},{"position":68,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1010.0,"score":0.036,"rank":60.82,"hrank":43.94,"imp":27.674,"train_time":0.79,"predict_time":0.123,"imputed_pct":25.49,"hardware":"GPU","elo_ci":"+58\/-66","imputed":true},{"position":69,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1007.0,"score":0.01,"rank":61.05,"hrank":53.39,"imp":26.231,"train_time":1.87,"predict_time":0.283,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-56","imputed":false},{"position":70,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1005.0,"score":0.008,"rank":61.17,"hrank":44.93,"imp":25.739,"train_time":2.86,"predict_time":0.373,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-63","imputed":false},{"position":71,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.002,"rank":61.5,"hrank":56.68,"imp":26.318,"train_time":0.43,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-40","imputed":false},{"position":72,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":992.0,"score":0.018,"rank":62.01,"hrank":42.86,"imp":27.874,"train_time":129.08,"predict_time":1.801,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-78","imputed":false},{"position":73,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":989.0,"score":0.075,"rank":62.24,"hrank":25.93,"imp":30.507,"train_time":384.75,"predict_time":5.348,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-95","imputed":false},{"position":74,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":960.0,"score":0.033,"rank":64.12,"hrank":45.89,"imp":33.71,"train_time":237.63,"predict_time":0.424,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-100","imputed":false},{"position":75,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":936.0,"score":0.023,"rank":65.53,"hrank":48.2,"imp":34.253,"train_time":237.63,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-106","imputed":false},{"position":76,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":931.0,"score":0.002,"rank":65.83,"hrank":50.56,"imp":31.803,"train_time":27.32,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-53","imputed":false},{"position":77,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":885.0,"score":0.01,"rank":68.3,"hrank":52.67,"imp":32.832,"train_time":129.08,"predict_time":0.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-92","imputed":false},{"position":78,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":862.0,"score":0.013,"rank":69.42,"hrank":53.13,"imp":36.861,"train_time":1.19,"predict_time":0.12,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-113","imputed":false},{"position":79,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":639.0,"score":0.0,"rank":76.61,"hrank":68.38,"imp":46.029,"train_time":0.19,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-88","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_all/pareto_front_explorer.html
index 5b79b1b78951b7e88e4a2610b3d35732b2a20375..45ef2fb8f5ed3e18fd758aa997e70eec789c9c9d 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.3429487991,"imp":18.5626343699,"elo":1269.5,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"imp":15.588528412,"elo":1356.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0380928654,"imp":14.9503537582,"elo":1384.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3442282032,"imp":14.72117124,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0325848325,"imp":18.7191566159,"elo":1250.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0336512776,"imp":17.6138003773,"elo":1291.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4040760969,"imp":17.2759095354,"elo":1314.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"imp":20.3154944935,"elo":1190.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0150308987,"imp":19.3748397702,"elo":1219.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1412900124,"imp":18.734016812,"elo":1253.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2834081321,"imp":26.2310351768,"elo":1006.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0712129939,"imp":20.6220594726,"elo":1178.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7662402325,"imp":19.8361396109,"elo":1204.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3731781389,"imp":25.7388501006,"elo":1004.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3063915412,"imp":21.1575769102,"elo":1137.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4666950703,"imp":19.7386961101,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1416787942,"imp":19.4051210577,"elo":1180.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3337603531,"imp":15.7521053174,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"imp":15.0989736718,"elo":1408.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.2423742585,"imp":15.5927312714,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.5047910539,"imp":17.2361832431,"elo":1311.5,"imputed":true,"imputed_pct":35.2941176471,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3073466784,"imp":19.2552732764,"elo":1239.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4747242315,"imp":16.1066519394,"elo":1337.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.1379168675,"imp":15.6613301431,"elo":1367.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.0799350228,"imp":23.4021360099,"elo":1146.0,"imputed":true,"imputed_pct":74.5098039216,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0799350228,"imp":23.1225771201,"elo":1156.0,"imputed":true,"imputed_pct":74.5098039216,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"imp":23.703379007,"elo":1085.8,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"imp":31.8025000078,"elo":931.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2581384468,"imp":27.8587730829,"elo":1049.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5997539679,"imp":26.5039613042,"elo":1084.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0535405758,"imp":26.3183513081,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852279186,"imp":21.5909038168,"elo":1135.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7709478321,"imp":20.884617726,"elo":1172.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.6887954948,"imp":18.6334957836,"elo":1218.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.663132055,"imp":14.2683471697,"elo":1402.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":11.9749853765,"imp":12.9492805492,"elo":1480.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"imp":11.7294546131,"elo":1504.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2178786994,"imp":11.1860512843,"elo":1533.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9076602204,"imp":10.4970952198,"elo":1573.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0745579697,"imp":18.340351508,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"imp":17.2910686343,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9597364603,"imp":15.0029132595,"elo":1384.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":286.5373788357,"imp":13.8553240305,"elo":1437.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"imp":14.0664166138,"elo":1437.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.9854290718,"imp":4.1459586596,"elo":1765.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1226521934,"imp":27.6738478693,"elo":1009.8,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.4796604156,"imp":17.1374245291,"elo":1307.2,"imputed":true,"imputed_pct":29.4117647059,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1505492146,"imp":10.2451337085,"elo":1570.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2435252328,"imp":17.0642482835,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2512939462,"imp":15.0177264339,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.2468658586,"imp":14.1476228811,"elo":1424.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5548034626,"imp":10.8199729106,"elo":1593.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"imp":9.3982489748,"elo":1640.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3129231924,"imp":18.9359143543,"elo":1241.3,"imputed":true,"imputed_pct":35.2941176471,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4619844041,"imp":18.2063464844,"elo":1268.2,"imputed":true,"imputed_pct":35.2941176471,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":21.4417453448,"imp":16.7184430035,"elo":1322.5,"imputed":true,"imputed_pct":35.2941176471,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3480868421,"imp":30.5067525537,"elo":989.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2680534632,"imp":25.6375872706,"elo":1087.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"imp":25.6533551916,"elo":1086.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719683769,"imp":16.1478294627,"elo":1333.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1258046788,"imp":22.8735512448,"elo":1073.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1311128328,"imp":18.9388451265,"elo":1201.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.95162543,"imp":17.2240102343,"elo":1269.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1226232881,"imp":18.851428559,"elo":1207.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3080760155,"imp":16.1287168289,"elo":1333.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.689445163,"imp":15.8177548362,"elo":1355.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":68.173296937,"imp":23.0801987545,"elo":1089.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":62.1299021172,"imp":17.211355754,"elo":1280.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":464.3696034625,"imp":15.3661612139,"elo":1382.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"imp":26.5020498731,"elo":1037.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1302743502,"imp":17.5142371104,"elo":1285.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.5492595349,"imp":16.3492250292,"elo":1330.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.3429487991,"x_train":205.88373935,"imp":18.5626343699,"elo":1269.5,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"x_train":5.876676442,"imp":15.588528412,"elo":1356.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0380928654,"x_train":1346.2106911803,"imp":14.9503537582,"elo":1384.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3442282032,"x_train":1346.2106911803,"imp":14.72117124,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0325848325,"x_train":1.9886321895,"imp":18.7191566159,"elo":1250.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0336512776,"x_train":427.391750384,"imp":17.6138003773,"elo":1291.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4040760969,"x_train":427.391750384,"imp":17.2759095354,"elo":1314.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"x_train":6.6731468509,"imp":20.3154944935,"elo":1190.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0150308987,"x_train":1711.2458519871,"imp":19.3748397702,"elo":1219.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1412900124,"x_train":1711.2458519871,"imp":18.734016812,"elo":1253.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2834081321,"x_train":1.8719729433,"imp":26.2310351768,"elo":1006.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0712129939,"x_train":263.0379305179,"imp":20.6220594726,"elo":1178.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7662402325,"x_train":263.0379305179,"imp":19.8361396109,"elo":1204.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3731781389,"x_train":2.8561491959,"imp":25.7388501006,"elo":1004.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3063915412,"x_train":593.2377884029,"imp":21.1575769102,"elo":1137.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4666950703,"x_train":593.2377884029,"imp":19.7386961101,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1416787942,"x_train":1.960032688,"imp":19.4051210577,"elo":1180.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3337603531,"x_train":417.0474735649,"imp":15.7521053174,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"x_train":417.0474735649,"imp":15.0989736718,"elo":1408.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.2423742585,"x_train":26.4597512682,"imp":15.5927312714,"elo":1344.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.5047910539,"x_train":87.6450813939,"imp":17.2361832431,"elo":1311.5,"imputed":true,"imputed_pct":35.2941176471,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3073466784,"x_train":14.8696863446,"imp":19.2552732764,"elo":1239.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4747242315,"x_train":4621.6656335635,"imp":16.1066519394,"elo":1337.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.1379168675,"x_train":4621.6656335635,"imp":15.6613301431,"elo":1367.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.0799350228,"x_train":0.5265501441,"imp":23.4021360099,"elo":1146.0,"imputed":true,"imputed_pct":74.5098039216,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0799350228,"x_train":0.5265501441,"imp":23.1225771201,"elo":1156.0,"imputed":true,"imputed_pct":74.5098039216,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"x_train":13.1170218359,"imp":23.703379007,"elo":1085.8,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"x_train":27.322832323,"imp":31.8025000078,"elo":931.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2581384468,"x_train":185.3067056521,"imp":27.8587730829,"elo":1049.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5997539679,"x_train":185.3067056521,"imp":26.5039613042,"elo":1084.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0535405758,"x_train":0.4343250827,"imp":26.3183513081,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852279186,"x_train":373.2371055096,"imp":21.5909038168,"elo":1135.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7709478321,"x_train":373.2371055096,"imp":20.884617726,"elo":1172.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.6887954948,"x_train":10.055649238,"imp":18.6334957836,"elo":1218.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.663132055,"x_train":2950.7244476948,"imp":14.2683471697,"elo":1402.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":11.9749853765,"x_train":2950.7244476948,"imp":12.9492805492,"elo":1480.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"x_train":5.8097985299,"imp":11.7294546131,"elo":1504.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2178786994,"x_train":2040.2228420973,"imp":11.1860512843,"elo":1533.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9076602204,"x_train":2040.2228420973,"imp":10.4970952198,"elo":1573.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0745579697,"x_train":14.1132668455,"imp":18.340351508,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"x_train":47.6466878891,"imp":17.2910686343,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9597364603,"x_train":4910.3773062373,"imp":15.0029132595,"elo":1384.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":286.5373788357,"x_train":4910.3773062373,"imp":13.8553240305,"elo":1437.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"x_train":2.0676840515,"imp":14.0664166138,"elo":1437.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.9854290718,"x_train":38.8059088678,"imp":4.1459586596,"elo":1765.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1226521934,"x_train":0.7886124945,"imp":27.6738478693,"elo":1009.8,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.4796604156,"x_train":6.6270115137,"imp":17.1374245291,"elo":1307.2,"imputed":true,"imputed_pct":29.4117647059,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1505492146,"x_train":2.0524528172,"imp":10.2451337085,"elo":1570.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2435252328,"x_train":7.4864639685,"imp":17.0642482835,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2512939462,"x_train":2450.1270774379,"imp":15.0177264339,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.2468658586,"x_train":2450.1270774379,"imp":14.1476228811,"elo":1424.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5548034626,"x_train":5.4825798778,"imp":10.8199729106,"elo":1593.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"x_train":3.65846503,"imp":9.3982489748,"elo":1640.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3129231924,"x_train":3.3572004182,"imp":18.9359143543,"elo":1241.3,"imputed":true,"imputed_pct":35.2941176471,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4619844041,"x_train":3031.4979650506,"imp":18.2063464844,"elo":1268.2,"imputed":true,"imputed_pct":35.2941176471,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":21.4417453448,"x_train":3031.4979650506,"imp":16.7184430035,"elo":1322.5,"imputed":true,"imputed_pct":35.2941176471,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3480868421,"x_train":384.7456301814,"imp":30.5067525537,"elo":989.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2680534632,"x_train":28729.7373655693,"imp":25.6375872706,"elo":1087.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"x_train":28729.7373655693,"imp":25.6533551916,"elo":1086.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719683769,"x_train":1.1753394519,"imp":16.1478294627,"elo":1333.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1258046788,"x_train":9.9925632138,"imp":22.8735512448,"elo":1073.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1311128328,"x_train":2875.5203407747,"imp":18.9388451265,"elo":1201.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.95162543,"x_train":2875.5203407747,"imp":17.2240102343,"elo":1269.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1226232881,"x_train":1.9409389396,"imp":18.851428559,"elo":1207.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3080760155,"x_train":693.4907982506,"imp":16.1287168289,"elo":1333.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.689445163,"x_train":693.4907982506,"imp":15.8177548362,"elo":1355.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":68.173296937,"x_train":296.6392917616,"imp":23.0801987545,"elo":1089.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":62.1299021172,"x_train":12685.0815501834,"imp":17.211355754,"elo":1280.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":464.3696034625,"x_train":12685.0815501834,"imp":15.3661612139,"elo":1382.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"x_train":3.2328671519,"imp":26.5020498731,"elo":1037.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1302743502,"x_train":846.8860702337,"imp":17.5142371104,"elo":1285.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.5492595349,"x_train":846.8860702337,"imp":16.3492250292,"elo":1330.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_all/tuning_trajectories_explorer.html
index 872f5e28d390ef8486c314bbdd69c24c87ab73d0..2726e4ad1de2938ed51e38b259b4f45df8edce8a 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":205.88373935,"imp":18.5813547735,"elo":1284.5,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":5.876676442,"imp":15.6065851096,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":11.6678251256,"imp":15.1793361507,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":31.9106517624,"imp":14.9100953425,"elo":1409.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":62.5721251757,"imp":14.8094734563,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":155.9049355669,"imp":14.7025535215,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":317.345916891,"imp":14.6930885341,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":662.4830210734,"imp":14.7098048719,"elo":1418.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":998.8498259224,"imp":14.7293145036,"elo":1416.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1346.2106911803,"imp":14.7362587188,"elo":1415.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.9886321895,"imp":18.7369326826,"elo":1261.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.4383850203,"imp":18.3959666487,"elo":1280.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.0701848574,"imp":18.1764299402,"elo":1292.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":20.4734178164,"imp":17.9857398173,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":52.7173757556,"imp":17.7136346524,"elo":1314.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":106.3941874905,"imp":17.562288704,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":212.8207478205,"imp":17.3993644083,"elo":1328.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":318.9957179838,"imp":17.3300682783,"elo":1331.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":427.391750384,"imp":17.2927290403,"elo":1333.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":6.6731468509,"imp":20.3329124963,"elo":1196.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":16.5765343457,"imp":19.9360827949,"elo":1210.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":47.7385589033,"imp":19.5112193553,"elo":1229.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":89.4110724699,"imp":19.3306842701,"elo":1238.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":218.448124708,"imp":19.1021913433,"elo":1248.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":440.2545735965,"imp":18.9607952035,"elo":1255.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":856.7647171053,"imp":18.8389024106,"elo":1262.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1266.0910927829,"imp":18.7955242029,"elo":1265.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1711.2458519871,"imp":18.7531569535,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.8719729433,"imp":26.2472013772,"elo":1007.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1259596548,"imp":24.4021328027,"elo":1056.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.7721912257,"imp":21.7371689645,"elo":1126.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":13.288092342,"imp":20.8155931069,"elo":1165.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":32.8419339595,"imp":20.1625420297,"elo":1194.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":65.4938634865,"imp":20.0002200669,"elo":1203.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":130.8369715186,"imp":19.8897585305,"elo":1209.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":196.2244804971,"imp":19.8467667636,"elo":1213.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":263.0379305179,"imp":19.8542401115,"elo":1214.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.8561491959,"imp":25.7552940791,"elo":1009.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.5343795004,"imp":24.5566884748,"elo":1045.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.5510344275,"imp":22.7093264429,"elo":1110.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":29.723343154,"imp":21.8028219279,"elo":1141.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":72.5444980812,"imp":20.9028755361,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":145.754371191,"imp":20.4489727956,"elo":1185.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":293.9475008279,"imp":20.1067260855,"elo":1195.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":441.8525399102,"imp":19.8847902506,"elo":1202.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":593.2377884029,"imp":19.7519391596,"elo":1207.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":1.960032688,"imp":19.4223268604,"elo":1183.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":4.1465368891,"imp":17.9127028318,"elo":1257.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":10.3907433568,"imp":16.745888127,"elo":1331.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":20.2701675373,"imp":16.1451075213,"elo":1369.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":51.487448334,"imp":15.5895757233,"elo":1401.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":103.6855742013,"imp":15.377461123,"elo":1415.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":207.5435114784,"imp":15.2170155689,"elo":1425.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":310.2926532568,"imp":15.1552693571,"elo":1428.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":417.0474735649,"imp":15.1181427797,"elo":1431.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":26.4597512682,"imp":15.5471950943,"elo":1365.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":87.6450813939,"imp":17.2541916329,"elo":1329.7,"imputed":true,"imputed_pct":35.2941176471,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.8696863446,"imp":19.2726417887,"elo":1250.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":40.5171687704,"imp":17.9821164502,"elo":1284.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":111.4498310885,"imp":16.0650273001,"elo":1364.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":224.929440425,"imp":15.5742135066,"elo":1387.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":554.8880523029,"imp":15.0656604708,"elo":1411.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1145.4516650488,"imp":15.1130109133,"elo":1408.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2295.4928354707,"imp":15.3720755011,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3433.7924746694,"imp":15.5711586763,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4621.6656335635,"imp":15.6848956215,"elo":1386.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":0.5265501441,"imp":23.4173207309,"elo":1152.3,"imputed":true,"imputed_pct":74.5098039216,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":0.5265501441,"imp":23.1378222847,"elo":1161.7,"imputed":true,"imputed_pct":74.5098039216,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_train":13.1170218359,"imp":23.7279885673,"elo":1094.0,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":27.322832323,"imp":32.3034146549,"elo":931.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":60.3304655332,"imp":30.0574375635,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":157.3958954151,"imp":26.6880817331,"elo":1085.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":185.3067056521,"imp":26.5163927326,"elo":1092.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4343250827,"imp":26.3443545072,"elo":997.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.3006968127,"imp":23.6832328225,"elo":1070.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":7.8694051654,"imp":22.3470002365,"elo":1109.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":17.2473724895,"imp":21.6511464138,"elo":1135.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":45.0373615697,"imp":21.1616221694,"elo":1163.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":91.3917233329,"imp":21.0298281684,"elo":1171.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":184.6996741609,"imp":20.9428721411,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":277.8099565315,"imp":20.9047047797,"elo":1178.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":373.2371055096,"imp":20.8997780504,"elo":1180.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.055649238,"imp":18.6508348206,"elo":1225.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.6705594408,"imp":16.2987011126,"elo":1330.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":69.643527334,"imp":14.6237723993,"elo":1417.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":141.1529977589,"imp":14.0061431083,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":364.6011469598,"imp":13.5307330372,"elo":1477.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":736.9754853516,"imp":13.3108256189,"elo":1490.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1469.2653973047,"imp":13.1285460202,"elo":1498.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2197.5320041687,"imp":13.0360888153,"elo":1504.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2950.7244476948,"imp":12.9686488422,"elo":1507.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.8097985299,"imp":11.7491247048,"elo":1521.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.6950892358,"imp":11.6797925771,"elo":1528.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":45.8180610371,"imp":11.3585061763,"elo":1545.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":96.056645332,"imp":11.1774072762,"elo":1562.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":245.1410909168,"imp":10.9349022701,"elo":1581.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":498.5667116885,"imp":10.7533566047,"elo":1591.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1015.1487598759,"imp":10.6345822919,"elo":1594.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1526.7130408425,"imp":10.5741990357,"elo":1595.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2040.2228420973,"imp":10.5220655594,"elo":1595.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":14.1132668455,"imp":18.3572436507,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":47.6466878891,"imp":17.3073951723,"elo":1324.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.407219275,"imp":16.4905430418,"elo":1350.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":132.9785409439,"imp":15.6764823719,"elo":1384.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":243.7764468154,"imp":15.3221828246,"elo":1399.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":599.1452639259,"imp":14.6518498381,"elo":1425.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1224.6320092925,"imp":14.3105565246,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2440.7040454743,"imp":14.0895799523,"elo":1448.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3652.693914187,"imp":13.9476888289,"elo":1453.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4910.3773062373,"imp":13.8761184088,"elo":1455.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0676840515,"imp":14.0835928363,"elo":1452.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.8059088678,"imp":4.1621324185,"elo":1785.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.7886124945,"imp":27.6878844468,"elo":1018.6,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":6.6270115137,"imp":17.1556578378,"elo":1324.8,"imputed":true,"imputed_pct":29.4117647059,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":2.0524528172,"imp":10.2650620686,"elo":1591.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":7.4864639685,"imp":17.082430493,"elo":1294.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":16.6863025317,"imp":16.5568126345,"elo":1319.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":56.5960636953,"imp":15.9189312228,"elo":1352.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":118.7995551511,"imp":15.3477939918,"elo":1384.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":301.7168598673,"imp":14.8558678329,"elo":1409.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":606.795758935,"imp":14.5680937883,"elo":1425.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1211.551032833,"imp":14.3647682509,"elo":1436.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1826.0696865509,"imp":14.2203370763,"elo":1442.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2450.1270774379,"imp":14.1652694081,"elo":1447.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.4825798778,"imp":10.8394230989,"elo":1620.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.65846503,"imp":9.4164243724,"elo":1662.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3.3572004182,"imp":18.9535426804,"elo":1254.5,"imputed":true,"imputed_pct":35.2941176471,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":23.1786189605,"imp":18.3742001687,"elo":1278.7,"imputed":true,"imputed_pct":35.2941176471,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":63.7281578816,"imp":17.787049067,"elo":1306.3,"imputed":true,"imputed_pct":35.2941176471,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":143.4678270825,"imp":17.4153044832,"elo":1319.4,"imputed":true,"imputed_pct":35.2941176471,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":360.1628265674,"imp":17.0817066168,"elo":1332.6,"imputed":true,"imputed_pct":35.2941176471,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":731.6454759573,"imp":16.9022597121,"elo":1339.6,"imputed":true,"imputed_pct":35.2941176471,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1485.885866909,"imp":16.7937604058,"elo":1343.2,"imputed":true,"imputed_pct":35.2941176471,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2238.2118339463,"imp":16.7495622267,"elo":1344.3,"imputed":true,"imputed_pct":35.2941176471,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3031.4979650506,"imp":16.7215605152,"elo":1343.8,"imputed":true,"imputed_pct":35.2941176471,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":384.7456301814,"imp":30.5222923408,"elo":995.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":579.5314694238,"imp":30.3953041638,"elo":1003.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":986.9188087142,"imp":30.129435193,"elo":1011.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1626.2057458405,"imp":29.4699808864,"elo":1028.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3523.6508469471,"imp":28.2744674139,"elo":1050.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":7090.4373429841,"imp":27.0759431861,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":14236.6030369928,"imp":26.4053258616,"elo":1085.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":21385.328840253,"imp":25.9929582132,"elo":1095.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":28729.7373655693,"imp":25.6703338597,"elo":1105.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1753394519,"imp":16.1657705633,"elo":1351.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.9925632138,"imp":22.8850366269,"elo":1081.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":23.3976730308,"imp":21.1661879002,"elo":1132.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":67.83522273,"imp":19.6430780766,"elo":1180.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":150.5961763502,"imp":18.9177332705,"elo":1212.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":358.8914252792,"imp":18.2366603106,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":711.8242682675,"imp":17.8219921337,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1402.3570975877,"imp":17.5279048071,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2125.0611186926,"imp":17.3712258363,"elo":1282.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2875.5203407747,"imp":17.237392444,"elo":1288.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.9409389396,"imp":18.8687466883,"elo":1211.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.7736976246,"imp":17.223916001,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":17.1658020762,"imp":16.6921946861,"elo":1323.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":34.0813643193,"imp":16.3749596179,"elo":1342.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.4155038646,"imp":16.1290997784,"elo":1356.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":172.6999358448,"imp":16.0259383344,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":344.2808053206,"imp":15.9436458826,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":515.1408947248,"imp":15.8844126064,"elo":1371.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":693.4907982506,"imp":15.836256277,"elo":1375.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":296.6392917616,"imp":23.0966015493,"elo":1098.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":761.0284006744,"imp":20.4062004234,"elo":1164.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2323.5909795411,"imp":18.8693887945,"elo":1266.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4948.4562156001,"imp":17.5999354536,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12120.2560802151,"imp":15.4129316692,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12685.0815501834,"imp":15.3839367442,"elo":1402.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.2328671519,"imp":26.5176586874,"elo":1045.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":7.7757953049,"imp":23.2835232661,"elo":1109.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":20.6204304219,"imp":19.3320675044,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":41.9962580807,"imp":17.5849304711,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":105.704979404,"imp":16.7547041225,"elo":1335.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":210.3194448262,"imp":16.4855249302,"elo":1346.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":422.0870812379,"imp":16.3830505277,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":631.9086612835,"imp":16.3618097985,"elo":1349.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":846.8860702337,"imp":16.358428191,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.3429487991,"x_train":205.88373935,"imp":18.5813547735,"elo":1284.5,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"x_train":5.876676442,"imp":15.6065851096,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0633653548,"x_train":11.6678251256,"imp":15.1793361507,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1177165074,"x_train":31.9106517624,"imp":14.9100953425,"elo":1409.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1714353465,"x_train":62.5721251757,"imp":14.8094734563,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2216220526,"x_train":155.9049355669,"imp":14.7025535215,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2666794695,"x_train":317.345916891,"imp":14.6930885341,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3105276001,"x_train":662.4830210734,"imp":14.7098048719,"elo":1418.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3407350551,"x_train":998.8498259224,"imp":14.7293145036,"elo":1416.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3448570028,"x_train":1346.2106911803,"imp":14.7362587188,"elo":1415.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0325848325,"x_train":1.9886321895,"imp":18.7369326826,"elo":1261.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0609444401,"x_train":4.4383850203,"imp":18.3959666487,"elo":1280.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1129020181,"x_train":10.0701848574,"imp":18.1764299402,"elo":1292.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1771034691,"x_train":20.4734178164,"imp":17.9857398173,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2454903367,"x_train":52.7173757556,"imp":17.7136346524,"elo":1314.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2926890222,"x_train":106.3941874905,"imp":17.562288704,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.312025413,"x_train":212.8207478205,"imp":17.3993644083,"elo":1328.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3237207716,"x_train":318.9957179838,"imp":17.3300682783,"elo":1331.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4040760969,"x_train":427.391750384,"imp":17.2927290403,"elo":1333.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"x_train":6.6731468509,"imp":20.3329124963,"elo":1196.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0304247242,"x_train":16.5765343457,"imp":19.9360827949,"elo":1210.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0528424778,"x_train":47.7385589033,"imp":19.5112193553,"elo":1229.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0702878348,"x_train":89.4110724699,"imp":19.3306842701,"elo":1238.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0974716988,"x_train":218.448124708,"imp":19.1021913433,"elo":1248.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1141121872,"x_train":440.2545735965,"imp":18.9607952035,"elo":1255.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1277215511,"x_train":856.7647171053,"imp":18.8389024106,"elo":1262.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1376216405,"x_train":1266.0910927829,"imp":18.7955242029,"elo":1265.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1384442291,"x_train":1711.2458519871,"imp":18.7531569535,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2834081321,"x_train":1.8719729433,"imp":26.2472013772,"elo":1007.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2992675583,"x_train":3.1259596548,"imp":24.4021328027,"elo":1056.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3233863551,"x_train":6.7721912257,"imp":21.7371689645,"elo":1126.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3916757047,"x_train":13.288092342,"imp":20.8155931069,"elo":1165.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4883309222,"x_train":32.8419339595,"imp":20.1625420297,"elo":1194.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5677931699,"x_train":65.4938634865,"imp":20.0002200669,"elo":1203.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6638268741,"x_train":130.8369715186,"imp":19.8897585305,"elo":1209.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7241674769,"x_train":196.2244804971,"imp":19.8467667636,"elo":1213.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7662402325,"x_train":263.0379305179,"imp":19.8542401115,"elo":1214.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3731781389,"x_train":2.8561491959,"imp":25.7552940791,"elo":1009.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6641420919,"x_train":5.5343795004,"imp":24.5566884748,"elo":1045.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1863927783,"x_train":14.5510344275,"imp":22.7093264429,"elo":1110.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.732344048,"x_train":29.723343154,"imp":21.8028219279,"elo":1141.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.3920814156,"x_train":72.5444980812,"imp":20.9028755361,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1392189601,"x_train":145.754371191,"imp":20.4489727956,"elo":1185.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.8048101374,"x_train":293.9475008279,"imp":20.1067260855,"elo":1195.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.2073695134,"x_train":441.8525399102,"imp":19.8847902506,"elo":1202.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4558194315,"x_train":593.2377884029,"imp":19.7519391596,"elo":1207.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1416787942,"x_train":1.960032688,"imp":19.4223268604,"elo":1183.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6924533472,"x_train":4.1465368891,"imp":17.9127028318,"elo":1257.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9306682769,"x_train":10.3907433568,"imp":16.745888127,"elo":1331.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1975461537,"x_train":20.2701675373,"imp":16.1451075213,"elo":1369.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5075319607,"x_train":51.487448334,"imp":15.5895757233,"elo":1401.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6740993927,"x_train":103.6855742013,"imp":15.377461123,"elo":1415.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.0960239899,"x_train":207.5435114784,"imp":15.2170155689,"elo":1425.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.295770736,"x_train":310.2926532568,"imp":15.1552693571,"elo":1428.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"x_train":417.0474735649,"imp":15.1181427797,"elo":1431.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":6.2423742585,"x_train":26.4597512682,"imp":15.5471950943,"elo":1365.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.5047910539,"x_train":87.6450813939,"imp":17.2541916329,"elo":1329.7,"imputed":true,"imputed_pct":35.2941176471,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3073466784,"x_train":14.8696863446,"imp":19.2726417887,"elo":1250.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8515440488,"x_train":40.5171687704,"imp":17.9821164502,"elo":1284.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.746831802,"x_train":111.4498310885,"imp":16.0650273001,"elo":1364.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.5416509891,"x_train":224.929440425,"imp":15.5742135066,"elo":1387.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.0322964811,"x_train":554.8880523029,"imp":15.0656604708,"elo":1411.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.4485225391,"x_train":1145.4516650488,"imp":15.1130109133,"elo":1408.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.931406374,"x_train":2295.4928354707,"imp":15.3720755011,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.7285966831,"x_train":3433.7924746694,"imp":15.5711586763,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.1379168675,"x_train":4621.6656335635,"imp":15.6848956215,"elo":1386.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.0799350228,"x_train":0.5265501441,"imp":23.4173207309,"elo":1152.3,"imputed":true,"imputed_pct":74.5098039216,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0799350228,"x_train":0.5265501441,"imp":23.1378222847,"elo":1161.7,"imputed":true,"imputed_pct":74.5098039216,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"x_train":13.1170218359,"imp":23.7279885673,"elo":1094.0,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"x_train":27.322832323,"imp":32.3034146549,"elo":931.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2337632354,"x_train":60.3304655332,"imp":30.0574375635,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5732905746,"x_train":157.3958954151,"imp":26.6880817331,"elo":1085.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5997539679,"x_train":185.3067056521,"imp":26.5163927326,"elo":1092.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0535405758,"x_train":0.4343250827,"imp":26.3443545072,"elo":997.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1187728047,"x_train":2.3006968127,"imp":23.6832328225,"elo":1070.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.235791088,"x_train":7.8694051654,"imp":22.3470002365,"elo":1109.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3406296664,"x_train":17.2473724895,"imp":21.6511464138,"elo":1135.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5043278961,"x_train":45.0373615697,"imp":21.1616221694,"elo":1163.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5878338984,"x_train":91.3917233329,"imp":21.0298281684,"elo":1171.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6965050642,"x_train":184.6996741609,"imp":20.9428721411,"elo":1176.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7595438385,"x_train":277.8099565315,"imp":20.9047047797,"elo":1178.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7784693874,"x_train":373.2371055096,"imp":20.8997780504,"elo":1180.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6887954948,"x_train":10.055649238,"imp":18.6508348206,"elo":1225.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.459301494,"x_train":24.6705594408,"imp":16.2987011126,"elo":1330.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.716219813,"x_train":69.643527334,"imp":14.6237723993,"elo":1417.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.6764824634,"x_train":141.1529977589,"imp":14.0061431083,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.8477187358,"x_train":364.6011469598,"imp":13.5307330372,"elo":1477.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.7676765361,"x_train":736.9754853516,"imp":13.3108256189,"elo":1490.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.259146773,"x_train":1469.2653973047,"imp":13.1285460202,"elo":1498.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.0531789205,"x_train":2197.5320041687,"imp":13.0360888153,"elo":1504.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.3778248403,"x_train":2950.7244476948,"imp":12.9686488422,"elo":1507.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"x_train":5.8097985299,"imp":11.7491247048,"elo":1521.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4031929556,"x_train":16.6950892358,"imp":11.6797925771,"elo":1528.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8742272233,"x_train":45.8180610371,"imp":11.3585061763,"elo":1545.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.0955253444,"x_train":96.056645332,"imp":11.1774072762,"elo":1562.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7534074644,"x_train":245.1410909168,"imp":10.9349022701,"elo":1581.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8500761173,"x_train":498.5667116885,"imp":10.7533566047,"elo":1591.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.7920736053,"x_train":1015.1487598759,"imp":10.6345822919,"elo":1594.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9997599059,"x_train":1526.7130408425,"imp":10.5741990357,"elo":1595.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.0306819261,"x_train":2040.2228420973,"imp":10.5220655594,"elo":1595.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0745579697,"x_train":14.1132668455,"imp":18.3572436507,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"x_train":47.6466878891,"imp":17.3073951723,"elo":1324.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":57.2989818274,"x_train":70.407219275,"imp":16.4905430418,"elo":1350.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":81.7804361754,"x_train":132.9785409439,"imp":15.6764823719,"elo":1384.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":113.4133274069,"x_train":243.7764468154,"imp":15.3221828246,"elo":1399.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":170.4030706025,"x_train":599.1452639259,"imp":14.6518498381,"elo":1425.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":211.9243576754,"x_train":1224.6320092925,"imp":14.3105565246,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":252.4669965135,"x_train":2440.7040454743,"imp":14.0895799523,"elo":1448.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":264.8770310031,"x_train":3652.693914187,"imp":13.9476888289,"elo":1453.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":289.5741517329,"x_train":4910.3773062373,"imp":13.8761184088,"elo":1455.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"x_train":2.0676840515,"imp":14.0835928363,"elo":1452.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.9854290718,"x_train":38.8059088678,"imp":4.1621324185,"elo":1785.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1226521934,"x_train":0.7886124945,"imp":27.6878844468,"elo":1018.6,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.4796604156,"x_train":6.6270115137,"imp":17.1556578378,"elo":1324.8,"imputed":true,"imputed_pct":29.4117647059,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1505492146,"x_train":2.0524528172,"imp":10.2650620686,"elo":1591.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2435252328,"x_train":7.4864639685,"imp":17.082430493,"elo":1294.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.486302031,"x_train":16.6863025317,"imp":16.5568126345,"elo":1319.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.8905707593,"x_train":56.5960636953,"imp":15.9189312228,"elo":1352.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.969667724,"x_train":118.7995551511,"imp":15.3477939918,"elo":1384.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.2874446083,"x_train":301.7168598673,"imp":14.8558678329,"elo":1409.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.5958040708,"x_train":606.795758935,"imp":14.5680937883,"elo":1425.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.9425270856,"x_train":1211.551032833,"imp":14.3647682509,"elo":1436.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.1383865311,"x_train":1826.0696865509,"imp":14.2203370763,"elo":1442.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.2361524691,"x_train":2450.1270774379,"imp":14.1652694081,"elo":1447.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5548034626,"x_train":5.4825798778,"imp":10.8394230989,"elo":1620.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"x_train":3.65846503,"imp":9.4164243724,"elo":1662.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3129231924,"x_train":3.3572004182,"imp":18.9535426804,"elo":1254.5,"imputed":true,"imputed_pct":35.2941176471,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":3.0953116875,"x_train":23.1786189605,"imp":18.3742001687,"elo":1278.7,"imputed":true,"imputed_pct":35.2941176471,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":5.683535796,"x_train":63.7281578816,"imp":17.787049067,"elo":1306.3,"imputed":true,"imputed_pct":35.2941176471,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":8.8242133888,"x_train":143.4678270825,"imp":17.4153044832,"elo":1319.4,"imputed":true,"imputed_pct":35.2941176471,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":11.2919613957,"x_train":360.1628265674,"imp":17.0817066168,"elo":1332.6,"imputed":true,"imputed_pct":35.2941176471,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":14.177371364,"x_train":731.6454759573,"imp":16.9022597121,"elo":1339.6,"imputed":true,"imputed_pct":35.2941176471,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.3997432641,"x_train":1485.885866909,"imp":16.7937604058,"elo":1343.2,"imputed":true,"imputed_pct":35.2941176471,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.5700586544,"x_train":2238.2118339463,"imp":16.7495622267,"elo":1344.3,"imputed":true,"imputed_pct":35.2941176471,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":20.6465366411,"x_train":3031.4979650506,"imp":16.7215605152,"elo":1343.8,"imputed":true,"imputed_pct":35.2941176471,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3480868421,"x_train":384.7456301814,"imp":30.5222923408,"elo":995.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.8647718906,"x_train":579.5314694238,"imp":30.3953041638,"elo":1003.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8842610276,"x_train":986.9188087142,"imp":30.129435193,"elo":1011.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.4371098465,"x_train":1626.2057458405,"imp":29.4699808864,"elo":1028.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.4116860286,"x_train":3523.6508469471,"imp":28.2744674139,"elo":1050.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.581168526,"x_train":7090.4373429841,"imp":27.0759431861,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.0586264094,"x_train":14236.6030369928,"imp":26.4053258616,"elo":1085.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3998340801,"x_train":21385.328840253,"imp":25.9929582132,"elo":1095.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"x_train":28729.7373655693,"imp":25.6703338597,"elo":1105.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719683769,"x_train":1.1753394519,"imp":16.1657705633,"elo":1351.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1258046788,"x_train":9.9925632138,"imp":22.8850366269,"elo":1081.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.250263975,"x_train":23.3976730308,"imp":21.1661879002,"elo":1132.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5082028825,"x_train":67.83522273,"imp":19.6430780766,"elo":1180.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7461273476,"x_train":150.5961763502,"imp":18.9177332705,"elo":1212.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1062947766,"x_train":358.8914252792,"imp":18.2366603106,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3608592506,"x_train":711.8242682675,"imp":17.8219921337,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5999240551,"x_train":1402.3570975877,"imp":17.5279048071,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.7712720949,"x_train":2125.0611186926,"imp":17.3712258363,"elo":1282.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.95162543,"x_train":2875.5203407747,"imp":17.237392444,"elo":1288.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1226232881,"x_train":1.9409389396,"imp":18.8687466883,"elo":1211.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5491058954,"x_train":5.7736976246,"imp":17.223916001,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9903293256,"x_train":17.1658020762,"imp":16.6921946861,"elo":1323.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1808822728,"x_train":34.0813643193,"imp":16.3749596179,"elo":1342.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.5144136771,"x_train":86.4155038646,"imp":16.1290997784,"elo":1356.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.6869916704,"x_train":172.6999358448,"imp":16.0259383344,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.732994711,"x_train":344.2808053206,"imp":15.9436458826,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.809783177,"x_train":515.1408947248,"imp":15.8844126064,"elo":1371.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7397156138,"x_train":693.4907982506,"imp":15.836256277,"elo":1375.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":68.173296937,"x_train":296.6392917616,"imp":23.0966015493,"elo":1098.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":133.585345468,"x_train":761.0284006744,"imp":20.4062004234,"elo":1164.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":253.2846359591,"x_train":2323.5909795411,"imp":18.8693887945,"elo":1266.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":348.3200767517,"x_train":4948.4562156001,"imp":17.5999354536,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":469.7064233571,"x_train":12120.2560802151,"imp":15.4129316692,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":463.2685781743,"x_train":12685.0815501834,"imp":15.3839367442,"elo":1402.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"x_train":3.2328671519,"imp":26.5176586874,"elo":1045.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8711410437,"x_train":7.7757953049,"imp":23.2835232661,"elo":1109.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1335512608,"x_train":20.6204304219,"imp":19.3320675044,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.4441868365,"x_train":41.9962580807,"imp":17.5849304711,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8072467962,"x_train":105.704979404,"imp":16.7547041225,"elo":1335.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.1086840459,"x_train":210.3194448262,"imp":16.4855249302,"elo":1346.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.3155613238,"x_train":422.0870812379,"imp":16.3830505277,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.5338332871,"x_train":631.9086612835,"imp":16.3618097985,"elo":1349.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.5660475163,"x_train":846.8860702337,"imp":16.358428191,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..d015e7f7f7c30d3b26e94f23c6ddc3ef583b959d
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_all/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (default)", "RealMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "AutoGluon 1.4 (best, 4h)", "TabDPT (tuned + ensembled)", "TabM (tuned + ensembled)", "TabDPT (tuned)", "RealMLP (tuned)", "TabPFNv2 (tuned + ensembled)", "Mitra (default)", "LightGBM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "TabM (tuned)", "LimiX (default)", "TabSwift (default)", "TabDPT (default)", "CatBoost (tuned)", "iLTM (tuned + ensembled)", "LightGBM (tuned)", "xRFM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabPFNv2 (tuned)", "CatBoost (default)", "TabPFNv2 (default)", "ModernNCA (tuned + ensembled)", "ModernNCA (tuned)", "XGBoost (tuned)", "SAP-RPT-OSS (default)", "xRFM (tuned)", "TabICL (default)", "TabM (default)", "BetaTabPFN (default)", "ChimeraBoost (tuned + ensembled)", "EBM (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "iLTM (tuned)", "ModernNCA (default)", "ChimeraBoost (tuned)", "RealMLP (default)", "EBM (tuned)", "ExtraTrees (tuned + ensembled)", "FastaiMLP (tuned + ensembled)", "ChimeraBoost (default)", "EBM (default)", "TorchMLP (tuned)", "ExtraTrees (tuned)", "Nori-30M (default)", "Nori (default)", "XGBoost (default)", "FastaiMLP (tuned)", "RandomForest (tuned + ensembled)", "LightGBM (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned)", "OrionMSP (default)", "TorchMLP (default)", "PerpetualBooster (tuned + ensembled)", "TabFlex (default)", "iLTM (default)", "xRFM (default)", "PerpetualBooster (tuned)", "TabSTAR (default)", "ExtraTrees (default)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "RandomForest (default)", "KNN (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (default)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7708333134651184, 0.751225471496582, 0.8088235259056091, 0.8174019455909729, 0.7873774766921997, 0.7977941036224365, 0.8357843160629272, 0.8651960492134094, 0.8799019455909729, 0.8897058963775635, 0.8823529481887817, 0.9031862616539001, 0.8958333134651184, 0.875, 0.8799019455909729, 0.8933823704719543, 0.9154411554336548, 0.9154411554336548, 0.9264705777168274, 0.90625, 0.8860294222831726, 0.8995097875595093, 0.9276960492134094, 0.9240196347236633, 0.9276960492134094, 0.9203431606292725, 0.9350489974021912, 0.8909313678741455, 0.9227941036224365, 0.8848039507865906, 0.9424019455909729, 0.9313725233078003, 0.9411764740943909, 0.8786764740943909, 0.9154411554336548, 0.9522058963775635, 0.9325980544090271, 0.9111519455909729, 0.9350489974021912, 0.9264705777168274, 0.9595588445663452, 0.9424019455909729, 0.938725471496582, 0.9399510025978088, 0.9473039507865906, 0.9375, 0.9534313678741455, 0.9264705777168274, 0.9546568393707275, 0.9350489974021912, 0.969362735748291, 0.9485294222831726, 0.8811274766921997, 0.9313725233078003, 0.9718137383460999, 0.9497548937797546, 0.969362735748291, 0.9718137383460999, 0.8725489974021912, 0.8713235259056091, 0.968137264251709, 0.9558823704719543, 0.9754902124404907, 0.9865196347236633, 0.9534313678741455, 0.9571078419685364, 0.970588207244873, 0.9852941036224365, 0.904411792755127, 0.9865196347236633, 0.9767156839370728, 0.9583333134651184, 0.9852941036224365, 0.969362735748291, 0.9632353186607361, 0.9877451062202454, 0.9754902124404907, 0.9718137383460999, 0.9938725233078003], [0.2291666716337204, NaN, 0.5796568393707275, 0.5551470518112183, 0.6348039507865906, 0.6188725233078003, 0.6703431606292725, 0.7071078419685364, 0.7450980544090271, 0.7745097875595093, 0.751225471496582, 0.7585784196853638, 0.7977941036224365, 0.7916666865348816, 0.779411792755127, 0.779411792755127, 0.7732843160629272, 0.8259803652763367, 0.8345588445663452, 0.8370097875595093, 0.7843137383460999, 0.8823529481887817, 0.8272058963775635, 0.8406862616539001, 0.8639705777168274, 0.8615196347236633, 0.8529411554336548, 0.8486519455909729, 0.8400735259056091, 0.8541666865348816, 0.8192402124404907, 0.8786764740943909, 0.8651960492134094, 0.8590686321258545, 0.7965686321258545, 0.8639705777168274, 0.8854166865348816, 0.8700980544090271, 0.8302696347236633, 0.8860294222831726, 0.8590686321258545, 0.9105392098426819, 0.9258578419685364, 0.905024528503418, 0.8860294222831726, 0.9252451062202454, 0.876225471496582, 0.9350489974021912, 0.8841911554336548, 0.8958333134651184, 0.8964460492134094, 0.938725471496582, 0.936274528503418, 0.8566176295280457, 0.8884803652763367, 0.9509803652763367, 0.9209558963775635, 0.938725471496582, 0.9595588445663452, 0.8511029481887817, 0.8492646813392639, 0.9546568393707275, 0.9448529481887817, 0.9656862616539001, 0.9473039507865906, 0.9473039507865906, 0.9607843160629272, 0.9620097875595093, 0.9644607901573181, 0.8897058963775635, 0.9754902124404907, 0.966911792755127, 0.9325980544090271, 0.9852941036224365, 0.9485294222831726, 0.938725471496582, 0.9779411554336548, 0.9632353186607361, 0.9656862616539001, 0.9950980544090271], [0.24877451360225677, 0.4203431308269501, NaN, 0.4791666567325592, 0.6200980544090271, 0.5539215803146362, 0.6090686321258545, 0.6348039507865906, 0.7775735259056091, 0.6715686321258545, 0.8051470518112183, 0.7377451062202454, 0.8259803652763367, 0.7610294222831726, 0.8186274766921997, 0.7156862616539001, 0.7444853186607361, 0.8615196347236633, 0.8639705777168274, 0.8388480544090271, 0.7769607901573181, 0.8357843160629272, 0.7659313678741455, 0.8670343160629272, 0.8909313678741455, 0.876225471496582, 0.875, 0.8884803652763367, 0.7990196347236633, 0.8792892098426819, 0.8088235259056091, 0.8860294222831726, 0.8970588445663452, 0.8946078419685364, 0.7720588445663452, 0.8694853186607361, 0.8645833134651184, 0.8835784196853638, 0.8382353186607361, 0.9166666865348816, 0.8897058963775635, 0.9252451062202454, 0.9154411554336548, 0.8958333134651184, 0.9240196347236633, 0.9117646813392639, 0.8982843160629272, 0.9240196347236633, 0.8982843160629272, 0.9301470518112183, 0.8995097875595093, 0.9448529481887817, 0.9166666865348816, 0.7953431606292725, 0.843137264251709, 0.9620097875595093, 0.9166666865348816, 0.9436274766921997, 0.9607843160629272, 0.8357843160629272, 0.8357843160629272, 0.9460784196853638, 0.9479166865348816, 0.969362735748291, 0.9620097875595093, 0.936274528503418, 0.9583333134651184, 0.9595588445663452, 0.967524528503418, 0.873774528503418, 0.9583333134651184, 0.9656862616539001, 0.9276960492134094, 0.9730392098426819, 0.9509803652763367, 0.9436274766921997, 0.9742646813392639, 0.9601715803146362, 0.9644607901573181, 0.9938725233078003], [0.19117647409439087, 0.44485294818878174, 0.5208333134651184, NaN, 0.5870097875595093, 0.5674019455909729, 0.6090686321258545, 0.6525735259056091, 0.7156862616539001, 0.7316176295280457, 0.7450980544090271, 0.7665441036224365, 0.7867646813392639, 0.7683823704719543, 0.7524510025978088, 0.7414215803146362, 0.750612735748291, 0.8088235259056091, 0.811274528503418, 0.8088235259056091, 0.7861519455909729, 0.8370097875595093, 0.7928921580314636, 0.8149510025978088, 0.843137264251709, 0.8382353186607361, 0.8247548937797546, 0.8247548937797546, 0.7922794222831726, 0.8382353186607361, 0.8014705777168274, 0.843137264251709, 0.8480392098426819, 0.8296568393707275, 0.780637264251709, 0.8204656839370728, 0.8792892098426819, 0.8566176295280457, 0.8235294222831726, 0.8566176295280457, 0.845588207244873, 0.8799019455909729, 0.875, 0.8841911554336548, 0.8664215803146362, 0.9019607901573181, 0.8639705777168274, 0.8976715803146362, 0.8700980544090271, 0.876225471496582, 0.8897058963775635, 0.908088207244873, 0.8995097875595093, 0.8272058963775635, 0.8578431606292725, 0.9178921580314636, 0.906862735748291, 0.9252451062202454, 0.9411764740943909, 0.844362735748291, 0.8394607901573181, 0.9276960492134094, 0.936274528503418, 0.9460784196853638, 0.9460784196853638, 0.9276960492134094, 0.9399510025978088, 0.9448529481887817, 0.9497548937797546, 0.8921568393707275, 0.9583333134651184, 0.9497548937797546, 0.9411764740943909, 0.968137264251709, 0.9546568393707275, 0.9473039507865906, 0.9718137383460999, 0.9662989974021912, 0.9644607901573181, 0.9950980544090271], [0.1825980395078659, 0.3651960790157318, 0.3799019753932953, 0.41299018263816833, NaN, 0.4828431308269501, 0.5435048937797546, 0.6139705777168274, 0.6911764740943909, 0.6433823704719543, 0.7095588445663452, 0.6727941036224365, 0.7757353186607361, 0.7022058963775635, 0.7634803652763367, 0.6911764740943909, 0.7475489974021912, 0.8075980544090271, 0.813725471496582, 0.8149510025978088, 0.751225471496582, 0.7990196347236633, 0.7218137383460999, 0.8149510025978088, 0.841911792755127, 0.8529411554336548, 0.8235294222831726, 0.8492646813392639, 0.7928921580314636, 0.8517156839370728, 0.7726715803146362, 0.8590686321258545, 0.8602941036224365, 0.8713235259056091, 0.7916666865348816, 0.843137264251709, 0.8088235259056091, 0.8682597875595093, 0.8198529481887817, 0.8946078419685364, 0.8615196347236633, 0.8958333134651184, 0.906862735748291, 0.8982843160629272, 0.9031862616539001, 0.9093137383460999, 0.8915441036224365, 0.908088207244873, 0.8725489974021912, 0.9105392098426819, 0.8872548937797546, 0.9350489974021912, 0.905024528503418, 0.7941176295280457, 0.8357843160629272, 0.9620097875595093, 0.9093137383460999, 0.9313725233078003, 0.9583333134651184, 0.8370097875595093, 0.8406862616539001, 0.9485294222831726, 0.9129902124404907, 0.9632353186607361, 0.9558823704719543, 0.9338235259056091, 0.9460784196853638, 0.9485294222831726, 0.9534313678741455, 0.8688725233078003, 0.9607843160629272, 0.9509803652763367, 0.9252451062202454, 0.9779411554336548, 0.9485294222831726, 0.9399510025978088, 0.9730392098426819, 0.9571078419685364, 0.9607843160629272, 0.9926470518112183], [0.21262255311012268, 0.3811274468898773, 0.44607841968536377, 0.4325980246067047, 0.5171568393707275, NaN, 0.6666666865348816, 0.6642156839370728, 0.6936274766921997, 0.623774528503418, 0.7352941036224365, 0.6642156839370728, 0.7610294222831726, 0.7022058963775635, 0.7475489974021912, 0.7647058963775635, 0.7450980544090271, 0.7928921580314636, 0.7977941036224365, 0.7990196347236633, 0.7598039507865906, 0.7965686321258545, 0.6985294222831726, 0.8088235259056091, 0.7922794222831726, 0.8204656839370728, 0.8174019455909729, 0.8284313678741455, 0.8406862616539001, 0.8284313678741455, 0.8247548937797546, 0.8088235259056091, 0.8517156839370728, 0.842524528503418, 0.78125, 0.8272058963775635, 0.78125, 0.8529411554336548, 0.8333333134651184, 0.8486519455909729, 0.8566176295280457, 0.8872548937797546, 0.873774528503418, 0.8627451062202454, 0.8664215803146362, 0.8823529481887817, 0.8780637383460999, 0.8774510025978088, 0.8682597875595093, 0.8872548937797546, 0.8799019455909729, 0.9203431606292725, 0.8799019455909729, 0.7990196347236633, 0.841911792755127, 0.9301470518112183, 0.9093137383460999, 0.9178921580314636, 0.9485294222831726, 0.8296568393707275, 0.8296568393707275, 0.9325980544090271, 0.9289215803146362, 0.9571078419685364, 0.9571078419685364, 0.9301470518112183, 0.936274528503418, 0.936274528503418, 0.9583333134651184, 0.8639705777168274, 0.9558823704719543, 0.9522058963775635, 0.9203431606292725, 0.9656862616539001, 0.9301470518112183, 0.9313725233078003, 0.9742646813392639, 0.9491421580314636, 0.9485294222831726, 0.9914215803146362], [0.20220588147640228, 0.3296568691730499, 0.3909313678741455, 0.3909313678741455, 0.45649510622024536, 0.3333333432674408, NaN, 0.5759803652763367, 0.6421568393707275, 0.5894607901573181, 0.6752451062202454, 0.6213235259056091, 0.7267156839370728, 0.6623774766921997, 0.7071078419685364, 0.7138480544090271, 0.7083333134651184, 0.7450980544090271, 0.7585784196853638, 0.780024528503418, 0.7242646813392639, 0.7640931606292725, 0.6740196347236633, 0.7647058963775635, 0.75, 0.7953431606292725, 0.7910539507865906, 0.7941176295280457, 0.8094362616539001, 0.7855392098426819, 0.7904411554336548, 0.7738970518112183, 0.7990196347236633, 0.8063725233078003, 0.7524510025978088, 0.8100489974021912, 0.7561274766921997, 0.8308823704719543, 0.8125, 0.8094362616539001, 0.8370097875595093, 0.8639705777168274, 0.8541666865348816, 0.8394607901573181, 0.8345588445663452, 0.8664215803146362, 0.8529411554336548, 0.8670343160629272, 0.8504902124404907, 0.8639705777168274, 0.8651960492134094, 0.904411792755127, 0.875, 0.7757353186607361, 0.8174019455909729, 0.9117646813392639, 0.9031862616539001, 0.8897058963775635, 0.9375, 0.8051470518112183, 0.8100489974021912, 0.9117646813392639, 0.9178921580314636, 0.9558823704719543, 0.938112735748291, 0.9154411554336548, 0.9301470518112183, 0.9264705777168274, 0.9448529481887817, 0.8566176295280457, 0.938725471496582, 0.9448529481887817, 0.9093137383460999, 0.9558823704719543, 0.9178921580314636, 0.9240196347236633, 0.9742646813392639, 0.9515931606292725, 0.9509803652763367, 0.9865196347236633], [0.16421568393707275, 0.2928921580314636, 0.3651960790157318, 0.34742647409439087, 0.3860294222831726, 0.33578431606292725, 0.42401960492134094, NaN, 0.5894607901573181, 0.5502451062202454, 0.6593137383460999, 0.5796568393707275, 0.7132353186607361, 0.6280637383460999, 0.6764705777168274, 0.6642156839370728, 0.7034313678741455, 0.7303921580314636, 0.7475489974021912, 0.7585784196853638, 0.7279411554336548, 0.7708333134651184, 0.6691176295280457, 0.7647058963775635, 0.7267156839370728, 0.7855392098426819, 0.7843137383460999, 0.7745097875595093, 0.7647058963775635, 0.7904411554336548, 0.7781862616539001, 0.7365196347236633, 0.783088207244873, 0.8039215803146362, 0.7634803652763367, 0.813725471496582, 0.7610294222831726, 0.8308823704719543, 0.813112735748291, 0.8161764740943909, 0.8198529481887817, 0.8639705777168274, 0.8584558963775635, 0.844362735748291, 0.8272058963775635, 0.8602941036224365, 0.8480392098426819, 0.8700980544090271, 0.873774528503418, 0.8664215803146362, 0.8639705777168274, 0.904411792755127, 0.8823529481887817, 0.7536764740943909, 0.7610294222831726, 0.9007353186607361, 0.905637264251709, 0.876225471496582, 0.9338235259056091, 0.8308823704719543, 0.8302696347236633, 0.9117646813392639, 0.936274528503418, 0.9718137383460999, 0.9350489974021912, 0.9319853186607361, 0.9399510025978088, 0.9338235259056091, 0.9424019455909729, 0.8713235259056091, 0.9424019455909729, 0.9485294222831726, 0.9215686321258545, 0.9546568393707275, 0.908088207244873, 0.9264705777168274, 0.969362735748291, 0.9620097875595093, 0.9473039507865906, 0.9852941036224365], [0.1348039209842682, 0.2549019753932953, 0.22242647409439087, 0.28431373834609985, 0.30882352590560913, 0.3063725531101227, 0.3578431308269501, 0.4105392098426819, NaN, 0.49754902720451355, 0.53125, 0.4908088147640228, 0.6151960492134094, 0.5349264740943909, 0.6942402124404907, 0.5287989974021912, 0.5735294222831726, 0.7034313678741455, 0.719362735748291, 0.6825980544090271, 0.5906862616539001, 0.595588207244873, 0.5882353186607361, 0.7254902124404907, 0.7401960492134094, 0.7579656839370728, 0.7671568393707275, 0.7953431606292725, 0.626225471496582, 0.7438725233078003, 0.6409313678741455, 0.7034313678741455, 0.7291666865348816, 0.813725471496582, 0.6715686321258545, 0.7720588445663452, 0.6446078419685364, 0.8149510025978088, 0.7316176295280457, 0.7696078419685364, 0.7977941036224365, 0.781862735748291, 0.8259803652763367, 0.7965686321258545, 0.7910539507865906, 0.9013480544090271, 0.8174019455909729, 0.8480392098426819, 0.8474264740943909, 0.8149510025978088, 0.8388480544090271, 0.8321078419685364, 0.8572303652763367, 0.7708333134651184, 0.8075980544090271, 0.9105392098426819, 0.8609068393707275, 0.9117646813392639, 0.9276960492134094, 0.781862735748291, 0.7855392098426819, 0.9313725233078003, 0.8406862616539001, 0.936274528503418, 0.9436274766921997, 0.8725489974021912, 0.9117646813392639, 0.938725471496582, 0.9509803652763367, 0.8370097875595093, 0.9607843160629272, 0.9522058963775635, 0.905637264251709, 0.9656862616539001, 0.9252451062202454, 0.906862735748291, 0.9754902124404907, 0.9203431606292725, 0.9460784196853638, 0.9938725233078003], [0.1200980395078659, 0.22549019753932953, 0.3284313678741455, 0.26838234066963196, 0.35661765933036804, 0.3762255012989044, 0.4105392098426819, 0.44975489377975464, 0.5024510025978088, NaN, 0.5563725233078003, 0.5147058963775635, 0.5741421580314636, 0.5808823704719543, 0.5839460492134094, 0.5600489974021912, 0.594362735748291, 0.6286764740943909, 0.6654411554336548, 0.6323529481887817, 0.6593137383460999, 0.7077205777168274, 0.7028186321258545, 0.6642156839370728, 0.7022058963775635, 0.6654411554336548, 0.6936274766921997, 0.6875, 0.6299019455909729, 0.6936274766921997, 0.6458333134651184, 0.6850489974021912, 0.7046568393707275, 0.7022058963775635, 0.6629902124404907, 0.7365196347236633, 0.6924019455909729, 0.716911792755127, 0.686274528503418, 0.7450980544090271, 0.7120097875595093, 0.7757353186607361, 0.7984068393707275, 0.7990196347236633, 0.7671568393707275, 0.8051470518112183, 0.7469362616539001, 0.8486519455909729, 0.813725471496582, 0.7916666865348816, 0.7904411554336548, 0.8284313678741455, 0.8713235259056091, 0.7549019455909729, 0.7781862616539001, 0.8553921580314636, 0.8627451062202454, 0.8713235259056091, 0.8823529481887817, 0.7647058963775635, 0.7671568393707275, 0.8823529481887817, 0.8995097875595093, 0.904411792755127, 0.9019607901573181, 0.9276960492134094, 0.8811274766921997, 0.906862735748291, 0.9276960492134094, 0.845588207244873, 0.9454656839370728, 0.938725471496582, 0.8725489974021912, 0.9620097875595093, 0.9252451062202454, 0.8811274766921997, 0.969362735748291, 0.9436274766921997, 0.9571078419685364, 0.9901960492134094], [0.11029411852359772, 0.24877451360225677, 0.19485294818878174, 0.2549019753932953, 0.2904411852359772, 0.2647058963775635, 0.32475489377975464, 0.34068626165390015, 0.46875, 0.4436274468898773, NaN, 0.4436274468898773, 0.5674019455909729, 0.47549018263816833, 0.5882353186607361, 0.4522058963775635, 0.48651960492134094, 0.720588207244873, 0.686274528503418, 0.6372548937797546, 0.5453431606292725, 0.5814951062202454, 0.5563725233078003, 0.7046568393707275, 0.7009803652763367, 0.7598039507865906, 0.6917892098426819, 0.7696078419685364, 0.5416666865348816, 0.75, 0.563725471496582, 0.688725471496582, 0.6997548937797546, 0.7696078419685364, 0.6176470518112183, 0.7377451062202454, 0.6470588445663452, 0.7463235259056091, 0.7071078419685364, 0.7971813678741455, 0.7732843160629272, 0.8186274766921997, 0.8075980544090271, 0.7855392098426819, 0.8149510025978088, 0.7977941036224365, 0.7928921580314636, 0.783088207244873, 0.8247548937797546, 0.8504902124404907, 0.8284313678741455, 0.8578431606292725, 0.8063725233078003, 0.7218137383460999, 0.7426470518112183, 0.9105392098426819, 0.8627451062202454, 0.8566176295280457, 0.9240196347236633, 0.7769607901573181, 0.779411792755127, 0.8817402124404907, 0.814338207244873, 0.9215686321258545, 0.936274528503418, 0.876225471496582, 0.9129902124404907, 0.904411792755127, 0.9375, 0.8321078419685364, 0.9301470518112183, 0.9473039507865906, 0.9025735259056091, 0.9411764740943909, 0.904411792755127, 0.9197303652763367, 0.9595588445663452, 0.9411764740943909, 0.9375, 0.9901960492134094], [0.11764705926179886, 0.24142156541347504, 0.26225489377975464, 0.23345588147640228, 0.3272058963775635, 0.33578431606292725, 0.37867647409439087, 0.4203431308269501, 0.5091911554336548, 0.4852941036224365, 0.5563725233078003, NaN, 0.5729166865348816, 0.6593137383460999, 0.5827205777168274, 0.5526960492134094, 0.5772058963775635, 0.6323529481887817, 0.6372548937797546, 0.624387264251709, 0.5986519455909729, 0.6439951062202454, 0.7046568393707275, 0.6470588445663452, 0.6973039507865906, 0.6617646813392639, 0.6991421580314636, 0.6734068393707275, 0.6415441036224365, 0.686274528503418, 0.6384803652763367, 0.6936274766921997, 0.6960784196853638, 0.7034313678741455, 0.6605392098426819, 0.7212010025978088, 0.6317402124404907, 0.7101715803146362, 0.6629902124404907, 0.7303921580314636, 0.7077205777168274, 0.749387264251709, 0.7892156839370728, 0.7787989974021912, 0.75, 0.7781862616539001, 0.7322303652763367, 0.8174019455909729, 0.7781862616539001, 0.7879902124404907, 0.748774528503418, 0.8100489974021912, 0.8370097875595093, 0.7843137383460999, 0.8051470518112183, 0.841911792755127, 0.8247548937797546, 0.8700980544090271, 0.8774510025978088, 0.751225471496582, 0.7524510025978088, 0.8786764740943909, 0.876225471496582, 0.8946078419685364, 0.9117646813392639, 0.8909313678741455, 0.8713235259056091, 0.9172794222831726, 0.9209558963775635, 0.8357843160629272, 0.9509803652763367, 0.9203431606292725, 0.8406862616539001, 0.9558823704719543, 0.9172794222831726, 0.8541666865348816, 0.9571078419685364, 0.904411792755127, 0.9436274766921997, 0.9938725233078003], [0.09681372344493866, 0.20220588147640228, 0.17401960492134094, 0.2132352888584137, 0.2242647111415863, 0.2389705926179886, 0.27328431606292725, 0.2867647111415863, 0.3848039209842682, 0.4258578419685364, 0.4325980246067047, 0.4270833432674408, NaN, 0.469362735748291, 0.49754902720451355, 0.44117647409439087, 0.4742647111415863, 0.591911792755127, 0.6017156839370728, 0.719362735748291, 0.5098039507865906, 0.5343137383460999, 0.5404411554336548, 0.6188725233078003, 0.6011029481887817, 0.6513480544090271, 0.5968137383460999, 0.6727941036224365, 0.5355392098426819, 0.6838235259056091, 0.5398284196853638, 0.6286764740943909, 0.6403186321258545, 0.7120097875595093, 0.623774528503418, 0.625, 0.6151960492134094, 0.8609068393707275, 0.720588207244873, 0.7218137383460999, 0.7328431606292725, 0.7340686321258545, 0.7267156839370728, 0.7267156839370728, 0.7585784196853638, 0.8229166865348816, 0.7769607901573181, 0.7426470518112183, 0.7757353186607361, 0.8075980544090271, 0.7861519455909729, 0.814338207244873, 0.748774528503418, 0.686274528503418, 0.7083333134651184, 0.8578431606292725, 0.8308823704719543, 0.7867646813392639, 0.8829656839370728, 0.7732843160629272, 0.7757353186607361, 0.8210784196853638, 0.8075980544090271, 0.9227941036224365, 0.8860294222831726, 0.8578431606292725, 0.905637264251709, 0.8590686321258545, 0.905637264251709, 0.8296568393707275, 0.8921568393707275, 0.9178921580314636, 0.9031862616539001, 0.9203431606292725, 0.8909313678741455, 0.9148284196853638, 0.9558823704719543, 0.9350489974021912, 0.9276960492134094, 0.9754902124404907], [0.1041666641831398, 0.2083333283662796, 0.2389705926179886, 0.23161764442920685, 0.2977941036224365, 0.2977941036224365, 0.3376225531101227, 0.37193626165390015, 0.46507352590560913, 0.41911765933036804, 0.5245097875595093, 0.34068626165390015, 0.530637264251709, NaN, 0.5257353186607361, 0.5214460492134094, 0.561274528503418, 0.5857843160629272, 0.595588207244873, 0.5870097875595093, 0.5759803652763367, 0.5882353186607361, 0.6458333134651184, 0.6060048937797546, 0.6311274766921997, 0.6225489974021912, 0.654411792755127, 0.6439951062202454, 0.5998774766921997, 0.6421568393707275, 0.5974264740943909, 0.6464460492134094, 0.6470588445663452, 0.654411792755127, 0.6139705777168274, 0.6691176295280457, 0.5900735259056091, 0.6703431606292725, 0.6213235259056091, 0.6960784196853638, 0.6666666865348816, 0.7095588445663452, 0.7340686321258545, 0.7242646813392639, 0.7095588445663452, 0.7340686321258545, 0.7058823704719543, 0.7732843160629272, 0.7365196347236633, 0.7475489974021912, 0.7242646813392639, 0.7634803652763367, 0.7916666865348816, 0.7377451062202454, 0.7720588445663452, 0.7855392098426819, 0.7953431606292725, 0.811274528503418, 0.8321078419685364, 0.7224264740943909, 0.7218137383460999, 0.8394607901573181, 0.8713235259056091, 0.8725489974021912, 0.8933823704719543, 0.8786764740943909, 0.8345588445663452, 0.8933823704719543, 0.9019607901573181, 0.8235294222831726, 0.9227941036224365, 0.905637264251709, 0.8265931606292725, 0.9301470518112183, 0.9105392098426819, 0.841911792755127, 0.9558823704719543, 0.8921568393707275, 0.9350489974021912, 0.9914215803146362], [0.125, 0.22058823704719543, 0.18137255311012268, 0.24754901230335236, 0.23651960492134094, 0.25245097279548645, 0.2928921580314636, 0.3235294222831726, 0.30575981736183167, 0.4160539209842682, 0.4117647111415863, 0.4172794222831726, 0.5024510025978088, 0.4742647111415863, NaN, 0.4528186321258545, 0.48651960492134094, 0.5833333134651184, 0.5882353186607361, 0.5857843160629272, 0.53125, 0.5183823704719543, 0.5355392098426819, 0.5968137383460999, 0.59375, 0.6421568393707275, 0.6200980544090271, 0.6764705777168274, 0.5404411554336548, 0.6421568393707275, 0.5686274766921997, 0.6078431606292725, 0.6421568393707275, 0.7040441036224365, 0.6078431606292725, 0.6721813678741455, 0.5894607901573181, 0.7316176295280457, 0.6899510025978088, 0.6666666865348816, 0.720588207244873, 0.7340686321258545, 0.7303921580314636, 0.7414215803146362, 0.7034313678741455, 0.8026960492134094, 0.7585784196853638, 0.7622548937797546, 0.7732843160629272, 0.75, 0.7585784196853638, 0.8051470518112183, 0.7843137383460999, 0.7071078419685364, 0.719362735748291, 0.845588207244873, 0.8296568393707275, 0.844362735748291, 0.8566176295280457, 0.7659313678741455, 0.7671568393707275, 0.8627451062202454, 0.7971813678741455, 0.905637264251709, 0.9007353186607361, 0.8584558963775635, 0.876225471496582, 0.8811274766921997, 0.908088207244873, 0.8210784196853638, 0.8897058963775635, 0.9307597875595093, 0.8651960492134094, 0.9203431606292725, 0.8897058963775635, 0.8719362616539001, 0.9620097875595093, 0.8995097875595093, 0.9166666865348816, 0.9828431606292725], [0.1200980395078659, 0.22058823704719543, 0.28431373834609985, 0.25857841968536377, 0.30882352590560913, 0.23529411852359772, 0.2861519753932953, 0.33578431606292725, 0.47120097279548645, 0.43995097279548645, 0.5477941036224365, 0.4473039209842682, 0.5588235259056091, 0.4785539209842682, 0.5471813678741455, NaN, 0.5232843160629272, 0.5863970518112183, 0.5980392098426819, 0.6011029481887817, 0.5232843160629272, 0.5575980544090271, 0.5196078419685364, 0.6047794222831726, 0.6139705777168274, 0.6164215803146362, 0.6458333134651184, 0.6378676295280457, 0.6470588445663452, 0.6421568393707275, 0.6464460492134094, 0.6519607901573181, 0.6470588445663452, 0.6519607901573181, 0.5765931606292725, 0.6599264740943909, 0.594975471496582, 0.6507353186607361, 0.6825980544090271, 0.6825980544090271, 0.6954656839370728, 0.6776960492134094, 0.6911764740943909, 0.6623774766921997, 0.6936274766921997, 0.7120097875595093, 0.719362735748291, 0.6917892098426819, 0.7408088445663452, 0.720588207244873, 0.7450980544090271, 0.7328431606292725, 0.7046568393707275, 0.65625, 0.686887264251709, 0.7671568393707275, 0.7677696347236633, 0.7414215803146362, 0.7732843160629272, 0.7144607901573181, 0.7156862616539001, 0.779411792755127, 0.7689951062202454, 0.8088235259056091, 0.8026960492134094, 0.810661792755127, 0.8063725233078003, 0.8602941036224365, 0.8272058963775635, 0.7769607901573181, 0.8909313678741455, 0.8651960492134094, 0.8553921580314636, 0.8651960492134094, 0.8541666865348816, 0.8713235259056091, 0.9252451062202454, 0.8860294222831726, 0.908088207244873, 0.9938725233078003], [0.10661764442920685, 0.22671568393707275, 0.2555147111415863, 0.2493872493505478, 0.25245097279548645, 0.2549019753932953, 0.2916666567325592, 0.2965686321258545, 0.4264705777168274, 0.405637264251709, 0.5134803652763367, 0.4227941036224365, 0.5257353186607361, 0.4387255012989044, 0.5134803652763367, 0.47671568393707275, NaN, 0.5735294222831726, 0.5735294222831726, 0.595588207244873, 0.532475471496582, 0.5355392098426819, 0.5122548937797546, 0.5729166865348816, 0.6035539507865906, 0.6115196347236633, 0.5821078419685364, 0.6145833134651184, 0.5594362616539001, 0.6421568393707275, 0.5735294222831726, 0.6348039507865906, 0.6433823704719543, 0.623774528503418, 0.5674019455909729, 0.6207107901573181, 0.6041666865348816, 0.6482843160629272, 0.6666666865348816, 0.6899510025978088, 0.7156862616539001, 0.719362735748291, 0.6764705777168274, 0.6850489974021912, 0.7009803652763367, 0.6936274766921997, 0.7450980544090271, 0.6789215803146362, 0.7573529481887817, 0.720588207244873, 0.7536764740943909, 0.7708333134651184, 0.6917892098426819, 0.6482843160629272, 0.6642156839370728, 0.7745097875595093, 0.8002451062202454, 0.7107843160629272, 0.7634803652763367, 0.7549019455909729, 0.751225471496582, 0.7567402124404907, 0.7738970518112183, 0.8198529481887817, 0.7977941036224365, 0.8235294222831726, 0.8259803652763367, 0.8468137383460999, 0.8161764740943909, 0.7941176295280457, 0.8480392098426819, 0.8639705777168274, 0.8676470518112183, 0.8639705777168274, 0.8504902124404907, 0.8835784196853638, 0.9313725233078003, 0.9105392098426819, 0.9099264740943909, 0.9901960492134094], [0.08455882221460342, 0.17401960492134094, 0.13848039507865906, 0.19117647409439087, 0.1924019604921341, 0.20710784196853638, 0.2549019753932953, 0.2696078419685364, 0.2965686321258545, 0.37132352590560913, 0.27941176295280457, 0.36764705181121826, 0.40808823704719543, 0.41421568393707275, 0.4166666567325592, 0.41360294818878174, 0.4264705777168274, NaN, 0.5183823704719543, 0.4944852888584137, 0.44485294818878174, 0.5257353186607361, 0.5134803652763367, 0.5520833134651184, 0.5275735259056091, 0.655024528503418, 0.5539215803146362, 0.6801470518112183, 0.4901960790157318, 0.6709558963775635, 0.4987744987010956, 0.5465686321258545, 0.5772058963775635, 0.716911792755127, 0.5563725233078003, 0.6151960492134094, 0.5784313678741455, 0.688112735748291, 0.6066176295280457, 0.7224264740943909, 0.7040441036224365, 0.6397058963775635, 0.6715686321258545, 0.6924019455909729, 0.7634803652763367, 0.7230392098426819, 0.7475489974021912, 0.7555146813392639, 0.7144607901573181, 0.8321078419685364, 0.8174019455909729, 0.7340686321258545, 0.7585784196853638, 0.7071078419685364, 0.7083333134651184, 0.8958333134651184, 0.7904411554336548, 0.8345588445663452, 0.9571078419685364, 0.7591911554336548, 0.7598039507865906, 0.873774528503418, 0.8039215803146362, 0.8492646813392639, 0.9142156839370728, 0.8357843160629272, 0.9191176295280457, 0.8321078419685364, 0.9289215803146362, 0.8284313678741455, 0.9178921580314636, 0.8823529481887817, 0.8811274766921997, 0.9485294222831726, 0.8982843160629272, 0.8958333134651184, 0.9595588445663452, 0.936274528503418, 0.9344362616539001, 0.9865196347236633], [0.08455882221460342, 0.16544117033481598, 0.1360294073820114, 0.18872548639774323, 0.18627451360225677, 0.20220588147640228, 0.24142156541347504, 0.25245097279548645, 0.280637264251709, 0.3345588147640228, 0.3137255012989044, 0.36274510622024536, 0.39828431606292725, 0.40441176295280457, 0.4117647111415863, 0.4019607901573181, 0.4264705777168274, 0.48161765933036804, NaN, 0.46568626165390015, 0.45588234066963196, 0.5171568393707275, 0.47181373834609985, 0.594975471496582, 0.5232843160629272, 0.5508578419685364, 0.5404411554336548, 0.626838207244873, 0.4791666567325592, 0.6482843160629272, 0.4901960790157318, 0.5747548937797546, 0.5851715803146362, 0.657475471496582, 0.5808823704719543, 0.6200980544090271, 0.5796568393707275, 0.6372548937797546, 0.6164215803146362, 0.6875, 0.6642156839370728, 0.655637264251709, 0.6948529481887817, 0.6740196347236633, 0.7242646813392639, 0.7071078419685364, 0.7064951062202454, 0.7561274766921997, 0.6924019455909729, 0.780637264251709, 0.7561274766921997, 0.7438725233078003, 0.7787989974021912, 0.7034313678741455, 0.6960784196853638, 0.8872548937797546, 0.7610294222831726, 0.843137264251709, 0.8921568393707275, 0.7585784196853638, 0.7549019455909729, 0.8780637383460999, 0.7745097875595093, 0.8566176295280457, 0.875, 0.8039215803146362, 0.8854166865348816, 0.8553921580314636, 0.8933823704719543, 0.811274528503418, 0.9019607901573181, 0.8774510025978088, 0.8590686321258545, 0.9424019455909729, 0.8995097875595093, 0.8713235259056091, 0.9325980544090271, 0.8958333134651184, 0.9264705777168274, 0.9877451062202454], [0.07352941483259201, 0.16299019753932953, 0.1611519604921341, 0.19117647409439087, 0.18504901230335236, 0.20098039507865906, 0.21997548639774323, 0.24142156541347504, 0.3174019753932953, 0.36764705181121826, 0.36274510622024536, 0.375612735748291, 0.280637264251709, 0.41299018263816833, 0.41421568393707275, 0.39889705181121826, 0.40441176295280457, 0.5055146813392639, 0.5343137383460999, NaN, 0.48713234066963196, 0.4828431308269501, 0.4944852888584137, 0.5502451062202454, 0.531862735748291, 0.5502451062202454, 0.5269607901573181, 0.5802696347236633, 0.4803921580314636, 0.6029411554336548, 0.5, 0.5674019455909729, 0.5833333134651184, 0.6041666865348816, 0.5465686321258545, 0.5735294222831726, 0.561274528503418, 0.720588207244873, 0.6446078419685364, 0.6593137383460999, 0.6433823704719543, 0.6740196347236633, 0.655637264251709, 0.6703431606292725, 0.6629902124404907, 0.7634803652763367, 0.6973039507865906, 0.6924019455909729, 0.7236519455909729, 0.7285539507865906, 0.7328431606292725, 0.7542892098426819, 0.7095588445663452, 0.6617646813392639, 0.6752451062202454, 0.8051470518112183, 0.7928921580314636, 0.751225471496582, 0.8002451062202454, 0.7561274766921997, 0.7604166865348816, 0.7696078419685364, 0.7769607901573181, 0.904411792755127, 0.8651960492134094, 0.8357843160629272, 0.8848039507865906, 0.8186274766921997, 0.875, 0.8100489974021912, 0.844362735748291, 0.8774510025978088, 0.8811274766921997, 0.8884803652763367, 0.8615196347236633, 0.8897058963775635, 0.9375, 0.9178921580314636, 0.908088207244873, 0.968137264251709], [0.09375, 0.21568627655506134, 0.22303920984268188, 0.2138480395078659, 0.24877451360225677, 0.2401960790157318, 0.2757352888584137, 0.2720588147640228, 0.40931373834609985, 0.34068626165390015, 0.4546568691730499, 0.4013480246067047, 0.4901960790157318, 0.42401960492134094, 0.46875, 0.47671568393707275, 0.4675244987010956, 0.5551470518112183, 0.5441176295280457, 0.5128676295280457, NaN, 0.4840686321258545, 0.46078431606292725, 0.5465686321258545, 0.5765931606292725, 0.5784313678741455, 0.5453431606292725, 0.5882353186607361, 0.5379902124404907, 0.6090686321258545, 0.5465686321258545, 0.5698529481887817, 0.6053921580314636, 0.6041666865348816, 0.5465686321258545, 0.5655637383460999, 0.5845588445663452, 0.5870097875595093, 0.6115196347236633, 0.6740196347236633, 0.6703431606292725, 0.6397058963775635, 0.6176470518112183, 0.6654411554336548, 0.688725471496582, 0.6642156839370728, 0.6985294222831726, 0.6985294222831726, 0.686887264251709, 0.7022058963775635, 0.7022058963775635, 0.6997548937797546, 0.7022058963775635, 0.6985294222831726, 0.6985294222831726, 0.7536764740943909, 0.7542892098426819, 0.7524510025978088, 0.7438725233078003, 0.7285539507865906, 0.7254902124404907, 0.7708333134651184, 0.7640931606292725, 0.781862735748291, 0.780637264251709, 0.7610294222831726, 0.8039215803146362, 0.8308823704719543, 0.7953431606292725, 0.7775735259056091, 0.813725471496582, 0.8394607901573181, 0.8566176295280457, 0.8602941036224365, 0.8370097875595093, 0.8713235259056091, 0.8713235259056091, 0.8909313678741455, 0.876225471496582, 0.968137264251709], [0.11397058516740799, 0.11764705926179886, 0.16421568393707275, 0.16299019753932953, 0.20098039507865906, 0.2034313678741455, 0.23590686917304993, 0.2291666716337204, 0.40441176295280457, 0.2922794222831726, 0.41850489377975464, 0.35600489377975464, 0.46568626165390015, 0.4117647111415863, 0.48161765933036804, 0.4424019753932953, 0.4644607901573181, 0.4742647111415863, 0.4828431308269501, 0.5171568393707275, 0.5159313678741455, NaN, 0.5061274766921997, 0.5116421580314636, 0.5196078419685364, 0.5208333134651184, 0.5428921580314636, 0.5441176295280457, 0.5281862616539001, 0.5367646813392639, 0.5465686321258545, 0.5453431606292725, 0.5710784196853638, 0.5833333134651184, 0.5821078419685364, 0.5796568393707275, 0.5159313678741455, 0.6176470518112183, 0.5980392098426819, 0.6017156839370728, 0.6225489974021912, 0.6666666865348816, 0.6666666865348816, 0.6776960492134094, 0.6323529481887817, 0.720588207244873, 0.658088207244873, 0.6850489974021912, 0.7671568393707275, 0.7046568393707275, 0.6789215803146362, 0.7254902124404907, 0.7071078419685364, 0.6813725233078003, 0.7083333134651184, 0.7389705777168274, 0.7928921580314636, 0.718137264251709, 0.7837010025978088, 0.7622548937797546, 0.7634803652763367, 0.7585784196853638, 0.875, 0.8529411554336548, 0.8541666865348816, 0.9031862616539001, 0.8321078419685364, 0.8345588445663452, 0.8725489974021912, 0.8345588445663452, 0.8615196347236633, 0.8946078419685364, 0.8259803652763367, 0.8897058963775635, 0.8823529481887817, 0.845588207244873, 0.9325980544090271, 0.9172794222831726, 0.9534313678741455, 0.9938725233078003], [0.10049019753932953, 0.17279411852359772, 0.2340686321258545, 0.20710784196853638, 0.27818626165390015, 0.3014705777168274, 0.32598039507865906, 0.33088234066963196, 0.4117647111415863, 0.2971813678741455, 0.4436274468898773, 0.2953431308269501, 0.4595588147640228, 0.3541666567325592, 0.4644607901573181, 0.4803921580314636, 0.48774510622024536, 0.48651960492134094, 0.5281862616539001, 0.5055146813392639, 0.5392156839370728, 0.4938725531101227, NaN, 0.5392156839370728, 0.5539215803146362, 0.5343137383460999, 0.5698529481887817, 0.5674019455909729, 0.5416666865348816, 0.5477941036224365, 0.5563725233078003, 0.5625, 0.5710784196853638, 0.5882353186607361, 0.5735294222831726, 0.5968137383460999, 0.53125, 0.6200980544090271, 0.5747548937797546, 0.6299019455909729, 0.5833333134651184, 0.6335784196853638, 0.6495097875595093, 0.6642156839370728, 0.6397058963775635, 0.6378676295280457, 0.6121323704719543, 0.7303921580314636, 0.6715686321258545, 0.6813725233078003, 0.656862735748291, 0.6911764740943909, 0.751225471496582, 0.6764705777168274, 0.7046568393707275, 0.718137264251709, 0.7316176295280457, 0.7536764740943909, 0.7598039507865906, 0.688725471496582, 0.6899510025978088, 0.7916666865348816, 0.811274528503418, 0.8161764740943909, 0.8259803652763367, 0.8848039507865906, 0.811274528503418, 0.8394607901573181, 0.8615196347236633, 0.7855392098426819, 0.8848039507865906, 0.876225471496582, 0.7781862616539001, 0.876225471496582, 0.875, 0.7990196347236633, 0.9215686321258545, 0.876225471496582, 0.906862735748291, 0.9926470518112183], [0.07230392098426819, 0.15931372344493866, 0.13296568393707275, 0.18504901230335236, 0.18504901230335236, 0.19117647409439087, 0.23529411852359772, 0.23529411852359772, 0.27450981736183167, 0.33578431606292725, 0.2953431308269501, 0.3529411852359772, 0.3811274468898773, 0.39399510622024536, 0.40318626165390015, 0.3952205777168274, 0.4270833432674408, 0.4479166567325592, 0.4050244987010956, 0.44975489377975464, 0.4534313678741455, 0.4883578419685364, 0.46078431606292725, NaN, 0.4950980246067047, 0.531862735748291, 0.5214460492134094, 0.6029411554336548, 0.4522058963775635, 0.6072303652763367, 0.4791666567325592, 0.5600489974021912, 0.563725471496582, 0.6409313678741455, 0.5698529481887817, 0.6017156839370728, 0.5674019455909729, 0.625612735748291, 0.59375, 0.6501225233078003, 0.655637264251709, 0.6495097875595093, 0.658088207244873, 0.656862735748291, 0.6991421580314636, 0.6948529481887817, 0.6936274766921997, 0.7365196347236633, 0.6740196347236633, 0.7659313678741455, 0.7549019455909729, 0.7310048937797546, 0.748774528503418, 0.6899510025978088, 0.6911764740943909, 0.8823529481887817, 0.751838207244873, 0.8284313678741455, 0.8664215803146362, 0.7585784196853638, 0.7573529481887817, 0.8602941036224365, 0.7696078419685364, 0.8406862616539001, 0.8651960492134094, 0.7965686321258545, 0.876225471496582, 0.8468137383460999, 0.8860294222831726, 0.8125, 0.8884803652763367, 0.8651960492134094, 0.8553921580314636, 0.9338235259056091, 0.8909313678741455, 0.8658088445663452, 0.936274528503418, 0.8909313678741455, 0.9301470518112183, 0.9865196347236633], [0.07598039507865906, 0.1360294073820114, 0.1090686246752739, 0.1568627506494522, 0.15808823704719543, 0.2077205926179886, 0.25, 0.27328431606292725, 0.2598039209842682, 0.2977941036224365, 0.29901960492134094, 0.3026960790157318, 0.39889705181121826, 0.3688725531101227, 0.40625, 0.3860294222831726, 0.3964460790157318, 0.47242647409439087, 0.47671568393707275, 0.468137264251709, 0.4234068691730499, 0.4803921580314636, 0.44607841968536377, 0.5049019455909729, NaN, 0.5453431606292725, 0.5496323704719543, 0.5680146813392639, 0.5, 0.5502451062202454, 0.5085784196853638, 0.5514705777168274, 0.5894607901573181, 0.6164215803146362, 0.5575980544090271, 0.6182597875595093, 0.5373774766921997, 0.5710784196853638, 0.6053921580314636, 0.6531862616539001, 0.6360294222831726, 0.6654411554336548, 0.7371323704719543, 0.7009803652763367, 0.6715686321258545, 0.7279411554336548, 0.6740196347236633, 0.7420343160629272, 0.7352941036224365, 0.71875, 0.7107843160629272, 0.7450980544090271, 0.7683823704719543, 0.688725471496582, 0.7022058963775635, 0.8308823704719543, 0.7861519455909729, 0.8272058963775635, 0.8700980544090271, 0.7218137383460999, 0.7224264740943909, 0.8553921580314636, 0.7689951062202454, 0.8688725233078003, 0.8541666865348816, 0.8517156839370728, 0.8823529481887817, 0.8946078419685364, 0.8884803652763367, 0.8063725233078003, 0.9093137383460999, 0.8970588445663452, 0.813725471496582, 0.9276960492134094, 0.8774510025978088, 0.8394607901573181, 0.9313725233078003, 0.8786764740943909, 0.9087010025978088, 0.9828431606292725], [0.07230392098426819, 0.13848039507865906, 0.12377450615167618, 0.1617647111415863, 0.14705882966518402, 0.17953431606292725, 0.20465686917304993, 0.21446079015731812, 0.24203431606292725, 0.3345588147640228, 0.2401960790157318, 0.3382352888584137, 0.3486519753932953, 0.37745097279548645, 0.3578431308269501, 0.38357841968536377, 0.38848039507865906, 0.3449755012989044, 0.4491421580314636, 0.44975489377975464, 0.4215686321258545, 0.4791666567325592, 0.46568626165390015, 0.468137264251709, 0.4546568691730499, NaN, 0.499387264251709, 0.560661792755127, 0.46078431606292725, 0.5900735259056091, 0.4803921580314636, 0.5024510025978088, 0.5196078419685364, 0.6041666865348816, 0.5183823704719543, 0.5502451062202454, 0.5453431606292725, 0.6397058963775635, 0.593137264251709, 0.6470588445663452, 0.6642156839370728, 0.6066176295280457, 0.6066176295280457, 0.6507353186607361, 0.6776960492134094, 0.6942402124404907, 0.7144607901573181, 0.7077205777168274, 0.6807597875595093, 0.7696078419685364, 0.7598039507865906, 0.688112735748291, 0.7322303652763367, 0.6875, 0.6899510025978088, 0.8609068393707275, 0.7414215803146362, 0.7769607901573181, 0.9160539507865906, 0.75, 0.751225471496582, 0.8284313678741455, 0.7904411554336548, 0.8284313678741455, 0.876838207244873, 0.8210784196853638, 0.9037989974021912, 0.8204656839370728, 0.8982843160629272, 0.813725471496582, 0.8933823704719543, 0.8676470518112183, 0.8615196347236633, 0.9252451062202454, 0.876225471496582, 0.8805146813392639, 0.9546568393707275, 0.9227941036224365, 0.9264705777168274, 0.9816176295280457], [0.07965686172246933, 0.14705882966518402, 0.125, 0.17524509131908417, 0.1764705926179886, 0.1825980395078659, 0.2089460790157318, 0.21568627655506134, 0.23284313082695007, 0.3063725531101227, 0.3082107901573181, 0.3008578419685364, 0.40318626165390015, 0.34558823704719543, 0.3799019753932953, 0.3541666567325592, 0.4178921580314636, 0.44607841968536377, 0.4595588147640228, 0.4730392098426819, 0.4546568691730499, 0.4571078419685364, 0.43014705181121826, 0.4785539209842682, 0.45036765933036804, 0.500612735748291, NaN, 0.5165441036224365, 0.469362735748291, 0.5349264740943909, 0.468137264251709, 0.5147058963775635, 0.5165441036224365, 0.5490196347236633, 0.5085784196853638, 0.6427696347236633, 0.5281862616539001, 0.5710784196853638, 0.5857843160629272, 0.6115196347236633, 0.6066176295280457, 0.591911792755127, 0.6207107901573181, 0.6225489974021912, 0.6299019455909729, 0.6397058963775635, 0.6605392098426819, 0.6838235259056091, 0.6636029481887817, 0.6593137383460999, 0.6850489974021912, 0.6764705777168274, 0.7230392098426819, 0.7083333134651184, 0.718137264251709, 0.7573529481887817, 0.7291666865348816, 0.7671568393707275, 0.7622548937797546, 0.7162989974021912, 0.7144607901573181, 0.7990196347236633, 0.7990196347236633, 0.8468137383460999, 0.8161764740943909, 0.8578431606292725, 0.8100489974021912, 0.9497548937797546, 0.8498774766921997, 0.781862735748291, 0.9350489974021912, 0.905637264251709, 0.8676470518112183, 0.9411764740943909, 0.90625, 0.8799019455909729, 0.9215686321258545, 0.9037989974021912, 0.9289215803146362, 0.9889705777168274], [0.06495098024606705, 0.1513480395078659, 0.11151960492134094, 0.17524509131908417, 0.1507352888584137, 0.1715686321258545, 0.20588235557079315, 0.22549019753932953, 0.20465686917304993, 0.3125, 0.23039215803146362, 0.3265931308269501, 0.3272058963775635, 0.35600489377975464, 0.3235294222831726, 0.36213234066963196, 0.3854166567325592, 0.31985294818878174, 0.37316176295280457, 0.41973039507865906, 0.4117647111415863, 0.45588234066963196, 0.4325980246067047, 0.3970588147640228, 0.4319852888584137, 0.43933823704719543, 0.4834558963775635, NaN, 0.4491421580314636, 0.5275735259056091, 0.468137264251709, 0.4779411852359772, 0.5183823704719543, 0.6047794222831726, 0.5147058963775635, 0.5686274766921997, 0.5343137383460999, 0.5857843160629272, 0.5735294222831726, 0.6029411554336548, 0.623774528503418, 0.6121323704719543, 0.6102941036224365, 0.6397058963775635, 0.6397058963775635, 0.6642156839370728, 0.6789215803146362, 0.6899510025978088, 0.6678921580314636, 0.7150735259056091, 0.7328431606292725, 0.6789215803146362, 0.7291666865348816, 0.6875, 0.6960784196853638, 0.8995097875595093, 0.720588207244873, 0.8180146813392639, 0.9105392098426819, 0.7377451062202454, 0.7340686321258545, 0.8412989974021912, 0.7689951062202454, 0.8186274766921997, 0.8321078419685364, 0.8045343160629272, 0.8541666865348816, 0.8272058963775635, 0.8609068393707275, 0.7916666865348816, 0.8786764740943909, 0.845588207244873, 0.8504902124404907, 0.9350489974021912, 0.8664215803146362, 0.8590686321258545, 0.9227941036224365, 0.8897058963775635, 0.9148284196853638, 0.9791666865348816], [0.1090686246752739, 0.15992647409439087, 0.20098039507865906, 0.2077205926179886, 0.20710784196853638, 0.15931372344493866, 0.19056372344493866, 0.23529411852359772, 0.3737744987010956, 0.3700980246067047, 0.4583333432674408, 0.3584558963775635, 0.4644607901573181, 0.4001225531101227, 0.4595588147640228, 0.3529411852359772, 0.44056373834609985, 0.5098039507865906, 0.5208333134651184, 0.5196078419685364, 0.46200981736183167, 0.47181373834609985, 0.4583333432674408, 0.5477941036224365, 0.5, 0.5392156839370728, 0.530637264251709, 0.5508578419685364, NaN, 0.561274528503418, 0.5343137383460999, 0.5588235259056091, 0.5649510025978088, 0.5551470518112183, 0.5079656839370728, 0.5502451062202454, 0.5422794222831726, 0.593137264251709, 0.626838207244873, 0.6053921580314636, 0.6335784196853638, 0.6164215803146362, 0.6072303652763367, 0.592524528503418, 0.626225471496582, 0.6464460492134094, 0.6629902124404907, 0.594362735748291, 0.6746323704719543, 0.654411792755127, 0.6789215803146362, 0.6727941036224365, 0.626225471496582, 0.592524528503418, 0.6096813678741455, 0.6740196347236633, 0.7144607901573181, 0.6311274766921997, 0.6825980544090271, 0.6924019455909729, 0.688725471496582, 0.6917892098426819, 0.7077205777168274, 0.7536764740943909, 0.7279411554336548, 0.7751225233078003, 0.781862735748291, 0.8051470518112183, 0.7432597875595093, 0.751225471496582, 0.8100489974021912, 0.811274528503418, 0.8308823704719543, 0.7990196347236633, 0.8039215803146362, 0.843137264251709, 0.8786764740943909, 0.8651960492134094, 0.8688725233078003, 0.9767156839370728], [0.07720588147640228, 0.1458333283662796, 0.12071078270673752, 0.1617647111415863, 0.14828431606292725, 0.1715686321258545, 0.21446079015731812, 0.20955882966518402, 0.2561274468898773, 0.3063725531101227, 0.25, 0.3137255012989044, 0.31617647409439087, 0.3578431308269501, 0.3578431308269501, 0.3578431308269501, 0.3578431308269501, 0.3290441036224365, 0.35171568393707275, 0.3970588147640228, 0.3909313678741455, 0.4632352888584137, 0.4522058963775635, 0.39276960492134094, 0.44975489377975464, 0.40992647409439087, 0.46507352590560913, 0.47242647409439087, 0.4387255012989044, NaN, 0.46017158031463623, 0.4779411852359772, 0.5036764740943909, 0.5128676295280457, 0.5122548937797546, 0.5575980544090271, 0.5196078419685364, 0.5723039507865906, 0.5379902124404907, 0.6280637383460999, 0.6286764740943909, 0.5992646813392639, 0.6115196347236633, 0.594362735748291, 0.6752451062202454, 0.656862735748291, 0.6838235259056091, 0.6875, 0.6654411554336548, 0.7757353186607361, 0.7328431606292725, 0.6924019455909729, 0.7046568393707275, 0.6678921580314636, 0.6691176295280457, 0.783088207244873, 0.7328431606292725, 0.7769607901573181, 0.8161764740943909, 0.750612735748291, 0.751225471496582, 0.8063725233078003, 0.75, 0.8125, 0.8480392098426819, 0.8002451062202454, 0.9093137383460999, 0.7941176295280457, 0.8799019455909729, 0.8149510025978088, 0.8664215803146362, 0.876225471496582, 0.8394607901573181, 0.9019607901573181, 0.8725489974021912, 0.8553921580314636, 0.9276960492134094, 0.905637264251709, 0.9325980544090271, 0.9840686321258545], [0.11519607901573181, 0.18075980246067047, 0.19117647409439087, 0.1985294073820114, 0.22732843458652496, 0.17524509131908417, 0.20955882966518402, 0.22181372344493866, 0.3590686321258545, 0.3541666567325592, 0.4362744987010956, 0.36151960492134094, 0.46017158031463623, 0.40257352590560913, 0.4313725531101227, 0.3535539209842682, 0.4264705777168274, 0.501225471496582, 0.5098039507865906, 0.5, 0.4534313678741455, 0.4534313678741455, 0.4436274468898773, 0.5208333134651184, 0.49142158031463623, 0.5196078419685364, 0.531862735748291, 0.531862735748291, 0.46568626165390015, 0.5398284196853638, NaN, 0.5171568393707275, 0.533088207244873, 0.5416666865348816, 0.48774510622024536, 0.5649510025978088, 0.5349264740943909, 0.5808823704719543, 0.6102941036224365, 0.5710784196853638, 0.6200980544090271, 0.5821078419685364, 0.5674019455909729, 0.5839460492134094, 0.5845588445663452, 0.6348039507865906, 0.6495097875595093, 0.6139705777168274, 0.6605392098426819, 0.591911792755127, 0.6666666865348816, 0.6372548937797546, 0.6286764740943909, 0.6041666865348816, 0.6176470518112183, 0.658088207244873, 0.7071078419685364, 0.6470588445663452, 0.6703431606292725, 0.7046568393707275, 0.7015931606292725, 0.7028186321258545, 0.7401960492134094, 0.7598039507865906, 0.7365196347236633, 0.8149510025978088, 0.7610294222831726, 0.8063725233078003, 0.748774528503418, 0.7598039507865906, 0.8051470518112183, 0.811274528503418, 0.8523284196853638, 0.813725471496582, 0.8149510025978088, 0.8578431606292725, 0.873774528503418, 0.8909313678741455, 0.8811274766921997, 0.9730392098426819], [0.057598039507865906, 0.12132352590560913, 0.11397058516740799, 0.1568627506494522, 0.1409313678741455, 0.19117647409439087, 0.22610294818878174, 0.26348039507865906, 0.2965686321258545, 0.31495097279548645, 0.3112744987010956, 0.3063725531101227, 0.37132352590560913, 0.3535539209842682, 0.3921568691730499, 0.3480392098426819, 0.3651960790157318, 0.4534313678741455, 0.42524510622024536, 0.4325980246067047, 0.43014705181121826, 0.4546568691730499, 0.4375, 0.43995097279548645, 0.4485294222831726, 0.49754902720451355, 0.4852941036224365, 0.5220588445663452, 0.44117647409439087, 0.5220588445663452, 0.4828431308269501, NaN, 0.6011029481887817, 0.5281862616539001, 0.49142158031463623, 0.5784313678741455, 0.5085784196853638, 0.5428921580314636, 0.5882353186607361, 0.5772058963775635, 0.5980392098426819, 0.5784313678741455, 0.5808823704719543, 0.7107843160629272, 0.6164215803146362, 0.6715686321258545, 0.6366421580314636, 0.6629902124404907, 0.6519607901573181, 0.6348039507865906, 0.654411792755127, 0.658088207244873, 0.6924019455909729, 0.6495097875595093, 0.658088207244873, 0.7622548937797546, 0.7107843160629272, 0.7303921580314636, 0.7426470518112183, 0.6789215803146362, 0.6764705777168274, 0.7414215803146362, 0.7132353186607361, 0.7769607901573181, 0.783088207244873, 0.7745097875595093, 0.7965686321258545, 0.8596813678741455, 0.7916666865348816, 0.7377451062202454, 0.843137264251709, 0.8210784196853638, 0.7671568393707275, 0.8553921580314636, 0.8235294222831726, 0.7781862616539001, 0.8578431606292725, 0.813725471496582, 0.8517156839370728, 0.9583333134651184], [0.06862745434045792, 0.1348039209842682, 0.10294117778539658, 0.15196079015731812, 0.13970588147640228, 0.14828431606292725, 0.20098039507865906, 0.21691176295280457, 0.2708333432674408, 0.2953431308269501, 0.30024510622024536, 0.30392158031463623, 0.3596813678741455, 0.3529411852359772, 0.3578431308269501, 0.3529411852359772, 0.35661765933036804, 0.4227941036224365, 0.41482841968536377, 0.4166666567325592, 0.3946078419685364, 0.42892158031463623, 0.42892158031463623, 0.4362744987010956, 0.4105392098426819, 0.4803921580314636, 0.4834558963775635, 0.48161765933036804, 0.43504902720451355, 0.49632352590560913, 0.46691176295280457, 0.39889705181121826, NaN, 0.49693626165390015, 0.49142158031463623, 0.5416666865348816, 0.5171568393707275, 0.5416666865348816, 0.5716911554336548, 0.5808823704719543, 0.5821078419685364, 0.5833333134651184, 0.5698529481887817, 0.6924019455909729, 0.59375, 0.6531862616539001, 0.6274510025978088, 0.6482843160629272, 0.6648284196853638, 0.6482843160629272, 0.654411792755127, 0.6997548937797546, 0.6691176295280457, 0.6593137383460999, 0.6776960492134094, 0.718137264251709, 0.7463235259056091, 0.7414215803146362, 0.7254902124404907, 0.7218137383460999, 0.7254902124404907, 0.7598039507865906, 0.7634803652763367, 0.8063725233078003, 0.7904411554336548, 0.8272058963775635, 0.8492646813392639, 0.8553921580314636, 0.8259803652763367, 0.7879902124404907, 0.8468137383460999, 0.8645833134651184, 0.8223039507865906, 0.8774510025978088, 0.8627451062202454, 0.844362735748291, 0.8958333134651184, 0.8884803652763367, 0.905637264251709, 0.9779411554336548], [0.05882352963089943, 0.1409313678741455, 0.10539215803146362, 0.17034313082695007, 0.12867647409439087, 0.15747548639774323, 0.19362744688987732, 0.19607843458652496, 0.18627451360225677, 0.2977941036224365, 0.23039215803146362, 0.2965686321258545, 0.28799018263816833, 0.34558823704719543, 0.2959558963775635, 0.3480392098426819, 0.3762255012989044, 0.28308823704719543, 0.3425244987010956, 0.3958333432674408, 0.3958333432674408, 0.4166666567325592, 0.4117647111415863, 0.3590686321258545, 0.38357841968536377, 0.3958333432674408, 0.45098039507865906, 0.3952205777168274, 0.44485294818878174, 0.48713234066963196, 0.4583333432674408, 0.47181373834609985, 0.5030637383460999, NaN, 0.5073529481887817, 0.532475471496582, 0.5147058963775635, 0.5514705777168274, 0.5575980544090271, 0.5484068393707275, 0.5980392098426819, 0.5857843160629272, 0.5747548937797546, 0.6078431606292725, 0.594362735748291, 0.6360294222831726, 0.656862735748291, 0.6832107901573181, 0.6433823704719543, 0.6727941036224365, 0.7022058963775635, 0.6593137383460999, 0.716911792755127, 0.6752451062202454, 0.6825980544090271, 0.8529411554336548, 0.7156862616539001, 0.7892156839370728, 0.8676470518112183, 0.7242646813392639, 0.7261029481887817, 0.8198529481887817, 0.751225471496582, 0.8063725233078003, 0.8051470518112183, 0.7928921580314636, 0.8480392098426819, 0.8161764740943909, 0.845588207244873, 0.7892156839370728, 0.876225471496582, 0.8357843160629272, 0.841911792755127, 0.9203431606292725, 0.8633578419685364, 0.8578431606292725, 0.9215686321258545, 0.8848039507865906, 0.9105392098426819, 0.9767156839370728], [0.12132352590560913, 0.2034313678741455, 0.22794117033481598, 0.2193627506494522, 0.2083333283662796, 0.21875, 0.24754901230335236, 0.23651960492134094, 0.3284313678741455, 0.33700981736183167, 0.38235294818878174, 0.3394607901573181, 0.3762255012989044, 0.3860294222831726, 0.3921568691730499, 0.4234068691730499, 0.4325980246067047, 0.4436274468898773, 0.41911765933036804, 0.4534313678741455, 0.4534313678741455, 0.4178921580314636, 0.4264705777168274, 0.43014705181121826, 0.4424019753932953, 0.48161765933036804, 0.49142158031463623, 0.4852941036224365, 0.49203431606292725, 0.48774510622024536, 0.5122548937797546, 0.5085784196853638, 0.5085784196853638, 0.49264705181121826, NaN, 0.4950980246067047, 0.5098039507865906, 0.5159313678741455, 0.5526960492134094, 0.5232843160629272, 0.5563725233078003, 0.5857843160629272, 0.530024528503418, 0.5833333134651184, 0.5441176295280457, 0.6225489974021912, 0.593137264251709, 0.625, 0.6703431606292725, 0.594362735748291, 0.6329656839370728, 0.626225471496582, 0.6433823704719543, 0.6004902124404907, 0.6127451062202454, 0.6470588445663452, 0.7267156839370728, 0.6770833134651184, 0.6801470518112183, 0.6507353186607361, 0.6458333134651184, 0.688112735748291, 0.6948529481887817, 0.7549019455909729, 0.7683823704719543, 0.7745097875595093, 0.7267156839370728, 0.7622548937797546, 0.7775735259056091, 0.7953431606292725, 0.7904411554336548, 0.8161764740943909, 0.8247548937797546, 0.7990196347236633, 0.7904411554336548, 0.8308823704719543, 0.8382353186607361, 0.8627451062202454, 0.8884803652763367, 0.9852941036224365], [0.08455882221460342, 0.1360294073820114, 0.1305147111415863, 0.17953431606292725, 0.1568627506494522, 0.17279411852359772, 0.18995098769664764, 0.18627451360225677, 0.22794117033481598, 0.26348039507865906, 0.26225489377975464, 0.27879902720451355, 0.375, 0.33088234066963196, 0.3278186321258545, 0.34007352590560913, 0.3792892098426819, 0.3848039209842682, 0.3799019753932953, 0.4264705777168274, 0.43443626165390015, 0.4203431308269501, 0.40318626165390015, 0.39828431606292725, 0.38174018263816833, 0.44975489377975464, 0.35723039507865906, 0.4313725531101227, 0.44975489377975464, 0.4424019753932953, 0.43504902720451355, 0.4215686321258545, 0.4583333432674408, 0.4675244987010956, 0.5049019455909729, NaN, 0.5171568393707275, 0.531862735748291, 0.5747548937797546, 0.5447303652763367, 0.563112735748291, 0.5698529481887817, 0.5661764740943909, 0.593137264251709, 0.5870097875595093, 0.5992646813392639, 0.6164215803146362, 0.6372548937797546, 0.6305146813392639, 0.623774528503418, 0.6397058963775635, 0.6458333134651184, 0.6850489974021912, 0.6495097875595093, 0.6495097875595093, 0.6764705777168274, 0.7058823704719543, 0.7022058963775635, 0.7046568393707275, 0.716911792755127, 0.719362735748291, 0.7438725233078003, 0.7745097875595093, 0.8382353186607361, 0.780024528503418, 0.8615196347236633, 0.813725471496582, 0.9019607901573181, 0.8204656839370728, 0.7904411554336548, 0.8688725233078003, 0.8860294222831726, 0.8627451062202454, 0.8909313678741455, 0.8933823704719543, 0.8799019455909729, 0.908088207244873, 0.9178921580314636, 0.9276960492134094, 0.9865196347236633], [0.04779411852359772, 0.1145833358168602, 0.1354166716337204, 0.12071078270673752, 0.19117647409439087, 0.21875, 0.24387255311012268, 0.2389705926179886, 0.3553921580314636, 0.3075980246067047, 0.3529411852359772, 0.36825981736183167, 0.3848039209842682, 0.40992647409439087, 0.4105392098426819, 0.4050244987010956, 0.3958333432674408, 0.4215686321258545, 0.4203431308269501, 0.4387255012989044, 0.4154411852359772, 0.4840686321258545, 0.46875, 0.4325980246067047, 0.4626225531101227, 0.4546568691730499, 0.47181373834609985, 0.46568626165390015, 0.4577205777168274, 0.4803921580314636, 0.46507352590560913, 0.49142158031463623, 0.4828431308269501, 0.4852941036224365, 0.4901960790157318, 0.4828431308269501, NaN, 0.5220588445663452, 0.5367646813392639, 0.5514705777168274, 0.5502451062202454, 0.5379902124404907, 0.5490196347236633, 0.5447303652763367, 0.5796568393707275, 0.5735294222831726, 0.5833333134651184, 0.6017156839370728, 0.6115196347236633, 0.6078431606292725, 0.6176470518112183, 0.623774528503418, 0.6145833134651184, 0.6936274766921997, 0.6911764740943909, 0.6372548937797546, 0.6531862616539001, 0.6041666865348816, 0.6783088445663452, 0.7034313678741455, 0.7009803652763367, 0.6838235259056091, 0.7365196347236633, 0.7144607901573181, 0.7389705777168274, 0.7524510025978088, 0.8125, 0.7365196347236633, 0.7573529481887817, 0.7745097875595093, 0.7879902124404907, 0.8180146813392639, 0.8357843160629272, 0.8259803652763367, 0.8700980544090271, 0.8480392098426819, 0.8566176295280457, 0.9283088445663452, 0.9215686321258545, 0.9730392098426819], [0.0674019604921341, 0.1299019604921341, 0.11642156541347504, 0.14338235557079315, 0.13174019753932953, 0.14705882966518402, 0.16911764442920685, 0.16911764442920685, 0.18504901230335236, 0.28308823704719543, 0.25367647409439087, 0.28982841968536377, 0.13909313082695007, 0.3296568691730499, 0.26838234066963196, 0.3492647111415863, 0.35171568393707275, 0.311887264251709, 0.36274510622024536, 0.27941176295280457, 0.41299018263816833, 0.38235294818878174, 0.3799019753932953, 0.374387264251709, 0.42892158031463623, 0.3602941036224365, 0.42892158031463623, 0.41421568393707275, 0.406862735748291, 0.4276960790157318, 0.41911765933036804, 0.4571078419685364, 0.4583333432674408, 0.4485294222831726, 0.4840686321258545, 0.468137264251709, 0.4779411852359772, NaN, 0.5, 0.4779411852359772, 0.5024510025978088, 0.5723039507865906, 0.5361519455909729, 0.5674019455909729, 0.5049019455909729, 0.6102941036224365, 0.5710784196853638, 0.6188725233078003, 0.6433823704719543, 0.5759803652763367, 0.5998774766921997, 0.658088207244873, 0.6397058963775635, 0.6115196347236633, 0.6139705777168274, 0.6715686321258545, 0.7242646813392639, 0.6875, 0.654411792755127, 0.7352941036224365, 0.7377451062202454, 0.6985294222831726, 0.7120097875595093, 0.7965686321258545, 0.7628676295280457, 0.7634803652763367, 0.8014705777168274, 0.7291666865348816, 0.7745097875595093, 0.8125, 0.7916666865348816, 0.8247548937797546, 0.8272058963775635, 0.8321078419685364, 0.8186274766921997, 0.8388480544090271, 0.8823529481887817, 0.8823529481887817, 0.9007353186607361, 0.9522058963775635], [0.0888480395078659, 0.16973039507865906, 0.1617647111415863, 0.1764705926179886, 0.18014705181121826, 0.1666666716337204, 0.1875, 0.1868872493505478, 0.26838234066963196, 0.3137255012989044, 0.2928921580314636, 0.33700981736183167, 0.27941176295280457, 0.37867647409439087, 0.31004902720451355, 0.3174019753932953, 0.3333333432674408, 0.39338234066963196, 0.38357841968536377, 0.3553921580314636, 0.38848039507865906, 0.4019607901573181, 0.42524510622024536, 0.40625, 0.3946078419685364, 0.406862735748291, 0.41421568393707275, 0.4264705777168274, 0.37316176295280457, 0.46200981736183167, 0.3897058963775635, 0.4117647111415863, 0.4283088147640228, 0.4424019753932953, 0.4473039209842682, 0.42524510622024536, 0.4632352888584137, 0.5, NaN, 0.5042892098426819, 0.5674019455909729, 0.5245097875595093, 0.469362735748291, 0.4950980246067047, 0.5281862616539001, 0.5661764740943909, 0.6115196347236633, 0.5367646813392639, 0.5851715803146362, 0.5759803652763367, 0.6531862616539001, 0.5992646813392639, 0.5526960492134094, 0.6311274766921997, 0.6286764740943909, 0.5870097875595093, 0.6482843160629272, 0.5575980544090271, 0.6023284196853638, 0.7303921580314636, 0.7291666865348816, 0.625, 0.6740196347236633, 0.7009803652763367, 0.7034313678741455, 0.7359068393707275, 0.8210784196853638, 0.6740196347236633, 0.7267156839370728, 0.8008578419685364, 0.719362735748291, 0.8002451062202454, 0.8860294222831726, 0.7634803652763367, 0.8602941036224365, 0.8995097875595093, 0.8333333134651184, 0.9473039507865906, 0.9338235259056091, 0.969362735748291], [0.06495098024606705, 0.11397058516740799, 0.0833333358168602, 0.14338235557079315, 0.10539215803146362, 0.1513480395078659, 0.19056372344493866, 0.18382352590560913, 0.23039215803146362, 0.2549019753932953, 0.2028186321258545, 0.2696078419685364, 0.27818626165390015, 0.30392158031463623, 0.3333333432674408, 0.3174019753932953, 0.31004902720451355, 0.27757352590560913, 0.3125, 0.34068626165390015, 0.32598039507865906, 0.39828431606292725, 0.3700980246067047, 0.3498774468898773, 0.34681373834609985, 0.3529411852359772, 0.38848039507865906, 0.3970588147640228, 0.3946078419685364, 0.37193626165390015, 0.42892158031463623, 0.4227941036224365, 0.41911765933036804, 0.4515931308269501, 0.47671568393707275, 0.45526960492134094, 0.4485294222831726, 0.5220588445663452, 0.4957107901573181, NaN, 0.5539215803146362, 0.5281862616539001, 0.5159313678741455, 0.47181373834609985, 0.6072303652763367, 0.5698529481887817, 0.595588207244873, 0.5747548937797546, 0.561274528503418, 0.7965686321258545, 0.6311274766921997, 0.591911792755127, 0.6102941036224365, 0.591911792755127, 0.595588207244873, 0.6776960492134094, 0.6084558963775635, 0.6776960492134094, 0.7322303652763367, 0.7095588445663452, 0.7046568393707275, 0.7156862616539001, 0.7083333134651184, 0.716911792755127, 0.7867646813392639, 0.7867646813392639, 0.8174019455909729, 0.7426470518112183, 0.8149510025978088, 0.7855392098426819, 0.7879902124404907, 0.8161764740943909, 0.7855392098426819, 0.8210784196853638, 0.7953431606292725, 0.8002451062202454, 0.8700980544090271, 0.8700980544090271, 0.8860294222831726, 0.9736519455909729], [0.07352941483259201, 0.1409313678741455, 0.11029411852359772, 0.15441176295280457, 0.13848039507865906, 0.14338235557079315, 0.16299019753932953, 0.18014705181121826, 0.20220588147640228, 0.28799018263816833, 0.22671568393707275, 0.2922794222831726, 0.2671568691730499, 0.3333333432674408, 0.27941176295280457, 0.30453431606292725, 0.28431373834609985, 0.2959558963775635, 0.33578431606292725, 0.35661765933036804, 0.3296568691730499, 0.37745097279548645, 0.4166666567325592, 0.344362735748291, 0.3639705777168274, 0.33578431606292725, 0.39338234066963196, 0.3762255012989044, 0.36642158031463623, 0.37132352590560913, 0.3799019753932953, 0.4019607901573181, 0.4178921580314636, 0.4019607901573181, 0.4436274468898773, 0.436887264251709, 0.44975489377975464, 0.49754902720451355, 0.4325980246067047, 0.44607841968536377, NaN, 0.5257353186607361, 0.4828431308269501, 0.5257353186607361, 0.48651960492134094, 0.5625, 0.7212010025978088, 0.6023284196853638, 0.5508578419685364, 0.5600489974021912, 0.7683823704719543, 0.6158088445663452, 0.6127451062202454, 0.6188725233078003, 0.6286764740943909, 0.6415441036224365, 0.623161792755127, 0.658088207244873, 0.6360294222831726, 0.7230392098426819, 0.7144607901573181, 0.6838235259056091, 0.6838235259056091, 0.7463235259056091, 0.7549019455909729, 0.7463235259056091, 0.8026960492134094, 0.7083333134651184, 0.7726715803146362, 0.8100489974021912, 0.7585784196853638, 0.8321078419685364, 0.8615196347236633, 0.7953431606292725, 0.8296568393707275, 0.875, 0.8848039507865906, 0.9007353186607361, 0.8878676295280457, 0.9522058963775635], [0.040441177785396576, 0.08946078270673752, 0.07475490123033524, 0.1200980395078659, 0.1041666641831398, 0.11274509876966476, 0.1360294073820114, 0.1360294073820114, 0.2181372493505478, 0.2242647111415863, 0.18137255311012268, 0.250612735748291, 0.2659313678741455, 0.2904411852359772, 0.2659313678741455, 0.3223039209842682, 0.280637264251709, 0.3602941036224365, 0.344362735748291, 0.32598039507865906, 0.3602941036224365, 0.3333333432674408, 0.36642158031463623, 0.35049018263816833, 0.3345588147640228, 0.39338234066963196, 0.40808823704719543, 0.38786765933036804, 0.38357841968536377, 0.4007352888584137, 0.4178921580314636, 0.4215686321258545, 0.4166666567325592, 0.41421568393707275, 0.41421568393707275, 0.43014705181121826, 0.46200981736183167, 0.4276960790157318, 0.47549018263816833, 0.47181373834609985, 0.4742647111415863, NaN, 0.5067402124404907, 0.5441176295280457, 0.5085784196853638, 0.5551470518112183, 0.5392156839370728, 0.6323529481887817, 0.6384803652763367, 0.5281862616539001, 0.5796568393707275, 0.748774528503418, 0.6470588445663452, 0.591911792755127, 0.593137264251709, 0.6531862616539001, 0.7432597875595093, 0.6875, 0.6482843160629272, 0.7034313678741455, 0.7101715803146362, 0.7120097875595093, 0.6727941036224365, 0.8921568393707275, 0.7218137383460999, 0.779411792755127, 0.8069853186607361, 0.7683823704719543, 0.7708333134651184, 0.8002451062202454, 0.7647058963775635, 0.8584558963775635, 0.811887264251709, 0.7965686321258545, 0.8198529481887817, 0.8370097875595093, 0.8823529481887817, 0.8982843160629272, 0.8933823704719543, 0.9595588445663452], [0.057598039507865906, 0.07414215803146362, 0.08455882221460342, 0.125, 0.09313725680112839, 0.12622548639774323, 0.1458333283662796, 0.14154411852359772, 0.17401960492134094, 0.20159313082695007, 0.1924019604921341, 0.21078431606292725, 0.27328431606292725, 0.2659313678741455, 0.2696078419685364, 0.30882352590560913, 0.3235294222831726, 0.3284313678741455, 0.30514705181121826, 0.344362735748291, 0.38235294818878174, 0.3333333432674408, 0.35049018263816833, 0.34191176295280457, 0.26286765933036804, 0.39338234066963196, 0.3792892098426819, 0.3897058963775635, 0.39276960492134094, 0.38848039507865906, 0.4325980246067047, 0.41911765933036804, 0.43014705181121826, 0.42524510622024536, 0.4699755012989044, 0.43382352590560913, 0.45098039507865906, 0.4638480246067047, 0.530637264251709, 0.4840686321258545, 0.5171568393707275, 0.49325981736183167, NaN, 0.560661792755127, 0.5349264740943909, 0.5710784196853638, 0.5698529481887817, 0.594362735748291, 0.6188725233078003, 0.5747548937797546, 0.6041666865348816, 0.6041666865348816, 0.6446078419685364, 0.5980392098426819, 0.5968137383460999, 0.6666666865348816, 0.6703431606292725, 0.6960784196853638, 0.6924019455909729, 0.6917892098426819, 0.688725471496582, 0.7291666865348816, 0.6924019455909729, 0.8100489974021912, 0.7230392098426819, 0.7763480544090271, 0.810661792755127, 0.811274528503418, 0.7708333134651184, 0.7598039507865906, 0.8161764740943909, 0.8468137383460999, 0.7616421580314636, 0.8468137383460999, 0.8149510025978088, 0.7916666865348816, 0.8651960492134094, 0.8492646813392639, 0.873161792755127, 0.9620097875595093], [0.061274509876966476, 0.09497549384832382, 0.1041666641831398, 0.11580882221460342, 0.10171568393707275, 0.13725490868091583, 0.16053920984268188, 0.1556372493505478, 0.2034313678741455, 0.20098039507865906, 0.21446079015731812, 0.22120098769664764, 0.27328431606292725, 0.2757352888584137, 0.25857841968536377, 0.3376225531101227, 0.31495097279548645, 0.3075980246067047, 0.32598039507865906, 0.3296568691730499, 0.3345588147640228, 0.3223039209842682, 0.33578431606292725, 0.343137264251709, 0.29901960492134094, 0.3492647111415863, 0.37745097279548645, 0.3602941036224365, 0.4074755012989044, 0.405637264251709, 0.4160539209842682, 0.28921568393707275, 0.3075980246067047, 0.3921568691730499, 0.4166666567325592, 0.406862735748291, 0.45526960492134094, 0.4325980246067047, 0.5049019455909729, 0.5281862616539001, 0.4742647111415863, 0.45588234066963196, 0.43933823704719543, NaN, 0.5441176295280457, 0.49264705181121826, 0.5196078419685364, 0.5686274766921997, 0.5477941036224365, 0.5857843160629272, 0.563725471496582, 0.5416666865348816, 0.6041666865348816, 0.6225489974021912, 0.6323529481887817, 0.5882353186607361, 0.6200980544090271, 0.6617646813392639, 0.623774528503418, 0.6789215803146362, 0.6838235259056091, 0.7218137383460999, 0.7524510025978088, 0.7732843160629272, 0.7242646813392639, 0.8100489974021912, 0.7928921580314636, 0.7647058963775635, 0.7769607901573181, 0.779411792755127, 0.813112735748291, 0.8247548937797546, 0.7977941036224365, 0.8578431606292725, 0.8651960492134094, 0.8180146813392639, 0.8639705777168274, 0.8982843160629272, 0.9093137383460999, 0.9803921580314636], [0.06004901975393295, 0.11397058516740799, 0.07598039507865906, 0.13357843458652496, 0.09681372344493866, 0.13357843458652496, 0.16544117033481598, 0.17279411852359772, 0.2089460790157318, 0.23284313082695007, 0.18504901230335236, 0.25, 0.24142156541347504, 0.2904411852359772, 0.2965686321258545, 0.3063725531101227, 0.29901960492134094, 0.23651960492134094, 0.2757352888584137, 0.33700981736183167, 0.3112744987010956, 0.36764705181121826, 0.3602941036224365, 0.3008578419685364, 0.3284313678741455, 0.3223039209842682, 0.3700980246067047, 0.3602941036224365, 0.3737744987010956, 0.32475489377975464, 0.4154411852359772, 0.38357841968536377, 0.40625, 0.405637264251709, 0.45588234066963196, 0.41299018263816833, 0.4203431308269501, 0.4950980246067047, 0.47181373834609985, 0.39276960492134094, 0.5134803652763367, 0.49142158031463623, 0.46507352590560913, 0.45588234066963196, NaN, 0.5526960492134094, 0.5667892098426819, 0.5526960492134094, 0.5379902124404907, 0.7156862616539001, 0.595588207244873, 0.5625, 0.593137264251709, 0.5723039507865906, 0.5772058963775635, 0.6501225233078003, 0.5876225233078003, 0.6495097875595093, 0.686274528503418, 0.6973039507865906, 0.6960784196853638, 0.6899510025978088, 0.686274528503418, 0.7071078419685364, 0.7469362616539001, 0.751225471496582, 0.7916666865348816, 0.7218137383460999, 0.7763480544090271, 0.7708333134651184, 0.7769607901573181, 0.7941176295280457, 0.7732843160629272, 0.7904411554336548, 0.7769607901573181, 0.779411792755127, 0.8633578419685364, 0.8566176295280457, 0.8639705777168274, 0.9620097875595093], [0.05269607901573181, 0.07475490123033524, 0.0882352963089943, 0.09803921729326248, 0.09068627655506134, 0.11764705926179886, 0.13357843458652496, 0.13970588147640228, 0.0986519604921341, 0.19485294818878174, 0.20220588147640228, 0.22181372344493866, 0.1770833283662796, 0.2659313678741455, 0.1973039209842682, 0.28799018263816833, 0.3063725531101227, 0.2769607901573181, 0.2928921580314636, 0.23651960492134094, 0.33578431606292725, 0.27941176295280457, 0.36213234066963196, 0.30514705181121826, 0.2720588147640228, 0.30575981736183167, 0.3602941036224365, 0.33578431606292725, 0.3535539209842682, 0.343137264251709, 0.3651960790157318, 0.3284313678741455, 0.34681373834609985, 0.3639705777168274, 0.37745097279548645, 0.4007352888584137, 0.4264705777168274, 0.3897058963775635, 0.43382352590560913, 0.43014705181121826, 0.4375, 0.44485294818878174, 0.42892158031463623, 0.5073529481887817, 0.4473039209842682, NaN, 0.5049019455909729, 0.5735294222831726, 0.5422794222831726, 0.5110294222831726, 0.5343137383460999, 0.5833333134651184, 0.6066176295280457, 0.5882353186607361, 0.595588207244873, 0.6139705777168274, 0.6531862616539001, 0.6531862616539001, 0.6102941036224365, 0.6740196347236633, 0.6813725233078003, 0.6875, 0.6593137383460999, 0.7763480544090271, 0.7156862616539001, 0.7861519455909729, 0.7720588445663452, 0.720588207244873, 0.7720588445663452, 0.7781862616539001, 0.7953431606292725, 0.8725489974021912, 0.7953431606292725, 0.8014705777168274, 0.8284313678741455, 0.8259803652763367, 0.876225471496582, 0.8872548937797546, 0.906862735748291, 0.9791666865348816], [0.0625, 0.12377450615167618, 0.10171568393707275, 0.1360294073820114, 0.10845588147640228, 0.12193627655506134, 0.14705882966518402, 0.15196079015731812, 0.1825980395078659, 0.25306373834609985, 0.20710784196853638, 0.26776960492134094, 0.22303920984268188, 0.29411765933036804, 0.24142156541347504, 0.280637264251709, 0.2549019753932953, 0.25245097279548645, 0.29350489377975464, 0.3026960790157318, 0.3014705777168274, 0.34191176295280457, 0.38786765933036804, 0.3063725531101227, 0.32598039507865906, 0.2855392098426819, 0.3394607901573181, 0.32107841968536377, 0.33700981736183167, 0.31617647409439087, 0.35049018263816833, 0.3633578419685364, 0.37254902720451355, 0.343137264251709, 0.406862735748291, 0.38357841968536377, 0.4166666567325592, 0.42892158031463623, 0.38848039507865906, 0.40441176295280457, 0.27879902720451355, 0.46078431606292725, 0.43014705181121826, 0.4803921580314636, 0.4332107901573181, 0.4950980246067047, NaN, 0.5477941036224365, 0.5024510025978088, 0.5036764740943909, 0.6623774766921997, 0.5674019455909729, 0.5821078419685364, 0.6017156839370728, 0.6164215803146362, 0.591911792755127, 0.594975471496582, 0.6115196347236633, 0.5845588445663452, 0.7034313678741455, 0.7003676295280457, 0.6519607901573181, 0.6452205777168274, 0.716911792755127, 0.7218137383460999, 0.71875, 0.7598039507865906, 0.686274528503418, 0.7463235259056091, 0.7953431606292725, 0.7426470518112183, 0.7990196347236633, 0.8345588445663452, 0.7683823704719543, 0.8026960492134094, 0.8504902124404907, 0.8639705777168274, 0.8823529481887817, 0.8860294222831726, 0.9460784196853638], [0.046568628400564194, 0.06495098024606705, 0.07598039507865906, 0.10232843458652496, 0.09191176295280457, 0.12254901975393295, 0.13296568393707275, 0.1299019604921341, 0.15196079015731812, 0.1513480395078659, 0.21691176295280457, 0.1825980395078659, 0.25735294818878174, 0.22671568393707275, 0.23774509131908417, 0.3082107901573181, 0.32107841968536377, 0.2444852888584137, 0.24387255311012268, 0.3075980246067047, 0.3014705777168274, 0.31495097279548645, 0.2696078419685364, 0.26348039507865906, 0.25796568393707275, 0.2922794222831726, 0.31617647409439087, 0.31004902720451355, 0.405637264251709, 0.3125, 0.3860294222831726, 0.33700981736183167, 0.35171568393707275, 0.3167892098426819, 0.375, 0.36274510622024536, 0.39828431606292725, 0.3811274468898773, 0.4632352888584137, 0.42524510622024536, 0.39767158031463623, 0.36764705181121826, 0.405637264251709, 0.4313725531101227, 0.4473039209842682, 0.4264705777168274, 0.4522058963775635, NaN, 0.49142158031463623, 0.49754902720451355, 0.4803921580314636, 0.44424018263816833, 0.6807597875595093, 0.6458333134651184, 0.6482843160629272, 0.5196078419685364, 0.5392156839370728, 0.6911764740943909, 0.5379902124404907, 0.595588207244873, 0.6053921580314636, 0.7438725233078003, 0.7022058963775635, 0.6127451062202454, 0.6507353186607361, 0.7450980544090271, 0.6985294222831726, 0.7132353186607361, 0.6838235259056091, 0.7322303652763367, 0.9191176295280457, 0.75, 0.7420343160629272, 0.8860294222831726, 0.8688725233078003, 0.7579656839370728, 0.780637264251709, 0.8308823704719543, 0.9129902124404907, 0.9914215803146362], [0.07352941483259201, 0.11580882221460342, 0.10171568393707275, 0.1299019604921341, 0.12745098769664764, 0.13174019753932953, 0.14950980246067047, 0.12622548639774323, 0.15257352590560913, 0.18627451360225677, 0.17524509131908417, 0.22181372344493866, 0.2242647111415863, 0.26348039507865906, 0.22671568393707275, 0.2591911852359772, 0.24264705181121826, 0.2855392098426819, 0.3075980246067047, 0.2763480246067047, 0.313112735748291, 0.23284313082695007, 0.3284313678741455, 0.32598039507865906, 0.2647058963775635, 0.31924018263816833, 0.33639705181121826, 0.3321078419685364, 0.32536765933036804, 0.3345588147640228, 0.3394607901573181, 0.3480392098426819, 0.33517158031463623, 0.35661765933036804, 0.3296568691730499, 0.3694852888584137, 0.38848039507865906, 0.35661765933036804, 0.41482841968536377, 0.4387255012989044, 0.4491421580314636, 0.36151960492134094, 0.3811274468898773, 0.4522058963775635, 0.46200981736183167, 0.4577205777168274, 0.49754902720451355, 0.5085784196853638, NaN, 0.4852941036224365, 0.5171568393707275, 0.4901960790157318, 0.5257353186607361, 0.5759803652763367, 0.5784313678741455, 0.5404411554336548, 0.6850489974021912, 0.6047794222831726, 0.5337010025978088, 0.6311274766921997, 0.6348039507865906, 0.6225489974021912, 0.5833333134651184, 0.6942402124404907, 0.6384803652763367, 0.7242646813392639, 0.7328431606292725, 0.7034313678741455, 0.6727941036224365, 0.7426470518112183, 0.7450980544090271, 0.9509803652763367, 0.8272058963775635, 0.7303921580314636, 0.7855392098426819, 0.8480392098426819, 0.8186274766921997, 0.8970588445663452, 0.8602941036224365, 0.9546568393707275], [0.04534313827753067, 0.1041666641831398, 0.06985294073820114, 0.12377450615167618, 0.08946078270673752, 0.11274509876966476, 0.1360294073820114, 0.13357843458652496, 0.18504901230335236, 0.2083333283662796, 0.14950980246067047, 0.21200980246067047, 0.1924019604921341, 0.25245097279548645, 0.25, 0.27941176295280457, 0.27941176295280457, 0.16789215803146362, 0.2193627506494522, 0.2714460790157318, 0.2977941036224365, 0.2953431308269501, 0.3186274468898773, 0.2340686321258545, 0.28125, 0.23039215803146362, 0.34068626165390015, 0.28492647409439087, 0.34558823704719543, 0.2242647111415863, 0.40808823704719543, 0.3651960790157318, 0.35171568393707275, 0.3272058963775635, 0.405637264251709, 0.3762255012989044, 0.3921568691730499, 0.42401960492134094, 0.42401960492134094, 0.2034313678741455, 0.43995097279548645, 0.47181373834609985, 0.42524510622024536, 0.41421568393707275, 0.28431373834609985, 0.4889705777168274, 0.49632352590560913, 0.5024510025978088, 0.5147058963775635, NaN, 0.5367646813392639, 0.531862735748291, 0.5232843160629272, 0.5367646813392639, 0.5441176295280457, 0.5863970518112183, 0.5526960492134094, 0.6127451062202454, 0.5912989974021912, 0.6727941036224365, 0.6776960492134094, 0.6286764740943909, 0.6433823704719543, 0.656862735748291, 0.6924019455909729, 0.6948529481887817, 0.7561274766921997, 0.6776960492134094, 0.7279411554336548, 0.7426470518112183, 0.6850489974021912, 0.7475489974021912, 0.7659313678741455, 0.7573529481887817, 0.7334558963775635, 0.7757353186607361, 0.8198529481887817, 0.8553921580314636, 0.8639705777168274, 0.9607843160629272], [0.06495098024606705, 0.10355392098426819, 0.10049019753932953, 0.11029411852359772, 0.11274509876966476, 0.1200980395078659, 0.1348039209842682, 0.1360294073820114, 0.1611519604921341, 0.20955882966518402, 0.1715686321258545, 0.2512255012989044, 0.2138480395078659, 0.2757352888584137, 0.24142156541347504, 0.2549019753932953, 0.24632352590560913, 0.1825980395078659, 0.24387255311012268, 0.2671568691730499, 0.2977941036224365, 0.32107841968536377, 0.343137264251709, 0.2450980395078659, 0.28921568393707275, 0.2401960790157318, 0.31495097279548645, 0.2671568691730499, 0.32107841968536377, 0.2671568691730499, 0.3333333432674408, 0.34558823704719543, 0.34558823704719543, 0.2977941036224365, 0.36703431606292725, 0.3602941036224365, 0.38235294818878174, 0.4001225531101227, 0.34681373834609985, 0.3688725531101227, 0.23161764442920685, 0.4203431308269501, 0.3958333432674408, 0.4362744987010956, 0.40441176295280457, 0.46568626165390015, 0.3376225531101227, 0.5196078419685364, 0.4828431308269501, 0.4632352888584137, NaN, 0.5379902124404907, 0.5379902124404907, 0.5845588445663452, 0.5968137383460999, 0.5281862616539001, 0.5686274766921997, 0.5772058963775635, 0.5428921580314636, 0.6727941036224365, 0.6703431606292725, 0.6090686321258545, 0.6311274766921997, 0.7058823704719543, 0.6801470518112183, 0.7107843160629272, 0.7561274766921997, 0.6433823704719543, 0.7071078419685364, 0.7781862616539001, 0.7279411554336548, 0.7879902124404907, 0.8198529481887817, 0.7414215803146362, 0.8186274766921997, 0.844975471496582, 0.8382353186607361, 0.8995097875595093, 0.8897058963775635, 0.9485294222831726], [0.030637254938483238, 0.061274509876966476, 0.05514705926179886, 0.09191176295280457, 0.06495098024606705, 0.07965686172246933, 0.09558823704719543, 0.09558823704719543, 0.16789215803146362, 0.1715686321258545, 0.14215686917304993, 0.18995098769664764, 0.18566176295280457, 0.23651960492134094, 0.19485294818878174, 0.2671568691730499, 0.2291666716337204, 0.2659313678741455, 0.2561274468898773, 0.24571079015731812, 0.30024510622024536, 0.27450981736183167, 0.30882352590560913, 0.26899510622024536, 0.2549019753932953, 0.311887264251709, 0.3235294222831726, 0.32107841968536377, 0.3272058963775635, 0.3075980246067047, 0.36274510622024536, 0.34191176295280457, 0.30024510622024536, 0.34068626165390015, 0.3737744987010956, 0.3541666567325592, 0.3762255012989044, 0.34191176295280457, 0.4007352888584137, 0.40808823704719543, 0.3841911852359772, 0.2512255012989044, 0.3958333432674408, 0.4583333432674408, 0.4375, 0.4166666567325592, 0.4325980246067047, 0.5557597875595093, 0.5098039507865906, 0.468137264251709, 0.46200981736183167, NaN, 0.5747548937797546, 0.5539215803146362, 0.5526960492134094, 0.5367646813392639, 0.6066176295280457, 0.6078431606292725, 0.5545343160629272, 0.6850489974021912, 0.686274528503418, 0.6372548937797546, 0.6102941036224365, 0.8082107901573181, 0.6531862616539001, 0.7463235259056091, 0.7438725233078003, 0.7083333134651184, 0.7009803652763367, 0.7781862616539001, 0.718137264251709, 0.781862735748291, 0.7426470518112183, 0.7450980544090271, 0.7775735259056091, 0.7604166865348816, 0.8235294222831726, 0.8700980544090271, 0.8566176295280457, 0.9473039507865906], [0.05147058889269829, 0.06372549384832382, 0.0833333358168602, 0.10049019753932953, 0.09497549384832382, 0.1200980395078659, 0.125, 0.11764705926179886, 0.14276960492134094, 0.12867647409439087, 0.19362744688987732, 0.16299019753932953, 0.2512255012989044, 0.2083333283662796, 0.21568627655506134, 0.2953431308269501, 0.3082107901573181, 0.24142156541347504, 0.22120098769664764, 0.2904411852359772, 0.2977941036224365, 0.2928921580314636, 0.24877451360225677, 0.2512255012989044, 0.23161764442920685, 0.26776960492134094, 0.2769607901573181, 0.2708333432674408, 0.3737744987010956, 0.2953431308269501, 0.37132352590560913, 0.3075980246067047, 0.33088234066963196, 0.28308823704719543, 0.35661765933036804, 0.31495097279548645, 0.3854166567325592, 0.3602941036224365, 0.4473039209842682, 0.3897058963775635, 0.38725489377975464, 0.3529411852359772, 0.3553921580314636, 0.3958333432674408, 0.406862735748291, 0.39338234066963196, 0.4178921580314636, 0.31924018263816833, 0.4742647111415863, 0.47671568393707275, 0.46200981736183167, 0.42524510622024536, NaN, 0.6158088445663452, 0.6145833134651184, 0.4644607901573181, 0.5220588445663452, 0.593137264251709, 0.4950980246067047, 0.5962010025978088, 0.6017156839370728, 0.6611519455909729, 0.6899510025978088, 0.5857843160629272, 0.6286764740943909, 0.7389705777168274, 0.6764705777168274, 0.6776960492134094, 0.6642156839370728, 0.7279411554336548, 0.8504902124404907, 0.7291666865348816, 0.7279411554336548, 0.8486519455909729, 0.8382353186607361, 0.7377451062202454, 0.7683823704719543, 0.8235294222831726, 0.9178921580314636, 0.9901960492134094], [0.11887254565954208, 0.14338235557079315, 0.20465686917304993, 0.17279411852359772, 0.20588235557079315, 0.20098039507865906, 0.2242647111415863, 0.24632352590560913, 0.2291666716337204, 0.2450980395078659, 0.27818626165390015, 0.21568627655506134, 0.3137255012989044, 0.26225489377975464, 0.2928921580314636, 0.34375, 0.35171568393707275, 0.2928921580314636, 0.2965686321258545, 0.3382352888584137, 0.3014705777168274, 0.3186274468898773, 0.3235294222831726, 0.31004902720451355, 0.3112744987010956, 0.3125, 0.2916666567325592, 0.3125, 0.4074755012989044, 0.3321078419685364, 0.3958333432674408, 0.35049018263816833, 0.34068626165390015, 0.32475489377975464, 0.39950981736183167, 0.35049018263816833, 0.3063725531101227, 0.38848039507865906, 0.3688725531101227, 0.40808823704719543, 0.3811274468898773, 0.40808823704719543, 0.4019607901573181, 0.37745097279548645, 0.4276960790157318, 0.4117647111415863, 0.39828431606292725, 0.3541666567325592, 0.42401960492134094, 0.4632352888584137, 0.4154411852359772, 0.44607841968536377, 0.3841911852359772, NaN, 0.5625, 0.4117647111415863, 0.4632352888584137, 0.36764705181121826, 0.46507352590560913, 0.4791666567325592, 0.4791666567325592, 0.44117647409439087, 0.5196078419685364, 0.533088207244873, 0.5747548937797546, 0.5808823704719543, 0.49754902720451355, 0.6409313678741455, 0.595588207244873, 0.6299019455909729, 0.6188725233078003, 0.6096813678741455, 0.6053921580314636, 0.6323529481887817, 0.6384803652763367, 0.6274510025978088, 0.6801470518112183, 0.7009803652763367, 0.780637264251709, 0.9632353186607361], [0.06862745434045792, 0.11151960492134094, 0.1568627506494522, 0.14215686917304993, 0.16421568393707275, 0.15808823704719543, 0.1825980395078659, 0.2389705926179886, 0.1924019604921341, 0.22181372344493866, 0.25735294818878174, 0.19485294818878174, 0.2916666567325592, 0.22794117033481598, 0.280637264251709, 0.313112735748291, 0.33578431606292725, 0.2916666567325592, 0.30392158031463623, 0.32475489377975464, 0.3014705777168274, 0.2916666567325592, 0.2953431308269501, 0.30882352590560913, 0.2977941036224365, 0.31004902720451355, 0.281862735748291, 0.30392158031463623, 0.3903186321258545, 0.33088234066963196, 0.38235294818878174, 0.34191176295280457, 0.3223039209842682, 0.3174019753932953, 0.38725489377975464, 0.35049018263816833, 0.30882352590560913, 0.3860294222831726, 0.37132352590560913, 0.40441176295280457, 0.37132352590560913, 0.406862735748291, 0.40318626165390015, 0.36764705181121826, 0.4227941036224365, 0.40441176295280457, 0.38357841968536377, 0.35171568393707275, 0.4215686321258545, 0.45588234066963196, 0.40318626165390015, 0.4473039209842682, 0.3854166567325592, 0.4375, NaN, 0.406862735748291, 0.46078431606292725, 0.3700980246067047, 0.4675244987010956, 0.4730392098426819, 0.47549018263816833, 0.4436274468898773, 0.5220588445663452, 0.533088207244873, 0.5649510025978088, 0.5833333134651184, 0.4987744987010956, 0.6409313678741455, 0.5906862616539001, 0.6286764740943909, 0.6225489974021912, 0.6096813678741455, 0.6053921580314636, 0.6348039507865906, 0.6372548937797546, 0.6274510025978088, 0.6789215803146362, 0.7009803652763367, 0.779411792755127, 0.9632353186607361], [0.02818627469241619, 0.04901960864663124, 0.03799019753932953, 0.08210784196853638, 0.03799019753932953, 0.06985294073820114, 0.0882352963089943, 0.0992647036910057, 0.08946078270673752, 0.14460784196853638, 0.08946078270673752, 0.15808823704719543, 0.14215686917304993, 0.21446079015731812, 0.15441176295280457, 0.23284313082695007, 0.22549019753932953, 0.1041666641831398, 0.11274509876966476, 0.19485294818878174, 0.24632352590560913, 0.2610294222831726, 0.281862735748291, 0.11764705926179886, 0.16911764442920685, 0.13909313082695007, 0.24264705181121826, 0.10049019753932953, 0.32598039507865906, 0.21691176295280457, 0.34191176295280457, 0.23774509131908417, 0.281862735748291, 0.14705882966518402, 0.3529411852359772, 0.3235294222831726, 0.36274510622024536, 0.3284313678741455, 0.41299018263816833, 0.3223039209842682, 0.3584558963775635, 0.34681373834609985, 0.3333333432674408, 0.4117647111415863, 0.3498774468898773, 0.3860294222831726, 0.40808823704719543, 0.4803921580314636, 0.4595588147640228, 0.41360294818878174, 0.47181373834609985, 0.4632352888584137, 0.5355392098426819, 0.5882353186607361, 0.593137264251709, NaN, 0.530637264251709, 0.5490196347236633, 0.48100489377975464, 0.6397058963775635, 0.6433823704719543, 0.6066176295280457, 0.656862735748291, 0.6360294222831726, 0.5735294222831726, 0.7107843160629272, 0.75, 0.7046568393707275, 0.6507353186607361, 0.7242646813392639, 0.751225471496582, 0.718137264251709, 0.7218137383460999, 0.8100489974021912, 0.7696078419685364, 0.7365196347236633, 0.7745097875595093, 0.8223039507865906, 0.844362735748291, 0.9448529481887817], [0.050245098769664764, 0.07904411852359772, 0.0833333358168602, 0.09313725680112839, 0.09068627655506134, 0.09068627655506134, 0.09681372344493866, 0.09436274319887161, 0.13909313082695007, 0.13725490868091583, 0.13725490868091583, 0.17524509131908417, 0.16911764442920685, 0.20465686917304993, 0.17034313082695007, 0.23223039507865906, 0.19975490868091583, 0.20955882966518402, 0.2389705926179886, 0.20710784196853638, 0.24571079015731812, 0.20710784196853638, 0.26838234066963196, 0.24816176295280457, 0.2138480395078659, 0.25857841968536377, 0.2708333432674408, 0.27941176295280457, 0.2855392098426819, 0.2671568691730499, 0.2928921580314636, 0.28921568393707275, 0.25367647409439087, 0.28431373834609985, 0.27328431606292725, 0.29411765933036804, 0.34681373834609985, 0.2757352888584137, 0.35171568393707275, 0.3915441036224365, 0.37683823704719543, 0.25674018263816833, 0.3296568691730499, 0.3799019753932953, 0.4123774468898773, 0.34681373834609985, 0.4050244987010956, 0.46078431606292725, 0.31495097279548645, 0.4473039209842682, 0.4313725531101227, 0.39338234066963196, 0.4779411852359772, 0.5367646813392639, 0.5392156839370728, 0.469362735748291, NaN, 0.532475471496582, 0.4644607901573181, 0.591911792755127, 0.5894607901573181, 0.5514705777168274, 0.5355392098426819, 0.6121323704719543, 0.5833333134651184, 0.6703431606292725, 0.6740196347236633, 0.6348039507865906, 0.626225471496582, 0.7218137383460999, 0.7009803652763367, 0.8333333134651184, 0.7647058963775635, 0.688725471496582, 0.7426470518112183, 0.7873774766921997, 0.7561274766921997, 0.8651960492134094, 0.8321078419685364, 0.9460784196853638], [0.030637254938483238, 0.061274509876966476, 0.05637254938483238, 0.07475490123033524, 0.06862745434045792, 0.08210784196853638, 0.11029411852359772, 0.12377450615167618, 0.0882352963089943, 0.12867647409439087, 0.14338235557079315, 0.1299019604921341, 0.2132352888584137, 0.18872548639774323, 0.1556372493505478, 0.25857841968536377, 0.28921568393707275, 0.16544117033481598, 0.1568627506494522, 0.24877451360225677, 0.24754901230335236, 0.281862735748291, 0.24632352590560913, 0.1715686321258545, 0.17279411852359772, 0.22303920984268188, 0.23284313082695007, 0.1819852888584137, 0.3688725531101227, 0.22303920984268188, 0.3529411852359772, 0.2696078419685364, 0.25857841968536377, 0.21078431606292725, 0.3229166567325592, 0.2977941036224365, 0.3958333432674408, 0.3125, 0.4424019753932953, 0.3223039209842682, 0.34191176295280457, 0.3125, 0.30392158031463623, 0.3382352888584137, 0.35049018263816833, 0.34681373834609985, 0.38848039507865906, 0.30882352590560913, 0.3952205777168274, 0.38725489377975464, 0.4227941036224365, 0.3921568691730499, 0.406862735748291, 0.6323529481887817, 0.6299019455909729, 0.45098039507865906, 0.4675244987010956, NaN, 0.468137264251709, 0.5661764740943909, 0.5674019455909729, 0.6789215803146362, 0.6801470518112183, 0.5698529481887817, 0.593137264251709, 0.716911792755127, 0.6274510025978088, 0.6605392098426819, 0.6286764740943909, 0.6954656839370728, 0.781862735748291, 0.7022058963775635, 0.7156862616539001, 0.8958333134651184, 0.8014705777168274, 0.7254902124404907, 0.7267156839370728, 0.8057597875595093, 0.8860294222831726, 0.9718137383460999], [0.02818627469241619, 0.040441177785396576, 0.03921568766236305, 0.05882352963089943, 0.0416666679084301, 0.05147058889269829, 0.0625, 0.06617647409439087, 0.07230392098426819, 0.11764705926179886, 0.07598039507865906, 0.12254901975393295, 0.11703431606292725, 0.16789215803146362, 0.14338235557079315, 0.22671568393707275, 0.23651960492134094, 0.04289215803146362, 0.10784313827753067, 0.19975490868091583, 0.2561274468898773, 0.21629901230335236, 0.2401960790157318, 0.13357843458652496, 0.1299019604921341, 0.08394607901573181, 0.23774509131908417, 0.08946078270673752, 0.3174019753932953, 0.18382352590560913, 0.3296568691730499, 0.25735294818878174, 0.27450981736183167, 0.13235294818878174, 0.31985294818878174, 0.2953431308269501, 0.3216911852359772, 0.34558823704719543, 0.39767158031463623, 0.26776960492134094, 0.3639705777168274, 0.35171568393707275, 0.3075980246067047, 0.3762255012989044, 0.3137255012989044, 0.3897058963775635, 0.4154411852359772, 0.46200981736183167, 0.46629902720451355, 0.40870097279548645, 0.4571078419685364, 0.44546568393707275, 0.5049019455909729, 0.5349264740943909, 0.532475471496582, 0.5189951062202454, 0.5355392098426819, 0.531862735748291, NaN, 0.6335784196853638, 0.6299019455909729, 0.5741421580314636, 0.6188725233078003, 0.6482843160629272, 0.6078431606292725, 0.6948529481887817, 0.7058823704719543, 0.686274528503418, 0.6691176295280457, 0.718137264251709, 0.7303921580314636, 0.7291666865348816, 0.6973039507865906, 0.7726715803146362, 0.7524510025978088, 0.7120097875595093, 0.8051470518112183, 0.813725471496582, 0.8492646813392639, 0.9552696347236633], [0.12745098769664764, 0.14889705181121826, 0.16421568393707275, 0.1556372493505478, 0.16299019753932953, 0.17034313082695007, 0.19485294818878174, 0.16911764442920685, 0.2181372493505478, 0.23529411852359772, 0.22303920984268188, 0.24877451360225677, 0.22671568393707275, 0.27757352590560913, 0.2340686321258545, 0.2855392098426819, 0.2450980395078659, 0.24080882966518402, 0.24142156541347504, 0.24387255311012268, 0.2714460790157318, 0.23774509131908417, 0.3112744987010956, 0.24142156541347504, 0.27818626165390015, 0.25, 0.28370097279548645, 0.26225489377975464, 0.3075980246067047, 0.2493872493505478, 0.2953431308269501, 0.32107841968536377, 0.27818626165390015, 0.2757352888584137, 0.3492647111415863, 0.28308823704719543, 0.2965686321258545, 0.2647058963775635, 0.2696078419685364, 0.2904411852359772, 0.2769607901573181, 0.2965686321258545, 0.3082107901573181, 0.32107841968536377, 0.3026960790157318, 0.32598039507865906, 0.2965686321258545, 0.40441176295280457, 0.3688725531101227, 0.3272058963775635, 0.3272058963775635, 0.31495097279548645, 0.40379902720451355, 0.5208333134651184, 0.5269607901573181, 0.3602941036224365, 0.40808823704719543, 0.43382352590560913, 0.36642158031463623, NaN, 0.53125, 0.47181373834609985, 0.4571078419685364, 0.4264705777168274, 0.5551470518112183, 0.529411792755127, 0.43014705181121826, 0.5257353186607361, 0.5772058963775635, 0.9534313678741455, 0.6225489974021912, 0.623774528503418, 0.6629902124404907, 0.6102941036224365, 0.6948529481887817, 0.6850489974021912, 0.6139705777168274, 0.7892156839370728, 0.8259803652763367, 0.9031862616539001], [0.12867647409439087, 0.1507352888584137, 0.16421568393707275, 0.16053920984268188, 0.15931372344493866, 0.17034313082695007, 0.18995098769664764, 0.16973039507865906, 0.21446079015731812, 0.23284313082695007, 0.22058823704719543, 0.24754901230335236, 0.2242647111415863, 0.27818626165390015, 0.23284313082695007, 0.28431373834609985, 0.24877451360225677, 0.2401960790157318, 0.2450980395078659, 0.2395833283662796, 0.27450981736183167, 0.23651960492134094, 0.31004902720451355, 0.24264705181121826, 0.27757352590560913, 0.24877451360225677, 0.2855392098426819, 0.2659313678741455, 0.3112744987010956, 0.24877451360225677, 0.2984068691730499, 0.3235294222831726, 0.27450981736183167, 0.27389705181121826, 0.3541666567325592, 0.280637264251709, 0.29901960492134094, 0.26225489377975464, 0.2708333432674408, 0.2953431308269501, 0.2855392098426819, 0.28982841968536377, 0.3112744987010956, 0.31617647409439087, 0.30392158031463623, 0.3186274468898773, 0.29963234066963196, 0.3946078419685364, 0.3651960790157318, 0.3223039209842682, 0.3296568691730499, 0.3137255012989044, 0.39828431606292725, 0.5208333134651184, 0.5245097875595093, 0.35661765933036804, 0.4105392098426819, 0.4325980246067047, 0.3700980246067047, 0.46875, NaN, 0.4730392098426819, 0.46078431606292725, 0.42524510622024536, 0.5514705777168274, 0.5281862616539001, 0.4325980246067047, 0.5232843160629272, 0.5723039507865906, 0.966911792755127, 0.623774528503418, 0.6200980544090271, 0.655637264251709, 0.6127451062202454, 0.686274528503418, 0.6825980544090271, 0.6127451062202454, 0.7892156839370728, 0.8210784196853638, 0.9031862616539001], [0.03186274692416191, 0.04534313827753067, 0.053921569138765335, 0.07230392098426819, 0.05147058889269829, 0.0674019604921341, 0.0882352963089943, 0.0882352963089943, 0.06862745434045792, 0.11764705926179886, 0.11825980246067047, 0.12132352590560913, 0.17892156541347504, 0.16053920984268188, 0.13725490868091583, 0.22058823704719543, 0.24325980246067047, 0.12622548639774323, 0.12193627655506134, 0.23039215803146362, 0.2291666716337204, 0.24142156541347504, 0.2083333283662796, 0.13970588147640228, 0.14460784196853638, 0.1715686321258545, 0.20098039507865906, 0.15870098769664764, 0.3082107901573181, 0.19362744688987732, 0.2971813678741455, 0.25857841968536377, 0.2401960790157318, 0.18014705181121826, 0.311887264251709, 0.2561274468898773, 0.31617647409439087, 0.3014705777168274, 0.375, 0.28431373834609985, 0.31617647409439087, 0.28799018263816833, 0.2708333432674408, 0.27818626165390015, 0.31004902720451355, 0.3125, 0.3480392098426819, 0.2561274468898773, 0.37745097279548645, 0.37132352590560913, 0.3909313678741455, 0.36274510622024536, 0.3388480246067047, 0.5588235259056091, 0.5563725233078003, 0.39338234066963196, 0.4485294222831726, 0.32107841968536377, 0.4258578419685364, 0.5281862616539001, 0.5269607901573181, NaN, 0.5968137383460999, 0.5220588445663452, 0.5674019455909729, 0.6495097875595093, 0.6078431606292725, 0.6164215803146362, 0.6053921580314636, 0.6776960492134094, 0.7058823704719543, 0.6948529481887817, 0.6948529481887817, 0.7647058963775635, 0.7702205777168274, 0.7071078419685364, 0.7083333134651184, 0.7953431606292725, 0.8786764740943909, 0.970588207244873], [0.04411764815449715, 0.05514705926179886, 0.0520833320915699, 0.06372549384832382, 0.08700980246067047, 0.07107843458652496, 0.08210784196853638, 0.06372549384832382, 0.15931372344493866, 0.10049019753932953, 0.18566176295280457, 0.12377450615167618, 0.1924019604921341, 0.12867647409439087, 0.2028186321258545, 0.23100490868091583, 0.22610294818878174, 0.19607843458652496, 0.22549019753932953, 0.22303920984268188, 0.23590686917304993, 0.125, 0.18872548639774323, 0.23039215803146362, 0.23100490868091583, 0.20955882966518402, 0.20098039507865906, 0.23100490868091583, 0.2922794222831726, 0.25, 0.2598039209842682, 0.2867647111415863, 0.23651960492134094, 0.24877451360225677, 0.30514705181121826, 0.22549019753932953, 0.26348039507865906, 0.28799018263816833, 0.32598039507865906, 0.2916666567325592, 0.31617647409439087, 0.3272058963775635, 0.3075980246067047, 0.24754901230335236, 0.3137255012989044, 0.34068626165390015, 0.3547794222831726, 0.2977941036224365, 0.4166666567325592, 0.35661765933036804, 0.3688725531101227, 0.3897058963775635, 0.31004902720451355, 0.4803921580314636, 0.4779411852359772, 0.343137264251709, 0.4644607901573181, 0.31985294818878174, 0.3811274468898773, 0.5428921580314636, 0.5392156839370728, 0.40318626165390015, NaN, 0.4828431308269501, 0.5404411554336548, 0.6139705777168274, 0.5833333134651184, 0.5674019455909729, 0.5759803652763367, 0.686274528503418, 0.5784313678741455, 0.6801470518112183, 0.7058823704719543, 0.6127451062202454, 0.6813725233078003, 0.7248774766921997, 0.7279411554336548, 0.8100489974021912, 0.7965686321258545, 0.9571078419685364], [0.02450980432331562, 0.03431372717022896, 0.030637254938483238, 0.053921569138765335, 0.036764707416296005, 0.04289215803146362, 0.04411764815449715, 0.02818627469241619, 0.06372549384832382, 0.09558823704719543, 0.0784313753247261, 0.10539215803146362, 0.07720588147640228, 0.12745098769664764, 0.09436274319887161, 0.19117647409439087, 0.18014705181121826, 0.1507352888584137, 0.14338235557079315, 0.09558823704719543, 0.2181372493505478, 0.14705882966518402, 0.18382352590560913, 0.15931372344493866, 0.13112744688987732, 0.1715686321258545, 0.15318627655506134, 0.18137255311012268, 0.24632352590560913, 0.1875, 0.2401960790157318, 0.22303920984268188, 0.19362744688987732, 0.19362744688987732, 0.2450980395078659, 0.1617647111415863, 0.2855392098426819, 0.2034313678741455, 0.29901960492134094, 0.28308823704719543, 0.25367647409439087, 0.10784313827753067, 0.18995098769664764, 0.22671568393707275, 0.2928921580314636, 0.2236519604921341, 0.28308823704719543, 0.38725489377975464, 0.30575981736183167, 0.343137264251709, 0.29411765933036804, 0.19178920984268188, 0.41421568393707275, 0.46691176295280457, 0.46691176295280457, 0.3639705777168274, 0.38786765933036804, 0.43014705181121826, 0.35171568393707275, 0.5735294222831726, 0.5747548937797546, 0.4779411852359772, 0.5171568393707275, NaN, 0.5122548937797546, 0.6109068393707275, 0.5453431606292725, 0.5845588445663452, 0.5539215803146362, 0.6911764740943909, 0.6446078419685364, 0.6470588445663452, 0.6715686321258545, 0.6421568393707275, 0.7083333134651184, 0.6924019455909729, 0.6678921580314636, 0.8026960492134094, 0.8272058963775635, 0.9301470518112183], [0.013480392284691334, 0.05269607901573181, 0.03799019753932953, 0.053921569138765335, 0.04411764815449715, 0.04289215803146362, 0.06188725307583809, 0.06495098024606705, 0.05637254938483238, 0.09803921729326248, 0.06372549384832382, 0.0882352963089943, 0.11397058516740799, 0.10661764442920685, 0.0992647036910057, 0.1973039209842682, 0.20220588147640228, 0.08578431606292725, 0.125, 0.1348039209842682, 0.2193627506494522, 0.1458333283662796, 0.17401960492134094, 0.1348039209842682, 0.1458333283662796, 0.12316176295280457, 0.18382352590560913, 0.16789215803146362, 0.2720588147640228, 0.15196079015731812, 0.26348039507865906, 0.21691176295280457, 0.20955882966518402, 0.19485294818878174, 0.23161764442920685, 0.21997548639774323, 0.2610294222831726, 0.23713235557079315, 0.2965686321258545, 0.2132352888584137, 0.2450980395078659, 0.27818626165390015, 0.2769607901573181, 0.2757352888584137, 0.25306373834609985, 0.28431373834609985, 0.27818626165390015, 0.3492647111415863, 0.36151960492134094, 0.3075980246067047, 0.31985294818878174, 0.34681373834609985, 0.37132352590560913, 0.42524510622024536, 0.43504902720451355, 0.4264705777168274, 0.4166666567325592, 0.406862735748291, 0.3921568691730499, 0.44485294818878174, 0.4485294222831726, 0.4325980246067047, 0.4595588147640228, 0.48774510622024536, NaN, 0.5514705777168274, 0.561274528503418, 0.563725471496582, 0.7395833134651184, 0.530637264251709, 0.5906862616539001, 0.5894607901573181, 0.6507353186607361, 0.6139705777168274, 0.6047794222831726, 0.6629902124404907, 0.843137264251709, 0.716911792755127, 0.7450980544090271, 0.9350489974021912], [0.046568628400564194, 0.05269607901573181, 0.06372549384832382, 0.07230392098426819, 0.06617647409439087, 0.06985294073820114, 0.08455882221460342, 0.0680147036910057, 0.12745098769664764, 0.07230392098426819, 0.12377450615167618, 0.1090686246752739, 0.14215686917304993, 0.12132352590560913, 0.14154411852359772, 0.18933823704719543, 0.1764705926179886, 0.16421568393707275, 0.19607843458652496, 0.16421568393707275, 0.2389705926179886, 0.09681372344493866, 0.11519607901573181, 0.2034313678741455, 0.14828431606292725, 0.17892156541347504, 0.14215686917304993, 0.19546568393707275, 0.22487744688987732, 0.19975490868091583, 0.18504901230335236, 0.22549019753932953, 0.17279411852359772, 0.20710784196853638, 0.22549019753932953, 0.13848039507865906, 0.24754901230335236, 0.23651960492134094, 0.2640931308269501, 0.2132352888584137, 0.25367647409439087, 0.22058823704719543, 0.2236519604921341, 0.18995098769664764, 0.24877451360225677, 0.2138480395078659, 0.28125, 0.2549019753932953, 0.2757352888584137, 0.30514705181121826, 0.28921568393707275, 0.25367647409439087, 0.2610294222831726, 0.41911765933036804, 0.4166666567325592, 0.28921568393707275, 0.3296568691730499, 0.28308823704719543, 0.30514705181121826, 0.47058823704719543, 0.47181373834609985, 0.35049018263816833, 0.3860294222831726, 0.3890931308269501, 0.4485294222831726, NaN, 0.5061274766921997, 0.4852941036224365, 0.501225471496582, 0.6115196347236633, 0.48651960492134094, 0.595588207244873, 0.626225471496582, 0.5514705777168274, 0.6286764740943909, 0.6488970518112183, 0.6225489974021912, 0.7450980544090271, 0.7647058963775635, 0.9203431606292725], [0.04289215803146362, 0.03921568766236305, 0.0416666679084301, 0.06004901975393295, 0.053921569138765335, 0.06372549384832382, 0.06985294073820114, 0.06004901975393295, 0.0882352963089943, 0.11887254565954208, 0.08700980246067047, 0.12867647409439087, 0.09436274319887161, 0.16544117033481598, 0.12377450615167618, 0.19362744688987732, 0.17401960492134094, 0.08088235557079315, 0.1145833358168602, 0.11519607901573181, 0.19607843458652496, 0.16789215803146362, 0.18872548639774323, 0.12377450615167618, 0.11764705926179886, 0.09620098024606705, 0.18995098769664764, 0.1458333283662796, 0.2181372493505478, 0.09068627655506134, 0.2389705926179886, 0.2034313678741455, 0.1507352888584137, 0.15196079015731812, 0.27328431606292725, 0.18627451360225677, 0.1875, 0.1985294073820114, 0.17892156541347504, 0.1825980395078659, 0.1973039209842682, 0.1930147111415863, 0.18933823704719543, 0.20710784196853638, 0.2083333283662796, 0.22794117033481598, 0.2401960790157318, 0.3014705777168274, 0.2671568691730499, 0.24387255311012268, 0.24387255311012268, 0.2561274468898773, 0.3235294222831726, 0.5024510025978088, 0.501225471496582, 0.25, 0.32598039507865906, 0.37254902720451355, 0.29411765933036804, 0.5698529481887817, 0.5674019455909729, 0.3921568691730499, 0.4166666567325592, 0.4546568691730499, 0.4387255012989044, 0.4938725531101227, NaN, 0.4901960790157318, 0.46078431606292725, 0.6531862616539001, 0.5024510025978088, 0.593137264251709, 0.654411792755127, 0.5674019455909729, 0.6924019455909729, 0.6715686321258545, 0.5735294222831726, 0.7610294222831726, 0.7953431606292725, 0.8970588445663452], [0.029411764815449715, 0.03799019753932953, 0.040441177785396576, 0.05514705926179886, 0.05147058889269829, 0.06372549384832382, 0.07352941483259201, 0.06617647409439087, 0.061274509876966476, 0.09313725680112839, 0.09558823704719543, 0.08272058516740799, 0.1409313678741455, 0.10661764442920685, 0.11887254565954208, 0.13970588147640228, 0.15318627655506134, 0.16789215803146362, 0.14460784196853638, 0.18137255311012268, 0.16911764442920685, 0.16544117033481598, 0.16053920984268188, 0.15318627655506134, 0.10539215803146362, 0.17953431606292725, 0.050245098769664764, 0.17279411852359772, 0.19485294818878174, 0.20588235557079315, 0.19362744688987732, 0.1403186321258545, 0.14460784196853638, 0.18382352590560913, 0.23774509131908417, 0.09803921729326248, 0.26348039507865906, 0.2708333432674408, 0.32598039507865906, 0.25735294818878174, 0.2916666567325592, 0.23161764442920685, 0.18872548639774323, 0.23529411852359772, 0.27818626165390015, 0.27941176295280457, 0.3137255012989044, 0.2867647111415863, 0.2965686321258545, 0.3223039209842682, 0.35661765933036804, 0.2916666567325592, 0.3223039209842682, 0.3590686321258545, 0.3590686321258545, 0.2953431308269501, 0.3651960790157318, 0.3394607901573181, 0.3137255012989044, 0.4742647111415863, 0.47671568393707275, 0.38357841968536377, 0.4325980246067047, 0.4154411852359772, 0.4362744987010956, 0.5147058963775635, 0.5098039507865906, NaN, 0.46629902720451355, 0.5735294222831726, 0.47549018263816833, 0.530637264251709, 0.5870097875595093, 0.5257353186607361, 0.5821078419685364, 0.593137264251709, 0.5906862616539001, 0.6752451062202454, 0.7083333134651184, 0.9203431606292725], [0.014705882407724857, 0.03553921729326248, 0.032475490123033524, 0.050245098769664764, 0.046568628400564194, 0.0416666679084301, 0.05514705926179886, 0.057598039507865906, 0.04901960864663124, 0.07230392098426819, 0.0625, 0.07904411852359772, 0.09436274319887161, 0.09803921729326248, 0.09191176295280457, 0.17279411852359772, 0.18382352590560913, 0.07107843458652496, 0.10661764442920685, 0.125, 0.20465686917304993, 0.12745098769664764, 0.13848039507865906, 0.11397058516740799, 0.11151960492134094, 0.10171568393707275, 0.15012255311012268, 0.13909313082695007, 0.25674018263816833, 0.1200980395078659, 0.2512255012989044, 0.2083333283662796, 0.17401960492134094, 0.15441176295280457, 0.22242647409439087, 0.17953431606292725, 0.24264705181121826, 0.22549019753932953, 0.27328431606292725, 0.18504901230335236, 0.22732843458652496, 0.2291666716337204, 0.2291666716337204, 0.22303920984268188, 0.2236519604921341, 0.22794117033481598, 0.25367647409439087, 0.31617647409439087, 0.3272058963775635, 0.2720588147640228, 0.2928921580314636, 0.29901960492134094, 0.33578431606292725, 0.40441176295280457, 0.40931373834609985, 0.3492647111415863, 0.3737744987010956, 0.37132352590560913, 0.33088234066963196, 0.4227941036224365, 0.4276960790157318, 0.3946078419685364, 0.42401960492134094, 0.44607841968536377, 0.2604166567325592, 0.4987744987010956, 0.5392156839370728, 0.5337010025978088, NaN, 0.5085784196853638, 0.5465686321258545, 0.5502451062202454, 0.6335784196853638, 0.5845588445663452, 0.5674019455909729, 0.6446078419685364, 0.7469362616539001, 0.6985294222831726, 0.716911792755127, 0.9142156839370728], [0.09558823704719543, 0.11029411852359772, 0.12622548639774323, 0.10784313827753067, 0.13112744688987732, 0.1360294073820114, 0.14338235557079315, 0.12867647409439087, 0.16299019753932953, 0.15441176295280457, 0.16789215803146362, 0.16421568393707275, 0.17034313082695007, 0.1764705926179886, 0.17892156541347504, 0.22303920984268188, 0.20588235557079315, 0.1715686321258545, 0.18872548639774323, 0.18995098769664764, 0.22242647409439087, 0.16544117033481598, 0.21446079015731812, 0.1875, 0.19362744688987732, 0.18627451360225677, 0.2181372493505478, 0.2083333283662796, 0.24877451360225677, 0.18504901230335236, 0.2401960790157318, 0.26225489377975464, 0.21200980246067047, 0.21078431606292725, 0.20465686917304993, 0.20955882966518402, 0.22549019753932953, 0.1875, 0.19914215803146362, 0.21446079015731812, 0.18995098769664764, 0.19975490868091583, 0.2401960790157318, 0.22058823704719543, 0.2291666716337204, 0.22181372344493866, 0.20465686917304993, 0.26776960492134094, 0.25735294818878174, 0.25735294818878174, 0.22181372344493866, 0.22181372344493866, 0.2720588147640228, 0.3700980246067047, 0.37132352590560913, 0.2757352888584137, 0.27818626165390015, 0.30453431606292725, 0.281862735748291, 0.046568628400564194, 0.033088237047195435, 0.3223039209842682, 0.3137255012989044, 0.30882352590560913, 0.469362735748291, 0.38848039507865906, 0.34681373834609985, 0.4264705777168274, 0.49142158031463623, NaN, 0.46568626165390015, 0.4522058963775635, 0.5575980544090271, 0.43382352590560913, 0.5526960492134094, 0.5674019455909729, 0.531862735748291, 0.6617646813392639, 0.6911764740943909, 0.8725489974021912], [0.013480392284691334, 0.02450980432331562, 0.0416666679084301, 0.0416666679084301, 0.03921568766236305, 0.04411764815449715, 0.061274509876966476, 0.057598039507865906, 0.03921568766236305, 0.05453431233763695, 0.06985294073820114, 0.04901960864663124, 0.10784313827753067, 0.07720588147640228, 0.11029411852359772, 0.1090686246752739, 0.15196079015731812, 0.08210784196853638, 0.09803921729326248, 0.1556372493505478, 0.18627451360225677, 0.13848039507865906, 0.11519607901573181, 0.11151960492134094, 0.09068627655506134, 0.10661764442920685, 0.06495098024606705, 0.12132352590560913, 0.18995098769664764, 0.13357843458652496, 0.19485294818878174, 0.1568627506494522, 0.15318627655506134, 0.12377450615167618, 0.20955882966518402, 0.13112744688987732, 0.21200980246067047, 0.2083333283662796, 0.280637264251709, 0.21200980246067047, 0.24142156541347504, 0.23529411852359772, 0.18382352590560913, 0.1868872493505478, 0.22303920984268188, 0.20465686917304993, 0.25735294818878174, 0.08088235557079315, 0.2549019753932953, 0.31495097279548645, 0.2720588147640228, 0.281862735748291, 0.14950980246067047, 0.3811274468898773, 0.37745097279548645, 0.24877451360225677, 0.29901960492134094, 0.2181372493505478, 0.2696078419685364, 0.37745097279548645, 0.3762255012989044, 0.29411765933036804, 0.4215686321258545, 0.3553921580314636, 0.40931373834609985, 0.5134803652763367, 0.49754902720451355, 0.5245097875595093, 0.4534313678741455, 0.5343137383460999, NaN, 0.5575980544090271, 0.595588207244873, 0.5453431606292725, 0.623774528503418, 0.6090686321258545, 0.5759803652763367, 0.7009803652763367, 0.7573529481887817, 0.9718137383460999], [0.023284314200282097, 0.033088237047195435, 0.03431372717022896, 0.050245098769664764, 0.04901960864663124, 0.04779411852359772, 0.05514705926179886, 0.05147058889269829, 0.04779411852359772, 0.061274509876966476, 0.05269607901573181, 0.07965686172246933, 0.08210784196853638, 0.09436274319887161, 0.06924019753932953, 0.1348039209842682, 0.1360294073820114, 0.11764705926179886, 0.12254901975393295, 0.12254901975393295, 0.16053920984268188, 0.10539215803146362, 0.12377450615167618, 0.1348039209842682, 0.10294117778539658, 0.13235294818878174, 0.09436274319887161, 0.15441176295280457, 0.18872548639774323, 0.12377450615167618, 0.18872548639774323, 0.17892156541347504, 0.1354166716337204, 0.16421568393707275, 0.18382352590560913, 0.11397058516740799, 0.1819852888584137, 0.17524509131908417, 0.19975490868091583, 0.18382352590560913, 0.16789215803146362, 0.14154411852359772, 0.15318627655506134, 0.17524509131908417, 0.20588235557079315, 0.12745098769664764, 0.20098039507865906, 0.25, 0.04901960864663124, 0.25245097279548645, 0.21200980246067047, 0.2181372493505478, 0.2708333432674408, 0.3903186321258545, 0.3903186321258545, 0.281862735748291, 0.1666666716337204, 0.2977941036224365, 0.2708333432674408, 0.3762255012989044, 0.3799019753932953, 0.30514705181121826, 0.31985294818878174, 0.3529411852359772, 0.4105392098426819, 0.40441176295280457, 0.406862735748291, 0.469362735748291, 0.44975489377975464, 0.5477941036224365, 0.4424019753932953, NaN, 0.6335784196853638, 0.47732841968536377, 0.5735294222831726, 0.6531862616539001, 0.5882353186607361, 0.7475489974021912, 0.7450980544090271, 0.9129902124404907], [0.0416666679084301, 0.0674019604921341, 0.07230392098426819, 0.05882352963089943, 0.07475490123033524, 0.07965686172246933, 0.09068627655506134, 0.0784313753247261, 0.09436274319887161, 0.12745098769664764, 0.09742647409439087, 0.15931372344493866, 0.09681372344493866, 0.17340686917304993, 0.1348039209842682, 0.14460784196853638, 0.13235294818878174, 0.11887254565954208, 0.1409313678741455, 0.11887254565954208, 0.14338235557079315, 0.17401960492134094, 0.22181372344493866, 0.14460784196853638, 0.18627451360225677, 0.13848039507865906, 0.13235294818878174, 0.14950980246067047, 0.16911764442920685, 0.16053920984268188, 0.14767156541347504, 0.23284313082695007, 0.1776960790157318, 0.15808823704719543, 0.17524509131908417, 0.13725490868091583, 0.16421568393707275, 0.17279411852359772, 0.11397058516740799, 0.21446079015731812, 0.13848039507865906, 0.1881127506494522, 0.23835784196853638, 0.20220588147640228, 0.22671568393707275, 0.20465686917304993, 0.16544117033481598, 0.25796568393707275, 0.17279411852359772, 0.2340686321258545, 0.18014705181121826, 0.25735294818878174, 0.2720588147640228, 0.3946078419685364, 0.3946078419685364, 0.27818626165390015, 0.23529411852359772, 0.28431373834609985, 0.3026960790157318, 0.33700981736183167, 0.344362735748291, 0.30514705181121826, 0.29411765933036804, 0.3284313678741455, 0.3492647111415863, 0.3737744987010956, 0.34558823704719543, 0.41299018263816833, 0.36642158031463623, 0.4424019753932953, 0.40441176295280457, 0.36642158031463623, NaN, 0.406862735748291, 0.44485294818878174, 0.6672794222831726, 0.46078431606292725, 0.8125, 0.5490196347236633, 0.7585784196853638], [0.014705882407724857, 0.014705882407724857, 0.026960784569382668, 0.03186274692416191, 0.022058824077248573, 0.03431372717022896, 0.04411764815449715, 0.04534313827753067, 0.03431372717022896, 0.03799019753932953, 0.05882352963089943, 0.04411764815449715, 0.07965686172246933, 0.06985294073820114, 0.07965686172246933, 0.1348039209842682, 0.1360294073820114, 0.05147058889269829, 0.057598039507865906, 0.11151960492134094, 0.13970588147640228, 0.11029411852359772, 0.12377450615167618, 0.06617647409439087, 0.07230392098426819, 0.07475490123033524, 0.05882352963089943, 0.06495098024606705, 0.20098039507865906, 0.09803921729326248, 0.18627451360225677, 0.14460784196853638, 0.12254901975393295, 0.07965686172246933, 0.20098039507865906, 0.1090686246752739, 0.17401960492134094, 0.16789215803146362, 0.23651960492134094, 0.17892156541347504, 0.20465686917304993, 0.2034313678741455, 0.15318627655506134, 0.14215686917304993, 0.20955882966518402, 0.1985294073820114, 0.23161764442920685, 0.11397058516740799, 0.2696078419685364, 0.24264705181121826, 0.25857841968536377, 0.2549019753932953, 0.1513480395078659, 0.36764705181121826, 0.3651960790157318, 0.18995098769664764, 0.3112744987010956, 0.1041666641831398, 0.22732843458652496, 0.3897058963775635, 0.38725489377975464, 0.23529411852359772, 0.38725489377975464, 0.3578431308269501, 0.3860294222831726, 0.4485294222831726, 0.4325980246067047, 0.4742647111415863, 0.4154411852359772, 0.5661764740943909, 0.4546568691730499, 0.5226715803146362, 0.593137264251709, NaN, 0.5882353186607361, 0.6151960492134094, 0.5674019455909729, 0.6960784196853638, 0.748774528503418, 0.9460784196853638], [0.030637254938483238, 0.05147058889269829, 0.04901960864663124, 0.04534313827753067, 0.05147058889269829, 0.06985294073820114, 0.08210784196853638, 0.09191176295280457, 0.07475490123033524, 0.07475490123033524, 0.09558823704719543, 0.08272058516740799, 0.1090686246752739, 0.08946078270673752, 0.11029411852359772, 0.1458333283662796, 0.14950980246067047, 0.10171568393707275, 0.10049019753932953, 0.13848039507865906, 0.16299019753932953, 0.11764705926179886, 0.125, 0.1090686246752739, 0.12254901975393295, 0.12377450615167618, 0.09375, 0.13357843458652496, 0.19607843458652496, 0.12745098769664764, 0.18504901230335236, 0.1764705926179886, 0.13725490868091583, 0.13664215803146362, 0.20955882966518402, 0.10661764442920685, 0.1299019604921341, 0.18137255311012268, 0.13970588147640228, 0.20465686917304993, 0.17034313082695007, 0.18014705181121826, 0.18504901230335236, 0.1348039209842682, 0.22303920984268188, 0.1715686321258545, 0.1973039209842682, 0.13112744688987732, 0.21446079015731812, 0.2665441036224365, 0.18137255311012268, 0.22242647409439087, 0.1617647111415863, 0.36151960492134094, 0.36274510622024536, 0.23039215803146362, 0.25735294818878174, 0.1985294073820114, 0.24754901230335236, 0.30514705181121826, 0.3137255012989044, 0.2297794073820114, 0.3186274468898773, 0.2916666567325592, 0.3952205777168274, 0.37132352590560913, 0.3075980246067047, 0.4178921580314636, 0.4325980246067047, 0.4473039209842682, 0.3762255012989044, 0.4264705777168274, 0.5551470518112183, 0.4117647111415863, NaN, 0.5827205777168274, 0.5392156839370728, 0.6985294222831726, 0.8866421580314636, 0.9742646813392639], [0.036764707416296005, 0.061274509876966476, 0.05637254938483238, 0.05269607901573181, 0.06004901975393295, 0.06862745434045792, 0.07598039507865906, 0.07352941483259201, 0.09313725680112839, 0.11887254565954208, 0.08026960492134094, 0.1458333283662796, 0.08517156541347504, 0.15808823704719543, 0.12806372344493866, 0.12867647409439087, 0.11642156541347504, 0.1041666641831398, 0.12867647409439087, 0.11029411852359772, 0.12867647409439087, 0.15441176295280457, 0.20098039507865906, 0.13419117033481598, 0.16053920984268188, 0.1194852963089943, 0.1200980395078659, 0.1409313678741455, 0.1568627506494522, 0.14460784196853638, 0.14215686917304993, 0.22181372344493866, 0.1556372493505478, 0.14215686917304993, 0.16911764442920685, 0.1200980395078659, 0.15196079015731812, 0.1611519604921341, 0.10049019753932953, 0.19975490868091583, 0.125, 0.16299019753932953, 0.2083333283662796, 0.1819852888584137, 0.22058823704719543, 0.17401960492134094, 0.14950980246067047, 0.24203431606292725, 0.15196079015731812, 0.2242647111415863, 0.15502451360225677, 0.2395833283662796, 0.26225489377975464, 0.37254902720451355, 0.37254902720451355, 0.26348039507865906, 0.21262255311012268, 0.27450981736183167, 0.28799018263816833, 0.31495097279548645, 0.3174019753932953, 0.2928921580314636, 0.2751225531101227, 0.3075980246067047, 0.33700981736183167, 0.35110294818878174, 0.3284313678741455, 0.406862735748291, 0.3553921580314636, 0.4325980246067047, 0.3909313678741455, 0.34681373834609985, 0.3327205777168274, 0.3848039209842682, 0.4172794222831726, NaN, 0.4546568691730499, 0.7438725233078003, 0.5367646813392639, 0.7524510025978088], [0.01225490216165781, 0.022058824077248573, 0.025735294446349144, 0.02818627469241619, 0.026960784569382668, 0.025735294446349144, 0.025735294446349144, 0.030637254938483238, 0.02450980432331562, 0.030637254938483238, 0.040441177785396576, 0.04289215803146362, 0.04411764815449715, 0.04411764815449715, 0.03799019753932953, 0.07475490123033524, 0.06862745434045792, 0.040441177785396576, 0.0674019604921341, 0.0625, 0.12867647409439087, 0.0674019604921341, 0.0784313753247261, 0.06372549384832382, 0.06862745434045792, 0.04534313827753067, 0.0784313753247261, 0.07720588147640228, 0.12132352590560913, 0.07230392098426819, 0.12622548639774323, 0.14215686917304993, 0.1041666641831398, 0.0784313753247261, 0.1617647111415863, 0.09191176295280457, 0.14338235557079315, 0.11764705926179886, 0.1666666716337204, 0.1299019604921341, 0.11519607901573181, 0.11764705926179886, 0.1348039209842682, 0.1360294073820114, 0.13664215803146362, 0.12377450615167618, 0.1360294073820114, 0.2193627506494522, 0.18137255311012268, 0.18014705181121826, 0.1617647111415863, 0.1764705926179886, 0.23161764442920685, 0.31985294818878174, 0.32107841968536377, 0.22549019753932953, 0.24387255311012268, 0.27328431606292725, 0.19485294818878174, 0.3860294222831726, 0.38725489377975464, 0.2916666567325592, 0.2720588147640228, 0.3321078419685364, 0.1568627506494522, 0.37745097279548645, 0.4264705777168274, 0.40931373834609985, 0.25306373834609985, 0.468137264251709, 0.42401960492134094, 0.4117647111415863, 0.5392156839370728, 0.4325980246067047, 0.46078431606292725, 0.5453431606292725, NaN, 0.6158088445663452, 0.6078431606292725, 0.8553921580314636], [0.02450980432331562, 0.036764707416296005, 0.039828430861234665, 0.03370098024606705, 0.04289215803146362, 0.05085784196853638, 0.04840686172246933, 0.03799019753932953, 0.07965686172246933, 0.05637254938483238, 0.05882352963089943, 0.09558823704719543, 0.06495098024606705, 0.10784313827753067, 0.10049019753932953, 0.11397058516740799, 0.08946078270673752, 0.06372549384832382, 0.1041666641831398, 0.08210784196853638, 0.1090686246752739, 0.08272058516740799, 0.12377450615167618, 0.1090686246752739, 0.12132352590560913, 0.07720588147640228, 0.09620098024606705, 0.11029411852359772, 0.1348039209842682, 0.09436274319887161, 0.1090686246752739, 0.18627451360225677, 0.11151960492134094, 0.11519607901573181, 0.13725490868091583, 0.08210784196853638, 0.07169117778539658, 0.11764705926179886, 0.05269607901573181, 0.1299019604921341, 0.0992647036910057, 0.10171568393707275, 0.1507352888584137, 0.10171568393707275, 0.14338235557079315, 0.11274509876966476, 0.11764705926179886, 0.16911764442920685, 0.10294117778539658, 0.14460784196853638, 0.10049019753932953, 0.1299019604921341, 0.1764705926179886, 0.29901960492134094, 0.29901960492134094, 0.1776960790157318, 0.1348039209842682, 0.19424019753932953, 0.18627451360225677, 0.21078431606292725, 0.21078431606292725, 0.20465686917304993, 0.18995098769664764, 0.1973039209842682, 0.28308823704719543, 0.2549019753932953, 0.2389705926179886, 0.32475489377975464, 0.3014705777168274, 0.3382352888584137, 0.29901960492134094, 0.25245097279548645, 0.1875, 0.30392158031463623, 0.3014705777168274, 0.2561274468898773, 0.3841911852359772, NaN, 0.46568626165390015, 0.7071078419685364], [0.02818627469241619, 0.03431372717022896, 0.03553921729326248, 0.03553921729326248, 0.03921568766236305, 0.05147058889269829, 0.04901960864663124, 0.05269607901573181, 0.053921569138765335, 0.04289215803146362, 0.0625, 0.05637254938483238, 0.07230392098426819, 0.06495098024606705, 0.0833333358168602, 0.09191176295280457, 0.09007352590560913, 0.06556372344493866, 0.07352941483259201, 0.09191176295280457, 0.12377450615167618, 0.046568628400564194, 0.09313725680112839, 0.06985294073820114, 0.09129901975393295, 0.07352941483259201, 0.07107843458652496, 0.08517156541347504, 0.13112744688987732, 0.0674019604921341, 0.11887254565954208, 0.14828431606292725, 0.09436274319887161, 0.08946078270673752, 0.11151960492134094, 0.07230392098426819, 0.0784313753247261, 0.0992647036910057, 0.06617647409439087, 0.11397058516740799, 0.11213235557079315, 0.10661764442920685, 0.12683823704719543, 0.09068627655506134, 0.1360294073820114, 0.09313725680112839, 0.11397058516740799, 0.08700980246067047, 0.13970588147640228, 0.1360294073820114, 0.11029411852359772, 0.14338235557079315, 0.08210784196853638, 0.2193627506494522, 0.22058823704719543, 0.1556372493505478, 0.16789215803146362, 0.11397058516740799, 0.1507352888584137, 0.17401960492134094, 0.17892156541347504, 0.12132352590560913, 0.2034313678741455, 0.17279411852359772, 0.2549019753932953, 0.23529411852359772, 0.20465686917304993, 0.2916666567325592, 0.28308823704719543, 0.30882352590560913, 0.24264705181121826, 0.2549019753932953, 0.45098039507865906, 0.2512255012989044, 0.11335784196853638, 0.4632352888584137, 0.3921568691730499, 0.5343137383460999, NaN, 0.9375], [0.006127451080828905, 0.0049019609577953815, 0.006127451080828905, 0.0049019609577953815, 0.007352941203862429, 0.00857843179255724, 0.013480392284691334, 0.014705882407724857, 0.006127451080828905, 0.009803921915590763, 0.009803921915590763, 0.006127451080828905, 0.02450980432331562, 0.00857843179255724, 0.01715686358511448, 0.006127451080828905, 0.009803921915590763, 0.013480392284691334, 0.01225490216165781, 0.03186274692416191, 0.03186274692416191, 0.006127451080828905, 0.007352941203862429, 0.013480392284691334, 0.01715686358511448, 0.018382353708148003, 0.011029412038624287, 0.02083333395421505, 0.023284314200282097, 0.015931373462080956, 0.026960784569382668, 0.0416666679084301, 0.022058824077248573, 0.023284314200282097, 0.014705882407724857, 0.013480392284691334, 0.026960784569382668, 0.04779411852359772, 0.030637254938483238, 0.026348039507865906, 0.04779411852359772, 0.040441177785396576, 0.03799019753932953, 0.019607843831181526, 0.03799019753932953, 0.02083333395421505, 0.053921569138765335, 0.00857843179255724, 0.04534313827753067, 0.03921568766236305, 0.05147058889269829, 0.05269607901573181, 0.009803921915590763, 0.036764707416296005, 0.036764707416296005, 0.05514705926179886, 0.053921569138765335, 0.02818627469241619, 0.04473039135336876, 0.09681372344493866, 0.09681372344493866, 0.029411764815449715, 0.04289215803146362, 0.06985294073820114, 0.06495098024606705, 0.07965686172246933, 0.10294117778539658, 0.07965686172246933, 0.08578431606292725, 0.12745098769664764, 0.02818627469241619, 0.08700980246067047, 0.24142156541347504, 0.053921569138765335, 0.025735294446349144, 0.24754901230335236, 0.14460784196853638, 0.2928921580314636, 0.0625, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9212661303},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8560824649},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8503040481},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8333953815},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8237388298},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8194651291},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7896190125},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.77962894},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7366669736},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.722294612},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7099156131},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7090779348},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6720650289},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6706533895},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6655963009},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6515186769},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.6429169748},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6409158589},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.627039897},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6162664457},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6151107623},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6149866556},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.612628753},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6122331814},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.611356727},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5986829856},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5892746336},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5810607492},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5792069996},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5753443778},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5727305178},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5635626091},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5633764582},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5588933364},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5549609076},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5527348568},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.5495470351},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5205618654},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5170560298},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5123091958},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5075856299},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5049950363},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5025440566},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.491809381},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4859533976},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4695954317},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4670358678},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4491964516},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4490180555},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4380119115},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4376008343},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4337614795},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4230888559},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.4079098395},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.3965236391},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3913734814},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.38962832},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3879995029},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3768382367},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3697645187},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3684847359},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3503195574},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3451306151},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3121975053},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3092346122},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2865397745},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2834760502},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2806993061},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2769142465},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2677153129},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.252466493},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2393816702},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2381716934},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2297794119},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2283987951},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2185948747},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1914169162},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1564361501},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1467873544},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0435747087}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_all/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..c4de287b41bae712bb5af755fa70bd38625010aa
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_all/winrate_matrix.csv
@@ -0,0 +1,81 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (default),RealMLP (tuned + ensembled),TabDPT-Turbo (default),"AutoGluon 1.4 (best, 4h)",TabDPT (tuned + ensembled),TabM (tuned + ensembled),TabDPT (tuned),RealMLP (tuned),TabPFNv2 (tuned + ensembled),Mitra (default),LightGBM (tuned + ensembled),CatBoost (tuned + ensembled),TabM (tuned),LimiX (default),TabSwift (default),TabDPT (default),CatBoost (tuned),iLTM (tuned + ensembled),LightGBM (tuned),xRFM (tuned + ensembled),XGBoost (tuned + ensembled),TabPFNv2 (tuned),CatBoost (default),TabPFNv2 (default),ModernNCA (tuned + ensembled),ModernNCA (tuned),XGBoost (tuned),SAP-RPT-OSS (default),xRFM (tuned),TabICL (default),TabM (default),BetaTabPFN (default),ChimeraBoost (tuned + ensembled),EBM (tuned + ensembled),TorchMLP (tuned + ensembled),iLTM (tuned),ModernNCA (default),ChimeraBoost (tuned),RealMLP (default),EBM (tuned),ExtraTrees (tuned + ensembled),FastaiMLP (tuned + ensembled),ChimeraBoost (default),EBM (default),TorchMLP (tuned),ExtraTrees (tuned),Nori-30M (default),Nori (default),XGBoost (default),FastaiMLP (tuned),RandomForest (tuned + ensembled),LightGBM (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),RandomForest (tuned),OrionMSP (default),TorchMLP (default),PerpetualBooster (tuned + ensembled),TabFlex (default),iLTM (default),xRFM (default),PerpetualBooster (tuned),TabSTAR (default),ExtraTrees (default),FastaiMLP (default),Linear (tuned + ensembled),RandomForest (default),KNN (tuned + ensembled),Linear (tuned),PerpetualBooster (default),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.7708333,0.7512255,0.8088235,0.81740195,0.7873775,0.7977941,0.8357843,0.86519605,0.87990195,0.8897059,0.88235295,0.90318626,0.8958333,0.875,0.87990195,0.8933824,0.91544116,0.91544116,0.9264706,0.90625,0.8860294,0.8995098,0.92769605,0.92401963,0.92769605,0.92034316,0.935049,0.89093137,0.9227941,0.88480395,0.94240195,0.9313725,0.9411765,0.8786765,0.91544116,0.9522059,0.93259805,0.91115195,0.935049,0.9264706,0.95955884,0.94240195,0.9387255,0.939951,0.94730395,0.9375,0.95343137,0.9264706,0.95465684,0.935049,0.96936274,0.9485294,0.8811275,0.9313725,0.97181374,0.9497549,0.96936274,0.97181374,0.872549,0.8713235,0.96813726,0.9558824,0.9754902,0.98651963,0.95343137,0.95710784,0.9705882,0.9852941,0.9044118,0.98651963,0.9767157,0.9583333,0.9852941,0.96936274,0.9632353,0.9877451,0.9754902,0.97181374,0.9938725
+TabPFN-3 (default),0.22916667,,0.57965684,0.55514705,0.63480395,0.6188725,0.67034316,0.70710784,0.74509805,0.7745098,0.7512255,0.7585784,0.7977941,0.7916667,0.7794118,0.7794118,0.7732843,0.82598037,0.83455884,0.8370098,0.78431374,0.88235295,0.8272059,0.84068626,0.8639706,0.86151963,0.85294116,0.84865195,0.8400735,0.8541667,0.8192402,0.8786765,0.86519605,0.85906863,0.79656863,0.8639706,0.8854167,0.87009805,0.83026963,0.8860294,0.85906863,0.9105392,0.92585784,0.9050245,0.8860294,0.9252451,0.8762255,0.935049,0.88419116,0.8958333,0.89644605,0.9387255,0.9362745,0.8566176,0.88848037,0.95098037,0.9209559,0.9387255,0.95955884,0.85110295,0.8492647,0.95465684,0.94485295,0.96568626,0.94730395,0.94730395,0.9607843,0.9620098,0.9644608,0.8897059,0.9754902,0.9669118,0.93259805,0.9852941,0.9485294,0.9387255,0.97794116,0.9632353,0.96568626,0.99509805
+"AutoGluon 1.5 (extreme, 4h)",0.24877451,0.42034313,,0.47916666,0.62009805,0.5539216,0.60906863,0.63480395,0.7775735,0.67156863,0.80514705,0.7377451,0.82598037,0.7610294,0.8186275,0.71568626,0.7444853,0.86151963,0.8639706,0.83884805,0.7769608,0.8357843,0.76593137,0.8670343,0.89093137,0.8762255,0.875,0.88848037,0.79901963,0.8792892,0.8088235,0.8860294,0.89705884,0.89460784,0.77205884,0.8694853,0.8645833,0.8835784,0.8382353,0.9166667,0.8897059,0.9252451,0.91544116,0.8958333,0.92401963,0.9117647,0.8982843,0.92401963,0.8982843,0.93014705,0.8995098,0.94485295,0.9166667,0.79534316,0.84313726,0.9620098,0.9166667,0.9436275,0.9607843,0.8357843,0.8357843,0.9460784,0.9479167,0.96936274,0.9620098,0.9362745,0.9583333,0.95955884,0.9675245,0.8737745,0.9583333,0.96568626,0.92769605,0.9730392,0.95098037,0.9436275,0.9742647,0.9601716,0.9644608,0.9938725
+TabICLv2 (default),0.19117647,0.44485295,0.5208333,,0.5870098,0.56740195,0.60906863,0.6525735,0.71568626,0.7316176,0.74509805,0.7665441,0.7867647,0.7683824,0.752451,0.7414216,0.75061274,0.8088235,0.8112745,0.8088235,0.78615195,0.8370098,0.79289216,0.814951,0.84313726,0.8382353,0.8247549,0.8247549,0.7922794,0.8382353,0.8014706,0.84313726,0.8480392,0.82965684,0.78063726,0.8204657,0.8792892,0.8566176,0.8235294,0.8566176,0.8455882,0.87990195,0.875,0.88419116,0.8664216,0.9019608,0.8639706,0.8976716,0.87009805,0.8762255,0.8897059,0.9080882,0.8995098,0.8272059,0.85784316,0.91789216,0.90686274,0.9252451,0.9411765,0.84436274,0.8394608,0.92769605,0.9362745,0.9460784,0.9460784,0.92769605,0.939951,0.94485295,0.9497549,0.89215684,0.9583333,0.9497549,0.9411765,0.96813726,0.95465684,0.94730395,0.97181374,0.966299,0.9644608,0.99509805
+TabPFN-2.6 (default),0.18259804,0.36519608,0.37990198,0.41299018,,0.48284313,0.5435049,0.6139706,0.6911765,0.6433824,0.70955884,0.6727941,0.7757353,0.7022059,0.76348037,0.6911765,0.747549,0.80759805,0.8137255,0.814951,0.7512255,0.79901963,0.72181374,0.814951,0.8419118,0.85294116,0.8235294,0.8492647,0.79289216,0.8517157,0.7726716,0.85906863,0.8602941,0.8713235,0.7916667,0.84313726,0.8088235,0.8682598,0.81985295,0.89460784,0.86151963,0.8958333,0.90686274,0.8982843,0.90318626,0.90931374,0.8915441,0.9080882,0.872549,0.9105392,0.8872549,0.935049,0.9050245,0.7941176,0.8357843,0.9620098,0.90931374,0.9313725,0.9583333,0.8370098,0.84068626,0.9485294,0.9129902,0.9632353,0.9558824,0.9338235,0.9460784,0.9485294,0.95343137,0.8688725,0.9607843,0.95098037,0.9252451,0.97794116,0.9485294,0.939951,0.9730392,0.95710784,0.9607843,0.99264705
+RealTabPFN-2.5 (tuned + ensembled),0.21262255,0.38112745,0.44607842,0.43259802,0.51715684,,0.6666667,0.6642157,0.6936275,0.6237745,0.7352941,0.6642157,0.7610294,0.7022059,0.747549,0.7647059,0.74509805,0.79289216,0.7977941,0.79901963,0.75980395,0.79656863,0.6985294,0.8088235,0.7922794,0.8204657,0.81740195,0.82843137,0.84068626,0.82843137,0.8247549,0.8088235,0.8517157,0.8425245,0.78125,0.8272059,0.78125,0.85294116,0.8333333,0.84865195,0.8566176,0.8872549,0.8737745,0.8627451,0.8664216,0.88235295,0.87806374,0.877451,0.8682598,0.8872549,0.87990195,0.92034316,0.87990195,0.79901963,0.8419118,0.93014705,0.90931374,0.91789216,0.9485294,0.82965684,0.82965684,0.93259805,0.9289216,0.95710784,0.95710784,0.93014705,0.9362745,0.9362745,0.9583333,0.8639706,0.9558824,0.9522059,0.92034316,0.96568626,0.93014705,0.9313725,0.9742647,0.94914216,0.9485294,0.9914216
+RealTabPFN-2.5 (tuned),0.20220588,0.32965687,0.39093137,0.39093137,0.4564951,0.33333334,,0.57598037,0.64215684,0.5894608,0.6752451,0.6213235,0.7267157,0.6623775,0.70710784,0.71384805,0.7083333,0.74509805,0.7585784,0.7800245,0.7242647,0.76409316,0.67401963,0.7647059,0.75,0.79534316,0.79105395,0.7941176,0.80943626,0.7855392,0.79044116,0.77389705,0.79901963,0.8063725,0.752451,0.810049,0.7561275,0.8308824,0.8125,0.80943626,0.8370098,0.8639706,0.8541667,0.8394608,0.83455884,0.8664216,0.85294116,0.8670343,0.8504902,0.8639706,0.86519605,0.9044118,0.875,0.7757353,0.81740195,0.9117647,0.90318626,0.8897059,0.9375,0.80514705,0.810049,0.9117647,0.91789216,0.9558824,0.93811274,0.91544116,0.93014705,0.9264706,0.94485295,0.8566176,0.9387255,0.94485295,0.90931374,0.9558824,0.91789216,0.92401963,0.9742647,0.95159316,0.95098037,0.98651963
+RealTabPFN-2.5 (default),0.16421568,0.29289216,0.36519608,0.34742647,0.38602942,0.33578432,0.4240196,,0.5894608,0.5502451,0.65931374,0.57965684,0.7132353,0.62806374,0.6764706,0.6642157,0.70343137,0.73039216,0.747549,0.7585784,0.72794116,0.7708333,0.6691176,0.7647059,0.7267157,0.7855392,0.78431374,0.7745098,0.7647059,0.79044116,0.77818626,0.73651963,0.7830882,0.8039216,0.76348037,0.8137255,0.7610294,0.8308824,0.81311274,0.8161765,0.81985295,0.8639706,0.8584559,0.84436274,0.8272059,0.8602941,0.8480392,0.87009805,0.8737745,0.8664216,0.8639706,0.9044118,0.88235295,0.7536765,0.7610294,0.9007353,0.90563726,0.8762255,0.9338235,0.8308824,0.83026963,0.9117647,0.9362745,0.97181374,0.935049,0.9319853,0.939951,0.9338235,0.94240195,0.8713235,0.94240195,0.9485294,0.92156863,0.95465684,0.9080882,0.9264706,0.96936274,0.9620098,0.94730395,0.9852941
+RealMLP (tuned + ensembled),0.13480392,0.25490198,0.22242647,0.28431374,0.30882353,0.30637255,0.35784313,0.4105392,,0.49754903,0.53125,0.4908088,0.61519605,0.5349265,0.6942402,0.528799,0.5735294,0.70343137,0.71936274,0.68259805,0.59068626,0.5955882,0.5882353,0.7254902,0.74019605,0.7579657,0.76715684,0.79534316,0.6262255,0.7438725,0.64093137,0.70343137,0.7291667,0.8137255,0.67156863,0.77205884,0.64460784,0.814951,0.7316176,0.76960784,0.7977941,0.78186274,0.82598037,0.79656863,0.79105395,0.90134805,0.81740195,0.8480392,0.8474265,0.814951,0.83884805,0.83210784,0.85723037,0.7708333,0.80759805,0.9105392,0.86090684,0.9117647,0.92769605,0.78186274,0.7855392,0.9313725,0.84068626,0.9362745,0.9436275,0.872549,0.9117647,0.9387255,0.95098037,0.8370098,0.9607843,0.9522059,0.90563726,0.96568626,0.9252451,0.90686274,0.9754902,0.92034316,0.9460784,0.9938725
+TabDPT-Turbo (default),0.12009804,0.2254902,0.32843137,0.26838234,0.35661766,0.3762255,0.4105392,0.4497549,0.502451,,0.5563725,0.5147059,0.57414216,0.5808824,0.58394605,0.560049,0.59436274,0.6286765,0.66544116,0.63235295,0.65931374,0.7077206,0.70281863,0.6642157,0.7022059,0.66544116,0.6936275,0.6875,0.62990195,0.6936275,0.6458333,0.685049,0.70465684,0.7022059,0.6629902,0.73651963,0.69240195,0.7169118,0.6862745,0.74509805,0.7120098,0.7757353,0.79840684,0.79901963,0.76715684,0.80514705,0.74693626,0.84865195,0.8137255,0.7916667,0.79044116,0.82843137,0.8713235,0.75490195,0.77818626,0.85539216,0.8627451,0.8713235,0.88235295,0.7647059,0.76715684,0.88235295,0.8995098,0.9044118,0.9019608,0.92769605,0.8811275,0.90686274,0.92769605,0.8455882,0.9454657,0.9387255,0.872549,0.9620098,0.9252451,0.8811275,0.96936274,0.9436275,0.95710784,0.99019605
+"AutoGluon 1.4 (best, 4h)",0.11029412,0.24877451,0.19485295,0.25490198,0.2904412,0.2647059,0.3247549,0.34068626,0.46875,0.44362745,,0.44362745,0.56740195,0.47549018,0.5882353,0.4522059,0.4865196,0.7205882,0.6862745,0.6372549,0.54534316,0.5814951,0.5563725,0.70465684,0.70098037,0.75980395,0.6917892,0.76960784,0.5416667,0.75,0.5637255,0.6887255,0.6997549,0.76960784,0.61764705,0.7377451,0.64705884,0.7463235,0.70710784,0.79718137,0.7732843,0.8186275,0.80759805,0.7855392,0.814951,0.7977941,0.79289216,0.7830882,0.8247549,0.8504902,0.82843137,0.85784316,0.8063725,0.72181374,0.74264705,0.9105392,0.8627451,0.8566176,0.92401963,0.7769608,0.7794118,0.8817402,0.8143382,0.92156863,0.9362745,0.8762255,0.9129902,0.9044118,0.9375,0.83210784,0.93014705,0.94730395,0.9025735,0.9411765,0.9044118,0.91973037,0.95955884,0.9411765,0.9375,0.99019605
+TabDPT (tuned + ensembled),0.11764706,0.24142157,0.2622549,0.23345588,0.3272059,0.33578432,0.37867647,0.42034313,0.50919116,0.4852941,0.5563725,,0.5729167,0.65931374,0.5827206,0.55269605,0.5772059,0.63235295,0.6372549,0.62438726,0.59865195,0.6439951,0.70465684,0.64705884,0.69730395,0.6617647,0.69914216,0.67340684,0.6415441,0.6862745,0.63848037,0.6936275,0.6960784,0.70343137,0.6605392,0.721201,0.6317402,0.7101716,0.6629902,0.73039216,0.7077206,0.74938726,0.7892157,0.778799,0.75,0.77818626,0.73223037,0.81740195,0.77818626,0.7879902,0.7487745,0.810049,0.8370098,0.78431374,0.80514705,0.8419118,0.8247549,0.87009805,0.877451,0.7512255,0.752451,0.8786765,0.8762255,0.89460784,0.9117647,0.89093137,0.8713235,0.9172794,0.9209559,0.8357843,0.95098037,0.92034316,0.84068626,0.9558824,0.9172794,0.8541667,0.95710784,0.9044118,0.9436275,0.9938725
+TabM (tuned + ensembled),0.09681372,0.20220588,0.1740196,0.21323529,0.22426471,0.2389706,0.27328432,0.2867647,0.38480392,0.42585784,0.43259802,0.42708334,,0.46936274,0.49754903,0.44117647,0.4742647,0.5919118,0.6017157,0.71936274,0.50980395,0.53431374,0.54044116,0.6188725,0.60110295,0.65134805,0.59681374,0.6727941,0.5355392,0.6838235,0.5398284,0.6286765,0.64031863,0.7120098,0.6237745,0.625,0.61519605,0.86090684,0.7205882,0.72181374,0.73284316,0.73406863,0.7267157,0.7267157,0.7585784,0.8229167,0.7769608,0.74264705,0.7757353,0.80759805,0.78615195,0.8143382,0.7487745,0.6862745,0.7083333,0.85784316,0.8308824,0.7867647,0.8829657,0.7732843,0.7757353,0.8210784,0.80759805,0.9227941,0.8860294,0.85784316,0.90563726,0.85906863,0.90563726,0.82965684,0.89215684,0.91789216,0.90318626,0.92034316,0.89093137,0.9148284,0.9558824,0.935049,0.92769605,0.9754902
+TabDPT (tuned),0.104166664,0.20833333,0.2389706,0.23161764,0.2977941,0.2977941,0.33762255,0.37193626,0.46507353,0.41911766,0.5245098,0.34068626,0.53063726,,0.5257353,0.52144605,0.5612745,0.5857843,0.5955882,0.5870098,0.57598037,0.5882353,0.6458333,0.6060049,0.6311275,0.622549,0.6544118,0.6439951,0.5998775,0.64215684,0.5974265,0.64644605,0.64705884,0.6544118,0.6139706,0.6691176,0.5900735,0.67034316,0.6213235,0.6960784,0.6666667,0.70955884,0.73406863,0.7242647,0.70955884,0.73406863,0.7058824,0.7732843,0.73651963,0.747549,0.7242647,0.76348037,0.7916667,0.7377451,0.77205884,0.7855392,0.79534316,0.8112745,0.83210784,0.7224265,0.72181374,0.8394608,0.8713235,0.872549,0.8933824,0.8786765,0.83455884,0.8933824,0.9019608,0.8235294,0.9227941,0.90563726,0.82659316,0.93014705,0.9105392,0.8419118,0.9558824,0.89215684,0.935049,0.9914216
+RealMLP (tuned),0.125,0.22058824,0.18137255,0.24754901,0.2365196,0.25245097,0.29289216,0.32352942,0.30575982,0.41605392,0.4117647,0.41727942,0.502451,0.4742647,,0.45281863,0.4865196,0.5833333,0.5882353,0.5857843,0.53125,0.5183824,0.5355392,0.59681374,0.59375,0.64215684,0.62009805,0.6764706,0.54044116,0.64215684,0.5686275,0.60784316,0.64215684,0.7040441,0.60784316,0.67218137,0.5894608,0.7316176,0.689951,0.6666667,0.7205882,0.73406863,0.73039216,0.7414216,0.70343137,0.80269605,0.7585784,0.7622549,0.7732843,0.75,0.7585784,0.80514705,0.78431374,0.70710784,0.71936274,0.8455882,0.82965684,0.84436274,0.8566176,0.76593137,0.76715684,0.8627451,0.79718137,0.90563726,0.9007353,0.8584559,0.8762255,0.8811275,0.9080882,0.8210784,0.8897059,0.9307598,0.86519605,0.92034316,0.8897059,0.87193626,0.9620098,0.8995098,0.9166667,0.98284316
+TabPFNv2 (tuned + ensembled),0.12009804,0.22058824,0.28431374,0.25857842,0.30882353,0.23529412,0.28615198,0.33578432,0.47120097,0.43995097,0.5477941,0.44730392,0.5588235,0.47855392,0.54718137,,0.5232843,0.58639705,0.5980392,0.60110295,0.5232843,0.55759805,0.51960784,0.6047794,0.6139706,0.6164216,0.6458333,0.6378676,0.64705884,0.64215684,0.64644605,0.6519608,0.64705884,0.6519608,0.57659316,0.6599265,0.5949755,0.6507353,0.68259805,0.68259805,0.6954657,0.67769605,0.6911765,0.6623775,0.6936275,0.7120098,0.71936274,0.6917892,0.74080884,0.7205882,0.74509805,0.73284316,0.70465684,0.65625,0.68688726,0.76715684,0.76776963,0.7414216,0.7732843,0.7144608,0.71568626,0.7794118,0.7689951,0.8088235,0.80269605,0.8106618,0.8063725,0.8602941,0.8272059,0.7769608,0.89093137,0.86519605,0.85539216,0.86519605,0.8541667,0.8713235,0.9252451,0.8860294,0.9080882,0.9938725
+Mitra (default),0.106617644,0.22671568,0.2555147,0.24938725,0.25245097,0.25490198,0.29166666,0.29656863,0.42647058,0.40563726,0.51348037,0.4227941,0.5257353,0.4387255,0.51348037,0.47671568,,0.5735294,0.5735294,0.5955882,0.5324755,0.5355392,0.5122549,0.5729167,0.60355395,0.61151963,0.58210784,0.6145833,0.55943626,0.64215684,0.5735294,0.63480395,0.6433824,0.6237745,0.56740195,0.6207108,0.6041667,0.6482843,0.6666667,0.689951,0.71568626,0.71936274,0.6764706,0.685049,0.70098037,0.6936275,0.74509805,0.6789216,0.75735295,0.7205882,0.7536765,0.7708333,0.6917892,0.6482843,0.6642157,0.7745098,0.8002451,0.7107843,0.76348037,0.75490195,0.7512255,0.7567402,0.77389705,0.81985295,0.7977941,0.8235294,0.82598037,0.84681374,0.8161765,0.7941176,0.8480392,0.8639706,0.86764705,0.8639706,0.8504902,0.8835784,0.9313725,0.9105392,0.9099265,0.99019605
+LightGBM (tuned + ensembled),0.08455882,0.1740196,0.1384804,0.19117647,0.19240196,0.20710784,0.25490198,0.26960784,0.29656863,0.37132353,0.27941176,0.36764705,0.40808824,0.41421568,0.41666666,0.41360295,0.42647058,,0.5183824,0.4944853,0.44485295,0.5257353,0.51348037,0.5520833,0.5275735,0.6550245,0.5539216,0.68014705,0.49019608,0.6709559,0.4987745,0.54656863,0.5772059,0.7169118,0.5563725,0.61519605,0.57843137,0.68811274,0.6066176,0.7224265,0.7040441,0.6397059,0.67156863,0.69240195,0.76348037,0.7230392,0.747549,0.7555147,0.7144608,0.83210784,0.81740195,0.73406863,0.7585784,0.70710784,0.7083333,0.8958333,0.79044116,0.83455884,0.95710784,0.75919116,0.75980395,0.8737745,0.8039216,0.8492647,0.9142157,0.8357843,0.9191176,0.83210784,0.9289216,0.82843137,0.91789216,0.88235295,0.8811275,0.9485294,0.8982843,0.8958333,0.95955884,0.9362745,0.93443626,0.98651963
+CatBoost (tuned + ensembled),0.08455882,0.16544117,0.1360294,0.18872549,0.18627451,0.20220588,0.24142157,0.25245097,0.28063726,0.3345588,0.3137255,0.3627451,0.39828432,0.40441176,0.4117647,0.4019608,0.42647058,0.48161766,,0.46568626,0.45588234,0.51715684,0.47181374,0.5949755,0.5232843,0.55085784,0.54044116,0.6268382,0.47916666,0.6482843,0.49019608,0.5747549,0.5851716,0.6574755,0.5808824,0.62009805,0.57965684,0.6372549,0.6164216,0.6875,0.6642157,0.65563726,0.69485295,0.67401963,0.7242647,0.70710784,0.7064951,0.7561275,0.69240195,0.78063726,0.7561275,0.7438725,0.778799,0.70343137,0.6960784,0.8872549,0.7610294,0.84313726,0.89215684,0.7585784,0.75490195,0.87806374,0.7745098,0.8566176,0.875,0.8039216,0.8854167,0.85539216,0.8933824,0.8112745,0.9019608,0.877451,0.85906863,0.94240195,0.8995098,0.8713235,0.93259805,0.8958333,0.9264706,0.9877451
+TabM (tuned),0.073529415,0.1629902,0.16115196,0.19117647,0.18504901,0.2009804,0.21997549,0.24142157,0.31740198,0.36764705,0.3627451,0.37561274,0.28063726,0.41299018,0.41421568,0.39889705,0.40441176,0.5055147,0.53431374,,0.48713234,0.48284313,0.4944853,0.5502451,0.53186274,0.5502451,0.5269608,0.58026963,0.48039216,0.60294116,0.5,0.56740195,0.5833333,0.6041667,0.54656863,0.5735294,0.5612745,0.7205882,0.64460784,0.65931374,0.6433824,0.67401963,0.65563726,0.67034316,0.6629902,0.76348037,0.69730395,0.69240195,0.72365195,0.72855395,0.73284316,0.7542892,0.70955884,0.6617647,0.6752451,0.80514705,0.79289216,0.7512255,0.8002451,0.7561275,0.7604167,0.76960784,0.7769608,0.9044118,0.86519605,0.8357843,0.88480395,0.8186275,0.875,0.810049,0.84436274,0.877451,0.8811275,0.88848037,0.86151963,0.8897059,0.9375,0.91789216,0.9080882,0.96813726
+LimiX (default),0.09375,0.21568628,0.22303921,0.21384804,0.24877451,0.24019608,0.2757353,0.2720588,0.40931374,0.34068626,0.45465687,0.40134802,0.49019608,0.4240196,0.46875,0.47671568,0.4675245,0.55514705,0.5441176,0.5128676,,0.48406863,0.46078432,0.54656863,0.57659316,0.57843137,0.54534316,0.5882353,0.5379902,0.60906863,0.54656863,0.56985295,0.60539216,0.6041667,0.54656863,0.56556374,0.58455884,0.5870098,0.61151963,0.67401963,0.67034316,0.6397059,0.61764705,0.66544116,0.6887255,0.6642157,0.6985294,0.6985294,0.68688726,0.7022059,0.7022059,0.6997549,0.7022059,0.6985294,0.6985294,0.7536765,0.7542892,0.752451,0.7438725,0.72855395,0.7254902,0.7708333,0.76409316,0.78186274,0.78063726,0.7610294,0.8039216,0.8308824,0.79534316,0.7775735,0.8137255,0.8394608,0.8566176,0.8602941,0.8370098,0.8713235,0.8713235,0.89093137,0.8762255,0.96813726
+TabSwift (default),0.113970585,0.11764706,0.16421568,0.1629902,0.2009804,0.20343137,0.23590687,0.22916667,0.40441176,0.29227942,0.4185049,0.3560049,0.46568626,0.4117647,0.48161766,0.44240198,0.4644608,0.4742647,0.48284313,0.51715684,0.51593137,,0.5061275,0.51164216,0.51960784,0.5208333,0.54289216,0.5441176,0.52818626,0.5367647,0.54656863,0.54534316,0.5710784,0.5833333,0.58210784,0.57965684,0.51593137,0.61764705,0.5980392,0.6017157,0.622549,0.6666667,0.6666667,0.67769605,0.63235295,0.7205882,0.6580882,0.685049,0.76715684,0.70465684,0.6789216,0.7254902,0.70710784,0.6813725,0.7083333,0.7389706,0.79289216,0.71813726,0.783701,0.7622549,0.76348037,0.7585784,0.875,0.85294116,0.8541667,0.90318626,0.83210784,0.83455884,0.872549,0.83455884,0.86151963,0.89460784,0.82598037,0.8897059,0.88235295,0.8455882,0.93259805,0.9172794,0.95343137,0.9938725
+TabDPT (default),0.1004902,0.17279412,0.23406863,0.20710784,0.27818626,0.30147058,0.3259804,0.33088234,0.4117647,0.29718137,0.44362745,0.29534313,0.4595588,0.35416666,0.4644608,0.48039216,0.4877451,0.4865196,0.52818626,0.5055147,0.5392157,0.49387255,,0.5392157,0.5539216,0.53431374,0.56985295,0.56740195,0.5416667,0.5477941,0.5563725,0.5625,0.5710784,0.5882353,0.5735294,0.59681374,0.53125,0.62009805,0.5747549,0.62990195,0.5833333,0.6335784,0.6495098,0.6642157,0.6397059,0.6378676,0.6121324,0.73039216,0.67156863,0.6813725,0.65686274,0.6911765,0.7512255,0.6764706,0.70465684,0.71813726,0.7316176,0.7536765,0.75980395,0.6887255,0.689951,0.7916667,0.8112745,0.8161765,0.82598037,0.88480395,0.8112745,0.8394608,0.86151963,0.7855392,0.88480395,0.8762255,0.77818626,0.8762255,0.875,0.79901963,0.92156863,0.8762255,0.90686274,0.99264705
+CatBoost (tuned),0.07230392,0.15931372,0.13296568,0.18504901,0.18504901,0.19117647,0.23529412,0.23529412,0.27450982,0.33578432,0.29534313,0.3529412,0.38112745,0.3939951,0.40318626,0.39522058,0.42708334,0.44791666,0.4050245,0.4497549,0.45343137,0.48835784,0.46078432,,0.49509802,0.53186274,0.52144605,0.60294116,0.4522059,0.60723037,0.47916666,0.560049,0.5637255,0.64093137,0.56985295,0.6017157,0.56740195,0.62561274,0.59375,0.6501225,0.65563726,0.6495098,0.6580882,0.65686274,0.69914216,0.69485295,0.6936275,0.73651963,0.67401963,0.76593137,0.75490195,0.7310049,0.7487745,0.689951,0.6911765,0.88235295,0.7518382,0.82843137,0.8664216,0.7585784,0.75735295,0.8602941,0.76960784,0.84068626,0.86519605,0.79656863,0.8762255,0.84681374,0.8860294,0.8125,0.88848037,0.86519605,0.85539216,0.9338235,0.89093137,0.86580884,0.9362745,0.89093137,0.93014705,0.98651963
+iLTM (tuned + ensembled),0.075980395,0.1360294,0.109068625,0.15686275,0.15808824,0.2077206,0.25,0.27328432,0.25980392,0.2977941,0.2990196,0.30269608,0.39889705,0.36887255,0.40625,0.38602942,0.39644608,0.47242647,0.47671568,0.46813726,0.42340687,0.48039216,0.44607842,0.50490195,,0.54534316,0.5496324,0.5680147,0.5,0.5502451,0.5085784,0.5514706,0.5894608,0.6164216,0.55759805,0.6182598,0.5373775,0.5710784,0.60539216,0.65318626,0.6360294,0.66544116,0.7371324,0.70098037,0.67156863,0.72794116,0.67401963,0.7420343,0.7352941,0.71875,0.7107843,0.74509805,0.7683824,0.6887255,0.7022059,0.8308824,0.78615195,0.8272059,0.87009805,0.72181374,0.7224265,0.85539216,0.7689951,0.8688725,0.8541667,0.8517157,0.88235295,0.89460784,0.88848037,0.8063725,0.90931374,0.89705884,0.8137255,0.92769605,0.877451,0.8394608,0.9313725,0.8786765,0.908701,0.98284316
+LightGBM (tuned),0.07230392,0.1384804,0.123774506,0.16176471,0.14705883,0.17953432,0.20465687,0.21446079,0.24203432,0.3345588,0.24019608,0.3382353,0.34865198,0.37745097,0.35784313,0.38357842,0.3884804,0.3449755,0.44914216,0.4497549,0.42156863,0.47916666,0.46568626,0.46813726,0.45465687,,0.49938726,0.5606618,0.46078432,0.5900735,0.48039216,0.502451,0.51960784,0.6041667,0.5183824,0.5502451,0.54534316,0.6397059,0.59313726,0.64705884,0.6642157,0.6066176,0.6066176,0.6507353,0.67769605,0.6942402,0.7144608,0.7077206,0.6807598,0.76960784,0.75980395,0.68811274,0.73223037,0.6875,0.689951,0.86090684,0.7414216,0.7769608,0.91605395,0.75,0.7512255,0.82843137,0.79044116,0.82843137,0.8768382,0.8210784,0.903799,0.8204657,0.8982843,0.8137255,0.8933824,0.86764705,0.86151963,0.9252451,0.8762255,0.8805147,0.95465684,0.9227941,0.9264706,0.9816176
+xRFM (tuned + ensembled),0.07965686,0.14705883,0.125,0.17524509,0.1764706,0.18259804,0.20894608,0.21568628,0.23284313,0.30637255,0.3082108,0.30085784,0.40318626,0.34558824,0.37990198,0.35416666,0.41789216,0.44607842,0.4595588,0.4730392,0.45465687,0.45710784,0.43014705,0.47855392,0.45036766,0.50061274,,0.5165441,0.46936274,0.5349265,0.46813726,0.5147059,0.5165441,0.54901963,0.5085784,0.64276963,0.52818626,0.5710784,0.5857843,0.61151963,0.6066176,0.5919118,0.6207108,0.622549,0.62990195,0.6397059,0.6605392,0.6838235,0.66360295,0.65931374,0.685049,0.6764706,0.7230392,0.7083333,0.71813726,0.75735295,0.7291667,0.76715684,0.7622549,0.716299,0.7144608,0.79901963,0.79901963,0.84681374,0.8161765,0.85784316,0.810049,0.9497549,0.8498775,0.78186274,0.935049,0.90563726,0.86764705,0.9411765,0.90625,0.87990195,0.92156863,0.903799,0.9289216,0.9889706
+XGBoost (tuned + ensembled),0.06495098,0.15134804,0.111519605,0.17524509,0.15073529,0.17156863,0.20588236,0.2254902,0.20465687,0.3125,0.23039216,0.32659313,0.3272059,0.3560049,0.32352942,0.36213234,0.38541666,0.31985295,0.37316176,0.4197304,0.4117647,0.45588234,0.43259802,0.3970588,0.4319853,0.43933824,0.4834559,,0.44914216,0.5275735,0.46813726,0.4779412,0.5183824,0.6047794,0.5147059,0.5686275,0.53431374,0.5857843,0.5735294,0.60294116,0.6237745,0.6121324,0.6102941,0.6397059,0.6397059,0.6642157,0.6789216,0.689951,0.66789216,0.7150735,0.73284316,0.6789216,0.7291667,0.6875,0.6960784,0.8995098,0.7205882,0.8180147,0.9105392,0.7377451,0.73406863,0.841299,0.7689951,0.8186275,0.83210784,0.8045343,0.8541667,0.8272059,0.86090684,0.7916667,0.8786765,0.8455882,0.8504902,0.935049,0.8664216,0.85906863,0.9227941,0.8897059,0.9148284,0.9791667
+TabPFNv2 (tuned),0.109068625,0.15992647,0.2009804,0.2077206,0.20710784,0.15931372,0.19056372,0.23529412,0.3737745,0.37009802,0.45833334,0.3584559,0.4644608,0.40012255,0.4595588,0.3529412,0.44056374,0.50980395,0.5208333,0.51960784,0.46200982,0.47181374,0.45833334,0.5477941,0.5,0.5392157,0.53063726,0.55085784,,0.5612745,0.53431374,0.5588235,0.564951,0.55514705,0.5079657,0.5502451,0.5422794,0.59313726,0.6268382,0.60539216,0.6335784,0.6164216,0.60723037,0.5925245,0.6262255,0.64644605,0.6629902,0.59436274,0.6746324,0.6544118,0.6789216,0.6727941,0.6262255,0.5925245,0.60968137,0.67401963,0.7144608,0.6311275,0.68259805,0.69240195,0.6887255,0.6917892,0.7077206,0.7536765,0.72794116,0.7751225,0.78186274,0.80514705,0.7432598,0.7512255,0.810049,0.8112745,0.8308824,0.79901963,0.8039216,0.84313726,0.8786765,0.86519605,0.8688725,0.9767157
+CatBoost (default),0.07720588,0.14583333,0.12071078,0.16176471,0.14828432,0.17156863,0.21446079,0.20955883,0.25612745,0.30637255,0.25,0.3137255,0.31617647,0.35784313,0.35784313,0.35784313,0.35784313,0.3290441,0.35171568,0.3970588,0.39093137,0.4632353,0.4522059,0.3927696,0.4497549,0.40992647,0.46507353,0.47242647,0.4387255,,0.46017158,0.4779412,0.5036765,0.5128676,0.5122549,0.55759805,0.51960784,0.57230395,0.5379902,0.62806374,0.6286765,0.5992647,0.61151963,0.59436274,0.6752451,0.65686274,0.6838235,0.6875,0.66544116,0.7757353,0.73284316,0.69240195,0.70465684,0.66789216,0.6691176,0.7830882,0.73284316,0.7769608,0.8161765,0.75061274,0.7512255,0.8063725,0.75,0.8125,0.8480392,0.8002451,0.90931374,0.7941176,0.87990195,0.814951,0.8664216,0.8762255,0.8394608,0.9019608,0.872549,0.85539216,0.92769605,0.90563726,0.93259805,0.98406863
+TabPFNv2 (default),0.11519608,0.1807598,0.19117647,0.1985294,0.22732843,0.17524509,0.20955883,0.22181372,0.35906863,0.35416666,0.4362745,0.3615196,0.46017158,0.40257353,0.43137255,0.35355392,0.42647058,0.5012255,0.50980395,0.5,0.45343137,0.45343137,0.44362745,0.5208333,0.49142158,0.51960784,0.53186274,0.53186274,0.46568626,0.5398284,,0.51715684,0.5330882,0.5416667,0.4877451,0.564951,0.5349265,0.5808824,0.6102941,0.5710784,0.62009805,0.58210784,0.56740195,0.58394605,0.58455884,0.63480395,0.6495098,0.6139706,0.6605392,0.5919118,0.6666667,0.6372549,0.6286765,0.6041667,0.61764705,0.6580882,0.70710784,0.64705884,0.67034316,0.70465684,0.70159316,0.70281863,0.74019605,0.75980395,0.73651963,0.814951,0.7610294,0.8063725,0.7487745,0.75980395,0.80514705,0.8112745,0.8523284,0.8137255,0.814951,0.85784316,0.8737745,0.89093137,0.8811275,0.9730392
+ModernNCA (tuned + ensembled),0.05759804,0.121323526,0.113970585,0.15686275,0.14093137,0.19117647,0.22610295,0.2634804,0.29656863,0.31495097,0.3112745,0.30637255,0.37132353,0.35355392,0.39215687,0.3480392,0.36519608,0.45343137,0.4252451,0.43259802,0.43014705,0.45465687,0.4375,0.43995097,0.44852942,0.49754903,0.4852941,0.52205884,0.44117647,0.52205884,0.48284313,,0.60110295,0.52818626,0.49142158,0.57843137,0.5085784,0.54289216,0.5882353,0.5772059,0.5980392,0.57843137,0.5808824,0.7107843,0.6164216,0.67156863,0.63664216,0.6629902,0.6519608,0.63480395,0.6544118,0.6580882,0.69240195,0.6495098,0.6580882,0.7622549,0.7107843,0.73039216,0.74264705,0.6789216,0.6764706,0.7414216,0.7132353,0.7769608,0.7830882,0.7745098,0.79656863,0.85968137,0.7916667,0.7377451,0.84313726,0.8210784,0.76715684,0.85539216,0.8235294,0.77818626,0.85784316,0.8137255,0.8517157,0.9583333
+ModernNCA (tuned),0.068627454,0.13480392,0.10294118,0.15196079,0.13970588,0.14828432,0.2009804,0.21691176,0.27083334,0.29534313,0.3002451,0.30392158,0.35968137,0.3529412,0.35784313,0.3529412,0.35661766,0.4227941,0.41482842,0.41666666,0.39460784,0.42892158,0.42892158,0.4362745,0.4105392,0.48039216,0.4834559,0.48161766,0.43504903,0.49632353,0.46691176,0.39889705,,0.49693626,0.49142158,0.5416667,0.51715684,0.5416667,0.57169116,0.5808824,0.58210784,0.5833333,0.56985295,0.69240195,0.59375,0.65318626,0.627451,0.6482843,0.6648284,0.6482843,0.6544118,0.6997549,0.6691176,0.65931374,0.67769605,0.71813726,0.7463235,0.7414216,0.7254902,0.72181374,0.7254902,0.75980395,0.76348037,0.8063725,0.79044116,0.8272059,0.8492647,0.85539216,0.82598037,0.7879902,0.84681374,0.8645833,0.82230395,0.877451,0.8627451,0.84436274,0.8958333,0.88848037,0.90563726,0.97794116
+XGBoost (tuned),0.05882353,0.14093137,0.10539216,0.17034313,0.12867647,0.15747549,0.19362745,0.19607843,0.18627451,0.2977941,0.23039216,0.29656863,0.28799018,0.34558824,0.2959559,0.3480392,0.3762255,0.28308824,0.3425245,0.39583334,0.39583334,0.41666666,0.4117647,0.35906863,0.38357842,0.39583334,0.4509804,0.39522058,0.44485295,0.48713234,0.45833334,0.47181374,0.50306374,,0.50735295,0.5324755,0.5147059,0.5514706,0.55759805,0.54840684,0.5980392,0.5857843,0.5747549,0.60784316,0.59436274,0.6360294,0.65686274,0.6832108,0.6433824,0.6727941,0.7022059,0.65931374,0.7169118,0.6752451,0.68259805,0.85294116,0.71568626,0.7892157,0.86764705,0.7242647,0.72610295,0.81985295,0.7512255,0.8063725,0.80514705,0.79289216,0.8480392,0.8161765,0.8455882,0.7892157,0.8762255,0.8357843,0.8419118,0.92034316,0.86335784,0.85784316,0.92156863,0.88480395,0.9105392,0.9767157
+SAP-RPT-OSS (default),0.121323526,0.20343137,0.22794117,0.21936275,0.20833333,0.21875,0.24754901,0.2365196,0.32843137,0.33700982,0.38235295,0.3394608,0.3762255,0.38602942,0.39215687,0.42340687,0.43259802,0.44362745,0.41911766,0.45343137,0.45343137,0.41789216,0.42647058,0.43014705,0.44240198,0.48161766,0.49142158,0.4852941,0.49203432,0.4877451,0.5122549,0.5085784,0.5085784,0.49264705,,0.49509802,0.50980395,0.51593137,0.55269605,0.5232843,0.5563725,0.5857843,0.5300245,0.5833333,0.5441176,0.622549,0.59313726,0.625,0.67034316,0.59436274,0.6329657,0.6262255,0.6433824,0.6004902,0.6127451,0.64705884,0.7267157,0.6770833,0.68014705,0.6507353,0.6458333,0.68811274,0.69485295,0.75490195,0.7683824,0.7745098,0.7267157,0.7622549,0.7775735,0.79534316,0.79044116,0.8161765,0.8247549,0.79901963,0.79044116,0.8308824,0.8382353,0.8627451,0.88848037,0.9852941
+xRFM (tuned),0.08455882,0.1360294,0.13051471,0.17953432,0.15686275,0.17279412,0.18995099,0.18627451,0.22794117,0.2634804,0.2622549,0.27879903,0.375,0.33088234,0.32781863,0.34007353,0.3792892,0.38480392,0.37990198,0.42647058,0.43443626,0.42034313,0.40318626,0.39828432,0.38174018,0.4497549,0.3572304,0.43137255,0.4497549,0.44240198,0.43504903,0.42156863,0.45833334,0.4675245,0.50490195,,0.51715684,0.53186274,0.5747549,0.54473037,0.56311274,0.56985295,0.5661765,0.59313726,0.5870098,0.5992647,0.6164216,0.6372549,0.6305147,0.6237745,0.6397059,0.6458333,0.685049,0.6495098,0.6495098,0.6764706,0.7058824,0.7022059,0.70465684,0.7169118,0.71936274,0.7438725,0.7745098,0.8382353,0.7800245,0.86151963,0.8137255,0.9019608,0.8204657,0.79044116,0.8688725,0.8860294,0.8627451,0.89093137,0.8933824,0.87990195,0.9080882,0.91789216,0.92769605,0.98651963
+TabICL (default),0.04779412,0.114583336,0.13541667,0.12071078,0.19117647,0.21875,0.24387255,0.2389706,0.35539216,0.30759802,0.3529412,0.36825982,0.38480392,0.40992647,0.4105392,0.4050245,0.39583334,0.42156863,0.42034313,0.4387255,0.4154412,0.48406863,0.46875,0.43259802,0.46262255,0.45465687,0.47181374,0.46568626,0.45772058,0.48039216,0.46507353,0.49142158,0.48284313,0.4852941,0.49019608,0.48284313,,0.52205884,0.5367647,0.5514706,0.5502451,0.5379902,0.54901963,0.54473037,0.57965684,0.5735294,0.5833333,0.6017157,0.61151963,0.60784316,0.61764705,0.6237745,0.6145833,0.6936275,0.6911765,0.6372549,0.65318626,0.6041667,0.67830884,0.70343137,0.70098037,0.6838235,0.73651963,0.7144608,0.7389706,0.752451,0.8125,0.73651963,0.75735295,0.7745098,0.7879902,0.8180147,0.8357843,0.82598037,0.87009805,0.8480392,0.8566176,0.92830884,0.92156863,0.9730392
+TabM (default),0.06740196,0.12990196,0.116421565,0.14338236,0.1317402,0.14705883,0.16911764,0.16911764,0.18504901,0.28308824,0.25367647,0.28982842,0.13909313,0.32965687,0.26838234,0.3492647,0.35171568,0.31188726,0.3627451,0.27941176,0.41299018,0.38235295,0.37990198,0.37438726,0.42892158,0.3602941,0.42892158,0.41421568,0.40686274,0.42769608,0.41911766,0.45710784,0.45833334,0.44852942,0.48406863,0.46813726,0.4779412,,0.5,0.4779412,0.502451,0.57230395,0.53615195,0.56740195,0.50490195,0.6102941,0.5710784,0.6188725,0.6433824,0.57598037,0.5998775,0.6580882,0.6397059,0.61151963,0.6139706,0.67156863,0.7242647,0.6875,0.6544118,0.7352941,0.7377451,0.6985294,0.7120098,0.79656863,0.7628676,0.76348037,0.8014706,0.7291667,0.7745098,0.8125,0.7916667,0.8247549,0.8272059,0.83210784,0.8186275,0.83884805,0.88235295,0.88235295,0.9007353,0.9522059
+BetaTabPFN (default),0.08884804,0.1697304,0.16176471,0.1764706,0.18014705,0.16666667,0.1875,0.18688725,0.26838234,0.3137255,0.29289216,0.33700982,0.27941176,0.37867647,0.31004903,0.31740198,0.33333334,0.39338234,0.38357842,0.35539216,0.3884804,0.4019608,0.4252451,0.40625,0.39460784,0.40686274,0.41421568,0.42647058,0.37316176,0.46200982,0.3897059,0.4117647,0.4283088,0.44240198,0.44730392,0.4252451,0.4632353,0.5,,0.5042892,0.56740195,0.5245098,0.46936274,0.49509802,0.52818626,0.5661765,0.61151963,0.5367647,0.5851716,0.57598037,0.65318626,0.5992647,0.55269605,0.6311275,0.6286765,0.5870098,0.6482843,0.55759805,0.6023284,0.73039216,0.7291667,0.625,0.67401963,0.70098037,0.70343137,0.73590684,0.8210784,0.67401963,0.7267157,0.80085784,0.71936274,0.8002451,0.8860294,0.76348037,0.8602941,0.8995098,0.8333333,0.94730395,0.9338235,0.96936274
+ChimeraBoost (tuned + ensembled),0.06495098,0.113970585,0.083333336,0.14338236,0.10539216,0.15134804,0.19056372,0.18382353,0.23039216,0.25490198,0.20281863,0.26960784,0.27818626,0.30392158,0.33333334,0.31740198,0.31004903,0.27757353,0.3125,0.34068626,0.3259804,0.39828432,0.37009802,0.34987745,0.34681374,0.3529412,0.3884804,0.3970588,0.39460784,0.37193626,0.42892158,0.4227941,0.41911766,0.45159313,0.47671568,0.4552696,0.44852942,0.52205884,0.4957108,,0.5539216,0.52818626,0.51593137,0.47181374,0.60723037,0.56985295,0.5955882,0.5747549,0.5612745,0.79656863,0.6311275,0.5919118,0.6102941,0.5919118,0.5955882,0.67769605,0.6084559,0.67769605,0.73223037,0.70955884,0.70465684,0.71568626,0.7083333,0.7169118,0.7867647,0.7867647,0.81740195,0.74264705,0.814951,0.7855392,0.7879902,0.8161765,0.7855392,0.8210784,0.79534316,0.8002451,0.87009805,0.87009805,0.8860294,0.97365195
+EBM (tuned + ensembled),0.073529415,0.14093137,0.11029412,0.15441176,0.1384804,0.14338236,0.1629902,0.18014705,0.20220588,0.28799018,0.22671568,0.29227942,0.26715687,0.33333334,0.27941176,0.30453432,0.28431374,0.2959559,0.33578432,0.35661766,0.32965687,0.37745097,0.41666666,0.34436274,0.36397058,0.33578432,0.39338234,0.3762255,0.36642158,0.37132353,0.37990198,0.4019608,0.41789216,0.4019608,0.44362745,0.43688726,0.4497549,0.49754903,0.43259802,0.44607842,,0.5257353,0.48284313,0.5257353,0.4865196,0.5625,0.721201,0.6023284,0.55085784,0.560049,0.7683824,0.61580884,0.6127451,0.6188725,0.6286765,0.6415441,0.6231618,0.6580882,0.6360294,0.7230392,0.7144608,0.6838235,0.6838235,0.7463235,0.75490195,0.7463235,0.80269605,0.7083333,0.7726716,0.810049,0.7585784,0.83210784,0.86151963,0.79534316,0.82965684,0.875,0.88480395,0.9007353,0.8878676,0.9522059
+TorchMLP (tuned + ensembled),0.040441178,0.08946078,0.0747549,0.12009804,0.104166664,0.1127451,0.1360294,0.1360294,0.21813725,0.22426471,0.18137255,0.25061274,0.26593137,0.2904412,0.26593137,0.32230392,0.28063726,0.3602941,0.34436274,0.3259804,0.3602941,0.33333334,0.36642158,0.35049018,0.3345588,0.39338234,0.40808824,0.38786766,0.38357842,0.4007353,0.41789216,0.42156863,0.41666666,0.41421568,0.41421568,0.43014705,0.46200982,0.42769608,0.47549018,0.47181374,0.4742647,,0.5067402,0.5441176,0.5085784,0.55514705,0.5392157,0.63235295,0.63848037,0.52818626,0.57965684,0.7487745,0.64705884,0.5919118,0.59313726,0.65318626,0.7432598,0.6875,0.6482843,0.70343137,0.7101716,0.7120098,0.6727941,0.89215684,0.72181374,0.7794118,0.8069853,0.7683824,0.7708333,0.8002451,0.7647059,0.8584559,0.81188726,0.79656863,0.81985295,0.8370098,0.88235295,0.8982843,0.8933824,0.95955884
+iLTM (tuned),0.05759804,0.07414216,0.08455882,0.125,0.09313726,0.12622549,0.14583333,0.14154412,0.1740196,0.20159313,0.19240196,0.21078432,0.27328432,0.26593137,0.26960784,0.30882353,0.32352942,0.32843137,0.30514705,0.34436274,0.38235295,0.33333334,0.35049018,0.34191176,0.26286766,0.39338234,0.3792892,0.3897059,0.3927696,0.3884804,0.43259802,0.41911766,0.43014705,0.4252451,0.4699755,0.43382353,0.4509804,0.46384802,0.53063726,0.48406863,0.51715684,0.49325982,,0.5606618,0.5349265,0.5710784,0.56985295,0.59436274,0.6188725,0.5747549,0.6041667,0.6041667,0.64460784,0.5980392,0.59681374,0.6666667,0.67034316,0.6960784,0.69240195,0.6917892,0.6887255,0.7291667,0.69240195,0.810049,0.7230392,0.77634805,0.8106618,0.8112745,0.7708333,0.75980395,0.8161765,0.84681374,0.76164216,0.84681374,0.814951,0.7916667,0.86519605,0.8492647,0.8731618,0.9620098
+ModernNCA (default),0.06127451,0.094975494,0.104166664,0.11580882,0.101715684,0.13725491,0.16053921,0.15563725,0.20343137,0.2009804,0.21446079,0.22120099,0.27328432,0.2757353,0.25857842,0.33762255,0.31495097,0.30759802,0.3259804,0.32965687,0.3345588,0.32230392,0.33578432,0.34313726,0.2990196,0.3492647,0.37745097,0.3602941,0.4074755,0.40563726,0.41605392,0.28921568,0.30759802,0.39215687,0.41666666,0.40686274,0.4552696,0.43259802,0.50490195,0.52818626,0.4742647,0.45588234,0.43933824,,0.5441176,0.49264705,0.51960784,0.5686275,0.5477941,0.5857843,0.5637255,0.5416667,0.6041667,0.622549,0.63235295,0.5882353,0.62009805,0.6617647,0.6237745,0.6789216,0.6838235,0.72181374,0.752451,0.7732843,0.7242647,0.810049,0.79289216,0.7647059,0.7769608,0.7794118,0.81311274,0.8247549,0.7977941,0.85784316,0.86519605,0.8180147,0.8639706,0.8982843,0.90931374,0.98039216
+ChimeraBoost (tuned),0.06004902,0.113970585,0.075980395,0.13357843,0.09681372,0.13357843,0.16544117,0.17279412,0.20894608,0.23284313,0.18504901,0.25,0.24142157,0.2904412,0.29656863,0.30637255,0.2990196,0.2365196,0.2757353,0.33700982,0.3112745,0.36764705,0.3602941,0.30085784,0.32843137,0.32230392,0.37009802,0.3602941,0.3737745,0.3247549,0.4154412,0.38357842,0.40625,0.40563726,0.45588234,0.41299018,0.42034313,0.49509802,0.47181374,0.3927696,0.51348037,0.49142158,0.46507353,0.45588234,,0.55269605,0.5667892,0.55269605,0.5379902,0.71568626,0.5955882,0.5625,0.59313726,0.57230395,0.5772059,0.6501225,0.5876225,0.6495098,0.6862745,0.69730395,0.6960784,0.689951,0.6862745,0.70710784,0.74693626,0.7512255,0.7916667,0.72181374,0.77634805,0.7708333,0.7769608,0.7941176,0.7732843,0.79044116,0.7769608,0.7794118,0.86335784,0.8566176,0.8639706,0.9620098
+RealMLP (default),0.05269608,0.0747549,0.0882353,0.09803922,0.09068628,0.11764706,0.13357843,0.13970588,0.09865196,0.19485295,0.20220588,0.22181372,0.17708333,0.26593137,0.19730392,0.28799018,0.30637255,0.2769608,0.29289216,0.2365196,0.33578432,0.27941176,0.36213234,0.30514705,0.2720588,0.30575982,0.3602941,0.33578432,0.35355392,0.34313726,0.36519608,0.32843137,0.34681374,0.36397058,0.37745097,0.4007353,0.42647058,0.3897059,0.43382353,0.43014705,0.4375,0.44485295,0.42892158,0.50735295,0.44730392,,0.50490195,0.5735294,0.5422794,0.5110294,0.53431374,0.5833333,0.6066176,0.5882353,0.5955882,0.6139706,0.65318626,0.65318626,0.6102941,0.67401963,0.6813725,0.6875,0.65931374,0.77634805,0.71568626,0.78615195,0.77205884,0.7205882,0.77205884,0.77818626,0.79534316,0.872549,0.79534316,0.8014706,0.82843137,0.82598037,0.8762255,0.8872549,0.90686274,0.9791667
+EBM (tuned),0.0625,0.123774506,0.101715684,0.1360294,0.10845588,0.12193628,0.14705883,0.15196079,0.18259804,0.25306374,0.20710784,0.2677696,0.22303921,0.29411766,0.24142157,0.28063726,0.25490198,0.25245097,0.2935049,0.30269608,0.30147058,0.34191176,0.38786766,0.30637255,0.3259804,0.2855392,0.3394608,0.32107842,0.33700982,0.31617647,0.35049018,0.36335784,0.37254903,0.34313726,0.40686274,0.38357842,0.41666666,0.42892158,0.3884804,0.40441176,0.27879903,0.46078432,0.43014705,0.48039216,0.4332108,0.49509802,,0.5477941,0.502451,0.5036765,0.6623775,0.56740195,0.58210784,0.6017157,0.6164216,0.5919118,0.5949755,0.61151963,0.58455884,0.70343137,0.7003676,0.6519608,0.6452206,0.7169118,0.72181374,0.71875,0.75980395,0.6862745,0.7463235,0.79534316,0.74264705,0.79901963,0.83455884,0.7683824,0.80269605,0.8504902,0.8639706,0.88235295,0.8860294,0.9460784
+ExtraTrees (tuned + ensembled),0.04656863,0.06495098,0.075980395,0.102328435,0.09191176,0.12254902,0.13296568,0.12990196,0.15196079,0.15134804,0.21691176,0.18259804,0.25735295,0.22671568,0.23774509,0.3082108,0.32107842,0.24448529,0.24387255,0.30759802,0.30147058,0.31495097,0.26960784,0.2634804,0.25796568,0.29227942,0.31617647,0.31004903,0.40563726,0.3125,0.38602942,0.33700982,0.35171568,0.3167892,0.375,0.3627451,0.39828432,0.38112745,0.4632353,0.4252451,0.39767158,0.36764705,0.40563726,0.43137255,0.44730392,0.42647058,0.4522059,,0.49142158,0.49754903,0.48039216,0.44424018,0.6807598,0.6458333,0.6482843,0.51960784,0.5392157,0.6911765,0.5379902,0.5955882,0.60539216,0.7438725,0.7022059,0.6127451,0.6507353,0.74509805,0.6985294,0.7132353,0.6838235,0.73223037,0.9191176,0.75,0.7420343,0.8860294,0.8688725,0.7579657,0.78063726,0.8308824,0.9129902,0.9914216
+FastaiMLP (tuned + ensembled),0.073529415,0.11580882,0.101715684,0.12990196,0.12745099,0.1317402,0.1495098,0.12622549,0.15257353,0.18627451,0.17524509,0.22181372,0.22426471,0.2634804,0.22671568,0.2591912,0.24264705,0.2855392,0.30759802,0.27634802,0.31311274,0.23284313,0.32843137,0.3259804,0.2647059,0.31924018,0.33639705,0.33210784,0.32536766,0.3345588,0.3394608,0.3480392,0.33517158,0.35661766,0.32965687,0.3694853,0.3884804,0.35661766,0.41482842,0.4387255,0.44914216,0.3615196,0.38112745,0.4522059,0.46200982,0.45772058,0.49754903,0.5085784,,0.4852941,0.51715684,0.49019608,0.5257353,0.57598037,0.57843137,0.54044116,0.685049,0.6047794,0.533701,0.6311275,0.63480395,0.622549,0.5833333,0.6942402,0.63848037,0.7242647,0.73284316,0.70343137,0.6727941,0.74264705,0.74509805,0.95098037,0.8272059,0.73039216,0.7855392,0.8480392,0.8186275,0.89705884,0.8602941,0.95465684
+ChimeraBoost (default),0.04534314,0.104166664,0.06985294,0.123774506,0.08946078,0.1127451,0.1360294,0.13357843,0.18504901,0.20833333,0.1495098,0.2120098,0.19240196,0.25245097,0.25,0.27941176,0.27941176,0.16789216,0.21936275,0.27144608,0.2977941,0.29534313,0.31862745,0.23406863,0.28125,0.23039216,0.34068626,0.28492647,0.34558824,0.22426471,0.40808824,0.36519608,0.35171568,0.3272059,0.40563726,0.3762255,0.39215687,0.4240196,0.4240196,0.20343137,0.43995097,0.47181374,0.4252451,0.41421568,0.28431374,0.48897058,0.49632353,0.502451,0.5147059,,0.5367647,0.53186274,0.5232843,0.5367647,0.5441176,0.58639705,0.55269605,0.6127451,0.591299,0.6727941,0.67769605,0.6286765,0.6433824,0.65686274,0.69240195,0.69485295,0.7561275,0.67769605,0.72794116,0.74264705,0.685049,0.747549,0.76593137,0.75735295,0.7334559,0.7757353,0.81985295,0.85539216,0.8639706,0.9607843
+EBM (default),0.06495098,0.10355392,0.1004902,0.11029412,0.1127451,0.12009804,0.13480392,0.1360294,0.16115196,0.20955883,0.17156863,0.2512255,0.21384804,0.2757353,0.24142157,0.25490198,0.24632353,0.18259804,0.24387255,0.26715687,0.2977941,0.32107842,0.34313726,0.24509804,0.28921568,0.24019608,0.31495097,0.26715687,0.32107842,0.26715687,0.33333334,0.34558824,0.34558824,0.2977941,0.36703432,0.3602941,0.38235295,0.40012255,0.34681374,0.36887255,0.23161764,0.42034313,0.39583334,0.4362745,0.40441176,0.46568626,0.33762255,0.51960784,0.48284313,0.4632353,,0.5379902,0.5379902,0.58455884,0.59681374,0.52818626,0.5686275,0.5772059,0.54289216,0.6727941,0.67034316,0.60906863,0.6311275,0.7058824,0.68014705,0.7107843,0.7561275,0.6433824,0.70710784,0.77818626,0.72794116,0.7879902,0.81985295,0.7414216,0.8186275,0.8449755,0.8382353,0.8995098,0.8897059,0.9485294
+TorchMLP (tuned),0.030637255,0.06127451,0.05514706,0.09191176,0.06495098,0.07965686,0.09558824,0.09558824,0.16789216,0.17156863,0.14215687,0.18995099,0.18566176,0.2365196,0.19485295,0.26715687,0.22916667,0.26593137,0.25612745,0.24571079,0.3002451,0.27450982,0.30882353,0.2689951,0.25490198,0.31188726,0.32352942,0.32107842,0.3272059,0.30759802,0.3627451,0.34191176,0.3002451,0.34068626,0.3737745,0.35416666,0.3762255,0.34191176,0.4007353,0.40808824,0.3841912,0.2512255,0.39583334,0.45833334,0.4375,0.41666666,0.43259802,0.5557598,0.50980395,0.46813726,0.46200982,,0.5747549,0.5539216,0.55269605,0.5367647,0.6066176,0.60784316,0.5545343,0.685049,0.6862745,0.6372549,0.6102941,0.8082108,0.65318626,0.7463235,0.7438725,0.7083333,0.70098037,0.77818626,0.71813726,0.78186274,0.74264705,0.74509805,0.7775735,0.7604167,0.8235294,0.87009805,0.8566176,0.94730395
+ExtraTrees (tuned),0.05147059,0.063725494,0.083333336,0.1004902,0.094975494,0.12009804,0.125,0.11764706,0.1427696,0.12867647,0.19362745,0.1629902,0.2512255,0.20833333,0.21568628,0.29534313,0.3082108,0.24142157,0.22120099,0.2904412,0.2977941,0.29289216,0.24877451,0.2512255,0.23161764,0.2677696,0.2769608,0.27083334,0.3737745,0.29534313,0.37132353,0.30759802,0.33088234,0.28308824,0.35661766,0.31495097,0.38541666,0.3602941,0.44730392,0.3897059,0.3872549,0.3529412,0.35539216,0.39583334,0.40686274,0.39338234,0.41789216,0.31924018,0.4742647,0.47671568,0.46200982,0.4252451,,0.61580884,0.6145833,0.4644608,0.52205884,0.59313726,0.49509802,0.596201,0.6017157,0.66115195,0.689951,0.5857843,0.6286765,0.7389706,0.6764706,0.67769605,0.6642157,0.72794116,0.8504902,0.7291667,0.72794116,0.84865195,0.8382353,0.7377451,0.7683824,0.8235294,0.91789216,0.99019605
+Nori-30M (default),0.118872546,0.14338236,0.20465687,0.17279412,0.20588236,0.2009804,0.22426471,0.24632353,0.22916667,0.24509804,0.27818626,0.21568628,0.3137255,0.2622549,0.29289216,0.34375,0.35171568,0.29289216,0.29656863,0.3382353,0.30147058,0.31862745,0.32352942,0.31004903,0.3112745,0.3125,0.29166666,0.3125,0.4074755,0.33210784,0.39583334,0.35049018,0.34068626,0.3247549,0.39950982,0.35049018,0.30637255,0.3884804,0.36887255,0.40808824,0.38112745,0.40808824,0.4019608,0.37745097,0.42769608,0.4117647,0.39828432,0.35416666,0.4240196,0.4632353,0.4154412,0.44607842,0.3841912,,0.5625,0.4117647,0.4632353,0.36764705,0.46507353,0.47916666,0.47916666,0.44117647,0.51960784,0.5330882,0.5747549,0.5808824,0.49754903,0.64093137,0.5955882,0.62990195,0.6188725,0.60968137,0.60539216,0.63235295,0.63848037,0.627451,0.68014705,0.70098037,0.78063726,0.9632353
+Nori (default),0.068627454,0.111519605,0.15686275,0.14215687,0.16421568,0.15808824,0.18259804,0.2389706,0.19240196,0.22181372,0.25735295,0.19485295,0.29166666,0.22794117,0.28063726,0.31311274,0.33578432,0.29166666,0.30392158,0.3247549,0.30147058,0.29166666,0.29534313,0.30882353,0.2977941,0.31004903,0.28186274,0.30392158,0.39031863,0.33088234,0.38235295,0.34191176,0.32230392,0.31740198,0.3872549,0.35049018,0.30882353,0.38602942,0.37132353,0.40441176,0.37132353,0.40686274,0.40318626,0.36764705,0.4227941,0.40441176,0.38357842,0.35171568,0.42156863,0.45588234,0.40318626,0.44730392,0.38541666,0.4375,,0.40686274,0.46078432,0.37009802,0.4675245,0.4730392,0.47549018,0.44362745,0.52205884,0.5330882,0.564951,0.5833333,0.4987745,0.64093137,0.59068626,0.6286765,0.622549,0.60968137,0.60539216,0.63480395,0.6372549,0.627451,0.6789216,0.70098037,0.7794118,0.9632353
+XGBoost (default),0.028186275,0.04901961,0.037990198,0.08210784,0.037990198,0.06985294,0.0882353,0.0992647,0.08946078,0.14460784,0.08946078,0.15808824,0.14215687,0.21446079,0.15441176,0.23284313,0.2254902,0.104166664,0.1127451,0.19485295,0.24632353,0.26102942,0.28186274,0.11764706,0.16911764,0.13909313,0.24264705,0.1004902,0.3259804,0.21691176,0.34191176,0.23774509,0.28186274,0.14705883,0.3529412,0.32352942,0.3627451,0.32843137,0.41299018,0.32230392,0.3584559,0.34681374,0.33333334,0.4117647,0.34987745,0.38602942,0.40808824,0.48039216,0.4595588,0.41360295,0.47181374,0.4632353,0.5355392,0.5882353,0.59313726,,0.53063726,0.54901963,0.4810049,0.6397059,0.6433824,0.6066176,0.65686274,0.6360294,0.5735294,0.7107843,0.75,0.70465684,0.6507353,0.7242647,0.7512255,0.71813726,0.72181374,0.810049,0.76960784,0.73651963,0.7745098,0.82230395,0.84436274,0.94485295
+FastaiMLP (tuned),0.0502451,0.07904412,0.083333336,0.09313726,0.09068628,0.09068628,0.09681372,0.09436274,0.13909313,0.13725491,0.13725491,0.17524509,0.16911764,0.20465687,0.17034313,0.2322304,0.19975491,0.20955883,0.2389706,0.20710784,0.24571079,0.20710784,0.26838234,0.24816176,0.21384804,0.25857842,0.27083334,0.27941176,0.2855392,0.26715687,0.29289216,0.28921568,0.25367647,0.28431374,0.27328432,0.29411766,0.34681374,0.2757353,0.35171568,0.3915441,0.37683824,0.25674018,0.32965687,0.37990198,0.41237745,0.34681374,0.4050245,0.46078432,0.31495097,0.44730392,0.43137255,0.39338234,0.4779412,0.5367647,0.5392157,0.46936274,,0.5324755,0.4644608,0.5919118,0.5894608,0.5514706,0.5355392,0.6121324,0.5833333,0.67034316,0.67401963,0.63480395,0.6262255,0.72181374,0.70098037,0.8333333,0.7647059,0.6887255,0.74264705,0.7873775,0.7561275,0.86519605,0.83210784,0.9460784
+RandomForest (tuned + ensembled),0.030637255,0.06127451,0.05637255,0.0747549,0.068627454,0.08210784,0.11029412,0.123774506,0.0882353,0.12867647,0.14338236,0.12990196,0.21323529,0.18872549,0.15563725,0.25857842,0.28921568,0.16544117,0.15686275,0.24877451,0.24754901,0.28186274,0.24632353,0.17156863,0.17279412,0.22303921,0.23284313,0.18198529,0.36887255,0.22303921,0.3529412,0.26960784,0.25857842,0.21078432,0.32291666,0.2977941,0.39583334,0.3125,0.44240198,0.32230392,0.34191176,0.3125,0.30392158,0.3382353,0.35049018,0.34681374,0.3884804,0.30882353,0.39522058,0.3872549,0.4227941,0.39215687,0.40686274,0.63235295,0.62990195,0.4509804,0.4675245,,0.46813726,0.5661765,0.56740195,0.6789216,0.68014705,0.56985295,0.59313726,0.7169118,0.627451,0.6605392,0.6286765,0.6954657,0.78186274,0.7022059,0.71568626,0.8958333,0.8014706,0.7254902,0.7267157,0.8057598,0.8860294,0.97181374
+LightGBM (default),0.028186275,0.040441178,0.039215688,0.05882353,0.041666668,0.05147059,0.0625,0.066176474,0.07230392,0.11764706,0.075980395,0.12254902,0.117034316,0.16789216,0.14338236,0.22671568,0.2365196,0.042892158,0.10784314,0.19975491,0.25612745,0.21629901,0.24019608,0.13357843,0.12990196,0.08394608,0.23774509,0.08946078,0.31740198,0.18382353,0.32965687,0.25735295,0.27450982,0.13235295,0.31985295,0.29534313,0.3216912,0.34558824,0.39767158,0.2677696,0.36397058,0.35171568,0.30759802,0.3762255,0.3137255,0.3897059,0.4154412,0.46200982,0.46629903,0.40870097,0.45710784,0.44546568,0.50490195,0.5349265,0.5324755,0.5189951,0.5355392,0.53186274,,0.6335784,0.62990195,0.57414216,0.6188725,0.6482843,0.60784316,0.69485295,0.7058824,0.6862745,0.6691176,0.71813726,0.73039216,0.7291667,0.69730395,0.7726716,0.752451,0.7120098,0.80514705,0.8137255,0.8492647,0.95526963
+TabSTAR (tuned),0.12745099,0.14889705,0.16421568,0.15563725,0.1629902,0.17034313,0.19485295,0.16911764,0.21813725,0.23529412,0.22303921,0.24877451,0.22671568,0.27757353,0.23406863,0.2855392,0.24509804,0.24080883,0.24142157,0.24387255,0.27144608,0.23774509,0.3112745,0.24142157,0.27818626,0.25,0.28370097,0.2622549,0.30759802,0.24938725,0.29534313,0.32107842,0.27818626,0.2757353,0.3492647,0.28308824,0.29656863,0.2647059,0.26960784,0.2904412,0.2769608,0.29656863,0.3082108,0.32107842,0.30269608,0.3259804,0.29656863,0.40441176,0.36887255,0.3272059,0.3272059,0.31495097,0.40379903,0.5208333,0.5269608,0.3602941,0.40808824,0.43382353,0.36642158,,0.53125,0.47181374,0.45710784,0.42647058,0.55514705,0.5294118,0.43014705,0.5257353,0.5772059,0.95343137,0.622549,0.6237745,0.6629902,0.6102941,0.69485295,0.685049,0.6139706,0.7892157,0.82598037,0.90318626
+TabSTAR (tuned + ensembled),0.12867647,0.15073529,0.16421568,0.16053921,0.15931372,0.17034313,0.18995099,0.1697304,0.21446079,0.23284313,0.22058824,0.24754901,0.22426471,0.27818626,0.23284313,0.28431374,0.24877451,0.24019608,0.24509804,0.23958333,0.27450982,0.2365196,0.31004903,0.24264705,0.27757353,0.24877451,0.2855392,0.26593137,0.3112745,0.24877451,0.29840687,0.32352942,0.27450982,0.27389705,0.35416666,0.28063726,0.2990196,0.2622549,0.27083334,0.29534313,0.2855392,0.28982842,0.3112745,0.31617647,0.30392158,0.31862745,0.29963234,0.39460784,0.36519608,0.32230392,0.32965687,0.3137255,0.39828432,0.5208333,0.5245098,0.35661766,0.4105392,0.43259802,0.37009802,0.46875,,0.4730392,0.46078432,0.4252451,0.5514706,0.52818626,0.43259802,0.5232843,0.57230395,0.9669118,0.6237745,0.62009805,0.65563726,0.6127451,0.6862745,0.68259805,0.6127451,0.7892157,0.8210784,0.90318626
+RandomForest (tuned),0.031862747,0.04534314,0.05392157,0.07230392,0.05147059,0.06740196,0.0882353,0.0882353,0.068627454,0.11764706,0.1182598,0.121323526,0.17892157,0.16053921,0.13725491,0.22058824,0.2432598,0.12622549,0.12193628,0.23039216,0.22916667,0.24142157,0.20833333,0.13970588,0.14460784,0.17156863,0.2009804,0.15870099,0.3082108,0.19362745,0.29718137,0.25857842,0.24019608,0.18014705,0.31188726,0.25612745,0.31617647,0.30147058,0.375,0.28431374,0.31617647,0.28799018,0.27083334,0.27818626,0.31004903,0.3125,0.3480392,0.25612745,0.37745097,0.37132353,0.39093137,0.3627451,0.33884802,0.5588235,0.5563725,0.39338234,0.44852942,0.32107842,0.42585784,0.52818626,0.5269608,,0.59681374,0.52205884,0.56740195,0.6495098,0.60784316,0.6164216,0.60539216,0.67769605,0.7058824,0.69485295,0.69485295,0.7647059,0.7702206,0.70710784,0.7083333,0.79534316,0.8786765,0.9705882
+OrionMSP (default),0.04411765,0.05514706,0.052083332,0.063725494,0.0870098,0.071078435,0.08210784,0.063725494,0.15931372,0.1004902,0.18566176,0.123774506,0.19240196,0.12867647,0.20281863,0.23100491,0.22610295,0.19607843,0.2254902,0.22303921,0.23590687,0.125,0.18872549,0.23039216,0.23100491,0.20955883,0.2009804,0.23100491,0.29227942,0.25,0.25980392,0.2867647,0.2365196,0.24877451,0.30514705,0.2254902,0.2634804,0.28799018,0.3259804,0.29166666,0.31617647,0.3272059,0.30759802,0.24754901,0.3137255,0.34068626,0.35477942,0.2977941,0.41666666,0.35661766,0.36887255,0.3897059,0.31004903,0.48039216,0.4779412,0.34313726,0.4644608,0.31985295,0.38112745,0.54289216,0.5392157,0.40318626,,0.48284313,0.54044116,0.6139706,0.5833333,0.56740195,0.57598037,0.6862745,0.57843137,0.68014705,0.7058824,0.6127451,0.6813725,0.7248775,0.72794116,0.810049,0.79656863,0.95710784
+TorchMLP (default),0.024509804,0.034313727,0.030637255,0.05392157,0.036764707,0.042892158,0.04411765,0.028186275,0.063725494,0.09558824,0.078431375,0.10539216,0.07720588,0.12745099,0.09436274,0.19117647,0.18014705,0.15073529,0.14338236,0.09558824,0.21813725,0.14705883,0.18382353,0.15931372,0.13112745,0.17156863,0.15318628,0.18137255,0.24632353,0.1875,0.24019608,0.22303921,0.19362745,0.19362745,0.24509804,0.16176471,0.2855392,0.20343137,0.2990196,0.28308824,0.25367647,0.10784314,0.18995099,0.22671568,0.29289216,0.22365196,0.28308824,0.3872549,0.30575982,0.34313726,0.29411766,0.19178921,0.41421568,0.46691176,0.46691176,0.36397058,0.38786766,0.43014705,0.35171568,0.5735294,0.5747549,0.4779412,0.51715684,,0.5122549,0.61090684,0.54534316,0.58455884,0.5539216,0.6911765,0.64460784,0.64705884,0.67156863,0.64215684,0.7083333,0.69240195,0.66789216,0.80269605,0.8272059,0.93014705
+PerpetualBooster (tuned + ensembled),0.013480392,0.05269608,0.037990198,0.05392157,0.04411765,0.042892158,0.061887253,0.06495098,0.05637255,0.09803922,0.063725494,0.0882353,0.113970585,0.106617644,0.0992647,0.19730392,0.20220588,0.085784316,0.125,0.13480392,0.21936275,0.14583333,0.1740196,0.13480392,0.14583333,0.12316176,0.18382353,0.16789216,0.2720588,0.15196079,0.2634804,0.21691176,0.20955883,0.19485295,0.23161764,0.21997549,0.26102942,0.23713236,0.29656863,0.21323529,0.24509804,0.27818626,0.2769608,0.2757353,0.25306374,0.28431374,0.27818626,0.3492647,0.3615196,0.30759802,0.31985295,0.34681374,0.37132353,0.4252451,0.43504903,0.42647058,0.41666666,0.40686274,0.39215687,0.44485295,0.44852942,0.43259802,0.4595588,0.4877451,,0.5514706,0.5612745,0.5637255,0.7395833,0.53063726,0.59068626,0.5894608,0.6507353,0.6139706,0.6047794,0.6629902,0.84313726,0.7169118,0.74509805,0.935049
+TabFlex (default),0.04656863,0.05269608,0.063725494,0.07230392,0.066176474,0.06985294,0.08455882,0.0680147,0.12745099,0.07230392,0.123774506,0.109068625,0.14215687,0.121323526,0.14154412,0.18933824,0.1764706,0.16421568,0.19607843,0.16421568,0.2389706,0.09681372,0.11519608,0.20343137,0.14828432,0.17892157,0.14215687,0.19546568,0.22487745,0.19975491,0.18504901,0.2254902,0.17279412,0.20710784,0.2254902,0.1384804,0.24754901,0.2365196,0.26409313,0.21323529,0.25367647,0.22058824,0.22365196,0.18995099,0.24877451,0.21384804,0.28125,0.25490198,0.2757353,0.30514705,0.28921568,0.25367647,0.26102942,0.41911766,0.41666666,0.28921568,0.32965687,0.28308824,0.30514705,0.47058824,0.47181374,0.35049018,0.38602942,0.38909313,0.44852942,,0.5061275,0.4852941,0.5012255,0.61151963,0.4865196,0.5955882,0.6262255,0.5514706,0.6286765,0.64889705,0.622549,0.74509805,0.7647059,0.92034316
+iLTM (default),0.042892158,0.039215688,0.041666668,0.06004902,0.05392157,0.063725494,0.06985294,0.06004902,0.0882353,0.118872546,0.0870098,0.12867647,0.09436274,0.16544117,0.123774506,0.19362745,0.1740196,0.080882356,0.114583336,0.11519608,0.19607843,0.16789216,0.18872549,0.123774506,0.11764706,0.09620098,0.18995099,0.14583333,0.21813725,0.09068628,0.2389706,0.20343137,0.15073529,0.15196079,0.27328432,0.18627451,0.1875,0.1985294,0.17892157,0.18259804,0.19730392,0.19301471,0.18933824,0.20710784,0.20833333,0.22794117,0.24019608,0.30147058,0.26715687,0.24387255,0.24387255,0.25612745,0.32352942,0.502451,0.5012255,0.25,0.3259804,0.37254903,0.29411766,0.56985295,0.56740195,0.39215687,0.41666666,0.45465687,0.4387255,0.49387255,,0.49019608,0.46078432,0.65318626,0.502451,0.59313726,0.6544118,0.56740195,0.69240195,0.67156863,0.5735294,0.7610294,0.79534316,0.89705884
+xRFM (default),0.029411765,0.037990198,0.040441178,0.05514706,0.05147059,0.063725494,0.073529415,0.066176474,0.06127451,0.09313726,0.09558824,0.082720585,0.14093137,0.106617644,0.118872546,0.13970588,0.15318628,0.16789216,0.14460784,0.18137255,0.16911764,0.16544117,0.16053921,0.15318628,0.10539216,0.17953432,0.0502451,0.17279412,0.19485295,0.20588236,0.19362745,0.14031863,0.14460784,0.18382353,0.23774509,0.09803922,0.2634804,0.27083334,0.3259804,0.25735295,0.29166666,0.23161764,0.18872549,0.23529412,0.27818626,0.27941176,0.3137255,0.2867647,0.29656863,0.32230392,0.35661766,0.29166666,0.32230392,0.35906863,0.35906863,0.29534313,0.36519608,0.3394608,0.3137255,0.4742647,0.47671568,0.38357842,0.43259802,0.4154412,0.4362745,0.5147059,0.50980395,,0.46629903,0.5735294,0.47549018,0.53063726,0.5870098,0.5257353,0.58210784,0.59313726,0.59068626,0.6752451,0.7083333,0.92034316
+PerpetualBooster (tuned),0.014705882,0.035539217,0.03247549,0.0502451,0.04656863,0.041666668,0.05514706,0.05759804,0.04901961,0.07230392,0.0625,0.07904412,0.09436274,0.09803922,0.09191176,0.17279412,0.18382353,0.071078435,0.106617644,0.125,0.20465687,0.12745099,0.1384804,0.113970585,0.111519605,0.101715684,0.15012255,0.13909313,0.25674018,0.12009804,0.2512255,0.20833333,0.1740196,0.15441176,0.22242647,0.17953432,0.24264705,0.2254902,0.27328432,0.18504901,0.22732843,0.22916667,0.22916667,0.22303921,0.22365196,0.22794117,0.25367647,0.31617647,0.3272059,0.2720588,0.29289216,0.2990196,0.33578432,0.40441176,0.40931374,0.3492647,0.3737745,0.37132353,0.33088234,0.4227941,0.42769608,0.39460784,0.4240196,0.44607842,0.26041666,0.4987745,0.5392157,0.533701,,0.5085784,0.54656863,0.5502451,0.6335784,0.58455884,0.56740195,0.64460784,0.74693626,0.6985294,0.7169118,0.9142157
+TabSTAR (default),0.09558824,0.11029412,0.12622549,0.10784314,0.13112745,0.1360294,0.14338236,0.12867647,0.1629902,0.15441176,0.16789216,0.16421568,0.17034313,0.1764706,0.17892157,0.22303921,0.20588236,0.17156863,0.18872549,0.18995099,0.22242647,0.16544117,0.21446079,0.1875,0.19362745,0.18627451,0.21813725,0.20833333,0.24877451,0.18504901,0.24019608,0.2622549,0.2120098,0.21078432,0.20465687,0.20955883,0.2254902,0.1875,0.19914216,0.21446079,0.18995099,0.19975491,0.24019608,0.22058824,0.22916667,0.22181372,0.20465687,0.2677696,0.25735295,0.25735295,0.22181372,0.22181372,0.2720588,0.37009802,0.37132353,0.2757353,0.27818626,0.30453432,0.28186274,0.04656863,0.033088237,0.32230392,0.3137255,0.30882353,0.46936274,0.3884804,0.34681374,0.42647058,0.49142158,,0.46568626,0.4522059,0.55759805,0.43382353,0.55269605,0.56740195,0.53186274,0.6617647,0.6911765,0.872549
+ExtraTrees (default),0.013480392,0.024509804,0.041666668,0.041666668,0.039215688,0.04411765,0.06127451,0.05759804,0.039215688,0.054534312,0.06985294,0.04901961,0.10784314,0.07720588,0.11029412,0.109068625,0.15196079,0.08210784,0.09803922,0.15563725,0.18627451,0.1384804,0.11519608,0.111519605,0.09068628,0.106617644,0.06495098,0.121323526,0.18995099,0.13357843,0.19485295,0.15686275,0.15318628,0.123774506,0.20955883,0.13112745,0.2120098,0.20833333,0.28063726,0.2120098,0.24142157,0.23529412,0.18382353,0.18688725,0.22303921,0.20465687,0.25735295,0.080882356,0.25490198,0.31495097,0.2720588,0.28186274,0.1495098,0.38112745,0.37745097,0.24877451,0.2990196,0.21813725,0.26960784,0.37745097,0.3762255,0.29411766,0.42156863,0.35539216,0.40931374,0.51348037,0.49754903,0.5245098,0.45343137,0.53431374,,0.55759805,0.5955882,0.54534316,0.6237745,0.60906863,0.57598037,0.70098037,0.75735295,0.97181374
+FastaiMLP (default),0.023284314,0.033088237,0.034313727,0.0502451,0.04901961,0.04779412,0.05514706,0.05147059,0.04779412,0.06127451,0.05269608,0.07965686,0.08210784,0.09436274,0.0692402,0.13480392,0.1360294,0.11764706,0.12254902,0.12254902,0.16053921,0.10539216,0.123774506,0.13480392,0.10294118,0.13235295,0.09436274,0.15441176,0.18872549,0.123774506,0.18872549,0.17892157,0.13541667,0.16421568,0.18382353,0.113970585,0.18198529,0.17524509,0.19975491,0.18382353,0.16789216,0.14154412,0.15318628,0.17524509,0.20588236,0.12745099,0.2009804,0.25,0.04901961,0.25245097,0.2120098,0.21813725,0.27083334,0.39031863,0.39031863,0.28186274,0.16666667,0.2977941,0.27083334,0.3762255,0.37990198,0.30514705,0.31985295,0.3529412,0.4105392,0.40441176,0.40686274,0.46936274,0.4497549,0.5477941,0.44240198,,0.6335784,0.47732842,0.5735294,0.65318626,0.5882353,0.747549,0.74509805,0.9129902
+Linear (tuned + ensembled),0.041666668,0.06740196,0.07230392,0.05882353,0.0747549,0.07965686,0.09068628,0.078431375,0.09436274,0.12745099,0.097426474,0.15931372,0.09681372,0.17340687,0.13480392,0.14460784,0.13235295,0.118872546,0.14093137,0.118872546,0.14338236,0.1740196,0.22181372,0.14460784,0.18627451,0.1384804,0.13235295,0.1495098,0.16911764,0.16053921,0.14767157,0.23284313,0.17769608,0.15808824,0.17524509,0.13725491,0.16421568,0.17279412,0.113970585,0.21446079,0.1384804,0.18811275,0.23835784,0.20220588,0.22671568,0.20465687,0.16544117,0.25796568,0.17279412,0.23406863,0.18014705,0.25735295,0.2720588,0.39460784,0.39460784,0.27818626,0.23529412,0.28431374,0.30269608,0.33700982,0.34436274,0.30514705,0.29411766,0.32843137,0.3492647,0.3737745,0.34558824,0.41299018,0.36642158,0.44240198,0.40441176,0.36642158,,0.40686274,0.44485295,0.6672794,0.46078432,0.8125,0.54901963,0.7585784
+RandomForest (default),0.014705882,0.014705882,0.026960785,0.031862747,0.022058824,0.034313727,0.04411765,0.04534314,0.034313727,0.037990198,0.05882353,0.04411765,0.07965686,0.06985294,0.07965686,0.13480392,0.1360294,0.05147059,0.05759804,0.111519605,0.13970588,0.11029412,0.123774506,0.066176474,0.07230392,0.0747549,0.05882353,0.06495098,0.2009804,0.09803922,0.18627451,0.14460784,0.12254902,0.07965686,0.2009804,0.109068625,0.1740196,0.16789216,0.2365196,0.17892157,0.20465687,0.20343137,0.15318628,0.14215687,0.20955883,0.1985294,0.23161764,0.113970585,0.26960784,0.24264705,0.25857842,0.25490198,0.15134804,0.36764705,0.36519608,0.18995099,0.3112745,0.104166664,0.22732843,0.3897059,0.3872549,0.23529412,0.3872549,0.35784313,0.38602942,0.44852942,0.43259802,0.4742647,0.4154412,0.5661765,0.45465687,0.5226716,0.59313726,,0.5882353,0.61519605,0.56740195,0.6960784,0.7487745,0.9460784
+KNN (tuned + ensembled),0.030637255,0.05147059,0.04901961,0.04534314,0.05147059,0.06985294,0.08210784,0.09191176,0.0747549,0.0747549,0.09558824,0.082720585,0.109068625,0.08946078,0.11029412,0.14583333,0.1495098,0.101715684,0.1004902,0.1384804,0.1629902,0.11764706,0.125,0.109068625,0.12254902,0.123774506,0.09375,0.13357843,0.19607843,0.12745099,0.18504901,0.1764706,0.13725491,0.13664216,0.20955883,0.106617644,0.12990196,0.18137255,0.13970588,0.20465687,0.17034313,0.18014705,0.18504901,0.13480392,0.22303921,0.17156863,0.19730392,0.13112745,0.21446079,0.2665441,0.18137255,0.22242647,0.16176471,0.3615196,0.3627451,0.23039216,0.25735295,0.1985294,0.24754901,0.30514705,0.3137255,0.2297794,0.31862745,0.29166666,0.39522058,0.37132353,0.30759802,0.41789216,0.43259802,0.44730392,0.3762255,0.42647058,0.55514705,0.4117647,,0.5827206,0.5392157,0.6985294,0.88664216,0.9742647
+Linear (tuned),0.036764707,0.06127451,0.05637255,0.05269608,0.06004902,0.068627454,0.075980395,0.073529415,0.09313726,0.118872546,0.080269605,0.14583333,0.085171565,0.15808824,0.12806372,0.12867647,0.116421565,0.104166664,0.12867647,0.11029412,0.12867647,0.15441176,0.2009804,0.13419117,0.16053921,0.1194853,0.12009804,0.14093137,0.15686275,0.14460784,0.14215687,0.22181372,0.15563725,0.14215687,0.16911764,0.12009804,0.15196079,0.16115196,0.1004902,0.19975491,0.125,0.1629902,0.20833333,0.18198529,0.22058824,0.1740196,0.1495098,0.24203432,0.15196079,0.22426471,0.15502451,0.23958333,0.2622549,0.37254903,0.37254903,0.2634804,0.21262255,0.27450982,0.28799018,0.31495097,0.31740198,0.29289216,0.27512255,0.30759802,0.33700982,0.35110295,0.32843137,0.40686274,0.35539216,0.43259802,0.39093137,0.34681374,0.33272058,0.38480392,0.41727942,,0.45465687,0.7438725,0.5367647,0.752451
+PerpetualBooster (default),0.012254902,0.022058824,0.025735294,0.028186275,0.026960785,0.025735294,0.025735294,0.030637255,0.024509804,0.030637255,0.040441178,0.042892158,0.04411765,0.04411765,0.037990198,0.0747549,0.068627454,0.040441178,0.06740196,0.0625,0.12867647,0.06740196,0.078431375,0.063725494,0.068627454,0.04534314,0.078431375,0.07720588,0.121323526,0.07230392,0.12622549,0.14215687,0.104166664,0.078431375,0.16176471,0.09191176,0.14338236,0.11764706,0.16666667,0.12990196,0.11519608,0.11764706,0.13480392,0.1360294,0.13664216,0.123774506,0.1360294,0.21936275,0.18137255,0.18014705,0.16176471,0.1764706,0.23161764,0.31985295,0.32107842,0.2254902,0.24387255,0.27328432,0.19485295,0.38602942,0.3872549,0.29166666,0.2720588,0.33210784,0.15686275,0.37745097,0.42647058,0.40931374,0.25306374,0.46813726,0.4240196,0.4117647,0.5392157,0.43259802,0.46078432,0.54534316,,0.61580884,0.60784316,0.85539216
+Linear (default),0.024509804,0.036764707,0.03982843,0.03370098,0.042892158,0.050857842,0.04840686,0.037990198,0.07965686,0.05637255,0.05882353,0.09558824,0.06495098,0.10784314,0.1004902,0.113970585,0.08946078,0.063725494,0.104166664,0.08210784,0.109068625,0.082720585,0.123774506,0.109068625,0.121323526,0.07720588,0.09620098,0.11029412,0.13480392,0.09436274,0.109068625,0.18627451,0.111519605,0.11519608,0.13725491,0.08210784,0.07169118,0.11764706,0.05269608,0.12990196,0.0992647,0.101715684,0.15073529,0.101715684,0.14338236,0.1127451,0.11764706,0.16911764,0.10294118,0.14460784,0.1004902,0.12990196,0.1764706,0.2990196,0.2990196,0.17769608,0.13480392,0.1942402,0.18627451,0.21078432,0.21078432,0.20465687,0.18995099,0.19730392,0.28308824,0.25490198,0.2389706,0.3247549,0.30147058,0.3382353,0.2990196,0.25245097,0.1875,0.30392158,0.30147058,0.25612745,0.3841912,,0.46568626,0.70710784
+KNN (tuned),0.028186275,0.034313727,0.035539217,0.035539217,0.039215688,0.05147059,0.04901961,0.05269608,0.05392157,0.042892158,0.0625,0.05637255,0.07230392,0.06495098,0.083333336,0.09191176,0.090073526,0.06556372,0.073529415,0.09191176,0.123774506,0.04656863,0.09313726,0.06985294,0.09129902,0.073529415,0.071078435,0.085171565,0.13112745,0.06740196,0.118872546,0.14828432,0.09436274,0.08946078,0.111519605,0.07230392,0.078431375,0.0992647,0.066176474,0.113970585,0.112132356,0.106617644,0.12683824,0.09068628,0.1360294,0.09313726,0.113970585,0.0870098,0.13970588,0.1360294,0.11029412,0.14338236,0.08210784,0.21936275,0.22058824,0.15563725,0.16789216,0.113970585,0.15073529,0.1740196,0.17892157,0.121323526,0.20343137,0.17279412,0.25490198,0.23529412,0.20465687,0.29166666,0.28308824,0.30882353,0.24264705,0.25490198,0.4509804,0.2512255,0.11335784,0.4632353,0.39215687,0.53431374,,0.9375
+KNN (default),0.006127451,0.004901961,0.006127451,0.004901961,0.007352941,0.008578432,0.013480392,0.014705882,0.006127451,0.009803922,0.009803922,0.006127451,0.024509804,0.008578432,0.017156864,0.006127451,0.009803922,0.013480392,0.012254902,0.031862747,0.031862747,0.006127451,0.007352941,0.013480392,0.017156864,0.018382354,0.011029412,0.020833334,0.023284314,0.015931373,0.026960785,0.041666668,0.022058824,0.023284314,0.014705882,0.013480392,0.026960785,0.04779412,0.030637255,0.02634804,0.04779412,0.040441178,0.037990198,0.019607844,0.037990198,0.020833334,0.05392157,0.008578432,0.04534314,0.039215688,0.05147059,0.05269608,0.009803922,0.036764707,0.036764707,0.05514706,0.05392157,0.028186275,0.04473039,0.09681372,0.09681372,0.029411765,0.042892158,0.06985294,0.06495098,0.07965686,0.10294118,0.07965686,0.085784316,0.12745099,0.028186275,0.0870098,0.24142157,0.05392157,0.025735294,0.24754901,0.14460784,0.29289216,0.0625,
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_medium/leaderboard_overview_explorer.html
index b5bda792cd7fa805fca9ba69996ab0b02c78d2fa..800d3a0cdad14c90e4f19ebf3dea4355e6fa5d06 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — All Tasks | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_all/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..6fed05eca8b63e7cc567098ed5f6d39f8077df79
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_all/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1972.0,"score":0.873,"rank":4.45,"hrank":1.69,"imp":2.683,"train_time":27.29,"predict_time":13.119,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+297\/-140","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1881.0,"score":0.716,"rank":6.54,"hrank":3.0,"imp":4.858,"train_time":191.18,"predict_time":2.207,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-48","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1780.0,"score":0.671,"rank":9.75,"hrank":3.26,"imp":5.878,"train_time":1.01,"predict_time":0.215,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+208\/-128","imputed":false},{"position":3,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1756.0,"score":0.59,"rank":10.68,"hrank":4.52,"imp":6.048,"train_time":432.36,"predict_time":4.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-57","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1707.0,"score":0.518,"rank":12.77,"hrank":8.94,"imp":7.846,"train_time":2.76,"predict_time":0.701,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-67","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1685.0,"score":0.553,"rank":13.82,"hrank":4.44,"imp":6.851,"train_time":0.6,"predict_time":0.119,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+177\/-107","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1633.0,"score":0.458,"rank":16.48,"hrank":8.58,"imp":8.576,"train_time":735.58,"predict_time":11.736,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-101","imputed":false},{"position":7,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1608.0,"score":0.427,"rank":17.89,"hrank":12.17,"imp":8.917,"train_time":1719.82,"predict_time":1.675,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-90","imputed":false},{"position":8,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1600.0,"score":0.413,"rank":18.38,"hrank":12.59,"imp":9.07,"train_time":2526.31,"predict_time":6.013,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-105","imputed":false},{"position":9,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1592.0,"score":0.394,"rank":18.82,"hrank":9.46,"imp":8.932,"train_time":489.93,"predict_time":0.249,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-67","imputed":false},{"position":10,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1588.0,"score":0.416,"rank":19.07,"hrank":10.62,"imp":9.218,"train_time":735.58,"predict_time":1.386,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-93","imputed":false},{"position":11,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1578.0,"score":0.386,"rank":19.64,"hrank":9.53,"imp":9.14,"train_time":489.93,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-69","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1572.0,"score":0.344,"rank":20.01,"hrank":13.85,"imp":10.466,"train_time":131.56,"predict_time":2.639,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-55","imputed":false},{"position":13,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1560.0,"score":0.392,"rank":20.78,"hrank":9.73,"imp":9.095,"train_time":2.47,"predict_time":0.01,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-102","imputed":false},{"position":14,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1542.0,"score":0.339,"rank":21.9,"hrank":16.03,"imp":10.221,"train_time":2067.95,"predict_time":65.195,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-55","imputed":false},{"position":15,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1533.0,"score":0.312,"rank":22.52,"hrank":15.19,"imp":10.546,"train_time":282.13,"predict_time":0.563,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-60","imputed":false},{"position":16,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1529.0,"score":0.308,"rank":22.79,"hrank":13.53,"imp":10.254,"train_time":1236.88,"predict_time":0.579,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+133\/-83","imputed":false},{"position":17,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1519.0,"score":0.335,"rank":23.41,"hrank":15.12,"imp":10.077,"train_time":1.88,"predict_time":0.645,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-77","imputed":false},{"position":18,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1518.0,"score":0.317,"rank":23.52,"hrank":14.38,"imp":10.104,"train_time":1719.82,"predict_time":0.081,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-85","imputed":false},{"position":19,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1511.0,"score":0.367,"rank":23.93,"hrank":11.27,"imp":10.809,"train_time":0.75,"predict_time":0.18,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+153\/-135","imputed":false},{"position":20,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1509.0,"score":0.335,"rank":24.1,"hrank":6.4,"imp":10.287,"train_time":4786.6,"predict_time":444.544,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-114","imputed":false},{"position":21,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1500.0,"score":0.328,"rank":24.66,"hrank":13.62,"imp":11.684,"train_time":95.92,"predict_time":0.297,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-122","imputed":false},{"position":22,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1493.0,"score":0.237,"rank":25.2,"hrank":18.98,"imp":11.603,"train_time":131.56,"predict_time":0.132,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-56","imputed":false},{"position":23,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1488.0,"score":0.253,"rank":25.5,"hrank":20.56,"imp":11.102,"train_time":282.13,"predict_time":0.066,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-52","imputed":false},{"position":24,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1467.0,"score":0.246,"rank":26.99,"hrank":16.07,"imp":11.089,"train_time":1236.88,"predict_time":0.054,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-73","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1462.0,"score":0.233,"rank":27.36,"hrank":21.43,"imp":11.215,"train_time":2526.31,"predict_time":0.349,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-77","imputed":false},{"position":26,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1455.0,"score":0.282,"rank":27.79,"hrank":15.73,"imp":12.146,"train_time":95.92,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-115","imputed":false},{"position":27,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1411.0,"score":0.238,"rank":30.99,"hrank":22.01,"imp":12.482,"train_time":566.74,"predict_time":0.951,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-105","imputed":false},{"position":28,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1411.0,"score":0.167,"rank":31.0,"hrank":19.5,"imp":12.112,"train_time":2067.95,"predict_time":10.881,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-64","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1406.0,"score":0.221,"rank":31.38,"hrank":19.77,"imp":13.176,"train_time":0.48,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-114","imputed":false},{"position":30,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1392.0,"score":0.23,"rank":32.4,"hrank":12.76,"imp":12.662,"train_time":4786.6,"predict_time":42.641,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-115","imputed":false},{"position":31,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1365.0,"score":0.139,"rank":34.36,"hrank":27.39,"imp":12.45,"train_time":1770.56,"predict_time":1.069,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-61","imputed":false},{"position":32,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1363.0,"score":0.151,"rank":34.49,"hrank":22.33,"imp":12.907,"train_time":0.49,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-98","imputed":false},{"position":33,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1361.0,"score":0.175,"rank":34.68,"hrank":28.48,"imp":13.982,"train_time":566.74,"predict_time":0.06,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-97","imputed":false},{"position":34,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1334.0,"score":0.168,"rank":36.7,"hrank":21.06,"imp":13.406,"train_time":3.97,"predict_time":1.596,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+140\/-111","imputed":true},{"position":35,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1333.0,"score":0.164,"rank":36.75,"hrank":25.34,"imp":13.337,"train_time":0.51,"predict_time":0.127,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+114\/-101","imputed":false},{"position":36,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1319.0,"score":0.185,"rank":37.81,"hrank":16.79,"imp":15.311,"train_time":46.62,"predict_time":43.738,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+117\/-134","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1318.0,"score":0.107,"rank":37.9,"hrank":30.93,"imp":13.982,"train_time":2.13,"predict_time":0.047,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+113\/-85","imputed":false},{"position":38,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1309.0,"score":0.151,"rank":38.51,"hrank":28.11,"imp":15.595,"train_time":695.61,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+121\/-124","imputed":false},{"position":39,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1292.0,"score":0.084,"rank":39.82,"hrank":35.46,"imp":14.33,"train_time":8.06,"predict_time":0.25,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-74","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1266.0,"score":0.057,"rank":41.73,"hrank":36.46,"imp":14.114,"train_time":1770.56,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-71","imputed":false},{"position":41,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1266.0,"score":0.174,"rank":41.77,"hrank":10.03,"imp":16.418,"train_time":6.55,"predict_time":5.409,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-151","imputed":false},{"position":42,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1264.0,"score":0.045,"rank":41.92,"hrank":38.09,"imp":15.091,"train_time":6.75,"predict_time":0.235,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-71","imputed":false},{"position":43,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1254.0,"score":0.033,"rank":42.59,"hrank":40.08,"imp":14.82,"train_time":0.29,"predict_time":0.045,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-59","imputed":false},{"position":44,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1253.0,"score":0.103,"rank":42.7,"hrank":34.05,"imp":16.441,"train_time":695.61,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-131","imputed":false},{"position":45,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1245.0,"score":0.034,"rank":43.29,"hrank":37.06,"imp":15.1,"train_time":66.0,"predict_time":0.341,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-76","imputed":false},{"position":46,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1223.0,"score":0.052,"rank":44.86,"hrank":37.07,"imp":17.01,"train_time":66.07,"predict_time":0.652,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+91\/-93","imputed":true},{"position":47,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1217.0,"score":0.027,"rank":45.34,"hrank":42.63,"imp":15.304,"train_time":49.61,"predict_time":0.199,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-83","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1213.0,"score":0.035,"rank":45.61,"hrank":39.26,"imp":15.744,"train_time":66.0,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-79","imputed":false},{"position":49,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1211.0,"score":0.071,"rank":45.77,"hrank":39.35,"imp":17.87,"train_time":2.18,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-133","imputed":false},{"position":50,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1201.0,"score":0.16,"rank":46.48,"hrank":26.24,"imp":18.639,"train_time":0.2,"predict_time":0.021,"imputed_pct":66.67,"hardware":"GPU","elo_ci":"+138\/-164","imputed":true},{"position":51,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1193.0,"score":0.031,"rank":47.05,"hrank":40.92,"imp":16.933,"train_time":11.44,"predict_time":10.416,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-95","imputed":false},{"position":52,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1186.0,"score":0.066,"rank":47.52,"hrank":38.06,"imp":17.192,"train_time":351.94,"predict_time":1.658,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-136","imputed":false},{"position":53,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1179.0,"score":0.134,"rank":48.05,"hrank":32.64,"imp":18.766,"train_time":0.2,"predict_time":0.021,"imputed_pct":66.67,"hardware":"GPU","elo_ci":"+126\/-168","imputed":true},{"position":54,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1178.0,"score":0.01,"rank":48.1,"hrank":45.59,"imp":16.303,"train_time":49.61,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-76","imputed":false},{"position":55,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1173.0,"score":0.043,"rank":48.44,"hrank":43.24,"imp":18.488,"train_time":88.27,"predict_time":9.62,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-113","imputed":false},{"position":56,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1127.0,"score":0.002,"rank":51.59,"hrank":49.49,"imp":18.145,"train_time":2.18,"predict_time":0.039,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-65","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1108.0,"score":0.025,"rank":52.86,"hrank":45.57,"imp":21.633,"train_time":20.26,"predict_time":0.515,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-124","imputed":false},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1093.0,"score":0.018,"rank":53.81,"hrank":47.54,"imp":19.294,"train_time":351.94,"predict_time":0.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-104","imputed":false},{"position":59,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1061.0,"score":0.009,"rank":55.79,"hrank":50.83,"imp":22.747,"train_time":20.26,"predict_time":0.2,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-118","imputed":false},{"position":60,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1040.0,"score":0.022,"rank":57.03,"hrank":49.91,"imp":25.98,"train_time":2.49,"predict_time":0.145,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+144\/-174","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1025.0,"score":0.021,"rank":57.94,"hrank":51.44,"imp":22.864,"train_time":1.51,"predict_time":0.111,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+132\/-160","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":59.34,"hrank":58.19,"imp":21.197,"train_time":0.2,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-85","imputed":false},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":987.0,"score":0.007,"rank":60.07,"hrank":52.62,"imp":24.353,"train_time":150.3,"predict_time":10.071,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-119","imputed":false},{"position":64,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":984.0,"score":0.0,"rank":60.21,"hrank":58.3,"imp":25.807,"train_time":3381.77,"predict_time":1.724,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-85","imputed":false},{"position":65,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":984.0,"score":0.0,"rank":60.23,"hrank":58.98,"imp":22.691,"train_time":10.1,"predict_time":3.716,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+57\/-84","imputed":true},{"position":66,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":983.0,"score":0.0,"rank":60.27,"hrank":58.49,"imp":25.781,"train_time":3381.77,"predict_time":5.283,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-80","imputed":false},{"position":67,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":973.0,"score":0.0,"rank":60.76,"hrank":58.53,"imp":22.789,"train_time":0.48,"predict_time":0.073,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-118","imputed":false},{"position":68,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":938.0,"score":0.008,"rank":62.53,"hrank":56.75,"imp":32.872,"train_time":43.64,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-216","imputed":false},{"position":69,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":938.0,"score":0.0,"rank":62.56,"hrank":61.2,"imp":27.703,"train_time":0.25,"predict_time":0.045,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+81\/-108","imputed":true},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":899.0,"score":0.003,"rank":64.34,"hrank":59.53,"imp":33.73,"train_time":43.64,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-237","imputed":false},{"position":71,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":894.0,"score":0.002,"rank":64.59,"hrank":62.22,"imp":29.424,"train_time":1.5,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-160","imputed":false},{"position":72,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":852.0,"score":0.0,"rank":66.3,"hrank":65.49,"imp":30.685,"train_time":110.78,"predict_time":1.964,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-88","imputed":false},{"position":73,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":815.0,"score":0.0,"rank":67.68,"hrank":67.31,"imp":29.973,"train_time":150.3,"predict_time":0.482,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-103","imputed":false},{"position":74,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":810.0,"score":0.0,"rank":67.86,"hrank":65.56,"imp":35.794,"train_time":0.19,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-231","imputed":false},{"position":75,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":582.0,"score":0.0,"rank":73.48,"hrank":73.36,"imp":43.501,"train_time":0.1,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-140","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_medium/pareto_front_explorer.html
index 25589ecf66303fc72affdf0afaebfdd898f58479..dbdb0e59ebebca885cc3a4021fe7471da2ae78fd 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.6520557525,"imp":17.0103274818,"elo":1223.3,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0102827644,"imp":9.0952095474,"elo":1560.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0196975147,"imp":9.1399516875,"elo":1578.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2485703935,"imp":8.9317687118,"elo":1592.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0225370945,"imp":13.175735213,"elo":1405.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0265970701,"imp":12.1456953066,"elo":1455.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2967122121,"imp":11.6842157983,"elo":1500.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"imp":17.8697973966,"elo":1210.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0061458745,"imp":16.4411619269,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0851920607,"imp":15.5948171656,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0733045085,"imp":22.7894839202,"elo":973.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0153132761,"imp":16.3030865094,"elo":1177.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1990111469,"imp":15.3042054388,"elo":1216.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"imp":22.8643090483,"elo":1024.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1141755222,"imp":19.2937753734,"elo":1092.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6581625315,"imp":17.1916474397,"elo":1186.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0447579838,"imp":14.819832856,"elo":1254.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1322357268,"imp":11.6033224024,"elo":1492.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"imp":10.4658385154,"elo":1572.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.4085730775,"imp":16.4177044552,"elo":1265.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"imp":14.3302915146,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3490410688,"imp":11.2147479695,"elo":1461.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.0133340655,"imp":9.0700293864,"elo":1599.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.020899149,"imp":18.7657004236,"elo":1178.6,"imputed":true,"imputed_pct":66.6666666667,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.020899149,"imp":18.638873156,"elo":1200.8,"imputed":true,"imputed_pct":66.6666666667,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.7156797489,"imp":22.690940904,"elo":983.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0149545776,"imp":29.4235354386,"elo":893.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1995710638,"imp":22.7471582873,"elo":1061.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5149235805,"imp":21.6330567052,"elo":1107.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.020899149,"imp":21.1973502954,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0243850814,"imp":15.7438782463,"elo":1212.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3407391303,"imp":15.0995978866,"elo":1244.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"imp":15.0910499881,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0807033173,"imp":10.103937795,"elo":1517.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.6746674551,"imp":8.9170097602,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"imp":10.077321119,"elo":1519.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3859818406,"imp":9.2178209823,"elo":1588.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.7359861516,"imp":8.5761380952,"elo":1633.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.4162023366,"imp":16.9330805211,"elo":1192.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"imp":15.3106058247,"elo":1318.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.6414441586,"imp":12.6616447819,"elo":1391.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":444.5439481393,"imp":10.2874153003,"elo":1508.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1795394575,"imp":10.8085465808,"elo":1511.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":13.1188916889,"imp":2.6827874305,"elo":1971.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0451670031,"imp":27.7026324503,"elo":937.7,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.5956993553,"imp":13.4063747264,"elo":1333.8,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1186281919,"imp":6.8505424386,"elo":1684.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0470610966,"imp":13.9816848431,"elo":1317.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0538978015,"imp":11.0894895039,"elo":1466.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5790353938,"imp":10.2538835957,"elo":1528.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"imp":7.8460750331,"elo":1707.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2147265792,"imp":5.8782765424,"elo":1780.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"imp":30.6853633341,"elo":852.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.723789106,"imp":25.807196939,"elo":984.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"imp":25.7806935994,"elo":982.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1271670825,"imp":13.3366678662,"elo":1333.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0388490278,"imp":18.1452782924,"elo":1126.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.050592863,"imp":14.1135657835,"elo":1266.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0691781292,"imp":12.4495254577,"elo":1365.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"imp":12.9074172378,"elo":1363.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0664942176,"imp":11.1023313966,"elo":1488.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.5626253891,"imp":10.5460036421,"elo":1532.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":9.6204056883,"imp":18.4877413347,"elo":1173.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.8813375711,"imp":12.1121479202,"elo":1410.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.1951531214,"imp":10.2205385819,"elo":1542.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1448756257,"imp":25.9796305322,"elo":1040.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0599221336,"imp":13.9821871779,"elo":1360.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.9511369135,"imp":12.4822741841,"elo":1410.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.6520557525,"x_train":66.0653250838,"imp":17.0103274818,"elo":1223.3,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0102827644,"x_train":2.4694138287,"imp":9.0952095474,"elo":1560.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0196975147,"x_train":489.9274101037,"imp":9.1399516875,"elo":1578.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2485703935,"x_train":489.9274101037,"imp":8.9317687118,"elo":1592.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0225370945,"x_train":0.4789788316,"imp":13.175735213,"elo":1405.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0265970701,"x_train":95.9169155889,"imp":12.1456953066,"elo":1455.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2967122121,"x_train":95.9169155889,"imp":11.6842157983,"elo":1500.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"x_train":2.1751921611,"imp":17.8697973966,"elo":1210.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0061458745,"x_train":695.6092947225,"imp":16.4411619269,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0851920607,"x_train":695.6092947225,"imp":15.5948171656,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0733045085,"x_train":0.4842583839,"imp":22.7894839202,"elo":973.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0153132761,"x_train":49.6127891307,"imp":16.3030865094,"elo":1177.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1990111469,"x_train":49.6127891307,"imp":15.3042054388,"elo":1216.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"x_train":1.5145310857,"imp":22.8643090483,"elo":1024.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1141755222,"x_train":351.9412405462,"imp":19.2937753734,"elo":1092.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6581625315,"x_train":351.9412405462,"imp":17.1916474397,"elo":1186.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0447579838,"x_train":0.2873206973,"imp":14.819832856,"elo":1254.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1322357268,"x_train":131.5620070704,"imp":11.6033224024,"elo":1492.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"x_train":131.5620070704,"imp":10.4658385154,"elo":1572.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.4085730775,"x_train":6.5532925928,"imp":16.4177044552,"elo":1265.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"x_train":8.0637708772,"imp":14.3302915146,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3490410688,"x_train":2526.3066806002,"imp":11.2147479695,"elo":1461.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.0133340655,"x_train":2526.3066806002,"imp":9.0700293864,"elo":1599.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.020899149,"x_train":0.1991527253,"imp":18.7657004236,"elo":1178.6,"imputed":true,"imputed_pct":66.6666666667,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.020899149,"x_train":0.1991527253,"imp":18.638873156,"elo":1200.8,"imputed":true,"imputed_pct":66.6666666667,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.7156797489,"x_train":10.0952284839,"imp":22.690940904,"elo":983.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0149545776,"x_train":1.4985629713,"imp":29.4235354386,"elo":893.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1995710638,"x_train":20.2619047675,"imp":22.7471582873,"elo":1061.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5149235805,"x_train":20.2619047675,"imp":21.6330567052,"elo":1107.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.020899149,"x_train":0.1991527253,"imp":21.1973502954,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0243850814,"x_train":65.9990639089,"imp":15.7438782463,"elo":1212.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3407391303,"x_train":65.9990639089,"imp":15.0995978866,"elo":1244.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"x_train":6.7532462417,"imp":15.0910499881,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0807033173,"x_train":1719.8202901016,"imp":10.103937795,"elo":1517.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.6746674551,"x_train":1719.8202901016,"imp":8.9170097602,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"x_train":1.8760518371,"imp":10.077321119,"elo":1519.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3859818406,"x_train":735.5848736649,"imp":9.2178209823,"elo":1588.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.7359861516,"x_train":735.5848736649,"imp":8.5761380952,"elo":1633.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.4162023366,"x_train":11.4380569036,"imp":16.9330805211,"elo":1192.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"x_train":46.6209023207,"imp":15.3106058247,"elo":1318.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.6414441586,"x_train":4786.5989331939,"imp":12.6616447819,"elo":1391.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":444.5439481393,"x_train":4786.5989331939,"imp":10.2874153003,"elo":1508.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1795394575,"x_train":0.7535798582,"imp":10.8085465808,"elo":1511.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":13.1188916889,"x_train":27.2884140106,"imp":2.6827874305,"elo":1971.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0451670031,"x_train":0.253543068,"imp":27.7026324503,"elo":937.7,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.5956993553,"x_train":3.9694329831,"imp":13.4063747264,"elo":1333.8,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1186281919,"x_train":0.6015671442,"imp":6.8505424386,"elo":1684.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0470610966,"x_train":2.1306936636,"imp":13.9816848431,"elo":1317.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0538978015,"x_train":1236.8778796591,"imp":11.0894895039,"elo":1466.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5790353938,"x_train":1236.8778796591,"imp":10.2538835957,"elo":1528.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"x_train":2.7558592937,"imp":7.8460750331,"elo":1707.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2147265792,"x_train":1.0099025842,"imp":5.8782765424,"elo":1780.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"x_train":110.7816266415,"imp":30.6853633341,"elo":852.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.723789106,"x_train":3381.7712162192,"imp":25.807196939,"elo":984.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"x_train":3381.7712162192,"imp":25.7806935994,"elo":982.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1271670825,"x_train":0.5069992699,"imp":13.3366678662,"elo":1333.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0388490278,"x_train":2.1848199639,"imp":18.1452782924,"elo":1126.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.050592863,"x_train":1770.5620854906,"imp":14.1135657835,"elo":1266.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0691781292,"x_train":1770.5620854906,"imp":12.4495254577,"elo":1365.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"x_train":0.4869354634,"imp":12.9074172378,"elo":1363.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0664942176,"x_train":282.1320041804,"imp":11.1023313966,"elo":1488.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.5626253891,"x_train":282.1320041804,"imp":10.5460036421,"elo":1532.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":9.6204056883,"x_train":88.2701787746,"imp":18.4877413347,"elo":1173.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.8813375711,"x_train":2067.9528996121,"imp":12.1121479202,"elo":1410.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.1951531214,"x_train":2067.9528996121,"imp":10.2205385819,"elo":1542.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1448756257,"x_train":2.4940806965,"imp":25.9796305322,"elo":1040.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0599221336,"x_train":566.7369825686,"imp":13.9821871779,"elo":1360.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.9511369135,"x_train":566.7369825686,"imp":12.4822741841,"elo":1410.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_medium/tuning_trajectories_explorer.html
index b056cc36a12eb466323563296b6fd8a74e2b3ad4..5e81d41417aff00dfdfc09f27c01611b0e843bd4 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":66.0653250838,"imp":17.0239882366,"elo":1235.1,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":2.4694138287,"imp":9.1083854124,"elo":1581.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":4.4162445685,"imp":9.1159938977,"elo":1586.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.8766122224,"imp":9.0779495775,"elo":1603.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":24.258974221,"imp":9.0588594541,"elo":1607.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":62.0516636378,"imp":8.9467666446,"elo":1619.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":122.5800174334,"imp":8.9070623506,"elo":1624.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":242.6410763635,"imp":8.9234658367,"elo":1623.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":361.1283580261,"imp":8.9312030363,"elo":1622.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":489.9274101037,"imp":8.939773823,"elo":1620.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4789788316,"imp":13.1894545746,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.9589536975,"imp":12.7804689305,"elo":1456.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.356217589,"imp":12.4744183072,"elo":1476.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.7221268742,"imp":12.2391809253,"elo":1493.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.9255203586,"imp":12.0497339763,"elo":1508.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":23.8768976028,"imp":11.913101781,"elo":1517.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":47.751833203,"imp":11.7755666991,"elo":1526.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":71.5860615637,"imp":11.7150159217,"elo":1530.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":95.9169155889,"imp":11.6957695725,"elo":1533.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.1751921611,"imp":17.8803428496,"elo":1217.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":6.6636925925,"imp":17.1941915211,"elo":1242.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":17.2339258054,"imp":16.6116876896,"elo":1269.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":34.5777558766,"imp":16.3560470223,"elo":1282.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":87.9245165634,"imp":16.0521398506,"elo":1299.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":174.9652611714,"imp":15.8994205919,"elo":1310.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":345.0671578857,"imp":15.7386714692,"elo":1321.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":520.7521321905,"imp":15.6825400755,"elo":1326.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":695.6092947225,"imp":15.6013580723,"elo":1332.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.4842583839,"imp":22.7998895097,"elo":971.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.846613752,"imp":20.6597293066,"elo":1035.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.6923175211,"imp":17.4621722506,"elo":1127.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9291798058,"imp":16.3110647365,"elo":1171.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.6147364724,"imp":15.5601823286,"elo":1203.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.7108167547,"imp":15.3929982195,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.9416823722,"imp":15.2669930228,"elo":1224.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.1384318226,"imp":15.2655490595,"elo":1226.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":49.6127891307,"imp":15.3160959667,"elo":1228.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.5145310857,"imp":22.8738415301,"elo":1032.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.1588915157,"imp":21.8537906596,"elo":1057.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.5956312383,"imp":20.5238950426,"elo":1104.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":17.7818624821,"imp":19.7509391562,"elo":1127.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":42.5300885718,"imp":18.7952090418,"elo":1154.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":85.6762822632,"imp":18.216855293,"elo":1170.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":173.75989872,"imp":17.7296246052,"elo":1184.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":261.7851807625,"imp":17.4018458318,"elo":1197.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":351.9412405462,"imp":17.1977570828,"elo":1206.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.2873206973,"imp":14.831406831,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.9316502247,"imp":14.2272850125,"elo":1313.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":2.7286702854,"imp":13.0771089423,"elo":1410.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":6.1477936876,"imp":12.2165878048,"elo":1477.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":16.5209724855,"imp":11.2062569778,"elo":1542.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":32.6797765909,"imp":10.7760390499,"elo":1578.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":65.3538276873,"imp":10.5834838941,"elo":1593.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":98.4212833586,"imp":10.5224169036,"elo":1598.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":131.5620070704,"imp":10.4847692361,"elo":1604.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":6.5532925928,"imp":16.1415758528,"elo":1288.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":8.0637708772,"imp":14.341711205,"elo":1307.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":21.7538395214,"imp":12.8782704772,"elo":1377.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.9177628832,"imp":11.027959171,"elo":1491.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":121.8592858048,"imp":10.3441528707,"elo":1537.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":292.3137710756,"imp":9.6118899681,"elo":1590.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":629.8900400305,"imp":9.3077354843,"elo":1612.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1243.5591974675,"imp":9.1817703032,"elo":1624.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1862.5519447854,"imp":9.1222507032,"elo":1627.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2526.3066806002,"imp":9.0917386137,"elo":1630.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":0.1991527253,"imp":18.7758956884,"elo":1190.9,"imputed":true,"imputed_pct":66.6666666667,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":0.1991527253,"imp":18.6490726329,"elo":1214.0,"imputed":true,"imputed_pct":66.6666666667,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_train":10.0952284839,"imp":22.736140275,"elo":986.4,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.4985629713,"imp":29.6890705123,"elo":905.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.4317773957,"imp":27.150126118,"elo":993.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":16.9493056602,"imp":21.7611748075,"elo":1107.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.2619047675,"imp":21.637241668,"elo":1116.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1991527253,"imp":21.2131448299,"elo":998.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.5021873111,"imp":17.6947708454,"elo":1117.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.4894550512,"imp":16.2525252885,"elo":1182.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":3.1307773192,"imp":15.7036900771,"elo":1212.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":8.0202567964,"imp":15.3404069654,"elo":1237.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":16.2377871584,"imp":15.2141741302,"elo":1246.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":32.6710934428,"imp":15.1267895029,"elo":1256.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":49.1719658028,"imp":15.0853480414,"elo":1260.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":65.9990639089,"imp":15.1111406521,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.7532462417,"imp":15.1005671513,"elo":1275.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.5275786908,"imp":12.9072260047,"elo":1405.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.1241530408,"imp":11.3635581269,"elo":1501.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":82.7422408171,"imp":10.5674293143,"elo":1541.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":213.2761385894,"imp":9.8481783226,"elo":1587.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":427.9091537602,"imp":9.5297368118,"elo":1603.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":848.9386425085,"imp":9.1601227196,"elo":1622.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1278.8937139162,"imp":9.0036425298,"elo":1633.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1719.8202901016,"imp":8.9287868289,"elo":1642.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.8760518371,"imp":10.0918725787,"elo":1534.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.1278647155,"imp":10.018663163,"elo":1544.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.2069927888,"imp":9.8482067878,"elo":1558.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":34.2536042268,"imp":9.7015761863,"elo":1575.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":88.9628219424,"imp":9.3584259435,"elo":1615.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":181.3952243454,"imp":9.000340936,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":363.9172361826,"imp":8.7779080241,"elo":1648.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":547.7650992279,"imp":8.6576561267,"elo":1652.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":735.5848736649,"imp":8.5856394282,"elo":1652.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":11.4380569036,"imp":16.9413141811,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.6209023207,"imp":15.3200410441,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.0348846629,"imp":14.1956495304,"elo":1363.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":129.2960112707,"imp":13.0943832644,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":230.7293949461,"imp":12.6784457778,"elo":1422.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":589.0616829286,"imp":11.7243258163,"elo":1467.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1187.7578308406,"imp":11.1912530117,"elo":1493.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2376.2279314343,"imp":10.8135073436,"elo":1510.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3574.1919132683,"imp":10.4831511393,"elo":1524.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4786.5989331939,"imp":10.296860814,"elo":1529.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.7535798582,"imp":10.8199022081,"elo":1530.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":27.2884140106,"imp":2.6842884759,"elo":1978.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.253543068,"imp":27.7103356025,"elo":939.8,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":3.9694329831,"imp":13.4201320779,"elo":1356.0,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6015671442,"imp":6.8659561097,"elo":1704.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":2.1306936636,"imp":13.9939671673,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.7201323021,"imp":13.1386024727,"elo":1376.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":26.071931656,"imp":12.07671084,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":56.6403905326,"imp":11.465419546,"elo":1472.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":151.6325084544,"imp":10.9150875666,"elo":1505.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":305.4050159003,"imp":10.6347881155,"elo":1526.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":598.166050268,"imp":10.4092043336,"elo":1541.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":909.6246432443,"imp":10.3205639671,"elo":1549.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1236.8778796591,"imp":10.2623069414,"elo":1557.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.7558592937,"imp":7.8605289769,"elo":1736.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.0099025842,"imp":5.8834864036,"elo":1789.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":110.7816266415,"imp":30.6917614221,"elo":846.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":125.5711234695,"imp":30.609802755,"elo":860.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":166.4991276653,"imp":30.4571522335,"elo":866.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":216.8260972869,"imp":30.0179135156,"elo":882.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":435.6268489691,"imp":29.0532719309,"elo":905.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":841.7488888841,"imp":27.8654166105,"elo":931.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1663.6865760139,"imp":27.0808278838,"elo":953.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2487.2049124756,"imp":26.3846780548,"elo":976.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3381.7712162192,"imp":25.7887166563,"elo":1001.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.5069992699,"imp":13.3506819035,"elo":1353.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":2.1848199639,"imp":18.1535807635,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.222603335,"imp":16.9138550883,"elo":1179.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":41.2363997061,"imp":15.5421029397,"elo":1234.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":91.3237652004,"imp":14.7262690578,"elo":1277.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":218.5473202022,"imp":13.8435434487,"elo":1318.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":447.0507548682,"imp":13.3314756024,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":866.7256429369,"imp":12.9284009506,"elo":1369.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1331.6601195387,"imp":12.685478784,"elo":1382.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1770.5620854906,"imp":12.4629978646,"elo":1393.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4869354634,"imp":12.9207034717,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.4341782987,"imp":11.9175965636,"elo":1459.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":7.341922179,"imp":11.3659122277,"elo":1506.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":13.4253918933,"imp":11.0012314263,"elo":1527.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":35.1370362218,"imp":10.7494419639,"elo":1542.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":69.8388019133,"imp":10.6449945049,"elo":1550.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":138.6843072833,"imp":10.5779558918,"elo":1557.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":209.6003170598,"imp":10.5660972584,"elo":1560.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":282.1320041804,"imp":10.5606176671,"elo":1564.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":88.2701787746,"imp":18.4983781039,"elo":1186.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":168.2367198699,"imp":16.3890454014,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":410.883351941,"imp":16.2186243252,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":805.5219751868,"imp":13.5594974448,"elo":1432.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1992.2815124104,"imp":10.2704282323,"elo":1567.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":2067.9528996121,"imp":10.2338269266,"elo":1573.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.4940806965,"imp":25.9875577867,"elo":1045.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.3133350441,"imp":21.3468517179,"elo":1123.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":13.8020981623,"imp":16.1353961932,"elo":1251.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":28.3270720096,"imp":13.9002791385,"elo":1364.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":70.6047440508,"imp":12.9743181235,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":140.5453323732,"imp":12.6248944307,"elo":1431.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":281.1244162931,"imp":12.4891082036,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":422.3532460518,"imp":12.4480188978,"elo":1438.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":566.7369825686,"imp":12.4628187149,"elo":1436.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.6520557525,"x_train":66.0653250838,"imp":17.0239882366,"elo":1235.1,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0102827644,"x_train":2.4694138287,"imp":9.1083854124,"elo":1581.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0406807345,"x_train":4.4162445685,"imp":9.1159938977,"elo":1586.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0688636844,"x_train":12.8766122224,"imp":9.0779495775,"elo":1603.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0926006369,"x_train":24.258974221,"imp":9.0588594541,"elo":1607.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1480097263,"x_train":62.0516636378,"imp":8.9467666446,"elo":1619.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1788978917,"x_train":122.5800174334,"imp":8.9070623506,"elo":1624.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2123832568,"x_train":242.6410763635,"imp":8.9234658367,"elo":1623.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.228573636,"x_train":361.1283580261,"imp":8.9312030363,"elo":1622.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2570153277,"x_train":489.9274101037,"imp":8.939773823,"elo":1620.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0225370945,"x_train":0.4789788316,"imp":13.1894545746,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0470795501,"x_train":0.9589536975,"imp":12.7804689305,"elo":1456.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0965149191,"x_train":2.356217589,"imp":12.4744183072,"elo":1476.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.14469563,"x_train":4.7221268742,"imp":12.2391809253,"elo":1493.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2016668384,"x_train":11.9255203586,"imp":12.0497339763,"elo":1508.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2410485968,"x_train":23.8768976028,"imp":11.913101781,"elo":1517.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2647783876,"x_train":47.751833203,"imp":11.7755666991,"elo":1526.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2890240067,"x_train":71.5860615637,"imp":11.7150159217,"elo":1530.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3064184086,"x_train":95.9169155889,"imp":11.6957695725,"elo":1533.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"x_train":2.1751921611,"imp":17.8803428496,"elo":1217.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0116608277,"x_train":6.6636925925,"imp":17.1941915211,"elo":1242.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0226990233,"x_train":17.2339258054,"imp":16.6116876896,"elo":1269.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333135949,"x_train":34.5777558766,"imp":16.3560470223,"elo":1282.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0497571688,"x_train":87.9245165634,"imp":16.0521398506,"elo":1299.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0593008852,"x_train":174.9652611714,"imp":15.8994205919,"elo":1310.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0709126161,"x_train":345.0671578857,"imp":15.7386714692,"elo":1321.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0763826625,"x_train":520.7521321905,"imp":15.6825400755,"elo":1326.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0851920607,"x_train":695.6092947225,"imp":15.6013580723,"elo":1332.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0733045085,"x_train":0.4842583839,"imp":22.7998895097,"elo":971.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0752221241,"x_train":0.846613752,"imp":20.6597293066,"elo":1035.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0926460136,"x_train":1.6923175211,"imp":17.4621722506,"elo":1127.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1010934443,"x_train":2.9291798058,"imp":16.3110647365,"elo":1171.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1082818488,"x_train":6.6147364724,"imp":15.5601823286,"elo":1203.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.143858289,"x_train":12.7108167547,"imp":15.3929982195,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1608782018,"x_train":24.9416823722,"imp":15.2669930228,"elo":1224.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1877156331,"x_train":37.1384318226,"imp":15.2655490595,"elo":1226.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1990111469,"x_train":49.6127891307,"imp":15.3160959667,"elo":1228.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"x_train":1.5145310857,"imp":22.8738415301,"elo":1032.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2219956901,"x_train":3.1588915157,"imp":21.8537906596,"elo":1057.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4051013545,"x_train":8.5956312383,"imp":20.5238950426,"elo":1104.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6441505685,"x_train":17.7818624821,"imp":19.7509391562,"elo":1127.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9551913221,"x_train":42.5300885718,"imp":18.7952090418,"elo":1154.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2793194595,"x_train":85.6762822632,"imp":18.216855293,"elo":1170.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.5591641512,"x_train":173.75989872,"imp":17.7296246052,"elo":1184.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6900218907,"x_train":261.7851807625,"imp":17.4018458318,"elo":1197.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6589060596,"x_train":351.9412405462,"imp":17.1977570828,"elo":1206.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0447579838,"x_train":0.2873206973,"imp":14.831406831,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4234159248,"x_train":0.9316502247,"imp":14.2272850125,"elo":1313.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7433885464,"x_train":2.7286702854,"imp":13.0771089423,"elo":1410.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9410923145,"x_train":6.1477936876,"imp":12.2165878048,"elo":1477.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3202136893,"x_train":16.5209724855,"imp":11.2062569778,"elo":1542.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4195738396,"x_train":32.6797765909,"imp":10.7760390499,"elo":1578.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.0760086735,"x_train":65.3538276873,"imp":10.5834838941,"elo":1593.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2773590032,"x_train":98.4212833586,"imp":10.5224169036,"elo":1598.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.638673831,"x_train":131.5620070704,"imp":10.4847692361,"elo":1604.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":5.4085730775,"x_train":6.5532925928,"imp":16.1415758528,"elo":1288.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"x_train":8.0637708772,"imp":14.341711205,"elo":1307.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5785744608,"x_train":21.7538395214,"imp":12.8782704772,"elo":1377.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.2947480655,"x_train":60.9177628832,"imp":11.027959171,"elo":1491.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1427076773,"x_train":121.8592858048,"imp":10.3441528707,"elo":1537.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7736882796,"x_train":292.3137710756,"imp":9.6118899681,"elo":1590.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.9630762477,"x_train":629.8900400305,"imp":9.3077354843,"elo":1612.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1679213756,"x_train":1243.5591974675,"imp":9.1817703032,"elo":1624.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.8768173804,"x_train":1862.5519447854,"imp":9.1222507032,"elo":1627.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.10416073,"x_train":2526.3066806002,"imp":9.0917386137,"elo":1630.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.020899149,"x_train":0.1991527253,"imp":18.7758956884,"elo":1190.9,"imputed":true,"imputed_pct":66.6666666667,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.020899149,"x_train":0.1991527253,"imp":18.6490726329,"elo":1214.0,"imputed":true,"imputed_pct":66.6666666667,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.7156797489,"x_train":10.0952284839,"imp":22.736140275,"elo":986.4,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0149545776,"x_train":1.4985629713,"imp":29.6890705123,"elo":905.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1460145839,"x_train":6.4317773957,"imp":27.150126118,"elo":993.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4447962725,"x_train":16.9493056602,"imp":21.7611748075,"elo":1107.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5149235805,"x_train":20.2619047675,"imp":21.637241668,"elo":1116.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.020899149,"x_train":0.1991527253,"imp":21.2131448299,"elo":998.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.045901974,"x_train":0.5021873111,"imp":17.6947708454,"elo":1117.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1128248591,"x_train":1.4894550512,"imp":16.2525252885,"elo":1182.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1367392873,"x_train":3.1307773192,"imp":15.7036900771,"elo":1212.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2025401714,"x_train":8.0202567964,"imp":15.3404069654,"elo":1237.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2523558649,"x_train":16.2377871584,"imp":15.2141741302,"elo":1246.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2972209594,"x_train":32.6710934428,"imp":15.1267895029,"elo":1256.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3310669279,"x_train":49.1719658028,"imp":15.0853480414,"elo":1260.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3414833614,"x_train":65.9990639089,"imp":15.1111406521,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"x_train":6.7532462417,"imp":15.1005671513,"elo":1275.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3080259999,"x_train":15.5275786908,"imp":12.9072260047,"elo":1405.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5106934439,"x_train":39.1241530408,"imp":11.3635581269,"elo":1501.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.7594361579,"x_train":82.7422408171,"imp":10.5674293143,"elo":1541.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.9843404045,"x_train":213.2761385894,"imp":9.8481783226,"elo":1587.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1860326968,"x_train":427.9091537602,"imp":9.5297368118,"elo":1603.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4580786339,"x_train":848.9386425085,"imp":9.1601227196,"elo":1622.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5889939537,"x_train":1278.8937139162,"imp":9.0036425298,"elo":1633.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6682080328,"x_train":1719.8202901016,"imp":8.9287868289,"elo":1642.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6449695981,"x_train":1.8760518371,"imp":10.0918725787,"elo":1534.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4937099645,"x_train":5.1278647155,"imp":10.018663163,"elo":1544.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2513268047,"x_train":16.2069927888,"imp":9.8482067878,"elo":1558.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.9918356458,"x_train":34.2536042268,"imp":9.7015761863,"elo":1575.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4620925312,"x_train":88.9628219424,"imp":9.3584259435,"elo":1615.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.4114827703,"x_train":181.3952243454,"imp":9.000340936,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6236180593,"x_train":363.9172361826,"imp":8.7779080241,"elo":1648.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.9955169691,"x_train":547.7650992279,"imp":8.6576561267,"elo":1652.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.7359861516,"x_train":735.5848736649,"imp":8.5856394282,"elo":1652.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.4162023366,"x_train":11.4380569036,"imp":16.9413141811,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7378122226,"x_train":46.6209023207,"imp":15.3200410441,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":60.7634572276,"x_train":70.0348846629,"imp":14.1956495304,"elo":1363.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":97.7966669957,"x_train":129.2960112707,"imp":13.0943832644,"elo":1405.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":126.1317081836,"x_train":230.7293949461,"imp":12.6784457778,"elo":1422.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":211.892083719,"x_train":589.0616829286,"imp":11.7243258163,"elo":1467.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":283.7834460479,"x_train":1187.7578308406,"imp":11.1912530117,"elo":1493.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":356.1524517375,"x_train":2376.2279314343,"imp":10.8135073436,"elo":1510.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":402.585313069,"x_train":3574.1919132683,"imp":10.4831511393,"elo":1524.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":433.0154440766,"x_train":4786.5989331939,"imp":10.296860814,"elo":1529.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1795394575,"x_train":0.7535798582,"imp":10.8199022081,"elo":1530.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":13.1188916889,"x_train":27.2884140106,"imp":2.6842884759,"elo":1978.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0451670031,"x_train":0.253543068,"imp":27.7103356025,"elo":939.8,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.5956993553,"x_train":3.9694329831,"imp":13.4201320779,"elo":1356.0,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1186281919,"x_train":0.6015671442,"imp":6.8659561097,"elo":1704.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0470610966,"x_train":2.1306936636,"imp":13.9939671673,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1016642016,"x_train":6.7201323021,"imp":13.1386024727,"elo":1376.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.1965833708,"x_train":26.071931656,"imp":12.07671084,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2690374607,"x_train":56.6403905326,"imp":11.465419546,"elo":1472.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3453390813,"x_train":151.6325084544,"imp":10.9150875666,"elo":1505.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4641659542,"x_train":305.4050159003,"imp":10.6347881155,"elo":1526.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.5371776272,"x_train":598.166050268,"imp":10.4092043336,"elo":1541.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.5567799537,"x_train":909.6246432443,"imp":10.3205639671,"elo":1549.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5682121276,"x_train":1236.8778796591,"imp":10.2623069414,"elo":1557.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"x_train":2.7558592937,"imp":7.8605289769,"elo":1736.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2147265792,"x_train":1.0099025842,"imp":5.8834864036,"elo":1789.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"x_train":110.7816266415,"imp":30.6917614221,"elo":846.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9334778523,"x_train":125.5711234695,"imp":30.609802755,"elo":860.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1406475162,"x_train":166.4991276653,"imp":30.4571522335,"elo":866.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.4886555763,"x_train":216.8260972869,"imp":30.0179135156,"elo":882.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.967901931,"x_train":435.6268489691,"imp":29.0532719309,"elo":905.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4310027255,"x_train":841.7488888841,"imp":27.8654166105,"elo":931.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.4436810211,"x_train":1663.6865760139,"imp":27.0808278838,"elo":953.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9435530344,"x_train":2487.2049124756,"imp":26.3846780548,"elo":976.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"x_train":3381.7712162192,"imp":25.7887166563,"elo":1001.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1271670825,"x_train":0.5069992699,"imp":13.3506819035,"elo":1353.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0388490278,"x_train":2.1848199639,"imp":18.1535807635,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0914656589,"x_train":9.222603335,"imp":16.9138550883,"elo":1179.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2165503089,"x_train":41.2363997061,"imp":15.5421029397,"elo":1234.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.357854022,"x_train":91.3237652004,"imp":14.7262690578,"elo":1277.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5218944893,"x_train":218.5473202022,"imp":13.8435434487,"elo":1318.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7429074477,"x_train":447.0507548682,"imp":13.3314756024,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9478582394,"x_train":866.7256429369,"imp":12.9284009506,"elo":1369.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0346365316,"x_train":1331.6601195387,"imp":12.685478784,"elo":1382.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0691781292,"x_train":1770.5620854906,"imp":12.4629978646,"elo":1393.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"x_train":0.4869354634,"imp":12.9207034717,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1793184212,"x_train":2.4341782987,"imp":11.9175965636,"elo":1459.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2826887094,"x_train":7.341922179,"imp":11.3659122277,"elo":1506.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2968304513,"x_train":13.4253918933,"imp":11.0012314263,"elo":1527.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4028365541,"x_train":35.1370362218,"imp":10.7494419639,"elo":1542.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4362798234,"x_train":69.8388019133,"imp":10.6449945049,"elo":1550.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4803960764,"x_train":138.6843072833,"imp":10.5779558918,"elo":1557.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5237722573,"x_train":209.6003170598,"imp":10.5660972584,"elo":1560.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5608426218,"x_train":282.1320041804,"imp":10.5606176671,"elo":1564.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":9.6204056883,"x_train":88.2701787746,"imp":18.4983781039,"elo":1186.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":18.9558180407,"x_train":168.2367198699,"imp":16.3890454014,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":33.2102432394,"x_train":410.883351941,"imp":16.2186243252,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":46.9242331846,"x_train":805.5219751868,"imp":13.5594974448,"elo":1432.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":66.5740410026,"x_train":1992.2815124104,"imp":10.2704282323,"elo":1567.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":65.1951531214,"x_train":2067.9528996121,"imp":10.2338269266,"elo":1573.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1448756257,"x_train":2.4940806965,"imp":25.9875577867,"elo":1045.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.1903662412,"x_train":5.3133350441,"imp":21.3468517179,"elo":1123.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.2632687113,"x_train":13.8020981623,"imp":16.1353961932,"elo":1251.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.381885611,"x_train":28.3270720096,"imp":13.9002791385,"elo":1364.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.5413359982,"x_train":70.6047440508,"imp":12.9743181235,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.6886364123,"x_train":140.5453323732,"imp":12.6248944307,"elo":1431.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.8346908135,"x_train":281.1244162931,"imp":12.4891082036,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":0.8992040717,"x_train":422.3532460518,"imp":12.4480188978,"elo":1438.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":0.9679699005,"x_train":566.7369825686,"imp":12.4628187149,"elo":1436.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..228ba2355eaf9b14ff17f159197ab29438a1dff3
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_all/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "AutoGluon 1.4 (best, 4h)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealMLP (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "CatBoost (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "CatBoost (tuned)", "LightGBM (tuned + ensembled)", "CatBoost (default)", "iLTM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabM (tuned + ensembled)", "RealTabPFN-2.5 (default)", "RealMLP (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "LightGBM (tuned)", "XGBoost (tuned)", "TabM (tuned)", "ModernNCA (tuned)", "ChimeraBoost (tuned)", "xRFM (tuned + ensembled)", "iLTM (tuned)", "ChimeraBoost (default)", "TabDPT (tuned)", "TorchMLP (tuned + ensembled)", "XGBoost (default)", "xRFM (tuned)", "TabICL (default)", "TabSwift (default)", "TabDPT (default)", "TabM (default)", "EBM (tuned + ensembled)", "ModernNCA (default)", "TorchMLP (tuned)", "LimiX (default)", "RealMLP (default)", "LightGBM (default)", "EBM (tuned)", "RandomForest (tuned + ensembled)", "BetaTabPFN (default)", "ExtraTrees (tuned + ensembled)", "RandomForest (tuned)", "EBM (default)", "Nori-30M (default)", "SAP-RPT-OSS (default)", "FastaiMLP (tuned + ensembled)", "Nori (default)", "ExtraTrees (tuned)", "iLTM (default)", "TorchMLP (default)", "PerpetualBooster (tuned + ensembled)", "FastaiMLP (tuned)", "PerpetualBooster (tuned)", "xRFM (default)", "FastaiMLP (default)", "RandomForest (default)", "KNN (tuned + ensembled)", "TabSTAR (tuned)", "OrionMSP (default)", "TabSTAR (tuned + ensembled)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "TabFlex (default)", "Linear (tuned)", "PerpetualBooster (default)", "TabSTAR (default)", "KNN (tuned)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.6666666865348816, 0.7851851582527161, 0.7555555701255798, 0.9333333373069763, 0.9037036895751953, 0.9111111164093018, 0.8814814686775208, 0.9259259104728699, 0.8518518805503845, 0.9185185432434082, 0.8814814686775208, 0.9037036895751953, 0.8592592477798462, 0.8962963223457336, 0.9185185432434082, 0.9925925731658936, 0.9851852059364319, 0.8888888955116272, 0.9629629850387573, 0.8740741014480591, 0.9185185432434082, 0.9777777791023254, 0.9407407641410828, 1.0, 0.9629629850387573, 0.9185185432434082, 0.9259259104728699, 0.9333333373069763, 0.9333333373069763, 0.9259259104728699, 0.9777777791023254, 0.9851852059364319, 0.9407407641410828, 0.970370352268219, 0.970370352268219, 0.9407407641410828, 1.0, 0.9925925731658936, 1.0, 1.0, 0.9037036895751953, 1.0, 1.0, 1.0, 0.9851852059364319, 1.0, 0.9925925731658936, 0.9851852059364319, 1.0, 0.9925925731658936, 0.9851852059364319, 0.9407407641410828, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 0.9851852059364319, 1.0, 0.970370352268219, 0.9851852059364319, 0.9925925731658936, 0.970370352268219, 1.0, 1.0, 1.0, 0.9925925731658936, 0.9851852059364319, 1.0, 0.9851852059364319, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3333333432674408, NaN, 0.3629629611968994, 0.8222222328186035, 0.7333333492279053, 0.5185185074806213, 0.7111111283302307, 0.8962963223457336, 0.8296296000480652, 0.9481481313705444, 0.770370364189148, 0.9407407641410828, 0.9555555582046509, 0.9333333373069763, 0.9555555582046509, 0.970370352268219, 0.8740741014480591, 0.7851851582527161, 0.9111111164093018, 0.7259259223937988, 0.7333333492279053, 0.9481481313705444, 0.9851852059364319, 0.9925925731658936, 0.9111111164093018, 0.9777777791023254, 0.9555555582046509, 0.9629629850387573, 0.9259259104728699, 0.9777777791023254, 0.7777777910232544, 0.9925925731658936, 0.9777777791023254, 0.9925925731658936, 0.9333333373069763, 0.9629629850387573, 0.8518518805503845, 1.0, 0.9925925731658936, 0.9851852059364319, 0.9851852059364319, 0.8592592477798462, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 0.8888888955116272, 0.970370352268219, 1.0, 0.9629629850387573, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.21481481194496155, 0.6370370388031006, NaN, 0.6518518328666687, 0.770370364189148, 0.6592592597007751, 0.8296296000480652, 0.7629629373550415, 0.8592592477798462, 0.7333333492279053, 0.8444444537162781, 0.7259259223937988, 0.7555555701255798, 0.7259259223937988, 0.7851851582527161, 0.7629629373550415, 0.8592592477798462, 0.9185185432434082, 0.770370364189148, 0.9185185432434082, 0.7333333492279053, 0.7851851582527161, 0.8666666746139526, 0.7851851582527161, 0.9037036895751953, 0.8518518805503845, 0.7777777910232544, 0.800000011920929, 0.9111111164093018, 0.7777777910232544, 0.8222222328186035, 0.8888888955116272, 0.8740741014480591, 0.8666666746139526, 0.9111111164093018, 0.9777777791023254, 0.8740741014480591, 0.9925925731658936, 0.8444444537162781, 0.9851852059364319, 0.9777777791023254, 0.8666666746139526, 0.9629629850387573, 0.9925925731658936, 0.8740741014480591, 0.9629629850387573, 0.9629629850387573, 0.9777777791023254, 0.9777777791023254, 0.9185185432434082, 0.9333333373069763, 0.9481481313705444, 0.9333333373069763, 0.9925925731658936, 0.9777777791023254, 0.9925925731658936, 0.9777777791023254, 0.9333333373069763, 0.9555555582046509, 0.9851852059364319, 0.9481481313705444, 0.9555555582046509, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 0.9481481313705444, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 0.970370352268219, 1.0], [0.24444444477558136, 0.17777778208255768, 0.3481481373310089, NaN, 0.5925925970077515, 0.5037037134170532, 0.5037037134170532, 0.7037037014961243, 0.6666666865348816, 0.8222222328186035, 0.6000000238418579, 0.8296296000480652, 0.8740741014480591, 0.8074073791503906, 0.8888888955116272, 0.9111111164093018, 0.7925925850868225, 0.644444465637207, 0.8740741014480591, 0.6296296119689941, 0.6296296119689941, 0.8518518805503845, 0.9111111164093018, 0.9259259104728699, 0.8518518805503845, 0.8370370268821716, 0.9111111164093018, 0.9185185432434082, 0.8962963223457336, 0.9259259104728699, 0.770370364189148, 0.9407407641410828, 0.9481481313705444, 0.970370352268219, 0.8444444537162781, 0.8444444537162781, 0.8296296000480652, 0.9481481313705444, 0.970370352268219, 0.9629629850387573, 0.9851852059364319, 0.8518518805503845, 0.9777777791023254, 0.970370352268219, 0.970370352268219, 0.9925925731658936, 0.9407407641410828, 0.9925925731658936, 0.9851852059364319, 0.970370352268219, 0.8296296000480652, 0.9407407641410828, 0.9777777791023254, 0.9333333373069763, 1.0, 0.9777777791023254, 0.9925925731658936, 0.9851852059364319, 0.9851852059364319, 0.9777777791023254, 1.0, 0.9925925731658936, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.2666666805744171, 0.22962963581085205, 0.40740740299224854, NaN, 0.4592592716217041, 0.5555555820465088, 0.6888889074325562, 0.6740740537643433, 0.6518518328666687, 0.644444465637207, 0.6518518328666687, 0.7333333492279053, 0.6518518328666687, 0.7037037014961243, 0.7185184955596924, 0.7333333492279053, 0.7777777910232544, 0.7925925850868225, 0.7185184955596924, 0.7407407164573669, 0.770370364189148, 0.8296296000480652, 0.8222222328186035, 0.8222222328186035, 0.8814814686775208, 0.800000011920929, 0.8074073791503906, 0.8370370268821716, 0.7851851582527161, 0.8296296000480652, 0.8962963223457336, 0.9407407641410828, 0.8666666746139526, 0.8518518805503845, 0.9111111164093018, 0.8296296000480652, 0.9481481313705444, 0.9185185432434082, 0.9851852059364319, 0.970370352268219, 0.8518518805503845, 1.0, 0.9629629850387573, 0.9777777791023254, 0.970370352268219, 0.9333333373069763, 0.9851852059364319, 0.9925925731658936, 0.9777777791023254, 0.8888888955116272, 0.970370352268219, 0.9111111164093018, 0.9407407641410828, 0.9925925731658936, 0.9629629850387573, 1.0, 0.9925925731658936, 0.9629629850387573, 0.9925925731658936, 0.9481481313705444, 0.9481481313705444, 1.0, 0.9629629850387573, 1.0, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0], [0.0962962955236435, 0.48148149251937866, 0.34074074029922485, 0.4962962865829468, 0.5407407283782959, NaN, 0.7333333492279053, 0.644444465637207, 0.7407407164573669, 0.6740740537643433, 0.7555555701255798, 0.6666666865348816, 0.7111111283302307, 0.6740740537643433, 0.6888889074325562, 0.6814814805984497, 0.7407407164573669, 0.8222222328186035, 0.6814814805984497, 0.7851851582527161, 0.770370364189148, 0.6814814805984497, 0.7777777910232544, 0.6962962746620178, 0.8148148059844971, 0.800000011920929, 0.6814814805984497, 0.7333333492279053, 0.770370364189148, 0.6962962746620178, 0.8222222328186035, 0.8074073791503906, 0.8444444537162781, 0.7481481432914734, 0.8888888955116272, 0.9407407641410828, 0.8222222328186035, 0.9629629850387573, 0.770370364189148, 0.9037036895751953, 0.8814814686775208, 0.8740741014480591, 0.9407407641410828, 0.9481481313705444, 0.8666666746139526, 0.9185185432434082, 0.9481481313705444, 0.9925925731658936, 0.9259259104728699, 0.8888888955116272, 0.9333333373069763, 0.9481481313705444, 0.8962963223457336, 0.9333333373069763, 0.9925925731658936, 0.9481481313705444, 0.9851852059364319, 0.9333333373069763, 0.9555555582046509, 0.9333333373069763, 0.9333333373069763, 0.9481481313705444, 1.0, 0.970370352268219, 0.9851852059364319, 1.0, 0.9851852059364319, 1.0, 0.9481481313705444, 1.0, 0.9629629850387573, 0.9333333373069763, 1.0, 1.0, 0.9777777791023254, 1.0], [0.08888889104127884, 0.2888889014720917, 0.17037037014961243, 0.4962962865829468, 0.4444444477558136, 0.2666666805744171, NaN, 0.6000000238418579, 0.5777778029441833, 0.585185170173645, 0.8222222328186035, 0.6074073910713196, 0.6296296119689941, 0.6000000238418579, 0.614814817905426, 0.6592592597007751, 0.614814817905426, 0.8074073791503906, 0.6888889074325562, 0.6296296119689941, 0.6592592597007751, 0.6074073910713196, 0.6962962746620178, 0.6814814805984497, 0.6962962746620178, 0.7925925850868225, 0.6666666865348816, 0.7629629373550415, 0.7481481432914734, 0.7407407164573669, 0.7925925850868225, 0.8074073791503906, 0.8296296000480652, 0.8074073791503906, 0.800000011920929, 0.9037036895751953, 0.800000011920929, 0.8296296000480652, 0.8222222328186035, 0.8814814686775208, 0.8814814686775208, 0.7777777910232544, 0.8814814686775208, 0.9481481313705444, 0.8814814686775208, 0.9407407641410828, 0.8666666746139526, 0.9777777791023254, 0.970370352268219, 0.9111111164093018, 0.9407407641410828, 0.9037036895751953, 0.8222222328186035, 0.9851852059364319, 0.9777777791023254, 0.9037036895751953, 0.970370352268219, 0.9777777791023254, 0.9185185432434082, 0.9851852059364319, 0.9481481313705444, 0.9333333373069763, 1.0, 0.9407407641410828, 1.0, 0.9925925731658936, 1.0, 0.9925925731658936, 0.8962963223457336, 1.0, 0.9037036895751953, 1.0, 1.0, 1.0, 0.970370352268219, 1.0], [0.11851851642131805, 0.10370370000600815, 0.2370370328426361, 0.29629629850387573, 0.31111112236976624, 0.35555556416511536, 0.4000000059604645, NaN, 0.5333333611488342, 0.5185185074806213, 0.4592592716217041, 0.5481481552124023, 0.6074073910713196, 0.5185185074806213, 0.6370370388031006, 0.6370370388031006, 0.5925925970077515, 0.5925925970077515, 0.770370364189148, 0.5925925970077515, 0.5777778029441833, 0.5185185074806213, 0.7037037014961243, 0.6888889074325562, 0.644444465637207, 0.7185184955596924, 0.614814817905426, 0.8592592477798462, 0.7481481432914734, 0.644444465637207, 0.7185184955596924, 0.8370370268821716, 0.7851851582527161, 0.9185185432434082, 0.7185184955596924, 0.7555555701255798, 0.7481481432914734, 0.9333333373069763, 0.8222222328186035, 0.9037036895751953, 0.8888888955116272, 0.8888888955116272, 1.0, 0.9555555582046509, 0.8814814686775208, 0.9185185432434082, 0.9037036895751953, 0.9333333373069763, 0.9111111164093018, 0.8962963223457336, 0.7925925850868225, 0.9259259104728699, 0.9851852059364319, 0.8666666746139526, 0.9185185432434082, 0.8740741014480591, 0.9925925731658936, 0.9851852059364319, 0.9851852059364319, 0.9925925731658936, 1.0, 0.9777777791023254, 0.9555555582046509, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07407407462596893, 0.17037037014961243, 0.14074073731899261, 0.3333333432674408, 0.32592591643333435, 0.25925925374031067, 0.42222222685813904, 0.46666666865348816, NaN, 0.5259259343147278, 0.4962962865829468, 0.5185185074806213, 0.5777778029441833, 0.5629629492759705, 0.614814817905426, 0.6296296119689941, 0.6000000238418579, 0.6370370388031006, 0.6740740537643433, 0.5481481552124023, 0.6370370388031006, 0.6222222447395325, 0.6296296119689941, 0.644444465637207, 0.6962962746620178, 0.8296296000480652, 0.6962962746620178, 0.7555555701255798, 0.7555555701255798, 0.7481481432914734, 0.7629629373550415, 0.8222222328186035, 0.7629629373550415, 0.8444444537162781, 0.7555555701255798, 0.7851851582527161, 0.8148148059844971, 0.8814814686775208, 0.8222222328186035, 0.970370352268219, 0.8962963223457336, 0.8814814686775208, 0.9259259104728699, 0.9333333373069763, 0.8370370268821716, 0.8592592477798462, 0.8814814686775208, 0.9333333373069763, 0.8518518805503845, 0.8814814686775208, 0.8814814686775208, 0.9037036895751953, 0.9111111164093018, 0.9037036895751953, 0.9259259104728699, 0.8962963223457336, 0.9259259104728699, 0.9925925731658936, 0.9555555582046509, 0.9925925731658936, 1.0, 0.9777777791023254, 1.0, 0.9925925731658936, 0.9481481313705444, 0.970370352268219, 0.9481481313705444, 0.9851852059364319, 0.9333333373069763, 0.9555555582046509, 0.9407407641410828, 1.0, 1.0, 1.0, 0.9851852059364319, 1.0], [0.14814814925193787, 0.051851850003004074, 0.2666666805744171, 0.17777778208255768, 0.3481481373310089, 0.32592591643333435, 0.4148148000240326, 0.48148149251937866, 0.4740740656852722, NaN, 0.4296296238899231, 0.614814817905426, 0.5185185074806213, 0.6666666865348816, 0.6000000238418579, 0.6888889074325562, 0.4888888895511627, 0.5555555820465088, 0.614814817905426, 0.46666666865348816, 0.5333333611488342, 0.6962962746620178, 0.644444465637207, 0.7629629373550415, 0.5555555820465088, 0.6703703999519348, 0.8074073791503906, 0.7185184955596924, 0.7481481432914734, 0.8592592477798462, 0.6370370388031006, 0.7925925850868225, 0.8740741014480591, 0.7925925850868225, 0.7925925850868225, 0.7481481432914734, 0.6888889074325562, 0.8222222328186035, 0.8962963223457336, 0.8222222328186035, 0.9037036895751953, 0.6814814805984497, 0.8888888955116272, 0.9629629850387573, 0.8814814686775208, 0.9259259104728699, 0.8962963223457336, 0.970370352268219, 0.9185185432434082, 0.9629629850387573, 0.7407407164573669, 0.9185185432434082, 0.9111111164093018, 0.7333333492279053, 0.970370352268219, 0.9481481313705444, 0.9629629850387573, 0.9777777791023254, 0.9777777791023254, 0.9925925731658936, 0.9259259104728699, 0.9851852059364319, 0.9925925731658936, 0.9925925731658936, 1.0, 0.9925925731658936, 1.0, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0, 1.0], [0.08148147910833359, 0.22962963581085205, 0.15555556118488312, 0.4000000059604645, 0.35555556416511536, 0.24444444477558136, 0.17777778208255768, 0.5407407283782959, 0.5037037134170532, 0.5703703761100769, NaN, 0.5777778029441833, 0.6000000238418579, 0.5777778029441833, 0.585185170173645, 0.6222222447395325, 0.6074073910713196, 0.7111111283302307, 0.6296296119689941, 0.5777778029441833, 0.6370370388031006, 0.5703703761100769, 0.7037037014961243, 0.6370370388031006, 0.6888889074325562, 0.7407407164573669, 0.6370370388031006, 0.7481481432914734, 0.7185184955596924, 0.7111111283302307, 0.7407407164573669, 0.770370364189148, 0.7851851582527161, 0.7925925850868225, 0.7777777910232544, 0.8888888955116272, 0.7555555701255798, 0.7851851582527161, 0.7777777910232544, 0.8592592477798462, 0.8740741014480591, 0.770370364189148, 0.8444444537162781, 0.9259259104728699, 0.8222222328186035, 0.9037036895751953, 0.8370370268821716, 0.9407407641410828, 0.9555555582046509, 0.8888888955116272, 0.9111111164093018, 0.8888888955116272, 0.800000011920929, 0.9629629850387573, 0.9555555582046509, 0.8814814686775208, 0.9407407641410828, 0.9777777791023254, 0.9185185432434082, 0.9629629850387573, 0.9333333373069763, 0.9259259104728699, 0.9925925731658936, 0.9259259104728699, 0.9925925731658936, 0.9925925731658936, 1.0, 0.9851852059364319, 0.8814814686775208, 0.9925925731658936, 0.9111111164093018, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0], [0.11851851642131805, 0.05925925821065903, 0.27407407760620117, 0.17037037014961243, 0.3481481373310089, 0.3333333432674408, 0.39259257912635803, 0.45185184478759766, 0.48148149251937866, 0.385185182094574, 0.42222222685813904, NaN, 0.4444444477558136, 0.585185170173645, 0.585185170173645, 0.6592592597007751, 0.4962962865829468, 0.5259259343147278, 0.6074073910713196, 0.4592592716217041, 0.5333333611488342, 0.7185184955596924, 0.6370370388031006, 0.7259259223937988, 0.5407407283782959, 0.6370370388031006, 0.8074073791503906, 0.7259259223937988, 0.7037037014961243, 0.8296296000480652, 0.6296296119689941, 0.7851851582527161, 0.8666666746139526, 0.7851851582527161, 0.800000011920929, 0.7407407164573669, 0.6814814805984497, 0.8444444537162781, 0.8888888955116272, 0.8074073791503906, 0.8962963223457336, 0.6814814805984497, 0.8962963223457336, 0.9407407641410828, 0.8814814686775208, 0.9185185432434082, 0.9037036895751953, 0.9481481313705444, 0.9185185432434082, 0.9777777791023254, 0.7407407164573669, 0.9111111164093018, 0.8814814686775208, 0.7333333492279053, 0.9629629850387573, 0.9555555582046509, 0.9555555582046509, 0.9777777791023254, 0.970370352268219, 0.9925925731658936, 0.9333333373069763, 0.9629629850387573, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9851852059364319, 0.9925925731658936, 0.9851852059364319, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0962962955236435, 0.04444444552063942, 0.24444444477558136, 0.1259259283542633, 0.2666666805744171, 0.2888889014720917, 0.37037035822868347, 0.39259257912635803, 0.42222222685813904, 0.48148149251937866, 0.4000000059604645, 0.5555555820465088, NaN, 0.5777778029441833, 0.5407407283782959, 0.6888889074325562, 0.4740740656852722, 0.4962962865829468, 0.5259259343147278, 0.5185185074806213, 0.4962962865829468, 0.644444465637207, 0.8222222328186035, 0.7777777910232544, 0.5925925970077515, 0.614814817905426, 0.7407407164573669, 0.644444465637207, 0.7259259223937988, 0.7481481432914734, 0.6296296119689941, 0.7555555701255798, 0.8814814686775208, 0.7185184955596924, 0.7777777910232544, 0.7259259223937988, 0.7481481432914734, 0.8518518805503845, 0.8444444537162781, 0.8444444537162781, 0.8740741014480591, 0.6592592597007751, 0.8814814686775208, 0.9851852059364319, 0.9111111164093018, 0.9333333373069763, 0.9111111164093018, 0.9851852059364319, 0.9407407641410828, 0.9777777791023254, 0.7481481432914734, 0.9037036895751953, 0.8444444537162781, 0.7481481432914734, 0.9851852059364319, 0.9851852059364319, 0.9555555582046509, 0.9851852059364319, 0.9407407641410828, 0.9851852059364319, 0.9037036895751953, 0.9111111164093018, 1.0, 0.9851852059364319, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9851852059364319, 1.0, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0], [0.14074073731899261, 0.06666667014360428, 0.27407407760620117, 0.192592591047287, 0.3481481373310089, 0.32592591643333435, 0.4000000059604645, 0.48148149251937866, 0.43703705072402954, 0.3333333432674408, 0.42222222685813904, 0.4148148000240326, 0.42222222685813904, NaN, 0.5629629492759705, 0.5777778029441833, 0.4888888895511627, 0.5407407283782959, 0.585185170173645, 0.4592592716217041, 0.5481481552124023, 0.6074073910713196, 0.6074073910713196, 0.6740740537643433, 0.5703703761100769, 0.6518518328666687, 0.7333333492279053, 0.7259259223937988, 0.7333333492279053, 0.800000011920929, 0.6370370388031006, 0.7925925850868225, 0.7925925850868225, 0.800000011920929, 0.8074073791503906, 0.770370364189148, 0.6962962746620178, 0.770370364189148, 0.8740741014480591, 0.8444444537162781, 0.9037036895751953, 0.7037037014961243, 0.8074073791503906, 0.8740741014480591, 0.8740741014480591, 0.8518518805503845, 0.9037036895751953, 0.8814814686775208, 0.8518518805503845, 0.9259259104728699, 0.7407407164573669, 0.8444444537162781, 0.8888888955116272, 0.7333333492279053, 0.8962963223457336, 0.9629629850387573, 0.9185185432434082, 0.9555555582046509, 0.9851852059364319, 0.970370352268219, 0.9259259104728699, 0.970370352268219, 0.9925925731658936, 0.9925925731658936, 1.0, 0.9925925731658936, 1.0, 0.9777777791023254, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10370370000600815, 0.04444444552063942, 0.21481481194496155, 0.1111111119389534, 0.29629629850387573, 0.31111112236976624, 0.385185182094574, 0.3629629611968994, 0.385185182094574, 0.4000000059604645, 0.4148148000240326, 0.4148148000240326, 0.4592592716217041, 0.43703705072402954, NaN, 0.5407407283782959, 0.4888888895511627, 0.5555555820465088, 0.4962962865829468, 0.43703705072402954, 0.5111111402511597, 0.5037037134170532, 0.5777778029441833, 0.6592592597007751, 0.614814817905426, 0.6000000238418579, 0.5777778029441833, 0.6296296119689941, 0.7555555701255798, 0.6222222447395325, 0.6296296119689941, 0.8814814686775208, 0.7481481432914734, 0.7333333492279053, 0.8148148059844971, 0.7037037014961243, 0.6592592597007751, 0.7925925850868225, 0.800000011920929, 0.8074073791503906, 0.9555555582046509, 0.6888889074325562, 0.9407407641410828, 0.9111111164093018, 0.8074073791503906, 0.9111111164093018, 0.9037036895751953, 0.9333333373069763, 0.9185185432434082, 0.8666666746139526, 0.6740740537643433, 0.9111111164093018, 0.970370352268219, 0.7111111283302307, 0.9555555582046509, 0.9851852059364319, 0.9925925731658936, 0.8666666746139526, 0.970370352268219, 0.9333333373069763, 0.9777777791023254, 0.9925925731658936, 0.9851852059364319, 0.9851852059364319, 0.9925925731658936, 0.9851852059364319, 0.9925925731658936, 0.9629629850387573, 0.9851852059364319, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936, 0.9925925731658936], [0.08148147910833359, 0.029629629105329514, 0.2370370328426361, 0.08888889104127884, 0.28148147463798523, 0.3185185194015503, 0.34074074029922485, 0.3629629611968994, 0.37037035822868347, 0.31111112236976624, 0.3777777850627899, 0.34074074029922485, 0.31111112236976624, 0.42222222685813904, 0.4592592716217041, NaN, 0.4888888895511627, 0.5037037134170532, 0.48148149251937866, 0.48148149251937866, 0.5259259343147278, 0.5333333611488342, 0.43703705072402954, 0.7555555701255798, 0.5407407283782959, 0.5629629492759705, 0.5777778029441833, 0.644444465637207, 0.6666666865348816, 0.7259259223937988, 0.6370370388031006, 0.800000011920929, 0.8814814686775208, 0.7407407164573669, 0.7333333492279053, 0.6518518328666687, 0.6814814805984497, 0.770370364189148, 0.8666666746139526, 0.8074073791503906, 0.8296296000480652, 0.6518518328666687, 0.8666666746139526, 0.970370352268219, 0.8962963223457336, 0.9333333373069763, 0.9185185432434082, 0.9481481313705444, 0.9333333373069763, 0.9851852059364319, 0.7037037014961243, 0.9037036895751953, 0.8666666746139526, 0.7333333492279053, 0.9629629850387573, 0.9777777791023254, 0.970370352268219, 0.9037036895751953, 0.9185185432434082, 0.9629629850387573, 0.9111111164093018, 0.9333333373069763, 1.0, 0.9925925731658936, 0.9925925731658936, 1.0, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.007407407276332378, 0.1259259283542633, 0.14074073731899261, 0.2074074000120163, 0.2666666805744171, 0.25925925374031067, 0.385185182094574, 0.40740740299224854, 0.4000000059604645, 0.5111111402511597, 0.39259257912635803, 0.5037037134170532, 0.5259259343147278, 0.5111111402511597, 0.5111111402511597, 0.5111111402511597, NaN, 0.39259257912635803, 0.5111111402511597, 0.48148149251937866, 0.46666666865348816, 0.5185185074806213, 0.6592592597007751, 0.5925925970077515, 0.8074073791503906, 0.614814817905426, 0.5703703761100769, 0.5629629492759705, 0.6888889074325562, 0.614814817905426, 0.614814817905426, 0.6666666865348816, 0.7111111283302307, 0.585185170173645, 0.7185184955596924, 0.7259259223937988, 0.6814814805984497, 0.9777777791023254, 0.7111111283302307, 0.770370364189148, 0.7851851582527161, 0.7111111283302307, 0.9851852059364319, 0.9333333373069763, 0.8296296000480652, 0.8222222328186035, 0.9407407641410828, 0.8888888955116272, 0.8222222328186035, 0.8222222328186035, 0.6814814805984497, 0.8666666746139526, 0.8814814686775208, 0.7407407164573669, 0.9111111164093018, 0.8888888955116272, 0.9851852059364319, 0.9185185432434082, 0.9407407641410828, 0.9333333373069763, 0.8888888955116272, 0.9333333373069763, 0.9925925731658936, 0.9629629850387573, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 0.9629629850387573, 0.9333333373069763, 1.0, 1.0, 0.9851852059364319, 1.0], [0.014814814552664757, 0.21481481194496155, 0.08148147910833359, 0.35555556416511536, 0.2222222238779068, 0.17777778208255768, 0.192592591047287, 0.40740740299224854, 0.3629629611968994, 0.4444444477558136, 0.2888889014720917, 0.4740740656852722, 0.5037037134170532, 0.4592592716217041, 0.4444444477558136, 0.4962962865829468, 0.6074073910713196, NaN, 0.5111111402511597, 0.3777777850627899, 0.5111111402511597, 0.5111111402511597, 0.5925925970077515, 0.5555555820465088, 0.6592592597007751, 0.614814817905426, 0.5481481552124023, 0.6814814805984497, 0.7111111283302307, 0.644444465637207, 0.6370370388031006, 0.7407407164573669, 0.7259259223937988, 0.7185184955596924, 0.7407407164573669, 0.9111111164093018, 0.6962962746620178, 0.7259259223937988, 0.6222222447395325, 0.800000011920929, 0.8666666746139526, 0.7259259223937988, 0.8074073791503906, 0.9185185432434082, 0.6888889074325562, 0.8296296000480652, 0.8592592477798462, 0.9481481313705444, 0.8666666746139526, 0.7481481432914734, 0.8222222328186035, 0.9333333373069763, 0.7777777910232544, 0.8518518805503845, 0.9629629850387573, 0.8888888955116272, 0.970370352268219, 0.9185185432434082, 0.8666666746139526, 0.9259259104728699, 0.9333333373069763, 0.8962963223457336, 0.9851852059364319, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9925925731658936, 0.8962963223457336, 1.0, 0.8962963223457336, 0.9407407641410828, 1.0, 1.0, 0.9481481313705444, 1.0], [0.1111111119389534, 0.08888889104127884, 0.22962963581085205, 0.1259259283542633, 0.2074074000120163, 0.3185185194015503, 0.31111112236976624, 0.22962963581085205, 0.32592591643333435, 0.385185182094574, 0.37037035822868347, 0.39259257912635803, 0.4740740656852722, 0.4148148000240326, 0.5037037134170532, 0.5185185074806213, 0.4888888895511627, 0.4888888895511627, NaN, 0.4592592716217041, 0.5333333611488342, 0.39259257912635803, 0.6000000238418579, 0.6000000238418579, 0.5259259343147278, 0.6222222447395325, 0.4592592716217041, 0.6962962746620178, 0.6518518328666687, 0.5629629492759705, 0.6666666865348816, 0.7777777910232544, 0.7259259223937988, 0.800000011920929, 0.6518518328666687, 0.6074073910713196, 0.6888889074325562, 0.8148148059844971, 0.7481481432914734, 0.8296296000480652, 0.8740741014480591, 0.8370370268821716, 0.9185185432434082, 0.8444444537162781, 0.800000011920929, 0.8888888955116272, 0.8518518805503845, 0.8148148059844971, 0.8814814686775208, 0.8148148059844971, 0.7777777910232544, 0.8592592477798462, 0.9185185432434082, 0.7481481432914734, 0.8592592477798462, 0.8222222328186035, 0.9555555582046509, 0.9481481313705444, 0.9777777791023254, 0.970370352268219, 0.9481481313705444, 0.970370352268219, 0.9407407641410828, 0.9777777791023254, 0.9851852059364319, 0.9851852059364319, 0.9851852059364319, 0.9629629850387573, 0.9851852059364319, 0.9851852059364319, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0], [0.03703703731298447, 0.27407407760620117, 0.08148147910833359, 0.37037035822868347, 0.28148147463798523, 0.21481481194496155, 0.37037035822868347, 0.40740740299224854, 0.45185184478759766, 0.5333333611488342, 0.42222222685813904, 0.5407407283782959, 0.48148149251937866, 0.5407407283782959, 0.5629629492759705, 0.5185185074806213, 0.5185185074806213, 0.6222222447395325, 0.5407407283782959, NaN, 0.46666666865348816, 0.5259259343147278, 0.5259259343147278, 0.5481481552124023, 0.6000000238418579, 0.5703703761100769, 0.585185170173645, 0.6592592597007751, 0.6666666865348816, 0.5777778029441833, 0.6592592597007751, 0.7629629373550415, 0.6666666865348816, 0.7407407164573669, 0.644444465637207, 0.8074073791503906, 0.7925925850868225, 0.7111111283302307, 0.6074073910713196, 0.7777777910232544, 0.8222222328186035, 0.7777777910232544, 0.800000011920929, 0.8592592477798462, 0.6518518328666687, 0.7851851582527161, 0.7407407164573669, 0.8444444537162781, 0.8370370268821716, 0.7555555701255798, 0.8296296000480652, 0.7851851582527161, 0.7629629373550415, 0.9333333373069763, 0.8814814686775208, 0.7925925850868225, 0.8666666746139526, 0.9111111164093018, 0.8296296000480652, 0.9629629850387573, 0.9407407641410828, 0.9185185432434082, 0.970370352268219, 0.9037036895751953, 0.9259259104728699, 0.9333333373069763, 0.9259259104728699, 0.9481481313705444, 0.8222222328186035, 0.9259259104728699, 0.8370370268821716, 1.0, 1.0, 1.0, 0.9185185432434082, 1.0], [0.1259259283542633, 0.2666666805744171, 0.2666666805744171, 0.37037035822868347, 0.25925925374031067, 0.22962963581085205, 0.34074074029922485, 0.42222222685813904, 0.3629629611968994, 0.46666666865348816, 0.3629629611968994, 0.46666666865348816, 0.5037037134170532, 0.45185184478759766, 0.4888888895511627, 0.4740740656852722, 0.5333333611488342, 0.4888888895511627, 0.46666666865348816, 0.5333333611488342, NaN, 0.46666666865348816, 0.5259259343147278, 0.5185185074806213, 0.614814817905426, 0.5555555820465088, 0.5037037134170532, 0.7037037014961243, 0.6296296119689941, 0.6000000238418579, 0.9111111164093018, 0.7185184955596924, 0.6370370388031006, 0.7333333492279053, 0.5703703761100769, 0.7111111283302307, 0.9555555582046509, 0.7259259223937988, 0.6296296119689941, 0.8370370268821716, 0.8148148059844971, 0.7259259223937988, 0.7555555701255798, 0.8518518805503845, 0.6814814805984497, 0.7777777910232544, 0.7481481432914734, 0.7851851582527161, 0.770370364189148, 0.7037037014961243, 0.8814814686775208, 0.8296296000480652, 0.7925925850868225, 0.9481481313705444, 0.8592592477798462, 0.7555555701255798, 0.8518518805503845, 0.9851852059364319, 0.8666666746139526, 0.9851852059364319, 0.9481481313705444, 0.8888888955116272, 0.9925925731658936, 0.8962963223457336, 0.9037036895751953, 0.9925925731658936, 0.9037036895751953, 0.9925925731658936, 0.8370370268821716, 0.9777777791023254, 0.8370370268821716, 1.0, 1.0, 1.0, 0.9259259104728699, 1.0], [0.08148147910833359, 0.051851850003004074, 0.21481481194496155, 0.14814814925193787, 0.22962963581085205, 0.3185185194015503, 0.39259257912635803, 0.48148149251937866, 0.3777777850627899, 0.3037036955356598, 0.4296296238899231, 0.28148147463798523, 0.35555556416511536, 0.39259257912635803, 0.4962962865829468, 0.46666666865348816, 0.48148149251937866, 0.4888888895511627, 0.6074073910713196, 0.4740740656852722, 0.5333333611488342, NaN, 0.4000000059604645, 0.5481481552124023, 0.5481481552124023, 0.6074073910713196, 0.7777777910232544, 0.7037037014961243, 0.6592592597007751, 0.9037036895751953, 0.6222222447395325, 0.6740740537643433, 0.7111111283302307, 0.7407407164573669, 0.6888889074325562, 0.644444465637207, 0.6222222447395325, 0.7407407164573669, 0.8148148059844971, 0.7407407164573669, 0.7777777910232544, 0.6296296119689941, 0.8148148059844971, 0.8444444537162781, 0.8370370268821716, 0.8592592477798462, 0.8518518805503845, 0.8370370268821716, 0.8592592477798462, 0.8888888955116272, 0.7333333492279053, 0.8518518805503845, 0.8370370268821716, 0.7333333492279053, 0.8444444537162781, 0.9037036895751953, 0.8222222328186035, 0.8444444537162781, 0.8592592477798462, 0.8592592477798462, 0.8592592477798462, 0.8518518805503845, 0.9333333373069763, 0.9629629850387573, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9259259104728699, 1.0, 0.9333333373069763, 1.0, 0.8814814686775208, 0.9481481313705444, 1.0, 1.0, 1.0], [0.02222222276031971, 0.014814814552664757, 0.13333334028720856, 0.08888889104127884, 0.17037037014961243, 0.2222222238779068, 0.3037036955356598, 0.29629629850387573, 0.37037035822868347, 0.35555556416511536, 0.29629629850387573, 0.3629629611968994, 0.17777778208255768, 0.39259257912635803, 0.42222222685813904, 0.5629629492759705, 0.34074074029922485, 0.40740740299224854, 0.4000000059604645, 0.4740740656852722, 0.4740740656852722, 0.6000000238418579, NaN, 0.6592592597007751, 0.48148149251937866, 0.5259259343147278, 0.6518518328666687, 0.6074073910713196, 0.6296296119689941, 0.6888889074325562, 0.585185170173645, 0.6518518328666687, 0.8296296000480652, 0.6222222447395325, 0.6814814805984497, 0.6222222447395325, 0.644444465637207, 0.8074073791503906, 0.7851851582527161, 0.7851851582527161, 0.7333333492279053, 0.6296296119689941, 0.8444444537162781, 0.9777777791023254, 0.8222222328186035, 0.8592592477798462, 0.8592592477798462, 0.8888888955116272, 0.8962963223457336, 0.9037036895751953, 0.7407407164573669, 0.8148148059844971, 0.8370370268821716, 0.7333333492279053, 0.9481481313705444, 0.9555555582046509, 0.9333333373069763, 0.9555555582046509, 0.8592592477798462, 0.9777777791023254, 0.8666666746139526, 0.8740741014480591, 0.9925925731658936, 0.9407407641410828, 1.0, 0.9925925731658936, 1.0, 0.9925925731658936, 0.9111111164093018, 0.9925925731658936, 0.9407407641410828, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0], [0.05925925821065903, 0.007407407276332378, 0.21481481194496155, 0.07407407462596893, 0.17777778208255768, 0.3037036955356598, 0.3185185194015503, 0.31111112236976624, 0.35555556416511536, 0.2370370328426361, 0.3629629611968994, 0.27407407760620117, 0.2222222238779068, 0.32592591643333435, 0.34074074029922485, 0.24444444477558136, 0.40740740299224854, 0.4444444477558136, 0.4000000059604645, 0.45185184478759766, 0.48148149251937866, 0.45185184478759766, 0.34074074029922485, NaN, 0.4962962865829468, 0.5555555820465088, 0.5037037134170532, 0.5925925970077515, 0.5925925970077515, 0.6740740537643433, 0.614814817905426, 0.7111111283302307, 0.7777777910232544, 0.6740740537643433, 0.6962962746620178, 0.6296296119689941, 0.6592592597007751, 0.7333333492279053, 0.8074073791503906, 0.7851851582527161, 0.8148148059844971, 0.6370370388031006, 0.8444444537162781, 0.9111111164093018, 0.8740741014480591, 0.9111111164093018, 0.9037036895751953, 0.9259259104728699, 0.9037036895751953, 0.9481481313705444, 0.7037037014961243, 0.8962963223457336, 0.8370370268821716, 0.7333333492279053, 0.9555555582046509, 0.9555555582046509, 0.970370352268219, 0.8592592477798462, 0.9111111164093018, 0.9555555582046509, 0.8666666746139526, 0.9037036895751953, 1.0, 0.9925925731658936, 0.9925925731658936, 1.0, 0.9925925731658936, 0.9925925731658936, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.08888889104127884, 0.0962962955236435, 0.14814814925193787, 0.17777778208255768, 0.18518517911434174, 0.3037036955356598, 0.35555556416511536, 0.3037036955356598, 0.4444444477558136, 0.31111112236976624, 0.4592592716217041, 0.40740740299224854, 0.4296296238899231, 0.385185182094574, 0.4592592716217041, 0.192592591047287, 0.34074074029922485, 0.4740740656852722, 0.4000000059604645, 0.385185182094574, 0.45185184478759766, 0.5185185074806213, 0.5037037134170532, NaN, 0.5185185074806213, 0.5111111402511597, 0.4888888895511627, 0.6518518328666687, 0.5259259343147278, 0.5407407283782959, 0.5777778029441833, 0.6888889074325562, 0.5777778029441833, 0.6592592597007751, 0.644444465637207, 0.585185170173645, 0.8888888955116272, 0.6370370388031006, 0.6888889074325562, 0.7333333492279053, 0.7111111283302307, 0.9333333373069763, 0.8296296000480652, 0.7777777910232544, 0.7851851582527161, 0.8666666746139526, 0.8296296000480652, 0.7925925850868225, 0.7925925850868225, 0.6592592597007751, 0.8074073791503906, 0.8666666746139526, 0.6888889074325562, 0.8888888955116272, 0.8370370268821716, 0.9851852059364319, 0.9185185432434082, 0.9185185432434082, 0.9259259104728699, 0.8814814686775208, 0.9407407641410828, 0.9777777791023254, 0.9259259104728699, 1.0, 1.0, 1.0, 0.9851852059364319, 0.9259259104728699, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9555555582046509, 1.0], [0.03703703731298447, 0.02222222276031971, 0.14814814925193787, 0.16296295821666718, 0.11851851642131805, 0.20000000298023224, 0.2074074000120163, 0.28148147463798523, 0.17037037014961243, 0.3296296298503876, 0.25925925374031067, 0.3629629611968994, 0.385185182094574, 0.3481481373310089, 0.4000000059604645, 0.43703705072402954, 0.385185182094574, 0.385185182094574, 0.3777777850627899, 0.4296296238899231, 0.4444444477558136, 0.39259257912635803, 0.4740740656852722, 0.4444444477558136, 0.48148149251937866, NaN, 0.4444444477558136, 0.5703703761100769, 0.644444465637207, 0.5259259343147278, 0.6000000238418579, 0.7185184955596924, 0.585185170173645, 0.6370370388031006, 0.6962962746620178, 0.7037037014961243, 0.7407407164573669, 0.7111111283302307, 0.6222222447395325, 0.8888888955116272, 0.8148148059844971, 0.7333333492279053, 0.7851851582527161, 0.7851851582527161, 0.7037037014961243, 0.7851851582527161, 0.7851851582527161, 0.8296296000480652, 0.800000011920929, 0.7555555701255798, 0.7555555701255798, 0.8222222328186035, 0.7777777910232544, 0.8592592477798462, 0.8370370268821716, 0.8444444537162781, 0.9037036895751953, 0.9037036895751953, 0.9481481313705444, 0.9481481313705444, 0.9185185432434082, 0.9407407641410828, 0.9777777791023254, 0.970370352268219, 0.970370352268219, 0.9851852059364319, 0.970370352268219, 0.9407407641410828, 0.9111111164093018, 0.9777777791023254, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9925925731658936, 1.0], [0.08148147910833359, 0.04444444552063942, 0.2222222238779068, 0.08888889104127884, 0.20000000298023224, 0.3185185194015503, 0.3333333432674408, 0.385185182094574, 0.3037036955356598, 0.192592591047287, 0.3629629611968994, 0.192592591047287, 0.25925925374031067, 0.2666666805744171, 0.42222222685813904, 0.42222222685813904, 0.4296296238899231, 0.45185184478759766, 0.5407407283782959, 0.4148148000240326, 0.4962962865829468, 0.2222222238779068, 0.3481481373310089, 0.4962962865829468, 0.4888888895511627, 0.5555555820465088, NaN, 0.6666666865348816, 0.6000000238418579, 0.7555555701255798, 0.6000000238418579, 0.6296296119689941, 0.6592592597007751, 0.7185184955596924, 0.6518518328666687, 0.6222222447395325, 0.585185170173645, 0.7037037014961243, 0.7333333492279053, 0.7037037014961243, 0.7481481432914734, 0.6222222447395325, 0.8074073791503906, 0.7851851582527161, 0.7851851582527161, 0.8444444537162781, 0.8370370268821716, 0.8074073791503906, 0.8444444537162781, 0.8370370268821716, 0.7111111283302307, 0.8222222328186035, 0.8074073791503906, 0.7259259223937988, 0.8222222328186035, 0.8666666746139526, 0.8148148059844971, 0.8296296000480652, 0.8370370268821716, 0.8444444537162781, 0.8518518805503845, 0.8444444537162781, 0.9259259104728699, 0.9481481313705444, 0.9259259104728699, 0.9259259104728699, 0.9259259104728699, 0.9185185432434082, 1.0, 0.9333333373069763, 1.0, 0.8888888955116272, 0.9481481313705444, 1.0, 1.0, 1.0], [0.07407407462596893, 0.03703703731298447, 0.20000000298023224, 0.08148147910833359, 0.192592591047287, 0.2666666805744171, 0.2370370328426361, 0.14074073731899261, 0.24444444477558136, 0.28148147463798523, 0.2518518567085266, 0.27407407760620117, 0.35555556416511536, 0.27407407760620117, 0.37037035822868347, 0.35555556416511536, 0.43703705072402954, 0.3185185194015503, 0.3037036955356598, 0.34074074029922485, 0.29629629850387573, 0.29629629850387573, 0.39259257912635803, 0.40740740299224854, 0.5111111402511597, 0.4296296238899231, 0.3333333432674408, NaN, 0.5037037134170532, 0.39259257912635803, 0.35555556416511536, 0.6000000238418579, 0.5111111402511597, 0.7851851582527161, 0.5629629492759705, 0.5111111402511597, 0.5333333611488342, 0.6962962746620178, 0.6296296119689941, 0.6370370388031006, 0.7185184955596924, 0.7037037014961243, 0.7481481432914734, 0.6370370388031006, 0.6888889074325562, 0.7629629373550415, 0.7555555701255798, 0.8074073791503906, 0.770370364189148, 0.7185184955596924, 0.6740740537643433, 0.7259259223937988, 0.8222222328186035, 0.7185184955596924, 0.8074073791503906, 0.6962962746620178, 0.8370370268821716, 0.8370370268821716, 0.8666666746139526, 0.9037036895751953, 1.0, 0.9111111164093018, 0.9185185432434082, 0.9777777791023254, 0.9037036895751953, 0.9777777791023254, 0.9037036895751953, 0.9851852059364319, 0.9851852059364319, 0.9851852059364319, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9777777791023254, 1.0, 1.0], [0.06666667014360428, 0.07407407462596893, 0.08888889104127884, 0.10370370000600815, 0.16296295821666718, 0.22962963581085205, 0.2518518567085266, 0.2518518567085266, 0.24444444477558136, 0.2518518567085266, 0.28148147463798523, 0.29629629850387573, 0.27407407760620117, 0.2666666805744171, 0.24444444477558136, 0.3333333432674408, 0.31111112236976624, 0.2888889014720917, 0.3481481373310089, 0.3333333432674408, 0.37037035822868347, 0.34074074029922485, 0.37037035822868347, 0.40740740299224854, 0.3481481373310089, 0.35555556416511536, 0.4000000059604645, 0.4962962865829468, NaN, 0.48148149251937866, 0.5259259343147278, 0.5629629492759705, 0.4962962865829468, 0.5407407283782959, 0.644444465637207, 0.5481481552124023, 0.5777778029441833, 0.614814817905426, 0.6074073910713196, 0.6296296119689941, 0.6814814805984497, 0.7111111283302307, 0.7259259223937988, 0.7851851582527161, 0.6518518328666687, 0.8222222328186035, 0.770370364189148, 0.7555555701255798, 0.8370370268821716, 0.6666666865348816, 0.6518518328666687, 0.8296296000480652, 0.8222222328186035, 0.644444465637207, 0.8444444537162781, 0.8370370268821716, 0.9259259104728699, 0.770370364189148, 0.9037036895751953, 0.8222222328186035, 0.9407407641410828, 0.9333333373069763, 0.9777777791023254, 0.9925925731658936, 0.9629629850387573, 0.9851852059364319, 0.9629629850387573, 0.9481481313705444, 0.9333333373069763, 0.9925925731658936, 0.9555555582046509, 0.9407407641410828, 0.9925925731658936, 1.0, 1.0, 1.0], [0.06666667014360428, 0.02222222276031971, 0.2222222238779068, 0.07407407462596893, 0.21481481194496155, 0.3037036955356598, 0.25925925374031067, 0.35555556416511536, 0.2518518567085266, 0.14074073731899261, 0.2888889014720917, 0.17037037014961243, 0.2518518567085266, 0.20000000298023224, 0.3777777850627899, 0.27407407760620117, 0.385185182094574, 0.35555556416511536, 0.43703705072402954, 0.42222222685813904, 0.4000000059604645, 0.0962962955236435, 0.31111112236976624, 0.32592591643333435, 0.4740740656852722, 0.4740740656852722, 0.24444444477558136, 0.6074073910713196, 0.5185185074806213, NaN, 0.5925925970077515, 0.614814817905426, 0.5629629492759705, 0.7037037014961243, 0.5703703761100769, 0.5925925970077515, 0.614814817905426, 0.6074073910713196, 0.6666666865348816, 0.6666666865348816, 0.6962962746620178, 0.6222222447395325, 0.6962962746620178, 0.6666666865348816, 0.7481481432914734, 0.7481481432914734, 0.770370364189148, 0.6592592597007751, 0.7407407164573669, 0.7777777910232544, 0.6814814805984497, 0.7777777910232544, 0.8074073791503906, 0.7333333492279053, 0.7185184955596924, 0.8222222328186035, 0.7851851582527161, 0.7925925850868225, 0.8444444537162781, 0.8296296000480652, 0.8518518805503845, 0.8444444537162781, 0.9333333373069763, 0.9333333373069763, 0.9259259104728699, 0.9333333373069763, 0.9333333373069763, 0.9185185432434082, 1.0, 0.9333333373069763, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9777777791023254, 1.0, 1.0], [0.07407407462596893, 0.2222222238779068, 0.17777778208255768, 0.22962963581085205, 0.17037037014961243, 0.17777778208255768, 0.2074074000120163, 0.28148147463798523, 0.2370370328426361, 0.3629629611968994, 0.25925925374031067, 0.37037035822868347, 0.37037035822868347, 0.3629629611968994, 0.37037035822868347, 0.3629629611968994, 0.385185182094574, 0.3629629611968994, 0.3333333432674408, 0.34074074029922485, 0.08888889104127884, 0.3777777850627899, 0.4148148000240326, 0.385185182094574, 0.4592592716217041, 0.4000000059604645, 0.4000000059604645, 0.644444465637207, 0.4740740656852722, 0.40740740299224854, NaN, 0.5259259343147278, 0.4888888895511627, 0.6222222447395325, 0.4296296238899231, 0.5481481552124023, 0.7740740776062012, 0.5629629492759705, 0.48148149251937866, 0.614814817905426, 0.6740740537643433, 0.6814814805984497, 0.6074073910713196, 0.7037037014961243, 0.6000000238418579, 0.6592592597007751, 0.5925925970077515, 0.7111111283302307, 0.7259259223937988, 0.6074073910713196, 0.7925925850868225, 0.6592592597007751, 0.6518518328666687, 0.8814814686775208, 0.7629629373550415, 0.6296296119689941, 0.7851851582527161, 0.9629629850387573, 0.7851851582527161, 0.970370352268219, 0.9333333373069763, 0.8518518805503845, 0.9333333373069763, 0.8666666746139526, 0.7777777910232544, 0.9407407641410828, 0.7851851582527161, 0.9555555582046509, 0.7925925850868225, 0.9333333373069763, 0.8296296000480652, 1.0, 0.9407407641410828, 0.9851852059364319, 0.8666666746139526, 1.0], [0.02222222276031971, 0.007407407276332378, 0.1111111119389534, 0.05925925821065903, 0.10370370000600815, 0.192592591047287, 0.192592591047287, 0.16296295821666718, 0.17777778208255768, 0.2074074000120163, 0.22962963581085205, 0.21481481194496155, 0.24444444477558136, 0.2074074000120163, 0.11851851642131805, 0.20000000298023224, 0.3333333432674408, 0.25925925374031067, 0.2222222238779068, 0.2370370328426361, 0.28148147463798523, 0.32592591643333435, 0.3481481373310089, 0.2888889014720917, 0.42222222685813904, 0.28148147463798523, 0.37037035822868347, 0.4000000059604645, 0.43703705072402954, 0.385185182094574, 0.4740740656852722, NaN, 0.4740740656852722, 0.45185184478759766, 0.5259259343147278, 0.5185185074806213, 0.5111111402511597, 0.5703703761100769, 0.5481481552124023, 0.5259259343147278, 0.9111111164093018, 0.6000000238418579, 0.6296296119689941, 0.6370370388031006, 0.6592592597007751, 0.7259259223937988, 0.6814814805984497, 0.7555555701255798, 0.7851851582527161, 0.6888889074325562, 0.6666666865348816, 0.8148148059844971, 0.7407407164573669, 0.6592592597007751, 0.800000011920929, 0.8148148059844971, 0.9925925731658936, 0.7333333492279053, 0.8814814686775208, 0.7555555701255798, 0.8222222328186035, 0.8888888955116272, 0.9407407641410828, 0.9407407641410828, 0.9481481313705444, 0.9481481313705444, 0.9481481313705444, 0.9037036895751953, 0.9407407641410828, 0.9481481313705444, 0.9407407641410828, 0.9259259104728699, 1.0, 1.0, 0.970370352268219, 1.0], [0.014814814552664757, 0.02222222276031971, 0.1259259283542633, 0.051851850003004074, 0.05925925821065903, 0.15555556118488312, 0.17037037014961243, 0.21481481194496155, 0.2370370328426361, 0.1259259283542633, 0.21481481194496155, 0.13333334028720856, 0.11851851642131805, 0.2074074000120163, 0.2518518567085266, 0.11851851642131805, 0.2888889014720917, 0.27407407760620117, 0.27407407760620117, 0.3333333432674408, 0.3629629611968994, 0.2888889014720917, 0.17037037014961243, 0.2222222238779068, 0.31111112236976624, 0.4148148000240326, 0.34074074029922485, 0.4888888895511627, 0.5037037134170532, 0.43703705072402954, 0.5111111402511597, 0.5259259343147278, NaN, 0.5481481552124023, 0.5925925970077515, 0.5259259343147278, 0.5185185074806213, 0.4740740656852722, 0.585185170173645, 0.614814817905426, 0.6962962746620178, 0.585185170173645, 0.5629629492759705, 0.5111111402511597, 0.6592592597007751, 0.7629629373550415, 0.7259259223937988, 0.7259259223937988, 0.7851851582527161, 0.7629629373550415, 0.6518518328666687, 0.7629629373550415, 0.7925925850868225, 0.6814814805984497, 0.7555555701255798, 0.8666666746139526, 0.8814814686775208, 0.7481481432914734, 0.8074073791503906, 0.8814814686775208, 0.7777777910232544, 0.8444444537162781, 0.9481481313705444, 0.9259259104728699, 0.9629629850387573, 0.9925925731658936, 0.970370352268219, 0.8740741014480591, 0.9185185432434082, 1.0, 0.9259259104728699, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0], [0.05925925821065903, 0.007407407276332378, 0.13333334028720856, 0.029629629105329514, 0.13333334028720856, 0.2518518567085266, 0.192592591047287, 0.08148147910833359, 0.15555556118488312, 0.2074074000120163, 0.2074074000120163, 0.21481481194496155, 0.28148147463798523, 0.20000000298023224, 0.2666666805744171, 0.25925925374031067, 0.4148148000240326, 0.28148147463798523, 0.20000000298023224, 0.25925925374031067, 0.2666666805744171, 0.25925925374031067, 0.3777777850627899, 0.32592591643333435, 0.42222222685813904, 0.3629629611968994, 0.28148147463798523, 0.21481481194496155, 0.4592592716217041, 0.29629629850387573, 0.3777777850627899, 0.5481481552124023, 0.45185184478759766, NaN, 0.5333333611488342, 0.5185185074806213, 0.4296296238899231, 0.644444465637207, 0.5407407283782959, 0.5777778029441833, 0.6888889074325562, 0.614814817905426, 0.7185184955596924, 0.5925925970077515, 0.6370370388031006, 0.7333333492279053, 0.7555555701255798, 0.7037037014961243, 0.770370364189148, 0.614814817905426, 0.5777778029441833, 0.800000011920929, 0.7777777910232544, 0.585185170173645, 0.7851851582527161, 0.6518518328666687, 0.8222222328186035, 0.8148148059844971, 0.8814814686775208, 0.8444444537162781, 0.9629629850387573, 0.9111111164093018, 0.8666666746139526, 0.9851852059364319, 0.8888888955116272, 0.9111111164093018, 0.8962963223457336, 0.9481481313705444, 0.970370352268219, 0.970370352268219, 0.9925925731658936, 0.9259259104728699, 0.9925925731658936, 1.0, 1.0, 1.0], [0.029629629105329514, 0.06666667014360428, 0.08888889104127884, 0.15555556118488312, 0.14814814925193787, 0.1111111119389534, 0.20000000298023224, 0.28148147463798523, 0.24444444477558136, 0.2074074000120163, 0.2222222238779068, 0.20000000298023224, 0.2222222238779068, 0.192592591047287, 0.18518517911434174, 0.2666666805744171, 0.28148147463798523, 0.25925925374031067, 0.3481481373310089, 0.35555556416511536, 0.4296296238899231, 0.31111112236976624, 0.3185185194015503, 0.3037036955356598, 0.34074074029922485, 0.3037036955356598, 0.3481481373310089, 0.43703705072402954, 0.35555556416511536, 0.4296296238899231, 0.5703703761100769, 0.4740740656852722, 0.40740740299224854, 0.46666666865348816, NaN, 0.5407407283782959, 0.614814817905426, 0.42222222685813904, 0.614814817905426, 0.5925925970077515, 0.5481481552124023, 0.5259259343147278, 0.5333333611488342, 0.5777778029441833, 0.644444465637207, 0.48148149251937866, 0.6259258985519409, 0.6000000238418579, 0.5777778029441833, 0.6740740537643433, 0.6740740537643433, 0.7111111283302307, 0.7481481432914734, 0.6666666865348816, 0.6222222447395325, 0.7407407164573669, 0.7037037014961243, 0.6666666865348816, 0.7851851582527161, 0.7037037014961243, 0.6370370388031006, 0.8518518805503845, 0.8333333134651184, 0.8814814686775208, 0.9481481313705444, 0.8333333134651184, 0.9555555582046509, 0.8518518805503845, 0.8370370268821716, 0.8185185194015503, 0.8370370268821716, 0.9037036895751953, 1.0, 1.0, 0.9555555582046509, 1.0], [0.029629629105329514, 0.03703703731298447, 0.02222222276031971, 0.15555556118488312, 0.08888889104127884, 0.05925925821065903, 0.0962962955236435, 0.24444444477558136, 0.21481481194496155, 0.2518518567085266, 0.1111111119389534, 0.25925925374031067, 0.27407407760620117, 0.22962963581085205, 0.29629629850387573, 0.3481481373310089, 0.27407407760620117, 0.08888889104127884, 0.39259257912635803, 0.192592591047287, 0.2888889014720917, 0.35555556416511536, 0.3777777850627899, 0.37037035822868347, 0.35555556416511536, 0.29629629850387573, 0.3777777850627899, 0.4888888895511627, 0.45185184478759766, 0.40740740299224854, 0.45185184478759766, 0.48148149251937866, 0.4740740656852722, 0.48148149251937866, 0.4592592716217041, NaN, 0.6222222447395325, 0.4888888895511627, 0.5481481552124023, 0.5629629492759705, 0.5777778029441833, 0.5925925970077515, 0.585185170173645, 0.5925925970077515, 0.585185170173645, 0.5555555820465088, 0.6074073910713196, 0.6222222447395325, 0.5777778029441833, 0.6000000238418579, 0.6962962746620178, 0.6666666865348816, 0.7481481432914734, 0.6814814805984497, 0.6666666865348816, 0.6962962746620178, 0.7185184955596924, 0.7925925850868225, 0.7851851582527161, 0.8296296000480652, 0.800000011920929, 0.7851851582527161, 0.8370370268821716, 0.8518518805503845, 0.8666666746139526, 0.8962963223457336, 0.8666666746139526, 0.8814814686775208, 0.8148148059844971, 0.8962963223457336, 0.8370370268821716, 0.9333333373069763, 0.9481481313705444, 0.9777777791023254, 0.8740741014480591, 1.0], [0.05925925821065903, 0.14814814925193787, 0.1259259283542633, 0.17037037014961243, 0.17037037014961243, 0.17777778208255768, 0.20000000298023224, 0.2518518567085266, 0.18518517911434174, 0.31111112236976624, 0.24444444477558136, 0.3185185194015503, 0.2518518567085266, 0.3037036955356598, 0.34074074029922485, 0.3185185194015503, 0.3185185194015503, 0.3037036955356598, 0.31111112236976624, 0.2074074000120163, 0.04444444552063942, 0.3777777850627899, 0.35555556416511536, 0.34074074029922485, 0.4148148000240326, 0.25925925374031067, 0.4148148000240326, 0.46666666865348816, 0.42222222685813904, 0.385185182094574, 0.22592592239379883, 0.4888888895511627, 0.48148149251937866, 0.5703703761100769, 0.385185182094574, 0.3777777850627899, NaN, 0.5555555820465088, 0.48148149251937866, 0.5407407283782959, 0.585185170173645, 0.5703703761100769, 0.5555555820465088, 0.6000000238418579, 0.5333333611488342, 0.6296296119689941, 0.614814817905426, 0.6222222447395325, 0.6592592597007751, 0.5407407283782959, 0.6074073910713196, 0.5555555820465088, 0.585185170173645, 0.7037037014961243, 0.644444465637207, 0.614814817905426, 0.7481481432914734, 0.770370364189148, 0.6740740537643433, 0.8666666746139526, 0.8148148059844971, 0.7555555701255798, 0.7851851582527161, 0.7407407164573669, 0.7777777910232544, 0.8148148059844971, 0.7777777910232544, 0.7777777910232544, 0.800000011920929, 0.9037036895751953, 0.8074073791503906, 0.9851852059364319, 0.8074073791503906, 0.8222222328186035, 0.8296296000480652, 1.0], [0.0, 0.0, 0.007407407276332378, 0.051851850003004074, 0.051851850003004074, 0.03703703731298447, 0.17037037014961243, 0.06666667014360428, 0.11851851642131805, 0.17777778208255768, 0.21481481194496155, 0.15555556118488312, 0.14814814925193787, 0.22962963581085205, 0.2074074000120163, 0.22962963581085205, 0.02222222276031971, 0.27407407760620117, 0.18518517911434174, 0.2888889014720917, 0.27407407760620117, 0.25925925374031067, 0.192592591047287, 0.2666666805744171, 0.1111111119389534, 0.2888889014720917, 0.29629629850387573, 0.3037036955356598, 0.385185182094574, 0.39259257912635803, 0.43703705072402954, 0.4296296238899231, 0.5259259343147278, 0.35555556416511536, 0.5777778029441833, 0.5111111402511597, 0.4444444477558136, NaN, 0.5333333611488342, 0.5629629492759705, 0.5407407283782959, 0.5703703761100769, 0.6814814805984497, 0.614814817905426, 0.6222222447395325, 0.6888889074325562, 0.7185184955596924, 0.6370370388031006, 0.6962962746620178, 0.6518518328666687, 0.5555555820465088, 0.6814814805984497, 0.770370364189148, 0.5481481552124023, 0.6592592597007751, 0.6740740537643433, 0.7259259223937988, 0.8592592477798462, 0.8518518805503845, 0.8814814686775208, 0.7407407164573669, 0.8740741014480591, 0.8666666746139526, 0.8962963223457336, 0.9333333373069763, 0.970370352268219, 0.9333333373069763, 0.9185185432434082, 0.8666666746139526, 0.9555555582046509, 0.8592592477798462, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0], [0.007407407276332378, 0.007407407276332378, 0.15555556118488312, 0.029629629105329514, 0.08148147910833359, 0.22962963581085205, 0.17777778208255768, 0.17777778208255768, 0.17777778208255768, 0.10370370000600815, 0.2222222238779068, 0.1111111119389534, 0.15555556118488312, 0.1259259283542633, 0.20000000298023224, 0.13333334028720856, 0.2888889014720917, 0.3777777850627899, 0.2518518567085266, 0.39259257912635803, 0.37037035822868347, 0.18518517911434174, 0.21481481194496155, 0.192592591047287, 0.3629629611968994, 0.3777777850627899, 0.2666666805744171, 0.37037035822868347, 0.39259257912635803, 0.3333333432674408, 0.5185185074806213, 0.45185184478759766, 0.4148148000240326, 0.4592592716217041, 0.385185182094574, 0.45185184478759766, 0.5185185074806213, 0.46666666865348816, NaN, 0.5481481552124023, 0.644444465637207, 0.46666666865348816, 0.5481481552124023, 0.5185185074806213, 0.8296296000480652, 0.614814817905426, 0.5925925970077515, 0.6666666865348816, 0.644444465637207, 0.9481481313705444, 0.6000000238418579, 0.6666666865348816, 0.644444465637207, 0.6518518328666687, 0.6962962746620178, 0.644444465637207, 0.800000011920929, 0.6518518328666687, 0.7037037014961243, 0.644444465637207, 0.6740740537643433, 0.7851851582527161, 0.7185184955596924, 0.8370370268821716, 0.9777777791023254, 0.7185184955596924, 0.9407407641410828, 0.7259259223937988, 0.9333333373069763, 0.7851851582527161, 0.9481481313705444, 0.7555555701255798, 1.0, 0.9259259104728699, 0.9629629850387573, 0.9333333373069763], [0.0, 0.014814814552664757, 0.014814814552664757, 0.03703703731298447, 0.014814814552664757, 0.0962962955236435, 0.11851851642131805, 0.0962962955236435, 0.029629629105329514, 0.17777778208255768, 0.14074073731899261, 0.192592591047287, 0.15555556118488312, 0.15555556118488312, 0.192592591047287, 0.192592591047287, 0.22962963581085205, 0.20000000298023224, 0.17037037014961243, 0.2222222238779068, 0.16296295821666718, 0.25925925374031067, 0.21481481194496155, 0.21481481194496155, 0.31111112236976624, 0.1111111119389534, 0.29629629850387573, 0.3629629611968994, 0.37037035822868347, 0.3333333432674408, 0.385185182094574, 0.4740740656852722, 0.385185182094574, 0.42222222685813904, 0.40740740299224854, 0.43703705072402954, 0.4592592716217041, 0.43703705072402954, 0.45185184478759766, NaN, 0.5555555820465088, 0.5555555820465088, 0.5259259343147278, 0.5925925970077515, 0.48148149251937866, 0.644444465637207, 0.585185170173645, 0.5777778029441833, 0.7111111283302307, 0.5037037134170532, 0.6888889074325562, 0.6740740537643433, 0.585185170173645, 0.7407407164573669, 0.6370370388031006, 0.6222222447395325, 0.8370370268821716, 0.770370364189148, 0.6962962746620178, 0.8444444537162781, 0.8666666746139526, 0.7851851582527161, 0.9555555582046509, 0.8888888955116272, 0.8296296000480652, 0.9259259104728699, 0.8370370268821716, 0.8888888955116272, 0.800000011920929, 0.9333333373069763, 0.8296296000480652, 0.9481481313705444, 1.0, 1.0, 0.9111111164093018, 1.0], [0.0, 0.014814814552664757, 0.02222222276031971, 0.014814814552664757, 0.029629629105329514, 0.11851851642131805, 0.11851851642131805, 0.1111111119389534, 0.10370370000600815, 0.0962962955236435, 0.1259259283542633, 0.10370370000600815, 0.1259259283542633, 0.0962962955236435, 0.04444444552063942, 0.17037037014961243, 0.21481481194496155, 0.13333334028720856, 0.1259259283542633, 0.17777778208255768, 0.18518517911434174, 0.2222222238779068, 0.2666666805744171, 0.18518517911434174, 0.2666666805744171, 0.18518517911434174, 0.2518518567085266, 0.28148147463798523, 0.3185185194015503, 0.3037036955356598, 0.32592591643333435, 0.08888889104127884, 0.3037036955356598, 0.31111112236976624, 0.45185184478759766, 0.42222222685813904, 0.4148148000240326, 0.4592592716217041, 0.35555556416511536, 0.4444444477558136, NaN, 0.46666666865348816, 0.43703705072402954, 0.5185185074806213, 0.4740740656852722, 0.5703703761100769, 0.5703703761100769, 0.614814817905426, 0.6296296119689941, 0.5185185074806213, 0.6222222447395325, 0.7037037014961243, 0.6370370388031006, 0.614814817905426, 0.6962962746620178, 0.7259259223937988, 0.9259259104728699, 0.7037037014961243, 0.7481481432914734, 0.7333333492279053, 0.770370364189148, 0.7777777910232544, 0.8518518805503845, 0.8962963223457336, 0.8962963223457336, 0.8888888955116272, 0.8962963223457336, 0.8148148059844971, 0.9185185432434082, 0.8962963223457336, 0.9185185432434082, 0.9037036895751953, 0.9851852059364319, 1.0, 0.9481481313705444, 1.0], [0.0962962955236435, 0.14074073731899261, 0.13333334028720856, 0.14814814925193787, 0.14814814925193787, 0.1259259283542633, 0.2222222238779068, 0.1111111119389534, 0.11851851642131805, 0.3185185194015503, 0.22962963581085205, 0.3185185194015503, 0.34074074029922485, 0.29629629850387573, 0.31111112236976624, 0.3481481373310089, 0.2888889014720917, 0.27407407760620117, 0.16296295821666718, 0.2222222238779068, 0.27407407760620117, 0.37037035822868347, 0.37037035822868347, 0.3629629611968994, 0.2888889014720917, 0.2666666805744171, 0.3777777850627899, 0.29629629850387573, 0.2888889014720917, 0.3777777850627899, 0.3185185194015503, 0.4000000059604645, 0.4148148000240326, 0.385185182094574, 0.4740740656852722, 0.40740740299224854, 0.4296296238899231, 0.4296296238899231, 0.5333333611488342, 0.4444444477558136, 0.5333333611488342, NaN, 0.4888888895511627, 0.40740740299224854, 0.5407407283782959, 0.5111111402511597, 0.4962962865829468, 0.48148149251937866, 0.5333333611488342, 0.5407407283782959, 0.6296296119689941, 0.6000000238418579, 0.6222222447395325, 0.5703703761100769, 0.5037037134170532, 0.5259259343147278, 0.5925925970077515, 0.5629629492759705, 0.7111111283302307, 0.6074073910713196, 0.6666666865348816, 0.7407407164573669, 0.770370364189148, 0.7111111283302307, 0.6814814805984497, 0.6888889074325562, 0.6888889074325562, 0.6962962746620178, 0.7629629373550415, 0.7037037014961243, 0.770370364189148, 0.7925925850868225, 0.7481481432914734, 0.7851851582527161, 0.770370364189148, 0.9259259104728699], [0.0, 0.0, 0.03703703731298447, 0.02222222276031971, 0.0, 0.05925925821065903, 0.11851851642131805, 0.0, 0.07407407462596893, 0.1111111119389534, 0.15555556118488312, 0.10370370000600815, 0.11851851642131805, 0.192592591047287, 0.05925925821065903, 0.13333334028720856, 0.014814814552664757, 0.192592591047287, 0.08148147910833359, 0.20000000298023224, 0.24444444477558136, 0.18518517911434174, 0.15555556118488312, 0.15555556118488312, 0.06666667014360428, 0.21481481194496155, 0.192592591047287, 0.2518518567085266, 0.27407407760620117, 0.3037036955356598, 0.39259257912635803, 0.37037035822868347, 0.43703705072402954, 0.28148147463798523, 0.46666666865348816, 0.4148148000240326, 0.4444444477558136, 0.3185185194015503, 0.45185184478759766, 0.4740740656852722, 0.5629629492759705, 0.5111111402511597, NaN, 0.48148149251937866, 0.5333333611488342, 0.614814817905426, 0.5555555820465088, 0.6222222447395325, 0.6740740537643433, 0.5555555820465088, 0.585185170173645, 0.6296296119689941, 0.6666666865348816, 0.5703703761100769, 0.6518518328666687, 0.6740740537643433, 0.7259259223937988, 0.7481481432914734, 0.8814814686775208, 0.7925925850868225, 0.6888889074325562, 0.8962963223457336, 0.8222222328186035, 0.8592592477798462, 0.8444444537162781, 0.8296296000480652, 0.8444444537162781, 0.9185185432434082, 0.8888888955116272, 0.9111111164093018, 0.9185185432434082, 0.9333333373069763, 0.9777777791023254, 1.0, 0.9407407641410828, 1.0], [0.0, 0.0, 0.007407407276332378, 0.029629629105329514, 0.03703703731298447, 0.051851850003004074, 0.051851850003004074, 0.04444444552063942, 0.06666667014360428, 0.03703703731298447, 0.07407407462596893, 0.05925925821065903, 0.014814814552664757, 0.1259259283542633, 0.08888889104127884, 0.029629629105329514, 0.06666667014360428, 0.08148147910833359, 0.15555556118488312, 0.14074073731899261, 0.14814814925193787, 0.15555556118488312, 0.02222222276031971, 0.08888889104127884, 0.17037037014961243, 0.21481481194496155, 0.21481481194496155, 0.3629629611968994, 0.21481481194496155, 0.3333333432674408, 0.29629629850387573, 0.3629629611968994, 0.4888888895511627, 0.40740740299224854, 0.42222222685813904, 0.40740740299224854, 0.4000000059604645, 0.385185182094574, 0.48148149251937866, 0.40740740299224854, 0.48148149251937866, 0.5925925970077515, 0.5185185074806213, NaN, 0.5407407283782959, 0.5925925970077515, 0.6074073910713196, 0.5703703761100769, 0.6000000238418579, 0.5629629492759705, 0.6000000238418579, 0.5777778029441833, 0.6518518328666687, 0.5925925970077515, 0.6370370388031006, 0.6740740537643433, 0.7629629373550415, 0.7333333492279053, 0.7333333492279053, 0.8148148059844971, 0.7925925850868225, 0.800000011920929, 0.9259259104728699, 0.8444444537162781, 0.8962963223457336, 0.9629629850387573, 0.9037036895751953, 0.9407407641410828, 0.7925925850868225, 0.9481481313705444, 0.800000011920929, 0.9333333373069763, 1.0, 1.0, 0.8740741014480591, 1.0], [0.0, 0.0, 0.1259259283542633, 0.029629629105329514, 0.02222222276031971, 0.13333334028720856, 0.11851851642131805, 0.11851851642131805, 0.16296295821666718, 0.11851851642131805, 0.17777778208255768, 0.11851851642131805, 0.08888889104127884, 0.1259259283542633, 0.192592591047287, 0.10370370000600815, 0.17037037014961243, 0.31111112236976624, 0.20000000298023224, 0.3481481373310089, 0.3185185194015503, 0.16296295821666718, 0.17777778208255768, 0.1259259283542633, 0.2222222238779068, 0.29629629850387573, 0.21481481194496155, 0.31111112236976624, 0.3481481373310089, 0.2518518567085266, 0.4000000059604645, 0.34074074029922485, 0.34074074029922485, 0.3629629611968994, 0.35555556416511536, 0.4148148000240326, 0.46666666865348816, 0.3777777850627899, 0.17037037014961243, 0.5185185074806213, 0.5259259343147278, 0.4592592716217041, 0.46666666865348816, 0.4592592716217041, NaN, 0.5333333611488342, 0.5259259343147278, 0.5925925970077515, 0.5555555820465088, 0.8074073791503906, 0.5629629492759705, 0.614814817905426, 0.5703703761100769, 0.6296296119689941, 0.6370370388031006, 0.5555555820465088, 0.7407407164573669, 0.6296296119689941, 0.6888889074325562, 0.6370370388031006, 0.6592592597007751, 0.7481481432914734, 0.6740740537643433, 0.7925925850868225, 0.9481481313705444, 0.6962962746620178, 0.9185185432434082, 0.6888889074325562, 0.9037036895751953, 0.770370364189148, 0.9259259104728699, 0.7333333492279053, 1.0, 0.9185185432434082, 0.9481481313705444, 0.9333333373069763], [0.014814814552664757, 0.0, 0.03703703731298447, 0.007407407276332378, 0.029629629105329514, 0.08148147910833359, 0.05925925821065903, 0.08148147910833359, 0.14074073731899261, 0.07407407462596893, 0.0962962955236435, 0.08148147910833359, 0.06666667014360428, 0.14814814925193787, 0.08888889104127884, 0.06666667014360428, 0.17777778208255768, 0.17037037014961243, 0.1111111119389534, 0.21481481194496155, 0.2222222238779068, 0.14074073731899261, 0.14074073731899261, 0.08888889104127884, 0.21481481194496155, 0.21481481194496155, 0.15555556118488312, 0.2370370328426361, 0.17777778208255768, 0.2518518567085266, 0.34074074029922485, 0.27407407760620117, 0.2370370328426361, 0.2666666805744171, 0.5185185074806213, 0.4444444477558136, 0.37037035822868347, 0.31111112236976624, 0.385185182094574, 0.35555556416511536, 0.4296296238899231, 0.4888888895511627, 0.385185182094574, 0.40740740299224854, 0.46666666865348816, NaN, 0.6222222447395325, 0.5555555820465088, 0.6962962746620178, 0.5185185074806213, 0.6518518328666687, 0.6666666865348816, 0.5925925970077515, 0.644444465637207, 0.6370370388031006, 0.6370370388031006, 0.7111111283302307, 0.6518518328666687, 0.7333333492279053, 0.6814814805984497, 0.6814814805984497, 0.800000011920929, 0.970370352268219, 0.9259259104728699, 0.8666666746139526, 0.9481481313705444, 0.8740741014480591, 0.8592592477798462, 0.8666666746139526, 0.970370352268219, 0.9037036895751953, 0.8444444537162781, 0.9555555582046509, 1.0, 0.970370352268219, 1.0], [0.0, 0.0, 0.03703703731298447, 0.05925925821065903, 0.06666667014360428, 0.051851850003004074, 0.13333334028720856, 0.0962962955236435, 0.11851851642131805, 0.10370370000600815, 0.16296295821666718, 0.0962962955236435, 0.08888889104127884, 0.0962962955236435, 0.0962962955236435, 0.08148147910833359, 0.05925925821065903, 0.14074073731899261, 0.14814814925193787, 0.25925925374031067, 0.2518518567085266, 0.14814814925193787, 0.14074073731899261, 0.0962962955236435, 0.13333334028720856, 0.21481481194496155, 0.16296295821666718, 0.24444444477558136, 0.22962963581085205, 0.22962963581085205, 0.40740740299224854, 0.3185185194015503, 0.27407407760620117, 0.24444444477558136, 0.3740740716457367, 0.39259257912635803, 0.385185182094574, 0.28148147463798523, 0.40740740299224854, 0.4148148000240326, 0.4296296238899231, 0.5037037134170532, 0.4444444477558136, 0.39259257912635803, 0.4740740656852722, 0.3777777850627899, NaN, 0.4740740656852722, 0.4888888895511627, 0.5703703761100769, 0.5185185074806213, 0.6222222447395325, 0.6888889074325562, 0.5111111402511597, 0.4888888895511627, 0.5925925970077515, 0.614814817905426, 0.5925925970077515, 0.800000011920929, 0.614814817905426, 0.5555555820465088, 0.8370370268821716, 0.6777777671813965, 0.9185185432434082, 0.9555555582046509, 0.7370370626449585, 0.9629629850387573, 0.7555555701255798, 0.8370370268821716, 0.8111110925674438, 0.8444444537162781, 0.8370370268821716, 1.0, 1.0, 0.9629629850387573, 1.0], [0.007407407276332378, 0.0, 0.02222222276031971, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.02222222276031971, 0.06666667014360428, 0.06666667014360428, 0.029629629105329514, 0.05925925821065903, 0.051851850003004074, 0.014814814552664757, 0.11851851642131805, 0.06666667014360428, 0.051851850003004074, 0.1111111119389534, 0.051851850003004074, 0.18518517911434174, 0.15555556118488312, 0.21481481194496155, 0.16296295821666718, 0.1111111119389534, 0.07407407462596893, 0.17037037014961243, 0.17037037014961243, 0.192592591047287, 0.192592591047287, 0.24444444477558136, 0.34074074029922485, 0.2888889014720917, 0.24444444477558136, 0.27407407760620117, 0.29629629850387573, 0.4000000059604645, 0.3777777850627899, 0.3777777850627899, 0.3629629611968994, 0.3333333432674408, 0.42222222685813904, 0.385185182094574, 0.5185185074806213, 0.3777777850627899, 0.4296296238899231, 0.40740740299224854, 0.4444444477558136, 0.5259259343147278, NaN, 0.46666666865348816, 0.4148148000240326, 0.6666666865348816, 0.5925925970077515, 0.5407407283782959, 0.6592592597007751, 0.8074073791503906, 0.6074073910713196, 0.5703703761100769, 0.6370370388031006, 0.6222222447395325, 0.6888889074325562, 0.6814814805984497, 0.7629629373550415, 0.9259259104728699, 0.8296296000480652, 0.8148148059844971, 0.8962963223457336, 0.8148148059844971, 0.9851852059364319, 0.8148148059844971, 0.9111111164093018, 0.8592592477798462, 0.8666666746139526, 0.9037036895751953, 0.9333333373069763, 0.9333333373069763, 1.0], [0.014814814552664757, 0.007407407276332378, 0.02222222276031971, 0.014814814552664757, 0.007407407276332378, 0.07407407462596893, 0.029629629105329514, 0.08888889104127884, 0.14814814925193787, 0.08148147910833359, 0.04444444552063942, 0.08148147910833359, 0.05925925821065903, 0.14814814925193787, 0.08148147910833359, 0.06666667014360428, 0.17777778208255768, 0.13333334028720856, 0.11851851642131805, 0.16296295821666718, 0.22962963581085205, 0.14074073731899261, 0.10370370000600815, 0.0962962955236435, 0.2074074000120163, 0.20000000298023224, 0.15555556118488312, 0.22962963581085205, 0.16296295821666718, 0.25925925374031067, 0.27407407760620117, 0.21481481194496155, 0.21481481194496155, 0.22962963581085205, 0.42222222685813904, 0.42222222685813904, 0.34074074029922485, 0.3037036955356598, 0.35555556416511536, 0.2888889014720917, 0.37037035822868347, 0.46666666865348816, 0.32592591643333435, 0.4000000059604645, 0.4444444477558136, 0.3037036955356598, 0.5111111402511597, 0.5333333611488342, NaN, 0.48148149251937866, 0.6333333253860474, 0.644444465637207, 0.5555555820465088, 0.6259258985519409, 0.6074073910713196, 0.6296296119689941, 0.6222222447395325, 0.6222222447395325, 0.6888889074325562, 0.6370370388031006, 0.6518518328666687, 0.770370364189148, 0.8629629611968994, 0.9037036895751953, 0.8592592477798462, 0.8518518805503845, 0.8666666746139526, 0.8222222328186035, 0.8074073791503906, 0.8888888955116272, 0.8518518805503845, 0.8370370268821716, 0.9555555582046509, 1.0, 0.9407407641410828, 1.0], [0.0, 0.0, 0.08148147910833359, 0.029629629105329514, 0.02222222276031971, 0.1111111119389534, 0.08888889104127884, 0.10370370000600815, 0.11851851642131805, 0.03703703731298447, 0.1111111119389534, 0.02222222276031971, 0.02222222276031971, 0.07407407462596893, 0.13333334028720856, 0.014814814552664757, 0.17777778208255768, 0.2518518567085266, 0.18518517911434174, 0.24444444477558136, 0.29629629850387573, 0.1111111119389534, 0.0962962955236435, 0.051851850003004074, 0.2074074000120163, 0.24444444477558136, 0.16296295821666718, 0.28148147463798523, 0.3333333432674408, 0.2222222238779068, 0.39259257912635803, 0.31111112236976624, 0.2370370328426361, 0.385185182094574, 0.32592591643333435, 0.4000000059604645, 0.4592592716217041, 0.3481481373310089, 0.051851850003004074, 0.4962962865829468, 0.48148149251937866, 0.4592592716217041, 0.4444444477558136, 0.43703705072402954, 0.192592591047287, 0.48148149251937866, 0.4296296238899231, 0.585185170173645, 0.5185185074806213, NaN, 0.5481481552124023, 0.5185185074806213, 0.5333333611488342, 0.614814817905426, 0.6370370388031006, 0.5629629492759705, 0.7333333492279053, 0.614814817905426, 0.6888889074325562, 0.6370370388031006, 0.6222222447395325, 0.7111111283302307, 0.6592592597007751, 0.7851851582527161, 0.8592592477798462, 0.6888889074325562, 0.8444444537162781, 0.6814814805984497, 0.8740741014480591, 0.7629629373550415, 0.8962963223457336, 0.7259259223937988, 1.0, 0.8814814686775208, 0.9407407641410828, 0.9333333373069763], [0.007407407276332378, 0.1111111119389534, 0.06666667014360428, 0.17037037014961243, 0.1111111119389534, 0.06666667014360428, 0.05925925821065903, 0.2074074000120163, 0.11851851642131805, 0.25925925374031067, 0.08888889104127884, 0.25925925374031067, 0.2518518567085266, 0.25925925374031067, 0.32592591643333435, 0.29629629850387573, 0.3185185194015503, 0.17777778208255768, 0.2222222238779068, 0.17037037014961243, 0.11851851642131805, 0.2666666805744171, 0.25925925374031067, 0.29629629850387573, 0.34074074029922485, 0.24444444477558136, 0.2888889014720917, 0.32592591643333435, 0.3481481373310089, 0.3185185194015503, 0.2074074000120163, 0.3333333432674408, 0.3481481373310089, 0.42222222685813904, 0.32592591643333435, 0.3037036955356598, 0.39259257912635803, 0.4444444477558136, 0.4000000059604645, 0.31111112236976624, 0.3777777850627899, 0.37037035822868347, 0.4148148000240326, 0.4000000059604645, 0.43703705072402954, 0.3481481373310089, 0.48148149251937866, 0.3333333432674408, 0.36666667461395264, 0.45185184478759766, NaN, 0.4296296238899231, 0.4444444477558136, 0.585185170173645, 0.35555556416511536, 0.42222222685813904, 0.4740740656852722, 0.6370370388031006, 0.4888888895511627, 0.6370370388031006, 0.8666666746139526, 0.5925925970077515, 0.6592592597007751, 0.5407407283782959, 0.6000000238418579, 0.6518518328666687, 0.6000000238418579, 0.7259259223937988, 0.6296296119689941, 0.7185184955596924, 0.6592592597007751, 0.8592592477798462, 0.6740740537643433, 0.770370364189148, 0.6740740537643433, 1.0], [0.014814814552664757, 0.029629629105329514, 0.051851850003004074, 0.05925925821065903, 0.029629629105329514, 0.051851850003004074, 0.0962962955236435, 0.07407407462596893, 0.0962962955236435, 0.08148147910833359, 0.1111111119389534, 0.08888889104127884, 0.0962962955236435, 0.15555556118488312, 0.08888889104127884, 0.0962962955236435, 0.13333334028720856, 0.06666667014360428, 0.14074073731899261, 0.21481481194496155, 0.17037037014961243, 0.14814814925193787, 0.18518517911434174, 0.10370370000600815, 0.192592591047287, 0.17777778208255768, 0.17777778208255768, 0.27407407760620117, 0.17037037014961243, 0.2222222238779068, 0.34074074029922485, 0.18518517911434174, 0.2370370328426361, 0.20000000298023224, 0.2888889014720917, 0.3333333432674408, 0.4444444477558136, 0.3185185194015503, 0.3333333432674408, 0.32592591643333435, 0.29629629850387573, 0.4000000059604645, 0.37037035822868347, 0.42222222685813904, 0.385185182094574, 0.3333333432674408, 0.3777777850627899, 0.40740740299224854, 0.35555556416511536, 0.48148149251937866, 0.5703703761100769, NaN, 0.5333333611488342, 0.614814817905426, 0.5185185074806213, 0.40740740299224854, 0.6000000238418579, 0.6814814805984497, 0.6962962746620178, 0.7037037014961243, 0.6222222447395325, 0.7777777910232544, 0.6888889074325562, 0.8592592477798462, 0.7111111283302307, 0.7185184955596924, 0.7259259223937988, 0.770370364189148, 0.8740741014480591, 0.7851851582527161, 0.8962963223457336, 0.800000011920929, 0.970370352268219, 1.0, 0.9851852059364319, 1.0], [0.05925925821065903, 0.0, 0.06666667014360428, 0.02222222276031971, 0.08888889104127884, 0.10370370000600815, 0.17777778208255768, 0.014814814552664757, 0.08888889104127884, 0.08888889104127884, 0.20000000298023224, 0.11851851642131805, 0.15555556118488312, 0.1111111119389534, 0.029629629105329514, 0.13333334028720856, 0.11851851642131805, 0.2222222238779068, 0.08148147910833359, 0.2370370328426361, 0.2074074000120163, 0.16296295821666718, 0.16296295821666718, 0.16296295821666718, 0.13333334028720856, 0.2222222238779068, 0.192592591047287, 0.17777778208255768, 0.17777778208255768, 0.192592591047287, 0.3481481373310089, 0.25925925374031067, 0.2074074000120163, 0.2222222238779068, 0.2518518567085266, 0.2518518567085266, 0.4148148000240326, 0.22962963581085205, 0.35555556416511536, 0.4148148000240326, 0.3629629611968994, 0.3777777850627899, 0.3333333432674408, 0.3481481373310089, 0.4296296238899231, 0.40740740299224854, 0.31111112236976624, 0.4592592716217041, 0.4444444477558136, 0.46666666865348816, 0.5555555820465088, 0.46666666865348816, NaN, 0.5555555820465088, 0.48148149251937866, 0.45185184478759766, 0.5333333611488342, 0.5703703761100769, 0.8296296000480652, 0.585185170173645, 0.6666666865348816, 0.9777777791023254, 0.6000000238418579, 0.7185184955596924, 0.7185184955596924, 0.5333333611488342, 0.7185184955596924, 0.7555555701255798, 0.9259259104728699, 0.6666666865348816, 0.9333333373069763, 0.8074073791503906, 0.8444444537162781, 0.7925925850868225, 0.9851852059364319, 1.0], [0.007407407276332378, 0.03703703731298447, 0.007407407276332378, 0.06666667014360428, 0.05925925821065903, 0.06666667014360428, 0.014814814552664757, 0.13333334028720856, 0.0962962955236435, 0.2666666805744171, 0.03703703731298447, 0.2666666805744171, 0.2518518567085266, 0.2666666805744171, 0.2888889014720917, 0.2666666805744171, 0.25925925374031067, 0.14814814925193787, 0.2518518567085266, 0.06666667014360428, 0.051851850003004074, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.31111112236976624, 0.14074073731899261, 0.27407407760620117, 0.28148147463798523, 0.35555556416511536, 0.2666666805744171, 0.11851851642131805, 0.34074074029922485, 0.3185185194015503, 0.4148148000240326, 0.3333333432674408, 0.3185185194015503, 0.29629629850387573, 0.45185184478759766, 0.3481481373310089, 0.25925925374031067, 0.385185182094574, 0.4296296238899231, 0.4296296238899231, 0.40740740299224854, 0.37037035822868347, 0.35555556416511536, 0.4888888895511627, 0.34074074029922485, 0.3740740716457367, 0.385185182094574, 0.4148148000240326, 0.385185182094574, 0.4444444477558136, NaN, 0.3629629611968994, 0.4296296238899231, 0.4740740656852722, 0.5777778029441833, 0.4888888895511627, 0.614814817905426, 0.8740741014480591, 0.5925925970077515, 0.6666666865348816, 0.5407407283782959, 0.6000000238418579, 0.6592592597007751, 0.6000000238418579, 0.7333333492279053, 0.6296296119689941, 0.7259259223937988, 0.6592592597007751, 0.8518518805503845, 0.6740740537643433, 0.770370364189148, 0.6740740537643433, 1.0], [0.007407407276332378, 0.0, 0.02222222276031971, 0.0, 0.007407407276332378, 0.007407407276332378, 0.02222222276031971, 0.08148147910833359, 0.07407407462596893, 0.029629629105329514, 0.04444444552063942, 0.03703703731298447, 0.014814814552664757, 0.10370370000600815, 0.04444444552063942, 0.03703703731298447, 0.08888889104127884, 0.03703703731298447, 0.14074073731899261, 0.11851851642131805, 0.14074073731899261, 0.15555556118488312, 0.051851850003004074, 0.04444444552063942, 0.1111111119389534, 0.16296295821666718, 0.17777778208255768, 0.192592591047287, 0.15555556118488312, 0.28148147463798523, 0.2370370328426361, 0.20000000298023224, 0.24444444477558136, 0.21481481194496155, 0.3777777850627899, 0.3333333432674408, 0.35555556416511536, 0.34074074029922485, 0.3037036955356598, 0.3629629611968994, 0.3037036955356598, 0.4962962865829468, 0.3481481373310089, 0.3629629611968994, 0.3629629611968994, 0.3629629611968994, 0.5111111402511597, 0.192592591047287, 0.39259257912635803, 0.3629629611968994, 0.644444465637207, 0.48148149251937866, 0.5185185074806213, 0.6370370388031006, NaN, 0.5037037134170532, 0.5111111402511597, 0.6074073910713196, 0.5925925970077515, 0.6666666865348816, 0.6592592597007751, 0.6962962746620178, 0.9037036895751953, 0.7925925850868225, 0.8074073791503906, 0.8888888955116272, 0.800000011920929, 0.9629629850387573, 0.8074073791503906, 0.9111111164093018, 0.8296296000480652, 0.8592592477798462, 0.8962963223457336, 0.9407407641410828, 0.9259259104728699, 1.0], [0.0, 0.0, 0.007407407276332378, 0.02222222276031971, 0.03703703731298447, 0.051851850003004074, 0.0962962955236435, 0.1259259283542633, 0.10370370000600815, 0.051851850003004074, 0.11851851642131805, 0.04444444552063942, 0.014814814552664757, 0.03703703731298447, 0.014814814552664757, 0.02222222276031971, 0.1111111119389534, 0.1111111119389534, 0.17777778208255768, 0.2074074000120163, 0.24444444477558136, 0.0962962955236435, 0.04444444552063942, 0.04444444552063942, 0.16296295821666718, 0.15555556118488312, 0.13333334028720856, 0.3037036955356598, 0.16296295821666718, 0.17777778208255768, 0.37037035822868347, 0.18518517911434174, 0.13333334028720856, 0.3481481373310089, 0.25925925374031067, 0.3037036955356598, 0.385185182094574, 0.32592591643333435, 0.35555556416511536, 0.3777777850627899, 0.27407407760620117, 0.4740740656852722, 0.32592591643333435, 0.32592591643333435, 0.4444444477558136, 0.3629629611968994, 0.40740740299224854, 0.39259257912635803, 0.37037035822868347, 0.43703705072402954, 0.5777778029441833, 0.5925925970077515, 0.5481481552124023, 0.5703703761100769, 0.4962962865829468, NaN, 0.5703703761100769, 0.5037037134170532, 0.6740740537643433, 0.4962962865829468, 0.6000000238418579, 0.7851851582527161, 0.6740740537643433, 0.800000011920929, 0.8740741014480591, 0.7185184955596924, 0.8814814686775208, 0.6370370388031006, 0.8370370268821716, 0.7481481432914734, 0.8666666746139526, 0.6222222447395325, 0.9925925731658936, 0.8666666746139526, 0.9259259104728699, 0.9555555582046509], [0.0, 0.0, 0.02222222276031971, 0.007407407276332378, 0.0, 0.014814814552664757, 0.029629629105329514, 0.007407407276332378, 0.07407407462596893, 0.03703703731298447, 0.05925925821065903, 0.04444444552063942, 0.04444444552063942, 0.08148147910833359, 0.007407407276332378, 0.029629629105329514, 0.014814814552664757, 0.029629629105329514, 0.04444444552063942, 0.13333334028720856, 0.14814814925193787, 0.17777778208255768, 0.06666667014360428, 0.029629629105329514, 0.014814814552664757, 0.0962962955236435, 0.18518517911434174, 0.16296295821666718, 0.07407407462596893, 0.21481481194496155, 0.21481481194496155, 0.007407407276332378, 0.11851851642131805, 0.17777778208255768, 0.29629629850387573, 0.28148147463798523, 0.2518518567085266, 0.27407407760620117, 0.20000000298023224, 0.16296295821666718, 0.07407407462596893, 0.40740740299224854, 0.27407407760620117, 0.2370370328426361, 0.25925925374031067, 0.2888889014720917, 0.385185182094574, 0.4296296238899231, 0.3777777850627899, 0.2666666805744171, 0.5259259343147278, 0.4000000059604645, 0.46666666865348816, 0.5259259343147278, 0.4888888895511627, 0.4296296238899231, NaN, 0.6000000238418579, 0.6000000238418579, 0.614814817905426, 0.6074073910713196, 0.6370370388031006, 0.6888889074325562, 0.8370370268821716, 0.770370364189148, 0.7777777910232544, 0.7629629373550415, 0.7777777910232544, 0.8222222328186035, 0.800000011920929, 0.8444444537162781, 0.770370364189148, 0.9185185432434082, 0.9925925731658936, 0.9185185432434082, 1.0], [0.0, 0.007407407276332378, 0.06666667014360428, 0.014814814552664757, 0.007407407276332378, 0.06666667014360428, 0.02222222276031971, 0.014814814552664757, 0.007407407276332378, 0.02222222276031971, 0.02222222276031971, 0.02222222276031971, 0.014814814552664757, 0.04444444552063942, 0.13333334028720856, 0.0962962955236435, 0.08148147910833359, 0.08148147910833359, 0.051851850003004074, 0.08888889104127884, 0.014814814552664757, 0.15555556118488312, 0.04444444552063942, 0.14074073731899261, 0.08148147910833359, 0.0962962955236435, 0.17037037014961243, 0.16296295821666718, 0.22962963581085205, 0.2074074000120163, 0.03703703731298447, 0.2666666805744171, 0.2518518567085266, 0.18518517911434174, 0.3333333432674408, 0.2074074000120163, 0.22962963581085205, 0.14074073731899261, 0.3481481373310089, 0.22962963581085205, 0.29629629850387573, 0.43703705072402954, 0.2518518567085266, 0.2666666805744171, 0.37037035822868347, 0.3481481373310089, 0.40740740299224854, 0.3629629611968994, 0.3777777850627899, 0.385185182094574, 0.3629629611968994, 0.3185185194015503, 0.4296296238899231, 0.42222222685813904, 0.39259257912635803, 0.4962962865829468, 0.4000000059604645, NaN, 0.46666666865348816, 0.8962963223457336, 0.644444465637207, 0.5333333611488342, 0.6740740537643433, 0.6296296119689941, 0.5111111402511597, 0.7185184955596924, 0.5111111402511597, 0.7111111283302307, 0.7037037014961243, 0.7259259223937988, 0.7111111283302307, 0.9481481313705444, 0.5777778029441833, 0.770370364189148, 0.7111111283302307, 0.970370352268219], [0.014814814552664757, 0.0, 0.04444444552063942, 0.014814814552664757, 0.03703703731298447, 0.04444444552063942, 0.08148147910833359, 0.014814814552664757, 0.04444444552063942, 0.02222222276031971, 0.08148147910833359, 0.029629629105329514, 0.05925925821065903, 0.014814814552664757, 0.029629629105329514, 0.08148147910833359, 0.05925925821065903, 0.13333334028720856, 0.02222222276031971, 0.17037037014961243, 0.13333334028720856, 0.14074073731899261, 0.14074073731899261, 0.08888889104127884, 0.08148147910833359, 0.051851850003004074, 0.16296295821666718, 0.13333334028720856, 0.0962962955236435, 0.15555556118488312, 0.21481481194496155, 0.11851851642131805, 0.192592591047287, 0.11851851642131805, 0.21481481194496155, 0.21481481194496155, 0.32592591643333435, 0.14814814925193787, 0.29629629850387573, 0.3037036955356598, 0.2518518567085266, 0.2888889014720917, 0.11851851642131805, 0.2666666805744171, 0.31111112236976624, 0.2666666805744171, 0.20000000298023224, 0.3777777850627899, 0.31111112236976624, 0.31111112236976624, 0.5111111402511597, 0.3037036955356598, 0.17037037014961243, 0.5111111402511597, 0.40740740299224854, 0.32592591643333435, 0.4000000059604645, 0.5333333611488342, NaN, 0.5629629492759705, 0.5925925970077515, 0.7777777910232544, 0.5333333611488342, 0.6074073910713196, 0.6370370388031006, 0.4740740656852722, 0.6370370388031006, 0.6666666865348816, 0.7777777910232544, 0.6000000238418579, 0.8222222328186035, 0.7333333492279053, 0.8518518805503845, 0.7851851582527161, 0.9333333373069763, 1.0], [0.0, 0.0, 0.014814814552664757, 0.02222222276031971, 0.007407407276332378, 0.06666667014360428, 0.014814814552664757, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.03703703731298447, 0.007407407276332378, 0.014814814552664757, 0.029629629105329514, 0.06666667014360428, 0.03703703731298447, 0.06666667014360428, 0.07407407462596893, 0.029629629105329514, 0.03703703731298447, 0.014814814552664757, 0.14074073731899261, 0.02222222276031971, 0.04444444552063942, 0.07407407462596893, 0.051851850003004074, 0.15555556118488312, 0.0962962955236435, 0.17777778208255768, 0.17037037014961243, 0.029629629105329514, 0.24444444477558136, 0.11851851642131805, 0.15555556118488312, 0.29629629850387573, 0.17037037014961243, 0.13333334028720856, 0.11851851642131805, 0.35555556416511536, 0.15555556118488312, 0.2666666805744171, 0.39259257912635803, 0.2074074000120163, 0.18518517911434174, 0.3629629611968994, 0.3185185194015503, 0.385185182094574, 0.31111112236976624, 0.3629629611968994, 0.3629629611968994, 0.3629629611968994, 0.29629629850387573, 0.4148148000240326, 0.385185182094574, 0.3333333432674408, 0.5037037134170532, 0.385185182094574, 0.10370370000600815, 0.43703705072402954, NaN, 0.585185170173645, 0.5185185074806213, 0.6518518328666687, 0.6000000238418579, 0.5111111402511597, 0.6962962746620178, 0.5111111402511597, 0.614814817905426, 0.6814814805984497, 0.6592592597007751, 0.6962962746620178, 0.9259259104728699, 0.5407407283782959, 0.7111111283302307, 0.6888889074325562, 0.970370352268219], [0.029629629105329514, 0.0, 0.051851850003004074, 0.0, 0.051851850003004074, 0.06666667014360428, 0.051851850003004074, 0.0, 0.0, 0.07407407462596893, 0.06666667014360428, 0.06666667014360428, 0.0962962955236435, 0.07407407462596893, 0.02222222276031971, 0.08888889104127884, 0.1111111119389534, 0.06666667014360428, 0.051851850003004074, 0.05925925821065903, 0.051851850003004074, 0.14074073731899261, 0.13333334028720856, 0.13333334028720856, 0.11851851642131805, 0.08148147910833359, 0.14814814925193787, 0.0, 0.05925925821065903, 0.14814814925193787, 0.06666667014360428, 0.17777778208255768, 0.2222222238779068, 0.03703703731298447, 0.3629629611968994, 0.20000000298023224, 0.18518517911434174, 0.25925925374031067, 0.32592591643333435, 0.13333334028720856, 0.22962963581085205, 0.3333333432674408, 0.31111112236976624, 0.2074074000120163, 0.34074074029922485, 0.3185185194015503, 0.4444444477558136, 0.3185185194015503, 0.3481481373310089, 0.3777777850627899, 0.13333334028720856, 0.3777777850627899, 0.3333333432674408, 0.1259259283542633, 0.34074074029922485, 0.4000000059604645, 0.39259257912635803, 0.35555556416511536, 0.40740740299224854, 0.4148148000240326, NaN, 0.4592592716217041, 0.4444444477558136, 0.5259259343147278, 0.5333333611488342, 0.5629629492759705, 0.5333333611488342, 0.4888888895511627, 0.48148149251937866, 0.6074073910713196, 0.5185185074806213, 0.614814817905426, 0.5777778029441833, 0.6592592597007751, 0.5259259343147278, 0.9111111164093018], [0.014814814552664757, 0.0, 0.04444444552063942, 0.007407407276332378, 0.051851850003004074, 0.051851850003004074, 0.06666667014360428, 0.02222222276031971, 0.02222222276031971, 0.014814814552664757, 0.07407407462596893, 0.03703703731298447, 0.08888889104127884, 0.029629629105329514, 0.007407407276332378, 0.06666667014360428, 0.06666667014360428, 0.10370370000600815, 0.029629629105329514, 0.08148147910833359, 0.1111111119389534, 0.14814814925193787, 0.1259259283542633, 0.0962962955236435, 0.05925925821065903, 0.05925925821065903, 0.15555556118488312, 0.08888889104127884, 0.06666667014360428, 0.15555556118488312, 0.14814814925193787, 0.1111111119389534, 0.15555556118488312, 0.08888889104127884, 0.14814814925193787, 0.21481481194496155, 0.24444444477558136, 0.1259259283542633, 0.21481481194496155, 0.21481481194496155, 0.2222222238779068, 0.25925925374031067, 0.10370370000600815, 0.20000000298023224, 0.2518518567085266, 0.20000000298023224, 0.16296295821666718, 0.2370370328426361, 0.22962963581085205, 0.2888889014720917, 0.40740740299224854, 0.2222222238779068, 0.02222222276031971, 0.40740740299224854, 0.3037036955356598, 0.21481481194496155, 0.3629629611968994, 0.46666666865348816, 0.2222222238779068, 0.48148149251937866, 0.5407407283782959, NaN, 0.4148148000240326, 0.5333333611488342, 0.5703703761100769, 0.4296296238899231, 0.5777778029441833, 0.43703705072402954, 0.7037037014961243, 0.5407407283782959, 0.7555555701255798, 0.614814817905426, 0.7333333492279053, 0.6740740537643433, 0.7851851582527161, 0.8666666746139526], [0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.0, 0.007407407276332378, 0.014814814552664757, 0.0, 0.007407407276332378, 0.014814814552664757, 0.05925925821065903, 0.029629629105329514, 0.007407407276332378, 0.06666667014360428, 0.007407407276332378, 0.0, 0.02222222276031971, 0.02222222276031971, 0.07407407462596893, 0.08148147910833359, 0.02222222276031971, 0.06666667014360428, 0.06666667014360428, 0.05925925821065903, 0.051851850003004074, 0.13333334028720856, 0.1666666716337204, 0.16296295821666718, 0.21481481194496155, 0.13333334028720856, 0.28148147463798523, 0.04444444552063942, 0.14814814925193787, 0.22962963581085205, 0.17777778208255768, 0.07407407462596893, 0.32592591643333435, 0.029629629105329514, 0.3222222328186035, 0.07407407462596893, 0.13703703880310059, 0.34074074029922485, 0.34074074029922485, 0.31111112236976624, 0.4000000059604645, 0.3333333432674408, 0.0962962955236435, 0.32592591643333435, 0.31111112236976624, 0.32592591643333435, 0.46666666865348816, 0.3481481373310089, 0.5555555820465088, 0.585185170173645, NaN, 0.5185185074806213, 0.5925925970077515, 0.49259260296821594, 0.5925925970077515, 0.585185170173645, 0.6296296119689941, 0.5592592358589172, 0.6592592597007751, 0.7629629373550415, 0.6740740537643433, 0.770370364189148, 0.6740740537643433, 1.0], [0.029629629105329514, 0.007407407276332378, 0.04444444552063942, 0.007407407276332378, 0.03703703731298447, 0.029629629105329514, 0.05925925821065903, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.07407407462596893, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.03703703731298447, 0.06666667014360428, 0.02222222276031971, 0.0962962955236435, 0.10370370000600815, 0.03703703731298447, 0.05925925821065903, 0.007407407276332378, 0.07407407462596893, 0.029629629105329514, 0.051851850003004074, 0.02222222276031971, 0.007407407276332378, 0.06666667014360428, 0.13333334028720856, 0.05925925821065903, 0.07407407462596893, 0.014814814552664757, 0.11851851642131805, 0.14814814925193787, 0.25925925374031067, 0.10370370000600815, 0.16296295821666718, 0.1111111119389534, 0.10370370000600815, 0.2888889014720917, 0.14074073731899261, 0.15555556118488312, 0.2074074000120163, 0.07407407462596893, 0.08148147910833359, 0.17037037014961243, 0.0962962955236435, 0.21481481194496155, 0.4592592716217041, 0.14074073731899261, 0.28148147463798523, 0.4592592716217041, 0.2074074000120163, 0.20000000298023224, 0.16296295821666718, 0.37037035822868347, 0.39259257912635803, 0.4000000059604645, 0.4740740656852722, 0.46666666865348816, 0.48148149251937866, NaN, 0.39259257912635803, 0.40740740299224854, 0.39259257912635803, 0.46666666865348816, 0.7185184955596924, 0.46666666865348816, 0.7555555701255798, 0.6666666865348816, 0.5925925970077515, 1.0, 0.8148148059844971, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.014814814552664757, 0.0, 0.0, 0.051851850003004074, 0.0, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.0, 0.007407407276332378, 0.007407407276332378, 0.0, 0.0, 0.014814814552664757, 0.07407407462596893, 0.0962962955236435, 0.06666667014360428, 0.0, 0.007407407276332378, 0.0, 0.029629629105329514, 0.07407407462596893, 0.0962962955236435, 0.03703703731298447, 0.07407407462596893, 0.2222222238779068, 0.051851850003004074, 0.03703703731298447, 0.1111111119389534, 0.051851850003004074, 0.13333334028720856, 0.2222222238779068, 0.06666667014360428, 0.02222222276031971, 0.17037037014961243, 0.10370370000600815, 0.3185185194015503, 0.15555556118488312, 0.10370370000600815, 0.051851850003004074, 0.13333334028720856, 0.04444444552063942, 0.18518517911434174, 0.14074073731899261, 0.14074073731899261, 0.4000000059604645, 0.2888889014720917, 0.28148147463798523, 0.4000000059604645, 0.192592591047287, 0.1259259283542633, 0.22962963581085205, 0.4888888895511627, 0.3629629611968994, 0.4888888895511627, 0.46666666865348816, 0.4296296238899231, 0.40740740299224854, 0.6074073910713196, NaN, 0.43703705072402954, 0.5185185074806213, 0.5555555820465088, 0.7111111283302307, 0.5925925970077515, 0.7185184955596924, 0.5333333611488342, 1.0, 0.7851851582527161, 0.8296296000480652, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.029629629105329514, 0.007407407276332378, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.007407407276332378, 0.014814814552664757, 0.0, 0.0, 0.0, 0.014814814552664757, 0.06666667014360428, 0.007407407276332378, 0.06666667014360428, 0.007407407276332378, 0.0, 0.0, 0.014814814552664757, 0.07407407462596893, 0.02222222276031971, 0.014814814552664757, 0.06666667014360428, 0.05925925821065903, 0.051851850003004074, 0.007407407276332378, 0.08888889104127884, 0.1666666716337204, 0.10370370000600815, 0.18518517911434174, 0.029629629105329514, 0.28148147463798523, 0.07407407462596893, 0.1111111119389534, 0.31111112236976624, 0.17037037014961243, 0.03703703731298447, 0.3037036955356598, 0.051851850003004074, 0.2629629671573639, 0.10370370000600815, 0.14814814925193787, 0.31111112236976624, 0.3481481373310089, 0.28148147463798523, 0.46666666865348816, 0.34074074029922485, 0.1111111119389534, 0.28148147463798523, 0.2222222238779068, 0.28148147463798523, 0.5259259343147278, 0.3037036955356598, 0.43703705072402954, 0.5703703761100769, 0.5074074268341064, 0.5925925970077515, 0.5629629492759705, NaN, 0.5629629492759705, 0.5925925970077515, 0.6074073910713196, 0.5888888835906982, 0.6222222447395325, 0.6666666865348816, 0.6962962746620178, 0.6592592597007751, 0.6666666865348816, 0.9629629850387573], [0.0, 0.0, 0.0, 0.0, 0.0, 0.014814814552664757, 0.0, 0.0, 0.051851850003004074, 0.0, 0.0, 0.007407407276332378, 0.007407407276332378, 0.0, 0.007407407276332378, 0.007407407276332378, 0.0, 0.0, 0.014814814552664757, 0.07407407462596893, 0.0962962955236435, 0.06666667014360428, 0.0, 0.007407407276332378, 0.0, 0.029629629105329514, 0.07407407462596893, 0.0962962955236435, 0.03703703731298447, 0.06666667014360428, 0.21481481194496155, 0.051851850003004074, 0.029629629105329514, 0.10370370000600815, 0.04444444552063942, 0.13333334028720856, 0.2222222238779068, 0.06666667014360428, 0.05925925821065903, 0.16296295821666718, 0.10370370000600815, 0.31111112236976624, 0.15555556118488312, 0.0962962955236435, 0.08148147910833359, 0.1259259283542633, 0.03703703731298447, 0.18518517911434174, 0.13333334028720856, 0.15555556118488312, 0.4000000059604645, 0.27407407760620117, 0.28148147463798523, 0.4000000059604645, 0.20000000298023224, 0.11851851642131805, 0.2370370328426361, 0.4888888895511627, 0.3629629611968994, 0.4888888895511627, 0.46666666865348816, 0.42222222685813904, 0.40740740299224854, 0.6074073910713196, 0.48148149251937866, 0.43703705072402954, NaN, 0.5555555820465088, 0.7111111283302307, 0.5925925970077515, 0.7185184955596924, 0.5333333611488342, 1.0, 0.7851851582527161, 0.8296296000480652, 0.800000011920929], [0.007407407276332378, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.007407407276332378, 0.0, 0.014814814552664757, 0.007407407276332378, 0.014814814552664757, 0.014814814552664757, 0.007407407276332378, 0.02222222276031971, 0.03703703731298447, 0.007407407276332378, 0.0, 0.007407407276332378, 0.03703703731298447, 0.051851850003004074, 0.007407407276332378, 0.07407407462596893, 0.007407407276332378, 0.007407407276332378, 0.014814814552664757, 0.05925925821065903, 0.08148147910833359, 0.014814814552664757, 0.051851850003004074, 0.08148147910833359, 0.04444444552063942, 0.0962962955236435, 0.1259259283542633, 0.051851850003004074, 0.14814814925193787, 0.11851851642131805, 0.2222222238779068, 0.08148147910833359, 0.27407407760620117, 0.1111111119389534, 0.18518517911434174, 0.3037036955356598, 0.08148147910833359, 0.05925925821065903, 0.31111112236976624, 0.14074073731899261, 0.24444444477558136, 0.014814814552664757, 0.17777778208255768, 0.3185185194015503, 0.27407407760620117, 0.22962963581085205, 0.24444444477558136, 0.2666666805744171, 0.03703703731298447, 0.3629629611968994, 0.2222222238779068, 0.2888889014720917, 0.3333333432674408, 0.385185182094574, 0.5111111402511597, 0.5629629492759705, 0.4148148000240326, 0.5333333611488342, 0.4444444477558136, 0.40740740299224854, 0.4444444477558136, NaN, 0.6370370388031006, 0.5407407283782959, 0.6592592597007751, 0.7259259223937988, 0.5703703761100769, 0.7037037014961243, 0.6814814805984497, 1.0], [0.014814814552664757, 0.0, 0.051851850003004074, 0.0, 0.02222222276031971, 0.051851850003004074, 0.10370370000600815, 0.0, 0.06666667014360428, 0.007407407276332378, 0.11851851642131805, 0.007407407276332378, 0.014814814552664757, 0.007407407276332378, 0.014814814552664757, 0.0, 0.04444444552063942, 0.10370370000600815, 0.014814814552664757, 0.17777778208255768, 0.16296295821666718, 0.0, 0.08888889104127884, 0.0, 0.07407407462596893, 0.08888889104127884, 0.0, 0.014814814552664757, 0.06666667014360428, 0.0, 0.2074074000120163, 0.05925925821065903, 0.08148147910833359, 0.029629629105329514, 0.16296295821666718, 0.18518517911434174, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.08148147910833359, 0.2370370328426361, 0.1111111119389534, 0.2074074000120163, 0.0962962955236435, 0.13333334028720856, 0.16296295821666718, 0.18518517911434174, 0.192592591047287, 0.1259259283542633, 0.37037035822868347, 0.1259259283542633, 0.07407407462596893, 0.37037035822868347, 0.192592591047287, 0.16296295821666718, 0.17777778208255768, 0.29629629850387573, 0.2222222238779068, 0.3185185194015503, 0.5185185074806213, 0.29629629850387573, 0.37037035822868347, 0.28148147463798523, 0.2888889014720917, 0.39259257912635803, 0.2888889014720917, 0.3629629611968994, NaN, 0.4592592716217041, 0.8370370268821716, 0.5111111402511597, 0.3333333432674408, 0.4148148000240326, 0.9555555582046509, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.007407407276332378, 0.007407407276332378, 0.007407407276332378, 0.0, 0.007407407276332378, 0.007407407276332378, 0.0, 0.0, 0.0, 0.014814814552664757, 0.07407407462596893, 0.02222222276031971, 0.06666667014360428, 0.007407407276332378, 0.0, 0.0, 0.02222222276031971, 0.06666667014360428, 0.014814814552664757, 0.007407407276332378, 0.06666667014360428, 0.06666667014360428, 0.051851850003004074, 0.0, 0.029629629105329514, 0.1814814805984497, 0.10370370000600815, 0.0962962955236435, 0.04444444552063942, 0.21481481194496155, 0.06666667014360428, 0.10370370000600815, 0.29629629850387573, 0.08888889104127884, 0.051851850003004074, 0.22962963581085205, 0.029629629105329514, 0.18888889253139496, 0.08888889104127884, 0.1111111119389534, 0.2370370328426361, 0.28148147463798523, 0.21481481194496155, 0.3333333432674408, 0.27407407760620117, 0.08888889104127884, 0.2518518567085266, 0.20000000298023224, 0.27407407760620117, 0.4000000059604645, 0.34074074029922485, 0.39259257912635803, 0.4592592716217041, 0.4407407343387604, 0.5333333611488342, 0.40740740299224854, 0.41111111640930176, 0.40740740299224854, 0.4592592716217041, 0.5407407283782959, NaN, 0.5629629492759705, 0.6666666865348816, 0.5481481552124023, 0.7037037014961243, 0.6518518328666687, 0.8666666746139526], [0.014814814552664757, 0.0, 0.04444444552063942, 0.0, 0.02222222276031971, 0.03703703731298447, 0.0962962955236435, 0.0, 0.05925925821065903, 0.0, 0.08888889104127884, 0.0, 0.007407407276332378, 0.0, 0.007407407276332378, 0.0, 0.03703703731298447, 0.10370370000600815, 0.007407407276332378, 0.16296295821666718, 0.16296295821666718, 0.0, 0.05925925821065903, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.04444444552063942, 0.0, 0.17037037014961243, 0.05925925821065903, 0.07407407462596893, 0.007407407276332378, 0.16296295821666718, 0.16296295821666718, 0.192592591047287, 0.14074073731899261, 0.051851850003004074, 0.17037037014961243, 0.08148147910833359, 0.22962963581085205, 0.08148147910833359, 0.20000000298023224, 0.07407407462596893, 0.0962962955236435, 0.15555556118488312, 0.14074073731899261, 0.14814814925193787, 0.10370370000600815, 0.34074074029922485, 0.10370370000600815, 0.06666667014360428, 0.34074074029922485, 0.17037037014961243, 0.13333334028720856, 0.15555556118488312, 0.2888889014720917, 0.17777778208255768, 0.3037036955356598, 0.48148149251937866, 0.24444444477558136, 0.34074074029922485, 0.24444444477558136, 0.28148147463798523, 0.3777777850627899, 0.28148147463798523, 0.34074074029922485, 0.16296295821666718, 0.43703705072402954, NaN, 0.4962962865829468, 0.3333333432674408, 0.4148148000240326, 0.8814814686775208, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.06666667014360428, 0.05925925821065903, 0.0, 0.0, 0.0, 0.11851851642131805, 0.0, 0.0, 0.06666667014360428, 0.0, 0.1111111119389534, 0.06666667014360428, 0.05925925821065903, 0.13333334028720856, 0.0, 0.07407407462596893, 0.0, 0.07407407462596893, 0.0962962955236435, 0.06666667014360428, 0.014814814552664757, 0.06666667014360428, 0.24444444477558136, 0.051851850003004074, 0.0962962955236435, 0.2074074000120163, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.15555556118488312, 0.16296295821666718, 0.13333334028720856, 0.16296295821666718, 0.27407407760620117, 0.14074073731899261, 0.20000000298023224, 0.192592591047287, 0.14814814925193787, 0.14074073731899261, 0.3777777850627899, 0.22962963581085205, 0.051851850003004074, 0.2666666805744171, 0.07407407462596893, 0.385185182094574, 0.385185182094574, 0.2370370328426361, 0.3333333432674408, 0.46666666865348816, 0.3333333432674408, 0.46666666865348816, 0.27407407760620117, 0.4888888895511627, 0.3333333432674408, 0.5037037134170532, NaN, 0.46666666865348816, 0.45185184478759766, 0.5037037134170532, 0.9185185432434082], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.051851850003004074, 0.0, 0.0, 0.0, 0.0, 0.051851850003004074, 0.06666667014360428, 0.007407407276332378, 0.06666667014360428, 0.05925925821065903, 0.0, 0.0, 0.007407407276332378, 0.0, 0.051851850003004074, 0.192592591047287, 0.0, 0.0, 0.0, 0.014814814552664757, 0.2518518567085266, 0.02222222276031971, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0962962955236435, 0.04444444552063942, 0.0, 0.32592591643333435, 0.029629629105329514, 0.15555556118488312, 0.32592591643333435, 0.10370370000600815, 0.007407407276332378, 0.08148147910833359, 0.42222222685813904, 0.14814814925193787, 0.4592592716217041, 0.42222222685813904, 0.2666666805744171, 0.32592591643333435, 0.40740740299224854, 0.0, 0.3037036955356598, 0.0, 0.4296296238899231, 0.6666666865348816, 0.45185184478759766, 0.6666666865348816, 0.5333333611488342, NaN, 0.6666666865348816, 0.6814814805984497, 0.7851851582527161], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.014814814552664757, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.17777778208255768, 0.0, 0.07407407462596893, 0.0, 0.0, 0.21481481194496155, 0.0, 0.0, 0.08148147910833359, 0.0, 0.0, 0.06666667014360428, 0.0, 0.11851851642131805, 0.22962963581085205, 0.0, 0.2074074000120163, 0.22962963581085205, 0.05925925821065903, 0.13333334028720856, 0.007407407276332378, 0.22962963581085205, 0.21481481194496155, 0.2888889014720917, 0.34074074029922485, 0.32592591643333435, 0.22962963581085205, 0.0, 0.21481481194496155, 0.34074074029922485, 0.21481481194496155, 0.29629629850387573, 0.585185170173645, 0.29629629850387573, 0.585185170173645, 0.5481481552124023, 0.3333333432674408, NaN, 0.585185170173645, 0.9925925731658936], [0.0, 0.0, 0.029629629105329514, 0.0, 0.02222222276031971, 0.02222222276031971, 0.029629629105329514, 0.0, 0.014814814552664757, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.014814814552664757, 0.051851850003004074, 0.0, 0.08148147910833359, 0.07407407462596893, 0.0, 0.02222222276031971, 0.0, 0.04444444552063942, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.13333334028720856, 0.029629629105329514, 0.06666667014360428, 0.0, 0.04444444552063942, 0.1259259283542633, 0.17037037014961243, 0.06666667014360428, 0.03703703731298447, 0.08888889104127884, 0.051851850003004074, 0.22962963581085205, 0.05925925821065903, 0.1259259283542633, 0.051851850003004074, 0.029629629105329514, 0.03703703731298447, 0.06666667014360428, 0.05925925821065903, 0.05925925821065903, 0.32592591643333435, 0.014814814552664757, 0.014814814552664757, 0.32592591643333435, 0.07407407462596893, 0.07407407462596893, 0.08148147910833359, 0.2888889014720917, 0.06666667014360428, 0.31111112236976624, 0.4740740656852722, 0.21481481194496155, 0.32592591643333435, 0.18518517911434174, 0.17037037014961243, 0.3333333432674408, 0.17037037014961243, 0.3185185194015503, 0.04444444552063942, 0.3481481373310089, 0.11851851642131805, 0.4962962865829468, 0.3185185194015503, 0.4148148000240326, NaN, 0.6296296119689941], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.007407407276332378, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.07407407462596893, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.029629629105329514, 0.0, 0.029629629105329514, 0.08888889104127884, 0.13333334028720856, 0.0, 0.0, 0.20000000298023224, 0.03703703731298447, 0.20000000298023224, 0.0, 0.3333333432674408, 0.13333334028720856, 0.3333333432674408, 0.08148147910833359, 0.21481481194496155, 0.007407407276332378, 0.37037035822868347, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9539753111},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9261234574},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8833580259},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8709135811},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8430617282},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8290370375},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7935802445},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7748148164},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7682962938},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7624197534},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7590123472},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7514074048},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7464691389},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7362962956},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7212839513},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7130864208},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7095308683},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.701135806},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6997530886},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6942222204},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6919506176},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6845432103},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6773333317},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6733827142},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6535308658},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6485432084},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.6427654305},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6001975304},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6000000005},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5949629634},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5812839499},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5552592592},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5534814802},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5509135809},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.5240493835},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.523358026},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5091851856},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5079506194},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4998518516},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4823703708},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.456888891},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4563950629},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4544197547},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4454321002},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4439506177},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4361481483},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.4152098781},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4087901233},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.405185187},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.40306173},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.3936296284},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3859753106},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3797530874},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.3726913602},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3720493826},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3674074063},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3255308659},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3085432114},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2959012377},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2695308645},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2529382734},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2407901242},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2221728385},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2124444439},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2105679014},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2102716056},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2097777784},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2031604945},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1795555554},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1791604946},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1554567902},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1520987672},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1293827167},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1109135799},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1085432093},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0335802474}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_all/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..234f59b45262b9aad5942f540f807c0157ddc56c
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,77 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),"AutoGluon 1.4 (best, 4h)",TabPFN-2.6 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),RealMLP (tuned + ensembled),ModernNCA (tuned + ensembled),CatBoost (tuned + ensembled),RealTabPFN-2.5 (tuned),CatBoost (tuned),LightGBM (tuned + ensembled),CatBoost (default),iLTM (tuned + ensembled),XGBoost (tuned + ensembled),TabM (tuned + ensembled),RealTabPFN-2.5 (default),RealMLP (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),ChimeraBoost (tuned + ensembled),LightGBM (tuned),XGBoost (tuned),TabM (tuned),ModernNCA (tuned),ChimeraBoost (tuned),xRFM (tuned + ensembled),iLTM (tuned),ChimeraBoost (default),TabDPT (tuned),TorchMLP (tuned + ensembled),XGBoost (default),xRFM (tuned),TabICL (default),TabSwift (default),TabDPT (default),TabM (default),EBM (tuned + ensembled),ModernNCA (default),TorchMLP (tuned),LimiX (default),RealMLP (default),LightGBM (default),EBM (tuned),RandomForest (tuned + ensembled),BetaTabPFN (default),ExtraTrees (tuned + ensembled),RandomForest (tuned),EBM (default),Nori-30M (default),SAP-RPT-OSS (default),FastaiMLP (tuned + ensembled),Nori (default),ExtraTrees (tuned),iLTM (default),TorchMLP (default),PerpetualBooster (tuned + ensembled),FastaiMLP (tuned),PerpetualBooster (tuned),xRFM (default),FastaiMLP (default),RandomForest (default),KNN (tuned + ensembled),TabSTAR (tuned),OrionMSP (default),TabSTAR (tuned + ensembled),ExtraTrees (default),Linear (tuned + ensembled),TabFlex (default),Linear (tuned),PerpetualBooster (default),TabSTAR (default),KNN (tuned),Linear (default),KNN (default)
+TabFM (default),,0.6666667,0.78518516,0.75555557,0.93333334,0.9037037,0.9111111,0.88148147,0.9259259,0.8518519,0.91851854,0.88148147,0.9037037,0.85925925,0.8962963,0.91851854,0.9925926,0.9851852,0.8888889,0.962963,0.8740741,0.91851854,0.9777778,0.94074076,1.0,0.962963,0.91851854,0.9259259,0.93333334,0.93333334,0.9259259,0.9777778,0.9851852,0.94074076,0.97037035,0.97037035,0.94074076,1.0,0.9925926,1.0,1.0,0.9037037,1.0,1.0,1.0,0.9851852,1.0,0.9925926,0.9851852,1.0,0.9925926,0.9851852,0.94074076,0.9925926,0.9925926,1.0,1.0,1.0,0.9851852,1.0,0.97037035,0.9851852,0.9925926,0.97037035,1.0,1.0,1.0,0.9925926,0.9851852,1.0,0.9851852,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.33333334,,0.36296296,0.82222223,0.73333335,0.5185185,0.7111111,0.8962963,0.8296296,0.94814813,0.77037036,0.94074076,0.95555556,0.93333334,0.95555556,0.97037035,0.8740741,0.78518516,0.9111111,0.7259259,0.73333335,0.94814813,0.9851852,0.9925926,0.9111111,0.9777778,0.95555556,0.962963,0.9259259,0.9777778,0.7777778,0.9925926,0.9777778,0.9925926,0.93333334,0.962963,0.8518519,1.0,0.9925926,0.9851852,0.9851852,0.85925925,1.0,1.0,1.0,1.0,1.0,1.0,0.9925926,1.0,0.8888889,0.97037035,1.0,0.962963,1.0,1.0,1.0,0.9925926,1.0,1.0,1.0,1.0,1.0,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.21481481,0.63703704,,0.65185183,0.77037036,0.65925926,0.8296296,0.76296294,0.85925925,0.73333335,0.84444445,0.7259259,0.75555557,0.7259259,0.78518516,0.76296294,0.85925925,0.91851854,0.77037036,0.91851854,0.73333335,0.78518516,0.8666667,0.78518516,0.9037037,0.8518519,0.7777778,0.8,0.9111111,0.7777778,0.82222223,0.8888889,0.8740741,0.8666667,0.9111111,0.9777778,0.8740741,0.9925926,0.84444445,0.9851852,0.9777778,0.8666667,0.962963,0.9925926,0.8740741,0.962963,0.962963,0.9777778,0.9777778,0.91851854,0.93333334,0.94814813,0.93333334,0.9925926,0.9777778,0.9925926,0.9777778,0.93333334,0.95555556,0.9851852,0.94814813,0.95555556,1.0,0.95555556,1.0,1.0,1.0,1.0,0.94814813,1.0,0.95555556,1.0,1.0,1.0,0.97037035,1.0
+"AutoGluon 1.4 (best, 4h)",0.24444444,0.17777778,0.34814814,,0.5925926,0.5037037,0.5037037,0.7037037,0.6666667,0.82222223,0.6,0.8296296,0.8740741,0.8074074,0.8888889,0.9111111,0.7925926,0.64444447,0.8740741,0.6296296,0.6296296,0.8518519,0.9111111,0.9259259,0.8518519,0.837037,0.9111111,0.91851854,0.8962963,0.9259259,0.77037036,0.94074076,0.94814813,0.97037035,0.84444445,0.84444445,0.8296296,0.94814813,0.97037035,0.962963,0.9851852,0.8518519,0.9777778,0.97037035,0.97037035,0.9925926,0.94074076,0.9925926,0.9851852,0.97037035,0.8296296,0.94074076,0.9777778,0.93333334,1.0,0.9777778,0.9925926,0.9851852,0.9851852,0.9777778,1.0,0.9925926,1.0,0.9925926,1.0,1.0,1.0,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.06666667,0.26666668,0.22962964,0.4074074,,0.45925927,0.5555556,0.6888889,0.67407405,0.65185183,0.64444447,0.65185183,0.73333335,0.65185183,0.7037037,0.7185185,0.73333335,0.7777778,0.7925926,0.7185185,0.7407407,0.77037036,0.8296296,0.82222223,0.82222223,0.88148147,0.8,0.8074074,0.837037,0.78518516,0.8296296,0.8962963,0.94074076,0.8666667,0.8518519,0.9111111,0.8296296,0.94814813,0.91851854,0.9851852,0.97037035,0.8518519,1.0,0.962963,0.9777778,0.97037035,0.93333334,0.9851852,0.9925926,0.9777778,0.8888889,0.97037035,0.9111111,0.94074076,0.9925926,0.962963,1.0,0.9925926,0.962963,0.9925926,0.94814813,0.94814813,1.0,0.962963,1.0,1.0,1.0,1.0,0.9777778,1.0,0.9777778,1.0,1.0,1.0,0.9777778,1.0
+TabICLv2 (default),0.096296296,0.4814815,0.34074074,0.4962963,0.5407407,,0.73333335,0.64444447,0.7407407,0.67407405,0.75555557,0.6666667,0.7111111,0.67407405,0.6888889,0.6814815,0.7407407,0.82222223,0.6814815,0.78518516,0.77037036,0.6814815,0.7777778,0.6962963,0.8148148,0.8,0.6814815,0.73333335,0.77037036,0.6962963,0.82222223,0.8074074,0.84444445,0.74814814,0.8888889,0.94074076,0.82222223,0.962963,0.77037036,0.9037037,0.88148147,0.8740741,0.94074076,0.94814813,0.8666667,0.91851854,0.94814813,0.9925926,0.9259259,0.8888889,0.93333334,0.94814813,0.8962963,0.93333334,0.9925926,0.94814813,0.9851852,0.93333334,0.95555556,0.93333334,0.93333334,0.94814813,1.0,0.97037035,0.9851852,1.0,0.9851852,1.0,0.94814813,1.0,0.962963,0.93333334,1.0,1.0,0.9777778,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.08888889,0.2888889,0.17037037,0.4962963,0.44444445,0.26666668,,0.6,0.5777778,0.58518517,0.82222223,0.6074074,0.6296296,0.6,0.6148148,0.65925926,0.6148148,0.8074074,0.6888889,0.6296296,0.65925926,0.6074074,0.6962963,0.6814815,0.6962963,0.7925926,0.6666667,0.76296294,0.74814814,0.7407407,0.7925926,0.8074074,0.8296296,0.8074074,0.8,0.9037037,0.8,0.8296296,0.82222223,0.88148147,0.88148147,0.7777778,0.88148147,0.94814813,0.88148147,0.94074076,0.8666667,0.9777778,0.97037035,0.9111111,0.94074076,0.9037037,0.82222223,0.9851852,0.9777778,0.9037037,0.97037035,0.9777778,0.91851854,0.9851852,0.94814813,0.93333334,1.0,0.94074076,1.0,0.9925926,1.0,0.9925926,0.8962963,1.0,0.9037037,1.0,1.0,1.0,0.97037035,1.0
+RealMLP (tuned + ensembled),0.11851852,0.1037037,0.23703703,0.2962963,0.31111112,0.35555556,0.4,,0.53333336,0.5185185,0.45925927,0.54814816,0.6074074,0.5185185,0.63703704,0.63703704,0.5925926,0.5925926,0.77037036,0.5925926,0.5777778,0.5185185,0.7037037,0.6888889,0.64444447,0.7185185,0.6148148,0.85925925,0.74814814,0.64444447,0.7185185,0.837037,0.78518516,0.91851854,0.7185185,0.75555557,0.74814814,0.93333334,0.82222223,0.9037037,0.8888889,0.8888889,1.0,0.95555556,0.88148147,0.91851854,0.9037037,0.93333334,0.9111111,0.8962963,0.7925926,0.9259259,0.9851852,0.8666667,0.91851854,0.8740741,0.9925926,0.9851852,0.9851852,0.9925926,1.0,0.9777778,0.95555556,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.074074075,0.17037037,0.14074074,0.33333334,0.32592592,0.25925925,0.42222223,0.46666667,,0.52592593,0.4962963,0.5185185,0.5777778,0.56296295,0.6148148,0.6296296,0.6,0.63703704,0.67407405,0.54814816,0.63703704,0.62222224,0.6296296,0.64444447,0.6962963,0.8296296,0.6962963,0.75555557,0.75555557,0.74814814,0.76296294,0.82222223,0.76296294,0.84444445,0.75555557,0.78518516,0.8148148,0.88148147,0.82222223,0.97037035,0.8962963,0.88148147,0.9259259,0.93333334,0.837037,0.85925925,0.88148147,0.93333334,0.8518519,0.88148147,0.88148147,0.9037037,0.9111111,0.9037037,0.9259259,0.8962963,0.9259259,0.9925926,0.95555556,0.9925926,1.0,0.9777778,1.0,0.9925926,0.94814813,0.97037035,0.94814813,0.9851852,0.93333334,0.95555556,0.94074076,1.0,1.0,1.0,0.9851852,1.0
+CatBoost (tuned + ensembled),0.14814815,0.05185185,0.26666668,0.17777778,0.34814814,0.32592592,0.4148148,0.4814815,0.47407407,,0.42962962,0.6148148,0.5185185,0.6666667,0.6,0.6888889,0.4888889,0.5555556,0.6148148,0.46666667,0.53333336,0.6962963,0.64444447,0.76296294,0.5555556,0.6703704,0.8074074,0.7185185,0.74814814,0.85925925,0.63703704,0.7925926,0.8740741,0.7925926,0.7925926,0.74814814,0.6888889,0.82222223,0.8962963,0.82222223,0.9037037,0.6814815,0.8888889,0.962963,0.88148147,0.9259259,0.8962963,0.97037035,0.91851854,0.962963,0.7407407,0.91851854,0.9111111,0.73333335,0.97037035,0.94814813,0.962963,0.9777778,0.9777778,0.9925926,0.9259259,0.9851852,0.9925926,0.9925926,1.0,0.9925926,1.0,0.9925926,0.9925926,0.9925926,1.0,1.0,1.0,0.9925926,1.0,1.0
+RealTabPFN-2.5 (tuned),0.08148148,0.22962964,0.15555556,0.4,0.35555556,0.24444444,0.17777778,0.5407407,0.5037037,0.5703704,,0.5777778,0.6,0.5777778,0.58518517,0.62222224,0.6074074,0.7111111,0.6296296,0.5777778,0.63703704,0.5703704,0.7037037,0.63703704,0.6888889,0.7407407,0.63703704,0.74814814,0.7185185,0.7111111,0.7407407,0.77037036,0.78518516,0.7925926,0.7777778,0.8888889,0.75555557,0.78518516,0.7777778,0.85925925,0.8740741,0.77037036,0.84444445,0.9259259,0.82222223,0.9037037,0.837037,0.94074076,0.95555556,0.8888889,0.9111111,0.8888889,0.8,0.962963,0.95555556,0.88148147,0.94074076,0.9777778,0.91851854,0.962963,0.93333334,0.9259259,0.9925926,0.9259259,0.9925926,0.9925926,1.0,0.9851852,0.88148147,0.9925926,0.9111111,1.0,1.0,1.0,0.95555556,1.0
+CatBoost (tuned),0.11851852,0.05925926,0.27407408,0.17037037,0.34814814,0.33333334,0.39259258,0.45185184,0.4814815,0.38518518,0.42222223,,0.44444445,0.58518517,0.58518517,0.65925926,0.4962963,0.52592593,0.6074074,0.45925927,0.53333336,0.7185185,0.63703704,0.7259259,0.5407407,0.63703704,0.8074074,0.7259259,0.7037037,0.8296296,0.6296296,0.78518516,0.8666667,0.78518516,0.8,0.7407407,0.6814815,0.84444445,0.8888889,0.8074074,0.8962963,0.6814815,0.8962963,0.94074076,0.88148147,0.91851854,0.9037037,0.94814813,0.91851854,0.9777778,0.7407407,0.9111111,0.88148147,0.73333335,0.962963,0.95555556,0.95555556,0.9777778,0.97037035,0.9925926,0.93333334,0.962963,0.9925926,0.9925926,0.9925926,0.9851852,0.9925926,0.9851852,0.9925926,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.096296296,0.044444446,0.24444444,0.12592593,0.26666668,0.2888889,0.37037036,0.39259258,0.42222223,0.4814815,0.4,0.5555556,,0.5777778,0.5407407,0.6888889,0.47407407,0.4962963,0.52592593,0.5185185,0.4962963,0.64444447,0.82222223,0.7777778,0.5925926,0.6148148,0.7407407,0.64444447,0.7259259,0.74814814,0.6296296,0.75555557,0.88148147,0.7185185,0.7777778,0.7259259,0.74814814,0.8518519,0.84444445,0.84444445,0.8740741,0.65925926,0.88148147,0.9851852,0.9111111,0.93333334,0.9111111,0.9851852,0.94074076,0.9777778,0.74814814,0.9037037,0.84444445,0.74814814,0.9851852,0.9851852,0.95555556,0.9851852,0.94074076,0.9851852,0.9037037,0.9111111,1.0,0.9851852,0.9925926,0.9925926,0.9925926,0.9925926,0.9851852,1.0,0.9925926,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.14074074,0.06666667,0.27407408,0.19259259,0.34814814,0.32592592,0.4,0.4814815,0.43703705,0.33333334,0.42222223,0.4148148,0.42222223,,0.56296295,0.5777778,0.4888889,0.5407407,0.58518517,0.45925927,0.54814816,0.6074074,0.6074074,0.67407405,0.5703704,0.65185183,0.73333335,0.7259259,0.73333335,0.8,0.63703704,0.7925926,0.7925926,0.8,0.8074074,0.77037036,0.6962963,0.77037036,0.8740741,0.84444445,0.9037037,0.7037037,0.8074074,0.8740741,0.8740741,0.8518519,0.9037037,0.88148147,0.8518519,0.9259259,0.7407407,0.84444445,0.8888889,0.73333335,0.8962963,0.962963,0.91851854,0.95555556,0.9851852,0.97037035,0.9259259,0.97037035,0.9925926,0.9925926,1.0,0.9925926,1.0,0.9777778,0.9925926,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.1037037,0.044444446,0.21481481,0.11111111,0.2962963,0.31111112,0.38518518,0.36296296,0.38518518,0.4,0.4148148,0.4148148,0.45925927,0.43703705,,0.5407407,0.4888889,0.5555556,0.4962963,0.43703705,0.51111114,0.5037037,0.5777778,0.65925926,0.6148148,0.6,0.5777778,0.6296296,0.75555557,0.62222224,0.6296296,0.88148147,0.74814814,0.73333335,0.8148148,0.7037037,0.65925926,0.7925926,0.8,0.8074074,0.95555556,0.6888889,0.94074076,0.9111111,0.8074074,0.9111111,0.9037037,0.93333334,0.91851854,0.8666667,0.67407405,0.9111111,0.97037035,0.7111111,0.95555556,0.9851852,0.9925926,0.8666667,0.97037035,0.93333334,0.9777778,0.9925926,0.9851852,0.9851852,0.9925926,0.9851852,0.9925926,0.962963,0.9851852,0.9925926,0.9925926,0.9925926,0.9925926,0.9925926,0.9925926,0.9925926
+XGBoost (tuned + ensembled),0.08148148,0.02962963,0.23703703,0.08888889,0.28148147,0.31851852,0.34074074,0.36296296,0.37037036,0.31111112,0.37777779,0.34074074,0.31111112,0.42222223,0.45925927,,0.4888889,0.5037037,0.4814815,0.4814815,0.52592593,0.53333336,0.43703705,0.75555557,0.5407407,0.56296295,0.5777778,0.64444447,0.6666667,0.7259259,0.63703704,0.8,0.88148147,0.7407407,0.73333335,0.65185183,0.6814815,0.77037036,0.8666667,0.8074074,0.8296296,0.65185183,0.8666667,0.97037035,0.8962963,0.93333334,0.91851854,0.94814813,0.93333334,0.9851852,0.7037037,0.9037037,0.8666667,0.73333335,0.962963,0.9777778,0.97037035,0.9037037,0.91851854,0.962963,0.9111111,0.93333334,1.0,0.9925926,0.9925926,1.0,0.9925926,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0074074073,0.12592593,0.14074074,0.2074074,0.26666668,0.25925925,0.38518518,0.4074074,0.4,0.51111114,0.39259258,0.5037037,0.52592593,0.51111114,0.51111114,0.51111114,,0.39259258,0.51111114,0.4814815,0.46666667,0.5185185,0.65925926,0.5925926,0.8074074,0.6148148,0.5703704,0.56296295,0.6888889,0.6148148,0.6148148,0.6666667,0.7111111,0.58518517,0.7185185,0.7259259,0.6814815,0.9777778,0.7111111,0.77037036,0.78518516,0.7111111,0.9851852,0.93333334,0.8296296,0.82222223,0.94074076,0.8888889,0.82222223,0.82222223,0.6814815,0.8666667,0.88148147,0.7407407,0.9111111,0.8888889,0.9851852,0.91851854,0.94074076,0.93333334,0.8888889,0.93333334,0.9925926,0.962963,1.0,1.0,1.0,1.0,0.95555556,1.0,0.962963,0.93333334,1.0,1.0,0.9851852,1.0
+RealTabPFN-2.5 (default),0.014814815,0.21481481,0.08148148,0.35555556,0.22222222,0.17777778,0.19259259,0.4074074,0.36296296,0.44444445,0.2888889,0.47407407,0.5037037,0.45925927,0.44444445,0.4962963,0.6074074,,0.51111114,0.37777779,0.51111114,0.51111114,0.5925926,0.5555556,0.65925926,0.6148148,0.54814816,0.6814815,0.7111111,0.64444447,0.63703704,0.7407407,0.7259259,0.7185185,0.7407407,0.9111111,0.6962963,0.7259259,0.62222224,0.8,0.8666667,0.7259259,0.8074074,0.91851854,0.6888889,0.8296296,0.85925925,0.94814813,0.8666667,0.74814814,0.82222223,0.93333334,0.7777778,0.8518519,0.962963,0.8888889,0.97037035,0.91851854,0.8666667,0.9259259,0.93333334,0.8962963,0.9851852,0.93333334,1.0,1.0,1.0,0.9925926,0.8962963,1.0,0.8962963,0.94074076,1.0,1.0,0.94814813,1.0
+RealMLP (tuned),0.11111111,0.08888889,0.22962964,0.12592593,0.2074074,0.31851852,0.31111112,0.22962964,0.32592592,0.38518518,0.37037036,0.39259258,0.47407407,0.4148148,0.5037037,0.5185185,0.4888889,0.4888889,,0.45925927,0.53333336,0.39259258,0.6,0.6,0.52592593,0.62222224,0.45925927,0.6962963,0.65185183,0.56296295,0.6666667,0.7777778,0.7259259,0.8,0.65185183,0.6074074,0.6888889,0.8148148,0.74814814,0.8296296,0.8740741,0.837037,0.91851854,0.84444445,0.8,0.8888889,0.8518519,0.8148148,0.88148147,0.8148148,0.7777778,0.85925925,0.91851854,0.74814814,0.85925925,0.82222223,0.95555556,0.94814813,0.9777778,0.97037035,0.94814813,0.97037035,0.94074076,0.9777778,0.9851852,0.9851852,0.9851852,0.962963,0.9851852,0.9851852,0.9925926,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.037037037,0.27407408,0.08148148,0.37037036,0.28148147,0.21481481,0.37037036,0.4074074,0.45185184,0.53333336,0.42222223,0.5407407,0.4814815,0.5407407,0.56296295,0.5185185,0.5185185,0.62222224,0.5407407,,0.46666667,0.52592593,0.52592593,0.54814816,0.6,0.5703704,0.58518517,0.65925926,0.6666667,0.5777778,0.65925926,0.76296294,0.6666667,0.7407407,0.64444447,0.8074074,0.7925926,0.7111111,0.6074074,0.7777778,0.82222223,0.7777778,0.8,0.85925925,0.65185183,0.78518516,0.7407407,0.84444445,0.837037,0.75555557,0.8296296,0.78518516,0.76296294,0.93333334,0.88148147,0.7925926,0.8666667,0.9111111,0.8296296,0.962963,0.94074076,0.91851854,0.97037035,0.9037037,0.9259259,0.93333334,0.9259259,0.94814813,0.82222223,0.9259259,0.837037,1.0,1.0,1.0,0.91851854,1.0
+TabDPT (tuned + ensembled),0.12592593,0.26666668,0.26666668,0.37037036,0.25925925,0.22962964,0.34074074,0.42222223,0.36296296,0.46666667,0.36296296,0.46666667,0.5037037,0.45185184,0.4888889,0.47407407,0.53333336,0.4888889,0.46666667,0.53333336,,0.46666667,0.52592593,0.5185185,0.6148148,0.5555556,0.5037037,0.7037037,0.6296296,0.6,0.9111111,0.7185185,0.63703704,0.73333335,0.5703704,0.7111111,0.95555556,0.7259259,0.6296296,0.837037,0.8148148,0.7259259,0.75555557,0.8518519,0.6814815,0.7777778,0.74814814,0.78518516,0.77037036,0.7037037,0.88148147,0.8296296,0.7925926,0.94814813,0.85925925,0.75555557,0.8518519,0.9851852,0.8666667,0.9851852,0.94814813,0.8888889,0.9925926,0.8962963,0.9037037,0.9925926,0.9037037,0.9925926,0.837037,0.9777778,0.837037,1.0,1.0,1.0,0.9259259,1.0
+ChimeraBoost (tuned + ensembled),0.08148148,0.05185185,0.21481481,0.14814815,0.22962964,0.31851852,0.39259258,0.4814815,0.37777779,0.3037037,0.42962962,0.28148147,0.35555556,0.39259258,0.4962963,0.46666667,0.4814815,0.4888889,0.6074074,0.47407407,0.53333336,,0.4,0.54814816,0.54814816,0.6074074,0.7777778,0.7037037,0.65925926,0.9037037,0.62222224,0.67407405,0.7111111,0.7407407,0.6888889,0.64444447,0.62222224,0.7407407,0.8148148,0.7407407,0.7777778,0.6296296,0.8148148,0.84444445,0.837037,0.85925925,0.8518519,0.837037,0.85925925,0.8888889,0.73333335,0.8518519,0.837037,0.73333335,0.84444445,0.9037037,0.82222223,0.84444445,0.85925925,0.85925925,0.85925925,0.8518519,0.93333334,0.962963,0.93333334,0.93333334,0.93333334,0.9259259,1.0,0.93333334,1.0,0.88148147,0.94814813,1.0,1.0,1.0
+LightGBM (tuned),0.022222223,0.014814815,0.13333334,0.08888889,0.17037037,0.22222222,0.3037037,0.2962963,0.37037036,0.35555556,0.2962963,0.36296296,0.17777778,0.39259258,0.42222223,0.56296295,0.34074074,0.4074074,0.4,0.47407407,0.47407407,0.6,,0.65925926,0.4814815,0.52592593,0.65185183,0.6074074,0.6296296,0.6888889,0.58518517,0.65185183,0.8296296,0.62222224,0.6814815,0.62222224,0.64444447,0.8074074,0.78518516,0.78518516,0.73333335,0.6296296,0.84444445,0.9777778,0.82222223,0.85925925,0.85925925,0.8888889,0.8962963,0.9037037,0.7407407,0.8148148,0.837037,0.73333335,0.94814813,0.95555556,0.93333334,0.95555556,0.85925925,0.9777778,0.8666667,0.8740741,0.9925926,0.94074076,1.0,0.9925926,1.0,0.9925926,0.9111111,0.9925926,0.94074076,1.0,1.0,1.0,0.9777778,1.0
+XGBoost (tuned),0.05925926,0.0074074073,0.21481481,0.074074075,0.17777778,0.3037037,0.31851852,0.31111112,0.35555556,0.23703703,0.36296296,0.27407408,0.22222222,0.32592592,0.34074074,0.24444444,0.4074074,0.44444445,0.4,0.45185184,0.4814815,0.45185184,0.34074074,,0.4962963,0.5555556,0.5037037,0.5925926,0.5925926,0.67407405,0.6148148,0.7111111,0.7777778,0.67407405,0.6962963,0.6296296,0.65925926,0.73333335,0.8074074,0.78518516,0.8148148,0.63703704,0.84444445,0.9111111,0.8740741,0.9111111,0.9037037,0.9259259,0.9037037,0.94814813,0.7037037,0.8962963,0.837037,0.73333335,0.95555556,0.95555556,0.97037035,0.85925925,0.9111111,0.95555556,0.8666667,0.9037037,1.0,0.9925926,0.9925926,1.0,0.9925926,0.9925926,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.08888889,0.096296296,0.14814815,0.17777778,0.18518518,0.3037037,0.35555556,0.3037037,0.44444445,0.31111112,0.45925927,0.4074074,0.42962962,0.38518518,0.45925927,0.19259259,0.34074074,0.47407407,0.4,0.38518518,0.45185184,0.5185185,0.5037037,,0.5185185,0.51111114,0.4888889,0.65185183,0.52592593,0.5407407,0.5777778,0.6888889,0.5777778,0.65925926,0.64444447,0.58518517,0.8888889,0.63703704,0.6888889,0.73333335,0.7111111,0.93333334,0.8296296,0.7777778,0.78518516,0.8666667,0.8296296,0.7925926,0.7925926,0.65925926,0.8074074,0.8666667,0.6888889,0.8888889,0.837037,0.9851852,0.91851854,0.91851854,0.9259259,0.88148147,0.94074076,0.9777778,0.9259259,1.0,1.0,1.0,0.9851852,0.9259259,1.0,0.93333334,0.93333334,1.0,1.0,0.95555556,1.0
+ModernNCA (tuned),0.037037037,0.022222223,0.14814815,0.16296296,0.11851852,0.2,0.2074074,0.28148147,0.17037037,0.32962963,0.25925925,0.36296296,0.38518518,0.34814814,0.4,0.43703705,0.38518518,0.38518518,0.37777779,0.42962962,0.44444445,0.39259258,0.47407407,0.44444445,0.4814815,,0.44444445,0.5703704,0.64444447,0.52592593,0.6,0.7185185,0.58518517,0.63703704,0.6962963,0.7037037,0.7407407,0.7111111,0.62222224,0.8888889,0.8148148,0.73333335,0.78518516,0.78518516,0.7037037,0.78518516,0.78518516,0.8296296,0.8,0.75555557,0.75555557,0.82222223,0.7777778,0.85925925,0.837037,0.84444445,0.9037037,0.9037037,0.94814813,0.94814813,0.91851854,0.94074076,0.9777778,0.97037035,0.97037035,0.9851852,0.97037035,0.94074076,0.9111111,0.9777778,0.93333334,1.0,1.0,1.0,0.9925926,1.0
+ChimeraBoost (tuned),0.08148148,0.044444446,0.22222222,0.08888889,0.2,0.31851852,0.33333334,0.38518518,0.3037037,0.19259259,0.36296296,0.19259259,0.25925925,0.26666668,0.42222223,0.42222223,0.42962962,0.45185184,0.5407407,0.4148148,0.4962963,0.22222222,0.34814814,0.4962963,0.4888889,0.5555556,,0.6666667,0.6,0.75555557,0.6,0.6296296,0.65925926,0.7185185,0.65185183,0.62222224,0.58518517,0.7037037,0.73333335,0.7037037,0.74814814,0.62222224,0.8074074,0.78518516,0.78518516,0.84444445,0.837037,0.8074074,0.84444445,0.837037,0.7111111,0.82222223,0.8074074,0.7259259,0.82222223,0.8666667,0.8148148,0.8296296,0.837037,0.84444445,0.8518519,0.84444445,0.9259259,0.94814813,0.9259259,0.9259259,0.9259259,0.91851854,1.0,0.93333334,1.0,0.8888889,0.94814813,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.074074075,0.037037037,0.2,0.08148148,0.19259259,0.26666668,0.23703703,0.14074074,0.24444444,0.28148147,0.25185186,0.27407408,0.35555556,0.27407408,0.37037036,0.35555556,0.43703705,0.31851852,0.3037037,0.34074074,0.2962963,0.2962963,0.39259258,0.4074074,0.51111114,0.42962962,0.33333334,,0.5037037,0.39259258,0.35555556,0.6,0.51111114,0.78518516,0.56296295,0.51111114,0.53333336,0.6962963,0.6296296,0.63703704,0.7185185,0.7037037,0.74814814,0.63703704,0.6888889,0.76296294,0.75555557,0.8074074,0.77037036,0.7185185,0.67407405,0.7259259,0.82222223,0.7185185,0.8074074,0.6962963,0.837037,0.837037,0.8666667,0.9037037,1.0,0.9111111,0.91851854,0.9777778,0.9037037,0.9777778,0.9037037,0.9851852,0.9851852,0.9851852,1.0,0.93333334,0.93333334,0.9777778,1.0,1.0
+iLTM (tuned),0.06666667,0.074074075,0.08888889,0.1037037,0.16296296,0.22962964,0.25185186,0.25185186,0.24444444,0.25185186,0.28148147,0.2962963,0.27407408,0.26666668,0.24444444,0.33333334,0.31111112,0.2888889,0.34814814,0.33333334,0.37037036,0.34074074,0.37037036,0.4074074,0.34814814,0.35555556,0.4,0.4962963,,0.4814815,0.52592593,0.56296295,0.4962963,0.5407407,0.64444447,0.54814816,0.5777778,0.6148148,0.6074074,0.6296296,0.6814815,0.7111111,0.7259259,0.78518516,0.65185183,0.82222223,0.77037036,0.75555557,0.837037,0.6666667,0.65185183,0.8296296,0.82222223,0.64444447,0.84444445,0.837037,0.9259259,0.77037036,0.9037037,0.82222223,0.94074076,0.93333334,0.9777778,0.9925926,0.962963,0.9851852,0.962963,0.94814813,0.93333334,0.9925926,0.95555556,0.94074076,0.9925926,1.0,1.0,1.0
+ChimeraBoost (default),0.06666667,0.022222223,0.22222222,0.074074075,0.21481481,0.3037037,0.25925925,0.35555556,0.25185186,0.14074074,0.2888889,0.17037037,0.25185186,0.2,0.37777779,0.27407408,0.38518518,0.35555556,0.43703705,0.42222223,0.4,0.096296296,0.31111112,0.32592592,0.47407407,0.47407407,0.24444444,0.6074074,0.5185185,,0.5925926,0.6148148,0.56296295,0.7037037,0.5703704,0.5925926,0.6148148,0.6074074,0.6666667,0.6666667,0.6962963,0.62222224,0.6962963,0.6666667,0.74814814,0.74814814,0.77037036,0.65925926,0.7407407,0.7777778,0.6814815,0.7777778,0.8074074,0.73333335,0.7185185,0.82222223,0.78518516,0.7925926,0.84444445,0.8296296,0.8518519,0.84444445,0.93333334,0.93333334,0.9259259,0.93333334,0.93333334,0.91851854,1.0,0.93333334,1.0,0.8666667,0.93333334,0.9777778,1.0,1.0
+TabDPT (tuned),0.074074075,0.22222222,0.17777778,0.22962964,0.17037037,0.17777778,0.2074074,0.28148147,0.23703703,0.36296296,0.25925925,0.37037036,0.37037036,0.36296296,0.37037036,0.36296296,0.38518518,0.36296296,0.33333334,0.34074074,0.08888889,0.37777779,0.4148148,0.38518518,0.45925927,0.4,0.4,0.64444447,0.47407407,0.4074074,,0.52592593,0.4888889,0.62222224,0.42962962,0.54814816,0.7740741,0.56296295,0.4814815,0.6148148,0.67407405,0.6814815,0.6074074,0.7037037,0.6,0.65925926,0.5925926,0.7111111,0.7259259,0.6074074,0.7925926,0.65925926,0.65185183,0.88148147,0.76296294,0.6296296,0.78518516,0.962963,0.78518516,0.97037035,0.93333334,0.8518519,0.93333334,0.8666667,0.7777778,0.94074076,0.78518516,0.95555556,0.7925926,0.93333334,0.8296296,1.0,0.94074076,0.9851852,0.8666667,1.0
+TorchMLP (tuned + ensembled),0.022222223,0.0074074073,0.11111111,0.05925926,0.1037037,0.19259259,0.19259259,0.16296296,0.17777778,0.2074074,0.22962964,0.21481481,0.24444444,0.2074074,0.11851852,0.2,0.33333334,0.25925925,0.22222222,0.23703703,0.28148147,0.32592592,0.34814814,0.2888889,0.42222223,0.28148147,0.37037036,0.4,0.43703705,0.38518518,0.47407407,,0.47407407,0.45185184,0.52592593,0.5185185,0.51111114,0.5703704,0.54814816,0.52592593,0.9111111,0.6,0.6296296,0.63703704,0.65925926,0.7259259,0.6814815,0.75555557,0.78518516,0.6888889,0.6666667,0.8148148,0.7407407,0.65925926,0.8,0.8148148,0.9925926,0.73333335,0.88148147,0.75555557,0.82222223,0.8888889,0.94074076,0.94074076,0.94814813,0.94814813,0.94814813,0.9037037,0.94074076,0.94814813,0.94074076,0.9259259,1.0,1.0,0.97037035,1.0
+XGBoost (default),0.014814815,0.022222223,0.12592593,0.05185185,0.05925926,0.15555556,0.17037037,0.21481481,0.23703703,0.12592593,0.21481481,0.13333334,0.11851852,0.2074074,0.25185186,0.11851852,0.2888889,0.27407408,0.27407408,0.33333334,0.36296296,0.2888889,0.17037037,0.22222222,0.31111112,0.4148148,0.34074074,0.4888889,0.5037037,0.43703705,0.51111114,0.52592593,,0.54814816,0.5925926,0.52592593,0.5185185,0.47407407,0.58518517,0.6148148,0.6962963,0.58518517,0.56296295,0.51111114,0.65925926,0.76296294,0.7259259,0.7259259,0.78518516,0.76296294,0.65185183,0.76296294,0.7925926,0.6814815,0.75555557,0.8666667,0.88148147,0.74814814,0.8074074,0.88148147,0.7777778,0.84444445,0.94814813,0.9259259,0.962963,0.9925926,0.97037035,0.8740741,0.91851854,1.0,0.9259259,1.0,1.0,1.0,0.93333334,1.0
+xRFM (tuned),0.05925926,0.0074074073,0.13333334,0.02962963,0.13333334,0.25185186,0.19259259,0.08148148,0.15555556,0.2074074,0.2074074,0.21481481,0.28148147,0.2,0.26666668,0.25925925,0.4148148,0.28148147,0.2,0.25925925,0.26666668,0.25925925,0.37777779,0.32592592,0.42222223,0.36296296,0.28148147,0.21481481,0.45925927,0.2962963,0.37777779,0.54814816,0.45185184,,0.53333336,0.5185185,0.42962962,0.64444447,0.5407407,0.5777778,0.6888889,0.6148148,0.7185185,0.5925926,0.63703704,0.73333335,0.75555557,0.7037037,0.77037036,0.6148148,0.5777778,0.8,0.7777778,0.58518517,0.78518516,0.65185183,0.82222223,0.8148148,0.88148147,0.84444445,0.962963,0.9111111,0.8666667,0.9851852,0.8888889,0.9111111,0.8962963,0.94814813,0.97037035,0.97037035,0.9925926,0.9259259,0.9925926,1.0,1.0,1.0
+TabICL (default),0.02962963,0.06666667,0.08888889,0.15555556,0.14814815,0.11111111,0.2,0.28148147,0.24444444,0.2074074,0.22222222,0.2,0.22222222,0.19259259,0.18518518,0.26666668,0.28148147,0.25925925,0.34814814,0.35555556,0.42962962,0.31111112,0.31851852,0.3037037,0.34074074,0.3037037,0.34814814,0.43703705,0.35555556,0.42962962,0.5703704,0.47407407,0.4074074,0.46666667,,0.5407407,0.6148148,0.42222223,0.6148148,0.5925926,0.54814816,0.52592593,0.53333336,0.5777778,0.64444447,0.4814815,0.6259259,0.6,0.5777778,0.67407405,0.67407405,0.7111111,0.74814814,0.6666667,0.62222224,0.7407407,0.7037037,0.6666667,0.78518516,0.7037037,0.63703704,0.8518519,0.8333333,0.88148147,0.94814813,0.8333333,0.95555556,0.8518519,0.837037,0.8185185,0.837037,0.9037037,1.0,1.0,0.95555556,1.0
+TabSwift (default),0.02962963,0.037037037,0.022222223,0.15555556,0.08888889,0.05925926,0.096296296,0.24444444,0.21481481,0.25185186,0.11111111,0.25925925,0.27407408,0.22962964,0.2962963,0.34814814,0.27407408,0.08888889,0.39259258,0.19259259,0.2888889,0.35555556,0.37777779,0.37037036,0.35555556,0.2962963,0.37777779,0.4888889,0.45185184,0.4074074,0.45185184,0.4814815,0.47407407,0.4814815,0.45925927,,0.62222224,0.4888889,0.54814816,0.56296295,0.5777778,0.5925926,0.58518517,0.5925926,0.58518517,0.5555556,0.6074074,0.62222224,0.5777778,0.6,0.6962963,0.6666667,0.74814814,0.6814815,0.6666667,0.6962963,0.7185185,0.7925926,0.78518516,0.8296296,0.8,0.78518516,0.837037,0.8518519,0.8666667,0.8962963,0.8666667,0.88148147,0.8148148,0.8962963,0.837037,0.93333334,0.94814813,0.9777778,0.8740741,1.0
+TabDPT (default),0.05925926,0.14814815,0.12592593,0.17037037,0.17037037,0.17777778,0.2,0.25185186,0.18518518,0.31111112,0.24444444,0.31851852,0.25185186,0.3037037,0.34074074,0.31851852,0.31851852,0.3037037,0.31111112,0.2074074,0.044444446,0.37777779,0.35555556,0.34074074,0.4148148,0.25925925,0.4148148,0.46666667,0.42222223,0.38518518,0.22592592,0.4888889,0.4814815,0.5703704,0.38518518,0.37777779,,0.5555556,0.4814815,0.5407407,0.58518517,0.5703704,0.5555556,0.6,0.53333336,0.6296296,0.6148148,0.62222224,0.65925926,0.5407407,0.6074074,0.5555556,0.58518517,0.7037037,0.64444447,0.6148148,0.74814814,0.77037036,0.67407405,0.8666667,0.8148148,0.75555557,0.78518516,0.7407407,0.7777778,0.8148148,0.7777778,0.7777778,0.8,0.9037037,0.8074074,0.9851852,0.8074074,0.82222223,0.8296296,1.0
+TabM (default),0.0,0.0,0.0074074073,0.05185185,0.05185185,0.037037037,0.17037037,0.06666667,0.11851852,0.17777778,0.21481481,0.15555556,0.14814815,0.22962964,0.2074074,0.22962964,0.022222223,0.27407408,0.18518518,0.2888889,0.27407408,0.25925925,0.19259259,0.26666668,0.11111111,0.2888889,0.2962963,0.3037037,0.38518518,0.39259258,0.43703705,0.42962962,0.52592593,0.35555556,0.5777778,0.51111114,0.44444445,,0.53333336,0.56296295,0.5407407,0.5703704,0.6814815,0.6148148,0.62222224,0.6888889,0.7185185,0.63703704,0.6962963,0.65185183,0.5555556,0.6814815,0.77037036,0.54814816,0.65925926,0.67407405,0.7259259,0.85925925,0.8518519,0.88148147,0.7407407,0.8740741,0.8666667,0.8962963,0.93333334,0.97037035,0.93333334,0.91851854,0.8666667,0.95555556,0.85925925,0.93333334,1.0,1.0,0.93333334,1.0
+EBM (tuned + ensembled),0.0074074073,0.0074074073,0.15555556,0.02962963,0.08148148,0.22962964,0.17777778,0.17777778,0.17777778,0.1037037,0.22222222,0.11111111,0.15555556,0.12592593,0.2,0.13333334,0.2888889,0.37777779,0.25185186,0.39259258,0.37037036,0.18518518,0.21481481,0.19259259,0.36296296,0.37777779,0.26666668,0.37037036,0.39259258,0.33333334,0.5185185,0.45185184,0.4148148,0.45925927,0.38518518,0.45185184,0.5185185,0.46666667,,0.54814816,0.64444447,0.46666667,0.54814816,0.5185185,0.8296296,0.6148148,0.5925926,0.6666667,0.64444447,0.94814813,0.6,0.6666667,0.64444447,0.65185183,0.6962963,0.64444447,0.8,0.65185183,0.7037037,0.64444447,0.67407405,0.78518516,0.7185185,0.837037,0.9777778,0.7185185,0.94074076,0.7259259,0.93333334,0.78518516,0.94814813,0.75555557,1.0,0.9259259,0.962963,0.93333334
+ModernNCA (default),0.0,0.014814815,0.014814815,0.037037037,0.014814815,0.096296296,0.11851852,0.096296296,0.02962963,0.17777778,0.14074074,0.19259259,0.15555556,0.15555556,0.19259259,0.19259259,0.22962964,0.2,0.17037037,0.22222222,0.16296296,0.25925925,0.21481481,0.21481481,0.31111112,0.11111111,0.2962963,0.36296296,0.37037036,0.33333334,0.38518518,0.47407407,0.38518518,0.42222223,0.4074074,0.43703705,0.45925927,0.43703705,0.45185184,,0.5555556,0.5555556,0.52592593,0.5925926,0.4814815,0.64444447,0.58518517,0.5777778,0.7111111,0.5037037,0.6888889,0.67407405,0.58518517,0.7407407,0.63703704,0.62222224,0.837037,0.77037036,0.6962963,0.84444445,0.8666667,0.78518516,0.95555556,0.8888889,0.8296296,0.9259259,0.837037,0.8888889,0.8,0.93333334,0.8296296,0.94814813,1.0,1.0,0.9111111,1.0
+TorchMLP (tuned),0.0,0.014814815,0.022222223,0.014814815,0.02962963,0.11851852,0.11851852,0.11111111,0.1037037,0.096296296,0.12592593,0.1037037,0.12592593,0.096296296,0.044444446,0.17037037,0.21481481,0.13333334,0.12592593,0.17777778,0.18518518,0.22222222,0.26666668,0.18518518,0.26666668,0.18518518,0.25185186,0.28148147,0.31851852,0.3037037,0.32592592,0.08888889,0.3037037,0.31111112,0.45185184,0.42222223,0.4148148,0.45925927,0.35555556,0.44444445,,0.46666667,0.43703705,0.5185185,0.47407407,0.5703704,0.5703704,0.6148148,0.6296296,0.5185185,0.62222224,0.7037037,0.63703704,0.6148148,0.6962963,0.7259259,0.9259259,0.7037037,0.74814814,0.73333335,0.77037036,0.7777778,0.8518519,0.8962963,0.8962963,0.8888889,0.8962963,0.8148148,0.91851854,0.8962963,0.91851854,0.9037037,0.9851852,1.0,0.94814813,1.0
+LimiX (default),0.096296296,0.14074074,0.13333334,0.14814815,0.14814815,0.12592593,0.22222222,0.11111111,0.11851852,0.31851852,0.22962964,0.31851852,0.34074074,0.2962963,0.31111112,0.34814814,0.2888889,0.27407408,0.16296296,0.22222222,0.27407408,0.37037036,0.37037036,0.36296296,0.2888889,0.26666668,0.37777779,0.2962963,0.2888889,0.37777779,0.31851852,0.4,0.4148148,0.38518518,0.47407407,0.4074074,0.42962962,0.42962962,0.53333336,0.44444445,0.53333336,,0.4888889,0.4074074,0.5407407,0.51111114,0.4962963,0.4814815,0.53333336,0.5407407,0.6296296,0.6,0.62222224,0.5703704,0.5037037,0.52592593,0.5925926,0.56296295,0.7111111,0.6074074,0.6666667,0.7407407,0.77037036,0.7111111,0.6814815,0.6888889,0.6888889,0.6962963,0.76296294,0.7037037,0.77037036,0.7925926,0.74814814,0.78518516,0.77037036,0.9259259
+RealMLP (default),0.0,0.0,0.037037037,0.022222223,0.0,0.05925926,0.11851852,0.0,0.074074075,0.11111111,0.15555556,0.1037037,0.11851852,0.19259259,0.05925926,0.13333334,0.014814815,0.19259259,0.08148148,0.2,0.24444444,0.18518518,0.15555556,0.15555556,0.06666667,0.21481481,0.19259259,0.25185186,0.27407408,0.3037037,0.39259258,0.37037036,0.43703705,0.28148147,0.46666667,0.4148148,0.44444445,0.31851852,0.45185184,0.47407407,0.56296295,0.51111114,,0.4814815,0.53333336,0.6148148,0.5555556,0.62222224,0.67407405,0.5555556,0.58518517,0.6296296,0.6666667,0.5703704,0.65185183,0.67407405,0.7259259,0.74814814,0.88148147,0.7925926,0.6888889,0.8962963,0.82222223,0.85925925,0.84444445,0.8296296,0.84444445,0.91851854,0.8888889,0.9111111,0.91851854,0.93333334,0.9777778,1.0,0.94074076,1.0
+LightGBM (default),0.0,0.0,0.0074074073,0.02962963,0.037037037,0.05185185,0.05185185,0.044444446,0.06666667,0.037037037,0.074074075,0.05925926,0.014814815,0.12592593,0.08888889,0.02962963,0.06666667,0.08148148,0.15555556,0.14074074,0.14814815,0.15555556,0.022222223,0.08888889,0.17037037,0.21481481,0.21481481,0.36296296,0.21481481,0.33333334,0.2962963,0.36296296,0.4888889,0.4074074,0.42222223,0.4074074,0.4,0.38518518,0.4814815,0.4074074,0.4814815,0.5925926,0.5185185,,0.5407407,0.5925926,0.6074074,0.5703704,0.6,0.56296295,0.6,0.5777778,0.65185183,0.5925926,0.63703704,0.67407405,0.76296294,0.73333335,0.73333335,0.8148148,0.7925926,0.8,0.9259259,0.84444445,0.8962963,0.962963,0.9037037,0.94074076,0.7925926,0.94814813,0.8,0.93333334,1.0,1.0,0.8740741,1.0
+EBM (tuned),0.0,0.0,0.12592593,0.02962963,0.022222223,0.13333334,0.11851852,0.11851852,0.16296296,0.11851852,0.17777778,0.11851852,0.08888889,0.12592593,0.19259259,0.1037037,0.17037037,0.31111112,0.2,0.34814814,0.31851852,0.16296296,0.17777778,0.12592593,0.22222222,0.2962963,0.21481481,0.31111112,0.34814814,0.25185186,0.4,0.34074074,0.34074074,0.36296296,0.35555556,0.4148148,0.46666667,0.37777779,0.17037037,0.5185185,0.52592593,0.45925927,0.46666667,0.45925927,,0.53333336,0.52592593,0.5925926,0.5555556,0.8074074,0.56296295,0.6148148,0.5703704,0.6296296,0.63703704,0.5555556,0.7407407,0.6296296,0.6888889,0.63703704,0.65925926,0.74814814,0.67407405,0.7925926,0.94814813,0.6962963,0.91851854,0.6888889,0.9037037,0.77037036,0.9259259,0.73333335,1.0,0.91851854,0.94814813,0.93333334
+RandomForest (tuned + ensembled),0.014814815,0.0,0.037037037,0.0074074073,0.02962963,0.08148148,0.05925926,0.08148148,0.14074074,0.074074075,0.096296296,0.08148148,0.06666667,0.14814815,0.08888889,0.06666667,0.17777778,0.17037037,0.11111111,0.21481481,0.22222222,0.14074074,0.14074074,0.08888889,0.21481481,0.21481481,0.15555556,0.23703703,0.17777778,0.25185186,0.34074074,0.27407408,0.23703703,0.26666668,0.5185185,0.44444445,0.37037036,0.31111112,0.38518518,0.35555556,0.42962962,0.4888889,0.38518518,0.4074074,0.46666667,,0.62222224,0.5555556,0.6962963,0.5185185,0.65185183,0.6666667,0.5925926,0.64444447,0.63703704,0.63703704,0.7111111,0.65185183,0.73333335,0.6814815,0.6814815,0.8,0.97037035,0.9259259,0.8666667,0.94814813,0.8740741,0.85925925,0.8666667,0.97037035,0.9037037,0.84444445,0.95555556,1.0,0.97037035,1.0
+BetaTabPFN (default),0.0,0.0,0.037037037,0.05925926,0.06666667,0.05185185,0.13333334,0.096296296,0.11851852,0.1037037,0.16296296,0.096296296,0.08888889,0.096296296,0.096296296,0.08148148,0.05925926,0.14074074,0.14814815,0.25925925,0.25185186,0.14814815,0.14074074,0.096296296,0.13333334,0.21481481,0.16296296,0.24444444,0.22962964,0.22962964,0.4074074,0.31851852,0.27407408,0.24444444,0.37407407,0.39259258,0.38518518,0.28148147,0.4074074,0.4148148,0.42962962,0.5037037,0.44444445,0.39259258,0.47407407,0.37777779,,0.47407407,0.4888889,0.5703704,0.5185185,0.62222224,0.6888889,0.51111114,0.4888889,0.5925926,0.6148148,0.5925926,0.8,0.6148148,0.5555556,0.837037,0.67777777,0.91851854,0.95555556,0.73703706,0.962963,0.75555557,0.837037,0.8111111,0.84444445,0.837037,1.0,1.0,0.962963,1.0
+ExtraTrees (tuned + ensembled),0.0074074073,0.0,0.022222223,0.0074074073,0.014814815,0.0074074073,0.022222223,0.06666667,0.06666667,0.02962963,0.05925926,0.05185185,0.014814815,0.11851852,0.06666667,0.05185185,0.11111111,0.05185185,0.18518518,0.15555556,0.21481481,0.16296296,0.11111111,0.074074075,0.17037037,0.17037037,0.19259259,0.19259259,0.24444444,0.34074074,0.2888889,0.24444444,0.27407408,0.2962963,0.4,0.37777779,0.37777779,0.36296296,0.33333334,0.42222223,0.38518518,0.5185185,0.37777779,0.42962962,0.4074074,0.44444445,0.52592593,,0.46666667,0.4148148,0.6666667,0.5925926,0.5407407,0.65925926,0.8074074,0.6074074,0.5703704,0.63703704,0.62222224,0.6888889,0.6814815,0.76296294,0.9259259,0.8296296,0.8148148,0.8962963,0.8148148,0.9851852,0.8148148,0.9111111,0.85925925,0.8666667,0.9037037,0.93333334,0.93333334,1.0
+RandomForest (tuned),0.014814815,0.0074074073,0.022222223,0.014814815,0.0074074073,0.074074075,0.02962963,0.08888889,0.14814815,0.08148148,0.044444446,0.08148148,0.05925926,0.14814815,0.08148148,0.06666667,0.17777778,0.13333334,0.11851852,0.16296296,0.22962964,0.14074074,0.1037037,0.096296296,0.2074074,0.2,0.15555556,0.22962964,0.16296296,0.25925925,0.27407408,0.21481481,0.21481481,0.22962964,0.42222223,0.42222223,0.34074074,0.3037037,0.35555556,0.2888889,0.37037036,0.46666667,0.32592592,0.4,0.44444445,0.3037037,0.51111114,0.53333336,,0.4814815,0.6333333,0.64444447,0.5555556,0.6259259,0.6074074,0.6296296,0.62222224,0.62222224,0.6888889,0.63703704,0.65185183,0.77037036,0.86296296,0.9037037,0.85925925,0.8518519,0.8666667,0.82222223,0.8074074,0.8888889,0.8518519,0.837037,0.95555556,1.0,0.94074076,1.0
+EBM (default),0.0,0.0,0.08148148,0.02962963,0.022222223,0.11111111,0.08888889,0.1037037,0.11851852,0.037037037,0.11111111,0.022222223,0.022222223,0.074074075,0.13333334,0.014814815,0.17777778,0.25185186,0.18518518,0.24444444,0.2962963,0.11111111,0.096296296,0.05185185,0.2074074,0.24444444,0.16296296,0.28148147,0.33333334,0.22222222,0.39259258,0.31111112,0.23703703,0.38518518,0.32592592,0.4,0.45925927,0.34814814,0.05185185,0.4962963,0.4814815,0.45925927,0.44444445,0.43703705,0.19259259,0.4814815,0.42962962,0.58518517,0.5185185,,0.54814816,0.5185185,0.53333336,0.6148148,0.63703704,0.56296295,0.73333335,0.6148148,0.6888889,0.63703704,0.62222224,0.7111111,0.65925926,0.78518516,0.85925925,0.6888889,0.84444445,0.6814815,0.8740741,0.76296294,0.8962963,0.7259259,1.0,0.88148147,0.94074076,0.93333334
+Nori-30M (default),0.0074074073,0.11111111,0.06666667,0.17037037,0.11111111,0.06666667,0.05925926,0.2074074,0.11851852,0.25925925,0.08888889,0.25925925,0.25185186,0.25925925,0.32592592,0.2962963,0.31851852,0.17777778,0.22222222,0.17037037,0.11851852,0.26666668,0.25925925,0.2962963,0.34074074,0.24444444,0.2888889,0.32592592,0.34814814,0.31851852,0.2074074,0.33333334,0.34814814,0.42222223,0.32592592,0.3037037,0.39259258,0.44444445,0.4,0.31111112,0.37777779,0.37037036,0.4148148,0.4,0.43703705,0.34814814,0.4814815,0.33333334,0.36666667,0.45185184,,0.42962962,0.44444445,0.58518517,0.35555556,0.42222223,0.47407407,0.63703704,0.4888889,0.63703704,0.8666667,0.5925926,0.65925926,0.5407407,0.6,0.65185183,0.6,0.7259259,0.6296296,0.7185185,0.65925926,0.85925925,0.67407405,0.77037036,0.67407405,1.0
+SAP-RPT-OSS (default),0.014814815,0.02962963,0.05185185,0.05925926,0.02962963,0.05185185,0.096296296,0.074074075,0.096296296,0.08148148,0.11111111,0.08888889,0.096296296,0.15555556,0.08888889,0.096296296,0.13333334,0.06666667,0.14074074,0.21481481,0.17037037,0.14814815,0.18518518,0.1037037,0.19259259,0.17777778,0.17777778,0.27407408,0.17037037,0.22222222,0.34074074,0.18518518,0.23703703,0.2,0.2888889,0.33333334,0.44444445,0.31851852,0.33333334,0.32592592,0.2962963,0.4,0.37037036,0.42222223,0.38518518,0.33333334,0.37777779,0.4074074,0.35555556,0.4814815,0.5703704,,0.53333336,0.6148148,0.5185185,0.4074074,0.6,0.6814815,0.6962963,0.7037037,0.62222224,0.7777778,0.6888889,0.85925925,0.7111111,0.7185185,0.7259259,0.77037036,0.8740741,0.78518516,0.8962963,0.8,0.97037035,1.0,0.9851852,1.0
+FastaiMLP (tuned + ensembled),0.05925926,0.0,0.06666667,0.022222223,0.08888889,0.1037037,0.17777778,0.014814815,0.08888889,0.08888889,0.2,0.11851852,0.15555556,0.11111111,0.02962963,0.13333334,0.11851852,0.22222222,0.08148148,0.23703703,0.2074074,0.16296296,0.16296296,0.16296296,0.13333334,0.22222222,0.19259259,0.17777778,0.17777778,0.19259259,0.34814814,0.25925925,0.2074074,0.22222222,0.25185186,0.25185186,0.4148148,0.22962964,0.35555556,0.4148148,0.36296296,0.37777779,0.33333334,0.34814814,0.42962962,0.4074074,0.31111112,0.45925927,0.44444445,0.46666667,0.5555556,0.46666667,,0.5555556,0.4814815,0.45185184,0.53333336,0.5703704,0.8296296,0.58518517,0.6666667,0.9777778,0.6,0.7185185,0.7185185,0.53333336,0.7185185,0.75555557,0.9259259,0.6666667,0.93333334,0.8074074,0.84444445,0.7925926,0.9851852,1.0
+Nori (default),0.0074074073,0.037037037,0.0074074073,0.06666667,0.05925926,0.06666667,0.014814815,0.13333334,0.096296296,0.26666668,0.037037037,0.26666668,0.25185186,0.26666668,0.2888889,0.26666668,0.25925925,0.14814815,0.25185186,0.06666667,0.05185185,0.26666668,0.26666668,0.26666668,0.31111112,0.14074074,0.27407408,0.28148147,0.35555556,0.26666668,0.11851852,0.34074074,0.31851852,0.4148148,0.33333334,0.31851852,0.2962963,0.45185184,0.34814814,0.25925925,0.38518518,0.42962962,0.42962962,0.4074074,0.37037036,0.35555556,0.4888889,0.34074074,0.37407407,0.38518518,0.4148148,0.38518518,0.44444445,,0.36296296,0.42962962,0.47407407,0.5777778,0.4888889,0.6148148,0.8740741,0.5925926,0.6666667,0.5407407,0.6,0.65925926,0.6,0.73333335,0.6296296,0.7259259,0.65925926,0.8518519,0.67407405,0.77037036,0.67407405,1.0
+ExtraTrees (tuned),0.0074074073,0.0,0.022222223,0.0,0.0074074073,0.0074074073,0.022222223,0.08148148,0.074074075,0.02962963,0.044444446,0.037037037,0.014814815,0.1037037,0.044444446,0.037037037,0.08888889,0.037037037,0.14074074,0.11851852,0.14074074,0.15555556,0.05185185,0.044444446,0.11111111,0.16296296,0.17777778,0.19259259,0.15555556,0.28148147,0.23703703,0.2,0.24444444,0.21481481,0.37777779,0.33333334,0.35555556,0.34074074,0.3037037,0.36296296,0.3037037,0.4962963,0.34814814,0.36296296,0.36296296,0.36296296,0.51111114,0.19259259,0.39259258,0.36296296,0.64444447,0.4814815,0.5185185,0.63703704,,0.5037037,0.51111114,0.6074074,0.5925926,0.6666667,0.65925926,0.6962963,0.9037037,0.7925926,0.8074074,0.8888889,0.8,0.962963,0.8074074,0.9111111,0.8296296,0.85925925,0.8962963,0.94074076,0.9259259,1.0
+iLTM (default),0.0,0.0,0.0074074073,0.022222223,0.037037037,0.05185185,0.096296296,0.12592593,0.1037037,0.05185185,0.11851852,0.044444446,0.014814815,0.037037037,0.014814815,0.022222223,0.11111111,0.11111111,0.17777778,0.2074074,0.24444444,0.096296296,0.044444446,0.044444446,0.16296296,0.15555556,0.13333334,0.3037037,0.16296296,0.17777778,0.37037036,0.18518518,0.13333334,0.34814814,0.25925925,0.3037037,0.38518518,0.32592592,0.35555556,0.37777779,0.27407408,0.47407407,0.32592592,0.32592592,0.44444445,0.36296296,0.4074074,0.39259258,0.37037036,0.43703705,0.5777778,0.5925926,0.54814816,0.5703704,0.4962963,,0.5703704,0.5037037,0.67407405,0.4962963,0.6,0.78518516,0.67407405,0.8,0.8740741,0.7185185,0.88148147,0.63703704,0.837037,0.74814814,0.8666667,0.62222224,0.9925926,0.8666667,0.9259259,0.95555556
+TorchMLP (default),0.0,0.0,0.022222223,0.0074074073,0.0,0.014814815,0.02962963,0.0074074073,0.074074075,0.037037037,0.05925926,0.044444446,0.044444446,0.08148148,0.0074074073,0.02962963,0.014814815,0.02962963,0.044444446,0.13333334,0.14814815,0.17777778,0.06666667,0.02962963,0.014814815,0.096296296,0.18518518,0.16296296,0.074074075,0.21481481,0.21481481,0.0074074073,0.11851852,0.17777778,0.2962963,0.28148147,0.25185186,0.27407408,0.2,0.16296296,0.074074075,0.4074074,0.27407408,0.23703703,0.25925925,0.2888889,0.38518518,0.42962962,0.37777779,0.26666668,0.52592593,0.4,0.46666667,0.52592593,0.4888889,0.42962962,,0.6,0.6,0.6148148,0.6074074,0.63703704,0.6888889,0.837037,0.77037036,0.7777778,0.76296294,0.7777778,0.82222223,0.8,0.84444445,0.77037036,0.91851854,0.9925926,0.91851854,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0074074073,0.06666667,0.014814815,0.0074074073,0.06666667,0.022222223,0.014814815,0.0074074073,0.022222223,0.022222223,0.022222223,0.014814815,0.044444446,0.13333334,0.096296296,0.08148148,0.08148148,0.05185185,0.08888889,0.014814815,0.15555556,0.044444446,0.14074074,0.08148148,0.096296296,0.17037037,0.16296296,0.22962964,0.2074074,0.037037037,0.26666668,0.25185186,0.18518518,0.33333334,0.2074074,0.22962964,0.14074074,0.34814814,0.22962964,0.2962963,0.43703705,0.25185186,0.26666668,0.37037036,0.34814814,0.4074074,0.36296296,0.37777779,0.38518518,0.36296296,0.31851852,0.42962962,0.42222223,0.39259258,0.4962963,0.4,,0.46666667,0.8962963,0.64444447,0.53333336,0.67407405,0.6296296,0.51111114,0.7185185,0.51111114,0.7111111,0.7037037,0.7259259,0.7111111,0.94814813,0.5777778,0.77037036,0.7111111,0.97037035
+FastaiMLP (tuned),0.014814815,0.0,0.044444446,0.014814815,0.037037037,0.044444446,0.08148148,0.014814815,0.044444446,0.022222223,0.08148148,0.02962963,0.05925926,0.014814815,0.02962963,0.08148148,0.05925926,0.13333334,0.022222223,0.17037037,0.13333334,0.14074074,0.14074074,0.08888889,0.08148148,0.05185185,0.16296296,0.13333334,0.096296296,0.15555556,0.21481481,0.11851852,0.19259259,0.11851852,0.21481481,0.21481481,0.32592592,0.14814815,0.2962963,0.3037037,0.25185186,0.2888889,0.11851852,0.26666668,0.31111112,0.26666668,0.2,0.37777779,0.31111112,0.31111112,0.51111114,0.3037037,0.17037037,0.51111114,0.4074074,0.32592592,0.4,0.53333336,,0.56296295,0.5925926,0.7777778,0.53333336,0.6074074,0.63703704,0.47407407,0.63703704,0.6666667,0.7777778,0.6,0.82222223,0.73333335,0.8518519,0.78518516,0.93333334,1.0
+PerpetualBooster (tuned),0.0,0.0,0.014814815,0.022222223,0.0074074073,0.06666667,0.014814815,0.0074074073,0.0074074073,0.0074074073,0.037037037,0.0074074073,0.014814815,0.02962963,0.06666667,0.037037037,0.06666667,0.074074075,0.02962963,0.037037037,0.014814815,0.14074074,0.022222223,0.044444446,0.074074075,0.05185185,0.15555556,0.096296296,0.17777778,0.17037037,0.02962963,0.24444444,0.11851852,0.15555556,0.2962963,0.17037037,0.13333334,0.11851852,0.35555556,0.15555556,0.26666668,0.39259258,0.2074074,0.18518518,0.36296296,0.31851852,0.38518518,0.31111112,0.36296296,0.36296296,0.36296296,0.2962963,0.4148148,0.38518518,0.33333334,0.5037037,0.38518518,0.1037037,0.43703705,,0.58518517,0.5185185,0.65185183,0.6,0.51111114,0.6962963,0.51111114,0.6148148,0.6814815,0.65925926,0.6962963,0.9259259,0.5407407,0.7111111,0.6888889,0.97037035
+xRFM (default),0.02962963,0.0,0.05185185,0.0,0.05185185,0.06666667,0.05185185,0.0,0.0,0.074074075,0.06666667,0.06666667,0.096296296,0.074074075,0.022222223,0.08888889,0.11111111,0.06666667,0.05185185,0.05925926,0.05185185,0.14074074,0.13333334,0.13333334,0.11851852,0.08148148,0.14814815,0.0,0.05925926,0.14814815,0.06666667,0.17777778,0.22222222,0.037037037,0.36296296,0.2,0.18518518,0.25925925,0.32592592,0.13333334,0.22962964,0.33333334,0.31111112,0.2074074,0.34074074,0.31851852,0.44444445,0.31851852,0.34814814,0.37777779,0.13333334,0.37777779,0.33333334,0.12592593,0.34074074,0.4,0.39259258,0.35555556,0.4074074,0.4148148,,0.45925927,0.44444445,0.52592593,0.53333336,0.56296295,0.53333336,0.4888889,0.4814815,0.6074074,0.5185185,0.6148148,0.5777778,0.65925926,0.52592593,0.9111111
+FastaiMLP (default),0.014814815,0.0,0.044444446,0.0074074073,0.05185185,0.05185185,0.06666667,0.022222223,0.022222223,0.014814815,0.074074075,0.037037037,0.08888889,0.02962963,0.0074074073,0.06666667,0.06666667,0.1037037,0.02962963,0.08148148,0.11111111,0.14814815,0.12592593,0.096296296,0.05925926,0.05925926,0.15555556,0.08888889,0.06666667,0.15555556,0.14814815,0.11111111,0.15555556,0.08888889,0.14814815,0.21481481,0.24444444,0.12592593,0.21481481,0.21481481,0.22222222,0.25925925,0.1037037,0.2,0.25185186,0.2,0.16296296,0.23703703,0.22962964,0.2888889,0.4074074,0.22222222,0.022222223,0.4074074,0.3037037,0.21481481,0.36296296,0.46666667,0.22222222,0.4814815,0.5407407,,0.4148148,0.53333336,0.5703704,0.42962962,0.5777778,0.43703705,0.7037037,0.5407407,0.75555557,0.6148148,0.73333335,0.67407405,0.78518516,0.8666667
+RandomForest (default),0.0074074073,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.0074074073,0.0074074073,0.0074074073,0.0,0.0074074073,0.014814815,0.0,0.0074074073,0.014814815,0.05925926,0.02962963,0.0074074073,0.06666667,0.0074074073,0.0,0.022222223,0.022222223,0.074074075,0.08148148,0.022222223,0.06666667,0.06666667,0.05925926,0.05185185,0.13333334,0.16666667,0.16296296,0.21481481,0.13333334,0.28148147,0.044444446,0.14814815,0.22962964,0.17777778,0.074074075,0.32592592,0.02962963,0.32222223,0.074074075,0.13703704,0.34074074,0.34074074,0.31111112,0.4,0.33333334,0.096296296,0.32592592,0.31111112,0.32592592,0.46666667,0.34814814,0.5555556,0.58518517,,0.5185185,0.5925926,0.4925926,0.5925926,0.58518517,0.6296296,0.55925924,0.65925926,0.76296294,0.67407405,0.77037036,0.67407405,1.0
+KNN (tuned + ensembled),0.02962963,0.0074074073,0.044444446,0.0074074073,0.037037037,0.02962963,0.05925926,0.0074074073,0.0074074073,0.0074074073,0.074074075,0.0074074073,0.014814815,0.0074074073,0.014814815,0.0074074073,0.037037037,0.06666667,0.022222223,0.096296296,0.1037037,0.037037037,0.05925926,0.0074074073,0.074074075,0.02962963,0.05185185,0.022222223,0.0074074073,0.06666667,0.13333334,0.05925926,0.074074075,0.014814815,0.11851852,0.14814815,0.25925925,0.1037037,0.16296296,0.11111111,0.1037037,0.2888889,0.14074074,0.15555556,0.2074074,0.074074075,0.08148148,0.17037037,0.096296296,0.21481481,0.45925927,0.14074074,0.28148147,0.45925927,0.2074074,0.2,0.16296296,0.37037036,0.39259258,0.4,0.47407407,0.46666667,0.4814815,,0.39259258,0.4074074,0.39259258,0.46666667,0.7185185,0.46666667,0.75555557,0.6666667,0.5925926,1.0,0.8148148,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.014814815,0.0,0.0,0.05185185,0.0,0.0074074073,0.0074074073,0.0074074073,0.0,0.0074074073,0.0074074073,0.0,0.0,0.014814815,0.074074075,0.096296296,0.06666667,0.0,0.0074074073,0.0,0.02962963,0.074074075,0.096296296,0.037037037,0.074074075,0.22222222,0.05185185,0.037037037,0.11111111,0.05185185,0.13333334,0.22222222,0.06666667,0.022222223,0.17037037,0.1037037,0.31851852,0.15555556,0.1037037,0.05185185,0.13333334,0.044444446,0.18518518,0.14074074,0.14074074,0.4,0.2888889,0.28148147,0.4,0.19259259,0.12592593,0.22962964,0.4888889,0.36296296,0.4888889,0.46666667,0.42962962,0.4074074,0.6074074,,0.43703705,0.5185185,0.5555556,0.7111111,0.5925926,0.7185185,0.53333336,1.0,0.78518516,0.8296296,0.8
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.02962963,0.0074074073,0.0074074073,0.014814815,0.0074074073,0.0074074073,0.014814815,0.0,0.0,0.0,0.014814815,0.06666667,0.0074074073,0.06666667,0.0074074073,0.0,0.0,0.014814815,0.074074075,0.022222223,0.014814815,0.06666667,0.05925926,0.05185185,0.0074074073,0.08888889,0.16666667,0.1037037,0.18518518,0.02962963,0.28148147,0.074074075,0.11111111,0.31111112,0.17037037,0.037037037,0.3037037,0.05185185,0.26296297,0.1037037,0.14814815,0.31111112,0.34814814,0.28148147,0.46666667,0.34074074,0.11111111,0.28148147,0.22222222,0.28148147,0.52592593,0.3037037,0.43703705,0.5703704,0.5074074,0.5925926,0.56296295,,0.56296295,0.5925926,0.6074074,0.5888889,0.62222224,0.6666667,0.6962963,0.65925926,0.6666667,0.962963
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.014814815,0.0,0.0,0.05185185,0.0,0.0,0.0074074073,0.0074074073,0.0,0.0074074073,0.0074074073,0.0,0.0,0.014814815,0.074074075,0.096296296,0.06666667,0.0,0.0074074073,0.0,0.02962963,0.074074075,0.096296296,0.037037037,0.06666667,0.21481481,0.05185185,0.02962963,0.1037037,0.044444446,0.13333334,0.22222222,0.06666667,0.05925926,0.16296296,0.1037037,0.31111112,0.15555556,0.096296296,0.08148148,0.12592593,0.037037037,0.18518518,0.13333334,0.15555556,0.4,0.27407408,0.28148147,0.4,0.2,0.11851852,0.23703703,0.4888889,0.36296296,0.4888889,0.46666667,0.42222223,0.4074074,0.6074074,0.4814815,0.43703705,,0.5555556,0.7111111,0.5925926,0.7185185,0.53333336,1.0,0.78518516,0.8296296,0.8
+ExtraTrees (default),0.0074074073,0.0,0.0,0.0074074073,0.0,0.0,0.0074074073,0.0,0.014814815,0.0074074073,0.014814815,0.014814815,0.0074074073,0.022222223,0.037037037,0.0074074073,0.0,0.0074074073,0.037037037,0.05185185,0.0074074073,0.074074075,0.0074074073,0.0074074073,0.014814815,0.05925926,0.08148148,0.014814815,0.05185185,0.08148148,0.044444446,0.096296296,0.12592593,0.05185185,0.14814815,0.11851852,0.22222222,0.08148148,0.27407408,0.11111111,0.18518518,0.3037037,0.08148148,0.05925926,0.31111112,0.14074074,0.24444444,0.014814815,0.17777778,0.31851852,0.27407408,0.22962964,0.24444444,0.26666668,0.037037037,0.36296296,0.22222222,0.2888889,0.33333334,0.38518518,0.51111114,0.56296295,0.4148148,0.53333336,0.44444445,0.4074074,0.44444445,,0.63703704,0.5407407,0.65925926,0.7259259,0.5703704,0.7037037,0.6814815,1.0
+Linear (tuned + ensembled),0.014814815,0.0,0.05185185,0.0,0.022222223,0.05185185,0.1037037,0.0,0.06666667,0.0074074073,0.11851852,0.0074074073,0.014814815,0.0074074073,0.014814815,0.0,0.044444446,0.1037037,0.014814815,0.17777778,0.16296296,0.0,0.08888889,0.0,0.074074075,0.08888889,0.0,0.014814815,0.06666667,0.0,0.2074074,0.05925926,0.08148148,0.02962963,0.16296296,0.18518518,0.2,0.13333334,0.06666667,0.2,0.08148148,0.23703703,0.11111111,0.2074074,0.096296296,0.13333334,0.16296296,0.18518518,0.19259259,0.12592593,0.37037036,0.12592593,0.074074075,0.37037036,0.19259259,0.16296296,0.17777778,0.2962963,0.22222222,0.31851852,0.5185185,0.2962963,0.37037036,0.28148147,0.2888889,0.39259258,0.2888889,0.36296296,,0.45925927,0.837037,0.51111114,0.33333334,0.4148148,0.95555556,0.6666667
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0074074073,0.0074074073,0.0074074073,0.0,0.0074074073,0.0074074073,0.0,0.0,0.0,0.014814815,0.074074075,0.022222223,0.06666667,0.0074074073,0.0,0.0,0.022222223,0.06666667,0.014814815,0.0074074073,0.06666667,0.06666667,0.05185185,0.0,0.02962963,0.18148148,0.1037037,0.096296296,0.044444446,0.21481481,0.06666667,0.1037037,0.2962963,0.08888889,0.05185185,0.22962964,0.02962963,0.18888889,0.08888889,0.11111111,0.23703703,0.28148147,0.21481481,0.33333334,0.27407408,0.08888889,0.25185186,0.2,0.27407408,0.4,0.34074074,0.39259258,0.45925927,0.44074073,0.53333336,0.4074074,0.41111112,0.4074074,0.45925927,0.5407407,,0.56296295,0.6666667,0.54814816,0.7037037,0.65185183,0.8666667
+Linear (tuned),0.014814815,0.0,0.044444446,0.0,0.022222223,0.037037037,0.096296296,0.0,0.05925926,0.0,0.08888889,0.0,0.0074074073,0.0,0.0074074073,0.0,0.037037037,0.1037037,0.0074074073,0.16296296,0.16296296,0.0,0.05925926,0.0,0.06666667,0.06666667,0.0,0.0,0.044444446,0.0,0.17037037,0.05925926,0.074074075,0.0074074073,0.16296296,0.16296296,0.19259259,0.14074074,0.05185185,0.17037037,0.08148148,0.22962964,0.08148148,0.2,0.074074075,0.096296296,0.15555556,0.14074074,0.14814815,0.1037037,0.34074074,0.1037037,0.06666667,0.34074074,0.17037037,0.13333334,0.15555556,0.2888889,0.17777778,0.3037037,0.4814815,0.24444444,0.34074074,0.24444444,0.28148147,0.37777779,0.28148147,0.34074074,0.16296296,0.43703705,,0.4962963,0.33333334,0.4148148,0.88148147,0.6666667
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.06666667,0.05925926,0.0,0.0,0.0,0.11851852,0.0,0.0,0.06666667,0.0,0.11111111,0.06666667,0.05925926,0.13333334,0.0,0.074074075,0.0,0.074074075,0.096296296,0.06666667,0.014814815,0.06666667,0.24444444,0.05185185,0.096296296,0.2074074,0.06666667,0.06666667,0.26666668,0.15555556,0.16296296,0.13333334,0.16296296,0.27407408,0.14074074,0.2,0.19259259,0.14814815,0.14074074,0.37777779,0.22962964,0.05185185,0.26666668,0.074074075,0.38518518,0.38518518,0.23703703,0.33333334,0.46666667,0.33333334,0.46666667,0.27407408,0.4888889,0.33333334,0.5037037,,0.46666667,0.45185184,0.5037037,0.91851854
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.0,0.0,0.0,0.0,0.0,0.05185185,0.0,0.0,0.0,0.0,0.05185185,0.06666667,0.0074074073,0.06666667,0.05925926,0.0,0.0,0.0074074073,0.0,0.05185185,0.19259259,0.0,0.0,0.0,0.014814815,0.25185186,0.022222223,0.0,0.0,0.044444446,0.0,0.096296296,0.044444446,0.0,0.32592592,0.02962963,0.15555556,0.32592592,0.1037037,0.0074074073,0.08148148,0.42222223,0.14814815,0.45925927,0.42222223,0.26666668,0.32592592,0.4074074,0.0,0.3037037,0.0,0.42962962,0.6666667,0.45185184,0.6666667,0.53333336,,0.6666667,0.6814815,0.78518516
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.022222223,0.014814815,0.0,0.0,0.0,0.0,0.022222223,0.17777778,0.0,0.074074075,0.0,0.0,0.21481481,0.0,0.0,0.08148148,0.0,0.0,0.06666667,0.0,0.11851852,0.22962964,0.0,0.2074074,0.22962964,0.05925926,0.13333334,0.0074074073,0.22962964,0.21481481,0.2888889,0.34074074,0.32592592,0.22962964,0.0,0.21481481,0.34074074,0.21481481,0.2962963,0.58518517,0.2962963,0.58518517,0.54814816,0.33333334,,0.58518517,0.9925926
+Linear (default),0.0,0.0,0.02962963,0.0,0.022222223,0.022222223,0.02962963,0.0,0.014814815,0.0,0.044444446,0.0,0.0,0.0,0.0074074073,0.0,0.014814815,0.05185185,0.0,0.08148148,0.074074075,0.0,0.022222223,0.0,0.044444446,0.0074074073,0.0,0.0,0.0,0.0,0.13333334,0.02962963,0.06666667,0.0,0.044444446,0.12592593,0.17037037,0.06666667,0.037037037,0.08888889,0.05185185,0.22962964,0.05925926,0.12592593,0.05185185,0.02962963,0.037037037,0.06666667,0.05925926,0.05925926,0.32592592,0.014814815,0.014814815,0.32592592,0.074074075,0.074074075,0.08148148,0.2888889,0.06666667,0.31111112,0.47407407,0.21481481,0.32592592,0.18518518,0.17037037,0.33333334,0.17037037,0.31851852,0.044444446,0.34814814,0.11851852,0.4962963,0.31851852,0.4148148,,0.6296296
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0074074073,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.074074075,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.02962963,0.0,0.02962963,0.08888889,0.13333334,0.0,0.0,0.2,0.037037037,0.2,0.0,0.33333334,0.13333334,0.33333334,0.08148148,0.21481481,0.0074074073,0.37037036,
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_small/leaderboard_overview_explorer.html
index 9819bab6a6d9c0094838acf0d18575bef78ebcaf..65f617c8675460ab2a847de5b57a8fab676b71d8 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — All Tasks | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_all/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..5d19ea53b48cbe3d7d68c13f0ea647f2b807cc81
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_all/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1728.0,"score":0.889,"rank":7.78,"hrank":1.84,"imp":4.756,"train_time":39.57,"predict_time":5.358,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-86","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1621.0,"score":0.665,"rank":12.23,"hrank":5.01,"imp":10.865,"train_time":9.17,"predict_time":0.416,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-52","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1617.0,"score":0.63,"rank":12.44,"hrank":5.37,"imp":9.378,"train_time":346.58,"predict_time":6.557,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-44","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1580.0,"score":0.605,"rank":14.39,"hrank":7.85,"imp":12.059,"train_time":7.03,"predict_time":0.548,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-45","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1580.0,"score":0.583,"rank":14.41,"hrank":6.49,"imp":11.297,"train_time":2289.39,"predict_time":8.048,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-61","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1556.0,"score":0.624,"rank":15.8,"hrank":5.81,"imp":11.66,"train_time":3.93,"predict_time":0.152,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-73","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1540.0,"score":0.538,"rank":16.8,"hrank":7.33,"imp":12.006,"train_time":2289.39,"predict_time":1.143,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-54","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1526.0,"score":0.526,"rank":17.7,"hrank":9.63,"imp":12.418,"train_time":6.76,"predict_time":0.636,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-48","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1460.0,"score":0.393,"rank":22.33,"hrank":12.93,"imp":14.629,"train_time":3771.85,"predict_time":21.902,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-44","imputed":false},{"position":9,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1455.0,"score":0.422,"rank":22.75,"hrank":10.47,"imp":14.85,"train_time":3528.13,"predict_time":48.589,"imputed_pct":8.33,"hardware":"GPU","elo_ci":"+87\/-60","imputed":true},{"position":10,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1436.0,"score":0.405,"rank":24.21,"hrank":11.48,"imp":15.586,"train_time":126.87,"predict_time":3.093,"imputed_pct":8.33,"hardware":"GPU","elo_ci":"+69\/-56","imputed":true},{"position":11,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1433.0,"score":0.386,"rank":24.47,"hrank":10.3,"imp":15.342,"train_time":5119.78,"predict_time":218.453,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-55","imputed":false},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1433.0,"score":0.384,"rank":24.49,"hrank":10.15,"imp":15.424,"train_time":4.69,"predict_time":0.192,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-51","imputed":false},{"position":13,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1424.0,"score":0.314,"rank":25.2,"hrank":15.29,"imp":15.693,"train_time":3716.57,"predict_time":2.547,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-46","imputed":false},{"position":14,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1409.0,"score":0.313,"rank":26.39,"hrank":13.82,"imp":15.77,"train_time":2585.58,"predict_time":3.838,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-44","imputed":false},{"position":15,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1401.0,"score":0.346,"rank":27.03,"hrank":12.25,"imp":15.978,"train_time":5119.78,"predict_time":28.351,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-55","imputed":false},{"position":16,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1392.0,"score":0.341,"rank":27.83,"hrank":10.93,"imp":15.249,"train_time":35.51,"predict_time":6.828,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-72","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1383.0,"score":0.286,"rank":28.6,"hrank":15.24,"imp":16.004,"train_time":3771.85,"predict_time":1.778,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-40","imputed":false},{"position":18,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1373.0,"score":0.225,"rank":29.47,"hrank":22.22,"imp":17.029,"train_time":892.49,"predict_time":2.57,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-34","imputed":false},{"position":19,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1370.0,"score":0.315,"rank":29.74,"hrank":13.22,"imp":16.958,"train_time":3528.13,"predict_time":0.817,"imputed_pct":8.33,"hardware":"GPU","elo_ci":"+70\/-62","imputed":true},{"position":20,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1356.0,"score":0.246,"rank":30.92,"hrank":15.78,"imp":16.654,"train_time":2585.58,"predict_time":0.436,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-48","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1352.0,"score":0.208,"rank":31.27,"hrank":22.78,"imp":17.133,"train_time":1739.17,"predict_time":0.396,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-39","imputed":false},{"position":22,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1352.0,"score":0.284,"rank":31.33,"hrank":19.0,"imp":17.319,"train_time":2.76,"predict_time":0.058,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-48","imputed":false},{"position":23,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1348.0,"score":0.195,"rank":31.68,"hrank":21.7,"imp":17.51,"train_time":22811.74,"predict_time":841.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+41\/-38","imputed":false},{"position":24,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1340.0,"score":0.187,"rank":32.39,"hrank":23.89,"imp":17.371,"train_time":1739.17,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+37\/-33","imputed":false},{"position":25,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1339.0,"score":0.176,"rank":32.45,"hrank":25.03,"imp":17.481,"train_time":892.49,"predict_time":0.346,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+31\/-31","imputed":false},{"position":26,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1330.0,"score":0.285,"rank":33.29,"hrank":15.8,"imp":17.994,"train_time":4.54,"predict_time":0.443,"imputed_pct":8.33,"hardware":"GPU","elo_ci":"+70\/-63","imputed":true},{"position":27,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1323.0,"score":0.27,"rank":33.9,"hrank":15.24,"imp":18.116,"train_time":50.33,"predict_time":43.707,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-68","imputed":false},{"position":28,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1319.0,"score":0.195,"rank":34.26,"hrank":20.65,"imp":17.96,"train_time":1126.97,"predict_time":2.94,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+45\/-41","imputed":false},{"position":29,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1314.0,"score":0.248,"rank":34.74,"hrank":17.32,"imp":18.692,"train_time":7.15,"predict_time":1.469,"imputed_pct":27.78,"hardware":"GPU","elo_ci":"+77\/-101","imputed":true},{"position":30,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1313.0,"score":0.151,"rank":34.82,"hrank":26.54,"imp":18.014,"train_time":884.48,"predict_time":2.367,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-35","imputed":false},{"position":31,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1310.0,"score":0.175,"rank":35.02,"hrank":17.43,"imp":18.145,"train_time":6523.55,"predict_time":0.505,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+40\/-41","imputed":false},{"position":32,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1310.0,"score":0.218,"rank":35.03,"hrank":14.21,"imp":18.408,"train_time":6523.55,"predict_time":8.39,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-55","imputed":false},{"position":33,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1308.0,"score":0.253,"rank":35.29,"hrank":12.29,"imp":18.927,"train_time":16.23,"predict_time":1.277,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-56","imputed":false},{"position":34,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1306.0,"score":0.155,"rank":35.39,"hrank":25.67,"imp":18.294,"train_time":8.46,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+36\/-36","imputed":false},{"position":35,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1300.0,"score":0.241,"rank":35.95,"hrank":19.35,"imp":19.209,"train_time":296.99,"predict_time":1.835,"imputed_pct":22.22,"hardware":"GPU","elo_ci":"+58\/-60","imputed":true},{"position":36,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1296.0,"score":0.124,"rank":36.37,"hrank":26.06,"imp":18.223,"train_time":884.48,"predict_time":0.393,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+33\/-40","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1283.0,"score":0.185,"rank":37.56,"hrank":21.97,"imp":18.349,"train_time":9.94,"predict_time":0.445,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-46","imputed":false},{"position":38,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1274.0,"score":0.148,"rank":38.39,"hrank":23.41,"imp":18.986,"train_time":1126.97,"predict_time":0.163,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+40\/-37","imputed":false},{"position":39,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1267.0,"score":0.129,"rank":39.04,"hrank":27.86,"imp":19.606,"train_time":1218.87,"predict_time":0.455,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-48","imputed":false},{"position":40,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1250.0,"score":0.105,"rank":40.62,"hrank":28.94,"imp":19.892,"train_time":1218.87,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-51","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1248.0,"score":0.125,"rank":40.73,"hrank":28.47,"imp":19.213,"train_time":3865.95,"predict_time":2.157,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-50","imputed":false},{"position":42,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1248.0,"score":0.093,"rank":40.75,"hrank":29.8,"imp":19.336,"train_time":22811.74,"predict_time":161.221,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+35\/-36","imputed":false},{"position":43,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1248.0,"score":0.102,"rank":40.8,"hrank":29.13,"imp":20.042,"train_time":2499.4,"predict_time":0.153,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-41","imputed":false},{"position":44,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1230.0,"score":0.092,"rank":42.42,"hrank":23.7,"imp":21.307,"train_time":16.28,"predict_time":0.321,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+45\/-49","imputed":false},{"position":45,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1220.0,"score":0.075,"rank":43.31,"hrank":30.56,"imp":20.597,"train_time":2499.4,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-43","imputed":false},{"position":46,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1213.0,"score":0.075,"rank":44.02,"hrank":30.87,"imp":20.11,"train_time":15.69,"predict_time":4.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+38\/-39","imputed":false},{"position":47,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1208.0,"score":0.084,"rank":44.48,"hrank":28.23,"imp":21.029,"train_time":5.49,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-59","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1206.0,"score":0.119,"rank":44.6,"hrank":24.98,"imp":20.8,"train_time":1679.4,"predict_time":8.069,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-59","imputed":false},{"position":49,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1206.0,"score":0.102,"rank":44.63,"hrank":27.86,"imp":21.724,"train_time":822.58,"predict_time":1.501,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-62","imputed":false},{"position":50,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1194.0,"score":0.07,"rank":45.74,"hrank":29.02,"imp":21.335,"train_time":7.39,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-50","imputed":false},{"position":51,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1187.0,"score":0.079,"rank":46.38,"hrank":33.71,"imp":20.949,"train_time":3865.95,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-49","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1184.0,"score":0.095,"rank":46.67,"hrank":25.48,"imp":22.422,"train_time":822.58,"predict_time":0.214,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-63","imputed":false},{"position":53,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1161.0,"score":0.018,"rank":48.73,"hrank":42.21,"imp":21.328,"train_time":3.29,"predict_time":0.253,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+26\/-38","imputed":false},{"position":54,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1160.0,"score":0.046,"rank":48.79,"hrank":39.83,"imp":21.316,"train_time":3.61,"predict_time":0.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-40","imputed":false},{"position":55,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1160.0,"score":0.058,"rank":48.81,"hrank":30.53,"imp":21.934,"train_time":1679.4,"predict_time":0.826,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-51","imputed":false},{"position":56,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1151.0,"score":0.071,"rank":49.62,"hrank":29.09,"imp":23.295,"train_time":590.04,"predict_time":1.46,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-65","imputed":false},{"position":57,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1149.0,"score":0.185,"rank":49.78,"hrank":9.61,"imp":24.991,"train_time":0.58,"predict_time":0.117,"imputed_pct":77.78,"hardware":"GPU","elo_ci":"+108\/-100","imputed":true},{"position":58,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1142.0,"score":0.167,"rank":50.37,"hrank":18.7,"imp":25.334,"train_time":0.58,"predict_time":0.117,"imputed_pct":77.78,"hardware":"GPU","elo_ci":"+98\/-97","imputed":true},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1126.0,"score":0.174,"rank":51.77,"hrank":13.03,"imp":25.567,"train_time":40848.87,"predict_time":5.606,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-77","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1125.0,"score":0.171,"rank":51.87,"hrank":12.42,"imp":25.6,"train_time":40848.87,"predict_time":25.936,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-77","imputed":false},{"position":61,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1121.0,"score":0.077,"rank":52.16,"hrank":35.98,"imp":24.125,"train_time":13.75,"predict_time":2.405,"imputed_pct":22.22,"hardware":"GPU","elo_ci":"+68\/-57","imputed":true},{"position":62,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1112.0,"score":0.043,"rank":52.93,"hrank":28.01,"imp":24.027,"train_time":590.04,"predict_time":0.15,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-62","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1080.0,"score":0.027,"rank":55.61,"hrank":41.34,"imp":28.534,"train_time":342.04,"predict_time":0.895,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-57","imputed":false},{"position":64,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1063.0,"score":0.038,"rank":56.94,"hrank":35.06,"imp":24.994,"train_time":524.48,"predict_time":161.813,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-69","imputed":false},{"position":65,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1057.0,"score":0.012,"rank":57.4,"hrank":42.17,"imp":24.844,"train_time":13.69,"predict_time":0.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-49","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1046.0,"score":0.016,"rank":58.24,"hrank":41.07,"imp":29.989,"train_time":342.04,"predict_time":0.313,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-56","imputed":false},{"position":67,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1041.0,"score":0.028,"rank":58.66,"hrank":49.38,"imp":26.72,"train_time":5.45,"predict_time":1.098,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-82","imputed":false},{"position":68,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1034.0,"score":0.05,"rank":59.17,"hrank":38.99,"imp":27.662,"train_time":1.58,"predict_time":0.15,"imputed_pct":22.22,"hardware":"GPU","elo_ci":"+72\/-70","imputed":true},{"position":69,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1030.0,"score":0.106,"rank":59.42,"hrank":20.62,"imp":30.432,"train_time":471.66,"predict_time":7.214,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-116","imputed":false},{"position":70,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1020.0,"score":0.014,"rank":60.19,"hrank":50.92,"imp":27.665,"train_time":4.19,"predict_time":0.566,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-71","imputed":false},{"position":71,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1001.0,"score":0.001,"rank":61.54,"hrank":42.32,"imp":26.937,"train_time":5.48,"predict_time":0.684,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-62","imputed":false},{"position":72,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.003,"rank":61.58,"hrank":55.34,"imp":28.452,"train_time":0.51,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-70","imputed":false},{"position":73,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":994.0,"score":0.023,"rank":61.96,"hrank":39.49,"imp":29.341,"train_time":111.87,"predict_time":1.214,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-92","imputed":false},{"position":74,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":968.0,"score":0.041,"rank":63.73,"hrank":42.15,"imp":34.059,"train_time":521.07,"predict_time":0.646,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-94","imputed":false},{"position":75,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":948.0,"score":0.029,"rank":64.93,"hrank":44.27,"imp":34.471,"train_time":521.07,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-96","imputed":false},{"position":76,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":946.0,"score":0.0,"rank":65.07,"hrank":46.34,"imp":32.794,"train_time":39.09,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-56","imputed":false},{"position":77,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":907.0,"score":0.014,"rank":67.28,"hrank":47.7,"imp":34.023,"train_time":111.87,"predict_time":0.111,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-90","imputed":false},{"position":78,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":874.0,"score":0.018,"rank":68.95,"hrank":48.75,"imp":37.306,"train_time":2.59,"predict_time":0.138,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-113","imputed":false},{"position":79,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":661.0,"score":0.0,"rank":76.25,"hrank":65.22,"imp":47.082,"train_time":0.37,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-127","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_small/pareto_front_explorer.html
index 1fa59d13f08935d27979ad7dc1e1bb228b2b869b..6c7e82f6d559308c56fb94493e055088059e7997 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8345788263,"imp":19.2094289066,"elo":1300.4,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0455896074,"imp":18.2940779389,"elo":1306.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0508532668,"imp":17.371354621,"elo":1339.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"imp":17.1334222934,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0413889487,"imp":21.0289155338,"elo":1207.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0359833093,"imp":19.8921774901,"elo":1249.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.454903865,"imp":19.6057819258,"elo":1266.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0171560587,"imp":21.3345349506,"elo":1194.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0172938338,"imp":20.5972055382,"elo":1220.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1529693961,"imp":20.0420166646,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5663988088,"imp":27.6650148671,"elo":1019.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2136923717,"imp":22.4216315405,"elo":1183.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5011279394,"imp":21.7244455159,"elo":1206.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6842938244,"imp":26.936575539,"elo":1000.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.82578692,"imp":21.9341608839,"elo":1159.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.0688374767,"imp":20.7999663895,"elo":1206.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1695753133,"imp":21.3156578084,"elo":1160.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3458375858,"imp":17.4807648653,"elo":1339.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.5704534161,"imp":17.0294466537,"elo":1372.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.8284826783,"imp":15.2489924448,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.0927295997,"imp":15.5856969713,"elo":1436.1,"imputed":true,"imputed_pct":8.3333333333,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.320600689,"imp":21.3073490105,"elo":1230.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5053503954,"imp":18.1449452601,"elo":1310.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3903226253,"imp":18.4077054585,"elo":1310.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.1169043969,"imp":25.3339841709,"elo":1142.1,"imputed":true,"imputed_pct":77.7777777778,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.1169043969,"imp":24.9907871052,"elo":1148.8,"imputed":true,"imputed_pct":77.7777777778,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.405321849,"imp":24.1252282165,"elo":1121.3,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0406570787,"imp":32.7937352449,"elo":945.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3125464363,"imp":29.988612581,"elo":1046.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.8948102984,"imp":28.5335048871,"elo":1079.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0885648768,"imp":28.45210173,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1503007896,"imp":24.0271644712,"elo":1112.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.4602306271,"imp":23.295042659,"elo":1150.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.6888676463,"imp":20.1095148651,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.7776750936,"imp":16.0035177424,"elo":1382.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.9024962791,"imp":14.629393378,"elo":1460.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6362272808,"imp":12.4178435689,"elo":1525.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1425914845,"imp":12.0061472435,"elo":1540.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0475948482,"imp":11.2974940217,"elo":1579.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2773508677,"imp":18.9267144191,"elo":1307.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"imp":18.1162614716,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":28.3513403832,"imp":15.9784417919,"elo":1401.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":218.4533640735,"imp":15.3419526681,"elo":1432.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1922706752,"imp":15.4238624608,"elo":1432.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.3578381333,"imp":4.7556133385,"elo":1728.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1496544963,"imp":27.6618542939,"elo":1033.7,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.4691971112,"imp":18.6920286136,"elo":1313.6,"imputed":true,"imputed_pct":27.7777777778,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1520731745,"imp":11.6595467377,"elo":1556.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.444937792,"imp":18.3486497169,"elo":1282.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4361574183,"imp":16.6544918214,"elo":1356.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.8381409686,"imp":15.7700142501,"elo":1409.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.548288662,"imp":12.0590970263,"elo":1580.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4158216639,"imp":10.864904155,"elo":1620.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4430066712,"imp":17.9936493789,"elo":1329.6,"imputed":true,"imputed_pct":8.3333333333,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.8166246618,"imp":16.9577521072,"elo":1369.7,"imputed":true,"imputed_pct":8.3333333333,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":48.5886271028,"imp":14.8498888427,"elo":1454.8,"imputed":true,"imputed_pct":8.3333333333,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.2137264423,"imp":30.4323313952,"elo":1030.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.6062080426,"imp":25.5669165755,"elo":1125.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9362072534,"imp":25.6002975216,"elo":1124.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0579126899,"imp":17.3191467946,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1550906392,"imp":24.8436649749,"elo":1057.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1474394112,"imp":20.9493781861,"elo":1186.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1570271396,"imp":19.2133788912,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2525411892,"imp":21.3280999428,"elo":1160.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3929683731,"imp":18.2230440923,"elo":1295.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.3674863949,"imp":18.0143178337,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":161.8134383214,"imp":24.9937226794,"elo":1062.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":161.2206474003,"imp":19.3360256848,"elo":1248.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"imp":17.5101706439,"elo":1347.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.0976028857,"imp":26.7197245984,"elo":1040.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1632446807,"imp":18.9859245823,"elo":1273.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9395549707,"imp":17.9604545479,"elo":1318.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8345788263,"x_train":296.99454477,"imp":19.2094289066,"elo":1300.4,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0455896074,"x_train":8.4625903355,"imp":18.2940779389,"elo":1306.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0508532668,"x_train":1739.1690497298,"imp":17.371354621,"elo":1339.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"x_train":1739.1690497298,"imp":17.1334222934,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0413889487,"x_train":5.4899520078,"imp":21.0289155338,"elo":1207.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0359833093,"x_train":1218.8724153796,"imp":19.8921774901,"elo":1249.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.454903865,"x_train":1218.8724153796,"imp":19.6057819258,"elo":1266.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0171560587,"x_train":7.3884202915,"imp":21.3345349506,"elo":1194.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0172938338,"x_train":2499.3997394449,"imp":20.5972055382,"elo":1220.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1529693961,"x_train":2499.3997394449,"imp":20.0420166646,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5663988088,"x_train":4.1941752843,"imp":27.6650148671,"elo":1019.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2136923717,"x_train":822.5846846043,"imp":22.4216315405,"elo":1183.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5011279394,"x_train":822.5846846043,"imp":21.7244455159,"elo":1206.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6842938244,"x_train":5.483894574,"imp":26.936575539,"elo":1000.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.82578692,"x_train":1679.3959423676,"imp":21.9341608839,"elo":1159.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.0688374767,"x_train":1679.3959423676,"imp":20.7999663895,"elo":1206.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1695753133,"x_train":3.6086515992,"imp":21.3156578084,"elo":1160.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3458375858,"x_train":892.485407479,"imp":17.4807648653,"elo":1339.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.5704534161,"x_train":892.485407479,"imp":17.0294466537,"elo":1372.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.8284826783,"x_train":35.5096105631,"imp":15.2489924448,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.0927295997,"x_train":126.8739666394,"imp":15.5856969713,"elo":1436.1,"imputed":true,"imputed_pct":8.3333333333,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.320600689,"x_train":16.2796239102,"imp":21.3073490105,"elo":1230.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5053503954,"x_train":6523.5503361576,"imp":18.1449452601,"elo":1310.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3903226253,"x_train":6523.5503361576,"imp":18.4077054585,"elo":1310.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.1169043969,"x_train":0.5832555357,"imp":25.3339841709,"elo":1142.1,"imputed":true,"imputed_pct":77.7777777778,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.1169043969,"x_train":0.5832555357,"imp":24.9907871052,"elo":1148.8,"imputed":true,"imputed_pct":77.7777777778,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.405321849,"x_train":13.7518758578,"imp":24.1252282165,"elo":1121.3,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0406570787,"x_train":39.0907116856,"imp":32.7937352449,"elo":945.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3125464363,"x_train":342.0390094668,"imp":29.988612581,"elo":1046.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.8948102984,"x_train":342.0390094668,"imp":28.5335048871,"elo":1079.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0885648768,"x_train":0.5088535014,"imp":28.45210173,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1503007896,"x_train":590.0394439851,"imp":24.0271644712,"elo":1112.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.4602306271,"x_train":590.0394439851,"imp":23.295042659,"elo":1150.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.6888676463,"x_train":15.693235166,"imp":20.1095148651,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.7776750936,"x_train":3771.8540221068,"imp":16.0035177424,"elo":1382.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.9024962791,"x_train":3771.8540221068,"imp":14.629393378,"elo":1460.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6362272808,"x_train":6.7587899978,"imp":12.4178435689,"elo":1525.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1425914845,"x_train":2289.3887157667,"imp":12.0061472435,"elo":1540.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0475948482,"x_train":2289.3887157667,"imp":11.2974940217,"elo":1579.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2773508677,"x_train":16.2304597591,"imp":18.9267144191,"elo":1307.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"x_train":50.3250241527,"imp":18.1162614716,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":28.3513403832,"x_train":5119.7769851031,"imp":15.9784417919,"elo":1401.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":218.4533640735,"x_train":5119.7769851031,"imp":15.3419526681,"elo":1432.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1922706752,"x_train":4.6911130946,"imp":15.4238624608,"elo":1432.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.3578381333,"x_train":39.5720422516,"imp":4.7556133385,"elo":1728.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1496544963,"x_train":1.5770540896,"imp":27.6618542939,"elo":1033.7,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.4691971112,"x_train":7.1465625212,"imp":18.6920286136,"elo":1313.6,"imputed":true,"imputed_pct":27.7777777778,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1520731745,"x_train":3.9320097545,"imp":11.6595467377,"elo":1556.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.444937792,"x_train":9.9360092189,"imp":18.3486497169,"elo":1282.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4361574183,"x_train":2585.5798636409,"imp":16.6544918214,"elo":1356.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.8381409686,"x_train":2585.5798636409,"imp":15.7700142501,"elo":1409.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.548288662,"x_train":7.0342890773,"imp":12.0590970263,"elo":1580.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4158216639,"x_train":9.1699299544,"imp":10.864904155,"elo":1620.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4430066712,"x_train":4.5443881432,"imp":17.9936493789,"elo":1329.6,"imputed":true,"imputed_pct":8.3333333333,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.8166246618,"x_train":3528.1289596077,"imp":16.9577521072,"elo":1369.7,"imputed":true,"imputed_pct":8.3333333333,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":48.5886271028,"x_train":3528.1289596077,"imp":14.8498888427,"elo":1454.8,"imputed":true,"imputed_pct":8.3333333333,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.2137264423,"x_train":471.6614585494,"imp":30.4323313952,"elo":1030.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.6062080426,"x_train":40848.8665748301,"imp":25.5669165755,"elo":1125.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9362072534,"x_train":40848.8665748301,"imp":25.6002975216,"elo":1124.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0579126899,"x_train":2.7560512439,"imp":17.3191467946,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1550906392,"x_train":13.688894957,"imp":24.8436649749,"elo":1057.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1474394112,"x_train":3865.9507169384,"imp":20.9493781861,"elo":1186.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1570271396,"x_train":3865.9507169384,"imp":19.2133788912,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2525411892,"x_train":3.2873922338,"imp":21.3280999428,"elo":1160.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3929683731,"x_train":884.4808277373,"imp":18.2230440923,"elo":1295.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.3674863949,"x_train":884.4808277373,"imp":18.0143178337,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":161.8134383214,"x_train":524.4765295004,"imp":24.9937226794,"elo":1062.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":161.2206474003,"x_train":22811.7440512431,"imp":19.3360256848,"elo":1248.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"x_train":22811.7440512431,"imp":17.5101706439,"elo":1347.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.0976028857,"x_train":5.4460301782,"imp":26.7197245984,"elo":1040.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1632446807,"x_train":1126.9686387245,"imp":18.9859245823,"elo":1273.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9395549707,"x_train":1126.9686387245,"imp":17.9604545479,"elo":1318.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_small/tuning_trajectories_explorer.html
index 2ace60abf1ab598b9e65395badbab0babff3827f..9f7a4f2a5d1eae80973fbb0bf6e74df9e979cb0c 100644
--- a/data/imputation_yes/splits_all/tasks_all/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":296.99454477,"imp":19.2302574972,"elo":1318.6,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":8.4625903355,"imp":18.3141683167,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":20.3461935557,"imp":17.705728756,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":49.7230862409,"imp":17.3401560779,"elo":1367.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":92.4850850095,"imp":17.2055626238,"elo":1372.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":216.0131399202,"imp":17.1007980535,"elo":1375.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":428.4029173435,"imp":17.1039327772,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":862.883798881,"imp":17.1207794699,"elo":1373.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1291.8438396447,"imp":17.1451942817,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1739.1690497298,"imp":17.1514607587,"elo":1370.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.4899520078,"imp":21.0483818943,"elo":1217.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.3341031504,"imp":20.7357573647,"elo":1233.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":28.8927967722,"imp":20.5522681207,"elo":1244.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":58.8720319788,"imp":20.3801393556,"elo":1251.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":150.365699836,"imp":20.0735932675,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":302.4787198089,"imp":19.9161165885,"elo":1270.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":606.008218265,"imp":19.7426134538,"elo":1277.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":908.7557510718,"imp":19.6696734269,"elo":1279.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1218.8724153796,"imp":19.6247954852,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.3884202915,"imp":21.3548165158,"elo":1202.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":19.1397994387,"imp":21.0785374923,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":59.6568017021,"imp":20.7193575493,"elo":1230.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":123.907431834,"imp":20.5701164566,"elo":1238.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":318.4600106407,"imp":20.373046132,"elo":1246.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":623.1113495163,"imp":20.2363679583,"elo":1253.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1256.5403166692,"imp":20.1306653029,"elo":1257.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1885.0271107114,"imp":20.0926009226,"elo":1260.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2499.3997394449,"imp":20.0664064873,"elo":1262.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.1941752843,"imp":27.6835813219,"elo":1021.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":8.7200784389,"imp":25.9614675928,"elo":1063.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":21.6852759737,"imp":23.5184175953,"elo":1129.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.1231699977,"imp":22.6924799279,"elo":1169.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":107.3953873718,"imp":22.0801919052,"elo":1199.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":210.0602463393,"imp":21.9198958366,"elo":1208.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":412.076486211,"imp":21.8159108254,"elo":1214.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":614.2406759256,"imp":21.7556074736,"elo":1219.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":822.5846846043,"imp":21.7451335052,"elo":1219.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.483894574,"imp":26.9558993079,"elo":1004.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.5302504742,"imp":25.6828958977,"elo":1046.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":42.0685403004,"imp":23.6199228597,"elo":1119.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":79.1969033107,"imp":22.6577730828,"elo":1155.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":202.916509576,"imp":21.7810699087,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":413.0147999434,"imp":21.379021755,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":829.4495951905,"imp":21.0971850356,"elo":1212.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1248.1221717005,"imp":20.9193504251,"elo":1218.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1679.3959423676,"imp":20.8161816916,"elo":1222.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":3.6086515992,"imp":21.3352102059,"elo":1164.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":8.8659554243,"imp":19.4482935898,"elo":1250.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":21.9215058782,"imp":18.2745461206,"elo":1320.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":43.5933949927,"imp":17.7819907366,"elo":1351.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":110.831400314,"imp":17.4159585339,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":222.3578942114,"imp":17.2947203201,"elo":1383.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":444.7316933954,"imp":17.1476537668,"elo":1392.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":664.9958936779,"imp":17.0856245461,"elo":1395.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":892.485407479,"imp":17.0487150895,"elo":1396.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":35.5096105631,"imp":15.2995364449,"elo":1413.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":126.8739666394,"imp":15.6069701436,"elo":1457.1,"imputed":true,"imputed_pct":8.3333333333,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":16.2796239102,"imp":21.3271961986,"elo":1242.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":54.2164838254,"imp":20.1087189389,"elo":1265.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":158.539941202,"imp":18.1638056873,"elo":1338.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":318.0880622722,"imp":17.7534054382,"elo":1355.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":784.3400464924,"imp":17.3380648469,"elo":1373.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1619.6093744476,"imp":17.5318756753,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3239.1118632438,"imp":17.9513693335,"elo":1347.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4838.9886069116,"imp":18.2582036651,"elo":1335.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6523.5503361576,"imp":18.4320443748,"elo":1327.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":0.5832555357,"imp":25.3512478319,"elo":1148.4,"imputed":true,"imputed_pct":77.7777777778,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":0.5832555357,"imp":25.0081346397,"elo":1154.1,"imputed":true,"imputed_pct":77.7777777778,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_train":13.7518758578,"imp":24.1412586891,"elo":1130.4,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":39.0907116856,"imp":33.3927247143,"elo":942.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":114.1968017072,"imp":31.2688173325,"elo":1001.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":285.3525964278,"imp":28.7409596188,"elo":1081.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":342.0390094668,"imp":28.5493723429,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.5088535014,"imp":28.4823585394,"elo":997.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.3556051422,"imp":26.1784253129,"elo":1051.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":12.849220133,"imp":24.8863647982,"elo":1084.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":27.9677019589,"imp":24.1292532208,"elo":1112.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":70.7546625501,"imp":23.5871285044,"elo":1143.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":143.9864831887,"imp":23.4530173511,"elo":1151.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":290.4393098655,"imp":23.3662399071,"elo":1156.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":439.3706578063,"imp":23.3294367541,"elo":1158.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":590.0394439851,"imp":23.3117102997,"elo":1159.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":15.693235166,"imp":20.1301130161,"elo":1221.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":33.8496808632,"imp":17.7118157409,"elo":1321.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":89.0225754036,"imp":15.9821950128,"elo":1411.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":186.9144346811,"imp":15.4389405224,"elo":1443.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":469.4672918765,"imp":15.0651308349,"elo":1466.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":941.4479936377,"imp":14.8862792885,"elo":1478.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1881.365821255,"imp":14.7820557288,"elo":1483.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2818.5782940854,"imp":14.7162747676,"elo":1486.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3771.8540221068,"imp":14.6519246811,"elo":1487.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.7587899978,"imp":12.439646424,"elo":1550.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":20.0220773915,"imp":12.371929833,"elo":1555.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":54.3500923902,"imp":11.9877975882,"elo":1575.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":108.9879012879,"imp":11.7923368969,"elo":1593.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":284.7684460337,"imp":11.5917674062,"elo":1603.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":566.4138152119,"imp":11.4837797999,"elo":1607.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1137.2039338041,"imp":11.4081965702,"elo":1608.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1710.4116369596,"imp":11.3727585811,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2289.3887157667,"imp":11.3289097808,"elo":1606.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":16.2304597591,"imp":18.9472142631,"elo":1326.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":50.3250241527,"imp":18.1354593924,"elo":1339.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":73.7698941637,"imp":17.4467486715,"elo":1364.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":139.8184770834,"imp":16.752357,"elo":1398.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":256.4808826589,"imp":16.4237399275,"elo":1414.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":631.0078513376,"imp":15.8716515139,"elo":1434.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1275.2299790654,"imp":15.6102663216,"elo":1445.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2539.0709530625,"imp":15.4546102059,"elo":1452.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3811.2080246053,"imp":15.3912461996,"elo":1454.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":5119.7769851031,"imp":15.3674757399,"elo":1454.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":4.6911130946,"imp":15.4434639313,"elo":1451.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":39.5720422516,"imp":4.7779007279,"elo":1751.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.5770540896,"imp":27.6785297985,"elo":1044.3,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":7.1465625212,"imp":18.7121269044,"elo":1332.0,"imputed":true,"imputed_pct":27.7777777778,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":3.9320097545,"imp":11.6813562181,"elo":1581.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.9360092189,"imp":18.3692902121,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":21.7462127279,"imp":17.9810668686,"elo":1315.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":59.9879738748,"imp":17.5198563823,"elo":1343.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":127.1136519479,"imp":16.9654500109,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":318.7083632316,"imp":16.4978596105,"elo":1398.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":637.1067463953,"imp":16.206971152,"elo":1413.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1278.8911702563,"imp":16.0129198831,"elo":1424.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1925.3471149571,"imp":15.8452425385,"elo":1429.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2585.5798636409,"imp":15.7915037693,"elo":1433.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":7.0342890773,"imp":12.080628983,"elo":1609.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":9.1699299544,"imp":10.8884818594,"elo":1648.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4.5443881432,"imp":18.0143841275,"elo":1348.6,"imputed":true,"imputed_pct":8.3333333333,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":28.502561259,"imp":17.1936489026,"elo":1381.9,"imputed":true,"imputed_pct":8.3333333333,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":75.7623836717,"imp":16.3618515085,"elo":1421.2,"imputed":true,"imputed_pct":8.3333333333,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":165.7269447804,"imp":15.8352133481,"elo":1440.9,"imputed":true,"imputed_pct":8.3333333333,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":418.0474340763,"imp":15.3626163708,"elo":1461.7,"imputed":true,"imputed_pct":8.3333333333,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":850.3707294443,"imp":15.1083999224,"elo":1473.2,"imputed":true,"imputed_pct":8.3333333333,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1734.1893554164,"imp":14.9546925718,"elo":1479.4,"imputed":true,"imputed_pct":8.3333333333,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2609.9882511574,"imp":14.8920784848,"elo":1481.2,"imputed":true,"imputed_pct":8.3333333333,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3528.1289596077,"imp":14.8524093934,"elo":1480.3,"imputed":true,"imputed_pct":8.3333333333,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":471.6614585494,"imp":30.4516802237,"elo":1039.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":797.2338293006,"imp":30.3059297508,"elo":1046.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1402.6890753282,"imp":29.9928864261,"elo":1054.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2175.0878519494,"imp":29.2416756242,"elo":1072.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5045.5225760468,"imp":27.9499655318,"elo":1095.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":10226.3553477645,"imp":26.7469959259,"elo":1114.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":20731.065214537,"imp":26.1238666857,"elo":1128.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":30617.5877879932,"imp":25.8297416126,"elo":1136.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":40848.8665748301,"imp":25.6210076945,"elo":1143.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":2.7560512439,"imp":17.3387241716,"elo":1370.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.688894957,"imp":24.85647657,"elo":1065.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":32.1799419222,"imp":22.9379932385,"elo":1122.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":92.7470191468,"imp":21.351817717,"elo":1172.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":198.4037882463,"imp":20.6641766925,"elo":1201.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":483.4739388196,"imp":20.0671256697,"elo":1227.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":974.0069917832,"imp":19.6930406883,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1885.3999284296,"imp":19.4443647473,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2882.8865617337,"imp":19.3236204414,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3865.9507169384,"imp":19.2267235187,"elo":1266.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":3.2873922338,"imp":21.3470980286,"elo":1163.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":8.2975106968,"imp":19.4348824332,"elo":1244.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":21.7087779063,"imp":18.9114790437,"elo":1275.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":42.6971413089,"imp":18.614013031,"elo":1295.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":109.8273979632,"imp":18.3706238678,"elo":1310.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":219.7348179637,"imp":18.2679982633,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":439.6020314301,"imp":18.1793500454,"elo":1322.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":659.3117486382,"imp":18.1003773348,"elo":1327.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":884.4808277373,"imp":18.0344390312,"elo":1331.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":524.4765295004,"imp":25.012527985,"elo":1072.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1309.3726391156,"imp":22.0800150159,"elo":1149.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":3961.6553268098,"imp":19.97387399,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":8841.9143914963,"imp":19.2834512906,"elo":1305.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":21737.9715611002,"imp":17.5556414346,"elo":1363.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":22811.7440512431,"imp":17.5298158348,"elo":1366.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":5.4460301782,"imp":26.7385340627,"elo":1048.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":11.0404626494,"imp":24.0904697445,"elo":1109.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":27.9509507616,"imp":20.664013884,"elo":1224.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":55.9413410319,"imp":19.1202018597,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":140.091009561,"imp":18.3298649555,"elo":1325.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":280.2176062527,"imp":18.0941209717,"elo":1335.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":560.4680551302,"imp":18.0055264961,"elo":1337.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":840.8353293445,"imp":17.9925560072,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1126.9686387245,"imp":17.981598806,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8345788263,"x_train":296.99454477,"imp":19.2302574972,"elo":1318.6,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0455896074,"x_train":8.4625903355,"imp":18.3141683167,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0966097048,"x_train":20.3461935557,"imp":17.705728756,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1806528582,"x_train":49.7230862409,"imp":17.3401560779,"elo":1367.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2398344067,"x_train":92.4850850095,"imp":17.2055626238,"elo":1372.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2815893316,"x_train":216.0131399202,"imp":17.1007980535,"elo":1375.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3179145648,"x_train":428.4029173435,"imp":17.1039327772,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806002107,"x_train":862.883798881,"imp":17.1207794699,"elo":1373.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3873615123,"x_train":1291.8438396447,"imp":17.1451942817,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"x_train":1739.1690497298,"imp":17.1514607587,"elo":1370.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0413889487,"x_train":5.4899520078,"imp":21.0483818943,"elo":1217.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0801821127,"x_train":11.3341031504,"imp":20.7357573647,"elo":1233.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1457038997,"x_train":28.8927967722,"imp":20.5522681207,"elo":1244.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2046887662,"x_train":58.8720319788,"imp":20.3801393556,"elo":1251.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2874258032,"x_train":150.365699836,"imp":20.0735932675,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3447163363,"x_train":302.4787198089,"imp":19.9161165885,"elo":1270.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4006058605,"x_train":606.008218265,"imp":19.7426134538,"elo":1277.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4327969131,"x_train":908.7557510718,"imp":19.6696734269,"elo":1279.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4496610991,"x_train":1218.8724153796,"imp":19.6247954852,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0171560587,"x_train":7.3884202915,"imp":21.3548165158,"elo":1202.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333155319,"x_train":19.1397994387,"imp":21.0785374923,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.060202021,"x_train":59.6568017021,"imp":20.7193575493,"elo":1230.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0794108609,"x_train":123.907431834,"imp":20.5701164566,"elo":1238.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1075090355,"x_train":318.4600106407,"imp":20.373046132,"elo":1246.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1237830888,"x_train":623.1113495163,"imp":20.2363679583,"elo":1253.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1440522632,"x_train":1256.5403166692,"imp":20.1306653029,"elo":1257.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1469719753,"x_train":1885.0271107114,"imp":20.0926009226,"elo":1260.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1440681396,"x_train":2499.3997394449,"imp":20.0664064873,"elo":1262.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5663988088,"x_train":4.1941752843,"imp":27.6835813219,"elo":1021.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7011757792,"x_train":8.7200784389,"imp":25.9614675928,"elo":1063.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7330007013,"x_train":21.6852759737,"imp":23.5184175953,"elo":1129.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7428670659,"x_train":43.1231699977,"imp":22.6924799279,"elo":1169.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0188559546,"x_train":107.3953873718,"imp":22.0801919052,"elo":1199.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2214373513,"x_train":210.0602463393,"imp":21.9198958366,"elo":1208.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4038021731,"x_train":412.076486211,"imp":21.8159108254,"elo":1214.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4626957587,"x_train":614.2406759256,"imp":21.7556074736,"elo":1219.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5011279394,"x_train":822.5846846043,"imp":21.7451335052,"elo":1219.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6842938244,"x_train":5.483894574,"imp":26.9558993079,"elo":1004.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4234649143,"x_train":14.5302504742,"imp":25.6828958977,"elo":1046.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7086381558,"x_train":42.0685403004,"imp":23.6199228597,"elo":1119.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.9064906779,"x_train":79.1969033107,"imp":22.6577730828,"elo":1155.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.5176917438,"x_train":202.916509576,"imp":21.7810699087,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.8255235607,"x_train":413.0147999434,"imp":21.379021755,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.5799140147,"x_train":829.4495951905,"imp":21.0971850356,"elo":1212.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.9133058467,"x_train":1248.1221717005,"imp":20.9193504251,"elo":1218.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.0772277344,"x_train":1679.3959423676,"imp":20.8161816916,"elo":1222.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1695753133,"x_train":3.6086515992,"imp":21.3352102059,"elo":1164.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7271936335,"x_train":8.8659554243,"imp":19.4482935898,"elo":1250.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1652455249,"x_train":21.9215058782,"imp":18.2745461206,"elo":1320.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4098790834,"x_train":43.5933949927,"imp":17.7819907366,"elo":1351.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.8172136654,"x_train":110.831400314,"imp":17.4159585339,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.0744273861,"x_train":222.3578942114,"imp":17.2947203201,"elo":1383.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.3211352586,"x_train":444.7316933954,"imp":17.1476537668,"elo":1392.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.4252767385,"x_train":664.9958936779,"imp":17.0856245461,"elo":1395.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.5817474842,"x_train":892.485407479,"imp":17.0487150895,"elo":1396.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":6.8284826783,"x_train":35.5096105631,"imp":15.2995364449,"elo":1413.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":3.0927295997,"x_train":126.8739666394,"imp":15.6069701436,"elo":1457.1,"imputed":true,"imputed_pct":8.3333333333,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.320600689,"x_train":16.2796239102,"imp":21.3271961986,"elo":1242.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8565664695,"x_train":54.2164838254,"imp":20.1087189389,"elo":1265.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9156693463,"x_train":158.539941202,"imp":18.1638056873,"elo":1338.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7858437586,"x_train":318.0880622722,"imp":17.7534054382,"elo":1355.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.2195144755,"x_train":784.3400464924,"imp":17.3380648469,"elo":1373.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5749647915,"x_train":1619.6093744476,"imp":17.5318756753,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.0637124303,"x_train":3239.1118632438,"imp":17.9513693335,"elo":1347.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8596602241,"x_train":4838.9886069116,"imp":18.2582036651,"elo":1335.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3848981892,"x_train":6523.5503361576,"imp":18.4320443748,"elo":1327.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.1169043969,"x_train":0.5832555357,"imp":25.3512478319,"elo":1148.4,"imputed":true,"imputed_pct":77.7777777778,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.1169043969,"x_train":0.5832555357,"imp":25.0081346397,"elo":1154.1,"imputed":true,"imputed_pct":77.7777777778,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.405321849,"x_train":13.7518758578,"imp":24.1412586891,"elo":1130.4,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0406570787,"x_train":39.0907116856,"imp":33.3927247143,"elo":942.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3252722558,"x_train":114.1968017072,"imp":31.2688173325,"elo":1001.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.783304892,"x_train":285.3525964278,"imp":28.7409596188,"elo":1081.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.8948102984,"x_train":342.0390094668,"imp":28.5493723429,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0885648768,"x_train":0.5088535014,"imp":28.4823585394,"elo":997.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2741812716,"x_train":3.3556051422,"imp":26.1784253129,"elo":1051.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4818339926,"x_train":12.849220133,"imp":24.8863647982,"elo":1084.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6789925729,"x_train":27.9677019589,"imp":24.1292532208,"elo":1112.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8688854,"x_train":70.7546625501,"imp":23.5871285044,"elo":1143.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0449947563,"x_train":143.9864831887,"imp":23.4530173511,"elo":1151.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2523875096,"x_train":290.4393098655,"imp":23.3662399071,"elo":1156.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.3737687487,"x_train":439.3706578063,"imp":23.3294367541,"elo":1158.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4599601668,"x_train":590.0394439851,"imp":23.3117102997,"elo":1159.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.6888676463,"x_train":15.693235166,"imp":20.1301130161,"elo":1221.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.699336819,"x_train":33.8496808632,"imp":17.7118157409,"elo":1321.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.361104313,"x_train":89.0225754036,"imp":15.9821950128,"elo":1411.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.4140492169,"x_train":186.9144346811,"imp":15.4389405224,"elo":1443.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.9271276382,"x_train":469.4672918765,"imp":15.0651308349,"elo":1466.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.2043254803,"x_train":941.4479936377,"imp":14.8862792885,"elo":1478.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.3143880217,"x_train":1881.365821255,"imp":14.7820557288,"elo":1483.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":20.5026820042,"x_train":2818.5782940854,"imp":14.7162747676,"elo":1486.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.9024962791,"x_train":3771.8540221068,"imp":14.6519246811,"elo":1487.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6362272808,"x_train":6.7587899978,"imp":12.439646424,"elo":1550.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3637258308,"x_train":20.0220773915,"imp":12.371929833,"elo":1555.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.7767959403,"x_train":54.3500923902,"imp":11.9877975882,"elo":1575.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.0029832468,"x_train":108.9879012879,"imp":11.7923368969,"elo":1593.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7106402076,"x_train":284.7684460337,"imp":11.5917674062,"elo":1603.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.7935919179,"x_train":566.4138152119,"imp":11.4837797999,"elo":1607.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.3891270541,"x_train":1137.2039338041,"imp":11.4081965702,"elo":1608.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8718643944,"x_train":1710.4116369596,"imp":11.3727585811,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3533448181,"x_train":2289.3887157667,"imp":11.3289097808,"elo":1606.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2773508677,"x_train":16.2304597591,"imp":18.9472142631,"elo":1326.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"x_train":50.3250241527,"imp":18.1354593924,"elo":1339.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.1120162901,"x_train":73.7698941637,"imp":17.4467486715,"elo":1364.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":66.0787878177,"x_train":139.8184770834,"imp":16.752357,"elo":1398.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":83.6221921837,"x_train":256.4808826589,"imp":16.4237399275,"elo":1414.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":110.2479942847,"x_train":631.0078513376,"imp":15.8716515139,"elo":1434.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":138.475374244,"x_train":1275.2299790654,"imp":15.6102663216,"elo":1445.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":167.4276056736,"x_train":2539.0709530625,"imp":15.4546102059,"elo":1452.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":188.4316559068,"x_train":3811.2080246053,"imp":15.3912461996,"elo":1454.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":214.0824864784,"x_train":5119.7769851031,"imp":15.3674757399,"elo":1454.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1922706752,"x_train":4.6911130946,"imp":15.4434639313,"elo":1451.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":5.3578381333,"x_train":39.5720422516,"imp":4.7779007279,"elo":1751.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1496544963,"x_train":1.5770540896,"imp":27.6785297985,"elo":1044.3,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.4691971112,"x_train":7.1465625212,"imp":18.7121269044,"elo":1332.0,"imputed":true,"imputed_pct":27.7777777778,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1520731745,"x_train":3.9320097545,"imp":11.6813562181,"elo":1581.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.444937792,"x_train":9.9360092189,"imp":18.3692902121,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.8480171202,"x_train":21.7462127279,"imp":17.9810668686,"elo":1315.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.6121786403,"x_train":59.9879738748,"imp":17.5198563823,"elo":1343.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.0828680317,"x_train":127.1136519479,"imp":16.9654500109,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.6698479453,"x_train":318.7083632316,"imp":16.4978596105,"elo":1398.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.1238178806,"x_train":637.1067463953,"imp":16.206971152,"elo":1413.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.4844153286,"x_train":1278.8911702563,"imp":16.0129198831,"elo":1424.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":3.6365060902,"x_train":1925.3471149571,"imp":15.8452425385,"elo":1429.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":3.8381409686,"x_train":2585.5798636409,"imp":15.7915037693,"elo":1433.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.548288662,"x_train":7.0342890773,"imp":12.080628983,"elo":1609.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4158216639,"x_train":9.1699299544,"imp":10.8884818594,"elo":1648.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4430066712,"x_train":4.5443881432,"imp":18.0143841275,"elo":1348.6,"imputed":true,"imputed_pct":8.3333333333,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":5.2947341072,"x_train":28.502561259,"imp":17.1936489026,"elo":1381.9,"imputed":true,"imputed_pct":8.3333333333,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":11.6163487935,"x_train":75.7623836717,"imp":16.3618515085,"elo":1421.2,"imputed":true,"imputed_pct":8.3333333333,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.1847797251,"x_train":165.7269447804,"imp":15.8352133481,"elo":1440.9,"imputed":true,"imputed_pct":8.3333333333,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":25.4663625727,"x_train":418.0474340763,"imp":15.3626163708,"elo":1461.7,"imputed":true,"imputed_pct":8.3333333333,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":31.3968342236,"x_train":850.3707294443,"imp":15.1083999224,"elo":1473.2,"imputed":true,"imputed_pct":8.3333333333,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":40.4418371235,"x_train":1734.1893554164,"imp":14.9546925718,"elo":1479.4,"imputed":true,"imputed_pct":8.3333333333,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":45.6113721285,"x_train":2609.9882511574,"imp":14.8920784848,"elo":1481.2,"imputed":true,"imputed_pct":8.3333333333,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":48.572405785,"x_train":3528.1289596077,"imp":14.8524093934,"elo":1480.3,"imputed":true,"imputed_pct":8.3333333333,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.2137264423,"x_train":471.6614585494,"imp":30.4516802237,"elo":1039.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.2461770633,"x_train":797.2338293006,"imp":30.3059297508,"elo":1046.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.3473767381,"x_train":1402.6890753282,"imp":29.9928864261,"elo":1054.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":11.9018815775,"x_train":2175.0878519494,"imp":29.2416756242,"elo":1072.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.2259261458,"x_train":5045.5225760468,"imp":27.9499655318,"elo":1095.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.6020662841,"x_train":10226.3553477645,"imp":26.7469959259,"elo":1114.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.9769855841,"x_train":20731.065214537,"imp":26.1238666857,"elo":1128.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":24.3740533341,"x_train":30617.5877879932,"imp":25.8297416126,"elo":1136.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9362072534,"x_train":40848.8665748301,"imp":25.6210076945,"elo":1143.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0579126899,"x_train":2.7560512439,"imp":17.3387241716,"elo":1370.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1550906392,"x_train":13.688894957,"imp":24.85647657,"elo":1065.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3108419787,"x_train":32.1799419222,"imp":22.9379932385,"elo":1122.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6204009277,"x_train":92.7470191468,"imp":21.351817717,"elo":1172.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8859862453,"x_train":198.4037882463,"imp":20.6641766925,"elo":1201.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2789048063,"x_train":483.4739388196,"imp":20.0671256697,"elo":1227.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6522358142,"x_train":974.0069917832,"imp":19.6930406883,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8962723592,"x_train":1885.3999284296,"imp":19.4443647473,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0769295602,"x_train":2882.8865617337,"imp":19.3236204414,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1462178945,"x_train":3865.9507169384,"imp":19.2267235187,"elo":1266.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2525411892,"x_train":3.2873922338,"imp":21.3470980286,"elo":1163.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.668721793,"x_train":8.2975106968,"imp":19.4348824332,"elo":1244.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.3135984732,"x_train":21.7087779063,"imp":18.9114790437,"elo":1275.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4736641563,"x_train":42.6971413089,"imp":18.614013031,"elo":1295.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8610108955,"x_train":109.8273979632,"imp":18.3706238678,"elo":1310.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0831926002,"x_train":219.7348179637,"imp":18.2679982633,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.2439698975,"x_train":439.6020314301,"imp":18.1793500454,"elo":1322.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3416077193,"x_train":659.3117486382,"imp":18.1003773348,"elo":1327.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.4462956888,"x_train":884.4808277373,"imp":18.0344390312,"elo":1331.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":161.8134383214,"x_train":524.4765295004,"imp":25.012527985,"elo":1072.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":315.1785851675,"x_train":1309.3726391156,"imp":22.0800150159,"elo":1149.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":530.068709775,"x_train":3961.6553268098,"imp":19.97387399,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":748.7519406764,"x_train":8841.9143914963,"imp":19.2834512906,"elo":1305.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":828.5067468066,"x_train":21737.9715611002,"imp":17.5556414346,"elo":1363.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"x_train":22811.7440512431,"imp":17.5298158348,"elo":1366.8,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.0976028857,"x_train":5.4460301782,"imp":26.7385340627,"elo":1048.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.2469331487,"x_train":11.0404626494,"imp":24.0904697445,"elo":1109.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.4842749757,"x_train":27.9509507616,"imp":20.664013884,"elo":1224.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.8164722651,"x_train":55.9413410319,"imp":19.1202018597,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8569378726,"x_train":140.091009561,"imp":18.3298649555,"elo":1325.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.3361472689,"x_train":280.2176062527,"imp":18.0941209717,"elo":1335.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.5518696321,"x_train":560.4680551302,"imp":18.0055264961,"elo":1337.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.795048319,"x_train":840.8353293445,"imp":17.9925560072,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.9237006693,"x_train":1126.9686387245,"imp":17.981598806,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_all/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..f398fb53fcf33d9549085054ba58e818dae7d951
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_all/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (default)", "TabPFNv2 (tuned + ensembled)", "RealMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "Mitra (default)", "TabDPT (tuned + ensembled)", "TabDPT (tuned)", "AutoGluon 1.4 (best, 4h)", "TabM (tuned + ensembled)", "RealMLP (tuned)", "TabPFNv2 (tuned)", "LimiX (default)", "TabPFNv2 (default)", "TabDPT (default)", "TabSwift (default)", "LightGBM (tuned + ensembled)", "TabM (tuned)", "CatBoost (tuned + ensembled)", "iLTM (tuned + ensembled)", "SAP-RPT-OSS (default)", "xRFM (tuned + ensembled)", "CatBoost (tuned)", "LightGBM (tuned)", "XGBoost (tuned + ensembled)", "TabICL (default)", "xRFM (tuned)", "ModernNCA (tuned)", "CatBoost (default)", "BetaTabPFN (default)", "XGBoost (tuned)", "ModernNCA (tuned + ensembled)", "TabM (default)", "EBM (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "ModernNCA (default)", "iLTM (tuned)", "ChimeraBoost (tuned + ensembled)", "EBM (tuned)", "RealMLP (default)", "FastaiMLP (tuned + ensembled)", "ChimeraBoost (tuned)", "ExtraTrees (tuned + ensembled)", "EBM (default)", "ExtraTrees (tuned)", "TorchMLP (tuned)", "Nori-30M (default)", "FastaiMLP (tuned)", "ChimeraBoost (default)", "TabSTAR (tuned)", "Nori (default)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned + ensembled)", "OrionMSP (default)", "LightGBM (default)", "XGBoost (default)", "RandomForest (tuned)", "TorchMLP (default)", "PerpetualBooster (tuned + ensembled)", "TabFlex (default)", "TabSTAR (default)", "xRFM (default)", "PerpetualBooster (tuned)", "iLTM (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "FastaiMLP (default)", "Linear (tuned)", "KNN (tuned + ensembled)", "RandomForest (default)", "PerpetualBooster (default)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7679882645606995, 0.7679882645606995, 0.7900146842002869, 0.7628487348556519, 0.7944199442863464, 0.7738619446754456, 0.8061674237251282, 0.8575624227523804, 0.8619676828384399, 0.8634361028671265, 0.8737151026725769, 0.8839941024780273, 0.8898678421974182, 0.91629958152771, 0.8854625821113586, 0.8722466826438904, 0.8707782626152039, 0.9067547917366028, 0.8634361028671265, 0.8913362622261047, 0.8693098425865173, 0.9177680015563965, 0.9118942618370056, 0.9280470013618469, 0.9295154213905334, 0.8575624227523804, 0.919236421585083, 0.9368575811386108, 0.9177680015563965, 0.9383260011672974, 0.9486050009727478, 0.9104258418083191, 0.9251101613044739, 0.9353891611099243, 0.8935388922691345, 0.9412628412246704, 0.9456681609153748, 0.919236421585083, 0.9133626818656921, 0.9559471607208252, 0.9265785813331604, 0.9441996812820435, 0.9383260011672974, 0.9251101613044739, 0.9368575811386108, 0.9236416816711426, 0.9441996812820435, 0.9456681609153748, 0.922173261642456, 0.9397944211959839, 0.9632892608642578, 0.8590308427810669, 0.9427312612533569, 0.9588840007781982, 0.8472834229469299, 0.919236421585083, 0.8458150029182434, 0.9662261605262756, 0.9471365809440613, 0.9662261605262756, 0.9691630005836487, 0.9647576808929443, 0.9706314206123352, 0.9838472604751587, 0.9441996812820435, 0.8854625821113586, 0.9706314206123352, 0.9823788404464722, 0.9486050009727478, 0.98531574010849, 0.9530102610588074, 0.9750367403030396, 0.9588840007781982, 0.9691630005836487, 0.9838472604751587, 0.98531574010849, 0.9706314206123352, 0.9662261605262756, 0.9926578402519226], [0.23201175034046173, NaN, 0.5682819485664368, 0.5345080494880676, 0.5770925283432007, 0.607929527759552, 0.6358296871185303, 0.6651982665061951, 0.7356828451156616, 0.7415565252304077, 0.7459618449211121, 0.7283406853675842, 0.7635829448699951, 0.7856094241142273, 0.7709251046180725, 0.7856094241142273, 0.781204104423523, 0.808370053768158, 0.7679882645606995, 0.7834067344665527, 0.8179148435592651, 0.8634361028671265, 0.8399412631988525, 0.8237885236740112, 0.8546255230903625, 0.8795888423919678, 0.7665198445320129, 0.8634361028671265, 0.8634361028671265, 0.8604992628097534, 0.865638792514801, 0.880323052406311, 0.8634361028671265, 0.8678414225578308, 0.8795888423919678, 0.8039647340774536, 0.8737151026725769, 0.8825256824493408, 0.8458150029182434, 0.8619676828384399, 0.9148311018943787, 0.889133632183075, 0.9287812113761902, 0.9060205817222595, 0.8766520023345947, 0.9177680015563965, 0.8744493126869202, 0.907489001750946, 0.9265785813331604, 0.892070472240448, 0.9280470013618469, 0.93098384141922, 0.8414096832275391, 0.9140968918800354, 0.919236421585083, 0.8215858936309814, 0.8678414225578308, 0.8193832635879517, 0.933920681476593, 0.933920681476593, 0.9530102610588074, 0.9662261605262756, 0.9500734210014343, 0.9632892608642578, 0.9500734210014343, 0.9368575811386108, 0.8678414225578308, 0.9647576808929443, 0.9603524208068848, 0.9544786810874939, 0.9706314206123352, 0.9295154213905334, 0.9691630005836487, 0.9353891611099243, 0.9471365809440613, 0.9823788404464722, 0.9735682606697083, 0.9618208408355713, 0.9588840007781982, 0.9941262602806091], [0.23201175034046173, 0.43171805143356323, NaN, 0.47136563062667847, 0.5227606296539307, 0.5976505279541016, 0.5770925283432007, 0.6049926280975342, 0.6593245267868042, 0.7540381550788879, 0.6607929468154907, 0.6938325762748718, 0.7386196851730347, 0.757709264755249, 0.8017621040344238, 0.8164464235305786, 0.8002936840057373, 0.7591776847839355, 0.7606461048126221, 0.7709251046180725, 0.7488986849784851, 0.8105726838111877, 0.8428781032562256, 0.8245227336883545, 0.8472834229469299, 0.8781204223632812, 0.7327459454536438, 0.8575624227523804, 0.8524228930473328, 0.8546255230903625, 0.8722466826438904, 0.8509544730186462, 0.8450807929039001, 0.8810572624206543, 0.8685756325721741, 0.8061674237251282, 0.8751835823059082, 0.8972100019454956, 0.8604992628097534, 0.8693098425865173, 0.9118942618370056, 0.8781204223632812, 0.9133626818656921, 0.9104258418083191, 0.8781204223632812, 0.8942731022834778, 0.8781204223632812, 0.9177680015563965, 0.9089574217796326, 0.8795888423919678, 0.9001468420028687, 0.9368575811386108, 0.7767988443374634, 0.9001468420028687, 0.9207048416137695, 0.8032305240631104, 0.8193832635879517, 0.8032305240631104, 0.9324522614479065, 0.9375917911529541, 0.9530102610588074, 0.9588840007781982, 0.9368575811386108, 0.9632892608642578, 0.9559471607208252, 0.9236416816711426, 0.8487518429756165, 0.9515418410301208, 0.961086630821228, 0.9500734210014343, 0.9500734210014343, 0.9133626818656921, 0.9588840007781982, 0.9324522614479065, 0.9427312612533569, 0.9676945805549622, 0.9691630005836487, 0.9522760510444641, 0.9574155807495117, 0.9926578402519226], [0.20998531579971313, 0.46549192070961, 0.5286343693733215, NaN, 0.5345080494880676, 0.596182107925415, 0.5800293684005737, 0.6189427375793457, 0.6901615262031555, 0.7298091053962708, 0.7209985256195068, 0.7011747360229492, 0.7657856345176697, 0.757709264755249, 0.7944199442863464, 0.7958884239196777, 0.7665198445320129, 0.7511013150215149, 0.7687224745750427, 0.7621145248413086, 0.7870778441429138, 0.8164464235305786, 0.8281938433647156, 0.8076358437538147, 0.838472843170166, 0.8737151026725769, 0.7474302649497986, 0.8428781032562256, 0.8443465232849121, 0.850220263004303, 0.853157103061676, 0.877386212348938, 0.8348017334938049, 0.8575624227523804, 0.8707782626152039, 0.7988252639770508, 0.8560940027236938, 0.8634361028671265, 0.835536003112793, 0.8604992628097534, 0.8942731022834778, 0.880323052406311, 0.8957415819168091, 0.8913362622261047, 0.8634361028671265, 0.8942731022834778, 0.864904522895813, 0.9030836820602417, 0.8788546323776245, 0.8898678421974182, 0.8810572624206543, 0.9133626818656921, 0.8061674237251282, 0.8972100019454956, 0.9118942618370056, 0.8164464235305786, 0.8428781032562256, 0.8105726838111877, 0.9265785813331604, 0.9236416816711426, 0.9397944211959839, 0.9324522614479065, 0.9280470013618469, 0.9383260011672974, 0.9486050009727478, 0.9133626818656921, 0.8707782626152039, 0.9471365809440613, 0.9530102610588074, 0.9383260011672974, 0.9500734210014343, 0.9397944211959839, 0.9500734210014343, 0.9441996812820435, 0.9515418410301208, 0.9618208408355713, 0.9794420003890991, 0.9640234708786011, 0.9574155807495117, 0.9941262602806091], [0.23715125024318695, 0.4229075014591217, 0.47723934054374695, 0.46549192070961, NaN, 0.5315712094306946, 0.6358296871185303, 0.6358296871185303, 0.7180616855621338, 0.7121879458427429, 0.622613787651062, 0.6945667862892151, 0.6651982665061951, 0.6842877864837646, 0.7826725244522095, 0.7900146842002869, 0.7591776847839355, 0.8091042637825012, 0.7562408447265625, 0.7900146842002869, 0.6784141063690186, 0.7753304243087769, 0.8252570033073425, 0.8193832635879517, 0.8399412631988525, 0.8274596333503723, 0.7569750547409058, 0.8281938433647156, 0.8487518429756165, 0.8450807929039001, 0.8619676828384399, 0.7775330543518066, 0.8311306834220886, 0.8634361028671265, 0.8737151026725769, 0.826725423336029, 0.8744493126869202, 0.8546255230903625, 0.8575624227523804, 0.8634361028671265, 0.9030836820602417, 0.8590308427810669, 0.8986784219741821, 0.8964757919311523, 0.877386212348938, 0.8825256824493408, 0.877386212348938, 0.9060205817222595, 0.8575624227523804, 0.8737151026725769, 0.8604992628097534, 0.9280470013618469, 0.7709251046180725, 0.907489001750946, 0.91629958152771, 0.7958884239196777, 0.8135095238685608, 0.7958884239196777, 0.9133626818656921, 0.91629958152771, 0.9486050009727478, 0.9500734210014343, 0.9251101613044739, 0.9544786810874939, 0.9530102610588074, 0.91629958152771, 0.8370044231414795, 0.933920681476593, 0.9530102610588074, 0.9427312612533569, 0.9486050009727478, 0.9251101613044739, 0.9559471607208252, 0.9368575811386108, 0.9280470013618469, 0.9588840007781982, 0.9691630005836487, 0.9449338912963867, 0.9383260011672974, 0.9897210001945496], [0.20558002591133118, 0.392070472240448, 0.40234947204589844, 0.40381792187690735, 0.4684287905693054, NaN, 0.5234948396682739, 0.5814977884292603, 0.6299559473991394, 0.691629946231842, 0.6284875273704529, 0.6975036859512329, 0.6593245267868042, 0.676945686340332, 0.769456684589386, 0.784140944480896, 0.757709264755249, 0.7518355250358582, 0.7312775254249573, 0.727606475353241, 0.700440526008606, 0.7767988443374634, 0.8223201036453247, 0.8135095238685608, 0.8458150029182434, 0.8693098425865173, 0.7562408447265625, 0.826725423336029, 0.8472834229469299, 0.8575624227523804, 0.8751835823059082, 0.8002936840057373, 0.838472843170166, 0.8560940027236938, 0.8913362622261047, 0.7973568439483643, 0.8810572624206543, 0.8957415819168091, 0.8524228930473328, 0.850220263004303, 0.8957415819168091, 0.8810572624206543, 0.9207048416137695, 0.919236421585083, 0.8744493126869202, 0.8913362622261047, 0.864904522895813, 0.9236416816711426, 0.8928046822547913, 0.8693098425865173, 0.8876652121543884, 0.9280470013618469, 0.7753304243087769, 0.8986784219741821, 0.9353891611099243, 0.8046989440917969, 0.8149779438972473, 0.8091042637825012, 0.9236416816711426, 0.8957415819168091, 0.9574155807495117, 0.9662261605262756, 0.9397944211959839, 0.9559471607208252, 0.9486050009727478, 0.9207048416137695, 0.8428781032562256, 0.9486050009727478, 0.9456681609153748, 0.9427312612533569, 0.9530102610588074, 0.9148311018943787, 0.9515418410301208, 0.9324522614479065, 0.9456681609153748, 0.9735682606697083, 0.9676945805549622, 0.9530102610588074, 0.9530102610588074, 0.9911894202232361], [0.22613802552223206, 0.3641703426837921, 0.4229075014591217, 0.41997063159942627, 0.3641703426837921, 0.4765051305294037, NaN, 0.5491923689842224, 0.6585903167724609, 0.6622613668441772, 0.5917767882347107, 0.6519823670387268, 0.6182085275650024, 0.646842896938324, 0.7298091053962708, 0.7503671050071716, 0.7224669456481934, 0.7731277346611023, 0.7151248455047607, 0.7503671050071716, 0.6578561067581177, 0.7393538951873779, 0.7738619446754456, 0.7980910539627075, 0.7958884239196777, 0.7826725244522095, 0.7254038453102112, 0.799559473991394, 0.8017621040344238, 0.8135095238685608, 0.8281938433647156, 0.7518355250358582, 0.8135095238685608, 0.8105726838111877, 0.826725423336029, 0.8076358437538147, 0.8399412631988525, 0.8274596333503723, 0.8399412631988525, 0.8487518429756165, 0.8825256824493408, 0.835536003112793, 0.8810572624206543, 0.8568282127380371, 0.8590308427810669, 0.8707782626152039, 0.8604992628097534, 0.8737151026725769, 0.8524228930473328, 0.8604992628097534, 0.8590308427810669, 0.9104258418083191, 0.7488986849784851, 0.9001468420028687, 0.8942731022834778, 0.7679882645606995, 0.7885462641716003, 0.772393524646759, 0.8869310021400452, 0.9030836820602417, 0.9397944211959839, 0.9368575811386108, 0.9030836820602417, 0.9588840007781982, 0.9302496314048767, 0.9001468420028687, 0.8281938433647156, 0.9251101613044739, 0.9412628412246704, 0.9397944211959839, 0.9295154213905334, 0.9148311018943787, 0.9486050009727478, 0.9265785813331604, 0.91629958152771, 0.9486050009727478, 0.9691630005836487, 0.9508076310157776, 0.9412628412246704, 0.9838472604751587], [0.1938326060771942, 0.3348017632961273, 0.39500734210014343, 0.3810572624206543, 0.3641703426837921, 0.41850221157073975, 0.4508076310157776, NaN, 0.6005873680114746, 0.6255506873130798, 0.5844346284866333, 0.6475771069526672, 0.5932452082633972, 0.6262848973274231, 0.7195301055908203, 0.7342143654823303, 0.7092511057853699, 0.7209985256195068, 0.7283406853675842, 0.7371512651443481, 0.6637297868728638, 0.7430249452590942, 0.7753304243087769, 0.7782672643661499, 0.8076358437538147, 0.7826725244522095, 0.7298091053962708, 0.8046989440917969, 0.8223201036453247, 0.8237885236740112, 0.8296622633934021, 0.7650514245033264, 0.8325991034507751, 0.8164464235305786, 0.8560940027236938, 0.8039647340774536, 0.853157103061676, 0.8105726838111877, 0.8516886830329895, 0.8590308427810669, 0.8883994221687317, 0.853157103061676, 0.8876652121543884, 0.8766520023345947, 0.8795888423919678, 0.8707782626152039, 0.8928046822547913, 0.8825256824493408, 0.8546255230903625, 0.8869310021400452, 0.8663730025291443, 0.9118942618370056, 0.7400881052017212, 0.9133626818656921, 0.9104258418083191, 0.7973568439483643, 0.7430249452590942, 0.796622633934021, 0.8854625821113586, 0.9236416816711426, 0.9368575811386108, 0.9353891611099243, 0.9207048416137695, 0.9720998406410217, 0.9383260011672974, 0.9185022115707397, 0.8458150029182434, 0.933920681476593, 0.9456681609153748, 0.9500734210014343, 0.9324522614479065, 0.9265785813331604, 0.9588840007781982, 0.9324522614479065, 0.9030836820602417, 0.9486050009727478, 0.9750367403030396, 0.9647576808929443, 0.9368575811386108, 0.9823788404464722], [0.14243759214878082, 0.26431718468666077, 0.3406754732131958, 0.3098384737968445, 0.2819383144378662, 0.3700440526008606, 0.34140968322753906, 0.3994126319885254, NaN, 0.5572687387466431, 0.5212922096252441, 0.5308369994163513, 0.5345080494880676, 0.5616739988327026, 0.6563876867294312, 0.6681351065635681, 0.6453744769096375, 0.6762114763259888, 0.5814977884292603, 0.6784141063690186, 0.5800293684005737, 0.6358296871185303, 0.7026431560516357, 0.715859055519104, 0.7151248455047607, 0.7327459454536438, 0.6292217373847961, 0.757709264755249, 0.7232011556625366, 0.7371512651443481, 0.7643171548843384, 0.6806167364120483, 0.7643171548843384, 0.7709251046180725, 0.7679882645606995, 0.7547723650932312, 0.781204104423523, 0.781204104423523, 0.7533039450645447, 0.7775330543518066, 0.8002936840057373, 0.7848751544952393, 0.8237885236740112, 0.8046989440917969, 0.7973568439483643, 0.8179148435592651, 0.808370053768158, 0.8164464235305786, 0.814243733882904, 0.8252570033073425, 0.8252570033073425, 0.8487518429756165, 0.7209985256195068, 0.8274596333503723, 0.850220263004303, 0.7386196851730347, 0.7591776847839355, 0.7400881052017212, 0.8825256824493408, 0.8245227336883545, 0.9133626818656921, 0.9089574217796326, 0.9030836820602417, 0.907489001750946, 0.8972100019454956, 0.8612334728240967, 0.7973568439483643, 0.9207048416137695, 0.922173261642456, 0.9016152620315552, 0.9515418410301208, 0.9001468420028687, 0.9207048416137695, 0.9133626818656921, 0.9207048416137695, 0.9405286312103271, 0.9574155807495117, 0.9280470013618469, 0.9353891611099243, 0.9926578402519226], [0.13803230226039886, 0.2584434747695923, 0.24596181511878967, 0.27019089460372925, 0.2878120541572571, 0.30837005376815796, 0.33773863315582275, 0.37444934248924255, 0.4427312910556793, NaN, 0.47870779037475586, 0.4977973699569702, 0.47356829047203064, 0.4985315799713135, 0.577826738357544, 0.619676947593689, 0.6791483163833618, 0.5594713687896729, 0.5315712094306946, 0.5785609483718872, 0.5565345287322998, 0.5638766288757324, 0.7224669456481934, 0.6901615262031555, 0.7591776847839355, 0.7606461048126221, 0.6211453676223755, 0.7488986849784851, 0.7606461048126221, 0.7687224745750427, 0.826725423336029, 0.6299559473991394, 0.7430249452590942, 0.7312775254249573, 0.7885462641716003, 0.6975036859512329, 0.838472843170166, 0.7371512651443481, 0.7914831042289734, 0.7929515242576599, 0.7709251046180725, 0.7753304243087769, 0.8414096832275391, 0.8193832635879517, 0.8046989440917969, 0.8817914724349976, 0.8201174736022949, 0.8259912133216858, 0.8311306834220886, 0.8274596333503723, 0.8450807929039001, 0.8208516836166382, 0.7665198445320129, 0.8362702131271362, 0.8487518429756165, 0.7386196851730347, 0.7958884239196777, 0.7430249452590942, 0.9104258418083191, 0.8091042637825012, 0.922173261642456, 0.9353891611099243, 0.9353891611099243, 0.9251101613044739, 0.9353891611099243, 0.8472834229469299, 0.8046989440917969, 0.9265785813331604, 0.9427312612533569, 0.919236421585083, 0.9530102610588074, 0.8869310021400452, 0.9471365809440613, 0.8883994221687317, 0.9118942618370056, 0.9676945805549622, 0.9706314206123352, 0.9045521020889282, 0.9353891611099243, 0.9926578402519226], [0.13656388223171234, 0.2540381848812103, 0.3392070531845093, 0.27900147438049316, 0.3773861825466156, 0.3715124726295471, 0.4082232117652893, 0.4155653417110443, 0.47870779037475586, 0.5212922096252441, NaN, 0.5198237895965576, 0.5242290496826172, 0.5653451085090637, 0.5932452082633972, 0.5851688981056213, 0.592510998249054, 0.5624082088470459, 0.6358296871185303, 0.5814977884292603, 0.6850220561027527, 0.6879588961601257, 0.6578561067581177, 0.6387665271759033, 0.691629946231842, 0.7298091053962708, 0.6387665271759033, 0.700440526008606, 0.688693106174469, 0.6930983662605286, 0.7209985256195068, 0.7019089460372925, 0.7356828451156616, 0.7312775254249573, 0.7239353656768799, 0.6754772663116455, 0.7327459454536438, 0.7312775254249573, 0.7180616855621338, 0.7327459454536438, 0.7782672643661499, 0.8032305240631104, 0.8245227336883545, 0.7885462641716003, 0.7657856345176697, 0.8061674237251282, 0.8237885236740112, 0.8032305240631104, 0.8494860529899597, 0.7973568439483643, 0.8693098425865173, 0.8296622633934021, 0.7400881052017212, 0.8693098425865173, 0.8340675234794617, 0.7327459454536438, 0.7474302649497986, 0.7356828451156616, 0.8883994221687317, 0.8928046822547913, 0.8869310021400452, 0.8928046822547913, 0.8913362622261047, 0.9118942618370056, 0.9001468420028687, 0.9280470013618469, 0.8149779438972473, 0.9001468420028687, 0.9207048416137695, 0.8986784219741821, 0.9449338912963867, 0.8825256824493408, 0.9427312612533569, 0.8898678421974182, 0.9295154213905334, 0.9603524208068848, 0.9632892608642578, 0.9486050009727478, 0.9486050009727478, 0.988252580165863], [0.1262848824262619, 0.27165931463241577, 0.3061673939228058, 0.2988252639770508, 0.3054331839084625, 0.3024963140487671, 0.3480176329612732, 0.35242289304733276, 0.4691630005836487, 0.5022026300430298, 0.4801762104034424, NaN, 0.5051395297050476, 0.5124816298484802, 0.6152716875076294, 0.6284875273704529, 0.6035242080688477, 0.5682819485664368, 0.592510998249054, 0.5881057381629944, 0.5712187886238098, 0.6093979477882385, 0.6872246861457825, 0.7092511057853699, 0.685756266117096, 0.7202643156051636, 0.6182085275650024, 0.6813509464263916, 0.6850220561027527, 0.7312775254249573, 0.7364170551300049, 0.6908957362174988, 0.7173274755477905, 0.7665198445320129, 0.7679882645606995, 0.7349485754966736, 0.7474302649497986, 0.7606461048126221, 0.7503671050071716, 0.8017621040344238, 0.850220263004303, 0.8120411038398743, 0.8061674237251282, 0.8135095238685608, 0.8281938433647156, 0.7958884239196777, 0.8281938433647156, 0.8252570033073425, 0.7988252639770508, 0.835536003112793, 0.8098384737968445, 0.8942731022834778, 0.7092511057853699, 0.8663730025291443, 0.850220263004303, 0.7870778441429138, 0.7298091053962708, 0.7826725244522095, 0.8458150029182434, 0.8296622633934021, 0.9001468420028687, 0.9177680015563965, 0.8795888423919678, 0.9207048416137695, 0.8913362622261047, 0.8759177923202515, 0.8179148435592651, 0.9045521020889282, 0.9089574217796326, 0.9251101613044739, 0.9001468420028687, 0.9148311018943787, 0.919236421585083, 0.9280470013618469, 0.91629958152771, 0.9361233711242676, 0.9647576808929443, 0.9574155807495117, 0.9375917911529541, 0.988252580165863], [0.11600587517023087, 0.2364170402288437, 0.26138031482696533, 0.2342143952846527, 0.3348017632961273, 0.3406754732131958, 0.38179147243499756, 0.4067547619342804, 0.46549192070961, 0.5264317393302917, 0.4757709205150604, 0.4948605000972748, NaN, 0.6093979477882385, 0.5932452082633972, 0.580763578414917, 0.6057268977165222, 0.5719529986381531, 0.5734214186668396, 0.5682819485664368, 0.6549192070960999, 0.6306901574134827, 0.6578561067581177, 0.6262848973274231, 0.6710719466209412, 0.7386196851730347, 0.6270191073417664, 0.6982378959655762, 0.6828193664550781, 0.688693106174469, 0.7129221558570862, 0.6439059972763062, 0.718795895576477, 0.7239353656768799, 0.7327459454536438, 0.6461086869239807, 0.7400881052017212, 0.7591776847839355, 0.7070484757423401, 0.7232011556625366, 0.7555066347122192, 0.7672540545463562, 0.8208516836166382, 0.7826725244522095, 0.742290735244751, 0.7826725244522095, 0.7753304243087769, 0.7988252639770508, 0.8237885236740112, 0.757709264755249, 0.8325991034507751, 0.8091042637825012, 0.7650514245033264, 0.8164464235305786, 0.8252570033073425, 0.7209985256195068, 0.7767988443374634, 0.7224669456481934, 0.8883994221687317, 0.853157103061676, 0.8825256824493408, 0.8825256824493408, 0.9001468420028687, 0.9030836820602417, 0.8972100019454956, 0.8737151026725769, 0.8032305240631104, 0.9111600518226624, 0.9082232117652893, 0.8942731022834778, 0.9427312612533569, 0.8414096832275391, 0.9265785813331604, 0.8575624227523804, 0.9214390516281128, 0.9486050009727478, 0.9486050009727478, 0.9001468420028687, 0.9324522614479065, 0.9926578402519226], [0.11013215780258179, 0.2143906056880951, 0.24229075014591217, 0.24229075014591217, 0.31571218371391296, 0.32305434346199036, 0.3531571328639984, 0.3737151324748993, 0.43832600116729736, 0.5014684200286865, 0.43465492129325867, 0.4875183701515198, 0.3906020522117615, NaN, 0.5829662084579468, 0.5594713687896729, 0.5638766288757324, 0.5323054194450378, 0.5550661087036133, 0.5308369994163513, 0.6204111576080322, 0.596182107925415, 0.6284875273704529, 0.6123347878456116, 0.6417033672332764, 0.6828193664550781, 0.6049926280975342, 0.6563876867294312, 0.6527165770530701, 0.6637297868728638, 0.6997063159942627, 0.6218795776367188, 0.6784141063690186, 0.6960352659225464, 0.6975036859512329, 0.6270191073417664, 0.7077826857566833, 0.727606475353241, 0.691629946231842, 0.703377366065979, 0.7459618449211121, 0.7459618449211121, 0.7856094241142273, 0.7591776847839355, 0.7268722653388977, 0.7591776847839355, 0.7533039450645447, 0.7709251046180725, 0.7856094241142273, 0.7474302649497986, 0.7973568439483643, 0.781204104423523, 0.7268722653388977, 0.7973568439483643, 0.8149779438972473, 0.7114537358283997, 0.7503671050071716, 0.7092511057853699, 0.8414096832275391, 0.8575624227523804, 0.8575624227523804, 0.8443465232849121, 0.8619676828384399, 0.8898678421974182, 0.8795888423919678, 0.8678414225578308, 0.8002936840057373, 0.8854625821113586, 0.8883994221687317, 0.8751835823059082, 0.91629958152771, 0.8333333134651184, 0.91629958152771, 0.8443465232849121, 0.919236421585083, 0.9295154213905334, 0.9471365809440613, 0.8972100019454956, 0.9251101613044739, 0.9897210001945496], [0.08370044082403183, 0.2290748953819275, 0.19823788106441498, 0.20558002591133118, 0.21732746064662933, 0.230543315410614, 0.27019089460372925, 0.2804698944091797, 0.34361234307289124, 0.42217326164245605, 0.4067547619342804, 0.384728342294693, 0.4067547619342804, 0.41703376173973083, NaN, 0.5227606296539307, 0.5315712094306946, 0.4508076310157776, 0.48458150029182434, 0.47723934054374695, 0.5022026300430298, 0.5293685793876648, 0.6901615262031555, 0.5947136282920837, 0.6593245267868042, 0.6637297868728638, 0.553597629070282, 0.646842896938324, 0.6798825263977051, 0.7298091053962708, 0.7415565252304077, 0.607929527759552, 0.691629946231842, 0.6725403666496277, 0.7386196851730347, 0.6607929468154907, 0.7386196851730347, 0.6930983662605286, 0.7063142657279968, 0.7342143654823303, 0.7944199442863464, 0.7503671050071716, 0.7900146842002869, 0.7863436341285706, 0.757709264755249, 0.7621145248413086, 0.7944199442863464, 0.7958884239196777, 0.7415565252304077, 0.8002936840057373, 0.7679882645606995, 0.8325991034507751, 0.700440526008606, 0.838472843170166, 0.835536003112793, 0.7327459454536438, 0.7048457860946655, 0.7356828451156616, 0.8296622633934021, 0.7775330543518066, 0.9148311018943787, 0.9030836820602417, 0.8612334728240967, 0.907489001750946, 0.9265785813331604, 0.8516886830329895, 0.7988252639770508, 0.8854625821113586, 0.9295154213905334, 0.9001468420028687, 0.9177680015563965, 0.8832598924636841, 0.9383260011672974, 0.903817892074585, 0.8869310021400452, 0.9295154213905334, 0.9515418410301208, 0.9295154213905334, 0.9251101613044739, 0.988252580165863], [0.11453744769096375, 0.2143906056880951, 0.18355359137058258, 0.20411160588264465, 0.20998531579971313, 0.21585902571678162, 0.24963289499282837, 0.2657856047153473, 0.3318648934364319, 0.38032305240631104, 0.41483113169670105, 0.3715124726295471, 0.419236421585083, 0.44052863121032715, 0.47723934054374695, NaN, 0.4948605000972748, 0.4449339210987091, 0.46989721059799194, 0.4500734210014343, 0.5124816298484802, 0.4963289201259613, 0.6049926280975342, 0.7019089460372925, 0.619676947593689, 0.6189427375793457, 0.5756241083145142, 0.6035242080688477, 0.6417033672332764, 0.649779736995697, 0.7048457860946655, 0.5947136282920837, 0.6328927874565125, 0.6453744769096375, 0.7180616855621338, 0.676945686340332, 0.7356828451156616, 0.6740087866783142, 0.8377386331558228, 0.7371512651443481, 0.7474302649497986, 0.7180616855621338, 0.7342143654823303, 0.7621145248413086, 0.7665198445320129, 0.7907488942146301, 0.7547723650932312, 0.7958884239196777, 0.7136563658714294, 0.7790014743804932, 0.7165932655334473, 0.8201174736022949, 0.6872246861457825, 0.8091042637825012, 0.8458150029182434, 0.7283406853675842, 0.7019089460372925, 0.7312775254249573, 0.7797356843948364, 0.769456684589386, 0.8729808926582336, 0.8869310021400452, 0.8208516836166382, 0.9104258418083191, 0.8795888423919678, 0.8296622633934021, 0.7958884239196777, 0.853157103061676, 0.9001468420028687, 0.9089574217796326, 0.8707782626152039, 0.8928046822547913, 0.9148311018943787, 0.9052863717079163, 0.8766520023345947, 0.9060205817222595, 0.9603524208068848, 0.9251101613044739, 0.9133626818656921, 0.9706314206123352], [0.12775330245494843, 0.21879589557647705, 0.1997063159942627, 0.23348017036914825, 0.24082231521606445, 0.24229075014591217, 0.27753305435180664, 0.2907488942146301, 0.35462555289268494, 0.3208516836166382, 0.40748897194862366, 0.39647576212882996, 0.39427313208580017, 0.4361233413219452, 0.4684287905693054, 0.5051395297050476, NaN, 0.4596182107925415, 0.4706314206123352, 0.4948605000972748, 0.5051395297050476, 0.5007342100143433, 0.6049926280975342, 0.5976505279541016, 0.6284875273704529, 0.6116005778312683, 0.5580029487609863, 0.6049926280975342, 0.6372981071472168, 0.6505139470100403, 0.7077826857566833, 0.5770925283432007, 0.646842896938324, 0.6461086869239807, 0.6872246861457825, 0.6578561067581177, 0.7246695756912231, 0.6637297868728638, 0.7151248455047607, 0.7151248455047607, 0.7254038453102112, 0.7239353656768799, 0.7459618449211121, 0.7209985256195068, 0.7503671050071716, 0.7797356843948364, 0.7444933652877808, 0.7518355250358582, 0.7518355250358582, 0.7474302649497986, 0.769456684589386, 0.7914831042289734, 0.6930983662605286, 0.8002936840057373, 0.7870778441429138, 0.7224669456481934, 0.7136563658714294, 0.7239353656768799, 0.835536003112793, 0.7599118947982788, 0.8590308427810669, 0.8693098425865173, 0.8590308427810669, 0.8957415819168091, 0.8913362622261047, 0.8333333134651184, 0.7856094241142273, 0.8678414225578308, 0.8957415819168091, 0.8869310021400452, 0.8751835823059082, 0.8414096832275391, 0.9229074716567993, 0.8480176329612732, 0.8722466826438904, 0.91629958152771, 0.9544786810874939, 0.8795888423919678, 0.9001468420028687, 0.9794420003890991], [0.12922173738479614, 0.19162996113300323, 0.24082231521606445, 0.2488986849784851, 0.19089573621749878, 0.24816446006298065, 0.2268722504377365, 0.27900147438049316, 0.3237885534763336, 0.44052863121032715, 0.4375917911529541, 0.43171805143356323, 0.4280470013618469, 0.46769458055496216, 0.5491923689842224, 0.5550661087036133, 0.5403817892074585, NaN, 0.5080763697624207, 0.5440528392791748, 0.5066079497337341, 0.5330396294593811, 0.610866367816925, 0.6182085275650024, 0.622613787651062, 0.596182107925415, 0.5469897389411926, 0.619676947593689, 0.6549192070960999, 0.6446402072906494, 0.6600587368011475, 0.6174743175506592, 0.6328927874565125, 0.6725403666496277, 0.6710719466209412, 0.6879588961601257, 0.6651982665061951, 0.6696035265922546, 0.6842877864837646, 0.703377366065979, 0.7268722653388977, 0.7011747360229492, 0.7232011556625366, 0.7121879458427429, 0.7298091053962708, 0.7393538951873779, 0.7290748953819275, 0.7356828451156616, 0.6975036859512329, 0.7459618449211121, 0.7312775254249573, 0.7767988443374634, 0.6446402072906494, 0.7635829448699951, 0.7709251046180725, 0.7121879458427429, 0.6666666865348816, 0.7077826857566833, 0.7503671050071716, 0.7511013150215149, 0.8046989440917969, 0.7973568439483643, 0.7980910539627075, 0.8414096832275391, 0.8076358437538147, 0.8186490535736084, 0.7665198445320129, 0.8546255230903625, 0.8215858936309814, 0.8722466826438904, 0.8546255230903625, 0.8707782626152039, 0.8560940027236938, 0.8795888423919678, 0.8604992628097534, 0.8612334728240967, 0.9016152620315552, 0.9030836820602417, 0.8883994221687317, 0.9720998406410217], [0.09324523061513901, 0.23201175034046173, 0.23935389518737793, 0.23127754032611847, 0.2437591850757599, 0.2687224745750427, 0.28487518429756165, 0.27165931463241577, 0.41850221157073975, 0.4684287905693054, 0.3641703426837921, 0.40748897194862366, 0.426578551530838, 0.4449339210987091, 0.515418529510498, 0.5301027894020081, 0.5293685793876648, 0.49192363023757935, NaN, 0.5022026300430298, 0.4669603407382965, 0.49926578998565674, 0.5976505279541016, 0.5572687387466431, 0.5888399481773376, 0.6292217373847961, 0.5359765291213989, 0.5947136282920837, 0.5917767882347107, 0.619676947593689, 0.6358296871185303, 0.6064611077308655, 0.6013215780258179, 0.6725403666496277, 0.6710719466209412, 0.634361207485199, 0.6519823670387268, 0.6593245267868042, 0.6182085275650024, 0.6975036859512329, 0.6872246861457825, 0.7092511057853699, 0.6828193664550781, 0.7342143654823303, 0.7298091053962708, 0.6989721059799194, 0.6997063159942627, 0.7503671050071716, 0.7415565252304077, 0.7342143654823303, 0.7415565252304077, 0.7327459454536438, 0.7121879458427429, 0.7628487348556519, 0.7665198445320129, 0.7378854751586914, 0.7239353656768799, 0.7327459454536438, 0.8002936840057373, 0.7790014743804932, 0.8105726838111877, 0.8208516836166382, 0.8179148435592651, 0.8193832635879517, 0.8237885236740112, 0.772393524646759, 0.7834067344665527, 0.8634361028671265, 0.8325991034507751, 0.8590308427810669, 0.8370044231414795, 0.8751835823059082, 0.8590308427810669, 0.8913362622261047, 0.8619676828384399, 0.8781204223632812, 0.8869310021400452, 0.9148311018943787, 0.8942731022834778, 0.9765051603317261], [0.13656388223171234, 0.21659325063228607, 0.2290748953819275, 0.2378854602575302, 0.20998531579971313, 0.27239352464675903, 0.24963289499282837, 0.26284876465797424, 0.32158589363098145, 0.4214390516281128, 0.41850221157073975, 0.4118942618370056, 0.43171805143356323, 0.4691630005836487, 0.5227606296539307, 0.5499265789985657, 0.5051395297050476, 0.4559471309185028, 0.4977973699569702, NaN, 0.4889867901802063, 0.5110132098197937, 0.6005873680114746, 0.5947136282920837, 0.6093979477882385, 0.5859031081199646, 0.5227606296539307, 0.6211453676223755, 0.622613787651062, 0.6211453676223755, 0.6372981071472168, 0.607929527759552, 0.6505139470100403, 0.634361207485199, 0.6453744769096375, 0.6674008965492249, 0.6490455269813538, 0.619676947593689, 0.6696035265922546, 0.6872246861457825, 0.685756266117096, 0.6908957362174988, 0.6754772663116455, 0.6710719466209412, 0.7136563658714294, 0.7254038453102112, 0.7121879458427429, 0.685756266117096, 0.7209985256195068, 0.7312775254249573, 0.7342143654823303, 0.7342143654823303, 0.6563876867294312, 0.7547723650932312, 0.6960352659225464, 0.7268722653388977, 0.6740087866783142, 0.7232011556625366, 0.769456684589386, 0.7892804741859436, 0.7885462641716003, 0.7782672643661499, 0.8149779438972473, 0.8487518429756165, 0.8179148435592651, 0.865638792514801, 0.7767988443374634, 0.8560940027236938, 0.8281938433647156, 0.8472834229469299, 0.8487518429756165, 0.8964757919311523, 0.8560940027236938, 0.8972100019454956, 0.8737151026725769, 0.8759177923202515, 0.8957415819168091, 0.933920681476593, 0.9030836820602417, 0.9676945805549622], [0.10866373032331467, 0.18208517134189606, 0.2511013150215149, 0.21292217075824738, 0.32158589363098145, 0.29955947399139404, 0.3421438932418823, 0.33627018332481384, 0.41997063159942627, 0.4434655010700226, 0.3149779736995697, 0.4287812113761902, 0.34508076310157776, 0.3795888423919678, 0.4977973699569702, 0.4875183701515198, 0.4948605000972748, 0.49339208006858826, 0.5330396294593811, 0.5110132098197937, NaN, 0.5168869495391846, 0.5330396294593811, 0.5234948396682739, 0.5712187886238098, 0.596182107925415, 0.5770925283432007, 0.5903083682060242, 0.5829662084579468, 0.5697503685951233, 0.6167401075363159, 0.5602055788040161, 0.6020557880401611, 0.6328927874565125, 0.596182107925415, 0.5668135285377502, 0.6372981071472168, 0.6372981071472168, 0.6328927874565125, 0.6035242080688477, 0.6622613668441772, 0.688693106174469, 0.6945667862892151, 0.6798825263977051, 0.6277533173561096, 0.6541849970817566, 0.688693106174469, 0.6842877864837646, 0.7518355250358582, 0.6798825263977051, 0.772393524646759, 0.7121879458427429, 0.6901615262031555, 0.7430249452590942, 0.7400881052017212, 0.6710719466209412, 0.7048457860946655, 0.6725403666496277, 0.7782672643661499, 0.8105726838111877, 0.7914831042289734, 0.7650514245033264, 0.8179148435592651, 0.8296622633934021, 0.8370044231414795, 0.8810572624206543, 0.781204104423523, 0.8443465232849121, 0.8604992628097534, 0.850220263004303, 0.9060205817222595, 0.7738619446754456, 0.9001468420028687, 0.7973568439483643, 0.9016152620315552, 0.8942731022834778, 0.9089574217796326, 0.8854625821113586, 0.9236416816711426, 0.9911894202232361], [0.13069015741348267, 0.13656388223171234, 0.18942731618881226, 0.18355359137058258, 0.22466960549354553, 0.22320117056369781, 0.26064610481262207, 0.2569750249385834, 0.3641703426837921, 0.4361233413219452, 0.31204110383987427, 0.3906020522117615, 0.36930984258651733, 0.40381792187690735, 0.4706314206123352, 0.5036710500717163, 0.49926578998565674, 0.4669603407382965, 0.5007342100143433, 0.4889867901802063, 0.4831130802631378, NaN, 0.5139500498771667, 0.5491923689842224, 0.5286343693733215, 0.5638766288757324, 0.5653451085090637, 0.553597629070282, 0.5616739988327026, 0.5491923689842224, 0.5829662084579468, 0.527165949344635, 0.5991189479827881, 0.6255506873130798, 0.5976505279541016, 0.596182107925415, 0.6255506873130798, 0.610866367816925, 0.6431717872619629, 0.6372981071472168, 0.703377366065979, 0.700440526008606, 0.7092511057853699, 0.6505139470100403, 0.6725403666496277, 0.7474302649497986, 0.7709251046180725, 0.6828193664550781, 0.6975036859512329, 0.6945667862892151, 0.7151248455047607, 0.7547723650932312, 0.6784141063690186, 0.7944199442863464, 0.7635829448699951, 0.7415565252304077, 0.7136563658714294, 0.7430249452590942, 0.7503671050071716, 0.8707782626152039, 0.8215858936309814, 0.7914831042289734, 0.7944199442863464, 0.8795888423919678, 0.8663730025291443, 0.9045521020889282, 0.8120411038398743, 0.8414096832275391, 0.8810572624206543, 0.8590308427810669, 0.8575624227523804, 0.8281938433647156, 0.91629958152771, 0.8472834229469299, 0.8883994221687317, 0.9001468420028687, 0.9324522614479065, 0.9258443713188171, 0.9486050009727478, 0.9926578402519226], [0.08223201334476471, 0.16005873680114746, 0.15712188184261322, 0.17180617153644562, 0.17474302649497986, 0.1776798814535141, 0.22613802552223206, 0.22466960549354553, 0.29735681414604187, 0.27753305435180664, 0.3421438932418823, 0.3127753436565399, 0.3421438932418823, 0.3715124726295471, 0.3098384737968445, 0.39500734210014343, 0.39500734210014343, 0.38913363218307495, 0.40234947204589844, 0.3994126319885254, 0.4669603407382965, 0.48604992032051086, NaN, 0.47503671050071716, 0.5256975293159485, 0.5249632596969604, 0.4875183701515198, 0.5359765291213989, 0.5513949990272522, 0.6218795776367188, 0.6784141063690186, 0.538913369178772, 0.5947136282920837, 0.5697503685951233, 0.6894273161888123, 0.5462555289268494, 0.7048457860946655, 0.5712187886238098, 0.6556534767150879, 0.6762114763259888, 0.6167401075363159, 0.6622613668441772, 0.6607929468154907, 0.7378854751586914, 0.7151248455047607, 0.691629946231842, 0.688693106174469, 0.7679882645606995, 0.7099853157997131, 0.7856094241142273, 0.7136563658714294, 0.7063142657279968, 0.6989721059799194, 0.7606461048126221, 0.8487518429756165, 0.7129221558570862, 0.700440526008606, 0.7136563658714294, 0.8149779438972473, 0.7665198445320129, 0.9515418410301208, 0.8986784219741821, 0.8604992628097534, 0.8281938433647156, 0.9001468420028687, 0.8032305240631104, 0.7944199442863464, 0.8179148435592651, 0.9177680015563965, 0.9060205817222595, 0.9030836820602417, 0.8604992628097534, 0.8766520023345947, 0.8766520023345947, 0.8810572624206543, 0.9383260011672974, 0.9515418410301208, 0.9236416816711426, 0.9214390516281128, 0.9838472604751587], [0.08810572326183319, 0.17621144652366638, 0.17547723650932312, 0.1923641711473465, 0.18061673641204834, 0.18649046123027802, 0.20190896093845367, 0.2217327505350113, 0.2841409742832184, 0.3098384737968445, 0.3612334728240967, 0.2907488942146301, 0.3737151324748993, 0.3876652121543884, 0.40528634190559387, 0.2980910539627075, 0.40234947204589844, 0.38179147243499756, 0.4427312910556793, 0.40528634190559387, 0.4765051305294037, 0.4508076310157776, 0.5249632596969604, NaN, 0.5521292090415955, 0.5609397888183594, 0.4948605000972748, 0.5345080494880676, 0.5682819485664368, 0.5565345287322998, 0.6042584180831909, 0.541850209236145, 0.5726872086524963, 0.596182107925415, 0.6372981071472168, 0.6005873680114746, 0.6240822076797485, 0.619676947593689, 0.6872246861457825, 0.6446402072906494, 0.6930983662605286, 0.6666666865348816, 0.6563876867294312, 0.700440526008606, 0.6813509464263916, 0.7298091053962708, 0.6953010559082031, 0.6930983662605286, 0.6651982665061951, 0.7209985256195068, 0.6740087866783142, 0.7584434747695923, 0.6622613668441772, 0.7679882645606995, 0.7687224745750427, 0.7077826857566833, 0.6725403666496277, 0.7129221558570862, 0.7444933652877808, 0.7327459454536438, 0.7944199442863464, 0.8281938433647156, 0.7650514245033264, 0.8883994221687317, 0.8546255230903625, 0.8032305240631104, 0.772393524646759, 0.8061674237251282, 0.864904522895813, 0.8942731022834778, 0.8164464235305786, 0.8722466826438904, 0.864904522895813, 0.8810572624206543, 0.8487518429756165, 0.8707782626152039, 0.9383260011672974, 0.9104258418083191, 0.8898678421974182, 0.9618208408355713], [0.07195301353931427, 0.14537444710731506, 0.15271659195423126, 0.16152717173099518, 0.16005873680114746, 0.15418502688407898, 0.20411160588264465, 0.1923641711473465, 0.28487518429756165, 0.24082231521606445, 0.30837005376815796, 0.31424376368522644, 0.32892805337905884, 0.35829663276672363, 0.3406754732131958, 0.38032305240631104, 0.3715124726295471, 0.3773861825466156, 0.41116005182266235, 0.3906020522117615, 0.4287812113761902, 0.47136563062667847, 0.4743025004863739, 0.44787079095840454, NaN, 0.5080763697624207, 0.5139500498771667, 0.5051395297050476, 0.5910425782203674, 0.5323054194450378, 0.6145374178886414, 0.5374449491500854, 0.5859031081199646, 0.5682819485664368, 0.6446402072906494, 0.5609397888183594, 0.6365638971328735, 0.5947136282920837, 0.6005873680114746, 0.6182085275650024, 0.6284875273704529, 0.6446402072906494, 0.6842877864837646, 0.685756266117096, 0.6718061566352844, 0.6710719466209412, 0.6490455269813538, 0.7077826857566833, 0.7136563658714294, 0.7151248455047607, 0.7408223152160645, 0.7121879458427429, 0.6960352659225464, 0.7180616855621338, 0.7650514245033264, 0.7107195258140564, 0.688693106174469, 0.7063142657279968, 0.826725423336029, 0.7312775254249573, 0.8781204223632812, 0.8898678421974182, 0.8700440526008606, 0.835536003112793, 0.8546255230903625, 0.7665198445320129, 0.7738619446754456, 0.8414096832275391, 0.8737151026725769, 0.8729808926582336, 0.8839941024780273, 0.8325991034507751, 0.8560940027236938, 0.8458150029182434, 0.8810572624206543, 0.9324522614479065, 0.919236421585083, 0.8751835823059082, 0.9133626818656921, 0.98531574010849], [0.07048457860946655, 0.12041115760803223, 0.12187959253787994, 0.1262848824262619, 0.17254038155078888, 0.13069015741348267, 0.21732746064662933, 0.21732746064662933, 0.2672540247440338, 0.23935389518737793, 0.27019089460372925, 0.2797356843948364, 0.26138031482696533, 0.3171806037425995, 0.33627018332481384, 0.3810572624206543, 0.3883994221687317, 0.40381792187690735, 0.37077826261520386, 0.4140969216823578, 0.40381792187690735, 0.4361233413219452, 0.47503671050071716, 0.4390602111816406, 0.49192363023757935, NaN, 0.4875183701515198, 0.5337738394737244, 0.5227606296539307, 0.538913369178772, 0.5734214186668396, 0.48237884044647217, 0.5954478979110718, 0.5873715281486511, 0.5726872086524963, 0.5462555289268494, 0.607929527759552, 0.5844346284866333, 0.527165949344635, 0.6035242080688477, 0.622613787651062, 0.6798825263977051, 0.7334801554679871, 0.6828193664550781, 0.6475771069526672, 0.685756266117096, 0.688693106174469, 0.6901615262031555, 0.7041115760803223, 0.6798825263977051, 0.7312775254249573, 0.703377366065979, 0.691629946231842, 0.7496328949928284, 0.7378854751586914, 0.6681351065635681, 0.700440526008606, 0.6688693165779114, 0.8105726838111877, 0.7261380553245544, 0.8619676828384399, 0.8472834229469299, 0.8428781032562256, 0.8443465232849121, 0.8516886830329895, 0.8237885236740112, 0.769456684589386, 0.8781204223632812, 0.8795888423919678, 0.8619676828384399, 0.8986784219741821, 0.7797356843948364, 0.8781204223632812, 0.8091042637825012, 0.8560940027236938, 0.91629958152771, 0.919236421585083, 0.8560940027236938, 0.892070472240448, 0.9809104204177856], [0.14243759214878082, 0.23348017036914825, 0.2672540247440338, 0.2525697648525238, 0.24302496016025543, 0.2437591850757599, 0.2745961844921112, 0.27019089460372925, 0.37077826261520386, 0.3788546323776245, 0.3612334728240967, 0.38179147243499756, 0.37298092246055603, 0.39500734210014343, 0.44640234112739563, 0.4243759214878082, 0.44199705123901367, 0.45301029086112976, 0.46402350068092346, 0.47723934054374695, 0.4229075014591217, 0.43465492129325867, 0.5124816298484802, 0.5051395297050476, 0.48604992032051086, 0.5124816298484802, NaN, 0.5345080494880676, 0.4977973699569702, 0.5403817892074585, 0.5624082088470459, 0.553597629070282, 0.553597629070282, 0.5741556286811829, 0.553597629070282, 0.5873715281486511, 0.5697503685951233, 0.5903083682060242, 0.5550661087036133, 0.6005873680114746, 0.6651982665061951, 0.634361207485199, 0.6013215780258179, 0.5976505279541016, 0.634361207485199, 0.6725403666496277, 0.6975036859512329, 0.6167401075363159, 0.6681351065635681, 0.6629955768585205, 0.6681351065635681, 0.691629946231842, 0.6064611077308655, 0.7327459454536438, 0.6681351065635681, 0.6387665271759033, 0.6123347878456116, 0.6299559473991394, 0.7452276349067688, 0.6901615262031555, 0.7312775254249573, 0.7283406853675842, 0.7540381550788879, 0.7856094241142273, 0.7856094241142273, 0.772393524646759, 0.7606461048126221, 0.7900146842002869, 0.7922173142433167, 0.7900146842002869, 0.7944199442863464, 0.8149779438972473, 0.8237885236740112, 0.8179148435592651, 0.7767988443374634, 0.8208516836166382, 0.8458150029182434, 0.838472843170166, 0.8663730025291443, 0.9823788404464722], [0.08076358586549759, 0.13656388223171234, 0.14243759214878082, 0.15712188184261322, 0.17180617153644562, 0.17327459156513214, 0.20044052600860596, 0.19530102610588074, 0.24229075014591217, 0.2511013150215149, 0.29955947399139404, 0.3186490535736084, 0.30176210403442383, 0.34361234307289124, 0.3531571328639984, 0.39647576212882996, 0.39500734210014343, 0.38032305240631104, 0.40528634190559387, 0.3788546323776245, 0.40969163179397583, 0.44640234112739563, 0.46402350068092346, 0.46549192070961, 0.4948605000972748, 0.46622613072395325, 0.46549192070961, NaN, 0.5190895795822144, 0.5220264196395874, 0.5484581589698792, 0.5212922096252441, 0.6145374178886414, 0.5337738394737244, 0.5866373181343079, 0.5521292090415955, 0.5770925283432007, 0.5682819485664368, 0.5462555289268494, 0.6020557880401611, 0.5903083682060242, 0.619676947593689, 0.6439059972763062, 0.6740087866783142, 0.6549192070960999, 0.6182085275650024, 0.6321585774421692, 0.688693106174469, 0.6593245267868042, 0.6784141063690186, 0.7063142657279968, 0.6681351065635681, 0.7151248455047607, 0.7019089460372925, 0.7121879458427429, 0.6791483163833618, 0.7180616855621338, 0.676945686340332, 0.7679882645606995, 0.7635829448699951, 0.7870778441429138, 0.8061674237251282, 0.8046989440917969, 0.8487518429756165, 0.8120411038398743, 0.8325991034507751, 0.7518355250358582, 0.9397944211959839, 0.8392070531845093, 0.8325991034507751, 0.9251101613044739, 0.8443465232849121, 0.9045521020889282, 0.8560940027236938, 0.892070472240448, 0.9456681609153748, 0.919236421585083, 0.8847283124923706, 0.919236421585083, 0.9867841601371765], [0.06314244121313095, 0.13656388223171234, 0.14757709205150604, 0.1556534469127655, 0.15124815702438354, 0.15271659195423126, 0.19823788106441498, 0.1776798814535141, 0.276798814535141, 0.23935389518737793, 0.311306893825531, 0.3149779736995697, 0.3171806037425995, 0.34728339314460754, 0.3201174736022949, 0.35829663276672363, 0.3627019226551056, 0.34508076310157776, 0.4082232117652893, 0.3773861825466156, 0.41703376173973083, 0.43832600116729736, 0.4486050009727478, 0.43171805143356323, 0.40895742177963257, 0.47723934054374695, 0.5022026300430298, 0.48091042041778564, NaN, 0.5110132098197937, 0.5917767882347107, 0.5212922096252441, 0.5653451085090637, 0.5491923689842224, 0.6116005778312683, 0.5323054194450378, 0.6240822076797485, 0.5756241083145142, 0.5822319984436035, 0.6093979477882385, 0.622613787651062, 0.6270191073417664, 0.6490455269813538, 0.6365638971328735, 0.6563876867294312, 0.6549192070960999, 0.6328927874565125, 0.6776798963546753, 0.6945667862892151, 0.7107195258140564, 0.7063142657279968, 0.6982378959655762, 0.6798825263977051, 0.7085168957710266, 0.7533039450645447, 0.7121879458427429, 0.6828193664550781, 0.7107195258140564, 0.8105726838111877, 0.7268722653388977, 0.8516886830329895, 0.8854625821113586, 0.8487518429756165, 0.8179148435592651, 0.8428781032562256, 0.757709264755249, 0.7753304243087769, 0.8296622633934021, 0.864904522895813, 0.8604992628097534, 0.8693098425865173, 0.8281938433647156, 0.8458150029182434, 0.8392070531845093, 0.8707782626152039, 0.922173261642456, 0.9236416816711426, 0.8693098425865173, 0.91629958152771, 0.9838472604751587], [0.08223201334476471, 0.13950073719024658, 0.14537444710731506, 0.14977973699569702, 0.15491923689842224, 0.14243759214878082, 0.18649046123027802, 0.17621144652366638, 0.26284876465797424, 0.23127754032611847, 0.30690160393714905, 0.2687224745750427, 0.311306893825531, 0.33627018332481384, 0.27019089460372925, 0.350220263004303, 0.3494860529899597, 0.3553597629070282, 0.38032305240631104, 0.3788546323776245, 0.4302496314048767, 0.4508076310157776, 0.37812042236328125, 0.4434655010700226, 0.46769458055496216, 0.461086630821228, 0.4596182107925415, 0.4779735803604126, 0.4889867901802063, NaN, 0.5602055788040161, 0.5183553695678711, 0.5359765291213989, 0.5183553695678711, 0.6292217373847961, 0.5403817892074585, 0.5932452082633972, 0.5286343693733215, 0.6064611077308655, 0.6402349472045898, 0.5976505279541016, 0.6240822076797485, 0.6020557880401611, 0.6563876867294312, 0.6930983662605286, 0.664463996887207, 0.649779736995697, 0.6828193664550781, 0.6718061566352844, 0.7312775254249573, 0.6894273161888123, 0.6791483163833618, 0.676945686340332, 0.7180616855621338, 0.7856094241142273, 0.700440526008606, 0.6813509464263916, 0.7019089460372925, 0.7606461048126221, 0.7503671050071716, 0.903817892074585, 0.8671072125434875, 0.8149779438972473, 0.8076358437538147, 0.8612334728240967, 0.7870778441429138, 0.7767988443374634, 0.811306893825531, 0.8825256824493408, 0.8935388922691345, 0.8737151026725769, 0.8516886830329895, 0.8663730025291443, 0.8685756325721741, 0.8634361028671265, 0.9118942618370056, 0.9456681609153748, 0.9118942618370056, 0.9118942618370056, 0.9779735803604126], [0.06167401000857353, 0.13436123728752136, 0.12775330245494843, 0.14684288203716278, 0.13803230226039886, 0.12481644749641418, 0.17180617153644562, 0.1703377366065979, 0.23568281531333923, 0.17327459156513214, 0.27900147438049316, 0.2635829746723175, 0.28707781434059143, 0.3002936840057373, 0.2584434747695923, 0.2951541841030121, 0.29221731424331665, 0.33994126319885254, 0.3641703426837921, 0.3627019226551056, 0.38325992226600647, 0.41703376173973083, 0.32158589363098145, 0.3957415521144867, 0.38546255230903625, 0.426578551530838, 0.4375917911529541, 0.45154184103012085, 0.4082232117652893, 0.4397944211959839, NaN, 0.4948605000972748, 0.5345080494880676, 0.5095447897911072, 0.5484581589698792, 0.5051395297050476, 0.5748898386955261, 0.49926578998565674, 0.5491923689842224, 0.5756241083145142, 0.5748898386955261, 0.6064611077308655, 0.5991189479827881, 0.6167401075363159, 0.6358296871185303, 0.6240822076797485, 0.6284875273704529, 0.6519823670387268, 0.6387665271759033, 0.6828193664550781, 0.6828193664550781, 0.6490455269813538, 0.6842877864837646, 0.6813509464263916, 0.7129221558570862, 0.6872246861457825, 0.688693106174469, 0.6828193664550781, 0.7951542139053345, 0.7232011556625366, 0.8986784219741821, 0.9030836820602417, 0.823054313659668, 0.7885462641716003, 0.8179148435592651, 0.7657856345176697, 0.7503671050071716, 0.8105726838111877, 0.8406754732131958, 0.8296622633934021, 0.8560940027236938, 0.8208516836166382, 0.8281938433647156, 0.8311306834220886, 0.8414096832275391, 0.922173261642456, 0.907489001750946, 0.8678414225578308, 0.8979442119598389, 0.9750367403030396], [0.051395006477832794, 0.11967694759368896, 0.14904552698135376, 0.1226138025522232, 0.22246696054935455, 0.1997063159942627, 0.24816446006298065, 0.23494860529899597, 0.31938326358795166, 0.3700440526008606, 0.2980910539627075, 0.3091042637825012, 0.35609397292137146, 0.37812042236328125, 0.392070472240448, 0.40528634190559387, 0.4229075014591217, 0.3825257122516632, 0.3935389220714569, 0.392070472240448, 0.4397944211959839, 0.4728340804576874, 0.461086630821228, 0.458149790763855, 0.46255508065223694, 0.5176211595535278, 0.44640234112739563, 0.47870779037475586, 0.47870779037475586, 0.4816446304321289, 0.5051395297050476, NaN, 0.48604992032051086, 0.5183553695678711, 0.5374449491500854, 0.5190895795822144, 0.5212922096252441, 0.5403817892074585, 0.541850209236145, 0.5374449491500854, 0.5506607890129089, 0.5352423191070557, 0.5873715281486511, 0.5991189479827881, 0.5712187886238098, 0.5814977884292603, 0.5844346284866333, 0.6255506873130798, 0.6020557880401611, 0.6064611077308655, 0.6130689978599548, 0.6387665271759033, 0.6975036859512329, 0.6270191073417664, 0.6431717872619629, 0.6549192070960999, 0.6960352659225464, 0.6505139470100403, 0.6284875273704529, 0.7173274755477905, 0.6982378959655762, 0.6828193664550781, 0.7048457860946655, 0.7165932655334473, 0.7533039450645447, 0.7393538951873779, 0.7298091053962708, 0.7562408447265625, 0.7679882645606995, 0.826725423336029, 0.7753304243087769, 0.835536003112793, 0.811306893825531, 0.850220263004303, 0.8678414225578308, 0.8245227336883545, 0.8472834229469299, 0.9229074716567993, 0.9060205817222595, 0.9676945805549622], [0.08957415819168091, 0.13656388223171234, 0.15491923689842224, 0.16519823670387268, 0.16886931657791138, 0.16152717173099518, 0.18649046123027802, 0.16740088164806366, 0.23568281531333923, 0.2569750249385834, 0.26431718468666077, 0.28267255425453186, 0.28120410442352295, 0.32158589363098145, 0.30837005376815796, 0.36710718274116516, 0.3531571328639984, 0.36710718274116516, 0.39867842197418213, 0.3494860529899597, 0.39794421195983887, 0.4008810520172119, 0.40528634190559387, 0.4273127615451813, 0.4140969216823578, 0.4045521318912506, 0.44640234112739563, 0.38546255230903625, 0.43465492129325867, 0.46402350068092346, 0.46549192070961, 0.5139500498771667, NaN, 0.47723934054374695, 0.4904552102088928, 0.538913369178772, 0.49559471011161804, 0.4743025004863739, 0.5095447897911072, 0.5675477385520935, 0.5741556286811829, 0.596182107925415, 0.5873715281486511, 0.6013215780258179, 0.6123347878456116, 0.5756241083145142, 0.6013215780258179, 0.6475771069526672, 0.6240822076797485, 0.6446402072906494, 0.6651982665061951, 0.6372981071472168, 0.6637297868728638, 0.6710719466209412, 0.688693106174469, 0.6828193664550781, 0.6622613668441772, 0.6842877864837646, 0.6960352659225464, 0.7474302649497986, 0.7268722653388977, 0.7209985256195068, 0.7386196851730347, 0.8414096832275391, 0.7731277346611023, 0.8399412631988525, 0.7503671050071716, 0.8898678421974182, 0.8157122135162354, 0.8458150029182434, 0.853157103061676, 0.8414096832275391, 0.8810572624206543, 0.8575624227523804, 0.8751835823059082, 0.8957415819168091, 0.9045521020889282, 0.9016152620315552, 0.9133626818656921, 0.9838472604751587], [0.07488986849784851, 0.13215859234333038, 0.1189427301287651, 0.14243759214878082, 0.13656388223171234, 0.14390602707862854, 0.18942731618881226, 0.18355359137058258, 0.2290748953819275, 0.2687224745750427, 0.2687224745750427, 0.23348017036914825, 0.27606460452079773, 0.303964763879776, 0.3274596035480499, 0.35462555289268494, 0.3538913428783417, 0.3274596035480499, 0.3274596035480499, 0.36563876271247864, 0.36710718274116516, 0.37444934248924255, 0.4302496314048767, 0.40381792187690735, 0.43171805143356323, 0.41262850165367126, 0.42584434151649475, 0.46622613072395325, 0.4508076310157776, 0.4816446304321289, 0.4904552102088928, 0.4816446304321289, 0.5227606296539307, NaN, 0.5256975293159485, 0.5293685793876648, 0.5073421597480774, 0.44419971108436584, 0.5080763697624207, 0.5741556286811829, 0.5565345287322998, 0.6534507870674133, 0.5550661087036133, 0.6182085275650024, 0.6123347878456116, 0.6270191073417664, 0.6424375772476196, 0.6233479976654053, 0.6123347878456116, 0.634361207485199, 0.6358296871185303, 0.676945686340332, 0.6402349472045898, 0.7063142657279968, 0.6725403666496277, 0.6725403666496277, 0.6417033672332764, 0.676945686340332, 0.7327459454536438, 0.7195301055908203, 0.7136563658714294, 0.7444933652877808, 0.7518355250358582, 0.7870778441429138, 0.7679882645606995, 0.7973568439483643, 0.7459618449211121, 0.8428781032562256, 0.8017621040344238, 0.850220263004303, 0.8281938433647156, 0.8046989440917969, 0.8494860529899597, 0.826725423336029, 0.8414096832275391, 0.8575624227523804, 0.8751835823059082, 0.8678414225578308, 0.8869310021400452, 0.9735682606697083], [0.06461086869239807, 0.12041115760803223, 0.13142438232898712, 0.12922173738479614, 0.1262848824262619, 0.10866373032331467, 0.17327459156513214, 0.14390602707862854, 0.23201175034046173, 0.21145375072956085, 0.27606460452079773, 0.23201175034046173, 0.2672540247440338, 0.3024963140487671, 0.26138031482696533, 0.2819383144378662, 0.3127753436565399, 0.32892805337905884, 0.32892805337905884, 0.35462555289268494, 0.40381792187690735, 0.40234947204589844, 0.31057268381118774, 0.3627019226551056, 0.3553597629070282, 0.4273127615451813, 0.44640234112739563, 0.4133627116680145, 0.3883994221687317, 0.37077826261520386, 0.45154184103012085, 0.46255508065223694, 0.5095447897911072, 0.4743025004863739, NaN, 0.46549192070961, 0.48091042041778564, 0.48604992032051086, 0.5330396294593811, 0.5800293684005737, 0.5609397888183594, 0.5447870492935181, 0.5873715281486511, 0.6321585774421692, 0.6461086869239807, 0.6270191073417664, 0.6211453676223755, 0.6637297868728638, 0.6490455269813538, 0.6945667862892151, 0.6666666865348816, 0.6505139470100403, 0.6534507870674133, 0.6828193664550781, 0.7709251046180725, 0.7011747360229492, 0.6563876867294312, 0.7019089460372925, 0.7621145248413086, 0.7019089460372925, 0.8046989440917969, 0.781204104423523, 0.7973568439483643, 0.7914831042289734, 0.826725423336029, 0.7621145248413086, 0.7782672643661499, 0.7679882645606995, 0.8619676828384399, 0.8986784219741821, 0.8443465232849121, 0.8091042637825012, 0.8575624227523804, 0.826725423336029, 0.8487518429756165, 0.8839941024780273, 0.9133626818656921, 0.8869310021400452, 0.919236421585083, 0.9809104204177856], [0.10646108537912369, 0.196035236120224, 0.1938326060771942, 0.20117473602294922, 0.17327459156513214, 0.20264317095279694, 0.1923641711473465, 0.196035236120224, 0.2452276051044464, 0.3024963140487671, 0.3245227634906769, 0.26505139470100403, 0.3538913428783417, 0.37298092246055603, 0.3392070531845093, 0.32305434346199036, 0.3421438932418823, 0.31204110383987427, 0.36563876271247864, 0.33259913325309753, 0.43318650126457214, 0.40381792187690735, 0.453744500875473, 0.3994126319885254, 0.4390602111816406, 0.453744500875473, 0.41262850165367126, 0.44787079095840454, 0.46769458055496216, 0.4596182107925415, 0.4948605000972748, 0.48091042041778564, 0.461086630821228, 0.4706314206123352, 0.5345080494880676, NaN, 0.5110132098197937, 0.46989721059799194, 0.5433186292648315, 0.5991189479827881, 0.5653451085090637, 0.5110132098197937, 0.5168869495391846, 0.5748898386955261, 0.6387665271759033, 0.5903083682060242, 0.5646108388900757, 0.6005873680114746, 0.5491923689842224, 0.6696035265922546, 0.5653451085090637, 0.6328927874565125, 0.6534507870674133, 0.6182085275650024, 0.6446402072906494, 0.685756266117096, 0.6519823670387268, 0.6828193664550781, 0.5932452082633972, 0.661527156829834, 0.6439059972763062, 0.6490455269813538, 0.6519823670387268, 0.7180616855621338, 0.7254038453102112, 0.7209985256195068, 0.7613803148269653, 0.6975036859512329, 0.7488986849784851, 0.8663730025291443, 0.7121879458427429, 0.8957415819168091, 0.7929515242576599, 0.9104258418083191, 0.8487518429756165, 0.7804698944091797, 0.8325991034507751, 0.9441996812820435, 0.9207048416137695, 0.9632892608642578], [0.05873715132474899, 0.1262848824262619, 0.12481644749641418, 0.14390602707862854, 0.12555065751075745, 0.1189427301287651, 0.16005873680114746, 0.14684288203716278, 0.21879589557647705, 0.16152717173099518, 0.2672540247440338, 0.2525697648525238, 0.2599118947982788, 0.29221731424331665, 0.26138031482696533, 0.26431718468666077, 0.27533039450645447, 0.3348017632961273, 0.3480176329612732, 0.35095447301864624, 0.3627019226551056, 0.37444934248924255, 0.2951541841030121, 0.3759177625179291, 0.36343613266944885, 0.392070472240448, 0.4302496314048767, 0.4229075014591217, 0.3759177625179291, 0.4067547619342804, 0.4251101315021515, 0.47870779037475586, 0.5044052600860596, 0.492657870054245, 0.5190895795822144, 0.4889867901802063, NaN, 0.4948605000972748, 0.515418529510498, 0.5565345287322998, 0.5609397888183594, 0.5726872086524963, 0.5712187886238098, 0.5675477385520935, 0.6138032078742981, 0.5947136282920837, 0.6049926280975342, 0.6123347878456116, 0.635095477104187, 0.6534507870674133, 0.6696035265922546, 0.6284875273704529, 0.6696035265922546, 0.676945686340332, 0.6725403666496277, 0.6710719466209412, 0.6725403666496277, 0.673274576663971, 0.7650514245033264, 0.7019089460372925, 0.8590308427810669, 0.8678414225578308, 0.8032305240631104, 0.7738619446754456, 0.7944199442863464, 0.7518355250358582, 0.7474302649497986, 0.8061674237251282, 0.8237885236740112, 0.826725423336029, 0.853157103061676, 0.8105726838111877, 0.8223201036453247, 0.8296622633934021, 0.8377386331558228, 0.9045521020889282, 0.9060205817222595, 0.8619676828384399, 0.8928046822547913, 0.9720998406410217], [0.05433186516165733, 0.11747430264949799, 0.10279001295566559, 0.13656388223171234, 0.14537444710731506, 0.10425844043493271, 0.17254038155078888, 0.18942731618881226, 0.21879589557647705, 0.26284876465797424, 0.2687224745750427, 0.23935389518737793, 0.24082231521606445, 0.27239352464675903, 0.30690160393714905, 0.3259911835193634, 0.33627018332481384, 0.33039647340774536, 0.3406754732131958, 0.38032305240631104, 0.3627019226551056, 0.38913363218307495, 0.4287812113761902, 0.38032305240631104, 0.40528634190559387, 0.4155653417110443, 0.40969163179397583, 0.43171805143356323, 0.4243759214878082, 0.47136563062667847, 0.5007342100143433, 0.4596182107925415, 0.5256975293159485, 0.5558003187179565, 0.5139500498771667, 0.5301027894020081, 0.5051395297050476, NaN, 0.4757709205150604, 0.553597629070282, 0.5301027894020081, 0.6593245267868042, 0.5462555289268494, 0.5682819485664368, 0.5969163179397583, 0.6211453676223755, 0.6005873680114746, 0.6005873680114746, 0.6093979477882385, 0.6093979477882385, 0.6461086869239807, 0.610866367816925, 0.6035242080688477, 0.6622613668441772, 0.6123347878456116, 0.6255506873130798, 0.6093979477882385, 0.622613787651062, 0.7048457860946655, 0.6622613668441772, 0.7048457860946655, 0.7621145248413086, 0.7195301055908203, 0.7474302649497986, 0.7415565252304077, 0.7386196851730347, 0.685756266117096, 0.8318648934364319, 0.7518355250358582, 0.7767988443374634, 0.8149779438972473, 0.7342143654823303, 0.7900146842002869, 0.7459618449211121, 0.7900146842002869, 0.826725423336029, 0.8296622633934021, 0.7797356843948364, 0.8223201036453247, 0.9500734210014343], [0.08076358586549759, 0.15418502688407898, 0.13950073719024658, 0.16446402668952942, 0.14243759214878082, 0.14757709205150604, 0.16005873680114746, 0.1483113020658493, 0.24669604003429413, 0.20851688086986542, 0.2819383144378662, 0.24963289499282837, 0.2929515540599823, 0.30837005376815796, 0.29368576407432556, 0.16226138174533844, 0.28487518429756165, 0.31571218371391296, 0.38179147243499756, 0.33039647340774536, 0.36710718274116516, 0.3568281829357147, 0.3443465530872345, 0.3127753436565399, 0.3994126319885254, 0.4728340804576874, 0.4449339210987091, 0.453744500875473, 0.4177680015563965, 0.3935389220714569, 0.4508076310157776, 0.458149790763855, 0.4904552102088928, 0.49192363023757935, 0.4669603407382965, 0.45668134093284607, 0.48458150029182434, 0.5242290496826172, NaN, 0.4963289201259613, 0.6005873680114746, 0.5682819485664368, 0.566079318523407, 0.5212922096252441, 0.5609397888183594, 0.596182107925415, 0.6182085275650024, 0.5462555289268494, 0.6152716875076294, 0.5895741581916809, 0.6358296871185303, 0.6813509464263916, 0.622613787651062, 0.6989721059799194, 0.6123347878456116, 0.6960352659225464, 0.6270191073417664, 0.6989721059799194, 0.6872246861457825, 0.6607929468154907, 0.6622613668441772, 0.700440526008606, 0.6989721059799194, 0.8105726838111877, 0.7437591552734375, 0.7254038453102112, 0.7753304243087769, 0.7268722653388977, 0.7533039450645447, 0.826725423336029, 0.7665198445320129, 0.8193832635879517, 0.8149779438972473, 0.8348017334938049, 0.8032305240631104, 0.8252570033073425, 0.8722466826438904, 0.8722466826438904, 0.8810572624206543, 0.9427312612533569], [0.08663729578256607, 0.13803230226039886, 0.13069015741348267, 0.13950073719024658, 0.13656388223171234, 0.14977973699569702, 0.15124815702438354, 0.1409691572189331, 0.22246696054935455, 0.2070484608411789, 0.2672540247440338, 0.19823788106441498, 0.276798814535141, 0.2966226041316986, 0.2657856047153473, 0.26284876465797424, 0.28487518429756165, 0.2966226041316986, 0.3024963140487671, 0.3127753436565399, 0.39647576212882996, 0.3627019226551056, 0.3237885534763336, 0.3553597629070282, 0.38179147243499756, 0.39647576212882996, 0.3994126319885254, 0.39794421195983887, 0.3906020522117615, 0.35976505279541016, 0.4243759214878082, 0.46255508065223694, 0.4324522614479065, 0.42584434151649475, 0.41997063159942627, 0.4008810520172119, 0.4434655010700226, 0.44640234112739563, 0.5036710500717163, NaN, 0.5403817892074585, 0.5212922096252441, 0.5007342100143433, 0.4977973699569702, 0.6997063159942627, 0.5653451085090637, 0.5323054194450378, 0.5301027894020081, 0.5895741581916809, 0.7327459454536438, 0.596182107925415, 0.6101321578025818, 0.622613787651062, 0.6071953177452087, 0.6049926280975342, 0.6725403666496277, 0.6240822076797485, 0.6696035265922546, 0.6666666865348816, 0.676945686340332, 0.6593245267868042, 0.6864904761314392, 0.691629946231842, 0.7356828451156616, 0.7753304243087769, 0.7386196851730347, 0.772393524646759, 0.7151248455047607, 0.7980910539627075, 0.8340675234794617, 0.7650514245033264, 0.8472834229469299, 0.8414096832275391, 0.8604992628097534, 0.8281938433647156, 0.8105726838111877, 0.9104258418083191, 0.8883994221687317, 0.880323052406311, 0.9559471607208252], [0.044052861630916595, 0.08516886830329895, 0.08810572326183319, 0.10572687536478043, 0.0969163030385971, 0.10425844043493271, 0.11747430264949799, 0.11160058528184891, 0.1997063159942627, 0.2290748953819275, 0.2217327505350113, 0.14977973699569702, 0.24449339509010315, 0.2540381848812103, 0.20558002591133118, 0.2525697648525238, 0.2745961844921112, 0.2731277644634247, 0.3127753436565399, 0.31424376368522644, 0.33773863315582275, 0.2966226041316986, 0.38325992226600647, 0.30690160393714905, 0.3715124726295471, 0.3773861825466156, 0.3348017632961273, 0.40969163179397583, 0.3773861825466156, 0.40234947204589844, 0.4251101315021515, 0.44933921098709106, 0.42584434151649475, 0.4434655010700226, 0.4390602111816406, 0.43465492129325867, 0.4390602111816406, 0.46989721059799194, 0.3994126319885254, 0.4596182107925415, NaN, 0.5477239489555359, 0.5205579996109009, 0.5007342100143433, 0.515418529510498, 0.5403817892074585, 0.6182085275650024, 0.5359765291213989, 0.607929527759552, 0.5580029487609863, 0.6167401075363159, 0.7165932655334473, 0.5770925283432007, 0.715859055519104, 0.5565345287322998, 0.6549192070960999, 0.5800293684005737, 0.6629955768585205, 0.6798825263977051, 0.6182085275650024, 0.6505139470100403, 0.688693106174469, 0.6975036859512329, 0.8722466826438904, 0.7195301055908203, 0.7459618449211121, 0.7606461048126221, 0.757709264755249, 0.7738619446754456, 0.8054332137107849, 0.7371512651443481, 0.7863436341285706, 0.8524228930473328, 0.8164464235305786, 0.7958884239196777, 0.7679882645606995, 0.8737151026725769, 0.8839941024780273, 0.8722466826438904, 0.9515418410301208], [0.07342144101858139, 0.11086637526750565, 0.12187959253787994, 0.11967694759368896, 0.1409691572189331, 0.1189427301287651, 0.16446402668952942, 0.14684288203716278, 0.21512481570243835, 0.22466960549354553, 0.19676946103572845, 0.18795888125896454, 0.232745960354805, 0.2540381848812103, 0.24963289499282837, 0.2819383144378662, 0.27606460452079773, 0.2988252639770508, 0.2907488942146301, 0.3091042637825012, 0.311306893825531, 0.29955947399139404, 0.33773863315582275, 0.3333333432674408, 0.3553597629070282, 0.3201174736022949, 0.36563876271247864, 0.38032305240631104, 0.37298092246055603, 0.3759177625179291, 0.3935389220714569, 0.4647577106952667, 0.40381792187690735, 0.3465491831302643, 0.45521292090415955, 0.4889867901802063, 0.4273127615451813, 0.3406754732131958, 0.43171805143356323, 0.47870779037475586, 0.4522760510444641, NaN, 0.45301029086112976, 0.5814977884292603, 0.527165949344635, 0.48604992032051086, 0.5403817892074585, 0.5932452082633972, 0.5668135285377502, 0.5756241083145142, 0.5976505279541016, 0.538913369178772, 0.6093979477882385, 0.6049926280975342, 0.6358296871185303, 0.6490455269813538, 0.610866367816925, 0.6534507870674133, 0.6651982665061951, 0.7180616855621338, 0.6299559473991394, 0.6284875273704529, 0.7239353656768799, 0.7606461048126221, 0.7151248455047607, 0.7856094241142273, 0.7356828451156616, 0.7444933652877808, 0.7635829448699951, 0.826725423336029, 0.7980910539627075, 0.7973568439483643, 0.8325991034507751, 0.8157122135162354, 0.8604992628097534, 0.838472843170166, 0.8472834229469299, 0.8957415819168091, 0.8913362622261047, 0.9765051603317261], [0.055800292640924454, 0.07121879607439041, 0.08663729578256607, 0.10425844043493271, 0.10132158547639847, 0.07929515093564987, 0.1189427301287651, 0.11233480274677277, 0.17621144652366638, 0.15859030187129974, 0.17547723650932312, 0.1938326060771942, 0.17914831638336182, 0.2143906056880951, 0.20998531579971313, 0.2657856047153473, 0.2540381848812103, 0.276798814535141, 0.3171806037425995, 0.3245227634906769, 0.3054331839084625, 0.2907488942146301, 0.3392070531845093, 0.34361234307289124, 0.31571218371391296, 0.26651981472969055, 0.39867842197418213, 0.35609397292137146, 0.35095447301864624, 0.39794421195983887, 0.4008810520172119, 0.41262850165367126, 0.41262850165367126, 0.4449339210987091, 0.41262850165367126, 0.4831130802631378, 0.4287812113761902, 0.453744500875473, 0.4339207112789154, 0.49926578998565674, 0.4794420003890991, 0.5469897389411926, NaN, 0.5124816298484802, 0.553597629070282, 0.5403817892074585, 0.5785609483718872, 0.5616739988327026, 0.5624082088470459, 0.5917767882347107, 0.6049926280975342, 0.5888399481773376, 0.5873715281486511, 0.6240822076797485, 0.5932452082633972, 0.6380323171615601, 0.5873715281486511, 0.634361207485199, 0.6710719466209412, 0.634361207485199, 0.6740087866783142, 0.700440526008606, 0.7077826857566833, 0.7870778441429138, 0.7136563658714294, 0.7334801554679871, 0.7136563658714294, 0.7856094241142273, 0.7606461048126221, 0.8054332137107849, 0.7900146842002869, 0.727606475353241, 0.8296622633934021, 0.7591776847839355, 0.7797356843948364, 0.8208516836166382, 0.850220263004303, 0.8193832635879517, 0.8480176329612732, 0.9544786810874939], [0.06167401000857353, 0.09397944062948227, 0.08957415819168091, 0.10866373032331467, 0.10352423042058945, 0.08076358586549759, 0.14317180216312408, 0.12334802001714706, 0.19530102610588074, 0.18061673641204834, 0.21145375072956085, 0.18649046123027802, 0.21732746064662933, 0.24082231521606445, 0.21365638077259064, 0.2378854602575302, 0.27900147438049316, 0.2878120541572571, 0.2657856047153473, 0.32892805337905884, 0.3201174736022949, 0.3494860529899597, 0.2621145248413086, 0.29955947399139404, 0.31424376368522644, 0.3171806037425995, 0.40234947204589844, 0.3259911835193634, 0.36343613266944885, 0.34361234307289124, 0.38325992226600647, 0.4008810520172119, 0.39867842197418213, 0.38179147243499756, 0.3678414225578308, 0.4251101315021515, 0.4324522614479065, 0.43171805143356323, 0.47870779037475586, 0.5022026300430298, 0.49926578998565674, 0.41850221157073975, 0.4875183701515198, NaN, 0.5477239489555359, 0.5212922096252441, 0.5066079497337341, 0.5734214186668396, 0.5227606296539307, 0.5800293684005737, 0.5638766288757324, 0.5550661087036133, 0.5638766288757324, 0.5587371587753296, 0.7753304243087769, 0.6651982665061951, 0.5682819485664368, 0.6593245267868042, 0.6417033672332764, 0.6637297868728638, 0.7099853157997131, 0.6710719466209412, 0.6872246861457825, 0.6960352659225464, 0.7753304243087769, 0.757709264755249, 0.7533039450645447, 0.7195301055908203, 0.8061674237251282, 0.8002936840057373, 0.7606461048126221, 0.7430249452590942, 0.8091042637825012, 0.7606461048126221, 0.7621145248413086, 0.7988252639770508, 0.8678414225578308, 0.8443465232849121, 0.8634361028671265, 0.9684287905693054], [0.07488986849784851, 0.12334802001714706, 0.12187959253787994, 0.13656388223171234, 0.1226138025522232, 0.12555065751075745, 0.1409691572189331, 0.12041115760803223, 0.20264317095279694, 0.19530102610588074, 0.2342143952846527, 0.17180617153644562, 0.257709264755249, 0.2731277644634247, 0.24229075014591217, 0.23348017036914825, 0.24963289499282837, 0.27019089460372925, 0.27019089460372925, 0.28634360432624817, 0.3722466826438904, 0.3274596035480499, 0.28487518429756165, 0.3186490535736084, 0.3281938433647156, 0.35242289304733276, 0.36563876271247864, 0.34508076310157776, 0.34361234307289124, 0.30690160393714905, 0.3641703426837921, 0.4287812113761902, 0.3876652121543884, 0.3876652121543884, 0.3538913428783417, 0.3612334728240967, 0.3861967623233795, 0.4030837118625641, 0.4390602111816406, 0.3002936840057373, 0.48458150029182434, 0.4728340804576874, 0.44640234112739563, 0.4522760510444641, NaN, 0.5007342100143433, 0.4889867901802063, 0.4765051305294037, 0.538913369178772, 0.6336269974708557, 0.5712187886238098, 0.5756241083145142, 0.6093979477882385, 0.5763583183288574, 0.553597629070282, 0.6549192070960999, 0.6138032078742981, 0.6571218967437744, 0.6270191073417664, 0.635095477104187, 0.6093979477882385, 0.6417033672332764, 0.6710719466209412, 0.7121879458427429, 0.7400881052017212, 0.7085168957710266, 0.7547723650932312, 0.691629946231842, 0.7679882645606995, 0.8002936840057373, 0.7533039450645447, 0.8208516836166382, 0.8091042637825012, 0.835536003112793, 0.8046989440917969, 0.7870778441429138, 0.8898678421974182, 0.8693098425865173, 0.8795888423919678, 0.9486050009727478], [0.06314244121313095, 0.08223201334476471, 0.10572687536478043, 0.10572687536478043, 0.11747430264949799, 0.10866373032331467, 0.12922173738479614, 0.12922173738479614, 0.18208517134189606, 0.11820852011442184, 0.1938326060771942, 0.20411160588264465, 0.21732746064662933, 0.24082231521606445, 0.2378854602575302, 0.20925110578536987, 0.22026431560516357, 0.26064610481262207, 0.30102789402008057, 0.2745961844921112, 0.345814973115921, 0.2525697648525238, 0.30837005376815796, 0.27019089460372925, 0.32892805337905884, 0.31424376368522644, 0.3274596035480499, 0.38179147243499756, 0.34508076310157776, 0.3355359733104706, 0.3759177625179291, 0.41850221157073975, 0.4243759214878082, 0.37298092246055603, 0.37298092246055603, 0.40969163179397583, 0.40528634190559387, 0.3788546323776245, 0.40381792187690735, 0.43465492129325867, 0.4596182107925415, 0.5139500498771667, 0.4596182107925415, 0.47870779037475586, 0.49926578998565674, NaN, 0.5176211595535278, 0.4977973699569702, 0.5638766288757324, 0.5301027894020081, 0.5976505279541016, 0.5873715281486511, 0.5888399481773376, 0.607929527759552, 0.5521292090415955, 0.6402349472045898, 0.6005873680114746, 0.6490455269813538, 0.6607929468154907, 0.6255506873130798, 0.6358296871185303, 0.6490455269813538, 0.6901615262031555, 0.7863436341285706, 0.7092511057853699, 0.7613803148269653, 0.7386196851730347, 0.7268722653388977, 0.7679882645606995, 0.7914831042289734, 0.7709251046180725, 0.7767988443374634, 0.8678414225578308, 0.8076358437538147, 0.8223201036453247, 0.7973568439483643, 0.864904522895813, 0.8766520023345947, 0.8883994221687317, 0.9750367403030396], [0.07635829597711563, 0.12555065751075745, 0.12187959253787994, 0.13509544730186462, 0.1226138025522232, 0.13509544730186462, 0.13950073719024658, 0.10719530284404755, 0.19162996113300323, 0.17988252639770508, 0.17621144652366638, 0.17180617153644562, 0.22466960549354553, 0.24669604003429413, 0.20558002591133118, 0.2452276051044464, 0.25550660490989685, 0.2709251046180725, 0.3002936840057373, 0.2878120541572571, 0.311306893825531, 0.2290748953819275, 0.311306893825531, 0.30469897389411926, 0.35095447301864624, 0.311306893825531, 0.3024963140487671, 0.3678414225578308, 0.36710718274116516, 0.350220263004303, 0.3715124726295471, 0.4155653417110443, 0.39867842197418213, 0.35756242275238037, 0.3788546323776245, 0.43538913130760193, 0.39500734210014343, 0.3994126319885254, 0.38179147243499756, 0.46769458055496216, 0.38179147243499756, 0.4596182107925415, 0.4214390516281128, 0.49339208006858826, 0.5110132098197937, 0.48237884044647217, NaN, 0.515418529510498, 0.5183553695678711, 0.527165949344635, 0.5345080494880676, 0.515418529510498, 0.5800293684005737, 0.6563876867294312, 0.5433186292648315, 0.6138032078742981, 0.5829662084579468, 0.6182085275650024, 0.6439059972763062, 0.5932452082633972, 0.5704845786094666, 0.6064611077308655, 0.6578561067581177, 0.7261380553245544, 0.6519823670387268, 0.7356828451156616, 0.7224669456481934, 0.7107195258140564, 0.6901615262031555, 0.7885462641716003, 0.7430249452590942, 0.8076358437538147, 0.9456681609153748, 0.8311306834220886, 0.7988252639770508, 0.7562408447265625, 0.8208516836166382, 0.8795888423919678, 0.8737151026725769, 0.9456681609153748], [0.055800292640924454, 0.09251101315021515, 0.08223201334476471, 0.0969163030385971, 0.09397944062948227, 0.07635829597711563, 0.1262848824262619, 0.11747430264949799, 0.18355359137058258, 0.1740088164806366, 0.19676946103572845, 0.17474302649497986, 0.20117473602294922, 0.2290748953819275, 0.20411160588264465, 0.20411160588264465, 0.24816446006298065, 0.26431718468666077, 0.24963289499282837, 0.31424376368522644, 0.31571218371391296, 0.3171806037425995, 0.23201175034046173, 0.30690160393714905, 0.29221731424331665, 0.3098384737968445, 0.38325992226600647, 0.311306893825531, 0.3223201036453247, 0.3171806037425995, 0.3480176329612732, 0.37444934248924255, 0.35242289304733276, 0.37665197253227234, 0.33627018332481384, 0.3994126319885254, 0.3876652121543884, 0.3994126319885254, 0.453744500875473, 0.46989721059799194, 0.46402350068092346, 0.4067547619342804, 0.43832600116729736, 0.426578551530838, 0.5234948396682739, 0.5022026300430298, 0.48458150029182434, NaN, 0.5022026300430298, 0.5477239489555359, 0.5477239489555359, 0.5256975293159485, 0.5447870492935181, 0.5381791591644287, 0.7077826857566833, 0.6519823670387268, 0.5477239489555359, 0.6505139470100403, 0.610866367816925, 0.6387665271759033, 0.6666666865348816, 0.6483113169670105, 0.6593245267868042, 0.685756266117096, 0.730543315410614, 0.7151248455047607, 0.7356828451156616, 0.6960352659225464, 0.7628487348556519, 0.7767988443374634, 0.7488986849784851, 0.7283406853675842, 0.784140944480896, 0.7356828451156616, 0.7430249452590942, 0.7635829448699951, 0.8582966327667236, 0.8281938433647156, 0.8370044231414795, 0.9544786810874939], [0.05433186516165733, 0.07342144101858139, 0.09104258567094803, 0.12114537507295609, 0.14243759214878082, 0.10719530284404755, 0.14757709205150604, 0.14537444710731506, 0.18575623631477356, 0.16886931657791138, 0.15051394701004028, 0.20117473602294922, 0.17621144652366638, 0.2143906056880951, 0.2584434747695923, 0.28634360432624817, 0.24816446006298065, 0.3024963140487671, 0.2584434747695923, 0.27900147438049316, 0.24816446006298065, 0.3024963140487671, 0.29001468420028687, 0.3348017632961273, 0.28634360432624817, 0.29588839411735535, 0.3318648934364319, 0.3406754732131958, 0.3054331839084625, 0.3281938433647156, 0.3612334728240967, 0.39794421195983887, 0.3759177625179291, 0.3876652121543884, 0.35095447301864624, 0.4508076310157776, 0.3649045526981354, 0.3906020522117615, 0.384728342294693, 0.4104258418083191, 0.392070472240448, 0.43318650126457214, 0.4375917911529541, 0.47723934054374695, 0.461086630821228, 0.4361233413219452, 0.4816446304321289, 0.4977973699569702, NaN, 0.49339208006858826, 0.6556534767150879, 0.4559471309185028, 0.6417033672332764, 0.5227606296539307, 0.5286343693733215, 0.5521292090415955, 0.6461086869239807, 0.5638766288757324, 0.7400881052017212, 0.6637297868728638, 0.5594713687896729, 0.5682819485664368, 0.7988252639770508, 0.6211453676223755, 0.6534507870674133, 0.7121879458427429, 0.6982378959655762, 0.7195301055908203, 0.6828193664550781, 0.7165932655334473, 0.9060205817222595, 0.727606475353241, 0.7474302649497986, 0.7378854751586914, 0.8766520023345947, 0.8781204223632812, 0.7635829448699951, 0.8105726838111877, 0.9089574217796326, 0.9897210001945496], [0.07782672345638275, 0.10792951285839081, 0.12041115760803223, 0.11013215780258179, 0.1262848824262619, 0.13069015741348267, 0.13950073719024658, 0.11306901276111603, 0.17474302649497986, 0.17254038155078888, 0.20264317095279694, 0.16446402668952942, 0.24229075014591217, 0.2525697648525238, 0.1997063159942627, 0.22099852561950684, 0.2525697648525238, 0.2540381848812103, 0.2657856047153473, 0.2687224745750427, 0.3201174736022949, 0.3054331839084625, 0.2143906056880951, 0.27900147438049316, 0.28487518429756165, 0.3201174736022949, 0.3370043933391571, 0.32158589363098145, 0.2892804741859436, 0.2687224745750427, 0.3171806037425995, 0.3935389220714569, 0.3553597629070282, 0.36563876271247864, 0.3054331839084625, 0.33039647340774536, 0.3465491831302643, 0.3906020522117615, 0.4104258418083191, 0.2672540247440338, 0.44199705123901367, 0.4243759214878082, 0.4082232117652893, 0.41997063159942627, 0.3663729727268219, 0.46989721059799194, 0.4728340804576874, 0.4522760510444641, 0.5066079497337341, NaN, 0.5183553695678711, 0.5491923689842224, 0.5917767882347107, 0.5447870492935181, 0.5110132098197937, 0.6358296871185303, 0.5932452082633972, 0.6358296871185303, 0.596182107925415, 0.619676947593689, 0.5638766288757324, 0.5859031081199646, 0.6270191073417664, 0.700440526008606, 0.6930983662605286, 0.700440526008606, 0.7342143654823303, 0.6475771069526672, 0.7209985256195068, 0.7944199442863464, 0.7371512651443481, 0.8091042637825012, 0.8032305240631104, 0.8348017334938049, 0.8252570033073425, 0.757709264755249, 0.8604992628097534, 0.8913362622261047, 0.8913362622261047, 0.9515418410301208], [0.06020557880401611, 0.07195301353931427, 0.09985315799713135, 0.1189427301287651, 0.13950073719024658, 0.11233480274677277, 0.1409691572189331, 0.1336270123720169, 0.17474302649497986, 0.15491923689842224, 0.13069015741348267, 0.19016152620315552, 0.16740088164806366, 0.20264317095279694, 0.23201175034046173, 0.2834067642688751, 0.230543315410614, 0.2687224745750427, 0.2584434747695923, 0.2657856047153473, 0.22760646045207977, 0.28487518429756165, 0.28634360432624817, 0.3259911835193634, 0.25917768478393555, 0.2687224745750427, 0.3318648934364319, 0.29368576407432556, 0.29368576407432556, 0.31057268381118774, 0.3171806037425995, 0.3869309723377228, 0.3348017632961273, 0.3641703426837921, 0.3333333432674408, 0.43465492129325867, 0.33039647340774536, 0.3538913428783417, 0.3641703426837921, 0.40381792187690735, 0.38325992226600647, 0.40234947204589844, 0.39500734210014343, 0.4361233413219452, 0.4287812113761902, 0.40234947204589844, 0.46549192070961, 0.4522760510444641, 0.3443465530872345, 0.4816446304321289, NaN, 0.44933921098709106, 0.6101321578025818, 0.5080763697624207, 0.515418529510498, 0.5543318390846252, 0.6101321578025818, 0.5624082088470459, 0.6387665271759033, 0.6505139470100403, 0.5212922096252441, 0.5080763697624207, 0.7143905758857727, 0.6005873680114746, 0.6328927874565125, 0.7048457860946655, 0.6945667862892151, 0.6813509464263916, 0.6637297868728638, 0.7107195258140564, 0.8281938433647156, 0.7121879458427429, 0.7356828451156616, 0.7195301055908203, 0.8472834229469299, 0.8377386331558228, 0.7503671050071716, 0.8032305240631104, 0.9133626818656921, 0.988252580165863], [0.036710720509290695, 0.06901615113019943, 0.06314244121313095, 0.08663729578256607, 0.07195301353931427, 0.07195301353931427, 0.08957415819168091, 0.08810572326183319, 0.15124815702438354, 0.17914831638336182, 0.1703377366065979, 0.10572687536478043, 0.19089573621749878, 0.21879589557647705, 0.16740088164806366, 0.17988252639770508, 0.20851688086986542, 0.22320117056369781, 0.2672540247440338, 0.2657856047153473, 0.2878120541572571, 0.2452276051044464, 0.29368576407432556, 0.2415565401315689, 0.2878120541572571, 0.2966226041316986, 0.30837005376815796, 0.3318648934364319, 0.30176210403442383, 0.3208516836166382, 0.35095447301864624, 0.3612334728240967, 0.3627019226551056, 0.32305434346199036, 0.3494860529899597, 0.36710718274116516, 0.3715124726295471, 0.38913363218307495, 0.3186490535736084, 0.3898678421974182, 0.2834067642688751, 0.461086630821228, 0.41116005182266235, 0.4449339210987091, 0.4243759214878082, 0.41262850165367126, 0.48458150029182434, 0.4743025004863739, 0.5440528392791748, 0.4508076310157776, 0.5506607890129089, NaN, 0.5403817892074585, 0.5785609483718872, 0.5007342100143433, 0.6431717872619629, 0.5403817892074585, 0.6446402072906494, 0.6152716875076294, 0.5550661087036133, 0.5616739988327026, 0.5829662084579468, 0.6387665271759033, 0.7848751544952393, 0.6431717872619629, 0.7165932655334473, 0.7371512651443481, 0.6960352659225464, 0.6945667862892151, 0.7474302649497986, 0.6989721059799194, 0.7077826857566833, 0.7826725244522095, 0.7290748953819275, 0.7540381550788879, 0.7239353656768799, 0.8076358437538147, 0.8546255230903625, 0.8281938433647156, 0.9368575811386108], [0.1409691572189331, 0.15859030187129974, 0.22320117056369781, 0.1938326060771942, 0.2290748953819275, 0.22466960549354553, 0.2511013150215149, 0.2599118947982788, 0.27900147438049316, 0.23348017036914825, 0.2599118947982788, 0.2907488942146301, 0.23494860529899597, 0.2731277644634247, 0.29955947399139404, 0.3127753436565399, 0.30690160393714905, 0.3553597629070282, 0.2878120541572571, 0.34361234307289124, 0.3098384737968445, 0.32158589363098145, 0.30102789402008057, 0.33773863315582275, 0.303964763879776, 0.30837005376815796, 0.3935389220714569, 0.28487518429756165, 0.3201174736022949, 0.32305434346199036, 0.31571218371391296, 0.3024963140487671, 0.33627018332481384, 0.35976505279541016, 0.3465491831302643, 0.3465491831302643, 0.33039647340774536, 0.39647576212882996, 0.3773861825466156, 0.3773861825466156, 0.4229075014591217, 0.3906020522117615, 0.41262850165367126, 0.4361233413219452, 0.3906020522117615, 0.41116005182266235, 0.41997063159942627, 0.45521292090415955, 0.35829663276672363, 0.4082232117652893, 0.3898678421974182, 0.4596182107925415, NaN, 0.458149790763855, 0.49192363023757935, 0.45521292090415955, 0.5580029487609863, 0.45521292090415955, 0.3715124726295471, 0.49339208006858826, 0.4779735803604126, 0.4243759214878082, 0.4559471309185028, 0.5447870492935181, 0.5624082088470459, 0.553597629070282, 0.6211453676223755, 0.596182107925415, 0.5873715281486511, 0.5124816298484802, 0.5976505279541016, 0.6005873680114746, 0.6130689978599548, 0.6211453676223755, 0.6578561067581177, 0.6270191073417664, 0.6446402072906494, 0.7063142657279968, 0.7826725244522095, 0.9559471607208252], [0.05726872384548187, 0.08590308576822281, 0.09985315799713135, 0.10279001295566559, 0.09251101315021515, 0.10132158547639847, 0.09985315799713135, 0.08663729578256607, 0.17254038155078888, 0.16372980177402496, 0.13069015741348267, 0.1336270123720169, 0.18355359137058258, 0.20264317095279694, 0.16152717173099518, 0.19089573621749878, 0.1997063159942627, 0.2364170402288437, 0.23715125024318695, 0.2452276051044464, 0.2569750249385834, 0.20558002591133118, 0.23935389518737793, 0.23201175034046173, 0.2819383144378662, 0.25036710500717163, 0.2672540247440338, 0.2980910539627075, 0.2914831042289734, 0.2819383144378662, 0.3186490535736084, 0.37298092246055603, 0.32892805337905884, 0.29368576407432556, 0.3171806037425995, 0.38179147243499756, 0.32305434346199036, 0.33773863315582275, 0.30102789402008057, 0.39280471205711365, 0.2841409742832184, 0.39500734210014343, 0.3759177625179291, 0.4412628412246704, 0.42364171147346497, 0.392070472240448, 0.34361234307289124, 0.4618208408355713, 0.47723934054374695, 0.45521292090415955, 0.49192363023757935, 0.4214390516281128, 0.541850209236145, NaN, 0.5051395297050476, 0.5829662084579468, 0.5447870492935181, 0.5800293684005737, 0.5851688981056213, 0.5477239489555359, 0.5036710500717163, 0.5242290496826172, 0.5991189479827881, 0.6541849970817566, 0.5932452082633972, 0.6842877864837646, 0.6960352659225464, 0.6431717872619629, 0.6387665271759033, 0.7430249452590942, 0.7077826857566833, 0.7621145248413086, 0.8443465232849121, 0.7804698944091797, 0.769456684589386, 0.7195301055908203, 0.7606461048126221, 0.8516886830329895, 0.8414096832275391, 0.9353891611099243], [0.041116006672382355, 0.08076358586549759, 0.07929515093564987, 0.08810572326183319, 0.08370044082403183, 0.06461086869239807, 0.10572687536478043, 0.08957415819168091, 0.14977973699569702, 0.15124815702438354, 0.16593244671821594, 0.14977973699569702, 0.17474302649497986, 0.1850220263004303, 0.16446402668952942, 0.15418502688407898, 0.21292217075824738, 0.2290748953819275, 0.23348017036914825, 0.303964763879776, 0.2599118947982788, 0.2364170402288437, 0.15124815702438354, 0.23127754032611847, 0.23494860529899597, 0.2621145248413086, 0.3318648934364319, 0.2878120541572571, 0.24669604003429413, 0.2143906056880951, 0.28707781434059143, 0.3568281829357147, 0.311306893825531, 0.3274596035480499, 0.2290748953819275, 0.3553597629070282, 0.3274596035480499, 0.3876652121543884, 0.3876652121543884, 0.39500734210014343, 0.4434655010700226, 0.3641703426837921, 0.4067547619342804, 0.22466960549354553, 0.44640234112739563, 0.44787079095840454, 0.45668134093284607, 0.29221731424331665, 0.47136563062667847, 0.4889867901802063, 0.48458150029182434, 0.49926578998565674, 0.5080763697624207, 0.4948605000972748, NaN, 0.622613787651062, 0.5066079497337341, 0.6270191073417664, 0.5859031081199646, 0.5859031081199646, 0.5763583183288574, 0.5910425782203674, 0.6064611077308655, 0.6314243674278259, 0.6725403666496277, 0.6475771069526672, 0.7048457860946655, 0.6431717872619629, 0.7077826857566833, 0.7430249452590942, 0.6387665271759033, 0.7195301055908203, 0.7283406853675842, 0.7312775254249573, 0.6938325762748718, 0.7224669456481934, 0.8105726838111877, 0.826725423336029, 0.8414096832275391, 0.9530102610588074], [0.15271659195423126, 0.17841409146785736, 0.19676946103572845, 0.18355359137058258, 0.20411160588264465, 0.19530102610588074, 0.23201175034046173, 0.20264317095279694, 0.26138031482696533, 0.26138031482696533, 0.2672540247440338, 0.21292217075824738, 0.27900147438049316, 0.28854626417160034, 0.2672540247440338, 0.27165931463241577, 0.27753305435180664, 0.2878120541572571, 0.2621145248413086, 0.2731277644634247, 0.32892805337905884, 0.2584434747695923, 0.28707781434059143, 0.29221731424331665, 0.2892804741859436, 0.3318648934364319, 0.3612334728240967, 0.3208516836166382, 0.2878120541572571, 0.29955947399139404, 0.3127753436565399, 0.34508076310157776, 0.3171806037425995, 0.3274596035480499, 0.2988252639770508, 0.31424376368522644, 0.32892805337905884, 0.37444934248924255, 0.303964763879776, 0.3274596035480499, 0.34508076310157776, 0.35095447301864624, 0.36196768283843994, 0.3348017632961273, 0.34508076310157776, 0.35976505279541016, 0.3861967623233795, 0.3480176329612732, 0.44787079095840454, 0.3641703426837921, 0.44566813111305237, 0.3568281829357147, 0.5447870492935181, 0.41703376173973083, 0.3773861825466156, NaN, 0.5521292090415955, 0.5337738394737244, 0.49339208006858826, 0.461086630821228, 0.41850221157073975, 0.4243759214878082, 0.5374449491500854, 0.46549192070961, 0.5682819485664368, 0.5168869495391846, 0.9441996812820435, 0.5374449491500854, 0.5947136282920837, 0.4904552102088928, 0.6358296871185303, 0.6534507870674133, 0.6622613668441772, 0.6784141063690186, 0.7121879458427429, 0.6505139470100403, 0.6299559473991394, 0.781204104423523, 0.8340675234794617, 0.9236416816711426], [0.08076358586549759, 0.13215859234333038, 0.18061673641204834, 0.15712188184261322, 0.18649046123027802, 0.1850220263004303, 0.21145375072956085, 0.2569750249385834, 0.24082231521606445, 0.20411160588264465, 0.2525697648525238, 0.27019089460372925, 0.22320117056369781, 0.24963289499282837, 0.2951541841030121, 0.2980910539627075, 0.28634360432624817, 0.3333333432674408, 0.27606460452079773, 0.3259911835193634, 0.2951541841030121, 0.28634360432624817, 0.29955947399139404, 0.3274596035480499, 0.311306893825531, 0.29955947399139404, 0.3876652121543884, 0.2819383144378662, 0.3171806037425995, 0.3186490535736084, 0.311306893825531, 0.303964763879776, 0.33773863315582275, 0.35829663276672363, 0.34361234307289124, 0.3480176329612732, 0.3274596035480499, 0.3906020522117615, 0.37298092246055603, 0.3759177625179291, 0.41997063159942627, 0.38913363218307495, 0.41262850165367126, 0.43171805143356323, 0.3861967623233795, 0.3994126319885254, 0.41703376173973083, 0.4522760510444641, 0.3538913428783417, 0.4067547619342804, 0.3898678421974182, 0.4596182107925415, 0.44199705123901367, 0.45521292090415955, 0.49339208006858826, 0.44787079095840454, NaN, 0.4508076310157776, 0.37298092246055603, 0.4948605000972748, 0.4794420003890991, 0.4243759214878082, 0.45741555094718933, 0.5447870492935181, 0.5624082088470459, 0.5550661087036133, 0.619676947593689, 0.5947136282920837, 0.5859031081199646, 0.5124816298484802, 0.6005873680114746, 0.6005873680114746, 0.6130689978599548, 0.6211453676223755, 0.6563876867294312, 0.6284875273704529, 0.6446402072906494, 0.7063142657279968, 0.781204104423523, 0.9559471607208252], [0.15418502688407898, 0.18061673641204834, 0.19676946103572845, 0.18942731618881226, 0.20411160588264465, 0.19089573621749878, 0.22760646045207977, 0.2033773809671402, 0.2599118947982788, 0.2569750249385834, 0.26431718468666077, 0.21732746064662933, 0.27753305435180664, 0.2907488942146301, 0.26431718468666077, 0.2687224745750427, 0.27606460452079773, 0.29221731424331665, 0.2672540247440338, 0.276798814535141, 0.3274596035480499, 0.2569750249385834, 0.28634360432624817, 0.28707781434059143, 0.29368576407432556, 0.3311306834220886, 0.3700440526008606, 0.32305434346199036, 0.2892804741859436, 0.2980910539627075, 0.3171806037425995, 0.3494860529899597, 0.31571218371391296, 0.32305434346199036, 0.2980910539627075, 0.3171806037425995, 0.32672539353370667, 0.3773861825466156, 0.30102789402008057, 0.33039647340774536, 0.3370043933391571, 0.3465491831302643, 0.36563876271247864, 0.3406754732131958, 0.342878133058548, 0.35095447301864624, 0.38179147243499756, 0.3494860529899597, 0.4361233413219452, 0.3641703426837921, 0.4375917911529541, 0.3553597629070282, 0.5447870492935181, 0.41997063159942627, 0.37298092246055603, 0.46622613072395325, 0.5491923689842224, NaN, 0.49339208006858826, 0.46549192070961, 0.4243759214878082, 0.4214390516281128, 0.5403817892074585, 0.46255508065223694, 0.5638766288757324, 0.515418529510498, 0.9603524208068848, 0.5345080494880676, 0.5888399481773376, 0.4948605000972748, 0.6372981071472168, 0.6446402072906494, 0.6593245267868042, 0.6754772663116455, 0.7019089460372925, 0.6534507870674133, 0.6284875273704529, 0.781204104423523, 0.8281938433647156, 0.9236416816711426], [0.033773861825466156, 0.06607929617166519, 0.06754772365093231, 0.07342144101858139, 0.08663729578256607, 0.07635829597711563, 0.11306901276111603, 0.11453744769096375, 0.11747430264949799, 0.08957415819168091, 0.11160058528184891, 0.15418502688407898, 0.11160058528184891, 0.15859030187129974, 0.1703377366065979, 0.22026431560516357, 0.16446402668952942, 0.24963289499282837, 0.1997063159942627, 0.230543315410614, 0.2217327505350113, 0.24963289499282837, 0.1850220263004303, 0.25550660490989685, 0.17327459156513214, 0.18942731618881226, 0.2547723948955536, 0.23201175034046173, 0.18942731618881226, 0.23935389518737793, 0.20484581589698792, 0.3715124726295471, 0.303964763879776, 0.2672540247440338, 0.2378854602575302, 0.4067547619342804, 0.23494860529899597, 0.2951541841030121, 0.3127753436565399, 0.3333333432674408, 0.3201174736022949, 0.3348017632961273, 0.32892805337905884, 0.35829663276672363, 0.37298092246055603, 0.3392070531845093, 0.35609397292137146, 0.38913363218307495, 0.2599118947982788, 0.40381792187690735, 0.3612334728240967, 0.384728342294693, 0.6284875273704529, 0.41483113169670105, 0.4140969216823578, 0.5066079497337341, 0.6270191073417664, 0.5066079497337341, NaN, 0.6270191073417664, 0.4801762104034424, 0.49339208006858826, 0.6754772663116455, 0.541850209236145, 0.5814977884292603, 0.6666666865348816, 0.6439059972763062, 0.6563876867294312, 0.6182085275650024, 0.6255506873130798, 0.7665198445320129, 0.685756266117096, 0.6828193664550781, 0.6901615262031555, 0.7767988443374634, 0.8810572624206543, 0.703377366065979, 0.7731277346611023, 0.8634361028671265, 0.9662261605262756], [0.05286343768239021, 0.06607929617166519, 0.06240822374820709, 0.07635829597711563, 0.08370044082403183, 0.10425844043493271, 0.0969163030385971, 0.07635829597711563, 0.17547723650932312, 0.19089573621749878, 0.10719530284404755, 0.1703377366065979, 0.14684288203716278, 0.14243759214878082, 0.22246696054935455, 0.230543315410614, 0.2400881052017212, 0.2488986849784851, 0.22099852561950684, 0.2107195258140564, 0.18942731618881226, 0.12922173738479614, 0.23348017036914825, 0.2672540247440338, 0.2687224745750427, 0.27386197447776794, 0.3098384737968445, 0.2364170402288437, 0.2731277644634247, 0.24963289499282837, 0.276798814535141, 0.28267255425453186, 0.2525697648525238, 0.2804698944091797, 0.2980910539627075, 0.338472843170166, 0.2980910539627075, 0.33773863315582275, 0.3392070531845093, 0.32305434346199036, 0.38179147243499756, 0.2819383144378662, 0.36563876271247864, 0.33627018332481384, 0.3649045526981354, 0.37444934248924255, 0.4067547619342804, 0.3612334728240967, 0.33627018332481384, 0.38032305240631104, 0.3494860529899597, 0.4449339210987091, 0.5066079497337341, 0.4522760510444641, 0.4140969216823578, 0.538913369178772, 0.5051395297050476, 0.5345080494880676, 0.37298092246055603, NaN, 0.44933921098709106, 0.40969163179397583, 0.453744500875473, 0.5345080494880676, 0.5917767882347107, 0.6189427375793457, 0.6842877864837646, 0.5932452082633972, 0.6299559473991394, 0.6431717872619629, 0.5756241083145142, 0.7254038453102112, 0.7019089460372925, 0.7452276349067688, 0.6989721059799194, 0.6336269974708557, 0.7400881052017212, 0.838472843170166, 0.8237885236740112, 0.9559471607208252], [0.033773861825466156, 0.046989720314741135, 0.046989720314741135, 0.06020557880401611, 0.051395006477832794, 0.042584434151649475, 0.06020557880401611, 0.06314244121313095, 0.08663729578256607, 0.07782672345638275, 0.11306901276111603, 0.09985315799713135, 0.11747430264949799, 0.14243759214878082, 0.08516886830329895, 0.12701909244060516, 0.1409691572189331, 0.19530102610588074, 0.18942731618881226, 0.21145375072956085, 0.20851688086986542, 0.17841409146785736, 0.04845815151929855, 0.20558002591133118, 0.12187959253787994, 0.13803230226039886, 0.2687224745750427, 0.21292217075824738, 0.1483113020658493, 0.09618208557367325, 0.10132158547639847, 0.30176210403442383, 0.2731277644634247, 0.28634360432624817, 0.19530102610588074, 0.35609397292137146, 0.1409691572189331, 0.2951541841030121, 0.33773863315582275, 0.3406754732131958, 0.3494860529899597, 0.3700440526008606, 0.3259911835193634, 0.29001468420028687, 0.3906020522117615, 0.3641703426837921, 0.42951542139053345, 0.3333333432674408, 0.44052863121032715, 0.4361233413219452, 0.47870779037475586, 0.43832600116729736, 0.5220264196395874, 0.4963289201259613, 0.42364171147346497, 0.5814977884292603, 0.5205579996109009, 0.5756241083145142, 0.5198237895965576, 0.5506607890129089, NaN, 0.5249632596969604, 0.56901615858078, 0.6255506873130798, 0.5829662084579468, 0.6446402072906494, 0.6622613668441772, 0.6651982665061951, 0.6402349472045898, 0.7121879458427429, 0.688693106174469, 0.6784141063690186, 0.7151248455047607, 0.6945667862892151, 0.7342143654823303, 0.742290735244751, 0.7797356843948364, 0.8017621040344238, 0.8193832635879517, 0.946402370929718], [0.030837005004286766, 0.033773861825466156, 0.041116006672382355, 0.06754772365093231, 0.049926578998565674, 0.033773861825466156, 0.06314244121313095, 0.06461086869239807, 0.09104258567094803, 0.06461086869239807, 0.10719530284404755, 0.08223201334476471, 0.11747430264949799, 0.1556534469127655, 0.0969163030385971, 0.11306901276111603, 0.13069015741348267, 0.20264317095279694, 0.17914831638336182, 0.2217327505350113, 0.23494860529899597, 0.20851688086986542, 0.10132158547639847, 0.17180617153644562, 0.11013215780258179, 0.15271659195423126, 0.27165931463241577, 0.1938326060771942, 0.11453744769096375, 0.13289280235767365, 0.0969163030385971, 0.3171806037425995, 0.27900147438049316, 0.25550660490989685, 0.21879589557647705, 0.35095447301864624, 0.13215859234333038, 0.2378854602575302, 0.29955947399139404, 0.3135095536708832, 0.311306893825531, 0.3715124726295471, 0.29955947399139404, 0.32892805337905884, 0.35829663276672363, 0.35095447301864624, 0.3935389220714569, 0.3516886830329895, 0.43171805143356323, 0.4140969216823578, 0.49192363023757935, 0.41703376173973083, 0.5756241083145142, 0.4757709205150604, 0.40895742177963257, 0.5756241083145142, 0.5756241083145142, 0.5785609483718872, 0.5066079497337341, 0.5903083682060242, 0.47503671050071716, NaN, 0.5712187886238098, 0.5873715281486511, 0.538913369178772, 0.6534507870674133, 0.6696035265922546, 0.6901615262031555, 0.6049926280975342, 0.7268722653388977, 0.7268722653388977, 0.6828193664550781, 0.6930983662605286, 0.6989721059799194, 0.7386196851730347, 0.7826725244522095, 0.7298091053962708, 0.8002936840057373, 0.8135095238685608, 0.933920681476593], [0.035242289304733276, 0.049926578998565674, 0.06314244121313095, 0.07195301353931427, 0.07488986849784851, 0.06020557880401611, 0.0969163030385971, 0.07929515093564987, 0.0969163030385971, 0.06461086869239807, 0.10866373032331467, 0.12041115760803223, 0.09985315799713135, 0.13803230226039886, 0.13876652717590332, 0.17914831638336182, 0.1409691572189331, 0.20190896093845367, 0.18208517134189606, 0.1850220263004303, 0.18208517134189606, 0.20558002591133118, 0.13950073719024658, 0.23494860529899597, 0.1299559473991394, 0.15712188184261322, 0.24596181511878967, 0.19530102610588074, 0.15124815702438354, 0.1850220263004303, 0.17694567143917084, 0.2951541841030121, 0.26138031482696533, 0.24816446006298065, 0.20264317095279694, 0.3480176329612732, 0.19676946103572845, 0.2804698944091797, 0.30102789402008057, 0.30837005376815796, 0.3024963140487671, 0.27606460452079773, 0.29221731424331665, 0.3127753436565399, 0.32892805337905884, 0.3098384737968445, 0.3421438932418823, 0.3406754732131958, 0.20117473602294922, 0.37298092246055603, 0.2856093943119049, 0.3612334728240967, 0.5440528392791748, 0.4008810520172119, 0.3935389220714569, 0.46255508065223694, 0.5425844192504883, 0.4596182107925415, 0.3245227634906769, 0.5462555289268494, 0.43098384141921997, 0.4287812113761902, NaN, 0.5022026300430298, 0.5565345287322998, 0.6020557880401611, 0.622613787651062, 0.6093979477882385, 0.5991189479827881, 0.6035242080688477, 0.6828193664550781, 0.6725403666496277, 0.6798825263977051, 0.6784141063690186, 0.7437591552734375, 0.7452276349067688, 0.6828193664550781, 0.7665198445320129, 0.8546255230903625, 0.9647576808929443], [0.029368575662374496, 0.036710720509290695, 0.036710720509290695, 0.06167401000857353, 0.045521292835474014, 0.044052861630916595, 0.041116006672382355, 0.027900146320462227, 0.09251101315021515, 0.07488986849784851, 0.08810572326183319, 0.07929515093564987, 0.0969163030385971, 0.11013215780258179, 0.09251101315021515, 0.08957415819168091, 0.10425844043493271, 0.15859030187129974, 0.18061673641204834, 0.15124815702438354, 0.1703377366065979, 0.12041115760803223, 0.17180617153644562, 0.11160058528184891, 0.16446402668952942, 0.1556534469127655, 0.2143906056880951, 0.15124815702438354, 0.18208517134189606, 0.1923641711473465, 0.21145375072956085, 0.2834067642688751, 0.15859030187129974, 0.21292217075824738, 0.20851688086986542, 0.2819383144378662, 0.22613802552223206, 0.2525697648525238, 0.18942731618881226, 0.26431718468666077, 0.12775330245494843, 0.23935389518737793, 0.21292217075824738, 0.303964763879776, 0.2878120541572571, 0.21365638077259064, 0.27386197447776794, 0.31424376368522644, 0.3788546323776245, 0.29955947399139404, 0.3994126319885254, 0.21512481570243835, 0.45521292090415955, 0.345814973115921, 0.3685756325721741, 0.5345080494880676, 0.45521292090415955, 0.5374449491500854, 0.458149790763855, 0.46549192070961, 0.37444934248924255, 0.41262850165367126, 0.4977973699569702, NaN, 0.4948605000972748, 0.5734214186668396, 0.6461086869239807, 0.5800293684005737, 0.541850209236145, 0.5682819485664368, 0.6182085275650024, 0.6417033672332764, 0.6490455269813538, 0.6622613668441772, 0.6828193664550781, 0.6328927874565125, 0.6475771069526672, 0.7797356843948364, 0.7944199442863464, 0.91629958152771], [0.016152717173099518, 0.049926578998565674, 0.044052861630916595, 0.051395006477832794, 0.046989720314741135, 0.051395006477832794, 0.06975036859512329, 0.06167401000857353, 0.10279001295566559, 0.06461086869239807, 0.09985315799713135, 0.10866373032331467, 0.10279001295566559, 0.12041115760803223, 0.07342144101858139, 0.12041115760803223, 0.10866373032331467, 0.1923641711473465, 0.17621144652366638, 0.18208517134189606, 0.1629955917596817, 0.1336270123720169, 0.09985315799713135, 0.14537444710731506, 0.14537444710731506, 0.1483113020658493, 0.2143906056880951, 0.18795888125896454, 0.15712188184261322, 0.13876652717590332, 0.18208517134189606, 0.24669604003429413, 0.2268722504377365, 0.23201175034046173, 0.17327459156513214, 0.2745961844921112, 0.20558002591133118, 0.2584434747695923, 0.2562408149242401, 0.22466960549354553, 0.2804698944091797, 0.28487518429756165, 0.28634360432624817, 0.22466960549354553, 0.2599118947982788, 0.2907488942146301, 0.3480176329612732, 0.269456684589386, 0.3465491831302643, 0.30690160393714905, 0.36710718274116516, 0.3568281829357147, 0.4375917911529541, 0.4067547619342804, 0.3274596035480499, 0.43171805143356323, 0.4375917911529541, 0.4361233413219452, 0.41850221157073975, 0.4082232117652893, 0.41703376173973083, 0.461086630821228, 0.4434655010700226, 0.5051395297050476, NaN, 0.5168869495391846, 0.5212922096252441, 0.5477239489555359, 0.7085168957710266, 0.5741556286811829, 0.5668135285377502, 0.6402349472045898, 0.6005873680114746, 0.6534507870674133, 0.5998531579971313, 0.6020557880401611, 0.8223201036453247, 0.7180616855621338, 0.7400881052017212, 0.9280470013618469], [0.055800292640924454, 0.06314244121313095, 0.07635829597711563, 0.08663729578256607, 0.08370044082403183, 0.07929515093564987, 0.09985315799713135, 0.08149779587984085, 0.13876652717590332, 0.15271659195423126, 0.07195301353931427, 0.12408223003149033, 0.1262848824262619, 0.13215859234333038, 0.1483113020658493, 0.1703377366065979, 0.1666666716337204, 0.1813509613275528, 0.22760646045207977, 0.13436123728752136, 0.1189427301287651, 0.09544786810874939, 0.19676946103572845, 0.19676946103572845, 0.23348017036914825, 0.17621144652366638, 0.22760646045207977, 0.16740088164806366, 0.24229075014591217, 0.21292217075824738, 0.2342143952846527, 0.26064610481262207, 0.16005873680114746, 0.20264317095279694, 0.2378854602575302, 0.27900147438049316, 0.24816446006298065, 0.26138031482696533, 0.2745961844921112, 0.26138031482696533, 0.2540381848812103, 0.2143906056880951, 0.26651981472969055, 0.24229075014591217, 0.2914831042289734, 0.23861967027187347, 0.26431718468666077, 0.28487518429756165, 0.2878120541572571, 0.29955947399139404, 0.2951541841030121, 0.2834067642688751, 0.44640234112739563, 0.31571218371391296, 0.35242289304733276, 0.4831130802631378, 0.4449339210987091, 0.48458150029182434, 0.3333333432674408, 0.3810572624206543, 0.3553597629070282, 0.3465491831302643, 0.39794421195983887, 0.426578551530838, 0.4831130802631378, NaN, 0.6240822076797485, 0.5036710500717163, 0.5330396294593811, 0.5565345287322998, 0.49192363023757935, 0.6431717872619629, 0.622613787651062, 0.6659324765205383, 0.6475771069526672, 0.5734214186668396, 0.6138032078742981, 0.7635829448699951, 0.7767988443374634, 0.93098384141922], [0.11453744769096375, 0.13215859234333038, 0.15124815702438354, 0.12922173738479614, 0.1629955917596817, 0.15712188184261322, 0.17180617153644562, 0.15418502688407898, 0.20264317095279694, 0.19530102610588074, 0.1850220263004303, 0.18208517134189606, 0.19676946103572845, 0.1997063159942627, 0.20117473602294922, 0.20411160588264465, 0.2143906056880951, 0.23348017036914825, 0.21659325063228607, 0.22320117056369781, 0.21879589557647705, 0.18795888125896454, 0.20558002591133118, 0.22760646045207977, 0.22613802552223206, 0.230543315410614, 0.23935389518737793, 0.24816446006298065, 0.22466960549354553, 0.22320117056369781, 0.24963289499282837, 0.27019089460372925, 0.24963289499282837, 0.2540381848812103, 0.2217327505350113, 0.23861967027187347, 0.2525697648525238, 0.31424376368522644, 0.22466960549354553, 0.22760646045207977, 0.23935389518737793, 0.26431718468666077, 0.28634360432624817, 0.24669604003429413, 0.2452276051044464, 0.26138031482696533, 0.27753305435180664, 0.26431718468666077, 0.30176210403442383, 0.2657856047153473, 0.3054331839084625, 0.26284876465797424, 0.3788546323776245, 0.303964763879776, 0.2951541841030121, 0.055800292640924454, 0.38032305240631104, 0.039647575467824936, 0.35609397292137146, 0.31571218371391296, 0.33773863315582275, 0.33039647340774536, 0.3773861825466156, 0.3538913428783417, 0.47870779037475586, 0.3759177625179291, NaN, 0.4273127615451813, 0.4977973699569702, 0.4140969216823578, 0.4728340804576874, 0.5359765291213989, 0.4889867901802063, 0.5477239489555359, 0.5814977884292603, 0.45521292090415955, 0.5315712094306946, 0.6578561067581177, 0.6960352659225464, 0.8898678421974182], [0.029368575662374496, 0.035242289304733276, 0.04845815151929855, 0.05286343768239021, 0.06607929617166519, 0.051395006477832794, 0.07488986849784851, 0.06607929617166519, 0.07929515093564987, 0.07342144101858139, 0.09985315799713135, 0.09544786810874939, 0.08883994072675705, 0.11453744769096375, 0.11453744769096375, 0.14684288203716278, 0.13215859234333038, 0.14537444710731506, 0.13656388223171234, 0.14390602707862854, 0.1556534469127655, 0.15859030187129974, 0.18208517134189606, 0.1938326060771942, 0.15859030187129974, 0.12187959253787994, 0.20998531579971313, 0.06020557880401611, 0.1703377366065979, 0.1886930912733078, 0.18942731618881226, 0.2437591850757599, 0.11013215780258179, 0.15712188184261322, 0.23201175034046173, 0.3024963140487671, 0.1938326060771942, 0.16813509166240692, 0.2731277644634247, 0.28487518429756165, 0.24229075014591217, 0.25550660490989685, 0.2143906056880951, 0.2804698944091797, 0.30837005376815796, 0.2731277644634247, 0.2892804741859436, 0.303964763879776, 0.2804698944091797, 0.35242289304733276, 0.3186490535736084, 0.303964763879776, 0.40381792187690735, 0.3568281829357147, 0.3568281829357147, 0.46255508065223694, 0.40528634190559387, 0.46549192070961, 0.34361234307289124, 0.4067547619342804, 0.3348017632961273, 0.3098384737968445, 0.3906020522117615, 0.41997063159942627, 0.4522760510444641, 0.4963289201259613, 0.5726872086524963, NaN, 0.4765051305294037, 0.5315712094306946, 0.4728340804576874, 0.607929527759552, 0.5447870492935181, 0.607929527759552, 0.5932452082633972, 0.541850209236145, 0.5859031081199646, 0.7048457860946655, 0.7180616855621338, 0.922173261642456], [0.017621144652366638, 0.039647575467824936, 0.038913361728191376, 0.046989720314741135, 0.046989720314741135, 0.05433186516165733, 0.05873715132474899, 0.05433186516165733, 0.07782672345638275, 0.05726872384548187, 0.07929515093564987, 0.09104258567094803, 0.09177679568529129, 0.11160058528184891, 0.07048457860946655, 0.09985315799713135, 0.10425844043493271, 0.17841409146785736, 0.16740088164806366, 0.17180617153644562, 0.13950073719024658, 0.1189427301287651, 0.08223201334476471, 0.13509544730186462, 0.1262848824262619, 0.12041115760803223, 0.20778267085552216, 0.16079294681549072, 0.13509544730186462, 0.11747430264949799, 0.1593245267868042, 0.23201175034046173, 0.18428781628608704, 0.19823788106441498, 0.13803230226039886, 0.2511013150215149, 0.17621144652366638, 0.24816446006298065, 0.24669604003429413, 0.20190896093845367, 0.22613802552223206, 0.2364170402288437, 0.23935389518737793, 0.1938326060771942, 0.23201175034046173, 0.23201175034046173, 0.3098384737968445, 0.23715125024318695, 0.3171806037425995, 0.27900147438049316, 0.33627018332481384, 0.3054331839084625, 0.41262850165367126, 0.3612334728240967, 0.29221731424331665, 0.40528634190559387, 0.4140969216823578, 0.41116005182266235, 0.38179147243499756, 0.3700440526008606, 0.35976505279541016, 0.39500734210014343, 0.4008810520172119, 0.458149790763855, 0.2914831042289734, 0.4669603407382965, 0.5022026300430298, 0.5234948396682739, NaN, 0.5462555289268494, 0.5330396294593811, 0.6240822076797485, 0.5565345287322998, 0.634361207485199, 0.5609397888183594, 0.5712187886238098, 0.7114537358283997, 0.700440526008606, 0.7180616855621338, 0.9030836820602417], [0.051395006477832794, 0.045521292835474014, 0.049926578998565674, 0.06167401000857353, 0.05726872384548187, 0.05726872384548187, 0.06020557880401611, 0.049926578998565674, 0.09838473051786423, 0.08076358586549759, 0.10132158547639847, 0.07488986849784851, 0.10572687536478043, 0.12481644749641418, 0.09985315799713135, 0.09104258567094803, 0.11306901276111603, 0.12775330245494843, 0.1409691572189331, 0.15271659195423126, 0.14977973699569702, 0.1409691572189331, 0.09397944062948227, 0.10572687536478043, 0.12701909244060516, 0.13803230226039886, 0.20998531579971313, 0.16740088164806366, 0.13950073719024658, 0.10646108537912369, 0.1703377366065979, 0.17327459156513214, 0.15418502688407898, 0.14977973699569702, 0.10132158547639847, 0.1336270123720169, 0.17327459156513214, 0.22320117056369781, 0.17327459156513214, 0.16593244671821594, 0.19456681609153748, 0.17327459156513214, 0.19456681609153748, 0.1997063159942627, 0.1997063159942627, 0.20851688086986542, 0.21145375072956085, 0.22320117056369781, 0.2834067642688751, 0.20558002591133118, 0.2892804741859436, 0.2525697648525238, 0.4875183701515198, 0.2569750249385834, 0.2569750249385834, 0.5095447897911072, 0.4875183701515198, 0.5051395297050476, 0.37444934248924255, 0.3568281829357147, 0.2878120541572571, 0.2731277644634247, 0.39647576212882996, 0.43171805143356323, 0.42584434151649475, 0.4434655010700226, 0.5859031081199646, 0.4684287905693054, 0.453744500875473, NaN, 0.4757709205150604, 0.6182085275650024, 0.5550661087036133, 0.6328927874565125, 0.6710719466209412, 0.5462555289268494, 0.5638766288757324, 0.7283406853675842, 0.781204104423523, 0.8854625821113586], [0.014684287831187248, 0.029368575662374496, 0.049926578998565674, 0.049926578998565674, 0.051395006477832794, 0.046989720314741135, 0.07048457860946655, 0.06754772365093231, 0.04845815151929855, 0.046989720314741135, 0.055066078901290894, 0.09985315799713135, 0.05726872384548187, 0.08370044082403183, 0.08223201334476471, 0.12922173738479614, 0.12481644749641418, 0.14537444710731506, 0.1629955917596817, 0.15124815702438354, 0.09397944062948227, 0.14243759214878082, 0.0969163030385971, 0.18355359137058258, 0.11600587517023087, 0.10132158547639847, 0.20558002591133118, 0.07488986849784851, 0.13069015741348267, 0.1262848824262619, 0.14390602707862854, 0.22466960549354553, 0.14684288203716278, 0.17180617153644562, 0.1556534469127655, 0.2878120541572571, 0.14684288203716278, 0.1850220263004303, 0.23348017036914825, 0.23494860529899597, 0.26284876465797424, 0.20190896093845367, 0.20998531579971313, 0.23935389518737793, 0.24669604003429413, 0.2290748953819275, 0.2569750249385834, 0.2511013150215149, 0.09397944062948227, 0.26284876465797424, 0.17180617153644562, 0.30102789402008057, 0.40234947204589844, 0.29221731424331665, 0.3612334728240967, 0.3641703426837921, 0.3994126319885254, 0.3627019226551056, 0.23348017036914825, 0.4243759214878082, 0.311306893825531, 0.2731277644634247, 0.3171806037425995, 0.38179147243499756, 0.43318650126457214, 0.5080763697624207, 0.527165949344635, 0.527165949344635, 0.4669603407382965, 0.5242290496826172, NaN, 0.5873715281486511, 0.5565345287322998, 0.5991189479827881, 0.6417033672332764, 0.5712187886238098, 0.5462555289268494, 0.7048457860946655, 0.7679882645606995, 0.9662261605262756], [0.046989720314741135, 0.07048457860946655, 0.08663729578256607, 0.06020557880401611, 0.07488986849784851, 0.08516886830329895, 0.08516886830329895, 0.07342144101858139, 0.09985315799713135, 0.11306901276111603, 0.11747430264949799, 0.08516886830329895, 0.15859030187129974, 0.1666666716337204, 0.11674008518457413, 0.10719530284404755, 0.15859030187129974, 0.12922173738479614, 0.12481644749641418, 0.10352423042058945, 0.22613802552223206, 0.17180617153644562, 0.13950073719024658, 0.12775330245494843, 0.16740088164806366, 0.22026431560516357, 0.1850220263004303, 0.1556534469127655, 0.17180617153644562, 0.1483113020658493, 0.17914831638336182, 0.16446402668952942, 0.15859030187129974, 0.19530102610588074, 0.19089573621749878, 0.10425844043493271, 0.18942731618881226, 0.2657856047153473, 0.18061673641204834, 0.15271659195423126, 0.21365638077259064, 0.20264317095279694, 0.27239352464675903, 0.2569750249385834, 0.17914831638336182, 0.22320117056369781, 0.1923641711473465, 0.27165931463241577, 0.27239352464675903, 0.19089573621749878, 0.2878120541572571, 0.29221731424331665, 0.3994126319885254, 0.2378854602575302, 0.2804698944091797, 0.3465491831302643, 0.3994126319885254, 0.3553597629070282, 0.31424376368522644, 0.2745961844921112, 0.32158589363098145, 0.3171806037425995, 0.3274596035480499, 0.35829663276672363, 0.35976505279541016, 0.3568281829357147, 0.46402350068092346, 0.392070472240448, 0.3759177625179291, 0.38179147243499756, 0.41262850165367126, NaN, 0.38032305240631104, 0.6336269974708557, 0.47723934054374695, 0.4140969216823578, 0.4508076310157776, 0.784140944480896, 0.5756241083145142, 0.7767988443374634], [0.024963289499282837, 0.030837005004286766, 0.041116006672382355, 0.049926578998565674, 0.044052861630916595, 0.04845815151929855, 0.051395006477832794, 0.041116006672382355, 0.07929515093564987, 0.05286343768239021, 0.05726872384548187, 0.08076358586549759, 0.07342144101858139, 0.08370044082403183, 0.06167401000857353, 0.08516886830329895, 0.07709251344203949, 0.14390602707862854, 0.1409691572189331, 0.14390602707862854, 0.09985315799713135, 0.08370044082403183, 0.12334802001714706, 0.13509544730186462, 0.14390602707862854, 0.12187959253787994, 0.17621144652366638, 0.09544786810874939, 0.15418502688407898, 0.1336270123720169, 0.17180617153644562, 0.1886930912733078, 0.1189427301287651, 0.15051394701004028, 0.14243759214878082, 0.2070484608411789, 0.1776798814535141, 0.20998531579971313, 0.1850220263004303, 0.15859030187129974, 0.14757709205150604, 0.16740088164806366, 0.1703377366065979, 0.19089573621749878, 0.19089573621749878, 0.13215859234333038, 0.05433186516165733, 0.21585902571678162, 0.2525697648525238, 0.19676946103572845, 0.26431718468666077, 0.21732746064662933, 0.3869309723377228, 0.1556534469127655, 0.27165931463241577, 0.33773863315582275, 0.3869309723377228, 0.3406754732131958, 0.3171806037425995, 0.2980910539627075, 0.28487518429756165, 0.30690160393714905, 0.3201174736022949, 0.35095447301864624, 0.3994126319885254, 0.3773861825466156, 0.5110132098197937, 0.45521292090415955, 0.4434655010700226, 0.4449339210987091, 0.4434655010700226, 0.619676947593689, NaN, 0.6328927874565125, 0.5814977884292603, 0.48972100019454956, 0.5829662084579468, 0.7400881052017212, 0.7591776847839355, 0.922173261642456], [0.041116006672382355, 0.06461086869239807, 0.06754772365093231, 0.055800292640924454, 0.06314244121313095, 0.06754772365093231, 0.07342144101858139, 0.06754772365093231, 0.08663729578256607, 0.11160058528184891, 0.11013215780258179, 0.07195301353931427, 0.14243759214878082, 0.1556534469127655, 0.09618208557367325, 0.09471365809440613, 0.151982381939888, 0.12041115760803223, 0.10866373032331467, 0.10279001295566559, 0.20264317095279694, 0.15271659195423126, 0.12334802001714706, 0.1189427301287651, 0.15418502688407898, 0.19089573621749878, 0.18208517134189606, 0.14390602707862854, 0.16079294681549072, 0.13142438232898712, 0.16886931657791138, 0.14977973699569702, 0.14243759214878082, 0.17327459156513214, 0.17327459156513214, 0.08957415819168091, 0.1703377366065979, 0.2540381848812103, 0.16519823670387268, 0.13950073719024658, 0.18355359137058258, 0.18428781628608704, 0.24082231521606445, 0.23935389518737793, 0.16446402668952942, 0.1923641711473465, 0.16886931657791138, 0.26431718468666077, 0.2621145248413086, 0.16519823670387268, 0.2804698944091797, 0.2709251046180725, 0.3788546323776245, 0.2195301055908203, 0.2687224745750427, 0.32158589363098145, 0.3788546323776245, 0.3245227634906769, 0.3098384737968445, 0.2547723948955536, 0.3054331839084625, 0.30102789402008057, 0.32158589363098145, 0.33773863315582275, 0.3465491831302643, 0.33406755328178406, 0.4522760510444641, 0.392070472240448, 0.36563876271247864, 0.36710718274116516, 0.4008810520172119, 0.3663729727268219, 0.36710718274116516, NaN, 0.45154184103012085, 0.3935389220714569, 0.44640234112739563, 0.7165932655334473, 0.5609397888183594, 0.769456684589386], [0.030837005004286766, 0.05286343768239021, 0.05726872384548187, 0.04845815151929855, 0.07195301353931427, 0.05433186516165733, 0.08370044082403183, 0.0969163030385971, 0.07929515093564987, 0.08810572326183319, 0.07048457860946655, 0.08370044082403183, 0.07856094092130661, 0.08076358586549759, 0.11306901276111603, 0.12334802001714706, 0.12775330245494843, 0.13950073719024658, 0.13803230226039886, 0.1262848824262619, 0.09838473051786423, 0.11160058528184891, 0.1189427301287651, 0.15124815702438354, 0.1189427301287651, 0.14390602707862854, 0.22320117056369781, 0.10792951285839081, 0.12922173738479614, 0.13656388223171234, 0.15859030187129974, 0.13215859234333038, 0.12481644749641418, 0.15859030187129974, 0.15124815702438354, 0.15124815702438354, 0.16226138174533844, 0.20998531579971313, 0.19676946103572845, 0.17180617153644562, 0.20411160588264465, 0.13950073719024658, 0.22026431560516357, 0.2378854602575302, 0.19530102610588074, 0.1776798814535141, 0.20117473602294922, 0.2569750249385834, 0.12334802001714706, 0.17474302649497986, 0.15271659195423126, 0.24596181511878967, 0.3421438932418823, 0.230543315410614, 0.3061673939228058, 0.2878120541572571, 0.34361234307289124, 0.2980910539627075, 0.22320117056369781, 0.30102789402008057, 0.2657856047153473, 0.26138031482696533, 0.2562408149242401, 0.3171806037425995, 0.40014684200286865, 0.35242289304733276, 0.41850221157073975, 0.4067547619342804, 0.4390602111816406, 0.32892805337905884, 0.35829663276672363, 0.5227606296539307, 0.41850221157073975, 0.5484581589698792, NaN, 0.39794421195983887, 0.5139500498771667, 0.6754772663116455, 0.8641703128814697, 0.9691630005836487], [0.016152717173099518, 0.017621144652366638, 0.032305434346199036, 0.038179147988557816, 0.041116006672382355, 0.026431718841195107, 0.051395006477832794, 0.051395006477832794, 0.05947136506438255, 0.032305434346199036, 0.039647575467824936, 0.06387665122747421, 0.051395006477832794, 0.07048457860946655, 0.07048457860946655, 0.09397944062948227, 0.08370044082403183, 0.13876652717590332, 0.12187959253787994, 0.12408223003149033, 0.10572687536478043, 0.09985315799713135, 0.06167401000857353, 0.12922173738479614, 0.06754772365093231, 0.08370044082403183, 0.17914831638336182, 0.05433186516165733, 0.07782672345638275, 0.08810572326183319, 0.07782672345638275, 0.17547723650932312, 0.10425844043493271, 0.14243759214878082, 0.11600587517023087, 0.2195301055908203, 0.09544786810874939, 0.17327459156513214, 0.17474302649497986, 0.18942731618881226, 0.23201175034046173, 0.16152717173099518, 0.17914831638336182, 0.20117473602294922, 0.21292217075824738, 0.20264317095279694, 0.2437591850757599, 0.2364170402288437, 0.12187959253787994, 0.24229075014591217, 0.16226138174533844, 0.27606460452079773, 0.37298092246055603, 0.2804698944091797, 0.27753305435180664, 0.3494860529899597, 0.3715124726295471, 0.3465491831302643, 0.1189427301287651, 0.3663729727268219, 0.257709264755249, 0.21732746064662933, 0.2547723948955536, 0.36710718274116516, 0.39794421195983887, 0.426578551530838, 0.5447870492935181, 0.458149790763855, 0.4287812113761902, 0.453744500875473, 0.4287812113761902, 0.5859031081199646, 0.5102789998054504, 0.6064611077308655, 0.6020557880401611, NaN, 0.5286343693733215, 0.700440526008606, 0.7444933652877808, 0.9353891611099243], [0.014684287831187248, 0.026431718841195107, 0.030837005004286766, 0.020558003336191177, 0.030837005004286766, 0.032305434346199036, 0.030837005004286766, 0.024963289499282837, 0.042584434151649475, 0.029368575662374496, 0.036710720509290695, 0.035242289304733276, 0.051395006477832794, 0.05286343768239021, 0.04845815151929855, 0.039647575467824936, 0.045521292835474014, 0.09838473051786423, 0.11306901276111603, 0.10425844043493271, 0.09104258567094803, 0.06754772365093231, 0.04845815151929855, 0.06167401000857353, 0.08076358586549759, 0.08076358586549759, 0.15418502688407898, 0.08076358586549759, 0.07635829597711563, 0.05433186516165733, 0.09251101315021515, 0.15271659195423126, 0.09544786810874939, 0.12481644749641418, 0.08663729578256607, 0.16740088164806366, 0.09397944062948227, 0.1703377366065979, 0.12775330245494843, 0.08957415819168091, 0.1262848824262619, 0.15271659195423126, 0.14977973699569702, 0.13215859234333038, 0.11013215780258179, 0.13509544730186462, 0.17914831638336182, 0.14170338213443756, 0.2364170402288437, 0.13950073719024658, 0.24963289499282837, 0.1923641711473465, 0.3553597629070282, 0.23935389518737793, 0.18942731618881226, 0.3700440526008606, 0.3553597629070282, 0.3715124726295471, 0.2966226041316986, 0.2599118947982788, 0.22026431560516357, 0.27019089460372925, 0.3171806037425995, 0.35242289304733276, 0.1776798814535141, 0.3861967623233795, 0.4684287905693054, 0.4140969216823578, 0.28854626417160034, 0.4361233413219452, 0.453744500875473, 0.5491923689842224, 0.41703376173973083, 0.553597629070282, 0.48604992032051086, 0.47136563062667847, NaN, 0.6380323171615601, 0.6387665271759033, 0.8428781032562256], [0.029368575662374496, 0.038179147988557816, 0.047723934054374695, 0.035976506769657135, 0.055066078901290894, 0.046989720314741135, 0.049192365258932114, 0.035242289304733276, 0.07195301353931427, 0.09544786810874939, 0.051395006477832794, 0.042584434151649475, 0.09985315799713135, 0.10279001295566559, 0.07048457860946655, 0.07488986849784851, 0.12041115760803223, 0.0969163030385971, 0.08516886830329895, 0.06607929617166519, 0.11453744769096375, 0.07415565103292465, 0.07635829597711563, 0.08957415819168091, 0.12481644749641418, 0.14390602707862854, 0.16152717173099518, 0.115271657705307, 0.13069015741348267, 0.08810572326183319, 0.13215859234333038, 0.07709251344203949, 0.09838473051786423, 0.13215859234333038, 0.11306901276111603, 0.055800292640924454, 0.13803230226039886, 0.22026431560516357, 0.12775330245494843, 0.11160058528184891, 0.11600587517023087, 0.10425844043493271, 0.18061673641204834, 0.1556534469127655, 0.13069015741348267, 0.12334802001714706, 0.12041115760803223, 0.17180617153644562, 0.18942731618881226, 0.10866373032331467, 0.19676946103572845, 0.14537444710731506, 0.29368576407432556, 0.1483113020658493, 0.17327459156513214, 0.21879589557647705, 0.29368576407432556, 0.21879589557647705, 0.2268722504377365, 0.16152717173099518, 0.19823788106441498, 0.1997063159942627, 0.23348017036914825, 0.22026431560516357, 0.2819383144378662, 0.2364170402288437, 0.3421438932418823, 0.2951541841030121, 0.29955947399139404, 0.27165931463241577, 0.2951541841030121, 0.21585902571678162, 0.2599118947982788, 0.2834067642688751, 0.3245227634906769, 0.29955947399139404, 0.36196768283843994, NaN, 0.4757709205150604, 0.7224669456481934], [0.033773861825466156, 0.041116006672382355, 0.042584434151649475, 0.042584434151649475, 0.06167401000857353, 0.046989720314741135, 0.05873715132474899, 0.06314244121313095, 0.06461086869239807, 0.06461086869239807, 0.051395006477832794, 0.06240822374820709, 0.06754772365093231, 0.07488986849784851, 0.07488986849784851, 0.08663729578256607, 0.09985315799713135, 0.11160058528184891, 0.10572687536478043, 0.0969163030385971, 0.07635829597711563, 0.051395006477832794, 0.07856094092130661, 0.11013215780258179, 0.08663729578256607, 0.10792951285839081, 0.1336270123720169, 0.08076358586549759, 0.08370044082403183, 0.08810572326183319, 0.10205580294132233, 0.09397944062948227, 0.08663729578256607, 0.11306901276111603, 0.08076358586549759, 0.07929515093564987, 0.10719530284404755, 0.1776798814535141, 0.1189427301287651, 0.11967694759368896, 0.12775330245494843, 0.10866373032331467, 0.151982381939888, 0.13656388223171234, 0.12041115760803223, 0.11160058528184891, 0.1262848824262619, 0.1629955917596817, 0.09104258567094803, 0.10866373032331467, 0.08663729578256607, 0.17180617153644562, 0.21732746064662933, 0.15859030187129974, 0.15859030187129974, 0.16593244671821594, 0.21879589557647705, 0.17180617153644562, 0.13656388223171234, 0.17621144652366638, 0.18061673641204834, 0.18649046123027802, 0.14537444710731506, 0.20558002591133118, 0.2599118947982788, 0.22320117056369781, 0.303964763879776, 0.2819383144378662, 0.2819383144378662, 0.21879589557647705, 0.23201175034046173, 0.4243759214878082, 0.24082231521606445, 0.4390602111816406, 0.13582965731620789, 0.25550660490989685, 0.3612334728240967, 0.5242290496826172, NaN, 0.9265785813331604], [0.007342143915593624, 0.005873715039342642, 0.007342143915593624, 0.005873715039342642, 0.010279001668095589, 0.008810572326183319, 0.016152717173099518, 0.017621144652366638, 0.007342143915593624, 0.007342143915593624, 0.011747430078685284, 0.011747430078685284, 0.007342143915593624, 0.010279001668095589, 0.011747430078685284, 0.029368575662374496, 0.020558003336191177, 0.027900146320462227, 0.023494860157370567, 0.032305434346199036, 0.008810572326183319, 0.007342143915593624, 0.016152717173099518, 0.038179147988557816, 0.014684287831187248, 0.019089573994278908, 0.017621144652366638, 0.013215859420597553, 0.016152717173099518, 0.022026430815458298, 0.024963289499282837, 0.032305434346199036, 0.016152717173099518, 0.026431718841195107, 0.019089573994278908, 0.036710720509290695, 0.027900146320462227, 0.049926578998565674, 0.05726872384548187, 0.044052861630916595, 0.04845815151929855, 0.023494860157370567, 0.045521292835474014, 0.031571220606565475, 0.051395006477832794, 0.024963289499282837, 0.05433186516165733, 0.045521292835474014, 0.010279001668095589, 0.04845815151929855, 0.011747430078685284, 0.06314244121313095, 0.044052861630916595, 0.06461086869239807, 0.046989720314741135, 0.07635829597711563, 0.044052861630916595, 0.07635829597711563, 0.033773861825466156, 0.044052861630916595, 0.05359765142202377, 0.06607929617166519, 0.035242289304733276, 0.08370044082403183, 0.07195301353931427, 0.06901615113019943, 0.11013215780258179, 0.07782672345638275, 0.0969163030385971, 0.11453744769096375, 0.033773861825466156, 0.22320117056369781, 0.07782672345638275, 0.230543315410614, 0.030837005004286766, 0.06461086869239807, 0.15712188184261322, 0.27753305435180664, 0.07342144101858139, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9143943221},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.849504637},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8345322392},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8325433546},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8225245855},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8183330554},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7933418845},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7923381479},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7340749825},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7272532988},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7250878262},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.7234614067},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7094555651},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6847989774},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6767040264},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6617966877},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6563690797},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.6474283905},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6434227389},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.6393613289},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.6303648748},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6300024147},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6174464211},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6056246371},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.597892155},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.586916113},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5854198025},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5839141989},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5822227172},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5799271362},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5520084025},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.5514786556},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5499265782},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5431885351},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5408650704},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.53459172},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5329188259},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5206509412},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5194613297},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5069239214},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.491161549},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4889310237},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4794327021},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4756686935},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4693024035},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4689120651},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4605383006},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4520251319},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4520158366},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4418762426},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.427870405},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.425212364},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.4080187363},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4058253847},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4035019238},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3993475692},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.3982416028},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3979627856},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3730924393},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3688637342},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3583802666},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3552761205},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3349411685},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3059071732},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.305702708},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.3051822512},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.293165302},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2842803764},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2745404171},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.263759551},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.26011636},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2478763526},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2373557119},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2292514726},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2288611306},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.228424321},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1983587052},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1637483964},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.152707299},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0458930467}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_all/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_all/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..68ab5e9c9f97f2934d51143610fc76d516323f97
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_all/datasets_small/winrate_matrix.csv
@@ -0,0 +1,81 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (default),TabPFNv2 (tuned + ensembled),RealMLP (tuned + ensembled),TabDPT-Turbo (default),Mitra (default),TabDPT (tuned + ensembled),TabDPT (tuned),"AutoGluon 1.4 (best, 4h)",TabM (tuned + ensembled),RealMLP (tuned),TabPFNv2 (tuned),LimiX (default),TabPFNv2 (default),TabDPT (default),TabSwift (default),LightGBM (tuned + ensembled),TabM (tuned),CatBoost (tuned + ensembled),iLTM (tuned + ensembled),SAP-RPT-OSS (default),xRFM (tuned + ensembled),CatBoost (tuned),LightGBM (tuned),XGBoost (tuned + ensembled),TabICL (default),xRFM (tuned),ModernNCA (tuned),CatBoost (default),BetaTabPFN (default),XGBoost (tuned),ModernNCA (tuned + ensembled),TabM (default),EBM (tuned + ensembled),TorchMLP (tuned + ensembled),ModernNCA (default),iLTM (tuned),ChimeraBoost (tuned + ensembled),EBM (tuned),RealMLP (default),FastaiMLP (tuned + ensembled),ChimeraBoost (tuned),ExtraTrees (tuned + ensembled),EBM (default),ExtraTrees (tuned),TorchMLP (tuned),Nori-30M (default),FastaiMLP (tuned),ChimeraBoost (default),TabSTAR (tuned),Nori (default),TabSTAR (tuned + ensembled),RandomForest (tuned + ensembled),OrionMSP (default),LightGBM (default),XGBoost (default),RandomForest (tuned),TorchMLP (default),PerpetualBooster (tuned + ensembled),TabFlex (default),TabSTAR (default),xRFM (default),PerpetualBooster (tuned),iLTM (default),ExtraTrees (default),Linear (tuned + ensembled),FastaiMLP (default),Linear (tuned),KNN (tuned + ensembled),RandomForest (default),PerpetualBooster (default),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.76798826,0.76798826,0.7900147,0.76284873,0.79441994,0.77386194,0.8061674,0.8575624,0.8619677,0.8634361,0.8737151,0.8839941,0.88986784,0.9162996,0.8854626,0.8722467,0.87077826,0.9067548,0.8634361,0.89133626,0.86930984,0.917768,0.91189426,0.928047,0.9295154,0.8575624,0.9192364,0.9368576,0.917768,0.938326,0.948605,0.91042584,0.92511016,0.93538916,0.8935389,0.94126284,0.94566816,0.9192364,0.9133627,0.95594716,0.9265786,0.9441997,0.938326,0.92511016,0.9368576,0.9236417,0.9441997,0.94566816,0.92217326,0.9397944,0.96328926,0.85903084,0.94273126,0.958884,0.8472834,0.9192364,0.845815,0.96622616,0.9471366,0.96622616,0.969163,0.9647577,0.9706314,0.98384726,0.9441997,0.8854626,0.9706314,0.98237884,0.948605,0.98531574,0.95301026,0.97503674,0.958884,0.969163,0.98384726,0.98531574,0.9706314,0.96622616,0.99265784
+TabPFN-3 (default),0.23201175,,0.56828195,0.53450805,0.5770925,0.6079295,0.6358297,0.66519827,0.73568285,0.7415565,0.74596184,0.7283407,0.76358294,0.7856094,0.7709251,0.7856094,0.7812041,0.80837005,0.76798826,0.78340673,0.81791484,0.8634361,0.83994126,0.8237885,0.8546255,0.87958884,0.76651984,0.8634361,0.8634361,0.86049926,0.8656388,0.88032305,0.8634361,0.8678414,0.87958884,0.80396473,0.8737151,0.8825257,0.845815,0.8619677,0.9148311,0.88913363,0.9287812,0.9060206,0.876652,0.917768,0.8744493,0.907489,0.9265786,0.8920705,0.928047,0.93098384,0.8414097,0.9140969,0.9192364,0.8215859,0.8678414,0.81938326,0.9339207,0.9339207,0.95301026,0.96622616,0.9500734,0.96328926,0.9500734,0.9368576,0.8678414,0.9647577,0.9603524,0.9544787,0.9706314,0.9295154,0.969163,0.93538916,0.9471366,0.98237884,0.97356826,0.96182084,0.958884,0.99412626
+"AutoGluon 1.5 (extreme, 4h)",0.23201175,0.43171805,,0.47136563,0.5227606,0.5976505,0.5770925,0.6049926,0.6593245,0.75403816,0.66079295,0.6938326,0.7386197,0.75770926,0.8017621,0.8164464,0.8002937,0.7591777,0.7606461,0.7709251,0.7488987,0.8105727,0.8428781,0.82452273,0.8472834,0.8781204,0.73274595,0.8575624,0.8524229,0.8546255,0.8722467,0.8509545,0.8450808,0.88105726,0.86857563,0.8061674,0.8751836,0.89721,0.86049926,0.86930984,0.91189426,0.8781204,0.9133627,0.91042584,0.8781204,0.8942731,0.8781204,0.917768,0.9089574,0.87958884,0.90014684,0.9368576,0.77679884,0.90014684,0.92070484,0.8032305,0.81938326,0.8032305,0.93245226,0.9375918,0.95301026,0.958884,0.9368576,0.96328926,0.95594716,0.9236417,0.84875184,0.95154184,0.96108663,0.9500734,0.9500734,0.9133627,0.958884,0.93245226,0.94273126,0.9676946,0.969163,0.95227605,0.9574156,0.99265784
+TabICLv2 (default),0.20998532,0.46549192,0.52863437,,0.53450805,0.5961821,0.58002937,0.61894274,0.6901615,0.7298091,0.7209985,0.70117474,0.76578563,0.75770926,0.79441994,0.7958884,0.76651984,0.7511013,0.7687225,0.7621145,0.78707784,0.8164464,0.82819384,0.80763584,0.83847284,0.8737151,0.74743026,0.8428781,0.8443465,0.85022026,0.8531571,0.8773862,0.83480173,0.8575624,0.87077826,0.79882526,0.856094,0.8634361,0.835536,0.86049926,0.8942731,0.88032305,0.8957416,0.89133626,0.8634361,0.8942731,0.8649045,0.9030837,0.87885463,0.88986784,0.88105726,0.9133627,0.8061674,0.89721,0.91189426,0.8164464,0.8428781,0.8105727,0.9265786,0.9236417,0.9397944,0.93245226,0.928047,0.938326,0.948605,0.9133627,0.87077826,0.9471366,0.95301026,0.938326,0.9500734,0.9397944,0.9500734,0.9441997,0.95154184,0.96182084,0.979442,0.9640235,0.9574156,0.99412626
+RealTabPFN-2.5 (tuned + ensembled),0.23715125,0.4229075,0.47723934,0.46549192,,0.5315712,0.6358297,0.6358297,0.7180617,0.71218795,0.6226138,0.6945668,0.66519827,0.6842878,0.7826725,0.7900147,0.7591777,0.80910426,0.75624084,0.7900147,0.6784141,0.7753304,0.825257,0.81938326,0.83994126,0.82745963,0.75697505,0.82819384,0.84875184,0.8450808,0.8619677,0.77753305,0.8311307,0.8634361,0.8737151,0.8267254,0.8744493,0.8546255,0.8575624,0.8634361,0.9030837,0.85903084,0.8986784,0.8964758,0.8773862,0.8825257,0.8773862,0.9060206,0.8575624,0.8737151,0.86049926,0.928047,0.7709251,0.907489,0.9162996,0.7958884,0.8135095,0.7958884,0.9133627,0.9162996,0.948605,0.9500734,0.92511016,0.9544787,0.95301026,0.9162996,0.8370044,0.9339207,0.95301026,0.94273126,0.948605,0.92511016,0.95594716,0.9368576,0.928047,0.958884,0.969163,0.9449339,0.938326,0.989721
+TabPFN-2.6 (default),0.20558003,0.39207047,0.40234947,0.40381792,0.4684288,,0.52349484,0.5814978,0.62995595,0.69162995,0.6284875,0.6975037,0.6593245,0.6769457,0.7694567,0.78414094,0.75770926,0.7518355,0.7312775,0.7276065,0.7004405,0.77679884,0.8223201,0.8135095,0.845815,0.86930984,0.75624084,0.8267254,0.8472834,0.8575624,0.8751836,0.8002937,0.83847284,0.856094,0.89133626,0.79735684,0.88105726,0.8957416,0.8524229,0.85022026,0.8957416,0.88105726,0.92070484,0.9192364,0.8744493,0.89133626,0.8649045,0.9236417,0.8928047,0.86930984,0.8876652,0.928047,0.7753304,0.8986784,0.93538916,0.80469894,0.81497794,0.80910426,0.9236417,0.8957416,0.9574156,0.96622616,0.9397944,0.95594716,0.948605,0.92070484,0.8428781,0.948605,0.94566816,0.94273126,0.95301026,0.9148311,0.95154184,0.93245226,0.94566816,0.97356826,0.9676946,0.95301026,0.95301026,0.9911894
+RealTabPFN-2.5 (tuned),0.22613803,0.36417034,0.4229075,0.41997063,0.36417034,0.47650513,,0.54919237,0.6585903,0.66226137,0.5917768,0.65198237,0.6182085,0.6468429,0.7298091,0.7503671,0.72246695,0.77312773,0.71512485,0.7503671,0.6578561,0.7393539,0.77386194,0.79809105,0.7958884,0.7826725,0.72540385,0.7995595,0.8017621,0.8135095,0.82819384,0.7518355,0.8135095,0.8105727,0.8267254,0.80763584,0.83994126,0.82745963,0.83994126,0.84875184,0.8825257,0.835536,0.88105726,0.8568282,0.85903084,0.87077826,0.86049926,0.8737151,0.8524229,0.86049926,0.85903084,0.91042584,0.7488987,0.90014684,0.8942731,0.76798826,0.78854626,0.7723935,0.886931,0.9030837,0.9397944,0.9368576,0.9030837,0.958884,0.93024963,0.90014684,0.82819384,0.92511016,0.94126284,0.9397944,0.9295154,0.9148311,0.948605,0.9265786,0.9162996,0.948605,0.969163,0.95080763,0.94126284,0.98384726
+RealTabPFN-2.5 (default),0.1938326,0.33480176,0.39500734,0.38105726,0.36417034,0.4185022,0.45080763,,0.60058737,0.6255507,0.5844346,0.6475771,0.5932452,0.6262849,0.7195301,0.73421437,0.7092511,0.7209985,0.7283407,0.73715127,0.6637298,0.74302495,0.7753304,0.77826726,0.80763584,0.7826725,0.7298091,0.80469894,0.8223201,0.8237885,0.82966226,0.7650514,0.8325991,0.8164464,0.856094,0.80396473,0.8531571,0.8105727,0.8516887,0.85903084,0.8883994,0.8531571,0.8876652,0.876652,0.87958884,0.87077826,0.8928047,0.8825257,0.8546255,0.886931,0.866373,0.91189426,0.7400881,0.9133627,0.91042584,0.79735684,0.74302495,0.79662263,0.8854626,0.9236417,0.9368576,0.93538916,0.92070484,0.97209984,0.938326,0.9185022,0.845815,0.9339207,0.94566816,0.9500734,0.93245226,0.9265786,0.958884,0.93245226,0.9030837,0.948605,0.97503674,0.9647577,0.9368576,0.98237884
+TabPFNv2 (tuned + ensembled),0.14243759,0.26431718,0.34067547,0.30983847,0.2819383,0.37004405,0.34140968,0.39941263,,0.55726874,0.5212922,0.530837,0.53450805,0.561674,0.6563877,0.6681351,0.6453745,0.6762115,0.5814978,0.6784141,0.58002937,0.6358297,0.70264316,0.71585906,0.71512485,0.73274595,0.62922174,0.75770926,0.72320116,0.73715127,0.76431715,0.68061674,0.76431715,0.7709251,0.76798826,0.75477237,0.7812041,0.7812041,0.75330395,0.77753305,0.8002937,0.78487515,0.8237885,0.80469894,0.79735684,0.81791484,0.80837005,0.8164464,0.81424373,0.825257,0.825257,0.84875184,0.7209985,0.82745963,0.85022026,0.7386197,0.7591777,0.7400881,0.8825257,0.82452273,0.9133627,0.9089574,0.9030837,0.907489,0.89721,0.8612335,0.79735684,0.92070484,0.92217326,0.90161526,0.95154184,0.90014684,0.92070484,0.9133627,0.92070484,0.94052863,0.9574156,0.928047,0.93538916,0.99265784
+RealMLP (tuned + ensembled),0.1380323,0.25844347,0.24596182,0.2701909,0.28781205,0.30837005,0.33773863,0.37444934,0.4427313,,0.4787078,0.49779737,0.4735683,0.49853158,0.57782674,0.61967695,0.6791483,0.55947137,0.5315712,0.57856095,0.5565345,0.5638766,0.72246695,0.6901615,0.7591777,0.7606461,0.62114537,0.7488987,0.7606461,0.7687225,0.8267254,0.62995595,0.74302495,0.7312775,0.78854626,0.6975037,0.83847284,0.73715127,0.7914831,0.7929515,0.7709251,0.7753304,0.8414097,0.81938326,0.80469894,0.8817915,0.8201175,0.8259912,0.8311307,0.82745963,0.8450808,0.8208517,0.76651984,0.8362702,0.84875184,0.7386197,0.7958884,0.74302495,0.91042584,0.80910426,0.92217326,0.93538916,0.93538916,0.92511016,0.93538916,0.8472834,0.80469894,0.9265786,0.94273126,0.9192364,0.95301026,0.886931,0.9471366,0.8883994,0.91189426,0.9676946,0.9706314,0.9045521,0.93538916,0.99265784
+TabDPT-Turbo (default),0.13656388,0.25403818,0.33920705,0.27900147,0.37738618,0.37151247,0.4082232,0.41556534,0.4787078,0.5212922,,0.5198238,0.52422905,0.5653451,0.5932452,0.5851689,0.592511,0.5624082,0.6358297,0.5814978,0.68502206,0.6879589,0.6578561,0.6387665,0.69162995,0.7298091,0.6387665,0.7004405,0.6886931,0.69309837,0.7209985,0.70190895,0.73568285,0.7312775,0.72393537,0.67547727,0.73274595,0.7312775,0.7180617,0.73274595,0.77826726,0.8032305,0.82452273,0.78854626,0.76578563,0.8061674,0.8237885,0.8032305,0.84948605,0.79735684,0.86930984,0.82966226,0.7400881,0.86930984,0.8340675,0.73274595,0.74743026,0.73568285,0.8883994,0.8928047,0.886931,0.8928047,0.89133626,0.91189426,0.90014684,0.928047,0.81497794,0.90014684,0.92070484,0.8986784,0.9449339,0.8825257,0.94273126,0.88986784,0.9295154,0.9603524,0.96328926,0.948605,0.948605,0.9882526
+Mitra (default),0.12628488,0.2716593,0.3061674,0.29882526,0.30543318,0.3024963,0.34801763,0.3524229,0.469163,0.50220263,0.4801762,,0.5051395,0.51248163,0.6152717,0.6284875,0.6035242,0.56828195,0.592511,0.58810574,0.5712188,0.60939795,0.6872247,0.7092511,0.68575627,0.7202643,0.6182085,0.68135095,0.68502206,0.7312775,0.73641706,0.69089574,0.7173275,0.76651984,0.76798826,0.7349486,0.74743026,0.7606461,0.7503671,0.8017621,0.85022026,0.8120411,0.8061674,0.8135095,0.82819384,0.7958884,0.82819384,0.825257,0.79882526,0.835536,0.8098385,0.8942731,0.7092511,0.866373,0.85022026,0.78707784,0.7298091,0.7826725,0.845815,0.82966226,0.90014684,0.917768,0.87958884,0.92070484,0.89133626,0.8759178,0.81791484,0.9045521,0.9089574,0.92511016,0.90014684,0.9148311,0.9192364,0.928047,0.9162996,0.9361234,0.9647577,0.9574156,0.9375918,0.9882526
+TabDPT (tuned + ensembled),0.116005875,0.23641704,0.2613803,0.2342144,0.33480176,0.34067547,0.38179147,0.40675476,0.46549192,0.52643174,0.47577092,0.4948605,,0.60939795,0.5932452,0.5807636,0.6057269,0.571953,0.5734214,0.56828195,0.6549192,0.63069016,0.6578561,0.6262849,0.67107195,0.7386197,0.6270191,0.6982379,0.68281937,0.6886931,0.71292216,0.643906,0.7187959,0.72393537,0.73274595,0.6461087,0.7400881,0.7591777,0.7070485,0.72320116,0.75550663,0.76725405,0.8208517,0.7826725,0.74229074,0.7826725,0.7753304,0.79882526,0.8237885,0.75770926,0.8325991,0.80910426,0.7650514,0.8164464,0.825257,0.7209985,0.77679884,0.72246695,0.8883994,0.8531571,0.8825257,0.8825257,0.90014684,0.9030837,0.89721,0.8737151,0.8032305,0.91116005,0.9082232,0.8942731,0.94273126,0.8414097,0.9265786,0.8575624,0.92143905,0.948605,0.948605,0.90014684,0.93245226,0.99265784
+TabDPT (tuned),0.11013216,0.2143906,0.24229075,0.24229075,0.31571218,0.32305434,0.35315713,0.37371513,0.438326,0.5014684,0.43465492,0.48751837,0.39060205,,0.5829662,0.55947137,0.5638766,0.5323054,0.5550661,0.530837,0.62041116,0.5961821,0.6284875,0.6123348,0.64170337,0.68281937,0.6049926,0.6563877,0.6527166,0.6637298,0.6997063,0.6218796,0.6784141,0.69603527,0.6975037,0.6270191,0.7077827,0.7276065,0.69162995,0.70337737,0.74596184,0.74596184,0.7856094,0.7591777,0.72687227,0.7591777,0.75330395,0.7709251,0.7856094,0.74743026,0.79735684,0.7812041,0.72687227,0.79735684,0.81497794,0.71145374,0.7503671,0.7092511,0.8414097,0.8575624,0.8575624,0.8443465,0.8619677,0.88986784,0.87958884,0.8678414,0.8002937,0.8854626,0.8883994,0.8751836,0.9162996,0.8333333,0.9162996,0.8443465,0.9192364,0.9295154,0.9471366,0.89721,0.92511016,0.989721
+"AutoGluon 1.4 (best, 4h)",0.08370044,0.2290749,0.19823788,0.20558003,0.21732746,0.23054332,0.2701909,0.2804699,0.34361234,0.42217326,0.40675476,0.38472834,0.40675476,0.41703376,,0.5227606,0.5315712,0.45080763,0.4845815,0.47723934,0.50220263,0.5293686,0.6901615,0.5947136,0.6593245,0.6637298,0.5535976,0.6468429,0.6798825,0.7298091,0.7415565,0.6079295,0.69162995,0.67254037,0.7386197,0.66079295,0.7386197,0.69309837,0.70631427,0.73421437,0.79441994,0.7503671,0.7900147,0.78634363,0.75770926,0.7621145,0.79441994,0.7958884,0.7415565,0.8002937,0.76798826,0.8325991,0.7004405,0.83847284,0.835536,0.73274595,0.7048458,0.73568285,0.82966226,0.77753305,0.9148311,0.9030837,0.8612335,0.907489,0.9265786,0.8516887,0.79882526,0.8854626,0.9295154,0.90014684,0.917768,0.8832599,0.938326,0.9038179,0.886931,0.9295154,0.95154184,0.9295154,0.92511016,0.9882526
+TabM (tuned + ensembled),0.11453745,0.2143906,0.18355359,0.2041116,0.20998532,0.21585903,0.2496329,0.2657856,0.3318649,0.38032305,0.41483113,0.37151247,0.41923642,0.44052863,0.47723934,,0.4948605,0.44493392,0.4698972,0.45007342,0.51248163,0.49632892,0.6049926,0.70190895,0.61967695,0.61894274,0.5756241,0.6035242,0.64170337,0.64977974,0.7048458,0.5947136,0.6328928,0.6453745,0.7180617,0.6769457,0.73568285,0.6740088,0.83773863,0.73715127,0.74743026,0.7180617,0.73421437,0.7621145,0.76651984,0.7907489,0.75477237,0.7958884,0.71365637,0.7790015,0.71659327,0.8201175,0.6872247,0.80910426,0.845815,0.7283407,0.70190895,0.7312775,0.7797357,0.7694567,0.8729809,0.886931,0.8208517,0.91042584,0.87958884,0.82966226,0.7958884,0.8531571,0.90014684,0.9089574,0.87077826,0.8928047,0.9148311,0.9052864,0.876652,0.9060206,0.9603524,0.92511016,0.9133627,0.9706314
+RealMLP (tuned),0.1277533,0.2187959,0.19970632,0.23348017,0.24082232,0.24229075,0.27753305,0.2907489,0.35462555,0.32085168,0.40748897,0.39647576,0.39427313,0.43612334,0.4684288,0.5051395,,0.4596182,0.47063142,0.4948605,0.5051395,0.5007342,0.6049926,0.5976505,0.6284875,0.6116006,0.55800295,0.6049926,0.6372981,0.65051395,0.7077827,0.5770925,0.6468429,0.6461087,0.6872247,0.6578561,0.7246696,0.6637298,0.71512485,0.71512485,0.72540385,0.72393537,0.74596184,0.7209985,0.7503671,0.7797357,0.74449337,0.7518355,0.7518355,0.74743026,0.7694567,0.7914831,0.69309837,0.8002937,0.78707784,0.72246695,0.71365637,0.72393537,0.835536,0.7599119,0.85903084,0.86930984,0.85903084,0.8957416,0.89133626,0.8333333,0.7856094,0.8678414,0.8957416,0.886931,0.8751836,0.8414097,0.9229075,0.84801763,0.8722467,0.9162996,0.9544787,0.87958884,0.90014684,0.979442
+TabPFNv2 (tuned),0.12922174,0.19162996,0.24082232,0.24889868,0.19089574,0.24816446,0.22687225,0.27900147,0.32378855,0.44052863,0.4375918,0.43171805,0.428047,0.46769458,0.54919237,0.5550661,0.5403818,,0.50807637,0.54405284,0.50660795,0.5330396,0.61086637,0.6182085,0.6226138,0.5961821,0.54698974,0.61967695,0.6549192,0.6446402,0.66005874,0.6174743,0.6328928,0.67254037,0.67107195,0.6879589,0.66519827,0.6696035,0.6842878,0.70337737,0.72687227,0.70117474,0.72320116,0.71218795,0.7298091,0.7393539,0.7290749,0.73568285,0.6975037,0.74596184,0.7312775,0.77679884,0.6446402,0.76358294,0.7709251,0.71218795,0.6666667,0.7077827,0.7503671,0.7511013,0.80469894,0.79735684,0.79809105,0.8414097,0.80763584,0.81864905,0.76651984,0.8546255,0.8215859,0.8722467,0.8546255,0.87077826,0.856094,0.87958884,0.86049926,0.8612335,0.90161526,0.9030837,0.8883994,0.97209984
+LimiX (default),0.09324523,0.23201175,0.2393539,0.23127754,0.24375919,0.26872247,0.28487518,0.2716593,0.4185022,0.4684288,0.36417034,0.40748897,0.42657855,0.44493392,0.5154185,0.5301028,0.5293686,0.49192363,,0.50220263,0.46696034,0.4992658,0.5976505,0.55726874,0.58883995,0.62922174,0.5359765,0.5947136,0.5917768,0.61967695,0.6358297,0.6064611,0.6013216,0.67254037,0.67107195,0.6343612,0.65198237,0.6593245,0.6182085,0.6975037,0.6872247,0.7092511,0.68281937,0.73421437,0.7298091,0.6989721,0.6997063,0.7503671,0.7415565,0.73421437,0.7415565,0.73274595,0.71218795,0.76284873,0.76651984,0.7378855,0.72393537,0.73274595,0.8002937,0.7790015,0.8105727,0.8208517,0.81791484,0.81938326,0.8237885,0.7723935,0.78340673,0.8634361,0.8325991,0.85903084,0.8370044,0.8751836,0.85903084,0.89133626,0.8619677,0.8781204,0.886931,0.9148311,0.8942731,0.97650516
+TabPFNv2 (default),0.13656388,0.21659325,0.2290749,0.23788546,0.20998532,0.27239352,0.2496329,0.26284876,0.3215859,0.42143905,0.4185022,0.41189426,0.43171805,0.469163,0.5227606,0.5499266,0.5051395,0.45594713,0.49779737,,0.4889868,0.5110132,0.60058737,0.5947136,0.60939795,0.5859031,0.5227606,0.62114537,0.6226138,0.62114537,0.6372981,0.6079295,0.65051395,0.6343612,0.6453745,0.6674009,0.6490455,0.61967695,0.6696035,0.6872247,0.68575627,0.69089574,0.67547727,0.67107195,0.71365637,0.72540385,0.71218795,0.68575627,0.7209985,0.7312775,0.73421437,0.73421437,0.6563877,0.75477237,0.69603527,0.72687227,0.6740088,0.72320116,0.7694567,0.7892805,0.78854626,0.77826726,0.81497794,0.84875184,0.81791484,0.8656388,0.77679884,0.856094,0.82819384,0.8472834,0.84875184,0.8964758,0.856094,0.89721,0.8737151,0.8759178,0.8957416,0.9339207,0.9030837,0.9676946
+TabDPT (default),0.10866373,0.18208517,0.25110132,0.21292217,0.3215859,0.29955947,0.3421439,0.33627018,0.41997063,0.4434655,0.31497797,0.4287812,0.34508076,0.37958884,0.49779737,0.48751837,0.4948605,0.49339208,0.5330396,0.5110132,,0.51688695,0.5330396,0.52349484,0.5712188,0.5961821,0.5770925,0.59030837,0.5829662,0.56975037,0.6167401,0.5602056,0.6020558,0.6328928,0.5961821,0.5668135,0.6372981,0.6372981,0.6328928,0.6035242,0.66226137,0.6886931,0.6945668,0.6798825,0.6277533,0.654185,0.6886931,0.6842878,0.7518355,0.6798825,0.7723935,0.71218795,0.6901615,0.74302495,0.7400881,0.67107195,0.7048458,0.67254037,0.77826726,0.8105727,0.7914831,0.7650514,0.81791484,0.82966226,0.8370044,0.88105726,0.7812041,0.8443465,0.86049926,0.85022026,0.9060206,0.77386194,0.90014684,0.79735684,0.90161526,0.8942731,0.9089574,0.8854626,0.9236417,0.9911894
+TabSwift (default),0.13069016,0.13656388,0.18942732,0.18355359,0.2246696,0.22320117,0.2606461,0.25697502,0.36417034,0.43612334,0.3120411,0.39060205,0.36930984,0.40381792,0.47063142,0.50367105,0.4992658,0.46696034,0.5007342,0.4889868,0.48311308,,0.51395005,0.54919237,0.52863437,0.5638766,0.5653451,0.5535976,0.561674,0.54919237,0.5829662,0.52716595,0.59911895,0.6255507,0.5976505,0.5961821,0.6255507,0.61086637,0.6431718,0.6372981,0.70337737,0.7004405,0.7092511,0.65051395,0.67254037,0.74743026,0.7709251,0.68281937,0.6975037,0.6945668,0.71512485,0.75477237,0.6784141,0.79441994,0.76358294,0.7415565,0.71365637,0.74302495,0.7503671,0.87077826,0.8215859,0.7914831,0.79441994,0.87958884,0.866373,0.9045521,0.8120411,0.8414097,0.88105726,0.85903084,0.8575624,0.82819384,0.9162996,0.8472834,0.8883994,0.90014684,0.93245226,0.9258444,0.948605,0.99265784
+LightGBM (tuned + ensembled),0.08223201,0.16005874,0.15712188,0.17180617,0.17474303,0.17767988,0.22613803,0.2246696,0.2973568,0.27753305,0.3421439,0.31277534,0.3421439,0.37151247,0.30983847,0.39500734,0.39500734,0.38913363,0.40234947,0.39941263,0.46696034,0.48604992,,0.4750367,0.5256975,0.52496326,0.48751837,0.5359765,0.551395,0.6218796,0.6784141,0.53891337,0.5947136,0.56975037,0.6894273,0.5462555,0.7048458,0.5712188,0.6556535,0.6762115,0.6167401,0.66226137,0.66079295,0.7378855,0.71512485,0.69162995,0.6886931,0.76798826,0.7099853,0.7856094,0.71365637,0.70631427,0.6989721,0.7606461,0.84875184,0.71292216,0.7004405,0.71365637,0.81497794,0.76651984,0.95154184,0.8986784,0.86049926,0.82819384,0.90014684,0.8032305,0.79441994,0.81791484,0.917768,0.9060206,0.9030837,0.86049926,0.876652,0.876652,0.88105726,0.938326,0.95154184,0.9236417,0.92143905,0.98384726
+TabM (tuned),0.08810572,0.17621145,0.17547724,0.19236417,0.18061674,0.18649046,0.20190896,0.22173275,0.28414097,0.30983847,0.36123347,0.2907489,0.37371513,0.3876652,0.40528634,0.29809105,0.40234947,0.38179147,0.4427313,0.40528634,0.47650513,0.45080763,0.52496326,,0.5521292,0.5609398,0.4948605,0.53450805,0.56828195,0.5565345,0.6042584,0.5418502,0.5726872,0.5961821,0.6372981,0.60058737,0.6240822,0.61967695,0.6872247,0.6446402,0.69309837,0.6666667,0.6563877,0.7004405,0.68135095,0.7298091,0.69530106,0.69309837,0.66519827,0.7209985,0.6740088,0.7584435,0.66226137,0.76798826,0.7687225,0.7077827,0.67254037,0.71292216,0.74449337,0.73274595,0.79441994,0.82819384,0.7650514,0.8883994,0.8546255,0.8032305,0.7723935,0.8061674,0.8649045,0.8942731,0.8164464,0.8722467,0.8649045,0.88105726,0.84875184,0.87077826,0.938326,0.91042584,0.88986784,0.96182084
+CatBoost (tuned + ensembled),0.07195301,0.14537445,0.15271659,0.16152717,0.16005874,0.15418503,0.2041116,0.19236417,0.28487518,0.24082232,0.30837005,0.31424376,0.32892805,0.35829663,0.34067547,0.38032305,0.37151247,0.37738618,0.41116005,0.39060205,0.4287812,0.47136563,0.4743025,0.4478708,,0.50807637,0.51395005,0.5051395,0.5910426,0.5323054,0.6145374,0.53744495,0.5859031,0.56828195,0.6446402,0.5609398,0.6365639,0.5947136,0.60058737,0.6182085,0.6284875,0.6446402,0.6842878,0.68575627,0.67180616,0.67107195,0.6490455,0.7077827,0.71365637,0.71512485,0.7408223,0.71218795,0.69603527,0.7180617,0.7650514,0.7107195,0.6886931,0.70631427,0.8267254,0.7312775,0.8781204,0.88986784,0.87004405,0.835536,0.8546255,0.76651984,0.77386194,0.8414097,0.8737151,0.8729809,0.8839941,0.8325991,0.856094,0.845815,0.88105726,0.93245226,0.9192364,0.8751836,0.9133627,0.98531574
+iLTM (tuned + ensembled),0.07048458,0.12041116,0.12187959,0.12628488,0.17254038,0.13069016,0.21732746,0.21732746,0.26725402,0.2393539,0.2701909,0.27973568,0.2613803,0.3171806,0.33627018,0.38105726,0.38839942,0.40381792,0.37077826,0.41409692,0.40381792,0.43612334,0.4750367,0.4390602,0.49192363,,0.48751837,0.53377384,0.5227606,0.53891337,0.5734214,0.48237884,0.5954479,0.5873715,0.5726872,0.5462555,0.6079295,0.5844346,0.52716595,0.6035242,0.6226138,0.6798825,0.73348016,0.68281937,0.6475771,0.68575627,0.6886931,0.6901615,0.7041116,0.6798825,0.7312775,0.70337737,0.69162995,0.7496329,0.7378855,0.6681351,0.7004405,0.6688693,0.8105727,0.72613806,0.8619677,0.8472834,0.8428781,0.8443465,0.8516887,0.8237885,0.7694567,0.8781204,0.87958884,0.8619677,0.8986784,0.7797357,0.8781204,0.80910426,0.856094,0.9162996,0.9192364,0.856094,0.8920705,0.9809104
+SAP-RPT-OSS (default),0.14243759,0.23348017,0.26725402,0.25256976,0.24302496,0.24375919,0.27459618,0.2701909,0.37077826,0.37885463,0.36123347,0.38179147,0.37298092,0.39500734,0.44640234,0.42437592,0.44199705,0.4530103,0.4640235,0.47723934,0.4229075,0.43465492,0.51248163,0.5051395,0.48604992,0.51248163,,0.53450805,0.49779737,0.5403818,0.5624082,0.5535976,0.5535976,0.5741556,0.5535976,0.5873715,0.56975037,0.59030837,0.5550661,0.60058737,0.66519827,0.6343612,0.6013216,0.5976505,0.6343612,0.67254037,0.6975037,0.6167401,0.6681351,0.6629956,0.6681351,0.69162995,0.6064611,0.73274595,0.6681351,0.6387665,0.6123348,0.62995595,0.74522763,0.6901615,0.7312775,0.7283407,0.75403816,0.7856094,0.7856094,0.7723935,0.7606461,0.7900147,0.7922173,0.7900147,0.79441994,0.81497794,0.8237885,0.81791484,0.77679884,0.8208517,0.845815,0.83847284,0.866373,0.98237884
+xRFM (tuned + ensembled),0.080763586,0.13656388,0.14243759,0.15712188,0.17180617,0.17327459,0.20044053,0.19530103,0.24229075,0.25110132,0.29955947,0.31864905,0.3017621,0.34361234,0.35315713,0.39647576,0.39500734,0.38032305,0.40528634,0.37885463,0.40969163,0.44640234,0.4640235,0.46549192,0.4948605,0.46622613,0.46549192,,0.5190896,0.5220264,0.54845816,0.5212922,0.6145374,0.53377384,0.5866373,0.5521292,0.5770925,0.56828195,0.5462555,0.6020558,0.59030837,0.61967695,0.643906,0.6740088,0.6549192,0.6182085,0.6321586,0.6886931,0.6593245,0.6784141,0.70631427,0.6681351,0.71512485,0.70190895,0.71218795,0.6791483,0.7180617,0.6769457,0.76798826,0.76358294,0.78707784,0.8061674,0.80469894,0.84875184,0.8120411,0.8325991,0.7518355,0.9397944,0.83920705,0.8325991,0.92511016,0.8443465,0.9045521,0.856094,0.8920705,0.94566816,0.9192364,0.8847283,0.9192364,0.98678416
+CatBoost (tuned),0.06314244,0.13656388,0.14757709,0.15565345,0.15124816,0.15271659,0.19823788,0.17767988,0.2767988,0.2393539,0.3113069,0.31497797,0.3171806,0.3472834,0.32011747,0.35829663,0.36270192,0.34508076,0.4082232,0.37738618,0.41703376,0.438326,0.448605,0.43171805,0.40895742,0.47723934,0.50220263,0.48091042,,0.5110132,0.5917768,0.5212922,0.5653451,0.54919237,0.6116006,0.5323054,0.6240822,0.5756241,0.582232,0.60939795,0.6226138,0.6270191,0.6490455,0.6365639,0.6563877,0.6549192,0.6328928,0.6776799,0.6945668,0.7107195,0.70631427,0.6982379,0.6798825,0.7085169,0.75330395,0.71218795,0.68281937,0.7107195,0.8105727,0.72687227,0.8516887,0.8854626,0.84875184,0.81791484,0.8428781,0.75770926,0.7753304,0.82966226,0.8649045,0.86049926,0.86930984,0.82819384,0.845815,0.83920705,0.87077826,0.92217326,0.9236417,0.86930984,0.9162996,0.98384726
+LightGBM (tuned),0.08223201,0.13950074,0.14537445,0.14977974,0.15491924,0.14243759,0.18649046,0.17621145,0.26284876,0.23127754,0.3069016,0.26872247,0.3113069,0.33627018,0.2701909,0.35022026,0.34948605,0.35535976,0.38032305,0.37885463,0.43024963,0.45080763,0.37812042,0.4434655,0.46769458,0.46108663,0.4596182,0.47797358,0.4889868,,0.5602056,0.51835537,0.5359765,0.51835537,0.62922174,0.5403818,0.5932452,0.52863437,0.6064611,0.64023495,0.5976505,0.6240822,0.6020558,0.6563877,0.69309837,0.664464,0.64977974,0.68281937,0.67180616,0.7312775,0.6894273,0.6791483,0.6769457,0.7180617,0.7856094,0.7004405,0.68135095,0.70190895,0.7606461,0.7503671,0.9038179,0.8671072,0.81497794,0.80763584,0.8612335,0.78707784,0.77679884,0.8113069,0.8825257,0.8935389,0.8737151,0.8516887,0.866373,0.86857563,0.8634361,0.91189426,0.94566816,0.91189426,0.91189426,0.9779736
+XGBoost (tuned + ensembled),0.06167401,0.13436124,0.1277533,0.14684288,0.1380323,0.12481645,0.17180617,0.17033774,0.23568282,0.17327459,0.27900147,0.26358297,0.2870778,0.30029368,0.25844347,0.29515418,0.2922173,0.33994126,0.36417034,0.36270192,0.38325992,0.41703376,0.3215859,0.39574155,0.38546255,0.42657855,0.4375918,0.45154184,0.4082232,0.43979442,,0.4948605,0.53450805,0.5095448,0.54845816,0.5051395,0.57488984,0.4992658,0.54919237,0.5756241,0.57488984,0.6064611,0.59911895,0.6167401,0.6358297,0.6240822,0.6284875,0.65198237,0.6387665,0.68281937,0.68281937,0.6490455,0.6842878,0.68135095,0.71292216,0.6872247,0.6886931,0.68281937,0.7951542,0.72320116,0.8986784,0.9030837,0.8230543,0.78854626,0.81791484,0.76578563,0.7503671,0.8105727,0.8406755,0.82966226,0.856094,0.8208517,0.82819384,0.8311307,0.8414097,0.92217326,0.907489,0.8678414,0.8979442,0.97503674
+TabICL (default),0.051395006,0.11967695,0.14904553,0.1226138,0.22246696,0.19970632,0.24816446,0.2349486,0.31938326,0.37004405,0.29809105,0.30910426,0.35609397,0.37812042,0.39207047,0.40528634,0.4229075,0.3825257,0.39353892,0.39207047,0.43979442,0.47283408,0.46108663,0.4581498,0.46255508,0.51762116,0.44640234,0.4787078,0.4787078,0.48164463,0.5051395,,0.48604992,0.51835537,0.53744495,0.5190896,0.5212922,0.5403818,0.5418502,0.53744495,0.5506608,0.5352423,0.5873715,0.59911895,0.5712188,0.5814978,0.5844346,0.6255507,0.6020558,0.6064611,0.613069,0.6387665,0.6975037,0.6270191,0.6431718,0.6549192,0.69603527,0.65051395,0.6284875,0.7173275,0.6982379,0.68281937,0.7048458,0.71659327,0.75330395,0.7393539,0.7298091,0.75624084,0.76798826,0.8267254,0.7753304,0.835536,0.8113069,0.85022026,0.8678414,0.82452273,0.8472834,0.9229075,0.9060206,0.9676946
+xRFM (tuned),0.08957416,0.13656388,0.15491924,0.16519824,0.16886932,0.16152717,0.18649046,0.16740088,0.23568282,0.25697502,0.26431718,0.28267255,0.2812041,0.3215859,0.30837005,0.36710718,0.35315713,0.36710718,0.39867842,0.34948605,0.3979442,0.40088105,0.40528634,0.42731276,0.41409692,0.40455213,0.44640234,0.38546255,0.43465492,0.4640235,0.46549192,0.51395005,,0.47723934,0.4904552,0.53891337,0.4955947,0.4743025,0.5095448,0.56754774,0.5741556,0.5961821,0.5873715,0.6013216,0.6123348,0.5756241,0.6013216,0.6475771,0.6240822,0.6446402,0.66519827,0.6372981,0.6637298,0.67107195,0.6886931,0.68281937,0.66226137,0.6842878,0.69603527,0.74743026,0.72687227,0.7209985,0.7386197,0.8414097,0.77312773,0.83994126,0.7503671,0.88986784,0.8157122,0.845815,0.8531571,0.8414097,0.88105726,0.8575624,0.8751836,0.8957416,0.9045521,0.90161526,0.9133627,0.98384726
+ModernNCA (tuned),0.07488987,0.13215859,0.11894273,0.14243759,0.13656388,0.14390603,0.18942732,0.18355359,0.2290749,0.26872247,0.26872247,0.23348017,0.2760646,0.30396476,0.3274596,0.35462555,0.35389134,0.3274596,0.3274596,0.36563876,0.36710718,0.37444934,0.43024963,0.40381792,0.43171805,0.4126285,0.42584434,0.46622613,0.45080763,0.48164463,0.4904552,0.48164463,0.5227606,,0.5256975,0.5293686,0.50734216,0.4441997,0.50807637,0.5741556,0.5565345,0.6534508,0.5550661,0.6182085,0.6123348,0.6270191,0.6424376,0.623348,0.6123348,0.6343612,0.6358297,0.6769457,0.64023495,0.70631427,0.67254037,0.67254037,0.64170337,0.6769457,0.73274595,0.7195301,0.71365637,0.74449337,0.7518355,0.78707784,0.76798826,0.79735684,0.74596184,0.8428781,0.8017621,0.85022026,0.82819384,0.80469894,0.84948605,0.8267254,0.8414097,0.8575624,0.8751836,0.8678414,0.886931,0.97356826
+CatBoost (default),0.06461087,0.12041116,0.13142438,0.12922174,0.12628488,0.10866373,0.17327459,0.14390603,0.23201175,0.21145375,0.2760646,0.23201175,0.26725402,0.3024963,0.2613803,0.2819383,0.31277534,0.32892805,0.32892805,0.35462555,0.40381792,0.40234947,0.31057268,0.36270192,0.35535976,0.42731276,0.44640234,0.4133627,0.38839942,0.37077826,0.45154184,0.46255508,0.5095448,0.4743025,,0.46549192,0.48091042,0.48604992,0.5330396,0.58002937,0.5609398,0.54478705,0.5873715,0.6321586,0.6461087,0.6270191,0.62114537,0.6637298,0.6490455,0.6945668,0.6666667,0.65051395,0.6534508,0.68281937,0.7709251,0.70117474,0.6563877,0.70190895,0.7621145,0.70190895,0.80469894,0.7812041,0.79735684,0.7914831,0.8267254,0.7621145,0.77826726,0.76798826,0.8619677,0.8986784,0.8443465,0.80910426,0.8575624,0.8267254,0.84875184,0.8839941,0.9133627,0.886931,0.9192364,0.9809104
+BetaTabPFN (default),0.106461085,0.19603524,0.1938326,0.20117474,0.17327459,0.20264317,0.19236417,0.19603524,0.2452276,0.3024963,0.32452276,0.2650514,0.35389134,0.37298092,0.33920705,0.32305434,0.3421439,0.3120411,0.36563876,0.33259913,0.4331865,0.40381792,0.4537445,0.39941263,0.4390602,0.4537445,0.4126285,0.4478708,0.46769458,0.4596182,0.4948605,0.48091042,0.46108663,0.47063142,0.53450805,,0.5110132,0.4698972,0.5433186,0.59911895,0.5653451,0.5110132,0.51688695,0.57488984,0.6387665,0.59030837,0.56461084,0.60058737,0.54919237,0.6696035,0.5653451,0.6328928,0.6534508,0.6182085,0.6446402,0.68575627,0.65198237,0.68281937,0.5932452,0.66152716,0.643906,0.6490455,0.65198237,0.7180617,0.72540385,0.7209985,0.7613803,0.6975037,0.7488987,0.866373,0.71218795,0.8957416,0.7929515,0.91042584,0.84875184,0.7804699,0.8325991,0.9441997,0.92070484,0.96328926
+XGBoost (tuned),0.05873715,0.12628488,0.12481645,0.14390603,0.12555066,0.11894273,0.16005874,0.14684288,0.2187959,0.16152717,0.26725402,0.25256976,0.2599119,0.2922173,0.2613803,0.26431718,0.2753304,0.33480176,0.34801763,0.35095447,0.36270192,0.37444934,0.29515418,0.37591776,0.36343613,0.39207047,0.43024963,0.4229075,0.37591776,0.40675476,0.42511013,0.4787078,0.50440526,0.49265787,0.5190896,0.4889868,,0.4948605,0.5154185,0.5565345,0.5609398,0.5726872,0.5712188,0.56754774,0.6138032,0.5947136,0.6049926,0.6123348,0.6350955,0.6534508,0.6696035,0.6284875,0.6696035,0.6769457,0.67254037,0.67107195,0.67254037,0.6732746,0.7650514,0.70190895,0.85903084,0.8678414,0.8032305,0.77386194,0.79441994,0.7518355,0.74743026,0.8061674,0.8237885,0.8267254,0.8531571,0.8105727,0.8223201,0.82966226,0.83773863,0.9045521,0.9060206,0.8619677,0.8928047,0.97209984
+ModernNCA (tuned + ensembled),0.054331865,0.1174743,0.10279001,0.13656388,0.14537445,0.10425844,0.17254038,0.18942732,0.2187959,0.26284876,0.26872247,0.2393539,0.24082232,0.27239352,0.3069016,0.32599118,0.33627018,0.33039647,0.34067547,0.38032305,0.36270192,0.38913363,0.4287812,0.38032305,0.40528634,0.41556534,0.40969163,0.43171805,0.42437592,0.47136563,0.5007342,0.4596182,0.5256975,0.5558003,0.51395005,0.5301028,0.5051395,,0.47577092,0.5535976,0.5301028,0.6593245,0.5462555,0.56828195,0.5969163,0.62114537,0.60058737,0.60058737,0.60939795,0.60939795,0.6461087,0.61086637,0.6035242,0.66226137,0.6123348,0.6255507,0.60939795,0.6226138,0.7048458,0.66226137,0.7048458,0.7621145,0.7195301,0.74743026,0.7415565,0.7386197,0.68575627,0.8318649,0.7518355,0.77679884,0.81497794,0.73421437,0.7900147,0.74596184,0.7900147,0.8267254,0.82966226,0.7797357,0.8223201,0.9500734
+TabM (default),0.080763586,0.15418503,0.13950074,0.16446403,0.14243759,0.14757709,0.16005874,0.1483113,0.24669604,0.20851688,0.2819383,0.2496329,0.29295155,0.30837005,0.29368576,0.16226138,0.28487518,0.31571218,0.38179147,0.33039647,0.36710718,0.35682818,0.34434655,0.31277534,0.39941263,0.47283408,0.44493392,0.4537445,0.417768,0.39353892,0.45080763,0.4581498,0.4904552,0.49192363,0.46696034,0.45668134,0.4845815,0.52422905,,0.49632892,0.60058737,0.56828195,0.5660793,0.5212922,0.5609398,0.5961821,0.6182085,0.5462555,0.6152717,0.58957416,0.6358297,0.68135095,0.6226138,0.6989721,0.6123348,0.69603527,0.6270191,0.6989721,0.6872247,0.66079295,0.66226137,0.7004405,0.6989721,0.8105727,0.74375916,0.72540385,0.7753304,0.72687227,0.75330395,0.8267254,0.76651984,0.81938326,0.81497794,0.83480173,0.8032305,0.825257,0.8722467,0.8722467,0.88105726,0.94273126
+EBM (tuned + ensembled),0.086637296,0.1380323,0.13069016,0.13950074,0.13656388,0.14977974,0.15124816,0.14096916,0.22246696,0.20704846,0.26725402,0.19823788,0.2767988,0.2966226,0.2657856,0.26284876,0.28487518,0.2966226,0.3024963,0.31277534,0.39647576,0.36270192,0.32378855,0.35535976,0.38179147,0.39647576,0.39941263,0.3979442,0.39060205,0.35976505,0.42437592,0.46255508,0.43245226,0.42584434,0.41997063,0.40088105,0.4434655,0.44640234,0.50367105,,0.5403818,0.5212922,0.5007342,0.49779737,0.6997063,0.5653451,0.5323054,0.5301028,0.58957416,0.73274595,0.5961821,0.61013216,0.6226138,0.6071953,0.6049926,0.67254037,0.6240822,0.6696035,0.6666667,0.6769457,0.6593245,0.6864905,0.69162995,0.73568285,0.7753304,0.7386197,0.7723935,0.71512485,0.79809105,0.8340675,0.7650514,0.8472834,0.8414097,0.86049926,0.82819384,0.8105727,0.91042584,0.8883994,0.88032305,0.95594716
+TorchMLP (tuned + ensembled),0.04405286,0.08516887,0.08810572,0.105726875,0.0969163,0.10425844,0.1174743,0.111600585,0.19970632,0.2290749,0.22173275,0.14977974,0.2444934,0.25403818,0.20558003,0.25256976,0.27459618,0.27312776,0.31277534,0.31424376,0.33773863,0.2966226,0.38325992,0.3069016,0.37151247,0.37738618,0.33480176,0.40969163,0.37738618,0.40234947,0.42511013,0.4493392,0.42584434,0.4434655,0.4390602,0.43465492,0.4390602,0.4698972,0.39941263,0.4596182,,0.54772395,0.520558,0.5007342,0.5154185,0.5403818,0.6182085,0.5359765,0.6079295,0.55800295,0.6167401,0.71659327,0.5770925,0.71585906,0.5565345,0.6549192,0.58002937,0.6629956,0.6798825,0.6182085,0.65051395,0.6886931,0.6975037,0.8722467,0.7195301,0.74596184,0.7606461,0.75770926,0.77386194,0.8054332,0.73715127,0.78634363,0.8524229,0.8164464,0.7958884,0.76798826,0.8737151,0.8839941,0.8722467,0.95154184
+ModernNCA (default),0.07342144,0.110866375,0.12187959,0.11967695,0.14096916,0.11894273,0.16446403,0.14684288,0.21512482,0.2246696,0.19676946,0.18795888,0.23274596,0.25403818,0.2496329,0.2819383,0.2760646,0.29882526,0.2907489,0.30910426,0.3113069,0.29955947,0.33773863,0.33333334,0.35535976,0.32011747,0.36563876,0.38032305,0.37298092,0.37591776,0.39353892,0.4647577,0.40381792,0.34654918,0.45521292,0.4889868,0.42731276,0.34067547,0.43171805,0.4787078,0.45227605,,0.4530103,0.5814978,0.52716595,0.48604992,0.5403818,0.5932452,0.5668135,0.5756241,0.5976505,0.53891337,0.60939795,0.6049926,0.6358297,0.6490455,0.61086637,0.6534508,0.66519827,0.7180617,0.62995595,0.6284875,0.72393537,0.7606461,0.71512485,0.7856094,0.73568285,0.74449337,0.76358294,0.8267254,0.79809105,0.79735684,0.8325991,0.8157122,0.86049926,0.83847284,0.8472834,0.8957416,0.89133626,0.97650516
+iLTM (tuned),0.055800293,0.071218796,0.086637296,0.10425844,0.101321585,0.07929515,0.11894273,0.1123348,0.17621145,0.1585903,0.17547724,0.1938326,0.17914832,0.2143906,0.20998532,0.2657856,0.25403818,0.2767988,0.3171806,0.32452276,0.30543318,0.2907489,0.33920705,0.34361234,0.31571218,0.2665198,0.39867842,0.35609397,0.35095447,0.3979442,0.40088105,0.4126285,0.4126285,0.44493392,0.4126285,0.48311308,0.4287812,0.4537445,0.4339207,0.4992658,0.479442,0.54698974,,0.51248163,0.5535976,0.5403818,0.57856095,0.561674,0.5624082,0.5917768,0.6049926,0.58883995,0.5873715,0.6240822,0.5932452,0.6380323,0.5873715,0.6343612,0.67107195,0.6343612,0.6740088,0.7004405,0.7077827,0.78707784,0.71365637,0.73348016,0.71365637,0.7856094,0.7606461,0.8054332,0.7900147,0.7276065,0.82966226,0.7591777,0.7797357,0.8208517,0.85022026,0.81938326,0.84801763,0.9544787
+ChimeraBoost (tuned + ensembled),0.06167401,0.09397944,0.08957416,0.10866373,0.10352423,0.080763586,0.1431718,0.12334802,0.19530103,0.18061674,0.21145375,0.18649046,0.21732746,0.24082232,0.21365638,0.23788546,0.27900147,0.28781205,0.2657856,0.32892805,0.32011747,0.34948605,0.26211452,0.29955947,0.31424376,0.3171806,0.40234947,0.32599118,0.36343613,0.34361234,0.38325992,0.40088105,0.39867842,0.38179147,0.36784142,0.42511013,0.43245226,0.43171805,0.4787078,0.50220263,0.4992658,0.4185022,0.48751837,,0.54772395,0.5212922,0.50660795,0.5734214,0.5227606,0.58002937,0.5638766,0.5550661,0.5638766,0.55873716,0.7753304,0.66519827,0.56828195,0.6593245,0.64170337,0.6637298,0.7099853,0.67107195,0.6872247,0.69603527,0.7753304,0.75770926,0.75330395,0.7195301,0.8061674,0.8002937,0.7606461,0.74302495,0.80910426,0.7606461,0.7621145,0.79882526,0.8678414,0.8443465,0.8634361,0.9684288
+EBM (tuned),0.07488987,0.12334802,0.12187959,0.13656388,0.1226138,0.12555066,0.14096916,0.12041116,0.20264317,0.19530103,0.2342144,0.17180617,0.25770926,0.27312776,0.24229075,0.23348017,0.2496329,0.2701909,0.2701909,0.2863436,0.37224668,0.3274596,0.28487518,0.31864905,0.32819384,0.3524229,0.36563876,0.34508076,0.34361234,0.3069016,0.36417034,0.4287812,0.3876652,0.3876652,0.35389134,0.36123347,0.38619676,0.4030837,0.4390602,0.30029368,0.4845815,0.47283408,0.44640234,0.45227605,,0.5007342,0.4889868,0.47650513,0.53891337,0.633627,0.5712188,0.5756241,0.60939795,0.5763583,0.5535976,0.6549192,0.6138032,0.6571219,0.6270191,0.6350955,0.60939795,0.64170337,0.67107195,0.71218795,0.7400881,0.7085169,0.75477237,0.69162995,0.76798826,0.8002937,0.75330395,0.8208517,0.80910426,0.835536,0.80469894,0.78707784,0.88986784,0.86930984,0.87958884,0.948605
+RealMLP (default),0.06314244,0.08223201,0.105726875,0.105726875,0.1174743,0.10866373,0.12922174,0.12922174,0.18208517,0.11820852,0.1938326,0.2041116,0.21732746,0.24082232,0.23788546,0.2092511,0.22026432,0.2606461,0.3010279,0.27459618,0.34581497,0.25256976,0.30837005,0.2701909,0.32892805,0.31424376,0.3274596,0.38179147,0.34508076,0.33553597,0.37591776,0.4185022,0.42437592,0.37298092,0.37298092,0.40969163,0.40528634,0.37885463,0.40381792,0.43465492,0.4596182,0.51395005,0.4596182,0.4787078,0.4992658,,0.51762116,0.49779737,0.5638766,0.5301028,0.5976505,0.5873715,0.58883995,0.6079295,0.5521292,0.64023495,0.60058737,0.6490455,0.66079295,0.6255507,0.6358297,0.6490455,0.6901615,0.78634363,0.7092511,0.7613803,0.7386197,0.72687227,0.76798826,0.7914831,0.7709251,0.77679884,0.8678414,0.80763584,0.8223201,0.79735684,0.8649045,0.876652,0.8883994,0.97503674
+FastaiMLP (tuned + ensembled),0.076358296,0.12555066,0.12187959,0.13509545,0.1226138,0.13509545,0.13950074,0.1071953,0.19162996,0.17988253,0.17621145,0.17180617,0.2246696,0.24669604,0.20558003,0.2452276,0.2555066,0.2709251,0.30029368,0.28781205,0.3113069,0.2290749,0.3113069,0.30469897,0.35095447,0.3113069,0.3024963,0.36784142,0.36710718,0.35022026,0.37151247,0.41556534,0.39867842,0.35756242,0.37885463,0.43538913,0.39500734,0.39941263,0.38179147,0.46769458,0.38179147,0.4596182,0.42143905,0.49339208,0.5110132,0.48237884,,0.5154185,0.51835537,0.52716595,0.53450805,0.5154185,0.58002937,0.6563877,0.5433186,0.6138032,0.5829662,0.6182085,0.643906,0.5932452,0.5704846,0.6064611,0.6578561,0.72613806,0.65198237,0.73568285,0.72246695,0.7107195,0.6901615,0.78854626,0.74302495,0.80763584,0.94566816,0.8311307,0.79882526,0.75624084,0.8208517,0.87958884,0.8737151,0.94566816
+ChimeraBoost (tuned),0.055800293,0.09251101,0.08223201,0.0969163,0.09397944,0.076358296,0.12628488,0.1174743,0.18355359,0.17400882,0.19676946,0.17474303,0.20117474,0.2290749,0.2041116,0.2041116,0.24816446,0.26431718,0.2496329,0.31424376,0.31571218,0.3171806,0.23201175,0.3069016,0.2922173,0.30983847,0.38325992,0.3113069,0.3223201,0.3171806,0.34801763,0.37444934,0.3524229,0.37665197,0.33627018,0.39941263,0.3876652,0.39941263,0.4537445,0.4698972,0.4640235,0.40675476,0.438326,0.42657855,0.52349484,0.50220263,0.4845815,,0.50220263,0.54772395,0.54772395,0.5256975,0.54478705,0.53817916,0.7077827,0.65198237,0.54772395,0.65051395,0.61086637,0.6387665,0.6666667,0.6483113,0.6593245,0.68575627,0.7305433,0.71512485,0.73568285,0.69603527,0.76284873,0.77679884,0.7488987,0.7283407,0.78414094,0.73568285,0.74302495,0.76358294,0.85829663,0.82819384,0.8370044,0.9544787
+ExtraTrees (tuned + ensembled),0.054331865,0.07342144,0.091042586,0.121145375,0.14243759,0.1071953,0.14757709,0.14537445,0.18575624,0.16886932,0.15051395,0.20117474,0.17621145,0.2143906,0.25844347,0.2863436,0.24816446,0.3024963,0.25844347,0.27900147,0.24816446,0.3024963,0.29001468,0.33480176,0.2863436,0.2958884,0.3318649,0.34067547,0.30543318,0.32819384,0.36123347,0.3979442,0.37591776,0.3876652,0.35095447,0.45080763,0.36490455,0.39060205,0.38472834,0.41042584,0.39207047,0.4331865,0.4375918,0.47723934,0.46108663,0.43612334,0.48164463,0.49779737,,0.49339208,0.6556535,0.45594713,0.64170337,0.5227606,0.52863437,0.5521292,0.6461087,0.5638766,0.7400881,0.6637298,0.55947137,0.56828195,0.79882526,0.62114537,0.6534508,0.71218795,0.6982379,0.7195301,0.68281937,0.71659327,0.9060206,0.7276065,0.74743026,0.7378855,0.876652,0.8781204,0.76358294,0.8105727,0.9089574,0.989721
+EBM (default),0.07782672,0.10792951,0.12041116,0.11013216,0.12628488,0.13069016,0.13950074,0.11306901,0.17474303,0.17254038,0.20264317,0.16446403,0.24229075,0.25256976,0.19970632,0.22099853,0.25256976,0.25403818,0.2657856,0.26872247,0.32011747,0.30543318,0.2143906,0.27900147,0.28487518,0.32011747,0.3370044,0.3215859,0.28928047,0.26872247,0.3171806,0.39353892,0.35535976,0.36563876,0.30543318,0.33039647,0.34654918,0.39060205,0.41042584,0.26725402,0.44199705,0.42437592,0.4082232,0.41997063,0.36637297,0.4698972,0.47283408,0.45227605,0.50660795,,0.51835537,0.54919237,0.5917768,0.54478705,0.5110132,0.6358297,0.5932452,0.6358297,0.5961821,0.61967695,0.5638766,0.5859031,0.6270191,0.7004405,0.69309837,0.7004405,0.73421437,0.6475771,0.7209985,0.79441994,0.73715127,0.80910426,0.8032305,0.83480173,0.825257,0.75770926,0.86049926,0.89133626,0.89133626,0.95154184
+ExtraTrees (tuned),0.06020558,0.07195301,0.09985316,0.11894273,0.13950074,0.1123348,0.14096916,0.13362701,0.17474303,0.15491924,0.13069016,0.19016153,0.16740088,0.20264317,0.23201175,0.28340676,0.23054332,0.26872247,0.25844347,0.2657856,0.22760646,0.28487518,0.2863436,0.32599118,0.25917768,0.26872247,0.3318649,0.29368576,0.29368576,0.31057268,0.3171806,0.38693097,0.33480176,0.36417034,0.33333334,0.43465492,0.33039647,0.35389134,0.36417034,0.40381792,0.38325992,0.40234947,0.39500734,0.43612334,0.4287812,0.40234947,0.46549192,0.45227605,0.34434655,0.48164463,,0.4493392,0.61013216,0.50807637,0.5154185,0.55433184,0.61013216,0.5624082,0.6387665,0.65051395,0.5212922,0.50807637,0.7143906,0.60058737,0.6328928,0.7048458,0.6945668,0.68135095,0.6637298,0.7107195,0.82819384,0.71218795,0.73568285,0.7195301,0.8472834,0.83773863,0.7503671,0.8032305,0.9133627,0.9882526
+TorchMLP (tuned),0.03671072,0.06901615,0.06314244,0.086637296,0.07195301,0.07195301,0.08957416,0.08810572,0.15124816,0.17914832,0.17033774,0.105726875,0.19089574,0.2187959,0.16740088,0.17988253,0.20851688,0.22320117,0.26725402,0.2657856,0.28781205,0.2452276,0.29368576,0.24155654,0.28781205,0.2966226,0.30837005,0.3318649,0.3017621,0.32085168,0.35095447,0.36123347,0.36270192,0.32305434,0.34948605,0.36710718,0.37151247,0.38913363,0.31864905,0.38986784,0.28340676,0.46108663,0.41116005,0.44493392,0.42437592,0.4126285,0.4845815,0.4743025,0.54405284,0.45080763,0.5506608,,0.5403818,0.57856095,0.5007342,0.6431718,0.5403818,0.6446402,0.6152717,0.5550661,0.561674,0.5829662,0.6387665,0.78487515,0.6431718,0.71659327,0.73715127,0.69603527,0.6945668,0.74743026,0.6989721,0.7077827,0.7826725,0.7290749,0.75403816,0.72393537,0.80763584,0.8546255,0.82819384,0.9368576
+Nori-30M (default),0.14096916,0.1585903,0.22320117,0.1938326,0.2290749,0.2246696,0.25110132,0.2599119,0.27900147,0.23348017,0.2599119,0.2907489,0.2349486,0.27312776,0.29955947,0.31277534,0.3069016,0.35535976,0.28781205,0.34361234,0.30983847,0.3215859,0.3010279,0.33773863,0.30396476,0.30837005,0.39353892,0.28487518,0.32011747,0.32305434,0.31571218,0.3024963,0.33627018,0.35976505,0.34654918,0.34654918,0.33039647,0.39647576,0.37738618,0.37738618,0.4229075,0.39060205,0.4126285,0.43612334,0.39060205,0.41116005,0.41997063,0.45521292,0.35829663,0.4082232,0.38986784,0.4596182,,0.4581498,0.49192363,0.45521292,0.55800295,0.45521292,0.37151247,0.49339208,0.47797358,0.42437592,0.45594713,0.54478705,0.5624082,0.5535976,0.62114537,0.5961821,0.5873715,0.51248163,0.5976505,0.60058737,0.613069,0.62114537,0.6578561,0.6270191,0.6446402,0.70631427,0.7826725,0.95594716
+FastaiMLP (tuned),0.057268724,0.085903086,0.09985316,0.10279001,0.09251101,0.101321585,0.09985316,0.086637296,0.17254038,0.1637298,0.13069016,0.13362701,0.18355359,0.20264317,0.16152717,0.19089574,0.19970632,0.23641704,0.23715125,0.2452276,0.25697502,0.20558003,0.2393539,0.23201175,0.2819383,0.2503671,0.26725402,0.29809105,0.2914831,0.2819383,0.31864905,0.37298092,0.32892805,0.29368576,0.3171806,0.38179147,0.32305434,0.33773863,0.3010279,0.3928047,0.28414097,0.39500734,0.37591776,0.44126284,0.4236417,0.39207047,0.34361234,0.46182084,0.47723934,0.45521292,0.49192363,0.42143905,0.5418502,,0.5051395,0.5829662,0.54478705,0.58002937,0.5851689,0.54772395,0.50367105,0.52422905,0.59911895,0.654185,0.5932452,0.6842878,0.69603527,0.6431718,0.6387665,0.74302495,0.7077827,0.7621145,0.8443465,0.7804699,0.7694567,0.7195301,0.7606461,0.8516887,0.8414097,0.93538916
+ChimeraBoost (default),0.041116007,0.080763586,0.07929515,0.08810572,0.08370044,0.06461087,0.105726875,0.08957416,0.14977974,0.15124816,0.16593245,0.14977974,0.17474303,0.18502203,0.16446403,0.15418503,0.21292217,0.2290749,0.23348017,0.30396476,0.2599119,0.23641704,0.15124816,0.23127754,0.2349486,0.26211452,0.3318649,0.28781205,0.24669604,0.2143906,0.2870778,0.35682818,0.3113069,0.3274596,0.2290749,0.35535976,0.3274596,0.3876652,0.3876652,0.39500734,0.4434655,0.36417034,0.40675476,0.2246696,0.44640234,0.4478708,0.45668134,0.2922173,0.47136563,0.4889868,0.4845815,0.4992658,0.50807637,0.4948605,,0.6226138,0.50660795,0.6270191,0.5859031,0.5859031,0.5763583,0.5910426,0.6064611,0.63142437,0.67254037,0.6475771,0.7048458,0.6431718,0.7077827,0.74302495,0.6387665,0.7195301,0.7283407,0.7312775,0.6938326,0.72246695,0.8105727,0.8267254,0.8414097,0.95301026
+TabSTAR (tuned),0.15271659,0.17841409,0.19676946,0.18355359,0.2041116,0.19530103,0.23201175,0.20264317,0.2613803,0.2613803,0.26725402,0.21292217,0.27900147,0.28854626,0.26725402,0.2716593,0.27753305,0.28781205,0.26211452,0.27312776,0.32892805,0.25844347,0.2870778,0.2922173,0.28928047,0.3318649,0.36123347,0.32085168,0.28781205,0.29955947,0.31277534,0.34508076,0.3171806,0.3274596,0.29882526,0.31424376,0.32892805,0.37444934,0.30396476,0.3274596,0.34508076,0.35095447,0.36196768,0.33480176,0.34508076,0.35976505,0.38619676,0.34801763,0.4478708,0.36417034,0.44566813,0.35682818,0.54478705,0.41703376,0.37738618,,0.5521292,0.53377384,0.49339208,0.46108663,0.4185022,0.42437592,0.53744495,0.46549192,0.56828195,0.51688695,0.9441997,0.53744495,0.5947136,0.4904552,0.6358297,0.6534508,0.66226137,0.6784141,0.71218795,0.65051395,0.62995595,0.7812041,0.8340675,0.9236417
+Nori (default),0.080763586,0.13215859,0.18061674,0.15712188,0.18649046,0.18502203,0.21145375,0.25697502,0.24082232,0.2041116,0.25256976,0.2701909,0.22320117,0.2496329,0.29515418,0.29809105,0.2863436,0.33333334,0.2760646,0.32599118,0.29515418,0.2863436,0.29955947,0.3274596,0.3113069,0.29955947,0.3876652,0.2819383,0.3171806,0.31864905,0.3113069,0.30396476,0.33773863,0.35829663,0.34361234,0.34801763,0.3274596,0.39060205,0.37298092,0.37591776,0.41997063,0.38913363,0.4126285,0.43171805,0.38619676,0.39941263,0.41703376,0.45227605,0.35389134,0.40675476,0.38986784,0.4596182,0.44199705,0.45521292,0.49339208,0.4478708,,0.45080763,0.37298092,0.4948605,0.479442,0.42437592,0.45741555,0.54478705,0.5624082,0.5550661,0.61967695,0.5947136,0.5859031,0.51248163,0.60058737,0.60058737,0.613069,0.62114537,0.6563877,0.6284875,0.6446402,0.70631427,0.7812041,0.95594716
+TabSTAR (tuned + ensembled),0.15418503,0.18061674,0.19676946,0.18942732,0.2041116,0.19089574,0.22760646,0.20337738,0.2599119,0.25697502,0.26431718,0.21732746,0.27753305,0.2907489,0.26431718,0.26872247,0.2760646,0.2922173,0.26725402,0.2767988,0.3274596,0.25697502,0.2863436,0.2870778,0.29368576,0.33113068,0.37004405,0.32305434,0.28928047,0.29809105,0.3171806,0.34948605,0.31571218,0.32305434,0.29809105,0.3171806,0.3267254,0.37738618,0.3010279,0.33039647,0.3370044,0.34654918,0.36563876,0.34067547,0.34287813,0.35095447,0.38179147,0.34948605,0.43612334,0.36417034,0.4375918,0.35535976,0.54478705,0.41997063,0.37298092,0.46622613,0.54919237,,0.49339208,0.46549192,0.42437592,0.42143905,0.5403818,0.46255508,0.5638766,0.5154185,0.9603524,0.53450805,0.58883995,0.4948605,0.6372981,0.6446402,0.6593245,0.67547727,0.70190895,0.6534508,0.6284875,0.7812041,0.82819384,0.9236417
+RandomForest (tuned + ensembled),0.03377386,0.066079296,0.06754772,0.07342144,0.086637296,0.076358296,0.11306901,0.11453745,0.1174743,0.08957416,0.111600585,0.15418503,0.111600585,0.1585903,0.17033774,0.22026432,0.16446403,0.2496329,0.19970632,0.23054332,0.22173275,0.2496329,0.18502203,0.2555066,0.17327459,0.18942732,0.2547724,0.23201175,0.18942732,0.2393539,0.20484582,0.37151247,0.30396476,0.26725402,0.23788546,0.40675476,0.2349486,0.29515418,0.31277534,0.33333334,0.32011747,0.33480176,0.32892805,0.35829663,0.37298092,0.33920705,0.35609397,0.38913363,0.2599119,0.40381792,0.36123347,0.38472834,0.6284875,0.41483113,0.41409692,0.50660795,0.6270191,0.50660795,,0.6270191,0.4801762,0.49339208,0.67547727,0.5418502,0.5814978,0.6666667,0.643906,0.6563877,0.6182085,0.6255507,0.76651984,0.68575627,0.68281937,0.6901615,0.77679884,0.88105726,0.70337737,0.77312773,0.8634361,0.96622616
+OrionMSP (default),0.052863438,0.066079296,0.062408224,0.076358296,0.08370044,0.10425844,0.0969163,0.076358296,0.17547724,0.19089574,0.1071953,0.17033774,0.14684288,0.14243759,0.22246696,0.23054332,0.2400881,0.24889868,0.22099853,0.21071953,0.18942732,0.12922174,0.23348017,0.26725402,0.26872247,0.27386197,0.30983847,0.23641704,0.27312776,0.2496329,0.2767988,0.28267255,0.25256976,0.2804699,0.29809105,0.33847284,0.29809105,0.33773863,0.33920705,0.32305434,0.38179147,0.2819383,0.36563876,0.33627018,0.36490455,0.37444934,0.40675476,0.36123347,0.33627018,0.38032305,0.34948605,0.44493392,0.50660795,0.45227605,0.41409692,0.53891337,0.5051395,0.53450805,0.37298092,,0.4493392,0.40969163,0.4537445,0.53450805,0.5917768,0.61894274,0.6842878,0.5932452,0.62995595,0.6431718,0.5756241,0.72540385,0.70190895,0.74522763,0.6989721,0.633627,0.7400881,0.83847284,0.8237885,0.95594716
+LightGBM (default),0.03377386,0.04698972,0.04698972,0.06020558,0.051395006,0.042584434,0.06020558,0.06314244,0.086637296,0.07782672,0.11306901,0.09985316,0.1174743,0.14243759,0.08516887,0.12701909,0.14096916,0.19530103,0.18942732,0.21145375,0.20851688,0.17841409,0.04845815,0.20558003,0.12187959,0.1380323,0.26872247,0.21292217,0.1483113,0.096182086,0.101321585,0.3017621,0.27312776,0.2863436,0.19530103,0.35609397,0.14096916,0.29515418,0.33773863,0.34067547,0.34948605,0.37004405,0.32599118,0.29001468,0.39060205,0.36417034,0.42951542,0.33333334,0.44052863,0.43612334,0.4787078,0.438326,0.5220264,0.49632892,0.4236417,0.5814978,0.520558,0.5756241,0.5198238,0.5506608,,0.52496326,0.56901616,0.6255507,0.5829662,0.6446402,0.66226137,0.66519827,0.64023495,0.71218795,0.6886931,0.6784141,0.71512485,0.6945668,0.73421437,0.74229074,0.7797357,0.8017621,0.81938326,0.9464024
+XGBoost (default),0.030837005,0.03377386,0.041116007,0.06754772,0.04992658,0.03377386,0.06314244,0.06461087,0.091042586,0.06461087,0.1071953,0.08223201,0.1174743,0.15565345,0.0969163,0.11306901,0.13069016,0.20264317,0.17914832,0.22173275,0.2349486,0.20851688,0.101321585,0.17180617,0.11013216,0.15271659,0.2716593,0.1938326,0.11453745,0.1328928,0.0969163,0.3171806,0.27900147,0.2555066,0.2187959,0.35095447,0.13215859,0.23788546,0.29955947,0.31350955,0.3113069,0.37151247,0.29955947,0.32892805,0.35829663,0.35095447,0.39353892,0.35168868,0.43171805,0.41409692,0.49192363,0.41703376,0.5756241,0.47577092,0.40895742,0.5756241,0.5756241,0.57856095,0.50660795,0.59030837,0.4750367,,0.5712188,0.5873715,0.53891337,0.6534508,0.6696035,0.6901615,0.6049926,0.72687227,0.72687227,0.68281937,0.69309837,0.6989721,0.7386197,0.7826725,0.7298091,0.8002937,0.8135095,0.9339207
+RandomForest (tuned),0.03524229,0.04992658,0.06314244,0.07195301,0.07488987,0.06020558,0.0969163,0.07929515,0.0969163,0.06461087,0.10866373,0.12041116,0.09985316,0.1380323,0.13876653,0.17914832,0.14096916,0.20190896,0.18208517,0.18502203,0.18208517,0.20558003,0.13950074,0.2349486,0.12995595,0.15712188,0.24596182,0.19530103,0.15124816,0.18502203,0.17694567,0.29515418,0.2613803,0.24816446,0.20264317,0.34801763,0.19676946,0.2804699,0.3010279,0.30837005,0.3024963,0.2760646,0.2922173,0.31277534,0.32892805,0.30983847,0.3421439,0.34067547,0.20117474,0.37298092,0.2856094,0.36123347,0.54405284,0.40088105,0.39353892,0.46255508,0.5425844,0.4596182,0.32452276,0.5462555,0.43098384,0.4287812,,0.50220263,0.5565345,0.6020558,0.6226138,0.60939795,0.59911895,0.6035242,0.68281937,0.67254037,0.6798825,0.6784141,0.74375916,0.74522763,0.68281937,0.76651984,0.8546255,0.9647577
+TorchMLP (default),0.029368576,0.03671072,0.03671072,0.06167401,0.045521293,0.04405286,0.041116007,0.027900146,0.09251101,0.07488987,0.08810572,0.07929515,0.0969163,0.11013216,0.09251101,0.08957416,0.10425844,0.1585903,0.18061674,0.15124816,0.17033774,0.12041116,0.17180617,0.111600585,0.16446403,0.15565345,0.2143906,0.15124816,0.18208517,0.19236417,0.21145375,0.28340676,0.1585903,0.21292217,0.20851688,0.2819383,0.22613803,0.25256976,0.18942732,0.26431718,0.1277533,0.2393539,0.21292217,0.30396476,0.28781205,0.21365638,0.27386197,0.31424376,0.37885463,0.29955947,0.39941263,0.21512482,0.45521292,0.34581497,0.36857563,0.53450805,0.45521292,0.53744495,0.4581498,0.46549192,0.37444934,0.4126285,0.49779737,,0.4948605,0.5734214,0.6461087,0.58002937,0.5418502,0.56828195,0.6182085,0.64170337,0.6490455,0.66226137,0.68281937,0.6328928,0.6475771,0.7797357,0.79441994,0.9162996
+PerpetualBooster (tuned + ensembled),0.016152717,0.04992658,0.04405286,0.051395006,0.04698972,0.051395006,0.06975037,0.06167401,0.10279001,0.06461087,0.09985316,0.10866373,0.10279001,0.12041116,0.07342144,0.12041116,0.10866373,0.19236417,0.17621145,0.18208517,0.16299559,0.13362701,0.09985316,0.14537445,0.14537445,0.1483113,0.2143906,0.18795888,0.15712188,0.13876653,0.18208517,0.24669604,0.22687225,0.23201175,0.17327459,0.27459618,0.20558003,0.25844347,0.25624081,0.2246696,0.2804699,0.28487518,0.2863436,0.2246696,0.2599119,0.2907489,0.34801763,0.26945668,0.34654918,0.3069016,0.36710718,0.35682818,0.4375918,0.40675476,0.3274596,0.43171805,0.4375918,0.43612334,0.4185022,0.4082232,0.41703376,0.46108663,0.4434655,0.5051395,,0.51688695,0.5212922,0.54772395,0.7085169,0.5741556,0.5668135,0.64023495,0.60058737,0.6534508,0.59985316,0.6020558,0.8223201,0.7180617,0.7400881,0.928047
+TabFlex (default),0.055800293,0.06314244,0.076358296,0.086637296,0.08370044,0.07929515,0.09985316,0.081497796,0.13876653,0.15271659,0.07195301,0.12408223,0.12628488,0.13215859,0.1483113,0.17033774,0.16666667,0.18135096,0.22760646,0.13436124,0.11894273,0.09544787,0.19676946,0.19676946,0.23348017,0.17621145,0.22760646,0.16740088,0.24229075,0.21292217,0.2342144,0.2606461,0.16005874,0.20264317,0.23788546,0.27900147,0.24816446,0.2613803,0.27459618,0.2613803,0.25403818,0.2143906,0.2665198,0.24229075,0.2914831,0.23861967,0.26431718,0.28487518,0.28781205,0.29955947,0.29515418,0.28340676,0.44640234,0.31571218,0.3524229,0.48311308,0.44493392,0.4845815,0.33333334,0.38105726,0.35535976,0.34654918,0.3979442,0.42657855,0.48311308,,0.6240822,0.50367105,0.5330396,0.5565345,0.49192363,0.6431718,0.6226138,0.6659325,0.6475771,0.5734214,0.6138032,0.76358294,0.77679884,0.93098384
+TabSTAR (default),0.11453745,0.13215859,0.15124816,0.12922174,0.16299559,0.15712188,0.17180617,0.15418503,0.20264317,0.19530103,0.18502203,0.18208517,0.19676946,0.19970632,0.20117474,0.2041116,0.2143906,0.23348017,0.21659325,0.22320117,0.2187959,0.18795888,0.20558003,0.22760646,0.22613803,0.23054332,0.2393539,0.24816446,0.2246696,0.22320117,0.2496329,0.2701909,0.2496329,0.25403818,0.22173275,0.23861967,0.25256976,0.31424376,0.2246696,0.22760646,0.2393539,0.26431718,0.2863436,0.24669604,0.2452276,0.2613803,0.27753305,0.26431718,0.3017621,0.2657856,0.30543318,0.26284876,0.37885463,0.30396476,0.29515418,0.055800293,0.38032305,0.039647575,0.35609397,0.31571218,0.33773863,0.33039647,0.37738618,0.35389134,0.4787078,0.37591776,,0.42731276,0.49779737,0.41409692,0.47283408,0.5359765,0.4889868,0.54772395,0.5814978,0.45521292,0.5315712,0.6578561,0.69603527,0.88986784
+xRFM (default),0.029368576,0.03524229,0.04845815,0.052863438,0.066079296,0.051395006,0.07488987,0.066079296,0.07929515,0.07342144,0.09985316,0.09544787,0.08883994,0.11453745,0.11453745,0.14684288,0.13215859,0.14537445,0.13656388,0.14390603,0.15565345,0.1585903,0.18208517,0.1938326,0.1585903,0.12187959,0.20998532,0.06020558,0.17033774,0.18869309,0.18942732,0.24375919,0.11013216,0.15712188,0.23201175,0.3024963,0.1938326,0.16813509,0.27312776,0.28487518,0.24229075,0.2555066,0.2143906,0.2804699,0.30837005,0.27312776,0.28928047,0.30396476,0.2804699,0.3524229,0.31864905,0.30396476,0.40381792,0.35682818,0.35682818,0.46255508,0.40528634,0.46549192,0.34361234,0.40675476,0.33480176,0.30983847,0.39060205,0.41997063,0.45227605,0.49632892,0.5726872,,0.47650513,0.5315712,0.47283408,0.6079295,0.54478705,0.6079295,0.5932452,0.5418502,0.5859031,0.7048458,0.7180617,0.92217326
+PerpetualBooster (tuned),0.017621145,0.039647575,0.03891336,0.04698972,0.04698972,0.054331865,0.05873715,0.054331865,0.07782672,0.057268724,0.07929515,0.091042586,0.091776796,0.111600585,0.07048458,0.09985316,0.10425844,0.17841409,0.16740088,0.17180617,0.13950074,0.11894273,0.08223201,0.13509545,0.12628488,0.12041116,0.20778267,0.16079295,0.13509545,0.1174743,0.15932453,0.23201175,0.18428782,0.19823788,0.1380323,0.25110132,0.17621145,0.24816446,0.24669604,0.20190896,0.22613803,0.23641704,0.2393539,0.1938326,0.23201175,0.23201175,0.30983847,0.23715125,0.3171806,0.27900147,0.33627018,0.30543318,0.4126285,0.36123347,0.2922173,0.40528634,0.41409692,0.41116005,0.38179147,0.37004405,0.35976505,0.39500734,0.40088105,0.4581498,0.2914831,0.46696034,0.50220263,0.52349484,,0.5462555,0.5330396,0.6240822,0.5565345,0.6343612,0.5609398,0.5712188,0.71145374,0.7004405,0.7180617,0.9030837
+iLTM (default),0.051395006,0.045521293,0.04992658,0.06167401,0.057268724,0.057268724,0.06020558,0.04992658,0.09838473,0.080763586,0.101321585,0.07488987,0.105726875,0.12481645,0.09985316,0.091042586,0.11306901,0.1277533,0.14096916,0.15271659,0.14977974,0.14096916,0.09397944,0.105726875,0.12701909,0.1380323,0.20998532,0.16740088,0.13950074,0.106461085,0.17033774,0.17327459,0.15418503,0.14977974,0.101321585,0.13362701,0.17327459,0.22320117,0.17327459,0.16593245,0.19456682,0.17327459,0.19456682,0.19970632,0.19970632,0.20851688,0.21145375,0.22320117,0.28340676,0.20558003,0.28928047,0.25256976,0.48751837,0.25697502,0.25697502,0.5095448,0.48751837,0.5051395,0.37444934,0.35682818,0.28781205,0.27312776,0.39647576,0.43171805,0.42584434,0.4434655,0.5859031,0.4684288,0.4537445,,0.47577092,0.6182085,0.5550661,0.6328928,0.67107195,0.5462555,0.5638766,0.7283407,0.7812041,0.8854626
+ExtraTrees (default),0.014684288,0.029368576,0.04992658,0.04992658,0.051395006,0.04698972,0.07048458,0.06754772,0.04845815,0.04698972,0.05506608,0.09985316,0.057268724,0.08370044,0.08223201,0.12922174,0.12481645,0.14537445,0.16299559,0.15124816,0.09397944,0.14243759,0.0969163,0.18355359,0.116005875,0.101321585,0.20558003,0.07488987,0.13069016,0.12628488,0.14390603,0.2246696,0.14684288,0.17180617,0.15565345,0.28781205,0.14684288,0.18502203,0.23348017,0.2349486,0.26284876,0.20190896,0.20998532,0.2393539,0.24669604,0.2290749,0.25697502,0.25110132,0.09397944,0.26284876,0.17180617,0.3010279,0.40234947,0.2922173,0.36123347,0.36417034,0.39941263,0.36270192,0.23348017,0.42437592,0.3113069,0.27312776,0.3171806,0.38179147,0.4331865,0.50807637,0.52716595,0.52716595,0.46696034,0.52422905,,0.5873715,0.5565345,0.59911895,0.64170337,0.5712188,0.5462555,0.7048458,0.76798826,0.96622616
+Linear (tuned + ensembled),0.04698972,0.07048458,0.086637296,0.06020558,0.07488987,0.08516887,0.08516887,0.07342144,0.09985316,0.11306901,0.1174743,0.08516887,0.1585903,0.16666667,0.116740085,0.1071953,0.1585903,0.12922174,0.12481645,0.10352423,0.22613803,0.17180617,0.13950074,0.1277533,0.16740088,0.22026432,0.18502203,0.15565345,0.17180617,0.1483113,0.17914832,0.16446403,0.1585903,0.19530103,0.19089574,0.10425844,0.18942732,0.2657856,0.18061674,0.15271659,0.21365638,0.20264317,0.27239352,0.25697502,0.17914832,0.22320117,0.19236417,0.2716593,0.27239352,0.19089574,0.28781205,0.2922173,0.39941263,0.23788546,0.2804699,0.34654918,0.39941263,0.35535976,0.31424376,0.27459618,0.3215859,0.3171806,0.3274596,0.35829663,0.35976505,0.35682818,0.4640235,0.39207047,0.37591776,0.38179147,0.4126285,,0.38032305,0.633627,0.47723934,0.41409692,0.45080763,0.78414094,0.5756241,0.77679884
+FastaiMLP (default),0.02496329,0.030837005,0.041116007,0.04992658,0.04405286,0.04845815,0.051395006,0.041116007,0.07929515,0.052863438,0.057268724,0.080763586,0.07342144,0.08370044,0.06167401,0.08516887,0.07709251,0.14390603,0.14096916,0.14390603,0.09985316,0.08370044,0.12334802,0.13509545,0.14390603,0.12187959,0.17621145,0.09544787,0.15418503,0.13362701,0.17180617,0.18869309,0.11894273,0.15051395,0.14243759,0.20704846,0.17767988,0.20998532,0.18502203,0.1585903,0.14757709,0.16740088,0.17033774,0.19089574,0.19089574,0.13215859,0.054331865,0.21585903,0.25256976,0.19676946,0.26431718,0.21732746,0.38693097,0.15565345,0.2716593,0.33773863,0.38693097,0.34067547,0.3171806,0.29809105,0.28487518,0.3069016,0.32011747,0.35095447,0.39941263,0.37738618,0.5110132,0.45521292,0.4434655,0.44493392,0.4434655,0.61967695,,0.6328928,0.5814978,0.489721,0.5829662,0.7400881,0.7591777,0.92217326
+Linear (tuned),0.041116007,0.06461087,0.06754772,0.055800293,0.06314244,0.06754772,0.07342144,0.06754772,0.086637296,0.111600585,0.11013216,0.07195301,0.14243759,0.15565345,0.096182086,0.09471366,0.15198238,0.12041116,0.10866373,0.10279001,0.20264317,0.15271659,0.12334802,0.11894273,0.15418503,0.19089574,0.18208517,0.14390603,0.16079295,0.13142438,0.16886932,0.14977974,0.14243759,0.17327459,0.17327459,0.08957416,0.17033774,0.25403818,0.16519824,0.13950074,0.18355359,0.18428782,0.24082232,0.2393539,0.16446403,0.19236417,0.16886932,0.26431718,0.26211452,0.16519824,0.2804699,0.2709251,0.37885463,0.2195301,0.26872247,0.3215859,0.37885463,0.32452276,0.30983847,0.2547724,0.30543318,0.3010279,0.3215859,0.33773863,0.34654918,0.33406755,0.45227605,0.39207047,0.36563876,0.36710718,0.40088105,0.36637297,0.36710718,,0.45154184,0.39353892,0.44640234,0.71659327,0.5609398,0.7694567
+KNN (tuned + ensembled),0.030837005,0.052863438,0.057268724,0.04845815,0.07195301,0.054331865,0.08370044,0.0969163,0.07929515,0.08810572,0.07048458,0.08370044,0.07856094,0.080763586,0.11306901,0.12334802,0.1277533,0.13950074,0.1380323,0.12628488,0.09838473,0.111600585,0.11894273,0.15124816,0.11894273,0.14390603,0.22320117,0.10792951,0.12922174,0.13656388,0.1585903,0.13215859,0.12481645,0.1585903,0.15124816,0.15124816,0.16226138,0.20998532,0.19676946,0.17180617,0.2041116,0.13950074,0.22026432,0.23788546,0.19530103,0.17767988,0.20117474,0.25697502,0.12334802,0.17474303,0.15271659,0.24596182,0.3421439,0.23054332,0.3061674,0.28781205,0.34361234,0.29809105,0.22320117,0.3010279,0.2657856,0.2613803,0.25624081,0.3171806,0.40014684,0.3524229,0.4185022,0.40675476,0.4390602,0.32892805,0.35829663,0.5227606,0.4185022,0.54845816,,0.3979442,0.51395005,0.67547727,0.8641703,0.969163
+RandomForest (default),0.016152717,0.017621145,0.032305434,0.038179148,0.041116007,0.026431719,0.051395006,0.051395006,0.059471365,0.032305434,0.039647575,0.06387665,0.051395006,0.07048458,0.07048458,0.09397944,0.08370044,0.13876653,0.12187959,0.12408223,0.105726875,0.09985316,0.06167401,0.12922174,0.06754772,0.08370044,0.17914832,0.054331865,0.07782672,0.08810572,0.07782672,0.17547724,0.10425844,0.14243759,0.116005875,0.2195301,0.09544787,0.17327459,0.17474303,0.18942732,0.23201175,0.16152717,0.17914832,0.20117474,0.21292217,0.20264317,0.24375919,0.23641704,0.12187959,0.24229075,0.16226138,0.2760646,0.37298092,0.2804699,0.27753305,0.34948605,0.37151247,0.34654918,0.11894273,0.36637297,0.25770926,0.21732746,0.2547724,0.36710718,0.3979442,0.42657855,0.54478705,0.4581498,0.4287812,0.4537445,0.4287812,0.5859031,0.510279,0.6064611,0.6020558,,0.52863437,0.7004405,0.74449337,0.93538916
+PerpetualBooster (default),0.014684288,0.026431719,0.030837005,0.020558003,0.030837005,0.032305434,0.030837005,0.02496329,0.042584434,0.029368576,0.03671072,0.03524229,0.051395006,0.052863438,0.04845815,0.039647575,0.045521293,0.09838473,0.11306901,0.10425844,0.091042586,0.06754772,0.04845815,0.06167401,0.080763586,0.080763586,0.15418503,0.080763586,0.076358296,0.054331865,0.09251101,0.15271659,0.09544787,0.12481645,0.086637296,0.16740088,0.09397944,0.17033774,0.1277533,0.08957416,0.12628488,0.15271659,0.14977974,0.13215859,0.11013216,0.13509545,0.17914832,0.14170338,0.23641704,0.13950074,0.2496329,0.19236417,0.35535976,0.2393539,0.18942732,0.37004405,0.35535976,0.37151247,0.2966226,0.2599119,0.22026432,0.2701909,0.3171806,0.3524229,0.17767988,0.38619676,0.4684288,0.41409692,0.28854626,0.43612334,0.4537445,0.54919237,0.41703376,0.5535976,0.48604992,0.47136563,,0.6380323,0.6387665,0.8428781
+Linear (default),0.029368576,0.038179148,0.047723934,0.035976507,0.05506608,0.04698972,0.049192365,0.03524229,0.07195301,0.09544787,0.051395006,0.042584434,0.09985316,0.10279001,0.07048458,0.07488987,0.12041116,0.0969163,0.08516887,0.066079296,0.11453745,0.07415565,0.076358296,0.08957416,0.12481645,0.14390603,0.16152717,0.11527166,0.13069016,0.08810572,0.13215859,0.07709251,0.09838473,0.13215859,0.11306901,0.055800293,0.1380323,0.22026432,0.1277533,0.111600585,0.116005875,0.10425844,0.18061674,0.15565345,0.13069016,0.12334802,0.12041116,0.17180617,0.18942732,0.10866373,0.19676946,0.14537445,0.29368576,0.1483113,0.17327459,0.2187959,0.29368576,0.2187959,0.22687225,0.16152717,0.19823788,0.19970632,0.23348017,0.22026432,0.2819383,0.23641704,0.3421439,0.29515418,0.29955947,0.2716593,0.29515418,0.21585903,0.2599119,0.28340676,0.32452276,0.29955947,0.36196768,,0.47577092,0.72246695
+KNN (tuned),0.03377386,0.041116007,0.042584434,0.042584434,0.06167401,0.04698972,0.05873715,0.06314244,0.06461087,0.06461087,0.051395006,0.062408224,0.06754772,0.07488987,0.07488987,0.086637296,0.09985316,0.111600585,0.105726875,0.0969163,0.076358296,0.051395006,0.07856094,0.11013216,0.086637296,0.10792951,0.13362701,0.080763586,0.08370044,0.08810572,0.1020558,0.09397944,0.086637296,0.11306901,0.080763586,0.07929515,0.1071953,0.17767988,0.11894273,0.11967695,0.1277533,0.10866373,0.15198238,0.13656388,0.12041116,0.111600585,0.12628488,0.16299559,0.091042586,0.10866373,0.086637296,0.17180617,0.21732746,0.1585903,0.1585903,0.16593245,0.2187959,0.17180617,0.13656388,0.17621145,0.18061674,0.18649046,0.14537445,0.20558003,0.2599119,0.22320117,0.30396476,0.2819383,0.2819383,0.2187959,0.23201175,0.42437592,0.24082232,0.4390602,0.13582966,0.2555066,0.36123347,0.52422905,,0.9265786
+KNN (default),0.007342144,0.005873715,0.007342144,0.005873715,0.010279002,0.008810572,0.016152717,0.017621145,0.007342144,0.007342144,0.01174743,0.01174743,0.007342144,0.010279002,0.01174743,0.029368576,0.020558003,0.027900146,0.02349486,0.032305434,0.008810572,0.007342144,0.016152717,0.038179148,0.014684288,0.019089574,0.017621145,0.013215859,0.016152717,0.02202643,0.02496329,0.032305434,0.016152717,0.026431719,0.019089574,0.03671072,0.027900146,0.04992658,0.057268724,0.04405286,0.04845815,0.02349486,0.045521293,0.03157122,0.051395006,0.02496329,0.054331865,0.045521293,0.010279002,0.04845815,0.01174743,0.06314244,0.04405286,0.06461087,0.04698972,0.076358296,0.04405286,0.076358296,0.03377386,0.04405286,0.05359765,0.066079296,0.03524229,0.08370044,0.07195301,0.06901615,0.11013216,0.07782672,0.0969163,0.11453745,0.03377386,0.22320117,0.07782672,0.23054332,0.030837005,0.06461087,0.15712188,0.27753305,0.07342144,
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_all/leaderboard_overview_explorer.html
index a00a64a9d3cf7acb7a3f28116c7e79faa008cb57..5e2cbc1ef4b1ce844675082b19b2613db0eeb290 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Binary | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_binary/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..d10f8c3aff2e306be93435d8f1e425acb37cf087
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_binary/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1780.0,"score":0.887,"rank":6.7,"hrank":1.69,"imp":4.088,"train_time":36.45,"predict_time":7.109,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-91","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1661.0,"score":0.657,"rank":11.06,"hrank":4.46,"imp":11.397,"train_time":2.63,"predict_time":0.361,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-64","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1658.0,"score":0.619,"rank":11.19,"hrank":4.12,"imp":8.864,"train_time":272.41,"predict_time":4.561,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-80","imputed":false},{"position":3,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1590.0,"score":0.595,"rank":14.65,"hrank":5.36,"imp":11.521,"train_time":1.77,"predict_time":0.158,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-58","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1584.0,"score":0.536,"rank":15.02,"hrank":8.01,"imp":12.779,"train_time":4.82,"predict_time":0.464,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-59","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1539.0,"score":0.459,"rank":17.81,"hrank":8.64,"imp":13.44,"train_time":1924.5,"predict_time":9.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-54","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1523.0,"score":0.465,"rank":18.86,"hrank":11.17,"imp":13.88,"train_time":4.91,"predict_time":0.694,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-45","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1511.0,"score":0.431,"rank":19.7,"hrank":9.31,"imp":14.019,"train_time":1924.5,"predict_time":1.329,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-48","imputed":false},{"position":8,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1482.0,"score":0.38,"rank":21.8,"hrank":7.85,"imp":14.33,"train_time":1152.97,"predict_time":1.783,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-65","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1466.0,"score":0.338,"rank":23.03,"hrank":14.78,"imp":15.256,"train_time":2775.98,"predict_time":8.849,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-49","imputed":false},{"position":10,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1452.0,"score":0.299,"rank":24.13,"hrank":11.85,"imp":15.704,"train_time":1612.07,"predict_time":1.884,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-48","imputed":false},{"position":11,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1443.0,"score":0.324,"rank":24.86,"hrank":13.59,"imp":15.43,"train_time":8.06,"predict_time":1.742,"imputed_pct":3.33,"hardware":"GPU","elo_ci":"+64\/-49","imputed":true},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1428.0,"score":0.328,"rank":26.07,"hrank":10.86,"imp":16.013,"train_time":1.44,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-59","imputed":false},{"position":13,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1411.0,"score":0.239,"rank":27.44,"hrank":18.49,"imp":16.799,"train_time":328.96,"predict_time":1.101,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-38","imputed":false},{"position":14,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1401.0,"score":0.248,"rank":28.3,"hrank":17.74,"imp":16.511,"train_time":8305.07,"predict_time":336.002,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-51","imputed":false},{"position":15,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1400.0,"score":0.257,"rank":28.4,"hrank":13.66,"imp":15.817,"train_time":4982.66,"predict_time":358.753,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-50","imputed":false},{"position":16,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1398.0,"score":0.239,"rank":28.55,"hrank":13.73,"imp":16.584,"train_time":1612.07,"predict_time":0.232,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-51","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1392.0,"score":0.244,"rank":29.05,"hrank":14.79,"imp":16.536,"train_time":738.26,"predict_time":0.246,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-50","imputed":false},{"position":18,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1391.0,"score":0.258,"rank":29.2,"hrank":14.08,"imp":16.504,"train_time":3.04,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-49","imputed":false},{"position":19,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1386.0,"score":0.238,"rank":29.64,"hrank":14.97,"imp":16.699,"train_time":738.26,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-52","imputed":false},{"position":20,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1379.0,"score":0.221,"rank":30.24,"hrank":16.32,"imp":16.839,"train_time":2775.98,"predict_time":0.415,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-54","imputed":false},{"position":21,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1365.0,"score":0.285,"rank":31.4,"hrank":12.28,"imp":18.382,"train_time":14.26,"predict_time":4.299,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-79","imputed":false},{"position":22,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1365.0,"score":0.262,"rank":31.45,"hrank":18.52,"imp":17.485,"train_time":1.13,"predict_time":0.059,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-67","imputed":false},{"position":23,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1364.0,"score":0.16,"rank":31.49,"hrank":23.25,"imp":17.599,"train_time":328.96,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-32","imputed":false},{"position":24,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1364.0,"score":0.233,"rank":31.51,"hrank":17.85,"imp":18.487,"train_time":250.39,"predict_time":1.708,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-51","imputed":false},{"position":25,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1362.0,"score":0.219,"rank":31.74,"hrank":17.71,"imp":17.985,"train_time":281.39,"predict_time":0.314,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-56","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1361.0,"score":0.196,"rank":31.76,"hrank":19.61,"imp":17.598,"train_time":473.04,"predict_time":0.621,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-46","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1348.0,"score":0.179,"rank":32.99,"hrank":18.32,"imp":17.933,"train_time":3383.91,"predict_time":0.51,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+43\/-56","imputed":false},{"position":28,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1345.0,"score":0.246,"rank":33.22,"hrank":16.21,"imp":18.096,"train_time":3383.91,"predict_time":8.39,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-90","imputed":false},{"position":29,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1340.0,"score":0.192,"rank":33.61,"hrank":15.35,"imp":17.225,"train_time":4982.66,"predict_time":42.199,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-52","imputed":false},{"position":30,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1334.0,"score":0.178,"rank":34.23,"hrank":19.67,"imp":18.453,"train_time":281.39,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-55","imputed":false},{"position":31,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1334.0,"score":0.295,"rank":34.24,"hrank":13.38,"imp":18.962,"train_time":103.88,"predict_time":2.615,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+88\/-82","imputed":true},{"position":32,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1330.0,"score":0.143,"rank":34.56,"hrank":23.18,"imp":18.088,"train_time":473.04,"predict_time":0.13,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-40","imputed":false},{"position":33,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1324.0,"score":0.253,"rank":35.12,"hrank":15.32,"imp":19.185,"train_time":3031.98,"predict_time":27.042,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+84\/-73","imputed":true},{"position":34,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1310.0,"score":0.15,"rank":36.35,"hrank":22.75,"imp":18.48,"train_time":642.4,"predict_time":2.766,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-54","imputed":false},{"position":35,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1308.0,"score":0.156,"rank":36.58,"hrank":20.12,"imp":19.076,"train_time":1.25,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-57","imputed":false},{"position":36,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1303.0,"score":0.14,"rank":37.02,"hrank":26.61,"imp":18.305,"train_time":2241.92,"predict_time":2.415,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-39","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1300.0,"score":0.152,"rank":37.32,"hrank":22.59,"imp":18.614,"train_time":5.03,"predict_time":0.208,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-56","imputed":false},{"position":38,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1294.0,"score":0.109,"rank":37.8,"hrank":26.81,"imp":19.703,"train_time":936.75,"predict_time":0.134,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-43","imputed":false},{"position":39,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1290.0,"score":0.111,"rank":38.19,"hrank":22.55,"imp":19.231,"train_time":8305.07,"predict_time":52.875,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+43\/-50","imputed":false},{"position":40,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1286.0,"score":0.121,"rank":38.59,"hrank":24.85,"imp":19.574,"train_time":642.4,"predict_time":0.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-60","imputed":false},{"position":41,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1282.0,"score":0.146,"rank":38.86,"hrank":18.25,"imp":20.794,"train_time":12.86,"predict_time":3.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-59","imputed":false},{"position":42,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1280.0,"score":0.204,"rank":39.11,"hrank":16.63,"imp":20.801,"train_time":3031.98,"predict_time":0.514,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+78\/-77","imputed":true},{"position":43,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1268.0,"score":0.155,"rank":40.22,"hrank":21.82,"imp":20.051,"train_time":49.67,"predict_time":43.824,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-68","imputed":false},{"position":44,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1262.0,"score":0.071,"rank":40.7,"hrank":29.43,"imp":20.375,"train_time":936.75,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-44","imputed":false},{"position":45,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1260.0,"score":0.099,"rank":40.92,"hrank":22.62,"imp":20.593,"train_time":13.51,"predict_time":0.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+44\/-43","imputed":false},{"position":46,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1255.0,"score":0.076,"rank":41.35,"hrank":28.75,"imp":19.498,"train_time":9.33,"predict_time":1.116,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+41\/-46","imputed":false},{"position":47,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1254.0,"score":0.109,"rank":41.48,"hrank":27.44,"imp":19.815,"train_time":559.38,"predict_time":4.079,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-58","imputed":false},{"position":48,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1253.0,"score":0.2,"rank":41.56,"hrank":17.98,"imp":21.57,"train_time":2.71,"predict_time":0.276,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+77\/-78","imputed":true},{"position":49,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1239.0,"score":0.069,"rank":42.82,"hrank":27.83,"imp":21.15,"train_time":3.27,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-50","imputed":false},{"position":50,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.072,"rank":44.28,"hrank":30.2,"imp":20.605,"train_time":1.24,"predict_time":0.11,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-55","imputed":false},{"position":51,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1220.0,"score":0.06,"rank":44.51,"hrank":33.64,"imp":20.327,"train_time":2241.92,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-37","imputed":false},{"position":52,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1187.0,"score":0.043,"rank":47.44,"hrank":31.32,"imp":21.732,"train_time":559.38,"predict_time":0.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-59","imputed":false},{"position":53,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1184.0,"score":0.028,"rank":47.7,"hrank":38.68,"imp":21.642,"train_time":182.35,"predict_time":0.686,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-54","imputed":false},{"position":54,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1178.0,"score":0.031,"rank":48.22,"hrank":41.02,"imp":21.33,"train_time":0.81,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-36","imputed":false},{"position":55,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1171.0,"score":0.063,"rank":48.86,"hrank":29.31,"imp":23.166,"train_time":210.38,"predict_time":43.847,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-66","imputed":false},{"position":56,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1164.0,"score":0.16,"rank":49.41,"hrank":12.73,"imp":24.476,"train_time":25373.34,"predict_time":3.667,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-86","imputed":false},{"position":57,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1161.0,"score":0.028,"rank":49.68,"hrank":37.94,"imp":22.454,"train_time":182.35,"predict_time":0.058,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-56","imputed":false},{"position":58,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1158.0,"score":0.153,"rank":49.96,"hrank":12.35,"imp":24.604,"train_time":25373.34,"predict_time":14.078,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-85","imputed":false},{"position":59,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1151.0,"score":0.021,"rank":50.53,"hrank":37.74,"imp":23.033,"train_time":169.75,"predict_time":0.603,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-60","imputed":false},{"position":60,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1141.0,"score":0.087,"rank":51.38,"hrank":33.89,"imp":24.152,"train_time":14.46,"predict_time":2.705,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-69","imputed":false},{"position":61,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1120.0,"score":0.007,"rank":53.13,"hrank":41.1,"imp":23.689,"train_time":169.75,"predict_time":0.059,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-59","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1099.0,"score":0.032,"rank":54.74,"hrank":39.18,"imp":25.54,"train_time":117.85,"predict_time":0.773,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-68","imputed":false},{"position":63,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1098.0,"score":0.012,"rank":54.82,"hrank":41.31,"imp":24.003,"train_time":4.95,"predict_time":0.091,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-46","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1067.0,"score":0.017,"rank":57.17,"hrank":42.33,"imp":26.913,"train_time":117.85,"predict_time":0.29,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-71","imputed":false},{"position":65,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1066.0,"score":0.102,"rank":57.28,"hrank":20.34,"imp":29.214,"train_time":391.11,"predict_time":4.052,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-104","imputed":false},{"position":66,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1066.0,"score":0.011,"rank":57.29,"hrank":37.19,"imp":25.327,"train_time":2.56,"predict_time":0.267,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-77","imputed":false},{"position":67,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1053.0,"score":0.055,"rank":58.23,"hrank":37.05,"imp":29.817,"train_time":162.68,"predict_time":0.405,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-102","imputed":false},{"position":68,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1044.0,"score":0.057,"rank":58.84,"hrank":36.58,"imp":29.096,"train_time":0.85,"predict_time":0.129,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-98","imputed":false},{"position":69,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1030.0,"score":0.013,"rank":59.81,"hrank":43.6,"imp":27.277,"train_time":139.88,"predict_time":3.917,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-94","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1028.0,"score":0.035,"rank":59.92,"hrank":39.41,"imp":30.439,"train_time":162.68,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-110","imputed":false},{"position":71,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1002.0,"score":0.009,"rank":61.62,"hrank":53.47,"imp":27.394,"train_time":1.09,"predict_time":0.191,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-92","imputed":false},{"position":72,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.004,"rank":61.76,"hrank":54.95,"imp":27.733,"train_time":0.25,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-77","imputed":false},{"position":73,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":984.0,"score":0.011,"rank":62.77,"hrank":54.15,"imp":30.546,"train_time":2.8,"predict_time":0.564,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-99","imputed":false},{"position":74,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":972.0,"score":0.021,"rank":63.44,"hrank":43.52,"imp":32.277,"train_time":0.84,"predict_time":0.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-109","imputed":false},{"position":75,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":947.0,"score":0.0,"rank":64.86,"hrank":52.52,"imp":30.975,"train_time":15.8,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-73","imputed":false},{"position":76,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":926.0,"score":0.001,"rank":65.99,"hrank":48.95,"imp":32.081,"train_time":139.88,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-124","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":622.0,"score":0.0,"rank":75.48,"hrank":68.79,"imp":46.014,"train_time":0.15,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+101\/-158","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_all/pareto_front_explorer.html
index 51e53c51ee8d51ea8a9f2ca5193cb2efda07f8c3..3de7b48b59c37bcc63aee81759b9a1b16be4edfb 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7076126445,"imp":18.4866613079,"elo":1364.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0203827935,"imp":16.5036930817,"elo":1390.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.028876079,"imp":16.6993938434,"elo":1385.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.245609885,"imp":16.5359003279,"elo":1392.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0261083185,"imp":19.0759241486,"elo":1307.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0271269609,"imp":18.4530000839,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3141545533,"imp":17.9849925604,"elo":1361.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0148652021,"imp":21.149722515,"elo":1239.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0160758722,"imp":20.3752775842,"elo":1262.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1335297064,"imp":19.7026453378,"elo":1294.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1905448885,"imp":27.3944661579,"elo":1002.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0583336488,"imp":22.453747647,"elo":1161.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6857611443,"imp":21.6423356218,"elo":1184.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2670364168,"imp":25.3266879236,"elo":1065.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2760208318,"imp":21.732386306,"elo":1187.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0790081457,"imp":19.8146595822,"elo":1253.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0925963767,"imp":21.3302201521,"elo":1178.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1264589403,"imp":17.5985878233,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1012944014,"imp":16.798785983,"elo":1411.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.2986429509,"imp":18.3824192374,"elo":1365.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.6152636131,"imp":18.9617472986,"elo":1333.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.310422718,"imp":20.5934045385,"elo":1260.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5103151766,"imp":17.9334509229,"elo":1347.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3903226253,"imp":18.0961291655,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7050443034,"imp":24.1522716371,"elo":1140.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0291154934,"imp":30.9746453444,"elo":947.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2900648816,"imp":26.9128783736,"elo":1067.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7725893154,"imp":25.5398251543,"elo":1099.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.02947866,"imp":27.7326064044,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0593784249,"imp":23.6890906685,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.6033358381,"imp":23.033027877,"elo":1151.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.1163701521,"imp":19.4983343958,"elo":1255.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.4146988006,"imp":16.8394057755,"elo":1378.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":8.8488311223,"imp":15.2564513691,"elo":1466.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6942704557,"imp":13.8798596518,"elo":1523.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3289068866,"imp":14.0187885268,"elo":1511.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.1029658055,"imp":13.4404211141,"elo":1539.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.166690434,"imp":20.7939330249,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"imp":20.051398761,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.1992341236,"imp":17.2246576566,"elo":1340.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":358.7531028481,"imp":15.8172984497,"elo":1400.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"imp":16.0127527493,"elo":1427.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"imp":4.088125843,"elo":1779.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.128947975,"imp":29.0960083209,"elo":1043.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.7422345138,"imp":15.4296572582,"elo":1442.7,"imputed":true,"imputed_pct":3.3333333333,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.158249445,"imp":11.521496407,"elo":1590.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2075493916,"imp":18.6141876837,"elo":1299.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2319353399,"imp":16.5838062669,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.8843761148,"imp":15.7036315908,"elo":1451.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4644557714,"imp":12.7786145967,"elo":1584.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3605764018,"imp":11.3967383289,"elo":1660.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.275897876,"imp":21.5704284384,"elo":1252.9,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5144045305,"imp":20.8010702984,"elo":1279.8,"imputed":true,"imputed_pct":33.3333333333,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":27.0418336147,"imp":19.1847130669,"elo":1323.7,"imputed":true,"imputed_pct":33.3333333333,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0517146513,"imp":29.2137160166,"elo":1065.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6668093955,"imp":24.4760442869,"elo":1164.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0783983006,"imp":24.6038335984,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0586280558,"imp":17.4848237608,"elo":1364.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0914684387,"imp":24.0028834199,"elo":1098.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1361580238,"imp":20.3274685266,"elo":1220.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4148361224,"imp":18.3054758967,"elo":1302.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1102748358,"imp":20.6054527626,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1302082399,"imp":18.087814544,"elo":1330.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6209770484,"imp":17.598318978,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":43.847258449,"imp":23.1660443357,"elo":1170.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":52.8754683564,"imp":19.23113942,"elo":1289.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":336.0020690595,"imp":16.511425795,"elo":1401.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.5638730428,"imp":30.5456425989,"elo":983.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.146245996,"imp":19.5744677973,"elo":1285.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.7662131303,"imp":18.4801869209,"elo":1310.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7076126445,"x_train":250.3856225541,"imp":18.4866613079,"elo":1364.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0203827935,"x_train":3.0406505418,"imp":16.5036930817,"elo":1390.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.028876079,"x_train":738.2612405475,"imp":16.6993938434,"elo":1385.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.245609885,"x_train":738.2612405475,"imp":16.5359003279,"elo":1392.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0261083185,"x_train":1.2471436483,"imp":19.0759241486,"elo":1307.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0271269609,"x_train":281.3883698637,"imp":18.4530000839,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3141545533,"x_train":281.3883698637,"imp":17.9849925604,"elo":1361.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0148652021,"x_train":3.2701795229,"imp":21.149722515,"elo":1239.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0160758722,"x_train":936.7523176703,"imp":20.3752775842,"elo":1262.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1335297064,"x_train":936.7523176703,"imp":19.7026453378,"elo":1294.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1905448885,"x_train":1.0867260517,"imp":27.3944661579,"elo":1002.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0583336488,"x_train":182.354525564,"imp":22.453747647,"elo":1161.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6857611443,"x_train":182.354525564,"imp":21.6423356218,"elo":1184.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2670364168,"x_train":2.556542435,"imp":25.3266879236,"elo":1065.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2760208318,"x_train":559.3818159963,"imp":21.732386306,"elo":1187.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0790081457,"x_train":559.3818159963,"imp":19.8146595822,"elo":1253.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0925963767,"x_train":0.8126978053,"imp":21.3302201521,"elo":1178.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1264589403,"x_train":328.9554302437,"imp":17.5985878233,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1012944014,"x_train":328.9554302437,"imp":16.798785983,"elo":1411.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.2986429509,"x_train":14.2586021738,"imp":18.3824192374,"elo":1365.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.6152636131,"x_train":103.8835881064,"imp":18.9617472986,"elo":1333.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.310422718,"x_train":13.5066187328,"imp":20.5934045385,"elo":1260.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5103151766,"x_train":3383.9090129865,"imp":17.9334509229,"elo":1347.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3903226253,"x_train":3383.9090129865,"imp":18.0961291655,"elo":1344.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7050443034,"x_train":14.4628692945,"imp":24.1522716371,"elo":1140.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0291154934,"x_train":15.7950255214,"imp":30.9746453444,"elo":947.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2900648816,"x_train":117.8541822739,"imp":26.9128783736,"elo":1067.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7725893154,"x_train":117.8541822739,"imp":25.5398251543,"elo":1099.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.02947866,"x_train":0.2544731196,"imp":27.7326064044,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0593784249,"x_train":169.7548111684,"imp":23.6890906685,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.6033358381,"x_train":169.7548111684,"imp":23.033027877,"elo":1151.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.1163701521,"x_train":9.3300735239,"imp":19.4983343958,"elo":1255.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.4146988006,"x_train":2775.9833627967,"imp":16.8394057755,"elo":1378.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":8.8488311223,"x_train":2775.9833627967,"imp":15.2564513691,"elo":1466.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6942704557,"x_train":4.9087216863,"imp":13.8798596518,"elo":1523.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3289068866,"x_train":1924.5007092935,"imp":14.0187885268,"elo":1511.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.1029658055,"x_train":1924.5007092935,"imp":13.4404211141,"elo":1539.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.166690434,"x_train":12.8582127474,"imp":20.7939330249,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"x_train":49.671607539,"imp":20.051398761,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.1992341236,"x_train":4982.6619115707,"imp":17.2246576566,"elo":1340.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":358.7531028481,"x_train":4982.6619115707,"imp":15.8172984497,"elo":1400.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"x_train":1.4436537137,"imp":16.0127527493,"elo":1427.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"x_train":36.4454770138,"imp":4.088125843,"elo":1779.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.128947975,"x_train":0.8504331063,"imp":29.0960083209,"elo":1043.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.7422345138,"x_train":8.0604387769,"imp":15.4296572582,"elo":1442.7,"imputed":true,"imputed_pct":3.3333333333,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.158249445,"x_train":1.7686337423,"imp":11.521496407,"elo":1590.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2075493916,"x_train":5.0292797696,"imp":18.6141876837,"elo":1299.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2319353399,"x_train":1612.0705710762,"imp":16.5838062669,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.8843761148,"x_train":1612.0705710762,"imp":15.7036315908,"elo":1451.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4644557714,"x_train":4.8153059659,"imp":12.7786145967,"elo":1584.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3605764018,"x_train":2.62612376,"imp":11.3967383289,"elo":1660.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.275897876,"x_train":2.7139173522,"imp":21.5704284384,"elo":1252.9,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5144045305,"x_train":3031.9779409514,"imp":20.8010702984,"elo":1279.8,"imputed":true,"imputed_pct":33.3333333333,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":27.0418336147,"x_train":3031.9779409514,"imp":19.1847130669,"elo":1323.7,"imputed":true,"imputed_pct":33.3333333333,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0517146513,"x_train":391.1080876361,"imp":29.2137160166,"elo":1065.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6668093955,"x_train":25373.3350332343,"imp":24.4760442869,"elo":1164.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0783983006,"x_train":25373.3350332343,"imp":24.6038335984,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0586280558,"x_train":1.1266327909,"imp":17.4848237608,"elo":1364.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0914684387,"x_train":4.9502582412,"imp":24.0028834199,"elo":1098.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1361580238,"x_train":2241.9244610212,"imp":20.3274685266,"elo":1220.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4148361224,"x_train":2241.9244610212,"imp":18.3054758967,"elo":1302.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1102748358,"x_train":1.2379804347,"imp":20.6054527626,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1302082399,"x_train":473.0394868517,"imp":18.087814544,"elo":1330.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6209770484,"x_train":473.0394868517,"imp":17.598318978,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":43.847258449,"x_train":210.3752328358,"imp":23.1660443357,"elo":1170.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":52.8754683564,"x_train":8305.0663055001,"imp":19.23113942,"elo":1289.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":336.0020690595,"x_train":8305.0663055001,"imp":16.511425795,"elo":1401.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.5638730428,"x_train":2.8035263333,"imp":30.5456425989,"elo":983.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.146245996,"x_train":642.4042496517,"imp":19.5744677973,"elo":1285.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.7662131303,"x_train":642.4042496517,"imp":18.4801869209,"elo":1310.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_all/tuning_trajectories_explorer.html
index 14d67b2cce599ae9b305743689b6b487d9dc1475..a10aaa51f0f9a656f9bcd4284e580ca858b7cb37 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":250.3856225541,"imp":18.5111970608,"elo":1385.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.0406505418,"imp":16.5270112491,"elo":1412.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":7.9183650076,"imp":16.4334019835,"elo":1418.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":18.1594886071,"imp":16.3811982127,"elo":1424.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":36.3465094416,"imp":16.4058771563,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":92.6292744451,"imp":16.4110662045,"elo":1423.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":181.7510780205,"imp":16.4252984486,"elo":1422.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":367.6200398568,"imp":16.4877137011,"elo":1417.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":549.0966445079,"imp":16.5264697626,"elo":1415.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":738.2612405475,"imp":16.553941136,"elo":1412.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.2471436483,"imp":19.0988083965,"elo":1325.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.6592018735,"imp":18.6974628967,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":6.827854614,"imp":18.5092763707,"elo":1362.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":13.7690635449,"imp":18.4053191774,"elo":1367.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":34.8312631967,"imp":18.2543960325,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":69.9560479767,"imp":18.1641596143,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":140.0107160319,"imp":18.0700408006,"elo":1384.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":209.935077018,"imp":18.0275380427,"elo":1384.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":281.3883698637,"imp":18.0062162685,"elo":1385.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":3.2701795229,"imp":21.1720780697,"elo":1246.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":8.5418580508,"imp":20.7633239845,"elo":1259.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":21.5154250642,"imp":20.384423713,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":44.8133243453,"imp":20.2339930905,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":115.6024980549,"imp":20.0743228795,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":235.0233869244,"imp":19.9364403384,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":467.0399506352,"imp":19.8109297103,"elo":1305.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":700.0647480226,"imp":19.7742501879,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":936.7523176703,"imp":19.7275399297,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.0867260517,"imp":27.4154470047,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.9830559575,"imp":25.6893174594,"elo":1050.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.7045699739,"imp":23.2640125217,"elo":1121.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.2122172375,"imp":22.3818774997,"elo":1157.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":22.7973551316,"imp":21.9286814389,"elo":1176.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":45.453079637,"imp":21.7649873007,"elo":1185.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":90.7421763607,"imp":21.6620665397,"elo":1192.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":136.0852569857,"imp":21.6342782361,"elo":1195.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":182.354525564,"imp":21.666879142,"elo":1195.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.556542435,"imp":25.3478826669,"elo":1075.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.2460669855,"imp":23.9270205672,"elo":1115.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":13.8120709411,"imp":22.0867101812,"elo":1182.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":28.1488717871,"imp":21.183249725,"elo":1214.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":68.1608030808,"imp":20.5016251032,"elo":1238.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":137.2584674121,"imp":20.1732327337,"elo":1253.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":277.1007160478,"imp":19.9811805776,"elo":1261.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":416.6378564081,"imp":19.8613802284,"elo":1267.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":559.3818159963,"imp":19.8304970891,"elo":1270.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.8126978053,"imp":21.3519787486,"elo":1182.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":2.859374793,"imp":19.3397609558,"elo":1276.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":7.6235569739,"imp":18.0741617029,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":15.609269256,"imp":17.4809829732,"elo":1397.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":40.8449535661,"imp":17.104912897,"elo":1419.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":81.9221899435,"imp":16.9740917089,"elo":1427.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":163.7675316385,"imp":16.8635815739,"elo":1433.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":244.6291470871,"imp":16.823642539,"elo":1434.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":328.9554302437,"imp":16.8241072376,"elo":1434.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":14.2586021738,"imp":18.2860274804,"elo":1389.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":103.8835881064,"imp":18.9842526789,"elo":1353.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":13.5066187328,"imp":20.6156999998,"elo":1274.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":29.3976468222,"imp":19.3998226824,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":85.706508499,"imp":17.4199530509,"elo":1388.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":168.9559526652,"imp":17.0254574242,"elo":1403.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":408.9287781105,"imp":16.6511532015,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":844.0960252495,"imp":16.9083960353,"elo":1408.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1679.9964089232,"imp":17.4406345999,"elo":1389.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":2502.3959153065,"imp":17.8863602236,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":3383.9090129865,"imp":18.128443934,"elo":1366.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":14.4628692945,"imp":24.1566797405,"elo":1154.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":15.7950255214,"imp":31.8068760738,"elo":944.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":37.0036028141,"imp":29.4429222479,"elo":1008.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":98.9757470004,"imp":25.6741135543,"elo":1106.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":117.8541822739,"imp":25.5571490291,"elo":1112.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.2544731196,"imp":27.7533860784,"elo":1000.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":1.1022482459,"imp":25.4576678318,"elo":1064.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":3.7036683225,"imp":24.3134445208,"elo":1104.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":7.9903720898,"imp":23.7552895931,"elo":1127.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":20.5657356408,"imp":23.3034561218,"elo":1147.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":41.6049917756,"imp":23.1679043014,"elo":1154.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":83.9433065314,"imp":23.0813286239,"elo":1159.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":126.3785182217,"imp":23.0412686419,"elo":1161.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":169.7548111684,"imp":23.052583209,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":9.3300735239,"imp":19.5198891163,"elo":1264.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":23.3699912915,"imp":17.6348857866,"elo":1362.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":64.116353982,"imp":16.3851967165,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":135.6738391001,"imp":15.9236429503,"elo":1456.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":343.9787328013,"imp":15.56949449,"elo":1476.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":692.1853967506,"imp":15.439256828,"elo":1484.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1383.1114451594,"imp":15.3432371953,"elo":1489.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2072.6413859802,"imp":15.3010681638,"elo":1492.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2775.9833627967,"imp":15.2811888887,"elo":1492.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":4.9087216863,"imp":13.9047820892,"elo":1541.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.4525053202,"imp":13.9152567863,"elo":1542.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":44.7436120415,"imp":13.7443412476,"elo":1547.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":91.7548974847,"imp":13.6802290466,"elo":1548.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":236.7117395728,"imp":13.5312350041,"elo":1556.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":474.1680372395,"imp":13.5015583458,"elo":1560.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":957.1717702231,"imp":13.4503391004,"elo":1562.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1438.8272594182,"imp":13.4504659659,"elo":1561.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1924.5007092935,"imp":13.4734310617,"elo":1558.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":12.8582127474,"imp":20.815132495,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":49.671607539,"imp":20.0729654842,"elo":1282.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":74.371713941,"imp":18.9534324664,"elo":1313.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":135.8465940794,"imp":17.8977681859,"elo":1352.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":245.5890330499,"imp":17.5332862751,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":603.2630243511,"imp":16.7059357489,"elo":1394.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1228.1552324622,"imp":16.3144914526,"elo":1408.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2466.8687838379,"imp":16.0819109683,"elo":1416.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3715.4039160262,"imp":15.9139047975,"elo":1421.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4982.6619115707,"imp":15.8457948453,"elo":1422.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":1.4436537137,"imp":16.0352282826,"elo":1444.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":36.4454770138,"imp":4.1073062652,"elo":1802.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.8504331063,"imp":29.1157731089,"elo":1060.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":8.0604387769,"imp":15.4537641455,"elo":1462.5,"imputed":true,"imputed_pct":3.3333333333,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.7686337423,"imp":11.5469105451,"elo":1613.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":5.0292797696,"imp":18.6370288069,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":12.3959630427,"imp":18.1247027206,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":37.7043962113,"imp":17.4375547709,"elo":1378.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":76.792938919,"imp":16.8386880009,"elo":1414.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":197.2453170668,"imp":16.3258651178,"elo":1439.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":398.0123545957,"imp":16.0516641462,"elo":1454.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":792.8708351593,"imp":15.8590111296,"elo":1465.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1198.5048285012,"imp":15.7707952166,"elo":1471.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1612.0705710762,"imp":15.7259220424,"elo":1475.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":4.8153059659,"imp":12.8031804013,"elo":1611.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":2.62612376,"imp":11.4163926831,"elo":1686.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2.7139173522,"imp":21.5920320722,"elo":1267.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":22.8408812219,"imp":20.9197666461,"elo":1293.9,"imputed":true,"imputed_pct":33.3333333333,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":63.3712068389,"imp":20.2778558895,"elo":1320.9,"imputed":true,"imputed_pct":33.3333333333,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":143.4062804674,"imp":19.8354791071,"elo":1333.0,"imputed":true,"imputed_pct":33.3333333333,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":359.6193496539,"imp":19.5379125503,"elo":1339.4,"imputed":true,"imputed_pct":33.3333333333,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":730.2395031165,"imp":19.3599517271,"elo":1344.1,"imputed":true,"imputed_pct":33.3333333333,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1487.9333897377,"imp":19.2627186282,"elo":1345.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2241.2008934422,"imp":19.2169418551,"elo":1346.3,"imputed":true,"imputed_pct":33.3333333333,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3031.9779409514,"imp":19.1924549253,"elo":1345.0,"imputed":true,"imputed_pct":33.3333333333,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":391.1080876361,"imp":29.2337776953,"elo":1077.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":569.6608345335,"imp":29.118754218,"elo":1083.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":976.9745445733,"imp":28.879253093,"elo":1089.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1509.8128718211,"imp":28.2236589123,"elo":1104.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3235.1736621912,"imp":27.0929236101,"elo":1125.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":6343.9144626669,"imp":25.9597966717,"elo":1144.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":12590.6623522261,"imp":25.3437389603,"elo":1159.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":18827.9881620466,"imp":24.9479584614,"elo":1168.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":25373.3350332343,"imp":24.6258376522,"elo":1178.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1266327909,"imp":17.508540473,"elo":1385.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":4.9502582412,"imp":24.0149190951,"elo":1109.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.9663495618,"imp":21.9720169683,"elo":1169.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":44.495543604,"imp":20.3310907669,"elo":1219.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":110.9528145485,"imp":19.6628416187,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":276.2432638721,"imp":19.0884718359,"elo":1279.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":564.391803329,"imp":18.7255189572,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1102.6775922424,"imp":18.5349571312,"elo":1311.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1669.2077718784,"imp":18.4238206861,"elo":1319.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2241.9244610212,"imp":18.3199705921,"elo":1324.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.2379804347,"imp":20.6276491525,"elo":1230.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":4.1675681573,"imp":18.7633778308,"elo":1317.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":11.5413095581,"imp":18.2737874481,"elo":1347.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":22.4971330897,"imp":18.0013496097,"elo":1362.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":58.5886356918,"imp":17.7903010927,"elo":1372.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":117.0352585562,"imp":17.7307810217,"elo":1375.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":234.6013258332,"imp":17.6810419709,"elo":1379.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":351.8557074643,"imp":17.6446804084,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":473.0394868517,"imp":17.6221934952,"elo":1383.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":210.3752328358,"imp":23.1872448085,"elo":1181.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":506.0952070543,"imp":21.1043870904,"elo":1232.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":1471.9943120247,"imp":18.803090198,"elo":1319.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":3204.4293403006,"imp":17.6648996565,"elo":1369.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":7937.3877327789,"imp":16.5574510924,"elo":1420.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":8305.0663055001,"imp":16.5335931357,"elo":1423.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.8035263333,"imp":30.5657922317,"elo":990.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":6.6149535103,"imp":26.2538947515,"elo":1063.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":16.126759698,"imp":21.0690407957,"elo":1207.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":31.9971507614,"imp":19.0904969924,"elo":1303.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":79.9925409669,"imp":18.4970070655,"elo":1335.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":159.8382634935,"imp":18.3596906649,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":319.7915016575,"imp":18.3965392375,"elo":1336.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":479.1842606686,"imp":18.4441154494,"elo":1333.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":642.4042496517,"imp":18.4883782672,"elo":1329.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7076126445,"x_train":250.3856225541,"imp":18.5111970608,"elo":1385.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0203827935,"x_train":3.0406505418,"imp":16.5270112491,"elo":1412.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0439021895,"x_train":7.9183650076,"imp":16.4334019835,"elo":1418.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0872974817,"x_train":18.1594886071,"imp":16.3811982127,"elo":1424.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1177578623,"x_train":36.3465094416,"imp":16.4058771563,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1576011846,"x_train":92.6292744451,"imp":16.4110662045,"elo":1423.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1905796825,"x_train":181.7510780205,"imp":16.4252984486,"elo":1422.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2216909575,"x_train":367.6200398568,"imp":16.4877137011,"elo":1417.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.233726561,"x_train":549.0966445079,"imp":16.5264697626,"elo":1415.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2505091082,"x_train":738.2612405475,"imp":16.553941136,"elo":1412.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0261083185,"x_train":1.2471436483,"imp":19.0988083965,"elo":1325.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0519030406,"x_train":2.6592018735,"imp":18.6974628967,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1047084686,"x_train":6.827854614,"imp":18.5092763707,"elo":1362.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1608995496,"x_train":13.7690635449,"imp":18.4053191774,"elo":1367.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2235785875,"x_train":34.8312631967,"imp":18.2543960325,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.268731958,"x_train":69.9560479767,"imp":18.1641596143,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2953259759,"x_train":140.0107160319,"imp":18.0700408006,"elo":1384.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3063723891,"x_train":209.935077018,"imp":18.0275380427,"elo":1384.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3135203066,"x_train":281.3883698637,"imp":18.0062162685,"elo":1385.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0148652021,"x_train":3.2701795229,"imp":21.1720780697,"elo":1246.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0298750396,"x_train":8.5418580508,"imp":20.7633239845,"elo":1259.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0523006922,"x_train":21.5154250642,"imp":20.384423713,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0707230342,"x_train":44.8133243453,"imp":20.2339930905,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0955594836,"x_train":115.6024980549,"imp":20.0743228795,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1108747895,"x_train":235.0233869244,"imp":19.9364403384,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1241247455,"x_train":467.0399506352,"imp":19.8109297103,"elo":1305.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1273889013,"x_train":700.0647480226,"imp":19.7742501879,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1283935846,"x_train":936.7523176703,"imp":19.7275399297,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1905448885,"x_train":1.0867260517,"imp":27.4154470047,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2264116831,"x_train":1.9830559575,"imp":25.6893174594,"elo":1050.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2899539484,"x_train":4.7045699739,"imp":23.2640125217,"elo":1121.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3380187835,"x_train":9.2122172375,"imp":22.3818774997,"elo":1157.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4229346143,"x_train":22.7973551316,"imp":21.9286814389,"elo":1176.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5030792517,"x_train":45.453079637,"imp":21.7649873007,"elo":1185.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5833685615,"x_train":90.7421763607,"imp":21.6620665397,"elo":1192.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6430515372,"x_train":136.0852569857,"imp":21.6342782361,"elo":1195.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6917689999,"x_train":182.354525564,"imp":21.666879142,"elo":1195.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2670364168,"x_train":2.556542435,"imp":25.3478826669,"elo":1075.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4726706959,"x_train":5.2460669855,"imp":23.9270205672,"elo":1115.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9145377265,"x_train":13.8120709411,"imp":22.0867101812,"elo":1182.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4648401665,"x_train":28.1488717871,"imp":21.183249725,"elo":1214.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0889905192,"x_train":68.1608030808,"imp":20.5016251032,"elo":1238.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.6339038812,"x_train":137.2584674121,"imp":20.1732327337,"elo":1253.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.3191961162,"x_train":277.1007160478,"imp":19.9811805776,"elo":1261.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.7369178253,"x_train":416.6378564081,"imp":19.8613802284,"elo":1267.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0812529125,"x_train":559.3818159963,"imp":19.8304970891,"elo":1270.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0925963767,"x_train":0.8126978053,"imp":21.3519787486,"elo":1182.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4060297968,"x_train":2.859374793,"imp":19.3397609558,"elo":1276.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6600545749,"x_train":7.6235569739,"imp":18.0741617029,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8849225155,"x_train":15.609269256,"imp":17.4809829732,"elo":1397.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1404380746,"x_train":40.8449535661,"imp":17.104912897,"elo":1419.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2644666214,"x_train":81.9221899435,"imp":16.9740917089,"elo":1427.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3003023736,"x_train":163.7675316385,"imp":16.8635815739,"elo":1433.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2279060483,"x_train":244.6291470871,"imp":16.823642539,"elo":1434.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1375151455,"x_train":328.9554302437,"imp":16.8241072376,"elo":1434.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.2986429509,"x_train":14.2586021738,"imp":18.2860274804,"elo":1389.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.6152636131,"x_train":103.8835881064,"imp":18.9842526789,"elo":1353.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.310422718,"x_train":13.5066187328,"imp":20.6156999998,"elo":1274.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8103899414,"x_train":29.3976468222,"imp":19.3998226824,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8318086975,"x_train":85.706508499,"imp":17.4199530509,"elo":1388.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7115644528,"x_train":168.9559526652,"imp":17.0254574242,"elo":1403.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.0880667933,"x_train":408.9287781105,"imp":16.6511532015,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.4716607961,"x_train":844.0960252495,"imp":16.9083960353,"elo":1408.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.9921274944,"x_train":1679.9964089232,"imp":17.4406345999,"elo":1389.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8596602241,"x_train":2502.3959153065,"imp":17.8863602236,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3848981892,"x_train":3383.9090129865,"imp":18.128443934,"elo":1366.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7050443034,"x_train":14.4628692945,"imp":24.1566797405,"elo":1154.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0291154934,"x_train":15.7950255214,"imp":31.8068760738,"elo":944.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2360786,"x_train":37.0036028141,"imp":29.4429222479,"elo":1008.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6732920785,"x_train":98.9757470004,"imp":25.6741135543,"elo":1106.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7725893154,"x_train":117.8541822739,"imp":25.5571490291,"elo":1112.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.02947866,"x_train":0.2544731196,"imp":27.7533860784,"elo":1000.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0884016237,"x_train":1.1022482459,"imp":25.4576678318,"elo":1064.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1791844925,"x_train":3.7036683225,"imp":24.3134445208,"elo":1104.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2574024682,"x_train":7.9903720898,"imp":23.7552895931,"elo":1127.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3681557693,"x_train":20.5657356408,"imp":23.3034561218,"elo":1147.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.462216789,"x_train":41.6049917756,"imp":23.1679043014,"elo":1154.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5500877206,"x_train":83.9433065314,"imp":23.0813286239,"elo":1159.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5963022541,"x_train":126.3785182217,"imp":23.0412686419,"elo":1161.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6073439144,"x_train":169.7548111684,"imp":23.052583209,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1163701521,"x_train":9.3300735239,"imp":19.5198891163,"elo":1264.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5367213253,"x_train":23.3699912915,"imp":17.6348857866,"elo":1362.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.4240428043,"x_train":64.116353982,"imp":16.3851967165,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.3826374553,"x_train":135.6738391001,"imp":15.9236429503,"elo":1456.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.7349492529,"x_train":343.9787328013,"imp":15.56949449,"elo":1476.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.2159141116,"x_train":692.1853967506,"imp":15.439256828,"elo":1484.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.3769693441,"x_train":1383.1114451594,"imp":15.3432371953,"elo":1489.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.3169488161,"x_train":2072.6413859802,"imp":15.3010681638,"elo":1492.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.7763731411,"x_train":2775.9833627967,"imp":15.2811888887,"elo":1492.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6942704557,"x_train":4.9087216863,"imp":13.9047820892,"elo":1541.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7020387497,"x_train":16.4525053202,"imp":13.9152567863,"elo":1542.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.0703931915,"x_train":44.7436120415,"imp":13.7443412476,"elo":1547.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1355887912,"x_train":91.7548974847,"imp":13.6802290466,"elo":1548.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7695458497,"x_train":236.7117395728,"imp":13.5312350041,"elo":1556.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.7430322432,"x_train":474.1680372395,"imp":13.5015583458,"elo":1560.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9819714023,"x_train":957.1717702231,"imp":13.4503391004,"elo":1562.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7612085046,"x_train":1438.8272594182,"imp":13.4504659659,"elo":1561.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.2115603982,"x_train":1924.5007092935,"imp":13.4734310617,"elo":1558.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.166690434,"x_train":12.8582127474,"imp":20.815132495,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"x_train":49.671607539,"imp":20.0729654842,"elo":1282.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":59.5820773592,"x_train":74.371713941,"imp":18.9534324664,"elo":1313.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":88.9924523268,"x_train":135.8465940794,"imp":17.8977681859,"elo":1352.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":119.9760691608,"x_train":245.5890330499,"imp":17.5332862751,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":184.7836530334,"x_train":603.2630243511,"imp":16.7059357489,"elo":1394.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":246.0445440803,"x_train":1228.1552324622,"imp":16.3144914526,"elo":1408.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":305.831978607,"x_train":2466.8687838379,"imp":16.0819109683,"elo":1416.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":338.3243282561,"x_train":3715.4039160262,"imp":15.9139047975,"elo":1421.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":357.7828705921,"x_train":4982.6619115707,"imp":15.8457948453,"elo":1422.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"x_train":1.4436537137,"imp":16.0352282826,"elo":1444.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"x_train":36.4454770138,"imp":4.1073062652,"elo":1802.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.128947975,"x_train":0.8504331063,"imp":29.1157731089,"elo":1060.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.7422345138,"x_train":8.0604387769,"imp":15.4537641455,"elo":1462.5,"imputed":true,"imputed_pct":3.3333333333,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.158249445,"x_train":1.7686337423,"imp":11.5469105451,"elo":1613.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2075493916,"x_train":5.0292797696,"imp":18.6370288069,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.3669465558,"x_train":12.3959630427,"imp":18.1247027206,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.641044718,"x_train":37.7043962113,"imp":17.4375547709,"elo":1378.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.8182662875,"x_train":76.792938919,"imp":16.8386880009,"elo":1414.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.0673724766,"x_train":197.2453170668,"imp":16.3258651178,"elo":1439.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.2587716295,"x_train":398.0123545957,"imp":16.0516641462,"elo":1454.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.5504237308,"x_train":792.8708351593,"imp":15.8590111296,"elo":1465.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.7273482778,"x_train":1198.5048285012,"imp":15.7707952166,"elo":1471.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":1.8903034491,"x_train":1612.0705710762,"imp":15.7259220424,"elo":1475.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4644557714,"x_train":4.8153059659,"imp":12.8031804013,"elo":1611.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3605764018,"x_train":2.62612376,"imp":11.4163926831,"elo":1686.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.275897876,"x_train":2.7139173522,"imp":21.5920320722,"elo":1267.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":3.1145347607,"x_train":22.8408812219,"imp":20.9197666461,"elo":1293.9,"imputed":true,"imputed_pct":33.3333333333,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":5.9565674909,"x_train":63.3712068389,"imp":20.2778558895,"elo":1320.9,"imputed":true,"imputed_pct":33.3333333333,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":9.7060065027,"x_train":143.4062804674,"imp":19.8354791071,"elo":1333.0,"imputed":true,"imputed_pct":33.3333333333,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":13.9847723718,"x_train":359.6193496539,"imp":19.5379125503,"elo":1339.4,"imputed":true,"imputed_pct":33.3333333333,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.1162730103,"x_train":730.2395031165,"imp":19.3599517271,"elo":1344.1,"imputed":true,"imputed_pct":33.3333333333,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":23.389743329,"x_train":1487.9333897377,"imp":19.2627186282,"elo":1345.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":25.2628659914,"x_train":2241.2008934422,"imp":19.2169418551,"elo":1346.3,"imputed":true,"imputed_pct":33.3333333333,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":26.5617852255,"x_train":3031.9779409514,"imp":19.1924549253,"elo":1345.0,"imputed":true,"imputed_pct":33.3333333333,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0517146513,"x_train":391.1080876361,"imp":29.2337776953,"elo":1077.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5372670149,"x_train":569.6608345335,"imp":29.118754218,"elo":1083.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.680399048,"x_train":976.9745445733,"imp":28.879253093,"elo":1089.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5405477526,"x_train":1509.8128718211,"imp":28.2236589123,"elo":1104.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5104627144,"x_train":3235.1736621912,"imp":27.0929236101,"elo":1125.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.5720442233,"x_train":6343.9144626669,"imp":25.9597966717,"elo":1144.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.437201778,"x_train":12590.6623522261,"imp":25.3437389603,"elo":1159.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":13.8939261019,"x_train":18827.9881620466,"imp":24.9479584614,"elo":1168.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0865396805,"x_train":25373.3350332343,"imp":24.6258376522,"elo":1178.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0586280558,"x_train":1.1266327909,"imp":17.508540473,"elo":1385.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0914684387,"x_train":4.9502582412,"imp":24.0149190951,"elo":1109.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1917076474,"x_train":13.9663495618,"imp":21.9720169683,"elo":1169.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4742459424,"x_train":44.495543604,"imp":20.3310907669,"elo":1219.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7489571843,"x_train":110.9528145485,"imp":19.6628416187,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1980104117,"x_train":276.2432638721,"imp":19.0884718359,"elo":1279.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6220136736,"x_train":564.391803329,"imp":18.7255189572,"elo":1299.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8953030401,"x_train":1102.6775922424,"imp":18.5349571312,"elo":1311.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.2031294127,"x_train":1669.2077718784,"imp":18.4238206861,"elo":1319.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4133616403,"x_train":2241.9244610212,"imp":18.3199705921,"elo":1324.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1102748358,"x_train":1.2379804347,"imp":20.6276491525,"elo":1230.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2662387933,"x_train":4.1675681573,"imp":18.7633778308,"elo":1317.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3317369138,"x_train":11.5413095581,"imp":18.2737874481,"elo":1347.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3713912294,"x_train":22.4971330897,"imp":18.0013496097,"elo":1362.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4711698488,"x_train":58.5886356918,"imp":17.7903010927,"elo":1372.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5335858253,"x_train":117.0352585562,"imp":17.7307810217,"elo":1375.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5867602292,"x_train":234.6013258332,"imp":17.6810419709,"elo":1379.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6015304474,"x_train":351.8557074643,"imp":17.6446804084,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6276540518,"x_train":473.0394868517,"imp":17.6221934952,"elo":1383.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":43.847258449,"x_train":210.3752328358,"imp":23.1872448085,"elo":1181.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":92.0045040051,"x_train":506.0952070543,"imp":21.1043870904,"elo":1232.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":160.2377449485,"x_train":1471.9943120247,"imp":18.803090198,"elo":1319.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":248.0978697609,"x_train":3204.4293403006,"imp":17.6648996565,"elo":1369.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":328.1842391713,"x_train":7937.3877327789,"imp":16.5574510924,"elo":1420.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":336.0020690595,"x_train":8305.0663055001,"imp":16.5335931357,"elo":1423.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.5638730428,"x_train":2.8035263333,"imp":30.5657922317,"elo":990.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.803767519,"x_train":6.6149535103,"imp":26.2538947515,"elo":1063.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.9568557255,"x_train":16.126759698,"imp":21.0690407957,"elo":1207.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.3168671243,"x_train":31.9971507614,"imp":19.0904969924,"elo":1303.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.7124473326,"x_train":79.9925409669,"imp":18.4970070655,"elo":1335.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2004675665,"x_train":159.8382634935,"imp":18.3596906649,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.497744625,"x_train":319.7915016575,"imp":18.3965392375,"elo":1336.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.6993276206,"x_train":479.1842606686,"imp":18.4441154494,"elo":1333.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.7587528196,"x_train":642.4042496517,"imp":18.4883782672,"elo":1329.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..8f2b4b2df39d006ecf852ac00e8acd73182492d2
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_binary/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "TabDPT-Turbo (default)", "AutoGluon 1.4 (best, 4h)", "RealMLP (tuned + ensembled)", "TabICL (default)", "TabM (tuned + ensembled)", "LimiX (default)", "TabDPT (tuned + ensembled)", "TabSwift (default)", "Mitra (default)", "LightGBM (tuned + ensembled)", "BetaTabPFN (default)", "TabM (tuned)", "TabPFNv2 (tuned + ensembled)", "TabDPT (tuned)", "RealMLP (tuned)", "LightGBM (tuned)", "iLTM (tuned + ensembled)", "CatBoost (default)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "TabPFNv2 (default)", "ModernNCA (tuned)", "TabDPT (default)", "xRFM (tuned + ensembled)", "xRFM (tuned)", "TabPFNv2 (tuned)", "XGBoost (tuned + ensembled)", "SAP-RPT-OSS (default)", "ChimeraBoost (tuned + ensembled)", "EBM (tuned + ensembled)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "ModernNCA (default)", "XGBoost (tuned)", "TabM (default)", "ChimeraBoost (tuned)", "RealMLP (default)", "EBM (tuned)", "ModernNCA (tuned + ensembled)", "EBM (default)", "ChimeraBoost (default)", "iLTM (tuned)", "FastaiMLP (tuned)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "TorchMLP (tuned)", "OrionMSP (default)", "ExtraTrees (tuned + ensembled)", "ExtraTrees (tuned)", "XGBoost (default)", "iLTM (default)", "TabFlex (default)", "LightGBM (default)", "TabSTAR (default)", "Linear (tuned + ensembled)", "TorchMLP (default)", "RandomForest (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "Linear (tuned)", "FastaiMLP (default)", "RandomForest (tuned)", "PerpetualBooster (tuned)", "Linear (default)", "KNN (tuned + ensembled)", "xRFM (default)", "ExtraTrees (default)", "RandomForest (default)", "PerpetualBooster (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7968036532402039, 0.7557077407836914, 0.810502290725708, 0.8196346759796143, 0.8150684833526611, 0.8299086689949036, 0.835616409778595, 0.8378995656967163, 0.8561643958091736, 0.8767123222351074, 0.922374427318573, 0.9063926935195923, 0.905251145362854, 0.8972602486610413, 0.8675799369812012, 0.8561643958091736, 0.888127863407135, 0.8664383292198181, 0.9383561611175537, 0.8972602486610413, 0.8926940560340881, 0.8721461296081543, 0.9109588861465454, 0.9041095972061157, 0.8949771523475647, 0.8926940560340881, 0.9018265008926392, 0.8310502171516418, 0.913241982460022, 0.9018265008926392, 0.9041095972061157, 0.8835616707801819, 0.8744292259216309, 0.9360730648040771, 0.8904109597206116, 0.9178082346916199, 0.9086757898330688, 0.8972602486610413, 0.9543378949165344, 0.8972602486610413, 0.9452054500579834, 0.9360730648040771, 0.9292237162590027, 0.913241982460022, 0.9178082346916199, 0.9406392574310303, 0.9109588861465454, 0.9383561611175537, 0.9269406199455261, 0.9269406199455261, 0.8196346759796143, 0.8219178318977356, 0.9680365324020386, 0.9429223537445068, 0.9520547986030579, 0.9429223537445068, 0.9703196287155151, 0.9406392574310303, 0.9269406199455261, 0.9726027250289917, 0.853881299495697, 0.9246575236320496, 0.9771689772605896, 0.9771689772605896, 0.9817351698875427, 0.9337899684906006, 0.956620991230011, 0.9748858213424683, 0.9840182662010193, 0.9543378949165344, 0.9726027250289917, 0.9748858213424683, 0.9840182662010193, 0.9771689772605896, 0.9885844588279724, 0.9726027250289917, 0.9931507110595703], [0.20319634675979614, NaN, 0.5958904027938843, 0.5114155411720276, 0.655251145362854, 0.7031963467597961, 0.6940639019012451, 0.7191780805587769, 0.7351598143577576, 0.7191780805587769, 0.7899543642997742, 0.8116438388824463, 0.7922374606132507, 0.7739726305007935, 0.8196346759796143, 0.8310502171516418, 0.7808219194412231, 0.7853881120681763, 0.7728310227394104, 0.8378995656967163, 0.7968036532402039, 0.8493150472640991, 0.810502290725708, 0.8424657583236694, 0.8493150472640991, 0.8310502171516418, 0.8242009282112122, 0.8333333134651184, 0.7796803712844849, 0.8607305884361267, 0.8470319509506226, 0.8424657583236694, 0.8150684833526611, 0.8504565954208374, 0.8481734991073608, 0.8287671208381653, 0.8744292259216309, 0.844748854637146, 0.8413242101669312, 0.888127863407135, 0.8504565954208374, 0.8721461296081543, 0.8904109597206116, 0.878995418548584, 0.8812785148620605, 0.8630136847496033, 0.9109588861465454, 0.8641552329063416, 0.8698630332946777, 0.905251145362854, 0.8732876777648926, 0.784246563911438, 0.7968036532402039, 0.9337899684906006, 0.9018265008926392, 0.9383561611175537, 0.9337899684906006, 0.9406392574310303, 0.9292237162590027, 0.9041095972061157, 0.9543378949165344, 0.8310502171516418, 0.878995418548584, 0.9474886059761047, 0.9383561611175537, 0.9360730648040771, 0.8904109597206116, 0.9406392574310303, 0.9611872434616089, 0.9520547986030579, 0.931506872177124, 0.9383561611175537, 0.9543378949165344, 0.9703196287155151, 0.9748858213424683, 0.9703196287155151, 0.9634703397750854, 0.9954338073730469], [0.2442922443151474, 0.4041095972061157, NaN, 0.456620991230011, 0.6484017968177795, 0.6301369667053223, 0.6506849527359009, 0.664383590221405, 0.6073059439659119, 0.77625572681427, 0.8162100315093994, 0.7819634675979614, 0.7990867495536804, 0.7465753555297852, 0.77625572681427, 0.767123281955719, 0.6906392574310303, 0.8310502171516418, 0.7579908967018127, 0.818493127822876, 0.7488584518432617, 0.7945205569267273, 0.8470319509506226, 0.8401826620101929, 0.8744292259216309, 0.8436073064804077, 0.853881299495697, 0.8436073064804077, 0.7420091032981873, 0.8698630332946777, 0.77625572681427, 0.8675799369812012, 0.8230593800544739, 0.7899543642997742, 0.878995418548584, 0.8036529421806335, 0.888127863407135, 0.8401826620101929, 0.8561643958091736, 0.8995434045791626, 0.835616409778595, 0.8926940560340881, 0.878995418548584, 0.8972602486610413, 0.853881299495697, 0.8470319509506226, 0.9337899684906006, 0.8561643958091736, 0.9018265008926392, 0.8926940560340881, 0.8652967810630798, 0.7465753555297852, 0.7488584518432617, 0.9200913310050964, 0.905251145362854, 0.9246575236320496, 0.9086757898330688, 0.9337899684906006, 0.922374427318573, 0.8812785148620605, 0.9474886059761047, 0.801369845867157, 0.8744292259216309, 0.9474886059761047, 0.9429223537445068, 0.9360730648040771, 0.9018265008926392, 0.9383561611175537, 0.9452054500579834, 0.9509132504463196, 0.9257990717887878, 0.9269406199455261, 0.9429223537445068, 0.9497717022895813, 0.9520547986030579, 0.9634703397750854, 0.9452054500579834, 0.9931507110595703], [0.18949772417545319, 0.4885844886302948, 0.543379008769989, NaN, 0.6004565954208374, 0.6312785148620605, 0.646118700504303, 0.664383590221405, 0.7191780805587769, 0.7191780805587769, 0.7442922592163086, 0.8025113940238953, 0.7579908967018127, 0.784246563911438, 0.8139269351959229, 0.8082191944122314, 0.7134703397750854, 0.7648401856422424, 0.7739726305007935, 0.7945205569267273, 0.7899543642997742, 0.8127853870391846, 0.7716894745826721, 0.8036529421806335, 0.8150684833526611, 0.7968036532402039, 0.7853881120681763, 0.7945205569267273, 0.7625570893287659, 0.8333333134651184, 0.8401826620101929, 0.8173515796661377, 0.7910959124565125, 0.8047945499420166, 0.810502290725708, 0.801369845867157, 0.8242009282112122, 0.8196346759796143, 0.8401826620101929, 0.878995418548584, 0.8504565954208374, 0.8287671208381653, 0.8607305884361267, 0.8424657583236694, 0.8698630332946777, 0.844748854637146, 0.8698630332946777, 0.8561643958091736, 0.8378995656967163, 0.8584474921226501, 0.8812785148620605, 0.7922374606132507, 0.7899543642997742, 0.9178082346916199, 0.9246575236320496, 0.9121004343032837, 0.9109588861465454, 0.913241982460022, 0.9086757898330688, 0.8972602486610413, 0.9406392574310303, 0.835616409778595, 0.8949771523475647, 0.956620991230011, 0.9383561611175537, 0.9429223537445068, 0.9086757898330688, 0.931506872177124, 0.9360730648040771, 0.9520547986030579, 0.9372146129608154, 0.9474886059761047, 0.9520547986030579, 0.9520547986030579, 0.956620991230011, 0.9794520735740662, 0.9589040875434875, 0.9954338073730469], [0.18036529421806335, 0.344748854637146, 0.35159817337989807, 0.3995433747768402, NaN, 0.5456621050834656, 0.534246563911438, 0.5673515796661377, 0.6050228476524353, 0.6575342416763306, 0.7214611768722534, 0.7191780805587769, 0.7214611768722534, 0.7054794430732727, 0.7214611768722534, 0.7146118879318237, 0.6826484203338623, 0.7557077407836914, 0.7214611768722534, 0.7785388231277466, 0.7191780805587769, 0.7305936217308044, 0.7831050157546997, 0.8127853870391846, 0.7945205569267273, 0.7739726305007935, 0.7808219194412231, 0.767123281955719, 0.715753436088562, 0.8242009282112122, 0.7191780805587769, 0.8036529421806335, 0.7831050157546997, 0.8059360980987549, 0.7853881120681763, 0.8059360980987549, 0.8493150472640991, 0.801369845867157, 0.7899543642997742, 0.853881299495697, 0.8401826620101929, 0.8310502171516418, 0.8458904027938843, 0.8630136847496033, 0.8561643958091736, 0.8390411138534546, 0.8812785148620605, 0.8242009282112122, 0.8584474921226501, 0.8584474921226501, 0.8424657583236694, 0.7442922592163086, 0.7557077407836914, 0.913241982460022, 0.8515982031822205, 0.913241982460022, 0.905251145362854, 0.9360730648040771, 0.8995434045791626, 0.878995418548584, 0.9474886059761047, 0.7899543642997742, 0.8675799369812012, 0.9497717022895813, 0.9360730648040771, 0.931506872177124, 0.8926940560340881, 0.9109588861465454, 0.9543378949165344, 0.931506872177124, 0.9200913310050964, 0.9269406199455261, 0.9383561611175537, 0.9452054500579834, 0.9611872434616089, 0.9611872434616089, 0.9474886059761047, 0.990867555141449], [0.18493150174617767, 0.29680365324020386, 0.36986300349235535, 0.36872145533561707, 0.4543378949165344, NaN, 0.4817351698875427, 0.5251141786575317, 0.5388127565383911, 0.6506849527359009, 0.6621004343032837, 0.664383590221405, 0.6780821681022644, 0.6849315166473389, 0.6826484203338623, 0.7534246444702148, 0.7009132504463196, 0.7214611768722534, 0.7294520735740662, 0.732876718044281, 0.7351598143577576, 0.715753436088562, 0.7442922592163086, 0.7648401856422424, 0.72374427318573, 0.7625570893287659, 0.7351598143577576, 0.7351598143577576, 0.767123281955719, 0.7557077407836914, 0.7397260069847107, 0.7945205569267273, 0.7876712083816528, 0.7968036532402039, 0.7602739930152893, 0.801369845867157, 0.7990867495536804, 0.7945205569267273, 0.8059360980987549, 0.8264840245246887, 0.8059360980987549, 0.7990867495536804, 0.8082191944122314, 0.8150684833526611, 0.8264840245246887, 0.835616409778595, 0.7831050157546997, 0.8196346759796143, 0.8333333134651184, 0.8458904027938843, 0.853881299495697, 0.7785388231277466, 0.7819634675979614, 0.8972602486610413, 0.913241982460022, 0.8926940560340881, 0.8972602486610413, 0.888127863407135, 0.8926940560340881, 0.9029680490493774, 0.9406392574310303, 0.8127853870391846, 0.853881299495697, 0.9611872434616089, 0.8972602486610413, 0.931506872177124, 0.8652967810630798, 0.9086757898330688, 0.9337899684906006, 0.9452054500579834, 0.9292237162590027, 0.9086757898330688, 0.9429223537445068, 0.9452054500579834, 0.956620991230011, 0.9703196287155151, 0.9452054500579834, 0.9931507110595703], [0.17009133100509644, 0.3059360682964325, 0.3493150770664215, 0.35388126969337463, 0.465753436088562, 0.5182648301124573, NaN, 0.6255707740783691, 0.5319634675979614, 0.6598173379898071, 0.6712328791618347, 0.6678082346916199, 0.6712328791618347, 0.7031963467597961, 0.6666666865348816, 0.7009132504463196, 0.7077625393867493, 0.7260273694992065, 0.72374427318573, 0.7283105254173279, 0.7534246444702148, 0.7168949842453003, 0.7260273694992065, 0.7659817337989807, 0.7340182662010193, 0.7557077407836914, 0.7420091032981873, 0.7465753555297852, 0.7602739930152893, 0.801369845867157, 0.6735159754753113, 0.7785388231277466, 0.7625570893287659, 0.810502290725708, 0.7625570893287659, 0.7728310227394104, 0.7865296602249146, 0.7785388231277466, 0.7773972749710083, 0.8219178318977356, 0.7945205569267273, 0.793379008769989, 0.7990867495536804, 0.8059360980987549, 0.810502290725708, 0.8093607425689697, 0.7990867495536804, 0.8127853870391846, 0.8242009282112122, 0.8196346759796143, 0.8470319509506226, 0.7465753555297852, 0.7534246444702148, 0.878995418548584, 0.8767123222351074, 0.8835616707801819, 0.8767123222351074, 0.888127863407135, 0.8812785148620605, 0.8744292259216309, 0.9337899684906006, 0.7899543642997742, 0.853881299495697, 0.9246575236320496, 0.913241982460022, 0.931506872177124, 0.8744292259216309, 0.913241982460022, 0.9246575236320496, 0.9383561611175537, 0.905251145362854, 0.9109588861465454, 0.9269406199455261, 0.9429223537445068, 0.9452054500579834, 0.9611872434616089, 0.931506872177124, 0.990867555141449], [0.16438356041908264, 0.28082191944122314, 0.33561643958091736, 0.33561643958091736, 0.4326483905315399, 0.47488585114479065, 0.37442922592163086, NaN, 0.5228310227394104, 0.6278538703918457, 0.6324200630187988, 0.6347032189369202, 0.6484017968177795, 0.6621004343032837, 0.6598173379898071, 0.6860730648040771, 0.6963470578193665, 0.7054794430732727, 0.7031963467597961, 0.7111872434616089, 0.7180365324020386, 0.7043378949165344, 0.7077625393867493, 0.7602739930152893, 0.7054794430732727, 0.7283105254173279, 0.7283105254173279, 0.72374427318573, 0.7374429106712341, 0.7465753555297852, 0.6757990717887878, 0.7522830963134766, 0.7442922592163086, 0.7819634675979614, 0.7442922592163086, 0.7602739930152893, 0.759132444858551, 0.7716894745826721, 0.7579908967018127, 0.8082191944122314, 0.7694063782691956, 0.767123281955719, 0.7853881120681763, 0.7922374606132507, 0.7968036532402039, 0.7876712083816528, 0.7888127565383911, 0.7945205569267273, 0.8059360980987549, 0.8036529421806335, 0.8310502171516418, 0.7374429106712341, 0.7488584518432617, 0.8767123222351074, 0.8721461296081543, 0.8687214851379395, 0.8721461296081543, 0.8812785148620605, 0.8698630332946777, 0.8607305884361267, 0.922374427318573, 0.7922374606132507, 0.8333333134651184, 0.922374427318573, 0.888127863407135, 0.9121004343032837, 0.8584474921226501, 0.8995434045791626, 0.913241982460022, 0.9246575236320496, 0.9098173379898071, 0.9041095972061157, 0.9200913310050964, 0.9337899684906006, 0.9406392574310303, 0.9634703397750854, 0.9337899684906006, 0.990867555141449], [0.16210044920444489, 0.26484018564224243, 0.39269405603408813, 0.28082191944122314, 0.3949771821498871, 0.4611872136592865, 0.4680365324020386, 0.4771689474582672, NaN, 0.534246563911438, 0.5730593800544739, 0.5502282977104187, 0.5422374606132507, 0.6232876777648926, 0.6506849527359009, 0.6883561611175537, 0.5753424763679504, 0.5890411138534546, 0.5730593800544739, 0.6073059439659119, 0.6141552329063416, 0.7077625393867493, 0.6678082346916199, 0.6232876777648926, 0.6529680490493774, 0.646118700504303, 0.6689497828483582, 0.664383590221405, 0.646118700504303, 0.6506849527359009, 0.8093607425689697, 0.6940639019012451, 0.7100456357002258, 0.6598173379898071, 0.6803653240203857, 0.6826484203338623, 0.6849315166473389, 0.6940639019012451, 0.7374429106712341, 0.7283105254173279, 0.7260273694992065, 0.698630154132843, 0.698630154132843, 0.7214611768722534, 0.7511415481567383, 0.7203196287155151, 0.6803653240203857, 0.7374429106712341, 0.7351598143577576, 0.759132444858551, 0.7945205569267273, 0.7077625393867493, 0.7168949842453003, 0.7945205569267273, 0.8401826620101929, 0.8458904027938843, 0.878995418548584, 0.8173515796661377, 0.7990867495536804, 0.8721461296081543, 0.8607305884361267, 0.7899543642997742, 0.8082191944122314, 0.878995418548584, 0.8767123222351074, 0.8812785148620605, 0.8287671208381653, 0.9063926935195923, 0.8904109597206116, 0.9041095972061157, 0.9063926935195923, 0.9041095972061157, 0.9292237162590027, 0.9303653240203857, 0.9360730648040771, 0.965753436088562, 0.9520547986030579, 0.9931507110595703], [0.1438356190919876, 0.28082191944122314, 0.22374428808689117, 0.28082191944122314, 0.34246575832366943, 0.3493150770664215, 0.34018266201019287, 0.3721461296081543, 0.465753436088562, NaN, 0.5970319509506226, 0.5, 0.5593607425689697, 0.4817351698875427, 0.543379008769989, 0.517123281955719, 0.47488585114479065, 0.6826484203338623, 0.5844748616218567, 0.6255707740783691, 0.47488585114479065, 0.568493127822876, 0.6963470578193665, 0.7214611768722534, 0.7123287916183472, 0.7100456357002258, 0.72374427318573, 0.732876718044281, 0.5616438388824463, 0.6575342416763306, 0.6050228476524353, 0.6952054500579834, 0.7077625393867493, 0.5616438388824463, 0.7831050157546997, 0.6164383292198181, 0.7454338073730469, 0.7283105254173279, 0.7351598143577576, 0.7831050157546997, 0.7031963467597961, 0.7831050157546997, 0.7442922592163086, 0.7785388231277466, 0.77625572681427, 0.7420091032981873, 0.72374427318573, 0.7648401856422424, 0.7945205569267273, 0.7945205569267273, 0.7785388231277466, 0.7031963467597961, 0.7123287916183472, 0.8287671208381653, 0.6974886059761047, 0.7990867495536804, 0.8082191944122314, 0.8972602486610413, 0.844748854637146, 0.7876712083816528, 0.9246575236320496, 0.7602739930152893, 0.836758017539978, 0.8858447670936584, 0.8926940560340881, 0.9178082346916199, 0.8641552329063416, 0.9086757898330688, 0.9029680490493774, 0.9200913310050964, 0.8926940560340881, 0.913241982460022, 0.9292237162590027, 0.9360730648040771, 0.9360730648040771, 0.9452054500579834, 0.9178082346916199, 0.9954338073730469], [0.12328767031431198, 0.21004566550254822, 0.1837899535894394, 0.2557077705860138, 0.2785388231277466, 0.3378995358943939, 0.3287671208381653, 0.3675799071788788, 0.4269406497478485, 0.40296804904937744, NaN, 0.4885844886302948, 0.5045661926269531, 0.46347030997276306, 0.5011415481567383, 0.4794520437717438, 0.4726027250289917, 0.6187214851379395, 0.5502282977104187, 0.5799086689949036, 0.49200913310050964, 0.5490867495536804, 0.7305936217308044, 0.6860730648040771, 0.6757990717887878, 0.6529680490493774, 0.6803653240203857, 0.6826484203338623, 0.5365296602249146, 0.6415525078773499, 0.5936073064804077, 0.6872146129608154, 0.6621004343032837, 0.5593607425689697, 0.732876718044281, 0.6255707740783691, 0.7054794430732727, 0.7351598143577576, 0.7477169036865234, 0.7283105254173279, 0.6757990717887878, 0.7739726305007935, 0.7420091032981873, 0.740867555141449, 0.8253424763679504, 0.7557077407836914, 0.6849315166473389, 0.7568492889404297, 0.7420091032981873, 0.7579908967018127, 0.7728310227394104, 0.6940639019012451, 0.7054794430732727, 0.7876712083816528, 0.7100456357002258, 0.8378995656967163, 0.827625572681427, 0.8721461296081543, 0.8561643958091736, 0.7648401856422424, 0.9018265008926392, 0.7488584518432617, 0.8264840245246887, 0.8926940560340881, 0.9109588861465454, 0.913241982460022, 0.8310502171516418, 0.9109588861465454, 0.9292237162590027, 0.9246575236320496, 0.8515982031822205, 0.9041095972061157, 0.9246575236320496, 0.9383561611175537, 0.9611872434616089, 0.965753436088562, 0.9246575236320496, 0.9931507110595703], [0.077625572681427, 0.1883561611175537, 0.21803653240203857, 0.19748859107494354, 0.28082191944122314, 0.33561643958091736, 0.3321917951107025, 0.3652968108654022, 0.4497717022895813, 0.5, 0.5114155411720276, NaN, 0.5114155411720276, 0.5045661926269531, 0.5559360980987549, 0.5593607425689697, 0.5239726305007935, 0.5730593800544739, 0.5547945499420166, 0.5821917653083801, 0.5662100315093994, 0.6061643958091736, 0.6073059439659119, 0.6255707740783691, 0.6495434045791626, 0.6438356041908264, 0.6278538703918457, 0.6347032189369202, 0.6187214851379395, 0.6598173379898071, 0.6472602486610413, 0.6803653240203857, 0.646118700504303, 0.621004581451416, 0.6917808055877686, 0.6735159754753113, 0.6963470578193665, 0.6735159754753113, 0.6963470578193665, 0.7123287916183472, 0.715753436088562, 0.7168949842453003, 0.6917808055877686, 0.7283105254173279, 0.7283105254173279, 0.689497709274292, 0.7351598143577576, 0.698630154132843, 0.7397260069847107, 0.7534246444702148, 0.7420091032981873, 0.7031963467597961, 0.7077625393867493, 0.7739726305007935, 0.7968036532402039, 0.8333333134651184, 0.827625572681427, 0.8515982031822205, 0.810502290725708, 0.8127853870391846, 0.8984017968177795, 0.7739726305007935, 0.7808219194412231, 0.8698630332946777, 0.8675799369812012, 0.8904109597206116, 0.7990867495536804, 0.8641552329063416, 0.8972602486610413, 0.9086757898330688, 0.8755707740783691, 0.8835616707801819, 0.922374427318573, 0.9520547986030579, 0.948630154132843, 0.9611872434616089, 0.9383561611175537, 0.990867555141449], [0.09360730648040771, 0.20776255428791046, 0.2009132355451584, 0.24200913310050964, 0.2785388231277466, 0.3219178020954132, 0.3287671208381653, 0.35159817337989807, 0.45776256918907166, 0.4406392574310303, 0.4954337775707245, 0.4885844886302948, NaN, 0.46347030997276306, 0.5376712083816528, 0.5456621050834656, 0.43607306480407715, 0.611872136592865, 0.5913242101669312, 0.6746575236320496, 0.4885844886302948, 0.5639269351959229, 0.6347032189369202, 0.6655251383781433, 0.6267123222351074, 0.6666666865348816, 0.6529680490493774, 0.6575342416763306, 0.5536529421806335, 0.6609588861465454, 0.5890411138534546, 0.6050228476524353, 0.5821917653083801, 0.5456621050834656, 0.6963470578193665, 0.6301369667053223, 0.7146118879318237, 0.72374427318573, 0.6735159754753113, 0.689497709274292, 0.6940639019012451, 0.7557077407836914, 0.8299086689949036, 0.767123281955719, 0.7682648301124573, 0.767123281955719, 0.7077625393867493, 0.7568492889404297, 0.7853881120681763, 0.7442922592163086, 0.7351598143577576, 0.7054794430732727, 0.7100456357002258, 0.7865296602249146, 0.6940639019012451, 0.8173515796661377, 0.7990867495536804, 0.853881299495697, 0.8493150472640991, 0.767123281955719, 0.905251145362854, 0.7534246444702148, 0.8333333134651184, 0.8904109597206116, 0.8652967810630798, 0.888127863407135, 0.8481734991073608, 0.8721461296081543, 0.8858447670936584, 0.9109588861465454, 0.878995418548584, 0.8812785148620605, 0.8858447670936584, 0.922374427318573, 0.922374427318573, 0.965753436088562, 0.9086757898330688, 0.9817351698875427], [0.09474886208772659, 0.22602739930152893, 0.25342464447021484, 0.21575342118740082, 0.2945205569267273, 0.31506848335266113, 0.29680365324020386, 0.3378995358943939, 0.3767123222351074, 0.5182648301124573, 0.5365296602249146, 0.4954337775707245, 0.5365296602249146, NaN, 0.4942922294139862, 0.47488585114479065, 0.5399543642997742, 0.6027397513389587, 0.5182648301124573, 0.5490867495536804, 0.577625572681427, 0.5068492889404297, 0.594748854637146, 0.621004581451416, 0.6952054500579834, 0.6301369667053223, 0.6415525078773499, 0.6392694115638733, 0.6164383292198181, 0.6278538703918457, 0.5616438388824463, 0.6347032189369202, 0.6130136847496033, 0.6347032189369202, 0.6803653240203857, 0.6369863152503967, 0.6917808055877686, 0.6689497828483582, 0.638127863407135, 0.6826484203338623, 0.646118700504303, 0.7031963467597961, 0.6506849527359009, 0.7123287916183472, 0.6621004343032837, 0.6849315166473389, 0.6917808055877686, 0.6803653240203857, 0.7191780805587769, 0.7009132504463196, 0.7111872434616089, 0.638127863407135, 0.6369863152503967, 0.7397260069847107, 0.7271689772605896, 0.7579908967018127, 0.7557077407836914, 0.77625572681427, 0.7694063782691956, 0.7146118879318237, 0.7968036532402039, 0.7043378949165344, 0.7716894745826721, 0.7808219194412231, 0.8036529421806335, 0.8219178318977356, 0.7968036532402039, 0.8287671208381653, 0.8242009282112122, 0.844748854637146, 0.8036529421806335, 0.8082191944122314, 0.888127863407135, 0.878995418548584, 0.8972602486610413, 0.8949771523475647, 0.8515982031822205, 0.965753436088562], [0.10273972898721695, 0.18036529421806335, 0.22374428808689117, 0.18607306480407715, 0.2785388231277466, 0.3173516094684601, 0.3333333432674408, 0.34018266201019287, 0.3493150770664215, 0.456620991230011, 0.4988584518432617, 0.4440639317035675, 0.4623287618160248, 0.5057077407836914, NaN, 0.517123281955719, 0.5068492889404297, 0.5251141786575317, 0.4954337775707245, 0.52625572681427, 0.5410959124565125, 0.6392694115638733, 0.5901826620101929, 0.5753424763679504, 0.5981734991073608, 0.5821917653083801, 0.5936073064804077, 0.5913242101669312, 0.5639269351959229, 0.6027397513389587, 0.6484017968177795, 0.628995418548584, 0.6175799369812012, 0.6312785148620605, 0.6015982031822205, 0.6415525078773499, 0.6232876777648926, 0.6221461296081543, 0.664383590221405, 0.681506872177124, 0.663241982460022, 0.6529680490493774, 0.628995418548584, 0.6575342416763306, 0.6621004343032837, 0.6518265008926392, 0.6849315166473389, 0.646118700504303, 0.689497709274292, 0.72374427318573, 0.7351598143577576, 0.655251145362854, 0.6575342416763306, 0.7374429106712341, 0.7831050157546997, 0.7945205569267273, 0.801369845867157, 0.7808219194412231, 0.801369845867157, 0.7968036532402039, 0.8310502171516418, 0.7602739930152893, 0.7511415481567383, 0.8401826620101929, 0.8561643958091736, 0.8584474921226501, 0.7716894745826721, 0.8721461296081543, 0.8561643958091736, 0.8732876777648926, 0.8378995656967163, 0.8664383292198181, 0.9166666865348816, 0.9269406199455261, 0.9246575236320496, 0.9360730648040771, 0.9086757898330688, 0.9931507110595703], [0.13242009282112122, 0.16894976794719696, 0.232876718044281, 0.19178082048892975, 0.28538814187049866, 0.24657534062862396, 0.2990867495536804, 0.31392693519592285, 0.3116438388824463, 0.482876718044281, 0.5205479264259338, 0.4406392574310303, 0.4543378949165344, 0.5251141786575317, 0.482876718044281, NaN, 0.49086758494377136, 0.5228310227394104, 0.4954337775707245, 0.509132444858551, 0.5456621050834656, 0.543379008769989, 0.5958904027938843, 0.5502282977104187, 0.5981734991073608, 0.5639269351959229, 0.5753424763679504, 0.5901826620101929, 0.5981734991073608, 0.621004581451416, 0.6141552329063416, 0.6506849527359009, 0.6164383292198181, 0.611872136592865, 0.611872136592865, 0.6232876777648926, 0.6278538703918457, 0.6324200630187988, 0.7054794430732727, 0.6598173379898071, 0.655251145362854, 0.6484017968177795, 0.6392694115638733, 0.6666666865348816, 0.6575342416763306, 0.655251145362854, 0.6598173379898071, 0.655251145362854, 0.7031963467597961, 0.7283105254173279, 0.7442922592163086, 0.7077625393867493, 0.7191780805587769, 0.7191780805587769, 0.844748854637146, 0.801369845867157, 0.7968036532402039, 0.7602739930152893, 0.7785388231277466, 0.888127863407135, 0.7888127565383911, 0.7876712083816528, 0.7899543642997742, 0.8607305884361267, 0.8059360980987549, 0.8652967810630798, 0.8150684833526611, 0.8470319509506226, 0.8242009282112122, 0.8652967810630798, 0.8573059439659119, 0.8515982031822205, 0.8904109597206116, 0.8858447670936584, 0.8949771523475647, 0.9246575236320496, 0.9383561611175537, 0.9931507110595703], [0.1438356190919876, 0.21917808055877686, 0.3093607425689697, 0.28652969002723694, 0.3173516094684601, 0.2990867495536804, 0.29223743081092834, 0.30365297198295593, 0.42465752363204956, 0.5251141786575317, 0.5273972749710083, 0.47602739930152893, 0.5639269351959229, 0.4600456655025482, 0.4931506812572479, 0.509132444858551, NaN, 0.5570776462554932, 0.5479452013969421, 0.6369863152503967, 0.5490867495536804, 0.5045661926269531, 0.6141552329063416, 0.6050228476524353, 0.6152967810630798, 0.6415525078773499, 0.6164383292198181, 0.6084474921226501, 0.6027397513389587, 0.6164383292198181, 0.586758017539978, 0.6095890402793884, 0.6152967810630798, 0.6027397513389587, 0.6495434045791626, 0.5958904027938843, 0.689497709274292, 0.6712328791618347, 0.6506849527359009, 0.6757990717887878, 0.6232876777648926, 0.655251145362854, 0.6575342416763306, 0.6940639019012451, 0.6529680490493774, 0.6849315166473389, 0.6666666865348816, 0.6849315166473389, 0.7077625393867493, 0.689497709274292, 0.6963470578193665, 0.6187214851379395, 0.621004581451416, 0.7442922592163086, 0.6929223537445068, 0.698630154132843, 0.706620991230011, 0.7625570893287659, 0.7534246444702148, 0.7739726305007935, 0.7488584518432617, 0.6803653240203857, 0.7648401856422424, 0.7808219194412231, 0.7351598143577576, 0.7831050157546997, 0.7990867495536804, 0.801369845867157, 0.7636986374855042, 0.8059360980987549, 0.8333333134651184, 0.7716894745826721, 0.9041095972061157, 0.8401826620101929, 0.8515982031822205, 0.9200913310050964, 0.8664383292198181, 0.9885844588279724], [0.11187214404344559, 0.21461187303066254, 0.16894976794719696, 0.23515981435775757, 0.2442922443151474, 0.2785388231277466, 0.27397260069847107, 0.2945205569267273, 0.4109589159488678, 0.3173516094684601, 0.38127854466438293, 0.4269406497478485, 0.388127863407135, 0.39726027846336365, 0.47488585114479065, 0.4771689474582672, 0.4429223835468292, NaN, 0.4452054798603058, 0.4714611768722534, 0.45776256918907166, 0.5205479264259338, 0.5114155411720276, 0.646118700504303, 0.5490867495536804, 0.6244292259216309, 0.621004581451416, 0.6244292259216309, 0.4885844886302948, 0.5525113940238953, 0.5936073064804077, 0.5913242101669312, 0.6095890402793884, 0.5068492889404297, 0.7351598143577576, 0.5890411138534546, 0.6061643958091736, 0.6541095972061157, 0.6004565954208374, 0.6073059439659119, 0.6141552329063416, 0.7853881120681763, 0.6952054500579834, 0.6621004343032837, 0.6621004343032837, 0.6963470578193665, 0.621004581451416, 0.7534246444702148, 0.7465753555297852, 0.6940639019012451, 0.6872146129608154, 0.681506872177124, 0.6849315166473389, 0.7260273694992065, 0.6757990717887878, 0.8070776462554932, 0.7899543642997742, 0.8698630332946777, 0.8493150472640991, 0.7077625393867493, 0.9497717022895813, 0.7602739930152893, 0.8173515796661377, 0.810502290725708, 0.8995434045791626, 0.8744292259216309, 0.8310502171516418, 0.7853881120681763, 0.922374427318573, 0.8972602486610413, 0.8812785148620605, 0.8995434045791626, 0.8470319509506226, 0.922374427318573, 0.9520547986030579, 0.9452054500579834, 0.9212328791618347, 0.990867555141449], [0.1335616409778595, 0.2271689474582672, 0.24200913310050964, 0.22602739930152893, 0.2785388231277466, 0.2705479562282562, 0.27625569701194763, 0.29680365324020386, 0.4269406497478485, 0.4155251085758209, 0.4497717022895813, 0.4452054798603058, 0.40867578983306885, 0.4817351698875427, 0.5045661926269531, 0.5045661926269531, 0.45205479860305786, 0.5547945499420166, NaN, 0.4954337775707245, 0.48059362173080444, 0.5593607425689697, 0.5205479264259338, 0.5616438388824463, 0.5479452013969421, 0.6050228476524353, 0.5730593800544739, 0.5970319509506226, 0.543379008769989, 0.5833333134651184, 0.5981734991073608, 0.5844748616218567, 0.5639269351959229, 0.5205479264259338, 0.6347032189369202, 0.6027397513389587, 0.628995418548584, 0.655251145362854, 0.6130136847496033, 0.6278538703918457, 0.6392694115638733, 0.655251145362854, 0.6803653240203857, 0.655251145362854, 0.655251145362854, 0.6780821681022644, 0.5958904027938843, 0.7305936217308044, 0.689497709274292, 0.6438356041908264, 0.7031963467597961, 0.7260273694992065, 0.732876718044281, 0.7123287916183472, 0.6940639019012451, 0.7922374606132507, 0.7922374606132507, 0.7648401856422424, 0.8127853870391846, 0.7614155411720276, 0.818493127822876, 0.7888127565383911, 0.8310502171516418, 0.810502290725708, 0.8424657583236694, 0.8287671208381653, 0.8561643958091736, 0.7922374606132507, 0.8424657583236694, 0.844748854637146, 0.9109588861465454, 0.9109588861465454, 0.8150684833526611, 0.8630136847496033, 0.8926940560340881, 0.9200913310050964, 0.9497717022895813, 0.9840182662010193], [0.06164383515715599, 0.16210044920444489, 0.18150684237480164, 0.2054794579744339, 0.2214611917734146, 0.267123281955719, 0.2716895043849945, 0.2888127863407135, 0.39269405603408813, 0.37442922592163086, 0.42009133100509644, 0.4178082048892975, 0.32534247636795044, 0.4509132504463196, 0.47374430298805237, 0.49086758494377136, 0.36301368474960327, 0.5285388231277466, 0.5045661926269531, NaN, 0.43721461296081543, 0.5114155411720276, 0.534246563911438, 0.5707762837409973, 0.5525113940238953, 0.6004565954208374, 0.6050228476524353, 0.6095890402793884, 0.5251141786575317, 0.6278538703918457, 0.551369845867157, 0.5319634675979614, 0.5365296602249146, 0.4840182662010193, 0.6267123222351074, 0.5547945499420166, 0.6666666865348816, 0.611872136592865, 0.6244292259216309, 0.646118700504303, 0.6506849527359009, 0.655251145362854, 0.7100456357002258, 0.6826484203338623, 0.7146118879318237, 0.6803653240203857, 0.6529680490493774, 0.6963470578193665, 0.715753436088562, 0.6712328791618347, 0.7031963467597961, 0.6872146129608154, 0.6906392574310303, 0.7340182662010193, 0.6666666865348816, 0.7511415481567383, 0.7579908967018127, 0.8082191944122314, 0.8150684833526611, 0.7511415481567383, 0.835616409778595, 0.732876718044281, 0.801369845867157, 0.8744292259216309, 0.8333333134651184, 0.8835616707801819, 0.8173515796661377, 0.8242009282112122, 0.8310502171516418, 0.8972602486610413, 0.8493150472640991, 0.853881299495697, 0.8424657583236694, 0.8835616707801819, 0.8926940560340881, 0.956620991230011, 0.8835616707801819, 0.9726027250289917], [0.10273972898721695, 0.20319634675979614, 0.2511415481567383, 0.21004566550254822, 0.28082191944122314, 0.26484018564224243, 0.24657534062862396, 0.2819634675979614, 0.38584473729133606, 0.5251141786575317, 0.5079908967018127, 0.4337899684906006, 0.5114155411720276, 0.422374427318573, 0.45890411734580994, 0.4543378949165344, 0.4509132504463196, 0.5422374606132507, 0.5194063782691956, 0.5627853870391846, NaN, 0.46917808055877686, 0.5833333134651184, 0.5616438388824463, 0.6004565954208374, 0.5936073064804077, 0.5844748616218567, 0.5833333134651184, 0.5936073064804077, 0.6141552329063416, 0.5365296602249146, 0.6095890402793884, 0.6130136847496033, 0.6152967810630798, 0.6244292259216309, 0.5308219194412231, 0.6506849527359009, 0.6335616707801819, 0.6061643958091736, 0.6278538703918457, 0.5993150472640991, 0.6392694115638733, 0.611872136592865, 0.6621004343032837, 0.6347032189369202, 0.6506849527359009, 0.6803653240203857, 0.6484017968177795, 0.6780821681022644, 0.655251145362854, 0.6404109597206116, 0.5958904027938843, 0.6004565954208374, 0.7054794430732727, 0.664383590221405, 0.6883561611175537, 0.6940639019012451, 0.7511415481567383, 0.7168949842453003, 0.732876718044281, 0.7557077407836914, 0.6757990717887878, 0.7351598143577576, 0.7557077407836914, 0.7488584518432617, 0.77625572681427, 0.7625570893287659, 0.7945205569267273, 0.7694063782691956, 0.8082191944122314, 0.7876712083816528, 0.7785388231277466, 0.9018265008926392, 0.8698630332946777, 0.8687214851379395, 0.9086757898330688, 0.853881299495697, 0.9931507110595703], [0.10730593651533127, 0.15068493783473969, 0.2054794579744339, 0.18721461296081543, 0.26940637826919556, 0.284246563911438, 0.2831050157546997, 0.2956621050834656, 0.29223743081092834, 0.43150684237480164, 0.4509132504463196, 0.3938356041908264, 0.43607306480407715, 0.4931506812572479, 0.3607305884361267, 0.456620991230011, 0.4954337775707245, 0.4794520437717438, 0.4406392574310303, 0.4885844886302948, 0.5308219194412231, NaN, 0.5136986374855042, 0.5136986374855042, 0.5365296602249146, 0.5479452013969421, 0.5547945499420166, 0.551369845867157, 0.5376712083816528, 0.5388127565383911, 0.6004565954208374, 0.586758017539978, 0.5502282977104187, 0.5993150472640991, 0.5673515796661377, 0.5707762837409973, 0.5844748616218567, 0.5958904027938843, 0.6141552329063416, 0.6347032189369202, 0.611872136592865, 0.5890411138534546, 0.577625572681427, 0.6050228476524353, 0.6255707740783691, 0.6278538703918457, 0.6267123222351074, 0.6278538703918457, 0.6232876777648926, 0.655251145362854, 0.6940639019012451, 0.603881299495697, 0.6050228476524353, 0.6963470578193665, 0.7785388231277466, 0.7397260069847107, 0.7602739930152893, 0.7123287916183472, 0.7351598143577576, 0.7785388231277466, 0.7511415481567383, 0.7374429106712341, 0.740867555141449, 0.810502290725708, 0.7853881120681763, 0.8378995656967163, 0.7716894745826721, 0.844748854637146, 0.8196346759796143, 0.8470319509506226, 0.8196346759796143, 0.8561643958091736, 0.8812785148620605, 0.8835616707801819, 0.8972602486610413, 0.9337899684906006, 0.8972602486610413, 0.9885844588279724], [0.1278538852930069, 0.18949772417545319, 0.15296803414821625, 0.2283104956150055, 0.2168949842453003, 0.2557077705860138, 0.27397260069847107, 0.29223743081092834, 0.3321917951107025, 0.30365297198295593, 0.26940637826919556, 0.39269405603408813, 0.3652968108654022, 0.405251145362854, 0.40981733798980713, 0.4041095972061157, 0.38584473729133606, 0.4885844886302948, 0.4794520437717438, 0.465753436088562, 0.4166666567325592, 0.48630136251449585, NaN, 0.5388127565383911, 0.4942922294139862, 0.543379008769989, 0.5136986374855042, 0.5228310227394104, 0.49086758494377136, 0.5228310227394104, 0.5228310227394104, 0.5388127565383911, 0.5559360980987549, 0.4817351698875427, 0.5936073064804077, 0.5479452013969421, 0.5570776462554932, 0.6141552329063416, 0.6324200630187988, 0.6324200630187988, 0.5981734991073608, 0.6495434045791626, 0.6392694115638733, 0.6027397513389587, 0.6940639019012451, 0.6598173379898071, 0.5821917653083801, 0.6278538703918457, 0.6506849527359009, 0.6438356041908264, 0.7100456357002258, 0.6712328791618347, 0.6826484203338623, 0.72374427318573, 0.6335616707801819, 0.7420091032981873, 0.7534246444702148, 0.77625572681427, 0.7922374606132507, 0.740867555141449, 0.810502290725708, 0.7351598143577576, 0.7557077407836914, 0.844748854637146, 0.8219178318977356, 0.8470319509506226, 0.7659817337989807, 0.8755707740783691, 0.8378995656967163, 0.8515982031822205, 0.8150684833526611, 0.844748854637146, 0.8675799369812012, 0.8493150472640991, 0.8812785148620605, 0.9474886059761047, 0.8721461296081543, 0.9794520735740662], [0.0890410989522934, 0.15753424167633057, 0.15981735289096832, 0.19634702801704407, 0.18721461296081543, 0.23515981435775757, 0.2340182662010193, 0.2397260218858719, 0.3767123222351074, 0.2785388231277466, 0.31392693519592285, 0.37442922592163086, 0.3344748914241791, 0.37899544835090637, 0.42465752363204956, 0.4497717022895813, 0.3949771821498871, 0.35388126969337463, 0.4383561611175537, 0.4292237460613251, 0.4383561611175537, 0.48630136251449585, 0.4611872136592865, NaN, 0.4817351698875427, 0.560502290725708, 0.5582191944122314, 0.5479452013969421, 0.4771689474582672, 0.4885844886302948, 0.5410959124565125, 0.5559360980987549, 0.5502282977104187, 0.49086758494377136, 0.6244292259216309, 0.5410959124565125, 0.5502282977104187, 0.621004581451416, 0.5696346759796143, 0.5570776462554932, 0.586758017539978, 0.6849315166473389, 0.6506849527359009, 0.5844748616218567, 0.6404109597206116, 0.6484017968177795, 0.5844748616218567, 0.6872146129608154, 0.689497709274292, 0.6324200630187988, 0.621004581451416, 0.664383590221405, 0.6689497828483582, 0.6655251383781433, 0.6575342416763306, 0.7568492889404297, 0.7659817337989807, 0.818493127822876, 0.8207762837409973, 0.6917808055877686, 0.914383590221405, 0.7420091032981873, 0.7831050157546997, 0.7922374606132507, 0.8401826620101929, 0.8504565954208374, 0.8070776462554932, 0.7648401856422424, 0.8744292259216309, 0.8721461296081543, 0.8584474921226501, 0.8630136847496033, 0.8253424763679504, 0.9063926935195923, 0.9246575236320496, 0.9429223537445068, 0.9086757898330688, 0.9931507110595703], [0.09589041024446487, 0.15068493783473969, 0.12557077407836914, 0.18493150174617767, 0.2054794579744339, 0.27625569701194763, 0.2659817337989807, 0.2945205569267273, 0.34703195095062256, 0.2876712381839752, 0.32420089840888977, 0.3504566252231598, 0.3732876777648926, 0.3047945201396942, 0.4018264710903168, 0.4018264710903168, 0.3847031891345978, 0.4509132504463196, 0.45205479860305786, 0.44748857617378235, 0.3995433747768402, 0.46347030997276306, 0.5057077407836914, 0.5182648301124573, NaN, 0.49771690368652344, 0.4771689474582672, 0.5068492889404297, 0.49771690368652344, 0.534246563911438, 0.5205479264259338, 0.5742009282112122, 0.5993150472640991, 0.5182648301124573, 0.5650684833526611, 0.5502282977104187, 0.5753424763679504, 0.5981734991073608, 0.6301369667053223, 0.6141552329063416, 0.6027397513389587, 0.6347032189369202, 0.5821917653083801, 0.5913242101669312, 0.664383590221405, 0.6324200630187988, 0.6050228476524353, 0.6392694115638733, 0.6358447670936584, 0.7111872434616089, 0.6883561611175537, 0.6347032189369202, 0.638127863407135, 0.7054794430732727, 0.6267123222351074, 0.7294520735740662, 0.7465753555297852, 0.810502290725708, 0.7899543642997742, 0.7351598143577576, 0.8584474921226501, 0.7191780805587769, 0.7191780805587769, 0.8173515796661377, 0.8401826620101929, 0.835616409778595, 0.7442922592163086, 0.8150684833526611, 0.8607305884361267, 0.8561643958091736, 0.7808219194412231, 0.8424657583236694, 0.9109588861465454, 0.878995418548584, 0.9041095972061157, 0.9086757898330688, 0.8801369667053223, 0.9748858213424683], [0.10502283275127411, 0.16894976794719696, 0.15639269351959229, 0.20319634675979614, 0.22602739930152893, 0.23744292557239532, 0.2442922443151474, 0.2716895043849945, 0.35388126969337463, 0.2899543344974518, 0.34703195095062256, 0.3561643958091736, 0.3333333432674408, 0.36986300349235535, 0.4178082048892975, 0.43607306480407715, 0.35844749212265015, 0.37557077407836914, 0.3949771821498871, 0.3995433747768402, 0.4063926935195923, 0.45205479860305786, 0.456620991230011, 0.439497709274292, 0.5022830963134766, NaN, 0.4931506812572479, 0.4942922294139862, 0.4668949842453003, 0.5068492889404297, 0.5136986374855042, 0.5376712083816528, 0.5547945499420166, 0.45205479860305786, 0.5696346759796143, 0.5662100315093994, 0.5650684833526611, 0.5890411138534546, 0.5639269351959229, 0.5936073064804077, 0.5456621050834656, 0.6130136847496033, 0.6187214851379395, 0.6255707740783691, 0.621004581451416, 0.6232876777648926, 0.586758017539978, 0.655251145362854, 0.6712328791618347, 0.664383590221405, 0.6484017968177795, 0.672374427318573, 0.6780821681022644, 0.7009132504463196, 0.6004565954208374, 0.7716894745826721, 0.7648401856422424, 0.8173515796661377, 0.8310502171516418, 0.6575342416763306, 0.8515982031822205, 0.7283105254173279, 0.7557077407836914, 0.7990867495536804, 0.8767123222351074, 0.7853881120681763, 0.77625572681427, 0.7876712083816528, 0.8812785148620605, 0.8242009282112122, 0.8242009282112122, 0.8858447670936584, 0.8310502171516418, 0.8744292259216309, 0.8995434045791626, 0.8926940560340881, 0.913241982460022, 0.9863013625144958], [0.10730593651533127, 0.17579908668994904, 0.14611871540546417, 0.21461187303066254, 0.21917808055877686, 0.26484018564224243, 0.25799086689949036, 0.2716895043849945, 0.33105021715164185, 0.27625569701194763, 0.31963470578193665, 0.3721461296081543, 0.34703195095062256, 0.35844749212265015, 0.4063926935195923, 0.42465752363204956, 0.3835616409778595, 0.37899544835090637, 0.4269406497478485, 0.3949771821498871, 0.4155251085758209, 0.4452054798603058, 0.48630136251449585, 0.44178083539009094, 0.5228310227394104, 0.5068492889404297, NaN, 0.5365296602249146, 0.45890411734580994, 0.5102739930152893, 0.5159817337989807, 0.5456621050834656, 0.5479452013969421, 0.456620991230011, 0.5559360980987549, 0.5844748616218567, 0.5593607425689697, 0.5616438388824463, 0.5639269351959229, 0.586758017539978, 0.5410959124565125, 0.6084474921226501, 0.6050228476524353, 0.6095890402793884, 0.5958904027938843, 0.5924657583236694, 0.5821917653083801, 0.621004581451416, 0.6506849527359009, 0.6598173379898071, 0.6301369667053223, 0.6598173379898071, 0.655251145362854, 0.6917808055877686, 0.5958904027938843, 0.7694063782691956, 0.7636986374855042, 0.8219178318977356, 0.7865296602249146, 0.6415525078773499, 0.8630136847496033, 0.7168949842453003, 0.7511415481567383, 0.7831050157546997, 0.8698630332946777, 0.7945205569267273, 0.7716894745826721, 0.7808219194412231, 0.8892694115638733, 0.8196346759796143, 0.8059360980987549, 0.8858447670936584, 0.8287671208381653, 0.8904109597206116, 0.9269406199455261, 0.8858447670936584, 0.8949771523475647, 0.990867555141449], [0.09817351400852203, 0.1666666716337204, 0.15639269351959229, 0.2054794579744339, 0.232876718044281, 0.26484018564224243, 0.25342464447021484, 0.27625569701194763, 0.33561643958091736, 0.267123281955719, 0.3173516094684601, 0.3652968108654022, 0.34246575832366943, 0.3607305884361267, 0.40867578983306885, 0.40981733798980713, 0.39155250787734985, 0.37557077407836914, 0.40296804904937744, 0.3904109597206116, 0.4166666567325592, 0.44863012433052063, 0.4771689474582672, 0.45205479860305786, 0.4931506812572479, 0.5057077407836914, 0.46347030997276306, NaN, 0.4611872136592865, 0.48630136251449585, 0.5068492889404297, 0.5422374606132507, 0.534246563911438, 0.4429223835468292, 0.5753424763679504, 0.5707762837409973, 0.5353881120681763, 0.5707762837409973, 0.534246563911438, 0.5799086689949036, 0.5273972749710083, 0.6187214851379395, 0.594748854637146, 0.5901826620101929, 0.5936073064804077, 0.5936073064804077, 0.5844748616218567, 0.6347032189369202, 0.6347032189369202, 0.6232876777648926, 0.6312785148620605, 0.6621004343032837, 0.6666666865348816, 0.681506872177124, 0.5913242101669312, 0.7465753555297852, 0.7374429106712341, 0.8333333134651184, 0.7694063782691956, 0.6324200630187988, 0.8401826620101929, 0.72374427318573, 0.7465753555297852, 0.7648401856422424, 0.8698630332946777, 0.7808219194412231, 0.7614155411720276, 0.7648401856422424, 0.8698630332946777, 0.810502290725708, 0.7968036532402039, 0.8812785148620605, 0.8219178318977356, 0.8812785148620605, 0.9178082346916199, 0.8949771523475647, 0.9041095972061157, 0.990867555141449], [0.16894976794719696, 0.22031962871551514, 0.25799086689949036, 0.23744292557239532, 0.284246563911438, 0.232876718044281, 0.2397260218858719, 0.26255708932876587, 0.35388126969337463, 0.4383561611175537, 0.46347030997276306, 0.38127854466438293, 0.44634702801704407, 0.3835616409778595, 0.43607306480407715, 0.4018264710903168, 0.39726027846336365, 0.5114155411720276, 0.456620991230011, 0.47488585114479065, 0.4063926935195923, 0.4623287618160248, 0.509132444858551, 0.5228310227394104, 0.5022830963134766, 0.5331050157546997, 0.5410959124565125, 0.5388127565383911, NaN, 0.5068492889404297, 0.5182648301124573, 0.543379008769989, 0.5456621050834656, 0.48059362173080444, 0.5479452013969421, 0.5045661926269531, 0.543379008769989, 0.5707762837409973, 0.5958904027938843, 0.5547945499420166, 0.5308219194412231, 0.5639269351959229, 0.5844748616218567, 0.5616438388824463, 0.5981734991073608, 0.5890411138534546, 0.5639269351959229, 0.6004565954208374, 0.5525113940238953, 0.5707762837409973, 0.6164383292198181, 0.6141552329063416, 0.6107305884361267, 0.6141552329063416, 0.6392694115638733, 0.6232876777648926, 0.6392694115638733, 0.6415525078773499, 0.6826484203338623, 0.7739726305007935, 0.6666666865348816, 0.6917808055877686, 0.7454338073730469, 0.7168949842453003, 0.6780821681022644, 0.7260273694992065, 0.7602739930152893, 0.7442922592163086, 0.706620991230011, 0.7305936217308044, 0.7990867495536804, 0.7465753555297852, 0.844748854637146, 0.7945205569267273, 0.7922374606132507, 0.8698630332946777, 0.8333333134651184, 0.9771689772605896], [0.08675798773765564, 0.1392694115638733, 0.13013698160648346, 0.1666666716337204, 0.17579908668994904, 0.2442922443151474, 0.19863013923168182, 0.25342464447021484, 0.3493150770664215, 0.34246575832366943, 0.35844749212265015, 0.34018266201019287, 0.3390410840511322, 0.3721461296081543, 0.39726027846336365, 0.37899544835090637, 0.3835616409778595, 0.44748857617378235, 0.4166666567325592, 0.3721461296081543, 0.38584473729133606, 0.4611872136592865, 0.4771689474582672, 0.5114155411720276, 0.465753436088562, 0.4931506812572479, 0.4897260367870331, 0.5136986374855042, 0.4931506812572479, NaN, 0.49771690368652344, 0.5285388231277466, 0.5525113940238953, 0.47031962871551514, 0.5502282977104187, 0.5182648301124573, 0.5616438388824463, 0.5525113940238953, 0.5787671208381653, 0.568493127822876, 0.6232876777648926, 0.5673515796661377, 0.543379008769989, 0.5742009282112122, 0.6301369667053223, 0.5913242101669312, 0.5525113940238953, 0.6027397513389587, 0.6095890402793884, 0.611872136592865, 0.6735159754753113, 0.6369863152503967, 0.6438356041908264, 0.6963470578193665, 0.621004581451416, 0.6849315166473389, 0.6940639019012451, 0.7100456357002258, 0.7465753555297852, 0.7305936217308044, 0.7465753555297852, 0.7283105254173279, 0.7168949842453003, 0.7648401856422424, 0.7694063782691956, 0.7808219194412231, 0.7420091032981873, 0.7956621050834656, 0.7785388231277466, 0.8082191944122314, 0.7945205569267273, 0.810502290725708, 0.8721461296081543, 0.8493150472640991, 0.8470319509506226, 0.8835616707801819, 0.8561643958091736, 0.9771689772605896], [0.09817351400852203, 0.15296803414821625, 0.22374428808689117, 0.15981735289096832, 0.28082191944122314, 0.2602739632129669, 0.3264840245246887, 0.32420089840888977, 0.19063927233219147, 0.3949771821498871, 0.4063926935195923, 0.35273972153663635, 0.4109589159488678, 0.4383561611175537, 0.35159817337989807, 0.38584473729133606, 0.41324201226234436, 0.4063926935195923, 0.4018264710903168, 0.44863012433052063, 0.46347030997276306, 0.3995433747768402, 0.4771689474582672, 0.45890411734580994, 0.4794520437717438, 0.48630136251449585, 0.4840182662010193, 0.4931506812572479, 0.4817351698875427, 0.5022830963134766, NaN, 0.5273972749710083, 0.5136986374855042, 0.5205479264259338, 0.5136986374855042, 0.5296803712844849, 0.5707762837409973, 0.5182648301124573, 0.543379008769989, 0.5707762837409973, 0.577625572681427, 0.5410959124565125, 0.5662100315093994, 0.5753424763679504, 0.5353881120681763, 0.5559360980987549, 0.5479452013969421, 0.5936073064804077, 0.577625572681427, 0.5799086689949036, 0.6369863152503967, 0.5890411138534546, 0.5936073064804077, 0.6415525078773499, 0.7146118879318237, 0.7260273694992065, 0.732876718044281, 0.6712328791618347, 0.6963470578193665, 0.7876712083816528, 0.6712328791618347, 0.6757990717887878, 0.7283105254173279, 0.7853881120681763, 0.7374429106712341, 0.7694063782691956, 0.7442922592163086, 0.801369845867157, 0.7694063782691956, 0.7990867495536804, 0.7922374606132507, 0.8264840245246887, 0.8264840245246887, 0.835616409778595, 0.8424657583236694, 0.8926940560340881, 0.8721461296081543, 0.990867555141449], [0.09589041024446487, 0.15753424167633057, 0.13242009282112122, 0.1826484054327011, 0.19634702801704407, 0.2054794579744339, 0.2214611917734146, 0.24771688878536224, 0.3059360682964325, 0.3047945201396942, 0.31278538703918457, 0.31963470578193665, 0.3949771821498871, 0.3652968108654022, 0.37100455164909363, 0.3493150770664215, 0.3904109597206116, 0.40867578983306885, 0.4155251085758209, 0.4680365324020386, 0.3904109597206116, 0.41324201226234436, 0.4611872136592865, 0.4440639317035675, 0.42579910159111023, 0.4623287618160248, 0.4543378949165344, 0.45776256918907166, 0.456620991230011, 0.4714611768722534, 0.4726027250289917, NaN, 0.5787671208381653, 0.4840182662010193, 0.4988584518432617, 0.4954337775707245, 0.509132444858551, 0.5707762837409973, 0.5810502171516418, 0.5114155411720276, 0.5273972749710083, 0.5319634675979614, 0.5570776462554932, 0.5296803712844849, 0.5913242101669312, 0.6164383292198181, 0.5730593800544739, 0.6095890402793884, 0.5388127565383911, 0.5559360980987549, 0.6575342416763306, 0.5993150472640991, 0.5981734991073608, 0.6232876777648926, 0.646118700504303, 0.7146118879318237, 0.7351598143577576, 0.7009132504463196, 0.7100456357002258, 0.7488584518432617, 0.7374429106712341, 0.689497709274292, 0.7694063782691956, 0.7990867495536804, 0.8036529421806335, 0.7785388231277466, 0.7899543642997742, 0.8493150472640991, 0.8242009282112122, 0.8321917653083801, 0.8230593800544739, 0.871004581451416, 0.9337899684906006, 0.9200913310050964, 0.9337899684906006, 0.913241982460022, 0.8812785148620605, 0.9885844588279724], [0.1164383590221405, 0.18493150174617767, 0.17694063484668732, 0.20890410244464874, 0.2168949842453003, 0.21232876181602478, 0.23744292557239532, 0.2557077705860138, 0.2899543344974518, 0.29223743081092834, 0.3378995358943939, 0.35388126969337463, 0.4178082048892975, 0.38698631525039673, 0.3824200928211212, 0.3835616409778595, 0.3847031891345978, 0.3904109597206116, 0.43607306480407715, 0.46347030997276306, 0.38698631525039673, 0.4497717022895813, 0.4440639317035675, 0.4497717022895813, 0.4006849229335785, 0.4452054798603058, 0.45205479860305786, 0.465753436088562, 0.4543378949165344, 0.44748857617378235, 0.48630136251449585, 0.4212328791618347, NaN, 0.5045661926269531, 0.4680365324020386, 0.534246563911438, 0.517123281955719, 0.551369845867157, 0.5582191944122314, 0.5251141786575317, 0.5547945499420166, 0.5148401856422424, 0.5502282977104187, 0.5388127565383911, 0.5799086689949036, 0.6073059439659119, 0.509132444858551, 0.6027397513389587, 0.5365296602249146, 0.5753424763679504, 0.6666666865348816, 0.6232876777648926, 0.6278538703918457, 0.6324200630187988, 0.621004581451416, 0.7191780805587769, 0.7488584518432617, 0.6369863152503967, 0.7260273694992065, 0.7648401856422424, 0.7009132504463196, 0.7168949842453003, 0.7945205569267273, 0.8150684833526611, 0.7511415481567383, 0.7568492889404297, 0.8242009282112122, 0.8378995656967163, 0.7899543642997742, 0.7979452013969421, 0.8493150472640991, 0.8698630332946777, 0.8972602486610413, 0.8630136847496033, 0.8652967810630798, 0.8949771523475647, 0.8926940560340881, 0.9885844588279724], [0.12557077407836914, 0.1495433747768402, 0.21004566550254822, 0.1952054798603058, 0.1940639317035675, 0.20319634675979614, 0.18949772417545319, 0.21803653240203857, 0.34018266201019287, 0.4383561611175537, 0.4406392574310303, 0.37899544835090637, 0.4543378949165344, 0.3652968108654022, 0.36872145533561707, 0.388127863407135, 0.39726027846336365, 0.4931506812572479, 0.4794520437717438, 0.5159817337989807, 0.3847031891345978, 0.4006849229335785, 0.5182648301124573, 0.509132444858551, 0.4817351698875427, 0.5479452013969421, 0.543379008769989, 0.5570776462554932, 0.5194063782691956, 0.5296803712844849, 0.4794520437717438, 0.5159817337989807, 0.4954337775707245, NaN, 0.5673515796661377, 0.517123281955719, 0.5913242101669312, 0.5593607425689697, 0.5490867495536804, 0.5616438388824463, 0.5399543642997742, 0.5730593800544739, 0.577625572681427, 0.611872136592865, 0.5764840245246887, 0.5844748616218567, 0.586758017539978, 0.586758017539978, 0.6301369667053223, 0.5924657583236694, 0.5821917653083801, 0.5821917653083801, 0.5799086689949036, 0.6301369667053223, 0.586758017539978, 0.5981734991073608, 0.6392694115638733, 0.6438356041908264, 0.6872146129608154, 0.6963470578193665, 0.6735159754753113, 0.6598173379898071, 0.7009132504463196, 0.6803653240203857, 0.646118700504303, 0.7123287916183472, 0.72374427318573, 0.7191780805587769, 0.6860730648040771, 0.7180365324020386, 0.7488584518432617, 0.7260273694992065, 0.835616409778595, 0.7899543642997742, 0.7773972749710083, 0.8493150472640991, 0.8082191944122314, 0.9840182662010193], [0.06392694264650345, 0.15182648599147797, 0.12100456655025482, 0.18949772417545319, 0.21461187303066254, 0.2397260218858719, 0.23744292557239532, 0.2557077705860138, 0.31963470578193665, 0.2168949842453003, 0.267123281955719, 0.30821916460990906, 0.30365297198295593, 0.31963470578193665, 0.39840182662010193, 0.388127863407135, 0.3504566252231598, 0.26484018564224243, 0.3652968108654022, 0.3732876777648926, 0.37557077407836914, 0.4326483905315399, 0.4063926935195923, 0.37557077407836914, 0.43493151664733887, 0.43036529421806335, 0.4440639317035675, 0.42465752363204956, 0.45205479860305786, 0.4497717022895813, 0.48630136251449585, 0.5011415481567383, 0.5319634675979614, 0.4326483905315399, NaN, 0.509132444858551, 0.45890411734580994, 0.5045661926269531, 0.5182648301124573, 0.5468036532402039, 0.5273972749710083, 0.6347032189369202, 0.5821917653083801, 0.5182648301124573, 0.5639269351959229, 0.5753424763679504, 0.5228310227394104, 0.5936073064804077, 0.560502290725708, 0.5844748616218567, 0.5753424763679504, 0.6347032189369202, 0.6369863152503967, 0.6187214851379395, 0.594748854637146, 0.7146118879318237, 0.7351598143577576, 0.8652967810630798, 0.7305936217308044, 0.6472602486610413, 0.8926940560340881, 0.698630154132843, 0.7397260069847107, 0.7465753555297852, 0.8801369667053223, 0.7397260069847107, 0.7511415481567383, 0.7214611768722534, 0.8755707740783691, 0.7773972749710083, 0.7945205569267273, 0.8401826620101929, 0.8333333134651184, 0.8744292259216309, 0.9200913310050964, 0.8767123222351074, 0.8824200630187988, 0.9794520735740662], [0.10958904027938843, 0.17123287916183472, 0.19634702801704407, 0.19863013923168182, 0.1940639317035675, 0.19863013923168182, 0.2271689474582672, 0.2397260218858719, 0.3173516094684601, 0.3835616409778595, 0.37442922592163086, 0.3264840245246887, 0.36986300349235535, 0.36301368474960327, 0.35844749212265015, 0.3767123222351074, 0.4041095972061157, 0.4109589159488678, 0.39726027846336365, 0.4452054798603058, 0.46917808055877686, 0.4292237460613251, 0.45205479860305786, 0.45890411734580994, 0.4497717022895813, 0.4337899684906006, 0.4155251085758209, 0.4292237460613251, 0.4954337775707245, 0.4817351698875427, 0.47031962871551514, 0.5045661926269531, 0.465753436088562, 0.482876718044281, 0.49086758494377136, NaN, 0.4840182662010193, 0.47488585114479065, 0.5616438388824463, 0.5570776462554932, 0.5410959124565125, 0.5068492889404297, 0.5228310227394104, 0.5159817337989807, 0.5913242101669312, 0.5114155411720276, 0.5319634675979614, 0.5468036532402039, 0.5319634675979614, 0.5308219194412231, 0.6392694115638733, 0.5639269351959229, 0.5639269351959229, 0.6324200630187988, 0.6301369667053223, 0.655251145362854, 0.689497709274292, 0.6255707740783691, 0.6529680490493774, 0.7465753555297852, 0.6780821681022644, 0.698630154132843, 0.7420091032981873, 0.7374429106712341, 0.706620991230011, 0.77625572681427, 0.7534246444702148, 0.7602739930152893, 0.7203196287155151, 0.7888127565383911, 0.8082191944122314, 0.7968036532402039, 0.8310502171516418, 0.8242009282112122, 0.8310502171516418, 0.8310502171516418, 0.878995418548584, 0.9794520735740662], [0.08219178020954132, 0.12557077407836914, 0.11187214404344559, 0.17579908668994904, 0.15068493783473969, 0.2009132355451584, 0.21347032487392426, 0.24086758494377136, 0.31506848335266113, 0.2545662224292755, 0.2945205569267273, 0.30365297198295593, 0.28538814187049866, 0.30821916460990906, 0.3767123222351074, 0.3721461296081543, 0.310502290725708, 0.3938356041908264, 0.37100455164909363, 0.3333333432674408, 0.3493150770664215, 0.4155251085758209, 0.4429223835468292, 0.4497717022895813, 0.42465752363204956, 0.43493151664733887, 0.4406392574310303, 0.46461185812950134, 0.456620991230011, 0.4383561611175537, 0.4292237460613251, 0.49086758494377136, 0.482876718044281, 0.40867578983306885, 0.5410959124565125, 0.5159817337989807, NaN, 0.5547945499420166, 0.5, 0.5205479264259338, 0.4817351698875427, 0.6015982031822205, 0.568493127822876, 0.6187214851379395, 0.5639269351959229, 0.5913242101669312, 0.5456621050834656, 0.5913242101669312, 0.7191780805587769, 0.5936073064804077, 0.560502290725708, 0.6575342416763306, 0.655251145362854, 0.6164383292198181, 0.5913242101669312, 0.6735159754753113, 0.6849315166473389, 0.767123281955719, 0.7374429106712341, 0.664383590221405, 0.8002282977104187, 0.7100456357002258, 0.7146118879318237, 0.7305936217308044, 0.7990867495536804, 0.7557077407836914, 0.7283105254173279, 0.7009132504463196, 0.8082191944122314, 0.7808219194412231, 0.767123281955719, 0.810502290725708, 0.7968036532402039, 0.835616409778595, 0.844748854637146, 0.8584474921226501, 0.8630136847496033, 0.9714611768722534], [0.09132420271635056, 0.155251145362854, 0.15981735289096832, 0.18036529421806335, 0.19863013923168182, 0.2054794579744339, 0.2214611917734146, 0.2283104956150055, 0.3059360682964325, 0.2716895043849945, 0.26484018564224243, 0.3264840245246887, 0.27625569701194763, 0.33105021715164185, 0.3778538703918457, 0.3675799071788788, 0.3287671208381653, 0.3458904027938843, 0.344748854637146, 0.388127863407135, 0.3664383590221405, 0.4041095972061157, 0.38584473729133606, 0.37899544835090637, 0.4018264710903168, 0.4109589159488678, 0.4383561611175537, 0.4292237460613251, 0.4292237460613251, 0.44748857617378235, 0.4817351698875427, 0.4292237460613251, 0.44863012433052063, 0.4406392574310303, 0.4954337775707245, 0.5251141786575317, 0.4452054798603058, NaN, 0.4942922294139862, 0.4931506812572479, 0.509132444858551, 0.5136986374855042, 0.5799086689949036, 0.4931506812572479, 0.5593607425689697, 0.6666666865348816, 0.4794520437717438, 0.6917808055877686, 0.543379008769989, 0.5365296602249146, 0.5582191944122314, 0.6438356041908264, 0.6415525078773499, 0.6152967810630798, 0.621004581451416, 0.7226027250289917, 0.7168949842453003, 0.7180365324020386, 0.7397260069847107, 0.6963470578193665, 0.7420091032981873, 0.7283105254173279, 0.7694063782691956, 0.7031963467597961, 0.7808219194412231, 0.7557077407836914, 0.7831050157546997, 0.7579908967018127, 0.7945205569267273, 0.7751141786575317, 0.8196346759796143, 0.8698630332946777, 0.7694063782691956, 0.8515982031822205, 0.8561643958091736, 0.8835616707801819, 0.8984017968177795, 0.9771689772605896], [0.10273972898721695, 0.15867580473423004, 0.1438356190919876, 0.15981735289096832, 0.21004566550254822, 0.1940639317035675, 0.2226027399301529, 0.24200913310050964, 0.26255708932876587, 0.26484018564224243, 0.25228309631347656, 0.30365297198295593, 0.3264840245246887, 0.361872136592865, 0.33561643958091736, 0.2945205569267273, 0.3493150770664215, 0.3995433747768402, 0.38698631525039673, 0.37557077407836914, 0.3938356041908264, 0.38584473729133606, 0.3675799071788788, 0.43036529421806335, 0.36986300349235535, 0.43607306480407715, 0.43607306480407715, 0.465753436088562, 0.4041095972061157, 0.4212328791618347, 0.456620991230011, 0.41894978284835815, 0.44178083539009094, 0.4509132504463196, 0.4817351698875427, 0.4383561611175537, 0.5, 0.5057077407836914, NaN, 0.47031962871551514, 0.5273972749710083, 0.5205479264259338, 0.4954337775707245, 0.5251141786575317, 0.5696346759796143, 0.5570776462554932, 0.5, 0.5662100315093994, 0.5159817337989807, 0.5182648301124573, 0.7054794430732727, 0.6164383292198181, 0.6232876777648926, 0.5936073064804077, 0.5844748616218567, 0.6598173379898071, 0.6598173379898071, 0.6415525078773499, 0.6803653240203857, 0.7305936217308044, 0.6860730648040771, 0.7009132504463196, 0.7534246444702148, 0.8047945499420166, 0.7568492889404297, 0.7876712083816528, 0.77625572681427, 0.9360730648040771, 0.767123281955719, 0.8082191944122314, 0.8082191944122314, 0.8584474921226501, 0.8561643958091736, 0.8378995656967163, 0.853881299495697, 0.9383561611175537, 0.8767123222351074, 0.9748858213424683], [0.04566210135817528, 0.11187214404344559, 0.1004566177725792, 0.12100456655025482, 0.14611871540546417, 0.17351597547531128, 0.1780821979045868, 0.19178082048892975, 0.2716895043849945, 0.2168949842453003, 0.2716895043849945, 0.2876712381839752, 0.310502290725708, 0.3173516094684601, 0.31849315762519836, 0.34018266201019287, 0.32420089840888977, 0.39269405603408813, 0.3721461296081543, 0.35388126969337463, 0.3721461296081543, 0.3652968108654022, 0.3675799071788788, 0.4429223835468292, 0.38584473729133606, 0.4063926935195923, 0.41324201226234436, 0.42009133100509644, 0.4452054798603058, 0.43150684237480164, 0.4292237460613251, 0.4885844886302948, 0.47488585114479065, 0.4383561611175537, 0.45319634675979614, 0.4429223835468292, 0.4794520437717438, 0.5068492889404297, 0.5296803712844849, NaN, 0.5182648301124573, 0.49771690368652344, 0.4885844886302948, 0.5296803712844849, 0.5593607425689697, 0.5570776462554932, 0.5205479264259338, 0.5730593800544739, 0.5410959124565125, 0.5559360980987549, 0.6449771523475647, 0.6187214851379395, 0.6267123222351074, 0.7477169036865234, 0.5799086689949036, 0.689497709274292, 0.6872146129608154, 0.7077625393867493, 0.7431507110595703, 0.7009132504463196, 0.7031963467597961, 0.7214611768722534, 0.7294520735740662, 0.8652967810630798, 0.7511415481567383, 0.7922374606132507, 0.7602739930152893, 0.8025113940238953, 0.7808219194412231, 0.8264840245246887, 0.8127853870391846, 0.8264840245246887, 0.8242009282112122, 0.8036529421806335, 0.8173515796661377, 0.9063926935195923, 0.8744292259216309, 0.9726027250289917], [0.10273972898721695, 0.1495433747768402, 0.16438356041908264, 0.1495433747768402, 0.15981735289096832, 0.1940639317035675, 0.2054794579744339, 0.23059360682964325, 0.27397260069847107, 0.29680365324020386, 0.32420089840888977, 0.284246563911438, 0.3059360682964325, 0.35388126969337463, 0.33675798773765564, 0.344748854637146, 0.3767123222351074, 0.38584473729133606, 0.3607305884361267, 0.3493150770664215, 0.4006849229335785, 0.388127863407135, 0.4018264710903168, 0.41324201226234436, 0.39726027846336365, 0.4543378949165344, 0.45890411734580994, 0.4726027250289917, 0.46917808055877686, 0.3767123222351074, 0.422374427318573, 0.4726027250289917, 0.4452054798603058, 0.4600456655025482, 0.4726027250289917, 0.45890411734580994, 0.5182648301124573, 0.49086758494377136, 0.4726027250289917, 0.4817351698875427, NaN, 0.5159817337989807, 0.4771689474582672, 0.5456621050834656, 0.5068492889404297, 0.5319634675979614, 0.4452054798603058, 0.5547945499420166, 0.5593607425689697, 0.5148401856422424, 0.5547945499420166, 0.6187214851379395, 0.6278538703918457, 0.5730593800544739, 0.6438356041908264, 0.6506849527359009, 0.6689497828483582, 0.6164383292198181, 0.7100456357002258, 0.7146118879318237, 0.6689497828483582, 0.7397260069847107, 0.7214611768722534, 0.7922374606132507, 0.7260273694992065, 0.7488584518432617, 0.7454338073730469, 0.7625570893287659, 0.7602739930152893, 0.7808219194412231, 0.8173515796661377, 0.8515982031822205, 0.7831050157546997, 0.8321917653083801, 0.8287671208381653, 0.853881299495697, 0.8949771523475647, 0.9817351698875427], [0.054794520139694214, 0.1278538852930069, 0.10730593651533127, 0.17123287916183472, 0.16894976794719696, 0.2009132355451584, 0.20662100613117218, 0.232876718044281, 0.30136987566947937, 0.2168949842453003, 0.22602739930152893, 0.2831050157546997, 0.2442922443151474, 0.29680365324020386, 0.34703195095062256, 0.35159817337989807, 0.344748854637146, 0.21461187303066254, 0.344748854637146, 0.344748854637146, 0.3607305884361267, 0.4109589159488678, 0.3504566252231598, 0.31506848335266113, 0.3652968108654022, 0.38698631525039673, 0.39155250787734985, 0.38127854466438293, 0.43607306480407715, 0.4326483905315399, 0.45890411734580994, 0.4680365324020386, 0.48515981435775757, 0.4269406497478485, 0.3652968108654022, 0.4931506812572479, 0.39840182662010193, 0.48630136251449585, 0.4794520437717438, 0.5022830963134766, 0.4840182662010193, NaN, 0.5388127565383911, 0.465753436088562, 0.5273972749710083, 0.5456621050834656, 0.5114155411720276, 0.5502282977104187, 0.5045661926269531, 0.5547945499420166, 0.5753424763679504, 0.6187214851379395, 0.6244292259216309, 0.5890411138534546, 0.568493127822876, 0.7089040875434875, 0.7260273694992065, 0.7968036532402039, 0.7191780805587769, 0.6347032189369202, 0.8401826620101929, 0.6940639019012451, 0.72374427318573, 0.7191780805587769, 0.8515982031822205, 0.7054794430732727, 0.7488584518432617, 0.7077625393867493, 0.853881299495697, 0.7648401856422424, 0.7853881120681763, 0.8413242101669312, 0.8173515796661377, 0.8721461296081543, 0.8995434045791626, 0.8721461296081543, 0.878995418548584, 0.9748858213424683], [0.06392694264650345, 0.10958904027938843, 0.12100456655025482, 0.1392694115638733, 0.15410958230495453, 0.19178082048892975, 0.2009132355451584, 0.21461187303066254, 0.30136987566947937, 0.2557077705860138, 0.25799086689949036, 0.30821916460990906, 0.17009133100509644, 0.3493150770664215, 0.37100455164909363, 0.3607305884361267, 0.34246575832366943, 0.3047945201396942, 0.31963470578193665, 0.2899543344974518, 0.388127863407135, 0.422374427318573, 0.3607305884361267, 0.3493150770664215, 0.4178082048892975, 0.38127854466438293, 0.3949771821498871, 0.405251145362854, 0.4155251085758209, 0.456620991230011, 0.4337899684906006, 0.4429223835468292, 0.4497717022895813, 0.422374427318573, 0.4178082048892975, 0.4771689474582672, 0.43150684237480164, 0.42009133100509644, 0.5045661926269531, 0.5114155411720276, 0.5228310227394104, 0.4611872136592865, NaN, 0.46347030997276306, 0.5251141786575317, 0.4954337775707245, 0.5022830963134766, 0.49200913310050964, 0.5228310227394104, 0.5376712083816528, 0.5936073064804077, 0.6506849527359009, 0.6575342416763306, 0.5913242101669312, 0.6004565954208374, 0.6803653240203857, 0.6940639019012451, 0.6255707740783691, 0.6826484203338623, 0.6392694115638733, 0.664383590221405, 0.7420091032981873, 0.7168949842453003, 0.7420091032981873, 0.7785388231277466, 0.7363013625144958, 0.7317351698875427, 0.7283105254173279, 0.77625572681427, 0.7465753555297852, 0.7831050157546997, 0.8242009282112122, 0.7557077407836914, 0.835616409778595, 0.8515982031822205, 0.8630136847496033, 0.8904109597206116, 0.9680365324020386], [0.07077625393867493, 0.12100456655025482, 0.10273972898721695, 0.15753424167633057, 0.13698630034923553, 0.18493150174617767, 0.1940639317035675, 0.20776255428791046, 0.2785388231277466, 0.2214611917734146, 0.25913241505622864, 0.2716895043849945, 0.232876718044281, 0.2876712381839752, 0.34246575832366943, 0.3333333432674408, 0.3059360682964325, 0.3378995358943939, 0.344748854637146, 0.3173516094684601, 0.3378995358943939, 0.3949771821498871, 0.39726027846336365, 0.4155251085758209, 0.40867578983306885, 0.37442922592163086, 0.3904109597206116, 0.40981733798980713, 0.4383561611175537, 0.42579910159111023, 0.42465752363204956, 0.47031962871551514, 0.4611872136592865, 0.388127863407135, 0.4817351698875427, 0.4840182662010193, 0.38127854466438293, 0.5068492889404297, 0.47488585114479065, 0.47031962871551514, 0.4543378949165344, 0.534246563911438, 0.5365296602249146, NaN, 0.534246563911438, 0.5468036532402039, 0.5045661926269531, 0.5525113940238953, 0.611872136592865, 0.5376712083816528, 0.5376712083816528, 0.6415525078773499, 0.6392694115638733, 0.5753424763679504, 0.5799086689949036, 0.6415525078773499, 0.6621004343032837, 0.7317351698875427, 0.6917808055877686, 0.6369863152503967, 0.7420091032981873, 0.7031963467597961, 0.689497709274292, 0.7191780805587769, 0.7694063782691956, 0.7043378949165344, 0.6963470578193665, 0.6849315166473389, 0.7716894745826721, 0.7317351698875427, 0.7465753555297852, 0.7831050157546997, 0.7716894745826721, 0.8264840245246887, 0.8150684833526611, 0.8527397513389587, 0.8333333134651184, 0.9589040875434875], [0.08675798773765564, 0.11872146278619766, 0.14611871540546417, 0.13013698160648346, 0.1438356190919876, 0.17351597547531128, 0.18949772417545319, 0.20319634675979614, 0.24885845184326172, 0.22374428808689117, 0.17465753853321075, 0.2716895043849945, 0.23173515498638153, 0.3378995358943939, 0.3378995358943939, 0.34246575832366943, 0.34703195095062256, 0.3378995358943939, 0.344748854637146, 0.28538814187049866, 0.3652968108654022, 0.37442922592163086, 0.3059360682964325, 0.3595890402793884, 0.33561643958091736, 0.37899544835090637, 0.4041095972061157, 0.4063926935195923, 0.4018264710903168, 0.36986300349235535, 0.46461185812950134, 0.40867578983306885, 0.42009133100509644, 0.4235159754753113, 0.43607306480407715, 0.40867578983306885, 0.43607306480407715, 0.4406392574310303, 0.43036529421806335, 0.4406392574310303, 0.4931506812572479, 0.4726027250289917, 0.47488585114479065, 0.465753436088562, NaN, 0.5205479264259338, 0.4337899684906006, 0.5068492889404297, 0.4931506812572479, 0.5022830963134766, 0.577625572681427, 0.6278538703918457, 0.646118700504303, 0.5936073064804077, 0.586758017539978, 0.6575342416763306, 0.6757990717887878, 0.6598173379898071, 0.6917808055877686, 0.7180365324020386, 0.689497709274292, 0.7031963467597961, 0.7054794430732727, 0.7751141786575317, 0.7305936217308044, 0.7557077407836914, 0.7420091032981873, 0.8150684833526611, 0.7579908967018127, 0.7922374606132507, 0.7922374606132507, 0.844748854637146, 0.7831050157546997, 0.8310502171516418, 0.8333333134651184, 0.9086757898330688, 0.8721461296081543, 0.9748858213424683], [0.08219178020954132, 0.13698630034923553, 0.15296803414821625, 0.155251145362854, 0.1609589010477066, 0.16438356041908264, 0.19063927233219147, 0.21232876181602478, 0.27968037128448486, 0.25799086689949036, 0.2442922443151474, 0.310502290725708, 0.232876718044281, 0.31506848335266113, 0.3481735289096832, 0.344748854637146, 0.31506848335266113, 0.30365297198295593, 0.3219178020954132, 0.31963470578193665, 0.3493150770664215, 0.3721461296081543, 0.34018266201019287, 0.35159817337989807, 0.3675799071788788, 0.3767123222351074, 0.40753424167633057, 0.4063926935195923, 0.4109589159488678, 0.40867578983306885, 0.4440639317035675, 0.3835616409778595, 0.39269405603408813, 0.4155251085758209, 0.42465752363204956, 0.4885844886302948, 0.40867578983306885, 0.3333333432674408, 0.4429223835468292, 0.4429223835468292, 0.4680365324020386, 0.4543378949165344, 0.5045661926269531, 0.45319634675979614, 0.4794520437717438, NaN, 0.44863012433052063, 0.6152967810630798, 0.4954337775707245, 0.4954337775707245, 0.5376712083816528, 0.6232876777648926, 0.628995418548584, 0.5616438388824463, 0.5901826620101929, 0.6621004343032837, 0.7009132504463196, 0.6780821681022644, 0.6849315166473389, 0.6678082346916199, 0.7077625393867493, 0.7077625393867493, 0.7374429106712341, 0.6826484203338623, 0.7420091032981873, 0.732876718044281, 0.7557077407836914, 0.7146118879318237, 0.7694063782691956, 0.7511415481567383, 0.7853881120681763, 0.8515982031822205, 0.7534246444702148, 0.8424657583236694, 0.8401826620101929, 0.8675799369812012, 0.8995434045791626, 0.9771689772605896], [0.05936073139309883, 0.0890410989522934, 0.06621004641056061, 0.13013698160648346, 0.11872146278619766, 0.2168949842453003, 0.2009132355451584, 0.2111872136592865, 0.31963470578193665, 0.27625569701194763, 0.31506848335266113, 0.26484018564224243, 0.29223743081092834, 0.30821916460990906, 0.31506848335266113, 0.34018266201019287, 0.3333333432674408, 0.37899544835090637, 0.4041095972061157, 0.34703195095062256, 0.31963470578193665, 0.3732876777648926, 0.4178082048892975, 0.4155251085758209, 0.3949771821498871, 0.41324201226234436, 0.4178082048892975, 0.4155251085758209, 0.43607306480407715, 0.44748857617378235, 0.45205479860305786, 0.4269406497478485, 0.49086758494377136, 0.41324201226234436, 0.4771689474582672, 0.4680365324020386, 0.4543378949165344, 0.5205479264259338, 0.5, 0.4794520437717438, 0.5547945499420166, 0.4885844886302948, 0.49771690368652344, 0.4954337775707245, 0.5662100315093994, 0.551369845867157, NaN, 0.5388127565383911, 0.49771690368652344, 0.5205479264259338, 0.5593607425689697, 0.543379008769989, 0.5479452013969421, 0.5707762837409973, 0.5, 0.5890411138534546, 0.6095890402793884, 0.6689497828483582, 0.6278538703918457, 0.6004565954208374, 0.6872146129608154, 0.6141552329063416, 0.5981734991073608, 0.655251145362854, 0.6575342416763306, 0.6689497828483582, 0.6027397513389587, 0.6940639019012451, 0.6598173379898071, 0.6735159754753113, 0.655251145362854, 0.72374427318573, 0.8139269351959229, 0.7648401856422424, 0.767123281955719, 0.7739726305007935, 0.7579908967018127, 0.931506872177124], [0.0890410989522934, 0.13584475219249725, 0.1438356190919876, 0.1438356190919876, 0.17579908668994904, 0.18036529421806335, 0.18721461296081543, 0.2054794579744339, 0.26255708932876587, 0.23515981435775757, 0.24315068125724792, 0.30136987566947937, 0.24315068125724792, 0.31963470578193665, 0.35388126969337463, 0.344748854637146, 0.31506848335266113, 0.24657534062862396, 0.26940637826919556, 0.30365297198295593, 0.35159817337989807, 0.3721461296081543, 0.3721461296081543, 0.31278538703918457, 0.3607305884361267, 0.344748854637146, 0.37899544835090637, 0.3652968108654022, 0.3995433747768402, 0.39726027846336365, 0.4063926935195923, 0.3904109597206116, 0.39726027846336365, 0.41324201226234436, 0.4063926935195923, 0.45319634675979614, 0.40867578983306885, 0.30821916460990906, 0.4337899684906006, 0.4269406497478485, 0.4452054798603058, 0.4497717022895813, 0.5079908967018127, 0.44748857617378235, 0.4931506812572479, 0.3847031891345978, 0.4611872136592865, NaN, 0.45890411734580994, 0.4680365324020386, 0.5365296602249146, 0.6095890402793884, 0.6050228476524353, 0.5639269351959229, 0.5936073064804077, 0.6666666865348816, 0.6598173379898071, 0.6369863152503967, 0.7031963467597961, 0.6780821681022644, 0.664383590221405, 0.7054794430732727, 0.7442922592163086, 0.7077625393867493, 0.7511415481567383, 0.7146118879318237, 0.7659817337989807, 0.7214611768722534, 0.7716894745826721, 0.7283105254173279, 0.8173515796661377, 0.8630136847496033, 0.732876718044281, 0.8310502171516418, 0.8219178318977356, 0.8515982031822205, 0.9086757898330688, 0.9748858213424683], [0.06164383515715599, 0.13013698160648346, 0.09817351400852203, 0.16210044920444489, 0.14155250787734985, 0.1666666716337204, 0.17579908668994904, 0.1940639317035675, 0.26484018564224243, 0.2054794579744339, 0.25799086689949036, 0.2602739632129669, 0.21461187303066254, 0.28082191944122314, 0.310502290725708, 0.29680365324020386, 0.29223743081092834, 0.25342464447021484, 0.310502290725708, 0.284246563911438, 0.3219178020954132, 0.3767123222351074, 0.3493150770664215, 0.310502290725708, 0.36415526270866394, 0.3287671208381653, 0.3493150770664215, 0.3652968108654022, 0.44748857617378235, 0.3904109597206116, 0.422374427318573, 0.4611872136592865, 0.46347030997276306, 0.36986300349235535, 0.439497709274292, 0.4680365324020386, 0.28082191944122314, 0.456620991230011, 0.4840182662010193, 0.45890411734580994, 0.4406392574310303, 0.4954337775707245, 0.4771689474582672, 0.388127863407135, 0.5068492889404297, 0.5045661926269531, 0.5022830963134766, 0.5410959124565125, NaN, 0.534246563911438, 0.5296803712844849, 0.6347032189369202, 0.6392694115638733, 0.5547945499420166, 0.5753424763679504, 0.6506849527359009, 0.6529680490493774, 0.7134703397750854, 0.698630154132843, 0.6324200630187988, 0.7271689772605896, 0.6940639019012451, 0.7031963467597961, 0.6917808055877686, 0.7534246444702148, 0.6803653240203857, 0.7191780805587769, 0.6735159754753113, 0.7602739930152893, 0.6963470578193665, 0.7625570893287659, 0.7773972749710083, 0.7739726305007935, 0.7990867495536804, 0.8082191944122314, 0.810502290725708, 0.8515982031822205, 0.965753436088562], [0.07305935770273209, 0.09474886208772659, 0.10730593651533127, 0.14155250787734985, 0.14155250787734985, 0.15410958230495453, 0.18036529421806335, 0.19634702801704407, 0.24086758494377136, 0.2054794579744339, 0.24200913310050964, 0.24657534062862396, 0.2557077705860138, 0.2990867495536804, 0.27625569701194763, 0.2716895043849945, 0.310502290725708, 0.3059360682964325, 0.3561643958091736, 0.3287671208381653, 0.344748854637146, 0.344748854637146, 0.3561643958091736, 0.3675799071788788, 0.2888127863407135, 0.33561643958091736, 0.34018266201019287, 0.3767123222351074, 0.4292237460613251, 0.388127863407135, 0.42009133100509644, 0.4440639317035675, 0.42465752363204956, 0.40753424167633057, 0.4155251085758209, 0.46917808055877686, 0.4063926935195923, 0.46347030997276306, 0.4817351698875427, 0.4440639317035675, 0.48515981435775757, 0.4452054798603058, 0.4623287618160248, 0.4623287618160248, 0.49771690368652344, 0.5045661926269531, 0.4794520437717438, 0.5319634675979614, 0.465753436088562, NaN, 0.5525113940238953, 0.6152967810630798, 0.6073059439659119, 0.5502282977104187, 0.5045661926269531, 0.6141552329063416, 0.6347032189369202, 0.6095890402793884, 0.663241982460022, 0.6244292259216309, 0.6506849527359009, 0.6780821681022644, 0.6335616707801819, 0.7557077407836914, 0.7100456357002258, 0.6826484203338623, 0.6689497828483582, 0.7534246444702148, 0.72374427318573, 0.7305936217308044, 0.7283105254173279, 0.7831050157546997, 0.8059360980987549, 0.7694063782691956, 0.7922374606132507, 0.8333333134651184, 0.8299086689949036, 0.9589040875434875], [0.07305935770273209, 0.12671232223510742, 0.13470318913459778, 0.11872146278619766, 0.15753424167633057, 0.14611871540546417, 0.15296803414821625, 0.16894976794719696, 0.2054794579744339, 0.2214611917734146, 0.2271689474582672, 0.25799086689949036, 0.26484018564224243, 0.2888127863407135, 0.26484018564224243, 0.2557077705860138, 0.30365297198295593, 0.31278538703918457, 0.29680365324020386, 0.29680365324020386, 0.3595890402793884, 0.3059360682964325, 0.2899543344974518, 0.37899544835090637, 0.3116438388824463, 0.35159817337989807, 0.36986300349235535, 0.36872145533561707, 0.3835616409778595, 0.3264840245246887, 0.36301368474960327, 0.34246575832366943, 0.3333333432674408, 0.4178082048892975, 0.42465752363204956, 0.3607305884361267, 0.439497709274292, 0.44178083539009094, 0.2945205569267273, 0.3550228178501129, 0.4452054798603058, 0.42465752363204956, 0.4063926935195923, 0.4623287618160248, 0.422374427318573, 0.4623287618160248, 0.4406392574310303, 0.46347030997276306, 0.47031962871551514, 0.44748857617378235, NaN, 0.5525113940238953, 0.5456621050834656, 0.5159817337989807, 0.5068492889404297, 0.5936073064804077, 0.5958904027938843, 0.5799086689949036, 0.5890411138534546, 0.6666666865348816, 0.6027397513389587, 0.6598173379898071, 0.6689497828483582, 0.6906392574310303, 0.6609588861465454, 0.7054794430732727, 0.6974886059761047, 0.7853881120681763, 0.6780821681022644, 0.7442922592163086, 0.7488584518432617, 0.7831050157546997, 0.7488584518432617, 0.767123281955719, 0.7876712083816528, 0.8607305884361267, 0.8401826620101929, 0.965753436088562], [0.18036529421806335, 0.21575342118740082, 0.25342464447021484, 0.20776255428791046, 0.2557077705860138, 0.2214611917734146, 0.25342464447021484, 0.26255708932876587, 0.29223743081092834, 0.29680365324020386, 0.3059360682964325, 0.29680365324020386, 0.2945205569267273, 0.361872136592865, 0.344748854637146, 0.29223743081092834, 0.38127854466438293, 0.31849315762519836, 0.27397260069847107, 0.31278538703918457, 0.4041095972061157, 0.39611873030662537, 0.3287671208381653, 0.33561643958091736, 0.3652968108654022, 0.327625572681427, 0.34018266201019287, 0.3378995358943939, 0.38584473729133606, 0.36301368474960327, 0.4109589159488678, 0.4006849229335785, 0.3767123222351074, 0.4178082048892975, 0.3652968108654022, 0.43607306480407715, 0.34246575832366943, 0.3561643958091736, 0.3835616409778595, 0.38127854466438293, 0.38127854466438293, 0.38127854466438293, 0.3493150770664215, 0.35844749212265015, 0.3721461296081543, 0.3767123222351074, 0.456620991230011, 0.3904109597206116, 0.3652968108654022, 0.3847031891345978, 0.44748857617378235, NaN, 0.6095890402793884, 0.4041095972061157, 0.5296803712844849, 0.568493127822876, 0.5673515796661377, 0.4817351698875427, 0.4155251085758209, 0.6164383292198181, 0.4885844886302948, 0.9292237162590027, 0.611872136592865, 0.5159817337989807, 0.6050228476524353, 0.6369863152503967, 0.6347032189369202, 0.6415525078773499, 0.6347032189369202, 0.6438356041908264, 0.6917808055877686, 0.7557077407836914, 0.7100456357002258, 0.7831050157546997, 0.7785388231277466, 0.7031963467597961, 0.8515982031822205, 0.931506872177124], [0.1780821979045868, 0.20319634675979614, 0.2511415481567383, 0.21004566550254822, 0.2442922443151474, 0.21803653240203857, 0.24657534062862396, 0.2511415481567383, 0.2831050157546997, 0.2876712381839752, 0.2945205569267273, 0.29223743081092834, 0.2899543344974518, 0.36301368474960327, 0.34246575832366943, 0.28082191944122314, 0.37899544835090637, 0.31506848335266113, 0.267123281955719, 0.3093607425689697, 0.3995433747768402, 0.3949771821498871, 0.3173516094684601, 0.33105021715164185, 0.361872136592865, 0.3219178020954132, 0.344748854637146, 0.3333333432674408, 0.3892694115638733, 0.3561643958091736, 0.4063926935195923, 0.4018264710903168, 0.3721461296081543, 0.42009133100509644, 0.36301368474960327, 0.43607306480407715, 0.344748854637146, 0.35844749212265015, 0.3767123222351074, 0.3732876777648926, 0.3721461296081543, 0.37557077407836914, 0.34246575832366943, 0.3607305884361267, 0.35388126969337463, 0.37100455164909363, 0.45205479860305786, 0.3949771821498871, 0.3607305884361267, 0.39269405603408813, 0.4543378949165344, 0.3904109597206116, NaN, 0.3995433747768402, 0.5205479264259338, 0.5479452013969421, 0.5502282977104187, 0.47031962871551514, 0.4178082048892975, 0.6073059439659119, 0.48630136251449585, 0.9474886059761047, 0.6073059439659119, 0.5136986374855042, 0.5958904027938843, 0.6347032189369202, 0.6347032189369202, 0.6392694115638733, 0.6255707740783691, 0.6392694115638733, 0.689497709274292, 0.7442922592163086, 0.7031963467597961, 0.77625572681427, 0.7739726305007935, 0.7009132504463196, 0.8401826620101929, 0.931506872177124], [0.031963471323251724, 0.06621004641056061, 0.07990867644548416, 0.08219178020954132, 0.08675798773765564, 0.10273972898721695, 0.12100456655025482, 0.12328767031431198, 0.2054794579744339, 0.17123287916183472, 0.21232876181602478, 0.22602739930152893, 0.21347032487392426, 0.2602739632129669, 0.26255708932876587, 0.28082191944122314, 0.2557077705860138, 0.27397260069847107, 0.2876712381839752, 0.2659817337989807, 0.2945205569267273, 0.30365297198295593, 0.27625569701194763, 0.3344748914241791, 0.2945205569267273, 0.2990867495536804, 0.30821916460990906, 0.31849315762519836, 0.38584473729133606, 0.30365297198295593, 0.35844749212265015, 0.3767123222351074, 0.3675799071788788, 0.36986300349235535, 0.38127854466438293, 0.3675799071788788, 0.3835616409778595, 0.3847031891345978, 0.4063926935195923, 0.25228309631347656, 0.4269406497478485, 0.4109589159488678, 0.40867578983306885, 0.42465752363204956, 0.4063926935195923, 0.4383561611175537, 0.4292237460613251, 0.43607306480407715, 0.4452054798603058, 0.4497717022895813, 0.4840182662010193, 0.5958904027938843, 0.6004565954208374, NaN, 0.5228310227394104, 0.6130136847496033, 0.6073059439659119, 0.5662100315093994, 0.6506849527359009, 0.6803653240203857, 0.594748854637146, 0.689497709274292, 0.6598173379898071, 0.7682648301124573, 0.6666666865348816, 0.6963470578193665, 0.6746575236320496, 0.7031963467597961, 0.6917808055877686, 0.7305936217308044, 0.7625570893287659, 0.7728310227394104, 0.7602739930152893, 0.7534246444702148, 0.7648401856422424, 0.8470319509506226, 0.810502290725708, 0.9611872434616089], [0.05707762390375137, 0.09817351400852203, 0.09474886208772659, 0.07534246891736984, 0.14840182662010193, 0.08675798773765564, 0.12328767031431198, 0.1278538852930069, 0.15981735289096832, 0.30251142382621765, 0.2899543344974518, 0.20319634675979614, 0.3059360682964325, 0.2728310525417328, 0.2168949842453003, 0.155251145362854, 0.3070776164531708, 0.32420089840888977, 0.3059360682964325, 0.3333333432674408, 0.33561643958091736, 0.2214611917734146, 0.3664383590221405, 0.34246575832366943, 0.3732876777648926, 0.3995433747768402, 0.4041095972061157, 0.40867578983306885, 0.3607305884361267, 0.37899544835090637, 0.28538814187049866, 0.35388126969337463, 0.37899544835090637, 0.41324201226234436, 0.405251145362854, 0.36986300349235535, 0.40867578983306885, 0.37899544835090637, 0.4155251085758209, 0.42009133100509644, 0.3561643958091736, 0.43150684237480164, 0.3995433747768402, 0.42009133100509644, 0.41324201226234436, 0.40981733798980713, 0.5, 0.4063926935195923, 0.42465752363204956, 0.4954337775707245, 0.4931506812572479, 0.47031962871551514, 0.4794520437717438, 0.4771689474582672, NaN, 0.49771690368652344, 0.5022830963134766, 0.5205479264259338, 0.5296803712844849, 0.6187214851379395, 0.5639269351959229, 0.6073059439659119, 0.6278538703918457, 0.5821917653083801, 0.5547945499420166, 0.6164383292198181, 0.6563926935195923, 0.655251145362854, 0.5890411138534546, 0.6324200630187988, 0.6917808055877686, 0.6621004343032837, 0.7420091032981873, 0.7260273694992065, 0.698630154132843, 0.7922374606132507, 0.7146118879318237, 0.9611872434616089], [0.04794520512223244, 0.06164383515715599, 0.07534246891736984, 0.08789954334497452, 0.08675798773765564, 0.10730593651533127, 0.1164383590221405, 0.13127854466438293, 0.15410958230495453, 0.2009132355451584, 0.16210044920444489, 0.1666666716337204, 0.1826484054327011, 0.24200913310050964, 0.2054794579744339, 0.19863013923168182, 0.30136987566947937, 0.19292236864566803, 0.20776255428791046, 0.24885845184326172, 0.3116438388824463, 0.2602739632129669, 0.25799086689949036, 0.24315068125724792, 0.2705479562282562, 0.2283104956150055, 0.23059360682964325, 0.25342464447021484, 0.3767123222351074, 0.31506848335266113, 0.27397260069847107, 0.28538814187049866, 0.28082191944122314, 0.4018264710903168, 0.28538814187049866, 0.344748854637146, 0.3264840245246887, 0.2773972749710083, 0.34018266201019287, 0.310502290725708, 0.3493150770664215, 0.29109588265419006, 0.31963470578193665, 0.35844749212265015, 0.34246575832366943, 0.3378995358943939, 0.4109589159488678, 0.3333333432674408, 0.3493150770664215, 0.38584473729133606, 0.4063926935195923, 0.43150684237480164, 0.45205479860305786, 0.38698631525039673, 0.5022830963134766, NaN, 0.6324200630187988, 0.4406392574310303, 0.4931506812572479, 0.5707762837409973, 0.4885844886302948, 0.6221461296081543, 0.5993150472640991, 0.5593607425689697, 0.6586757898330688, 0.577625572681427, 0.628995418548584, 0.6004565954208374, 0.7009132504463196, 0.5981734991073608, 0.7031963467597961, 0.7853881120681763, 0.7077625393867493, 0.8721461296081543, 0.8378995656967163, 0.7260273694992065, 0.8424657583236694, 0.9885844588279724], [0.05707762390375137, 0.06621004641056061, 0.09132420271635056, 0.0890410989522934, 0.09474886208772659, 0.10273972898721695, 0.12328767031431198, 0.1278538852930069, 0.12100456655025482, 0.19178082048892975, 0.172374427318573, 0.172374427318573, 0.2009132355451584, 0.2442922443151474, 0.19863013923168182, 0.20319634675979614, 0.293379008769989, 0.21004566550254822, 0.20776255428791046, 0.24200913310050964, 0.3059360682964325, 0.2397260218858719, 0.24657534062862396, 0.2340182662010193, 0.25342464447021484, 0.23515981435775757, 0.23630136251449585, 0.26255708932876587, 0.3607305884361267, 0.3059360682964325, 0.267123281955719, 0.26484018564224243, 0.2511415481567383, 0.3607305884361267, 0.26484018564224243, 0.310502290725708, 0.31506848335266113, 0.2831050157546997, 0.34018266201019287, 0.31278538703918457, 0.33105021715164185, 0.27397260069847107, 0.3059360682964325, 0.3378995358943939, 0.32420089840888977, 0.2990867495536804, 0.3904109597206116, 0.34018266201019287, 0.34703195095062256, 0.3652968108654022, 0.4041095972061157, 0.4326483905315399, 0.4497717022895813, 0.39269405603408813, 0.49771690368652344, 0.3675799071788788, NaN, 0.4109589159488678, 0.49086758494377136, 0.5707762837409973, 0.4383561611175537, 0.6164383292198181, 0.5913242101669312, 0.543379008769989, 0.6027397513389587, 0.5456621050834656, 0.6164383292198181, 0.5730593800544739, 0.646118700504303, 0.5844748616218567, 0.6963470578193665, 0.7374429106712341, 0.6803653240203857, 0.8059360980987549, 0.8002282977104187, 0.7054794430732727, 0.8493150472640991, 0.9863013625144958], [0.029680365696549416, 0.05936073139309883, 0.06621004641056061, 0.08675798773765564, 0.06392694264650345, 0.11187214404344559, 0.11187214404344559, 0.11872146278619766, 0.1826484054327011, 0.10273972898721695, 0.1278538852930069, 0.14840182662010193, 0.14611871540546417, 0.22374428808689117, 0.21917808055877686, 0.2397260218858719, 0.23744292557239532, 0.13013698160648346, 0.23515981435775757, 0.19178082048892975, 0.24885845184326172, 0.2876712381839752, 0.22374428808689117, 0.18150684237480164, 0.18949772417545319, 0.1826484054327011, 0.1780821979045868, 0.1666666716337204, 0.35844749212265015, 0.2899543344974518, 0.3287671208381653, 0.2990867495536804, 0.36301368474960327, 0.3561643958091736, 0.13470318913459778, 0.37442922592163086, 0.232876718044281, 0.2819634675979614, 0.35844749212265015, 0.29223743081092834, 0.3835616409778595, 0.20319634675979614, 0.37442922592163086, 0.2682648301124573, 0.34018266201019287, 0.3219178020954132, 0.33105021715164185, 0.36301368474960327, 0.28652969002723694, 0.3904109597206116, 0.42009133100509644, 0.5182648301124573, 0.5296803712844849, 0.4337899684906006, 0.4794520437717438, 0.5593607425689697, 0.5890411138534546, NaN, 0.5730593800544739, 0.5456621050834656, 0.5011415481567383, 0.6187214851379395, 0.6004565954208374, 0.5981734991073608, 0.6278538703918457, 0.5182648301124573, 0.6141552329063416, 0.586758017539978, 0.6575342416763306, 0.5753424763679504, 0.6757990717887878, 0.7168949842453003, 0.7191780805587769, 0.7557077407836914, 0.7716894745826721, 0.7077625393867493, 0.7853881120681763, 0.9383561611175537], [0.05936073139309883, 0.07077625393867493, 0.077625572681427, 0.09132420271635056, 0.1004566177725792, 0.10730593651533127, 0.11872146278619766, 0.13013698160648346, 0.2009132355451584, 0.155251145362854, 0.1438356190919876, 0.18949772417545319, 0.15068493783473969, 0.23059360682964325, 0.19863013923168182, 0.2214611917734146, 0.24657534062862396, 0.15068493783473969, 0.18721461296081543, 0.18493150174617767, 0.2831050157546997, 0.26484018564224243, 0.20776255428791046, 0.17922374606132507, 0.21004566550254822, 0.16894976794719696, 0.21347032487392426, 0.23059360682964325, 0.3173516094684601, 0.25342464447021484, 0.30365297198295593, 0.2899543344974518, 0.27397260069847107, 0.31278538703918457, 0.26940637826919556, 0.34703195095062256, 0.26255708932876587, 0.2602739632129669, 0.31963470578193665, 0.2568493187427521, 0.2899543344974518, 0.28082191944122314, 0.3173516094684601, 0.30821916460990906, 0.30821916460990906, 0.31506848335266113, 0.3721461296081543, 0.29680365324020386, 0.30136987566947937, 0.33675798773765564, 0.4109589159488678, 0.5844748616218567, 0.5821917653083801, 0.3493150770664215, 0.47031962871551514, 0.5068492889404297, 0.509132444858551, 0.4269406497478485, NaN, 0.543379008769989, 0.46347030997276306, 0.6529680490493774, 0.5981734991073608, 0.5479452013969421, 0.6141552329063416, 0.5182648301124573, 0.6301369667053223, 0.5844748616218567, 0.6347032189369202, 0.5319634675979614, 0.6940639019012451, 0.7579908967018127, 0.6621004343032837, 0.698630154132843, 0.7305936217308044, 0.6415525078773499, 0.8219178318977356, 0.9360730648040771], [0.07305935770273209, 0.09589041024446487, 0.11872146278619766, 0.10273972898721695, 0.12100456655025482, 0.09703196585178375, 0.12557077407836914, 0.1392694115638733, 0.1278538852930069, 0.21232876181602478, 0.23515981435775757, 0.18721461296081543, 0.232876718044281, 0.28538814187049866, 0.20319634675979614, 0.11187214404344559, 0.22602739930152893, 0.29223743081092834, 0.2385844737291336, 0.24885845184326172, 0.267123281955719, 0.2214611917734146, 0.25913241505622864, 0.30821916460990906, 0.26484018564224243, 0.34246575832366943, 0.35844749212265015, 0.3675799071788788, 0.22602739930152893, 0.26940637826919556, 0.21232876181602478, 0.2511415481567383, 0.23515981435775757, 0.30365297198295593, 0.35273972153663635, 0.25342464447021484, 0.33561643958091736, 0.30365297198295593, 0.26940637826919556, 0.2990867495536804, 0.28538814187049866, 0.3652968108654022, 0.3607305884361267, 0.36301368474960327, 0.2819634675979614, 0.3321917951107025, 0.3995433747768402, 0.3219178020954132, 0.3675799071788788, 0.37557077407836914, 0.3333333432674408, 0.3835616409778595, 0.39269405603408813, 0.31963470578193665, 0.38127854466438293, 0.4292237460613251, 0.4292237460613251, 0.4543378949165344, 0.456620991230011, NaN, 0.47031962871551514, 0.5296803712844849, 0.5251141786575317, 0.4783104956150055, 0.5022830963134766, 0.5639269351959229, 0.5650684833526611, 0.5456621050834656, 0.5136986374855042, 0.6027397513389587, 0.6141552329063416, 0.611872136592865, 0.6324200630187988, 0.6232876777648926, 0.6529680490493774, 0.698630154132843, 0.6940639019012451, 0.8926940560340881], [0.027397260069847107, 0.04566210135817528, 0.052511416375637054, 0.05936073139309883, 0.052511416375637054, 0.05936073139309883, 0.06621004641056061, 0.077625572681427, 0.1392694115638733, 0.07534246891736984, 0.09817351400852203, 0.10159817337989807, 0.09474886208772659, 0.20319634675979614, 0.16894976794719696, 0.2111872136592865, 0.2511415481567383, 0.0502283088862896, 0.18150684237480164, 0.16438356041908264, 0.2442922443151474, 0.24885845184326172, 0.18949772417545319, 0.08561643958091736, 0.14155250787734985, 0.14840182662010193, 0.13698630034923553, 0.15981735289096832, 0.3333333432674408, 0.25342464447021484, 0.3287671208381653, 0.26255708932876587, 0.2990867495536804, 0.3264840245246887, 0.10730593651533127, 0.3219178020954132, 0.1997716873884201, 0.25799086689949036, 0.31392693519592285, 0.29680365324020386, 0.33105021715164185, 0.15981735289096832, 0.33561643958091736, 0.25799086689949036, 0.310502290725708, 0.29223743081092834, 0.31278538703918457, 0.33561643958091736, 0.2728310525417328, 0.3493150770664215, 0.39726027846336365, 0.5114155411720276, 0.5136986374855042, 0.405251145362854, 0.43607306480407715, 0.5114155411720276, 0.5616438388824463, 0.4988584518432617, 0.5365296602249146, 0.5296803712844849, NaN, 0.6073059439659119, 0.577625572681427, 0.586758017539978, 0.5913242101669312, 0.5182648301124573, 0.5890411138534546, 0.5707762837409973, 0.6198630332946777, 0.586758017539978, 0.6621004343032837, 0.6917808055877686, 0.7260273694992065, 0.7511415481567383, 0.7682648301124573, 0.732876718044281, 0.7945205569267273, 0.9577625393867493], [0.14611871540546417, 0.16894976794719696, 0.19863013923168182, 0.16438356041908264, 0.21004566550254822, 0.18721461296081543, 0.21004566550254822, 0.20776255428791046, 0.21004566550254822, 0.2397260218858719, 0.2511415481567383, 0.22602739930152893, 0.24657534062862396, 0.2956621050834656, 0.2397260218858719, 0.21232876181602478, 0.31963470578193665, 0.2397260218858719, 0.2111872136592865, 0.267123281955719, 0.32420089840888977, 0.26255708932876587, 0.26484018564224243, 0.25799086689949036, 0.28082191944122314, 0.2716895043849945, 0.2831050157546997, 0.27625569701194763, 0.30821916460990906, 0.2716895043849945, 0.32420089840888977, 0.310502290725708, 0.2831050157546997, 0.34018266201019287, 0.30136987566947937, 0.30136987566947937, 0.2899543344974518, 0.2716895043849945, 0.2990867495536804, 0.2785388231277466, 0.2602739632129669, 0.3059360682964325, 0.25799086689949036, 0.29680365324020386, 0.29680365324020386, 0.29223743081092834, 0.38584473729133606, 0.2945205569267273, 0.3059360682964325, 0.3219178020954132, 0.34018266201019287, 0.07077625393867493, 0.052511416375637054, 0.310502290725708, 0.39269405603408813, 0.3778538703918457, 0.3835616409778595, 0.38127854466438293, 0.34703195095062256, 0.47031962871551514, 0.39269405603408813, NaN, 0.4885844886302948, 0.3904109597206116, 0.42579910159111023, 0.5410959124565125, 0.5022830963134766, 0.4543378949165344, 0.4497717022895813, 0.5456621050834656, 0.5616438388824463, 0.6050228476524353, 0.6255707740783691, 0.6187214851379395, 0.5936073064804077, 0.6232876777648926, 0.7260273694992065, 0.922374427318573], [0.07534246891736984, 0.12100456655025482, 0.12557077407836914, 0.10502283275127411, 0.13242009282112122, 0.14611871540546417, 0.14611871540546417, 0.1666666716337204, 0.19178082048892975, 0.16324201226234436, 0.17351597547531128, 0.21917808055877686, 0.1666666716337204, 0.2283104956150055, 0.24885845184326172, 0.21004566550254822, 0.23515981435775757, 0.1826484054327011, 0.16894976794719696, 0.19863013923168182, 0.26484018564224243, 0.25913241505622864, 0.2442922443151474, 0.2168949842453003, 0.28082191944122314, 0.2442922443151474, 0.24885845184326172, 0.25342464447021484, 0.2545662224292755, 0.2831050157546997, 0.2716895043849945, 0.23059360682964325, 0.2054794579744339, 0.2990867495536804, 0.2602739632129669, 0.25799086689949036, 0.28538814187049866, 0.23059360682964325, 0.24657534062862396, 0.2705479562282562, 0.2785388231277466, 0.27625569701194763, 0.2831050157546997, 0.310502290725708, 0.2945205569267273, 0.26255708932876587, 0.4018264710903168, 0.2557077705860138, 0.29680365324020386, 0.3664383590221405, 0.33105021715164185, 0.388127863407135, 0.39269405603408813, 0.34018266201019287, 0.3721461296081543, 0.4006849229335785, 0.40867578983306885, 0.3995433747768402, 0.4018264710903168, 0.47488585114479065, 0.422374427318573, 0.5114155411720276, NaN, 0.4429223835468292, 0.43150684237480164, 0.4794520437717438, 0.6678082346916199, 0.4931506812572479, 0.45890411734580994, 0.5022830963134766, 0.7579908967018127, 0.5753424763679504, 0.5981734991073608, 0.5936073064804077, 0.5662100315093994, 0.621004581451416, 0.6780821681022644, 0.8926940560340881], [0.02283105067908764, 0.052511416375637054, 0.052511416375637054, 0.04337899386882782, 0.0502283088862896, 0.0388127863407135, 0.07534246891736984, 0.077625572681427, 0.12100456655025482, 0.11415524780750275, 0.10730593651533127, 0.13013698160648346, 0.10958904027938843, 0.21917808055877686, 0.15981735289096832, 0.1392694115638733, 0.21917808055877686, 0.18949772417545319, 0.18949772417545319, 0.12557077407836914, 0.2442922443151474, 0.18949772417545319, 0.155251145362854, 0.20776255428791046, 0.1826484054327011, 0.2009132355451584, 0.2168949842453003, 0.23515981435775757, 0.2831050157546997, 0.23515981435775757, 0.21461187303066254, 0.2009132355451584, 0.18493150174617767, 0.31963470578193665, 0.25342464447021484, 0.26255708932876587, 0.26940637826919556, 0.29680365324020386, 0.1952054798603058, 0.13470318913459778, 0.20776255428791046, 0.28082191944122314, 0.25799086689949036, 0.28082191944122314, 0.22488585114479065, 0.3173516094684601, 0.344748854637146, 0.29223743081092834, 0.30821916460990906, 0.2442922443151474, 0.3093607425689697, 0.4840182662010193, 0.48630136251449585, 0.23173515498638153, 0.4178082048892975, 0.4406392574310303, 0.456620991230011, 0.4018264710903168, 0.45205479860305786, 0.5216894745826721, 0.41324201226234436, 0.6095890402793884, 0.5570776462554932, NaN, 0.49086758494377136, 0.5456621050834656, 0.5799086689949036, 0.577625572681427, 0.5479452013969421, 0.5844748616218567, 0.6598173379898071, 0.698630154132843, 0.6529680490493774, 0.664383590221405, 0.664383590221405, 0.7260273694992065, 0.7785388231277466, 0.9406392574310303], [0.02283105067908764, 0.06164383515715599, 0.05707762390375137, 0.06164383515715599, 0.06392694264650345, 0.10273972898721695, 0.08675798773765564, 0.11187214404344559, 0.12328767031431198, 0.10730593651533127, 0.0890410989522934, 0.13242009282112122, 0.13470318913459778, 0.19634702801704407, 0.1438356190919876, 0.1940639317035675, 0.26484018564224243, 0.1004566177725792, 0.15753424167633057, 0.1666666716337204, 0.2511415481567383, 0.21461187303066254, 0.1780821979045868, 0.15981735289096832, 0.15981735289096832, 0.12328767031431198, 0.13013698160648346, 0.13013698160648346, 0.3219178020954132, 0.23059360682964325, 0.26255708932876587, 0.19634702801704407, 0.24885845184326172, 0.35388126969337463, 0.11986301094293594, 0.293379008769989, 0.2009132355451584, 0.21917808055877686, 0.24315068125724792, 0.24885845184326172, 0.27397260069847107, 0.14840182662010193, 0.2214611917734146, 0.23059360682964325, 0.26940637826919556, 0.25799086689949036, 0.34246575832366943, 0.24885845184326172, 0.24657534062862396, 0.2899543344974518, 0.3390410840511322, 0.3949771821498871, 0.4041095972061157, 0.3333333432674408, 0.4452054798603058, 0.34132421016693115, 0.39726027846336365, 0.3721461296081543, 0.38584473729133606, 0.49771690368652344, 0.40867578983306885, 0.5742009282112122, 0.568493127822876, 0.509132444858551, NaN, 0.5159817337989807, 0.5844748616218567, 0.5319634675979614, 0.6324200630187988, 0.5410959124565125, 0.6655251383781433, 0.7009132504463196, 0.6849315166473389, 0.7511415481567383, 0.8173515796661377, 0.6278538703918457, 0.8150684833526611, 0.9634703397750854], [0.01826483942568302, 0.06392694264650345, 0.06392694264650345, 0.05707762390375137, 0.06849315017461777, 0.06849315017461777, 0.06849315017461777, 0.08789954334497452, 0.11872146278619766, 0.08219178020954132, 0.08675798773765564, 0.10958904027938843, 0.11187214404344559, 0.1780821979045868, 0.14155250787734985, 0.13470318913459778, 0.2168949842453003, 0.12557077407836914, 0.17123287916183472, 0.1164383590221405, 0.22374428808689117, 0.16210044920444489, 0.15296803414821625, 0.1495433747768402, 0.16438356041908264, 0.21461187303066254, 0.2054794579744339, 0.21917808055877686, 0.27397260069847107, 0.21917808055877686, 0.23059360682964325, 0.2214611917734146, 0.24315068125724792, 0.2876712381839752, 0.2602739632129669, 0.22374428808689117, 0.2442922443151474, 0.2442922443151474, 0.21232876181602478, 0.20776255428791046, 0.2511415481567383, 0.2945205569267273, 0.26369863748550415, 0.2956621050834656, 0.2442922443151474, 0.267123281955719, 0.33105021715164185, 0.28538814187049866, 0.31963470578193665, 0.3173516094684601, 0.2945205569267273, 0.36301368474960327, 0.3652968108654022, 0.30365297198295593, 0.3835616409778595, 0.422374427318573, 0.4543378949165344, 0.4817351698875427, 0.4817351698875427, 0.43607306480407715, 0.4817351698875427, 0.45890411734580994, 0.5205479264259338, 0.4543378949165344, 0.4840182662010193, NaN, 0.5319634675979614, 0.44748857617378235, 0.509132444858551, 0.6860730648040771, 0.5890411138534546, 0.6130136847496033, 0.6415525078773499, 0.664383590221405, 0.6826484203338623, 0.7922374606132507, 0.7031963467597961, 0.9429223537445068], [0.06621004641056061, 0.10958904027938843, 0.09817351400852203, 0.09132420271635056, 0.10730593651533127, 0.13470318913459778, 0.12557077407836914, 0.14155250787734985, 0.17123287916183472, 0.13584475219249725, 0.16894976794719696, 0.2009132355451584, 0.15182648599147797, 0.20319634675979614, 0.2283104956150055, 0.18493150174617767, 0.2009132355451584, 0.16894976794719696, 0.1438356190919876, 0.1826484054327011, 0.23744292557239532, 0.2283104956150055, 0.2340182662010193, 0.19292236864566803, 0.2557077705860138, 0.22374428808689117, 0.2283104956150055, 0.2385844737291336, 0.2397260218858719, 0.25799086689949036, 0.2557077705860138, 0.21004566550254822, 0.17579908668994904, 0.27625569701194763, 0.24885845184326172, 0.24657534062862396, 0.2716895043849945, 0.2168949842453003, 0.22374428808689117, 0.2397260218858719, 0.2545662224292755, 0.2511415481567383, 0.2682648301124573, 0.30365297198295593, 0.25799086689949036, 0.2442922443151474, 0.39726027846336365, 0.2340182662010193, 0.28082191944122314, 0.33105021715164185, 0.30251142382621765, 0.3652968108654022, 0.3652968108654022, 0.32534247636795044, 0.3436073064804077, 0.37100455164909363, 0.3835616409778595, 0.38584473729133606, 0.36986300349235535, 0.43493151664733887, 0.4109589159488678, 0.49771690368652344, 0.3321917951107025, 0.42009133100509644, 0.4155251085758209, 0.4680365324020386, NaN, 0.47031962871551514, 0.4337899684906006, 0.4885844886302948, 0.6757990717887878, 0.52625572681427, 0.5890411138534546, 0.568493127822876, 0.5456621050834656, 0.6187214851379395, 0.6598173379898071, 0.8835616707801819], [0.04337899386882782, 0.05936073139309883, 0.06164383515715599, 0.06849315017461777, 0.0890410989522934, 0.09132420271635056, 0.08675798773765564, 0.1004566177725792, 0.09360730648040771, 0.09132420271635056, 0.0890410989522934, 0.13584475219249725, 0.1278538852930069, 0.17123287916183472, 0.1278538852930069, 0.15296803414821625, 0.19863013923168182, 0.21461187303066254, 0.20776255428791046, 0.17579908668994904, 0.2054794579744339, 0.155251145362854, 0.12442922592163086, 0.23515981435775757, 0.18493150174617767, 0.21232876181602478, 0.21917808055877686, 0.23515981435775757, 0.2557077705860138, 0.20433789491653442, 0.19863013923168182, 0.15068493783473969, 0.16210044920444489, 0.28082191944122314, 0.2785388231277466, 0.2397260218858719, 0.2990867495536804, 0.24200913310050964, 0.06392694264650345, 0.19748859107494354, 0.23744292557239532, 0.29223743081092834, 0.2716895043849945, 0.31506848335266113, 0.18493150174617767, 0.28538814187049866, 0.3059360682964325, 0.2785388231277466, 0.3264840245246887, 0.24657534062862396, 0.21461187303066254, 0.35844749212265015, 0.3607305884361267, 0.29680365324020386, 0.344748854637146, 0.3995433747768402, 0.4269406497478485, 0.41324201226234436, 0.4155251085758209, 0.4543378949165344, 0.4292237460613251, 0.5456621050834656, 0.5068492889404297, 0.422374427318573, 0.4680365324020386, 0.5525113940238953, 0.5296803712844849, NaN, 0.47488585114479065, 0.5799086689949036, 0.5662100315093994, 0.6415525078773499, 0.6232876777648926, 0.6278538703918457, 0.6130136847496033, 0.7146118879318237, 0.7397260069847107, 0.9429223537445068], [0.0251141544431448, 0.0388127863407135, 0.054794520139694214, 0.06392694264650345, 0.04566210135817528, 0.06621004641056061, 0.07534246891736984, 0.08675798773765564, 0.10958904027938843, 0.09703196585178375, 0.07077625393867493, 0.10273972898721695, 0.11415524780750275, 0.17579908668994904, 0.1438356190919876, 0.17579908668994904, 0.23630136251449585, 0.077625572681427, 0.15753424167633057, 0.16894976794719696, 0.23059360682964325, 0.18036529421806335, 0.16210044920444489, 0.12557077407836914, 0.1392694115638733, 0.11872146278619766, 0.1107305958867073, 0.13013698160648346, 0.293379008769989, 0.2214611917734146, 0.23059360682964325, 0.17579908668994904, 0.21004566550254822, 0.31392693519592285, 0.12442922592163086, 0.27968037128448486, 0.19178082048892975, 0.2054794579744339, 0.232876718044281, 0.21917808055877686, 0.2397260218858719, 0.14611871540546417, 0.22374428808689117, 0.2283104956150055, 0.24200913310050964, 0.23059360682964325, 0.34018266201019287, 0.2283104956150055, 0.2397260218858719, 0.27625569701194763, 0.3219178020954132, 0.3652968108654022, 0.37442922592163086, 0.30821916460990906, 0.4109589159488678, 0.2990867495536804, 0.35388126969337463, 0.34246575832366943, 0.3652968108654022, 0.48630136251449585, 0.38013699650764465, 0.5502282977104187, 0.5410959124565125, 0.45205479860305786, 0.3675799071788788, 0.49086758494377136, 0.5662100315093994, 0.5251141786575317, NaN, 0.5273972749710083, 0.6529680490493774, 0.6769406199455261, 0.664383590221405, 0.7146118879318237, 0.7363013625144958, 0.611872136592865, 0.8082191944122314, 0.965753436088562], [0.015981735661625862, 0.04794520512223244, 0.04908675700426102, 0.04794520512223244, 0.06849315017461777, 0.054794520139694214, 0.06164383515715599, 0.07534246891736984, 0.09589041024446487, 0.07990867644548416, 0.07534246891736984, 0.09132420271635056, 0.0890410989522934, 0.155251145362854, 0.12671232223510742, 0.13470318913459778, 0.1940639317035675, 0.10273972898721695, 0.155251145362854, 0.10273972898721695, 0.19178082048892975, 0.15296803414821625, 0.14840182662010193, 0.1278538852930069, 0.1438356190919876, 0.17579908668994904, 0.18036529421806335, 0.18949772417545319, 0.26940637826919556, 0.19178082048892975, 0.2009132355451584, 0.16780821979045868, 0.20205479860305786, 0.2819634675979614, 0.2226027399301529, 0.2111872136592865, 0.21917808055877686, 0.22488585114479065, 0.19178082048892975, 0.17351597547531128, 0.21917808055877686, 0.23515981435775757, 0.25342464447021484, 0.2682648301124573, 0.20776255428791046, 0.24885845184326172, 0.3264840245246887, 0.2716895043849945, 0.30365297198295593, 0.26940637826919556, 0.2557077705860138, 0.3561643958091736, 0.3607305884361267, 0.26940637826919556, 0.3675799071788788, 0.4018264710903168, 0.4155251085758209, 0.42465752363204956, 0.4680365324020386, 0.39726027846336365, 0.41324201226234436, 0.4543378949165344, 0.49771690368652344, 0.4155251085758209, 0.45890411734580994, 0.31392693519592285, 0.5114155411720276, 0.42009133100509644, 0.4726027250289917, NaN, 0.5662100315093994, 0.5799086689949036, 0.6221461296081543, 0.6415525078773499, 0.6575342416763306, 0.7340182662010193, 0.664383590221405, 0.9155251383781433], [0.04566210135817528, 0.06849315017461777, 0.07420091331005096, 0.06278538703918457, 0.07990867644548416, 0.07077625393867493, 0.09474886208772659, 0.09018264710903168, 0.09360730648040771, 0.10730593651533127, 0.14840182662010193, 0.12442922592163086, 0.12100456655025482, 0.19634702801704407, 0.16210044920444489, 0.14269407093524933, 0.1666666716337204, 0.11872146278619766, 0.0890410989522934, 0.15068493783473969, 0.21232876181602478, 0.18036529421806335, 0.18493150174617767, 0.14155250787734985, 0.21917808055877686, 0.17579908668994904, 0.1940639317035675, 0.20319634675979614, 0.2009132355451584, 0.2054794579744339, 0.20776255428791046, 0.17694063484668732, 0.15068493783473969, 0.2511415481567383, 0.2054794579744339, 0.19178082048892975, 0.232876718044281, 0.18036529421806335, 0.19178082048892975, 0.18721461296081543, 0.1826484054327011, 0.21461187303066254, 0.2168949842453003, 0.25342464447021484, 0.20776255428791046, 0.21461187303066254, 0.344748854637146, 0.1826484054327011, 0.23744292557239532, 0.2716895043849945, 0.2511415481567383, 0.30821916460990906, 0.310502290725708, 0.23744292557239532, 0.30821916460990906, 0.29680365324020386, 0.30365297198295593, 0.32420089840888977, 0.3059360682964325, 0.38584473729133606, 0.3378995358943939, 0.4383561611175537, 0.24200913310050964, 0.34018266201019287, 0.3344748914241791, 0.4109589159488678, 0.32420089840888977, 0.4337899684906006, 0.34703195095062256, 0.4337899684906006, NaN, 0.4429223835468292, 0.5639269351959229, 0.5068492889404297, 0.5114155411720276, 0.5559360980987549, 0.5936073064804077, 0.8333333134651184], [0.027397260069847107, 0.06164383515715599, 0.07305935770273209, 0.052511416375637054, 0.07305935770273209, 0.09132420271635056, 0.0890410989522934, 0.09589041024446487, 0.09589041024446487, 0.08675798773765564, 0.09589041024446487, 0.1164383590221405, 0.11872146278619766, 0.19178082048892975, 0.1335616409778595, 0.14840182662010193, 0.2283104956150055, 0.1004566177725792, 0.0890410989522934, 0.14611871540546417, 0.2214611917734146, 0.1438356190919876, 0.155251145362854, 0.13698630034923553, 0.15753424167633057, 0.11415524780750275, 0.11415524780750275, 0.11872146278619766, 0.25342464447021484, 0.18949772417545319, 0.17351597547531128, 0.12899543344974518, 0.13013698160648346, 0.27397260069847107, 0.15981735289096832, 0.20319634675979614, 0.18949772417545319, 0.13013698160648346, 0.14155250787734985, 0.17351597547531128, 0.14840182662010193, 0.15867580473423004, 0.17579908668994904, 0.2168949842453003, 0.155251145362854, 0.14840182662010193, 0.27625569701194763, 0.13698630034923553, 0.2226027399301529, 0.2168949842453003, 0.2168949842453003, 0.2442922443151474, 0.2557077705860138, 0.2271689474582672, 0.3378995358943939, 0.21461187303066254, 0.26255708932876587, 0.2831050157546997, 0.24200913310050964, 0.388127863407135, 0.30821916460990906, 0.3949771821498871, 0.42465752363204956, 0.30136987566947937, 0.2990867495536804, 0.38698631525039673, 0.47374430298805237, 0.35844749212265015, 0.3230593502521515, 0.42009133100509644, 0.5570776462554932, NaN, 0.5844748616218567, 0.49086758494377136, 0.5251141786575317, 0.5479452013969421, 0.8458904027938843, 0.9680365324020386], [0.0251141544431448, 0.04566210135817528, 0.05707762390375137, 0.04794520512223244, 0.06164383515715599, 0.05707762390375137, 0.07305935770273209, 0.07990867644548416, 0.07077625393867493, 0.07077625393867493, 0.07534246891736984, 0.077625572681427, 0.11415524780750275, 0.11187214404344559, 0.0833333358168602, 0.10958904027938843, 0.09589041024446487, 0.15296803414821625, 0.18493150174617767, 0.15753424167633057, 0.09817351400852203, 0.11872146278619766, 0.13242009282112122, 0.17465753853321075, 0.0890410989522934, 0.16894976794719696, 0.17123287916183472, 0.1780821979045868, 0.155251145362854, 0.1278538852930069, 0.17351597547531128, 0.06621004641056061, 0.10273972898721695, 0.16438356041908264, 0.1666666716337204, 0.16894976794719696, 0.20319634675979614, 0.23059360682964325, 0.1438356190919876, 0.17579908668994904, 0.2168949842453003, 0.1826484054327011, 0.2442922443151474, 0.2283104956150055, 0.2168949842453003, 0.24657534062862396, 0.18607306480407715, 0.267123281955719, 0.22602739930152893, 0.1940639317035675, 0.2511415481567383, 0.2899543344974518, 0.29680365324020386, 0.2397260218858719, 0.25799086689949036, 0.29223743081092834, 0.31963470578193665, 0.28082191944122314, 0.3378995358943939, 0.3675799071788788, 0.27397260069847107, 0.37442922592163086, 0.4018264710903168, 0.34703195095062256, 0.31506848335266113, 0.35844749212265015, 0.4109589159488678, 0.3767123222351074, 0.33561643958091736, 0.3778538703918457, 0.43607306480407715, 0.4155251085758209, NaN, 0.4497717022895813, 0.41324201226234436, 0.5136986374855042, 0.5228310227394104, 0.8630136847496033], [0.015981735661625862, 0.029680365696549416, 0.0502283088862896, 0.04794520512223244, 0.054794520139694214, 0.054794520139694214, 0.05707762390375137, 0.06621004641056061, 0.06963470578193665, 0.06392694264650345, 0.06164383515715599, 0.04794520512223244, 0.077625572681427, 0.12100456655025482, 0.07305935770273209, 0.11415524780750275, 0.15981735289096832, 0.077625572681427, 0.13698630034923553, 0.1164383590221405, 0.13013698160648346, 0.1164383590221405, 0.15068493783473969, 0.09360730648040771, 0.12100456655025482, 0.12557077407836914, 0.10958904027938843, 0.11872146278619766, 0.2054794579744339, 0.15068493783473969, 0.16438356041908264, 0.07990867644548416, 0.13698630034923553, 0.21004566550254822, 0.12557077407836914, 0.17579908668994904, 0.16438356041908264, 0.14840182662010193, 0.16210044920444489, 0.19634702801704407, 0.16780821979045868, 0.1278538852930069, 0.16438356041908264, 0.17351597547531128, 0.16894976794719696, 0.15753424167633057, 0.23515981435775757, 0.16894976794719696, 0.2009132355451584, 0.23059360682964325, 0.232876718044281, 0.2168949842453003, 0.22374428808689117, 0.24657534062862396, 0.27397260069847107, 0.1278538852930069, 0.1940639317035675, 0.2442922443151474, 0.30136987566947937, 0.3767123222351074, 0.24885845184326172, 0.38127854466438293, 0.4063926935195923, 0.33561643958091736, 0.24885845184326172, 0.33561643958091736, 0.43150684237480164, 0.3721461296081543, 0.28538814187049866, 0.35844749212265015, 0.4931506812572479, 0.509132444858551, 0.5502282977104187, NaN, 0.5228310227394104, 0.49771690368652344, 0.6415525078773499, 0.956620991230011], [0.02283105067908764, 0.0251141544431448, 0.04794520512223244, 0.04337899386882782, 0.0388127863407135, 0.04337899386882782, 0.054794520139694214, 0.05936073139309883, 0.06392694264650345, 0.06392694264650345, 0.0388127863407135, 0.051369864493608475, 0.077625572681427, 0.10273972898721695, 0.07534246891736984, 0.10502283275127411, 0.14840182662010193, 0.04794520512223244, 0.10730593651533127, 0.10730593651533127, 0.13127854466438293, 0.10273972898721695, 0.11872146278619766, 0.07534246891736984, 0.09589041024446487, 0.1004566177725792, 0.07305935770273209, 0.08219178020954132, 0.20776255428791046, 0.15296803414821625, 0.15753424167633057, 0.06621004641056061, 0.13470318913459778, 0.2226027399301529, 0.07990867644548416, 0.16894976794719696, 0.155251145362854, 0.1438356190919876, 0.14611871540546417, 0.1826484054327011, 0.17123287916183472, 0.1004566177725792, 0.14840182662010193, 0.18493150174617767, 0.1666666716337204, 0.15981735289096832, 0.232876718044281, 0.1780821979045868, 0.19178082048892975, 0.20776255428791046, 0.21232876181602478, 0.2214611917734146, 0.22602739930152893, 0.23515981435775757, 0.30136987566947937, 0.16210044920444489, 0.1997716873884201, 0.2283104956150055, 0.26940637826919556, 0.34703195095062256, 0.23173515498638153, 0.4063926935195923, 0.4337899684906006, 0.33561643958091736, 0.1826484054327011, 0.3173516094684601, 0.4543378949165344, 0.38698631525039673, 0.26369863748550415, 0.34246575832366943, 0.4885844886302948, 0.47488585114479065, 0.586758017539978, 0.4771689474582672, NaN, 0.5045661926269531, 0.6347032189369202, 0.9406392574310303], [0.01141552533954382, 0.029680365696549416, 0.03652967885136604, 0.02054794505238533, 0.0388127863407135, 0.029680365696549416, 0.0388127863407135, 0.03652967885136604, 0.034246575087308884, 0.054794520139694214, 0.034246575087308884, 0.0388127863407135, 0.034246575087308884, 0.10502283275127411, 0.06392694264650345, 0.07534246891736984, 0.07990867644548416, 0.054794520139694214, 0.07990867644548416, 0.04337899386882782, 0.09132420271635056, 0.06621004641056061, 0.052511416375637054, 0.05707762390375137, 0.09132420271635056, 0.10730593651533127, 0.11415524780750275, 0.10502283275127411, 0.13013698160648346, 0.1164383590221405, 0.10730593651533127, 0.08675798773765564, 0.10502283275127411, 0.15068493783473969, 0.12328767031431198, 0.16894976794719696, 0.14155250787734985, 0.1164383590221405, 0.06164383515715599, 0.09360730648040771, 0.14611871540546417, 0.1278538852930069, 0.13698630034923553, 0.14726027846336365, 0.09132420271635056, 0.13242009282112122, 0.22602739930152893, 0.14840182662010193, 0.18949772417545319, 0.1666666716337204, 0.1392694115638733, 0.29680365324020386, 0.2990867495536804, 0.15296803414821625, 0.20776255428791046, 0.27397260069847107, 0.2945205569267273, 0.29223743081092834, 0.35844749212265015, 0.30136987566947937, 0.267123281955719, 0.3767123222351074, 0.37899544835090637, 0.27397260069847107, 0.3721461296081543, 0.20776255428791046, 0.38127854466438293, 0.28538814187049866, 0.388127863407135, 0.2659817337989807, 0.4440639317035675, 0.45205479860305786, 0.48630136251449585, 0.5022830963134766, 0.4954337775707245, NaN, 0.5525113940238953, 0.8926940560340881], [0.027397260069847107, 0.03652967885136604, 0.054794520139694214, 0.04109589010477066, 0.052511416375637054, 0.054794520139694214, 0.06849315017461777, 0.06621004641056061, 0.04794520512223244, 0.08219178020954132, 0.07534246891736984, 0.06164383515715599, 0.09132420271635056, 0.14840182662010193, 0.09132420271635056, 0.06164383515715599, 0.1335616409778595, 0.07876712083816528, 0.0502283088862896, 0.1164383590221405, 0.14611871540546417, 0.10273972898721695, 0.1278538852930069, 0.09132420271635056, 0.11986301094293594, 0.08675798773765564, 0.10502283275127411, 0.09589041024446487, 0.1666666716337204, 0.1438356190919876, 0.1278538852930069, 0.11872146278619766, 0.10730593651533127, 0.19178082048892975, 0.11757990717887878, 0.12100456655025482, 0.13698630034923553, 0.10159817337989807, 0.12328767031431198, 0.12557077407836914, 0.10502283275127411, 0.12100456655025482, 0.10958904027938843, 0.1666666716337204, 0.1278538852930069, 0.1004566177725792, 0.24200913310050964, 0.09132420271635056, 0.14840182662010193, 0.17009133100509644, 0.15981735289096832, 0.14840182662010193, 0.15981735289096832, 0.18949772417545319, 0.28538814187049866, 0.15753424167633057, 0.15068493783473969, 0.21461187303066254, 0.1780821979045868, 0.3059360682964325, 0.2054794579744339, 0.27397260069847107, 0.3219178020954132, 0.2214611917734146, 0.18493150174617767, 0.29680365324020386, 0.34018266201019287, 0.2602739632129669, 0.19178082048892975, 0.33561643958091736, 0.4063926935195923, 0.15410958230495453, 0.4771689474582672, 0.35844749212265015, 0.3652968108654022, 0.44748857617378235, NaN, 0.9497717022895813], [0.006849315017461777, 0.004566209856420755, 0.006849315017461777, 0.004566209856420755, 0.00913241971284151, 0.006849315017461777, 0.00913241971284151, 0.00913241971284151, 0.006849315017461777, 0.004566209856420755, 0.006849315017461777, 0.00913241971284151, 0.01826483942568302, 0.034246575087308884, 0.006849315017461777, 0.006849315017461777, 0.01141552533954382, 0.00913241971284151, 0.015981735661625862, 0.027397260069847107, 0.006849315017461777, 0.01141552533954382, 0.02054794505238533, 0.006849315017461777, 0.0251141544431448, 0.013698630034923553, 0.00913241971284151, 0.00913241971284151, 0.02283105067908764, 0.02283105067908764, 0.00913241971284151, 0.01141552533954382, 0.01141552533954382, 0.015981735661625862, 0.02054794505238533, 0.02054794505238533, 0.028538811951875687, 0.02283105067908764, 0.0251141544431448, 0.027397260069847107, 0.01826483942568302, 0.0251141544431448, 0.031963471323251724, 0.04109589010477066, 0.0251141544431448, 0.02283105067908764, 0.06849315017461777, 0.0251141544431448, 0.034246575087308884, 0.04109589010477066, 0.034246575087308884, 0.06849315017461777, 0.06849315017461777, 0.0388127863407135, 0.0388127863407135, 0.01141552533954382, 0.013698630034923553, 0.06164383515715599, 0.06392694264650345, 0.10730593651533127, 0.04223744198679924, 0.077625572681427, 0.10730593651533127, 0.05936073139309883, 0.03652967885136604, 0.05707762390375137, 0.1164383590221405, 0.05707762390375137, 0.034246575087308884, 0.08447488397359848, 0.1666666716337204, 0.031963471323251724, 0.13698630034923553, 0.04337899386882782, 0.05936073139309883, 0.10730593651533127, 0.0502283088862896, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.910173156},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8433997492},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8296269951},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8238747551},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.7918075095},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.775840595},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7696435974},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7503706337},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7074067506},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6964656355},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6781414919},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6737680166},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6712180477},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6404109592},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6399217228},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6384243599},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.6316640004},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6293808921},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6273942946},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6165124841},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6057492734},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5988258285},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.5899157901},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5861353272},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5821176526},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5754610672},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5728814556},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5660766175},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5580857499},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5562177562},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5553727094},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5517997979},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5497983744},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5441499145},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5364407283},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5317262641},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5291170024},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5256923448},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5220156561},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5168564329},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5142619934},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5061969996},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4987398443},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4960564521},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4908082759},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4894295195},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4798671645},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4785180581},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4752120048},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4698155711},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4478147405},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4426258669},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.4354355707},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4322332917},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.4198689434},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.381100634},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3663938799},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3641849006},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3567573972},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.3496857014},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3380329728},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.3359129455},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.3265284949},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.316536203},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3116734866},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3053282347},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.301206785},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2979155538},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.288797961},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2772934827},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.2481913062},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2358862602},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2219652485},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.210282868},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2016989868},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1841457631},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1691128513},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0346023839}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_binary/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..6cb6487075a258f604fc0e63bd64be3d55dbdc35
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_all/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),TabDPT-Turbo (default),"AutoGluon 1.4 (best, 4h)",RealMLP (tuned + ensembled),TabICL (default),TabM (tuned + ensembled),LimiX (default),TabDPT (tuned + ensembled),TabSwift (default),Mitra (default),LightGBM (tuned + ensembled),BetaTabPFN (default),TabM (tuned),TabPFNv2 (tuned + ensembled),TabDPT (tuned),RealMLP (tuned),LightGBM (tuned),iLTM (tuned + ensembled),CatBoost (default),CatBoost (tuned + ensembled),CatBoost (tuned),TabPFNv2 (default),ModernNCA (tuned),TabDPT (default),xRFM (tuned + ensembled),xRFM (tuned),TabPFNv2 (tuned),XGBoost (tuned + ensembled),SAP-RPT-OSS (default),ChimeraBoost (tuned + ensembled),EBM (tuned + ensembled),FastaiMLP (tuned + ensembled),TorchMLP (tuned + ensembled),ModernNCA (default),XGBoost (tuned),TabM (default),ChimeraBoost (tuned),RealMLP (default),EBM (tuned),ModernNCA (tuned + ensembled),EBM (default),ChimeraBoost (default),iLTM (tuned),FastaiMLP (tuned),TabSTAR (tuned),TabSTAR (tuned + ensembled),TorchMLP (tuned),OrionMSP (default),ExtraTrees (tuned + ensembled),ExtraTrees (tuned),XGBoost (default),iLTM (default),TabFlex (default),LightGBM (default),TabSTAR (default),Linear (tuned + ensembled),TorchMLP (default),RandomForest (tuned + ensembled),PerpetualBooster (tuned + ensembled),Linear (tuned),FastaiMLP (default),RandomForest (tuned),PerpetualBooster (tuned),Linear (default),KNN (tuned + ensembled),xRFM (default),ExtraTrees (default),RandomForest (default),PerpetualBooster (default),KNN (tuned),KNN (default)
+TabFM (default),,0.79680365,0.75570774,0.8105023,0.8196347,0.8150685,0.82990867,0.8356164,0.83789957,0.8561644,0.8767123,0.9223744,0.9063927,0.90525115,0.89726025,0.86757994,0.8561644,0.88812786,0.8664383,0.93835616,0.89726025,0.89269406,0.8721461,0.9109589,0.9041096,0.89497715,0.89269406,0.9018265,0.8310502,0.913242,0.9018265,0.9041096,0.8835617,0.8744292,0.93607306,0.89041096,0.91780823,0.9086758,0.89726025,0.9543379,0.89726025,0.94520545,0.93607306,0.9292237,0.913242,0.91780823,0.94063926,0.9109589,0.93835616,0.9269406,0.9269406,0.8196347,0.82191783,0.96803653,0.94292235,0.9520548,0.94292235,0.9703196,0.94063926,0.9269406,0.9726027,0.8538813,0.9246575,0.977169,0.977169,0.98173517,0.93378997,0.956621,0.9748858,0.98401827,0.9543379,0.9726027,0.9748858,0.98401827,0.977169,0.98858446,0.9726027,0.9931507
+TabPFN-3 (default),0.20319635,,0.5958904,0.51141554,0.65525115,0.70319635,0.6940639,0.7191781,0.7351598,0.7191781,0.78995436,0.81164384,0.79223746,0.77397263,0.8196347,0.8310502,0.7808219,0.7853881,0.772831,0.83789957,0.79680365,0.84931505,0.8105023,0.84246576,0.84931505,0.8310502,0.8242009,0.8333333,0.7796804,0.8607306,0.84703195,0.84246576,0.8150685,0.8504566,0.8481735,0.8287671,0.8744292,0.84474885,0.8413242,0.88812786,0.8504566,0.8721461,0.89041096,0.8789954,0.8812785,0.8630137,0.9109589,0.86415523,0.86986303,0.90525115,0.8732877,0.78424656,0.79680365,0.93378997,0.9018265,0.93835616,0.93378997,0.94063926,0.9292237,0.9041096,0.9543379,0.8310502,0.8789954,0.9474886,0.93835616,0.93607306,0.89041096,0.94063926,0.96118724,0.9520548,0.9315069,0.93835616,0.9543379,0.9703196,0.9748858,0.9703196,0.96347034,0.9954338
+"AutoGluon 1.5 (extreme, 4h)",0.24429224,0.4041096,,0.456621,0.6484018,0.63013697,0.65068495,0.6643836,0.60730594,0.7762557,0.81621003,0.78196347,0.79908675,0.74657536,0.7762557,0.7671233,0.69063926,0.8310502,0.7579909,0.8184931,0.74885845,0.79452056,0.84703195,0.84018266,0.8744292,0.8436073,0.8538813,0.8436073,0.7420091,0.86986303,0.7762557,0.86757994,0.8230594,0.78995436,0.8789954,0.80365294,0.88812786,0.84018266,0.8561644,0.8995434,0.8356164,0.89269406,0.8789954,0.89726025,0.8538813,0.84703195,0.93378997,0.8561644,0.9018265,0.89269406,0.8652968,0.74657536,0.74885845,0.92009133,0.90525115,0.9246575,0.9086758,0.93378997,0.9223744,0.8812785,0.9474886,0.80136985,0.8744292,0.9474886,0.94292235,0.93607306,0.9018265,0.93835616,0.94520545,0.95091325,0.9257991,0.9269406,0.94292235,0.9497717,0.9520548,0.96347034,0.94520545,0.9931507
+TabICLv2 (default),0.18949772,0.4885845,0.543379,,0.6004566,0.6312785,0.6461187,0.6643836,0.7191781,0.7191781,0.74429226,0.8025114,0.7579909,0.78424656,0.81392694,0.8082192,0.71347034,0.7648402,0.77397263,0.79452056,0.78995436,0.8127854,0.7716895,0.80365294,0.8150685,0.79680365,0.7853881,0.79452056,0.7625571,0.8333333,0.84018266,0.8173516,0.7910959,0.80479455,0.8105023,0.80136985,0.8242009,0.8196347,0.84018266,0.8789954,0.8504566,0.8287671,0.8607306,0.84246576,0.86986303,0.84474885,0.86986303,0.8561644,0.83789957,0.8584475,0.8812785,0.79223746,0.78995436,0.91780823,0.9246575,0.91210043,0.9109589,0.913242,0.9086758,0.89726025,0.94063926,0.8356164,0.89497715,0.956621,0.93835616,0.94292235,0.9086758,0.9315069,0.93607306,0.9520548,0.9372146,0.9474886,0.9520548,0.9520548,0.956621,0.9794521,0.9589041,0.9954338
+TabPFN-2.6 (default),0.1803653,0.34474885,0.35159817,0.39954337,,0.5456621,0.53424656,0.5673516,0.60502285,0.65753424,0.7214612,0.7191781,0.7214612,0.70547944,0.7214612,0.7146119,0.6826484,0.75570774,0.7214612,0.7785388,0.7191781,0.7305936,0.783105,0.8127854,0.79452056,0.77397263,0.7808219,0.7671233,0.71575344,0.8242009,0.7191781,0.80365294,0.783105,0.8059361,0.7853881,0.8059361,0.84931505,0.80136985,0.78995436,0.8538813,0.84018266,0.8310502,0.8458904,0.8630137,0.8561644,0.8390411,0.8812785,0.8242009,0.8584475,0.8584475,0.84246576,0.74429226,0.75570774,0.913242,0.8515982,0.913242,0.90525115,0.93607306,0.8995434,0.8789954,0.9474886,0.78995436,0.86757994,0.9497717,0.93607306,0.9315069,0.89269406,0.9109589,0.9543379,0.9315069,0.92009133,0.9269406,0.93835616,0.94520545,0.96118724,0.96118724,0.9474886,0.99086756
+RealTabPFN-2.5 (default),0.1849315,0.29680365,0.369863,0.36872146,0.4543379,,0.48173517,0.5251142,0.53881276,0.65068495,0.66210043,0.6643836,0.67808217,0.6849315,0.6826484,0.75342464,0.70091325,0.7214612,0.7294521,0.7328767,0.7351598,0.71575344,0.74429226,0.7648402,0.7237443,0.7625571,0.7351598,0.7351598,0.7671233,0.75570774,0.739726,0.79452056,0.7876712,0.79680365,0.760274,0.80136985,0.79908675,0.79452056,0.8059361,0.826484,0.8059361,0.79908675,0.8082192,0.8150685,0.826484,0.8356164,0.783105,0.8196347,0.8333333,0.8458904,0.8538813,0.7785388,0.78196347,0.89726025,0.913242,0.89269406,0.89726025,0.88812786,0.89269406,0.90296805,0.94063926,0.8127854,0.8538813,0.96118724,0.89726025,0.9315069,0.8652968,0.9086758,0.93378997,0.94520545,0.9292237,0.9086758,0.94292235,0.94520545,0.956621,0.9703196,0.94520545,0.9931507
+RealTabPFN-2.5 (tuned + ensembled),0.17009133,0.30593607,0.34931508,0.35388127,0.46575344,0.51826483,,0.6255708,0.53196347,0.65981734,0.6712329,0.66780823,0.6712329,0.70319635,0.6666667,0.70091325,0.70776254,0.72602737,0.7237443,0.7283105,0.75342464,0.716895,0.72602737,0.76598173,0.73401827,0.75570774,0.7420091,0.74657536,0.760274,0.80136985,0.673516,0.7785388,0.7625571,0.8105023,0.7625571,0.772831,0.78652966,0.7785388,0.7773973,0.82191783,0.79452056,0.793379,0.79908675,0.8059361,0.8105023,0.80936074,0.79908675,0.8127854,0.8242009,0.8196347,0.84703195,0.74657536,0.75342464,0.8789954,0.8767123,0.8835617,0.8767123,0.88812786,0.8812785,0.8744292,0.93378997,0.78995436,0.8538813,0.9246575,0.913242,0.9315069,0.8744292,0.913242,0.9246575,0.93835616,0.90525115,0.9109589,0.9269406,0.94292235,0.94520545,0.96118724,0.9315069,0.99086756
+RealTabPFN-2.5 (tuned),0.16438356,0.28082192,0.33561644,0.33561644,0.4326484,0.47488585,0.37442923,,0.522831,0.6278539,0.63242006,0.6347032,0.6484018,0.66210043,0.65981734,0.68607306,0.69634706,0.70547944,0.70319635,0.71118724,0.71803653,0.7043379,0.70776254,0.760274,0.70547944,0.7283105,0.7283105,0.7237443,0.7374429,0.74657536,0.6757991,0.7522831,0.74429226,0.78196347,0.74429226,0.760274,0.75913244,0.7716895,0.7579909,0.8082192,0.7694064,0.7671233,0.7853881,0.79223746,0.79680365,0.7876712,0.78881276,0.79452056,0.8059361,0.80365294,0.8310502,0.7374429,0.74885845,0.8767123,0.8721461,0.8687215,0.8721461,0.8812785,0.86986303,0.8607306,0.9223744,0.79223746,0.8333333,0.9223744,0.88812786,0.91210043,0.8584475,0.8995434,0.913242,0.9246575,0.90981734,0.9041096,0.92009133,0.93378997,0.94063926,0.96347034,0.93378997,0.99086756
+TabDPT-Turbo (default),0.16210045,0.2648402,0.39269406,0.28082192,0.39497718,0.4611872,0.46803653,0.47716895,,0.53424656,0.5730594,0.5502283,0.54223746,0.6232877,0.65068495,0.68835616,0.5753425,0.5890411,0.5730594,0.60730594,0.61415523,0.70776254,0.66780823,0.6232877,0.65296805,0.6461187,0.6689498,0.6643836,0.6461187,0.65068495,0.80936074,0.6940639,0.71004564,0.65981734,0.6803653,0.6826484,0.6849315,0.6940639,0.7374429,0.7283105,0.72602737,0.69863015,0.69863015,0.7214612,0.75114155,0.7203196,0.6803653,0.7374429,0.7351598,0.75913244,0.79452056,0.70776254,0.716895,0.79452056,0.84018266,0.8458904,0.8789954,0.8173516,0.79908675,0.8721461,0.8607306,0.78995436,0.8082192,0.8789954,0.8767123,0.8812785,0.8287671,0.9063927,0.89041096,0.9041096,0.9063927,0.9041096,0.9292237,0.9303653,0.93607306,0.96575344,0.9520548,0.9931507
+"AutoGluon 1.4 (best, 4h)",0.14383562,0.28082192,0.22374429,0.28082192,0.34246576,0.34931508,0.34018266,0.37214613,0.46575344,,0.59703195,0.5,0.55936074,0.48173517,0.543379,0.5171233,0.47488585,0.6826484,0.58447486,0.6255708,0.47488585,0.5684931,0.69634706,0.7214612,0.7123288,0.71004564,0.7237443,0.7328767,0.56164384,0.65753424,0.60502285,0.69520545,0.70776254,0.56164384,0.783105,0.6164383,0.7454338,0.7283105,0.7351598,0.783105,0.70319635,0.783105,0.74429226,0.7785388,0.7762557,0.7420091,0.7237443,0.7648402,0.79452056,0.79452056,0.7785388,0.70319635,0.7123288,0.8287671,0.6974886,0.79908675,0.8082192,0.89726025,0.84474885,0.7876712,0.9246575,0.760274,0.836758,0.88584477,0.89269406,0.91780823,0.86415523,0.9086758,0.90296805,0.92009133,0.89269406,0.913242,0.9292237,0.93607306,0.93607306,0.94520545,0.91780823,0.9954338
+RealMLP (tuned + ensembled),0.12328767,0.21004567,0.18378995,0.25570777,0.27853882,0.33789954,0.32876712,0.3675799,0.42694065,0.40296805,,0.4885845,0.5045662,0.4634703,0.50114155,0.47945204,0.47260273,0.6187215,0.5502283,0.57990867,0.49200913,0.54908675,0.7305936,0.68607306,0.6757991,0.65296805,0.6803653,0.6826484,0.53652966,0.6415525,0.5936073,0.6872146,0.66210043,0.55936074,0.7328767,0.6255708,0.70547944,0.7351598,0.7477169,0.7283105,0.6757991,0.77397263,0.7420091,0.74086756,0.8253425,0.75570774,0.6849315,0.7568493,0.7420091,0.7579909,0.772831,0.6940639,0.70547944,0.7876712,0.71004564,0.83789957,0.8276256,0.8721461,0.8561644,0.7648402,0.9018265,0.74885845,0.826484,0.89269406,0.9109589,0.913242,0.8310502,0.9109589,0.9292237,0.9246575,0.8515982,0.9041096,0.9246575,0.93835616,0.96118724,0.96575344,0.9246575,0.9931507
+TabICL (default),0.07762557,0.18835616,0.21803653,0.19748859,0.28082192,0.33561644,0.3321918,0.3652968,0.4497717,0.5,0.51141554,,0.51141554,0.5045662,0.5559361,0.55936074,0.52397263,0.5730594,0.55479455,0.58219177,0.56621003,0.6061644,0.60730594,0.6255708,0.6495434,0.6438356,0.6278539,0.6347032,0.6187215,0.65981734,0.64726025,0.6803653,0.6461187,0.6210046,0.6917808,0.673516,0.69634706,0.673516,0.69634706,0.7123288,0.71575344,0.716895,0.6917808,0.7283105,0.7283105,0.6894977,0.7351598,0.69863015,0.739726,0.75342464,0.7420091,0.70319635,0.70776254,0.77397263,0.79680365,0.8333333,0.8276256,0.8515982,0.8105023,0.8127854,0.8984018,0.77397263,0.7808219,0.86986303,0.86757994,0.89041096,0.79908675,0.86415523,0.89726025,0.9086758,0.8755708,0.8835617,0.9223744,0.9520548,0.94863015,0.96118724,0.93835616,0.99086756
+TabM (tuned + ensembled),0.09360731,0.20776255,0.20091324,0.24200913,0.27853882,0.3219178,0.32876712,0.35159817,0.45776257,0.44063926,0.49543378,0.4885845,,0.4634703,0.5376712,0.5456621,0.43607306,0.61187214,0.5913242,0.6746575,0.4885845,0.56392694,0.6347032,0.66552514,0.6267123,0.6666667,0.65296805,0.65753424,0.55365294,0.6609589,0.5890411,0.60502285,0.58219177,0.5456621,0.69634706,0.63013697,0.7146119,0.7237443,0.673516,0.6894977,0.6940639,0.75570774,0.82990867,0.7671233,0.76826483,0.7671233,0.70776254,0.7568493,0.7853881,0.74429226,0.7351598,0.70547944,0.71004564,0.78652966,0.6940639,0.8173516,0.79908675,0.8538813,0.84931505,0.7671233,0.90525115,0.75342464,0.8333333,0.89041096,0.8652968,0.88812786,0.8481735,0.8721461,0.88584477,0.9109589,0.8789954,0.8812785,0.88584477,0.9223744,0.9223744,0.96575344,0.9086758,0.98173517
+LimiX (default),0.09474886,0.2260274,0.25342464,0.21575342,0.29452056,0.31506848,0.29680365,0.33789954,0.37671232,0.51826483,0.53652966,0.49543378,0.53652966,,0.49429223,0.47488585,0.53995436,0.60273975,0.51826483,0.54908675,0.5776256,0.5068493,0.59474885,0.6210046,0.69520545,0.63013697,0.6415525,0.6392694,0.6164383,0.6278539,0.56164384,0.6347032,0.6130137,0.6347032,0.6803653,0.6369863,0.6917808,0.6689498,0.63812786,0.6826484,0.6461187,0.70319635,0.65068495,0.7123288,0.66210043,0.6849315,0.6917808,0.6803653,0.7191781,0.70091325,0.71118724,0.63812786,0.6369863,0.739726,0.727169,0.7579909,0.75570774,0.7762557,0.7694064,0.7146119,0.79680365,0.7043379,0.7716895,0.7808219,0.80365294,0.82191783,0.79680365,0.8287671,0.8242009,0.84474885,0.80365294,0.8082192,0.88812786,0.8789954,0.89726025,0.89497715,0.8515982,0.96575344
+TabDPT (tuned + ensembled),0.10273973,0.1803653,0.22374429,0.18607306,0.27853882,0.3173516,0.33333334,0.34018266,0.34931508,0.456621,0.49885845,0.44406393,0.46232876,0.50570774,,0.5171233,0.5068493,0.5251142,0.49543378,0.5262557,0.5410959,0.6392694,0.59018266,0.5753425,0.5981735,0.58219177,0.5936073,0.5913242,0.56392694,0.60273975,0.6484018,0.6289954,0.61757994,0.6312785,0.6015982,0.6415525,0.6232877,0.6221461,0.6643836,0.6815069,0.663242,0.65296805,0.6289954,0.65753424,0.66210043,0.6518265,0.6849315,0.6461187,0.6894977,0.7237443,0.7351598,0.65525115,0.65753424,0.7374429,0.783105,0.79452056,0.80136985,0.7808219,0.80136985,0.79680365,0.8310502,0.760274,0.75114155,0.84018266,0.8561644,0.8584475,0.7716895,0.8721461,0.8561644,0.8732877,0.83789957,0.8664383,0.9166667,0.9269406,0.9246575,0.93607306,0.9086758,0.9931507
+TabSwift (default),0.1324201,0.16894977,0.23287672,0.19178082,0.28538814,0.24657534,0.29908675,0.31392694,0.31164384,0.48287672,0.5205479,0.44063926,0.4543379,0.5251142,0.48287672,,0.49086758,0.522831,0.49543378,0.50913244,0.5456621,0.543379,0.5958904,0.5502283,0.5981735,0.56392694,0.5753425,0.59018266,0.5981735,0.6210046,0.61415523,0.65068495,0.6164383,0.61187214,0.61187214,0.6232877,0.6278539,0.63242006,0.70547944,0.65981734,0.65525115,0.6484018,0.6392694,0.6666667,0.65753424,0.65525115,0.65981734,0.65525115,0.70319635,0.7283105,0.74429226,0.70776254,0.7191781,0.7191781,0.84474885,0.80136985,0.79680365,0.760274,0.7785388,0.88812786,0.78881276,0.7876712,0.78995436,0.8607306,0.8059361,0.8652968,0.8150685,0.84703195,0.8242009,0.8652968,0.85730594,0.8515982,0.89041096,0.88584477,0.89497715,0.9246575,0.93835616,0.9931507
+Mitra (default),0.14383562,0.21917808,0.30936074,0.2865297,0.3173516,0.29908675,0.29223743,0.30365297,0.42465752,0.5251142,0.5273973,0.4760274,0.56392694,0.46004567,0.49315068,0.50913244,,0.55707765,0.5479452,0.6369863,0.54908675,0.5045662,0.61415523,0.60502285,0.6152968,0.6415525,0.6164383,0.6084475,0.60273975,0.6164383,0.586758,0.60958904,0.6152968,0.60273975,0.6495434,0.5958904,0.6894977,0.6712329,0.65068495,0.6757991,0.6232877,0.65525115,0.65753424,0.6940639,0.65296805,0.6849315,0.6666667,0.6849315,0.70776254,0.6894977,0.69634706,0.6187215,0.6210046,0.74429226,0.69292235,0.69863015,0.706621,0.7625571,0.75342464,0.77397263,0.74885845,0.6803653,0.7648402,0.7808219,0.7351598,0.783105,0.79908675,0.80136985,0.76369864,0.8059361,0.8333333,0.7716895,0.9041096,0.84018266,0.8515982,0.92009133,0.8664383,0.98858446
+LightGBM (tuned + ensembled),0.111872144,0.21461187,0.16894977,0.23515981,0.24429224,0.27853882,0.2739726,0.29452056,0.41095892,0.3173516,0.38127854,0.42694065,0.38812786,0.39726028,0.47488585,0.47716895,0.44292238,,0.44520548,0.47146118,0.45776257,0.5205479,0.51141554,0.6461187,0.54908675,0.6244292,0.6210046,0.6244292,0.4885845,0.5525114,0.5936073,0.5913242,0.60958904,0.5068493,0.7351598,0.5890411,0.6061644,0.6541096,0.6004566,0.60730594,0.61415523,0.7853881,0.69520545,0.66210043,0.66210043,0.69634706,0.6210046,0.75342464,0.74657536,0.6940639,0.6872146,0.6815069,0.6849315,0.72602737,0.6757991,0.80707765,0.78995436,0.86986303,0.84931505,0.70776254,0.9497717,0.760274,0.8173516,0.8105023,0.8995434,0.8744292,0.8310502,0.7853881,0.9223744,0.89726025,0.8812785,0.8995434,0.84703195,0.9223744,0.9520548,0.94520545,0.9212329,0.99086756
+BetaTabPFN (default),0.13356164,0.22716895,0.24200913,0.2260274,0.27853882,0.27054796,0.2762557,0.29680365,0.42694065,0.4155251,0.4497717,0.44520548,0.4086758,0.48173517,0.5045662,0.5045662,0.4520548,0.55479455,,0.49543378,0.48059362,0.55936074,0.5205479,0.56164384,0.5479452,0.60502285,0.5730594,0.59703195,0.543379,0.5833333,0.5981735,0.58447486,0.56392694,0.5205479,0.6347032,0.60273975,0.6289954,0.65525115,0.6130137,0.6278539,0.6392694,0.65525115,0.6803653,0.65525115,0.65525115,0.67808217,0.5958904,0.7305936,0.6894977,0.6438356,0.70319635,0.72602737,0.7328767,0.7123288,0.6940639,0.79223746,0.79223746,0.7648402,0.8127854,0.76141554,0.8184931,0.78881276,0.8310502,0.8105023,0.84246576,0.8287671,0.8561644,0.79223746,0.84246576,0.84474885,0.9109589,0.9109589,0.8150685,0.8630137,0.89269406,0.92009133,0.9497717,0.98401827
+TabM (tuned),0.061643835,0.16210045,0.18150684,0.20547946,0.22146119,0.26712328,0.2716895,0.2888128,0.39269406,0.37442923,0.42009133,0.4178082,0.32534248,0.45091325,0.4737443,0.49086758,0.36301368,0.5285388,0.5045662,,0.4372146,0.51141554,0.53424656,0.5707763,0.5525114,0.6004566,0.60502285,0.60958904,0.5251142,0.6278539,0.55136985,0.53196347,0.53652966,0.48401827,0.6267123,0.55479455,0.6666667,0.61187214,0.6244292,0.6461187,0.65068495,0.65525115,0.71004564,0.6826484,0.7146119,0.6803653,0.65296805,0.69634706,0.71575344,0.6712329,0.70319635,0.6872146,0.69063926,0.73401827,0.6666667,0.75114155,0.7579909,0.8082192,0.8150685,0.75114155,0.8356164,0.7328767,0.80136985,0.8744292,0.8333333,0.8835617,0.8173516,0.8242009,0.8310502,0.89726025,0.84931505,0.8538813,0.84246576,0.8835617,0.89269406,0.956621,0.8835617,0.9726027
+TabPFNv2 (tuned + ensembled),0.10273973,0.20319635,0.25114155,0.21004567,0.28082192,0.2648402,0.24657534,0.28196347,0.38584474,0.5251142,0.5079909,0.43378997,0.51141554,0.42237443,0.45890412,0.4543379,0.45091325,0.54223746,0.5194064,0.5627854,,0.46917808,0.5833333,0.56164384,0.6004566,0.5936073,0.58447486,0.5833333,0.5936073,0.61415523,0.53652966,0.60958904,0.6130137,0.6152968,0.6244292,0.5308219,0.65068495,0.6335617,0.6061644,0.6278539,0.59931505,0.6392694,0.61187214,0.66210043,0.6347032,0.65068495,0.6803653,0.6484018,0.67808217,0.65525115,0.64041096,0.5958904,0.6004566,0.70547944,0.6643836,0.68835616,0.6940639,0.75114155,0.716895,0.7328767,0.75570774,0.6757991,0.7351598,0.75570774,0.74885845,0.7762557,0.7625571,0.79452056,0.7694064,0.8082192,0.7876712,0.7785388,0.9018265,0.86986303,0.8687215,0.9086758,0.8538813,0.9931507
+TabDPT (tuned),0.10730594,0.15068494,0.20547946,0.18721461,0.26940638,0.28424656,0.28310502,0.2956621,0.29223743,0.43150684,0.45091325,0.3938356,0.43607306,0.49315068,0.3607306,0.456621,0.49543378,0.47945204,0.44063926,0.4885845,0.5308219,,0.51369864,0.51369864,0.53652966,0.5479452,0.55479455,0.55136985,0.5376712,0.53881276,0.6004566,0.586758,0.5502283,0.59931505,0.5673516,0.5707763,0.58447486,0.5958904,0.61415523,0.6347032,0.61187214,0.5890411,0.5776256,0.60502285,0.6255708,0.6278539,0.6267123,0.6278539,0.6232877,0.65525115,0.6940639,0.6038813,0.60502285,0.69634706,0.7785388,0.739726,0.760274,0.7123288,0.7351598,0.7785388,0.75114155,0.7374429,0.74086756,0.8105023,0.7853881,0.83789957,0.7716895,0.84474885,0.8196347,0.84703195,0.8196347,0.8561644,0.8812785,0.8835617,0.89726025,0.93378997,0.89726025,0.98858446
+RealMLP (tuned),0.12785389,0.18949772,0.15296803,0.2283105,0.21689498,0.25570777,0.2739726,0.29223743,0.3321918,0.30365297,0.26940638,0.39269406,0.3652968,0.40525115,0.40981734,0.4041096,0.38584474,0.4885845,0.47945204,0.46575344,0.41666666,0.48630136,,0.53881276,0.49429223,0.543379,0.51369864,0.522831,0.49086758,0.522831,0.522831,0.53881276,0.5559361,0.48173517,0.5936073,0.5479452,0.55707765,0.61415523,0.63242006,0.63242006,0.5981735,0.6495434,0.6392694,0.60273975,0.6940639,0.65981734,0.58219177,0.6278539,0.65068495,0.6438356,0.71004564,0.6712329,0.6826484,0.7237443,0.6335617,0.7420091,0.75342464,0.7762557,0.79223746,0.74086756,0.8105023,0.7351598,0.75570774,0.84474885,0.82191783,0.84703195,0.76598173,0.8755708,0.83789957,0.8515982,0.8150685,0.84474885,0.86757994,0.84931505,0.8812785,0.9474886,0.8721461,0.9794521
+LightGBM (tuned),0.0890411,0.15753424,0.15981735,0.19634703,0.18721461,0.23515981,0.23401827,0.23972602,0.37671232,0.27853882,0.31392694,0.37442923,0.3344749,0.37899545,0.42465752,0.4497717,0.39497718,0.35388127,0.43835616,0.42922375,0.43835616,0.48630136,0.4611872,,0.48173517,0.5605023,0.5582192,0.5479452,0.47716895,0.4885845,0.5410959,0.5559361,0.5502283,0.49086758,0.6244292,0.5410959,0.5502283,0.6210046,0.5696347,0.55707765,0.586758,0.6849315,0.65068495,0.58447486,0.64041096,0.6484018,0.58447486,0.6872146,0.6894977,0.63242006,0.6210046,0.6643836,0.6689498,0.66552514,0.65753424,0.7568493,0.76598173,0.8184931,0.8207763,0.6917808,0.9143836,0.7420091,0.783105,0.79223746,0.84018266,0.8504566,0.80707765,0.7648402,0.8744292,0.8721461,0.8584475,0.8630137,0.8253425,0.9063927,0.9246575,0.94292235,0.9086758,0.9931507
+iLTM (tuned + ensembled),0.09589041,0.15068494,0.12557077,0.1849315,0.20547946,0.2762557,0.26598173,0.29452056,0.34703195,0.28767124,0.3242009,0.35045663,0.37328768,0.30479452,0.40182647,0.40182647,0.3847032,0.45091325,0.4520548,0.44748858,0.39954337,0.4634703,0.50570774,0.51826483,,0.4977169,0.47716895,0.5068493,0.4977169,0.53424656,0.5205479,0.5742009,0.59931505,0.51826483,0.5650685,0.5502283,0.5753425,0.5981735,0.63013697,0.61415523,0.60273975,0.6347032,0.58219177,0.5913242,0.6643836,0.63242006,0.60502285,0.6392694,0.63584477,0.71118724,0.68835616,0.6347032,0.63812786,0.70547944,0.6267123,0.7294521,0.74657536,0.8105023,0.78995436,0.7351598,0.8584475,0.7191781,0.7191781,0.8173516,0.84018266,0.8356164,0.74429226,0.8150685,0.8607306,0.8561644,0.7808219,0.84246576,0.9109589,0.8789954,0.9041096,0.9086758,0.88013697,0.9748858
+CatBoost (default),0.10502283,0.16894977,0.1563927,0.20319635,0.2260274,0.23744293,0.24429224,0.2716895,0.35388127,0.28995433,0.34703195,0.3561644,0.33333334,0.369863,0.4178082,0.43607306,0.3584475,0.37557077,0.39497718,0.39954337,0.4063927,0.4520548,0.456621,0.4394977,0.5022831,,0.49315068,0.49429223,0.46689498,0.5068493,0.51369864,0.5376712,0.55479455,0.4520548,0.5696347,0.56621003,0.5650685,0.5890411,0.56392694,0.5936073,0.5456621,0.6130137,0.6187215,0.6255708,0.6210046,0.6232877,0.586758,0.65525115,0.6712329,0.6643836,0.6484018,0.6723744,0.67808217,0.70091325,0.6004566,0.7716895,0.7648402,0.8173516,0.8310502,0.65753424,0.8515982,0.7283105,0.75570774,0.79908675,0.8767123,0.7853881,0.7762557,0.7876712,0.8812785,0.8242009,0.8242009,0.88584477,0.8310502,0.8744292,0.8995434,0.89269406,0.913242,0.98630136
+CatBoost (tuned + ensembled),0.10730594,0.17579909,0.14611872,0.21461187,0.21917808,0.2648402,0.25799087,0.2716895,0.33105022,0.2762557,0.3196347,0.37214613,0.34703195,0.3584475,0.4063927,0.42465752,0.38356164,0.37899545,0.42694065,0.39497718,0.4155251,0.44520548,0.48630136,0.44178084,0.522831,0.5068493,,0.53652966,0.45890412,0.510274,0.51598173,0.5456621,0.5479452,0.456621,0.5559361,0.58447486,0.55936074,0.56164384,0.56392694,0.586758,0.5410959,0.6084475,0.60502285,0.60958904,0.5958904,0.59246576,0.58219177,0.6210046,0.65068495,0.65981734,0.63013697,0.65981734,0.65525115,0.6917808,0.5958904,0.7694064,0.76369864,0.82191783,0.78652966,0.6415525,0.8630137,0.716895,0.75114155,0.783105,0.86986303,0.79452056,0.7716895,0.7808219,0.8892694,0.8196347,0.8059361,0.88584477,0.8287671,0.89041096,0.9269406,0.88584477,0.89497715,0.99086756
+CatBoost (tuned),0.098173514,0.16666667,0.1563927,0.20547946,0.23287672,0.2648402,0.25342464,0.2762557,0.33561644,0.26712328,0.3173516,0.3652968,0.34246576,0.3607306,0.4086758,0.40981734,0.3915525,0.37557077,0.40296805,0.39041096,0.41666666,0.44863012,0.47716895,0.4520548,0.49315068,0.50570774,0.4634703,,0.4611872,0.48630136,0.5068493,0.54223746,0.53424656,0.44292238,0.5753425,0.5707763,0.5353881,0.5707763,0.53424656,0.57990867,0.5273973,0.6187215,0.59474885,0.59018266,0.5936073,0.5936073,0.58447486,0.6347032,0.6347032,0.6232877,0.6312785,0.66210043,0.6666667,0.6815069,0.5913242,0.74657536,0.7374429,0.8333333,0.7694064,0.63242006,0.84018266,0.7237443,0.74657536,0.7648402,0.86986303,0.7808219,0.76141554,0.7648402,0.86986303,0.8105023,0.79680365,0.8812785,0.82191783,0.8812785,0.91780823,0.89497715,0.9041096,0.99086756
+TabPFNv2 (default),0.16894977,0.22031963,0.25799087,0.23744293,0.28424656,0.23287672,0.23972602,0.2625571,0.35388127,0.43835616,0.4634703,0.38127854,0.44634703,0.38356164,0.43607306,0.40182647,0.39726028,0.51141554,0.456621,0.47488585,0.4063927,0.46232876,0.50913244,0.522831,0.5022831,0.533105,0.5410959,0.53881276,,0.5068493,0.51826483,0.543379,0.5456621,0.48059362,0.5479452,0.5045662,0.543379,0.5707763,0.5958904,0.55479455,0.5308219,0.56392694,0.58447486,0.56164384,0.5981735,0.5890411,0.56392694,0.6004566,0.5525114,0.5707763,0.6164383,0.61415523,0.6107306,0.61415523,0.6392694,0.6232877,0.6392694,0.6415525,0.6826484,0.77397263,0.6666667,0.6917808,0.7454338,0.716895,0.67808217,0.72602737,0.760274,0.74429226,0.706621,0.7305936,0.79908675,0.74657536,0.84474885,0.79452056,0.79223746,0.86986303,0.8333333,0.977169
+ModernNCA (tuned),0.08675799,0.13926941,0.13013698,0.16666667,0.17579909,0.24429224,0.19863014,0.25342464,0.34931508,0.34246576,0.3584475,0.34018266,0.33904108,0.37214613,0.39726028,0.37899545,0.38356164,0.44748858,0.41666666,0.37214613,0.38584474,0.4611872,0.47716895,0.51141554,0.46575344,0.49315068,0.48972604,0.51369864,0.49315068,,0.4977169,0.5285388,0.5525114,0.47031963,0.5502283,0.51826483,0.56164384,0.5525114,0.5787671,0.5684931,0.6232877,0.5673516,0.543379,0.5742009,0.63013697,0.5913242,0.5525114,0.60273975,0.60958904,0.61187214,0.673516,0.6369863,0.6438356,0.69634706,0.6210046,0.6849315,0.6940639,0.71004564,0.74657536,0.7305936,0.74657536,0.7283105,0.716895,0.7648402,0.7694064,0.7808219,0.7420091,0.7956621,0.7785388,0.8082192,0.79452056,0.8105023,0.8721461,0.84931505,0.84703195,0.8835617,0.8561644,0.977169
+TabDPT (default),0.098173514,0.15296803,0.22374429,0.15981735,0.28082192,0.26027396,0.32648402,0.3242009,0.19063927,0.39497718,0.4063927,0.35273972,0.41095892,0.43835616,0.35159817,0.38584474,0.413242,0.4063927,0.40182647,0.44863012,0.4634703,0.39954337,0.47716895,0.45890412,0.47945204,0.48630136,0.48401827,0.49315068,0.48173517,0.5022831,,0.5273973,0.51369864,0.5205479,0.51369864,0.5296804,0.5707763,0.51826483,0.543379,0.5707763,0.5776256,0.5410959,0.56621003,0.5753425,0.5353881,0.5559361,0.5479452,0.5936073,0.5776256,0.57990867,0.6369863,0.5890411,0.5936073,0.6415525,0.7146119,0.72602737,0.7328767,0.6712329,0.69634706,0.7876712,0.6712329,0.6757991,0.7283105,0.7853881,0.7374429,0.7694064,0.74429226,0.80136985,0.7694064,0.79908675,0.79223746,0.826484,0.826484,0.8356164,0.84246576,0.89269406,0.8721461,0.99086756
+xRFM (tuned + ensembled),0.09589041,0.15753424,0.1324201,0.1826484,0.19634703,0.20547946,0.22146119,0.24771689,0.30593607,0.30479452,0.3127854,0.3196347,0.39497718,0.3652968,0.37100455,0.34931508,0.39041096,0.4086758,0.4155251,0.46803653,0.39041096,0.413242,0.4611872,0.44406393,0.4257991,0.46232876,0.4543379,0.45776257,0.456621,0.47146118,0.47260273,,0.5787671,0.48401827,0.49885845,0.49543378,0.50913244,0.5707763,0.5810502,0.51141554,0.5273973,0.53196347,0.55707765,0.5296804,0.5913242,0.6164383,0.5730594,0.60958904,0.53881276,0.5559361,0.65753424,0.59931505,0.5981735,0.6232877,0.6461187,0.7146119,0.7351598,0.70091325,0.71004564,0.74885845,0.7374429,0.6894977,0.7694064,0.79908675,0.80365294,0.7785388,0.78995436,0.84931505,0.8242009,0.83219177,0.8230594,0.8710046,0.93378997,0.92009133,0.93378997,0.913242,0.8812785,0.98858446
+xRFM (tuned),0.11643836,0.1849315,0.17694063,0.2089041,0.21689498,0.21232876,0.23744293,0.25570777,0.28995433,0.29223743,0.33789954,0.35388127,0.4178082,0.38698632,0.3824201,0.38356164,0.3847032,0.39041096,0.43607306,0.4634703,0.38698632,0.4497717,0.44406393,0.4497717,0.40068492,0.44520548,0.4520548,0.46575344,0.4543379,0.44748858,0.48630136,0.42123288,,0.5045662,0.46803653,0.53424656,0.5171233,0.55136985,0.5582192,0.5251142,0.55479455,0.5148402,0.5502283,0.53881276,0.57990867,0.60730594,0.50913244,0.60273975,0.53652966,0.5753425,0.6666667,0.6232877,0.6278539,0.63242006,0.6210046,0.7191781,0.74885845,0.6369863,0.72602737,0.7648402,0.70091325,0.716895,0.79452056,0.8150685,0.75114155,0.7568493,0.8242009,0.83789957,0.78995436,0.7979452,0.84931505,0.86986303,0.89726025,0.8630137,0.8652968,0.89497715,0.89269406,0.98858446
+TabPFNv2 (tuned),0.12557077,0.14954337,0.21004567,0.19520548,0.19406393,0.20319635,0.18949772,0.21803653,0.34018266,0.43835616,0.44063926,0.37899545,0.4543379,0.3652968,0.36872146,0.38812786,0.39726028,0.49315068,0.47945204,0.51598173,0.3847032,0.40068492,0.51826483,0.50913244,0.48173517,0.5479452,0.543379,0.55707765,0.5194064,0.5296804,0.47945204,0.51598173,0.49543378,,0.5673516,0.5171233,0.5913242,0.55936074,0.54908675,0.56164384,0.53995436,0.5730594,0.5776256,0.61187214,0.576484,0.58447486,0.586758,0.586758,0.63013697,0.59246576,0.58219177,0.58219177,0.57990867,0.63013697,0.586758,0.5981735,0.6392694,0.6438356,0.6872146,0.69634706,0.673516,0.65981734,0.70091325,0.6803653,0.6461187,0.7123288,0.7237443,0.7191781,0.68607306,0.71803653,0.74885845,0.72602737,0.8356164,0.78995436,0.7773973,0.84931505,0.8082192,0.98401827
+XGBoost (tuned + ensembled),0.06392694,0.15182649,0.12100457,0.18949772,0.21461187,0.23972602,0.23744293,0.25570777,0.3196347,0.21689498,0.26712328,0.30821916,0.30365297,0.3196347,0.39840183,0.38812786,0.35045663,0.2648402,0.3652968,0.37328768,0.37557077,0.4326484,0.4063927,0.37557077,0.43493152,0.4303653,0.44406393,0.42465752,0.4520548,0.4497717,0.48630136,0.50114155,0.53196347,0.4326484,,0.50913244,0.45890412,0.5045662,0.51826483,0.54680365,0.5273973,0.6347032,0.58219177,0.51826483,0.56392694,0.5753425,0.522831,0.5936073,0.5605023,0.58447486,0.5753425,0.6347032,0.6369863,0.6187215,0.59474885,0.7146119,0.7351598,0.8652968,0.7305936,0.64726025,0.89269406,0.69863015,0.739726,0.74657536,0.88013697,0.739726,0.75114155,0.7214612,0.8755708,0.7773973,0.79452056,0.84018266,0.8333333,0.8744292,0.92009133,0.8767123,0.88242006,0.9794521
+SAP-RPT-OSS (default),0.10958904,0.17123288,0.19634703,0.19863014,0.19406393,0.19863014,0.22716895,0.23972602,0.3173516,0.38356164,0.37442923,0.32648402,0.369863,0.36301368,0.3584475,0.37671232,0.4041096,0.41095892,0.39726028,0.44520548,0.46917808,0.42922375,0.4520548,0.45890412,0.4497717,0.43378997,0.4155251,0.42922375,0.49543378,0.48173517,0.47031963,0.5045662,0.46575344,0.48287672,0.49086758,,0.48401827,0.47488585,0.56164384,0.55707765,0.5410959,0.5068493,0.522831,0.51598173,0.5913242,0.51141554,0.53196347,0.54680365,0.53196347,0.5308219,0.6392694,0.56392694,0.56392694,0.63242006,0.63013697,0.65525115,0.6894977,0.6255708,0.65296805,0.74657536,0.67808217,0.69863015,0.7420091,0.7374429,0.706621,0.7762557,0.75342464,0.760274,0.7203196,0.78881276,0.8082192,0.79680365,0.8310502,0.8242009,0.8310502,0.8310502,0.8789954,0.9794521
+ChimeraBoost (tuned + ensembled),0.08219178,0.12557077,0.111872144,0.17579909,0.15068494,0.20091324,0.21347032,0.24086758,0.31506848,0.25456622,0.29452056,0.30365297,0.28538814,0.30821916,0.37671232,0.37214613,0.3105023,0.3938356,0.37100455,0.33333334,0.34931508,0.4155251,0.44292238,0.4497717,0.42465752,0.43493152,0.44063926,0.46461186,0.456621,0.43835616,0.42922375,0.49086758,0.48287672,0.4086758,0.5410959,0.51598173,,0.55479455,0.5,0.5205479,0.48173517,0.6015982,0.5684931,0.6187215,0.56392694,0.5913242,0.5456621,0.5913242,0.7191781,0.5936073,0.5605023,0.65753424,0.65525115,0.6164383,0.5913242,0.673516,0.6849315,0.7671233,0.7374429,0.6643836,0.8002283,0.71004564,0.7146119,0.7305936,0.79908675,0.75570774,0.7283105,0.70091325,0.8082192,0.7808219,0.7671233,0.8105023,0.79680365,0.8356164,0.84474885,0.8584475,0.8630137,0.9714612
+EBM (tuned + ensembled),0.0913242,0.15525115,0.15981735,0.1803653,0.19863014,0.20547946,0.22146119,0.2283105,0.30593607,0.2716895,0.2648402,0.32648402,0.2762557,0.33105022,0.37785387,0.3675799,0.32876712,0.3458904,0.34474885,0.38812786,0.36643836,0.4041096,0.38584474,0.37899545,0.40182647,0.41095892,0.43835616,0.42922375,0.42922375,0.44748858,0.48173517,0.42922375,0.44863012,0.44063926,0.49543378,0.5251142,0.44520548,,0.49429223,0.49315068,0.50913244,0.51369864,0.57990867,0.49315068,0.55936074,0.6666667,0.47945204,0.6917808,0.543379,0.53652966,0.5582192,0.6438356,0.6415525,0.6152968,0.6210046,0.7226027,0.716895,0.71803653,0.739726,0.69634706,0.7420091,0.7283105,0.7694064,0.70319635,0.7808219,0.75570774,0.783105,0.7579909,0.79452056,0.7751142,0.8196347,0.86986303,0.7694064,0.8515982,0.8561644,0.8835617,0.8984018,0.977169
+FastaiMLP (tuned + ensembled),0.10273973,0.1586758,0.14383562,0.15981735,0.21004567,0.19406393,0.22260274,0.24200913,0.2625571,0.2648402,0.2522831,0.30365297,0.32648402,0.36187214,0.33561644,0.29452056,0.34931508,0.39954337,0.38698632,0.37557077,0.3938356,0.38584474,0.3675799,0.4303653,0.369863,0.43607306,0.43607306,0.46575344,0.4041096,0.42123288,0.456621,0.41894978,0.44178084,0.45091325,0.48173517,0.43835616,0.5,0.50570774,,0.47031963,0.5273973,0.5205479,0.49543378,0.5251142,0.5696347,0.55707765,0.5,0.56621003,0.51598173,0.51826483,0.70547944,0.6164383,0.6232877,0.5936073,0.58447486,0.65981734,0.65981734,0.6415525,0.6803653,0.7305936,0.68607306,0.70091325,0.75342464,0.80479455,0.7568493,0.7876712,0.7762557,0.93607306,0.7671233,0.8082192,0.8082192,0.8584475,0.8561644,0.83789957,0.8538813,0.93835616,0.8767123,0.9748858
+TorchMLP (tuned + ensembled),0.0456621,0.111872144,0.10045662,0.12100457,0.14611872,0.17351598,0.1780822,0.19178082,0.2716895,0.21689498,0.2716895,0.28767124,0.3105023,0.3173516,0.31849316,0.34018266,0.3242009,0.39269406,0.37214613,0.35388127,0.37214613,0.3652968,0.3675799,0.44292238,0.38584474,0.4063927,0.413242,0.42009133,0.44520548,0.43150684,0.42922375,0.4885845,0.47488585,0.43835616,0.45319635,0.44292238,0.47945204,0.5068493,0.5296804,,0.51826483,0.4977169,0.4885845,0.5296804,0.55936074,0.55707765,0.5205479,0.5730594,0.5410959,0.5559361,0.64497715,0.6187215,0.6267123,0.7477169,0.57990867,0.6894977,0.6872146,0.70776254,0.7431507,0.70091325,0.70319635,0.7214612,0.7294521,0.8652968,0.75114155,0.79223746,0.760274,0.8025114,0.7808219,0.826484,0.8127854,0.826484,0.8242009,0.80365294,0.8173516,0.9063927,0.8744292,0.9726027
+ModernNCA (default),0.10273973,0.14954337,0.16438356,0.14954337,0.15981735,0.19406393,0.20547946,0.2305936,0.2739726,0.29680365,0.3242009,0.28424656,0.30593607,0.35388127,0.336758,0.34474885,0.37671232,0.38584474,0.3607306,0.34931508,0.40068492,0.38812786,0.40182647,0.413242,0.39726028,0.4543379,0.45890412,0.47260273,0.46917808,0.37671232,0.42237443,0.47260273,0.44520548,0.46004567,0.47260273,0.45890412,0.51826483,0.49086758,0.47260273,0.48173517,,0.51598173,0.47716895,0.5456621,0.5068493,0.53196347,0.44520548,0.55479455,0.55936074,0.5148402,0.55479455,0.6187215,0.6278539,0.5730594,0.6438356,0.65068495,0.6689498,0.6164383,0.71004564,0.7146119,0.6689498,0.739726,0.7214612,0.79223746,0.72602737,0.74885845,0.7454338,0.7625571,0.760274,0.7808219,0.8173516,0.8515982,0.783105,0.83219177,0.8287671,0.8538813,0.89497715,0.98173517
+XGBoost (tuned),0.05479452,0.12785389,0.10730594,0.17123288,0.16894977,0.20091324,0.206621,0.23287672,0.30136988,0.21689498,0.2260274,0.28310502,0.24429224,0.29680365,0.34703195,0.35159817,0.34474885,0.21461187,0.34474885,0.34474885,0.3607306,0.41095892,0.35045663,0.31506848,0.3652968,0.38698632,0.3915525,0.38127854,0.43607306,0.4326484,0.45890412,0.46803653,0.4851598,0.42694065,0.3652968,0.49315068,0.39840183,0.48630136,0.47945204,0.5022831,0.48401827,,0.53881276,0.46575344,0.5273973,0.5456621,0.51141554,0.5502283,0.5045662,0.55479455,0.5753425,0.6187215,0.6244292,0.5890411,0.5684931,0.7089041,0.72602737,0.79680365,0.7191781,0.6347032,0.84018266,0.6940639,0.7237443,0.7191781,0.8515982,0.70547944,0.74885845,0.70776254,0.8538813,0.7648402,0.7853881,0.8413242,0.8173516,0.8721461,0.8995434,0.8721461,0.8789954,0.9748858
+TabM (default),0.06392694,0.10958904,0.12100457,0.13926941,0.15410958,0.19178082,0.20091324,0.21461187,0.30136988,0.25570777,0.25799087,0.30821916,0.17009133,0.34931508,0.37100455,0.3607306,0.34246576,0.30479452,0.3196347,0.28995433,0.38812786,0.42237443,0.3607306,0.34931508,0.4178082,0.38127854,0.39497718,0.40525115,0.4155251,0.456621,0.43378997,0.44292238,0.4497717,0.42237443,0.4178082,0.47716895,0.43150684,0.42009133,0.5045662,0.51141554,0.522831,0.4611872,,0.4634703,0.5251142,0.49543378,0.5022831,0.49200913,0.522831,0.5376712,0.5936073,0.65068495,0.65753424,0.5913242,0.6004566,0.6803653,0.6940639,0.6255708,0.6826484,0.6392694,0.6643836,0.7420091,0.716895,0.7420091,0.7785388,0.73630136,0.73173517,0.7283105,0.7762557,0.74657536,0.783105,0.8242009,0.75570774,0.8356164,0.8515982,0.8630137,0.89041096,0.96803653
+ChimeraBoost (tuned),0.070776254,0.12100457,0.10273973,0.15753424,0.1369863,0.1849315,0.19406393,0.20776255,0.27853882,0.22146119,0.25913242,0.2716895,0.23287672,0.28767124,0.34246576,0.33333334,0.30593607,0.33789954,0.34474885,0.3173516,0.33789954,0.39497718,0.39726028,0.4155251,0.4086758,0.37442923,0.39041096,0.40981734,0.43835616,0.4257991,0.42465752,0.47031963,0.4611872,0.38812786,0.48173517,0.48401827,0.38127854,0.5068493,0.47488585,0.47031963,0.4543379,0.53424656,0.53652966,,0.53424656,0.54680365,0.5045662,0.5525114,0.61187214,0.5376712,0.5376712,0.6415525,0.6392694,0.5753425,0.57990867,0.6415525,0.66210043,0.73173517,0.6917808,0.6369863,0.7420091,0.70319635,0.6894977,0.7191781,0.7694064,0.7043379,0.69634706,0.6849315,0.7716895,0.73173517,0.74657536,0.783105,0.7716895,0.826484,0.8150685,0.85273975,0.8333333,0.9589041
+RealMLP (default),0.08675799,0.11872146,0.14611872,0.13013698,0.14383562,0.17351598,0.18949772,0.20319635,0.24885845,0.22374429,0.17465754,0.2716895,0.23173515,0.33789954,0.33789954,0.34246576,0.34703195,0.33789954,0.34474885,0.28538814,0.3652968,0.37442923,0.30593607,0.35958904,0.33561644,0.37899545,0.4041096,0.4063927,0.40182647,0.369863,0.46461186,0.4086758,0.42009133,0.42351598,0.43607306,0.4086758,0.43607306,0.44063926,0.4303653,0.44063926,0.49315068,0.47260273,0.47488585,0.46575344,,0.5205479,0.43378997,0.5068493,0.49315068,0.5022831,0.5776256,0.6278539,0.6461187,0.5936073,0.586758,0.65753424,0.6757991,0.65981734,0.6917808,0.71803653,0.6894977,0.70319635,0.70547944,0.7751142,0.7305936,0.75570774,0.7420091,0.8150685,0.7579909,0.79223746,0.79223746,0.84474885,0.783105,0.8310502,0.8333333,0.9086758,0.8721461,0.9748858
+EBM (tuned),0.08219178,0.1369863,0.15296803,0.15525115,0.1609589,0.16438356,0.19063927,0.21232876,0.27968037,0.25799087,0.24429224,0.3105023,0.23287672,0.31506848,0.34817353,0.34474885,0.31506848,0.30365297,0.3219178,0.3196347,0.34931508,0.37214613,0.34018266,0.35159817,0.3675799,0.37671232,0.40753424,0.4063927,0.41095892,0.4086758,0.44406393,0.38356164,0.39269406,0.4155251,0.42465752,0.4885845,0.4086758,0.33333334,0.44292238,0.44292238,0.46803653,0.4543379,0.5045662,0.45319635,0.47945204,,0.44863012,0.6152968,0.49543378,0.49543378,0.5376712,0.6232877,0.6289954,0.56164384,0.59018266,0.66210043,0.70091325,0.67808217,0.6849315,0.66780823,0.70776254,0.70776254,0.7374429,0.6826484,0.7420091,0.7328767,0.75570774,0.7146119,0.7694064,0.75114155,0.7853881,0.8515982,0.75342464,0.84246576,0.84018266,0.86757994,0.8995434,0.977169
+ModernNCA (tuned + ensembled),0.05936073,0.0890411,0.06621005,0.13013698,0.11872146,0.21689498,0.20091324,0.21118721,0.3196347,0.2762557,0.31506848,0.2648402,0.29223743,0.30821916,0.31506848,0.34018266,0.33333334,0.37899545,0.4041096,0.34703195,0.3196347,0.37328768,0.4178082,0.4155251,0.39497718,0.413242,0.4178082,0.4155251,0.43607306,0.44748858,0.4520548,0.42694065,0.49086758,0.413242,0.47716895,0.46803653,0.4543379,0.5205479,0.5,0.47945204,0.55479455,0.4885845,0.4977169,0.49543378,0.56621003,0.55136985,,0.53881276,0.4977169,0.5205479,0.55936074,0.543379,0.5479452,0.5707763,0.5,0.5890411,0.60958904,0.6689498,0.6278539,0.6004566,0.6872146,0.61415523,0.5981735,0.65525115,0.65753424,0.6689498,0.60273975,0.6940639,0.65981734,0.673516,0.65525115,0.7237443,0.81392694,0.7648402,0.7671233,0.77397263,0.7579909,0.9315069
+EBM (default),0.0890411,0.13584475,0.14383562,0.14383562,0.17579909,0.1803653,0.18721461,0.20547946,0.2625571,0.23515981,0.24315068,0.30136988,0.24315068,0.3196347,0.35388127,0.34474885,0.31506848,0.24657534,0.26940638,0.30365297,0.35159817,0.37214613,0.37214613,0.3127854,0.3607306,0.34474885,0.37899545,0.3652968,0.39954337,0.39726028,0.4063927,0.39041096,0.39726028,0.413242,0.4063927,0.45319635,0.4086758,0.30821916,0.43378997,0.42694065,0.44520548,0.4497717,0.5079909,0.44748858,0.49315068,0.3847032,0.4611872,,0.45890412,0.46803653,0.53652966,0.60958904,0.60502285,0.56392694,0.5936073,0.6666667,0.65981734,0.6369863,0.70319635,0.67808217,0.6643836,0.70547944,0.74429226,0.70776254,0.75114155,0.7146119,0.76598173,0.7214612,0.7716895,0.7283105,0.8173516,0.8630137,0.7328767,0.8310502,0.82191783,0.8515982,0.9086758,0.9748858
+ChimeraBoost (default),0.061643835,0.13013698,0.098173514,0.16210045,0.14155251,0.16666667,0.17579909,0.19406393,0.2648402,0.20547946,0.25799087,0.26027396,0.21461187,0.28082192,0.3105023,0.29680365,0.29223743,0.25342464,0.3105023,0.28424656,0.3219178,0.37671232,0.34931508,0.3105023,0.36415526,0.32876712,0.34931508,0.3652968,0.44748858,0.39041096,0.42237443,0.4611872,0.4634703,0.369863,0.4394977,0.46803653,0.28082192,0.456621,0.48401827,0.45890412,0.44063926,0.49543378,0.47716895,0.38812786,0.5068493,0.5045662,0.5022831,0.5410959,,0.53424656,0.5296804,0.6347032,0.6392694,0.55479455,0.5753425,0.65068495,0.65296805,0.71347034,0.69863015,0.63242006,0.727169,0.6940639,0.70319635,0.6917808,0.75342464,0.6803653,0.7191781,0.673516,0.760274,0.69634706,0.7625571,0.7773973,0.77397263,0.79908675,0.8082192,0.8105023,0.8515982,0.96575344
+iLTM (tuned),0.07305936,0.09474886,0.10730594,0.14155251,0.14155251,0.15410958,0.1803653,0.19634703,0.24086758,0.20547946,0.24200913,0.24657534,0.25570777,0.29908675,0.2762557,0.2716895,0.3105023,0.30593607,0.3561644,0.32876712,0.34474885,0.34474885,0.3561644,0.3675799,0.2888128,0.33561644,0.34018266,0.37671232,0.42922375,0.38812786,0.42009133,0.44406393,0.42465752,0.40753424,0.4155251,0.46917808,0.4063927,0.4634703,0.48173517,0.44406393,0.4851598,0.44520548,0.46232876,0.46232876,0.4977169,0.5045662,0.47945204,0.53196347,0.46575344,,0.5525114,0.6152968,0.60730594,0.5502283,0.5045662,0.61415523,0.6347032,0.60958904,0.663242,0.6244292,0.65068495,0.67808217,0.6335617,0.75570774,0.71004564,0.6826484,0.6689498,0.75342464,0.7237443,0.7305936,0.7283105,0.783105,0.8059361,0.7694064,0.79223746,0.8333333,0.82990867,0.9589041
+FastaiMLP (tuned),0.07305936,0.12671232,0.13470319,0.11872146,0.15753424,0.14611872,0.15296803,0.16894977,0.20547946,0.22146119,0.22716895,0.25799087,0.2648402,0.2888128,0.2648402,0.25570777,0.30365297,0.3127854,0.29680365,0.29680365,0.35958904,0.30593607,0.28995433,0.37899545,0.31164384,0.35159817,0.369863,0.36872146,0.38356164,0.32648402,0.36301368,0.34246576,0.33333334,0.4178082,0.42465752,0.3607306,0.4394977,0.44178084,0.29452056,0.35502282,0.44520548,0.42465752,0.4063927,0.46232876,0.42237443,0.46232876,0.44063926,0.4634703,0.47031963,0.44748858,,0.5525114,0.5456621,0.51598173,0.5068493,0.5936073,0.5958904,0.57990867,0.5890411,0.6666667,0.60273975,0.65981734,0.6689498,0.69063926,0.6609589,0.70547944,0.6974886,0.7853881,0.67808217,0.74429226,0.74885845,0.783105,0.74885845,0.7671233,0.7876712,0.8607306,0.84018266,0.96575344
+TabSTAR (tuned),0.1803653,0.21575342,0.25342464,0.20776255,0.25570777,0.22146119,0.25342464,0.2625571,0.29223743,0.29680365,0.30593607,0.29680365,0.29452056,0.36187214,0.34474885,0.29223743,0.38127854,0.31849316,0.2739726,0.3127854,0.4041096,0.39611873,0.32876712,0.33561644,0.3652968,0.32762557,0.34018266,0.33789954,0.38584474,0.36301368,0.41095892,0.40068492,0.37671232,0.4178082,0.3652968,0.43607306,0.34246576,0.3561644,0.38356164,0.38127854,0.38127854,0.38127854,0.34931508,0.3584475,0.37214613,0.37671232,0.456621,0.39041096,0.3652968,0.3847032,0.44748858,,0.60958904,0.4041096,0.5296804,0.5684931,0.5673516,0.48173517,0.4155251,0.6164383,0.4885845,0.9292237,0.61187214,0.51598173,0.60502285,0.6369863,0.6347032,0.6415525,0.6347032,0.6438356,0.6917808,0.75570774,0.71004564,0.783105,0.7785388,0.70319635,0.8515982,0.9315069
+TabSTAR (tuned + ensembled),0.1780822,0.20319635,0.25114155,0.21004567,0.24429224,0.21803653,0.24657534,0.25114155,0.28310502,0.28767124,0.29452056,0.29223743,0.28995433,0.36301368,0.34246576,0.28082192,0.37899545,0.31506848,0.26712328,0.30936074,0.39954337,0.39497718,0.3173516,0.33105022,0.36187214,0.3219178,0.34474885,0.33333334,0.3892694,0.3561644,0.4063927,0.40182647,0.37214613,0.42009133,0.36301368,0.43607306,0.34474885,0.3584475,0.37671232,0.37328768,0.37214613,0.37557077,0.34246576,0.3607306,0.35388127,0.37100455,0.4520548,0.39497718,0.3607306,0.39269406,0.4543379,0.39041096,,0.39954337,0.5205479,0.5479452,0.5502283,0.47031963,0.4178082,0.60730594,0.48630136,0.9474886,0.60730594,0.51369864,0.5958904,0.6347032,0.6347032,0.6392694,0.6255708,0.6392694,0.6894977,0.74429226,0.70319635,0.7762557,0.77397263,0.70091325,0.84018266,0.9315069
+TorchMLP (tuned),0.03196347,0.06621005,0.07990868,0.08219178,0.08675799,0.10273973,0.12100457,0.12328767,0.20547946,0.17123288,0.21232876,0.2260274,0.21347032,0.26027396,0.2625571,0.28082192,0.25570777,0.2739726,0.28767124,0.26598173,0.29452056,0.30365297,0.2762557,0.3344749,0.29452056,0.29908675,0.30821916,0.31849316,0.38584474,0.30365297,0.3584475,0.37671232,0.3675799,0.369863,0.38127854,0.3675799,0.38356164,0.3847032,0.4063927,0.2522831,0.42694065,0.41095892,0.4086758,0.42465752,0.4063927,0.43835616,0.42922375,0.43607306,0.44520548,0.4497717,0.48401827,0.5958904,0.6004566,,0.522831,0.6130137,0.60730594,0.56621003,0.65068495,0.6803653,0.59474885,0.6894977,0.65981734,0.76826483,0.6666667,0.69634706,0.6746575,0.70319635,0.6917808,0.7305936,0.7625571,0.772831,0.760274,0.75342464,0.7648402,0.84703195,0.8105023,0.96118724
+OrionMSP (default),0.057077624,0.098173514,0.09474886,0.07534247,0.14840183,0.08675799,0.12328767,0.12785389,0.15981735,0.30251142,0.28995433,0.20319635,0.30593607,0.27283105,0.21689498,0.15525115,0.30707762,0.3242009,0.30593607,0.33333334,0.33561644,0.22146119,0.36643836,0.34246576,0.37328768,0.39954337,0.4041096,0.4086758,0.3607306,0.37899545,0.28538814,0.35388127,0.37899545,0.413242,0.40525115,0.369863,0.4086758,0.37899545,0.4155251,0.42009133,0.3561644,0.43150684,0.39954337,0.42009133,0.413242,0.40981734,0.5,0.4063927,0.42465752,0.49543378,0.49315068,0.47031963,0.47945204,0.47716895,,0.4977169,0.5022831,0.5205479,0.5296804,0.6187215,0.56392694,0.60730594,0.6278539,0.58219177,0.55479455,0.6164383,0.6563927,0.65525115,0.5890411,0.63242006,0.6917808,0.66210043,0.7420091,0.72602737,0.69863015,0.79223746,0.7146119,0.96118724
+ExtraTrees (tuned + ensembled),0.047945205,0.061643835,0.07534247,0.08789954,0.08675799,0.10730594,0.11643836,0.13127854,0.15410958,0.20091324,0.16210045,0.16666667,0.1826484,0.24200913,0.20547946,0.19863014,0.30136988,0.19292237,0.20776255,0.24885845,0.31164384,0.26027396,0.25799087,0.24315068,0.27054796,0.2283105,0.2305936,0.25342464,0.37671232,0.31506848,0.2739726,0.28538814,0.28082192,0.40182647,0.28538814,0.34474885,0.32648402,0.27739727,0.34018266,0.3105023,0.34931508,0.29109588,0.3196347,0.3584475,0.34246576,0.33789954,0.41095892,0.33333334,0.34931508,0.38584474,0.4063927,0.43150684,0.4520548,0.38698632,0.5022831,,0.63242006,0.44063926,0.49315068,0.5707763,0.4885845,0.6221461,0.59931505,0.55936074,0.6586758,0.5776256,0.6289954,0.6004566,0.70091325,0.5981735,0.70319635,0.7853881,0.70776254,0.8721461,0.83789957,0.72602737,0.84246576,0.98858446
+ExtraTrees (tuned),0.057077624,0.06621005,0.0913242,0.0890411,0.09474886,0.10273973,0.12328767,0.12785389,0.12100457,0.19178082,0.17237443,0.17237443,0.20091324,0.24429224,0.19863014,0.20319635,0.293379,0.21004567,0.20776255,0.24200913,0.30593607,0.23972602,0.24657534,0.23401827,0.25342464,0.23515981,0.23630136,0.2625571,0.3607306,0.30593607,0.26712328,0.2648402,0.25114155,0.3607306,0.2648402,0.3105023,0.31506848,0.28310502,0.34018266,0.3127854,0.33105022,0.2739726,0.30593607,0.33789954,0.3242009,0.29908675,0.39041096,0.34018266,0.34703195,0.3652968,0.4041096,0.4326484,0.4497717,0.39269406,0.4977169,0.3675799,,0.41095892,0.49086758,0.5707763,0.43835616,0.6164383,0.5913242,0.543379,0.60273975,0.5456621,0.6164383,0.5730594,0.6461187,0.58447486,0.69634706,0.7374429,0.6803653,0.8059361,0.8002283,0.70547944,0.84931505,0.98630136
+XGBoost (default),0.029680366,0.05936073,0.06621005,0.08675799,0.06392694,0.111872144,0.111872144,0.11872146,0.1826484,0.10273973,0.12785389,0.14840183,0.14611872,0.22374429,0.21917808,0.23972602,0.23744293,0.13013698,0.23515981,0.19178082,0.24885845,0.28767124,0.22374429,0.18150684,0.18949772,0.1826484,0.1780822,0.16666667,0.3584475,0.28995433,0.32876712,0.29908675,0.36301368,0.3561644,0.13470319,0.37442923,0.23287672,0.28196347,0.3584475,0.29223743,0.38356164,0.20319635,0.37442923,0.26826483,0.34018266,0.3219178,0.33105022,0.36301368,0.2865297,0.39041096,0.42009133,0.51826483,0.5296804,0.43378997,0.47945204,0.55936074,0.5890411,,0.5730594,0.5456621,0.50114155,0.6187215,0.6004566,0.5981735,0.6278539,0.51826483,0.61415523,0.586758,0.65753424,0.5753425,0.6757991,0.716895,0.7191781,0.75570774,0.7716895,0.70776254,0.7853881,0.93835616
+iLTM (default),0.05936073,0.070776254,0.07762557,0.0913242,0.10045662,0.10730594,0.11872146,0.13013698,0.20091324,0.15525115,0.14383562,0.18949772,0.15068494,0.2305936,0.19863014,0.22146119,0.24657534,0.15068494,0.18721461,0.1849315,0.28310502,0.2648402,0.20776255,0.17922375,0.21004567,0.16894977,0.21347032,0.2305936,0.3173516,0.25342464,0.30365297,0.28995433,0.2739726,0.3127854,0.26940638,0.34703195,0.2625571,0.26027396,0.3196347,0.25684932,0.28995433,0.28082192,0.3173516,0.30821916,0.30821916,0.31506848,0.37214613,0.29680365,0.30136988,0.336758,0.41095892,0.58447486,0.58219177,0.34931508,0.47031963,0.5068493,0.50913244,0.42694065,,0.543379,0.4634703,0.65296805,0.5981735,0.5479452,0.61415523,0.51826483,0.63013697,0.58447486,0.6347032,0.53196347,0.6940639,0.7579909,0.66210043,0.69863015,0.7305936,0.6415525,0.82191783,0.93607306
+TabFlex (default),0.07305936,0.09589041,0.11872146,0.10273973,0.12100457,0.097031966,0.12557077,0.13926941,0.12785389,0.21232876,0.23515981,0.18721461,0.23287672,0.28538814,0.20319635,0.111872144,0.2260274,0.29223743,0.23858447,0.24885845,0.26712328,0.22146119,0.25913242,0.30821916,0.2648402,0.34246576,0.3584475,0.3675799,0.2260274,0.26940638,0.21232876,0.25114155,0.23515981,0.30365297,0.35273972,0.25342464,0.33561644,0.30365297,0.26940638,0.29908675,0.28538814,0.3652968,0.3607306,0.36301368,0.28196347,0.3321918,0.39954337,0.3219178,0.3675799,0.37557077,0.33333334,0.38356164,0.39269406,0.3196347,0.38127854,0.42922375,0.42922375,0.4543379,0.456621,,0.47031963,0.5296804,0.5251142,0.4783105,0.5022831,0.56392694,0.5650685,0.5456621,0.51369864,0.60273975,0.61415523,0.61187214,0.63242006,0.6232877,0.65296805,0.69863015,0.6940639,0.89269406
+LightGBM (default),0.02739726,0.0456621,0.052511416,0.05936073,0.052511416,0.05936073,0.06621005,0.07762557,0.13926941,0.07534247,0.098173514,0.10159817,0.09474886,0.20319635,0.16894977,0.21118721,0.25114155,0.05022831,0.18150684,0.16438356,0.24429224,0.24885845,0.18949772,0.08561644,0.14155251,0.14840183,0.1369863,0.15981735,0.33333334,0.25342464,0.32876712,0.2625571,0.29908675,0.32648402,0.10730594,0.3219178,0.19977169,0.25799087,0.31392694,0.29680365,0.33105022,0.15981735,0.33561644,0.25799087,0.3105023,0.29223743,0.3127854,0.33561644,0.27283105,0.34931508,0.39726028,0.51141554,0.51369864,0.40525115,0.43607306,0.51141554,0.56164384,0.49885845,0.53652966,0.5296804,,0.60730594,0.5776256,0.586758,0.5913242,0.51826483,0.5890411,0.5707763,0.61986303,0.586758,0.66210043,0.6917808,0.72602737,0.75114155,0.76826483,0.7328767,0.79452056,0.95776254
+TabSTAR (default),0.14611872,0.16894977,0.19863014,0.16438356,0.21004567,0.18721461,0.21004567,0.20776255,0.21004567,0.23972602,0.25114155,0.2260274,0.24657534,0.2956621,0.23972602,0.21232876,0.3196347,0.23972602,0.21118721,0.26712328,0.3242009,0.2625571,0.2648402,0.25799087,0.28082192,0.2716895,0.28310502,0.2762557,0.30821916,0.2716895,0.3242009,0.3105023,0.28310502,0.34018266,0.30136988,0.30136988,0.28995433,0.2716895,0.29908675,0.27853882,0.26027396,0.30593607,0.25799087,0.29680365,0.29680365,0.29223743,0.38584474,0.29452056,0.30593607,0.3219178,0.34018266,0.070776254,0.052511416,0.3105023,0.39269406,0.37785387,0.38356164,0.38127854,0.34703195,0.47031963,0.39269406,,0.4885845,0.39041096,0.4257991,0.5410959,0.5022831,0.4543379,0.4497717,0.5456621,0.56164384,0.60502285,0.6255708,0.6187215,0.5936073,0.6232877,0.72602737,0.9223744
+Linear (tuned + ensembled),0.07534247,0.12100457,0.12557077,0.10502283,0.1324201,0.14611872,0.14611872,0.16666667,0.19178082,0.16324201,0.17351598,0.21917808,0.16666667,0.2283105,0.24885845,0.21004567,0.23515981,0.1826484,0.16894977,0.19863014,0.2648402,0.25913242,0.24429224,0.21689498,0.28082192,0.24429224,0.24885845,0.25342464,0.25456622,0.28310502,0.2716895,0.2305936,0.20547946,0.29908675,0.26027396,0.25799087,0.28538814,0.2305936,0.24657534,0.27054796,0.27853882,0.2762557,0.28310502,0.3105023,0.29452056,0.2625571,0.40182647,0.25570777,0.29680365,0.36643836,0.33105022,0.38812786,0.39269406,0.34018266,0.37214613,0.40068492,0.4086758,0.39954337,0.40182647,0.47488585,0.42237443,0.51141554,,0.44292238,0.43150684,0.47945204,0.66780823,0.49315068,0.45890412,0.5022831,0.7579909,0.5753425,0.5981735,0.5936073,0.56621003,0.6210046,0.67808217,0.89269406
+TorchMLP (default),0.02283105,0.052511416,0.052511416,0.043378994,0.05022831,0.038812786,0.07534247,0.07762557,0.12100457,0.11415525,0.10730594,0.13013698,0.10958904,0.21917808,0.15981735,0.13926941,0.21917808,0.18949772,0.18949772,0.12557077,0.24429224,0.18949772,0.15525115,0.20776255,0.1826484,0.20091324,0.21689498,0.23515981,0.28310502,0.23515981,0.21461187,0.20091324,0.1849315,0.3196347,0.25342464,0.2625571,0.26940638,0.29680365,0.19520548,0.13470319,0.20776255,0.28082192,0.25799087,0.28082192,0.22488585,0.3173516,0.34474885,0.29223743,0.30821916,0.24429224,0.30936074,0.48401827,0.48630136,0.23173515,0.4178082,0.44063926,0.456621,0.40182647,0.4520548,0.5216895,0.413242,0.60958904,0.55707765,,0.49086758,0.5456621,0.57990867,0.5776256,0.5479452,0.58447486,0.65981734,0.69863015,0.65296805,0.6643836,0.6643836,0.72602737,0.7785388,0.94063926
+RandomForest (tuned + ensembled),0.02283105,0.061643835,0.057077624,0.061643835,0.06392694,0.10273973,0.08675799,0.111872144,0.12328767,0.10730594,0.0890411,0.1324201,0.13470319,0.19634703,0.14383562,0.19406393,0.2648402,0.10045662,0.15753424,0.16666667,0.25114155,0.21461187,0.1780822,0.15981735,0.15981735,0.12328767,0.13013698,0.13013698,0.3219178,0.2305936,0.2625571,0.19634703,0.24885845,0.35388127,0.11986301,0.293379,0.20091324,0.21917808,0.24315068,0.24885845,0.2739726,0.14840183,0.22146119,0.2305936,0.26940638,0.25799087,0.34246576,0.24885845,0.24657534,0.28995433,0.33904108,0.39497718,0.4041096,0.33333334,0.44520548,0.3413242,0.39726028,0.37214613,0.38584474,0.4977169,0.4086758,0.5742009,0.5684931,0.50913244,,0.51598173,0.58447486,0.53196347,0.63242006,0.5410959,0.66552514,0.70091325,0.6849315,0.75114155,0.8173516,0.6278539,0.8150685,0.96347034
+PerpetualBooster (tuned + ensembled),0.01826484,0.06392694,0.06392694,0.057077624,0.06849315,0.06849315,0.06849315,0.08789954,0.11872146,0.08219178,0.08675799,0.10958904,0.111872144,0.1780822,0.14155251,0.13470319,0.21689498,0.12557077,0.17123288,0.11643836,0.22374429,0.16210045,0.15296803,0.14954337,0.16438356,0.21461187,0.20547946,0.21917808,0.2739726,0.21917808,0.2305936,0.22146119,0.24315068,0.28767124,0.26027396,0.22374429,0.24429224,0.24429224,0.21232876,0.20776255,0.25114155,0.29452056,0.26369864,0.2956621,0.24429224,0.26712328,0.33105022,0.28538814,0.3196347,0.3173516,0.29452056,0.36301368,0.3652968,0.30365297,0.38356164,0.42237443,0.4543379,0.48173517,0.48173517,0.43607306,0.48173517,0.45890412,0.5205479,0.4543379,0.48401827,,0.53196347,0.44748858,0.50913244,0.68607306,0.5890411,0.6130137,0.6415525,0.6643836,0.6826484,0.79223746,0.70319635,0.94292235
+Linear (tuned),0.06621005,0.10958904,0.098173514,0.0913242,0.10730594,0.13470319,0.12557077,0.14155251,0.17123288,0.13584475,0.16894977,0.20091324,0.15182649,0.20319635,0.2283105,0.1849315,0.20091324,0.16894977,0.14383562,0.1826484,0.23744293,0.2283105,0.23401827,0.19292237,0.25570777,0.22374429,0.2283105,0.23858447,0.23972602,0.25799087,0.25570777,0.21004567,0.17579909,0.2762557,0.24885845,0.24657534,0.2716895,0.21689498,0.22374429,0.23972602,0.25456622,0.25114155,0.26826483,0.30365297,0.25799087,0.24429224,0.39726028,0.23401827,0.28082192,0.33105022,0.30251142,0.3652968,0.3652968,0.32534248,0.3436073,0.37100455,0.38356164,0.38584474,0.369863,0.43493152,0.41095892,0.4977169,0.3321918,0.42009133,0.4155251,0.46803653,,0.47031963,0.43378997,0.4885845,0.6757991,0.5262557,0.5890411,0.5684931,0.5456621,0.6187215,0.65981734,0.8835617
+FastaiMLP (default),0.043378994,0.05936073,0.061643835,0.06849315,0.0890411,0.0913242,0.08675799,0.10045662,0.09360731,0.0913242,0.0890411,0.13584475,0.12785389,0.17123288,0.12785389,0.15296803,0.19863014,0.21461187,0.20776255,0.17579909,0.20547946,0.15525115,0.124429226,0.23515981,0.1849315,0.21232876,0.21917808,0.23515981,0.25570777,0.2043379,0.19863014,0.15068494,0.16210045,0.28082192,0.27853882,0.23972602,0.29908675,0.24200913,0.06392694,0.19748859,0.23744293,0.29223743,0.2716895,0.31506848,0.1849315,0.28538814,0.30593607,0.27853882,0.32648402,0.24657534,0.21461187,0.3584475,0.3607306,0.29680365,0.34474885,0.39954337,0.42694065,0.413242,0.4155251,0.4543379,0.42922375,0.5456621,0.5068493,0.42237443,0.46803653,0.5525114,0.5296804,,0.47488585,0.57990867,0.56621003,0.6415525,0.6232877,0.6278539,0.6130137,0.7146119,0.739726,0.94292235
+RandomForest (tuned),0.025114154,0.038812786,0.05479452,0.06392694,0.0456621,0.06621005,0.07534247,0.08675799,0.10958904,0.097031966,0.070776254,0.10273973,0.11415525,0.17579909,0.14383562,0.17579909,0.23630136,0.07762557,0.15753424,0.16894977,0.2305936,0.1803653,0.16210045,0.12557077,0.13926941,0.11872146,0.110730596,0.13013698,0.293379,0.22146119,0.2305936,0.17579909,0.21004567,0.31392694,0.124429226,0.27968037,0.19178082,0.20547946,0.23287672,0.21917808,0.23972602,0.14611872,0.22374429,0.2283105,0.24200913,0.2305936,0.34018266,0.2283105,0.23972602,0.2762557,0.3219178,0.3652968,0.37442923,0.30821916,0.41095892,0.29908675,0.35388127,0.34246576,0.3652968,0.48630136,0.380137,0.5502283,0.5410959,0.4520548,0.3675799,0.49086758,0.56621003,0.5251142,,0.5273973,0.65296805,0.6769406,0.6643836,0.7146119,0.73630136,0.61187214,0.8082192,0.96575344
+PerpetualBooster (tuned),0.015981736,0.047945205,0.049086757,0.047945205,0.06849315,0.05479452,0.061643835,0.07534247,0.09589041,0.07990868,0.07534247,0.0913242,0.0890411,0.15525115,0.12671232,0.13470319,0.19406393,0.10273973,0.15525115,0.10273973,0.19178082,0.15296803,0.14840183,0.12785389,0.14383562,0.17579909,0.1803653,0.18949772,0.26940638,0.19178082,0.20091324,0.16780822,0.2020548,0.28196347,0.22260274,0.21118721,0.21917808,0.22488585,0.19178082,0.17351598,0.21917808,0.23515981,0.25342464,0.26826483,0.20776255,0.24885845,0.32648402,0.2716895,0.30365297,0.26940638,0.25570777,0.3561644,0.3607306,0.26940638,0.3675799,0.40182647,0.4155251,0.42465752,0.46803653,0.39726028,0.413242,0.4543379,0.4977169,0.4155251,0.45890412,0.31392694,0.51141554,0.42009133,0.47260273,,0.56621003,0.57990867,0.6221461,0.6415525,0.65753424,0.73401827,0.6643836,0.91552514
+Linear (default),0.0456621,0.06849315,0.07420091,0.06278539,0.07990868,0.070776254,0.09474886,0.09018265,0.09360731,0.10730594,0.14840183,0.124429226,0.12100457,0.19634703,0.16210045,0.14269407,0.16666667,0.11872146,0.0890411,0.15068494,0.21232876,0.1803653,0.1849315,0.14155251,0.21917808,0.17579909,0.19406393,0.20319635,0.20091324,0.20547946,0.20776255,0.17694063,0.15068494,0.25114155,0.20547946,0.19178082,0.23287672,0.1803653,0.19178082,0.18721461,0.1826484,0.21461187,0.21689498,0.25342464,0.20776255,0.21461187,0.34474885,0.1826484,0.23744293,0.2716895,0.25114155,0.30821916,0.3105023,0.23744293,0.30821916,0.29680365,0.30365297,0.3242009,0.30593607,0.38584474,0.33789954,0.43835616,0.24200913,0.34018266,0.3344749,0.41095892,0.3242009,0.43378997,0.34703195,0.43378997,,0.44292238,0.56392694,0.5068493,0.51141554,0.5559361,0.5936073,0.8333333
+KNN (tuned + ensembled),0.02739726,0.061643835,0.07305936,0.052511416,0.07305936,0.0913242,0.0890411,0.09589041,0.09589041,0.08675799,0.09589041,0.11643836,0.11872146,0.19178082,0.13356164,0.14840183,0.2283105,0.10045662,0.0890411,0.14611872,0.22146119,0.14383562,0.15525115,0.1369863,0.15753424,0.11415525,0.11415525,0.11872146,0.25342464,0.18949772,0.17351598,0.12899543,0.13013698,0.2739726,0.15981735,0.20319635,0.18949772,0.13013698,0.14155251,0.17351598,0.14840183,0.1586758,0.17579909,0.21689498,0.15525115,0.14840183,0.2762557,0.1369863,0.22260274,0.21689498,0.21689498,0.24429224,0.25570777,0.22716895,0.33789954,0.21461187,0.2625571,0.28310502,0.24200913,0.38812786,0.30821916,0.39497718,0.42465752,0.30136988,0.29908675,0.38698632,0.4737443,0.3584475,0.32305935,0.42009133,0.55707765,,0.58447486,0.49086758,0.5251142,0.5479452,0.8458904,0.96803653
+xRFM (default),0.025114154,0.0456621,0.057077624,0.047945205,0.061643835,0.057077624,0.07305936,0.07990868,0.070776254,0.070776254,0.07534247,0.07762557,0.11415525,0.111872144,0.083333336,0.10958904,0.09589041,0.15296803,0.1849315,0.15753424,0.098173514,0.11872146,0.1324201,0.17465754,0.0890411,0.16894977,0.17123288,0.1780822,0.15525115,0.12785389,0.17351598,0.06621005,0.10273973,0.16438356,0.16666667,0.16894977,0.20319635,0.2305936,0.14383562,0.17579909,0.21689498,0.1826484,0.24429224,0.2283105,0.21689498,0.24657534,0.18607306,0.26712328,0.2260274,0.19406393,0.25114155,0.28995433,0.29680365,0.23972602,0.25799087,0.29223743,0.3196347,0.28082192,0.33789954,0.3675799,0.2739726,0.37442923,0.40182647,0.34703195,0.31506848,0.3584475,0.41095892,0.37671232,0.33561644,0.37785387,0.43607306,0.4155251,,0.4497717,0.413242,0.51369864,0.522831,0.8630137
+ExtraTrees (default),0.015981736,0.029680366,0.05022831,0.047945205,0.05479452,0.05479452,0.057077624,0.06621005,0.069634706,0.06392694,0.061643835,0.047945205,0.07762557,0.12100457,0.07305936,0.11415525,0.15981735,0.07762557,0.1369863,0.11643836,0.13013698,0.11643836,0.15068494,0.09360731,0.12100457,0.12557077,0.10958904,0.11872146,0.20547946,0.15068494,0.16438356,0.07990868,0.1369863,0.21004567,0.12557077,0.17579909,0.16438356,0.14840183,0.16210045,0.19634703,0.16780822,0.12785389,0.16438356,0.17351598,0.16894977,0.15753424,0.23515981,0.16894977,0.20091324,0.2305936,0.23287672,0.21689498,0.22374429,0.24657534,0.2739726,0.12785389,0.19406393,0.24429224,0.30136988,0.37671232,0.24885845,0.38127854,0.4063927,0.33561644,0.24885845,0.33561644,0.43150684,0.37214613,0.28538814,0.3584475,0.49315068,0.50913244,0.5502283,,0.522831,0.4977169,0.6415525,0.956621
+RandomForest (default),0.02283105,0.025114154,0.047945205,0.043378994,0.038812786,0.043378994,0.05479452,0.05936073,0.06392694,0.06392694,0.038812786,0.051369864,0.07762557,0.10273973,0.07534247,0.10502283,0.14840183,0.047945205,0.10730594,0.10730594,0.13127854,0.10273973,0.11872146,0.07534247,0.09589041,0.10045662,0.07305936,0.08219178,0.20776255,0.15296803,0.15753424,0.06621005,0.13470319,0.22260274,0.07990868,0.16894977,0.15525115,0.14383562,0.14611872,0.1826484,0.17123288,0.10045662,0.14840183,0.1849315,0.16666667,0.15981735,0.23287672,0.1780822,0.19178082,0.20776255,0.21232876,0.22146119,0.2260274,0.23515981,0.30136988,0.16210045,0.19977169,0.2283105,0.26940638,0.34703195,0.23173515,0.4063927,0.43378997,0.33561644,0.1826484,0.3173516,0.4543379,0.38698632,0.26369864,0.34246576,0.4885845,0.47488585,0.586758,0.47716895,,0.5045662,0.6347032,0.94063926
+PerpetualBooster (default),0.011415525,0.029680366,0.03652968,0.020547945,0.038812786,0.029680366,0.038812786,0.03652968,0.034246575,0.05479452,0.034246575,0.038812786,0.034246575,0.10502283,0.06392694,0.07534247,0.07990868,0.05479452,0.07990868,0.043378994,0.0913242,0.06621005,0.052511416,0.057077624,0.0913242,0.10730594,0.11415525,0.10502283,0.13013698,0.11643836,0.10730594,0.08675799,0.10502283,0.15068494,0.12328767,0.16894977,0.14155251,0.11643836,0.061643835,0.09360731,0.14611872,0.12785389,0.1369863,0.14726028,0.0913242,0.1324201,0.2260274,0.14840183,0.18949772,0.16666667,0.13926941,0.29680365,0.29908675,0.15296803,0.20776255,0.2739726,0.29452056,0.29223743,0.3584475,0.30136988,0.26712328,0.37671232,0.37899545,0.2739726,0.37214613,0.20776255,0.38127854,0.28538814,0.38812786,0.26598173,0.44406393,0.4520548,0.48630136,0.5022831,0.49543378,,0.5525114,0.89269406
+KNN (tuned),0.02739726,0.03652968,0.05479452,0.04109589,0.052511416,0.05479452,0.06849315,0.06621005,0.047945205,0.08219178,0.07534247,0.061643835,0.0913242,0.14840183,0.0913242,0.061643835,0.13356164,0.07876712,0.05022831,0.11643836,0.14611872,0.10273973,0.12785389,0.0913242,0.11986301,0.08675799,0.10502283,0.09589041,0.16666667,0.14383562,0.12785389,0.11872146,0.10730594,0.19178082,0.11757991,0.12100457,0.1369863,0.10159817,0.12328767,0.12557077,0.10502283,0.12100457,0.10958904,0.16666667,0.12785389,0.10045662,0.24200913,0.0913242,0.14840183,0.17009133,0.15981735,0.14840183,0.15981735,0.18949772,0.28538814,0.15753424,0.15068494,0.21461187,0.1780822,0.30593607,0.20547946,0.2739726,0.3219178,0.22146119,0.1849315,0.29680365,0.34018266,0.26027396,0.19178082,0.33561644,0.4063927,0.15410958,0.47716895,0.3584475,0.3652968,0.44748858,,0.9497717
+KNN (default),0.006849315,0.00456621,0.006849315,0.00456621,0.00913242,0.006849315,0.00913242,0.00913242,0.006849315,0.00456621,0.006849315,0.00913242,0.01826484,0.034246575,0.006849315,0.006849315,0.011415525,0.00913242,0.015981736,0.02739726,0.006849315,0.011415525,0.020547945,0.006849315,0.025114154,0.01369863,0.00913242,0.00913242,0.02283105,0.02283105,0.00913242,0.011415525,0.011415525,0.015981736,0.020547945,0.020547945,0.028538812,0.02283105,0.025114154,0.02739726,0.01826484,0.025114154,0.03196347,0.04109589,0.025114154,0.02283105,0.06849315,0.025114154,0.034246575,0.04109589,0.034246575,0.06849315,0.06849315,0.038812786,0.038812786,0.011415525,0.01369863,0.061643835,0.06392694,0.10730594,0.042237442,0.07762557,0.10730594,0.05936073,0.03652968,0.057077624,0.11643836,0.057077624,0.034246575,0.084474884,0.16666667,0.03196347,0.1369863,0.043378994,0.05936073,0.10730594,0.05022831,
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
index f362223e455fe934aeefc9b64f000bcd78fe882c..333ecbd4b15605d10c785664ecf58df868fffc0b 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Binary | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..64885a4d758e4130745210dedc04cafd567ff869
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_binary/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2061.0,"score":0.714,"rank":5.23,"hrank":2.76,"imp":5.398,"train_time":128.64,"predict_time":2.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+171\/-60","imputed":false},{"position":1,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2055.0,"score":0.878,"rank":5.36,"hrank":1.55,"imp":3.02,"train_time":42.17,"predict_time":19.818,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+380\/-164","imputed":false},{"position":2,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1904.0,"score":0.616,"rank":9.74,"hrank":3.48,"imp":6.355,"train_time":325.92,"predict_time":2.156,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+184\/-74","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1869.0,"score":0.591,"rank":11.12,"hrank":3.99,"imp":8.448,"train_time":1.18,"predict_time":0.306,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+263\/-167","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1787.0,"score":0.456,"rank":14.86,"hrank":9.5,"imp":9.828,"train_time":2.76,"predict_time":0.886,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-92","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1782.0,"score":0.481,"rank":15.12,"hrank":4.82,"imp":8.27,"train_time":0.65,"predict_time":0.198,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+178\/-112","imputed":false},{"position":6,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1761.0,"score":0.49,"rank":16.19,"hrank":7.23,"imp":8.289,"train_time":1.38,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+140\/-84","imputed":false},{"position":7,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1760.0,"score":0.446,"rank":16.25,"hrank":12.41,"imp":9.784,"train_time":1883.82,"predict_time":80.768,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+114\/-62","imputed":false},{"position":8,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1755.0,"score":0.47,"rank":16.54,"hrank":7.94,"imp":8.625,"train_time":329.05,"predict_time":0.166,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+158\/-91","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1745.0,"score":0.416,"rank":17.05,"hrank":11.77,"imp":9.151,"train_time":1633.8,"predict_time":2.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-49","imputed":false},{"position":10,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1740.0,"score":0.461,"rank":17.33,"hrank":7.91,"imp":8.922,"train_time":329.05,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+173\/-106","imputed":false},{"position":11,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1720.0,"score":0.426,"rank":18.44,"hrank":10.47,"imp":10.523,"train_time":95.92,"predict_time":0.332,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+174\/-133","imputed":false},{"position":12,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1715.0,"score":0.342,"rank":18.73,"hrank":10.42,"imp":10.352,"train_time":583.55,"predict_time":0.579,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+175\/-109","imputed":false},{"position":13,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1712.0,"score":0.381,"rank":18.91,"hrank":12.0,"imp":10.95,"train_time":110.13,"predict_time":0.386,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-88","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1698.0,"score":0.36,"rank":19.77,"hrank":13.22,"imp":11.42,"train_time":86.85,"predict_time":0.668,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+155\/-77","imputed":false},{"position":15,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1675.0,"score":0.351,"rank":21.1,"hrank":13.51,"imp":10.709,"train_time":2139.17,"predict_time":6.41,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-132","imputed":false},{"position":16,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1655.0,"score":0.358,"rank":22.36,"hrank":12.13,"imp":11.186,"train_time":95.92,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+158\/-106","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1651.0,"score":0.305,"rank":22.63,"hrank":13.23,"imp":10.511,"train_time":1633.8,"predict_time":0.088,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-66","imputed":false},{"position":18,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1650.0,"score":0.297,"rank":22.64,"hrank":12.56,"imp":11.894,"train_time":842.25,"predict_time":15.118,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+163\/-100","imputed":false},{"position":19,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1640.0,"score":0.291,"rank":23.3,"hrank":18.18,"imp":11.775,"train_time":110.13,"predict_time":0.061,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-73","imputed":false},{"position":20,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1636.0,"score":0.279,"rank":23.59,"hrank":12.45,"imp":11.248,"train_time":583.55,"predict_time":0.056,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+168\/-112","imputed":false},{"position":21,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1617.0,"score":0.292,"rank":24.78,"hrank":15.21,"imp":11.861,"train_time":0.48,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+163\/-120","imputed":false},{"position":22,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1609.0,"score":0.262,"rank":25.32,"hrank":13.71,"imp":12.444,"train_time":842.25,"predict_time":1.465,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+167\/-106","imputed":false},{"position":23,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1590.0,"score":0.257,"rank":26.57,"hrank":16.26,"imp":12.197,"train_time":2.3,"predict_time":0.896,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+133\/-113","imputed":false},{"position":24,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1583.0,"score":0.206,"rank":26.99,"hrank":20.04,"imp":13.106,"train_time":86.85,"predict_time":0.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+141\/-78","imputed":false},{"position":25,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1572.0,"score":0.246,"rank":27.72,"hrank":15.39,"imp":12.869,"train_time":8.53,"predict_time":3.578,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-147","imputed":false},{"position":26,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1568.0,"score":0.217,"rank":28.01,"hrank":15.13,"imp":12.775,"train_time":1883.82,"predict_time":10.881,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-109","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1557.0,"score":0.171,"rank":28.72,"hrank":21.09,"imp":13.195,"train_time":2139.17,"predict_time":0.349,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-72","imputed":false},{"position":28,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1539.0,"score":0.191,"rank":30.0,"hrank":17.42,"imp":12.773,"train_time":4860.88,"predict_time":475.78,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-141","imputed":false},{"position":29,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1538.0,"score":0.186,"rank":30.05,"hrank":22.88,"imp":12.583,"train_time":1156.69,"predict_time":1.558,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-107","imputed":false},{"position":30,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1530.0,"score":0.226,"rank":30.57,"hrank":13.99,"imp":14.371,"train_time":0.76,"predict_time":0.203,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+196\/-140","imputed":false},{"position":31,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1486.0,"score":0.189,"rank":33.67,"hrank":23.26,"imp":13.749,"train_time":572.82,"predict_time":1.935,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-125","imputed":false},{"position":32,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1482.0,"score":0.146,"rank":33.91,"hrank":27.23,"imp":15.71,"train_time":188.8,"predict_time":0.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+132\/-110","imputed":false},{"position":33,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1473.0,"score":0.154,"rank":34.56,"hrank":19.67,"imp":14.168,"train_time":0.49,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-129","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1470.0,"score":0.141,"rank":34.74,"hrank":26.92,"imp":13.804,"train_time":1.5,"predict_time":0.054,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-146","imputed":false},{"position":35,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1455.0,"score":0.149,"rank":35.79,"hrank":23.08,"imp":14.857,"train_time":0.65,"predict_time":0.154,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+178\/-163","imputed":false},{"position":36,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1454.0,"score":0.154,"rank":35.91,"hrank":28.34,"imp":15.465,"train_time":572.82,"predict_time":0.109,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-124","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1437.0,"score":0.06,"rank":37.1,"hrank":33.97,"imp":14.238,"train_time":6.41,"predict_time":0.247,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-59","imputed":false},{"position":38,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1427.0,"score":0.106,"rank":37.81,"hrank":30.36,"imp":14.727,"train_time":371.59,"predict_time":2.386,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-149","imputed":false},{"position":39,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1426.0,"score":0.077,"rank":37.86,"hrank":30.44,"imp":16.866,"train_time":96.61,"predict_time":0.754,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-135","imputed":false},{"position":40,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1409.0,"score":0.079,"rank":39.05,"hrank":34.48,"imp":16.781,"train_time":188.8,"predict_time":0.008,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-104","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1392.0,"score":0.066,"rank":40.19,"hrank":33.63,"imp":14.805,"train_time":1156.69,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-118","imputed":false},{"position":42,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1375.0,"score":0.055,"rank":41.35,"hrank":27.17,"imp":16.118,"train_time":4860.88,"predict_time":42.641,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+119\/-143","imputed":false},{"position":43,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1369.0,"score":0.05,"rank":41.8,"hrank":37.07,"imp":18.373,"train_time":0.71,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-112","imputed":false},{"position":44,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1359.0,"score":0.037,"rank":42.48,"hrank":39.09,"imp":15.811,"train_time":0.29,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-112","imputed":false},{"position":45,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1338.0,"score":0.054,"rank":43.9,"hrank":38.92,"imp":19.541,"train_time":82.93,"predict_time":8.231,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-117","imputed":false},{"position":46,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1324.0,"score":0.013,"rank":44.81,"hrank":41.52,"imp":17.083,"train_time":8.34,"predict_time":0.25,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-84","imputed":false},{"position":47,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1323.0,"score":0.003,"rank":44.85,"hrank":42.66,"imp":17.01,"train_time":42.95,"predict_time":0.226,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-67","imputed":false},{"position":48,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1311.0,"score":0.012,"rank":45.67,"hrank":41.98,"imp":18.698,"train_time":16.41,"predict_time":16.24,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-114","imputed":false},{"position":49,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1291.0,"score":0.005,"rank":46.93,"hrank":44.82,"imp":16.58,"train_time":59.3,"predict_time":0.156,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-106","imputed":false},{"position":50,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1287.0,"score":0.006,"rank":47.15,"hrank":43.65,"imp":17.883,"train_time":42.95,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-76","imputed":false},{"position":51,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1284.0,"score":0.029,"rank":47.4,"hrank":40.87,"imp":18.045,"train_time":371.59,"predict_time":0.102,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-139","imputed":false},{"position":52,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1283.0,"score":0.019,"rank":47.44,"hrank":37.01,"imp":20.148,"train_time":47.03,"predict_time":45.876,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+160\/-190","imputed":false},{"position":53,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1267.0,"score":0.076,"rank":48.46,"hrank":31.46,"imp":22.562,"train_time":5.11,"predict_time":2.699,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+184\/-316","imputed":false},{"position":54,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.0,"rank":49.85,"hrank":47.99,"imp":18.065,"train_time":1.86,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-85","imputed":false},{"position":55,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1244.0,"score":0.0,"rank":49.85,"hrank":47.62,"imp":17.828,"train_time":59.3,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-103","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1220.0,"score":0.033,"rank":51.28,"hrank":44.39,"imp":20.939,"train_time":1.52,"predict_time":0.115,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-208","imputed":false},{"position":57,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1157.0,"score":0.009,"rank":54.91,"hrank":49.11,"imp":25.91,"train_time":43.64,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+171\/-229","imputed":false},{"position":58,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1141.0,"score":0.01,"rank":55.78,"hrank":46.38,"imp":23.022,"train_time":513.06,"predict_time":78.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-172","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1141.0,"score":0.0,"rank":55.78,"hrank":53.92,"imp":25.608,"train_time":3265.68,"predict_time":1.948,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-117","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1132.0,"score":0.0,"rank":56.28,"hrank":54.48,"imp":25.685,"train_time":3265.68,"predict_time":5.38,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-123","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1116.0,"score":0.005,"rank":57.12,"hrank":52.24,"imp":26.784,"train_time":43.64,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+177\/-221","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1077.0,"score":0.0,"rank":59.01,"hrank":50.8,"imp":25.041,"train_time":20.26,"predict_time":0.641,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+134\/-222","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1042.0,"score":0.0,"rank":60.64,"hrank":55.35,"imp":26.165,"train_time":20.26,"predict_time":0.258,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+133\/-240","imputed":false},{"position":64,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1011.0,"score":0.0,"rank":62.01,"hrank":60.53,"imp":24.431,"train_time":0.34,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-232","imputed":false},{"position":65,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1005.0,"score":0.0,"rank":62.26,"hrank":59.87,"imp":28.451,"train_time":0.22,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-238","imputed":false},{"position":66,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":62.46,"hrank":61.48,"imp":24.92,"train_time":0.12,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-220","imputed":false},{"position":67,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":990.0,"score":0.0,"rank":62.85,"hrank":61.48,"imp":26.242,"train_time":17.63,"predict_time":7.0,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+109\/-205","imputed":false},{"position":68,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":988.0,"score":0.023,"rank":62.95,"hrank":55.55,"imp":30.971,"train_time":2.62,"predict_time":0.186,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+183\/-376","imputed":false},{"position":69,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":967.0,"score":0.0,"rank":63.75,"hrank":62.86,"imp":31.128,"train_time":101.08,"predict_time":2.021,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-120","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":936.0,"score":0.0,"rank":64.89,"hrank":63.78,"imp":32.252,"train_time":0.24,"predict_time":0.036,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-163","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":917.0,"score":0.0,"rank":65.57,"hrank":65.27,"imp":29.572,"train_time":513.06,"predict_time":2.568,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-112","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":856.0,"score":0.0,"rank":67.43,"hrank":66.17,"imp":34.486,"train_time":1.21,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+158\/-266","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":572.0,"score":0.0,"rank":72.49,"hrank":72.46,"imp":46.783,"train_time":0.11,"predict_time":0.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-272","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/pareto_front_explorer.html
index 38429986c5ad78e12aae8481fe627d99c558efb8..eaed902f01d15f0965895084a463915edefb913b 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.754458801,"imp":16.8661638698,"elo":1425.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.01764933,"imp":8.2891264052,"elo":1761.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0196975147,"imp":8.9223628261,"elo":1740.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1663811446,"imp":8.6254985581,"elo":1754.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0233879248,"imp":11.8612633831,"elo":1617.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0265970701,"imp":11.1861216719,"elo":1655.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3315968945,"imp":10.5225022079,"elo":1720.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"imp":18.3734807884,"elo":1368.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0083401831,"imp":16.781436104,"elo":1408.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0872584538,"imp":15.7095805742,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0570892384,"imp":24.4306517864,"elo":1010.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0153132761,"imp":17.8275871335,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1561119171,"imp":16.5804781208,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1146303549,"imp":20.9390549114,"elo":1220.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1021965345,"imp":18.0452245989,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.385543961,"imp":14.7272167972,"elo":1426.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0329178148,"imp":15.8110309462,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0776085786,"imp":13.1055792349,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.6682734383,"imp":11.4198756165,"elo":1697.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.6994157224,"imp":22.5618336989,"elo":1266.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"imp":17.0831560703,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3490410688,"imp":13.1951996828,"elo":1557.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.4096341011,"imp":10.7090216199,"elo":1675.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.9998247538,"imp":26.2416785073,"elo":990.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0296504802,"imp":34.4859755804,"elo":855.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2581384468,"imp":26.1649335607,"elo":1042.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6407534322,"imp":25.0413536419,"elo":1077.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0137007533,"imp":24.9196784107,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0211930497,"imp":17.8834807855,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2259131636,"imp":17.0100195458,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2472765634,"imp":14.2382516089,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0879217199,"imp":10.5105993564,"elo":1650.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.053386569,"imp":9.1511055435,"elo":1745.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8963303312,"imp":12.1973167677,"elo":1589.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4652611394,"imp":12.4444200979,"elo":1608.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":15.1184451567,"imp":11.8937892778,"elo":1650.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.2398511784,"imp":18.6984349634,"elo":1310.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.8758317963,"imp":20.1484844796,"elo":1283.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.6414441586,"imp":16.11836212,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":475.7800751889,"imp":12.7728075069,"elo":1538.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2033080244,"imp":14.3714973635,"elo":1530.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":19.8176830022,"imp":3.020015071,"elo":2054.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0356431156,"imp":32.2515992944,"elo":936.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.5782128678,"imp":12.8687318099,"elo":1572.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1979712895,"imp":8.2697512852,"elo":1781.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0535231299,"imp":13.8042192486,"elo":1470.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0562011083,"imp":11.2477717069,"elo":1635.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5790353938,"imp":10.3518431999,"elo":1715.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8857412969,"imp":9.8284104744,"elo":1787.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3059932611,"imp":8.4484775587,"elo":1869.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0205027027,"imp":31.1278628307,"elo":967.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9483778428,"imp":25.6077503055,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3796059211,"imp":25.685200574,"elo":1131.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.154334046,"imp":14.8573655394,"elo":1455.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0463224066,"imp":18.0651304935,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0539785703,"imp":14.8048858983,"elo":1392.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5584373713,"imp":12.5825653736,"elo":1537.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"imp":14.168487644,"elo":1473.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0608575349,"imp":11.7747720598,"elo":1640.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3857911875,"imp":10.9496331211,"elo":1712.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":8.2311408315,"imp":19.5407183337,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.8813375711,"imp":12.7751500563,"elo":1567.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":80.7682550616,"imp":9.7844509775,"elo":1760.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1859407599,"imp":30.97120717,"elo":988.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1086014839,"imp":15.4647952687,"elo":1453.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.9349365505,"imp":13.7487303387,"elo":1485.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.754458801,"x_train":96.610665791,"imp":16.8661638698,"elo":1425.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.01764933,"x_train":1.3820827827,"imp":8.2891264052,"elo":1761.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0196975147,"x_train":329.0507531312,"imp":8.9223628261,"elo":1740.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1663811446,"x_train":329.0507531312,"imp":8.6254985581,"elo":1754.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0233879248,"x_train":0.4789788316,"imp":11.8612633831,"elo":1617.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0265970701,"x_train":95.9169155889,"imp":11.1861216719,"elo":1655.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3315968945,"x_train":95.9169155889,"imp":10.5225022079,"elo":1720.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"x_train":0.7112587015,"imp":18.3734807884,"elo":1368.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0083401831,"x_train":188.7956405878,"imp":16.781436104,"elo":1408.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0872584538,"x_train":188.7956405878,"imp":15.7095805742,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0570892384,"x_train":0.3400693672,"imp":24.4306517864,"elo":1010.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0153132761,"x_train":59.3011906074,"imp":17.8275871335,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1561119171,"x_train":59.3011906074,"imp":16.5804781208,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1146303549,"x_train":1.5179051995,"imp":20.9390549114,"elo":1220.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1021965345,"x_train":371.5880936781,"imp":18.0452245989,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.385543961,"x_train":371.5880936781,"imp":14.7272167972,"elo":1426.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0329178148,"x_train":0.2873206973,"imp":15.8110309462,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0776085786,"x_train":86.8470118165,"imp":13.1055792349,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.6682734383,"x_train":86.8470118165,"imp":11.4198756165,"elo":1697.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.6994157224,"x_train":5.1116723286,"imp":22.5618336989,"elo":1266.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"x_train":8.3438180493,"imp":17.0831560703,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3490410688,"x_train":2139.1684153663,"imp":13.1951996828,"elo":1557.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.4096341011,"x_train":2139.1684153663,"imp":10.7090216199,"elo":1675.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.9998247538,"x_train":17.6345194999,"imp":26.2416785073,"elo":990.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0296504802,"x_train":1.213148609,"imp":34.4859755804,"elo":855.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2581384468,"x_train":20.2619047675,"imp":26.1649335607,"elo":1042.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6407534322,"x_train":20.2619047675,"imp":25.0413536419,"elo":1077.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0137007533,"x_train":0.1197809845,"imp":24.9196784107,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0211930497,"x_train":42.9504760952,"imp":17.8834807855,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2259131636,"x_train":42.9504760952,"imp":17.0100195458,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2472765634,"x_train":6.412776503,"imp":14.2382516089,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0879217199,"x_train":1633.8040014148,"imp":10.5105993564,"elo":1650.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.053386569,"x_train":1633.8040014148,"imp":9.1511055435,"elo":1745.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8963303312,"x_train":2.2999820948,"imp":12.1973167677,"elo":1589.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4652611394,"x_train":842.247738377,"imp":12.4444200979,"elo":1608.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":15.1184451567,"x_train":842.247738377,"imp":11.8937892778,"elo":1650.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.2398511784,"x_train":16.4124685619,"imp":18.6984349634,"elo":1310.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.8758317963,"x_train":47.0281853011,"imp":20.1484844796,"elo":1283.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.6414441586,"x_train":4860.878886905,"imp":16.11836212,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":475.7800751889,"x_train":4860.878886905,"imp":12.7728075069,"elo":1538.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2033080244,"x_train":0.7584924197,"imp":14.3714973635,"elo":1530.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":19.8176830022,"x_train":42.1723102051,"imp":3.020015071,"elo":2054.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0356431156,"x_train":0.2354314672,"imp":32.2515992944,"elo":936.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.5782128678,"x_train":8.5347633004,"imp":12.8687318099,"elo":1572.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1979712895,"x_train":0.6451456984,"imp":8.2697512852,"elo":1781.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0535231299,"x_train":1.4982282715,"imp":13.8042192486,"elo":1470.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0562011083,"x_train":583.5491237683,"imp":11.2477717069,"elo":1635.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5790353938,"x_train":583.5491237683,"imp":10.3518431999,"elo":1715.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8857412969,"x_train":2.7558592937,"imp":9.8284104744,"elo":1787.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3059932611,"x_train":1.1838833522,"imp":8.4484775587,"elo":1869.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0205027027,"x_train":101.0759279672,"imp":31.1278628307,"elo":967.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9483778428,"x_train":3265.6782173608,"imp":25.6077503055,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3796059211,"x_train":3265.6782173608,"imp":25.685200574,"elo":1131.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.154334046,"x_train":0.6533526697,"imp":14.8573655394,"elo":1455.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0463224066,"x_train":1.8588422186,"imp":18.0651304935,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0539785703,"x_train":1156.6875333303,"imp":14.8048858983,"elo":1392.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5584373713,"x_train":1156.6875333303,"imp":12.5825653736,"elo":1537.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"x_train":0.4869354634,"imp":14.168487644,"elo":1473.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0608575349,"x_train":110.1339438717,"imp":11.7747720598,"elo":1640.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3857911875,"x_train":110.1339438717,"imp":10.9496331211,"elo":1712.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":8.2311408315,"x_train":82.933952473,"imp":19.5407183337,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.8813375711,"x_train":1883.8223518872,"imp":12.7751500563,"elo":1567.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":80.7682550616,"x_train":1883.8223518872,"imp":9.7844509775,"elo":1760.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1859407599,"x_train":2.6170575688,"imp":30.97120717,"elo":988.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1086014839,"x_train":572.8239309061,"imp":15.4647952687,"elo":1453.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.9349365505,"x_train":572.8239309061,"imp":13.7487303387,"elo":1485.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
index 73305d60787b3f260cf1b39bfa3a01e11c8d528e..3c7ad446a8c67e4c2065afec80ebf20f9e8bd8cd 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":96.610665791,"imp":16.8869456784,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.3820827827,"imp":8.3086430794,"elo":1811.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.6835494217,"imp":8.4706547398,"elo":1806.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.7546497969,"imp":8.5841259108,"elo":1814.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":23.6295991191,"imp":8.6424368764,"elo":1813.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":47.3749289529,"imp":8.5691655817,"elo":1822.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":86.7772971055,"imp":8.5420983708,"elo":1825.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":167.2712430131,"imp":8.5965913098,"elo":1817.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":246.0389923704,"imp":8.6173993693,"elo":1813.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":329.0507531312,"imp":8.6369523227,"elo":1809.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4789788316,"imp":11.8817963637,"elo":1662.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.9589536975,"imp":11.2914262643,"elo":1720.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.356217589,"imp":11.0046004786,"elo":1744.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.7221268742,"imp":10.8577960514,"elo":1757.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.9255203586,"imp":10.7741504301,"elo":1765.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":23.8768976028,"imp":10.691764247,"elo":1769.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":47.751833203,"imp":10.5870641658,"elo":1776.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":71.5860615637,"imp":10.546056945,"elo":1777.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":95.9169155889,"imp":10.5395976031,"elo":1777.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":0.7112587015,"imp":18.3887323699,"elo":1389.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":1.8058880969,"imp":17.5681117799,"elo":1422.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":4.573941659,"imp":16.9302588345,"elo":1451.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":9.1250887196,"imp":16.6193459022,"elo":1465.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":23.2616910123,"imp":16.3041625121,"elo":1483.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":47.2291260007,"imp":16.1193353693,"elo":1494.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":94.0295132726,"imp":15.9112062584,"elo":1507.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":141.071175821,"imp":15.8388892899,"elo":1511.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":188.7956405878,"imp":15.7173309074,"elo":1519.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3400693672,"imp":24.4462285072,"elo":1014.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.5790262037,"imp":22.0416278917,"elo":1087.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.5414359953,"imp":18.6787831286,"elo":1207.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9353651221,"imp":17.5257408655,"elo":1260.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.3111220646,"imp":16.855997351,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.5437419558,"imp":16.6465273424,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":29.1909327794,"imp":16.4874499806,"elo":1320.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.978271297,"imp":16.5043302134,"elo":1321.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":59.3011906074,"imp":16.5978412796,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.5179051995,"imp":20.9534199059,"elo":1248.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.2504528706,"imp":19.3327356804,"elo":1285.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":9.005873763,"imp":17.5571979893,"elo":1349.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":18.7663631809,"imp":16.7222137808,"elo":1380.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":44.7672715291,"imp":15.8951142548,"elo":1415.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":90.2243280036,"imp":15.4651613267,"elo":1436.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":183.9710430372,"imp":15.0996327621,"elo":1451.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":276.7808531494,"imp":14.8700614389,"elo":1464.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":371.5880936781,"imp":14.7354431653,"elo":1471.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.2873206973,"imp":15.8281297505,"elo":1387.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.7523280043,"imp":14.9384872067,"elo":1465.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":1.9490903848,"imp":13.7308752861,"elo":1577.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":4.0385354064,"imp":12.8004998388,"elo":1661.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":10.406685589,"imp":12.0514195339,"elo":1713.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":21.6557265527,"imp":11.735781379,"elo":1737.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":43.0306651896,"imp":11.5390381014,"elo":1747.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":64.0073656446,"imp":11.4725564601,"elo":1749.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":86.8470118165,"imp":11.4489706957,"elo":1752.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":5.1116723286,"imp":22.0964299162,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":8.3438180493,"imp":17.0999174747,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.3687146046,"imp":15.0875586421,"elo":1457.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.9177628832,"imp":12.6189562445,"elo":1615.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":112.4005567224,"imp":11.8503204406,"elo":1664.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":260.3290997978,"imp":11.1003331005,"elo":1715.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":532.4813413639,"imp":10.8147952016,"elo":1734.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1051.303381757,"imp":10.7709785782,"elo":1739.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1569.7102742529,"imp":10.7545797716,"elo":1737.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2139.1684153663,"imp":10.7426084688,"elo":1734.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":17.6345194999,"imp":26.2069426341,"elo":1006.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.213148609,"imp":34.9227236672,"elo":868.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.4317773957,"imp":32.6695542912,"elo":936.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":16.9493056602,"imp":25.175368626,"elo":1096.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.2619047675,"imp":25.0463843754,"elo":1104.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1197809845,"imp":24.9344281054,"elo":1001.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3353669751,"imp":20.1928378008,"elo":1163.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.0588012865,"imp":18.3428197494,"elo":1258.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":2.1551843555,"imp":17.7161521778,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":5.269352713,"imp":17.2837785806,"elo":1328.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":10.5968016873,"imp":17.1321851807,"elo":1340.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":21.2572302017,"imp":17.0271739624,"elo":1351.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":32.0070168617,"imp":16.9741463147,"elo":1356.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":42.9504760952,"imp":17.0267485259,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.412776503,"imp":14.2522040484,"elo":1468.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.4470659835,"imp":12.2716889306,"elo":1612.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.1186736736,"imp":11.2130847519,"elo":1687.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":79.8382984311,"imp":10.5621920308,"elo":1723.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":204.957931587,"imp":9.911741712,"elo":1763.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":413.9931126009,"imp":9.6468682449,"elo":1775.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":813.6028057178,"imp":9.3433666543,"elo":1791.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1220.4699358016,"imp":9.1915960063,"elo":1801.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1633.8040014148,"imp":9.168219635,"elo":1806.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2.2999820948,"imp":12.21913411,"elo":1628.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.6597534236,"imp":12.1879102238,"elo":1636.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":21.5428942424,"imp":12.1199442425,"elo":1647.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":39.0796097222,"imp":12.0869199997,"elo":1654.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":88.9628219424,"imp":11.9639017638,"elo":1678.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":188.8767701417,"imp":11.9693640401,"elo":1691.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":396.9834212212,"imp":11.961761151,"elo":1698.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":604.8684512841,"imp":11.9607451636,"elo":1697.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":842.247738377,"imp":11.9072122952,"elo":1695.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":16.4124685619,"imp":18.7099466002,"elo":1341.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":47.0281853011,"imp":20.1619369601,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.0348846629,"imp":18.390424005,"elo":1358.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":127.5381763694,"imp":16.6621751076,"elo":1425.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":230.7293949461,"imp":16.0702194481,"elo":1448.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":589.1524400719,"imp":14.7273331995,"elo":1512.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1198.3439089076,"imp":13.9901152254,"elo":1547.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2408.9332260936,"imp":13.5007187905,"elo":1566.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3630.8641964758,"imp":13.0556520127,"elo":1585.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4860.878886905,"imp":12.786090517,"elo":1590.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.7584924197,"imp":14.3878981343,"elo":1573.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":42.1723102051,"imp":3.0198899691,"elo":2095.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.2354314672,"imp":32.2629483255,"elo":947.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":8.5347633004,"imp":12.8892694171,"elo":1620.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6451456984,"imp":8.2928759373,"elo":1827.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.4982282715,"imp":13.8226609481,"elo":1506.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":3.9284284946,"imp":13.0585880383,"elo":1551.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":12.4876393146,"imp":12.1100926661,"elo":1617.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":27.3834783781,"imp":11.5365261569,"elo":1667.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":71.1997682157,"imp":10.9618621861,"elo":1706.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":143.8112757102,"imp":10.6986168262,"elo":1731.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":285.6452634754,"imp":10.4860402965,"elo":1748.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":432.2819705079,"imp":10.4082446029,"elo":1756.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":583.5491237683,"imp":10.3634823237,"elo":1763.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.7558592937,"imp":9.8498615705,"elo":1840.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.1838833522,"imp":8.4535084303,"elo":1915.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":101.0759279672,"imp":31.1369080501,"elo":973.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":111.2079544113,"imp":31.05854612,"elo":988.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":141.6266688837,"imp":30.8939189678,"elo":996.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":187.760864208,"imp":30.4005168147,"elo":1017.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":414.9961090703,"imp":29.3533425922,"elo":1047.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":831.557341273,"imp":28.0743782128,"elo":1081.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1652.4911170928,"imp":27.2274709966,"elo":1111.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2441.373538211,"imp":26.4148787287,"elo":1140.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3265.6782173608,"imp":25.6967661148,"elo":1171.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.6533526697,"imp":14.8784132282,"elo":1495.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":1.8588422186,"imp":18.0767520838,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.7947477355,"imp":16.6067405067,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":26.9428583764,"imp":15.2096234808,"elo":1408.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":62.6027434746,"imp":14.5141871393,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":150.4831797618,"imp":13.7553693218,"elo":1508.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":283.6255748988,"imp":13.2666108735,"elo":1542.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":555.0904532489,"imp":12.9692715177,"elo":1566.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":859.203668693,"imp":12.761614669,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1156.6875333303,"imp":12.6027470128,"elo":1587.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4869354634,"imp":14.1881619037,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":1.1317321011,"imp":12.5919116327,"elo":1644.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":2.9964300137,"imp":11.9134244829,"elo":1706.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":5.3381407276,"imp":11.4190066821,"elo":1736.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":13.9223759351,"imp":11.1330668545,"elo":1751.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":27.5993202576,"imp":11.0312057564,"elo":1757.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":55.0233345183,"imp":10.9726161464,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":81.9474032161,"imp":10.9693188118,"elo":1765.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":110.1339438717,"imp":10.9720087259,"elo":1767.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":82.933952473,"imp":19.5561303695,"elo":1369.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":155.9256875342,"imp":17.2867604146,"elo":1427.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":374.1243102893,"imp":14.4138746108,"elo":1550.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":733.7366585123,"imp":12.2405692825,"elo":1665.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1813.3707967819,"imp":9.8456285395,"elo":1814.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1883.8223518872,"imp":9.8041530963,"elo":1820.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.6170575688,"imp":30.9829138539,"elo":993.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.9970579818,"imp":24.2891420854,"elo":1112.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":14.4751935279,"imp":17.1529800491,"elo":1327.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":28.4433104494,"imp":14.6789332659,"elo":1478.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":71.1413647457,"imp":13.8532051581,"elo":1532.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":142.1475184089,"imp":13.571857429,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":285.1331865035,"imp":13.5502598296,"elo":1546.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":427.4360460386,"imp":13.6114489793,"elo":1541.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":572.8239309061,"imp":13.7136725988,"elo":1533.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.754458801,"x_train":96.610665791,"imp":16.8869456784,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.01764933,"x_train":1.3820827827,"imp":8.3086430794,"elo":1811.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0357929968,"x_train":3.6835494217,"imp":8.4706547398,"elo":1806.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0590771616,"x_train":12.7546497969,"imp":8.5841259108,"elo":1814.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0772621764,"x_train":23.6295991191,"imp":8.6424368764,"elo":1813.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0947269938,"x_train":47.3749289529,"imp":8.5691655817,"elo":1822.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1111539594,"x_train":86.7772971055,"imp":8.5420983708,"elo":1825.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1397193986,"x_train":167.2712430131,"imp":8.5965913098,"elo":1817.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1540251012,"x_train":246.0389923704,"imp":8.6173993693,"elo":1813.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1663811446,"x_train":329.0507531312,"imp":8.6369523227,"elo":1809.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0233879248,"x_train":0.4789788316,"imp":11.8817963637,"elo":1662.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0478671165,"x_train":0.9589536975,"imp":11.2914262643,"elo":1720.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1129020181,"x_train":2.356217589,"imp":11.0046004786,"elo":1744.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1771034691,"x_train":4.7221268742,"imp":10.8577960514,"elo":1757.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2454903367,"x_train":11.9255203586,"imp":10.7741504301,"elo":1765.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2926890222,"x_train":23.8768976028,"imp":10.691764247,"elo":1769.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.312025413,"x_train":47.751833203,"imp":10.5870641658,"elo":1776.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3237207716,"x_train":71.5860615637,"imp":10.546056945,"elo":1777.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3206222047,"x_train":95.9169155889,"imp":10.5395976031,"elo":1777.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0054230738,"x_train":0.7112587015,"imp":18.3887323699,"elo":1389.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0116608277,"x_train":1.8058880969,"imp":17.5681117799,"elo":1422.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0226990233,"x_train":4.573941659,"imp":16.9302588345,"elo":1451.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333135949,"x_train":9.1250887196,"imp":16.6193459022,"elo":1465.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0501304583,"x_train":23.2616910123,"imp":16.3041625121,"elo":1483.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0610135476,"x_train":47.2291260007,"imp":16.1193353693,"elo":1494.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0728842068,"x_train":94.0295132726,"imp":15.9112062584,"elo":1507.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0813406855,"x_train":141.071175821,"imp":15.8388892899,"elo":1511.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0885616373,"x_train":188.7956405878,"imp":15.7173309074,"elo":1519.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0570892384,"x_train":0.3400693672,"imp":24.4462285072,"elo":1014.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0703457475,"x_train":0.5790262037,"imp":22.0416278917,"elo":1087.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0928856349,"x_train":1.5414359953,"imp":18.6787831286,"elo":1207.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1010934443,"x_train":2.9353651221,"imp":17.5257408655,"elo":1260.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1082818488,"x_train":7.3111220646,"imp":16.855997351,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1223033865,"x_train":14.5437419558,"imp":16.6465273424,"elo":1308.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1386914014,"x_train":29.1909327794,"imp":16.4874499806,"elo":1320.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1459534487,"x_train":43.978271297,"imp":16.5043302134,"elo":1321.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1524033833,"x_train":59.3011906074,"imp":16.5978412796,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1146303549,"x_train":1.5179051995,"imp":20.9534199059,"elo":1248.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.224806949,"x_train":3.2504528706,"imp":19.3327356804,"elo":1285.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5565998143,"x_train":9.005873763,"imp":17.5571979893,"elo":1349.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9690827943,"x_train":18.7663631809,"imp":16.7222137808,"elo":1380.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3936753554,"x_train":44.7672715291,"imp":15.8951142548,"elo":1415.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6932007533,"x_train":90.2243280036,"imp":15.4651613267,"elo":1436.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0269078994,"x_train":183.9710430372,"imp":15.0996327621,"elo":1451.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.2179323369,"x_train":276.7808531494,"imp":14.8700614389,"elo":1464.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.3776360952,"x_train":371.5880936781,"imp":14.7354431653,"elo":1471.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0329178148,"x_train":0.2873206973,"imp":15.8281297505,"elo":1387.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.3022864661,"x_train":0.7523280043,"imp":14.9384872067,"elo":1465.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5169466094,"x_train":1.9490903848,"imp":13.7308752861,"elo":1577.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7806727427,"x_train":4.0385354064,"imp":12.8004998388,"elo":1661.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7659565508,"x_train":10.406685589,"imp":12.0514195339,"elo":1713.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7608339553,"x_train":21.6557265527,"imp":11.735781379,"elo":1737.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6794432016,"x_train":43.0306651896,"imp":11.5390381014,"elo":1747.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.642292477,"x_train":64.0073656446,"imp":11.4725564601,"elo":1749.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6494090854,"x_train":86.8470118165,"imp":11.4489706957,"elo":1752.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":2.6994157224,"x_train":5.1116723286,"imp":22.0964299162,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2496013467,"x_train":8.3438180493,"imp":17.0999174747,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5785744608,"x_train":25.3687146046,"imp":15.0875586421,"elo":1457.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.2947480655,"x_train":60.9177628832,"imp":12.6189562445,"elo":1615.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1427076773,"x_train":112.4005567224,"imp":11.8503204406,"elo":1664.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.2223339105,"x_train":260.3290997978,"imp":11.1003331005,"elo":1715.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.4032452473,"x_train":532.4813413639,"imp":10.8147952016,"elo":1734.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5772898048,"x_train":1051.303381757,"imp":10.7709785782,"elo":1739.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.1475244152,"x_train":1569.7102742529,"imp":10.7545797716,"elo":1737.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3256045535,"x_train":2139.1684153663,"imp":10.7426084688,"elo":1734.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.9998247538,"x_train":17.6345194999,"imp":26.2069426341,"elo":1006.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0296504802,"x_train":1.213148609,"imp":34.9227236672,"elo":868.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2293591891,"x_train":6.4317773957,"imp":32.6695542912,"elo":936.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5854122012,"x_train":16.9493056602,"imp":25.175368626,"elo":1096.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6407534322,"x_train":20.2619047675,"imp":25.0463843754,"elo":1104.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0137007533,"x_train":0.1197809845,"imp":24.9344281054,"elo":1001.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0410613688,"x_train":0.3353669751,"imp":20.1928378008,"elo":1163.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0866608789,"x_train":1.0588012865,"imp":18.3428197494,"elo":1258.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1305578206,"x_train":2.1551843555,"imp":17.7161521778,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1467222069,"x_train":5.269352713,"imp":17.2837785806,"elo":1328.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1724734092,"x_train":10.5968016873,"imp":17.1321851807,"elo":1340.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2013078773,"x_train":21.2572302017,"imp":17.0271739624,"elo":1351.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2097607806,"x_train":32.0070168617,"imp":16.9741463147,"elo":1356.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2232886539,"x_train":42.9504760952,"imp":17.0267485259,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2472765634,"x_train":6.412776503,"imp":14.2522040484,"elo":1468.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4377557757,"x_train":15.4470659835,"imp":12.2716889306,"elo":1612.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.6660949816,"x_train":39.1186736736,"imp":11.2130847519,"elo":1687.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.8989691909,"x_train":79.8382984311,"imp":10.5621920308,"elo":1723.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1399419366,"x_train":204.957931587,"imp":9.911741712,"elo":1763.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.3802752953,"x_train":413.9931126009,"imp":9.6468682449,"elo":1775.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6732415593,"x_train":813.6028057178,"imp":9.3433666543,"elo":1791.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.9229356811,"x_train":1220.4699358016,"imp":9.1915960063,"elo":1801.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.0414922251,"x_train":1633.8040014148,"imp":9.168219635,"elo":1806.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8963303312,"x_train":2.2999820948,"imp":12.21913411,"elo":1628.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.927863305,"x_train":6.6597534236,"imp":12.1879102238,"elo":1636.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.2372972886,"x_train":21.5428942424,"imp":12.1199442425,"elo":1647.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.3430098245,"x_train":39.0796097222,"imp":12.0869199997,"elo":1654.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1180506729,"x_train":88.9628219424,"imp":11.9639017638,"elo":1678.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.5551205244,"x_train":188.8767701417,"imp":11.9693640401,"elo":1691.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.3131673972,"x_train":396.9834212212,"imp":11.961761151,"elo":1698.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":13.5329260857,"x_train":604.8684512841,"imp":11.9607451636,"elo":1697.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":14.6273407585,"x_train":842.247738377,"imp":11.9072122952,"elo":1695.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.2398511784,"x_train":16.4124685619,"imp":18.7099466002,"elo":1341.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":45.8758317963,"x_train":47.0281853011,"imp":20.1619369601,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":66.3525228473,"x_train":70.0348846629,"imp":18.390424005,"elo":1358.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":98.2456271023,"x_train":127.5381763694,"imp":16.6621751076,"elo":1425.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":142.0328330544,"x_train":230.7293949461,"imp":16.0702194481,"elo":1448.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":218.1812948024,"x_train":589.1524400719,"imp":14.7273331995,"elo":1512.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":285.9879390033,"x_train":1198.3439089076,"imp":13.9901152254,"elo":1547.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":390.5997136252,"x_train":2408.9332260936,"imp":13.5007187905,"elo":1566.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":442.8772138735,"x_train":3630.8641964758,"imp":13.0556520127,"elo":1585.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":481.8516025188,"x_train":4860.878886905,"imp":12.786090517,"elo":1590.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2033080244,"x_train":0.7584924197,"imp":14.3878981343,"elo":1573.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":19.8176830022,"x_train":42.1723102051,"imp":3.0198899691,"elo":2095.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0356431156,"x_train":0.2354314672,"imp":32.2629483255,"elo":947.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":3.5782128678,"x_train":8.5347633004,"imp":12.8892694171,"elo":1620.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1979712895,"x_train":0.6451456984,"imp":8.2928759373,"elo":1827.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0535231299,"x_train":1.4982282715,"imp":13.8226609481,"elo":1506.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1154003256,"x_train":3.9284284946,"imp":13.0585880383,"elo":1551.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2000419921,"x_train":12.4876393146,"imp":12.1100926661,"elo":1617.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2767821239,"x_train":27.3834783781,"imp":11.5365261569,"elo":1667.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.36488929,"x_train":71.1997682157,"imp":10.9618621861,"elo":1706.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4641659542,"x_train":143.8112757102,"imp":10.6986168262,"elo":1731.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.5371776272,"x_train":285.6452634754,"imp":10.4860402965,"elo":1748.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.5567799537,"x_train":432.2819705079,"imp":10.4082446029,"elo":1756.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5682121276,"x_train":583.5491237683,"imp":10.3634823237,"elo":1763.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8857412969,"x_train":2.7558592937,"imp":9.8498615705,"elo":1840.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3059932611,"x_train":1.1838833522,"imp":8.4535084303,"elo":1915.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0205027027,"x_train":101.0759279672,"imp":31.1369080501,"elo":973.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0983982624,"x_train":111.2079544113,"imp":31.05854612,"elo":988.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.2224011522,"x_train":141.6266688837,"imp":30.8939189678,"elo":996.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5850970839,"x_train":187.760864208,"imp":30.4005168147,"elo":1017.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.2850393528,"x_train":414.9961090703,"imp":29.3533425922,"elo":1047.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4241099579,"x_train":831.557341273,"imp":28.0743782128,"elo":1081.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.4436810211,"x_train":1652.4911170928,"imp":27.2274709966,"elo":1111.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.0366030881,"x_train":2441.373538211,"imp":26.4148787287,"elo":1140.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.365156164,"x_train":3265.6782173608,"imp":25.6967661148,"elo":1171.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.154334046,"x_train":0.6533526697,"imp":14.8784132282,"elo":1495.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0463224066,"x_train":1.8588422186,"imp":18.0767520838,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1000380446,"x_train":6.7947477355,"imp":16.6067405067,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.262302914,"x_train":26.9428583764,"imp":15.2096234808,"elo":1408.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4175923212,"x_train":62.6027434746,"imp":14.5141871393,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6171906817,"x_train":150.4831797618,"imp":13.7553693218,"elo":1508.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9111530149,"x_train":283.6255748988,"imp":13.2666108735,"elo":1542.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1543711789,"x_train":555.0904532489,"imp":12.9692715177,"elo":1566.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3269752577,"x_train":859.203668693,"imp":12.761614669,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5455206258,"x_train":1156.6875333303,"imp":12.6027470128,"elo":1587.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0534163689,"x_train":0.4869354634,"imp":14.1881619037,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1297403015,"x_train":1.1317321011,"imp":12.5919116327,"elo":1644.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2494283602,"x_train":2.9964300137,"imp":11.9134244829,"elo":1706.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2835102125,"x_train":5.3381407276,"imp":11.4190066821,"elo":1736.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3287421252,"x_train":13.9223759351,"imp":11.1330668545,"elo":1751.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3407457713,"x_train":27.5993202576,"imp":11.0312057564,"elo":1757.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3388091225,"x_train":55.0233345183,"imp":10.9726161464,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3715591178,"x_train":81.9474032161,"imp":10.9693188118,"elo":1765.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3857911875,"x_train":110.1339438717,"imp":10.9720087259,"elo":1767.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":8.2311408315,"x_train":82.933952473,"imp":19.5561303695,"elo":1369.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":17.9776071875,"x_train":155.9256875342,"imp":17.2867604146,"elo":1427.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":33.2102432394,"x_train":374.1243102893,"imp":14.4138746108,"elo":1550.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":49.0550574883,"x_train":733.7366585123,"imp":12.2405692825,"elo":1665.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":80.6306891729,"x_train":1813.3707967819,"imp":9.8456285395,"elo":1814.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":81.4550745619,"x_train":1883.8223518872,"imp":9.8041530963,"elo":1820.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1859407599,"x_train":2.6170575688,"imp":30.9829138539,"elo":993.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.292827281,"x_train":5.9970579818,"imp":24.2891420854,"elo":1112.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.4370737653,"x_train":14.4751935279,"imp":17.1529800491,"elo":1327.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.6030397775,"x_train":28.4433104494,"imp":14.6789332659,"elo":1478.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.893684495,"x_train":71.1413647457,"imp":13.8532051581,"elo":1532.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.1268006825,"x_train":142.1475184089,"imp":13.571857429,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.4622467576,"x_train":285.1331865035,"imp":13.5502598296,"elo":1546.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.7229354568,"x_train":427.4360460386,"imp":13.6114489793,"elo":1541.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9204374353,"x_train":572.8239309061,"imp":13.7136725988,"elo":1533.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..309516737ea60938bf9955c5e3ee5293355c1813
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_binary/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "TabFM (default)", "AutoGluon 1.4 (best, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "CatBoost (default)", "iLTM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "RealMLP (tuned + ensembled)", "CatBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "TabM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "LightGBM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "ChimeraBoost (tuned)", "RealMLP (tuned)", "RealTabPFN-2.5 (tuned + ensembled)", "XGBoost (tuned)", "TabM (tuned)", "ChimeraBoost (default)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (default)", "LightGBM (tuned)", "TabICL (default)", "iLTM (tuned)", "ModernNCA (tuned)", "TabDPT (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "xRFM (tuned + ensembled)", "EBM (tuned + ensembled)", "XGBoost (default)", "TabM (default)", "TabSwift (default)", "xRFM (tuned)", "RealMLP (default)", "FastaiMLP (tuned + ensembled)", "BetaTabPFN (default)", "EBM (tuned)", "TorchMLP (tuned)", "TabDPT (tuned)", "EBM (default)", "LightGBM (default)", "iLTM (default)", "ModernNCA (default)", "RandomForest (tuned + ensembled)", "SAP-RPT-OSS (default)", "ExtraTrees (tuned + ensembled)", "RandomForest (tuned)", "FastaiMLP (tuned)", "TabDPT (default)", "LimiX (default)", "TorchMLP (default)", "ExtraTrees (tuned)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "KNN (tuned + ensembled)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "PerpetualBooster (tuned)", "ExtraTrees (default)", "Linear (default)", "RandomForest (default)", "OrionMSP (default)", "xRFM (default)", "TabSTAR (default)", "TabFlex (default)", "KNN (tuned)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.3333333432674408, 0.7160493731498718, 0.45679011940956116, 0.8024691343307495, 0.6419752836227417, 0.8888888955116272, 0.9259259104728699, 0.9135802388191223, 0.9753086566925049, 0.9012345671653748, 0.9135802388191223, 0.790123462677002, 0.9506173133850098, 0.9259259104728699, 0.9629629850387573, 0.9259259104728699, 0.9629629850387573, 0.9135802388191223, 0.9876543283462524, 0.8518518805503845, 0.9629629850387573, 0.9135802388191223, 0.9135802388191223, 0.9753086566925049, 0.8888888955116272, 0.8765432238578796, 0.9876543283462524, 0.9135802388191223, 0.9876543283462524, 0.9012345671653748, 0.9753086566925049, 0.9876543283462524, 0.9629629850387573, 1.0, 0.9382715821266174, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9506173133850098, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9876543283462524, 1.0, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.6666666865348816, NaN, 0.6790123581886292, 0.9012345671653748, 0.9506173133850098, 0.9012345671653748, 0.7654321193695068, 0.8271604776382446, 0.7530864477157593, 0.8024691343307495, 0.8024691343307495, 0.8641975522041321, 0.9876543283462524, 0.8641975522041321, 0.8395061492919922, 0.8765432238578796, 0.8641975522041321, 0.8148148059844971, 0.9629629850387573, 0.9012345671653748, 1.0, 0.8888888955116272, 0.9629629850387573, 0.9753086566925049, 0.9629629850387573, 0.9506173133850098, 0.8888888955116272, 0.9382715821266174, 0.9629629850387573, 0.9629629850387573, 0.9629629850387573, 0.8765432238578796, 0.9876543283462524, 0.9753086566925049, 1.0, 0.9506173133850098, 0.9012345671653748, 1.0, 0.9012345671653748, 1.0, 1.0, 1.0, 0.9629629850387573, 1.0, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9753086566925049, 0.9506173133850098, 1.0, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9506173133850098, 1.0, 1.0, 0.9753086566925049, 1.0, 1.0, 0.9876543283462524, 1.0, 0.9876543283462524, 1.0, 0.9506173133850098, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2839506268501282, 0.32098764181137085, NaN, 0.4691357910633087, 0.6790123581886292, 0.6172839403152466, 0.6913580298423767, 0.8271604776382446, 0.7407407164573669, 0.8518518805503845, 0.7530864477157593, 0.7530864477157593, 0.7160493731498718, 0.8518518805503845, 0.8271604776382446, 0.7654321193695068, 0.8518518805503845, 0.9135802388191223, 0.7037037014961243, 0.8765432238578796, 0.7777777910232544, 0.8765432238578796, 0.7530864477157593, 0.7407407164573669, 0.8888888955116272, 0.7407407164573669, 0.8395061492919922, 0.8888888955116272, 0.7777777910232544, 0.9012345671653748, 0.790123462677002, 0.9506173133850098, 0.9506173133850098, 0.9135802388191223, 0.9135802388191223, 0.8271604776382446, 0.9629629850387573, 0.9629629850387573, 0.9629629850387573, 0.9012345671653748, 0.9506173133850098, 0.9753086566925049, 0.9629629850387573, 0.9506173133850098, 0.9506173133850098, 0.9629629850387573, 0.9876543283462524, 0.9876543283462524, 0.9629629850387573, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9506173133850098, 0.9753086566925049, 0.9876543283462524, 1.0, 0.9876543283462524, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9629629850387573, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5432098507881165, 0.09876543283462524, 0.5308641791343689, NaN, 0.7283950448036194, 0.5308641791343689, 0.654321014881134, 0.6790123581886292, 0.6666666865348816, 0.7160493731498718, 0.654321014881134, 0.7530864477157593, 0.7654321193695068, 0.7160493731498718, 0.7037037014961243, 0.8765432238578796, 0.7407407164573669, 0.7283950448036194, 0.8518518805503845, 0.7530864477157593, 0.8395061492919922, 0.7407407164573669, 0.8641975522041321, 0.8641975522041321, 0.8888888955116272, 0.8518518805503845, 0.8518518805503845, 0.8641975522041321, 0.8765432238578796, 0.8148148059844971, 0.9506173133850098, 0.7654321193695068, 0.8518518805503845, 0.9012345671653748, 0.9876543283462524, 0.9629629850387573, 0.7777777910232544, 0.9382715821266174, 0.8888888955116272, 0.9382715821266174, 0.9012345671653748, 0.9629629850387573, 0.9629629850387573, 0.9506173133850098, 0.9876543283462524, 0.9876543283462524, 0.9753086566925049, 0.9382715821266174, 0.9135802388191223, 0.9629629850387573, 0.9629629850387573, 0.9259259104728699, 0.9629629850387573, 0.8888888955116272, 0.9629629850387573, 0.9629629850387573, 0.9259259104728699, 0.9135802388191223, 0.9259259104728699, 1.0, 1.0, 0.9259259104728699, 1.0, 1.0, 1.0, 0.9506173133850098, 1.0, 1.0, 0.9135802388191223, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1975308656692505, 0.04938271641731262, 0.32098764181137085, 0.2716049253940582, NaN, 0.5185185074806213, 0.4691357910633087, 0.5061728358268738, 0.5185185074806213, 0.5802469253540039, 0.5185185074806213, 0.6419752836227417, 0.5802469253540039, 0.5308641791343689, 0.654321014881134, 0.6666666865348816, 0.6790123581886292, 0.6790123581886292, 0.7777777910232544, 0.7037037014961243, 0.7160493731498718, 0.6419752836227417, 0.7777777910232544, 0.8395061492919922, 0.8024691343307495, 0.7530864477157593, 0.7283950448036194, 0.8024691343307495, 0.8395061492919922, 0.8271604776382446, 0.8395061492919922, 0.7407407164573669, 0.8641975522041321, 0.9012345671653748, 0.9135802388191223, 0.8518518805503845, 0.790123462677002, 1.0, 0.8518518805503845, 0.8888888955116272, 0.9629629850387573, 0.9506173133850098, 0.9506173133850098, 0.9629629850387573, 0.9382715821266174, 0.9382715821266174, 0.9753086566925049, 0.9506173133850098, 0.9506173133850098, 0.9753086566925049, 0.9876543283462524, 0.9382715821266174, 0.9382715821266174, 0.8641975522041321, 1.0, 0.9876543283462524, 0.9135802388191223, 0.9629629850387573, 0.9382715821266174, 1.0, 1.0, 0.9629629850387573, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9629629850387573, 1.0, 1.0, 0.9135802388191223, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3580246865749359, 0.09876543283462524, 0.3827160596847534, 0.4691357910633087, 0.48148149251937866, NaN, 0.5679012537002563, 0.5925925970077515, 0.5679012537002563, 0.5308641791343689, 0.5679012537002563, 0.5802469253540039, 0.5679012537002563, 0.5802469253540039, 0.654321014881134, 0.6790123581886292, 0.5802469253540039, 0.5925925970077515, 0.8148148059844971, 0.604938268661499, 0.6913580298423767, 0.604938268661499, 0.8395061492919922, 0.8271604776382446, 0.7530864477157593, 0.8148148059844971, 0.7283950448036194, 0.7777777910232544, 0.8888888955116272, 0.790123462677002, 0.8641975522041321, 0.6666666865348816, 0.7530864477157593, 0.8518518805503845, 0.9382715821266174, 0.9012345671653748, 0.6913580298423767, 0.9012345671653748, 0.8271604776382446, 0.9135802388191223, 0.9012345671653748, 0.9012345671653748, 0.9382715821266174, 0.9259259104728699, 0.9382715821266174, 0.9135802388191223, 0.9506173133850098, 0.9629629850387573, 0.9135802388191223, 0.9876543283462524, 0.9629629850387573, 0.9259259104728699, 0.9506173133850098, 0.9012345671653748, 0.9753086566925049, 0.9876543283462524, 0.9135802388191223, 0.9135802388191223, 0.9506173133850098, 0.9753086566925049, 0.9753086566925049, 0.9382715821266174, 1.0, 1.0, 1.0, 0.9629629850387573, 1.0, 1.0, 0.9135802388191223, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.23456789553165436, 0.3086419701576233, 0.34567901492118835, 0.5308641791343689, 0.43209877610206604, NaN, 0.5308641791343689, 0.45679011940956116, 0.5802469253540039, 0.5308641791343689, 0.5925925970077515, 0.5555555820465088, 0.5925925970077515, 0.604938268661499, 0.5925925970077515, 0.7160493731498718, 0.6419752836227417, 0.5802469253540039, 0.7283950448036194, 0.6172839403152466, 0.790123462677002, 0.604938268661499, 0.7160493731498718, 0.7654321193695068, 0.7037037014961243, 0.7283950448036194, 0.6913580298423767, 0.654321014881134, 0.7654321193695068, 0.654321014881134, 0.9382715821266174, 0.9012345671653748, 0.8765432238578796, 0.7654321193695068, 0.790123462677002, 0.9506173133850098, 0.8024691343307495, 0.8148148059844971, 0.8395061492919922, 0.9012345671653748, 0.9506173133850098, 0.790123462677002, 0.9259259104728699, 0.9259259104728699, 0.9382715821266174, 0.9876543283462524, 0.9753086566925049, 0.9506173133850098, 0.9629629850387573, 0.9629629850387573, 0.9753086566925049, 0.8888888955116272, 0.8395061492919922, 0.9753086566925049, 0.9876543283462524, 0.9506173133850098, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9876543283462524, 0.9876543283462524, 0.9753086566925049, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.07407407462596893, 0.17283950746059418, 0.17283950746059418, 0.32098764181137085, 0.4938271641731262, 0.40740740299224854, 0.4691357910633087, NaN, 0.45679011940956116, 0.4938271641731262, 0.4691357910633087, 0.4938271641731262, 0.5308641791343689, 0.5925925970077515, 0.6296296119689941, 0.5679012537002563, 0.5802469253540039, 0.6172839403152466, 0.6419752836227417, 0.7530864477157593, 0.6172839403152466, 0.604938268661499, 0.6913580298423767, 0.7407407164573669, 0.7530864477157593, 0.7160493731498718, 0.6790123581886292, 0.7654321193695068, 0.7160493731498718, 0.8518518805503845, 0.6790123581886292, 0.8765432238578796, 0.9012345671653748, 0.8765432238578796, 0.8148148059844971, 0.7407407164573669, 0.9506173133850098, 0.9506173133850098, 0.9629629850387573, 0.8518518805503845, 0.9135802388191223, 0.9506173133850098, 0.9012345671653748, 0.9382715821266174, 0.9135802388191223, 1.0, 0.9753086566925049, 0.9753086566925049, 0.9753086566925049, 0.9506173133850098, 0.9876543283462524, 0.9629629850387573, 0.9259259104728699, 0.8395061492919922, 1.0, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9629629850387573, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.08641975373029709, 0.2469135820865631, 0.25925925374031067, 0.3333333432674408, 0.48148149251937866, 0.43209877610206604, 0.5432098507881165, 0.5432098507881165, NaN, 0.5802469253540039, 0.5802469253540039, 0.5432098507881165, 0.5555555820465088, 0.6296296119689941, 0.5679012537002563, 0.5432098507881165, 0.7037037014961243, 0.6790123581886292, 0.5802469253540039, 0.7037037014961243, 0.604938268661499, 0.7654321193695068, 0.5925925970077515, 0.7283950448036194, 0.7037037014961243, 0.654321014881134, 0.7160493731498718, 0.709876537322998, 0.6296296119689941, 0.7654321193695068, 0.6296296119689941, 0.9135802388191223, 0.9259259104728699, 0.8518518805503845, 0.7407407164573669, 0.7530864477157593, 0.9382715821266174, 0.8148148059844971, 0.8518518805503845, 0.8271604776382446, 0.9012345671653748, 0.9506173133850098, 0.790123462677002, 0.9382715821266174, 0.9506173133850098, 0.9135802388191223, 0.9506173133850098, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9629629850387573, 0.9629629850387573, 0.8765432238578796, 0.8024691343307495, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0], [0.02469135820865631, 0.1975308656692505, 0.14814814925193787, 0.2839506268501282, 0.4197530746459961, 0.4691357910633087, 0.4197530746459961, 0.5061728358268738, 0.4197530746459961, NaN, 0.4691357910633087, 0.4197530746459961, 0.4691357910633087, 0.5555555820465088, 0.5802469253540039, 0.5925925970077515, 0.5432098507881165, 0.7530864477157593, 0.6172839403152466, 0.6666666865348816, 0.5308641791343689, 0.5432098507881165, 0.654321014881134, 0.6790123581886292, 0.7530864477157593, 0.6419752836227417, 0.6913580298423767, 0.7654321193695068, 0.6790123581886292, 0.8395061492919922, 0.7407407164573669, 0.8271604776382446, 0.8395061492919922, 0.8271604776382446, 0.8888888955116272, 0.7283950448036194, 0.8765432238578796, 1.0, 0.9753086566925049, 0.8395061492919922, 0.9135802388191223, 0.9259259104728699, 0.9012345671653748, 0.9382715821266174, 0.9506173133850098, 0.9012345671653748, 0.9506173133850098, 0.9753086566925049, 0.9629629850387573, 1.0, 0.9629629850387573, 0.9753086566925049, 0.9012345671653748, 0.9382715821266174, 0.9876543283462524, 0.9753086566925049, 0.9629629850387573, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.09876543283462524, 0.1975308656692505, 0.2469135820865631, 0.34567901492118835, 0.48148149251937866, 0.43209877610206604, 0.4691357910633087, 0.5308641791343689, 0.4197530746459961, 0.5308641791343689, NaN, 0.5802469253540039, 0.5679012537002563, 0.6296296119689941, 0.5185185074806213, 0.5555555820465088, 0.7037037014961243, 0.6666666865348816, 0.5432098507881165, 0.7283950448036194, 0.5802469253540039, 0.7283950448036194, 0.5925925970077515, 0.6790123581886292, 0.7160493731498718, 0.6666666865348816, 0.6790123581886292, 0.654321014881134, 0.6296296119689941, 0.7530864477157593, 0.6172839403152466, 0.9259259104728699, 0.9135802388191223, 0.8395061492919922, 0.7777777910232544, 0.7407407164573669, 0.9259259104728699, 0.8271604776382446, 0.8024691343307495, 0.8395061492919922, 0.9012345671653748, 0.9382715821266174, 0.7777777910232544, 0.9629629850387573, 0.9135802388191223, 0.9259259104728699, 0.9259259104728699, 0.9753086566925049, 0.9629629850387573, 0.9506173133850098, 0.9629629850387573, 0.9506173133850098, 0.8641975522041321, 0.8024691343307495, 0.9753086566925049, 0.9629629850387573, 0.9382715821266174, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9753086566925049, 0.9876543283462524, 0.9753086566925049, 1.0, 0.9876543283462524, 0.9753086566925049, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.08641975373029709, 0.1358024626970291, 0.2469135820865631, 0.2469135820865631, 0.3580246865749359, 0.4197530746459961, 0.40740740299224854, 0.5061728358268738, 0.45679011940956116, 0.5802469253540039, 0.4197530746459961, NaN, 0.5679012537002563, 0.5308641791343689, 0.5802469253540039, 0.5061728358268738, 0.7777777910232544, 0.6913580298423767, 0.5679012537002563, 0.654321014881134, 0.654321014881134, 0.8641975522041321, 0.6296296119689941, 0.6296296119689941, 0.6296296119689941, 0.6172839403152466, 0.7037037014961243, 0.6913580298423767, 0.6666666865348816, 0.6790123581886292, 0.6790123581886292, 0.8888888955116272, 0.9135802388191223, 0.7530864477157593, 0.8024691343307495, 0.6296296119689941, 0.9135802388191223, 0.8395061492919922, 0.8395061492919922, 0.8641975522041321, 0.9506173133850098, 0.8518518805503845, 0.8148148059844971, 0.9753086566925049, 0.8641975522041321, 0.9506173133850098, 0.9012345671653748, 0.9876543283462524, 0.9753086566925049, 0.8765432238578796, 0.9876543283462524, 0.8765432238578796, 0.8148148059844971, 0.8271604776382446, 0.9012345671653748, 0.8765432238578796, 0.8641975522041321, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 0.8641975522041321, 0.8765432238578796, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9135802388191223, 1.0], [0.20987653732299805, 0.012345679104328156, 0.2839506268501282, 0.23456789553165436, 0.4197530746459961, 0.43209877610206604, 0.4444444477558136, 0.4691357910633087, 0.4444444477558136, 0.5308641791343689, 0.43209877610206604, 0.43209877610206604, NaN, 0.4197530746459961, 0.5185185074806213, 0.5802469253540039, 0.5061728358268738, 0.6172839403152466, 0.6296296119689941, 0.5555555820465088, 0.7283950448036194, 0.5802469253540039, 0.654321014881134, 0.6296296119689941, 0.6790123581886292, 0.6419752836227417, 0.7037037014961243, 0.8148148059844971, 0.7283950448036194, 0.6666666865348816, 0.790123462677002, 0.6172839403152466, 0.7407407164573669, 0.7407407164573669, 0.9629629850387573, 0.8395061492919922, 0.6172839403152466, 0.9753086566925049, 0.8024691343307495, 0.9012345671653748, 0.9382715821266174, 0.8641975522041321, 0.8888888955116272, 0.9259259104728699, 1.0, 0.9259259104728699, 0.9629629850387573, 0.9259259104728699, 0.9012345671653748, 0.9259259104728699, 0.9259259104728699, 0.9012345671653748, 0.8765432238578796, 0.8518518805503845, 0.9753086566925049, 0.9506173133850098, 0.8888888955116272, 0.9259259104728699, 0.9382715821266174, 1.0, 1.0, 0.9382715821266174, 0.9753086566925049, 1.0, 1.0, 0.9753086566925049, 1.0, 1.0, 0.9012345671653748, 1.0, 1.0, 1.0, 1.0, 1.0], [0.04938271641731262, 0.1358024626970291, 0.14814814925193787, 0.2839506268501282, 0.4691357910633087, 0.4197530746459961, 0.40740740299224854, 0.40740740299224854, 0.37037035822868347, 0.4444444477558136, 0.37037035822868347, 0.4691357910633087, 0.5802469253540039, NaN, 0.5061728358268738, 0.4691357910633087, 0.5308641791343689, 0.5432098507881165, 0.5555555820465088, 0.8641975522041321, 0.6419752836227417, 0.6296296119689941, 0.5925925970077515, 0.6419752836227417, 0.6666666865348816, 0.5555555820465088, 0.5925925970077515, 0.604938268661499, 0.654321014881134, 0.7777777910232544, 0.654321014881134, 0.8024691343307495, 0.9012345671653748, 0.9382715821266174, 0.790123462677002, 0.654321014881134, 0.9012345671653748, 0.8271604776382446, 0.7777777910232544, 0.8641975522041321, 0.9382715821266174, 0.8271604776382446, 0.8271604776382446, 0.9753086566925049, 0.9629629850387573, 0.9629629850387573, 0.9876543283462524, 1.0, 0.9506173133850098, 0.9876543283462524, 0.9876543283462524, 0.8641975522041321, 0.8765432238578796, 0.7530864477157593, 1.0, 0.9876543283462524, 0.8888888955116272, 1.0, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9135802388191223, 0.9629629850387573, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07407407462596893, 0.16049382090568542, 0.17283950746059418, 0.29629629850387573, 0.34567901492118835, 0.34567901492118835, 0.395061731338501, 0.37037035822868347, 0.43209877610206604, 0.4197530746459961, 0.48148149251937866, 0.4197530746459961, 0.48148149251937866, 0.4938271641731262, NaN, 0.45679011940956116, 0.5679012537002563, 0.5185185074806213, 0.5185185074806213, 0.654321014881134, 0.5925925970077515, 0.5802469253540039, 0.5555555820465088, 0.5802469253540039, 0.8641975522041321, 0.6296296119689941, 0.6666666865348816, 0.6172839403152466, 0.5679012537002563, 0.7037037014961243, 0.6913580298423767, 0.7407407164573669, 0.8148148059844971, 0.8024691343307495, 0.7777777910232544, 0.7160493731498718, 0.8148148059844971, 0.8024691343307495, 0.7407407164573669, 0.8518518805503845, 0.9012345671653748, 0.9012345671653748, 0.7777777910232544, 0.9629629850387573, 0.9753086566925049, 0.9753086566925049, 0.9753086566925049, 0.9629629850387573, 0.9506173133850098, 0.9753086566925049, 0.9506173133850098, 0.9012345671653748, 0.9506173133850098, 0.7654321193695068, 0.9753086566925049, 0.9753086566925049, 0.8518518805503845, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 0.9876543283462524, 0.9506173133850098, 1.0, 1.0, 1.0, 1.0, 1.0], [0.03703703731298447, 0.12345679104328156, 0.23456789553165436, 0.12345679104328156, 0.3333333432674408, 0.32098764181137085, 0.40740740299224854, 0.43209877610206604, 0.45679011940956116, 0.40740740299224854, 0.4444444477558136, 0.4938271641731262, 0.4197530746459961, 0.5308641791343689, 0.5432098507881165, NaN, 0.5925925970077515, 0.5432098507881165, 0.5925925970077515, 0.5555555820465088, 0.5185185074806213, 0.5925925970077515, 0.604938268661499, 0.6172839403152466, 0.5925925970077515, 0.604938268661499, 0.604938268661499, 0.7160493731498718, 0.654321014881134, 0.7037037014961243, 0.6790123581886292, 0.6790123581886292, 0.8148148059844971, 0.7283950448036194, 0.8024691343307495, 0.6666666865348816, 0.7530864477157593, 0.8765432238578796, 0.8518518805503845, 0.8024691343307495, 0.8395061492919922, 0.8271604776382446, 0.8641975522041321, 0.8518518805503845, 0.9012345671653748, 0.8271604776382446, 0.9506173133850098, 0.8765432238578796, 0.8395061492919922, 0.8888888955116272, 0.8641975522041321, 0.9259259104728699, 0.9135802388191223, 0.9506173133850098, 0.8765432238578796, 0.8765432238578796, 0.9629629850387573, 0.8888888955116272, 0.9876543283462524, 0.9135802388191223, 0.9135802388191223, 0.9012345671653748, 0.9876543283462524, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 1.0, 0.9506173133850098, 1.0, 1.0, 0.9259259104728699, 1.0, 1.0, 1.0], [0.07407407462596893, 0.1358024626970291, 0.14814814925193787, 0.25925925374031067, 0.32098764181137085, 0.4197530746459961, 0.2839506268501282, 0.4197530746459961, 0.29629629850387573, 0.45679011940956116, 0.29629629850387573, 0.2222222238779068, 0.4938271641731262, 0.4691357910633087, 0.43209877610206604, 0.40740740299224854, NaN, 0.6419752836227417, 0.5308641791343689, 0.5802469253540039, 0.5555555820465088, 0.6790123581886292, 0.5432098507881165, 0.5679012537002563, 0.5679012537002563, 0.5432098507881165, 0.604938268661499, 0.6419752836227417, 0.6172839403152466, 0.604938268661499, 0.5802469253540039, 0.8518518805503845, 0.8024691343307495, 0.7283950448036194, 0.7654321193695068, 0.5925925970077515, 0.8888888955116272, 0.8271604776382446, 0.790123462677002, 0.8395061492919922, 0.8641975522041321, 0.8024691343307495, 0.790123462677002, 0.9012345671653748, 0.790123462677002, 0.8888888955116272, 0.8395061492919922, 0.9629629850387573, 0.9259259104728699, 0.8518518805503845, 0.9629629850387573, 0.8395061492919922, 0.7530864477157593, 0.8148148059844971, 0.9012345671653748, 0.8518518805503845, 0.8518518805503845, 1.0, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 1.0, 0.8641975522041321, 0.8641975522041321, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 0.9259259104728699, 1.0], [0.03703703731298447, 0.18518517911434174, 0.08641975373029709, 0.2716049253940582, 0.32098764181137085, 0.40740740299224854, 0.3580246865749359, 0.3827160596847534, 0.32098764181137085, 0.2469135820865631, 0.3333333432674408, 0.3086419701576233, 0.3827160596847534, 0.45679011940956116, 0.48148149251937866, 0.45679011940956116, 0.3580246865749359, NaN, 0.4938271641731262, 0.5925925970077515, 0.4444444477558136, 0.48148149251937866, 0.5308641791343689, 0.5802469253540039, 0.6666666865348816, 0.5308641791343689, 0.6419752836227417, 0.6419752836227417, 0.6172839403152466, 0.7530864477157593, 0.5432098507881165, 0.7283950448036194, 0.7160493731498718, 0.7530864477157593, 0.7407407164573669, 0.604938268661499, 0.7777777910232544, 0.8888888955116272, 0.8641975522041321, 0.7530864477157593, 0.790123462677002, 0.9012345671653748, 0.8148148059844971, 0.8148148059844971, 0.8395061492919922, 0.8148148059844971, 0.8395061492919922, 0.9382715821266174, 0.8765432238578796, 0.8888888955116272, 0.9135802388191223, 0.9629629850387573, 0.8271604776382446, 0.8888888955116272, 0.9259259104728699, 0.9506173133850098, 0.9506173133850098, 0.9753086566925049, 0.9629629850387573, 0.9753086566925049, 0.9753086566925049, 0.9876543283462524, 0.9629629850387573, 0.9506173133850098, 0.9876543283462524, 1.0, 0.9876543283462524, 0.9753086566925049, 0.9753086566925049, 1.0, 0.9753086566925049, 1.0, 1.0, 1.0], [0.08641975373029709, 0.03703703731298447, 0.29629629850387573, 0.14814814925193787, 0.2222222238779068, 0.18518517911434174, 0.4197530746459961, 0.3580246865749359, 0.4197530746459961, 0.3827160596847534, 0.45679011940956116, 0.43209877610206604, 0.37037035822868347, 0.4444444477558136, 0.48148149251937866, 0.40740740299224854, 0.4691357910633087, 0.5061728358268738, NaN, 0.4938271641731262, 0.5061728358268738, 0.5679012537002563, 0.7777777910232544, 0.6790123581886292, 0.604938268661499, 0.6666666865348816, 0.5802469253540039, 0.654321014881134, 0.654321014881134, 0.6790123581886292, 0.604938268661499, 0.6172839403152466, 0.7037037014961243, 0.7407407164573669, 0.7160493731498718, 0.8395061492919922, 0.6790123581886292, 0.8024691343307495, 0.7037037014961243, 0.7777777910232544, 0.8024691343307495, 0.8024691343307495, 0.8765432238578796, 0.8518518805503845, 0.9135802388191223, 0.8395061492919922, 0.8024691343307495, 0.9012345671653748, 0.8395061492919922, 0.9629629850387573, 0.9506173133850098, 0.8641975522041321, 0.8888888955116272, 0.7407407164573669, 0.9506173133850098, 0.9629629850387573, 0.8888888955116272, 0.8271604776382446, 0.9012345671653748, 1.0, 1.0, 0.8395061492919922, 0.9629629850387573, 0.9753086566925049, 0.9876543283462524, 0.9506173133850098, 1.0, 0.9876543283462524, 0.9135802388191223, 1.0, 1.0, 1.0, 1.0, 1.0], [0.012345679104328156, 0.09876543283462524, 0.12345679104328156, 0.2469135820865631, 0.29629629850387573, 0.395061731338501, 0.2716049253940582, 0.2469135820865631, 0.29629629850387573, 0.3333333432674408, 0.2716049253940582, 0.34567901492118835, 0.4444444477558136, 0.1358024626970291, 0.34567901492118835, 0.4444444477558136, 0.4197530746459961, 0.40740740299224854, 0.5061728358268738, NaN, 0.5679012537002563, 0.5555555820465088, 0.5185185074806213, 0.5555555820465088, 0.5679012537002563, 0.4938271641731262, 0.5185185074806213, 0.5925925970077515, 0.5679012537002563, 0.6419752836227417, 0.604938268661499, 0.7160493731498718, 0.8024691343307495, 0.8148148059844971, 0.7530864477157593, 0.6296296119689941, 0.790123462677002, 0.8024691343307495, 0.7283950448036194, 0.8395061492919922, 0.9012345671653748, 0.8024691343307495, 0.7777777910232544, 0.9135802388191223, 0.9259259104728699, 0.9259259104728699, 0.9629629850387573, 0.9629629850387573, 0.9382715821266174, 0.9506173133850098, 0.9382715821266174, 0.8518518805503845, 0.8395061492919922, 0.7283950448036194, 1.0, 0.9753086566925049, 0.8395061492919922, 1.0, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 1.0, 0.9012345671653748, 0.9629629850387573, 0.9876543283462524, 1.0, 1.0, 1.0, 0.9506173133850098, 1.0, 1.0, 1.0, 1.0, 1.0], [0.14814814925193787, 0.0, 0.2222222238779068, 0.16049382090568542, 0.2839506268501282, 0.3086419701576233, 0.3827160596847534, 0.3827160596847534, 0.395061731338501, 0.4691357910633087, 0.4197530746459961, 0.34567901492118835, 0.2716049253940582, 0.3580246865749359, 0.40740740299224854, 0.48148149251937866, 0.4444444477558136, 0.5555555820465088, 0.4938271641731262, 0.43209877610206604, NaN, 0.45679011940956116, 0.5185185074806213, 0.5432098507881165, 0.4938271641731262, 0.5432098507881165, 0.654321014881134, 0.7037037014961243, 0.604938268661499, 0.6172839403152466, 0.654321014881134, 0.5679012537002563, 0.6172839403152466, 0.7283950448036194, 0.8271604776382446, 0.7283950448036194, 0.6172839403152466, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8518518805503845, 0.8271604776382446, 0.8024691343307495, 0.8765432238578796, 0.8641975522041321, 0.8395061492919922, 0.9012345671653748, 0.8641975522041321, 0.8518518805503845, 0.8518518805503845, 0.8765432238578796, 0.8641975522041321, 0.8271604776382446, 0.8518518805503845, 0.9753086566925049, 0.9135802388191223, 0.9012345671653748, 0.8765432238578796, 0.8765432238578796, 1.0, 1.0, 0.8888888955116272, 0.9753086566925049, 0.9876543283462524, 0.9753086566925049, 0.9259259104728699, 0.9876543283462524, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.03703703731298447, 0.1111111119389534, 0.12345679104328156, 0.25925925374031067, 0.3580246865749359, 0.395061731338501, 0.20987653732299805, 0.395061731338501, 0.23456789553165436, 0.45679011940956116, 0.2716049253940582, 0.1358024626970291, 0.4197530746459961, 0.37037035822868347, 0.4197530746459961, 0.40740740299224854, 0.32098764181137085, 0.5185185074806213, 0.43209877610206604, 0.4444444477558136, 0.5432098507881165, NaN, 0.48148149251937866, 0.48148149251937866, 0.5061728358268738, 0.5061728358268738, 0.6172839403152466, 0.604938268661499, 0.5555555820465088, 0.604938268661499, 0.5925925970077515, 0.7777777910232544, 0.7283950448036194, 0.6790123581886292, 0.6419752836227417, 0.604938268661499, 0.8641975522041321, 0.7283950448036194, 0.790123462677002, 0.7283950448036194, 0.8024691343307495, 0.7530864477157593, 0.8148148059844971, 0.8518518805503845, 0.7283950448036194, 0.9135802388191223, 0.790123462677002, 0.9382715821266174, 0.9506173133850098, 0.8518518805503845, 0.9012345671653748, 0.8518518805503845, 0.8024691343307495, 0.8148148059844971, 0.8518518805503845, 0.8395061492919922, 0.8518518805503845, 1.0, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 0.8641975522041321, 0.8518518805503845, 0.9876543283462524, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0], [0.08641975373029709, 0.03703703731298447, 0.2469135820865631, 0.1358024626970291, 0.2222222238779068, 0.16049382090568542, 0.395061731338501, 0.3086419701576233, 0.40740740299224854, 0.34567901492118835, 0.40740740299224854, 0.37037035822868347, 0.34567901492118835, 0.40740740299224854, 0.4444444477558136, 0.395061731338501, 0.45679011940956116, 0.4691357910633087, 0.2222222238779068, 0.48148149251937866, 0.48148149251937866, 0.5185185074806213, NaN, 0.6296296119689941, 0.604938268661499, 0.6296296119689941, 0.5308641791343689, 0.5679012537002563, 0.654321014881134, 0.6172839403152466, 0.5432098507881165, 0.6172839403152466, 0.6296296119689941, 0.7283950448036194, 0.6419752836227417, 0.8148148059844971, 0.654321014881134, 0.7407407164573669, 0.6666666865348816, 0.7283950448036194, 0.7037037014961243, 0.790123462677002, 0.8024691343307495, 0.8148148059844971, 0.8765432238578796, 0.8024691343307495, 0.7654321193695068, 0.8395061492919922, 0.8148148059844971, 0.9012345671653748, 0.9259259104728699, 0.8641975522041321, 0.8271604776382446, 0.7283950448036194, 0.9012345671653748, 0.9259259104728699, 0.8765432238578796, 0.8024691343307495, 0.8765432238578796, 0.9876543283462524, 1.0, 0.8518518805503845, 0.9629629850387573, 0.9382715821266174, 0.9753086566925049, 0.9259259104728699, 0.9876543283462524, 0.9876543283462524, 0.8888888955116272, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.08641975373029709, 0.02469135820865631, 0.25925925374031067, 0.1358024626970291, 0.16049382090568542, 0.17283950746059418, 0.2839506268501282, 0.25925925374031067, 0.2716049253940582, 0.32098764181137085, 0.32098764181137085, 0.37037035822868347, 0.37037035822868347, 0.3580246865749359, 0.4197530746459961, 0.3827160596847534, 0.43209877610206604, 0.4197530746459961, 0.32098764181137085, 0.4444444477558136, 0.45679011940956116, 0.5185185074806213, 0.37037035822868347, NaN, 0.5555555820465088, 0.5802469253540039, 0.5802469253540039, 0.5061728358268738, 0.5802469253540039, 0.5802469253540039, 0.4938271641731262, 0.6419752836227417, 0.5925925970077515, 0.7283950448036194, 0.5432098507881165, 0.8518518805503845, 0.654321014881134, 0.6790123581886292, 0.6296296119689941, 0.7654321193695068, 0.7037037014961243, 0.7777777910232544, 0.7530864477157593, 0.7530864477157593, 0.8641975522041321, 0.8148148059844971, 0.7777777910232544, 0.8271604776382446, 0.8888888955116272, 0.9135802388191223, 0.8888888955116272, 0.7777777910232544, 0.8395061492919922, 0.7407407164573669, 0.9506173133850098, 0.9382715821266174, 0.8271604776382446, 0.8271604776382446, 0.8888888955116272, 1.0, 1.0, 0.8271604776382446, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 0.9135802388191223, 0.9753086566925049, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.02469135820865631, 0.03703703731298447, 0.1111111119389534, 0.1111111119389534, 0.1975308656692505, 0.2469135820865631, 0.23456789553165436, 0.2469135820865631, 0.29629629850387573, 0.2469135820865631, 0.2839506268501282, 0.37037035822868347, 0.32098764181137085, 0.3333333432674408, 0.1358024626970291, 0.40740740299224854, 0.43209877610206604, 0.3333333432674408, 0.395061731338501, 0.43209877610206604, 0.5061728358268738, 0.4938271641731262, 0.395061731338501, 0.4444444477558136, NaN, 0.4691357910633087, 0.4938271641731262, 0.5061728358268738, 0.5308641791343689, 0.5308641791343689, 0.6172839403152466, 0.6913580298423767, 0.7283950448036194, 0.7160493731498718, 0.7283950448036194, 0.5925925970077515, 0.6913580298423767, 0.7407407164573669, 0.7283950448036194, 0.7654321193695068, 0.7654321193695068, 0.6666666865348816, 0.7160493731498718, 0.8395061492919922, 0.9629629850387573, 0.9259259104728699, 0.9259259104728699, 0.8518518805503845, 0.8024691343307495, 0.8148148059844971, 0.8765432238578796, 0.7654321193695068, 0.8148148059844971, 0.7160493731498718, 0.9259259104728699, 0.9135802388191223, 0.790123462677002, 0.8518518805503845, 0.9012345671653748, 1.0, 1.0, 0.9012345671653748, 0.9382715821266174, 0.9629629850387573, 0.9876543283462524, 0.9629629850387573, 0.9876543283462524, 0.9876543283462524, 0.9012345671653748, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.1111111119389534, 0.04938271641731262, 0.25925925374031067, 0.14814814925193787, 0.2469135820865631, 0.18518517911434174, 0.29629629850387573, 0.2839506268501282, 0.34567901492118835, 0.3580246865749359, 0.3333333432674408, 0.3827160596847534, 0.3580246865749359, 0.4444444477558136, 0.37037035822868347, 0.395061731338501, 0.45679011940956116, 0.4691357910633087, 0.3333333432674408, 0.5061728358268738, 0.45679011940956116, 0.4938271641731262, 0.37037035822868347, 0.4197530746459961, 0.5308641791343689, NaN, 0.48148149251937866, 0.45679011940956116, 0.604938268661499, 0.5925925970077515, 0.5925925970077515, 0.6172839403152466, 0.6913580298423767, 0.6790123581886292, 0.5679012537002563, 0.6172839403152466, 0.6666666865348816, 0.6419752836227417, 0.654321014881134, 0.654321014881134, 0.7037037014961243, 0.6419752836227417, 0.8395061492919922, 0.7283950448036194, 0.8518518805503845, 0.7407407164573669, 0.8518518805503845, 0.790123462677002, 0.7530864477157593, 0.7777777910232544, 0.8024691343307495, 0.6790123581886292, 0.9135802388191223, 0.7654321193695068, 0.7777777910232544, 0.8148148059844971, 0.7654321193695068, 0.7283950448036194, 0.8395061492919922, 0.9259259104728699, 0.9382715821266174, 0.7283950448036194, 0.9753086566925049, 1.0, 1.0, 0.9259259104728699, 1.0, 1.0, 0.9259259104728699, 1.0, 0.9753086566925049, 1.0, 1.0, 1.0], [0.12345679104328156, 0.1111111119389534, 0.16049382090568542, 0.14814814925193787, 0.2716049253940582, 0.2716049253940582, 0.2716049253940582, 0.32098764181137085, 0.2839506268501282, 0.3086419701576233, 0.32098764181137085, 0.29629629850387573, 0.29629629850387573, 0.40740740299224854, 0.3333333432674408, 0.395061731338501, 0.395061731338501, 0.3580246865749359, 0.4197530746459961, 0.48148149251937866, 0.34567901492118835, 0.3827160596847534, 0.4691357910633087, 0.4197530746459961, 0.5061728358268738, 0.5185185074806213, NaN, 0.48148149251937866, 0.5061728358268738, 0.4444444477558136, 0.5555555820465088, 0.6790123581886292, 0.5679012537002563, 0.5555555820465088, 0.5802469253540039, 0.5679012537002563, 0.6790123581886292, 0.6172839403152466, 0.7037037014961243, 0.6172839403152466, 0.6419752836227417, 0.5802469253540039, 0.7654321193695068, 0.654321014881134, 0.7777777910232544, 0.7407407164573669, 0.8271604776382446, 0.8148148059844971, 0.8271604776382446, 0.8148148059844971, 0.8395061492919922, 0.8395061492919922, 0.8395061492919922, 0.8518518805503845, 0.8765432238578796, 0.8518518805503845, 0.8888888955116272, 0.8888888955116272, 0.9876543283462524, 0.9382715821266174, 0.9382715821266174, 0.9259259104728699, 0.9259259104728699, 0.9135802388191223, 0.9629629850387573, 1.0, 0.9629629850387573, 0.9753086566925049, 0.9876543283462524, 0.9876543283462524, 0.9876543283462524, 1.0, 1.0, 1.0], [0.012345679104328156, 0.06172839552164078, 0.1111111119389534, 0.1358024626970291, 0.1975308656692505, 0.2222222238779068, 0.3086419701576233, 0.23456789553165436, 0.29012346267700195, 0.23456789553165436, 0.34567901492118835, 0.3086419701576233, 0.18518517911434174, 0.395061731338501, 0.3827160596847534, 0.2839506268501282, 0.3580246865749359, 0.3580246865749359, 0.34567901492118835, 0.40740740299224854, 0.29629629850387573, 0.395061731338501, 0.43209877610206604, 0.4938271641731262, 0.4938271641731262, 0.5432098507881165, 0.5185185074806213, NaN, 0.5308641791343689, 0.5679012537002563, 0.5925925970077515, 0.5061728358268738, 0.5802469253540039, 0.5925925970077515, 0.5802469253540039, 0.604938268661499, 0.5802469253540039, 0.6913580298423767, 0.6296296119689941, 0.654321014881134, 0.654321014881134, 0.6913580298423767, 0.7530864477157593, 0.6913580298423767, 0.7407407164573669, 0.7407407164573669, 0.8888888955116272, 0.8024691343307495, 0.7654321193695068, 0.8271604776382446, 0.790123462677002, 0.9135802388191223, 0.8518518805503845, 0.8024691343307495, 0.8395061492919922, 0.8395061492919922, 0.9012345671653748, 0.8518518805503845, 0.9506173133850098, 0.9506173133850098, 0.9506173133850098, 0.8888888955116272, 0.9382715821266174, 0.9506173133850098, 0.9629629850387573, 0.9876543283462524, 0.9753086566925049, 0.9876543283462524, 0.9506173133850098, 1.0, 0.9753086566925049, 1.0, 1.0, 1.0], [0.08641975373029709, 0.03703703731298447, 0.2222222238779068, 0.12345679104328156, 0.16049382090568542, 0.1111111119389534, 0.34567901492118835, 0.2839506268501282, 0.37037035822868347, 0.32098764181137085, 0.37037035822868347, 0.3333333432674408, 0.2716049253940582, 0.34567901492118835, 0.43209877610206604, 0.34567901492118835, 0.3827160596847534, 0.3827160596847534, 0.34567901492118835, 0.43209877610206604, 0.395061731338501, 0.4444444477558136, 0.34567901492118835, 0.4197530746459961, 0.4691357910633087, 0.395061731338501, 0.4938271641731262, 0.4691357910633087, NaN, 0.5802469253540039, 0.45679011940956116, 0.5679012537002563, 0.5555555820465088, 0.604938268661499, 0.5555555820465088, 0.6419752836227417, 0.6172839403152466, 0.604938268661499, 0.654321014881134, 0.5802469253540039, 0.5925925970077515, 0.7283950448036194, 0.8765432238578796, 0.6296296119689941, 0.7530864477157593, 0.7037037014961243, 0.7654321193695068, 0.7407407164573669, 0.7160493731498718, 0.7530864477157593, 0.7283950448036194, 0.7777777910232544, 0.9382715821266174, 0.7407407164573669, 0.7530864477157593, 0.790123462677002, 0.8148148059844971, 0.7283950448036194, 0.8271604776382446, 0.8395061492919922, 0.8395061492919922, 0.7283950448036194, 0.9876543283462524, 0.9753086566925049, 0.9876543283462524, 0.8765432238578796, 0.9876543283462524, 0.9876543283462524, 0.9135802388191223, 1.0, 0.9629629850387573, 1.0, 1.0, 1.0], [0.012345679104328156, 0.03703703731298447, 0.09876543283462524, 0.18518517911434174, 0.17283950746059418, 0.20987653732299805, 0.23456789553165436, 0.14814814925193787, 0.23456789553165436, 0.16049382090568542, 0.2469135820865631, 0.32098764181137085, 0.3333333432674408, 0.2222222238779068, 0.29629629850387573, 0.29629629850387573, 0.395061731338501, 0.2469135820865631, 0.32098764181137085, 0.3580246865749359, 0.3827160596847534, 0.395061731338501, 0.3827160596847534, 0.4197530746459961, 0.4691357910633087, 0.40740740299224854, 0.5555555820465088, 0.43209877610206604, 0.4197530746459961, NaN, 0.395061731338501, 0.5555555820465088, 0.5185185074806213, 0.6296296119689941, 0.5802469253540039, 0.6419752836227417, 0.6172839403152466, 0.6172839403152466, 0.6172839403152466, 0.5555555820465088, 0.6666666865348816, 0.9259259104728699, 0.6790123581886292, 0.7037037014961243, 0.7283950448036194, 0.8271604776382446, 0.7160493731498718, 0.7777777910232544, 0.8024691343307495, 0.8765432238578796, 0.8395061492919922, 0.8148148059844971, 0.7407407164573669, 0.7407407164573669, 0.9876543283462524, 0.9012345671653748, 0.8148148059844971, 0.9012345671653748, 0.9012345671653748, 0.9135802388191223, 0.9135802388191223, 0.9012345671653748, 0.9753086566925049, 0.9629629850387573, 0.9876543283462524, 0.9506173133850098, 0.9876543283462524, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.09876543283462524, 0.03703703731298447, 0.20987653732299805, 0.04938271641731262, 0.16049382090568542, 0.1358024626970291, 0.34567901492118835, 0.32098764181137085, 0.37037035822868347, 0.25925925374031067, 0.3827160596847534, 0.32098764181137085, 0.20987653732299805, 0.34567901492118835, 0.3086419701576233, 0.32098764181137085, 0.4197530746459961, 0.45679011940956116, 0.395061731338501, 0.395061731338501, 0.34567901492118835, 0.40740740299224854, 0.45679011940956116, 0.5061728358268738, 0.3827160596847534, 0.40740740299224854, 0.4444444477558136, 0.40740740299224854, 0.5432098507881165, 0.604938268661499, NaN, 0.5802469253540039, 0.5308641791343689, 0.5925925970077515, 0.5185185074806213, 0.6790123581886292, 0.6172839403152466, 0.6666666865348816, 0.604938268661499, 0.5679012537002563, 0.5802469253540039, 0.7037037014961243, 0.8148148059844971, 0.6790123581886292, 0.7654321193695068, 0.654321014881134, 0.6666666865348816, 0.7160493731498718, 0.6666666865348816, 0.7407407164573669, 0.7777777910232544, 0.7160493731498718, 0.9012345671653748, 0.8518518805503845, 0.7777777910232544, 0.8024691343307495, 0.8641975522041321, 0.7037037014961243, 0.8395061492919922, 0.8765432238578796, 0.8765432238578796, 0.7283950448036194, 0.9629629850387573, 0.9629629850387573, 0.9135802388191223, 0.8641975522041321, 0.9506173133850098, 0.8888888955116272, 0.9012345671653748, 1.0, 0.8765432238578796, 1.0, 1.0, 1.0], [0.02469135820865631, 0.12345679104328156, 0.04938271641731262, 0.23456789553165436, 0.25925925374031067, 0.3333333432674408, 0.06172839552164078, 0.12345679104328156, 0.08641975373029709, 0.17283950746059418, 0.07407407462596893, 0.1111111119389534, 0.3827160596847534, 0.1975308656692505, 0.25925925374031067, 0.32098764181137085, 0.14814814925193787, 0.2716049253940582, 0.3827160596847534, 0.2839506268501282, 0.43209877610206604, 0.2222222238779068, 0.3827160596847534, 0.3580246865749359, 0.3086419701576233, 0.3827160596847534, 0.32098764181137085, 0.4938271641731262, 0.43209877610206604, 0.4444444477558136, 0.4197530746459961, NaN, 0.604938268661499, 0.40740740299224854, 0.5061728358268738, 0.4197530746459961, 0.6419752836227417, 0.5802469253540039, 0.7037037014961243, 0.5925925970077515, 0.6790123581886292, 0.6419752836227417, 0.5185185074806213, 0.6913580298423767, 0.5185185074806213, 0.604938268661499, 0.6296296119689941, 0.7160493731498718, 0.7654321193695068, 0.8148148059844971, 0.7283950448036194, 0.7777777910232544, 0.654321014881134, 0.7654321193695068, 0.7407407164573669, 0.8024691343307495, 0.8518518805503845, 0.9753086566925049, 0.9629629850387573, 0.8395061492919922, 0.8395061492919922, 1.0, 0.790123462677002, 0.8395061492919922, 0.9753086566925049, 1.0, 0.9753086566925049, 0.9629629850387573, 1.0, 0.8888888955116272, 0.9753086566925049, 0.9629629850387573, 0.8888888955116272, 1.0], [0.012345679104328156, 0.012345679104328156, 0.04938271641731262, 0.14814814925193787, 0.1358024626970291, 0.2469135820865631, 0.09876543283462524, 0.09876543283462524, 0.07407407462596893, 0.16049382090568542, 0.08641975373029709, 0.08641975373029709, 0.25925925374031067, 0.09876543283462524, 0.18518517911434174, 0.18518517911434174, 0.1975308656692505, 0.2839506268501282, 0.29629629850387573, 0.1975308656692505, 0.3827160596847534, 0.2716049253940582, 0.37037035822868347, 0.40740740299224854, 0.2716049253940582, 0.3086419701576233, 0.43209877610206604, 0.4197530746459961, 0.4444444477558136, 0.48148149251937866, 0.4691357910633087, 0.395061731338501, NaN, 0.5679012537002563, 0.5555555820465088, 0.5308641791343689, 0.5432098507881165, 0.6913580298423767, 0.6296296119689941, 0.654321014881134, 0.7530864477157593, 0.7654321193695068, 0.6666666865348816, 0.9135802388191223, 0.6419752836227417, 0.6913580298423767, 0.6913580298423767, 0.7407407164573669, 0.8641975522041321, 0.7777777910232544, 0.7777777910232544, 0.7283950448036194, 0.6419752836227417, 0.6666666865348816, 0.8518518805503845, 0.8271604776382446, 0.7777777910232544, 0.8888888955116272, 0.9012345671653748, 0.9753086566925049, 0.9629629850387573, 0.9135802388191223, 0.8641975522041321, 0.8518518805503845, 0.8765432238578796, 0.9382715821266174, 0.8641975522041321, 0.8641975522041321, 0.8024691343307495, 1.0, 0.9753086566925049, 1.0, 0.8888888955116272, 1.0], [0.03703703731298447, 0.02469135820865631, 0.08641975373029709, 0.09876543283462524, 0.09876543283462524, 0.14814814925193787, 0.12345679104328156, 0.12345679104328156, 0.14814814925193787, 0.17283950746059418, 0.16049382090568542, 0.2469135820865631, 0.25925925374031067, 0.06172839552164078, 0.1975308656692505, 0.2716049253940582, 0.2716049253940582, 0.2469135820865631, 0.25925925374031067, 0.18518517911434174, 0.2716049253940582, 0.32098764181137085, 0.2716049253940582, 0.2716049253940582, 0.2839506268501282, 0.32098764181137085, 0.4444444477558136, 0.40740740299224854, 0.395061731338501, 0.37037035822868347, 0.40740740299224854, 0.5925925970077515, 0.43209877610206604, NaN, 0.48148149251937866, 0.4938271641731262, 0.6419752836227417, 0.4938271641731262, 0.654321014881134, 0.5432098507881165, 0.5432098507881165, 0.6296296119689941, 0.6172839403152466, 0.604938268661499, 0.48148149251937866, 0.790123462677002, 0.6913580298423767, 0.8395061492919922, 0.8148148059844971, 0.8518518805503845, 0.8518518805503845, 0.6790123581886292, 0.6172839403152466, 0.6790123581886292, 0.9135802388191223, 0.8888888955116272, 0.7407407164573669, 0.8641975522041321, 0.8765432238578796, 0.9382715821266174, 0.9506173133850098, 0.8765432238578796, 0.8641975522041321, 0.9259259104728699, 0.9135802388191223, 0.8888888955116272, 0.9135802388191223, 0.9876543283462524, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.08641975373029709, 0.012345679104328156, 0.08641975373029709, 0.06172839552164078, 0.23456789553165436, 0.18518517911434174, 0.25925925374031067, 0.1111111119389534, 0.2222222238779068, 0.1975308656692505, 0.03703703731298447, 0.20987653732299805, 0.2222222238779068, 0.1975308656692505, 0.23456789553165436, 0.25925925374031067, 0.2839506268501282, 0.2469135820865631, 0.17283950746059418, 0.3580246865749359, 0.3580246865749359, 0.45679011940956116, 0.2716049253940582, 0.43209877610206604, 0.4197530746459961, 0.4197530746459961, 0.4444444477558136, 0.4197530746459961, 0.48148149251937866, 0.4938271641731262, 0.4444444477558136, 0.5185185074806213, NaN, 0.5061728358268738, 0.5308641791343689, 0.5679012537002563, 0.6172839403152466, 0.5555555820465088, 0.5925925970077515, 0.5679012537002563, 0.6790123581886292, 0.6419752836227417, 0.6913580298423767, 0.6790123581886292, 0.8024691343307495, 0.7407407164573669, 0.6913580298423767, 0.7160493731498718, 0.7407407164573669, 0.7530864477157593, 0.6913580298423767, 0.6419752836227417, 0.8024691343307495, 0.7530864477157593, 0.790123462677002, 0.7777777910232544, 0.8271604776382446, 0.8888888955116272, 0.8888888955116272, 0.7654321193695068, 0.9753086566925049, 0.9753086566925049, 0.9382715821266174, 0.8888888955116272, 0.9135802388191223, 0.9876543283462524, 0.8888888955116272, 1.0, 0.9506173133850098, 1.0, 1.0, 1.0], [0.06172839552164078, 0.04938271641731262, 0.17283950746059418, 0.03703703731298447, 0.14814814925193787, 0.09876543283462524, 0.20987653732299805, 0.25925925374031067, 0.2469135820865631, 0.2716049253940582, 0.25925925374031067, 0.37037035822868347, 0.16049382090568542, 0.34567901492118835, 0.2839506268501282, 0.3333333432674408, 0.40740740299224854, 0.395061731338501, 0.16049382090568542, 0.37037035822868347, 0.2716049253940582, 0.395061731338501, 0.18518517911434174, 0.14814814925193787, 0.40740740299224854, 0.3827160596847534, 0.43209877610206604, 0.395061731338501, 0.3580246865749359, 0.3580246865749359, 0.32098764181137085, 0.5802469253540039, 0.4691357910633087, 0.5061728358268738, 0.4938271641731262, NaN, 0.5308641791343689, 0.4197530746459961, 0.5802469253540039, 0.5185185074806213, 0.5308641791343689, 0.5185185074806213, 0.5802469253540039, 0.5555555820465088, 0.5802469253540039, 0.6790123581886292, 0.6172839403152466, 0.5925925970077515, 0.5925925970077515, 0.6913580298423767, 0.6172839403152466, 0.6419752836227417, 0.8395061492919922, 0.7530864477157593, 0.654321014881134, 0.6913580298423767, 0.6419752836227417, 0.6913580298423767, 0.7654321193695068, 0.7777777910232544, 0.7777777910232544, 0.7283950448036194, 0.9506173133850098, 0.9753086566925049, 0.9135802388191223, 0.790123462677002, 0.9012345671653748, 1.0, 0.8888888955116272, 0.9135802388191223, 1.0, 0.9629629850387573, 1.0, 1.0], [0.012345679104328156, 0.09876543283462524, 0.03703703731298447, 0.2222222238779068, 0.20987653732299805, 0.3086419701576233, 0.04938271641731262, 0.04938271641731262, 0.06172839552164078, 0.12345679104328156, 0.07407407462596893, 0.08641975373029709, 0.3827160596847534, 0.09876543283462524, 0.18518517911434174, 0.2469135820865631, 0.1111111119389534, 0.2222222238779068, 0.32098764181137085, 0.20987653732299805, 0.3827160596847534, 0.1358024626970291, 0.34567901492118835, 0.34567901492118835, 0.3086419701576233, 0.3333333432674408, 0.32098764181137085, 0.4197530746459961, 0.3827160596847534, 0.3827160596847534, 0.3827160596847534, 0.3580246865749359, 0.45679011940956116, 0.3580246865749359, 0.4691357910633087, 0.4691357910633087, NaN, 0.5555555820465088, 0.6296296119689941, 0.5925925970077515, 0.6172839403152466, 0.604938268661499, 0.5679012537002563, 0.5802469253540039, 0.5061728358268738, 0.5308641791343689, 0.6296296119689941, 0.6913580298423767, 0.8888888955116272, 0.7283950448036194, 0.7283950448036194, 0.8024691343307495, 0.6296296119689941, 0.6913580298423767, 0.7530864477157593, 0.8271604776382446, 0.8518518805503845, 0.9506173133850098, 0.9753086566925049, 0.8148148059844971, 0.8271604776382446, 0.9876543283462524, 0.8395061492919922, 0.8148148059844971, 0.9382715821266174, 1.0, 0.8888888955116272, 0.8641975522041321, 0.9876543283462524, 0.9876543283462524, 0.9506173133850098, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.03703703731298447, 0.06172839552164078, 0.0, 0.09876543283462524, 0.1975308656692505, 0.04938271641731262, 0.18518517911434174, 0.0, 0.17283950746059418, 0.16049382090568542, 0.02469135820865631, 0.17283950746059418, 0.1975308656692505, 0.12345679104328156, 0.17283950746059418, 0.1111111119389534, 0.1975308656692505, 0.1975308656692505, 0.1111111119389534, 0.2716049253940582, 0.25925925374031067, 0.32098764181137085, 0.25925925374031067, 0.3580246865749359, 0.3827160596847534, 0.3086419701576233, 0.395061731338501, 0.3827160596847534, 0.3333333432674408, 0.4197530746459961, 0.3086419701576233, 0.5061728358268738, 0.43209877610206604, 0.5802469253540039, 0.4444444477558136, NaN, 0.4444444477558136, 0.5061728358268738, 0.4444444477558136, 0.6419752836227417, 0.6296296119689941, 0.48148149251937866, 0.6419752836227417, 0.6790123581886292, 0.7037037014961243, 0.6666666865348816, 0.6913580298423767, 0.7654321193695068, 0.7407407164573669, 0.8024691343307495, 0.7037037014961243, 0.6790123581886292, 0.8765432238578796, 0.790123462677002, 0.8271604776382446, 0.8148148059844971, 0.8518518805503845, 0.9629629850387573, 0.9629629850387573, 0.8641975522041321, 0.9506173133850098, 0.9382715821266174, 0.9506173133850098, 0.9012345671653748, 0.9506173133850098, 0.9506173133850098, 0.8641975522041321, 1.0, 0.9876543283462524, 1.0, 1.0, 1.0], [0.0, 0.09876543283462524, 0.03703703731298447, 0.1111111119389534, 0.14814814925193787, 0.17283950746059418, 0.18518517911434174, 0.03703703731298447, 0.14814814925193787, 0.02469135820865631, 0.1975308656692505, 0.16049382090568542, 0.1975308656692505, 0.2222222238779068, 0.25925925374031067, 0.14814814925193787, 0.20987653732299805, 0.1358024626970291, 0.29629629850387573, 0.2716049253940582, 0.2222222238779068, 0.20987653732299805, 0.3333333432674408, 0.37037035822868347, 0.2716049253940582, 0.34567901492118835, 0.29629629850387573, 0.37037035822868347, 0.34567901492118835, 0.3827160596847534, 0.395061731338501, 0.29629629850387573, 0.37037035822868347, 0.34567901492118835, 0.3827160596847534, 0.4197530746459961, 0.37037035822868347, 0.5555555820465088, NaN, 0.4444444477558136, 0.4938271641731262, 0.5432098507881165, 0.5802469253540039, 0.5308641791343689, 0.5802469253540039, 0.4938271641731262, 0.6913580298423767, 0.6419752836227417, 0.6790123581886292, 0.654321014881134, 0.6666666865348816, 0.9259259104728699, 0.6913580298423767, 0.6172839403152466, 0.7530864477157593, 0.6913580298423767, 0.9629629850387573, 0.8765432238578796, 0.8641975522041321, 0.7530864477157593, 0.7530864477157593, 0.8888888955116272, 0.8641975522041321, 0.8518518805503845, 0.9259259104728699, 0.9753086566925049, 0.9259259104728699, 0.8148148059844971, 1.0, 0.8888888955116272, 0.9259259104728699, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.09876543283462524, 0.06172839552164078, 0.1111111119389534, 0.08641975373029709, 0.16049382090568542, 0.14814814925193787, 0.17283950746059418, 0.16049382090568542, 0.16049382090568542, 0.1358024626970291, 0.09876543283462524, 0.1358024626970291, 0.14814814925193787, 0.1975308656692505, 0.16049382090568542, 0.2469135820865631, 0.2222222238779068, 0.16049382090568542, 0.2222222238779068, 0.2716049253940582, 0.2716049253940582, 0.23456789553165436, 0.23456789553165436, 0.34567901492118835, 0.3827160596847534, 0.34567901492118835, 0.4197530746459961, 0.4444444477558136, 0.43209877610206604, 0.40740740299224854, 0.34567901492118835, 0.45679011940956116, 0.4444444477558136, 0.48148149251937866, 0.40740740299224854, 0.4938271641731262, 0.5555555820465088, NaN, 0.4197530746459961, 0.5555555820465088, 0.6790123581886292, 0.5802469253540039, 0.654321014881134, 0.604938268661499, 0.6666666865348816, 0.6172839403152466, 0.7160493731498718, 0.6790123581886292, 0.654321014881134, 0.7037037014961243, 0.6419752836227417, 0.7283950448036194, 0.7407407164573669, 0.7037037014961243, 0.7407407164573669, 0.7283950448036194, 0.9012345671653748, 0.9753086566925049, 0.9876543283462524, 0.7407407164573669, 0.8518518805503845, 0.8518518805503845, 0.8395061492919922, 0.9382715821266174, 0.8518518805503845, 0.8765432238578796, 0.790123462677002, 1.0, 0.9629629850387573, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.04938271641731262, 0.09876543283462524, 0.03703703731298447, 0.09876543283462524, 0.09876543283462524, 0.08641975373029709, 0.09876543283462524, 0.08641975373029709, 0.09876543283462524, 0.04938271641731262, 0.06172839552164078, 0.06172839552164078, 0.09876543283462524, 0.16049382090568542, 0.1358024626970291, 0.20987653732299805, 0.1975308656692505, 0.09876543283462524, 0.14814814925193787, 0.1975308656692505, 0.29629629850387573, 0.29629629850387573, 0.23456789553165436, 0.29629629850387573, 0.3580246865749359, 0.34567901492118835, 0.40740740299224854, 0.3333333432674408, 0.4197530746459961, 0.32098764181137085, 0.2469135820865631, 0.45679011940956116, 0.40740740299224854, 0.4691357910633087, 0.3827160596847534, 0.5555555820465088, 0.5061728358268738, 0.5802469253540039, NaN, 0.5802469253540039, 0.5432098507881165, 0.7160493731498718, 0.5432098507881165, 0.5925925970077515, 0.6419752836227417, 0.6666666865348816, 0.8024691343307495, 0.654321014881134, 0.6913580298423767, 0.7037037014961243, 0.5555555820465088, 0.654321014881134, 0.8271604776382446, 0.7283950448036194, 0.7160493731498718, 0.8395061492919922, 0.8765432238578796, 0.9382715821266174, 0.9382715821266174, 0.8765432238578796, 0.8518518805503845, 0.8395061492919922, 0.8148148059844971, 0.9135802388191223, 0.8271604776382446, 0.8641975522041321, 0.790123462677002, 1.0, 0.9876543283462524, 1.0, 0.8888888955116272, 1.0], [0.02469135820865631, 0.0, 0.02469135820865631, 0.03703703731298447, 0.04938271641731262, 0.09876543283462524, 0.04938271641731262, 0.04938271641731262, 0.04938271641731262, 0.07407407462596893, 0.06172839552164078, 0.14814814925193787, 0.1358024626970291, 0.17283950746059418, 0.09876543283462524, 0.17283950746059418, 0.1975308656692505, 0.09876543283462524, 0.1975308656692505, 0.1975308656692505, 0.17283950746059418, 0.2469135820865631, 0.20987653732299805, 0.2222222238779068, 0.3333333432674408, 0.3580246865749359, 0.4197530746459961, 0.3086419701576233, 0.2716049253940582, 0.07407407462596893, 0.29629629850387573, 0.3580246865749359, 0.23456789553165436, 0.37037035822868347, 0.43209877610206604, 0.48148149251937866, 0.395061731338501, 0.3580246865749359, 0.45679011940956116, 0.4444444477558136, 0.4197530746459961, NaN, 0.43209877610206604, 0.4444444477558136, 0.5308641791343689, 0.7037037014961243, 0.6172839403152466, 0.604938268661499, 0.6172839403152466, 0.6790123581886292, 0.6913580298423767, 0.604938268661499, 0.5802469253540039, 0.5802469253540039, 0.8765432238578796, 0.7530864477157593, 0.654321014881134, 0.8641975522041321, 0.8271604776382446, 0.8271604776382446, 0.8271604776382446, 0.8641975522041321, 0.9506173133850098, 0.9506173133850098, 0.8888888955116272, 0.9135802388191223, 0.9135802388191223, 0.9753086566925049, 0.8765432238578796, 0.9753086566925049, 0.9876543283462524, 1.0, 1.0, 1.0], [0.04938271641731262, 0.03703703731298447, 0.03703703731298447, 0.03703703731298447, 0.04938271641731262, 0.06172839552164078, 0.20987653732299805, 0.09876543283462524, 0.20987653732299805, 0.09876543283462524, 0.2222222238779068, 0.18518517911434174, 0.1111111119389534, 0.17283950746059418, 0.2222222238779068, 0.1358024626970291, 0.20987653732299805, 0.18518517911434174, 0.12345679104328156, 0.2222222238779068, 0.1975308656692505, 0.18518517911434174, 0.1975308656692505, 0.2469135820865631, 0.2839506268501282, 0.16049382090568542, 0.23456789553165436, 0.2469135820865631, 0.12345679104328156, 0.32098764181137085, 0.18518517911434174, 0.48148149251937866, 0.3333333432674408, 0.3827160596847534, 0.32098764181137085, 0.4197530746459961, 0.43209877610206604, 0.37037035822868347, 0.4197530746459961, 0.32098764181137085, 0.45679011940956116, 0.5679012537002563, NaN, 0.4691357910633087, 0.5185185074806213, 0.4938271641731262, 0.5185185074806213, 0.5432098507881165, 0.4938271641731262, 0.6419752836227417, 0.654321014881134, 0.6419752836227417, 0.7592592835426331, 0.6913580298423767, 0.6419752836227417, 0.7160493731498718, 0.7530864477157593, 0.654321014881134, 0.7777777910232544, 0.6296296119689941, 0.6419752836227417, 0.7160493731498718, 0.9629629850387573, 0.9629629850387573, 0.9259259104728699, 0.7777777910232544, 0.9382715821266174, 0.9012345671653748, 0.8888888955116272, 0.9012345671653748, 0.8888888955116272, 0.9753086566925049, 1.0, 1.0], [0.0, 0.0, 0.04938271641731262, 0.04938271641731262, 0.03703703731298447, 0.07407407462596893, 0.07407407462596893, 0.06172839552164078, 0.06172839552164078, 0.06172839552164078, 0.03703703731298447, 0.02469135820865631, 0.07407407462596893, 0.02469135820865631, 0.03703703731298447, 0.14814814925193787, 0.09876543283462524, 0.18518517911434174, 0.14814814925193787, 0.08641975373029709, 0.12345679104328156, 0.14814814925193787, 0.18518517911434174, 0.2469135820865631, 0.16049382090568542, 0.2716049253940582, 0.34567901492118835, 0.3086419701576233, 0.37037035822868347, 0.29629629850387573, 0.32098764181137085, 0.3086419701576233, 0.08641975373029709, 0.395061731338501, 0.3580246865749359, 0.4444444477558136, 0.4197530746459961, 0.5185185074806213, 0.4691357910633087, 0.4197530746459961, 0.2839506268501282, 0.5555555820465088, 0.5308641791343689, NaN, 0.5061728358268738, 0.6172839403152466, 0.604938268661499, 0.6172839403152466, 0.6419752836227417, 0.6419752836227417, 0.6790123581886292, 0.7037037014961243, 0.5555555820465088, 0.654321014881134, 0.8148148059844971, 0.7283950448036194, 0.7037037014961243, 0.790123462677002, 0.8641975522041321, 0.8271604776382446, 0.8271604776382446, 0.8271604776382446, 0.8518518805503845, 0.8395061492919922, 0.8024691343307495, 0.9012345671653748, 0.8024691343307495, 0.8518518805503845, 0.7777777910232544, 1.0, 0.9753086566925049, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.04938271641731262, 0.012345679104328156, 0.06172839552164078, 0.06172839552164078, 0.07407407462596893, 0.08641975373029709, 0.04938271641731262, 0.04938271641731262, 0.08641975373029709, 0.1358024626970291, 0.0, 0.03703703731298447, 0.02469135820865631, 0.09876543283462524, 0.20987653732299805, 0.16049382090568542, 0.08641975373029709, 0.07407407462596893, 0.1358024626970291, 0.2716049253940582, 0.12345679104328156, 0.1358024626970291, 0.03703703731298447, 0.14814814925193787, 0.2222222238779068, 0.25925925374031067, 0.2469135820865631, 0.2716049253940582, 0.23456789553165436, 0.48148149251937866, 0.3580246865749359, 0.5185185074806213, 0.3086419701576233, 0.4197530746459961, 0.4938271641731262, 0.3580246865749359, 0.4197530746459961, 0.34567901492118835, 0.45679011940956116, 0.4691357910633087, 0.48148149251937866, 0.4938271641731262, NaN, 0.5679012537002563, 0.5555555820465088, 0.6172839403152466, 0.5432098507881165, 0.6172839403152466, 0.6419752836227417, 0.5555555820465088, 0.6296296119689941, 0.6666666865348816, 0.7160493731498718, 0.7160493731498718, 0.6666666865348816, 0.654321014881134, 0.7407407164573669, 0.8271604776382446, 0.8395061492919922, 0.6666666865348816, 0.8641975522041321, 0.9506173133850098, 0.9382715821266174, 0.790123462677002, 0.9629629850387573, 0.9382715821266174, 0.8888888955116272, 1.0, 0.9135802388191223, 1.0, 1.0, 1.0], [0.0, 0.0, 0.03703703731298447, 0.012345679104328156, 0.06172839552164078, 0.08641975373029709, 0.06172839552164078, 0.0, 0.08641975373029709, 0.09876543283462524, 0.07407407462596893, 0.04938271641731262, 0.07407407462596893, 0.03703703731298447, 0.02469135820865631, 0.17283950746059418, 0.1111111119389534, 0.18518517911434174, 0.16049382090568542, 0.07407407462596893, 0.16049382090568542, 0.08641975373029709, 0.1975308656692505, 0.18518517911434174, 0.07407407462596893, 0.25925925374031067, 0.25925925374031067, 0.25925925374031067, 0.29629629850387573, 0.17283950746059418, 0.34567901492118835, 0.395061731338501, 0.3086419701576233, 0.20987653732299805, 0.32098764181137085, 0.32098764181137085, 0.4691357910633087, 0.32098764181137085, 0.5061728358268738, 0.395061731338501, 0.40740740299224854, 0.29629629850387573, 0.5061728358268738, 0.3827160596847534, 0.43209877610206604, NaN, 0.5185185074806213, 0.5432098507881165, 0.6419752836227417, 0.5308641791343689, 0.5555555820465088, 0.6913580298423767, 0.5308641791343689, 0.6296296119689941, 0.604938268661499, 0.604938268661499, 0.7407407164573669, 0.7283950448036194, 0.8641975522041321, 0.8765432238578796, 0.8888888955116272, 0.7777777910232544, 0.7283950448036194, 0.7160493731498718, 0.8271604776382446, 0.8765432238578796, 0.8395061492919922, 0.9135802388191223, 0.8765432238578796, 0.9876543283462524, 0.9629629850387573, 0.8888888955116272, 0.7777777910232544, 1.0], [0.0, 0.0, 0.012345679104328156, 0.02469135820865631, 0.02469135820865631, 0.04938271641731262, 0.012345679104328156, 0.02469135820865631, 0.04938271641731262, 0.04938271641731262, 0.07407407462596893, 0.09876543283462524, 0.03703703731298447, 0.012345679104328156, 0.02469135820865631, 0.04938271641731262, 0.16049382090568542, 0.16049382090568542, 0.1975308656692505, 0.03703703731298447, 0.09876543283462524, 0.20987653732299805, 0.23456789553165436, 0.2222222238779068, 0.07407407462596893, 0.14814814925193787, 0.17283950746059418, 0.1111111119389534, 0.23456789553165436, 0.2839506268501282, 0.3333333432674408, 0.37037035822868347, 0.3086419701576233, 0.3086419701576233, 0.1975308656692505, 0.3827160596847534, 0.37037035822868347, 0.29629629850387573, 0.3086419701576233, 0.3333333432674408, 0.3580246865749359, 0.3827160596847534, 0.48148149251937866, 0.395061731338501, 0.4444444477558136, 0.48148149251937866, NaN, 0.5802469253540039, 0.5308641791343689, 0.5185185074806213, 0.654321014881134, 0.4938271641731262, 0.5308641791343689, 0.5925925970077515, 0.7283950448036194, 0.604938268661499, 0.6419752836227417, 0.6666666865348816, 0.8148148059844971, 0.7160493731498718, 0.7283950448036194, 0.7160493731498718, 0.8271604776382446, 0.8518518805503845, 0.9259259104728699, 0.8518518805503845, 0.9506173133850098, 0.9012345671653748, 0.8888888955116272, 1.0, 0.9135802388191223, 1.0, 0.9135802388191223, 1.0], [0.0, 0.02469135820865631, 0.012345679104328156, 0.06172839552164078, 0.04938271641731262, 0.03703703731298447, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.012345679104328156, 0.07407407462596893, 0.0, 0.03703703731298447, 0.12345679104328156, 0.03703703731298447, 0.06172839552164078, 0.09876543283462524, 0.03703703731298447, 0.1358024626970291, 0.06172839552164078, 0.16049382090568542, 0.17283950746059418, 0.14814814925193787, 0.20987653732299805, 0.18518517911434174, 0.1975308656692505, 0.25925925374031067, 0.2222222238779068, 0.2839506268501282, 0.2839506268501282, 0.25925925374031067, 0.16049382090568542, 0.25925925374031067, 0.40740740299224854, 0.3086419701576233, 0.3333333432674408, 0.3580246865749359, 0.3827160596847534, 0.3333333432674408, 0.395061731338501, 0.45679011940956116, 0.3827160596847534, 0.3827160596847534, 0.45679011940956116, 0.4197530746459961, NaN, 0.5679012537002563, 0.5925925970077515, 0.7283950448036194, 0.5925925970077515, 0.5061728358268738, 0.5925925970077515, 0.6666666865348816, 0.654321014881134, 0.6666666865348816, 0.7777777910232544, 0.8765432238578796, 0.7777777910232544, 0.790123462677002, 0.8395061492919922, 0.8395061492919922, 0.8148148059844971, 0.9629629850387573, 0.9506173133850098, 0.9629629850387573, 0.9259259104728699, 0.8765432238578796, 0.9259259104728699, 0.9629629850387573, 1.0, 0.8888888955116272, 1.0], [0.0, 0.02469135820865631, 0.03703703731298447, 0.08641975373029709, 0.04938271641731262, 0.08641975373029709, 0.04938271641731262, 0.02469135820865631, 0.02469135820865631, 0.03703703731298447, 0.03703703731298447, 0.02469135820865631, 0.09876543283462524, 0.04938271641731262, 0.04938271641731262, 0.16049382090568542, 0.07407407462596893, 0.12345679104328156, 0.16049382090568542, 0.06172839552164078, 0.14814814925193787, 0.04938271641731262, 0.18518517911434174, 0.1111111119389534, 0.1975308656692505, 0.2469135820865631, 0.17283950746059418, 0.23456789553165436, 0.2839506268501282, 0.1975308656692505, 0.3333333432674408, 0.23456789553165436, 0.1358024626970291, 0.18518517911434174, 0.3086419701576233, 0.40740740299224854, 0.1111111119389534, 0.3086419701576233, 0.32098764181137085, 0.2839506268501282, 0.1975308656692505, 0.3827160596847534, 0.5061728358268738, 0.3580246865749359, 0.45679011940956116, 0.3580246865749359, 0.4691357910633087, 0.43209877610206604, NaN, 0.5679012537002563, 0.4691357910633087, 0.5679012537002563, 0.5802469253540039, 0.654321014881134, 0.604938268661499, 0.7530864477157593, 0.6790123581886292, 0.790123462677002, 0.8765432238578796, 0.6296296119689941, 0.654321014881134, 0.8271604776382446, 0.8271604776382446, 0.8518518805503845, 0.8395061492919922, 0.9753086566925049, 0.8395061492919922, 0.8518518805503845, 0.8148148059844971, 0.9506173133850098, 0.9629629850387573, 1.0, 0.8888888955116272, 1.0], [0.0, 0.012345679104328156, 0.012345679104328156, 0.03703703731298447, 0.02469135820865631, 0.012345679104328156, 0.03703703731298447, 0.04938271641731262, 0.012345679104328156, 0.0, 0.04938271641731262, 0.12345679104328156, 0.07407407462596893, 0.012345679104328156, 0.02469135820865631, 0.1111111119389534, 0.14814814925193787, 0.1111111119389534, 0.03703703731298447, 0.04938271641731262, 0.14814814925193787, 0.14814814925193787, 0.09876543283462524, 0.08641975373029709, 0.18518517911434174, 0.2222222238779068, 0.18518517911434174, 0.17283950746059418, 0.2469135820865631, 0.12345679104328156, 0.25925925374031067, 0.18518517911434174, 0.2222222238779068, 0.14814814925193787, 0.2839506268501282, 0.3086419701576233, 0.2716049253940582, 0.23456789553165436, 0.34567901492118835, 0.32098764181137085, 0.34567901492118835, 0.32098764181137085, 0.3580246865749359, 0.3580246865749359, 0.3827160596847534, 0.4691357910633087, 0.48148149251937866, 0.40740740299224854, 0.43209877610206604, NaN, 0.5061728358268738, 0.48148149251937866, 0.48148149251937866, 0.5308641791343689, 0.604938268661499, 0.7283950448036194, 0.604938268661499, 0.6913580298423767, 0.7160493731498718, 0.6913580298423767, 0.6913580298423767, 0.7654321193695068, 0.7283950448036194, 0.7777777910232544, 0.9753086566925049, 0.8888888955116272, 0.9876543283462524, 0.9382715821266174, 0.9012345671653748, 0.8395061492919922, 0.9629629850387573, 0.8888888955116272, 1.0, 1.0], [0.012345679104328156, 0.02469135820865631, 0.02469135820865631, 0.03703703731298447, 0.012345679104328156, 0.03703703731298447, 0.03703703731298447, 0.012345679104328156, 0.03703703731298447, 0.03703703731298447, 0.03703703731298447, 0.012345679104328156, 0.07407407462596893, 0.012345679104328156, 0.04938271641731262, 0.1358024626970291, 0.03703703731298447, 0.08641975373029709, 0.04938271641731262, 0.06172839552164078, 0.12345679104328156, 0.09876543283462524, 0.07407407462596893, 0.1111111119389534, 0.12345679104328156, 0.1975308656692505, 0.16049382090568542, 0.20987653732299805, 0.2716049253940582, 0.16049382090568542, 0.2222222238779068, 0.2716049253940582, 0.2222222238779068, 0.14814814925193787, 0.25925925374031067, 0.3827160596847534, 0.2716049253940582, 0.25925925374031067, 0.3333333432674408, 0.34567901492118835, 0.3086419701576233, 0.3086419701576233, 0.34567901492118835, 0.32098764181137085, 0.3580246865749359, 0.4444444477558136, 0.34567901492118835, 0.2716049253940582, 0.5308641791343689, 0.4938271641731262, NaN, 0.5185185074806213, 0.45679011940956116, 0.5802469253540039, 0.5308641791343689, 0.5925925970077515, 0.6296296119689941, 0.6790123581886292, 0.8395061492919922, 0.7654321193695068, 0.7777777910232544, 0.7530864477157593, 0.8148148059844971, 0.8148148059844971, 0.9135802388191223, 0.9012345671653748, 0.9320987462997437, 0.9135802388191223, 0.8395061492919922, 0.9259259104728699, 0.9753086566925049, 1.0, 0.8888888955116272, 1.0], [0.0, 0.02469135820865631, 0.02469135820865631, 0.07407407462596893, 0.06172839552164078, 0.07407407462596893, 0.02469135820865631, 0.03703703731298447, 0.03703703731298447, 0.02469135820865631, 0.04938271641731262, 0.12345679104328156, 0.09876543283462524, 0.1358024626970291, 0.09876543283462524, 0.07407407462596893, 0.16049382090568542, 0.03703703731298447, 0.1358024626970291, 0.14814814925193787, 0.1358024626970291, 0.14814814925193787, 0.1358024626970291, 0.2222222238779068, 0.23456789553165436, 0.32098764181137085, 0.16049382090568542, 0.08641975373029709, 0.2222222238779068, 0.18518517911434174, 0.2839506268501282, 0.2222222238779068, 0.2716049253940582, 0.32098764181137085, 0.2469135820865631, 0.3580246865749359, 0.1975308656692505, 0.1975308656692505, 0.07407407462596893, 0.29629629850387573, 0.29629629850387573, 0.395061731338501, 0.3580246865749359, 0.29629629850387573, 0.4444444477558136, 0.3086419701576233, 0.5061728358268738, 0.40740740299224854, 0.43209877610206604, 0.5185185074806213, 0.48148149251937866, NaN, 0.5432098507881165, 0.4691357910633087, 0.5432098507881165, 0.5679012537002563, 0.6790123581886292, 0.6296296119689941, 0.6790123581886292, 0.6172839403152466, 0.6172839403152466, 0.7037037014961243, 0.8024691343307495, 0.8148148059844971, 0.8024691343307495, 0.8888888955116272, 0.8518518805503845, 0.7530864477157593, 0.8765432238578796, 0.8765432238578796, 0.8518518805503845, 0.8765432238578796, 0.9506173133850098, 1.0], [0.02469135820865631, 0.02469135820865631, 0.04938271641731262, 0.03703703731298447, 0.06172839552164078, 0.04938271641731262, 0.1111111119389534, 0.07407407462596893, 0.12345679104328156, 0.09876543283462524, 0.1358024626970291, 0.18518517911434174, 0.12345679104328156, 0.12345679104328156, 0.04938271641731262, 0.08641975373029709, 0.2469135820865631, 0.17283950746059418, 0.1111111119389534, 0.16049382090568542, 0.17283950746059418, 0.1975308656692505, 0.17283950746059418, 0.16049382090568542, 0.18518517911434174, 0.08641975373029709, 0.16049382090568542, 0.14814814925193787, 0.06172839552164078, 0.25925925374031067, 0.09876543283462524, 0.34567901492118835, 0.3580246865749359, 0.3827160596847534, 0.3086419701576233, 0.16049382090568542, 0.37037035822868347, 0.29629629850387573, 0.3086419701576233, 0.3580246865749359, 0.4444444477558136, 0.4197530746459961, 0.24074074625968933, 0.4444444477558136, 0.37037035822868347, 0.4691357910633087, 0.4691357910633087, 0.4938271641731262, 0.4197530746459961, 0.5185185074806213, 0.5432098507881165, 0.45679011940956116, NaN, 0.5061728358268738, 0.5802469253540039, 0.5555555820465088, 0.5925925970077515, 0.6666666865348816, 0.5679012537002563, 0.6296296119689941, 0.6296296119689941, 0.6790123581886292, 0.7654321193695068, 0.8148148059844971, 0.6419752836227417, 0.7160493731498718, 0.6666666865348816, 0.6913580298423767, 0.7283950448036194, 0.6790123581886292, 0.8395061492919922, 0.7037037014961243, 0.9753086566925049, 1.0], [0.012345679104328156, 0.04938271641731262, 0.02469135820865631, 0.1111111119389534, 0.1358024626970291, 0.09876543283462524, 0.16049382090568542, 0.16049382090568542, 0.1975308656692505, 0.06172839552164078, 0.1975308656692505, 0.17283950746059418, 0.14814814925193787, 0.2469135820865631, 0.23456789553165436, 0.04938271641731262, 0.18518517911434174, 0.1111111119389534, 0.25925925374031067, 0.2716049253940582, 0.14814814925193787, 0.18518517911434174, 0.2716049253940582, 0.25925925374031067, 0.2839506268501282, 0.23456789553165436, 0.14814814925193787, 0.1975308656692505, 0.25925925374031067, 0.25925925374031067, 0.14814814925193787, 0.23456789553165436, 0.3333333432674408, 0.32098764181137085, 0.3580246865749359, 0.2469135820865631, 0.3086419701576233, 0.32098764181137085, 0.3827160596847534, 0.2716049253940582, 0.34567901492118835, 0.4197530746459961, 0.3086419701576233, 0.34567901492118835, 0.3333333432674408, 0.37037035822868347, 0.40740740299224854, 0.40740740299224854, 0.34567901492118835, 0.4691357910633087, 0.4197530746459961, 0.5308641791343689, 0.4938271641731262, NaN, 0.43209877610206604, 0.5061728358268738, 0.5679012537002563, 0.604938268661499, 0.5185185074806213, 0.4691357910633087, 0.48148149251937866, 0.6172839403152466, 0.5925925970077515, 0.6419752836227417, 0.6790123581886292, 0.6172839403152466, 0.7283950448036194, 0.5925925970077515, 0.7530864477157593, 0.5802469253540039, 0.6172839403152466, 0.6419752836227417, 0.7654321193695068, 0.8765432238578796], [0.0, 0.0, 0.012345679104328156, 0.03703703731298447, 0.0, 0.02469135820865631, 0.02469135820865631, 0.0, 0.012345679104328156, 0.012345679104328156, 0.02469135820865631, 0.09876543283462524, 0.02469135820865631, 0.0, 0.02469135820865631, 0.12345679104328156, 0.09876543283462524, 0.07407407462596893, 0.04938271641731262, 0.0, 0.02469135820865631, 0.14814814925193787, 0.09876543283462524, 0.04938271641731262, 0.07407407462596893, 0.2222222238779068, 0.12345679104328156, 0.16049382090568542, 0.2469135820865631, 0.012345679104328156, 0.2222222238779068, 0.25925925374031067, 0.14814814925193787, 0.08641975373029709, 0.1975308656692505, 0.34567901492118835, 0.2469135820865631, 0.12345679104328156, 0.2469135820865631, 0.25925925374031067, 0.17283950746059418, 0.12345679104328156, 0.3580246865749359, 0.18518517911434174, 0.2839506268501282, 0.395061731338501, 0.2716049253940582, 0.3333333432674408, 0.395061731338501, 0.395061731338501, 0.4691357910633087, 0.45679011940956116, 0.4197530746459961, 0.5679012537002563, NaN, 0.4938271641731262, 0.5061728358268738, 0.7037037014961243, 0.7777777910232544, 0.7037037014961243, 0.7037037014961243, 0.7407407164573669, 0.7777777910232544, 0.8024691343307495, 0.8518518805503845, 0.8641975522041321, 0.8765432238578796, 0.9259259104728699, 0.8765432238578796, 0.8888888955116272, 0.9506173133850098, 0.9876543283462524, 0.9506173133850098, 1.0], [0.0, 0.012345679104328156, 0.0, 0.03703703731298447, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.02469135820865631, 0.02469135820865631, 0.03703703731298447, 0.12345679104328156, 0.04938271641731262, 0.012345679104328156, 0.02469135820865631, 0.12345679104328156, 0.14814814925193787, 0.04938271641731262, 0.03703703731298447, 0.02469135820865631, 0.08641975373029709, 0.16049382090568542, 0.07407407462596893, 0.06172839552164078, 0.08641975373029709, 0.18518517911434174, 0.14814814925193787, 0.16049382090568542, 0.20987653732299805, 0.09876543283462524, 0.1975308656692505, 0.1975308656692505, 0.17283950746059418, 0.1111111119389534, 0.2469135820865631, 0.3086419701576233, 0.17283950746059418, 0.20987653732299805, 0.3086419701576233, 0.29629629850387573, 0.2716049253940582, 0.2469135820865631, 0.2839506268501282, 0.2716049253940582, 0.2839506268501282, 0.395061731338501, 0.395061731338501, 0.34567901492118835, 0.2469135820865631, 0.2716049253940582, 0.40740740299224854, 0.43209877610206604, 0.4444444477558136, 0.4938271641731262, 0.5061728358268738, NaN, 0.4938271641731262, 0.6790123581886292, 0.654321014881134, 0.6790123581886292, 0.6666666865348816, 0.7160493731498718, 0.6790123581886292, 0.7530864477157593, 0.9382715821266174, 0.8765432238578796, 0.9506173133850098, 0.9259259104728699, 0.8888888955116272, 0.8271604776382446, 0.9629629850387573, 0.9012345671653748, 0.9876543283462524, 1.0], [0.0, 0.02469135820865631, 0.012345679104328156, 0.07407407462596893, 0.08641975373029709, 0.08641975373029709, 0.04938271641731262, 0.0, 0.02469135820865631, 0.03703703731298447, 0.06172839552164078, 0.1358024626970291, 0.1111111119389534, 0.1111111119389534, 0.14814814925193787, 0.03703703731298447, 0.14814814925193787, 0.04938271641731262, 0.1111111119389534, 0.16049382090568542, 0.09876543283462524, 0.14814814925193787, 0.12345679104328156, 0.17283950746059418, 0.20987653732299805, 0.23456789553165436, 0.1111111119389534, 0.09876543283462524, 0.18518517911434174, 0.18518517911434174, 0.1358024626970291, 0.14814814925193787, 0.2222222238779068, 0.25925925374031067, 0.20987653732299805, 0.3580246865749359, 0.14814814925193787, 0.17283950746059418, 0.03703703731298447, 0.25925925374031067, 0.2839506268501282, 0.34567901492118835, 0.2469135820865631, 0.29629629850387573, 0.3333333432674408, 0.25925925374031067, 0.3580246865749359, 0.3333333432674408, 0.32098764181137085, 0.395061731338501, 0.37037035822868347, 0.32098764181137085, 0.40740740299224854, 0.43209877610206604, 0.4938271641731262, 0.5061728358268738, NaN, 0.5432098507881165, 0.5555555820465088, 0.5308641791343689, 0.5432098507881165, 0.5925925970077515, 0.7654321193695068, 0.7777777910232544, 0.6790123581886292, 0.6419752836227417, 0.6790123581886292, 0.7037037014961243, 0.790123462677002, 0.7777777910232544, 0.7777777910232544, 0.790123462677002, 0.8765432238578796, 1.0], [0.0, 0.02469135820865631, 0.0, 0.08641975373029709, 0.03703703731298447, 0.08641975373029709, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.07407407462596893, 0.0, 0.02469135820865631, 0.1111111119389534, 0.0, 0.02469135820865631, 0.17283950746059418, 0.0, 0.12345679104328156, 0.0, 0.1975308656692505, 0.17283950746059418, 0.14814814925193787, 0.2716049253940582, 0.1111111119389534, 0.14814814925193787, 0.2716049253940582, 0.09876543283462524, 0.29629629850387573, 0.02469135820865631, 0.1111111119389534, 0.1358024626970291, 0.2222222238779068, 0.3086419701576233, 0.04938271641731262, 0.18518517911434174, 0.12345679104328156, 0.2716049253940582, 0.16049382090568542, 0.1358024626970291, 0.34567901492118835, 0.20987653732299805, 0.34567901492118835, 0.2716049253940582, 0.3333333432674408, 0.2222222238779068, 0.20987653732299805, 0.3086419701576233, 0.32098764181137085, 0.37037035822868347, 0.3333333432674408, 0.395061731338501, 0.29629629850387573, 0.32098764181137085, 0.45679011940956116, NaN, 0.3580246865749359, 0.48148149251937866, 0.48148149251937866, 0.7962962985038757, 0.48148149251937866, 0.5061728358268738, 0.604938268661499, 0.9259259104728699, 0.6172839403152466, 0.654321014881134, 0.7654321193695068, 0.5555555820465088, 0.654321014881134, 0.5802469253540039, 0.7407407164573669, 0.8888888955116272], [0.012345679104328156, 0.04938271641731262, 0.012345679104328156, 0.07407407462596893, 0.06172839552164078, 0.04938271641731262, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.06172839552164078, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.02469135820865631, 0.03703703731298447, 0.09876543283462524, 0.012345679104328156, 0.12345679104328156, 0.02469135820865631, 0.12345679104328156, 0.1111111119389534, 0.09876543283462524, 0.16049382090568542, 0.012345679104328156, 0.04938271641731262, 0.17283950746059418, 0.09876543283462524, 0.16049382090568542, 0.03703703731298447, 0.09876543283462524, 0.12345679104328156, 0.17283950746059418, 0.23456789553165436, 0.02469135820865631, 0.14814814925193787, 0.1358024626970291, 0.09876543283462524, 0.12345679104328156, 0.17283950746059418, 0.2222222238779068, 0.1358024626970291, 0.25925925374031067, 0.1358024626970291, 0.18518517911434174, 0.12345679104328156, 0.12345679104328156, 0.2839506268501282, 0.16049382090568542, 0.32098764181137085, 0.43209877610206604, 0.48148149251937866, 0.2222222238779068, 0.34567901492118835, 0.4444444477558136, 0.6419752836227417, NaN, 0.3827160596847534, 0.3827160596847534, 0.7037037014961243, 0.5925925970077515, 0.6419752836227417, 0.6296296119689941, 0.8024691343307495, 0.7654321193695068, 0.6419752836227417, 0.7654321193695068, 0.6666666865348816, 0.7160493731498718, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.08641975373029709, 0.012345679104328156, 0.02469135820865631, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.012345679104328156, 0.0, 0.0, 0.07407407462596893, 0.06172839552164078, 0.04938271641731262, 0.16049382090568542, 0.08641975373029709, 0.12345679104328156, 0.16049382090568542, 0.02469135820865631, 0.06172839552164078, 0.1111111119389534, 0.2222222238779068, 0.18518517911434174, 0.03703703731298447, 0.2469135820865631, 0.02469135820865631, 0.06172839552164078, 0.17283950746059418, 0.37037035822868347, 0.17283950746059418, 0.17283950746059418, 0.12345679104328156, 0.2839506268501282, 0.2222222238779068, 0.37037035822868347, 0.3086419701576233, 0.23456789553165436, 0.3827160596847534, 0.37037035822868347, 0.5308641791343689, 0.29629629850387573, 0.32098764181137085, 0.4691357910633087, 0.5185185074806213, 0.6172839403152466, NaN, 0.6604938507080078, 0.5308641791343689, 0.7037037014961243, 0.7037037014961243, 0.6666666865348816, 0.7160493731498718, 0.6666666865348816, 0.654321014881134, 0.6666666865348816, 1.0, 0.8641975522041321, 0.8641975522041321, 0.7777777910232544, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.08641975373029709, 0.012345679104328156, 0.02469135820865631, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06172839552164078, 0.06172839552164078, 0.04938271641731262, 0.16049382090568542, 0.08641975373029709, 0.12345679104328156, 0.16049382090568542, 0.03703703731298447, 0.04938271641731262, 0.1111111119389534, 0.2222222238779068, 0.17283950746059418, 0.03703703731298447, 0.2469135820865631, 0.012345679104328156, 0.06172839552164078, 0.17283950746059418, 0.3580246865749359, 0.17283950746059418, 0.16049382090568542, 0.1111111119389534, 0.2716049253940582, 0.20987653732299805, 0.34567901492118835, 0.3086419701576233, 0.2222222238779068, 0.3827160596847534, 0.37037035822868347, 0.5185185074806213, 0.29629629850387573, 0.3333333432674408, 0.45679011940956116, 0.5185185074806213, 0.6172839403152466, 0.3395061790943146, NaN, 0.5308641791343689, 0.7037037014961243, 0.7037037014961243, 0.6666666865348816, 0.7160493731498718, 0.6666666865348816, 0.654321014881134, 0.6666666865348816, 1.0, 0.8641975522041321, 0.8641975522041321, 0.7777777910232544, 0.8888888955116272], [0.0, 0.02469135820865631, 0.0, 0.07407407462596893, 0.03703703731298447, 0.06172839552164078, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06172839552164078, 0.0, 0.012345679104328156, 0.09876543283462524, 0.0, 0.012345679104328156, 0.16049382090568542, 0.0, 0.1111111119389534, 0.0, 0.14814814925193787, 0.17283950746059418, 0.09876543283462524, 0.2716049253940582, 0.07407407462596893, 0.1111111119389534, 0.2716049253940582, 0.09876543283462524, 0.2716049253940582, 0.0, 0.08641975373029709, 0.12345679104328156, 0.23456789553165436, 0.2716049253940582, 0.012345679104328156, 0.1358024626970291, 0.1111111119389534, 0.25925925374031067, 0.12345679104328156, 0.1358024626970291, 0.2839506268501282, 0.17283950746059418, 0.3333333432674408, 0.2222222238779068, 0.2839506268501282, 0.16049382090568542, 0.17283950746059418, 0.23456789553165436, 0.2469135820865631, 0.29629629850387573, 0.32098764181137085, 0.3827160596847534, 0.25925925374031067, 0.2839506268501282, 0.40740740299224854, 0.20370370149612427, 0.29629629850387573, 0.4691357910633087, 0.4691357910633087, NaN, 0.48148149251937866, 0.4938271641731262, 0.5679012537002563, 0.8641975522041321, 0.5679012537002563, 0.6296296119689941, 0.7283950448036194, 0.5555555820465088, 0.6172839403152466, 0.5802469253540039, 0.7407407164573669, 0.8888888955116272], [0.012345679104328156, 0.0, 0.02469135820865631, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.02469135820865631, 0.1358024626970291, 0.02469135820865631, 0.08641975373029709, 0.02469135820865631, 0.012345679104328156, 0.1358024626970291, 0.03703703731298447, 0.03703703731298447, 0.09876543283462524, 0.02469135820865631, 0.1358024626970291, 0.03703703731298447, 0.02469135820865631, 0.06172839552164078, 0.02469135820865631, 0.07407407462596893, 0.06172839552164078, 0.012345679104328156, 0.02469135820865631, 0.03703703731298447, 0.20987653732299805, 0.1358024626970291, 0.1358024626970291, 0.02469135820865631, 0.04938271641731262, 0.16049382090568542, 0.04938271641731262, 0.1358024626970291, 0.14814814925193787, 0.14814814925193787, 0.04938271641731262, 0.03703703731298447, 0.14814814925193787, 0.1358024626970291, 0.2716049253940582, 0.17283950746059418, 0.16049382090568542, 0.17283950746059418, 0.2716049253940582, 0.18518517911434174, 0.1975308656692505, 0.23456789553165436, 0.40740740299224854, 0.2222222238779068, 0.32098764181137085, 0.23456789553165436, 0.5185185074806213, 0.40740740299224854, 0.29629629850387573, 0.29629629850387573, 0.5185185074806213, NaN, 0.8395061492919922, 0.5802469253540039, 0.5185185074806213, 0.5925925970077515, 0.6666666865348816, 0.6296296119689941, 0.3827160596847534, 0.5679012537002563, 0.6172839403152466, 0.9135802388191223, 0.9506173133850098], [0.0, 0.0, 0.03703703731298447, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.03703703731298447, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.12345679104328156, 0.0, 0.03703703731298447, 0.02469135820865631, 0.012345679104328156, 0.1358024626970291, 0.04938271641731262, 0.02469135820865631, 0.03703703731298447, 0.012345679104328156, 0.14814814925193787, 0.06172839552164078, 0.012345679104328156, 0.03703703731298447, 0.0, 0.08641975373029709, 0.04938271641731262, 0.02469135820865631, 0.03703703731298447, 0.03703703731298447, 0.16049382090568542, 0.14814814925193787, 0.07407407462596893, 0.02469135820865631, 0.02469135820865631, 0.18518517911434174, 0.06172839552164078, 0.14814814925193787, 0.14814814925193787, 0.16049382090568542, 0.04938271641731262, 0.03703703731298447, 0.16049382090568542, 0.04938271641731262, 0.2839506268501282, 0.14814814925193787, 0.18518517911434174, 0.14814814925193787, 0.2222222238779068, 0.18518517911434174, 0.18518517911434174, 0.18518517911434174, 0.3580246865749359, 0.1975308656692505, 0.2469135820865631, 0.2222222238779068, 0.4938271641731262, 0.3580246865749359, 0.29629629850387573, 0.29629629850387573, 0.5061728358268738, 0.16049382090568542, NaN, 0.5555555820465088, 0.4938271641731262, 0.5925925970077515, 0.6666666865348816, 0.6296296119689941, 0.34567901492118835, 0.5061728358268738, 0.5308641791343689, 0.8888888955116272, 0.9506173133850098], [0.0, 0.012345679104328156, 0.012345679104328156, 0.0, 0.0, 0.0, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.0, 0.02469135820865631, 0.012345679104328156, 0.0, 0.012345679104328156, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.0, 0.03703703731298447, 0.03703703731298447, 0.012345679104328156, 0.012345679104328156, 0.08641975373029709, 0.02469135820865631, 0.12345679104328156, 0.08641975373029709, 0.06172839552164078, 0.08641975373029709, 0.06172839552164078, 0.04938271641731262, 0.07407407462596893, 0.16049382090568542, 0.18518517911434174, 0.1111111119389534, 0.07407407462596893, 0.1975308656692505, 0.06172839552164078, 0.17283950746059418, 0.07407407462596893, 0.03703703731298447, 0.16049382090568542, 0.02469135820865631, 0.08641975373029709, 0.1975308656692505, 0.3580246865749359, 0.32098764181137085, 0.14814814925193787, 0.06172839552164078, 0.32098764181137085, 0.395061731338501, 0.37037035822868347, 0.3333333432674408, 0.3333333432674408, 0.43209877610206604, 0.4197530746459961, 0.4444444477558136, NaN, 0.4691357910633087, 0.4444444477558136, 0.43209877610206604, 0.6790123581886292, 0.5061728358268738, 0.5432098507881165, 0.6172839403152466, 0.7654321193695068, 1.0], [0.0, 0.0, 0.0, 0.04938271641731262, 0.03703703731298447, 0.03703703731298447, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.04938271641731262, 0.0, 0.07407407462596893, 0.0, 0.07407407462596893, 0.08641975373029709, 0.03703703731298447, 0.07407407462596893, 0.0, 0.012345679104328156, 0.12345679104328156, 0.04938271641731262, 0.1358024626970291, 0.0, 0.06172839552164078, 0.1111111119389534, 0.1111111119389534, 0.20987653732299805, 0.0, 0.09876543283462524, 0.02469135820865631, 0.06172839552164078, 0.08641975373029709, 0.08641975373029709, 0.2222222238779068, 0.09876543283462524, 0.20987653732299805, 0.12345679104328156, 0.14814814925193787, 0.04938271641731262, 0.02469135820865631, 0.1111111119389534, 0.09876543283462524, 0.1111111119389534, 0.2839506268501282, 0.3827160596847534, 0.1358024626970291, 0.12345679104328156, 0.3580246865749359, 0.07407407462596893, 0.1975308656692505, 0.2839506268501282, 0.2839506268501282, 0.1358024626970291, 0.48148149251937866, 0.5061728358268738, 0.5308641791343689, NaN, 0.5432098507881165, 0.5555555820465088, 0.7160493731498718, 0.5308641791343689, 0.48148149251937866, 0.5802469253540039, 0.7407407164573669, 0.8765432238578796], [0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.0, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.0, 0.03703703731298447, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.04938271641731262, 0.02469135820865631, 0.1358024626970291, 0.08641975373029709, 0.08641975373029709, 0.09876543283462524, 0.1111111119389534, 0.04938271641731262, 0.07407407462596893, 0.14814814925193787, 0.17283950746059418, 0.08641975373029709, 0.06172839552164078, 0.1975308656692505, 0.03703703731298447, 0.16049382090568542, 0.04938271641731262, 0.03703703731298447, 0.16049382090568542, 0.012345679104328156, 0.06790123134851456, 0.14814814925193787, 0.3333333432674408, 0.2716049253940582, 0.12345679104328156, 0.04938271641731262, 0.32098764181137085, 0.3827160596847534, 0.23456789553165436, 0.3333333432674408, 0.3333333432674408, 0.43209877610206604, 0.40740740299224854, 0.40740740299224854, 0.5555555820465088, 0.45679011940956116, NaN, 0.43209877610206604, 0.790123462677002, 0.45679011940956116, 0.5432098507881165, 0.6172839403152466, 0.7654321193695068, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.012345679104328156, 0.012345679104328156, 0.0, 0.02469135820865631, 0.0, 0.0, 0.0, 0.012345679104328156, 0.04938271641731262, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.02469135820865631, 0.012345679104328156, 0.012345679104328156, 0.0, 0.1111111119389534, 0.03703703731298447, 0.1358024626970291, 0.012345679104328156, 0.012345679104328156, 0.0, 0.1358024626970291, 0.04938271641731262, 0.18518517911434174, 0.12345679104328156, 0.1358024626970291, 0.02469135820865631, 0.09876543283462524, 0.14814814925193787, 0.06172839552164078, 0.08641975373029709, 0.09876543283462524, 0.07407407462596893, 0.14814814925193787, 0.06172839552164078, 0.08641975373029709, 0.2469135820865631, 0.3086419701576233, 0.40740740299224854, 0.07407407462596893, 0.07407407462596893, 0.29629629850387573, 0.34567901492118835, 0.3580246865749359, 0.34567901492118835, 0.34567901492118835, 0.37037035822868347, 0.3333333432674408, 0.3333333432674408, 0.5679012537002563, 0.4444444477558136, 0.5679012537002563, NaN, 0.5925925970077515, 0.4938271641731262, 0.5925925970077515, 0.43209877610206604, 0.604938268661499, 0.9382715821266174], [0.0, 0.04938271641731262, 0.0, 0.08641975373029709, 0.08641975373029709, 0.08641975373029709, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.09876543283462524, 0.012345679104328156, 0.04938271641731262, 0.0, 0.0, 0.02469135820865631, 0.08641975373029709, 0.04938271641731262, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.09876543283462524, 0.07407407462596893, 0.012345679104328156, 0.04938271641731262, 0.08641975373029709, 0.1111111119389534, 0.09876543283462524, 0.0, 0.1975308656692505, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.012345679104328156, 0.1358024626970291, 0.0, 0.20987653732299805, 0.20987653732299805, 0.12345679104328156, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.12345679104328156, 0.1111111119389534, 0.12345679104328156, 0.18518517911434174, 0.09876543283462524, 0.16049382090568542, 0.12345679104328156, 0.2716049253940582, 0.2469135820865631, 0.12345679104328156, 0.1111111119389534, 0.20987653732299805, 0.23456789553165436, 0.23456789553165436, 0.3333333432674408, 0.3333333432674408, 0.2716049253940582, 0.37037035822868347, 0.37037035822868347, 0.32098764181137085, 0.2839506268501282, 0.20987653732299805, 0.40740740299224854, NaN, 0.40740740299224854, 0.40740740299224854, 0.43209877610206604, 0.5308641791343689, 0.8518518805503845], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.08641975373029709, 0.012345679104328156, 0.0, 0.1111111119389534, 0.0, 0.0, 0.02469135820865631, 0.09876543283462524, 0.0, 0.0, 0.012345679104328156, 0.0, 0.07407407462596893, 0.04938271641731262, 0.16049382090568542, 0.07407407462596893, 0.12345679104328156, 0.32098764181137085, 0.4197530746459961, 0.1111111119389534, 0.17283950746059418, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.0, 0.0, 0.4444444477558136, 0.6172839403152466, 0.654321014881134, 0.4938271641731262, 0.4691357910633087, 0.5432098507881165, 0.5061728358268738, 0.5925925970077515, NaN, 0.6419752836227417, 0.6666666865348816, 0.7777777910232544, 0.8641975522041321], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.07407407462596893, 0.0, 0.02469135820865631, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.012345679104328156, 0.02469135820865631, 0.012345679104328156, 0.02469135820865631, 0.03703703731298447, 0.0, 0.12345679104328156, 0.02469135820865631, 0.02469135820865631, 0.0, 0.04938271641731262, 0.0, 0.04938271641731262, 0.012345679104328156, 0.07407407462596893, 0.03703703731298447, 0.012345679104328156, 0.012345679104328156, 0.1111111119389534, 0.02469135820865631, 0.08641975373029709, 0.03703703731298447, 0.08641975373029709, 0.03703703731298447, 0.03703703731298447, 0.03703703731298447, 0.02469135820865631, 0.14814814925193787, 0.16049382090568542, 0.3827160596847534, 0.04938271641731262, 0.03703703731298447, 0.2222222238779068, 0.34567901492118835, 0.2839506268501282, 0.1358024626970291, 0.1358024626970291, 0.3827160596847534, 0.43209877610206604, 0.4938271641731262, 0.45679011940956116, 0.5185185074806213, 0.45679011940956116, 0.40740740299224854, 0.5925925970077515, 0.3580246865749359, NaN, 0.5061728358268738, 0.6666666865348816, 0.7777777910232544], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.012345679104328156, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03703703731298447, 0.0, 0.0, 0.0, 0.03703703731298447, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.02469135820865631, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.12345679104328156, 0.29629629850387573, 0.3580246865749359, 0.012345679104328156, 0.09876543283462524, 0.20987653732299805, 0.4197530746459961, 0.0, 0.1358024626970291, 0.1358024626970291, 0.4197530746459961, 0.3827160596847534, 0.4691357910633087, 0.3827160596847534, 0.4197530746459961, 0.3827160596847534, 0.5679012537002563, 0.5679012537002563, 0.3333333432674408, 0.4938271641731262, NaN, 0.790123462677002, 0.9876543283462524], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08641975373029709, 0.0, 0.0, 0.0, 0.0, 0.07407407462596893, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.08641975373029709, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.04938271641731262, 0.02469135820865631, 0.23456789553165436, 0.04938271641731262, 0.012345679104328156, 0.12345679104328156, 0.25925925374031067, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.25925925374031067, 0.08641975373029709, 0.1111111119389534, 0.23456789553165436, 0.25925925374031067, 0.23456789553165436, 0.395061731338501, 0.4691357910633087, 0.2222222238779068, 0.3333333432674408, 0.20987653732299805, NaN, 0.8641975522041321], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.12345679104328156, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.04938271641731262, 0.04938271641731262, 0.0, 0.12345679104328156, 0.0, 0.06172839552164078, 0.14814814925193787, 0.1358024626970291, 0.2222222238779068, 0.012345679104328156, 0.1358024626970291, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9419922233},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9403010403},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8802638327},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8613225148},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8100794876},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8065279926},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7919837676},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7911381736},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7871638797},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7801454429},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7762557097},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7610350113},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.757145272},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7546084925},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.742939293},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7246744487},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.7074243212},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7037037056},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7035345874},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6945712845},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6905124332},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.674277018},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.6668357841},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6497547779},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6440047355},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6340267208},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6299678678},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6202435348},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6027397267},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6020632498},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.5949602589},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5525114164},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5491290403},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5403348547},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5377980728},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5234229667},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5217317777},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5054963666},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4956874704},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.4950109936},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4787755787},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4632166432},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4472349052},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4410620656},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4317605267},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.4123118533},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3997970572},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3992897028},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3881278559},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3708777267},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3677490257},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3644512097},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.3638592919},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.3499069826},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3307965518},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3307965515},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.3111787586},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2615423643},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2496194822},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2495349237},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2427701684},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2312700826},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2053103317},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1829866398},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1642144424},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1608320649},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.1580416036},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.1527143586},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1513614064},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1403686793},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1248097419},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1155082027},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0899712498},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0206325046}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..149c36da91037844344e538b861f66877eda9677
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,"AutoGluon 1.5 (extreme, 4h)",TabFM (default),"AutoGluon 1.4 (best, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),TabICLv2 (default),CatBoost (default),iLTM (tuned + ensembled),CatBoost (tuned + ensembled),RealMLP (tuned + ensembled),CatBoost (tuned),ChimeraBoost (tuned + ensembled),TabM (tuned + ensembled),XGBoost (tuned + ensembled),LightGBM (tuned + ensembled),ModernNCA (tuned + ensembled),ChimeraBoost (tuned),RealMLP (tuned),RealTabPFN-2.5 (tuned + ensembled),XGBoost (tuned),TabM (tuned),ChimeraBoost (default),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (default),LightGBM (tuned),TabICL (default),iLTM (tuned),ModernNCA (tuned),TabDPT (tuned + ensembled),TorchMLP (tuned + ensembled),TabDPT-Turbo (default),xRFM (tuned + ensembled),EBM (tuned + ensembled),XGBoost (default),TabM (default),TabSwift (default),xRFM (tuned),RealMLP (default),FastaiMLP (tuned + ensembled),BetaTabPFN (default),EBM (tuned),TorchMLP (tuned),TabDPT (tuned),EBM (default),LightGBM (default),iLTM (default),ModernNCA (default),RandomForest (tuned + ensembled),SAP-RPT-OSS (default),ExtraTrees (tuned + ensembled),RandomForest (tuned),FastaiMLP (tuned),TabDPT (default),LimiX (default),TorchMLP (default),ExtraTrees (tuned),FastaiMLP (default),Linear (tuned + ensembled),KNN (tuned + ensembled),TabSTAR (tuned),TabSTAR (tuned + ensembled),Linear (tuned),PerpetualBooster (tuned + ensembled),PerpetualBooster (tuned),ExtraTrees (default),Linear (default),RandomForest (default),OrionMSP (default),xRFM (default),TabSTAR (default),TabFlex (default),KNN (tuned),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.33333334,0.7160494,0.45679012,0.80246913,0.6419753,0.8888889,0.9259259,0.91358024,0.97530866,0.90123457,0.91358024,0.79012346,0.9506173,0.9259259,0.962963,0.9259259,0.962963,0.91358024,0.9876543,0.8518519,0.962963,0.91358024,0.91358024,0.97530866,0.8888889,0.8765432,0.9876543,0.91358024,0.9876543,0.90123457,0.97530866,0.9876543,0.962963,1.0,0.9382716,0.9876543,1.0,1.0,1.0,1.0,0.97530866,0.9506173,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9876543,1.0,0.97530866,0.9876543,1.0,1.0,1.0,1.0,0.9876543,1.0,1.0,1.0,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.6666667,,0.67901236,0.90123457,0.9506173,0.90123457,0.7654321,0.8271605,0.75308645,0.80246913,0.80246913,0.86419755,0.9876543,0.86419755,0.83950615,0.8765432,0.86419755,0.8148148,0.962963,0.90123457,1.0,0.8888889,0.962963,0.97530866,0.962963,0.9506173,0.8888889,0.9382716,0.962963,0.962963,0.962963,0.8765432,0.9876543,0.97530866,1.0,0.9506173,0.90123457,1.0,0.90123457,1.0,1.0,1.0,0.962963,1.0,1.0,1.0,1.0,0.97530866,0.97530866,0.9876543,0.97530866,0.97530866,0.97530866,0.9506173,1.0,0.9876543,0.97530866,0.97530866,0.9506173,1.0,1.0,0.97530866,1.0,1.0,0.9876543,1.0,0.9876543,1.0,0.9506173,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.28395063,0.32098764,,0.4691358,0.67901236,0.61728394,0.69135803,0.8271605,0.7407407,0.8518519,0.75308645,0.75308645,0.7160494,0.8518519,0.8271605,0.7654321,0.8518519,0.91358024,0.7037037,0.8765432,0.7777778,0.8765432,0.75308645,0.7407407,0.8888889,0.7407407,0.83950615,0.8888889,0.7777778,0.90123457,0.79012346,0.9506173,0.9506173,0.91358024,0.91358024,0.8271605,0.962963,0.962963,0.962963,0.90123457,0.9506173,0.97530866,0.962963,0.9506173,0.9506173,0.962963,0.9876543,0.9876543,0.962963,0.9876543,0.97530866,0.97530866,0.9506173,0.97530866,0.9876543,1.0,0.9876543,1.0,0.9876543,1.0,1.0,1.0,0.97530866,0.962963,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.54320985,0.09876543,0.5308642,,0.72839504,0.5308642,0.654321,0.67901236,0.6666667,0.7160494,0.654321,0.75308645,0.7654321,0.7160494,0.7037037,0.8765432,0.7407407,0.72839504,0.8518519,0.75308645,0.83950615,0.7407407,0.86419755,0.86419755,0.8888889,0.8518519,0.8518519,0.86419755,0.8765432,0.8148148,0.9506173,0.7654321,0.8518519,0.90123457,0.9876543,0.962963,0.7777778,0.9382716,0.8888889,0.9382716,0.90123457,0.962963,0.962963,0.9506173,0.9876543,0.9876543,0.97530866,0.9382716,0.91358024,0.962963,0.962963,0.9259259,0.962963,0.8888889,0.962963,0.962963,0.9259259,0.91358024,0.9259259,1.0,1.0,0.9259259,1.0,1.0,1.0,0.9506173,1.0,1.0,0.91358024,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.19753087,0.049382716,0.32098764,0.27160493,,0.5185185,0.4691358,0.50617284,0.5185185,0.5802469,0.5185185,0.6419753,0.5802469,0.5308642,0.654321,0.6666667,0.67901236,0.67901236,0.7777778,0.7037037,0.7160494,0.6419753,0.7777778,0.83950615,0.80246913,0.75308645,0.72839504,0.80246913,0.83950615,0.8271605,0.83950615,0.7407407,0.86419755,0.90123457,0.91358024,0.8518519,0.79012346,1.0,0.8518519,0.8888889,0.962963,0.9506173,0.9506173,0.962963,0.9382716,0.9382716,0.97530866,0.9506173,0.9506173,0.97530866,0.9876543,0.9382716,0.9382716,0.86419755,1.0,0.9876543,0.91358024,0.962963,0.9382716,1.0,1.0,0.962963,0.9876543,0.9876543,1.0,0.962963,1.0,1.0,0.91358024,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.3580247,0.09876543,0.38271606,0.4691358,0.4814815,,0.56790125,0.5925926,0.56790125,0.5308642,0.56790125,0.5802469,0.56790125,0.5802469,0.654321,0.67901236,0.5802469,0.5925926,0.8148148,0.60493827,0.69135803,0.60493827,0.83950615,0.8271605,0.75308645,0.8148148,0.72839504,0.7777778,0.8888889,0.79012346,0.86419755,0.6666667,0.75308645,0.8518519,0.9382716,0.90123457,0.69135803,0.90123457,0.8271605,0.91358024,0.90123457,0.90123457,0.9382716,0.9259259,0.9382716,0.91358024,0.9506173,0.962963,0.91358024,0.9876543,0.962963,0.9259259,0.9506173,0.90123457,0.97530866,0.9876543,0.91358024,0.91358024,0.9506173,0.97530866,0.97530866,0.9382716,1.0,1.0,1.0,0.962963,1.0,1.0,0.91358024,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.11111111,0.2345679,0.30864197,0.34567901,0.5308642,0.43209878,,0.5308642,0.45679012,0.5802469,0.5308642,0.5925926,0.5555556,0.5925926,0.60493827,0.5925926,0.7160494,0.6419753,0.5802469,0.72839504,0.61728394,0.79012346,0.60493827,0.7160494,0.7654321,0.7037037,0.72839504,0.69135803,0.654321,0.7654321,0.654321,0.9382716,0.90123457,0.8765432,0.7654321,0.79012346,0.9506173,0.80246913,0.8148148,0.83950615,0.90123457,0.9506173,0.79012346,0.9259259,0.9259259,0.9382716,0.9876543,0.97530866,0.9506173,0.962963,0.962963,0.97530866,0.8888889,0.83950615,0.97530866,0.9876543,0.9506173,0.9876543,0.9876543,1.0,1.0,1.0,0.9876543,0.9876543,0.97530866,1.0,0.9876543,0.9876543,1.0,1.0,0.9876543,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.074074075,0.17283951,0.17283951,0.32098764,0.49382716,0.4074074,0.4691358,,0.45679012,0.49382716,0.4691358,0.49382716,0.5308642,0.5925926,0.6296296,0.56790125,0.5802469,0.61728394,0.6419753,0.75308645,0.61728394,0.60493827,0.69135803,0.7407407,0.75308645,0.7160494,0.67901236,0.7654321,0.7160494,0.8518519,0.67901236,0.8765432,0.90123457,0.8765432,0.8148148,0.7407407,0.9506173,0.9506173,0.962963,0.8518519,0.91358024,0.9506173,0.90123457,0.9382716,0.91358024,1.0,0.97530866,0.97530866,0.97530866,0.9506173,0.9876543,0.962963,0.9259259,0.83950615,1.0,0.9876543,1.0,0.9876543,0.9876543,1.0,1.0,1.0,0.97530866,0.962963,0.9876543,1.0,0.9876543,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.086419754,0.24691358,0.25925925,0.33333334,0.4814815,0.43209878,0.54320985,0.54320985,,0.5802469,0.5802469,0.54320985,0.5555556,0.6296296,0.56790125,0.54320985,0.7037037,0.67901236,0.5802469,0.7037037,0.60493827,0.7654321,0.5925926,0.72839504,0.7037037,0.654321,0.7160494,0.70987654,0.6296296,0.7654321,0.6296296,0.91358024,0.9259259,0.8518519,0.7407407,0.75308645,0.9382716,0.8148148,0.8518519,0.8271605,0.90123457,0.9506173,0.79012346,0.9382716,0.9506173,0.91358024,0.9506173,0.97530866,0.97530866,0.9876543,0.962963,0.962963,0.8765432,0.80246913,0.9876543,0.97530866,0.97530866,0.9876543,0.9876543,1.0,1.0,1.0,0.97530866,0.9876543,0.9876543,1.0,0.9876543,0.9876543,1.0,1.0,0.9876543,0.9876543,1.0,1.0
+RealMLP (tuned + ensembled),0.024691358,0.19753087,0.14814815,0.28395063,0.41975307,0.4691358,0.41975307,0.50617284,0.41975307,,0.4691358,0.41975307,0.4691358,0.5555556,0.5802469,0.5925926,0.54320985,0.75308645,0.61728394,0.6666667,0.5308642,0.54320985,0.654321,0.67901236,0.75308645,0.6419753,0.69135803,0.7654321,0.67901236,0.83950615,0.7407407,0.8271605,0.83950615,0.8271605,0.8888889,0.72839504,0.8765432,1.0,0.97530866,0.83950615,0.91358024,0.9259259,0.90123457,0.9382716,0.9506173,0.90123457,0.9506173,0.97530866,0.962963,1.0,0.962963,0.97530866,0.90123457,0.9382716,0.9876543,0.97530866,0.962963,1.0,0.9876543,1.0,1.0,1.0,0.97530866,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.09876543,0.19753087,0.24691358,0.34567901,0.4814815,0.43209878,0.4691358,0.5308642,0.41975307,0.5308642,,0.5802469,0.56790125,0.6296296,0.5185185,0.5555556,0.7037037,0.6666667,0.54320985,0.72839504,0.5802469,0.72839504,0.5925926,0.67901236,0.7160494,0.6666667,0.67901236,0.654321,0.6296296,0.75308645,0.61728394,0.9259259,0.91358024,0.83950615,0.7777778,0.7407407,0.9259259,0.8271605,0.80246913,0.83950615,0.90123457,0.9382716,0.7777778,0.962963,0.91358024,0.9259259,0.9259259,0.97530866,0.962963,0.9506173,0.962963,0.9506173,0.86419755,0.80246913,0.97530866,0.962963,0.9382716,0.9876543,0.9876543,0.9876543,0.9876543,1.0,0.97530866,0.9876543,0.97530866,1.0,0.9876543,0.97530866,1.0,1.0,0.9876543,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.086419754,0.13580246,0.24691358,0.24691358,0.3580247,0.41975307,0.4074074,0.50617284,0.45679012,0.5802469,0.41975307,,0.56790125,0.5308642,0.5802469,0.50617284,0.7777778,0.69135803,0.56790125,0.654321,0.654321,0.86419755,0.6296296,0.6296296,0.6296296,0.61728394,0.7037037,0.69135803,0.6666667,0.67901236,0.67901236,0.8888889,0.91358024,0.75308645,0.80246913,0.6296296,0.91358024,0.83950615,0.83950615,0.86419755,0.9506173,0.8518519,0.8148148,0.97530866,0.86419755,0.9506173,0.90123457,0.9876543,0.97530866,0.8765432,0.9876543,0.8765432,0.8148148,0.8271605,0.90123457,0.8765432,0.86419755,1.0,0.9876543,1.0,1.0,1.0,0.86419755,0.8765432,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.91358024,1.0
+TabM (tuned + ensembled),0.20987654,0.012345679,0.28395063,0.2345679,0.41975307,0.43209878,0.44444445,0.4691358,0.44444445,0.5308642,0.43209878,0.43209878,,0.41975307,0.5185185,0.5802469,0.50617284,0.61728394,0.6296296,0.5555556,0.72839504,0.5802469,0.654321,0.6296296,0.67901236,0.6419753,0.7037037,0.8148148,0.72839504,0.6666667,0.79012346,0.61728394,0.7407407,0.7407407,0.962963,0.83950615,0.61728394,0.97530866,0.80246913,0.90123457,0.9382716,0.86419755,0.8888889,0.9259259,1.0,0.9259259,0.962963,0.9259259,0.90123457,0.9259259,0.9259259,0.90123457,0.8765432,0.8518519,0.97530866,0.9506173,0.8888889,0.9259259,0.9382716,1.0,1.0,0.9382716,0.97530866,1.0,1.0,0.97530866,1.0,1.0,0.90123457,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.049382716,0.13580246,0.14814815,0.28395063,0.4691358,0.41975307,0.4074074,0.4074074,0.37037036,0.44444445,0.37037036,0.4691358,0.5802469,,0.50617284,0.4691358,0.5308642,0.54320985,0.5555556,0.86419755,0.6419753,0.6296296,0.5925926,0.6419753,0.6666667,0.5555556,0.5925926,0.60493827,0.654321,0.7777778,0.654321,0.80246913,0.90123457,0.9382716,0.79012346,0.654321,0.90123457,0.8271605,0.7777778,0.86419755,0.9382716,0.8271605,0.8271605,0.97530866,0.962963,0.962963,0.9876543,1.0,0.9506173,0.9876543,0.9876543,0.86419755,0.8765432,0.75308645,1.0,0.9876543,0.8888889,1.0,0.9876543,0.9876543,0.9876543,1.0,0.91358024,0.962963,0.9876543,1.0,1.0,1.0,0.9876543,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.074074075,0.16049382,0.17283951,0.2962963,0.34567901,0.34567901,0.39506173,0.37037036,0.43209878,0.41975307,0.4814815,0.41975307,0.4814815,0.49382716,,0.45679012,0.56790125,0.5185185,0.5185185,0.654321,0.5925926,0.5802469,0.5555556,0.5802469,0.86419755,0.6296296,0.6666667,0.61728394,0.56790125,0.7037037,0.69135803,0.7407407,0.8148148,0.80246913,0.7777778,0.7160494,0.8148148,0.80246913,0.7407407,0.8518519,0.90123457,0.90123457,0.7777778,0.962963,0.97530866,0.97530866,0.97530866,0.962963,0.9506173,0.97530866,0.9506173,0.90123457,0.9506173,0.7654321,0.97530866,0.97530866,0.8518519,0.97530866,0.97530866,0.9876543,0.9876543,0.9876543,0.97530866,0.97530866,0.9876543,1.0,1.0,0.9876543,0.9506173,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.037037037,0.12345679,0.2345679,0.12345679,0.33333334,0.32098764,0.4074074,0.43209878,0.45679012,0.4074074,0.44444445,0.49382716,0.41975307,0.5308642,0.54320985,,0.5925926,0.54320985,0.5925926,0.5555556,0.5185185,0.5925926,0.60493827,0.61728394,0.5925926,0.60493827,0.60493827,0.7160494,0.654321,0.7037037,0.67901236,0.67901236,0.8148148,0.72839504,0.80246913,0.6666667,0.75308645,0.8765432,0.8518519,0.80246913,0.83950615,0.8271605,0.86419755,0.8518519,0.90123457,0.8271605,0.9506173,0.8765432,0.83950615,0.8888889,0.86419755,0.9259259,0.91358024,0.9506173,0.8765432,0.8765432,0.962963,0.8888889,0.9876543,0.91358024,0.91358024,0.90123457,0.9876543,0.9876543,0.97530866,0.97530866,1.0,0.9506173,1.0,1.0,0.9259259,1.0,1.0,1.0
+ChimeraBoost (tuned),0.074074075,0.13580246,0.14814815,0.25925925,0.32098764,0.41975307,0.28395063,0.41975307,0.2962963,0.45679012,0.2962963,0.22222222,0.49382716,0.4691358,0.43209878,0.4074074,,0.6419753,0.5308642,0.5802469,0.5555556,0.67901236,0.54320985,0.56790125,0.56790125,0.54320985,0.60493827,0.6419753,0.61728394,0.60493827,0.5802469,0.8518519,0.80246913,0.72839504,0.7654321,0.5925926,0.8888889,0.8271605,0.79012346,0.83950615,0.86419755,0.80246913,0.79012346,0.90123457,0.79012346,0.8888889,0.83950615,0.962963,0.9259259,0.8518519,0.962963,0.83950615,0.75308645,0.8148148,0.90123457,0.8518519,0.8518519,1.0,0.97530866,0.9876543,0.9876543,1.0,0.86419755,0.86419755,0.9876543,1.0,0.9876543,0.9876543,1.0,1.0,1.0,1.0,0.9259259,1.0
+RealMLP (tuned),0.037037037,0.18518518,0.086419754,0.27160493,0.32098764,0.4074074,0.3580247,0.38271606,0.32098764,0.24691358,0.33333334,0.30864197,0.38271606,0.45679012,0.4814815,0.45679012,0.3580247,,0.49382716,0.5925926,0.44444445,0.4814815,0.5308642,0.5802469,0.6666667,0.5308642,0.6419753,0.6419753,0.61728394,0.75308645,0.54320985,0.72839504,0.7160494,0.75308645,0.7407407,0.60493827,0.7777778,0.8888889,0.86419755,0.75308645,0.79012346,0.90123457,0.8148148,0.8148148,0.83950615,0.8148148,0.83950615,0.9382716,0.8765432,0.8888889,0.91358024,0.962963,0.8271605,0.8888889,0.9259259,0.9506173,0.9506173,0.97530866,0.962963,0.97530866,0.97530866,0.9876543,0.962963,0.9506173,0.9876543,1.0,0.9876543,0.97530866,0.97530866,1.0,0.97530866,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.086419754,0.037037037,0.2962963,0.14814815,0.22222222,0.18518518,0.41975307,0.3580247,0.41975307,0.38271606,0.45679012,0.43209878,0.37037036,0.44444445,0.4814815,0.4074074,0.4691358,0.50617284,,0.49382716,0.50617284,0.56790125,0.7777778,0.67901236,0.60493827,0.6666667,0.5802469,0.654321,0.654321,0.67901236,0.60493827,0.61728394,0.7037037,0.7407407,0.7160494,0.83950615,0.67901236,0.80246913,0.7037037,0.7777778,0.80246913,0.80246913,0.8765432,0.8518519,0.91358024,0.83950615,0.80246913,0.90123457,0.83950615,0.962963,0.9506173,0.86419755,0.8888889,0.7407407,0.9506173,0.962963,0.8888889,0.8271605,0.90123457,1.0,1.0,0.83950615,0.962963,0.97530866,0.9876543,0.9506173,1.0,0.9876543,0.91358024,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.012345679,0.09876543,0.12345679,0.24691358,0.2962963,0.39506173,0.27160493,0.24691358,0.2962963,0.33333334,0.27160493,0.34567901,0.44444445,0.13580246,0.34567901,0.44444445,0.41975307,0.4074074,0.50617284,,0.56790125,0.5555556,0.5185185,0.5555556,0.56790125,0.49382716,0.5185185,0.5925926,0.56790125,0.6419753,0.60493827,0.7160494,0.80246913,0.8148148,0.75308645,0.6296296,0.79012346,0.80246913,0.72839504,0.83950615,0.90123457,0.80246913,0.7777778,0.91358024,0.9259259,0.9259259,0.962963,0.962963,0.9382716,0.9506173,0.9382716,0.8518519,0.83950615,0.72839504,1.0,0.97530866,0.83950615,1.0,0.9876543,0.9876543,0.9876543,1.0,0.90123457,0.962963,0.9876543,1.0,1.0,1.0,0.9506173,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.14814815,0.0,0.22222222,0.16049382,0.28395063,0.30864197,0.38271606,0.38271606,0.39506173,0.4691358,0.41975307,0.34567901,0.27160493,0.3580247,0.4074074,0.4814815,0.44444445,0.5555556,0.49382716,0.43209878,,0.45679012,0.5185185,0.54320985,0.49382716,0.54320985,0.654321,0.7037037,0.60493827,0.61728394,0.654321,0.56790125,0.61728394,0.72839504,0.8271605,0.72839504,0.61728394,0.8888889,0.7777778,0.7777778,0.8518519,0.8271605,0.80246913,0.8765432,0.86419755,0.83950615,0.90123457,0.86419755,0.8518519,0.8518519,0.8765432,0.86419755,0.8271605,0.8518519,0.97530866,0.91358024,0.90123457,0.8765432,0.8765432,1.0,1.0,0.8888889,0.97530866,0.9876543,0.97530866,0.9259259,0.9876543,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.037037037,0.11111111,0.12345679,0.25925925,0.3580247,0.39506173,0.20987654,0.39506173,0.2345679,0.45679012,0.27160493,0.13580246,0.41975307,0.37037036,0.41975307,0.4074074,0.32098764,0.5185185,0.43209878,0.44444445,0.54320985,,0.4814815,0.4814815,0.50617284,0.50617284,0.61728394,0.60493827,0.5555556,0.60493827,0.5925926,0.7777778,0.72839504,0.67901236,0.6419753,0.60493827,0.86419755,0.72839504,0.79012346,0.72839504,0.80246913,0.75308645,0.8148148,0.8518519,0.72839504,0.91358024,0.79012346,0.9382716,0.9506173,0.8518519,0.90123457,0.8518519,0.80246913,0.8148148,0.8518519,0.83950615,0.8518519,1.0,0.97530866,0.9876543,1.0,1.0,0.86419755,0.8518519,0.9876543,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0
+RealTabPFN-2.5 (tuned),0.086419754,0.037037037,0.24691358,0.13580246,0.22222222,0.16049382,0.39506173,0.30864197,0.4074074,0.34567901,0.4074074,0.37037036,0.34567901,0.4074074,0.44444445,0.39506173,0.45679012,0.4691358,0.22222222,0.4814815,0.4814815,0.5185185,,0.6296296,0.60493827,0.6296296,0.5308642,0.56790125,0.654321,0.61728394,0.54320985,0.61728394,0.6296296,0.72839504,0.6419753,0.8148148,0.654321,0.7407407,0.6666667,0.72839504,0.7037037,0.79012346,0.80246913,0.8148148,0.8765432,0.80246913,0.7654321,0.83950615,0.8148148,0.90123457,0.9259259,0.86419755,0.8271605,0.72839504,0.90123457,0.9259259,0.8765432,0.80246913,0.8765432,0.9876543,1.0,0.8518519,0.962963,0.9382716,0.97530866,0.9259259,0.9876543,0.9876543,0.8888889,1.0,0.9876543,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.086419754,0.024691358,0.25925925,0.13580246,0.16049382,0.17283951,0.28395063,0.25925925,0.27160493,0.32098764,0.32098764,0.37037036,0.37037036,0.3580247,0.41975307,0.38271606,0.43209878,0.41975307,0.32098764,0.44444445,0.45679012,0.5185185,0.37037036,,0.5555556,0.5802469,0.5802469,0.50617284,0.5802469,0.5802469,0.49382716,0.6419753,0.5925926,0.72839504,0.54320985,0.8518519,0.654321,0.67901236,0.6296296,0.7654321,0.7037037,0.7777778,0.75308645,0.75308645,0.86419755,0.8148148,0.7777778,0.8271605,0.8888889,0.91358024,0.8888889,0.7777778,0.83950615,0.7407407,0.9506173,0.9382716,0.8271605,0.8271605,0.8888889,1.0,1.0,0.8271605,0.97530866,0.9876543,0.9876543,0.91358024,0.97530866,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.024691358,0.037037037,0.11111111,0.11111111,0.19753087,0.24691358,0.2345679,0.24691358,0.2962963,0.24691358,0.28395063,0.37037036,0.32098764,0.33333334,0.13580246,0.4074074,0.43209878,0.33333334,0.39506173,0.43209878,0.50617284,0.49382716,0.39506173,0.44444445,,0.4691358,0.49382716,0.50617284,0.5308642,0.5308642,0.61728394,0.69135803,0.72839504,0.7160494,0.72839504,0.5925926,0.69135803,0.7407407,0.72839504,0.7654321,0.7654321,0.6666667,0.7160494,0.83950615,0.962963,0.9259259,0.9259259,0.8518519,0.80246913,0.8148148,0.8765432,0.7654321,0.8148148,0.7160494,0.9259259,0.91358024,0.79012346,0.8518519,0.90123457,1.0,1.0,0.90123457,0.9382716,0.962963,0.9876543,0.962963,0.9876543,0.9876543,0.90123457,1.0,0.9876543,1.0,1.0,1.0
+TabICL (default),0.11111111,0.049382716,0.25925925,0.14814815,0.24691358,0.18518518,0.2962963,0.28395063,0.34567901,0.3580247,0.33333334,0.38271606,0.3580247,0.44444445,0.37037036,0.39506173,0.45679012,0.4691358,0.33333334,0.50617284,0.45679012,0.49382716,0.37037036,0.41975307,0.5308642,,0.4814815,0.45679012,0.60493827,0.5925926,0.5925926,0.61728394,0.69135803,0.67901236,0.56790125,0.61728394,0.6666667,0.6419753,0.654321,0.654321,0.7037037,0.6419753,0.83950615,0.72839504,0.8518519,0.7407407,0.8518519,0.79012346,0.75308645,0.7777778,0.80246913,0.67901236,0.91358024,0.7654321,0.7777778,0.8148148,0.7654321,0.72839504,0.83950615,0.9259259,0.9382716,0.72839504,0.97530866,1.0,1.0,0.9259259,1.0,1.0,0.9259259,1.0,0.97530866,1.0,1.0,1.0
+iLTM (tuned),0.12345679,0.11111111,0.16049382,0.14814815,0.27160493,0.27160493,0.27160493,0.32098764,0.28395063,0.30864197,0.32098764,0.2962963,0.2962963,0.4074074,0.33333334,0.39506173,0.39506173,0.3580247,0.41975307,0.4814815,0.34567901,0.38271606,0.4691358,0.41975307,0.50617284,0.5185185,,0.4814815,0.50617284,0.44444445,0.5555556,0.67901236,0.56790125,0.5555556,0.5802469,0.56790125,0.67901236,0.61728394,0.7037037,0.61728394,0.6419753,0.5802469,0.7654321,0.654321,0.7777778,0.7407407,0.8271605,0.8148148,0.8271605,0.8148148,0.83950615,0.83950615,0.83950615,0.8518519,0.8765432,0.8518519,0.8888889,0.8888889,0.9876543,0.9382716,0.9382716,0.9259259,0.9259259,0.91358024,0.962963,1.0,0.962963,0.97530866,0.9876543,0.9876543,0.9876543,1.0,1.0,1.0
+ModernNCA (tuned),0.012345679,0.061728396,0.11111111,0.13580246,0.19753087,0.22222222,0.30864197,0.2345679,0.29012346,0.2345679,0.34567901,0.30864197,0.18518518,0.39506173,0.38271606,0.28395063,0.3580247,0.3580247,0.34567901,0.4074074,0.2962963,0.39506173,0.43209878,0.49382716,0.49382716,0.54320985,0.5185185,,0.5308642,0.56790125,0.5925926,0.50617284,0.5802469,0.5925926,0.5802469,0.60493827,0.5802469,0.69135803,0.6296296,0.654321,0.654321,0.69135803,0.75308645,0.69135803,0.7407407,0.7407407,0.8888889,0.80246913,0.7654321,0.8271605,0.79012346,0.91358024,0.8518519,0.80246913,0.83950615,0.83950615,0.90123457,0.8518519,0.9506173,0.9506173,0.9506173,0.8888889,0.9382716,0.9506173,0.962963,0.9876543,0.97530866,0.9876543,0.9506173,1.0,0.97530866,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.086419754,0.037037037,0.22222222,0.12345679,0.16049382,0.11111111,0.34567901,0.28395063,0.37037036,0.32098764,0.37037036,0.33333334,0.27160493,0.34567901,0.43209878,0.34567901,0.38271606,0.38271606,0.34567901,0.43209878,0.39506173,0.44444445,0.34567901,0.41975307,0.4691358,0.39506173,0.49382716,0.4691358,,0.5802469,0.45679012,0.56790125,0.5555556,0.60493827,0.5555556,0.6419753,0.61728394,0.60493827,0.654321,0.5802469,0.5925926,0.72839504,0.8765432,0.6296296,0.75308645,0.7037037,0.7654321,0.7407407,0.7160494,0.75308645,0.72839504,0.7777778,0.9382716,0.7407407,0.75308645,0.79012346,0.8148148,0.72839504,0.8271605,0.83950615,0.83950615,0.72839504,0.9876543,0.97530866,0.9876543,0.8765432,0.9876543,0.9876543,0.91358024,1.0,0.962963,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.012345679,0.037037037,0.09876543,0.18518518,0.17283951,0.20987654,0.2345679,0.14814815,0.2345679,0.16049382,0.24691358,0.32098764,0.33333334,0.22222222,0.2962963,0.2962963,0.39506173,0.24691358,0.32098764,0.3580247,0.38271606,0.39506173,0.38271606,0.41975307,0.4691358,0.4074074,0.5555556,0.43209878,0.41975307,,0.39506173,0.5555556,0.5185185,0.6296296,0.5802469,0.6419753,0.61728394,0.61728394,0.61728394,0.5555556,0.6666667,0.9259259,0.67901236,0.7037037,0.72839504,0.8271605,0.7160494,0.7777778,0.80246913,0.8765432,0.83950615,0.8148148,0.7407407,0.7407407,0.9876543,0.90123457,0.8148148,0.90123457,0.90123457,0.91358024,0.91358024,0.90123457,0.97530866,0.962963,0.9876543,0.9506173,0.9876543,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.09876543,0.037037037,0.20987654,0.049382716,0.16049382,0.13580246,0.34567901,0.32098764,0.37037036,0.25925925,0.38271606,0.32098764,0.20987654,0.34567901,0.30864197,0.32098764,0.41975307,0.45679012,0.39506173,0.39506173,0.34567901,0.4074074,0.45679012,0.50617284,0.38271606,0.4074074,0.44444445,0.4074074,0.54320985,0.60493827,,0.5802469,0.5308642,0.5925926,0.5185185,0.67901236,0.61728394,0.6666667,0.60493827,0.56790125,0.5802469,0.7037037,0.8148148,0.67901236,0.7654321,0.654321,0.6666667,0.7160494,0.6666667,0.7407407,0.7777778,0.7160494,0.90123457,0.8518519,0.7777778,0.80246913,0.86419755,0.7037037,0.83950615,0.8765432,0.8765432,0.72839504,0.962963,0.962963,0.91358024,0.86419755,0.9506173,0.8888889,0.90123457,1.0,0.8765432,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.024691358,0.12345679,0.049382716,0.2345679,0.25925925,0.33333334,0.061728396,0.12345679,0.086419754,0.17283951,0.074074075,0.11111111,0.38271606,0.19753087,0.25925925,0.32098764,0.14814815,0.27160493,0.38271606,0.28395063,0.43209878,0.22222222,0.38271606,0.3580247,0.30864197,0.38271606,0.32098764,0.49382716,0.43209878,0.44444445,0.41975307,,0.60493827,0.4074074,0.50617284,0.41975307,0.6419753,0.5802469,0.7037037,0.5925926,0.67901236,0.6419753,0.5185185,0.69135803,0.5185185,0.60493827,0.6296296,0.7160494,0.7654321,0.8148148,0.72839504,0.7777778,0.654321,0.7654321,0.7407407,0.80246913,0.8518519,0.97530866,0.962963,0.83950615,0.83950615,1.0,0.79012346,0.83950615,0.97530866,1.0,0.97530866,0.962963,1.0,0.8888889,0.97530866,0.962963,0.8888889,1.0
+EBM (tuned + ensembled),0.012345679,0.012345679,0.049382716,0.14814815,0.13580246,0.24691358,0.09876543,0.09876543,0.074074075,0.16049382,0.086419754,0.086419754,0.25925925,0.09876543,0.18518518,0.18518518,0.19753087,0.28395063,0.2962963,0.19753087,0.38271606,0.27160493,0.37037036,0.4074074,0.27160493,0.30864197,0.43209878,0.41975307,0.44444445,0.4814815,0.4691358,0.39506173,,0.56790125,0.5555556,0.5308642,0.54320985,0.69135803,0.6296296,0.654321,0.75308645,0.7654321,0.6666667,0.91358024,0.6419753,0.69135803,0.69135803,0.7407407,0.86419755,0.7777778,0.7777778,0.72839504,0.6419753,0.6666667,0.8518519,0.8271605,0.7777778,0.8888889,0.90123457,0.97530866,0.962963,0.91358024,0.86419755,0.8518519,0.8765432,0.9382716,0.86419755,0.86419755,0.80246913,1.0,0.97530866,1.0,0.8888889,1.0
+XGBoost (default),0.037037037,0.024691358,0.086419754,0.09876543,0.09876543,0.14814815,0.12345679,0.12345679,0.14814815,0.17283951,0.16049382,0.24691358,0.25925925,0.061728396,0.19753087,0.27160493,0.27160493,0.24691358,0.25925925,0.18518518,0.27160493,0.32098764,0.27160493,0.27160493,0.28395063,0.32098764,0.44444445,0.4074074,0.39506173,0.37037036,0.4074074,0.5925926,0.43209878,,0.4814815,0.49382716,0.6419753,0.49382716,0.654321,0.54320985,0.54320985,0.6296296,0.61728394,0.60493827,0.4814815,0.79012346,0.69135803,0.83950615,0.8148148,0.8518519,0.8518519,0.67901236,0.61728394,0.67901236,0.91358024,0.8888889,0.7407407,0.86419755,0.8765432,0.9382716,0.9506173,0.8765432,0.86419755,0.9259259,0.91358024,0.8888889,0.91358024,0.9876543,0.8888889,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.086419754,0.012345679,0.086419754,0.061728396,0.2345679,0.18518518,0.25925925,0.11111111,0.22222222,0.19753087,0.037037037,0.20987654,0.22222222,0.19753087,0.2345679,0.25925925,0.28395063,0.24691358,0.17283951,0.3580247,0.3580247,0.45679012,0.27160493,0.43209878,0.41975307,0.41975307,0.44444445,0.41975307,0.4814815,0.49382716,0.44444445,0.5185185,,0.50617284,0.5308642,0.56790125,0.61728394,0.5555556,0.5925926,0.56790125,0.67901236,0.6419753,0.69135803,0.67901236,0.80246913,0.7407407,0.69135803,0.7160494,0.7407407,0.75308645,0.69135803,0.6419753,0.80246913,0.75308645,0.79012346,0.7777778,0.8271605,0.8888889,0.8888889,0.7654321,0.97530866,0.97530866,0.9382716,0.8888889,0.91358024,0.9876543,0.8888889,1.0,0.9506173,1.0,1.0,1.0
+TabSwift (default),0.061728396,0.049382716,0.17283951,0.037037037,0.14814815,0.09876543,0.20987654,0.25925925,0.24691358,0.27160493,0.25925925,0.37037036,0.16049382,0.34567901,0.28395063,0.33333334,0.4074074,0.39506173,0.16049382,0.37037036,0.27160493,0.39506173,0.18518518,0.14814815,0.4074074,0.38271606,0.43209878,0.39506173,0.3580247,0.3580247,0.32098764,0.5802469,0.4691358,0.50617284,0.49382716,,0.5308642,0.41975307,0.5802469,0.5185185,0.5308642,0.5185185,0.5802469,0.5555556,0.5802469,0.67901236,0.61728394,0.5925926,0.5925926,0.69135803,0.61728394,0.6419753,0.83950615,0.75308645,0.654321,0.69135803,0.6419753,0.69135803,0.7654321,0.7777778,0.7777778,0.72839504,0.9506173,0.97530866,0.91358024,0.79012346,0.90123457,1.0,0.8888889,0.91358024,1.0,0.962963,1.0,1.0
+xRFM (tuned),0.012345679,0.09876543,0.037037037,0.22222222,0.20987654,0.30864197,0.049382716,0.049382716,0.061728396,0.12345679,0.074074075,0.086419754,0.38271606,0.09876543,0.18518518,0.24691358,0.11111111,0.22222222,0.32098764,0.20987654,0.38271606,0.13580246,0.34567901,0.34567901,0.30864197,0.33333334,0.32098764,0.41975307,0.38271606,0.38271606,0.38271606,0.3580247,0.45679012,0.3580247,0.4691358,0.4691358,,0.5555556,0.6296296,0.5925926,0.61728394,0.60493827,0.56790125,0.5802469,0.50617284,0.5308642,0.6296296,0.69135803,0.8888889,0.72839504,0.72839504,0.80246913,0.6296296,0.69135803,0.75308645,0.8271605,0.8518519,0.9506173,0.97530866,0.8148148,0.8271605,0.9876543,0.83950615,0.8148148,0.9382716,1.0,0.8888889,0.86419755,0.9876543,0.9876543,0.9506173,1.0,0.8888889,1.0
+RealMLP (default),0.0,0.0,0.037037037,0.061728396,0.0,0.09876543,0.19753087,0.049382716,0.18518518,0.0,0.17283951,0.16049382,0.024691358,0.17283951,0.19753087,0.12345679,0.17283951,0.11111111,0.19753087,0.19753087,0.11111111,0.27160493,0.25925925,0.32098764,0.25925925,0.3580247,0.38271606,0.30864197,0.39506173,0.38271606,0.33333334,0.41975307,0.30864197,0.50617284,0.43209878,0.5802469,0.44444445,,0.44444445,0.50617284,0.44444445,0.6419753,0.6296296,0.4814815,0.6419753,0.67901236,0.7037037,0.6666667,0.69135803,0.7654321,0.7407407,0.80246913,0.7037037,0.67901236,0.8765432,0.79012346,0.8271605,0.8148148,0.8518519,0.962963,0.962963,0.86419755,0.9506173,0.9382716,0.9506173,0.90123457,0.9506173,0.9506173,0.86419755,1.0,0.9876543,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.09876543,0.037037037,0.11111111,0.14814815,0.17283951,0.18518518,0.037037037,0.14814815,0.024691358,0.19753087,0.16049382,0.19753087,0.22222222,0.25925925,0.14814815,0.20987654,0.13580246,0.2962963,0.27160493,0.22222222,0.20987654,0.33333334,0.37037036,0.27160493,0.34567901,0.2962963,0.37037036,0.34567901,0.38271606,0.39506173,0.2962963,0.37037036,0.34567901,0.38271606,0.41975307,0.37037036,0.5555556,,0.44444445,0.49382716,0.54320985,0.5802469,0.5308642,0.5802469,0.49382716,0.69135803,0.6419753,0.67901236,0.654321,0.6666667,0.9259259,0.69135803,0.61728394,0.75308645,0.69135803,0.962963,0.8765432,0.86419755,0.75308645,0.75308645,0.8888889,0.86419755,0.8518519,0.9259259,0.97530866,0.9259259,0.8148148,1.0,0.8888889,0.9259259,0.8888889,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.09876543,0.061728396,0.11111111,0.086419754,0.16049382,0.14814815,0.17283951,0.16049382,0.16049382,0.13580246,0.09876543,0.13580246,0.14814815,0.19753087,0.16049382,0.24691358,0.22222222,0.16049382,0.22222222,0.27160493,0.27160493,0.2345679,0.2345679,0.34567901,0.38271606,0.34567901,0.41975307,0.44444445,0.43209878,0.4074074,0.34567901,0.45679012,0.44444445,0.4814815,0.4074074,0.49382716,0.5555556,,0.41975307,0.5555556,0.67901236,0.5802469,0.654321,0.60493827,0.6666667,0.61728394,0.7160494,0.67901236,0.654321,0.7037037,0.6419753,0.72839504,0.7407407,0.7037037,0.7407407,0.72839504,0.90123457,0.97530866,0.9876543,0.7407407,0.8518519,0.8518519,0.83950615,0.9382716,0.8518519,0.8765432,0.79012346,1.0,0.962963,1.0,0.8888889,1.0
+EBM (tuned),0.0,0.0,0.049382716,0.09876543,0.037037037,0.09876543,0.09876543,0.086419754,0.09876543,0.086419754,0.09876543,0.049382716,0.061728396,0.061728396,0.09876543,0.16049382,0.13580246,0.20987654,0.19753087,0.09876543,0.14814815,0.19753087,0.2962963,0.2962963,0.2345679,0.2962963,0.3580247,0.34567901,0.4074074,0.33333334,0.41975307,0.32098764,0.24691358,0.45679012,0.4074074,0.4691358,0.38271606,0.5555556,0.50617284,0.5802469,,0.5802469,0.54320985,0.7160494,0.54320985,0.5925926,0.6419753,0.6666667,0.80246913,0.654321,0.69135803,0.7037037,0.5555556,0.654321,0.8271605,0.72839504,0.7160494,0.83950615,0.8765432,0.9382716,0.9382716,0.8765432,0.8518519,0.83950615,0.8148148,0.91358024,0.8271605,0.86419755,0.79012346,1.0,0.9876543,1.0,0.8888889,1.0
+TorchMLP (tuned),0.024691358,0.0,0.024691358,0.037037037,0.049382716,0.09876543,0.049382716,0.049382716,0.049382716,0.074074075,0.061728396,0.14814815,0.13580246,0.17283951,0.09876543,0.17283951,0.19753087,0.09876543,0.19753087,0.19753087,0.17283951,0.24691358,0.20987654,0.22222222,0.33333334,0.3580247,0.41975307,0.30864197,0.27160493,0.074074075,0.2962963,0.3580247,0.2345679,0.37037036,0.43209878,0.4814815,0.39506173,0.3580247,0.45679012,0.44444445,0.41975307,,0.43209878,0.44444445,0.5308642,0.7037037,0.61728394,0.60493827,0.61728394,0.67901236,0.69135803,0.60493827,0.5802469,0.5802469,0.8765432,0.75308645,0.654321,0.86419755,0.8271605,0.8271605,0.8271605,0.86419755,0.9506173,0.9506173,0.8888889,0.91358024,0.91358024,0.97530866,0.8765432,0.97530866,0.9876543,1.0,1.0,1.0
+TabDPT (tuned),0.049382716,0.037037037,0.037037037,0.037037037,0.049382716,0.061728396,0.20987654,0.09876543,0.20987654,0.09876543,0.22222222,0.18518518,0.11111111,0.17283951,0.22222222,0.13580246,0.20987654,0.18518518,0.12345679,0.22222222,0.19753087,0.18518518,0.19753087,0.24691358,0.28395063,0.16049382,0.2345679,0.24691358,0.12345679,0.32098764,0.18518518,0.4814815,0.33333334,0.38271606,0.32098764,0.41975307,0.43209878,0.37037036,0.41975307,0.32098764,0.45679012,0.56790125,,0.4691358,0.5185185,0.49382716,0.5185185,0.54320985,0.49382716,0.6419753,0.654321,0.6419753,0.7592593,0.69135803,0.6419753,0.7160494,0.75308645,0.654321,0.7777778,0.6296296,0.6419753,0.7160494,0.962963,0.962963,0.9259259,0.7777778,0.9382716,0.90123457,0.8888889,0.90123457,0.8888889,0.97530866,1.0,1.0
+EBM (default),0.0,0.0,0.049382716,0.049382716,0.037037037,0.074074075,0.074074075,0.061728396,0.061728396,0.061728396,0.037037037,0.024691358,0.074074075,0.024691358,0.037037037,0.14814815,0.09876543,0.18518518,0.14814815,0.086419754,0.12345679,0.14814815,0.18518518,0.24691358,0.16049382,0.27160493,0.34567901,0.30864197,0.37037036,0.2962963,0.32098764,0.30864197,0.086419754,0.39506173,0.3580247,0.44444445,0.41975307,0.5185185,0.4691358,0.41975307,0.28395063,0.5555556,0.5308642,,0.50617284,0.61728394,0.60493827,0.61728394,0.6419753,0.6419753,0.67901236,0.7037037,0.5555556,0.654321,0.8148148,0.72839504,0.7037037,0.79012346,0.86419755,0.8271605,0.8271605,0.8271605,0.8518519,0.83950615,0.80246913,0.90123457,0.80246913,0.8518519,0.7777778,1.0,0.97530866,1.0,0.8888889,1.0
+LightGBM (default),0.0,0.0,0.049382716,0.012345679,0.061728396,0.061728396,0.074074075,0.086419754,0.049382716,0.049382716,0.086419754,0.13580246,0.0,0.037037037,0.024691358,0.09876543,0.20987654,0.16049382,0.086419754,0.074074075,0.13580246,0.27160493,0.12345679,0.13580246,0.037037037,0.14814815,0.22222222,0.25925925,0.24691358,0.27160493,0.2345679,0.4814815,0.3580247,0.5185185,0.30864197,0.41975307,0.49382716,0.3580247,0.41975307,0.34567901,0.45679012,0.4691358,0.4814815,0.49382716,,0.56790125,0.5555556,0.61728394,0.54320985,0.61728394,0.6419753,0.5555556,0.6296296,0.6666667,0.7160494,0.7160494,0.6666667,0.654321,0.7407407,0.8271605,0.83950615,0.6666667,0.86419755,0.9506173,0.9382716,0.79012346,0.962963,0.9382716,0.8888889,1.0,0.91358024,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.037037037,0.012345679,0.061728396,0.086419754,0.061728396,0.0,0.086419754,0.09876543,0.074074075,0.049382716,0.074074075,0.037037037,0.024691358,0.17283951,0.11111111,0.18518518,0.16049382,0.074074075,0.16049382,0.086419754,0.19753087,0.18518518,0.074074075,0.25925925,0.25925925,0.25925925,0.2962963,0.17283951,0.34567901,0.39506173,0.30864197,0.20987654,0.32098764,0.32098764,0.4691358,0.32098764,0.50617284,0.39506173,0.4074074,0.2962963,0.50617284,0.38271606,0.43209878,,0.5185185,0.54320985,0.6419753,0.5308642,0.5555556,0.69135803,0.5308642,0.6296296,0.60493827,0.60493827,0.7407407,0.72839504,0.86419755,0.8765432,0.8888889,0.7777778,0.72839504,0.7160494,0.8271605,0.8765432,0.83950615,0.91358024,0.8765432,0.9876543,0.962963,0.8888889,0.7777778,1.0
+ModernNCA (default),0.0,0.0,0.012345679,0.024691358,0.024691358,0.049382716,0.012345679,0.024691358,0.049382716,0.049382716,0.074074075,0.09876543,0.037037037,0.012345679,0.024691358,0.049382716,0.16049382,0.16049382,0.19753087,0.037037037,0.09876543,0.20987654,0.2345679,0.22222222,0.074074075,0.14814815,0.17283951,0.11111111,0.2345679,0.28395063,0.33333334,0.37037036,0.30864197,0.30864197,0.19753087,0.38271606,0.37037036,0.2962963,0.30864197,0.33333334,0.3580247,0.38271606,0.4814815,0.39506173,0.44444445,0.4814815,,0.5802469,0.5308642,0.5185185,0.654321,0.49382716,0.5308642,0.5925926,0.72839504,0.60493827,0.6419753,0.6666667,0.8148148,0.7160494,0.72839504,0.7160494,0.8271605,0.8518519,0.9259259,0.8518519,0.9506173,0.90123457,0.8888889,1.0,0.91358024,1.0,0.91358024,1.0
+RandomForest (tuned + ensembled),0.0,0.024691358,0.012345679,0.061728396,0.049382716,0.037037037,0.024691358,0.024691358,0.024691358,0.024691358,0.024691358,0.012345679,0.074074075,0.0,0.037037037,0.12345679,0.037037037,0.061728396,0.09876543,0.037037037,0.13580246,0.061728396,0.16049382,0.17283951,0.14814815,0.20987654,0.18518518,0.19753087,0.25925925,0.22222222,0.28395063,0.28395063,0.25925925,0.16049382,0.25925925,0.4074074,0.30864197,0.33333334,0.3580247,0.38271606,0.33333334,0.39506173,0.45679012,0.38271606,0.38271606,0.45679012,0.41975307,,0.56790125,0.5925926,0.72839504,0.5925926,0.50617284,0.5925926,0.6666667,0.654321,0.6666667,0.7777778,0.8765432,0.7777778,0.79012346,0.83950615,0.83950615,0.8148148,0.962963,0.9506173,0.962963,0.9259259,0.8765432,0.9259259,0.962963,1.0,0.8888889,1.0
+SAP-RPT-OSS (default),0.0,0.024691358,0.037037037,0.086419754,0.049382716,0.086419754,0.049382716,0.024691358,0.024691358,0.037037037,0.037037037,0.024691358,0.09876543,0.049382716,0.049382716,0.16049382,0.074074075,0.12345679,0.16049382,0.061728396,0.14814815,0.049382716,0.18518518,0.11111111,0.19753087,0.24691358,0.17283951,0.2345679,0.28395063,0.19753087,0.33333334,0.2345679,0.13580246,0.18518518,0.30864197,0.4074074,0.11111111,0.30864197,0.32098764,0.28395063,0.19753087,0.38271606,0.50617284,0.3580247,0.45679012,0.3580247,0.4691358,0.43209878,,0.56790125,0.4691358,0.56790125,0.5802469,0.654321,0.60493827,0.75308645,0.67901236,0.79012346,0.8765432,0.6296296,0.654321,0.8271605,0.8271605,0.8518519,0.83950615,0.97530866,0.83950615,0.8518519,0.8148148,0.9506173,0.962963,1.0,0.8888889,1.0
+ExtraTrees (tuned + ensembled),0.0,0.012345679,0.012345679,0.037037037,0.024691358,0.012345679,0.037037037,0.049382716,0.012345679,0.0,0.049382716,0.12345679,0.074074075,0.012345679,0.024691358,0.11111111,0.14814815,0.11111111,0.037037037,0.049382716,0.14814815,0.14814815,0.09876543,0.086419754,0.18518518,0.22222222,0.18518518,0.17283951,0.24691358,0.12345679,0.25925925,0.18518518,0.22222222,0.14814815,0.28395063,0.30864197,0.27160493,0.2345679,0.34567901,0.32098764,0.34567901,0.32098764,0.3580247,0.3580247,0.38271606,0.4691358,0.4814815,0.4074074,0.43209878,,0.50617284,0.4814815,0.4814815,0.5308642,0.60493827,0.72839504,0.60493827,0.69135803,0.7160494,0.69135803,0.69135803,0.7654321,0.72839504,0.7777778,0.97530866,0.8888889,0.9876543,0.9382716,0.90123457,0.83950615,0.962963,0.8888889,1.0,1.0
+RandomForest (tuned),0.012345679,0.024691358,0.024691358,0.037037037,0.012345679,0.037037037,0.037037037,0.012345679,0.037037037,0.037037037,0.037037037,0.012345679,0.074074075,0.012345679,0.049382716,0.13580246,0.037037037,0.086419754,0.049382716,0.061728396,0.12345679,0.09876543,0.074074075,0.11111111,0.12345679,0.19753087,0.16049382,0.20987654,0.27160493,0.16049382,0.22222222,0.27160493,0.22222222,0.14814815,0.25925925,0.38271606,0.27160493,0.25925925,0.33333334,0.34567901,0.30864197,0.30864197,0.34567901,0.32098764,0.3580247,0.44444445,0.34567901,0.27160493,0.5308642,0.49382716,,0.5185185,0.45679012,0.5802469,0.5308642,0.5925926,0.6296296,0.67901236,0.83950615,0.7654321,0.7777778,0.75308645,0.8148148,0.8148148,0.91358024,0.90123457,0.93209875,0.91358024,0.83950615,0.9259259,0.97530866,1.0,0.8888889,1.0
+FastaiMLP (tuned),0.0,0.024691358,0.024691358,0.074074075,0.061728396,0.074074075,0.024691358,0.037037037,0.037037037,0.024691358,0.049382716,0.12345679,0.09876543,0.13580246,0.09876543,0.074074075,0.16049382,0.037037037,0.13580246,0.14814815,0.13580246,0.14814815,0.13580246,0.22222222,0.2345679,0.32098764,0.16049382,0.086419754,0.22222222,0.18518518,0.28395063,0.22222222,0.27160493,0.32098764,0.24691358,0.3580247,0.19753087,0.19753087,0.074074075,0.2962963,0.2962963,0.39506173,0.3580247,0.2962963,0.44444445,0.30864197,0.50617284,0.4074074,0.43209878,0.5185185,0.4814815,,0.54320985,0.4691358,0.54320985,0.56790125,0.67901236,0.6296296,0.67901236,0.61728394,0.61728394,0.7037037,0.80246913,0.8148148,0.80246913,0.8888889,0.8518519,0.75308645,0.8765432,0.8765432,0.8518519,0.8765432,0.9506173,1.0
+TabDPT (default),0.024691358,0.024691358,0.049382716,0.037037037,0.061728396,0.049382716,0.11111111,0.074074075,0.12345679,0.09876543,0.13580246,0.18518518,0.12345679,0.12345679,0.049382716,0.086419754,0.24691358,0.17283951,0.11111111,0.16049382,0.17283951,0.19753087,0.17283951,0.16049382,0.18518518,0.086419754,0.16049382,0.14814815,0.061728396,0.25925925,0.09876543,0.34567901,0.3580247,0.38271606,0.30864197,0.16049382,0.37037036,0.2962963,0.30864197,0.3580247,0.44444445,0.41975307,0.24074075,0.44444445,0.37037036,0.4691358,0.4691358,0.49382716,0.41975307,0.5185185,0.54320985,0.45679012,,0.50617284,0.5802469,0.5555556,0.5925926,0.6666667,0.56790125,0.6296296,0.6296296,0.67901236,0.7654321,0.8148148,0.6419753,0.7160494,0.6666667,0.69135803,0.72839504,0.67901236,0.83950615,0.7037037,0.97530866,1.0
+LimiX (default),0.012345679,0.049382716,0.024691358,0.11111111,0.13580246,0.09876543,0.16049382,0.16049382,0.19753087,0.061728396,0.19753087,0.17283951,0.14814815,0.24691358,0.2345679,0.049382716,0.18518518,0.11111111,0.25925925,0.27160493,0.14814815,0.18518518,0.27160493,0.25925925,0.28395063,0.2345679,0.14814815,0.19753087,0.25925925,0.25925925,0.14814815,0.2345679,0.33333334,0.32098764,0.3580247,0.24691358,0.30864197,0.32098764,0.38271606,0.27160493,0.34567901,0.41975307,0.30864197,0.34567901,0.33333334,0.37037036,0.4074074,0.4074074,0.34567901,0.4691358,0.41975307,0.5308642,0.49382716,,0.43209878,0.50617284,0.56790125,0.60493827,0.5185185,0.4691358,0.4814815,0.61728394,0.5925926,0.6419753,0.67901236,0.61728394,0.72839504,0.5925926,0.75308645,0.5802469,0.61728394,0.6419753,0.7654321,0.8765432
+TorchMLP (default),0.0,0.0,0.012345679,0.037037037,0.0,0.024691358,0.024691358,0.0,0.012345679,0.012345679,0.024691358,0.09876543,0.024691358,0.0,0.024691358,0.12345679,0.09876543,0.074074075,0.049382716,0.0,0.024691358,0.14814815,0.09876543,0.049382716,0.074074075,0.22222222,0.12345679,0.16049382,0.24691358,0.012345679,0.22222222,0.25925925,0.14814815,0.086419754,0.19753087,0.34567901,0.24691358,0.12345679,0.24691358,0.25925925,0.17283951,0.12345679,0.3580247,0.18518518,0.28395063,0.39506173,0.27160493,0.33333334,0.39506173,0.39506173,0.4691358,0.45679012,0.41975307,0.56790125,,0.49382716,0.50617284,0.7037037,0.7777778,0.7037037,0.7037037,0.7407407,0.7777778,0.80246913,0.8518519,0.86419755,0.8765432,0.9259259,0.8765432,0.8888889,0.9506173,0.9876543,0.9506173,1.0
+ExtraTrees (tuned),0.0,0.012345679,0.0,0.037037037,0.012345679,0.012345679,0.012345679,0.012345679,0.024691358,0.024691358,0.037037037,0.12345679,0.049382716,0.012345679,0.024691358,0.12345679,0.14814815,0.049382716,0.037037037,0.024691358,0.086419754,0.16049382,0.074074075,0.061728396,0.086419754,0.18518518,0.14814815,0.16049382,0.20987654,0.09876543,0.19753087,0.19753087,0.17283951,0.11111111,0.24691358,0.30864197,0.17283951,0.20987654,0.30864197,0.2962963,0.27160493,0.24691358,0.28395063,0.27160493,0.28395063,0.39506173,0.39506173,0.34567901,0.24691358,0.27160493,0.4074074,0.43209878,0.44444445,0.49382716,0.50617284,,0.49382716,0.67901236,0.654321,0.67901236,0.6666667,0.7160494,0.67901236,0.75308645,0.9382716,0.8765432,0.9506173,0.9259259,0.8888889,0.8271605,0.962963,0.90123457,0.9876543,1.0
+FastaiMLP (default),0.0,0.024691358,0.012345679,0.074074075,0.086419754,0.086419754,0.049382716,0.0,0.024691358,0.037037037,0.061728396,0.13580246,0.11111111,0.11111111,0.14814815,0.037037037,0.14814815,0.049382716,0.11111111,0.16049382,0.09876543,0.14814815,0.12345679,0.17283951,0.20987654,0.2345679,0.11111111,0.09876543,0.18518518,0.18518518,0.13580246,0.14814815,0.22222222,0.25925925,0.20987654,0.3580247,0.14814815,0.17283951,0.037037037,0.25925925,0.28395063,0.34567901,0.24691358,0.2962963,0.33333334,0.25925925,0.3580247,0.33333334,0.32098764,0.39506173,0.37037036,0.32098764,0.4074074,0.43209878,0.49382716,0.50617284,,0.54320985,0.5555556,0.5308642,0.54320985,0.5925926,0.7654321,0.7777778,0.67901236,0.6419753,0.67901236,0.7037037,0.79012346,0.7777778,0.7777778,0.79012346,0.8765432,1.0
+Linear (tuned + ensembled),0.0,0.024691358,0.0,0.086419754,0.037037037,0.086419754,0.012345679,0.012345679,0.012345679,0.0,0.012345679,0.0,0.074074075,0.0,0.024691358,0.11111111,0.0,0.024691358,0.17283951,0.0,0.12345679,0.0,0.19753087,0.17283951,0.14814815,0.27160493,0.11111111,0.14814815,0.27160493,0.09876543,0.2962963,0.024691358,0.11111111,0.13580246,0.22222222,0.30864197,0.049382716,0.18518518,0.12345679,0.27160493,0.16049382,0.13580246,0.34567901,0.20987654,0.34567901,0.27160493,0.33333334,0.22222222,0.20987654,0.30864197,0.32098764,0.37037036,0.33333334,0.39506173,0.2962963,0.32098764,0.45679012,,0.3580247,0.4814815,0.4814815,0.7962963,0.4814815,0.50617284,0.60493827,0.9259259,0.61728394,0.654321,0.7654321,0.5555556,0.654321,0.5802469,0.7407407,0.8888889
+KNN (tuned + ensembled),0.012345679,0.049382716,0.012345679,0.074074075,0.061728396,0.049382716,0.012345679,0.012345679,0.012345679,0.012345679,0.012345679,0.012345679,0.061728396,0.012345679,0.024691358,0.012345679,0.024691358,0.037037037,0.09876543,0.012345679,0.12345679,0.024691358,0.12345679,0.11111111,0.09876543,0.16049382,0.012345679,0.049382716,0.17283951,0.09876543,0.16049382,0.037037037,0.09876543,0.12345679,0.17283951,0.2345679,0.024691358,0.14814815,0.13580246,0.09876543,0.12345679,0.17283951,0.22222222,0.13580246,0.25925925,0.13580246,0.18518518,0.12345679,0.12345679,0.28395063,0.16049382,0.32098764,0.43209878,0.4814815,0.22222222,0.34567901,0.44444445,0.6419753,,0.38271606,0.38271606,0.7037037,0.5925926,0.6419753,0.6296296,0.80246913,0.7654321,0.6419753,0.7654321,0.6666667,0.7160494,1.0,0.8888889,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.024691358,0.0,0.0,0.0,0.0,0.012345679,0.0,0.0,0.012345679,0.012345679,0.086419754,0.012345679,0.024691358,0.0,0.012345679,0.0,0.012345679,0.012345679,0.0,0.0,0.074074075,0.061728396,0.049382716,0.16049382,0.086419754,0.12345679,0.16049382,0.024691358,0.061728396,0.11111111,0.22222222,0.18518518,0.037037037,0.24691358,0.024691358,0.061728396,0.17283951,0.37037036,0.17283951,0.17283951,0.12345679,0.28395063,0.22222222,0.37037036,0.30864197,0.2345679,0.38271606,0.37037036,0.5308642,0.2962963,0.32098764,0.4691358,0.5185185,0.61728394,,0.66049385,0.5308642,0.7037037,0.7037037,0.6666667,0.7160494,0.6666667,0.654321,0.6666667,1.0,0.86419755,0.86419755,0.7777778,0.8888889
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.024691358,0.0,0.0,0.0,0.0,0.012345679,0.0,0.0,0.012345679,0.012345679,0.086419754,0.012345679,0.024691358,0.0,0.012345679,0.0,0.0,0.0,0.0,0.0,0.061728396,0.061728396,0.049382716,0.16049382,0.086419754,0.12345679,0.16049382,0.037037037,0.049382716,0.11111111,0.22222222,0.17283951,0.037037037,0.24691358,0.012345679,0.061728396,0.17283951,0.3580247,0.17283951,0.16049382,0.11111111,0.27160493,0.20987654,0.34567901,0.30864197,0.22222222,0.38271606,0.37037036,0.5185185,0.2962963,0.33333334,0.45679012,0.5185185,0.61728394,0.33950618,,0.5308642,0.7037037,0.7037037,0.6666667,0.7160494,0.6666667,0.654321,0.6666667,1.0,0.86419755,0.86419755,0.7777778,0.8888889
+Linear (tuned),0.0,0.024691358,0.0,0.074074075,0.037037037,0.061728396,0.0,0.0,0.0,0.0,0.0,0.0,0.061728396,0.0,0.012345679,0.09876543,0.0,0.012345679,0.16049382,0.0,0.11111111,0.0,0.14814815,0.17283951,0.09876543,0.27160493,0.074074075,0.11111111,0.27160493,0.09876543,0.27160493,0.0,0.086419754,0.12345679,0.2345679,0.27160493,0.012345679,0.13580246,0.11111111,0.25925925,0.12345679,0.13580246,0.28395063,0.17283951,0.33333334,0.22222222,0.28395063,0.16049382,0.17283951,0.2345679,0.24691358,0.2962963,0.32098764,0.38271606,0.25925925,0.28395063,0.4074074,0.2037037,0.2962963,0.4691358,0.4691358,,0.4814815,0.49382716,0.56790125,0.86419755,0.56790125,0.6296296,0.72839504,0.5555556,0.61728394,0.5802469,0.7407407,0.8888889
+PerpetualBooster (tuned + ensembled),0.012345679,0.0,0.024691358,0.0,0.012345679,0.0,0.012345679,0.024691358,0.024691358,0.024691358,0.024691358,0.13580246,0.024691358,0.086419754,0.024691358,0.012345679,0.13580246,0.037037037,0.037037037,0.09876543,0.024691358,0.13580246,0.037037037,0.024691358,0.061728396,0.024691358,0.074074075,0.061728396,0.012345679,0.024691358,0.037037037,0.20987654,0.13580246,0.13580246,0.024691358,0.049382716,0.16049382,0.049382716,0.13580246,0.14814815,0.14814815,0.049382716,0.037037037,0.14814815,0.13580246,0.27160493,0.17283951,0.16049382,0.17283951,0.27160493,0.18518518,0.19753087,0.2345679,0.4074074,0.22222222,0.32098764,0.2345679,0.5185185,0.4074074,0.2962963,0.2962963,0.5185185,,0.83950615,0.5802469,0.5185185,0.5925926,0.6666667,0.6296296,0.38271606,0.56790125,0.61728394,0.91358024,0.9506173
+PerpetualBooster (tuned),0.0,0.0,0.037037037,0.0,0.012345679,0.0,0.012345679,0.037037037,0.012345679,0.012345679,0.012345679,0.12345679,0.0,0.037037037,0.024691358,0.012345679,0.13580246,0.049382716,0.024691358,0.037037037,0.012345679,0.14814815,0.061728396,0.012345679,0.037037037,0.0,0.086419754,0.049382716,0.024691358,0.037037037,0.037037037,0.16049382,0.14814815,0.074074075,0.024691358,0.024691358,0.18518518,0.061728396,0.14814815,0.14814815,0.16049382,0.049382716,0.037037037,0.16049382,0.049382716,0.28395063,0.14814815,0.18518518,0.14814815,0.22222222,0.18518518,0.18518518,0.18518518,0.3580247,0.19753087,0.24691358,0.22222222,0.49382716,0.3580247,0.2962963,0.2962963,0.50617284,0.16049382,,0.5555556,0.49382716,0.5925926,0.6666667,0.6296296,0.34567901,0.50617284,0.5308642,0.8888889,0.9506173
+ExtraTrees (default),0.0,0.012345679,0.012345679,0.0,0.0,0.0,0.024691358,0.012345679,0.012345679,0.0,0.024691358,0.012345679,0.0,0.012345679,0.012345679,0.024691358,0.012345679,0.012345679,0.012345679,0.012345679,0.024691358,0.012345679,0.024691358,0.012345679,0.012345679,0.0,0.037037037,0.037037037,0.012345679,0.012345679,0.086419754,0.024691358,0.12345679,0.086419754,0.061728396,0.086419754,0.061728396,0.049382716,0.074074075,0.16049382,0.18518518,0.11111111,0.074074075,0.19753087,0.061728396,0.17283951,0.074074075,0.037037037,0.16049382,0.024691358,0.086419754,0.19753087,0.3580247,0.32098764,0.14814815,0.061728396,0.32098764,0.39506173,0.37037036,0.33333334,0.33333334,0.43209878,0.41975307,0.44444445,,0.4691358,0.44444445,0.43209878,0.67901236,0.50617284,0.54320985,0.61728394,0.7654321,1.0
+Linear (default),0.0,0.0,0.0,0.049382716,0.037037037,0.037037037,0.0,0.0,0.0,0.0,0.0,0.0,0.024691358,0.0,0.0,0.024691358,0.0,0.0,0.049382716,0.0,0.074074075,0.0,0.074074075,0.086419754,0.037037037,0.074074075,0.0,0.012345679,0.12345679,0.049382716,0.13580246,0.0,0.061728396,0.11111111,0.11111111,0.20987654,0.0,0.09876543,0.024691358,0.061728396,0.086419754,0.086419754,0.22222222,0.09876543,0.20987654,0.12345679,0.14814815,0.049382716,0.024691358,0.11111111,0.09876543,0.11111111,0.28395063,0.38271606,0.13580246,0.12345679,0.3580247,0.074074075,0.19753087,0.28395063,0.28395063,0.13580246,0.4814815,0.50617284,0.5308642,,0.54320985,0.5555556,0.7160494,0.5308642,0.4814815,0.5802469,0.7407407,0.8765432
+RandomForest (default),0.0,0.012345679,0.0,0.0,0.0,0.0,0.012345679,0.012345679,0.012345679,0.0,0.012345679,0.0,0.0,0.0,0.0,0.0,0.012345679,0.012345679,0.0,0.0,0.012345679,0.0,0.012345679,0.024691358,0.012345679,0.0,0.037037037,0.024691358,0.012345679,0.012345679,0.049382716,0.024691358,0.13580246,0.086419754,0.086419754,0.09876543,0.11111111,0.049382716,0.074074075,0.14814815,0.17283951,0.086419754,0.061728396,0.19753087,0.037037037,0.16049382,0.049382716,0.037037037,0.16049382,0.012345679,0.06790123,0.14814815,0.33333334,0.27160493,0.12345679,0.049382716,0.32098764,0.38271606,0.2345679,0.33333334,0.33333334,0.43209878,0.4074074,0.4074074,0.5555556,0.45679012,,0.43209878,0.79012346,0.45679012,0.54320985,0.61728394,0.7654321,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.012345679,0.012345679,0.012345679,0.0,0.024691358,0.0,0.0,0.0,0.012345679,0.049382716,0.012345679,0.024691358,0.012345679,0.0,0.0,0.0,0.012345679,0.0,0.012345679,0.0,0.024691358,0.012345679,0.012345679,0.0,0.11111111,0.037037037,0.13580246,0.012345679,0.012345679,0.0,0.13580246,0.049382716,0.18518518,0.12345679,0.13580246,0.024691358,0.09876543,0.14814815,0.061728396,0.086419754,0.09876543,0.074074075,0.14814815,0.061728396,0.086419754,0.24691358,0.30864197,0.4074074,0.074074075,0.074074075,0.2962963,0.34567901,0.3580247,0.34567901,0.34567901,0.37037036,0.33333334,0.33333334,0.56790125,0.44444445,0.56790125,,0.5925926,0.49382716,0.5925926,0.43209878,0.60493827,0.9382716
+xRFM (default),0.0,0.049382716,0.0,0.086419754,0.086419754,0.086419754,0.0,0.0,0.0,0.0,0.0,0.0,0.09876543,0.012345679,0.049382716,0.0,0.0,0.024691358,0.086419754,0.049382716,0.11111111,0.0,0.11111111,0.11111111,0.09876543,0.074074075,0.012345679,0.049382716,0.086419754,0.11111111,0.09876543,0.0,0.19753087,0.11111111,0.11111111,0.11111111,0.012345679,0.13580246,0.0,0.20987654,0.20987654,0.12345679,0.11111111,0.22222222,0.11111111,0.12345679,0.11111111,0.12345679,0.18518518,0.09876543,0.16049382,0.12345679,0.27160493,0.24691358,0.12345679,0.11111111,0.20987654,0.2345679,0.2345679,0.33333334,0.33333334,0.27160493,0.37037036,0.37037036,0.32098764,0.28395063,0.20987654,0.4074074,,0.4074074,0.4074074,0.43209878,0.5308642,0.8518519
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.012345679,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.086419754,0.012345679,0.0,0.11111111,0.0,0.0,0.024691358,0.09876543,0.0,0.0,0.012345679,0.0,0.074074075,0.049382716,0.16049382,0.074074075,0.12345679,0.32098764,0.41975307,0.11111111,0.17283951,0.22222222,0.44444445,0.33333334,0.0,0.0,0.44444445,0.61728394,0.654321,0.49382716,0.4691358,0.54320985,0.50617284,0.5925926,,0.6419753,0.6666667,0.7777778,0.86419755
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.012345679,0.0,0.012345679,0.0,0.012345679,0.0,0.0,0.0,0.0,0.074074075,0.0,0.024691358,0.0,0.0,0.0,0.0,0.012345679,0.0,0.012345679,0.024691358,0.012345679,0.024691358,0.037037037,0.0,0.12345679,0.024691358,0.024691358,0.0,0.049382716,0.0,0.049382716,0.012345679,0.074074075,0.037037037,0.012345679,0.012345679,0.11111111,0.024691358,0.086419754,0.037037037,0.086419754,0.037037037,0.037037037,0.037037037,0.024691358,0.14814815,0.16049382,0.38271606,0.049382716,0.037037037,0.22222222,0.34567901,0.28395063,0.13580246,0.13580246,0.38271606,0.43209878,0.49382716,0.45679012,0.5185185,0.45679012,0.4074074,0.5925926,0.3580247,,0.50617284,0.6666667,0.7777778
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.012345679,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.037037037,0.0,0.0,0.0,0.037037037,0.0,0.0,0.11111111,0.0,0.0,0.0,0.024691358,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.12345679,0.2962963,0.3580247,0.012345679,0.09876543,0.20987654,0.41975307,0.0,0.13580246,0.13580246,0.41975307,0.38271606,0.4691358,0.38271606,0.41975307,0.38271606,0.56790125,0.56790125,0.33333334,0.49382716,,0.79012346,0.9876543
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.086419754,0.0,0.0,0.0,0.0,0.074074075,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.22222222,0.086419754,0.11111111,0.11111111,0.0,0.11111111,0.049382716,0.024691358,0.2345679,0.049382716,0.012345679,0.12345679,0.25925925,0.11111111,0.22222222,0.22222222,0.25925925,0.086419754,0.11111111,0.2345679,0.25925925,0.2345679,0.39506173,0.4691358,0.22222222,0.33333334,0.20987654,,0.86419755
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.12345679,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.049382716,0.049382716,0.0,0.12345679,0.0,0.061728396,0.14814815,0.13580246,0.22222222,0.012345679,0.13580246,
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_small/leaderboard_overview_explorer.html
index f19d1773df2a5cc83b747cdf04f49c1614a5862d..d31f0e5a2c9b4b2ced50754b6632a4100e2197d7 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Binary | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_binary/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..58ef65dcf85fa35bd47ff8c2a81bac566a6765c6
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_binary/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1742.0,"score":0.89,"rank":7.25,"hrank":1.75,"imp":4.546,"train_time":36.24,"predict_time":5.432,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+155\/-108","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1642.0,"score":0.679,"rank":11.03,"hrank":4.69,"imp":12.66,"train_time":3.85,"predict_time":0.374,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-59","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1588.0,"score":0.573,"rank":13.74,"hrank":5.21,"imp":10.35,"train_time":368.19,"predict_time":6.655,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-78","imputed":false},{"position":3,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1575.0,"score":0.633,"rank":14.45,"hrank":5.63,"imp":12.915,"train_time":2.09,"predict_time":0.152,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-86","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1564.0,"score":0.561,"rank":15.09,"hrank":7.5,"imp":14.043,"train_time":5.48,"predict_time":0.433,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-64","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1556.0,"score":0.543,"rank":15.51,"hrank":9.85,"imp":14.601,"train_time":5.96,"predict_time":0.554,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-41","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1553.0,"score":0.519,"rank":15.73,"hrank":7.62,"imp":14.103,"train_time":2052.28,"predict_time":8.252,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-62","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1528.0,"score":0.492,"rank":17.27,"hrank":8.19,"imp":14.694,"train_time":2052.28,"predict_time":1.265,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+116\/-55","imputed":false},{"position":8,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1470.0,"score":0.422,"rank":21.25,"hrank":9.99,"imp":16.408,"train_time":143.16,"predict_time":3.756,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+84\/-54","imputed":true},{"position":9,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1452.0,"score":0.362,"rank":22.6,"hrank":11.57,"imp":16.727,"train_time":3422.32,"predict_time":45.829,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+92\/-55","imputed":true},{"position":10,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1438.0,"score":0.371,"rank":23.62,"hrank":9.73,"imp":16.591,"train_time":19.82,"predict_time":4.463,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-79","imputed":false},{"position":11,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1438.0,"score":0.347,"rank":23.64,"hrank":12.94,"imp":16.527,"train_time":7.78,"predict_time":1.706,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+61\/-60","imputed":true},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1433.0,"score":0.363,"rank":24.05,"hrank":9.9,"imp":16.716,"train_time":2.3,"predict_time":0.114,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-62","imputed":false},{"position":13,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1414.0,"score":0.293,"rank":25.59,"hrank":16.59,"imp":17.873,"train_time":2950.72,"predict_time":15.589,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-49","imputed":false},{"position":14,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1403.0,"score":0.269,"rank":26.44,"hrank":12.6,"imp":17.997,"train_time":2289.91,"predict_time":2.709,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-51","imputed":false},{"position":15,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1397.0,"score":0.271,"rank":26.97,"hrank":17.03,"imp":17.748,"train_time":1921.63,"predict_time":1.717,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-67","imputed":false},{"position":16,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1388.0,"score":0.274,"rank":27.7,"hrank":12.5,"imp":17.122,"train_time":5351.93,"predict_time":268.579,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-67","imputed":false},{"position":17,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1381.0,"score":0.291,"rank":28.31,"hrank":12.64,"imp":19.036,"train_time":3422.32,"predict_time":0.982,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+116\/-65","imputed":true},{"position":18,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1378.0,"score":0.288,"rank":28.53,"hrank":15.14,"imp":19.181,"train_time":317.92,"predict_time":1.982,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-61","imputed":false},{"position":19,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1368.0,"score":0.305,"rank":29.43,"hrank":17.06,"imp":18.611,"train_time":1.23,"predict_time":0.052,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-71","imputed":false},{"position":20,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1359.0,"score":0.248,"rank":30.18,"hrank":12.93,"imp":17.699,"train_time":5351.93,"predict_time":37.606,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-69","imputed":false},{"position":21,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1354.0,"score":0.21,"rank":30.65,"hrank":14.36,"imp":18.871,"train_time":2289.91,"predict_time":0.266,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-50","imputed":false},{"position":22,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1353.0,"score":0.175,"rank":30.73,"hrank":22.3,"imp":19.104,"train_time":512.67,"predict_time":1.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-36","imputed":false},{"position":23,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1342.0,"score":0.286,"rank":31.71,"hrank":13.75,"imp":20.135,"train_time":4.05,"predict_time":0.421,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+96\/-74","imputed":true},{"position":24,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1322.0,"score":0.126,"rank":33.4,"hrank":24.96,"imp":19.524,"train_time":512.67,"predict_time":0.243,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-34","imputed":false},{"position":25,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1322.0,"score":0.152,"rank":33.45,"hrank":21.74,"imp":19.394,"train_time":15277.76,"predict_time":543.233,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-56","imputed":false},{"position":26,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1321.0,"score":0.171,"rank":33.48,"hrank":18.13,"imp":19.552,"train_time":2950.72,"predict_time":1.225,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-55","imputed":false},{"position":27,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1311.0,"score":0.137,"rank":34.39,"hrank":23.47,"imp":19.926,"train_time":1260.85,"predict_time":0.312,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-43","imputed":false},{"position":28,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1307.0,"score":0.148,"rank":34.75,"hrank":23.71,"imp":20.024,"train_time":3.83,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-44","imputed":false},{"position":29,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1307.0,"score":0.171,"rank":34.77,"hrank":17.34,"imp":19.964,"train_time":4806.04,"predict_time":0.512,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-61","imputed":false},{"position":30,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1306.0,"score":0.132,"rank":34.9,"hrank":24.23,"imp":20.032,"train_time":1260.85,"predict_time":0.031,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-42","imputed":false},{"position":31,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1297.0,"score":0.202,"rank":35.67,"hrank":14.67,"imp":21.692,"train_time":9.05,"predict_time":1.763,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-73","imputed":false},{"position":32,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1287.0,"score":0.207,"rank":36.56,"hrank":18.51,"imp":20.01,"train_time":51.83,"predict_time":43.611,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-77","imputed":false},{"position":33,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1280.0,"score":0.106,"rank":37.27,"hrank":26.93,"imp":20.448,"train_time":693.49,"predict_time":1.397,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-48","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1278.0,"score":0.121,"rank":37.44,"hrank":25.17,"imp":21.183,"train_time":550.77,"predict_time":0.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-56","imputed":false},{"position":35,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1277.0,"score":0.129,"rank":37.46,"hrank":22.53,"imp":20.508,"train_time":711.98,"predict_time":2.983,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-66","imputed":false},{"position":36,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1268.0,"score":0.146,"rank":38.27,"hrank":21.12,"imp":20.676,"train_time":6.8,"predict_time":0.248,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-60","imputed":false},{"position":37,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1267.0,"score":0.191,"rank":38.42,"hrank":17.74,"imp":21.262,"train_time":4806.04,"predict_time":8.45,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+94\/-103","imputed":false},{"position":38,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1259.0,"score":0.129,"rank":39.17,"hrank":18.91,"imp":22.098,"train_time":14.38,"predict_time":0.334,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-65","imputed":false},{"position":39,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1258.0,"score":0.08,"rank":39.23,"hrank":26.59,"imp":21.414,"train_time":1353.2,"predict_time":0.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-48","imputed":false},{"position":40,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1257.0,"score":0.09,"rank":39.3,"hrank":26.8,"imp":21.567,"train_time":550.77,"predict_time":0.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-52","imputed":false},{"position":41,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1256.0,"score":0.067,"rank":39.38,"hrank":26.29,"imp":20.793,"train_time":693.49,"predict_time":0.194,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-39","imputed":false},{"position":42,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1254.0,"score":0.102,"rank":39.55,"hrank":23.58,"imp":21.336,"train_time":711.98,"predict_time":0.162,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-65","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1250.0,"score":0.107,"rank":39.98,"hrank":28.61,"imp":20.758,"train_time":2665.48,"predict_time":2.619,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-52","imputed":false},{"position":44,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1237.0,"score":0.059,"rank":41.11,"hrank":27.63,"imp":21.915,"train_time":1353.2,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-48","imputed":false},{"position":45,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1232.0,"score":0.085,"rank":41.64,"hrank":23.35,"imp":22.168,"train_time":2.58,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-58","imputed":false},{"position":46,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1222.0,"score":0.058,"rank":42.49,"hrank":28.53,"imp":21.998,"train_time":15277.76,"predict_time":119.284,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-49","imputed":false},{"position":47,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1218.0,"score":0.068,"rank":42.92,"hrank":25.09,"imp":22.34,"train_time":6.67,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-55","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1217.0,"score":0.106,"rank":42.93,"hrank":26.34,"imp":21.995,"train_time":644.68,"predict_time":5.974,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-79","imputed":false},{"position":49,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1216.0,"score":0.071,"rank":43.08,"hrank":26.95,"imp":21.753,"train_time":10.9,"predict_time":2.248,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-51","imputed":false},{"position":50,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1186.0,"score":0.125,"rank":45.72,"hrank":28.3,"imp":23.257,"train_time":14.16,"predict_time":2.52,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-74","imputed":false},{"position":51,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1182.0,"score":0.229,"rank":46.1,"hrank":9.58,"imp":23.991,"train_time":34059.64,"predict_time":3.908,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+117\/-95","imputed":false},{"position":52,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1181.0,"score":0.051,"rank":46.22,"hrank":33.6,"imp":22.694,"train_time":2665.48,"predict_time":0.143,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-45","imputed":false},{"position":53,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1176.0,"score":0.219,"rank":46.67,"hrank":9.27,"imp":24.14,"train_time":34059.64,"predict_time":18.811,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-94","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1170.0,"score":0.046,"rank":47.21,"hrank":28.42,"imp":23.313,"train_time":644.68,"predict_time":0.398,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-83","imputed":false},{"position":55,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1160.0,"score":0.035,"rank":48.01,"hrank":36.52,"imp":23.812,"train_time":346.84,"predict_time":0.936,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-81","imputed":false},{"position":56,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1157.0,"score":0.027,"rank":48.3,"hrank":39.14,"imp":23.364,"train_time":1.94,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-55","imputed":false},{"position":57,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1143.0,"score":0.038,"rank":49.52,"hrank":34.87,"imp":24.436,"train_time":346.84,"predict_time":0.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-82","imputed":false},{"position":58,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1130.0,"score":0.019,"rank":50.61,"hrank":41.88,"imp":23.696,"train_time":2.44,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-55","imputed":false},{"position":59,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1129.0,"score":0.062,"rank":50.71,"hrank":26.46,"imp":24.72,"train_time":335.81,"predict_time":85.05,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-68","imputed":false},{"position":60,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1111.0,"score":0.044,"rank":52.21,"hrank":35.51,"imp":25.753,"train_time":218.34,"predict_time":0.904,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-81","imputed":false},{"position":61,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1103.0,"score":0.025,"rank":52.9,"hrank":35.93,"imp":25.614,"train_time":484.44,"predict_time":1.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-96","imputed":false},{"position":62,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1092.0,"score":0.145,"rank":53.73,"hrank":15.73,"imp":28.393,"train_time":460.12,"predict_time":6.641,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-127","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1077.0,"score":0.024,"rank":54.98,"hrank":38.25,"imp":27.233,"train_time":218.34,"predict_time":0.322,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-81","imputed":false},{"position":64,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1072.0,"score":0.082,"rank":55.31,"hrank":30.75,"imp":27.744,"train_time":1.3,"predict_time":0.15,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-126","imputed":false},{"position":65,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1070.0,"score":0.005,"rank":55.55,"hrank":40.02,"imp":26.177,"train_time":484.44,"predict_time":0.086,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-94","imputed":false},{"position":66,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1054.0,"score":0.017,"rank":56.74,"hrank":38.89,"imp":26.548,"train_time":9.39,"predict_time":0.13,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-89","imputed":false},{"position":67,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1023.0,"score":0.069,"rank":58.99,"hrank":33.37,"imp":31.491,"train_time":278.97,"predict_time":0.873,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-133","imputed":false},{"position":68,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1018.0,"score":0.0,"rank":59.32,"hrank":34.63,"imp":27.207,"train_time":2.97,"predict_time":0.388,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-92","imputed":false},{"position":69,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1003.0,"score":0.046,"rank":60.38,"hrank":35.47,"imp":32.005,"train_time":278.97,"predict_time":0.103,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-132","imputed":false},{"position":70,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1001.0,"score":0.013,"rank":60.51,"hrank":50.33,"imp":28.665,"train_time":2.07,"predict_time":0.314,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-128","imputed":false},{"position":71,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.005,"rank":60.57,"hrank":51.92,"imp":28.938,"train_time":0.42,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-123","imputed":false},{"position":72,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":992.0,"score":0.013,"rank":61.13,"hrank":42.34,"imp":29.1,"train_time":90.74,"predict_time":1.622,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-118","imputed":false},{"position":73,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":988.0,"score":0.005,"rank":61.34,"hrank":52.79,"imp":30.363,"train_time":3.71,"predict_time":0.919,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-115","imputed":false},{"position":74,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":971.0,"score":0.0,"rank":62.45,"hrank":47.68,"imp":29.47,"train_time":30.83,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-68","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":959.0,"score":0.03,"rank":63.17,"hrank":38.72,"imp":33.917,"train_time":1.48,"predict_time":0.139,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-165","imputed":false},{"position":76,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":925.0,"score":0.001,"rank":65.11,"hrank":43.79,"imp":33.156,"train_time":90.74,"predict_time":0.125,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-141","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":634.0,"score":0.0,"rank":75.05,"hrank":65.95,"imp":45.684,"train_time":0.22,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-195","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_small/pareto_front_explorer.html
index 88d68599d12af8493f518e272e6c8eb09bcbe6a1..3e08c805e89885c59d2025cbde1276647d0539f7 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.9820731122,"imp":19.1811602099,"elo":1378.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"imp":20.0242216574,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0307451934,"imp":20.0324071366,"elo":1305.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3119174081,"imp":19.926072515,"elo":1311.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0288287122,"imp":22.1679216195,"elo":1231.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0276568517,"imp":21.5673765461,"elo":1257.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2363905932,"imp":21.1832027115,"elo":1277.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0179394275,"imp":22.3395403978,"elo":1217.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0173877944,"imp":21.9154953614,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1553298525,"imp":21.4139588079,"elo":1258.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3137781461,"imp":28.664672317,"elo":1000.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0776142438,"imp":24.436387867,"elo":1143.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9356355775,"imp":23.8117031222,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3877560139,"imp":27.2071020716,"elo":1018.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3977238019,"imp":23.3125984662,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.9735168391,"imp":21.9949922043,"elo":1217.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1213716865,"imp":23.6955869546,"elo":1130.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.2429152387,"imp":19.5241629327,"elo":1322.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1797709942,"imp":19.1040332829,"elo":1352.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.4627845186,"imp":16.591241611,"elo":1438.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.755821965,"imp":16.4083482505,"elo":1469.9,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3338546996,"imp":22.0977967392,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5122402643,"imp":19.9641300257,"elo":1307.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4495512241,"imp":21.2620323993,"elo":1266.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"imp":23.2568115498,"elo":1186.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"imp":29.469789529,"elo":970.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3219913165,"imp":27.2334261506,"elo":1076.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9044251985,"imp":25.7534558024,"elo":1111.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0408261303,"imp":28.9381469731,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0857054373,"imp":26.1772091898,"elo":1069.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.0254370493,"imp":25.6143171618,"elo":1102.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":2.2477938016,"imp":21.7526555902,"elo":1215.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.2247970899,"imp":19.5517513837,"elo":1321.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":15.5885181427,"imp":17.8730281514,"elo":1413.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5540470709,"imp":14.6009494593,"elo":1556.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2654117448,"imp":14.6935178534,"elo":1527.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.2520254135,"imp":14.1032633297,"elo":1552.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7629525797,"imp":21.6920036226,"elo":1297.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.6109974271,"imp":20.0097905959,"elo":1287.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":37.6060491721,"imp":17.6987843151,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":268.5791952447,"imp":17.1220802823,"elo":1388.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1141716321,"imp":16.7161479147,"elo":1433.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.4319322507,"imp":4.5458876024,"elo":1741.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1499608083,"imp":27.7436121895,"elo":1072.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.7059832513,"imp":16.527196736,"elo":1438.2,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1522267247,"imp":12.9151014592,"elo":1574.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2482801599,"imp":20.6756027273,"elo":1268.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2655194186,"imp":18.8706782213,"elo":1353.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.7085526712,"imp":17.9972551869,"elo":1403.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4326735827,"imp":14.0429877919,"elo":1563.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.374483993,"imp":12.660278659,"elo":1642.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4207897504,"imp":20.135035593,"elo":1341.5,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.9822813165,"imp":19.035952536,"elo":1380.9,"imputed":true,"imputed_pct":4.7619047619,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":45.8293605146,"imp":16.7268707767,"elo":1451.7,"imputed":true,"imputed_pct":4.7619047619,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.6409697056,"imp":28.3933673819,"elo":1092.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.9075519343,"imp":23.9910274218,"elo":1182.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"imp":24.1403906089,"elo":1175.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0519986869,"imp":18.6108772842,"elo":1367.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1303337018,"imp":26.5476346741,"elo":1053.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1432880461,"imp":22.694289653,"elo":1180.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.6194496632,"imp":20.7581518352,"elo":1249.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209029857,"imp":23.3641520991,"elo":1157.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1936851087,"imp":20.7934041802,"elo":1256.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.3974981596,"imp":20.4477557738,"elo":1279.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":85.0504523595,"imp":24.7197554793,"elo":1128.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":119.2837367952,"imp":21.9979920045,"elo":1222.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":543.2330416203,"imp":19.3944150025,"elo":1321.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"imp":30.3632577827,"elo":988.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1622176419,"imp":21.3357560238,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9831667257,"imp":20.5079540276,"elo":1277.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.9820731122,"x_train":317.9218245308,"imp":19.1811602099,"elo":1378.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"x_train":3.8348511861,"imp":20.0242216574,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0307451934,"x_train":1260.8478030837,"imp":20.0324071366,"elo":1305.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3119174081,"x_train":1260.8478030837,"imp":19.926072515,"elo":1311.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0288287122,"x_train":2.5798987548,"imp":22.1679216195,"elo":1231.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0276568517,"x_train":550.7724074523,"imp":21.5673765461,"elo":1257.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2363905932,"x_train":550.7724074523,"imp":21.1832027115,"elo":1277.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0179394275,"x_train":6.6731468509,"imp":22.3395403978,"elo":1217.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0173877944,"x_train":1353.2049878904,"imp":21.9154953614,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1553298525,"x_train":1353.2049878904,"imp":21.4139588079,"elo":1258.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3137781461,"x_train":2.0680378675,"imp":28.664672317,"elo":1000.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0776142438,"x_train":346.8355167309,"imp":24.436387867,"elo":1143.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9356355775,"x_train":346.8355167309,"imp":23.8117031222,"elo":1160.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3877560139,"x_train":2.9685174704,"imp":27.2071020716,"elo":1018.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3977238019,"x_train":644.6819040616,"imp":23.3125984662,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.9735168391,"x_train":644.6819040616,"imp":21.9949922043,"elo":1217.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1213716865,"x_train":2.4426615648,"imp":23.6955869546,"elo":1130.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.2429152387,"x_train":512.6694116522,"imp":19.5241629327,"elo":1322.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1797709942,"x_train":512.6694116522,"imp":19.1040332829,"elo":1352.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.4627845186,"x_train":19.824824253,"imp":16.591241611,"elo":1438.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.755821965,"x_train":143.1617088461,"imp":16.4083482505,"elo":1469.9,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3338546996,"x_train":14.3767255664,"imp":22.0977967392,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5122402643,"x_train":4806.0384822712,"imp":19.9641300257,"elo":1307.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4495512241,"x_train":4806.0384822712,"imp":21.2620323993,"elo":1266.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"x_train":14.1568412145,"imp":23.2568115498,"elo":1186.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"x_train":30.8283410858,"imp":29.469789529,"elo":970.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3219913165,"x_train":218.3366110643,"imp":27.2334261506,"elo":1076.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9044251985,"x_train":218.3366110643,"imp":25.7534558024,"elo":1111.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0408261303,"x_train":0.4188243081,"imp":28.9381469731,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0857054373,"x_train":484.4354636329,"imp":26.1772091898,"elo":1069.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.0254370493,"x_train":484.4354636329,"imp":25.6143171618,"elo":1102.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":2.2477938016,"x_train":10.8961223125,"imp":21.7526555902,"elo":1215.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.2247970899,"x_train":2950.7244476948,"imp":19.5517513837,"elo":1321.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":15.5885181427,"x_train":2950.7244476948,"imp":17.8730281514,"elo":1413.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5540470709,"x_train":5.9621453989,"imp":14.6009494593,"elo":1556.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2654117448,"x_train":2052.2840837184,"imp":14.6935178534,"elo":1527.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.2520254135,"x_train":2052.2840837184,"imp":14.1032633297,"elo":1552.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7629525797,"x_train":9.0524779449,"imp":21.6920036226,"elo":1297.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.6109974271,"x_train":51.8297758697,"imp":20.0097905959,"elo":1287.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":37.6060491721,"x_train":5351.9343480662,"imp":17.6987843151,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":268.5791952447,"x_train":5351.9343480662,"imp":17.1220802823,"elo":1388.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1141716321,"x_train":2.2957537095,"imp":16.7161479147,"elo":1433.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.4319322507,"x_train":36.2449142975,"imp":4.5458876024,"elo":1741.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1499608083,"x_train":1.2985567013,"imp":27.7436121895,"elo":1072.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.7059832513,"x_train":7.7819495177,"imp":16.527196736,"elo":1438.2,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1522267247,"x_train":2.0883166208,"imp":12.9151014592,"elo":1574.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2482801599,"x_train":6.8016802788,"imp":20.6756027273,"elo":1268.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2655194186,"x_train":2289.910634932,"imp":18.8706782213,"elo":1353.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.7085526712,"x_train":2289.910634932,"imp":17.9972551869,"elo":1403.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4326735827,"x_train":5.4825798778,"imp":14.0429877919,"elo":1563.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.374483993,"x_train":3.8458762973,"imp":12.660278659,"elo":1642.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4207897504,"x_train":4.0473610858,"imp":20.135035593,"elo":1341.5,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.9822813165,"x_train":3422.3190042951,"imp":19.035952536,"elo":1380.9,"imputed":true,"imputed_pct":4.7619047619,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":45.8293605146,"x_train":3422.3190042951,"imp":16.7268707767,"elo":1451.7,"imputed":true,"imputed_pct":4.7619047619,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.6409697056,"x_train":460.1168119192,"imp":28.3933673819,"elo":1092.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.9075519343,"x_train":34059.6410419582,"imp":23.9910274218,"elo":1182.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"x_train":34059.6410419582,"imp":24.1403906089,"elo":1175.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0519986869,"x_train":1.2303610879,"imp":18.6108772842,"elo":1367.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1303337018,"x_train":9.3924084625,"imp":26.5476346741,"elo":1053.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1432880461,"x_train":2665.4817950505,"imp":22.694289653,"elo":1180.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.6194496632,"x_train":2665.4817950505,"imp":20.7581518352,"elo":1249.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209029857,"x_train":1.9409389396,"imp":23.3641520991,"elo":1157.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1936851087,"x_train":693.4907982506,"imp":20.7934041802,"elo":1256.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.3974981596,"x_train":693.4907982506,"imp":20.4477557738,"elo":1279.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":85.0504523595,"x_train":335.8142132441,"imp":24.7197554793,"elo":1128.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":119.2837367952,"x_train":15277.755542326,"imp":21.9979920045,"elo":1222.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":543.2330416203,"x_train":15277.755542326,"imp":19.3944150025,"elo":1321.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"x_train":3.7137863239,"imp":30.3632577827,"elo":988.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1622176419,"x_train":711.9845683972,"imp":21.3357560238,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9831667257,"x_train":711.9845683972,"imp":20.5079540276,"elo":1277.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_small/tuning_trajectories_explorer.html
index 7a069b0ca6b201ab62f78fbaadee18301058ffe8..1a3ba4f4343ecd66f1885546acd59e8c3766e9a9 100644
--- a/data/imputation_yes/splits_all/tasks_binary/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":317.9218245308,"imp":19.2073047961,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.8348511861,"imp":20.0491690361,"elo":1326.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":10.7408674121,"imp":19.8460079452,"elo":1336.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":27.0793552473,"imp":19.7228006279,"elo":1342.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":55.6913166543,"imp":19.7330658478,"elo":1342.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":153.0098769834,"imp":19.7718807572,"elo":1340.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":311.0672039337,"imp":19.8038127676,"elo":1337.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":627.0754171345,"imp":19.8696232973,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":940.3702560612,"imp":19.9160713597,"elo":1330.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1260.8478030837,"imp":19.9469363418,"elo":1327.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.5798987548,"imp":22.1918135534,"elo":1244.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.3455889861,"imp":21.8714785964,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":13.3365552104,"imp":21.7255660388,"elo":1271.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":26.9265316391,"imp":21.6399719457,"elo":1275.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":68.2541543508,"imp":21.4602155764,"elo":1284.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":136.870208803,"imp":21.3666147717,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":274.0188126874,"imp":21.2770307869,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":410.8168496029,"imp":21.2338870846,"elo":1295.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":550.7724074523,"imp":21.2061956966,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":6.6731468509,"imp":22.3649405125,"elo":1227.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":13.6833132635,"imp":22.1327006436,"elo":1235.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":33.7629667518,"imp":21.8647800895,"elo":1249.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":67.3034244419,"imp":21.7831275998,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":168.595185677,"imp":21.6901058942,"elo":1259.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":338.3045915746,"imp":21.572342468,"elo":1266.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":673.6824905284,"imp":21.4822397611,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1009.6693952358,"imp":21.4608334299,"elo":1272.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1353.2049878904,"imp":21.4462009392,"elo":1274.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.0680378675,"imp":28.6879692178,"elo":1002.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.8007843383,"imp":27.2526129884,"elo":1037.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.0291964114,"imp":25.229110833,"elo":1100.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":17.6458570814,"imp":24.4630789144,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.5325190163,"imp":24.1026889051,"elo":1153.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":86.5677692966,"imp":23.9586129971,"elo":1162.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":172.7997856037,"imp":23.8797593507,"elo":1168.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":258.9344533948,"imp":23.8328273886,"elo":1173.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":346.8355167309,"imp":23.8393239401,"elo":1173.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.9685174704,"imp":27.2312238502,"elo":1026.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.1141929845,"imp":25.8959998044,"elo":1072.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":16.0049294587,"imp":24.0279296921,"elo":1147.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":32.4300042816,"imp":23.0951222725,"elo":1183.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":78.9900203661,"imp":22.4758440382,"elo":1206.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":158.4909250486,"imp":22.1909776224,"elo":1220.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":319.6026300848,"imp":22.0732724985,"elo":1226.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":480.1317591961,"imp":22.0005168524,"elo":1231.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":644.6819040616,"imp":22.0140916279,"elo":1232.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.4426615648,"imp":23.719342605,"elo":1134.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":5.5591739929,"imp":21.226021134,"elo":1242.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":13.0753440884,"imp":19.9355701673,"elo":1321.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":25.7142984261,"imp":19.4869043166,"elo":1349.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":63.8902151248,"imp":19.2706957669,"elo":1364.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":127.6621046473,"imp":19.2190818503,"elo":1369.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":255.1492455474,"imp":19.1455287764,"elo":1374.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":382.2249600708,"imp":19.1169651442,"elo":1376.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":512.6694116522,"imp":19.1277371842,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":19.824824253,"imp":16.6529978651,"elo":1457.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":143.1617088461,"imp":16.4341774961,"elo":1490.5,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.3767255664,"imp":22.1224639392,"elo":1275.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":49.2187024257,"imp":21.2479358426,"elo":1284.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":133.7824058533,"imp":19.4775231108,"elo":1351.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":249.6696643307,"imp":19.2433732744,"elo":1357.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":597.3147049605,"imp":19.030076102,"elo":1363.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1201.702225649,"imp":19.5199392498,"elo":1342.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2376.3135821935,"imp":20.2990586092,"elo":1313.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3550.4691289242,"imp":20.9428375602,"elo":1293.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4806.0384822712,"imp":21.2938019906,"elo":1280.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":14.1568412145,"imp":23.2779956432,"elo":1198.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":30.8283410858,"imp":30.4715128195,"elo":963.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":72.2336761586,"imp":28.0600799437,"elo":1026.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":181.7482066596,"imp":25.8878613807,"elo":1115.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":218.3366110643,"imp":25.7760481664,"elo":1121.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4188243081,"imp":28.961510924,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.8140094679,"imp":27.7140235594,"elo":1033.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":10.1111434028,"imp":26.8722837086,"elo":1059.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":22.2070189832,"imp":26.3434913426,"elo":1081.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":58.3686895535,"imp":25.8833179252,"elo":1101.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":118.7702866394,"imp":25.7546410674,"elo":1108.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":239.7007801605,"imp":25.675966336,"elo":1111.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":360.8189481305,"imp":25.641463925,"elo":1113.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":484.4354636329,"imp":25.6350837875,"elo":1114.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.8961223125,"imp":21.7774684311,"elo":1225.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.6705594408,"imp":19.9333987249,"elo":1313.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":69.643527334,"imp":18.6018161298,"elo":1390.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":141.1529977589,"imp":18.2214076301,"elo":1414.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":364.6011469598,"imp":17.9942456806,"elo":1429.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":736.9754853516,"imp":17.9217090779,"elo":1436.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1469.2653973047,"imp":17.9146102843,"elo":1439.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2197.5320041687,"imp":17.9194133741,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2950.7244476948,"imp":17.9010328545,"elo":1438.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.9621453989,"imp":14.6272026518,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.7696901807,"imp":14.6555481702,"elo":1580.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":46.2247717016,"imp":14.4405113927,"elo":1582.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":96.4776470867,"imp":14.363075781,"elo":1581.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":247.5589870969,"imp":14.20294925,"elo":1581.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":503.7582274337,"imp":14.1582130482,"elo":1581.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1020.2013695698,"imp":14.0883010787,"elo":1580.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1526.9732493853,"imp":14.0889177383,"elo":1579.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2052.2840837184,"imp":14.1446676759,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":9.0524779449,"imp":21.7173550214,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.8297758697,"imp":20.0348348517,"elo":1304.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":77.5384458131,"imp":19.194721807,"elo":1332.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":145.4125168687,"imp":18.4273080767,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":257.902985689,"imp":18.1603149152,"elo":1374.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":651.1867262454,"imp":17.5539082701,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1324.8715392081,"imp":17.3106526929,"elo":1402.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2657.8350777127,"imp":17.1881361874,"elo":1407.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3995.7346684077,"imp":17.1388702767,"elo":1409.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":5351.9343480662,"imp":17.1570967003,"elo":1409.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.2957537095,"imp":16.7412269175,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":36.2449142975,"imp":4.5733418206,"elo":1766.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.2985567013,"imp":27.7669837303,"elo":1089.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":7.7819495177,"imp":16.5528333149,"elo":1458.8,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":2.0883166208,"imp":12.9414968056,"elo":1599.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.8016802788,"imp":20.7003293178,"elo":1283.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":15.442951022,"imp":20.2958947274,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":49.8362170361,"imp":19.7207528158,"elo":1335.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":110.6090004093,"imp":19.1110430769,"elo":1372.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":279.2571002214,"imp":18.6247235172,"elo":1396.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":563.1030178809,"imp":18.3458272833,"elo":1409.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1110.4089598597,"imp":18.1617129153,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1689.4279279682,"imp":18.0690311939,"elo":1424.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2289.910634932,"imp":18.0241104932,"elo":1427.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.4825798778,"imp":14.0688884717,"elo":1590.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.8458762973,"imp":12.68620022,"elo":1671.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4.0473610858,"imp":20.1595766294,"elo":1362.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":26.1321442475,"imp":19.1991974493,"elo":1398.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":70.8191350853,"imp":18.2821820827,"elo":1436.8,"imputed":true,"imputed_pct":4.7619047619,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":154.3706019343,"imp":17.6502152507,"elo":1455.6,"imputed":true,"imputed_pct":4.7619047619,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":393.1065181502,"imp":17.2251201695,"elo":1466.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":810.6442685856,"imp":16.9708904221,"elo":1473.8,"imputed":true,"imputed_pct":4.7619047619,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1663.0513945281,"imp":16.8319859952,"elo":1476.6,"imputed":true,"imputed_pct":4.7619047619,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2512.506403524,"imp":16.766590605,"elo":1477.6,"imputed":true,"imputed_pct":4.7619047619,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3422.3190042951,"imp":16.7316092766,"elo":1475.3,"imputed":true,"imputed_pct":4.7619047619,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":460.1168119192,"imp":28.4181504004,"elo":1103.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":734.7515368462,"imp":28.2874148314,"elo":1108.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1321.1026922005,"imp":28.0158248609,"elo":1115.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1962.2511632163,"imp":27.2907198112,"elo":1129.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4372.5625009241,"imp":26.1241726178,"elo":1150.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":8707.0506318291,"imp":25.0535474398,"elo":1169.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":17339.6234835031,"imp":24.5364252305,"elo":1182.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":25624.7257379004,"imp":24.3192783469,"elo":1188.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34059.6410419582,"imp":24.1668683111,"elo":1194.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.2303610879,"imp":18.6357378636,"elo":1387.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.3924084625,"imp":26.5598478142,"elo":1066.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":19.9499332005,"imp":24.2714211661,"elo":1131.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":60.5341760166,"imp":22.5260053181,"elo":1181.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":136.8122971072,"imp":21.8694078242,"elo":1208.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":328.250602853,"imp":21.3740871991,"elo":1232.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":672.0587261642,"imp":21.065050993,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1298.4117955909,"imp":20.9202509656,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1997.6434256187,"imp":20.8504804077,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2665.4817950505,"imp":20.770209269,"elo":1268.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.9409389396,"imp":23.3874294021,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.5392416815,"imp":21.4082919157,"elo":1239.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":16.764226514,"imp":20.9996572903,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":31.9140091666,"imp":20.8223537215,"elo":1274.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":85.9052711013,"imp":20.6434014804,"elo":1284.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":172.6999358448,"imp":20.6020275639,"elo":1287.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":344.2808053206,"imp":20.5560816099,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":515.1408947248,"imp":20.5055496641,"elo":1293.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":693.4907982506,"imp":20.472272682,"elo":1296.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":335.8142132441,"imp":24.743436711,"elo":1140.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":882.5688189042,"imp":22.7405128087,"elo":1194.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2676.7112743199,"imp":20.6841825925,"elo":1273.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":5942.9139958052,"imp":19.9896126739,"elo":1306.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":14596.1261611529,"imp":19.4339464722,"elo":1338.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":15277.755542326,"imp":19.4176388669,"elo":1339.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.7137863239,"imp":30.3870258221,"elo":995.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":7.2328490388,"imp":27.0959316084,"elo":1054.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.7783258681,"imp":22.7473525442,"elo":1190.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":35.5509910734,"imp":20.9811671609,"elo":1276.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.843717188,"imp":20.487207883,"elo":1303.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":177.5290085781,"imp":20.4116191946,"elo":1305.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":354.4498168115,"imp":20.4735161266,"elo":1299.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":530.9324752986,"imp":20.5152582222,"elo":1296.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":711.9845683972,"imp":20.5346806965,"elo":1293.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.9820731122,"x_train":317.9218245308,"imp":19.2073047961,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0248961767,"x_train":3.8348511861,"imp":20.0491690361,"elo":1326.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0505639318,"x_train":10.7408674121,"imp":19.8460079452,"elo":1336.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1055185035,"x_train":27.0793552473,"imp":19.7228006279,"elo":1342.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1439083199,"x_train":55.6913166543,"imp":19.7330658478,"elo":1342.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1870134039,"x_train":153.0098769834,"imp":19.7718807572,"elo":1340.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2323100656,"x_train":311.0672039337,"imp":19.8038127676,"elo":1337.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2769297222,"x_train":627.0754171345,"imp":19.8696232973,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3021056593,"x_train":940.3702560612,"imp":19.9160713597,"elo":1330.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3157761587,"x_train":1260.8478030837,"imp":19.9469363418,"elo":1327.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0288287122,"x_train":2.5798987548,"imp":22.1918135534,"elo":1244.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0559389647,"x_train":5.3455889861,"imp":21.8714785964,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.095923574,"x_train":13.3365552104,"imp":21.7255660388,"elo":1271.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1330256679,"x_train":26.9265316391,"imp":21.6399719457,"elo":1275.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1721658573,"x_train":68.2541543508,"imp":21.4602155764,"elo":1284.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1894366664,"x_train":136.870208803,"imp":21.3666147717,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2208324522,"x_train":274.0188126874,"imp":21.2770307869,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.226853687,"x_train":410.8168496029,"imp":21.2338870846,"elo":1295.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2454393854,"x_train":550.7724074523,"imp":21.2061956966,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0179394275,"x_train":6.6731468509,"imp":22.3649405125,"elo":1227.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.034525777,"x_train":13.6833132635,"imp":22.1327006436,"elo":1235.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.058506366,"x_train":33.7629667518,"imp":21.8647800895,"elo":1249.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0797047657,"x_train":67.3034244419,"imp":21.7831275998,"elo":1254.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1112174743,"x_train":168.595185677,"imp":21.6901058942,"elo":1259.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.128617728,"x_train":338.3045915746,"imp":21.572342468,"elo":1266.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1479697023,"x_train":673.6824905284,"imp":21.4822397611,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1471018087,"x_train":1009.6693952358,"imp":21.4608334299,"elo":1272.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1455019881,"x_train":1353.2049878904,"imp":21.4462009392,"elo":1274.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3137781461,"x_train":2.0680378675,"imp":28.6879692178,"elo":1002.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3830515747,"x_train":3.8007843383,"imp":27.2526129884,"elo":1037.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4375192148,"x_train":9.0291964114,"imp":25.229110833,"elo":1100.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4630669777,"x_train":17.6458570814,"imp":24.4630789144,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5956078235,"x_train":43.5325190163,"imp":24.1026889051,"elo":1153.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7076605511,"x_train":86.5677692966,"imp":23.9586129971,"elo":1162.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7982051709,"x_train":172.7997856037,"imp":23.8797593507,"elo":1168.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8552846086,"x_train":258.9344533948,"imp":23.8328273886,"elo":1173.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9356355775,"x_train":346.8355167309,"imp":23.8393239401,"elo":1173.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3877560139,"x_train":2.9685174704,"imp":27.2312238502,"elo":1026.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6804478637,"x_train":6.1141929845,"imp":25.8959998044,"elo":1072.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3000093411,"x_train":16.0049294587,"imp":24.0279296921,"elo":1147.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0596118244,"x_train":32.4300042816,"imp":23.0951222725,"elo":1183.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1608857062,"x_train":78.9900203661,"imp":22.4758440382,"elo":1206.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.8825453215,"x_train":158.4909250486,"imp":22.1909776224,"elo":1220.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.8073759391,"x_train":319.6026300848,"imp":22.0732724985,"elo":1226.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6151177765,"x_train":480.1317591961,"imp":22.0005168524,"elo":1231.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.0018475561,"x_train":644.6819040616,"imp":22.0140916279,"elo":1232.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1213716865,"x_train":2.4426615648,"imp":23.719342605,"elo":1134.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4646453215,"x_train":5.5591739929,"imp":21.226021134,"elo":1242.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7052079417,"x_train":13.0753440884,"imp":19.9355701673,"elo":1321.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9222412801,"x_train":25.7142984261,"imp":19.4869043166,"elo":1349.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1944952069,"x_train":63.8902151248,"imp":19.2706957669,"elo":1364.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3691324541,"x_train":127.6621046473,"imp":19.2190818503,"elo":1369.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3439559328,"x_train":255.1492455474,"imp":19.1455287764,"elo":1374.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3558664616,"x_train":382.2249600708,"imp":19.1169651442,"elo":1376.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.190246053,"x_train":512.6694116522,"imp":19.1277371842,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.4627845186,"x_train":19.824824253,"imp":16.6529978651,"elo":1457.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":3.755821965,"x_train":143.1617088461,"imp":16.4341774961,"elo":1490.5,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3338546996,"x_train":14.3767255664,"imp":22.1224639392,"elo":1275.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8515440488,"x_train":49.2187024257,"imp":21.2479358426,"elo":1284.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9302543292,"x_train":133.7824058533,"imp":19.4775231108,"elo":1351.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.8814779166,"x_train":249.6696643307,"imp":19.2433732744,"elo":1357.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1438371055,"x_train":597.3147049605,"imp":19.030076102,"elo":1363.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.494799053,"x_train":1201.702225649,"imp":19.5199392498,"elo":1342.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.0528486148,"x_train":2376.3135821935,"imp":20.2990586092,"elo":1313.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.9073730127,"x_train":3550.4691289242,"imp":20.9428375602,"elo":1293.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4341438441,"x_train":4806.0384822712,"imp":21.2938019906,"elo":1280.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5200162431,"x_train":14.1568412145,"imp":23.2779956432,"elo":1198.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0285805066,"x_train":30.8283410858,"imp":30.4715128195,"elo":963.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2383939645,"x_train":72.2336761586,"imp":28.0600799437,"elo":1026.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7611719559,"x_train":181.7482066596,"imp":25.8878613807,"elo":1115.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9044251985,"x_train":218.3366110643,"imp":25.7760481664,"elo":1121.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0408261303,"x_train":0.4188243081,"imp":28.961510924,"elo":999.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1187728047,"x_train":2.8140094679,"imp":27.7140235594,"elo":1033.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2923837664,"x_train":10.1111434028,"imp":26.8722837086,"elo":1059.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4489618597,"x_train":22.2070189832,"imp":26.3434913426,"elo":1081.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.643034933,"x_train":58.3686895535,"imp":25.8833179252,"elo":1101.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7794781581,"x_train":118.7702866394,"imp":25.7546410674,"elo":1108.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9035741347,"x_train":239.7007801605,"imp":25.675966336,"elo":1111.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9768570151,"x_train":360.8189481305,"imp":25.641463925,"elo":1113.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0127633212,"x_train":484.4354636329,"imp":25.6350837875,"elo":1114.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.2477938016,"x_train":10.8961223125,"imp":21.7774684311,"elo":1225.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.7014714901,"x_train":24.6705594408,"imp":19.9333987249,"elo":1313.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.7878975485,"x_train":69.643527334,"imp":18.6018161298,"elo":1390.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.822444877,"x_train":141.1529977589,"imp":18.2214076301,"elo":1414.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.0337121308,"x_train":364.6011469598,"imp":17.9942456806,"elo":1429.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.0024192762,"x_train":736.9754853516,"imp":17.9217090779,"elo":1436.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.6952744743,"x_train":1469.2653973047,"imp":17.9146102843,"elo":1439.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.7348675092,"x_train":2197.5320041687,"imp":17.9194133741,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.535629646,"x_train":2950.7244476948,"imp":17.9010328545,"elo":1438.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5540470709,"x_train":5.9621453989,"imp":14.6272026518,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3798151402,"x_train":16.7696901807,"imp":14.6555481702,"elo":1580.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.6778202239,"x_train":46.2247717016,"imp":14.4405113927,"elo":1582.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.9104411493,"x_train":96.4776470867,"imp":14.363075781,"elo":1581.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.6891164694,"x_train":247.5589870969,"imp":14.20294925,"elo":1581.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8180134178,"x_train":503.7582274337,"imp":14.1582130482,"elo":1581.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6847264435,"x_train":1020.2013695698,"imp":14.0883010787,"elo":1580.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.2438200441,"x_train":1526.9732493853,"imp":14.0889177383,"elo":1579.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.660934628,"x_train":2052.2840837184,"imp":14.1446676759,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7629525797,"x_train":9.0524779449,"imp":21.7173550214,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.6109974271,"x_train":51.8297758697,"imp":20.0348348517,"elo":1304.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.9837454782,"x_train":77.5384458131,"imp":19.194721807,"elo":1332.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":67.6378270251,"x_train":145.4125168687,"imp":18.4273080767,"elo":1364.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":86.9150727201,"x_train":257.902985689,"imp":18.1603149152,"elo":1374.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":128.1526540176,"x_train":651.1867262454,"imp":17.5539082701,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":168.6065426596,"x_train":1324.8715392081,"imp":17.3106526929,"elo":1402.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":206.1759779511,"x_train":2657.8350777127,"imp":17.1881361874,"elo":1407.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":240.0720860454,"x_train":3995.7346684077,"imp":17.1388702767,"elo":1409.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":264.8900452851,"x_train":5351.9343480662,"imp":17.1570967003,"elo":1409.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1141716321,"x_train":2.2957537095,"imp":16.7412269175,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":5.4319322507,"x_train":36.2449142975,"imp":4.5733418206,"elo":1766.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1499608083,"x_train":1.2985567013,"imp":27.7669837303,"elo":1089.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.7059832513,"x_train":7.7819495177,"imp":16.5528333149,"elo":1458.8,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1522267247,"x_train":2.0883166208,"imp":12.9414968056,"elo":1599.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2482801599,"x_train":6.8016802788,"imp":20.7003293178,"elo":1283.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.5017988445,"x_train":15.442951022,"imp":20.2958947274,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9596655794,"x_train":49.8362170361,"imp":19.7207528158,"elo":1335.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.3053184018,"x_train":110.6090004093,"imp":19.1110430769,"elo":1372.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.7312799575,"x_train":279.2571002214,"imp":18.6247235172,"elo":1396.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":2.0837991009,"x_train":563.1030178809,"imp":18.3458272833,"elo":1409.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":2.3607400079,"x_train":1110.4089598597,"imp":18.1617129153,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.5984736372,"x_train":1689.4279279682,"imp":18.0690311939,"elo":1424.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.955519638,"x_train":2289.910634932,"imp":18.0241104932,"elo":1427.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4326735827,"x_train":5.4825798778,"imp":14.0688884717,"elo":1590.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.374483993,"x_train":3.8458762973,"imp":12.68620022,"elo":1671.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4207897504,"x_train":4.0473610858,"imp":20.1595766294,"elo":1362.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.6529810635,"x_train":26.1321442475,"imp":19.1991974493,"elo":1398.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":9.2207743835,"x_train":70.8191350853,"imp":18.2821820827,"elo":1436.8,"imputed":true,"imputed_pct":4.7619047619,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":13.7772167659,"x_train":154.3706019343,"imp":17.6502152507,"elo":1455.6,"imputed":true,"imputed_pct":4.7619047619,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":22.316366756,"x_train":393.1065181502,"imp":17.2251201695,"elo":1466.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":28.8069789863,"x_train":810.6442685856,"imp":16.9708904221,"elo":1473.8,"imputed":true,"imputed_pct":4.7619047619,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":36.1369043899,"x_train":1663.0513945281,"imp":16.8319859952,"elo":1476.6,"imputed":true,"imputed_pct":4.7619047619,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":41.1023939991,"x_train":2512.506403524,"imp":16.766590605,"elo":1477.6,"imputed":true,"imputed_pct":4.7619047619,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":46.3634886593,"x_train":3422.3190042951,"imp":16.7316092766,"elo":1475.3,"imputed":true,"imputed_pct":4.7619047619,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.6409697056,"x_train":460.1168119192,"imp":28.4181504004,"elo":1103.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.3541500481,"x_train":734.7515368462,"imp":28.2874148314,"elo":1108.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1404765981,"x_train":1321.1026922005,"imp":28.0158248609,"elo":1115.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.0682608485,"x_train":1962.2511632163,"imp":27.2907198112,"elo":1129.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.4116860286,"x_train":4372.5625009241,"imp":26.1241726178,"elo":1150.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.581168526,"x_train":8707.0506318291,"imp":25.0535474398,"elo":1169.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.0586264094,"x_train":17339.6234835031,"imp":24.5364252305,"elo":1182.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8750179092,"x_train":25624.7257379004,"imp":24.3192783469,"elo":1188.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8114333153,"x_train":34059.6410419582,"imp":24.1668683111,"elo":1194.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0519986869,"x_train":1.2303610879,"imp":18.6357378636,"elo":1387.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1303337018,"x_train":9.3924084625,"imp":26.5598478142,"elo":1066.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2624076267,"x_train":19.9499332005,"imp":24.2714211661,"elo":1131.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5692050445,"x_train":60.5341760166,"imp":22.5260053181,"elo":1181.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8653863265,"x_train":136.8122971072,"imp":21.8694078242,"elo":1208.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2483118673,"x_train":328.250602853,"imp":21.3740871991,"elo":1232.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6672661515,"x_train":672.0587261642,"imp":21.065050993,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1766444304,"x_train":1298.4117955909,"imp":20.9202509656,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4248878932,"x_train":1997.6434256187,"imp":20.8504804077,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.6309014567,"x_train":2665.4817950505,"imp":20.770209269,"elo":1268.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209029857,"x_train":1.9409389396,"imp":23.3874294021,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4595679196,"x_train":5.5392416815,"imp":21.4082919157,"elo":1239.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6794374605,"x_train":16.764226514,"imp":20.9996572903,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7612793276,"x_train":31.9140091666,"imp":20.8223537215,"elo":1274.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9549360649,"x_train":85.9052711013,"imp":20.6434014804,"elo":1284.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0795695279,"x_train":172.6999358448,"imp":20.6020275639,"elo":1287.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1949903748,"x_train":344.2808053206,"imp":20.5560816099,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.3595882138,"x_train":515.1408947248,"imp":20.5055496641,"elo":1293.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.3974981596,"x_train":693.4907982506,"imp":20.472272682,"elo":1296.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":85.0504523595,"x_train":335.8142132441,"imp":24.743436711,"elo":1140.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":166.0881040231,"x_train":882.5688189042,"imp":22.7405128087,"elo":1194.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":280.1665667632,"x_train":2676.7112743199,"imp":20.6841825925,"elo":1273.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":413.7384149023,"x_train":5942.9139958052,"imp":19.9896126739,"elo":1306.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":528.3767220648,"x_train":14596.1261611529,"imp":19.4339464722,"elo":1338.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":543.2330416203,"x_train":15277.755542326,"imp":19.4176388669,"elo":1339.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9186215876,"x_train":3.7137863239,"imp":30.3870258221,"elo":995.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.0338429173,"x_train":7.2328490388,"imp":27.0959316084,"elo":1054.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1335512608,"x_train":17.7783258681,"imp":22.7473525442,"elo":1190.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.4441868365,"x_train":35.5509910734,"imp":20.9811671609,"elo":1276.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.7785012767,"x_train":88.843717188,"imp":20.487207883,"elo":1303.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2922510872,"x_train":177.5290085781,"imp":20.4116191946,"elo":1305.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.6799279262,"x_train":354.4498168115,"imp":20.4735161266,"elo":1299.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.8648219542,"x_train":530.9324752986,"imp":20.5152582222,"elo":1296.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.9514581229,"x_train":711.9845683972,"imp":20.5346806965,"elo":1293.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_binary/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..d801fd0aa8e5b3813735c47bc46401037e4f8f8b
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_binary/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "TabICLv2 (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "Mitra (default)", "TabDPT-Turbo (default)", "TabPFNv2 (tuned + ensembled)", "LimiX (default)", "TabICL (default)", "TabSwift (default)", "AutoGluon 1.4 (best, 4h)", "BetaTabPFN (default)", "RealMLP (tuned + ensembled)", "TabM (tuned + ensembled)", "TabPFNv2 (default)", "TabDPT (tuned + ensembled)", "TabDPT (tuned)", "TabPFNv2 (tuned)", "LightGBM (tuned + ensembled)", "TabM (tuned)", "TabDPT (default)", "LightGBM (tuned)", "RealMLP (tuned)", "SAP-RPT-OSS (default)", "xRFM (tuned)", "xRFM (tuned + ensembled)", "ModernNCA (tuned)", "ModernNCA (default)", "iLTM (tuned + ensembled)", "CatBoost (default)", "FastaiMLP (tuned + ensembled)", "CatBoost (tuned + ensembled)", "EBM (tuned + ensembled)", "CatBoost (tuned)", "TorchMLP (tuned + ensembled)", "EBM (tuned)", "TabM (default)", "XGBoost (tuned + ensembled)", "EBM (default)", "RealMLP (default)", "TabSTAR (tuned)", "OrionMSP (default)", "TabSTAR (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "FastaiMLP (tuned)", "XGBoost (tuned)", "ChimeraBoost (tuned)", "iLTM (tuned)", "ChimeraBoost (default)", "ModernNCA (tuned + ensembled)", "TorchMLP (tuned)", "TabFlex (default)", "ExtraTrees (tuned + ensembled)", "TabSTAR (default)", "ExtraTrees (tuned)", "iLTM (default)", "Linear (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "XGBoost (default)", "Linear (tuned)", "LightGBM (default)", "TorchMLP (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "RandomForest (tuned + ensembled)", "RandomForest (tuned)", "Linear (default)", "xRFM (default)", "KNN (tuned + ensembled)", "ExtraTrees (default)", "RandomForest (default)", "PerpetualBooster (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7731092572212219, 0.7899159789085388, 0.7759103775024414, 0.7787114977836609, 0.7899159789085388, 0.7997198700904846, 0.8067227005958557, 0.8263305425643921, 0.8095238208770752, 0.8767507076263428, 0.894957959651947, 0.9159663915634155, 0.848739504814148, 0.8963585495948792, 0.8361344337463379, 0.8935574293136597, 0.8879551887512207, 0.7955182194709778, 0.8823529481887817, 0.8767507076263428, 0.848739504814148, 0.8991596698760986, 0.924369752407074, 0.8851540684700012, 0.8991596698760986, 0.8851540684700012, 0.8711484670639038, 0.8795518279075623, 0.9103641510009766, 0.9075630307197571, 0.8739495873451233, 0.9215686321258545, 0.924369752407074, 0.8963585495948792, 0.924369752407074, 0.8907563090324402, 0.924369752407074, 0.9523809552192688, 0.8991596698760986, 0.9215686321258545, 0.9523809552192688, 0.8907563090324402, 0.8935574293136597, 0.7787114977836609, 0.9299719929695129, 0.7815126180648804, 0.9299719929695129, 0.9159663915634155, 0.9551820755004883, 0.9439775943756104, 0.9355742335319519, 0.9495798349380493, 0.9551820755004883, 0.9607843160629272, 0.9103641510009766, 0.9439775943756104, 0.8207283020019531, 0.9327731132507324, 0.9271708726882935, 0.913165271282196, 0.9775910377502441, 0.9691876769065857, 0.924369752407074, 0.9663865566253662, 0.9719887971878052, 0.9803921580314636, 0.9523809552192688, 0.9775910377502441, 0.9747899174690247, 0.9439775943756104, 0.9803921580314636, 0.9775910377502441, 0.9831932783126831, 0.9747899174690247, 0.9859943985939026, 0.9663865566253662, 0.9915966391563416], [0.22689075767993927, NaN, 0.5070028305053711, 0.6078431606292725, 0.6666666865348816, 0.6386554837226868, 0.6582633256912231, 0.686274528503418, 0.7310924530029297, 0.686274528503418, 0.7507002949714661, 0.7478991746902466, 0.8025209903717041, 0.8011204600334167, 0.761904776096344, 0.7352941036224365, 0.8067227005958557, 0.7983193397521973, 0.7296918630599976, 0.8067227005958557, 0.8235294222831726, 0.8165265917778015, 0.8039215803146362, 0.83753502368927, 0.8207283020019531, 0.831932783126831, 0.8291316628456116, 0.8095238208770752, 0.8235294222831726, 0.8599439859390259, 0.8599439859390259, 0.8221288323402405, 0.8879551887512207, 0.8711484670639038, 0.8305321931838989, 0.8599439859390259, 0.843137264251709, 0.8739495873451233, 0.9047619104385376, 0.8543417453765869, 0.8683473467826843, 0.8781512379646301, 0.8445377945899963, 0.8683473467826843, 0.7352941036224365, 0.8795518279075623, 0.7507002949714661, 0.9019607901573181, 0.8613445162773132, 0.8991596698760986, 0.9103641510009766, 0.9173669219017029, 0.8991596698760986, 0.918767511844635, 0.9271708726882935, 0.8823529481887817, 0.9327731132507324, 0.7927170991897583, 0.9271708726882935, 0.9159663915634155, 0.8711484670639038, 0.9215686321258545, 0.9495798349380493, 0.8823529481887817, 0.9467787146568298, 0.9439775943756104, 0.9411764740943909, 0.9439775943756104, 0.9383753538131714, 0.9607843160629272, 0.9271708726882935, 0.9635854363441467, 0.9411764740943909, 0.9635854363441467, 0.9691876769065857, 0.9635854363441467, 0.9551820755004883, 0.994397759437561], [0.21008403599262238, 0.4929971992969513, NaN, 0.5854341983795166, 0.586834728717804, 0.6274510025978088, 0.6078431606292725, 0.6246498823165894, 0.6484593749046326, 0.686274528503418, 0.7422969341278076, 0.7577030658721924, 0.7997198700904846, 0.7871148586273193, 0.7955182194709778, 0.7422969341278076, 0.7927170991897583, 0.8011204600334167, 0.7086834907531738, 0.7969187498092651, 0.7843137383460999, 0.7605041861534119, 0.7899159789085388, 0.8179271817207336, 0.8151260614395142, 0.8151260614395142, 0.8123249411582947, 0.7759103775024414, 0.813725471496582, 0.8515406250953674, 0.8459383845329285, 0.8277310729026794, 0.8655462265014648, 0.848739504814148, 0.843137264251709, 0.8347339034080505, 0.8347339034080505, 0.8459383845329285, 0.8991596698760986, 0.831932783126831, 0.843137264251709, 0.8627451062202454, 0.8403361439704895, 0.8627451062202454, 0.7507002949714661, 0.9075630307197571, 0.7478991746902466, 0.8795518279075623, 0.8711484670639038, 0.8795518279075623, 0.9019607901573181, 0.8879551887512207, 0.8907563090324402, 0.913165271282196, 0.9215686321258545, 0.8739495873451233, 0.894957959651947, 0.7983193397521973, 0.8935574293136597, 0.9075630307197571, 0.8907563090324402, 0.9299719929695129, 0.9271708726882935, 0.9019607901573181, 0.9411764740943909, 0.9523809552192688, 0.9411764740943909, 0.9355742335319519, 0.9327731132507324, 0.9299719929695129, 0.9313725233078003, 0.9607843160629272, 0.9467787146568298, 0.9411764740943909, 0.9467787146568298, 0.9747899174690247, 0.9495798349380493, 0.994397759437561], [0.2240896373987198, 0.3921568691730499, 0.4145658314228058, NaN, 0.5658263564109802, 0.6134454011917114, 0.5910364389419556, 0.6078431606292725, 0.6204481720924377, 0.5406162738800049, 0.6918767690658569, 0.6918767690658569, 0.7577030658721924, 0.7282913327217102, 0.7899159789085388, 0.7030812501907349, 0.7801120281219482, 0.8011204600334167, 0.6834734082221985, 0.7450980544090271, 0.7591036558151245, 0.7422969341278076, 0.8095238208770752, 0.8109243512153625, 0.7310924530029297, 0.8095238208770752, 0.8207283020019531, 0.7591036558151245, 0.7857142686843872, 0.843137264251709, 0.843137264251709, 0.7983193397521973, 0.8627451062202454, 0.8333333134651184, 0.8235294222831726, 0.8403361439704895, 0.8067227005958557, 0.8305321931838989, 0.8795518279075623, 0.8123249411582947, 0.8515406250953674, 0.8627451062202454, 0.8235294222831726, 0.8207283020019531, 0.6890756487846375, 0.8837534785270691, 0.6918767690658569, 0.8823529481887817, 0.8347339034080505, 0.8711484670639038, 0.8907563090324402, 0.8963585495948792, 0.8879551887512207, 0.9271708726882935, 0.9075630307197571, 0.8543417453765869, 0.9075630307197571, 0.756302535533905, 0.8879551887512207, 0.9047619104385376, 0.8459383845329285, 0.924369752407074, 0.9271708726882935, 0.8795518279075623, 0.9355742335319519, 0.9355742335319519, 0.9397758841514587, 0.924369752407074, 0.9299719929695129, 0.9355742335319519, 0.9089635610580444, 0.9299719929695129, 0.913165271282196, 0.9383753538131714, 0.9411764740943909, 0.9551820755004883, 0.9327731132507324, 0.9915966391563416], [0.2212885171175003, 0.3333333432674408, 0.41316527128219604, 0.43417367339134216, NaN, 0.5210084319114685, 0.518207311630249, 0.5602241158485413, 0.6386554837226868, 0.5490196347236633, 0.680672287940979, 0.6722689270973206, 0.6834734082221985, 0.7310924530029297, 0.7394958138465881, 0.7212885022163391, 0.7394958138465881, 0.7478991746902466, 0.7198879718780518, 0.7058823704719543, 0.7072829008102417, 0.756302535533905, 0.7899159789085388, 0.7955182194709778, 0.7170868515968323, 0.8123249411582947, 0.8179271817207336, 0.7815126180648804, 0.8179271817207336, 0.8291316628456116, 0.8123249411582947, 0.8123249411582947, 0.8291316628456116, 0.8711484670639038, 0.8459383845329285, 0.8403361439704895, 0.8403361439704895, 0.8291316628456116, 0.8823529481887817, 0.8655462265014648, 0.8683473467826843, 0.8515406250953674, 0.8347339034080505, 0.8599439859390259, 0.7282913327217102, 0.8935574293136597, 0.732492983341217, 0.8963585495948792, 0.8711484670639038, 0.8795518279075623, 0.9019607901573181, 0.906162440776825, 0.9047619104385376, 0.8739495873451233, 0.924369752407074, 0.8809523582458496, 0.8879551887512207, 0.7703081369400024, 0.8879551887512207, 0.9103641510009766, 0.8599439859390259, 0.9215686321258545, 0.924369752407074, 0.8739495873451233, 0.9579831957817078, 0.9635854363441467, 0.9355742335319519, 0.9271708726882935, 0.913165271282196, 0.9439775943756104, 0.9327731132507324, 0.9551820755004883, 0.913165271282196, 0.9355742335319519, 0.9523809552192688, 0.9635854363441467, 0.9327731132507324, 0.9915966391563416], [0.21008403599262238, 0.3613445460796356, 0.37254902720451355, 0.38655462861061096, 0.4789915978908539, NaN, 0.4789915978908539, 0.5196078419685364, 0.6106442809104919, 0.5518207550048828, 0.6554622054100037, 0.6694678068161011, 0.7114846110343933, 0.6834734082221985, 0.7338935732841492, 0.6834734082221985, 0.7535014152526855, 0.7535014152526855, 0.651260495185852, 0.6946778893470764, 0.680672287940979, 0.761904776096344, 0.7787114977836609, 0.7927170991897583, 0.6694678068161011, 0.8151260614395142, 0.8067227005958557, 0.7731092572212219, 0.7815126180648804, 0.8179271817207336, 0.8291316628456116, 0.8095238208770752, 0.8599439859390259, 0.843137264251709, 0.7759103775024414, 0.8403361439704895, 0.7871148586273193, 0.8235294222831726, 0.8599439859390259, 0.8109243512153625, 0.8305321931838989, 0.843137264251709, 0.7927170991897583, 0.8235294222831726, 0.686274528503418, 0.8179271817207336, 0.7002801299095154, 0.8963585495948792, 0.8207283020019531, 0.8599439859390259, 0.9047619104385376, 0.8879551887512207, 0.9075630307197571, 0.9299719929695129, 0.9047619104385376, 0.8515406250953674, 0.8991596698760986, 0.7422969341278076, 0.8865545988082886, 0.8907563090324402, 0.8459383845329285, 0.918767511844635, 0.9439775943756104, 0.8767507076263428, 0.9495798349380493, 0.9383753538131714, 0.918767511844635, 0.9103641510009766, 0.9327731132507324, 0.9467787146568298, 0.9103641510009766, 0.9439775943756104, 0.924369752407074, 0.9327731132507324, 0.9523809552192688, 0.9523809552192688, 0.9355742335319519, 0.9887955188751221], [0.2002801150083542, 0.34173670411109924, 0.3921568691730499, 0.4089635908603668, 0.48179271817207336, 0.5210084319114685, NaN, 0.5910364389419556, 0.6414566040039062, 0.5154061913490295, 0.6974790096282959, 0.6946778893470764, 0.668067216873169, 0.6694678068161011, 0.7422969341278076, 0.7114846110343933, 0.7366946935653687, 0.7394958138465881, 0.7058823704719543, 0.6694678068161011, 0.680672287940979, 0.767507016658783, 0.7815126180648804, 0.7787114977836609, 0.6246498823165894, 0.8025209903717041, 0.7759103775024414, 0.7577030658721924, 0.7815126180648804, 0.8151260614395142, 0.8347339034080505, 0.7927170991897583, 0.819327712059021, 0.831932783126831, 0.7941176295280457, 0.8151260614395142, 0.7955182194709778, 0.8123249411582947, 0.8543417453765869, 0.8109243512153625, 0.8179271817207336, 0.8347339034080505, 0.8039215803146362, 0.8123249411582947, 0.6890756487846375, 0.8515406250953674, 0.6974790096282959, 0.8669467568397522, 0.843137264251709, 0.8613445162773132, 0.8823529481887817, 0.8739495873451233, 0.8823529481887817, 0.8879551887512207, 0.8963585495948792, 0.8459383845329285, 0.8655462265014648, 0.7422969341278076, 0.8571428656578064, 0.8907563090324402, 0.8599439859390259, 0.924369752407074, 0.9215686321258545, 0.8823529481887817, 0.9383753538131714, 0.918767511844635, 0.9299719929695129, 0.918767511844635, 0.9159663915634155, 0.918767511844635, 0.894957959651947, 0.9299719929695129, 0.913165271282196, 0.9327731132507324, 0.9327731132507324, 0.9523809552192688, 0.9159663915634155, 0.9887955188751221], [0.19327731430530548, 0.3137255012989044, 0.37535014748573303, 0.3921568691730499, 0.43977591395378113, 0.4803921580314636, 0.4089635908603668, NaN, 0.6302521228790283, 0.518207311630249, 0.656862735748291, 0.6470588445663452, 0.6358543634414673, 0.656862735748291, 0.7142857313156128, 0.6974790096282959, 0.6974790096282959, 0.7170868515968323, 0.680672287940979, 0.6610644459724426, 0.6820728182792664, 0.7352941036224365, 0.7647058963775635, 0.7633053064346313, 0.6414566040039062, 0.7955182194709778, 0.761904776096344, 0.7478991746902466, 0.7647058963775635, 0.7829131484031677, 0.7871148586273193, 0.7703081369400024, 0.7955182194709778, 0.8039215803146362, 0.7787114977836609, 0.8011204600334167, 0.8039215803146362, 0.7955182194709778, 0.8515406250953674, 0.8067227005958557, 0.8179271817207336, 0.8207283020019531, 0.7899159789085388, 0.8095238208770752, 0.680672287940979, 0.8459383845329285, 0.6918767690658569, 0.8473389148712158, 0.8235294222831726, 0.831932783126831, 0.8683473467826843, 0.8655462265014648, 0.8711484670639038, 0.8781512379646301, 0.8963585495948792, 0.831932783126831, 0.8613445162773132, 0.7450980544090271, 0.8599439859390259, 0.8851540684700012, 0.8403361439704895, 0.900560200214386, 0.9159663915634155, 0.8599439859390259, 0.9327731132507324, 0.9271708726882935, 0.9215686321258545, 0.9047619104385376, 0.8991596698760986, 0.9103641510009766, 0.906162440776825, 0.9271708726882935, 0.9103641510009766, 0.924369752407074, 0.9299719929695129, 0.9551820755004883, 0.918767511844635, 0.9887955188751221], [0.1736694723367691, 0.2689075767993927, 0.35154062509536743, 0.37955182790756226, 0.3613445460796356, 0.38935574889183044, 0.35854342579841614, 0.36974790692329407, NaN, 0.5098039507865906, 0.5686274766921997, 0.5028011202812195, 0.5840336084365845, 0.6022409200668335, 0.6442577242851257, 0.6386554837226868, 0.6470588445663452, 0.6918767690658569, 0.6260504126548767, 0.6022409200668335, 0.605042040348053, 0.6344537734985352, 0.6834734082221985, 0.7787114977836609, 0.6442577242851257, 0.7394958138465881, 0.7507002949714661, 0.6946778893470764, 0.7296918630599976, 0.7422969341278076, 0.7507002949714661, 0.7535014152526855, 0.7521008253097534, 0.7843137383460999, 0.7815126180648804, 0.7535014152526855, 0.7927170991897583, 0.743697464466095, 0.8263305425643921, 0.8011204600334167, 0.7871148586273193, 0.7969187498092651, 0.7955182194709778, 0.7899159789085388, 0.6834734082221985, 0.7521008253097534, 0.686274528503418, 0.8459383845329285, 0.8207283020019531, 0.8039215803146362, 0.848739504814148, 0.83753502368927, 0.8683473467826843, 0.8179271817207336, 0.8935574293136597, 0.8263305425643921, 0.8543417453765869, 0.7310924530029297, 0.8557422757148743, 0.8879551887512207, 0.8515406250953674, 0.8683473467826843, 0.9159663915634155, 0.8823529481887817, 0.9103641510009766, 0.9299719929695129, 0.8963585495948792, 0.9103641510009766, 0.8935574293136597, 0.9215686321258545, 0.918767511844635, 0.9299719929695129, 0.8935574293136597, 0.9047619104385376, 0.9313725233078003, 0.9551820755004883, 0.9229691624641418, 0.9859943985939026], [0.190476194024086, 0.3137255012989044, 0.3137255012989044, 0.4593837559223175, 0.45098039507865906, 0.4481792747974396, 0.48459383845329285, 0.48179271817207336, 0.4901960790157318, NaN, 0.5378151535987854, 0.5714285969734192, 0.5826330780982971, 0.6904761791229248, 0.6078431606292725, 0.5742297172546387, 0.6442577242851257, 0.6176470518112183, 0.5770308375358582, 0.67507004737854, 0.6834734082221985, 0.593837559223175, 0.6526610851287842, 0.6666666865348816, 0.7885153889656067, 0.6778711676597595, 0.7156862616539001, 0.686274528503418, 0.7310924530029297, 0.7198879718780518, 0.7058823704719543, 0.7394958138465881, 0.7282913327217102, 0.7142857313156128, 0.767507016658783, 0.7366946935653687, 0.7310924530029297, 0.7282913327217102, 0.756302535533905, 0.7521008253097534, 0.7394958138465881, 0.756302535533905, 0.7507002949714661, 0.7703081369400024, 0.6694678068161011, 0.8291316628456116, 0.680672287940979, 0.767507016658783, 0.8123249411582947, 0.767507016658783, 0.7899159789085388, 0.8305321931838989, 0.8095238208770752, 0.761904776096344, 0.8151260614395142, 0.8711484670639038, 0.8697478771209717, 0.7422969341278076, 0.8963585495948792, 0.831932783126831, 0.831932783126831, 0.8627451062202454, 0.8683473467826843, 0.8515406250953674, 0.8823529481887817, 0.9019607901573181, 0.8907563090324402, 0.9159663915634155, 0.913165271282196, 0.9159663915634155, 0.9159663915634155, 0.9355742335319519, 0.918767511844635, 0.9341736435890198, 0.9327731132507324, 0.9579831957817078, 0.9411764740943909, 0.9915966391563416], [0.12324929982423782, 0.24929971992969513, 0.25770309567451477, 0.30812326073646545, 0.3193277418613434, 0.3445378243923187, 0.3025210201740265, 0.343137264251709, 0.4313725531101227, 0.462184876203537, NaN, 0.456582635641098, 0.5322129130363464, 0.5350140333175659, 0.6442577242851257, 0.6036414504051208, 0.6232492923736572, 0.6274510025978088, 0.6064425706863403, 0.5602241158485413, 0.5616246461868286, 0.6414566040039062, 0.6652660965919495, 0.6876750588417053, 0.5826330780982971, 0.686274528503418, 0.7128851413726807, 0.6148459315299988, 0.7268907427787781, 0.7422969341278076, 0.7478991746902466, 0.7240896224975586, 0.7338935732841492, 0.7254902124404907, 0.7268907427787781, 0.7142857313156128, 0.7464985847473145, 0.7128851413726807, 0.767507016658783, 0.7591036558151245, 0.7310924530029297, 0.7661064267158508, 0.7507002949714661, 0.767507016658783, 0.6554622054100037, 0.7170868515968323, 0.6610644459724426, 0.7983193397521973, 0.7521008253097534, 0.7843137383460999, 0.8095238208770752, 0.7955182194709778, 0.831932783126831, 0.8347339034080505, 0.8459383845329285, 0.7759103775024414, 0.8417366743087769, 0.7254902124404907, 0.8403361439704895, 0.843137264251709, 0.8151260614395142, 0.8599439859390259, 0.9019607901573181, 0.83753502368927, 0.918767511844635, 0.8991596698760986, 0.8991596698760986, 0.9019607901573181, 0.9103641510009766, 0.9215686321258545, 0.8627451062202454, 0.9271708726882935, 0.9019607901573181, 0.9411764740943909, 0.9439775943756104, 0.9411764740943909, 0.9075630307197571, 0.9915966391563416], [0.10504201799631119, 0.2521008551120758, 0.24229691922664642, 0.30812326073646545, 0.32773110270500183, 0.3305322229862213, 0.30532214045524597, 0.3529411852359772, 0.4971988797187805, 0.4285714328289032, 0.5434173941612244, NaN, 0.5546218752861023, 0.5266106724739075, 0.6302521228790283, 0.5742297172546387, 0.6442577242851257, 0.6246498823165894, 0.5910364389419556, 0.5476190447807312, 0.5518207550048828, 0.6134454011917114, 0.686274528503418, 0.6400560140609741, 0.5770308375358582, 0.6974790096282959, 0.7044817805290222, 0.7030812501907349, 0.6820728182792664, 0.7254902124404907, 0.7254902124404907, 0.7002801299095154, 0.8165265917778015, 0.7366946935653687, 0.6960784196853638, 0.7422969341278076, 0.7450980544090271, 0.7394958138465881, 0.7787114977836609, 0.761904776096344, 0.7170868515968323, 0.7787114977836609, 0.756302535533905, 0.7394958138465881, 0.6764705777168274, 0.7577030658721924, 0.6722689270973206, 0.8095238208770752, 0.7521008253097534, 0.8011204600334167, 0.831932783126831, 0.8263305425643921, 0.8403361439704895, 0.83753502368927, 0.8123249411582947, 0.7366946935653687, 0.8235294222831726, 0.732492983341217, 0.8123249411582947, 0.8599439859390259, 0.8095238208770752, 0.8739495873451233, 0.8795518279075623, 0.83753502368927, 0.9019607901573181, 0.8599439859390259, 0.8907563090324402, 0.8879551887512207, 0.8935574293136597, 0.9159663915634155, 0.8459383845329285, 0.918767511844635, 0.8739495873451233, 0.924369752407074, 0.9355742335319519, 0.924369752407074, 0.8991596698760986, 0.9859943985939026], [0.08403361588716507, 0.1974789947271347, 0.2002801150083542, 0.24229691922664642, 0.3165266215801239, 0.2885154187679291, 0.33193278312683105, 0.3641456663608551, 0.4159663915634155, 0.41736695170402527, 0.46778711676597595, 0.4453781545162201, NaN, 0.5462185144424438, 0.5546218752861023, 0.5322129130363464, 0.5462185144424438, 0.5462185144424438, 0.5322129130363464, 0.5448179244995117, 0.5532212853431702, 0.5350140333175659, 0.6190476417541504, 0.6106442809104919, 0.586834728717804, 0.6470588445663452, 0.6386554837226868, 0.6554622054100037, 0.6414566040039062, 0.6946778893470764, 0.7058823704719543, 0.6848739385604858, 0.732492983341217, 0.7226890921592712, 0.7058823704719543, 0.6918767690658569, 0.6694678068161011, 0.7030812501907349, 0.7394958138465881, 0.686274528503418, 0.7198879718780518, 0.7478991746902466, 0.6918767690658569, 0.7478991746902466, 0.6526610851287842, 0.7507002949714661, 0.6554622054100037, 0.767507016658783, 0.756302535533905, 0.7647058963775635, 0.7899159789085388, 0.8151260614395142, 0.7955182194709778, 0.8123249411582947, 0.8039215803146362, 0.7759103775024414, 0.8459383845329285, 0.7226890921592712, 0.8305321931838989, 0.8263305425643921, 0.7927170991897583, 0.8711484670639038, 0.8907563090324402, 0.8151260614395142, 0.9089635610580444, 0.8907563090324402, 0.8879551887512207, 0.8865545988082886, 0.8851540684700012, 0.918767511844635, 0.8641456365585327, 0.9215686321258545, 0.8935574293136597, 0.9411764740943909, 0.9369747638702393, 0.9523809552192688, 0.924369752407074, 0.9887955188751221], [0.15126051008701324, 0.19887955486774445, 0.21288515627384186, 0.2717086970806122, 0.2689075767993927, 0.3165266215801239, 0.3305322229862213, 0.343137264251709, 0.3977591097354889, 0.3095238208770752, 0.46498599648475647, 0.4733893573284149, 0.45378151535987854, NaN, 0.5532212853431702, 0.4901960790157318, 0.5770308375358582, 0.5210084319114685, 0.529411792755127, 0.5112044811248779, 0.5350140333175659, 0.5462185144424438, 0.5770308375358582, 0.5630252361297607, 0.5630252361297607, 0.5826330780982971, 0.6414566040039062, 0.6302521228790283, 0.6358543634414673, 0.6666666865348816, 0.6722689270973206, 0.6638655662536621, 0.67507004737854, 0.6442577242851257, 0.7338935732841492, 0.6498599648475647, 0.6694678068161011, 0.6652660965919495, 0.7282913327217102, 0.6834734082221985, 0.6722689270973206, 0.6722689270973206, 0.6778711676597595, 0.7114846110343933, 0.6918767690658569, 0.8095238208770752, 0.7058823704719543, 0.686274528503418, 0.767507016658783, 0.7114846110343933, 0.7254902124404907, 0.7955182194709778, 0.7731092572212219, 0.7338935732841492, 0.7647058963775635, 0.8627451062202454, 0.8263305425643921, 0.7591036558151245, 0.8207283020019531, 0.8011204600334167, 0.8123249411582947, 0.8459383845329285, 0.8179271817207336, 0.8347339034080505, 0.8361344337463379, 0.9075630307197571, 0.8403361439704895, 0.8935574293136597, 0.8543417453765869, 0.8711484670639038, 0.8725489974021912, 0.8907563090324402, 0.8711484670639038, 0.8795518279075623, 0.8935574293136597, 0.9075630307197571, 0.9327731132507324, 0.9915966391563416], [0.10364145785570145, 0.2380952388048172, 0.2044817954301834, 0.21008403599262238, 0.26050421595573425, 0.2661064565181732, 0.25770309567451477, 0.2857142984867096, 0.35574230551719666, 0.3921568691730499, 0.35574230551719666, 0.36974790692329407, 0.4453781545162201, 0.44677871465682983, NaN, 0.5126050710678101, 0.5392156839370728, 0.523809552192688, 0.462184876203537, 0.4901960790157318, 0.4789915978908539, 0.462184876203537, 0.6498599648475647, 0.5910364389419556, 0.5266106724739075, 0.6834734082221985, 0.6470588445663452, 0.5378151535987854, 0.6498599648475647, 0.6372548937797546, 0.605042040348053, 0.6386554837226868, 0.686274528503418, 0.7142857313156128, 0.6834734082221985, 0.7198879718780518, 0.6778711676597595, 0.7282913327217102, 0.756302535533905, 0.6946778893470764, 0.7058823704719543, 0.767507016658783, 0.7226890921592712, 0.7338935732841492, 0.6358543634414673, 0.6288515329360962, 0.6470588445663452, 0.743697464466095, 0.7338935732841492, 0.761904776096344, 0.761904776096344, 0.7843137383460999, 0.7759103775024414, 0.7142857313156128, 0.7955182194709778, 0.7394958138465881, 0.756302535533905, 0.7058823704719543, 0.7647058963775635, 0.8179271817207336, 0.7997198700904846, 0.9047619104385376, 0.8935574293136597, 0.8333333134651184, 0.918767511844635, 0.8627451062202454, 0.9103641510009766, 0.8907563090324402, 0.8711484670639038, 0.8865545988082886, 0.8683473467826843, 0.913165271282196, 0.8963585495948792, 0.924369752407074, 0.9215686321258545, 0.9327731132507324, 0.8991596698760986, 0.994397759437561], [0.16386555135250092, 0.2647058963775635, 0.25770309567451477, 0.2969187796115875, 0.2787114977836609, 0.3165266215801239, 0.2885154187679291, 0.3025210201740265, 0.3613445460796356, 0.4257703125476837, 0.39635854959487915, 0.4257703125476837, 0.46778711676597595, 0.5098039507865906, 0.48739495873451233, NaN, 0.5154061913490295, 0.4789915978908539, 0.4733893573284149, 0.523809552192688, 0.5322129130363464, 0.4453781545162201, 0.6470588445663452, 0.5574229955673218, 0.5882353186607361, 0.6358543634414673, 0.5826330780982971, 0.5770308375358582, 0.599439799785614, 0.6246498823165894, 0.6372548937797546, 0.6330532431602478, 0.6386554837226868, 0.7058823704719543, 0.6260504126548767, 0.6638655662536621, 0.7254902124404907, 0.6960784196853638, 0.6694678068161011, 0.7366946935653687, 0.7338935732841492, 0.7478991746902466, 0.7647058963775635, 0.6918767690658569, 0.6694678068161011, 0.6526610851287842, 0.67507004737854, 0.7408963441848755, 0.7030812501907349, 0.767507016658783, 0.767507016658783, 0.7030812501907349, 0.7843137383460999, 0.686274528503418, 0.7478991746902466, 0.7156862616539001, 0.8179271817207336, 0.7408963441848755, 0.8123249411582947, 0.8599439859390259, 0.8543417453765869, 0.8235294222831726, 0.8347339034080505, 0.8823529481887817, 0.8557422757148743, 0.8263305425643921, 0.843137264251709, 0.8039215803146362, 0.8935574293136597, 0.8851540684700012, 0.9047619104385376, 0.8207283020019531, 0.913165271282196, 0.8683473467826843, 0.9019607901573181, 0.9271708726882935, 0.9383753538131714, 0.9803921580314636], [0.10644257813692093, 0.19327731430530548, 0.2072829157114029, 0.21988795697689056, 0.26050421595573425, 0.24649859964847565, 0.26330533623695374, 0.3025210201740265, 0.3529411852359772, 0.35574230551719666, 0.3767507076263428, 0.35574230551719666, 0.45378151535987854, 0.42296919226646423, 0.46078431606292725, 0.48459383845329285, NaN, 0.5126050710678101, 0.4313725531101227, 0.46078431606292725, 0.4691876769065857, 0.4593837559223175, 0.6274510025978088, 0.5910364389419556, 0.523809552192688, 0.6708683371543884, 0.7254902124404907, 0.5490196347236633, 0.6134454011917114, 0.6554622054100037, 0.6134454011917114, 0.6134454011917114, 0.7142857313156128, 0.7058823704719543, 0.6960784196853638, 0.7394958138465881, 0.7114846110343933, 0.7310924530029297, 0.7030812501907349, 0.7198879718780518, 0.7086834907531738, 0.7731092572212219, 0.7156862616539001, 0.7857142686843872, 0.6246498823165894, 0.6442577242851257, 0.6386554837226868, 0.7703081369400024, 0.7268907427787781, 0.7983193397521973, 0.7857142686843872, 0.7731092572212219, 0.7871148586273193, 0.7058823704719543, 0.756302535533905, 0.7114846110343933, 0.8011204600334167, 0.6918767690658569, 0.7941176295280457, 0.8459383845329285, 0.7871148586273193, 0.8991596698760986, 0.8823529481887817, 0.7927170991897583, 0.8907563090324402, 0.8711484670639038, 0.9103641510009766, 0.8991596698760986, 0.8963585495948792, 0.9215686321258545, 0.8179271817207336, 0.9075630307197571, 0.8851540684700012, 0.924369752407074, 0.9523809552192688, 0.9579831957817078, 0.9075630307197571, 0.9915966391563416], [0.1120448186993599, 0.20168067514896393, 0.19887955486774445, 0.19887955486774445, 0.2521008551120758, 0.24649859964847565, 0.26050421595573425, 0.2829131782054901, 0.30812326073646545, 0.38235294818878174, 0.37254902720451355, 0.37535014748573303, 0.45378151535987854, 0.4789915978908539, 0.4761904776096344, 0.5210084319114685, 0.48739495873451233, NaN, 0.4523809552192688, 0.49439775943756104, 0.4901960790157318, 0.4425770342350006, 0.6330532431602478, 0.66246497631073, 0.523809552192688, 0.66246497631073, 0.6386554837226868, 0.5686274766921997, 0.5742297172546387, 0.6022409200668335, 0.6260504126548767, 0.6330532431602478, 0.66246497631073, 0.7170868515968323, 0.6442577242851257, 0.7002801299095154, 0.7198879718780518, 0.7086834907531738, 0.6946778893470764, 0.7282913327217102, 0.7997198700904846, 0.7591036558151245, 0.7184873819351196, 0.7212885022163391, 0.6386554837226868, 0.6246498823165894, 0.6442577242851257, 0.7787114977836609, 0.6974790096282959, 0.8011204600334167, 0.8263305425643921, 0.7535014152526855, 0.831932783126831, 0.7366946935653687, 0.7689075469970703, 0.7142857313156128, 0.7927170991897583, 0.6974790096282959, 0.7647058963775635, 0.831932783126831, 0.8123249411582947, 0.8683473467826843, 0.8795518279075623, 0.8277310729026794, 0.8837534785270691, 0.8711484670639038, 0.8907563090324402, 0.8683473467826843, 0.8515406250953674, 0.8767507076263428, 0.8571428656578064, 0.8823529481887817, 0.8683473467826843, 0.9047619104385376, 0.9047619104385376, 0.9579831957817078, 0.8879551887512207, 0.9775910377502441], [0.2044817954301834, 0.27030813694000244, 0.29131653904914856, 0.3165266215801239, 0.28011205792427063, 0.34873950481414795, 0.29411765933036804, 0.3193277418613434, 0.3739495873451233, 0.42296919226646423, 0.39355742931365967, 0.4089635908603668, 0.46778711676597595, 0.47058823704719543, 0.5378151535987854, 0.5266106724739075, 0.5686274766921997, 0.5476190447807312, NaN, 0.5322129130363464, 0.5532212853431702, 0.48459383845329285, 0.6274510025978088, 0.5798319578170776, 0.5602241158485413, 0.6386554837226868, 0.6218487620353699, 0.5826330780982971, 0.6442577242851257, 0.6610644459724426, 0.6162465214729309, 0.6400560140609741, 0.6134454011917114, 0.651260495185852, 0.7142857313156128, 0.6610644459724426, 0.6694678068161011, 0.6582633256912231, 0.6778711676597595, 0.6834734082221985, 0.6974790096282959, 0.6722689270973206, 0.6918767690658569, 0.7226890921592712, 0.6778711676597595, 0.686274528503418, 0.6736694574356079, 0.6666666865348816, 0.7226890921592712, 0.6918767690658569, 0.686274528503418, 0.6918767690658569, 0.6778711676597595, 0.6918767690658569, 0.7338935732841492, 0.8263305425643921, 0.761904776096344, 0.7450980544090271, 0.7731092572212219, 0.8011204600334167, 0.8277310729026794, 0.7983193397521973, 0.767507016658783, 0.8347339034080505, 0.8095238208770752, 0.8515406250953674, 0.8039215803146362, 0.8403361439704895, 0.8235294222831726, 0.8515406250953674, 0.8767507076263428, 0.8571428656578064, 0.8627451062202454, 0.848739504814148, 0.8585433959960938, 0.8935574293136597, 0.8823529481887817, 0.9719887971878052], [0.11764705926179886, 0.19327731430530548, 0.20308123528957367, 0.2549019753932953, 0.29411765933036804, 0.30532214045524597, 0.3305322229862213, 0.33893558382987976, 0.3977591097354889, 0.32492998242378235, 0.43977591395378113, 0.4523809552192688, 0.4551820755004883, 0.48879551887512207, 0.5098039507865906, 0.4761904776096344, 0.5392156839370728, 0.505602240562439, 0.46778711676597595, NaN, 0.5854341983795166, 0.5504201650619507, 0.5462185144424438, 0.5560224056243896, 0.5826330780982971, 0.599439799785614, 0.6372548937797546, 0.6246498823165894, 0.6176470518112183, 0.6428571343421936, 0.6330532431602478, 0.6400560140609741, 0.6694678068161011, 0.6358543634414673, 0.6666666865348816, 0.6442577242851257, 0.6372548937797546, 0.6414566040039062, 0.7044817805290222, 0.6652660965919495, 0.6456582546234131, 0.6596638560295105, 0.6498599648475647, 0.67507004737854, 0.6134454011917114, 0.7366946935653687, 0.6162465214729309, 0.6890756487846375, 0.7254902124404907, 0.7030812501907349, 0.7198879718780518, 0.7759103775024414, 0.7450980544090271, 0.761904776096344, 0.7394958138465881, 0.7591036558151245, 0.8039215803146362, 0.7058823704719543, 0.8039215803146362, 0.8235294222831726, 0.756302535533905, 0.8291316628456116, 0.8207283020019531, 0.7815126180648804, 0.848739504814148, 0.8599439859390259, 0.8501400351524353, 0.8851540684700012, 0.8823529481887817, 0.8851540684700012, 0.8291316628456116, 0.9173669219017029, 0.8753501176834106, 0.913165271282196, 0.9103641510009766, 0.9215686321258545, 0.8879551887512207, 0.9915966391563416], [0.12324929982423782, 0.1764705926179886, 0.21568627655506134, 0.24089635908603668, 0.2927170991897583, 0.3193277418613434, 0.3193277418613434, 0.31792718172073364, 0.3949579894542694, 0.3165266215801239, 0.4383753538131714, 0.4481792747974396, 0.44677871465682983, 0.46498599648475647, 0.5210084319114685, 0.46778711676597595, 0.5308123230934143, 0.5098039507865906, 0.44677871465682983, 0.4145658314228058, NaN, 0.5224089622497559, 0.5378151535987854, 0.5546218752861023, 0.5644257664680481, 0.5658263564109802, 0.5882353186607361, 0.5882353186607361, 0.5770308375358582, 0.6106442809104919, 0.605042040348053, 0.6330532431602478, 0.6358543634414673, 0.6246498823165894, 0.6582633256912231, 0.6330532431602478, 0.6554622054100037, 0.6260504126548767, 0.7058823704719543, 0.6666666865348816, 0.6358543634414673, 0.656862735748291, 0.6638655662536621, 0.6834734082221985, 0.5980392098426819, 0.7507002949714661, 0.5966386795043945, 0.67507004737854, 0.7058823704719543, 0.6722689270973206, 0.6946778893470764, 0.7507002949714661, 0.7226890921592712, 0.738095223903656, 0.7254902124404907, 0.7535014152526855, 0.761904776096344, 0.7002801299095154, 0.7703081369400024, 0.7899159789085388, 0.7605041861534119, 0.8095238208770752, 0.7871148586273193, 0.7843137383460999, 0.8039215803146362, 0.848739504814148, 0.8207283020019531, 0.8655462265014648, 0.8403361439704895, 0.8571428656578064, 0.8291316628456116, 0.8795518279075623, 0.8739495873451233, 0.8739495873451233, 0.8879551887512207, 0.918767511844635, 0.8795518279075623, 0.9859943985939026], [0.15126051008701324, 0.1834733933210373, 0.23949579894542694, 0.25770309567451477, 0.24369747936725616, 0.2380952388048172, 0.23249299824237823, 0.2647058963775635, 0.36554622650146484, 0.40616247057914734, 0.35854342579841614, 0.38655462861061096, 0.46498599648475647, 0.45378151535987854, 0.5378151535987854, 0.5546218752861023, 0.5406162738800049, 0.5574229955673218, 0.5154061913490295, 0.4495798349380493, 0.47759103775024414, NaN, 0.605042040348053, 0.6302521228790283, 0.5126050710678101, 0.6218487620353699, 0.6330532431602478, 0.5980392098426819, 0.5826330780982971, 0.6274510025978088, 0.6442577242851257, 0.651260495185852, 0.5882353186607361, 0.6694678068161011, 0.656862735748291, 0.6638655662536621, 0.6554622054100037, 0.680672287940979, 0.686274528503418, 0.6778711676597595, 0.6890756487846375, 0.6960784196853638, 0.67507004737854, 0.6960784196853638, 0.6386554837226868, 0.6218487620353699, 0.6358543634414673, 0.7254902124404907, 0.680672287940979, 0.7030812501907349, 0.7478991746902466, 0.7184873819351196, 0.7731092572212219, 0.7198879718780518, 0.7535014152526855, 0.7310924530029297, 0.7310924530029297, 0.7058823704719543, 0.7731092572212219, 0.8067227005958557, 0.7731092572212219, 0.7815126180648804, 0.7703081369400024, 0.7899159789085388, 0.8179271817207336, 0.8067227005958557, 0.7885153889656067, 0.8095238208770752, 0.7843137383460999, 0.819327712059021, 0.8151260614395142, 0.8459383845329285, 0.83753502368927, 0.843137264251709, 0.831932783126831, 0.8683473467826843, 0.8515406250953674, 0.9803921580314636], [0.10084033757448196, 0.19607843458652496, 0.21008403599262238, 0.190476194024086, 0.21008403599262238, 0.2212885171175003, 0.21848739683628082, 0.23529411852359772, 0.3165266215801239, 0.3473389446735382, 0.33473390340805054, 0.3137255012989044, 0.380952388048172, 0.42296919226646423, 0.3501400649547577, 0.3529411852359772, 0.37254902720451355, 0.3669467866420746, 0.37254902720451355, 0.45378151535987854, 0.462184876203537, 0.3949579894542694, NaN, 0.44397759437561035, 0.5126050710678101, 0.5966386795043945, 0.5098039507865906, 0.5070028305053711, 0.5630252361297607, 0.5574229955673218, 0.5378151535987854, 0.5322129130363464, 0.5896358489990234, 0.6764705777168274, 0.5686274766921997, 0.6638655662536621, 0.6176470518112183, 0.656862735748291, 0.5854341983795166, 0.6498599648475647, 0.6764705777168274, 0.7899159789085388, 0.7058823704719543, 0.6302521228790283, 0.6120448112487793, 0.605042040348053, 0.6162465214729309, 0.6484593749046326, 0.6386554837226868, 0.8151260614395142, 0.6834734082221985, 0.7002801299095154, 0.7843137383460999, 0.6582633256912231, 0.686274528503418, 0.6414566040039062, 0.7689075469970703, 0.7058823704719543, 0.7478991746902466, 0.8207283020019531, 0.7815126180648804, 0.8515406250953674, 0.8851540684700012, 0.7955182194709778, 0.9439775943756104, 0.7731092572212219, 0.8795518279075623, 0.7703081369400024, 0.8851540684700012, 0.9159663915634155, 0.8543417453765869, 0.8235294222831726, 0.8823529481887817, 0.9075630307197571, 0.9411764740943909, 0.9327731132507324, 0.9033613204956055, 0.9887955188751221], [0.07563025504350662, 0.16246499121189117, 0.18207283318042755, 0.18907563388347626, 0.2044817954301834, 0.2072829157114029, 0.2212885171175003, 0.23669467866420746, 0.2212885171175003, 0.3333333432674408, 0.3123249411582947, 0.3599439859390259, 0.38935574889183044, 0.43697479367256165, 0.4089635908603668, 0.4425770342350006, 0.4089635908603668, 0.33753502368927, 0.42016807198524475, 0.44397759437561035, 0.4453781545162201, 0.36974790692329407, 0.5560224056243896, NaN, 0.48879551887512207, 0.5882353186607361, 0.529411792755127, 0.48739495873451233, 0.518207311630249, 0.523809552192688, 0.6106442809104919, 0.593837559223175, 0.5910364389419556, 0.6498599648475647, 0.5896358489990234, 0.6526610851287842, 0.6106442809104919, 0.6526610851287842, 0.6526610851287842, 0.6414566040039062, 0.6834734082221985, 0.6876750588417053, 0.6554622054100037, 0.67507004737854, 0.6162465214729309, 0.5910364389419556, 0.6204481720924377, 0.7394958138465881, 0.6666666865348816, 0.7058823704719543, 0.7366946935653687, 0.67507004737854, 0.7745097875595093, 0.6918767690658569, 0.7128851413726807, 0.6946778893470764, 0.7282913327217102, 0.6722689270973206, 0.7226890921592712, 0.8095238208770752, 0.7843137383460999, 0.8627451062202454, 0.8263305425643921, 0.8011204600334167, 0.8291316628456116, 0.8515406250953674, 0.8767507076263428, 0.8067227005958557, 0.8263305425643921, 0.8207283020019531, 0.831932783126831, 0.831932783126831, 0.848739504814148, 0.8627451062202454, 0.8711484670639038, 0.9467787146568298, 0.8571428656578064, 0.9663865566253662], [0.11484593898057938, 0.17927171289920807, 0.18487395346164703, 0.2689075767993927, 0.2829131782054901, 0.3305322229862213, 0.37535014748573303, 0.35854342579841614, 0.35574230551719666, 0.21148459613323212, 0.41736695170402527, 0.42296919226646423, 0.41316527128219604, 0.43697479367256165, 0.4733893573284149, 0.4117647111415863, 0.4761904776096344, 0.4761904776096344, 0.43977591395378113, 0.41736695170402527, 0.4355742335319519, 0.48739495873451233, 0.48739495873451233, 0.5112044811248779, NaN, 0.5210084319114685, 0.5462185144424438, 0.5546218752861023, 0.5462185144424438, 0.5686274766921997, 0.5826330780982971, 0.6022409200668335, 0.5714285969734192, 0.5714285969734192, 0.5966386795043945, 0.5658263564109802, 0.5546218752861023, 0.5742297172546387, 0.6414566040039062, 0.581232488155365, 0.6246498823165894, 0.6022409200668335, 0.6274510025978088, 0.5896358489990234, 0.5798319578170776, 0.7198879718780518, 0.5854341983795166, 0.6582633256912231, 0.6778711676597595, 0.6274510025978088, 0.6498599648475647, 0.67507004737854, 0.6638655662536621, 0.6526610851287842, 0.6918767690658569, 0.7759103775024414, 0.7731092572212219, 0.67507004737854, 0.7731092572212219, 0.7478991746902466, 0.7422969341278076, 0.7703081369400024, 0.7366946935653687, 0.7591036558151245, 0.7394958138465881, 0.831932783126831, 0.7955182194709778, 0.848739504814148, 0.7927170991897583, 0.8207283020019531, 0.8095238208770752, 0.848739504814148, 0.8851540684700012, 0.8795518279075623, 0.8823529481887817, 0.8739495873451233, 0.9103641510009766, 0.9887955188751221], [0.10084033757448196, 0.16806723177433014, 0.18487395346164703, 0.190476194024086, 0.18767507374286652, 0.18487395346164703, 0.1974789947271347, 0.2044817954301834, 0.26050421595573425, 0.32212886214256287, 0.3137255012989044, 0.3025210201740265, 0.3529411852359772, 0.41736695170402527, 0.3165266215801239, 0.3641456663608551, 0.3291316628456116, 0.33753502368927, 0.3613445460796356, 0.4005602300167084, 0.43417367339134216, 0.3781512677669525, 0.40336135029792786, 0.4117647111415863, 0.4789915978908539, NaN, 0.4901960790157318, 0.48179271817207336, 0.518207311630249, 0.5252100825309753, 0.48459383845329285, 0.5098039507865906, 0.5350140333175659, 0.6344537734985352, 0.5336134433746338, 0.6176470518112183, 0.5966386795043945, 0.6078431606292725, 0.5630252361297607, 0.6218487620353699, 0.6330532431602478, 0.6904761791229248, 0.6526610851287842, 0.6176470518112183, 0.5882353186607361, 0.5826330780982971, 0.593837559223175, 0.5910364389419556, 0.5882353186607361, 0.7422969341278076, 0.6190476417541504, 0.6638655662536621, 0.7338935732841492, 0.6246498823165894, 0.6652660965919495, 0.6246498823165894, 0.743697464466095, 0.6834734082221985, 0.732492983341217, 0.7969187498092651, 0.767507016658783, 0.8305321931838989, 0.8417366743087769, 0.7857142686843872, 0.9033613204956055, 0.761904776096344, 0.8515406250953674, 0.7591036558151245, 0.83753502368927, 0.8739495873451233, 0.8347339034080505, 0.8081232309341431, 0.8543417453765869, 0.8879551887512207, 0.9103641510009766, 0.9299719929695129, 0.8879551887512207, 0.9915966391563416], [0.11484593898057938, 0.17086835205554962, 0.18767507374286652, 0.17927171289920807, 0.18207283318042755, 0.19327731430530548, 0.2240896373987198, 0.2380952388048172, 0.24929971992969513, 0.28431373834609985, 0.28711485862731934, 0.2955182194709778, 0.3613445460796356, 0.35854342579841614, 0.3529411852359772, 0.41736695170402527, 0.27450981736183167, 0.3613445460796356, 0.3781512677669525, 0.36274510622024536, 0.4117647111415863, 0.3669467866420746, 0.4901960790157318, 0.47058823704719543, 0.45378151535987854, 0.5098039507865906, NaN, 0.4733893573284149, 0.505602240562439, 0.4957983195781708, 0.4957983195781708, 0.5434173941612244, 0.5196078419685364, 0.5854341983795166, 0.5798319578170776, 0.5574229955673218, 0.5910364389419556, 0.5658263564109802, 0.605042040348053, 0.6302521228790283, 0.6162465214729309, 0.6246498823165894, 0.5854341983795166, 0.6498599648475647, 0.6022409200668335, 0.5560224056243896, 0.6162465214729309, 0.6134454011917114, 0.6526610851287842, 0.66246497631073, 0.6582633256912231, 0.6442577242851257, 0.6890756487846375, 0.6106442809104919, 0.6834734082221985, 0.6876750588417053, 0.7086834907531738, 0.67507004737854, 0.7086834907531738, 0.7871148586273193, 0.7058823704719543, 0.8207283020019531, 0.7815126180648804, 0.7156862616539001, 0.8039215803146362, 0.8263305425643921, 0.8291316628456116, 0.8585433959960938, 0.7955182194709778, 0.8207283020019531, 0.7731092572212219, 0.843137264251709, 0.8179271817207336, 0.8179271817207336, 0.8571428656578064, 0.9355742335319519, 0.843137264251709, 0.9747899174690247], [0.12885154783725739, 0.190476194024086, 0.2240896373987198, 0.24089635908603668, 0.21848739683628082, 0.22689075767993927, 0.24229691922664642, 0.2521008551120758, 0.30532214045524597, 0.3137255012989044, 0.3851540684700012, 0.2969187796115875, 0.3445378243923187, 0.36974790692329407, 0.462184876203537, 0.42296919226646423, 0.45098039507865906, 0.4313725531101227, 0.41736695170402527, 0.37535014748573303, 0.4117647111415863, 0.4019607901573181, 0.4929971992969513, 0.5126050710678101, 0.4453781545162201, 0.518207311630249, 0.5266106724739075, NaN, 0.5462185144424438, 0.5658263564109802, 0.5378151535987854, 0.5574229955673218, 0.5462185144424438, 0.5210084319114685, 0.6162465214729309, 0.5042017102241516, 0.5518207550048828, 0.518207311630249, 0.6386554837226868, 0.5826330780982971, 0.5714285969734192, 0.5910364389419556, 0.5896358489990234, 0.6554622054100037, 0.5490196347236633, 0.5798319578170776, 0.5434173941612244, 0.5882353186607361, 0.6554622054100037, 0.6078431606292725, 0.6162465214729309, 0.6120448112487793, 0.6414566040039062, 0.6162465214729309, 0.6890756487846375, 0.6974790096282959, 0.67507004737854, 0.6414566040039062, 0.67507004737854, 0.7198879718780518, 0.7310924530029297, 0.7647058963775635, 0.7254902124404907, 0.7366946935653687, 0.7282913327217102, 0.767507016658783, 0.7745097875595093, 0.7787114977836609, 0.7689075469970703, 0.7773109078407288, 0.7703081369400024, 0.8347339034080505, 0.7787114977836609, 0.8207283020019531, 0.8291316628456116, 0.8179271817207336, 0.8515406250953674, 0.9747899174690247], [0.12044817954301834, 0.1764705926179886, 0.18627451360225677, 0.2142857164144516, 0.18207283318042755, 0.21848739683628082, 0.21848739683628082, 0.23529411852359772, 0.27030813694000244, 0.2689075767993927, 0.2731092572212219, 0.31792718172073364, 0.35854342579841614, 0.3641456663608551, 0.3501400649547577, 0.4005602300167084, 0.38655462861061096, 0.4257703125476837, 0.35574230551719666, 0.38235294818878174, 0.42296919226646423, 0.41736695170402527, 0.43697479367256165, 0.48179271817207336, 0.45378151535987854, 0.48179271817207336, 0.49439775943756104, 0.45378151535987854, NaN, 0.4355742335319519, 0.45378151535987854, 0.5378151535987854, 0.4803921580314636, 0.5350140333175659, 0.5420168042182922, 0.5406162738800049, 0.5728291273117065, 0.5546218752861023, 0.5574229955673218, 0.605042040348053, 0.5686274766921997, 0.5518207550048828, 0.6078431606292725, 0.5854341983795166, 0.5798319578170776, 0.5658263564109802, 0.5826330780982971, 0.6148459315299988, 0.6358543634414673, 0.5840336084365845, 0.6358543634414673, 0.6330532431602478, 0.6274510025978088, 0.5686274766921997, 0.6386554837226868, 0.7226890921592712, 0.7170868515968323, 0.6554622054100037, 0.7310924530029297, 0.7703081369400024, 0.7591036558151245, 0.738095223903656, 0.7002801299095154, 0.7871148586273193, 0.7450980544090271, 0.8291316628456116, 0.7941176295280457, 0.8347339034080505, 0.7647058963775635, 0.8039215803146362, 0.8151260614395142, 0.8767507076263428, 0.8459383845329285, 0.8459383845329285, 0.8599439859390259, 0.8963585495948792, 0.8683473467826843, 0.9859943985939026], [0.08963585644960403, 0.14005602896213531, 0.14845938980579376, 0.1568627506494522, 0.17086835205554962, 0.18207283318042755, 0.18487395346164703, 0.21708683669567108, 0.25770309567451477, 0.28011205792427063, 0.25770309567451477, 0.27450981736183167, 0.30532214045524597, 0.3333333432674408, 0.36274510622024536, 0.37535014748573303, 0.3445378243923187, 0.3977591097354889, 0.33893558382987976, 0.3571428656578064, 0.38935574889183044, 0.37254902720451355, 0.4425770342350006, 0.4761904776096344, 0.4313725531101227, 0.47478991746902466, 0.5042017102241516, 0.43417367339134216, 0.5644257664680481, NaN, 0.4663865566253662, 0.5042017102241516, 0.49439775943756104, 0.5532212853431702, 0.5532212853431702, 0.5378151535987854, 0.5630252361297607, 0.5448179244995117, 0.5266106724739075, 0.6022409200668335, 0.5686274766921997, 0.5672268867492676, 0.5910364389419556, 0.593837559223175, 0.5448179244995117, 0.5742297172546387, 0.5434173941612244, 0.599439799785614, 0.6302521228790283, 0.5882353186607361, 0.6162465214729309, 0.6092436909675598, 0.6106442809104919, 0.6302521228790283, 0.6190476417541504, 0.6974790096282959, 0.6918767690658569, 0.6442577242851257, 0.7198879718780518, 0.7338935732841492, 0.7226890921592712, 0.7759103775024414, 0.767507016658783, 0.7422969341278076, 0.7871148586273193, 0.8123249411582947, 0.8305321931838989, 0.848739504814148, 0.8235294222831726, 0.8459383845329285, 0.7829131484031677, 0.918767511844635, 0.8501400351524353, 0.9075630307197571, 0.924369752407074, 0.918767511844635, 0.8627451062202454, 0.9859943985939026], [0.09243697673082352, 0.14005602896213531, 0.15406163036823273, 0.1568627506494522, 0.18767507374286652, 0.17086835205554962, 0.16526611149311066, 0.21288515627384186, 0.24929971992969513, 0.29411765933036804, 0.2521008551120758, 0.27450981736183167, 0.29411765933036804, 0.32773110270500183, 0.3949579894542694, 0.36274510622024536, 0.38655462861061096, 0.3739495873451233, 0.3837535083293915, 0.3669467866420746, 0.3949579894542694, 0.35574230551719666, 0.462184876203537, 0.38935574889183044, 0.41736695170402527, 0.5154061913490295, 0.5042017102241516, 0.462184876203537, 0.5462185144424438, 0.5336134433746338, NaN, 0.5630252361297607, 0.518207311630249, 0.5350140333175659, 0.5672268867492676, 0.5350140333175659, 0.5462185144424438, 0.5518207550048828, 0.5686274766921997, 0.5770308375358582, 0.5350140333175659, 0.5854341983795166, 0.5826330780982971, 0.6162465214729309, 0.5658263564109802, 0.5378151535987854, 0.5742297172546387, 0.6190476417541504, 0.6190476417541504, 0.6036414504051208, 0.6232492923736572, 0.6330532431602478, 0.6582633256912231, 0.6134454011917114, 0.6974790096282959, 0.67507004737854, 0.6526610851287842, 0.6666666865348816, 0.6610644459724426, 0.7478991746902466, 0.686274528503418, 0.7450980544090271, 0.7366946935653687, 0.7086834907531738, 0.7478991746902466, 0.7478991746902466, 0.7759103775024414, 0.7717086672782898, 0.761904776096344, 0.7759103775024414, 0.7507002949714661, 0.8543417453765869, 0.7787114977836609, 0.8235294222831726, 0.8179271817207336, 0.8571428656578064, 0.8235294222831726, 0.9719887971878052], [0.1260504275560379, 0.17787115275859833, 0.17226891219615936, 0.20168067514896393, 0.18767507374286652, 0.190476194024086, 0.2072829157114029, 0.22969187796115875, 0.24649859964847565, 0.26050421595573425, 0.2759103775024414, 0.299719899892807, 0.31512606143951416, 0.3361344635486603, 0.3613445460796356, 0.3669467866420746, 0.38655462861061096, 0.3669467866420746, 0.3599439859390259, 0.3599439859390259, 0.3669467866420746, 0.34873950481414795, 0.46778711676597595, 0.40616247057914734, 0.3977591097354889, 0.4901960790157318, 0.456582635641098, 0.4425770342350006, 0.462184876203537, 0.4957983195781708, 0.43697479367256165, NaN, 0.48179271817207336, 0.5546218752861023, 0.5098039507865906, 0.5518207550048828, 0.5322129130363464, 0.5630252361297607, 0.5266106724739075, 0.5714285969734192, 0.5406162738800049, 0.5770308375358582, 0.5910364389419556, 0.5546218752861023, 0.5966386795043945, 0.5854341983795166, 0.605042040348053, 0.6134454011917114, 0.5686274766921997, 0.6246498823165894, 0.6330532431602478, 0.5924369692802429, 0.6386554837226868, 0.5350140333175659, 0.6162465214729309, 0.6694678068161011, 0.680672287940979, 0.680672287940979, 0.6834734082221985, 0.761904776096344, 0.7338935732841492, 0.7310924530029297, 0.686274528503418, 0.7521008253097534, 0.7198879718780518, 0.8067227005958557, 0.7647058963775635, 0.7899159789085388, 0.7591036558151245, 0.7843137383460999, 0.8095238208770752, 0.7591036558151245, 0.8599439859390259, 0.8109243512153625, 0.8011204600334167, 0.8403361439704895, 0.8711484670639038, 0.9775910377502441], [0.0784313753247261, 0.1120448186993599, 0.13445378839969635, 0.13725490868091583, 0.17086835205554962, 0.14005602896213531, 0.1806722730398178, 0.2044817954301834, 0.2478991597890854, 0.2717086970806122, 0.2661064565181732, 0.1834733933210373, 0.26750701665878296, 0.32492998242378235, 0.3137255012989044, 0.3613445460796356, 0.2857142984867096, 0.33753502368927, 0.38655462861061096, 0.3305322229862213, 0.3641456663608551, 0.4117647111415863, 0.41036415100097656, 0.4089635908603668, 0.4285714328289032, 0.46498599648475647, 0.4803921580314636, 0.45378151535987854, 0.5196078419685364, 0.505602240562439, 0.48179271817207336, 0.518207311630249, NaN, 0.5042017102241516, 0.5546218752861023, 0.48179271817207336, 0.529411792755127, 0.5154061913490295, 0.5602241158485413, 0.5686274766921997, 0.529411792755127, 0.5588235259056091, 0.5714285969734192, 0.599439799785614, 0.5518207550048828, 0.5448179244995117, 0.5560224056243896, 0.593837559223175, 0.6260504126548767, 0.6078431606292725, 0.593837559223175, 0.7184873819351196, 0.6428571343421936, 0.6134454011917114, 0.6498599648475647, 0.67507004737854, 0.6792716979980469, 0.6554622054100037, 0.6918767690658569, 0.7422969341278076, 0.6582633256912231, 0.8039215803146362, 0.7955182194709778, 0.686274528503418, 0.8459383845329285, 0.7759103775024414, 0.831932783126831, 0.7731092572212219, 0.8095238208770752, 0.831932783126831, 0.7310924530029297, 0.8907563090324402, 0.8095238208770752, 0.8543417453765869, 0.8851540684700012, 0.8879551887512207, 0.8529411554336548, 0.9691876769065857], [0.07563025504350662, 0.12885154783725739, 0.15126051008701324, 0.1666666716337204, 0.12885154783725739, 0.1568627506494522, 0.16806723177433014, 0.19607843458652496, 0.21568627655506134, 0.2857142984867096, 0.27450981736183167, 0.26330533623695374, 0.27731093764305115, 0.35574230551719666, 0.2857142984867096, 0.29411765933036804, 0.29411765933036804, 0.2829131782054901, 0.34873950481414795, 0.3641456663608551, 0.37535014748573303, 0.3305322229862213, 0.3235294222831726, 0.3501400649547577, 0.4285714328289032, 0.36554622650146484, 0.4145658314228058, 0.4789915978908539, 0.46498599648475647, 0.44677871465682983, 0.46498599648475647, 0.4453781545162201, 0.4957983195781708, NaN, 0.5070028305053711, 0.5014005899429321, 0.518207311630249, 0.4859943985939026, 0.5546218752861023, 0.5602241158485413, 0.5854341983795166, 0.5644257664680481, 0.593837559223175, 0.5798319578170776, 0.5980392098426819, 0.5126050710678101, 0.605042040348053, 0.5588235259056091, 0.5742297172546387, 0.586834728717804, 0.605042040348053, 0.6498599648475647, 0.6442577242851257, 0.5854341983795166, 0.6442577242851257, 0.5826330780982971, 0.7282913327217102, 0.6666666865348816, 0.7142857313156128, 0.8067227005958557, 0.7030812501907349, 0.7394958138465881, 0.8039215803146362, 0.7254902124404907, 0.8347339034080505, 0.7591036558151245, 0.7871148586273193, 0.7507002949714661, 0.8543417453765869, 0.8627451062202454, 0.7843137383460999, 0.7927170991897583, 0.8627451062202454, 0.8515406250953674, 0.8795518279075623, 0.8683473467826843, 0.8935574293136597, 0.9831932783126831], [0.10364145785570145, 0.16946779191493988, 0.1568627506494522, 0.1764705926179886, 0.15406163036823273, 0.2240896373987198, 0.20588235557079315, 0.2212885171175003, 0.21848739683628082, 0.23249299824237823, 0.2731092572212219, 0.30392158031463623, 0.29411765933036804, 0.2661064565181732, 0.3165266215801239, 0.3739495873451233, 0.30392158031463623, 0.35574230551719666, 0.2857142984867096, 0.3333333432674408, 0.34173670411109924, 0.343137264251709, 0.4313725531101227, 0.41036415100097656, 0.40336135029792786, 0.4663865566253662, 0.42016807198524475, 0.3837535083293915, 0.45798319578170776, 0.44677871465682983, 0.4327731132507324, 0.4901960790157318, 0.4453781545162201, 0.4929971992969513, NaN, 0.5014005899429321, 0.5364145636558533, 0.5266106724739075, 0.4901960790157318, 0.5714285969734192, 0.5210084319114685, 0.5406162738800049, 0.5742297172546387, 0.5728291273117065, 0.5854341983795166, 0.5322129130363464, 0.593837559223175, 0.5770308375358582, 0.6554622054100037, 0.5770308375358582, 0.5966386795043945, 0.5686274766921997, 0.5854341983795166, 0.5798319578170776, 0.605042040348053, 0.686274528503418, 0.6610644459724426, 0.6582633256912231, 0.6526610851287842, 0.7226890921592712, 0.7254902124404907, 0.7703081369400024, 0.7086834907531738, 0.7507002949714661, 0.7100840210914612, 0.8165265917778015, 0.7983193397521973, 0.9299719929695129, 0.7829131484031677, 0.7899159789085388, 0.7703081369400024, 0.8235294222831726, 0.8571428656578064, 0.8179271817207336, 0.83753502368927, 0.924369752407074, 0.8739495873451233, 0.9691876769065857], [0.07563025504350662, 0.14005602896213531, 0.16526611149311066, 0.1596638709306717, 0.1596638709306717, 0.1596638709306717, 0.18487395346164703, 0.19887955486774445, 0.24649859964847565, 0.26330533623695374, 0.2857142984867096, 0.25770309567451477, 0.30812326073646545, 0.3501400649547577, 0.28011205792427063, 0.3361344635486603, 0.26050421595573425, 0.299719899892807, 0.33893558382987976, 0.35574230551719666, 0.3669467866420746, 0.3361344635486603, 0.3361344635486603, 0.3473389446735382, 0.43417367339134216, 0.38235294818878174, 0.4425770342350006, 0.4957983195781708, 0.4593837559223175, 0.462184876203537, 0.46498599648475647, 0.4481792747974396, 0.518207311630249, 0.49859943985939026, 0.49859943985939026, NaN, 0.4789915978908539, 0.5266106724739075, 0.5462185144424438, 0.5224089622497559, 0.5742297172546387, 0.5392156839370728, 0.5490196347236633, 0.5462185144424438, 0.5826330780982971, 0.5070028305053711, 0.5770308375358582, 0.5630252361297607, 0.5546218752861023, 0.586834728717804, 0.5882353186607361, 0.6470588445663452, 0.6246498823165894, 0.5910364389419556, 0.6330532431602478, 0.5630252361297607, 0.7198879718780518, 0.6526610851287842, 0.7156862616539001, 0.7577030658721924, 0.6974790096282959, 0.7535014152526855, 0.8151260614395142, 0.7198879718780518, 0.843137264251709, 0.7366946935653687, 0.7815126180648804, 0.7366946935653687, 0.8459383845329285, 0.8725489974021912, 0.761904776096344, 0.7899159789085388, 0.8627451062202454, 0.8683473467826843, 0.913165271282196, 0.8599439859390259, 0.8739495873451233, 0.9887955188751221], [0.10924369841814041, 0.1568627506494522, 0.16526611149311066, 0.19327731430530548, 0.1596638709306717, 0.21288515627384186, 0.2044817954301834, 0.19607843458652496, 0.2072829157114029, 0.2689075767993927, 0.25350141525268555, 0.2549019753932953, 0.3305322229862213, 0.3305322229862213, 0.32212886214256287, 0.27450981736183167, 0.2885154187679291, 0.28011205792427063, 0.3305322229862213, 0.36274510622024536, 0.3445378243923187, 0.3445378243923187, 0.38235294818878174, 0.38935574889183044, 0.4453781545162201, 0.40336135029792786, 0.4089635908603668, 0.4481792747974396, 0.42717087268829346, 0.43697479367256165, 0.45378151535987854, 0.46778711676597595, 0.47058823704719543, 0.48179271817207336, 0.46358543634414673, 0.5210084319114685, NaN, 0.5070028305053711, 0.4957983195781708, 0.6470588445663452, 0.5854341983795166, 0.5854341983795166, 0.6414566040039062, 0.529411792755127, 0.5686274766921997, 0.5658263564109802, 0.5686274766921997, 0.5266106724739075, 0.5196078419685364, 0.5854341983795166, 0.5602241158485413, 0.5602241158485413, 0.605042040348053, 0.5462185144424438, 0.581232488155365, 0.6330532431602478, 0.7100840210914612, 0.6666666865348816, 0.6918767690658569, 0.7507002949714661, 0.7422969341278076, 0.7310924530029297, 0.7521008253097534, 0.7535014152526855, 0.7647058963775635, 0.6694678068161011, 0.7577030658721924, 0.7535014152526855, 0.7899159789085388, 0.7983193397521973, 0.7927170991897583, 0.761904776096344, 0.8627451062202454, 0.8459383845329285, 0.8543417453765869, 0.8823529481887817, 0.8753501176834106, 0.9719887971878052], [0.07563025504350662, 0.1260504275560379, 0.15406163036823273, 0.16946779191493988, 0.17086835205554962, 0.1764705926179886, 0.18767507374286652, 0.2044817954301834, 0.25630253553390503, 0.2717086970806122, 0.28711485862731934, 0.26050421595573425, 0.2969187796115875, 0.33473390340805054, 0.2717086970806122, 0.30392158031463623, 0.2689075767993927, 0.29131653904914856, 0.34173670411109924, 0.35854342579841614, 0.3739495873451233, 0.3193277418613434, 0.343137264251709, 0.3473389446735382, 0.4257703125476837, 0.3921568691730499, 0.43417367339134216, 0.48179271817207336, 0.4453781545162201, 0.4551820755004883, 0.4481792747974396, 0.43697479367256165, 0.48459383845329285, 0.5140056014060974, 0.4733893573284149, 0.4733893573284149, 0.4929971992969513, NaN, 0.5406162738800049, 0.523809552192688, 0.5532212853431702, 0.5630252361297607, 0.5602241158485413, 0.5406162738800049, 0.5882353186607361, 0.5042017102241516, 0.593837559223175, 0.5252100825309753, 0.5588235259056091, 0.593837559223175, 0.5644257664680481, 0.6106442809104919, 0.6134454011917114, 0.5910364389419556, 0.6232492923736572, 0.5518207550048828, 0.7002801299095154, 0.6610644459724426, 0.686274528503418, 0.7338935732841492, 0.6918767690658569, 0.7366946935653687, 0.831932783126831, 0.7072829008102417, 0.8235294222831726, 0.7170868515968323, 0.7703081369400024, 0.7254902124404907, 0.8459383845329285, 0.848739504814148, 0.7507002949714661, 0.7815126180648804, 0.8571428656578064, 0.8599439859390259, 0.9019607901573181, 0.8711484670639038, 0.8823529481887817, 0.9887955188751221], [0.0476190485060215, 0.095238097012043, 0.10084033757448196, 0.12044817954301834, 0.11764705926179886, 0.14005602896213531, 0.14565826952457428, 0.14845938980579376, 0.1736694723367691, 0.24369747936725616, 0.23249299824237823, 0.2212885171175003, 0.26050421595573425, 0.2717086970806122, 0.24369747936725616, 0.3305322229862213, 0.2969187796115875, 0.30532214045524597, 0.32212886214256287, 0.2955182194709778, 0.29411765933036804, 0.3137255012989044, 0.4145658314228058, 0.3473389446735382, 0.35854342579841614, 0.43697479367256165, 0.3949579894542694, 0.3613445460796356, 0.4425770342350006, 0.4733893573284149, 0.4313725531101227, 0.4733893573284149, 0.43977591395378113, 0.4453781545162201, 0.5098039507865906, 0.45378151535987854, 0.5042017102241516, 0.4593837559223175, NaN, 0.5322129130363464, 0.46778711676597595, 0.505602240562439, 0.5434173941612244, 0.5462185144424438, 0.5518207550048828, 0.48459383845329285, 0.5616246461868286, 0.5154061913490295, 0.6064425706863403, 0.529411792755127, 0.5602241158485413, 0.5560224056243896, 0.5742297172546387, 0.5714285969734192, 0.7072829008102417, 0.6330532431602478, 0.6470588445663452, 0.6582633256912231, 0.6386554837226868, 0.7240896224975586, 0.6904761791229248, 0.7507002949714661, 0.7254902124404907, 0.7282913327217102, 0.6974790096282959, 0.83753502368927, 0.7955182194709778, 0.7997198700904846, 0.7450980544090271, 0.767507016658783, 0.7815126180648804, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.7787114977836609, 0.8851540684700012, 0.8459383845329285, 0.9663865566253662], [0.10084033757448196, 0.14565826952457428, 0.16806723177433014, 0.18767507374286652, 0.13445378839969635, 0.18907563388347626, 0.18907563388347626, 0.19327731430530548, 0.19887955486774445, 0.2478991597890854, 0.24089635908603668, 0.2380952388048172, 0.3137255012989044, 0.3165266215801239, 0.30532214045524597, 0.26330533623695374, 0.28011205792427063, 0.2717086970806122, 0.3165266215801239, 0.33473390340805054, 0.3333333432674408, 0.32212886214256287, 0.3501400649547577, 0.35854342579841614, 0.418767511844635, 0.3781512677669525, 0.36974790692329407, 0.41736695170402527, 0.3949579894542694, 0.3977591097354889, 0.42296919226646423, 0.4285714328289032, 0.4313725531101227, 0.43977591395378113, 0.4285714328289032, 0.47759103775024414, 0.3529411852359772, 0.4761904776096344, 0.46778711676597595, NaN, 0.5266106724739075, 0.5070028305053711, 0.5924369692802429, 0.462184876203537, 0.5518207550048828, 0.5280112028121948, 0.5588235259056091, 0.4901960790157318, 0.5, 0.5350140333175659, 0.5252100825309753, 0.5266106724739075, 0.5630252361297607, 0.5140056014060974, 0.5574229955673218, 0.5952380895614624, 0.6638655662536621, 0.6414566040039062, 0.6946778893470764, 0.7058823704719543, 0.7142857313156128, 0.7058823704719543, 0.7282913327217102, 0.7282913327217102, 0.7450980544090271, 0.6498599648475647, 0.7310924530029297, 0.7142857313156128, 0.7591036558151245, 0.7871148586273193, 0.756302535533905, 0.7450980544090271, 0.8459383845329285, 0.848739504814148, 0.843137264251709, 0.8627451062202454, 0.8767507076263428, 0.9719887971878052], [0.0784313753247261, 0.13165266811847687, 0.1568627506494522, 0.14845938980579376, 0.13165266811847687, 0.16946779191493988, 0.18207283318042755, 0.18207283318042755, 0.21288515627384186, 0.26050421595573425, 0.2689075767993927, 0.2829131782054901, 0.28011205792427063, 0.32773110270500183, 0.29411765933036804, 0.2661064565181732, 0.29131653904914856, 0.2002801150083542, 0.3025210201740265, 0.3543417453765869, 0.3641456663608551, 0.31092438101768494, 0.3235294222831726, 0.3165266215801239, 0.37535014748573303, 0.3669467866420746, 0.3837535083293915, 0.4285714328289032, 0.4313725531101227, 0.4313725531101227, 0.46498599648475647, 0.4593837559223175, 0.47058823704719543, 0.4145658314228058, 0.4789915978908539, 0.4257703125476837, 0.4145658314228058, 0.44677871465682983, 0.5322129130363464, 0.4733893573284149, NaN, 0.46498599648475647, 0.45798319578170776, 0.5154061913490295, 0.5966386795043945, 0.5126050710678101, 0.605042040348053, 0.48459383845329285, 0.5574229955673218, 0.5098039507865906, 0.5154061913490295, 0.5644257664680481, 0.5602241158485413, 0.5714285969734192, 0.5966386795043945, 0.5686274766921997, 0.6722689270973206, 0.6834734082221985, 0.680672287940979, 0.6834734082221985, 0.7030812501907349, 0.6820728182792664, 0.6498599648475647, 0.7240896224975586, 0.6582633256912231, 0.7282913327217102, 0.6946778893470764, 0.7142857313156128, 0.7871148586273193, 0.7843137383460999, 0.7591036558151245, 0.7254902124404907, 0.8235294222831726, 0.8123249411582947, 0.83753502368927, 0.831932783126831, 0.8655462265014648, 0.9607843160629272], [0.0476190485060215, 0.12184873968362808, 0.13725490868091583, 0.13725490868091583, 0.14845938980579376, 0.1568627506494522, 0.16526611149311066, 0.17927171289920807, 0.20308123528957367, 0.24369747936725616, 0.23389355838298798, 0.2212885171175003, 0.2521008551120758, 0.32773110270500183, 0.23249299824237823, 0.2521008551120758, 0.22689075767993927, 0.24089635908603668, 0.32773110270500183, 0.3403361439704895, 0.343137264251709, 0.30392158031463623, 0.21008403599262238, 0.3123249411582947, 0.3977591097354889, 0.3095238208770752, 0.37535014748573303, 0.4089635908603668, 0.4481792747974396, 0.4327731132507324, 0.4145658314228058, 0.42296919226646423, 0.44117647409439087, 0.4355742335319519, 0.4593837559223175, 0.46078431606292725, 0.4145658314228058, 0.43697479367256165, 0.49439775943756104, 0.4929971992969513, 0.5350140333175659, NaN, 0.5070028305053711, 0.5042017102241516, 0.5546218752861023, 0.5028011202812195, 0.5574229955673218, 0.456582635641098, 0.5098039507865906, 0.5826330780982971, 0.5154061913490295, 0.5826330780982971, 0.5448179244995117, 0.5350140333175659, 0.5714285969734192, 0.5672268867492676, 0.6526610851287842, 0.6302521228790283, 0.6778711676597595, 0.6778711676597595, 0.680672287940979, 0.7002801299095154, 0.848739504814148, 0.6946778893470764, 0.8767507076263428, 0.6890756487846375, 0.7352941036224365, 0.6834734082221985, 0.8529411554336548, 0.8501400351524353, 0.7478991746902466, 0.7983193397521973, 0.8067227005958557, 0.848739504814148, 0.9019607901573181, 0.848739504814148, 0.8557422757148743, 0.9747899174690247], [0.10924369841814041, 0.15546219050884247, 0.1596638709306717, 0.1764705926179886, 0.16526611149311066, 0.2072829157114029, 0.19607843458652496, 0.21008403599262238, 0.2044817954301834, 0.24929971992969513, 0.24929971992969513, 0.24369747936725616, 0.30812326073646545, 0.32212886214256287, 0.27731093764305115, 0.23529411852359772, 0.28431373834609985, 0.28151261806488037, 0.30812326073646545, 0.3501400649547577, 0.3361344635486603, 0.32492998242378235, 0.29411765933036804, 0.3445378243923187, 0.37254902720451355, 0.3473389446735382, 0.4145658314228058, 0.41036415100097656, 0.3921568691730499, 0.4089635908603668, 0.41736695170402527, 0.4089635908603668, 0.4285714328289032, 0.40616247057914734, 0.4257703125476837, 0.45098039507865906, 0.35854342579841614, 0.43977591395378113, 0.456582635641098, 0.4075630307197571, 0.5420168042182922, 0.4929971992969513, NaN, 0.48739495873451233, 0.5602241158485413, 0.5350140333175659, 0.5546218752861023, 0.4957983195781708, 0.49859943985939026, 0.5322129130363464, 0.5266106724739075, 0.4957983195781708, 0.529411792755127, 0.5322129130363464, 0.5658263564109802, 0.6106442809104919, 0.6722689270973206, 0.6386554837226868, 0.6442577242851257, 0.7226890921592712, 0.7338935732841492, 0.6834734082221985, 0.6918767690658569, 0.7521008253097534, 0.7002801299095154, 0.6834734082221985, 0.7030812501907349, 0.7254902124404907, 0.7815126180648804, 0.7927170991897583, 0.7983193397521973, 0.7226890921592712, 0.8627451062202454, 0.83753502368927, 0.8263305425643921, 0.843137264251709, 0.8879551887512207, 0.9691876769065857], [0.10644257813692093, 0.13165266811847687, 0.13725490868091583, 0.17927171289920807, 0.14005602896213531, 0.1764705926179886, 0.18767507374286652, 0.190476194024086, 0.21008403599262238, 0.22969187796115875, 0.23249299824237823, 0.26050421595573425, 0.2521008551120758, 0.2885154187679291, 0.2661064565181732, 0.30812326073646545, 0.2142857164144516, 0.2787114977836609, 0.27731093764305115, 0.32492998242378235, 0.3165266215801239, 0.30392158031463623, 0.36974790692329407, 0.32492998242378235, 0.41036415100097656, 0.38235294818878174, 0.3501400649547577, 0.3445378243923187, 0.4145658314228058, 0.40616247057914734, 0.3837535083293915, 0.4453781545162201, 0.4005602300167084, 0.42016807198524475, 0.42717087268829346, 0.45378151535987854, 0.47058823704719543, 0.4593837559223175, 0.45378151535987854, 0.5378151535987854, 0.48459383845329285, 0.4957983195781708, 0.5126050710678101, NaN, 0.5518207550048828, 0.5042017102241516, 0.5742297172546387, 0.49859943985939026, 0.5266106724739075, 0.5350140333175659, 0.5322129130363464, 0.529411792755127, 0.5434173941612244, 0.5042017102241516, 0.5826330780982971, 0.656862735748291, 0.6330532431602478, 0.6358543634414673, 0.6498599648475647, 0.6946778893470764, 0.680672287940979, 0.7114846110343933, 0.6946778893470764, 0.7142857313156128, 0.7002801299095154, 0.7521008253097534, 0.7591036558151245, 0.8123249411582947, 0.7450980544090271, 0.761904776096344, 0.767507016658783, 0.7647058963775635, 0.843137264251709, 0.8039215803146362, 0.8067227005958557, 0.8879551887512207, 0.843137264251709, 0.9691876769065857], [0.2212885171175003, 0.2647058963775635, 0.24929971992969513, 0.31092438101768494, 0.2717086970806122, 0.3137255012989044, 0.31092438101768494, 0.3193277418613434, 0.3165266215801239, 0.3305322229862213, 0.3445378243923187, 0.3235294222831726, 0.3473389446735382, 0.30812326073646545, 0.3641456663608551, 0.3305322229862213, 0.37535014748573303, 0.3613445460796356, 0.32212886214256287, 0.38655462861061096, 0.4019607901573181, 0.3613445460796356, 0.3879551887512207, 0.3837535083293915, 0.42016807198524475, 0.4117647111415863, 0.3977591097354889, 0.45098039507865906, 0.42016807198524475, 0.4551820755004883, 0.43417367339134216, 0.40336135029792786, 0.4481792747974396, 0.4019607901573181, 0.4145658314228058, 0.41736695170402527, 0.4313725531101227, 0.4117647111415863, 0.4481792747974396, 0.4481792747974396, 0.40336135029792786, 0.4453781545162201, 0.43977591395378113, 0.4481792747974396, NaN, 0.5014005899429321, 0.5980392098426819, 0.42016807198524475, 0.462184876203537, 0.46498599648475647, 0.43697479367256165, 0.45798319578170776, 0.4453781545162201, 0.5406162738800049, 0.456582635641098, 0.5602241158485413, 0.6274510025978088, 0.913165271282196, 0.6232492923736572, 0.48179271817207336, 0.6330532431602478, 0.6218487620353699, 0.5770308375358582, 0.6582633256912231, 0.5602241158485413, 0.5658263564109802, 0.6302521228790283, 0.680672287940979, 0.6918767690658569, 0.7254902124404907, 0.686274528503418, 0.7198879718780518, 0.7871148586273193, 0.8095238208770752, 0.8039215803146362, 0.686274528503418, 0.848739504814148, 0.9411764740943909], [0.07002801448106766, 0.12044817954301834, 0.09243697673082352, 0.11624649912118912, 0.10644257813692093, 0.18207283318042755, 0.14845938980579376, 0.15406163036823273, 0.2478991597890854, 0.17086835205554962, 0.2829131782054901, 0.24229691922664642, 0.24929971992969513, 0.190476194024086, 0.3711484670639038, 0.3473389446735382, 0.35574230551719666, 0.37535014748573303, 0.3137255012989044, 0.26330533623695374, 0.24929971992969513, 0.3781512677669525, 0.3949579894542694, 0.4089635908603668, 0.28011205792427063, 0.41736695170402527, 0.44397759437561035, 0.42016807198524475, 0.43417367339134216, 0.4257703125476837, 0.462184876203537, 0.4145658314228058, 0.4551820755004883, 0.48739495873451233, 0.46778711676597595, 0.4929971992969513, 0.43417367339134216, 0.4957983195781708, 0.5154061913490295, 0.4719887971878052, 0.48739495873451233, 0.4971988797187805, 0.46498599648475647, 0.4957983195781708, 0.49859943985939026, NaN, 0.5098039507865906, 0.5014005899429321, 0.5490196347236633, 0.529411792755127, 0.5126050710678101, 0.6022409200668335, 0.5210084319114685, 0.6022409200668335, 0.5798319578170776, 0.6246498823165894, 0.5966386795043945, 0.6330532431602478, 0.599439799785614, 0.6302521228790283, 0.6918767690658569, 0.680672287940979, 0.6358543634414673, 0.7212885022163391, 0.6778711676597595, 0.6974790096282959, 0.7002801299095154, 0.7366946935653687, 0.6638655662536621, 0.7030812501907349, 0.7478991746902466, 0.7759103775024414, 0.7310924530029297, 0.761904776096344, 0.7282913327217102, 0.8347339034080505, 0.7787114977836609, 0.9663865566253662], [0.21848739683628082, 0.24929971992969513, 0.2521008551120758, 0.30812326073646545, 0.26750701665878296, 0.299719899892807, 0.3025210201740265, 0.30812326073646545, 0.3137255012989044, 0.3193277418613434, 0.33893558382987976, 0.32773110270500183, 0.3445378243923187, 0.29411765933036804, 0.3529411852359772, 0.32492998242378235, 0.3613445460796356, 0.35574230551719666, 0.3263305425643921, 0.3837535083293915, 0.40336135029792786, 0.3641456663608551, 0.3837535083293915, 0.37955182790756226, 0.4145658314228058, 0.40616247057914734, 0.3837535083293915, 0.456582635641098, 0.41736695170402527, 0.456582635641098, 0.4257703125476837, 0.3949579894542694, 0.44397759437561035, 0.3949579894542694, 0.40616247057914734, 0.42296919226646423, 0.4313725531101227, 0.40616247057914734, 0.4383753538131714, 0.44117647409439087, 0.3949579894542694, 0.4425770342350006, 0.4453781545162201, 0.4257703125476837, 0.4019607901573181, 0.4901960790157318, NaN, 0.42296919226646423, 0.47058823704719543, 0.45798319578170776, 0.43977591395378113, 0.46778711676597595, 0.4425770342350006, 0.5350140333175659, 0.45098039507865906, 0.5490196347236633, 0.6022409200668335, 0.9355742335319519, 0.599439799785614, 0.48739495873451233, 0.6274510025978088, 0.6190476417541504, 0.5658263564109802, 0.6582633256912231, 0.5602241158485413, 0.5630252361297607, 0.6246498823165894, 0.680672287940979, 0.6834734082221985, 0.7170868515968323, 0.6834734082221985, 0.7114846110343933, 0.7731092572212219, 0.8011204600334167, 0.7983193397521973, 0.6834734082221985, 0.8347339034080505, 0.9411764740943909], [0.07002801448106766, 0.09803921729326248, 0.12044817954301834, 0.11764705926179886, 0.10364145785570145, 0.10364145785570145, 0.1330532282590866, 0.15266107022762299, 0.15406163036823273, 0.23249299824237823, 0.20168067514896393, 0.190476194024086, 0.23249299824237823, 0.3137255012989044, 0.25630253553390503, 0.2591036558151245, 0.22969187796115875, 0.2212885171175003, 0.3333333432674408, 0.31092438101768494, 0.32492998242378235, 0.27450981736183167, 0.35154062509536743, 0.26050421595573425, 0.34173670411109924, 0.4089635908603668, 0.38655462861061096, 0.4117647111415863, 0.3851540684700012, 0.4005602300167084, 0.380952388048172, 0.38655462861061096, 0.40616247057914734, 0.44117647409439087, 0.42296919226646423, 0.43697479367256165, 0.4733893573284149, 0.47478991746902466, 0.48459383845329285, 0.5098039507865906, 0.5154061913490295, 0.5434173941612244, 0.5042017102241516, 0.5014005899429321, 0.5798319578170776, 0.49859943985939026, 0.5770308375358582, NaN, 0.48879551887512207, 0.5896358489990234, 0.5826330780982971, 0.5686274766921997, 0.686274528503418, 0.5546218752861023, 0.5630252361297607, 0.5882353186607361, 0.6274510025978088, 0.6442577242851257, 0.6414566040039062, 0.6890756487846375, 0.6498599648475647, 0.7310924530029297, 0.7703081369400024, 0.6666666865348816, 0.7857142686843872, 0.6918767690658569, 0.7591036558151245, 0.6638655662536621, 0.756302535533905, 0.767507016658783, 0.7142857313156128, 0.7507002949714661, 0.7703081369400024, 0.8011204600334167, 0.8095238208770752, 0.8459383845329285, 0.831932783126831, 0.9649859666824341], [0.08403361588716507, 0.13865546882152557, 0.12885154783725739, 0.16526611149311066, 0.12885154783725739, 0.17927171289920807, 0.1568627506494522, 0.1764705926179886, 0.17927171289920807, 0.18767507374286652, 0.2478991597890854, 0.2478991597890854, 0.24369747936725616, 0.23249299824237823, 0.2661064565181732, 0.2969187796115875, 0.2731092572212219, 0.3025210201740265, 0.27731093764305115, 0.27450981736183167, 0.29411765933036804, 0.3193277418613434, 0.3613445460796356, 0.3333333432674408, 0.32212886214256287, 0.4117647111415863, 0.3473389446735382, 0.3445378243923187, 0.3641456663608551, 0.36974790692329407, 0.380952388048172, 0.4313725531101227, 0.3739495873451233, 0.4257703125476837, 0.3445378243923187, 0.4453781545162201, 0.4803921580314636, 0.44117647409439087, 0.39355742931365967, 0.5, 0.4425770342350006, 0.4901960790157318, 0.5014005899429321, 0.4733893573284149, 0.5378151535987854, 0.45098039507865906, 0.529411792755127, 0.5112044811248779, NaN, 0.48739495873451233, 0.5308123230934143, 0.5126050710678101, 0.5434173941612244, 0.523809552192688, 0.5434173941612244, 0.6246498823165894, 0.6106442809104919, 0.6106442809104919, 0.6022409200668335, 0.6526610851287842, 0.6778711676597595, 0.6834734082221985, 0.6386554837226868, 0.6960784196853638, 0.6386554837226868, 0.7240896224975586, 0.7282913327217102, 0.8095238208770752, 0.7184873819351196, 0.7226890921592712, 0.7170868515968323, 0.7198879718780518, 0.8067227005958557, 0.7591036558151245, 0.7731092572212219, 0.8403361439704895, 0.831932783126831, 0.9579831957817078], [0.04481792822480202, 0.10084033757448196, 0.12044817954301834, 0.12885154783725739, 0.12044817954301834, 0.14005602896213531, 0.13865546882152557, 0.16806723177433014, 0.19607843458652496, 0.23249299824237823, 0.21568627655506134, 0.19887955486774445, 0.23529411852359772, 0.2885154187679291, 0.2380952388048172, 0.23249299824237823, 0.20168067514896393, 0.19887955486774445, 0.30812326073646545, 0.2969187796115875, 0.32773110270500183, 0.2969187796115875, 0.18487395346164703, 0.29411765933036804, 0.37254902720451355, 0.25770309567451477, 0.33753502368927, 0.3921568691730499, 0.4159663915634155, 0.4117647111415863, 0.39635854959487915, 0.37535014748573303, 0.3921568691730499, 0.41316527128219604, 0.42296919226646423, 0.41316527128219604, 0.4145658314228058, 0.40616247057914734, 0.47058823704719543, 0.46498599648475647, 0.4901960790157318, 0.41736695170402527, 0.46778711676597595, 0.46498599648475647, 0.5350140333175659, 0.47058823704719543, 0.5420168042182922, 0.41036415100097656, 0.5126050710678101, NaN, 0.4761904776096344, 0.5630252361297607, 0.4929971992969513, 0.5266106724739075, 0.5406162738800049, 0.5518207550048828, 0.6540616154670715, 0.6246498823165894, 0.6694678068161011, 0.6722689270973206, 0.6610644459724426, 0.6610644459724426, 0.7927170991897583, 0.6918767690658569, 0.8207283020019531, 0.6554622054100037, 0.7198879718780518, 0.6778711676597595, 0.8263305425643921, 0.8347339034080505, 0.7366946935653687, 0.7871148586273193, 0.8081232309341431, 0.8459383845329285, 0.8767507076263428, 0.843137264251709, 0.8515406250953674, 0.9691876769065857], [0.05602240934967995, 0.08963585644960403, 0.09803921729326248, 0.10924369841814041, 0.09803921729326248, 0.095238097012043, 0.11764705926179886, 0.13165266811847687, 0.15126051008701324, 0.21008403599262238, 0.190476194024086, 0.16806723177433014, 0.21008403599262238, 0.27450981736183167, 0.2380952388048172, 0.23249299824237823, 0.2142857164144516, 0.1736694723367691, 0.3137255012989044, 0.28011205792427063, 0.30532214045524597, 0.2521008551120758, 0.3165266215801239, 0.26330533623695374, 0.3501400649547577, 0.380952388048172, 0.34173670411109924, 0.3837535083293915, 0.3641456663608551, 0.3837535083293915, 0.3767507076263428, 0.3669467866420746, 0.40616247057914734, 0.3949579894542694, 0.40336135029792786, 0.4117647111415863, 0.43977591395378113, 0.4355742335319519, 0.43977591395378113, 0.47478991746902466, 0.48459383845329285, 0.48459383845329285, 0.4733893573284149, 0.46778711676597595, 0.5630252361297607, 0.48739495873451233, 0.5602241158485413, 0.41736695170402527, 0.4691876769065857, 0.523809552192688, NaN, 0.5224089622497559, 0.5966386795043945, 0.5266106724739075, 0.523809552192688, 0.5546218752861023, 0.593837559223175, 0.6358543634414673, 0.6190476417541504, 0.6470588445663452, 0.6190476417541504, 0.668067216873169, 0.732492983341217, 0.6274510025978088, 0.7310924530029297, 0.6778711676597595, 0.7016806602478027, 0.6470588445663452, 0.7254902124404907, 0.7282913327217102, 0.6890756487846375, 0.7198879718780518, 0.7394958138465881, 0.7899159789085388, 0.7759103775024414, 0.8361344337463379, 0.7955182194709778, 0.9495798349380493], [0.06442577391862869, 0.08263305574655533, 0.1120448186993599, 0.10364145785570145, 0.09383753687143326, 0.1120448186993599, 0.1260504275560379, 0.13445378839969635, 0.16246499121189117, 0.16946779191493988, 0.2044817954301834, 0.1736694723367691, 0.18487395346164703, 0.2044817954301834, 0.21568627655506134, 0.2969187796115875, 0.22689075767993927, 0.24649859964847565, 0.30812326073646545, 0.2240896373987198, 0.24929971992969513, 0.28151261806488037, 0.299719899892807, 0.32492998242378235, 0.32492998242378235, 0.3361344635486603, 0.35574230551719666, 0.3879551887512207, 0.3669467866420746, 0.3907563090324402, 0.3669467866420746, 0.4075630307197571, 0.28151261806488037, 0.3501400649547577, 0.4313725531101227, 0.3529411852359772, 0.43977591395378113, 0.38935574889183044, 0.44397759437561035, 0.4733893573284149, 0.4355742335319519, 0.41736695170402527, 0.5042017102241516, 0.47058823704719543, 0.5420168042182922, 0.3977591097354889, 0.5322129130363464, 0.4313725531101227, 0.48739495873451233, 0.43697479367256165, 0.47759103775024414, NaN, 0.48459383845329285, 0.49859943985939026, 0.5434173941612244, 0.5420168042182922, 0.5686274766921997, 0.6078431606292725, 0.5854341983795166, 0.6456582546234131, 0.575630247592926, 0.6274510025978088, 0.6218487620353699, 0.6106442809104919, 0.6218487620353699, 0.7282913327217102, 0.6890756487846375, 0.7226890921592712, 0.686274528503418, 0.6974790096282959, 0.6666666865348816, 0.7647058963775635, 0.7366946935653687, 0.7254902124404907, 0.7535014152526855, 0.7955182194709778, 0.7913165092468262, 0.9495798349380493], [0.05042016878724098, 0.10084033757448196, 0.10924369841814041, 0.1120448186993599, 0.095238097012043, 0.09243697673082352, 0.11764705926179886, 0.12885154783725739, 0.13165266811847687, 0.190476194024086, 0.16806723177433014, 0.1596638709306717, 0.2044817954301834, 0.22689075767993927, 0.2240896373987198, 0.21568627655506134, 0.21288515627384186, 0.16806723177433014, 0.32212886214256287, 0.2549019753932953, 0.27731093764305115, 0.22689075767993927, 0.21568627655506134, 0.22549019753932953, 0.3361344635486603, 0.2661064565181732, 0.31092438101768494, 0.35854342579841614, 0.37254902720451355, 0.38935574889183044, 0.34173670411109924, 0.3613445460796356, 0.3571428656578064, 0.35574230551719666, 0.4145658314228058, 0.37535014748573303, 0.3949579894542694, 0.38655462861061096, 0.4257703125476837, 0.43697479367256165, 0.43977591395378113, 0.4551820755004883, 0.47058823704719543, 0.456582635641098, 0.5546218752861023, 0.4789915978908539, 0.5574229955673218, 0.3137255012989044, 0.456582635641098, 0.5070028305053711, 0.40336135029792786, 0.5154061913490295, NaN, 0.523809552192688, 0.5098039507865906, 0.5490196347236633, 0.605042040348053, 0.6246498823165894, 0.6106442809104919, 0.6498599648475647, 0.6358543634414673, 0.6386554837226868, 0.7212885022163391, 0.6554622054100037, 0.7268907427787781, 0.6554622054100037, 0.6610644459724426, 0.6330532431602478, 0.7114846110343933, 0.7282913327217102, 0.7086834907531738, 0.7226890921592712, 0.732492983341217, 0.756302535533905, 0.7647058963775635, 0.7927170991897583, 0.8179271817207336, 0.9579831957817078], [0.04481792822480202, 0.08123249560594559, 0.08683473616838455, 0.07282913476228714, 0.1260504275560379, 0.07002801448106766, 0.1120448186993599, 0.12184873968362808, 0.18207283318042755, 0.2380952388048172, 0.16526611149311066, 0.16246499121189117, 0.18767507374286652, 0.2661064565181732, 0.2857142984867096, 0.3137255012989044, 0.29411765933036804, 0.26330533623695374, 0.30812326073646545, 0.2380952388048172, 0.261904776096344, 0.28011205792427063, 0.34173670411109924, 0.30812326073646545, 0.3473389446735382, 0.37535014748573303, 0.38935574889183044, 0.3837535083293915, 0.4313725531101227, 0.36974790692329407, 0.38655462861061096, 0.46498599648475647, 0.38655462861061096, 0.4145658314228058, 0.42016807198524475, 0.4089635908603668, 0.45378151535987854, 0.4089635908603668, 0.4285714328289032, 0.4859943985939026, 0.4285714328289032, 0.46498599648475647, 0.46778711676597595, 0.4957983195781708, 0.4593837559223175, 0.3977591097354889, 0.46498599648475647, 0.4453781545162201, 0.4761904776096344, 0.4733893573284149, 0.4733893573284149, 0.5014005899429321, 0.4761904776096344, NaN, 0.5126050710678101, 0.5266106724739075, 0.5210084319114685, 0.5266106724739075, 0.5490196347236633, 0.5826330780982971, 0.5322129130363464, 0.5966386795043945, 0.6554622054100037, 0.5350140333175659, 0.6386554837226868, 0.605042040348053, 0.6022409200668335, 0.6330532431602478, 0.6078431606292725, 0.6134454011917114, 0.5826330780982971, 0.7717086672782898, 0.6638655662536621, 0.7170868515968323, 0.7142857313156128, 0.7226890921592712, 0.7030812501907349, 0.9159663915634155], [0.03921568766236305, 0.07282913476228714, 0.0784313753247261, 0.09243697673082352, 0.07563025504350662, 0.095238097012043, 0.10364145785570145, 0.10364145785570145, 0.10644257813692093, 0.18487395346164703, 0.15406163036823273, 0.18767507374286652, 0.19607843458652496, 0.23529411852359772, 0.2044817954301834, 0.2521008551120758, 0.24369747936725616, 0.2310924381017685, 0.2661064565181732, 0.26050421595573425, 0.27450981736183167, 0.24649859964847565, 0.3137255012989044, 0.28711485862731934, 0.30812326073646545, 0.33473390340805054, 0.3165266215801239, 0.31092438101768494, 0.3613445460796356, 0.380952388048172, 0.3025210201740265, 0.3837535083293915, 0.3501400649547577, 0.35574230551719666, 0.3949579894542694, 0.3669467866420746, 0.418767511844635, 0.3767507076263428, 0.2927170991897583, 0.4425770342350006, 0.40336135029792786, 0.4285714328289032, 0.43417367339134216, 0.41736695170402527, 0.5434173941612244, 0.42016807198524475, 0.5490196347236633, 0.43697479367256165, 0.456582635641098, 0.4593837559223175, 0.4761904776096344, 0.456582635641098, 0.4901960790157318, 0.48739495873451233, NaN, 0.6106442809104919, 0.5980392098426819, 0.6246498823165894, 0.5742297172546387, 0.6386554837226868, 0.6134454011917114, 0.6386554837226868, 0.6106442809104919, 0.6316526532173157, 0.6092436909675598, 0.743697464466095, 0.680672287940979, 0.7142857313156128, 0.680672287940979, 0.6918767690658569, 0.7282913327217102, 0.7338935732841492, 0.7605041861534119, 0.7226890921592712, 0.7310924530029297, 0.8123249411582947, 0.767507016658783, 0.9523809552192688], [0.08963585644960403, 0.11764705926179886, 0.1260504275560379, 0.14565826952457428, 0.1190476194024086, 0.14845938980579376, 0.15406163036823273, 0.16806723177433014, 0.1736694723367691, 0.12885154783725739, 0.2240896373987198, 0.26330533623695374, 0.2240896373987198, 0.13725490868091583, 0.26050421595573425, 0.28431373834609985, 0.2885154187679291, 0.2857142984867096, 0.1736694723367691, 0.24089635908603668, 0.24649859964847565, 0.2689075767993927, 0.35854342579841614, 0.30532214045524597, 0.2240896373987198, 0.37535014748573303, 0.3123249411582947, 0.3025210201740265, 0.27731093764305115, 0.3025210201740265, 0.32492998242378235, 0.3305322229862213, 0.32492998242378235, 0.41736695170402527, 0.3137255012989044, 0.43697479367256165, 0.3669467866420746, 0.4481792747974396, 0.3669467866420746, 0.4047619104385376, 0.4313725531101227, 0.4327731132507324, 0.38935574889183044, 0.343137264251709, 0.43977591395378113, 0.37535014748573303, 0.45098039507865906, 0.4117647111415863, 0.37535014748573303, 0.4481792747974396, 0.4453781545162201, 0.45798319578170776, 0.45098039507865906, 0.4733893573284149, 0.38935574889183044, NaN, 0.518207311630249, 0.5686274766921997, 0.518207311630249, 0.5518207550048828, 0.5658263564109802, 0.593837559223175, 0.5574229955673218, 0.6064425706863403, 0.5574229955673218, 0.575630247592926, 0.6274510025978088, 0.6190476417541504, 0.6078431606292725, 0.6246498823165894, 0.6358543634414673, 0.6414566040039062, 0.686274528503418, 0.6610644459724426, 0.6974790096282959, 0.7058823704719543, 0.7366946935653687, 0.918767511844635], [0.05602240934967995, 0.06722689419984818, 0.10504201799631119, 0.09243697673082352, 0.1120448186993599, 0.10084033757448196, 0.13445378839969635, 0.13865546882152557, 0.14565826952457428, 0.13025210797786713, 0.15826331079006195, 0.1764705926179886, 0.15406163036823273, 0.1736694723367691, 0.24369747936725616, 0.18207283318042755, 0.19887955486774445, 0.2072829157114029, 0.2380952388048172, 0.19607843458652496, 0.2380952388048172, 0.2689075767993927, 0.2310924381017685, 0.2717086970806122, 0.22689075767993927, 0.25630253553390503, 0.29131653904914856, 0.32492998242378235, 0.2829131782054901, 0.30812326073646545, 0.3473389446735382, 0.3193277418613434, 0.3207283020019531, 0.2717086970806122, 0.33893558382987976, 0.28011205792427063, 0.2899159789085388, 0.299719899892807, 0.3529411852359772, 0.3361344635486603, 0.32773110270500183, 0.3473389446735382, 0.32773110270500183, 0.3669467866420746, 0.37254902720451355, 0.40336135029792786, 0.3977591097354889, 0.37254902720451355, 0.38935574889183044, 0.34593838453292847, 0.40616247057914734, 0.4313725531101227, 0.3949579894542694, 0.4789915978908539, 0.4019607901573181, 0.48179271817207336, NaN, 0.5728291273117065, 0.6106442809104919, 0.49859943985939026, 0.5784313678741455, 0.5434173941612244, 0.5070028305053711, 0.5980392098426819, 0.5126050710678101, 0.5490196347236633, 0.5574229955673218, 0.599439799785614, 0.7156862616539001, 0.7450980544090271, 0.6610644459724426, 0.6638655662536621, 0.8011204600334167, 0.848739504814148, 0.8039215803146362, 0.6638655662536621, 0.831932783126831, 0.9859943985939026], [0.17927171289920807, 0.2072829157114029, 0.20168067514896393, 0.24369747936725616, 0.22969187796115875, 0.25770309567451477, 0.25770309567451477, 0.2549019753932953, 0.2689075767993927, 0.25770309567451477, 0.27450981736183167, 0.26750701665878296, 0.27731093764305115, 0.24089635908603668, 0.29411765933036804, 0.2591036558151245, 0.30812326073646545, 0.3025210201740265, 0.2549019753932953, 0.29411765933036804, 0.299719899892807, 0.29411765933036804, 0.29411765933036804, 0.32773110270500183, 0.32492998242378235, 0.3165266215801239, 0.32492998242378235, 0.35854342579841614, 0.3445378243923187, 0.35574230551719666, 0.3333333432674408, 0.3193277418613434, 0.3445378243923187, 0.3333333432674408, 0.34173670411109924, 0.3473389446735382, 0.3333333432674408, 0.33893558382987976, 0.34173670411109924, 0.35854342579841614, 0.3165266215801239, 0.36974790692329407, 0.3613445460796356, 0.3641456663608551, 0.08683473616838455, 0.3669467866420746, 0.06442577391862869, 0.35574230551719666, 0.38935574889183044, 0.37535014748573303, 0.3641456663608551, 0.3921568691730499, 0.37535014748573303, 0.4733893573284149, 0.37535014748573303, 0.4313725531101227, 0.42717087268829346, NaN, 0.4313725531101227, 0.42296919226646423, 0.49859943985939026, 0.523809552192688, 0.46778711676597595, 0.5154061913490295, 0.48179271817207336, 0.45378151535987854, 0.5210084319114685, 0.5070028305053711, 0.505602240562439, 0.5350140333175659, 0.5826330780982971, 0.6330532431602478, 0.6666666865348816, 0.6470588445663452, 0.605042040348053, 0.5882353186607361, 0.7394958138465881, 0.9355742335319519], [0.06722689419984818, 0.07282913476228714, 0.10644257813692093, 0.1120448186993599, 0.1120448186993599, 0.11344537883996964, 0.1428571492433548, 0.14005602896213531, 0.14425770938396454, 0.10364145785570145, 0.1596638709306717, 0.18767507374286652, 0.16946779191493988, 0.17927171289920807, 0.23529411852359772, 0.18767507374286652, 0.20588235557079315, 0.23529411852359772, 0.22689075767993927, 0.19607843458652496, 0.22969187796115875, 0.22689075767993927, 0.2521008551120758, 0.27731093764305115, 0.22689075767993927, 0.26750701665878296, 0.29131653904914856, 0.32492998242378235, 0.2689075767993927, 0.28011205792427063, 0.33893558382987976, 0.3165266215801239, 0.30812326073646545, 0.2857142984867096, 0.3473389446735382, 0.28431373834609985, 0.30812326073646545, 0.3137255012989044, 0.3613445460796356, 0.30532214045524597, 0.3193277418613434, 0.32212886214256287, 0.35574230551719666, 0.3501400649547577, 0.3767507076263428, 0.4005602300167084, 0.4005602300167084, 0.35854342579841614, 0.3977591097354889, 0.3305322229862213, 0.380952388048172, 0.4145658314228058, 0.38935574889183044, 0.45098039507865906, 0.4257703125476837, 0.48179271817207336, 0.38935574889183044, 0.5686274766921997, NaN, 0.5126050710678101, 0.5714285969734192, 0.5154061913490295, 0.4789915978908539, 0.593837559223175, 0.4733893573284149, 0.5518207550048828, 0.5462185144424438, 0.5910364389419556, 0.6610644459724426, 0.7002801299095154, 0.6554622054100037, 0.6330532431602478, 0.756302535533905, 0.7759103775024414, 0.7661064267158508, 0.6414566040039062, 0.83753502368927, 0.9831932783126831], [0.07282913476228714, 0.08403361588716507, 0.09243697673082352, 0.095238097012043, 0.08963585644960403, 0.10924369841814041, 0.10924369841814041, 0.11484593898057938, 0.1120448186993599, 0.16806723177433014, 0.1568627506494522, 0.14005602896213531, 0.1736694723367691, 0.19887955486774445, 0.18207283318042755, 0.14005602896213531, 0.15406163036823273, 0.16806723177433014, 0.19887955486774445, 0.1764705926179886, 0.21008403599262238, 0.19327731430530548, 0.17927171289920807, 0.190476194024086, 0.2521008551120758, 0.20308123528957367, 0.21288515627384186, 0.28011205792427063, 0.22969187796115875, 0.2661064565181732, 0.2521008551120758, 0.2380952388048172, 0.25770309567451477, 0.19327731430530548, 0.27731093764305115, 0.24229691922664642, 0.24929971992969513, 0.2661064565181732, 0.2759103775024414, 0.29411765933036804, 0.3165266215801239, 0.32212886214256287, 0.27731093764305115, 0.30532214045524597, 0.518207311630249, 0.36974790692329407, 0.5126050710678101, 0.31092438101768494, 0.3473389446735382, 0.32773110270500183, 0.3529411852359772, 0.3543417453765869, 0.3501400649547577, 0.41736695170402527, 0.3613445460796356, 0.4481792747974396, 0.5014005899429321, 0.5770308375358582, 0.48739495873451233, NaN, 0.5686274766921997, 0.47058823704719543, 0.4761904776096344, 0.5966386795043945, 0.47058823704719543, 0.5350140333175659, 0.4901960790157318, 0.5490196347236633, 0.6302521228790283, 0.6526610851287842, 0.6526610851287842, 0.6134454011917114, 0.7338935732841492, 0.6694678068161011, 0.7058823704719543, 0.6106442809104919, 0.8067227005958557, 0.9215686321258545], [0.08683473616838455, 0.12885154783725739, 0.10924369841814041, 0.15406163036823273, 0.14005602896213531, 0.15406163036823273, 0.14005602896213531, 0.1596638709306717, 0.14845938980579376, 0.16806723177433014, 0.18487395346164703, 0.190476194024086, 0.2072829157114029, 0.18767507374286652, 0.2002801150083542, 0.14565826952457428, 0.21288515627384186, 0.18767507374286652, 0.17226891219615936, 0.24369747936725616, 0.23949579894542694, 0.22689075767993927, 0.21848739683628082, 0.21568627655506134, 0.25770309567451477, 0.23249299824237823, 0.29411765933036804, 0.2689075767993927, 0.24089635908603668, 0.27731093764305115, 0.3137255012989044, 0.2661064565181732, 0.34173670411109924, 0.2969187796115875, 0.27450981736183167, 0.3025210201740265, 0.25770309567451477, 0.30812326073646545, 0.3095238208770752, 0.2857142984867096, 0.2969187796115875, 0.3193277418613434, 0.2661064565181732, 0.3193277418613434, 0.3669467866420746, 0.30812326073646545, 0.37254902720451355, 0.3501400649547577, 0.32212886214256287, 0.33893558382987976, 0.380952388048172, 0.424369752407074, 0.3641456663608551, 0.46778711676597595, 0.38655462861061096, 0.43417367339134216, 0.4215686321258545, 0.5014005899429321, 0.4285714328289032, 0.4313725531101227, NaN, 0.4789915978908539, 0.4593837559223175, 0.6386554837226868, 0.43977591395378113, 0.4761904776096344, 0.5014005899429321, 0.5014005899429321, 0.4789915978908539, 0.4901960790157318, 0.7198879718780518, 0.5602241158485413, 0.6246498823165894, 0.5910364389419556, 0.5546218752861023, 0.593837559223175, 0.7002801299095154, 0.8935574293136597], [0.02240896411240101, 0.0784313753247261, 0.07002801448106766, 0.07563025504350662, 0.0784313753247261, 0.08123249560594559, 0.07563025504350662, 0.09943977743387222, 0.13165266811847687, 0.13725490868091583, 0.14005602896213531, 0.1260504275560379, 0.12885154783725739, 0.15406163036823273, 0.095238097012043, 0.1764705926179886, 0.10084033757448196, 0.13165266811847687, 0.20168067514896393, 0.17086835205554962, 0.190476194024086, 0.21848739683628082, 0.14845938980579376, 0.13725490868091583, 0.22969187796115875, 0.16946779191493988, 0.17927171289920807, 0.23529411852359772, 0.261904776096344, 0.2240896373987198, 0.2549019753932953, 0.2689075767993927, 0.19607843458652496, 0.26050421595573425, 0.22969187796115875, 0.24649859964847565, 0.2689075767993927, 0.26330533623695374, 0.24929971992969513, 0.29411765933036804, 0.31792718172073364, 0.299719899892807, 0.3165266215801239, 0.2885154187679291, 0.3781512677669525, 0.3193277418613434, 0.380952388048172, 0.2689075767993927, 0.3165266215801239, 0.33893558382987976, 0.33193278312683105, 0.37254902720451355, 0.3613445460796356, 0.40336135029792786, 0.3613445460796356, 0.40616247057914734, 0.456582635641098, 0.4761904776096344, 0.48459383845329285, 0.529411792755127, 0.5210084319114685, NaN, 0.5602241158485413, 0.5350140333175659, 0.5602241158485413, 0.5070028305053711, 0.651260495185852, 0.4957983195781708, 0.5574229955673218, 0.5826330780982971, 0.605042040348053, 0.6442577242851257, 0.6596638560295105, 0.6834734082221985, 0.7030812501907349, 0.7647058963775635, 0.7226890921592712, 0.9411764740943909], [0.030812324956059456, 0.05042016878724098, 0.07282913476228714, 0.07282913476228714, 0.07563025504350662, 0.05602240934967995, 0.0784313753247261, 0.08403361588716507, 0.08403361588716507, 0.13165266811847687, 0.09803921729326248, 0.12044817954301834, 0.10924369841814041, 0.18207283318042755, 0.10644257813692093, 0.16526611149311066, 0.11764705926179886, 0.12044817954301834, 0.23249299824237823, 0.17927171289920807, 0.21288515627384186, 0.22969187796115875, 0.11484593898057938, 0.1736694723367691, 0.26330533623695374, 0.15826331079006195, 0.21848739683628082, 0.27450981736183167, 0.299719899892807, 0.23249299824237823, 0.26330533623695374, 0.3137255012989044, 0.2044817954301834, 0.19607843458652496, 0.29131653904914856, 0.18487395346164703, 0.2478991597890854, 0.16806723177433014, 0.27450981736183167, 0.2717086970806122, 0.3501400649547577, 0.15126051008701324, 0.30812326073646545, 0.30532214045524597, 0.42296919226646423, 0.3641456663608551, 0.43417367339134216, 0.22969187796115875, 0.3613445460796356, 0.2072829157114029, 0.26750701665878296, 0.3781512677669525, 0.2787114977836609, 0.3445378243923187, 0.38935574889183044, 0.4425770342350006, 0.4929971992969513, 0.5322129130363464, 0.5210084319114685, 0.523809552192688, 0.5406162738800049, 0.43977591395378113, NaN, 0.5546218752861023, 0.505602240562439, 0.5266106724739075, 0.4957983195781708, 0.5518207550048828, 0.5798319578170776, 0.6134454011917114, 0.6274510025978088, 0.680672287940979, 0.680672287940979, 0.7198879718780518, 0.7394958138465881, 0.6414566040039062, 0.7366946935653687, 0.924369752407074], [0.07563025504350662, 0.11764705926179886, 0.09803921729326248, 0.12044817954301834, 0.1260504275560379, 0.12324929982423782, 0.11764705926179886, 0.14005602896213531, 0.11764705926179886, 0.14845938980579376, 0.16246499121189117, 0.16246499121189117, 0.18487395346164703, 0.16526611149311066, 0.1666666716337204, 0.11764705926179886, 0.2072829157114029, 0.17226891219615936, 0.16526611149311066, 0.21848739683628082, 0.21568627655506134, 0.21008403599262238, 0.2044817954301834, 0.19887955486774445, 0.24089635908603668, 0.2142857164144516, 0.28431373834609985, 0.26330533623695374, 0.21288515627384186, 0.25770309567451477, 0.29131653904914856, 0.2478991597890854, 0.3137255012989044, 0.27450981736183167, 0.24929971992969513, 0.28011205792427063, 0.24649859964847565, 0.2927170991897583, 0.2717086970806122, 0.2717086970806122, 0.2759103775024414, 0.30532214045524597, 0.2478991597890854, 0.2857142984867096, 0.34173670411109924, 0.2787114977836609, 0.34173670411109924, 0.3333333432674408, 0.30392158031463623, 0.30812326073646545, 0.37254902720451355, 0.38935574889183044, 0.3445378243923187, 0.46498599648475647, 0.3683473467826843, 0.39355742931365967, 0.4019607901573181, 0.48459383845329285, 0.40616247057914734, 0.40336135029792786, 0.3613445460796356, 0.46498599648475647, 0.4453781545162201, NaN, 0.4285714328289032, 0.456582635641098, 0.48739495873451233, 0.48459383845329285, 0.4733893573284149, 0.4761904776096344, 0.6330532431602478, 0.5574229955673218, 0.5784313678741455, 0.5686274766921997, 0.5406162738800049, 0.5910364389419556, 0.6778711676597595, 0.8823529481887817], [0.03361344709992409, 0.053221289068460464, 0.05882352963089943, 0.06442577391862869, 0.042016807943582535, 0.05042016878724098, 0.06162464991211891, 0.06722689419984818, 0.08963585644960403, 0.11764705926179886, 0.08123249560594559, 0.09803921729326248, 0.09103641659021378, 0.16386555135250092, 0.08123249560594559, 0.14425770938396454, 0.10924369841814041, 0.11624649912118912, 0.190476194024086, 0.15126051008701324, 0.19607843458652496, 0.18207283318042755, 0.05602240934967995, 0.17086835205554962, 0.26050421595573425, 0.09663865715265274, 0.19607843458652496, 0.2717086970806122, 0.2549019753932953, 0.21288515627384186, 0.2521008551120758, 0.28011205792427063, 0.15406163036823273, 0.16526611149311066, 0.2899159789085388, 0.1568627506494522, 0.23529411852359772, 0.1764705926179886, 0.3025210201740265, 0.2549019753932953, 0.34173670411109924, 0.12324929982423782, 0.299719899892807, 0.299719899892807, 0.43977591395378113, 0.32212886214256287, 0.43977591395378113, 0.2142857164144516, 0.3613445460796356, 0.17927171289920807, 0.2689075767993927, 0.3781512677669525, 0.2731092572212219, 0.3613445460796356, 0.3907563090324402, 0.4425770342350006, 0.48739495873451233, 0.518207311630249, 0.5266106724739075, 0.529411792755127, 0.5602241158485413, 0.43977591395378113, 0.49439775943756104, 0.5714285969734192, NaN, 0.5574229955673218, 0.5042017102241516, 0.5490196347236633, 0.5854341983795166, 0.6148459315299988, 0.6330532431602478, 0.6890756487846375, 0.680672287940979, 0.7086834907531738, 0.7240896224975586, 0.6722689270973206, 0.7478991746902466, 0.9481792449951172], [0.028011204674839973, 0.05602240934967995, 0.0476190485060215, 0.06442577391862869, 0.03641456738114357, 0.06162464991211891, 0.08123249560594559, 0.07282913476228714, 0.07002801448106766, 0.09803921729326248, 0.10084033757448196, 0.14005602896213531, 0.10924369841814041, 0.09243697673082352, 0.13725490868091583, 0.1736694723367691, 0.12885154783725739, 0.12885154783725739, 0.14845938980579376, 0.14005602896213531, 0.15126051008701324, 0.19327731430530548, 0.22689075767993927, 0.14845938980579376, 0.16806723177433014, 0.2380952388048172, 0.1736694723367691, 0.23249299824237823, 0.17086835205554962, 0.18767507374286652, 0.2521008551120758, 0.19327731430530548, 0.2240896373987198, 0.24089635908603668, 0.1834733933210373, 0.26330533623695374, 0.3305322229862213, 0.2829131782054901, 0.16246499121189117, 0.3501400649547577, 0.2717086970806122, 0.31092438101768494, 0.3165266215801239, 0.2478991597890854, 0.43417367339134216, 0.3025210201740265, 0.43697479367256165, 0.30812326073646545, 0.2759103775024414, 0.3445378243923187, 0.32212886214256287, 0.2717086970806122, 0.3445378243923187, 0.3949579894542694, 0.25630253553390503, 0.424369752407074, 0.45098039507865906, 0.5462185144424438, 0.4481792747974396, 0.46498599648475647, 0.523809552192688, 0.4929971992969513, 0.4733893573284149, 0.5434173941612244, 0.4425770342350006, NaN, 0.5350140333175659, 0.593837559223175, 0.5266106724739075, 0.5658263564109802, 0.6134454011917114, 0.6022409200668335, 0.680672287940979, 0.6218487620353699, 0.6162465214729309, 0.67507004737854, 0.7310924530029297, 0.9271708726882935], [0.019607843831181526, 0.05882352963089943, 0.05882352963089943, 0.06022408977150917, 0.06442577391862869, 0.08123249560594559, 0.07002801448106766, 0.0784313753247261, 0.10364145785570145, 0.10924369841814041, 0.10084033757448196, 0.10924369841814041, 0.1120448186993599, 0.1596638709306717, 0.08963585644960403, 0.1568627506494522, 0.08963585644960403, 0.10924369841814041, 0.19607843458652496, 0.1498599499464035, 0.17927171289920807, 0.21148459613323212, 0.12044817954301834, 0.12324929982423782, 0.2044817954301834, 0.14845938980579376, 0.17086835205554962, 0.22549019753932953, 0.20588235557079315, 0.16946779191493988, 0.2240896373987198, 0.23529411852359772, 0.16806723177433014, 0.21288515627384186, 0.20168067514896393, 0.21848739683628082, 0.24229691922664642, 0.22969187796115875, 0.2044817954301834, 0.2689075767993927, 0.30532214045524597, 0.2647058963775635, 0.2969187796115875, 0.24089635908603668, 0.36974790692329407, 0.299719899892807, 0.37535014748573303, 0.24089635908603668, 0.2717086970806122, 0.28011205792427063, 0.29831933975219727, 0.31092438101768494, 0.33893558382987976, 0.3977591097354889, 0.3193277418613434, 0.37254902720451355, 0.4425770342350006, 0.4789915978908539, 0.45378151535987854, 0.5098039507865906, 0.49859943985939026, 0.34873950481414795, 0.5042017102241516, 0.5126050710678101, 0.4957983195781708, 0.46498599648475647, NaN, 0.46498599648475647, 0.5210084319114685, 0.5378151535987854, 0.5826330780982971, 0.6204481720924377, 0.6302521228790283, 0.6610644459724426, 0.6722689270973206, 0.6988795399665833, 0.6946778893470764, 0.9075630307197571], [0.0476190485060215, 0.05602240934967995, 0.06442577391862869, 0.07563025504350662, 0.07282913476228714, 0.08963585644960403, 0.08123249560594559, 0.095238097012043, 0.08963585644960403, 0.08403361588716507, 0.09803921729326248, 0.1120448186993599, 0.11344537883996964, 0.10644257813692093, 0.10924369841814041, 0.19607843458652496, 0.10084033757448196, 0.13165266811847687, 0.1596638709306717, 0.11484593898057938, 0.13445378839969635, 0.190476194024086, 0.22969187796115875, 0.19327731430530548, 0.15126051008701324, 0.24089635908603668, 0.14145658910274506, 0.2212885171175003, 0.16526611149311066, 0.15126051008701324, 0.228291317820549, 0.21008403599262238, 0.22689075767993927, 0.24929971992969513, 0.07002801448106766, 0.26330533623695374, 0.24649859964847565, 0.27450981736183167, 0.2002801150083542, 0.2857142984867096, 0.2857142984867096, 0.3165266215801239, 0.27450981736183167, 0.18767507374286652, 0.3193277418613434, 0.26330533623695374, 0.3193277418613434, 0.3361344635486603, 0.190476194024086, 0.32212886214256287, 0.3529411852359772, 0.27731093764305115, 0.3669467866420746, 0.3669467866420746, 0.2857142984867096, 0.380952388048172, 0.4005602300167084, 0.4929971992969513, 0.4089635908603668, 0.45098039507865906, 0.49859943985939026, 0.5042017102241516, 0.4481792747974396, 0.5154061913490295, 0.45098039507865906, 0.40616247057914734, 0.5350140333175659, NaN, 0.49859943985939026, 0.49859943985939026, 0.5490196347236633, 0.5854341983795166, 0.6610644459724426, 0.6162465214729309, 0.5980392098426819, 0.6778711676597595, 0.7282913327217102, 0.9299719929695129], [0.02240896411240101, 0.06162464991211891, 0.06722689419984818, 0.07002801448106766, 0.08683473616838455, 0.06722689419984818, 0.08403361588716507, 0.10084033757448196, 0.10644257813692093, 0.08683473616838455, 0.08963585644960403, 0.10644257813692093, 0.11484593898057938, 0.14565826952457428, 0.12885154783725739, 0.10644257813692093, 0.10364145785570145, 0.14845938980579376, 0.1764705926179886, 0.11764705926179886, 0.1596638709306717, 0.21568627655506134, 0.11484593898057938, 0.1736694723367691, 0.2072829157114029, 0.16246499121189117, 0.2044817954301834, 0.2310924381017685, 0.23529411852359772, 0.1764705926179886, 0.2380952388048172, 0.24089635908603668, 0.190476194024086, 0.14565826952457428, 0.21708683669567108, 0.15406163036823273, 0.21008403599262238, 0.15406163036823273, 0.2549019753932953, 0.24089635908603668, 0.21288515627384186, 0.14705882966518402, 0.21848739683628082, 0.2549019753932953, 0.30812326073646545, 0.3361344635486603, 0.3165266215801239, 0.24369747936725616, 0.28151261806488037, 0.1736694723367691, 0.27450981736183167, 0.3137255012989044, 0.2885154187679291, 0.3921568691730499, 0.3193277418613434, 0.3921568691730499, 0.28431373834609985, 0.49439775943756104, 0.33893558382987976, 0.36974790692329407, 0.5210084319114685, 0.4425770342350006, 0.42016807198524475, 0.5266106724739075, 0.4145658314228058, 0.4733893573284149, 0.4789915978908539, 0.5014005899429321, NaN, 0.6106442809104919, 0.6008403301239014, 0.6414566040039062, 0.6610644459724426, 0.7030812501907349, 0.7843137383460999, 0.5686274766921997, 0.7731092572212219, 0.9551820755004883], [0.02521008439362049, 0.03921568766236305, 0.07002801448106766, 0.06442577391862869, 0.05602240934967995, 0.053221289068460464, 0.08123249560594559, 0.08963585644960403, 0.0784313753247261, 0.08403361588716507, 0.0784313753247261, 0.08403361588716507, 0.08123249560594559, 0.12885154783725739, 0.11344537883996964, 0.11484593898057938, 0.0784313753247261, 0.12324929982423782, 0.14845938980579376, 0.11484593898057938, 0.1428571492433548, 0.1806722730398178, 0.08403361588716507, 0.17927171289920807, 0.17927171289920807, 0.1260504275560379, 0.17927171289920807, 0.22268907725811005, 0.19607843458652496, 0.15406163036823273, 0.2240896373987198, 0.21568627655506134, 0.16806723177433014, 0.13725490868091583, 0.21008403599262238, 0.12745098769664764, 0.20168067514896393, 0.15126051008701324, 0.23249299824237823, 0.21288515627384186, 0.21568627655506134, 0.1498599499464035, 0.2072829157114029, 0.2380952388048172, 0.27450981736183167, 0.2969187796115875, 0.2829131782054901, 0.23249299824237823, 0.27731093764305115, 0.16526611149311066, 0.2717086970806122, 0.3025210201740265, 0.2717086970806122, 0.38655462861061096, 0.30812326073646545, 0.37535014748573303, 0.2549019753932953, 0.46498599648475647, 0.299719899892807, 0.3473389446735382, 0.5098039507865906, 0.41736695170402527, 0.38655462861061096, 0.523809552192688, 0.3851540684700012, 0.43417367339134216, 0.462184876203537, 0.5014005899429321, 0.38935574889183044, NaN, 0.5966386795043945, 0.6246498823165894, 0.6400560140609741, 0.6694678068161011, 0.6918767690658569, 0.5490196347236633, 0.7647058963775635, 0.9579831957817078], [0.05602240934967995, 0.07282913476228714, 0.06862745434045792, 0.09103641659021378, 0.06722689419984818, 0.08963585644960403, 0.10504201799631119, 0.09383753687143326, 0.08123249560594559, 0.08403361588716507, 0.13725490868091583, 0.15406163036823273, 0.1358543485403061, 0.12745098769664764, 0.13165266811847687, 0.095238097012043, 0.18207283318042755, 0.1428571492433548, 0.12324929982423782, 0.17086835205554962, 0.17086835205554962, 0.18487395346164703, 0.14565826952457428, 0.16806723177433014, 0.190476194024086, 0.16526611149311066, 0.22689075767993927, 0.22969187796115875, 0.18487395346164703, 0.21708683669567108, 0.24929971992969513, 0.190476194024086, 0.2689075767993927, 0.21568627655506134, 0.22969187796115875, 0.2380952388048172, 0.2072829157114029, 0.24929971992969513, 0.21848739683628082, 0.24369747936725616, 0.24089635908603668, 0.2521008551120758, 0.20168067514896393, 0.23249299824237823, 0.3137255012989044, 0.2521008551120758, 0.3165266215801239, 0.2857142984867096, 0.2829131782054901, 0.26330533623695374, 0.31092438101768494, 0.3333333432674408, 0.29131653904914856, 0.41736695170402527, 0.2717086970806122, 0.3641456663608551, 0.33893558382987976, 0.41736695170402527, 0.3445378243923187, 0.3473389446735382, 0.28011205792427063, 0.3949579894542694, 0.37254902720451355, 0.3669467866420746, 0.3669467866420746, 0.38655462861061096, 0.41736695170402527, 0.45098039507865906, 0.39915966987609863, 0.40336135029792786, NaN, 0.529411792755127, 0.49859943985939026, 0.5014005899429321, 0.5042017102241516, 0.5140056014060974, 0.5966386795043945, 0.8235294222831726], [0.019607843831181526, 0.03641456738114357, 0.03921568766236305, 0.07002801448106766, 0.04481792822480202, 0.05602240934967995, 0.07002801448106766, 0.07282913476228714, 0.07002801448106766, 0.06442577391862869, 0.07282913476228714, 0.08123249560594559, 0.0784313753247261, 0.10924369841814041, 0.08683473616838455, 0.17927171289920807, 0.09243697673082352, 0.11764705926179886, 0.1428571492433548, 0.08263305574655533, 0.12044817954301834, 0.15406163036823273, 0.1764705926179886, 0.16806723177433014, 0.15126051008701324, 0.19187675416469574, 0.1568627506494522, 0.16526611149311066, 0.12324929982423782, 0.08123249560594559, 0.14565826952457428, 0.24089635908603668, 0.10924369841814041, 0.2072829157114029, 0.1764705926179886, 0.21008403599262238, 0.2380952388048172, 0.21848739683628082, 0.190476194024086, 0.2549019753932953, 0.27450981736183167, 0.20168067514896393, 0.27731093764305115, 0.23529411852359772, 0.28011205792427063, 0.2240896373987198, 0.2885154187679291, 0.24929971992969513, 0.28011205792427063, 0.21288515627384186, 0.28011205792427063, 0.23529411852359772, 0.27731093764305115, 0.228291317820549, 0.2661064565181732, 0.35854342579841614, 0.3361344635486603, 0.3669467866420746, 0.3669467866420746, 0.38655462861061096, 0.43977591395378113, 0.35574230551719666, 0.3193277418613434, 0.4425770342350006, 0.31092438101768494, 0.3977591097354889, 0.37955182790756226, 0.4145658314228058, 0.35854342579841614, 0.37535014748573303, 0.47058823704719543, NaN, 0.456582635641098, 0.4789915978908539, 0.4593837559223175, 0.5098039507865906, 0.5434173941612244, 0.8655462265014648], [0.02240896411240101, 0.05882352963089943, 0.053221289068460464, 0.08683473616838455, 0.08683473616838455, 0.07563025504350662, 0.08683473616838455, 0.08963585644960403, 0.10644257813692093, 0.08123249560594559, 0.09803921729326248, 0.1260504275560379, 0.10644257813692093, 0.12885154783725739, 0.10364145785570145, 0.08683473616838455, 0.11484593898057938, 0.13165266811847687, 0.13725490868091583, 0.12464985996484756, 0.1260504275560379, 0.16246499121189117, 0.11764705926179886, 0.15126051008701324, 0.11484593898057938, 0.14565826952457428, 0.18207283318042755, 0.2212885171175003, 0.15406163036823273, 0.1498599499464035, 0.2212885171175003, 0.14005602896213531, 0.190476194024086, 0.13725490868091583, 0.1428571492433548, 0.13725490868091583, 0.13725490868091583, 0.1428571492433548, 0.190476194024086, 0.15406163036823273, 0.1764705926179886, 0.19327731430530548, 0.13725490868091583, 0.1568627506494522, 0.21288515627384186, 0.2689075767993927, 0.22689075767993927, 0.22969187796115875, 0.19327731430530548, 0.19187675416469574, 0.26050421595573425, 0.26330533623695374, 0.26750701665878296, 0.3361344635486603, 0.23949579894542694, 0.3137255012989044, 0.19887955486774445, 0.3333333432674408, 0.24369747936725616, 0.2661064565181732, 0.37535014748573303, 0.3403361439704895, 0.3193277418613434, 0.4215686321258545, 0.3193277418613434, 0.3193277418613434, 0.36974790692329407, 0.33893558382987976, 0.33893558382987976, 0.3599439859390259, 0.5014005899429321, 0.5434173941612244, NaN, 0.4593837559223175, 0.47058823704719543, 0.47058823704719543, 0.8109243512153625, 0.9607843160629272], [0.016806723549962044, 0.03641456738114357, 0.05882352963089943, 0.06162464991211891, 0.06442577391862869, 0.06722689419984818, 0.06722689419984818, 0.07563025504350662, 0.095238097012043, 0.06582633405923843, 0.05882352963089943, 0.07563025504350662, 0.05882352963089943, 0.12044817954301834, 0.07563025504350662, 0.13165266811847687, 0.07563025504350662, 0.095238097012043, 0.15126051008701324, 0.08683473616838455, 0.1260504275560379, 0.1568627506494522, 0.09243697673082352, 0.13725490868091583, 0.12044817954301834, 0.1120448186993599, 0.18207283318042755, 0.17927171289920807, 0.15406163036823273, 0.09243697673082352, 0.1764705926179886, 0.18907563388347626, 0.14565826952457428, 0.14845938980579376, 0.18207283318042755, 0.13165266811847687, 0.15406163036823273, 0.14005602896213531, 0.2380952388048172, 0.15126051008701324, 0.18767507374286652, 0.15126051008701324, 0.16246499121189117, 0.19607843458652496, 0.190476194024086, 0.2380952388048172, 0.19887955486774445, 0.19887955486774445, 0.24089635908603668, 0.15406163036823273, 0.21008403599262238, 0.27450981736183167, 0.24369747936725616, 0.2829131782054901, 0.27731093764305115, 0.33893558382987976, 0.15126051008701324, 0.3529411852359772, 0.2240896373987198, 0.3305322229862213, 0.4089635908603668, 0.3165266215801239, 0.28011205792427063, 0.4313725531101227, 0.29131653904914856, 0.3781512677669525, 0.33893558382987976, 0.3837535083293915, 0.2969187796115875, 0.3305322229862213, 0.49859943985939026, 0.5210084319114685, 0.5406162738800049, NaN, 0.5406162738800049, 0.43697479367256165, 0.6470588445663452, 0.9467787146568298], [0.02521008439362049, 0.030812324956059456, 0.053221289068460464, 0.05882352963089943, 0.0476190485060215, 0.0476190485060215, 0.06722689419984818, 0.07002801448106766, 0.06862745434045792, 0.06722689419984818, 0.05602240934967995, 0.06442577391862869, 0.06302521377801895, 0.10644257813692093, 0.0784313753247261, 0.09803921729326248, 0.0476190485060215, 0.095238097012043, 0.14145658910274506, 0.08963585644960403, 0.1120448186993599, 0.16806723177433014, 0.05882352963089943, 0.12885154783725739, 0.11764705926179886, 0.08963585644960403, 0.1428571492433548, 0.17086835205554962, 0.14005602896213531, 0.07563025504350662, 0.18207283318042755, 0.19887955486774445, 0.11484593898057938, 0.12044817954301834, 0.16246499121189117, 0.08683473616838455, 0.14565826952457428, 0.09803921729326248, 0.2212885171175003, 0.1568627506494522, 0.16246499121189117, 0.09803921729326248, 0.1736694723367691, 0.19327731430530548, 0.19607843458652496, 0.2717086970806122, 0.20168067514896393, 0.190476194024086, 0.22689075767993927, 0.12324929982423782, 0.2240896373987198, 0.24649859964847565, 0.23529411852359772, 0.2857142984867096, 0.2689075767993927, 0.3025210201740265, 0.19607843458652496, 0.3949579894542694, 0.23389355838298798, 0.29411765933036804, 0.4453781545162201, 0.2969187796115875, 0.26050421595573425, 0.4593837559223175, 0.2759103775024414, 0.3837535083293915, 0.32773110270500183, 0.4019607901573181, 0.21568627655506134, 0.30812326073646545, 0.4957983195781708, 0.5406162738800049, 0.529411792755127, 0.4593837559223175, NaN, 0.4453781545162201, 0.6386554837226868, 0.9271708726882935], [0.014005602337419987, 0.03641456738114357, 0.02521008439362049, 0.04481792822480202, 0.03641456738114357, 0.0476190485060215, 0.0476190485060215, 0.04481792822480202, 0.04481792822480202, 0.042016807943582535, 0.05882352963089943, 0.07563025504350662, 0.0476190485060215, 0.09243697673082352, 0.06722689419984818, 0.07282913476228714, 0.042016807943582535, 0.042016807943582535, 0.10644257813692093, 0.0784313753247261, 0.08123249560594559, 0.13165266811847687, 0.06722689419984818, 0.053221289068460464, 0.1260504275560379, 0.07002801448106766, 0.06442577391862869, 0.18207283318042755, 0.10364145785570145, 0.08123249560594559, 0.1428571492433548, 0.1596638709306717, 0.1120448186993599, 0.13165266811847687, 0.07563025504350662, 0.14005602896213531, 0.11764705926179886, 0.12885154783725739, 0.11484593898057938, 0.13725490868091583, 0.16806723177433014, 0.15126051008701324, 0.1568627506494522, 0.1120448186993599, 0.3137255012989044, 0.16526611149311066, 0.3165266215801239, 0.15406163036823273, 0.1596638709306717, 0.1568627506494522, 0.16386555135250092, 0.2044817954301834, 0.2072829157114029, 0.27731093764305115, 0.18767507374286652, 0.29411765933036804, 0.3361344635486603, 0.4117647111415863, 0.35854342579841614, 0.38935574889183044, 0.40616247057914734, 0.23529411852359772, 0.35854342579841614, 0.4089635908603668, 0.32773110270500183, 0.32492998242378235, 0.30112046003341675, 0.32212886214256287, 0.4313725531101227, 0.45098039507865906, 0.4859943985939026, 0.4901960790157318, 0.529411792755127, 0.5630252361297607, 0.5546218752861023, NaN, 0.6302521228790283, 0.8991596698760986], [0.03361344709992409, 0.04481792822480202, 0.05042016878724098, 0.06722689419984818, 0.06722689419984818, 0.06442577391862869, 0.08403361588716507, 0.08123249560594559, 0.07703081518411636, 0.05882352963089943, 0.09243697673082352, 0.10084033757448196, 0.07563025504350662, 0.06722689419984818, 0.10084033757448196, 0.06162464991211891, 0.09243697673082352, 0.1120448186993599, 0.11764705926179886, 0.1120448186993599, 0.12044817954301834, 0.14845938980579376, 0.09663865715265274, 0.1428571492433548, 0.08963585644960403, 0.1120448186993599, 0.1568627506494522, 0.14845938980579376, 0.13165266811847687, 0.13725490868091583, 0.1764705926179886, 0.12885154783725739, 0.14705882966518402, 0.10644257813692093, 0.1260504275560379, 0.1260504275560379, 0.12464985996484756, 0.11764705926179886, 0.15406163036823273, 0.12324929982423782, 0.13445378839969635, 0.14425770938396454, 0.1120448186993599, 0.1568627506494522, 0.15126051008701324, 0.2212885171175003, 0.16526611149311066, 0.16806723177433014, 0.16806723177433014, 0.14845938980579376, 0.2044817954301834, 0.20868347585201263, 0.18207283318042755, 0.2969187796115875, 0.23249299824237823, 0.26330533623695374, 0.16806723177433014, 0.26050421595573425, 0.16246499121189117, 0.19327731430530548, 0.299719899892807, 0.27731093764305115, 0.26330533623695374, 0.32212886214256287, 0.2521008551120758, 0.2689075767993927, 0.30532214045524597, 0.2717086970806122, 0.22689075767993927, 0.23529411852359772, 0.40336135029792786, 0.456582635641098, 0.18907563388347626, 0.3529411852359772, 0.3613445460796356, 0.36974790692329407, NaN, 0.9411764740943909], [0.008403361774981022, 0.005602241028100252, 0.005602241028100252, 0.008403361774981022, 0.008403361774981022, 0.011204482056200504, 0.011204482056200504, 0.011204482056200504, 0.014005602337419987, 0.008403361774981022, 0.008403361774981022, 0.014005602337419987, 0.011204482056200504, 0.008403361774981022, 0.005602241028100252, 0.019607843831181526, 0.008403361774981022, 0.02240896411240101, 0.028011204674839973, 0.008403361774981022, 0.014005602337419987, 0.019607843831181526, 0.011204482056200504, 0.03361344709992409, 0.011204482056200504, 0.008403361774981022, 0.02521008439362049, 0.02521008439362049, 0.014005602337419987, 0.014005602337419987, 0.028011204674839973, 0.02240896411240101, 0.030812324956059456, 0.016806723549962044, 0.030812324956059456, 0.011204482056200504, 0.028011204674839973, 0.011204482056200504, 0.03361344709992409, 0.028011204674839973, 0.03921568766236305, 0.02521008439362049, 0.030812324956059456, 0.030812324956059456, 0.05882352963089943, 0.03361344709992409, 0.05882352963089943, 0.03501400724053383, 0.042016807943582535, 0.030812324956059456, 0.05042016878724098, 0.05042016878724098, 0.042016807943582535, 0.08403361588716507, 0.0476190485060215, 0.08123249560594559, 0.014005602337419987, 0.06442577391862869, 0.016806723549962044, 0.0784313753247261, 0.10644257813692093, 0.05882352963089943, 0.07563025504350662, 0.11764705926179886, 0.05182072892785072, 0.07282913476228714, 0.09243697673082352, 0.07002801448106766, 0.04481792822480202, 0.042016807943582535, 0.1764705926179886, 0.13445378839969635, 0.03921568766236305, 0.053221289068460464, 0.07282913476228714, 0.10084033757448196, 0.05882352963089943, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9027792982},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8376987207},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8255302176},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8034486587},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8006111618},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.7854232686},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7811488306},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7655425875},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.7353123146},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7287278642},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7028811592},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7018625728},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6784713983},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6600640379},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6533522601},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6532249375},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6524064263},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6488595521},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.6450398466},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.643821173},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.627432803},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.6273054794},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6005856986},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.5962203169},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5952562975},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5689548621},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.5607515842},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.558987245},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5518389281},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5466550359},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5375059252},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5337407815},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5323765973},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5240278048},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5229182691},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5218996797},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5166612207},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5158972786},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4929790192},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4877405603},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4854487362},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4840481736},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4827567491},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4822292668},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4815198903},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.4755902467},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.4741533077},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4736803921},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4609844037},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.4598566785},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4447961092},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4295536481},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.426206856},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4210775325},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4198952389},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.396795092},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3761504687},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.3755320417},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3671650572},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3391174749},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.3370802958},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3234566643},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3198188455},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.3131070681},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3105060287},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3068682107},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2938630064},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2905707807},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2851504306},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.2644148638},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.2635054089},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2372949235},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2266906817},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2174324334},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2077922127},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2038088017},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1781258007},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.038015207}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_binary/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_binary/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..a0e7e44d0cdd8fa22c4dbee9c4aff67c646249b8
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_binary/datasets_small/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),TabPFN-3 (default),TabICLv2 (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),Mitra (default),TabDPT-Turbo (default),TabPFNv2 (tuned + ensembled),LimiX (default),TabICL (default),TabSwift (default),"AutoGluon 1.4 (best, 4h)",BetaTabPFN (default),RealMLP (tuned + ensembled),TabM (tuned + ensembled),TabPFNv2 (default),TabDPT (tuned + ensembled),TabDPT (tuned),TabPFNv2 (tuned),LightGBM (tuned + ensembled),TabM (tuned),TabDPT (default),LightGBM (tuned),RealMLP (tuned),SAP-RPT-OSS (default),xRFM (tuned),xRFM (tuned + ensembled),ModernNCA (tuned),ModernNCA (default),iLTM (tuned + ensembled),CatBoost (default),FastaiMLP (tuned + ensembled),CatBoost (tuned + ensembled),EBM (tuned + ensembled),CatBoost (tuned),TorchMLP (tuned + ensembled),EBM (tuned),TabM (default),XGBoost (tuned + ensembled),EBM (default),RealMLP (default),TabSTAR (tuned),OrionMSP (default),TabSTAR (tuned + ensembled),ChimeraBoost (tuned + ensembled),FastaiMLP (tuned),XGBoost (tuned),ChimeraBoost (tuned),iLTM (tuned),ChimeraBoost (default),ModernNCA (tuned + ensembled),TorchMLP (tuned),TabFlex (default),ExtraTrees (tuned + ensembled),TabSTAR (default),ExtraTrees (tuned),iLTM (default),Linear (tuned + ensembled),PerpetualBooster (tuned + ensembled),XGBoost (default),Linear (tuned),LightGBM (default),TorchMLP (default),PerpetualBooster (tuned),FastaiMLP (default),RandomForest (tuned + ensembled),RandomForest (tuned),Linear (default),xRFM (default),KNN (tuned + ensembled),ExtraTrees (default),RandomForest (default),PerpetualBooster (default),KNN (tuned),KNN (default)
+TabFM (default),,0.77310926,0.789916,0.7759104,0.7787115,0.789916,0.79971987,0.8067227,0.82633054,0.8095238,0.8767507,0.89495796,0.9159664,0.8487395,0.89635855,0.83613443,0.8935574,0.8879552,0.7955182,0.88235295,0.8767507,0.8487395,0.89915967,0.92436975,0.88515407,0.89915967,0.88515407,0.87114847,0.8795518,0.91036415,0.90756303,0.8739496,0.92156863,0.92436975,0.89635855,0.92436975,0.8907563,0.92436975,0.95238096,0.89915967,0.92156863,0.95238096,0.8907563,0.8935574,0.7787115,0.929972,0.7815126,0.929972,0.9159664,0.9551821,0.9439776,0.93557423,0.94957983,0.9551821,0.9607843,0.91036415,0.9439776,0.8207283,0.9327731,0.9271709,0.9131653,0.97759104,0.9691877,0.92436975,0.96638656,0.9719888,0.98039216,0.95238096,0.97759104,0.9747899,0.9439776,0.98039216,0.97759104,0.9831933,0.9747899,0.9859944,0.96638656,0.99159664
+TabPFN-3 (default),0.22689076,,0.50700283,0.60784316,0.6666667,0.6386555,0.6582633,0.6862745,0.73109245,0.6862745,0.7507003,0.7478992,0.802521,0.80112046,0.7619048,0.7352941,0.8067227,0.79831934,0.72969186,0.8067227,0.8235294,0.8165266,0.8039216,0.837535,0.8207283,0.8319328,0.82913166,0.8095238,0.8235294,0.859944,0.859944,0.82212883,0.8879552,0.87114847,0.8305322,0.859944,0.84313726,0.8739496,0.9047619,0.85434175,0.86834735,0.87815124,0.8445378,0.86834735,0.7352941,0.8795518,0.7507003,0.9019608,0.8613445,0.89915967,0.91036415,0.9173669,0.89915967,0.9187675,0.9271709,0.88235295,0.9327731,0.7927171,0.9271709,0.9159664,0.87114847,0.92156863,0.94957983,0.88235295,0.9467787,0.9439776,0.9411765,0.9439776,0.93837535,0.9607843,0.9271709,0.96358544,0.9411765,0.96358544,0.9691877,0.96358544,0.9551821,0.99439776
+TabICLv2 (default),0.21008404,0.4929972,,0.5854342,0.5868347,0.627451,0.60784316,0.6246499,0.6484594,0.6862745,0.74229693,0.75770307,0.79971987,0.78711486,0.7955182,0.74229693,0.7927171,0.80112046,0.7086835,0.79691875,0.78431374,0.7605042,0.789916,0.8179272,0.81512606,0.81512606,0.81232494,0.7759104,0.8137255,0.8515406,0.8459384,0.8277311,0.8655462,0.8487395,0.84313726,0.8347339,0.8347339,0.8459384,0.89915967,0.8319328,0.84313726,0.8627451,0.84033614,0.8627451,0.7507003,0.90756303,0.7478992,0.8795518,0.87114847,0.8795518,0.9019608,0.8879552,0.8907563,0.9131653,0.92156863,0.8739496,0.89495796,0.79831934,0.8935574,0.90756303,0.8907563,0.929972,0.9271709,0.9019608,0.9411765,0.95238096,0.9411765,0.93557423,0.9327731,0.929972,0.9313725,0.9607843,0.9467787,0.9411765,0.9467787,0.9747899,0.94957983,0.99439776
+"AutoGluon 1.5 (extreme, 4h)",0.22408964,0.39215687,0.41456583,,0.56582636,0.6134454,0.59103644,0.60784316,0.6204482,0.5406163,0.69187677,0.69187677,0.75770307,0.72829133,0.789916,0.70308125,0.780112,0.80112046,0.6834734,0.74509805,0.75910366,0.74229693,0.8095238,0.81092435,0.73109245,0.8095238,0.8207283,0.75910366,0.78571427,0.84313726,0.84313726,0.79831934,0.8627451,0.8333333,0.8235294,0.84033614,0.8067227,0.8305322,0.8795518,0.81232494,0.8515406,0.8627451,0.8235294,0.8207283,0.68907565,0.8837535,0.69187677,0.88235295,0.8347339,0.87114847,0.8907563,0.89635855,0.8879552,0.9271709,0.90756303,0.85434175,0.90756303,0.75630254,0.8879552,0.9047619,0.8459384,0.92436975,0.9271709,0.8795518,0.93557423,0.93557423,0.9397759,0.92436975,0.929972,0.93557423,0.90896356,0.929972,0.9131653,0.93837535,0.9411765,0.9551821,0.9327731,0.99159664
+RealTabPFN-2.5 (default),0.22128852,0.33333334,0.41316527,0.43417367,,0.52100843,0.5182073,0.5602241,0.6386555,0.54901963,0.6806723,0.6722689,0.6834734,0.73109245,0.7394958,0.7212885,0.7394958,0.7478992,0.719888,0.7058824,0.7072829,0.75630254,0.789916,0.7955182,0.71708685,0.81232494,0.8179272,0.7815126,0.8179272,0.82913166,0.81232494,0.81232494,0.82913166,0.87114847,0.8459384,0.84033614,0.84033614,0.82913166,0.88235295,0.8655462,0.86834735,0.8515406,0.8347339,0.859944,0.72829133,0.8935574,0.732493,0.89635855,0.87114847,0.8795518,0.9019608,0.90616244,0.9047619,0.8739496,0.92436975,0.88095236,0.8879552,0.77030814,0.8879552,0.91036415,0.859944,0.92156863,0.92436975,0.8739496,0.9579832,0.96358544,0.93557423,0.9271709,0.9131653,0.9439776,0.9327731,0.9551821,0.9131653,0.93557423,0.95238096,0.96358544,0.9327731,0.99159664
+TabPFN-2.6 (default),0.21008404,0.36134455,0.37254903,0.38655463,0.4789916,,0.4789916,0.51960784,0.6106443,0.55182076,0.6554622,0.6694678,0.7114846,0.6834734,0.7338936,0.6834734,0.7535014,0.7535014,0.6512605,0.6946779,0.6806723,0.7619048,0.7787115,0.7927171,0.6694678,0.81512606,0.8067227,0.77310926,0.7815126,0.8179272,0.82913166,0.8095238,0.859944,0.84313726,0.7759104,0.84033614,0.78711486,0.8235294,0.859944,0.81092435,0.8305322,0.84313726,0.7927171,0.8235294,0.6862745,0.8179272,0.70028013,0.89635855,0.8207283,0.859944,0.9047619,0.8879552,0.90756303,0.929972,0.9047619,0.8515406,0.89915967,0.74229693,0.8865546,0.8907563,0.8459384,0.9187675,0.9439776,0.8767507,0.94957983,0.93837535,0.9187675,0.91036415,0.9327731,0.9467787,0.91036415,0.9439776,0.92436975,0.9327731,0.95238096,0.95238096,0.93557423,0.9887955
+RealTabPFN-2.5 (tuned + ensembled),0.20028012,0.3417367,0.39215687,0.4089636,0.48179272,0.52100843,,0.59103644,0.6414566,0.5154062,0.697479,0.6946779,0.6680672,0.6694678,0.74229693,0.7114846,0.7366947,0.7394958,0.7058824,0.6694678,0.6806723,0.767507,0.7815126,0.7787115,0.6246499,0.802521,0.7759104,0.75770307,0.7815126,0.81512606,0.8347339,0.7927171,0.8193277,0.8319328,0.7941176,0.81512606,0.7955182,0.81232494,0.85434175,0.81092435,0.8179272,0.8347339,0.8039216,0.81232494,0.68907565,0.8515406,0.697479,0.86694676,0.84313726,0.8613445,0.88235295,0.8739496,0.88235295,0.8879552,0.89635855,0.8459384,0.8655462,0.74229693,0.85714287,0.8907563,0.859944,0.92436975,0.92156863,0.88235295,0.93837535,0.9187675,0.929972,0.9187675,0.9159664,0.9187675,0.89495796,0.929972,0.9131653,0.9327731,0.9327731,0.95238096,0.9159664,0.9887955
+RealTabPFN-2.5 (tuned),0.19327731,0.3137255,0.37535015,0.39215687,0.4397759,0.48039216,0.4089636,,0.6302521,0.5182073,0.65686274,0.64705884,0.63585436,0.65686274,0.71428573,0.697479,0.697479,0.71708685,0.6806723,0.66106445,0.6820728,0.7352941,0.7647059,0.7633053,0.6414566,0.7955182,0.7619048,0.7478992,0.7647059,0.78291315,0.78711486,0.77030814,0.7955182,0.8039216,0.7787115,0.80112046,0.8039216,0.7955182,0.8515406,0.8067227,0.8179272,0.8207283,0.789916,0.8095238,0.6806723,0.8459384,0.69187677,0.8473389,0.8235294,0.8319328,0.86834735,0.8655462,0.87114847,0.87815124,0.89635855,0.8319328,0.8613445,0.74509805,0.859944,0.88515407,0.84033614,0.9005602,0.9159664,0.859944,0.9327731,0.9271709,0.92156863,0.9047619,0.89915967,0.91036415,0.90616244,0.9271709,0.91036415,0.92436975,0.929972,0.9551821,0.9187675,0.9887955
+Mitra (default),0.17366947,0.26890758,0.35154063,0.37955183,0.36134455,0.38935575,0.35854343,0.3697479,,0.50980395,0.5686275,0.5028011,0.5840336,0.6022409,0.6442577,0.6386555,0.64705884,0.69187677,0.6260504,0.6022409,0.60504204,0.6344538,0.6834734,0.7787115,0.6442577,0.7394958,0.7507003,0.6946779,0.72969186,0.74229693,0.7507003,0.7535014,0.7521008,0.78431374,0.7815126,0.7535014,0.7927171,0.74369746,0.82633054,0.80112046,0.78711486,0.79691875,0.7955182,0.789916,0.6834734,0.7521008,0.6862745,0.8459384,0.8207283,0.8039216,0.8487395,0.837535,0.86834735,0.8179272,0.8935574,0.82633054,0.85434175,0.73109245,0.8557423,0.8879552,0.8515406,0.86834735,0.9159664,0.88235295,0.91036415,0.929972,0.89635855,0.91036415,0.8935574,0.92156863,0.9187675,0.929972,0.8935574,0.9047619,0.9313725,0.9551821,0.92296916,0.9859944
+TabDPT-Turbo (default),0.1904762,0.3137255,0.3137255,0.45938376,0.4509804,0.44817927,0.48459384,0.48179272,0.49019608,,0.53781515,0.5714286,0.5826331,0.6904762,0.60784316,0.5742297,0.6442577,0.61764705,0.57703084,0.67507005,0.6834734,0.59383756,0.6526611,0.6666667,0.7885154,0.67787117,0.71568626,0.6862745,0.73109245,0.719888,0.7058824,0.7394958,0.72829133,0.71428573,0.767507,0.7366947,0.73109245,0.72829133,0.75630254,0.7521008,0.7394958,0.75630254,0.7507003,0.77030814,0.6694678,0.82913166,0.6806723,0.767507,0.81232494,0.767507,0.789916,0.8305322,0.8095238,0.7619048,0.81512606,0.87114847,0.8697479,0.74229693,0.89635855,0.8319328,0.8319328,0.8627451,0.86834735,0.8515406,0.88235295,0.9019608,0.8907563,0.9159664,0.9131653,0.9159664,0.9159664,0.93557423,0.9187675,0.93417364,0.9327731,0.9579832,0.9411765,0.99159664
+TabPFNv2 (tuned + ensembled),0.1232493,0.24929972,0.2577031,0.30812326,0.31932774,0.34453782,0.30252102,0.34313726,0.43137255,0.46218488,,0.45658264,0.5322129,0.53501403,0.6442577,0.60364145,0.6232493,0.627451,0.6064426,0.5602241,0.56162465,0.6414566,0.6652661,0.68767506,0.5826331,0.6862745,0.71288514,0.61484593,0.72689074,0.74229693,0.7478992,0.7240896,0.7338936,0.7254902,0.72689074,0.71428573,0.7464986,0.71288514,0.767507,0.75910366,0.73109245,0.7661064,0.7507003,0.767507,0.6554622,0.71708685,0.66106445,0.79831934,0.7521008,0.78431374,0.8095238,0.7955182,0.8319328,0.8347339,0.8459384,0.7759104,0.8417367,0.7254902,0.84033614,0.84313726,0.81512606,0.859944,0.9019608,0.837535,0.9187675,0.89915967,0.89915967,0.9019608,0.91036415,0.92156863,0.8627451,0.9271709,0.9019608,0.9411765,0.9439776,0.9411765,0.90756303,0.99159664
+LimiX (default),0.10504202,0.25210086,0.24229692,0.30812326,0.3277311,0.33053222,0.30532214,0.3529412,0.49719888,0.42857143,0.5434174,,0.5546219,0.5266107,0.6302521,0.5742297,0.6442577,0.6246499,0.59103644,0.54761904,0.55182076,0.6134454,0.6862745,0.640056,0.57703084,0.697479,0.7044818,0.70308125,0.6820728,0.7254902,0.7254902,0.70028013,0.8165266,0.7366947,0.6960784,0.74229693,0.74509805,0.7394958,0.7787115,0.7619048,0.71708685,0.7787115,0.75630254,0.7394958,0.6764706,0.75770307,0.6722689,0.8095238,0.7521008,0.80112046,0.8319328,0.82633054,0.84033614,0.837535,0.81232494,0.7366947,0.8235294,0.732493,0.81232494,0.859944,0.8095238,0.8739496,0.8795518,0.837535,0.9019608,0.859944,0.8907563,0.8879552,0.8935574,0.9159664,0.8459384,0.9187675,0.8739496,0.92436975,0.93557423,0.92436975,0.89915967,0.9859944
+TabICL (default),0.084033616,0.197479,0.20028012,0.24229692,0.31652662,0.28851542,0.33193278,0.36414567,0.4159664,0.41736695,0.46778712,0.44537815,,0.5462185,0.5546219,0.5322129,0.5462185,0.5462185,0.5322129,0.5448179,0.5532213,0.53501403,0.61904764,0.6106443,0.5868347,0.64705884,0.6386555,0.6554622,0.6414566,0.6946779,0.7058824,0.68487394,0.732493,0.7226891,0.7058824,0.69187677,0.6694678,0.70308125,0.7394958,0.6862745,0.719888,0.7478992,0.69187677,0.7478992,0.6526611,0.7507003,0.6554622,0.767507,0.75630254,0.7647059,0.789916,0.81512606,0.7955182,0.81232494,0.8039216,0.7759104,0.8459384,0.7226891,0.8305322,0.82633054,0.7927171,0.87114847,0.8907563,0.81512606,0.90896356,0.8907563,0.8879552,0.8865546,0.88515407,0.9187675,0.86414564,0.92156863,0.8935574,0.9411765,0.93697476,0.95238096,0.92436975,0.9887955
+TabSwift (default),0.15126051,0.19887955,0.21288516,0.2717087,0.26890758,0.31652662,0.33053222,0.34313726,0.3977591,0.30952382,0.464986,0.47338936,0.45378152,,0.5532213,0.49019608,0.57703084,0.52100843,0.5294118,0.5112045,0.53501403,0.5462185,0.57703084,0.56302524,0.56302524,0.5826331,0.6414566,0.6302521,0.63585436,0.6666667,0.6722689,0.66386557,0.67507005,0.6442577,0.7338936,0.64985996,0.6694678,0.6652661,0.72829133,0.6834734,0.6722689,0.6722689,0.67787117,0.7114846,0.69187677,0.8095238,0.7058824,0.6862745,0.767507,0.7114846,0.7254902,0.7955182,0.77310926,0.7338936,0.7647059,0.8627451,0.82633054,0.75910366,0.8207283,0.80112046,0.81232494,0.8459384,0.8179272,0.8347339,0.83613443,0.90756303,0.84033614,0.8935574,0.85434175,0.87114847,0.872549,0.8907563,0.87114847,0.8795518,0.8935574,0.90756303,0.9327731,0.99159664
+"AutoGluon 1.4 (best, 4h)",0.10364146,0.23809524,0.2044818,0.21008404,0.26050422,0.26610646,0.2577031,0.2857143,0.3557423,0.39215687,0.3557423,0.3697479,0.44537815,0.4467787,,0.5126051,0.5392157,0.52380955,0.46218488,0.49019608,0.4789916,0.46218488,0.64985996,0.59103644,0.5266107,0.6834734,0.64705884,0.53781515,0.64985996,0.6372549,0.60504204,0.6386555,0.6862745,0.71428573,0.6834734,0.719888,0.67787117,0.72829133,0.75630254,0.6946779,0.7058824,0.767507,0.7226891,0.7338936,0.63585436,0.62885153,0.64705884,0.74369746,0.7338936,0.7619048,0.7619048,0.78431374,0.7759104,0.71428573,0.7955182,0.7394958,0.75630254,0.7058824,0.7647059,0.8179272,0.79971987,0.9047619,0.8935574,0.8333333,0.9187675,0.8627451,0.91036415,0.8907563,0.87114847,0.8865546,0.86834735,0.9131653,0.89635855,0.92436975,0.92156863,0.9327731,0.89915967,0.99439776
+BetaTabPFN (default),0.16386555,0.2647059,0.2577031,0.29691878,0.2787115,0.31652662,0.28851542,0.30252102,0.36134455,0.4257703,0.39635855,0.4257703,0.46778712,0.50980395,0.48739496,,0.5154062,0.4789916,0.47338936,0.52380955,0.5322129,0.44537815,0.64705884,0.557423,0.5882353,0.63585436,0.5826331,0.57703084,0.5994398,0.6246499,0.6372549,0.63305324,0.6386555,0.7058824,0.6260504,0.66386557,0.7254902,0.6960784,0.6694678,0.7366947,0.7338936,0.7478992,0.7647059,0.69187677,0.6694678,0.6526611,0.67507005,0.74089634,0.70308125,0.767507,0.767507,0.70308125,0.78431374,0.6862745,0.7478992,0.71568626,0.8179272,0.74089634,0.81232494,0.859944,0.85434175,0.8235294,0.8347339,0.88235295,0.8557423,0.82633054,0.84313726,0.8039216,0.8935574,0.88515407,0.9047619,0.8207283,0.9131653,0.86834735,0.9019608,0.9271709,0.93837535,0.98039216
+RealMLP (tuned + ensembled),0.10644258,0.19327731,0.20728292,0.21988796,0.26050422,0.2464986,0.26330534,0.30252102,0.3529412,0.3557423,0.3767507,0.3557423,0.45378152,0.4229692,0.46078432,0.48459384,,0.5126051,0.43137255,0.46078432,0.46918768,0.45938376,0.627451,0.59103644,0.52380955,0.67086834,0.7254902,0.54901963,0.6134454,0.6554622,0.6134454,0.6134454,0.71428573,0.7058824,0.6960784,0.7394958,0.7114846,0.73109245,0.70308125,0.719888,0.7086835,0.77310926,0.71568626,0.78571427,0.6246499,0.6442577,0.6386555,0.77030814,0.72689074,0.79831934,0.78571427,0.77310926,0.78711486,0.7058824,0.75630254,0.7114846,0.80112046,0.69187677,0.7941176,0.8459384,0.78711486,0.89915967,0.88235295,0.7927171,0.8907563,0.87114847,0.91036415,0.89915967,0.89635855,0.92156863,0.8179272,0.90756303,0.88515407,0.92436975,0.95238096,0.9579832,0.90756303,0.99159664
+TabM (tuned + ensembled),0.11204482,0.20168068,0.19887955,0.19887955,0.25210086,0.2464986,0.26050422,0.28291318,0.30812326,0.38235295,0.37254903,0.37535015,0.45378152,0.4789916,0.47619048,0.52100843,0.48739496,,0.45238096,0.49439776,0.49019608,0.44257703,0.63305324,0.662465,0.52380955,0.662465,0.6386555,0.5686275,0.5742297,0.6022409,0.6260504,0.63305324,0.662465,0.71708685,0.6442577,0.70028013,0.719888,0.7086835,0.6946779,0.72829133,0.79971987,0.75910366,0.7184874,0.7212885,0.6386555,0.6246499,0.6442577,0.7787115,0.697479,0.80112046,0.82633054,0.7535014,0.8319328,0.7366947,0.76890755,0.71428573,0.7927171,0.697479,0.7647059,0.8319328,0.81232494,0.86834735,0.8795518,0.8277311,0.8837535,0.87114847,0.8907563,0.86834735,0.8515406,0.8767507,0.85714287,0.88235295,0.86834735,0.9047619,0.9047619,0.9579832,0.8879552,0.97759104
+TabPFNv2 (default),0.2044818,0.27030814,0.29131654,0.31652662,0.28011206,0.3487395,0.29411766,0.31932774,0.3739496,0.4229692,0.39355743,0.4089636,0.46778712,0.47058824,0.53781515,0.5266107,0.5686275,0.54761904,,0.5322129,0.5532213,0.48459384,0.627451,0.57983196,0.5602241,0.6386555,0.62184876,0.5826331,0.6442577,0.66106445,0.6162465,0.640056,0.6134454,0.6512605,0.71428573,0.66106445,0.6694678,0.6582633,0.67787117,0.6834734,0.697479,0.6722689,0.69187677,0.7226891,0.67787117,0.6862745,0.67366946,0.6666667,0.7226891,0.69187677,0.6862745,0.69187677,0.67787117,0.69187677,0.7338936,0.82633054,0.7619048,0.74509805,0.77310926,0.80112046,0.8277311,0.79831934,0.767507,0.8347339,0.8095238,0.8515406,0.8039216,0.84033614,0.8235294,0.8515406,0.8767507,0.85714287,0.8627451,0.8487395,0.8585434,0.8935574,0.88235295,0.9719888
+TabDPT (tuned + ensembled),0.11764706,0.19327731,0.20308124,0.25490198,0.29411766,0.30532214,0.33053222,0.33893558,0.3977591,0.32492998,0.4397759,0.45238096,0.45518208,0.48879552,0.50980395,0.47619048,0.5392157,0.50560224,0.46778712,,0.5854342,0.55042017,0.5462185,0.5560224,0.5826331,0.5994398,0.6372549,0.6246499,0.61764705,0.64285713,0.63305324,0.640056,0.6694678,0.63585436,0.6666667,0.6442577,0.6372549,0.6414566,0.7044818,0.6652661,0.64565825,0.65966386,0.64985996,0.67507005,0.6134454,0.7366947,0.6162465,0.68907565,0.7254902,0.70308125,0.719888,0.7759104,0.74509805,0.7619048,0.7394958,0.75910366,0.8039216,0.7058824,0.8039216,0.8235294,0.75630254,0.82913166,0.8207283,0.7815126,0.8487395,0.859944,0.85014004,0.88515407,0.88235295,0.88515407,0.82913166,0.9173669,0.8753501,0.9131653,0.91036415,0.92156863,0.8879552,0.99159664
+TabDPT (tuned),0.1232493,0.1764706,0.21568628,0.24089636,0.2927171,0.31932774,0.31932774,0.31792718,0.394958,0.31652662,0.43837535,0.44817927,0.4467787,0.464986,0.52100843,0.46778712,0.5308123,0.50980395,0.4467787,0.41456583,,0.52240896,0.53781515,0.5546219,0.56442577,0.56582636,0.5882353,0.5882353,0.57703084,0.6106443,0.60504204,0.63305324,0.63585436,0.6246499,0.6582633,0.63305324,0.6554622,0.6260504,0.7058824,0.6666667,0.63585436,0.65686274,0.66386557,0.6834734,0.5980392,0.7507003,0.5966387,0.67507005,0.7058824,0.6722689,0.6946779,0.7507003,0.7226891,0.7380952,0.7254902,0.7535014,0.7619048,0.70028013,0.77030814,0.789916,0.7605042,0.8095238,0.78711486,0.78431374,0.8039216,0.8487395,0.8207283,0.8655462,0.84033614,0.85714287,0.82913166,0.8795518,0.8739496,0.8739496,0.8879552,0.9187675,0.8795518,0.9859944
+TabPFNv2 (tuned),0.15126051,0.1834734,0.2394958,0.2577031,0.24369748,0.23809524,0.232493,0.2647059,0.36554623,0.40616247,0.35854343,0.38655463,0.464986,0.45378152,0.53781515,0.5546219,0.5406163,0.557423,0.5154062,0.44957983,0.47759104,,0.60504204,0.6302521,0.5126051,0.62184876,0.63305324,0.5980392,0.5826331,0.627451,0.6442577,0.6512605,0.5882353,0.6694678,0.65686274,0.66386557,0.6554622,0.6806723,0.6862745,0.67787117,0.68907565,0.6960784,0.67507005,0.6960784,0.6386555,0.62184876,0.63585436,0.7254902,0.6806723,0.70308125,0.7478992,0.7184874,0.77310926,0.719888,0.7535014,0.73109245,0.73109245,0.7058824,0.77310926,0.8067227,0.77310926,0.7815126,0.77030814,0.789916,0.8179272,0.8067227,0.7885154,0.8095238,0.78431374,0.8193277,0.81512606,0.8459384,0.837535,0.84313726,0.8319328,0.86834735,0.8515406,0.98039216
+LightGBM (tuned + ensembled),0.10084034,0.19607843,0.21008404,0.1904762,0.21008404,0.22128852,0.2184874,0.23529412,0.31652662,0.34733894,0.3347339,0.3137255,0.3809524,0.4229692,0.35014006,0.3529412,0.37254903,0.3669468,0.37254903,0.45378152,0.46218488,0.394958,,0.4439776,0.5126051,0.5966387,0.50980395,0.50700283,0.56302524,0.557423,0.53781515,0.5322129,0.58963585,0.6764706,0.5686275,0.66386557,0.61764705,0.65686274,0.5854342,0.64985996,0.6764706,0.789916,0.7058824,0.6302521,0.6120448,0.60504204,0.6162465,0.6484594,0.6386555,0.81512606,0.6834734,0.70028013,0.78431374,0.6582633,0.6862745,0.6414566,0.76890755,0.7058824,0.7478992,0.8207283,0.7815126,0.8515406,0.88515407,0.7955182,0.9439776,0.77310926,0.8795518,0.77030814,0.88515407,0.9159664,0.85434175,0.8235294,0.88235295,0.90756303,0.9411765,0.9327731,0.9033613,0.9887955
+TabM (tuned),0.075630255,0.16246499,0.18207283,0.18907563,0.2044818,0.20728292,0.22128852,0.23669468,0.22128852,0.33333334,0.31232494,0.359944,0.38935575,0.4369748,0.4089636,0.44257703,0.4089636,0.33753502,0.42016807,0.4439776,0.44537815,0.3697479,0.5560224,,0.48879552,0.5882353,0.5294118,0.48739496,0.5182073,0.52380955,0.6106443,0.59383756,0.59103644,0.64985996,0.58963585,0.6526611,0.6106443,0.6526611,0.6526611,0.6414566,0.6834734,0.68767506,0.6554622,0.67507005,0.6162465,0.59103644,0.6204482,0.7394958,0.6666667,0.7058824,0.7366947,0.67507005,0.7745098,0.69187677,0.71288514,0.6946779,0.72829133,0.6722689,0.7226891,0.8095238,0.78431374,0.8627451,0.82633054,0.80112046,0.82913166,0.8515406,0.8767507,0.8067227,0.82633054,0.8207283,0.8319328,0.8319328,0.8487395,0.8627451,0.87114847,0.9467787,0.85714287,0.96638656
+TabDPT (default),0.11484594,0.17927171,0.18487395,0.26890758,0.28291318,0.33053222,0.37535015,0.35854343,0.3557423,0.2114846,0.41736695,0.4229692,0.41316527,0.4369748,0.47338936,0.4117647,0.47619048,0.47619048,0.4397759,0.41736695,0.43557423,0.48739496,0.48739496,0.5112045,,0.52100843,0.5462185,0.5546219,0.5462185,0.5686275,0.5826331,0.6022409,0.5714286,0.5714286,0.5966387,0.56582636,0.5546219,0.5742297,0.6414566,0.5812325,0.6246499,0.6022409,0.627451,0.58963585,0.57983196,0.719888,0.5854342,0.6582633,0.67787117,0.627451,0.64985996,0.67507005,0.66386557,0.6526611,0.69187677,0.7759104,0.77310926,0.67507005,0.77310926,0.7478992,0.74229693,0.77030814,0.7366947,0.75910366,0.7394958,0.8319328,0.7955182,0.8487395,0.7927171,0.8207283,0.8095238,0.8487395,0.88515407,0.8795518,0.88235295,0.8739496,0.91036415,0.9887955
+LightGBM (tuned),0.10084034,0.16806723,0.18487395,0.1904762,0.18767507,0.18487395,0.197479,0.2044818,0.26050422,0.32212886,0.3137255,0.30252102,0.3529412,0.41736695,0.31652662,0.36414567,0.32913166,0.33753502,0.36134455,0.40056023,0.43417367,0.37815127,0.40336135,0.4117647,0.4789916,,0.49019608,0.48179272,0.5182073,0.5252101,0.48459384,0.50980395,0.53501403,0.6344538,0.53361344,0.61764705,0.5966387,0.60784316,0.56302524,0.62184876,0.63305324,0.6904762,0.6526611,0.61764705,0.5882353,0.5826331,0.59383756,0.59103644,0.5882353,0.74229693,0.61904764,0.66386557,0.7338936,0.6246499,0.6652661,0.6246499,0.74369746,0.6834734,0.732493,0.79691875,0.767507,0.8305322,0.8417367,0.78571427,0.9033613,0.7619048,0.8515406,0.75910366,0.837535,0.8739496,0.8347339,0.80812323,0.85434175,0.8879552,0.91036415,0.929972,0.8879552,0.99159664
+RealMLP (tuned),0.11484594,0.17086835,0.18767507,0.17927171,0.18207283,0.19327731,0.22408964,0.23809524,0.24929972,0.28431374,0.28711486,0.29551822,0.36134455,0.35854343,0.3529412,0.41736695,0.27450982,0.36134455,0.37815127,0.3627451,0.4117647,0.3669468,0.49019608,0.47058824,0.45378152,0.50980395,,0.47338936,0.50560224,0.49579832,0.49579832,0.5434174,0.51960784,0.5854342,0.57983196,0.557423,0.59103644,0.56582636,0.60504204,0.6302521,0.6162465,0.6246499,0.5854342,0.64985996,0.6022409,0.5560224,0.6162465,0.6134454,0.6526611,0.662465,0.6582633,0.6442577,0.68907565,0.6106443,0.6834734,0.68767506,0.7086835,0.67507005,0.7086835,0.78711486,0.7058824,0.8207283,0.7815126,0.71568626,0.8039216,0.82633054,0.82913166,0.8585434,0.7955182,0.8207283,0.77310926,0.84313726,0.8179272,0.8179272,0.85714287,0.93557423,0.84313726,0.9747899
+SAP-RPT-OSS (default),0.12885155,0.1904762,0.22408964,0.24089636,0.2184874,0.22689076,0.24229692,0.25210086,0.30532214,0.3137255,0.38515407,0.29691878,0.34453782,0.3697479,0.46218488,0.4229692,0.4509804,0.43137255,0.41736695,0.37535015,0.4117647,0.4019608,0.4929972,0.5126051,0.44537815,0.5182073,0.5266107,,0.5462185,0.56582636,0.53781515,0.557423,0.5462185,0.52100843,0.6162465,0.5042017,0.55182076,0.5182073,0.6386555,0.5826331,0.5714286,0.59103644,0.58963585,0.6554622,0.54901963,0.57983196,0.5434174,0.5882353,0.6554622,0.60784316,0.6162465,0.6120448,0.6414566,0.6162465,0.68907565,0.697479,0.67507005,0.6414566,0.67507005,0.719888,0.73109245,0.7647059,0.7254902,0.7366947,0.72829133,0.767507,0.7745098,0.7787115,0.76890755,0.7773109,0.77030814,0.8347339,0.7787115,0.8207283,0.82913166,0.8179272,0.8515406,0.9747899
+xRFM (tuned),0.12044818,0.1764706,0.18627451,0.21428572,0.18207283,0.2184874,0.2184874,0.23529412,0.27030814,0.26890758,0.27310926,0.31792718,0.35854343,0.36414567,0.35014006,0.40056023,0.38655463,0.4257703,0.3557423,0.38235295,0.4229692,0.41736695,0.4369748,0.48179272,0.45378152,0.48179272,0.49439776,0.45378152,,0.43557423,0.45378152,0.53781515,0.48039216,0.53501403,0.5420168,0.5406163,0.5728291,0.5546219,0.557423,0.60504204,0.5686275,0.55182076,0.60784316,0.5854342,0.57983196,0.56582636,0.5826331,0.61484593,0.63585436,0.5840336,0.63585436,0.63305324,0.627451,0.5686275,0.6386555,0.7226891,0.71708685,0.6554622,0.73109245,0.77030814,0.75910366,0.7380952,0.70028013,0.78711486,0.74509805,0.82913166,0.7941176,0.8347339,0.7647059,0.8039216,0.81512606,0.8767507,0.8459384,0.8459384,0.859944,0.89635855,0.86834735,0.9859944
+xRFM (tuned + ensembled),0.08963586,0.14005603,0.14845939,0.15686275,0.17086835,0.18207283,0.18487395,0.21708684,0.2577031,0.28011206,0.2577031,0.27450982,0.30532214,0.33333334,0.3627451,0.37535015,0.34453782,0.3977591,0.33893558,0.35714287,0.38935575,0.37254903,0.44257703,0.47619048,0.43137255,0.47478992,0.5042017,0.43417367,0.56442577,,0.46638656,0.5042017,0.49439776,0.5532213,0.5532213,0.53781515,0.56302524,0.5448179,0.5266107,0.6022409,0.5686275,0.5672269,0.59103644,0.59383756,0.5448179,0.5742297,0.5434174,0.5994398,0.6302521,0.5882353,0.6162465,0.6092437,0.6106443,0.6302521,0.61904764,0.697479,0.69187677,0.6442577,0.719888,0.7338936,0.7226891,0.7759104,0.767507,0.74229693,0.78711486,0.81232494,0.8305322,0.8487395,0.8235294,0.8459384,0.78291315,0.9187675,0.85014004,0.90756303,0.92436975,0.9187675,0.8627451,0.9859944
+ModernNCA (tuned),0.09243698,0.14005603,0.15406163,0.15686275,0.18767507,0.17086835,0.16526611,0.21288516,0.24929972,0.29411766,0.25210086,0.27450982,0.29411766,0.3277311,0.394958,0.3627451,0.38655463,0.3739496,0.3837535,0.3669468,0.394958,0.3557423,0.46218488,0.38935575,0.41736695,0.5154062,0.5042017,0.46218488,0.5462185,0.53361344,,0.56302524,0.5182073,0.53501403,0.5672269,0.53501403,0.5462185,0.55182076,0.5686275,0.57703084,0.53501403,0.5854342,0.5826331,0.6162465,0.56582636,0.53781515,0.5742297,0.61904764,0.61904764,0.60364145,0.6232493,0.63305324,0.6582633,0.6134454,0.697479,0.67507005,0.6526611,0.6666667,0.66106445,0.7478992,0.6862745,0.74509805,0.7366947,0.7086835,0.7478992,0.7478992,0.7759104,0.77170867,0.7619048,0.7759104,0.7507003,0.85434175,0.7787115,0.8235294,0.8179272,0.85714287,0.8235294,0.9719888
+ModernNCA (default),0.12605043,0.17787115,0.17226891,0.20168068,0.18767507,0.1904762,0.20728292,0.22969188,0.2464986,0.26050422,0.27591038,0.2997199,0.31512606,0.33613446,0.36134455,0.3669468,0.38655463,0.3669468,0.359944,0.359944,0.3669468,0.3487395,0.46778712,0.40616247,0.3977591,0.49019608,0.45658264,0.44257703,0.46218488,0.49579832,0.4369748,,0.48179272,0.5546219,0.50980395,0.55182076,0.5322129,0.56302524,0.5266107,0.5714286,0.5406163,0.57703084,0.59103644,0.5546219,0.5966387,0.5854342,0.60504204,0.6134454,0.5686275,0.6246499,0.63305324,0.59243697,0.6386555,0.53501403,0.6162465,0.6694678,0.6806723,0.6806723,0.6834734,0.7619048,0.7338936,0.73109245,0.6862745,0.7521008,0.719888,0.8067227,0.7647059,0.789916,0.75910366,0.78431374,0.8095238,0.75910366,0.859944,0.81092435,0.80112046,0.84033614,0.87114847,0.97759104
+iLTM (tuned + ensembled),0.078431375,0.11204482,0.13445379,0.13725491,0.17086835,0.14005603,0.18067227,0.2044818,0.24789916,0.2717087,0.26610646,0.1834734,0.26750702,0.32492998,0.3137255,0.36134455,0.2857143,0.33753502,0.38655463,0.33053222,0.36414567,0.4117647,0.41036415,0.4089636,0.42857143,0.464986,0.48039216,0.45378152,0.51960784,0.50560224,0.48179272,0.5182073,,0.5042017,0.5546219,0.48179272,0.5294118,0.5154062,0.5602241,0.5686275,0.5294118,0.5588235,0.5714286,0.5994398,0.55182076,0.5448179,0.5560224,0.59383756,0.6260504,0.60784316,0.59383756,0.7184874,0.64285713,0.6134454,0.64985996,0.67507005,0.6792717,0.6554622,0.69187677,0.74229693,0.6582633,0.8039216,0.7955182,0.6862745,0.8459384,0.7759104,0.8319328,0.77310926,0.8095238,0.8319328,0.73109245,0.8907563,0.8095238,0.85434175,0.88515407,0.8879552,0.85294116,0.9691877
+CatBoost (default),0.075630255,0.12885155,0.15126051,0.16666667,0.12885155,0.15686275,0.16806723,0.19607843,0.21568628,0.2857143,0.27450982,0.26330534,0.27731094,0.3557423,0.2857143,0.29411766,0.29411766,0.28291318,0.3487395,0.36414567,0.37535015,0.33053222,0.32352942,0.35014006,0.42857143,0.36554623,0.41456583,0.4789916,0.464986,0.4467787,0.464986,0.44537815,0.49579832,,0.50700283,0.5014006,0.5182073,0.4859944,0.5546219,0.5602241,0.5854342,0.56442577,0.59383756,0.57983196,0.5980392,0.5126051,0.60504204,0.5588235,0.5742297,0.5868347,0.60504204,0.64985996,0.6442577,0.5854342,0.6442577,0.5826331,0.72829133,0.6666667,0.71428573,0.8067227,0.70308125,0.7394958,0.8039216,0.7254902,0.8347339,0.75910366,0.78711486,0.7507003,0.85434175,0.8627451,0.78431374,0.7927171,0.8627451,0.8515406,0.8795518,0.86834735,0.8935574,0.9831933
+FastaiMLP (tuned + ensembled),0.10364146,0.16946779,0.15686275,0.1764706,0.15406163,0.22408964,0.20588236,0.22128852,0.2184874,0.232493,0.27310926,0.30392158,0.29411766,0.26610646,0.31652662,0.3739496,0.30392158,0.3557423,0.2857143,0.33333334,0.3417367,0.34313726,0.43137255,0.41036415,0.40336135,0.46638656,0.42016807,0.3837535,0.4579832,0.4467787,0.4327731,0.49019608,0.44537815,0.4929972,,0.5014006,0.53641456,0.5266107,0.49019608,0.5714286,0.52100843,0.5406163,0.5742297,0.5728291,0.5854342,0.5322129,0.59383756,0.57703084,0.6554622,0.57703084,0.5966387,0.5686275,0.5854342,0.57983196,0.60504204,0.6862745,0.66106445,0.6582633,0.6526611,0.7226891,0.7254902,0.77030814,0.7086835,0.7507003,0.710084,0.8165266,0.79831934,0.929972,0.78291315,0.789916,0.77030814,0.8235294,0.85714287,0.8179272,0.837535,0.92436975,0.8739496,0.9691877
+CatBoost (tuned + ensembled),0.075630255,0.14005603,0.16526611,0.15966387,0.15966387,0.15966387,0.18487395,0.19887955,0.2464986,0.26330534,0.2857143,0.2577031,0.30812326,0.35014006,0.28011206,0.33613446,0.26050422,0.2997199,0.33893558,0.3557423,0.3669468,0.33613446,0.33613446,0.34733894,0.43417367,0.38235295,0.44257703,0.49579832,0.45938376,0.46218488,0.464986,0.44817927,0.5182073,0.49859944,0.49859944,,0.4789916,0.5266107,0.5462185,0.52240896,0.5742297,0.5392157,0.54901963,0.5462185,0.5826331,0.50700283,0.57703084,0.56302524,0.5546219,0.5868347,0.5882353,0.64705884,0.6246499,0.59103644,0.63305324,0.56302524,0.719888,0.6526611,0.71568626,0.75770307,0.697479,0.7535014,0.81512606,0.719888,0.84313726,0.7366947,0.7815126,0.7366947,0.8459384,0.872549,0.7619048,0.789916,0.8627451,0.86834735,0.9131653,0.859944,0.8739496,0.9887955
+EBM (tuned + ensembled),0.1092437,0.15686275,0.16526611,0.19327731,0.15966387,0.21288516,0.2044818,0.19607843,0.20728292,0.26890758,0.25350142,0.25490198,0.33053222,0.33053222,0.32212886,0.27450982,0.28851542,0.28011206,0.33053222,0.3627451,0.34453782,0.34453782,0.38235295,0.38935575,0.44537815,0.40336135,0.4089636,0.44817927,0.42717087,0.4369748,0.45378152,0.46778712,0.47058824,0.48179272,0.46358544,0.52100843,,0.50700283,0.49579832,0.64705884,0.5854342,0.5854342,0.6414566,0.5294118,0.5686275,0.56582636,0.5686275,0.5266107,0.51960784,0.5854342,0.5602241,0.5602241,0.60504204,0.5462185,0.5812325,0.63305324,0.710084,0.6666667,0.69187677,0.7507003,0.74229693,0.73109245,0.7521008,0.7535014,0.7647059,0.6694678,0.75770307,0.7535014,0.789916,0.79831934,0.7927171,0.7619048,0.8627451,0.8459384,0.85434175,0.88235295,0.8753501,0.9719888
+CatBoost (tuned),0.075630255,0.12605043,0.15406163,0.16946779,0.17086835,0.1764706,0.18767507,0.2044818,0.25630254,0.2717087,0.28711486,0.26050422,0.29691878,0.3347339,0.2717087,0.30392158,0.26890758,0.29131654,0.3417367,0.35854343,0.3739496,0.31932774,0.34313726,0.34733894,0.4257703,0.39215687,0.43417367,0.48179272,0.44537815,0.45518208,0.44817927,0.4369748,0.48459384,0.5140056,0.47338936,0.47338936,0.4929972,,0.5406163,0.52380955,0.5532213,0.56302524,0.5602241,0.5406163,0.5882353,0.5042017,0.59383756,0.5252101,0.5588235,0.59383756,0.56442577,0.6106443,0.6134454,0.59103644,0.6232493,0.55182076,0.70028013,0.66106445,0.6862745,0.7338936,0.69187677,0.7366947,0.8319328,0.7072829,0.8235294,0.71708685,0.77030814,0.7254902,0.8459384,0.8487395,0.7507003,0.7815126,0.85714287,0.859944,0.9019608,0.87114847,0.88235295,0.9887955
+TorchMLP (tuned + ensembled),0.04761905,0.0952381,0.10084034,0.12044818,0.11764706,0.14005603,0.14565827,0.14845939,0.17366947,0.24369748,0.232493,0.22128852,0.26050422,0.2717087,0.24369748,0.33053222,0.29691878,0.30532214,0.32212886,0.29551822,0.29411766,0.3137255,0.41456583,0.34733894,0.35854343,0.4369748,0.394958,0.36134455,0.44257703,0.47338936,0.43137255,0.47338936,0.4397759,0.44537815,0.50980395,0.45378152,0.5042017,0.45938376,,0.5322129,0.46778712,0.50560224,0.5434174,0.5462185,0.55182076,0.48459384,0.56162465,0.5154062,0.6064426,0.5294118,0.5602241,0.5560224,0.5742297,0.5714286,0.7072829,0.63305324,0.64705884,0.6582633,0.6386555,0.7240896,0.6904762,0.7507003,0.7254902,0.72829133,0.697479,0.837535,0.7955182,0.79971987,0.74509805,0.767507,0.7815126,0.8095238,0.8095238,0.7619048,0.7787115,0.88515407,0.8459384,0.96638656
+EBM (tuned),0.10084034,0.14565827,0.16806723,0.18767507,0.13445379,0.18907563,0.18907563,0.19327731,0.19887955,0.24789916,0.24089636,0.23809524,0.3137255,0.31652662,0.30532214,0.26330534,0.28011206,0.2717087,0.31652662,0.3347339,0.33333334,0.32212886,0.35014006,0.35854343,0.4187675,0.37815127,0.3697479,0.41736695,0.394958,0.3977591,0.4229692,0.42857143,0.43137255,0.4397759,0.42857143,0.47759104,0.3529412,0.47619048,0.46778712,,0.5266107,0.50700283,0.59243697,0.46218488,0.55182076,0.5280112,0.5588235,0.49019608,0.5,0.53501403,0.5252101,0.5266107,0.56302524,0.5140056,0.557423,0.5952381,0.66386557,0.6414566,0.6946779,0.7058824,0.71428573,0.7058824,0.72829133,0.72829133,0.74509805,0.64985996,0.73109245,0.71428573,0.75910366,0.78711486,0.75630254,0.74509805,0.8459384,0.8487395,0.84313726,0.8627451,0.8767507,0.9719888
+TabM (default),0.078431375,0.13165267,0.15686275,0.14845939,0.13165267,0.16946779,0.18207283,0.18207283,0.21288516,0.26050422,0.26890758,0.28291318,0.28011206,0.3277311,0.29411766,0.26610646,0.29131654,0.20028012,0.30252102,0.35434175,0.36414567,0.31092438,0.32352942,0.31652662,0.37535015,0.3669468,0.3837535,0.42857143,0.43137255,0.43137255,0.464986,0.45938376,0.47058824,0.41456583,0.4789916,0.4257703,0.41456583,0.4467787,0.5322129,0.47338936,,0.464986,0.4579832,0.5154062,0.5966387,0.5126051,0.60504204,0.48459384,0.557423,0.50980395,0.5154062,0.56442577,0.5602241,0.5714286,0.5966387,0.5686275,0.6722689,0.6834734,0.6806723,0.6834734,0.70308125,0.6820728,0.64985996,0.7240896,0.6582633,0.72829133,0.6946779,0.71428573,0.78711486,0.78431374,0.75910366,0.7254902,0.8235294,0.81232494,0.837535,0.8319328,0.8655462,0.9607843
+XGBoost (tuned + ensembled),0.04761905,0.12184874,0.13725491,0.13725491,0.14845939,0.15686275,0.16526611,0.17927171,0.20308124,0.24369748,0.23389356,0.22128852,0.25210086,0.3277311,0.232493,0.25210086,0.22689076,0.24089636,0.3277311,0.34033614,0.34313726,0.30392158,0.21008404,0.31232494,0.3977591,0.30952382,0.37535015,0.4089636,0.44817927,0.4327731,0.41456583,0.4229692,0.44117647,0.43557423,0.45938376,0.46078432,0.41456583,0.4369748,0.49439776,0.4929972,0.53501403,,0.50700283,0.5042017,0.5546219,0.5028011,0.557423,0.45658264,0.50980395,0.5826331,0.5154062,0.5826331,0.5448179,0.53501403,0.5714286,0.5672269,0.6526611,0.6302521,0.67787117,0.67787117,0.6806723,0.70028013,0.8487395,0.6946779,0.8767507,0.68907565,0.7352941,0.6834734,0.85294116,0.85014004,0.7478992,0.79831934,0.8067227,0.8487395,0.9019608,0.8487395,0.8557423,0.9747899
+EBM (default),0.1092437,0.15546219,0.15966387,0.1764706,0.16526611,0.20728292,0.19607843,0.21008404,0.2044818,0.24929972,0.24929972,0.24369748,0.30812326,0.32212886,0.27731094,0.23529412,0.28431374,0.28151262,0.30812326,0.35014006,0.33613446,0.32492998,0.29411766,0.34453782,0.37254903,0.34733894,0.41456583,0.41036415,0.39215687,0.4089636,0.41736695,0.4089636,0.42857143,0.40616247,0.4257703,0.4509804,0.35854343,0.4397759,0.45658264,0.40756303,0.5420168,0.4929972,,0.48739496,0.5602241,0.53501403,0.5546219,0.49579832,0.49859944,0.5322129,0.5266107,0.49579832,0.5294118,0.5322129,0.56582636,0.6106443,0.6722689,0.6386555,0.6442577,0.7226891,0.7338936,0.6834734,0.69187677,0.7521008,0.70028013,0.6834734,0.70308125,0.7254902,0.7815126,0.7927171,0.79831934,0.7226891,0.8627451,0.837535,0.82633054,0.84313726,0.8879552,0.9691877
+RealMLP (default),0.10644258,0.13165267,0.13725491,0.17927171,0.14005603,0.1764706,0.18767507,0.1904762,0.21008404,0.22969188,0.232493,0.26050422,0.25210086,0.28851542,0.26610646,0.30812326,0.21428572,0.2787115,0.27731094,0.32492998,0.31652662,0.30392158,0.3697479,0.32492998,0.41036415,0.38235295,0.35014006,0.34453782,0.41456583,0.40616247,0.3837535,0.44537815,0.40056023,0.42016807,0.42717087,0.45378152,0.47058824,0.45938376,0.45378152,0.53781515,0.48459384,0.49579832,0.5126051,,0.55182076,0.5042017,0.5742297,0.49859944,0.5266107,0.53501403,0.5322129,0.5294118,0.5434174,0.5042017,0.5826331,0.65686274,0.63305324,0.63585436,0.64985996,0.6946779,0.6806723,0.7114846,0.6946779,0.71428573,0.70028013,0.7521008,0.75910366,0.81232494,0.74509805,0.7619048,0.767507,0.7647059,0.84313726,0.8039216,0.8067227,0.8879552,0.84313726,0.9691877
+TabSTAR (tuned),0.22128852,0.2647059,0.24929972,0.31092438,0.2717087,0.3137255,0.31092438,0.31932774,0.31652662,0.33053222,0.34453782,0.32352942,0.34733894,0.30812326,0.36414567,0.33053222,0.37535015,0.36134455,0.32212886,0.38655463,0.4019608,0.36134455,0.3879552,0.3837535,0.42016807,0.4117647,0.3977591,0.4509804,0.42016807,0.45518208,0.43417367,0.40336135,0.44817927,0.4019608,0.41456583,0.41736695,0.43137255,0.4117647,0.44817927,0.44817927,0.40336135,0.44537815,0.4397759,0.44817927,,0.5014006,0.5980392,0.42016807,0.46218488,0.464986,0.4369748,0.4579832,0.44537815,0.5406163,0.45658264,0.5602241,0.627451,0.9131653,0.6232493,0.48179272,0.63305324,0.62184876,0.57703084,0.6582633,0.5602241,0.56582636,0.6302521,0.6806723,0.69187677,0.7254902,0.6862745,0.719888,0.78711486,0.8095238,0.8039216,0.6862745,0.8487395,0.9411765
+OrionMSP (default),0.070028014,0.12044818,0.09243698,0.1162465,0.10644258,0.18207283,0.14845939,0.15406163,0.24789916,0.17086835,0.28291318,0.24229692,0.24929972,0.1904762,0.37114847,0.34733894,0.3557423,0.37535015,0.3137255,0.26330534,0.24929972,0.37815127,0.394958,0.4089636,0.28011206,0.41736695,0.4439776,0.42016807,0.43417367,0.4257703,0.46218488,0.41456583,0.45518208,0.48739496,0.46778712,0.4929972,0.43417367,0.49579832,0.5154062,0.4719888,0.48739496,0.49719888,0.464986,0.49579832,0.49859944,,0.50980395,0.5014006,0.54901963,0.5294118,0.5126051,0.6022409,0.52100843,0.6022409,0.57983196,0.6246499,0.5966387,0.63305324,0.5994398,0.6302521,0.69187677,0.6806723,0.63585436,0.7212885,0.67787117,0.697479,0.70028013,0.7366947,0.66386557,0.70308125,0.7478992,0.7759104,0.73109245,0.7619048,0.72829133,0.8347339,0.7787115,0.96638656
+TabSTAR (tuned + ensembled),0.2184874,0.24929972,0.25210086,0.30812326,0.26750702,0.2997199,0.30252102,0.30812326,0.3137255,0.31932774,0.33893558,0.3277311,0.34453782,0.29411766,0.3529412,0.32492998,0.36134455,0.3557423,0.32633054,0.3837535,0.40336135,0.36414567,0.3837535,0.37955183,0.41456583,0.40616247,0.3837535,0.45658264,0.41736695,0.45658264,0.4257703,0.394958,0.4439776,0.394958,0.40616247,0.4229692,0.43137255,0.40616247,0.43837535,0.44117647,0.394958,0.44257703,0.44537815,0.4257703,0.4019608,0.49019608,,0.4229692,0.47058824,0.4579832,0.4397759,0.46778712,0.44257703,0.53501403,0.4509804,0.54901963,0.6022409,0.93557423,0.5994398,0.48739496,0.627451,0.61904764,0.56582636,0.6582633,0.5602241,0.56302524,0.6246499,0.6806723,0.6834734,0.71708685,0.6834734,0.7114846,0.77310926,0.80112046,0.79831934,0.6834734,0.8347339,0.9411765
+ChimeraBoost (tuned + ensembled),0.070028014,0.09803922,0.12044818,0.11764706,0.10364146,0.10364146,0.13305323,0.15266107,0.15406163,0.232493,0.20168068,0.1904762,0.232493,0.3137255,0.25630254,0.25910366,0.22969188,0.22128852,0.33333334,0.31092438,0.32492998,0.27450982,0.35154063,0.26050422,0.3417367,0.4089636,0.38655463,0.4117647,0.38515407,0.40056023,0.3809524,0.38655463,0.40616247,0.44117647,0.4229692,0.4369748,0.47338936,0.47478992,0.48459384,0.50980395,0.5154062,0.5434174,0.5042017,0.5014006,0.57983196,0.49859944,0.57703084,,0.48879552,0.58963585,0.5826331,0.5686275,0.6862745,0.5546219,0.56302524,0.5882353,0.627451,0.6442577,0.6414566,0.68907565,0.64985996,0.73109245,0.77030814,0.6666667,0.78571427,0.69187677,0.75910366,0.66386557,0.75630254,0.767507,0.71428573,0.7507003,0.77030814,0.80112046,0.8095238,0.8459384,0.8319328,0.96498597
+FastaiMLP (tuned),0.084033616,0.13865547,0.12885155,0.16526611,0.12885155,0.17927171,0.15686275,0.1764706,0.17927171,0.18767507,0.24789916,0.24789916,0.24369748,0.232493,0.26610646,0.29691878,0.27310926,0.30252102,0.27731094,0.27450982,0.29411766,0.31932774,0.36134455,0.33333334,0.32212886,0.4117647,0.34733894,0.34453782,0.36414567,0.3697479,0.3809524,0.43137255,0.3739496,0.4257703,0.34453782,0.44537815,0.48039216,0.44117647,0.39355743,0.5,0.44257703,0.49019608,0.5014006,0.47338936,0.53781515,0.4509804,0.5294118,0.5112045,,0.48739496,0.5308123,0.5126051,0.5434174,0.52380955,0.5434174,0.6246499,0.6106443,0.6106443,0.6022409,0.6526611,0.67787117,0.6834734,0.6386555,0.6960784,0.6386555,0.7240896,0.72829133,0.8095238,0.7184874,0.7226891,0.71708685,0.719888,0.8067227,0.75910366,0.77310926,0.84033614,0.8319328,0.9579832
+XGBoost (tuned),0.04481793,0.10084034,0.12044818,0.12885155,0.12044818,0.14005603,0.13865547,0.16806723,0.19607843,0.232493,0.21568628,0.19887955,0.23529412,0.28851542,0.23809524,0.232493,0.20168068,0.19887955,0.30812326,0.29691878,0.3277311,0.29691878,0.18487395,0.29411766,0.37254903,0.2577031,0.33753502,0.39215687,0.4159664,0.4117647,0.39635855,0.37535015,0.39215687,0.41316527,0.4229692,0.41316527,0.41456583,0.40616247,0.47058824,0.464986,0.49019608,0.41736695,0.46778712,0.464986,0.53501403,0.47058824,0.5420168,0.41036415,0.5126051,,0.47619048,0.56302524,0.4929972,0.5266107,0.5406163,0.55182076,0.6540616,0.6246499,0.6694678,0.6722689,0.66106445,0.66106445,0.7927171,0.69187677,0.8207283,0.6554622,0.719888,0.67787117,0.82633054,0.8347339,0.7366947,0.78711486,0.80812323,0.8459384,0.8767507,0.84313726,0.8515406,0.9691877
+ChimeraBoost (tuned),0.05602241,0.08963586,0.09803922,0.1092437,0.09803922,0.0952381,0.11764706,0.13165267,0.15126051,0.21008404,0.1904762,0.16806723,0.21008404,0.27450982,0.23809524,0.232493,0.21428572,0.17366947,0.3137255,0.28011206,0.30532214,0.25210086,0.31652662,0.26330534,0.35014006,0.3809524,0.3417367,0.3837535,0.36414567,0.3837535,0.3767507,0.3669468,0.40616247,0.394958,0.40336135,0.4117647,0.4397759,0.43557423,0.4397759,0.47478992,0.48459384,0.48459384,0.47338936,0.46778712,0.56302524,0.48739496,0.5602241,0.41736695,0.46918768,0.52380955,,0.52240896,0.5966387,0.5266107,0.52380955,0.5546219,0.59383756,0.63585436,0.61904764,0.64705884,0.61904764,0.6680672,0.732493,0.627451,0.73109245,0.67787117,0.70168066,0.64705884,0.7254902,0.72829133,0.68907565,0.719888,0.7394958,0.789916,0.7759104,0.83613443,0.7955182,0.94957983
+iLTM (tuned),0.064425774,0.082633056,0.11204482,0.10364146,0.09383754,0.11204482,0.12605043,0.13445379,0.16246499,0.16946779,0.2044818,0.17366947,0.18487395,0.2044818,0.21568628,0.29691878,0.22689076,0.2464986,0.30812326,0.22408964,0.24929972,0.28151262,0.2997199,0.32492998,0.32492998,0.33613446,0.3557423,0.3879552,0.3669468,0.3907563,0.3669468,0.40756303,0.28151262,0.35014006,0.43137255,0.3529412,0.4397759,0.38935575,0.4439776,0.47338936,0.43557423,0.41736695,0.5042017,0.47058824,0.5420168,0.3977591,0.5322129,0.43137255,0.48739496,0.4369748,0.47759104,,0.48459384,0.49859944,0.5434174,0.5420168,0.5686275,0.60784316,0.5854342,0.64565825,0.57563025,0.627451,0.62184876,0.6106443,0.62184876,0.72829133,0.68907565,0.7226891,0.6862745,0.697479,0.6666667,0.7647059,0.7366947,0.7254902,0.7535014,0.7955182,0.7913165,0.94957983
+ChimeraBoost (default),0.05042017,0.10084034,0.1092437,0.11204482,0.0952381,0.09243698,0.11764706,0.12885155,0.13165267,0.1904762,0.16806723,0.15966387,0.2044818,0.22689076,0.22408964,0.21568628,0.21288516,0.16806723,0.32212886,0.25490198,0.27731094,0.22689076,0.21568628,0.2254902,0.33613446,0.26610646,0.31092438,0.35854343,0.37254903,0.38935575,0.3417367,0.36134455,0.35714287,0.3557423,0.41456583,0.37535015,0.394958,0.38655463,0.4257703,0.4369748,0.4397759,0.45518208,0.47058824,0.45658264,0.5546219,0.4789916,0.557423,0.3137255,0.45658264,0.50700283,0.40336135,0.5154062,,0.52380955,0.50980395,0.54901963,0.60504204,0.6246499,0.6106443,0.64985996,0.63585436,0.6386555,0.7212885,0.6554622,0.72689074,0.6554622,0.66106445,0.63305324,0.7114846,0.72829133,0.7086835,0.7226891,0.732493,0.75630254,0.7647059,0.7927171,0.8179272,0.9579832
+ModernNCA (tuned + ensembled),0.04481793,0.081232496,0.086834736,0.072829135,0.12605043,0.070028014,0.11204482,0.12184874,0.18207283,0.23809524,0.16526611,0.16246499,0.18767507,0.26610646,0.2857143,0.3137255,0.29411766,0.26330534,0.30812326,0.23809524,0.26190478,0.28011206,0.3417367,0.30812326,0.34733894,0.37535015,0.38935575,0.3837535,0.43137255,0.3697479,0.38655463,0.464986,0.38655463,0.41456583,0.42016807,0.4089636,0.45378152,0.4089636,0.42857143,0.4859944,0.42857143,0.464986,0.46778712,0.49579832,0.45938376,0.3977591,0.464986,0.44537815,0.47619048,0.47338936,0.47338936,0.5014006,0.47619048,,0.5126051,0.5266107,0.52100843,0.5266107,0.54901963,0.5826331,0.5322129,0.5966387,0.6554622,0.53501403,0.6386555,0.60504204,0.6022409,0.63305324,0.60784316,0.6134454,0.5826331,0.77170867,0.66386557,0.71708685,0.71428573,0.7226891,0.70308125,0.9159664
+TorchMLP (tuned),0.039215688,0.072829135,0.078431375,0.09243698,0.075630255,0.0952381,0.10364146,0.10364146,0.10644258,0.18487395,0.15406163,0.18767507,0.19607843,0.23529412,0.2044818,0.25210086,0.24369748,0.23109244,0.26610646,0.26050422,0.27450982,0.2464986,0.3137255,0.28711486,0.30812326,0.3347339,0.31652662,0.31092438,0.36134455,0.3809524,0.30252102,0.3837535,0.35014006,0.3557423,0.394958,0.3669468,0.4187675,0.3767507,0.2927171,0.44257703,0.40336135,0.42857143,0.43417367,0.41736695,0.5434174,0.42016807,0.54901963,0.4369748,0.45658264,0.45938376,0.47619048,0.45658264,0.49019608,0.48739496,,0.6106443,0.5980392,0.6246499,0.5742297,0.6386555,0.6134454,0.6386555,0.6106443,0.63165265,0.6092437,0.74369746,0.6806723,0.71428573,0.6806723,0.69187677,0.72829133,0.7338936,0.7605042,0.7226891,0.73109245,0.81232494,0.767507,0.95238096
+TabFlex (default),0.08963586,0.11764706,0.12605043,0.14565827,0.11904762,0.14845939,0.15406163,0.16806723,0.17366947,0.12885155,0.22408964,0.26330534,0.22408964,0.13725491,0.26050422,0.28431374,0.28851542,0.2857143,0.17366947,0.24089636,0.2464986,0.26890758,0.35854343,0.30532214,0.22408964,0.37535015,0.31232494,0.30252102,0.27731094,0.30252102,0.32492998,0.33053222,0.32492998,0.41736695,0.3137255,0.4369748,0.3669468,0.44817927,0.3669468,0.4047619,0.43137255,0.4327731,0.38935575,0.34313726,0.4397759,0.37535015,0.4509804,0.4117647,0.37535015,0.44817927,0.44537815,0.4579832,0.4509804,0.47338936,0.38935575,,0.5182073,0.5686275,0.5182073,0.55182076,0.56582636,0.59383756,0.557423,0.6064426,0.557423,0.57563025,0.627451,0.61904764,0.60784316,0.6246499,0.63585436,0.6414566,0.6862745,0.66106445,0.697479,0.7058824,0.7366947,0.9187675
+ExtraTrees (tuned + ensembled),0.05602241,0.067226894,0.10504202,0.09243698,0.11204482,0.10084034,0.13445379,0.13865547,0.14565827,0.13025211,0.15826331,0.1764706,0.15406163,0.17366947,0.24369748,0.18207283,0.19887955,0.20728292,0.23809524,0.19607843,0.23809524,0.26890758,0.23109244,0.2717087,0.22689076,0.25630254,0.29131654,0.32492998,0.28291318,0.30812326,0.34733894,0.31932774,0.3207283,0.2717087,0.33893558,0.28011206,0.28991598,0.2997199,0.3529412,0.33613446,0.3277311,0.34733894,0.3277311,0.3669468,0.37254903,0.40336135,0.3977591,0.37254903,0.38935575,0.34593838,0.40616247,0.43137255,0.394958,0.4789916,0.4019608,0.48179272,,0.5728291,0.6106443,0.49859944,0.57843137,0.5434174,0.50700283,0.5980392,0.5126051,0.54901963,0.557423,0.5994398,0.71568626,0.74509805,0.66106445,0.66386557,0.80112046,0.8487395,0.8039216,0.66386557,0.8319328,0.9859944
+TabSTAR (default),0.17927171,0.20728292,0.20168068,0.24369748,0.22969188,0.2577031,0.2577031,0.25490198,0.26890758,0.2577031,0.27450982,0.26750702,0.27731094,0.24089636,0.29411766,0.25910366,0.30812326,0.30252102,0.25490198,0.29411766,0.2997199,0.29411766,0.29411766,0.3277311,0.32492998,0.31652662,0.32492998,0.35854343,0.34453782,0.3557423,0.33333334,0.31932774,0.34453782,0.33333334,0.3417367,0.34733894,0.33333334,0.33893558,0.3417367,0.35854343,0.31652662,0.3697479,0.36134455,0.36414567,0.086834736,0.3669468,0.064425774,0.3557423,0.38935575,0.37535015,0.36414567,0.39215687,0.37535015,0.47338936,0.37535015,0.43137255,0.42717087,,0.43137255,0.4229692,0.49859944,0.52380955,0.46778712,0.5154062,0.48179272,0.45378152,0.52100843,0.50700283,0.50560224,0.53501403,0.5826331,0.63305324,0.6666667,0.64705884,0.60504204,0.5882353,0.7394958,0.93557423
+ExtraTrees (tuned),0.067226894,0.072829135,0.10644258,0.11204482,0.11204482,0.11344538,0.14285715,0.14005603,0.14425771,0.10364146,0.15966387,0.18767507,0.16946779,0.17927171,0.23529412,0.18767507,0.20588236,0.23529412,0.22689076,0.19607843,0.22969188,0.22689076,0.25210086,0.27731094,0.22689076,0.26750702,0.29131654,0.32492998,0.26890758,0.28011206,0.33893558,0.31652662,0.30812326,0.2857143,0.34733894,0.28431374,0.30812326,0.3137255,0.36134455,0.30532214,0.31932774,0.32212886,0.3557423,0.35014006,0.3767507,0.40056023,0.40056023,0.35854343,0.3977591,0.33053222,0.3809524,0.41456583,0.38935575,0.4509804,0.4257703,0.48179272,0.38935575,0.5686275,,0.5126051,0.5714286,0.5154062,0.4789916,0.59383756,0.47338936,0.55182076,0.5462185,0.59103644,0.66106445,0.70028013,0.6554622,0.63305324,0.75630254,0.7759104,0.7661064,0.6414566,0.837535,0.9831933
+iLTM (default),0.072829135,0.084033616,0.09243698,0.0952381,0.08963586,0.1092437,0.1092437,0.11484594,0.11204482,0.16806723,0.15686275,0.14005603,0.17366947,0.19887955,0.18207283,0.14005603,0.15406163,0.16806723,0.19887955,0.1764706,0.21008404,0.19327731,0.17927171,0.1904762,0.25210086,0.20308124,0.21288516,0.28011206,0.22969188,0.26610646,0.25210086,0.23809524,0.2577031,0.19327731,0.27731094,0.24229692,0.24929972,0.26610646,0.27591038,0.29411766,0.31652662,0.32212886,0.27731094,0.30532214,0.5182073,0.3697479,0.5126051,0.31092438,0.34733894,0.3277311,0.3529412,0.35434175,0.35014006,0.41736695,0.36134455,0.44817927,0.5014006,0.57703084,0.48739496,,0.5686275,0.47058824,0.47619048,0.5966387,0.47058824,0.53501403,0.49019608,0.54901963,0.6302521,0.6526611,0.6526611,0.6134454,0.7338936,0.6694678,0.7058824,0.6106443,0.8067227,0.92156863
+Linear (tuned + ensembled),0.086834736,0.12885155,0.1092437,0.15406163,0.14005603,0.15406163,0.14005603,0.15966387,0.14845939,0.16806723,0.18487395,0.1904762,0.20728292,0.18767507,0.20028012,0.14565827,0.21288516,0.18767507,0.17226891,0.24369748,0.2394958,0.22689076,0.2184874,0.21568628,0.2577031,0.232493,0.29411766,0.26890758,0.24089636,0.27731094,0.3137255,0.26610646,0.3417367,0.29691878,0.27450982,0.30252102,0.2577031,0.30812326,0.30952382,0.2857143,0.29691878,0.31932774,0.26610646,0.31932774,0.3669468,0.30812326,0.37254903,0.35014006,0.32212886,0.33893558,0.3809524,0.42436975,0.36414567,0.46778712,0.38655463,0.43417367,0.42156863,0.5014006,0.42857143,0.43137255,,0.4789916,0.45938376,0.6386555,0.4397759,0.47619048,0.5014006,0.5014006,0.4789916,0.49019608,0.719888,0.5602241,0.6246499,0.59103644,0.5546219,0.59383756,0.70028013,0.8935574
+PerpetualBooster (tuned + ensembled),0.022408964,0.078431375,0.070028014,0.075630255,0.078431375,0.081232496,0.075630255,0.09943978,0.13165267,0.13725491,0.14005603,0.12605043,0.12885155,0.15406163,0.0952381,0.1764706,0.10084034,0.13165267,0.20168068,0.17086835,0.1904762,0.2184874,0.14845939,0.13725491,0.22969188,0.16946779,0.17927171,0.23529412,0.26190478,0.22408964,0.25490198,0.26890758,0.19607843,0.26050422,0.22969188,0.2464986,0.26890758,0.26330534,0.24929972,0.29411766,0.31792718,0.2997199,0.31652662,0.28851542,0.37815127,0.31932774,0.3809524,0.26890758,0.31652662,0.33893558,0.33193278,0.37254903,0.36134455,0.40336135,0.36134455,0.40616247,0.45658264,0.47619048,0.48459384,0.5294118,0.52100843,,0.5602241,0.53501403,0.5602241,0.50700283,0.6512605,0.49579832,0.557423,0.5826331,0.60504204,0.6442577,0.65966386,0.6834734,0.70308125,0.7647059,0.7226891,0.9411765
+XGBoost (default),0.030812325,0.05042017,0.072829135,0.072829135,0.075630255,0.05602241,0.078431375,0.084033616,0.084033616,0.13165267,0.09803922,0.12044818,0.1092437,0.18207283,0.10644258,0.16526611,0.11764706,0.12044818,0.232493,0.17927171,0.21288516,0.22969188,0.11484594,0.17366947,0.26330534,0.15826331,0.2184874,0.27450982,0.2997199,0.232493,0.26330534,0.3137255,0.2044818,0.19607843,0.29131654,0.18487395,0.24789916,0.16806723,0.27450982,0.2717087,0.35014006,0.15126051,0.30812326,0.30532214,0.4229692,0.36414567,0.43417367,0.22969188,0.36134455,0.20728292,0.26750702,0.37815127,0.2787115,0.34453782,0.38935575,0.44257703,0.4929972,0.5322129,0.52100843,0.52380955,0.5406163,0.4397759,,0.5546219,0.50560224,0.5266107,0.49579832,0.55182076,0.57983196,0.6134454,0.627451,0.6806723,0.6806723,0.719888,0.7394958,0.6414566,0.7366947,0.92436975
+Linear (tuned),0.075630255,0.11764706,0.09803922,0.12044818,0.12605043,0.1232493,0.11764706,0.14005603,0.11764706,0.14845939,0.16246499,0.16246499,0.18487395,0.16526611,0.16666667,0.11764706,0.20728292,0.17226891,0.16526611,0.2184874,0.21568628,0.21008404,0.2044818,0.19887955,0.24089636,0.21428572,0.28431374,0.26330534,0.21288516,0.2577031,0.29131654,0.24789916,0.3137255,0.27450982,0.24929972,0.28011206,0.2464986,0.2927171,0.2717087,0.2717087,0.27591038,0.30532214,0.24789916,0.2857143,0.3417367,0.2787115,0.3417367,0.33333334,0.30392158,0.30812326,0.37254903,0.38935575,0.34453782,0.464986,0.36834735,0.39355743,0.4019608,0.48459384,0.40616247,0.40336135,0.36134455,0.464986,0.44537815,,0.42857143,0.45658264,0.48739496,0.48459384,0.47338936,0.47619048,0.63305324,0.557423,0.57843137,0.5686275,0.5406163,0.59103644,0.67787117,0.88235295
+LightGBM (default),0.033613447,0.05322129,0.05882353,0.064425774,0.042016808,0.05042017,0.06162465,0.067226894,0.08963586,0.11764706,0.081232496,0.09803922,0.09103642,0.16386555,0.081232496,0.14425771,0.1092437,0.1162465,0.1904762,0.15126051,0.19607843,0.18207283,0.05602241,0.17086835,0.26050422,0.09663866,0.19607843,0.2717087,0.25490198,0.21288516,0.25210086,0.28011206,0.15406163,0.16526611,0.28991598,0.15686275,0.23529412,0.1764706,0.30252102,0.25490198,0.3417367,0.1232493,0.2997199,0.2997199,0.4397759,0.32212886,0.4397759,0.21428572,0.36134455,0.17927171,0.26890758,0.37815127,0.27310926,0.36134455,0.3907563,0.44257703,0.48739496,0.5182073,0.5266107,0.5294118,0.5602241,0.4397759,0.49439776,0.5714286,,0.557423,0.5042017,0.54901963,0.5854342,0.61484593,0.63305324,0.68907565,0.6806723,0.7086835,0.7240896,0.6722689,0.7478992,0.94817924
+TorchMLP (default),0.028011205,0.05602241,0.04761905,0.064425774,0.036414567,0.06162465,0.081232496,0.072829135,0.070028014,0.09803922,0.10084034,0.14005603,0.1092437,0.09243698,0.13725491,0.17366947,0.12885155,0.12885155,0.14845939,0.14005603,0.15126051,0.19327731,0.22689076,0.14845939,0.16806723,0.23809524,0.17366947,0.232493,0.17086835,0.18767507,0.25210086,0.19327731,0.22408964,0.24089636,0.1834734,0.26330534,0.33053222,0.28291318,0.16246499,0.35014006,0.2717087,0.31092438,0.31652662,0.24789916,0.43417367,0.30252102,0.4369748,0.30812326,0.27591038,0.34453782,0.32212886,0.2717087,0.34453782,0.394958,0.25630254,0.42436975,0.4509804,0.5462185,0.44817927,0.464986,0.52380955,0.4929972,0.47338936,0.5434174,0.44257703,,0.53501403,0.59383756,0.5266107,0.56582636,0.6134454,0.6022409,0.6806723,0.62184876,0.6162465,0.67507005,0.73109245,0.9271709
+PerpetualBooster (tuned),0.019607844,0.05882353,0.05882353,0.06022409,0.064425774,0.081232496,0.070028014,0.078431375,0.10364146,0.1092437,0.10084034,0.1092437,0.11204482,0.15966387,0.08963586,0.15686275,0.08963586,0.1092437,0.19607843,0.14985995,0.17927171,0.2114846,0.12044818,0.1232493,0.2044818,0.14845939,0.17086835,0.2254902,0.20588236,0.16946779,0.22408964,0.23529412,0.16806723,0.21288516,0.20168068,0.2184874,0.24229692,0.22969188,0.2044818,0.26890758,0.30532214,0.2647059,0.29691878,0.24089636,0.3697479,0.2997199,0.37535015,0.24089636,0.2717087,0.28011206,0.29831934,0.31092438,0.33893558,0.3977591,0.31932774,0.37254903,0.44257703,0.4789916,0.45378152,0.50980395,0.49859944,0.3487395,0.5042017,0.5126051,0.49579832,0.464986,,0.464986,0.52100843,0.53781515,0.5826331,0.6204482,0.6302521,0.66106445,0.6722689,0.69887954,0.6946779,0.90756303
+FastaiMLP (default),0.04761905,0.05602241,0.064425774,0.075630255,0.072829135,0.08963586,0.081232496,0.0952381,0.08963586,0.084033616,0.09803922,0.11204482,0.11344538,0.10644258,0.1092437,0.19607843,0.10084034,0.13165267,0.15966387,0.11484594,0.13445379,0.1904762,0.22969188,0.19327731,0.15126051,0.24089636,0.14145659,0.22128852,0.16526611,0.15126051,0.22829132,0.21008404,0.22689076,0.24929972,0.070028014,0.26330534,0.2464986,0.27450982,0.20028012,0.2857143,0.2857143,0.31652662,0.27450982,0.18767507,0.31932774,0.26330534,0.31932774,0.33613446,0.1904762,0.32212886,0.3529412,0.27731094,0.3669468,0.3669468,0.2857143,0.3809524,0.40056023,0.4929972,0.4089636,0.4509804,0.49859944,0.5042017,0.44817927,0.5154062,0.4509804,0.40616247,0.53501403,,0.49859944,0.49859944,0.54901963,0.5854342,0.66106445,0.6162465,0.5980392,0.67787117,0.72829133,0.929972
+RandomForest (tuned + ensembled),0.022408964,0.06162465,0.067226894,0.070028014,0.086834736,0.067226894,0.084033616,0.10084034,0.10644258,0.086834736,0.08963586,0.10644258,0.11484594,0.14565827,0.12885155,0.10644258,0.10364146,0.14845939,0.1764706,0.11764706,0.15966387,0.21568628,0.11484594,0.17366947,0.20728292,0.16246499,0.2044818,0.23109244,0.23529412,0.1764706,0.23809524,0.24089636,0.1904762,0.14565827,0.21708684,0.15406163,0.21008404,0.15406163,0.25490198,0.24089636,0.21288516,0.14705883,0.2184874,0.25490198,0.30812326,0.33613446,0.31652662,0.24369748,0.28151262,0.17366947,0.27450982,0.3137255,0.28851542,0.39215687,0.31932774,0.39215687,0.28431374,0.49439776,0.33893558,0.3697479,0.52100843,0.44257703,0.42016807,0.5266107,0.41456583,0.47338936,0.4789916,0.5014006,,0.6106443,0.60084033,0.6414566,0.66106445,0.70308125,0.78431374,0.5686275,0.77310926,0.9551821
+RandomForest (tuned),0.025210084,0.039215688,0.070028014,0.064425774,0.05602241,0.05322129,0.081232496,0.08963586,0.078431375,0.084033616,0.078431375,0.084033616,0.081232496,0.12885155,0.11344538,0.11484594,0.078431375,0.1232493,0.14845939,0.11484594,0.14285715,0.18067227,0.084033616,0.17927171,0.17927171,0.12605043,0.17927171,0.22268908,0.19607843,0.15406163,0.22408964,0.21568628,0.16806723,0.13725491,0.21008404,0.12745099,0.20168068,0.15126051,0.232493,0.21288516,0.21568628,0.14985995,0.20728292,0.23809524,0.27450982,0.29691878,0.28291318,0.232493,0.27731094,0.16526611,0.2717087,0.30252102,0.2717087,0.38655463,0.30812326,0.37535015,0.25490198,0.464986,0.2997199,0.34733894,0.50980395,0.41736695,0.38655463,0.52380955,0.38515407,0.43417367,0.46218488,0.5014006,0.38935575,,0.5966387,0.6246499,0.640056,0.6694678,0.69187677,0.54901963,0.7647059,0.9579832
+Linear (default),0.05602241,0.072829135,0.068627454,0.09103642,0.067226894,0.08963586,0.10504202,0.09383754,0.081232496,0.084033616,0.13725491,0.15406163,0.13585435,0.12745099,0.13165267,0.0952381,0.18207283,0.14285715,0.1232493,0.17086835,0.17086835,0.18487395,0.14565827,0.16806723,0.1904762,0.16526611,0.22689076,0.22969188,0.18487395,0.21708684,0.24929972,0.1904762,0.26890758,0.21568628,0.22969188,0.23809524,0.20728292,0.24929972,0.2184874,0.24369748,0.24089636,0.25210086,0.20168068,0.232493,0.3137255,0.25210086,0.31652662,0.2857143,0.28291318,0.26330534,0.31092438,0.33333334,0.29131654,0.41736695,0.2717087,0.36414567,0.33893558,0.41736695,0.34453782,0.34733894,0.28011206,0.394958,0.37254903,0.3669468,0.3669468,0.38655463,0.41736695,0.4509804,0.39915967,0.40336135,,0.5294118,0.49859944,0.5014006,0.5042017,0.5140056,0.5966387,0.8235294
+xRFM (default),0.019607844,0.036414567,0.039215688,0.070028014,0.04481793,0.05602241,0.070028014,0.072829135,0.070028014,0.064425774,0.072829135,0.081232496,0.078431375,0.1092437,0.086834736,0.17927171,0.09243698,0.11764706,0.14285715,0.082633056,0.12044818,0.15406163,0.1764706,0.16806723,0.15126051,0.19187675,0.15686275,0.16526611,0.1232493,0.081232496,0.14565827,0.24089636,0.1092437,0.20728292,0.1764706,0.21008404,0.23809524,0.2184874,0.1904762,0.25490198,0.27450982,0.20168068,0.27731094,0.23529412,0.28011206,0.22408964,0.28851542,0.24929972,0.28011206,0.21288516,0.28011206,0.23529412,0.27731094,0.22829132,0.26610646,0.35854343,0.33613446,0.3669468,0.3669468,0.38655463,0.4397759,0.3557423,0.31932774,0.44257703,0.31092438,0.3977591,0.37955183,0.41456583,0.35854343,0.37535015,0.47058824,,0.45658264,0.4789916,0.45938376,0.50980395,0.5434174,0.8655462
+KNN (tuned + ensembled),0.022408964,0.05882353,0.05322129,0.086834736,0.086834736,0.075630255,0.086834736,0.08963586,0.10644258,0.081232496,0.09803922,0.12605043,0.10644258,0.12885155,0.10364146,0.086834736,0.11484594,0.13165267,0.13725491,0.12464986,0.12605043,0.16246499,0.11764706,0.15126051,0.11484594,0.14565827,0.18207283,0.22128852,0.15406163,0.14985995,0.22128852,0.14005603,0.1904762,0.13725491,0.14285715,0.13725491,0.13725491,0.14285715,0.1904762,0.15406163,0.1764706,0.19327731,0.13725491,0.15686275,0.21288516,0.26890758,0.22689076,0.22969188,0.19327731,0.19187675,0.26050422,0.26330534,0.26750702,0.33613446,0.2394958,0.3137255,0.19887955,0.33333334,0.24369748,0.26610646,0.37535015,0.34033614,0.31932774,0.42156863,0.31932774,0.31932774,0.3697479,0.33893558,0.33893558,0.359944,0.5014006,0.5434174,,0.45938376,0.47058824,0.47058824,0.81092435,0.9607843
+ExtraTrees (default),0.016806724,0.036414567,0.05882353,0.06162465,0.064425774,0.067226894,0.067226894,0.075630255,0.0952381,0.065826334,0.05882353,0.075630255,0.05882353,0.12044818,0.075630255,0.13165267,0.075630255,0.0952381,0.15126051,0.086834736,0.12605043,0.15686275,0.09243698,0.13725491,0.12044818,0.11204482,0.18207283,0.17927171,0.15406163,0.09243698,0.1764706,0.18907563,0.14565827,0.14845939,0.18207283,0.13165267,0.15406163,0.14005603,0.23809524,0.15126051,0.18767507,0.15126051,0.16246499,0.19607843,0.1904762,0.23809524,0.19887955,0.19887955,0.24089636,0.15406163,0.21008404,0.27450982,0.24369748,0.28291318,0.27731094,0.33893558,0.15126051,0.3529412,0.22408964,0.33053222,0.4089636,0.31652662,0.28011206,0.43137255,0.29131654,0.37815127,0.33893558,0.3837535,0.29691878,0.33053222,0.49859944,0.52100843,0.5406163,,0.5406163,0.4369748,0.64705884,0.9467787
+RandomForest (default),0.025210084,0.030812325,0.05322129,0.05882353,0.04761905,0.04761905,0.067226894,0.070028014,0.068627454,0.067226894,0.05602241,0.064425774,0.063025214,0.10644258,0.078431375,0.09803922,0.04761905,0.0952381,0.14145659,0.08963586,0.11204482,0.16806723,0.05882353,0.12885155,0.11764706,0.08963586,0.14285715,0.17086835,0.14005603,0.075630255,0.18207283,0.19887955,0.11484594,0.12044818,0.16246499,0.086834736,0.14565827,0.09803922,0.22128852,0.15686275,0.16246499,0.09803922,0.17366947,0.19327731,0.19607843,0.2717087,0.20168068,0.1904762,0.22689076,0.1232493,0.22408964,0.2464986,0.23529412,0.2857143,0.26890758,0.30252102,0.19607843,0.394958,0.23389356,0.29411766,0.44537815,0.29691878,0.26050422,0.45938376,0.27591038,0.3837535,0.3277311,0.4019608,0.21568628,0.30812326,0.49579832,0.5406163,0.5294118,0.45938376,,0.44537815,0.6386555,0.9271709
+PerpetualBooster (default),0.014005602,0.036414567,0.025210084,0.04481793,0.036414567,0.04761905,0.04761905,0.04481793,0.04481793,0.042016808,0.05882353,0.075630255,0.04761905,0.09243698,0.067226894,0.072829135,0.042016808,0.042016808,0.10644258,0.078431375,0.081232496,0.13165267,0.067226894,0.05322129,0.12605043,0.070028014,0.064425774,0.18207283,0.10364146,0.081232496,0.14285715,0.15966387,0.11204482,0.13165267,0.075630255,0.14005603,0.11764706,0.12885155,0.11484594,0.13725491,0.16806723,0.15126051,0.15686275,0.11204482,0.3137255,0.16526611,0.31652662,0.15406163,0.15966387,0.15686275,0.16386555,0.2044818,0.20728292,0.27731094,0.18767507,0.29411766,0.33613446,0.4117647,0.35854343,0.38935575,0.40616247,0.23529412,0.35854343,0.4089636,0.3277311,0.32492998,0.30112046,0.32212886,0.43137255,0.4509804,0.4859944,0.49019608,0.5294118,0.56302524,0.5546219,,0.6302521,0.89915967
+KNN (tuned),0.033613447,0.04481793,0.05042017,0.067226894,0.067226894,0.064425774,0.084033616,0.081232496,0.077030815,0.05882353,0.09243698,0.10084034,0.075630255,0.067226894,0.10084034,0.06162465,0.09243698,0.11204482,0.11764706,0.11204482,0.12044818,0.14845939,0.09663866,0.14285715,0.08963586,0.11204482,0.15686275,0.14845939,0.13165267,0.13725491,0.1764706,0.12885155,0.14705883,0.10644258,0.12605043,0.12605043,0.12464986,0.11764706,0.15406163,0.1232493,0.13445379,0.14425771,0.11204482,0.15686275,0.15126051,0.22128852,0.16526611,0.16806723,0.16806723,0.14845939,0.2044818,0.20868348,0.18207283,0.29691878,0.232493,0.26330534,0.16806723,0.26050422,0.16246499,0.19327731,0.2997199,0.27731094,0.26330534,0.32212886,0.25210086,0.26890758,0.30532214,0.2717087,0.22689076,0.23529412,0.40336135,0.45658264,0.18907563,0.3529412,0.36134455,0.3697479,,0.9411765
+KNN (default),0.008403362,0.005602241,0.005602241,0.008403362,0.008403362,0.011204482,0.011204482,0.011204482,0.014005602,0.008403362,0.008403362,0.014005602,0.011204482,0.008403362,0.005602241,0.019607844,0.008403362,0.022408964,0.028011205,0.008403362,0.014005602,0.019607844,0.011204482,0.033613447,0.011204482,0.008403362,0.025210084,0.025210084,0.014005602,0.014005602,0.028011205,0.022408964,0.030812325,0.016806724,0.030812325,0.011204482,0.028011205,0.011204482,0.033613447,0.028011205,0.039215688,0.025210084,0.030812325,0.030812325,0.05882353,0.033613447,0.05882353,0.035014007,0.042016808,0.030812325,0.05042017,0.05042017,0.042016808,0.084033616,0.04761905,0.081232496,0.014005602,0.064425774,0.016806724,0.078431375,0.10644258,0.05882353,0.075630255,0.11764706,0.05182073,0.072829135,0.09243698,0.070028014,0.04481793,0.042016808,0.1764706,0.13445379,0.039215688,0.05322129,0.072829135,0.10084034,0.05882353,
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_all/leaderboard_overview_explorer.html
index b8ff039f9778a76155c7ef45dd88d088536606bb..379b300566938f1f2603e244d26efec59f27af0c 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Classification | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_classification/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..ef30ef33f39dbc11331e8ae00f45c8e13f148a5b
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_classification/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1744.0,"score":0.871,"rank":7.74,"hrank":1.74,"imp":5.032,"train_time":38.57,"predict_time":7.109,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-115","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1663.0,"score":0.638,"rank":10.87,"hrank":4.34,"imp":9.484,"train_time":267.31,"predict_time":3.981,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-75","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1634.0,"score":0.637,"rank":12.22,"hrank":4.74,"imp":11.818,"train_time":3.64,"predict_time":0.391,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-65","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1586.0,"score":0.552,"rank":14.79,"hrank":8.17,"imp":13.174,"train_time":5.17,"predict_time":0.539,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-59","imputed":false},{"position":4,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1575.0,"score":0.582,"rank":15.46,"hrank":5.66,"imp":12.424,"train_time":2.07,"predict_time":0.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-69","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1562.0,"score":0.51,"rank":16.23,"hrank":7.23,"imp":12.958,"train_time":2046.25,"predict_time":8.969,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-69","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1539.0,"score":0.488,"rank":17.68,"hrank":7.89,"imp":13.628,"train_time":2046.25,"predict_time":1.329,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-61","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1528.0,"score":0.483,"rank":18.42,"hrank":9.91,"imp":13.852,"train_time":5.76,"predict_time":0.794,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-54","imputed":false},{"position":8,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1482.0,"score":0.379,"rank":21.65,"hrank":8.27,"imp":15.075,"train_time":1684.65,"predict_time":1.969,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-53","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1466.0,"score":0.356,"rank":22.88,"hrank":13.52,"imp":15.678,"train_time":2879.46,"predict_time":12.484,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-38","imputed":false},{"position":10,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1446.0,"score":0.313,"rank":24.46,"hrank":12.25,"imp":16.655,"train_time":2168.86,"predict_time":2.347,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-44","imputed":false},{"position":11,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1422.0,"score":0.326,"rank":26.4,"hrank":11.02,"imp":17.124,"train_time":1.94,"predict_time":0.19,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-57","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1414.0,"score":0.255,"rank":27.04,"hrank":18.61,"imp":17.402,"train_time":382.09,"predict_time":1.487,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-33","imputed":false},{"position":13,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1410.0,"score":0.3,"rank":27.36,"hrank":14.82,"imp":17.565,"train_time":8.68,"predict_time":1.742,"imputed_pct":5.26,"hardware":"GPU","elo_ci":"+50\/-56","imputed":true},{"position":14,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1398.0,"score":0.276,"rank":28.43,"hrank":13.82,"imp":17.169,"train_time":4941.98,"predict_time":307.735,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-48","imputed":false},{"position":15,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1397.0,"score":0.258,"rank":28.45,"hrank":15.93,"imp":17.051,"train_time":1107.0,"predict_time":0.329,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-45","imputed":false},{"position":16,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1396.0,"score":0.255,"rank":28.58,"hrank":14.15,"imp":17.511,"train_time":2168.86,"predict_time":0.258,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-45","imputed":false},{"position":17,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1395.0,"score":0.336,"rank":28.66,"hrank":9.3,"imp":17.401,"train_time":18.41,"predict_time":4.667,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-77","imputed":false},{"position":18,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1392.0,"score":0.235,"rank":28.91,"hrank":18.99,"imp":17.453,"train_time":12597.42,"predict_time":502.448,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-43","imputed":false},{"position":19,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1391.0,"score":0.259,"rank":28.96,"hrank":15.23,"imp":17.118,"train_time":2879.46,"predict_time":0.598,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-42","imputed":false},{"position":20,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1387.0,"score":0.242,"rank":29.3,"hrank":16.17,"imp":17.311,"train_time":1107.0,"predict_time":0.038,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-43","imputed":false},{"position":21,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1374.0,"score":0.189,"rank":30.43,"hrank":22.68,"imp":18.081,"train_time":382.09,"predict_time":0.254,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-32","imputed":false},{"position":22,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1374.0,"score":0.241,"rank":30.49,"hrank":15.84,"imp":17.658,"train_time":5.41,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-43","imputed":false},{"position":23,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1363.0,"score":0.203,"rank":31.37,"hrank":20.29,"imp":18.188,"train_time":685.87,"predict_time":1.454,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-47","imputed":false},{"position":24,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1359.0,"score":0.243,"rank":31.77,"hrank":18.65,"imp":19.478,"train_time":282.82,"predict_time":1.886,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+49\/-44","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1350.0,"score":0.183,"rank":32.57,"hrank":17.59,"imp":18.472,"train_time":4880.28,"predict_time":0.525,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-36","imputed":false},{"position":26,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1349.0,"score":0.234,"rank":32.64,"hrank":17.26,"imp":18.544,"train_time":4880.28,"predict_time":8.736,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-75","imputed":false},{"position":27,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1349.0,"score":0.317,"rank":32.66,"hrank":13.5,"imp":19.639,"train_time":121.93,"predict_time":2.774,"imputed_pct":31.58,"hardware":"GPU","elo_ci":"+78\/-81","imputed":true},{"position":28,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1348.0,"score":0.304,"rank":32.7,"hrank":12.77,"imp":19.499,"train_time":3009.72,"predict_time":20.842,"imputed_pct":31.58,"hardware":"GPU","elo_ci":"+84\/-72","imputed":true},{"position":29,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1341.0,"score":0.211,"rank":33.31,"hrank":15.51,"imp":18.558,"train_time":4941.98,"predict_time":41.595,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-49","imputed":false},{"position":30,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1340.0,"score":0.232,"rank":33.44,"hrank":20.57,"imp":19.214,"train_time":1.2,"predict_time":0.067,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-55","imputed":false},{"position":31,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1339.0,"score":0.162,"rank":33.49,"hrank":22.97,"imp":18.556,"train_time":685.87,"predict_time":0.205,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-41","imputed":false},{"position":32,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1333.0,"score":0.204,"rank":34.03,"hrank":19.83,"imp":19.342,"train_time":470.37,"predict_time":0.436,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-58","imputed":false},{"position":33,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1315.0,"score":0.173,"rank":35.63,"hrank":22.02,"imp":19.405,"train_time":934.27,"predict_time":3.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+46\/-45","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1313.0,"score":0.174,"rank":35.87,"hrank":21.43,"imp":19.698,"train_time":470.37,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-57","imputed":false},{"position":35,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1303.0,"score":0.173,"rank":36.78,"hrank":22.03,"imp":19.702,"train_time":6.85,"predict_time":0.246,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-43","imputed":false},{"position":36,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1302.0,"score":0.13,"rank":36.85,"hrank":26.49,"imp":20.206,"train_time":1499.05,"predict_time":0.15,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+36\/-36","imputed":false},{"position":37,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1300.0,"score":0.236,"rank":37.04,"hrank":15.46,"imp":21.139,"train_time":3009.72,"predict_time":0.514,"imputed_pct":31.58,"hardware":"GPU","elo_ci":"+71\/-76","imputed":true},{"position":38,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1298.0,"score":0.136,"rank":37.15,"hrank":27.22,"imp":19.39,"train_time":2389.47,"predict_time":2.157,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-37","imputed":false},{"position":39,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1289.0,"score":0.19,"rank":37.97,"hrank":14.13,"imp":20.931,"train_time":13.7,"predict_time":2.033,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-66","imputed":false},{"position":40,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1285.0,"score":0.108,"rank":38.35,"hrank":24.31,"imp":19.945,"train_time":12597.42,"predict_time":70.851,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+40\/-36","imputed":false},{"position":41,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1278.0,"score":0.13,"rank":38.97,"hrank":23.84,"imp":20.742,"train_time":934.27,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-52","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1276.0,"score":0.148,"rank":39.2,"hrank":22.32,"imp":20.847,"train_time":1.86,"predict_time":0.035,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-63","imputed":false},{"position":43,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1276.0,"score":0.22,"rank":39.2,"hrank":17.66,"imp":21.649,"train_time":3.37,"predict_time":0.315,"imputed_pct":31.58,"hardware":"GPU","elo_ci":"+69\/-72","imputed":true},{"position":44,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1268.0,"score":0.092,"rank":39.87,"hrank":29.14,"imp":20.886,"train_time":1499.05,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-38","imputed":false},{"position":45,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1262.0,"score":0.157,"rank":40.42,"hrank":20.68,"imp":21.315,"train_time":49.21,"predict_time":43.824,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-54","imputed":false},{"position":46,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1251.0,"score":0.127,"rank":41.42,"hrank":23.88,"imp":21.264,"train_time":618.96,"predict_time":4.766,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-61","imputed":false},{"position":47,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.082,"rank":42.05,"hrank":24.79,"imp":22.181,"train_time":14.78,"predict_time":0.346,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+40\/-42","imputed":false},{"position":48,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1244.0,"score":0.085,"rank":42.1,"hrank":28.09,"imp":21.835,"train_time":5.72,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-43","imputed":false},{"position":49,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1232.0,"score":0.06,"rank":43.12,"hrank":31.24,"imp":21.435,"train_time":10.48,"predict_time":1.707,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+36\/-32","imputed":false},{"position":50,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1225.0,"score":0.068,"rank":43.76,"hrank":31.48,"imp":21.202,"train_time":1.77,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-51","imputed":false},{"position":51,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1221.0,"score":0.07,"rank":44.11,"hrank":33.0,"imp":21.425,"train_time":2389.47,"predict_time":0.152,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-36","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1209.0,"score":0.071,"rank":45.2,"hrank":30.1,"imp":22.216,"train_time":286.48,"predict_time":0.764,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-49","imputed":false},{"position":53,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1188.0,"score":0.054,"rank":47.01,"hrank":30.03,"imp":23.005,"train_time":618.96,"predict_time":0.298,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-59","imputed":false},{"position":54,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1184.0,"score":0.053,"rank":47.3,"hrank":38.73,"imp":22.169,"train_time":1.79,"predict_time":0.12,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-39","imputed":false},{"position":55,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1181.0,"score":0.072,"rank":47.58,"hrank":28.05,"imp":23.269,"train_time":323.8,"predict_time":0.743,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-59","imputed":false},{"position":56,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1180.0,"score":0.063,"rank":47.66,"hrank":28.07,"imp":23.194,"train_time":286.48,"predict_time":0.073,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-52","imputed":false},{"position":57,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1148.0,"score":0.054,"rank":50.41,"hrank":32.38,"imp":24.512,"train_time":262.93,"predict_time":64.506,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-54","imputed":false},{"position":58,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1147.0,"score":0.05,"rank":50.48,"hrank":27.1,"imp":24.053,"train_time":323.8,"predict_time":0.076,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-59","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1133.0,"score":0.126,"rank":51.6,"hrank":15.2,"imp":26.329,"train_time":29479.47,"predict_time":4.088,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-88","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1129.0,"score":0.121,"rank":51.95,"hrank":14.74,"imp":26.406,"train_time":29479.47,"predict_time":18.315,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-86","imputed":false},{"position":61,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1116.0,"score":0.069,"rank":53.03,"hrank":36.95,"imp":26.378,"train_time":16.32,"predict_time":2.859,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-59","imputed":false},{"position":62,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1090.0,"score":0.01,"rank":55.01,"hrank":41.55,"imp":25.554,"train_time":6.83,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-50","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1064.0,"score":0.025,"rank":57.02,"hrank":42.0,"imp":31.069,"train_time":187.03,"predict_time":0.955,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-60","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1043.0,"score":0.009,"rank":58.46,"hrank":40.41,"imp":27.726,"train_time":2.91,"predict_time":0.368,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-70","imputed":false},{"position":65,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1038.0,"score":0.043,"rank":58.8,"hrank":40.05,"imp":32.44,"train_time":252.6,"predict_time":0.528,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-85","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1030.0,"score":0.08,"rank":59.36,"hrank":23.79,"imp":31.861,"train_time":439.45,"predict_time":5.883,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-108","imputed":false},{"position":67,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1030.0,"score":0.014,"rank":59.36,"hrank":42.03,"imp":32.523,"train_time":187.03,"predict_time":0.338,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-57","imputed":false},{"position":68,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1025.0,"score":0.023,"rank":59.69,"hrank":39.01,"imp":30.255,"train_time":139.88,"predict_time":3.615,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-89","imputed":false},{"position":69,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1014.0,"score":0.045,"rank":60.47,"hrank":40.36,"imp":31.706,"train_time":1.11,"predict_time":0.189,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-90","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1013.0,"score":0.028,"rank":60.5,"hrank":42.39,"imp":33.111,"train_time":252.6,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-84","imputed":false},{"position":71,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1006.0,"score":0.019,"rank":60.93,"hrank":52.55,"imp":30.886,"train_time":3.61,"predict_time":0.938,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-89","imputed":false},{"position":72,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.003,"rank":61.34,"hrank":55.09,"imp":29.887,"train_time":0.38,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-59","imputed":false},{"position":73,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":992.0,"score":0.007,"rank":61.82,"hrank":54.76,"imp":29.888,"train_time":1.68,"predict_time":0.295,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-67","imputed":false},{"position":74,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":937.0,"score":0.017,"rank":65.0,"hrank":47.3,"imp":35.792,"train_time":1.29,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-99","imputed":false},{"position":75,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":924.0,"score":0.0,"rank":65.67,"hrank":47.35,"imp":36.631,"train_time":26.07,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-76","imputed":false},{"position":76,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":912.0,"score":0.012,"rank":66.28,"hrank":48.69,"imp":36.098,"train_time":139.88,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-107","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":585.0,"score":0.0,"rank":75.85,"hrank":70.4,"imp":51.396,"train_time":0.18,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-106","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_all/pareto_front_explorer.html
index a9db3af6b4d402b4b3a207baa663ce893ba319f3..fd09cf7034f4782b48d0b2c836f1456783c8028e 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8858780824,"imp":19.4781511855,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0255239609,"imp":17.6584995763,"elo":1373.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0377219356,"imp":17.3114054341,"elo":1387.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3286022328,"imp":17.0508763065,"elo":1397.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0349046894,"imp":20.8467320099,"elo":1275.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0333867141,"imp":19.6977805958,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4361928292,"imp":19.3417219344,"elo":1333.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0170265849,"imp":21.8346748016,"elo":1243.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0172938338,"imp":20.8858910267,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1503678683,"imp":20.2063216946,"elo":1301.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2946660708,"imp":29.8882976088,"elo":992.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0734711664,"imp":23.1936262865,"elo":1180.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7641423489,"imp":22.2156155364,"elo":1208.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3681069516,"imp":27.7264256769,"elo":1043.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2978802219,"imp":23.0045987086,"elo":1187.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.7655686359,"imp":21.2638385308,"elo":1251.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1204998116,"imp":22.1687992369,"elo":1184.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.253630478,"imp":18.0805448227,"elo":1374.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.4869408199,"imp":17.4017806563,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.6674634265,"imp":17.401470661,"elo":1394.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.774247901,"imp":19.6391627954,"elo":1348.6,"imputed":true,"imputed_pct":31.5789473684,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"imp":22.1813308679,"elo":1244.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5246167968,"imp":18.472106187,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7364459708,"imp":18.5442362137,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.8592422653,"imp":26.3775716195,"elo":1115.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0410553489,"imp":36.6310009628,"elo":924.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3381841475,"imp":32.5230163415,"elo":1030.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9547740596,"imp":31.0687661551,"elo":1063.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0372103395,"imp":29.8871397078,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0764179217,"imp":24.0529799071,"elo":1147.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7428875097,"imp":23.2692645989,"elo":1181.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7071575933,"imp":21.4353948092,"elo":1232.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5977456464,"imp":17.1182598636,"elo":1391.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":12.4836618956,"imp":15.6779789904,"elo":1465.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7935168864,"imp":13.8523997141,"elo":1527.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3289068866,"imp":13.627711878,"elo":1539.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9691710732,"imp":12.9579453132,"elo":1561.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0331410808,"imp":20.9314144248,"elo":1289.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"imp":21.3151500481,"elo":1262.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":41.5948089023,"imp":18.5580210863,"elo":1341.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":307.7353108496,"imp":17.1691130683,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1897902902,"imp":17.1244350047,"elo":1421.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"imp":5.0315562877,"elo":1744.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1888600498,"imp":31.7063587768,"elo":1013.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.7422345138,"imp":17.5653695571,"elo":1410.2,"imputed":true,"imputed_pct":5.2631578947,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"imp":12.4243948176,"elo":1574.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2459026963,"imp":19.7015760384,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2584066824,"imp":17.5108473161,"elo":1395.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.3471511219,"imp":16.654643155,"elo":1445.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"imp":13.1736739051,"elo":1586.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3910008352,"imp":11.8177075259,"elo":1634.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3152166325,"imp":21.6494904848,"elo":1275.7,"imputed":true,"imputed_pct":31.5789473684,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5144045305,"imp":21.1385984442,"elo":1299.7,"imputed":true,"imputed_pct":31.5789473684,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":20.8416670968,"imp":19.4992163191,"elo":1348.2,"imputed":true,"imputed_pct":31.5789473684,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.8830380236,"imp":31.8613750421,"elo":1030.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0878026987,"imp":26.3285807626,"elo":1133.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3145858894,"imp":26.4060718241,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0665397304,"imp":19.2136673684,"elo":1339.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1470302095,"imp":25.5543683762,"elo":1090.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1517792148,"imp":21.4252312884,"elo":1221.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1570271396,"imp":19.390051276,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.117054282,"imp":21.2018496986,"elo":1225.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2050537617,"imp":18.5558278444,"elo":1339.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.4544829944,"imp":18.1882373275,"elo":1363.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":64.5055925989,"imp":24.511556692,"elo":1147.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":70.8512833361,"imp":19.9445022629,"elo":1285.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":502.4479719176,"imp":17.4528856256,"elo":1391.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9382527357,"imp":30.8864343028,"elo":1006.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1687223142,"imp":20.741676615,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.0619945914,"imp":19.4047627702,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8858780824,"x_train":282.8166166302,"imp":19.4781511855,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0255239609,"x_train":5.4086527682,"imp":17.6584995763,"elo":1373.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0377219356,"x_train":1106.9971386445,"imp":17.3114054341,"elo":1387.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3286022328,"x_train":1106.9971386445,"imp":17.0508763065,"elo":1397.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0349046894,"x_train":1.857962315,"imp":20.8467320099,"elo":1275.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0333867141,"x_train":470.3718666937,"imp":19.6977805958,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4361928292,"x_train":470.3718666937,"imp":19.3417219344,"elo":1333.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0170265849,"x_train":5.7213239382,"imp":21.8346748016,"elo":1243.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0172938338,"x_train":1499.0498932557,"imp":20.8858910267,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1503678683,"x_train":1499.0498932557,"imp":20.2063216946,"elo":1301.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2946660708,"x_train":1.6841584145,"imp":29.8882976088,"elo":992.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0734711664,"x_train":286.4783954015,"imp":23.1936262865,"elo":1180.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7641423489,"x_train":286.4783954015,"imp":22.2156155364,"elo":1208.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3681069516,"x_train":2.9123333331,"imp":27.7264256769,"elo":1043.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2978802219,"x_train":618.9598462323,"imp":23.0045987086,"elo":1187.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.7655686359,"x_train":618.9598462323,"imp":21.2638385308,"elo":1251.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1204998116,"x_train":1.7915100586,"imp":22.1687992369,"elo":1184.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.253630478,"x_train":382.0935074289,"imp":18.0805448227,"elo":1374.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.4869408199,"x_train":382.0935074289,"imp":17.4017806563,"elo":1414.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.6674634265,"x_train":18.4149364165,"imp":17.401470661,"elo":1394.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.774247901,"x_train":121.9251095684,"imp":19.6391627954,"elo":1348.6,"imputed":true,"imputed_pct":31.5789473684,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"x_train":14.7795222568,"imp":22.1813308679,"elo":1244.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5246167968,"x_train":4880.2786288265,"imp":18.472106187,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7364459708,"x_train":4880.2786288265,"imp":18.5442362137,"elo":1348.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.8592422653,"x_train":16.3156181032,"imp":26.3775716195,"elo":1115.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0410553489,"x_train":26.0689936509,"imp":36.6310009628,"elo":924.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3381841475,"x_train":187.0264910689,"imp":32.5230163415,"elo":1030.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9547740596,"x_train":187.0264910689,"imp":31.0687661551,"elo":1063.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0372103395,"x_train":0.3814409026,"imp":29.8871397078,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0764179217,"x_train":323.8039447074,"imp":24.0529799071,"elo":1147.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7428875097,"x_train":323.8039447074,"imp":23.2692645989,"elo":1181.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7071575933,"x_train":10.4758857753,"imp":21.4353948092,"elo":1232.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5977456464,"x_train":2879.4580429663,"imp":17.1182598636,"elo":1391.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":12.4836618956,"x_train":2879.4580429663,"imp":15.6779789904,"elo":1465.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7935168864,"x_train":5.7589146636,"imp":13.8523997141,"elo":1527.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3289068866,"x_train":2046.2534629078,"imp":13.627711878,"elo":1539.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9691710732,"x_train":2046.2534629078,"imp":12.9579453132,"elo":1561.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0331410808,"x_train":13.7043041236,"imp":20.9314144248,"elo":1289.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"x_train":49.2127895844,"imp":21.3151500481,"elo":1262.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":41.5948089023,"x_train":4941.9777879175,"imp":18.5580210863,"elo":1341.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":307.7353108496,"x_train":4941.9777879175,"imp":17.1691130683,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1897902902,"x_train":1.9424208612,"imp":17.1244350047,"elo":1421.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"x_train":38.5654426559,"imp":5.0315562877,"elo":1744.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1888600498,"x_train":1.1127929815,"imp":31.7063587768,"elo":1013.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.7422345138,"x_train":8.6842463409,"imp":17.5653695571,"elo":1410.2,"imputed":true,"imputed_pct":5.2631578947,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"x_train":2.070384719,"imp":12.4243948176,"elo":1574.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2459026963,"x_train":6.8460080835,"imp":19.7015760384,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2584066824,"x_train":2168.8617901524,"imp":17.5108473161,"elo":1395.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.3471511219,"x_train":2168.8617901524,"imp":16.654643155,"elo":1445.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"x_train":5.1697320864,"imp":13.1736739051,"elo":1586.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3910008352,"x_train":3.6410702833,"imp":11.8177075259,"elo":1634.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3152166325,"x_train":3.3689389916,"imp":21.6494904848,"elo":1275.7,"imputed":true,"imputed_pct":31.5789473684,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5144045305,"x_train":3009.7234478495,"imp":21.1385984442,"elo":1299.7,"imputed":true,"imputed_pct":31.5789473684,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":20.8416670968,"x_train":3009.7234478495,"imp":19.4992163191,"elo":1348.2,"imputed":true,"imputed_pct":31.5789473684,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.8830380236,"x_train":439.4466040087,"imp":31.8613750421,"elo":1030.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0878026987,"x_train":29479.468606441,"imp":26.3285807626,"elo":1133.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3145858894,"x_train":29479.468606441,"imp":26.4060718241,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0665397304,"x_train":1.2028502699,"imp":19.2136673684,"elo":1339.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1470302095,"x_train":6.834853082,"imp":25.5543683762,"elo":1090.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1517792148,"x_train":2389.4741835732,"imp":21.4252312884,"elo":1221.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1570271396,"x_train":2389.4741835732,"imp":19.390051276,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.117054282,"x_train":1.7712088618,"imp":21.2018496986,"elo":1225.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2050537617,"x_train":685.8651054054,"imp":18.5558278444,"elo":1339.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.4544829944,"x_train":685.8651054054,"imp":18.1882373275,"elo":1363.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":64.5055925989,"x_train":262.9267967989,"imp":24.511556692,"elo":1147.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":70.8512833361,"x_train":12597.4213063423,"imp":19.9445022629,"elo":1285.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":502.4479719176,"x_train":12597.4213063423,"imp":17.4528856256,"elo":1391.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9382527357,"x_train":3.607105251,"imp":30.8864343028,"elo":1006.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1687223142,"x_train":934.2696678038,"imp":20.741676615,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.0619945914,"x_train":934.2696678038,"imp":19.4047627702,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_all/tuning_trajectories_explorer.html
index 10f4f0c242e0fe05f64cb0133cc3987d61ba3e5a..274a25f96c78017c2a87c383047eab79717040e3 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":282.8166166302,"imp":19.5021442391,"elo":1376.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":5.4086527682,"imp":17.681452422,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":10.084780712,"imp":17.2753848455,"elo":1407.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":24.9344211319,"imp":17.0934208517,"elo":1418.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":49.1678167274,"imp":17.0428486335,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":137.1191573934,"imp":16.9828456014,"elo":1423.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":271.2351529076,"imp":16.9898441141,"elo":1422.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":546.8626199056,"imp":17.0256041083,"elo":1418.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":821.0408680788,"imp":17.054980386,"elo":1417.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1106.9971386445,"imp":17.0697102602,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.857962315,"imp":20.8693569126,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.2122341586,"imp":20.4521027083,"elo":1311.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.9434770322,"imp":20.221097504,"elo":1322.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.49430198,"imp":20.0417645626,"elo":1329.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":57.9467632245,"imp":19.7650171022,"elo":1338.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":116.8428883394,"imp":19.6245873802,"elo":1344.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":234.1123581247,"imp":19.4642893894,"elo":1349.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":351.0170231939,"imp":19.3980454986,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":470.3718666937,"imp":19.3631465938,"elo":1353.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":5.7213239382,"imp":21.8568875958,"elo":1248.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":13.9986026157,"imp":21.4502454757,"elo":1261.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":37.2821101451,"imp":21.003002484,"elo":1280.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":74.1266951378,"imp":20.8338243362,"elo":1287.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":187.3186015742,"imp":20.6260411535,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":376.118404862,"imp":20.4658550113,"elo":1304.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":747.8066316207,"imp":20.3264793205,"elo":1311.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1119.3701634608,"imp":20.2794273524,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1499.0498932557,"imp":20.2307891086,"elo":1316.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.6841584145,"imp":29.9088374079,"elo":991.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1012271428,"imp":27.6242473627,"elo":1044.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.347596203,"imp":24.3333806455,"elo":1123.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.4431649541,"imp":23.2363303896,"elo":1168.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":35.7497456308,"imp":22.570456269,"elo":1198.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":71.3524964885,"imp":22.3960848662,"elo":1207.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":142.6147774706,"imp":22.2738464802,"elo":1214.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":213.7904556538,"imp":22.2246003471,"elo":1218.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":286.4783954015,"imp":22.2384970602,"elo":1218.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.9123333331,"imp":27.747484742,"elo":1050.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.8242862424,"imp":26.2294819393,"elo":1093.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":15.2779819431,"imp":24.0402651406,"elo":1165.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":31.0766737178,"imp":22.9904481327,"elo":1200.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":75.7672592237,"imp":22.1129694021,"elo":1231.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":152.1226481198,"imp":21.6998655135,"elo":1248.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":306.7750654563,"imp":21.4685399955,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":460.9921495532,"imp":21.3328669657,"elo":1262.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":618.9598462323,"imp":21.2803442406,"elo":1265.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":1.7915100586,"imp":22.190553817,"elo":1187.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":3.9851860484,"imp":20.3587740118,"elo":1273.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":9.4702073578,"imp":19.1194044693,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":18.3526064623,"imp":18.480777701,"elo":1385.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":46.5150486041,"imp":17.9152581739,"elo":1410.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":94.2919114747,"imp":17.6848486667,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":189.6982542539,"imp":17.516978021,"elo":1431.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":283.010042869,"imp":17.4574036598,"elo":1433.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":382.0935074289,"imp":17.4259974052,"elo":1435.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":18.4149364165,"imp":17.3388666384,"elo":1414.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":121.9251095684,"imp":19.6618995321,"elo":1367.4,"imputed":true,"imputed_pct":31.5789473684,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.7795222568,"imp":22.2033664552,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":44.6910607231,"imp":20.7198362942,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":123.8073281883,"imp":18.5488024098,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":240.4444302535,"imp":18.0700179842,"elo":1389.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":579.7896207628,"imp":17.586829204,"elo":1408.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1211.3449017346,"imp":17.7156541155,"elo":1401.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2400.9271967121,"imp":18.121320015,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3593.5581996112,"imp":18.4154364699,"elo":1375.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4880.2786288265,"imp":18.574597391,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":16.3156181032,"imp":26.4094685413,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":26.0689936509,"imp":37.3022427898,"elo":920.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":61.4802624456,"imp":35.1077084247,"elo":978.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":157.8952583504,"imp":31.2693145373,"elo":1068.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":187.0264910689,"imp":31.0843452702,"elo":1074.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3814409026,"imp":29.9209070406,"elo":997.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.0007658996,"imp":26.5685711497,"elo":1072.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":6.8718138048,"imp":24.9409902689,"elo":1115.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":14.972210111,"imp":24.1559839536,"elo":1143.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":39.102050789,"imp":23.5841139233,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":79.423821505,"imp":23.4357719073,"elo":1182.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":160.2640051076,"imp":23.3370691239,"elo":1187.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":241.0813778596,"imp":23.2919590494,"elo":1189.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":323.8039447074,"imp":23.2884320035,"elo":1190.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.4758857753,"imp":21.4571602864,"elo":1237.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.912673739,"imp":18.927462289,"elo":1346.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":67.3088373681,"imp":17.3346844655,"elo":1420.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":140.0742273376,"imp":16.7328061306,"elo":1447.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":358.3731533565,"imp":16.2624876847,"elo":1468.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":721.2122449282,"imp":16.0525248752,"elo":1478.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1434.2417543333,"imp":15.8640818198,"elo":1484.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2147.1635778763,"imp":15.7712978901,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2879.4580429663,"imp":15.7023392906,"elo":1489.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.7589146636,"imp":13.8772239734,"elo":1543.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.7323897082,"imp":13.9214383398,"elo":1545.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":46.0214163694,"imp":13.7294707272,"elo":1555.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":96.2671462093,"imp":13.6176059808,"elo":1565.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":246.3500390069,"imp":13.4178456471,"elo":1576.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":501.1624695611,"imp":13.2341708315,"elo":1581.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1017.6750647229,"imp":13.1108733505,"elo":1582.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1526.8431451139,"imp":13.0476620208,"elo":1581.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2046.2534629078,"imp":12.9897488878,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.7043041236,"imp":20.952758968,"elo":1303.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":49.2127895844,"imp":21.3354649151,"elo":1273.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":71.8416880018,"imp":20.3761596473,"elo":1302.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":133.3966980893,"imp":19.4002219227,"elo":1341.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":245.8810811274,"imp":19.0079250636,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":600.6176271602,"imp":18.1802093421,"elo":1385.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1226.0480499304,"imp":17.7552605732,"elo":1400.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2452.678200464,"imp":17.4768423776,"elo":1410.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3677.3248587818,"imp":17.2905410635,"elo":1415.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4941.9777879175,"imp":17.1955748357,"elo":1417.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":1.9424208612,"imp":17.1459428188,"elo":1437.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.5654426559,"imp":5.0516807424,"elo":1761.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.1127929815,"imp":31.7240656427,"elo":1026.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":8.6842463409,"imp":17.588708878,"elo":1427.9,"imputed":true,"imputed_pct":5.2631578947,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":2.070384719,"imp":12.4495759161,"elo":1594.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.8460080835,"imp":19.7245104543,"elo":1314.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":15.9731517787,"imp":19.1587737375,"elo":1340.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":48.3100942842,"imp":18.4099433312,"elo":1376.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":104.5645791303,"imp":17.8142159975,"elo":1409.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":264.7537616657,"imp":17.3117788731,"elo":1432.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":533.594254715,"imp":17.023615961,"elo":1446.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1058.8890383576,"imp":16.8216716507,"elo":1457.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1605.9821816147,"imp":16.7322425642,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2168.8617901524,"imp":16.6770245123,"elo":1466.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.1697320864,"imp":13.1982203172,"elo":1610.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.6410702833,"imp":11.8381062408,"elo":1655.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3.3689389916,"imp":21.6715869141,"elo":1289.9,"imputed":true,"imputed_pct":31.5789473684,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":22.8408812219,"imp":21.1335243364,"elo":1315.0,"imputed":true,"imputed_pct":31.5789473684,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":63.3712068389,"imp":20.5597852103,"elo":1342.0,"imputed":true,"imputed_pct":31.5789473684,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":143.4062804674,"imp":20.1535706193,"elo":1354.0,"imputed":true,"imputed_pct":31.5789473684,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":359.6193496539,"imp":19.8375857059,"elo":1362.8,"imputed":true,"imputed_pct":31.5789473684,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":730.2395031165,"imp":19.6592172618,"elo":1368.3,"imputed":true,"imputed_pct":31.5789473684,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1484.9037988086,"imp":19.5670748892,"elo":1369.9,"imputed":true,"imputed_pct":31.5789473684,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2231.7160487129,"imp":19.5250745447,"elo":1370.3,"imputed":true,"imputed_pct":31.5789473684,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3009.7234478495,"imp":19.5017676353,"elo":1369.1,"imputed":true,"imputed_pct":31.5789473684,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":439.4466040087,"imp":31.8812607842,"elo":1038.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":670.2438710225,"imp":31.7426000562,"elo":1045.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1014.6754881516,"imp":31.4554004745,"elo":1051.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1638.3361981714,"imp":30.7301729607,"elo":1067.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3535.8153662444,"imp":29.3747032409,"elo":1090.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":7084.5550409751,"imp":28.0029570837,"elo":1111.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":14613.5276339936,"imp":27.2507251208,"elo":1126.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":21800.1101862642,"imp":26.7912606708,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":29479.468606441,"imp":26.4278269859,"elo":1147.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.2028502699,"imp":19.2361448754,"elo":1357.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.834853082,"imp":25.5684336816,"elo":1098.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":16.5429640262,"imp":23.4765199655,"elo":1157.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":54.7346146645,"imp":21.7767985052,"elo":1208.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":118.7324150095,"imp":20.983642463,"elo":1243.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":292.558905056,"imp":20.3163156133,"elo":1271.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":595.2571207683,"imp":19.914008022,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1187.6701815634,"imp":19.6779081554,"elo":1303.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1789.3376646728,"imp":19.5396221068,"elo":1311.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2389.4741835732,"imp":19.4066856353,"elo":1317.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.7712088618,"imp":21.2238353336,"elo":1229.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.6564696531,"imp":19.5080345279,"elo":1310.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":16.9650142951,"imp":19.0308823467,"elo":1339.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":32.997686743,"imp":18.7281864288,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.160387483,"imp":18.493258509,"elo":1367.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":172.4620957623,"imp":18.3988160493,"elo":1372.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":343.1961190306,"imp":18.3211345065,"elo":1376.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":511.4517561336,"imp":18.2638274509,"elo":1380.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":685.8651054054,"imp":18.2117061522,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":262.9267967989,"imp":24.5324865248,"elo":1155.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":698.1262023376,"imp":22.330374633,"elo":1207.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2185.405093286,"imp":19.9492900804,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4868.6453358671,"imp":18.7710008474,"elo":1352.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12038.5904051722,"imp":17.5022100654,"elo":1409.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12597.4213063423,"imp":17.4751652248,"elo":1411.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.607105251,"imp":30.9061240061,"elo":1011.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":9.0537118684,"imp":27.0168064854,"elo":1082.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":23.1440570954,"imp":22.305394137,"elo":1213.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":46.3767226799,"imp":20.4114497336,"elo":1296.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":116.4989862715,"imp":19.6799802589,"elo":1330.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":232.7740378227,"imp":19.4556420191,"elo":1338.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":465.4926636368,"imp":19.4015327486,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":697.1621157189,"imp":19.4044640333,"elo":1335.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":934.2696678038,"imp":19.4156500684,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8858780824,"x_train":282.8166166302,"imp":19.5021442391,"elo":1376.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0255239609,"x_train":5.4086527682,"imp":17.681452422,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0519447497,"x_train":10.084780712,"imp":17.2753848455,"elo":1407.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1089050066,"x_train":24.9344211319,"imp":17.0934208517,"elo":1418.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1494108066,"x_train":49.1678167274,"imp":17.0428486335,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2091965714,"x_train":137.1191573934,"imp":16.9828456014,"elo":1423.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2443429842,"x_train":271.2351529076,"imp":16.9898441141,"elo":1422.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2858799211,"x_train":546.8626199056,"imp":17.0256041083,"elo":1418.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3088118892,"x_train":821.0408680788,"imp":17.054980386,"elo":1417.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.33304612,"x_train":1106.9971386445,"imp":17.0697102602,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0349046894,"x_train":1.857962315,"imp":20.8693569126,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0727101961,"x_train":4.2122341586,"imp":20.4521027083,"elo":1311.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1347246305,"x_train":10.9434770322,"imp":20.221097504,"elo":1322.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1933333513,"x_train":22.49430198,"imp":20.0417645626,"elo":1329.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2718991335,"x_train":57.9467632245,"imp":19.7650171022,"elo":1338.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3291196359,"x_train":116.8428883394,"imp":19.6245873802,"elo":1344.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3904301416,"x_train":234.1123581247,"imp":19.4642893894,"elo":1349.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4239634557,"x_train":351.0170231939,"imp":19.3980454986,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4227445213,"x_train":470.3718666937,"imp":19.3631465938,"elo":1353.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0170265849,"x_train":5.7213239382,"imp":21.8568875958,"elo":1248.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333155319,"x_train":13.9986026157,"imp":21.4502454757,"elo":1261.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0583218862,"x_train":37.2821101451,"imp":21.003002484,"elo":1280.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.078412591,"x_train":74.1266951378,"imp":20.8338243362,"elo":1287.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1031236538,"x_train":187.3186015742,"imp":20.6260411535,"elo":1296.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1212226385,"x_train":376.118404862,"imp":20.4658550113,"elo":1304.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1403339145,"x_train":747.8066316207,"imp":20.3264793205,"elo":1311.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1466477215,"x_train":1119.3701634608,"imp":20.2794273524,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1440681396,"x_train":1499.0498932557,"imp":20.2307891086,"elo":1316.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2946660708,"x_train":1.6841584145,"imp":29.9088374079,"elo":991.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3055075075,"x_train":3.1012271428,"imp":27.6242473627,"elo":1044.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3457056156,"x_train":7.347596203,"imp":24.3333806455,"elo":1123.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3944436772,"x_train":14.4431649541,"imp":23.2363303896,"elo":1168.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5014192887,"x_train":35.7497456308,"imp":22.570456269,"elo":1198.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5959923774,"x_train":71.3524964885,"imp":22.3960848662,"elo":1207.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6717315754,"x_train":142.6147774706,"imp":22.2738464802,"elo":1214.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7207892432,"x_train":213.7904556538,"imp":22.2246003471,"elo":1218.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7589964064,"x_train":286.4783954015,"imp":22.2384970602,"elo":1218.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3681069516,"x_train":2.9123333331,"imp":27.747484742,"elo":1050.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.608277041,"x_train":5.8242862424,"imp":26.2294819393,"elo":1093.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1293887702,"x_train":15.2779819431,"imp":24.0402651406,"elo":1165.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7335087318,"x_train":31.0766737178,"imp":22.9904481327,"elo":1200.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4971865346,"x_train":75.7672592237,"imp":22.1129694021,"elo":1231.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.2671043117,"x_train":152.1226481198,"imp":21.6998655135,"elo":1248.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0044565763,"x_train":306.7750654563,"imp":21.4685399955,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4204326806,"x_train":460.9921495532,"imp":21.3328669657,"elo":1262.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.7390931271,"x_train":618.9598462323,"imp":21.2803442406,"elo":1265.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1204998116,"x_train":1.7915100586,"imp":22.190553817,"elo":1187.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4762849552,"x_train":3.9851860484,"imp":20.3587740118,"elo":1273.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7292175393,"x_train":9.4702073578,"imp":19.1194044693,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9391818478,"x_train":18.3526064623,"imp":18.480777701,"elo":1385.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2648300791,"x_train":46.5150486041,"imp":17.9152581739,"elo":1410.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4080976208,"x_train":94.2919114747,"imp":17.6848486667,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4975096766,"x_train":189.6982542539,"imp":17.516978021,"elo":1431.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4963379792,"x_train":283.010042869,"imp":17.4574036598,"elo":1433.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5345635664,"x_train":382.0935074289,"imp":17.4259974052,"elo":1435.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.6674634265,"x_train":18.4149364165,"imp":17.3388666384,"elo":1414.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.774247901,"x_train":121.9251095684,"imp":19.6618995321,"elo":1367.4,"imputed":true,"imputed_pct":31.5789473684,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"x_train":14.7795222568,"imp":22.2033664552,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8804757883,"x_train":44.6910607231,"imp":20.7198362942,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9515166073,"x_train":123.8073281883,"imp":18.5488024098,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9139871186,"x_train":240.4444302535,"imp":18.0700179842,"elo":1389.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.2991192179,"x_train":579.7896207628,"imp":17.586829204,"elo":1408.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.6324379245,"x_train":1211.3449017346,"imp":17.7156541155,"elo":1401.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.2170906616,"x_train":2400.9271967121,"imp":18.121320015,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.2545907476,"x_train":3593.5581996112,"imp":18.4154364699,"elo":1375.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7287422808,"x_train":4880.2786288265,"imp":18.574597391,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.8592422653,"x_train":16.3156181032,"imp":26.4094685413,"elo":1126.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0410553489,"x_train":26.0689936509,"imp":37.3022427898,"elo":920.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3573866152,"x_train":61.4802624456,"imp":35.1077084247,"elo":978.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.8506460976,"x_train":157.8952583504,"imp":31.2693145373,"elo":1068.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9547740596,"x_train":187.0264910689,"imp":31.0843452702,"elo":1074.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0372103395,"x_train":0.3814409026,"imp":29.9209070406,"elo":997.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1067923159,"x_train":2.0007658996,"imp":26.5685711497,"elo":1072.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2162743727,"x_train":6.8718138048,"imp":24.9409902689,"elo":1115.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3276673648,"x_train":14.972210111,"imp":24.1559839536,"elo":1143.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4915865441,"x_train":39.102050789,"imp":23.5841139233,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.582457396,"x_train":79.423821505,"imp":23.4357719073,"elo":1182.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6587344304,"x_train":160.2640051076,"imp":23.3370691239,"elo":1187.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7122437923,"x_train":241.0813778596,"imp":23.2919590494,"elo":1189.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7428776826,"x_train":323.8039447074,"imp":23.2884320035,"elo":1190.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7071575933,"x_train":10.4758857753,"imp":21.4571602864,"elo":1237.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.4603108225,"x_train":24.912673739,"imp":18.927462289,"elo":1346.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.7287224795,"x_train":67.3088373681,"imp":17.3346844655,"elo":1420.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.7139024305,"x_train":140.0742273376,"imp":16.7328061306,"elo":1447.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.8490124424,"x_train":358.3731533565,"imp":16.2624876847,"elo":1468.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.7712471898,"x_train":721.2122449282,"imp":16.0525248752,"elo":1478.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.4105608019,"x_train":1434.2417543333,"imp":15.8640818198,"elo":1484.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.2019759137,"x_train":2147.1635778763,"imp":15.7712978901,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.6005448312,"x_train":2879.4580429663,"imp":15.7023392906,"elo":1489.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7935168864,"x_train":5.7589146636,"imp":13.8772239734,"elo":1543.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.5040919517,"x_train":16.7323897082,"imp":13.9214383398,"elo":1545.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.0181434206,"x_train":46.0214163694,"imp":13.7294707272,"elo":1555.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1961228236,"x_train":96.2671462093,"imp":13.6176059808,"elo":1565.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7695458497,"x_train":246.3500390069,"imp":13.4178456471,"elo":1576.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.0980977495,"x_train":501.1624695611,"imp":13.2341708315,"elo":1581.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3132120072,"x_train":1017.6750647229,"imp":13.1108733505,"elo":1582.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.2061952514,"x_train":1526.8431451139,"imp":13.0476620208,"elo":1581.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.0822605356,"x_train":2046.2534629078,"imp":12.9897488878,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0331410808,"x_train":13.7043041236,"imp":20.952758968,"elo":1303.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.8239520304,"x_train":49.2127895844,"imp":21.3354649151,"elo":1273.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":57.3203633186,"x_train":71.8416880018,"imp":20.3761596473,"elo":1302.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":83.2470200097,"x_train":133.3966980893,"imp":19.4002219227,"elo":1341.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":114.641595606,"x_train":245.8810811274,"imp":19.0079250636,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":172.6551856784,"x_train":600.6176271602,"imp":18.1802093421,"elo":1385.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":232.5924596656,"x_train":1226.0480499304,"imp":17.7552605732,"elo":1400.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":270.3155274697,"x_train":2452.678200464,"imp":17.4768423776,"elo":1410.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":289.6259857616,"x_train":3677.3248587818,"imp":17.2905410635,"elo":1415.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":311.2606556608,"x_train":4941.9777879175,"imp":17.1955748357,"elo":1417.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1897902902,"x_train":1.9424208612,"imp":17.1459428188,"elo":1437.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.1085269615,"x_train":38.5654426559,"imp":5.0516807424,"elo":1761.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1888600498,"x_train":1.1127929815,"imp":31.7240656427,"elo":1026.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.7422345138,"x_train":8.6842463409,"imp":17.588708878,"elo":1427.9,"imputed":true,"imputed_pct":5.2631578947,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"x_train":2.070384719,"imp":12.4495759161,"elo":1594.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2459026963,"x_train":6.8460080835,"imp":19.7245104543,"elo":1314.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4940504378,"x_train":15.9731517787,"imp":19.1587737375,"elo":1340.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9205220996,"x_train":48.3100942842,"imp":18.4099433312,"elo":1376.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.1374930629,"x_train":104.5645791303,"imp":17.8142159975,"elo":1409.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.5093622829,"x_train":264.7537616657,"imp":17.3117788731,"elo":1432.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.8290582337,"x_train":533.594254715,"imp":17.023615961,"elo":1446.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":2.1248042643,"x_train":1058.8890383576,"imp":16.8216716507,"elo":1457.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.297931986,"x_train":1605.9821816147,"imp":16.7322425642,"elo":1462.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.3421030774,"x_train":2168.8617901524,"imp":16.6770245123,"elo":1466.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"x_train":5.1697320864,"imp":13.1982203172,"elo":1610.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3910008352,"x_train":3.6410702833,"imp":11.8381062408,"elo":1655.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3152166325,"x_train":3.3689389916,"imp":21.6715869141,"elo":1289.9,"imputed":true,"imputed_pct":31.5789473684,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":2.8456403193,"x_train":22.8408812219,"imp":21.1335243364,"elo":1315.0,"imputed":true,"imputed_pct":31.5789473684,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.9849078236,"x_train":63.3712068389,"imp":20.5597852103,"elo":1342.0,"imputed":true,"imputed_pct":31.5789473684,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":7.8279061651,"x_train":143.4062804674,"imp":20.1535706193,"elo":1354.0,"imputed":true,"imputed_pct":31.5789473684,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.6982581862,"x_train":359.6193496539,"imp":19.8375857059,"elo":1362.8,"imputed":true,"imputed_pct":31.5789473684,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":13.5571446724,"x_train":730.2395031165,"imp":19.6592172618,"elo":1368.3,"imputed":true,"imputed_pct":31.5789473684,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":16.8502824402,"x_train":1484.9037988086,"imp":19.5670748892,"elo":1369.9,"imputed":true,"imputed_pct":31.5789473684,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.41186521,"x_train":2231.7160487129,"imp":19.5250745447,"elo":1370.3,"imputed":true,"imputed_pct":31.5789473684,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":20.4122539884,"x_train":3009.7234478495,"imp":19.5017676353,"elo":1369.1,"imputed":true,"imputed_pct":31.5789473684,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.8830380236,"x_train":439.4466040087,"imp":31.8812607842,"elo":1038.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.4248868953,"x_train":670.2438710225,"imp":31.7426000562,"elo":1045.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8753585219,"x_train":1014.6754881516,"imp":31.4554004745,"elo":1051.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.2526853475,"x_train":1638.3361981714,"imp":30.7301729607,"elo":1067.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.0721037925,"x_train":3535.8153662444,"imp":29.3747032409,"elo":1090.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":13.3360172525,"x_train":7084.5550409751,"imp":28.0029570837,"elo":1111.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":15.3290278651,"x_train":14613.5276339936,"imp":27.2507251208,"elo":1126.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.4140156092,"x_train":21800.1101862642,"imp":26.7912606708,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3145858894,"x_train":29479.468606441,"imp":26.4278269859,"elo":1147.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0665397304,"x_train":1.2028502699,"imp":19.2361448754,"elo":1357.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1470302095,"x_train":6.834853082,"imp":25.5684336816,"elo":1098.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2938262905,"x_train":16.5429640262,"imp":23.4765199655,"elo":1157.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5650998098,"x_train":54.7346146645,"imp":21.7767985052,"elo":1208.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8070919893,"x_train":118.7324150095,"imp":20.983642463,"elo":1243.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2465421293,"x_train":292.558905056,"imp":20.3163156133,"elo":1271.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6220136736,"x_train":595.2571207683,"imp":19.914008022,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8705493288,"x_train":1187.6701815634,"imp":19.6779081554,"elo":1303.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0769295602,"x_train":1789.3376646728,"imp":19.5396221068,"elo":1311.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1462178945,"x_train":2389.4741835732,"imp":19.4066856353,"elo":1317.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.117054282,"x_train":1.7712088618,"imp":21.2238353336,"elo":1229.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5009884397,"x_train":5.6564696531,"imp":19.5080345279,"elo":1310.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7880470326,"x_train":16.9650142951,"imp":19.0308823467,"elo":1339.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8826892134,"x_train":32.997686743,"imp":18.7281864288,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0482676981,"x_train":86.160387483,"imp":18.493258509,"elo":1367.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1873182694,"x_train":172.4620957623,"imp":18.3988160493,"elo":1372.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2414220916,"x_train":343.1961190306,"imp":18.3211345065,"elo":1376.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4140647498,"x_train":511.4517561336,"imp":18.2638274509,"elo":1380.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4519965713,"x_train":685.8651054054,"imp":18.2117061522,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":64.5055925989,"x_train":262.9267967989,"imp":24.5324865248,"elo":1155.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":130.1486843843,"x_train":698.1262023376,"imp":22.330374633,"elo":1207.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":260.8212573113,"x_train":2185.405093286,"imp":19.9492900804,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":378.933825373,"x_train":4868.6453358671,"imp":18.7710008474,"elo":1352.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":496.0296677873,"x_train":12038.5904051722,"imp":17.5022100654,"elo":1409.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":502.4479719176,"x_train":12597.4213063423,"imp":17.4751652248,"elo":1411.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9382527357,"x_train":3.607105251,"imp":30.9061240061,"elo":1011.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.9524919805,"x_train":9.0537118684,"imp":27.0168064854,"elo":1082.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1739792718,"x_train":23.1440570954,"imp":22.305394137,"elo":1213.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.4724798591,"x_train":46.3767226799,"imp":20.4114497336,"elo":1296.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8253774488,"x_train":116.4989862715,"imp":19.6799802589,"elo":1330.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2506435468,"x_train":232.7740378227,"imp":19.4556420191,"elo":1338.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.5518696321,"x_train":465.4926636368,"imp":19.4015327486,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.795048319,"x_train":697.1621157189,"imp":19.4044640333,"elo":1335.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.0515616235,"x_train":934.2696678038,"imp":19.4156500684,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..de665de066a7c966135e1e4af4ba913462ae0860
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_classification/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned)", "TabDPT-Turbo (default)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "LimiX (default)", "TabM (tuned + ensembled)", "Mitra (default)", "TabICL (default)", "TabDPT (tuned + ensembled)", "TabPFNv2 (tuned + ensembled)", "RealMLP (tuned)", "LightGBM (tuned + ensembled)", "TabM (tuned)", "BetaTabPFN (default)", "TabDPT (tuned)", "TabSwift (default)", "CatBoost (tuned + ensembled)", "LightGBM (tuned)", "iLTM (tuned + ensembled)", "TabPFNv2 (tuned)", "TabPFNv2 (default)", "CatBoost (tuned)", "ModernNCA (tuned)", "CatBoost (default)", "xRFM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabDPT (default)", "SAP-RPT-OSS (default)", "XGBoost (tuned)", "xRFM (tuned)", "EBM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "TabM (default)", "FastaiMLP (tuned + ensembled)", "ModernNCA (default)", "ChimeraBoost (tuned + ensembled)", "EBM (tuned)", "iLTM (tuned)", "EBM (default)", "ChimeraBoost (tuned)", "RealMLP (default)", "TorchMLP (tuned)", "FastaiMLP (tuned)", "ExtraTrees (tuned + ensembled)", "ChimeraBoost (default)", "ExtraTrees (tuned)", "XGBoost (default)", "TabSTAR (tuned)", "OrionMSP (default)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned + ensembled)", "LightGBM (default)", "RandomForest (tuned)", "iLTM (default)", "TorchMLP (default)", "Linear (tuned + ensembled)", "TabFlex (default)", "TabSTAR (default)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "FastaiMLP (default)", "xRFM (default)", "PerpetualBooster (tuned)", "KNN (tuned + ensembled)", "ExtraTrees (default)", "RandomForest (default)", "Linear (default)", "PerpetualBooster (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7979797720909119, 0.7592592835426331, 0.8333333134651184, 0.8181818127632141, 0.7853535413742065, 0.7996633052825928, 0.7912458181381226, 0.8518518805503845, 0.8518518805503845, 0.8636363744735718, 0.8762626051902771, 0.8838383555412292, 0.8737373948097229, 0.936026930809021, 0.8989899158477783, 0.8821548819541931, 0.8484848737716675, 0.8939393758773804, 0.9107744097709656, 0.8796296119689941, 0.8922559022903442, 0.8821548819541931, 0.8973063826560974, 0.9090909361839294, 0.9090909361839294, 0.8855218887329102, 0.8535353541374207, 0.9090909361839294, 0.9090909361839294, 0.9057239294052124, 0.9090909361839294, 0.9242424368858337, 0.9040403962135315, 0.8585858345031738, 0.9292929172515869, 0.8956229090690613, 0.9090909361839294, 0.936026930809021, 0.9478114247322083, 0.9158248901367188, 0.8989899158477783, 0.9191918969154358, 0.9208754301071167, 0.9208754301071167, 0.9292929172515869, 0.9158248901367188, 0.9292929172515869, 0.9309764504432678, 0.9595959782600403, 0.9309764504432678, 0.9461279511451721, 0.939393937587738, 0.9377104640007019, 0.9612794518470764, 0.8518518805503845, 0.941077470779419, 0.8518518805503845, 0.9595959782600403, 0.9612794518470764, 0.9579124450683594, 0.941077470779419, 0.9663299918174744, 0.9427609443664551, 0.9377104640007019, 0.8905723690986633, 0.9494949579238892, 0.9848484992980957, 0.9680134654045105, 0.9629629850387573, 0.9814814925193787, 0.9629629850387573, 0.9882155060768127, 0.9814814925193787, 0.9663299918174744, 0.9831649661064148, 0.9629629850387573, 0.9949495196342468], [0.20202019810676575, NaN, 0.5555555820465088, 0.5336700081825256, 0.6329966187477112, 0.6178451180458069, 0.6430976390838623, 0.6397306323051453, 0.7356902360916138, 0.7272727489471436, 0.7171717286109924, 0.7306397557258606, 0.7626262903213501, 0.7508417367935181, 0.8442760705947876, 0.7996633052825928, 0.747474730014801, 0.7626262903213501, 0.8030303120613098, 0.804713785648346, 0.7685185074806213, 0.8299663066864014, 0.8619528412818909, 0.8265992999076843, 0.8434343338012695, 0.8518518805503845, 0.8072390556335449, 0.7651515007019043, 0.8333333134651184, 0.8535353541374207, 0.8468013405799866, 0.8434343338012695, 0.8341751098632812, 0.8417508602142334, 0.7676767706871033, 0.8501683473587036, 0.8400673270225525, 0.8434343338012695, 0.8872053623199463, 0.8872053623199463, 0.8451178669929504, 0.8409090638160706, 0.881313145160675, 0.8888888955116272, 0.8619528412818909, 0.9082491397857666, 0.8796296119689941, 0.8872053623199463, 0.9090909361839294, 0.9242424368858337, 0.8914141654968262, 0.9309764504432678, 0.8872053623199463, 0.9292929172515869, 0.9511784315109253, 0.8409090638160706, 0.9259259104728699, 0.8451178669929504, 0.9242424368858337, 0.9478114247322083, 0.9461279511451721, 0.9461279511451721, 0.9528619647026062, 0.9074074029922485, 0.9292929172515869, 0.875420868396759, 0.9158248901367188, 0.9461279511451721, 0.9545454382896423, 0.9612794518470764, 0.9579124450683594, 0.941077470779419, 0.9781144857406616, 0.9814814925193787, 0.9494949579238892, 0.9696969985961914, 0.9595959782600403, 0.996632993221283], [0.24074074625968933, 0.4444444477558136, NaN, 0.4932659864425659, 0.6414141654968262, 0.5892255902290344, 0.5993266105651855, 0.6094276309013367, 0.6447811722755432, 0.7954545617103577, 0.7845118045806885, 0.7457912564277649, 0.7946127653121948, 0.7011784315109253, 0.813973069190979, 0.7895622849464417, 0.7053872346878052, 0.8215488195419312, 0.8518518805503845, 0.8072390556335449, 0.7777777910232544, 0.8131313323974609, 0.8249158263206482, 0.8636363744735718, 0.8552188277244568, 0.8855218887329102, 0.7693602442741394, 0.7659932374954224, 0.8594276309013367, 0.875420868396759, 0.8712121248245239, 0.8787878751754761, 0.875420868396759, 0.7946127653121948, 0.7626262903213501, 0.8787878751754761, 0.8543770909309387, 0.8619528412818909, 0.9175084233283997, 0.9023569226264954, 0.8619528412818909, 0.8619528412818909, 0.8703703880310059, 0.9141414165496826, 0.8737373948097229, 0.9057239294052124, 0.8787878751754761, 0.9175084233283997, 0.8872053623199463, 0.9292929172515869, 0.8855218887329102, 0.9191918969154358, 0.9242424368858337, 0.9074074029922485, 0.9494949579238892, 0.81144779920578, 0.9284511804580688, 0.8131313323974609, 0.9343434572219849, 0.9545454382896423, 0.936026930809021, 0.9427609443664551, 0.9579124450683594, 0.9006733894348145, 0.9124578833580017, 0.8518518805503845, 0.9225589036941528, 0.9511784315109253, 0.9528619647026062, 0.9545454382896423, 0.9570707082748413, 0.941077470779419, 0.9612794518470764, 0.9629629850387573, 0.945286214351654, 0.9646464586257935, 0.9545454382896423, 0.9949495196342468], [0.1666666716337204, 0.466329962015152, 0.5067340135574341, NaN, 0.5774410963058472, 0.560606062412262, 0.5850168466567993, 0.5757575631141663, 0.7121211886405945, 0.7020202279090881, 0.7070707082748413, 0.7314814925193787, 0.7390572428703308, 0.7129629850387573, 0.8442760705947876, 0.7971380352973938, 0.7255892157554626, 0.7289562225341797, 0.7845118045806885, 0.7676767706871033, 0.7676767706871033, 0.7962962985038757, 0.8383838534355164, 0.7861952781677246, 0.8181818127632141, 0.8316498398780823, 0.7584174871444702, 0.7508417367935181, 0.7929292917251587, 0.8299663066864014, 0.8215488195419312, 0.81144779920578, 0.8030303120613098, 0.819865345954895, 0.7323232293128967, 0.8131313323974609, 0.8005050420761108, 0.8215488195419312, 0.8400673270225525, 0.8670033812522888, 0.8316498398780823, 0.8367003202438354, 0.877946138381958, 0.8518518805503845, 0.8484848737716675, 0.8653198480606079, 0.8703703880310059, 0.8636363744735718, 0.8855218887329102, 0.9090909361839294, 0.8872053623199463, 0.8897306323051453, 0.8636363744735718, 0.8939393758773804, 0.9175084233283997, 0.8299663066864014, 0.9225589036941528, 0.8282828330993652, 0.9208754301071167, 0.9377104640007019, 0.9208754301071167, 0.9175084233283997, 0.941077470779419, 0.9191918969154358, 0.9107744097709656, 0.8771044015884399, 0.9276094436645508, 0.9511784315109253, 0.9461279511451721, 0.9377104640007019, 0.9562289714813232, 0.9444444179534912, 0.9579124450683594, 0.9663299918174744, 0.9537037014961243, 0.9764309525489807, 0.9545454382896423, 0.996632993221283], [0.1818181872367859, 0.3670033812522888, 0.3585858643054962, 0.4225589334964752, NaN, 0.48653197288513184, 0.5471380352973938, 0.5126262903213501, 0.627946138381958, 0.7037037014961243, 0.6750841736793518, 0.680134654045105, 0.7239057421684265, 0.695286214351654, 0.7390572428703308, 0.7020202279090881, 0.6767676472663879, 0.7491582632064819, 0.7828282713890076, 0.7744107842445374, 0.7542087435722351, 0.7272727489471436, 0.7777777910232544, 0.7895622849464417, 0.8333333134651184, 0.8299663066864014, 0.7693602442741394, 0.7281144857406616, 0.7828282713890076, 0.8282828330993652, 0.8215488195419312, 0.8030303120613098, 0.816498339176178, 0.7373737096786499, 0.7710437774658203, 0.8451178669929504, 0.8030303120613098, 0.819865345954895, 0.8771044015884399, 0.872053861618042, 0.8341751098632812, 0.8316498398780823, 0.8653198480606079, 0.875420868396759, 0.8577440977096558, 0.8821548819541931, 0.8518518805503845, 0.8821548819541931, 0.8855218887329102, 0.9158248901367188, 0.8787878751754761, 0.8973063826560974, 0.8821548819541931, 0.8897306323051453, 0.9478114247322083, 0.8097643256187439, 0.8821548819541931, 0.8181818127632141, 0.9090909361839294, 0.9444444179534912, 0.932659924030304, 0.9259259104728699, 0.9494949579238892, 0.8973063826560974, 0.9107744097709656, 0.8434343338012695, 0.9175084233283997, 0.941077470779419, 0.9343434572219849, 0.9377104640007019, 0.9377104640007019, 0.9343434572219849, 0.9595959782600403, 0.9713804721832275, 0.941077470779419, 0.9629629850387573, 0.9511784315109253, 0.9932659864425659], [0.21464645862579346, 0.3821548819541931, 0.4107744097709656, 0.43939393758773804, 0.5134680271148682, NaN, 0.5824915766716003, 0.6094276309013367, 0.5774410963058472, 0.691919207572937, 0.7121211886405945, 0.7020202279090881, 0.7070707082748413, 0.7121211886405945, 0.7028619647026062, 0.6851851940155029, 0.7575757503509521, 0.7356902360916138, 0.7710437774658203, 0.7491582632064819, 0.7744107842445374, 0.7272727489471436, 0.7727272510528564, 0.7946127653121948, 0.8021885752677917, 0.7752525210380554, 0.8097643256187439, 0.7912458181381226, 0.7946127653121948, 0.8232323527336121, 0.804713785648346, 0.8063973188400269, 0.804713785648346, 0.7037037014961243, 0.7685185074806213, 0.8223905563354492, 0.804713785648346, 0.819865345954895, 0.816498339176178, 0.8535353541374207, 0.8131313323974609, 0.8190235495567322, 0.8316498398780823, 0.8308081030845642, 0.8442760705947876, 0.8552188277244568, 0.8451178669929504, 0.8417508602142334, 0.8569023609161377, 0.8973063826560974, 0.875420868396759, 0.875420868396759, 0.8552188277244568, 0.8771044015884399, 0.9141414165496826, 0.81144779920578, 0.9057239294052124, 0.816498339176178, 0.9040403962135315, 0.9343434572219849, 0.9208754301071167, 0.9124578833580017, 0.941077470779419, 0.8905723690986633, 0.9074074029922485, 0.8434343338012695, 0.9057239294052124, 0.9461279511451721, 0.9343434572219849, 0.9377104640007019, 0.9478114247322083, 0.9158248901367188, 0.9562289714813232, 0.9562289714813232, 0.930134654045105, 0.9646464586257935, 0.9377104640007019, 0.9932659864425659], [0.20033669471740723, 0.3569023609161377, 0.40067338943481445, 0.4149831533432007, 0.4528619647026062, 0.41750842332839966, NaN, 0.4654882252216339, 0.5521885752677917, 0.6616161465644836, 0.6750841736793518, 0.6851851940155029, 0.7104377150535583, 0.68855220079422, 0.691919207572937, 0.6818181872367859, 0.6936026811599731, 0.7255892157554626, 0.7424242496490479, 0.755892276763916, 0.7634680271148682, 0.7112794518470764, 0.8013467788696289, 0.7592592835426331, 0.7811447978019714, 0.7356902360916138, 0.7744107842445374, 0.7912458181381226, 0.7659932374954224, 0.7626262903213501, 0.7929292917251587, 0.8030303120613098, 0.7777777910232544, 0.7340067625045776, 0.7828282713890076, 0.808080792427063, 0.81144779920578, 0.8097643256187439, 0.7659932374954224, 0.8468013405799866, 0.8131313323974609, 0.8367003202438354, 0.8333333134651184, 0.8215488195419312, 0.8417508602142334, 0.8611111044883728, 0.8400673270225525, 0.8316498398780823, 0.8703703880310059, 0.8989899158477783, 0.8771044015884399, 0.872053861618042, 0.8602693676948547, 0.8855218887329102, 0.8939393758773804, 0.8333333134651184, 0.932659924030304, 0.8341751098632812, 0.8653198480606079, 0.9343434572219849, 0.9141414165496826, 0.9175084233283997, 0.9612794518470764, 0.8922559022903442, 0.9267676472663879, 0.8619528412818909, 0.8989899158477783, 0.9461279511451721, 0.9309764504432678, 0.941077470779419, 0.9528619647026062, 0.9023569226264954, 0.9579124450683594, 0.9579124450683594, 0.9478114247322083, 0.9713804721832275, 0.9461279511451721, 0.9949495196342468], [0.20875421166419983, 0.36026936769485474, 0.3905723989009857, 0.42424243688583374, 0.4873737394809723, 0.3905723989009857, 0.5345118045806885, NaN, 0.5723905563354492, 0.6649831533432007, 0.6868686676025391, 0.6700336933135986, 0.7020202279090881, 0.7037037014961243, 0.6784511804580688, 0.6818181872367859, 0.7196969985961914, 0.7272727489471436, 0.7542087435722351, 0.75, 0.755892276763916, 0.7264309525489807, 0.7617844939231873, 0.7811447978019714, 0.8013467788696289, 0.752525269985199, 0.7904040217399597, 0.7760942578315735, 0.7777777910232544, 0.7794612646102905, 0.7912458181381226, 0.7904040217399597, 0.7878788113594055, 0.7104377150535583, 0.7609427571296692, 0.804713785648346, 0.7912458181381226, 0.819865345954895, 0.8072390556335449, 0.8451178669929504, 0.8063973188400269, 0.8030303120613098, 0.816498339176178, 0.813973069190979, 0.8333333134651184, 0.8434343338012695, 0.8350168466567993, 0.8350168466567993, 0.8484848737716675, 0.8956229090690613, 0.8670033812522888, 0.869528591632843, 0.8468013405799866, 0.8787878751754761, 0.9074074029922485, 0.8030303120613098, 0.9006733894348145, 0.81144779920578, 0.8872053623199463, 0.9292929172515869, 0.9124578833580017, 0.9040403962135315, 0.941077470779419, 0.875420868396759, 0.8973063826560974, 0.8434343338012695, 0.8956229090690613, 0.9318181872367859, 0.9259259104728699, 0.936026930809021, 0.9377104640007019, 0.9124578833580017, 0.9494949579238892, 0.9528619647026062, 0.933501660823822, 0.9646464586257935, 0.9427609443664551, 0.9932659864425659], [0.14814814925193787, 0.26430976390838623, 0.3552188575267792, 0.28787878155708313, 0.372053861618042, 0.4225589334964752, 0.44781145453453064, 0.4276094138622284, NaN, 0.5454545617103577, 0.5286195278167725, 0.5925925970077515, 0.5361952781677246, 0.5505050420761108, 0.5824915766716003, 0.624579131603241, 0.5656565427780151, 0.618686854839325, 0.6094276309013367, 0.5892255902290344, 0.5740740895271301, 0.680134654045105, 0.7264309525489807, 0.6531986594200134, 0.6212121248245239, 0.6936026811599731, 0.6111111044883728, 0.6329966187477112, 0.6464646458625793, 0.6565656661987305, 0.6750841736793518, 0.6767676472663879, 0.6632996797561646, 0.8072390556335449, 0.6414141654968262, 0.6767676472663879, 0.7053872346878052, 0.6818181872367859, 0.680134654045105, 0.7255892157554626, 0.6683501601219177, 0.7508417367935181, 0.755892276763916, 0.7390572428703308, 0.7196969985961914, 0.7735690474510193, 0.7491582632064819, 0.7659932374954224, 0.7777777910232544, 0.7912458181381226, 0.8148148059844971, 0.8257575631141663, 0.7760942578315735, 0.8535353541374207, 0.8265992999076843, 0.7542087435722351, 0.8670033812522888, 0.7609427571296692, 0.8451178669929504, 0.8602693676948547, 0.8602693676948547, 0.8367003202438354, 0.872053861618042, 0.8249158263206482, 0.9057239294052124, 0.8299663066864014, 0.8367003202438354, 0.8973063826560974, 0.9158248901367188, 0.8922559022903442, 0.9141414165496826, 0.9090909361839294, 0.9486532211303711, 0.9528619647026062, 0.9225589036941528, 0.9612794518470764, 0.9511784315109253, 0.9949495196342468], [0.14814814925193787, 0.27272728085517883, 0.20454545319080353, 0.29797980189323425, 0.29629629850387573, 0.3080808222293854, 0.33838382363319397, 0.3350168466567993, 0.4545454680919647, NaN, 0.4722222089767456, 0.469696968793869, 0.5538720488548279, 0.48821547627449036, 0.5134680271148682, 0.505892276763916, 0.4789562225341797, 0.6776094436645508, 0.6734007000923157, 0.6111111044883728, 0.6329966187477112, 0.5429292917251587, 0.5488215684890747, 0.695286214351654, 0.7230639457702637, 0.6969696879386902, 0.563973069190979, 0.5454545617103577, 0.7053872346878052, 0.6666666865348816, 0.7121211886405945, 0.7255892157554626, 0.7693602442741394, 0.5892255902290344, 0.6262626051902771, 0.7929292917251587, 0.7138047218322754, 0.7693602442741394, 0.6851851940155029, 0.7356902360916138, 0.7777777910232544, 0.7904040217399597, 0.7323232293128967, 0.7542087435722351, 0.7861952781677246, 0.7828282713890076, 0.7887205481529236, 0.7786195278167725, 0.869528591632843, 0.7929292917251587, 0.810606062412262, 0.8131313323974609, 0.7794612646102905, 0.8223905563354492, 0.8888888955116272, 0.7676767706871033, 0.7828282713890076, 0.7744107842445374, 0.8855218887329102, 0.9057239294052124, 0.9107744097709656, 0.8787878751754761, 0.9141414165496826, 0.8703703880310059, 0.8265992999076843, 0.8131313323974609, 0.872053861618042, 0.9309764504432678, 0.9343434572219849, 0.9276094436645508, 0.9377104640007019, 0.9090909361839294, 0.9545454382896423, 0.9545454382896423, 0.8905723690986633, 0.9663299918174744, 0.932659924030304, 0.9949495196342468], [0.13636364042758942, 0.28282827138900757, 0.21548821032047272, 0.2929292917251587, 0.3249158263206482, 0.28787878155708313, 0.3249158263206482, 0.31313130259513855, 0.47138047218322754, 0.5277777910232544, NaN, 0.4444444477558136, 0.5454545617103577, 0.4427609443664551, 0.5235690474510193, 0.5252525210380554, 0.4124579131603241, 0.6144781112670898, 0.6969696879386902, 0.6043770909309387, 0.6060606241226196, 0.5521885752677917, 0.5715488195419312, 0.691919207572937, 0.7222222089767456, 0.7053872346878052, 0.5050504803657532, 0.5404040217399597, 0.7171717286109924, 0.6548821330070496, 0.7407407164573669, 0.691077470779419, 0.747474730014801, 0.6094276309013367, 0.5959596037864685, 0.7390572428703308, 0.7171717286109924, 0.744107723236084, 0.695286214351654, 0.7760942578315735, 0.7289562225341797, 0.7609427571296692, 0.7356902360916138, 0.7836700081825256, 0.7609427571296692, 0.7845118045806885, 0.7929292917251587, 0.8063973188400269, 0.8030303120613098, 0.8316498398780823, 0.81144779920578, 0.7508417367935181, 0.8299663066864014, 0.7794612646102905, 0.8838383555412292, 0.7659932374954224, 0.753367006778717, 0.7727272510528564, 0.8350168466567993, 0.9090909361839294, 0.8678451180458069, 0.880471408367157, 0.8939393758773804, 0.8661616444587708, 0.8383838534355164, 0.8148148059844971, 0.8897306323051453, 0.9242424368858337, 0.9276094436645508, 0.8956229090690613, 0.9276094436645508, 0.8922559022903442, 0.9343434572219849, 0.9276094436645508, 0.9191918969154358, 0.9478114247322083, 0.9276094436645508, 0.996632993221283], [0.12373737245798111, 0.2693602740764618, 0.2542087435722351, 0.26851850748062134, 0.31986531615257263, 0.29797980189323425, 0.31481480598449707, 0.32996633648872375, 0.40740740299224854, 0.5303030014038086, 0.5555555820465088, NaN, 0.567340075969696, 0.5277777910232544, 0.5454545617103577, 0.5075757503509521, 0.5505050420761108, 0.5749158263206482, 0.6548821330070496, 0.5782828330993652, 0.5824915766716003, 0.5387205481529236, 0.5656565427780151, 0.6683501601219177, 0.6700336933135986, 0.7129629850387573, 0.6127946376800537, 0.6212121248245239, 0.6700336933135986, 0.6683501601219177, 0.6969696879386902, 0.6902356743812561, 0.7037037014961243, 0.5791245698928833, 0.6296296119689941, 0.7222222089767456, 0.689393937587738, 0.7121211886405945, 0.6818181872367859, 0.691919207572937, 0.6700336933135986, 0.6759259104728699, 0.7239057421684265, 0.7575757503509521, 0.7373737096786499, 0.7373737096786499, 0.7390572428703308, 0.7659932374954224, 0.7457912564277649, 0.7626262903213501, 0.7584174871444702, 0.7929292917251587, 0.7794612646102905, 0.7962962985038757, 0.8215488195419312, 0.7281144857406616, 0.7920875549316406, 0.7239057421684265, 0.8316498398780823, 0.8350168466567993, 0.8451178669929504, 0.8232323527336121, 0.8282828330993652, 0.8282828330993652, 0.7878788113594055, 0.7819865345954895, 0.8484848737716675, 0.8636363744735718, 0.8737373948097229, 0.9107744097709656, 0.8787878751754761, 0.8451178669929504, 0.9107744097709656, 0.9242424368858337, 0.8552188277244568, 0.9141414165496826, 0.8787878751754761, 0.9747474789619446], [0.11616161465644836, 0.2373737394809723, 0.2053872048854828, 0.2609427571296692, 0.2760942876338959, 0.2929292917251587, 0.28956228494644165, 0.29797980189323425, 0.4638047218322754, 0.4461279511451721, 0.4545454680919647, 0.43265992403030396, NaN, 0.4292929172515869, 0.49494948983192444, 0.5092592835426331, 0.43265992403030396, 0.567340075969696, 0.6144781112670898, 0.6666666865348816, 0.6397306323051453, 0.5488215684890747, 0.5723905563354492, 0.6127946376800537, 0.6557239294052124, 0.6170033812522888, 0.5134680271148682, 0.5126262903213501, 0.6329966187477112, 0.6388888955116272, 0.6969696879386902, 0.6144781112670898, 0.6902356743812561, 0.5959596037864685, 0.6329966187477112, 0.7272727489471436, 0.6178451180458069, 0.747474730014801, 0.6734007000923157, 0.7020202279090881, 0.8442760705947876, 0.7138047218322754, 0.7171717286109924, 0.7457912564277649, 0.7861952781677246, 0.7323232293128967, 0.7786195278167725, 0.7777777910232544, 0.817340075969696, 0.7954545617103577, 0.7744107842445374, 0.7457912564277649, 0.8148148059844971, 0.747474730014801, 0.8484848737716675, 0.7676767706871033, 0.7592592835426331, 0.7744107842445374, 0.7962962985038757, 0.9015151262283325, 0.8400673270225525, 0.8737373948097229, 0.9057239294052124, 0.8670033812522888, 0.8282828330993652, 0.816498339176178, 0.8829966187477112, 0.9057239294052124, 0.9006733894348145, 0.8670033812522888, 0.9259259104728699, 0.8838383555412292, 0.9208754301071167, 0.9141414165496826, 0.9107744097709656, 0.9629629850387573, 0.9242424368858337, 0.9865319728851318], [0.1262626200914383, 0.24915824830532074, 0.2988215386867523, 0.28703704476356506, 0.30471381545066833, 0.28787878155708313, 0.31144779920578003, 0.29629629850387573, 0.44949495792388916, 0.5117844939231873, 0.5572390556335449, 0.4722222089767456, 0.5707070827484131, NaN, 0.5168350338935852, 0.503367006778717, 0.5210437774658203, 0.5942760705947876, 0.6212121248245239, 0.6329966187477112, 0.6026936173439026, 0.5269360542297363, 0.5925925970077515, 0.6329966187477112, 0.6531986594200134, 0.6658248901367188, 0.5993266105651855, 0.6060606241226196, 0.625420868396759, 0.6447811722755432, 0.7003366947174072, 0.6515151262283325, 0.6675084233283997, 0.6043770909309387, 0.5707070827484131, 0.6632996797561646, 0.6691918969154358, 0.6969696879386902, 0.6818181872367859, 0.7053872346878052, 0.6767676472663879, 0.6986532211303711, 0.6767676472663879, 0.7356902360916138, 0.7171717286109924, 0.7255892157554626, 0.7289562225341797, 0.7407407164573669, 0.7138047218322754, 0.7659932374954224, 0.747474730014801, 0.7070707082748413, 0.7542087435722351, 0.7213804721832275, 0.7929292917251587, 0.7003366947174072, 0.75, 0.7020202279090881, 0.744107723236084, 0.7962962985038757, 0.7718855142593384, 0.7878788113594055, 0.808080792427063, 0.8215488195419312, 0.8181818127632141, 0.747474730014801, 0.8434343338012695, 0.8282828330993652, 0.8367003202438354, 0.8872053623199463, 0.8451178669929504, 0.816498339176178, 0.8670033812522888, 0.8737373948097229, 0.8771044015884399, 0.9309764504432678, 0.8863636255264282, 0.9915825128555298], [0.06397306174039841, 0.15572389960289001, 0.186026930809021, 0.15572389960289001, 0.2609427571296692, 0.2971380352973938, 0.3080808222293854, 0.32154881954193115, 0.41750842332839966, 0.48653197288513184, 0.4764309823513031, 0.4545454680919647, 0.5050504803657532, 0.4831649959087372, NaN, 0.505892276763916, 0.5084174871444702, 0.560606062412262, 0.5690235495567322, 0.560606062412262, 0.5505050420761108, 0.5597643256187439, 0.6262626051902771, 0.5723905563354492, 0.6060606241226196, 0.6271044015884399, 0.558922529220581, 0.5732323527336121, 0.5858585834503174, 0.627946138381958, 0.6380471587181091, 0.6397306323051453, 0.6329966187477112, 0.6422559022903442, 0.6127946376800537, 0.6531986594200134, 0.6481481194496155, 0.6329966187477112, 0.6616161465644836, 0.6582491397857666, 0.6582491397857666, 0.680134654045105, 0.7163299918174744, 0.7205387353897095, 0.6616161465644836, 0.7306397557258606, 0.6936026811599731, 0.7390572428703308, 0.7205387353897095, 0.752525269985199, 0.7323232293128967, 0.7979797720909119, 0.7744107842445374, 0.810606062412262, 0.816498339176178, 0.744107723236084, 0.8249158263206482, 0.747474730014801, 0.8249158263206482, 0.8880471587181091, 0.8552188277244568, 0.8333333134651184, 0.8484848737716675, 0.7912458181381226, 0.8468013405799866, 0.804713785648346, 0.808080792427063, 0.9057239294052124, 0.869528591632843, 0.8787878751754761, 0.9175084233283997, 0.8973063826560974, 0.9478114247322083, 0.9478114247322083, 0.9082491397857666, 0.9579124450683594, 0.939393937587738, 0.9932659864425659], [0.10101009905338287, 0.20033669471740723, 0.21043771505355835, 0.202861949801445, 0.29797980189323425, 0.31481480598449707, 0.3181818127632141, 0.3181818127632141, 0.37542086839675903, 0.49410775303840637, 0.4747474789619446, 0.49242424964904785, 0.49074074625968933, 0.49663299322128296, 0.49410775303840637, NaN, 0.5084174871444702, 0.5715488195419312, 0.563973069190979, 0.5395622849464417, 0.5370370149612427, 0.6498316526412964, 0.6001683473587036, 0.5808081030845642, 0.5959596037864685, 0.627946138381958, 0.5833333134651184, 0.5656565427780151, 0.5875421166419983, 0.6127946376800537, 0.624579131603241, 0.6355218887329102, 0.6052188277244568, 0.6784511804580688, 0.6094276309013367, 0.6380471587181091, 0.6540403962135315, 0.6405723690986633, 0.6666666865348816, 0.6792929172515869, 0.6355218887329102, 0.695286214351654, 0.7146464586257935, 0.6868686676025391, 0.6708754301071167, 0.7289562225341797, 0.6734007000923157, 0.7053872346878052, 0.7154881954193115, 0.7542087435722351, 0.7592592835426331, 0.7676767706871033, 0.7390572428703308, 0.7979797720909119, 0.7979797720909119, 0.7222222089767456, 0.8299663066864014, 0.7239057421684265, 0.8282828330993652, 0.8400673270225525, 0.8383838534355164, 0.8232323527336121, 0.8552188277244568, 0.7811447978019714, 0.8501683473587036, 0.8063973188400269, 0.7996633052825928, 0.8855218887329102, 0.8905723690986633, 0.9015151262283325, 0.8964646458625793, 0.8930976390838623, 0.9461279511451721, 0.939393937587738, 0.868686854839325, 0.941077470779419, 0.9259259104728699, 0.9949495196342468], [0.11784511804580688, 0.2525252401828766, 0.2946127951145172, 0.27441078424453735, 0.3232323229312897, 0.24242424964904785, 0.30639731884002686, 0.280303031206131, 0.4343434274196625, 0.5210437774658203, 0.5875421166419983, 0.44949495792388916, 0.567340075969696, 0.4789562225341797, 0.4915824830532074, 0.4915824830532074, NaN, 0.5900673270225525, 0.6136363744735718, 0.6085858345031738, 0.6119528412818909, 0.5143097639083862, 0.5572390556335449, 0.6329966187477112, 0.6329966187477112, 0.6430976390838623, 0.624579131603241, 0.6077440977096558, 0.6388888955116272, 0.6464646458625793, 0.6683501601219177, 0.6700336933135986, 0.6691918969154358, 0.567340075969696, 0.5496633052825928, 0.6767676472663879, 0.6742424368858337, 0.6843434572219849, 0.680134654045105, 0.680134654045105, 0.6683501601219177, 0.6792929172515869, 0.6557239294052124, 0.7087541818618774, 0.7003366947174072, 0.7053872346878052, 0.7053872346878052, 0.7188552021980286, 0.7104377150535583, 0.7407407164573669, 0.7095959782600403, 0.7028619647026062, 0.7306397557258606, 0.7188552021980286, 0.7811447978019714, 0.683501660823822, 0.7306397557258606, 0.6868686676025391, 0.752525269985199, 0.7962962985038757, 0.7760942578315735, 0.7609427571296692, 0.7962962985038757, 0.8013467788696289, 0.7878788113594055, 0.744107723236084, 0.8232323527336121, 0.8282828330993652, 0.8333333134651184, 0.9006733894348145, 0.8501683473587036, 0.8097643256187439, 0.8888888955116272, 0.8627946376800537, 0.8434343338012695, 0.9242424368858337, 0.8771044015884399, 0.9949495196342468], [0.1515151560306549, 0.2373737394809723, 0.17845118045806885, 0.2710437774658203, 0.25084173679351807, 0.26430976390838623, 0.27441078424453735, 0.27272728085517883, 0.38131314516067505, 0.3223905861377716, 0.38552188873291016, 0.4250841736793518, 0.43265992403030396, 0.40572389960289, 0.43939393758773804, 0.42845118045806885, 0.4099326729774475, NaN, 0.5555555820465088, 0.5168350338935852, 0.5774410963058472, 0.4932659864425659, 0.4932659864425659, 0.563973069190979, 0.5959596037864685, 0.554713785648346, 0.48653197288513184, 0.5067340135574341, 0.5723905563354492, 0.5723905563354492, 0.6212121248245239, 0.6010100841522217, 0.6498316526412964, 0.5387205481529236, 0.5622895359992981, 0.6843434572219849, 0.630471408367157, 0.6750841736793518, 0.6111111044883728, 0.6767676472663879, 0.6851851940155029, 0.6969696879386902, 0.6683501601219177, 0.6430976390838623, 0.7188552021980286, 0.6936026811599731, 0.6902356743812561, 0.6717171669006348, 0.7693602442741394, 0.755892276763916, 0.7693602442741394, 0.7407407164573669, 0.7138047218322754, 0.7592592835426331, 0.8097643256187439, 0.7508417367935181, 0.7247474789619446, 0.755892276763916, 0.8131313323974609, 0.8350168466567993, 0.8299663066864014, 0.8299663066864014, 0.8771044015884399, 0.816498339176178, 0.808922529220581, 0.8013467788696289, 0.8257575631141663, 0.8838383555412292, 0.9082491397857666, 0.8787878751754761, 0.8838383555412292, 0.8670033812522888, 0.8838383555412292, 0.8939393758773804, 0.8636363744735718, 0.9511784315109253, 0.8989899158477783, 0.9848484992980957], [0.10606060922145844, 0.19696970283985138, 0.14814814925193787, 0.21548821032047272, 0.21717171370983124, 0.2289562225341797, 0.25757575035095215, 0.2457912415266037, 0.3905723989009857, 0.3265993297100067, 0.3030303120613098, 0.34511783719062805, 0.38552188873291016, 0.3787878751754761, 0.43097642064094543, 0.436026930809021, 0.3863636255264282, 0.4444444477558136, NaN, 0.46717172861099243, 0.469696968793869, 0.48821547627449036, 0.5202020406723022, 0.563973069190979, 0.6085858345031738, 0.5412458181381226, 0.4511784613132477, 0.4545454680919647, 0.5799663066864014, 0.5269360542297363, 0.6557239294052124, 0.5808081030845642, 0.6818181872367859, 0.5909090638160706, 0.5808081030845642, 0.7138047218322754, 0.6346801519393921, 0.6641414165496826, 0.563973069190979, 0.6127946376800537, 0.6742424368858337, 0.6178451180458069, 0.6548821330070496, 0.7062289714813232, 0.7138047218322754, 0.6734007000923157, 0.7710437774658203, 0.7373737096786499, 0.7087541818618774, 0.7306397557258606, 0.7154881954193115, 0.7331649661064148, 0.808080792427063, 0.7373737096786499, 0.8653198480606079, 0.7516834735870361, 0.7407407164573669, 0.7542087435722351, 0.8131313323974609, 0.9478114247322083, 0.8653198480606079, 0.8888888955116272, 0.8316498398780823, 0.8367003202438354, 0.7845118045806885, 0.8215488195419312, 0.8569023609161377, 0.8922559022903442, 0.8417508602142334, 0.8265992999076843, 0.9124578833580017, 0.8888888955116272, 0.9292929172515869, 0.939393937587738, 0.9124578833580017, 0.9444444179534912, 0.930134654045105, 0.9932659864425659], [0.08922559022903442, 0.19528619945049286, 0.19276094436645508, 0.23232322931289673, 0.22558923065662384, 0.25084173679351807, 0.24410773813724518, 0.25, 0.4107744097709656, 0.3888888955116272, 0.3956229090690613, 0.42171716690063477, 0.3333333432674408, 0.3670033812522888, 0.43939393758773804, 0.46043771505355835, 0.3914141356945038, 0.4831649959087372, 0.5328282713890076, NaN, 0.5538720488548279, 0.49663299322128296, 0.5269360542297363, 0.560606062412262, 0.563973069190979, 0.563973069190979, 0.4545454680919647, 0.4764309823513031, 0.5791245698928833, 0.5993266105651855, 0.631313145160675, 0.5555555820465088, 0.6136363744735718, 0.5648148059844971, 0.5555555820465088, 0.631313145160675, 0.5824915766716003, 0.6498316526412964, 0.6212121248245239, 0.6632996797561646, 0.7222222089767456, 0.6691918969154358, 0.6750841736793518, 0.6936026811599731, 0.7138047218322754, 0.6632996797561646, 0.7289562225341797, 0.7003366947174072, 0.7676767706871033, 0.7483165264129639, 0.7407407164573669, 0.695286214351654, 0.753367006778717, 0.7121211886405945, 0.804713785648346, 0.7508417367935181, 0.7356902360916138, 0.7601010203361511, 0.7744107842445374, 0.8249158263206482, 0.7912458181381226, 0.8484848737716675, 0.8905723690986633, 0.8468013405799866, 0.816498339176178, 0.7996633052825928, 0.8585858345031738, 0.9040403962135315, 0.8636363744735718, 0.8282828330993652, 0.9124578833580017, 0.8535353541374207, 0.8771044015884399, 0.8888888955116272, 0.8888888955116272, 0.9545454382896423, 0.9057239294052124, 0.9797979593276978], [0.12037037312984467, 0.23148147761821747, 0.2222222238779068, 0.23232322931289673, 0.2457912415266037, 0.22558923065662384, 0.23653198778629303, 0.24410773813724518, 0.42592594027519226, 0.3670033812522888, 0.39393940567970276, 0.41750842332839966, 0.36026936769485474, 0.3973063826560974, 0.44949495792388916, 0.46296295523643494, 0.38804712891578674, 0.4225589334964752, 0.5303030014038086, 0.4461279511451721, NaN, 0.5168350338935852, 0.5134680271148682, 0.5218855142593384, 0.5404040217399597, 0.5336700081825256, 0.4427609443664551, 0.469696968793869, 0.5496633052825928, 0.5530303120613098, 0.6127946376800537, 0.560606062412262, 0.5791245698928833, 0.5824915766716003, 0.5538720488548279, 0.5942760705947876, 0.5690235495567322, 0.6565656661987305, 0.5521885752677917, 0.6397306323051453, 0.627946138381958, 0.6439393758773804, 0.6481481194496155, 0.6557239294052124, 0.7003366947174072, 0.6212121248245239, 0.7424242496490479, 0.6683501601219177, 0.7104377150535583, 0.7188552021980286, 0.7255892157554626, 0.7087541818618774, 0.7306397557258606, 0.7255892157554626, 0.747474730014801, 0.7811447978019714, 0.7390572428703308, 0.7861952781677246, 0.7609427571296692, 0.7836700081825256, 0.7744107842445374, 0.8451178669929504, 0.8299663066864014, 0.8602693676948547, 0.8240740895271301, 0.8409090638160706, 0.8787878751754761, 0.8569023609161377, 0.8451178669929504, 0.7929292917251587, 0.875420868396759, 0.8838383555412292, 0.8535353541374207, 0.8619528412818909, 0.9343434572219849, 0.9259259104728699, 0.9562289714813232, 0.9882155060768127], [0.10774410516023636, 0.17003366351127625, 0.18686868250370026, 0.20370370149612427, 0.27272728085517883, 0.27272728085517883, 0.2887205481529236, 0.2735690176486969, 0.31986531615257263, 0.4570707082748413, 0.44781145453453064, 0.4612794518470764, 0.4511784613132477, 0.47306397557258606, 0.4402357041835785, 0.3501683473587036, 0.48569023609161377, 0.5067340135574341, 0.5117844939231873, 0.503367006778717, 0.4831649959087372, NaN, 0.5404040217399597, 0.5336700081825256, 0.5404040217399597, 0.5555555820465088, 0.5412458181381226, 0.5260942578315735, 0.5412458181381226, 0.5538720488548279, 0.5774410963058472, 0.5858585834503174, 0.5732323527336121, 0.6346801519393921, 0.5505050420761108, 0.5824915766716003, 0.5875421166419983, 0.5942760705947876, 0.6069023609161377, 0.624579131603241, 0.5942760705947876, 0.6397306323051453, 0.6414141654968262, 0.6481481194496155, 0.6346801519393921, 0.6515151262283325, 0.6430976390838623, 0.6582491397857666, 0.6616161465644836, 0.6936026811599731, 0.7188552021980286, 0.7121211886405945, 0.6868686676025391, 0.7390572428703308, 0.7239057421684265, 0.6843434572219849, 0.8265992999076843, 0.6851851940155029, 0.752525269985199, 0.7777777910232544, 0.7895622849464417, 0.7727272510528564, 0.8249158263206482, 0.7617844939231873, 0.8367003202438354, 0.7878788113594055, 0.7828282713890076, 0.8636363744735718, 0.8703703880310059, 0.8670033812522888, 0.8737373948097229, 0.8838383555412292, 0.9090909361839294, 0.9074074029922485, 0.8518518805503845, 0.939393937587738, 0.9141414165496826, 0.9915825128555298], [0.11784511804580688, 0.13804714381694794, 0.1750841736793518, 0.16161616146564484, 0.2222222238779068, 0.22727273404598236, 0.1986531913280487, 0.23821549117565155, 0.2735690176486969, 0.4511784613132477, 0.42845118045806885, 0.4343434274196625, 0.4276094138622284, 0.40740740299224854, 0.3737373650074005, 0.3998316526412964, 0.4427609443664551, 0.5067340135574341, 0.47979798913002014, 0.47306397557258606, 0.48653197288513184, 0.4595959484577179, NaN, 0.4915824830532074, 0.5016834735870361, 0.5353535413742065, 0.5050504803657532, 0.5101010203361511, 0.5143097639083862, 0.5370370149612427, 0.5420875549316406, 0.5622895359992981, 0.5538720488548279, 0.567340075969696, 0.5841751098632812, 0.5841751098632812, 0.5622895359992981, 0.5858585834503174, 0.5723905563354492, 0.619528591632843, 0.6111111044883728, 0.6936026811599731, 0.627946138381958, 0.6127946376800537, 0.6178451180458069, 0.6616161465644836, 0.6262626051902771, 0.6430976390838623, 0.6632996797561646, 0.691919207572937, 0.7272727489471436, 0.6734007000923157, 0.7188552021980286, 0.695286214351654, 0.695286214351654, 0.752525269985199, 0.8670033812522888, 0.7575757503509521, 0.6936026811599731, 0.7819865345954895, 0.747474730014801, 0.7861952781677246, 0.8181818127632141, 0.7609427571296692, 0.9057239294052124, 0.8265992999076843, 0.7878788113594055, 0.8838383555412292, 0.8552188277244568, 0.819865345954895, 0.8872053623199463, 0.8518518805503845, 0.8569023609161377, 0.8872053623199463, 0.8863636255264282, 0.9309764504432678, 0.9427609443664551, 0.9949495196342468], [0.1026936024427414, 0.1734006702899933, 0.13636364042758942, 0.2138047069311142, 0.21043771505355835, 0.2053872048854828, 0.24074074625968933, 0.21885521709918976, 0.3468013405799866, 0.30471381545066833, 0.3080808222293854, 0.3316498398780823, 0.3872053921222687, 0.3670033812522888, 0.4276094138622284, 0.4191919267177582, 0.3670033812522888, 0.436026930809021, 0.436026930809021, 0.43939393758773804, 0.4781144857406616, 0.466329962015152, 0.5084174871444702, NaN, 0.48400673270225525, 0.5269360542297363, 0.43097642064094543, 0.4444444477558136, 0.5782828330993652, 0.5210437774658203, 0.5925925970077515, 0.560606062412262, 0.5749158263206482, 0.5370370149612427, 0.5723905563354492, 0.6102693676948547, 0.6077440977096558, 0.5993266105651855, 0.5656565427780151, 0.6060606241226196, 0.6212121248245239, 0.6010100841522217, 0.6178451180458069, 0.6531986594200134, 0.6439393758773804, 0.6683501601219177, 0.68855220079422, 0.680134654045105, 0.68855220079422, 0.7205387353897095, 0.683501660823822, 0.7390572428703308, 0.7222222089767456, 0.7567340135574341, 0.8484848737716675, 0.744107723236084, 0.695286214351654, 0.7390572428703308, 0.8148148059844971, 0.8703703880310059, 0.8611111044883728, 0.8425925970077515, 0.8299663066864014, 0.8063973188400269, 0.7356902360916138, 0.7895622849464417, 0.8232323527336121, 0.8417508602142334, 0.8333333134651184, 0.8484848737716675, 0.8602693676948547, 0.8821548819541931, 0.9057239294052124, 0.9259259104728699, 0.8569023609161377, 0.9074074029922485, 0.9107744097709656, 0.9932659864425659], [0.09090909361839294, 0.15656565129756927, 0.14478114247322083, 0.1818181872367859, 0.1666666716337204, 0.19781145453453064, 0.21885521709918976, 0.1986531913280487, 0.3787878751754761, 0.27693602442741394, 0.2777777910232544, 0.32996633648872375, 0.34427610039711, 0.3468013405799866, 0.39393940567970276, 0.4040403962135315, 0.3670033812522888, 0.4040403962135315, 0.3914141356945038, 0.436026930809021, 0.4595959484577179, 0.4595959484577179, 0.4983164966106415, 0.5159932374954224, NaN, 0.48653197288513184, 0.436026930809021, 0.44781145453453064, 0.5252525210380554, 0.47979798913002014, 0.6119528412818909, 0.5496633052825928, 0.5867003202438354, 0.5538720488548279, 0.5420875549316406, 0.624579131603241, 0.5841751098632812, 0.6329966187477112, 0.5336700081825256, 0.5824915766716003, 0.6380471587181091, 0.5917508602142334, 0.6296296119689941, 0.6548821330070496, 0.6784511804580688, 0.624579131603241, 0.7239057421684265, 0.6767676472663879, 0.689393937587738, 0.686026930809021, 0.6616161465644836, 0.691077470779419, 0.7626262903213501, 0.7230639457702637, 0.8291245698928833, 0.7407407164573669, 0.7306397557258606, 0.7457912564277649, 0.7575757503509521, 0.9166666865348816, 0.8249158263206482, 0.8678451180458069, 0.8131313323974609, 0.8131313323974609, 0.7727272510528564, 0.808080792427063, 0.8392255902290344, 0.874579131603241, 0.8265992999076843, 0.813973069190979, 0.8939393758773804, 0.8636363744735718, 0.9175084233283997, 0.9158248901367188, 0.8956229090690613, 0.9427609443664551, 0.9208754301071167, 0.9949495196342468], [0.09090909361839294, 0.14814814925193787, 0.11447811126708984, 0.16835017502307892, 0.17003366351127625, 0.22474747896194458, 0.26430976390838623, 0.24747474491596222, 0.30639731884002686, 0.3030303120613098, 0.2946127951145172, 0.28703704476356506, 0.3829966187477112, 0.33417508006095886, 0.37289562821388245, 0.372053861618042, 0.3569023609161377, 0.44528618454933167, 0.45875421166419983, 0.436026930809021, 0.466329962015152, 0.4444444477558136, 0.46464645862579346, 0.47306397557258606, 0.5134680271148682, NaN, 0.4680134654045105, 0.48148149251937866, 0.503367006778717, 0.5420875549316406, 0.5404040217399597, 0.5782828330993652, 0.5429292917251587, 0.5303030014038086, 0.5471380352973938, 0.5976430773735046, 0.625420868396759, 0.6010100841522217, 0.5707070827484131, 0.5976430773735046, 0.5656565427780151, 0.6430976390838623, 0.6565656661987305, 0.6565656661987305, 0.6414141654968262, 0.7062289714813232, 0.6717171669006348, 0.6666666865348816, 0.7070707082748413, 0.7037037014961243, 0.7146464586257935, 0.7062289714813232, 0.7146464586257935, 0.7390572428703308, 0.8030303120613098, 0.6969696879386902, 0.691077470779419, 0.6978114247322083, 0.7861952781677246, 0.8535353541374207, 0.8232323527336121, 0.8434343338012695, 0.8299663066864014, 0.747474730014801, 0.804713785648346, 0.7861952781677246, 0.7828282713890076, 0.8518518805503845, 0.8619528412818909, 0.8787878751754761, 0.872053861618042, 0.8468013405799866, 0.9090909361839294, 0.9090909361839294, 0.8367003202438354, 0.9124578833580017, 0.8863636255264282, 0.9814814925193787], [0.11447811126708984, 0.19276094436645508, 0.2306397259235382, 0.2415824979543686, 0.2306397259235382, 0.1902356892824173, 0.22558923065662384, 0.2095959633588791, 0.3888888955116272, 0.436026930809021, 0.49494948983192444, 0.3872053921222687, 0.48653197288513184, 0.40067338943481445, 0.44107744097709656, 0.4166666567325592, 0.37542086839675903, 0.5134680271148682, 0.5488215684890747, 0.5454545617103577, 0.5572390556335449, 0.45875421166419983, 0.49494948983192444, 0.5690235495567322, 0.563973069190979, 0.5319865345954895, NaN, 0.5328282713890076, 0.5976430773735046, 0.5656565427780151, 0.6077440977096558, 0.5707070827484131, 0.5917508602142334, 0.5202020406723022, 0.5210437774658203, 0.5909090638160706, 0.5808081030845642, 0.6144781112670898, 0.5909090638160706, 0.619528591632843, 0.6178451180458069, 0.6203703880310059, 0.5968013405799866, 0.6498316526412964, 0.6363636255264282, 0.6422559022903442, 0.6481481194496155, 0.6649831533432007, 0.6658248901367188, 0.680134654045105, 0.6582491397857666, 0.619528591632843, 0.68855220079422, 0.6616161465644836, 0.6868686676025391, 0.6683501601219177, 0.6683501601219177, 0.6632996797561646, 0.6582491397857666, 0.7188552021980286, 0.7095959782600403, 0.7390572428703308, 0.7390572428703308, 0.7710437774658203, 0.7609427571296692, 0.7323232293128967, 0.7878788113594055, 0.7794612646102905, 0.7727272510528564, 0.8417508602142334, 0.7819865345954895, 0.7676767706871033, 0.8265992999076843, 0.7937710285186768, 0.8148148059844971, 0.8821548819541931, 0.8434343338012695, 0.9882155060768127], [0.14646464586257935, 0.2348484843969345, 0.23400673270225525, 0.24915824830532074, 0.2718855142593384, 0.20875421166419983, 0.20875421166419983, 0.22390572726726532, 0.3670033812522888, 0.4545454680919647, 0.4595959484577179, 0.3787878751754761, 0.4873737394809723, 0.39393940567970276, 0.4267676770687103, 0.4343434274196625, 0.39225590229034424, 0.4932659864425659, 0.5454545617103577, 0.5235690474510193, 0.5303030014038086, 0.4739057123661041, 0.4898989796638489, 0.5555555820465088, 0.5521885752677917, 0.5185185074806213, 0.46717172861099243, NaN, 0.5707070827484131, 0.5269360542297363, 0.5917508602142334, 0.5841751098632812, 0.5740740895271301, 0.5218855142593384, 0.5016834735870361, 0.5824915766716003, 0.6111111044883728, 0.6161616444587708, 0.5437710285186768, 0.5841751098632812, 0.619528591632843, 0.6262626051902771, 0.5833333134651184, 0.6094276309013367, 0.6363636255264282, 0.5925925970077515, 0.6531986594200134, 0.6212121248245239, 0.6666666865348816, 0.6430976390838623, 0.6700336933135986, 0.6346801519393921, 0.6262626051902771, 0.6531986594200134, 0.6784511804580688, 0.6936026811599731, 0.7087541818618774, 0.689393937587738, 0.6767676472663879, 0.7171717286109924, 0.7196969985961914, 0.7239057421684265, 0.7575757503509521, 0.7971380352973938, 0.81144779920578, 0.7542087435722351, 0.804713785648346, 0.7929292917251587, 0.7878788113594055, 0.8434343338012695, 0.7929292917251587, 0.7777777910232544, 0.8131313323974609, 0.8097643256187439, 0.8501683473587036, 0.8956229090690613, 0.8602693676948547, 0.9831649661064148], [0.09090909361839294, 0.1666666716337204, 0.14057238399982452, 0.2070707082748413, 0.21717171370983124, 0.2053872048854828, 0.23400673270225525, 0.2222222238779068, 0.35353535413742065, 0.2946127951145172, 0.28282827138900757, 0.32996633648872375, 0.3670033812522888, 0.37457913160324097, 0.4141414165496826, 0.4124579131603241, 0.3611111044883728, 0.4276094138622284, 0.42003366351127625, 0.4208754301071167, 0.4503366947174072, 0.45875421166419983, 0.48569023609161377, 0.42171716690063477, 0.4747474789619446, 0.49663299322128296, 0.402356892824173, 0.4292929172515869, NaN, 0.5, 0.5782828330993652, 0.5462962985038757, 0.5757575631141663, 0.5269360542297363, 0.5572390556335449, 0.6077440977096558, 0.5942760705947876, 0.6026936173439026, 0.558922529220581, 0.6043770909309387, 0.6102693676948547, 0.5757575631141663, 0.5993266105651855, 0.618686854839325, 0.631313145160675, 0.6346801519393921, 0.691919207572937, 0.6590909361839294, 0.6767676472663879, 0.7079124450683594, 0.6742424368858337, 0.7171717286109924, 0.7104377150535583, 0.7289562225341797, 0.8417508602142334, 0.7424242496490479, 0.691919207572937, 0.744107723236084, 0.8063973188400269, 0.8434343338012695, 0.8400673270225525, 0.8299663066864014, 0.8097643256187439, 0.8013467788696289, 0.7289562225341797, 0.7946127653121948, 0.8156565427780151, 0.8316498398780823, 0.819865345954895, 0.8367003202438354, 0.8535353541374207, 0.8737373948097229, 0.8939393758773804, 0.9175084233283997, 0.8501683473587036, 0.9141414165496826, 0.9175084233283997, 0.9932659864425659], [0.09090909361839294, 0.14646464586257935, 0.12457912415266037, 0.17003366351127625, 0.17171716690063477, 0.17676767706871033, 0.2373737394809723, 0.22053872048854828, 0.34343433380126953, 0.3333333432674408, 0.34511783719062805, 0.3316498398780823, 0.3611111044883728, 0.3552188575267792, 0.372053861618042, 0.3872053921222687, 0.35353535413742065, 0.4276094138622284, 0.47306397557258606, 0.40067338943481445, 0.4469696879386902, 0.4461279511451721, 0.46296295523643494, 0.4789562225341797, 0.5202020406723022, 0.4579124450683594, 0.4343434274196625, 0.47306397557258606, 0.5, NaN, 0.5370370149612427, 0.5496633052825928, 0.5370370149612427, 0.5168350338935852, 0.5303030014038086, 0.5446127653121948, 0.6010100841522217, 0.5740740895271301, 0.5008417367935181, 0.5757575631141663, 0.5505050420761108, 0.5951178669929504, 0.6750841736793518, 0.6212121248245239, 0.619528591632843, 0.6144781112670898, 0.6363636255264282, 0.6287878751754761, 0.680134654045105, 0.7154881954193115, 0.6936026811599731, 0.680134654045105, 0.6902356743812561, 0.7037037014961243, 0.7222222089767456, 0.7138047218322754, 0.7104377150535583, 0.7205387353897095, 0.744107723236084, 0.7626262903213501, 0.7643097639083862, 0.8063973188400269, 0.804713785648346, 0.7626262903213501, 0.7979797720909119, 0.7912458181381226, 0.7929292917251587, 0.8299663066864014, 0.8341751098632812, 0.8703703880310059, 0.8501683473587036, 0.8400673270225525, 0.880471408367157, 0.8670033812522888, 0.8468013405799866, 0.9006733894348145, 0.8872053623199463, 0.9831649661064148], [0.09427609294652939, 0.15319865942001343, 0.12878787517547607, 0.17845118045806885, 0.17845118045806885, 0.19528619945049286, 0.2070707082748413, 0.20875421166419983, 0.3249158263206482, 0.28787878155708313, 0.25925925374031067, 0.3030303120613098, 0.3030303120613098, 0.2996633052825928, 0.36195287108421326, 0.37542086839675903, 0.3316498398780823, 0.3787878751754761, 0.34427610039711, 0.36868685483932495, 0.3872053921222687, 0.4225589334964752, 0.4579124450683594, 0.40740740299224854, 0.38804712891578674, 0.4595959484577179, 0.39225590229034424, 0.408249169588089, 0.42171716690063477, 0.46296295523643494, NaN, 0.4957912564277649, 0.47727271914482117, 0.5286195278167725, 0.5454545617103577, 0.5143097639083862, 0.5622895359992981, 0.563973069190979, 0.5117844939231873, 0.5723905563354492, 0.5656565427780151, 0.5707070827484131, 0.567340075969696, 0.625420868396759, 0.624579131603241, 0.6178451180458069, 0.6649831533432007, 0.6683501601219177, 0.6481481194496155, 0.68855220079422, 0.6582491397857666, 0.680134654045105, 0.7424242496490479, 0.7003366947174072, 0.7693602442741394, 0.7382155060768127, 0.6784511804580688, 0.7407407164573669, 0.7659932374954224, 0.8232323527336121, 0.8013467788696289, 0.875420868396759, 0.816498339176178, 0.7794612646102905, 0.747474730014801, 0.7979797720909119, 0.8013467788696289, 0.8215488195419312, 0.8350168466567993, 0.8013467788696289, 0.8569023609161377, 0.8653198480606079, 0.8787878751754761, 0.8872053623199463, 0.8703703880310059, 0.9006733894348145, 0.9225589036941528, 0.9898989796638489], [0.09090909361839294, 0.15656565129756927, 0.12121212482452393, 0.18855218589305878, 0.19696970283985138, 0.19360269606113434, 0.19696970283985138, 0.2095959633588791, 0.3232323229312897, 0.27441078424453735, 0.30892255902290344, 0.3097642958164215, 0.38552188873291016, 0.3484848439693451, 0.36026936769485474, 0.36447811126708984, 0.32996633648872375, 0.39898988604545593, 0.4191919267177582, 0.4444444477558136, 0.43939393758773804, 0.4141414165496826, 0.4377104341983795, 0.43939393758773804, 0.4503366947174072, 0.42171716690063477, 0.4292929172515869, 0.41582491993904114, 0.45370370149612427, 0.4503366947174072, 0.5042087435722351, NaN, 0.48232322931289673, 0.4898989796638489, 0.48821547627449036, 0.5050504803657532, 0.6085858345031738, 0.5471380352973938, 0.5117844939231873, 0.5235690474510193, 0.5454545617103577, 0.5833333134651184, 0.5555555820465088, 0.5942760705947876, 0.6144781112670898, 0.566498339176178, 0.6363636255264282, 0.6043770909309387, 0.6262626051902771, 0.631313145160675, 0.6632996797561646, 0.6565656661987305, 0.6296296119689941, 0.7003366947174072, 0.7020202279090881, 0.6961279511451721, 0.7323232293128967, 0.695286214351654, 0.7407407164573669, 0.7188552021980286, 0.7811447978019714, 0.7457912564277649, 0.8131313323974609, 0.819865345954895, 0.8131313323974609, 0.7676767706871033, 0.8367003202438354, 0.8148148059844971, 0.880471408367157, 0.941077470779419, 0.8493266105651855, 0.8829966187477112, 0.939393937587738, 0.9276094436645508, 0.869528591632843, 0.9090909361839294, 0.9090909361839294, 0.9915825128555298], [0.07575757801532745, 0.16582491993904114, 0.12457912415266037, 0.19696970283985138, 0.1835016906261444, 0.19528619945049286, 0.2222222238779068, 0.21212121844291687, 0.33670035004615784, 0.2306397259235382, 0.2525252401828766, 0.29629629850387573, 0.3097642958164215, 0.33249157667160034, 0.3670033812522888, 0.3947811424732208, 0.3308080732822418, 0.3501683473587036, 0.3181818127632141, 0.3863636255264282, 0.4208754301071167, 0.4267676770687103, 0.4461279511451721, 0.4250841736793518, 0.41329964995384216, 0.4570707082748413, 0.408249169588089, 0.42592594027519226, 0.42424243688583374, 0.46296295523643494, 0.5227272510528564, 0.5176767706871033, NaN, 0.5, 0.5252525210380554, 0.5917508602142334, 0.5757575631141663, 0.560606062412262, 0.503367006778717, 0.5782828330993652, 0.5774410963058472, 0.563973069190979, 0.5976430773735046, 0.5740740895271301, 0.624579131603241, 0.6010100841522217, 0.6616161465644836, 0.6144781112670898, 0.6397306323051453, 0.6531986594200134, 0.6228955984115601, 0.6649831533432007, 0.6574074029922485, 0.7087541818618774, 0.8821548819541931, 0.7205387353897095, 0.689393937587738, 0.7205387353897095, 0.7971380352973938, 0.8872053623199463, 0.8223905563354492, 0.7996633052825928, 0.7929292917251587, 0.7946127653121948, 0.7382155060768127, 0.7744107842445374, 0.8063973188400269, 0.7979797720909119, 0.7912458181381226, 0.8232323527336121, 0.8240740895271301, 0.8434343338012695, 0.8855218887329102, 0.9175084233283997, 0.8484848737716675, 0.8989899158477783, 0.8964646458625793, 0.9848484992980957], [0.09595959633588791, 0.1582491546869278, 0.2053872048854828, 0.18013468384742737, 0.2626262605190277, 0.29629629850387573, 0.26599326729774475, 0.28956228494644165, 0.19276094436645508, 0.4107744097709656, 0.3905723989009857, 0.4208754301071167, 0.4040403962135315, 0.3956229090690613, 0.35774409770965576, 0.32154881954193115, 0.43265992403030396, 0.4612794518470764, 0.40909090638160706, 0.43518519401550293, 0.41750842332839966, 0.3653198778629303, 0.43265992403030396, 0.46296295523643494, 0.4461279511451721, 0.469696968793869, 0.47979798913002014, 0.4781144857406616, 0.47306397557258606, 0.4831649959087372, 0.47138047218322754, 0.5101010203361511, 0.5, NaN, 0.5151515007019043, 0.5185185074806213, 0.5168350338935852, 0.4983164966106415, 0.5151515007019043, 0.5336700081825256, 0.5437710285186768, 0.5538720488548279, 0.5707070827484131, 0.5774410963058472, 0.5328282713890076, 0.5572390556335449, 0.5622895359992981, 0.5875421166419983, 0.5513467788696289, 0.6026936173439026, 0.6363636255264282, 0.6548821330070496, 0.6296296119689941, 0.6818181872367859, 0.6329966187477112, 0.6380471587181091, 0.7424242496490479, 0.6380471587181091, 0.6717171669006348, 0.691919207572937, 0.7222222089767456, 0.7407407164573669, 0.7508417367935181, 0.695286214351654, 0.8434343338012695, 0.7340067625045776, 0.7239057421684265, 0.7929292917251587, 0.8299663066864014, 0.7962962985038757, 0.819865345954895, 0.8350168466567993, 0.8585858345031738, 0.8316498398780823, 0.8299663066864014, 0.8922559022903442, 0.875420868396759, 0.9932659864425659], [0.14141413569450378, 0.23232322931289673, 0.2373737394809723, 0.26767677068710327, 0.2289562225341797, 0.23148147761821747, 0.21717171370983124, 0.2390572428703308, 0.3585858643054962, 0.3737373650074005, 0.4040403962135315, 0.37037035822868347, 0.3670033812522888, 0.4292929172515869, 0.3872053921222687, 0.3905723989009857, 0.4503366947174072, 0.4377104341983795, 0.4191919267177582, 0.4444444477558136, 0.4461279511451721, 0.44949495792388916, 0.41582491993904114, 0.4276094138622284, 0.4579124450683594, 0.4528619647026062, 0.4789562225341797, 0.4983164966106415, 0.4427609443664551, 0.469696968793869, 0.4545454680919647, 0.5117844939231873, 0.4747474789619446, 0.4848484992980957, NaN, 0.4831649959087372, 0.5, 0.4932659864425659, 0.5084174871444702, 0.5808081030845642, 0.5336700081825256, 0.5909090638160706, 0.5420875549316406, 0.4831649959087372, 0.5269360542297363, 0.5227272510528564, 0.5597643256187439, 0.5117844939231873, 0.6346801519393921, 0.6397306323051453, 0.6649831533432007, 0.6127946376800537, 0.5505050420761108, 0.6380471587181091, 0.5942760705947876, 0.6127946376800537, 0.6414141654968262, 0.6094276309013367, 0.6523569226264954, 0.6616161465644836, 0.6675084233283997, 0.6599326729774475, 0.7356902360916138, 0.7609427571296692, 0.7508417367935181, 0.7643097639083862, 0.7693602442741394, 0.7744107842445374, 0.7710437774658203, 0.7727272510528564, 0.7819865345954895, 0.7424242496490479, 0.7794612646102905, 0.7845118045806885, 0.81144779920578, 0.8131313323974609, 0.8518518805503845, 0.9848484992980957], [0.07070706784725189, 0.1498316526412964, 0.12121212482452393, 0.18686868250370026, 0.15488214790821075, 0.1776094287633896, 0.19191919267177582, 0.19528619945049286, 0.3232323229312897, 0.2070707082748413, 0.2609427571296692, 0.2777777910232544, 0.27272728085517883, 0.33670035004615784, 0.3468013405799866, 0.36195287108421326, 0.3232323229312897, 0.3156565725803375, 0.2861952781677246, 0.36868685483932495, 0.40572389960289, 0.41750842332839966, 0.41582491993904114, 0.38973063230514526, 0.37542086839675903, 0.402356892824173, 0.40909090638160706, 0.41750842332839966, 0.39225590229034424, 0.4553872048854828, 0.48569023609161377, 0.49494948983192444, 0.408249169588089, 0.48148149251937866, 0.5168350338935852, NaN, 0.5412458181381226, 0.5404040217399597, 0.4983164966106415, 0.5454545617103577, 0.5488215684890747, 0.5336700081825256, 0.5656565427780151, 0.5277777910232544, 0.5959596037864685, 0.5774410963058472, 0.6296296119689941, 0.5774410963058472, 0.6127946376800537, 0.6296296119689941, 0.6212121248245239, 0.6658248901367188, 0.6161616444587708, 0.7037037014961243, 0.8249158263206482, 0.7070707082748413, 0.6717171669006348, 0.7095959782600403, 0.7744107842445374, 0.8518518805503845, 0.8131313323974609, 0.7912458181381226, 0.7727272510528564, 0.7828282713890076, 0.7289562225341797, 0.7710437774658203, 0.804713785648346, 0.7727272510528564, 0.7811447978019714, 0.81144779920578, 0.816498339176178, 0.8442760705947876, 0.8872053623199463, 0.9040403962135315, 0.8417508602142334, 0.8973063826560974, 0.8939393758773804, 0.9814814925193787], [0.10437710583209991, 0.15993265807628632, 0.14562289416790009, 0.19949494302272797, 0.19696970283985138, 0.19528619945049286, 0.18855218589305878, 0.20875421166419983, 0.2946127951145172, 0.2861952781677246, 0.28282827138900757, 0.31060606241226196, 0.3821548819541931, 0.3308080732822418, 0.35185185074806213, 0.3459596037864685, 0.32575756311416626, 0.3695286214351654, 0.3653198778629303, 0.41750842332839966, 0.43097642064094543, 0.4124579131603241, 0.4377104341983795, 0.39225590229034424, 0.41582491993904114, 0.37457913160324097, 0.4191919267177582, 0.3888888955116272, 0.40572389960289, 0.39898988604545593, 0.4377104341983795, 0.3914141356945038, 0.42424243688583374, 0.4831649959087372, 0.5, 0.45875421166419983, NaN, 0.505892276763916, 0.42592594027519226, 0.5067340135574341, 0.5235690474510193, 0.5496633052825928, 0.5404040217399597, 0.558080792427063, 0.5740740895271301, 0.5404040217399597, 0.5976430773735046, 0.5808081030845642, 0.5808081030845642, 0.6127946376800537, 0.6397306323051453, 0.6397306323051453, 0.6144781112670898, 0.6851851940155029, 0.6094276309013367, 0.7037037014961243, 0.7053872346878052, 0.7087541818618774, 0.6717171669006348, 0.6734007000923157, 0.7272727489471436, 0.747474730014801, 0.8063973188400269, 0.816498339176178, 0.8249158263206482, 0.7861952781677246, 0.8400673270225525, 0.7937710285186768, 0.8619528412818909, 0.9040403962135315, 0.8308081030845642, 0.872053861618042, 0.8855218887329102, 0.8653198480606079, 0.8872053623199463, 0.8989899158477783, 0.9141414165496826, 0.9915825128555298], [0.09090909361839294, 0.15656565129756927, 0.13804714381694794, 0.17845118045806885, 0.18013468384742737, 0.18013468384742737, 0.1902356892824173, 0.18013468384742737, 0.3181818127632141, 0.2306397259235382, 0.25589224696159363, 0.28787878155708313, 0.2525252401828766, 0.3030303120613098, 0.3670033812522888, 0.3594276010990143, 0.3156565725803375, 0.3249158263206482, 0.3358585834503174, 0.3501683473587036, 0.34343433380126953, 0.40572389960289, 0.4141414165496826, 0.40067338943481445, 0.3670033812522888, 0.39898988604545593, 0.38552188873291016, 0.38383838534355164, 0.3973063826560974, 0.42592594027519226, 0.436026930809021, 0.4528619647026062, 0.43939393758773804, 0.5016834735870361, 0.5067340135574341, 0.4595959484577179, 0.49410775303840637, NaN, 0.4444444477558136, 0.5319865345954895, 0.5303030014038086, 0.5176767706871033, 0.5437710285186768, 0.5185185074806213, 0.689393937587738, 0.5370370149612427, 0.7340067625045776, 0.5505050420761108, 0.6094276309013367, 0.6388888955116272, 0.5934343338012695, 0.6641414165496826, 0.6212121248245239, 0.6700336933135986, 0.6927609443664551, 0.7205387353897095, 0.68855220079422, 0.7205387353897095, 0.7104377150535583, 0.7087541818618774, 0.7323232293128967, 0.7946127653121948, 0.7609427571296692, 0.8148148059844971, 0.7744107842445374, 0.7962962985038757, 0.8333333134651184, 0.7996633052825928, 0.816498339176178, 0.7643097639083862, 0.817340075969696, 0.8417508602142334, 0.8417508602142334, 0.8417508602142334, 0.8653198480606079, 0.8973063826560974, 0.9099326729774475, 0.9831649661064148], [0.06397306174039841, 0.11279461532831192, 0.08249158412218094, 0.15993265807628632, 0.12289562076330185, 0.1835016906261444, 0.23400673270225525, 0.19276094436645508, 0.31986531615257263, 0.31481480598449707, 0.30471381545066833, 0.3181818127632141, 0.3265993297100067, 0.3181818127632141, 0.33838382363319397, 0.3333333432674408, 0.31986531615257263, 0.3888888955116272, 0.436026930809021, 0.3787878751754761, 0.44781145453453064, 0.3930976390838623, 0.4276094138622284, 0.4343434274196625, 0.466329962015152, 0.4292929172515869, 0.40909090638160706, 0.45622894167900085, 0.44107744097709656, 0.49915826320648193, 0.48821547627449036, 0.48821547627449036, 0.49663299322128296, 0.4848484992980957, 0.4915824830532074, 0.5016834735870361, 0.5740740895271301, 0.5555555820465088, NaN, 0.5151515007019043, 0.5168350338935852, 0.5488215684890747, 0.6481481194496155, 0.5538720488548279, 0.5951178669929504, 0.5656565427780151, 0.6026936173439026, 0.5858585834503174, 0.6515151262283325, 0.624579131603241, 0.6262626051902771, 0.6296296119689941, 0.6010100841522217, 0.6632996797561646, 0.7087541818618774, 0.6414141654968262, 0.619528591632843, 0.6430976390838623, 0.683501660823822, 0.7205387353897095, 0.68855220079422, 0.7205387353897095, 0.7272727489471436, 0.6818181872367859, 0.7037037014961243, 0.7070707082748413, 0.6969696879386902, 0.7508417367935181, 0.7609427571296692, 0.8358585834503174, 0.752525269985199, 0.7744107842445374, 0.8232323527336121, 0.8148148059844971, 0.744107723236084, 0.8215488195419312, 0.804713785648346, 0.9494949579238892], [0.05218855291604996, 0.11279461532831192, 0.09764309972524643, 0.13299663364887238, 0.12794612348079681, 0.14646464586257935, 0.15319865942001343, 0.15488214790821075, 0.27441078424453735, 0.26430976390838623, 0.22390572726726532, 0.3080808222293854, 0.29797980189323425, 0.2946127951145172, 0.341750830411911, 0.3207070827484131, 0.31986531615257263, 0.3232323229312897, 0.3872053921222687, 0.33670035004615784, 0.36026936769485474, 0.37542086839675903, 0.3804713785648346, 0.39393940567970276, 0.41750842332839966, 0.402356892824173, 0.3804713785648346, 0.41582491993904114, 0.3956229090690613, 0.42424243688583374, 0.4276094138622284, 0.4764309823513031, 0.42171716690063477, 0.466329962015152, 0.4191919267177582, 0.4545454680919647, 0.4932659864425659, 0.4680134654045105, 0.4848484992980957, NaN, 0.46296295523643494, 0.5774410963058472, 0.5538720488548279, 0.5151515007019043, 0.5303030014038086, 0.554713785648346, 0.563973069190979, 0.5521885752677917, 0.5993266105651855, 0.7592592835426331, 0.6877104640007019, 0.6464646458625793, 0.5723905563354492, 0.6599326729774475, 0.6750841736793518, 0.6936026811599731, 0.631313145160675, 0.7011784315109253, 0.6969696879386902, 0.691919207572937, 0.7222222089767456, 0.7752525210380554, 0.8737373948097229, 0.7685185074806213, 0.7777777910232544, 0.7878788113594055, 0.8030303120613098, 0.8063973188400269, 0.8442760705947876, 0.7845118045806885, 0.8417508602142334, 0.804713785648346, 0.8030303120613098, 0.8013467788696289, 0.8602693676948547, 0.8989899158477783, 0.8922559022903442, 0.9797979593276978], [0.08417508751153946, 0.15488214790821075, 0.13804714381694794, 0.16835017502307892, 0.16582491993904114, 0.18686868250370026, 0.18686868250370026, 0.19360269606113434, 0.3316498398780823, 0.2222222238779068, 0.2710437774658203, 0.32996633648872375, 0.15572389960289001, 0.3232323229312897, 0.341750830411911, 0.36447811126708984, 0.3316498398780823, 0.31481480598449707, 0.32575756311416626, 0.2777777910232544, 0.372053861618042, 0.40572389960289, 0.3888888955116272, 0.3787878751754761, 0.36195287108421326, 0.4343434274196625, 0.3821548819541931, 0.3804713785648346, 0.38973063230514526, 0.44949495792388916, 0.4343434274196625, 0.4545454680919647, 0.4225589334964752, 0.45622894167900085, 0.466329962015152, 0.4511784613132477, 0.4764309823513031, 0.469696968793869, 0.4831649959087372, 0.5370370149612427, NaN, 0.5690235495567322, 0.5488215684890747, 0.5, 0.5437710285186768, 0.5361952781677246, 0.5513467788696289, 0.5235690474510193, 0.6111111044883728, 0.627946138381958, 0.6515151262283325, 0.6161616444587708, 0.5875421166419983, 0.6414141654968262, 0.6363636255264282, 0.7239057421684265, 0.6632996797561646, 0.7306397557258606, 0.7037037014961243, 0.6498316526412964, 0.7154881954193115, 0.747474730014801, 0.7811447978019714, 0.7693602442741394, 0.7340067625045776, 0.804713785648346, 0.7853535413742065, 0.7904040217399597, 0.7946127653121948, 0.7424242496490479, 0.7962962985038757, 0.81144779920578, 0.8299663066864014, 0.8282828330993652, 0.8400673270225525, 0.880471408367157, 0.9090909361839294, 0.9764309525489807], [0.10101009905338287, 0.15909090638160706, 0.13804714381694794, 0.16329966485500336, 0.16835017502307892, 0.18097643554210663, 0.16329966485500336, 0.19696970283985138, 0.24915824830532074, 0.2095959633588791, 0.2390572428703308, 0.32407405972480774, 0.2861952781677246, 0.3013468086719513, 0.31986531615257263, 0.30471381545066833, 0.3207070827484131, 0.3030303120613098, 0.3821548819541931, 0.3308080732822418, 0.35606059432029724, 0.36026936769485474, 0.30639731884002686, 0.39898988604545593, 0.408249169588089, 0.3569023609161377, 0.37962964177131653, 0.3737373650074005, 0.42424243688583374, 0.40488216280937195, 0.4292929172515869, 0.4166666567325592, 0.436026930809021, 0.4461279511451721, 0.40909090638160706, 0.466329962015152, 0.4503366947174072, 0.48232322931289673, 0.4511784613132477, 0.4225589334964752, 0.43097642064094543, NaN, 0.5370370149612427, 0.5353535413742065, 0.5319865345954895, 0.4747474789619446, 0.5437710285186768, 0.5572390556335449, 0.566498339176178, 0.5622895359992981, 0.7087541818618774, 0.6026936173439026, 0.560606062412262, 0.6127946376800537, 0.6111111044883728, 0.6447811722755432, 0.5875421166419983, 0.6498316526412964, 0.6877104640007019, 0.6506733894348145, 0.6986532211303711, 0.6969696879386902, 0.7601010203361511, 0.7710437774658203, 0.7811447978019714, 0.7626262903213501, 0.7996633052825928, 0.7609427571296692, 0.9511784315109253, 0.7845118045806885, 0.7895622849464417, 0.804713785648346, 0.8265992999076843, 0.7895622849464417, 0.8585858345031738, 0.8855218887329102, 0.8821548819541931, 0.9797979593276978], [0.08080808073282242, 0.11868686974048615, 0.12962962687015533, 0.12205386906862259, 0.1346801370382309, 0.16835017502307892, 0.1666666716337204, 0.1835016906261444, 0.24410773813724518, 0.26767677068710327, 0.26430976390838623, 0.2760942876338959, 0.28282827138900757, 0.3232323229312897, 0.283670037984848, 0.28535354137420654, 0.34427610039711, 0.3316498398780823, 0.34511783719062805, 0.3249158263206482, 0.35185185074806213, 0.3585858643054962, 0.372053861618042, 0.3821548819541931, 0.37037035822868347, 0.34343433380126953, 0.4031986594200134, 0.4166666567325592, 0.40067338943481445, 0.3249158263206482, 0.43265992403030396, 0.4444444477558136, 0.402356892824173, 0.4292929172515869, 0.4579124450683594, 0.4343434274196625, 0.4595959484577179, 0.45622894167900085, 0.35185185074806213, 0.4461279511451721, 0.4511784613132477, 0.46296295523643494, NaN, 0.5707070827484131, 0.5117844939231873, 0.46717172861099243, 0.5505050420761108, 0.5858585834503174, 0.5134680271148682, 0.5420875549316406, 0.5437710285186768, 0.5757575631141663, 0.6228955984115601, 0.6228955984115601, 0.5791245698928833, 0.6649831533432007, 0.691919207572937, 0.6717171669006348, 0.6515151262283325, 0.6700336933135986, 0.7171717286109924, 0.7306397557258606, 0.7794612646102905, 0.7373737096786499, 0.7710437774658203, 0.7744107842445374, 0.7651515007019043, 0.7845118045806885, 0.7895622849464417, 0.7575757503509521, 0.8148148059844971, 0.8451178669929504, 0.8358585834503174, 0.8367003202438354, 0.8602693676948547, 0.8670033812522888, 0.8973063826560974, 0.9865319728851318], [0.0791245773434639, 0.1111111119389534, 0.08585858345031738, 0.14814814925193787, 0.12457912415266037, 0.16919191181659698, 0.17845118045806885, 0.186026930809021, 0.2609427571296692, 0.2457912415266037, 0.21632996201515198, 0.24242424964904785, 0.2542087435722351, 0.26430976390838623, 0.2794612646102905, 0.31313130259513855, 0.29124578833580017, 0.3569023609161377, 0.29377105832099915, 0.30639731884002686, 0.34427610039711, 0.35185185074806213, 0.3872053921222687, 0.3468013405799866, 0.34511783719062805, 0.34343433380126953, 0.3501683473587036, 0.3905723989009857, 0.38131314516067505, 0.3787878751754761, 0.37457913160324097, 0.40572389960289, 0.42592594027519226, 0.4225589334964752, 0.5168350338935852, 0.4722222089767456, 0.4419191777706146, 0.48148149251937866, 0.4461279511451721, 0.4848484992980957, 0.5, 0.46464645862579346, 0.4292929172515869, NaN, 0.5202020406723022, 0.5185185074806213, 0.5370370149612427, 0.5951178669929504, 0.5387205481529236, 0.5791245698928833, 0.5277777910232544, 0.5572390556335449, 0.7744107842445374, 0.5976430773735046, 0.6447811722755432, 0.695286214351654, 0.6363636255264282, 0.695286214351654, 0.6481481194496155, 0.7314814925193787, 0.6936026811599731, 0.7643097639083862, 0.7003366947174072, 0.7070707082748413, 0.744107723236084, 0.7710437774658203, 0.7272727489471436, 0.7491582632064819, 0.7710437774658203, 0.7659932374954224, 0.7878788113594055, 0.7693602442741394, 0.804713785648346, 0.7912458181381226, 0.8232323527336121, 0.8400673270225525, 0.8636363744735718, 0.9789562225341797], [0.0791245773434639, 0.13804714381694794, 0.1262626200914383, 0.1515151560306549, 0.14225588738918304, 0.15572389960289001, 0.1582491546869278, 0.1666666716337204, 0.280303031206131, 0.2138047069311142, 0.2390572428703308, 0.2626262605190277, 0.2138047069311142, 0.28282827138900757, 0.33838382363319397, 0.3291245698928833, 0.2996633052825928, 0.28114476799964905, 0.2861952781677246, 0.2861952781677246, 0.2996633052825928, 0.3653198778629303, 0.3821548819541931, 0.35606059432029724, 0.32154881954193115, 0.3585858643054962, 0.3636363744735718, 0.3636363744735718, 0.36868685483932495, 0.3804713785648346, 0.37542086839675903, 0.38552188873291016, 0.37542086839675903, 0.46717172861099243, 0.47306397557258606, 0.4040403962135315, 0.42592594027519226, 0.31060606241226196, 0.40488216280937195, 0.469696968793869, 0.45622894167900085, 0.4680134654045105, 0.48821547627449036, 0.47979798913002014, NaN, 0.48148149251937866, 0.6422559022903442, 0.505892276763916, 0.5286195278167725, 0.5858585834503174, 0.566498339176178, 0.6010100841522217, 0.5740740895271301, 0.6397306323051453, 0.6414141654968262, 0.7020202279090881, 0.6523569226264954, 0.7079124450683594, 0.6649831533432007, 0.6599326729774475, 0.7037037014961243, 0.744107723236084, 0.7356902360916138, 0.7794612646102905, 0.753367006778717, 0.7794612646102905, 0.8013467788696289, 0.7777777910232544, 0.7811447978019714, 0.7407407164573669, 0.7946127653121948, 0.8148148059844971, 0.8282828330993652, 0.8215488195419312, 0.8400673270225525, 0.8787878751754761, 0.9107744097709656, 0.9831649661064148], [0.07070706784725189, 0.0917508453130722, 0.09427609294652939, 0.1346801370382309, 0.11784511804580688, 0.14478114247322083, 0.1388888955116272, 0.15656565129756927, 0.2264309823513031, 0.21717171370983124, 0.21548821032047272, 0.2626262605190277, 0.26767677068710327, 0.27441078424453735, 0.2693602740764618, 0.2710437774658203, 0.2946127951145172, 0.30639731884002686, 0.3265993297100067, 0.33670035004615784, 0.3787878751754761, 0.3484848439693451, 0.33838382363319397, 0.3316498398780823, 0.37542086839675903, 0.29377105832099915, 0.35774409770965576, 0.40740740299224854, 0.3653198778629303, 0.38552188873291016, 0.3821548819541931, 0.4335016906261444, 0.39898988604545593, 0.4427609443664551, 0.47727271914482117, 0.4225589334964752, 0.4595959484577179, 0.46296295523643494, 0.4343434274196625, 0.44528618454933167, 0.4638047218322754, 0.5252525210380554, 0.5328282713890076, 0.48148149251937866, 0.5185185074806213, NaN, 0.558922529220581, 0.5328282713890076, 0.5622895359992981, 0.5707070827484131, 0.5892255902290344, 0.5707070827484131, 0.563973069190979, 0.6296296119689941, 0.6161616444587708, 0.6776094436645508, 0.6010100841522217, 0.6717171669006348, 0.6649831533432007, 0.6700336933135986, 0.7104377150535583, 0.746632993221283, 0.7828282713890076, 0.6725589036941528, 0.7163299918174744, 0.7491582632064819, 0.7138047218322754, 0.7390572428703308, 0.8030303120613098, 0.7895622849464417, 0.7794612646102905, 0.7811447978019714, 0.81144779920578, 0.8131313323974609, 0.7929292917251587, 0.8535353541374207, 0.8459596037864685, 0.9696969985961914], [0.08417508751153946, 0.12037037312984467, 0.12121212482452393, 0.12962962687015533, 0.14814814925193787, 0.15488214790821075, 0.15993265807628632, 0.16498316824436188, 0.25084173679351807, 0.2112794667482376, 0.2070707082748413, 0.2609427571296692, 0.22138047218322754, 0.2710437774658203, 0.30639731884002686, 0.3265993297100067, 0.2946127951145172, 0.3097642958164215, 0.2289562225341797, 0.2710437774658203, 0.25757575035095215, 0.3569023609161377, 0.3737373650074005, 0.31144779920578003, 0.2760942876338959, 0.32828283309936523, 0.35185185074806213, 0.3468013405799866, 0.3080808222293854, 0.3636363744735718, 0.3350168466567993, 0.3636363744735718, 0.33838382363319397, 0.4377104341983795, 0.4402357041835785, 0.37037035822868347, 0.402356892824173, 0.26599326729774475, 0.3973063826560974, 0.436026930809021, 0.4486531913280487, 0.45622894167900085, 0.44949495792388916, 0.46296295523643494, 0.35774409770965576, 0.44107744097709656, NaN, 0.48821547627449036, 0.5084174871444702, 0.5841751098632812, 0.5521885752677917, 0.5808081030845642, 0.5387205481529236, 0.5993266105651855, 0.5656565427780151, 0.68855220079422, 0.6481481194496155, 0.6868686676025391, 0.6414141654968262, 0.6228955984115601, 0.6717171669006348, 0.7457912564277649, 0.7272727489471436, 0.7575757503509521, 0.7575757503509521, 0.7760942578315735, 0.7920875549316406, 0.7592592835426331, 0.7861952781677246, 0.7087541818618774, 0.7744107842445374, 0.8383838534355164, 0.8148148059844971, 0.7996633052825928, 0.8653198480606079, 0.8585858345031738, 0.9259259104728699, 0.9814814925193787], [0.07070706784725189, 0.11279461532831192, 0.08249158412218094, 0.13636364042758942, 0.11784511804580688, 0.1582491546869278, 0.16835017502307892, 0.16498316824436188, 0.23400673270225525, 0.22138047218322754, 0.19360269606113434, 0.23400673270225525, 0.2222222238779068, 0.25925925374031067, 0.2609427571296692, 0.2946127951145172, 0.28114476799964905, 0.32828283309936523, 0.2626262605190277, 0.2996633052825928, 0.3316498398780823, 0.341750830411911, 0.3569023609161377, 0.31986531615257263, 0.3232323229312897, 0.3333333432674408, 0.3350168466567993, 0.3787878751754761, 0.34090909361839294, 0.3712121248245239, 0.3316498398780823, 0.3956229090690613, 0.38552188873291016, 0.4124579131603241, 0.48821547627449036, 0.4225589334964752, 0.4191919267177582, 0.44949495792388916, 0.4141414165496826, 0.44781145453453064, 0.4764309823513031, 0.4427609443664551, 0.4141414165496826, 0.40488216280937195, 0.49410775303840637, 0.46717172861099243, 0.5117844939231873, NaN, 0.5218855142593384, 0.5488215684890747, 0.5042087435722351, 0.5353535413742065, 0.695286214351654, 0.5824915766716003, 0.6271044015884399, 0.6818181872367859, 0.6262626051902771, 0.6818181872367859, 0.624579131603241, 0.691919207572937, 0.6683501601219177, 0.7289562225341797, 0.6936026811599731, 0.691919207572937, 0.7154881954193115, 0.7592592835426331, 0.7020202279090881, 0.7112794518470764, 0.7542087435722351, 0.7424242496490479, 0.746632993221283, 0.7491582632064819, 0.7946127653121948, 0.7693602442741394, 0.8063973188400269, 0.8324915766716003, 0.8417508602142334, 0.9696969985961914], [0.06902357190847397, 0.09090909361839294, 0.11279461532831192, 0.11447811126708984, 0.11447811126708984, 0.1430976390838623, 0.12962962687015533, 0.1515151560306549, 0.2222222238779068, 0.1304713785648346, 0.19696970283985138, 0.2542087435722351, 0.18265993893146515, 0.2861952781677246, 0.2794612646102905, 0.2845117747783661, 0.28956228494644165, 0.2306397259235382, 0.29124578833580017, 0.23232322931289673, 0.28956228494644165, 0.33838382363319397, 0.33670035004615784, 0.31144779920578003, 0.31060606241226196, 0.2929292917251587, 0.33417508006095886, 0.3333333432674408, 0.3232323229312897, 0.31986531615257263, 0.35185185074806213, 0.3737373650074005, 0.36026936769485474, 0.4486531913280487, 0.3653198778629303, 0.3872053921222687, 0.4191919267177582, 0.3905723989009857, 0.3484848439693451, 0.40067338943481445, 0.3888888955116272, 0.4335016906261444, 0.48653197288513184, 0.4612794518470764, 0.47138047218322754, 0.4377104341983795, 0.4915824830532074, 0.4781144857406616, NaN, 0.558922529220581, 0.5690235495567322, 0.567340075969696, 0.5252525210380554, 0.6060606241226196, 0.5909090638160706, 0.6700336933135986, 0.5993266105651855, 0.683501660823822, 0.6397306323051453, 0.6346801519393921, 0.6784511804580688, 0.7154881954193115, 0.7550504803657532, 0.7289562225341797, 0.7735690474510193, 0.7659932374954224, 0.7710437774658203, 0.7777777910232544, 0.8468013405799866, 0.7306397557258606, 0.819865345954895, 0.808080792427063, 0.8131313323974609, 0.7946127653121948, 0.8451178669929504, 0.8939393758773804, 0.8872053623199463, 0.9814814925193787], [0.04040404036641121, 0.07575757801532745, 0.07070706784725189, 0.09090909361839294, 0.08417508751153946, 0.1026936024427414, 0.10101009905338287, 0.10437710583209991, 0.20875421166419983, 0.2070707082748413, 0.16835017502307892, 0.2373737394809723, 0.20454545319080353, 0.23400673270225525, 0.24747474491596222, 0.2457912415266037, 0.25925925374031067, 0.24410773813724518, 0.2693602740764618, 0.2516835033893585, 0.28114476799964905, 0.30639731884002686, 0.3080808222293854, 0.2794612646102905, 0.313973069190979, 0.29629629850387573, 0.31986531615257263, 0.3569023609161377, 0.2920875549316406, 0.2845117747783661, 0.31144779920578003, 0.36868685483932495, 0.3468013405799866, 0.3973063826560974, 0.36026936769485474, 0.37037035822868347, 0.3872053921222687, 0.3611111044883728, 0.37542086839675903, 0.24074074625968933, 0.372053861618042, 0.4377104341983795, 0.4579124450683594, 0.4208754301071167, 0.4141414165496826, 0.4292929172515869, 0.41582491993904114, 0.4511784613132477, 0.44107744097709656, NaN, 0.5303030014038086, 0.5530303120613098, 0.49663299322128296, 0.5757575631141663, 0.5404040217399597, 0.6683501601219177, 0.5690235495567322, 0.6700336933135986, 0.6077440977096558, 0.5799663066864014, 0.6430976390838623, 0.6936026811599731, 0.7819865345954895, 0.683501660823822, 0.755892276763916, 0.7592592835426331, 0.7079124450683594, 0.7222222089767456, 0.7626262903213501, 0.7239057421684265, 0.7592592835426331, 0.7550504803657532, 0.752525269985199, 0.7542087435722351, 0.8215488195419312, 0.8417508602142334, 0.8451178669929504, 0.9713804721832275], [0.06902357190847397, 0.10858585685491562, 0.11447811126708984, 0.11279461532831192, 0.12121212482452393, 0.12457912415266037, 0.12289562076330185, 0.13299663364887238, 0.18518517911434174, 0.18939393758773804, 0.18855218589305878, 0.2415824979543686, 0.22558923065662384, 0.2525252401828766, 0.26767677068710327, 0.24074074625968933, 0.2904040515422821, 0.2306397259235382, 0.2845117747783661, 0.25925925374031067, 0.27441078424453735, 0.28114476799964905, 0.27272728085517883, 0.3164983093738556, 0.33838382363319397, 0.28535354137420654, 0.341750830411911, 0.32996633648872375, 0.32575756311416626, 0.30639731884002686, 0.341750830411911, 0.33670035004615784, 0.37710437178611755, 0.3636363744735718, 0.3350168466567993, 0.3787878751754761, 0.36026936769485474, 0.40656566619873047, 0.3737373650074005, 0.3122895658016205, 0.3484848439693451, 0.29124578833580017, 0.45622894167900085, 0.4722222089767456, 0.4335016906261444, 0.4107744097709656, 0.44781145453453064, 0.4957912564277649, 0.43097642064094543, 0.469696968793869, NaN, 0.5336700081825256, 0.5202020406723022, 0.5538720488548279, 0.5454545617103577, 0.5892255902290344, 0.5269360542297363, 0.5875421166419983, 0.6069023609161377, 0.5757575631141663, 0.6212121248245239, 0.6296296119689941, 0.6691918969154358, 0.6936026811599731, 0.7121211886405945, 0.7323232293128967, 0.7264309525489807, 0.695286214351654, 0.8383838534355164, 0.7070707082748413, 0.7373737096786499, 0.755892276763916, 0.7676767706871033, 0.7373737096786499, 0.8148148059844971, 0.8249158263206482, 0.8552188277244568, 0.9730639457702637], [0.05387205258011818, 0.06902357190847397, 0.08080808073282242, 0.11026936024427414, 0.1026936024427414, 0.12457912415266037, 0.12794612348079681, 0.1304713785648346, 0.17424242198467255, 0.18686868250370026, 0.24915824830532074, 0.2070707082748413, 0.2542087435722351, 0.2929292917251587, 0.20202019810676575, 0.23232322931289673, 0.2971380352973938, 0.25925925374031067, 0.2668350040912628, 0.30471381545066833, 0.29124578833580017, 0.28787878155708313, 0.3265993297100067, 0.2609427571296692, 0.30892255902290344, 0.29377105832099915, 0.3804713785648346, 0.3653198778629303, 0.28282827138900757, 0.31986531615257263, 0.31986531615257263, 0.34343433380126953, 0.3350168466567993, 0.34511783719062805, 0.3872053921222687, 0.33417508006095886, 0.36026936769485474, 0.3358585834503174, 0.37037035822868347, 0.35353535413742065, 0.38383838534355164, 0.3973063826560974, 0.42424243688583374, 0.4427609443664551, 0.39898988604545593, 0.4292929172515869, 0.4191919267177582, 0.46464645862579346, 0.43265992403030396, 0.4469696879386902, 0.466329962015152, NaN, 0.48821547627449036, 0.6717171669006348, 0.4983164966106415, 0.558922529220581, 0.619528591632843, 0.5740740895271301, 0.6632996797561646, 0.5707070827484131, 0.7272727489471436, 0.6026936173439026, 0.619528591632843, 0.6456229090690613, 0.6784511804580688, 0.7045454382896423, 0.6675084233283997, 0.6700336933135986, 0.680134654045105, 0.7373737096786499, 0.691919207572937, 0.8299663066864014, 0.9057239294052124, 0.872053861618042, 0.7676767706871033, 0.7828282713890076, 0.8838383555412292, 0.9915825128555298], [0.06060606241226196, 0.11279461532831192, 0.07575757801532745, 0.13636364042758942, 0.11784511804580688, 0.14478114247322083, 0.13973063230514526, 0.15319865942001343, 0.22390572726726532, 0.22053872048854828, 0.17003366351127625, 0.22053872048854828, 0.18518517911434174, 0.2457912415266037, 0.22558923065662384, 0.2609427571296692, 0.2693602740764618, 0.2861952781677246, 0.19191919267177582, 0.24663299322128296, 0.2693602740764618, 0.31313130259513855, 0.28114476799964905, 0.2777777910232544, 0.2373737394809723, 0.28535354137420654, 0.31144779920578003, 0.3737373650074005, 0.28956228494644165, 0.3097642958164215, 0.25757575035095215, 0.37037035822868347, 0.34259259700775146, 0.37037035822868347, 0.44949495792388916, 0.38383838534355164, 0.38552188873291016, 0.3787878751754761, 0.39898988604545593, 0.4276094138622284, 0.4124579131603241, 0.43939393758773804, 0.37710437178611755, 0.22558923065662384, 0.42592594027519226, 0.436026930809021, 0.4612794518470764, 0.30471381545066833, 0.4747474789619446, 0.503367006778717, 0.47979798913002014, 0.5117844939231873, NaN, 0.5319865345954895, 0.5597643256187439, 0.6565656661987305, 0.5707070827484131, 0.6649831533432007, 0.5925925970077515, 0.6102693676948547, 0.6111111044883728, 0.683501660823822, 0.6346801519393921, 0.6784511804580688, 0.6414141654968262, 0.7272727489471436, 0.691919207572937, 0.6717171669006348, 0.6936026811599731, 0.695286214351654, 0.6969696879386902, 0.6944444179534912, 0.6717171669006348, 0.7272727489471436, 0.8013467788696289, 0.7760942578315735, 0.8518518805503845, 0.9747474789619446], [0.062289562076330185, 0.07070706784725189, 0.09259258955717087, 0.10606060922145844, 0.11026936024427414, 0.12289562076330185, 0.11447811126708984, 0.12121212482452393, 0.14646464586257935, 0.1776094287633896, 0.22053872048854828, 0.20370370149612427, 0.2525252401828766, 0.27861952781677246, 0.18939393758773804, 0.20202019810676575, 0.28114476799964905, 0.24074074625968933, 0.2626262605190277, 0.28787878155708313, 0.27441078424453735, 0.2609427571296692, 0.30471381545066833, 0.24326598644256592, 0.27693602442741394, 0.2609427571296692, 0.33838382363319397, 0.3468013405799866, 0.2710437774658203, 0.29629629850387573, 0.2996633052825928, 0.2996633052825928, 0.29124578833580017, 0.3181818127632141, 0.36195287108421326, 0.29629629850387573, 0.31481480598449707, 0.32996633648872375, 0.33670035004615784, 0.3400673270225525, 0.3585858643054962, 0.3872053921222687, 0.37710437178611755, 0.402356892824173, 0.36026936769485474, 0.37037035822868347, 0.40067338943481445, 0.41750842332839966, 0.39393940567970276, 0.42424243688583374, 0.4461279511451721, 0.32828283309936523, 0.4680134654045105, NaN, 0.43939393758773804, 0.5597643256187439, 0.6077440977096558, 0.5707070827484131, 0.5555555820465088, 0.5202020406723022, 0.6338383555412292, 0.5723905563354492, 0.5892255902290344, 0.6262626051902771, 0.6750841736793518, 0.7003366947174072, 0.6397306323051453, 0.6430976390838623, 0.6565656661987305, 0.6936026811599731, 0.6717171669006348, 0.7895622849464417, 0.8282828330993652, 0.8257575631141663, 0.7575757503509521, 0.7626262903213501, 0.8888888955116272, 0.9898989796638489], [0.03872053697705269, 0.048821549862623215, 0.05050504952669144, 0.08249158412218094, 0.05218855291604996, 0.08585858345031738, 0.10606060922145844, 0.09259258955717087, 0.1734006702899933, 0.1111111119389534, 0.11616161465644836, 0.17845118045806885, 0.1515151560306549, 0.2070707082748413, 0.1835016906261444, 0.20202019810676575, 0.21885521709918976, 0.1902356892824173, 0.1346801370382309, 0.19528619945049286, 0.2525252401828766, 0.2760942876338959, 0.30471381545066833, 0.1515151560306549, 0.1708754152059555, 0.19696970283985138, 0.31313130259513855, 0.32154881954193115, 0.1582491546869278, 0.2777777910232544, 0.2306397259235382, 0.29797980189323425, 0.11784511804580688, 0.3670033812522888, 0.40572389960289, 0.1750841736793518, 0.3905723989009857, 0.3072390556335449, 0.29124578833580017, 0.3249158263206482, 0.3636363744735718, 0.3888888955116272, 0.4208754301071167, 0.3552188575267792, 0.3585858643054962, 0.38383838534355164, 0.4343434274196625, 0.37289562821388245, 0.40909090638160706, 0.4595959484577179, 0.4545454680919647, 0.5016834735870361, 0.4402357041835785, 0.560606062412262, NaN, 0.6144781112670898, 0.5875421166419983, 0.619528591632843, 0.5538720488548279, 0.5530303120613098, 0.6161616444587708, 0.6750841736793518, 0.6430976390838623, 0.6363636255264282, 0.6616161465644836, 0.7154881954193115, 0.6565656661987305, 0.6178451180458069, 0.6818181872367859, 0.7121211886405945, 0.6683501601219177, 0.7424242496490479, 0.7861952781677246, 0.7979797720909119, 0.7609427571296692, 0.7643097639083862, 0.8232323527336121, 0.9545454382896423], [0.14814814925193787, 0.15909090638160706, 0.18855218589305878, 0.17003366351127625, 0.1902356892824173, 0.18855218589305878, 0.1666666716337204, 0.19696970283985138, 0.2457912415266037, 0.23232322931289673, 0.23400673270225525, 0.2718855142593384, 0.23232322931289673, 0.2996633052825928, 0.25589224696159363, 0.2777777910232544, 0.3164983093738556, 0.24915824830532074, 0.24831649661064148, 0.24915824830532074, 0.21885521709918976, 0.3156565725803375, 0.24747474491596222, 0.25589224696159363, 0.25925925374031067, 0.3030303120613098, 0.3316498398780823, 0.30639731884002686, 0.25757575035095215, 0.2861952781677246, 0.26178452372550964, 0.3038720488548279, 0.2794612646102905, 0.36195287108421326, 0.3872053921222687, 0.2929292917251587, 0.29629629850387573, 0.2794612646102905, 0.3585858643054962, 0.30639731884002686, 0.2760942876338959, 0.3552188575267792, 0.3350168466567993, 0.30471381545066833, 0.29797980189323425, 0.3223905861377716, 0.31144779920578003, 0.3181818127632141, 0.32996633648872375, 0.3316498398780823, 0.4107744097709656, 0.44107744097709656, 0.34343433380126953, 0.4402357041835785, 0.38552188873291016, NaN, 0.4764309823513031, 0.5917508602142334, 0.46464645862579346, 0.40909090638160706, 0.4983164966106415, 0.38383838534355164, 0.44107744097709656, 0.6010100841522217, 0.5757575631141663, 0.9478114247322083, 0.631313145160675, 0.6111111044883728, 0.6329966187477112, 0.5858585834503174, 0.6262626051902771, 0.7003366947174072, 0.68855220079422, 0.6868686676025391, 0.7491582632064819, 0.6649831533432007, 0.8535353541374207, 0.9461279511451721], [0.05892255902290344, 0.07407407462596893, 0.07154881954193115, 0.07744107395410538, 0.11784511804580688, 0.09427609294652939, 0.06734006851911545, 0.09932659566402435, 0.13299663364887238, 0.21717171370983124, 0.24663299322128296, 0.20791245996952057, 0.24074074625968933, 0.25, 0.1750841736793518, 0.17003366351127625, 0.2693602740764618, 0.2752525210380554, 0.25925925374031067, 0.26430976390838623, 0.2609427571296692, 0.1734006702899933, 0.13299663364887238, 0.30471381545066833, 0.2693602740764618, 0.30892255902290344, 0.3316498398780823, 0.29124578833580017, 0.3080808222293854, 0.28956228494644165, 0.32154881954193115, 0.26767677068710327, 0.31060606241226196, 0.25757575035095215, 0.3585858643054962, 0.32828283309936523, 0.2946127951145172, 0.31144779920578003, 0.3804713785648346, 0.36868685483932495, 0.33670035004615784, 0.4124579131603241, 0.3080808222293854, 0.3636363744735718, 0.347643107175827, 0.39898988604545593, 0.35185185074806213, 0.3737373650074005, 0.40067338943481445, 0.43097642064094543, 0.47306397557258606, 0.3804713785648346, 0.4292929172515869, 0.39225590229034424, 0.4124579131603241, 0.5235690474510193, NaN, 0.5252525210380554, 0.4343434274196625, 0.47979798913002014, 0.469696968793869, 0.5185185074806213, 0.5303030014038086, 0.6127946376800537, 0.6565656661987305, 0.683501660823822, 0.6388888955116272, 0.6329966187477112, 0.680134654045105, 0.6464646458625793, 0.6683501601219177, 0.6380471587181091, 0.6599326729774475, 0.6548821330070496, 0.7457912564277649, 0.7811447978019714, 0.7676767706871033, 0.9713804721832275], [0.14814814925193787, 0.15488214790821075, 0.18686868250370026, 0.17171716690063477, 0.1818181872367859, 0.1835016906261444, 0.16582491993904114, 0.18855218589305878, 0.2390572428703308, 0.22558923065662384, 0.22727273404598236, 0.2760942876338959, 0.22558923065662384, 0.29797980189323425, 0.2525252401828766, 0.2760942876338959, 0.31313130259513855, 0.24410773813724518, 0.2457912415266037, 0.23989899456501007, 0.2138047069311142, 0.31481480598449707, 0.24242424964904785, 0.2609427571296692, 0.2542087435722351, 0.30218854546546936, 0.33670035004615784, 0.31060606241226196, 0.25589224696159363, 0.2794612646102905, 0.25925925374031067, 0.30471381545066833, 0.2794612646102905, 0.36195287108421326, 0.3905723989009857, 0.2904040515422821, 0.29124578833580017, 0.2794612646102905, 0.3569023609161377, 0.2988215386867523, 0.2693602740764618, 0.3501683473587036, 0.32828283309936523, 0.30471381545066833, 0.2920875549316406, 0.32828283309936523, 0.31313130259513855, 0.3181818127632141, 0.3164983093738556, 0.32996633648872375, 0.4124579131603241, 0.42592594027519226, 0.3350168466567993, 0.4292929172515869, 0.3804713785648346, 0.408249169588089, 0.4747474789619446, NaN, 0.4579124450683594, 0.4141414165496826, 0.4932659864425659, 0.3872053921222687, 0.4377104341983795, 0.5909090638160706, 0.567340075969696, 0.9612794518470764, 0.627946138381958, 0.6060606241226196, 0.627946138381958, 0.5824915766716003, 0.619528591632843, 0.68855220079422, 0.6818181872367859, 0.683501660823822, 0.7424242496490479, 0.6632996797561646, 0.8451178669929504, 0.9461279511451721], [0.04040404036641121, 0.07575757801532745, 0.06565656512975693, 0.0791245773434639, 0.09090909361839294, 0.09595959633588791, 0.1346801370382309, 0.11279461532831192, 0.15488214790821075, 0.11447811126708984, 0.16498316824436188, 0.16835017502307892, 0.20370370149612427, 0.25589224696159363, 0.1750841736793518, 0.17171716690063477, 0.24747474491596222, 0.18686868250370026, 0.18686868250370026, 0.22558923065662384, 0.2390572428703308, 0.24747474491596222, 0.30639731884002686, 0.18518517911434174, 0.24242424964904785, 0.2138047069311142, 0.341750830411911, 0.3232323229312897, 0.19360269606113434, 0.25589224696159363, 0.23400673270225525, 0.25925925374031067, 0.202861949801445, 0.32828283309936523, 0.347643107175827, 0.22558923065662384, 0.32828283309936523, 0.28956228494644165, 0.3164983093738556, 0.3030303120613098, 0.29629629850387573, 0.3122895658016205, 0.3484848439693451, 0.35185185074806213, 0.3350168466567993, 0.3350168466567993, 0.3585858643054962, 0.37542086839675903, 0.36026936769485474, 0.39225590229034424, 0.3930976390838623, 0.33670035004615784, 0.40740740299224854, 0.4444444477558136, 0.4461279511451721, 0.5353535413742065, 0.5656565427780151, 0.5420875549316406, NaN, 0.49663299322128296, 0.6776094436645508, 0.5084174871444702, 0.5707070827484131, 0.6094276309013367, 0.6161616444587708, 0.6708754301071167, 0.6228955984115601, 0.6212121248245239, 0.6262626051902771, 0.7053872346878052, 0.6447811722755432, 0.755892276763916, 0.804713785648346, 0.8619528412818909, 0.7331649661064148, 0.7087541818618774, 0.8619528412818909, 0.9730639457702637], [0.03872053697705269, 0.05218855291604996, 0.04545454680919647, 0.062289562076330185, 0.0555555559694767, 0.06565656512975693, 0.06565656512975693, 0.07070706784725189, 0.13973063230514526, 0.09427609294652939, 0.09090909361839294, 0.16498316824436188, 0.09848485141992569, 0.20370370149612427, 0.11195286363363266, 0.15993265807628632, 0.20370370149612427, 0.16498316824436188, 0.05218855291604996, 0.1750841736793518, 0.21632996201515198, 0.2222222238779068, 0.2180134654045105, 0.12962962687015533, 0.0833333358168602, 0.14646464586257935, 0.28114476799964905, 0.28282827138900757, 0.15656565129756927, 0.2373737394809723, 0.17676767706871033, 0.28114476799964905, 0.11279461532831192, 0.3080808222293854, 0.33838382363319397, 0.14814814925193787, 0.3265993297100067, 0.29124578833580017, 0.2794612646102905, 0.3080808222293854, 0.3501683473587036, 0.34932661056518555, 0.32996633648872375, 0.26851850748062134, 0.3400673270225525, 0.32996633648872375, 0.37710437178611755, 0.3080808222293854, 0.3653198778629303, 0.42003366351127625, 0.42424243688583374, 0.4292929172515869, 0.38973063230514526, 0.47979798913002014, 0.4469696879386902, 0.5909090638160706, 0.5202020406723022, 0.5858585834503174, 0.503367006778717, NaN, 0.5446127653121948, 0.5993266105651855, 0.6077440977096558, 0.5892255902290344, 0.624579131603241, 0.695286214351654, 0.6094276309013367, 0.6127946376800537, 0.6548821330070496, 0.6767676472663879, 0.6734007000923157, 0.7138047218322754, 0.7239057421684265, 0.7314814925193787, 0.7457912564277649, 0.7727272510528564, 0.8299663066864014, 0.9688552021980286], [0.04208754375576973, 0.05387205258011818, 0.06397306174039841, 0.0791245773434639, 0.06734006851911545, 0.0791245773434639, 0.08585858345031738, 0.0875420868396759, 0.13973063230514526, 0.08922559022903442, 0.13215488195419312, 0.15488214790821075, 0.15993265807628632, 0.22811447083950043, 0.14478114247322083, 0.16161616146564484, 0.22390572726726532, 0.17003366351127625, 0.1346801370382309, 0.20875421166419983, 0.22558923065662384, 0.21043771505355835, 0.2525252401828766, 0.1388888955116272, 0.1750841736793518, 0.17676767706871033, 0.2904040515422821, 0.280303031206131, 0.15993265807628632, 0.23569023609161377, 0.1986531913280487, 0.21885521709918976, 0.1776094287633896, 0.2777777910232544, 0.33249157667160034, 0.18686868250370026, 0.27272728085517883, 0.26767677068710327, 0.31144779920578003, 0.2777777910232544, 0.2845117747783661, 0.3013468086719513, 0.28282827138900757, 0.30639731884002686, 0.29629629850387573, 0.28956228494644165, 0.32828283309936523, 0.3316498398780823, 0.32154881954193115, 0.3569023609161377, 0.3787878751754761, 0.27272728085517883, 0.3888888955116272, 0.36616161465644836, 0.38383838534355164, 0.5016834735870361, 0.5303030014038086, 0.5067340135574341, 0.3223905861377716, 0.4553872048854828, NaN, 0.48148149251937866, 0.5117844939231873, 0.5808081030845642, 0.6026936173439026, 0.6481481194496155, 0.5976430773735046, 0.5976430773735046, 0.6178451180458069, 0.6582491397857666, 0.631313145160675, 0.7230639457702637, 0.7373737096786499, 0.7609427571296692, 0.7188552021980286, 0.691919207572937, 0.8518518805503845, 0.9747474789619446], [0.05892255902290344, 0.05387205258011818, 0.05723905563354492, 0.08249158412218094, 0.07407407462596893, 0.0875420868396759, 0.08249158412218094, 0.09595959633588791, 0.16329966485500336, 0.12121212482452393, 0.1195286214351654, 0.17676767706871033, 0.1262626200914383, 0.21212121844291687, 0.1666666716337204, 0.17676767706871033, 0.2390572428703308, 0.17003366351127625, 0.1111111119389534, 0.1515151560306549, 0.15488214790821075, 0.22727273404598236, 0.2138047069311142, 0.15740740299224854, 0.13215488195419312, 0.15656565129756927, 0.2609427571296692, 0.2760942876338959, 0.17003366351127625, 0.19360269606113434, 0.12457912415266037, 0.2542087435722351, 0.20033669471740723, 0.25925925374031067, 0.3400673270225525, 0.20875421166419983, 0.2525252401828766, 0.2053872048854828, 0.2794612646102905, 0.22474747896194458, 0.2525252401828766, 0.3030303120613098, 0.2693602740764618, 0.23569023609161377, 0.25589224696159363, 0.25336700677871704, 0.2542087435722351, 0.2710437774658203, 0.2845117747783661, 0.30639731884002686, 0.37037035822868347, 0.3973063826560974, 0.3164983093738556, 0.4276094138622284, 0.3249158263206482, 0.6161616444587708, 0.48148149251937866, 0.6127946376800537, 0.4915824830532074, 0.40067338943481445, 0.5185185074806213, NaN, 0.5370370149612427, 0.5909090638160706, 0.5875421166419983, 0.68855220079422, 0.6161616444587708, 0.558922529220581, 0.6094276309013367, 0.6060606241226196, 0.5757575631141663, 0.7020202279090881, 0.631313145160675, 0.68855220079422, 0.7171717286109924, 0.6565656661987305, 0.808080792427063, 0.936026930809021], [0.033670034259557724, 0.047138046473264694, 0.04208754375576973, 0.05892255902290344, 0.05050504952669144, 0.05892255902290344, 0.03872053697705269, 0.05892255902290344, 0.12794612348079681, 0.08585858345031738, 0.10606060922145844, 0.17171716690063477, 0.09427609294652939, 0.19191919267177582, 0.1515151560306549, 0.14478114247322083, 0.20370370149612427, 0.12289562076330185, 0.16835017502307892, 0.10942760854959488, 0.17003366351127625, 0.1750841736793518, 0.1818181872367859, 0.17003366351127625, 0.18686868250370026, 0.17003366351127625, 0.2609427571296692, 0.24242424964904785, 0.1902356892824173, 0.19528619945049286, 0.1835016906261444, 0.18686868250370026, 0.2070707082748413, 0.24915824830532074, 0.26430976390838623, 0.22727273404598236, 0.19360269606113434, 0.2390572428703308, 0.27272728085517883, 0.1262626200914383, 0.21885521709918976, 0.23989899456501007, 0.22053872048854828, 0.2996633052825928, 0.26430976390838623, 0.21717171370983124, 0.27272728085517883, 0.30639731884002686, 0.24494948983192444, 0.2180134654045105, 0.3308080732822418, 0.3804713785648346, 0.3653198778629303, 0.4107744097709656, 0.3569023609161377, 0.558922529220581, 0.469696968793869, 0.5622895359992981, 0.4292929172515869, 0.39225590229034424, 0.48821547627449036, 0.46296295523643494, NaN, 0.5909090638160706, 0.5984848737716675, 0.6851851940155029, 0.619528591632843, 0.5723905563354492, 0.6515151262283325, 0.6212121248245239, 0.6178451180458069, 0.683501660823822, 0.6632996797561646, 0.6414141654968262, 0.7340067625045776, 0.7154881954193115, 0.8181818127632141, 0.9562289714813232], [0.05723905563354492, 0.09259258955717087, 0.09932659566402435, 0.08080808073282242, 0.1026936024427414, 0.10942760854959488, 0.10774410516023636, 0.12457912415266037, 0.1750841736793518, 0.12962962687015533, 0.13383838534355164, 0.17171716690063477, 0.13299663364887238, 0.17845118045806885, 0.20875421166419983, 0.21885521709918976, 0.1986531913280487, 0.1835016906261444, 0.16329966485500336, 0.15319865942001343, 0.13973063230514526, 0.23821549117565155, 0.2390572428703308, 0.19360269606113434, 0.18686868250370026, 0.2525252401828766, 0.2289562225341797, 0.202861949801445, 0.1986531913280487, 0.2373737394809723, 0.22053872048854828, 0.18013468384742737, 0.2053872048854828, 0.30471381545066833, 0.2390572428703308, 0.21717171370983124, 0.1835016906261444, 0.18518517911434174, 0.3181818127632141, 0.23148147761821747, 0.2306397259235382, 0.2289562225341797, 0.2626262605190277, 0.2929292917251587, 0.22053872048854828, 0.3274410665035248, 0.24242424964904785, 0.3080808222293854, 0.2710437774658203, 0.3164983093738556, 0.30639731884002686, 0.3543770909309387, 0.32154881954193115, 0.3737373650074005, 0.3636363744735718, 0.39898988604545593, 0.3872053921222687, 0.40909090638160706, 0.3905723989009857, 0.4107744097709656, 0.4191919267177582, 0.40909090638160706, 0.40909090638160706, NaN, 0.49663299322128296, 0.5420875549316406, 0.691919207572937, 0.47306397557258606, 0.48148149251937866, 0.5387205481529236, 0.49494948983192444, 0.5656565427780151, 0.5538720488548279, 0.5420875549316406, 0.808080792427063, 0.6010100841522217, 0.7020202279090881, 0.9040403962135315], [0.062289562076330185, 0.07070706784725189, 0.0875420868396759, 0.08922559022903442, 0.08922559022903442, 0.09259258955717087, 0.07323232293128967, 0.1026936024427414, 0.09427609294652939, 0.1734006702899933, 0.16161616146564484, 0.21212121844291687, 0.17171716690063477, 0.1818181872367859, 0.15319865942001343, 0.1498316526412964, 0.21212121844291687, 0.19107744097709656, 0.21548821032047272, 0.1835016906261444, 0.17592592537403107, 0.16329966485500336, 0.09427609294652939, 0.26430976390838623, 0.22727273404598236, 0.19528619945049286, 0.2390572428703308, 0.18855218589305878, 0.2710437774658203, 0.20202019810676575, 0.2525252401828766, 0.18686868250370026, 0.26178452372550964, 0.15656565129756927, 0.24915824830532074, 0.2710437774658203, 0.1750841736793518, 0.22558923065662384, 0.29629629850387573, 0.2222222238779068, 0.26599326729774475, 0.21885521709918976, 0.2289562225341797, 0.25589224696159363, 0.24663299322128296, 0.283670037984848, 0.24242424964904785, 0.2845117747783661, 0.2264309823513031, 0.24410773813724518, 0.28787878155708313, 0.32154881954193115, 0.3585858643054962, 0.3249158263206482, 0.33838382363319397, 0.42424243688583374, 0.34343433380126953, 0.43265992403030396, 0.38383838534355164, 0.37542086839675903, 0.3973063826560974, 0.4124579131603241, 0.4015151560306549, 0.503367006778717, NaN, 0.5808081030845642, 0.5345118045806885, 0.5067340135574341, 0.563973069190979, 0.5336700081825256, 0.5656565427780151, 0.5656565427780151, 0.5336700081825256, 0.5707070827484131, 0.6565656661987305, 0.6363636255264282, 0.7239057421684265, 0.9208754301071167], [0.10942760854959488, 0.12457912415266037, 0.14814814925193787, 0.12289562076330185, 0.15656565129756927, 0.15656565129756927, 0.13804714381694794, 0.15656565129756927, 0.17003366351127625, 0.18686868250370026, 0.18518517911434174, 0.2180134654045105, 0.1835016906261444, 0.2525252401828766, 0.19528619945049286, 0.19360269606113434, 0.25589224696159363, 0.1986531913280487, 0.17845118045806885, 0.20033669471740723, 0.15909090638160706, 0.21212121844291687, 0.1734006702899933, 0.21043771505355835, 0.19191919267177582, 0.2138047069311142, 0.26767677068710327, 0.2457912415266037, 0.2053872048854828, 0.20875421166419983, 0.20202019810676575, 0.23232322931289673, 0.22558923065662384, 0.26599326729774475, 0.23569023609161377, 0.2289562225341797, 0.2138047069311142, 0.20370370149612427, 0.2929292917251587, 0.21212121844291687, 0.19528619945049286, 0.2373737394809723, 0.22558923065662384, 0.2289562225341797, 0.22053872048854828, 0.25084173679351807, 0.22390572726726532, 0.24074074625968933, 0.23400673270225525, 0.24074074625968933, 0.26767677068710327, 0.2954545319080353, 0.27272728085517883, 0.2996633052825928, 0.2845117747783661, 0.05218855291604996, 0.3164983093738556, 0.03872053697705269, 0.3291245698928833, 0.30471381545066833, 0.35185185074806213, 0.31144779920578003, 0.31481480598449707, 0.4579124450683594, 0.4191919267177582, NaN, 0.47138047218322754, 0.5218855142593384, 0.40909090638160706, 0.48148149251937866, 0.5404040217399597, 0.5404040217399597, 0.5269360542297363, 0.48148149251937866, 0.5976430773735046, 0.5892255902290344, 0.7053872346878052, 0.932659924030304], [0.05050504952669144, 0.08417508751153946, 0.07744107395410538, 0.07239057123661041, 0.08249158412218094, 0.09427609294652939, 0.10101009905338287, 0.10437710583209991, 0.16329966485500336, 0.12794612348079681, 0.11026936024427414, 0.1515151560306549, 0.11700336635112762, 0.15656565129756927, 0.19191919267177582, 0.20033669471740723, 0.17676767706871033, 0.17424242198467255, 0.1430976390838623, 0.14141413569450378, 0.12121212482452393, 0.21717171370983124, 0.21212121844291687, 0.17676767706871033, 0.16077440977096558, 0.21717171370983124, 0.21212121844291687, 0.19528619945049286, 0.18434342741966248, 0.2070707082748413, 0.1986531913280487, 0.16329966485500336, 0.19360269606113434, 0.2760942876338959, 0.2306397259235382, 0.19528619945049286, 0.15993265807628632, 0.1666666716337204, 0.3030303120613098, 0.19696970283985138, 0.21464645862579346, 0.20033669471740723, 0.2348484843969345, 0.27272728085517883, 0.1986531913280487, 0.2861952781677246, 0.20791245996952057, 0.29797980189323425, 0.2289562225341797, 0.2920875549316406, 0.2735690176486969, 0.33249157667160034, 0.3080808222293854, 0.36026936769485474, 0.34343433380126953, 0.36868685483932495, 0.3611111044883728, 0.372053861618042, 0.37710437178611755, 0.3905723989009857, 0.402356892824173, 0.38383838534355164, 0.3804713785648346, 0.3080808222293854, 0.4654882252216339, 0.5286195278167725, NaN, 0.45622894167900085, 0.4545454680919647, 0.5269360542297363, 0.47979798913002014, 0.5277777910232544, 0.5353535413742065, 0.5117844939231873, 0.7239057421684265, 0.5925925970077515, 0.6851851940155029, 0.8973063826560974], [0.01515151560306549, 0.05387205258011818, 0.048821549862623215, 0.048821549862623215, 0.05892255902290344, 0.05387205258011818, 0.05387205258011818, 0.06818182021379471, 0.1026936024427414, 0.06902357190847397, 0.07575757801532745, 0.13636364042758942, 0.09427609294652939, 0.17171716690063477, 0.09427609294652939, 0.11447811126708984, 0.17171716690063477, 0.11616161465644836, 0.10774410516023636, 0.09595959633588791, 0.1430976390838623, 0.13636364042758942, 0.11616161465644836, 0.1582491546869278, 0.12542086839675903, 0.14814814925193787, 0.22053872048854828, 0.2070707082748413, 0.16835017502307892, 0.17003366351127625, 0.17845118045806885, 0.18518517911434174, 0.20202019810676575, 0.2070707082748413, 0.22558923065662384, 0.22727273404598236, 0.20622895658016205, 0.20033669471740723, 0.24915824830532074, 0.19360269606113434, 0.2095959633588791, 0.2390572428703308, 0.21548821032047272, 0.25084173679351807, 0.2222222238779068, 0.2609427571296692, 0.24074074625968933, 0.2887205481529236, 0.2222222238779068, 0.2777777910232544, 0.30471381545066833, 0.32996633648872375, 0.32828283309936523, 0.3569023609161377, 0.3821548819541931, 0.3888888955116272, 0.3670033812522888, 0.39393940567970276, 0.3787878751754761, 0.3872053921222687, 0.402356892824173, 0.44107744097709656, 0.4276094138622284, 0.5269360542297363, 0.4932659864425659, 0.4781144857406616, 0.5437710285186768, NaN, 0.48821547627449036, 0.5505050420761108, 0.7129629850387573, 0.5446127653121948, 0.5909090638160706, 0.5791245698928833, 0.6127946376800537, 0.7979797720909119, 0.68855220079422, 0.9427609443664551], [0.031986530870199203, 0.04545454680919647, 0.047138046473264694, 0.05387205258011818, 0.06565656512975693, 0.06565656512975693, 0.06902357190847397, 0.07407407462596893, 0.08417508751153946, 0.06565656512975693, 0.07239057123661041, 0.1262626200914383, 0.09932659566402435, 0.16329966485500336, 0.1304713785648346, 0.10942760854959488, 0.1666666716337204, 0.0917508453130722, 0.1582491546869278, 0.13636364042758942, 0.15488214790821075, 0.12962962687015533, 0.14478114247322083, 0.1666666716337204, 0.1734006702899933, 0.13804714381694794, 0.22727273404598236, 0.21212121844291687, 0.18013468384742737, 0.16582491993904114, 0.16498316824436188, 0.1195286214351654, 0.20875421166419983, 0.17003366351127625, 0.2289562225341797, 0.21885521709918976, 0.13804714381694794, 0.1835016906261444, 0.2390572428703308, 0.15572389960289001, 0.2053872048854828, 0.048821549862623215, 0.21043771505355835, 0.2289562225341797, 0.21885521709918976, 0.19696970283985138, 0.2138047069311142, 0.2457912415266037, 0.15319865942001343, 0.2373737394809723, 0.16161616146564484, 0.31986531615257263, 0.30639731884002686, 0.34343433380126953, 0.3181818127632141, 0.3670033812522888, 0.31986531615257263, 0.372053861618042, 0.3737373650074005, 0.34511783719062805, 0.3821548819541931, 0.3905723989009857, 0.3484848439693451, 0.5185185074806213, 0.436026930809021, 0.5909090638160706, 0.5454545617103577, 0.5117844939231873, NaN, 0.5420875549316406, 0.5404040217399597, 0.5791245698928833, 0.5202020406723022, 0.5361952781677246, 0.6548821330070496, 0.6565656661987305, 0.7542087435722351, 0.9562289714813232], [0.03703703731298447, 0.03872053697705269, 0.04545454680919647, 0.062289562076330185, 0.062289562076330185, 0.062289562076330185, 0.05892255902290344, 0.06397306174039841, 0.10774410516023636, 0.07239057123661041, 0.10437710583209991, 0.08922559022903442, 0.13299663364887238, 0.11279461532831192, 0.12121212482452393, 0.09848485141992569, 0.09932659566402435, 0.12121212482452393, 0.1734006702899933, 0.17171716690063477, 0.2070707082748413, 0.13299663364887238, 0.18013468384742737, 0.1515151560306549, 0.186026930809021, 0.12121212482452393, 0.1582491546869278, 0.15656565129756927, 0.16329966485500336, 0.12962962687015533, 0.1986531913280487, 0.05892255902290344, 0.17676767706871033, 0.20370370149612427, 0.22727273404598236, 0.18855218589305878, 0.09595959633588791, 0.23569023609161377, 0.16414141654968262, 0.21548821032047272, 0.25757575035095215, 0.21548821032047272, 0.24242424964904785, 0.23400673270225525, 0.25925925374031067, 0.21043771505355835, 0.29124578833580017, 0.25757575035095215, 0.2693602740764618, 0.2760942876338959, 0.2929292917251587, 0.2626262605190277, 0.30471381545066833, 0.30639731884002686, 0.28787878155708313, 0.4141414165496826, 0.35353535413742065, 0.41750842332839966, 0.2946127951145172, 0.3232323229312897, 0.341750830411911, 0.39393940567970276, 0.3787878751754761, 0.4612794518470764, 0.466329962015152, 0.5185185074806213, 0.47306397557258606, 0.44949495792388916, 0.4579124450683594, NaN, 0.4739057123661041, 0.47306397557258606, 0.46296295523643494, 0.48148149251937866, 0.5572390556335449, 0.5791245698928833, 0.627946138381958, 0.8989899158477783], [0.018518518656492233, 0.04208754375576973, 0.04292929172515869, 0.04377104341983795, 0.062289562076330185, 0.05218855291604996, 0.047138046473264694, 0.062289562076330185, 0.08585858345031738, 0.062289562076330185, 0.07239057123661041, 0.12121212482452393, 0.07407407462596893, 0.15488214790821075, 0.08249158412218094, 0.10353535413742065, 0.1498316526412964, 0.11616161465644836, 0.0875420868396759, 0.0875420868396759, 0.12457912415266037, 0.1262626200914383, 0.11279461532831192, 0.13973063230514526, 0.10606060922145844, 0.12794612348079681, 0.2180134654045105, 0.2070707082748413, 0.14646464586257935, 0.1498316526412964, 0.1430976390838623, 0.15067340433597565, 0.17592592537403107, 0.18013468384742737, 0.2180134654045105, 0.1835016906261444, 0.16919191181659698, 0.18265993893146515, 0.24747474491596222, 0.1582491546869278, 0.20370370149612427, 0.21043771505355835, 0.18518517911434174, 0.21212121844291687, 0.2053872048854828, 0.22053872048854828, 0.22558923065662384, 0.25336700677871704, 0.18013468384742737, 0.24074074625968933, 0.2626262605190277, 0.3080808222293854, 0.3030303120613098, 0.32828283309936523, 0.3316498398780823, 0.3737373650074005, 0.3316498398780823, 0.3804713785648346, 0.3552188575267792, 0.3265993297100067, 0.36868685483932495, 0.42424243688583374, 0.3821548819541931, 0.5050504803657532, 0.4343434274196625, 0.4595959484577179, 0.5202020406723022, 0.28703704476356506, 0.4595959484577179, 0.5260942578315735, NaN, 0.5101010203361511, 0.5471380352973938, 0.5521885752677917, 0.5892255902290344, 0.7171717286109924, 0.6582491397857666, 0.9208754301071167], [0.03703703731298447, 0.05892255902290344, 0.05892255902290344, 0.0555555559694767, 0.06565656512975693, 0.08417508751153946, 0.09764309972524643, 0.0875420868396759, 0.09090909361839294, 0.09090909361839294, 0.10774410516023636, 0.15488214790821075, 0.11616161465644836, 0.1835016906261444, 0.1026936024427414, 0.1069023534655571, 0.1902356892824173, 0.13299663364887238, 0.1111111119389534, 0.14646464586257935, 0.11616161465644836, 0.11616161465644836, 0.14814814925193787, 0.11784511804580688, 0.13636364042758942, 0.15319865942001343, 0.23232322931289673, 0.2222222238779068, 0.1262626200914383, 0.15993265807628632, 0.1346801370382309, 0.11700336635112762, 0.15656565129756927, 0.16498316824436188, 0.25757575035095215, 0.15572389960289001, 0.12794612348079681, 0.1582491546869278, 0.22558923065662384, 0.19528619945049286, 0.18855218589305878, 0.19528619945049286, 0.15488214790821075, 0.2306397259235382, 0.18518517911434174, 0.21885521709918976, 0.16161616146564484, 0.25084173679351807, 0.19191919267177582, 0.24494948983192444, 0.24410773813724518, 0.17003366351127625, 0.3055555522441864, 0.21043771505355835, 0.25757575035095215, 0.2996633052825928, 0.36195287108421326, 0.31144779920578003, 0.24410773813724518, 0.2861952781677246, 0.27693602442741394, 0.29797980189323425, 0.3164983093738556, 0.4343434274196625, 0.4343434274196625, 0.4595959484577179, 0.4722222089767456, 0.4553872048854828, 0.4208754301071167, 0.5269360542297363, 0.4898989796638489, NaN, 0.47138047218322754, 0.4898989796638489, 0.5959596037864685, 0.5875421166419983, 0.869528591632843, 0.9747474789619446], [0.011784511618316174, 0.021885521709918976, 0.03872053697705269, 0.04208754375576973, 0.04040404036641121, 0.04377104341983795, 0.04208754375576973, 0.05050504952669144, 0.0513468012213707, 0.04545454680919647, 0.06565656512975693, 0.08922559022903442, 0.0791245773434639, 0.13299663364887238, 0.05218855291604996, 0.05387205258011818, 0.1111111119389534, 0.11616161465644836, 0.07070706784725189, 0.12289562076330185, 0.14646464586257935, 0.09090909361839294, 0.1430976390838623, 0.09427609294652939, 0.08249158412218094, 0.09090909361839294, 0.1734006702899933, 0.18686868250370026, 0.10606060922145844, 0.1195286214351654, 0.12121212482452393, 0.06060606241226196, 0.11447811126708984, 0.14141413569450378, 0.22053872048854828, 0.11279461532831192, 0.11447811126708984, 0.1582491546869278, 0.17676767706871033, 0.19696970283985138, 0.17003366351127625, 0.1734006702899933, 0.16414141654968262, 0.19528619945049286, 0.17171716690063477, 0.18855218589305878, 0.18518517911434174, 0.2053872048854828, 0.18686868250370026, 0.24747474491596222, 0.23232322931289673, 0.09427609294652939, 0.32828283309936523, 0.17171716690063477, 0.2138047069311142, 0.31144779920578003, 0.3400673270225525, 0.3181818127632141, 0.19528619945049286, 0.2760942876338959, 0.2626262605190277, 0.36868685483932495, 0.33670035004615784, 0.4461279511451721, 0.466329962015152, 0.47306397557258606, 0.46464645862579346, 0.40909090638160706, 0.47979798913002014, 0.5370370149612427, 0.4528619647026062, 0.5286195278167725, NaN, 0.5101010203361511, 0.5892255902290344, 0.5538720488548279, 0.695286214351654, 0.9680134654045105], [0.018518518656492233, 0.018518518656492233, 0.03703703731298447, 0.033670034259557724, 0.02861952781677246, 0.04377104341983795, 0.04208754375576973, 0.047138046473264694, 0.047138046473264694, 0.04545454680919647, 0.07239057123661041, 0.07575757801532745, 0.08585858345031738, 0.1262626200914383, 0.05218855291604996, 0.06060606241226196, 0.13720539212226868, 0.10606060922145844, 0.06060606241226196, 0.1111111119389534, 0.13804714381694794, 0.09259258955717087, 0.11279461532831192, 0.07407407462596893, 0.08417508751153946, 0.09090909361839294, 0.20622895658016205, 0.1902356892824173, 0.08249158412218094, 0.13299663364887238, 0.11279461532831192, 0.07239057123661041, 0.08249158412218094, 0.16835017502307892, 0.21548821032047272, 0.09595959633588791, 0.1346801370382309, 0.1582491546869278, 0.18518517911434174, 0.1986531913280487, 0.17171716690063477, 0.21043771505355835, 0.16329966485500336, 0.20875421166419983, 0.17845118045806885, 0.18686868250370026, 0.20033669471740723, 0.2306397259235382, 0.2053872048854828, 0.2457912415266037, 0.2626262605190277, 0.12794612348079681, 0.27272728085517883, 0.17424242198467255, 0.20202019810676575, 0.31313130259513855, 0.34511783719062805, 0.3164983093738556, 0.13804714381694794, 0.26851850748062134, 0.2390572428703308, 0.31144779920578003, 0.3585858643054962, 0.4579124450683594, 0.4292929172515869, 0.5185185074806213, 0.48821547627449036, 0.4208754301071167, 0.4638047218322754, 0.5185185074806213, 0.44781145453453064, 0.5101010203361511, 0.4898989796638489, NaN, 0.5892255902290344, 0.560606062412262, 0.7020202279090881, 0.9562289714813232], [0.033670034259557724, 0.05050504952669144, 0.05471380427479744, 0.046296294778585434, 0.05892255902290344, 0.06986531615257263, 0.05218855291604996, 0.06649831682443619, 0.07744107395410538, 0.10942760854959488, 0.08080808073282242, 0.14478114247322083, 0.08922559022903442, 0.12289562076330185, 0.0917508453130722, 0.13131313025951385, 0.15656565129756927, 0.13636364042758942, 0.0875420868396759, 0.1111111119389534, 0.06565656512975693, 0.14814814925193787, 0.11363636702299118, 0.1430976390838623, 0.10437710583209991, 0.16329966485500336, 0.18518517911434174, 0.1498316526412964, 0.1498316526412964, 0.15319865942001343, 0.12962962687015533, 0.1304713785648346, 0.1515151560306549, 0.17003366351127625, 0.18855218589305878, 0.1582491546869278, 0.11279461532831192, 0.1346801370382309, 0.25589224696159363, 0.13973063230514526, 0.15993265807628632, 0.14141413569450378, 0.13973063230514526, 0.17676767706871033, 0.15993265807628632, 0.2070707082748413, 0.1346801370382309, 0.19360269606113434, 0.15488214790821075, 0.17845118045806885, 0.18518517911434174, 0.23232322931289673, 0.1986531913280487, 0.24242424964904785, 0.2390572428703308, 0.25084173679351807, 0.2542087435722351, 0.25757575035095215, 0.2668350040912628, 0.2542087435722351, 0.28114476799964905, 0.28282827138900757, 0.26599326729774475, 0.19191919267177582, 0.34343433380126953, 0.402356892824173, 0.2760942876338959, 0.3872053921222687, 0.34511783719062805, 0.4427609443664551, 0.4107744097709656, 0.4040403962135315, 0.4107744097709656, 0.4107744097709656, NaN, 0.5176767706871033, 0.6077440977096558, 0.8552188277244568], [0.016835017129778862, 0.03030303120613098, 0.035353533923625946, 0.023569023236632347, 0.03703703731298447, 0.035353533923625946, 0.02861952781677246, 0.035353533923625946, 0.03872053697705269, 0.033670034259557724, 0.05218855291604996, 0.08585858345031738, 0.03703703731298447, 0.06902357190847397, 0.04208754375576973, 0.05892255902290344, 0.07575757801532745, 0.048821549862623215, 0.0555555559694767, 0.04545454680919647, 0.07407407462596893, 0.06060606241226196, 0.06902357190847397, 0.09259258955717087, 0.05723905563354492, 0.0875420868396759, 0.11784511804580688, 0.10437710583209991, 0.08585858345031738, 0.09932659566402435, 0.09932659566402435, 0.09090909361839294, 0.10101009905338287, 0.10774410516023636, 0.18686868250370026, 0.1026936024427414, 0.10101009905338287, 0.1026936024427414, 0.17845118045806885, 0.10101009905338287, 0.1195286214351654, 0.11447811126708984, 0.13299663364887238, 0.15993265807628632, 0.12121212482452393, 0.14646464586257935, 0.14141413569450378, 0.16750842332839966, 0.10606060922145844, 0.1582491546869278, 0.1750841736793518, 0.21717171370983124, 0.22390572726726532, 0.2373737394809723, 0.23569023609161377, 0.3350168466567993, 0.21885521709918976, 0.33670035004615784, 0.29124578833580017, 0.22727273404598236, 0.3080808222293854, 0.34343433380126953, 0.2845117747783661, 0.39898988604545593, 0.3636363744735718, 0.4107744097709656, 0.40740740299224854, 0.20202019810676575, 0.34343433380126953, 0.4208754301071167, 0.28282827138900757, 0.4124579131603241, 0.4461279511451721, 0.43939393758773804, 0.48232322931289673, NaN, 0.5690235495567322, 0.8569023609161377], [0.03703703731298447, 0.04040404036641121, 0.04545454680919647, 0.04545454680919647, 0.048821549862623215, 0.062289562076330185, 0.05387205258011818, 0.05723905563354492, 0.048821549862623215, 0.06734006851911545, 0.07239057123661041, 0.12121212482452393, 0.07575757801532745, 0.11363636702299118, 0.06060606241226196, 0.07407407462596893, 0.12289562076330185, 0.10101009905338287, 0.06986531615257263, 0.09427609294652939, 0.04377104341983795, 0.08585858345031738, 0.05723905563354492, 0.08922559022903442, 0.0791245773434639, 0.11363636702299118, 0.15656565129756927, 0.13973063230514526, 0.08249158412218094, 0.11279461532831192, 0.07744107395410538, 0.09090909361839294, 0.10353535413742065, 0.12457912415266037, 0.14814814925193787, 0.10606060922145844, 0.08585858345031738, 0.09006734192371368, 0.19528619945049286, 0.10774410516023636, 0.09090909361839294, 0.11784511804580688, 0.1026936024427414, 0.13636364042758942, 0.08922559022903442, 0.1540404111146927, 0.07407407462596893, 0.1582491546869278, 0.11279461532831192, 0.15488214790821075, 0.14478114247322083, 0.11616161465644836, 0.14814814925193787, 0.1111111119389534, 0.17676767706871033, 0.14646464586257935, 0.23232322931289673, 0.15488214790821075, 0.13804714381694794, 0.17003366351127625, 0.14814814925193787, 0.19191919267177582, 0.1818181872367859, 0.29797980189323425, 0.2760942876338959, 0.2946127951145172, 0.31481480598449707, 0.31144779920578003, 0.2457912415266037, 0.372053861618042, 0.341750830411911, 0.1304713785648346, 0.30471381545066833, 0.29797980189323425, 0.39225590229034424, 0.43097642064094543, NaN, 0.9444444179534912], [0.005050505045801401, 0.003367003286257386, 0.005050505045801401, 0.003367003286257386, 0.006734006572514772, 0.006734006572514772, 0.005050505045801401, 0.006734006572514772, 0.005050505045801401, 0.005050505045801401, 0.003367003286257386, 0.02525252476334572, 0.013468013145029545, 0.008417508564889431, 0.006734006572514772, 0.005050505045801401, 0.005050505045801401, 0.01515151560306549, 0.006734006572514772, 0.020202020183205605, 0.011784511618316174, 0.008417508564889431, 0.005050505045801401, 0.006734006572514772, 0.005050505045801401, 0.018518518656492233, 0.011784511618316174, 0.016835017129778862, 0.006734006572514772, 0.016835017129778862, 0.010101010091602802, 0.008417508564889431, 0.01515151560306549, 0.006734006572514772, 0.01515151560306549, 0.018518518656492233, 0.008417508564889431, 0.016835017129778862, 0.05050504952669144, 0.020202020183205605, 0.023569023236632347, 0.020202020183205605, 0.013468013145029545, 0.021043771877884865, 0.016835017129778862, 0.03030303120613098, 0.018518518656492233, 0.03030303120613098, 0.018518518656492233, 0.02861952781677246, 0.02693602629005909, 0.008417508564889431, 0.02525252476334572, 0.010101010091602802, 0.04545454680919647, 0.05387205258011818, 0.02861952781677246, 0.05387205258011818, 0.02693602629005909, 0.031144781038165092, 0.02525252476334572, 0.06397306174039841, 0.04377104341983795, 0.09595959633588791, 0.0791245773434639, 0.06734006851911545, 0.1026936024427414, 0.05723905563354492, 0.04377104341983795, 0.10101009905338287, 0.0791245773434639, 0.02525252476334572, 0.031986530870199203, 0.04377104341983795, 0.14478114247322083, 0.1430976390838623, 0.0555555559694767, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.90971403},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8397940424},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8396737955},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8179194558},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8022978732},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7983514821},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7871682214},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7844352629},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.706502251},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7018780896},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.694324195},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6757728818},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6738598142},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.6585552491},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6567624272},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6553631556},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6486510127},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6314880387},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6252787599},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6245791268},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6131990894},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6119419308},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5986378934},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5972604844},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5918164321},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5877060639},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5869845609},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5858039285},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5849840386},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.572707592},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5590974643},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5587039234},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5580152175},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5493134807},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5379115846},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5372010156},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.535539378},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5289693469},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5251760009},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5179282018},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5145065373},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5020879803},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4963269034},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4901722864},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4882810781},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.483536666},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4671061253},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4664064913},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4613450506},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4334032961},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4312825206},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4303751782},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4229087409},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4017993785},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3856858629},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3818378588},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.376262626},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3762298312},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3706655298},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3486269617},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3346342198},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3243255061},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3122021066},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.3022869383},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2958699547},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2772093214},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2765096861},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2764550264},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2601010094},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2526564344},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2485351352},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2414294448},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2187677636},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2177073764},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.20238095},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1784949054},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1520726742},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0282478462}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_classification/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..337ac5e875f0ab32885ef0e738ed282523d6212c
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_all/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned),TabDPT-Turbo (default),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",LimiX (default),TabM (tuned + ensembled),Mitra (default),TabICL (default),TabDPT (tuned + ensembled),TabPFNv2 (tuned + ensembled),RealMLP (tuned),LightGBM (tuned + ensembled),TabM (tuned),BetaTabPFN (default),TabDPT (tuned),TabSwift (default),CatBoost (tuned + ensembled),LightGBM (tuned),iLTM (tuned + ensembled),TabPFNv2 (tuned),TabPFNv2 (default),CatBoost (tuned),ModernNCA (tuned),CatBoost (default),xRFM (tuned + ensembled),XGBoost (tuned + ensembled),TabDPT (default),SAP-RPT-OSS (default),XGBoost (tuned),xRFM (tuned),EBM (tuned + ensembled),ModernNCA (tuned + ensembled),TorchMLP (tuned + ensembled),TabM (default),FastaiMLP (tuned + ensembled),ModernNCA (default),ChimeraBoost (tuned + ensembled),EBM (tuned),iLTM (tuned),EBM (default),ChimeraBoost (tuned),RealMLP (default),TorchMLP (tuned),FastaiMLP (tuned),ExtraTrees (tuned + ensembled),ChimeraBoost (default),ExtraTrees (tuned),XGBoost (default),TabSTAR (tuned),OrionMSP (default),TabSTAR (tuned + ensembled),RandomForest (tuned + ensembled),LightGBM (default),RandomForest (tuned),iLTM (default),TorchMLP (default),Linear (tuned + ensembled),TabFlex (default),TabSTAR (default),Linear (tuned),PerpetualBooster (tuned + ensembled),FastaiMLP (default),xRFM (default),PerpetualBooster (tuned),KNN (tuned + ensembled),ExtraTrees (default),RandomForest (default),Linear (default),PerpetualBooster (default),KNN (tuned),KNN (default)
+TabFM (default),,0.7979798,0.7592593,0.8333333,0.8181818,0.78535354,0.7996633,0.7912458,0.8518519,0.8518519,0.8636364,0.8762626,0.88383836,0.8737374,0.93602693,0.8989899,0.8821549,0.8484849,0.8939394,0.9107744,0.8796296,0.8922559,0.8821549,0.8973064,0.90909094,0.90909094,0.8855219,0.85353535,0.90909094,0.90909094,0.9057239,0.90909094,0.92424244,0.9040404,0.85858583,0.9292929,0.8956229,0.90909094,0.93602693,0.9478114,0.9158249,0.8989899,0.9191919,0.92087543,0.92087543,0.9292929,0.9158249,0.9292929,0.93097645,0.959596,0.93097645,0.94612795,0.93939394,0.93771046,0.96127945,0.8518519,0.9410775,0.8518519,0.959596,0.96127945,0.95791245,0.9410775,0.96633,0.94276094,0.93771046,0.89057237,0.94949496,0.9848485,0.96801347,0.962963,0.9814815,0.962963,0.9882155,0.9814815,0.96633,0.98316497,0.962963,0.9949495
+TabPFN-3 (default),0.2020202,,0.5555556,0.53367,0.6329966,0.6178451,0.64309764,0.63973063,0.73569024,0.72727275,0.7171717,0.73063976,0.7626263,0.75084174,0.8442761,0.7996633,0.74747473,0.7626263,0.8030303,0.8047138,0.7685185,0.8299663,0.86195284,0.8265993,0.84343433,0.8518519,0.80723906,0.7651515,0.8333333,0.85353535,0.84680134,0.84343433,0.8341751,0.84175086,0.7676768,0.85016835,0.8400673,0.84343433,0.88720536,0.88720536,0.84511787,0.84090906,0.88131315,0.8888889,0.86195284,0.90824914,0.8796296,0.88720536,0.90909094,0.92424244,0.89141417,0.93097645,0.88720536,0.9292929,0.95117843,0.84090906,0.9259259,0.84511787,0.92424244,0.9478114,0.94612795,0.94612795,0.95286196,0.9074074,0.9292929,0.87542087,0.9158249,0.94612795,0.95454544,0.96127945,0.95791245,0.9410775,0.9781145,0.9814815,0.94949496,0.969697,0.959596,0.996633
+"AutoGluon 1.5 (extreme, 4h)",0.24074075,0.44444445,,0.493266,0.64141417,0.5892256,0.5993266,0.60942763,0.6447812,0.79545456,0.7845118,0.74579126,0.79461277,0.70117843,0.81397307,0.7895623,0.70538723,0.8215488,0.8518519,0.80723906,0.7777778,0.81313133,0.8249158,0.8636364,0.8552188,0.8855219,0.76936024,0.76599324,0.85942763,0.87542087,0.8712121,0.8787879,0.87542087,0.79461277,0.7626263,0.8787879,0.8543771,0.86195284,0.9175084,0.9023569,0.86195284,0.86195284,0.8703704,0.9141414,0.8737374,0.9057239,0.8787879,0.9175084,0.88720536,0.9292929,0.8855219,0.9191919,0.92424244,0.9074074,0.94949496,0.8114478,0.9284512,0.81313133,0.93434346,0.95454544,0.93602693,0.94276094,0.95791245,0.9006734,0.9124579,0.8518519,0.9225589,0.95117843,0.95286196,0.95454544,0.9570707,0.9410775,0.96127945,0.962963,0.9452862,0.96464646,0.95454544,0.9949495
+TabICLv2 (default),0.16666667,0.46632996,0.506734,,0.5774411,0.56060606,0.58501685,0.57575756,0.7121212,0.7020202,0.7070707,0.7314815,0.73905724,0.712963,0.8442761,0.79713804,0.7255892,0.7289562,0.7845118,0.7676768,0.7676768,0.7962963,0.83838385,0.7861953,0.8181818,0.83164984,0.7584175,0.75084174,0.7929293,0.8299663,0.8215488,0.8114478,0.8030303,0.81986535,0.7323232,0.81313133,0.80050504,0.8215488,0.8400673,0.8670034,0.83164984,0.8367003,0.87794614,0.8518519,0.8484849,0.86531985,0.8703704,0.8636364,0.8855219,0.90909094,0.88720536,0.88973063,0.8636364,0.8939394,0.9175084,0.8299663,0.9225589,0.82828283,0.92087543,0.93771046,0.92087543,0.9175084,0.9410775,0.9191919,0.9107744,0.8771044,0.92760944,0.95117843,0.94612795,0.93771046,0.956229,0.9444444,0.95791245,0.96633,0.9537037,0.97643095,0.95454544,0.996633
+TabPFN-2.6 (default),0.18181819,0.36700338,0.35858586,0.42255893,,0.48653197,0.54713804,0.5126263,0.62794614,0.7037037,0.6750842,0.68013465,0.72390574,0.6952862,0.73905724,0.7020202,0.67676765,0.74915826,0.7828283,0.7744108,0.75420874,0.72727275,0.7777778,0.7895623,0.8333333,0.8299663,0.76936024,0.7281145,0.7828283,0.82828283,0.8215488,0.8030303,0.81649834,0.7373737,0.7710438,0.84511787,0.8030303,0.81986535,0.8771044,0.87205386,0.8341751,0.83164984,0.86531985,0.87542087,0.8577441,0.8821549,0.8518519,0.8821549,0.8855219,0.9158249,0.8787879,0.8973064,0.8821549,0.88973063,0.9478114,0.8097643,0.8821549,0.8181818,0.90909094,0.9444444,0.9326599,0.9259259,0.94949496,0.8973064,0.9107744,0.84343433,0.9175084,0.9410775,0.93434346,0.93771046,0.93771046,0.93434346,0.959596,0.9713805,0.9410775,0.962963,0.95117843,0.993266
+RealTabPFN-2.5 (tuned + ensembled),0.21464646,0.38215488,0.4107744,0.43939394,0.513468,,0.5824916,0.60942763,0.5774411,0.6919192,0.7121212,0.7020202,0.7070707,0.7121212,0.70286196,0.6851852,0.75757575,0.73569024,0.7710438,0.74915826,0.7744108,0.72727275,0.77272725,0.79461277,0.8021886,0.7752525,0.8097643,0.7912458,0.79461277,0.82323235,0.8047138,0.8063973,0.8047138,0.7037037,0.7685185,0.82239056,0.8047138,0.81986535,0.81649834,0.85353535,0.81313133,0.81902355,0.83164984,0.8308081,0.8442761,0.8552188,0.84511787,0.84175086,0.85690236,0.8973064,0.87542087,0.87542087,0.8552188,0.8771044,0.9141414,0.8114478,0.9057239,0.81649834,0.9040404,0.93434346,0.92087543,0.9124579,0.9410775,0.89057237,0.9074074,0.84343433,0.9057239,0.94612795,0.93434346,0.93771046,0.9478114,0.9158249,0.956229,0.956229,0.93013465,0.96464646,0.93771046,0.993266
+RealTabPFN-2.5 (default),0.2003367,0.35690236,0.4006734,0.41498315,0.45286196,0.41750842,,0.46548823,0.5521886,0.66161615,0.6750842,0.6851852,0.7104377,0.6885522,0.6919192,0.6818182,0.6936027,0.7255892,0.74242425,0.7558923,0.763468,0.71127945,0.8013468,0.7592593,0.7811448,0.73569024,0.7744108,0.7912458,0.76599324,0.7626263,0.7929293,0.8030303,0.7777778,0.73400676,0.7828283,0.8080808,0.8114478,0.8097643,0.76599324,0.84680134,0.81313133,0.8367003,0.8333333,0.8215488,0.84175086,0.8611111,0.8400673,0.83164984,0.8703704,0.8989899,0.8771044,0.87205386,0.86026937,0.8855219,0.8939394,0.8333333,0.9326599,0.8341751,0.86531985,0.93434346,0.9141414,0.9175084,0.96127945,0.8922559,0.92676765,0.86195284,0.8989899,0.94612795,0.93097645,0.9410775,0.95286196,0.9023569,0.95791245,0.95791245,0.9478114,0.9713805,0.94612795,0.9949495
+RealTabPFN-2.5 (tuned),0.20875421,0.36026937,0.3905724,0.42424244,0.48737374,0.3905724,0.5345118,,0.57239056,0.66498315,0.68686867,0.6700337,0.7020202,0.7037037,0.6784512,0.6818182,0.719697,0.72727275,0.75420874,0.75,0.7558923,0.72643095,0.7617845,0.7811448,0.8013468,0.75252527,0.790404,0.77609426,0.7777778,0.77946126,0.7912458,0.790404,0.7878788,0.7104377,0.76094276,0.8047138,0.7912458,0.81986535,0.80723906,0.84511787,0.8063973,0.8030303,0.81649834,0.81397307,0.8333333,0.84343433,0.83501685,0.83501685,0.8484849,0.8956229,0.8670034,0.8695286,0.84680134,0.8787879,0.9074074,0.8030303,0.9006734,0.8114478,0.88720536,0.9292929,0.9124579,0.9040404,0.9410775,0.87542087,0.8973064,0.84343433,0.8956229,0.9318182,0.9259259,0.93602693,0.93771046,0.9124579,0.94949496,0.95286196,0.93350166,0.96464646,0.94276094,0.993266
+TabDPT-Turbo (default),0.14814815,0.26430976,0.35521886,0.28787878,0.37205386,0.42255893,0.44781145,0.4276094,,0.54545456,0.5286195,0.5925926,0.5361953,0.55050504,0.5824916,0.62457913,0.56565654,0.61868685,0.60942763,0.5892256,0.5740741,0.68013465,0.72643095,0.65319866,0.6212121,0.6936027,0.6111111,0.6329966,0.64646465,0.65656567,0.6750842,0.67676765,0.6632997,0.80723906,0.64141417,0.67676765,0.70538723,0.6818182,0.68013465,0.7255892,0.66835016,0.75084174,0.7558923,0.73905724,0.719697,0.77356905,0.74915826,0.76599324,0.7777778,0.7912458,0.8148148,0.82575756,0.77609426,0.85353535,0.8265993,0.75420874,0.8670034,0.76094276,0.84511787,0.86026937,0.86026937,0.8367003,0.87205386,0.8249158,0.9057239,0.8299663,0.8367003,0.8973064,0.9158249,0.8922559,0.9141414,0.90909094,0.9486532,0.95286196,0.9225589,0.96127945,0.95117843,0.9949495
+RealMLP (tuned + ensembled),0.14814815,0.27272728,0.20454545,0.2979798,0.2962963,0.30808082,0.33838382,0.33501685,0.45454547,,0.4722222,0.46969697,0.55387205,0.48821548,0.513468,0.5058923,0.47895622,0.67760944,0.6734007,0.6111111,0.6329966,0.5429293,0.54882157,0.6952862,0.72306395,0.6969697,0.56397307,0.54545456,0.70538723,0.6666667,0.7121212,0.7255892,0.76936024,0.5892256,0.6262626,0.7929293,0.7138047,0.76936024,0.6851852,0.73569024,0.7777778,0.790404,0.7323232,0.75420874,0.7861953,0.7828283,0.78872055,0.7786195,0.8695286,0.7929293,0.81060606,0.81313133,0.77946126,0.82239056,0.8888889,0.7676768,0.7828283,0.7744108,0.8855219,0.9057239,0.9107744,0.8787879,0.9141414,0.8703704,0.8265993,0.81313133,0.87205386,0.93097645,0.93434346,0.92760944,0.93771046,0.90909094,0.95454544,0.95454544,0.89057237,0.96633,0.9326599,0.9949495
+"AutoGluon 1.4 (best, 4h)",0.13636364,0.28282827,0.21548821,0.2929293,0.32491583,0.28787878,0.32491583,0.3131313,0.47138047,0.5277778,,0.44444445,0.54545456,0.44276094,0.52356905,0.5252525,0.4124579,0.6144781,0.6969697,0.6043771,0.6060606,0.5521886,0.5715488,0.6919192,0.7222222,0.70538723,0.5050505,0.540404,0.7171717,0.65488213,0.7407407,0.6910775,0.74747473,0.60942763,0.5959596,0.73905724,0.7171717,0.7441077,0.6952862,0.77609426,0.7289562,0.76094276,0.73569024,0.78367,0.76094276,0.7845118,0.7929293,0.8063973,0.8030303,0.83164984,0.8114478,0.75084174,0.8299663,0.77946126,0.88383836,0.76599324,0.753367,0.77272725,0.83501685,0.90909094,0.8678451,0.8804714,0.8939394,0.86616164,0.83838385,0.8148148,0.88973063,0.92424244,0.92760944,0.8956229,0.92760944,0.8922559,0.93434346,0.92760944,0.9191919,0.9478114,0.92760944,0.996633
+LimiX (default),0.12373737,0.26936027,0.25420874,0.2685185,0.31986532,0.2979798,0.3148148,0.32996634,0.4074074,0.530303,0.5555556,,0.5673401,0.5277778,0.54545456,0.50757575,0.55050504,0.5749158,0.65488213,0.57828283,0.5824916,0.53872055,0.56565654,0.66835016,0.6700337,0.712963,0.61279464,0.6212121,0.6700337,0.66835016,0.6969697,0.6902357,0.7037037,0.57912457,0.6296296,0.7222222,0.68939394,0.7121212,0.6818182,0.6919192,0.6700337,0.6759259,0.72390574,0.75757575,0.7373737,0.7373737,0.73905724,0.76599324,0.74579126,0.7626263,0.7584175,0.7929293,0.77946126,0.7962963,0.8215488,0.7281145,0.79208755,0.72390574,0.83164984,0.83501685,0.84511787,0.82323235,0.82828283,0.82828283,0.7878788,0.78198653,0.8484849,0.8636364,0.8737374,0.9107744,0.8787879,0.84511787,0.9107744,0.92424244,0.8552188,0.9141414,0.8787879,0.9747475
+TabM (tuned + ensembled),0.116161615,0.23737374,0.2053872,0.26094276,0.2760943,0.2929293,0.28956228,0.2979798,0.46380472,0.44612795,0.45454547,0.43265992,,0.42929292,0.4949495,0.5092593,0.43265992,0.5673401,0.6144781,0.6666667,0.63973063,0.54882157,0.57239056,0.61279464,0.6557239,0.6170034,0.513468,0.5126263,0.6329966,0.6388889,0.6969697,0.6144781,0.6902357,0.5959596,0.6329966,0.72727275,0.6178451,0.74747473,0.6734007,0.7020202,0.8442761,0.7138047,0.7171717,0.74579126,0.7861953,0.7323232,0.7786195,0.7777778,0.8173401,0.79545456,0.7744108,0.74579126,0.8148148,0.74747473,0.8484849,0.7676768,0.7592593,0.7744108,0.7962963,0.9015151,0.8400673,0.8737374,0.9057239,0.8670034,0.82828283,0.81649834,0.8829966,0.9057239,0.9006734,0.8670034,0.9259259,0.88383836,0.92087543,0.9141414,0.9107744,0.962963,0.92424244,0.986532
+Mitra (default),0.12626262,0.24915825,0.29882154,0.28703704,0.30471382,0.28787878,0.3114478,0.2962963,0.44949496,0.5117845,0.55723906,0.4722222,0.5707071,,0.51683503,0.503367,0.5210438,0.5942761,0.6212121,0.6329966,0.6026936,0.52693605,0.5925926,0.6329966,0.65319866,0.6658249,0.5993266,0.6060606,0.62542087,0.6447812,0.7003367,0.6515151,0.6675084,0.6043771,0.5707071,0.6632997,0.6691919,0.6969697,0.6818182,0.70538723,0.67676765,0.6986532,0.67676765,0.73569024,0.7171717,0.7255892,0.7289562,0.7407407,0.7138047,0.76599324,0.74747473,0.7070707,0.75420874,0.7213805,0.7929293,0.7003367,0.75,0.7020202,0.7441077,0.7962963,0.7718855,0.7878788,0.8080808,0.8215488,0.8181818,0.74747473,0.84343433,0.82828283,0.8367003,0.88720536,0.84511787,0.81649834,0.8670034,0.8737374,0.8771044,0.93097645,0.8863636,0.9915825
+TabICL (default),0.06397306,0.1557239,0.18602693,0.1557239,0.26094276,0.29713804,0.30808082,0.32154882,0.41750842,0.48653197,0.47643098,0.45454547,0.5050505,0.483165,,0.5058923,0.5084175,0.56060606,0.56902355,0.56060606,0.55050504,0.5597643,0.6262626,0.57239056,0.6060606,0.6271044,0.5589225,0.57323235,0.5858586,0.62794614,0.63804716,0.63973063,0.6329966,0.6422559,0.61279464,0.65319866,0.6481481,0.6329966,0.66161615,0.65824914,0.65824914,0.68013465,0.71633,0.72053874,0.66161615,0.73063976,0.6936027,0.73905724,0.72053874,0.75252527,0.7323232,0.7979798,0.7744108,0.81060606,0.81649834,0.7441077,0.8249158,0.74747473,0.8249158,0.88804716,0.8552188,0.8333333,0.8484849,0.7912458,0.84680134,0.8047138,0.8080808,0.9057239,0.8695286,0.8787879,0.9175084,0.8973064,0.9478114,0.9478114,0.90824914,0.95791245,0.93939394,0.993266
+TabDPT (tuned + ensembled),0.1010101,0.2003367,0.21043772,0.20286195,0.2979798,0.3148148,0.3181818,0.3181818,0.37542087,0.49410775,0.47474748,0.49242425,0.49074075,0.496633,0.49410775,,0.5084175,0.5715488,0.56397307,0.5395623,0.537037,0.64983165,0.60016835,0.5808081,0.5959596,0.62794614,0.5833333,0.56565654,0.5875421,0.61279464,0.62457913,0.6355219,0.6052188,0.6784512,0.60942763,0.63804716,0.6540404,0.64057237,0.6666667,0.6792929,0.6355219,0.6952862,0.71464646,0.68686867,0.67087543,0.7289562,0.6734007,0.70538723,0.7154882,0.75420874,0.7592593,0.7676768,0.73905724,0.7979798,0.7979798,0.7222222,0.8299663,0.72390574,0.82828283,0.8400673,0.83838385,0.82323235,0.8552188,0.7811448,0.85016835,0.8063973,0.7996633,0.8855219,0.89057237,0.9015151,0.89646465,0.89309764,0.94612795,0.93939394,0.86868685,0.9410775,0.9259259,0.9949495
+TabPFNv2 (tuned + ensembled),0.11784512,0.25252524,0.2946128,0.27441078,0.32323232,0.24242425,0.30639732,0.28030303,0.43434343,0.5210438,0.5875421,0.44949496,0.5673401,0.47895622,0.49158248,0.49158248,,0.5900673,0.6136364,0.60858583,0.61195284,0.51430976,0.55723906,0.6329966,0.6329966,0.64309764,0.62457913,0.6077441,0.6388889,0.64646465,0.66835016,0.6700337,0.6691919,0.5673401,0.5496633,0.67676765,0.67424244,0.68434346,0.68013465,0.68013465,0.66835016,0.6792929,0.6557239,0.7087542,0.7003367,0.70538723,0.70538723,0.7188552,0.7104377,0.7407407,0.709596,0.70286196,0.73063976,0.7188552,0.7811448,0.68350166,0.73063976,0.68686867,0.75252527,0.7962963,0.77609426,0.76094276,0.7962963,0.8013468,0.7878788,0.7441077,0.82323235,0.82828283,0.8333333,0.9006734,0.85016835,0.8097643,0.8888889,0.86279464,0.84343433,0.92424244,0.8771044,0.9949495
+RealMLP (tuned),0.15151516,0.23737374,0.17845118,0.27104378,0.25084174,0.26430976,0.27441078,0.27272728,0.38131315,0.3223906,0.3855219,0.42508417,0.43265992,0.4057239,0.43939394,0.42845118,0.40993267,,0.5555556,0.51683503,0.5774411,0.493266,0.493266,0.56397307,0.5959596,0.5547138,0.48653197,0.506734,0.57239056,0.57239056,0.6212121,0.6010101,0.64983165,0.53872055,0.56228954,0.68434346,0.6304714,0.6750842,0.6111111,0.67676765,0.6851852,0.6969697,0.66835016,0.64309764,0.7188552,0.6936027,0.6902357,0.67171717,0.76936024,0.7558923,0.76936024,0.7407407,0.7138047,0.7592593,0.8097643,0.75084174,0.7247475,0.7558923,0.81313133,0.83501685,0.8299663,0.8299663,0.8771044,0.81649834,0.8089225,0.8013468,0.82575756,0.88383836,0.90824914,0.8787879,0.88383836,0.8670034,0.88383836,0.8939394,0.8636364,0.95117843,0.8989899,0.9848485
+LightGBM (tuned + ensembled),0.10606061,0.1969697,0.14814815,0.21548821,0.21717171,0.22895622,0.25757575,0.24579124,0.3905724,0.32659933,0.3030303,0.34511784,0.3855219,0.37878788,0.43097642,0.43602693,0.38636363,0.44444445,,0.46717173,0.46969697,0.48821548,0.52020204,0.56397307,0.60858583,0.5412458,0.45117846,0.45454547,0.5799663,0.52693605,0.6557239,0.5808081,0.6818182,0.59090906,0.5808081,0.7138047,0.63468015,0.6641414,0.56397307,0.61279464,0.67424244,0.6178451,0.65488213,0.706229,0.7138047,0.6734007,0.7710438,0.7373737,0.7087542,0.73063976,0.7154882,0.73316497,0.8080808,0.7373737,0.86531985,0.7516835,0.7407407,0.75420874,0.81313133,0.9478114,0.86531985,0.8888889,0.83164984,0.8367003,0.7845118,0.8215488,0.85690236,0.8922559,0.84175086,0.8265993,0.9124579,0.8888889,0.9292929,0.93939394,0.9124579,0.9444444,0.93013465,0.993266
+TabM (tuned),0.08922559,0.1952862,0.19276094,0.23232323,0.22558923,0.25084174,0.24410774,0.25,0.4107744,0.3888889,0.3956229,0.42171717,0.33333334,0.36700338,0.43939394,0.46043772,0.39141414,0.483165,0.5328283,,0.55387205,0.496633,0.52693605,0.56060606,0.56397307,0.56397307,0.45454547,0.47643098,0.57912457,0.5993266,0.63131315,0.5555556,0.6136364,0.5648148,0.5555556,0.63131315,0.5824916,0.64983165,0.6212121,0.6632997,0.7222222,0.6691919,0.6750842,0.6936027,0.7138047,0.6632997,0.7289562,0.7003367,0.7676768,0.7483165,0.7407407,0.6952862,0.753367,0.7121212,0.8047138,0.75084174,0.73569024,0.760101,0.7744108,0.8249158,0.7912458,0.8484849,0.89057237,0.84680134,0.81649834,0.7996633,0.85858583,0.9040404,0.8636364,0.82828283,0.9124579,0.85353535,0.8771044,0.8888889,0.8888889,0.95454544,0.9057239,0.97979796
+BetaTabPFN (default),0.12037037,0.23148148,0.22222222,0.23232323,0.24579124,0.22558923,0.23653199,0.24410774,0.42592594,0.36700338,0.3939394,0.41750842,0.36026937,0.39730638,0.44949496,0.46296296,0.38804713,0.42255893,0.530303,0.44612795,,0.51683503,0.513468,0.5218855,0.540404,0.53367,0.44276094,0.46969697,0.5496633,0.5530303,0.61279464,0.56060606,0.57912457,0.5824916,0.55387205,0.5942761,0.56902355,0.65656567,0.5521886,0.63973063,0.62794614,0.6439394,0.6481481,0.6557239,0.7003367,0.6212121,0.74242425,0.66835016,0.7104377,0.7188552,0.7255892,0.7087542,0.73063976,0.7255892,0.74747473,0.7811448,0.73905724,0.7861953,0.76094276,0.78367,0.7744108,0.84511787,0.8299663,0.86026937,0.8240741,0.84090906,0.8787879,0.85690236,0.84511787,0.7929293,0.87542087,0.88383836,0.85353535,0.86195284,0.93434346,0.9259259,0.956229,0.9882155
+TabDPT (tuned),0.107744105,0.17003366,0.18686868,0.2037037,0.27272728,0.27272728,0.28872055,0.27356902,0.31986532,0.4570707,0.44781145,0.46127945,0.45117846,0.47306398,0.4402357,0.35016835,0.48569024,0.506734,0.5117845,0.503367,0.483165,,0.540404,0.53367,0.540404,0.5555556,0.5412458,0.52609426,0.5412458,0.55387205,0.5774411,0.5858586,0.57323235,0.63468015,0.55050504,0.5824916,0.5875421,0.5942761,0.60690236,0.62457913,0.5942761,0.63973063,0.64141417,0.6481481,0.63468015,0.6515151,0.64309764,0.65824914,0.66161615,0.6936027,0.7188552,0.7121212,0.68686867,0.73905724,0.72390574,0.68434346,0.8265993,0.6851852,0.75252527,0.7777778,0.7895623,0.77272725,0.8249158,0.7617845,0.8367003,0.7878788,0.7828283,0.8636364,0.8703704,0.8670034,0.8737374,0.88383836,0.90909094,0.9074074,0.8518519,0.93939394,0.9141414,0.9915825
+TabSwift (default),0.11784512,0.13804714,0.17508417,0.16161616,0.22222222,0.22727273,0.19865319,0.23821549,0.27356902,0.45117846,0.42845118,0.43434343,0.4276094,0.4074074,0.37373737,0.39983165,0.44276094,0.506734,0.479798,0.47306398,0.48653197,0.45959595,,0.49158248,0.5016835,0.53535354,0.5050505,0.510101,0.51430976,0.537037,0.54208755,0.56228954,0.55387205,0.5673401,0.5841751,0.5841751,0.56228954,0.5858586,0.57239056,0.6195286,0.6111111,0.6936027,0.62794614,0.61279464,0.6178451,0.66161615,0.6262626,0.64309764,0.6632997,0.6919192,0.72727275,0.6734007,0.7188552,0.6952862,0.6952862,0.75252527,0.8670034,0.75757575,0.6936027,0.78198653,0.74747473,0.7861953,0.8181818,0.76094276,0.9057239,0.8265993,0.7878788,0.88383836,0.8552188,0.81986535,0.88720536,0.8518519,0.85690236,0.88720536,0.8863636,0.93097645,0.94276094,0.9949495
+CatBoost (tuned + ensembled),0.1026936,0.17340067,0.13636364,0.2138047,0.21043772,0.2053872,0.24074075,0.21885522,0.34680134,0.30471382,0.30808082,0.33164984,0.3872054,0.36700338,0.4276094,0.41919193,0.36700338,0.43602693,0.43602693,0.43939394,0.4781145,0.46632996,0.5084175,,0.48400673,0.52693605,0.43097642,0.44444445,0.57828283,0.5210438,0.5925926,0.56060606,0.5749158,0.537037,0.57239056,0.61026937,0.6077441,0.5993266,0.56565654,0.6060606,0.6212121,0.6010101,0.6178451,0.65319866,0.6439394,0.66835016,0.6885522,0.68013465,0.6885522,0.72053874,0.68350166,0.73905724,0.7222222,0.756734,0.8484849,0.7441077,0.6952862,0.73905724,0.8148148,0.8703704,0.8611111,0.8425926,0.8299663,0.8063973,0.73569024,0.7895623,0.82323235,0.84175086,0.8333333,0.8484849,0.86026937,0.8821549,0.9057239,0.9259259,0.85690236,0.9074074,0.9107744,0.993266
+LightGBM (tuned),0.09090909,0.15656565,0.14478114,0.18181819,0.16666667,0.19781145,0.21885522,0.19865319,0.37878788,0.27693602,0.2777778,0.32996634,0.3442761,0.34680134,0.3939394,0.4040404,0.36700338,0.4040404,0.39141414,0.43602693,0.45959595,0.45959595,0.4983165,0.51599324,,0.48653197,0.43602693,0.44781145,0.5252525,0.479798,0.61195284,0.5496633,0.5867003,0.55387205,0.54208755,0.62457913,0.5841751,0.6329966,0.53367,0.5824916,0.63804716,0.59175086,0.6296296,0.65488213,0.6784512,0.62457913,0.72390574,0.67676765,0.68939394,0.68602693,0.66161615,0.6910775,0.7626263,0.72306395,0.82912457,0.7407407,0.73063976,0.74579126,0.75757575,0.9166667,0.8249158,0.8678451,0.81313133,0.81313133,0.77272725,0.8080808,0.8392256,0.87457913,0.8265993,0.81397307,0.8939394,0.8636364,0.9175084,0.9158249,0.8956229,0.94276094,0.92087543,0.9949495
+iLTM (tuned + ensembled),0.09090909,0.14814815,0.11447811,0.16835018,0.17003366,0.22474748,0.26430976,0.24747474,0.30639732,0.3030303,0.2946128,0.28703704,0.38299662,0.33417508,0.37289563,0.37205386,0.35690236,0.44528618,0.4587542,0.43602693,0.46632996,0.44444445,0.46464646,0.47306398,0.513468,,0.46801347,0.4814815,0.503367,0.54208755,0.540404,0.57828283,0.5429293,0.530303,0.54713804,0.5976431,0.62542087,0.6010101,0.5707071,0.5976431,0.56565654,0.64309764,0.65656567,0.65656567,0.64141417,0.706229,0.67171717,0.6666667,0.7070707,0.7037037,0.71464646,0.706229,0.71464646,0.73905724,0.8030303,0.6969697,0.6910775,0.6978114,0.7861953,0.85353535,0.82323235,0.84343433,0.8299663,0.74747473,0.8047138,0.7861953,0.7828283,0.8518519,0.86195284,0.8787879,0.87205386,0.84680134,0.90909094,0.90909094,0.8367003,0.9124579,0.8863636,0.9814815
+TabPFNv2 (tuned),0.11447811,0.19276094,0.23063973,0.2415825,0.23063973,0.19023569,0.22558923,0.20959596,0.3888889,0.43602693,0.4949495,0.3872054,0.48653197,0.4006734,0.44107744,0.41666666,0.37542087,0.513468,0.54882157,0.54545456,0.55723906,0.4587542,0.4949495,0.56902355,0.56397307,0.53198653,,0.5328283,0.5976431,0.56565654,0.6077441,0.5707071,0.59175086,0.52020204,0.5210438,0.59090906,0.5808081,0.6144781,0.59090906,0.6195286,0.6178451,0.6203704,0.59680134,0.64983165,0.6363636,0.6422559,0.6481481,0.66498315,0.6658249,0.68013465,0.65824914,0.6195286,0.6885522,0.66161615,0.68686867,0.66835016,0.66835016,0.6632997,0.65824914,0.7188552,0.709596,0.73905724,0.73905724,0.7710438,0.76094276,0.7323232,0.7878788,0.77946126,0.77272725,0.84175086,0.78198653,0.7676768,0.8265993,0.793771,0.8148148,0.8821549,0.84343433,0.9882155
+TabPFNv2 (default),0.14646465,0.23484848,0.23400673,0.24915825,0.2718855,0.20875421,0.20875421,0.22390573,0.36700338,0.45454547,0.45959595,0.37878788,0.48737374,0.3939394,0.42676768,0.43434343,0.3922559,0.493266,0.54545456,0.52356905,0.530303,0.4739057,0.48989898,0.5555556,0.5521886,0.5185185,0.46717173,,0.5707071,0.52693605,0.59175086,0.5841751,0.5740741,0.5218855,0.5016835,0.5824916,0.6111111,0.61616164,0.543771,0.5841751,0.6195286,0.6262626,0.5833333,0.60942763,0.6363636,0.5925926,0.65319866,0.6212121,0.6666667,0.64309764,0.6700337,0.63468015,0.6262626,0.65319866,0.6784512,0.6936027,0.7087542,0.68939394,0.67676765,0.7171717,0.719697,0.72390574,0.75757575,0.79713804,0.8114478,0.75420874,0.8047138,0.7929293,0.7878788,0.84343433,0.7929293,0.7777778,0.81313133,0.8097643,0.85016835,0.8956229,0.86026937,0.98316497
+CatBoost (tuned),0.09090909,0.16666667,0.14057238,0.20707071,0.21717171,0.2053872,0.23400673,0.22222222,0.35353535,0.2946128,0.28282827,0.32996634,0.36700338,0.37457913,0.41414142,0.4124579,0.3611111,0.4276094,0.42003366,0.42087543,0.4503367,0.4587542,0.48569024,0.42171717,0.47474748,0.496633,0.4023569,0.42929292,,0.5,0.57828283,0.5462963,0.57575756,0.52693605,0.55723906,0.6077441,0.5942761,0.6026936,0.5589225,0.6043771,0.61026937,0.57575756,0.5993266,0.61868685,0.63131315,0.63468015,0.6919192,0.65909094,0.67676765,0.70791245,0.67424244,0.7171717,0.7104377,0.7289562,0.84175086,0.74242425,0.6919192,0.7441077,0.8063973,0.84343433,0.8400673,0.8299663,0.8097643,0.8013468,0.7289562,0.79461277,0.81565654,0.83164984,0.81986535,0.8367003,0.85353535,0.8737374,0.8939394,0.9175084,0.85016835,0.9141414,0.9175084,0.993266
+ModernNCA (tuned),0.09090909,0.14646465,0.124579124,0.17003366,0.17171717,0.17676768,0.23737374,0.22053872,0.34343433,0.33333334,0.34511784,0.33164984,0.3611111,0.35521886,0.37205386,0.3872054,0.35353535,0.4276094,0.47306398,0.4006734,0.4469697,0.44612795,0.46296296,0.47895622,0.52020204,0.45791245,0.43434343,0.47306398,0.5,,0.537037,0.5496633,0.537037,0.51683503,0.530303,0.54461277,0.6010101,0.5740741,0.50084174,0.57575756,0.55050504,0.59511787,0.6750842,0.6212121,0.6195286,0.6144781,0.6363636,0.6287879,0.68013465,0.7154882,0.6936027,0.68013465,0.6902357,0.7037037,0.7222222,0.7138047,0.7104377,0.72053874,0.7441077,0.7626263,0.76430976,0.8063973,0.8047138,0.7626263,0.7979798,0.7912458,0.7929293,0.8299663,0.8341751,0.8703704,0.85016835,0.8400673,0.8804714,0.8670034,0.84680134,0.9006734,0.88720536,0.98316497
+CatBoost (default),0.09427609,0.15319866,0.12878788,0.17845118,0.17845118,0.1952862,0.20707071,0.20875421,0.32491583,0.28787878,0.25925925,0.3030303,0.3030303,0.2996633,0.36195287,0.37542087,0.33164984,0.37878788,0.3442761,0.36868685,0.3872054,0.42255893,0.45791245,0.4074074,0.38804713,0.45959595,0.3922559,0.40824917,0.42171717,0.46296296,,0.49579126,0.47727272,0.5286195,0.54545456,0.51430976,0.56228954,0.56397307,0.5117845,0.57239056,0.56565654,0.5707071,0.5673401,0.62542087,0.62457913,0.6178451,0.66498315,0.66835016,0.6481481,0.6885522,0.65824914,0.68013465,0.74242425,0.7003367,0.76936024,0.7382155,0.6784512,0.7407407,0.76599324,0.82323235,0.8013468,0.87542087,0.81649834,0.77946126,0.74747473,0.7979798,0.8013468,0.8215488,0.83501685,0.8013468,0.85690236,0.86531985,0.8787879,0.88720536,0.8703704,0.9006734,0.9225589,0.989899
+xRFM (tuned + ensembled),0.09090909,0.15656565,0.121212125,0.18855219,0.1969697,0.1936027,0.1969697,0.20959596,0.32323232,0.27441078,0.30892256,0.3097643,0.3855219,0.34848484,0.36026937,0.3644781,0.32996634,0.3989899,0.41919193,0.44444445,0.43939394,0.41414142,0.43771043,0.43939394,0.4503367,0.42171717,0.42929292,0.41582492,0.4537037,0.4503367,0.50420874,,0.48232323,0.48989898,0.48821548,0.5050505,0.60858583,0.54713804,0.5117845,0.52356905,0.54545456,0.5833333,0.5555556,0.5942761,0.6144781,0.56649834,0.6363636,0.6043771,0.6262626,0.63131315,0.6632997,0.65656567,0.6296296,0.7003367,0.7020202,0.69612795,0.7323232,0.6952862,0.7407407,0.7188552,0.7811448,0.74579126,0.81313133,0.81986535,0.81313133,0.7676768,0.8367003,0.8148148,0.8804714,0.9410775,0.8493266,0.8829966,0.93939394,0.92760944,0.8695286,0.90909094,0.90909094,0.9915825
+XGBoost (tuned + ensembled),0.07575758,0.16582492,0.124579124,0.1969697,0.18350169,0.1952862,0.22222222,0.21212122,0.33670035,0.23063973,0.25252524,0.2962963,0.3097643,0.33249158,0.36700338,0.39478114,0.33080807,0.35016835,0.3181818,0.38636363,0.42087543,0.42676768,0.44612795,0.42508417,0.41329965,0.4570707,0.40824917,0.42592594,0.42424244,0.46296296,0.52272725,0.5176768,,0.5,0.5252525,0.59175086,0.57575756,0.56060606,0.503367,0.57828283,0.5774411,0.56397307,0.5976431,0.5740741,0.62457913,0.6010101,0.66161615,0.6144781,0.63973063,0.65319866,0.6228956,0.66498315,0.6574074,0.7087542,0.8821549,0.72053874,0.68939394,0.72053874,0.79713804,0.88720536,0.82239056,0.7996633,0.7929293,0.79461277,0.7382155,0.7744108,0.8063973,0.7979798,0.7912458,0.82323235,0.8240741,0.84343433,0.8855219,0.9175084,0.8484849,0.8989899,0.89646465,0.9848485
+TabDPT (default),0.0959596,0.15824915,0.2053872,0.18013468,0.26262626,0.2962963,0.26599327,0.28956228,0.19276094,0.4107744,0.3905724,0.42087543,0.4040404,0.3956229,0.3577441,0.32154882,0.43265992,0.46127945,0.4090909,0.4351852,0.41750842,0.36531988,0.43265992,0.46296296,0.44612795,0.46969697,0.479798,0.4781145,0.47306398,0.483165,0.47138047,0.510101,0.5,,0.5151515,0.5185185,0.51683503,0.4983165,0.5151515,0.53367,0.543771,0.55387205,0.5707071,0.5774411,0.5328283,0.55723906,0.56228954,0.5875421,0.5513468,0.6026936,0.6363636,0.65488213,0.6296296,0.6818182,0.6329966,0.63804716,0.74242425,0.63804716,0.67171717,0.6919192,0.7222222,0.7407407,0.75084174,0.6952862,0.84343433,0.73400676,0.72390574,0.7929293,0.8299663,0.7962963,0.81986535,0.83501685,0.85858583,0.83164984,0.8299663,0.8922559,0.87542087,0.993266
+SAP-RPT-OSS (default),0.14141414,0.23232323,0.23737374,0.26767677,0.22895622,0.23148148,0.21717171,0.23905724,0.35858586,0.37373737,0.4040404,0.37037036,0.36700338,0.42929292,0.3872054,0.3905724,0.4503367,0.43771043,0.41919193,0.44444445,0.44612795,0.44949496,0.41582492,0.4276094,0.45791245,0.45286196,0.47895622,0.4983165,0.44276094,0.46969697,0.45454547,0.5117845,0.47474748,0.4848485,,0.483165,0.5,0.493266,0.5084175,0.5808081,0.53367,0.59090906,0.54208755,0.483165,0.52693605,0.52272725,0.5597643,0.5117845,0.63468015,0.63973063,0.66498315,0.61279464,0.55050504,0.63804716,0.5942761,0.61279464,0.64141417,0.60942763,0.6523569,0.66161615,0.6675084,0.6599327,0.73569024,0.76094276,0.75084174,0.76430976,0.76936024,0.7744108,0.7710438,0.77272725,0.78198653,0.74242425,0.77946126,0.7845118,0.8114478,0.81313133,0.8518519,0.9848485
+XGBoost (tuned),0.07070707,0.14983165,0.121212125,0.18686868,0.15488215,0.17760943,0.1919192,0.1952862,0.32323232,0.20707071,0.26094276,0.2777778,0.27272728,0.33670035,0.34680134,0.36195287,0.32323232,0.31565657,0.28619528,0.36868685,0.4057239,0.41750842,0.41582492,0.38973063,0.37542087,0.4023569,0.4090909,0.41750842,0.3922559,0.4553872,0.48569024,0.4949495,0.40824917,0.4814815,0.51683503,,0.5412458,0.540404,0.4983165,0.54545456,0.54882157,0.53367,0.56565654,0.5277778,0.5959596,0.5774411,0.6296296,0.5774411,0.61279464,0.6296296,0.6212121,0.6658249,0.61616164,0.7037037,0.8249158,0.7070707,0.67171717,0.709596,0.7744108,0.8518519,0.81313133,0.7912458,0.77272725,0.7828283,0.7289562,0.7710438,0.8047138,0.77272725,0.7811448,0.8114478,0.81649834,0.8442761,0.88720536,0.9040404,0.84175086,0.8973064,0.8939394,0.9814815
+xRFM (tuned),0.104377106,0.15993266,0.1456229,0.19949494,0.1969697,0.1952862,0.18855219,0.20875421,0.2946128,0.28619528,0.28282827,0.31060606,0.38215488,0.33080807,0.35185185,0.3459596,0.32575756,0.36952862,0.36531988,0.41750842,0.43097642,0.4124579,0.43771043,0.3922559,0.41582492,0.37457913,0.41919193,0.3888889,0.4057239,0.3989899,0.43771043,0.39141414,0.42424244,0.483165,0.5,0.4587542,,0.5058923,0.42592594,0.506734,0.52356905,0.5496633,0.540404,0.5580808,0.5740741,0.540404,0.5976431,0.5808081,0.5808081,0.61279464,0.63973063,0.63973063,0.6144781,0.6851852,0.60942763,0.7037037,0.70538723,0.7087542,0.67171717,0.6734007,0.72727275,0.74747473,0.8063973,0.81649834,0.8249158,0.7861953,0.8400673,0.793771,0.86195284,0.9040404,0.8308081,0.87205386,0.8855219,0.86531985,0.88720536,0.8989899,0.9141414,0.9915825
+EBM (tuned + ensembled),0.09090909,0.15656565,0.13804714,0.17845118,0.18013468,0.18013468,0.19023569,0.18013468,0.3181818,0.23063973,0.25589225,0.28787878,0.25252524,0.3030303,0.36700338,0.3594276,0.31565657,0.32491583,0.33585858,0.35016835,0.34343433,0.4057239,0.41414142,0.4006734,0.36700338,0.3989899,0.3855219,0.3838384,0.39730638,0.42592594,0.43602693,0.45286196,0.43939394,0.5016835,0.506734,0.45959595,0.49410775,,0.44444445,0.53198653,0.530303,0.5176768,0.543771,0.5185185,0.68939394,0.537037,0.73400676,0.55050504,0.60942763,0.6388889,0.59343433,0.6641414,0.6212121,0.6700337,0.69276094,0.72053874,0.6885522,0.72053874,0.7104377,0.7087542,0.7323232,0.79461277,0.76094276,0.8148148,0.7744108,0.7962963,0.8333333,0.7996633,0.81649834,0.76430976,0.8173401,0.84175086,0.84175086,0.84175086,0.86531985,0.8973064,0.9099327,0.98316497
+ModernNCA (tuned + ensembled),0.06397306,0.112794615,0.082491584,0.15993266,0.12289562,0.18350169,0.23400673,0.19276094,0.31986532,0.3148148,0.30471382,0.3181818,0.32659933,0.3181818,0.33838382,0.33333334,0.31986532,0.3888889,0.43602693,0.37878788,0.44781145,0.39309764,0.4276094,0.43434343,0.46632996,0.42929292,0.4090909,0.45622894,0.44107744,0.49915826,0.48821548,0.48821548,0.496633,0.4848485,0.49158248,0.5016835,0.5740741,0.5555556,,0.5151515,0.51683503,0.54882157,0.6481481,0.55387205,0.59511787,0.56565654,0.6026936,0.5858586,0.6515151,0.62457913,0.6262626,0.6296296,0.6010101,0.6632997,0.7087542,0.64141417,0.6195286,0.64309764,0.68350166,0.72053874,0.6885522,0.72053874,0.72727275,0.6818182,0.7037037,0.7070707,0.6969697,0.75084174,0.76094276,0.8358586,0.75252527,0.7744108,0.82323235,0.8148148,0.7441077,0.8215488,0.8047138,0.94949496
+TorchMLP (tuned + ensembled),0.052188553,0.112794615,0.0976431,0.13299663,0.12794612,0.14646465,0.15319866,0.15488215,0.27441078,0.26430976,0.22390573,0.30808082,0.2979798,0.2946128,0.34175083,0.32070708,0.31986532,0.32323232,0.3872054,0.33670035,0.36026937,0.37542087,0.38047138,0.3939394,0.41750842,0.4023569,0.38047138,0.41582492,0.3956229,0.42424244,0.4276094,0.47643098,0.42171717,0.46632996,0.41919193,0.45454547,0.493266,0.46801347,0.4848485,,0.46296296,0.5774411,0.55387205,0.5151515,0.530303,0.5547138,0.56397307,0.5521886,0.5993266,0.7592593,0.68771046,0.64646465,0.57239056,0.6599327,0.6750842,0.6936027,0.63131315,0.70117843,0.6969697,0.6919192,0.7222222,0.7752525,0.8737374,0.7685185,0.7777778,0.7878788,0.8030303,0.8063973,0.8442761,0.7845118,0.84175086,0.8047138,0.8030303,0.8013468,0.86026937,0.8989899,0.8922559,0.97979796
+TabM (default),0.08417509,0.15488215,0.13804714,0.16835018,0.16582492,0.18686868,0.18686868,0.1936027,0.33164984,0.22222222,0.27104378,0.32996634,0.1557239,0.32323232,0.34175083,0.3644781,0.33164984,0.3148148,0.32575756,0.2777778,0.37205386,0.4057239,0.3888889,0.37878788,0.36195287,0.43434343,0.38215488,0.38047138,0.38973063,0.44949496,0.43434343,0.45454547,0.42255893,0.45622894,0.46632996,0.45117846,0.47643098,0.46969697,0.483165,0.537037,,0.56902355,0.54882157,0.5,0.543771,0.5361953,0.5513468,0.52356905,0.6111111,0.62794614,0.6515151,0.61616164,0.5875421,0.64141417,0.6363636,0.72390574,0.6632997,0.73063976,0.7037037,0.64983165,0.7154882,0.74747473,0.7811448,0.76936024,0.73400676,0.8047138,0.78535354,0.790404,0.79461277,0.74242425,0.7962963,0.8114478,0.8299663,0.82828283,0.8400673,0.8804714,0.90909094,0.97643095
+FastaiMLP (tuned + ensembled),0.1010101,0.1590909,0.13804714,0.16329966,0.16835018,0.18097644,0.16329966,0.1969697,0.24915825,0.20959596,0.23905724,0.32407406,0.28619528,0.3013468,0.31986532,0.30471382,0.32070708,0.3030303,0.38215488,0.33080807,0.3560606,0.36026937,0.30639732,0.3989899,0.40824917,0.35690236,0.37962964,0.37373737,0.42424244,0.40488216,0.42929292,0.41666666,0.43602693,0.44612795,0.4090909,0.46632996,0.4503367,0.48232323,0.45117846,0.42255893,0.43097642,,0.537037,0.53535354,0.53198653,0.47474748,0.543771,0.55723906,0.56649834,0.56228954,0.7087542,0.6026936,0.56060606,0.61279464,0.6111111,0.6447812,0.5875421,0.64983165,0.68771046,0.6506734,0.6986532,0.6969697,0.760101,0.7710438,0.7811448,0.7626263,0.7996633,0.76094276,0.95117843,0.7845118,0.7895623,0.8047138,0.8265993,0.7895623,0.85858583,0.8855219,0.8821549,0.97979796
+ModernNCA (default),0.08080808,0.11868687,0.12962963,0.12205387,0.13468014,0.16835018,0.16666667,0.18350169,0.24410774,0.26767677,0.26430976,0.2760943,0.28282827,0.32323232,0.28367004,0.28535354,0.3442761,0.33164984,0.34511784,0.32491583,0.35185185,0.35858586,0.37205386,0.38215488,0.37037036,0.34343433,0.40319866,0.41666666,0.4006734,0.32491583,0.43265992,0.44444445,0.4023569,0.42929292,0.45791245,0.43434343,0.45959595,0.45622894,0.35185185,0.44612795,0.45117846,0.46296296,,0.5707071,0.5117845,0.46717173,0.55050504,0.5858586,0.513468,0.54208755,0.543771,0.57575756,0.6228956,0.6228956,0.57912457,0.66498315,0.6919192,0.67171717,0.6515151,0.6700337,0.7171717,0.73063976,0.77946126,0.7373737,0.7710438,0.7744108,0.7651515,0.7845118,0.7895623,0.75757575,0.8148148,0.84511787,0.8358586,0.8367003,0.86026937,0.8670034,0.8973064,0.986532
+ChimeraBoost (tuned + ensembled),0.07912458,0.11111111,0.08585858,0.14814815,0.124579124,0.16919191,0.17845118,0.18602693,0.26094276,0.24579124,0.21632996,0.24242425,0.25420874,0.26430976,0.27946126,0.3131313,0.2912458,0.35690236,0.29377106,0.30639732,0.3442761,0.35185185,0.3872054,0.34680134,0.34511784,0.34343433,0.35016835,0.3905724,0.38131315,0.37878788,0.37457913,0.4057239,0.42592594,0.42255893,0.51683503,0.4722222,0.44191918,0.4814815,0.44612795,0.4848485,0.5,0.46464646,0.42929292,,0.52020204,0.5185185,0.537037,0.59511787,0.53872055,0.57912457,0.5277778,0.55723906,0.7744108,0.5976431,0.6447812,0.6952862,0.6363636,0.6952862,0.6481481,0.7314815,0.6936027,0.76430976,0.7003367,0.7070707,0.7441077,0.7710438,0.72727275,0.74915826,0.7710438,0.76599324,0.7878788,0.76936024,0.8047138,0.7912458,0.82323235,0.8400673,0.8636364,0.9789562
+EBM (tuned),0.07912458,0.13804714,0.12626262,0.15151516,0.14225589,0.1557239,0.15824915,0.16666667,0.28030303,0.2138047,0.23905724,0.26262626,0.2138047,0.28282827,0.33838382,0.32912457,0.2996633,0.28114477,0.28619528,0.28619528,0.2996633,0.36531988,0.38215488,0.3560606,0.32154882,0.35858586,0.36363637,0.36363637,0.36868685,0.38047138,0.37542087,0.3855219,0.37542087,0.46717173,0.47306398,0.4040404,0.42592594,0.31060606,0.40488216,0.46969697,0.45622894,0.46801347,0.48821548,0.479798,,0.4814815,0.6422559,0.5058923,0.5286195,0.5858586,0.56649834,0.6010101,0.5740741,0.63973063,0.64141417,0.7020202,0.6523569,0.70791245,0.66498315,0.6599327,0.7037037,0.7441077,0.73569024,0.77946126,0.753367,0.77946126,0.8013468,0.7777778,0.7811448,0.7407407,0.79461277,0.8148148,0.82828283,0.8215488,0.8400673,0.8787879,0.9107744,0.98316497
+iLTM (tuned),0.07070707,0.091750845,0.09427609,0.13468014,0.11784512,0.14478114,0.1388889,0.15656565,0.22643098,0.21717171,0.21548821,0.26262626,0.26767677,0.27441078,0.26936027,0.27104378,0.2946128,0.30639732,0.32659933,0.33670035,0.37878788,0.34848484,0.33838382,0.33164984,0.37542087,0.29377106,0.3577441,0.4074074,0.36531988,0.3855219,0.38215488,0.4335017,0.3989899,0.44276094,0.47727272,0.42255893,0.45959595,0.46296296,0.43434343,0.44528618,0.46380472,0.5252525,0.5328283,0.4814815,0.5185185,,0.5589225,0.5328283,0.56228954,0.5707071,0.5892256,0.5707071,0.56397307,0.6296296,0.61616164,0.67760944,0.6010101,0.67171717,0.66498315,0.6700337,0.7104377,0.746633,0.7828283,0.6725589,0.71633,0.74915826,0.7138047,0.73905724,0.8030303,0.7895623,0.77946126,0.7811448,0.8114478,0.81313133,0.7929293,0.85353535,0.8459596,0.969697
+EBM (default),0.08417509,0.12037037,0.121212125,0.12962963,0.14814815,0.15488215,0.15993266,0.16498317,0.25084174,0.21127947,0.20707071,0.26094276,0.22138047,0.27104378,0.30639732,0.32659933,0.2946128,0.3097643,0.22895622,0.27104378,0.25757575,0.35690236,0.37373737,0.3114478,0.2760943,0.32828283,0.35185185,0.34680134,0.30808082,0.36363637,0.33501685,0.36363637,0.33838382,0.43771043,0.4402357,0.37037036,0.4023569,0.26599327,0.39730638,0.43602693,0.4486532,0.45622894,0.44949496,0.46296296,0.3577441,0.44107744,,0.48821548,0.5084175,0.5841751,0.5521886,0.5808081,0.53872055,0.5993266,0.56565654,0.6885522,0.6481481,0.68686867,0.64141417,0.6228956,0.67171717,0.74579126,0.72727275,0.75757575,0.75757575,0.77609426,0.79208755,0.7592593,0.7861953,0.7087542,0.7744108,0.83838385,0.8148148,0.7996633,0.86531985,0.85858583,0.9259259,0.9814815
+ChimeraBoost (tuned),0.07070707,0.112794615,0.082491584,0.13636364,0.11784512,0.15824915,0.16835018,0.16498317,0.23400673,0.22138047,0.1936027,0.23400673,0.22222222,0.25925925,0.26094276,0.2946128,0.28114477,0.32828283,0.26262626,0.2996633,0.33164984,0.34175083,0.35690236,0.31986532,0.32323232,0.33333334,0.33501685,0.37878788,0.3409091,0.37121212,0.33164984,0.3956229,0.3855219,0.4124579,0.48821548,0.42255893,0.41919193,0.44949496,0.41414142,0.44781145,0.47643098,0.44276094,0.41414142,0.40488216,0.49410775,0.46717173,0.5117845,,0.5218855,0.54882157,0.50420874,0.53535354,0.6952862,0.5824916,0.6271044,0.6818182,0.6262626,0.6818182,0.62457913,0.6919192,0.66835016,0.7289562,0.6936027,0.6919192,0.7154882,0.7592593,0.7020202,0.71127945,0.75420874,0.74242425,0.746633,0.74915826,0.79461277,0.76936024,0.8063973,0.8324916,0.84175086,0.969697
+RealMLP (default),0.06902357,0.09090909,0.112794615,0.11447811,0.11447811,0.14309764,0.12962963,0.15151516,0.22222222,0.13047138,0.1969697,0.25420874,0.18265994,0.28619528,0.27946126,0.28451177,0.28956228,0.23063973,0.2912458,0.23232323,0.28956228,0.33838382,0.33670035,0.3114478,0.31060606,0.2929293,0.33417508,0.33333334,0.32323232,0.31986532,0.35185185,0.37373737,0.36026937,0.4486532,0.36531988,0.3872054,0.41919193,0.3905724,0.34848484,0.4006734,0.3888889,0.4335017,0.48653197,0.46127945,0.47138047,0.43771043,0.49158248,0.4781145,,0.5589225,0.56902355,0.5673401,0.5252525,0.6060606,0.59090906,0.6700337,0.5993266,0.68350166,0.63973063,0.63468015,0.6784512,0.7154882,0.7550505,0.7289562,0.77356905,0.76599324,0.7710438,0.7777778,0.84680134,0.73063976,0.81986535,0.8080808,0.81313133,0.79461277,0.84511787,0.8939394,0.88720536,0.9814815
+TorchMLP (tuned),0.04040404,0.07575758,0.07070707,0.09090909,0.08417509,0.1026936,0.1010101,0.104377106,0.20875421,0.20707071,0.16835018,0.23737374,0.20454545,0.23400673,0.24747474,0.24579124,0.25925925,0.24410774,0.26936027,0.2516835,0.28114477,0.30639732,0.30808082,0.27946126,0.31397307,0.2962963,0.31986532,0.35690236,0.29208755,0.28451177,0.3114478,0.36868685,0.34680134,0.39730638,0.36026937,0.37037036,0.3872054,0.3611111,0.37542087,0.24074075,0.37205386,0.43771043,0.45791245,0.42087543,0.41414142,0.42929292,0.41582492,0.45117846,0.44107744,,0.530303,0.5530303,0.496633,0.57575756,0.540404,0.66835016,0.56902355,0.6700337,0.6077441,0.5799663,0.64309764,0.6936027,0.78198653,0.68350166,0.7558923,0.7592593,0.70791245,0.7222222,0.7626263,0.72390574,0.7592593,0.7550505,0.75252527,0.75420874,0.8215488,0.84175086,0.84511787,0.9713805
+FastaiMLP (tuned),0.06902357,0.10858586,0.11447811,0.112794615,0.121212125,0.124579124,0.12289562,0.13299663,0.18518518,0.18939394,0.18855219,0.2415825,0.22558923,0.25252524,0.26767677,0.24074075,0.29040405,0.23063973,0.28451177,0.25925925,0.27441078,0.28114477,0.27272728,0.3164983,0.33838382,0.28535354,0.34175083,0.32996634,0.32575756,0.30639732,0.34175083,0.33670035,0.37710437,0.36363637,0.33501685,0.37878788,0.36026937,0.40656567,0.37373737,0.31228957,0.34848484,0.2912458,0.45622894,0.4722222,0.4335017,0.4107744,0.44781145,0.49579126,0.43097642,0.46969697,,0.53367,0.52020204,0.55387205,0.54545456,0.5892256,0.52693605,0.5875421,0.60690236,0.57575756,0.6212121,0.6296296,0.6691919,0.6936027,0.7121212,0.7323232,0.72643095,0.6952862,0.83838385,0.7070707,0.7373737,0.7558923,0.7676768,0.7373737,0.8148148,0.8249158,0.8552188,0.97306395
+ExtraTrees (tuned + ensembled),0.053872053,0.06902357,0.08080808,0.11026936,0.1026936,0.124579124,0.12794612,0.13047138,0.17424242,0.18686868,0.24915825,0.20707071,0.25420874,0.2929293,0.2020202,0.23232323,0.29713804,0.25925925,0.266835,0.30471382,0.2912458,0.28787878,0.32659933,0.26094276,0.30892256,0.29377106,0.38047138,0.36531988,0.28282827,0.31986532,0.31986532,0.34343433,0.33501685,0.34511784,0.3872054,0.33417508,0.36026937,0.33585858,0.37037036,0.35353535,0.3838384,0.39730638,0.42424244,0.44276094,0.3989899,0.42929292,0.41919193,0.46464646,0.43265992,0.4469697,0.46632996,,0.48821548,0.67171717,0.4983165,0.5589225,0.6195286,0.5740741,0.6632997,0.5707071,0.72727275,0.6026936,0.6195286,0.6456229,0.6784512,0.70454544,0.6675084,0.6700337,0.68013465,0.7373737,0.6919192,0.8299663,0.9057239,0.87205386,0.7676768,0.7828283,0.88383836,0.9915825
+ChimeraBoost (default),0.060606062,0.112794615,0.07575758,0.13636364,0.11784512,0.14478114,0.13973063,0.15319866,0.22390573,0.22053872,0.17003366,0.22053872,0.18518518,0.24579124,0.22558923,0.26094276,0.26936027,0.28619528,0.1919192,0.246633,0.26936027,0.3131313,0.28114477,0.2777778,0.23737374,0.28535354,0.3114478,0.37373737,0.28956228,0.3097643,0.25757575,0.37037036,0.3425926,0.37037036,0.44949496,0.3838384,0.3855219,0.37878788,0.3989899,0.4276094,0.4124579,0.43939394,0.37710437,0.22558923,0.42592594,0.43602693,0.46127945,0.30471382,0.47474748,0.503367,0.479798,0.5117845,,0.53198653,0.5597643,0.65656567,0.5707071,0.66498315,0.5925926,0.61026937,0.6111111,0.68350166,0.63468015,0.6784512,0.64141417,0.72727275,0.6919192,0.67171717,0.6936027,0.6952862,0.6969697,0.6944444,0.67171717,0.72727275,0.8013468,0.77609426,0.8518519,0.9747475
+ExtraTrees (tuned),0.062289562,0.07070707,0.09259259,0.10606061,0.11026936,0.12289562,0.11447811,0.121212125,0.14646465,0.17760943,0.22053872,0.2037037,0.25252524,0.27861953,0.18939394,0.2020202,0.28114477,0.24074075,0.26262626,0.28787878,0.27441078,0.26094276,0.30471382,0.24326599,0.27693602,0.26094276,0.33838382,0.34680134,0.27104378,0.2962963,0.2996633,0.2996633,0.2912458,0.3181818,0.36195287,0.2962963,0.3148148,0.32996634,0.33670035,0.34006733,0.35858586,0.3872054,0.37710437,0.4023569,0.36026937,0.37037036,0.4006734,0.41750842,0.3939394,0.42424244,0.44612795,0.32828283,0.46801347,,0.43939394,0.5597643,0.6077441,0.5707071,0.5555556,0.52020204,0.63383836,0.57239056,0.5892256,0.6262626,0.6750842,0.7003367,0.63973063,0.64309764,0.65656567,0.6936027,0.67171717,0.7895623,0.82828283,0.82575756,0.75757575,0.7626263,0.8888889,0.989899
+XGBoost (default),0.038720537,0.04882155,0.05050505,0.082491584,0.052188553,0.08585858,0.10606061,0.09259259,0.17340067,0.11111111,0.116161615,0.17845118,0.15151516,0.20707071,0.18350169,0.2020202,0.21885522,0.19023569,0.13468014,0.1952862,0.25252524,0.2760943,0.30471382,0.15151516,0.17087542,0.1969697,0.3131313,0.32154882,0.15824915,0.2777778,0.23063973,0.2979798,0.11784512,0.36700338,0.4057239,0.17508417,0.3905724,0.30723906,0.2912458,0.32491583,0.36363637,0.3888889,0.42087543,0.35521886,0.35858586,0.3838384,0.43434343,0.37289563,0.4090909,0.45959595,0.45454547,0.5016835,0.4402357,0.56060606,,0.6144781,0.5875421,0.6195286,0.55387205,0.5530303,0.61616164,0.6750842,0.64309764,0.6363636,0.66161615,0.7154882,0.65656567,0.6178451,0.6818182,0.7121212,0.66835016,0.74242425,0.7861953,0.7979798,0.76094276,0.76430976,0.82323235,0.95454544
+TabSTAR (tuned),0.14814815,0.1590909,0.18855219,0.17003366,0.19023569,0.18855219,0.16666667,0.1969697,0.24579124,0.23232323,0.23400673,0.2718855,0.23232323,0.2996633,0.25589225,0.2777778,0.3164983,0.24915825,0.2483165,0.24915825,0.21885522,0.31565657,0.24747474,0.25589225,0.25925925,0.3030303,0.33164984,0.30639732,0.25757575,0.28619528,0.26178452,0.30387205,0.27946126,0.36195287,0.3872054,0.2929293,0.2962963,0.27946126,0.35858586,0.30639732,0.2760943,0.35521886,0.33501685,0.30471382,0.2979798,0.3223906,0.3114478,0.3181818,0.32996634,0.33164984,0.4107744,0.44107744,0.34343433,0.4402357,0.3855219,,0.47643098,0.59175086,0.46464646,0.4090909,0.4983165,0.3838384,0.44107744,0.6010101,0.57575756,0.9478114,0.63131315,0.6111111,0.6329966,0.5858586,0.6262626,0.7003367,0.6885522,0.68686867,0.74915826,0.66498315,0.85353535,0.94612795
+OrionMSP (default),0.05892256,0.074074075,0.07154882,0.077441074,0.11784512,0.09427609,0.06734007,0.099326596,0.13299663,0.21717171,0.246633,0.20791246,0.24074075,0.25,0.17508417,0.17003366,0.26936027,0.27525252,0.25925925,0.26430976,0.26094276,0.17340067,0.13299663,0.30471382,0.26936027,0.30892256,0.33164984,0.2912458,0.30808082,0.28956228,0.32154882,0.26767677,0.31060606,0.25757575,0.35858586,0.32828283,0.2946128,0.3114478,0.38047138,0.36868685,0.33670035,0.4124579,0.30808082,0.36363637,0.3476431,0.3989899,0.35185185,0.37373737,0.4006734,0.43097642,0.47306398,0.38047138,0.42929292,0.3922559,0.4124579,0.52356905,,0.5252525,0.43434343,0.479798,0.46969697,0.5185185,0.530303,0.61279464,0.65656567,0.68350166,0.6388889,0.6329966,0.68013465,0.64646465,0.66835016,0.63804716,0.6599327,0.65488213,0.74579126,0.7811448,0.7676768,0.9713805
+TabSTAR (tuned + ensembled),0.14814815,0.15488215,0.18686868,0.17171717,0.18181819,0.18350169,0.16582492,0.18855219,0.23905724,0.22558923,0.22727273,0.2760943,0.22558923,0.2979798,0.25252524,0.2760943,0.3131313,0.24410774,0.24579124,0.239899,0.2138047,0.3148148,0.24242425,0.26094276,0.25420874,0.30218855,0.33670035,0.31060606,0.25589225,0.27946126,0.25925925,0.30471382,0.27946126,0.36195287,0.3905724,0.29040405,0.2912458,0.27946126,0.35690236,0.29882154,0.26936027,0.35016835,0.32828283,0.30471382,0.29208755,0.32828283,0.3131313,0.3181818,0.3164983,0.32996634,0.4124579,0.42592594,0.33501685,0.42929292,0.38047138,0.40824917,0.47474748,,0.45791245,0.41414142,0.493266,0.3872054,0.43771043,0.59090906,0.5673401,0.96127945,0.62794614,0.6060606,0.62794614,0.5824916,0.6195286,0.6885522,0.6818182,0.68350166,0.74242425,0.6632997,0.84511787,0.94612795
+RandomForest (tuned + ensembled),0.04040404,0.07575758,0.065656565,0.07912458,0.09090909,0.0959596,0.13468014,0.112794615,0.15488215,0.11447811,0.16498317,0.16835018,0.2037037,0.25589225,0.17508417,0.17171717,0.24747474,0.18686868,0.18686868,0.22558923,0.23905724,0.24747474,0.30639732,0.18518518,0.24242425,0.2138047,0.34175083,0.32323232,0.1936027,0.25589225,0.23400673,0.25925925,0.20286195,0.32828283,0.3476431,0.22558923,0.32828283,0.28956228,0.3164983,0.3030303,0.2962963,0.31228957,0.34848484,0.35185185,0.33501685,0.33501685,0.35858586,0.37542087,0.36026937,0.3922559,0.39309764,0.33670035,0.4074074,0.44444445,0.44612795,0.53535354,0.56565654,0.54208755,,0.496633,0.67760944,0.5084175,0.5707071,0.60942763,0.61616164,0.67087543,0.6228956,0.6212121,0.6262626,0.70538723,0.6447812,0.7558923,0.8047138,0.86195284,0.73316497,0.7087542,0.86195284,0.97306395
+LightGBM (default),0.038720537,0.052188553,0.045454547,0.062289562,0.055555556,0.065656565,0.065656565,0.07070707,0.13973063,0.09427609,0.09090909,0.16498317,0.09848485,0.2037037,0.11195286,0.15993266,0.2037037,0.16498317,0.052188553,0.17508417,0.21632996,0.22222222,0.21801347,0.12962963,0.083333336,0.14646465,0.28114477,0.28282827,0.15656565,0.23737374,0.17676768,0.28114477,0.112794615,0.30808082,0.33838382,0.14814815,0.32659933,0.2912458,0.27946126,0.30808082,0.35016835,0.3493266,0.32996634,0.2685185,0.34006733,0.32996634,0.37710437,0.30808082,0.36531988,0.42003366,0.42424244,0.42929292,0.38973063,0.479798,0.4469697,0.59090906,0.52020204,0.5858586,0.503367,,0.54461277,0.5993266,0.6077441,0.5892256,0.62457913,0.6952862,0.60942763,0.61279464,0.65488213,0.67676765,0.6734007,0.7138047,0.72390574,0.7314815,0.74579126,0.77272725,0.8299663,0.9688552
+RandomForest (tuned),0.042087544,0.053872053,0.06397306,0.07912458,0.06734007,0.07912458,0.08585858,0.08754209,0.13973063,0.08922559,0.13215488,0.15488215,0.15993266,0.22811447,0.14478114,0.16161616,0.22390573,0.17003366,0.13468014,0.20875421,0.22558923,0.21043772,0.25252524,0.1388889,0.17508417,0.17676768,0.29040405,0.28030303,0.15993266,0.23569024,0.19865319,0.21885522,0.17760943,0.2777778,0.33249158,0.18686868,0.27272728,0.26767677,0.3114478,0.2777778,0.28451177,0.3013468,0.28282827,0.30639732,0.2962963,0.28956228,0.32828283,0.33164984,0.32154882,0.35690236,0.37878788,0.27272728,0.3888889,0.3661616,0.3838384,0.5016835,0.530303,0.506734,0.3223906,0.4553872,,0.4814815,0.5117845,0.5808081,0.6026936,0.6481481,0.5976431,0.5976431,0.6178451,0.65824914,0.63131315,0.72306395,0.7373737,0.76094276,0.7188552,0.6919192,0.8518519,0.9747475
+iLTM (default),0.05892256,0.053872053,0.057239056,0.082491584,0.074074075,0.08754209,0.082491584,0.0959596,0.16329966,0.121212125,0.11952862,0.17676768,0.12626262,0.21212122,0.16666667,0.17676768,0.23905724,0.17003366,0.11111111,0.15151516,0.15488215,0.22727273,0.2138047,0.1574074,0.13215488,0.15656565,0.26094276,0.2760943,0.17003366,0.1936027,0.124579124,0.25420874,0.2003367,0.25925925,0.34006733,0.20875421,0.25252524,0.2053872,0.27946126,0.22474748,0.25252524,0.3030303,0.26936027,0.23569024,0.25589225,0.253367,0.25420874,0.27104378,0.28451177,0.30639732,0.37037036,0.39730638,0.3164983,0.4276094,0.32491583,0.61616164,0.4814815,0.61279464,0.49158248,0.4006734,0.5185185,,0.537037,0.59090906,0.5875421,0.6885522,0.61616164,0.5589225,0.60942763,0.6060606,0.57575756,0.7020202,0.63131315,0.6885522,0.7171717,0.65656567,0.8080808,0.93602693
+TorchMLP (default),0.033670034,0.047138046,0.042087544,0.05892256,0.05050505,0.05892256,0.038720537,0.05892256,0.12794612,0.08585858,0.10606061,0.17171717,0.09427609,0.1919192,0.15151516,0.14478114,0.2037037,0.12289562,0.16835018,0.10942761,0.17003366,0.17508417,0.18181819,0.17003366,0.18686868,0.17003366,0.26094276,0.24242425,0.19023569,0.1952862,0.18350169,0.18686868,0.20707071,0.24915825,0.26430976,0.22727273,0.1936027,0.23905724,0.27272728,0.12626262,0.21885522,0.239899,0.22053872,0.2996633,0.26430976,0.21717171,0.27272728,0.30639732,0.24494949,0.21801347,0.33080807,0.38047138,0.36531988,0.4107744,0.35690236,0.5589225,0.46969697,0.56228954,0.42929292,0.3922559,0.48821548,0.46296296,,0.59090906,0.5984849,0.6851852,0.6195286,0.57239056,0.6515151,0.6212121,0.6178451,0.68350166,0.6632997,0.64141417,0.73400676,0.7154882,0.8181818,0.956229
+Linear (tuned + ensembled),0.057239056,0.09259259,0.099326596,0.08080808,0.1026936,0.10942761,0.107744105,0.124579124,0.17508417,0.12962963,0.13383839,0.17171717,0.13299663,0.17845118,0.20875421,0.21885522,0.19865319,0.18350169,0.16329966,0.15319866,0.13973063,0.23821549,0.23905724,0.1936027,0.18686868,0.25252524,0.22895622,0.20286195,0.19865319,0.23737374,0.22053872,0.18013468,0.2053872,0.30471382,0.23905724,0.21717171,0.18350169,0.18518518,0.3181818,0.23148148,0.23063973,0.22895622,0.26262626,0.2929293,0.22053872,0.32744107,0.24242425,0.30808082,0.27104378,0.3164983,0.30639732,0.3543771,0.32154882,0.37373737,0.36363637,0.3989899,0.3872054,0.4090909,0.3905724,0.4107744,0.41919193,0.4090909,0.4090909,,0.496633,0.54208755,0.6919192,0.47306398,0.4814815,0.53872055,0.4949495,0.56565654,0.55387205,0.54208755,0.8080808,0.6010101,0.7020202,0.9040404
+TabFlex (default),0.062289562,0.07070707,0.08754209,0.08922559,0.08922559,0.09259259,0.07323232,0.1026936,0.09427609,0.17340067,0.16161616,0.21212122,0.17171717,0.18181819,0.15319866,0.14983165,0.21212122,0.19107744,0.21548821,0.18350169,0.17592593,0.16329966,0.09427609,0.26430976,0.22727273,0.1952862,0.23905724,0.18855219,0.27104378,0.2020202,0.25252524,0.18686868,0.26178452,0.15656565,0.24915825,0.27104378,0.17508417,0.22558923,0.2962963,0.22222222,0.26599327,0.21885522,0.22895622,0.25589225,0.246633,0.28367004,0.24242425,0.28451177,0.22643098,0.24410774,0.28787878,0.32154882,0.35858586,0.32491583,0.33838382,0.42424244,0.34343433,0.43265992,0.3838384,0.37542087,0.39730638,0.4124579,0.40151516,0.503367,,0.5808081,0.5345118,0.506734,0.56397307,0.53367,0.56565654,0.56565654,0.53367,0.5707071,0.65656567,0.6363636,0.72390574,0.92087543
+TabSTAR (default),0.10942761,0.124579124,0.14814815,0.12289562,0.15656565,0.15656565,0.13804714,0.15656565,0.17003366,0.18686868,0.18518518,0.21801347,0.18350169,0.25252524,0.1952862,0.1936027,0.25589225,0.19865319,0.17845118,0.2003367,0.1590909,0.21212122,0.17340067,0.21043772,0.1919192,0.2138047,0.26767677,0.24579124,0.2053872,0.20875421,0.2020202,0.23232323,0.22558923,0.26599327,0.23569024,0.22895622,0.2138047,0.2037037,0.2929293,0.21212122,0.1952862,0.23737374,0.22558923,0.22895622,0.22053872,0.25084174,0.22390573,0.24074075,0.23400673,0.24074075,0.26767677,0.29545453,0.27272728,0.2996633,0.28451177,0.052188553,0.3164983,0.038720537,0.32912457,0.30471382,0.35185185,0.3114478,0.3148148,0.45791245,0.41919193,,0.47138047,0.5218855,0.4090909,0.4814815,0.540404,0.540404,0.52693605,0.4814815,0.5976431,0.5892256,0.70538723,0.9326599
+Linear (tuned),0.05050505,0.08417509,0.077441074,0.07239057,0.082491584,0.09427609,0.1010101,0.104377106,0.16329966,0.12794612,0.11026936,0.15151516,0.11700337,0.15656565,0.1919192,0.2003367,0.17676768,0.17424242,0.14309764,0.14141414,0.121212125,0.21717171,0.21212122,0.17676768,0.16077441,0.21717171,0.21212122,0.1952862,0.18434343,0.20707071,0.19865319,0.16329966,0.1936027,0.2760943,0.23063973,0.1952862,0.15993266,0.16666667,0.3030303,0.1969697,0.21464646,0.2003367,0.23484848,0.27272728,0.19865319,0.28619528,0.20791246,0.2979798,0.22895622,0.29208755,0.27356902,0.33249158,0.30808082,0.36026937,0.34343433,0.36868685,0.3611111,0.37205386,0.37710437,0.3905724,0.4023569,0.3838384,0.38047138,0.30808082,0.46548823,0.5286195,,0.45622894,0.45454547,0.52693605,0.479798,0.5277778,0.53535354,0.5117845,0.72390574,0.5925926,0.6851852,0.8973064
+PerpetualBooster (tuned + ensembled),0.015151516,0.053872053,0.04882155,0.04882155,0.05892256,0.053872053,0.053872053,0.06818182,0.1026936,0.06902357,0.07575758,0.13636364,0.09427609,0.17171717,0.09427609,0.11447811,0.17171717,0.116161615,0.107744105,0.0959596,0.14309764,0.13636364,0.116161615,0.15824915,0.12542087,0.14814815,0.22053872,0.20707071,0.16835018,0.17003366,0.17845118,0.18518518,0.2020202,0.20707071,0.22558923,0.22727273,0.20622896,0.2003367,0.24915825,0.1936027,0.20959596,0.23905724,0.21548821,0.25084174,0.22222222,0.26094276,0.24074075,0.28872055,0.22222222,0.2777778,0.30471382,0.32996634,0.32828283,0.35690236,0.38215488,0.3888889,0.36700338,0.3939394,0.37878788,0.3872054,0.4023569,0.44107744,0.4276094,0.52693605,0.493266,0.4781145,0.543771,,0.48821548,0.55050504,0.712963,0.54461277,0.59090906,0.57912457,0.61279464,0.7979798,0.6885522,0.94276094
+FastaiMLP (default),0.03198653,0.045454547,0.047138046,0.053872053,0.065656565,0.065656565,0.06902357,0.074074075,0.08417509,0.065656565,0.07239057,0.12626262,0.099326596,0.16329966,0.13047138,0.10942761,0.16666667,0.091750845,0.15824915,0.13636364,0.15488215,0.12962963,0.14478114,0.16666667,0.17340067,0.13804714,0.22727273,0.21212122,0.18013468,0.16582492,0.16498317,0.11952862,0.20875421,0.17003366,0.22895622,0.21885522,0.13804714,0.18350169,0.23905724,0.1557239,0.2053872,0.04882155,0.21043772,0.22895622,0.21885522,0.1969697,0.2138047,0.24579124,0.15319866,0.23737374,0.16161616,0.31986532,0.30639732,0.34343433,0.3181818,0.36700338,0.31986532,0.37205386,0.37373737,0.34511784,0.38215488,0.3905724,0.34848484,0.5185185,0.43602693,0.59090906,0.54545456,0.5117845,,0.54208755,0.540404,0.57912457,0.52020204,0.5361953,0.65488213,0.65656567,0.75420874,0.956229
+xRFM (default),0.037037037,0.038720537,0.045454547,0.062289562,0.062289562,0.062289562,0.05892256,0.06397306,0.107744105,0.07239057,0.104377106,0.08922559,0.13299663,0.112794615,0.121212125,0.09848485,0.099326596,0.121212125,0.17340067,0.17171717,0.20707071,0.13299663,0.18013468,0.15151516,0.18602693,0.121212125,0.15824915,0.15656565,0.16329966,0.12962963,0.19865319,0.05892256,0.17676768,0.2037037,0.22727273,0.18855219,0.0959596,0.23569024,0.16414142,0.21548821,0.25757575,0.21548821,0.24242425,0.23400673,0.25925925,0.21043772,0.2912458,0.25757575,0.26936027,0.2760943,0.2929293,0.26262626,0.30471382,0.30639732,0.28787878,0.41414142,0.35353535,0.41750842,0.2946128,0.32323232,0.34175083,0.3939394,0.37878788,0.46127945,0.46632996,0.5185185,0.47306398,0.44949496,0.45791245,,0.4739057,0.47306398,0.46296296,0.4814815,0.55723906,0.57912457,0.62794614,0.8989899
+PerpetualBooster (tuned),0.018518519,0.042087544,0.04292929,0.043771043,0.062289562,0.052188553,0.047138046,0.062289562,0.08585858,0.062289562,0.07239057,0.121212125,0.074074075,0.15488215,0.082491584,0.103535354,0.14983165,0.116161615,0.08754209,0.08754209,0.124579124,0.12626262,0.112794615,0.13973063,0.10606061,0.12794612,0.21801347,0.20707071,0.14646465,0.14983165,0.14309764,0.1506734,0.17592593,0.18013468,0.21801347,0.18350169,0.16919191,0.18265994,0.24747474,0.15824915,0.2037037,0.21043772,0.18518518,0.21212122,0.2053872,0.22053872,0.22558923,0.253367,0.18013468,0.24074075,0.26262626,0.30808082,0.3030303,0.32828283,0.33164984,0.37373737,0.33164984,0.38047138,0.35521886,0.32659933,0.36868685,0.42424244,0.38215488,0.5050505,0.43434343,0.45959595,0.52020204,0.28703704,0.45959595,0.52609426,,0.510101,0.54713804,0.5521886,0.5892256,0.7171717,0.65824914,0.92087543
+KNN (tuned + ensembled),0.037037037,0.05892256,0.05892256,0.055555556,0.065656565,0.08417509,0.0976431,0.08754209,0.09090909,0.09090909,0.107744105,0.15488215,0.116161615,0.18350169,0.1026936,0.10690235,0.19023569,0.13299663,0.11111111,0.14646465,0.116161615,0.116161615,0.14814815,0.11784512,0.13636364,0.15319866,0.23232323,0.22222222,0.12626262,0.15993266,0.13468014,0.11700337,0.15656565,0.16498317,0.25757575,0.1557239,0.12794612,0.15824915,0.22558923,0.1952862,0.18855219,0.1952862,0.15488215,0.23063973,0.18518518,0.21885522,0.16161616,0.25084174,0.1919192,0.24494949,0.24410774,0.17003366,0.30555555,0.21043772,0.25757575,0.2996633,0.36195287,0.3114478,0.24410774,0.28619528,0.27693602,0.2979798,0.3164983,0.43434343,0.43434343,0.45959595,0.4722222,0.4553872,0.42087543,0.52693605,0.48989898,,0.47138047,0.48989898,0.5959596,0.5875421,0.8695286,0.9747475
+ExtraTrees (default),0.011784512,0.021885522,0.038720537,0.042087544,0.04040404,0.043771043,0.042087544,0.05050505,0.0513468,0.045454547,0.065656565,0.08922559,0.07912458,0.13299663,0.052188553,0.053872053,0.11111111,0.116161615,0.07070707,0.12289562,0.14646465,0.09090909,0.14309764,0.09427609,0.082491584,0.09090909,0.17340067,0.18686868,0.10606061,0.11952862,0.121212125,0.060606062,0.11447811,0.14141414,0.22053872,0.112794615,0.11447811,0.15824915,0.17676768,0.1969697,0.17003366,0.17340067,0.16414142,0.1952862,0.17171717,0.18855219,0.18518518,0.2053872,0.18686868,0.24747474,0.23232323,0.09427609,0.32828283,0.17171717,0.2138047,0.3114478,0.34006733,0.3181818,0.1952862,0.2760943,0.26262626,0.36868685,0.33670035,0.44612795,0.46632996,0.47306398,0.46464646,0.4090909,0.479798,0.537037,0.45286196,0.5286195,,0.510101,0.5892256,0.55387205,0.6952862,0.96801347
+RandomForest (default),0.018518519,0.018518519,0.037037037,0.033670034,0.028619528,0.043771043,0.042087544,0.047138046,0.047138046,0.045454547,0.07239057,0.07575758,0.08585858,0.12626262,0.052188553,0.060606062,0.13720539,0.10606061,0.060606062,0.11111111,0.13804714,0.09259259,0.112794615,0.074074075,0.08417509,0.09090909,0.20622896,0.19023569,0.082491584,0.13299663,0.112794615,0.07239057,0.082491584,0.16835018,0.21548821,0.0959596,0.13468014,0.15824915,0.18518518,0.19865319,0.17171717,0.21043772,0.16329966,0.20875421,0.17845118,0.18686868,0.2003367,0.23063973,0.2053872,0.24579124,0.26262626,0.12794612,0.27272728,0.17424242,0.2020202,0.3131313,0.34511784,0.3164983,0.13804714,0.2685185,0.23905724,0.3114478,0.35858586,0.45791245,0.42929292,0.5185185,0.48821548,0.42087543,0.46380472,0.5185185,0.44781145,0.510101,0.48989898,,0.5892256,0.56060606,0.7020202,0.956229
+Linear (default),0.033670034,0.05050505,0.054713804,0.046296295,0.05892256,0.069865316,0.052188553,0.06649832,0.077441074,0.10942761,0.08080808,0.14478114,0.08922559,0.12289562,0.091750845,0.13131313,0.15656565,0.13636364,0.08754209,0.11111111,0.065656565,0.14814815,0.11363637,0.14309764,0.104377106,0.16329966,0.18518518,0.14983165,0.14983165,0.15319866,0.12962963,0.13047138,0.15151516,0.17003366,0.18855219,0.15824915,0.112794615,0.13468014,0.25589225,0.13973063,0.15993266,0.14141414,0.13973063,0.17676768,0.15993266,0.20707071,0.13468014,0.1936027,0.15488215,0.17845118,0.18518518,0.23232323,0.19865319,0.24242425,0.23905724,0.25084174,0.25420874,0.25757575,0.266835,0.25420874,0.28114477,0.28282827,0.26599327,0.1919192,0.34343433,0.4023569,0.2760943,0.3872054,0.34511784,0.44276094,0.4107744,0.4040404,0.4107744,0.4107744,,0.5176768,0.6077441,0.8552188
+PerpetualBooster (default),0.016835017,0.030303031,0.035353534,0.023569023,0.037037037,0.035353534,0.028619528,0.035353534,0.038720537,0.033670034,0.052188553,0.08585858,0.037037037,0.06902357,0.042087544,0.05892256,0.07575758,0.04882155,0.055555556,0.045454547,0.074074075,0.060606062,0.06902357,0.09259259,0.057239056,0.08754209,0.11784512,0.104377106,0.08585858,0.099326596,0.099326596,0.09090909,0.1010101,0.107744105,0.18686868,0.1026936,0.1010101,0.1026936,0.17845118,0.1010101,0.11952862,0.11447811,0.13299663,0.15993266,0.121212125,0.14646465,0.14141414,0.16750842,0.10606061,0.15824915,0.17508417,0.21717171,0.22390573,0.23737374,0.23569024,0.33501685,0.21885522,0.33670035,0.2912458,0.22727273,0.30808082,0.34343433,0.28451177,0.3989899,0.36363637,0.4107744,0.4074074,0.2020202,0.34343433,0.42087543,0.28282827,0.4124579,0.44612795,0.43939394,0.48232323,,0.56902355,0.85690236
+KNN (tuned),0.037037037,0.04040404,0.045454547,0.045454547,0.04882155,0.062289562,0.053872053,0.057239056,0.04882155,0.06734007,0.07239057,0.121212125,0.07575758,0.11363637,0.060606062,0.074074075,0.12289562,0.1010101,0.069865316,0.09427609,0.043771043,0.08585858,0.057239056,0.08922559,0.07912458,0.11363637,0.15656565,0.13973063,0.082491584,0.112794615,0.077441074,0.09090909,0.103535354,0.124579124,0.14814815,0.10606061,0.08585858,0.09006734,0.1952862,0.107744105,0.09090909,0.11784512,0.1026936,0.13636364,0.08922559,0.15404041,0.074074075,0.15824915,0.112794615,0.15488215,0.14478114,0.116161615,0.14814815,0.11111111,0.17676768,0.14646465,0.23232323,0.15488215,0.13804714,0.17003366,0.14814815,0.1919192,0.18181819,0.2979798,0.2760943,0.2946128,0.3148148,0.3114478,0.24579124,0.37205386,0.34175083,0.13047138,0.30471382,0.2979798,0.3922559,0.43097642,,0.9444444
+KNN (default),0.005050505,0.0033670033,0.005050505,0.0033670033,0.0067340066,0.0067340066,0.005050505,0.0067340066,0.005050505,0.005050505,0.0033670033,0.025252525,0.013468013,0.008417509,0.0067340066,0.005050505,0.005050505,0.015151516,0.0067340066,0.02020202,0.011784512,0.008417509,0.005050505,0.0067340066,0.005050505,0.018518519,0.011784512,0.016835017,0.0067340066,0.016835017,0.01010101,0.008417509,0.015151516,0.0067340066,0.015151516,0.018518519,0.008417509,0.016835017,0.05050505,0.02020202,0.023569023,0.02020202,0.013468013,0.021043772,0.016835017,0.030303031,0.018518519,0.030303031,0.018518519,0.028619528,0.026936026,0.008417509,0.025252525,0.01010101,0.045454547,0.053872053,0.028619528,0.053872053,0.026936026,0.031144781,0.025252525,0.06397306,0.043771043,0.0959596,0.07912458,0.06734007,0.1026936,0.057239056,0.043771043,0.1010101,0.07912458,0.025252525,0.03198653,0.043771043,0.14478114,0.14309764,0.055555556,
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
index 144e85d6c88b5487c4ba67b1e167776f975a35b4..395681d7c4de874584e2927242d5eee363abfcc5 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Classification | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..770b35ed53fc4e72f0195924c7bd02c7a6e661b8
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_classification/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2031.0,"score":0.728,"rank":5.0,"hrank":2.76,"imp":5.25,"train_time":147.57,"predict_time":1.973,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-51","imputed":false},{"position":1,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2006.0,"score":0.852,"rank":5.54,"hrank":1.68,"imp":3.722,"train_time":38.86,"predict_time":18.759,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+293\/-161","imputed":false},{"position":2,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1875.0,"score":0.632,"rank":9.31,"hrank":3.6,"imp":6.35,"train_time":314.18,"predict_time":2.249,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+170\/-80","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1850.0,"score":0.622,"rank":10.26,"hrank":3.71,"imp":7.876,"train_time":1.1,"predict_time":0.289,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+280\/-175","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1761.0,"score":0.477,"rank":14.12,"hrank":9.24,"imp":9.735,"train_time":2.41,"predict_time":0.701,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-87","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1751.0,"score":0.483,"rank":14.6,"hrank":5.07,"imp":8.717,"train_time":0.63,"predict_time":0.17,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+190\/-110","imputed":false},{"position":6,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1711.0,"score":0.46,"rank":16.71,"hrank":8.41,"imp":9.316,"train_time":378.5,"predict_time":0.168,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+142\/-85","imputed":false},{"position":7,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1705.0,"score":0.47,"rank":17.02,"hrank":7.77,"imp":9.178,"train_time":1.49,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-76","imputed":false},{"position":8,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1704.0,"score":0.43,"rank":17.04,"hrank":13.03,"imp":10.529,"train_time":1607.74,"predict_time":72.982,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-68","imputed":false},{"position":9,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1697.0,"score":0.452,"rank":17.43,"hrank":8.37,"imp":9.588,"train_time":378.5,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+155\/-95","imputed":false},{"position":10,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1673.0,"score":0.381,"rank":18.8,"hrank":12.4,"imp":11.387,"train_time":207.65,"predict_time":0.398,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-92","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1669.0,"score":0.367,"rank":19.06,"hrank":13.13,"imp":11.7,"train_time":93.21,"predict_time":0.758,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-76","imputed":false},{"position":12,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1668.0,"score":0.411,"rank":19.11,"hrank":11.12,"imp":11.214,"train_time":90.17,"predict_time":0.314,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-131","imputed":false},{"position":13,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1652.0,"score":0.374,"rank":20.02,"hrank":12.72,"imp":10.95,"train_time":1624.55,"predict_time":1.763,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+122\/-90","imputed":false},{"position":14,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1640.0,"score":0.35,"rank":20.76,"hrank":10.11,"imp":11.175,"train_time":783.29,"predict_time":13.79,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+190\/-114","imputed":false},{"position":15,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1636.0,"score":0.308,"rank":21.0,"hrank":11.26,"imp":11.793,"train_time":860.28,"predict_time":0.554,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+159\/-104","imputed":false},{"position":16,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1628.0,"score":0.344,"rank":21.53,"hrank":14.17,"imp":11.383,"train_time":2338.19,"predict_time":14.248,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-124","imputed":false},{"position":17,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1610.0,"score":0.35,"rank":22.66,"hrank":12.79,"imp":11.808,"train_time":90.17,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+144\/-110","imputed":false},{"position":18,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1606.0,"score":0.299,"rank":22.92,"hrank":18.3,"imp":12.162,"train_time":207.65,"predict_time":0.056,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-73","imputed":false},{"position":19,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1599.0,"score":0.31,"rank":23.37,"hrank":12.04,"imp":11.899,"train_time":783.29,"predict_time":1.924,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+180\/-114","imputed":false},{"position":20,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1571.0,"score":0.275,"rank":25.16,"hrank":14.26,"imp":12.216,"train_time":1624.55,"predict_time":0.083,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-94","imputed":false},{"position":21,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1565.0,"score":0.251,"rank":25.57,"hrank":13.41,"imp":12.672,"train_time":860.28,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-125","imputed":false},{"position":22,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1562.0,"score":0.227,"rank":25.74,"hrank":19.22,"imp":13.259,"train_time":93.21,"predict_time":0.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-73","imputed":false},{"position":23,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1561.0,"score":0.268,"rank":25.82,"hrank":16.13,"imp":12.843,"train_time":0.38,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-116","imputed":false},{"position":24,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1552.0,"score":0.258,"rank":26.4,"hrank":16.78,"imp":12.737,"train_time":2.3,"predict_time":0.82,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+133\/-99","imputed":false},{"position":25,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1531.0,"score":0.244,"rank":27.8,"hrank":16.13,"imp":13.434,"train_time":7.51,"predict_time":3.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-135","imputed":false},{"position":26,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1522.0,"score":0.205,"rank":28.46,"hrank":15.98,"imp":13.587,"train_time":1607.74,"predict_time":10.661,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-91","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1518.0,"score":0.177,"rank":28.68,"hrank":21.64,"imp":13.717,"train_time":2338.19,"predict_time":0.54,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-76","imputed":false},{"position":28,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1514.0,"score":0.247,"rank":28.99,"hrank":13.95,"imp":14.355,"train_time":0.79,"predict_time":0.23,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+182\/-170","imputed":false},{"position":29,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1490.0,"score":0.174,"rank":30.63,"hrank":18.37,"imp":13.722,"train_time":4805.68,"predict_time":461.79,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-129","imputed":false},{"position":30,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1488.0,"score":0.168,"rank":30.77,"hrank":23.79,"imp":13.578,"train_time":1169.32,"predict_time":1.442,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-77","imputed":false},{"position":31,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1474.0,"score":0.176,"rank":31.72,"hrank":23.99,"imp":15.517,"train_time":442.2,"predict_time":0.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+136\/-112","imputed":false},{"position":32,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1456.0,"score":0.175,"rank":33.02,"hrank":19.6,"imp":14.317,"train_time":0.47,"predict_time":0.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-108","imputed":false},{"position":33,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1424.0,"score":0.17,"rank":35.31,"hrank":24.58,"imp":15.536,"train_time":709.86,"predict_time":2.929,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-144","imputed":false},{"position":34,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1418.0,"score":0.141,"rank":35.67,"hrank":23.87,"imp":15.521,"train_time":0.62,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+151\/-150","imputed":false},{"position":35,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1407.0,"score":0.115,"rank":36.5,"hrank":29.68,"imp":16.508,"train_time":442.2,"predict_time":0.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+139\/-105","imputed":false},{"position":36,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1407.0,"score":0.127,"rank":36.51,"hrank":28.3,"imp":15.92,"train_time":1.56,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-128","imputed":false},{"position":37,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1393.0,"score":0.139,"rank":37.49,"hrank":29.68,"imp":17.32,"train_time":709.86,"predict_time":0.19,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+142\/-146","imputed":false},{"position":38,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1369.0,"score":0.054,"rank":39.18,"hrank":35.44,"imp":16.81,"train_time":6.98,"predict_time":0.233,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-89","imputed":false},{"position":39,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1363.0,"score":0.069,"rank":39.58,"hrank":31.87,"imp":18.84,"train_time":84.36,"predict_time":0.919,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-125","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1359.0,"score":0.06,"rank":39.86,"hrank":33.92,"imp":15.588,"train_time":1169.32,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-79","imputed":false},{"position":41,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1356.0,"score":0.095,"rank":40.08,"hrank":31.95,"imp":17.692,"train_time":389.62,"predict_time":2.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-164","imputed":false},{"position":42,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1354.0,"score":0.071,"rank":40.21,"hrank":35.46,"imp":18.314,"train_time":1.44,"predict_time":0.007,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-108","imputed":false},{"position":43,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1335.0,"score":0.049,"rank":41.54,"hrank":28.22,"imp":17.081,"train_time":4805.68,"predict_time":43.211,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-117","imputed":false},{"position":44,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1334.0,"score":0.048,"rank":41.57,"hrank":33.67,"imp":16.69,"train_time":52.45,"predict_time":0.219,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-104","imputed":false},{"position":45,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1315.0,"score":0.058,"rank":42.86,"hrank":38.29,"imp":19.676,"train_time":68.97,"predict_time":8.043,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-105","imputed":false},{"position":46,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1312.0,"score":0.033,"rank":43.04,"hrank":39.84,"imp":17.073,"train_time":0.28,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-99","imputed":false},{"position":47,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1302.0,"score":0.168,"rank":43.71,"hrank":7.78,"imp":20.306,"train_time":5.06,"predict_time":2.932,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+231\/-279","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1300.0,"score":0.049,"rank":43.87,"hrank":35.85,"imp":17.509,"train_time":52.45,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-111","imputed":false},{"position":49,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1292.0,"score":0.012,"rank":44.38,"hrank":41.41,"imp":17.813,"train_time":10.5,"predict_time":0.606,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-75","imputed":false},{"position":50,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1277.0,"score":0.018,"rank":45.38,"hrank":42.99,"imp":16.935,"train_time":50.19,"predict_time":0.154,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-94","imputed":false},{"position":51,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1276.0,"score":0.011,"rank":45.48,"hrank":42.15,"imp":19.434,"train_time":14.85,"predict_time":13.554,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-105","imputed":false},{"position":52,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1251.0,"score":0.017,"rank":47.08,"hrank":37.59,"imp":20.743,"train_time":46.9,"predict_time":45.623,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-146","imputed":false},{"position":53,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1227.0,"score":0.026,"rank":48.6,"hrank":42.19,"imp":20.621,"train_time":389.62,"predict_time":0.112,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-148","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1226.0,"score":0.0,"rank":48.66,"hrank":46.4,"imp":18.338,"train_time":50.19,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-94","imputed":false},{"position":55,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1203.0,"score":0.0,"rank":50.08,"hrank":48.37,"imp":19.659,"train_time":1.87,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-73","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1159.0,"score":0.029,"rank":52.64,"hrank":45.84,"imp":24.202,"train_time":1.61,"predict_time":0.112,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-214","imputed":false},{"position":57,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1103.0,"score":0.0,"rank":55.76,"hrank":54.08,"imp":26.789,"train_time":2951.8,"predict_time":1.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-104","imputed":false},{"position":58,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1099.0,"score":0.008,"rank":55.96,"hrank":50.37,"imp":28.713,"train_time":43.46,"predict_time":0.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-242","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1094.0,"score":0.0,"rank":56.22,"hrank":54.6,"imp":26.863,"train_time":2951.8,"predict_time":4.991,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-104","imputed":false},{"position":60,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1073.0,"score":0.009,"rank":57.31,"hrank":48.05,"imp":26.652,"train_time":331.68,"predict_time":44.947,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+101\/-218","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1056.0,"score":0.004,"rank":58.16,"hrank":53.45,"imp":29.813,"train_time":43.46,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+157\/-250","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1032.0,"score":0.0,"rank":59.3,"hrank":51.72,"imp":27.448,"train_time":22.82,"predict_time":0.606,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+125\/-201","imputed":false},{"position":63,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":60.76,"hrank":59.36,"imp":25.121,"train_time":0.12,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-180","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":994.0,"score":0.0,"rank":61.03,"hrank":56.14,"imp":28.773,"train_time":22.82,"predict_time":0.253,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-228","imputed":false},{"position":65,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":968.0,"score":0.0,"rank":62.12,"hrank":60.83,"imp":27.361,"train_time":17.17,"predict_time":6.883,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+99\/-178","imputed":false},{"position":66,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":968.0,"score":0.0,"rank":62.14,"hrank":60.79,"imp":27.159,"train_time":0.34,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-217","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":950.0,"score":0.0,"rank":62.88,"hrank":60.63,"imp":31.351,"train_time":0.2,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-229","imputed":false},{"position":68,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":935.0,"score":0.0,"rank":63.46,"hrank":62.64,"imp":32.487,"train_time":86.28,"predict_time":1.898,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-106","imputed":false},{"position":69,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":932.0,"score":0.021,"rank":63.54,"hrank":56.64,"imp":33.606,"train_time":2.8,"predict_time":0.224,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+189\/-424","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":880.0,"score":0.0,"rank":65.42,"hrank":64.37,"imp":34.878,"train_time":0.2,"predict_time":0.033,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-164","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":852.0,"score":0.0,"rank":66.31,"hrank":65.97,"imp":33.83,"train_time":331.68,"predict_time":1.626,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-124","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":804.0,"score":0.0,"rank":67.69,"hrank":66.53,"imp":36.946,"train_time":1.64,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-282","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":512.0,"score":0.0,"rank":72.64,"hrank":72.61,"imp":50.452,"train_time":0.12,"predict_time":0.101,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-308","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/pareto_front_explorer.html
index e3fa8097feb5778d8bc801cfe57e1b9277f31e3c..5c92ef4eb6b7cf7fcf309bc31640239e6f0d95a3 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.9189171706,"imp":18.840007406,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0190099757,"imp":9.1783534259,"elo":1704.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0199434097,"imp":9.5878875099,"elo":1697.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1678973097,"imp":9.315864629,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0229625096,"imp":12.843489519,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0260723508,"imp":11.8083619918,"elo":1609.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3141545533,"imp":11.2140890816,"elo":1667.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0069074234,"imp":18.313628056,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0091418408,"imp":16.5078148968,"elo":1406.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0944568378,"imp":15.5170264457,"elo":1474.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0569778809,"imp":27.1590909048,"elo":967.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0150080481,"imp":18.3381917932,"elo":1225.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1535912971,"imp":16.9352042076,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1123933797,"imp":24.2020361148,"elo":1159.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1117599745,"imp":20.620988534,"elo":1226.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.1135199956,"imp":17.6917823414,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.033160957,"imp":17.0730560129,"elo":1312.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0872426084,"imp":13.2588415734,"elo":1562.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7582948004,"imp":11.7004556275,"elo":1668.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.9321947194,"imp":20.305650329,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.6055455764,"imp":17.8133281766,"elo":1292.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5401190643,"imp":13.7174094198,"elo":1518.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":14.2476482136,"imp":11.3829611751,"elo":1627.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8827204019,"imp":27.3609275393,"elo":968.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0226836019,"imp":36.946006529,"elo":804.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2526043838,"imp":28.7726864862,"elo":993.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6062271231,"imp":27.4482786558,"elo":1032.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0133914852,"imp":25.1205416265,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0207808341,"imp":17.5089228838,"elo":1300.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2185753382,"imp":16.6904413475,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.232511973,"imp":16.8103385739,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0834429515,"imp":12.215693912,"elo":1571.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.762731528,"imp":10.9499682998,"elo":1652.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8199508222,"imp":12.7365093479,"elo":1552.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.9240530403,"imp":11.8993811355,"elo":1598.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":13.789522936,"imp":11.1745533202,"elo":1640.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.5537416022,"imp":19.4340799677,"elo":1275.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.622831341,"imp":20.7430151958,"elo":1250.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.2108859387,"imp":17.0805751385,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":461.7902853066,"imp":13.722448173,"elo":1489.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2302397985,"imp":14.3547062199,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":18.7591592077,"imp":3.7223594114,"elo":2005.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0330391224,"imp":34.8784648588,"elo":879.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.1457473701,"imp":13.434078273,"elo":1531.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1703078893,"imp":8.7168375107,"elo":1751.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0548347304,"imp":15.9198153896,"elo":1406.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0550494549,"imp":12.6721520809,"elo":1564.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.554086605,"imp":11.7928607556,"elo":1636.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7011909333,"imp":9.7349142793,"elo":1760.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2888892556,"imp":7.8760094387,"elo":1849.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.898348797,"imp":32.4874771321,"elo":934.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6889181125,"imp":26.7888300543,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9908559667,"imp":26.8629927897,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1693477495,"imp":15.5212999312,"elo":1418.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0462763773,"imp":19.6594845905,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0534965201,"imp":15.5879677352,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4424042458,"imp":13.577761493,"elo":1488.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.052450313,"imp":14.3165551195,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0561708961,"imp":12.1622111016,"elo":1605.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3981946384,"imp":11.387418587,"elo":1673.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":8.0432165428,"imp":19.6755609009,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.6611582164,"imp":13.586548919,"elo":1521.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":72.9817040915,"imp":10.5294798486,"elo":1704.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.2236113698,"imp":33.606488769,"elo":932.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1897669991,"imp":17.3202372496,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9293738561,"imp":15.5356888427,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.9189171706,"x_train":84.3646061894,"imp":18.840007406,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0190099757,"x_train":1.4934673843,"imp":9.1783534259,"elo":1704.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0199434097,"x_train":378.4994428039,"imp":9.5878875099,"elo":1697.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1678973097,"x_train":378.4994428039,"imp":9.315864629,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0229625096,"x_train":0.3843823491,"imp":12.843489519,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0260723508,"x_train":90.1654472213,"imp":11.8083619918,"elo":1609.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3141545533,"x_train":90.1654472213,"imp":11.2140890816,"elo":1667.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0069074234,"x_train":1.4432254313,"imp":18.313628056,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0091418408,"x_train":442.2024676551,"imp":16.5078148968,"elo":1406.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0944568378,"x_train":442.2024676551,"imp":15.5170264457,"elo":1474.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0569778809,"x_train":0.3394264702,"imp":27.1590909048,"elo":967.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0150080481,"x_train":50.1933853639,"imp":18.3381917932,"elo":1225.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1535912971,"x_train":50.1933853639,"imp":16.9352042076,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1123933797,"x_train":1.6121148448,"imp":24.2020361148,"elo":1159.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1117599745,"x_train":389.6192568267,"imp":20.620988534,"elo":1226.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.1135199956,"x_train":389.6192568267,"imp":17.6917823414,"elo":1355.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.033160957,"x_train":0.2814937163,"imp":17.0730560129,"elo":1312.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0872426084,"x_train":93.2136338393,"imp":13.2588415734,"elo":1562.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7582948004,"x_train":93.2136338393,"imp":11.7004556275,"elo":1668.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.9321947194,"x_train":5.058611156,"imp":20.305650329,"elo":1302.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.6055455764,"x_train":10.5029291491,"imp":17.8133281766,"elo":1292.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5401190643,"x_train":2338.1908652586,"imp":13.7174094198,"elo":1518.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":14.2476482136,"x_train":2338.1908652586,"imp":11.3829611751,"elo":1627.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8827204019,"x_train":17.1713211146,"imp":27.3609275393,"elo":968.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0226836019,"x_train":1.6442071407,"imp":36.946006529,"elo":804.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2526043838,"x_train":22.8161312471,"imp":28.7726864862,"elo":993.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6062271231,"x_train":22.8161312471,"imp":27.4482786558,"elo":1032.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0133914852,"x_train":0.1211565882,"imp":25.1205416265,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0207808341,"x_train":52.4497680024,"imp":17.5089228838,"elo":1300.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2185753382,"x_train":52.4497680024,"imp":16.6904413475,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.232511973,"x_train":6.9801966051,"imp":16.8103385739,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0834429515,"x_train":1624.5504043652,"imp":12.215693912,"elo":1571.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.762731528,"x_train":1624.5504043652,"imp":10.9499682998,"elo":1652.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8199508222,"x_train":2.2973997315,"imp":12.7365093479,"elo":1552.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.9240530403,"x_train":783.2895068512,"imp":11.8993811355,"elo":1598.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":13.789522936,"x_train":783.2895068512,"imp":11.1745533202,"elo":1640.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.5537416022,"x_train":14.8539049818,"imp":19.4340799677,"elo":1275.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.622831341,"x_train":46.8983759891,"imp":20.7430151958,"elo":1250.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.2108859387,"x_train":4805.6782161313,"imp":17.0805751385,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":461.7902853066,"x_train":4805.6782161313,"imp":13.722448173,"elo":1489.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2302397985,"x_train":0.7869111961,"imp":14.3547062199,"elo":1513.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":18.7591592077,"x_train":38.8556954397,"imp":3.7223594114,"elo":2005.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0330391224,"x_train":0.1957996411,"imp":34.8784648588,"elo":879.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.1457473701,"x_train":7.5078983053,"imp":13.434078273,"elo":1531.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1703078893,"x_train":0.6258098026,"imp":8.7168375107,"elo":1751.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0548347304,"x_train":1.5586488241,"imp":15.9198153896,"elo":1406.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0550494549,"x_train":860.2847801702,"imp":12.6721520809,"elo":1564.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.554086605,"x_train":860.2847801702,"imp":11.7928607556,"elo":1636.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7011909333,"x_train":2.4084538179,"imp":9.7349142793,"elo":1760.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2888892556,"x_train":1.0968929682,"imp":7.8760094387,"elo":1849.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.898348797,"x_train":86.2757906839,"imp":32.4874771321,"elo":934.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6889181125,"x_train":2951.7955583902,"imp":26.7888300543,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9908559667,"x_train":2951.7955583902,"imp":26.8629927897,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1693477495,"x_train":0.6181258773,"imp":15.5212999312,"elo":1418.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0462763773,"x_train":1.8659541839,"imp":19.6594845905,"elo":1202.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0534965201,"x_train":1169.3150011596,"imp":15.5879677352,"elo":1359.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4424042458,"x_train":1169.3150011596,"imp":13.577761493,"elo":1488.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.052450313,"x_train":0.4700674902,"imp":14.3165551195,"elo":1455.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0561708961,"x_train":207.6491745508,"imp":12.1622111016,"elo":1605.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3981946384,"x_train":207.6491745508,"imp":11.387418587,"elo":1673.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":8.0432165428,"x_train":68.9669811874,"imp":19.6755609009,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":10.6611582164,"x_train":1607.7353976095,"imp":13.586548919,"elo":1521.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":72.9817040915,"x_train":1607.7353976095,"imp":10.5294798486,"elo":1704.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.2236113698,"x_train":2.8035263333,"imp":33.606488769,"elo":932.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1897669991,"x_train":709.8550005699,"imp":17.3202372496,"elo":1392.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9293738561,"x_train":709.8550005699,"imp":15.5356888427,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
index d668fbf08564a149ddb280915189026f6f3ff1b5..ac4ea238010fe448ef05af479526ec44b90c30f9 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":84.3646061894,"imp":18.8599955331,"elo":1384.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.4934673843,"imp":9.1976006657,"elo":1740.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":4.0498969951,"imp":9.3187381366,"elo":1737.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.8156310096,"imp":9.3725589058,"elo":1748.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":23.9442866701,"imp":9.3921018347,"elo":1749.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":52.2059341627,"imp":9.2916172381,"elo":1760.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":97.6448146128,"imp":9.2566381211,"elo":1764.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":190.15396393,"imp":9.2967413641,"elo":1758.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":281.6941410441,"imp":9.3113324842,"elo":1756.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":378.4994428039,"imp":9.3273441341,"elo":1753.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.3843823491,"imp":12.8635534115,"elo":1590.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.8231266994,"imp":12.2327635752,"elo":1643.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.0711841529,"imp":11.8377408404,"elo":1670.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.198580454,"imp":11.6213761336,"elo":1685.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.0736254563,"imp":11.5012152174,"elo":1694.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.535781385,"imp":11.4021767141,"elo":1699.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":44.9181007897,"imp":11.2853772779,"elo":1707.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":67.208700505,"imp":11.2408277916,"elo":1709.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":90.1654472213,"imp":11.2307808149,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":1.4432254313,"imp":18.3289887858,"elo":1366.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":4.2347903447,"imp":17.4720804235,"elo":1400.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":10.3173350008,"imp":16.7542431694,"elo":1433.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":21.8514222981,"imp":16.4427611222,"elo":1447.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":55.5931037878,"imp":16.1033145209,"elo":1466.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":111.097193586,"imp":15.9162222115,"elo":1477.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":219.5483355792,"imp":15.709674653,"elo":1490.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":330.9116540058,"imp":15.6383543743,"elo":1494.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":442.2024676551,"imp":15.52589414,"elo":1502.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3394264702,"imp":27.1741880295,"elo":961.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.561280236,"imp":24.2835226498,"elo":1037.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.3509100065,"imp":19.8198892022,"elo":1167.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.5531799736,"imp":18.2046872025,"elo":1232.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.2677435481,"imp":17.2453969381,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.4252027007,"imp":17.0238417989,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.8317026638,"imp":16.8635575395,"elo":1296.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.332646039,"imp":16.8714868407,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":50.1933853639,"imp":16.9523835395,"elo":1299.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.6121148448,"imp":24.2159078135,"elo":1174.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.4214998387,"imp":22.7745276103,"elo":1208.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":9.5320935926,"imp":20.9902035951,"elo":1270.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":19.7206332887,"imp":19.9998660631,"elo":1299.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":47.120631421,"imp":19.1321650838,"elo":1332.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":95.0331726255,"imp":18.6307722458,"elo":1351.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":192.9295597629,"imp":18.1898274964,"elo":1365.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":290.1521255355,"imp":17.891970254,"elo":1378.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":389.6192568267,"imp":17.7004215583,"elo":1386.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.2814937163,"imp":17.0899008251,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.8419891145,"imp":16.2689403096,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":2.1642156548,"imp":14.9266707853,"elo":1504.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":4.3828192424,"imp":13.8517848473,"elo":1586.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":11.3294435796,"imp":12.6195613378,"elo":1649.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":23.1362359393,"imp":12.0578006895,"elo":1690.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":46.1918681929,"imp":11.8306177522,"elo":1704.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":68.9587529747,"imp":11.7603011738,"elo":1706.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":93.2136338393,"imp":11.7281550364,"elo":1710.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":5.058611156,"imp":19.8907338623,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":10.5029291491,"imp":17.8299401868,"elo":1315.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.3830456847,"imp":15.8929011855,"elo":1407.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":65.7796741249,"imp":13.5293238,"elo":1549.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":124.0730628064,"imp":12.7321692308,"elo":1595.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":283.3505795288,"imp":11.9230807327,"elo":1644.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":583.6843020141,"imp":11.5856976387,"elo":1664.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1153.7001044299,"imp":11.4878041124,"elo":1671.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1719.5734803795,"imp":11.4420520926,"elo":1671.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2338.1908652586,"imp":11.4152796867,"elo":1670.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":17.1713211146,"imp":27.4282235907,"elo":974.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.6442071407,"imp":37.3438261467,"elo":804.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.5769737451,"imp":34.813557491,"elo":883.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":19.1886688806,"imp":27.6055384276,"elo":1038.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":22.8161312471,"imp":27.4540417274,"elo":1046.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1211565882,"imp":25.1437303937,"elo":998.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.39306177,"imp":20.0876024453,"elo":1167.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.261863005,"imp":18.1118560726,"elo":1266.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":2.6044216897,"imp":17.448305273,"elo":1303.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":6.4126782162,"imp":16.9889325644,"elo":1333.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":12.9417195679,"imp":16.8332836861,"elo":1345.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":25.9261667663,"imp":16.7203484605,"elo":1355.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":39.086650906,"imp":16.6653861195,"elo":1360.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":52.4497680024,"imp":16.7072446847,"elo":1364.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.9801966051,"imp":16.8241005098,"elo":1387.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.4217554801,"imp":14.7107487455,"elo":1517.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.0914867245,"imp":13.5019228572,"elo":1584.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":79.6483382915,"imp":12.714245834,"elo":1617.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":203.6966152959,"imp":11.9655799475,"elo":1653.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":410.5193453014,"imp":11.6538509223,"elo":1665.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":808.8154712049,"imp":11.2152910283,"elo":1680.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1213.5228253628,"imp":11.0331396261,"elo":1690.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1624.5504043652,"imp":10.9669183657,"elo":1696.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2.2973997315,"imp":12.757790933,"elo":1576.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.7558525217,"imp":12.7040141437,"elo":1584.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":18.4525573231,"imp":12.5678686593,"elo":1599.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":36.2094345784,"imp":12.5003093797,"elo":1607.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":88.4354783372,"imp":12.1553628698,"elo":1640.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":183.3566697473,"imp":11.7002365934,"elo":1661.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":377.7921394838,"imp":11.4270632498,"elo":1669.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":572.0185730315,"imp":11.2797380522,"elo":1669.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":783.2895068512,"imp":11.188360367,"elo":1668.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":14.8539049818,"imp":19.4459273431,"elo":1297.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.8983759891,"imp":20.7566063243,"elo":1273.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":69.8822600854,"imp":19.1332286663,"elo":1313.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":127.4067152197,"imp":17.5591056151,"elo":1372.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":229.7746727932,"imp":16.995992756,"elo":1392.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":589.1070615002,"imp":15.6827832788,"elo":1452.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1193.0508698741,"imp":14.9633943872,"elo":1484.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2388.3823364922,"imp":14.4617778749,"elo":1503.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3593.4863060271,"imp":14.0010964181,"elo":1521.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4805.6782161313,"imp":13.7360803063,"elo":1526.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.7869111961,"imp":14.371186543,"elo":1544.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.8556954397,"imp":3.7240503537,"elo":2039.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1957996411,"imp":34.8895165819,"elo":880.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":7.5078983053,"imp":13.454211295,"elo":1564.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6258098026,"imp":8.7394017522,"elo":1785.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.5586488241,"imp":15.9377258692,"elo":1426.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":5.192079831,"imp":14.9598444176,"elo":1470.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":19.2797854853,"imp":13.7607602335,"elo":1531.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":41.1266447543,"imp":13.0968145146,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":106.1594477641,"imp":12.4863504012,"elo":1615.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":213.4318571252,"imp":12.1925610385,"elo":1638.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":422.7331976766,"imp":11.9493041925,"elo":1654.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":638.9519356459,"imp":11.8594998702,"elo":1662.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":860.2847801702,"imp":11.8049633358,"elo":1670.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.4084538179,"imp":9.7560536872,"elo":1801.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.0968929682,"imp":7.8824971032,"elo":1878.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":86.2757906839,"imp":32.4967224715,"elo":934.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":96.7574708184,"imp":32.41479258,"elo":949.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":132.620297349,"imp":32.2594861504,"elo":955.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":182.3615021409,"imp":31.786012596,"elo":975.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":379.7267185426,"imp":30.7058200242,"elo":1005.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":737.6739956698,"imp":29.3722505198,"elo":1037.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1465.1527881613,"imp":28.4744557852,"elo":1064.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2179.2889039095,"imp":27.6256143537,"elo":1092.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2951.7955583902,"imp":26.8746550395,"elo":1121.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.6181258773,"imp":15.5418068987,"elo":1445.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":1.8659541839,"imp":19.6714412852,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.8422781275,"imp":18.2754375515,"elo":1279.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":28.1225860036,"imp":16.8936521045,"elo":1343.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":62.6333146165,"imp":16.0329912533,"elo":1393.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":153.2940248569,"imp":15.1322927453,"elo":1438.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":287.4960670101,"imp":14.5504300077,"elo":1473.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":565.1252028376,"imp":14.1383335222,"elo":1498.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":871.368400213,"imp":13.8720944137,"elo":1513.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1169.3150011596,"imp":13.597536455,"elo":1523.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4700674902,"imp":14.3359708531,"elo":1484.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.0561352068,"imp":12.8895219142,"elo":1600.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":5.7242090848,"imp":12.2710877404,"elo":1655.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":9.9945517672,"imp":11.8250868461,"elo":1683.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":26.428780062,"imp":11.5620514204,"elo":1698.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":52.4146823069,"imp":11.4682434169,"elo":1704.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":103.8233704142,"imp":11.4119948439,"elo":1710.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":154.5284843647,"imp":11.4080903645,"elo":1713.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":207.6491745508,"imp":11.4089279264,"elo":1716.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":68.9669811874,"imp":19.6910335019,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":130.6527630186,"imp":17.5680857216,"elo":1390.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":316.404564055,"imp":14.9245895934,"elo":1499.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":624.6899718112,"imp":12.8882163313,"elo":1603.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1547.1736962358,"imp":10.591051261,"elo":1743.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1607.7353976095,"imp":10.5488846537,"elo":1749.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.8035263333,"imp":33.6178632238,"elo":926.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":6.8864266433,"imp":27.0655054285,"elo":1049.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.5478119749,"imp":19.7955614958,"elo":1258.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":35.2197842651,"imp":16.9393451699,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.4231720748,"imp":15.9543591258,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":176.2334816176,"imp":15.5960824642,"elo":1464.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":353.6101338707,"imp":15.4855978336,"elo":1463.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":529.672353661,"imp":15.4670973384,"elo":1461.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":709.8550005699,"imp":15.5056797319,"elo":1454.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.9189171706,"x_train":84.3646061894,"imp":18.8599955331,"elo":1384.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0190099757,"x_train":1.4934673843,"imp":9.1976006657,"elo":1740.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.038142376,"x_train":4.0498969951,"imp":9.3187381366,"elo":1737.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0612481453,"x_train":12.8156310096,"imp":9.3725589058,"elo":1748.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0782544909,"x_train":23.9442866701,"imp":9.3921018347,"elo":1749.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.095374063,"x_train":52.2059341627,"imp":9.2916172381,"elo":1760.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1151706511,"x_train":97.6448146128,"imp":9.2566381211,"elo":1764.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1432773038,"x_train":190.15396393,"imp":9.2967413641,"elo":1758.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1564459455,"x_train":281.6941410441,"imp":9.3113324842,"elo":1756.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1678973097,"x_train":378.4994428039,"imp":9.3273441341,"elo":1753.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0229625096,"x_train":0.3843823491,"imp":12.8635534115,"elo":1590.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0474733333,"x_train":0.8231266994,"imp":12.2327635752,"elo":1643.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1047084686,"x_train":2.0711841529,"imp":11.8377408404,"elo":1670.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1608995496,"x_train":4.198580454,"imp":11.6213761336,"elo":1685.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2235785875,"x_train":11.0736254563,"imp":11.5012152174,"elo":1694.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.268731958,"x_train":22.535781385,"imp":11.4021767141,"elo":1699.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2953259759,"x_train":44.9181007897,"imp":11.2853772779,"elo":1707.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3063723891,"x_train":67.208700505,"imp":11.2408277916,"elo":1709.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3135203066,"x_train":90.1654472213,"imp":11.2307808149,"elo":1710.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0069074234,"x_train":1.4432254313,"imp":18.3289887858,"elo":1366.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0179091186,"x_train":4.2347903447,"imp":17.4720804235,"elo":1400.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0310818868,"x_train":10.3173350008,"imp":16.7542431694,"elo":1433.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0444548419,"x_train":21.8514222981,"imp":16.4427611222,"elo":1447.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0642533403,"x_train":55.5931037878,"imp":16.1033145209,"elo":1466.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0756378371,"x_train":111.097193586,"imp":15.9162222115,"elo":1477.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.08260308,"x_train":219.5483355792,"imp":15.709674653,"elo":1490.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.08968946,"x_train":330.9116540058,"imp":15.6383543743,"elo":1494.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0948963283,"x_train":442.2024676551,"imp":15.52589414,"elo":1502.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0569778809,"x_train":0.3394264702,"imp":27.1741880295,"elo":961.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0649760559,"x_train":0.561280236,"imp":24.2835226498,"elo":1037.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0823683729,"x_train":1.3509100065,"imp":19.8198892022,"elo":1167.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0925044765,"x_train":2.5531799736,"imp":18.2046872025,"elo":1232.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1048375565,"x_train":6.2677435481,"imp":17.2453969381,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1198241829,"x_train":12.4252027007,"imp":17.0238417989,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1381552995,"x_train":24.8317026638,"imp":16.8635575395,"elo":1296.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1436546335,"x_train":37.332646039,"imp":16.8714868407,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1517370302,"x_train":50.1933853639,"imp":16.9523835395,"elo":1299.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1123933797,"x_train":1.6121148448,"imp":24.2159078135,"elo":1174.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2234013195,"x_train":3.4214998387,"imp":22.7745276103,"elo":1208.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5317849911,"x_train":9.5320935926,"imp":20.9902035951,"elo":1270.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8880009049,"x_train":19.7206332887,"imp":19.9998660631,"elo":1299.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2374143618,"x_train":47.120631421,"imp":19.1321650838,"elo":1332.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4963820182,"x_train":95.0331726255,"imp":18.6307722458,"elo":1351.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.871926073,"x_train":192.9295597629,"imp":18.1898274964,"elo":1365.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.9971010285,"x_train":290.1521255355,"imp":17.891970254,"elo":1378.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0921384936,"x_train":389.6192568267,"imp":17.7004215583,"elo":1386.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.033160957,"x_train":0.2814937163,"imp":17.0899008251,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.3076291887,"x_train":0.8419891145,"imp":16.2689403096,"elo":1399.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5603369831,"x_train":2.1642156548,"imp":14.9266707853,"elo":1504.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8536127137,"x_train":4.3828192424,"imp":13.8517848473,"elo":1586.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.935960933,"x_train":11.3294435796,"imp":12.6195613378,"elo":1649.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9476843576,"x_train":23.1362359393,"imp":12.0578006895,"elo":1690.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8576552911,"x_train":46.1918681929,"imp":11.8306177522,"elo":1704.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7970201041,"x_train":68.9587529747,"imp":11.7603011738,"elo":1706.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.756028165,"x_train":93.2136338393,"imp":11.7281550364,"elo":1710.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":2.9321947194,"x_train":5.058611156,"imp":19.8907338623,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.6055455764,"x_train":10.5029291491,"imp":17.8299401868,"elo":1315.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.1645670464,"x_train":25.3830456847,"imp":15.8929011855,"elo":1407.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.5989848543,"x_train":65.7796741249,"imp":13.5293238,"elo":1549.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1554075804,"x_train":124.0730628064,"imp":12.7321692308,"elo":1595.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.8948147244,"x_train":283.3505795288,"imp":11.9230807327,"elo":1644.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.4555428919,"x_train":583.6843020141,"imp":11.5856976387,"elo":1664.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":11.9967506884,"x_train":1153.7001044299,"imp":11.4878041124,"elo":1671.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":13.3726529631,"x_train":1719.5734803795,"imp":11.4420520926,"elo":1671.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":14.1602029953,"x_train":2338.1908652586,"imp":11.4152796867,"elo":1670.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8827204019,"x_train":17.1713211146,"imp":27.4282235907,"elo":974.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0226836019,"x_train":1.6442071407,"imp":37.3438261467,"elo":804.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1912576397,"x_train":6.5769737451,"imp":34.813557491,"elo":883.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5361050995,"x_train":19.1886688806,"imp":27.6055384276,"elo":1038.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6062271231,"x_train":22.8161312471,"imp":27.4540417274,"elo":1046.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0133914852,"x_train":0.1211565882,"imp":25.1437303937,"elo":998.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.040619544,"x_train":0.39306177,"imp":20.0876024453,"elo":1167.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0822556302,"x_train":1.261863005,"imp":18.1118560726,"elo":1266.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1187605168,"x_train":2.6044216897,"imp":17.448305273,"elo":1303.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1439903713,"x_train":6.4126782162,"imp":16.9889325644,"elo":1333.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1661672354,"x_train":12.9417195679,"imp":16.8332836861,"elo":1345.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1936671381,"x_train":25.9261667663,"imp":16.7203484605,"elo":1355.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2030888401,"x_train":39.086650906,"imp":16.6653861195,"elo":1360.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.215523944,"x_train":52.4497680024,"imp":16.7072446847,"elo":1364.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.232511973,"x_train":6.9801966051,"imp":16.8241005098,"elo":1387.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.366818368,"x_train":15.4217554801,"imp":14.7107487455,"elo":1517.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5883942127,"x_train":39.0914867245,"imp":13.5019228572,"elo":1584.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.8292026744,"x_train":79.6483382915,"imp":12.714245834,"elo":1617.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.0621411706,"x_train":203.6966152959,"imp":11.9655799475,"elo":1653.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.2831539961,"x_train":410.5193453014,"imp":11.6538509223,"elo":1665.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5130210219,"x_train":808.8154712049,"imp":11.2152910283,"elo":1680.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6874000841,"x_train":1213.5228253628,"imp":11.0331396261,"elo":1690.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7535445217,"x_train":1624.5504043652,"imp":10.9669183657,"elo":1696.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8199508222,"x_train":2.2973997315,"imp":12.757790933,"elo":1576.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7781160341,"x_train":5.7558525217,"imp":12.7040141437,"elo":1584.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.528542093,"x_train":18.4525573231,"imp":12.5678686593,"elo":1599.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.5801414596,"x_train":36.2094345784,"imp":12.5003093797,"elo":1607.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.0669182071,"x_train":88.4354783372,"imp":12.1553628698,"elo":1640.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.3558901149,"x_train":183.3566697473,"imp":11.7002365934,"elo":1661.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.0333907004,"x_train":377.7921394838,"imp":11.4270632498,"elo":1669.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.9376757094,"x_train":572.0185730315,"imp":11.2797380522,"elo":1669.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":13.6073656299,"x_train":783.2895068512,"imp":11.188360367,"elo":1668.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.5537416022,"x_train":14.8539049818,"imp":19.4459273431,"elo":1297.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":45.622831341,"x_train":46.8983759891,"imp":20.7566063243,"elo":1273.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":66.3156018301,"x_train":69.8822600854,"imp":19.1332286663,"elo":1313.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":98.021147049,"x_train":127.4067152197,"imp":17.5591056151,"elo":1372.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":134.082270619,"x_train":229.7746727932,"imp":16.995992756,"elo":1392.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":215.0366892607,"x_train":589.1070615002,"imp":15.6827832788,"elo":1452.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":284.8856925256,"x_train":1193.0508698741,"imp":14.9633943872,"elo":1484.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":380.3864389015,"x_train":2388.3823364922,"imp":14.4617778749,"elo":1503.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":441.6863563071,"x_train":3593.4863060271,"imp":14.0010964181,"elo":1521.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":464.2805045187,"x_train":4805.6782161313,"imp":13.7360803063,"elo":1526.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2302397985,"x_train":0.7869111961,"imp":14.371186543,"elo":1544.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":18.7591592077,"x_train":38.8556954397,"imp":3.7240503537,"elo":2039.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0330391224,"x_train":0.1957996411,"imp":34.8895165819,"elo":880.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":3.1457473701,"x_train":7.5078983053,"imp":13.454211295,"elo":1564.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1703078893,"x_train":0.6258098026,"imp":8.7394017522,"elo":1785.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0548347304,"x_train":1.5586488241,"imp":15.9377258692,"elo":1426.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1165748723,"x_train":5.192079831,"imp":14.9598444176,"elo":1470.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2008241117,"x_train":19.2797854853,"imp":13.7607602335,"elo":1531.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2729097923,"x_train":41.1266447543,"imp":13.0968145146,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3551141856,"x_train":106.1594477641,"imp":12.4863504012,"elo":1615.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4346764878,"x_train":213.4318571252,"imp":12.1925610385,"elo":1638.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4982388933,"x_train":422.7331976766,"imp":11.9493041925,"elo":1654.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.5348355254,"x_train":638.9519356459,"imp":11.8594998702,"elo":1662.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5548771281,"x_train":860.2847801702,"imp":11.8049633358,"elo":1670.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7011909333,"x_train":2.4084538179,"imp":9.7560536872,"elo":1801.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2888892556,"x_train":1.0968929682,"imp":7.8824971032,"elo":1878.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.898348797,"x_train":86.2757906839,"imp":32.4967224715,"elo":934.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.938220505,"x_train":96.7574708184,"imp":32.41479258,"elo":949.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.0360121046,"x_train":132.620297349,"imp":32.2594861504,"elo":955.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.3732929292,"x_train":182.3615021409,"imp":31.786012596,"elo":975.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9911165015,"x_train":379.7267185426,"imp":30.7058200242,"elo":1005.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4182120307,"x_train":737.6739956698,"imp":29.3722505198,"elo":1037.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2639960398,"x_train":1465.1527881613,"imp":28.4744557852,"elo":1064.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9050217868,"x_train":2179.2889039095,"imp":27.6256143537,"elo":1092.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9836310881,"x_train":2951.7955583902,"imp":26.8746550395,"elo":1121.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1693477495,"x_train":0.6181258773,"imp":15.5418068987,"elo":1445.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0462763773,"x_train":1.8659541839,"imp":19.6714412852,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0986462001,"x_train":6.8422781275,"imp":18.2754375515,"elo":1279.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2529547278,"x_train":28.1225860036,"imp":16.8936521045,"elo":1343.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4092168594,"x_train":62.6333146165,"imp":16.0329912533,"elo":1393.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6013598742,"x_train":153.2940248569,"imp":15.1322927453,"elo":1438.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8316842129,"x_train":287.4960670101,"imp":14.5504300077,"elo":1473.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.09015958,"x_train":565.1252028376,"imp":14.1383335222,"elo":1498.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2700433543,"x_train":871.368400213,"imp":13.8720944137,"elo":1513.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4406412634,"x_train":1169.3150011596,"imp":13.597536455,"elo":1523.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.052450313,"x_train":0.4700674902,"imp":14.3359708531,"elo":1484.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1462358449,"x_train":2.0561352068,"imp":12.8895219142,"elo":1600.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2538924294,"x_train":5.7242090848,"imp":12.2710877404,"elo":1655.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2894544358,"x_train":9.9945517672,"imp":11.8250868461,"elo":1683.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3450335972,"x_train":26.428780062,"imp":11.5620514204,"elo":1698.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3624432873,"x_train":52.4146823069,"imp":11.4682434169,"elo":1704.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3639681589,"x_train":103.8233704142,"imp":11.4119948439,"elo":1710.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3851024669,"x_train":154.5284843647,"imp":11.4080903645,"elo":1713.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.398537392,"x_train":207.6491745508,"imp":11.4089279264,"elo":1716.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":8.0432165428,"x_train":68.9669811874,"imp":19.6910335019,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":15.6237886223,"x_train":130.6527630186,"imp":17.5680857216,"elo":1390.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":29.3242995716,"x_train":316.404564055,"imp":14.9245895934,"elo":1499.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":45.5811393322,"x_train":624.6899718112,"imp":12.8882163313,"elo":1603.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":71.7349528943,"x_train":1547.1736962358,"imp":10.591051261,"elo":1743.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":73.3251138416,"x_train":1607.7353976095,"imp":10.5488846537,"elo":1749.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.2236113698,"x_train":2.8035263333,"imp":33.6178632238,"elo":926.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.390344083,"x_train":6.8864266433,"imp":27.0655054285,"elo":1049.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.6801837529,"x_train":17.5478119749,"imp":19.7955614958,"elo":1258.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.9680498121,"x_train":35.2197842651,"imp":16.9393451699,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.4009052838,"x_train":88.4231720748,"imp":15.9543591258,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.8435618428,"x_train":176.2334816176,"imp":15.5960824642,"elo":1464.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.3657351437,"x_train":353.6101338707,"imp":15.4855978336,"elo":1463.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.6852473948,"x_train":529.672353661,"imp":15.4670973384,"elo":1461.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.884834451,"x_train":709.8550005699,"imp":15.5056797319,"elo":1454.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..6a41fd3a41324c3e469c51121bdbe75f253d8d89
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_classification/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "TabFM (default)", "AutoGluon 1.4 (best, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "CatBoost (tuned + ensembled)", "CatBoost (default)", "iLTM (tuned + ensembled)", "CatBoost (tuned)", "XGBoost (tuned + ensembled)", "LightGBM (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "RealMLP (tuned + ensembled)", "RealTabPFN-2.5 (tuned + ensembled)", "TabM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "ChimeraBoost (tuned)", "XGBoost (tuned)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned)", "TabM (tuned)", "LightGBM (tuned)", "ChimeraBoost (default)", "RealTabPFN-2.5 (default)", "TabICL (default)", "iLTM (tuned)", "ModernNCA (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "EBM (tuned + ensembled)", "XGBoost (default)", "xRFM (tuned + ensembled)", "TabSwift (default)", "EBM (tuned)", "TabM (default)", "xRFM (tuned)", "RealMLP (default)", "BetaTabPFN (default)", "TorchMLP (tuned)", "FastaiMLP (tuned + ensembled)", "EBM (default)", "TabDPT (tuned)", "RandomForest (tuned + ensembled)", "iLTM (default)", "LightGBM (default)", "LimiX (default)", "RandomForest (tuned)", "ModernNCA (default)", "ExtraTrees (tuned + ensembled)", "SAP-RPT-OSS (default)", "TabDPT (default)", "FastaiMLP (tuned)", "ExtraTrees (tuned)", "TorchMLP (default)", "FastaiMLP (default)", "TabSTAR (tuned)", "Linear (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "KNN (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "RandomForest (default)", "PerpetualBooster (tuned)", "OrionMSP (default)", "ExtraTrees (default)", "Linear (default)", "TabSTAR (default)", "xRFM (default)", "TabFlex (default)", "KNN (tuned)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.4000000059604645, 0.7444444298744202, 0.4444444477558136, 0.8222222328186035, 0.6777777671813965, 0.9222221970558167, 0.8999999761581421, 0.9333333373069763, 0.9111111164093018, 0.9555555582046509, 0.9333333373069763, 0.9222221970558167, 0.9777777791023254, 0.8999999761581421, 0.8111110925674438, 0.9666666388511658, 0.9333333373069763, 0.9888888597488403, 0.9222221970558167, 0.9666666388511658, 0.8666666746139526, 0.9777777791023254, 0.9666666388511658, 0.9222221970558167, 0.8999999761581421, 0.8888888955116272, 0.9888888597488403, 0.9111111164093018, 0.9222221970558167, 0.9888888597488403, 0.9888888597488403, 0.9666666388511658, 0.9777777791023254, 0.9444444179534912, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 0.8888888955116272, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.6000000238418579, NaN, 0.6333333253860474, 0.8111110925674438, 0.8999999761581421, 0.9111111164093018, 0.7777777910232544, 0.7888888716697693, 0.8444444537162781, 0.8222222328186035, 0.8777777552604675, 0.855555534362793, 0.8777777552604675, 0.8222222328186035, 0.8666666746139526, 0.9888888597488403, 0.8888888955116272, 0.8777777552604675, 0.9111111164093018, 0.8777777552604675, 0.8333333134651184, 1.0, 0.9666666388511658, 0.8999999761581421, 0.9777777791023254, 0.9555555582046509, 0.8999999761581421, 0.9444444179534912, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9888888597488403, 0.9777777791023254, 0.8888888955116272, 0.9555555582046509, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 0.9111111164093018, 1.0, 0.9666666388511658, 0.9777777791023254, 1.0, 1.0, 0.855555534362793, 0.9777777791023254, 1.0, 0.9888888597488403, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9888888597488403, 1.0, 0.9777777791023254, 1.0, 0.9777777791023254, 1.0, 0.9555555582046509, 0.9777777791023254, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0], [0.25555557012557983, 0.36666667461395264, NaN, 0.42222222685813904, 0.7111111283302307, 0.6555555462837219, 0.7666666507720947, 0.7222222089767456, 0.8444444537162781, 0.7777777910232544, 0.8666666746139526, 0.8444444537162781, 0.7777777910232544, 0.8666666746139526, 0.644444465637207, 0.7444444298744202, 0.7888888716697693, 0.8666666746139526, 0.8888888955116272, 0.7444444298744202, 0.9222221970558167, 0.800000011920929, 0.8999999761581421, 0.8888888955116272, 0.7666666507720947, 0.7666666507720947, 0.855555534362793, 0.8999999761581421, 0.8111110925674438, 0.800000011920929, 0.9111111164093018, 0.9555555582046509, 0.9222221970558167, 0.9555555582046509, 0.8444444537162781, 0.9555555582046509, 0.9222221970558167, 0.9666666388511658, 0.9666666388511658, 0.9111111164093018, 0.9777777791023254, 0.9666666388511658, 0.9555555582046509, 0.9666666388511658, 0.9888888597488403, 0.9666666388511658, 0.9555555582046509, 0.8777777552604675, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.9666666388511658, 0.9555555582046509, 0.9777777791023254, 1.0, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.9777777791023254, 1.0, 0.9666666388511658, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5555555820465088, 0.18888889253139496, 0.5777778029441833, NaN, 0.7555555701255798, 0.5777778029441833, 0.699999988079071, 0.6888889074325562, 0.7111111283302307, 0.6888889074325562, 0.7444444298744202, 0.7333333492279053, 0.7777777910232544, 0.7444444298744202, 0.855555534362793, 0.7888888716697693, 0.8888888955116272, 0.7666666507720947, 0.7777777910232544, 0.8777777552604675, 0.7555555701255798, 0.855555534362793, 0.8999999761581421, 0.7666666507720947, 0.8777777552604675, 0.8666666746139526, 0.8666666746139526, 0.8777777552604675, 0.9555555582046509, 0.8888888955116272, 0.8333333134651184, 0.8666666746139526, 0.9111111164093018, 0.7888888716697693, 0.9666666388511658, 0.9111111164093018, 0.9888888597488403, 0.800000011920929, 0.9444444179534912, 0.9444444179534912, 0.9666666388511658, 0.8999999761581421, 0.9555555582046509, 0.9666666388511658, 0.9444444179534912, 0.9888888597488403, 0.9888888597488403, 0.800000011920929, 0.9666666388511658, 0.9777777791023254, 0.9666666388511658, 0.9222221970558167, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 1.0, 0.9222221970558167, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 0.9222221970558167, 1.0, 1.0, 1.0, 1.0], [0.17777778208255768, 0.10000000149011612, 0.2888889014720917, 0.24444444477558136, NaN, 0.5666666626930237, 0.5666666626930237, 0.5222222208976746, 0.5555555820465088, 0.5666666626930237, 0.5777778029441833, 0.6888889074325562, 0.6777777671813965, 0.6222222447395325, 0.699999988079071, 0.6222222447395325, 0.699999988079071, 0.7111111283302307, 0.7333333492279053, 0.699999988079071, 0.7111111283302307, 0.7444444298744202, 0.8222222328186035, 0.6777777671813965, 0.855555534362793, 0.7777777910232544, 0.7555555701255798, 0.8222222328186035, 0.855555534362793, 0.855555534362793, 0.8444444537162781, 0.8777777552604675, 0.9111111164093018, 0.7666666507720947, 0.8666666746139526, 0.9666666388511658, 0.9222221970558167, 0.8111110925674438, 1.0, 0.8999999761581421, 0.9555555582046509, 0.8666666746139526, 0.9666666388511658, 0.9555555582046509, 0.9555555582046509, 0.9444444179534912, 0.9444444179534912, 0.7777777910232544, 0.9888888597488403, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 0.9444444179534912, 0.9444444179534912, 0.9888888597488403, 1.0, 0.9222221970558167, 1.0, 0.9666666388511658, 1.0, 0.9444444179534912, 0.9666666388511658, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9666666388511658, 1.0, 0.9222221970558167, 1.0, 1.0, 1.0, 1.0], [0.3222222328186035, 0.08888889104127884, 0.3444444537162781, 0.42222222685813904, 0.4333333373069763, NaN, 0.6111111044883728, 0.6111111044883728, 0.6333333253860474, 0.6000000238418579, 0.6222222447395325, 0.6666666865348816, 0.6222222447395325, 0.5777778029441833, 0.7333333492279053, 0.6111111044883728, 0.7111111283302307, 0.6222222447395325, 0.644444465637207, 0.7555555701255798, 0.6333333253860474, 0.7222222089767456, 0.7666666507720947, 0.644444465637207, 0.8444444537162781, 0.8333333134651184, 0.7555555701255798, 0.800000011920929, 0.8777777552604675, 0.8999999761581421, 0.8111110925674438, 0.7555555701255798, 0.8666666746139526, 0.699999988079071, 0.9111111164093018, 0.8999999761581421, 0.9444444179534912, 0.7222222089767456, 0.9111111164093018, 0.9222221970558167, 0.9111111164093018, 0.8444444537162781, 0.9333333373069763, 0.9444444179534912, 0.9555555582046509, 0.9222221970558167, 0.9444444179534912, 0.8111110925674438, 0.9666666388511658, 0.9555555582046509, 0.9888888597488403, 0.9222221970558167, 0.9555555582046509, 0.9333333373069763, 0.9888888597488403, 0.9777777791023254, 0.9222221970558167, 0.9777777791023254, 0.9222221970558167, 0.9777777791023254, 0.9555555582046509, 0.9444444179534912, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9666666388511658, 1.0, 0.9222221970558167, 1.0, 1.0, 1.0, 1.0], [0.07777778059244156, 0.2222222238779068, 0.23333333432674408, 0.30000001192092896, 0.4333333373069763, 0.3888888955116272, NaN, 0.5888888835906982, 0.5777778029441833, 0.5888888835906982, 0.6111111044883728, 0.5111111402511597, 0.5888888835906982, 0.6222222447395325, 0.5222222208976746, 0.6000000238418579, 0.5888888835906982, 0.7333333492279053, 0.6888889074325562, 0.5333333611488342, 0.7111111283302307, 0.644444465637207, 0.644444465637207, 0.7888888716697693, 0.7333333492279053, 0.6888889074325562, 0.7444444298744202, 0.7388888597488403, 0.6000000238418579, 0.6666666865348816, 0.7888888716697693, 0.8444444537162781, 0.8222222328186035, 0.9222221970558167, 0.7777777910232544, 0.8222222328186035, 0.7666666507720947, 0.9444444179534912, 0.8333333134651184, 0.8444444537162781, 0.9555555582046509, 0.8666666746139526, 0.9444444179534912, 0.8111110925674438, 0.8888888955116272, 0.9222221970558167, 0.9555555582046509, 0.7222222089767456, 0.8777777552604675, 0.9555555582046509, 0.9888888597488403, 0.9777777791023254, 0.8888888955116272, 0.9666666388511658, 0.9777777791023254, 0.9888888597488403, 0.9777777791023254, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0], [0.10000000149011612, 0.21111111342906952, 0.2777777910232544, 0.31111112236976624, 0.47777777910232544, 0.3888888955116272, 0.41111111640930176, NaN, 0.5222222208976746, 0.47777777910232544, 0.5333333611488342, 0.5444444417953491, 0.6000000238418579, 0.6222222447395325, 0.5222222208976746, 0.6000000238418579, 0.5888888835906982, 0.7222222089767456, 0.6555555462837219, 0.5444444417953491, 0.6777777671813965, 0.6555555462837219, 0.6888889074325562, 0.8111110925674438, 0.7111111283302307, 0.7111111283302307, 0.7555555701255798, 0.7111111283302307, 0.5888888835906982, 0.6888889074325562, 0.7888888716697693, 0.8111110925674438, 0.7888888716697693, 0.9444444179534912, 0.8111110925674438, 0.8111110925674438, 0.7888888716697693, 0.9555555582046509, 0.8222222328186035, 0.855555534362793, 0.9555555582046509, 0.8333333134651184, 0.8888888955116272, 0.8111110925674438, 0.8777777552604675, 0.9444444179534912, 0.9333333373069763, 0.7555555701255798, 0.8777777552604675, 0.9888888597488403, 0.9666666388511658, 0.9555555582046509, 0.8999999761581421, 0.9777777791023254, 0.9888888597488403, 0.9777777791023254, 0.9555555582046509, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0], [0.06666667014360428, 0.15555556118488312, 0.15555556118488312, 0.2888889014720917, 0.4444444477558136, 0.36666667461395264, 0.42222222685813904, 0.47777777910232544, NaN, 0.4444444477558136, 0.5333333611488342, 0.5666666626930237, 0.5, 0.5444444417953491, 0.5777778029441833, 0.5777778029441833, 0.5666666626930237, 0.5888888835906982, 0.6777777671813965, 0.6222222447395325, 0.6555555462837219, 0.6555555462837219, 0.6777777671813965, 0.644444465637207, 0.7333333492279053, 0.7333333492279053, 0.7111111283302307, 0.7777777910232544, 0.6111111044883728, 0.7444444298744202, 0.8666666746139526, 0.8111110925674438, 0.7888888716697693, 0.8888888955116272, 0.7666666507720947, 0.8222222328186035, 0.8333333134651184, 0.9555555582046509, 0.9555555582046509, 0.8666666746139526, 0.9555555582046509, 0.9666666388511658, 0.9111111164093018, 0.9111111164093018, 0.8777777552604675, 1.0, 0.9222221970558167, 0.7555555701255798, 0.8888888955116272, 0.9777777791023254, 0.9555555582046509, 0.9777777791023254, 0.9333333373069763, 0.9666666388511658, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9666666388511658, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.08888889104127884, 0.17777778208255768, 0.2222222238779068, 0.31111112236976624, 0.4333333373069763, 0.4000000059604645, 0.41111111640930176, 0.5222222208976746, 0.5555555820465088, NaN, 0.6222222447395325, 0.46666666865348816, 0.6222222447395325, 0.5777778029441833, 0.4888888895511627, 0.6111111044883728, 0.6000000238418579, 0.7333333492279053, 0.7111111283302307, 0.5333333611488342, 0.699999988079071, 0.6222222447395325, 0.6666666865348816, 0.7555555701255798, 0.6888889074325562, 0.699999988079071, 0.7111111283302307, 0.6888889074325562, 0.5888888835906982, 0.6666666865348816, 0.7777777910232544, 0.8333333134651184, 0.8111110925674438, 0.9333333373069763, 0.7666666507720947, 0.8222222328186035, 0.800000011920929, 0.9333333373069763, 0.8444444537162781, 0.855555534362793, 0.9444444179534912, 0.8222222328186035, 0.9666666388511658, 0.800000011920929, 0.8888888955116272, 0.9333333373069763, 0.9222221970558167, 0.7222222089767456, 0.8777777552604675, 0.9333333373069763, 0.9555555582046509, 0.9666666388511658, 0.8777777552604675, 0.9555555582046509, 0.9666666388511658, 0.9777777791023254, 0.9444444179534912, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0], [0.04444444552063942, 0.12222222238779068, 0.13333334028720856, 0.25555557012557983, 0.42222222685813904, 0.3777777850627899, 0.3888888955116272, 0.46666666865348816, 0.46666666865348816, 0.3777777850627899, NaN, 0.4555555582046509, 0.5222222208976746, 0.5, 0.5, 0.6222222447395325, 0.5222222208976746, 0.5777778029441833, 0.8666666746139526, 0.5333333611488342, 0.5888888835906982, 0.6777777671813965, 0.6222222447395325, 0.6666666865348816, 0.6555555462837219, 0.6000000238418579, 0.6333333253860474, 0.644444465637207, 0.6222222447395325, 0.6888889074325562, 0.800000011920929, 0.8111110925674438, 0.9333333373069763, 0.8222222328186035, 0.6888889074325562, 0.8444444537162781, 0.8111110925674438, 0.9111111164093018, 0.8444444537162781, 0.8777777552604675, 0.8444444537162781, 0.800000011920929, 0.9777777791023254, 0.8444444537162781, 0.8999999761581421, 0.9666666388511658, 0.9666666388511658, 0.6777777671813965, 0.8999999761581421, 0.9888888597488403, 0.9888888597488403, 0.9555555582046509, 0.8888888955116272, 0.8777777552604675, 0.9888888597488403, 1.0, 0.8999999761581421, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9888888597488403, 1.0, 0.9222221970558167, 1.0, 0.9666666388511658, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.14444445073604584, 0.15555556118488312, 0.2666666805744171, 0.31111112236976624, 0.3333333432674408, 0.4888888895511627, 0.4555555582046509, 0.4333333373069763, 0.5333333611488342, 0.5444444417953491, NaN, 0.47777777910232544, 0.47777777910232544, 0.46666666865348816, 0.5333333611488342, 0.5111111402511597, 0.6111111044883728, 0.6888889074325562, 0.5, 0.5666666626930237, 0.6333333253860474, 0.8666666746139526, 0.6222222447395325, 0.6222222447395325, 0.6666666865348816, 0.699999988079071, 0.6555555462837219, 0.6666666865348816, 0.6111111044883728, 0.7333333492279053, 0.7666666507720947, 0.8222222328186035, 0.7666666507720947, 0.7444444298744202, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 0.8222222328186035, 0.8666666746139526, 0.9111111164093018, 0.7666666507720947, 0.9666666388511658, 0.800000011920929, 0.8999999761581421, 0.9777777791023254, 0.9777777791023254, 0.6888889074325562, 0.9111111164093018, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 0.9555555582046509, 0.9111111164093018, 0.9777777791023254, 0.9777777791023254, 0.8666666746139526, 0.9888888597488403, 0.9777777791023254, 0.9888888597488403, 0.9777777791023254, 0.9888888597488403, 0.9777777791023254, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0], [0.07777778059244156, 0.12222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3222222328186035, 0.3777777850627899, 0.41111111640930176, 0.4000000059604645, 0.5, 0.3777777850627899, 0.47777777910232544, 0.5222222208976746, NaN, 0.6222222447395325, 0.5111111402511597, 0.6111111044883728, 0.5222222208976746, 0.7555555701255798, 0.5888888835906982, 0.5666666626930237, 0.7222222089767456, 0.6888889074325562, 0.5666666626930237, 0.8777777552604675, 0.6333333253860474, 0.6333333253860474, 0.7333333492279053, 0.7111111283302307, 0.6111111044883728, 0.699999988079071, 0.7111111283302307, 0.8222222328186035, 0.6777777671813965, 0.8999999761581421, 0.6666666865348816, 0.855555534362793, 0.8222222328186035, 0.9222221970558167, 0.855555534362793, 0.8777777552604675, 0.8666666746139526, 0.855555534362793, 0.9333333373069763, 0.8333333134651184, 0.8888888955116272, 0.9555555582046509, 0.8777777552604675, 0.7444444298744202, 0.8888888955116272, 0.9111111164093018, 0.8888888955116272, 0.9777777791023254, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.9111111164093018, 0.8777777552604675, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.8777777552604675, 1.0, 0.8888888955116272, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9222221970558167, 1.0], [0.02222222276031971, 0.17777778208255768, 0.13333334028720856, 0.25555557012557983, 0.3777777850627899, 0.42222222685813904, 0.3777777850627899, 0.3777777850627899, 0.4555555582046509, 0.42222222685813904, 0.5, 0.5222222208976746, 0.3777777850627899, NaN, 0.5555555820465088, 0.42222222685813904, 0.5333333611488342, 0.4888888895511627, 0.6000000238418579, 0.5888888835906982, 0.7555555701255798, 0.47777777910232544, 0.6777777671813965, 0.4888888895511627, 0.6111111044883728, 0.5777778029441833, 0.6222222447395325, 0.6888889074325562, 0.6666666865348816, 0.6111111044883728, 0.7555555701255798, 0.7555555701255798, 0.7444444298744202, 0.8444444537162781, 0.6555555462837219, 0.8222222328186035, 0.8999999761581421, 0.8888888955116272, 1.0, 0.855555534362793, 0.8333333134651184, 0.9777777791023254, 0.8444444537162781, 0.8111110925674438, 0.8777777552604675, 0.8111110925674438, 0.9333333373069763, 0.8444444537162781, 0.8666666746139526, 0.855555534362793, 0.8999999761581421, 0.8888888955116272, 0.8222222328186035, 0.9777777791023254, 0.8777777552604675, 0.9888888597488403, 0.9666666388511658, 1.0, 1.0, 1.0, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9333333373069763, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.13333334028720856, 0.35555556416511536, 0.14444445073604584, 0.30000001192092896, 0.2666666805744171, 0.47777777910232544, 0.47777777910232544, 0.42222222685813904, 0.5111111402511597, 0.5, 0.5333333611488342, 0.4888888895511627, 0.4444444477558136, NaN, 0.4333333373069763, 0.46666666865348816, 0.5222222208976746, 0.5444444417953491, 0.800000011920929, 0.5555555820465088, 0.5555555820465088, 0.644444465637207, 0.6111111044883728, 0.7111111283302307, 0.699999988079071, 0.6222222447395325, 0.6888889074325562, 0.644444465637207, 0.6888889074325562, 0.7111111283302307, 0.7333333492279053, 0.7666666507720947, 0.6555555462837219, 0.855555534362793, 0.8222222328186035, 0.7444444298744202, 0.7111111283302307, 0.8222222328186035, 0.800000011920929, 0.8222222328186035, 0.7333333492279053, 0.8666666746139526, 0.8888888955116272, 0.9111111164093018, 0.855555534362793, 0.9222221970558167, 0.6666666865348816, 0.9555555582046509, 0.8222222328186035, 0.9666666388511658, 0.855555534362793, 0.8999999761581421, 0.8777777552604675, 0.9666666388511658, 0.9555555582046509, 0.8999999761581421, 1.0, 0.8444444537162781, 1.0, 0.9111111164093018, 0.855555534362793, 0.9666666388511658, 1.0, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.9555555582046509, 1.0, 0.9222221970558167, 1.0, 1.0, 1.0, 1.0], [0.18888889253139496, 0.011111111380159855, 0.25555557012557983, 0.21111111342906952, 0.3777777850627899, 0.3888888955116272, 0.4000000059604645, 0.4000000059604645, 0.42222222685813904, 0.3888888955116272, 0.3777777850627899, 0.46666666865348816, 0.3888888955116272, 0.5777778029441833, 0.5666666626930237, NaN, 0.5222222208976746, 0.4555555582046509, 0.5, 0.5888888835906982, 0.6555555462837219, 0.7333333492279053, 0.6111111044883728, 0.5222222208976746, 0.5666666626930237, 0.5777778029441833, 0.6333333253860474, 0.7333333492279053, 0.7111111283302307, 0.6555555462837219, 0.6111111044883728, 0.6666666865348816, 0.6666666865348816, 0.6555555462837219, 0.7555555701255798, 0.8444444537162781, 0.9666666388511658, 0.6555555462837219, 0.9777777791023254, 0.9111111164093018, 0.7777777910232544, 0.8222222328186035, 0.8333333134651184, 0.8777777552604675, 0.8333333134651184, 0.8333333134651184, 1.0, 0.7666666507720947, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.8777777552604675, 0.8777777552604675, 0.9111111164093018, 0.8666666746139526, 0.9777777791023254, 0.8999999761581421, 1.0, 0.9333333373069763, 1.0, 0.9444444179534912, 0.9444444179534912, 0.9777777791023254, 0.9888888597488403, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0], [0.03333333507180214, 0.1111111119389534, 0.21111111342906952, 0.1111111119389534, 0.30000001192092896, 0.2888889014720917, 0.41111111640930176, 0.41111111640930176, 0.4333333373069763, 0.4000000059604645, 0.47777777910232544, 0.4888888895511627, 0.47777777910232544, 0.46666666865348816, 0.5333333611488342, 0.47777777910232544, NaN, 0.5555555820465088, 0.5, 0.5444444417953491, 0.5888888835906982, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.6222222447395325, 0.6333333253860474, 0.644444465637207, 0.7444444298744202, 0.6111111044883728, 0.6888889074325562, 0.7333333492279053, 0.7333333492279053, 0.6555555462837219, 0.7111111283302307, 0.699999988079071, 0.7555555701255798, 0.8222222328186035, 0.7777777910232544, 0.8888888955116272, 0.8222222328186035, 0.8444444537162781, 0.8666666746139526, 0.8222222328186035, 0.8777777552604675, 0.7888888716697693, 0.8444444537162781, 0.9111111164093018, 0.855555534362793, 0.7777777910232544, 0.9555555582046509, 0.8999999761581421, 0.855555534362793, 0.9222221970558167, 0.9333333373069763, 0.8888888955116272, 0.8888888955116272, 0.9666666388511658, 0.9222221970558167, 0.8999999761581421, 0.9222221970558167, 0.9888888597488403, 0.9111111164093018, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.06666667014360428, 0.12222222238779068, 0.13333334028720856, 0.23333333432674408, 0.2888889014720917, 0.3777777850627899, 0.2666666805744171, 0.2777777910232544, 0.41111111640930176, 0.2666666805744171, 0.42222222685813904, 0.3888888955116272, 0.24444444477558136, 0.5111111402511597, 0.47777777910232544, 0.5444444417953491, 0.4444444477558136, NaN, 0.5222222208976746, 0.4888888895511627, 0.6777777671813965, 0.6000000238418579, 0.5111111402511597, 0.7111111283302307, 0.5777778029441833, 0.5777778029441833, 0.644444465637207, 0.6666666865348816, 0.5222222208976746, 0.6555555462837219, 0.644444465637207, 0.7222222089767456, 0.6555555462837219, 0.8666666746139526, 0.6333333253860474, 0.7777777910232544, 0.7888888716697693, 0.8999999761581421, 0.8444444537162781, 0.855555534362793, 0.8222222328186035, 0.8111110925674438, 0.855555534362793, 0.8111110925674438, 0.8666666746139526, 0.8999999761581421, 0.8111110925674438, 0.7333333492279053, 0.8666666746139526, 0.855555534362793, 0.8666666746139526, 0.9333333373069763, 0.7777777910232544, 0.855555534362793, 0.8666666746139526, 0.9111111164093018, 0.8666666746139526, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9777777791023254, 1.0, 0.8777777552604675, 0.9888888597488403, 0.8777777552604675, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0], [0.011111111380159855, 0.08888889104127884, 0.1111111119389534, 0.2222222238779068, 0.2666666805744171, 0.35555556416511536, 0.31111112236976624, 0.3444444537162781, 0.3222222328186035, 0.2888889014720917, 0.13333334028720856, 0.31111112236976624, 0.41111111640930176, 0.4000000059604645, 0.4555555582046509, 0.5, 0.5, 0.47777777910232544, NaN, 0.46666666865348816, 0.46666666865348816, 0.6111111044883728, 0.5111111402511597, 0.6000000238418579, 0.5666666626930237, 0.5444444417953491, 0.5666666626930237, 0.6333333253860474, 0.5777778029441833, 0.6111111044883728, 0.6777777671813965, 0.7222222089767456, 0.7666666507720947, 0.7444444298744202, 0.6666666865348816, 0.8111110925674438, 0.7777777910232544, 0.8111110925674438, 0.8222222328186035, 0.855555534362793, 0.8222222328186035, 0.7555555701255798, 0.9222221970558167, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.6555555462837219, 0.855555534362793, 0.9666666388511658, 0.9555555582046509, 0.9444444179534912, 0.855555534362793, 0.8666666746139526, 0.9777777791023254, 1.0, 0.855555534362793, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9888888597488403, 1.0, 0.9111111164093018, 1.0, 0.9666666388511658, 1.0, 0.9888888597488403, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0], [0.07777778059244156, 0.12222222238779068, 0.25555557012557983, 0.12222222238779068, 0.30000001192092896, 0.24444444477558136, 0.46666666865348816, 0.4555555582046509, 0.3777777850627899, 0.46666666865348816, 0.46666666865348816, 0.5, 0.4333333373069763, 0.41111111640930176, 0.20000000298023224, 0.41111111640930176, 0.4555555582046509, 0.5111111402511597, 0.5333333611488342, NaN, 0.5222222208976746, 0.5333333611488342, 0.644444465637207, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.5777778029441833, 0.6111111044883728, 0.5888888835906982, 0.6888889074325562, 0.6555555462837219, 0.6666666865348816, 0.7555555701255798, 0.6555555462837219, 0.8333333134651184, 0.7333333492279053, 0.6777777671813965, 0.6888889074325562, 0.7666666507720947, 0.7555555701255798, 0.8111110925674438, 0.699999988079071, 0.8333333134651184, 0.8222222328186035, 0.855555534362793, 0.8222222328186035, 0.8888888955116272, 0.6555555462837219, 0.9333333373069763, 0.7888888716697693, 0.9111111164093018, 0.8333333134651184, 0.8444444537162781, 0.8777777552604675, 0.9333333373069763, 0.9111111164093018, 0.8888888955116272, 0.9888888597488403, 0.8222222328186035, 1.0, 0.8888888955116272, 0.8666666746139526, 0.9666666388511658, 0.9888888597488403, 0.9444444179534912, 0.9888888597488403, 0.9777777791023254, 0.9333333373069763, 1.0, 0.8999999761581421, 0.9888888597488403, 1.0, 1.0, 1.0], [0.03333333507180214, 0.1666666716337204, 0.07777778059244156, 0.24444444477558136, 0.2888889014720917, 0.36666667461395264, 0.2888889014720917, 0.3222222328186035, 0.3444444537162781, 0.30000001192092896, 0.41111111640930176, 0.4333333373069763, 0.2777777910232544, 0.24444444477558136, 0.4444444477558136, 0.3444444537162781, 0.41111111640930176, 0.3222222328186035, 0.5333333611488342, 0.47777777910232544, NaN, 0.4000000059604645, 0.6000000238418579, 0.4333333373069763, 0.5222222208976746, 0.47777777910232544, 0.5777778029441833, 0.5777778029441833, 0.4888888895511627, 0.5555555820465088, 0.6777777671813965, 0.644444465637207, 0.6777777671813965, 0.7555555701255798, 0.5444444417953491, 0.7111111283302307, 0.7666666507720947, 0.800000011920929, 0.8999999761581421, 0.7777777910232544, 0.8111110925674438, 0.8777777552604675, 0.7333333492279053, 0.7333333492279053, 0.8444444537162781, 0.7333333492279053, 0.8222222328186035, 0.800000011920929, 0.8222222328186035, 0.7555555701255798, 0.800000011920929, 0.7888888716697693, 0.7444444298744202, 0.9666666388511658, 0.855555534362793, 0.9333333373069763, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9666666388511658, 0.9888888597488403, 0.9666666388511658, 0.9111111164093018, 0.9555555582046509, 0.9777777791023254, 0.9888888597488403, 1.0, 1.0, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.20000000298023224, 0.14444445073604584, 0.25555557012557983, 0.2777777910232544, 0.35555556416511536, 0.3444444537162781, 0.3444444537162781, 0.3777777850627899, 0.3222222328186035, 0.36666667461395264, 0.31111112236976624, 0.5222222208976746, 0.4444444477558136, 0.2666666805744171, 0.4333333373069763, 0.4000000059604645, 0.3888888955116272, 0.46666666865348816, 0.6000000238418579, NaN, 0.4444444477558136, 0.41111111640930176, 0.4888888895511627, 0.4888888895511627, 0.5888888835906982, 0.6333333253860474, 0.5888888835906982, 0.5444444417953491, 0.5666666626930237, 0.5555555820465088, 0.6555555462837219, 0.6111111044883728, 0.6555555462837219, 0.7666666507720947, 0.8444444537162781, 0.6555555462837219, 0.8999999761581421, 0.800000011920929, 0.7444444298744202, 0.800000011920929, 0.7888888716697693, 0.7666666507720947, 0.7777777910232544, 0.7555555701255798, 0.8666666746139526, 0.7666666507720947, 0.7888888716697693, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8777777552604675, 0.8333333134651184, 0.9777777791023254, 0.9111111164093018, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8999999761581421, 0.9777777791023254, 0.9666666388511658, 0.9888888597488403, 1.0, 0.9777777791023254, 0.9333333373069763, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.02222222276031971, 0.03333333507180214, 0.10000000149011612, 0.10000000149011612, 0.17777778208255768, 0.23333333432674408, 0.35555556416511536, 0.31111112236976624, 0.3222222328186035, 0.3333333432674408, 0.3777777850627899, 0.13333334028720856, 0.4333333373069763, 0.3222222328186035, 0.35555556416511536, 0.3888888955116272, 0.46666666865348816, 0.4888888895511627, 0.4888888895511627, 0.35555556416511536, 0.4000000059604645, 0.5555555820465088, NaN, 0.5444444417953491, 0.4888888895511627, 0.5222222208976746, 0.5444444417953491, 0.5555555820465088, 0.6000000238418579, 0.5777778029441833, 0.5777778029441833, 0.6777777671813965, 0.7444444298744202, 0.7222222089767456, 0.6333333253860474, 0.7333333492279053, 0.7555555701255798, 0.7222222089767456, 0.7666666507720947, 0.7888888716697693, 0.699999988079071, 0.7555555701255798, 0.855555534362793, 0.7444444298744202, 0.7888888716697693, 0.9333333373069763, 0.9666666388511658, 0.644444465637207, 0.8444444537162781, 0.9333333373069763, 0.8333333134651184, 0.8222222328186035, 0.8333333134651184, 0.7888888716697693, 0.9222221970558167, 0.9333333373069763, 0.8111110925674438, 1.0, 0.8666666746139526, 1.0, 0.9111111164093018, 0.9111111164093018, 0.9444444179534912, 0.9888888597488403, 0.9666666388511658, 0.9888888597488403, 0.9888888597488403, 0.9666666388511658, 1.0, 0.9111111164093018, 0.9888888597488403, 1.0, 1.0, 1.0], [0.03333333507180214, 0.10000000149011612, 0.1111111119389534, 0.23333333432674408, 0.3222222328186035, 0.35555556416511536, 0.21111111342906952, 0.18888889253139496, 0.35555556416511536, 0.24444444477558136, 0.3333333432674408, 0.3777777850627899, 0.12222222238779068, 0.5111111402511597, 0.3888888955116272, 0.47777777910232544, 0.36666667461395264, 0.2888889014720917, 0.4000000059604645, 0.4333333373069763, 0.5666666626930237, 0.5888888835906982, 0.4555555582046509, NaN, 0.4333333373069763, 0.4555555582046509, 0.5666666626930237, 0.5444444417953491, 0.5333333611488342, 0.5666666626930237, 0.6222222447395325, 0.6555555462837219, 0.6111111044883728, 0.800000011920929, 0.6000000238418579, 0.7222222089767456, 0.6777777671813965, 0.8777777552604675, 0.7555555701255798, 0.7555555701255798, 0.7444444298744202, 0.8111110925674438, 0.7666666507720947, 0.8333333134651184, 0.8444444537162781, 0.8333333134651184, 0.7555555701255798, 0.7333333492279053, 0.8111110925674438, 0.8111110925674438, 0.7666666507720947, 0.9555555582046509, 0.8222222328186035, 0.8666666746139526, 0.8444444537162781, 0.8666666746139526, 0.8666666746139526, 0.9888888597488403, 1.0, 1.0, 0.9777777791023254, 1.0, 0.8777777552604675, 1.0, 0.8666666746139526, 1.0, 0.9888888597488403, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0], [0.07777778059244156, 0.02222222276031971, 0.23333333432674408, 0.12222222238779068, 0.14444445073604584, 0.15555556118488312, 0.2666666805744171, 0.2888889014720917, 0.2666666805744171, 0.31111112236976624, 0.3444444537162781, 0.3777777850627899, 0.36666667461395264, 0.3888888955116272, 0.2888889014720917, 0.4333333373069763, 0.3777777850627899, 0.42222222685813904, 0.4333333373069763, 0.3333333432674408, 0.47777777910232544, 0.5111111402511597, 0.5111111402511597, 0.5666666626930237, NaN, 0.6111111044883728, 0.6222222447395325, 0.5222222208976746, 0.4444444477558136, 0.6222222447395325, 0.6222222447395325, 0.5333333611488342, 0.6888889074325562, 0.6777777671813965, 0.8666666746139526, 0.6333333253860474, 0.5888888835906982, 0.6888889074325562, 0.7111111283302307, 0.7888888716697693, 0.800000011920929, 0.6666666865348816, 0.7222222089767456, 0.7777777910232544, 0.7444444298744202, 0.8333333134651184, 0.8777777552604675, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.9222221970558167, 0.8999999761581421, 0.855555534362793, 0.800000011920929, 0.9444444179534912, 0.9555555582046509, 0.8444444537162781, 1.0, 0.8444444537162781, 1.0, 0.8999999761581421, 0.8444444537162781, 0.9777777791023254, 0.9777777791023254, 0.9888888597488403, 1.0, 0.9888888597488403, 0.9222221970558167, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.04444444552063942, 0.23333333432674408, 0.13333334028720856, 0.2222222238779068, 0.1666666716337204, 0.31111112236976624, 0.2888889014720917, 0.2666666805744171, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.36666667461395264, 0.42222222685813904, 0.30000001192092896, 0.42222222685813904, 0.36666667461395264, 0.42222222685813904, 0.4555555582046509, 0.3333333432674408, 0.5222222208976746, 0.5111111402511597, 0.47777777910232544, 0.5444444417953491, 0.3888888955116272, NaN, 0.5333333611488342, 0.4444444477558136, 0.5333333611488342, 0.644444465637207, 0.6333333253860474, 0.6222222447395325, 0.6111111044883728, 0.6555555462837219, 0.6555555462837219, 0.6333333253860474, 0.6111111044883728, 0.699999988079071, 0.6777777671813965, 0.6888889074325562, 0.6777777671813965, 0.6888889074325562, 0.6777777671813965, 0.855555534362793, 0.7111111283302307, 0.7666666507720947, 0.8666666746139526, 0.6888889074325562, 0.7222222089767456, 0.8666666746139526, 0.800000011920929, 0.7777777910232544, 0.9222221970558167, 0.7111111283302307, 0.8333333134651184, 0.800000011920929, 0.7888888716697693, 0.9333333373069763, 0.7555555701255798, 0.9444444179534912, 0.855555534362793, 0.7555555701255798, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9777777791023254, 1.0, 1.0, 1.0], [0.1111111119389534, 0.10000000149011612, 0.14444445073604584, 0.13333334028720856, 0.24444444477558136, 0.24444444477558136, 0.25555557012557983, 0.24444444477558136, 0.2888889014720917, 0.2888889014720917, 0.36666667461395264, 0.30000001192092896, 0.2666666805744171, 0.3777777850627899, 0.3777777850627899, 0.36666667461395264, 0.35555556416511536, 0.35555556416511536, 0.4333333373069763, 0.42222222685813904, 0.42222222685813904, 0.41111111640930176, 0.4555555582046509, 0.4333333373069763, 0.3777777850627899, 0.46666666865348816, NaN, 0.4333333373069763, 0.5, 0.5555555820465088, 0.4888888895511627, 0.5111111402511597, 0.5, 0.7111111283302307, 0.5777778029441833, 0.5777778029441833, 0.6222222447395325, 0.7111111283302307, 0.6555555462837219, 0.6555555462837219, 0.6111111044883728, 0.7333333492279053, 0.6000000238418579, 0.7888888716697693, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.7555555701255798, 0.8444444537162781, 0.7444444298744202, 0.8444444537162781, 0.855555534362793, 0.855555534362793, 0.8666666746139526, 0.8888888955116272, 0.8999999761581421, 0.9444444179534912, 0.8999999761581421, 0.9444444179534912, 0.9888888597488403, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9222221970558167, 0.9777777791023254, 0.9666666388511658, 1.0, 0.9888888597488403, 0.9888888597488403, 0.9888888597488403, 1.0, 1.0, 1.0], [0.011111111380159855, 0.0555555559694767, 0.10000000149011612, 0.12222222238779068, 0.17777778208255768, 0.20000000298023224, 0.2611111104488373, 0.2888889014720917, 0.2222222238779068, 0.31111112236976624, 0.35555556416511536, 0.3444444537162781, 0.2888889014720917, 0.31111112236976624, 0.31111112236976624, 0.2666666805744171, 0.25555557012557983, 0.3333333432674408, 0.36666667461395264, 0.3888888955116272, 0.42222222685813904, 0.36666667461395264, 0.4444444477558136, 0.4555555582046509, 0.47777777910232544, 0.5555555820465088, 0.5666666626930237, NaN, 0.5333333611488342, 0.5777778029441833, 0.6111111044883728, 0.5222222208976746, 0.5333333611488342, 0.5555555820465088, 0.644444465637207, 0.5888888835906982, 0.6222222447395325, 0.6222222447395325, 0.7222222089767456, 0.6888889074325562, 0.7222222089767456, 0.6666666865348816, 0.644444465637207, 0.7777777910232544, 0.7222222089767456, 0.7666666507720947, 0.7666666507720947, 0.7222222089767456, 0.7111111283302307, 0.8999999761581421, 0.8444444537162781, 0.7888888716697693, 0.8666666746139526, 0.9222221970558167, 0.855555534362793, 0.855555534362793, 0.9111111164093018, 0.9555555582046509, 0.8666666746139526, 0.9555555582046509, 0.9555555582046509, 0.8999999761581421, 0.9444444179534912, 0.9777777791023254, 0.9555555582046509, 0.9888888597488403, 0.9666666388511658, 0.9888888597488403, 1.0, 0.9555555582046509, 0.9777777791023254, 1.0, 1.0, 1.0], [0.08888889104127884, 0.03333333507180214, 0.18888889253139496, 0.04444444552063942, 0.14444445073604584, 0.12222222238779068, 0.4000000059604645, 0.41111111640930176, 0.3888888955116272, 0.41111111640930176, 0.3777777850627899, 0.3333333432674408, 0.3888888955116272, 0.3333333432674408, 0.35555556416511536, 0.2888889014720917, 0.3888888955116272, 0.47777777910232544, 0.42222222685813904, 0.41111111640930176, 0.5111111402511597, 0.41111111640930176, 0.4000000059604645, 0.46666666865348816, 0.5555555820465088, 0.46666666865348816, 0.5, 0.46666666865348816, NaN, 0.5888888835906982, 0.644444465637207, 0.5111111402511597, 0.6000000238418579, 0.6222222447395325, 0.7111111283302307, 0.5777778029441833, 0.5666666626930237, 0.6555555462837219, 0.699999988079071, 0.6111111044883728, 0.7333333492279053, 0.644444465637207, 0.7111111283302307, 0.8333333134651184, 0.6777777671813965, 0.6888889074325562, 0.7888888716697693, 0.7666666507720947, 0.7555555701255798, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.9111111164093018, 0.7444444298744202, 0.8222222328186035, 0.800000011920929, 0.8777777552604675, 0.8888888955116272, 0.7333333492279053, 0.8888888955116272, 0.855555534362793, 0.7555555701255798, 0.9666666388511658, 0.9555555582046509, 0.9666666388511658, 0.8999999761581421, 0.9222221970558167, 0.8777777552604675, 1.0, 0.9111111164093018, 0.8888888955116272, 1.0, 1.0, 1.0], [0.07777778059244156, 0.03333333507180214, 0.20000000298023224, 0.1111111119389534, 0.14444445073604584, 0.10000000149011612, 0.3333333432674408, 0.31111112236976624, 0.25555557012557983, 0.3333333432674408, 0.31111112236976624, 0.3888888955116272, 0.30000001192092896, 0.3888888955116272, 0.31111112236976624, 0.3444444537162781, 0.31111112236976624, 0.3444444537162781, 0.3888888955116272, 0.31111112236976624, 0.4444444477558136, 0.4555555582046509, 0.42222222685813904, 0.4333333373069763, 0.3777777850627899, 0.35555556416511536, 0.4444444477558136, 0.42222222685813904, 0.41111111640930176, NaN, 0.5777778029441833, 0.5, 0.5444444417953491, 0.6111111044883728, 0.6111111044883728, 0.5333333611488342, 0.6000000238418579, 0.6555555462837219, 0.644444465637207, 0.6222222447395325, 0.7222222089767456, 0.6888889074325562, 0.5666666626930237, 0.8888888955116272, 0.6666666865348816, 0.6333333253860474, 0.7777777910232544, 0.6666666865348816, 0.6555555462837219, 0.7888888716697693, 0.6777777671813965, 0.7444444298744202, 0.9444444179534912, 0.800000011920929, 0.7888888716697693, 0.7777777910232544, 0.8333333134651184, 0.855555534362793, 0.7555555701255798, 0.855555534362793, 0.8444444537162781, 0.7555555701255798, 0.9888888597488403, 0.9888888597488403, 0.9777777791023254, 0.9888888597488403, 0.9888888597488403, 0.8888888955116272, 1.0, 0.9222221970558167, 0.9666666388511658, 1.0, 1.0, 1.0], [0.011111111380159855, 0.03333333507180214, 0.08888889104127884, 0.1666666716337204, 0.15555556118488312, 0.18888889253139496, 0.21111111342906952, 0.21111111342906952, 0.13333334028720856, 0.2222222238779068, 0.20000000298023224, 0.2666666805744171, 0.2888889014720917, 0.24444444477558136, 0.2888889014720917, 0.3888888955116272, 0.2666666805744171, 0.35555556416511536, 0.3222222328186035, 0.3444444537162781, 0.3222222328186035, 0.4333333373069763, 0.42222222685813904, 0.3777777850627899, 0.3777777850627899, 0.36666667461395264, 0.5111111402511597, 0.3888888955116272, 0.35555556416511536, 0.42222222685813904, NaN, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.5888888835906982, 0.6000000238418579, 0.6222222447395325, 0.6555555462837219, 0.6555555462837219, 0.6000000238418579, 0.8777777552604675, 0.6555555462837219, 0.6333333253860474, 0.7111111283302307, 0.699999988079071, 0.7555555701255798, 0.7555555701255798, 0.6666666865348816, 0.7555555701255798, 0.7333333492279053, 0.800000011920929, 0.8222222328186035, 0.7666666507720947, 0.8333333134651184, 0.8666666746139526, 0.9888888597488403, 0.8333333134651184, 0.9222221970558167, 0.9111111164093018, 0.9222221970558167, 0.9111111164093018, 0.9111111164093018, 0.9777777791023254, 0.9888888597488403, 0.9666666388511658, 1.0, 0.9888888597488403, 0.9555555582046509, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.011111111380159855, 0.011111111380159855, 0.04444444552063942, 0.13333334028720856, 0.12222222238779068, 0.24444444477558136, 0.15555556118488312, 0.18888889253139496, 0.18888889253139496, 0.1666666716337204, 0.18888889253139496, 0.23333333432674408, 0.17777778208255768, 0.24444444477558136, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.35555556416511536, 0.4444444477558136, 0.3222222328186035, 0.3444444537162781, 0.46666666865348816, 0.3777777850627899, 0.4888888895511627, 0.47777777910232544, 0.4888888895511627, 0.5, 0.5333333611488342, NaN, 0.6111111044883728, 0.4555555582046509, 0.5777778029441833, 0.7444444298744202, 0.6000000238418579, 0.5888888835906982, 0.7222222089767456, 0.6888889074325562, 0.7888888716697693, 0.6666666865348816, 0.9222221970558167, 0.699999988079071, 0.7222222089767456, 0.7222222089767456, 0.6777777671813965, 0.6000000238418579, 0.7666666507720947, 0.7222222089767456, 0.800000011920929, 0.8777777552604675, 0.6777777671813965, 0.7555555701255798, 0.8444444537162781, 0.8666666746139526, 0.800000011920929, 0.9777777791023254, 0.8999999761581421, 0.9666666388511658, 0.9111111164093018, 0.9222221970558167, 0.8777777552604675, 0.8777777552604675, 0.8666666746139526, 0.8777777552604675, 0.8888888955116272, 0.9444444179534912, 1.0, 0.8222222328186035, 0.9777777791023254, 1.0, 0.8999999761581421, 1.0], [0.03333333507180214, 0.02222222276031971, 0.07777778059244156, 0.08888889104127884, 0.08888889104127884, 0.13333334028720856, 0.17777778208255768, 0.21111111342906952, 0.21111111342906952, 0.18888889253139496, 0.06666667014360428, 0.17777778208255768, 0.3222222328186035, 0.25555557012557983, 0.23333333432674408, 0.3333333432674408, 0.3444444537162781, 0.3444444537162781, 0.23333333432674408, 0.24444444477558136, 0.3222222328186035, 0.3444444537162781, 0.25555557012557983, 0.3888888955116272, 0.31111112236976624, 0.3888888955116272, 0.5, 0.46666666865348816, 0.4000000059604645, 0.4555555582046509, 0.4333333373069763, 0.3888888955116272, NaN, 0.6333333253860474, 0.5444444417953491, 0.4888888895511627, 0.5333333611488342, 0.6777777671813965, 0.5444444417953491, 0.5888888835906982, 0.6666666865348816, 0.6888889074325562, 0.644444465637207, 0.6555555462837219, 0.7555555701255798, 0.8111110925674438, 0.5333333611488342, 0.6111111044883728, 0.7777777910232544, 0.7222222089767456, 0.8666666746139526, 0.8333333134651184, 0.6555555462837219, 0.7111111283302307, 0.8999999761581421, 0.9222221970558167, 0.7666666507720947, 0.9444444179534912, 0.8777777552604675, 0.9555555582046509, 0.8888888955116272, 0.8888888955116272, 0.8777777552604675, 0.9222221970558167, 0.9333333373069763, 0.9888888597488403, 0.9222221970558167, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.02222222276031971, 0.1111111119389534, 0.04444444552063942, 0.21111111342906952, 0.23333333432674408, 0.30000001192092896, 0.07777778059244156, 0.0555555559694767, 0.1111111119389534, 0.06666667014360428, 0.17777778208255768, 0.23333333432674408, 0.10000000149011612, 0.15555556118488312, 0.3444444537162781, 0.3444444537162781, 0.2888889014720917, 0.13333334028720856, 0.25555557012557983, 0.3444444537162781, 0.24444444477558136, 0.3888888955116272, 0.2777777910232544, 0.20000000298023224, 0.3222222328186035, 0.3444444537162781, 0.2888889014720917, 0.4444444477558136, 0.3777777850627899, 0.3888888955116272, 0.4000000059604645, 0.5444444417953491, 0.36666667461395264, NaN, 0.3777777850627899, 0.6111111044883728, 0.5555555820465088, 0.6777777671813965, 0.6222222447395325, 0.6333333253860474, 0.5777778029441833, 0.7333333492279053, 0.6222222447395325, 0.46666666865348816, 0.644444465637207, 0.5444444417953491, 0.46666666865348816, 0.6888889074325562, 0.6555555462837219, 0.5666666626930237, 0.7333333492279053, 0.6888889074325562, 0.5888888835906982, 0.800000011920929, 0.7222222089767456, 0.7555555701255798, 0.8666666746139526, 0.855555534362793, 0.9777777791023254, 0.855555534362793, 0.9666666388511658, 1.0, 0.8111110925674438, 0.8777777552604675, 0.855555534362793, 0.9666666388511658, 0.9777777791023254, 1.0, 0.8999999761581421, 1.0, 0.9777777791023254, 0.9666666388511658, 0.8999999761581421, 1.0], [0.0555555559694767, 0.04444444552063942, 0.15555556118488312, 0.03333333507180214, 0.13333334028720856, 0.08888889104127884, 0.2222222238779068, 0.18888889253139496, 0.23333333432674408, 0.23333333432674408, 0.31111112236976624, 0.25555557012557983, 0.3333333432674408, 0.3444444537162781, 0.14444445073604584, 0.24444444477558136, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.1666666716337204, 0.4555555582046509, 0.3444444537162781, 0.36666667461395264, 0.4000000059604645, 0.13333334028720856, 0.3444444537162781, 0.42222222685813904, 0.35555556416511536, 0.2888889014720917, 0.3888888955116272, 0.41111111640930176, 0.42222222685813904, 0.4555555582046509, 0.6222222447395325, NaN, 0.47777777910232544, 0.5444444417953491, 0.5777778029441833, 0.47777777910232544, 0.5666666626930237, 0.5555555820465088, 0.6222222447395325, 0.5, 0.6222222447395325, 0.5333333611488342, 0.644444465637207, 0.6222222447395325, 0.6777777671813965, 0.5555555820465088, 0.6555555462837219, 0.6333333253860474, 0.6333333253860474, 0.855555534362793, 0.6777777671813965, 0.699999988079071, 0.6888889074325562, 0.6777777671813965, 0.800000011920929, 0.7222222089767456, 0.800000011920929, 0.7888888716697693, 0.7555555701255798, 0.9555555582046509, 0.9111111164093018, 0.9777777791023254, 1.0, 0.9222221970558167, 0.8111110925674438, 0.9222221970558167, 0.8999999761581421, 1.0, 0.9666666388511658, 1.0, 1.0], [0.0, 0.0, 0.04444444552063942, 0.08888889104127884, 0.03333333507180214, 0.10000000149011612, 0.17777778208255768, 0.18888889253139496, 0.17777778208255768, 0.17777778208255768, 0.15555556118488312, 0.13333334028720856, 0.14444445073604584, 0.17777778208255768, 0.17777778208255768, 0.15555556118488312, 0.24444444477558136, 0.2222222238779068, 0.18888889253139496, 0.2666666805744171, 0.2888889014720917, 0.23333333432674408, 0.2666666805744171, 0.2777777910232544, 0.36666667461395264, 0.36666667461395264, 0.42222222685813904, 0.41111111640930176, 0.42222222685813904, 0.46666666865348816, 0.4000000059604645, 0.25555557012557983, 0.5111111402511597, 0.3888888955116272, 0.5222222208976746, NaN, 0.46666666865348816, 0.4444444477558136, 0.6000000238418579, 0.6222222447395325, 0.6222222447395325, 0.5555555820465088, 0.7444444298744202, 0.5888888835906982, 0.6333333253860474, 0.6333333253860474, 0.5888888835906982, 0.5888888835906982, 0.6666666865348816, 0.6777777671813965, 0.6888889074325562, 0.8222222328186035, 0.6000000238418579, 0.7333333492279053, 0.7555555701255798, 0.8444444537162781, 0.7444444298744202, 0.9444444179534912, 0.855555534362793, 0.9444444179534912, 0.8888888955116272, 0.8888888955116272, 0.8666666746139526, 0.8444444537162781, 0.855555534362793, 0.8777777552604675, 0.8333333134651184, 0.9222221970558167, 1.0, 0.8111110925674438, 0.9888888597488403, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.07777778059244156, 0.011111111380159855, 0.07777778059244156, 0.0555555559694767, 0.23333333432674408, 0.21111111342906952, 0.1666666716337204, 0.20000000298023224, 0.18888889253139496, 0.20000000298023224, 0.17777778208255768, 0.10000000149011612, 0.25555557012557983, 0.03333333507180214, 0.17777778208255768, 0.21111111342906952, 0.2222222238779068, 0.3222222328186035, 0.23333333432674408, 0.15555556118488312, 0.24444444477558136, 0.3222222328186035, 0.41111111640930176, 0.3888888955116272, 0.3777777850627899, 0.3777777850627899, 0.4333333373069763, 0.4000000059604645, 0.3777777850627899, 0.4000000059604645, 0.46666666865348816, 0.4444444477558136, 0.4555555582046509, 0.5333333611488342, NaN, 0.4888888895511627, 0.6111111044883728, 0.6000000238418579, 0.5111111402511597, 0.6555555462837219, 0.5777778029441833, 0.6111111044883728, 0.6666666865348816, 0.6111111044883728, 0.6222222447395325, 0.5777778029441833, 0.6666666865348816, 0.7222222089767456, 0.644444465637207, 0.6222222447395325, 0.6222222447395325, 0.7777777910232544, 0.6777777671813965, 0.7777777910232544, 0.8111110925674438, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8444444537162781, 0.7888888716697693, 0.9777777791023254, 0.8222222328186035, 0.9777777791023254, 0.9777777791023254, 0.9444444179534912, 0.8999999761581421, 1.0, 0.8999999761581421, 0.9555555582046509, 1.0, 1.0, 1.0], [0.011111111380159855, 0.08888889104127884, 0.03333333507180214, 0.20000000298023224, 0.18888889253139496, 0.2777777910232544, 0.0555555559694767, 0.04444444552063942, 0.04444444552063942, 0.06666667014360428, 0.08888889104127884, 0.1666666716337204, 0.07777778059244156, 0.1111111119389534, 0.2888889014720917, 0.3444444537162781, 0.2222222238779068, 0.10000000149011612, 0.18888889253139496, 0.31111112236976624, 0.20000000298023224, 0.3444444537162781, 0.2777777910232544, 0.12222222238779068, 0.31111112236976624, 0.30000001192092896, 0.2888889014720917, 0.3777777850627899, 0.3444444537162781, 0.3444444537162781, 0.3444444537162781, 0.41111111640930176, 0.3222222328186035, 0.3222222328186035, 0.42222222685813904, 0.5555555820465088, 0.5111111402511597, NaN, 0.6000000238418579, 0.6333333253860474, 0.5444444417953491, 0.6666666865348816, 0.5222222208976746, 0.5111111402511597, 0.6222222447395325, 0.47777777910232544, 0.4555555582046509, 0.6222222447395325, 0.6555555462837219, 0.5666666626930237, 0.6555555462837219, 0.800000011920929, 0.5666666626930237, 0.8222222328186035, 0.7444444298744202, 0.7333333492279053, 0.8666666746139526, 0.8333333134651184, 0.9555555582046509, 0.8444444537162781, 0.9777777791023254, 0.9888888597488403, 0.855555534362793, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.9444444179534912, 1.0, 0.9888888597488403, 0.9888888597488403, 0.9555555582046509, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.03333333507180214, 0.0555555559694767, 0.0, 0.08888889104127884, 0.1666666716337204, 0.17777778208255768, 0.04444444552063942, 0.15555556118488312, 0.15555556118488312, 0.17777778208255768, 0.14444445073604584, 0.0, 0.17777778208255768, 0.02222222276031971, 0.1111111119389534, 0.15555556118488312, 0.17777778208255768, 0.23333333432674408, 0.10000000149011612, 0.10000000149011612, 0.23333333432674408, 0.24444444477558136, 0.2888889014720917, 0.3222222328186035, 0.3444444537162781, 0.2777777910232544, 0.30000001192092896, 0.35555556416511536, 0.3444444537162781, 0.2777777910232544, 0.4555555582046509, 0.3777777850627899, 0.5222222208976746, 0.4000000059604645, 0.3888888955116272, 0.4000000059604645, NaN, 0.4555555582046509, 0.5777778029441833, 0.5, 0.4333333373069763, 0.5666666626930237, 0.6000000238418579, 0.6111111044883728, 0.5777778029441833, 0.6111111044883728, 0.6666666865348816, 0.6333333253860474, 0.6888889074325562, 0.6222222447395325, 0.6333333253860474, 0.8222222328186035, 0.7111111283302307, 0.7888888716697693, 0.8444444537162781, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8777777552604675, 0.9555555582046509, 0.855555534362793, 0.9444444179534912, 0.8666666746139526, 0.9555555582046509, 0.9111111164093018, 1.0, 0.8777777552604675, 0.9888888597488403, 1.0, 1.0, 1.0], [0.0, 0.0, 0.08888889104127884, 0.0555555559694767, 0.10000000149011612, 0.07777778059244156, 0.15555556118488312, 0.14444445073604584, 0.13333334028720856, 0.14444445073604584, 0.12222222238779068, 0.13333334028720856, 0.12222222238779068, 0.14444445073604584, 0.20000000298023224, 0.08888889104127884, 0.17777778208255768, 0.14444445073604584, 0.14444445073604584, 0.24444444477558136, 0.2222222238779068, 0.20000000298023224, 0.21111111342906952, 0.24444444477558136, 0.21111111342906952, 0.31111112236976624, 0.3444444537162781, 0.31111112236976624, 0.3888888955116272, 0.3777777850627899, 0.4000000059604645, 0.31111112236976624, 0.41111111640930176, 0.36666667461395264, 0.4333333373069763, 0.3777777850627899, 0.4000000059604645, 0.36666667461395264, 0.5444444417953491, NaN, 0.5, 0.6000000238418579, 0.5222222208976746, 0.6111111044883728, 0.5555555820465088, 0.5444444417953491, 0.5888888835906982, 0.6555555462837219, 0.5888888835906982, 0.6000000238418579, 0.6111111044883728, 0.644444465637207, 0.5777778029441833, 0.7333333492279053, 0.6333333253860474, 0.6666666865348816, 0.7666666507720947, 0.9444444179534912, 0.7555555701255798, 0.9555555582046509, 0.9111111164093018, 0.7666666507720947, 0.8666666746139526, 0.7666666507720947, 0.8666666746139526, 0.855555534362793, 0.855555534362793, 0.9444444179534912, 1.0, 0.8111110925674438, 0.9666666388511658, 1.0, 0.8999999761581421, 1.0], [0.02222222276031971, 0.0, 0.02222222276031971, 0.03333333507180214, 0.04444444552063942, 0.08888889104127884, 0.04444444552063942, 0.04444444552063942, 0.04444444552063942, 0.0555555559694767, 0.15555556118488312, 0.08888889104127884, 0.13333334028720856, 0.1666666716337204, 0.17777778208255768, 0.2222222238779068, 0.15555556118488312, 0.17777778208255768, 0.17777778208255768, 0.18888889253139496, 0.18888889253139496, 0.25555557012557983, 0.30000001192092896, 0.25555557012557983, 0.20000000298023224, 0.3222222328186035, 0.3888888955116272, 0.2777777910232544, 0.2666666805744171, 0.2777777910232544, 0.12222222238779068, 0.21111111342906952, 0.3333333432674408, 0.42222222685813904, 0.4444444477558136, 0.3777777850627899, 0.4888888895511627, 0.4555555582046509, 0.42222222685813904, 0.5, NaN, 0.5111111402511597, 0.4000000059604645, 0.4888888895511627, 0.5444444417953491, 0.644444465637207, 0.5777778029441833, 0.5222222208976746, 0.6222222447395325, 0.6555555462837219, 0.6111111044883728, 0.6555555462837219, 0.6222222447395325, 0.644444465637207, 0.7333333492279053, 0.8888888955116272, 0.6888889074325562, 0.8444444537162781, 0.8777777552604675, 0.8444444537162781, 0.8444444537162781, 0.8777777552604675, 0.9555555582046509, 0.9222221970558167, 0.9555555582046509, 0.9777777791023254, 0.8999999761581421, 0.9222221970558167, 0.9777777791023254, 0.8888888955116272, 0.9888888597488403, 1.0, 1.0, 1.0], [0.0, 0.08888889104127884, 0.03333333507180214, 0.10000000149011612, 0.13333334028720856, 0.15555556118488312, 0.13333334028720856, 0.1666666716337204, 0.03333333507180214, 0.17777778208255768, 0.20000000298023224, 0.23333333432674408, 0.14444445073604584, 0.02222222276031971, 0.2666666805744171, 0.17777778208255768, 0.13333334028720856, 0.18888889253139496, 0.24444444477558136, 0.30000001192092896, 0.12222222238779068, 0.20000000298023224, 0.24444444477558136, 0.18888889253139496, 0.3333333432674408, 0.31111112236976624, 0.2666666805744171, 0.3333333432674408, 0.35555556416511536, 0.31111112236976624, 0.3444444537162781, 0.3333333432674408, 0.31111112236976624, 0.2666666805744171, 0.3777777850627899, 0.4444444477558136, 0.3444444537162781, 0.3333333432674408, 0.5, 0.4000000059604645, 0.4888888895511627, NaN, 0.47777777910232544, 0.5222222208976746, 0.5777778029441833, 0.4444444477558136, 0.5222222208976746, 0.5555555820465088, 0.6000000238418579, 0.6222222447395325, 0.5888888835906982, 0.6111111044883728, 0.6222222447395325, 0.8444444537162781, 0.6222222447395325, 0.6777777671813965, 0.9666666388511658, 0.6777777671813965, 0.8888888955116272, 0.6777777671813965, 0.8777777552604675, 0.8999999761581421, 0.855555534362793, 0.8333333134651184, 0.8666666746139526, 0.7333333492279053, 0.9333333373069763, 0.9777777791023254, 0.8666666746139526, 1.0, 0.9333333373069763, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.04444444552063942, 0.04444444552063942, 0.03333333507180214, 0.06666667014360428, 0.0555555559694767, 0.1111111119389534, 0.08888889104127884, 0.03333333507180214, 0.02222222276031971, 0.03333333507180214, 0.06666667014360428, 0.15555556118488312, 0.13333334028720856, 0.1666666716337204, 0.17777778208255768, 0.14444445073604584, 0.07777778059244156, 0.1666666716337204, 0.2666666805744171, 0.21111111342906952, 0.14444445073604584, 0.23333333432674408, 0.2777777910232544, 0.3222222328186035, 0.4000000059604645, 0.35555556416511536, 0.2888889014720917, 0.4333333373069763, 0.36666667461395264, 0.07777778059244156, 0.35555556416511536, 0.3777777850627899, 0.5, 0.25555557012557983, 0.42222222685813904, 0.47777777910232544, 0.5666666626930237, 0.47777777910232544, 0.6000000238418579, 0.5222222208976746, NaN, 0.5777778029441833, 0.5555555820465088, 0.644444465637207, 0.5555555820465088, 0.5888888835906982, 0.6111111044883728, 0.644444465637207, 0.6777777671813965, 0.6777777671813965, 0.6000000238418579, 0.7333333492279053, 0.7555555701255798, 0.8333333134651184, 0.7333333492279053, 0.8444444537162781, 0.8111110925674438, 0.8444444537162781, 0.8777777552604675, 0.8444444537162781, 0.8666666746139526, 0.8222222328186035, 0.855555534362793, 0.8666666746139526, 0.8222222328186035, 0.9111111164093018, 1.0, 0.800000011920929, 0.9777777791023254, 1.0, 0.8999999761581421, 1.0], [0.04444444552063942, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.04444444552063942, 0.0555555559694767, 0.18888889253139496, 0.18888889253139496, 0.08888889104127884, 0.20000000298023224, 0.15555556118488312, 0.20000000298023224, 0.1666666716337204, 0.18888889253139496, 0.1111111119389534, 0.12222222238779068, 0.12222222238779068, 0.18888889253139496, 0.20000000298023224, 0.17777778208255768, 0.2666666805744171, 0.23333333432674408, 0.25555557012557983, 0.1666666716337204, 0.2222222238779068, 0.14444445073604584, 0.21111111342906952, 0.2222222238779068, 0.1666666716337204, 0.1111111119389534, 0.2888889014720917, 0.30000001192092896, 0.3444444537162781, 0.5333333611488342, 0.3777777850627899, 0.41111111640930176, 0.3888888955116272, 0.4888888895511627, 0.4333333373069763, 0.3888888955116272, 0.5111111402511597, 0.47777777910232544, 0.42222222685813904, NaN, 0.4888888895511627, 0.4444444477558136, 0.5666666626930237, 0.6222222447395325, 0.5888888835906982, 0.46666666865348816, 0.5777778029441833, 0.5222222208976746, 0.75, 0.6777777671813965, 0.644444465637207, 0.6777777671813965, 0.7777777910232544, 0.6666666865348816, 0.6888889074325562, 0.6777777671813965, 0.800000011920929, 0.7444444298744202, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 0.9111111164093018, 0.9333333373069763, 0.800000011920929, 0.9111111164093018, 0.8999999761581421, 0.8999999761581421, 0.9777777791023254, 1.0, 1.0], [0.0, 0.02222222276031971, 0.011111111380159855, 0.0555555559694767, 0.04444444552063942, 0.04444444552063942, 0.1111111119389534, 0.12222222238779068, 0.12222222238779068, 0.1111111119389534, 0.10000000149011612, 0.10000000149011612, 0.1111111119389534, 0.12222222238779068, 0.08888889104127884, 0.1666666716337204, 0.21111111342906952, 0.13333334028720856, 0.13333334028720856, 0.14444445073604584, 0.15555556118488312, 0.2222222238779068, 0.21111111342906952, 0.15555556118488312, 0.25555557012557983, 0.2888889014720917, 0.2666666805744171, 0.2777777910232544, 0.3222222328186035, 0.3333333432674408, 0.30000001192092896, 0.2777777910232544, 0.24444444477558136, 0.35555556416511536, 0.46666666865348816, 0.36666667461395264, 0.3333333432674408, 0.3777777850627899, 0.4000000059604645, 0.4444444477558136, 0.4555555582046509, 0.42222222685813904, 0.4444444477558136, 0.5111111402511597, NaN, 0.5111111402511597, 0.4444444477558136, 0.5333333611488342, 0.7444444298744202, 0.47777777910232544, 0.6333333253860474, 0.6111111044883728, 0.5555555820465088, 0.6333333253860474, 0.6888889074325562, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8111110925674438, 0.8888888955116272, 0.855555534362793, 0.855555534362793, 0.9666666388511658, 0.8333333134651184, 0.9333333373069763, 0.9666666388511658, 0.9555555582046509, 0.9333333373069763, 0.8888888955116272, 0.9666666388511658, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.03333333507180214, 0.011111111380159855, 0.0555555559694767, 0.07777778059244156, 0.07777778059244156, 0.0555555559694767, 0.0, 0.06666667014360428, 0.03333333507180214, 0.02222222276031971, 0.04444444552063942, 0.18888889253139496, 0.14444445073604584, 0.1666666716337204, 0.15555556118488312, 0.10000000149011612, 0.06666667014360428, 0.17777778208255768, 0.2666666805744171, 0.24444444477558136, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.31111112236976624, 0.36666667461395264, 0.24444444477558136, 0.2777777910232544, 0.18888889253139496, 0.4555555582046509, 0.35555556416511536, 0.36666667461395264, 0.3888888955116272, 0.5222222208976746, 0.3888888955116272, 0.4555555582046509, 0.35555556416511536, 0.5555555820465088, 0.35555556416511536, 0.5555555820465088, 0.4888888895511627, NaN, 0.4888888895511627, 0.5666666626930237, 0.5, 0.5666666626930237, 0.4888888895511627, 0.6777777671813965, 0.5777778029441833, 0.7222222089767456, 0.644444465637207, 0.644444465637207, 0.7666666507720947, 0.8888888955116272, 0.7555555701255798, 0.8999999761581421, 0.8777777552604675, 0.800000011920929, 0.7555555701255798, 0.855555534362793, 0.7444444298744202, 0.9222221970558167, 0.8444444537162781, 0.8888888955116272, 0.9888888597488403, 0.8888888955116272, 0.9666666388511658, 0.8999999761581421, 0.800000011920929, 1.0], [0.0, 0.0, 0.04444444552063942, 0.011111111380159855, 0.0555555559694767, 0.0555555559694767, 0.04444444552063942, 0.06666667014360428, 0.07777778059244156, 0.07777778059244156, 0.03333333507180214, 0.02222222276031971, 0.12222222238779068, 0.06666667014360428, 0.07777778059244156, 0.0, 0.08888889104127884, 0.18888889253139496, 0.06666667014360428, 0.1111111119389534, 0.17777778208255768, 0.13333334028720856, 0.03333333507180214, 0.24444444477558136, 0.12222222238779068, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.21111111342906952, 0.2222222238779068, 0.24444444477558136, 0.3222222328186035, 0.46666666865348816, 0.5333333611488342, 0.3777777850627899, 0.41111111640930176, 0.3777777850627899, 0.5444444417953491, 0.42222222685813904, 0.41111111640930176, 0.42222222685813904, 0.47777777910232544, 0.4444444477558136, 0.4333333373069763, 0.5555555820465088, 0.5111111402511597, NaN, 0.6000000238418579, 0.5777778029441833, 0.5, 0.5555555820465088, 0.4888888895511627, 0.5666666626930237, 0.6000000238418579, 0.644444465637207, 0.7444444298744202, 0.699999988079071, 0.8444444537162781, 0.6888889074325562, 0.855555534362793, 0.7666666507720947, 0.699999988079071, 0.8777777552604675, 0.8888888955116272, 0.9555555582046509, 0.9444444179534912, 0.9444444179534912, 0.8111110925674438, 1.0, 0.8999999761581421, 0.9222221970558167, 1.0, 1.0, 1.0], [0.1111111119389534, 0.14444445073604584, 0.12222222238779068, 0.20000000298023224, 0.2222222238779068, 0.18888889253139496, 0.2777777910232544, 0.24444444477558136, 0.24444444477558136, 0.2777777910232544, 0.3222222328186035, 0.31111112236976624, 0.25555557012557983, 0.15555556118488312, 0.3333333432674408, 0.23333333432674408, 0.14444445073604584, 0.2666666805744171, 0.3444444537162781, 0.3444444537162781, 0.20000000298023224, 0.23333333432674408, 0.35555556416511536, 0.2666666805744171, 0.3333333432674408, 0.31111112236976624, 0.23333333432674408, 0.2777777910232544, 0.23333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3888888955116272, 0.31111112236976624, 0.3222222328186035, 0.41111111640930176, 0.42222222685813904, 0.3777777850627899, 0.3888888955116272, 0.3444444537162781, 0.47777777910232544, 0.4444444477558136, 0.41111111640930176, 0.3777777850627899, 0.46666666865348816, 0.4333333373069763, 0.4000000059604645, NaN, 0.47777777910232544, 0.46666666865348816, 0.5222222208976746, 0.41111111640930176, 0.5444444417953491, 0.5777778029441833, 0.5555555820465088, 0.4888888895511627, 0.6111111044883728, 0.5222222208976746, 0.644444465637207, 0.5333333611488342, 0.5666666626930237, 0.6555555462837219, 0.6333333253860474, 0.7555555701255798, 0.6777777671813965, 0.6333333253860474, 0.7111111283302307, 0.6555555462837219, 0.6222222447395325, 0.7777777910232544, 0.6555555462837219, 0.6777777671813965, 0.7888888716697693, 0.8888888955116272], [0.011111111380159855, 0.02222222276031971, 0.02222222276031971, 0.03333333507180214, 0.011111111380159855, 0.03333333507180214, 0.12222222238779068, 0.12222222238779068, 0.1111111119389534, 0.12222222238779068, 0.10000000149011612, 0.08888889104127884, 0.1111111119389534, 0.13333334028720856, 0.04444444552063942, 0.1666666716337204, 0.2222222238779068, 0.13333334028720856, 0.14444445073604584, 0.06666667014360428, 0.17777778208255768, 0.21111111342906952, 0.15555556118488312, 0.18888889253139496, 0.20000000298023224, 0.2777777910232544, 0.24444444477558136, 0.2888889014720917, 0.24444444477558136, 0.3444444537162781, 0.24444444477558136, 0.23333333432674408, 0.2222222238779068, 0.3444444537162781, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3444444537162781, 0.3333333432674408, 0.41111111640930176, 0.3777777850627899, 0.4000000059604645, 0.3888888955116272, 0.41111111640930176, 0.25555557012557983, 0.5, 0.42222222685813904, 0.5222222208976746, NaN, 0.41111111640930176, 0.5444444417953491, 0.5777778029441833, 0.5111111402511597, 0.5666666626930237, 0.6333333253860474, 0.5777778029441833, 0.6666666865348816, 0.7888888716697693, 0.7111111283302307, 0.800000011920929, 0.855555534362793, 0.7777777910232544, 0.8333333134651184, 0.9388889074325562, 0.8333333134651184, 0.9222221970558167, 0.9222221970558167, 0.9111111164093018, 0.9333333373069763, 0.855555534362793, 0.9777777791023254, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.011111111380159855, 0.02222222276031971, 0.02222222276031971, 0.04444444552063942, 0.04444444552063942, 0.011111111380159855, 0.02222222276031971, 0.06666667014360428, 0.011111111380159855, 0.02222222276031971, 0.08888889104127884, 0.14444445073604584, 0.17777778208255768, 0.10000000149011612, 0.04444444552063942, 0.14444445073604584, 0.03333333507180214, 0.21111111342906952, 0.24444444477558136, 0.1666666716337204, 0.06666667014360428, 0.18888889253139496, 0.20000000298023224, 0.13333334028720856, 0.15555556118488312, 0.10000000149011612, 0.30000001192092896, 0.21111111342906952, 0.2666666805744171, 0.2777777910232544, 0.2777777910232544, 0.4333333373069763, 0.3444444537162781, 0.3222222328186035, 0.2777777910232544, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.3444444537162781, 0.3777777850627899, 0.35555556416511536, 0.5333333611488342, 0.5222222208976746, 0.4333333373069763, 0.5, 0.5333333611488342, 0.5888888835906982, NaN, 0.46666666865348816, 0.5777778029441833, 0.5666666626930237, 0.5444444417953491, 0.5555555820465088, 0.7555555701255798, 0.6777777671813965, 0.7444444298744202, 0.699999988079071, 0.7555555701255798, 0.8333333134651184, 0.7444444298744202, 0.8444444537162781, 0.9555555582046509, 0.8666666746139526, 0.9111111164093018, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8999999761581421, 0.9222221970558167, 1.0, 0.9222221970558167, 1.0], [0.0, 0.011111111380159855, 0.011111111380159855, 0.03333333507180214, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.03333333507180214, 0.04444444552063942, 0.04444444552063942, 0.011111111380159855, 0.02222222276031971, 0.1111111119389534, 0.10000000149011612, 0.03333333507180214, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.04444444552063942, 0.08888889104127884, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.23333333432674408, 0.07777778059244156, 0.20000000298023224, 0.25555557012557983, 0.15555556118488312, 0.23333333432674408, 0.3222222328186035, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.36666667461395264, 0.31111112236976624, 0.35555556416511536, 0.3444444537162781, 0.31111112236976624, 0.3888888955116272, 0.3888888955116272, 0.41111111640930176, 0.3222222328186035, 0.42222222685813904, 0.36666667461395264, 0.5111111402511597, 0.4444444477558136, 0.47777777910232544, 0.4555555582046509, 0.5333333611488342, NaN, 0.4888888895511627, 0.5333333611488342, 0.5333333611488342, 0.7555555701255798, 0.644444465637207, 0.644444465637207, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.7444444298744202, 0.7888888716697693, 0.7555555701255798, 0.9888888597488403, 0.800000011920929, 0.9444444179534912, 0.9777777791023254, 0.8999999761581421, 0.855555534362793, 0.9111111164093018, 0.9666666388511658, 0.8999999761581421, 1.0, 1.0], [0.0, 0.02222222276031971, 0.03333333507180214, 0.07777778059244156, 0.04444444552063942, 0.07777778059244156, 0.02222222276031971, 0.04444444552063942, 0.02222222276031971, 0.03333333507180214, 0.04444444552063942, 0.04444444552063942, 0.02222222276031971, 0.1111111119389534, 0.14444445073604584, 0.12222222238779068, 0.14444445073604584, 0.06666667014360428, 0.0555555559694767, 0.1666666716337204, 0.21111111342906952, 0.20000000298023224, 0.17777778208255768, 0.04444444552063942, 0.10000000149011612, 0.2222222238779068, 0.15555556118488312, 0.21111111342906952, 0.30000001192092896, 0.25555557012557983, 0.17777778208255768, 0.12222222238779068, 0.1666666716337204, 0.31111112236976624, 0.36666667461395264, 0.17777778208255768, 0.3777777850627899, 0.20000000298023224, 0.3777777850627899, 0.35555556416511536, 0.3444444537162781, 0.3888888955116272, 0.3222222328186035, 0.47777777910232544, 0.3888888955116272, 0.3222222328186035, 0.5111111402511597, 0.5888888835906982, 0.42222222685813904, 0.42222222685813904, 0.5111111402511597, NaN, 0.5777778029441833, 0.6111111044883728, 0.6777777671813965, 0.644444465637207, 0.7111111283302307, 0.6666666865348816, 0.8111110925674438, 0.6888889074325562, 0.8888888955116272, 0.8444444537162781, 0.8444444537162781, 0.8222222328186035, 0.8666666746139526, 0.8666666746139526, 0.855555534362793, 0.9777777791023254, 0.9555555582046509, 0.8333333134651184, 0.9666666388511658, 1.0, 0.8999999761581421, 1.0], [0.02222222276031971, 0.02222222276031971, 0.04444444552063942, 0.03333333507180214, 0.0555555559694767, 0.04444444552063942, 0.1111111119389534, 0.10000000149011612, 0.06666667014360428, 0.12222222238779068, 0.1111111119389534, 0.04444444552063942, 0.1666666716337204, 0.17777778208255768, 0.10000000149011612, 0.12222222238779068, 0.07777778059244156, 0.2222222238779068, 0.14444445073604584, 0.15555556118488312, 0.25555557012557983, 0.20000000298023224, 0.1666666716337204, 0.17777778208255768, 0.14444445073604584, 0.07777778059244156, 0.14444445073604584, 0.13333334028720856, 0.08888889104127884, 0.0555555559694767, 0.23333333432674408, 0.3222222328186035, 0.3444444537162781, 0.41111111640930176, 0.14444445073604584, 0.4000000059604645, 0.3777777850627899, 0.4333333373069763, 0.36666667461395264, 0.42222222685813904, 0.3777777850627899, 0.3777777850627899, 0.4000000059604645, 0.25, 0.4444444477558136, 0.42222222685813904, 0.4333333373069763, 0.4555555582046509, 0.4888888895511627, 0.4333333373069763, 0.46666666865348816, 0.42222222685813904, NaN, 0.5111111402511597, 0.5, 0.6222222447395325, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.6111111044883728, 0.7111111283302307, 0.7888888716697693, 0.6777777671813965, 0.8333333134651184, 0.7222222089767456, 0.6777777671813965, 0.7444444298744202, 0.7111111283302307, 0.7555555701255798, 0.855555534362793, 0.7333333492279053, 0.9777777791023254, 1.0], [0.0, 0.02222222276031971, 0.02222222276031971, 0.06666667014360428, 0.0555555559694767, 0.06666667014360428, 0.03333333507180214, 0.02222222276031971, 0.03333333507180214, 0.04444444552063942, 0.12222222238779068, 0.08888889104127884, 0.1111111119389534, 0.02222222276031971, 0.12222222238779068, 0.08888889104127884, 0.06666667014360428, 0.14444445073604584, 0.13333334028720856, 0.12222222238779068, 0.03333333507180214, 0.12222222238779068, 0.21111111342906952, 0.13333334028720856, 0.20000000298023224, 0.2888889014720917, 0.14444445073604584, 0.07777778059244156, 0.25555557012557983, 0.20000000298023224, 0.1666666716337204, 0.24444444477558136, 0.2888889014720917, 0.20000000298023224, 0.3222222328186035, 0.2666666805744171, 0.2222222238779068, 0.17777778208255768, 0.17777778208255768, 0.2666666805744171, 0.35555556416511536, 0.15555556118488312, 0.2666666805744171, 0.3222222328186035, 0.36666667461395264, 0.2777777910232544, 0.4000000059604645, 0.42222222685813904, 0.4333333373069763, 0.4555555582046509, 0.46666666865348816, 0.3888888955116272, 0.4888888895511627, NaN, 0.5111111402511597, 0.4888888895511627, 0.7111111283302307, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.7111111283302307, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.6777777671813965, 0.8222222328186035, 0.8999999761581421, 0.8777777552604675, 0.8888888955116272, 0.8666666746139526, 0.8888888955116272, 0.9555555582046509, 1.0], [0.0, 0.011111111380159855, 0.0, 0.03333333507180214, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.03333333507180214, 0.011111111380159855, 0.02222222276031971, 0.1111111119389534, 0.12222222238779068, 0.03333333507180214, 0.13333334028720856, 0.1111111119389534, 0.13333334028720856, 0.02222222276031971, 0.06666667014360428, 0.14444445073604584, 0.1666666716337204, 0.07777778059244156, 0.15555556118488312, 0.0555555559694767, 0.1666666716337204, 0.13333334028720856, 0.14444445073604584, 0.17777778208255768, 0.21111111342906952, 0.13333334028720856, 0.15555556118488312, 0.10000000149011612, 0.2777777910232544, 0.30000001192092896, 0.24444444477558136, 0.3222222328186035, 0.25555557012557983, 0.2888889014720917, 0.36666667461395264, 0.2666666805744171, 0.3777777850627899, 0.24444444477558136, 0.35555556416511536, 0.31111112236976624, 0.35555556416511536, 0.35555556416511536, 0.4444444477558136, 0.36666667461395264, 0.4444444477558136, 0.24444444477558136, 0.3222222328186035, 0.5, 0.4888888895511627, NaN, 0.5555555820465088, 0.5444444417953491, 0.7111111283302307, 0.7111111283302307, 0.699999988079071, 0.6888889074325562, 0.7444444298744202, 0.7111111283302307, 0.9555555582046509, 0.7777777910232544, 0.9333333373069763, 0.9444444179534912, 0.8888888955116272, 0.8444444537162781, 0.8999999761581421, 0.9666666388511658, 0.9111111164093018, 0.9888888597488403, 1.0], [0.0, 0.0, 0.011111111380159855, 0.03333333507180214, 0.0, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.08888889104127884, 0.011111111380159855, 0.04444444552063942, 0.02222222276031971, 0.1111111119389534, 0.08888889104127884, 0.0, 0.08888889104127884, 0.06666667014360428, 0.02222222276031971, 0.06666667014360428, 0.13333334028720856, 0.04444444552063942, 0.20000000298023224, 0.1111111119389534, 0.14444445073604584, 0.20000000298023224, 0.2222222238779068, 0.011111111380159855, 0.13333334028720856, 0.07777778059244156, 0.24444444477558136, 0.31111112236976624, 0.15555556118488312, 0.2222222238779068, 0.2666666805744171, 0.21111111342906952, 0.3333333432674408, 0.1111111119389534, 0.3222222328186035, 0.1666666716337204, 0.3222222328186035, 0.30000001192092896, 0.35555556416511536, 0.25555557012557983, 0.5111111402511597, 0.42222222685813904, 0.24444444477558136, 0.35555556416511536, 0.35555556416511536, 0.3777777850627899, 0.5111111402511597, 0.4444444477558136, NaN, 0.5555555820465088, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.7888888716697693, 0.8222222328186035, 0.9222221970558167, 0.8666666746139526, 0.8777777552604675, 0.8999999761581421, 0.8888888955116272, 0.9555555582046509, 0.9888888597488403, 0.9555555582046509, 1.0], [0.0, 0.02222222276031971, 0.011111111380159855, 0.06666667014360428, 0.07777778059244156, 0.07777778059244156, 0.02222222276031971, 0.04444444552063942, 0.0, 0.0555555559694767, 0.10000000149011612, 0.13333334028720856, 0.12222222238779068, 0.03333333507180214, 0.10000000149011612, 0.10000000149011612, 0.03333333507180214, 0.13333334028720856, 0.14444445073604584, 0.1111111119389534, 0.04444444552063942, 0.08888889104127884, 0.18888889253139496, 0.13333334028720856, 0.15555556118488312, 0.21111111342906952, 0.10000000149011612, 0.08888889104127884, 0.12222222238779068, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.13333334028720856, 0.3222222328186035, 0.25555557012557983, 0.18888889253139496, 0.13333334028720856, 0.15555556118488312, 0.23333333432674408, 0.31111112236976624, 0.03333333507180214, 0.2666666805744171, 0.2222222238779068, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.3888888955116272, 0.3333333432674408, 0.3222222328186035, 0.35555556416511536, 0.2888889014720917, 0.36666667461395264, 0.2888889014720917, 0.4555555582046509, 0.4444444477558136, NaN, 0.47777777910232544, 0.5555555820465088, 0.4888888895511627, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.6111111044883728, 0.7222222089767456, 0.6333333253860474, 0.6222222447395325, 0.6777777671813965, 0.699999988079071, 0.8111110925674438, 0.800000011920929, 0.8111110925674438, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.07777778059244156, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.0, 0.0, 0.011111111380159855, 0.0, 0.06666667014360428, 0.0555555559694767, 0.04444444552063942, 0.1111111119389534, 0.14444445073604584, 0.07777778059244156, 0.02222222276031971, 0.0555555559694767, 0.14444445073604584, 0.20000000298023224, 0.0555555559694767, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.0555555559694767, 0.15555556118488312, 0.3222222328186035, 0.15555556118488312, 0.3333333432674408, 0.20000000298023224, 0.1111111119389534, 0.15555556118488312, 0.47777777910232544, 0.21111111342906952, 0.25555557012557983, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2888889014720917, 0.2666666805744171, 0.5222222208976746, NaN, 0.5666666626930237, 0.6555555462837219, 0.6555555462837219, 0.5777778029441833, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.644444465637207, 0.699999988079071, 0.7444444298744202, 1.0, 0.699999988079071, 0.8777777552604675, 0.8777777552604675, 0.800000011920929, 0.8999999761581421], [0.0, 0.02222222276031971, 0.0, 0.07777778059244156, 0.03333333507180214, 0.07777778059244156, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.02222222276031971, 0.0, 0.0, 0.15555556118488312, 0.06666667014360428, 0.10000000149011612, 0.0, 0.0, 0.17777778208255768, 0.02222222276031971, 0.1111111119389534, 0.13333334028720856, 0.0, 0.15555556118488312, 0.24444444477558136, 0.10000000149011612, 0.13333334028720856, 0.2666666805744171, 0.24444444477558136, 0.08888889104127884, 0.10000000149011612, 0.12222222238779068, 0.02222222276031971, 0.2777777910232544, 0.14444445073604584, 0.20000000298023224, 0.04444444552063942, 0.1666666716337204, 0.24444444477558136, 0.12222222238779068, 0.1111111119389534, 0.18888889253139496, 0.31111112236976624, 0.20000000298023224, 0.24444444477558136, 0.31111112236976624, 0.35555556416511536, 0.2888889014720917, 0.30000001192092896, 0.2777777910232544, 0.18888889253139496, 0.30000001192092896, 0.3333333432674408, 0.2888889014720917, 0.2666666805744171, 0.4444444477558136, 0.4333333373069763, NaN, 0.4333333373069763, 0.42222222685813904, 0.8166666626930237, 0.4444444477558136, 0.5555555820465088, 0.47777777910232544, 0.5888888835906982, 0.5444444417953491, 0.9333333373069763, 0.5, 0.7777777910232544, 0.6888889074325562, 0.6222222447395325, 0.7666666507720947, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.07777778059244156, 0.011111111380159855, 0.011111111380159855, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0555555559694767, 0.0555555559694767, 0.04444444552063942, 0.1111111119389534, 0.14444445073604584, 0.07777778059244156, 0.03333333507180214, 0.04444444552063942, 0.14444445073604584, 0.20000000298023224, 0.0555555559694767, 0.10000000149011612, 0.15555556118488312, 0.13333334028720856, 0.04444444552063942, 0.15555556118488312, 0.3222222328186035, 0.15555556118488312, 0.3222222328186035, 0.18888889253139496, 0.10000000149011612, 0.14444445073604584, 0.46666666865348816, 0.20000000298023224, 0.24444444477558136, 0.2777777910232544, 0.31111112236976624, 0.3333333432674408, 0.4444444477558136, 0.30000001192092896, 0.2666666805744171, 0.5111111402511597, 0.3444444537162781, 0.5666666626930237, NaN, 0.6555555462837219, 0.5777778029441833, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.644444465637207, 0.699999988079071, 0.7444444298744202, 1.0, 0.699999988079071, 0.8777777552604675, 0.8777777552604675, 0.800000011920929, 0.8999999761581421], [0.011111111380159855, 0.04444444552063942, 0.011111111380159855, 0.06666667014360428, 0.0555555559694767, 0.04444444552063942, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.08888889104127884, 0.0555555559694767, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.1111111119389534, 0.03333333507180214, 0.1111111119389534, 0.08888889104127884, 0.02222222276031971, 0.10000000149011612, 0.14444445073604584, 0.011111111380159855, 0.04444444552063942, 0.14444445073604584, 0.15555556118488312, 0.08888889104127884, 0.08888889104127884, 0.1111111119389534, 0.03333333507180214, 0.21111111342906952, 0.1111111119389534, 0.15555556118488312, 0.02222222276031971, 0.13333334028720856, 0.08888889104127884, 0.15555556118488312, 0.12222222238779068, 0.12222222238779068, 0.20000000298023224, 0.1111111119389534, 0.12222222238779068, 0.23333333432674408, 0.4333333373069763, 0.14444445073604584, 0.1666666716337204, 0.25555557012557983, 0.1111111119389534, 0.3888888955116272, 0.2888889014720917, 0.31111112236976624, 0.20000000298023224, 0.4000000059604645, 0.3444444537162781, 0.5777778029441833, 0.3444444537162781, NaN, 0.6333333253860474, 0.5333333611488342, 0.6888889074325562, 0.5777778029441833, 0.5777778029441833, 0.5666666626930237, 0.7222222089767456, 0.6000000238418579, 0.7111111283302307, 0.6666666865348816, 1.0, 0.8444444537162781, 1.0], [0.0, 0.02222222276031971, 0.0, 0.06666667014360428, 0.03333333507180214, 0.0555555559694767, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.0, 0.0, 0.14444445073604584, 0.0555555559694767, 0.08888889104127884, 0.0, 0.0, 0.13333334028720856, 0.011111111380159855, 0.10000000149011612, 0.08888889104127884, 0.0, 0.15555556118488312, 0.24444444477558136, 0.06666667014360428, 0.10000000149011612, 0.24444444477558136, 0.24444444477558136, 0.08888889104127884, 0.07777778059244156, 0.1111111119389534, 0.0, 0.24444444477558136, 0.1111111119389534, 0.21111111342906952, 0.011111111380159855, 0.12222222238779068, 0.23333333432674408, 0.12222222238779068, 0.10000000149011612, 0.15555556118488312, 0.25555557012557983, 0.14444445073604584, 0.20000000298023224, 0.30000001192092896, 0.3444444537162781, 0.2222222238779068, 0.25555557012557983, 0.21111111342906952, 0.15555556118488312, 0.2888889014720917, 0.2666666805744171, 0.25555557012557983, 0.23333333432674408, 0.36666667461395264, 0.42222222685813904, 0.18333333730697632, 0.42222222685813904, 0.36666667461395264, NaN, 0.4333333373069763, 0.5111111402511597, 0.4555555582046509, 0.5666666626930237, 0.5111111402511597, 0.8666666746139526, 0.5, 0.7222222089767456, 0.6555555462837219, 0.6222222447395325, 0.7444444298744202, 0.8999999761581421], [0.011111111380159855, 0.0, 0.02222222276031971, 0.0, 0.011111111380159855, 0.0, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.02222222276031971, 0.07777778059244156, 0.02222222276031971, 0.12222222238779068, 0.02222222276031971, 0.03333333507180214, 0.02222222276031971, 0.011111111380159855, 0.12222222238779068, 0.08888889104127884, 0.03333333507180214, 0.03333333507180214, 0.02222222276031971, 0.0555555559694767, 0.12222222238779068, 0.02222222276031971, 0.02222222276031971, 0.06666667014360428, 0.0555555559694767, 0.03333333507180214, 0.011111111380159855, 0.02222222276031971, 0.12222222238779068, 0.12222222238779068, 0.18888889253139496, 0.04444444552063942, 0.13333334028720856, 0.02222222276031971, 0.14444445073604584, 0.04444444552063942, 0.13333334028720856, 0.04444444552063942, 0.14444445073604584, 0.13333334028720856, 0.03333333507180214, 0.14444445073604584, 0.24444444477558136, 0.12222222238779068, 0.36666667461395264, 0.1666666716337204, 0.15555556118488312, 0.24444444477558136, 0.15555556118488312, 0.21111111342906952, 0.20000000298023224, 0.2888889014720917, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.5555555820465088, 0.2666666805744171, 0.46666666865348816, 0.5666666626930237, NaN, 0.5333333611488342, 0.855555534362793, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.36666667461395264, 0.6555555462837219, 0.6111111044883728, 0.6555555462837219, 0.9222221970558167, 0.9555555582046509], [0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.011111111380159855, 0.0, 0.011111111380159855, 0.0, 0.011111111380159855, 0.08888889104127884, 0.03333333507180214, 0.011111111380159855, 0.0, 0.02222222276031971, 0.0, 0.03333333507180214, 0.02222222276031971, 0.04444444552063942, 0.011111111380159855, 0.011111111380159855, 0.12222222238779068, 0.07777778059244156, 0.12222222238779068, 0.08888889104127884, 0.15555556118488312, 0.17777778208255768, 0.20000000298023224, 0.14444445073604584, 0.23333333432674408, 0.07777778059244156, 0.1666666716337204, 0.17777778208255768, 0.10000000149011612, 0.03333333507180214, 0.14444445073604584, 0.1111111119389534, 0.24444444477558136, 0.06111111119389534, 0.04444444552063942, 0.011111111380159855, 0.17777778208255768, 0.3222222328186035, 0.23333333432674408, 0.04444444552063942, 0.21111111342906952, 0.3888888955116272, 0.4000000059604645, 0.4444444477558136, 0.4000000059604645, 0.31111112236976624, 0.4888888895511627, 0.46666666865348816, NaN, 0.46666666865348816, 0.4888888895511627, 0.6000000238418579, 0.5111111402511597, 0.5111111402511597, 0.8111110925674438, 0.5888888835906982, 0.6555555462837219, 0.7888888716697693, 1.0], [0.0, 0.0, 0.03333333507180214, 0.0, 0.011111111380159855, 0.0, 0.011111111380159855, 0.011111111380159855, 0.03333333507180214, 0.011111111380159855, 0.03333333507180214, 0.02222222276031971, 0.1111111119389534, 0.011111111380159855, 0.02222222276031971, 0.0, 0.011111111380159855, 0.12222222238779068, 0.03333333507180214, 0.0555555559694767, 0.04444444552063942, 0.011111111380159855, 0.03333333507180214, 0.13333334028720856, 0.011111111380159855, 0.0, 0.07777778059244156, 0.04444444552063942, 0.03333333507180214, 0.02222222276031971, 0.03333333507180214, 0.13333334028720856, 0.06666667014360428, 0.14444445073604584, 0.02222222276031971, 0.14444445073604584, 0.02222222276031971, 0.1666666716337204, 0.0555555559694767, 0.13333334028720856, 0.04444444552063942, 0.13333334028720856, 0.14444445073604584, 0.03333333507180214, 0.1666666716337204, 0.25555557012557983, 0.04444444552063942, 0.3222222328186035, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.2222222238779068, 0.17777778208255768, 0.2777777910232544, 0.2666666805744171, 0.5222222208976746, 0.2666666805744171, 0.42222222685813904, 0.5444444417953491, 0.14444445073604584, 0.5333333611488342, NaN, 0.6000000238418579, 0.5444444417953491, 0.5333333611488342, 0.31111112236976624, 0.644444465637207, 0.5444444417953491, 0.5777778029441833, 0.8999999761581421, 0.9555555582046509], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.0, 0.011111111380159855, 0.0, 0.0, 0.011111111380159855, 0.0, 0.04444444552063942, 0.011111111380159855, 0.0, 0.011111111380159855, 0.02222222276031971, 0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.02222222276031971, 0.011111111380159855, 0.10000000149011612, 0.011111111380159855, 0.0, 0.12222222238779068, 0.011111111380159855, 0.03333333507180214, 0.0, 0.12222222238779068, 0.02222222276031971, 0.13333334028720856, 0.13333334028720856, 0.14444445073604584, 0.02222222276031971, 0.2666666805744171, 0.13333334028720856, 0.08888889104127884, 0.06666667014360428, 0.07777778059244156, 0.0555555559694767, 0.36666667461395264, 0.07777778059244156, 0.08888889104127884, 0.0555555559694767, 0.13333334028720856, 0.2777777910232544, 0.3222222328186035, 0.06666667014360428, 0.07777778059244156, 0.36666667461395264, 0.35555556416511536, 0.41111111640930176, 0.35555556416511536, 0.42222222685813904, 0.4333333373069763, 0.4000000059604645, 0.5111111402511597, 0.4000000059604645, NaN, 0.6111111044883728, 0.5, 0.5444444417953491, 0.6333333253860474, 0.6333333253860474, 0.4888888895511627, 0.644444465637207, 0.9444444179534912], [0.0, 0.011111111380159855, 0.011111111380159855, 0.0, 0.0, 0.0, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.011111111380159855, 0.0, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.011111111380159855, 0.011111111380159855, 0.011111111380159855, 0.0, 0.03333333507180214, 0.03333333507180214, 0.07777778059244156, 0.011111111380159855, 0.011111111380159855, 0.1111111119389534, 0.07777778059244156, 0.02222222276031971, 0.07777778059244156, 0.1666666716337204, 0.0555555559694767, 0.0555555559694767, 0.04444444552063942, 0.14444445073604584, 0.10000000149011612, 0.06666667014360428, 0.17777778208255768, 0.06666667014360428, 0.03333333507180214, 0.15555556118488312, 0.0555555559694767, 0.2888889014720917, 0.07777778059244156, 0.06666667014360428, 0.02222222276031971, 0.14444445073604584, 0.3222222328186035, 0.17777778208255768, 0.0555555559694767, 0.13333334028720856, 0.3777777850627899, 0.30000001192092896, 0.4555555582046509, 0.30000001192092896, 0.4333333373069763, 0.4888888895511627, 0.4000000059604645, 0.4000000059604645, 0.4555555582046509, 0.3888888955116272, NaN, 0.5222222208976746, 0.4555555582046509, 0.7111111283302307, 0.5888888835906982, 0.6555555462837219, 0.7888888716697693, 1.0], [0.0, 0.0, 0.0, 0.04444444552063942, 0.03333333507180214, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.02222222276031971, 0.02222222276031971, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.03333333507180214, 0.0, 0.07777778059244156, 0.06666667014360428, 0.0, 0.011111111380159855, 0.12222222238779068, 0.1111111119389534, 0.04444444552063942, 0.0555555559694767, 0.10000000149011612, 0.0, 0.18888889253139496, 0.07777778059244156, 0.10000000149011612, 0.0, 0.08888889104127884, 0.0555555559694767, 0.07777778059244156, 0.02222222276031971, 0.08888889104127884, 0.20000000298023224, 0.04444444552063942, 0.1111111119389534, 0.18888889253139496, 0.3444444537162781, 0.08888889104127884, 0.13333334028720856, 0.10000000149011612, 0.02222222276031971, 0.25555557012557983, 0.10000000149011612, 0.1111111119389534, 0.12222222238779068, 0.3222222328186035, 0.25555557012557983, 0.06666667014360428, 0.25555557012557983, 0.2777777910232544, 0.13333334028720856, 0.4333333373069763, 0.4888888895511627, 0.46666666865348816, 0.5, 0.47777777910232544, NaN, 0.47777777910232544, 0.7111111283302307, 0.5222222208976746, 0.6222222447395325, 0.7444444298744202, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.07777778059244156, 0.0, 0.0, 0.011111111380159855, 0.0, 0.0, 0.02222222276031971, 0.13333334028720856, 0.0, 0.08888889104127884, 0.06666667014360428, 0.011111111380159855, 0.0, 0.3777777850627899, 0.06666667014360428, 0.0, 0.14444445073604584, 0.04444444552063942, 0.2888889014720917, 0.12222222238779068, 0.15555556118488312, 0.10000000149011612, 0.30000001192092896, 0.0, 0.5, 0.0, 0.4000000059604645, 0.5, 0.6333333253860474, 0.4888888895511627, 0.6888889074325562, 0.4555555582046509, 0.5444444417953491, 0.5222222208976746, NaN, 0.6333333253860474, 0.6777777671813965, 0.699999988079071, 0.800000011920929, 0.8777777552604675], [0.0, 0.04444444552063942, 0.0, 0.07777778059244156, 0.07777778059244156, 0.07777778059244156, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.04444444552063942, 0.0, 0.0, 0.07777778059244156, 0.08888889104127884, 0.0, 0.0, 0.04444444552063942, 0.10000000149011612, 0.02222222276031971, 0.10000000149011612, 0.08888889104127884, 0.0, 0.10000000149011612, 0.06666667014360428, 0.011111111380159855, 0.04444444552063942, 0.08888889104127884, 0.07777778059244156, 0.10000000149011612, 0.17777778208255768, 0.10000000149011612, 0.0, 0.10000000149011612, 0.18888889253139496, 0.10000000149011612, 0.011111111380159855, 0.12222222238779068, 0.18888889253139496, 0.1111111119389534, 0.0, 0.20000000298023224, 0.10000000149011612, 0.1111111119389534, 0.1111111119389534, 0.10000000149011612, 0.2222222238779068, 0.14444445073604584, 0.10000000149011612, 0.08888889104127884, 0.1666666716337204, 0.24444444477558136, 0.1111111119389534, 0.10000000149011612, 0.1111111119389534, 0.18888889253139496, 0.30000001192092896, 0.2222222238779068, 0.30000001192092896, 0.2888889014720917, 0.2777777910232544, 0.3444444537162781, 0.18888889253139496, 0.35555556416511536, 0.36666667461395264, 0.2888889014720917, 0.2888889014720917, 0.36666667461395264, NaN, 0.4333333373069763, 0.4888888895511627, 0.5333333611488342, 0.8666666746139526], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.011111111380159855, 0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.011111111380159855, 0.02222222276031971, 0.0, 0.011111111380159855, 0.0, 0.0, 0.02222222276031971, 0.011111111380159855, 0.02222222276031971, 0.1111111119389534, 0.03333333507180214, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.0, 0.011111111380159855, 0.04444444552063942, 0.04444444552063942, 0.011111111380159855, 0.03333333507180214, 0.011111111380159855, 0.06666667014360428, 0.02222222276031971, 0.10000000149011612, 0.03333333507180214, 0.03333333507180214, 0.07777778059244156, 0.3444444537162781, 0.02222222276031971, 0.07777778059244156, 0.03333333507180214, 0.03333333507180214, 0.14444445073604584, 0.13333334028720856, 0.03333333507180214, 0.04444444552063942, 0.20000000298023224, 0.12222222238779068, 0.31111112236976624, 0.12222222238779068, 0.3333333432674408, 0.3444444537162781, 0.3888888955116272, 0.41111111640930176, 0.4555555582046509, 0.36666667461395264, 0.41111111640930176, 0.47777777910232544, 0.3222222328186035, 0.5666666626930237, NaN, 0.5555555820465088, 0.644444465637207, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011111111380159855, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.02222222276031971, 0.0, 0.10000000149011612, 0.0, 0.3222222328186035, 0.0, 0.0, 0.10000000149011612, 0.0, 0.2666666805744171, 0.1111111119389534, 0.08888889104127884, 0.011111111380159855, 0.18888889253139496, 0.12222222238779068, 0.3777777850627899, 0.12222222238779068, 0.0, 0.3777777850627899, 0.3444444537162781, 0.3444444537162781, 0.42222222685813904, 0.5111111402511597, 0.3444444537162781, 0.3777777850627899, 0.30000001192092896, 0.5111111402511597, 0.4444444477558136, NaN, 0.7111111283302307, 0.9888888597488403], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07777778059244156, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.0, 0.21111111342906952, 0.10000000149011612, 0.07777778059244156, 0.0, 0.10000000149011612, 0.02222222276031971, 0.04444444552063942, 0.011111111380159855, 0.04444444552063942, 0.1111111119389534, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.15555556118488312, 0.25555557012557983, 0.07777778059244156, 0.21111111342906952, 0.10000000149011612, 0.35555556416511536, 0.21111111342906952, 0.25555557012557983, 0.20000000298023224, 0.46666666865348816, 0.35555556416511536, 0.2888889014720917, NaN, 0.8777777552604675], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.04444444552063942, 0.0, 0.04444444552063942, 0.0555555559694767, 0.0, 0.1111111119389534, 0.12222222238779068, 0.13333334028720856, 0.20000000298023224, 0.011111111380159855, 0.12222222238779068, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9452054721},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9377473299},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8861491562},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8732115616},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8202435257},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8136986271},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7848554006},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7805174988},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7802130879},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7748858436},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7561643813},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7526636235},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7519025854},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7394216134},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7293759524},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7260273956},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7187214613},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.7033485538},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6996955841},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.6936073061},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6691019812},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6634703191},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6610350058},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.6599695588},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6520547965},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6328767152},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6238964996},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6207762578},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6165905655},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5940639282},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5922374447},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5791476407},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5613394215},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.529984781},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5251141585},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5136986321},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5135464266},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.50015221},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4770167455},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.4715372923},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4677321191},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4646879792},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4628614962},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4446727578},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4442922394},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.4266362271},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4240487081},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4149162916},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4127092872},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4057838691},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3920852384},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3907153768},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.3687214643},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3479452103},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3471841738},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3277016777},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2925418603},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2499238986},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2471080708},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.243531205},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2286149208},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2169710838},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2013698657},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.1813546444},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.177625575},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.1627092868},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1624048727},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1523592111},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.144444445},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1432267917},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1175038078},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1053272472},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0864535782},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0185692545}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..a4112e79d76d37df84069fdf77068b27756b6027
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,"AutoGluon 1.5 (extreme, 4h)",TabFM (default),"AutoGluon 1.4 (best, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),TabICLv2 (default),CatBoost (tuned + ensembled),CatBoost (default),iLTM (tuned + ensembled),CatBoost (tuned),XGBoost (tuned + ensembled),LightGBM (tuned + ensembled),ChimeraBoost (tuned + ensembled),RealMLP (tuned + ensembled),RealTabPFN-2.5 (tuned + ensembled),TabM (tuned + ensembled),ModernNCA (tuned + ensembled),ChimeraBoost (tuned),XGBoost (tuned),RealTabPFN-2.5 (tuned),RealMLP (tuned),TabM (tuned),LightGBM (tuned),ChimeraBoost (default),RealTabPFN-2.5 (default),TabICL (default),iLTM (tuned),ModernNCA (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),TorchMLP (tuned + ensembled),EBM (tuned + ensembled),XGBoost (default),xRFM (tuned + ensembled),TabSwift (default),EBM (tuned),TabM (default),xRFM (tuned),RealMLP (default),BetaTabPFN (default),TorchMLP (tuned),FastaiMLP (tuned + ensembled),EBM (default),TabDPT (tuned),RandomForest (tuned + ensembled),iLTM (default),LightGBM (default),LimiX (default),RandomForest (tuned),ModernNCA (default),ExtraTrees (tuned + ensembled),SAP-RPT-OSS (default),TabDPT (default),FastaiMLP (tuned),ExtraTrees (tuned),TorchMLP (default),FastaiMLP (default),TabSTAR (tuned),Linear (tuned + ensembled),TabSTAR (tuned + ensembled),KNN (tuned + ensembled),Linear (tuned),PerpetualBooster (tuned + ensembled),RandomForest (default),PerpetualBooster (tuned),OrionMSP (default),ExtraTrees (default),Linear (default),TabSTAR (default),xRFM (default),TabFlex (default),KNN (tuned),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.4,0.74444443,0.44444445,0.82222223,0.67777777,0.9222222,0.9,0.93333334,0.9111111,0.95555556,0.93333334,0.9222222,0.9777778,0.9,0.8111111,0.96666664,0.93333334,0.98888886,0.9222222,0.96666664,0.8666667,0.9777778,0.96666664,0.9222222,0.9,0.8888889,0.98888886,0.9111111,0.9222222,0.98888886,0.98888886,0.96666664,0.9777778,0.9444444,1.0,1.0,0.98888886,1.0,1.0,0.9777778,1.0,1.0,0.95555556,1.0,1.0,1.0,0.8888889,0.98888886,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.98888886,1.0,0.98888886,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.6,,0.6333333,0.8111111,0.9,0.9111111,0.7777778,0.7888889,0.84444445,0.82222223,0.87777776,0.85555553,0.87777776,0.82222223,0.8666667,0.98888886,0.8888889,0.87777776,0.9111111,0.87777776,0.8333333,1.0,0.96666664,0.9,0.9777778,0.95555556,0.9,0.9444444,0.96666664,0.96666664,0.96666664,0.98888886,0.9777778,0.8888889,0.95555556,1.0,1.0,0.9111111,1.0,1.0,1.0,0.9111111,1.0,0.96666664,0.9777778,1.0,1.0,0.85555553,0.9777778,1.0,0.98888886,0.9777778,0.9777778,0.9777778,0.98888886,1.0,0.9777778,1.0,0.9777778,1.0,0.95555556,0.9777778,1.0,0.98888886,1.0,1.0,0.98888886,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.25555557,0.36666667,,0.42222223,0.7111111,0.65555555,0.76666665,0.7222222,0.84444445,0.7777778,0.8666667,0.84444445,0.7777778,0.8666667,0.64444447,0.74444443,0.7888889,0.8666667,0.8888889,0.74444443,0.9222222,0.8,0.9,0.8888889,0.76666665,0.76666665,0.85555553,0.9,0.8111111,0.8,0.9111111,0.95555556,0.9222222,0.95555556,0.84444445,0.95555556,0.9222222,0.96666664,0.96666664,0.9111111,0.9777778,0.96666664,0.95555556,0.96666664,0.98888886,0.96666664,0.95555556,0.87777776,0.9777778,0.98888886,0.98888886,0.96666664,0.95555556,0.9777778,1.0,0.98888886,0.98888886,1.0,1.0,1.0,0.98888886,1.0,0.9777778,1.0,0.96666664,1.0,0.98888886,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.5555556,0.18888889,0.5777778,,0.75555557,0.5777778,0.7,0.6888889,0.7111111,0.6888889,0.74444443,0.73333335,0.7777778,0.74444443,0.85555553,0.7888889,0.8888889,0.76666665,0.7777778,0.87777776,0.75555557,0.85555553,0.9,0.76666665,0.87777776,0.8666667,0.8666667,0.87777776,0.95555556,0.8888889,0.8333333,0.8666667,0.9111111,0.7888889,0.96666664,0.9111111,0.98888886,0.8,0.9444444,0.9444444,0.96666664,0.9,0.95555556,0.96666664,0.9444444,0.98888886,0.98888886,0.8,0.96666664,0.9777778,0.96666664,0.9222222,0.96666664,0.93333334,0.96666664,0.96666664,0.93333334,1.0,0.9222222,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,0.9222222,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.17777778,0.1,0.2888889,0.24444444,,0.56666666,0.56666666,0.5222222,0.5555556,0.56666666,0.5777778,0.6888889,0.67777777,0.62222224,0.7,0.62222224,0.7,0.7111111,0.73333335,0.7,0.7111111,0.74444443,0.82222223,0.67777777,0.85555553,0.7777778,0.75555557,0.82222223,0.85555553,0.85555553,0.84444445,0.87777776,0.9111111,0.76666665,0.8666667,0.96666664,0.9222222,0.8111111,1.0,0.9,0.95555556,0.8666667,0.96666664,0.95555556,0.95555556,0.9444444,0.9444444,0.7777778,0.98888886,0.9777778,0.9777778,0.95555556,0.9444444,0.9444444,0.98888886,1.0,0.9222222,1.0,0.96666664,1.0,0.9444444,0.96666664,0.98888886,1.0,0.98888886,1.0,1.0,0.96666664,1.0,0.9222222,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.32222223,0.08888889,0.34444445,0.42222223,0.43333334,,0.6111111,0.6111111,0.6333333,0.6,0.62222224,0.6666667,0.62222224,0.5777778,0.73333335,0.6111111,0.7111111,0.62222224,0.64444447,0.75555557,0.6333333,0.7222222,0.76666665,0.64444447,0.84444445,0.8333333,0.75555557,0.8,0.87777776,0.9,0.8111111,0.75555557,0.8666667,0.7,0.9111111,0.9,0.9444444,0.7222222,0.9111111,0.9222222,0.9111111,0.84444445,0.93333334,0.9444444,0.95555556,0.9222222,0.9444444,0.8111111,0.96666664,0.95555556,0.98888886,0.9222222,0.95555556,0.93333334,0.98888886,0.9777778,0.9222222,0.9777778,0.9222222,0.9777778,0.95555556,0.9444444,1.0,1.0,1.0,1.0,1.0,0.96666664,1.0,0.9222222,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.07777778,0.22222222,0.23333333,0.3,0.43333334,0.3888889,,0.5888889,0.5777778,0.5888889,0.6111111,0.51111114,0.5888889,0.62222224,0.5222222,0.6,0.5888889,0.73333335,0.6888889,0.53333336,0.7111111,0.64444447,0.64444447,0.7888889,0.73333335,0.6888889,0.74444443,0.73888886,0.6,0.6666667,0.7888889,0.84444445,0.82222223,0.9222222,0.7777778,0.82222223,0.76666665,0.9444444,0.8333333,0.84444445,0.95555556,0.8666667,0.9444444,0.8111111,0.8888889,0.9222222,0.95555556,0.7222222,0.87777776,0.95555556,0.98888886,0.9777778,0.8888889,0.96666664,0.9777778,0.98888886,0.9777778,1.0,0.98888886,1.0,0.98888886,1.0,0.9777778,0.98888886,0.98888886,0.98888886,0.98888886,1.0,1.0,1.0,0.98888886,0.98888886,1.0,1.0
+CatBoost (default),0.1,0.21111111,0.2777778,0.31111112,0.47777778,0.3888889,0.41111112,,0.5222222,0.47777778,0.53333336,0.54444444,0.6,0.62222224,0.5222222,0.6,0.5888889,0.7222222,0.65555555,0.54444444,0.67777777,0.65555555,0.6888889,0.8111111,0.7111111,0.7111111,0.75555557,0.7111111,0.5888889,0.6888889,0.7888889,0.8111111,0.7888889,0.9444444,0.8111111,0.8111111,0.7888889,0.95555556,0.82222223,0.85555553,0.95555556,0.8333333,0.8888889,0.8111111,0.87777776,0.9444444,0.93333334,0.75555557,0.87777776,0.98888886,0.96666664,0.95555556,0.9,0.9777778,0.98888886,0.9777778,0.95555556,1.0,0.98888886,1.0,0.98888886,1.0,0.98888886,0.98888886,0.98888886,0.98888886,0.9777778,1.0,1.0,1.0,0.98888886,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.06666667,0.15555556,0.15555556,0.2888889,0.44444445,0.36666667,0.42222223,0.47777778,,0.44444445,0.53333336,0.56666666,0.5,0.54444444,0.5777778,0.5777778,0.56666666,0.5888889,0.67777777,0.62222224,0.65555555,0.65555555,0.67777777,0.64444447,0.73333335,0.73333335,0.7111111,0.7777778,0.6111111,0.74444443,0.8666667,0.8111111,0.7888889,0.8888889,0.76666665,0.82222223,0.8333333,0.95555556,0.95555556,0.8666667,0.95555556,0.96666664,0.9111111,0.9111111,0.87777776,1.0,0.9222222,0.75555557,0.8888889,0.9777778,0.95555556,0.9777778,0.93333334,0.96666664,0.98888886,1.0,1.0,1.0,0.98888886,1.0,0.98888886,1.0,0.9777778,0.98888886,0.96666664,0.98888886,0.98888886,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.08888889,0.17777778,0.22222222,0.31111112,0.43333334,0.4,0.41111112,0.5222222,0.5555556,,0.62222224,0.46666667,0.62222224,0.5777778,0.4888889,0.6111111,0.6,0.73333335,0.7111111,0.53333336,0.7,0.62222224,0.6666667,0.75555557,0.6888889,0.7,0.7111111,0.6888889,0.5888889,0.6666667,0.7777778,0.8333333,0.8111111,0.93333334,0.76666665,0.82222223,0.8,0.93333334,0.84444445,0.85555553,0.9444444,0.82222223,0.96666664,0.8,0.8888889,0.93333334,0.9222222,0.7222222,0.87777776,0.93333334,0.95555556,0.96666664,0.87777776,0.95555556,0.96666664,0.9777778,0.9444444,0.98888886,0.98888886,0.98888886,0.98888886,1.0,0.9777778,0.98888886,0.98888886,0.9777778,0.9777778,1.0,1.0,1.0,0.98888886,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.044444446,0.12222222,0.13333334,0.25555557,0.42222223,0.37777779,0.3888889,0.46666667,0.46666667,0.37777779,,0.45555556,0.5222222,0.5,0.5,0.62222224,0.5222222,0.5777778,0.8666667,0.53333336,0.5888889,0.67777777,0.62222224,0.6666667,0.65555555,0.6,0.6333333,0.64444447,0.62222224,0.6888889,0.8,0.8111111,0.93333334,0.82222223,0.6888889,0.84444445,0.8111111,0.9111111,0.84444445,0.87777776,0.84444445,0.8,0.9777778,0.84444445,0.9,0.96666664,0.96666664,0.67777777,0.9,0.98888886,0.98888886,0.95555556,0.8888889,0.87777776,0.98888886,1.0,0.9,0.98888886,1.0,0.98888886,0.98888886,1.0,0.9222222,1.0,0.96666664,1.0,0.98888886,1.0,1.0,0.98888886,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.06666667,0.14444445,0.15555556,0.26666668,0.31111112,0.33333334,0.4888889,0.45555556,0.43333334,0.53333336,0.54444444,,0.47777778,0.47777778,0.46666667,0.53333336,0.51111114,0.6111111,0.6888889,0.5,0.56666666,0.6333333,0.8666667,0.62222224,0.62222224,0.6666667,0.7,0.65555555,0.6666667,0.6111111,0.73333335,0.76666665,0.82222223,0.76666665,0.74444443,0.8666667,0.8,0.8333333,0.82222223,0.8666667,0.9111111,0.76666665,0.96666664,0.8,0.9,0.9777778,0.9777778,0.6888889,0.9111111,0.9777778,0.9777778,0.95555556,0.95555556,0.9111111,0.9777778,0.9777778,0.8666667,0.98888886,0.9777778,0.98888886,0.9777778,0.98888886,0.9777778,1.0,0.9777778,0.98888886,0.98888886,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.07777778,0.12222222,0.22222222,0.22222222,0.32222223,0.37777779,0.41111112,0.4,0.5,0.37777779,0.47777778,0.5222222,,0.62222224,0.51111114,0.6111111,0.5222222,0.75555557,0.5888889,0.56666666,0.7222222,0.6888889,0.56666666,0.87777776,0.6333333,0.6333333,0.73333335,0.7111111,0.6111111,0.7,0.7111111,0.82222223,0.67777777,0.9,0.6666667,0.85555553,0.82222223,0.9222222,0.85555553,0.87777776,0.8666667,0.85555553,0.93333334,0.8333333,0.8888889,0.95555556,0.87777776,0.74444443,0.8888889,0.9111111,0.8888889,0.9777778,0.8333333,0.8888889,0.8888889,0.9111111,0.87777776,1.0,1.0,1.0,0.98888886,1.0,0.87777776,1.0,0.8888889,1.0,0.98888886,1.0,1.0,1.0,1.0,1.0,0.9222222,1.0
+RealMLP (tuned + ensembled),0.022222223,0.17777778,0.13333334,0.25555557,0.37777779,0.42222223,0.37777779,0.37777779,0.45555556,0.42222223,0.5,0.5222222,0.37777779,,0.5555556,0.42222223,0.53333336,0.4888889,0.6,0.5888889,0.75555557,0.47777778,0.67777777,0.4888889,0.6111111,0.5777778,0.62222224,0.6888889,0.6666667,0.6111111,0.75555557,0.75555557,0.74444443,0.84444445,0.65555555,0.82222223,0.9,0.8888889,1.0,0.85555553,0.8333333,0.9777778,0.84444445,0.8111111,0.87777776,0.8111111,0.93333334,0.84444445,0.8666667,0.85555553,0.9,0.8888889,0.82222223,0.9777778,0.87777776,0.98888886,0.96666664,1.0,1.0,1.0,0.98888886,1.0,0.9777778,0.93333334,0.98888886,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.1,0.13333334,0.35555556,0.14444445,0.3,0.26666668,0.47777778,0.47777778,0.42222223,0.51111114,0.5,0.53333336,0.4888889,0.44444445,,0.43333334,0.46666667,0.5222222,0.54444444,0.8,0.5555556,0.5555556,0.64444447,0.6111111,0.7111111,0.7,0.62222224,0.6888889,0.64444447,0.6888889,0.7111111,0.73333335,0.76666665,0.65555555,0.85555553,0.82222223,0.74444443,0.7111111,0.82222223,0.8,0.82222223,0.73333335,0.8666667,0.8888889,0.9111111,0.85555553,0.9222222,0.6666667,0.95555556,0.82222223,0.96666664,0.85555553,0.9,0.87777776,0.96666664,0.95555556,0.9,1.0,0.84444445,1.0,0.9111111,0.85555553,0.96666664,1.0,0.9777778,0.98888886,0.98888886,0.95555556,1.0,0.9222222,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.18888889,0.011111111,0.25555557,0.21111111,0.37777779,0.3888889,0.4,0.4,0.42222223,0.3888889,0.37777779,0.46666667,0.3888889,0.5777778,0.56666666,,0.5222222,0.45555556,0.5,0.5888889,0.65555555,0.73333335,0.6111111,0.5222222,0.56666666,0.5777778,0.6333333,0.73333335,0.7111111,0.65555555,0.6111111,0.6666667,0.6666667,0.65555555,0.75555557,0.84444445,0.96666664,0.65555555,0.9777778,0.9111111,0.7777778,0.82222223,0.8333333,0.87777776,0.8333333,0.8333333,1.0,0.76666665,0.8333333,0.9,0.8333333,0.87777776,0.87777776,0.9111111,0.8666667,0.9777778,0.9,1.0,0.93333334,1.0,0.9444444,0.9444444,0.9777778,0.98888886,1.0,1.0,1.0,0.9777778,1.0,0.9111111,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.033333335,0.11111111,0.21111111,0.11111111,0.3,0.2888889,0.41111112,0.41111112,0.43333334,0.4,0.47777778,0.4888889,0.47777778,0.46666667,0.53333336,0.47777778,,0.5555556,0.5,0.54444444,0.5888889,0.56666666,0.53333336,0.6333333,0.62222224,0.6333333,0.64444447,0.74444443,0.6111111,0.6888889,0.73333335,0.73333335,0.65555555,0.7111111,0.7,0.75555557,0.82222223,0.7777778,0.8888889,0.82222223,0.84444445,0.8666667,0.82222223,0.87777776,0.7888889,0.84444445,0.9111111,0.85555553,0.7777778,0.95555556,0.9,0.85555553,0.9222222,0.93333334,0.8888889,0.8888889,0.96666664,0.9222222,0.9,0.9222222,0.98888886,0.9111111,0.98888886,1.0,0.98888886,0.95555556,0.9777778,0.9777778,1.0,1.0,0.93333334,1.0,1.0,1.0
+ChimeraBoost (tuned),0.06666667,0.12222222,0.13333334,0.23333333,0.2888889,0.37777779,0.26666668,0.2777778,0.41111112,0.26666668,0.42222223,0.3888889,0.24444444,0.51111114,0.47777778,0.54444444,0.44444445,,0.5222222,0.4888889,0.67777777,0.6,0.51111114,0.7111111,0.5777778,0.5777778,0.64444447,0.6666667,0.5222222,0.65555555,0.64444447,0.7222222,0.65555555,0.8666667,0.6333333,0.7777778,0.7888889,0.9,0.84444445,0.85555553,0.82222223,0.8111111,0.85555553,0.8111111,0.8666667,0.9,0.8111111,0.73333335,0.8666667,0.85555553,0.8666667,0.93333334,0.7777778,0.85555553,0.8666667,0.9111111,0.8666667,0.98888886,1.0,0.98888886,0.9777778,1.0,0.87777776,0.98888886,0.87777776,0.98888886,0.98888886,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0
+XGBoost (tuned),0.011111111,0.08888889,0.11111111,0.22222222,0.26666668,0.35555556,0.31111112,0.34444445,0.32222223,0.2888889,0.13333334,0.31111112,0.41111112,0.4,0.45555556,0.5,0.5,0.47777778,,0.46666667,0.46666667,0.6111111,0.51111114,0.6,0.56666666,0.54444444,0.56666666,0.6333333,0.5777778,0.6111111,0.67777777,0.7222222,0.76666665,0.74444443,0.6666667,0.8111111,0.7777778,0.8111111,0.82222223,0.85555553,0.82222223,0.75555557,0.9222222,0.8,0.8666667,0.93333334,0.93333334,0.65555555,0.85555553,0.96666664,0.95555556,0.9444444,0.85555553,0.8666667,0.9777778,1.0,0.85555553,0.98888886,1.0,0.98888886,0.98888886,1.0,0.9111111,1.0,0.96666664,1.0,0.98888886,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.07777778,0.12222222,0.25555557,0.12222222,0.3,0.24444444,0.46666667,0.45555556,0.37777779,0.46666667,0.46666667,0.5,0.43333334,0.41111112,0.2,0.41111112,0.45555556,0.51111114,0.53333336,,0.5222222,0.53333336,0.64444447,0.56666666,0.6666667,0.6666667,0.5777778,0.6111111,0.5888889,0.6888889,0.65555555,0.6666667,0.75555557,0.65555555,0.8333333,0.73333335,0.67777777,0.6888889,0.76666665,0.75555557,0.8111111,0.7,0.8333333,0.82222223,0.85555553,0.82222223,0.8888889,0.65555555,0.93333334,0.7888889,0.9111111,0.8333333,0.84444445,0.87777776,0.93333334,0.9111111,0.8888889,0.98888886,0.82222223,1.0,0.8888889,0.8666667,0.96666664,0.98888886,0.9444444,0.98888886,0.9777778,0.93333334,1.0,0.9,0.98888886,1.0,1.0,1.0
+RealMLP (tuned),0.033333335,0.16666667,0.07777778,0.24444444,0.2888889,0.36666667,0.2888889,0.32222223,0.34444445,0.3,0.41111112,0.43333334,0.2777778,0.24444444,0.44444445,0.34444445,0.41111112,0.32222223,0.53333336,0.47777778,,0.4,0.6,0.43333334,0.5222222,0.47777778,0.5777778,0.5777778,0.4888889,0.5555556,0.67777777,0.64444447,0.67777777,0.75555557,0.54444444,0.7111111,0.76666665,0.8,0.9,0.7777778,0.8111111,0.87777776,0.73333335,0.73333335,0.84444445,0.73333335,0.82222223,0.8,0.82222223,0.75555557,0.8,0.7888889,0.74444443,0.96666664,0.85555553,0.93333334,0.95555556,0.9777778,0.9777778,0.9777778,0.96666664,0.98888886,0.96666664,0.9111111,0.95555556,0.9777778,0.98888886,1.0,1.0,0.9777778,0.9777778,1.0,1.0,1.0
+TabM (tuned),0.13333334,0.0,0.2,0.14444445,0.25555557,0.2777778,0.35555556,0.34444445,0.34444445,0.37777779,0.32222223,0.36666667,0.31111112,0.5222222,0.44444445,0.26666668,0.43333334,0.4,0.3888889,0.46666667,0.6,,0.44444445,0.41111112,0.4888889,0.4888889,0.5888889,0.6333333,0.5888889,0.54444444,0.56666666,0.5555556,0.65555555,0.6111111,0.65555555,0.76666665,0.84444445,0.65555555,0.9,0.8,0.74444443,0.8,0.7888889,0.76666665,0.7777778,0.75555557,0.8666667,0.76666665,0.7888889,0.8333333,0.76666665,0.8,0.8,0.87777776,0.8333333,0.9777778,0.9111111,1.0,0.8888889,1.0,0.8888889,0.9,0.9777778,0.96666664,0.98888886,1.0,0.9777778,0.93333334,1.0,0.9,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.022222223,0.033333335,0.1,0.1,0.17777778,0.23333333,0.35555556,0.31111112,0.32222223,0.33333334,0.37777779,0.13333334,0.43333334,0.32222223,0.35555556,0.3888889,0.46666667,0.4888889,0.4888889,0.35555556,0.4,0.5555556,,0.54444444,0.4888889,0.5222222,0.54444444,0.5555556,0.6,0.5777778,0.5777778,0.67777777,0.74444443,0.7222222,0.6333333,0.73333335,0.75555557,0.7222222,0.76666665,0.7888889,0.7,0.75555557,0.85555553,0.74444443,0.7888889,0.93333334,0.96666664,0.64444447,0.84444445,0.93333334,0.8333333,0.82222223,0.8333333,0.7888889,0.9222222,0.93333334,0.8111111,1.0,0.8666667,1.0,0.9111111,0.9111111,0.9444444,0.98888886,0.96666664,0.98888886,0.98888886,0.96666664,1.0,0.9111111,0.98888886,1.0,1.0,1.0
+ChimeraBoost (default),0.033333335,0.1,0.11111111,0.23333333,0.32222223,0.35555556,0.21111111,0.18888889,0.35555556,0.24444444,0.33333334,0.37777779,0.12222222,0.51111114,0.3888889,0.47777778,0.36666667,0.2888889,0.4,0.43333334,0.56666666,0.5888889,0.45555556,,0.43333334,0.45555556,0.56666666,0.54444444,0.53333336,0.56666666,0.62222224,0.65555555,0.6111111,0.8,0.6,0.7222222,0.67777777,0.87777776,0.75555557,0.75555557,0.74444443,0.8111111,0.76666665,0.8333333,0.84444445,0.8333333,0.75555557,0.73333335,0.8111111,0.8111111,0.76666665,0.95555556,0.82222223,0.8666667,0.84444445,0.8666667,0.8666667,0.98888886,1.0,1.0,0.9777778,1.0,0.87777776,1.0,0.8666667,1.0,0.98888886,1.0,1.0,1.0,1.0,1.0,0.9,1.0
+RealTabPFN-2.5 (default),0.07777778,0.022222223,0.23333333,0.12222222,0.14444445,0.15555556,0.26666668,0.2888889,0.26666668,0.31111112,0.34444445,0.37777779,0.36666667,0.3888889,0.2888889,0.43333334,0.37777779,0.42222223,0.43333334,0.33333334,0.47777778,0.51111114,0.51111114,0.56666666,,0.6111111,0.62222224,0.5222222,0.44444445,0.62222224,0.62222224,0.53333336,0.6888889,0.67777777,0.8666667,0.6333333,0.5888889,0.6888889,0.7111111,0.7888889,0.8,0.6666667,0.7222222,0.7777778,0.74444443,0.8333333,0.87777776,0.6666667,0.8,0.8,0.9222222,0.9,0.85555553,0.8,0.9444444,0.95555556,0.84444445,1.0,0.84444445,1.0,0.9,0.84444445,0.9777778,0.9777778,0.98888886,1.0,0.98888886,0.9222222,1.0,0.9,1.0,1.0,1.0,1.0
+TabICL (default),0.1,0.044444446,0.23333333,0.13333334,0.22222222,0.16666667,0.31111112,0.2888889,0.26666668,0.3,0.4,0.33333334,0.36666667,0.42222223,0.3,0.42222223,0.36666667,0.42222223,0.45555556,0.33333334,0.5222222,0.51111114,0.47777778,0.54444444,0.3888889,,0.53333336,0.44444445,0.53333336,0.64444447,0.6333333,0.62222224,0.6111111,0.65555555,0.65555555,0.6333333,0.6111111,0.7,0.67777777,0.6888889,0.67777777,0.6888889,0.67777777,0.85555553,0.7111111,0.76666665,0.8666667,0.6888889,0.7222222,0.8666667,0.8,0.7777778,0.9222222,0.7111111,0.8333333,0.8,0.7888889,0.93333334,0.75555557,0.9444444,0.85555553,0.75555557,0.9777778,1.0,1.0,1.0,1.0,0.93333334,1.0,0.93333334,0.9777778,1.0,1.0,1.0
+iLTM (tuned),0.11111111,0.1,0.14444445,0.13333334,0.24444444,0.24444444,0.25555557,0.24444444,0.2888889,0.2888889,0.36666667,0.3,0.26666668,0.37777779,0.37777779,0.36666667,0.35555556,0.35555556,0.43333334,0.42222223,0.42222223,0.41111112,0.45555556,0.43333334,0.37777779,0.46666667,,0.43333334,0.5,0.5555556,0.4888889,0.51111114,0.5,0.7111111,0.5777778,0.5777778,0.62222224,0.7111111,0.65555555,0.65555555,0.6111111,0.73333335,0.6,0.7888889,0.73333335,0.76666665,0.8,0.76666665,0.75555557,0.84444445,0.74444443,0.84444445,0.85555553,0.85555553,0.8666667,0.8888889,0.9,0.9444444,0.9,0.9444444,0.98888886,0.93333334,0.93333334,0.96666664,0.9222222,0.9777778,0.96666664,1.0,0.98888886,0.98888886,0.98888886,1.0,1.0,1.0
+ModernNCA (tuned),0.011111111,0.055555556,0.1,0.12222222,0.17777778,0.2,0.2611111,0.2888889,0.22222222,0.31111112,0.35555556,0.34444445,0.2888889,0.31111112,0.31111112,0.26666668,0.25555557,0.33333334,0.36666667,0.3888889,0.42222223,0.36666667,0.44444445,0.45555556,0.47777778,0.5555556,0.56666666,,0.53333336,0.5777778,0.6111111,0.5222222,0.53333336,0.5555556,0.64444447,0.5888889,0.62222224,0.62222224,0.7222222,0.6888889,0.7222222,0.6666667,0.64444447,0.7777778,0.7222222,0.76666665,0.76666665,0.7222222,0.7111111,0.9,0.84444445,0.7888889,0.8666667,0.9222222,0.85555553,0.85555553,0.9111111,0.95555556,0.8666667,0.95555556,0.95555556,0.9,0.9444444,0.9777778,0.95555556,0.98888886,0.96666664,0.98888886,1.0,0.95555556,0.9777778,1.0,1.0,1.0
+TabDPT-Turbo (default),0.08888889,0.033333335,0.18888889,0.044444446,0.14444445,0.12222222,0.4,0.41111112,0.3888889,0.41111112,0.37777779,0.33333334,0.3888889,0.33333334,0.35555556,0.2888889,0.3888889,0.47777778,0.42222223,0.41111112,0.51111114,0.41111112,0.4,0.46666667,0.5555556,0.46666667,0.5,0.46666667,,0.5888889,0.64444447,0.51111114,0.6,0.62222224,0.7111111,0.5777778,0.56666666,0.65555555,0.7,0.6111111,0.73333335,0.64444447,0.7111111,0.8333333,0.67777777,0.6888889,0.7888889,0.76666665,0.75555557,0.7,0.76666665,0.7,0.9111111,0.74444443,0.82222223,0.8,0.87777776,0.8888889,0.73333335,0.8888889,0.85555553,0.75555557,0.96666664,0.95555556,0.96666664,0.9,0.9222222,0.87777776,1.0,0.9111111,0.8888889,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.07777778,0.033333335,0.2,0.11111111,0.14444445,0.1,0.33333334,0.31111112,0.25555557,0.33333334,0.31111112,0.3888889,0.3,0.3888889,0.31111112,0.34444445,0.31111112,0.34444445,0.3888889,0.31111112,0.44444445,0.45555556,0.42222223,0.43333334,0.37777779,0.35555556,0.44444445,0.42222223,0.41111112,,0.5777778,0.5,0.54444444,0.6111111,0.6111111,0.53333336,0.6,0.65555555,0.64444447,0.62222224,0.7222222,0.6888889,0.56666666,0.8888889,0.6666667,0.6333333,0.7777778,0.6666667,0.65555555,0.7888889,0.67777777,0.74444443,0.9444444,0.8,0.7888889,0.7777778,0.8333333,0.85555553,0.75555557,0.85555553,0.84444445,0.75555557,0.98888886,0.98888886,0.9777778,0.98888886,0.98888886,0.8888889,1.0,0.9222222,0.96666664,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.011111111,0.033333335,0.08888889,0.16666667,0.15555556,0.18888889,0.21111111,0.21111111,0.13333334,0.22222222,0.2,0.26666668,0.2888889,0.24444444,0.2888889,0.3888889,0.26666668,0.35555556,0.32222223,0.34444445,0.32222223,0.43333334,0.42222223,0.37777779,0.37777779,0.36666667,0.51111114,0.3888889,0.35555556,0.42222223,,0.46666667,0.56666666,0.6,0.5888889,0.6,0.62222224,0.65555555,0.65555555,0.6,0.87777776,0.65555555,0.6333333,0.7111111,0.7,0.75555557,0.75555557,0.6666667,0.75555557,0.73333335,0.8,0.82222223,0.76666665,0.8333333,0.8666667,0.98888886,0.8333333,0.9222222,0.9111111,0.9222222,0.9111111,0.9111111,0.9777778,0.98888886,0.96666664,1.0,0.98888886,0.95555556,1.0,0.9,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.011111111,0.011111111,0.044444446,0.13333334,0.12222222,0.24444444,0.15555556,0.18888889,0.18888889,0.16666667,0.18888889,0.23333333,0.17777778,0.24444444,0.26666668,0.33333334,0.26666668,0.2777778,0.2777778,0.33333334,0.35555556,0.44444445,0.32222223,0.34444445,0.46666667,0.37777779,0.4888889,0.47777778,0.4888889,0.5,0.53333336,,0.6111111,0.45555556,0.5777778,0.74444443,0.6,0.5888889,0.7222222,0.6888889,0.7888889,0.6666667,0.9222222,0.7,0.7222222,0.7222222,0.67777777,0.6,0.76666665,0.7222222,0.8,0.87777776,0.67777777,0.75555557,0.84444445,0.8666667,0.8,0.9777778,0.9,0.96666664,0.9111111,0.9222222,0.87777776,0.87777776,0.8666667,0.87777776,0.8888889,0.9444444,1.0,0.82222223,0.9777778,1.0,0.9,1.0
+XGBoost (default),0.033333335,0.022222223,0.07777778,0.08888889,0.08888889,0.13333334,0.17777778,0.21111111,0.21111111,0.18888889,0.06666667,0.17777778,0.32222223,0.25555557,0.23333333,0.33333334,0.34444445,0.34444445,0.23333333,0.24444444,0.32222223,0.34444445,0.25555557,0.3888889,0.31111112,0.3888889,0.5,0.46666667,0.4,0.45555556,0.43333334,0.3888889,,0.6333333,0.54444444,0.4888889,0.53333336,0.67777777,0.54444444,0.5888889,0.6666667,0.6888889,0.64444447,0.65555555,0.75555557,0.8111111,0.53333336,0.6111111,0.7777778,0.7222222,0.8666667,0.8333333,0.65555555,0.7111111,0.9,0.9222222,0.76666665,0.9444444,0.87777776,0.95555556,0.8888889,0.8888889,0.87777776,0.9222222,0.93333334,0.98888886,0.9222222,0.9,1.0,0.9,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.022222223,0.11111111,0.044444446,0.21111111,0.23333333,0.3,0.07777778,0.055555556,0.11111111,0.06666667,0.17777778,0.23333333,0.1,0.15555556,0.34444445,0.34444445,0.2888889,0.13333334,0.25555557,0.34444445,0.24444444,0.3888889,0.2777778,0.2,0.32222223,0.34444445,0.2888889,0.44444445,0.37777779,0.3888889,0.4,0.54444444,0.36666667,,0.37777779,0.6111111,0.5555556,0.67777777,0.62222224,0.6333333,0.5777778,0.73333335,0.62222224,0.46666667,0.64444447,0.54444444,0.46666667,0.6888889,0.65555555,0.56666666,0.73333335,0.6888889,0.5888889,0.8,0.7222222,0.75555557,0.8666667,0.85555553,0.9777778,0.85555553,0.96666664,1.0,0.8111111,0.87777776,0.85555553,0.96666664,0.9777778,1.0,0.9,1.0,0.9777778,0.96666664,0.9,1.0
+TabSwift (default),0.055555556,0.044444446,0.15555556,0.033333335,0.13333334,0.08888889,0.22222222,0.18888889,0.23333333,0.23333333,0.31111112,0.25555557,0.33333334,0.34444445,0.14444445,0.24444444,0.3,0.36666667,0.33333334,0.16666667,0.45555556,0.34444445,0.36666667,0.4,0.13333334,0.34444445,0.42222223,0.35555556,0.2888889,0.3888889,0.41111112,0.42222223,0.45555556,0.62222224,,0.47777778,0.54444444,0.5777778,0.47777778,0.56666666,0.5555556,0.62222224,0.5,0.62222224,0.53333336,0.64444447,0.62222224,0.67777777,0.5555556,0.65555555,0.6333333,0.6333333,0.85555553,0.67777777,0.7,0.6888889,0.67777777,0.8,0.7222222,0.8,0.7888889,0.75555557,0.95555556,0.9111111,0.9777778,1.0,0.9222222,0.8111111,0.9222222,0.9,1.0,0.96666664,1.0,1.0
+EBM (tuned),0.0,0.0,0.044444446,0.08888889,0.033333335,0.1,0.17777778,0.18888889,0.17777778,0.17777778,0.15555556,0.13333334,0.14444445,0.17777778,0.17777778,0.15555556,0.24444444,0.22222222,0.18888889,0.26666668,0.2888889,0.23333333,0.26666668,0.2777778,0.36666667,0.36666667,0.42222223,0.41111112,0.42222223,0.46666667,0.4,0.25555557,0.51111114,0.3888889,0.5222222,,0.46666667,0.44444445,0.6,0.62222224,0.62222224,0.5555556,0.74444443,0.5888889,0.6333333,0.6333333,0.5888889,0.5888889,0.6666667,0.67777777,0.6888889,0.82222223,0.6,0.73333335,0.75555557,0.84444445,0.74444443,0.9444444,0.85555553,0.9444444,0.8888889,0.8888889,0.8666667,0.84444445,0.85555553,0.87777776,0.8333333,0.9222222,1.0,0.8111111,0.98888886,1.0,0.9,1.0
+TabM (default),0.0,0.0,0.07777778,0.011111111,0.07777778,0.055555556,0.23333333,0.21111111,0.16666667,0.2,0.18888889,0.2,0.17777778,0.1,0.25555557,0.033333335,0.17777778,0.21111111,0.22222222,0.32222223,0.23333333,0.15555556,0.24444444,0.32222223,0.41111112,0.3888889,0.37777779,0.37777779,0.43333334,0.4,0.37777779,0.4,0.46666667,0.44444445,0.45555556,0.53333336,,0.4888889,0.6111111,0.6,0.51111114,0.65555555,0.5777778,0.6111111,0.6666667,0.6111111,0.62222224,0.5777778,0.6666667,0.7222222,0.64444447,0.62222224,0.62222224,0.7777778,0.67777777,0.7777778,0.8111111,0.9,0.8,0.9,0.84444445,0.7888889,0.9777778,0.82222223,0.9777778,0.9777778,0.9444444,0.9,1.0,0.9,0.95555556,1.0,1.0,1.0
+xRFM (tuned),0.011111111,0.08888889,0.033333335,0.2,0.18888889,0.2777778,0.055555556,0.044444446,0.044444446,0.06666667,0.08888889,0.16666667,0.07777778,0.11111111,0.2888889,0.34444445,0.22222222,0.1,0.18888889,0.31111112,0.2,0.34444445,0.2777778,0.12222222,0.31111112,0.3,0.2888889,0.37777779,0.34444445,0.34444445,0.34444445,0.41111112,0.32222223,0.32222223,0.42222223,0.5555556,0.51111114,,0.6,0.6333333,0.54444444,0.6666667,0.5222222,0.51111114,0.62222224,0.47777778,0.45555556,0.62222224,0.65555555,0.56666666,0.65555555,0.8,0.56666666,0.82222223,0.74444443,0.73333335,0.8666667,0.8333333,0.95555556,0.84444445,0.9777778,0.98888886,0.85555553,0.8,0.8333333,0.8666667,0.9444444,1.0,0.98888886,0.98888886,0.95555556,1.0,0.9,1.0
+RealMLP (default),0.0,0.0,0.033333335,0.055555556,0.0,0.08888889,0.16666667,0.17777778,0.044444446,0.15555556,0.15555556,0.17777778,0.14444445,0.0,0.17777778,0.022222223,0.11111111,0.15555556,0.17777778,0.23333333,0.1,0.1,0.23333333,0.24444444,0.2888889,0.32222223,0.34444445,0.2777778,0.3,0.35555556,0.34444445,0.2777778,0.45555556,0.37777779,0.5222222,0.4,0.3888889,0.4,,0.45555556,0.5777778,0.5,0.43333334,0.56666666,0.6,0.6111111,0.5777778,0.6111111,0.6666667,0.6333333,0.6888889,0.62222224,0.6333333,0.82222223,0.7111111,0.7888889,0.84444445,0.8666667,0.8333333,0.8666667,0.8666667,0.87777776,0.95555556,0.85555553,0.9444444,0.8666667,0.95555556,0.9111111,1.0,0.87777776,0.98888886,1.0,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.08888889,0.055555556,0.1,0.07777778,0.15555556,0.14444445,0.13333334,0.14444445,0.12222222,0.13333334,0.12222222,0.14444445,0.2,0.08888889,0.17777778,0.14444445,0.14444445,0.24444444,0.22222222,0.2,0.21111111,0.24444444,0.21111111,0.31111112,0.34444445,0.31111112,0.3888889,0.37777779,0.4,0.31111112,0.41111112,0.36666667,0.43333334,0.37777779,0.4,0.36666667,0.54444444,,0.5,0.6,0.5222222,0.6111111,0.5555556,0.54444444,0.5888889,0.65555555,0.5888889,0.6,0.6111111,0.64444447,0.5777778,0.73333335,0.6333333,0.6666667,0.76666665,0.9444444,0.75555557,0.95555556,0.9111111,0.76666665,0.8666667,0.76666665,0.8666667,0.85555553,0.85555553,0.9444444,1.0,0.8111111,0.96666664,1.0,0.9,1.0
+TorchMLP (tuned),0.022222223,0.0,0.022222223,0.033333335,0.044444446,0.08888889,0.044444446,0.044444446,0.044444446,0.055555556,0.15555556,0.08888889,0.13333334,0.16666667,0.17777778,0.22222222,0.15555556,0.17777778,0.17777778,0.18888889,0.18888889,0.25555557,0.3,0.25555557,0.2,0.32222223,0.3888889,0.2777778,0.26666668,0.2777778,0.12222222,0.21111111,0.33333334,0.42222223,0.44444445,0.37777779,0.4888889,0.45555556,0.42222223,0.5,,0.51111114,0.4,0.4888889,0.54444444,0.64444447,0.5777778,0.5222222,0.62222224,0.65555555,0.6111111,0.65555555,0.62222224,0.64444447,0.73333335,0.8888889,0.6888889,0.84444445,0.87777776,0.84444445,0.84444445,0.87777776,0.95555556,0.9222222,0.95555556,0.9777778,0.9,0.9222222,0.9777778,0.8888889,0.98888886,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.08888889,0.033333335,0.1,0.13333334,0.15555556,0.13333334,0.16666667,0.033333335,0.17777778,0.2,0.23333333,0.14444445,0.022222223,0.26666668,0.17777778,0.13333334,0.18888889,0.24444444,0.3,0.12222222,0.2,0.24444444,0.18888889,0.33333334,0.31111112,0.26666668,0.33333334,0.35555556,0.31111112,0.34444445,0.33333334,0.31111112,0.26666668,0.37777779,0.44444445,0.34444445,0.33333334,0.5,0.4,0.4888889,,0.47777778,0.5222222,0.5777778,0.44444445,0.5222222,0.5555556,0.6,0.62222224,0.5888889,0.6111111,0.62222224,0.84444445,0.62222224,0.67777777,0.96666664,0.67777777,0.8888889,0.67777777,0.87777776,0.9,0.85555553,0.8333333,0.8666667,0.73333335,0.93333334,0.9777778,0.8666667,1.0,0.93333334,0.9,1.0,1.0
+EBM (default),0.0,0.0,0.044444446,0.044444446,0.033333335,0.06666667,0.055555556,0.11111111,0.08888889,0.033333335,0.022222223,0.033333335,0.06666667,0.15555556,0.13333334,0.16666667,0.17777778,0.14444445,0.07777778,0.16666667,0.26666668,0.21111111,0.14444445,0.23333333,0.2777778,0.32222223,0.4,0.35555556,0.2888889,0.43333334,0.36666667,0.07777778,0.35555556,0.37777779,0.5,0.25555557,0.42222223,0.47777778,0.56666666,0.47777778,0.6,0.5222222,,0.5777778,0.5555556,0.64444447,0.5555556,0.5888889,0.6111111,0.64444447,0.67777777,0.67777777,0.6,0.73333335,0.75555557,0.8333333,0.73333335,0.84444445,0.8111111,0.84444445,0.87777776,0.84444445,0.8666667,0.82222223,0.85555553,0.8666667,0.82222223,0.9111111,1.0,0.8,0.9777778,1.0,0.9,1.0
+TabDPT (tuned),0.044444446,0.033333335,0.033333335,0.033333335,0.044444446,0.055555556,0.18888889,0.18888889,0.08888889,0.2,0.15555556,0.2,0.16666667,0.18888889,0.11111111,0.12222222,0.12222222,0.18888889,0.2,0.17777778,0.26666668,0.23333333,0.25555557,0.16666667,0.22222222,0.14444445,0.21111111,0.22222222,0.16666667,0.11111111,0.2888889,0.3,0.34444445,0.53333336,0.37777779,0.41111112,0.3888889,0.4888889,0.43333334,0.3888889,0.51111114,0.47777778,0.42222223,,0.4888889,0.44444445,0.56666666,0.62222224,0.5888889,0.46666667,0.5777778,0.5222222,0.75,0.67777777,0.64444447,0.67777777,0.7777778,0.6666667,0.6888889,0.67777777,0.8,0.74444443,0.96666664,0.9,0.96666664,0.9111111,0.93333334,0.8,0.9111111,0.9,0.9,0.9777778,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.022222223,0.011111111,0.055555556,0.044444446,0.044444446,0.11111111,0.12222222,0.12222222,0.11111111,0.1,0.1,0.11111111,0.12222222,0.08888889,0.16666667,0.21111111,0.13333334,0.13333334,0.14444445,0.15555556,0.22222222,0.21111111,0.15555556,0.25555557,0.2888889,0.26666668,0.2777778,0.32222223,0.33333334,0.3,0.2777778,0.24444444,0.35555556,0.46666667,0.36666667,0.33333334,0.37777779,0.4,0.44444445,0.45555556,0.42222223,0.44444445,0.51111114,,0.51111114,0.44444445,0.53333336,0.74444443,0.47777778,0.6333333,0.6111111,0.5555556,0.6333333,0.6888889,0.7,0.7,0.8,0.8,0.8111111,0.8888889,0.85555553,0.85555553,0.96666664,0.8333333,0.93333334,0.96666664,0.95555556,0.93333334,0.8888889,0.96666664,1.0,0.9,1.0
+iLTM (default),0.0,0.0,0.033333335,0.011111111,0.055555556,0.07777778,0.07777778,0.055555556,0.0,0.06666667,0.033333335,0.022222223,0.044444446,0.18888889,0.14444445,0.16666667,0.15555556,0.1,0.06666667,0.17777778,0.26666668,0.24444444,0.06666667,0.16666667,0.16666667,0.23333333,0.23333333,0.23333333,0.31111112,0.36666667,0.24444444,0.2777778,0.18888889,0.45555556,0.35555556,0.36666667,0.3888889,0.5222222,0.3888889,0.45555556,0.35555556,0.5555556,0.35555556,0.5555556,0.4888889,,0.4888889,0.56666666,0.5,0.56666666,0.4888889,0.67777777,0.5777778,0.7222222,0.64444447,0.64444447,0.76666665,0.8888889,0.75555557,0.9,0.87777776,0.8,0.75555557,0.85555553,0.74444443,0.9222222,0.84444445,0.8888889,0.98888886,0.8888889,0.96666664,0.9,0.8,1.0
+LightGBM (default),0.0,0.0,0.044444446,0.011111111,0.055555556,0.055555556,0.044444446,0.06666667,0.07777778,0.07777778,0.033333335,0.022222223,0.12222222,0.06666667,0.07777778,0.0,0.08888889,0.18888889,0.06666667,0.11111111,0.17777778,0.13333334,0.033333335,0.24444444,0.12222222,0.13333334,0.2,0.23333333,0.21111111,0.22222222,0.24444444,0.32222223,0.46666667,0.53333336,0.37777779,0.41111112,0.37777779,0.54444444,0.42222223,0.41111112,0.42222223,0.47777778,0.44444445,0.43333334,0.5555556,0.51111114,,0.6,0.5777778,0.5,0.5555556,0.4888889,0.56666666,0.6,0.64444447,0.74444443,0.7,0.84444445,0.6888889,0.85555553,0.76666665,0.7,0.87777776,0.8888889,0.95555556,0.9444444,0.9444444,0.8111111,1.0,0.9,0.9222222,1.0,1.0,1.0
+LimiX (default),0.11111111,0.14444445,0.12222222,0.2,0.22222222,0.18888889,0.2777778,0.24444444,0.24444444,0.2777778,0.32222223,0.31111112,0.25555557,0.15555556,0.33333334,0.23333333,0.14444445,0.26666668,0.34444445,0.34444445,0.2,0.23333333,0.35555556,0.26666668,0.33333334,0.31111112,0.23333333,0.2777778,0.23333333,0.33333334,0.33333334,0.4,0.3888889,0.31111112,0.32222223,0.41111112,0.42222223,0.37777779,0.3888889,0.34444445,0.47777778,0.44444445,0.41111112,0.37777779,0.46666667,0.43333334,0.4,,0.47777778,0.46666667,0.5222222,0.41111112,0.54444444,0.5777778,0.5555556,0.4888889,0.6111111,0.5222222,0.64444447,0.53333336,0.56666666,0.65555555,0.6333333,0.75555557,0.67777777,0.6333333,0.7111111,0.65555555,0.62222224,0.7777778,0.65555555,0.67777777,0.7888889,0.8888889
+RandomForest (tuned),0.011111111,0.022222223,0.022222223,0.033333335,0.011111111,0.033333335,0.12222222,0.12222222,0.11111111,0.12222222,0.1,0.08888889,0.11111111,0.13333334,0.044444446,0.16666667,0.22222222,0.13333334,0.14444445,0.06666667,0.17777778,0.21111111,0.15555556,0.18888889,0.2,0.2777778,0.24444444,0.2888889,0.24444444,0.34444445,0.24444444,0.23333333,0.22222222,0.34444445,0.44444445,0.33333334,0.33333334,0.34444445,0.33333334,0.41111112,0.37777779,0.4,0.3888889,0.41111112,0.25555557,0.5,0.42222223,0.5222222,,0.41111112,0.54444444,0.5777778,0.51111114,0.56666666,0.6333333,0.5777778,0.6666667,0.7888889,0.7111111,0.8,0.85555553,0.7777778,0.8333333,0.9388889,0.8333333,0.9222222,0.9222222,0.9111111,0.93333334,0.85555553,0.9777778,1.0,0.9,1.0
+ModernNCA (default),0.0,0.0,0.011111111,0.022222223,0.022222223,0.044444446,0.044444446,0.011111111,0.022222223,0.06666667,0.011111111,0.022222223,0.08888889,0.14444445,0.17777778,0.1,0.044444446,0.14444445,0.033333335,0.21111111,0.24444444,0.16666667,0.06666667,0.18888889,0.2,0.13333334,0.15555556,0.1,0.3,0.21111111,0.26666668,0.2777778,0.2777778,0.43333334,0.34444445,0.32222223,0.2777778,0.43333334,0.36666667,0.4,0.34444445,0.37777779,0.35555556,0.53333336,0.5222222,0.43333334,0.5,0.53333336,0.5888889,,0.46666667,0.5777778,0.56666666,0.54444444,0.5555556,0.75555557,0.67777777,0.74444443,0.7,0.75555557,0.8333333,0.74444443,0.84444445,0.95555556,0.8666667,0.9111111,0.93333334,0.8666667,1.0,0.9,0.9222222,1.0,0.9222222,1.0
+ExtraTrees (tuned + ensembled),0.0,0.011111111,0.011111111,0.033333335,0.022222223,0.011111111,0.011111111,0.033333335,0.044444446,0.044444446,0.011111111,0.022222223,0.11111111,0.1,0.033333335,0.16666667,0.1,0.13333334,0.044444446,0.08888889,0.2,0.23333333,0.16666667,0.23333333,0.07777778,0.2,0.25555557,0.15555556,0.23333333,0.32222223,0.2,0.2,0.13333334,0.26666668,0.36666667,0.31111112,0.35555556,0.34444445,0.31111112,0.3888889,0.3888889,0.41111112,0.32222223,0.42222223,0.36666667,0.51111114,0.44444445,0.47777778,0.45555556,0.53333336,,0.4888889,0.53333336,0.53333336,0.75555557,0.64444447,0.64444447,0.7222222,0.7222222,0.7222222,0.74444443,0.7888889,0.75555557,0.98888886,0.8,0.9444444,0.9777778,0.9,0.85555553,0.9111111,0.96666664,0.9,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.022222223,0.033333335,0.07777778,0.044444446,0.07777778,0.022222223,0.044444446,0.022222223,0.033333335,0.044444446,0.044444446,0.022222223,0.11111111,0.14444445,0.12222222,0.14444445,0.06666667,0.055555556,0.16666667,0.21111111,0.2,0.17777778,0.044444446,0.1,0.22222222,0.15555556,0.21111111,0.3,0.25555557,0.17777778,0.12222222,0.16666667,0.31111112,0.36666667,0.17777778,0.37777779,0.2,0.37777779,0.35555556,0.34444445,0.3888889,0.32222223,0.47777778,0.3888889,0.32222223,0.51111114,0.5888889,0.42222223,0.42222223,0.51111114,,0.5777778,0.6111111,0.67777777,0.64444447,0.7111111,0.6666667,0.8111111,0.6888889,0.8888889,0.84444445,0.84444445,0.82222223,0.8666667,0.8666667,0.85555553,0.9777778,0.95555556,0.8333333,0.96666664,1.0,0.9,1.0
+TabDPT (default),0.022222223,0.022222223,0.044444446,0.033333335,0.055555556,0.044444446,0.11111111,0.1,0.06666667,0.12222222,0.11111111,0.044444446,0.16666667,0.17777778,0.1,0.12222222,0.07777778,0.22222222,0.14444445,0.15555556,0.25555557,0.2,0.16666667,0.17777778,0.14444445,0.07777778,0.14444445,0.13333334,0.08888889,0.055555556,0.23333333,0.32222223,0.34444445,0.41111112,0.14444445,0.4,0.37777779,0.43333334,0.36666667,0.42222223,0.37777779,0.37777779,0.4,0.25,0.44444445,0.42222223,0.43333334,0.45555556,0.4888889,0.43333334,0.46666667,0.42222223,,0.51111114,0.5,0.62222224,0.6333333,0.6666667,0.7,0.6666667,0.6111111,0.7111111,0.7888889,0.67777777,0.8333333,0.7222222,0.67777777,0.74444443,0.7111111,0.75555557,0.85555553,0.73333335,0.9777778,1.0
+FastaiMLP (tuned),0.0,0.022222223,0.022222223,0.06666667,0.055555556,0.06666667,0.033333335,0.022222223,0.033333335,0.044444446,0.12222222,0.08888889,0.11111111,0.022222223,0.12222222,0.08888889,0.06666667,0.14444445,0.13333334,0.12222222,0.033333335,0.12222222,0.21111111,0.13333334,0.2,0.2888889,0.14444445,0.07777778,0.25555557,0.2,0.16666667,0.24444444,0.2888889,0.2,0.32222223,0.26666668,0.22222222,0.17777778,0.17777778,0.26666668,0.35555556,0.15555556,0.26666668,0.32222223,0.36666667,0.2777778,0.4,0.42222223,0.43333334,0.45555556,0.46666667,0.3888889,0.4888889,,0.51111114,0.4888889,0.7111111,0.5555556,0.6666667,0.5555556,0.7111111,0.73333335,0.8,0.76666665,0.8333333,0.67777777,0.82222223,0.9,0.87777776,0.8888889,0.8666667,0.8888889,0.95555556,1.0
+ExtraTrees (tuned),0.0,0.011111111,0.0,0.033333335,0.011111111,0.011111111,0.022222223,0.011111111,0.011111111,0.033333335,0.011111111,0.022222223,0.11111111,0.12222222,0.033333335,0.13333334,0.11111111,0.13333334,0.022222223,0.06666667,0.14444445,0.16666667,0.07777778,0.15555556,0.055555556,0.16666667,0.13333334,0.14444445,0.17777778,0.21111111,0.13333334,0.15555556,0.1,0.2777778,0.3,0.24444444,0.32222223,0.25555557,0.2888889,0.36666667,0.26666668,0.37777779,0.24444444,0.35555556,0.31111112,0.35555556,0.35555556,0.44444445,0.36666667,0.44444445,0.24444444,0.32222223,0.5,0.4888889,,0.5555556,0.54444444,0.7111111,0.7111111,0.7,0.6888889,0.74444443,0.7111111,0.95555556,0.7777778,0.93333334,0.9444444,0.8888889,0.84444445,0.9,0.96666664,0.9111111,0.98888886,1.0
+TorchMLP (default),0.0,0.0,0.011111111,0.033333335,0.0,0.022222223,0.011111111,0.022222223,0.0,0.022222223,0.0,0.022222223,0.08888889,0.011111111,0.044444446,0.022222223,0.11111111,0.08888889,0.0,0.08888889,0.06666667,0.022222223,0.06666667,0.13333334,0.044444446,0.2,0.11111111,0.14444445,0.2,0.22222222,0.011111111,0.13333334,0.07777778,0.24444444,0.31111112,0.15555556,0.22222222,0.26666668,0.21111111,0.33333334,0.11111111,0.32222223,0.16666667,0.32222223,0.3,0.35555556,0.25555557,0.51111114,0.42222223,0.24444444,0.35555556,0.35555556,0.37777779,0.51111114,0.44444445,,0.5555556,0.73333335,0.73333335,0.73333335,0.8,0.76666665,0.8,0.7888889,0.82222223,0.9222222,0.8666667,0.87777776,0.9,0.8888889,0.95555556,0.98888886,0.95555556,1.0
+FastaiMLP (default),0.0,0.022222223,0.011111111,0.06666667,0.07777778,0.07777778,0.022222223,0.044444446,0.0,0.055555556,0.1,0.13333334,0.12222222,0.033333335,0.1,0.1,0.033333335,0.13333334,0.14444445,0.11111111,0.044444446,0.08888889,0.18888889,0.13333334,0.15555556,0.21111111,0.1,0.08888889,0.12222222,0.16666667,0.16666667,0.2,0.23333333,0.13333334,0.32222223,0.25555557,0.18888889,0.13333334,0.15555556,0.23333333,0.31111112,0.033333335,0.26666668,0.22222222,0.3,0.23333333,0.3,0.3888889,0.33333334,0.32222223,0.35555556,0.2888889,0.36666667,0.2888889,0.45555556,0.44444445,,0.47777778,0.5555556,0.4888889,0.6,0.6333333,0.7,0.6111111,0.7222222,0.6333333,0.62222224,0.67777777,0.7,0.8111111,0.8,0.8111111,0.8888889,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.011111111,0.011111111,0.011111111,0.0,0.0,0.0,0.0,0.07777778,0.011111111,0.011111111,0.011111111,0.022222223,0.0,0.0,0.011111111,0.0,0.06666667,0.055555556,0.044444446,0.11111111,0.14444445,0.07777778,0.022222223,0.055555556,0.14444445,0.2,0.055555556,0.1,0.16666667,0.13333334,0.055555556,0.15555556,0.32222223,0.15555556,0.33333334,0.2,0.11111111,0.15555556,0.47777778,0.21111111,0.25555557,0.2777778,0.33333334,0.33333334,0.44444445,0.2888889,0.26666668,0.5222222,,0.56666666,0.65555555,0.65555555,0.5777778,0.73333335,0.6,0.73333335,0.64444447,0.7,0.74444443,1.0,0.7,0.87777776,0.87777776,0.8,0.9
+Linear (tuned + ensembled),0.0,0.022222223,0.0,0.07777778,0.033333335,0.07777778,0.011111111,0.011111111,0.011111111,0.011111111,0.0,0.022222223,0.0,0.0,0.15555556,0.06666667,0.1,0.0,0.0,0.17777778,0.022222223,0.11111111,0.13333334,0.0,0.15555556,0.24444444,0.1,0.13333334,0.26666668,0.24444444,0.08888889,0.1,0.12222222,0.022222223,0.2777778,0.14444445,0.2,0.044444446,0.16666667,0.24444444,0.12222222,0.11111111,0.18888889,0.31111112,0.2,0.24444444,0.31111112,0.35555556,0.2888889,0.3,0.2777778,0.18888889,0.3,0.33333334,0.2888889,0.26666668,0.44444445,0.43333334,,0.43333334,0.42222223,0.81666666,0.44444445,0.5555556,0.47777778,0.5888889,0.54444444,0.93333334,0.5,0.7777778,0.6888889,0.62222224,0.76666665,0.9
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.011111111,0.011111111,0.011111111,0.0,0.0,0.0,0.0,0.07777778,0.011111111,0.011111111,0.0,0.022222223,0.0,0.0,0.0,0.0,0.055555556,0.055555556,0.044444446,0.11111111,0.14444445,0.07777778,0.033333335,0.044444446,0.14444445,0.2,0.055555556,0.1,0.15555556,0.13333334,0.044444446,0.15555556,0.32222223,0.15555556,0.32222223,0.18888889,0.1,0.14444445,0.46666667,0.2,0.24444444,0.2777778,0.31111112,0.33333334,0.44444445,0.3,0.26666668,0.51111114,0.34444445,0.56666666,,0.65555555,0.5777778,0.73333335,0.6,0.73333335,0.64444447,0.7,0.74444443,1.0,0.7,0.87777776,0.87777776,0.8,0.9
+KNN (tuned + ensembled),0.011111111,0.044444446,0.011111111,0.06666667,0.055555556,0.044444446,0.011111111,0.011111111,0.011111111,0.011111111,0.011111111,0.022222223,0.011111111,0.011111111,0.08888889,0.055555556,0.011111111,0.022222223,0.011111111,0.11111111,0.033333335,0.11111111,0.08888889,0.022222223,0.1,0.14444445,0.011111111,0.044444446,0.14444445,0.15555556,0.08888889,0.08888889,0.11111111,0.033333335,0.21111111,0.11111111,0.15555556,0.022222223,0.13333334,0.08888889,0.15555556,0.12222222,0.12222222,0.2,0.11111111,0.12222222,0.23333333,0.43333334,0.14444445,0.16666667,0.25555557,0.11111111,0.3888889,0.2888889,0.31111112,0.2,0.4,0.34444445,0.5777778,0.34444445,,0.6333333,0.53333336,0.6888889,0.5777778,0.5777778,0.56666666,0.7222222,0.6,0.7111111,0.6666667,1.0,0.84444445,1.0
+Linear (tuned),0.0,0.022222223,0.0,0.06666667,0.033333335,0.055555556,0.0,0.0,0.0,0.0,0.0,0.011111111,0.0,0.0,0.14444445,0.055555556,0.08888889,0.0,0.0,0.13333334,0.011111111,0.1,0.08888889,0.0,0.15555556,0.24444444,0.06666667,0.1,0.24444444,0.24444444,0.08888889,0.07777778,0.11111111,0.0,0.24444444,0.11111111,0.21111111,0.011111111,0.12222222,0.23333333,0.12222222,0.1,0.15555556,0.25555557,0.14444445,0.2,0.3,0.34444445,0.22222222,0.25555557,0.21111111,0.15555556,0.2888889,0.26666668,0.25555557,0.23333333,0.36666667,0.42222223,0.18333334,0.42222223,0.36666667,,0.43333334,0.51111114,0.45555556,0.56666666,0.51111114,0.8666667,0.5,0.7222222,0.65555555,0.62222224,0.74444443,0.9
+PerpetualBooster (tuned + ensembled),0.011111111,0.0,0.022222223,0.0,0.011111111,0.0,0.022222223,0.011111111,0.022222223,0.022222223,0.07777778,0.022222223,0.12222222,0.022222223,0.033333335,0.022222223,0.011111111,0.12222222,0.08888889,0.033333335,0.033333335,0.022222223,0.055555556,0.12222222,0.022222223,0.022222223,0.06666667,0.055555556,0.033333335,0.011111111,0.022222223,0.12222222,0.12222222,0.18888889,0.044444446,0.13333334,0.022222223,0.14444445,0.044444446,0.13333334,0.044444446,0.14444445,0.13333334,0.033333335,0.14444445,0.24444444,0.12222222,0.36666667,0.16666667,0.15555556,0.24444444,0.15555556,0.21111111,0.2,0.2888889,0.2,0.3,0.26666668,0.5555556,0.26666668,0.46666667,0.56666666,,0.53333336,0.85555553,0.6,0.6,0.56666666,0.36666667,0.65555555,0.6111111,0.65555555,0.9222222,0.95555556
+RandomForest (default),0.0,0.011111111,0.0,0.0,0.0,0.0,0.011111111,0.011111111,0.011111111,0.011111111,0.0,0.0,0.0,0.06666667,0.0,0.011111111,0.0,0.011111111,0.0,0.011111111,0.08888889,0.033333335,0.011111111,0.0,0.022222223,0.0,0.033333335,0.022222223,0.044444446,0.011111111,0.011111111,0.12222222,0.07777778,0.12222222,0.08888889,0.15555556,0.17777778,0.2,0.14444445,0.23333333,0.07777778,0.16666667,0.17777778,0.1,0.033333335,0.14444445,0.11111111,0.24444444,0.06111111,0.044444446,0.011111111,0.17777778,0.32222223,0.23333333,0.044444446,0.21111111,0.3888889,0.4,0.44444445,0.4,0.31111112,0.4888889,0.46666667,,0.46666667,0.4888889,0.6,0.51111114,0.51111114,0.8111111,0.5888889,0.65555555,0.7888889,1.0
+PerpetualBooster (tuned),0.0,0.0,0.033333335,0.0,0.011111111,0.0,0.011111111,0.011111111,0.033333335,0.011111111,0.033333335,0.022222223,0.11111111,0.011111111,0.022222223,0.0,0.011111111,0.12222222,0.033333335,0.055555556,0.044444446,0.011111111,0.033333335,0.13333334,0.011111111,0.0,0.07777778,0.044444446,0.033333335,0.022222223,0.033333335,0.13333334,0.06666667,0.14444445,0.022222223,0.14444445,0.022222223,0.16666667,0.055555556,0.13333334,0.044444446,0.13333334,0.14444445,0.033333335,0.16666667,0.25555557,0.044444446,0.32222223,0.16666667,0.13333334,0.2,0.13333334,0.16666667,0.16666667,0.22222222,0.17777778,0.2777778,0.26666668,0.5222222,0.26666668,0.42222223,0.54444444,0.14444445,0.53333336,,0.6,0.54444444,0.53333336,0.31111112,0.64444447,0.54444444,0.5777778,0.9,0.95555556
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.011111111,0.011111111,0.011111111,0.022222223,0.0,0.011111111,0.0,0.0,0.011111111,0.0,0.044444446,0.011111111,0.0,0.011111111,0.022222223,0.0,0.011111111,0.0,0.0,0.0,0.022222223,0.011111111,0.1,0.011111111,0.0,0.12222222,0.011111111,0.033333335,0.0,0.12222222,0.022222223,0.13333334,0.13333334,0.14444445,0.022222223,0.26666668,0.13333334,0.08888889,0.06666667,0.07777778,0.055555556,0.36666667,0.07777778,0.08888889,0.055555556,0.13333334,0.2777778,0.32222223,0.06666667,0.07777778,0.36666667,0.35555556,0.41111112,0.35555556,0.42222223,0.43333334,0.4,0.51111114,0.4,,0.6111111,0.5,0.54444444,0.6333333,0.6333333,0.4888889,0.64444447,0.9444444
+ExtraTrees (default),0.0,0.011111111,0.011111111,0.0,0.0,0.0,0.011111111,0.022222223,0.011111111,0.022222223,0.011111111,0.011111111,0.011111111,0.0,0.011111111,0.0,0.022222223,0.011111111,0.011111111,0.022222223,0.011111111,0.022222223,0.011111111,0.011111111,0.011111111,0.0,0.033333335,0.033333335,0.07777778,0.011111111,0.011111111,0.11111111,0.07777778,0.022222223,0.07777778,0.16666667,0.055555556,0.055555556,0.044444446,0.14444445,0.1,0.06666667,0.17777778,0.06666667,0.033333335,0.15555556,0.055555556,0.2888889,0.07777778,0.06666667,0.022222223,0.14444445,0.32222223,0.17777778,0.055555556,0.13333334,0.37777779,0.3,0.45555556,0.3,0.43333334,0.4888889,0.4,0.4,0.45555556,0.3888889,,0.5222222,0.45555556,0.7111111,0.5888889,0.65555555,0.7888889,1.0
+Linear (default),0.0,0.0,0.0,0.044444446,0.033333335,0.033333335,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.022222223,0.022222223,0.0,0.0,0.06666667,0.0,0.06666667,0.033333335,0.0,0.07777778,0.06666667,0.0,0.011111111,0.12222222,0.11111111,0.044444446,0.055555556,0.1,0.0,0.18888889,0.07777778,0.1,0.0,0.08888889,0.055555556,0.07777778,0.022222223,0.08888889,0.2,0.044444446,0.11111111,0.18888889,0.34444445,0.08888889,0.13333334,0.1,0.022222223,0.25555557,0.1,0.11111111,0.12222222,0.32222223,0.25555557,0.06666667,0.25555557,0.2777778,0.13333334,0.43333334,0.4888889,0.46666667,0.5,0.47777778,,0.47777778,0.7111111,0.5222222,0.62222224,0.74444443,0.8888889
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.011111111,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.07777778,0.0,0.0,0.011111111,0.0,0.0,0.022222223,0.13333334,0.0,0.08888889,0.06666667,0.011111111,0.0,0.37777779,0.06666667,0.0,0.14444445,0.044444446,0.2888889,0.12222222,0.15555556,0.1,0.3,0.0,0.5,0.0,0.4,0.5,0.6333333,0.4888889,0.6888889,0.45555556,0.54444444,0.5222222,,0.6333333,0.67777777,0.7,0.8,0.87777776
+xRFM (default),0.0,0.044444446,0.0,0.07777778,0.07777778,0.07777778,0.0,0.0,0.0,0.0,0.011111111,0.044444446,0.0,0.0,0.07777778,0.08888889,0.0,0.0,0.044444446,0.1,0.022222223,0.1,0.08888889,0.0,0.1,0.06666667,0.011111111,0.044444446,0.08888889,0.07777778,0.1,0.17777778,0.1,0.0,0.1,0.18888889,0.1,0.011111111,0.12222222,0.18888889,0.11111111,0.0,0.2,0.1,0.11111111,0.11111111,0.1,0.22222222,0.14444445,0.1,0.08888889,0.16666667,0.24444444,0.11111111,0.1,0.11111111,0.18888889,0.3,0.22222222,0.3,0.2888889,0.2777778,0.34444445,0.18888889,0.35555556,0.36666667,0.2888889,0.2888889,0.36666667,,0.43333334,0.4888889,0.53333336,0.8666667
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.011111111,0.011111111,0.0,0.011111111,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.011111111,0.022222223,0.0,0.011111111,0.0,0.0,0.022222223,0.011111111,0.022222223,0.11111111,0.033333335,0.0,0.022222223,0.0,0.022222223,0.0,0.011111111,0.044444446,0.044444446,0.011111111,0.033333335,0.011111111,0.06666667,0.022222223,0.1,0.033333335,0.033333335,0.07777778,0.34444445,0.022222223,0.07777778,0.033333335,0.033333335,0.14444445,0.13333334,0.033333335,0.044444446,0.2,0.12222222,0.31111112,0.12222222,0.33333334,0.34444445,0.3888889,0.41111112,0.45555556,0.36666667,0.41111112,0.47777778,0.32222223,0.56666666,,0.5555556,0.64444447,0.8
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.011111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.033333335,0.033333335,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.022222223,0.0,0.1,0.0,0.32222223,0.0,0.0,0.1,0.0,0.26666668,0.11111111,0.08888889,0.011111111,0.18888889,0.12222222,0.37777779,0.12222222,0.0,0.37777779,0.34444445,0.34444445,0.42222223,0.51111114,0.34444445,0.37777779,0.3,0.51111114,0.44444445,,0.7111111,0.98888886
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07777778,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.1,0.0,0.1,0.0,0.1,0.0,0.0,0.1,0.0,0.1,0.2,0.0,0.21111111,0.1,0.07777778,0.0,0.1,0.022222223,0.044444446,0.011111111,0.044444446,0.11111111,0.2,0.23333333,0.2,0.15555556,0.25555557,0.07777778,0.21111111,0.1,0.35555556,0.21111111,0.25555557,0.2,0.46666667,0.35555556,0.2888889,,0.87777776
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.0,0.1,0.044444446,0.0,0.044444446,0.055555556,0.0,0.11111111,0.12222222,0.13333334,0.2,0.011111111,0.12222222,
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_small/leaderboard_overview_explorer.html
index bfe8ff7cabb2a369b618327e40b09441d2281830..81da7ac05917f67fbbe1dd6c6eeff853dfc02714 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Classification | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_classification/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..f76305124ee070d7c77b9c768543aa9fec3bcf6e
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_classification/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1713.0,"score":0.876,"rank":8.51,"hrank":1.76,"imp":5.499,"train_time":38.57,"predict_time":6.317,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+162\/-136","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1612.0,"score":0.637,"rank":12.93,"hrank":5.27,"imp":13.225,"train_time":7.17,"predict_time":0.407,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+99\/-75","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1611.0,"score":0.601,"rank":12.96,"hrank":5.47,"imp":10.997,"train_time":363.16,"predict_time":6.598,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-78","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1581.0,"score":0.56,"rank":14.62,"hrank":6.57,"imp":13.595,"train_time":2289.39,"predict_time":8.114,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-76","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1573.0,"score":0.571,"rank":15.03,"hrank":7.84,"imp":14.402,"train_time":6.26,"predict_time":0.539,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-65","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1565.0,"score":0.553,"rank":15.53,"hrank":8.64,"imp":14.251,"train_time":6.08,"predict_time":0.762,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-69","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1563.0,"score":0.544,"rank":15.64,"hrank":7.02,"imp":14.245,"train_time":2289.39,"predict_time":1.253,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-65","imputed":false},{"position":7,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1561.0,"score":0.609,"rank":15.76,"hrank":5.9,"imp":13.749,"train_time":3.22,"predict_time":0.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-80","imputed":false},{"position":8,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1469.0,"score":0.43,"rank":21.93,"hrank":10.56,"imp":17.682,"train_time":144.31,"predict_time":3.769,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+84\/-72","imputed":true},{"position":9,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1469.0,"score":0.413,"rank":21.94,"hrank":9.95,"imp":17.489,"train_time":3303.01,"predict_time":45.78,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+82\/-74","imputed":true},{"position":10,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1456.0,"score":0.393,"rank":22.93,"hrank":9.99,"imp":16.364,"train_time":27.77,"predict_time":5.044,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-78","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1445.0,"score":0.341,"rank":23.82,"hrank":13.82,"imp":17.367,"train_time":3423.61,"predict_time":18.933,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-51","imputed":false},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1425.0,"score":0.346,"rank":25.41,"hrank":10.25,"imp":18.114,"train_time":4.35,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-61","imputed":false},{"position":13,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1421.0,"score":0.306,"rank":25.67,"hrank":12.64,"imp":18.391,"train_time":2469.48,"predict_time":3.418,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-53","imputed":false},{"position":14,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1417.0,"score":0.283,"rank":26.06,"hrank":15.41,"imp":18.192,"train_time":3144.62,"predict_time":1.882,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-53","imputed":false},{"position":15,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1403.0,"score":0.311,"rank":27.2,"hrank":14.4,"imp":19.041,"train_time":8.89,"predict_time":1.714,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+72\/-54","imputed":true},{"position":16,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1398.0,"score":0.301,"rank":27.63,"hrank":12.7,"imp":18.4,"train_time":5271.47,"predict_time":231.832,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-60","imputed":false},{"position":17,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1395.0,"score":0.321,"rank":27.82,"hrank":12.19,"imp":19.713,"train_time":3303.01,"predict_time":0.929,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+97\/-75","imputed":true},{"position":18,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1385.0,"score":0.297,"rank":28.64,"hrank":16.22,"imp":19.706,"train_time":331.0,"predict_time":2.359,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-48","imputed":false},{"position":19,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1374.0,"score":0.249,"rank":29.6,"hrank":14.43,"imp":19.239,"train_time":2469.48,"predict_time":0.408,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-60","imputed":false},{"position":20,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1371.0,"score":0.205,"rank":29.89,"hrank":21.87,"imp":19.438,"train_time":735.37,"predict_time":1.9,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-36","imputed":false},{"position":21,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1367.0,"score":0.244,"rank":30.2,"hrank":15.6,"imp":18.869,"train_time":3423.61,"predict_time":1.533,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-54","imputed":false},{"position":22,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1367.0,"score":0.266,"rank":30.24,"hrank":13.35,"imp":19.086,"train_time":5271.47,"predict_time":33.978,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-60","imputed":false},{"position":23,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1360.0,"score":0.299,"rank":30.8,"hrank":14.08,"imp":20.41,"train_time":4.15,"predict_time":0.42,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+78\/-68","imputed":true},{"position":24,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1346.0,"score":0.164,"rank":32.08,"hrank":24.23,"imp":19.803,"train_time":735.37,"predict_time":0.311,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-34","imputed":false},{"position":25,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1341.0,"score":0.259,"rank":32.51,"hrank":19.6,"imp":20.532,"train_time":2.24,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-59","imputed":false},{"position":26,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1339.0,"score":0.176,"rank":32.62,"hrank":23.4,"imp":19.813,"train_time":1542.95,"predict_time":0.396,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-35","imputed":false},{"position":27,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1334.0,"score":0.156,"rank":33.13,"hrank":22.69,"imp":19.926,"train_time":21290.02,"predict_time":663.263,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-48","imputed":false},{"position":28,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1329.0,"score":0.159,"rank":33.52,"hrank":24.24,"imp":20.07,"train_time":1542.95,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-31","imputed":false},{"position":29,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1325.0,"score":0.176,"rank":33.92,"hrank":16.48,"imp":20.17,"train_time":6049.28,"predict_time":0.525,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-50","imputed":false},{"position":30,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1312.0,"score":0.254,"rank":35.03,"hrank":11.41,"imp":21.466,"train_time":13.27,"predict_time":1.624,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-68","imputed":false},{"position":31,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1309.0,"score":0.15,"rank":35.28,"hrank":25.17,"imp":20.687,"train_time":7.18,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-32","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1306.0,"score":0.171,"rank":35.57,"hrank":21.21,"imp":20.787,"train_time":1053.94,"predict_time":3.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-61","imputed":false},{"position":33,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1303.0,"score":0.13,"rank":35.86,"hrank":26.26,"imp":20.617,"train_time":798.08,"predict_time":2.227,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-45","imputed":false},{"position":34,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1295.0,"score":0.186,"rank":36.61,"hrank":18.72,"imp":21.102,"train_time":6049.28,"predict_time":8.736,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-75","imputed":false},{"position":35,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1294.0,"score":0.181,"rank":36.62,"hrank":20.38,"imp":21.052,"train_time":8.79,"predict_time":0.411,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-63","imputed":false},{"position":36,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1287.0,"score":0.102,"rank":37.26,"hrank":25.28,"imp":20.839,"train_time":798.08,"predict_time":0.353,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+37\/-40","imputed":false},{"position":37,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1284.0,"score":0.202,"rank":37.58,"hrank":17.78,"imp":21.519,"train_time":51.45,"predict_time":43.707,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-79","imputed":false},{"position":38,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1274.0,"score":0.102,"rank":38.5,"hrank":27.48,"imp":21.881,"train_time":2499.4,"predict_time":0.174,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-38","imputed":false},{"position":39,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1266.0,"score":0.124,"rank":39.22,"hrank":22.23,"imp":21.964,"train_time":1053.94,"predict_time":0.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-53","imputed":false},{"position":40,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1264.0,"score":0.122,"rank":39.36,"hrank":27.52,"imp":22.244,"train_time":1142.82,"predict_time":0.536,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-55","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1264.0,"score":0.114,"rank":39.42,"hrank":28.7,"imp":21.466,"train_time":2977.91,"predict_time":2.415,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-50","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1251.0,"score":0.101,"rank":40.57,"hrank":28.23,"imp":22.515,"train_time":1142.82,"predict_time":0.043,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-49","imputed":false},{"position":43,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1248.0,"score":0.076,"rank":40.85,"hrank":28.9,"imp":22.449,"train_time":2499.4,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-46","imputed":false},{"position":44,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.102,"rank":41.16,"hrank":21.68,"imp":23.741,"train_time":15.55,"predict_time":0.346,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-56","imputed":false},{"position":45,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1239.0,"score":0.136,"rank":41.66,"hrank":21.88,"imp":22.54,"train_time":1424.22,"predict_time":9.206,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-60","imputed":false},{"position":46,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1237.0,"score":0.067,"rank":41.84,"hrank":29.85,"imp":22.215,"train_time":21290.02,"predict_time":161.221,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+42\/-38","imputed":false},{"position":47,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1229.0,"score":0.08,"rank":42.52,"hrank":26.11,"imp":23.092,"train_time":7.09,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-50","imputed":false},{"position":48,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1213.0,"score":0.093,"rank":43.98,"hrank":25.87,"imp":23.705,"train_time":4.96,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-62","imputed":false},{"position":49,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1209.0,"score":0.053,"rank":44.32,"hrank":29.91,"imp":23.087,"train_time":14.89,"predict_time":4.432,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+43\/-39","imputed":false},{"position":50,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1200.0,"score":0.087,"rank":45.13,"hrank":27.18,"imp":24.101,"train_time":703.03,"predict_time":1.431,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-73","imputed":false},{"position":51,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1195.0,"score":0.068,"rank":45.52,"hrank":32.65,"imp":23.51,"train_time":2977.91,"predict_time":0.156,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-57","imputed":false},{"position":52,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1189.0,"score":0.063,"rank":46.11,"hrank":27.16,"imp":23.856,"train_time":1424.22,"predict_time":0.749,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-55","imputed":false},{"position":53,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1176.0,"score":0.083,"rank":47.24,"hrank":24.59,"imp":24.928,"train_time":703.03,"predict_time":0.184,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-73","imputed":false},{"position":54,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1173.0,"score":0.02,"rank":47.48,"hrank":40.13,"imp":23.661,"train_time":2.8,"predict_time":0.238,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-42","imputed":false},{"position":55,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1159.0,"score":0.053,"rank":48.67,"hrank":38.27,"imp":23.989,"train_time":3.15,"predict_time":0.141,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-52","imputed":false},{"position":56,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1154.0,"score":0.094,"rank":49.08,"hrank":32.24,"imp":26.026,"train_time":15.45,"predict_time":2.611,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-74","imputed":false},{"position":57,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1149.0,"score":0.172,"rank":49.54,"hrank":12.08,"imp":26.164,"train_time":40588.15,"predict_time":5.606,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-106","imputed":false},{"position":58,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1147.0,"score":0.076,"rank":49.68,"hrank":26.46,"imp":25.619,"train_time":571.73,"predict_time":1.389,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-90","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1145.0,"score":0.164,"rank":49.86,"hrank":11.68,"imp":26.243,"train_time":40588.15,"predict_time":22.575,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-107","imputed":false},{"position":60,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1110.0,"score":0.047,"rank":52.73,"hrank":24.9,"imp":26.39,"train_time":571.73,"predict_time":0.129,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-80","imputed":false},{"position":61,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1108.0,"score":0.047,"rank":52.9,"hrank":30.64,"imp":26.239,"train_time":463.05,"predict_time":147.937,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-67","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1075.0,"score":0.033,"rank":55.54,"hrank":39.14,"imp":32.362,"train_time":391.09,"predict_time":1.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-80","imputed":false},{"position":63,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1062.0,"score":0.013,"rank":56.47,"hrank":39.41,"imp":27.66,"train_time":11.29,"predict_time":0.204,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-61","imputed":false},{"position":64,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1053.0,"score":0.109,"rank":57.16,"hrank":19.42,"imp":31.638,"train_time":502.34,"predict_time":7.328,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-143","imputed":false},{"position":65,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1044.0,"score":0.061,"rank":57.86,"hrank":35.41,"imp":30.573,"train_time":2.59,"predict_time":0.533,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-96","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1040.0,"score":0.018,"rank":58.09,"hrank":38.37,"imp":33.862,"train_time":391.09,"predict_time":0.404,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-83","imputed":false},{"position":67,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1030.0,"score":0.018,"rank":58.84,"hrank":50.62,"imp":29.915,"train_time":4.95,"predict_time":1.012,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-109","imputed":false},{"position":68,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1025.0,"score":0.052,"rank":59.18,"hrank":37.14,"imp":33.771,"train_time":483.07,"predict_time":0.874,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-94","imputed":false},{"position":69,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1013.0,"score":0.0,"rank":59.98,"hrank":38.6,"imp":28.985,"train_time":5.04,"predict_time":0.645,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-79","imputed":false},{"position":70,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1011.0,"score":0.027,"rank":60.1,"hrank":36.42,"imp":31.542,"train_time":121.17,"predict_time":1.63,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-107","imputed":false},{"position":71,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1003.0,"score":0.035,"rank":60.64,"hrank":39.25,"imp":34.289,"train_time":483.07,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-95","imputed":false},{"position":72,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.004,"rank":60.84,"hrank":53.16,"imp":31.589,"train_time":0.46,"predict_time":0.068,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-81","imputed":false},{"position":73,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.01,"rank":60.85,"hrank":52.3,"imp":30.863,"train_time":3.61,"predict_time":0.552,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-99","imputed":false},{"position":74,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":951.0,"score":0.0,"rank":63.83,"hrank":42.53,"imp":36.518,"train_time":56.11,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-81","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":930.0,"score":0.023,"rank":65.03,"hrank":43.58,"imp":37.379,"train_time":2.39,"predict_time":0.143,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-125","imputed":false},{"position":76,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":924.0,"score":0.016,"rank":65.33,"hrank":44.15,"imp":36.908,"train_time":121.17,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-108","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":601.0,"score":0.0,"rank":75.56,"hrank":68.42,"imp":51.733,"train_time":0.28,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-138","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_small/pareto_front_explorer.html
index c8301fd5c50b40227e5e052b5622307b3379201c..07f0a6cf0d46b79f763d3d6749408a1f2649020d 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3591113273,"imp":19.7060596781,"elo":1385.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404555565,"imp":20.6871232014,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.044036937,"imp":20.0698046927,"elo":1329.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"imp":19.8133804771,"elo":1339.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.044213771,"imp":23.7050328995,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0431349373,"imp":22.5154300972,"elo":1250.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5357489286,"imp":22.2444479533,"elo":1264.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0222777214,"imp":23.0921914965,"elo":1229.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0223863596,"imp":22.4494896445,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1740181026,"imp":21.8810699978,"elo":1273.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5520616377,"imp":30.8630142888,"elo":999.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1838318082,"imp":24.9277100341,"elo":1175.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4313148571,"imp":24.1014767252,"elo":1199.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6452165158,"imp":28.9851362347,"elo":1013.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7488523806,"imp":23.8558880567,"elo":1188.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.2058236264,"imp":22.5395728842,"elo":1238.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.140680637,"imp":23.9887075311,"elo":1159.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3111174713,"imp":19.8025816975,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.8999116636,"imp":19.4379681667,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.0441668928,"imp":16.3642636367,"elo":1456.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.7692980502,"imp":17.6815274987,"elo":1469.4,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"imp":23.741331829,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5246167968,"imp":20.1702121753,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7364459708,"imp":21.1018344418,"elo":1294.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6112675715,"imp":26.0263730767,"elo":1154.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0479211725,"imp":36.5184989749,"elo":951.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4043784213,"imp":33.8624198612,"elo":1040.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0518021797,"imp":32.3617974049,"elo":1074.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0681202832,"imp":31.5894961654,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1287782959,"imp":26.3901431297,"elo":1110.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.3894088878,"imp":25.6188443315,"elo":1147.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.4321040652,"imp":23.0872006076,"elo":1208.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.5333309154,"imp":18.8691762749,"elo":1367.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":18.9330661418,"imp":17.3665542371,"elo":1444.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7616019805,"imp":14.2509319878,"elo":1564.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2533832566,"imp":14.2449728574,"elo":1563.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1139940547,"imp":13.594871025,"elo":1580.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6235508255,"imp":21.4661767309,"elo":1312.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"imp":21.5194839239,"elo":1283.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":33.9783644772,"imp":19.0856803533,"elo":1366.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":231.8320376913,"imp":18.4000648166,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"imp":18.1136238564,"elo":1424.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.3168064912,"imp":5.4991266007,"elo":1712.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5331077226,"imp":30.5734637475,"elo":1043.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.7136697978,"imp":19.04083073,"elo":1402.6,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"imp":13.7485224273,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4108430406,"imp":21.0522048416,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4078633109,"imp":19.2389527572,"elo":1374.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.4179947198,"imp":18.390994012,"elo":1421.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"imp":14.4018023429,"elo":1573.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4074238608,"imp":13.2254568427,"elo":1611.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4197455135,"imp":20.4098293628,"elo":1360.2,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.9285088661,"imp":19.7134637319,"elo":1395.2,"imputed":true,"imputed_pct":7.1428571429,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":45.7795702843,"imp":17.4885879907,"elo":1469.2,"imputed":true,"imputed_pct":7.1428571429,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.3276990403,"imp":31.6377671528,"elo":1053.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.6062080426,"imp":26.1642060156,"elo":1148.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.5750090197,"imp":26.2428857649,"elo":1145.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0545537099,"imp":20.5323700246,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.203890602,"imp":27.6596840139,"elo":1062.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.155776825,"imp":23.5099682716,"elo":1195.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4148361224,"imp":21.4658690556,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2382075678,"imp":23.6608834769,"elo":1172.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3532173564,"imp":20.8392623954,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.2269115228,"imp":20.6171011633,"elo":1303.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":147.9369405767,"imp":26.238698046,"elo":1108.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":161.2206474003,"imp":22.2151998858,"elo":1236.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":663.2627455245,"imp":19.9255305459,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.0121457497,"imp":29.9149862792,"elo":1029.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1687223142,"imp":21.9636192456,"elo":1265.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.0619945914,"imp":20.7865748872,"elo":1306.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3591113273,"x_train":331.0009616013,"imp":19.7060596781,"elo":1385.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404555565,"x_train":7.1761452725,"imp":20.6871232014,"elo":1309.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.044036937,"x_train":1542.9492660346,"imp":20.0698046927,"elo":1329.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"x_train":1542.9492660346,"imp":19.8133804771,"elo":1339.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.044213771,"x_train":4.9608581751,"imp":23.7050328995,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0431349373,"x_train":1142.8199158523,"imp":22.5154300972,"elo":1250.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5357489286,"x_train":1142.8199158523,"imp":22.2444479533,"elo":1264.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0222777214,"x_train":7.0915913087,"imp":23.0921914965,"elo":1229.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0223863596,"x_train":2499.3997394449,"imp":22.4494896445,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1740181026,"x_train":2499.3997394449,"imp":21.8810699978,"elo":1273.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5520616377,"x_train":3.6063372166,"imp":30.8630142888,"elo":999.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1838318082,"x_train":703.0330785914,"imp":24.9277100341,"elo":1175.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4313148571,"x_train":703.0330785914,"imp":24.1014767252,"elo":1199.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6452165158,"x_train":5.0370258118,"imp":28.9851362347,"elo":1013.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7488523806,"x_train":1424.216740031,"imp":23.8558880567,"elo":1188.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.2058236264,"x_train":1424.216740031,"imp":22.5395728842,"elo":1238.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.140680637,"x_train":3.149413842,"imp":23.9887075311,"elo":1159.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3111174713,"x_train":735.3700351286,"imp":19.8025816975,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.8999116636,"x_train":735.3700351286,"imp":19.4379681667,"elo":1370.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.0441668928,"x_train":27.7685088788,"imp":16.3642636367,"elo":1456.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.7692980502,"x_train":144.3147662312,"imp":17.6815274987,"elo":1469.4,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"x_train":15.5476029367,"imp":23.741331829,"elo":1244.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5246167968,"x_train":6049.2787189098,"imp":20.1702121753,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7364459708,"x_train":6049.2787189098,"imp":21.1018344418,"elo":1294.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6112675715,"x_train":15.4475945518,"imp":26.0263730767,"elo":1154.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0479211725,"x_train":56.1148745356,"imp":36.5184989749,"elo":951.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4043784213,"x_train":391.0862000393,"imp":33.8624198612,"elo":1040.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0518021797,"x_train":391.0862000393,"imp":32.3617974049,"elo":1074.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0681202832,"x_train":0.4628071975,"imp":31.5894961654,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1287782959,"x_train":571.7337775834,"imp":26.3901431297,"elo":1110.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.3894088878,"x_train":571.7337775834,"imp":25.6188443315,"elo":1147.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.4321040652,"x_train":14.8938989136,"imp":23.0872006076,"elo":1208.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.5333309154,"x_train":3423.6100841542,"imp":18.8691762749,"elo":1367.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":18.9330661418,"x_train":3423.6100841542,"imp":17.3665542371,"elo":1444.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7616019805,"x_train":6.0754426474,"imp":14.2509319878,"elo":1564.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2533832566,"x_train":2289.3887157667,"imp":14.2449728574,"elo":1563.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1139940547,"x_train":2289.3887157667,"imp":13.594871025,"elo":1580.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6235508255,"x_train":13.2671754693,"imp":21.4661767309,"elo":1312.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"x_train":51.4453139381,"imp":21.5194839239,"elo":1283.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":33.9783644772,"x_train":5271.4650532739,"imp":19.0856803533,"elo":1366.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":231.8320376913,"x_train":5271.4650532739,"imp":18.4000648166,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"x_train":4.3495748865,"imp":18.1136238564,"elo":1424.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.3168064912,"x_train":38.5654426559,"imp":5.4991266007,"elo":1712.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5331077226,"x_train":2.5922354756,"imp":30.5734637475,"elo":1043.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.7136697978,"x_train":8.8926032519,"imp":19.04083073,"elo":1402.6,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"x_train":3.2221111875,"imp":13.7485224273,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4108430406,"x_train":8.785094234,"imp":21.0522048416,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4078633109,"x_train":2469.4800149006,"imp":19.2389527572,"elo":1374.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.4179947198,"x_train":2469.4800149006,"imp":18.390994012,"elo":1421.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"x_train":6.2559884135,"imp":14.4018023429,"elo":1573.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4074238608,"x_train":7.1656661471,"imp":13.2254568427,"elo":1611.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4197455135,"x_train":4.1476660437,"imp":20.4098293628,"elo":1360.2,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.9285088661,"x_train":3303.0148093272,"imp":19.7134637319,"elo":1395.2,"imputed":true,"imputed_pct":7.1428571429,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":45.7795702843,"x_train":3303.0148093272,"imp":17.4885879907,"elo":1469.2,"imputed":true,"imputed_pct":7.1428571429,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.3276990403,"x_train":502.344886049,"imp":31.6377671528,"elo":1053.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.6062080426,"x_train":40588.1540174129,"imp":26.1642060156,"elo":1148.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.5750090197,"x_train":40588.1540174129,"imp":26.2428857649,"elo":1145.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0545537099,"x_train":2.2399715166,"imp":20.5323700246,"elo":1340.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.203890602,"x_train":11.2922428764,"imp":27.6596840139,"elo":1062.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.155776825,"x_train":2977.9095379662,"imp":23.5099682716,"elo":1195.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4148361224,"x_train":2977.9095379662,"imp":21.4658690556,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2382075678,"x_train":2.7957309689,"imp":23.6608834769,"elo":1172.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3532173564,"x_train":798.0801278695,"imp":20.8392623954,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.2269115228,"x_train":798.0801278695,"imp":20.6171011633,"elo":1303.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":147.9369405767,"x_train":463.0485235354,"imp":26.238698046,"elo":1108.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":161.2206474003,"x_train":21290.0189805033,"imp":22.2151998858,"elo":1236.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":663.2627455245,"x_train":21290.0189805033,"imp":19.9255305459,"elo":1333.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.0121457497,"x_train":4.9505740724,"imp":29.9149862792,"elo":1029.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1687223142,"x_train":1053.935912149,"imp":21.9636192456,"elo":1265.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.0619945914,"x_train":1053.935912149,"imp":20.7865748872,"elo":1306.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_small/tuning_trajectories_explorer.html
index f5558e92c7509afed3504ca67ca6801bb2e6af99..72ef543b774b18905fe37c2f3a47e636b426eb3a 100644
--- a/data/imputation_yes/splits_all/tasks_classification/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":331.0009616013,"imp":19.7314830627,"elo":1404.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":7.1761452725,"imp":20.7113994778,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":13.6550050397,"imp":20.1170443844,"elo":1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":34.7005566859,"imp":19.8508715466,"elo":1360.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":69.9246996151,"imp":19.7752582045,"elo":1363.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":185.5351574323,"imp":19.729712874,"elo":1364.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":379.2610359226,"imp":19.7517033973,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":771.7715864818,"imp":19.7859122312,"elo":1359.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1152.9335256343,"imp":19.8205689224,"elo":1357.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1542.9492660346,"imp":19.8348410196,"elo":1355.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.9608581751,"imp":23.7285724487,"elo":1220.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.8616521277,"imp":23.3875809701,"elo":1238.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":27.0121608835,"imp":23.2151534553,"elo":1245.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":54.6281407695,"imp":23.0490461444,"elo":1251.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":140.6370736031,"imp":22.7163749183,"elo":1262.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":283.4636990992,"imp":22.5611626181,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":569.3917360284,"imp":22.3853294293,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":853.7391955308,"imp":22.3113375367,"elo":1276.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1142.8199158523,"imp":22.2675629434,"elo":1278.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.0915913087,"imp":23.1168514565,"elo":1236.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":16.5198837522,"imp":22.8710187086,"elo":1243.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":59.6568017021,"imp":22.5204165249,"elo":1259.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":123.907431834,"imp":22.4020611983,"elo":1264.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":318.4600106407,"imp":22.2413006651,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":623.1113495163,"imp":22.0907238683,"elo":1278.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1256.5403166692,"imp":21.9753381303,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1885.0271107114,"imp":21.936953416,"elo":1286.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2499.3997394449,"imp":21.9111087403,"elo":1287.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.6063372166,"imp":30.8854979002,"elo":999.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.3528355841,"imp":28.8173633315,"elo":1043.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":18.0710159165,"imp":25.9453418752,"elo":1116.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":35.5062409712,"imp":25.0333458135,"elo":1160.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":87.9242300599,"imp":24.472263173,"elo":1191.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":175.4018189887,"imp":24.3147431045,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":350.0379937538,"imp":24.2060925304,"elo":1206.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":524.8382968951,"imp":24.1364265994,"elo":1211.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":703.0330785914,"imp":24.1263947462,"elo":1211.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.0370258118,"imp":29.0087622165,"elo":1018.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":12.7722389468,"imp":27.4633941996,"elo":1068.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":35.5543374641,"imp":25.1295728354,"elo":1149.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":67.9976554587,"imp":24.0585131576,"elo":1190.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":172.4339504664,"imp":23.177542373,"elo":1222.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":350.2703386124,"imp":22.795970252,"elo":1240.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":703.7735059836,"imp":22.6395087451,"elo":1247.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1059.2142650988,"imp":22.5617586484,"elo":1252.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1424.216740031,"imp":22.5588880557,"elo":1253.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":3.149413842,"imp":24.0122155998,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":7.1207733159,"imp":21.8194289055,"elo":1258.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":17.9053372504,"imp":20.6168093564,"elo":1329.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":35.9737069678,"imp":20.1339894344,"elo":1356.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":91.1663943012,"imp":19.8065784725,"elo":1374.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":182.6006011031,"imp":19.6945086585,"elo":1381.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":365.5882548593,"imp":19.5478209742,"elo":1389.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":548.151079493,"imp":19.4920831191,"elo":1391.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":735.3700351286,"imp":19.4609411083,"elo":1392.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":27.7685088788,"imp":16.427485487,"elo":1474.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":144.3147662312,"imp":17.7071143689,"elo":1489.2,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":15.5476029367,"imp":23.7653044082,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":52.4814325925,"imp":22.4437416901,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":150.689630028,"imp":20.3414733419,"elo":1346.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":293.311977236,"imp":19.976392539,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":719.031159336,"imp":19.6095965152,"elo":1371.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1498.353139385,"imp":19.9049242858,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3004.008741271,"imp":20.4904328373,"elo":1334.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4473.3946349608,"imp":20.9059308904,"elo":1318.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6049.2787189098,"imp":21.1314965711,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":15.4475945518,"imp":26.0456274523,"elo":1163.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":56.1148745356,"imp":37.2873915909,"elo":945.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":121.64643757,"imp":35.2127623295,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":327.8283737012,"imp":32.5778060051,"elo":1078.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":391.0862000393,"imp":32.3808822497,"elo":1084.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4628071975,"imp":31.6270415573,"elo":997.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.2094586922,"imp":28.8832028298,"elo":1045.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":11.9623791751,"imp":27.3799667675,"elo":1078.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":26.2110927853,"imp":26.5515834824,"elo":1107.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":68.7900402557,"imp":25.9395358372,"elo":1142.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":140.1670296597,"imp":25.7938034149,"elo":1150.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":282.7281586439,"imp":25.7001836465,"elo":1154.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":425.7671432604,"imp":25.6585922387,"elo":1156.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":571.7337775834,"imp":25.638856046,"elo":1156.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":14.8938989136,"imp":23.1118244924,"elo":1215.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":30.7592659452,"imp":20.4334314116,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":78.4494180107,"imp":18.703527897,"elo":1406.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":164.8792540854,"imp":18.1680062365,"elo":1433.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":421.7704708255,"imp":17.7970975909,"elo":1453.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":853.4354612721,"imp":17.6234798583,"elo":1463.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1707.2754627676,"imp":17.5243642454,"elo":1467.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2551.8451865301,"imp":17.4634972701,"elo":1469.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3423.6100841542,"imp":17.3935610495,"elo":1468.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.0754426474,"imp":14.2770214878,"elo":1587.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":20.0220773915,"imp":14.3562326955,"elo":1587.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":54.3500923902,"imp":14.1443286086,"elo":1596.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":108.9572713635,"imp":14.0166404811,"elo":1608.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":283.232361067,"imp":13.8687323532,"elo":1610.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":566.4138152119,"imp":13.7820044879,"elo":1609.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1137.2039338041,"imp":13.7122341007,"elo":1607.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1710.4116369596,"imp":13.6790634381,"elo":1606.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2289.3887157667,"imp":13.6331019309,"elo":1603.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.2671754693,"imp":21.4909131198,"elo":1329.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.4453139381,"imp":21.5422001261,"elo":1297.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":75.9360943292,"imp":20.8200635691,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":142.0530215894,"imp":20.0577634612,"elo":1358.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":257.4275777592,"imp":19.7264723164,"elo":1372.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":641.1662937545,"imp":19.0721472219,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1303.1613548368,"imp":18.7523556396,"elo":1405.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2614.4217732811,"imp":18.5536511286,"elo":1413.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3931.8869009419,"imp":18.4653427226,"elo":1416.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":5271.4650532739,"imp":18.4311085962,"elo":1417.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":4.3495748865,"imp":18.1369272031,"elo":1440.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.5654426559,"imp":5.5258344526,"elo":1730.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":2.5922354756,"imp":30.5935474501,"elo":1057.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":8.8926032519,"imp":19.0653151577,"elo":1421.1,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":3.2221111875,"imp":13.7746381175,"elo":1581.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":8.785094234,"imp":21.0769335204,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":20.4659364224,"imp":20.6583913518,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":57.4905112627,"imp":20.0703658661,"elo":1360.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":119.8455247692,"imp":19.4990022415,"elo":1391.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":304.1351923333,"imp":19.0351461844,"elo":1413.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":611.19598978,"imp":18.7489927191,"elo":1426.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1222.0069753852,"imp":18.5618028857,"elo":1436.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1840.5789024835,"imp":18.4725078121,"elo":1440.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2469.4800149006,"imp":18.417046361,"elo":1444.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":6.2559884135,"imp":14.4275655422,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":7.1656661471,"imp":13.2508237899,"elo":1636.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4.1476660437,"imp":20.4345472445,"elo":1379.3,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":26.5792281881,"imp":19.7043194604,"elo":1412.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":70.532340662,"imp":18.9256735036,"elo":1449.5,"imputed":true,"imputed_pct":7.1428571429,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":153.4472635413,"imp":18.3743822729,"elo":1467.6,"imputed":true,"imputed_pct":7.1428571429,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":384.6503124909,"imp":17.9455456048,"elo":1481.7,"imputed":true,"imputed_pct":7.1428571429,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":782.4457026513,"imp":17.7034741449,"elo":1490.6,"imputed":true,"imputed_pct":7.1428571429,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1599.9482137872,"imp":17.5784237822,"elo":1493.5,"imputed":true,"imputed_pct":7.1428571429,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2421.1383522149,"imp":17.5214233146,"elo":1494.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3303.0148093272,"imp":17.4897925089,"elo":1492.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":502.344886049,"imp":31.6614530388,"elo":1061.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":797.2338293006,"imp":31.5025312977,"elo":1067.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1342.9528314957,"imp":31.1682270189,"elo":1074.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2054.553219683,"imp":30.3530873767,"elo":1090.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4867.9697698991,"imp":28.8993043897,"elo":1112.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":9985.461750605,"imp":27.5139237136,"elo":1132.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":20581.7115313909,"imp":26.8136784549,"elo":1146.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":30569.2328830115,"imp":26.4932772127,"elo":1154.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":40588.1540174129,"imp":26.2682455382,"elo":1161.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":2.2399715166,"imp":20.5555512957,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":11.2922428764,"imp":27.6745023945,"elo":1071.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":26.0516500508,"imp":25.334049399,"elo":1136.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":70.9730313123,"imp":23.5207793625,"elo":1188.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":152.8912289988,"imp":22.7517321807,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":376.394024437,"imp":22.1677523519,"elo":1245.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":738.844201808,"imp":21.8295715986,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1456.0991336408,"imp":21.6563276672,"elo":1269.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2205.8455340006,"imp":21.5637391401,"elo":1275.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2977.9095379662,"imp":21.4813817712,"elo":1280.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.7957309689,"imp":23.6837869338,"elo":1175.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":7.086534818,"imp":21.8717890328,"elo":1249.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":19.7630506746,"imp":21.4450947061,"elo":1273.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":38.4377569942,"imp":21.1935791369,"elo":1288.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":99.4892942614,"imp":20.9686896121,"elo":1300.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":198.7048872189,"imp":20.8740205609,"elo":1306.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":397.0783434511,"imp":20.788684386,"elo":1311.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":594.6588102741,"imp":20.7123049817,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":798.0801278695,"imp":20.6412698043,"elo":1319.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":463.0485235354,"imp":26.2615768901,"elo":1116.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1277.7801478637,"imp":24.0311921014,"elo":1174.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":3759.8173217043,"imp":21.7438259686,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":8270.172943981,"imp":20.8719953174,"elo":1305.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":20344.7602093409,"imp":19.970481067,"elo":1349.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":21290.0189805033,"imp":19.9488368573,"elo":1351.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":4.9505740724,"imp":29.9376457141,"elo":1035.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":10.1548540276,"imp":26.9994140057,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":26.178393176,"imp":23.2017629375,"elo":1217.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":52.4324179391,"imp":21.6514870778,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":131.4097749593,"imp":21.010559235,"elo":1321.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":262.5999052849,"imp":20.8340561459,"elo":1327.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":524.6656622172,"imp":20.8000809325,"elo":1326.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":786.5490444842,"imp":20.8106664244,"elo":1325.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1053.935912149,"imp":20.8120680457,"elo":1323.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3591113273,"x_train":331.0009616013,"imp":19.7314830627,"elo":1404.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404555565,"x_train":7.1761452725,"imp":20.7113994778,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0841906456,"x_train":13.6550050397,"imp":20.1170443844,"elo":1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.158469367,"x_train":34.7005566859,"imp":19.8508715466,"elo":1360.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2195178645,"x_train":69.9246996151,"imp":19.7752582045,"elo":1363.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2815893316,"x_train":185.5351574323,"imp":19.729712874,"elo":1364.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3179145648,"x_train":379.2610359226,"imp":19.7517033973,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806002107,"x_train":771.7715864818,"imp":19.7859122312,"elo":1359.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3873615123,"x_train":1152.9335256343,"imp":19.8205689224,"elo":1357.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3960267603,"x_train":1542.9492660346,"imp":19.8348410196,"elo":1355.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.044213771,"x_train":4.9608581751,"imp":23.7285724487,"elo":1220.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0842869021,"x_train":10.8616521277,"imp":23.3875809701,"elo":1238.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1733534673,"x_train":27.0121608835,"imp":23.2151534553,"elo":1245.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2454079283,"x_train":54.6281407695,"imp":23.0490461444,"elo":1251.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3360644036,"x_train":140.6370736031,"imp":22.7163749183,"elo":1262.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3991994324,"x_train":283.4636990992,"imp":22.5611626181,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4584564684,"x_train":569.3917360284,"imp":22.3853294293,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5006577968,"x_train":853.7391955308,"imp":22.3113375367,"elo":1276.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5312563547,"x_train":1142.8199158523,"imp":22.2675629434,"elo":1278.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0222777214,"x_train":7.0915913087,"imp":23.1168514565,"elo":1236.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.051273724,"x_train":16.5198837522,"imp":22.8710187086,"elo":1243.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0788065997,"x_train":59.6568017021,"imp":22.5204165249,"elo":1259.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.097778986,"x_train":123.907431834,"imp":22.4020611983,"elo":1264.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1309347475,"x_train":318.4600106407,"imp":22.2413006651,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1473544657,"x_train":623.1113495163,"imp":22.0907238683,"elo":1278.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1683065578,"x_train":1256.5403166692,"imp":21.9753381303,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1758856014,"x_train":1885.0271107114,"imp":21.936953416,"elo":1286.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.181036219,"x_train":2499.3997394449,"imp":21.9111087403,"elo":1287.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5520616377,"x_train":3.6063372166,"imp":30.8854979002,"elo":999.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6246844205,"x_train":7.3528355841,"imp":28.8173633315,"elo":1043.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6183400546,"x_train":18.0710159165,"imp":25.9453418752,"elo":1116.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7242510207,"x_train":35.5062409712,"imp":25.0333458135,"elo":1160.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9536012796,"x_train":87.9242300599,"imp":24.472263173,"elo":1191.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1056205394,"x_train":175.4018189887,"imp":24.3147431045,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2895955274,"x_train":350.0379937538,"imp":24.2060925304,"elo":1206.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3699672148,"x_train":524.8382968951,"imp":24.1364265994,"elo":1211.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4314907954,"x_train":703.0330785914,"imp":24.1263947462,"elo":1211.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6452165158,"x_train":5.0370258118,"imp":29.0087622165,"elo":1018.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3091545347,"x_train":12.7722389468,"imp":27.4633941996,"elo":1068.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7086381558,"x_train":35.5543374641,"imp":25.1295728354,"elo":1149.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.7948756042,"x_train":67.9976554587,"imp":24.0585131576,"elo":1190.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.5085868791,"x_train":172.4339504664,"imp":23.177542373,"elo":1222.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.5711464723,"x_train":350.2703386124,"imp":22.795970252,"elo":1240.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7445847733,"x_train":703.7735059836,"imp":22.6395087451,"elo":1247.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.6616951336,"x_train":1059.2142650988,"imp":22.5617586484,"elo":1252.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.1991520011,"x_train":1424.216740031,"imp":22.5588880557,"elo":1253.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.140680637,"x_train":3.149413842,"imp":24.0122155998,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.557564348,"x_train":7.1207733159,"imp":21.8194289055,"elo":1258.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8204403589,"x_train":17.9053372504,"imp":20.6168093564,"elo":1329.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0270516849,"x_train":35.9737069678,"imp":20.1339894344,"elo":1356.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3215357247,"x_train":91.1663943012,"imp":19.8065784725,"elo":1374.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5580626458,"x_train":182.6006011031,"imp":19.6945086585,"elo":1381.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6848179893,"x_train":365.5882548593,"imp":19.5478209742,"elo":1389.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6963950347,"x_train":548.151079493,"imp":19.4920831191,"elo":1391.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.8237866308,"x_train":735.3700351286,"imp":19.4609411083,"elo":1392.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":5.0441668928,"x_train":27.7685088788,"imp":16.427485487,"elo":1474.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":3.7692980502,"x_train":144.3147662312,"imp":17.7071143689,"elo":1489.2,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3462658876,"x_train":15.5476029367,"imp":23.7653044082,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8804757883,"x_train":52.4814325925,"imp":22.4437416901,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9515166073,"x_train":150.689630028,"imp":20.3414733419,"elo":1346.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9139871186,"x_train":293.311977236,"imp":19.976392539,"elo":1358.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.2991192179,"x_train":719.031159336,"imp":19.6095965152,"elo":1371.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.6324379245,"x_train":1498.353139385,"imp":19.9049242858,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.2170906616,"x_train":3004.008741271,"imp":20.4904328373,"elo":1334.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.2545907476,"x_train":4473.3946349608,"imp":20.9059308904,"elo":1318.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7287422808,"x_train":6049.2787189098,"imp":21.1314965711,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6112675715,"x_train":15.4475945518,"imp":26.0456274523,"elo":1163.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0479211725,"x_train":56.1148745356,"imp":37.2873915909,"elo":945.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4213204184,"x_train":121.64643757,"imp":35.2127623295,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.9452456188,"x_train":327.8283737012,"imp":32.5778060051,"elo":1078.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0563537894,"x_train":391.0862000393,"imp":32.3808822497,"elo":1084.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0681202832,"x_train":0.4628071975,"imp":31.6270415573,"elo":997.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2374252662,"x_train":3.2094586922,"imp":28.8832028298,"elo":1045.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.441254012,"x_train":11.9623791751,"imp":27.3799667675,"elo":1078.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.648284888,"x_train":26.2110927853,"imp":26.5515834824,"elo":1107.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8046320893,"x_train":68.7900402557,"imp":25.9395358372,"elo":1142.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9831168585,"x_train":140.1670296597,"imp":25.7938034149,"elo":1150.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2033557443,"x_train":282.7281586439,"imp":25.7001836465,"elo":1154.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.305785724,"x_train":425.7671432604,"imp":25.6585922387,"elo":1156.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.388919598,"x_train":571.7337775834,"imp":25.638856046,"elo":1156.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.4321040652,"x_train":14.8938989136,"imp":23.1118244924,"elo":1215.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.5278018741,"x_train":30.7592659452,"imp":20.4334314116,"elo":1322.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.9302452502,"x_train":78.4494180107,"imp":18.703527897,"elo":1406.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.8524000954,"x_train":164.8792540854,"imp":18.1680062365,"elo":1433.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.294458449,"x_train":421.7704708255,"imp":17.7970975909,"elo":1453.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4003571581,"x_train":853.4354612721,"imp":17.6234798583,"elo":1463.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.0745491458,"x_train":1707.2754627676,"imp":17.5243642454,"elo":1467.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.8336064118,"x_train":2551.8451865301,"imp":17.4634972701,"elo":1469.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.0794572094,"x_train":3423.6100841542,"imp":17.3935610495,"elo":1468.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7616019805,"x_train":6.0754426474,"imp":14.2770214878,"elo":1587.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3608475094,"x_train":20.0220773915,"imp":14.3562326955,"elo":1587.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.9479323005,"x_train":54.3500923902,"imp":14.1443286086,"elo":1596.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1961228236,"x_train":108.9572713635,"imp":14.0166404811,"elo":1608.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7212619669,"x_train":283.232361067,"imp":13.8687323532,"elo":1610.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8314060794,"x_train":566.4138152119,"imp":13.7820044879,"elo":1609.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5661153905,"x_train":1137.2039338041,"imp":13.7122341007,"elo":1607.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0685248517,"x_train":1710.4116369596,"imp":13.6790634381,"elo":1606.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.5533679569,"x_train":2289.3887157667,"imp":13.6331019309,"elo":1603.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6235508255,"x_train":13.2671754693,"imp":21.4909131198,"elo":1329.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7074010359,"x_train":51.4453139381,"imp":21.5422001261,"elo":1297.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.7846214148,"x_train":75.9360943292,"imp":20.8200635691,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":68.4502119688,"x_train":142.0530215894,"imp":20.0577634612,"elo":1358.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":88.2706830363,"x_train":257.4275777592,"imp":19.7264723164,"elo":1372.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":120.5529066959,"x_train":641.1662937545,"imp":19.0721472219,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":148.5135318639,"x_train":1303.1613548368,"imp":18.7523556396,"elo":1405.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":181.9231911908,"x_train":2614.4217732811,"imp":18.5536511286,"elo":1413.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":200.1306803402,"x_train":3931.8869009419,"imp":18.4653427226,"elo":1416.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":228.0648501265,"x_train":5271.4650532739,"imp":18.4311085962,"elo":1417.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1689771704,"x_train":4.3495748865,"imp":18.1369272031,"elo":1440.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.3168064912,"x_train":38.5654426559,"imp":5.5258344526,"elo":1730.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5331077226,"x_train":2.5922354756,"imp":30.5935474501,"elo":1057.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.7136697978,"x_train":8.8926032519,"imp":19.0653151577,"elo":1421.1,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1671872076,"x_train":3.2221111875,"imp":13.7746381175,"elo":1581.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4108430406,"x_train":8.785094234,"imp":21.0769335204,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.7817725977,"x_train":20.4659364224,"imp":20.6583913518,"elo":1329.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.405202684,"x_train":57.4905112627,"imp":20.0703658661,"elo":1360.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.9090624077,"x_train":119.8455247692,"imp":19.4990022415,"elo":1391.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.5663201327,"x_train":304.1351923333,"imp":19.0351461844,"elo":1413.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":2.956697172,"x_train":611.19598978,"imp":18.7489927191,"elo":1426.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.3235732502,"x_train":1222.0069753852,"imp":18.5618028857,"elo":1436.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":3.405584308,"x_train":1840.5789024835,"imp":18.4725078121,"elo":1440.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":3.4143908721,"x_train":2469.4800149006,"imp":18.417046361,"elo":1444.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5385860219,"x_train":6.2559884135,"imp":14.4275655422,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4074238608,"x_train":7.1656661471,"imp":13.2508237899,"elo":1636.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4197455135,"x_train":4.1476660437,"imp":20.4345472445,"elo":1379.3,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.2036758824,"x_train":26.5792281881,"imp":19.7043194604,"elo":1412.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":8.7175627184,"x_train":70.532340662,"imp":18.9256735036,"elo":1449.5,"imputed":true,"imputed_pct":7.1428571429,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":13.3656054107,"x_train":153.4472635413,"imp":18.3743822729,"elo":1467.6,"imputed":true,"imputed_pct":7.1428571429,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":21.2732430653,"x_train":384.6503124909,"imp":17.9455456048,"elo":1481.7,"imputed":true,"imputed_pct":7.1428571429,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":28.7801538267,"x_train":782.4457026513,"imp":17.7034741449,"elo":1490.6,"imputed":true,"imputed_pct":7.1428571429,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":35.5433238381,"x_train":1599.9482137872,"imp":17.5784237822,"elo":1493.5,"imputed":true,"imputed_pct":7.1428571429,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":40.1153864785,"x_train":2421.1383522149,"imp":17.5214233146,"elo":1494.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":45.9942643145,"x_train":3303.0148093272,"imp":17.4897925089,"elo":1492.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.3276990403,"x_train":502.344886049,"imp":31.6614530388,"elo":1061.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.2461770633,"x_train":797.2338293006,"imp":31.5025312977,"elo":1067.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.851933236,"x_train":1342.9528314957,"imp":31.1682270189,"elo":1074.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.6518831225,"x_train":2054.553219683,"imp":30.3530873767,"elo":1090.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0496115636,"x_train":4867.9697698991,"imp":28.8993043897,"elo":1112.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.7051263351,"x_train":9985.461750605,"imp":27.5139237136,"elo":1132.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.2001866591,"x_train":20581.7115313909,"imp":26.8136784549,"elo":1146.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.3628807226,"x_train":30569.2328830115,"imp":26.4932772127,"elo":1154.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.2209155837,"x_train":40588.1540174129,"imp":26.2682455382,"elo":1161.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0545537099,"x_train":2.2399715166,"imp":20.5555512957,"elo":1358.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.203890602,"x_train":11.2922428764,"imp":27.6745023945,"elo":1071.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3965671437,"x_train":26.0516500508,"imp":25.334049399,"elo":1136.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7770360585,"x_train":70.9730313123,"imp":23.5207793625,"elo":1188.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0521362816,"x_train":152.8912289988,"imp":22.7517321807,"elo":1220.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.364215168,"x_train":376.394024437,"imp":22.1677523519,"elo":1245.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6688707494,"x_train":738.844201808,"imp":21.8295715986,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0488352505,"x_train":1456.0991336408,"imp":21.6563276672,"elo":1269.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.2724182677,"x_train":2205.8455340006,"imp":21.5637391401,"elo":1275.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.4133616403,"x_train":2977.9095379662,"imp":21.4813817712,"elo":1280.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2382075678,"x_train":2.7957309689,"imp":23.6837869338,"elo":1175.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6139127796,"x_train":7.086534818,"imp":21.8717890328,"elo":1249.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2441928973,"x_train":19.7630506746,"imp":21.4450947061,"elo":1273.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.450807434,"x_train":38.4377569942,"imp":21.1935791369,"elo":1288.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.643707399,"x_train":99.4892942614,"imp":20.9686896121,"elo":1300.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8330479874,"x_train":198.7048872189,"imp":20.8740205609,"elo":1306.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0008744691,"x_train":397.0783434511,"imp":20.788684386,"elo":1311.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0954188348,"x_train":594.6588102741,"imp":20.7123049817,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.1816683976,"x_train":798.0801278695,"imp":20.6412698043,"elo":1319.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":147.9369405767,"x_train":463.0485235354,"imp":26.2615768901,"elo":1116.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":301.0343409385,"x_train":1277.7801478637,"imp":24.0311921014,"elo":1174.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":433.1102919089,"x_train":3759.8173217043,"imp":21.7438259686,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":548.6068914868,"x_train":8270.172943981,"imp":20.8719953174,"elo":1305.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":636.7319509309,"x_train":20344.7602093409,"imp":19.970481067,"elo":1349.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":654.0373562851,"x_train":21290.0189805033,"imp":19.9488368573,"elo":1351.5,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.0121457497,"x_train":4.9505740724,"imp":29.9376457141,"elo":1035.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.0736404567,"x_train":10.1548540276,"imp":26.9994140057,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.2590656982,"x_train":26.178393176,"imp":23.2017629375,"elo":1217.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.6225412385,"x_train":52.4324179391,"imp":21.6514870778,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8253774488,"x_train":131.4097749593,"imp":21.010559235,"elo":1321.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2506435468,"x_train":262.5999052849,"imp":20.8340561459,"elo":1327.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.5518696321,"x_train":524.6656622172,"imp":20.8000809325,"elo":1326.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.795048319,"x_train":786.5490444842,"imp":20.8106664244,"elo":1325.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.0515616235,"x_train":1053.935912149,"imp":20.8120680457,"elo":1323.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_classification/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..ffcc5c6c7bc870f8d63cb3cc02e3f789d1e5c730
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_classification/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "TabPFN-2.6 (default)", "Mitra (default)", "TabPFNv2 (tuned + ensembled)", "TabDPT-Turbo (default)", "LimiX (default)", "RealMLP (tuned + ensembled)", "TabDPT (tuned + ensembled)", "TabM (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabICL (default)", "TabPFNv2 (tuned)", "TabPFNv2 (default)", "TabDPT (tuned)", "BetaTabPFN (default)", "RealMLP (tuned)", "TabM (tuned)", "TabSwift (default)", "LightGBM (tuned + ensembled)", "TabDPT (default)", "LightGBM (tuned)", "CatBoost (tuned + ensembled)", "ModernNCA (tuned)", "xRFM (tuned + ensembled)", "SAP-RPT-OSS (default)", "iLTM (tuned + ensembled)", "CatBoost (tuned)", "xRFM (tuned)", "XGBoost (tuned + ensembled)", "CatBoost (default)", "EBM (tuned + ensembled)", "TabM (default)", "ModernNCA (default)", "XGBoost (tuned)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "EBM (tuned)", "EBM (default)", "iLTM (tuned)", "RealMLP (default)", "FastaiMLP (tuned)", "ChimeraBoost (tuned + ensembled)", "ExtraTrees (tuned + ensembled)", "TorchMLP (tuned)", "ChimeraBoost (tuned)", "OrionMSP (default)", "ExtraTrees (tuned)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "ChimeraBoost (default)", "RandomForest (tuned + ensembled)", "XGBoost (default)", "LightGBM (default)", "TabFlex (default)", "RandomForest (tuned)", "Linear (tuned + ensembled)", "TorchMLP (default)", "iLTM (default)", "TabSTAR (default)", "PerpetualBooster (tuned + ensembled)", "Linear (tuned)", "xRFM (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "ExtraTrees (default)", "RandomForest (default)", "Linear (default)", "PerpetualBooster (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.795634925365448, 0.7876983880996704, 0.8194444179534912, 0.7678571343421936, 0.7708333134651184, 0.7757936716079712, 0.8035714030265808, 0.85317462682724, 0.863095223903656, 0.8313491940498352, 0.879960298538208, 0.8571428656578064, 0.886904776096344, 0.8650793433189392, 0.9047619104385376, 0.932539701461792, 0.8670634627342224, 0.829365074634552, 0.8789682388305664, 0.858134925365448, 0.851190447807312, 0.8948412537574768, 0.8690476417541504, 0.9007936716079712, 0.8908730149269104, 0.898809552192688, 0.9186508059501648, 0.9027777910232544, 0.9126983880996704, 0.8373016119003296, 0.920634925365448, 0.9246031641960144, 0.8928571343421936, 0.932539701461792, 0.9265872836112976, 0.8948412537574768, 0.9007936716079712, 0.9047619104385376, 0.932539701461792, 0.89682537317276, 0.9444444179534912, 0.9444444179534912, 0.9067460298538208, 0.9007936716079712, 0.9345238208770752, 0.9186508059501648, 0.9226190447807312, 0.9285714030265808, 0.9384920597076416, 0.9523809552192688, 0.9384920597076416, 0.9305555820465088, 0.9285714030265808, 0.8253968358039856, 0.8253968358039856, 0.9464285969734192, 0.9563491940498352, 0.9583333134651184, 0.954365074634552, 0.9265872836112976, 0.954365074634552, 0.9365079402923584, 0.9603174328804016, 0.9305555820465088, 0.8710317611694336, 0.9821428656578064, 0.9444444179534912, 0.9642857313156128, 0.97817462682724, 0.966269850730896, 0.9642857313156128, 0.988095223903656, 0.9801587462425232, 0.9603174328804016, 0.9801587462425232, 0.9563491940498352, 0.9940476417541504], [0.204365074634552, NaN, 0.5555555820465088, 0.5257936716079712, 0.601190447807312, 0.5753968358039856, 0.5972222089767456, 0.6111111044883728, 0.7063491940498352, 0.7023809552192688, 0.6964285969734192, 0.7182539701461792, 0.7242063283920288, 0.783730149269104, 0.7579365372657776, 0.7420634627342224, 0.8402777910232544, 0.7728174328804016, 0.7232142686843872, 0.8055555820465088, 0.7371031641960144, 0.7638888955116272, 0.795634925365448, 0.8432539701461792, 0.8154761791229248, 0.8194444179534912, 0.8333333134651184, 0.8492063283920288, 0.8492063283920288, 0.85317462682724, 0.7400793433189392, 0.8769841194152832, 0.8591269850730896, 0.8472222089767456, 0.8501983880996704, 0.875, 0.8392857313156128, 0.8194444179534912, 0.8640872836112976, 0.863095223903656, 0.8303571343421936, 0.89682537317276, 0.886904776096344, 0.85317462682724, 0.8660714030265808, 0.91567462682724, 0.9027777910232544, 0.8839285969734192, 0.908730149269104, 0.9246031641960144, 0.9166666865348816, 0.908730149269104, 0.9126983880996704, 0.9226190447807312, 0.8125, 0.817460298538208, 0.908730149269104, 0.920634925365448, 0.9583333134651184, 0.9404761791229248, 0.9166666865348816, 0.942460298538208, 0.9047619104385376, 0.9503968358039856, 0.9384920597076416, 0.85317462682724, 0.9365079402923584, 0.9126983880996704, 0.9682539701461792, 0.9503968358039856, 0.9583333134651184, 0.942460298538208, 0.9742063283920288, 0.97817462682724, 0.9484127163887024, 0.9642857313156128, 0.9523809552192688, 0.9960317611694336], [0.2123015820980072, 0.4444444477558136, NaN, 0.460317462682724, 0.5416666865348816, 0.533730149269104, 0.5535714030265808, 0.6091269850730896, 0.6478174328804016, 0.6527777910232544, 0.5972222089767456, 0.7202380895614624, 0.7628968358039856, 0.7658730149269104, 0.7916666865348816, 0.7916666865348816, 0.7986111044883728, 0.72817462682724, 0.7242063283920288, 0.7876983880996704, 0.738095223903656, 0.795634925365448, 0.7966269850730896, 0.8035714030265808, 0.8373016119003296, 0.761904776096344, 0.8333333134651184, 0.85317462682724, 0.8551587462425232, 0.8611111044883728, 0.7202380895614624, 0.8769841194152832, 0.8501983880996704, 0.8303571343421936, 0.8611111044883728, 0.8660714030265808, 0.8392857313156128, 0.8373016119003296, 0.8472222089767456, 0.8591269850730896, 0.8373016119003296, 0.886904776096344, 0.908730149269104, 0.851190447807312, 0.8571428656578064, 0.908730149269104, 0.8670634627342224, 0.8650793433189392, 0.9126983880996704, 0.9047619104385376, 0.920634925365448, 0.9146825671195984, 0.91567462682724, 0.8908730149269104, 0.7777777910232544, 0.7797619104385376, 0.9166666865348816, 0.9226190447807312, 0.9464285969734192, 0.9464285969734192, 0.89682537317276, 0.9265872836112976, 0.8829365372657776, 0.9503968358039856, 0.932539701461792, 0.8253968358039856, 0.9444444179534912, 0.908730149269104, 0.9464285969734192, 0.949404776096344, 0.9444444179534912, 0.932539701461792, 0.954365074634552, 0.9563491940498352, 0.9355158805847168, 0.9583333134651184, 0.9464285969734192, 0.9940476417541504], [0.1805555522441864, 0.4742063581943512, 0.5396825671195984, NaN, 0.538690447807312, 0.5297619104385376, 0.5436508059501648, 0.60317462682724, 0.6617063283920288, 0.6765872836112976, 0.682539701461792, 0.7172619104385376, 0.7242063283920288, 0.778769850730896, 0.761904776096344, 0.77182537317276, 0.846230149269104, 0.7152777910232544, 0.7063491940498352, 0.7698412537574768, 0.7400793433189392, 0.7460317611694336, 0.7757936716079712, 0.8253968358039856, 0.8055555820465088, 0.795634925365448, 0.8273809552192688, 0.817460298538208, 0.8353174328804016, 0.8313491940498352, 0.6984127163887024, 0.8670634627342224, 0.8273809552192688, 0.8144841194152832, 0.8353174328804016, 0.8591269850730896, 0.8333333134651184, 0.8115079402923584, 0.8640872836112976, 0.8432539701461792, 0.8353174328804016, 0.8769841194152832, 0.863095223903656, 0.8392857313156128, 0.8591269850730896, 0.8849206566810608, 0.8809523582458496, 0.8789682388305664, 0.8928571343421936, 0.8720238208770752, 0.908730149269104, 0.9067460298538208, 0.908730149269104, 0.8769841194152832, 0.8035714030265808, 0.8015872836112976, 0.9027777910232544, 0.9146825671195984, 0.9265872836112976, 0.9365079402923584, 0.8948412537574768, 0.9126983880996704, 0.9186508059501648, 0.9345238208770752, 0.9166666865348816, 0.8551587462425232, 0.942460298538208, 0.9246031641960144, 0.9404761791229248, 0.9484127163887024, 0.9503968358039856, 0.942460298538208, 0.9503968358039856, 0.9603174328804016, 0.9513888955116272, 0.9722222089767456, 0.9464285969734192, 0.9960317611694336], [0.2321428507566452, 0.3988095223903656, 0.4583333432674408, 0.4613095223903656, NaN, 0.4404761791229248, 0.48908731341362, 0.5079365372657776, 0.636904776096344, 0.6428571343421936, 0.5714285969734192, 0.6884920597076416, 0.7103174328804016, 0.692460298538208, 0.7599206566810608, 0.7539682388305664, 0.7063491940498352, 0.738095223903656, 0.7579365372657776, 0.699404776096344, 0.7589285969734192, 0.7698412537574768, 0.7996031641960144, 0.7896825671195984, 0.807539701461792, 0.7123016119003296, 0.829365074634552, 0.8472222089767456, 0.8055555820465088, 0.8253968358039856, 0.761904776096344, 0.8194444179534912, 0.8472222089767456, 0.8333333134651184, 0.8551587462425232, 0.8829365372657776, 0.8591269850730896, 0.85317462682724, 0.8392857313156128, 0.875, 0.8670634627342224, 0.886904776096344, 0.8353174328804016, 0.8789682388305664, 0.8611111044883728, 0.903769850730896, 0.898809552192688, 0.8908730149269104, 0.9027777910232544, 0.863095223903656, 0.9166666865348816, 0.9047619104385376, 0.920634925365448, 0.875, 0.8035714030265808, 0.8045634627342224, 0.9126983880996704, 0.886904776096344, 0.9305555820465088, 0.9444444179534912, 0.913690447807312, 0.9345238208770752, 0.9007936716079712, 0.9623016119003296, 0.932539701461792, 0.8373016119003296, 0.9404761791229248, 0.908730149269104, 0.9484127163887024, 0.9464285969734192, 0.9464285969734192, 0.9027777910232544, 0.9523809552192688, 0.954365074634552, 0.9523809552192688, 0.966269850730896, 0.9365079402923584, 0.9940476417541504], [0.2291666716337204, 0.4246031641960144, 0.466269850730896, 0.4702380895614624, 0.5595238208770752, NaN, 0.5753968358039856, 0.5515872836112976, 0.6607142686843872, 0.7142857313156128, 0.5654761791229248, 0.7083333134651184, 0.7361111044883728, 0.6845238208770752, 0.7559523582458496, 0.7757936716079712, 0.7033730149269104, 0.7757936716079712, 0.7539682388305664, 0.6984127163887024, 0.7698412537574768, 0.7678571343421936, 0.783730149269104, 0.7579365372657776, 0.8134920597076416, 0.6686508059501648, 0.8303571343421936, 0.851190447807312, 0.8472222089767456, 0.8333333134651184, 0.7529761791229248, 0.8382936716079712, 0.8452380895614624, 0.8214285969734192, 0.8591269850730896, 0.863095223903656, 0.8353174328804016, 0.8253968358039856, 0.8333333134651184, 0.8720238208770752, 0.83432537317276, 0.8789682388305664, 0.8789682388305664, 0.8482142686843872, 0.841269850730896, 0.89682537317276, 0.863095223903656, 0.875, 0.891865074634552, 0.8591269850730896, 0.9107142686843872, 0.898809552192688, 0.8908730149269104, 0.8611111044883728, 0.7777777910232544, 0.783730149269104, 0.898809552192688, 0.9027777910232544, 0.9404761791229248, 0.9365079402923584, 0.8908730149269104, 0.9146825671195984, 0.898809552192688, 0.9384920597076416, 0.9226190447807312, 0.8154761791229248, 0.942460298538208, 0.9146825671195984, 0.9404761791229248, 0.942460298538208, 0.9404761791229248, 0.9166666865348816, 0.9503968358039856, 0.9484127163887024, 0.925595223903656, 0.9583333134651184, 0.9265872836112976, 0.9920634627342224], [0.22420634329319, 0.4027777910232544, 0.4464285671710968, 0.4563491940498352, 0.5109127163887024, 0.4246031641960144, NaN, 0.5208333134651184, 0.6527777910232544, 0.6716269850730896, 0.5694444179534912, 0.6726190447807312, 0.7103174328804016, 0.6805555820465088, 0.7539682388305664, 0.7638888955116272, 0.6805555820465088, 0.754960298538208, 0.738095223903656, 0.70932537317276, 0.7559523582458496, 0.7638888955116272, 0.788690447807312, 0.7490079402923584, 0.7996031641960144, 0.6865079402923584, 0.829365074634552, 0.8373016119003296, 0.8095238208770752, 0.8144841194152832, 0.7480158805847168, 0.8194444179534912, 0.8333333134651184, 0.8095238208770752, 0.8452380895614624, 0.851190447807312, 0.8472222089767456, 0.829365074634552, 0.8214285969734192, 0.85317462682724, 0.8214285969734192, 0.8789682388305664, 0.870039701461792, 0.851190447807312, 0.8353174328804016, 0.8908730149269104, 0.863095223903656, 0.8650793433189392, 0.8819444179534912, 0.8621031641960144, 0.9107142686843872, 0.8928571343421936, 0.8849206566810608, 0.8690476417541504, 0.7698412537574768, 0.7777777910232544, 0.89682537317276, 0.8928571343421936, 0.9345238208770752, 0.9365079402923584, 0.8809523582458496, 0.908730149269104, 0.8849206566810608, 0.9464285969734192, 0.9186508059501648, 0.8154761791229248, 0.925595223903656, 0.9007936716079712, 0.942460298538208, 0.9365079402923584, 0.932539701461792, 0.9166666865348816, 0.9444444179534912, 0.9464285969734192, 0.9335317611694336, 0.9583333134651184, 0.932539701461792, 0.9920634627342224], [0.1964285671710968, 0.3888888955116272, 0.3908730149269104, 0.3968254029750824, 0.4920634925365448, 0.44841268658638, 0.4791666567325592, NaN, 0.6408730149269104, 0.6190476417541504, 0.5873016119003296, 0.6626983880996704, 0.7182539701461792, 0.6746031641960144, 0.7420634627342224, 0.7440476417541504, 0.7321428656578064, 0.72817462682724, 0.6795634627342224, 0.6865079402923584, 0.72817462682724, 0.7559523582458496, 0.7797619104385376, 0.761904776096344, 0.7996031641960144, 0.7003968358039856, 0.8353174328804016, 0.829365074634552, 0.829365074634552, 0.8095238208770752, 0.738095223903656, 0.8789682388305664, 0.8214285969734192, 0.8015872836112976, 0.8591269850730896, 0.875, 0.8095238208770752, 0.8184523582458496, 0.8452380895614624, 0.8650793433189392, 0.8253968358039856, 0.8769841194152832, 0.908730149269104, 0.8382936716079712, 0.8313491940498352, 0.9047619104385376, 0.8650793433189392, 0.8670634627342224, 0.9107142686843872, 0.8829365372657776, 0.908730149269104, 0.9126983880996704, 0.8611111044883728, 0.8720238208770752, 0.7757936716079712, 0.7857142686843872, 0.9186508059501648, 0.9007936716079712, 0.954365074634552, 0.9444444179534912, 0.8948412537574768, 0.9226190447807312, 0.8849206566810608, 0.9404761791229248, 0.9226190447807312, 0.8154761791229248, 0.932539701461792, 0.908730149269104, 0.9404761791229248, 0.9285714030265808, 0.9365079402923584, 0.932539701461792, 0.9523809552192688, 0.966269850730896, 0.9365079402923584, 0.9563491940498352, 0.942460298538208, 0.9920634627342224], [0.1468254029750824, 0.2936508059501648, 0.352182537317276, 0.3382936418056488, 0.363095223903656, 0.3392857015132904, 0.3472222089767456, 0.3591269850730896, NaN, 0.52182537317276, 0.52182537317276, 0.5128968358039856, 0.591269850730896, 0.591269850730896, 0.670634925365448, 0.6567460298538208, 0.6091269850730896, 0.6140872836112976, 0.625, 0.60317462682724, 0.6686508059501648, 0.6845238208770752, 0.7400793433189392, 0.682539701461792, 0.7321428656578064, 0.6547619104385376, 0.7678571343421936, 0.7440476417541504, 0.7559523582458496, 0.7460317611694336, 0.6408730149269104, 0.7827380895614624, 0.7351190447807312, 0.7529761791229248, 0.7867063283920288, 0.8234127163887024, 0.7996031641960144, 0.7658730149269104, 0.7896825671195984, 0.7817460298538208, 0.7936508059501648, 0.829365074634552, 0.8035714030265808, 0.817460298538208, 0.8273809552192688, 0.8492063283920288, 0.8154761791229248, 0.8392857313156128, 0.8670634627342224, 0.8313491940498352, 0.8888888955116272, 0.8710317611694336, 0.7966269850730896, 0.8422619104385376, 0.7539682388305664, 0.7559523582458496, 0.8888888955116272, 0.8710317611694336, 0.920634925365448, 0.9186508059501648, 0.8591269850730896, 0.898809552192688, 0.8888888955116272, 0.9146825671195984, 0.9027777910232544, 0.7896825671195984, 0.8928571343421936, 0.9067460298538208, 0.9007936716079712, 0.9126983880996704, 0.9166666865348816, 0.9067460298538208, 0.9146825671195984, 0.9404761791229248, 0.942460298538208, 0.9563491940498352, 0.9275793433189392, 0.9900793433189392], [0.1369047611951828, 0.2976190447807312, 0.3472222089767456, 0.32341268658638, 0.3571428656578064, 0.2857142984867096, 0.3283730149269104, 0.380952388048172, 0.4781745970249176, NaN, 0.5039682388305664, 0.4861111044883728, 0.6041666865348816, 0.5773809552192688, 0.6666666865348816, 0.692460298538208, 0.579365074634552, 0.64682537317276, 0.629960298538208, 0.5902777910232544, 0.6795634627342224, 0.6815476417541504, 0.711309552192688, 0.6408730149269104, 0.7232142686843872, 0.6111111044883728, 0.7440476417541504, 0.7440476417541504, 0.7579365372657776, 0.7678571343421936, 0.6160714030265808, 0.7559523582458496, 0.7509920597076416, 0.7589285969734192, 0.788690447807312, 0.7857142686843872, 0.7847222089767456, 0.7559523582458496, 0.7648809552192688, 0.7976190447807312, 0.7708333134651184, 0.7996031641960144, 0.8015872836112976, 0.7976190447807312, 0.7996031641960144, 0.8253968358039856, 0.8115079402923584, 0.7946428656578064, 0.8353174328804016, 0.8263888955116272, 0.8591269850730896, 0.8452380895614624, 0.773809552192688, 0.8392857313156128, 0.7341269850730896, 0.738095223903656, 0.8611111044883728, 0.8809523582458496, 0.9067460298538208, 0.920634925365448, 0.8234127163887024, 0.898809552192688, 0.8650793433189392, 0.9007936716079712, 0.8710317611694336, 0.7857142686843872, 0.8928571343421936, 0.8829365372657776, 0.9166666865348816, 0.9186508059501648, 0.9126983880996704, 0.898809552192688, 0.9404761791229248, 0.9305555820465088, 0.9027777910232544, 0.9484127163887024, 0.9166666865348816, 0.9940476417541504], [0.1686507910490036, 0.3035714328289032, 0.4027777910232544, 0.3174603283405304, 0.4285714328289032, 0.4345238208770752, 0.4305555522441864, 0.4126984179019928, 0.4781745970249176, 0.4960317313671112, NaN, 0.5615079402923584, 0.5833333134651184, 0.6309523582458496, 0.5803571343421936, 0.5892857313156128, 0.60317462682724, 0.5496031641960144, 0.5753968358039856, 0.6527777910232544, 0.567460298538208, 0.6378968358039856, 0.6210317611694336, 0.7291666865348816, 0.658730149269104, 0.788690447807312, 0.6607142686843872, 0.6984127163887024, 0.6904761791229248, 0.6865079402923584, 0.6309523582458496, 0.7480158805847168, 0.6884920597076416, 0.7142857313156128, 0.7142857313156128, 0.7222222089767456, 0.7123016119003296, 0.6865079402923584, 0.7658730149269104, 0.7222222089767456, 0.7698412537574768, 0.7400793433189392, 0.7321428656578064, 0.745039701461792, 0.7559523582458496, 0.8224206566810608, 0.7916666865348816, 0.8273809552192688, 0.8015872836112976, 0.836309552192688, 0.8015872836112976, 0.817460298538208, 0.8611111044883728, 0.8591269850730896, 0.7301587462425232, 0.738095223903656, 0.8313491940498352, 0.875, 0.8670634627342224, 0.8730158805847168, 0.908730149269104, 0.8789682388305664, 0.841269850730896, 0.8849206566810608, 0.863095223903656, 0.7996031641960144, 0.8849206566810608, 0.851190447807312, 0.8888888955116272, 0.9047619104385376, 0.9226190447807312, 0.9186508059501648, 0.9533730149269104, 0.9523809552192688, 0.9305555820465088, 0.954365074634552, 0.942460298538208, 0.9940476417541504], [0.1200396791100502, 0.2817460298538208, 0.2797619104385376, 0.2827380895614624, 0.3115079402923584, 0.2916666567325592, 0.3273809552192688, 0.3373015820980072, 0.4871031641960144, 0.5138888955116272, 0.4384920597076416, NaN, 0.5972222089767456, 0.538690447807312, 0.6269841194152832, 0.6329365372657776, 0.5873016119003296, 0.5873016119003296, 0.5972222089767456, 0.567460298538208, 0.625, 0.641865074634552, 0.6398809552192688, 0.6091269850730896, 0.716269850730896, 0.5853174328804016, 0.726190447807312, 0.738095223903656, 0.738095223903656, 0.7579365372657776, 0.6686508059501648, 0.7966269850730896, 0.7400793433189392, 0.745039701461792, 0.77182537317276, 0.7777777910232544, 0.7678571343421936, 0.7142857313156128, 0.7698412537574768, 0.7896825671195984, 0.7172619104385376, 0.7559523582458496, 0.7777777910232544, 0.795634925365448, 0.7976190447807312, 0.8273809552192688, 0.8095238208770752, 0.79067462682724, 0.8472222089767456, 0.841269850730896, 0.8134920597076416, 0.8551587462425232, 0.8204365372657776, 0.8392857313156128, 0.7648809552192688, 0.7579365372657776, 0.8710317611694336, 0.89682537317276, 0.898809552192688, 0.9126983880996704, 0.8115079402923584, 0.9107142686843872, 0.8611111044883728, 0.8888888955116272, 0.8928571343421936, 0.8105158805847168, 0.9047619104385376, 0.8829365372657776, 0.9345238208770752, 0.9146825671195984, 0.920634925365448, 0.8948412537574768, 0.9464285969734192, 0.954365074634552, 0.8908730149269104, 0.9365079402923584, 0.9146825671195984, 0.9900793433189392], [0.1428571492433548, 0.2757936418056488, 0.2371031790971756, 0.2757936418056488, 0.289682537317276, 0.2638888955116272, 0.289682537317276, 0.2817460298538208, 0.408730149269104, 0.3958333432674408, 0.4166666567325592, 0.4027777910232544, NaN, 0.4871031641960144, 0.5773809552192688, 0.5327380895614624, 0.5019841194152832, 0.4960317313671112, 0.4761904776096344, 0.4950396716594696, 0.5932539701461792, 0.663690447807312, 0.6349206566810608, 0.5297619104385376, 0.7003968358039856, 0.5476190447807312, 0.7311508059501648, 0.7519841194152832, 0.6626983880996704, 0.704365074634552, 0.579365074634552, 0.7400793433189392, 0.7559523582458496, 0.682539701461792, 0.817460298538208, 0.77182537317276, 0.77182537317276, 0.7559523582458496, 0.7103174328804016, 0.8273809552192688, 0.7569444179534912, 0.7321428656578064, 0.7123016119003296, 0.7797619104385376, 0.778769850730896, 0.8115079402923584, 0.846230149269104, 0.7807539701461792, 0.8214285969734192, 0.7976190447807312, 0.7857142686843872, 0.8303571343421936, 0.7440476417541504, 0.8125, 0.726190447807312, 0.7341269850730896, 0.8313491940498352, 0.886904776096344, 0.9146825671195984, 0.9007936716079712, 0.795634925365448, 0.9186508059501648, 0.8472222089767456, 0.9007936716079712, 0.8908730149269104, 0.7797619104385376, 0.9226190447807312, 0.8492063283920288, 0.9146825671195984, 0.9285714030265808, 0.9285714030265808, 0.8948412537574768, 0.9464285969734192, 0.9583333134651184, 0.8710317611694336, 0.9603174328804016, 0.920634925365448, 0.9940476417541504], [0.1130952388048172, 0.2162698358297348, 0.2341269850730896, 0.2212301641702652, 0.3075396716594696, 0.3154761791229248, 0.3194444477558136, 0.3253968358039856, 0.408730149269104, 0.4226190447807312, 0.369047611951828, 0.4613095223903656, 0.5128968358039856, NaN, 0.516865074634552, 0.523809552192688, 0.5188491940498352, 0.5109127163887024, 0.4900793731212616, 0.6071428656578064, 0.52182537317276, 0.5942460298538208, 0.5545634627342224, 0.5982142686843872, 0.5952380895614624, 0.6309523582458496, 0.6269841194152832, 0.625, 0.64682537317276, 0.6398809552192688, 0.5853174328804016, 0.6944444179534912, 0.6329365372657776, 0.653769850730896, 0.6577380895614624, 0.6805555820465088, 0.66567462682724, 0.641865074634552, 0.7013888955116272, 0.682539701461792, 0.6964285969734192, 0.6974206566810608, 0.7301587462425232, 0.6954365372657776, 0.692460298538208, 0.7797619104385376, 0.72817462682724, 0.7519841194152832, 0.7559523582458496, 0.783730149269104, 0.7599206566810608, 0.7698412537574768, 0.8015872836112976, 0.7996031641960144, 0.6984127163887024, 0.7003968358039856, 0.7936508059501648, 0.8571428656578064, 0.8432539701461792, 0.851190447807312, 0.829365074634552, 0.8710317611694336, 0.7857142686843872, 0.8690476417541504, 0.8571428656578064, 0.77182537317276, 0.8670634627342224, 0.807539701461792, 0.8978174328804016, 0.8819444179534912, 0.9007936716079712, 0.9017857313156128, 0.9384920597076416, 0.9305555820465088, 0.8650793433189392, 0.9305555820465088, 0.9126983880996704, 0.9940476417541504], [0.1349206417798996, 0.2420634925365448, 0.2083333283662796, 0.2380952388048172, 0.2400793582201004, 0.244047611951828, 0.2460317462682724, 0.2579365074634552, 0.329365074634552, 0.3333333432674408, 0.4196428656578064, 0.3730158805847168, 0.4226190447807312, 0.483134925365448, NaN, 0.4900793731212616, 0.4801587164402008, 0.4285714328289032, 0.4275793731212616, 0.4900793731212616, 0.591269850730896, 0.5515872836112976, 0.6547619104385376, 0.5396825671195984, 0.6408730149269104, 0.545634925365448, 0.663690447807312, 0.6507936716079712, 0.6220238208770752, 0.6071428656578064, 0.5892857313156128, 0.6517857313156128, 0.6765872836112976, 0.6111111044883728, 0.7460317611694336, 0.75, 0.761904776096344, 0.8224206566810608, 0.6845238208770752, 0.7678571343421936, 0.6944444179534912, 0.7182539701461792, 0.7003968358039856, 0.7757936716079712, 0.7688491940498352, 0.75, 0.788690447807312, 0.75, 0.8095238208770752, 0.7301587462425232, 0.7986111044883728, 0.8353174328804016, 0.716269850730896, 0.726190447807312, 0.726190447807312, 0.7341269850730896, 0.8670634627342224, 0.7896825671195984, 0.8809523582458496, 0.8839285969734192, 0.7976190447807312, 0.841269850730896, 0.8551587462425232, 0.8928571343421936, 0.8809523582458496, 0.783730149269104, 0.8928571343421936, 0.8720238208770752, 0.8591269850730896, 0.9126983880996704, 0.9007936716079712, 0.8730158805847168, 0.9067460298538208, 0.9007936716079712, 0.898809552192688, 0.9563491940498352, 0.9107142686843872, 0.9841269850730896], [0.095238097012043, 0.2579365074634552, 0.2083333283662796, 0.2281745970249176, 0.2460317462682724, 0.22420634329319, 0.2361111044883728, 0.255952388048172, 0.3432539701461792, 0.3075396716594696, 0.4107142984867096, 0.3670634925365448, 0.4672619104385376, 0.4761904776096344, 0.5099206566810608, NaN, 0.4801587164402008, 0.4166666567325592, 0.4583333432674408, 0.4781745970249176, 0.5515872836112976, 0.5595238208770752, 0.5694444179534912, 0.5228174328804016, 0.670634925365448, 0.5476190447807312, 0.6904761791229248, 0.6785714030265808, 0.6111111044883728, 0.6438491940498352, 0.5297619104385376, 0.6805555820465088, 0.7063491940498352, 0.6726190447807312, 0.726190447807312, 0.7440476417541504, 0.7063491940498352, 0.6944444179534912, 0.6904761791229248, 0.7123016119003296, 0.7242063283920288, 0.7519841194152832, 0.6785714030265808, 0.726190447807312, 0.7638888955116272, 0.77182537317276, 0.773809552192688, 0.7817460298538208, 0.7847222089767456, 0.7083333134651184, 0.8055555820465088, 0.795634925365448, 0.70932537317276, 0.7400793433189392, 0.7242063283920288, 0.7321428656578064, 0.8194444179534912, 0.807539701461792, 0.8769841194152832, 0.9007936716079712, 0.8095238208770752, 0.8482142686843872, 0.8422619104385376, 0.8769841194152832, 0.8650793433189392, 0.7817460298538208, 0.9146825671195984, 0.870039701461792, 0.8769841194152832, 0.920634925365448, 0.9166666865348816, 0.875, 0.9246031641960144, 0.9146825671195984, 0.9047619104385376, 0.9384920597076416, 0.9146825671195984, 0.9960317611694336], [0.0674603208899498, 0.1597222238779068, 0.2013888955116272, 0.1537698358297348, 0.2936508059501648, 0.2966269850730896, 0.3194444477558136, 0.2678571343421936, 0.3908730149269104, 0.420634925365448, 0.3968254029750824, 0.4126984179019928, 0.4980158805847168, 0.4811508059501648, 0.5198412537574768, 0.5198412537574768, NaN, 0.4801587164402008, 0.4970238208770752, 0.5069444179534912, 0.5257936716079712, 0.567460298538208, 0.5694444179534912, 0.6210317611694336, 0.6111111044883728, 0.5922619104385376, 0.6289682388305664, 0.6190476417541504, 0.6607142686843872, 0.636904776096344, 0.5833333134651184, 0.6914682388305664, 0.636904776096344, 0.6388888955116272, 0.6746031641960144, 0.7003968358039856, 0.6349206566810608, 0.6666666865348816, 0.6894841194152832, 0.6884920597076416, 0.6785714030265808, 0.6626983880996704, 0.7142857313156128, 0.6666666865348816, 0.6964285969734192, 0.7658730149269104, 0.72817462682724, 0.7361111044883728, 0.783730149269104, 0.7976190447807312, 0.7658730149269104, 0.795634925365448, 0.7936508059501648, 0.8065476417541504, 0.7103174328804016, 0.7123016119003296, 0.8154761791229248, 0.8452380895614624, 0.85317462682724, 0.891865074634552, 0.8234127163887024, 0.8789682388305664, 0.7976190447807312, 0.8571428656578064, 0.8452380895614624, 0.7698412537574768, 0.8928571343421936, 0.817460298538208, 0.8690476417541504, 0.9027777910232544, 0.8839285969734192, 0.9047619104385376, 0.9384920597076416, 0.9384920597076416, 0.903769850730896, 0.9503968358039856, 0.9285714030265808, 0.9920634627342224], [0.1329365074634552, 0.227182537317276, 0.2718254029750824, 0.2847222089767456, 0.261904776096344, 0.22420634329319, 0.2450396865606308, 0.2718254029750824, 0.38591268658638, 0.3531745970249176, 0.4503968358039856, 0.4126984179019928, 0.5039682388305664, 0.48908731341362, 0.5714285969734192, 0.5833333134651184, 0.5198412537574768, NaN, 0.5416666865348816, 0.5248016119003296, 0.6150793433189392, 0.591269850730896, 0.636904776096344, 0.567460298538208, 0.64682537317276, 0.5555555820465088, 0.6626983880996704, 0.6686508059501648, 0.6626983880996704, 0.6507936716079712, 0.5823412537574768, 0.625, 0.7023809552192688, 0.648809552192688, 0.6974206566810608, 0.7142857313156128, 0.7023809552192688, 0.6964285969734192, 0.6954365372657776, 0.6964285969734192, 0.7013888955116272, 0.72817462682724, 0.6964285969734192, 0.7222222089767456, 0.7321428656578064, 0.7509920597076416, 0.7589285969734192, 0.7341269850730896, 0.7658730149269104, 0.72817462682724, 0.7876983880996704, 0.7817460298538208, 0.7003968358039856, 0.77182537317276, 0.716269850730896, 0.7103174328804016, 0.8115079402923584, 0.7698412537574768, 0.795634925365448, 0.829365074634552, 0.7916666865348816, 0.8204365372657776, 0.829365074634552, 0.8333333134651184, 0.8452380895614624, 0.77182537317276, 0.8353174328804016, 0.841269850730896, 0.8472222089767456, 0.8382936716079712, 0.841269850730896, 0.8492063283920288, 0.8670634627342224, 0.8492063283920288, 0.8690476417541504, 0.898809552192688, 0.8769841194152832, 0.9861111044883728], [0.170634925365448, 0.2767857015132904, 0.2757936418056488, 0.2936508059501648, 0.2420634925365448, 0.2460317462682724, 0.261904776096344, 0.3204365074634552, 0.375, 0.3700396716594696, 0.4246031641960144, 0.4027777910232544, 0.523809552192688, 0.5099206566810608, 0.5724206566810608, 0.5416666865348816, 0.5029761791229248, 0.4583333432674408, NaN, 0.54067462682724, 0.5833333134651184, 0.5654761791229248, 0.6111111044883728, 0.5615079402923584, 0.6428571343421936, 0.557539701461792, 0.648809552192688, 0.6527777910232544, 0.6170634627342224, 0.6666666865348816, 0.5595238208770752, 0.6091269850730896, 0.670634925365448, 0.6845238208770752, 0.6765872836112976, 0.6954365372657776, 0.704365074634552, 0.6984127163887024, 0.6795634627342224, 0.6865079402923584, 0.7083333134651184, 0.6865079402923584, 0.6408730149269104, 0.7222222089767456, 0.738095223903656, 0.692460298538208, 0.7599206566810608, 0.7480158805847168, 0.7182539701461792, 0.7460317611694336, 0.7440476417541504, 0.7301587462425232, 0.7480158805847168, 0.761904776096344, 0.7460317611694336, 0.7410714030265808, 0.738095223903656, 0.7916666865348816, 0.7857142686843872, 0.8253968358039856, 0.851190447807312, 0.8373016119003296, 0.8601190447807312, 0.8551587462425232, 0.8273809552192688, 0.7976190447807312, 0.851190447807312, 0.8611111044883728, 0.8492063283920288, 0.851190447807312, 0.8591269850730896, 0.8611111044883728, 0.851190447807312, 0.8650793433189392, 0.9107142686843872, 0.9146825671195984, 0.89682537317276, 0.9801587462425232], [0.1210317462682724, 0.1944444477558136, 0.2123015820980072, 0.230158731341362, 0.300595223903656, 0.30158731341362, 0.2906745970249176, 0.3134920597076416, 0.3968254029750824, 0.4097222089767456, 0.3472222089767456, 0.4325396716594696, 0.504960298538208, 0.3928571343421936, 0.5099206566810608, 0.52182537317276, 0.4930555522441864, 0.4751984179019928, 0.4593254029750824, NaN, 0.5, 0.5496031641960144, 0.5515872836112976, 0.5694444179534912, 0.567460298538208, 0.6140872836112976, 0.591269850730896, 0.5952380895614624, 0.613095223903656, 0.5952380895614624, 0.5555555820465088, 0.6388888955116272, 0.6021825671195984, 0.6051587462425232, 0.6478174328804016, 0.64682537317276, 0.64682537317276, 0.6309523582458496, 0.6726190447807312, 0.6507936716079712, 0.6686508059501648, 0.6845238208770752, 0.6934523582458496, 0.6746031641960144, 0.682539701461792, 0.7301587462425232, 0.7023809552192688, 0.726190447807312, 0.7341269850730896, 0.7361111044883728, 0.726190447807312, 0.7420634627342224, 0.8115079402923584, 0.7559523582458496, 0.6875, 0.6865079402923584, 0.7797619104385376, 0.7996031641960144, 0.7916666865348816, 0.8154761791229248, 0.8253968358039856, 0.8253968358039856, 0.7748016119003296, 0.851190447807312, 0.8313491940498352, 0.7658730149269104, 0.8452380895614624, 0.7896825671195984, 0.8611111044883728, 0.8571428656578064, 0.886904776096344, 0.898809552192688, 0.9047619104385376, 0.908730149269104, 0.8611111044883728, 0.9285714030265808, 0.9027777910232544, 0.9900793433189392], [0.141865074634552, 0.2628968358039856, 0.261904776096344, 0.2599206268787384, 0.2410714328289032, 0.230158731341362, 0.244047611951828, 0.2718254029750824, 0.3313491940498352, 0.3204365074634552, 0.4325396716594696, 0.375, 0.4067460298538208, 0.4781745970249176, 0.408730149269104, 0.44841268658638, 0.4742063581943512, 0.3849206268787384, 0.4166666567325592, 0.5, NaN, 0.4583333432674408, 0.4900793731212616, 0.5277777910232544, 0.601190447807312, 0.5833333134651184, 0.5992063283920288, 0.5873016119003296, 0.596230149269104, 0.5952380895614624, 0.5376983880996704, 0.6051587462425232, 0.6220238208770752, 0.6051587462425232, 0.6607142686843872, 0.6964285969734192, 0.7182539701461792, 0.6686508059501648, 0.6567460298538208, 0.6746031641960144, 0.6517857313156128, 0.682539701461792, 0.6190476417541504, 0.7579365372657776, 0.7817460298538208, 0.670634925365448, 0.7400793433189392, 0.7242063283920288, 0.7509920597076416, 0.726190447807312, 0.7579365372657776, 0.761904776096344, 0.7182539701461792, 0.7420634627342224, 0.7519841194152832, 0.7559523582458496, 0.817460298538208, 0.7976190447807312, 0.807539701461792, 0.8184523582458496, 0.7986111044883728, 0.807539701461792, 0.8789682388305664, 0.8591269850730896, 0.898809552192688, 0.8125, 0.8551587462425232, 0.898809552192688, 0.7896825671195984, 0.8769841194152832, 0.8591269850730896, 0.8789682388305664, 0.85317462682724, 0.8789682388305664, 0.932539701461792, 0.9305555820465088, 0.9484127163887024, 0.9861111044883728], [0.1488095223903656, 0.2361111044883728, 0.204365074634552, 0.2539682686328888, 0.230158731341362, 0.2321428507566452, 0.2361111044883728, 0.244047611951828, 0.3154761791229248, 0.318452388048172, 0.3621031641960144, 0.358134925365448, 0.3363095223903656, 0.4057539701461792, 0.44841268658638, 0.4404761791229248, 0.4325396716594696, 0.408730149269104, 0.4345238208770752, 0.4503968358039856, 0.5416666865348816, NaN, 0.5376983880996704, 0.4841269850730896, 0.5773809552192688, 0.5019841194152832, 0.5952380895614624, 0.613095223903656, 0.5714285969734192, 0.5734127163887024, 0.52182537317276, 0.5922619104385376, 0.6210317611694336, 0.6001983880996704, 0.692460298538208, 0.6746031641960144, 0.6805555820465088, 0.670634925365448, 0.6527777910232544, 0.711309552192688, 0.6646825671195984, 0.6765872836112976, 0.64682537317276, 0.7202380895614624, 0.682539701461792, 0.7142857313156128, 0.7460317611694336, 0.7341269850730896, 0.7083333134651184, 0.7301587462425232, 0.7460317611694336, 0.7341269850730896, 0.6795634627342224, 0.7420634627342224, 0.7103174328804016, 0.716269850730896, 0.7638888955116272, 0.807539701461792, 0.8333333134651184, 0.8373016119003296, 0.778769850730896, 0.8313491940498352, 0.7876983880996704, 0.8670634627342224, 0.8472222089767456, 0.7658730149269104, 0.8690476417541504, 0.7966269850730896, 0.8611111044883728, 0.8710317611694336, 0.8998016119003296, 0.8492063283920288, 0.8650793433189392, 0.8908730149269104, 0.8392857313156128, 0.942460298538208, 0.8809523582458496, 0.9821428656578064], [0.105158731341362, 0.204365074634552, 0.2033730149269104, 0.22420634329319, 0.2003968209028244, 0.2162698358297348, 0.2113095223903656, 0.2202380895614624, 0.2599206268787384, 0.2886904776096344, 0.3789682686328888, 0.3601190447807312, 0.3650793731212616, 0.4454365074634552, 0.3452380895614624, 0.4305555522441864, 0.4305555522441864, 0.363095223903656, 0.3888888955116272, 0.44841268658638, 0.5099206566810608, 0.4623015820980072, NaN, 0.5039682388305664, 0.5625, 0.5228174328804016, 0.5853174328804016, 0.5972222089767456, 0.5932539701461792, 0.545634925365448, 0.511904776096344, 0.60317462682724, 0.6150793433189392, 0.5694444179534912, 0.66567462682724, 0.682539701461792, 0.6666666865348816, 0.7003968358039856, 0.64682537317276, 0.6746031641960144, 0.6458333134651184, 0.6805555820465088, 0.6547619104385376, 0.704365074634552, 0.7182539701461792, 0.6765872836112976, 0.7440476417541504, 0.716269850730896, 0.761904776096344, 0.682539701461792, 0.7490079402923584, 0.7539682388305664, 0.6884920597076416, 0.6904761791229248, 0.7063491940498352, 0.7172619104385376, 0.8144841194152832, 0.773809552192688, 0.8313491940498352, 0.817460298538208, 0.783730149269104, 0.7916666865348816, 0.8392857313156128, 0.875, 0.8650793433189392, 0.7638888955116272, 0.8908730149269104, 0.851190447807312, 0.8154761791229248, 0.898809552192688, 0.8551587462425232, 0.8472222089767456, 0.8591269850730896, 0.875, 0.8809523582458496, 0.9464285969734192, 0.8888888955116272, 0.976190447807312], [0.130952388048172, 0.1567460298538208, 0.1964285671710968, 0.1746031790971756, 0.210317462682724, 0.2420634925365448, 0.2509920597076416, 0.2380952388048172, 0.3174603283405304, 0.3591269850730896, 0.2708333432674408, 0.3908730149269104, 0.4702380895614624, 0.4017857015132904, 0.460317462682724, 0.477182537317276, 0.3789682686328888, 0.4325396716594696, 0.4384920597076416, 0.4305555522441864, 0.4722222089767456, 0.5158730149269104, 0.4960317313671112, NaN, 0.5198412537574768, 0.5158730149269104, 0.5257936716079712, 0.5396825671195984, 0.5694444179534912, 0.5515872836112976, 0.5753968358039856, 0.5892857313156128, 0.5644841194152832, 0.5595238208770752, 0.5972222089767456, 0.6051587462425232, 0.6150793433189392, 0.6230158805847168, 0.6230158805847168, 0.6289682388305664, 0.7063491940498352, 0.6567460298538208, 0.6210317611694336, 0.6428571343421936, 0.648809552192688, 0.704365074634552, 0.6964285969734192, 0.7361111044883728, 0.6626983880996704, 0.6805555820465088, 0.716269850730896, 0.692460298538208, 0.8432539701461792, 0.6944444179534912, 0.7440476417541504, 0.75, 0.7757936716079712, 0.7222222089767456, 0.738095223903656, 0.8105158805847168, 0.8888888955116272, 0.7817460298538208, 0.7678571343421936, 0.841269850730896, 0.8115079402923584, 0.8095238208770752, 0.8710317611694336, 0.7936508059501648, 0.8055555820465088, 0.8710317611694336, 0.886904776096344, 0.863095223903656, 0.8452380895614624, 0.8829365372657776, 0.8998016119003296, 0.9186508059501648, 0.9384920597076416, 0.9940476417541504], [0.0992063507437706, 0.184523805975914, 0.1626984179019928, 0.1944444477558136, 0.1924603134393692, 0.1865079402923584, 0.2003968209028244, 0.2003968209028244, 0.2678571343421936, 0.2767857015132904, 0.341269850730896, 0.283730149269104, 0.2996031641960144, 0.4047619104385376, 0.3591269850730896, 0.329365074634552, 0.3888888955116272, 0.3531745970249176, 0.3571428656578064, 0.4325396716594696, 0.3988095223903656, 0.4226190447807312, 0.4375, 0.4801587164402008, NaN, 0.5257936716079712, 0.5625, 0.5773809552192688, 0.5039682388305664, 0.5476190447807312, 0.5138888955116272, 0.5605158805847168, 0.5882936716079712, 0.5992063283920288, 0.7063491940498352, 0.6914682388305664, 0.6458333134651184, 0.6517857313156128, 0.5972222089767456, 0.7182539701461792, 0.591269850730896, 0.591269850730896, 0.5734127163887024, 0.6865079402923584, 0.7361111044883728, 0.6686508059501648, 0.6884920597076416, 0.6805555820465088, 0.7470238208770752, 0.6894841194152832, 0.6984127163887024, 0.7599206566810608, 0.6964285969734192, 0.6944444179534912, 0.70932537317276, 0.7123016119003296, 0.841269850730896, 0.7976190447807312, 0.8730158805847168, 0.942460298538208, 0.7460317611694336, 0.8571428656578064, 0.8115079402923584, 0.8055555820465088, 0.8730158805847168, 0.7896825671195984, 0.8769841194152832, 0.8333333134651184, 0.8035714030265808, 0.9007936716079712, 0.8373016119003296, 0.8730158805847168, 0.9186508059501648, 0.9285714030265808, 0.89682537317276, 0.9345238208770752, 0.9176587462425232, 0.9920634627342224], [0.1091269850730896, 0.1805555522441864, 0.2380952388048172, 0.204365074634552, 0.2876984179019928, 0.3313491940498352, 0.3134920597076416, 0.2996031641960144, 0.3452380895614624, 0.3888888955116272, 0.2113095223903656, 0.414682537317276, 0.4523809552192688, 0.369047611951828, 0.454365074634552, 0.4523809552192688, 0.4077380895614624, 0.4444444477558136, 0.4424603283405304, 0.38591268658638, 0.4166666567325592, 0.4980158805847168, 0.477182537317276, 0.4841269850730896, 0.4742063581943512, NaN, 0.4960317313671112, 0.5257936716079712, 0.545634925365448, 0.5277777910232544, 0.5317460298538208, 0.5416666865348816, 0.5357142686843872, 0.5317460298538208, 0.5694444179534912, 0.5376983880996704, 0.5297619104385376, 0.5734127163887024, 0.5952380895614624, 0.5853174328804016, 0.5853174328804016, 0.5873016119003296, 0.5932539701461792, 0.5565476417541504, 0.591269850730896, 0.6309523582458496, 0.58432537317276, 0.658730149269104, 0.6507936716079712, 0.6884920597076416, 0.6428571343421936, 0.6527777910232544, 0.7460317611694336, 0.7142857313156128, 0.6329365372657776, 0.6329365372657776, 0.7103174328804016, 0.7123016119003296, 0.6845238208770752, 0.738095223903656, 0.841269850730896, 0.7638888955116272, 0.6944444179534912, 0.773809552192688, 0.7976190447807312, 0.738095223903656, 0.7936508059501648, 0.726190447807312, 0.8035714030265808, 0.817460298538208, 0.8650793433189392, 0.875, 0.8908730149269104, 0.8591269850730896, 0.8452380895614624, 0.8769841194152832, 0.9007936716079712, 0.9920634627342224], [0.1011904776096344, 0.1666666716337204, 0.1666666716337204, 0.1726190447807312, 0.170634925365448, 0.1696428507566452, 0.170634925365448, 0.164682537317276, 0.2321428507566452, 0.255952388048172, 0.3392857015132904, 0.2738095223903656, 0.2688491940498352, 0.3730158805847168, 0.3363095223903656, 0.3095238208770752, 0.3710317313671112, 0.3373015820980072, 0.3511904776096344, 0.408730149269104, 0.4007936418056488, 0.4047619104385376, 0.414682537317276, 0.4742063581943512, 0.4375, 0.5039682388305664, NaN, 0.5446428656578064, 0.466269850730896, 0.5188491940498352, 0.4920634925365448, 0.5158730149269104, 0.5595238208770752, 0.5595238208770752, 0.6240079402923584, 0.66567462682724, 0.625, 0.6170634627342224, 0.5753968358039856, 0.648809552192688, 0.5625, 0.5833333134651184, 0.545634925365448, 0.6686508059501648, 0.7003968358039856, 0.6388888955116272, 0.675595223903656, 0.6388888955116272, 0.6944444179534912, 0.66567462682724, 0.6835317611694336, 0.7103174328804016, 0.6845238208770752, 0.6875, 0.6944444179534912, 0.7003968358039856, 0.8015872836112976, 0.7519841194152832, 0.8442460298538208, 0.9077380895614624, 0.7341269850730896, 0.8214285969734192, 0.8035714030265808, 0.7916666865348816, 0.8561508059501648, 0.773809552192688, 0.8621031641960144, 0.8263888955116272, 0.7966269850730896, 0.8809523582458496, 0.829365074634552, 0.8551587462425232, 0.9047619104385376, 0.9027777910232544, 0.8829365372657776, 0.932539701461792, 0.9067460298538208, 0.9940476417541504], [0.0813492089509964, 0.15079365670681, 0.1468254029750824, 0.1825396865606308, 0.1527777761220932, 0.1488095223903656, 0.1626984179019928, 0.170634925365448, 0.255952388048172, 0.255952388048172, 0.30158731341362, 0.261904776096344, 0.2480158656835556, 0.375, 0.3492063581943512, 0.3214285671710968, 0.380952388048172, 0.3313491940498352, 0.3472222089767456, 0.4047619104385376, 0.4126984179019928, 0.386904776096344, 0.4027777910232544, 0.460317462682724, 0.4226190447807312, 0.4742063581943512, 0.4553571343421936, NaN, 0.4821428656578064, 0.4960317313671112, 0.5, 0.5178571343421936, 0.5763888955116272, 0.5476190447807312, 0.5684523582458496, 0.5932539701461792, 0.5555555820465088, 0.5952380895614624, 0.557539701461792, 0.596230149269104, 0.5535714030265808, 0.5734127163887024, 0.5615079402923584, 0.6121031641960144, 0.6428571343421936, 0.6547619104385376, 0.6626983880996704, 0.6329365372657776, 0.6646825671195984, 0.6944444179534912, 0.6785714030265808, 0.670634925365448, 0.6428571343421936, 0.7172619104385376, 0.6984127163887024, 0.692460298538208, 0.7103174328804016, 0.8015872836112976, 0.85317462682724, 0.8551587462425232, 0.6904761791229248, 0.858134925365448, 0.773809552192688, 0.8015872836112976, 0.8283730149269104, 0.7519841194152832, 0.817460298538208, 0.7916666865348816, 0.8214285969734192, 0.8373016119003296, 0.807539701461792, 0.863095223903656, 0.8908730149269104, 0.9146825671195984, 0.8313491940498352, 0.8908730149269104, 0.89682537317276, 0.9920634627342224], [0.0972222238779068, 0.15079365670681, 0.144841268658638, 0.164682537317276, 0.1944444477558136, 0.1527777761220932, 0.190476194024086, 0.170634925365448, 0.244047611951828, 0.2420634925365448, 0.3095238208770752, 0.261904776096344, 0.3373015820980072, 0.3531745970249176, 0.3779761791229248, 0.3888888955116272, 0.3392857015132904, 0.3373015820980072, 0.3829365074634552, 0.386904776096344, 0.403769850730896, 0.4285714328289032, 0.4067460298538208, 0.4305555522441864, 0.4960317313671112, 0.454365074634552, 0.533730149269104, 0.5178571343421936, NaN, 0.5486111044883728, 0.4841269850730896, 0.5, 0.533730149269104, 0.5972222089767456, 0.5694444179534912, 0.5813491940498352, 0.5833333134651184, 0.5376983880996704, 0.6349206566810608, 0.5763888955116272, 0.5823412537574768, 0.5694444179534912, 0.5446428656578064, 0.625, 0.6349206566810608, 0.6230158805847168, 0.6726190447807312, 0.6527777910232544, 0.6805555820465088, 0.6507936716079712, 0.7142857313156128, 0.6815476417541504, 0.6607142686843872, 0.6765872836112976, 0.670634925365448, 0.6785714030265808, 0.7321428656578064, 0.7480158805847168, 0.7559523582458496, 0.761904776096344, 0.7658730149269104, 0.773809552192688, 0.7440476417541504, 0.795634925365448, 0.8134920597076416, 0.7539682388305664, 0.8095238208770752, 0.773809552192688, 0.8551587462425232, 0.8313491940498352, 0.8204365372657776, 0.8194444179534912, 0.8650793433189392, 0.8472222089767456, 0.8214285969734192, 0.8829365372657776, 0.8670634627342224, 0.9801587462425232], [0.0873015895485878, 0.1468254029750824, 0.1388888955116272, 0.1686507910490036, 0.1746031790971756, 0.1666666716337204, 0.1855158656835556, 0.190476194024086, 0.2539682686328888, 0.2321428507566452, 0.3134920597076416, 0.2420634925365448, 0.295634925365448, 0.3601190447807312, 0.3928571343421936, 0.3561508059501648, 0.363095223903656, 0.3492063581943512, 0.3333333432674408, 0.4047619104385376, 0.4047619104385376, 0.42658731341362, 0.454365074634552, 0.44841268658638, 0.4523809552192688, 0.4722222089767456, 0.4811508059501648, 0.5039682388305664, 0.4513888955116272, NaN, 0.4523809552192688, 0.477182537317276, 0.5228174328804016, 0.596230149269104, 0.5367063283920288, 0.58432537317276, 0.5476190447807312, 0.5436508059501648, 0.5535714030265808, 0.5496031641960144, 0.5565476417541504, 0.545634925365448, 0.5515872836112976, 0.6150793433189392, 0.6388888955116272, 0.6160714030265808, 0.6269841194152832, 0.6388888955116272, 0.682539701461792, 0.6428571343421936, 0.6408730149269104, 0.6884920597076416, 0.6904761791229248, 0.6964285969734192, 0.6676587462425232, 0.6666666865348816, 0.7063491940498352, 0.7579365372657776, 0.761904776096344, 0.7638888955116272, 0.783730149269104, 0.8035714030265808, 0.7916666865348816, 0.8234127163887024, 0.7817460298538208, 0.7440476417541504, 0.8154761791229248, 0.807539701461792, 0.9305555820465088, 0.8482142686843872, 0.8829365372657776, 0.8680555820465088, 0.932539701461792, 0.9365079402923584, 0.846230149269104, 0.9107142686843872, 0.898809552192688, 0.9900793433189392], [0.1626984179019928, 0.2599206268787384, 0.2797619104385376, 0.30158731341362, 0.2380952388048172, 0.247023805975914, 0.2519841194152832, 0.261904776096344, 0.3591269850730896, 0.3839285671710968, 0.369047611951828, 0.3313491940498352, 0.420634925365448, 0.414682537317276, 0.4107142984867096, 0.4702380895614624, 0.4166666567325592, 0.4176587164402008, 0.4404761791229248, 0.4444444477558136, 0.4623015820980072, 0.4781745970249176, 0.488095223903656, 0.4246031641960144, 0.4861111044883728, 0.4682539701461792, 0.5079365372657776, 0.5, 0.5158730149269104, 0.5476190447807312, NaN, 0.5297619104385376, 0.5158730149269104, 0.5535714030265808, 0.5515872836112976, 0.5277777910232544, 0.5595238208770752, 0.5615079402923584, 0.5634920597076416, 0.5595238208770752, 0.6269841194152832, 0.6527777910232544, 0.5734127163887024, 0.5892857313156128, 0.6021825671195984, 0.5882936716079712, 0.6805555820465088, 0.6746031641960144, 0.5654761791229248, 0.6309523582458496, 0.692460298538208, 0.591269850730896, 0.601190447807312, 0.6309523582458496, 0.60317462682724, 0.5952380895614624, 0.6408730149269104, 0.699404776096344, 0.670634925365448, 0.6884920597076416, 0.7123016119003296, 0.711309552192688, 0.7519841194152832, 0.7519841194152832, 0.7202380895614624, 0.7301587462425232, 0.761904776096344, 0.7559523582458496, 0.761904776096344, 0.766865074634552, 0.7817460298538208, 0.716269850730896, 0.7658730149269104, 0.7777777910232544, 0.7817460298538208, 0.7976190447807312, 0.8253968358039856, 0.9821428656578064], [0.0793650820851326, 0.1230158731341362, 0.1230158731341362, 0.1329365074634552, 0.1805555522441864, 0.16170634329319, 0.1805555522441864, 0.1210317462682724, 0.2172619104385376, 0.244047611951828, 0.2519841194152832, 0.2033730149269104, 0.2599206268787384, 0.3055555522441864, 0.3482142984867096, 0.3194444477558136, 0.3085317313671112, 0.375, 0.3908730149269104, 0.3611111044883728, 0.3948412835597992, 0.4077380895614624, 0.3968254029750824, 0.4107142984867096, 0.4394841194152832, 0.4583333432674408, 0.4841269850730896, 0.4821428656578064, 0.5, 0.5228174328804016, 0.4702380895614624, NaN, 0.5138888955116272, 0.5664682388305664, 0.5446428656578064, 0.5515872836112976, 0.5634920597076416, 0.5178571343421936, 0.5992063283920288, 0.5833333134651184, 0.5853174328804016, 0.5496031641960144, 0.5714285969734192, 0.6091269850730896, 0.6289682388305664, 0.7053571343421936, 0.6626983880996704, 0.6696428656578064, 0.6845238208770752, 0.6617063283920288, 0.658730149269104, 0.6805555820465088, 0.6378968358039856, 0.6944444179534912, 0.6428571343421936, 0.6438491940498352, 0.7271825671195984, 0.7698412537574768, 0.8055555820465088, 0.841269850730896, 0.7698412537574768, 0.8115079402923584, 0.704365074634552, 0.7996031641960144, 0.8154761791229248, 0.7480158805847168, 0.829365074634552, 0.7440476417541504, 0.8571428656578064, 0.8551587462425232, 0.8373016119003296, 0.8214285969734192, 0.8948412537574768, 0.8948412537574768, 0.807539701461792, 0.89682537317276, 0.8660714030265808, 0.97817462682724], [0.075396828353405, 0.1408730149269104, 0.1498015820980072, 0.1726190447807312, 0.1527777761220932, 0.1547619104385376, 0.1666666716337204, 0.1785714328289032, 0.2648809552192688, 0.2490079402923584, 0.3115079402923584, 0.2599206268787384, 0.244047611951828, 0.3670634925365448, 0.32341268658638, 0.2936508059501648, 0.363095223903656, 0.2976190447807312, 0.329365074634552, 0.397817462682724, 0.3779761791229248, 0.3789682686328888, 0.3849206268787384, 0.4355158805847168, 0.4117063581943512, 0.4642857015132904, 0.4404761791229248, 0.4236111044883728, 0.466269850730896, 0.477182537317276, 0.4841269850730896, 0.4861111044883728, NaN, 0.533730149269104, 0.567460298538208, 0.5882936716079712, 0.5615079402923584, 0.5763888955116272, 0.5396825671195984, 0.5892857313156128, 0.5317460298538208, 0.5734127163887024, 0.5515872836112976, 0.5972222089767456, 0.6428571343421936, 0.6210317611694336, 0.64682537317276, 0.6240079402923584, 0.6180555820465088, 0.6746031641960144, 0.66567462682724, 0.6458333134651184, 0.6408730149269104, 0.6865079402923584, 0.6984127163887024, 0.7003968358039856, 0.7023809552192688, 0.7916666865348816, 0.8472222089767456, 0.829365074634552, 0.682539701461792, 0.8333333134651184, 0.7678571343421936, 0.7797619104385376, 0.8115079402923584, 0.7579365372657776, 0.8055555820465088, 0.7827380895614624, 0.807539701461792, 0.829365074634552, 0.7976190447807312, 0.85317462682724, 0.8789682388305664, 0.9047619104385376, 0.8234127163887024, 0.898809552192688, 0.9027777910232544, 0.9920634627342224], [0.1071428582072258, 0.1527777761220932, 0.1696428507566452, 0.1855158656835556, 0.1666666716337204, 0.1785714328289032, 0.190476194024086, 0.1984127014875412, 0.247023805975914, 0.2410714328289032, 0.2857142984867096, 0.2549603283405304, 0.3174603283405304, 0.346230149269104, 0.3888888955116272, 0.3273809552192688, 0.3611111044883728, 0.3511904776096344, 0.3154761791229248, 0.3948412835597992, 0.3948412835597992, 0.3998015820980072, 0.4305555522441864, 0.4404761791229248, 0.4007936418056488, 0.4682539701461792, 0.4404761791229248, 0.4523809552192688, 0.4027777910232544, 0.403769850730896, 0.4464285671710968, 0.4335317313671112, 0.466269850730896, NaN, 0.4841269850730896, 0.5079365372657776, 0.5228174328804016, 0.5257936716079712, 0.5357142686843872, 0.5069444179534912, 0.528769850730896, 0.5357142686843872, 0.4623015820980072, 0.5773809552192688, 0.6111111044883728, 0.5853174328804016, 0.5773809552192688, 0.6071428656578064, 0.6438491940498352, 0.636904776096344, 0.625, 0.6666666865348816, 0.6765872836112976, 0.6746031641960144, 0.6805555820465088, 0.6845238208770752, 0.7023809552192688, 0.6805555820465088, 0.6607142686843872, 0.7123016119003296, 0.8015872836112976, 0.7400793433189392, 0.7916666865348816, 0.8194444179534912, 0.795634925365448, 0.75, 0.7827380895614624, 0.8134920597076416, 0.8888888955116272, 0.8303571343421936, 0.8611111044883728, 0.85317462682724, 0.875, 0.8769841194152832, 0.8670634627342224, 0.898809552192688, 0.898809552192688, 0.9900793433189392], [0.0674603208899498, 0.1498015820980072, 0.1388888955116272, 0.164682537317276, 0.144841268658638, 0.1408730149269104, 0.1547619104385376, 0.1408730149269104, 0.21329365670681, 0.2113095223903656, 0.2857142984867096, 0.2281745970249176, 0.1825396865606308, 0.3422619104385376, 0.2539682686328888, 0.2738095223903656, 0.3253968358039856, 0.3025793731212616, 0.32341268658638, 0.352182537317276, 0.3392857015132904, 0.3075396716594696, 0.3343254029750824, 0.4027777910232544, 0.2936508059501648, 0.4305555522441864, 0.3759920597076416, 0.431547611951828, 0.4305555522441864, 0.4632936418056488, 0.44841268658638, 0.4553571343421936, 0.4325396716594696, 0.5158730149269104, NaN, 0.5327380895614624, 0.5158730149269104, 0.5357142686843872, 0.5277777910232544, 0.5426587462425232, 0.52182537317276, 0.538690447807312, 0.5, 0.5853174328804016, 0.6051587462425232, 0.5952380895614624, 0.60317462682724, 0.5773809552192688, 0.5833333134651184, 0.6071428656578064, 0.6190476417541504, 0.6210317611694336, 0.6339285969734192, 0.658730149269104, 0.6726190447807312, 0.6726190447807312, 0.6557539701461792, 0.778769850730896, 0.8730158805847168, 0.8730158805847168, 0.6914682388305664, 0.8085317611694336, 0.7579365372657776, 0.7559523582458496, 0.7698412537574768, 0.7341269850730896, 0.7757936716079712, 0.77182537317276, 0.7936508059501648, 0.7986111044883728, 0.77182537317276, 0.817460298538208, 0.8670634627342224, 0.9027777910232544, 0.8214285969734192, 0.8809523582458496, 0.8779761791229248, 0.9821428656578064], [0.0734127014875412, 0.125, 0.1339285671710968, 0.1408730149269104, 0.1170634925365448, 0.1369047611951828, 0.1488095223903656, 0.125, 0.1765872985124588, 0.2142857164144516, 0.2777777910232544, 0.2222222238779068, 0.2281745970249176, 0.3194444477558136, 0.25, 0.255952388048172, 0.2996031641960144, 0.2857142984867096, 0.3045634925365448, 0.3531745970249176, 0.3035714328289032, 0.3253968358039856, 0.3174603283405304, 0.3948412835597992, 0.3085317313671112, 0.4623015820980072, 0.3343254029750824, 0.4067460298538208, 0.4186508059501648, 0.4156745970249176, 0.4722222089767456, 0.44841268658638, 0.4117063581943512, 0.4920634925365448, 0.4672619104385376, NaN, 0.5198412537574768, 0.5257936716079712, 0.4920634925365448, 0.48908731341362, 0.523809552192688, 0.533730149269104, 0.4980158805847168, 0.591269850730896, 0.625, 0.5932539701461792, 0.6170634627342224, 0.601190447807312, 0.629960298538208, 0.6289682388305664, 0.6408730149269104, 0.658730149269104, 0.6230158805847168, 0.648809552192688, 0.6914682388305664, 0.6944444179534912, 0.7301587462425232, 0.7460317611694336, 0.7658730149269104, 0.8035714030265808, 0.704365074634552, 0.7876983880996704, 0.7420634627342224, 0.7876983880996704, 0.863095223903656, 0.761904776096344, 0.7916666865348816, 0.7658730149269104, 0.7658730149269104, 0.8333333134651184, 0.8134920597076416, 0.8432539701461792, 0.8611111044883728, 0.8690476417541504, 0.8472222089767456, 0.8829365372657776, 0.908730149269104, 0.988095223903656], [0.105158731341362, 0.1607142835855484, 0.1607142835855484, 0.1666666716337204, 0.1408730149269104, 0.164682537317276, 0.1527777761220932, 0.190476194024086, 0.2003968209028244, 0.2152777761220932, 0.2876984179019928, 0.2321428507566452, 0.2281745970249176, 0.3343254029750824, 0.2380952388048172, 0.2936508059501648, 0.3650793731212616, 0.2976190447807312, 0.295634925365448, 0.3531745970249176, 0.2817460298538208, 0.3194444477558136, 0.3333333432674408, 0.3849206268787384, 0.3541666567325592, 0.4702380895614624, 0.375, 0.4444444477558136, 0.4166666567325592, 0.4523809552192688, 0.4404761791229248, 0.4365079402923584, 0.4384920597076416, 0.477182537317276, 0.4841269850730896, 0.4801587164402008, NaN, 0.5178571343421936, 0.511904776096344, 0.4920634925365448, 0.4910714328289032, 0.5317460298538208, 0.4761904776096344, 0.6795634627342224, 0.7003968358039856, 0.545634925365448, 0.5892857313156128, 0.5644841194152832, 0.579365074634552, 0.6398809552192688, 0.6121031641960144, 0.5992063283920288, 0.6547619104385376, 0.6388888955116272, 0.6746031641960144, 0.6765872836112976, 0.670634925365448, 0.7083333134651184, 0.7073412537574768, 0.7142857313156128, 0.738095223903656, 0.726190447807312, 0.7996031641960144, 0.7420634627342224, 0.807539701461792, 0.7599206566810608, 0.7857142686843872, 0.817460298538208, 0.7539682388305664, 0.8085317611694336, 0.8194444179534912, 0.829365074634552, 0.8333333134651184, 0.8353174328804016, 0.851190447807312, 0.89682537317276, 0.8938491940498352, 0.9801587462425232], [0.0992063507437706, 0.1805555522441864, 0.1626984179019928, 0.1884920597076416, 0.1468254029750824, 0.1746031790971756, 0.170634925365448, 0.181547611951828, 0.2341269850730896, 0.244047611951828, 0.3134920597076416, 0.2857142984867096, 0.244047611951828, 0.358134925365448, 0.1775793582201004, 0.3055555522441864, 0.3333333432674408, 0.3035714328289032, 0.30158731341362, 0.369047611951828, 0.3313491940498352, 0.329365074634552, 0.2996031641960144, 0.3769841194152832, 0.3482142984867096, 0.42658731341362, 0.3829365074634552, 0.4047619104385376, 0.4623015820980072, 0.4563491940498352, 0.4384920597076416, 0.4821428656578064, 0.4236111044883728, 0.4742063581943512, 0.4642857015132904, 0.4742063581943512, 0.4821428656578064, NaN, 0.5178571343421936, 0.4920634925365448, 0.5535714030265808, 0.5654761791229248, 0.5376983880996704, 0.545634925365448, 0.5466269850730896, 0.5644841194152832, 0.6111111044883728, 0.6289682388305664, 0.557539701461792, 0.6111111044883728, 0.648809552192688, 0.579365074634552, 0.6071428656578064, 0.6349206566810608, 0.692460298538208, 0.7003968358039856, 0.6349206566810608, 0.7103174328804016, 0.6666666865348816, 0.6547619104385376, 0.6944444179534912, 0.7242063283920288, 0.7638888955116272, 0.7817460298538208, 0.77182537317276, 0.7698412537574768, 0.7569444179534912, 0.7847222089767456, 0.7142857313156128, 0.7638888955116272, 0.7916666865348816, 0.8055555820465088, 0.8095238208770752, 0.829365074634552, 0.829365074634552, 0.8591269850730896, 0.8928571343421936, 0.9722222089767456], [0.095238097012043, 0.1359127014875412, 0.1527777761220932, 0.1359127014875412, 0.1607142835855484, 0.1666666716337204, 0.1785714328289032, 0.1547619104385376, 0.210317462682724, 0.2351190447807312, 0.2341269850730896, 0.230158731341362, 0.289682537317276, 0.2986111044883728, 0.3154761791229248, 0.3095238208770752, 0.3105158805847168, 0.3045634925365448, 0.3204365074634552, 0.3273809552192688, 0.3432539701461792, 0.3472222089767456, 0.3531745970249176, 0.3769841194152832, 0.4027777910232544, 0.4047619104385376, 0.4246031641960144, 0.4424603283405304, 0.3650793731212616, 0.4464285671710968, 0.4365079402923584, 0.4007936418056488, 0.460317462682724, 0.4642857015132904, 0.4722222089767456, 0.5079365372657776, 0.488095223903656, 0.4821428656578064, NaN, 0.5059523582458496, 0.4781745970249176, 0.4781745970249176, 0.4067460298538208, 0.545634925365448, 0.5853174328804016, 0.5228174328804016, 0.5396825671195984, 0.5436508059501648, 0.6567460298538208, 0.5952380895614624, 0.5773809552192688, 0.6646825671195984, 0.6527777910232544, 0.6349206566810608, 0.6507936716079712, 0.6567460298538208, 0.7003968358039856, 0.6746031641960144, 0.6329365372657776, 0.7003968358039856, 0.7440476417541504, 0.7400793433189392, 0.7440476417541504, 0.783730149269104, 0.783730149269104, 0.7341269850730896, 0.773809552192688, 0.7688491940498352, 0.7321428656578064, 0.8055555820465088, 0.8095238208770752, 0.8472222089767456, 0.8184523582458496, 0.8154761791229248, 0.8591269850730896, 0.8571428656578064, 0.8789682388305664, 0.9841269850730896], [0.0674603208899498, 0.1369047611951828, 0.1408730149269104, 0.1567460298538208, 0.125, 0.127976194024086, 0.1468254029750824, 0.1349206417798996, 0.2182539701461792, 0.2023809552192688, 0.2777777910232544, 0.210317462682724, 0.1726190447807312, 0.3174603283405304, 0.2321428507566452, 0.2876984179019928, 0.3115079402923584, 0.3035714328289032, 0.3134920597076416, 0.3492063581943512, 0.3253968358039856, 0.2886904776096344, 0.3253968358039856, 0.3710317313671112, 0.2817460298538208, 0.414682537317276, 0.3511904776096344, 0.403769850730896, 0.4236111044883728, 0.4503968358039856, 0.4404761791229248, 0.4166666567325592, 0.4107142984867096, 0.4930555522441864, 0.4573412835597992, 0.5109127163887024, 0.5079365372657776, 0.5079365372657776, 0.494047611951828, NaN, 0.494047611951828, 0.52182537317276, 0.4980158805847168, 0.557539701461792, 0.5773809552192688, 0.579365074634552, 0.5753968358039856, 0.5773809552192688, 0.5486111044883728, 0.6140872836112976, 0.5952380895614624, 0.5952380895614624, 0.613095223903656, 0.6547619104385376, 0.6567460298538208, 0.6597222089767456, 0.6190476417541504, 0.7579365372657776, 0.8353174328804016, 0.8373016119003296, 0.6805555820465088, 0.8055555820465088, 0.7440476417541504, 0.7321428656578064, 0.7658730149269104, 0.7301587462425232, 0.7480158805847168, 0.7698412537574768, 0.7857142686843872, 0.7896825671195984, 0.7678571343421936, 0.8184523582458496, 0.8690476417541504, 0.886904776096344, 0.8134920597076416, 0.8789682388305664, 0.875, 0.97817462682724], [0.1031746044754982, 0.1696428507566452, 0.1626984179019928, 0.164682537317276, 0.1329365074634552, 0.1656745970249176, 0.1785714328289032, 0.1746031790971756, 0.2063492089509964, 0.2291666716337204, 0.230158731341362, 0.2827380895614624, 0.2430555522441864, 0.3035714328289032, 0.3055555522441864, 0.2757936418056488, 0.3214285671710968, 0.2986111044883728, 0.2916666567325592, 0.3313491940498352, 0.3482142984867096, 0.335317462682724, 0.3541666567325592, 0.2936508059501648, 0.408730149269104, 0.414682537317276, 0.4375, 0.4464285671710968, 0.4176587164402008, 0.443452388048172, 0.3730158805847168, 0.414682537317276, 0.4682539701461792, 0.471230149269104, 0.4781745970249176, 0.4761904776096344, 0.5089285969734192, 0.4464285671710968, 0.52182537317276, 0.5059523582458496, NaN, 0.4365079402923584, 0.5079365372657776, 0.5476190447807312, 0.5555555820465088, 0.511904776096344, 0.5783730149269104, 0.6845238208770752, 0.6051587462425232, 0.6051587462425232, 0.5753968358039856, 0.6230158805847168, 0.5615079402923584, 0.6111111044883728, 0.6388888955116272, 0.6448412537574768, 0.6269841194152832, 0.7073412537574768, 0.6646825671195984, 0.6736111044883728, 0.7539682388305664, 0.716269850730896, 0.75, 0.7748016119003296, 0.7420634627342224, 0.7440476417541504, 0.7440476417541504, 0.7817460298538208, 0.7460317611694336, 0.7757936716079712, 0.9484127163887024, 0.7916666865348816, 0.807539701461792, 0.7817460298538208, 0.8373016119003296, 0.8650793433189392, 0.8789682388305664, 0.976190447807312], [0.0555555559694767, 0.1031746044754982, 0.1130952388048172, 0.1230158731341362, 0.1130952388048172, 0.1210317462682724, 0.1210317462682724, 0.1230158731341362, 0.170634925365448, 0.2003968209028244, 0.2599206268787384, 0.244047611951828, 0.2678571343421936, 0.3025793731212616, 0.2817460298538208, 0.2480158656835556, 0.3373015820980072, 0.2718254029750824, 0.3134920597076416, 0.3154761791229248, 0.3174603283405304, 0.32341268658638, 0.3194444477558136, 0.3432539701461792, 0.408730149269104, 0.4126984179019928, 0.4166666567325592, 0.42658731341362, 0.4305555522441864, 0.454365074634552, 0.3472222089767456, 0.4503968358039856, 0.42658731341362, 0.4642857015132904, 0.4613095223903656, 0.466269850730896, 0.4682539701461792, 0.4345238208770752, 0.52182537317276, 0.4781745970249176, 0.5634920597076416, NaN, 0.523809552192688, 0.5178571343421936, 0.5515872836112976, 0.5625, 0.5892857313156128, 0.6617063283920288, 0.5555555820465088, 0.6190476417541504, 0.738095223903656, 0.5873016119003296, 0.5654761791229248, 0.6230158805847168, 0.6527777910232544, 0.6617063283920288, 0.6071428656578064, 0.6964285969734192, 0.6944444179534912, 0.6805555820465088, 0.738095223903656, 0.716269850730896, 0.7430555820465088, 0.85317462682724, 0.778769850730896, 0.75, 0.7757936716079712, 0.783730149269104, 0.7638888955116272, 0.8194444179534912, 0.846230149269104, 0.7857142686843872, 0.7698412537574768, 0.7678571343421936, 0.8432539701461792, 0.8809523582458496, 0.8730158805847168, 0.976190447807312], [0.0555555559694767, 0.1130952388048172, 0.0912698432803154, 0.1369047611951828, 0.164682537317276, 0.1210317462682724, 0.1299603134393692, 0.0912698432803154, 0.1964285671710968, 0.1984127014875412, 0.2678571343421936, 0.2222222238779068, 0.2876984179019928, 0.2698412835597992, 0.2996031641960144, 0.3214285671710968, 0.2857142984867096, 0.3035714328289032, 0.3591269850730896, 0.306547611951828, 0.380952388048172, 0.3531745970249176, 0.3452380895614624, 0.3789682686328888, 0.42658731341362, 0.4067460298538208, 0.454365074634552, 0.4384920597076416, 0.4553571343421936, 0.44841268658638, 0.42658731341362, 0.4285714328289032, 0.44841268658638, 0.5376983880996704, 0.5, 0.5019841194152832, 0.523809552192688, 0.4623015820980072, 0.5932539701461792, 0.5019841194152832, 0.4920634925365448, 0.4761904776096344, NaN, 0.5664682388305664, 0.5634920597076416, 0.5515872836112976, 0.6091269850730896, 0.5714285969734192, 0.567460298538208, 0.5813491940498352, 0.5853174328804016, 0.591269850730896, 0.5595238208770752, 0.6230158805847168, 0.591269850730896, 0.5932539701461792, 0.5952380895614624, 0.6646825671195984, 0.7182539701461792, 0.6865079402923584, 0.6626983880996704, 0.6726190447807312, 0.6428571343421936, 0.6984127163887024, 0.6984127163887024, 0.6547619104385376, 0.7083333134651184, 0.658730149269104, 0.8065476417541504, 0.7103174328804016, 0.7242063283920288, 0.7361111044883728, 0.795634925365448, 0.7817460298538208, 0.7023809552192688, 0.7896825671195984, 0.7698412537574768, 0.9404761791229248], [0.0932539701461792, 0.1468254029750824, 0.1488095223903656, 0.1607142835855484, 0.1210317462682724, 0.1517857164144516, 0.1488095223903656, 0.16170634329319, 0.1825396865606308, 0.2023809552192688, 0.2549603283405304, 0.204365074634552, 0.2202380895614624, 0.3045634925365448, 0.22420634329319, 0.2738095223903656, 0.3333333432674408, 0.2777777910232544, 0.2777777910232544, 0.3253968358039856, 0.2420634925365448, 0.2797619104385376, 0.295634925365448, 0.3571428656578064, 0.3134920597076416, 0.443452388048172, 0.3313491940498352, 0.3878968358039856, 0.375, 0.3849206268787384, 0.4107142984867096, 0.3908730149269104, 0.4027777910232544, 0.4226190447807312, 0.414682537317276, 0.408730149269104, 0.3204365074634552, 0.454365074634552, 0.454365074634552, 0.4424603283405304, 0.4523809552192688, 0.4821428656578064, 0.4335317313671112, NaN, 0.6240079402923584, 0.4920634925365448, 0.5158730149269104, 0.5367063283920288, 0.5396825671195984, 0.5853174328804016, 0.579365074634552, 0.5565476417541504, 0.6121031641960144, 0.6190476417541504, 0.658730149269104, 0.66567462682724, 0.6269841194152832, 0.670634925365448, 0.6646825671195984, 0.6726190447807312, 0.711309552192688, 0.7103174328804016, 0.7658730149269104, 0.716269850730896, 0.7638888955116272, 0.7400793433189392, 0.761904776096344, 0.7857142686843872, 0.72817462682724, 0.783730149269104, 0.7876983880996704, 0.8015872836112976, 0.8273809552192688, 0.817460298538208, 0.8253968358039856, 0.875, 0.8948412537574768, 0.9801587462425232], [0.0992063507437706, 0.1339285671710968, 0.1428571492433548, 0.1408730149269104, 0.1388888955116272, 0.1587301641702652, 0.164682537317276, 0.1686507910490036, 0.1726190447807312, 0.2003968209028244, 0.244047611951828, 0.2023809552192688, 0.2212301641702652, 0.3075396716594696, 0.2311507910490036, 0.2361111044883728, 0.3035714328289032, 0.2678571343421936, 0.261904776096344, 0.3174603283405304, 0.2182539701461792, 0.3174603283405304, 0.2817460298538208, 0.3511904776096344, 0.2638888955116272, 0.408730149269104, 0.2996031641960144, 0.3571428656578064, 0.3650793731212616, 0.3611111044883728, 0.397817462682724, 0.3710317313671112, 0.3571428656578064, 0.3888888955116272, 0.3948412835597992, 0.375, 0.2996031641960144, 0.4533730149269104, 0.414682537317276, 0.4226190447807312, 0.4444444477558136, 0.44841268658638, 0.4365079402923584, 0.3759920597076416, NaN, 0.44841268658638, 0.4980158805847168, 0.5198412537574768, 0.533730149269104, 0.5634920597076416, 0.5813491940498352, 0.5496031641960144, 0.6091269850730896, 0.5714285969734192, 0.6607142686843872, 0.658730149269104, 0.5932539701461792, 0.6567460298538208, 0.60317462682724, 0.6349206566810608, 0.7182539701461792, 0.682539701461792, 0.7480158805847168, 0.7083333134651184, 0.7638888955116272, 0.7361111044883728, 0.7400793433189392, 0.7827380895614624, 0.692460298538208, 0.7599206566810608, 0.795634925365448, 0.8313491940498352, 0.8134920597076416, 0.795634925365448, 0.8571428656578064, 0.851190447807312, 0.9126983880996704, 0.97817462682724], [0.065476194024086, 0.0843253955245018, 0.0912698432803154, 0.115079365670681, 0.0962301567196846, 0.1031746044754982, 0.1091269850730896, 0.095238097012043, 0.15079365670681, 0.1746031790971756, 0.1775793582201004, 0.1726190447807312, 0.1884920597076416, 0.2202380895614624, 0.25, 0.2281745970249176, 0.2341269850730896, 0.2490079402923584, 0.3075396716594696, 0.2698412835597992, 0.329365074634552, 0.2857142984867096, 0.32341268658638, 0.295634925365448, 0.3313491940498352, 0.369047611951828, 0.3611111044883728, 0.3452380895614624, 0.3769841194152832, 0.3839285671710968, 0.4117063581943512, 0.2946428656578064, 0.3789682686328888, 0.414682537317276, 0.4047619104385376, 0.4067460298538208, 0.454365074634552, 0.4355158805847168, 0.477182537317276, 0.420634925365448, 0.488095223903656, 0.4375, 0.44841268658638, 0.5079365372657776, 0.5515872836112976, NaN, 0.545634925365448, 0.5416666865348816, 0.5198412537574768, 0.5396825671195984, 0.5634920597076416, 0.5644841194152832, 0.533730149269104, 0.5873016119003296, 0.629960298538208, 0.6230158805847168, 0.5873016119003296, 0.6527777910232544, 0.636904776096344, 0.64682537317276, 0.6676587462425232, 0.7023809552192688, 0.6319444179534912, 0.7638888955116272, 0.7430555820465088, 0.7063491940498352, 0.704365074634552, 0.6746031641960144, 0.7539682388305664, 0.7539682388305664, 0.7857142686843872, 0.7440476417541504, 0.783730149269104, 0.7857142686843872, 0.7559523582458496, 0.8273809552192688, 0.8184523582458496, 0.9642857313156128], [0.0813492089509964, 0.0972222238779068, 0.1329365074634552, 0.1190476194024086, 0.1011904776096344, 0.1369047611951828, 0.1369047611951828, 0.1349206417798996, 0.184523805975914, 0.1884920597076416, 0.2083333283662796, 0.190476194024086, 0.1537698358297348, 0.2718254029750824, 0.2113095223903656, 0.2261904776096344, 0.2718254029750824, 0.2410714328289032, 0.2400793582201004, 0.2976190447807312, 0.2599206268787384, 0.2539682686328888, 0.255952388048172, 0.3035714328289032, 0.3115079402923584, 0.4156745970249176, 0.324404776096344, 0.3373015820980072, 0.3273809552192688, 0.3730158805847168, 0.3194444477558136, 0.3373015820980072, 0.3531745970249176, 0.4226190447807312, 0.3968254029750824, 0.3829365074634552, 0.4107142984867096, 0.3888888955116272, 0.460317462682724, 0.4246031641960144, 0.4216269850730896, 0.4107142984867096, 0.3908730149269104, 0.4841269850730896, 0.5019841194152832, 0.454365074634552, NaN, 0.523809552192688, 0.5178571343421936, 0.545634925365448, 0.5555555820465088, 0.5357142686843872, 0.5515872836112976, 0.5873016119003296, 0.6349206566810608, 0.6507936716079712, 0.5753968358039856, 0.64682537317276, 0.6150793433189392, 0.6448412537574768, 0.7351190447807312, 0.6805555820465088, 0.7103174328804016, 0.7490079402923584, 0.7341269850730896, 0.7242063283920288, 0.7460317611694336, 0.7519841194152832, 0.704365074634552, 0.7976190447807312, 0.8472222089767456, 0.7976190447807312, 0.7876983880996704, 0.783730149269104, 0.8333333134651184, 0.875, 0.8670634627342224, 0.97817462682724], [0.0773809552192688, 0.1160714253783226, 0.1349206417798996, 0.1210317462682724, 0.1091269850730896, 0.125, 0.1349206417798996, 0.1329365074634552, 0.1607142835855484, 0.2053571492433548, 0.1726190447807312, 0.2093254029750824, 0.2192460298538208, 0.2480158656835556, 0.25, 0.2182539701461792, 0.2638888955116272, 0.2658730149269104, 0.2519841194152832, 0.2738095223903656, 0.2757936418056488, 0.2658730149269104, 0.283730149269104, 0.2638888955116272, 0.3194444477558136, 0.341269850730896, 0.3611111044883728, 0.3670634925365448, 0.3472222089767456, 0.3611111044883728, 0.3253968358039856, 0.3303571343421936, 0.3759920597076416, 0.3928571343421936, 0.4226190447807312, 0.3988095223903656, 0.4355158805847168, 0.3710317313671112, 0.4563491940498352, 0.4226190447807312, 0.3154761791229248, 0.3382936418056488, 0.4285714328289032, 0.4632936418056488, 0.4801587164402008, 0.4583333432674408, 0.4761904776096344, NaN, 0.5367063283920288, 0.545634925365448, 0.4900793731212616, 0.5585317611694336, 0.5, 0.5615079402923584, 0.5952380895614624, 0.5932539701461792, 0.5892857313156128, 0.6498016119003296, 0.591269850730896, 0.6071428656578064, 0.6845238208770752, 0.6547619104385376, 0.6984127163887024, 0.7013888955116272, 0.692460298538208, 0.7063491940498352, 0.6765872836112976, 0.7251983880996704, 0.6746031641960144, 0.7202380895614624, 0.8611111044883728, 0.7638888955116272, 0.7579365372657776, 0.7321428656578064, 0.7996031641960144, 0.8015872836112976, 0.8492063283920288, 0.9682539701461792], [0.0714285746216774, 0.0912698432803154, 0.0873015895485878, 0.1071428582072258, 0.0972222238779068, 0.1081349179148674, 0.1180555522441864, 0.0892857164144516, 0.1329365074634552, 0.164682537317276, 0.1984127014875412, 0.1527777761220932, 0.1785714328289032, 0.244047611951828, 0.190476194024086, 0.2152777761220932, 0.2162698358297348, 0.2341269850730896, 0.2817460298538208, 0.2658730149269104, 0.2490079402923584, 0.2916666567325592, 0.2380952388048172, 0.3373015820980072, 0.2529761791229248, 0.3492063581943512, 0.3055555522441864, 0.335317462682724, 0.3194444477558136, 0.3174603283405304, 0.4345238208770752, 0.3154761791229248, 0.3819444477558136, 0.3561508059501648, 0.4166666567325592, 0.3700396716594696, 0.420634925365448, 0.4424603283405304, 0.3432539701461792, 0.4513888955116272, 0.3948412835597992, 0.4444444477558136, 0.4325396716594696, 0.460317462682724, 0.466269850730896, 0.4801587164402008, 0.4821428656578064, 0.4632936418056488, NaN, 0.4980158805847168, 0.5277777910232544, 0.5664682388305664, 0.5714285969734192, 0.545634925365448, 0.6408730149269104, 0.6408730149269104, 0.7559523582458496, 0.6051587462425232, 0.6388888955116272, 0.7053571343421936, 0.6984127163887024, 0.658730149269104, 0.6547619104385376, 0.6626983880996704, 0.7301587462425232, 0.7301587462425232, 0.726190447807312, 0.6785714030265808, 0.7242063283920288, 0.7698412537574768, 0.7519841194152832, 0.7301587462425232, 0.77182537317276, 0.7539682388305664, 0.7916666865348816, 0.8253968358039856, 0.8392857313156128, 0.9751983880996704], [0.0615079365670681, 0.075396828353405, 0.095238097012043, 0.127976194024086, 0.1369047611951828, 0.1408730149269104, 0.1378968209028244, 0.1170634925365448, 0.1686507910490036, 0.1736111044883728, 0.1636904776096344, 0.1587301641702652, 0.2023809552192688, 0.2162698358297348, 0.2698412835597992, 0.2916666567325592, 0.2023809552192688, 0.2718254029750824, 0.2539682686328888, 0.2638888955116272, 0.2738095223903656, 0.2698412835597992, 0.3174603283405304, 0.3194444477558136, 0.3105158805847168, 0.3115079402923584, 0.3343254029750824, 0.3055555522441864, 0.3492063581943512, 0.3571428656578064, 0.369047611951828, 0.3382936418056488, 0.3253968358039856, 0.363095223903656, 0.3928571343421936, 0.3710317313671112, 0.3601190447807312, 0.3888888955116272, 0.4047619104385376, 0.38591268658638, 0.3948412835597992, 0.380952388048172, 0.4186508059501648, 0.414682537317276, 0.4365079402923584, 0.460317462682724, 0.454365074634552, 0.454365074634552, 0.5019841194152832, NaN, 0.4573412835597992, 0.523809552192688, 0.5615079402923584, 0.6567460298538208, 0.5297619104385376, 0.5476190447807312, 0.533730149269104, 0.716269850730896, 0.5634920597076416, 0.5932539701461792, 0.6269841194152832, 0.7757936716079712, 0.6319444179534912, 0.6150793433189392, 0.6190476417541504, 0.6775793433189392, 0.6547619104385376, 0.6458333134651184, 0.7063491940498352, 0.6726190447807312, 0.6865079402923584, 0.8452380895614624, 0.8928571343421936, 0.851190447807312, 0.7440476417541504, 0.7440476417541504, 0.8809523582458496, 0.9900793433189392], [0.0476190485060215, 0.0833333358168602, 0.0793650820851326, 0.0912698432803154, 0.0833333358168602, 0.0892857164144516, 0.0892857164144516, 0.0912698432803154, 0.1111111119389534, 0.1408730149269104, 0.1984127014875412, 0.1865079402923584, 0.2142857164144516, 0.2400793582201004, 0.2013888955116272, 0.1944444477558136, 0.2341269850730896, 0.2123015820980072, 0.255952388048172, 0.2738095223903656, 0.2420634925365448, 0.2539682686328888, 0.2509920597076416, 0.283730149269104, 0.30158731341362, 0.3571428656578064, 0.3164682686328888, 0.3214285671710968, 0.2857142984867096, 0.3591269850730896, 0.3075396716594696, 0.341269850730896, 0.3343254029750824, 0.375, 0.380952388048172, 0.3591269850730896, 0.3878968358039856, 0.3511904776096344, 0.4226190447807312, 0.4047619104385376, 0.4246031641960144, 0.261904776096344, 0.414682537317276, 0.420634925365448, 0.4186508059501648, 0.4365079402923584, 0.4444444477558136, 0.5099206566810608, 0.4722222089767456, 0.5426587462425232, NaN, 0.5, 0.4960317313671112, 0.5476190447807312, 0.636904776096344, 0.6388888955116272, 0.5396825671195984, 0.6190476417541504, 0.5773809552192688, 0.5803571343421936, 0.7142857313156128, 0.64682537317276, 0.648809552192688, 0.7628968358039856, 0.7023809552192688, 0.7202380895614624, 0.6805555820465088, 0.6775793433189392, 0.6944444179534912, 0.7242063283920288, 0.7757936716079712, 0.7390872836112976, 0.726190447807312, 0.7242063283920288, 0.8035714030265808, 0.8134920597076416, 0.817460298538208, 0.966269850730896], [0.0615079365670681, 0.0912698432803154, 0.085317462682724, 0.0932539701461792, 0.095238097012043, 0.1011904776096344, 0.1071428582072258, 0.0873015895485878, 0.1289682537317276, 0.1547619104385376, 0.1825396865606308, 0.144841268658638, 0.1696428507566452, 0.230158731341362, 0.164682537317276, 0.204365074634552, 0.204365074634552, 0.2182539701461792, 0.2698412835597992, 0.2579365074634552, 0.2380952388048172, 0.2658730149269104, 0.2460317462682724, 0.3075396716594696, 0.2400793582201004, 0.3472222089767456, 0.289682537317276, 0.329365074634552, 0.318452388048172, 0.3115079402923584, 0.408730149269104, 0.3194444477558136, 0.3541666567325592, 0.3333333432674408, 0.3789682686328888, 0.341269850730896, 0.4007936418056488, 0.420634925365448, 0.335317462682724, 0.4047619104385376, 0.3769841194152832, 0.4126984179019928, 0.408730149269104, 0.443452388048172, 0.4503968358039856, 0.4355158805847168, 0.4642857015132904, 0.4414682686328888, 0.4335317313671112, 0.4761904776096344, 0.5, NaN, 0.5615079402923584, 0.5317460298538208, 0.6269841194152832, 0.6269841194152832, 0.692460298538208, 0.5813491940498352, 0.6220238208770752, 0.670634925365448, 0.6646825671195984, 0.6329365372657776, 0.636904776096344, 0.6547619104385376, 0.6984127163887024, 0.716269850730896, 0.6815476417541504, 0.648809552192688, 0.6964285969734192, 0.7232142686843872, 0.7341269850730896, 0.7083333134651184, 0.7599206566810608, 0.7301587462425232, 0.77182537317276, 0.8144841194152832, 0.8134920597076416, 0.9642857313156128], [0.0694444477558136, 0.0873015895485878, 0.0843253955245018, 0.0912698432803154, 0.0793650820851326, 0.1091269850730896, 0.115079365670681, 0.1388888955116272, 0.2033730149269104, 0.2261904776096344, 0.1388888955116272, 0.1795634925365448, 0.255952388048172, 0.1984127014875412, 0.283730149269104, 0.2906745970249176, 0.2063492089509964, 0.2996031641960144, 0.2519841194152832, 0.1884920597076416, 0.2817460298538208, 0.3204365074634552, 0.3115079402923584, 0.1567460298538208, 0.3035714328289032, 0.2539682686328888, 0.3154761791229248, 0.3571428656578064, 0.3392857015132904, 0.3095238208770752, 0.3988095223903656, 0.3621031641960144, 0.3591269850730896, 0.32341268658638, 0.3660714328289032, 0.3769841194152832, 0.3452380895614624, 0.3928571343421936, 0.3472222089767456, 0.386904776096344, 0.4384920597076416, 0.4345238208770752, 0.4404761791229248, 0.3878968358039856, 0.3908730149269104, 0.466269850730896, 0.44841268658638, 0.5, 0.4285714328289032, 0.4384920597076416, 0.5039682388305664, 0.4384920597076416, NaN, 0.4503968358039856, 0.5535714030265808, 0.5555555820465088, 0.5059523582458496, 0.5, 0.4841269850730896, 0.5555555820465088, 0.6607142686843872, 0.5396825671195984, 0.648809552192688, 0.6111111044883728, 0.5972222089767456, 0.7083333134651184, 0.6746031641960144, 0.675595223903656, 0.648809552192688, 0.716269850730896, 0.7361111044883728, 0.6765872836112976, 0.6686508059501648, 0.6805555820465088, 0.7896825671195984, 0.8055555820465088, 0.817460298538208, 0.976190447807312], [0.0714285746216774, 0.0773809552192688, 0.1091269850730896, 0.1230158731341362, 0.125, 0.1388888955116272, 0.130952388048172, 0.127976194024086, 0.1577380895614624, 0.1607142835855484, 0.1408730149269104, 0.1607142835855484, 0.1875, 0.2003968209028244, 0.2738095223903656, 0.2599206268787384, 0.193452388048172, 0.2281745970249176, 0.2380952388048172, 0.244047611951828, 0.2579365074634552, 0.2579365074634552, 0.3095238208770752, 0.3055555522441864, 0.3055555522441864, 0.2857142984867096, 0.3125, 0.2827380895614624, 0.32341268658638, 0.3035714328289032, 0.369047611951828, 0.3055555522441864, 0.3134920597076416, 0.3253968358039856, 0.341269850730896, 0.3511904776096344, 0.3611111044883728, 0.3650793731212616, 0.3650793731212616, 0.3452380895614624, 0.3888888955116272, 0.3769841194152832, 0.3769841194152832, 0.380952388048172, 0.4285714328289032, 0.4126984179019928, 0.4126984179019928, 0.4384920597076416, 0.454365074634552, 0.3432539701461792, 0.4523809552192688, 0.4682539701461792, 0.5496031641960144, NaN, 0.5327380895614624, 0.5476190447807312, 0.523809552192688, 0.5992063283920288, 0.5, 0.5496031641960144, 0.6230158805847168, 0.6815476417541504, 0.6111111044883728, 0.5952380895614624, 0.6111111044883728, 0.6746031641960144, 0.6309523582458496, 0.6210317611694336, 0.6567460298538208, 0.6527777910232544, 0.6765872836112976, 0.807539701461792, 0.807539701461792, 0.8025793433189392, 0.7341269850730896, 0.7222222089767456, 0.8849206566810608, 0.988095223903656], [0.1746031790971756, 0.1875, 0.2222222238779068, 0.1964285671710968, 0.1964285671710968, 0.2222222238779068, 0.230158731341362, 0.22420634329319, 0.2460317462682724, 0.2658730149269104, 0.2698412835597992, 0.2351190447807312, 0.2738095223903656, 0.30158731341362, 0.2738095223903656, 0.2757936418056488, 0.289682537317276, 0.283730149269104, 0.2539682686328888, 0.3125, 0.2480158656835556, 0.289682537317276, 0.2936508059501648, 0.255952388048172, 0.2906745970249176, 0.3670634925365448, 0.3055555522441864, 0.30158731341362, 0.329365074634552, 0.3323412835597992, 0.3968254029750824, 0.3571428656578064, 0.30158731341362, 0.3194444477558136, 0.3273809552192688, 0.3085317313671112, 0.3253968358039856, 0.3075396716594696, 0.3492063581943512, 0.3432539701461792, 0.3611111044883728, 0.3472222089767456, 0.408730149269104, 0.341269850730896, 0.3392857015132904, 0.3700396716594696, 0.3650793731212616, 0.4047619104385376, 0.3591269850730896, 0.4702380895614624, 0.363095223903656, 0.3730158805847168, 0.4464285671710968, 0.4672619104385376, NaN, 0.5803571343421936, 0.4027777910232544, 0.511904776096344, 0.4444444477558136, 0.454365074634552, 0.52182537317276, 0.5496031641960144, 0.6071428656578064, 0.4722222089767456, 0.4325396716594696, 0.9384920597076416, 0.5892857313156128, 0.6408730149269104, 0.5654761791229248, 0.6071428656578064, 0.6527777910232544, 0.7083333134651184, 0.6865079402923584, 0.7023809552192688, 0.75, 0.6408730149269104, 0.8492063283920288, 0.954365074634552], [0.1746031790971756, 0.1825396865606308, 0.2202380895614624, 0.1984127014875412, 0.1954365074634552, 0.2162698358297348, 0.2222222238779068, 0.2142857164144516, 0.244047611951828, 0.261904776096344, 0.261904776096344, 0.2420634925365448, 0.2658730149269104, 0.2996031641960144, 0.2658730149269104, 0.2678571343421936, 0.2876984179019928, 0.289682537317276, 0.2589285671710968, 0.3134920597076416, 0.244047611951828, 0.283730149269104, 0.2827380895614624, 0.25, 0.2876984179019928, 0.3670634925365448, 0.2996031641960144, 0.3075396716594696, 0.3214285671710968, 0.3333333432674408, 0.4047619104385376, 0.3561508059501648, 0.2996031641960144, 0.3154761791229248, 0.3273809552192688, 0.3055555522441864, 0.32341268658638, 0.2996031641960144, 0.3432539701461792, 0.3402777910232544, 0.3551587164402008, 0.3382936418056488, 0.4067460298538208, 0.3343254029750824, 0.341269850730896, 0.3769841194152832, 0.3492063581943512, 0.4067460298538208, 0.3591269850730896, 0.4523809552192688, 0.3611111044883728, 0.3730158805847168, 0.4444444477558136, 0.4523809552192688, 0.4196428656578064, NaN, 0.3948412835597992, 0.5059523582458496, 0.4404761791229248, 0.4623015820980072, 0.511904776096344, 0.545634925365448, 0.5952380895614624, 0.4682539701461792, 0.4384920597076416, 0.954365074634552, 0.5833333134651184, 0.636904776096344, 0.5615079402923584, 0.5992063283920288, 0.648809552192688, 0.6944444179534912, 0.6785714030265808, 0.6984127163887024, 0.7420634627342224, 0.6388888955116272, 0.8392857313156128, 0.954365074634552], [0.0535714291036129, 0.0912698432803154, 0.0833333358168602, 0.0972222238779068, 0.0873015895485878, 0.1011904776096344, 0.1031746044754982, 0.0813492089509964, 0.1111111119389534, 0.1388888955116272, 0.1686507910490036, 0.1289682537317276, 0.1686507910490036, 0.2063492089509964, 0.1329365074634552, 0.1805555522441864, 0.184523805975914, 0.1884920597076416, 0.261904776096344, 0.2202380895614624, 0.1825396865606308, 0.2361111044883728, 0.1855158656835556, 0.22420634329319, 0.1587301641702652, 0.289682537317276, 0.1984127014875412, 0.289682537317276, 0.2678571343421936, 0.2936508059501648, 0.3591269850730896, 0.272817462682724, 0.2976190447807312, 0.2976190447807312, 0.3442460298538208, 0.2698412835597992, 0.329365074634552, 0.3650793731212616, 0.2996031641960144, 0.380952388048172, 0.3730158805847168, 0.3928571343421936, 0.4047619104385376, 0.3730158805847168, 0.4067460298538208, 0.4126984179019928, 0.4246031641960144, 0.4107142984867096, 0.244047611951828, 0.466269850730896, 0.460317462682724, 0.3075396716594696, 0.494047611951828, 0.4761904776096344, 0.5972222089767456, 0.6051587462425232, NaN, 0.5476190447807312, 0.550595223903656, 0.58432537317276, 0.5773809552192688, 0.5753968358039856, 0.6210317611694336, 0.5932539701461792, 0.6567460298538208, 0.6785714030265808, 0.6349206566810608, 0.636904776096344, 0.6408730149269104, 0.6666666865348816, 0.6626983880996704, 0.6438491940498352, 0.6150793433189392, 0.6785714030265808, 0.7658730149269104, 0.7539682388305664, 0.8253968358039856, 0.9702380895614624], [0.0436507947742939, 0.0793650820851326, 0.0773809552192688, 0.085317462682724, 0.1130952388048172, 0.0972222238779068, 0.1071428582072258, 0.0992063507437706, 0.1289682537317276, 0.1190476194024086, 0.125, 0.1031746044754982, 0.1130952388048172, 0.1428571492433548, 0.210317462682724, 0.1924603134393692, 0.1547619104385376, 0.230158731341362, 0.2083333283662796, 0.2003968209028244, 0.2023809552192688, 0.1924603134393692, 0.2261904776096344, 0.2777777910232544, 0.2023809552192688, 0.2876984179019928, 0.2480158656835556, 0.1984127014875412, 0.2519841194152832, 0.2420634925365448, 0.300595223903656, 0.230158731341362, 0.2083333283662796, 0.3194444477558136, 0.2212301641702652, 0.2539682686328888, 0.2916666567325592, 0.289682537317276, 0.3253968358039856, 0.2420634925365448, 0.2926587164402008, 0.3035714328289032, 0.335317462682724, 0.329365074634552, 0.3432539701461792, 0.3472222089767456, 0.3531745970249176, 0.3501984179019928, 0.3948412835597992, 0.283730149269104, 0.380952388048172, 0.4186508059501648, 0.5, 0.4007936418056488, 0.488095223903656, 0.494047611951828, 0.4523809552192688, NaN, 0.4821428656578064, 0.5059523582458496, 0.5535714030265808, 0.66567462682724, 0.5753968358039856, 0.5476190447807312, 0.5079365372657776, 0.6240079402923584, 0.579365074634552, 0.5813491940498352, 0.6726190447807312, 0.6111111044883728, 0.613095223903656, 0.7321428656578064, 0.7757936716079712, 0.8432539701461792, 0.6934523582458496, 0.6746031641960144, 0.8373016119003296, 0.9682539701461792], [0.0416666679084301, 0.0416666679084301, 0.0535714291036129, 0.0734127014875412, 0.0694444477558136, 0.0595238097012043, 0.065476194024086, 0.0456349216401577, 0.0793650820851326, 0.0932539701461792, 0.1329365074634552, 0.1011904776096344, 0.085317462682724, 0.1567460298538208, 0.1190476194024086, 0.1230158731341362, 0.1468254029750824, 0.204365074634552, 0.2142857164144516, 0.2083333283662796, 0.1924603134393692, 0.1666666716337204, 0.1686507910490036, 0.261904776096344, 0.1269841343164444, 0.3154761791229248, 0.1557539701461792, 0.1468254029750824, 0.244047611951828, 0.2380952388048172, 0.329365074634552, 0.1944444477558136, 0.1527777761220932, 0.3392857015132904, 0.1269841343164444, 0.2341269850730896, 0.2926587164402008, 0.3333333432674408, 0.3670634925365448, 0.164682537317276, 0.335317462682724, 0.3055555522441864, 0.2817460298538208, 0.335317462682724, 0.3968254029750824, 0.363095223903656, 0.3849206268787384, 0.408730149269104, 0.3611111044883728, 0.4365079402923584, 0.4226190447807312, 0.3779761791229248, 0.5158730149269104, 0.5, 0.5555555820465088, 0.5595238208770752, 0.449404776096344, 0.5178571343421936, NaN, 0.5565476417541504, 0.601190447807312, 0.5873016119003296, 0.5932539701461792, 0.5932539701461792, 0.6507936716079712, 0.6646825671195984, 0.5714285969734192, 0.6150793433189392, 0.6785714030265808, 0.6210317611694336, 0.6666666865348816, 0.716269850730896, 0.761904776096344, 0.7757936716079712, 0.7361111044883728, 0.7222222089767456, 0.7916666865348816, 0.9464285969734192], [0.0456349216401577, 0.0595238097012043, 0.0535714291036129, 0.0634920671582222, 0.0555555559694767, 0.0634920671582222, 0.0634920671582222, 0.0555555559694767, 0.0813492089509964, 0.0793650820851326, 0.1269841343164444, 0.0873015895485878, 0.0992063507437706, 0.1488095223903656, 0.1160714253783226, 0.0992063507437706, 0.1081349179148674, 0.170634925365448, 0.1746031790971756, 0.184523805975914, 0.181547611951828, 0.1626984179019928, 0.1825396865606308, 0.1894841343164444, 0.0575396828353405, 0.261904776096344, 0.092261902987957, 0.144841268658638, 0.2380952388048172, 0.2361111044883728, 0.3115079402923584, 0.1587301641702652, 0.170634925365448, 0.2876984179019928, 0.1269841343164444, 0.1964285671710968, 0.2857142984867096, 0.3452380895614624, 0.2996031641960144, 0.1626984179019928, 0.3263888955116272, 0.3194444477558136, 0.3134920597076416, 0.3273809552192688, 0.3650793731212616, 0.3531745970249176, 0.3551587164402008, 0.3928571343421936, 0.2946428656578064, 0.4067460298538208, 0.4196428656578064, 0.329365074634552, 0.4444444477558136, 0.4503968358039856, 0.545634925365448, 0.5376983880996704, 0.4156745970249176, 0.494047611951828, 0.443452388048172, NaN, 0.5714285969734192, 0.538690447807312, 0.5714285969734192, 0.5833333134651184, 0.6150793433189392, 0.6408730149269104, 0.5654761791229248, 0.5932539701461792, 0.636904776096344, 0.6230158805847168, 0.64682537317276, 0.704365074634552, 0.6845238208770752, 0.7033730149269104, 0.7341269850730896, 0.7321428656578064, 0.7996031641960144, 0.9632936716079712], [0.0734127014875412, 0.0833333358168602, 0.1031746044754982, 0.105158731341362, 0.0863095223903656, 0.1091269850730896, 0.1190476194024086, 0.105158731341362, 0.1408730149269104, 0.1765872985124588, 0.0912698432803154, 0.1884920597076416, 0.204365074634552, 0.170634925365448, 0.2023809552192688, 0.190476194024086, 0.1765872985124588, 0.2083333283662796, 0.1488095223903656, 0.1746031790971756, 0.2013888955116272, 0.2212301641702652, 0.2162698358297348, 0.1111111119389534, 0.2539682686328888, 0.1587301641702652, 0.2658730149269104, 0.3095238208770752, 0.2341269850730896, 0.2162698358297348, 0.2876984179019928, 0.230158731341362, 0.3174603283405304, 0.1984127014875412, 0.3085317313671112, 0.295634925365448, 0.261904776096344, 0.3055555522441864, 0.255952388048172, 0.3194444477558136, 0.2460317462682724, 0.261904776096344, 0.3373015820980072, 0.2886904776096344, 0.2817460298538208, 0.3323412835597992, 0.2648809552192688, 0.3154761791229248, 0.30158731341362, 0.3730158805847168, 0.2857142984867096, 0.335317462682724, 0.3392857015132904, 0.3769841194152832, 0.4781745970249176, 0.488095223903656, 0.4226190447807312, 0.4464285671710968, 0.3988095223903656, 0.4285714328289032, NaN, 0.4642857015132904, 0.5376983880996704, 0.4652777910232544, 0.4801587164402008, 0.6269841194152832, 0.5277777910232544, 0.5684523582458496, 0.5277777910232544, 0.5853174328804016, 0.6289682388305664, 0.6071428656578064, 0.5555555820465088, 0.5992063283920288, 0.6884920597076416, 0.6349206566810608, 0.7539682388305664, 0.942460298538208], [0.0456349216401577, 0.0575396828353405, 0.0734127014875412, 0.0873015895485878, 0.065476194024086, 0.085317462682724, 0.0912698432803154, 0.0773809552192688, 0.1011904776096344, 0.1011904776096344, 0.1210317462682724, 0.0892857164144516, 0.0813492089509964, 0.1289682537317276, 0.1587301641702652, 0.1517857164144516, 0.1210317462682724, 0.1795634925365448, 0.1626984179019928, 0.1746031790971756, 0.1924603134393692, 0.1686507910490036, 0.2083333283662796, 0.2182539701461792, 0.1428571492433548, 0.2361111044883728, 0.1785714328289032, 0.141865074634552, 0.2261904776096344, 0.1964285671710968, 0.2886904776096344, 0.1884920597076416, 0.1666666716337204, 0.2599206268787384, 0.1914682537317276, 0.2123015820980072, 0.2738095223903656, 0.2757936418056488, 0.2599206268787384, 0.1944444477558136, 0.283730149269104, 0.283730149269104, 0.3273809552192688, 0.289682537317276, 0.3174603283405304, 0.2976190447807312, 0.3194444477558136, 0.3452380895614624, 0.341269850730896, 0.22420634329319, 0.3531745970249176, 0.3670634925365448, 0.460317462682724, 0.318452388048172, 0.4503968358039856, 0.454365074634552, 0.4246031641960144, 0.3343254029750824, 0.4126984179019928, 0.4613095223903656, 0.5357142686843872, NaN, 0.557539701461792, 0.5, 0.4781745970249176, 0.5972222089767456, 0.5555555820465088, 0.5654761791229248, 0.6230158805847168, 0.5952380895614624, 0.6091269850730896, 0.699404776096344, 0.704365074634552, 0.7291666865348816, 0.6845238208770752, 0.6547619104385376, 0.8253968358039856, 0.9702380895614624], [0.0634920671582222, 0.095238097012043, 0.1170634925365448, 0.0813492089509964, 0.0992063507437706, 0.1011904776096344, 0.115079365670681, 0.115079365670681, 0.1111111119389534, 0.1349206417798996, 0.1587301641702652, 0.1388888955116272, 0.1527777761220932, 0.2142857164144516, 0.144841268658638, 0.1577380895614624, 0.2023809552192688, 0.170634925365448, 0.1398809552192688, 0.2251984179019928, 0.1210317462682724, 0.2123015820980072, 0.1607142835855484, 0.2321428507566452, 0.1884920597076416, 0.3055555522441864, 0.1964285671710968, 0.2261904776096344, 0.255952388048172, 0.2083333283662796, 0.2480158656835556, 0.295634925365448, 0.2321428507566452, 0.2083333283662796, 0.2420634925365448, 0.2579365074634552, 0.2003968209028244, 0.2361111044883728, 0.255952388048172, 0.255952388048172, 0.25, 0.2569444477558136, 0.3571428656578064, 0.2341269850730896, 0.2519841194152832, 0.3680555522441864, 0.289682537317276, 0.30158731341362, 0.3452380895614624, 0.3680555522441864, 0.3511904776096344, 0.363095223903656, 0.3511904776096344, 0.3888888955116272, 0.3928571343421936, 0.4047619104385376, 0.3789682686328888, 0.4246031641960144, 0.4067460298538208, 0.4285714328289032, 0.4623015820980072, 0.4424603283405304, NaN, 0.4345238208770752, 0.4384920597076416, 0.5496031641960144, 0.4781745970249176, 0.6696428656578064, 0.4960317313671112, 0.4980158805847168, 0.488095223903656, 0.591269850730896, 0.5555555820465088, 0.5396825671195984, 0.7857142686843872, 0.5714285969734192, 0.716269850730896, 0.9047619104385376], [0.0396825410425663, 0.0496031753718853, 0.0496031753718853, 0.065476194024086, 0.0376984141767025, 0.0615079365670681, 0.0535714291036129, 0.0595238097012043, 0.085317462682724, 0.0992063507437706, 0.115079365670681, 0.1111111119389534, 0.0992063507437706, 0.130952388048172, 0.1071428582072258, 0.1230158731341362, 0.1428571492433548, 0.1666666716337204, 0.144841268658638, 0.1488095223903656, 0.1408730149269104, 0.1329365074634552, 0.125, 0.1587301641702652, 0.1944444477558136, 0.2261904776096344, 0.2083333283662796, 0.1984127014875412, 0.204365074634552, 0.1765872985124588, 0.2480158656835556, 0.2003968209028244, 0.2202380895614624, 0.1805555522441864, 0.244047611951828, 0.2123015820980072, 0.2579365074634552, 0.2182539701461792, 0.2162698358297348, 0.2678571343421936, 0.2251984179019928, 0.1468254029750824, 0.30158731341362, 0.283730149269104, 0.2916666567325592, 0.2361111044883728, 0.2509920597076416, 0.2986111044883728, 0.3373015820980072, 0.3849206268787384, 0.2371031790971756, 0.3452380895614624, 0.3888888955116272, 0.4047619104385376, 0.5277777910232544, 0.5317460298538208, 0.4067460298538208, 0.4523809552192688, 0.4067460298538208, 0.4166666567325592, 0.5347222089767456, 0.5, 0.5654761791229248, NaN, 0.4821428656578064, 0.64682537317276, 0.5317460298538208, 0.5932539701461792, 0.5734127163887024, 0.5813491940498352, 0.6686508059501648, 0.6626983880996704, 0.6269841194152832, 0.6150793433189392, 0.7083333134651184, 0.6726190447807312, 0.7876983880996704, 0.9484127163887024], [0.0694444477558136, 0.0615079365670681, 0.0674603208899498, 0.0833333358168602, 0.0674603208899498, 0.0773809552192688, 0.0813492089509964, 0.0773809552192688, 0.0972222238779068, 0.1289682537317276, 0.1369047611951828, 0.1071428582072258, 0.1091269850730896, 0.1428571492433548, 0.1190476194024086, 0.1349206417798996, 0.1547619104385376, 0.1547619104385376, 0.1726190447807312, 0.1686507910490036, 0.1011904776096344, 0.1527777761220932, 0.1349206417798996, 0.1884920597076416, 0.1269841343164444, 0.2023809552192688, 0.1438492089509964, 0.1716269850730896, 0.1865079402923584, 0.2182539701461792, 0.2797619104385376, 0.184523805975914, 0.1884920597076416, 0.204365074634552, 0.230158731341362, 0.1369047611951828, 0.1924603134393692, 0.2281745970249176, 0.2162698358297348, 0.2341269850730896, 0.2579365074634552, 0.2212301641702652, 0.30158731341362, 0.2361111044883728, 0.2361111044883728, 0.2569444477558136, 0.2658730149269104, 0.3075396716594696, 0.2698412835597992, 0.380952388048172, 0.2976190447807312, 0.30158731341362, 0.4027777910232544, 0.3888888955116272, 0.567460298538208, 0.5615079402923584, 0.3432539701461792, 0.4920634925365448, 0.3492063581943512, 0.3849206268787384, 0.5198412537574768, 0.52182537317276, 0.5615079402923584, 0.5178571343421936, NaN, 0.6349206566810608, 0.523809552192688, 0.5833333134651184, 0.5555555820465088, 0.545634925365448, 0.5813491940498352, 0.670634925365448, 0.5932539701461792, 0.658730149269104, 0.6865079402923584, 0.6309523582458496, 0.7916666865348816, 0.9246031641960144], [0.1289682537317276, 0.1468254029750824, 0.1746031790971756, 0.144841268658638, 0.1626984179019928, 0.184523805975914, 0.184523805975914, 0.184523805975914, 0.210317462682724, 0.2142857164144516, 0.2003968209028244, 0.1894841343164444, 0.2202380895614624, 0.2281745970249176, 0.2162698358297348, 0.2182539701461792, 0.230158731341362, 0.2281745970249176, 0.2023809552192688, 0.2341269850730896, 0.1875, 0.2341269850730896, 0.2361111044883728, 0.190476194024086, 0.210317462682724, 0.261904776096344, 0.2261904776096344, 0.2480158656835556, 0.2460317462682724, 0.255952388048172, 0.2698412835597992, 0.2519841194152832, 0.2420634925365448, 0.25, 0.2658730149269104, 0.2380952388048172, 0.2400793582201004, 0.230158731341362, 0.2658730149269104, 0.2698412835597992, 0.255952388048172, 0.25, 0.3452380895614624, 0.2599206268787384, 0.2638888955116272, 0.2936508059501648, 0.2757936418056488, 0.2936508059501648, 0.2698412835597992, 0.3224206268787384, 0.2797619104385376, 0.283730149269104, 0.2916666567325592, 0.3253968358039856, 0.0615079365670681, 0.0456349216401577, 0.3214285671710968, 0.3759920597076416, 0.335317462682724, 0.3591269850730896, 0.3730158805847168, 0.4027777910232544, 0.4503968358039856, 0.3531745970249176, 0.3650793731212616, NaN, 0.5019841194152832, 0.466269850730896, 0.454365074634552, 0.5138888955116272, 0.4285714328289032, 0.5654761791229248, 0.523809552192688, 0.4801587164402008, 0.6111111044883728, 0.5515872836112976, 0.7063491940498352, 0.942460298538208], [0.01785714365541935, 0.0634920671582222, 0.0555555559694767, 0.0575396828353405, 0.0595238097012043, 0.0575396828353405, 0.0744047611951828, 0.0674603208899498, 0.1071428582072258, 0.1071428582072258, 0.115079365670681, 0.095238097012043, 0.0773809552192688, 0.1329365074634552, 0.1071428582072258, 0.085317462682724, 0.1071428582072258, 0.164682537317276, 0.1488095223903656, 0.1547619104385376, 0.144841268658638, 0.130952388048172, 0.1091269850730896, 0.1289682537317276, 0.1230158731341362, 0.2063492089509964, 0.1378968209028244, 0.1825396865606308, 0.190476194024086, 0.184523805975914, 0.2380952388048172, 0.170634925365448, 0.1944444477558136, 0.2172619104385376, 0.22420634329319, 0.2083333283662796, 0.2142857164144516, 0.2430555522441864, 0.2261904776096344, 0.2519841194152832, 0.255952388048172, 0.22420634329319, 0.2916666567325592, 0.2380952388048172, 0.2599206268787384, 0.295634925365448, 0.2539682686328888, 0.32341268658638, 0.2738095223903656, 0.3452380895614624, 0.3194444477558136, 0.318452388048172, 0.3253968358039856, 0.369047611951828, 0.4107142984867096, 0.4166666567325592, 0.3650793731212616, 0.420634925365448, 0.4285714328289032, 0.4345238208770752, 0.4722222089767456, 0.4444444477558136, 0.52182537317276, 0.4682539701461792, 0.4761904776096344, 0.4980158805847168, NaN, 0.5396825671195984, 0.5317460298538208, 0.6875, 0.52182537317276, 0.5585317611694336, 0.5892857313156128, 0.5873016119003296, 0.6210317611694336, 0.7757936716079712, 0.6944444179534912, 0.9404761791229248], [0.0555555559694767, 0.0873015895485878, 0.0912698432803154, 0.075396828353405, 0.0912698432803154, 0.085317462682724, 0.0992063507437706, 0.0912698432803154, 0.0932539701461792, 0.1170634925365448, 0.1488095223903656, 0.1170634925365448, 0.15079365670681, 0.1924603134393692, 0.127976194024086, 0.1299603134393692, 0.1825396865606308, 0.1587301641702652, 0.1388888955116272, 0.210317462682724, 0.1011904776096344, 0.2033730149269104, 0.1488095223903656, 0.2063492089509964, 0.1666666716337204, 0.2738095223903656, 0.1736111044883728, 0.2083333283662796, 0.2261904776096344, 0.1924603134393692, 0.244047611951828, 0.255952388048172, 0.2172619104385376, 0.1865079402923584, 0.2281745970249176, 0.2341269850730896, 0.1825396865606308, 0.2152777761220932, 0.2311507910490036, 0.230158731341362, 0.2182539701461792, 0.2162698358297348, 0.341269850730896, 0.2142857164144516, 0.2172619104385376, 0.3253968358039856, 0.2480158656835556, 0.2748015820980072, 0.3214285671710968, 0.3541666567325592, 0.3224206268787384, 0.3511904776096344, 0.324404776096344, 0.3789682686328888, 0.3591269850730896, 0.363095223903656, 0.363095223903656, 0.4186508059501648, 0.3849206268787384, 0.4067460298538208, 0.431547611951828, 0.4345238208770752, 0.3303571343421936, 0.4067460298538208, 0.4166666567325592, 0.533730149269104, 0.460317462682724, NaN, 0.4920634925365448, 0.4841269850730896, 0.4702380895614624, 0.5565476417541504, 0.5396825671195984, 0.511904776096344, 0.6984127163887024, 0.5654761791229248, 0.6964285969734192, 0.89682537317276], [0.0357142873108387, 0.0317460335791111, 0.0535714291036129, 0.0595238097012043, 0.0515873022377491, 0.0595238097012043, 0.0575396828353405, 0.0595238097012043, 0.0992063507437706, 0.0833333358168602, 0.1111111119389534, 0.065476194024086, 0.085317462682724, 0.102182537317276, 0.1408730149269104, 0.1230158731341362, 0.130952388048172, 0.1527777761220932, 0.15079365670681, 0.1388888955116272, 0.210317462682724, 0.1388888955116272, 0.184523805975914, 0.1944444477558136, 0.1964285671710968, 0.1964285671710968, 0.2033730149269104, 0.1785714328289032, 0.144841268658638, 0.0694444477558136, 0.2380952388048172, 0.1428571492433548, 0.1924603134393692, 0.1111111119389534, 0.2063492089509964, 0.2341269850730896, 0.2460317462682724, 0.2857142984867096, 0.2678571343421936, 0.2142857164144516, 0.2539682686328888, 0.2361111044883728, 0.193452388048172, 0.2718254029750824, 0.3075396716594696, 0.2460317462682724, 0.295634925365448, 0.3253968358039856, 0.2757936418056488, 0.2936508059501648, 0.3055555522441864, 0.3035714328289032, 0.3511904776096344, 0.3432539701461792, 0.4345238208770752, 0.4384920597076416, 0.3591269850730896, 0.3273809552192688, 0.3214285671710968, 0.363095223903656, 0.4722222089767456, 0.3769841194152832, 0.5039682388305664, 0.42658731341362, 0.4444444477558136, 0.545634925365448, 0.4682539701461792, 0.5079365372657776, NaN, 0.4950396716594696, 0.5059523582458496, 0.5059523582458496, 0.494047611951828, 0.533730149269104, 0.6051587462425232, 0.5873016119003296, 0.6527777910232544, 0.9047619104385376], [0.02182539738714695, 0.0496031753718853, 0.0505952388048172, 0.0515873022377491, 0.0535714291036129, 0.0575396828353405, 0.0634920671582222, 0.0714285746216774, 0.0873015895485878, 0.0813492089509964, 0.095238097012043, 0.085317462682724, 0.0714285746216774, 0.1180555522441864, 0.0873015895485878, 0.0793650820851326, 0.0972222238779068, 0.16170634329319, 0.1488095223903656, 0.1428571492433548, 0.1230158731341362, 0.1289682537317276, 0.1011904776096344, 0.1289682537317276, 0.0992063507437706, 0.1825396865606308, 0.1190476194024086, 0.1626984179019928, 0.1686507910490036, 0.1517857164144516, 0.233134925365448, 0.144841268658638, 0.170634925365448, 0.1696428507566452, 0.2013888955116272, 0.1666666716337204, 0.1914682537317276, 0.2361111044883728, 0.1944444477558136, 0.210317462682724, 0.22420634329319, 0.1805555522441864, 0.289682537317276, 0.2162698358297348, 0.2400793582201004, 0.2460317462682724, 0.2023809552192688, 0.2797619104385376, 0.230158731341362, 0.3273809552192688, 0.2757936418056488, 0.2767857015132904, 0.283730149269104, 0.3472222089767456, 0.3928571343421936, 0.4007936418056488, 0.3333333432674408, 0.3888888955116272, 0.3789682686328888, 0.3769841194152832, 0.414682537317276, 0.4047619104385376, 0.5019841194152832, 0.4186508059501648, 0.454365074634552, 0.4861111044883728, 0.3125, 0.5158730149269104, 0.504960298538208, NaN, 0.4920634925365448, 0.5257936716079712, 0.5476190447807312, 0.5555555820465088, 0.5992063283920288, 0.6845238208770752, 0.6726190447807312, 0.9146825671195984], [0.0337301604449749, 0.0416666679084301, 0.0555555559694767, 0.0496031753718853, 0.0535714291036129, 0.0595238097012043, 0.0674603208899498, 0.0634920671582222, 0.0833333358168602, 0.0873015895485878, 0.0773809552192688, 0.0793650820851326, 0.0714285746216774, 0.0992063507437706, 0.0992063507437706, 0.0833333358168602, 0.1160714253783226, 0.1587301641702652, 0.1408730149269104, 0.1130952388048172, 0.1408730149269104, 0.1001984104514122, 0.144841268658638, 0.1130952388048172, 0.1626984179019928, 0.1349206417798996, 0.170634925365448, 0.1924603134393692, 0.1795634925365448, 0.1170634925365448, 0.2182539701461792, 0.1626984179019928, 0.2023809552192688, 0.1388888955116272, 0.2281745970249176, 0.1865079402923584, 0.1805555522441864, 0.2083333283662796, 0.190476194024086, 0.2321428507566452, 0.0515873022377491, 0.1537698358297348, 0.2757936418056488, 0.2123015820980072, 0.204365074634552, 0.2142857164144516, 0.1527777761220932, 0.1388888955116272, 0.2480158656835556, 0.3134920597076416, 0.22420634329319, 0.2658730149269104, 0.2638888955116272, 0.32341268658638, 0.3472222089767456, 0.3511904776096344, 0.3373015820980072, 0.386904776096344, 0.3333333432674408, 0.3531745970249176, 0.3710317313671112, 0.3908730149269104, 0.511904776096344, 0.3313491940498352, 0.4186508059501648, 0.5714285969734192, 0.4781745970249176, 0.5297619104385376, 0.494047611951828, 0.5079365372657776, NaN, 0.5753968358039856, 0.5019841194152832, 0.5228174328804016, 0.6507936716079712, 0.6150793433189392, 0.7440476417541504, 0.9484127163887024], [0.0357142873108387, 0.0575396828353405, 0.0674603208899498, 0.0575396828353405, 0.0972222238779068, 0.0833333358168602, 0.0833333358168602, 0.0674603208899498, 0.0932539701461792, 0.1011904776096344, 0.0813492089509964, 0.105158731341362, 0.105158731341362, 0.0982142835855484, 0.1269841343164444, 0.125, 0.095238097012043, 0.15079365670681, 0.1388888955116272, 0.1011904776096344, 0.1210317462682724, 0.15079365670681, 0.1527777761220932, 0.1369047611951828, 0.1269841343164444, 0.125, 0.144841268658638, 0.1369047611951828, 0.1805555522441864, 0.1319444477558136, 0.283730149269104, 0.1785714328289032, 0.1468254029750824, 0.1468254029750824, 0.1825396865606308, 0.1567460298538208, 0.170634925365448, 0.1944444477558136, 0.1527777761220932, 0.181547611951828, 0.2083333283662796, 0.2142857164144516, 0.2638888955116272, 0.1984127014875412, 0.1686507910490036, 0.255952388048172, 0.2023809552192688, 0.2361111044883728, 0.2698412835597992, 0.1547619104385376, 0.26091268658638, 0.2916666567325592, 0.32341268658638, 0.1924603134393692, 0.2916666567325592, 0.3055555522441864, 0.3561508059501648, 0.2678571343421936, 0.283730149269104, 0.295634925365448, 0.3928571343421936, 0.300595223903656, 0.408730149269104, 0.3373015820980072, 0.329365074634552, 0.4345238208770752, 0.4414682686328888, 0.443452388048172, 0.494047611951828, 0.4742063581943512, 0.4246031641960144, NaN, 0.454365074634552, 0.454365074634552, 0.5734127163887024, 0.5416666865348816, 0.846230149269104, 0.9702380895614624], [0.011904762126505375, 0.02579365111887455, 0.0456349216401577, 0.0496031753718853, 0.0476190485060215, 0.0496031753718853, 0.0555555559694767, 0.0476190485060215, 0.085317462682724, 0.0595238097012043, 0.0466269850730896, 0.0535714291036129, 0.0535714291036129, 0.0615079365670681, 0.0932539701461792, 0.075396828353405, 0.0615079365670681, 0.1329365074634552, 0.1488095223903656, 0.095238097012043, 0.1468254029750824, 0.1349206417798996, 0.1408730149269104, 0.1547619104385376, 0.0813492089509964, 0.1091269850730896, 0.095238097012043, 0.1091269850730896, 0.1349206417798996, 0.0674603208899498, 0.2341269850730896, 0.105158731341362, 0.1210317462682724, 0.125, 0.1329365074634552, 0.1388888955116272, 0.1666666716337204, 0.190476194024086, 0.181547611951828, 0.130952388048172, 0.1924603134393692, 0.230158731341362, 0.204365074634552, 0.1726190447807312, 0.1865079402923584, 0.2162698358297348, 0.2123015820980072, 0.2420634925365448, 0.2281745970249176, 0.1071428582072258, 0.2738095223903656, 0.2400793582201004, 0.3313491940498352, 0.1924603134393692, 0.3134920597076416, 0.3214285671710968, 0.3849206268787384, 0.22420634329319, 0.2380952388048172, 0.3154761791229248, 0.4444444477558136, 0.295634925365448, 0.4444444477558136, 0.3730158805847168, 0.4067460298538208, 0.4761904776096344, 0.4107142984867096, 0.460317462682724, 0.5059523582458496, 0.4523809552192688, 0.4980158805847168, 0.545634925365448, NaN, 0.5297619104385376, 0.601190447807312, 0.511904776096344, 0.7023809552192688, 0.9623016119003296], [0.01984127052128315, 0.02182539738714695, 0.0436507947742939, 0.0396825410425663, 0.0456349216401577, 0.0515873022377491, 0.0535714291036129, 0.0337301604449749, 0.0595238097012043, 0.0694444477558136, 0.0476190485060215, 0.0456349216401577, 0.0416666679084301, 0.0694444477558136, 0.0992063507437706, 0.085317462682724, 0.0615079365670681, 0.15079365670681, 0.1349206417798996, 0.0912698432803154, 0.1210317462682724, 0.1091269850730896, 0.125, 0.1170634925365448, 0.0714285746216774, 0.1408730149269104, 0.0972222238779068, 0.085317462682724, 0.1527777761220932, 0.0634920671582222, 0.2222222238779068, 0.105158731341362, 0.095238097012043, 0.1230158731341362, 0.0972222238779068, 0.130952388048172, 0.164682537317276, 0.170634925365448, 0.184523805975914, 0.1130952388048172, 0.2182539701461792, 0.2321428507566452, 0.2182539701461792, 0.1825396865606308, 0.204365074634552, 0.2142857164144516, 0.2162698358297348, 0.2678571343421936, 0.2460317462682724, 0.1488095223903656, 0.2757936418056488, 0.2698412835597992, 0.3194444477558136, 0.1974206417798996, 0.2976190447807312, 0.30158731341362, 0.3214285671710968, 0.1567460298538208, 0.22420634329319, 0.2966269850730896, 0.4007936418056488, 0.2708333432674408, 0.460317462682724, 0.3849206268787384, 0.341269850730896, 0.5198412537574768, 0.4126984179019928, 0.488095223903656, 0.466269850730896, 0.4444444477558136, 0.477182537317276, 0.545634925365448, 0.4702380895614624, NaN, 0.60317462682724, 0.5198412537574768, 0.7103174328804016, 0.9484127163887024], [0.0396825410425663, 0.0515873022377491, 0.0644841268658638, 0.0486111119389534, 0.0476190485060215, 0.0744047611951828, 0.0664682537317276, 0.0634920671582222, 0.0575396828353405, 0.0972222238779068, 0.0694444477558136, 0.1091269850730896, 0.1289682537317276, 0.1349206417798996, 0.1011904776096344, 0.095238097012043, 0.0962301567196846, 0.130952388048172, 0.0892857164144516, 0.1388888955116272, 0.0674603208899498, 0.1607142835855484, 0.1190476194024086, 0.1001984104514122, 0.1031746044754982, 0.1547619104385376, 0.1170634925365448, 0.1686507910490036, 0.1785714328289032, 0.1537698358297348, 0.2182539701461792, 0.1924603134393692, 0.1765872985124588, 0.1329365074634552, 0.1785714328289032, 0.1527777761220932, 0.1488095223903656, 0.170634925365448, 0.1408730149269104, 0.1865079402923584, 0.1626984179019928, 0.1567460298538208, 0.2976190447807312, 0.1746031790971756, 0.1428571492433548, 0.244047611951828, 0.1666666716337204, 0.2003968209028244, 0.2083333283662796, 0.255952388048172, 0.1964285671710968, 0.2281745970249176, 0.210317462682724, 0.2658730149269104, 0.25, 0.2579365074634552, 0.2341269850730896, 0.306547611951828, 0.2638888955116272, 0.2658730149269104, 0.3115079402923584, 0.3154761791229248, 0.2142857164144516, 0.2916666567325592, 0.3134920597076416, 0.3888888955116272, 0.3789682686328888, 0.30158731341362, 0.3948412835597992, 0.4007936418056488, 0.3492063581943512, 0.42658731341362, 0.3988095223903656, 0.3968254029750824, NaN, 0.477182537317276, 0.6051587462425232, 0.8492063283920288], [0.01984127052128315, 0.0357142873108387, 0.0416666679084301, 0.02777777798473835, 0.0337301604449749, 0.0416666679084301, 0.0416666679084301, 0.0436507947742939, 0.0436507947742939, 0.0515873022377491, 0.0456349216401577, 0.0634920671582222, 0.0396825410425663, 0.0694444477558136, 0.0436507947742939, 0.0615079365670681, 0.0496031753718853, 0.1011904776096344, 0.085317462682724, 0.0714285746216774, 0.0694444477558136, 0.0575396828353405, 0.0535714291036129, 0.0813492089509964, 0.065476194024086, 0.1230158731341362, 0.0674603208899498, 0.1091269850730896, 0.1170634925365448, 0.0892857164144516, 0.2023809552192688, 0.1031746044754982, 0.1011904776096344, 0.1011904776096344, 0.1190476194024086, 0.1170634925365448, 0.1031746044754982, 0.1408730149269104, 0.1428571492433548, 0.1210317462682724, 0.1349206417798996, 0.1190476194024086, 0.210317462682724, 0.125, 0.1488095223903656, 0.1726190447807312, 0.125, 0.1984127014875412, 0.1746031790971756, 0.255952388048172, 0.1865079402923584, 0.1855158656835556, 0.1944444477558136, 0.2777777910232544, 0.3591269850730896, 0.3611111044883728, 0.2460317462682724, 0.3253968358039856, 0.2777777910232544, 0.2678571343421936, 0.3650793731212616, 0.3452380895614624, 0.4285714328289032, 0.3273809552192688, 0.369047611951828, 0.44841268658638, 0.22420634329319, 0.4345238208770752, 0.4126984179019928, 0.3154761791229248, 0.3849206268787384, 0.4583333432674408, 0.488095223903656, 0.4801587164402008, 0.5228174328804016, NaN, 0.6190476417541504, 0.85317462682724], [0.0436507947742939, 0.0476190485060215, 0.0535714291036129, 0.0535714291036129, 0.0634920671582222, 0.0734127014875412, 0.0674603208899498, 0.0575396828353405, 0.072420634329319, 0.0833333358168602, 0.0575396828353405, 0.085317462682724, 0.0793650820851326, 0.0873015895485878, 0.0892857164144516, 0.085317462682724, 0.0714285746216774, 0.1230158731341362, 0.1031746044754982, 0.0972222238779068, 0.0515873022377491, 0.1190476194024086, 0.1111111119389534, 0.0615079365670681, 0.082341268658638, 0.0992063507437706, 0.0932539701461792, 0.1031746044754982, 0.1329365074634552, 0.1011904776096344, 0.1746031790971756, 0.1339285671710968, 0.0972222238779068, 0.1011904776096344, 0.122023805975914, 0.0912698432803154, 0.1061507910490036, 0.1071428582072258, 0.1210317462682724, 0.125, 0.1210317462682724, 0.1269841343164444, 0.230158731341362, 0.105158731341362, 0.0873015895485878, 0.181547611951828, 0.1329365074634552, 0.15079365670681, 0.1607142835855484, 0.1190476194024086, 0.1825396865606308, 0.1865079402923584, 0.1825396865606308, 0.115079365670681, 0.15079365670681, 0.1607142835855484, 0.1746031790971756, 0.1626984179019928, 0.2083333283662796, 0.2003968209028244, 0.2460317462682724, 0.1746031790971756, 0.283730149269104, 0.2123015820980072, 0.2083333283662796, 0.2936508059501648, 0.3055555522441864, 0.3035714328289032, 0.3472222089767456, 0.3273809552192688, 0.255952388048172, 0.1537698358297348, 0.2976190447807312, 0.289682537317276, 0.3948412835597992, 0.380952388048172, NaN, 0.9365079402923584], [0.0059523810632526875, 0.003968254197388887, 0.0059523810632526875, 0.003968254197388887, 0.0059523810632526875, 0.007936508394777775, 0.007936508394777775, 0.007936508394777775, 0.009920635260641575, 0.0059523810632526875, 0.0059523810632526875, 0.009920635260641575, 0.0059523810632526875, 0.0059523810632526875, 0.01587301678955555, 0.003968254197388887, 0.007936508394777775, 0.013888888992369175, 0.01984127052128315, 0.009920635260641575, 0.013888888992369175, 0.01785714365541935, 0.02380952425301075, 0.0059523810632526875, 0.007936508394777775, 0.007936508394777775, 0.0059523810632526875, 0.007936508394777775, 0.01984127052128315, 0.009920635260641575, 0.01785714365541935, 0.02182539738714695, 0.007936508394777775, 0.009920635260641575, 0.01785714365541935, 0.011904762126505375, 0.01984127052128315, 0.02777777798473835, 0.01587301678955555, 0.02182539738714695, 0.02380952425301075, 0.02380952425301075, 0.0595238097012043, 0.01984127052128315, 0.02182539738714695, 0.0357142873108387, 0.02182539738714695, 0.0317460335791111, 0.02480158768594265, 0.009920635260641575, 0.0337301604449749, 0.0357142873108387, 0.02380952425301075, 0.011904762126505375, 0.0456349216401577, 0.0456349216401577, 0.02976190485060215, 0.0317460335791111, 0.0535714291036129, 0.0367063507437706, 0.0575396828353405, 0.02976190485060215, 0.095238097012043, 0.0515873022377491, 0.075396828353405, 0.0575396828353405, 0.0595238097012043, 0.1031746044754982, 0.095238097012043, 0.085317462682724, 0.0515873022377491, 0.02976190485060215, 0.0376984141767025, 0.0515873022377491, 0.15079365670681, 0.1468254029750824, 0.0634920671582222, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9042336623},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8326504819},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8203205532},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8169449576},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8082740707},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8081581095},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7979153757},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.797425786},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.7407364458},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7292955067},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.719413523},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7191944979},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6933235394},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6626468771},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6621057521},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6590136033},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6576221402},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.656617194},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.6549938164},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6375360721},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6357580921},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6224747463},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6148216852},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6081864586},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6002370656},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5786951132},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5764146583},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5618686875},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5608122041},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5606060617},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5601937755},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.551394041},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5490878189},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5390769945},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5203695109},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5176252307},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5172387098},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5118145733},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5073953837},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5053983748},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5053468373},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5009791793},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4880050497},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4806740888},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4645305085},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4552927226},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4550350441},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.4422799414},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4411332711},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4316764586},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.423057103},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4214466109},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.4111652238},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4059085755},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4021464642},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3966192526},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.3774221803},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3526721296},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3509714514},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3307951975},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.324997423},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3159400131},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.3092790157},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3051561516},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3020768923},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2977221195},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.286783654},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2844130086},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2717738621},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2578978566},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2513528147},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2394480524},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2266285306},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2211657396},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.208191611},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1937744806},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1582019181},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0301484238}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_classification/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_classification/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..0c555f0635ed1fe13d8cfc339744d4e651d8abb6
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_classification/datasets_small/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),TabPFN-2.6 (default),Mitra (default),TabPFNv2 (tuned + ensembled),TabDPT-Turbo (default),LimiX (default),RealMLP (tuned + ensembled),TabDPT (tuned + ensembled),TabM (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabICL (default),TabPFNv2 (tuned),TabPFNv2 (default),TabDPT (tuned),BetaTabPFN (default),RealMLP (tuned),TabM (tuned),TabSwift (default),LightGBM (tuned + ensembled),TabDPT (default),LightGBM (tuned),CatBoost (tuned + ensembled),ModernNCA (tuned),xRFM (tuned + ensembled),SAP-RPT-OSS (default),iLTM (tuned + ensembled),CatBoost (tuned),xRFM (tuned),XGBoost (tuned + ensembled),CatBoost (default),EBM (tuned + ensembled),TabM (default),ModernNCA (default),XGBoost (tuned),FastaiMLP (tuned + ensembled),TorchMLP (tuned + ensembled),ModernNCA (tuned + ensembled),EBM (tuned),EBM (default),iLTM (tuned),RealMLP (default),FastaiMLP (tuned),ChimeraBoost (tuned + ensembled),ExtraTrees (tuned + ensembled),TorchMLP (tuned),ChimeraBoost (tuned),OrionMSP (default),ExtraTrees (tuned),TabSTAR (tuned),TabSTAR (tuned + ensembled),ChimeraBoost (default),RandomForest (tuned + ensembled),XGBoost (default),LightGBM (default),TabFlex (default),RandomForest (tuned),Linear (tuned + ensembled),TorchMLP (default),iLTM (default),TabSTAR (default),PerpetualBooster (tuned + ensembled),Linear (tuned),xRFM (default),PerpetualBooster (tuned),FastaiMLP (default),KNN (tuned + ensembled),ExtraTrees (default),RandomForest (default),Linear (default),PerpetualBooster (default),KNN (tuned),KNN (default)
+TabFM (default),,0.7956349,0.7876984,0.8194444,0.76785713,0.7708333,0.7757937,0.8035714,0.8531746,0.8630952,0.8313492,0.8799603,0.85714287,0.8869048,0.86507934,0.9047619,0.9325397,0.86706346,0.8293651,0.87896824,0.8581349,0.85119045,0.89484125,0.86904764,0.9007937,0.890873,0.89880955,0.9186508,0.9027778,0.9126984,0.8373016,0.9206349,0.92460316,0.89285713,0.9325397,0.9265873,0.89484125,0.9007937,0.9047619,0.9325397,0.8968254,0.9444444,0.9444444,0.90674603,0.9007937,0.9345238,0.9186508,0.92261904,0.9285714,0.93849206,0.95238096,0.93849206,0.9305556,0.9285714,0.82539684,0.82539684,0.9464286,0.9563492,0.9583333,0.9543651,0.9265873,0.9543651,0.93650794,0.96031743,0.9305556,0.87103176,0.98214287,0.9444444,0.96428573,0.9781746,0.96626985,0.96428573,0.9880952,0.98015875,0.96031743,0.98015875,0.9563492,0.99404764
+TabPFN-3 (default),0.20436507,,0.5555556,0.5257937,0.60119045,0.57539684,0.5972222,0.6111111,0.7063492,0.70238096,0.6964286,0.71825397,0.7242063,0.78373015,0.75793654,0.74206346,0.8402778,0.77281743,0.72321427,0.8055556,0.73710316,0.7638889,0.7956349,0.84325397,0.8154762,0.8194444,0.8333333,0.8492063,0.8492063,0.8531746,0.74007934,0.8769841,0.859127,0.8472222,0.8501984,0.875,0.83928573,0.8194444,0.8640873,0.8630952,0.83035713,0.8968254,0.8869048,0.8531746,0.8660714,0.9156746,0.9027778,0.8839286,0.90873015,0.92460316,0.9166667,0.90873015,0.9126984,0.92261904,0.8125,0.8174603,0.90873015,0.9206349,0.9583333,0.9404762,0.9166667,0.9424603,0.9047619,0.95039684,0.93849206,0.8531746,0.93650794,0.9126984,0.96825397,0.95039684,0.9583333,0.9424603,0.9742063,0.9781746,0.9484127,0.96428573,0.95238096,0.99603176
+"AutoGluon 1.5 (extreme, 4h)",0.21230158,0.44444445,,0.46031746,0.5416667,0.53373015,0.5535714,0.609127,0.64781743,0.6527778,0.5972222,0.7202381,0.76289684,0.765873,0.7916667,0.7916667,0.7986111,0.7281746,0.7242063,0.7876984,0.7380952,0.7956349,0.796627,0.8035714,0.8373016,0.7619048,0.8333333,0.8531746,0.85515875,0.8611111,0.7202381,0.8769841,0.8501984,0.83035713,0.8611111,0.8660714,0.83928573,0.8373016,0.8472222,0.859127,0.8373016,0.8869048,0.90873015,0.85119045,0.85714287,0.90873015,0.86706346,0.86507934,0.9126984,0.9047619,0.9206349,0.91468257,0.9156746,0.890873,0.7777778,0.7797619,0.9166667,0.92261904,0.9464286,0.9464286,0.8968254,0.9265873,0.88293654,0.95039684,0.9325397,0.82539684,0.9444444,0.90873015,0.9464286,0.9494048,0.9444444,0.9325397,0.9543651,0.9563492,0.9355159,0.9583333,0.9464286,0.99404764
+TabICLv2 (default),0.18055555,0.47420636,0.53968257,,0.53869045,0.5297619,0.5436508,0.6031746,0.6617063,0.6765873,0.6825397,0.7172619,0.7242063,0.77876985,0.7619048,0.7718254,0.84623015,0.7152778,0.7063492,0.76984125,0.74007934,0.74603176,0.7757937,0.82539684,0.8055556,0.7956349,0.82738096,0.8174603,0.83531743,0.8313492,0.6984127,0.86706346,0.82738096,0.8144841,0.83531743,0.859127,0.8333333,0.81150794,0.8640873,0.84325397,0.83531743,0.8769841,0.8630952,0.83928573,0.859127,0.88492066,0.88095236,0.87896824,0.89285713,0.8720238,0.90873015,0.90674603,0.90873015,0.8769841,0.8035714,0.8015873,0.9027778,0.91468257,0.9265873,0.93650794,0.89484125,0.9126984,0.9186508,0.9345238,0.9166667,0.85515875,0.9424603,0.92460316,0.9404762,0.9484127,0.95039684,0.9424603,0.95039684,0.96031743,0.9513889,0.9722222,0.9464286,0.99603176
+RealTabPFN-2.5 (default),0.23214285,0.39880952,0.45833334,0.46130952,,0.44047618,0.4890873,0.50793654,0.6369048,0.64285713,0.5714286,0.68849206,0.71031743,0.6924603,0.75992066,0.75396824,0.7063492,0.7380952,0.75793654,0.6994048,0.7589286,0.76984125,0.79960316,0.78968257,0.8075397,0.7123016,0.8293651,0.8472222,0.8055556,0.82539684,0.7619048,0.8194444,0.8472222,0.8333333,0.85515875,0.88293654,0.859127,0.8531746,0.83928573,0.875,0.86706346,0.8869048,0.83531743,0.87896824,0.8611111,0.90376985,0.89880955,0.890873,0.9027778,0.8630952,0.9166667,0.9047619,0.9206349,0.875,0.8035714,0.80456346,0.9126984,0.8869048,0.9305556,0.9444444,0.91369045,0.9345238,0.9007937,0.9623016,0.9325397,0.8373016,0.9404762,0.90873015,0.9484127,0.9464286,0.9464286,0.9027778,0.95238096,0.9543651,0.95238096,0.96626985,0.93650794,0.99404764
+RealTabPFN-2.5 (tuned + ensembled),0.22916667,0.42460316,0.46626985,0.4702381,0.5595238,,0.57539684,0.5515873,0.66071427,0.71428573,0.5654762,0.7083333,0.7361111,0.6845238,0.75595236,0.7757937,0.703373,0.7757937,0.75396824,0.6984127,0.76984125,0.76785713,0.78373015,0.75793654,0.81349206,0.6686508,0.83035713,0.85119045,0.8472222,0.8333333,0.7529762,0.8382937,0.8452381,0.8214286,0.859127,0.8630952,0.83531743,0.82539684,0.8333333,0.8720238,0.8343254,0.87896824,0.87896824,0.84821427,0.84126985,0.8968254,0.8630952,0.875,0.8918651,0.859127,0.91071427,0.89880955,0.890873,0.8611111,0.7777778,0.78373015,0.89880955,0.9027778,0.9404762,0.93650794,0.890873,0.91468257,0.89880955,0.93849206,0.92261904,0.8154762,0.9424603,0.91468257,0.9404762,0.9424603,0.9404762,0.9166667,0.95039684,0.9484127,0.9255952,0.9583333,0.9265873,0.99206346
+RealTabPFN-2.5 (tuned),0.22420634,0.4027778,0.44642857,0.4563492,0.5109127,0.42460316,,0.5208333,0.6527778,0.671627,0.5694444,0.67261904,0.71031743,0.6805556,0.75396824,0.7638889,0.6805556,0.7549603,0.7380952,0.7093254,0.75595236,0.7638889,0.78869045,0.74900794,0.79960316,0.68650794,0.8293651,0.8373016,0.8095238,0.8144841,0.7480159,0.8194444,0.8333333,0.8095238,0.8452381,0.85119045,0.8472222,0.8293651,0.8214286,0.8531746,0.8214286,0.87896824,0.8700397,0.85119045,0.83531743,0.890873,0.8630952,0.86507934,0.8819444,0.86210316,0.91071427,0.89285713,0.88492066,0.86904764,0.76984125,0.7777778,0.8968254,0.89285713,0.9345238,0.93650794,0.88095236,0.90873015,0.88492066,0.9464286,0.9186508,0.8154762,0.9255952,0.9007937,0.9424603,0.93650794,0.9325397,0.9166667,0.9444444,0.9464286,0.93353176,0.9583333,0.9325397,0.99206346
+TabPFN-2.6 (default),0.19642857,0.3888889,0.39087301,0.3968254,0.4920635,0.4484127,0.47916666,,0.640873,0.61904764,0.5873016,0.6626984,0.71825397,0.67460316,0.74206346,0.74404764,0.73214287,0.7281746,0.67956346,0.68650794,0.7281746,0.75595236,0.7797619,0.7619048,0.79960316,0.70039684,0.83531743,0.8293651,0.8293651,0.8095238,0.7380952,0.87896824,0.8214286,0.8015873,0.859127,0.875,0.8095238,0.81845236,0.8452381,0.86507934,0.82539684,0.8769841,0.90873015,0.8382937,0.8313492,0.9047619,0.86507934,0.86706346,0.91071427,0.88293654,0.90873015,0.9126984,0.8611111,0.8720238,0.7757937,0.78571427,0.9186508,0.9007937,0.9543651,0.9444444,0.89484125,0.92261904,0.88492066,0.9404762,0.92261904,0.8154762,0.9325397,0.90873015,0.9404762,0.9285714,0.93650794,0.9325397,0.95238096,0.96626985,0.93650794,0.9563492,0.9424603,0.99206346
+Mitra (default),0.1468254,0.2936508,0.35218254,0.33829364,0.36309522,0.3392857,0.3472222,0.35912699,,0.5218254,0.5218254,0.51289684,0.59126985,0.59126985,0.6706349,0.65674603,0.609127,0.6140873,0.625,0.6031746,0.6686508,0.6845238,0.74007934,0.6825397,0.73214287,0.6547619,0.76785713,0.74404764,0.75595236,0.74603176,0.640873,0.7827381,0.73511904,0.7529762,0.7867063,0.8234127,0.79960316,0.765873,0.78968257,0.78174603,0.7936508,0.8293651,0.8035714,0.8174603,0.82738096,0.8492063,0.8154762,0.83928573,0.86706346,0.8313492,0.8888889,0.87103176,0.796627,0.8422619,0.75396824,0.75595236,0.8888889,0.87103176,0.9206349,0.9186508,0.859127,0.89880955,0.8888889,0.91468257,0.9027778,0.78968257,0.89285713,0.90674603,0.9007937,0.9126984,0.9166667,0.90674603,0.91468257,0.9404762,0.9424603,0.9563492,0.92757934,0.99007934
+TabPFNv2 (tuned + ensembled),0.13690476,0.29761904,0.3472222,0.3234127,0.35714287,0.2857143,0.32837301,0.3809524,0.4781746,,0.50396824,0.4861111,0.6041667,0.57738096,0.6666667,0.6924603,0.5793651,0.6468254,0.6299603,0.5902778,0.67956346,0.68154764,0.71130955,0.640873,0.72321427,0.6111111,0.74404764,0.74404764,0.75793654,0.76785713,0.6160714,0.75595236,0.75099206,0.7589286,0.78869045,0.78571427,0.7847222,0.75595236,0.76488096,0.79761904,0.7708333,0.79960316,0.8015873,0.79761904,0.79960316,0.82539684,0.81150794,0.79464287,0.83531743,0.8263889,0.859127,0.8452381,0.77380955,0.83928573,0.734127,0.7380952,0.8611111,0.88095236,0.90674603,0.9206349,0.8234127,0.89880955,0.86507934,0.9007937,0.87103176,0.78571427,0.89285713,0.88293654,0.9166667,0.9186508,0.9126984,0.89880955,0.9404762,0.9305556,0.9027778,0.9484127,0.9166667,0.99404764
+TabDPT-Turbo (default),0.16865079,0.30357143,0.4027778,0.31746033,0.42857143,0.43452382,0.43055555,0.41269842,0.4781746,0.49603173,,0.56150794,0.5833333,0.63095236,0.58035713,0.58928573,0.6031746,0.54960316,0.57539684,0.6527778,0.5674603,0.63789684,0.62103176,0.7291667,0.65873015,0.78869045,0.66071427,0.6984127,0.6904762,0.68650794,0.63095236,0.7480159,0.68849206,0.71428573,0.71428573,0.7222222,0.7123016,0.68650794,0.765873,0.7222222,0.76984125,0.74007934,0.73214287,0.7450397,0.75595236,0.82242066,0.7916667,0.82738096,0.8015873,0.83630955,0.8015873,0.8174603,0.8611111,0.859127,0.73015875,0.7380952,0.8313492,0.875,0.86706346,0.8730159,0.90873015,0.87896824,0.84126985,0.88492066,0.8630952,0.79960316,0.88492066,0.85119045,0.8888889,0.9047619,0.92261904,0.9186508,0.953373,0.95238096,0.9305556,0.9543651,0.9424603,0.99404764
+LimiX (default),0.12003968,0.28174603,0.2797619,0.2827381,0.31150794,0.29166666,0.32738096,0.33730158,0.48710316,0.5138889,0.43849206,,0.5972222,0.53869045,0.6269841,0.63293654,0.5873016,0.5873016,0.5972222,0.5674603,0.625,0.6418651,0.63988096,0.609127,0.71626985,0.58531743,0.72619045,0.7380952,0.7380952,0.75793654,0.6686508,0.796627,0.74007934,0.7450397,0.7718254,0.7777778,0.76785713,0.71428573,0.76984125,0.78968257,0.7172619,0.75595236,0.7777778,0.7956349,0.79761904,0.82738096,0.8095238,0.7906746,0.8472222,0.84126985,0.81349206,0.85515875,0.82043654,0.83928573,0.76488096,0.75793654,0.87103176,0.8968254,0.89880955,0.9126984,0.81150794,0.91071427,0.8611111,0.8888889,0.89285713,0.8105159,0.9047619,0.88293654,0.9345238,0.91468257,0.9206349,0.89484125,0.9464286,0.9543651,0.890873,0.93650794,0.91468257,0.99007934
+RealMLP (tuned + ensembled),0.14285715,0.27579364,0.23710318,0.27579364,0.28968254,0.2638889,0.28968254,0.28174603,0.40873015,0.39583334,0.41666666,0.4027778,,0.48710316,0.57738096,0.5327381,0.5019841,0.49603173,0.47619048,0.49503967,0.59325397,0.66369045,0.63492066,0.5297619,0.70039684,0.54761904,0.7311508,0.7519841,0.6626984,0.7043651,0.5793651,0.74007934,0.75595236,0.6825397,0.8174603,0.7718254,0.7718254,0.75595236,0.71031743,0.82738096,0.7569444,0.73214287,0.7123016,0.7797619,0.77876985,0.81150794,0.84623015,0.78075397,0.8214286,0.79761904,0.78571427,0.83035713,0.74404764,0.8125,0.72619045,0.734127,0.8313492,0.8869048,0.91468257,0.9007937,0.7956349,0.9186508,0.8472222,0.9007937,0.890873,0.7797619,0.92261904,0.8492063,0.91468257,0.9285714,0.9285714,0.89484125,0.9464286,0.9583333,0.87103176,0.96031743,0.9206349,0.99404764
+TabDPT (tuned + ensembled),0.11309524,0.21626984,0.23412699,0.22123016,0.30753967,0.31547618,0.31944445,0.32539684,0.40873015,0.42261904,0.3690476,0.46130952,0.51289684,,0.5168651,0.52380955,0.5188492,0.5109127,0.49007937,0.60714287,0.5218254,0.59424603,0.55456346,0.59821427,0.5952381,0.63095236,0.6269841,0.625,0.6468254,0.63988096,0.58531743,0.6944444,0.63293654,0.65376985,0.6577381,0.6805556,0.6656746,0.6418651,0.7013889,0.6825397,0.6964286,0.69742066,0.73015875,0.69543654,0.6924603,0.7797619,0.7281746,0.7519841,0.75595236,0.78373015,0.75992066,0.76984125,0.8015873,0.79960316,0.6984127,0.70039684,0.7936508,0.85714287,0.84325397,0.85119045,0.8293651,0.87103176,0.78571427,0.86904764,0.85714287,0.7718254,0.86706346,0.8075397,0.89781743,0.8819444,0.9007937,0.90178573,0.93849206,0.9305556,0.86507934,0.9305556,0.9126984,0.99404764
+TabM (tuned + ensembled),0.13492064,0.24206349,0.20833333,0.23809524,0.24007936,0.24404761,0.24603175,0.2579365,0.32936507,0.33333334,0.41964287,0.37301588,0.42261904,0.48313493,,0.49007937,0.48015872,0.42857143,0.42757937,0.49007937,0.59126985,0.5515873,0.6547619,0.53968257,0.640873,0.5456349,0.66369045,0.6507937,0.6220238,0.60714287,0.58928573,0.65178573,0.6765873,0.6111111,0.74603176,0.75,0.7619048,0.82242066,0.6845238,0.76785713,0.6944444,0.71825397,0.70039684,0.7757937,0.7688492,0.75,0.78869045,0.75,0.8095238,0.73015875,0.7986111,0.83531743,0.71626985,0.72619045,0.72619045,0.734127,0.86706346,0.78968257,0.88095236,0.8839286,0.79761904,0.84126985,0.85515875,0.89285713,0.88095236,0.78373015,0.89285713,0.8720238,0.859127,0.9126984,0.9007937,0.8730159,0.90674603,0.9007937,0.89880955,0.9563492,0.91071427,0.984127
+"AutoGluon 1.4 (best, 4h)",0.0952381,0.2579365,0.20833333,0.2281746,0.24603175,0.22420634,0.2361111,0.2559524,0.34325397,0.30753967,0.4107143,0.3670635,0.4672619,0.47619048,0.50992066,,0.48015872,0.41666666,0.45833334,0.4781746,0.5515873,0.5595238,0.5694444,0.52281743,0.6706349,0.54761904,0.6904762,0.6785714,0.6111111,0.6438492,0.5297619,0.6805556,0.7063492,0.67261904,0.72619045,0.74404764,0.7063492,0.6944444,0.6904762,0.7123016,0.7242063,0.7519841,0.6785714,0.72619045,0.7638889,0.7718254,0.77380955,0.78174603,0.7847222,0.7083333,0.8055556,0.7956349,0.7093254,0.74007934,0.7242063,0.73214287,0.8194444,0.8075397,0.8769841,0.9007937,0.8095238,0.84821427,0.8422619,0.8769841,0.86507934,0.78174603,0.91468257,0.8700397,0.8769841,0.9206349,0.9166667,0.875,0.92460316,0.91468257,0.9047619,0.93849206,0.91468257,0.99603176
+TabICL (default),0.06746032,0.15972222,0.2013889,0.15376984,0.2936508,0.29662699,0.31944445,0.26785713,0.39087301,0.42063493,0.3968254,0.41269842,0.49801588,0.4811508,0.51984125,0.51984125,,0.48015872,0.49702382,0.5069444,0.5257937,0.5674603,0.5694444,0.62103176,0.6111111,0.5922619,0.62896824,0.61904764,0.66071427,0.6369048,0.5833333,0.69146824,0.6369048,0.6388889,0.67460316,0.70039684,0.63492066,0.6666667,0.6894841,0.68849206,0.6785714,0.6626984,0.71428573,0.6666667,0.6964286,0.765873,0.7281746,0.7361111,0.78373015,0.79761904,0.765873,0.7956349,0.7936508,0.80654764,0.71031743,0.7123016,0.8154762,0.8452381,0.8531746,0.8918651,0.8234127,0.87896824,0.79761904,0.85714287,0.8452381,0.76984125,0.89285713,0.8174603,0.86904764,0.9027778,0.8839286,0.9047619,0.93849206,0.93849206,0.90376985,0.95039684,0.9285714,0.99206346
+TabPFNv2 (tuned),0.13293651,0.22718254,0.2718254,0.2847222,0.26190478,0.22420634,0.24503969,0.2718254,0.3859127,0.3531746,0.45039684,0.41269842,0.50396824,0.4890873,0.5714286,0.5833333,0.51984125,,0.5416667,0.5248016,0.61507934,0.59126985,0.6369048,0.5674603,0.6468254,0.5555556,0.6626984,0.6686508,0.6626984,0.6507937,0.58234125,0.625,0.70238096,0.64880955,0.69742066,0.71428573,0.70238096,0.6964286,0.69543654,0.6964286,0.7013889,0.7281746,0.6964286,0.7222222,0.73214287,0.75099206,0.7589286,0.734127,0.765873,0.7281746,0.7876984,0.78174603,0.70039684,0.7718254,0.71626985,0.71031743,0.81150794,0.76984125,0.7956349,0.8293651,0.7916667,0.82043654,0.8293651,0.8333333,0.8452381,0.7718254,0.83531743,0.84126985,0.8472222,0.8382937,0.84126985,0.8492063,0.86706346,0.8492063,0.86904764,0.89880955,0.8769841,0.9861111
+TabPFNv2 (default),0.17063493,0.2767857,0.27579364,0.2936508,0.24206349,0.24603175,0.26190478,0.3204365,0.375,0.37003967,0.42460316,0.4027778,0.52380955,0.50992066,0.57242066,0.5416667,0.5029762,0.45833334,,0.5406746,0.5833333,0.5654762,0.6111111,0.56150794,0.64285713,0.5575397,0.64880955,0.6527778,0.61706346,0.6666667,0.5595238,0.609127,0.6706349,0.6845238,0.6765873,0.69543654,0.7043651,0.6984127,0.67956346,0.68650794,0.7083333,0.68650794,0.640873,0.7222222,0.7380952,0.6924603,0.75992066,0.7480159,0.71825397,0.74603176,0.74404764,0.73015875,0.7480159,0.7619048,0.74603176,0.7410714,0.7380952,0.7916667,0.78571427,0.82539684,0.85119045,0.8373016,0.86011904,0.85515875,0.82738096,0.79761904,0.85119045,0.8611111,0.8492063,0.85119045,0.859127,0.8611111,0.85119045,0.86507934,0.91071427,0.91468257,0.8968254,0.98015875
+TabDPT (tuned),0.121031746,0.19444445,0.21230158,0.23015873,0.30059522,0.3015873,0.2906746,0.31349206,0.3968254,0.4097222,0.3472222,0.43253967,0.5049603,0.39285713,0.50992066,0.5218254,0.49305555,0.47519842,0.4593254,,0.5,0.54960316,0.5515873,0.5694444,0.5674603,0.6140873,0.59126985,0.5952381,0.6130952,0.5952381,0.5555556,0.6388889,0.60218257,0.60515875,0.64781743,0.6468254,0.6468254,0.63095236,0.67261904,0.6507937,0.6686508,0.6845238,0.69345236,0.67460316,0.6825397,0.73015875,0.70238096,0.72619045,0.734127,0.7361111,0.72619045,0.74206346,0.81150794,0.75595236,0.6875,0.68650794,0.7797619,0.79960316,0.7916667,0.8154762,0.82539684,0.82539684,0.7748016,0.85119045,0.8313492,0.765873,0.8452381,0.78968257,0.8611111,0.85714287,0.8869048,0.89880955,0.9047619,0.90873015,0.8611111,0.9285714,0.9027778,0.99007934
+BetaTabPFN (default),0.14186507,0.26289684,0.26190478,0.25992063,0.24107143,0.23015873,0.24404761,0.2718254,0.3313492,0.3204365,0.43253967,0.375,0.40674603,0.4781746,0.40873015,0.4484127,0.47420636,0.38492063,0.41666666,0.5,,0.45833334,0.49007937,0.5277778,0.60119045,0.5833333,0.5992063,0.5873016,0.59623015,0.5952381,0.5376984,0.60515875,0.6220238,0.60515875,0.66071427,0.6964286,0.71825397,0.6686508,0.65674603,0.67460316,0.65178573,0.6825397,0.61904764,0.75793654,0.78174603,0.6706349,0.74007934,0.7242063,0.75099206,0.72619045,0.75793654,0.7619048,0.71825397,0.74206346,0.7519841,0.75595236,0.8174603,0.79761904,0.8075397,0.81845236,0.7986111,0.8075397,0.87896824,0.859127,0.89880955,0.8125,0.85515875,0.89880955,0.78968257,0.8769841,0.859127,0.87896824,0.8531746,0.87896824,0.9325397,0.9305556,0.9484127,0.9861111
+RealMLP (tuned),0.14880952,0.2361111,0.20436507,0.25396827,0.23015873,0.23214285,0.2361111,0.24404761,0.31547618,0.3184524,0.36210316,0.35813493,0.33630952,0.40575397,0.4484127,0.44047618,0.43253967,0.40873015,0.43452382,0.45039684,0.5416667,,0.5376984,0.48412699,0.57738096,0.5019841,0.5952381,0.6130952,0.5714286,0.5734127,0.5218254,0.5922619,0.62103176,0.6001984,0.6924603,0.67460316,0.6805556,0.6706349,0.6527778,0.71130955,0.66468257,0.6765873,0.6468254,0.7202381,0.6825397,0.71428573,0.74603176,0.734127,0.7083333,0.73015875,0.74603176,0.734127,0.67956346,0.74206346,0.71031743,0.71626985,0.7638889,0.8075397,0.8333333,0.8373016,0.77876985,0.8313492,0.7876984,0.86706346,0.8472222,0.765873,0.86904764,0.796627,0.8611111,0.87103176,0.8998016,0.8492063,0.86507934,0.890873,0.83928573,0.9424603,0.88095236,0.98214287
+TabM (tuned),0.10515873,0.20436507,0.20337301,0.22420634,0.20039682,0.21626984,0.21130952,0.22023809,0.25992063,0.28869048,0.37896827,0.36011904,0.36507937,0.4454365,0.3452381,0.43055555,0.43055555,0.36309522,0.3888889,0.4484127,0.50992066,0.46230158,,0.50396824,0.5625,0.52281743,0.58531743,0.5972222,0.59325397,0.5456349,0.5119048,0.6031746,0.61507934,0.5694444,0.6656746,0.6825397,0.6666667,0.70039684,0.6468254,0.67460316,0.6458333,0.6805556,0.6547619,0.7043651,0.71825397,0.6765873,0.74404764,0.71626985,0.7619048,0.6825397,0.74900794,0.75396824,0.68849206,0.6904762,0.7063492,0.7172619,0.8144841,0.77380955,0.8313492,0.8174603,0.78373015,0.7916667,0.83928573,0.875,0.86507934,0.7638889,0.890873,0.85119045,0.8154762,0.89880955,0.85515875,0.8472222,0.859127,0.875,0.88095236,0.9464286,0.8888889,0.97619045
+TabSwift (default),0.13095239,0.15674603,0.19642857,0.17460318,0.21031746,0.24206349,0.25099206,0.23809524,0.31746033,0.35912699,0.27083334,0.39087301,0.4702381,0.4017857,0.46031746,0.47718254,0.37896827,0.43253967,0.43849206,0.43055555,0.4722222,0.515873,0.49603173,,0.51984125,0.515873,0.5257937,0.53968257,0.5694444,0.5515873,0.57539684,0.58928573,0.5644841,0.5595238,0.5972222,0.60515875,0.61507934,0.6230159,0.6230159,0.62896824,0.7063492,0.65674603,0.62103176,0.64285713,0.64880955,0.7043651,0.6964286,0.7361111,0.6626984,0.6805556,0.71626985,0.6924603,0.84325397,0.6944444,0.74404764,0.75,0.7757937,0.7222222,0.7380952,0.8105159,0.8888889,0.78174603,0.76785713,0.84126985,0.81150794,0.8095238,0.87103176,0.7936508,0.8055556,0.87103176,0.8869048,0.8630952,0.8452381,0.88293654,0.8998016,0.9186508,0.93849206,0.99404764
+LightGBM (tuned + ensembled),0.09920635,0.1845238,0.16269842,0.19444445,0.19246031,0.18650794,0.20039682,0.20039682,0.26785713,0.2767857,0.34126985,0.28373015,0.29960316,0.4047619,0.35912699,0.32936507,0.3888889,0.3531746,0.35714287,0.43253967,0.39880952,0.42261904,0.4375,0.48015872,,0.5257937,0.5625,0.57738096,0.50396824,0.54761904,0.5138889,0.5605159,0.5882937,0.5992063,0.7063492,0.69146824,0.6458333,0.65178573,0.5972222,0.71825397,0.59126985,0.59126985,0.5734127,0.68650794,0.7361111,0.6686508,0.68849206,0.6805556,0.7470238,0.6894841,0.6984127,0.75992066,0.6964286,0.6944444,0.7093254,0.7123016,0.84126985,0.79761904,0.8730159,0.9424603,0.74603176,0.85714287,0.81150794,0.8055556,0.8730159,0.78968257,0.8769841,0.8333333,0.8035714,0.9007937,0.8373016,0.8730159,0.9186508,0.9285714,0.8968254,0.9345238,0.91765875,0.99206346
+TabDPT (default),0.109126985,0.18055555,0.23809524,0.20436507,0.28769842,0.3313492,0.31349206,0.29960316,0.3452381,0.3888889,0.21130952,0.41468254,0.45238096,0.3690476,0.45436507,0.45238096,0.4077381,0.44444445,0.44246033,0.3859127,0.41666666,0.49801588,0.47718254,0.48412699,0.47420636,,0.49603173,0.5257937,0.5456349,0.5277778,0.53174603,0.5416667,0.53571427,0.53174603,0.5694444,0.5376984,0.5297619,0.5734127,0.5952381,0.58531743,0.58531743,0.5873016,0.59325397,0.55654764,0.59126985,0.63095236,0.5843254,0.65873015,0.6507937,0.68849206,0.64285713,0.6527778,0.74603176,0.71428573,0.63293654,0.63293654,0.71031743,0.7123016,0.6845238,0.7380952,0.84126985,0.7638889,0.6944444,0.77380955,0.79761904,0.7380952,0.7936508,0.72619045,0.8035714,0.8174603,0.86507934,0.875,0.890873,0.859127,0.8452381,0.8769841,0.9007937,0.99206346
+LightGBM (tuned),0.10119048,0.16666667,0.16666667,0.17261904,0.17063493,0.16964285,0.17063493,0.16468254,0.23214285,0.2559524,0.3392857,0.27380952,0.2688492,0.37301588,0.33630952,0.30952382,0.37103173,0.33730158,0.35119048,0.40873015,0.40079364,0.4047619,0.41468254,0.47420636,0.4375,0.50396824,,0.54464287,0.46626985,0.5188492,0.4920635,0.515873,0.5595238,0.5595238,0.62400794,0.6656746,0.625,0.61706346,0.57539684,0.64880955,0.5625,0.5833333,0.5456349,0.6686508,0.70039684,0.6388889,0.6755952,0.6388889,0.6944444,0.6656746,0.68353176,0.71031743,0.6845238,0.6875,0.6944444,0.70039684,0.8015873,0.7519841,0.84424603,0.9077381,0.734127,0.8214286,0.8035714,0.7916667,0.8561508,0.77380955,0.86210316,0.8263889,0.796627,0.88095236,0.8293651,0.85515875,0.9047619,0.9027778,0.88293654,0.9325397,0.90674603,0.99404764
+CatBoost (tuned + ensembled),0.08134921,0.15079366,0.1468254,0.18253969,0.15277778,0.14880952,0.16269842,0.17063493,0.2559524,0.2559524,0.3015873,0.26190478,0.24801587,0.375,0.34920636,0.32142857,0.3809524,0.3313492,0.3472222,0.4047619,0.41269842,0.38690478,0.4027778,0.46031746,0.42261904,0.47420636,0.45535713,,0.48214287,0.49603173,0.5,0.51785713,0.5763889,0.54761904,0.56845236,0.59325397,0.5555556,0.5952381,0.5575397,0.59623015,0.5535714,0.5734127,0.56150794,0.61210316,0.64285713,0.6547619,0.6626984,0.63293654,0.66468257,0.6944444,0.6785714,0.6706349,0.64285713,0.7172619,0.6984127,0.6924603,0.71031743,0.8015873,0.8531746,0.85515875,0.6904762,0.8581349,0.77380955,0.8015873,0.828373,0.7519841,0.8174603,0.7916667,0.8214286,0.8373016,0.8075397,0.8630952,0.890873,0.91468257,0.8313492,0.890873,0.8968254,0.99206346
+ModernNCA (tuned),0.097222224,0.15079366,0.14484127,0.16468254,0.19444445,0.15277778,0.1904762,0.17063493,0.24404761,0.24206349,0.30952382,0.26190478,0.33730158,0.3531746,0.37797618,0.3888889,0.3392857,0.33730158,0.3829365,0.38690478,0.40376985,0.42857143,0.40674603,0.43055555,0.49603173,0.45436507,0.53373015,0.51785713,,0.5486111,0.48412699,0.5,0.53373015,0.5972222,0.5694444,0.5813492,0.5833333,0.5376984,0.63492066,0.5763889,0.58234125,0.5694444,0.54464287,0.625,0.63492066,0.6230159,0.67261904,0.6527778,0.6805556,0.6507937,0.71428573,0.68154764,0.66071427,0.6765873,0.6706349,0.6785714,0.73214287,0.7480159,0.75595236,0.7619048,0.765873,0.77380955,0.74404764,0.7956349,0.81349206,0.75396824,0.8095238,0.77380955,0.85515875,0.8313492,0.82043654,0.8194444,0.86507934,0.8472222,0.8214286,0.88293654,0.86706346,0.98015875
+xRFM (tuned + ensembled),0.08730159,0.1468254,0.1388889,0.16865079,0.17460318,0.16666667,0.18551587,0.1904762,0.25396827,0.23214285,0.31349206,0.24206349,0.29563493,0.36011904,0.39285713,0.3561508,0.36309522,0.34920636,0.33333334,0.4047619,0.4047619,0.4265873,0.45436507,0.4484127,0.45238096,0.4722222,0.4811508,0.50396824,0.4513889,,0.45238096,0.47718254,0.52281743,0.59623015,0.5367063,0.5843254,0.54761904,0.5436508,0.5535714,0.54960316,0.55654764,0.5456349,0.5515873,0.61507934,0.6388889,0.6160714,0.6269841,0.6388889,0.6825397,0.64285713,0.640873,0.68849206,0.6904762,0.6964286,0.66765875,0.6666667,0.7063492,0.75793654,0.7619048,0.7638889,0.78373015,0.8035714,0.7916667,0.8234127,0.78174603,0.74404764,0.8154762,0.8075397,0.9305556,0.84821427,0.88293654,0.8680556,0.9325397,0.93650794,0.84623015,0.91071427,0.89880955,0.99007934
+SAP-RPT-OSS (default),0.16269842,0.25992063,0.2797619,0.3015873,0.23809524,0.2470238,0.25198412,0.26190478,0.35912699,0.38392857,0.3690476,0.3313492,0.42063493,0.41468254,0.4107143,0.4702381,0.41666666,0.41765872,0.44047618,0.44444445,0.46230158,0.4781746,0.48809522,0.42460316,0.4861111,0.46825397,0.50793654,0.5,0.515873,0.54761904,,0.5297619,0.515873,0.5535714,0.5515873,0.5277778,0.5595238,0.56150794,0.56349206,0.5595238,0.6269841,0.6527778,0.5734127,0.58928573,0.60218257,0.5882937,0.6805556,0.67460316,0.5654762,0.63095236,0.6924603,0.59126985,0.60119045,0.63095236,0.6031746,0.5952381,0.640873,0.6994048,0.6706349,0.68849206,0.7123016,0.71130955,0.7519841,0.7519841,0.7202381,0.73015875,0.7619048,0.75595236,0.7619048,0.7668651,0.78174603,0.71626985,0.765873,0.7777778,0.78174603,0.79761904,0.82539684,0.98214287
+iLTM (tuned + ensembled),0.07936508,0.12301587,0.12301587,0.13293651,0.18055555,0.16170634,0.18055555,0.121031746,0.21726191,0.24404761,0.25198412,0.20337301,0.25992063,0.30555555,0.3482143,0.31944445,0.30853173,0.375,0.39087301,0.3611111,0.39484128,0.4077381,0.3968254,0.4107143,0.43948412,0.45833334,0.48412699,0.48214287,0.5,0.52281743,0.4702381,,0.5138889,0.56646824,0.54464287,0.5515873,0.56349206,0.51785713,0.5992063,0.5833333,0.58531743,0.54960316,0.5714286,0.609127,0.62896824,0.70535713,0.6626984,0.66964287,0.6845238,0.6617063,0.65873015,0.6805556,0.63789684,0.6944444,0.64285713,0.6438492,0.72718257,0.76984125,0.8055556,0.84126985,0.76984125,0.81150794,0.7043651,0.79960316,0.8154762,0.7480159,0.8293651,0.74404764,0.85714287,0.85515875,0.8373016,0.8214286,0.89484125,0.89484125,0.8075397,0.8968254,0.8660714,0.9781746
+CatBoost (tuned),0.07539683,0.14087301,0.14980158,0.17261904,0.15277778,0.15476191,0.16666667,0.17857143,0.26488096,0.24900794,0.31150794,0.25992063,0.24404761,0.3670635,0.3234127,0.2936508,0.36309522,0.29761904,0.32936507,0.39781746,0.37797618,0.37896827,0.38492063,0.43551588,0.41170636,0.4642857,0.44047618,0.4236111,0.46626985,0.47718254,0.48412699,0.4861111,,0.53373015,0.5674603,0.5882937,0.56150794,0.5763889,0.53968257,0.58928573,0.53174603,0.5734127,0.5515873,0.5972222,0.64285713,0.62103176,0.6468254,0.62400794,0.6180556,0.67460316,0.6656746,0.6458333,0.640873,0.68650794,0.6984127,0.70039684,0.70238096,0.7916667,0.8472222,0.8293651,0.6825397,0.8333333,0.76785713,0.7797619,0.81150794,0.75793654,0.8055556,0.7827381,0.8075397,0.8293651,0.79761904,0.8531746,0.87896824,0.9047619,0.8234127,0.89880955,0.9027778,0.99206346
+xRFM (tuned),0.10714286,0.15277778,0.16964285,0.18551587,0.16666667,0.17857143,0.1904762,0.1984127,0.2470238,0.24107143,0.2857143,0.25496033,0.31746033,0.34623015,0.3888889,0.32738096,0.3611111,0.35119048,0.31547618,0.39484128,0.39484128,0.39980158,0.43055555,0.44047618,0.40079364,0.46825397,0.44047618,0.45238096,0.4027778,0.40376985,0.44642857,0.43353173,0.46626985,,0.48412699,0.50793654,0.52281743,0.5257937,0.53571427,0.5069444,0.52876985,0.53571427,0.46230158,0.57738096,0.6111111,0.58531743,0.57738096,0.60714287,0.6438492,0.6369048,0.625,0.6666667,0.6765873,0.67460316,0.6805556,0.6845238,0.70238096,0.6805556,0.66071427,0.7123016,0.8015873,0.74007934,0.7916667,0.8194444,0.7956349,0.75,0.7827381,0.81349206,0.8888889,0.83035713,0.8611111,0.8531746,0.875,0.8769841,0.86706346,0.89880955,0.89880955,0.99007934
+XGBoost (tuned + ensembled),0.06746032,0.14980158,0.1388889,0.16468254,0.14484127,0.14087301,0.15476191,0.14087301,0.21329366,0.21130952,0.2857143,0.2281746,0.18253969,0.3422619,0.25396827,0.27380952,0.32539684,0.30257937,0.3234127,0.35218254,0.3392857,0.30753967,0.3343254,0.4027778,0.2936508,0.43055555,0.37599206,0.4315476,0.43055555,0.46329364,0.4484127,0.45535713,0.43253967,0.515873,,0.5327381,0.515873,0.53571427,0.5277778,0.54265875,0.5218254,0.53869045,0.5,0.58531743,0.60515875,0.5952381,0.6031746,0.57738096,0.5833333,0.60714287,0.61904764,0.62103176,0.6339286,0.65873015,0.67261904,0.67261904,0.65575397,0.77876985,0.8730159,0.8730159,0.69146824,0.80853176,0.75793654,0.75595236,0.76984125,0.734127,0.7757937,0.7718254,0.7936508,0.7986111,0.7718254,0.8174603,0.86706346,0.9027778,0.8214286,0.88095236,0.8779762,0.98214287
+CatBoost (default),0.0734127,0.125,0.13392857,0.14087301,0.11706349,0.13690476,0.14880952,0.125,0.1765873,0.21428572,0.2777778,0.22222222,0.2281746,0.31944445,0.25,0.2559524,0.29960316,0.2857143,0.3045635,0.3531746,0.30357143,0.32539684,0.31746033,0.39484128,0.30853173,0.46230158,0.3343254,0.40674603,0.4186508,0.4156746,0.4722222,0.4484127,0.41170636,0.4920635,0.4672619,,0.51984125,0.5257937,0.4920635,0.4890873,0.52380955,0.53373015,0.49801588,0.59126985,0.625,0.59325397,0.61706346,0.60119045,0.6299603,0.62896824,0.640873,0.65873015,0.6230159,0.64880955,0.69146824,0.6944444,0.73015875,0.74603176,0.765873,0.8035714,0.7043651,0.7876984,0.74206346,0.7876984,0.8630952,0.7619048,0.7916667,0.765873,0.765873,0.8333333,0.81349206,0.84325397,0.8611111,0.86904764,0.8472222,0.88293654,0.90873015,0.9880952
+EBM (tuned + ensembled),0.10515873,0.16071428,0.16071428,0.16666667,0.14087301,0.16468254,0.15277778,0.1904762,0.20039682,0.21527778,0.28769842,0.23214285,0.2281746,0.3343254,0.23809524,0.2936508,0.36507937,0.29761904,0.29563493,0.3531746,0.28174603,0.31944445,0.33333334,0.38492063,0.35416666,0.4702381,0.375,0.44444445,0.41666666,0.45238096,0.44047618,0.43650794,0.43849206,0.47718254,0.48412699,0.48015872,,0.51785713,0.5119048,0.4920635,0.49107143,0.53174603,0.47619048,0.67956346,0.70039684,0.5456349,0.58928573,0.5644841,0.5793651,0.63988096,0.61210316,0.5992063,0.6547619,0.6388889,0.67460316,0.6765873,0.6706349,0.7083333,0.70734125,0.71428573,0.7380952,0.72619045,0.79960316,0.74206346,0.8075397,0.75992066,0.78571427,0.8174603,0.75396824,0.80853176,0.8194444,0.8293651,0.8333333,0.83531743,0.85119045,0.8968254,0.8938492,0.98015875
+TabM (default),0.09920635,0.18055555,0.16269842,0.18849206,0.1468254,0.17460318,0.17063493,0.18154761,0.23412699,0.24404761,0.31349206,0.2857143,0.24404761,0.35813493,0.17757936,0.30555555,0.33333334,0.30357143,0.3015873,0.3690476,0.3313492,0.32936507,0.29960316,0.37698412,0.3482143,0.4265873,0.3829365,0.4047619,0.46230158,0.4563492,0.43849206,0.48214287,0.4236111,0.47420636,0.4642857,0.47420636,0.48214287,,0.51785713,0.4920635,0.5535714,0.5654762,0.5376984,0.5456349,0.546627,0.5644841,0.6111111,0.62896824,0.5575397,0.6111111,0.64880955,0.5793651,0.60714287,0.63492066,0.6924603,0.70039684,0.63492066,0.71031743,0.6666667,0.6547619,0.6944444,0.7242063,0.7638889,0.78174603,0.7718254,0.76984125,0.7569444,0.7847222,0.71428573,0.7638889,0.7916667,0.8055556,0.8095238,0.8293651,0.8293651,0.859127,0.89285713,0.9722222
+ModernNCA (default),0.0952381,0.1359127,0.15277778,0.1359127,0.16071428,0.16666667,0.17857143,0.15476191,0.21031746,0.23511904,0.23412699,0.23015873,0.28968254,0.2986111,0.31547618,0.30952382,0.31051588,0.3045635,0.3204365,0.32738096,0.34325397,0.3472222,0.3531746,0.37698412,0.4027778,0.4047619,0.42460316,0.44246033,0.36507937,0.44642857,0.43650794,0.40079364,0.46031746,0.4642857,0.4722222,0.50793654,0.48809522,0.48214287,,0.50595236,0.4781746,0.4781746,0.40674603,0.5456349,0.58531743,0.52281743,0.53968257,0.5436508,0.65674603,0.5952381,0.57738096,0.66468257,0.6527778,0.63492066,0.6507937,0.65674603,0.70039684,0.67460316,0.63293654,0.70039684,0.74404764,0.74007934,0.74404764,0.78373015,0.78373015,0.734127,0.77380955,0.7688492,0.73214287,0.8055556,0.8095238,0.8472222,0.81845236,0.8154762,0.859127,0.85714287,0.87896824,0.984127
+XGBoost (tuned),0.06746032,0.13690476,0.14087301,0.15674603,0.125,0.1279762,0.1468254,0.13492064,0.21825397,0.20238096,0.2777778,0.21031746,0.17261904,0.31746033,0.23214285,0.28769842,0.31150794,0.30357143,0.31349206,0.34920636,0.32539684,0.28869048,0.32539684,0.37103173,0.28174603,0.41468254,0.35119048,0.40376985,0.4236111,0.45039684,0.44047618,0.41666666,0.4107143,0.49305555,0.45734128,0.5109127,0.50793654,0.50793654,0.4940476,,0.4940476,0.5218254,0.49801588,0.5575397,0.57738096,0.5793651,0.57539684,0.57738096,0.5486111,0.6140873,0.5952381,0.5952381,0.6130952,0.6547619,0.65674603,0.6597222,0.61904764,0.75793654,0.83531743,0.8373016,0.6805556,0.8055556,0.74404764,0.73214287,0.765873,0.73015875,0.7480159,0.76984125,0.78571427,0.78968257,0.76785713,0.81845236,0.86904764,0.8869048,0.81349206,0.87896824,0.875,0.9781746
+FastaiMLP (tuned + ensembled),0.103174604,0.16964285,0.16269842,0.16468254,0.13293651,0.1656746,0.17857143,0.17460318,0.20634921,0.22916667,0.23015873,0.2827381,0.24305555,0.30357143,0.30555555,0.27579364,0.32142857,0.2986111,0.29166666,0.3313492,0.3482143,0.33531746,0.35416666,0.2936508,0.40873015,0.41468254,0.4375,0.44642857,0.41765872,0.4434524,0.37301588,0.41468254,0.46825397,0.47123015,0.4781746,0.47619048,0.5089286,0.44642857,0.5218254,0.50595236,,0.43650794,0.50793654,0.54761904,0.5555556,0.5119048,0.578373,0.6845238,0.60515875,0.60515875,0.57539684,0.6230159,0.56150794,0.6111111,0.6388889,0.64484125,0.6269841,0.70734125,0.66468257,0.6736111,0.75396824,0.71626985,0.75,0.7748016,0.74206346,0.74404764,0.74404764,0.78174603,0.74603176,0.7757937,0.9484127,0.7916667,0.8075397,0.78174603,0.8373016,0.86507934,0.87896824,0.97619045
+TorchMLP (tuned + ensembled),0.055555556,0.103174604,0.11309524,0.12301587,0.11309524,0.121031746,0.121031746,0.12301587,0.17063493,0.20039682,0.25992063,0.24404761,0.26785713,0.30257937,0.28174603,0.24801587,0.33730158,0.2718254,0.31349206,0.31547618,0.31746033,0.3234127,0.31944445,0.34325397,0.40873015,0.41269842,0.41666666,0.4265873,0.43055555,0.45436507,0.3472222,0.45039684,0.4265873,0.4642857,0.46130952,0.46626985,0.46825397,0.43452382,0.5218254,0.4781746,0.56349206,,0.52380955,0.51785713,0.5515873,0.5625,0.58928573,0.6617063,0.5555556,0.61904764,0.7380952,0.5873016,0.5654762,0.6230159,0.6527778,0.6617063,0.60714287,0.6964286,0.6944444,0.6805556,0.7380952,0.71626985,0.7430556,0.8531746,0.77876985,0.75,0.7757937,0.78373015,0.7638889,0.8194444,0.84623015,0.78571427,0.76984125,0.76785713,0.84325397,0.88095236,0.8730159,0.97619045
+ModernNCA (tuned + ensembled),0.055555556,0.11309524,0.09126984,0.13690476,0.16468254,0.121031746,0.12996031,0.09126984,0.19642857,0.1984127,0.26785713,0.22222222,0.28769842,0.26984128,0.29960316,0.32142857,0.2857143,0.30357143,0.35912699,0.3065476,0.3809524,0.3531746,0.3452381,0.37896827,0.4265873,0.40674603,0.45436507,0.43849206,0.45535713,0.4484127,0.4265873,0.42857143,0.4484127,0.5376984,0.5,0.5019841,0.52380955,0.46230158,0.59325397,0.5019841,0.4920635,0.47619048,,0.56646824,0.56349206,0.5515873,0.609127,0.5714286,0.5674603,0.5813492,0.58531743,0.59126985,0.5595238,0.6230159,0.59126985,0.59325397,0.5952381,0.66468257,0.71825397,0.68650794,0.6626984,0.67261904,0.64285713,0.6984127,0.6984127,0.6547619,0.7083333,0.65873015,0.80654764,0.71031743,0.7242063,0.7361111,0.7956349,0.78174603,0.70238096,0.78968257,0.76984125,0.9404762
+EBM (tuned),0.09325397,0.1468254,0.14880952,0.16071428,0.121031746,0.15178572,0.14880952,0.16170634,0.18253969,0.20238096,0.25496033,0.20436507,0.22023809,0.3045635,0.22420634,0.27380952,0.33333334,0.2777778,0.2777778,0.32539684,0.24206349,0.2797619,0.29563493,0.35714287,0.31349206,0.4434524,0.3313492,0.38789684,0.375,0.38492063,0.4107143,0.39087301,0.4027778,0.42261904,0.41468254,0.40873015,0.3204365,0.45436507,0.45436507,0.44246033,0.45238096,0.48214287,0.43353173,,0.62400794,0.4920635,0.515873,0.5367063,0.53968257,0.58531743,0.5793651,0.55654764,0.61210316,0.61904764,0.65873015,0.6656746,0.6269841,0.6706349,0.66468257,0.67261904,0.71130955,0.71031743,0.765873,0.71626985,0.7638889,0.74007934,0.7619048,0.78571427,0.7281746,0.78373015,0.7876984,0.8015873,0.82738096,0.8174603,0.82539684,0.875,0.89484125,0.98015875
+EBM (default),0.09920635,0.13392857,0.14285715,0.14087301,0.1388889,0.15873016,0.16468254,0.16865079,0.17261904,0.20039682,0.24404761,0.20238096,0.22123016,0.30753967,0.23115079,0.2361111,0.30357143,0.26785713,0.26190478,0.31746033,0.21825397,0.31746033,0.28174603,0.35119048,0.2638889,0.40873015,0.29960316,0.35714287,0.36507937,0.3611111,0.39781746,0.37103173,0.35714287,0.3888889,0.39484128,0.375,0.29960316,0.45337301,0.41468254,0.42261904,0.44444445,0.4484127,0.43650794,0.37599206,,0.4484127,0.49801588,0.51984125,0.53373015,0.56349206,0.5813492,0.54960316,0.609127,0.5714286,0.66071427,0.65873015,0.59325397,0.65674603,0.6031746,0.63492066,0.71825397,0.6825397,0.7480159,0.7083333,0.7638889,0.7361111,0.74007934,0.7827381,0.6924603,0.75992066,0.7956349,0.8313492,0.81349206,0.7956349,0.85714287,0.85119045,0.9126984,0.9781746
+iLTM (tuned),0.065476194,0.084325396,0.09126984,0.115079366,0.09623016,0.103174604,0.109126985,0.0952381,0.15079366,0.17460318,0.17757936,0.17261904,0.18849206,0.22023809,0.25,0.2281746,0.23412699,0.24900794,0.30753967,0.26984128,0.32936507,0.2857143,0.3234127,0.29563493,0.3313492,0.3690476,0.3611111,0.3452381,0.37698412,0.38392857,0.41170636,0.29464287,0.37896827,0.41468254,0.4047619,0.40674603,0.45436507,0.43551588,0.47718254,0.42063493,0.48809522,0.4375,0.4484127,0.50793654,0.5515873,,0.5456349,0.5416667,0.51984125,0.53968257,0.56349206,0.5644841,0.53373015,0.5873016,0.6299603,0.6230159,0.5873016,0.6527778,0.6369048,0.6468254,0.66765875,0.70238096,0.6319444,0.7638889,0.7430556,0.7063492,0.7043651,0.67460316,0.75396824,0.75396824,0.78571427,0.74404764,0.78373015,0.78571427,0.75595236,0.82738096,0.81845236,0.96428573
+RealMLP (default),0.08134921,0.097222224,0.13293651,0.11904762,0.10119048,0.13690476,0.13690476,0.13492064,0.1845238,0.18849206,0.20833333,0.1904762,0.15376984,0.2718254,0.21130952,0.22619048,0.2718254,0.24107143,0.24007936,0.29761904,0.25992063,0.25396827,0.2559524,0.30357143,0.31150794,0.4156746,0.32440478,0.33730158,0.32738096,0.37301588,0.31944445,0.33730158,0.3531746,0.42261904,0.3968254,0.3829365,0.4107143,0.3888889,0.46031746,0.42460316,0.42162699,0.4107143,0.39087301,0.48412699,0.5019841,0.45436507,,0.52380955,0.51785713,0.5456349,0.5555556,0.53571427,0.5515873,0.5873016,0.63492066,0.6507937,0.57539684,0.6468254,0.61507934,0.64484125,0.73511904,0.6805556,0.71031743,0.74900794,0.734127,0.7242063,0.74603176,0.7519841,0.7043651,0.79761904,0.8472222,0.79761904,0.7876984,0.78373015,0.8333333,0.875,0.86706346,0.9781746
+FastaiMLP (tuned),0.077380955,0.116071425,0.13492064,0.121031746,0.109126985,0.125,0.13492064,0.13293651,0.16071428,0.20535715,0.17261904,0.2093254,0.21924603,0.24801587,0.25,0.21825397,0.2638889,0.26587301,0.25198412,0.27380952,0.27579364,0.26587301,0.28373015,0.2638889,0.31944445,0.34126985,0.3611111,0.3670635,0.3472222,0.3611111,0.32539684,0.33035713,0.37599206,0.39285713,0.42261904,0.39880952,0.43551588,0.37103173,0.4563492,0.42261904,0.31547618,0.33829364,0.42857143,0.46329364,0.48015872,0.45833334,0.47619048,,0.5367063,0.5456349,0.49007937,0.55853176,0.5,0.56150794,0.5952381,0.59325397,0.58928573,0.6498016,0.59126985,0.60714287,0.6845238,0.6547619,0.6984127,0.7013889,0.6924603,0.7063492,0.6765873,0.7251984,0.67460316,0.7202381,0.8611111,0.7638889,0.75793654,0.73214287,0.79960316,0.8015873,0.8492063,0.96825397
+ChimeraBoost (tuned + ensembled),0.071428575,0.09126984,0.08730159,0.10714286,0.097222224,0.10813492,0.11805555,0.08928572,0.13293651,0.16468254,0.1984127,0.15277778,0.17857143,0.24404761,0.1904762,0.21527778,0.21626984,0.23412699,0.28174603,0.26587301,0.24900794,0.29166666,0.23809524,0.33730158,0.25297618,0.34920636,0.30555555,0.33531746,0.31944445,0.31746033,0.43452382,0.31547618,0.38194445,0.3561508,0.41666666,0.37003967,0.42063493,0.44246033,0.34325397,0.4513889,0.39484128,0.44444445,0.43253967,0.46031746,0.46626985,0.48015872,0.48214287,0.46329364,,0.49801588,0.5277778,0.56646824,0.5714286,0.5456349,0.640873,0.640873,0.75595236,0.60515875,0.6388889,0.70535713,0.6984127,0.65873015,0.6547619,0.6626984,0.73015875,0.73015875,0.72619045,0.6785714,0.7242063,0.76984125,0.7519841,0.73015875,0.7718254,0.75396824,0.7916667,0.82539684,0.83928573,0.9751984
+ExtraTrees (tuned + ensembled),0.061507937,0.07539683,0.0952381,0.1279762,0.13690476,0.14087301,0.13789682,0.11706349,0.16865079,0.1736111,0.16369048,0.15873016,0.20238096,0.21626984,0.26984128,0.29166666,0.20238096,0.2718254,0.25396827,0.2638889,0.27380952,0.26984128,0.31746033,0.31944445,0.31051588,0.31150794,0.3343254,0.30555555,0.34920636,0.35714287,0.3690476,0.33829364,0.32539684,0.36309522,0.39285713,0.37103173,0.36011904,0.3888889,0.4047619,0.3859127,0.39484128,0.3809524,0.4186508,0.41468254,0.43650794,0.46031746,0.45436507,0.45436507,0.5019841,,0.45734128,0.52380955,0.56150794,0.65674603,0.5297619,0.54761904,0.53373015,0.71626985,0.56349206,0.59325397,0.6269841,0.7757937,0.6319444,0.61507934,0.61904764,0.67757934,0.6547619,0.6458333,0.7063492,0.67261904,0.68650794,0.8452381,0.89285713,0.85119045,0.74404764,0.74404764,0.88095236,0.99007934
+TorchMLP (tuned),0.04761905,0.083333336,0.07936508,0.09126984,0.083333336,0.08928572,0.08928572,0.09126984,0.11111111,0.14087301,0.1984127,0.18650794,0.21428572,0.24007936,0.2013889,0.19444445,0.23412699,0.21230158,0.2559524,0.27380952,0.24206349,0.25396827,0.25099206,0.28373015,0.3015873,0.35714287,0.31646827,0.32142857,0.2857143,0.35912699,0.30753967,0.34126985,0.3343254,0.375,0.3809524,0.35912699,0.38789684,0.35119048,0.42261904,0.4047619,0.42460316,0.26190478,0.41468254,0.42063493,0.4186508,0.43650794,0.44444445,0.50992066,0.4722222,0.54265875,,0.5,0.49603173,0.54761904,0.6369048,0.6388889,0.53968257,0.61904764,0.57738096,0.58035713,0.71428573,0.6468254,0.64880955,0.76289684,0.70238096,0.7202381,0.6805556,0.67757934,0.6944444,0.7242063,0.7757937,0.7390873,0.72619045,0.7242063,0.8035714,0.81349206,0.8174603,0.96626985
+ChimeraBoost (tuned),0.061507937,0.09126984,0.08531746,0.09325397,0.0952381,0.10119048,0.10714286,0.08730159,0.12896825,0.15476191,0.18253969,0.14484127,0.16964285,0.23015873,0.16468254,0.20436507,0.20436507,0.21825397,0.26984128,0.2579365,0.23809524,0.26587301,0.24603175,0.30753967,0.24007936,0.3472222,0.28968254,0.32936507,0.3184524,0.31150794,0.40873015,0.31944445,0.35416666,0.33333334,0.37896827,0.34126985,0.40079364,0.42063493,0.33531746,0.4047619,0.37698412,0.41269842,0.40873015,0.4434524,0.45039684,0.43551588,0.4642857,0.44146827,0.43353173,0.47619048,0.5,,0.56150794,0.53174603,0.6269841,0.6269841,0.6924603,0.5813492,0.6220238,0.6706349,0.66468257,0.63293654,0.6369048,0.6547619,0.6984127,0.71626985,0.68154764,0.64880955,0.6964286,0.72321427,0.734127,0.7083333,0.75992066,0.73015875,0.7718254,0.8144841,0.81349206,0.96428573
+OrionMSP (default),0.06944445,0.08730159,0.084325396,0.09126984,0.07936508,0.109126985,0.115079366,0.1388889,0.20337301,0.22619048,0.1388889,0.17956349,0.2559524,0.1984127,0.28373015,0.2906746,0.20634921,0.29960316,0.25198412,0.18849206,0.28174603,0.3204365,0.31150794,0.15674603,0.30357143,0.25396827,0.31547618,0.35714287,0.3392857,0.30952382,0.39880952,0.36210316,0.35912699,0.3234127,0.36607143,0.37698412,0.3452381,0.39285713,0.3472222,0.38690478,0.43849206,0.43452382,0.44047618,0.38789684,0.39087301,0.46626985,0.4484127,0.5,0.42857143,0.43849206,0.50396824,0.43849206,,0.45039684,0.5535714,0.5555556,0.50595236,0.5,0.48412699,0.5555556,0.66071427,0.53968257,0.64880955,0.6111111,0.5972222,0.7083333,0.67460316,0.6755952,0.64880955,0.71626985,0.7361111,0.6765873,0.6686508,0.6805556,0.78968257,0.8055556,0.8174603,0.97619045
+ExtraTrees (tuned),0.071428575,0.077380955,0.109126985,0.12301587,0.125,0.1388889,0.13095239,0.1279762,0.15773809,0.16071428,0.14087301,0.16071428,0.1875,0.20039682,0.27380952,0.25992063,0.19345239,0.2281746,0.23809524,0.24404761,0.2579365,0.2579365,0.30952382,0.30555555,0.30555555,0.2857143,0.3125,0.2827381,0.3234127,0.30357143,0.3690476,0.30555555,0.31349206,0.32539684,0.34126985,0.35119048,0.3611111,0.36507937,0.36507937,0.3452381,0.3888889,0.37698412,0.37698412,0.3809524,0.42857143,0.41269842,0.41269842,0.43849206,0.45436507,0.34325397,0.45238096,0.46825397,0.54960316,,0.5327381,0.54761904,0.52380955,0.5992063,0.5,0.54960316,0.6230159,0.68154764,0.6111111,0.5952381,0.6111111,0.67460316,0.63095236,0.62103176,0.65674603,0.6527778,0.6765873,0.8075397,0.8075397,0.80257934,0.734127,0.7222222,0.88492066,0.9880952
+TabSTAR (tuned),0.17460318,0.1875,0.22222222,0.19642857,0.19642857,0.22222222,0.23015873,0.22420634,0.24603175,0.26587301,0.26984128,0.23511904,0.27380952,0.3015873,0.27380952,0.27579364,0.28968254,0.28373015,0.25396827,0.3125,0.24801587,0.28968254,0.2936508,0.2559524,0.2906746,0.3670635,0.30555555,0.3015873,0.32936507,0.33234128,0.3968254,0.35714287,0.3015873,0.31944445,0.32738096,0.30853173,0.32539684,0.30753967,0.34920636,0.34325397,0.3611111,0.3472222,0.40873015,0.34126985,0.3392857,0.37003967,0.36507937,0.4047619,0.35912699,0.4702381,0.36309522,0.37301588,0.44642857,0.4672619,,0.58035713,0.4027778,0.5119048,0.44444445,0.45436507,0.5218254,0.54960316,0.60714287,0.4722222,0.43253967,0.93849206,0.58928573,0.640873,0.5654762,0.60714287,0.6527778,0.7083333,0.68650794,0.70238096,0.75,0.640873,0.8492063,0.9543651
+TabSTAR (tuned + ensembled),0.17460318,0.18253969,0.22023809,0.1984127,0.19543651,0.21626984,0.22222222,0.21428572,0.24404761,0.26190478,0.26190478,0.24206349,0.26587301,0.29960316,0.26587301,0.26785713,0.28769842,0.28968254,0.25892857,0.31349206,0.24404761,0.28373015,0.2827381,0.25,0.28769842,0.3670635,0.29960316,0.30753967,0.32142857,0.33333334,0.4047619,0.3561508,0.29960316,0.31547618,0.32738096,0.30555555,0.3234127,0.29960316,0.34325397,0.3402778,0.35515872,0.33829364,0.40674603,0.3343254,0.34126985,0.37698412,0.34920636,0.40674603,0.35912699,0.45238096,0.3611111,0.37301588,0.44444445,0.45238096,0.41964287,,0.39484128,0.50595236,0.44047618,0.46230158,0.5119048,0.5456349,0.5952381,0.46825397,0.43849206,0.9543651,0.5833333,0.6369048,0.56150794,0.5992063,0.64880955,0.6944444,0.6785714,0.6984127,0.74206346,0.6388889,0.83928573,0.9543651
+ChimeraBoost (default),0.05357143,0.09126984,0.083333336,0.097222224,0.08730159,0.10119048,0.103174604,0.08134921,0.11111111,0.1388889,0.16865079,0.12896825,0.16865079,0.20634921,0.13293651,0.18055555,0.1845238,0.18849206,0.26190478,0.22023809,0.18253969,0.2361111,0.18551587,0.22420634,0.15873016,0.28968254,0.1984127,0.28968254,0.26785713,0.2936508,0.35912699,0.27281746,0.29761904,0.29761904,0.34424603,0.26984128,0.32936507,0.36507937,0.29960316,0.3809524,0.37301588,0.39285713,0.4047619,0.37301588,0.40674603,0.41269842,0.42460316,0.4107143,0.24404761,0.46626985,0.46031746,0.30753967,0.4940476,0.47619048,0.5972222,0.60515875,,0.54761904,0.5505952,0.5843254,0.57738096,0.57539684,0.62103176,0.59325397,0.65674603,0.6785714,0.63492066,0.6369048,0.640873,0.6666667,0.6626984,0.6438492,0.61507934,0.6785714,0.765873,0.75396824,0.82539684,0.9702381
+RandomForest (tuned + ensembled),0.043650795,0.07936508,0.077380955,0.08531746,0.11309524,0.097222224,0.10714286,0.09920635,0.12896825,0.11904762,0.125,0.103174604,0.11309524,0.14285715,0.21031746,0.19246031,0.15476191,0.23015873,0.20833333,0.20039682,0.20238096,0.19246031,0.22619048,0.2777778,0.20238096,0.28769842,0.24801587,0.1984127,0.25198412,0.24206349,0.30059522,0.23015873,0.20833333,0.31944445,0.22123016,0.25396827,0.29166666,0.28968254,0.32539684,0.24206349,0.29265872,0.30357143,0.33531746,0.32936507,0.34325397,0.3472222,0.3531746,0.35019842,0.39484128,0.28373015,0.3809524,0.4186508,0.5,0.40079364,0.48809522,0.4940476,0.45238096,,0.48214287,0.50595236,0.5535714,0.6656746,0.57539684,0.54761904,0.50793654,0.62400794,0.5793651,0.5813492,0.67261904,0.6111111,0.6130952,0.73214287,0.7757937,0.84325397,0.69345236,0.67460316,0.8373016,0.96825397
+XGBoost (default),0.041666668,0.041666668,0.05357143,0.0734127,0.06944445,0.05952381,0.065476194,0.04563492,0.07936508,0.09325397,0.13293651,0.10119048,0.08531746,0.15674603,0.11904762,0.12301587,0.1468254,0.20436507,0.21428572,0.20833333,0.19246031,0.16666667,0.16865079,0.26190478,0.12698413,0.31547618,0.15575397,0.1468254,0.24404761,0.23809524,0.32936507,0.19444445,0.15277778,0.3392857,0.12698413,0.23412699,0.29265872,0.33333334,0.3670635,0.16468254,0.33531746,0.30555555,0.28174603,0.33531746,0.3968254,0.36309522,0.38492063,0.40873015,0.3611111,0.43650794,0.42261904,0.37797618,0.515873,0.5,0.5555556,0.5595238,0.44940478,0.51785713,,0.55654764,0.60119045,0.5873016,0.59325397,0.59325397,0.6507937,0.66468257,0.5714286,0.61507934,0.6785714,0.62103176,0.6666667,0.71626985,0.7619048,0.7757937,0.7361111,0.7222222,0.7916667,0.9464286
+LightGBM (default),0.04563492,0.05952381,0.05357143,0.06349207,0.055555556,0.06349207,0.06349207,0.055555556,0.08134921,0.07936508,0.12698413,0.08730159,0.09920635,0.14880952,0.116071425,0.09920635,0.10813492,0.17063493,0.17460318,0.1845238,0.18154761,0.16269842,0.18253969,0.18948413,0.057539683,0.26190478,0.0922619,0.14484127,0.23809524,0.2361111,0.31150794,0.15873016,0.17063493,0.28769842,0.12698413,0.19642857,0.2857143,0.3452381,0.29960316,0.16269842,0.3263889,0.31944445,0.31349206,0.32738096,0.36507937,0.3531746,0.35515872,0.39285713,0.29464287,0.40674603,0.41964287,0.32936507,0.44444445,0.45039684,0.5456349,0.5376984,0.4156746,0.4940476,0.4434524,,0.5714286,0.53869045,0.5714286,0.5833333,0.61507934,0.640873,0.5654762,0.59325397,0.6369048,0.6230159,0.6468254,0.7043651,0.6845238,0.703373,0.734127,0.73214287,0.79960316,0.9632937
+TabFlex (default),0.0734127,0.083333336,0.103174604,0.10515873,0.08630952,0.109126985,0.11904762,0.10515873,0.14087301,0.1765873,0.09126984,0.18849206,0.20436507,0.17063493,0.20238096,0.1904762,0.1765873,0.20833333,0.14880952,0.17460318,0.2013889,0.22123016,0.21626984,0.11111111,0.25396827,0.15873016,0.26587301,0.30952382,0.23412699,0.21626984,0.28769842,0.23015873,0.31746033,0.1984127,0.30853173,0.29563493,0.26190478,0.30555555,0.2559524,0.31944445,0.24603175,0.26190478,0.33730158,0.28869048,0.28174603,0.33234128,0.26488096,0.31547618,0.3015873,0.37301588,0.2857143,0.33531746,0.3392857,0.37698412,0.4781746,0.48809522,0.42261904,0.44642857,0.39880952,0.42857143,,0.4642857,0.5376984,0.4652778,0.48015872,0.6269841,0.5277778,0.56845236,0.5277778,0.58531743,0.62896824,0.60714287,0.5555556,0.5992063,0.68849206,0.63492066,0.75396824,0.9424603
+RandomForest (tuned),0.04563492,0.057539683,0.0734127,0.08730159,0.065476194,0.08531746,0.09126984,0.077380955,0.10119048,0.10119048,0.121031746,0.08928572,0.08134921,0.12896825,0.15873016,0.15178572,0.121031746,0.17956349,0.16269842,0.17460318,0.19246031,0.16865079,0.20833333,0.21825397,0.14285715,0.2361111,0.17857143,0.14186507,0.22619048,0.19642857,0.28869048,0.18849206,0.16666667,0.25992063,0.19146825,0.21230158,0.27380952,0.27579364,0.25992063,0.19444445,0.28373015,0.28373015,0.32738096,0.28968254,0.31746033,0.29761904,0.31944445,0.3452381,0.34126985,0.22420634,0.3531746,0.3670635,0.46031746,0.3184524,0.45039684,0.45436507,0.42460316,0.3343254,0.41269842,0.46130952,0.53571427,,0.5575397,0.5,0.4781746,0.5972222,0.5555556,0.5654762,0.6230159,0.5952381,0.609127,0.6994048,0.7043651,0.7291667,0.6845238,0.6547619,0.82539684,0.9702381
+Linear (tuned + ensembled),0.06349207,0.0952381,0.11706349,0.08134921,0.09920635,0.10119048,0.115079366,0.115079366,0.11111111,0.13492064,0.15873016,0.1388889,0.15277778,0.21428572,0.14484127,0.15773809,0.20238096,0.17063493,0.13988096,0.22519842,0.121031746,0.21230158,0.16071428,0.23214285,0.18849206,0.30555555,0.19642857,0.22619048,0.2559524,0.20833333,0.24801587,0.29563493,0.23214285,0.20833333,0.24206349,0.2579365,0.20039682,0.2361111,0.2559524,0.2559524,0.25,0.25694445,0.35714287,0.23412699,0.25198412,0.36805555,0.28968254,0.3015873,0.3452381,0.36805555,0.35119048,0.36309522,0.35119048,0.3888889,0.39285713,0.4047619,0.37896827,0.42460316,0.40674603,0.42857143,0.46230158,0.44246033,,0.43452382,0.43849206,0.54960316,0.4781746,0.66964287,0.49603173,0.49801588,0.48809522,0.59126985,0.5555556,0.53968257,0.78571427,0.5714286,0.71626985,0.9047619
+TorchMLP (default),0.03968254,0.049603175,0.049603175,0.065476194,0.037698414,0.061507937,0.05357143,0.05952381,0.08531746,0.09920635,0.115079366,0.11111111,0.09920635,0.13095239,0.10714286,0.12301587,0.14285715,0.16666667,0.14484127,0.14880952,0.14087301,0.13293651,0.125,0.15873016,0.19444445,0.22619048,0.20833333,0.1984127,0.20436507,0.1765873,0.24801587,0.20039682,0.22023809,0.18055555,0.24404761,0.21230158,0.2579365,0.21825397,0.21626984,0.26785713,0.22519842,0.1468254,0.3015873,0.28373015,0.29166666,0.2361111,0.25099206,0.2986111,0.33730158,0.38492063,0.23710318,0.3452381,0.3888889,0.4047619,0.5277778,0.53174603,0.40674603,0.45238096,0.40674603,0.41666666,0.5347222,0.5,0.5654762,,0.48214287,0.6468254,0.53174603,0.59325397,0.5734127,0.5813492,0.6686508,0.6626984,0.6269841,0.61507934,0.7083333,0.67261904,0.7876984,0.9484127
+iLTM (default),0.06944445,0.061507937,0.06746032,0.083333336,0.06746032,0.077380955,0.08134921,0.077380955,0.097222224,0.12896825,0.13690476,0.10714286,0.109126985,0.14285715,0.11904762,0.13492064,0.15476191,0.15476191,0.17261904,0.16865079,0.10119048,0.15277778,0.13492064,0.18849206,0.12698413,0.20238096,0.14384921,0.17162699,0.18650794,0.21825397,0.2797619,0.1845238,0.18849206,0.20436507,0.23015873,0.13690476,0.19246031,0.2281746,0.21626984,0.23412699,0.2579365,0.22123016,0.3015873,0.2361111,0.2361111,0.25694445,0.26587301,0.30753967,0.26984128,0.3809524,0.29761904,0.3015873,0.4027778,0.3888889,0.5674603,0.56150794,0.34325397,0.4920635,0.34920636,0.38492063,0.51984125,0.5218254,0.56150794,0.51785713,,0.63492066,0.52380955,0.5833333,0.5555556,0.5456349,0.5813492,0.6706349,0.59325397,0.65873015,0.68650794,0.63095236,0.7916667,0.92460316
+TabSTAR (default),0.12896825,0.1468254,0.17460318,0.14484127,0.16269842,0.1845238,0.1845238,0.1845238,0.21031746,0.21428572,0.20039682,0.18948413,0.22023809,0.2281746,0.21626984,0.21825397,0.23015873,0.2281746,0.20238096,0.23412699,0.1875,0.23412699,0.2361111,0.1904762,0.21031746,0.26190478,0.22619048,0.24801587,0.24603175,0.2559524,0.26984128,0.25198412,0.24206349,0.25,0.26587301,0.23809524,0.24007936,0.23015873,0.26587301,0.26984128,0.2559524,0.25,0.3452381,0.25992063,0.2638889,0.2936508,0.27579364,0.2936508,0.26984128,0.32242063,0.2797619,0.28373015,0.29166666,0.32539684,0.061507937,0.04563492,0.32142857,0.37599206,0.33531746,0.35912699,0.37301588,0.4027778,0.45039684,0.3531746,0.36507937,,0.5019841,0.46626985,0.45436507,0.5138889,0.42857143,0.5654762,0.52380955,0.48015872,0.6111111,0.5515873,0.7063492,0.9424603
+PerpetualBooster (tuned + ensembled),0.017857144,0.06349207,0.055555556,0.057539683,0.05952381,0.057539683,0.07440476,0.06746032,0.10714286,0.10714286,0.115079366,0.0952381,0.077380955,0.13293651,0.10714286,0.08531746,0.10714286,0.16468254,0.14880952,0.15476191,0.14484127,0.13095239,0.109126985,0.12896825,0.12301587,0.20634921,0.13789682,0.18253969,0.1904762,0.1845238,0.23809524,0.17063493,0.19444445,0.21726191,0.22420634,0.20833333,0.21428572,0.24305555,0.22619048,0.25198412,0.2559524,0.22420634,0.29166666,0.23809524,0.25992063,0.29563493,0.25396827,0.3234127,0.27380952,0.3452381,0.31944445,0.3184524,0.32539684,0.3690476,0.4107143,0.41666666,0.36507937,0.42063493,0.42857143,0.43452382,0.4722222,0.44444445,0.5218254,0.46825397,0.47619048,0.49801588,,0.53968257,0.53174603,0.6875,0.5218254,0.55853176,0.58928573,0.5873016,0.62103176,0.7757937,0.6944444,0.9404762
+Linear (tuned),0.055555556,0.08730159,0.09126984,0.07539683,0.09126984,0.08531746,0.09920635,0.09126984,0.09325397,0.11706349,0.14880952,0.11706349,0.15079366,0.19246031,0.1279762,0.12996031,0.18253969,0.15873016,0.1388889,0.21031746,0.10119048,0.20337301,0.14880952,0.20634921,0.16666667,0.27380952,0.1736111,0.20833333,0.22619048,0.19246031,0.24404761,0.2559524,0.21726191,0.18650794,0.2281746,0.23412699,0.18253969,0.21527778,0.23115079,0.23015873,0.21825397,0.21626984,0.34126985,0.21428572,0.21726191,0.32539684,0.24801587,0.27480158,0.32142857,0.35416666,0.32242063,0.35119048,0.32440478,0.37896827,0.35912699,0.36309522,0.36309522,0.4186508,0.38492063,0.40674603,0.4315476,0.43452382,0.33035713,0.40674603,0.41666666,0.53373015,0.46031746,,0.4920635,0.48412699,0.4702381,0.55654764,0.53968257,0.5119048,0.6984127,0.5654762,0.6964286,0.8968254
+xRFM (default),0.035714287,0.031746034,0.05357143,0.05952381,0.051587302,0.05952381,0.057539683,0.05952381,0.09920635,0.083333336,0.11111111,0.065476194,0.08531746,0.10218254,0.14087301,0.12301587,0.13095239,0.15277778,0.15079366,0.1388889,0.21031746,0.1388889,0.1845238,0.19444445,0.19642857,0.19642857,0.20337301,0.17857143,0.14484127,0.06944445,0.23809524,0.14285715,0.19246031,0.11111111,0.20634921,0.23412699,0.24603175,0.2857143,0.26785713,0.21428572,0.25396827,0.2361111,0.19345239,0.2718254,0.30753967,0.24603175,0.29563493,0.32539684,0.27579364,0.2936508,0.30555555,0.30357143,0.35119048,0.34325397,0.43452382,0.43849206,0.35912699,0.32738096,0.32142857,0.36309522,0.4722222,0.37698412,0.50396824,0.4265873,0.44444445,0.5456349,0.46825397,0.50793654,,0.49503967,0.50595236,0.50595236,0.4940476,0.53373015,0.60515875,0.5873016,0.6527778,0.9047619
+PerpetualBooster (tuned),0.021825397,0.049603175,0.05059524,0.051587302,0.05357143,0.057539683,0.06349207,0.071428575,0.08730159,0.08134921,0.0952381,0.08531746,0.071428575,0.11805555,0.08730159,0.07936508,0.097222224,0.16170634,0.14880952,0.14285715,0.12301587,0.12896825,0.10119048,0.12896825,0.09920635,0.18253969,0.11904762,0.16269842,0.16865079,0.15178572,0.23313493,0.14484127,0.17063493,0.16964285,0.2013889,0.16666667,0.19146825,0.2361111,0.19444445,0.21031746,0.22420634,0.18055555,0.28968254,0.21626984,0.24007936,0.24603175,0.20238096,0.2797619,0.23015873,0.32738096,0.27579364,0.2767857,0.28373015,0.3472222,0.39285713,0.40079364,0.33333334,0.3888889,0.37896827,0.37698412,0.41468254,0.4047619,0.5019841,0.4186508,0.45436507,0.4861111,0.3125,0.515873,0.5049603,,0.4920635,0.5257937,0.54761904,0.5555556,0.5992063,0.6845238,0.67261904,0.91468257
+FastaiMLP (default),0.03373016,0.041666668,0.055555556,0.049603175,0.05357143,0.05952381,0.06746032,0.06349207,0.083333336,0.08730159,0.077380955,0.07936508,0.071428575,0.09920635,0.09920635,0.083333336,0.116071425,0.15873016,0.14087301,0.11309524,0.14087301,0.10019841,0.14484127,0.11309524,0.16269842,0.13492064,0.17063493,0.19246031,0.17956349,0.11706349,0.21825397,0.16269842,0.20238096,0.1388889,0.2281746,0.18650794,0.18055555,0.20833333,0.1904762,0.23214285,0.051587302,0.15376984,0.27579364,0.21230158,0.20436507,0.21428572,0.15277778,0.1388889,0.24801587,0.31349206,0.22420634,0.26587301,0.2638889,0.3234127,0.3472222,0.35119048,0.33730158,0.38690478,0.33333334,0.3531746,0.37103173,0.39087301,0.5119048,0.3313492,0.4186508,0.5714286,0.4781746,0.5297619,0.4940476,0.50793654,,0.57539684,0.5019841,0.52281743,0.6507937,0.61507934,0.74404764,0.9484127
+KNN (tuned + ensembled),0.035714287,0.057539683,0.06746032,0.057539683,0.097222224,0.083333336,0.083333336,0.06746032,0.09325397,0.10119048,0.08134921,0.10515873,0.10515873,0.09821428,0.12698413,0.125,0.0952381,0.15079366,0.1388889,0.10119048,0.121031746,0.15079366,0.15277778,0.13690476,0.12698413,0.125,0.14484127,0.13690476,0.18055555,0.13194445,0.28373015,0.17857143,0.1468254,0.1468254,0.18253969,0.15674603,0.17063493,0.19444445,0.15277778,0.18154761,0.20833333,0.21428572,0.2638889,0.1984127,0.16865079,0.2559524,0.20238096,0.2361111,0.26984128,0.15476191,0.2609127,0.29166666,0.3234127,0.19246031,0.29166666,0.30555555,0.3561508,0.26785713,0.28373015,0.29563493,0.39285713,0.30059522,0.40873015,0.33730158,0.32936507,0.43452382,0.44146827,0.4434524,0.4940476,0.47420636,0.42460316,,0.45436507,0.45436507,0.5734127,0.5416667,0.84623015,0.9702381
+ExtraTrees (default),0.011904762,0.025793651,0.04563492,0.049603175,0.04761905,0.049603175,0.055555556,0.04761905,0.08531746,0.05952381,0.046626985,0.05357143,0.05357143,0.061507937,0.09325397,0.07539683,0.061507937,0.13293651,0.14880952,0.0952381,0.1468254,0.13492064,0.14087301,0.15476191,0.08134921,0.109126985,0.0952381,0.109126985,0.13492064,0.06746032,0.23412699,0.10515873,0.121031746,0.125,0.13293651,0.1388889,0.16666667,0.1904762,0.18154761,0.13095239,0.19246031,0.23015873,0.20436507,0.17261904,0.18650794,0.21626984,0.21230158,0.24206349,0.2281746,0.10714286,0.27380952,0.24007936,0.3313492,0.19246031,0.31349206,0.32142857,0.38492063,0.22420634,0.23809524,0.31547618,0.44444445,0.29563493,0.44444445,0.37301588,0.40674603,0.47619048,0.4107143,0.46031746,0.50595236,0.45238096,0.49801588,0.5456349,,0.5297619,0.60119045,0.5119048,0.70238096,0.9623016
+RandomForest (default),0.01984127,0.021825397,0.043650795,0.03968254,0.04563492,0.051587302,0.05357143,0.03373016,0.05952381,0.06944445,0.04761905,0.04563492,0.041666668,0.06944445,0.09920635,0.08531746,0.061507937,0.15079366,0.13492064,0.09126984,0.121031746,0.109126985,0.125,0.11706349,0.071428575,0.14087301,0.097222224,0.08531746,0.15277778,0.06349207,0.22222222,0.10515873,0.0952381,0.12301587,0.097222224,0.13095239,0.16468254,0.17063493,0.1845238,0.11309524,0.21825397,0.23214285,0.21825397,0.18253969,0.20436507,0.21428572,0.21626984,0.26785713,0.24603175,0.14880952,0.27579364,0.26984128,0.31944445,0.19742064,0.29761904,0.3015873,0.32142857,0.15674603,0.22420634,0.29662699,0.40079364,0.27083334,0.46031746,0.38492063,0.34126985,0.51984125,0.41269842,0.48809522,0.46626985,0.44444445,0.47718254,0.5456349,0.4702381,,0.6031746,0.51984125,0.71031743,0.9484127
+Linear (default),0.03968254,0.051587302,0.06448413,0.048611112,0.04761905,0.07440476,0.066468254,0.06349207,0.057539683,0.097222224,0.06944445,0.109126985,0.12896825,0.13492064,0.10119048,0.0952381,0.09623016,0.13095239,0.08928572,0.1388889,0.06746032,0.16071428,0.11904762,0.10019841,0.103174604,0.15476191,0.11706349,0.16865079,0.17857143,0.15376984,0.21825397,0.19246031,0.1765873,0.13293651,0.17857143,0.15277778,0.14880952,0.17063493,0.14087301,0.18650794,0.16269842,0.15674603,0.29761904,0.17460318,0.14285715,0.24404761,0.16666667,0.20039682,0.20833333,0.2559524,0.19642857,0.2281746,0.21031746,0.26587301,0.25,0.2579365,0.23412699,0.3065476,0.2638889,0.26587301,0.31150794,0.31547618,0.21428572,0.29166666,0.31349206,0.3888889,0.37896827,0.3015873,0.39484128,0.40079364,0.34920636,0.4265873,0.39880952,0.3968254,,0.47718254,0.60515875,0.8492063
+PerpetualBooster (default),0.01984127,0.035714287,0.041666668,0.027777778,0.03373016,0.041666668,0.041666668,0.043650795,0.043650795,0.051587302,0.04563492,0.06349207,0.03968254,0.06944445,0.043650795,0.061507937,0.049603175,0.10119048,0.08531746,0.071428575,0.06944445,0.057539683,0.05357143,0.08134921,0.065476194,0.12301587,0.06746032,0.109126985,0.11706349,0.08928572,0.20238096,0.103174604,0.10119048,0.10119048,0.11904762,0.11706349,0.103174604,0.14087301,0.14285715,0.121031746,0.13492064,0.11904762,0.21031746,0.125,0.14880952,0.17261904,0.125,0.1984127,0.17460318,0.2559524,0.18650794,0.18551587,0.19444445,0.2777778,0.35912699,0.3611111,0.24603175,0.32539684,0.2777778,0.26785713,0.36507937,0.3452381,0.42857143,0.32738096,0.3690476,0.4484127,0.22420634,0.43452382,0.41269842,0.31547618,0.38492063,0.45833334,0.48809522,0.48015872,0.52281743,,0.61904764,0.8531746
+KNN (tuned),0.043650795,0.04761905,0.05357143,0.05357143,0.06349207,0.0734127,0.06746032,0.057539683,0.072420634,0.083333336,0.057539683,0.08531746,0.07936508,0.08730159,0.08928572,0.08531746,0.071428575,0.12301587,0.103174604,0.097222224,0.051587302,0.11904762,0.11111111,0.061507937,0.08234127,0.09920635,0.09325397,0.103174604,0.13293651,0.10119048,0.17460318,0.13392857,0.097222224,0.10119048,0.122023806,0.09126984,0.10615079,0.10714286,0.121031746,0.125,0.121031746,0.12698413,0.23015873,0.10515873,0.08730159,0.18154761,0.13293651,0.15079366,0.16071428,0.11904762,0.18253969,0.18650794,0.18253969,0.115079366,0.15079366,0.16071428,0.17460318,0.16269842,0.20833333,0.20039682,0.24603175,0.17460318,0.28373015,0.21230158,0.20833333,0.2936508,0.30555555,0.30357143,0.3472222,0.32738096,0.2559524,0.15376984,0.29761904,0.28968254,0.39484128,0.3809524,,0.93650794
+KNN (default),0.005952381,0.003968254,0.005952381,0.003968254,0.005952381,0.007936508,0.007936508,0.007936508,0.009920635,0.005952381,0.005952381,0.009920635,0.005952381,0.005952381,0.015873017,0.003968254,0.007936508,0.013888889,0.01984127,0.009920635,0.013888889,0.017857144,0.023809524,0.005952381,0.007936508,0.007936508,0.005952381,0.007936508,0.01984127,0.009920635,0.017857144,0.021825397,0.007936508,0.009920635,0.017857144,0.011904762,0.01984127,0.027777778,0.015873017,0.021825397,0.023809524,0.023809524,0.05952381,0.01984127,0.021825397,0.035714287,0.021825397,0.031746034,0.024801588,0.009920635,0.03373016,0.035714287,0.023809524,0.011904762,0.04563492,0.04563492,0.029761905,0.031746034,0.05357143,0.03670635,0.057539683,0.029761905,0.0952381,0.051587302,0.07539683,0.057539683,0.05952381,0.103174604,0.0952381,0.08531746,0.051587302,0.029761905,0.037698414,0.051587302,0.15079366,0.1468254,0.06349207,
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
index 6b54f1584a8bd235b09907fab778608c1057033f..278a45e8f105174085494cd4b1cf422c31f6fea5 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Multiclass | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..0e4f204a6ac7a66d8f0be0b4a3c113b70865a769
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_multiclass/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1719.0,"score":0.706,"rank":9.65,"hrank":5.49,"imp":11.809,"train_time":251.72,"predict_time":2.753,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+186\/-76","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1708.0,"score":0.705,"rank":10.11,"hrank":5.01,"imp":12.161,"train_time":2710.49,"predict_time":1.346,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+180\/-82","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1702.0,"score":0.703,"rank":10.33,"hrank":4.49,"imp":11.149,"train_time":2710.49,"predict_time":8.095,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+228\/-121","imputed":false},{"position":3,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1673.0,"score":0.81,"rank":11.65,"hrank":1.99,"imp":8.569,"train_time":55.13,"predict_time":7.685,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+358\/-220","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1628.0,"score":0.61,"rank":13.92,"hrank":8.84,"imp":14.655,"train_time":9.94,"predict_time":0.613,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+198\/-106","imputed":false},{"position":5,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1581.0,"score":0.56,"rank":16.59,"hrank":6.26,"imp":13.396,"train_time":11.66,"predict_time":0.489,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+210\/-143","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1579.0,"score":0.549,"rank":16.75,"hrank":6.95,"imp":13.749,"train_time":10.23,"predict_time":0.844,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+200\/-134","imputed":false},{"position":7,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1553.0,"score":0.526,"rank":18.37,"hrank":4.87,"imp":13.723,"train_time":44.33,"predict_time":7.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+211\/-133","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1552.0,"score":0.535,"rank":18.48,"hrank":7.15,"imp":15.81,"train_time":5.64,"predict_time":0.197,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+239\/-178","imputed":false},{"position":9,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1513.0,"score":0.375,"rank":21.07,"hrank":10.35,"imp":17.871,"train_time":6219.75,"predict_time":3.78,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+172\/-93","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1495.0,"score":0.425,"rank":22.35,"hrank":10.24,"imp":17.259,"train_time":4150.82,"predict_time":19.128,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+165\/-115","imputed":false},{"position":11,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1477.0,"score":0.496,"rank":23.64,"hrank":7.85,"imp":20.679,"train_time":2590.72,"predict_time":10.803,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+270\/-221","imputed":true},{"position":12,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1470.0,"score":0.404,"rank":24.14,"hrank":12.18,"imp":18.164,"train_time":4150.82,"predict_time":3.42,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+126\/-91","imputed":false},{"position":13,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1452.0,"score":0.315,"rank":25.54,"hrank":19.1,"imp":19.663,"train_time":1377.48,"predict_time":5.201,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+107\/-76","imputed":false},{"position":14,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1450.0,"score":0.365,"rank":25.68,"hrank":13.99,"imp":20.221,"train_time":3881.83,"predict_time":4.213,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-100","imputed":false},{"position":15,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1443.0,"score":0.307,"rank":26.18,"hrank":22.36,"imp":18.982,"train_time":2635.98,"predict_time":1.181,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-40","imputed":false},{"position":16,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1440.0,"score":0.299,"rank":26.45,"hrank":20.77,"imp":19.888,"train_time":1377.48,"predict_time":1.259,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-67","imputed":false},{"position":17,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1436.0,"score":0.399,"rank":26.77,"hrank":13.97,"imp":22.179,"train_time":161.66,"predict_time":3.703,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+206\/-175","imputed":true},{"position":18,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1424.0,"score":0.319,"rank":27.64,"hrank":11.69,"imp":21.293,"train_time":5.46,"predict_time":0.271,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+216\/-130","imputed":false},{"position":19,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1420.0,"score":0.26,"rank":28.02,"hrank":23.12,"imp":19.606,"train_time":2635.98,"predict_time":0.173,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-47","imputed":false},{"position":20,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1413.0,"score":0.345,"rank":28.55,"hrank":14.47,"imp":22.238,"train_time":4635.17,"predict_time":212.235,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+170\/-126","imputed":false},{"position":21,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1411.0,"score":0.318,"rank":28.69,"hrank":15.96,"imp":20.987,"train_time":3881.83,"predict_time":0.698,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-90","imputed":false},{"position":22,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1404.0,"score":0.358,"rank":29.25,"hrank":12.22,"imp":22.404,"train_time":2590.72,"predict_time":0.54,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+210\/-171","imputed":true},{"position":23,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1401.0,"score":0.234,"rank":29.48,"hrank":22.21,"imp":20.311,"train_time":2210.17,"predict_time":0.974,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-72","imputed":false},{"position":24,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1396.0,"score":0.228,"rank":29.9,"hrank":23.34,"imp":20.4,"train_time":2210.17,"predict_time":3.984,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-74","imputed":false},{"position":25,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1390.0,"score":0.295,"rank":30.38,"hrank":16.55,"imp":21.946,"train_time":6.47,"predict_time":0.363,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+179\/-145","imputed":true},{"position":26,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1389.0,"score":0.19,"rank":30.44,"hrank":22.8,"imp":20.225,"train_time":7128.67,"predict_time":9.211,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-72","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1382.0,"score":0.201,"rank":31.0,"hrank":15.29,"imp":20.492,"train_time":7128.67,"predict_time":0.776,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-56","imputed":false},{"position":28,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1380.0,"score":0.187,"rank":31.18,"hrank":25.83,"imp":20.983,"train_time":28234.45,"predict_time":841.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-82","imputed":false},{"position":29,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1368.0,"score":0.282,"rank":32.2,"hrank":16.16,"imp":23.558,"train_time":4635.17,"predict_time":34.761,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+190\/-135","imputed":false},{"position":30,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1361.0,"score":0.283,"rank":32.73,"hrank":22.47,"imp":23.196,"train_time":507.78,"predict_time":4.427,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-126","imputed":false},{"position":31,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1359.0,"score":0.26,"rank":32.92,"hrank":19.66,"imp":22.872,"train_time":1980.61,"predict_time":3.604,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+148\/-139","imputed":false},{"position":32,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1355.0,"score":0.209,"rank":33.27,"hrank":25.35,"imp":22.095,"train_time":27107.4,"predict_time":0.271,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+109\/-93","imputed":false},{"position":33,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1339.0,"score":0.357,"rank":34.61,"hrank":7.65,"imp":21.447,"train_time":22.52,"predict_time":1.681,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+189\/-234","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1337.0,"score":0.251,"rank":34.76,"hrank":20.15,"imp":23.779,"train_time":12.83,"predict_time":0.673,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-130","imputed":false},{"position":35,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1330.0,"score":0.177,"rank":35.33,"hrank":29.88,"imp":21.989,"train_time":26.14,"predict_time":0.202,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-74","imputed":false},{"position":36,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1324.0,"score":0.236,"rank":35.84,"hrank":16.43,"imp":24.365,"train_time":984.66,"predict_time":1.487,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+180\/-149","imputed":false},{"position":37,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1316.0,"score":0.261,"rank":36.52,"hrank":14.29,"imp":24.155,"train_time":738.02,"predict_time":1.862,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+174\/-169","imputed":false},{"position":38,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1314.0,"score":0.211,"rank":36.72,"hrank":22.49,"imp":25.574,"train_time":10.98,"predict_time":1.898,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+166\/-154","imputed":true},{"position":39,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1313.0,"score":0.17,"rank":36.74,"hrank":28.11,"imp":22.801,"train_time":27107.4,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+109\/-104","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1303.0,"score":0.123,"rank":37.65,"hrank":29.79,"imp":23.457,"train_time":3902.08,"predict_time":2.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+107\/-117","imputed":false},{"position":41,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1287.0,"score":0.1,"rank":38.96,"hrank":34.38,"imp":22.62,"train_time":28234.45,"predict_time":276.786,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-82","imputed":false},{"position":42,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1282.0,"score":0.142,"rank":39.41,"hrank":29.13,"imp":24.403,"train_time":13.33,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-127","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1274.0,"score":0.195,"rank":40.08,"hrank":14.21,"imp":25.968,"train_time":984.66,"predict_time":0.279,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+177\/-150","imputed":false},{"position":44,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1270.0,"score":0.166,"rank":40.4,"hrank":20.69,"imp":25.119,"train_time":1980.61,"predict_time":0.61,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+144\/-122","imputed":false},{"position":45,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1268.0,"score":0.211,"rank":40.53,"hrank":11.9,"imp":25.418,"train_time":738.02,"predict_time":0.316,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+195\/-201","imputed":false},{"position":46,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1264.0,"score":0.118,"rank":40.87,"hrank":35.23,"imp":25.697,"train_time":4.32,"predict_time":0.084,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-112","imputed":false},{"position":47,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1261.0,"score":0.164,"rank":41.15,"hrank":17.29,"imp":26.054,"train_time":43.39,"predict_time":42.555,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+210\/-134","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1260.0,"score":0.198,"rank":41.19,"hrank":16.07,"imp":26.698,"train_time":2644.99,"predict_time":12.291,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+157\/-174","imputed":false},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1253.0,"score":0.05,"rank":41.8,"hrank":37.44,"imp":23.438,"train_time":4.72,"predict_time":0.542,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-73","imputed":false},{"position":50,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1250.0,"score":0.158,"rank":42.0,"hrank":32.26,"imp":24.366,"train_time":1599.92,"predict_time":0.205,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-142","imputed":false},{"position":51,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1243.0,"score":0.111,"rank":42.6,"hrank":30.81,"imp":25.542,"train_time":3902.08,"predict_time":0.223,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-148","imputed":false},{"position":52,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1243.0,"score":0.148,"rank":42.62,"hrank":35.85,"imp":24.429,"train_time":1599.92,"predict_time":1.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-136","imputed":false},{"position":53,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1228.0,"score":0.138,"rank":43.87,"hrank":32.01,"imp":25.313,"train_time":5.04,"predict_time":0.429,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+154\/-132","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1208.0,"score":0.097,"rank":45.39,"hrank":26.01,"imp":27.775,"train_time":2644.99,"predict_time":1.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+132\/-171","imputed":false},{"position":55,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1197.0,"score":0.019,"rank":46.29,"hrank":38.8,"imp":28.136,"train_time":17.95,"predict_time":0.691,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-92","imputed":false},{"position":56,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1161.0,"score":0.116,"rank":49.02,"hrank":37.86,"imp":27.487,"train_time":7.2,"predict_time":0.19,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-188","imputed":false},{"position":57,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1151.0,"score":0.0,"rank":49.76,"hrank":46.25,"imp":28.699,"train_time":19.2,"predict_time":7.62,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-86","imputed":false},{"position":58,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1091.0,"score":0.048,"rank":54.05,"hrank":47.3,"imp":32.164,"train_time":9.43,"predict_time":1.675,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-207","imputed":false},{"position":59,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1066.0,"score":0.0,"rank":55.72,"hrank":42.47,"imp":31.372,"train_time":13.69,"predict_time":0.314,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-134","imputed":false},{"position":60,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1058.0,"score":0.019,"rank":56.2,"hrank":53.24,"imp":29.557,"train_time":514.72,"predict_time":211.976,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-157","imputed":false},{"position":61,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1009.0,"score":0.0,"rank":59.22,"hrank":55.88,"imp":34.722,"train_time":21.06,"predict_time":3.367,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-146","imputed":false},{"position":62,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1009.0,"score":0.061,"rank":59.27,"hrank":27.97,"imp":41.425,"train_time":151.75,"predict_time":2.345,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+219\/-324","imputed":false},{"position":63,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1005.0,"score":0.0,"rank":59.45,"hrank":53.84,"imp":33.164,"train_time":40588.15,"predict_time":31.228,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-193","imputed":false},{"position":64,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":59.77,"hrank":55.61,"imp":37.967,"train_time":0.74,"predict_time":0.137,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-154","imputed":false},{"position":65,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":999.0,"score":0.0,"rank":59.8,"hrank":55.77,"imp":33.276,"train_time":40588.15,"predict_time":8.466,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-177","imputed":false},{"position":66,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":979.0,"score":0.0,"rank":60.97,"hrank":57.56,"imp":42.276,"train_time":893.09,"predict_time":0.818,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-242","imputed":false},{"position":67,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":950.0,"score":0.0,"rank":62.56,"hrank":60.19,"imp":39.24,"train_time":4.85,"predict_time":0.786,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-154","imputed":false},{"position":68,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":947.0,"score":0.0,"rank":62.71,"hrank":59.09,"imp":43.131,"train_time":893.09,"predict_time":0.146,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-250","imputed":false},{"position":69,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":944.0,"score":0.0,"rank":62.83,"hrank":59.9,"imp":36.725,"train_time":8.52,"predict_time":0.734,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-177","imputed":false},{"position":70,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":888.0,"score":0.0,"rank":65.57,"hrank":57.46,"imp":51.802,"train_time":1409.51,"predict_time":10.372,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-200","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":865.0,"score":0.0,"rank":66.58,"hrank":65.93,"imp":41.495,"train_time":5.06,"predict_time":1.013,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-173","imputed":false},{"position":72,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":852.0,"score":0.0,"rank":67.14,"hrank":65.29,"imp":41.79,"train_time":724.31,"predict_time":8.133,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-216","imputed":false},{"position":73,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":847.0,"score":0.053,"rank":67.34,"hrank":47.75,"imp":51.162,"train_time":151.75,"predict_time":0.258,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+271\/-439","imputed":false},{"position":74,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":841.0,"score":0.0,"rank":67.58,"hrank":40.94,"imp":53.561,"train_time":1409.51,"predict_time":3.696,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-212","imputed":false},{"position":75,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":812.0,"score":0.0,"rank":68.71,"hrank":34.58,"imp":57.842,"train_time":228.95,"predict_time":0.465,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-280","imputed":false},{"position":76,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":747.0,"score":0.0,"rank":70.86,"hrank":70.23,"imp":48.974,"train_time":4.53,"predict_time":0.164,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-240","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":352.0,"score":0.0,"rank":77.21,"hrank":77.18,"imp":71.577,"train_time":0.52,"predict_time":0.076,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-299","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/pareto_front_explorer.html
index 46abe7e262f6b88f6396026ef708190618573b44..9ee39254ceb086b7c5ce079155e3f0df0916fec7 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.4269406194,"imp":23.1962382265,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2017044414,"imp":21.9890239308,"elo":1330.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1731293093,"imp":19.6064488989,"elo":1419.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.1808204254,"imp":18.9820362264,"elo":1443.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1904913614,"imp":27.4872614898,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2050159281,"imp":24.3657075156,"elo":1250.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2759753363,"imp":24.429457087,"elo":1242.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0400823469,"imp":24.4032458767,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0510244193,"imp":22.8006914361,"elo":1313.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.2710569482,"imp":22.0951080329,"elo":1354.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7862858499,"imp":39.2401655498,"elo":949.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2793367209,"imp":25.9681711846,"elo":1273.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4872412963,"imp":24.365415216,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.733891056,"imp":36.7254422517,"elo":944.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1478371705,"imp":27.7753952183,"elo":1208.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.2912040793,"imp":26.6982595882,"elo":1260.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.4290808604,"imp":25.3134708048,"elo":1227.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.2591317348,"imp":19.8878835704,"elo":1439.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":5.2006257965,"imp":19.6630106814,"elo":1451.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.0619559967,"imp":13.7229134995,"elo":1553.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.7027625172,"imp":22.1794709087,"elo":1435.6,"imputed":true,"imputed_pct":25.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.690602132,"imp":28.136054603,"elo":1196.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7756352585,"imp":20.4920634274,"elo":1382.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2105111376,"imp":20.2246376446,"elo":1389.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3671495151,"imp":34.7224465533,"elo":1009.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4648882862,"imp":57.8423345319,"elo":811.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":3.6964435846,"imp":53.561033721,"elo":841.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":10.3718427289,"imp":51.8022949082,"elo":888.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1372541163,"imp":37.9666395955,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3155208059,"imp":25.4175645517,"elo":1268.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.8624917599,"imp":24.1551523057,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.6199152504,"imp":28.6993713596,"elo":1151.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.4203022504,"imp":18.1639626938,"elo":1470.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":19.128322436,"imp":17.2587075706,"elo":1494.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8441387842,"imp":13.7494249477,"elo":1578.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3463854051,"imp":12.1611744449,"elo":1707.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0946622325,"imp":11.1486610598,"elo":1702.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.681335577,"imp":21.4469696745,"elo":1338.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.5545582952,"imp":26.0542173745,"elo":1260.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":34.7609767403,"imp":23.5581339476,"elo":1367.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":212.2351742215,"imp":22.2384178881,"elo":1412.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2707653455,"imp":21.2932434623,"elo":1424.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.6845942368,"imp":8.5694204554,"elo":1673.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":1.0132382115,"imp":41.4951729861,"elo":865.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.8981443546,"imp":25.574290678,"elo":1313.6,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1967767849,"imp":15.8102638575,"elo":1551.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.6734553632,"imp":23.7792823687,"elo":1336.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.6978379022,"imp":20.9872512504,"elo":1411.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.2126633842,"imp":20.2209365211,"elo":1449.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6126587114,"imp":14.6551463119,"elo":1628.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4885714957,"imp":13.3963420146,"elo":1581.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3630834262,"imp":21.9459731589,"elo":1390.0,"imputed":true,"imputed_pct":25.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.539717246,"imp":22.4043289909,"elo":1404.0,"imputed":true,"imputed_pct":25.0,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.8027038192,"imp":20.6786035148,"elo":1477.1,"imputed":true,"imputed_pct":25.0,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1330110917,"imp":41.7900963877,"elo":851.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.4659125188,"imp":33.2755925465,"elo":999.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.2276662759,"imp":33.1644651704,"elo":1005.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0836134922,"imp":25.6968308973,"elo":1264.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.314314201,"imp":31.3724369621,"elo":1065.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.22258711,"imp":25.5418416449,"elo":1243.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0277285638,"imp":23.4572089482,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.5423707001,"imp":23.4383382088,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.9738781112,"imp":20.3108777206,"elo":1401.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.9840757918,"imp":20.4004311379,"elo":1395.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":211.9764977363,"imp":29.5572280283,"elo":1058.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":276.7864889801,"imp":22.6196129239,"elo":1287.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"imp":20.9833599902,"elo":1380.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.6749433874,"imp":32.1644031925,"elo":1090.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.6100366829,"imp":25.1187096816,"elo":1269.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.603859777,"imp":22.8719222053,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.4269406194,"x_train":507.7823899342,"imp":23.1962382265,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2017044414,"x_train":26.1374278937,"imp":21.9890239308,"elo":1330.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1731293093,"x_train":2635.9752136464,"imp":19.6064488989,"elo":1419.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.1808204254,"x_train":2635.9752136464,"imp":18.9820362264,"elo":1443.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1904913614,"x_train":7.2011684966,"imp":27.4872614898,"elo":1161.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2050159281,"x_train":1599.9245048672,"imp":24.3657075156,"elo":1250.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2759753363,"x_train":1599.9245048672,"imp":24.429457087,"elo":1242.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0400823469,"x_train":13.333580501,"imp":24.4032458767,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0510244193,"x_train":27107.3972485932,"imp":22.8006914361,"elo":1313.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.2710569482,"x_train":27107.3972485932,"imp":22.0951080329,"elo":1354.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7862858499,"x_train":4.8501982351,"imp":39.2401655498,"elo":949.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2793367209,"x_train":984.6610737813,"imp":25.9681711846,"elo":1273.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4872412963,"x_train":984.6610737813,"imp":24.365415216,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.733891056,"x_train":8.5155691544,"imp":36.7254422517,"elo":944.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1478371705,"x_train":2644.9909431875,"imp":27.7753952183,"elo":1208.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.2912040793,"x_train":2644.9909431875,"imp":26.6982595882,"elo":1260.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.4290808604,"x_train":5.0434629122,"imp":25.3134708048,"elo":1227.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.2591317348,"x_train":1377.4770257446,"imp":19.8878835704,"elo":1439.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":5.2006257965,"x_train":1377.4770257446,"imp":19.6630106814,"elo":1451.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.0619559967,"x_train":44.325578779,"imp":13.7229134995,"elo":1553.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.7027625172,"x_train":161.6635737868,"imp":22.1794709087,"elo":1435.6,"imputed":true,"imputed_pct":25.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.690602132,"x_train":17.9457268623,"imp":28.136054603,"elo":1196.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7756352585,"x_train":7128.6715875046,"imp":20.4920634274,"elo":1382.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2105111376,"x_train":7128.6715875046,"imp":20.2246376446,"elo":1389.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3671495151,"x_train":21.0623624205,"imp":34.7224465533,"elo":1009.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4648882862,"x_train":228.9454185462,"imp":57.8423345319,"elo":811.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":3.6964435846,"x_train":1409.5097915247,"imp":53.561033721,"elo":841.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":10.3718427289,"x_train":1409.5097915247,"imp":51.8022949082,"elo":888.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1372541163,"x_train":0.7350357633,"imp":37.9666395955,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3155208059,"x_train":738.0209327471,"imp":25.4175645517,"elo":1268.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.8624917599,"x_train":738.0209327471,"imp":24.1551523057,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.6199152504,"x_train":19.2044919241,"imp":28.6993713596,"elo":1151.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.4203022504,"x_train":4150.8158923996,"imp":18.1639626938,"elo":1470.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":19.128322436,"x_train":4150.8158923996,"imp":17.2587075706,"elo":1494.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8441387842,"x_train":10.2256680645,"imp":13.7494249477,"elo":1578.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3463854051,"x_train":2710.4880101354,"imp":12.1611744449,"elo":1707.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0946622325,"x_train":2710.4880101354,"imp":11.1486610598,"elo":1702.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.681335577,"x_train":22.5239093941,"imp":21.4469696745,"elo":1338.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.5545582952,"x_train":43.394404174,"imp":26.0542173745,"elo":1260.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":34.7609767403,"x_train":4635.1714883924,"imp":23.5581339476,"elo":1367.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":212.2351742215,"x_train":4635.1714883924,"imp":22.2384178881,"elo":1412.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2707653455,"x_train":5.4556230963,"imp":21.2932434623,"elo":1424.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.6845942368,"x_train":55.129309117,"imp":8.5694204554,"elo":1673.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":1.0132382115,"x_train":5.0558157998,"imp":41.4951729861,"elo":865.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.8981443546,"x_train":10.9804562909,"imp":25.574290678,"elo":1313.6,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1967767849,"x_train":5.6364435241,"imp":15.8102638575,"elo":1551.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.6734553632,"x_train":12.8284122123,"imp":23.7792823687,"elo":1336.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.6978379022,"x_train":3881.8318340105,"imp":20.9872512504,"elo":1411.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.2126633842,"x_train":3881.8318340105,"imp":20.2209365211,"elo":1449.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6126587114,"x_train":9.9408095472,"imp":14.6551463119,"elo":1628.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4885714957,"x_train":11.6635229296,"imp":13.3963420146,"elo":1581.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3630834262,"x_train":6.4658277198,"imp":21.9459731589,"elo":1390.0,"imputed":true,"imputed_pct":25.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.539717246,"x_train":2590.7221425729,"imp":22.4043289909,"elo":1404.0,"imputed":true,"imputed_pct":25.0,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.8027038192,"x_train":2590.7221425729,"imp":20.6786035148,"elo":1477.1,"imputed":true,"imputed_pct":25.0,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1330110917,"x_train":724.3138183772,"imp":41.7900963877,"elo":851.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.4659125188,"x_train":40588.1540174129,"imp":33.2755925465,"elo":999.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.2276662759,"x_train":40588.1540174129,"imp":33.1644651704,"elo":1005.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0836134922,"x_train":4.3206513889,"imp":25.6968308973,"elo":1264.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.314314201,"x_train":13.688894957,"imp":31.3724369621,"elo":1065.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.22258711,"x_train":3902.0763863206,"imp":25.5418416449,"elo":1243.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0277285638,"x_train":3902.0763863206,"imp":23.4572089482,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.5423707001,"x_train":4.7244136606,"imp":23.4383382088,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.9738781112,"x_train":2210.16539451,"imp":20.3108777206,"elo":1401.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.9840757918,"x_train":2210.16539451,"imp":20.4004311379,"elo":1395.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":211.9764977363,"x_train":514.7174339635,"imp":29.5572280283,"elo":1058.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":276.7864889801,"x_train":28234.4532410041,"imp":22.6196129239,"elo":1287.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"x_train":28234.4532410041,"imp":20.9833599902,"elo":1380.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.6749433874,"x_train":9.4251823487,"imp":32.1644031925,"elo":1090.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.6100366829,"x_train":1980.6125379489,"imp":25.1187096816,"elo":1269.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.603859777,"x_train":1980.6125379489,"imp":22.8719222053,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
index 6887540b3a6c56264db1d9b90798044fa082a989..b9081dee2ebc44c812400b5123cce8af8acf907b 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":507.7823899342,"imp":23.2181961578,"elo":1365.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":26.1374278937,"imp":22.0106068204,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":49.1954460239,"imp":20.432820578,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":117.0085419585,"imp":19.7642557476,"elo":1421.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":233.7248660534,"imp":19.4314916727,"elo":1435.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":454.7581834642,"imp":19.1270183398,"elo":1449.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":753.831728936,"imp":19.1068903597,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1385.7187287572,"imp":19.0426931353,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2005.179917238,"imp":19.0368952238,"elo":1454.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":2635.9752136464,"imp":19.003844476,"elo":1453.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":7.2011684966,"imp":27.508913848,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":15.101549461,"imp":27.0320020016,"elo":1173.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":38.0760065843,"imp":26.640426754,"elo":1185.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":76.9297326944,"imp":26.178434757,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":196.6496559406,"imp":25.4298461138,"elo":1214.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":396.6608033968,"imp":25.1011915024,"elo":1224.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":796.7427370653,"imp":24.6927215976,"elo":1235.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1194.3788869413,"imp":24.5374484578,"elo":1241.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1599.9245048672,"imp":24.4516353135,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":13.333580501,"imp":24.4249233188,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":155.2301034791,"imp":24.0262010676,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":423.1216867301,"imp":23.3226728751,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":1126.0038118286,"imp":23.0831915073,"elo":1324.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":3121.6567594042,"imp":22.6949846807,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":6636.6428563278,"imp":22.4511600345,"elo":1345.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":13596.8410765133,"imp":22.2597903586,"elo":1353.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":20505.2305784284,"imp":22.173841719,"elo":1358.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":27107.3972485932,"imp":22.1179735297,"elo":1360.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.8501982351,"imp":39.2590514202,"elo":942.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.7103499513,"imp":34.8802344999,"elo":1022.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.7095068985,"imp":28.3435111097,"elo":1145.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":49.0400181801,"imp":26.4405287266,"elo":1235.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":122.0237750052,"imp":24.9771118821,"elo":1309.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":244.0803421306,"imp":24.7627007367,"elo":1317.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":488.408774127,"imp":24.5680212571,"elo":1325.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":733.7086039727,"imp":24.4383082635,"elo":1331.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":984.6610737813,"imp":24.3820642536,"elo":1333.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.5155691544,"imp":36.7459925237,"elo":940.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":21.559937581,"imp":34.8637120844,"elo":999.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":62.1360327566,"imp":31.3660962381,"elo":1110.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":125.9632889407,"imp":29.7674421618,"elo":1168.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":322.1066821739,"imp":28.1555105231,"elo":1225.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":651.6805621192,"imp":27.4247384378,"elo":1252.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1305.7711888267,"imp":27.0461378125,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1966.2811643084,"imp":26.8509422305,"elo":1267.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2644.9909431875,"imp":26.7172710588,"elo":1270.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":5.0434629122,"imp":25.3352103232,"elo":1230.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":12.7823939958,"imp":24.1800729719,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":32.375532788,"imp":23.0390648431,"elo":1331.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":66.0618660449,"imp":22.23000793,"elo":1360.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":169.9915970428,"imp":20.9540529621,"elo":1403.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":340.9497698538,"imp":20.3501872582,"elo":1434.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":684.7234530331,"imp":19.9672146978,"elo":1453.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":1025.1619507673,"imp":19.8340078628,"elo":1459.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1377.4770257446,"imp":19.6830855333,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":44.325578779,"imp":13.7870134808,"elo":1562.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":161.6635737868,"imp":22.2030752315,"elo":1446.8,"imputed":true,"imputed_pct":25.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":17.9457268623,"imp":28.157115663,"elo":1197.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":57.1222738852,"imp":25.6698873382,"elo":1238.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":171.10400172,"imp":22.7819875057,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":346.9635750688,"imp":21.9871200842,"elo":1353.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":858.4265706116,"imp":21.0956142132,"elo":1389.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1767.2813809485,"imp":20.7428719159,"elo":1397.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3533.1581339321,"imp":20.6738903214,"elo":1397.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":5301.5364312197,"imp":20.3994723935,"elo":1397.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":7128.6715875046,"imp":20.2476728546,"elo":1396.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":21.0623624205,"imp":34.8574265444,"elo":1003.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":228.9454185462,"imp":57.9098679747,"elo":810.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":466.150787739,"imp":56.3506565873,"elo":838.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1174.4301363506,"imp":52.2513182237,"elo":877.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1409.5097915247,"imp":51.8113311743,"elo":886.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.7350357633,"imp":38.0491106487,"elo":987.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.5934411859,"imp":30.7344585918,"elo":1107.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":15.6940857589,"imp":27.2942868241,"elo":1174.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":34.1407666441,"imp":25.6585878052,"elo":1225.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":89.3035790361,"imp":24.6365806787,"elo":1303.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":181.3479112981,"imp":24.4402754295,"elo":1314.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":365.5123989512,"imp":24.2960959989,"elo":1321.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":549.883689003,"imp":24.2320480777,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":738.0209327471,"imp":24.172864983,"elo":1326.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":19.2044919241,"imp":28.7219271744,"elo":1150.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":41.019845821,"imp":23.7746241727,"elo":1305.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":103.1070784858,"imp":20.8952635244,"elo":1403.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":206.5111037718,"imp":19.7671680566,"elo":1440.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":515.981423509,"imp":18.8612121649,"elo":1470.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1035.4018073353,"imp":18.3522800521,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2068.879284709,"imp":17.8172491619,"elo":1500.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3101.33807397,"imp":17.5346593638,"elo":1507.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4150.8158923996,"imp":17.2816532978,"elo":1510.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":10.2256680645,"imp":13.7738810388,"elo":1583.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":22.8990925067,"imp":13.9446191652,"elo":1592.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":62.9203931706,"imp":13.6737062754,"elo":1624.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":129.5212073934,"imp":13.3827694838,"elo":1675.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":331.7365286813,"imp":12.992635558,"elo":1705.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":668.706695891,"imp":12.2314676528,"elo":1721.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1349.1531554668,"imp":11.8378767882,"elo":1718.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2024.0527372009,"imp":11.5371472269,"elo":1717.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2710.4880101354,"imp":11.1759407357,"elo":1714.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":22.5239093941,"imp":21.4688582417,"elo":1340.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":43.394404174,"imp":26.0698377808,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.8442293296,"imp":25.711386576,"elo":1280.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":122.0779658592,"imp":25.0344234357,"elo":1317.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":246.0408977706,"imp":24.5378205209,"elo":1344.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":571.7774981312,"imp":23.7087353167,"elo":1372.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1153.6863993273,"imp":23.1581447753,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2305.5325709939,"imp":22.7078351625,"elo":1411.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3450.9693523062,"imp":22.4529270609,"elo":1418.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4635.1714883924,"imp":22.2572497999,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":5.4556230963,"imp":21.3111223298,"elo":1434.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":55.129309117,"imp":8.593085032,"elo":1676.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":5.0558157998,"imp":41.5051626445,"elo":858.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":10.9804562909,"imp":25.5947516248,"elo":1320.1,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":5.6364435241,"imp":15.8345710574,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":12.8284122123,"imp":23.8025666322,"elo":1336.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":30.6749168115,"imp":23.0365400507,"elo":1361.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":92.4143274115,"imp":22.0564004325,"elo":1391.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":190.6444632314,"imp":21.4724459851,"elo":1413.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":476.8006390989,"imp":21.0089554552,"elo":1431.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":956.8892667394,"imp":20.6684352668,"elo":1446.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1912.1428059667,"imp":20.4316486045,"elo":1454.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2886.0992294917,"imp":20.3376701178,"elo":1457.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3881.8318340105,"imp":20.2436587744,"elo":1462.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":9.9408095472,"imp":14.6796200017,"elo":1645.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":11.6635229296,"imp":13.4195320821,"elo":1587.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":6.4658277198,"imp":21.9699175713,"elo":1396.8,"imputed":true,"imputed_pct":25.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":24.4325217028,"imp":21.9351156749,"elo":1417.6,"imputed":true,"imputed_pct":25.0,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":60.616973651,"imp":21.6170201631,"elo":1448.0,"imputed":true,"imputed_pct":25.0,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":127.8381423789,"imp":21.3464137897,"elo":1462.0,"imputed":true,"imputed_pct":25.0,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":315.8789741575,"imp":20.9613600395,"elo":1483.6,"imputed":true,"imputed_pct":25.0,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":633.6979487098,"imp":20.7814630166,"elo":1494.0,"imputed":true,"imputed_pct":25.0,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1285.7447177186,"imp":20.7084108681,"elo":1496.4,"imputed":true,"imputed_pct":25.0,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1930.4265884275,"imp":20.6805721307,"elo":1495.5,"imputed":true,"imputed_pct":25.0,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2590.7221425729,"imp":20.6616902977,"elo":1494.8,"imputed":true,"imputed_pct":25.0,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":724.3138183772,"imp":41.8093223676,"elo":842.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":912.4748239301,"imp":41.5820219496,"elo":854.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1342.7568425834,"imp":41.1159531554,"elo":865.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2054.553219683,"imp":40.1296006426,"elo":889.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4867.9697698991,"imp":37.9313768562,"elo":924.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":9985.461750605,"imp":35.6648086285,"elo":954.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":20581.7115313909,"imp":34.4019232225,"elo":980.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":30569.2328830115,"imp":33.7036439561,"elo":997.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":40588.1540174129,"imp":33.1852869871,"elo":1013.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":4.3206513889,"imp":25.7146613846,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.688894957,"imp":31.3941133808,"elo":1062.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":28.9516297239,"imp":29.1184062049,"elo":1125.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":95.1897765055,"imp":27.1982025237,"elo":1187.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":204.1284404869,"imp":25.936645629,"elo":1236.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":488.13995387,"imp":24.9207297786,"elo":1264.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":949.9701252728,"imp":24.3708420151,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1876.0678595078,"imp":23.963974496,"elo":1294.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2872.3862481998,"imp":23.7238774346,"elo":1303.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3902.0763863206,"imp":23.4818670473,"elo":1309.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":4.7244136606,"imp":23.4595335124,"elo":1249.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":15.9463545731,"imp":22.300497142,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":50.0714874559,"imp":21.8699882167,"elo":1329.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":100.9432671475,"imp":21.4538245005,"elo":1349.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":268.6777580855,"imp":21.1293488203,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":544.7091931075,"imp":20.903947403,"elo":1380.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":1096.7820562853,"imp":20.7214815152,"elo":1390.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":1644.1757072757,"imp":20.5856288602,"elo":1398.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":2210.16539451,"imp":20.4223786159,"elo":1408.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":514.7174339635,"imp":29.5771429609,"elo":1053.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1597.2950547029,"imp":26.9278279178,"elo":1128.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":4783.0314636057,"imp":24.2475396394,"elo":1245.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":10691.0627275524,"imp":22.9188803133,"elo":1308.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":26917.4383401199,"imp":21.0450562142,"elo":1386.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":28234.4532410041,"imp":21.0060605589,"elo":1390.7,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":9.4251823487,"imp":32.1823681604,"elo":1084.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":19.1707317375,"imp":29.8777254875,"elo":1159.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":47.812888724,"imp":26.9417191671,"elo":1259.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":97.2172107229,"imp":25.3650225131,"elo":1294.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":246.1477821059,"imp":24.1161297339,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":491.5550544333,"imp":23.5654595973,"elo":1350.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":987.2594279642,"imp":23.1702584153,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1477.605852992,"imp":23.0057712232,"elo":1364.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1980.6125379489,"imp":22.8929193229,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.4269406194,"x_train":507.7823899342,"imp":23.2181961578,"elo":1365.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2017044414,"x_train":26.1374278937,"imp":22.0106068204,"elo":1332.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3433286258,"x_train":49.1954460239,"imp":20.432820578,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.520712217,"x_train":117.0085419585,"imp":19.7642557476,"elo":1421.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.6542443518,"x_train":233.7248660534,"imp":19.4314916727,"elo":1435.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.8258809333,"x_train":454.7581834642,"imp":19.1270183398,"elo":1449.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.9640813791,"x_train":753.831728936,"imp":19.1068903597,"elo":1449.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.0748368064,"x_train":1385.7187287572,"imp":19.0426931353,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1178203711,"x_train":2005.179917238,"imp":19.0368952238,"elo":1454.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1808204254,"x_train":2635.9752136464,"imp":19.003844476,"elo":1453.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1904913614,"x_train":7.2011684966,"imp":27.508913848,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3136490334,"x_train":15.101549461,"imp":27.0320020016,"elo":1173.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4619268878,"x_train":38.0760065843,"imp":26.640426754,"elo":1185.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.599982434,"x_train":76.9297326944,"imp":26.178434757,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.7770406751,"x_train":196.6496559406,"imp":25.4298461138,"elo":1214.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.924512349,"x_train":396.6608033968,"imp":25.1011915024,"elo":1224.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.0877966046,"x_train":796.7427370653,"imp":24.6927215976,"elo":1235.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1969191452,"x_train":1194.3788869413,"imp":24.5374484578,"elo":1241.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2759753363,"x_train":1599.9245048672,"imp":24.4516353135,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0400823469,"x_train":13.333580501,"imp":24.4249233188,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0986631642,"x_train":155.2301034791,"imp":24.0262010676,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.1470099052,"x_train":423.1216867301,"imp":23.3226728751,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.1694069784,"x_train":1126.0038118286,"imp":23.0831915073,"elo":1324.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.2138690235,"x_train":3121.6567594042,"imp":22.6949846807,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.2330722424,"x_train":6636.6428563278,"imp":22.4511600345,"elo":1345.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.251670447,"x_train":13596.8410765133,"imp":22.2597903586,"elo":1353.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.2739203723,"x_train":20505.2305784284,"imp":22.173841719,"elo":1358.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.2710569482,"x_train":27107.3972485932,"imp":22.1179735297,"elo":1360.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7862858499,"x_train":4.8501982351,"imp":39.2590514202,"elo":942.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9602823864,"x_train":9.7103499513,"imp":34.8802344999,"elo":1022.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0018135834,"x_train":24.7095068985,"imp":28.3435111097,"elo":1145.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0557265009,"x_train":49.0400181801,"imp":26.4405287266,"elo":1235.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2034774794,"x_train":122.0237750052,"imp":24.9771118821,"elo":1309.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2879477145,"x_train":244.0803421306,"imp":24.7627007367,"elo":1317.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4038021731,"x_train":488.408774127,"imp":24.5680212571,"elo":1325.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4626957587,"x_train":733.7086039727,"imp":24.4383082635,"elo":1331.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4872412963,"x_train":984.6610737813,"imp":24.3820642536,"elo":1333.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.733891056,"x_train":8.5155691544,"imp":36.7459925237,"elo":940.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7139282714,"x_train":21.559937581,"imp":34.8637120844,"elo":999.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.4892327247,"x_train":62.1360327566,"imp":31.3660962381,"elo":1110.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0003746992,"x_train":125.9632889407,"imp":29.7674421618,"elo":1168.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.3351232966,"x_train":322.1066821739,"imp":28.1555105231,"elo":1225.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7914380124,"x_train":651.6805621192,"imp":27.4247384378,"elo":1252.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.8304499119,"x_train":1305.7711888267,"imp":27.0461378125,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":11.5305755623,"x_train":1966.2811643084,"imp":26.8509422305,"elo":1267.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.2697346691,"x_train":2644.9909431875,"imp":26.7172710588,"elo":1270.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4290808604,"x_train":5.0434629122,"imp":25.3352103232,"elo":1230.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6502979595,"x_train":12.7823939958,"imp":24.1800729719,"elo":1281.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.4043257832,"x_train":32.375532788,"imp":23.0390648431,"elo":1331.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.1150961465,"x_train":66.0618660449,"imp":22.23000793,"elo":1360.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.9668372585,"x_train":169.9915970428,"imp":20.9540529621,"elo":1403.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.5089803244,"x_train":340.9497698538,"imp":20.3501872582,"elo":1434.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.9926620726,"x_train":684.7234530331,"imp":19.9672146978,"elo":1453.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.0269000283,"x_train":1025.1619507673,"imp":19.8340078628,"elo":1459.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.1757953638,"x_train":1377.4770257446,"imp":19.6830855333,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":7.0619559967,"x_train":44.325578779,"imp":13.7870134808,"elo":1562.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":3.7027625172,"x_train":161.6635737868,"imp":22.2030752315,"elo":1446.8,"imputed":true,"imputed_pct":25.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.690602132,"x_train":17.9457268623,"imp":28.157115663,"elo":1197.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.2899674847,"x_train":57.1222738852,"imp":25.6698873382,"elo":1238.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.0939148169,"x_train":171.10400172,"imp":22.7819875057,"elo":1322.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.0865111274,"x_train":346.9635750688,"imp":21.9871200842,"elo":1353.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.7950901544,"x_train":858.4265706116,"imp":21.0956142132,"elo":1389.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.224511556,"x_train":1767.2813809485,"imp":20.7428719159,"elo":1397.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8020549495,"x_train":3533.1581339321,"imp":20.6738903214,"elo":1397.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.6927772436,"x_train":5301.5364312197,"imp":20.3994723935,"elo":1397.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2105111376,"x_train":7128.6715875046,"imp":20.2476728546,"elo":1396.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3671495151,"x_train":21.0623624205,"imp":34.8574265444,"elo":1003.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4648882862,"x_train":228.9454185462,"imp":57.9098679747,"elo":810.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":4.1872771761,"x_train":466.150787739,"imp":56.3506565873,"elo":838.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":9.9244331069,"x_train":1174.4301363506,"imp":52.2513182237,"elo":877.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":10.3718427289,"x_train":1409.5097915247,"imp":51.8113311743,"elo":886.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1372541163,"x_train":0.7350357633,"imp":38.0491106487,"elo":987.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4864845935,"x_train":4.5934411859,"imp":30.7344585918,"elo":1107.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8733492574,"x_train":15.6940857589,"imp":27.2942868241,"elo":1174.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.1516799057,"x_train":34.1407666441,"imp":25.6585878052,"elo":1225.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4009230269,"x_train":89.3035790361,"imp":24.6365806787,"elo":1303.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.5852633134,"x_train":181.3479112981,"imp":24.4402754295,"elo":1314.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7100231398,"x_train":365.5123989512,"imp":24.2960959989,"elo":1321.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.824130377,"x_train":549.883689003,"imp":24.2320480777,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.8696383358,"x_train":738.0209327471,"imp":24.172864983,"elo":1326.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.6199152504,"x_train":19.2044919241,"imp":28.7219271744,"elo":1150.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.7479796856,"x_train":41.019845821,"imp":23.7746241727,"elo":1305.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.2922134144,"x_train":103.1070784858,"imp":20.8952635244,"elo":1403.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.8666522912,"x_train":206.5111037718,"imp":19.7671680566,"elo":1440.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.1922827589,"x_train":515.981423509,"imp":18.8612121649,"elo":1470.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4563530196,"x_train":1035.4018073353,"imp":18.3522800521,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.0745491458,"x_train":2068.879284709,"imp":17.8172491619,"elo":1500.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.8336064118,"x_train":3101.33807397,"imp":17.5346593638,"elo":1507.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.128322436,"x_train":4150.8158923996,"imp":17.2816532978,"elo":1510.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8441387842,"x_train":10.2256680645,"imp":13.7738810388,"elo":1583.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.321516012,"x_train":22.8990925067,"imp":13.9446191652,"elo":1592.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.9479323005,"x_train":62.9203931706,"imp":13.6737062754,"elo":1624.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1961228236,"x_train":129.5212073934,"imp":13.3827694838,"elo":1675.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7707065307,"x_train":331.7365286813,"imp":12.992635558,"elo":1705.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4473075753,"x_train":668.706695891,"imp":12.2314676528,"elo":1721.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4240520347,"x_train":1349.1531554668,"imp":11.8378767882,"elo":1718.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6603879275,"x_train":2024.0527372009,"imp":11.5371472269,"elo":1717.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0946622325,"x_train":2710.4880101354,"imp":11.1759407357,"elo":1714.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.681335577,"x_train":22.5239093941,"imp":21.4688582417,"elo":1340.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":42.5545582952,"x_train":43.394404174,"imp":26.0698377808,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":51.1925760104,"x_train":63.8442293296,"imp":25.711386576,"elo":1280.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":76.2941155935,"x_train":122.0779658592,"imp":25.0344234357,"elo":1317.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":103.6508586102,"x_train":246.0408977706,"imp":24.5378205209,"elo":1344.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":130.9325740919,"x_train":571.7774981312,"imp":23.7087353167,"elo":1372.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":148.5135318639,"x_train":1153.6863993273,"imp":23.1581447753,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":181.9231911908,"x_train":2305.5325709939,"imp":22.7078351625,"elo":1411.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":200.1306803402,"x_train":3450.9693523062,"imp":22.4529270609,"elo":1418.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":212.2117190867,"x_train":4635.1714883924,"imp":22.2572497999,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2707653455,"x_train":5.4556230963,"imp":21.3111223298,"elo":1434.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.6845942368,"x_train":55.129309117,"imp":8.593085032,"elo":1676.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":1.0132382115,"x_train":5.0558157998,"imp":41.5051626445,"elo":858.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.8981443546,"x_train":10.9804562909,"imp":25.5947516248,"elo":1320.1,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1967767849,"x_train":5.6364435241,"imp":15.8345710574,"elo":1560.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.6734553632,"x_train":12.8284122123,"imp":23.8025666322,"elo":1336.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":1.2218434244,"x_train":30.6749168115,"imp":23.0365400507,"elo":1361.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.9529233417,"x_train":92.4143274115,"imp":22.0564004325,"elo":1391.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.354244056,"x_train":190.6444632314,"imp":21.4724459851,"elo":1413.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.8659233046,"x_train":476.8006390989,"imp":21.0089554552,"elo":1431.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.2670450085,"x_train":956.8892667394,"imp":20.6684352668,"elo":1446.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.7302509696,"x_train":1912.1428059667,"imp":20.4316486045,"elo":1454.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":3.9794024446,"x_train":2886.0992294917,"imp":20.3376701178,"elo":1457.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.2126633842,"x_train":3881.8318340105,"imp":20.2436587744,"elo":1462.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6126587114,"x_train":9.9408095472,"imp":14.6796200017,"elo":1645.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4885714957,"x_train":11.6635229296,"imp":13.4195320821,"elo":1587.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3630834262,"x_train":6.4658277198,"imp":21.9699175713,"elo":1396.8,"imputed":true,"imputed_pct":25.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":2.2752660768,"x_train":24.4325217028,"imp":21.9351156749,"elo":1417.6,"imputed":true,"imputed_pct":25.0,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.0816659307,"x_train":60.616973651,"imp":21.6170201631,"elo":1448.0,"imputed":true,"imputed_pct":25.0,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":5.9607949115,"x_train":127.8381423789,"imp":21.3464137897,"elo":1462.0,"imputed":true,"imputed_pct":25.0,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":7.0811072222,"x_train":315.8789741575,"imp":20.9613600395,"elo":1483.6,"imputed":true,"imputed_pct":25.0,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":8.1144193144,"x_train":633.6979487098,"imp":20.7814630166,"elo":1494.0,"imputed":true,"imputed_pct":25.0,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":9.2802561425,"x_train":1285.7447177186,"imp":20.7084108681,"elo":1496.4,"imputed":true,"imputed_pct":25.0,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.189822774,"x_train":1930.4265884275,"imp":20.6805721307,"elo":1495.5,"imputed":true,"imputed_pct":25.0,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.7913570177,"x_train":2590.7221425729,"imp":20.6616902977,"elo":1494.8,"imputed":true,"imputed_pct":25.0,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1330110917,"x_train":724.3138183772,"imp":41.8093223676,"elo":842.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.7458480887,"x_train":912.4748239301,"imp":41.5820219496,"elo":854.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.2200939284,"x_train":1342.7568425834,"imp":41.1159531554,"elo":865.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.5493158009,"x_train":2054.553219683,"imp":40.1296006426,"elo":889.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.646418544,"x_train":4867.9697698991,"imp":37.9313768562,"elo":924.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.164374429,"x_train":9985.461750605,"imp":35.6648086285,"elo":954.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9920396608,"x_train":20581.7115313909,"imp":34.4019232225,"elo":980.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":28.1227862886,"x_train":30569.2328830115,"imp":33.7036439561,"elo":997.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.2276662759,"x_train":40588.1540174129,"imp":33.1852869871,"elo":1013.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0836134922,"x_train":4.3206513889,"imp":25.7146613846,"elo":1268.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.314314201,"x_train":13.688894957,"imp":31.3941133808,"elo":1062.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5651033733,"x_train":28.9516297239,"imp":29.1184062049,"elo":1125.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9390551091,"x_train":95.1897765055,"imp":27.1982025237,"elo":1187.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1240569211,"x_train":204.1284404869,"imp":25.936645629,"elo":1236.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.364215168,"x_train":488.13995387,"imp":24.9207297786,"elo":1264.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5390396052,"x_train":949.9701252728,"imp":24.3708420151,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.725093994,"x_train":1876.0678595078,"imp":23.963974496,"elo":1294.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8858986032,"x_train":2872.3862481998,"imp":23.7238774346,"elo":1303.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0241196975,"x_train":3902.0763863206,"imp":23.4818670473,"elo":1309.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5423707001,"x_train":4.7244136606,"imp":23.4595335124,"elo":1249.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.1711649272,"x_train":15.9463545731,"imp":22.300497142,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.7546089687,"x_train":50.0714874559,"imp":21.8699882167,"elo":1329.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.6509055616,"x_train":100.9432671475,"imp":21.4538245005,"elo":1349.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.4198280312,"x_train":268.6777580855,"imp":21.1293488203,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.6537513112,"x_train":544.7091931075,"imp":20.903947403,"elo":1380.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.8978167069,"x_train":1096.7820562853,"imp":20.7214815152,"elo":1390.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.9192814316,"x_train":1644.1757072757,"imp":20.5856288602,"elo":1398.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.9829120694,"x_train":2210.16539451,"imp":20.4223786159,"elo":1408.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":211.9764977363,"x_train":514.7174339635,"imp":29.5771429609,"elo":1053.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":443.2333462935,"x_train":1597.2950547029,"imp":26.9278279178,"elo":1128.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":708.6888571295,"x_train":4783.0314636057,"imp":24.2475396394,"elo":1245.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":775.3272509789,"x_train":10691.0627275524,"imp":22.9188803133,"elo":1308.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":828.5067468066,"x_train":26917.4383401199,"imp":21.0450562142,"elo":1386.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":841.4458907481,"x_train":28234.4532410041,"imp":21.0060605589,"elo":1390.7,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.6749433874,"x_train":9.4251823487,"imp":32.1823681604,"elo":1084.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":2.2180011085,"x_train":19.1707317375,"imp":29.8777254875,"elo":1159.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.5522410409,"x_train":47.812888724,"imp":26.9417191671,"elo":1259.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":2.8792809641,"x_train":97.2172107229,"imp":25.3650225131,"elo":1294.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.1924978095,"x_train":246.1477821059,"imp":24.1161297339,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2945397285,"x_train":491.5550544333,"imp":23.5654595973,"elo":1350.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.8546394203,"x_train":987.2594279642,"imp":23.1702584153,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":3.2333264707,"x_train":1477.605852992,"imp":23.0057712232,"elo":1364.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.6075158555,"x_train":1980.6125379489,"imp":22.8929193229,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..a694288a36ddd1c700ef011e1760c0ef726aa5ec
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_multiclass/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "TabPFN-3 (default)", "RealTabPFN-2.5 (default)", "TabICLv2 (default)", "LimiX (default)", "TabPFNv2 (tuned + ensembled)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "Mitra (default)", "TabPFNv2 (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "TabPFNv2 (default)", "ModernNCA (tuned + ensembled)", "TabDPT (tuned)", "TabM (tuned)", "CatBoost (tuned)", "XGBoost (tuned)", "ModernNCA (tuned)", "XGBoost (tuned + ensembled)", "LightGBM (tuned + ensembled)", "TabICL (default)", "LightGBM (tuned)", "iLTM (tuned + ensembled)", "xRFM (tuned + ensembled)", "BetaTabPFN (default)", "ExtraTrees (tuned + ensembled)", "TabM (default)", "SAP-RPT-OSS (default)", "EBM (tuned + ensembled)", "RandomForest (tuned + ensembled)", "TabDPT (default)", "iLTM (tuned)", "TorchMLP (tuned + ensembled)", "CatBoost (default)", "ExtraTrees (tuned)", "xRFM (tuned)", "TabSwift (default)", "EBM (tuned)", "RandomForest (tuned)", "FastaiMLP (tuned + ensembled)", "XGBoost (default)", "ModernNCA (default)", "TorchMLP (tuned)", "EBM (default)", "FastaiMLP (tuned)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "RealMLP (default)", "LightGBM (default)", "xRFM (default)", "TorchMLP (default)", "ChimeraBoost (default)", "RandomForest (default)", "KNN (tuned + ensembled)", "OrionMSP (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "iLTM (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "PerpetualBooster (tuned)", "PerpetualBooster (default)", "FastaiMLP (default)", "TabFlex (default)", "TabSTAR (default)", "KNN (tuned)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.6666666865348816, 0.6602563858032227, 0.7692307829856873, 0.8141025900840759, 0.8012820482254028, 0.7564102411270142, 0.8974359035491943, 0.7948718070983887, 0.8397436141967773, 0.7820512652397156, 0.7820512652397156, 0.9230769276618958, 0.9166666865348816, 0.8910256624221802, 0.9038461446762085, 0.8846153616905212, 0.8205128312110901, 0.9102563858032227, 0.9166666865348816, 0.9230769276618958, 0.8910256624221802, 0.8333333134651184, 0.9294871687889099, 0.8846153616905212, 0.8974359035491943, 0.8910256624221802, 0.9102563858032227, 0.9743589758872986, 0.9038461446762085, 0.9230769276618958, 0.9230769276618958, 0.9166666865348816, 0.9294871687889099, 0.8589743375778198, 0.7692307829856873, 0.9102563858032227, 0.9102563858032227, 0.9102563858032227, 0.9358974099159241, 0.9294871687889099, 0.9358974099159241, 0.9230769276618958, 0.9294871687889099, 0.9230769276618958, 0.9294871687889099, 0.9102563858032227, 0.9038461446762085, 0.9358974099159241, 0.9807692170143127, 0.9358974099159241, 0.9294871687889099, 0.942307710647583, 0.9294871687889099, 0.9294871687889099, 0.9807692170143127, 0.9294871687889099, 0.9294871687889099, 0.9358974099159241, 0.942307710647583, 0.9935897588729858, 0.9358974099159241, 0.9358974099159241, 1.0, 0.9935897588729858, 0.942307710647583, 0.942307710647583, 0.9358974099159241, 0.9935897588729858, 0.9935897588729858, 0.9743589758872986, 0.9679487347602844, 1.0, 0.9679487347602844, 0.9935897588729858, 0.9358974099159241, 1.0, 1.0], [0.3333333432674408, NaN, 0.43589743971824646, 0.5448718070983887, 0.6410256624221802, 0.5833333134651184, 0.7019230723381042, 0.6730769276618958, 0.692307710647583, 0.7243589758872986, 0.7564102411270142, 0.7820512652397156, 0.7243589758872986, 0.8141025900840759, 0.7115384340286255, 0.7435897588729858, 0.8525640964508057, 0.8525640964508057, 0.9294871687889099, 0.8846153616905212, 0.8589743375778198, 0.7884615659713745, 0.8589743375778198, 0.9294871687889099, 0.9102563858032227, 0.8717948794364929, 0.9102563858032227, 0.8910256624221802, 0.8012820482254028, 0.9166666865348816, 0.8846153616905212, 0.8974359035491943, 0.9038461446762085, 0.8717948794364929, 0.8653846383094788, 0.7628205418586731, 0.9551281929016113, 0.8846153616905212, 0.807692289352417, 0.9551281929016113, 0.9487179517745972, 0.9679487347602844, 0.8974359035491943, 0.9230769276618958, 0.9743589758872986, 0.9615384340286255, 0.9102563858032227, 0.9294871687889099, 0.9807692170143127, 0.9487179517745972, 0.9487179517745972, 0.9487179517745972, 0.9679487347602844, 0.9551281929016113, 0.9679487347602844, 0.9935897588729858, 0.9487179517745972, 0.9807692170143127, 0.9935897588729858, 0.9615384340286255, 0.9871794581413269, 0.9358974099159241, 0.9807692170143127, 0.9935897588729858, 0.9935897588729858, 1.0, 0.9871794581413269, 0.9871794581413269, 1.0, 0.9871794581413269, 0.9743589758872986, 0.9679487347602844, 1.0, 1.0, 0.9871794581413269, 0.9679487347602844, 1.0, 1.0], [0.33974358439445496, 0.5641025900840759, NaN, 0.5833333134651184, 0.6474359035491943, 0.5961538553237915, 0.7628205418586731, 0.6794871687889099, 0.6987179517745972, 0.7692307829856873, 0.75, 0.7628205418586731, 0.7243589758872986, 0.807692289352417, 0.7051281929016113, 0.7371794581413269, 0.8589743375778198, 0.807692289352417, 0.942307710647583, 0.8782051205635071, 0.8653846383094788, 0.7564102411270142, 0.807692289352417, 0.9294871687889099, 0.9038461446762085, 0.8846153616905212, 0.9230769276618958, 0.8974359035491943, 0.8012820482254028, 0.9038461446762085, 0.8910256624221802, 0.8846153616905212, 0.9166666865348816, 0.8525640964508057, 0.8525640964508057, 0.7564102411270142, 0.9358974099159241, 0.8782051205635071, 0.7884615659713745, 0.9551281929016113, 0.942307710647583, 0.942307710647583, 0.8782051205635071, 0.9230769276618958, 0.9743589758872986, 0.942307710647583, 0.9102563858032227, 0.9358974099159241, 0.9871794581413269, 0.9358974099159241, 0.9487179517745972, 0.9358974099159241, 0.9551281929016113, 0.942307710647583, 0.9551281929016113, 0.9871794581413269, 0.9358974099159241, 0.9679487347602844, 0.9871794581413269, 0.942307710647583, 0.9871794581413269, 0.9294871687889099, 0.9871794581413269, 0.9935897588729858, 0.9935897588729858, 1.0, 0.9935897588729858, 0.9935897588729858, 0.9935897588729858, 0.9871794581413269, 0.9743589758872986, 0.9743589758872986, 0.9935897588729858, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.23076923191547394, 0.45512819290161133, 0.4166666567325592, NaN, 0.6217948794364929, 0.557692289352417, 0.5128205418586731, 0.5961538553237915, 0.7435897588729858, 0.5833333134651184, 0.7371794581413269, 0.75, 0.7307692170143127, 0.7115384340286255, 0.75, 0.8269230723381042, 0.807692289352417, 0.7820512652397156, 0.8910256624221802, 0.8333333134651184, 0.8717948794364929, 0.8653846383094788, 0.7756410241127014, 0.9038461446762085, 0.8397436141967773, 0.8910256624221802, 0.8653846383094788, 0.9102563858032227, 0.9038461446762085, 0.8974359035491943, 0.9166666865348816, 0.9102563858032227, 0.8333333134651184, 0.9038461446762085, 0.8141025900840759, 0.6474359035491943, 0.9230769276618958, 0.9102563858032227, 0.8461538553237915, 0.942307710647583, 0.9102563858032227, 0.9487179517745972, 0.9038461446762085, 0.942307710647583, 0.9871794581413269, 0.9487179517745972, 0.9102563858032227, 0.8782051205635071, 0.9935897588729858, 0.9679487347602844, 0.9551281929016113, 0.942307710647583, 0.942307710647583, 0.9743589758872986, 0.9871794581413269, 0.9807692170143127, 0.9743589758872986, 0.9871794581413269, 0.9871794581413269, 0.9871794581413269, 0.9935897588729858, 0.9807692170143127, 0.9935897588729858, 0.9935897588729858, 0.9743589758872986, 1.0, 0.9935897588729858, 0.9935897588729858, 0.9807692170143127, 0.9935897588729858, 0.9743589758872986, 0.9679487347602844, 0.9935897588729858, 1.0, 0.9935897588729858, 0.9807692170143127, 1.0, 1.0], [0.18589743971824646, 0.3589743673801422, 0.35256409645080566, 0.3782051205635071, NaN, 0.4294871687889099, 0.5512820482254028, 0.4871794879436493, 0.6089743375778198, 0.557692289352417, 0.6538461446762085, 0.6538461446762085, 0.7307692170143127, 0.6666666865348816, 0.692307710647583, 0.6474359035491943, 0.7243589758872986, 0.7307692170143127, 0.8141025900840759, 0.7628205418586731, 0.8653846383094788, 0.7179487347602844, 0.7628205418586731, 0.8269230723381042, 0.8846153616905212, 0.8397436141967773, 0.9038461446762085, 0.8589743375778198, 0.7948718070983887, 0.8910256624221802, 0.9294871687889099, 0.8012820482254028, 0.8461538553237915, 0.8525640964508057, 0.8012820482254028, 0.6730769276618958, 0.8717948794364929, 0.8333333134651184, 0.7884615659713745, 0.9487179517745972, 0.9230769276618958, 0.9551281929016113, 0.8461538553237915, 0.8589743375778198, 0.9551281929016113, 0.9102563858032227, 0.8717948794364929, 0.9487179517745972, 0.9807692170143127, 0.9358974099159241, 0.9230769276618958, 0.9294871687889099, 0.9807692170143127, 0.9358974099159241, 0.9487179517745972, 0.9679487347602844, 0.9358974099159241, 0.9358974099159241, 0.9487179517745972, 0.9487179517745972, 1.0, 0.9551281929016113, 0.9679487347602844, 1.0, 0.9807692170143127, 1.0, 0.9935897588729858, 0.9935897588729858, 0.9871794581413269, 0.9679487347602844, 0.9551281929016113, 0.9679487347602844, 1.0, 1.0, 0.9935897588729858, 0.9615384340286255, 1.0, 1.0], [0.19871795177459717, 0.4166666567325592, 0.4038461446762085, 0.4423076808452606, 0.5705128312110901, NaN, 0.4743589758872986, 0.5961538553237915, 0.6089743375778198, 0.6089743375778198, 0.5512820482254028, 0.6282051205635071, 0.6666666865348816, 0.6858974099159241, 0.7371794581413269, 0.7435897588729858, 0.7115384340286255, 0.6794871687889099, 0.8333333134651184, 0.7243589758872986, 0.8205128312110901, 0.7756410241127014, 0.7115384340286255, 0.8333333134651184, 0.7884615659713745, 0.8333333134651184, 0.7948718070983887, 0.8525640964508057, 0.9358974099159241, 0.8461538553237915, 0.8589743375778198, 0.8461538553237915, 0.7564102411270142, 0.9102563858032227, 0.7179487347602844, 0.5961538553237915, 0.8397436141967773, 0.8846153616905212, 0.8269230723381042, 0.9166666865348816, 0.8846153616905212, 0.8910256624221802, 0.9166666865348816, 0.9102563858032227, 0.9487179517745972, 0.8589743375778198, 0.9038461446762085, 0.8397436141967773, 0.9807692170143127, 0.9679487347602844, 0.8974359035491943, 0.9230769276618958, 0.942307710647583, 0.9102563858032227, 0.9294871687889099, 0.9871794581413269, 0.9294871687889099, 0.9807692170143127, 0.9679487347602844, 0.9358974099159241, 1.0, 0.9487179517745972, 0.9935897588729858, 1.0, 0.9871794581413269, 0.9935897588729858, 1.0, 0.9807692170143127, 0.9871794581413269, 0.9743589758872986, 0.9743589758872986, 0.9679487347602844, 0.9935897588729858, 1.0, 1.0, 0.9487179517745972, 1.0, 1.0], [0.24358974397182465, 0.29807692766189575, 0.2371794879436493, 0.4871794879436493, 0.44871795177459717, 0.5256410241127014, NaN, 0.5448718070983887, 0.6858974099159241, 0.5769230723381042, 0.6602563858032227, 0.6730769276618958, 0.6538461446762085, 0.7115384340286255, 0.5897436141967773, 0.6794871687889099, 0.7435897588729858, 0.8012820482254028, 0.8269230723381042, 0.8589743375778198, 0.7179487347602844, 0.6987179517745972, 0.8205128312110901, 0.8525640964508057, 0.8333333134651184, 0.7820512652397156, 0.8269230723381042, 0.8012820482254028, 0.7692307829856873, 0.8269230723381042, 0.7692307829856873, 0.8269230723381042, 0.8589743375778198, 0.8141025900840759, 0.8269230723381042, 0.7307692170143127, 0.8525640964508057, 0.7756410241127014, 0.7179487347602844, 0.9038461446762085, 0.9038461446762085, 0.8782051205635071, 0.8525640964508057, 0.8782051205635071, 0.9358974099159241, 0.8589743375778198, 0.8589743375778198, 0.9230769276618958, 0.9102563858032227, 0.9102563858032227, 0.9038461446762085, 0.8974359035491943, 0.942307710647583, 0.8782051205635071, 0.8846153616905212, 0.9935897588729858, 0.9166666865348816, 0.9358974099159241, 0.9615384340286255, 0.9358974099159241, 0.9615384340286255, 0.8846153616905212, 0.9871794581413269, 0.9935897588729858, 1.0, 0.9871794581413269, 0.9871794581413269, 0.9807692170143127, 0.9935897588729858, 0.9871794581413269, 0.9743589758872986, 0.9743589758872986, 0.9935897588729858, 0.9935897588729858, 1.0, 0.9487179517745972, 1.0, 1.0], [0.10256410390138626, 0.32692307233810425, 0.3205128312110901, 0.4038461446762085, 0.5128205418586731, 0.4038461446762085, 0.45512819290161133, NaN, 0.5833333134651184, 0.5448718070983887, 0.5833333134651184, 0.6089743375778198, 0.7115384340286255, 0.6282051205635071, 0.692307710647583, 0.75, 0.6730769276618958, 0.6858974099159241, 0.7884615659713745, 0.7179487347602844, 0.7564102411270142, 0.75, 0.692307710647583, 0.7884615659713745, 0.7692307829856873, 0.8205128312110901, 0.7820512652397156, 0.8397436141967773, 0.9615384340286255, 0.8589743375778198, 0.8782051205635071, 0.7948718070983887, 0.75, 0.8269230723381042, 0.75, 0.5384615659713745, 0.8269230723381042, 0.8717948794364929, 0.7628205418586731, 0.8846153616905212, 0.8333333134651184, 0.8910256624221802, 0.8461538553237915, 0.8269230723381042, 0.9230769276618958, 0.8589743375778198, 0.8782051205635071, 0.8269230723381042, 0.9294871687889099, 0.9551281929016113, 0.8846153616905212, 0.9102563858032227, 0.9038461446762085, 0.9230769276618958, 0.9294871687889099, 0.9294871687889099, 0.9294871687889099, 0.8974359035491943, 0.8974359035491943, 0.9358974099159241, 0.9935897588729858, 0.9358974099159241, 0.9166666865348816, 0.9743589758872986, 0.9871794581413269, 0.942307710647583, 0.9358974099159241, 0.9358974099159241, 0.9807692170143127, 0.9743589758872986, 0.9679487347602844, 0.9679487347602844, 0.9871794581413269, 0.9487179517745972, 0.9935897588729858, 0.942307710647583, 1.0, 1.0], [0.20512820780277252, 0.3076923191547394, 0.30128204822540283, 0.25641027092933655, 0.3910256326198578, 0.3910256326198578, 0.31410256028175354, 0.4166666567325592, NaN, 0.4743589758872986, 0.5128205418586731, 0.5192307829856873, 0.49358972907066345, 0.5512820482254028, 0.49358972907066345, 0.5448718070983887, 0.6602563858032227, 0.6538461446762085, 0.7435897588729858, 0.6346153616905212, 0.6538461446762085, 0.6282051205635071, 0.6602563858032227, 0.7564102411270142, 0.7756410241127014, 0.7820512652397156, 0.7692307829856873, 0.8012820482254028, 0.6858974099159241, 0.807692289352417, 0.7628205418586731, 0.8461538553237915, 0.7628205418586731, 0.8910256624221802, 0.7243589758872986, 0.6089743375778198, 0.8333333134651184, 0.9102563858032227, 0.6282051205635071, 0.8397436141967773, 0.7179487347602844, 0.8846153616905212, 0.9102563858032227, 0.9038461446762085, 0.8205128312110901, 0.8846153616905212, 0.9038461446762085, 0.7820512652397156, 0.9487179517745972, 0.942307710647583, 0.8269230723381042, 0.9038461446762085, 0.8910256624221802, 0.9166666865348816, 0.942307710647583, 0.9807692170143127, 0.942307710647583, 0.9743589758872986, 0.9615384340286255, 0.9487179517745972, 1.0, 0.9487179517745972, 0.9743589758872986, 1.0, 0.9871794581413269, 0.9743589758872986, 0.9807692170143127, 0.9679487347602844, 0.9935897588729858, 0.9807692170143127, 0.9743589758872986, 0.9679487347602844, 1.0, 0.9935897588729858, 1.0, 0.9551281929016113, 1.0, 1.0], [0.16025641560554504, 0.2756410241127014, 0.23076923191547394, 0.4166666567325592, 0.4423076808452606, 0.3910256326198578, 0.42307692766189575, 0.45512819290161133, 0.5256410241127014, NaN, 0.557692289352417, 0.6089743375778198, 0.557692289352417, 0.6506410241127014, 0.5705128312110901, 0.5833333134651184, 0.7628205418586731, 0.7243589758872986, 0.7692307829856873, 0.6474359035491943, 0.6794871687889099, 0.6410256624221802, 0.7371794581413269, 0.7948718070983887, 0.7820512652397156, 0.7371794581413269, 0.7948718070983887, 0.8141025900840759, 0.6538461446762085, 0.8333333134651184, 0.7628205418586731, 0.8397436141967773, 0.8717948794364929, 0.7435897588729858, 0.8269230723381042, 0.6025640964508057, 0.8269230723381042, 0.7628205418586731, 0.6538461446762085, 0.8461538553237915, 0.8269230723381042, 0.8782051205635071, 0.7884615659713745, 0.8461538553237915, 0.8461538553237915, 0.8397436141967773, 0.7948718070983887, 0.8846153616905212, 0.8653846383094788, 0.8141025900840759, 0.8397436141967773, 0.8653846383094788, 0.9038461446762085, 0.8782051205635071, 0.8717948794364929, 0.9230769276618958, 0.9102563858032227, 0.8974359035491943, 0.9102563858032227, 0.8782051205635071, 0.8461538553237915, 0.8974359035491943, 0.9166666865348816, 0.942307710647583, 0.9871794581413269, 0.8846153616905212, 0.9294871687889099, 0.9294871687889099, 0.9935897588729858, 0.9743589758872986, 0.9679487347602844, 0.9679487347602844, 0.942307710647583, 0.942307710647583, 0.9358974099159241, 0.942307710647583, 1.0, 1.0], [0.21794871985912323, 0.24358974397182465, 0.25, 0.2628205120563507, 0.3461538553237915, 0.44871795177459717, 0.33974358439445496, 0.4166666567325592, 0.4871794879436493, 0.4423076808452606, NaN, 0.5288461446762085, 0.5320512652397156, 0.5769230723381042, 0.5320512652397156, 0.5192307829856873, 0.6666666865348816, 0.692307710647583, 0.7371794581413269, 0.5705128312110901, 0.6858974099159241, 0.5256410241127014, 0.6987179517745972, 0.7692307829856873, 0.8461538553237915, 0.7371794581413269, 0.8717948794364929, 0.8269230723381042, 0.5833333134651184, 0.8269230723381042, 0.7564102411270142, 0.8333333134651184, 0.8653846383094788, 0.7435897588729858, 0.8782051205635071, 0.6282051205635071, 0.8653846383094788, 0.8141025900840759, 0.5769230723381042, 0.8525640964508057, 0.7564102411270142, 0.8782051205635071, 0.807692289352417, 0.8589743375778198, 0.7435897588729858, 0.8717948794364929, 0.8589743375778198, 0.9102563858032227, 0.9358974099159241, 0.8910256624221802, 0.807692289352417, 0.8782051205635071, 0.9166666865348816, 0.8846153616905212, 0.8910256624221802, 0.9935897588729858, 0.9166666865348816, 0.9358974099159241, 0.9743589758872986, 0.8846153616905212, 0.9358974099159241, 0.9230769276618958, 0.9871794581413269, 1.0, 0.9935897588729858, 0.942307710647583, 0.9743589758872986, 0.9679487347602844, 0.9871794581413269, 0.9807692170143127, 0.9743589758872986, 0.9679487347602844, 1.0, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.21794871985912323, 0.21794871985912323, 0.2371794879436493, 0.25, 0.3461538553237915, 0.3717948794364929, 0.32692307233810425, 0.3910256326198578, 0.48076921701431274, 0.3910256326198578, 0.4711538553237915, NaN, 0.4615384638309479, 0.5, 0.5192307829856873, 0.48076921701431274, 0.6153846383094788, 0.6217948794364929, 0.7051281929016113, 0.5512820482254028, 0.692307710647583, 0.5128205418586731, 0.6602563858032227, 0.7115384340286255, 0.7820512652397156, 0.7115384340286255, 0.807692289352417, 0.7435897588729858, 0.5705128312110901, 0.7564102411270142, 0.7243589758872986, 0.7756410241127014, 0.8525640964508057, 0.7371794581413269, 0.8141025900840759, 0.6025640964508057, 0.8461538553237915, 0.7884615659713745, 0.5833333134651184, 0.8333333134651184, 0.8012820482254028, 0.8397436141967773, 0.7756410241127014, 0.8397436141967773, 0.7435897588729858, 0.8846153616905212, 0.807692289352417, 0.8782051205635071, 0.9038461446762085, 0.8653846383094788, 0.8461538553237915, 0.8653846383094788, 0.9358974099159241, 0.8653846383094788, 0.8846153616905212, 0.9807692170143127, 0.9038461446762085, 0.9102563858032227, 0.9679487347602844, 0.8910256624221802, 0.9294871687889099, 0.9294871687889099, 0.9807692170143127, 0.9807692170143127, 0.9871794581413269, 0.9358974099159241, 0.9743589758872986, 0.9615384340286255, 0.9935897588729858, 0.9871794581413269, 0.9743589758872986, 0.9615384340286255, 1.0, 1.0, 0.9871794581413269, 0.9743589758872986, 1.0, 1.0], [0.07692307978868484, 0.2756410241127014, 0.2756410241127014, 0.26923078298568726, 0.26923078298568726, 0.3333333432674408, 0.3461538553237915, 0.2884615361690521, 0.5064102411270142, 0.4423076808452606, 0.46794870495796204, 0.5384615659713745, NaN, 0.5897436141967773, 0.5192307829856873, 0.5320512652397156, 0.6474359035491943, 0.5897436141967773, 0.6794871687889099, 0.6153846383094788, 0.7243589758872986, 0.5897436141967773, 0.6217948794364929, 0.6730769276618958, 0.6858974099159241, 0.7243589758872986, 0.7179487347602844, 0.8012820482254028, 0.6314102411270142, 0.7884615659713745, 0.807692289352417, 0.7692307829856873, 0.7564102411270142, 0.7307692170143127, 0.7307692170143127, 0.5, 0.7692307829856873, 0.7692307829856873, 0.6538461446762085, 0.8269230723381042, 0.7884615659713745, 0.8653846383094788, 0.7628205418586731, 0.8205128312110901, 0.8269230723381042, 0.807692289352417, 0.7948718070983887, 0.8333333134651184, 0.8782051205635071, 0.8269230723381042, 0.8269230723381042, 0.8525640964508057, 0.8910256624221802, 0.8717948794364929, 0.8653846383094788, 0.8846153616905212, 0.9294871687889099, 0.8397436141967773, 0.8846153616905212, 0.8846153616905212, 0.9358974099159241, 0.942307710647583, 0.9102563858032227, 0.942307710647583, 0.9807692170143127, 0.8846153616905212, 0.9294871687889099, 0.9294871687889099, 0.9679487347602844, 0.9551281929016113, 0.9551281929016113, 0.9615384340286255, 0.9358974099159241, 0.942307710647583, 0.9358974099159241, 0.942307710647583, 1.0, 1.0], [0.0833333358168602, 0.18589743971824646, 0.19230769574642181, 0.2884615361690521, 0.3333333432674408, 0.31410256028175354, 0.2884615361690521, 0.3717948794364929, 0.44871795177459717, 0.3493589758872986, 0.42307692766189575, 0.5, 0.41025641560554504, NaN, 0.5256410241127014, 0.5512820482254028, 0.6538461446762085, 0.5769230723381042, 0.6410256624221802, 0.5705128312110901, 0.6025640964508057, 0.6217948794364929, 0.6282051205635071, 0.7115384340286255, 0.6410256624221802, 0.6666666865348816, 0.6602563858032227, 0.7051281929016113, 0.6153846383094788, 0.7179487347602844, 0.6730769276618958, 0.7243589758872986, 0.7756410241127014, 0.6794871687889099, 0.7307692170143127, 0.5320512652397156, 0.7692307829856873, 0.692307710647583, 0.6346153616905212, 0.7820512652397156, 0.7820512652397156, 0.7756410241127014, 0.7243589758872986, 0.8205128312110901, 0.7948718070983887, 0.7820512652397156, 0.7756410241127014, 0.8205128312110901, 0.807692289352417, 0.7564102411270142, 0.8205128312110901, 0.8205128312110901, 0.8717948794364929, 0.8141025900840759, 0.8141025900840759, 0.9166666865348816, 0.8461538553237915, 0.8589743375778198, 0.9038461446762085, 0.8525640964508057, 0.8397436141967773, 0.8846153616905212, 0.8974359035491943, 0.9294871687889099, 0.9679487347602844, 0.8846153616905212, 0.9102563858032227, 0.8974359035491943, 0.9679487347602844, 0.9679487347602844, 0.9615384340286255, 0.9743589758872986, 0.9230769276618958, 0.942307710647583, 0.9358974099159241, 0.942307710647583, 1.0, 1.0], [0.10897435992956161, 0.2884615361690521, 0.29487180709838867, 0.25, 0.3076923191547394, 0.2628205120563507, 0.41025641560554504, 0.3076923191547394, 0.5064102411270142, 0.4294871687889099, 0.46794870495796204, 0.48076921701431274, 0.48076921701431274, 0.4743589758872986, NaN, 0.5512820482254028, 0.5128205418586731, 0.5192307829856873, 0.6089743375778198, 0.5961538553237915, 0.6794871687889099, 0.6025640964508057, 0.5384615659713745, 0.5961538553237915, 0.6153846383094788, 0.6730769276618958, 0.6153846383094788, 0.6666666865348816, 0.6730769276618958, 0.6153846383094788, 0.807692289352417, 0.6282051205635071, 0.5769230723381042, 0.7692307829856873, 0.5833333134651184, 0.5256410241127014, 0.6474359035491943, 0.7564102411270142, 0.8012820482254028, 0.8141025900840759, 0.7179487347602844, 0.7564102411270142, 0.7820512652397156, 0.692307710647583, 0.8333333134651184, 0.7179487347602844, 0.7756410241127014, 0.7884615659713745, 0.8525640964508057, 0.8397436141967773, 0.7820512652397156, 0.7820512652397156, 0.8717948794364929, 0.8910256624221802, 0.8910256624221802, 0.8525640964508057, 0.8589743375778198, 0.7884615659713745, 0.8525640964508057, 0.8910256624221802, 1.0, 0.9230769276618958, 0.942307710647583, 1.0, 0.8717948794364929, 0.942307710647583, 0.8846153616905212, 0.8846153616905212, 0.8589743375778198, 0.942307710647583, 0.942307710647583, 0.9487179517745972, 0.942307710647583, 1.0, 0.942307710647583, 0.9487179517745972, 0.9679487347602844, 1.0], [0.09615384787321091, 0.25641027092933655, 0.2628205120563507, 0.17307692766189575, 0.35256409645080566, 0.25641027092933655, 0.3205128312110901, 0.25, 0.45512819290161133, 0.4166666567325592, 0.48076921701431274, 0.5192307829856873, 0.46794870495796204, 0.44871795177459717, 0.44871795177459717, NaN, 0.5256410241127014, 0.5705128312110901, 0.5448718070983887, 0.5705128312110901, 0.6153846383094788, 0.6794871687889099, 0.5769230723381042, 0.5769230723381042, 0.5961538553237915, 0.6410256624221802, 0.6153846383094788, 0.6730769276618958, 0.6346153616905212, 0.6538461446762085, 0.7115384340286255, 0.6538461446762085, 0.6538461446762085, 0.692307710647583, 0.6538461446762085, 0.5192307829856873, 0.692307710647583, 0.75, 0.7628205418586731, 0.7435897588729858, 0.6730769276618958, 0.7435897588729858, 0.7884615659713745, 0.7564102411270142, 0.8333333134651184, 0.7243589758872986, 0.7884615659713745, 0.7820512652397156, 0.8461538553237915, 0.8589743375778198, 0.8012820482254028, 0.75, 0.8269230723381042, 0.8397436141967773, 0.8653846383094788, 0.8653846383094788, 0.8653846383094788, 0.8589743375778198, 0.8974359035491943, 0.8782051205635071, 0.9807692170143127, 0.9679487347602844, 0.9615384340286255, 1.0, 0.8653846383094788, 0.8846153616905212, 0.9102563858032227, 0.9102563858032227, 0.8782051205635071, 0.9615384340286255, 0.9615384340286255, 0.9551281929016113, 0.942307710647583, 1.0, 0.9358974099159241, 0.9743589758872986, 0.9551281929016113, 1.0], [0.11538461595773697, 0.14743590354919434, 0.14102564752101898, 0.19230769574642181, 0.2756410241127014, 0.2884615361690521, 0.25641027092933655, 0.32692307233810425, 0.33974358439445496, 0.2371794879436493, 0.3333333432674408, 0.38461539149284363, 0.35256409645080566, 0.3461538553237915, 0.4871794879436493, 0.4743589758872986, NaN, 0.5064102411270142, 0.6025640964508057, 0.48076921701431274, 0.6153846383094788, 0.5064102411270142, 0.5448718070983887, 0.6730769276618958, 0.6153846383094788, 0.6474359035491943, 0.6474359035491943, 0.7371794581413269, 0.5897436141967773, 0.7243589758872986, 0.6858974099159241, 0.6794871687889099, 0.6666666865348816, 0.6153846383094788, 0.6858974099159241, 0.5384615659713745, 0.7884615659713745, 0.6730769276618958, 0.6217948794364929, 0.7564102411270142, 0.7564102411270142, 0.8269230723381042, 0.6987179517745972, 0.7435897588729858, 0.7243589758872986, 0.8141025900840759, 0.7692307829856873, 0.8333333134651184, 0.8461538553237915, 0.8269230723381042, 0.8397436141967773, 0.8717948794364929, 0.9038461446762085, 0.8846153616905212, 0.8910256624221802, 0.8782051205635071, 0.8653846383094788, 0.8012820482254028, 0.9166666865348816, 0.9294871687889099, 0.9038461446762085, 0.8333333134651184, 0.9102563858032227, 0.9294871687889099, 0.9487179517745972, 0.9807692170143127, 0.942307710647583, 0.942307710647583, 0.9615384340286255, 0.942307710647583, 0.9487179517745972, 0.9551281929016113, 0.9807692170143127, 0.9807692170143127, 0.9679487347602844, 0.9551281929016113, 0.9935897588729858, 1.0], [0.1794871836900711, 0.14743590354919434, 0.19230769574642181, 0.21794871985912323, 0.26923078298568726, 0.3205128312110901, 0.19871795177459717, 0.31410256028175354, 0.3461538553237915, 0.2756410241127014, 0.3076923191547394, 0.3782051205635071, 0.41025641560554504, 0.42307692766189575, 0.48076921701431274, 0.4294871687889099, 0.49358972907066345, NaN, 0.5, 0.39743590354919434, 0.5769230723381042, 0.5064102411270142, 0.6442307829856873, 0.5641025900840759, 0.6474359035491943, 0.5769230723381042, 0.6730769276618958, 0.6217948794364929, 0.5128205418586731, 0.6282051205635071, 0.5897436141967773, 0.6410256624221802, 0.7756410241127014, 0.5448718070983887, 0.8846153616905212, 0.6410256624221802, 0.8141025900840759, 0.6025640964508057, 0.6153846383094788, 0.6987179517745972, 0.7371794581413269, 0.7820512652397156, 0.6025640964508057, 0.7179487347602844, 0.6474359035491943, 0.8397436141967773, 0.7115384340286255, 0.8269230723381042, 0.8333333134651184, 0.7820512652397156, 0.8205128312110901, 0.8397436141967773, 0.8846153616905212, 0.807692289352417, 0.8333333134651184, 0.9551281929016113, 0.8910256624221802, 0.8141025900840759, 0.9487179517745972, 0.8974359035491943, 0.8910256624221802, 0.8910256624221802, 0.942307710647583, 0.9166666865348816, 0.9615384340286255, 0.942307710647583, 0.942307710647583, 0.9551281929016113, 0.9807692170143127, 0.9551281929016113, 0.9679487347602844, 0.9551281929016113, 0.9807692170143127, 1.0, 0.9935897588729858, 0.9679487347602844, 1.0, 1.0], [0.08974359184503555, 0.07051282376050949, 0.057692307978868484, 0.10897435992956161, 0.18589743971824646, 0.1666666716337204, 0.17307692766189575, 0.21153846383094788, 0.25641027092933655, 0.23076923191547394, 0.2628205120563507, 0.29487180709838867, 0.3205128312110901, 0.3589743673801422, 0.3910256326198578, 0.45512819290161133, 0.39743590354919434, 0.5, NaN, 0.4038461446762085, 0.5192307829856873, 0.5256410241127014, 0.5641025900840759, 0.6955128312110901, 0.6153846383094788, 0.5512820482254028, 0.6282051205635071, 0.5961538553237915, 0.5833333134651184, 0.6025640964508057, 0.5384615659713745, 0.6025640964508057, 0.6217948794364929, 0.6538461446762085, 0.6666666865348816, 0.5384615659713745, 0.7051281929016113, 0.6602563858032227, 0.5961538553237915, 0.692307710647583, 0.6602563858032227, 0.8333333134651184, 0.7371794581413269, 0.7756410241127014, 0.7435897588729858, 0.7884615659713745, 0.7820512652397156, 0.7051281929016113, 0.9230769276618958, 0.8333333134651184, 0.8012820482254028, 0.8782051205635071, 0.8333333134651184, 0.8782051205635071, 0.9166666865348816, 0.9487179517745972, 0.8910256624221802, 0.9038461446762085, 0.9615384340286255, 0.9230769276618958, 0.9230769276618958, 0.8717948794364929, 0.9743589758872986, 0.9487179517745972, 0.9615384340286255, 1.0, 0.9807692170143127, 0.9743589758872986, 0.9679487347602844, 0.9743589758872986, 0.9743589758872986, 0.9679487347602844, 0.9807692170143127, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.0833333358168602, 0.11538461595773697, 0.12179487198591232, 0.1666666716337204, 0.2371794879436493, 0.2756410241127014, 0.14102564752101898, 0.28205129504203796, 0.36538460850715637, 0.35256409645080566, 0.4294871687889099, 0.44871795177459717, 0.38461539149284363, 0.4294871687889099, 0.4038461446762085, 0.4294871687889099, 0.5192307829856873, 0.6025640964508057, 0.5961538553237915, NaN, 0.4871794879436493, 0.5064102411270142, 0.6602563858032227, 0.6602563858032227, 0.6346153616905212, 0.5833333134651184, 0.6474359035491943, 0.6410256624221802, 0.5544871687889099, 0.6346153616905212, 0.5641025900840759, 0.6987179517745972, 0.7371794581413269, 0.6666666865348816, 0.7179487347602844, 0.49358972907066345, 0.7435897588729858, 0.6730769276618958, 0.5320512652397156, 0.6538461446762085, 0.6666666865348816, 0.7564102411270142, 0.692307710647583, 0.7948718070983887, 0.7371794581413269, 0.7692307829856873, 0.7564102411270142, 0.7115384340286255, 0.7820512652397156, 0.7307692170143127, 0.7243589758872986, 0.8012820482254028, 0.8205128312110901, 0.7884615659713745, 0.7948718070983887, 0.8589743375778198, 0.8589743375778198, 0.8397436141967773, 0.8717948794364929, 0.8333333134651184, 0.8589743375778198, 0.8653846383094788, 0.9038461446762085, 0.8653846383094788, 0.942307710647583, 0.8397436141967773, 0.9166666865348816, 0.9102563858032227, 0.9294871687889099, 0.9807692170143127, 0.9679487347602844, 0.9679487347602844, 0.9102563858032227, 0.9166666865348816, 0.9294871687889099, 0.9358974099159241, 0.9935897588729858, 1.0], [0.07692307978868484, 0.14102564752101898, 0.13461539149284363, 0.12820513546466827, 0.13461539149284363, 0.1794871836900711, 0.28205129504203796, 0.24358974397182465, 0.3461538553237915, 0.3205128312110901, 0.31410256028175354, 0.3076923191547394, 0.2756410241127014, 0.39743590354919434, 0.3205128312110901, 0.38461539149284363, 0.38461539149284363, 0.42307692766189575, 0.48076921701431274, 0.5128205418586731, NaN, 0.44871795177459717, 0.46794870495796204, 0.5128205418586731, 0.5384615659713745, 0.6442307829856873, 0.5512820482254028, 0.5961538553237915, 0.5448718070983887, 0.6089743375778198, 0.5256410241127014, 0.6602563858032227, 0.5705128312110901, 0.7435897588729858, 0.5705128312110901, 0.557692289352417, 0.6538461446762085, 0.7564102411270142, 0.5769230723381042, 0.692307710647583, 0.6153846383094788, 0.6987179517745972, 0.8141025900840759, 0.807692289352417, 0.6730769276618958, 0.7179487347602844, 0.7692307829856873, 0.6858974099159241, 0.8205128312110901, 0.9102563858032227, 0.7756410241127014, 0.7820512652397156, 0.8141025900840759, 0.8397436141967773, 0.8333333134651184, 0.8910256624221802, 0.8141025900840759, 0.8974359035491943, 0.9294871687889099, 0.8910256624221802, 0.9487179517745972, 0.9166666865348816, 0.9551281929016113, 0.9871794581413269, 0.9166666865348816, 0.9807692170143127, 0.9166666865348816, 0.9102563858032227, 0.9615384340286255, 0.9807692170143127, 0.9743589758872986, 0.9551281929016113, 0.9487179517745972, 0.9935897588729858, 0.9679487347602844, 0.9358974099159241, 0.9935897588729858, 1.0], [0.10897435992956161, 0.21153846383094788, 0.24358974397182465, 0.13461539149284363, 0.28205129504203796, 0.22435897588729858, 0.30128204822540283, 0.25, 0.3717948794364929, 0.3589743673801422, 0.4743589758872986, 0.4871794879436493, 0.41025641560554504, 0.3782051205635071, 0.39743590354919434, 0.3205128312110901, 0.49358972907066345, 0.49358972907066345, 0.4743589758872986, 0.49358972907066345, 0.5512820482254028, NaN, 0.5448718070983887, 0.5128205418586731, 0.5641025900840759, 0.5961538553237915, 0.5897436141967773, 0.6025640964508057, 0.5705128312110901, 0.6153846383094788, 0.6089743375778198, 0.5833333134651184, 0.6025640964508057, 0.6346153616905212, 0.6410256624221802, 0.49358972907066345, 0.5897436141967773, 0.6602563858032227, 0.7307692170143127, 0.6410256624221802, 0.5961538553237915, 0.6602563858032227, 0.6794871687889099, 0.692307710647583, 0.7756410241127014, 0.6538461446762085, 0.7051281929016113, 0.7115384340286255, 0.7564102411270142, 0.7243589758872986, 0.6858974099159241, 0.6858974099159241, 0.7884615659713745, 0.807692289352417, 0.8269230723381042, 0.7628205418586731, 0.8525640964508057, 0.8269230723381042, 0.8653846383094788, 0.8653846383094788, 0.9358974099159241, 0.9615384340286255, 0.9615384340286255, 0.9807692170143127, 0.8205128312110901, 0.8782051205635071, 0.9102563858032227, 0.9102563858032227, 0.8141025900840759, 0.9358974099159241, 0.9487179517745972, 0.9551281929016113, 0.942307710647583, 1.0, 0.9294871687889099, 0.9615384340286255, 0.942307710647583, 1.0], [0.1666666716337204, 0.14102564752101898, 0.19230769574642181, 0.22435897588729858, 0.2371794879436493, 0.2884615361690521, 0.1794871836900711, 0.3076923191547394, 0.33974358439445496, 0.2628205120563507, 0.30128204822540283, 0.33974358439445496, 0.3782051205635071, 0.3717948794364929, 0.4615384638309479, 0.42307692766189575, 0.45512819290161133, 0.35576921701431274, 0.43589743971824646, 0.33974358439445496, 0.5320512652397156, 0.45512819290161133, NaN, 0.49358972907066345, 0.5641025900840759, 0.5192307829856873, 0.5769230723381042, 0.5448718070983887, 0.5, 0.5448718070983887, 0.5961538553237915, 0.6217948794364929, 0.692307710647583, 0.5384615659713745, 0.7564102411270142, 0.557692289352417, 0.7564102411270142, 0.6089743375778198, 0.6025640964508057, 0.6410256624221802, 0.7115384340286255, 0.7179487347602844, 0.5833333134651184, 0.7115384340286255, 0.6282051205635071, 0.807692289352417, 0.6794871687889099, 0.7948718070983887, 0.7948718070983887, 0.7435897588729858, 0.7884615659713745, 0.8205128312110901, 0.8461538553237915, 0.75, 0.7692307829856873, 0.9166666865348816, 0.7948718070983887, 0.7884615659713745, 0.9358974099159241, 0.8589743375778198, 0.8782051205635071, 0.8525640964508057, 0.9294871687889099, 0.8589743375778198, 0.9743589758872986, 0.942307710647583, 0.9294871687889099, 0.9551281929016113, 0.9743589758872986, 0.9615384340286255, 0.9551281929016113, 0.9487179517745972, 0.9743589758872986, 1.0, 0.9871794581413269, 0.9679487347602844, 1.0, 1.0], [0.07051282376050949, 0.07051282376050949, 0.07051282376050949, 0.09615384787321091, 0.17307692766189575, 0.1666666716337204, 0.14743590354919434, 0.21153846383094788, 0.24358974397182465, 0.20512820780277252, 0.23076923191547394, 0.2884615361690521, 0.32692307233810425, 0.2884615361690521, 0.4038461446762085, 0.42307692766189575, 0.32692307233810425, 0.43589743971824646, 0.3044871687889099, 0.33974358439445496, 0.4871794879436493, 0.4871794879436493, 0.5064102411270142, NaN, 0.5769230723381042, 0.5384615659713745, 0.5769230723381042, 0.5448718070983887, 0.5512820482254028, 0.5384615659713745, 0.5064102411270142, 0.557692289352417, 0.5833333134651184, 0.6346153616905212, 0.6538461446762085, 0.5192307829856873, 0.692307710647583, 0.6282051205635071, 0.5833333134651184, 0.6666666865348816, 0.6730769276618958, 0.7820512652397156, 0.7051281929016113, 0.7628205418586731, 0.6987179517745972, 0.7371794581413269, 0.7564102411270142, 0.692307710647583, 0.8653846383094788, 0.8012820482254028, 0.7820512652397156, 0.8525640964508057, 0.7948718070983887, 0.8525640964508057, 0.8525640964508057, 0.9102563858032227, 0.8525640964508057, 0.8782051205635071, 0.9358974099159241, 0.9230769276618958, 0.9166666865348816, 0.8525640964508057, 0.9743589758872986, 0.9294871687889099, 0.9551281929016113, 1.0, 0.9679487347602844, 0.9615384340286255, 0.9679487347602844, 0.9743589758872986, 0.9743589758872986, 0.9679487347602844, 0.9743589758872986, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.11538461595773697, 0.08974359184503555, 0.09615384787321091, 0.16025641560554504, 0.11538461595773697, 0.21153846383094788, 0.1666666716337204, 0.23076923191547394, 0.22435897588729858, 0.21794871985912323, 0.1538461595773697, 0.21794871985912323, 0.31410256028175354, 0.3589743673801422, 0.38461539149284363, 0.4038461446762085, 0.38461539149284363, 0.35256409645080566, 0.38461539149284363, 0.36538460850715637, 0.4615384638309479, 0.43589743971824646, 0.43589743971824646, 0.42307692766189575, NaN, 0.5192307829856873, 0.5288461446762085, 0.4871794879436493, 0.5256410241127014, 0.5448718070983887, 0.5064102411270142, 0.5705128312110901, 0.5769230723381042, 0.5448718070983887, 0.5769230723381042, 0.5833333134651184, 0.692307710647583, 0.557692289352417, 0.5448718070983887, 0.6410256624221802, 0.6666666865348816, 0.7628205418586731, 0.6410256624221802, 0.6987179517745972, 0.5961538553237915, 0.7371794581413269, 0.6987179517745972, 0.6858974099159241, 0.9038461446762085, 0.7948718070983887, 0.7435897588729858, 0.8525640964508057, 0.75, 0.8910256624221802, 0.8910256624221802, 0.8525640964508057, 0.8846153616905212, 0.7948718070983887, 0.9230769276618958, 0.9294871687889099, 0.9166666865348816, 0.8525640964508057, 0.9615384340286255, 0.9294871687889099, 0.9487179517745972, 0.9935897588729858, 0.9551281929016113, 0.9487179517745972, 0.9615384340286255, 0.9615384340286255, 0.9615384340286255, 0.9679487347602844, 0.9871794581413269, 0.9935897588729858, 0.9871794581413269, 0.9358974099159241, 1.0, 1.0], [0.10256410390138626, 0.12820513546466827, 0.11538461595773697, 0.10897435992956161, 0.16025641560554504, 0.1666666716337204, 0.21794871985912323, 0.1794871836900711, 0.21794871985912323, 0.2628205120563507, 0.2628205120563507, 0.2884615361690521, 0.2756410241127014, 0.3333333432674408, 0.32692307233810425, 0.3589743673801422, 0.35256409645080566, 0.42307692766189575, 0.44871795177459717, 0.4166666567325592, 0.35576921701431274, 0.4038461446762085, 0.48076921701431274, 0.4615384638309479, 0.48076921701431274, NaN, 0.5, 0.5448718070983887, 0.4615384638309479, 0.5448718070983887, 0.43589743971824646, 0.6089743375778198, 0.5320512652397156, 0.6666666865348816, 0.5705128312110901, 0.5641025900840759, 0.6346153616905212, 0.6730769276618958, 0.5705128312110901, 0.6217948794364929, 0.5961538553237915, 0.6602563858032227, 0.7307692170143127, 0.7371794581413269, 0.6987179517745972, 0.6987179517745972, 0.7243589758872986, 0.6410256624221802, 0.7564102411270142, 0.8205128312110901, 0.7692307829856873, 0.7307692170143127, 0.75, 0.7820512652397156, 0.7884615659713745, 0.8205128312110901, 0.807692289352417, 0.8653846383094788, 0.9166666865348816, 0.9166666865348816, 0.9230769276618958, 0.9230769276618958, 0.9615384340286255, 0.9679487347602844, 0.8910256624221802, 0.9743589758872986, 0.9294871687889099, 0.9358974099159241, 0.9358974099159241, 0.9679487347602844, 0.9679487347602844, 0.9487179517745972, 0.942307710647583, 0.9871794581413269, 0.9679487347602844, 0.9743589758872986, 0.9935897588729858, 1.0], [0.10897435992956161, 0.08974359184503555, 0.07692307978868484, 0.13461539149284363, 0.09615384787321091, 0.20512820780277252, 0.17307692766189575, 0.21794871985912323, 0.23076923191547394, 0.20512820780277252, 0.12820513546466827, 0.19230769574642181, 0.28205129504203796, 0.33974358439445496, 0.38461539149284363, 0.38461539149284363, 0.35256409645080566, 0.32692307233810425, 0.3717948794364929, 0.35256409645080566, 0.44871795177459717, 0.41025641560554504, 0.42307692766189575, 0.42307692766189575, 0.4711538553237915, 0.5, NaN, 0.46794870495796204, 0.5320512652397156, 0.5192307829856873, 0.5192307829856873, 0.5641025900840759, 0.5769230723381042, 0.5256410241127014, 0.5641025900840759, 0.5705128312110901, 0.7179487347602844, 0.5641025900840759, 0.5384615659713745, 0.6474359035491943, 0.6666666865348816, 0.7820512652397156, 0.6346153616905212, 0.6987179517745972, 0.6089743375778198, 0.7628205418586731, 0.6730769276618958, 0.692307710647583, 0.9294871687889099, 0.7948718070983887, 0.75, 0.8525640964508057, 0.7564102411270142, 0.8846153616905212, 0.8974359035491943, 0.8525640964508057, 0.8717948794364929, 0.7948718070983887, 0.9230769276618958, 0.9294871687889099, 0.9102563858032227, 0.8525640964508057, 0.9551281929016113, 0.9166666865348816, 0.9487179517745972, 0.9935897588729858, 0.9615384340286255, 0.9551281929016113, 0.9615384340286255, 0.9615384340286255, 0.9551281929016113, 0.9615384340286255, 0.9871794581413269, 0.9935897588729858, 0.9871794581413269, 0.9358974099159241, 1.0, 1.0], [0.08974359184503555, 0.10897435992956161, 0.10256410390138626, 0.08974359184503555, 0.14102564752101898, 0.14743590354919434, 0.19871795177459717, 0.16025641560554504, 0.19871795177459717, 0.18589743971824646, 0.17307692766189575, 0.25641027092933655, 0.19871795177459717, 0.29487180709838867, 0.3333333432674408, 0.32692307233810425, 0.2628205120563507, 0.3782051205635071, 0.4038461446762085, 0.3589743673801422, 0.4038461446762085, 0.39743590354919434, 0.45512819290161133, 0.45512819290161133, 0.5128205418586731, 0.45512819290161133, 0.5320512652397156, NaN, 0.4423076808452606, 0.5032051205635071, 0.5192307829856873, 0.5512820482254028, 0.5384615659713745, 0.5256410241127014, 0.6153846383094788, 0.557692289352417, 0.692307710647583, 0.5705128312110901, 0.5833333134651184, 0.6153846383094788, 0.6282051205635071, 0.7435897588729858, 0.5897436141967773, 0.7051281929016113, 0.6410256624221802, 0.7628205418586731, 0.7051281929016113, 0.6666666865348816, 0.8525640964508057, 0.7692307829856873, 0.7435897588729858, 0.8205128312110901, 0.7948718070983887, 0.9487179517745972, 0.9871794581413269, 0.8397436141967773, 0.942307710647583, 0.7692307829856873, 0.8910256624221802, 0.9807692170143127, 0.9038461446762085, 0.8589743375778198, 0.9230769276618958, 0.9487179517745972, 0.8910256624221802, 1.0, 0.9487179517745972, 0.9487179517745972, 0.9294871687889099, 0.942307710647583, 0.9551281929016113, 0.942307710647583, 1.0, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.025641025975346565, 0.19871795177459717, 0.19871795177459717, 0.09615384787321091, 0.20512820780277252, 0.06410256773233414, 0.23076923191547394, 0.03846153989434242, 0.31410256028175354, 0.3461538553237915, 0.4166666567325592, 0.4294871687889099, 0.36858972907066345, 0.38461539149284363, 0.32692307233810425, 0.36538460850715637, 0.41025641560554504, 0.4871794879436493, 0.4166666567325592, 0.4455128312110901, 0.45512819290161133, 0.4294871687889099, 0.5, 0.44871795177459717, 0.4743589758872986, 0.5384615659713745, 0.46794870495796204, 0.557692289352417, NaN, 0.5512820482254028, 0.5641025900840759, 0.5256410241127014, 0.5384615659713745, 0.6987179517745972, 0.5641025900840759, 0.4423076808452606, 0.5192307829856873, 0.7051281929016113, 0.6282051205635071, 0.6666666865348816, 0.5064102411270142, 0.6217948794364929, 0.7628205418586731, 0.6538461446762085, 0.8141025900840759, 0.5833333134651184, 0.7371794581413269, 0.6346153616905212, 0.7179487347602844, 0.7179487347602844, 0.692307710647583, 0.6794871687889099, 0.7051281929016113, 0.7692307829856873, 0.7884615659713745, 0.6987179517745972, 0.8589743375778198, 0.7564102411270142, 0.7884615659713745, 0.8717948794364929, 0.9455128312110901, 0.9358974099159241, 0.9038461446762085, 0.9358974099159241, 0.8205128312110901, 0.8974359035491943, 0.8589743375778198, 0.8589743375778198, 0.8333333134651184, 0.9487179517745972, 0.942307710647583, 0.9487179517745972, 0.8846153616905212, 0.942307710647583, 0.8910256624221802, 0.942307710647583, 1.0, 1.0], [0.09615384787321091, 0.0833333358168602, 0.09615384787321091, 0.10256410390138626, 0.10897435992956161, 0.1538461595773697, 0.17307692766189575, 0.14102564752101898, 0.19230769574642181, 0.1666666716337204, 0.17307692766189575, 0.24358974397182465, 0.21153846383094788, 0.28205129504203796, 0.38461539149284363, 0.3461538553237915, 0.2756410241127014, 0.3717948794364929, 0.39743590354919434, 0.36538460850715637, 0.3910256326198578, 0.38461539149284363, 0.45512819290161133, 0.4615384638309479, 0.45512819290161133, 0.45512819290161133, 0.48076921701431274, 0.4967948794364929, 0.44871795177459717, NaN, 0.5, 0.5320512652397156, 0.5192307829856873, 0.5064102411270142, 0.6025640964508057, 0.5448718070983887, 0.6666666865348816, 0.5256410241127014, 0.5897436141967773, 0.6025640964508057, 0.6538461446762085, 0.7564102411270142, 0.6025640964508057, 0.6794871687889099, 0.6346153616905212, 0.7628205418586731, 0.6858974099159241, 0.6538461446762085, 0.8589743375778198, 0.75, 0.7435897588729858, 0.8269230723381042, 0.7756410241127014, 0.9358974099159241, 0.9487179517745972, 0.8269230723381042, 0.9230769276618958, 0.7820512652397156, 0.8717948794364929, 0.9679487347602844, 0.8910256624221802, 0.8653846383094788, 0.9358974099159241, 0.9487179517745972, 0.8974359035491943, 1.0, 0.9551281929016113, 0.9615384340286255, 0.9294871687889099, 0.942307710647583, 0.9551281929016113, 0.942307710647583, 1.0, 1.0, 0.9935897588729858, 0.9551281929016113, 1.0, 1.0], [0.07692307978868484, 0.11538461595773697, 0.10897435992956161, 0.0833333358168602, 0.07051282376050949, 0.14102564752101898, 0.23076923191547394, 0.12179487198591232, 0.2371794879436493, 0.2371794879436493, 0.24358974397182465, 0.2756410241127014, 0.19230769574642181, 0.32692307233810425, 0.19230769574642181, 0.2884615361690521, 0.31410256028175354, 0.41025641560554504, 0.4615384638309479, 0.43589743971824646, 0.4743589758872986, 0.3910256326198578, 0.4038461446762085, 0.49358972907066345, 0.49358972907066345, 0.5641025900840759, 0.48076921701431274, 0.48076921701431274, 0.43589743971824646, 0.5, NaN, 0.5897436141967773, 0.5064102411270142, 0.6410256624221802, 0.5192307829856873, 0.5384615659713745, 0.6089743375778198, 0.6346153616905212, 0.557692289352417, 0.692307710647583, 0.5512820482254028, 0.6602563858032227, 0.7179487347602844, 0.6987179517745972, 0.6410256624221802, 0.6666666865348816, 0.7179487347602844, 0.6794871687889099, 0.7820512652397156, 0.807692289352417, 0.6987179517745972, 0.7628205418586731, 0.7884615659713745, 0.8782051205635071, 0.8846153616905212, 0.8269230723381042, 0.8397436141967773, 0.7884615659713745, 0.8653846383094788, 0.9358974099159241, 0.9230769276618958, 0.8589743375778198, 0.8717948794364929, 0.9935897588729858, 0.8269230723381042, 0.9935897588729858, 0.8717948794364929, 0.8653846383094788, 0.8910256624221802, 0.8974359035491943, 0.9166666865348816, 0.9230769276618958, 0.9935897588729858, 1.0, 0.9743589758872986, 0.9038461446762085, 0.9935897588729858, 1.0], [0.07692307978868484, 0.10256410390138626, 0.11538461595773697, 0.08974359184503555, 0.19871795177459717, 0.1538461595773697, 0.17307692766189575, 0.20512820780277252, 0.1538461595773697, 0.16025641560554504, 0.1666666716337204, 0.22435897588729858, 0.23076923191547394, 0.2756410241127014, 0.3717948794364929, 0.3461538553237915, 0.3205128312110901, 0.3589743673801422, 0.39743590354919434, 0.30128204822540283, 0.33974358439445496, 0.4166666567325592, 0.3782051205635071, 0.4423076808452606, 0.4294871687889099, 0.3910256326198578, 0.43589743971824646, 0.44871795177459717, 0.4743589758872986, 0.46794870495796204, 0.41025641560554504, NaN, 0.5064102411270142, 0.49358972907066345, 0.5128205418586731, 0.46794870495796204, 0.48076921701431274, 0.5641025900840759, 0.5384615659713745, 0.5961538553237915, 0.557692289352417, 0.6217948794364929, 0.6025640964508057, 0.692307710647583, 0.6858974099159241, 0.6089743375778198, 0.6602563858032227, 0.5897436141967773, 0.7051281929016113, 0.6346153616905212, 0.6538461446762085, 0.7115384340286255, 0.6794871687889099, 0.8141025900840759, 0.8333333134651184, 0.7243589758872986, 0.6666666865348816, 0.9615384340286255, 0.8525640964508057, 0.8846153616905212, 0.9102563858032227, 0.9166666865348816, 0.9743589758872986, 0.9935897588729858, 0.9615384340286255, 0.8461538553237915, 0.9679487347602844, 0.9679487347602844, 0.9679487347602844, 0.9166666865348816, 0.8974359035491943, 0.8974359035491943, 0.9679487347602844, 0.9935897588729858, 0.9871794581413269, 0.9871794581413269, 1.0, 1.0], [0.0833333358168602, 0.09615384787321091, 0.0833333358168602, 0.1666666716337204, 0.1538461595773697, 0.24358974397182465, 0.14102564752101898, 0.25, 0.2371794879436493, 0.12820513546466827, 0.13461539149284363, 0.14743590354919434, 0.24358974397182465, 0.22435897588729858, 0.42307692766189575, 0.3461538553237915, 0.3333333432674408, 0.22435897588729858, 0.3782051205635071, 0.2628205120563507, 0.4294871687889099, 0.39743590354919434, 0.3076923191547394, 0.4166666567325592, 0.42307692766189575, 0.46794870495796204, 0.42307692766189575, 0.4615384638309479, 0.4615384638309479, 0.48076921701431274, 0.49358972907066345, 0.49358972907066345, NaN, 0.4743589758872986, 0.48076921701431274, 0.4166666567325592, 0.6602563858032227, 0.5320512652397156, 0.5384615659713745, 0.557692289352417, 0.6730769276618958, 0.6346153616905212, 0.5384615659713745, 0.5833333134651184, 0.5384615659713745, 0.7628205418586731, 0.5833333134651184, 0.7307692170143127, 0.6987179517745972, 0.6730769276618958, 0.7371794581413269, 0.7756410241127014, 0.7884615659713745, 0.7051281929016113, 0.7307692170143127, 0.8653846383094788, 0.6858974099159241, 0.7307692170143127, 0.8846153616905212, 0.8461538553237915, 0.7756410241127014, 0.807692289352417, 0.8653846383094788, 0.8269230723381042, 0.942307710647583, 0.9358974099159241, 0.9358974099159241, 0.9358974099159241, 0.942307710647583, 0.9358974099159241, 0.9615384340286255, 0.942307710647583, 0.9935897588729858, 1.0, 0.9871794581413269, 0.9743589758872986, 1.0, 1.0], [0.07051282376050949, 0.12820513546466827, 0.14743590354919434, 0.09615384787321091, 0.14743590354919434, 0.08974359184503555, 0.18589743971824646, 0.17307692766189575, 0.10897435992956161, 0.25641027092933655, 0.25641027092933655, 0.2628205120563507, 0.26923078298568726, 0.3205128312110901, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.45512819290161133, 0.3461538553237915, 0.3333333432674408, 0.25641027092933655, 0.36538460850715637, 0.4615384638309479, 0.36538460850715637, 0.45512819290161133, 0.3333333432674408, 0.4743589758872986, 0.4743589758872986, 0.30128204822540283, 0.49358972907066345, 0.3589743673801422, 0.5064102411270142, 0.5256410241127014, NaN, 0.5641025900840759, 0.5064102411270142, 0.5, 0.6762820482254028, 0.5448718070983887, 0.5512820482254028, 0.4743589758872986, 0.5769230723381042, 0.7820512652397156, 0.5833333134651184, 0.6858974099159241, 0.5705128312110901, 0.8012820482254028, 0.557692289352417, 0.6602563858032227, 0.6346153616905212, 0.6153846383094788, 0.6602563858032227, 0.6346153616905212, 0.7628205418586731, 0.7692307829856873, 0.6858974099159241, 0.8012820482254028, 0.8205128312110901, 0.7884615659713745, 0.8782051205635071, 0.9679487347602844, 0.9551281929016113, 0.9487179517745972, 1.0, 0.7756410241127014, 0.9102563858032227, 0.9166666865348816, 0.9166666865348816, 0.7756410241127014, 0.9294871687889099, 0.9551281929016113, 0.942307710647583, 0.9038461446762085, 0.9807692170143127, 0.9358974099159241, 1.0, 0.9487179517745972, 1.0], [0.14102564752101898, 0.13461539149284363, 0.14743590354919434, 0.18589743971824646, 0.19871795177459717, 0.28205129504203796, 0.17307692766189575, 0.25, 0.2756410241127014, 0.17307692766189575, 0.12179487198591232, 0.18589743971824646, 0.26923078298568726, 0.26923078298568726, 0.4166666567325592, 0.3461538553237915, 0.31410256028175354, 0.11538461595773697, 0.3333333432674408, 0.28205129504203796, 0.4294871687889099, 0.3589743673801422, 0.24358974397182465, 0.3461538553237915, 0.42307692766189575, 0.4294871687889099, 0.43589743971824646, 0.38461539149284363, 0.43589743971824646, 0.39743590354919434, 0.48076921701431274, 0.4871794879436493, 0.5192307829856873, 0.43589743971824646, NaN, 0.43589743971824646, 0.6089743375778198, 0.49358972907066345, 0.5192307829856873, 0.5320512652397156, 0.6089743375778198, 0.5833333134651184, 0.49358972907066345, 0.5512820482254028, 0.46794870495796204, 0.6794871687889099, 0.5448718070983887, 0.75, 0.6666666865348816, 0.6217948794364929, 0.7307692170143127, 0.7179487347602844, 0.8141025900840759, 0.692307710647583, 0.692307710647583, 0.8525640964508057, 0.6089743375778198, 0.7051281929016113, 0.8910256624221802, 0.7692307829856873, 0.7628205418586731, 0.7756410241127014, 0.8397436141967773, 0.8141025900840759, 0.9166666865348816, 0.9294871687889099, 0.9294871687889099, 0.9358974099159241, 0.9358974099159241, 0.942307710647583, 0.9358974099159241, 0.9294871687889099, 0.9807692170143127, 1.0, 0.9807692170143127, 0.9615384340286255, 1.0, 1.0], [0.23076923191547394, 0.2371794879436493, 0.24358974397182465, 0.35256409645080566, 0.32692307233810425, 0.4038461446762085, 0.26923078298568726, 0.4615384638309479, 0.3910256326198578, 0.39743590354919434, 0.3717948794364929, 0.39743590354919434, 0.5, 0.46794870495796204, 0.4743589758872986, 0.48076921701431274, 0.4615384638309479, 0.3589743673801422, 0.4615384638309479, 0.5064102411270142, 0.4423076808452606, 0.5064102411270142, 0.4423076808452606, 0.48076921701431274, 0.4166666567325592, 0.43589743971824646, 0.4294871687889099, 0.4423076808452606, 0.557692289352417, 0.45512819290161133, 0.4615384638309479, 0.5320512652397156, 0.5833333134651184, 0.49358972907066345, 0.5641025900840759, NaN, 0.5448718070983887, 0.5, 0.5256410241127014, 0.5, 0.6474359035491943, 0.5128205418586731, 0.49358972907066345, 0.5961538553237915, 0.5256410241127014, 0.5705128312110901, 0.5192307829856873, 0.6730769276618958, 0.5064102411270142, 0.5448718070983887, 0.6602563858032227, 0.5961538553237915, 0.7371794581413269, 0.5, 0.48076921701431274, 0.7564102411270142, 0.6153846383094788, 0.6089743375778198, 0.7307692170143127, 0.6025640964508057, 0.6538461446762085, 0.5897436141967773, 0.6730769276618958, 0.6538461446762085, 0.8141025900840759, 0.6794871687889099, 0.75, 0.7371794581413269, 0.8141025900840759, 0.7692307829856873, 0.7628205418586731, 0.7628205418586731, 0.8012820482254028, 0.7628205418586731, 0.9487179517745972, 0.7756410241127014, 0.8205128312110901, 1.0], [0.08974359184503555, 0.044871795922517776, 0.06410256773233414, 0.07692307978868484, 0.12820513546466827, 0.16025641560554504, 0.14743590354919434, 0.17307692766189575, 0.1666666716337204, 0.17307692766189575, 0.13461539149284363, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.35256409645080566, 0.3076923191547394, 0.21153846383094788, 0.18589743971824646, 0.29487180709838867, 0.25641027092933655, 0.3461538553237915, 0.41025641560554504, 0.24358974397182465, 0.3076923191547394, 0.3076923191547394, 0.36538460850715637, 0.28205129504203796, 0.3076923191547394, 0.48076921701431274, 0.3333333432674408, 0.3910256326198578, 0.5192307829856873, 0.33974358439445496, 0.5, 0.3910256326198578, 0.45512819290161133, NaN, 0.5128205418586731, 0.557692289352417, 0.5384615659713745, 0.6410256624221802, 0.5064102411270142, 0.5384615659713745, 0.6217948794364929, 0.5448718070983887, 0.7532051205635071, 0.557692289352417, 0.5833333134651184, 0.6217948794364929, 0.6410256624221802, 0.7051281929016113, 0.8525640964508057, 0.692307710647583, 0.7115384340286255, 0.7243589758872986, 0.75, 0.6153846383094788, 0.75, 0.9230769276618958, 0.8397436141967773, 0.8012820482254028, 0.7628205418586731, 0.8782051205635071, 0.8141025900840759, 0.942307710647583, 0.9487179517745972, 0.9358974099159241, 0.942307710647583, 0.9743589758872986, 0.9230769276618958, 0.9358974099159241, 0.9358974099159241, 0.9807692170143127, 0.9935897588729858, 0.9871794581413269, 0.942307710647583, 0.9935897588729858, 1.0], [0.08974359184503555, 0.11538461595773697, 0.12179487198591232, 0.08974359184503555, 0.1666666716337204, 0.11538461595773697, 0.22435897588729858, 0.12820513546466827, 0.08974359184503555, 0.2371794879436493, 0.18589743971824646, 0.21153846383094788, 0.23076923191547394, 0.3076923191547394, 0.24358974397182465, 0.25, 0.32692307233810425, 0.39743590354919434, 0.33974358439445496, 0.32692307233810425, 0.24358974397182465, 0.33974358439445496, 0.3910256326198578, 0.3717948794364929, 0.4423076808452606, 0.32692307233810425, 0.43589743971824646, 0.4294871687889099, 0.29487180709838867, 0.4743589758872986, 0.36538460850715637, 0.43589743971824646, 0.46794870495796204, 0.32371795177459717, 0.5064102411270142, 0.5, 0.4871794879436493, NaN, 0.5128205418586731, 0.4615384638309479, 0.45512819290161133, 0.5448718070983887, 0.5769230723381042, 0.5512820482254028, 0.6217948794364929, 0.5512820482254028, 0.8044871687889099, 0.5064102411270142, 0.6538461446762085, 0.557692289352417, 0.557692289352417, 0.6666666865348816, 0.5448718070983887, 0.7820512652397156, 0.7756410241127014, 0.6153846383094788, 0.7435897588729858, 0.7628205418586731, 0.7435897588729858, 0.8589743375778198, 0.9871794581413269, 0.9102563858032227, 0.9038461446762085, 0.9551281929016113, 0.7243589758872986, 0.8525640964508057, 0.9294871687889099, 0.9294871687889099, 0.7307692170143127, 0.9166666865348816, 0.9358974099159241, 0.9358974099159241, 0.8910256624221802, 0.9487179517745972, 0.942307710647583, 0.9935897588729858, 0.9230769276618958, 1.0], [0.08974359184503555, 0.19230769574642181, 0.21153846383094788, 0.1538461595773697, 0.21153846383094788, 0.17307692766189575, 0.28205129504203796, 0.2371794879436493, 0.3717948794364929, 0.3461538553237915, 0.42307692766189575, 0.4166666567325592, 0.3461538553237915, 0.36538460850715637, 0.19871795177459717, 0.2371794879436493, 0.3782051205635071, 0.38461539149284363, 0.4038461446762085, 0.46794870495796204, 0.42307692766189575, 0.26923078298568726, 0.39743590354919434, 0.4166666567325592, 0.45512819290161133, 0.4294871687889099, 0.4615384638309479, 0.4166666567325592, 0.3717948794364929, 0.41025641560554504, 0.4423076808452606, 0.4615384638309479, 0.4615384638309479, 0.45512819290161133, 0.48076921701431274, 0.4743589758872986, 0.4423076808452606, 0.4871794879436493, NaN, 0.49358972907066345, 0.4294871687889099, 0.4294871687889099, 0.5384615659713745, 0.5256410241127014, 0.5641025900840759, 0.46794870495796204, 0.5897436141967773, 0.5833333134651184, 0.5256410241127014, 0.5512820482254028, 0.49358972907066345, 0.4743589758872986, 0.6346153616905212, 0.6217948794364929, 0.5961538553237915, 0.5961538553237915, 0.75, 0.7115384340286255, 0.6538461446762085, 0.7756410241127014, 0.8012820482254028, 0.8589743375778198, 0.8205128312110901, 0.9230769276618958, 0.6025640964508057, 0.8653846383094788, 0.7756410241127014, 0.7628205418586731, 0.6666666865348816, 0.8589743375778198, 0.8782051205635071, 0.8910256624221802, 0.9102563858032227, 1.0, 0.8974359035491943, 0.8846153616905212, 0.9358974099159241, 1.0], [0.06410256773233414, 0.044871795922517776, 0.044871795922517776, 0.057692307978868484, 0.05128205195069313, 0.0833333358168602, 0.09615384787321091, 0.11538461595773697, 0.16025641560554504, 0.1538461595773697, 0.14743590354919434, 0.1666666716337204, 0.17307692766189575, 0.21794871985912323, 0.18589743971824646, 0.25641027092933655, 0.24358974397182465, 0.30128204822540283, 0.3076923191547394, 0.3461538553237915, 0.3076923191547394, 0.3589743673801422, 0.3589743673801422, 0.3333333432674408, 0.3589743673801422, 0.3782051205635071, 0.35256409645080566, 0.38461539149284363, 0.3333333432674408, 0.39743590354919434, 0.3076923191547394, 0.4038461446762085, 0.4423076808452606, 0.44871795177459717, 0.46794870495796204, 0.5, 0.4615384638309479, 0.5384615659713745, 0.5064102411270142, NaN, 0.44871795177459717, 0.5128205418586731, 0.6153846383094788, 0.557692289352417, 0.5256410241127014, 0.557692289352417, 0.6730769276618958, 0.6474359035491943, 0.6346153616905212, 0.6666666865348816, 0.6282051205635071, 0.6346153616905212, 0.692307710647583, 0.7307692170143127, 0.692307710647583, 0.7435897588729858, 0.7243589758872986, 0.7435897588729858, 0.8589743375778198, 0.8397436141967773, 0.8717948794364929, 0.7756410241127014, 0.8717948794364929, 0.9294871687889099, 0.7820512652397156, 0.9807692170143127, 0.8525640964508057, 0.8525640964508057, 0.8397436141967773, 0.8974359035491943, 0.9166666865348816, 0.9102563858032227, 0.942307710647583, 0.9743589758872986, 0.9487179517745972, 0.8910256624221802, 0.9743589758872986, 1.0], [0.07051282376050949, 0.05128205195069313, 0.057692307978868484, 0.08974359184503555, 0.07692307978868484, 0.11538461595773697, 0.09615384787321091, 0.1666666716337204, 0.28205129504203796, 0.17307692766189575, 0.24358974397182465, 0.19871795177459717, 0.21153846383094788, 0.21794871985912323, 0.28205129504203796, 0.32692307233810425, 0.24358974397182465, 0.2628205120563507, 0.33974358439445496, 0.3333333432674408, 0.38461539149284363, 0.4038461446762085, 0.2884615361690521, 0.32692307233810425, 0.3333333432674408, 0.4038461446762085, 0.3333333432674408, 0.3717948794364929, 0.49358972907066345, 0.3461538553237915, 0.44871795177459717, 0.4423076808452606, 0.32692307233810425, 0.5256410241127014, 0.3910256326198578, 0.35256409645080566, 0.3589743673801422, 0.5448718070983887, 0.5705128312110901, 0.5512820482254028, NaN, 0.4871794879436493, 0.5833333134651184, 0.5448718070983887, 0.49358972907066345, 0.45512819290161133, 0.557692289352417, 0.7115384340286255, 0.5833333134651184, 0.6538461446762085, 0.7916666865348816, 0.5384615659713745, 0.807692289352417, 0.6153846383094788, 0.6153846383094788, 0.7115384340286255, 0.6602563858032227, 0.6730769276618958, 0.8974359035491943, 0.6602563858032227, 0.7564102411270142, 0.7435897588729858, 0.7756410241127014, 0.8012820482254028, 0.8782051205635071, 0.8653846383094788, 0.9038461446762085, 0.9102563858032227, 0.9230769276618958, 0.8461538553237915, 0.8846153616905212, 0.8782051205635071, 0.9615384340286255, 0.9935897588729858, 0.9743589758872986, 0.942307710647583, 0.9935897588729858, 1.0], [0.06410256773233414, 0.03205128386616707, 0.057692307978868484, 0.05128205195069313, 0.044871795922517776, 0.10897435992956161, 0.12179487198591232, 0.10897435992956161, 0.11538461595773697, 0.12179487198591232, 0.12179487198591232, 0.16025641560554504, 0.13461539149284363, 0.22435897588729858, 0.24358974397182465, 0.25641027092933655, 0.17307692766189575, 0.21794871985912323, 0.1666666716337204, 0.24358974397182465, 0.30128204822540283, 0.33974358439445496, 0.28205129504203796, 0.21794871985912323, 0.2371794879436493, 0.33974358439445496, 0.21794871985912323, 0.25641027092933655, 0.3782051205635071, 0.24358974397182465, 0.33974358439445496, 0.3782051205635071, 0.36538460850715637, 0.42307692766189575, 0.4166666567325592, 0.4871794879436493, 0.49358972907066345, 0.45512819290161133, 0.5705128312110901, 0.4871794879436493, 0.5128205418586731, NaN, 0.5192307829856873, 0.5833333134651184, 0.5192307829856873, 0.6282051205635071, 0.5769230723381042, 0.5897436141967773, 0.6346153616905212, 0.6282051205635071, 0.6538461446762085, 0.692307710647583, 0.6858974099159241, 0.7884615659713745, 0.7948718070983887, 0.7243589758872986, 0.7435897588729858, 0.7179487347602844, 0.8653846383094788, 0.942307710647583, 0.8525640964508057, 0.807692289352417, 0.8974359035491943, 0.8910256624221802, 0.8461538553237915, 1.0, 0.9230769276618958, 0.9166666865348816, 0.8717948794364929, 0.9230769276618958, 0.9487179517745972, 0.9230769276618958, 0.9679487347602844, 1.0, 0.9935897588729858, 0.9487179517745972, 1.0, 1.0], [0.07692307978868484, 0.10256410390138626, 0.12179487198591232, 0.09615384787321091, 0.1538461595773697, 0.0833333358168602, 0.14743590354919434, 0.1538461595773697, 0.08974359184503555, 0.21153846383094788, 0.19230769574642181, 0.22435897588729858, 0.2371794879436493, 0.2756410241127014, 0.21794871985912323, 0.21153846383094788, 0.30128204822540283, 0.39743590354919434, 0.2628205120563507, 0.3076923191547394, 0.18589743971824646, 0.3205128312110901, 0.4166666567325592, 0.29487180709838867, 0.3589743673801422, 0.26923078298568726, 0.36538460850715637, 0.41025641560554504, 0.2371794879436493, 0.39743590354919434, 0.28205129504203796, 0.39743590354919434, 0.4615384638309479, 0.21794871985912323, 0.5064102411270142, 0.5064102411270142, 0.4615384638309479, 0.42307692766189575, 0.4615384638309479, 0.38461539149284363, 0.4166666567325592, 0.48076921701431274, NaN, 0.49358972907066345, 0.5897436141967773, 0.5320512652397156, 0.5993589758872986, 0.5192307829856873, 0.5192307829856873, 0.5064102411270142, 0.5128205418586731, 0.5705128312110901, 0.5641025900840759, 0.6410256624221802, 0.6474359035491943, 0.5897436141967773, 0.75, 0.7307692170143127, 0.7179487347602844, 0.807692289352417, 0.8974359035491943, 0.9358974099159241, 0.9166666865348816, 0.8910256624221802, 0.7243589758872986, 0.8012820482254028, 0.9166666865348816, 0.9102563858032227, 0.7051281929016113, 0.9166666865348816, 0.9166666865348816, 0.9230769276618958, 0.8910256624221802, 0.9679487347602844, 0.9358974099159241, 1.0, 0.9294871687889099, 1.0], [0.07051282376050949, 0.07692307978868484, 0.07692307978868484, 0.057692307978868484, 0.14102564752101898, 0.08974359184503555, 0.12179487198591232, 0.17307692766189575, 0.09615384787321091, 0.1538461595773697, 0.14102564752101898, 0.16025641560554504, 0.1794871836900711, 0.1794871836900711, 0.3076923191547394, 0.24358974397182465, 0.25641027092933655, 0.28205129504203796, 0.22435897588729858, 0.20512820780277252, 0.19230769574642181, 0.3076923191547394, 0.2884615361690521, 0.2371794879436493, 0.30128204822540283, 0.2628205120563507, 0.30128204822540283, 0.29487180709838867, 0.3461538553237915, 0.3205128312110901, 0.30128204822540283, 0.3076923191547394, 0.4166666567325592, 0.4166666567325592, 0.44871795177459717, 0.4038461446762085, 0.3782051205635071, 0.44871795177459717, 0.4743589758872986, 0.4423076808452606, 0.45512819290161133, 0.4166666567325592, 0.5064102411270142, NaN, 0.5897436141967773, 0.48076921701431274, 0.5512820482254028, 0.5256410241127014, 0.5320512652397156, 0.5, 0.557692289352417, 0.5833333134651184, 0.5641025900840759, 0.6987179517745972, 0.6730769276618958, 0.5833333134651184, 0.5961538553237915, 0.9230769276618958, 0.7820512652397156, 0.8333333134651184, 0.8653846383094788, 0.8782051205635071, 0.942307710647583, 0.9487179517745972, 0.8782051205635071, 0.807692289352417, 0.9294871687889099, 0.9358974099159241, 0.8846153616905212, 0.8974359035491943, 0.9230769276618958, 0.9102563858032227, 0.9294871687889099, 0.9935897588729858, 0.9807692170143127, 0.9743589758872986, 0.9935897588729858, 1.0], [0.07692307978868484, 0.025641025975346565, 0.025641025975346565, 0.012820512987673283, 0.044871795922517776, 0.05128205195069313, 0.06410256773233414, 0.07692307978868484, 0.1794871836900711, 0.1538461595773697, 0.25641027092933655, 0.25641027092933655, 0.17307692766189575, 0.20512820780277252, 0.1666666716337204, 0.1666666716337204, 0.2756410241127014, 0.35256409645080566, 0.25641027092933655, 0.2628205120563507, 0.32692307233810425, 0.22435897588729858, 0.3717948794364929, 0.30128204822540283, 0.4038461446762085, 0.30128204822540283, 0.3910256326198578, 0.3589743673801422, 0.18589743971824646, 0.36538460850715637, 0.3589743673801422, 0.31410256028175354, 0.4615384638309479, 0.31410256028175354, 0.5320512652397156, 0.4743589758872986, 0.45512819290161133, 0.3782051205635071, 0.43589743971824646, 0.4743589758872986, 0.5064102411270142, 0.48076921701431274, 0.41025641560554504, 0.41025641560554504, NaN, 0.5128205418586731, 0.5320512652397156, 0.6602563858032227, 0.5128205418586731, 0.5512820482254028, 0.6153846383094788, 0.5448718070983887, 0.6794871687889099, 0.5769230723381042, 0.5705128312110901, 0.6794871687889099, 0.7628205418586731, 0.6217948794364929, 0.6987179517745972, 0.7628205418586731, 0.8653846383094788, 0.8525640964508057, 0.9294871687889099, 0.7756410241127014, 0.6794871687889099, 0.807692289352417, 0.8782051205635071, 0.8653846383094788, 0.7115384340286255, 0.9358974099159241, 0.9487179517745972, 0.9487179517745972, 0.8782051205635071, 0.9551281929016113, 0.9358974099159241, 0.9551281929016113, 0.9679487347602844, 1.0], [0.07051282376050949, 0.03846153989434242, 0.057692307978868484, 0.05128205195069313, 0.08974359184503555, 0.14102564752101898, 0.14102564752101898, 0.14102564752101898, 0.11538461595773697, 0.16025641560554504, 0.12820513546466827, 0.11538461595773697, 0.19230769574642181, 0.21794871985912323, 0.28205129504203796, 0.2756410241127014, 0.18589743971824646, 0.16025641560554504, 0.21153846383094788, 0.23076923191547394, 0.28205129504203796, 0.3461538553237915, 0.19230769574642181, 0.2628205120563507, 0.2628205120563507, 0.30128204822540283, 0.2371794879436493, 0.2371794879436493, 0.4166666567325592, 0.2371794879436493, 0.3333333432674408, 0.3910256326198578, 0.2371794879436493, 0.4294871687889099, 0.3205128312110901, 0.4294871687889099, 0.24679486453533173, 0.44871795177459717, 0.5320512652397156, 0.4423076808452606, 0.5448718070983887, 0.3717948794364929, 0.46794870495796204, 0.5192307829856873, 0.4871794879436493, NaN, 0.5192307829856873, 0.5384615659713745, 0.5384615659713745, 0.5448718070983887, 0.6538461446762085, 0.7179487347602844, 0.6474359035491943, 0.6538461446762085, 0.6794871687889099, 0.6666666865348816, 0.5256410241127014, 0.7051281929016113, 0.8846153616905212, 0.7948718070983887, 0.7692307829856873, 0.7115384340286255, 0.8269230723381042, 0.7884615659713745, 0.8974359035491943, 0.9102563858032227, 0.9230769276618958, 0.9294871687889099, 0.9294871687889099, 0.9038461446762085, 0.9166666865348816, 0.9102563858032227, 0.9679487347602844, 0.9935897588729858, 0.9807692170143127, 0.942307710647583, 0.9935897588729858, 1.0], [0.08974359184503555, 0.08974359184503555, 0.08974359184503555, 0.08974359184503555, 0.12820513546466827, 0.09615384787321091, 0.14102564752101898, 0.12179487198591232, 0.09615384787321091, 0.20512820780277252, 0.14102564752101898, 0.19230769574642181, 0.20512820780277252, 0.22435897588729858, 0.22435897588729858, 0.21153846383094788, 0.23076923191547394, 0.2884615361690521, 0.21794871985912323, 0.24358974397182465, 0.23076923191547394, 0.29487180709838867, 0.3205128312110901, 0.24358974397182465, 0.30128204822540283, 0.2756410241127014, 0.32692307233810425, 0.29487180709838867, 0.2628205120563507, 0.31410256028175354, 0.28205129504203796, 0.33974358439445496, 0.4166666567325592, 0.19871795177459717, 0.45512819290161133, 0.48076921701431274, 0.4423076808452606, 0.1955128163099289, 0.41025641560554504, 0.32692307233810425, 0.4423076808452606, 0.42307692766189575, 0.4006410241127014, 0.44871795177459717, 0.46794870495796204, 0.48076921701431274, NaN, 0.49358972907066345, 0.5, 0.4038461446762085, 0.49358972907066345, 0.6089743375778198, 0.5384615659713745, 0.6217948794364929, 0.6282051205635071, 0.5448718070983887, 0.6666666865348816, 0.6410256624221802, 0.6794871687889099, 0.807692289352417, 0.8301281929016113, 0.8525640964508057, 0.8653846383094788, 0.8012820482254028, 0.692307710647583, 0.807692289352417, 0.8846153616905212, 0.8782051205635071, 0.6858974099159241, 0.8974359035491943, 0.9230769276618958, 0.9166666865348816, 0.8782051205635071, 0.9294871687889099, 0.9230769276618958, 0.9743589758872986, 0.9038461446762085, 1.0], [0.09615384787321091, 0.07051282376050949, 0.06410256773233414, 0.12179487198591232, 0.05128205195069313, 0.16025641560554504, 0.07692307978868484, 0.17307692766189575, 0.21794871985912323, 0.11538461595773697, 0.08974359184503555, 0.12179487198591232, 0.1666666716337204, 0.1794871836900711, 0.21153846383094788, 0.21794871985912323, 0.1666666716337204, 0.17307692766189575, 0.29487180709838867, 0.2884615361690521, 0.31410256028175354, 0.2884615361690521, 0.20512820780277252, 0.3076923191547394, 0.31410256028175354, 0.3589743673801422, 0.3076923191547394, 0.3333333432674408, 0.36538460850715637, 0.3461538553237915, 0.3205128312110901, 0.41025641560554504, 0.26923078298568726, 0.4423076808452606, 0.25, 0.32692307233810425, 0.4166666567325592, 0.49358972907066345, 0.4166666567325592, 0.35256409645080566, 0.2884615361690521, 0.41025641560554504, 0.48076921701431274, 0.4743589758872986, 0.33974358439445496, 0.4615384638309479, 0.5064102411270142, NaN, 0.5256410241127014, 0.5641025900840759, 0.4743589758872986, 0.48076921701431274, 0.7179487347602844, 0.6474359035491943, 0.6346153616905212, 0.557692289352417, 0.5512820482254028, 0.5833333134651184, 0.6346153616905212, 0.6858974099159241, 0.6089743375778198, 0.6538461446762085, 0.5961538553237915, 0.7948718070983887, 0.8205128312110901, 0.7435897588729858, 0.7243589758872986, 0.7243589758872986, 0.8653846383094788, 0.6858974099159241, 0.7371794581413269, 0.7371794581413269, 0.9935897588729858, 0.9230769276618958, 0.9358974099159241, 0.8974359035491943, 1.0, 0.9935897588729858], [0.06410256773233414, 0.01923076994717121, 0.012820512987673283, 0.006410256493836641, 0.01923076994717121, 0.01923076994717121, 0.08974359184503555, 0.07051282376050949, 0.05128205195069313, 0.13461539149284363, 0.06410256773233414, 0.09615384787321091, 0.12179487198591232, 0.19230769574642181, 0.14743590354919434, 0.1538461595773697, 0.1538461595773697, 0.1666666716337204, 0.07692307978868484, 0.21794871985912323, 0.1794871836900711, 0.24358974397182465, 0.20512820780277252, 0.13461539149284363, 0.09615384787321091, 0.24358974397182465, 0.07051282376050949, 0.14743590354919434, 0.28205129504203796, 0.14102564752101898, 0.21794871985912323, 0.29487180709838867, 0.30128204822540283, 0.33974358439445496, 0.3333333432674408, 0.49358972907066345, 0.3782051205635071, 0.3461538553237915, 0.4743589758872986, 0.36538460850715637, 0.4166666567325592, 0.36538460850715637, 0.48076921701431274, 0.46794870495796204, 0.4871794879436493, 0.4615384638309479, 0.5, 0.4743589758872986, NaN, 0.5256410241127014, 0.5320512652397156, 0.6346153616905212, 0.5512820482254028, 0.6666666865348816, 0.6987179517745972, 0.6025640964508057, 0.6987179517745972, 0.692307710647583, 0.7692307829856873, 0.8717948794364929, 0.8717948794364929, 0.8141025900840759, 0.8910256624221802, 0.8717948794364929, 0.7371794581413269, 0.9615384340286255, 0.8846153616905212, 0.8717948794364929, 0.7756410241127014, 0.8974359035491943, 0.9294871687889099, 0.9230769276618958, 0.9487179517745972, 0.9871794581413269, 0.9871794581413269, 0.9294871687889099, 1.0, 1.0], [0.01923076994717121, 0.05128205195069313, 0.06410256773233414, 0.03205128386616707, 0.06410256773233414, 0.03205128386616707, 0.08974359184503555, 0.044871795922517776, 0.057692307978868484, 0.18589743971824646, 0.10897435992956161, 0.13461539149284363, 0.17307692766189575, 0.24358974397182465, 0.16025641560554504, 0.14102564752101898, 0.17307692766189575, 0.21794871985912323, 0.1666666716337204, 0.26923078298568726, 0.08974359184503555, 0.2756410241127014, 0.25641027092933655, 0.19871795177459717, 0.20512820780277252, 0.1794871836900711, 0.20512820780277252, 0.23076923191547394, 0.28205129504203796, 0.25, 0.19230769574642181, 0.36538460850715637, 0.32692307233810425, 0.36538460850715637, 0.3782051205635071, 0.45512819290161133, 0.3589743673801422, 0.4423076808452606, 0.44871795177459717, 0.3333333432674408, 0.3461538553237915, 0.3717948794364929, 0.49358972907066345, 0.5, 0.44871795177459717, 0.45512819290161133, 0.5961538553237915, 0.43589743971824646, 0.4743589758872986, NaN, 0.45512819290161133, 0.5384615659713745, 0.5128205418586731, 0.6987179517745972, 0.7179487347602844, 0.5320512652397156, 0.6730769276618958, 0.6858974099159241, 0.7435897588729858, 0.8012820482254028, 0.8589743375778198, 0.8269230723381042, 0.8269230723381042, 0.8461538553237915, 0.7820512652397156, 0.7884615659713745, 0.7948718070983887, 0.7948718070983887, 0.8205128312110901, 0.8846153616905212, 0.9102563858032227, 0.9038461446762085, 0.8653846383094788, 0.9294871687889099, 0.8717948794364929, 0.9038461446762085, 0.9807692170143127, 1.0], [0.06410256773233414, 0.05128205195069313, 0.05128205195069313, 0.044871795922517776, 0.07692307978868484, 0.10256410390138626, 0.09615384787321091, 0.11538461595773697, 0.17307692766189575, 0.16025641560554504, 0.19230769574642181, 0.1538461595773697, 0.17307692766189575, 0.1794871836900711, 0.21794871985912323, 0.19871795177459717, 0.16025641560554504, 0.1794871836900711, 0.19871795177459717, 0.2756410241127014, 0.22435897588729858, 0.31410256028175354, 0.21153846383094788, 0.21794871985912323, 0.25641027092933655, 0.23076923191547394, 0.25, 0.25641027092933655, 0.3076923191547394, 0.25641027092933655, 0.30128204822540283, 0.3461538553237915, 0.2628205120563507, 0.38461539149284363, 0.26923078298568726, 0.33974358439445496, 0.29487180709838867, 0.4423076808452606, 0.5064102411270142, 0.3717948794364929, 0.2083333283662796, 0.3461538553237915, 0.4871794879436493, 0.4423076808452606, 0.38461539149284363, 0.3461538553237915, 0.5064102411270142, 0.5256410241127014, 0.46794870495796204, 0.5448718070983887, NaN, 0.3589743673801422, 0.6602563858032227, 0.5256410241127014, 0.5256410241127014, 0.5384615659713745, 0.5384615659713745, 0.6217948794364929, 0.8205128312110901, 0.6410256624221802, 0.7243589758872986, 0.7051281929016113, 0.6987179517745972, 0.75, 0.75, 0.8141025900840759, 0.8717948794364929, 0.8653846383094788, 0.8012820482254028, 0.7948718070983887, 0.8397436141967773, 0.8269230723381042, 0.9294871687889099, 0.9679487347602844, 0.9551281929016113, 0.942307710647583, 0.9871794581413269, 1.0], [0.07051282376050949, 0.05128205195069313, 0.06410256773233414, 0.057692307978868484, 0.07051282376050949, 0.07692307978868484, 0.10256410390138626, 0.08974359184503555, 0.09615384787321091, 0.13461539149284363, 0.12179487198591232, 0.13461539149284363, 0.14743590354919434, 0.1794871836900711, 0.21794871985912323, 0.25, 0.12820513546466827, 0.16025641560554504, 0.12179487198591232, 0.19871795177459717, 0.21794871985912323, 0.31410256028175354, 0.1794871836900711, 0.14743590354919434, 0.14743590354919434, 0.26923078298568726, 0.14743590354919434, 0.1794871836900711, 0.3205128312110901, 0.17307692766189575, 0.2371794879436493, 0.2884615361690521, 0.22435897588729858, 0.33974358439445496, 0.28205129504203796, 0.4038461446762085, 0.14743590354919434, 0.3333333432674408, 0.5256410241127014, 0.36538460850715637, 0.4615384638309479, 0.3076923191547394, 0.4294871687889099, 0.4166666567325592, 0.45512819290161133, 0.28205129504203796, 0.3910256326198578, 0.5192307829856873, 0.36538460850715637, 0.4615384638309479, 0.6410256624221802, NaN, 0.5961538553237915, 0.6025640964508057, 0.6153846383094788, 0.5512820482254028, 0.5064102411270142, 0.6410256624221802, 0.7820512652397156, 0.7628205418586731, 0.7371794581413269, 0.7692307829856873, 0.8012820482254028, 0.7692307829856873, 0.7948718070983887, 0.8653846383094788, 0.9102563858032227, 0.9166666865348816, 0.8653846383094788, 0.8846153616905212, 0.9038461446762085, 0.8782051205635071, 0.9679487347602844, 0.9807692170143127, 0.9743589758872986, 0.9743589758872986, 1.0, 1.0], [0.057692307978868484, 0.03205128386616707, 0.044871795922517776, 0.057692307978868484, 0.01923076994717121, 0.057692307978868484, 0.057692307978868484, 0.09615384787321091, 0.10897435992956161, 0.09615384787321091, 0.0833333358168602, 0.06410256773233414, 0.10897435992956161, 0.12820513546466827, 0.12820513546466827, 0.17307692766189575, 0.09615384787321091, 0.11538461595773697, 0.1666666716337204, 0.1794871836900711, 0.18589743971824646, 0.21153846383094788, 0.1538461595773697, 0.20512820780277252, 0.25, 0.25, 0.24358974397182465, 0.20512820780277252, 0.29487180709838867, 0.22435897588729858, 0.21153846383094788, 0.3205128312110901, 0.21153846383094788, 0.36538460850715637, 0.18589743971824646, 0.2628205120563507, 0.3076923191547394, 0.45512819290161133, 0.36538460850715637, 0.3076923191547394, 0.19230769574642181, 0.31410256028175354, 0.43589743971824646, 0.43589743971824646, 0.3205128312110901, 0.35256409645080566, 0.4615384638309479, 0.28205129504203796, 0.44871795177459717, 0.4871794879436493, 0.33974358439445496, 0.4038461446762085, NaN, 0.5897436141967773, 0.5641025900840759, 0.45512819290161133, 0.5, 0.5897436141967773, 0.6089743375778198, 0.6602563858032227, 0.5961538553237915, 0.6794871687889099, 0.5833333134651184, 0.7692307829856873, 0.7628205418586731, 0.7435897588729858, 0.692307710647583, 0.7051281929016113, 0.807692289352417, 0.6666666865348816, 0.7179487347602844, 0.7243589758872986, 0.9871794581413269, 0.8397436141967773, 0.9358974099159241, 0.8974359035491943, 1.0, 0.9935897588729858], [0.07051282376050949, 0.044871795922517776, 0.057692307978868484, 0.025641025975346565, 0.06410256773233414, 0.08974359184503555, 0.12179487198591232, 0.07692307978868484, 0.0833333358168602, 0.12179487198591232, 0.11538461595773697, 0.13461539149284363, 0.12820513546466827, 0.18589743971824646, 0.10897435992956161, 0.16025641560554504, 0.11538461595773697, 0.19230769574642181, 0.12179487198591232, 0.21153846383094788, 0.16025641560554504, 0.19230769574642181, 0.25, 0.14743590354919434, 0.10897435992956161, 0.21794871985912323, 0.11538461595773697, 0.05128205195069313, 0.23076923191547394, 0.06410256773233414, 0.12179487198591232, 0.18589743971824646, 0.29487180709838867, 0.2371794879436493, 0.3076923191547394, 0.5, 0.2884615361690521, 0.21794871985912323, 0.3782051205635071, 0.26923078298568726, 0.38461539149284363, 0.21153846383094788, 0.3589743673801422, 0.30128204822540283, 0.42307692766189575, 0.3461538553237915, 0.3782051205635071, 0.35256409645080566, 0.3333333432674408, 0.30128204822540283, 0.4743589758872986, 0.39743590354919434, 0.41025641560554504, NaN, 0.4711538553237915, 0.4871794879436493, 0.5512820482254028, 0.6602563858032227, 0.6217948794364929, 0.9294871687889099, 0.6410256624221802, 0.6538461446762085, 0.7564102411270142, 0.7051281929016113, 0.6987179517745972, 0.8333333134651184, 0.7948718070983887, 0.8012820482254028, 0.7179487347602844, 0.7307692170143127, 0.7884615659713745, 0.7756410241127014, 0.9487179517745972, 0.9358974099159241, 0.9166666865348816, 0.8653846383094788, 0.9743589758872986, 1.0], [0.07051282376050949, 0.03205128386616707, 0.044871795922517776, 0.012820512987673283, 0.05128205195069313, 0.07051282376050949, 0.11538461595773697, 0.07051282376050949, 0.057692307978868484, 0.12820513546466827, 0.10897435992956161, 0.11538461595773697, 0.13461539149284363, 0.18589743971824646, 0.10897435992956161, 0.13461539149284363, 0.10897435992956161, 0.1666666716337204, 0.0833333358168602, 0.20512820780277252, 0.1666666716337204, 0.17307692766189575, 0.23076923191547394, 0.14743590354919434, 0.10897435992956161, 0.21153846383094788, 0.10256410390138626, 0.012820512987673283, 0.21153846383094788, 0.05128205195069313, 0.11538461595773697, 0.1666666716337204, 0.26923078298568726, 0.23076923191547394, 0.3076923191547394, 0.5192307829856873, 0.2756410241127014, 0.22435897588729858, 0.4038461446762085, 0.3076923191547394, 0.38461539149284363, 0.20512820780277252, 0.35256409645080566, 0.32692307233810425, 0.4294871687889099, 0.3205128312110901, 0.3717948794364929, 0.36538460850715637, 0.30128204822540283, 0.28205129504203796, 0.4743589758872986, 0.38461539149284363, 0.43589743971824646, 0.5288461446762085, NaN, 0.46794870495796204, 0.5384615659713745, 0.6794871687889099, 0.6153846383094788, 0.9294871687889099, 0.6410256624221802, 0.6538461446762085, 0.7628205418586731, 0.7179487347602844, 0.6858974099159241, 0.8397436141967773, 0.8012820482254028, 0.807692289352417, 0.7243589758872986, 0.7307692170143127, 0.807692289352417, 0.7884615659713745, 0.9679487347602844, 0.9679487347602844, 0.942307710647583, 0.8653846383094788, 0.9807692170143127, 1.0], [0.01923076994717121, 0.006410256493836641, 0.012820512987673283, 0.01923076994717121, 0.03205128386616707, 0.012820512987673283, 0.006410256493836641, 0.07051282376050949, 0.01923076994717121, 0.07692307978868484, 0.006410256493836641, 0.01923076994717121, 0.11538461595773697, 0.0833333358168602, 0.14743590354919434, 0.13461539149284363, 0.12179487198591232, 0.044871795922517776, 0.05128205195069313, 0.14102564752101898, 0.10897435992956161, 0.2371794879436493, 0.0833333358168602, 0.08974359184503555, 0.14743590354919434, 0.1794871836900711, 0.14743590354919434, 0.16025641560554504, 0.30128204822540283, 0.17307692766189575, 0.17307692766189575, 0.2756410241127014, 0.13461539149284363, 0.31410256028175354, 0.14743590354919434, 0.24358974397182465, 0.25, 0.38461539149284363, 0.4038461446762085, 0.25641027092933655, 0.2884615361690521, 0.2756410241127014, 0.41025641560554504, 0.4166666567325592, 0.3205128312110901, 0.3333333432674408, 0.45512819290161133, 0.4423076808452606, 0.39743590354919434, 0.46794870495796204, 0.4615384638309479, 0.44871795177459717, 0.5448718070983887, 0.5128205418586731, 0.5320512652397156, NaN, 0.48076921701431274, 0.5833333134651184, 0.6987179517745972, 0.6153846383094788, 0.6858974099159241, 0.7051281929016113, 0.6346153616905212, 0.7628205418586731, 0.7948718070983887, 0.7820512652397156, 0.7884615659713745, 0.7884615659713745, 0.8525640964508057, 0.8397436141967773, 0.8974359035491943, 0.8525640964508057, 0.9358974099159241, 0.9294871687889099, 0.942307710647583, 0.9294871687889099, 0.9935897588729858, 1.0], [0.07051282376050949, 0.05128205195069313, 0.06410256773233414, 0.025641025975346565, 0.06410256773233414, 0.07051282376050949, 0.0833333358168602, 0.07051282376050949, 0.057692307978868484, 0.08974359184503555, 0.0833333358168602, 0.09615384787321091, 0.07051282376050949, 0.1538461595773697, 0.14102564752101898, 0.13461539149284363, 0.13461539149284363, 0.10897435992956161, 0.10897435992956161, 0.14102564752101898, 0.18589743971824646, 0.14743590354919434, 0.20512820780277252, 0.14743590354919434, 0.11538461595773697, 0.19230769574642181, 0.12820513546466827, 0.057692307978868484, 0.14102564752101898, 0.07692307978868484, 0.16025641560554504, 0.3333333432674408, 0.31410256028175354, 0.19871795177459717, 0.3910256326198578, 0.38461539149284363, 0.38461539149284363, 0.25641027092933655, 0.25, 0.2756410241127014, 0.33974358439445496, 0.25641027092933655, 0.25, 0.4038461446762085, 0.2371794879436493, 0.4743589758872986, 0.3333333432674408, 0.44871795177459717, 0.30128204822540283, 0.32692307233810425, 0.4615384638309479, 0.49358972907066345, 0.5, 0.44871795177459717, 0.4615384638309479, 0.5192307829856873, NaN, 0.5384615659713745, 0.6666666865348816, 0.7179487347602844, 0.6282051205635071, 0.7756410241127014, 0.7564102411270142, 0.6474359035491943, 0.6217948794364929, 0.7756410241127014, 0.8141025900840759, 0.7884615659713745, 0.6666666865348816, 0.8782051205635071, 0.9166666865348816, 0.8846153616905212, 0.8910256624221802, 0.8910256624221802, 0.942307710647583, 0.9294871687889099, 0.9807692170143127, 1.0], [0.07051282376050949, 0.01923076994717121, 0.03205128386616707, 0.012820512987673283, 0.06410256773233414, 0.01923076994717121, 0.06410256773233414, 0.10256410390138626, 0.025641025975346565, 0.10256410390138626, 0.06410256773233414, 0.08974359184503555, 0.16025641560554504, 0.14102564752101898, 0.21153846383094788, 0.14102564752101898, 0.19871795177459717, 0.18589743971824646, 0.09615384787321091, 0.16025641560554504, 0.10256410390138626, 0.17307692766189575, 0.21153846383094788, 0.12179487198591232, 0.20512820780277252, 0.13461539149284363, 0.20512820780277252, 0.23076923191547394, 0.24358974397182465, 0.21794871985912323, 0.21153846383094788, 0.03846153989434242, 0.26923078298568726, 0.1794871836900711, 0.29487180709838867, 0.3910256326198578, 0.25, 0.2371794879436493, 0.2884615361690521, 0.25641027092933655, 0.32692307233810425, 0.28205129504203796, 0.26923078298568726, 0.07692307978868484, 0.3782051205635071, 0.29487180709838867, 0.3589743673801422, 0.4166666567325592, 0.3076923191547394, 0.31410256028175354, 0.3782051205635071, 0.3589743673801422, 0.41025641560554504, 0.33974358439445496, 0.3205128312110901, 0.4166666567325592, 0.4615384638309479, NaN, 0.46794870495796204, 0.5256410241127014, 0.6730769276618958, 0.6346153616905212, 0.6217948794364929, 0.5, 0.6282051205635071, 0.5512820482254028, 0.7628205418586731, 0.7564102411270142, 0.6474359035491943, 0.7051281929016113, 0.7435897588729858, 0.7628205418586731, 0.6858974099159241, 0.7435897588729858, 0.9230769276618958, 0.9230769276618958, 0.8974359035491943, 1.0], [0.06410256773233414, 0.006410256493836641, 0.012820512987673283, 0.012820512987673283, 0.05128205195069313, 0.03205128386616707, 0.03846153989434242, 0.10256410390138626, 0.03846153989434242, 0.08974359184503555, 0.025641025975346565, 0.03205128386616707, 0.11538461595773697, 0.09615384787321091, 0.14743590354919434, 0.10256410390138626, 0.0833333358168602, 0.05128205195069313, 0.03846153989434242, 0.12820513546466827, 0.07051282376050949, 0.13461539149284363, 0.06410256773233414, 0.06410256773233414, 0.07692307978868484, 0.0833333358168602, 0.07692307978868484, 0.10897435992956161, 0.21153846383094788, 0.12820513546466827, 0.13461539149284363, 0.14743590354919434, 0.11538461595773697, 0.21153846383094788, 0.10897435992956161, 0.26923078298568726, 0.07692307978868484, 0.25641027092933655, 0.3461538553237915, 0.14102564752101898, 0.10256410390138626, 0.13461539149284363, 0.28205129504203796, 0.21794871985912323, 0.30128204822540283, 0.11538461595773697, 0.3205128312110901, 0.36538460850715637, 0.23076923191547394, 0.25641027092933655, 0.1794871836900711, 0.21794871985912323, 0.3910256326198578, 0.3782051205635071, 0.38461539149284363, 0.30128204822540283, 0.3333333432674408, 0.5320512652397156, NaN, 0.5256410241127014, 0.5769230723381042, 0.6410256624221802, 0.6153846383094788, 0.6602563858032227, 0.6858974099159241, 0.49358972907066345, 0.7692307829856873, 0.7756410241127014, 0.7307692170143127, 0.6474359035491943, 0.7115384340286255, 0.6858974099159241, 0.8589743375778198, 0.8141025900840759, 0.8974359035491943, 0.9294871687889099, 0.942307710647583, 1.0], [0.057692307978868484, 0.03846153989434242, 0.057692307978868484, 0.012820512987673283, 0.05128205195069313, 0.06410256773233414, 0.06410256773233414, 0.06410256773233414, 0.05128205195069313, 0.12179487198591232, 0.11538461595773697, 0.10897435992956161, 0.11538461595773697, 0.14743590354919434, 0.10897435992956161, 0.12179487198591232, 0.07051282376050949, 0.10256410390138626, 0.07692307978868484, 0.1666666716337204, 0.10897435992956161, 0.13461539149284363, 0.14102564752101898, 0.07692307978868484, 0.07051282376050949, 0.0833333358168602, 0.07051282376050949, 0.01923076994717121, 0.12820513546466827, 0.03205128386616707, 0.06410256773233414, 0.11538461595773697, 0.1538461595773697, 0.12179487198591232, 0.23076923191547394, 0.39743590354919434, 0.16025641560554504, 0.14102564752101898, 0.22435897588729858, 0.16025641560554504, 0.33974358439445496, 0.057692307978868484, 0.19230769574642181, 0.1666666716337204, 0.2371794879436493, 0.20512820780277252, 0.19230769574642181, 0.31410256028175354, 0.12820513546466827, 0.19871795177459717, 0.3589743673801422, 0.2371794879436493, 0.33974358439445496, 0.07051282376050949, 0.07051282376050949, 0.38461539149284363, 0.28205129504203796, 0.4743589758872986, 0.4743589758872986, NaN, 0.5, 0.4615384638309479, 0.557692289352417, 0.31410256028175354, 0.6089743375778198, 0.6410256624221802, 0.7179487347602844, 0.7371794581413269, 0.6153846383094788, 0.6474359035491943, 0.6987179517745972, 0.6794871687889099, 0.75, 0.6666666865348816, 0.8205128312110901, 0.8525640964508057, 0.9102563858032227, 1.0], [0.006410256493836641, 0.012820512987673283, 0.012820512987673283, 0.006410256493836641, 0.0, 0.0, 0.03846153989434242, 0.006410256493836641, 0.0, 0.1538461595773697, 0.06410256773233414, 0.07051282376050949, 0.06410256773233414, 0.16025641560554504, 0.0, 0.01923076994717121, 0.09615384787321091, 0.10897435992956161, 0.07692307978868484, 0.14102564752101898, 0.05128205195069313, 0.06410256773233414, 0.12179487198591232, 0.0833333358168602, 0.0833333358168602, 0.07692307978868484, 0.08974359184503555, 0.09615384787321091, 0.05448717996478081, 0.10897435992956161, 0.07692307978868484, 0.08974359184503555, 0.22435897588729858, 0.03205128386616707, 0.2371794879436493, 0.3461538553237915, 0.19871795177459717, 0.012820512987673283, 0.19871795177459717, 0.12820513546466827, 0.24358974397182465, 0.14743590354919434, 0.10256410390138626, 0.13461539149284363, 0.13461539149284363, 0.23076923191547394, 0.16987179219722748, 0.3910256326198578, 0.12820513546466827, 0.14102564752101898, 0.2756410241127014, 0.2628205120563507, 0.4038461446762085, 0.3589743673801422, 0.3589743673801422, 0.31410256028175354, 0.3717948794364929, 0.32692307233810425, 0.42307692766189575, 0.5, NaN, 0.6089743375778198, 0.46794870495796204, 0.5256410241127014, 0.5256410241127014, 0.4294871687889099, 0.5705128312110901, 0.5705128312110901, 0.5833333134651184, 0.7115384340286255, 0.7435897588729858, 0.7179487347602844, 0.6794871687889099, 0.6602563858032227, 0.8333333134651184, 0.8910256624221802, 0.8717948794364929, 1.0], [0.06410256773233414, 0.06410256773233414, 0.07051282376050949, 0.01923076994717121, 0.044871795922517776, 0.05128205195069313, 0.11538461595773697, 0.06410256773233414, 0.05128205195069313, 0.10256410390138626, 0.07692307978868484, 0.07051282376050949, 0.057692307978868484, 0.11538461595773697, 0.07692307978868484, 0.03205128386616707, 0.1666666716337204, 0.10897435992956161, 0.12820513546466827, 0.13461539149284363, 0.0833333358168602, 0.03846153989434242, 0.14743590354919434, 0.14743590354919434, 0.14743590354919434, 0.07692307978868484, 0.14743590354919434, 0.14102564752101898, 0.06410256773233414, 0.13461539149284363, 0.14102564752101898, 0.0833333358168602, 0.19230769574642181, 0.044871795922517776, 0.22435897588729858, 0.41025641560554504, 0.2371794879436493, 0.08974359184503555, 0.14102564752101898, 0.22435897588729858, 0.25641027092933655, 0.19230769574642181, 0.06410256773233414, 0.12179487198591232, 0.14743590354919434, 0.2884615361690521, 0.14743590354919434, 0.3461538553237915, 0.18589743971824646, 0.17307692766189575, 0.29487180709838867, 0.23076923191547394, 0.3205128312110901, 0.3461538553237915, 0.3461538553237915, 0.29487180709838867, 0.22435897588729858, 0.36538460850715637, 0.3589743673801422, 0.5384615659713745, 0.3910256326198578, NaN, 0.4294871687889099, 0.4166666567325592, 0.4615384638309479, 0.45512819290161133, 0.45512819290161133, 0.46794870495796204, 0.46794870495796204, 0.6474359035491943, 0.6858974099159241, 0.6987179517745972, 0.5961538553237915, 0.5641025900840759, 0.6410256624221802, 0.9358974099159241, 0.7051281929016113, 0.9935897588729858], [0.06410256773233414, 0.01923076994717121, 0.012820512987673283, 0.006410256493836641, 0.03205128386616707, 0.006410256493836641, 0.012820512987673283, 0.0833333358168602, 0.025641025975346565, 0.0833333358168602, 0.012820512987673283, 0.01923076994717121, 0.08974359184503555, 0.10256410390138626, 0.057692307978868484, 0.03846153989434242, 0.08974359184503555, 0.057692307978868484, 0.025641025975346565, 0.09615384787321091, 0.044871795922517776, 0.03846153989434242, 0.07051282376050949, 0.025641025975346565, 0.03846153989434242, 0.03846153989434242, 0.044871795922517776, 0.07692307978868484, 0.09615384787321091, 0.06410256773233414, 0.12820513546466827, 0.025641025975346565, 0.13461539149284363, 0.05128205195069313, 0.16025641560554504, 0.32692307233810425, 0.12179487198591232, 0.09615384787321091, 0.1794871836900711, 0.12820513546466827, 0.22435897588729858, 0.10256410390138626, 0.0833333358168602, 0.057692307978868484, 0.07051282376050949, 0.17307692766189575, 0.13461539149284363, 0.4038461446762085, 0.10897435992956161, 0.17307692766189575, 0.30128204822540283, 0.19871795177459717, 0.4166666567325592, 0.24358974397182465, 0.2371794879436493, 0.36538460850715637, 0.24358974397182465, 0.3782051205635071, 0.38461539149284363, 0.4423076808452606, 0.5320512652397156, 0.5705128312110901, NaN, 0.4743589758872986, 0.5705128312110901, 0.4871794879436493, 0.6730769276618958, 0.6538461446762085, 0.5897436141967773, 0.6794871687889099, 0.7692307829856873, 0.75, 0.75, 0.7628205418586731, 0.8974359035491943, 0.9166666865348816, 0.8974359035491943, 1.0], [0.0, 0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.0, 0.0, 0.006410256493836641, 0.025641025975346565, 0.0, 0.057692307978868484, 0.0, 0.01923076994717121, 0.057692307978868484, 0.07051282376050949, 0.0, 0.0, 0.07051282376050949, 0.0833333358168602, 0.05128205195069313, 0.13461539149284363, 0.012820512987673283, 0.01923076994717121, 0.14102564752101898, 0.07051282376050949, 0.07051282376050949, 0.03205128386616707, 0.0833333358168602, 0.05128205195069313, 0.06410256773233414, 0.05128205195069313, 0.006410256493836641, 0.006410256493836641, 0.17307692766189575, 0.0, 0.18589743971824646, 0.3461538553237915, 0.18589743971824646, 0.044871795922517776, 0.07692307978868484, 0.07051282376050949, 0.19871795177459717, 0.10897435992956161, 0.10897435992956161, 0.05128205195069313, 0.22435897588729858, 0.21153846383094788, 0.19871795177459717, 0.20512820780277252, 0.12820513546466827, 0.1538461595773697, 0.25, 0.23076923191547394, 0.23076923191547394, 0.29487180709838867, 0.28205129504203796, 0.2371794879436493, 0.35256409645080566, 0.5, 0.33974358439445496, 0.6858974099159241, 0.4743589758872986, 0.5833333134651184, 0.5256410241127014, NaN, 0.557692289352417, 0.557692289352417, 0.5769230723381042, 0.5833333134651184, 0.557692289352417, 0.6153846383094788, 0.7179487347602844, 0.7115384340286255, 0.7820512652397156, 0.7179487347602844, 0.7307692170143127, 0.8461538553237915, 0.8589743375778198, 1.0], [0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.025641025975346565, 0.01923076994717121, 0.012820512987673283, 0.0, 0.012820512987673283, 0.012820512987673283, 0.012820512987673283, 0.006410256493836641, 0.012820512987673283, 0.01923076994717121, 0.03205128386616707, 0.12820513546466827, 0.13461539149284363, 0.05128205195069313, 0.03846153989434242, 0.03846153989434242, 0.057692307978868484, 0.0833333358168602, 0.1794871836900711, 0.025641025975346565, 0.044871795922517776, 0.05128205195069313, 0.10897435992956161, 0.05128205195069313, 0.10897435992956161, 0.1794871836900711, 0.10256410390138626, 0.17307692766189575, 0.03846153989434242, 0.057692307978868484, 0.22435897588729858, 0.0833333358168602, 0.18589743971824646, 0.057692307978868484, 0.2756410241127014, 0.39743590354919434, 0.21794871985912323, 0.12179487198591232, 0.1538461595773697, 0.2756410241127014, 0.12179487198591232, 0.3205128312110901, 0.10256410390138626, 0.3076923191547394, 0.1794871836900711, 0.2628205120563507, 0.21794871985912323, 0.25, 0.20512820780277252, 0.2371794879436493, 0.30128204822540283, 0.31410256028175354, 0.20512820780277252, 0.3782051205635071, 0.3717948794364929, 0.31410256028175354, 0.3910256326198578, 0.4743589758872986, 0.5384615659713745, 0.4294871687889099, 0.4423076808452606, NaN, 0.4294871687889099, 0.4294871687889099, 0.45512819290161133, 0.7596153616905212, 0.45512819290161133, 0.4743589758872986, 0.5448718070983887, 0.44871795177459717, 0.557692289352417, 0.6282051205635071, 0.7692307829856873, 0.9487179517745972, 0.9358974099159241], [0.057692307978868484, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.012820512987673283, 0.057692307978868484, 0.025641025975346565, 0.11538461595773697, 0.057692307978868484, 0.06410256773233414, 0.11538461595773697, 0.11538461595773697, 0.057692307978868484, 0.11538461595773697, 0.01923076994717121, 0.057692307978868484, 0.0, 0.16025641560554504, 0.01923076994717121, 0.12179487198591232, 0.057692307978868484, 0.0, 0.006410256493836641, 0.025641025975346565, 0.006410256493836641, 0.0, 0.10256410390138626, 0.0, 0.006410256493836641, 0.1538461595773697, 0.06410256773233414, 0.08974359184503555, 0.07051282376050949, 0.3205128312110901, 0.05128205195069313, 0.14743590354919434, 0.13461539149284363, 0.01923076994717121, 0.13461539149284363, 0.0, 0.19871795177459717, 0.19230769574642181, 0.19230769574642181, 0.08974359184503555, 0.19230769574642181, 0.25641027092933655, 0.03846153989434242, 0.21153846383094788, 0.18589743971824646, 0.13461539149284363, 0.25641027092933655, 0.1666666716337204, 0.16025641560554504, 0.21794871985912323, 0.22435897588729858, 0.44871795177459717, 0.5064102411270142, 0.3589743673801422, 0.5705128312110901, 0.5448718070983887, 0.5128205418586731, 0.4423076808452606, 0.5705128312110901, NaN, 0.7051281929016113, 0.6987179517745972, 0.5769230723381042, 0.6730769276618958, 0.6987179517745972, 0.6987179517745972, 0.6794871687889099, 0.7115384340286255, 0.7884615659713745, 0.7692307829856873, 0.7820512652397156, 0.9358974099159241], [0.057692307978868484, 0.012820512987673283, 0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.0, 0.012820512987673283, 0.06410256773233414, 0.01923076994717121, 0.07051282376050949, 0.025641025975346565, 0.025641025975346565, 0.07051282376050949, 0.08974359184503555, 0.11538461595773697, 0.08974359184503555, 0.057692307978868484, 0.057692307978868484, 0.01923076994717121, 0.0833333358168602, 0.0833333358168602, 0.08974359184503555, 0.07051282376050949, 0.03205128386616707, 0.044871795922517776, 0.07051282376050949, 0.03846153989434242, 0.05128205195069313, 0.14102564752101898, 0.044871795922517776, 0.12820513546466827, 0.03205128386616707, 0.06410256773233414, 0.0833333358168602, 0.07051282376050949, 0.25, 0.06410256773233414, 0.07051282376050949, 0.22435897588729858, 0.14743590354919434, 0.09615384787321091, 0.07692307978868484, 0.0833333358168602, 0.07051282376050949, 0.12179487198591232, 0.07692307978868484, 0.11538461595773697, 0.2756410241127014, 0.11538461595773697, 0.20512820780277252, 0.12820513546466827, 0.08974359184503555, 0.3076923191547394, 0.20512820780277252, 0.19871795177459717, 0.21153846383094788, 0.18589743971824646, 0.2371794879436493, 0.23076923191547394, 0.28205129504203796, 0.4294871687889099, 0.5448718070983887, 0.32692307233810425, 0.42307692766189575, 0.5705128312110901, 0.29487180709838867, NaN, 0.5416666865348816, 0.6217948794364929, 0.5384615659713745, 0.5769230723381042, 0.557692289352417, 0.6089743375778198, 0.4615384638309479, 1.0, 0.8589743375778198, 0.9102563858032227, 0.9871794581413269], [0.06410256773233414, 0.012820512987673283, 0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.01923076994717121, 0.01923076994717121, 0.06410256773233414, 0.03205128386616707, 0.07051282376050949, 0.03205128386616707, 0.03846153989434242, 0.07051282376050949, 0.10256410390138626, 0.11538461595773697, 0.08974359184503555, 0.057692307978868484, 0.044871795922517776, 0.025641025975346565, 0.08974359184503555, 0.08974359184503555, 0.08974359184503555, 0.044871795922517776, 0.03846153989434242, 0.05128205195069313, 0.06410256773233414, 0.044871795922517776, 0.05128205195069313, 0.14102564752101898, 0.03846153989434242, 0.13461539149284363, 0.03205128386616707, 0.06410256773233414, 0.0833333358168602, 0.06410256773233414, 0.2628205120563507, 0.057692307978868484, 0.07051282376050949, 0.2371794879436493, 0.14743590354919434, 0.08974359184503555, 0.0833333358168602, 0.08974359184503555, 0.06410256773233414, 0.13461539149284363, 0.07051282376050949, 0.12179487198591232, 0.2756410241127014, 0.12820513546466827, 0.20512820780277252, 0.13461539149284363, 0.0833333358168602, 0.29487180709838867, 0.19871795177459717, 0.19230769574642181, 0.21153846383094788, 0.21153846383094788, 0.24358974397182465, 0.22435897588729858, 0.2628205120563507, 0.4294871687889099, 0.5320512652397156, 0.3461538553237915, 0.4166666567325592, 0.5448718070983887, 0.30128204822540283, 0.4583333432674408, NaN, 0.6089743375778198, 0.5256410241127014, 0.5641025900840759, 0.557692289352417, 0.5961538553237915, 0.45512819290161133, 1.0, 0.8589743375778198, 0.8910256624221802, 0.9871794581413269], [0.006410256493836641, 0.0, 0.006410256493836641, 0.01923076994717121, 0.012820512987673283, 0.012820512987673283, 0.006410256493836641, 0.01923076994717121, 0.006410256493836641, 0.006410256493836641, 0.012820512987673283, 0.006410256493836641, 0.03205128386616707, 0.03205128386616707, 0.14102564752101898, 0.12179487198591232, 0.03846153989434242, 0.01923076994717121, 0.03205128386616707, 0.07051282376050949, 0.03846153989434242, 0.18589743971824646, 0.025641025975346565, 0.03205128386616707, 0.03846153989434242, 0.06410256773233414, 0.03846153989434242, 0.07051282376050949, 0.1666666716337204, 0.07051282376050949, 0.10897435992956161, 0.03205128386616707, 0.057692307978868484, 0.22435897588729858, 0.06410256773233414, 0.18589743971824646, 0.025641025975346565, 0.26923078298568726, 0.3333333432674408, 0.16025641560554504, 0.07692307978868484, 0.12820513546466827, 0.29487180709838867, 0.11538461595773697, 0.2884615361690521, 0.07051282376050949, 0.31410256028175354, 0.13461539149284363, 0.22435897588729858, 0.1794871836900711, 0.19871795177459717, 0.13461539149284363, 0.19230769574642181, 0.28205129504203796, 0.2756410241127014, 0.14743590354919434, 0.3333333432674408, 0.35256409645080566, 0.26923078298568726, 0.38461539149284363, 0.4166666567325592, 0.5320512652397156, 0.41025641560554504, 0.4423076808452606, 0.24038460850715637, 0.42307692766189575, 0.3782051205635071, 0.3910256326198578, NaN, 0.42307692766189575, 0.45512819290161133, 0.5192307829856873, 0.41025641560554504, 0.5512820482254028, 0.6153846383094788, 0.7564102411270142, 0.8589743375778198, 0.9358974099159241], [0.006410256493836641, 0.012820512987673283, 0.012820512987673283, 0.006410256493836641, 0.03205128386616707, 0.025641025975346565, 0.012820512987673283, 0.025641025975346565, 0.01923076994717121, 0.025641025975346565, 0.01923076994717121, 0.012820512987673283, 0.044871795922517776, 0.03205128386616707, 0.057692307978868484, 0.03846153989434242, 0.057692307978868484, 0.044871795922517776, 0.025641025975346565, 0.01923076994717121, 0.01923076994717121, 0.06410256773233414, 0.03846153989434242, 0.025641025975346565, 0.03846153989434242, 0.03205128386616707, 0.03846153989434242, 0.057692307978868484, 0.05128205195069313, 0.057692307978868484, 0.10256410390138626, 0.0833333358168602, 0.06410256773233414, 0.07051282376050949, 0.057692307978868484, 0.23076923191547394, 0.07692307978868484, 0.0833333358168602, 0.14102564752101898, 0.10256410390138626, 0.1538461595773697, 0.07692307978868484, 0.0833333358168602, 0.10256410390138626, 0.06410256773233414, 0.09615384787321091, 0.10256410390138626, 0.31410256028175354, 0.10256410390138626, 0.11538461595773697, 0.20512820780277252, 0.11538461595773697, 0.3333333432674408, 0.26923078298568726, 0.26923078298568726, 0.16025641560554504, 0.12179487198591232, 0.29487180709838867, 0.35256409645080566, 0.35256409645080566, 0.2884615361690521, 0.35256409645080566, 0.3205128312110901, 0.38461539149284363, 0.5448718070983887, 0.32692307233810425, 0.4615384638309479, 0.4743589758872986, 0.5769230723381042, NaN, 0.7884615659713745, 0.8141025900840759, 0.6025640964508057, 0.6538461446762085, 0.5320512652397156, 0.6474359035491943, 0.6794871687889099, 0.942307710647583], [0.025641025975346565, 0.025641025975346565, 0.025641025975346565, 0.025641025975346565, 0.044871795922517776, 0.025641025975346565, 0.025641025975346565, 0.03205128386616707, 0.025641025975346565, 0.03205128386616707, 0.025641025975346565, 0.025641025975346565, 0.044871795922517776, 0.03846153989434242, 0.057692307978868484, 0.03846153989434242, 0.05128205195069313, 0.03205128386616707, 0.025641025975346565, 0.03205128386616707, 0.025641025975346565, 0.05128205195069313, 0.044871795922517776, 0.025641025975346565, 0.03846153989434242, 0.03205128386616707, 0.044871795922517776, 0.044871795922517776, 0.057692307978868484, 0.044871795922517776, 0.0833333358168602, 0.10256410390138626, 0.03846153989434242, 0.044871795922517776, 0.06410256773233414, 0.2371794879436493, 0.06410256773233414, 0.06410256773233414, 0.12179487198591232, 0.0833333358168602, 0.11538461595773697, 0.05128205195069313, 0.0833333358168602, 0.07692307978868484, 0.05128205195069313, 0.0833333358168602, 0.07692307978868484, 0.2628205120563507, 0.07051282376050949, 0.08974359184503555, 0.16025641560554504, 0.09615384787321091, 0.28205129504203796, 0.21153846383094788, 0.19230769574642181, 0.10256410390138626, 0.0833333358168602, 0.25641027092933655, 0.2884615361690521, 0.30128204822540283, 0.25641027092933655, 0.31410256028175354, 0.23076923191547394, 0.28205129504203796, 0.5256410241127014, 0.30128204822540283, 0.42307692766189575, 0.43589743971824646, 0.5448718070983887, 0.21153846383094788, NaN, 0.6698718070983887, 0.5705128312110901, 0.5384615659713745, 0.4743589758872986, 0.6410256624221802, 0.6538461446762085, 0.9358974099159241], [0.03205128386616707, 0.03205128386616707, 0.025641025975346565, 0.03205128386616707, 0.03205128386616707, 0.03205128386616707, 0.025641025975346565, 0.03205128386616707, 0.03205128386616707, 0.03205128386616707, 0.03205128386616707, 0.03846153989434242, 0.03846153989434242, 0.025641025975346565, 0.05128205195069313, 0.044871795922517776, 0.044871795922517776, 0.044871795922517776, 0.03205128386616707, 0.03205128386616707, 0.044871795922517776, 0.044871795922517776, 0.05128205195069313, 0.03205128386616707, 0.03205128386616707, 0.05128205195069313, 0.03846153989434242, 0.057692307978868484, 0.05128205195069313, 0.057692307978868484, 0.07692307978868484, 0.10256410390138626, 0.057692307978868484, 0.057692307978868484, 0.07051282376050949, 0.2371794879436493, 0.06410256773233414, 0.06410256773233414, 0.10897435992956161, 0.08974359184503555, 0.12179487198591232, 0.07692307978868484, 0.07692307978868484, 0.08974359184503555, 0.05128205195069313, 0.08974359184503555, 0.0833333358168602, 0.2628205120563507, 0.07692307978868484, 0.09615384787321091, 0.17307692766189575, 0.12179487198591232, 0.2756410241127014, 0.22435897588729858, 0.21153846383094788, 0.14743590354919434, 0.11538461595773697, 0.2371794879436493, 0.31410256028175354, 0.3205128312110901, 0.28205129504203796, 0.30128204822540283, 0.25, 0.2884615361690521, 0.45512819290161133, 0.30128204822540283, 0.4423076808452606, 0.4423076808452606, 0.48076921701431274, 0.18589743971824646, 0.33012819290161133, NaN, 0.5064102411270142, 0.5384615659713745, 0.5064102411270142, 0.6153846383094788, 0.5897436141967773, 0.7564102411270142], [0.0, 0.0, 0.006410256493836641, 0.006410256493836641, 0.0, 0.006410256493836641, 0.006410256493836641, 0.012820512987673283, 0.0, 0.057692307978868484, 0.0, 0.0, 0.06410256773233414, 0.07692307978868484, 0.057692307978868484, 0.057692307978868484, 0.01923076994717121, 0.01923076994717121, 0.01923076994717121, 0.08974359184503555, 0.05128205195069313, 0.057692307978868484, 0.025641025975346565, 0.025641025975346565, 0.012820512987673283, 0.057692307978868484, 0.012820512987673283, 0.0, 0.11538461595773697, 0.0, 0.006410256493836641, 0.03205128386616707, 0.006410256493836641, 0.09615384787321091, 0.01923076994717121, 0.19871795177459717, 0.01923076994717121, 0.10897435992956161, 0.08974359184503555, 0.057692307978868484, 0.03846153989434242, 0.03205128386616707, 0.10897435992956161, 0.07051282376050949, 0.12179487198591232, 0.03205128386616707, 0.12179487198591232, 0.006410256493836641, 0.05128205195069313, 0.13461539149284363, 0.07051282376050949, 0.03205128386616707, 0.012820512987673283, 0.05128205195069313, 0.03205128386616707, 0.06410256773233414, 0.10897435992956161, 0.31410256028175354, 0.14102564752101898, 0.25, 0.3205128312110901, 0.4038461446762085, 0.25, 0.21794871985912323, 0.5512820482254028, 0.3205128312110901, 0.3910256326198578, 0.4038461446762085, 0.5897436141967773, 0.39743590354919434, 0.4294871687889099, 0.49358972907066345, NaN, 0.38461539149284363, 0.7179487347602844, 0.7948718070983887, 0.9038461446762085, 0.9935897588729858], [0.03205128386616707, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.05128205195069313, 0.006410256493836641, 0.057692307978868484, 0.0, 0.0, 0.057692307978868484, 0.057692307978868484, 0.0, 0.0, 0.01923076994717121, 0.0, 0.0, 0.0833333358168602, 0.006410256493836641, 0.0, 0.0, 0.0, 0.006410256493836641, 0.012820512987673283, 0.006410256493836641, 0.0, 0.057692307978868484, 0.0, 0.0, 0.006410256493836641, 0.0, 0.01923076994717121, 0.0, 0.2371794879436493, 0.006410256493836641, 0.05128205195069313, 0.0, 0.025641025975346565, 0.006410256493836641, 0.0, 0.03205128386616707, 0.006410256493836641, 0.044871795922517776, 0.006410256493836641, 0.07051282376050949, 0.07692307978868484, 0.012820512987673283, 0.07051282376050949, 0.03205128386616707, 0.01923076994717121, 0.16025641560554504, 0.06410256773233414, 0.03205128386616707, 0.07051282376050949, 0.10897435992956161, 0.25641027092933655, 0.18589743971824646, 0.3333333432674408, 0.33974358439445496, 0.43589743971824646, 0.2371794879436493, 0.28205129504203796, 0.4423076808452606, 0.2884615361690521, 0.5384615659713745, 0.5448718070983887, 0.44871795177459717, 0.3461538553237915, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, NaN, 0.7243589758872986, 0.807692289352417, 0.7756410241127014, 1.0], [0.006410256493836641, 0.012820512987673283, 0.006410256493836641, 0.006410256493836641, 0.006410256493836641, 0.0, 0.0, 0.006410256493836641, 0.0, 0.06410256773233414, 0.006410256493836641, 0.012820512987673283, 0.06410256773233414, 0.06410256773233414, 0.057692307978868484, 0.06410256773233414, 0.03205128386616707, 0.006410256493836641, 0.006410256493836641, 0.07051282376050949, 0.03205128386616707, 0.07051282376050949, 0.012820512987673283, 0.006410256493836641, 0.012820512987673283, 0.03205128386616707, 0.012820512987673283, 0.006410256493836641, 0.10897435992956161, 0.006410256493836641, 0.025641025975346565, 0.012820512987673283, 0.012820512987673283, 0.06410256773233414, 0.01923076994717121, 0.05128205195069313, 0.012820512987673283, 0.057692307978868484, 0.10256410390138626, 0.05128205195069313, 0.025641025975346565, 0.006410256493836641, 0.06410256773233414, 0.01923076994717121, 0.06410256773233414, 0.01923076994717121, 0.07692307978868484, 0.06410256773233414, 0.012820512987673283, 0.12820513546466827, 0.044871795922517776, 0.025641025975346565, 0.06410256773233414, 0.0833333358168602, 0.057692307978868484, 0.057692307978868484, 0.057692307978868484, 0.07692307978868484, 0.10256410390138626, 0.1794871836900711, 0.1666666716337204, 0.3589743673801422, 0.10256410390138626, 0.26923078298568726, 0.3717948794364929, 0.21153846383094788, 0.0, 0.0, 0.38461539149284363, 0.46794870495796204, 0.5256410241127014, 0.49358972907066345, 0.28205129504203796, 0.2756410241127014, NaN, 0.6474359035491943, 0.6987179517745972, 0.9615384340286255], [0.06410256773233414, 0.03205128386616707, 0.044871795922517776, 0.01923076994717121, 0.03846153989434242, 0.05128205195069313, 0.05128205195069313, 0.057692307978868484, 0.044871795922517776, 0.057692307978868484, 0.044871795922517776, 0.025641025975346565, 0.057692307978868484, 0.057692307978868484, 0.05128205195069313, 0.025641025975346565, 0.044871795922517776, 0.03205128386616707, 0.044871795922517776, 0.06410256773233414, 0.06410256773233414, 0.03846153989434242, 0.03205128386616707, 0.044871795922517776, 0.06410256773233414, 0.025641025975346565, 0.06410256773233414, 0.044871795922517776, 0.057692307978868484, 0.044871795922517776, 0.09615384787321091, 0.012820512987673283, 0.025641025975346565, 0.0, 0.03846153989434242, 0.22435897588729858, 0.057692307978868484, 0.006410256493836641, 0.11538461595773697, 0.10897435992956161, 0.057692307978868484, 0.05128205195069313, 0.0, 0.025641025975346565, 0.044871795922517776, 0.057692307978868484, 0.025641025975346565, 0.10256410390138626, 0.07051282376050949, 0.09615384787321091, 0.057692307978868484, 0.025641025975346565, 0.10256410390138626, 0.13461539149284363, 0.13461539149284363, 0.07051282376050949, 0.07051282376050949, 0.07692307978868484, 0.07051282376050949, 0.14743590354919434, 0.10897435992956161, 0.06410256773233414, 0.0833333358168602, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.14102564752101898, 0.14102564752101898, 0.24358974397182465, 0.35256409645080566, 0.3589743673801422, 0.38461539149284363, 0.20512820780277252, 0.19230769574642181, 0.35256409645080566, NaN, 0.35256409645080566, 0.9294871687889099], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03205128386616707, 0.044871795922517776, 0.006410256493836641, 0.0, 0.0, 0.006410256493836641, 0.006410256493836641, 0.057692307978868484, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.05128205195069313, 0.0, 0.1794871836900711, 0.006410256493836641, 0.07692307978868484, 0.06410256773233414, 0.025641025975346565, 0.006410256493836641, 0.0, 0.07051282376050949, 0.006410256493836641, 0.03205128386616707, 0.006410256493836641, 0.09615384787321091, 0.0, 0.0, 0.01923076994717121, 0.012820512987673283, 0.0, 0.0, 0.025641025975346565, 0.01923076994717121, 0.006410256493836641, 0.01923076994717121, 0.10256410390138626, 0.057692307978868484, 0.08974359184503555, 0.12820513546466827, 0.29487180709838867, 0.10256410390138626, 0.14102564752101898, 0.05128205195069313, 0.21794871985912323, 0.08974359184503555, 0.10897435992956161, 0.14102564752101898, 0.3205128312110901, 0.3461538553237915, 0.41025641560554504, 0.09615384787321091, 0.22435897588729858, 0.30128204822540283, 0.6474359035491943, NaN, 0.9166666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006410256493836641, 0.0, 0.0, 0.06410256773233414, 0.06410256773233414, 0.012820512987673283, 0.012820512987673283, 0.06410256773233414, 0.057692307978868484, 0.06410256773233414, 0.24358974397182465, 0.006410256493836641, 0.0, 0.03846153989434242, 0.07051282376050949, 0.0833333358168602, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9084249054},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8800782523},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8789543761},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8678821169},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8317515807},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8296703263},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8189726888},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8011987992},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7750582736},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.769105896},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7685231412},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7482101201},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.7340576092},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.7072510829},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7039627072},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.698717949},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6883116896},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6812770589},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6657092913},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.6636280374},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6523892803},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6487678972},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6472277715},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6380702621},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6242507491},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.619005995},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6185897434},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6137612423},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6090159847},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6077672323},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6033966061},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5780885775},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5733433212},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5687229434},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5587745595},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5552780547},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5381701664},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5362970355},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5323010312},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5220612748},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5209373952},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5131535161},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.5012071293},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.4955044949},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.4869297372},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.485056612},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4633283376},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4461371953},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.446053946},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4459706972},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.436688315},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4350649376},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3848651372},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.3831585101},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.380827509},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.3786213801},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3783716323},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3388278412},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3000333014},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.276057278},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2626540133},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2569930096},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2538295058},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2425907423},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2342241092},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2332667351},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2111638375},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2099983361},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2071678334},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1953879475},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1677905452},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1626290386},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1539294052},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1447718964},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1123876135},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1042291057},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0737595753},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0104062607}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..0d5b516964bfd7218575541184e33ea6003c3b2a
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_all/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),TabPFN-3 (default),RealTabPFN-2.5 (default),TabICLv2 (default),LimiX (default),TabPFNv2 (tuned + ensembled),RealMLP (tuned + ensembled),RealMLP (tuned),Mitra (default),TabPFNv2 (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabM (tuned + ensembled),CatBoost (tuned + ensembled),TabPFNv2 (default),ModernNCA (tuned + ensembled),TabDPT (tuned),TabM (tuned),CatBoost (tuned),XGBoost (tuned),ModernNCA (tuned),XGBoost (tuned + ensembled),LightGBM (tuned + ensembled),TabICL (default),LightGBM (tuned),iLTM (tuned + ensembled),xRFM (tuned + ensembled),BetaTabPFN (default),ExtraTrees (tuned + ensembled),TabM (default),SAP-RPT-OSS (default),EBM (tuned + ensembled),RandomForest (tuned + ensembled),TabDPT (default),iLTM (tuned),TorchMLP (tuned + ensembled),CatBoost (default),ExtraTrees (tuned),xRFM (tuned),TabSwift (default),EBM (tuned),RandomForest (tuned),FastaiMLP (tuned + ensembled),XGBoost (default),ModernNCA (default),TorchMLP (tuned),EBM (default),FastaiMLP (tuned),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),RealMLP (default),LightGBM (default),xRFM (default),TorchMLP (default),ChimeraBoost (default),RandomForest (default),KNN (tuned + ensembled),OrionMSP (default),ExtraTrees (default),Linear (tuned + ensembled),iLTM (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),Linear (tuned),PerpetualBooster (tuned + ensembled),PerpetualBooster (tuned),PerpetualBooster (default),FastaiMLP (default),TabFlex (default),TabSTAR (default),KNN (tuned),Linear (default),KNN (default)
+TabFM (default),,0.6666667,0.6602564,0.7692308,0.8141026,0.80128205,0.75641024,0.8974359,0.7948718,0.8397436,0.78205127,0.78205127,0.9230769,0.9166667,0.89102566,0.90384614,0.88461536,0.82051283,0.9102564,0.9166667,0.9230769,0.89102566,0.8333333,0.92948717,0.88461536,0.8974359,0.89102566,0.9102564,0.974359,0.90384614,0.9230769,0.9230769,0.9166667,0.92948717,0.85897434,0.7692308,0.9102564,0.9102564,0.9102564,0.9358974,0.92948717,0.9358974,0.9230769,0.92948717,0.9230769,0.92948717,0.9102564,0.90384614,0.9358974,0.9807692,0.9358974,0.92948717,0.9423077,0.92948717,0.92948717,0.9807692,0.92948717,0.92948717,0.9358974,0.9423077,0.99358976,0.9358974,0.9358974,1.0,0.99358976,0.9423077,0.9423077,0.9358974,0.99358976,0.99358976,0.974359,0.96794873,1.0,0.96794873,0.99358976,0.9358974,1.0,1.0
+RealTabPFN-2.5 (tuned),0.33333334,,0.43589744,0.5448718,0.64102566,0.5833333,0.7019231,0.6730769,0.6923077,0.724359,0.75641024,0.78205127,0.724359,0.8141026,0.71153843,0.74358976,0.8525641,0.8525641,0.92948717,0.88461536,0.85897434,0.78846157,0.85897434,0.92948717,0.9102564,0.8717949,0.9102564,0.89102566,0.80128205,0.9166667,0.88461536,0.8974359,0.90384614,0.8717949,0.86538464,0.76282054,0.9551282,0.88461536,0.8076923,0.9551282,0.94871795,0.96794873,0.8974359,0.9230769,0.974359,0.96153843,0.9102564,0.92948717,0.9807692,0.94871795,0.94871795,0.94871795,0.96794873,0.9551282,0.96794873,0.99358976,0.94871795,0.9807692,0.99358976,0.96153843,0.98717946,0.9358974,0.9807692,0.99358976,0.99358976,1.0,0.98717946,0.98717946,1.0,0.98717946,0.974359,0.96794873,1.0,1.0,0.98717946,0.96794873,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.33974358,0.5641026,,0.5833333,0.6474359,0.59615386,0.76282054,0.67948717,0.69871795,0.7692308,0.75,0.76282054,0.724359,0.8076923,0.7051282,0.73717946,0.85897434,0.8076923,0.9423077,0.8782051,0.86538464,0.75641024,0.8076923,0.92948717,0.90384614,0.88461536,0.9230769,0.8974359,0.80128205,0.90384614,0.89102566,0.88461536,0.9166667,0.8525641,0.8525641,0.75641024,0.9358974,0.8782051,0.78846157,0.9551282,0.9423077,0.9423077,0.8782051,0.9230769,0.974359,0.9423077,0.9102564,0.9358974,0.98717946,0.9358974,0.94871795,0.9358974,0.9551282,0.9423077,0.9551282,0.98717946,0.9358974,0.96794873,0.98717946,0.9423077,0.98717946,0.92948717,0.98717946,0.99358976,0.99358976,1.0,0.99358976,0.99358976,0.99358976,0.98717946,0.974359,0.974359,0.99358976,1.0,0.99358976,0.9551282,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.23076923,0.4551282,0.41666666,,0.6217949,0.5576923,0.51282054,0.59615386,0.74358976,0.5833333,0.73717946,0.75,0.7307692,0.71153843,0.75,0.8269231,0.8076923,0.78205127,0.89102566,0.8333333,0.8717949,0.86538464,0.775641,0.90384614,0.8397436,0.89102566,0.86538464,0.9102564,0.90384614,0.8974359,0.9166667,0.9102564,0.8333333,0.90384614,0.8141026,0.6474359,0.9230769,0.9102564,0.84615386,0.9423077,0.9102564,0.94871795,0.90384614,0.9423077,0.98717946,0.94871795,0.9102564,0.8782051,0.99358976,0.96794873,0.9551282,0.9423077,0.9423077,0.974359,0.98717946,0.9807692,0.974359,0.98717946,0.98717946,0.98717946,0.99358976,0.9807692,0.99358976,0.99358976,0.974359,1.0,0.99358976,0.99358976,0.9807692,0.99358976,0.974359,0.96794873,0.99358976,1.0,0.99358976,0.9807692,1.0,1.0
+TabPFN-2.6 (default),0.18589744,0.35897437,0.3525641,0.37820512,,0.42948717,0.55128205,0.4871795,0.60897434,0.5576923,0.65384614,0.65384614,0.7307692,0.6666667,0.6923077,0.6474359,0.724359,0.7307692,0.8141026,0.76282054,0.86538464,0.71794873,0.76282054,0.8269231,0.88461536,0.8397436,0.90384614,0.85897434,0.7948718,0.89102566,0.92948717,0.80128205,0.84615386,0.8525641,0.80128205,0.6730769,0.8717949,0.8333333,0.78846157,0.94871795,0.9230769,0.9551282,0.84615386,0.85897434,0.9551282,0.9102564,0.8717949,0.94871795,0.9807692,0.9358974,0.9230769,0.92948717,0.9807692,0.9358974,0.94871795,0.96794873,0.9358974,0.9358974,0.94871795,0.94871795,1.0,0.9551282,0.96794873,1.0,0.9807692,1.0,0.99358976,0.99358976,0.98717946,0.96794873,0.9551282,0.96794873,1.0,1.0,0.99358976,0.96153843,1.0,1.0
+TabPFN-3 (default),0.19871795,0.41666666,0.40384614,0.44230768,0.57051283,,0.47435898,0.59615386,0.60897434,0.60897434,0.55128205,0.6282051,0.6666667,0.6858974,0.73717946,0.74358976,0.71153843,0.67948717,0.8333333,0.724359,0.82051283,0.775641,0.71153843,0.8333333,0.78846157,0.8333333,0.7948718,0.8525641,0.9358974,0.84615386,0.85897434,0.84615386,0.75641024,0.9102564,0.71794873,0.59615386,0.8397436,0.88461536,0.8269231,0.9166667,0.88461536,0.89102566,0.9166667,0.9102564,0.94871795,0.85897434,0.90384614,0.8397436,0.9807692,0.96794873,0.8974359,0.9230769,0.9423077,0.9102564,0.92948717,0.98717946,0.92948717,0.9807692,0.96794873,0.9358974,1.0,0.94871795,0.99358976,1.0,0.98717946,0.99358976,1.0,0.9807692,0.98717946,0.974359,0.974359,0.96794873,0.99358976,1.0,1.0,0.94871795,1.0,1.0
+RealTabPFN-2.5 (default),0.24358974,0.29807693,0.23717949,0.4871795,0.44871795,0.525641,,0.5448718,0.6858974,0.5769231,0.6602564,0.6730769,0.65384614,0.71153843,0.5897436,0.67948717,0.74358976,0.80128205,0.8269231,0.85897434,0.71794873,0.69871795,0.82051283,0.8525641,0.8333333,0.78205127,0.8269231,0.80128205,0.7692308,0.8269231,0.7692308,0.8269231,0.85897434,0.8141026,0.8269231,0.7307692,0.8525641,0.775641,0.71794873,0.90384614,0.90384614,0.8782051,0.8525641,0.8782051,0.9358974,0.85897434,0.85897434,0.9230769,0.9102564,0.9102564,0.90384614,0.8974359,0.9423077,0.8782051,0.88461536,0.99358976,0.9166667,0.9358974,0.96153843,0.9358974,0.96153843,0.88461536,0.98717946,0.99358976,1.0,0.98717946,0.98717946,0.9807692,0.99358976,0.98717946,0.974359,0.974359,0.99358976,0.99358976,1.0,0.94871795,1.0,1.0
+TabICLv2 (default),0.102564104,0.32692307,0.32051283,0.40384614,0.51282054,0.40384614,0.4551282,,0.5833333,0.5448718,0.5833333,0.60897434,0.71153843,0.6282051,0.6923077,0.75,0.6730769,0.6858974,0.78846157,0.71794873,0.75641024,0.75,0.6923077,0.78846157,0.7692308,0.82051283,0.78205127,0.8397436,0.96153843,0.85897434,0.8782051,0.7948718,0.75,0.8269231,0.75,0.53846157,0.8269231,0.8717949,0.76282054,0.88461536,0.8333333,0.89102566,0.84615386,0.8269231,0.9230769,0.85897434,0.8782051,0.8269231,0.92948717,0.9551282,0.88461536,0.9102564,0.90384614,0.9230769,0.92948717,0.92948717,0.92948717,0.8974359,0.8974359,0.9358974,0.99358976,0.9358974,0.9166667,0.974359,0.98717946,0.9423077,0.9358974,0.9358974,0.9807692,0.974359,0.96794873,0.96794873,0.98717946,0.94871795,0.99358976,0.9423077,1.0,1.0
+LimiX (default),0.20512821,0.30769232,0.30128205,0.25641027,0.39102563,0.39102563,0.31410256,0.41666666,,0.47435898,0.51282054,0.5192308,0.49358973,0.55128205,0.49358973,0.5448718,0.6602564,0.65384614,0.74358976,0.63461536,0.65384614,0.6282051,0.6602564,0.75641024,0.775641,0.78205127,0.7692308,0.80128205,0.6858974,0.8076923,0.76282054,0.84615386,0.76282054,0.89102566,0.724359,0.60897434,0.8333333,0.9102564,0.6282051,0.8397436,0.71794873,0.88461536,0.9102564,0.90384614,0.82051283,0.88461536,0.90384614,0.78205127,0.94871795,0.9423077,0.8269231,0.90384614,0.89102566,0.9166667,0.9423077,0.9807692,0.9423077,0.974359,0.96153843,0.94871795,1.0,0.94871795,0.974359,1.0,0.98717946,0.974359,0.9807692,0.96794873,0.99358976,0.9807692,0.974359,0.96794873,1.0,0.99358976,1.0,0.9551282,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.16025642,0.27564102,0.23076923,0.41666666,0.44230768,0.39102563,0.42307693,0.4551282,0.525641,,0.5576923,0.60897434,0.5576923,0.650641,0.57051283,0.5833333,0.76282054,0.724359,0.7692308,0.6474359,0.67948717,0.64102566,0.73717946,0.7948718,0.78205127,0.73717946,0.7948718,0.8141026,0.65384614,0.8333333,0.76282054,0.8397436,0.8717949,0.74358976,0.8269231,0.6025641,0.8269231,0.76282054,0.65384614,0.84615386,0.8269231,0.8782051,0.78846157,0.84615386,0.84615386,0.8397436,0.7948718,0.88461536,0.86538464,0.8141026,0.8397436,0.86538464,0.90384614,0.8782051,0.8717949,0.9230769,0.9102564,0.8974359,0.9102564,0.8782051,0.84615386,0.8974359,0.9166667,0.9423077,0.98717946,0.88461536,0.92948717,0.92948717,0.99358976,0.974359,0.96794873,0.96794873,0.9423077,0.9423077,0.9358974,0.9423077,1.0,1.0
+RealMLP (tuned + ensembled),0.21794872,0.24358974,0.25,0.2628205,0.34615386,0.44871795,0.33974358,0.41666666,0.4871795,0.44230768,,0.52884614,0.53205127,0.5769231,0.53205127,0.5192308,0.6666667,0.6923077,0.73717946,0.57051283,0.6858974,0.525641,0.69871795,0.7692308,0.84615386,0.73717946,0.8717949,0.8269231,0.5833333,0.8269231,0.75641024,0.8333333,0.86538464,0.74358976,0.8782051,0.6282051,0.86538464,0.8141026,0.5769231,0.8525641,0.75641024,0.8782051,0.8076923,0.85897434,0.74358976,0.8717949,0.85897434,0.9102564,0.9358974,0.89102566,0.8076923,0.8782051,0.9166667,0.88461536,0.89102566,0.99358976,0.9166667,0.9358974,0.974359,0.88461536,0.9358974,0.9230769,0.98717946,1.0,0.99358976,0.9423077,0.974359,0.96794873,0.98717946,0.9807692,0.974359,0.96794873,1.0,1.0,0.99358976,0.9551282,1.0,1.0
+RealMLP (tuned),0.21794872,0.21794872,0.23717949,0.25,0.34615386,0.37179488,0.32692307,0.39102563,0.48076922,0.39102563,0.47115386,,0.46153846,0.5,0.5192308,0.48076922,0.61538464,0.6217949,0.7051282,0.55128205,0.6923077,0.51282054,0.6602564,0.71153843,0.78205127,0.71153843,0.8076923,0.74358976,0.57051283,0.75641024,0.724359,0.775641,0.8525641,0.73717946,0.8141026,0.6025641,0.84615386,0.78846157,0.5833333,0.8333333,0.80128205,0.8397436,0.775641,0.8397436,0.74358976,0.88461536,0.8076923,0.8782051,0.90384614,0.86538464,0.84615386,0.86538464,0.9358974,0.86538464,0.88461536,0.9807692,0.90384614,0.9102564,0.96794873,0.89102566,0.92948717,0.92948717,0.9807692,0.9807692,0.98717946,0.9358974,0.974359,0.96153843,0.99358976,0.98717946,0.974359,0.96153843,1.0,1.0,0.98717946,0.974359,1.0,1.0
+Mitra (default),0.07692308,0.27564102,0.27564102,0.26923078,0.26923078,0.33333334,0.34615386,0.28846154,0.50641024,0.44230768,0.4679487,0.53846157,,0.5897436,0.5192308,0.53205127,0.6474359,0.5897436,0.67948717,0.61538464,0.724359,0.5897436,0.6217949,0.6730769,0.6858974,0.724359,0.71794873,0.80128205,0.63141024,0.78846157,0.8076923,0.7692308,0.75641024,0.7307692,0.7307692,0.5,0.7692308,0.7692308,0.65384614,0.8269231,0.78846157,0.86538464,0.76282054,0.82051283,0.8269231,0.8076923,0.7948718,0.8333333,0.8782051,0.8269231,0.8269231,0.8525641,0.89102566,0.8717949,0.86538464,0.88461536,0.92948717,0.8397436,0.88461536,0.88461536,0.9358974,0.9423077,0.9102564,0.9423077,0.9807692,0.88461536,0.92948717,0.92948717,0.96794873,0.9551282,0.9551282,0.96153843,0.9358974,0.9423077,0.9358974,0.9423077,1.0,1.0
+TabPFNv2 (tuned),0.083333336,0.18589744,0.1923077,0.28846154,0.33333334,0.31410256,0.28846154,0.37179488,0.44871795,0.34935898,0.42307693,0.5,0.41025642,,0.525641,0.55128205,0.65384614,0.5769231,0.64102566,0.57051283,0.6025641,0.6217949,0.6282051,0.71153843,0.64102566,0.6666667,0.6602564,0.7051282,0.61538464,0.71794873,0.6730769,0.724359,0.775641,0.67948717,0.7307692,0.53205127,0.7692308,0.6923077,0.63461536,0.78205127,0.78205127,0.775641,0.724359,0.82051283,0.7948718,0.78205127,0.775641,0.82051283,0.8076923,0.75641024,0.82051283,0.82051283,0.8717949,0.8141026,0.8141026,0.9166667,0.84615386,0.85897434,0.90384614,0.8525641,0.8397436,0.88461536,0.8974359,0.92948717,0.96794873,0.88461536,0.9102564,0.8974359,0.96794873,0.96794873,0.96153843,0.974359,0.9230769,0.9423077,0.9358974,0.9423077,1.0,1.0
+TabDPT-Turbo (default),0.10897436,0.28846154,0.2948718,0.25,0.30769232,0.2628205,0.41025642,0.30769232,0.50641024,0.42948717,0.4679487,0.48076922,0.48076922,0.47435898,,0.55128205,0.51282054,0.5192308,0.60897434,0.59615386,0.67948717,0.6025641,0.53846157,0.59615386,0.61538464,0.6730769,0.61538464,0.6666667,0.6730769,0.61538464,0.8076923,0.6282051,0.5769231,0.7692308,0.5833333,0.525641,0.6474359,0.75641024,0.80128205,0.8141026,0.71794873,0.75641024,0.78205127,0.6923077,0.8333333,0.71794873,0.775641,0.78846157,0.8525641,0.8397436,0.78205127,0.78205127,0.8717949,0.89102566,0.89102566,0.8525641,0.85897434,0.78846157,0.8525641,0.89102566,1.0,0.9230769,0.9423077,1.0,0.8717949,0.9423077,0.88461536,0.88461536,0.85897434,0.9423077,0.9423077,0.94871795,0.9423077,1.0,0.9423077,0.94871795,0.96794873,1.0
+TabDPT (tuned + ensembled),0.09615385,0.25641027,0.2628205,0.17307693,0.3525641,0.25641027,0.32051283,0.25,0.4551282,0.41666666,0.48076922,0.5192308,0.4679487,0.44871795,0.44871795,,0.525641,0.57051283,0.5448718,0.57051283,0.61538464,0.67948717,0.5769231,0.5769231,0.59615386,0.64102566,0.61538464,0.6730769,0.63461536,0.65384614,0.71153843,0.65384614,0.65384614,0.6923077,0.65384614,0.5192308,0.6923077,0.75,0.76282054,0.74358976,0.6730769,0.74358976,0.78846157,0.75641024,0.8333333,0.724359,0.78846157,0.78205127,0.84615386,0.85897434,0.80128205,0.75,0.8269231,0.8397436,0.86538464,0.86538464,0.86538464,0.85897434,0.8974359,0.8782051,0.9807692,0.96794873,0.96153843,1.0,0.86538464,0.88461536,0.9102564,0.9102564,0.8782051,0.96153843,0.96153843,0.9551282,0.9423077,1.0,0.9358974,0.974359,0.9551282,1.0
+"AutoGluon 1.4 (best, 4h)",0.115384616,0.1474359,0.14102565,0.1923077,0.27564102,0.28846154,0.25641027,0.32692307,0.33974358,0.23717949,0.33333334,0.3846154,0.3525641,0.34615386,0.4871795,0.47435898,,0.50641024,0.6025641,0.48076922,0.61538464,0.50641024,0.5448718,0.6730769,0.61538464,0.6474359,0.6474359,0.73717946,0.5897436,0.724359,0.6858974,0.67948717,0.6666667,0.61538464,0.6858974,0.53846157,0.78846157,0.6730769,0.6217949,0.75641024,0.75641024,0.8269231,0.69871795,0.74358976,0.724359,0.8141026,0.7692308,0.8333333,0.84615386,0.8269231,0.8397436,0.8717949,0.90384614,0.88461536,0.89102566,0.8782051,0.86538464,0.80128205,0.9166667,0.92948717,0.90384614,0.8333333,0.9102564,0.92948717,0.94871795,0.9807692,0.9423077,0.9423077,0.96153843,0.9423077,0.94871795,0.9551282,0.9807692,0.9807692,0.96794873,0.9551282,0.99358976,1.0
+TabM (tuned + ensembled),0.17948718,0.1474359,0.1923077,0.21794872,0.26923078,0.32051283,0.19871795,0.31410256,0.34615386,0.27564102,0.30769232,0.37820512,0.41025642,0.42307693,0.48076922,0.42948717,0.49358973,,0.5,0.3974359,0.5769231,0.50641024,0.6442308,0.5641026,0.6474359,0.5769231,0.6730769,0.6217949,0.51282054,0.6282051,0.5897436,0.64102566,0.775641,0.5448718,0.88461536,0.64102566,0.8141026,0.6025641,0.61538464,0.69871795,0.73717946,0.78205127,0.6025641,0.71794873,0.6474359,0.8397436,0.71153843,0.8269231,0.8333333,0.78205127,0.82051283,0.8397436,0.88461536,0.8076923,0.8333333,0.9551282,0.89102566,0.8141026,0.94871795,0.8974359,0.89102566,0.89102566,0.9423077,0.9166667,0.96153843,0.9423077,0.9423077,0.9551282,0.9807692,0.9551282,0.96794873,0.9551282,0.9807692,1.0,0.99358976,0.96794873,1.0,1.0
+CatBoost (tuned + ensembled),0.08974359,0.070512824,0.057692308,0.10897436,0.18589744,0.16666667,0.17307693,0.21153846,0.25641027,0.23076923,0.2628205,0.2948718,0.32051283,0.35897437,0.39102563,0.4551282,0.3974359,0.5,,0.40384614,0.5192308,0.525641,0.5641026,0.69551283,0.61538464,0.55128205,0.6282051,0.59615386,0.5833333,0.6025641,0.53846157,0.6025641,0.6217949,0.65384614,0.6666667,0.53846157,0.7051282,0.6602564,0.59615386,0.6923077,0.6602564,0.8333333,0.73717946,0.775641,0.74358976,0.78846157,0.78205127,0.7051282,0.9230769,0.8333333,0.80128205,0.8782051,0.8333333,0.8782051,0.9166667,0.94871795,0.89102566,0.90384614,0.96153843,0.9230769,0.9230769,0.8717949,0.974359,0.94871795,0.96153843,1.0,0.9807692,0.974359,0.96794873,0.974359,0.974359,0.96794873,0.9807692,1.0,0.99358976,0.9551282,1.0,1.0
+TabPFNv2 (default),0.083333336,0.115384616,0.12179487,0.16666667,0.23717949,0.27564102,0.14102565,0.2820513,0.3653846,0.3525641,0.42948717,0.44871795,0.3846154,0.42948717,0.40384614,0.42948717,0.5192308,0.6025641,0.59615386,,0.4871795,0.50641024,0.6602564,0.6602564,0.63461536,0.5833333,0.6474359,0.64102566,0.55448717,0.63461536,0.5641026,0.69871795,0.73717946,0.6666667,0.71794873,0.49358973,0.74358976,0.6730769,0.53205127,0.65384614,0.6666667,0.75641024,0.6923077,0.7948718,0.73717946,0.7692308,0.75641024,0.71153843,0.78205127,0.7307692,0.724359,0.80128205,0.82051283,0.78846157,0.7948718,0.85897434,0.85897434,0.8397436,0.8717949,0.8333333,0.85897434,0.86538464,0.90384614,0.86538464,0.9423077,0.8397436,0.9166667,0.9102564,0.92948717,0.9807692,0.96794873,0.96794873,0.9102564,0.9166667,0.92948717,0.9358974,0.99358976,1.0
+ModernNCA (tuned + ensembled),0.07692308,0.14102565,0.13461539,0.12820514,0.13461539,0.17948718,0.2820513,0.24358974,0.34615386,0.32051283,0.31410256,0.30769232,0.27564102,0.3974359,0.32051283,0.3846154,0.3846154,0.42307693,0.48076922,0.51282054,,0.44871795,0.4679487,0.51282054,0.53846157,0.6442308,0.55128205,0.59615386,0.5448718,0.60897434,0.525641,0.6602564,0.57051283,0.74358976,0.57051283,0.5576923,0.65384614,0.75641024,0.5769231,0.6923077,0.61538464,0.69871795,0.8141026,0.8076923,0.6730769,0.71794873,0.7692308,0.6858974,0.82051283,0.9102564,0.775641,0.78205127,0.8141026,0.8397436,0.8333333,0.89102566,0.8141026,0.8974359,0.92948717,0.89102566,0.94871795,0.9166667,0.9551282,0.98717946,0.9166667,0.9807692,0.9166667,0.9102564,0.96153843,0.9807692,0.974359,0.9551282,0.94871795,0.99358976,0.96794873,0.9358974,0.99358976,1.0
+TabDPT (tuned),0.10897436,0.21153846,0.24358974,0.13461539,0.2820513,0.22435898,0.30128205,0.25,0.37179488,0.35897437,0.47435898,0.4871795,0.41025642,0.37820512,0.3974359,0.32051283,0.49358973,0.49358973,0.47435898,0.49358973,0.55128205,,0.5448718,0.51282054,0.5641026,0.59615386,0.5897436,0.6025641,0.57051283,0.61538464,0.60897434,0.5833333,0.6025641,0.63461536,0.64102566,0.49358973,0.5897436,0.6602564,0.7307692,0.64102566,0.59615386,0.6602564,0.67948717,0.6923077,0.775641,0.65384614,0.7051282,0.71153843,0.75641024,0.724359,0.6858974,0.6858974,0.78846157,0.8076923,0.8269231,0.76282054,0.8525641,0.8269231,0.86538464,0.86538464,0.9358974,0.96153843,0.96153843,0.9807692,0.82051283,0.8782051,0.9102564,0.9102564,0.8141026,0.9358974,0.94871795,0.9551282,0.9423077,1.0,0.92948717,0.96153843,0.9423077,1.0
+TabM (tuned),0.16666667,0.14102565,0.1923077,0.22435898,0.23717949,0.28846154,0.17948718,0.30769232,0.33974358,0.2628205,0.30128205,0.33974358,0.37820512,0.37179488,0.46153846,0.42307693,0.4551282,0.35576922,0.43589744,0.33974358,0.53205127,0.4551282,,0.49358973,0.5641026,0.5192308,0.5769231,0.5448718,0.5,0.5448718,0.59615386,0.6217949,0.6923077,0.53846157,0.75641024,0.5576923,0.75641024,0.60897434,0.6025641,0.64102566,0.71153843,0.71794873,0.5833333,0.71153843,0.6282051,0.8076923,0.67948717,0.7948718,0.7948718,0.74358976,0.78846157,0.82051283,0.84615386,0.75,0.7692308,0.9166667,0.7948718,0.78846157,0.9358974,0.85897434,0.8782051,0.8525641,0.92948717,0.85897434,0.974359,0.9423077,0.92948717,0.9551282,0.974359,0.96153843,0.9551282,0.94871795,0.974359,1.0,0.98717946,0.96794873,1.0,1.0
+CatBoost (tuned),0.070512824,0.070512824,0.070512824,0.09615385,0.17307693,0.16666667,0.1474359,0.21153846,0.24358974,0.20512821,0.23076923,0.28846154,0.32692307,0.28846154,0.40384614,0.42307693,0.32692307,0.43589744,0.30448717,0.33974358,0.4871795,0.4871795,0.50641024,,0.5769231,0.53846157,0.5769231,0.5448718,0.55128205,0.53846157,0.50641024,0.5576923,0.5833333,0.63461536,0.65384614,0.5192308,0.6923077,0.6282051,0.5833333,0.6666667,0.6730769,0.78205127,0.7051282,0.76282054,0.69871795,0.73717946,0.75641024,0.6923077,0.86538464,0.80128205,0.78205127,0.8525641,0.7948718,0.8525641,0.8525641,0.9102564,0.8525641,0.8782051,0.9358974,0.9230769,0.9166667,0.8525641,0.974359,0.92948717,0.9551282,1.0,0.96794873,0.96153843,0.96794873,0.974359,0.974359,0.96794873,0.974359,1.0,0.99358976,0.9551282,1.0,1.0
+XGBoost (tuned),0.115384616,0.08974359,0.09615385,0.16025642,0.115384616,0.21153846,0.16666667,0.23076923,0.22435898,0.21794872,0.15384616,0.21794872,0.31410256,0.35897437,0.3846154,0.40384614,0.3846154,0.3525641,0.3846154,0.3653846,0.46153846,0.43589744,0.43589744,0.42307693,,0.5192308,0.52884614,0.4871795,0.525641,0.5448718,0.50641024,0.57051283,0.5769231,0.5448718,0.5769231,0.5833333,0.6923077,0.5576923,0.5448718,0.64102566,0.6666667,0.76282054,0.64102566,0.69871795,0.59615386,0.73717946,0.69871795,0.6858974,0.90384614,0.7948718,0.74358976,0.8525641,0.75,0.89102566,0.89102566,0.8525641,0.88461536,0.7948718,0.9230769,0.92948717,0.9166667,0.8525641,0.96153843,0.92948717,0.94871795,0.99358976,0.9551282,0.94871795,0.96153843,0.96153843,0.96153843,0.96794873,0.98717946,0.99358976,0.98717946,0.9358974,1.0,1.0
+ModernNCA (tuned),0.102564104,0.12820514,0.115384616,0.10897436,0.16025642,0.16666667,0.21794872,0.17948718,0.21794872,0.2628205,0.2628205,0.28846154,0.27564102,0.33333334,0.32692307,0.35897437,0.3525641,0.42307693,0.44871795,0.41666666,0.35576922,0.40384614,0.48076922,0.46153846,0.48076922,,0.5,0.5448718,0.46153846,0.5448718,0.43589744,0.60897434,0.53205127,0.6666667,0.57051283,0.5641026,0.63461536,0.6730769,0.57051283,0.6217949,0.59615386,0.6602564,0.7307692,0.73717946,0.69871795,0.69871795,0.724359,0.64102566,0.75641024,0.82051283,0.7692308,0.7307692,0.75,0.78205127,0.78846157,0.82051283,0.8076923,0.86538464,0.9166667,0.9166667,0.9230769,0.9230769,0.96153843,0.96794873,0.89102566,0.974359,0.92948717,0.9358974,0.9358974,0.96794873,0.96794873,0.94871795,0.9423077,0.98717946,0.96794873,0.974359,0.99358976,1.0
+XGBoost (tuned + ensembled),0.10897436,0.08974359,0.07692308,0.13461539,0.09615385,0.20512821,0.17307693,0.21794872,0.23076923,0.20512821,0.12820514,0.1923077,0.2820513,0.33974358,0.3846154,0.3846154,0.3525641,0.32692307,0.37179488,0.3525641,0.44871795,0.41025642,0.42307693,0.42307693,0.47115386,0.5,,0.4679487,0.53205127,0.5192308,0.5192308,0.5641026,0.5769231,0.525641,0.5641026,0.57051283,0.71794873,0.5641026,0.53846157,0.6474359,0.6666667,0.78205127,0.63461536,0.69871795,0.60897434,0.76282054,0.6730769,0.6923077,0.92948717,0.7948718,0.75,0.8525641,0.75641024,0.88461536,0.8974359,0.8525641,0.8717949,0.7948718,0.9230769,0.92948717,0.9102564,0.8525641,0.9551282,0.9166667,0.94871795,0.99358976,0.96153843,0.9551282,0.96153843,0.96153843,0.9551282,0.96153843,0.98717946,0.99358976,0.98717946,0.9358974,1.0,1.0
+LightGBM (tuned + ensembled),0.08974359,0.10897436,0.102564104,0.08974359,0.14102565,0.1474359,0.19871795,0.16025642,0.19871795,0.18589744,0.17307693,0.25641027,0.19871795,0.2948718,0.33333334,0.32692307,0.2628205,0.37820512,0.40384614,0.35897437,0.40384614,0.3974359,0.4551282,0.4551282,0.51282054,0.4551282,0.53205127,,0.44230768,0.5032051,0.5192308,0.55128205,0.53846157,0.525641,0.61538464,0.5576923,0.6923077,0.57051283,0.5833333,0.61538464,0.6282051,0.74358976,0.5897436,0.7051282,0.64102566,0.76282054,0.7051282,0.6666667,0.8525641,0.7692308,0.74358976,0.82051283,0.7948718,0.94871795,0.98717946,0.8397436,0.9423077,0.7692308,0.89102566,0.9807692,0.90384614,0.85897434,0.9230769,0.94871795,0.89102566,1.0,0.94871795,0.94871795,0.92948717,0.9423077,0.9551282,0.9423077,1.0,1.0,0.99358976,0.9551282,1.0,1.0
+TabICL (default),0.025641026,0.19871795,0.19871795,0.09615385,0.20512821,0.06410257,0.23076923,0.03846154,0.31410256,0.34615386,0.41666666,0.42948717,0.36858973,0.3846154,0.32692307,0.3653846,0.41025642,0.4871795,0.41666666,0.44551283,0.4551282,0.42948717,0.5,0.44871795,0.47435898,0.53846157,0.4679487,0.5576923,,0.55128205,0.5641026,0.525641,0.53846157,0.69871795,0.5641026,0.44230768,0.5192308,0.7051282,0.6282051,0.6666667,0.50641024,0.6217949,0.76282054,0.65384614,0.8141026,0.5833333,0.73717946,0.63461536,0.71794873,0.71794873,0.6923077,0.67948717,0.7051282,0.7692308,0.78846157,0.69871795,0.85897434,0.75641024,0.78846157,0.8717949,0.94551283,0.9358974,0.90384614,0.9358974,0.82051283,0.8974359,0.85897434,0.85897434,0.8333333,0.94871795,0.9423077,0.94871795,0.88461536,0.9423077,0.89102566,0.9423077,1.0,1.0
+LightGBM (tuned),0.09615385,0.083333336,0.09615385,0.102564104,0.10897436,0.15384616,0.17307693,0.14102565,0.1923077,0.16666667,0.17307693,0.24358974,0.21153846,0.2820513,0.3846154,0.34615386,0.27564102,0.37179488,0.3974359,0.3653846,0.39102563,0.3846154,0.4551282,0.46153846,0.4551282,0.4551282,0.48076922,0.49679488,0.44871795,,0.5,0.53205127,0.5192308,0.50641024,0.6025641,0.5448718,0.6666667,0.525641,0.5897436,0.6025641,0.65384614,0.75641024,0.6025641,0.67948717,0.63461536,0.76282054,0.6858974,0.65384614,0.85897434,0.75,0.74358976,0.8269231,0.775641,0.9358974,0.94871795,0.8269231,0.9230769,0.78205127,0.8717949,0.96794873,0.89102566,0.86538464,0.9358974,0.94871795,0.8974359,1.0,0.9551282,0.96153843,0.92948717,0.9423077,0.9551282,0.9423077,1.0,1.0,0.99358976,0.9551282,1.0,1.0
+iLTM (tuned + ensembled),0.07692308,0.115384616,0.10897436,0.083333336,0.070512824,0.14102565,0.23076923,0.12179487,0.23717949,0.23717949,0.24358974,0.27564102,0.1923077,0.32692307,0.1923077,0.28846154,0.31410256,0.41025642,0.46153846,0.43589744,0.47435898,0.39102563,0.40384614,0.49358973,0.49358973,0.5641026,0.48076922,0.48076922,0.43589744,0.5,,0.5897436,0.50641024,0.64102566,0.5192308,0.53846157,0.60897434,0.63461536,0.5576923,0.6923077,0.55128205,0.6602564,0.71794873,0.69871795,0.64102566,0.6666667,0.71794873,0.67948717,0.78205127,0.8076923,0.69871795,0.76282054,0.78846157,0.8782051,0.88461536,0.8269231,0.8397436,0.78846157,0.86538464,0.9358974,0.9230769,0.85897434,0.8717949,0.99358976,0.8269231,0.99358976,0.8717949,0.86538464,0.89102566,0.8974359,0.9166667,0.9230769,0.99358976,1.0,0.974359,0.90384614,0.99358976,1.0
+xRFM (tuned + ensembled),0.07692308,0.102564104,0.115384616,0.08974359,0.19871795,0.15384616,0.17307693,0.20512821,0.15384616,0.16025642,0.16666667,0.22435898,0.23076923,0.27564102,0.37179488,0.34615386,0.32051283,0.35897437,0.3974359,0.30128205,0.33974358,0.41666666,0.37820512,0.44230768,0.42948717,0.39102563,0.43589744,0.44871795,0.47435898,0.4679487,0.41025642,,0.50641024,0.49358973,0.51282054,0.4679487,0.48076922,0.5641026,0.53846157,0.59615386,0.5576923,0.6217949,0.6025641,0.6923077,0.6858974,0.60897434,0.6602564,0.5897436,0.7051282,0.63461536,0.65384614,0.71153843,0.67948717,0.8141026,0.8333333,0.724359,0.6666667,0.96153843,0.8525641,0.88461536,0.9102564,0.9166667,0.974359,0.99358976,0.96153843,0.84615386,0.96794873,0.96794873,0.96794873,0.9166667,0.8974359,0.8974359,0.96794873,0.99358976,0.98717946,0.98717946,1.0,1.0
+BetaTabPFN (default),0.083333336,0.09615385,0.083333336,0.16666667,0.15384616,0.24358974,0.14102565,0.25,0.23717949,0.12820514,0.13461539,0.1474359,0.24358974,0.22435898,0.42307693,0.34615386,0.33333334,0.22435898,0.37820512,0.2628205,0.42948717,0.3974359,0.30769232,0.41666666,0.42307693,0.4679487,0.42307693,0.46153846,0.46153846,0.48076922,0.49358973,0.49358973,,0.47435898,0.48076922,0.41666666,0.6602564,0.53205127,0.53846157,0.5576923,0.6730769,0.63461536,0.53846157,0.5833333,0.53846157,0.76282054,0.5833333,0.7307692,0.69871795,0.6730769,0.73717946,0.775641,0.78846157,0.7051282,0.7307692,0.86538464,0.6858974,0.7307692,0.88461536,0.84615386,0.775641,0.8076923,0.86538464,0.8269231,0.9423077,0.9358974,0.9358974,0.9358974,0.9423077,0.9358974,0.96153843,0.9423077,0.99358976,1.0,0.98717946,0.974359,1.0,1.0
+ExtraTrees (tuned + ensembled),0.070512824,0.12820514,0.1474359,0.09615385,0.1474359,0.08974359,0.18589744,0.17307693,0.10897436,0.25641027,0.25641027,0.2628205,0.26923078,0.32051283,0.23076923,0.30769232,0.3846154,0.4551282,0.34615386,0.33333334,0.25641027,0.3653846,0.46153846,0.3653846,0.4551282,0.33333334,0.47435898,0.47435898,0.30128205,0.49358973,0.35897437,0.50641024,0.525641,,0.5641026,0.50641024,0.5,0.67628205,0.5448718,0.55128205,0.47435898,0.5769231,0.78205127,0.5833333,0.6858974,0.57051283,0.80128205,0.5576923,0.6602564,0.63461536,0.61538464,0.6602564,0.63461536,0.76282054,0.7692308,0.6858974,0.80128205,0.82051283,0.78846157,0.8782051,0.96794873,0.9551282,0.94871795,1.0,0.775641,0.9102564,0.9166667,0.9166667,0.775641,0.92948717,0.9551282,0.9423077,0.90384614,0.9807692,0.9358974,1.0,0.94871795,1.0
+TabM (default),0.14102565,0.13461539,0.1474359,0.18589744,0.19871795,0.2820513,0.17307693,0.25,0.27564102,0.17307693,0.12179487,0.18589744,0.26923078,0.26923078,0.41666666,0.34615386,0.31410256,0.115384616,0.33333334,0.2820513,0.42948717,0.35897437,0.24358974,0.34615386,0.42307693,0.42948717,0.43589744,0.3846154,0.43589744,0.3974359,0.48076922,0.4871795,0.5192308,0.43589744,,0.43589744,0.60897434,0.49358973,0.5192308,0.53205127,0.60897434,0.5833333,0.49358973,0.55128205,0.4679487,0.67948717,0.5448718,0.75,0.6666667,0.6217949,0.7307692,0.71794873,0.8141026,0.6923077,0.6923077,0.8525641,0.60897434,0.7051282,0.89102566,0.7692308,0.76282054,0.775641,0.8397436,0.8141026,0.9166667,0.92948717,0.92948717,0.9358974,0.9358974,0.9423077,0.9358974,0.92948717,0.9807692,1.0,0.9807692,0.96153843,1.0,1.0
+SAP-RPT-OSS (default),0.23076923,0.23717949,0.24358974,0.3525641,0.32692307,0.40384614,0.26923078,0.46153846,0.39102563,0.3974359,0.37179488,0.3974359,0.5,0.4679487,0.47435898,0.48076922,0.46153846,0.35897437,0.46153846,0.50641024,0.44230768,0.50641024,0.44230768,0.48076922,0.41666666,0.43589744,0.42948717,0.44230768,0.5576923,0.4551282,0.46153846,0.53205127,0.5833333,0.49358973,0.5641026,,0.5448718,0.5,0.525641,0.5,0.6474359,0.51282054,0.49358973,0.59615386,0.525641,0.57051283,0.5192308,0.6730769,0.50641024,0.5448718,0.6602564,0.59615386,0.73717946,0.5,0.48076922,0.75641024,0.61538464,0.60897434,0.7307692,0.6025641,0.65384614,0.5897436,0.6730769,0.65384614,0.8141026,0.67948717,0.75,0.73717946,0.8141026,0.7692308,0.76282054,0.76282054,0.80128205,0.76282054,0.94871795,0.775641,0.82051283,1.0
+EBM (tuned + ensembled),0.08974359,0.044871796,0.06410257,0.07692308,0.12820514,0.16025642,0.1474359,0.17307693,0.16666667,0.17307693,0.13461539,0.15384616,0.23076923,0.23076923,0.3525641,0.30769232,0.21153846,0.18589744,0.2948718,0.25641027,0.34615386,0.41025642,0.24358974,0.30769232,0.30769232,0.3653846,0.2820513,0.30769232,0.48076922,0.33333334,0.39102563,0.5192308,0.33974358,0.5,0.39102563,0.4551282,,0.51282054,0.5576923,0.53846157,0.64102566,0.50641024,0.53846157,0.6217949,0.5448718,0.7532051,0.5576923,0.5833333,0.6217949,0.64102566,0.7051282,0.8525641,0.6923077,0.71153843,0.724359,0.75,0.61538464,0.75,0.9230769,0.8397436,0.80128205,0.76282054,0.8782051,0.8141026,0.9423077,0.94871795,0.9358974,0.9423077,0.974359,0.9230769,0.9358974,0.9358974,0.9807692,0.99358976,0.98717946,0.9423077,0.99358976,1.0
+RandomForest (tuned + ensembled),0.08974359,0.115384616,0.12179487,0.08974359,0.16666667,0.115384616,0.22435898,0.12820514,0.08974359,0.23717949,0.18589744,0.21153846,0.23076923,0.30769232,0.24358974,0.25,0.32692307,0.3974359,0.33974358,0.32692307,0.24358974,0.33974358,0.39102563,0.37179488,0.44230768,0.32692307,0.43589744,0.42948717,0.2948718,0.47435898,0.3653846,0.43589744,0.4679487,0.32371795,0.50641024,0.5,0.4871795,,0.51282054,0.46153846,0.4551282,0.5448718,0.5769231,0.55128205,0.6217949,0.55128205,0.80448717,0.50641024,0.65384614,0.5576923,0.5576923,0.6666667,0.5448718,0.78205127,0.775641,0.61538464,0.74358976,0.76282054,0.74358976,0.85897434,0.98717946,0.9102564,0.90384614,0.9551282,0.724359,0.8525641,0.92948717,0.92948717,0.7307692,0.9166667,0.9358974,0.9358974,0.89102566,0.94871795,0.9423077,0.99358976,0.9230769,1.0
+TabDPT (default),0.08974359,0.1923077,0.21153846,0.15384616,0.21153846,0.17307693,0.2820513,0.23717949,0.37179488,0.34615386,0.42307693,0.41666666,0.34615386,0.3653846,0.19871795,0.23717949,0.37820512,0.3846154,0.40384614,0.4679487,0.42307693,0.26923078,0.3974359,0.41666666,0.4551282,0.42948717,0.46153846,0.41666666,0.37179488,0.41025642,0.44230768,0.46153846,0.46153846,0.4551282,0.48076922,0.47435898,0.44230768,0.4871795,,0.49358973,0.42948717,0.42948717,0.53846157,0.525641,0.5641026,0.4679487,0.5897436,0.5833333,0.525641,0.55128205,0.49358973,0.47435898,0.63461536,0.6217949,0.59615386,0.59615386,0.75,0.71153843,0.65384614,0.775641,0.80128205,0.85897434,0.82051283,0.9230769,0.6025641,0.86538464,0.775641,0.76282054,0.6666667,0.85897434,0.8782051,0.89102566,0.9102564,1.0,0.8974359,0.88461536,0.9358974,1.0
+iLTM (tuned),0.06410257,0.044871796,0.044871796,0.057692308,0.051282052,0.083333336,0.09615385,0.115384616,0.16025642,0.15384616,0.1474359,0.16666667,0.17307693,0.21794872,0.18589744,0.25641027,0.24358974,0.30128205,0.30769232,0.34615386,0.30769232,0.35897437,0.35897437,0.33333334,0.35897437,0.37820512,0.3525641,0.3846154,0.33333334,0.3974359,0.30769232,0.40384614,0.44230768,0.44871795,0.4679487,0.5,0.46153846,0.53846157,0.50641024,,0.44871795,0.51282054,0.61538464,0.5576923,0.525641,0.5576923,0.6730769,0.6474359,0.63461536,0.6666667,0.6282051,0.63461536,0.6923077,0.7307692,0.6923077,0.74358976,0.724359,0.74358976,0.85897434,0.8397436,0.8717949,0.775641,0.8717949,0.92948717,0.78205127,0.9807692,0.8525641,0.8525641,0.8397436,0.8974359,0.9166667,0.9102564,0.9423077,0.974359,0.94871795,0.89102566,0.974359,1.0
+TorchMLP (tuned + ensembled),0.070512824,0.051282052,0.057692308,0.08974359,0.07692308,0.115384616,0.09615385,0.16666667,0.2820513,0.17307693,0.24358974,0.19871795,0.21153846,0.21794872,0.2820513,0.32692307,0.24358974,0.2628205,0.33974358,0.33333334,0.3846154,0.40384614,0.28846154,0.32692307,0.33333334,0.40384614,0.33333334,0.37179488,0.49358973,0.34615386,0.44871795,0.44230768,0.32692307,0.525641,0.39102563,0.3525641,0.35897437,0.5448718,0.57051283,0.55128205,,0.4871795,0.5833333,0.5448718,0.49358973,0.4551282,0.5576923,0.71153843,0.5833333,0.65384614,0.7916667,0.53846157,0.8076923,0.61538464,0.61538464,0.71153843,0.6602564,0.6730769,0.8974359,0.6602564,0.75641024,0.74358976,0.775641,0.80128205,0.8782051,0.86538464,0.90384614,0.9102564,0.9230769,0.84615386,0.88461536,0.8782051,0.96153843,0.99358976,0.974359,0.9423077,0.99358976,1.0
+CatBoost (default),0.06410257,0.032051284,0.057692308,0.051282052,0.044871796,0.10897436,0.12179487,0.10897436,0.115384616,0.12179487,0.12179487,0.16025642,0.13461539,0.22435898,0.24358974,0.25641027,0.17307693,0.21794872,0.16666667,0.24358974,0.30128205,0.33974358,0.2820513,0.21794872,0.23717949,0.33974358,0.21794872,0.25641027,0.37820512,0.24358974,0.33974358,0.37820512,0.3653846,0.42307693,0.41666666,0.4871795,0.49358973,0.4551282,0.57051283,0.4871795,0.51282054,,0.5192308,0.5833333,0.5192308,0.6282051,0.5769231,0.5897436,0.63461536,0.6282051,0.65384614,0.6923077,0.6858974,0.78846157,0.7948718,0.724359,0.74358976,0.71794873,0.86538464,0.9423077,0.8525641,0.8076923,0.8974359,0.89102566,0.84615386,1.0,0.9230769,0.9166667,0.8717949,0.9230769,0.94871795,0.9230769,0.96794873,1.0,0.99358976,0.94871795,1.0,1.0
+ExtraTrees (tuned),0.07692308,0.102564104,0.12179487,0.09615385,0.15384616,0.083333336,0.1474359,0.15384616,0.08974359,0.21153846,0.1923077,0.22435898,0.23717949,0.27564102,0.21794872,0.21153846,0.30128205,0.3974359,0.2628205,0.30769232,0.18589744,0.32051283,0.41666666,0.2948718,0.35897437,0.26923078,0.3653846,0.41025642,0.23717949,0.3974359,0.2820513,0.3974359,0.46153846,0.21794872,0.50641024,0.50641024,0.46153846,0.42307693,0.46153846,0.3846154,0.41666666,0.48076922,,0.49358973,0.5897436,0.53205127,0.599359,0.5192308,0.5192308,0.50641024,0.51282054,0.57051283,0.5641026,0.64102566,0.6474359,0.5897436,0.75,0.7307692,0.71794873,0.8076923,0.8974359,0.9358974,0.9166667,0.89102566,0.724359,0.80128205,0.9166667,0.9102564,0.7051282,0.9166667,0.9166667,0.9230769,0.89102566,0.96794873,0.9358974,1.0,0.92948717,1.0
+xRFM (tuned),0.070512824,0.07692308,0.07692308,0.057692308,0.14102565,0.08974359,0.12179487,0.17307693,0.09615385,0.15384616,0.14102565,0.16025642,0.17948718,0.17948718,0.30769232,0.24358974,0.25641027,0.2820513,0.22435898,0.20512821,0.1923077,0.30769232,0.28846154,0.23717949,0.30128205,0.2628205,0.30128205,0.2948718,0.34615386,0.32051283,0.30128205,0.30769232,0.41666666,0.41666666,0.44871795,0.40384614,0.37820512,0.44871795,0.47435898,0.44230768,0.4551282,0.41666666,0.50641024,,0.5897436,0.48076922,0.55128205,0.525641,0.53205127,0.5,0.5576923,0.5833333,0.5641026,0.69871795,0.6730769,0.5833333,0.59615386,0.9230769,0.78205127,0.8333333,0.86538464,0.8782051,0.9423077,0.94871795,0.8782051,0.8076923,0.92948717,0.9358974,0.88461536,0.8974359,0.9230769,0.9102564,0.92948717,0.99358976,0.9807692,0.974359,0.99358976,1.0
+TabSwift (default),0.07692308,0.025641026,0.025641026,0.012820513,0.044871796,0.051282052,0.06410257,0.07692308,0.17948718,0.15384616,0.25641027,0.25641027,0.17307693,0.20512821,0.16666667,0.16666667,0.27564102,0.3525641,0.25641027,0.2628205,0.32692307,0.22435898,0.37179488,0.30128205,0.40384614,0.30128205,0.39102563,0.35897437,0.18589744,0.3653846,0.35897437,0.31410256,0.46153846,0.31410256,0.53205127,0.47435898,0.4551282,0.37820512,0.43589744,0.47435898,0.50641024,0.48076922,0.41025642,0.41025642,,0.51282054,0.53205127,0.6602564,0.51282054,0.55128205,0.61538464,0.5448718,0.67948717,0.5769231,0.57051283,0.67948717,0.76282054,0.6217949,0.69871795,0.76282054,0.86538464,0.8525641,0.92948717,0.775641,0.67948717,0.8076923,0.8782051,0.86538464,0.71153843,0.9358974,0.94871795,0.94871795,0.8782051,0.9551282,0.9358974,0.9551282,0.96794873,1.0
+EBM (tuned),0.070512824,0.03846154,0.057692308,0.051282052,0.08974359,0.14102565,0.14102565,0.14102565,0.115384616,0.16025642,0.12820514,0.115384616,0.1923077,0.21794872,0.2820513,0.27564102,0.18589744,0.16025642,0.21153846,0.23076923,0.2820513,0.34615386,0.1923077,0.2628205,0.2628205,0.30128205,0.23717949,0.23717949,0.41666666,0.23717949,0.33333334,0.39102563,0.23717949,0.42948717,0.32051283,0.42948717,0.24679486,0.44871795,0.53205127,0.44230768,0.5448718,0.37179488,0.4679487,0.5192308,0.4871795,,0.5192308,0.53846157,0.53846157,0.5448718,0.65384614,0.71794873,0.6474359,0.65384614,0.67948717,0.6666667,0.525641,0.7051282,0.88461536,0.7948718,0.7692308,0.71153843,0.8269231,0.78846157,0.8974359,0.9102564,0.9230769,0.92948717,0.92948717,0.90384614,0.9166667,0.9102564,0.96794873,0.99358976,0.9807692,0.9423077,0.99358976,1.0
+RandomForest (tuned),0.08974359,0.08974359,0.08974359,0.08974359,0.12820514,0.09615385,0.14102565,0.12179487,0.09615385,0.20512821,0.14102565,0.1923077,0.20512821,0.22435898,0.22435898,0.21153846,0.23076923,0.28846154,0.21794872,0.24358974,0.23076923,0.2948718,0.32051283,0.24358974,0.30128205,0.27564102,0.32692307,0.2948718,0.2628205,0.31410256,0.2820513,0.33974358,0.41666666,0.19871795,0.4551282,0.48076922,0.44230768,0.19551282,0.41025642,0.32692307,0.44230768,0.42307693,0.40064102,0.44871795,0.4679487,0.48076922,,0.49358973,0.5,0.40384614,0.49358973,0.60897434,0.53846157,0.6217949,0.6282051,0.5448718,0.6666667,0.64102566,0.67948717,0.8076923,0.8301282,0.8525641,0.86538464,0.80128205,0.6923077,0.8076923,0.88461536,0.8782051,0.6858974,0.8974359,0.9230769,0.9166667,0.8782051,0.92948717,0.9230769,0.974359,0.90384614,1.0
+FastaiMLP (tuned + ensembled),0.09615385,0.070512824,0.06410257,0.12179487,0.051282052,0.16025642,0.07692308,0.17307693,0.21794872,0.115384616,0.08974359,0.12179487,0.16666667,0.17948718,0.21153846,0.21794872,0.16666667,0.17307693,0.2948718,0.28846154,0.31410256,0.28846154,0.20512821,0.30769232,0.31410256,0.35897437,0.30769232,0.33333334,0.3653846,0.34615386,0.32051283,0.41025642,0.26923078,0.44230768,0.25,0.32692307,0.41666666,0.49358973,0.41666666,0.3525641,0.28846154,0.41025642,0.48076922,0.47435898,0.33974358,0.46153846,0.50641024,,0.525641,0.5641026,0.47435898,0.48076922,0.71794873,0.6474359,0.63461536,0.5576923,0.55128205,0.5833333,0.63461536,0.6858974,0.60897434,0.65384614,0.59615386,0.7948718,0.82051283,0.74358976,0.724359,0.724359,0.86538464,0.6858974,0.73717946,0.73717946,0.99358976,0.9230769,0.9358974,0.8974359,1.0,0.99358976
+XGBoost (default),0.06410257,0.01923077,0.012820513,0.0064102565,0.01923077,0.01923077,0.08974359,0.070512824,0.051282052,0.13461539,0.06410257,0.09615385,0.12179487,0.1923077,0.1474359,0.15384616,0.15384616,0.16666667,0.07692308,0.21794872,0.17948718,0.24358974,0.20512821,0.13461539,0.09615385,0.24358974,0.070512824,0.1474359,0.2820513,0.14102565,0.21794872,0.2948718,0.30128205,0.33974358,0.33333334,0.49358973,0.37820512,0.34615386,0.47435898,0.3653846,0.41666666,0.3653846,0.48076922,0.4679487,0.4871795,0.46153846,0.5,0.47435898,,0.525641,0.53205127,0.63461536,0.55128205,0.6666667,0.69871795,0.6025641,0.69871795,0.6923077,0.7692308,0.8717949,0.8717949,0.8141026,0.89102566,0.8717949,0.73717946,0.96153843,0.88461536,0.8717949,0.775641,0.8974359,0.92948717,0.9230769,0.94871795,0.98717946,0.98717946,0.92948717,1.0,1.0
+ModernNCA (default),0.01923077,0.051282052,0.06410257,0.032051284,0.06410257,0.032051284,0.08974359,0.044871796,0.057692308,0.18589744,0.10897436,0.13461539,0.17307693,0.24358974,0.16025642,0.14102565,0.17307693,0.21794872,0.16666667,0.26923078,0.08974359,0.27564102,0.25641027,0.19871795,0.20512821,0.17948718,0.20512821,0.23076923,0.2820513,0.25,0.1923077,0.3653846,0.32692307,0.3653846,0.37820512,0.4551282,0.35897437,0.44230768,0.44871795,0.33333334,0.34615386,0.37179488,0.49358973,0.5,0.44871795,0.4551282,0.59615386,0.43589744,0.47435898,,0.4551282,0.53846157,0.51282054,0.69871795,0.71794873,0.53205127,0.6730769,0.6858974,0.74358976,0.80128205,0.85897434,0.8269231,0.8269231,0.84615386,0.78205127,0.78846157,0.7948718,0.7948718,0.82051283,0.88461536,0.9102564,0.90384614,0.86538464,0.92948717,0.8717949,0.90384614,0.9807692,1.0
+TorchMLP (tuned),0.06410257,0.051282052,0.051282052,0.044871796,0.07692308,0.102564104,0.09615385,0.115384616,0.17307693,0.16025642,0.1923077,0.15384616,0.17307693,0.17948718,0.21794872,0.19871795,0.16025642,0.17948718,0.19871795,0.27564102,0.22435898,0.31410256,0.21153846,0.21794872,0.25641027,0.23076923,0.25,0.25641027,0.30769232,0.25641027,0.30128205,0.34615386,0.2628205,0.3846154,0.26923078,0.33974358,0.2948718,0.44230768,0.50641024,0.37179488,0.20833333,0.34615386,0.4871795,0.44230768,0.3846154,0.34615386,0.50641024,0.525641,0.4679487,0.5448718,,0.35897437,0.6602564,0.525641,0.525641,0.53846157,0.53846157,0.6217949,0.82051283,0.64102566,0.724359,0.7051282,0.69871795,0.75,0.75,0.8141026,0.8717949,0.86538464,0.80128205,0.7948718,0.8397436,0.8269231,0.92948717,0.96794873,0.9551282,0.9423077,0.98717946,1.0
+EBM (default),0.070512824,0.051282052,0.06410257,0.057692308,0.070512824,0.07692308,0.102564104,0.08974359,0.09615385,0.13461539,0.12179487,0.13461539,0.1474359,0.17948718,0.21794872,0.25,0.12820514,0.16025642,0.12179487,0.19871795,0.21794872,0.31410256,0.17948718,0.1474359,0.1474359,0.26923078,0.1474359,0.17948718,0.32051283,0.17307693,0.23717949,0.28846154,0.22435898,0.33974358,0.2820513,0.40384614,0.1474359,0.33333334,0.525641,0.3653846,0.46153846,0.30769232,0.42948717,0.41666666,0.4551282,0.2820513,0.39102563,0.5192308,0.3653846,0.46153846,0.64102566,,0.59615386,0.6025641,0.61538464,0.55128205,0.50641024,0.64102566,0.78205127,0.76282054,0.73717946,0.7692308,0.80128205,0.7692308,0.7948718,0.86538464,0.9102564,0.9166667,0.86538464,0.88461536,0.90384614,0.8782051,0.96794873,0.9807692,0.974359,0.974359,1.0,1.0
+FastaiMLP (tuned),0.057692308,0.032051284,0.044871796,0.057692308,0.01923077,0.057692308,0.057692308,0.09615385,0.10897436,0.09615385,0.083333336,0.06410257,0.10897436,0.12820514,0.12820514,0.17307693,0.09615385,0.115384616,0.16666667,0.17948718,0.18589744,0.21153846,0.15384616,0.20512821,0.25,0.25,0.24358974,0.20512821,0.2948718,0.22435898,0.21153846,0.32051283,0.21153846,0.3653846,0.18589744,0.2628205,0.30769232,0.4551282,0.3653846,0.30769232,0.1923077,0.31410256,0.43589744,0.43589744,0.32051283,0.3525641,0.46153846,0.2820513,0.44871795,0.4871795,0.33974358,0.40384614,,0.5897436,0.5641026,0.4551282,0.5,0.5897436,0.60897434,0.6602564,0.59615386,0.67948717,0.5833333,0.7692308,0.76282054,0.74358976,0.6923077,0.7051282,0.8076923,0.6666667,0.71794873,0.724359,0.98717946,0.8397436,0.9358974,0.8974359,1.0,0.99358976
+ChimeraBoost (tuned),0.070512824,0.044871796,0.057692308,0.025641026,0.06410257,0.08974359,0.12179487,0.07692308,0.083333336,0.12179487,0.115384616,0.13461539,0.12820514,0.18589744,0.10897436,0.16025642,0.115384616,0.1923077,0.12179487,0.21153846,0.16025642,0.1923077,0.25,0.1474359,0.10897436,0.21794872,0.115384616,0.051282052,0.23076923,0.06410257,0.12179487,0.18589744,0.2948718,0.23717949,0.30769232,0.5,0.28846154,0.21794872,0.37820512,0.26923078,0.3846154,0.21153846,0.35897437,0.30128205,0.42307693,0.34615386,0.37820512,0.3525641,0.33333334,0.30128205,0.47435898,0.3974359,0.41025642,,0.47115386,0.4871795,0.55128205,0.6602564,0.6217949,0.92948717,0.64102566,0.65384614,0.75641024,0.7051282,0.69871795,0.8333333,0.7948718,0.80128205,0.71794873,0.7307692,0.78846157,0.775641,0.94871795,0.9358974,0.9166667,0.86538464,0.974359,1.0
+ChimeraBoost (tuned + ensembled),0.070512824,0.032051284,0.044871796,0.012820513,0.051282052,0.070512824,0.115384616,0.070512824,0.057692308,0.12820514,0.10897436,0.115384616,0.13461539,0.18589744,0.10897436,0.13461539,0.10897436,0.16666667,0.083333336,0.20512821,0.16666667,0.17307693,0.23076923,0.1474359,0.10897436,0.21153846,0.102564104,0.012820513,0.21153846,0.051282052,0.115384616,0.16666667,0.26923078,0.23076923,0.30769232,0.5192308,0.27564102,0.22435898,0.40384614,0.30769232,0.3846154,0.20512821,0.3525641,0.32692307,0.42948717,0.32051283,0.37179488,0.3653846,0.30128205,0.2820513,0.47435898,0.3846154,0.43589744,0.52884614,,0.4679487,0.53846157,0.67948717,0.61538464,0.92948717,0.64102566,0.65384614,0.76282054,0.71794873,0.6858974,0.8397436,0.80128205,0.8076923,0.724359,0.7307692,0.8076923,0.78846157,0.96794873,0.96794873,0.9423077,0.86538464,0.9807692,1.0
+RealMLP (default),0.01923077,0.0064102565,0.012820513,0.01923077,0.032051284,0.012820513,0.0064102565,0.070512824,0.01923077,0.07692308,0.0064102565,0.01923077,0.115384616,0.083333336,0.1474359,0.13461539,0.12179487,0.044871796,0.051282052,0.14102565,0.10897436,0.23717949,0.083333336,0.08974359,0.1474359,0.17948718,0.1474359,0.16025642,0.30128205,0.17307693,0.17307693,0.27564102,0.13461539,0.31410256,0.1474359,0.24358974,0.25,0.3846154,0.40384614,0.25641027,0.28846154,0.27564102,0.41025642,0.41666666,0.32051283,0.33333334,0.4551282,0.44230768,0.3974359,0.4679487,0.46153846,0.44871795,0.5448718,0.51282054,0.53205127,,0.48076922,0.5833333,0.69871795,0.61538464,0.6858974,0.7051282,0.63461536,0.76282054,0.7948718,0.78205127,0.78846157,0.78846157,0.8525641,0.8397436,0.8974359,0.8525641,0.9358974,0.92948717,0.9423077,0.92948717,0.99358976,1.0
+LightGBM (default),0.070512824,0.051282052,0.06410257,0.025641026,0.06410257,0.070512824,0.083333336,0.070512824,0.057692308,0.08974359,0.083333336,0.09615385,0.070512824,0.15384616,0.14102565,0.13461539,0.13461539,0.10897436,0.10897436,0.14102565,0.18589744,0.1474359,0.20512821,0.1474359,0.115384616,0.1923077,0.12820514,0.057692308,0.14102565,0.07692308,0.16025642,0.33333334,0.31410256,0.19871795,0.39102563,0.3846154,0.3846154,0.25641027,0.25,0.27564102,0.33974358,0.25641027,0.25,0.40384614,0.23717949,0.47435898,0.33333334,0.44871795,0.30128205,0.32692307,0.46153846,0.49358973,0.5,0.44871795,0.46153846,0.5192308,,0.53846157,0.6666667,0.71794873,0.6282051,0.775641,0.75641024,0.6474359,0.6217949,0.775641,0.8141026,0.78846157,0.6666667,0.8782051,0.9166667,0.88461536,0.89102566,0.89102566,0.9423077,0.92948717,0.9807692,1.0
+xRFM (default),0.070512824,0.01923077,0.032051284,0.012820513,0.06410257,0.01923077,0.06410257,0.102564104,0.025641026,0.102564104,0.06410257,0.08974359,0.16025642,0.14102565,0.21153846,0.14102565,0.19871795,0.18589744,0.09615385,0.16025642,0.102564104,0.17307693,0.21153846,0.12179487,0.20512821,0.13461539,0.20512821,0.23076923,0.24358974,0.21794872,0.21153846,0.03846154,0.26923078,0.17948718,0.2948718,0.39102563,0.25,0.23717949,0.28846154,0.25641027,0.32692307,0.2820513,0.26923078,0.07692308,0.37820512,0.2948718,0.35897437,0.41666666,0.30769232,0.31410256,0.37820512,0.35897437,0.41025642,0.33974358,0.32051283,0.41666666,0.46153846,,0.4679487,0.525641,0.6730769,0.63461536,0.6217949,0.5,0.6282051,0.55128205,0.76282054,0.75641024,0.6474359,0.7051282,0.74358976,0.76282054,0.6858974,0.74358976,0.9230769,0.9230769,0.8974359,1.0
+TorchMLP (default),0.06410257,0.0064102565,0.012820513,0.012820513,0.051282052,0.032051284,0.03846154,0.102564104,0.03846154,0.08974359,0.025641026,0.032051284,0.115384616,0.09615385,0.1474359,0.102564104,0.083333336,0.051282052,0.03846154,0.12820514,0.070512824,0.13461539,0.06410257,0.06410257,0.07692308,0.083333336,0.07692308,0.10897436,0.21153846,0.12820514,0.13461539,0.1474359,0.115384616,0.21153846,0.10897436,0.26923078,0.07692308,0.25641027,0.34615386,0.14102565,0.102564104,0.13461539,0.2820513,0.21794872,0.30128205,0.115384616,0.32051283,0.3653846,0.23076923,0.25641027,0.17948718,0.21794872,0.39102563,0.37820512,0.3846154,0.30128205,0.33333334,0.53205127,,0.525641,0.5769231,0.64102566,0.61538464,0.6602564,0.6858974,0.49358973,0.7692308,0.775641,0.7307692,0.6474359,0.71153843,0.6858974,0.85897434,0.8141026,0.8974359,0.92948717,0.9423077,1.0
+ChimeraBoost (default),0.057692308,0.03846154,0.057692308,0.012820513,0.051282052,0.06410257,0.06410257,0.06410257,0.051282052,0.12179487,0.115384616,0.10897436,0.115384616,0.1474359,0.10897436,0.12179487,0.070512824,0.102564104,0.07692308,0.16666667,0.10897436,0.13461539,0.14102565,0.07692308,0.070512824,0.083333336,0.070512824,0.01923077,0.12820514,0.032051284,0.06410257,0.115384616,0.15384616,0.12179487,0.23076923,0.3974359,0.16025642,0.14102565,0.22435898,0.16025642,0.33974358,0.057692308,0.1923077,0.16666667,0.23717949,0.20512821,0.1923077,0.31410256,0.12820514,0.19871795,0.35897437,0.23717949,0.33974358,0.070512824,0.070512824,0.3846154,0.2820513,0.47435898,0.47435898,,0.5,0.46153846,0.5576923,0.31410256,0.60897434,0.64102566,0.71794873,0.73717946,0.61538464,0.6474359,0.69871795,0.67948717,0.75,0.6666667,0.82051283,0.8525641,0.9102564,1.0
+RandomForest (default),0.0064102565,0.012820513,0.012820513,0.0064102565,0.0,0.0,0.03846154,0.0064102565,0.0,0.15384616,0.06410257,0.070512824,0.06410257,0.16025642,0.0,0.01923077,0.09615385,0.10897436,0.07692308,0.14102565,0.051282052,0.06410257,0.12179487,0.083333336,0.083333336,0.07692308,0.08974359,0.09615385,0.05448718,0.10897436,0.07692308,0.08974359,0.22435898,0.032051284,0.23717949,0.34615386,0.19871795,0.012820513,0.19871795,0.12820514,0.24358974,0.1474359,0.102564104,0.13461539,0.13461539,0.23076923,0.16987179,0.39102563,0.12820514,0.14102565,0.27564102,0.2628205,0.40384614,0.35897437,0.35897437,0.31410256,0.37179488,0.32692307,0.42307693,0.5,,0.60897434,0.4679487,0.525641,0.525641,0.42948717,0.57051283,0.57051283,0.5833333,0.71153843,0.74358976,0.71794873,0.67948717,0.6602564,0.8333333,0.89102566,0.8717949,1.0
+KNN (tuned + ensembled),0.06410257,0.06410257,0.070512824,0.01923077,0.044871796,0.051282052,0.115384616,0.06410257,0.051282052,0.102564104,0.07692308,0.070512824,0.057692308,0.115384616,0.07692308,0.032051284,0.16666667,0.10897436,0.12820514,0.13461539,0.083333336,0.03846154,0.1474359,0.1474359,0.1474359,0.07692308,0.1474359,0.14102565,0.06410257,0.13461539,0.14102565,0.083333336,0.1923077,0.044871796,0.22435898,0.41025642,0.23717949,0.08974359,0.14102565,0.22435898,0.25641027,0.1923077,0.06410257,0.12179487,0.1474359,0.28846154,0.1474359,0.34615386,0.18589744,0.17307693,0.2948718,0.23076923,0.32051283,0.34615386,0.34615386,0.2948718,0.22435898,0.3653846,0.35897437,0.53846157,0.39102563,,0.42948717,0.41666666,0.46153846,0.4551282,0.4551282,0.4679487,0.4679487,0.6474359,0.6858974,0.69871795,0.59615386,0.5641026,0.64102566,0.9358974,0.7051282,0.99358976
+OrionMSP (default),0.06410257,0.01923077,0.012820513,0.0064102565,0.032051284,0.0064102565,0.012820513,0.083333336,0.025641026,0.083333336,0.012820513,0.01923077,0.08974359,0.102564104,0.057692308,0.03846154,0.08974359,0.057692308,0.025641026,0.09615385,0.044871796,0.03846154,0.070512824,0.025641026,0.03846154,0.03846154,0.044871796,0.07692308,0.09615385,0.06410257,0.12820514,0.025641026,0.13461539,0.051282052,0.16025642,0.32692307,0.12179487,0.09615385,0.17948718,0.12820514,0.22435898,0.102564104,0.083333336,0.057692308,0.070512824,0.17307693,0.13461539,0.40384614,0.10897436,0.17307693,0.30128205,0.19871795,0.41666666,0.24358974,0.23717949,0.3653846,0.24358974,0.37820512,0.3846154,0.44230768,0.53205127,0.57051283,,0.47435898,0.57051283,0.4871795,0.6730769,0.65384614,0.5897436,0.67948717,0.7692308,0.75,0.75,0.76282054,0.8974359,0.9166667,0.8974359,1.0
+ExtraTrees (default),0.0,0.0064102565,0.0064102565,0.0064102565,0.0,0.0,0.0064102565,0.025641026,0.0,0.057692308,0.0,0.01923077,0.057692308,0.070512824,0.0,0.0,0.070512824,0.083333336,0.051282052,0.13461539,0.012820513,0.01923077,0.14102565,0.070512824,0.070512824,0.032051284,0.083333336,0.051282052,0.06410257,0.051282052,0.0064102565,0.0064102565,0.17307693,0.0,0.18589744,0.34615386,0.18589744,0.044871796,0.07692308,0.070512824,0.19871795,0.10897436,0.10897436,0.051282052,0.22435898,0.21153846,0.19871795,0.20512821,0.12820514,0.15384616,0.25,0.23076923,0.23076923,0.2948718,0.2820513,0.23717949,0.3525641,0.5,0.33974358,0.6858974,0.47435898,0.5833333,0.525641,,0.5576923,0.5576923,0.5769231,0.5833333,0.5576923,0.61538464,0.71794873,0.71153843,0.78205127,0.71794873,0.7307692,0.84615386,0.85897434,1.0
+Linear (tuned + ensembled),0.0064102565,0.0064102565,0.0064102565,0.025641026,0.01923077,0.012820513,0.0,0.012820513,0.012820513,0.012820513,0.0064102565,0.012820513,0.01923077,0.032051284,0.12820514,0.13461539,0.051282052,0.03846154,0.03846154,0.057692308,0.083333336,0.17948718,0.025641026,0.044871796,0.051282052,0.10897436,0.051282052,0.10897436,0.17948718,0.102564104,0.17307693,0.03846154,0.057692308,0.22435898,0.083333336,0.18589744,0.057692308,0.27564102,0.3974359,0.21794872,0.12179487,0.15384616,0.27564102,0.12179487,0.32051283,0.102564104,0.30769232,0.17948718,0.2628205,0.21794872,0.25,0.20512821,0.23717949,0.30128205,0.31410256,0.20512821,0.37820512,0.37179488,0.31410256,0.39102563,0.47435898,0.53846157,0.42948717,0.44230768,,0.42948717,0.42948717,0.4551282,0.75961536,0.4551282,0.47435898,0.5448718,0.44871795,0.5576923,0.6282051,0.7692308,0.94871795,0.9358974
+iLTM (default),0.057692308,0.0,0.0,0.0,0.0,0.0064102565,0.012820513,0.057692308,0.025641026,0.115384616,0.057692308,0.06410257,0.115384616,0.115384616,0.057692308,0.115384616,0.01923077,0.057692308,0.0,0.16025642,0.01923077,0.12179487,0.057692308,0.0,0.0064102565,0.025641026,0.0064102565,0.0,0.102564104,0.0,0.0064102565,0.15384616,0.06410257,0.08974359,0.070512824,0.32051283,0.051282052,0.1474359,0.13461539,0.01923077,0.13461539,0.0,0.19871795,0.1923077,0.1923077,0.08974359,0.1923077,0.25641027,0.03846154,0.21153846,0.18589744,0.13461539,0.25641027,0.16666667,0.16025642,0.21794872,0.22435898,0.44871795,0.50641024,0.35897437,0.57051283,0.5448718,0.51282054,0.44230768,0.57051283,,0.7051282,0.69871795,0.5769231,0.6730769,0.69871795,0.69871795,0.67948717,0.71153843,0.78846157,0.7692308,0.78205127,0.9358974
+TabSTAR (tuned),0.057692308,0.012820513,0.0064102565,0.0064102565,0.0064102565,0.0,0.012820513,0.06410257,0.01923077,0.070512824,0.025641026,0.025641026,0.070512824,0.08974359,0.115384616,0.08974359,0.057692308,0.057692308,0.01923077,0.083333336,0.083333336,0.08974359,0.070512824,0.032051284,0.044871796,0.070512824,0.03846154,0.051282052,0.14102565,0.044871796,0.12820514,0.032051284,0.06410257,0.083333336,0.070512824,0.25,0.06410257,0.070512824,0.22435898,0.1474359,0.09615385,0.07692308,0.083333336,0.070512824,0.12179487,0.07692308,0.115384616,0.27564102,0.115384616,0.20512821,0.12820514,0.08974359,0.30769232,0.20512821,0.19871795,0.21153846,0.18589744,0.23717949,0.23076923,0.2820513,0.42948717,0.5448718,0.32692307,0.42307693,0.57051283,0.2948718,,0.5416667,0.6217949,0.53846157,0.5769231,0.5576923,0.60897434,0.46153846,1.0,0.85897434,0.9102564,0.98717946
+TabSTAR (tuned + ensembled),0.06410257,0.012820513,0.0064102565,0.0064102565,0.0064102565,0.01923077,0.01923077,0.06410257,0.032051284,0.070512824,0.032051284,0.03846154,0.070512824,0.102564104,0.115384616,0.08974359,0.057692308,0.044871796,0.025641026,0.08974359,0.08974359,0.08974359,0.044871796,0.03846154,0.051282052,0.06410257,0.044871796,0.051282052,0.14102565,0.03846154,0.13461539,0.032051284,0.06410257,0.083333336,0.06410257,0.2628205,0.057692308,0.070512824,0.23717949,0.1474359,0.08974359,0.083333336,0.08974359,0.06410257,0.13461539,0.070512824,0.12179487,0.27564102,0.12820514,0.20512821,0.13461539,0.083333336,0.2948718,0.19871795,0.1923077,0.21153846,0.21153846,0.24358974,0.22435898,0.2628205,0.42948717,0.53205127,0.34615386,0.41666666,0.5448718,0.30128205,0.45833334,,0.60897434,0.525641,0.5641026,0.5576923,0.59615386,0.4551282,1.0,0.85897434,0.89102566,0.98717946
+Linear (tuned),0.0064102565,0.0,0.0064102565,0.01923077,0.012820513,0.012820513,0.0064102565,0.01923077,0.0064102565,0.0064102565,0.012820513,0.0064102565,0.032051284,0.032051284,0.14102565,0.12179487,0.03846154,0.01923077,0.032051284,0.070512824,0.03846154,0.18589744,0.025641026,0.032051284,0.03846154,0.06410257,0.03846154,0.070512824,0.16666667,0.070512824,0.10897436,0.032051284,0.057692308,0.22435898,0.06410257,0.18589744,0.025641026,0.26923078,0.33333334,0.16025642,0.07692308,0.12820514,0.2948718,0.115384616,0.28846154,0.070512824,0.31410256,0.13461539,0.22435898,0.17948718,0.19871795,0.13461539,0.1923077,0.2820513,0.27564102,0.1474359,0.33333334,0.3525641,0.26923078,0.3846154,0.41666666,0.53205127,0.41025642,0.44230768,0.24038461,0.42307693,0.37820512,0.39102563,,0.42307693,0.4551282,0.5192308,0.41025642,0.55128205,0.61538464,0.75641024,0.85897434,0.9358974
+PerpetualBooster (tuned + ensembled),0.0064102565,0.012820513,0.012820513,0.0064102565,0.032051284,0.025641026,0.012820513,0.025641026,0.01923077,0.025641026,0.01923077,0.012820513,0.044871796,0.032051284,0.057692308,0.03846154,0.057692308,0.044871796,0.025641026,0.01923077,0.01923077,0.06410257,0.03846154,0.025641026,0.03846154,0.032051284,0.03846154,0.057692308,0.051282052,0.057692308,0.102564104,0.083333336,0.06410257,0.070512824,0.057692308,0.23076923,0.07692308,0.083333336,0.14102565,0.102564104,0.15384616,0.07692308,0.083333336,0.102564104,0.06410257,0.09615385,0.102564104,0.31410256,0.102564104,0.115384616,0.20512821,0.115384616,0.33333334,0.26923078,0.26923078,0.16025642,0.12179487,0.2948718,0.3525641,0.3525641,0.28846154,0.3525641,0.32051283,0.3846154,0.5448718,0.32692307,0.46153846,0.47435898,0.5769231,,0.78846157,0.8141026,0.6025641,0.65384614,0.53205127,0.6474359,0.67948717,0.9423077
+PerpetualBooster (tuned),0.025641026,0.025641026,0.025641026,0.025641026,0.044871796,0.025641026,0.025641026,0.032051284,0.025641026,0.032051284,0.025641026,0.025641026,0.044871796,0.03846154,0.057692308,0.03846154,0.051282052,0.032051284,0.025641026,0.032051284,0.025641026,0.051282052,0.044871796,0.025641026,0.03846154,0.032051284,0.044871796,0.044871796,0.057692308,0.044871796,0.083333336,0.102564104,0.03846154,0.044871796,0.06410257,0.23717949,0.06410257,0.06410257,0.12179487,0.083333336,0.115384616,0.051282052,0.083333336,0.07692308,0.051282052,0.083333336,0.07692308,0.2628205,0.070512824,0.08974359,0.16025642,0.09615385,0.2820513,0.21153846,0.1923077,0.102564104,0.083333336,0.25641027,0.28846154,0.30128205,0.25641027,0.31410256,0.23076923,0.2820513,0.525641,0.30128205,0.42307693,0.43589744,0.5448718,0.21153846,,0.6698718,0.57051283,0.53846157,0.47435898,0.64102566,0.65384614,0.9358974
+PerpetualBooster (default),0.032051284,0.032051284,0.025641026,0.032051284,0.032051284,0.032051284,0.025641026,0.032051284,0.032051284,0.032051284,0.032051284,0.03846154,0.03846154,0.025641026,0.051282052,0.044871796,0.044871796,0.044871796,0.032051284,0.032051284,0.044871796,0.044871796,0.051282052,0.032051284,0.032051284,0.051282052,0.03846154,0.057692308,0.051282052,0.057692308,0.07692308,0.102564104,0.057692308,0.057692308,0.070512824,0.23717949,0.06410257,0.06410257,0.10897436,0.08974359,0.12179487,0.07692308,0.07692308,0.08974359,0.051282052,0.08974359,0.083333336,0.2628205,0.07692308,0.09615385,0.17307693,0.12179487,0.27564102,0.22435898,0.21153846,0.1474359,0.115384616,0.23717949,0.31410256,0.32051283,0.2820513,0.30128205,0.25,0.28846154,0.4551282,0.30128205,0.44230768,0.44230768,0.48076922,0.18589744,0.3301282,,0.50641024,0.53846157,0.50641024,0.61538464,0.5897436,0.75641024
+FastaiMLP (default),0.0,0.0,0.0064102565,0.0064102565,0.0,0.0064102565,0.0064102565,0.012820513,0.0,0.057692308,0.0,0.0,0.06410257,0.07692308,0.057692308,0.057692308,0.01923077,0.01923077,0.01923077,0.08974359,0.051282052,0.057692308,0.025641026,0.025641026,0.012820513,0.057692308,0.012820513,0.0,0.115384616,0.0,0.0064102565,0.032051284,0.0064102565,0.09615385,0.01923077,0.19871795,0.01923077,0.10897436,0.08974359,0.057692308,0.03846154,0.032051284,0.10897436,0.070512824,0.12179487,0.032051284,0.12179487,0.0064102565,0.051282052,0.13461539,0.070512824,0.032051284,0.012820513,0.051282052,0.032051284,0.06410257,0.10897436,0.31410256,0.14102565,0.25,0.32051283,0.40384614,0.25,0.21794872,0.55128205,0.32051283,0.39102563,0.40384614,0.5897436,0.3974359,0.42948717,0.49358973,,0.3846154,0.71794873,0.7948718,0.90384614,0.99358976
+TabFlex (default),0.032051284,0.0,0.0,0.0,0.0,0.0,0.0064102565,0.051282052,0.0064102565,0.057692308,0.0,0.0,0.057692308,0.057692308,0.0,0.0,0.01923077,0.0,0.0,0.083333336,0.0064102565,0.0,0.0,0.0,0.0064102565,0.012820513,0.0064102565,0.0,0.057692308,0.0,0.0,0.0064102565,0.0,0.01923077,0.0,0.23717949,0.0064102565,0.051282052,0.0,0.025641026,0.0064102565,0.0,0.032051284,0.0064102565,0.044871796,0.0064102565,0.070512824,0.07692308,0.012820513,0.070512824,0.032051284,0.01923077,0.16025642,0.06410257,0.032051284,0.070512824,0.10897436,0.25641027,0.18589744,0.33333334,0.33974358,0.43589744,0.23717949,0.2820513,0.44230768,0.28846154,0.53846157,0.5448718,0.44871795,0.34615386,0.46153846,0.46153846,0.61538464,,0.724359,0.8076923,0.775641,1.0
+TabSTAR (default),0.0064102565,0.012820513,0.0064102565,0.0064102565,0.0064102565,0.0,0.0,0.0064102565,0.0,0.06410257,0.0064102565,0.012820513,0.06410257,0.06410257,0.057692308,0.06410257,0.032051284,0.0064102565,0.0064102565,0.070512824,0.032051284,0.070512824,0.012820513,0.0064102565,0.012820513,0.032051284,0.012820513,0.0064102565,0.10897436,0.0064102565,0.025641026,0.012820513,0.012820513,0.06410257,0.01923077,0.051282052,0.012820513,0.057692308,0.102564104,0.051282052,0.025641026,0.0064102565,0.06410257,0.01923077,0.06410257,0.01923077,0.07692308,0.06410257,0.012820513,0.12820514,0.044871796,0.025641026,0.06410257,0.083333336,0.057692308,0.057692308,0.057692308,0.07692308,0.102564104,0.17948718,0.16666667,0.35897437,0.102564104,0.26923078,0.37179488,0.21153846,0.0,0.0,0.3846154,0.4679487,0.525641,0.49358973,0.2820513,0.27564102,,0.6474359,0.69871795,0.96153843
+KNN (tuned),0.06410257,0.032051284,0.044871796,0.01923077,0.03846154,0.051282052,0.051282052,0.057692308,0.044871796,0.057692308,0.044871796,0.025641026,0.057692308,0.057692308,0.051282052,0.025641026,0.044871796,0.032051284,0.044871796,0.06410257,0.06410257,0.03846154,0.032051284,0.044871796,0.06410257,0.025641026,0.06410257,0.044871796,0.057692308,0.044871796,0.09615385,0.012820513,0.025641026,0.0,0.03846154,0.22435898,0.057692308,0.0064102565,0.115384616,0.10897436,0.057692308,0.051282052,0.0,0.025641026,0.044871796,0.057692308,0.025641026,0.102564104,0.070512824,0.09615385,0.057692308,0.025641026,0.102564104,0.13461539,0.13461539,0.070512824,0.070512824,0.07692308,0.070512824,0.1474359,0.10897436,0.06410257,0.083333336,0.15384616,0.23076923,0.23076923,0.14102565,0.14102565,0.24358974,0.3525641,0.35897437,0.3846154,0.20512821,0.1923077,0.3525641,,0.3525641,0.92948717
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.032051284,0.044871796,0.0064102565,0.0,0.0,0.0064102565,0.0064102565,0.057692308,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.051282052,0.0,0.17948718,0.0064102565,0.07692308,0.06410257,0.025641026,0.0064102565,0.0,0.070512824,0.0064102565,0.032051284,0.0064102565,0.09615385,0.0,0.0,0.01923077,0.012820513,0.0,0.0,0.025641026,0.01923077,0.0064102565,0.01923077,0.102564104,0.057692308,0.08974359,0.12820514,0.2948718,0.102564104,0.14102565,0.051282052,0.21794872,0.08974359,0.10897436,0.14102565,0.32051283,0.34615386,0.41025642,0.09615385,0.22435898,0.30128205,0.6474359,,0.9166667
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0064102565,0.0,0.0,0.06410257,0.06410257,0.012820513,0.012820513,0.06410257,0.057692308,0.06410257,0.24358974,0.0064102565,0.0,0.03846154,0.070512824,0.083333336,
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
index c5179037fddb7141190daf29bd88f7d116b412c4..01bc369a534fbab0d17dc9879d1266c0b0e2370a 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Multiclass | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..ea8efd2506e095317003281d1615e7477a7cb31d
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_multiclass/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":9121.0,"score":1.0,"rank":1.0,"hrank":1.0,"imp":0.0,"train_time":3.93,"predict_time":3.165,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":8156.0,"score":0.905,"rank":2.44,"hrank":2.3,"imp":2.724,"train_time":0.91,"predict_time":0.272,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":8028.0,"score":0.861,"rank":2.89,"hrank":2.7,"imp":3.919,"train_time":257.86,"predict_time":1.392,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":7765.0,"score":0.832,"rank":3.78,"hrank":3.67,"imp":4.701,"train_time":133.98,"predict_time":8.253,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":4,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":7084.0,"score":0.77,"rank":5.44,"hrank":5.29,"imp":6.303,"train_time":276.88,"predict_time":4.662,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":6998.0,"score":0.743,"rank":5.78,"hrank":5.74,"imp":6.994,"train_time":133.98,"predict_time":2.816,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":6,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":6643.0,"score":0.617,"rank":7.22,"hrank":7.05,"imp":10.043,"train_time":35.54,"predict_time":17.701,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":7,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":6586.0,"score":0.665,"rank":7.44,"hrank":7.41,"imp":8.893,"train_time":1.83,"predict_time":0.517,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":5356.0,"score":0.5,"rank":9.89,"hrank":9.76,"imp":12.741,"train_time":0.48,"predict_time":0.143,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":9,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":5104.0,"score":0.452,"rank":12.0,"hrank":11.6,"imp":13.784,"train_time":6148.01,"predict_time":0.15,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":10,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":5104.0,"score":0.452,"rank":12.0,"hrank":11.63,"imp":13.814,"train_time":61.95,"predict_time":0.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":5047.0,"score":0.432,"rank":12.67,"hrank":12.37,"imp":14.226,"train_time":157.72,"predict_time":3.649,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":12,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":4974.0,"score":0.44,"rank":13.56,"hrank":13.17,"imp":14.045,"train_time":6148.01,"predict_time":0.011,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":13,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":4915.0,"score":0.437,"rank":14.33,"hrank":13.74,"imp":14.138,"train_time":61.95,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":4897.0,"score":0.412,"rank":14.56,"hrank":14.09,"imp":14.638,"train_time":157.72,"predict_time":0.374,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":15,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":4879.0,"score":0.433,"rank":14.78,"hrank":13.63,"imp":14.204,"train_time":0.82,"predict_time":0.277,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":16,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":4651.0,"score":0.38,"rank":17.78,"hrank":17.71,"imp":15.327,"train_time":466.47,"predict_time":1.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":4614.0,"score":0.37,"rank":18.22,"hrank":17.84,"imp":15.529,"train_time":1199.98,"predict_time":0.344,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":4605.0,"score":0.367,"rank":18.33,"hrank":17.35,"imp":15.578,"train_time":1199.98,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":19,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":4530.0,"score":0.365,"rank":19.22,"hrank":19.06,"imp":15.649,"train_time":0.39,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":4502.0,"score":0.365,"rank":19.56,"hrank":19.45,"imp":15.649,"train_time":466.47,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":21,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":4102.0,"score":0.287,"rank":24.22,"hrank":23.87,"imp":17.235,"train_time":1222.57,"predict_time":21.653,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":22,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":4076.0,"score":0.289,"rank":24.56,"hrank":24.32,"imp":17.181,"train_time":3.47,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":23,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":4050.0,"score":0.268,"rank":24.89,"hrank":23.57,"imp":17.589,"train_time":1.48,"predict_time":0.53,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":24,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":4032.0,"score":0.277,"rank":25.11,"hrank":24.99,"imp":17.438,"train_time":68.36,"predict_time":0.189,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":25,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":4015.0,"score":0.278,"rank":25.33,"hrank":25.19,"imp":17.409,"train_time":68.36,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":26,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":4006.0,"score":0.276,"rank":25.44,"hrank":25.37,"imp":17.448,"train_time":4423.3,"predict_time":29.199,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":27,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":3971.0,"score":0.259,"rank":25.89,"hrank":25.51,"imp":17.775,"train_time":11.15,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":28,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":3743.0,"score":0.227,"rank":28.33,"hrank":28.28,"imp":18.417,"train_time":4423.3,"predict_time":2.237,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":29,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":3717.0,"score":0.221,"rank":28.56,"hrank":28.46,"imp":18.522,"train_time":2.82,"predict_time":1.073,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":30,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":2844.0,"score":0.137,"rank":31.44,"hrank":31.42,"imp":20.128,"train_time":26.01,"predict_time":0.151,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2640.0,"score":0.095,"rank":32.44,"hrank":32.4,"imp":20.889,"train_time":1222.57,"predict_time":7.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":32,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2486.0,"score":0.095,"rank":33.44,"hrank":33.4,"imp":20.889,"train_time":52.0,"predict_time":7.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":33,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":2344.0,"score":0.061,"rank":34.56,"hrank":34.41,"imp":21.497,"train_time":0.58,"predict_time":0.2,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":2266.0,"score":0.052,"rank":35.22,"hrank":35.16,"imp":21.684,"train_time":0.23,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":35,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2142.0,"score":0.018,"rank":36.33,"hrank":36.29,"imp":22.269,"train_time":4638.97,"predict_time":343.207,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":36,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":2080.0,"score":0.0,"rank":36.89,"hrank":36.76,"imp":22.636,"train_time":2389.34,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":37,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":2042.0,"score":0.003,"rank":37.22,"hrank":37.03,"imp":22.535,"train_time":2389.34,"predict_time":0.267,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":38,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1962.0,"score":0.0,"rank":37.89,"hrank":37.75,"imp":22.934,"train_time":26.01,"predict_time":0.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":39,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1587.0,"score":0.0,"rank":40.44,"hrank":40.44,"imp":24.385,"train_time":14.87,"predict_time":1.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":40,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1449.0,"score":0.0,"rank":41.44,"hrank":41.39,"imp":24.762,"train_time":1137.02,"predict_time":0.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":41,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1244.0,"score":0.0,"rank":43.22,"hrank":43.16,"imp":25.74,"train_time":4638.97,"predict_time":43.78,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":42,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1232.0,"score":0.0,"rank":43.33,"hrank":43.2,"imp":25.492,"train_time":1137.02,"predict_time":0.054,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":43,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1184.0,"score":0.0,"rank":43.78,"hrank":43.69,"imp":26.055,"train_time":9.87,"predict_time":9.237,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":44,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1172.0,"score":0.0,"rank":43.89,"hrank":43.81,"imp":26.094,"train_time":40.54,"predict_time":38.74,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":45,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":45.44,"hrank":45.32,"imp":26.928,"train_time":0.18,"predict_time":0.008,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":46,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":831.0,"score":0.0,"rank":46.78,"hrank":46.76,"imp":27.14,"train_time":1294.95,"predict_time":0.417,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":47,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":662.0,"score":0.0,"rank":47.89,"hrank":47.87,"imp":27.562,"train_time":1294.95,"predict_time":0.06,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":48,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":623.0,"score":0.0,"rank":48.11,"hrank":48.07,"imp":28.431,"train_time":0.18,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":49,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":29.0,"score":0.0,"rank":50.11,"hrank":50.11,"imp":31.618,"train_time":1863.61,"predict_time":6.302,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":50,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-412.0,"score":0.0,"rank":51.67,"hrank":51.66,"imp":34.019,"train_time":1863.61,"predict_time":1.131,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":51,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-502.0,"score":0.0,"rank":52.11,"hrank":52.08,"imp":34.009,"train_time":3.96,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":52,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-568.0,"score":0.0,"rank":52.44,"hrank":52.43,"imp":34.96,"train_time":3.96,"predict_time":0.077,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":53,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-1053.0,"score":0.0,"rank":55.0,"hrank":54.98,"imp":36.605,"train_time":66.07,"predict_time":1.792,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-1131.0,"score":0.0,"rank":55.5,"hrank":55.48,"imp":37.419,"train_time":1904.43,"predict_time":1.396,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":55,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-1140.0,"score":0.0,"rank":55.56,"hrank":55.48,"imp":37.434,"train_time":16.71,"predict_time":6.766,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":56,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-1166.0,"score":0.0,"rank":55.72,"hrank":55.71,"imp":37.463,"train_time":1904.43,"predict_time":2.971,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":57,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-1661.0,"score":0.0,"rank":57.89,"hrank":57.89,"imp":39.959,"train_time":7.55,"predict_time":0.149,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-2186.0,"score":0.0,"rank":59.44,"hrank":59.44,"imp":43.803,"train_time":569.0,"predict_time":0.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-2398.0,"score":0.0,"rank":60.44,"hrank":60.44,"imp":44.373,"train_time":569.0,"predict_time":1.658,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":60,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-2460.0,"score":0.0,"rank":60.78,"hrank":60.76,"imp":44.724,"train_time":65.83,"predict_time":1.583,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-2662.0,"score":0.0,"rank":61.89,"hrank":61.83,"imp":49.111,"train_time":28.2,"predict_time":0.515,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":62,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":-2916.0,"score":0.0,"rank":63.33,"hrank":63.32,"imp":51.715,"train_time":0.21,"predict_time":0.039,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-3106.0,"score":0.0,"rank":64.56,"hrank":64.48,"imp":52.242,"train_time":28.2,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-3155.0,"score":0.0,"rank":64.89,"hrank":64.88,"imp":53.569,"train_time":2.86,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":65,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-3236.0,"score":0.0,"rank":65.44,"hrank":65.43,"imp":53.943,"train_time":30.41,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":66,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-3542.0,"score":0.0,"rank":67.56,"hrank":67.55,"imp":57.076,"train_time":30.41,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-3661.0,"score":0.0,"rank":68.44,"hrank":68.43,"imp":57.449,"train_time":0.13,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":68,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-3719.0,"score":0.0,"rank":68.89,"hrank":68.8,"imp":57.324,"train_time":3.09,"predict_time":0.997,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-3865.0,"score":0.0,"rank":70.0,"hrank":69.96,"imp":59.086,"train_time":2.7,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-3896.0,"score":0.0,"rank":70.22,"hrank":70.21,"imp":58.52,"train_time":0.11,"predict_time":0.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":71,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-4035.0,"score":0.0,"rank":71.11,"hrank":71.1,"imp":59.325,"train_time":76.27,"predict_time":3.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":72,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-4973.0,"score":0.0,"rank":73.0,"hrank":73.0,"imp":72.158,"train_time":76.27,"predict_time":0.37,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-5748.0,"score":0.0,"rank":74.0,"hrank":74.0,"imp":83.474,"train_time":0.14,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/pareto_front_explorer.html
index 5ca97b2d459e916f09cb79eb13660c2ea657424c..91c315ca80aaf4a85c1b075bcff311933ea46bb0 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7916536996,"imp":36.6045992319,"elo":-1052.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0212757285,"imp":17.181396612,"elo":4075.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0810120956,"imp":15.5776096649,"elo":4604.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3436400049,"imp":15.529159267,"elo":4614.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135873471,"imp":21.6835247423,"elo":2266.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0169735693,"imp":17.4085248711,"elo":4014.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.189181224,"imp":17.4383709448,"elo":4032.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"imp":17.7749534644,"elo":3971.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0112425999,"imp":14.0452240321,"elo":4974.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1501267968,"imp":13.7840392894,"elo":5103.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.039263078,"imp":51.7150429706,"elo":-2916.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0094929976,"imp":22.9336337304,"elo":1961.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1510706771,"imp":20.1277389889,"elo":2843.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0849591873,"imp":53.5688669452,"elo":-3154.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1696430964,"imp":43.8028639504,"elo":-2186.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6581625315,"imp":44.3728722396,"elo":-2397.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0398739264,"imp":28.431281613,"elo":623.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3738761825,"imp":14.6382026202,"elo":4897.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":3.6487831267,"imp":14.225675726,"elo":5046.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":3.1649737164,"imp":0.0,"elo":9121.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8552799147,"imp":24.3848771338,"elo":1586.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2366352943,"imp":18.4172970522,"elo":3743.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":29.1989921988,"imp":17.4484171722,"elo":4006.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7656160501,"imp":37.434168827,"elo":-1139.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0157167235,"imp":59.0862850659,"elo":-3864.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1481536883,"imp":52.242462816,"elo":-3105.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5145997597,"imp":49.1106037807,"elo":-2662.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0076988881,"imp":26.928310568,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0164622104,"imp":14.1379017686,"elo":4914.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1699002263,"imp":13.8142375627,"elo":5103.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.1492926143,"imp":39.9591212587,"elo":-1661.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0604026988,"imp":27.5615449118,"elo":662.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":0.4165686723,"imp":27.1397331062,"elo":831.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5296965383,"imp":17.5892425696,"elo":4049.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8157300323,"imp":6.9940304738,"elo":6997.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.253098869,"imp":4.7014297019,"elo":7765.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2370616309,"imp":26.0548850064,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.7402173897,"imp":26.0937916411,"elo":1172.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7803277189,"imp":25.7404923043,"elo":1243.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":343.2065516381,"imp":22.2692141684,"elo":2142.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2773446206,"imp":14.2035859276,"elo":4879.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":17.7006354133,"imp":10.0434584757,"elo":6643.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0228731968,"imp":58.5202549384,"elo":-3895.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.0732772508,"imp":18.5221964405,"elo":3716.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426444891,"imp":12.7406135408,"elo":5355.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0767204065,"imp":34.9601806584,"elo":-568.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0538978015,"imp":25.4915754476,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.4463981702,"imp":24.7620187568,"elo":1448.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5166405697,"imp":8.8934485237,"elo":6585.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.27178525,"imp":2.7237963588,"elo":8156.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.5831571832,"imp":44.7240058444,"elo":-2459.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.3962170912,"imp":37.4185477938,"elo":-1131.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9708756879,"imp":37.4631227318,"elo":-1166.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.19959583,"imp":21.4967094579,"elo":2343.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0246185533,"imp":34.0086714629,"elo":-502.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.023043642,"imp":22.6357042665,"elo":2080.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2669741002,"imp":22.5345265676,"elo":2042.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"imp":15.6491623994,"elo":4529.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"imp":15.6491624778,"elo":4502.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0358030998,"imp":15.3274877808,"elo":4650.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"imp":20.8891440058,"elo":2486.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"imp":20.8891386829,"elo":2639.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":21.6530909246,"imp":17.2347396887,"elo":4101.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9973054408,"imp":57.3240231598,"elo":-3718.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":1.1308430496,"imp":34.019215077,"elo":-412.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":6.3018889681,"imp":31.6183153791,"elo":29.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7916536996,"x_train":66.0653250838,"imp":36.6045992319,"elo":-1052.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0212757285,"x_train":3.4684114436,"imp":17.181396612,"elo":4075.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0810120956,"x_train":1199.9825110223,"imp":15.5776096649,"elo":4604.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3436400049,"x_train":1199.9825110223,"imp":15.529159267,"elo":4614.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135873471,"x_train":0.2327292548,"imp":21.6835247423,"elo":2266.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0169735693,"x_train":68.3588435593,"imp":17.4085248711,"elo":4014.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.189181224,"x_train":68.3588435593,"imp":17.4383709448,"elo":4032.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"x_train":11.1470283442,"imp":17.7749534644,"elo":3971.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0112425999,"x_train":6148.0105869876,"imp":14.0452240321,"elo":4974.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1501267968,"x_train":6148.0105869876,"imp":13.7840392894,"elo":5103.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.039263078,"x_train":0.210649465,"imp":51.7150429706,"elo":-2916.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0094929976,"x_train":26.0128124598,"imp":22.9336337304,"elo":1961.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1510706771,"x_train":26.0128124598,"imp":20.1277389889,"elo":2843.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0849591873,"x_train":2.8561491959,"imp":53.5688669452,"elo":-3154.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1696430964,"x_train":569.0042450609,"imp":43.8028639504,"elo":-2186.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6581625315,"x_train":569.0042450609,"imp":44.3728722396,"elo":-2397.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0398739264,"x_train":0.177779185,"imp":28.431281613,"elo":623.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3738761825,"x_train":157.7188520507,"imp":14.6382026202,"elo":4897.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":3.6487831267,"x_train":157.7188520507,"imp":14.225675726,"elo":5046.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":3.1649737164,"x_train":3.9298653893,"imp":0.0,"elo":9121.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8552799147,"x_train":14.8696863446,"imp":24.3848771338,"elo":1586.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2366352943,"x_train":4423.2983658455,"imp":18.4172970522,"elo":3743.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":29.1989921988,"x_train":4423.2983658455,"imp":17.4484171722,"elo":4006.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7656160501,"x_train":16.7081227293,"imp":37.434168827,"elo":-1139.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0157167235,"x_train":2.7029391458,"imp":59.0862850659,"elo":-3864.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1481536883,"x_train":28.1974045386,"imp":52.242462816,"elo":-3105.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5145997597,"x_train":28.1974045386,"imp":49.1106037807,"elo":-2662.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0076988881,"x_train":0.1832795453,"imp":26.928310568,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0164622104,"x_train":61.9490599096,"imp":14.1379017686,"elo":4914.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1699002263,"x_train":61.9490599096,"imp":13.8142375627,"elo":5103.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.1492926143,"x_train":7.5476167071,"imp":39.9591212587,"elo":-1661.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0604026988,"x_train":1294.9500816771,"imp":27.5615449118,"elo":662.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":0.4165686723,"x_train":1294.9500816771,"imp":27.1397331062,"elo":831.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5296965383,"x_train":1.4775118336,"imp":17.5892425696,"elo":4049.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8157300323,"x_train":133.9809627068,"imp":6.9940304738,"elo":6997.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.253098869,"x_train":133.9809627068,"imp":4.7014297019,"elo":7765.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2370616309,"x_train":9.8664582493,"imp":26.0548850064,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.7402173897,"x_train":40.5370562747,"imp":26.0937916411,"elo":1172.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.7803277189,"x_train":4638.9676267895,"imp":25.7404923043,"elo":1243.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":343.2065516381,"x_train":4638.9676267895,"imp":22.2692141684,"elo":2142.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2773446206,"x_train":0.8153299726,"imp":14.2035859276,"elo":4879.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":17.7006354133,"x_train":35.5390806742,"imp":10.0434584757,"elo":6643.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0228731968,"x_train":0.1112663146,"imp":58.5202549384,"elo":-3895.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.0732772508,"x_train":2.8164229235,"imp":18.5221964405,"elo":3716.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426444891,"x_train":0.4844563774,"imp":12.7406135408,"elo":5355.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0767204065,"x_train":3.9605489519,"imp":34.9601806584,"elo":-568.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0538978015,"x_train":1137.0204365721,"imp":25.4915754476,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.4463981702,"x_train":1137.0204365721,"imp":24.7620187568,"elo":1448.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5166405697,"x_train":1.8291240885,"imp":8.8934485237,"elo":6585.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.27178525,"x_train":0.9127892597,"imp":2.7237963588,"elo":8156.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.5831571832,"x_train":65.832151552,"imp":44.7240058444,"elo":-2459.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.3962170912,"x_train":1904.4279322336,"imp":37.4185477938,"elo":-1131.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9708756879,"x_train":1904.4279322336,"imp":37.4631227318,"elo":-1166.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.19959583,"x_train":0.5828990849,"imp":21.4967094579,"elo":2343.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0246185533,"x_train":3.9645298444,"imp":34.0086714629,"elo":-502.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.023043642,"x_train":2389.3401785968,"imp":22.6357042665,"elo":2080.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2669741002,"x_train":2389.3401785968,"imp":22.5345265676,"elo":2042.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"x_train":0.3879892054,"imp":15.6491623994,"elo":4529.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"x_train":466.4747671387,"imp":15.6491624778,"elo":4502.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0358030998,"x_train":466.4747671387,"imp":15.3274877808,"elo":4650.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"x_train":52.0002317866,"imp":20.8891440058,"elo":2486.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"x_train":1222.5730299669,"imp":20.8891386829,"elo":2639.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":21.6530909246,"x_train":1222.5730299669,"imp":17.2347396887,"elo":4101.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9973054408,"x_train":3.0877205487,"imp":57.3240231598,"elo":-3718.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":1.1308430496,"x_train":1863.6070699824,"imp":34.019215077,"elo":-412.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":6.3018889681,"x_train":1863.6070699824,"imp":31.6183153791,"elo":29.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
index ec79b61c778223bada527ee5296d47c4bfd4364d..f63dfdc4a5020d7a98a5b55557ba26abbc776e02 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":66.0653250838,"imp":36.6174442254,"elo":-478.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.4684114436,"imp":17.1982189422,"elo":3638.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":9.4286940119,"imp":16.9514887084,"elo":3738.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":26.7881730821,"imp":16.4684558604,"elo":3901.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":53.5391935238,"imp":16.1390864594,"elo":4025.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":152.7281555404,"imp":15.7936821457,"elo":4178.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":297.8045100678,"imp":15.6874958736,"elo":4214.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":592.1492998961,"imp":15.5980918525,"elo":4269.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":887.2283364787,"imp":15.5567305187,"elo":4330.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1199.9825110223,"imp":15.5408704363,"elo":4352.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.2327292548,"imp":21.6993668415,"elo":2167.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.5808796011,"imp":20.704799374,"elo":2378.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.5148941182,"imp":19.3360040969,"elo":2908.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.1362950218,"imp":18.4935968731,"elo":3188.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":8.2937811532,"imp":18.0447983035,"elo":3317.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":16.9925774777,"imp":17.7958889184,"elo":3429.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":34.110705437,"imp":17.5701952871,"elo":3539.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":51.0567278918,"imp":17.4937654108,"elo":3595.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":68.3588435593,"imp":17.451429721,"elo":3640.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":11.1470283442,"imp":17.7912965293,"elo":3505.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":47.4246188425,"imp":16.6077982163,"elo":3870.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":136.0261588349,"imp":15.1701021831,"elo":4395.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":277.1128819248,"imp":14.8534981024,"elo":4477.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":742.3379842722,"imp":14.2956825998,"elo":4686.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":1521.2363697091,"imp":14.0882037921,"elo":4780.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":3056.5759955925,"imp":13.8958902042,"elo":4881.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":4572.6784881051,"imp":13.8335401341,"elo":4930.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":6148.0105869876,"imp":13.8029632333,"elo":4953.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.210649465,"imp":51.7258237302,"elo":-1819.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3400177245,"imp":44.4605754723,"elo":-1257.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.7306475516,"imp":30.0898438644,"elo":402.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.3716768748,"imp":24.3152042357,"elo":1561.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.3038879991,"imp":20.7499932217,"elo":2358.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.5251049409,"imp":20.419671907,"elo":2475.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.9730278534,"imp":20.2485255705,"elo":2550.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":19.420546809,"imp":20.1758964858,"elo":2604.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":26.0128124598,"imp":20.1432638787,"elo":2673.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.8561491959,"imp":53.5782989817,"elo":-2031.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.5274992778,"imp":53.7506549797,"elo":-2091.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.4182504903,"imp":51.8872540473,"elo":-1836.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":29.2437710823,"imp":49.4987366034,"elo":-1700.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":68.0373867319,"imp":48.2656225446,"elo":-1630.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":137.9784811137,"imp":47.1212705176,"elo":-1563.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":282.4399735782,"imp":46.0015801052,"elo":-1505.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":424.0861878909,"imp":45.0891495896,"elo":-1401.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":569.0042450609,"imp":44.3852270948,"elo":-1229.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.177779185,"imp":28.4458404969,"elo":650.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":1.2034968977,"imp":28.243018236,"elo":709.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":3.3940605174,"imp":25.6888302783,"elo":1217.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":7.2354516871,"imp":23.3133499239,"elo":1837.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":19.3256413209,"imp":17.7328375726,"elo":3450.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":38.6974931332,"imp":14.9559744843,"elo":4491.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":77.3490317839,"imp":14.4548346099,"elo":4717.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":116.3662460744,"imp":14.3500035973,"elo":4774.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":157.7188520507,"imp":14.2408141027,"elo":4855.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":3.9298653893,"imp":0.0394693775,"elo":7327.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.8696863446,"imp":24.4001445962,"elo":1550.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":38.3994266998,"imp":23.1409840761,"elo":1858.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":107.9937237268,"imp":21.7226317994,"elo":2113.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":214.6834801346,"imp":20.6688083433,"elo":2416.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":519.7617286711,"imp":19.327809423,"elo":2913.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1089.2077290986,"imp":18.5238195726,"elo":3197.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2164.8113675288,"imp":17.9392339201,"elo":3390.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3248.9873614597,"imp":17.6293029817,"elo":3491.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4423.2983658455,"imp":17.4693206477,"elo":3601.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":16.7081227293,"imp":38.4197521996,"elo":-676.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":2.7029391458,"imp":59.1337484624,"elo":-2952.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":8.4856383099,"imp":54.1095862893,"elo":-2149.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":23.7831181563,"imp":49.4770666427,"elo":-1708.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":28.1974045386,"imp":49.1229578957,"elo":-1619.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1832795453,"imp":27.0274509886,"elo":902.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4706668018,"imp":19.1404842461,"elo":2981.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.4894550512,"imp":16.0331829814,"elo":4148.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":3.053659024,"imp":15.0376831298,"elo":4387.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":7.5560037194,"imp":14.3353184179,"elo":4627.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":15.2866374484,"imp":14.1431702342,"elo":4714.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":30.5951033309,"imp":13.9589189437,"elo":4836.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":46.1662849504,"imp":13.8865443621,"elo":4887.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":61.9490599096,"imp":13.8317101134,"elo":4959.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":7.5476167071,"imp":39.9711686616,"elo":-840.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":13.2940077986,"imp":36.6622870804,"elo":-496.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":31.6350352615,"imp":34.1014658056,"elo":-170.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":62.0439269442,"imp":32.0827300636,"elo":139.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":161.8790996214,"imp":30.4501240663,"elo":312.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":328.4372725404,"imp":29.7166950188,"elo":435.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":646.6627533516,"imp":28.0626103943,"elo":650.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":968.2000565056,"imp":27.607032205,"elo":750.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1294.9500816771,"imp":27.1552069413,"elo":862.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.4775118336,"imp":17.6057023404,"elo":3588.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.9115484001,"imp":17.348949422,"elo":3692.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":3.2307266592,"imp":16.5991884106,"elo":3988.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.3145738811,"imp":16.2208138,"elo":4100.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":13.6646267949,"imp":13.878512824,"elo":4717.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":29.585585492,"imp":9.278089573,"elo":5800.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":63.4593657803,"imp":6.6147821389,"elo":6182.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":95.8186666767,"imp":5.1506740501,"elo":6348.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":133.9809627068,"imp":4.7186930132,"elo":6591.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":9.8664582493,"imp":26.0697540295,"elo":1145.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":40.5370562747,"imp":26.1086306019,"elo":1111.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.6468711363,"imp":25.8184706186,"elo":1211.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":120.4467271947,"imp":25.6314801824,"elo":1249.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":216.0600802864,"imp":25.327952527,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":558.8453433529,"imp":24.2818339933,"elo":1581.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1141.3485940959,"imp":23.7229068436,"elo":1717.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2295.535465635,"imp":23.1113096347,"elo":1858.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3458.2236513386,"imp":22.5100960665,"elo":1975.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4638.9676267895,"imp":22.2859884097,"elo":2066.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.8153299726,"imp":14.2207822206,"elo":4623.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":35.5390806742,"imp":10.0614938152,"elo":5836.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1112663146,"imp":58.5286308898,"elo":-2913.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":2.8164229235,"imp":18.5386881963,"elo":3234.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.4844563774,"imp":12.7581340857,"elo":5190.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":3.9605489519,"imp":34.9733101595,"elo":-215.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.0361437663,"imp":32.0711518311,"elo":144.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":26.071931656,"imp":28.6167683404,"elo":581.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":54.8698111305,"imp":27.1394097339,"elo":879.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":141.1191273125,"imp":26.2067443373,"elo":1097.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":283.0524385403,"imp":25.6380589495,"elo":1196.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":559.8211318778,"imp":25.1186792563,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":845.621900784,"imp":24.9207972757,"elo":1368.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1137.0204365721,"imp":24.7782924451,"elo":1439.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":1.8291240885,"imp":8.9117827376,"elo":5836.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":0.9127892597,"imp":2.7433951595,"elo":6822.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":65.832151552,"imp":44.7350522638,"elo":-1355.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":73.4850259904,"imp":44.6210107202,"elo":-1274.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":96.0490631917,"imp":44.5495907933,"elo":-1269.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":123.4027199075,"imp":44.2554746282,"elo":-1167.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":246.4013576646,"imp":42.8781169115,"elo":-1010.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":469.2886989974,"imp":41.053101283,"elo":-905.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":931.600810241,"imp":39.6973188826,"elo":-801.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1393.3493582788,"imp":38.5222349779,"elo":-709.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1904.4279322336,"imp":37.4756553614,"elo":-574.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.5828990849,"imp":21.512349933,"elo":2266.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":3.9645298444,"imp":34.0236440978,"elo":-170.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.6616998352,"imp":33.2937109549,"elo":-15.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":43.5410231146,"imp":32.0499097177,"elo":135.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":110.0430207107,"imp":29.7022282795,"elo":435.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":278.6743902359,"imp":27.524603557,"elo":768.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":589.8673134165,"imp":26.1048022154,"elo":1109.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1190.1861421136,"imp":24.659891563,"elo":1456.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1804.3563357667,"imp":23.8664121165,"elo":1672.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2389.3401785968,"imp":22.5506414343,"elo":1979.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.3879892054,"imp":15.6662513974,"elo":4163.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":3.362783389,"imp":15.5680144478,"elo":4186.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":10.9221958575,"imp":15.4900570576,"elo":4209.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":20.5995586615,"imp":15.4798083228,"elo":4224.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":57.8450913396,"imp":15.4229125134,"elo":4276.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":116.3010606073,"imp":15.4015823622,"elo":4305.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":231.8962989377,"imp":15.366403122,"elo":4340.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":345.7526810549,"imp":15.3570343393,"elo":4358.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":466.4747671387,"imp":15.3412007302,"elo":4401.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":52.0002317866,"imp":20.9051616936,"elo":2401.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":96.7947394331,"imp":20.1000134846,"elo":2651.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":239.4028626458,"imp":19.5210244374,"elo":2841.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":475.5145519061,"imp":18.7170397709,"elo":3206.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1177.5846933433,"imp":17.2998557543,"elo":3646.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1222.5730299669,"imp":17.2514686705,"elo":3695.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.0877205487,"imp":57.3324075538,"elo":-2746.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":12.6641668459,"imp":52.0527755162,"elo":-1914.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":42.6272411378,"imp":43.5787945161,"elo":-1172.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":88.9979412195,"imp":37.2830523058,"elo":-560.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":230.8719772304,"imp":34.8647448355,"elo":-288.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":461.2572637645,"imp":33.8141077812,"elo":-152.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":931.0175207888,"imp":32.9036398699,"elo":-19.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1389.8161177677,"imp":32.1679325704,"elo":97.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1863.6070699824,"imp":31.63374393,"elo":192.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7916536996,"x_train":66.0653250838,"imp":36.6174442254,"elo":-478.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0212757285,"x_train":3.4684114436,"imp":17.1982189422,"elo":3638.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0478586519,"x_train":9.4286940119,"imp":16.9514887084,"elo":3738.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.10020597,"x_train":26.7881730821,"imp":16.4684558604,"elo":3901.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1219205427,"x_train":53.5391935238,"imp":16.1390864594,"elo":4025.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2112305874,"x_train":152.7281555404,"imp":15.7936821457,"elo":4178.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2478990887,"x_train":297.8045100678,"imp":15.6874958736,"elo":4214.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2868644888,"x_train":592.1492998961,"imp":15.5980918525,"elo":4269.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3131577283,"x_train":887.2283364787,"imp":15.5567305187,"elo":4330.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.348787563,"x_train":1199.9825110223,"imp":15.5408704363,"elo":4352.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135873471,"x_train":0.2327292548,"imp":21.6993668415,"elo":2167.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306763366,"x_train":0.5808796011,"imp":20.704799374,"elo":2378.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.058031433,"x_train":1.5148941182,"imp":19.3360040969,"elo":2908.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0799589954,"x_train":3.1362950218,"imp":18.4935968731,"elo":3188.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.131080448,"x_train":8.2937811532,"imp":18.0447983035,"elo":3317.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1586631664,"x_train":16.9925774777,"imp":17.7958889184,"elo":3429.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1774151287,"x_train":34.110705437,"imp":17.5701952871,"elo":3539.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1906943335,"x_train":51.0567278918,"imp":17.4937654108,"elo":3595.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1929433857,"x_train":68.3588435593,"imp":17.451429721,"elo":3640.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0143528648,"x_train":11.1470283442,"imp":17.7912965293,"elo":3505.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0304247242,"x_train":47.4246188425,"imp":16.6077982163,"elo":3870.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.054592842,"x_train":136.0261588349,"imp":15.1701021831,"elo":4395.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0693827636,"x_train":277.1128819248,"imp":14.8534981024,"elo":4477.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1024467109,"x_train":742.3379842722,"imp":14.2956825998,"elo":4686.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1234968274,"x_train":1521.2363697091,"imp":14.0882037921,"elo":4780.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1405330049,"x_train":3056.5759955925,"imp":13.8958902042,"elo":4881.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1464533011,"x_train":4572.6784881051,"imp":13.8335401341,"elo":4930.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1501267968,"x_train":6148.0105869876,"imp":13.8029632333,"elo":4953.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.039263078,"x_train":0.210649465,"imp":51.7258237302,"elo":-1819.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0420610938,"x_train":0.3400177245,"imp":44.4605754723,"elo":-1257.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.035991066,"x_train":0.7306475516,"imp":30.0898438644,"elo":402.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0285236942,"x_train":1.3716768748,"imp":24.3152042357,"elo":1561.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0404946536,"x_train":3.3038879991,"imp":20.7499932217,"elo":2358.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0642507934,"x_train":6.5251049409,"imp":20.419671907,"elo":2475.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1069767422,"x_train":12.9730278534,"imp":20.2485255705,"elo":2550.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1321511543,"x_train":19.420546809,"imp":20.1758964858,"elo":2604.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1510706771,"x_train":26.0128124598,"imp":20.1432638787,"elo":2673.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0849591873,"x_train":2.8561491959,"imp":53.5782989817,"elo":-2031.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1669188087,"x_train":5.5274992778,"imp":53.7506549797,"elo":-2091.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3945555164,"x_train":14.4182504903,"imp":51.8872540473,"elo":-1836.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6441505685,"x_train":29.2437710823,"imp":49.4987366034,"elo":-1700.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9551913221,"x_train":68.0373867319,"imp":48.2656225446,"elo":-1630.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2793194595,"x_train":137.9784811137,"imp":47.1212705176,"elo":-1563.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7169442466,"x_train":282.4399735782,"imp":46.0015801052,"elo":-1505.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7762697202,"x_train":424.0861878909,"imp":45.0891495896,"elo":-1401.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6589060596,"x_train":569.0042450609,"imp":44.3852270948,"elo":-1229.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0398739264,"x_train":0.177779185,"imp":28.4458404969,"elo":650.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.707636671,"x_train":1.2034968977,"imp":28.243018236,"elo":709.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7934990344,"x_train":3.3940605174,"imp":25.6888302783,"elo":1217.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4799330005,"x_train":7.2354516871,"imp":23.3133499239,"elo":1837.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3202136893,"x_train":19.3256413209,"imp":17.7328375726,"elo":3450.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3734167029,"x_train":38.6974931332,"imp":14.9559744843,"elo":4491.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.0760086735,"x_train":77.3490317839,"imp":14.4548346099,"elo":4717.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.5418049787,"x_train":116.3662460744,"imp":14.3500035973,"elo":4774.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.6487831267,"x_train":157.7188520507,"imp":14.2408141027,"elo":4855.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":3.1649737164,"x_train":3.9298653893,"imp":0.0394693775,"elo":7327.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8552799147,"x_train":14.8696863446,"imp":24.4001445962,"elo":1550.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.8092665406,"x_train":38.3994266998,"imp":23.1409840761,"elo":1858.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.9564161978,"x_train":107.9937237268,"imp":21.7226317994,"elo":2113.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":11.1778615785,"x_train":214.6834801346,"imp":20.6688083433,"elo":2416.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":16.8970014449,"x_train":519.7617286711,"imp":19.327809423,"elo":2913.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":20.2753612803,"x_train":1089.2077290986,"imp":18.5238195726,"elo":3197.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":23.4597615325,"x_train":2164.8113675288,"imp":17.9392339201,"elo":3390.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":26.8100352995,"x_train":3248.9873614597,"imp":17.6293029817,"elo":3491.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":29.2376658843,"x_train":4423.2983658455,"imp":17.4693206477,"elo":3601.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7656160501,"x_train":16.7081227293,"imp":38.4197521996,"elo":-676.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0157167235,"x_train":2.7029391458,"imp":59.1337484624,"elo":-2952.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1460145839,"x_train":8.4856383099,"imp":54.1095862893,"elo":-2149.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4447962725,"x_train":23.7831181563,"imp":49.4770666427,"elo":-1708.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5145997597,"x_train":28.1974045386,"imp":49.1229578957,"elo":-1619.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0076988881,"x_train":0.1832795453,"imp":27.0274509886,"elo":902.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0211762886,"x_train":0.4706668018,"imp":19.1404842461,"elo":2981.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0401206899,"x_train":1.4894550512,"imp":16.0331829814,"elo":4148.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0553042309,"x_train":3.053659024,"imp":15.0376831298,"elo":4387.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0810592215,"x_train":7.5560037194,"imp":14.3353184179,"elo":4627.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1070101817,"x_train":15.2866374484,"imp":14.1431702342,"elo":4714.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1387414025,"x_train":30.5951033309,"imp":13.9589189437,"elo":4836.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1516373222,"x_train":46.1662849504,"imp":13.8865443621,"elo":4887.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1706958863,"x_train":61.9490599096,"imp":13.8317101134,"elo":4959.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.1492926143,"x_train":7.5476167071,"imp":39.9711686616,"elo":-840.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.209324614,"x_train":13.2940077986,"imp":36.6622870804,"elo":-496.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2344952565,"x_train":31.6350352615,"imp":34.1014658056,"elo":-170.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2811166627,"x_train":62.0439269442,"imp":32.0827300636,"elo":139.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3605099605,"x_train":161.8790996214,"imp":30.4501240663,"elo":312.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4238472839,"x_train":328.4372725404,"imp":29.7166950188,"elo":435.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3839306352,"x_train":646.6627533516,"imp":28.0626103943,"elo":650.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.406885263,"x_train":968.2000565056,"imp":27.607032205,"elo":750.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4165686723,"x_train":1294.9500816771,"imp":27.1552069413,"elo":862.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5296965383,"x_train":1.4775118336,"imp":17.6057023404,"elo":3588.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6843170741,"x_train":1.9115484001,"imp":17.348949422,"elo":3692.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9983103811,"x_train":3.2307266592,"imp":16.5991884106,"elo":3988.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7351012291,"x_train":5.3145738811,"imp":16.2208138,"elo":4100.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2130894628,"x_train":13.6646267949,"imp":13.878512824,"elo":4717.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.4628158113,"x_train":29.585585492,"imp":9.278089573,"elo":5800.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.400599732,"x_train":63.4593657803,"imp":6.6147821389,"elo":6182.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4275461958,"x_train":95.8186666767,"imp":5.1506740501,"elo":6348.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.253098869,"x_train":133.9809627068,"imp":4.7186930132,"elo":6591.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2370616309,"x_train":9.8664582493,"imp":26.0697540295,"elo":1145.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":38.7402173897,"x_train":40.5370562747,"imp":26.1086306019,"elo":1111.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.185823341,"x_train":63.6468711363,"imp":25.8184706186,"elo":1211.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":81.0004561794,"x_train":120.4467271947,"imp":25.6314801824,"elo":1249.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":114.1751843382,"x_train":216.0600802864,"imp":25.327952527,"elo":1325.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":170.4030706025,"x_train":558.8453433529,"imp":24.2818339933,"elo":1581.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":231.5103412116,"x_train":1141.3485940959,"imp":23.7229068436,"elo":1717.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":294.9832486808,"x_train":2295.535465635,"imp":23.1113096347,"elo":1858.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":314.3749405201,"x_train":3458.2236513386,"imp":22.5100960665,"elo":1975.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":343.2065516381,"x_train":4638.9676267895,"imp":22.2859884097,"elo":2066.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2773446206,"x_train":0.8153299726,"imp":14.2207822206,"elo":4623.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":17.7006354133,"x_train":35.5390806742,"imp":10.0614938152,"elo":5836.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0228731968,"x_train":0.1112663146,"imp":58.5286308898,"elo":-2913.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.0732772508,"x_train":2.8164229235,"imp":18.5386881963,"elo":3234.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426444891,"x_train":0.4844563774,"imp":12.7581340857,"elo":5190.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0767204065,"x_train":3.9605489519,"imp":34.9733101595,"elo":-215.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1313391893,"x_train":9.0361437663,"imp":32.0711518311,"elo":144.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2016062313,"x_train":26.071931656,"imp":28.6167683404,"elo":581.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2690374607,"x_train":54.8698111305,"imp":27.1394097339,"elo":879.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3287615415,"x_train":141.1191273125,"imp":26.2067443373,"elo":1097.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.3928311571,"x_train":283.0524385403,"imp":25.6380589495,"elo":1196.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4060980443,"x_train":559.8211318778,"imp":25.1186792563,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.4244678038,"x_train":845.621900784,"imp":24.9207972757,"elo":1368.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.4463981702,"x_train":1137.0204365721,"imp":24.7782924451,"elo":1439.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5166405697,"x_train":1.8291240885,"imp":8.9117827376,"elo":5836.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.27178525,"x_train":0.9127892597,"imp":2.7433951595,"elo":6822.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.5831571832,"x_train":65.832151552,"imp":44.7350522638,"elo":-1355.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6334117897,"x_train":73.4850259904,"imp":44.6210107202,"elo":-1274.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.7494818728,"x_train":96.0490631917,"imp":44.5495907933,"elo":-1269.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.8701977363,"x_train":123.4027199075,"imp":44.2554746282,"elo":-1167.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.3864186433,"x_train":246.4013576646,"imp":42.8781169115,"elo":-1010.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.8424963274,"x_train":469.2886989974,"imp":41.053101283,"elo":-905.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6043703251,"x_train":931.600810241,"imp":39.6973188826,"elo":-801.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.506791696,"x_train":1393.3493582788,"imp":38.5222349779,"elo":-709.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9708756879,"x_train":1904.4279322336,"imp":37.4756553614,"elo":-574.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.19959583,"x_train":0.5828990849,"imp":21.512349933,"elo":2266.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0246185533,"x_train":3.9645298444,"imp":34.0236440978,"elo":-170.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0576358482,"x_train":14.6616998352,"imp":33.2937109549,"elo":-15.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1132573234,"x_train":43.5410231146,"imp":32.0499097177,"elo":135.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.152027174,"x_train":110.0430207107,"imp":29.7022282795,"elo":435.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2158473884,"x_train":278.6743902359,"imp":27.524603557,"elo":768.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2573350856,"x_train":589.8673134165,"imp":26.1048022154,"elo":1109.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.277175668,"x_train":1190.1861421136,"imp":24.659891563,"elo":1456.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2880694779,"x_train":1804.3563357667,"imp":23.8664121165,"elo":1672.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2668851689,"x_train":2389.3401785968,"imp":22.5506414343,"elo":1979.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0514842572,"x_train":0.3879892054,"imp":15.6662513974,"elo":4163.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.366384192,"x_train":3.362783389,"imp":15.5680144478,"elo":4186.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6186026273,"x_train":10.9221958575,"imp":15.4900570576,"elo":4209.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6552489612,"x_train":20.5995586615,"imp":15.4798083228,"elo":4224.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8977020343,"x_train":57.8450913396,"imp":15.4229125134,"elo":4276.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9105374137,"x_train":116.3010606073,"imp":15.4015823622,"elo":4305.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8753171531,"x_train":231.8962989377,"imp":15.366403122,"elo":4340.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9589825129,"x_train":345.7526810549,"imp":15.3570343393,"elo":4358.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0370834578,"x_train":466.4747671387,"imp":15.3412007302,"elo":4401.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":7.8552922541,"x_train":52.0002317866,"imp":20.9051616936,"elo":2401.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":11.8109670804,"x_train":96.7947394331,"imp":20.1000134846,"elo":2651.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":16.6639574996,"x_train":239.4028626458,"imp":19.5210244374,"elo":2841.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":18.6752813789,"x_train":475.5145519061,"imp":18.7170397709,"elo":3206.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":21.4685568314,"x_train":1177.5846933433,"imp":17.2998557543,"elo":3646.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":21.6530909246,"x_train":1222.5730299669,"imp":17.2514686705,"elo":3695.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9973054408,"x_train":3.0877205487,"imp":57.3324075538,"elo":-2746.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.9976912644,"x_train":12.6641668459,"imp":52.0527755162,"elo":-1914.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.9612924274,"x_train":42.6272411378,"imp":43.5787945161,"elo":-1172.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":3.5953994677,"x_train":88.9979412195,"imp":37.2830523058,"elo":-560.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":5.3075739184,"x_train":230.8719772304,"imp":34.8647448355,"elo":-288.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":6.4637514204,"x_train":461.2572637645,"imp":33.8141077812,"elo":-152.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":7.0798232983,"x_train":931.0175207888,"imp":32.9036398699,"elo":-19.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":7.046435977,"x_train":1389.8161177677,"imp":32.1679325704,"elo":97.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":6.3018889681,"x_train":1863.6070699824,"imp":31.63374393,"elo":192.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..86d1ad8bd826f8c5086bcfe83fd9cb6bfbc87f4c
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_multiclass/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["LimiX (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (tuned)", "TabFM (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "EBM (tuned + ensembled)", "RandomForest (tuned + ensembled)", "LightGBM (tuned + ensembled)", "EBM (tuned)", "RandomForest (tuned)", "LightGBM (tuned)", "TabDPT-Turbo (default)", "XGBoost (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "XGBoost (default)", "XGBoost (tuned)", "iLTM (tuned + ensembled)", "CatBoost (default)", "RealTabPFN-2.5 (default)", "ChimeraBoost (tuned + ensembled)", "ChimeraBoost (tuned)", "ModernNCA (tuned + ensembled)", "EBM (default)", "ModernNCA (tuned)", "TabICL (default)", "ExtraTrees (tuned + ensembled)", "iLTM (tuned)", "iLTM (default)", "TabSwift (default)", "ChimeraBoost (default)", "TabDPT (tuned + ensembled)", "TorchMLP (tuned)", "TorchMLP (tuned + ensembled)", "ExtraTrees (tuned)", "ModernNCA (default)", "TabM (tuned + ensembled)", "TabDPT (tuned)", "TabM (tuned)", "SAP-RPT-OSS (default)", "TabDPT (default)", "RandomForest (default)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "LightGBM (default)", "xRFM (tuned + ensembled)", "xRFM (tuned)", "TorchMLP (default)", "TabM (default)", "BetaTabPFN (default)", "TabSTAR (tuned)", "OrionMSP (default)", "TabSTAR (tuned + ensembled)", "RealMLP (default)", "FastaiMLP (tuned)", "FastaiMLP (tuned + ensembled)", "TabSTAR (default)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "xRFM (default)", "PerpetualBooster (default)", "TabFlex (default)", "KNN (tuned + ensembled)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, NaN, 0.6666666865348816, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.3333333432674408, NaN, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.2222222238779068, NaN, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, NaN, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.3333333432674408, NaN, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, NaN, 0.4444444477558136, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5555555820465088, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.4444444477558136, NaN, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, NaN, 0.5555555820465088, 0.5555555820465088, 0.8888888955116272, 0.4444444477558136, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.4444444477558136, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, NaN, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, NaN, 0.5555555820465088, 0.4444444477558136, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, NaN, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.8888888955116272, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, NaN, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, NaN, 0.6666666865348816, 1.0, 1.0, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, NaN, 1.0, 1.0, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.0, 0.0, NaN, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4444444477558136, NaN, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, NaN, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.6666666865348816, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, NaN, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.2222222238779068, 0.6666666865348816, 0.4444444477558136, NaN, 0.7777777910232544, 0.4444444477558136, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.3333333432674408, 0.2222222238779068, NaN, 0.5555555820465088, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, NaN, 0.7777777910232544, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, NaN, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, NaN, 1.0, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, NaN, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, NaN, 0.4444444477558136, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.5555555820465088, NaN, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.3333333432674408, NaN, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, NaN, 0.5555555820465088, 0.5555555820465088, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, NaN, 0.5555555820465088, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, NaN, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, NaN, 0.6666666865348816, 1.0, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, NaN, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, NaN, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, NaN, 0.3333333432674408, 0.5555555820465088, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.2222222238779068, 0.6666666865348816, NaN, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, NaN, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, NaN, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, NaN, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, NaN, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.5555555820465088, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.4444444477558136, NaN, 0.4444444477558136, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.5555555820465088, NaN, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, NaN, 0.5555555820465088, 0.6111111044883728, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, NaN, 0.4444444477558136, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.3888888955116272, 0.5555555820465088, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, NaN, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, NaN, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, NaN, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, NaN, 0.5555555820465088, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4444444477558136, NaN, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, NaN, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.3333333432674408, NaN, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, NaN, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, NaN, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.0, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, NaN, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, NaN, 0.7777777910232544, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, NaN, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":1.0},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.9802130902},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9741248101},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9619482497},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9391171999},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.9345509896},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9147640792},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.9117199395},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8782343995},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8493150703},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8493150698},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8401826496},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.8280060892},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.8173515994},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.8143074592},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.8112633212},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7701674286},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7640791491},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7625570794},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.7503805183},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.7458143081},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6818873689},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6773211586},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6727549482},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6697108084},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.6666666676},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6651445976},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.6590563181},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6255707767},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6225266365},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5829528162},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5692541863},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.5555555561},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5403348564},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5312024365},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5159817362},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.5083713861},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5038051758},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4946727554},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4596651452},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4459665154},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4216133954},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.4200913251},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4140030454},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4124809746},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.3911719948},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3729071543},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.3576864539},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3546423137},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.3272450534},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.3059360736},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2998477932},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.2952815834},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.2602739734},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2534246579},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2526636229},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2503805181},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.2207001522},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.1993911724},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.1856925423},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1811263321},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1659056322},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1461187219},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1293759522},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1248097416},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1171993914},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0882800614},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0761035013},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.0700152218},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0547945214},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.0517503809},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.0395738205},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.0136986301},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..c46e757463f155afa7e64c9397d3004d23e910e0
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,LimiX (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (tuned),TabFM (default),TabPFN-2.6 (default),TabICLv2 (default),EBM (tuned + ensembled),RandomForest (tuned + ensembled),LightGBM (tuned + ensembled),EBM (tuned),RandomForest (tuned),LightGBM (tuned),TabDPT-Turbo (default),XGBoost (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),XGBoost (default),XGBoost (tuned),iLTM (tuned + ensembled),CatBoost (default),RealTabPFN-2.5 (default),ChimeraBoost (tuned + ensembled),ChimeraBoost (tuned),ModernNCA (tuned + ensembled),EBM (default),ModernNCA (tuned),TabICL (default),ExtraTrees (tuned + ensembled),iLTM (tuned),iLTM (default),TabSwift (default),ChimeraBoost (default),TabDPT (tuned + ensembled),TorchMLP (tuned),TorchMLP (tuned + ensembled),ExtraTrees (tuned),ModernNCA (default),TabM (tuned + ensembled),TabDPT (tuned),TabM (tuned),SAP-RPT-OSS (default),TabDPT (default),RandomForest (default),RealMLP (tuned + ensembled),RealMLP (tuned),LightGBM (default),xRFM (tuned + ensembled),xRFM (tuned),TorchMLP (default),TabM (default),BetaTabPFN (default),TabSTAR (tuned),OrionMSP (default),TabSTAR (tuned + ensembled),RealMLP (default),FastaiMLP (tuned),FastaiMLP (tuned + ensembled),TabSTAR (default),PerpetualBooster (tuned + ensembled),ExtraTrees (default),PerpetualBooster (tuned),FastaiMLP (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),xRFM (default),PerpetualBooster (default),TabFlex (default),KNN (tuned + ensembled),KNN (tuned),KNN (default)
+LimiX (default),,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.0,,0.6666667,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.0,0.33333334,,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,0.11111111,0.22222222,,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.0,0.0,0.11111111,,0.6666667,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.0,0.0,0.0,0.33333334,,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.0,0.0,0.0,0.0,0.22222222,0.11111111,,0.44444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.5555556,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.7777778,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.44444445,,0.6666667,0.6666667,0.8888889,0.7777778,0.6666667,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.33333334,0.33333334,,0.5555556,0.5555556,0.8888889,0.44444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.33333334,0.33333334,0.44444445,,0.44444445,0.5555556,0.44444445,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.11111111,0.44444445,0.5555556,,0.44444445,0.44444445,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.22222222,0.11111111,0.44444445,0.5555556,,0.44444445,0.7777778,0.8888889,0.7777778,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.33333334,0.5555556,0.5555556,0.5555556,0.5555556,,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.33333334,,0.5555556,0.44444445,0.8888889,0.8888889,1.0,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.44444445,,0.6666667,0.5555556,0.5555556,0.8888889,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.22222222,0.33333334,0.5555556,0.33333334,,0.5555556,0.5555556,0.7777778,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.33333334,0.11111111,0.44444445,0.44444445,,0.6666667,1.0,1.0,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.33333334,0.11111111,0.44444445,0.44444445,0.33333334,,1.0,1.0,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.0,0.0,,0.5555556,0.6666667,0.5555556,0.6666667,0.5555556,0.6666667,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.44444445,,0.6666667,0.6666667,0.7777778,0.5555556,0.5555556,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.33333334,,0.33333334,0.33333334,0.33333334,0.44444445,0.6666667,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.44444445,0.33333334,0.6666667,,0.5555556,0.6666667,0.5555556,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.22222222,0.6666667,0.44444445,,0.7777778,0.44444445,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.44444445,0.44444445,0.6666667,0.33333334,0.22222222,,0.5555556,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.44444445,0.5555556,0.44444445,0.5555556,0.44444445,,0.7777778,0.7777778,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.33333334,0.11111111,0.11111111,0.0,0.22222222,,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICL (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.33333334,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,,1.0,0.6666667,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,,0.6666667,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.33333334,0.33333334,,0.44444445,0.6666667,0.8888889,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.5555556,,0.6666667,0.6666667,0.7777778,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.33333334,,0.6666667,0.5555556,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.33333334,,0.5555556,0.5555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.44444445,0.44444445,,0.5555556,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.11111111,0.22222222,0.44444445,0.44444445,,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,,0.6666667,1.0,0.7777778,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.33333334,,0.7777778,0.7777778,0.6666667,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,,0.5555556,0.7777778,0.6666667,0.5555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.22222222,0.22222222,0.44444445,,0.33333334,0.5555556,0.7777778,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.22222222,0.6666667,,0.5555556,0.6666667,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.33333334,0.44444445,0.44444445,,0.7777778,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.44444445,0.22222222,0.33333334,0.22222222,,0.6666667,0.7777778,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.11111111,0.33333334,,0.7777778,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.22222222,,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.22222222,0.22222222,0.33333334,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.5555556,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.44444445,,0.44444445,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.5555556,,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.6666667,0.6666667,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,,0.5555556,0.6111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.44444445,,0.44444445,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.33333334,0.3888889,0.5555556,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,0.8888889,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,,0.6666667,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.33333334,,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.22222222,0.22222222,,0.7777778,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,,0.5555556,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.44444445,,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.8888889,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.22222222,,0.6666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.22222222,0.33333334,,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,,0.8888889,0.6666667,0.7777778,1.0,1.0,1.0,1.0
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,,0.6666667,0.7777778,0.8888889,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.22222222,0.0,0.11111111,0.33333334,0.33333334,,0.5555556,0.6666667,0.7777778,1.0,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.22222222,0.44444445,,0.5555556,0.5555556,1.0,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.33333334,0.44444445,,0.7777778,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.44444445,0.22222222,,1.0,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
index 073b493cf59fe5cd999702fe85850a464683d99f..feeb3d4cebff4c8f193e25f2ddca5d211295a3b6 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Multiclass | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..82d740fb1a9143eb975d6e11a5ddf58db9680cfb
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_multiclass/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1726.0,"score":0.683,"rank":10.62,"hrank":6.44,"imp":12.937,"train_time":245.58,"predict_time":2.838,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+158\/-58","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1723.0,"score":0.698,"rank":10.73,"hrank":4.93,"imp":12.899,"train_time":2742.8,"predict_time":1.218,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+195\/-76","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1711.0,"score":0.684,"rank":11.27,"hrank":4.64,"imp":12.07,"train_time":2742.8,"predict_time":7.936,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+244\/-110","imputed":false},{"position":3,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1690.0,"score":0.834,"rank":12.28,"hrank":1.8,"imp":8.359,"train_time":64.25,"predict_time":6.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+434\/-236","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1642.0,"score":0.599,"rank":14.85,"hrank":9.09,"imp":15.478,"train_time":11.1,"predict_time":0.682,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+200\/-97","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1629.0,"score":0.583,"rank":15.59,"hrank":6.31,"imp":13.201,"train_time":11.21,"predict_time":0.845,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+265\/-149","imputed":false},{"position":6,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1582.0,"score":0.486,"rank":18.51,"hrank":9.2,"imp":15.847,"train_time":4471.59,"predict_time":21.354,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+188\/-110","imputed":false},{"position":7,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1580.0,"score":0.51,"rank":18.61,"hrank":8.31,"imp":14.921,"train_time":12.32,"predict_time":0.524,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+199\/-135","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1564.0,"score":0.535,"rank":19.7,"hrank":6.89,"imp":16.249,"train_time":6.2,"predict_time":0.211,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+248\/-173","imputed":false},{"position":9,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1560.0,"score":0.566,"rank":19.96,"hrank":7.01,"imp":19.774,"train_time":2987.95,"predict_time":15.005,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+321\/-224","imputed":true},{"position":10,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1554.0,"score":0.462,"rank":20.34,"hrank":10.98,"imp":16.821,"train_time":4471.59,"predict_time":3.755,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-53","imputed":false},{"position":11,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1547.0,"score":0.459,"rank":20.86,"hrank":10.87,"imp":15.683,"train_time":46.82,"predict_time":7.43,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+159\/-116","imputed":false},{"position":12,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1512.0,"score":0.317,"rank":23.3,"hrank":12.0,"imp":19.524,"train_time":6549.61,"predict_time":2.898,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+188\/-84","imputed":false},{"position":13,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1512.0,"score":0.417,"rank":23.37,"hrank":12.77,"imp":19.572,"train_time":3919.29,"predict_time":4.506,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-81","imputed":false},{"position":14,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1504.0,"score":0.456,"rank":23.96,"hrank":12.71,"imp":21.501,"train_time":200.35,"predict_time":4.415,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+244\/-180","imputed":true},{"position":15,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1472.0,"score":0.409,"rank":26.38,"hrank":11.04,"imp":21.746,"train_time":2987.95,"predict_time":0.629,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+229\/-197","imputed":true},{"position":16,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1470.0,"score":0.364,"rank":26.47,"hrank":14.63,"imp":20.344,"train_time":3919.29,"predict_time":0.749,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+119\/-74","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1459.0,"score":0.293,"rank":27.32,"hrank":23.2,"imp":19.475,"train_time":3415.61,"predict_time":1.368,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-34","imputed":false},{"position":18,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1459.0,"score":0.294,"rank":27.38,"hrank":20.7,"imp":20.44,"train_time":1502.45,"predict_time":6.327,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-90","imputed":false},{"position":19,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1458.0,"score":0.384,"rank":27.44,"hrank":13.33,"imp":22.234,"train_time":4631.38,"predict_time":190.548,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+183\/-162","imputed":false},{"position":20,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1450.0,"score":0.337,"rank":28.09,"hrank":15.15,"imp":21.234,"train_time":7.26,"predict_time":0.403,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+206\/-167","imputed":true},{"position":21,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1449.0,"score":0.277,"rank":28.14,"hrank":22.28,"imp":20.638,"train_time":1502.45,"predict_time":1.367,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+131\/-77","imputed":false},{"position":22,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1438.0,"score":0.324,"rank":28.97,"hrank":20.64,"imp":21.281,"train_time":684.78,"predict_time":4.813,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-104","imputed":false},{"position":23,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1433.0,"score":0.239,"rank":29.4,"hrank":24.27,"imp":20.182,"train_time":3415.61,"predict_time":0.217,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-36","imputed":false},{"position":24,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1432.0,"score":0.298,"rank":29.48,"hrank":11.46,"imp":22.306,"train_time":6.75,"predict_time":0.264,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+212\/-172","imputed":false},{"position":25,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1427.0,"score":0.297,"rank":29.89,"hrank":18.02,"imp":21.622,"train_time":2097.62,"predict_time":3.141,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-139","imputed":false},{"position":26,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1420.0,"score":0.322,"rank":30.4,"hrank":14.82,"imp":23.246,"train_time":4631.38,"predict_time":30.351,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+197\/-163","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1414.0,"score":0.21,"rank":30.89,"hrank":22.67,"imp":20.977,"train_time":2551.65,"predict_time":1.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-68","imputed":false},{"position":28,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1411.0,"score":0.171,"rank":31.16,"hrank":22.48,"imp":20.621,"train_time":7359.32,"predict_time":9.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-81","imputed":false},{"position":29,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1408.0,"score":0.191,"rank":31.38,"hrank":14.35,"imp":20.788,"train_time":7359.32,"predict_time":0.596,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-61","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1405.0,"score":0.202,"rank":31.64,"hrank":24.45,"imp":21.125,"train_time":2551.65,"predict_time":4.526,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-66","imputed":false},{"position":31,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1405.0,"score":0.287,"rank":31.66,"hrank":18.46,"imp":22.182,"train_time":13.71,"predict_time":0.733,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-110","imputed":false},{"position":32,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1399.0,"score":0.167,"rank":32.18,"hrank":26.13,"imp":21.519,"train_time":34415.89,"predict_time":1003.647,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-72","imputed":false},{"position":33,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1387.0,"score":0.408,"rank":33.11,"hrank":6.84,"imp":20.789,"train_time":22.56,"predict_time":1.484,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+278\/-222","imputed":false},{"position":34,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1349.0,"score":0.17,"rank":36.31,"hrank":30.52,"imp":23.282,"train_time":45754.93,"predict_time":0.315,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-87","imputed":false},{"position":35,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1347.0,"score":0.242,"rank":36.47,"hrank":15.38,"imp":24.971,"train_time":1076.45,"predict_time":1.541,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+194\/-186","imputed":false},{"position":36,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1342.0,"score":0.156,"rank":36.87,"hrank":30.88,"imp":22.676,"train_time":29.23,"predict_time":0.249,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-73","imputed":false},{"position":37,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1332.0,"score":0.132,"rank":37.72,"hrank":28.98,"imp":23.589,"train_time":4142.9,"predict_time":2.104,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-117","imputed":false},{"position":38,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1330.0,"score":0.226,"rank":37.87,"hrank":14.51,"imp":24.173,"train_time":2775.63,"predict_time":12.586,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+215\/-175","imputed":false},{"position":39,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1330.0,"score":0.203,"rank":37.89,"hrank":21.83,"imp":26.582,"train_time":11.18,"predict_time":2.132,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+203\/-166","imputed":true},{"position":40,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1326.0,"score":0.189,"rank":38.21,"hrank":18.99,"imp":23.847,"train_time":2097.62,"predict_time":0.514,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-116","imputed":false},{"position":41,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1306.0,"score":0.093,"rank":39.89,"hrank":34.68,"imp":22.867,"train_time":34415.89,"predict_time":299.32,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-99","imputed":false},{"position":42,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1305.0,"score":0.229,"rank":40.02,"hrank":14.78,"imp":25.632,"train_time":755.27,"predict_time":2.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+191\/-209","imputed":false},{"position":43,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1304.0,"score":0.127,"rank":40.05,"hrank":33.54,"imp":24.051,"train_time":45754.93,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-94","imputed":false},{"position":44,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1300.0,"score":0.218,"rank":40.39,"hrank":13.05,"imp":26.402,"train_time":1076.45,"predict_time":0.31,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+193\/-207","imputed":false},{"position":45,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1298.0,"score":0.187,"rank":40.63,"hrank":15.9,"imp":26.049,"train_time":46.25,"predict_time":45.719,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+222\/-194","imputed":false},{"position":46,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1289.0,"score":0.119,"rank":41.34,"hrank":29.74,"imp":25.35,"train_time":15.52,"predict_time":0.045,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+140\/-129","imputed":false},{"position":47,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1284.0,"score":0.118,"rank":41.78,"hrank":35.35,"imp":26.297,"train_time":5.04,"predict_time":0.071,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-152","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1271.0,"score":0.111,"rank":42.81,"hrank":23.99,"imp":25.486,"train_time":2775.63,"predict_time":1.431,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+185\/-156","imputed":false},{"position":49,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1271.0,"score":0.158,"rank":42.85,"hrank":30.41,"imp":24.868,"train_time":5.32,"predict_time":0.43,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+166\/-175","imputed":false},{"position":50,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1264.0,"score":0.118,"rank":43.42,"hrank":30.12,"imp":25.957,"train_time":4142.9,"predict_time":0.287,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+129\/-157","imputed":false},{"position":51,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1253.0,"score":0.174,"rank":44.27,"hrank":11.67,"imp":27.029,"train_time":755.27,"predict_time":0.317,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+188\/-234","imputed":false},{"position":52,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1252.0,"score":0.135,"rank":44.38,"hrank":33.61,"imp":25.36,"train_time":2032.73,"predict_time":0.268,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-191","imputed":false},{"position":53,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1244.0,"score":0.0,"rank":45.03,"hrank":43.43,"imp":24.551,"train_time":5.48,"predict_time":0.599,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+26\/-43","imputed":false},{"position":54,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1242.0,"score":0.124,"rank":45.12,"hrank":38.22,"imp":25.428,"train_time":2032.73,"predict_time":1.367,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+144\/-173","imputed":false},{"position":55,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1217.0,"score":0.022,"rank":47.12,"hrank":38.58,"imp":28.672,"train_time":20.96,"predict_time":0.482,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-99","imputed":false},{"position":56,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1205.0,"score":0.0,"rank":48.02,"hrank":44.64,"imp":27.091,"train_time":19.64,"predict_time":7.682,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-75","imputed":false},{"position":57,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1166.0,"score":0.117,"rank":50.99,"hrank":38.28,"imp":28.316,"train_time":8.45,"predict_time":0.239,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+181\/-245","imputed":false},{"position":58,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1160.0,"score":0.055,"rank":51.36,"hrank":45.05,"imp":28.57,"train_time":10.84,"predict_time":2.189,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-175","imputed":false},{"position":59,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1099.0,"score":0.0,"rank":55.67,"hrank":41.05,"imp":30.996,"train_time":14.75,"predict_time":0.337,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-149","imputed":false},{"position":60,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1078.0,"score":0.069,"rank":57.0,"hrank":25.67,"imp":38.868,"train_time":173.46,"predict_time":1.637,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+212\/-333","imputed":false},{"position":61,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1043.0,"score":0.0,"rank":59.17,"hrank":55.4,"imp":34.335,"train_time":22.95,"predict_time":3.32,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-153","imputed":false},{"position":62,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1039.0,"score":0.0,"rank":59.42,"hrank":53.1,"imp":32.55,"train_time":42561.15,"predict_time":35.622,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+148\/-201","imputed":false},{"position":63,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1038.0,"score":0.0,"rank":59.46,"hrank":58.17,"imp":30.796,"train_time":581.49,"predict_time":277.193,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-165","imputed":false},{"position":64,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1033.0,"score":0.0,"rank":59.76,"hrank":56.19,"imp":40.609,"train_time":905.41,"predict_time":0.875,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-204","imputed":false},{"position":65,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1032.0,"score":0.0,"rank":59.85,"hrank":55.28,"imp":32.684,"train_time":42561.15,"predict_time":9.171,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+139\/-198","imputed":false},{"position":66,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1004.0,"score":0.0,"rank":61.44,"hrank":57.65,"imp":41.139,"train_time":905.41,"predict_time":0.175,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-213","imputed":false},{"position":67,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":61.67,"hrank":57.25,"imp":39.544,"train_time":0.85,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-205","imputed":false},{"position":68,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":996.0,"score":0.0,"rank":61.88,"hrank":59.29,"imp":37.458,"train_time":5.99,"predict_time":1.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-214","imputed":false},{"position":69,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":995.0,"score":0.0,"rank":61.96,"hrank":58.8,"imp":34.319,"train_time":8.97,"predict_time":0.742,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+149\/-183","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":925.0,"score":0.0,"rank":65.49,"hrank":64.9,"imp":39.063,"train_time":5.83,"predict_time":1.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-123","imputed":false},{"position":71,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":924.0,"score":0.0,"rank":65.52,"hrank":56.44,"imp":52.187,"train_time":1838.94,"predict_time":13.381,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-225","imputed":false},{"position":72,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":915.0,"score":0.06,"rank":65.96,"hrank":45.29,"imp":48.163,"train_time":173.46,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+266\/-427","imputed":false},{"position":73,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":880.0,"score":0.0,"rank":67.44,"hrank":38.71,"imp":53.749,"train_time":1838.94,"predict_time":5.572,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-242","imputed":false},{"position":74,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":880.0,"score":0.0,"rank":67.48,"hrank":65.36,"imp":41.371,"train_time":801.38,"predict_time":8.877,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-249","imputed":false},{"position":75,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":868.0,"score":0.0,"rank":67.96,"hrank":32.13,"imp":57.665,"train_time":295.36,"predict_time":0.573,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-279","imputed":false},{"position":76,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":792.0,"score":0.0,"rank":70.63,"hrank":69.93,"imp":47.764,"train_time":4.72,"predict_time":0.188,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-241","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":409.0,"score":0.0,"rank":77.09,"hrank":77.07,"imp":69.878,"train_time":0.57,"predict_time":0.067,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-267","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/pareto_front_explorer.html
index ad3f1f21cbe84a35d7789f1c5fbc9c786efc7213..75972ce14ccb990f68696d2d2c9362cc37dfe01c 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.8133362736,"imp":21.2807580829,"elo":1438.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2489310712,"imp":22.6758278335,"elo":1342.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.216888627,"imp":20.1819973609,"elo":1433.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.3675697397,"imp":19.4753043635,"elo":1459.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2385689265,"imp":28.3163667394,"elo":1165.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2679377135,"imp":25.3595907505,"elo":1251.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3667157604,"imp":25.4281836787,"elo":1242.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0454217787,"imp":25.3501447927,"elo":1289.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0696988938,"imp":24.0514724938,"elo":1304.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.3149796315,"imp":23.2824035677,"elo":1349.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0094189939,"imp":37.458040204,"elo":996.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3100216521,"imp":26.4016765352,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5409284286,"imp":24.9707975341,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7424241286,"imp":34.319238724,"elo":994.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4309052415,"imp":25.485756828,"elo":1271.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5856962357,"imp":24.1733149238,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.430301758,"imp":24.8680692608,"elo":1270.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.3668411392,"imp":20.6378379919,"elo":1448.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":6.3267197862,"imp":20.4397728178,"elo":1458.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.4299917889,"imp":15.6833297137,"elo":1546.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":4.4147014618,"imp":21.5010652431,"elo":1503.5,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4819080747,"imp":28.6719370986,"elo":1216.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5959001895,"imp":20.7884586238,"elo":1408.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.0233407175,"imp":20.6212405693,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3196375471,"imp":34.3350576571,"elo":1043.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.573021293,"imp":57.6646273127,"elo":867.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":5.5720509293,"imp":53.7494009931,"elo":880.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":13.3808302986,"imp":52.1868222122,"elo":924.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.14819125,"imp":39.5435437423,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3169111952,"imp":27.0289449493,"elo":1253.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":2.1547385269,"imp":25.6324258404,"elo":1304.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.6818049083,"imp":27.0908356597,"elo":1205.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.7554737985,"imp":16.8214509484,"elo":1554.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.354012373,"imp":15.8471324941,"elo":1581.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8448151087,"imp":13.2008795732,"elo":1628.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2178786994,"imp":12.8993378694,"elo":1723.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.9362255959,"imp":12.0696941109,"elo":1711.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.4841490714,"imp":20.7886960556,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.7189326692,"imp":26.0485639078,"elo":1297.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":30.3506797823,"imp":23.246368468,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":190.5482422391,"imp":22.2340184195,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2641860705,"imp":22.3060516815,"elo":1432.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4463921345,"imp":8.3588435954,"elo":1689.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":1.1673847834,"imp":39.0630184215,"elo":925.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":2.1323059202,"imp":26.5817327119,"elo":1330.3,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2109379604,"imp":16.2487853313,"elo":1563.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.732957515,"imp":22.1820111845,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7493943858,"imp":20.3437763651,"elo":1470.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.5061538419,"imp":19.5722104874,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6817692836,"imp":15.478245996,"elo":1641.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5239137837,"imp":14.920991394,"elo":1579.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4029933935,"imp":21.2342106719,"elo":1449.6,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.6293860489,"imp":21.7459973197,"elo":1471.5,"imputed":true,"imputed_pct":14.2857142857,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":15.0045974485,"imp":19.7737396327,"elo":1559.6,"imputed":true,"imputed_pct":14.2857142857,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.8772316642,"imp":41.3709664653,"elo":879.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.1707473319,"imp":32.6837417968,"elo":1031.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":35.6221458071,"imp":32.5503712331,"elo":1039.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0709157018,"imp":26.2968482458,"elo":1283.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3365764797,"imp":30.9958320334,"elo":1098.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2873908265,"imp":25.9570041275,"elo":1263.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1038316976,"imp":23.5890207169,"elo":1332.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.599155634,"imp":24.5510776102,"elo":1243.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0214870028,"imp":20.976837041,"elo":1414.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":4.5260005112,"imp":21.1251373318,"elo":1405.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":277.1925819839,"imp":30.7955257458,"elo":1038.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":299.3204233014,"imp":22.8668235297,"elo":1306.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1003.6468494405,"imp":21.5188771762,"elo":1398.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":2.1887529296,"imp":28.5701717686,"elo":1160.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.5143737594,"imp":23.8472089109,"elo":1326.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.1408224571,"imp":21.6224374661,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.8133362736,"x_train":684.7823129486,"imp":21.2807580829,"elo":1438.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2489310712,"x_train":29.2262634158,"imp":22.6758278335,"elo":1342.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.216888627,"x_train":3415.6125164831,"imp":20.1819973609,"elo":1433.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.3675697397,"x_train":3415.6125164831,"imp":19.4753043635,"elo":1459.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2385689265,"x_train":8.4538911064,"imp":28.3163667394,"elo":1165.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2679377135,"x_train":2032.7286825124,"imp":25.3595907505,"elo":1251.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3667157604,"x_train":2032.7286825124,"imp":25.4281836787,"elo":1242.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0454217787,"x_train":15.5201326578,"imp":25.3501447927,"elo":1289.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0696988938,"x_train":45754.9282597275,"imp":24.0514724938,"elo":1304.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.3149796315,"x_train":45754.9282597275,"imp":23.2824035677,"elo":1349.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0094189939,"x_train":5.9912431514,"imp":37.458040204,"elo":996.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3100216521,"x_train":1076.4479804348,"imp":26.4016765352,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5409284286,"x_train":1076.4479804348,"imp":24.9707975341,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7424241286,"x_train":8.9709107352,"imp":34.319238724,"elo":994.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4309052415,"x_train":2775.6292399688,"imp":25.485756828,"elo":1271.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5856962357,"x_train":2775.6292399688,"imp":24.1733149238,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.430301758,"x_train":5.3219291517,"imp":24.8680692608,"elo":1270.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.3668411392,"x_train":1502.4464127054,"imp":20.6378379919,"elo":1448.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":6.3267197862,"x_train":1502.4464127054,"imp":20.4397728178,"elo":1458.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.4299917889,"x_train":46.8234528507,"imp":15.6833297137,"elo":1546.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":4.4147014618,"x_train":200.3453678171,"imp":21.5010652431,"elo":1503.5,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4819080747,"x_train":20.9645091386,"imp":28.6719370986,"elo":1216.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5959001895,"x_train":7359.3230736405,"imp":20.7884586238,"elo":1408.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.0233407175,"x_train":7359.3230736405,"imp":20.6212405693,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3196375471,"x_train":22.9524183555,"imp":34.3350576571,"elo":1043.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.573021293,"x_train":295.3582726767,"imp":57.6646273127,"elo":867.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":5.5720509293,"x_train":1838.9438886021,"imp":53.7494009931,"elo":880.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":13.3808302986,"x_train":1838.9438886021,"imp":52.1868222122,"elo":924.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.14819125,"x_train":0.8546409522,"imp":39.5435437423,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3169111952,"x_train":755.269328828,"imp":27.0289449493,"elo":1253.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":2.1547385269,"x_train":755.269328828,"imp":25.6324258404,"elo":1304.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.6818049083,"x_train":19.6357631766,"imp":27.0908356597,"elo":1205.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.7554737985,"x_train":4471.5866070717,"imp":16.8214509484,"elo":1554.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.354012373,"x_train":4471.5866070717,"imp":15.8471324941,"elo":1581.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8448151087,"x_train":11.2142729689,"imp":13.2008795732,"elo":1628.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2178786994,"x_train":2742.7953513314,"imp":12.8993378694,"elo":1723.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.9362255959,"x_train":2742.7953513314,"imp":12.0696941109,"elo":1711.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.4841490714,"x_train":22.5634771017,"imp":20.7886960556,"elo":1387.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":45.7189326692,"x_train":46.2517520732,"imp":26.0485639078,"elo":1297.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":30.3506797823,"x_train":4631.3753499953,"imp":23.246368468,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":190.5482422391,"x_train":4631.3753499953,"imp":22.2340184195,"elo":1457.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2641860705,"x_train":6.7467122095,"imp":22.3060516815,"elo":1432.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4463921345,"x_train":64.2501249469,"imp":8.3588435954,"elo":1689.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":1.1673847834,"x_train":5.8322784268,"imp":39.0630184215,"elo":925.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":2.1323059202,"x_train":11.1754391604,"imp":26.5817327119,"elo":1330.3,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2109379604,"x_train":6.2039007574,"imp":16.2487853313,"elo":1563.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.732957515,"x_train":13.7118099506,"imp":22.1820111845,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7493943858,"x_train":3919.285334302,"imp":20.3437763651,"elo":1470.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.5061538419,"x_train":3919.285334302,"imp":19.5722104874,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6817692836,"x_train":11.1028664446,"imp":15.478245996,"elo":1641.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5239137837,"x_train":12.3203746583,"imp":14.920991394,"elo":1579.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4029933935,"x_train":7.2556258071,"imp":21.2342106719,"elo":1449.6,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.6293860489,"x_train":2987.9489306485,"imp":21.7459973197,"elo":1471.5,"imputed":true,"imputed_pct":14.2857142857,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":15.0045974485,"x_train":2987.9489306485,"imp":19.7737396327,"elo":1559.6,"imputed":true,"imputed_pct":14.2857142857,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.8772316642,"x_train":801.375360716,"imp":41.3709664653,"elo":879.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.1707473319,"x_train":42561.154121861,"imp":32.6837417968,"elo":1031.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":35.6221458071,"x_train":42561.154121861,"imp":32.5503712331,"elo":1039.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0709157018,"x_train":5.0404725813,"imp":26.2968482458,"elo":1283.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3365764797,"x_train":14.7477895176,"imp":30.9958320334,"elo":1098.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2873908265,"x_train":4142.8989350983,"imp":25.9570041275,"elo":1263.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1038316976,"x_train":4142.8989350983,"imp":23.5890207169,"elo":1332.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.599155634,"x_train":5.4806602189,"imp":24.5510776102,"elo":1243.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0214870028,"x_train":2551.6500015348,"imp":20.976837041,"elo":1414.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":4.5260005112,"x_train":2551.6500015348,"imp":21.1251373318,"elo":1405.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":277.1925819839,"x_train":581.4914553919,"imp":30.7955257458,"elo":1038.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":299.3204233014,"x_train":34415.8855588805,"imp":22.8668235297,"elo":1306.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1003.6468494405,"x_train":34415.8855588805,"imp":21.5188771762,"elo":1398.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":2.1887529296,"x_train":10.8402948127,"imp":28.5701717686,"elo":1160.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.5143737594,"x_train":2097.6180059154,"imp":23.8472089109,"elo":1326.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.1408224571,"x_train":2097.6180059154,"imp":21.6224374661,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
index 8daea71023cf2500c001c5f8b0e94c25481ea5ec..3b6bee666a7b86a347003337a3bf9dcdf0985e0c 100644
--- a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":684.7823129486,"imp":21.3040178624,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":29.2262634158,"imp":22.698090803,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":61.1709820217,"imp":20.9301537023,"elo":1412.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":144.6176954198,"imp":20.2350843029,"elo":1445.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":300.2328281021,"imp":19.9018352746,"elo":1459.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":478.2713322387,"imp":19.6032092247,"elo":1472.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":841.9181732231,"imp":19.5953752863,"elo":1471.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1705.1687538882,"imp":19.5347790329,"elo":1473.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2553.0942626612,"imp":19.5340616102,"elo":1473.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":3415.6125164831,"imp":19.4985550531,"elo":1472.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":8.4538911064,"imp":28.3388491346,"elo":1163.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":18.6841630733,"imp":27.9358880912,"elo":1178.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":48.4911153287,"imp":27.6839157051,"elo":1185.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":98.5757062009,"imp":27.2762687405,"elo":1198.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":250.870284713,"imp":26.4848529439,"elo":1215.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":503.8763822777,"imp":26.1448061572,"elo":1225.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1010.9316172533,"imp":25.7102253562,"elo":1236.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1516.2919521693,"imp":25.5436888931,"elo":1241.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2032.7286825124,"imp":25.4516646839,"elo":1245.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":15.5201326578,"imp":25.3725842887,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":263.0355881156,"imp":25.0859729035,"elo":1296.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":710.2172146253,"imp":24.4873258311,"elo":1314.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":1974.8947417325,"imp":24.2588619937,"elo":1323.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":5239.4557597159,"imp":23.894884978,"elo":1332.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":11351.8958788799,"imp":23.6458680692,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":22821.1893610533,"imp":23.4546332378,"elo":1349.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":34610.401754972,"imp":23.365313374,"elo":1354.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":45754.9282597275,"imp":23.3058321435,"elo":1356.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":5.9912431514,"imp":37.4780839474,"elo":990.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":11.3518702812,"imp":33.511614361,"elo":1069.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":27.5426636663,"imp":28.0940350019,"elo":1184.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":54.3053391392,"imp":26.744146511,"elo":1262.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":134.5396659696,"imp":25.5809859765,"elo":1338.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":268.4042300737,"imp":25.3831334267,"elo":1346.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":535.5515141355,"imp":25.1850920695,"elo":1354.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":803.2668643965,"imp":25.0472242317,"elo":1361.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1076.4479804348,"imp":24.9876071642,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.9709107352,"imp":34.3413773154,"elo":993.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":23.3777616357,"imp":32.1655773851,"elo":1058.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":65.8044502368,"imp":28.4345022654,"elo":1175.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":132.7278242503,"imp":26.9486858129,"elo":1236.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":338.4412410661,"imp":25.2826373771,"elo":1299.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":684.2594302606,"imp":24.6109481407,"elo":1328.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1371.7911574391,"imp":24.338217485,"elo":1340.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2064.6207054809,"imp":24.2454840363,"elo":1344.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2775.6292399688,"imp":24.1932773393,"elo":1345.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":5.3219291517,"imp":24.8908345841,"elo":1276.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":12.9118030008,"imp":23.5996522198,"elo":1334.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":34.395160853,"imp":22.6605269237,"elo":1381.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":71.4857515779,"imp":22.075244788,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":185.2078470973,"imp":21.4142265892,"elo":1433.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":372.6794951088,"imp":21.1207890831,"elo":1448.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":746.9586248726,"imp":20.7546975675,"elo":1465.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":1118.5590552235,"imp":20.6174370436,"elo":1470.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1502.4464127054,"imp":20.4605528805,"elo":1477.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":46.8234528507,"imp":15.7509483527,"elo":1565.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":200.3453678171,"imp":21.5259249871,"elo":1523.7,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":20.9645091386,"imp":28.6938258154,"elo":1219.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":58.5698168814,"imp":26.0311592328,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":173.2669344789,"imp":22.9333240351,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":350.5420088728,"imp":22.1754503329,"elo":1389.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":860.8181117489,"imp":21.3481577547,"elo":1424.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1818.8588293534,"imp":21.0598793935,"elo":1430.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3621.9510865846,"imp":21.0645555216,"elo":1427.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":5444.0858786703,"imp":20.7952108809,"elo":1426.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":7359.3230736405,"imp":20.6445803127,"elo":1423.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":22.9524183555,"imp":34.3485228794,"elo":1040.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":295.3582726767,"imp":57.735027905,"elo":868.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":606.9204093001,"imp":56.6708094871,"elo":883.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1533.0305097908,"imp":52.6476398782,"elo":916.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1838.9438886021,"imp":52.1953844998,"elo":924.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.8546409522,"imp":39.6236334572,"elo":991.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.9583883142,"imp":32.3907406412,"elo":1089.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":16.3951125276,"imp":28.9030159445,"elo":1150.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":35.2860398048,"imp":27.1758599017,"elo":1207.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":91.6231833671,"imp":26.1081895731,"elo":1295.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":185.806167626,"imp":25.9112904574,"elo":1306.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":374.1216863756,"imp":25.7728355782,"elo":1313.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":562.8425864643,"imp":25.7099771799,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":755.269328828,"imp":25.6501728215,"elo":1316.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":19.6357631766,"imp":27.1148926763,"elo":1206.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":44.7978169276,"imp":21.9335294716,"elo":1377.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":111.6196798565,"imp":19.0086631985,"elo":1491.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":222.0642460075,"imp":18.0078020556,"elo":1531.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":554.5469212766,"imp":17.2056533218,"elo":1566.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1115.1182819708,"imp":16.7287921997,"elo":1588.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2225.9959788897,"imp":16.3536261286,"elo":1598.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3338.8866547043,"imp":16.095748958,"elo":1606.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4471.5866070717,"imp":15.8711456345,"elo":1607.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":11.2142729689,"imp":13.2264779957,"elo":1641.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":25.4079172215,"imp":13.4582862714,"elo":1648.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":64.4678080693,"imp":13.2557802561,"elo":1679.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":130.8710194046,"imp":12.9773345815,"elo":1737.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":334.8480217005,"imp":12.8660816628,"elo":1747.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":675.6180238549,"imp":12.6533788071,"elo":1746.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1365.6295149721,"imp":12.5840331667,"elo":1738.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2047.0350964005,"imp":12.4495005378,"elo":1736.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2742.7953513314,"imp":12.098404696,"elo":1731.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":22.5634771017,"imp":20.8115874148,"elo":1394.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.2517520732,"imp":26.0642959493,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":64.0415875229,"imp":25.6960888556,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":123.7092045238,"imp":24.9491296147,"elo":1364.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":248.3053487257,"imp":24.42494452,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":584.7096529094,"imp":23.6268640772,"elo":1422.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1166.0242045586,"imp":23.0774644798,"elo":1443.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2315.5296763529,"imp":22.6501959522,"elo":1463.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3443.7150532739,"imp":22.4447600602,"elo":1470.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4631.3753499953,"imp":22.2531442842,"elo":1475.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":6.7467122095,"imp":22.3240280597,"elo":1445.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":64.2501249469,"imp":8.3833123487,"elo":1696.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":5.8322784268,"imp":39.0732386095,"elo":920.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":11.1754391604,"imp":26.602760686,"elo":1341.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":6.2039007574,"imp":16.2740620534,"elo":1575.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":13.7118099506,"imp":22.2067461283,"elo":1411.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":31.4151026664,"imp":21.745881225,"elo":1437.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":93.9419399758,"imp":21.1192050171,"elo":1466.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":192.8564173001,"imp":20.6628797353,"elo":1486.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":479.3804315766,"imp":20.2664141863,"elo":1502.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":961.3391911829,"imp":19.9584890265,"elo":1517.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1926.418861227,"imp":19.7620727971,"elo":1524.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2910.0650077596,"imp":19.6829376667,"elo":1527.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3919.285334302,"imp":19.5958539643,"elo":1532.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":11.1028664446,"imp":15.5035967537,"elo":1663.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":12.3203746583,"imp":14.9446944996,"elo":1594.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":7.2556258071,"imp":21.2594590898,"elo":1463.8,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":28.4871428687,"imp":21.2196854939,"elo":1488.7,"imputed":true,"imputed_pct":14.2857142857,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":70.2455462386,"imp":20.8561477661,"elo":1525.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":147.8920298682,"imp":20.5468833394,"elo":1542.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":366.1171890068,"imp":20.1068219105,"elo":1568.9,"imputed":true,"imputed_pct":14.2857142857,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":733.8269234633,"imp":19.901225313,"elo":1581.9,"imputed":true,"imputed_pct":14.2857142857,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1483.9217307082,"imp":19.8177371432,"elo":1585.0,"imputed":true,"imputed_pct":14.2857142857,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2225.2202634795,"imp":19.7859214434,"elo":1583.8,"imputed":true,"imputed_pct":14.2857142857,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2987.9489306485,"imp":19.7643422057,"elo":1582.9,"imputed":true,"imputed_pct":14.2857142857,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":801.375360716,"imp":41.3913609539,"elo":867.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1026.0399950622,"imp":41.1478806966,"elo":878.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1364.8029707909,"imp":40.6254334929,"elo":892.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2058.9341404243,"imp":39.5401900732,"elo":917.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5065.9505852434,"imp":37.2246997054,"elo":954.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":10436.7196251803,"imp":34.895052535,"elo":988.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":21636.9871111389,"imp":33.6454381282,"elo":1015.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":32013.360321672,"imp":33.0152738101,"elo":1033.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":42561.154121861,"imp":32.5723772194,"elo":1049.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":5.0404725813,"imp":26.314991592,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.7477895176,"imp":31.0184661355,"elo":1097.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":31.2590378107,"imp":28.5219340978,"elo":1167.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":101.8350811731,"imp":26.505101496,"elo":1235.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":218.6755099117,"imp":25.3987052504,"elo":1287.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":533.9506807148,"imp":24.5487478102,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1017.6084144692,"imp":24.1231334151,"elo":1327.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1968.6547247706,"imp":23.8645577721,"elo":1333.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3052.495945877,"imp":23.7035153372,"elo":1341.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4142.8989350983,"imp":23.6148992777,"elo":1343.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":5.4806602189,"imp":24.5728595288,"elo":1241.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":18.5840857617,"imp":23.2622803841,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":57.5257824541,"imp":22.7814069538,"elo":1335.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":117.1284034437,"imp":22.307255383,"elo":1357.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":309.9745400782,"imp":21.944554007,"elo":1377.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":629.246307752,"imp":21.6899995516,"elo":1390.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":1266.4892033336,"imp":21.4864927143,"elo":1401.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":1899.605994117,"imp":21.3325709346,"elo":1410.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":2551.6500015348,"imp":21.148261171,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":581.4914553919,"imp":30.8159974277,"elo":1031.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1918.622222682,"imp":27.9032299797,"elo":1120.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":5770.5281133302,"imp":24.9227560968,"elo":1258.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":12892.6565755789,"imp":23.519143248,"elo":1327.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":32775.6937551607,"imp":21.5800848513,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":34415.8855588805,"imp":21.5424308287,"elo":1414.7,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":10.8402948127,"imp":28.5895053899,"elo":1158.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":22.047469012,"imp":26.7098611977,"elo":1230.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":52.9985363102,"imp":24.5649941172,"elo":1332.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":105.4364802264,"imp":23.6624468284,"elo":1366.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":261.4235869814,"imp":22.5806132908,"elo":1405.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":521.852845102,"imp":22.1013669996,"elo":1426.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":1043.5013351397,"imp":21.7797753504,"elo":1438.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1565.3955882162,"imp":21.6968910307,"elo":1441.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2097.6180059154,"imp":21.6442300933,"elo":1443.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.8133362736,"x_train":684.7823129486,"imp":21.3040178624,"elo":1448.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2489310712,"x_train":29.2262634158,"imp":22.698090803,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3979294404,"x_train":61.1709820217,"imp":20.9301537023,"elo":1412.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.602962146,"x_train":144.6176954198,"imp":20.2350843029,"elo":1445.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.7875363724,"x_train":300.2328281021,"imp":19.9018352746,"elo":1459.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.0219051485,"x_train":478.2713322387,"imp":19.6032092247,"elo":1472.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1647414856,"x_train":841.9181732231,"imp":19.5953752863,"elo":1471.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2641468505,"x_train":1705.1687538882,"imp":19.5347790329,"elo":1473.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.3046821467,"x_train":2553.0942626612,"imp":19.5340616102,"elo":1473.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.3675697397,"x_train":3415.6125164831,"imp":19.4985550531,"elo":1472.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2385689265,"x_train":8.4538911064,"imp":28.3388491346,"elo":1163.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3630820441,"x_train":18.6841630733,"imp":27.9358880912,"elo":1178.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4757621068,"x_train":48.4911153287,"imp":27.6839157051,"elo":1185.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.6044354328,"x_train":98.5757062009,"imp":27.2762687405,"elo":1198.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.7940641756,"x_train":250.870284713,"imp":26.4848529439,"elo":1215.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.9459377238,"x_train":503.8763822777,"imp":26.1448061572,"elo":1225.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1537015916,"x_train":1010.9316172533,"imp":25.7102253562,"elo":1236.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2872609782,"x_train":1516.2919521693,"imp":25.5436888931,"elo":1241.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3667157604,"x_train":2032.7286825124,"imp":25.4516646839,"elo":1245.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0454217787,"x_train":15.5201326578,"imp":25.3725842887,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.1056375523,"x_train":263.0355881156,"imp":25.0859729035,"elo":1296.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.1472799815,"x_train":710.2172146253,"imp":24.4873258311,"elo":1314.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.1808874215,"x_train":1974.8947417325,"imp":24.2588619937,"elo":1323.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.250317583,"x_train":5239.4557597159,"imp":23.894884978,"elo":1332.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.2758121216,"x_train":11351.8958788799,"imp":23.6458680692,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.3045866294,"x_train":22821.1893610533,"imp":23.4546332378,"elo":1349.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.3279012754,"x_train":34610.401754972,"imp":23.365313374,"elo":1354.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.3149796315,"x_train":45754.9282597275,"imp":23.3058321435,"elo":1356.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0094189939,"x_train":5.9912431514,"imp":37.4780839474,"elo":990.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.315155335,"x_train":11.3518702812,"imp":33.511614361,"elo":1069.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3571607009,"x_train":27.5426636663,"imp":28.0940350019,"elo":1184.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3746513091,"x_train":54.3053391392,"imp":26.744146511,"elo":1262.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4170856465,"x_train":134.5396659696,"imp":25.5809859765,"elo":1338.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4416512416,"x_train":268.4042300737,"imp":25.3831334267,"elo":1346.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4692731875,"x_train":535.5515141355,"imp":25.1850920695,"elo":1354.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5288396415,"x_train":803.2668643965,"imp":25.0472242317,"elo":1361.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5409284286,"x_train":1076.4479804348,"imp":24.9876071642,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7424241286,"x_train":8.9709107352,"imp":34.3413773154,"elo":993.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7463780049,"x_train":23.3777616357,"imp":32.1655773851,"elo":1058.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.6779868038,"x_train":65.8044502368,"imp":28.4345022654,"elo":1175.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0030791892,"x_train":132.7278242503,"imp":26.9486858129,"elo":1236.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.8189974652,"x_train":338.4412410661,"imp":25.2826373771,"elo":1299.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.1925971466,"x_train":684.2594302606,"imp":24.6109481407,"elo":1328.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":10.1501412354,"x_train":1371.7911574391,"imp":24.338217485,"elo":1340.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":11.6323451448,"x_train":2064.6207054809,"imp":24.2454840363,"elo":1344.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5427574154,"x_train":2775.6292399688,"imp":24.1932773393,"elo":1345.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.430301758,"x_train":5.3219291517,"imp":24.8908345841,"elo":1276.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6903836077,"x_train":12.9118030008,"imp":23.5996522198,"elo":1334.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.5870394193,"x_train":34.395160853,"imp":22.6605269237,"elo":1381.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.7196389123,"x_train":71.4857515779,"imp":22.075244788,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.1691382103,"x_train":185.2078470973,"imp":21.4142265892,"elo":1433.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.8740373686,"x_train":372.6794951088,"imp":21.1207890831,"elo":1448.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.1746943313,"x_train":746.9586248726,"imp":20.7546975675,"elo":1465.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.9914919439,"x_train":1118.5590552235,"imp":20.6174370436,"elo":1470.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.2770589207,"x_train":1502.4464127054,"imp":20.4605528805,"elo":1477.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":7.4299917889,"x_train":46.8234528507,"imp":15.7509483527,"elo":1565.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":4.4147014618,"x_train":200.3453678171,"imp":21.5259249871,"elo":1523.7,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4819080747,"x_train":20.9645091386,"imp":28.6938258154,"elo":1219.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9993489222,"x_train":58.5698168814,"imp":26.0311592328,"elo":1263.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.0885736068,"x_train":173.2669344789,"imp":22.9333240351,"elo":1356.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9464963206,"x_train":350.5420088728,"imp":22.1754503329,"elo":1389.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.4544013304,"x_train":860.8181117489,"imp":21.3481577547,"elo":1424.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.770076796,"x_train":1818.8588293534,"imp":21.0598793935,"elo":1430.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.3813327083,"x_train":3621.9510865846,"imp":21.0645555216,"elo":1427.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.6018084824,"x_train":5444.0858786703,"imp":20.7952108809,"elo":1426.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.0233407175,"x_train":7359.3230736405,"imp":20.6445803127,"elo":1423.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3196375471,"x_train":22.9524183555,"imp":34.3485228794,"elo":1040.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.573021293,"x_train":295.3582726767,"imp":57.735027905,"elo":868.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":5.6054982404,"x_train":606.9204093001,"imp":56.6708094871,"elo":883.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":12.7962153938,"x_train":1533.0305097908,"imp":52.6476398782,"elo":916.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":13.3808302986,"x_train":1838.9438886021,"imp":52.1953844998,"elo":924.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.14819125,"x_train":0.8546409522,"imp":39.6236334572,"elo":991.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5196360618,"x_train":4.9583883142,"imp":32.3907406412,"elo":1089.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9536404312,"x_train":16.3951125276,"imp":28.9030159445,"elo":1150.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2230254852,"x_train":35.2860398048,"imp":27.1758599017,"elo":1207.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4290952433,"x_train":91.6231833671,"imp":26.1081895731,"elo":1295.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7349630625,"x_train":185.806167626,"imp":25.9112904574,"elo":1306.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.8945842103,"x_train":374.1216863756,"imp":25.7728355782,"elo":1313.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":2.0822348565,"x_train":562.8425864643,"imp":25.7099771799,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":2.1678728195,"x_train":755.269328828,"imp":25.6501728215,"elo":1316.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.6818049083,"x_train":19.6357631766,"imp":27.1148926763,"elo":1206.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.9471265792,"x_train":44.7978169276,"imp":21.9335294716,"elo":1377.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.2236776102,"x_train":111.6196798565,"imp":19.0086631985,"elo":1491.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.3629273246,"x_train":222.0642460075,"imp":18.0078020556,"elo":1531.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.0306005118,"x_train":554.5469212766,"imp":17.2056533218,"elo":1566.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.9685177259,"x_train":1115.1182819708,"imp":16.7287921997,"elo":1588.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.5293488548,"x_train":2225.9959788897,"imp":16.3536261286,"elo":1598.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.4116068115,"x_train":3338.8866547043,"imp":16.095748958,"elo":1606.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.354012373,"x_train":4471.5866070717,"imp":15.8711456345,"elo":1607.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8448151087,"x_train":11.2142729689,"imp":13.2264779957,"elo":1641.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3418798785,"x_train":25.4079172215,"imp":13.4582862714,"elo":1648.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2164999438,"x_train":64.4678080693,"imp":13.2557802561,"elo":1679.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.2967203027,"x_train":130.8710194046,"imp":12.9773345815,"elo":1737.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.2104520673,"x_train":334.8480217005,"imp":12.8660816628,"elo":1747.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8500761173,"x_train":675.6180238549,"imp":12.6533788071,"elo":1746.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4475043375,"x_train":1365.6295149721,"imp":12.5840331667,"elo":1738.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8932296593,"x_train":2047.0350964005,"imp":12.4495005378,"elo":1736.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.9362255959,"x_train":2742.7953513314,"imp":12.098404696,"elo":1731.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.4841490714,"x_train":22.5634771017,"imp":20.8115874148,"elo":1394.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":45.7189326692,"x_train":46.2517520732,"imp":26.0642959493,"elo":1302.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":50.1993286798,"x_train":64.0415875229,"imp":25.6960888556,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":71.5877750075,"x_train":123.7092045238,"imp":24.9491296147,"elo":1364.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":93.1265328822,"x_train":248.3053487257,"imp":24.42494452,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":112.9531593741,"x_train":584.7096529094,"imp":23.6268640772,"elo":1422.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":141.5252995103,"x_train":1166.0242045586,"imp":23.0774644798,"elo":1443.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":180.2303493058,"x_train":2315.5296763529,"imp":22.6501959522,"elo":1463.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":188.8136842543,"x_train":3443.7150532739,"imp":22.4447600602,"elo":1470.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":190.4954431622,"x_train":4631.3753499953,"imp":22.2531442842,"elo":1475.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2641860705,"x_train":6.7467122095,"imp":22.3240280597,"elo":1445.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.4463921345,"x_train":64.2501249469,"imp":8.3833123487,"elo":1696.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":1.1673847834,"x_train":5.8322784268,"imp":39.0732386095,"elo":920.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":2.1323059202,"x_train":11.1754391604,"imp":26.602760686,"elo":1341.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2109379604,"x_train":6.2039007574,"imp":16.2740620534,"elo":1575.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.732957515,"x_train":13.7118099506,"imp":22.2067461283,"elo":1411.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":1.251031111,"x_train":31.4151026664,"imp":21.745881225,"elo":1437.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":2.1831848653,"x_train":93.9419399758,"imp":21.1192050171,"elo":1466.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.5316249023,"x_train":192.8564173001,"imp":20.6628797353,"elo":1486.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.9885413426,"x_train":479.3804315766,"imp":20.2664141863,"elo":1502.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.2794460528,"x_train":961.3391911829,"imp":19.9584890265,"elo":1517.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.8992617594,"x_train":1926.418861227,"imp":19.7620727971,"elo":1524.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.3258184146,"x_train":2910.0650077596,"imp":19.6829376667,"elo":1527.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.5061538419,"x_train":3919.285334302,"imp":19.5958539643,"elo":1532.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6817692836,"x_train":11.1028664446,"imp":15.5035967537,"elo":1663.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5239137837,"x_train":12.3203746583,"imp":14.9446944996,"elo":1594.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4029933935,"x_train":7.2556258071,"imp":21.2594590898,"elo":1463.8,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":2.433582774,"x_train":28.4871428687,"imp":21.2196854939,"elo":1488.7,"imputed":true,"imputed_pct":14.2857142857,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.2862798512,"x_train":70.2455462386,"imp":20.8561477661,"elo":1525.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":6.8315989414,"x_train":147.8920298682,"imp":20.5468833394,"elo":1542.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":8.2026307059,"x_train":366.1171890068,"imp":20.1068219105,"elo":1568.9,"imputed":true,"imputed_pct":14.2857142857,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":9.9621575083,"x_train":733.8269234633,"imp":19.901225313,"elo":1581.9,"imputed":true,"imputed_pct":14.2857142857,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.1635180712,"x_train":1483.9217307082,"imp":19.8177371432,"elo":1585.0,"imputed":true,"imputed_pct":14.2857142857,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":13.9747772779,"x_train":2225.2202634795,"imp":19.7859214434,"elo":1583.8,"imputed":true,"imputed_pct":14.2857142857,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":15.0045974485,"x_train":2987.9489306485,"imp":19.7643422057,"elo":1582.9,"imputed":true,"imputed_pct":14.2857142857,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.8772316642,"x_train":801.375360716,"imp":41.3913609539,"elo":867.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.4381519904,"x_train":1026.0399950622,"imp":41.1478806966,"elo":878.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.1527950165,"x_train":1364.8029707909,"imp":40.6254334929,"elo":892.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.9581917636,"x_train":2058.9341404243,"imp":39.5401900732,"elo":917.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.771475197,"x_train":5065.9505852434,"imp":37.2246997054,"elo":954.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.9040613643,"x_train":10436.7196251803,"imp":34.895052535,"elo":988.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":29.4903892803,"x_train":21636.9871111389,"imp":33.6454381282,"elo":1015.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.6709522759,"x_train":32013.360321672,"imp":33.0152738101,"elo":1033.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":35.6221458071,"x_train":42561.154121861,"imp":32.5723772194,"elo":1049.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0709157018,"x_train":5.0404725813,"imp":26.314991592,"elo":1291.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3365764797,"x_train":14.7477895176,"imp":31.0184661355,"elo":1097.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6499035356,"x_train":31.2590378107,"imp":28.5219340978,"elo":1167.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0476437019,"x_train":101.8350811731,"imp":26.505101496,"elo":1235.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2034747343,"x_train":218.6755099117,"imp":25.3987052504,"elo":1287.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4189325906,"x_train":533.9506807148,"imp":24.5487478102,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6704753472,"x_train":1017.6084144692,"imp":24.1231334151,"elo":1327.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8715186479,"x_train":1968.6547247706,"imp":23.8645577721,"elo":1333.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0005251114,"x_train":3052.495945877,"imp":23.7035153372,"elo":1341.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.096613965,"x_train":4142.8989350983,"imp":23.6148992777,"elo":1343.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.599155634,"x_train":5.4806602189,"imp":24.5728595288,"elo":1241.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.0378638981,"x_train":18.5840857617,"imp":23.2622803841,"elo":1307.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.8012686319,"x_train":57.5257824541,"imp":22.7814069538,"elo":1335.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.9499124215,"x_train":117.1284034437,"imp":22.307255383,"elo":1357.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.0226060034,"x_train":309.9745400782,"imp":21.944554007,"elo":1377.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.442578599,"x_train":629.246307752,"imp":21.6899995516,"elo":1390.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.7654993382,"x_train":1266.4892033336,"imp":21.4864927143,"elo":1401.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.559561221,"x_train":1899.605994117,"imp":21.3325709346,"elo":1410.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.5236730663,"x_train":2551.6500015348,"imp":21.148261171,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":277.1925819839,"x_train":581.4914553919,"imp":30.8159974277,"elo":1031.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":597.2305576187,"x_train":1918.622222682,"imp":27.9032299797,"elo":1120.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":953.3210135052,"x_train":5770.5281133302,"imp":24.9227560968,"elo":1258.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":968.9952282534,"x_train":12892.6565755789,"imp":23.519143248,"elo":1327.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":993.7285537193,"x_train":32775.6937551607,"imp":21.5800848513,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":1003.6468494405,"x_train":34415.8855588805,"imp":21.5424308287,"elo":1414.7,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":2.1887529296,"x_train":10.8402948127,"imp":28.5895053899,"elo":1158.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":2.4383109526,"x_train":22.047469012,"imp":26.7098611977,"elo":1230.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.3346191076,"x_train":52.9985363102,"imp":24.5649941172,"elo":1332.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":2.1631624605,"x_train":105.4364802264,"imp":23.6624468284,"elo":1366.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.9055625999,"x_train":261.4235869814,"imp":22.5806132908,"elo":1405.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2090360064,"x_train":521.852845102,"imp":22.1013669996,"elo":1426.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.4238113379,"x_train":1043.5013351397,"imp":21.7797753504,"elo":1438.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.7252746839,"x_train":1565.3955882162,"imp":21.6968910307,"elo":1441.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.1516651241,"x_train":2097.6180059154,"imp":21.6442300933,"elo":1443.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..584a0f0c29584c520bdf1c0316c8c33c5b71c7c4
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_multiclass/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (default)", "TabPFN-2.6 (default)", "TabPFN-3 (default)", "TabICLv2 (default)", "TabPFNv2 (tuned + ensembled)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "LimiX (default)", "Mitra (default)", "TabPFNv2 (tuned)", "TabDPT (tuned + ensembled)", "TabDPT-Turbo (default)", "TabM (tuned + ensembled)", "TabPFNv2 (default)", "AutoGluon 1.4 (best, 4h)", "TabDPT (tuned)", "TabM (tuned)", "CatBoost (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "CatBoost (tuned)", "ModernNCA (tuned)", "XGBoost (tuned)", "XGBoost (tuned + ensembled)", "TabICL (default)", "LightGBM (tuned + ensembled)", "iLTM (tuned + ensembled)", "LightGBM (tuned)", "xRFM (tuned + ensembled)", "BetaTabPFN (default)", "TabM (default)", "ExtraTrees (tuned + ensembled)", "SAP-RPT-OSS (default)", "TabDPT (default)", "TorchMLP (tuned + ensembled)", "EBM (tuned + ensembled)", "iLTM (tuned)", "RandomForest (tuned + ensembled)", "xRFM (tuned)", "CatBoost (default)", "ExtraTrees (tuned)", "TabSwift (default)", "EBM (tuned)", "FastaiMLP (tuned + ensembled)", "ModernNCA (default)", "RandomForest (tuned)", "TorchMLP (tuned)", "XGBoost (default)", "EBM (default)", "FastaiMLP (tuned)", "RealMLP (default)", "LightGBM (default)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "xRFM (default)", "TorchMLP (default)", "KNN (tuned + ensembled)", "ChimeraBoost (default)", "OrionMSP (default)", "RandomForest (default)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "iLTM (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "PerpetualBooster (tuned)", "PerpetualBooster (default)", "FastaiMLP (default)", "TabFlex (default)", "KNN (tuned)", "TabSTAR (default)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.7006802558898926, 0.7006802558898926, 0.8163265585899353, 0.7414966225624084, 0.8367347121238708, 0.8503401279449463, 0.8911564350128174, 0.8299319744110107, 0.7687074542045593, 0.7687074542045593, 0.8435373902320862, 0.918367326259613, 0.9115646481513977, 0.8979591727256775, 0.884353756904602, 0.8095238208770752, 0.9115646481513977, 0.9251700639724731, 0.884353756904602, 0.8231292366981506, 0.9047619104385376, 0.918367326259613, 0.9251700639724731, 0.8911564350128174, 0.8775510191917419, 0.884353756904602, 0.9727891087532043, 0.9047619104385376, 0.918367326259613, 0.8979591727256775, 0.918367326259613, 0.9115646481513977, 0.8503401279449463, 0.9251700639724731, 0.7551020383834839, 0.9047619104385376, 0.9251700639724731, 0.9047619104385376, 0.9319728016853333, 0.9047619104385376, 0.9251700639724731, 0.9319728016853333, 0.918367326259613, 0.918367326259613, 0.9251700639724731, 0.8979591727256775, 0.9795918464660645, 0.9047619104385376, 0.9319728016853333, 0.9319728016853333, 0.9251700639724731, 0.9387755393981934, 0.9795918464660645, 0.9251700639724731, 0.9251700639724731, 0.9251700639724731, 0.9251700639724731, 0.9319728016853333, 0.9319728016853333, 0.9387755393981934, 0.9319728016853333, 0.9931972622871399, 1.0, 0.9931972622871399, 0.9931972622871399, 0.9387755393981934, 0.9387755393981934, 0.9319728016853333, 0.9931972622871399, 0.9727891087532043, 0.9659863710403442, 1.0, 0.9659863710403442, 0.9319728016853333, 0.9931972622871399, 1.0, 1.0], [0.29931971430778503, NaN, 0.4625850319862366, 0.5782312750816345, 0.6836734414100647, 0.6190476417541504, 0.6190476417541504, 0.6530612111091614, 0.7074829936027527, 0.7414966225624084, 0.7687074542045593, 0.7346938848495483, 0.7074829936027527, 0.8027210831642151, 0.7278911471366882, 0.6938775777816772, 0.8435373902320862, 0.8775510191917419, 0.884353756904602, 0.7755101919174194, 0.8503401279449463, 0.9251700639724731, 0.8503401279449463, 0.9251700639724731, 0.8639456033706665, 0.9047619104385376, 0.9047619104385376, 0.7891156673431396, 0.884353756904602, 0.8775510191917419, 0.9115646481513977, 0.8911564350128174, 0.8979591727256775, 0.8571428656578064, 0.8639456033706665, 0.7482993006706238, 0.795918345451355, 0.9455782175064087, 0.9523809552192688, 0.9523809552192688, 0.8775510191917419, 0.918367326259613, 0.9659863710403442, 0.8911564350128174, 0.9727891087532043, 0.9591836929321289, 0.9251700639724731, 0.9455782175064087, 0.9047619104385376, 0.9455782175064087, 0.9795918464660645, 0.9455782175064087, 0.9659863710403442, 0.9931972622871399, 0.9455782175064087, 0.9523809552192688, 0.9659863710403442, 0.9795918464660645, 0.9931972622871399, 0.9319728016853333, 0.9591836929321289, 0.9795918464660645, 0.9863945841789246, 0.9931972622871399, 0.9931972622871399, 1.0, 1.0, 0.9863945841789246, 0.9863945841789246, 0.9863945841789246, 0.9727891087532043, 0.9659863710403442, 1.0, 1.0, 0.9659863710403442, 0.9863945841789246, 1.0, 1.0], [0.29931971430778503, 0.5374149680137634, NaN, 0.6054421663284302, 0.7482993006706238, 0.6258503198623657, 0.6258503198623657, 0.6598639488220215, 0.7551020383834839, 0.7346938848495483, 0.7482993006706238, 0.7414966225624084, 0.7074829936027527, 0.795918345451355, 0.7210884094238281, 0.6870748400688171, 0.795918345451355, 0.8707482814788818, 0.8571428656578064, 0.7414966225624084, 0.795918345451355, 0.9387755393981934, 0.8571428656578064, 0.9251700639724731, 0.8775510191917419, 0.8979591727256775, 0.918367326259613, 0.7891156673431396, 0.8911564350128174, 0.884353756904602, 0.8979591727256775, 0.8775510191917419, 0.9115646481513977, 0.8435373902320862, 0.8435373902320862, 0.7414966225624084, 0.7755101919174194, 0.9387755393981934, 0.9319728016853333, 0.9523809552192688, 0.8707482814788818, 0.918367326259613, 0.9387755393981934, 0.8707482814788818, 0.9727891087532043, 0.9387755393981934, 0.9319728016853333, 0.9319728016853333, 0.9047619104385376, 0.9455782175064087, 0.9863945841789246, 0.9319728016853333, 0.9523809552192688, 0.9863945841789246, 0.9319728016853333, 0.9387755393981934, 0.9523809552192688, 0.9659863710403442, 0.9863945841789246, 0.9251700639724731, 0.9387755393981934, 0.9863945841789246, 0.9863945841789246, 0.9931972622871399, 0.9931972622871399, 0.9931972622871399, 1.0, 0.9931972622871399, 0.9931972622871399, 0.9863945841789246, 0.9727891087532043, 0.9727891087532043, 0.9931972622871399, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.18367347121238708, 0.4217686951160431, 0.3945578336715698, NaN, 0.4829931855201721, 0.5986394286155701, 0.5714285969734192, 0.5714285969734192, 0.557823121547699, 0.7210884094238281, 0.7346938848495483, 0.7891156673431396, 0.7142857313156128, 0.6938775777816772, 0.8163265585899353, 0.7346938848495483, 0.7687074542045593, 0.8231292366981506, 0.795918345451355, 0.8571428656578064, 0.761904776096344, 0.884353756904602, 0.8639456033706665, 0.8979591727256775, 0.884353756904602, 0.8299319744110107, 0.8571428656578064, 0.8979591727256775, 0.9047619104385376, 0.9115646481513977, 0.8911564350128174, 0.9047619104385376, 0.8231292366981506, 0.8027210831642151, 0.8979591727256775, 0.6258503198623657, 0.8367347121238708, 0.9047619104385376, 0.918367326259613, 0.9387755393981934, 0.9047619104385376, 0.9387755393981934, 0.9455782175064087, 0.8979591727256775, 0.9863945841789246, 0.9455782175064087, 0.8707482814788818, 0.9659863710403442, 0.9047619104385376, 0.9523809552192688, 0.9931972622871399, 0.9387755393981934, 0.9387755393981934, 0.9795918464660645, 0.9727891087532043, 0.9727891087532043, 0.9863945841789246, 0.9863945841789246, 0.9863945841789246, 0.9795918464660645, 0.9863945841789246, 0.9931972622871399, 0.9931972622871399, 0.9931972622871399, 0.9727891087532043, 0.9795918464660645, 1.0, 0.9931972622871399, 0.9931972622871399, 0.9931972622871399, 0.9727891087532043, 0.9659863710403442, 0.9931972622871399, 1.0, 0.9795918464660645, 0.9931972622871399, 1.0, 1.0], [0.25850340723991394, 0.3163265287876129, 0.25170066952705383, 0.5170068144798279, NaN, 0.4761904776096344, 0.557823121547699, 0.5782312750816345, 0.5510203838348389, 0.6394557952880859, 0.6530612111091614, 0.7278911471366882, 0.6326530575752258, 0.6938775777816772, 0.6598639488220215, 0.6258503198623657, 0.7891156673431396, 0.8503401279449463, 0.7891156673431396, 0.680272102355957, 0.8095238208770752, 0.8639456033706665, 0.7414966225624084, 0.8911564350128174, 0.7891156673431396, 0.8639456033706665, 0.8639456033706665, 0.761904776096344, 0.8503401279449463, 0.795918345451355, 0.8707482814788818, 0.8163265585899353, 0.8503401279449463, 0.8163265585899353, 0.8027210831642151, 0.7142857313156128, 0.7006802558898926, 0.8979591727256775, 0.9047619104385376, 0.8979591727256775, 0.8231292366981506, 0.8707482814788818, 0.9115646481513977, 0.8435373902320862, 0.9319728016853333, 0.9115646481513977, 0.918367326259613, 0.9047619104385376, 0.9115646481513977, 0.8979591727256775, 0.9455782175064087, 0.9251700639724731, 0.9387755393981934, 0.9931972622871399, 0.9115646481513977, 0.9115646481513977, 0.918367326259613, 0.9319728016853333, 0.9591836929321289, 0.8775510191917419, 0.9319728016853333, 0.9863945841789246, 0.9591836929321289, 0.9931972622871399, 1.0, 0.9931972622871399, 0.9863945841789246, 0.9863945841789246, 0.9795918464660645, 0.9863945841789246, 0.9727891087532043, 0.9727891087532043, 0.9931972622871399, 0.9931972622871399, 0.9455782175064087, 1.0, 1.0, 1.0], [0.16326530277729034, 0.380952388048172, 0.3741496503353119, 0.40136054158210754, 0.523809552192688, NaN, 0.45578232407569885, 0.45578232407569885, 0.5306122303009033, 0.6326530575752258, 0.6326530575752258, 0.646258533000946, 0.7142857313156128, 0.646258533000946, 0.6258503198623657, 0.6734693646430969, 0.7142857313156128, 0.7482993006706238, 0.7687074542045593, 0.7006802558898926, 0.7482993006706238, 0.8027210831642151, 0.8571428656578064, 0.8163265585899353, 0.8299319744110107, 0.8775510191917419, 0.8979591727256775, 0.7823129296302795, 0.8503401279449463, 0.9251700639724731, 0.884353756904602, 0.7891156673431396, 0.8367347121238708, 0.7891156673431396, 0.8435373902320862, 0.6530612111091614, 0.7755101919174194, 0.918367326259613, 0.8639456033706665, 0.9455782175064087, 0.8231292366981506, 0.8503401279449463, 0.9523809552192688, 0.8367347121238708, 0.9523809552192688, 0.9047619104385376, 0.9455782175064087, 0.9319728016853333, 0.8639456033706665, 0.918367326259613, 0.9795918464660645, 0.9251700639724731, 0.9795918464660645, 0.9659863710403442, 0.9319728016853333, 0.9319728016853333, 0.9455782175064087, 0.9319728016853333, 0.9455782175064087, 0.9523809552192688, 0.9455782175064087, 0.9659863710403442, 1.0, 1.0, 0.9795918464660645, 0.9863945841789246, 1.0, 0.9931972622871399, 0.9931972622871399, 0.9659863710403442, 0.9523809552192688, 0.9659863710403442, 1.0, 1.0, 0.9591836929321289, 0.9931972622871399, 1.0, 1.0], [0.14965985715389252, 0.380952388048172, 0.3741496503353119, 0.4285714328289032, 0.442176878452301, 0.5442177057266235, NaN, 0.5714285969734192, 0.5850340127944946, 0.523809552192688, 0.6054421663284302, 0.646258533000946, 0.646258533000946, 0.6666666865348816, 0.7278911471366882, 0.7210884094238281, 0.6598639488220215, 0.7074829936027527, 0.6938775777816772, 0.761904776096344, 0.6938775777816772, 0.8231292366981506, 0.8095238208770752, 0.8231292366981506, 0.8231292366981506, 0.7755101919174194, 0.7823129296302795, 0.9319728016853333, 0.8435373902320862, 0.8503401279449463, 0.8367347121238708, 0.8367347121238708, 0.7414966225624084, 0.7006802558898926, 0.9047619104385376, 0.5714285969734192, 0.8163265585899353, 0.8775510191917419, 0.8299319744110107, 0.9115646481513977, 0.8775510191917419, 0.9047619104385376, 0.884353756904602, 0.9115646481513977, 0.9455782175064087, 0.8503401279449463, 0.8299319744110107, 0.9659863710403442, 0.8979591727256775, 0.8911564350128174, 0.9795918464660645, 0.918367326259613, 0.9387755393981934, 0.9863945841789246, 0.9251700639724731, 0.9047619104385376, 0.9251700639724731, 0.9795918464660645, 0.9659863710403442, 0.9455782175064087, 0.9319728016853333, 0.9931972622871399, 1.0, 1.0, 0.9863945841789246, 0.9863945841789246, 0.9931972622871399, 1.0, 0.9795918464660645, 0.9727891087532043, 0.9727891087532043, 0.9659863710403442, 0.9931972622871399, 1.0, 0.9455782175064087, 1.0, 1.0, 1.0], [0.10884353518486023, 0.3469387888908386, 0.3401360511779785, 0.4285714328289032, 0.4217686951160431, 0.5442177057266235, 0.4285714328289032, NaN, 0.5170068144798279, 0.557823121547699, 0.5850340127944946, 0.6190476417541504, 0.6938775777816772, 0.6054421663284302, 0.7346938848495483, 0.6734693646430969, 0.6666666865348816, 0.7006802558898926, 0.7142857313156128, 0.7346938848495483, 0.6734693646430969, 0.7755101919174194, 0.7414966225624084, 0.7823129296302795, 0.8095238208770752, 0.7551020383834839, 0.7687074542045593, 0.9591836929321289, 0.8435373902320862, 0.8707482814788818, 0.8571428656578064, 0.7823129296302795, 0.7346938848495483, 0.7346938848495483, 0.8163265585899353, 0.5102040767669678, 0.7482993006706238, 0.8231292366981506, 0.8299319744110107, 0.8775510191917419, 0.8707482814788818, 0.8163265585899353, 0.884353756904602, 0.8367347121238708, 0.918367326259613, 0.8571428656578064, 0.8163265585899353, 0.9523809552192688, 0.8707482814788818, 0.8775510191917419, 0.9251700639724731, 0.9047619104385376, 0.8979591727256775, 0.9251700639724731, 0.9251700639724731, 0.918367326259613, 0.9251700639724731, 0.8911564350128174, 0.8911564350128174, 0.9319728016853333, 0.9319728016853333, 0.9115646481513977, 0.9931972622871399, 0.9727891087532043, 0.9863945841789246, 0.9795918464660645, 0.9387755393981934, 0.9319728016853333, 0.9319728016853333, 0.9727891087532043, 0.9659863710403442, 0.9659863710403442, 0.9863945841789246, 0.9455782175064087, 0.9387755393981934, 0.9931972622871399, 1.0, 1.0], [0.17006802558898926, 0.2925170063972473, 0.2448979616165161, 0.442176878452301, 0.44897958636283875, 0.4693877696990967, 0.41496598720550537, 0.4829931855201721, NaN, 0.557823121547699, 0.6054421663284302, 0.557823121547699, 0.5918367505073547, 0.6598639488220215, 0.6190476417541504, 0.6054421663284302, 0.761904776096344, 0.6870748400688171, 0.8095238208770752, 0.6598639488220215, 0.7687074542045593, 0.8163265585899353, 0.7210884094238281, 0.8435373902320862, 0.7823129296302795, 0.8299319744110107, 0.8435373902320862, 0.6938775777816772, 0.8639456033706665, 0.8095238208770752, 0.884353756904602, 0.8299319744110107, 0.8639456033706665, 0.8163265585899353, 0.7891156673431396, 0.6190476417541504, 0.680272102355957, 0.8775510191917419, 0.8775510191917419, 0.8979591727256775, 0.8095238208770752, 0.8367347121238708, 0.9319728016853333, 0.8367347121238708, 0.8979591727256775, 0.8911564350128174, 0.8775510191917419, 0.8639456033706665, 0.8435373902320862, 0.8911564350128174, 0.918367326259613, 0.918367326259613, 0.8979591727256775, 0.918367326259613, 0.9251700639724731, 0.9319728016853333, 0.9251700639724731, 0.8911564350128174, 0.9047619104385376, 0.8911564350128174, 0.9319728016853333, 0.9115646481513977, 0.8979591727256775, 0.9387755393981934, 0.9863945841789246, 0.9931972622871399, 0.9387755393981934, 0.9251700639724731, 0.9251700639724731, 0.9727891087532043, 0.9659863710403442, 0.9659863710403442, 0.9387755393981934, 0.9387755393981934, 0.9387755393981934, 0.9319728016853333, 1.0, 1.0], [0.2312925159931183, 0.25850340723991394, 0.26530611515045166, 0.2789115607738495, 0.36054420471191406, 0.36734694242477417, 0.4761904776096344, 0.442176878452301, 0.442176878452301, NaN, 0.5136054158210754, 0.5170068144798279, 0.5442177057266235, 0.5850340127944946, 0.5510203838348389, 0.5646258592605591, 0.7346938848495483, 0.5850340127944946, 0.7074829936027527, 0.557823121547699, 0.7414966225624084, 0.7823129296302795, 0.7278911471366882, 0.8163265585899353, 0.7823129296302795, 0.8979591727256775, 0.9251700639724731, 0.6190476417541504, 0.8775510191917419, 0.8027210831642151, 0.8775510191917419, 0.8231292366981506, 0.8571428656578064, 0.8707482814788818, 0.7891156673431396, 0.6530612111091614, 0.6054421663284302, 0.8027210831642151, 0.918367326259613, 0.9047619104385376, 0.8639456033706665, 0.8503401279449463, 0.9319728016853333, 0.8571428656578064, 0.7891156673431396, 0.9251700639724731, 0.9047619104385376, 0.9455782175064087, 0.9115646481513977, 0.8571428656578064, 0.9931972622871399, 0.9319728016853333, 0.9115646481513977, 0.9931972622871399, 0.9251700639724731, 0.9387755393981934, 0.9455782175064087, 0.9319728016853333, 0.9727891087532043, 0.918367326259613, 0.9387755393981934, 0.9863945841789246, 0.9727891087532043, 1.0, 0.9931972622871399, 0.9863945841789246, 1.0, 0.9727891087532043, 0.9659863710403442, 0.9795918464660645, 0.9727891087532043, 0.9659863710403442, 1.0, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.2312925159931183, 0.2312925159931183, 0.25170066952705383, 0.26530611515045166, 0.3469387888908386, 0.36734694242477417, 0.3945578336715698, 0.41496598720550537, 0.3945578336715698, 0.4863945543766022, NaN, 0.5102040767669678, 0.4761904776096344, 0.5102040767669678, 0.5102040767669678, 0.5510203838348389, 0.6598639488220215, 0.5714285969734192, 0.6530612111091614, 0.5442177057266235, 0.7006802558898926, 0.7482993006706238, 0.7346938848495483, 0.7551020383834839, 0.7551020383834839, 0.8299319744110107, 0.8571428656578064, 0.6054421663284302, 0.7891156673431396, 0.7687074542045593, 0.8027210831642151, 0.761904776096344, 0.8435373902320862, 0.8027210831642151, 0.7823129296302795, 0.6394557952880859, 0.6190476417541504, 0.8503401279449463, 0.8979591727256775, 0.884353756904602, 0.8367347121238708, 0.8299319744110107, 0.8911564350128174, 0.8231292366981506, 0.7891156673431396, 0.9387755393981934, 0.8707482814788818, 0.918367326259613, 0.8571428656578064, 0.8979591727256775, 0.9591836929321289, 0.918367326259613, 0.9319728016853333, 0.9795918464660645, 0.918367326259613, 0.918367326259613, 0.9387755393981934, 0.9047619104385376, 0.9659863710403442, 0.9251700639724731, 0.9455782175064087, 0.9795918464660645, 0.9727891087532043, 0.9795918464660645, 0.9863945841789246, 0.9931972622871399, 0.9931972622871399, 0.9727891087532043, 0.9591836929321289, 0.9863945841789246, 0.9727891087532043, 0.9591836929321289, 1.0, 1.0, 0.9727891087532043, 0.9863945841789246, 1.0, 1.0], [0.15646257996559143, 0.26530611515045166, 0.25850340723991394, 0.21088434755802155, 0.27210885286331177, 0.35374149680137634, 0.35374149680137634, 0.380952388048172, 0.442176878452301, 0.4829931855201721, 0.4897959232330322, NaN, 0.4625850319862366, 0.523809552192688, 0.5170068144798279, 0.4625850319862366, 0.6326530575752258, 0.6122449040412903, 0.6394557952880859, 0.6054421663284302, 0.6394557952880859, 0.7278911471366882, 0.6326530575752258, 0.7414966225624084, 0.7687074542045593, 0.761904776096344, 0.7551020383834839, 0.6666666865348816, 0.7891156673431396, 0.7482993006706238, 0.795918345451355, 0.8367347121238708, 0.7482993006706238, 0.7074829936027527, 0.884353756904602, 0.5850340127944946, 0.6054421663284302, 0.7006802558898926, 0.8231292366981506, 0.8299319744110107, 0.9047619104385376, 0.8979591727256775, 0.8775510191917419, 0.9047619104385376, 0.8095238208770752, 0.8775510191917419, 0.7687074542045593, 0.9387755393981934, 0.8979591727256775, 0.8163265585899353, 0.9455782175064087, 0.8979591727256775, 0.884353756904602, 0.9795918464660645, 0.9387755393981934, 0.9115646481513977, 0.9387755393981934, 0.9727891087532043, 0.9591836929321289, 0.9455782175064087, 0.9455782175064087, 0.9727891087532043, 1.0, 1.0, 0.9863945841789246, 0.9931972622871399, 0.9727891087532043, 0.9795918464660645, 0.9659863710403442, 0.9795918464660645, 0.9727891087532043, 0.9659863710403442, 1.0, 0.9931972622871399, 0.9523809552192688, 1.0, 1.0, 1.0], [0.08163265138864517, 0.2925170063972473, 0.2925170063972473, 0.2857142984867096, 0.36734694242477417, 0.2857142984867096, 0.35374149680137634, 0.30612245202064514, 0.40816327929496765, 0.45578232407569885, 0.523809552192688, 0.5374149680137634, NaN, 0.5646258592605591, 0.5646258592605591, 0.5510203838348389, 0.6190476417541504, 0.6224489808082581, 0.6870748400688171, 0.5986394286155701, 0.646258533000946, 0.7210884094238281, 0.7687074542045593, 0.7142857313156128, 0.7687074542045593, 0.7278911471366882, 0.761904776096344, 0.6700680255889893, 0.8503401279449463, 0.8571428656578064, 0.8367347121238708, 0.7551020383834839, 0.7414966225624084, 0.7142857313156128, 0.7755101919174194, 0.5102040767669678, 0.680272102355957, 0.8367347121238708, 0.8163265585899353, 0.8775510191917419, 0.8163265585899353, 0.8095238208770752, 0.918367326259613, 0.8095238208770752, 0.8775510191917419, 0.8571428656578064, 0.8231292366981506, 0.8775510191917419, 0.8435373902320862, 0.8775510191917419, 0.9319728016853333, 0.9047619104385376, 0.884353756904602, 0.8775510191917419, 0.9387755393981934, 0.9251700639724731, 0.918367326259613, 0.8299319744110107, 0.8775510191917419, 0.9387755393981934, 0.9387755393981934, 0.9047619104385376, 0.9625850319862366, 0.9387755393981934, 0.9795918464660645, 0.9659863710403442, 0.9387755393981934, 0.9251700639724731, 0.9251700639724731, 0.9523809552192688, 0.9523809552192688, 0.9591836929321289, 0.9319728016853333, 0.9387755393981934, 0.9387755393981934, 0.9319728016853333, 1.0, 1.0], [0.08843537420034409, 0.1972789168357849, 0.20408163964748383, 0.30612245202064514, 0.30612245202064514, 0.35374149680137634, 0.3333333432674408, 0.3945578336715698, 0.3401360511779785, 0.41496598720550537, 0.4897959232330322, 0.4761904776096344, 0.4353741407394409, NaN, 0.5850340127944946, 0.557823121547699, 0.6054421663284302, 0.6054421663284302, 0.6938775777816772, 0.6394557952880859, 0.6530612111091614, 0.680272102355957, 0.6394557952880859, 0.7551020383834839, 0.7074829936027527, 0.680272102355957, 0.7006802558898926, 0.6530612111091614, 0.7482993006706238, 0.7142857313156128, 0.761904776096344, 0.7074829936027527, 0.761904776096344, 0.7142857313156128, 0.7210884094238281, 0.5442177057266235, 0.6598639488220215, 0.8299319744110107, 0.8163265585899353, 0.8299319744110107, 0.7346938848495483, 0.8095238208770752, 0.8231292366981506, 0.7687074542045593, 0.8435373902320862, 0.8299319744110107, 0.8095238208770752, 0.8027210831642151, 0.8231292366981506, 0.8707482814788818, 0.8571428656578064, 0.8707482814788818, 0.8639456033706665, 0.9115646481513977, 0.8571428656578064, 0.8639456033706665, 0.8639456033706665, 0.8503401279449463, 0.8979591727256775, 0.8775510191917419, 0.9047619104385376, 0.8911564350128174, 0.8911564350128174, 0.9251700639724731, 0.9659863710403442, 0.9659863710403442, 0.9387755393981934, 0.9047619104385376, 0.8911564350128174, 0.9659863710403442, 0.9591836929321289, 0.9727891087532043, 0.918367326259613, 0.9387755393981934, 0.9387755393981934, 0.9319728016853333, 1.0, 1.0], [0.10204081982374191, 0.27210885286331177, 0.2789115607738495, 0.18367347121238708, 0.3401360511779785, 0.3741496503353119, 0.27210885286331177, 0.26530611515045166, 0.380952388048172, 0.44897958636283875, 0.4897959232330322, 0.4829931855201721, 0.4353741407394409, 0.41496598720550537, NaN, 0.4761904776096344, 0.5442177057266235, 0.5442177057266235, 0.557823121547699, 0.6598639488220215, 0.5510203838348389, 0.5782312750816345, 0.6530612111091614, 0.6122449040412903, 0.680272102355957, 0.6326530575752258, 0.6530612111091614, 0.6734693646430969, 0.7142857313156128, 0.7551020383834839, 0.6938775777816772, 0.6326530575752258, 0.6326530575752258, 0.6326530575752258, 0.7346938848495483, 0.4897959232330322, 0.7482993006706238, 0.680272102355957, 0.7346938848495483, 0.7891156673431396, 0.795918345451355, 0.7414966225624084, 0.7891156673431396, 0.7891156673431396, 0.8639456033706665, 0.7687074542045593, 0.7687074542045593, 0.8503401279449463, 0.8367347121238708, 0.8095238208770752, 0.8979591727256775, 0.795918345451355, 0.8163265585899353, 0.8571428656578064, 0.8571428656578064, 0.8911564350128174, 0.918367326259613, 0.8503401279449463, 0.8911564350128174, 0.9659863710403442, 0.9115646481513977, 0.9591836929321289, 0.9795918464660645, 1.0, 0.8571428656578064, 0.8707482814788818, 0.9387755393981934, 0.9047619104385376, 0.9047619104385376, 0.9591836929321289, 0.9591836929321289, 0.9523809552192688, 0.9387755393981934, 1.0, 0.9727891087532043, 0.9319728016853333, 0.9523809552192688, 1.0], [0.11564625799655914, 0.30612245202064514, 0.31292515993118286, 0.26530611515045166, 0.3741496503353119, 0.3265306055545807, 0.2789115607738495, 0.3265306055545807, 0.3945578336715698, 0.4353741407394409, 0.44897958636283875, 0.5374149680137634, 0.44897958636283875, 0.442176878452301, 0.523809552192688, NaN, 0.4897959232330322, 0.5714285969734192, 0.5442177057266235, 0.5782312750816345, 0.5102040767669678, 0.6054421663284302, 0.6598639488220215, 0.5918367505073547, 0.6530612111091614, 0.6122449040412903, 0.6122449040412903, 0.6530612111091614, 0.6734693646430969, 0.795918345451355, 0.6190476417541504, 0.6054421663284302, 0.5510203838348389, 0.557823121547699, 0.7551020383834839, 0.49659863114356995, 0.7891156673431396, 0.7006802558898926, 0.6666666865348816, 0.8027210831642151, 0.7823129296302795, 0.6734693646430969, 0.7414966225624084, 0.7687074542045593, 0.8231292366981506, 0.7278911471366882, 0.7755101919174194, 0.8299319744110107, 0.7891156673431396, 0.7687074542045593, 0.8639456033706665, 0.7687074542045593, 0.8639456033706665, 0.8435373902320862, 0.8503401279449463, 0.884353756904602, 0.884353756904602, 0.7755101919174194, 0.8435373902320862, 0.918367326259613, 0.884353756904602, 0.9387755393981934, 1.0, 1.0, 0.8639456033706665, 0.8503401279449463, 0.9387755393981934, 0.8775510191917419, 0.8775510191917419, 0.9387755393981934, 0.9387755393981934, 0.9455782175064087, 0.9387755393981934, 1.0, 0.9455782175064087, 0.9387755393981934, 0.9659863710403442, 1.0], [0.190476194024086, 0.15646257996559143, 0.20408163964748383, 0.2312925159931183, 0.21088434755802155, 0.2857142984867096, 0.3401360511779785, 0.3333333432674408, 0.2380952388048172, 0.26530611515045166, 0.3401360511779785, 0.36734694242477417, 0.380952388048172, 0.3945578336715698, 0.45578232407569885, 0.5102040767669678, NaN, 0.36734694242477417, 0.523809552192688, 0.4897959232330322, 0.6360543966293335, 0.5306122303009033, 0.6122449040412903, 0.5986394286155701, 0.6122449040412903, 0.6870748400688171, 0.7142857313156128, 0.5442177057266235, 0.6598639488220215, 0.6258503198623657, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.8775510191917419, 0.5782312750816345, 0.6394557952880859, 0.5986394286155701, 0.7755101919174194, 0.8639456033706665, 0.7414966225624084, 0.6394557952880859, 0.7006802558898926, 0.8299319744110107, 0.6326530575752258, 0.6870748400688171, 0.8911564350128174, 0.8163265585899353, 0.8095238208770752, 0.7551020383834839, 0.8707482814788818, 0.884353756904602, 0.8911564350128174, 0.8775510191917419, 0.9523809552192688, 0.884353756904602, 0.8571428656578064, 0.884353756904602, 0.8027210831642151, 0.9455782175064087, 0.884353756904602, 0.9523809552192688, 0.9387755393981934, 0.8911564350128174, 0.9115646481513977, 0.9591836929321289, 0.9795918464660645, 1.0, 0.9387755393981934, 0.9523809552192688, 0.9523809552192688, 0.9659863710403442, 0.9523809552192688, 0.9795918464660645, 1.0, 0.9659863710403442, 0.9931972622871399, 1.0, 1.0], [0.08843537420034409, 0.12244898080825806, 0.12925170361995697, 0.17687074840068817, 0.14965985715389252, 0.25170066952705383, 0.2925170063972473, 0.29931971430778503, 0.31292515993118286, 0.41496598720550537, 0.4285714328289032, 0.3877550959587097, 0.37755101919174194, 0.3945578336715698, 0.45578232407569885, 0.4285714328289032, 0.6326530575752258, NaN, 0.5510203838348389, 0.5102040767669678, 0.6870748400688171, 0.6326530575752258, 0.5170068144798279, 0.7006802558898926, 0.6190476417541504, 0.6734693646430969, 0.6870748400688171, 0.5884353518486023, 0.680272102355957, 0.5986394286155701, 0.6734693646430969, 0.680272102355957, 0.7210884094238281, 0.7006802558898926, 0.7074829936027527, 0.5034013390541077, 0.5510203838348389, 0.7074829936027527, 0.7891156673431396, 0.6938775777816772, 0.7142857313156128, 0.7823129296302795, 0.8027210831642151, 0.7346938848495483, 0.7823129296302795, 0.8163265585899353, 0.6938775777816772, 0.7755101919174194, 0.8027210831642151, 0.7687074542045593, 0.8299319744110107, 0.8503401279449463, 0.8095238208770752, 0.8503401279449463, 0.8639456033706665, 0.8367347121238708, 0.8435373902320862, 0.8299319744110107, 0.8639456033706665, 0.8571428656578064, 0.884353756904602, 0.8979591727256775, 0.8809523582458496, 0.8571428656578064, 0.9387755393981934, 0.9251700639724731, 0.8911564350128174, 0.9115646481513977, 0.9047619104385376, 0.9795918464660645, 0.9659863710403442, 0.9659863710403442, 0.9047619104385376, 0.9115646481513977, 0.9319728016853333, 0.9251700639724731, 0.9931972622871399, 1.0], [0.07482992857694626, 0.11564625799655914, 0.1428571492433548, 0.20408163964748383, 0.21088434755802155, 0.2312925159931183, 0.30612245202064514, 0.2857142984867096, 0.190476194024086, 0.2925170063972473, 0.3469387888908386, 0.36054420471191406, 0.31292515993118286, 0.30612245202064514, 0.442176878452301, 0.45578232407569885, 0.4761904776096344, 0.44897958636283875, NaN, 0.4761904776096344, 0.5170068144798279, 0.5782312750816345, 0.5918367505073547, 0.6530612111091614, 0.6258503198623657, 0.5918367505073547, 0.6258503198623657, 0.5646258592605591, 0.7210884094238281, 0.6666666865348816, 0.7074829936027527, 0.6598639488220215, 0.646258533000946, 0.6666666865348816, 0.5918367505073547, 0.5102040767669678, 0.5986394286155701, 0.7414966225624084, 0.7755101919174194, 0.7414966225624084, 0.6530612111091614, 0.7278911471366882, 0.8163265585899353, 0.680272102355957, 0.7074829936027527, 0.8027210831642151, 0.8231292366981506, 0.8163265585899353, 0.7551020383834839, 0.8299319744110107, 0.8367347121238708, 0.8639456033706665, 0.8979591727256775, 0.8707482814788818, 0.8571428656578064, 0.8775510191917419, 0.884353756904602, 0.7891156673431396, 0.9115646481513977, 0.8231292366981506, 0.9251700639724731, 0.9047619104385376, 0.8979591727256775, 0.9251700639724731, 0.9455782175064087, 0.9591836929321289, 0.9795918464660645, 0.9387755393981934, 0.9387755393981934, 0.9387755393981934, 0.9455782175064087, 0.9523809552192688, 0.9795918464660645, 0.9795918464660645, 0.9523809552192688, 0.9659863710403442, 0.9931972622871399, 1.0], [0.11564625799655914, 0.22448979318141937, 0.25850340723991394, 0.1428571492433548, 0.31972789764404297, 0.29931971430778503, 0.2380952388048172, 0.26530611515045166, 0.3401360511779785, 0.442176878452301, 0.45578232407569885, 0.3945578336715698, 0.40136054158210754, 0.36054420471191406, 0.3401360511779785, 0.4217686951160431, 0.5102040767669678, 0.4897959232330322, 0.523809552192688, NaN, 0.5442177057266235, 0.5034013390541077, 0.5850340127944946, 0.5442177057266235, 0.6326530575752258, 0.5986394286155701, 0.6258503198623657, 0.6054421663284302, 0.6394557952880859, 0.646258533000946, 0.6530612111091614, 0.557823121547699, 0.5782312750816345, 0.6190476417541504, 0.6734693646430969, 0.4761904776096344, 0.7346938848495483, 0.6326530575752258, 0.6258503198623657, 0.680272102355957, 0.7006802558898926, 0.6734693646430969, 0.7006802558898926, 0.7210884094238281, 0.8231292366981506, 0.6938775777816772, 0.6938775777816772, 0.7687074542045593, 0.7482993006706238, 0.7278911471366882, 0.8027210831642151, 0.7278911471366882, 0.7755101919174194, 0.7482993006706238, 0.8435373902320862, 0.8571428656578064, 0.8775510191917419, 0.8163265585899353, 0.8571428656578064, 0.9591836929321289, 0.918367326259613, 0.9591836929321289, 0.9591836929321289, 0.9795918464660645, 0.8095238208770752, 0.8027210831642151, 0.9319728016853333, 0.9047619104385376, 0.9047619104385376, 0.9319728016853333, 0.9455782175064087, 0.9523809552192688, 0.9387755393981934, 1.0, 0.9591836929321289, 0.9251700639724731, 0.9387755393981934, 1.0], [0.17687074840068817, 0.14965985715389252, 0.20408163964748383, 0.2380952388048172, 0.190476194024086, 0.25170066952705383, 0.30612245202064514, 0.3265306055545807, 0.2312925159931183, 0.25850340723991394, 0.29931971430778503, 0.36054420471191406, 0.35374149680137634, 0.3469387888908386, 0.44897958636283875, 0.4897959232330322, 0.3639455735683441, 0.31292515993118286, 0.4829931855201721, 0.45578232407569885, NaN, 0.4625850319862366, 0.5646258592605591, 0.523809552192688, 0.5510203838348389, 0.5986394286155701, 0.6122449040412903, 0.5306122303009033, 0.5782312750816345, 0.6326530575752258, 0.5782312750816345, 0.5986394286155701, 0.6734693646430969, 0.7414966225624084, 0.5714285969734192, 0.5714285969734192, 0.6054421663284302, 0.7482993006706238, 0.8027210831642151, 0.680272102355957, 0.646258533000946, 0.6938775777816772, 0.761904776096344, 0.6122449040412903, 0.6666666865348816, 0.8571428656578064, 0.7823129296302795, 0.7755101919174194, 0.7210884094238281, 0.8367347121238708, 0.8435373902320862, 0.8707482814788818, 0.8367347121238708, 0.9115646481513977, 0.7891156673431396, 0.795918345451355, 0.8163265585899353, 0.7755101919174194, 0.9319728016853333, 0.8435373902320862, 0.9115646481513977, 0.9251700639724731, 0.884353756904602, 0.8503401279449463, 0.9727891087532043, 0.9727891087532043, 1.0, 0.9251700639724731, 0.9523809552192688, 0.9591836929321289, 0.9523809552192688, 0.9455782175064087, 0.9727891087532043, 1.0, 0.9659863710403442, 0.9863945841789246, 1.0, 1.0], [0.095238097012043, 0.07482992857694626, 0.06122449040412903, 0.11564625799655914, 0.13605442643165588, 0.1972789168357849, 0.17687074840068817, 0.22448979318141937, 0.18367347121238708, 0.21768707036972046, 0.25170066952705383, 0.27210885286331177, 0.2789115607738495, 0.31972789764404297, 0.4217686951160431, 0.3945578336715698, 0.4693877696990967, 0.36734694242477417, 0.4217686951160431, 0.49659863114356995, 0.5374149680137634, NaN, 0.4897959232330322, 0.6972789168357849, 0.523809552192688, 0.6190476417541504, 0.6394557952880859, 0.557823121547699, 0.6326530575752258, 0.5170068144798279, 0.6326530575752258, 0.5782312750816345, 0.5986394286155701, 0.646258533000946, 0.6326530575752258, 0.5102040767669678, 0.5714285969734192, 0.6394557952880859, 0.7414966225624084, 0.6734693646430969, 0.6938775777816772, 0.761904776096344, 0.8231292366981506, 0.7210884094238281, 0.7278911471366882, 0.8299319744110107, 0.6870748400688171, 0.8231292366981506, 0.8231292366981506, 0.7891156673431396, 0.9455782175064087, 0.8707482814788818, 0.8231292366981506, 0.9455782175064087, 0.884353756904602, 0.8707482814788818, 0.9115646481513977, 0.8979591727256775, 0.9591836929321289, 0.8639456033706665, 0.918367326259613, 0.9727891087532043, 0.918367326259613, 0.9455782175064087, 0.9591836929321289, 0.9659863710403442, 1.0, 0.9795918464660645, 0.9727891087532043, 0.9727891087532043, 0.9727891087532043, 0.9659863710403442, 0.9795918464660645, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.08163265138864517, 0.14965985715389252, 0.1428571492433548, 0.13605442643165588, 0.25850340723991394, 0.1428571492433548, 0.190476194024086, 0.25850340723991394, 0.2789115607738495, 0.27210885286331177, 0.26530611515045166, 0.36734694242477417, 0.2312925159931183, 0.36054420471191406, 0.3469387888908386, 0.3401360511779785, 0.3877550959587097, 0.4829931855201721, 0.40816327929496765, 0.41496598720550537, 0.4353741407394409, 0.5102040767669678, NaN, 0.5442177057266235, 0.6224489808082581, 0.5714285969734192, 0.5850340127944946, 0.523809552192688, 0.6326530575752258, 0.5306122303009033, 0.646258533000946, 0.6394557952880859, 0.5442177057266235, 0.5442177057266235, 0.7278911471366882, 0.5306122303009033, 0.5510203838348389, 0.5918367505073547, 0.6938775777816772, 0.6734693646430969, 0.8027210831642151, 0.795918345451355, 0.7142857313156128, 0.8027210831642151, 0.6530612111091614, 0.761904776096344, 0.6666666865348816, 0.9047619104385376, 0.8163265585899353, 0.761904776096344, 0.8707482814788818, 0.795918345451355, 0.8027210831642151, 0.884353756904602, 0.8027210831642151, 0.8775510191917419, 0.8639456033706665, 0.8911564350128174, 0.9251700639724731, 0.9115646481513977, 0.884353756904602, 0.9523809552192688, 0.9455782175064087, 0.9863945841789246, 0.9115646481513977, 0.9591836929321289, 0.9795918464660645, 0.9115646481513977, 0.9047619104385376, 0.9795918464660645, 0.9727891087532043, 0.9523809552192688, 0.9455782175064087, 0.9931972622871399, 0.9319728016853333, 0.9659863710403442, 0.9931972622871399, 1.0], [0.07482992857694626, 0.07482992857694626, 0.07482992857694626, 0.10204081982374191, 0.10884353518486023, 0.18367347121238708, 0.17687074840068817, 0.21768707036972046, 0.15646257996559143, 0.18367347121238708, 0.2448979616165161, 0.25850340723991394, 0.2857142984867096, 0.2448979616165161, 0.3877550959587097, 0.40816327929496765, 0.40136054158210754, 0.29931971430778503, 0.3469387888908386, 0.45578232407569885, 0.4761904776096344, 0.3027210831642151, 0.45578232407569885, NaN, 0.5102040767669678, 0.5782312750816345, 0.5782312750816345, 0.523809552192688, 0.5782312750816345, 0.4897959232330322, 0.557823121547699, 0.5306122303009033, 0.557823121547699, 0.6326530575752258, 0.6122449040412903, 0.4897959232330322, 0.557823121547699, 0.6530612111091614, 0.7278911471366882, 0.646258533000946, 0.6598639488220215, 0.7482993006706238, 0.7687074542045593, 0.6870748400688171, 0.680272102355957, 0.7755101919174194, 0.6734693646430969, 0.7891156673431396, 0.795918345451355, 0.7687074542045593, 0.884353756904602, 0.8435373902320862, 0.7823129296302795, 0.9047619104385376, 0.8435373902320862, 0.8435373902320862, 0.8435373902320862, 0.8707482814788818, 0.9319728016853333, 0.8435373902320862, 0.918367326259613, 0.9727891087532043, 0.9115646481513977, 0.9251700639724731, 0.9523809552192688, 0.9659863710403442, 1.0, 0.9659863710403442, 0.9591836929321289, 0.9727891087532043, 0.9727891087532043, 0.9659863710403442, 0.9727891087532043, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.10884353518486023, 0.13605442643165588, 0.12244898080825806, 0.11564625799655914, 0.21088434755802155, 0.17006802558898926, 0.17687074840068817, 0.190476194024086, 0.21768707036972046, 0.21768707036972046, 0.2448979616165161, 0.2312925159931183, 0.2312925159931183, 0.2925170063972473, 0.31972789764404297, 0.3469387888908386, 0.3877550959587097, 0.380952388048172, 0.3741496503353119, 0.36734694242477417, 0.44897958636283875, 0.4761904776096344, 0.37755101919174194, 0.4897959232330322, NaN, 0.5102040767669678, 0.5306122303009033, 0.44897958636283875, 0.5782312750816345, 0.45578232407569885, 0.5782312750816345, 0.5850340127944946, 0.5034013390541077, 0.5442177057266235, 0.646258533000946, 0.5374149680137634, 0.5442177057266235, 0.5714285969734192, 0.6734693646430969, 0.5986394286155701, 0.7142857313156128, 0.7210884094238281, 0.6938775777816772, 0.7142857313156128, 0.680272102355957, 0.7414966225624084, 0.6190476417541504, 0.8095238208770752, 0.7687074542045593, 0.7551020383834839, 0.8027210831642151, 0.761904776096344, 0.7346938848495483, 0.8095238208770752, 0.795918345451355, 0.8231292366981506, 0.8299319744110107, 0.8571428656578064, 0.9115646481513977, 0.918367326259613, 0.9115646481513977, 0.9591836929321289, 0.918367326259613, 0.9659863710403442, 0.884353756904602, 0.9319728016853333, 0.9727891087532043, 0.9251700639724731, 0.9319728016853333, 0.9659863710403442, 0.9659863710403442, 0.9455782175064087, 0.9387755393981934, 0.9863945841789246, 0.9727891087532043, 0.9659863710403442, 0.9931972622871399, 1.0], [0.12244898080825806, 0.095238097012043, 0.10204081982374191, 0.17006802558898926, 0.13605442643165588, 0.12244898080825806, 0.22448979318141937, 0.2448979616165161, 0.17006802558898926, 0.10204081982374191, 0.17006802558898926, 0.2380952388048172, 0.27210885286331177, 0.31972789764404297, 0.36734694242477417, 0.3877550959587097, 0.31292515993118286, 0.3265306055545807, 0.40816327929496765, 0.40136054158210754, 0.40136054158210754, 0.380952388048172, 0.4285714328289032, 0.4217686951160431, 0.4897959232330322, NaN, 0.5544217824935913, 0.49659863114356995, 0.5170068144798279, 0.4761904776096344, 0.5782312750816345, 0.5442177057266235, 0.5510203838348389, 0.5510203838348389, 0.5170068144798279, 0.557823121547699, 0.5170068144798279, 0.646258533000946, 0.7346938848495483, 0.6190476417541504, 0.5918367505073547, 0.680272102355957, 0.7482993006706238, 0.6190476417541504, 0.5714285969734192, 0.7823129296302795, 0.6666666865348816, 0.7823129296302795, 0.7346938848495483, 0.7278911471366882, 0.9387755393981934, 0.8435373902320862, 0.7346938848495483, 0.8435373902320862, 0.8775510191917419, 0.884353756904602, 0.884353756904602, 0.7823129296302795, 0.918367326259613, 0.8435373902320862, 0.9251700639724731, 0.9591836929321289, 0.9115646481513977, 0.9251700639724731, 0.9455782175064087, 0.9591836929321289, 0.9931972622871399, 0.9523809552192688, 0.9455782175064087, 0.9591836929321289, 0.9591836929321289, 0.9659863710403442, 0.9863945841789246, 0.9931972622871399, 0.9319728016853333, 0.9863945841789246, 1.0, 1.0], [0.11564625799655914, 0.095238097012043, 0.08163265138864517, 0.1428571492433548, 0.13605442643165588, 0.10204081982374191, 0.21768707036972046, 0.2312925159931183, 0.15646257996559143, 0.07482992857694626, 0.1428571492433548, 0.2448979616165161, 0.2380952388048172, 0.29931971430778503, 0.3469387888908386, 0.3877550959587097, 0.2857142984867096, 0.31292515993118286, 0.3741496503353119, 0.3741496503353119, 0.3877550959587097, 0.36054420471191406, 0.41496598720550537, 0.4217686951160431, 0.4693877696990967, 0.4455782175064087, NaN, 0.5034013390541077, 0.49659863114356995, 0.4897959232330322, 0.5374149680137634, 0.5374149680137634, 0.5510203838348389, 0.5374149680137634, 0.49659863114356995, 0.5442177057266235, 0.5102040767669678, 0.646258533000946, 0.761904776096344, 0.6258503198623657, 0.5986394286155701, 0.680272102355957, 0.7687074542045593, 0.6122449040412903, 0.5850340127944946, 0.8095238208770752, 0.6734693646430969, 0.7823129296302795, 0.7074829936027527, 0.7346938848495483, 0.9319728016853333, 0.8435373902320862, 0.7414966225624084, 0.8435373902320862, 0.8639456033706665, 0.8775510191917419, 0.8911564350128174, 0.7823129296302795, 0.918367326259613, 0.8435373902320862, 0.9251700639724731, 0.9523809552192688, 0.9047619104385376, 0.9115646481513977, 0.9455782175064087, 0.9591836929321289, 0.9931972622871399, 0.9591836929321289, 0.9523809552192688, 0.9591836929321289, 0.9523809552192688, 0.9591836929321289, 0.9863945841789246, 0.9931972622871399, 0.9319728016853333, 0.9863945841789246, 1.0, 1.0], [0.027210883796215057, 0.21088434755802155, 0.21088434755802155, 0.10204081982374191, 0.2380952388048172, 0.21768707036972046, 0.06802721321582794, 0.040816325694322586, 0.30612245202064514, 0.380952388048172, 0.3945578336715698, 0.3333333432674408, 0.32993197441101074, 0.3469387888908386, 0.3265306055545807, 0.3469387888908386, 0.45578232407569885, 0.4115646183490753, 0.4353741407394409, 0.3945578336715698, 0.4693877696990967, 0.442176878452301, 0.4761904776096344, 0.4761904776096344, 0.5510203838348389, 0.5034013390541077, 0.49659863114356995, NaN, 0.5918367505073547, 0.5918367505073547, 0.5850340127944946, 0.49659863114356995, 0.5102040767669678, 0.5374149680137634, 0.680272102355957, 0.40816327929496765, 0.6054421663284302, 0.4761904776096344, 0.5510203838348389, 0.646258533000946, 0.7482993006706238, 0.6326530575752258, 0.646258533000946, 0.7482993006706238, 0.8027210831642151, 0.6190476417541504, 0.6122449040412903, 0.7006802558898926, 0.7823129296302795, 0.6734693646430969, 0.761904776096344, 0.7074829936027527, 0.6870748400688171, 0.680272102355957, 0.8503401279449463, 0.8095238208770752, 0.8231292366981506, 0.7414966225624084, 0.7755101919174194, 0.9319728016853333, 0.8639456033706665, 0.8979591727256775, 0.942176878452301, 0.9319728016853333, 0.8095238208770752, 0.8231292366981506, 0.8911564350128174, 0.8503401279449463, 0.8503401279449463, 0.9455782175064087, 0.9387755393981934, 0.9455782175064087, 0.8775510191917419, 0.9387755393981934, 0.9387755393981934, 0.884353756904602, 1.0, 1.0], [0.095238097012043, 0.11564625799655914, 0.10884353518486023, 0.095238097012043, 0.14965985715389252, 0.14965985715389252, 0.15646257996559143, 0.15646257996559143, 0.13605442643165588, 0.12244898080825806, 0.21088434755802155, 0.21088434755802155, 0.14965985715389252, 0.25170066952705383, 0.2857142984867096, 0.3265306055545807, 0.3401360511779785, 0.31972789764404297, 0.2789115607738495, 0.36054420471191406, 0.4217686951160431, 0.36734694242477417, 0.36734694242477417, 0.4217686951160431, 0.4217686951160431, 0.4829931855201721, 0.5034013390541077, 0.40816327929496765, NaN, 0.4897959232330322, 0.47959184646606445, 0.523809552192688, 0.5102040767669678, 0.5918367505073547, 0.49659863114356995, 0.5306122303009033, 0.557823121547699, 0.6054421663284302, 0.7142857313156128, 0.5918367505073547, 0.5850340127944946, 0.6870748400688171, 0.7278911471366882, 0.5646258592605591, 0.6190476417541504, 0.7755101919174194, 0.646258533000946, 0.7551020383834839, 0.7142857313156128, 0.7278911471366882, 0.8435373902320862, 0.8095238208770752, 0.7823129296302795, 0.8299319744110107, 0.9387755393981934, 0.9455782175064087, 0.9863945841789246, 0.7551020383834839, 0.884353756904602, 0.8503401279449463, 0.9795918464660645, 0.918367326259613, 0.8979591727256775, 0.9455782175064087, 0.884353756904602, 0.9251700639724731, 1.0, 0.9455782175064087, 0.9455782175064087, 0.9387755393981934, 0.9523809552192688, 0.9387755393981934, 1.0, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.08163265138864517, 0.12244898080825806, 0.11564625799655914, 0.08843537420034409, 0.20408163964748383, 0.07482992857694626, 0.14965985715389252, 0.12925170361995697, 0.190476194024086, 0.1972789168357849, 0.2312925159931183, 0.25170066952705383, 0.1428571492433548, 0.2857142984867096, 0.2448979616165161, 0.20408163964748383, 0.3741496503353119, 0.40136054158210754, 0.3333333432674408, 0.35374149680137634, 0.36734694242477417, 0.4829931855201721, 0.4693877696990967, 0.5102040767669678, 0.5442177057266235, 0.523809552192688, 0.5102040767669678, 0.40816327929496765, 0.5102040767669678, NaN, 0.5306122303009033, 0.5646258592605591, 0.4761904776096344, 0.4897959232330322, 0.6190476417541504, 0.5102040767669678, 0.5306122303009033, 0.523809552192688, 0.646258533000946, 0.6734693646430969, 0.6734693646430969, 0.680272102355957, 0.6666666865348816, 0.7006802558898926, 0.6190476417541504, 0.7074829936027527, 0.6598639488220215, 0.795918345451355, 0.761904776096344, 0.680272102355957, 0.8299319744110107, 0.7687074542045593, 0.7755101919174194, 0.8163265585899353, 0.8299319744110107, 0.8911564350128174, 0.9047619104385376, 0.7755101919174194, 0.8571428656578064, 0.8503401279449463, 0.9319728016853333, 0.8639456033706665, 0.918367326259613, 0.9931972622871399, 0.8163265585899353, 0.884353756904602, 0.9931972622871399, 0.8639456033706665, 0.8571428656578064, 0.8911564350128174, 0.9115646481513977, 0.918367326259613, 0.9931972622871399, 1.0, 0.8979591727256775, 0.9727891087532043, 0.9931972622871399, 1.0], [0.10204081982374191, 0.08843537420034409, 0.10204081982374191, 0.10884353518486023, 0.12925170361995697, 0.11564625799655914, 0.16326530277729034, 0.1428571492433548, 0.11564625799655914, 0.12244898080825806, 0.1972789168357849, 0.20408163964748383, 0.16326530277729034, 0.2380952388048172, 0.30612245202064514, 0.380952388048172, 0.3333333432674408, 0.3265306055545807, 0.2925170063972473, 0.3469387888908386, 0.4217686951160431, 0.36734694242477417, 0.35374149680137634, 0.442176878452301, 0.4217686951160431, 0.4217686951160431, 0.4625850319862366, 0.41496598720550537, 0.5204081535339355, 0.4693877696990967, NaN, 0.5034013390541077, 0.4897959232330322, 0.5782312750816345, 0.4761904776096344, 0.5170068144798279, 0.5646258592605591, 0.6326530575752258, 0.6938775777816772, 0.5782312750816345, 0.5442177057266235, 0.6598639488220215, 0.7414966225624084, 0.5782312750816345, 0.6122449040412903, 0.7823129296302795, 0.6326530575752258, 0.7346938848495483, 0.6938775777816772, 0.7278911471366882, 0.8503401279449463, 0.8163265585899353, 0.761904776096344, 0.8163265585899353, 0.918367326259613, 0.9319728016853333, 0.9455782175064087, 0.7687074542045593, 0.8639456033706665, 0.8571428656578064, 0.9659863710403442, 0.9319728016853333, 0.884353756904602, 0.9455782175064087, 0.8911564350128174, 0.9251700639724731, 1.0, 0.9523809552192688, 0.9591836929321289, 0.9387755393981934, 0.9523809552192688, 0.9387755393981934, 1.0, 1.0, 0.9523809552192688, 0.9931972622871399, 1.0, 1.0], [0.08163265138864517, 0.10884353518486023, 0.12244898080825806, 0.095238097012043, 0.18367347121238708, 0.21088434755802155, 0.16326530277729034, 0.21768707036972046, 0.17006802558898926, 0.17687074840068817, 0.2380952388048172, 0.16326530277729034, 0.2448979616165161, 0.2925170063972473, 0.36734694242477417, 0.3945578336715698, 0.380952388048172, 0.31972789764404297, 0.3401360511779785, 0.442176878452301, 0.40136054158210754, 0.4217686951160431, 0.36054420471191406, 0.4693877696990967, 0.41496598720550537, 0.45578232407569885, 0.4625850319862366, 0.5034013390541077, 0.4761904776096344, 0.4353741407394409, 0.49659863114356995, NaN, 0.4761904776096344, 0.4829931855201721, 0.523809552192688, 0.49659863114356995, 0.5714285969734192, 0.5918367505073547, 0.5102040767669678, 0.6326530575752258, 0.5986394286155701, 0.6734693646430969, 0.6598639488220215, 0.6394557952880859, 0.7278911471366882, 0.646258533000946, 0.5646258592605591, 0.6734693646430969, 0.7006802558898926, 0.6938775777816772, 0.7482993006706238, 0.7551020383834839, 0.6598639488220215, 0.7074829936027527, 0.7074829936027527, 0.8639456033706665, 0.884353756904602, 0.9591836929321289, 0.8503401279449463, 0.9115646481513977, 0.9387755393981934, 0.9727891087532043, 0.9659863710403442, 0.9931972622871399, 0.9591836929321289, 0.9659863710403442, 0.8979591727256775, 0.9659863710403442, 0.9659863710403442, 0.9115646481513977, 0.8911564350128174, 0.8911564350128174, 0.9659863710403442, 0.9931972622871399, 0.9863945841789246, 0.9863945841789246, 1.0, 1.0], [0.08843537420034409, 0.10204081982374191, 0.08843537420034409, 0.17687074840068817, 0.14965985715389252, 0.16326530277729034, 0.25850340723991394, 0.26530611515045166, 0.13605442643165588, 0.1428571492433548, 0.15646257996559143, 0.25170066952705383, 0.25850340723991394, 0.2380952388048172, 0.36734694242477417, 0.44897958636283875, 0.2380952388048172, 0.2789115607738495, 0.35374149680137634, 0.4217686951160431, 0.3265306055545807, 0.40136054158210754, 0.45578232407569885, 0.442176878452301, 0.49659863114356995, 0.44897958636283875, 0.44897958636283875, 0.4897959232330322, 0.4897959232330322, 0.523809552192688, 0.5102040767669678, 0.523809552192688, NaN, 0.5102040767669678, 0.5034013390541077, 0.442176878452301, 0.5714285969734192, 0.7142857313156128, 0.7006802558898926, 0.5918367505073547, 0.5646258592605591, 0.6190476417541504, 0.6734693646430969, 0.5714285969734192, 0.5714285969734192, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.7823129296302795, 0.7414966225624084, 0.8231292366981506, 0.7755101919174194, 0.8571428656578064, 0.7278911471366882, 0.7482993006706238, 0.7755101919174194, 0.7142857313156128, 0.9387755393981934, 0.795918345451355, 0.8979591727256775, 0.8775510191917419, 0.8231292366981506, 0.8163265585899353, 0.9387755393981934, 0.9387755393981934, 0.9931972622871399, 0.9523809552192688, 0.9523809552192688, 0.9319728016853333, 0.9591836929321289, 0.9387755393981934, 0.9931972622871399, 1.0, 0.9727891087532043, 0.9863945841789246, 1.0, 1.0], [0.14965985715389252, 0.1428571492433548, 0.15646257996559143, 0.1972789168357849, 0.18367347121238708, 0.21088434755802155, 0.29931971430778503, 0.26530611515045166, 0.18367347121238708, 0.12925170361995697, 0.1972789168357849, 0.2925170063972473, 0.2857142984867096, 0.2857142984867096, 0.36734694242477417, 0.442176878452301, 0.12244898080825806, 0.29931971430778503, 0.3333333432674408, 0.380952388048172, 0.25850340723991394, 0.35374149680137634, 0.45578232407569885, 0.36734694242477417, 0.45578232407569885, 0.44897958636283875, 0.4625850319862366, 0.4625850319862366, 0.40816327929496765, 0.5102040767669678, 0.4217686951160431, 0.5170068144798279, 0.4897959232330322, NaN, 0.4625850319862366, 0.4625850319862366, 0.5510203838348389, 0.646258533000946, 0.646258533000946, 0.5646258592605591, 0.523809552192688, 0.5782312750816345, 0.6190476417541504, 0.523809552192688, 0.49659863114356995, 0.7210884094238281, 0.7346938848495483, 0.6598639488220215, 0.5782312750816345, 0.7755101919174194, 0.7074829936027527, 0.761904776096344, 0.8027210831642151, 0.8435373902320862, 0.646258533000946, 0.7346938848495483, 0.7346938848495483, 0.6870748400688171, 0.9115646481513977, 0.761904776096344, 0.8163265585899353, 0.8367347121238708, 0.8095238208770752, 0.8027210831642151, 0.9115646481513977, 0.9319728016853333, 0.9863945841789246, 0.9251700639724731, 0.9319728016853333, 0.9387755393981934, 0.9319728016853333, 0.9251700639724731, 0.9795918464660645, 1.0, 0.9591836929321289, 0.9795918464660645, 1.0, 1.0], [0.07482992857694626, 0.13605442643165588, 0.15646257996559143, 0.10204081982374191, 0.1972789168357849, 0.15646257996559143, 0.095238097012043, 0.18367347121238708, 0.21088434755802155, 0.21088434755802155, 0.21768707036972046, 0.11564625799655914, 0.22448979318141937, 0.2789115607738495, 0.26530611515045166, 0.2448979616165161, 0.4217686951160431, 0.2925170063972473, 0.40816327929496765, 0.3265306055545807, 0.4285714328289032, 0.36734694242477417, 0.27210885286331177, 0.3877550959587097, 0.35374149680137634, 0.4829931855201721, 0.5034013390541077, 0.31972789764404297, 0.5034013390541077, 0.380952388048172, 0.523809552192688, 0.4761904776096344, 0.49659863114356995, 0.5374149680137634, NaN, 0.4761904776096344, 0.5170068144798279, 0.44897958636283875, 0.5306122303009033, 0.5306122303009033, 0.7176870703697205, 0.557823121547699, 0.6122449040412903, 0.7687074542045593, 0.6734693646430969, 0.6054421663284302, 0.5306122303009033, 0.6122449040412903, 0.8503401279449463, 0.5918367505073547, 0.7006802558898926, 0.7006802558898926, 0.6122449040412903, 0.6666666865348816, 0.7891156673431396, 0.8095238208770752, 0.8163265585899353, 0.8095238208770752, 0.7755101919174194, 0.9523809552192688, 0.8707482814788818, 0.9455782175064087, 0.9659863710403442, 1.0, 0.761904776096344, 0.761904776096344, 0.9115646481513977, 0.9115646481513977, 0.9115646481513977, 0.9251700639724731, 0.9523809552192688, 0.9387755393981934, 0.8979591727256775, 0.9795918464660645, 1.0, 0.9319728016853333, 0.9455782175064087, 1.0], [0.2448979616165161, 0.25170066952705383, 0.25850340723991394, 0.3741496503353119, 0.2857142984867096, 0.3469387888908386, 0.4285714328289032, 0.4897959232330322, 0.380952388048172, 0.3469387888908386, 0.36054420471191406, 0.41496598720550537, 0.4897959232330322, 0.45578232407569885, 0.5102040767669678, 0.5034013390541077, 0.36054420471191406, 0.49659863114356995, 0.4897959232330322, 0.523809552192688, 0.4285714328289032, 0.4897959232330322, 0.4693877696990967, 0.5102040767669678, 0.4625850319862366, 0.442176878452301, 0.45578232407569885, 0.5918367505073547, 0.4693877696990967, 0.4897959232330322, 0.4829931855201721, 0.5034013390541077, 0.557823121547699, 0.5374149680137634, 0.523809552192688, NaN, 0.523809552192688, 0.6870748400688171, 0.5782312750816345, 0.5306122303009033, 0.5306122303009033, 0.5714285969734192, 0.5442177057266235, 0.523809552192688, 0.557823121547699, 0.6054421663284302, 0.6530612111091614, 0.5782312750816345, 0.5510203838348389, 0.7006802558898926, 0.5374149680137634, 0.6326530575752258, 0.7210884094238281, 0.7414966225624084, 0.5918367505073547, 0.5306122303009033, 0.5102040767669678, 0.5850340127944946, 0.7142857313156128, 0.5646258592605591, 0.6394557952880859, 0.6530612111091614, 0.6530612111091614, 0.6326530575752258, 0.8027210831642151, 0.8027210831642151, 0.7210884094238281, 0.7346938848495483, 0.7210884094238281, 0.7551020383834839, 0.7482993006706238, 0.7482993006706238, 0.7891156673431396, 0.7482993006706238, 0.761904776096344, 0.9455782175064087, 0.8095238208770752, 1.0], [0.095238097012043, 0.20408163964748383, 0.22448979318141937, 0.16326530277729034, 0.29931971430778503, 0.22448979318141937, 0.18367347121238708, 0.25170066952705383, 0.31972789764404297, 0.3945578336715698, 0.380952388048172, 0.3945578336715698, 0.31972789764404297, 0.3401360511779785, 0.25170066952705383, 0.21088434755802155, 0.40136054158210754, 0.44897958636283875, 0.40136054158210754, 0.26530611515045166, 0.3945578336715698, 0.4285714328289032, 0.44897958636283875, 0.442176878452301, 0.45578232407569885, 0.4829931855201721, 0.4897959232330322, 0.3945578336715698, 0.442176878452301, 0.4693877696990967, 0.4353741407394409, 0.4285714328289032, 0.4285714328289032, 0.44897958636283875, 0.4829931855201721, 0.4761904776096344, NaN, 0.45578232407569885, 0.4693877696990967, 0.523809552192688, 0.5170068144798279, 0.49659863114356995, 0.45578232407569885, 0.5714285969734192, 0.5986394286155701, 0.49659863114356995, 0.557823121547699, 0.5782312750816345, 0.6258503198623657, 0.523809552192688, 0.557823121547699, 0.5034013390541077, 0.6122449040412903, 0.5714285969734192, 0.7346938848495483, 0.6598639488220215, 0.6326530575752258, 0.6938775777816772, 0.6326530575752258, 0.8503401279449463, 0.8231292366981506, 0.8095238208770752, 0.8027210831642151, 0.918367326259613, 0.5782312750816345, 0.646258533000946, 0.918367326259613, 0.761904776096344, 0.7482993006706238, 0.8503401279449463, 0.8707482814788818, 0.884353756904602, 0.9047619104385376, 1.0, 0.8775510191917419, 0.8911564350128174, 0.9319728016853333, 1.0], [0.07482992857694626, 0.054421767592430115, 0.06122449040412903, 0.095238097012043, 0.10204081982374191, 0.08163265138864517, 0.12244898080825806, 0.17687074840068817, 0.12244898080825806, 0.1972789168357849, 0.14965985715389252, 0.29931971430778503, 0.16326530277729034, 0.17006802558898926, 0.31972789764404297, 0.29931971430778503, 0.22448979318141937, 0.2925170063972473, 0.25850340723991394, 0.36734694242477417, 0.25170066952705383, 0.36054420471191406, 0.40816327929496765, 0.3469387888908386, 0.4285714328289032, 0.35374149680137634, 0.35374149680137634, 0.523809552192688, 0.3945578336715698, 0.4761904776096344, 0.36734694242477417, 0.40816327929496765, 0.2857142984867096, 0.35374149680137634, 0.5510203838348389, 0.31292515993118286, 0.5442177057266235, NaN, 0.380952388048172, 0.5782312750816345, 0.5782312750816345, 0.5170068144798279, 0.5170068144798279, 0.5850340127944946, 0.5170068144798279, 0.4829931855201721, 0.6938775777816772, 0.6394557952880859, 0.5918367505073547, 0.8129251599311829, 0.6190476417541504, 0.5714285969734192, 0.795918345451355, 0.6938775777816772, 0.6394557952880859, 0.6530612111091614, 0.6530612111091614, 0.6530612111091614, 0.8911564350128174, 0.7278911471366882, 0.6870748400688171, 0.761904776096344, 0.7414966225624084, 0.7891156673431396, 0.8707482814788818, 0.918367326259613, 0.9115646481513977, 0.8979591727256775, 0.9047619104385376, 0.8367347121238708, 0.8775510191917419, 0.8707482814788818, 0.9591836929321289, 0.9931972622871399, 0.9387755393981934, 0.9727891087532043, 0.9931972622871399, 1.0], [0.095238097012043, 0.0476190485060215, 0.06802721321582794, 0.08163265138864517, 0.095238097012043, 0.13605442643165588, 0.17006802558898926, 0.17006802558898926, 0.12244898080825806, 0.08163265138864517, 0.10204081982374191, 0.17687074840068817, 0.18367347121238708, 0.18367347121238708, 0.26530611515045166, 0.3333333432674408, 0.13605442643165588, 0.21088434755802155, 0.22448979318141937, 0.3741496503353119, 0.1972789168357849, 0.25850340723991394, 0.30612245202064514, 0.27210885286331177, 0.3265306055545807, 0.26530611515045166, 0.2380952388048172, 0.44897958636283875, 0.2857142984867096, 0.35374149680137634, 0.30612245202064514, 0.4897959232330322, 0.29931971430778503, 0.35374149680137634, 0.4693877696990967, 0.4217686951160431, 0.5306122303009033, 0.6190476417541504, NaN, 0.5102040767669678, 0.5102040767669678, 0.5986394286155701, 0.4761904776096344, 0.5102040767669678, 0.5170068144798279, 0.7585033774375916, 0.557823121547699, 0.6190476417541504, 0.5510203838348389, 0.6870748400688171, 0.5986394286155701, 0.8435373902320862, 0.6734693646430969, 0.7346938848495483, 0.5918367505073547, 0.6938775777816772, 0.7074829936027527, 0.7346938848495483, 0.918367326259613, 0.7482993006706238, 0.8299319744110107, 0.8707482814788818, 0.7891156673431396, 0.8027210831642151, 0.9387755393981934, 0.9727891087532043, 0.9455782175064087, 0.9319728016853333, 0.9387755393981934, 0.918367326259613, 0.9319728016853333, 0.9319728016853333, 0.9795918464660645, 0.9931972622871399, 0.9387755393981934, 0.9863945841789246, 0.9931972622871399, 1.0], [0.06802721321582794, 0.0476190485060215, 0.0476190485060215, 0.06122449040412903, 0.10204081982374191, 0.054421767592430115, 0.08843537420034409, 0.12244898080825806, 0.10204081982374191, 0.095238097012043, 0.11564625799655914, 0.17006802558898926, 0.12244898080825806, 0.17006802558898926, 0.21088434755802155, 0.1972789168357849, 0.25850340723991394, 0.30612245202064514, 0.25850340723991394, 0.31972789764404297, 0.31972789764404297, 0.3265306055545807, 0.3265306055545807, 0.35374149680137634, 0.40136054158210754, 0.380952388048172, 0.3741496503353119, 0.35374149680137634, 0.40816327929496765, 0.3265306055545807, 0.4217686951160431, 0.36734694242477417, 0.40816327929496765, 0.4353741407394409, 0.4693877696990967, 0.4693877696990967, 0.4761904776096344, 0.4217686951160431, 0.4897959232330322, NaN, 0.5714285969734192, 0.5306122303009033, 0.5442177057266235, 0.5918367505073547, 0.5170068144798279, 0.5918367505073547, 0.6258503198623657, 0.646258533000946, 0.7142857313156128, 0.6122449040412903, 0.6734693646430969, 0.6666666865348816, 0.6734693646430969, 0.7278911471366882, 0.7074829936027527, 0.7755101919174194, 0.7346938848495483, 0.7278911471366882, 0.8503401279449463, 0.761904776096344, 0.8367347121238708, 0.8639456033706665, 0.8639456033706665, 0.9251700639724731, 0.7687074542045593, 0.8299319744110107, 0.9795918464660645, 0.8435373902320862, 0.8435373902320862, 0.8911564350128174, 0.9115646481513977, 0.9047619104385376, 0.9387755393981934, 0.9727891087532043, 0.884353756904602, 0.9455782175064087, 0.9727891087532043, 1.0], [0.095238097012043, 0.12244898080825806, 0.12925170361995697, 0.095238097012043, 0.17687074840068817, 0.17687074840068817, 0.12244898080825806, 0.12925170361995697, 0.190476194024086, 0.13605442643165588, 0.16326530277729034, 0.095238097012043, 0.18367347121238708, 0.26530611515045166, 0.20408163964748383, 0.21768707036972046, 0.36054420471191406, 0.2857142984867096, 0.3469387888908386, 0.29931971430778503, 0.35374149680137634, 0.30612245202064514, 0.1972789168357849, 0.3401360511779785, 0.2857142984867096, 0.40816327929496765, 0.40136054158210754, 0.25170066952705383, 0.41496598720550537, 0.3265306055545807, 0.45578232407569885, 0.40136054158210754, 0.4353741407394409, 0.4761904776096344, 0.28231292963027954, 0.4693877696990967, 0.4829931855201721, 0.4217686951160431, 0.4897959232330322, 0.4285714328289032, NaN, 0.523809552192688, 0.5170068144798279, 0.5510203838348389, 0.5986394286155701, 0.5442177057266235, 0.4761904776096344, 0.5306122303009033, 0.7993197441101074, 0.5306122303009033, 0.6326530575752258, 0.646258533000946, 0.5170068144798279, 0.5918367505073547, 0.7278911471366882, 0.7687074542045593, 0.761904776096344, 0.7482993006706238, 0.7278911471366882, 0.9047619104385376, 0.8503401279449463, 0.8979591727256775, 0.9863945841789246, 0.9523809552192688, 0.7074829936027527, 0.7142857313156128, 0.8435373902320862, 0.9251700639724731, 0.9251700639724731, 0.9115646481513977, 0.9319728016853333, 0.9319728016853333, 0.884353756904602, 0.9455782175064087, 0.9931972622871399, 0.9387755393981934, 0.918367326259613, 1.0], [0.07482992857694626, 0.08163265138864517, 0.08163265138864517, 0.06122449040412903, 0.12925170361995697, 0.14965985715389252, 0.095238097012043, 0.18367347121238708, 0.16326530277729034, 0.14965985715389252, 0.17006802558898926, 0.10204081982374191, 0.190476194024086, 0.190476194024086, 0.25850340723991394, 0.3265306055545807, 0.29931971430778503, 0.21768707036972046, 0.27210885286331177, 0.3265306055545807, 0.30612245202064514, 0.2380952388048172, 0.20408163964748383, 0.25170066952705383, 0.2789115607738495, 0.31972789764404297, 0.31972789764404297, 0.36734694242477417, 0.31292515993118286, 0.31972789764404297, 0.3401360511779785, 0.3265306055545807, 0.380952388048172, 0.4217686951160431, 0.442176878452301, 0.4285714328289032, 0.5034013390541077, 0.4829931855201721, 0.40136054158210754, 0.4693877696990967, 0.4761904776096344, NaN, 0.442176878452301, 0.5374149680137634, 0.6258503198623657, 0.5102040767669678, 0.49659863114356995, 0.5306122303009033, 0.5850340127944946, 0.5918367505073547, 0.5646258592605591, 0.6190476417541504, 0.5374149680137634, 0.557823121547699, 0.6326530575752258, 0.7414966225624084, 0.7142857313156128, 0.918367326259613, 0.795918345451355, 0.8707482814788818, 0.884353756904602, 0.9455782175064087, 0.918367326259613, 0.9455782175064087, 0.8707482814788818, 0.8775510191917419, 0.8571428656578064, 0.9251700639724731, 0.9319728016853333, 0.8911564350128174, 0.918367326259613, 0.9047619104385376, 0.9251700639724731, 0.9931972622871399, 0.9727891087532043, 0.9795918464660645, 0.9931972622871399, 1.0], [0.06802721321582794, 0.03401360660791397, 0.06122449040412903, 0.054421767592430115, 0.08843537420034409, 0.0476190485060215, 0.11564625799655914, 0.11564625799655914, 0.06802721321582794, 0.06802721321582794, 0.10884353518486023, 0.12244898080825806, 0.08163265138864517, 0.17687074840068817, 0.21088434755802155, 0.25850340723991394, 0.17006802558898926, 0.1972789168357849, 0.18367347121238708, 0.29931971430778503, 0.2380952388048172, 0.17687074840068817, 0.2857142984867096, 0.2312925159931183, 0.30612245202064514, 0.25170066952705383, 0.2312925159931183, 0.35374149680137634, 0.27210885286331177, 0.3333333432674408, 0.25850340723991394, 0.3401360511779785, 0.3265306055545807, 0.380952388048172, 0.3877550959587097, 0.45578232407569885, 0.5442177057266235, 0.4829931855201721, 0.523809552192688, 0.45578232407569885, 0.4829931855201721, 0.557823121547699, NaN, 0.4897959232330322, 0.4897959232330322, 0.6666666865348816, 0.5646258592605591, 0.6054421663284302, 0.6054421663284302, 0.6326530575752258, 0.6734693646430969, 0.7006802558898926, 0.6666666865348816, 0.7074829936027527, 0.7278911471366882, 0.7891156673431396, 0.8027210831642151, 0.7006802558898926, 0.8571428656578064, 0.795918345451355, 0.9387755393981934, 0.8911564350128174, 0.8435373902320862, 0.884353756904602, 0.8367347121238708, 0.8639456033706665, 1.0, 0.918367326259613, 0.9115646481513977, 0.918367326259613, 0.9455782175064087, 0.918367326259613, 0.9659863710403442, 1.0, 0.9455782175064087, 0.9931972622871399, 1.0, 1.0], [0.08163265138864517, 0.10884353518486023, 0.12925170361995697, 0.10204081982374191, 0.15646257996559143, 0.16326530277729034, 0.08843537420034409, 0.16326530277729034, 0.16326530277729034, 0.1428571492433548, 0.17687074840068817, 0.095238097012043, 0.190476194024086, 0.2312925159931183, 0.21088434755802155, 0.2312925159931183, 0.36734694242477417, 0.26530611515045166, 0.31972789764404297, 0.2789115607738495, 0.3877550959587097, 0.2789115607738495, 0.1972789168357849, 0.31292515993118286, 0.2857142984867096, 0.380952388048172, 0.3877550959587097, 0.25170066952705383, 0.4353741407394409, 0.29931971430778503, 0.4217686951160431, 0.36054420471191406, 0.4285714328289032, 0.4761904776096344, 0.2312925159931183, 0.4761904776096344, 0.4285714328289032, 0.41496598720550537, 0.4897959232330322, 0.40816327929496765, 0.44897958636283875, 0.4625850319862366, 0.5102040767669678, NaN, 0.6122449040412903, 0.5646258592605591, 0.4897959232330322, 0.4829931855201721, 0.6360543966293335, 0.5170068144798279, 0.5510203838348389, 0.6054421663284302, 0.5374149680137634, 0.5646258592605591, 0.7346938848495483, 0.680272102355957, 0.6870748400688171, 0.7142857313156128, 0.7006802558898926, 0.9319728016853333, 0.8503401279449463, 0.9115646481513977, 0.8911564350128174, 0.884353756904602, 0.7074829936027527, 0.6870748400688171, 0.8503401279449463, 0.9115646481513977, 0.9047619104385376, 0.9115646481513977, 0.9115646481513977, 0.918367326259613, 0.884353756904602, 0.9659863710403442, 1.0, 0.9319728016853333, 0.9251700639724731, 1.0], [0.08163265138864517, 0.027210883796215057, 0.027210883796215057, 0.013605441898107529, 0.06802721321582794, 0.0476190485060215, 0.054421767592430115, 0.08163265138864517, 0.10204081982374191, 0.21088434755802155, 0.21088434755802155, 0.190476194024086, 0.12244898080825806, 0.15646257996559143, 0.13605442643165588, 0.17687074840068817, 0.31292515993118286, 0.21768707036972046, 0.2925170063972473, 0.17687074840068817, 0.3333333432674408, 0.27210885286331177, 0.3469387888908386, 0.31972789764404297, 0.31972789764404297, 0.4285714328289032, 0.41496598720550537, 0.1972789168357849, 0.380952388048172, 0.380952388048172, 0.3877550959587097, 0.27210885286331177, 0.4285714328289032, 0.5034013390541077, 0.3265306055545807, 0.442176878452301, 0.40136054158210754, 0.4829931855201721, 0.4829931855201721, 0.4829931855201721, 0.40136054158210754, 0.3741496503353119, 0.5102040767669678, 0.3877550959587097, NaN, 0.5442177057266235, 0.6394557952880859, 0.523809552192688, 0.5646258592605591, 0.5986394286155701, 0.5442177057266235, 0.5782312750816345, 0.6598639488220215, 0.6598639488220215, 0.7482993006706238, 0.6122449040412903, 0.6054421663284302, 0.5986394286155701, 0.680272102355957, 0.8435373902320862, 0.7823129296302795, 0.9251700639724731, 0.8571428656578064, 0.761904776096344, 0.6598639488220215, 0.6938775777816772, 0.8367347121238708, 0.8707482814788818, 0.8571428656578064, 0.9319728016853333, 0.9455782175064087, 0.9455782175064087, 0.8707482814788818, 0.9523809552192688, 0.9523809552192688, 0.9319728016853333, 0.9659863710403442, 1.0], [0.07482992857694626, 0.040816325694322586, 0.06122449040412903, 0.054421767592430115, 0.08843537420034409, 0.095238097012043, 0.14965985715389252, 0.1428571492433548, 0.10884353518486023, 0.07482992857694626, 0.06122449040412903, 0.12244898080825806, 0.1428571492433548, 0.17006802558898926, 0.2312925159931183, 0.27210885286331177, 0.10884353518486023, 0.18367347121238708, 0.1972789168357849, 0.30612245202064514, 0.1428571492433548, 0.17006802558898926, 0.2380952388048172, 0.22448979318141937, 0.25850340723991394, 0.21768707036972046, 0.190476194024086, 0.380952388048172, 0.22448979318141937, 0.2925170063972473, 0.21768707036972046, 0.35374149680137634, 0.190476194024086, 0.2789115607738495, 0.3945578336715698, 0.3945578336715698, 0.5034013390541077, 0.5170068144798279, 0.24149659276008606, 0.40816327929496765, 0.45578232407569885, 0.4897959232330322, 0.3333333432674408, 0.4353741407394409, 0.45578232407569885, NaN, 0.5102040767669678, 0.5170068144798279, 0.523809552192688, 0.6326530575752258, 0.5102040767669678, 0.7006802558898926, 0.6258503198623657, 0.646258533000946, 0.49659863114356995, 0.6326530575752258, 0.6598639488220215, 0.6870748400688171, 0.8775510191917419, 0.6938775777816772, 0.7823129296302795, 0.8163265585899353, 0.7551020383834839, 0.7755101919174194, 0.8911564350128174, 0.9251700639724731, 0.9047619104385376, 0.918367326259613, 0.9251700639724731, 0.8979591727256775, 0.9115646481513977, 0.9047619104385376, 0.9659863710403442, 0.9931972622871399, 0.9387755393981934, 0.9795918464660645, 0.9931972622871399, 1.0], [0.10204081982374191, 0.07482992857694626, 0.06802721321582794, 0.12925170361995697, 0.08163265138864517, 0.054421767592430115, 0.17006802558898926, 0.18367347121238708, 0.12244898080825806, 0.095238097012043, 0.12925170361995697, 0.2312925159931183, 0.17687074840068817, 0.190476194024086, 0.2312925159931183, 0.22448979318141937, 0.18367347121238708, 0.30612245202064514, 0.17687074840068817, 0.30612245202064514, 0.21768707036972046, 0.31292515993118286, 0.3333333432674408, 0.3265306055545807, 0.380952388048172, 0.3333333432674408, 0.3265306055545807, 0.3877550959587097, 0.35374149680137634, 0.3401360511779785, 0.36734694242477417, 0.4353741407394409, 0.2857142984867096, 0.26530611515045166, 0.4693877696990967, 0.3469387888908386, 0.442176878452301, 0.30612245202064514, 0.442176878452301, 0.3741496503353119, 0.523809552192688, 0.5034013390541077, 0.4353741407394409, 0.5102040767669678, 0.36054420471191406, 0.4897959232330322, NaN, 0.5986394286155701, 0.5374149680137634, 0.5034013390541077, 0.557823121547699, 0.5102040767669678, 0.7551020383834839, 0.5918367505073547, 0.5850340127944946, 0.6870748400688171, 0.6734693646430969, 0.557823121547699, 0.6734693646430969, 0.6326530575752258, 0.7278911471366882, 0.6326530575752258, 0.646258533000946, 0.7823129296302795, 0.8095238208770752, 0.8571428656578064, 0.7891156673431396, 0.7687074542045593, 0.7687074542045593, 0.680272102355957, 0.7210884094238281, 0.7210884094238281, 0.9931972622871399, 0.918367326259613, 0.8911564350128174, 0.9523809552192688, 1.0, 0.9931972622871399], [0.020408162847161293, 0.054421767592430115, 0.06802721321582794, 0.03401360660791397, 0.095238097012043, 0.06802721321582794, 0.03401360660791397, 0.0476190485060215, 0.13605442643165588, 0.054421767592430115, 0.08163265138864517, 0.06122449040412903, 0.12244898080825806, 0.1972789168357849, 0.14965985715389252, 0.17006802558898926, 0.190476194024086, 0.22448979318141937, 0.18367347121238708, 0.2312925159931183, 0.22448979318141937, 0.17687074840068817, 0.095238097012043, 0.21088434755802155, 0.190476194024086, 0.21768707036972046, 0.21768707036972046, 0.29931971430778503, 0.2448979616165161, 0.20408163964748383, 0.26530611515045166, 0.3265306055545807, 0.2857142984867096, 0.3401360511779785, 0.3877550959587097, 0.4217686951160431, 0.4217686951160431, 0.36054420471191406, 0.380952388048172, 0.35374149680137634, 0.4693877696990967, 0.4693877696990967, 0.3945578336715698, 0.5170068144798279, 0.4761904776096344, 0.4829931855201721, 0.40136054158210754, NaN, 0.6326530575752258, 0.4829931855201721, 0.5034013390541077, 0.5714285969734192, 0.4829931855201721, 0.5034013390541077, 0.6530612111091614, 0.7414966225624084, 0.761904776096344, 0.6666666865348816, 0.7278911471366882, 0.8163265585899353, 0.8503401279449463, 0.8163265585899353, 0.8503401279449463, 0.8367347121238708, 0.7687074542045593, 0.8095238208770752, 0.8367347121238708, 0.7823129296302795, 0.7823129296302795, 0.8775510191917419, 0.9047619104385376, 0.8979591727256775, 0.8571428656578064, 0.9251700639724731, 0.8979591727256775, 0.8639456033706665, 0.9795918464660645, 1.0], [0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.08843537420034409, 0.13605442643165588, 0.10204081982374191, 0.12925170361995697, 0.15646257996559143, 0.08843537420034409, 0.1428571492433548, 0.10204081982374191, 0.15646257996559143, 0.17687074840068817, 0.16326530277729034, 0.21088434755802155, 0.2448979616165161, 0.1972789168357849, 0.2448979616165161, 0.25170066952705383, 0.2789115607738495, 0.17687074840068817, 0.18367347121238708, 0.20408163964748383, 0.2312925159931183, 0.26530611515045166, 0.2925170063972473, 0.21768707036972046, 0.2857142984867096, 0.2380952388048172, 0.30612245202064514, 0.29931971430778503, 0.380952388048172, 0.4217686951160431, 0.14965985715389252, 0.44897958636283875, 0.3741496503353119, 0.40816327929496765, 0.44897958636283875, 0.2857142984867096, 0.20068027079105377, 0.41496598720550537, 0.3945578336715698, 0.3639455735683441, 0.4353741407394409, 0.4761904776096344, 0.4625850319862366, 0.36734694242477417, NaN, 0.4625850319862366, 0.4761904776096344, 0.5850340127944946, 0.5102040767669678, 0.5170068144798279, 0.646258533000946, 0.5986394286155701, 0.6054421663284302, 0.6190476417541504, 0.6598639488220215, 0.8435373902320862, 0.795918345451355, 0.8571428656578064, 0.819727897644043, 0.7891156673431396, 0.6734693646430969, 0.6666666865348816, 0.795918345451355, 0.8775510191917419, 0.8707482814788818, 0.8911564350128174, 0.918367326259613, 0.9115646481513977, 0.8707482814788818, 0.9251700639724731, 0.9727891087532043, 0.918367326259613, 0.8979591727256775, 1.0], [0.06802721321582794, 0.054421767592430115, 0.054421767592430115, 0.0476190485060215, 0.10204081982374191, 0.08163265138864517, 0.10884353518486023, 0.12244898080825806, 0.10884353518486023, 0.1428571492433548, 0.10204081982374191, 0.18367347121238708, 0.12244898080825806, 0.12925170361995697, 0.190476194024086, 0.2312925159931183, 0.12925170361995697, 0.2312925159931183, 0.17006802558898926, 0.27210885286331177, 0.16326530277729034, 0.21088434755802155, 0.2380952388048172, 0.2312925159931183, 0.2448979616165161, 0.27210885286331177, 0.26530611515045166, 0.3265306055545807, 0.27210885286331177, 0.31972789764404297, 0.27210885286331177, 0.30612245202064514, 0.21768707036972046, 0.22448979318141937, 0.40816327929496765, 0.29931971430778503, 0.4761904776096344, 0.18707482516765594, 0.31292515993118286, 0.3877550959587097, 0.4693877696990967, 0.40816327929496765, 0.36734694242477417, 0.4829931855201721, 0.40136054158210754, 0.36734694242477417, 0.49659863114356995, 0.5170068144798279, 0.5374149680137634, NaN, 0.49659863114356995, 0.380952388048172, 0.6394557952880859, 0.5102040767669678, 0.5102040767669678, 0.557823121547699, 0.557823121547699, 0.5986394286155701, 0.8095238208770752, 0.6870748400688171, 0.6598639488220215, 0.680272102355957, 0.7074829936027527, 0.7346938848495483, 0.7346938848495483, 0.7891156673431396, 0.8571428656578064, 0.8639456033706665, 0.8571428656578064, 0.7823129296302795, 0.8299319744110107, 0.8163265585899353, 0.9251700639724731, 0.9659863710403442, 0.9387755393981934, 0.9523809552192688, 0.9863945841789246, 1.0], [0.06802721321582794, 0.020408162847161293, 0.013605441898107529, 0.006802720949053764, 0.054421767592430115, 0.020408162847161293, 0.020408162847161293, 0.07482992857694626, 0.08163265138864517, 0.006802720949053764, 0.040816325694322586, 0.054421767592430115, 0.06802721321582794, 0.1428571492433548, 0.10204081982374191, 0.13605442643165588, 0.11564625799655914, 0.17006802558898926, 0.16326530277729034, 0.1972789168357849, 0.15646257996559143, 0.054421767592430115, 0.12925170361995697, 0.11564625799655914, 0.1972789168357849, 0.06122449040412903, 0.06802721321582794, 0.2380952388048172, 0.15646257996559143, 0.17006802558898926, 0.14965985715389252, 0.25170066952705383, 0.25850340723991394, 0.2925170063972473, 0.29931971430778503, 0.4625850319862366, 0.442176878452301, 0.380952388048172, 0.40136054158210754, 0.3265306055545807, 0.36734694242477417, 0.4353741407394409, 0.3265306055545807, 0.44897958636283875, 0.45578232407569885, 0.4897959232330322, 0.442176878452301, 0.49659863114356995, 0.523809552192688, 0.5034013390541077, NaN, 0.6122449040412903, 0.523809552192688, 0.5782312750816345, 0.680272102355957, 0.646258533000946, 0.680272102355957, 0.6734693646430969, 0.7551020383834839, 0.8027210831642151, 0.8639456033706665, 0.884353756904602, 0.8639456033706665, 0.8639456033706665, 0.7210884094238281, 0.761904776096344, 0.9591836929321289, 0.8775510191917419, 0.8639456033706665, 0.8911564350128174, 0.9251700639724731, 0.918367326259613, 0.9455782175064087, 0.9863945841789246, 0.9251700639724731, 0.9863945841789246, 1.0, 1.0], [0.07482992857694626, 0.054421767592430115, 0.06802721321582794, 0.06122449040412903, 0.07482992857694626, 0.07482992857694626, 0.08163265138864517, 0.095238097012043, 0.08163265138864517, 0.06802721321582794, 0.08163265138864517, 0.10204081982374191, 0.095238097012043, 0.12925170361995697, 0.20408163964748383, 0.2312925159931183, 0.10884353518486023, 0.14965985715389252, 0.13605442643165588, 0.27210885286331177, 0.12925170361995697, 0.12925170361995697, 0.20408163964748383, 0.15646257996559143, 0.2380952388048172, 0.15646257996559143, 0.15646257996559143, 0.2925170063972473, 0.190476194024086, 0.2312925159931183, 0.18367347121238708, 0.2448979616165161, 0.17687074840068817, 0.2380952388048172, 0.29931971430778503, 0.36734694242477417, 0.49659863114356995, 0.4285714328289032, 0.15646257996559143, 0.3333333432674408, 0.35374149680137634, 0.380952388048172, 0.29931971430778503, 0.3945578336715698, 0.4217686951160431, 0.29931971430778503, 0.4897959232330322, 0.4285714328289032, 0.41496598720550537, 0.6190476417541504, 0.3877550959587097, NaN, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.6054421663284302, 0.6258503198623657, 0.6190476417541504, 0.7687074542045593, 0.7551020383834839, 0.7482993006706238, 0.7891156673431396, 0.7210884094238281, 0.7551020383834839, 0.7823129296302795, 0.8571428656578064, 0.8639456033706665, 0.9047619104385376, 0.9115646481513977, 0.8775510191917419, 0.8979591727256775, 0.8707482814788818, 0.9659863710403442, 0.9795918464660645, 0.9727891087532043, 0.9727891087532043, 1.0, 1.0], [0.06122449040412903, 0.03401360660791397, 0.0476190485060215, 0.06122449040412903, 0.06122449040412903, 0.020408162847161293, 0.06122449040412903, 0.10204081982374191, 0.10204081982374191, 0.08843537420034409, 0.06802721321582794, 0.11564625799655914, 0.11564625799655914, 0.13605442643165588, 0.18367347121238708, 0.13605442643165588, 0.12244898080825806, 0.190476194024086, 0.10204081982374191, 0.22448979318141937, 0.16326530277729034, 0.17687074840068817, 0.1972789168357849, 0.21768707036972046, 0.26530611515045166, 0.26530611515045166, 0.25850340723991394, 0.31292515993118286, 0.21768707036972046, 0.22448979318141937, 0.2380952388048172, 0.3401360511779785, 0.22448979318141937, 0.1972789168357849, 0.3877550959587097, 0.2789115607738495, 0.3877550959587097, 0.20408163964748383, 0.3265306055545807, 0.3265306055545807, 0.4829931855201721, 0.4625850319862366, 0.3333333432674408, 0.4625850319862366, 0.3401360511779785, 0.3741496503353119, 0.2448979616165161, 0.5170068144798279, 0.4897959232330322, 0.36054420471191406, 0.4761904776096344, 0.4285714328289032, NaN, 0.4829931855201721, 0.5306122303009033, 0.6258503198623657, 0.5986394286155701, 0.5646258592605591, 0.646258533000946, 0.6598639488220215, 0.7006802558898926, 0.6190476417541504, 0.6326530575752258, 0.7551020383834839, 0.7482993006706238, 0.795918345451355, 0.7891156673431396, 0.7346938848495483, 0.7482993006706238, 0.6598639488220215, 0.7006802558898926, 0.7074829936027527, 0.9863945841789246, 0.8299319744110107, 0.8911564350128174, 0.9387755393981934, 1.0, 0.9931972622871399], [0.020408162847161293, 0.006802720949053764, 0.013605441898107529, 0.020408162847161293, 0.006802720949053764, 0.03401360660791397, 0.013605441898107529, 0.07482992857694626, 0.08163265138864517, 0.006802720949053764, 0.020408162847161293, 0.020408162847161293, 0.12244898080825806, 0.08843537420034409, 0.1428571492433548, 0.15646257996559143, 0.0476190485060215, 0.14965985715389252, 0.12925170361995697, 0.25170066952705383, 0.08843537420034409, 0.054421767592430115, 0.11564625799655914, 0.095238097012043, 0.190476194024086, 0.15646257996559143, 0.15646257996559143, 0.31972789764404297, 0.17006802558898926, 0.18367347121238708, 0.18367347121238708, 0.2925170063972473, 0.1428571492433548, 0.15646257996559143, 0.3333333432674408, 0.25850340723991394, 0.4285714328289032, 0.30612245202064514, 0.26530611515045166, 0.27210885286331177, 0.40816327929496765, 0.442176878452301, 0.2925170063972473, 0.4353741407394409, 0.3401360511779785, 0.35374149680137634, 0.40816327929496765, 0.49659863114356995, 0.4829931855201721, 0.4897959232330322, 0.4217686951160431, 0.4761904776096344, 0.5170068144798279, NaN, 0.5102040767669678, 0.5442177057266235, 0.5646258592605591, 0.557823121547699, 0.7414966225624084, 0.6870748400688171, 0.6530612111091614, 0.6666666865348816, 0.7278911471366882, 0.7482993006706238, 0.7823129296302795, 0.8435373902320862, 0.8299319744110107, 0.8367347121238708, 0.8367347121238708, 0.8299319744110107, 0.8911564350128174, 0.8435373902320862, 0.9319728016853333, 0.9251700639724731, 0.9251700639724731, 0.9387755393981934, 0.9931972622871399, 1.0], [0.07482992857694626, 0.054421767592430115, 0.06802721321582794, 0.027210883796215057, 0.08843537420034409, 0.06802721321582794, 0.07482992857694626, 0.07482992857694626, 0.07482992857694626, 0.07482992857694626, 0.08163265138864517, 0.06122449040412903, 0.06122449040412903, 0.1428571492433548, 0.1428571492433548, 0.14965985715389252, 0.11564625799655914, 0.13605442643165588, 0.1428571492433548, 0.15646257996559143, 0.21088434755802155, 0.11564625799655914, 0.1972789168357849, 0.15646257996559143, 0.20408163964748383, 0.12244898080825806, 0.13605442643165588, 0.14965985715389252, 0.06122449040412903, 0.17006802558898926, 0.08163265138864517, 0.2925170063972473, 0.27210885286331177, 0.35374149680137634, 0.21088434755802155, 0.40816327929496765, 0.26530611515045166, 0.36054420471191406, 0.40816327929496765, 0.2925170063972473, 0.27210885286331177, 0.36734694242477417, 0.27210885286331177, 0.26530611515045166, 0.25170066952705383, 0.5034013390541077, 0.41496598720550537, 0.3469387888908386, 0.35374149680137634, 0.4897959232330322, 0.31972789764404297, 0.523809552192688, 0.4693877696990967, 0.4897959232330322, NaN, 0.4761904776096344, 0.4897959232330322, 0.5102040767669678, 0.646258533000946, 0.761904776096344, 0.761904776096344, 0.7414966225624084, 0.6530612111091614, 0.6258503198623657, 0.5986394286155701, 0.646258533000946, 0.8231292366981506, 0.8027210831642151, 0.7755101919174194, 0.8707482814788818, 0.9115646481513977, 0.8775510191917419, 0.884353756904602, 0.884353756904602, 0.9251700639724731, 0.9387755393981934, 0.9795918464660645, 1.0], [0.07482992857694626, 0.0476190485060215, 0.06122449040412903, 0.027210883796215057, 0.08843537420034409, 0.06802721321582794, 0.095238097012043, 0.08163265138864517, 0.06802721321582794, 0.06122449040412903, 0.08163265138864517, 0.08843537420034409, 0.07482992857694626, 0.13605442643165588, 0.10884353518486023, 0.11564625799655914, 0.1428571492433548, 0.16326530277729034, 0.12244898080825806, 0.1428571492433548, 0.20408163964748383, 0.12925170361995697, 0.12244898080825806, 0.15646257996559143, 0.17687074840068817, 0.11564625799655914, 0.12244898080825806, 0.190476194024086, 0.054421767592430115, 0.10884353518486023, 0.06802721321582794, 0.13605442643165588, 0.25170066952705383, 0.26530611515045166, 0.190476194024086, 0.4693877696990967, 0.3401360511779785, 0.3469387888908386, 0.30612245202064514, 0.22448979318141937, 0.2312925159931183, 0.25850340723991394, 0.21088434755802155, 0.31972789764404297, 0.3877550959587097, 0.36734694242477417, 0.31292515993118286, 0.25850340723991394, 0.40136054158210754, 0.442176878452301, 0.35374149680137634, 0.3945578336715698, 0.3741496503353119, 0.45578232407569885, 0.523809552192688, NaN, 0.47278910875320435, 0.6394557952880859, 0.5986394286155701, 0.6326530575752258, 0.9251700639724731, 0.7414966225624084, 0.6190476417541504, 0.6870748400688171, 0.680272102355957, 0.7006802558898926, 0.8231292366981506, 0.7823129296302795, 0.7891156673431396, 0.7142857313156128, 0.7755101919174194, 0.761904776096344, 0.9455782175064087, 0.9319728016853333, 0.8571428656578064, 0.9115646481513977, 0.9727891087532043, 1.0], [0.07482992857694626, 0.03401360660791397, 0.0476190485060215, 0.013605441898107529, 0.08163265138864517, 0.054421767592430115, 0.07482992857694626, 0.07482992857694626, 0.07482992857694626, 0.054421767592430115, 0.06122449040412903, 0.06122449040412903, 0.08163265138864517, 0.13605442643165588, 0.08163265138864517, 0.11564625799655914, 0.11564625799655914, 0.15646257996559143, 0.11564625799655914, 0.12244898080825806, 0.18367347121238708, 0.08843537420034409, 0.13605442643165588, 0.15646257996559143, 0.17006802558898926, 0.11564625799655914, 0.10884353518486023, 0.17687074840068817, 0.013605441898107529, 0.095238097012043, 0.054421767592430115, 0.11564625799655914, 0.22448979318141937, 0.26530611515045166, 0.18367347121238708, 0.4897959232330322, 0.36734694242477417, 0.3469387888908386, 0.2925170063972473, 0.26530611515045166, 0.2380952388048172, 0.2857142984867096, 0.1972789168357849, 0.31292515993118286, 0.3945578336715698, 0.3401360511779785, 0.3265306055545807, 0.2380952388048172, 0.3945578336715698, 0.442176878452301, 0.31972789764404297, 0.3741496503353119, 0.40136054158210754, 0.4353741407394409, 0.5102040767669678, 0.5272108912467957, NaN, 0.6598639488220215, 0.5918367505073547, 0.6326530575752258, 0.9251700639724731, 0.7482993006706238, 0.6190476417541504, 0.7006802558898926, 0.6666666865348816, 0.7074829936027527, 0.8299319744110107, 0.7891156673431396, 0.795918345451355, 0.7142857313156128, 0.795918345451355, 0.7755101919174194, 0.9659863710403442, 0.9659863710403442, 0.8571428656578064, 0.9387755393981934, 0.9795918464660645, 1.0], [0.07482992857694626, 0.020408162847161293, 0.03401360660791397, 0.013605441898107529, 0.06802721321582794, 0.06802721321582794, 0.020408162847161293, 0.10884353518486023, 0.10884353518486023, 0.06802721321582794, 0.095238097012043, 0.027210883796215057, 0.17006802558898926, 0.14965985715389252, 0.14965985715389252, 0.22448979318141937, 0.1972789168357849, 0.17006802558898926, 0.21088434755802155, 0.18367347121238708, 0.22448979318141937, 0.10204081982374191, 0.10884353518486023, 0.12925170361995697, 0.1428571492433548, 0.21768707036972046, 0.21768707036972046, 0.25850340723991394, 0.2448979616165161, 0.22448979318141937, 0.2312925159931183, 0.040816325694322586, 0.2857142984867096, 0.31292515993118286, 0.190476194024086, 0.41496598720550537, 0.30612245202064514, 0.3469387888908386, 0.26530611515045166, 0.27210885286331177, 0.25170066952705383, 0.08163265138864517, 0.29931971430778503, 0.2857142984867096, 0.40136054158210754, 0.31292515993118286, 0.442176878452301, 0.3333333432674408, 0.380952388048172, 0.40136054158210754, 0.3265306055545807, 0.380952388048172, 0.4353741407394409, 0.442176878452301, 0.4897959232330322, 0.36054420471191406, 0.3401360511779785, NaN, 0.49659863114356995, 0.6258503198623657, 0.557823121547699, 0.6598639488220215, 0.7142857313156128, 0.5306122303009033, 0.6598639488220215, 0.6666666865348816, 0.5850340127944946, 0.8095238208770752, 0.8027210831642151, 0.7414966225624084, 0.7755101919174194, 0.7755101919174194, 0.7278911471366882, 0.7482993006706238, 0.918367326259613, 0.9795918464660645, 0.9319728016853333, 1.0], [0.06802721321582794, 0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.040816325694322586, 0.054421767592430115, 0.03401360660791397, 0.10884353518486023, 0.095238097012043, 0.027210883796215057, 0.03401360660791397, 0.040816325694322586, 0.12244898080825806, 0.10204081982374191, 0.10884353518486023, 0.15646257996559143, 0.054421767592430115, 0.13605442643165588, 0.08843537420034409, 0.1428571492433548, 0.06802721321582794, 0.040816325694322586, 0.07482992857694626, 0.06802721321582794, 0.08843537420034409, 0.08163265138864517, 0.08163265138864517, 0.22448979318141937, 0.11564625799655914, 0.1428571492433548, 0.13605442643165588, 0.14965985715389252, 0.06122449040412903, 0.08843537420034409, 0.22448979318141937, 0.2857142984867096, 0.36734694242477417, 0.10884353518486023, 0.08163265138864517, 0.14965985715389252, 0.27210885286331177, 0.20408163964748383, 0.1428571492433548, 0.29931971430778503, 0.31972789764404297, 0.12244898080825806, 0.3265306055545807, 0.27210885286331177, 0.3401360511779785, 0.190476194024086, 0.2448979616165161, 0.2312925159931183, 0.35374149680137634, 0.25850340723991394, 0.35374149680137634, 0.40136054158210754, 0.40816327929496765, 0.5034013390541077, NaN, 0.6190476417541504, 0.557823121547699, 0.5986394286155701, 0.6122449040412903, 0.6394557952880859, 0.6666666865348816, 0.7142857313156128, 0.523809552192688, 0.7551020383834839, 0.761904776096344, 0.6258503198623657, 0.6938775777816772, 0.6666666865348816, 0.8503401279449463, 0.8027210831642151, 0.9251700639724731, 0.8911564350128174, 0.9387755393981934, 1.0], [0.06802721321582794, 0.06802721321582794, 0.07482992857694626, 0.020408162847161293, 0.12244898080825806, 0.0476190485060215, 0.054421767592430115, 0.06802721321582794, 0.10884353518486023, 0.08163265138864517, 0.07482992857694626, 0.054421767592430115, 0.06122449040412903, 0.12244898080825806, 0.03401360660791397, 0.08163265138864517, 0.11564625799655914, 0.1428571492433548, 0.17687074840068817, 0.040816325694322586, 0.15646257996559143, 0.13605442643165588, 0.08843537420034409, 0.15646257996559143, 0.08163265138864517, 0.15646257996559143, 0.15646257996559143, 0.06802721321582794, 0.14965985715389252, 0.14965985715389252, 0.1428571492433548, 0.08843537420034409, 0.20408163964748383, 0.2380952388048172, 0.0476190485060215, 0.4353741407394409, 0.14965985715389252, 0.27210885286331177, 0.25170066952705383, 0.2380952388048172, 0.095238097012043, 0.12925170361995697, 0.20408163964748383, 0.06802721321582794, 0.15646257996559143, 0.30612245202064514, 0.36734694242477417, 0.18367347121238708, 0.15646257996559143, 0.31292515993118286, 0.1972789168357849, 0.2448979616165161, 0.3401360511779785, 0.31292515993118286, 0.2380952388048172, 0.36734694242477417, 0.36734694242477417, 0.3741496503353119, 0.380952388048172, NaN, 0.5714285969734192, 0.45578232407569885, 0.41496598720550537, 0.442176878452301, 0.4897959232330322, 0.49659863114356995, 0.4829931855201721, 0.4829931855201721, 0.49659863114356995, 0.6870748400688171, 0.7278911471366882, 0.7142857313156128, 0.6326530575752258, 0.5850340127944946, 0.9319728016853333, 0.680272102355957, 0.7482993006706238, 0.9931972622871399], [0.06122449040412903, 0.040816325694322586, 0.06122449040412903, 0.013605441898107529, 0.06802721321582794, 0.054421767592430115, 0.06802721321582794, 0.06802721321582794, 0.06802721321582794, 0.06122449040412903, 0.054421767592430115, 0.054421767592430115, 0.06122449040412903, 0.095238097012043, 0.08843537420034409, 0.11564625799655914, 0.0476190485060215, 0.11564625799655914, 0.07482992857694626, 0.08163265138864517, 0.08843537420034409, 0.08163265138864517, 0.11564625799655914, 0.08163265138864517, 0.08843537420034409, 0.07482992857694626, 0.07482992857694626, 0.13605442643165588, 0.020408162847161293, 0.06802721321582794, 0.03401360660791397, 0.06122449040412903, 0.10204081982374191, 0.18367347121238708, 0.12925170361995697, 0.36054420471191406, 0.17687074840068817, 0.31292515993118286, 0.17006802558898926, 0.16326530277729034, 0.14965985715389252, 0.11564625799655914, 0.06122449040412903, 0.14965985715389252, 0.21768707036972046, 0.21768707036972046, 0.27210885286331177, 0.14965985715389252, 0.20408163964748383, 0.3401360511779785, 0.13605442643165588, 0.25170066952705383, 0.29931971430778503, 0.3469387888908386, 0.2380952388048172, 0.07482992857694626, 0.07482992857694626, 0.442176878452301, 0.442176878452301, 0.4285714328289032, NaN, 0.5306122303009033, 0.4693877696990967, 0.27210885286331177, 0.5850340127944946, 0.5918367505073547, 0.6734693646430969, 0.7006802558898926, 0.7210884094238281, 0.6258503198623657, 0.680272102355957, 0.6598639488220215, 0.7346938848495483, 0.646258533000946, 0.8435373902320862, 0.8095238208770752, 0.9047619104385376, 1.0], [0.06802721321582794, 0.020408162847161293, 0.013605441898107529, 0.006802720949053764, 0.013605441898107529, 0.03401360660791397, 0.006802720949053764, 0.08843537420034409, 0.08843537420034409, 0.013605441898107529, 0.020408162847161293, 0.027210883796215057, 0.095238097012043, 0.10884353518486023, 0.040816325694322586, 0.06122449040412903, 0.06122449040412903, 0.10204081982374191, 0.095238097012043, 0.040816325694322586, 0.07482992857694626, 0.027210883796215057, 0.0476190485060215, 0.027210883796215057, 0.040816325694322586, 0.040816325694322586, 0.0476190485060215, 0.10204081982374191, 0.08163265138864517, 0.13605442643165588, 0.06802721321582794, 0.027210883796215057, 0.12244898080825806, 0.16326530277729034, 0.054421767592430115, 0.3469387888908386, 0.190476194024086, 0.2380952388048172, 0.12925170361995697, 0.13605442643165588, 0.10204081982374191, 0.054421767592430115, 0.10884353518486023, 0.08843537420034409, 0.07482992857694626, 0.18367347121238708, 0.36734694242477417, 0.18367347121238708, 0.1428571492433548, 0.31972789764404297, 0.11564625799655914, 0.21088434755802155, 0.380952388048172, 0.3333333432674408, 0.25850340723991394, 0.25850340723991394, 0.25170066952705383, 0.3401360511779785, 0.40136054158210754, 0.5442177057266235, 0.4693877696990967, NaN, 0.5646258592605591, 0.442176878452301, 0.5442177057266235, 0.5646258592605591, 0.5170068144798279, 0.6870748400688171, 0.6666666865348816, 0.6598639488220215, 0.7551020383834839, 0.7346938848495483, 0.7346938848495483, 0.7482993006706238, 0.9115646481513977, 0.8911564350128174, 0.8911564350128174, 1.0], [0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.006802720949053764, 0.040816325694322586, 0.0, 0.0, 0.006802720949053764, 0.10204081982374191, 0.027210883796215057, 0.027210883796215057, 0.0, 0.03741496428847313, 0.10884353518486023, 0.020408162847161293, 0.0, 0.10884353518486023, 0.1190476194024086, 0.10204081982374191, 0.040816325694322586, 0.11564625799655914, 0.08163265138864517, 0.054421767592430115, 0.08843537420034409, 0.08163265138864517, 0.08843537420034409, 0.095238097012043, 0.05782312899827957, 0.10204081982374191, 0.08163265138864517, 0.11564625799655914, 0.03401360660791397, 0.17687074840068817, 0.190476194024086, 0.03401360660791397, 0.3469387888908386, 0.1972789168357849, 0.25850340723991394, 0.21088434755802155, 0.13605442643165588, 0.013605441898107529, 0.08163265138864517, 0.15646257996559143, 0.10884353518486023, 0.1428571492433548, 0.2448979616165161, 0.35374149680137634, 0.14965985715389252, 0.18027210235595703, 0.2925170063972473, 0.13605442643165588, 0.2789115607738495, 0.36734694242477417, 0.27210885286331177, 0.3469387888908386, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.3877550959587097, 0.5850340127944946, 0.5306122303009033, 0.4353741407394409, NaN, 0.49659863114356995, 0.49659863114356995, 0.557823121547699, 0.45578232407569885, 0.5442177057266235, 0.5442177057266235, 0.6938775777816772, 0.7278911471366882, 0.7006802558898926, 0.6598639488220215, 0.6394557952880859, 0.884353756904602, 0.8231292366981506, 0.8639456033706665, 1.0], [0.0, 0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.0, 0.0, 0.027210883796215057, 0.06122449040412903, 0.0, 0.020408162847161293, 0.0, 0.06122449040412903, 0.07482992857694626, 0.0, 0.0, 0.08843537420034409, 0.1428571492433548, 0.07482992857694626, 0.020408162847161293, 0.14965985715389252, 0.054421767592430115, 0.013605441898107529, 0.07482992857694626, 0.03401360660791397, 0.07482992857694626, 0.08843537420034409, 0.06802721321582794, 0.054421767592430115, 0.006802720949053764, 0.054421767592430115, 0.006802720949053764, 0.18367347121238708, 0.1972789168357849, 0.0, 0.36734694242477417, 0.08163265138864517, 0.21088434755802155, 0.1972789168357849, 0.07482992857694626, 0.0476190485060215, 0.054421767592430115, 0.11564625799655914, 0.11564625799655914, 0.2380952388048172, 0.22448979318141937, 0.21768707036972046, 0.16326530277729034, 0.21088434755802155, 0.26530611515045166, 0.13605442643165588, 0.2448979616165161, 0.2448979616165161, 0.25170066952705383, 0.3741496503353119, 0.31292515993118286, 0.29931971430778503, 0.4693877696990967, 0.36054420471191406, 0.557823121547699, 0.7278911471366882, 0.557823121547699, 0.5034013390541077, NaN, 0.5306122303009033, 0.5306122303009033, 0.5918367505073547, 0.6122449040412903, 0.6190476417541504, 0.6394557952880859, 0.7278911471366882, 0.6938775777816772, 0.7755101919174194, 0.7006802558898926, 0.8367347121238708, 0.7755101919174194, 0.8503401279449463, 1.0], [0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.027210883796215057, 0.0, 0.020408162847161293, 0.013605441898107529, 0.013605441898107529, 0.013605441898107529, 0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.020408162847161293, 0.03401360660791397, 0.1428571492433548, 0.13605442643165588, 0.040816325694322586, 0.06122449040412903, 0.054421767592430115, 0.190476194024086, 0.027210883796215057, 0.040816325694322586, 0.08843537420034409, 0.0476190485060215, 0.11564625799655914, 0.054421767592430115, 0.054421767592430115, 0.190476194024086, 0.11564625799655914, 0.18367347121238708, 0.10884353518486023, 0.040816325694322586, 0.06122449040412903, 0.08843537420034409, 0.2380952388048172, 0.1972789168357849, 0.4217686951160431, 0.12925170361995697, 0.06122449040412903, 0.2312925159931183, 0.2925170063972473, 0.12925170361995697, 0.16326530277729034, 0.2925170063972473, 0.3401360511779785, 0.10884353518486023, 0.190476194024086, 0.2312925159931183, 0.3265306055545807, 0.26530611515045166, 0.2789115607738495, 0.21768707036972046, 0.25170066952705383, 0.21768707036972046, 0.40136054158210754, 0.31972789764404297, 0.3333333432674408, 0.3401360511779785, 0.3333333432674408, 0.5102040767669678, 0.41496598720550537, 0.45578232407569885, 0.5034013390541077, 0.4693877696990967, NaN, 0.7448979616165161, 0.45578232407569885, 0.45578232407569885, 0.4829931855201721, 0.4761904776096344, 0.4897959232330322, 0.5170068144798279, 0.45578232407569885, 0.5306122303009033, 0.7551020383834839, 0.6666666865348816, 0.9455782175064087, 0.9319728016853333], [0.006802720949053764, 0.0, 0.006802720949053764, 0.020408162847161293, 0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.020408162847161293, 0.006802720949053764, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.03401360660791397, 0.03401360660791397, 0.12925170361995697, 0.14965985715389252, 0.020408162847161293, 0.07482992857694626, 0.040816325694322586, 0.1972789168357849, 0.027210883796215057, 0.03401360660791397, 0.040816325694322586, 0.03401360660791397, 0.06802721321582794, 0.040816325694322586, 0.040816325694322586, 0.17687074840068817, 0.07482992857694626, 0.11564625799655914, 0.07482992857694626, 0.03401360660791397, 0.06122449040412903, 0.06802721321582794, 0.2380952388048172, 0.1972789168357849, 0.35374149680137634, 0.08163265138864517, 0.027210883796215057, 0.17006802558898926, 0.2857142984867096, 0.12244898080825806, 0.13605442643165588, 0.31292515993118286, 0.30612245202064514, 0.07482992857694626, 0.1428571492433548, 0.190476194024086, 0.3333333432674408, 0.21088434755802155, 0.2380952388048172, 0.1428571492433548, 0.20408163964748383, 0.15646257996559143, 0.35374149680137634, 0.29931971430778503, 0.2925170063972473, 0.3333333432674408, 0.2857142984867096, 0.5034013390541077, 0.40816327929496765, 0.4353741407394409, 0.442176878452301, 0.4693877696990967, 0.2551020383834839, NaN, 0.44897958636283875, 0.40136054158210754, 0.41496598720550537, 0.44897958636283875, 0.4761904776096344, 0.5034013390541077, 0.4353741407394409, 0.523809552192688, 0.7414966225624084, 0.6530612111091614, 0.8571428656578064, 0.9319728016853333], [0.06122449040412903, 0.0, 0.0, 0.0, 0.013605441898107529, 0.0, 0.006802720949053764, 0.06122449040412903, 0.06122449040412903, 0.0, 0.006802720949053764, 0.027210883796215057, 0.06122449040412903, 0.06122449040412903, 0.06122449040412903, 0.06122449040412903, 0.0, 0.10884353518486023, 0.020408162847161293, 0.06802721321582794, 0.0, 0.0, 0.020408162847161293, 0.0, 0.027210883796215057, 0.006802720949053764, 0.006802720949053764, 0.10884353518486023, 0.0, 0.006802720949053764, 0.0, 0.10204081982374191, 0.006802720949053764, 0.013605441898107529, 0.08843537420034409, 0.2789115607738495, 0.08163265138864517, 0.08843537420034409, 0.054421767592430115, 0.020408162847161293, 0.15646257996559143, 0.1428571492433548, 0.0, 0.14965985715389252, 0.16326530277729034, 0.095238097012043, 0.21088434755802155, 0.16326530277729034, 0.20408163964748383, 0.1428571492433548, 0.040816325694322586, 0.13605442643165588, 0.21088434755802155, 0.17006802558898926, 0.17687074840068817, 0.17687074840068817, 0.17006802558898926, 0.41496598720550537, 0.4761904776096344, 0.5170068144798279, 0.3265306055545807, 0.4829931855201721, 0.5442177057266235, 0.40816327929496765, 0.5442177057266235, 0.5510203838348389, NaN, 0.6870748400688171, 0.680272102355957, 0.6530612111091614, 0.680272102355957, 0.680272102355957, 0.6598639488220215, 0.6938775777816772, 0.7551020383834839, 0.7755101919174194, 0.7687074542045593, 0.9319728016853333], [0.06122449040412903, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.013605441898107529, 0.006802720949053764, 0.0, 0.06802721321582794, 0.07482992857694626, 0.027210883796215057, 0.027210883796215057, 0.020408162847161293, 0.07482992857694626, 0.095238097012043, 0.095238097012043, 0.12244898080825806, 0.06122449040412903, 0.08843537420034409, 0.06122449040412903, 0.095238097012043, 0.07482992857694626, 0.020408162847161293, 0.08843537420034409, 0.03401360660791397, 0.07482992857694626, 0.0476190485060215, 0.040816325694322586, 0.14965985715389252, 0.054421767592430115, 0.13605442643165588, 0.0476190485060215, 0.03401360660791397, 0.0476190485060215, 0.07482992857694626, 0.08843537420034409, 0.26530611515045166, 0.2380952388048172, 0.10204081982374191, 0.06802721321582794, 0.15646257996559143, 0.07482992857694626, 0.07482992857694626, 0.08163265138864517, 0.08843537420034409, 0.12925170361995697, 0.08163265138864517, 0.2312925159931183, 0.21768707036972046, 0.12244898080825806, 0.13605442643165588, 0.12244898080825806, 0.095238097012043, 0.26530611515045166, 0.16326530277729034, 0.1972789168357849, 0.21768707036972046, 0.21088434755802155, 0.190476194024086, 0.2448979616165161, 0.5170068144798279, 0.29931971430778503, 0.31292515993118286, 0.45578232407569885, 0.3877550959587097, 0.5442177057266235, 0.5986394286155701, 0.31292515993118286, NaN, 0.5374149680137634, 0.5102040767669678, 0.5510203838348389, 0.5306122303009033, 0.5850340127944946, 0.4285714328289032, 0.8503401279449463, 1.0, 0.9047619104385376, 0.9863945841789246], [0.06802721321582794, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.020408162847161293, 0.006802720949053764, 0.020408162847161293, 0.06802721321582794, 0.07482992857694626, 0.03401360660791397, 0.040816325694322586, 0.03401360660791397, 0.07482992857694626, 0.10884353518486023, 0.095238097012043, 0.12244898080825806, 0.0476190485060215, 0.095238097012043, 0.06122449040412903, 0.095238097012043, 0.0476190485060215, 0.027210883796215057, 0.095238097012043, 0.040816325694322586, 0.06802721321582794, 0.054421767592430115, 0.0476190485060215, 0.14965985715389252, 0.054421767592430115, 0.1428571492433548, 0.040816325694322586, 0.03401360660791397, 0.0476190485060215, 0.06802721321582794, 0.08843537420034409, 0.2789115607738495, 0.25170066952705383, 0.095238097012043, 0.06122449040412903, 0.15646257996559143, 0.07482992857694626, 0.06802721321582794, 0.08843537420034409, 0.095238097012043, 0.1428571492433548, 0.07482992857694626, 0.2312925159931183, 0.21768707036972046, 0.12925170361995697, 0.1428571492433548, 0.13605442643165588, 0.08843537420034409, 0.25170066952705383, 0.16326530277729034, 0.22448979318141937, 0.21088434755802155, 0.20408163964748383, 0.1972789168357849, 0.2380952388048172, 0.5034013390541077, 0.2789115607738495, 0.3333333432674408, 0.45578232407569885, 0.380952388048172, 0.5170068144798279, 0.5850340127944946, 0.31972789764404297, 0.4625850319862366, NaN, 0.49659863114356995, 0.5374149680137634, 0.5306122303009033, 0.5714285969734192, 0.4217686951160431, 0.8503401279449463, 1.0, 0.884353756904602, 0.9863945841789246], [0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.006802720949053764, 0.013605441898107529, 0.03401360660791397, 0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.020408162847161293, 0.013605441898107529, 0.020408162847161293, 0.0476190485060215, 0.03401360660791397, 0.040816325694322586, 0.06122449040412903, 0.0476190485060215, 0.020408162847161293, 0.06122449040412903, 0.06802721321582794, 0.040816325694322586, 0.027210883796215057, 0.020408162847161293, 0.027210883796215057, 0.03401360660791397, 0.040816325694322586, 0.040816325694322586, 0.054421767592430115, 0.06122449040412903, 0.10884353518486023, 0.06122449040412903, 0.08843537420034409, 0.06802721321582794, 0.06122449040412903, 0.07482992857694626, 0.2448979616165161, 0.14965985715389252, 0.16326530277729034, 0.08163265138864517, 0.10884353518486023, 0.08843537420034409, 0.10884353518486023, 0.08163265138864517, 0.08843537420034409, 0.06802721321582794, 0.10204081982374191, 0.31972789764404297, 0.12244898080825806, 0.10884353518486023, 0.21768707036972046, 0.10884353518486023, 0.12244898080825806, 0.3401360511779785, 0.17006802558898926, 0.12925170361995697, 0.2857142984867096, 0.2857142984867096, 0.25850340723991394, 0.3741496503353119, 0.31292515993118286, 0.3741496503353119, 0.3401360511779785, 0.30612245202064514, 0.36054420471191406, 0.523809552192688, 0.5510203838348389, 0.3469387888908386, 0.4897959232330322, 0.5034013390541077, NaN, 0.7755101919174194, 0.8027210831642151, 0.5850340127944946, 0.6326530575752258, 0.6258503198623657, 0.5510203838348389, 0.6598639488220215, 0.9387755393981934], [0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.0476190485060215, 0.027210883796215057, 0.03401360660791397, 0.03401360660791397, 0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.0476190485060215, 0.040816325694322586, 0.040816325694322586, 0.06122449040412903, 0.03401360660791397, 0.03401360660791397, 0.054421767592430115, 0.054421767592430115, 0.0476190485060215, 0.027210883796215057, 0.027210883796215057, 0.027210883796215057, 0.03401360660791397, 0.040816325694322586, 0.0476190485060215, 0.06122449040412903, 0.0476190485060215, 0.08843537420034409, 0.0476190485060215, 0.10884353518486023, 0.040816325694322586, 0.06802721321582794, 0.0476190485060215, 0.25170066952705383, 0.12925170361995697, 0.12244898080825806, 0.06802721321582794, 0.08843537420034409, 0.06802721321582794, 0.08163265138864517, 0.054421767592430115, 0.08843537420034409, 0.054421767592430115, 0.08843537420034409, 0.2789115607738495, 0.095238097012043, 0.08163265138864517, 0.17006802558898926, 0.07482992857694626, 0.10204081982374191, 0.29931971430778503, 0.10884353518486023, 0.08843537420034409, 0.22448979318141937, 0.20408163964748383, 0.22448979318141937, 0.30612245202064514, 0.27210885286331177, 0.31972789764404297, 0.2448979616165161, 0.27210885286331177, 0.27210885286331177, 0.5102040767669678, 0.523809552192688, 0.31972789764404297, 0.44897958636283875, 0.4625850319862366, 0.22448979318141937, NaN, 0.6496598720550537, 0.557823121547699, 0.5170068144798279, 0.6190476417541504, 0.5034013390541077, 0.6394557952880859, 0.9319728016853333], [0.03401360660791397, 0.03401360660791397, 0.027210883796215057, 0.03401360660791397, 0.027210883796215057, 0.03401360660791397, 0.03401360660791397, 0.03401360660791397, 0.03401360660791397, 0.03401360660791397, 0.040816325694322586, 0.03401360660791397, 0.040816325694322586, 0.027210883796215057, 0.0476190485060215, 0.054421767592430115, 0.0476190485060215, 0.03401360660791397, 0.0476190485060215, 0.0476190485060215, 0.054421767592430115, 0.03401360660791397, 0.0476190485060215, 0.03401360660791397, 0.054421767592430115, 0.03401360660791397, 0.040816325694322586, 0.054421767592430115, 0.06122449040412903, 0.08163265138864517, 0.06122449040412903, 0.10884353518486023, 0.06122449040412903, 0.07482992857694626, 0.06122449040412903, 0.25170066952705383, 0.11564625799655914, 0.12925170361995697, 0.06802721321582794, 0.095238097012043, 0.06802721321582794, 0.095238097012043, 0.08163265138864517, 0.08163265138864517, 0.054421767592430115, 0.095238097012043, 0.2789115607738495, 0.10204081982374191, 0.08843537420034409, 0.18367347121238708, 0.08163265138864517, 0.12925170361995697, 0.2925170063972473, 0.15646257996559143, 0.12244898080825806, 0.2380952388048172, 0.22448979318141937, 0.22448979318141937, 0.3333333432674408, 0.2857142984867096, 0.3401360511779785, 0.26530611515045166, 0.29931971430778503, 0.30612245202064514, 0.4829931855201721, 0.49659863114356995, 0.31972789764404297, 0.4693877696990967, 0.4693877696990967, 0.1972789168357849, 0.3503401279449463, NaN, 0.5374149680137634, 0.5374149680137634, 0.5918367505073547, 0.5374149680137634, 0.6122449040412903, 0.7414966225624084], [0.0, 0.0, 0.006802720949053764, 0.006802720949053764, 0.006802720949053764, 0.0, 0.006802720949053764, 0.013605441898107529, 0.06122449040412903, 0.0, 0.0, 0.0, 0.06802721321582794, 0.08163265138864517, 0.06122449040412903, 0.06122449040412903, 0.020408162847161293, 0.095238097012043, 0.020408162847161293, 0.06122449040412903, 0.027210883796215057, 0.020408162847161293, 0.054421767592430115, 0.027210883796215057, 0.06122449040412903, 0.013605441898107529, 0.013605441898107529, 0.12244898080825806, 0.0, 0.006802720949053764, 0.0, 0.03401360660791397, 0.006802720949053764, 0.020408162847161293, 0.10204081982374191, 0.21088434755802155, 0.095238097012043, 0.040816325694322586, 0.020408162847161293, 0.06122449040412903, 0.11564625799655914, 0.07482992857694626, 0.03401360660791397, 0.11564625799655914, 0.12925170361995697, 0.03401360660791397, 0.006802720949053764, 0.1428571492433548, 0.12925170361995697, 0.07482992857694626, 0.054421767592430115, 0.03401360660791397, 0.013605441898107529, 0.06802721321582794, 0.11564625799655914, 0.054421767592430115, 0.03401360660791397, 0.27210885286331177, 0.14965985715389252, 0.36734694242477417, 0.26530611515045166, 0.26530611515045166, 0.3401360511779785, 0.22448979318141937, 0.5442177057266235, 0.5646258592605591, 0.3401360511779785, 0.41496598720550537, 0.4285714328289032, 0.41496598720550537, 0.442176878452301, 0.4625850319862366, NaN, 0.3469387888908386, 0.7823129296302795, 0.761904776096344, 0.8979591727256775, 0.9931972622871399], [0.03401360660791397, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.054421767592430115, 0.06122449040412903, 0.0, 0.0, 0.006802720949053764, 0.06122449040412903, 0.06122449040412903, 0.0, 0.0, 0.0, 0.08843537420034409, 0.020408162847161293, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.06122449040412903, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.020408162847161293, 0.25170066952705383, 0.0, 0.006802720949053764, 0.006802720949053764, 0.027210883796215057, 0.054421767592430115, 0.006802720949053764, 0.0, 0.03401360660791397, 0.0476190485060215, 0.006802720949053764, 0.08163265138864517, 0.07482992857694626, 0.07482992857694626, 0.03401360660791397, 0.013605441898107529, 0.020408162847161293, 0.17006802558898926, 0.07482992857694626, 0.11564625799655914, 0.06802721321582794, 0.03401360660791397, 0.25170066952705383, 0.1972789168357849, 0.41496598720550537, 0.35374149680137634, 0.25170066952705383, 0.36054420471191406, 0.29931971430778503, 0.4693877696990967, 0.4761904776096344, 0.30612245202064514, 0.5714285969734192, 0.5782312750816345, 0.36734694242477417, 0.4829931855201721, 0.4625850319862366, 0.6530612111091614, NaN, 0.795918345451355, 0.7687074542045593, 0.8163265585899353, 1.0], [0.06802721321582794, 0.03401360660791397, 0.0476190485060215, 0.020408162847161293, 0.054421767592430115, 0.040816325694322586, 0.054421767592430115, 0.06122449040412903, 0.06122449040412903, 0.0476190485060215, 0.027210883796215057, 0.0476190485060215, 0.06122449040412903, 0.06122449040412903, 0.027210883796215057, 0.054421767592430115, 0.03401360660791397, 0.06802721321582794, 0.0476190485060215, 0.040816325694322586, 0.03401360660791397, 0.0476190485060215, 0.06802721321582794, 0.0476190485060215, 0.027210883796215057, 0.06802721321582794, 0.06802721321582794, 0.06122449040412903, 0.0476190485060215, 0.10204081982374191, 0.0476190485060215, 0.013605441898107529, 0.027210883796215057, 0.040816325694322586, 0.0, 0.2380952388048172, 0.12244898080825806, 0.06122449040412903, 0.06122449040412903, 0.11564625799655914, 0.006802720949053764, 0.027210883796215057, 0.054421767592430115, 0.0, 0.0476190485060215, 0.06122449040412903, 0.10884353518486023, 0.10204081982374191, 0.027210883796215057, 0.06122449040412903, 0.07482992857694626, 0.027210883796215057, 0.10884353518486023, 0.07482992857694626, 0.07482992857694626, 0.1428571492433548, 0.1428571492433548, 0.08163265138864517, 0.07482992857694626, 0.06802721321582794, 0.15646257996559143, 0.08843537420034409, 0.11564625799655914, 0.16326530277729034, 0.2448979616165161, 0.25850340723991394, 0.2448979616165161, 0.14965985715389252, 0.14965985715389252, 0.3741496503353119, 0.380952388048172, 0.40816327929496765, 0.21768707036972046, 0.20408163964748383, NaN, 0.3741496503353119, 0.3741496503353119, 0.9251700639724731], [0.006802720949053764, 0.013605441898107529, 0.006802720949053764, 0.006802720949053764, 0.0, 0.006802720949053764, 0.0, 0.006802720949053764, 0.06802721321582794, 0.006802720949053764, 0.013605441898107529, 0.0, 0.06802721321582794, 0.06802721321582794, 0.06802721321582794, 0.06122449040412903, 0.006802720949053764, 0.07482992857694626, 0.03401360660791397, 0.07482992857694626, 0.013605441898107529, 0.006802720949053764, 0.03401360660791397, 0.006802720949053764, 0.03401360660791397, 0.013605441898107529, 0.013605441898107529, 0.11564625799655914, 0.006802720949053764, 0.027210883796215057, 0.006802720949053764, 0.013605441898107529, 0.013605441898107529, 0.020408162847161293, 0.06802721321582794, 0.054421767592430115, 0.10884353518486023, 0.027210883796215057, 0.013605441898107529, 0.054421767592430115, 0.06122449040412903, 0.020408162847161293, 0.006802720949053764, 0.06802721321582794, 0.06802721321582794, 0.020408162847161293, 0.0476190485060215, 0.13605442643165588, 0.08163265138864517, 0.0476190485060215, 0.013605441898107529, 0.027210883796215057, 0.06122449040412903, 0.06122449040412903, 0.06122449040412903, 0.08843537420034409, 0.06122449040412903, 0.020408162847161293, 0.10884353518486023, 0.31972789764404297, 0.190476194024086, 0.10884353518486023, 0.17687074840068817, 0.22448979318141937, 0.3333333432674408, 0.3469387888908386, 0.22448979318141937, 0.0, 0.0, 0.44897958636283875, 0.49659863114356995, 0.4625850319862366, 0.2380952388048172, 0.2312925159931183, 0.6258503198623657, NaN, 0.680272102355957, 0.9591836929321289], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.03401360660791397, 0.0, 0.006802720949053764, 0.006802720949053764, 0.06122449040412903, 0.0, 0.0, 0.006802720949053764, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.054421767592430115, 0.190476194024086, 0.06802721321582794, 0.006802720949053764, 0.006802720949053764, 0.027210883796215057, 0.08163265138864517, 0.006802720949053764, 0.0, 0.07482992857694626, 0.03401360660791397, 0.006802720949053764, 0.0, 0.020408162847161293, 0.10204081982374191, 0.013605441898107529, 0.0, 0.0, 0.0, 0.006802720949053764, 0.020408162847161293, 0.027210883796215057, 0.020408162847161293, 0.06802721321582794, 0.06122449040412903, 0.25170066952705383, 0.095238097012043, 0.10884353518486023, 0.13605442643165588, 0.14965985715389252, 0.054421767592430115, 0.1428571492433548, 0.2312925159931183, 0.095238097012043, 0.11564625799655914, 0.3401360511779785, 0.36054420471191406, 0.3877550959587097, 0.10204081982374191, 0.18367347121238708, 0.6258503198623657, 0.31972789764404297, NaN, 0.9115646481513977], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.006802720949053764, 0.0, 0.0, 0.0, 0.0, 0.06802721321582794, 0.06802721321582794, 0.06802721321582794, 0.013605441898107529, 0.013605441898107529, 0.06122449040412903, 0.06802721321582794, 0.25850340723991394, 0.006802720949053764, 0.0, 0.07482992857694626, 0.040816325694322586, 0.08843537420034409, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9077657028},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8765350282},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8737520982},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8612951683},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8268840027},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8265747852},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8203904964},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.7960950624},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7934446524},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7926937017},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7723738849},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7612863336},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.7539093608},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.7278028097},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7083664644},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6967930029},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6942751148},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.67916777},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6727626143},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.662072621},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.659996466},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6589363014},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6505433368},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6296934317},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6174131992},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.615999649},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6085784958},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6069882513},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5993904059},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.597579292},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.594531321},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.594487145},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5933386379},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5758459278},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5665253113},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5631239507},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5384751306},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.520408165},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5186412228},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5178019276},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5166534155},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5080837508},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5020761546},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.5},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.4821980736},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4635126789},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4626733785},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4434137301},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4410725313},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4307359329},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4266277952},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4202668081},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3968548446},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.3889919608},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3800689126},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.3647407033},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3620902907},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3555084367},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.300998323},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2704302496},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.2589451357},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.254704481},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2536443162},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2489619212},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2417616401},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2147274502},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2121212119},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2093824534},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2083222905},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1977206459},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1705539368},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1694054253},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1561091974},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1506316802},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1098153547},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1087551905},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0738581145},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0110433785}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d423ec9b007e7ff62d026ac3fa052a4599f49477
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_multiclass/datasets_small/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (default),TabPFN-2.6 (default),TabPFN-3 (default),TabICLv2 (default),TabPFNv2 (tuned + ensembled),RealMLP (tuned + ensembled),RealMLP (tuned),LimiX (default),Mitra (default),TabPFNv2 (tuned),TabDPT (tuned + ensembled),TabDPT-Turbo (default),TabM (tuned + ensembled),TabPFNv2 (default),"AutoGluon 1.4 (best, 4h)",TabDPT (tuned),TabM (tuned),CatBoost (tuned + ensembled),ModernNCA (tuned + ensembled),CatBoost (tuned),ModernNCA (tuned),XGBoost (tuned),XGBoost (tuned + ensembled),TabICL (default),LightGBM (tuned + ensembled),iLTM (tuned + ensembled),LightGBM (tuned),xRFM (tuned + ensembled),BetaTabPFN (default),TabM (default),ExtraTrees (tuned + ensembled),SAP-RPT-OSS (default),TabDPT (default),TorchMLP (tuned + ensembled),EBM (tuned + ensembled),iLTM (tuned),RandomForest (tuned + ensembled),xRFM (tuned),CatBoost (default),ExtraTrees (tuned),TabSwift (default),EBM (tuned),FastaiMLP (tuned + ensembled),ModernNCA (default),RandomForest (tuned),TorchMLP (tuned),XGBoost (default),EBM (default),FastaiMLP (tuned),RealMLP (default),LightGBM (default),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),xRFM (default),TorchMLP (default),KNN (tuned + ensembled),ChimeraBoost (default),OrionMSP (default),RandomForest (default),ExtraTrees (default),Linear (tuned + ensembled),Linear (tuned),iLTM (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),PerpetualBooster (tuned + ensembled),PerpetualBooster (tuned),PerpetualBooster (default),FastaiMLP (default),TabFlex (default),KNN (tuned),TabSTAR (default),Linear (default),KNN (default)
+TabFM (default),,0.70068026,0.70068026,0.81632656,0.7414966,0.8367347,0.8503401,0.89115644,0.829932,0.76870745,0.76870745,0.8435374,0.9183673,0.91156465,0.8979592,0.88435376,0.8095238,0.91156465,0.92517006,0.88435376,0.82312924,0.9047619,0.9183673,0.92517006,0.89115644,0.877551,0.88435376,0.9727891,0.9047619,0.9183673,0.8979592,0.9183673,0.91156465,0.8503401,0.92517006,0.75510204,0.9047619,0.92517006,0.9047619,0.9319728,0.9047619,0.92517006,0.9319728,0.9183673,0.9183673,0.92517006,0.8979592,0.97959185,0.9047619,0.9319728,0.9319728,0.92517006,0.93877554,0.97959185,0.92517006,0.92517006,0.92517006,0.92517006,0.9319728,0.9319728,0.93877554,0.9319728,0.99319726,1.0,0.99319726,0.99319726,0.93877554,0.93877554,0.9319728,0.99319726,0.9727891,0.9659864,1.0,0.9659864,0.9319728,0.99319726,1.0,1.0
+RealTabPFN-2.5 (tuned),0.2993197,,0.46258503,0.5782313,0.68367344,0.61904764,0.61904764,0.6530612,0.707483,0.7414966,0.76870745,0.7346939,0.707483,0.8027211,0.72789115,0.6938776,0.8435374,0.877551,0.88435376,0.7755102,0.8503401,0.92517006,0.8503401,0.92517006,0.8639456,0.9047619,0.9047619,0.78911567,0.88435376,0.877551,0.91156465,0.89115644,0.8979592,0.85714287,0.8639456,0.7482993,0.79591835,0.9455782,0.95238096,0.95238096,0.877551,0.9183673,0.9659864,0.89115644,0.9727891,0.9591837,0.92517006,0.9455782,0.9047619,0.9455782,0.97959185,0.9455782,0.9659864,0.99319726,0.9455782,0.95238096,0.9659864,0.97959185,0.99319726,0.9319728,0.9591837,0.97959185,0.9863946,0.99319726,0.99319726,1.0,1.0,0.9863946,0.9863946,0.9863946,0.9727891,0.9659864,1.0,1.0,0.9659864,0.9863946,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.2993197,0.53741497,,0.60544217,0.7482993,0.6258503,0.6258503,0.65986395,0.75510204,0.7346939,0.7482993,0.7414966,0.707483,0.79591835,0.7210884,0.68707484,0.79591835,0.8707483,0.85714287,0.7414966,0.79591835,0.93877554,0.85714287,0.92517006,0.877551,0.8979592,0.9183673,0.78911567,0.89115644,0.88435376,0.8979592,0.877551,0.91156465,0.8435374,0.8435374,0.7414966,0.7755102,0.93877554,0.9319728,0.95238096,0.8707483,0.9183673,0.93877554,0.8707483,0.9727891,0.93877554,0.9319728,0.9319728,0.9047619,0.9455782,0.9863946,0.9319728,0.95238096,0.9863946,0.9319728,0.93877554,0.95238096,0.9659864,0.9863946,0.92517006,0.93877554,0.9863946,0.9863946,0.99319726,0.99319726,0.99319726,1.0,0.99319726,0.99319726,0.9863946,0.9727891,0.9727891,0.99319726,1.0,0.95238096,0.99319726,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.18367347,0.4217687,0.39455783,,0.4829932,0.5986394,0.5714286,0.5714286,0.5578231,0.7210884,0.7346939,0.78911567,0.71428573,0.6938776,0.81632656,0.7346939,0.76870745,0.82312924,0.79591835,0.85714287,0.7619048,0.88435376,0.8639456,0.8979592,0.88435376,0.829932,0.85714287,0.8979592,0.9047619,0.91156465,0.89115644,0.9047619,0.82312924,0.8027211,0.8979592,0.6258503,0.8367347,0.9047619,0.9183673,0.93877554,0.9047619,0.93877554,0.9455782,0.8979592,0.9863946,0.9455782,0.8707483,0.9659864,0.9047619,0.95238096,0.99319726,0.93877554,0.93877554,0.97959185,0.9727891,0.9727891,0.9863946,0.9863946,0.9863946,0.97959185,0.9863946,0.99319726,0.99319726,0.99319726,0.9727891,0.97959185,1.0,0.99319726,0.99319726,0.99319726,0.9727891,0.9659864,0.99319726,1.0,0.97959185,0.99319726,1.0,1.0
+RealTabPFN-2.5 (default),0.2585034,0.31632653,0.25170067,0.5170068,,0.47619048,0.5578231,0.5782313,0.5510204,0.6394558,0.6530612,0.72789115,0.63265306,0.6938776,0.65986395,0.6258503,0.78911567,0.8503401,0.78911567,0.6802721,0.8095238,0.8639456,0.7414966,0.89115644,0.78911567,0.8639456,0.8639456,0.7619048,0.8503401,0.79591835,0.8707483,0.81632656,0.8503401,0.81632656,0.8027211,0.71428573,0.70068026,0.8979592,0.9047619,0.8979592,0.82312924,0.8707483,0.91156465,0.8435374,0.9319728,0.91156465,0.9183673,0.9047619,0.91156465,0.8979592,0.9455782,0.92517006,0.93877554,0.99319726,0.91156465,0.91156465,0.9183673,0.9319728,0.9591837,0.877551,0.9319728,0.9863946,0.9591837,0.99319726,1.0,0.99319726,0.9863946,0.9863946,0.97959185,0.9863946,0.9727891,0.9727891,0.99319726,0.99319726,0.9455782,1.0,1.0,1.0
+TabPFN-2.6 (default),0.1632653,0.3809524,0.37414965,0.40136054,0.52380955,,0.45578232,0.45578232,0.53061223,0.63265306,0.63265306,0.64625853,0.71428573,0.64625853,0.6258503,0.67346936,0.71428573,0.7482993,0.76870745,0.70068026,0.7482993,0.8027211,0.85714287,0.81632656,0.829932,0.877551,0.8979592,0.7823129,0.8503401,0.92517006,0.88435376,0.78911567,0.8367347,0.78911567,0.8435374,0.6530612,0.7755102,0.9183673,0.8639456,0.9455782,0.82312924,0.8503401,0.95238096,0.8367347,0.95238096,0.9047619,0.9455782,0.9319728,0.8639456,0.9183673,0.97959185,0.92517006,0.97959185,0.9659864,0.9319728,0.9319728,0.9455782,0.9319728,0.9455782,0.95238096,0.9455782,0.9659864,1.0,1.0,0.97959185,0.9863946,1.0,0.99319726,0.99319726,0.9659864,0.95238096,0.9659864,1.0,1.0,0.9591837,0.99319726,1.0,1.0
+TabPFN-3 (default),0.14965986,0.3809524,0.37414965,0.42857143,0.44217688,0.5442177,,0.5714286,0.585034,0.52380955,0.60544217,0.64625853,0.64625853,0.6666667,0.72789115,0.7210884,0.65986395,0.707483,0.6938776,0.7619048,0.6938776,0.82312924,0.8095238,0.82312924,0.82312924,0.7755102,0.7823129,0.9319728,0.8435374,0.8503401,0.8367347,0.8367347,0.7414966,0.70068026,0.9047619,0.5714286,0.81632656,0.877551,0.829932,0.91156465,0.877551,0.9047619,0.88435376,0.91156465,0.9455782,0.8503401,0.829932,0.9659864,0.8979592,0.89115644,0.97959185,0.9183673,0.93877554,0.9863946,0.92517006,0.9047619,0.92517006,0.97959185,0.9659864,0.9455782,0.9319728,0.99319726,1.0,1.0,0.9863946,0.9863946,0.99319726,1.0,0.97959185,0.9727891,0.9727891,0.9659864,0.99319726,1.0,0.9455782,1.0,1.0,1.0
+TabICLv2 (default),0.108843535,0.3469388,0.34013605,0.42857143,0.4217687,0.5442177,0.42857143,,0.5170068,0.5578231,0.585034,0.61904764,0.6938776,0.60544217,0.7346939,0.67346936,0.6666667,0.70068026,0.71428573,0.7346939,0.67346936,0.7755102,0.7414966,0.7823129,0.8095238,0.75510204,0.76870745,0.9591837,0.8435374,0.8707483,0.85714287,0.7823129,0.7346939,0.7346939,0.81632656,0.5102041,0.7482993,0.82312924,0.829932,0.877551,0.8707483,0.81632656,0.88435376,0.8367347,0.9183673,0.85714287,0.81632656,0.95238096,0.8707483,0.877551,0.92517006,0.9047619,0.8979592,0.92517006,0.92517006,0.9183673,0.92517006,0.89115644,0.89115644,0.9319728,0.9319728,0.91156465,0.99319726,0.9727891,0.9863946,0.97959185,0.93877554,0.9319728,0.9319728,0.9727891,0.9659864,0.9659864,0.9863946,0.9455782,0.93877554,0.99319726,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.17006803,0.292517,0.24489796,0.44217688,0.4489796,0.46938777,0.414966,0.4829932,,0.5578231,0.60544217,0.5578231,0.59183675,0.65986395,0.61904764,0.60544217,0.7619048,0.68707484,0.8095238,0.65986395,0.76870745,0.81632656,0.7210884,0.8435374,0.7823129,0.829932,0.8435374,0.6938776,0.8639456,0.8095238,0.88435376,0.829932,0.8639456,0.81632656,0.78911567,0.61904764,0.6802721,0.877551,0.877551,0.8979592,0.8095238,0.8367347,0.9319728,0.8367347,0.8979592,0.89115644,0.877551,0.8639456,0.8435374,0.89115644,0.9183673,0.9183673,0.8979592,0.9183673,0.92517006,0.9319728,0.92517006,0.89115644,0.9047619,0.89115644,0.9319728,0.91156465,0.8979592,0.93877554,0.9863946,0.99319726,0.93877554,0.92517006,0.92517006,0.9727891,0.9659864,0.9659864,0.93877554,0.93877554,0.93877554,0.9319728,1.0,1.0
+RealMLP (tuned + ensembled),0.23129252,0.2585034,0.26530612,0.27891156,0.3605442,0.36734694,0.47619048,0.44217688,0.44217688,,0.5136054,0.5170068,0.5442177,0.585034,0.5510204,0.56462586,0.7346939,0.585034,0.707483,0.5578231,0.7414966,0.7823129,0.72789115,0.81632656,0.7823129,0.8979592,0.92517006,0.61904764,0.877551,0.8027211,0.877551,0.82312924,0.85714287,0.8707483,0.78911567,0.6530612,0.60544217,0.8027211,0.9183673,0.9047619,0.8639456,0.8503401,0.9319728,0.85714287,0.78911567,0.92517006,0.9047619,0.9455782,0.91156465,0.85714287,0.99319726,0.9319728,0.91156465,0.99319726,0.92517006,0.93877554,0.9455782,0.9319728,0.9727891,0.9183673,0.93877554,0.9863946,0.9727891,1.0,0.99319726,0.9863946,1.0,0.9727891,0.9659864,0.97959185,0.9727891,0.9659864,1.0,1.0,0.95238096,0.99319726,1.0,1.0
+RealMLP (tuned),0.23129252,0.23129252,0.25170067,0.26530612,0.3469388,0.36734694,0.39455783,0.414966,0.39455783,0.48639455,,0.5102041,0.47619048,0.5102041,0.5102041,0.5510204,0.65986395,0.5714286,0.6530612,0.5442177,0.70068026,0.7482993,0.7346939,0.75510204,0.75510204,0.829932,0.85714287,0.60544217,0.78911567,0.76870745,0.8027211,0.7619048,0.8435374,0.8027211,0.7823129,0.6394558,0.61904764,0.8503401,0.8979592,0.88435376,0.8367347,0.829932,0.89115644,0.82312924,0.78911567,0.93877554,0.8707483,0.9183673,0.85714287,0.8979592,0.9591837,0.9183673,0.9319728,0.97959185,0.9183673,0.9183673,0.93877554,0.9047619,0.9659864,0.92517006,0.9455782,0.97959185,0.9727891,0.97959185,0.9863946,0.99319726,0.99319726,0.9727891,0.9591837,0.9863946,0.9727891,0.9591837,1.0,1.0,0.9727891,0.9863946,1.0,1.0
+LimiX (default),0.15646258,0.26530612,0.2585034,0.21088435,0.27210885,0.3537415,0.3537415,0.3809524,0.44217688,0.4829932,0.48979592,,0.46258503,0.52380955,0.5170068,0.46258503,0.63265306,0.6122449,0.6394558,0.60544217,0.6394558,0.72789115,0.63265306,0.7414966,0.76870745,0.7619048,0.75510204,0.6666667,0.78911567,0.7482993,0.79591835,0.8367347,0.7482993,0.707483,0.88435376,0.585034,0.60544217,0.70068026,0.82312924,0.829932,0.9047619,0.8979592,0.877551,0.9047619,0.8095238,0.877551,0.76870745,0.93877554,0.8979592,0.81632656,0.9455782,0.8979592,0.88435376,0.97959185,0.93877554,0.91156465,0.93877554,0.9727891,0.9591837,0.9455782,0.9455782,0.9727891,1.0,1.0,0.9863946,0.99319726,0.9727891,0.97959185,0.9659864,0.97959185,0.9727891,0.9659864,1.0,0.99319726,0.95238096,1.0,1.0,1.0
+Mitra (default),0.08163265,0.292517,0.292517,0.2857143,0.36734694,0.2857143,0.3537415,0.30612245,0.40816328,0.45578232,0.52380955,0.53741497,,0.56462586,0.56462586,0.5510204,0.61904764,0.622449,0.68707484,0.5986394,0.64625853,0.7210884,0.76870745,0.71428573,0.76870745,0.72789115,0.7619048,0.670068,0.8503401,0.85714287,0.8367347,0.75510204,0.7414966,0.71428573,0.7755102,0.5102041,0.6802721,0.8367347,0.81632656,0.877551,0.81632656,0.8095238,0.9183673,0.8095238,0.877551,0.85714287,0.82312924,0.877551,0.8435374,0.877551,0.9319728,0.9047619,0.88435376,0.877551,0.93877554,0.92517006,0.9183673,0.829932,0.877551,0.93877554,0.93877554,0.9047619,0.96258503,0.93877554,0.97959185,0.9659864,0.93877554,0.92517006,0.92517006,0.95238096,0.95238096,0.9591837,0.9319728,0.93877554,0.93877554,0.9319728,1.0,1.0
+TabPFNv2 (tuned),0.088435374,0.19727892,0.20408164,0.30612245,0.30612245,0.3537415,0.33333334,0.39455783,0.34013605,0.414966,0.48979592,0.47619048,0.43537414,,0.585034,0.5578231,0.60544217,0.60544217,0.6938776,0.6394558,0.6530612,0.6802721,0.6394558,0.75510204,0.707483,0.6802721,0.70068026,0.6530612,0.7482993,0.71428573,0.7619048,0.707483,0.7619048,0.71428573,0.7210884,0.5442177,0.65986395,0.829932,0.81632656,0.829932,0.7346939,0.8095238,0.82312924,0.76870745,0.8435374,0.829932,0.8095238,0.8027211,0.82312924,0.8707483,0.85714287,0.8707483,0.8639456,0.91156465,0.85714287,0.8639456,0.8639456,0.8503401,0.8979592,0.877551,0.9047619,0.89115644,0.89115644,0.92517006,0.9659864,0.9659864,0.93877554,0.9047619,0.89115644,0.9659864,0.9591837,0.9727891,0.9183673,0.93877554,0.93877554,0.9319728,1.0,1.0
+TabDPT (tuned + ensembled),0.10204082,0.27210885,0.27891156,0.18367347,0.34013605,0.37414965,0.27210885,0.26530612,0.3809524,0.4489796,0.48979592,0.4829932,0.43537414,0.414966,,0.47619048,0.5442177,0.5442177,0.5578231,0.65986395,0.5510204,0.5782313,0.6530612,0.6122449,0.6802721,0.63265306,0.6530612,0.67346936,0.71428573,0.75510204,0.6938776,0.63265306,0.63265306,0.63265306,0.7346939,0.48979592,0.7482993,0.6802721,0.7346939,0.78911567,0.79591835,0.7414966,0.78911567,0.78911567,0.8639456,0.76870745,0.76870745,0.8503401,0.8367347,0.8095238,0.8979592,0.79591835,0.81632656,0.85714287,0.85714287,0.89115644,0.9183673,0.8503401,0.89115644,0.9659864,0.91156465,0.9591837,0.97959185,1.0,0.85714287,0.8707483,0.93877554,0.9047619,0.9047619,0.9591837,0.9591837,0.95238096,0.93877554,1.0,0.9727891,0.9319728,0.95238096,1.0
+TabDPT-Turbo (default),0.11564626,0.30612245,0.31292516,0.26530612,0.37414965,0.3265306,0.27891156,0.3265306,0.39455783,0.43537414,0.4489796,0.53741497,0.4489796,0.44217688,0.52380955,,0.48979592,0.5714286,0.5442177,0.5782313,0.5102041,0.60544217,0.65986395,0.59183675,0.6530612,0.6122449,0.6122449,0.6530612,0.67346936,0.79591835,0.61904764,0.60544217,0.5510204,0.5578231,0.75510204,0.49659863,0.78911567,0.70068026,0.6666667,0.8027211,0.7823129,0.67346936,0.7414966,0.76870745,0.82312924,0.72789115,0.7755102,0.829932,0.78911567,0.76870745,0.8639456,0.76870745,0.8639456,0.8435374,0.8503401,0.88435376,0.88435376,0.7755102,0.8435374,0.9183673,0.88435376,0.93877554,1.0,1.0,0.8639456,0.8503401,0.93877554,0.877551,0.877551,0.93877554,0.93877554,0.9455782,0.93877554,1.0,0.9455782,0.93877554,0.9659864,1.0
+TabM (tuned + ensembled),0.1904762,0.15646258,0.20408164,0.23129252,0.21088435,0.2857143,0.34013605,0.33333334,0.23809524,0.26530612,0.34013605,0.36734694,0.3809524,0.39455783,0.45578232,0.5102041,,0.36734694,0.52380955,0.48979592,0.6360544,0.53061223,0.6122449,0.5986394,0.6122449,0.68707484,0.71428573,0.5442177,0.65986395,0.6258503,0.6666667,0.61904764,0.7619048,0.877551,0.5782313,0.6394558,0.5986394,0.7755102,0.8639456,0.7414966,0.6394558,0.70068026,0.829932,0.63265306,0.68707484,0.89115644,0.81632656,0.8095238,0.75510204,0.8707483,0.88435376,0.89115644,0.877551,0.95238096,0.88435376,0.85714287,0.88435376,0.8027211,0.9455782,0.88435376,0.95238096,0.93877554,0.89115644,0.91156465,0.9591837,0.97959185,1.0,0.93877554,0.95238096,0.95238096,0.9659864,0.95238096,0.97959185,1.0,0.9659864,0.99319726,1.0,1.0
+TabPFNv2 (default),0.088435374,0.12244898,0.1292517,0.17687075,0.14965986,0.25170067,0.292517,0.2993197,0.31292516,0.414966,0.42857143,0.3877551,0.37755102,0.39455783,0.45578232,0.42857143,0.63265306,,0.5510204,0.5102041,0.68707484,0.63265306,0.5170068,0.70068026,0.61904764,0.67346936,0.68707484,0.58843535,0.6802721,0.5986394,0.67346936,0.6802721,0.7210884,0.70068026,0.707483,0.50340134,0.5510204,0.707483,0.78911567,0.6938776,0.71428573,0.7823129,0.8027211,0.7346939,0.7823129,0.81632656,0.6938776,0.7755102,0.8027211,0.76870745,0.829932,0.8503401,0.8095238,0.8503401,0.8639456,0.8367347,0.8435374,0.829932,0.8639456,0.85714287,0.88435376,0.8979592,0.88095236,0.85714287,0.93877554,0.92517006,0.89115644,0.91156465,0.9047619,0.97959185,0.9659864,0.9659864,0.9047619,0.91156465,0.9319728,0.92517006,0.99319726,1.0
+"AutoGluon 1.4 (best, 4h)",0.07482993,0.11564626,0.14285715,0.20408164,0.21088435,0.23129252,0.30612245,0.2857143,0.1904762,0.292517,0.3469388,0.3605442,0.31292516,0.30612245,0.44217688,0.45578232,0.47619048,0.4489796,,0.47619048,0.5170068,0.5782313,0.59183675,0.6530612,0.6258503,0.59183675,0.6258503,0.56462586,0.7210884,0.6666667,0.707483,0.65986395,0.64625853,0.6666667,0.59183675,0.5102041,0.5986394,0.7414966,0.7755102,0.7414966,0.6530612,0.72789115,0.81632656,0.6802721,0.707483,0.8027211,0.82312924,0.81632656,0.75510204,0.829932,0.8367347,0.8639456,0.8979592,0.8707483,0.85714287,0.877551,0.88435376,0.78911567,0.91156465,0.82312924,0.92517006,0.9047619,0.8979592,0.92517006,0.9455782,0.9591837,0.97959185,0.93877554,0.93877554,0.93877554,0.9455782,0.95238096,0.97959185,0.97959185,0.95238096,0.9659864,0.99319726,1.0
+TabDPT (tuned),0.11564626,0.2244898,0.2585034,0.14285715,0.3197279,0.2993197,0.23809524,0.26530612,0.34013605,0.44217688,0.45578232,0.39455783,0.40136054,0.3605442,0.34013605,0.4217687,0.5102041,0.48979592,0.52380955,,0.5442177,0.50340134,0.585034,0.5442177,0.63265306,0.5986394,0.6258503,0.60544217,0.6394558,0.64625853,0.6530612,0.5578231,0.5782313,0.61904764,0.67346936,0.47619048,0.7346939,0.63265306,0.6258503,0.6802721,0.70068026,0.67346936,0.70068026,0.7210884,0.82312924,0.6938776,0.6938776,0.76870745,0.7482993,0.72789115,0.8027211,0.72789115,0.7755102,0.7482993,0.8435374,0.85714287,0.877551,0.81632656,0.85714287,0.9591837,0.9183673,0.9591837,0.9591837,0.97959185,0.8095238,0.8027211,0.9319728,0.9047619,0.9047619,0.9319728,0.9455782,0.95238096,0.93877554,1.0,0.9591837,0.92517006,0.93877554,1.0
+TabM (tuned),0.17687075,0.14965986,0.20408164,0.23809524,0.1904762,0.25170067,0.30612245,0.3265306,0.23129252,0.2585034,0.2993197,0.3605442,0.3537415,0.3469388,0.4489796,0.48979592,0.36394557,0.31292516,0.4829932,0.45578232,,0.46258503,0.56462586,0.52380955,0.5510204,0.5986394,0.6122449,0.53061223,0.5782313,0.63265306,0.5782313,0.5986394,0.67346936,0.7414966,0.5714286,0.5714286,0.60544217,0.7482993,0.8027211,0.6802721,0.64625853,0.6938776,0.7619048,0.6122449,0.6666667,0.85714287,0.7823129,0.7755102,0.7210884,0.8367347,0.8435374,0.8707483,0.8367347,0.91156465,0.78911567,0.79591835,0.81632656,0.7755102,0.9319728,0.8435374,0.91156465,0.92517006,0.88435376,0.8503401,0.9727891,0.9727891,1.0,0.92517006,0.95238096,0.9591837,0.95238096,0.9455782,0.9727891,1.0,0.9659864,0.9863946,1.0,1.0
+CatBoost (tuned + ensembled),0.0952381,0.07482993,0.06122449,0.11564626,0.13605443,0.19727892,0.17687075,0.2244898,0.18367347,0.21768707,0.25170067,0.27210885,0.27891156,0.3197279,0.4217687,0.39455783,0.46938777,0.36734694,0.4217687,0.49659863,0.53741497,,0.48979592,0.6972789,0.52380955,0.61904764,0.6394558,0.5578231,0.63265306,0.5170068,0.63265306,0.5782313,0.5986394,0.64625853,0.63265306,0.5102041,0.5714286,0.6394558,0.7414966,0.67346936,0.6938776,0.7619048,0.82312924,0.7210884,0.72789115,0.829932,0.68707484,0.82312924,0.82312924,0.78911567,0.9455782,0.8707483,0.82312924,0.9455782,0.88435376,0.8707483,0.91156465,0.8979592,0.9591837,0.8639456,0.9183673,0.9727891,0.9183673,0.9455782,0.9591837,0.9659864,1.0,0.97959185,0.9727891,0.9727891,0.9727891,0.9659864,0.97959185,1.0,0.95238096,0.99319726,1.0,1.0
+ModernNCA (tuned + ensembled),0.08163265,0.14965986,0.14285715,0.13605443,0.2585034,0.14285715,0.1904762,0.2585034,0.27891156,0.27210885,0.26530612,0.36734694,0.23129252,0.3605442,0.3469388,0.34013605,0.3877551,0.4829932,0.40816328,0.414966,0.43537414,0.5102041,,0.5442177,0.622449,0.5714286,0.585034,0.52380955,0.63265306,0.53061223,0.64625853,0.6394558,0.5442177,0.5442177,0.72789115,0.53061223,0.5510204,0.59183675,0.6938776,0.67346936,0.8027211,0.79591835,0.71428573,0.8027211,0.6530612,0.7619048,0.6666667,0.9047619,0.81632656,0.7619048,0.8707483,0.79591835,0.8027211,0.88435376,0.8027211,0.877551,0.8639456,0.89115644,0.92517006,0.91156465,0.88435376,0.95238096,0.9455782,0.9863946,0.91156465,0.9591837,0.97959185,0.91156465,0.9047619,0.97959185,0.9727891,0.95238096,0.9455782,0.99319726,0.9319728,0.9659864,0.99319726,1.0
+CatBoost (tuned),0.07482993,0.07482993,0.07482993,0.10204082,0.108843535,0.18367347,0.17687075,0.21768707,0.15646258,0.18367347,0.24489796,0.2585034,0.2857143,0.24489796,0.3877551,0.40816328,0.40136054,0.2993197,0.3469388,0.45578232,0.47619048,0.30272108,0.45578232,,0.5102041,0.5782313,0.5782313,0.52380955,0.5782313,0.48979592,0.5578231,0.53061223,0.5578231,0.63265306,0.6122449,0.48979592,0.5578231,0.6530612,0.72789115,0.64625853,0.65986395,0.7482993,0.76870745,0.68707484,0.6802721,0.7755102,0.67346936,0.78911567,0.79591835,0.76870745,0.88435376,0.8435374,0.7823129,0.9047619,0.8435374,0.8435374,0.8435374,0.8707483,0.9319728,0.8435374,0.9183673,0.9727891,0.91156465,0.92517006,0.95238096,0.9659864,1.0,0.9659864,0.9591837,0.9727891,0.9727891,0.9659864,0.9727891,1.0,0.95238096,0.99319726,1.0,1.0
+ModernNCA (tuned),0.108843535,0.13605443,0.12244898,0.11564626,0.21088435,0.17006803,0.17687075,0.1904762,0.21768707,0.21768707,0.24489796,0.23129252,0.23129252,0.292517,0.3197279,0.3469388,0.3877551,0.3809524,0.37414965,0.36734694,0.4489796,0.47619048,0.37755102,0.48979592,,0.5102041,0.53061223,0.4489796,0.5782313,0.45578232,0.5782313,0.585034,0.50340134,0.5442177,0.64625853,0.53741497,0.5442177,0.5714286,0.67346936,0.5986394,0.71428573,0.7210884,0.6938776,0.71428573,0.6802721,0.7414966,0.61904764,0.8095238,0.76870745,0.75510204,0.8027211,0.7619048,0.7346939,0.8095238,0.79591835,0.82312924,0.829932,0.85714287,0.91156465,0.9183673,0.91156465,0.9591837,0.9183673,0.9659864,0.88435376,0.9319728,0.9727891,0.92517006,0.9319728,0.9659864,0.9659864,0.9455782,0.93877554,0.9863946,0.9727891,0.9659864,0.99319726,1.0
+XGBoost (tuned),0.12244898,0.0952381,0.10204082,0.17006803,0.13605443,0.12244898,0.2244898,0.24489796,0.17006803,0.10204082,0.17006803,0.23809524,0.27210885,0.3197279,0.36734694,0.3877551,0.31292516,0.3265306,0.40816328,0.40136054,0.40136054,0.3809524,0.42857143,0.4217687,0.48979592,,0.5544218,0.49659863,0.5170068,0.47619048,0.5782313,0.5442177,0.5510204,0.5510204,0.5170068,0.5578231,0.5170068,0.64625853,0.7346939,0.61904764,0.59183675,0.6802721,0.7482993,0.61904764,0.5714286,0.7823129,0.6666667,0.7823129,0.7346939,0.72789115,0.93877554,0.8435374,0.7346939,0.8435374,0.877551,0.88435376,0.88435376,0.7823129,0.9183673,0.8435374,0.92517006,0.9591837,0.91156465,0.92517006,0.9455782,0.9591837,0.99319726,0.95238096,0.9455782,0.9591837,0.9591837,0.9659864,0.9863946,0.99319726,0.9319728,0.9863946,1.0,1.0
+XGBoost (tuned + ensembled),0.11564626,0.0952381,0.08163265,0.14285715,0.13605443,0.10204082,0.21768707,0.23129252,0.15646258,0.07482993,0.14285715,0.24489796,0.23809524,0.2993197,0.3469388,0.3877551,0.2857143,0.31292516,0.37414965,0.37414965,0.3877551,0.3605442,0.414966,0.4217687,0.46938777,0.44557822,,0.50340134,0.49659863,0.48979592,0.53741497,0.53741497,0.5510204,0.53741497,0.49659863,0.5442177,0.5102041,0.64625853,0.7619048,0.6258503,0.5986394,0.6802721,0.76870745,0.6122449,0.585034,0.8095238,0.67346936,0.7823129,0.707483,0.7346939,0.9319728,0.8435374,0.7414966,0.8435374,0.8639456,0.877551,0.89115644,0.7823129,0.9183673,0.8435374,0.92517006,0.95238096,0.9047619,0.91156465,0.9455782,0.9591837,0.99319726,0.9591837,0.95238096,0.9591837,0.95238096,0.9591837,0.9863946,0.99319726,0.9319728,0.9863946,1.0,1.0
+TabICL (default),0.027210884,0.21088435,0.21088435,0.10204082,0.23809524,0.21768707,0.06802721,0.040816326,0.30612245,0.3809524,0.39455783,0.33333334,0.32993197,0.3469388,0.3265306,0.3469388,0.45578232,0.41156462,0.43537414,0.39455783,0.46938777,0.44217688,0.47619048,0.47619048,0.5510204,0.50340134,0.49659863,,0.59183675,0.59183675,0.585034,0.49659863,0.5102041,0.53741497,0.6802721,0.40816328,0.60544217,0.47619048,0.5510204,0.64625853,0.7482993,0.63265306,0.64625853,0.7482993,0.8027211,0.61904764,0.6122449,0.70068026,0.7823129,0.67346936,0.7619048,0.707483,0.68707484,0.6802721,0.8503401,0.8095238,0.82312924,0.7414966,0.7755102,0.9319728,0.8639456,0.8979592,0.9421769,0.9319728,0.8095238,0.82312924,0.89115644,0.8503401,0.8503401,0.9455782,0.93877554,0.9455782,0.877551,0.93877554,0.93877554,0.88435376,1.0,1.0
+LightGBM (tuned + ensembled),0.0952381,0.11564626,0.108843535,0.0952381,0.14965986,0.14965986,0.15646258,0.15646258,0.13605443,0.12244898,0.21088435,0.21088435,0.14965986,0.25170067,0.2857143,0.3265306,0.34013605,0.3197279,0.27891156,0.3605442,0.4217687,0.36734694,0.36734694,0.4217687,0.4217687,0.4829932,0.50340134,0.40816328,,0.48979592,0.47959185,0.52380955,0.5102041,0.59183675,0.49659863,0.53061223,0.5578231,0.60544217,0.71428573,0.59183675,0.585034,0.68707484,0.72789115,0.56462586,0.61904764,0.7755102,0.64625853,0.75510204,0.71428573,0.72789115,0.8435374,0.8095238,0.7823129,0.829932,0.93877554,0.9455782,0.9863946,0.75510204,0.88435376,0.8503401,0.97959185,0.9183673,0.8979592,0.9455782,0.88435376,0.92517006,1.0,0.9455782,0.9455782,0.93877554,0.95238096,0.93877554,1.0,1.0,0.95238096,0.99319726,1.0,1.0
+iLTM (tuned + ensembled),0.08163265,0.12244898,0.11564626,0.088435374,0.20408164,0.07482993,0.14965986,0.1292517,0.1904762,0.19727892,0.23129252,0.25170067,0.14285715,0.2857143,0.24489796,0.20408164,0.37414965,0.40136054,0.33333334,0.3537415,0.36734694,0.4829932,0.46938777,0.5102041,0.5442177,0.52380955,0.5102041,0.40816328,0.5102041,,0.53061223,0.56462586,0.47619048,0.48979592,0.61904764,0.5102041,0.53061223,0.52380955,0.64625853,0.67346936,0.67346936,0.6802721,0.6666667,0.70068026,0.61904764,0.707483,0.65986395,0.79591835,0.7619048,0.6802721,0.829932,0.76870745,0.7755102,0.81632656,0.829932,0.89115644,0.9047619,0.7755102,0.85714287,0.8503401,0.9319728,0.8639456,0.9183673,0.99319726,0.81632656,0.88435376,0.99319726,0.8639456,0.85714287,0.89115644,0.91156465,0.9183673,0.99319726,1.0,0.8979592,0.9727891,0.99319726,1.0
+LightGBM (tuned),0.10204082,0.088435374,0.10204082,0.108843535,0.1292517,0.11564626,0.1632653,0.14285715,0.11564626,0.12244898,0.19727892,0.20408164,0.1632653,0.23809524,0.30612245,0.3809524,0.33333334,0.3265306,0.292517,0.3469388,0.4217687,0.36734694,0.3537415,0.44217688,0.4217687,0.4217687,0.46258503,0.414966,0.52040815,0.46938777,,0.50340134,0.48979592,0.5782313,0.47619048,0.5170068,0.56462586,0.63265306,0.6938776,0.5782313,0.5442177,0.65986395,0.7414966,0.5782313,0.6122449,0.7823129,0.63265306,0.7346939,0.6938776,0.72789115,0.8503401,0.81632656,0.7619048,0.81632656,0.9183673,0.9319728,0.9455782,0.76870745,0.8639456,0.85714287,0.9659864,0.9319728,0.88435376,0.9455782,0.89115644,0.92517006,1.0,0.95238096,0.9591837,0.93877554,0.95238096,0.93877554,1.0,1.0,0.95238096,0.99319726,1.0,1.0
+xRFM (tuned + ensembled),0.08163265,0.108843535,0.12244898,0.0952381,0.18367347,0.21088435,0.1632653,0.21768707,0.17006803,0.17687075,0.23809524,0.1632653,0.24489796,0.292517,0.36734694,0.39455783,0.3809524,0.3197279,0.34013605,0.44217688,0.40136054,0.4217687,0.3605442,0.46938777,0.414966,0.45578232,0.46258503,0.50340134,0.47619048,0.43537414,0.49659863,,0.47619048,0.4829932,0.52380955,0.49659863,0.5714286,0.59183675,0.5102041,0.63265306,0.5986394,0.67346936,0.65986395,0.6394558,0.72789115,0.64625853,0.56462586,0.67346936,0.70068026,0.6938776,0.7482993,0.75510204,0.65986395,0.707483,0.707483,0.8639456,0.88435376,0.9591837,0.8503401,0.91156465,0.93877554,0.9727891,0.9659864,0.99319726,0.9591837,0.9659864,0.8979592,0.9659864,0.9659864,0.91156465,0.89115644,0.89115644,0.9659864,0.99319726,0.9863946,0.9863946,1.0,1.0
+BetaTabPFN (default),0.088435374,0.10204082,0.088435374,0.17687075,0.14965986,0.1632653,0.2585034,0.26530612,0.13605443,0.14285715,0.15646258,0.25170067,0.2585034,0.23809524,0.36734694,0.4489796,0.23809524,0.27891156,0.3537415,0.4217687,0.3265306,0.40136054,0.45578232,0.44217688,0.49659863,0.4489796,0.4489796,0.48979592,0.48979592,0.52380955,0.5102041,0.52380955,,0.5102041,0.50340134,0.44217688,0.5714286,0.71428573,0.70068026,0.59183675,0.56462586,0.61904764,0.67346936,0.5714286,0.5714286,0.8095238,0.71428573,0.71428573,0.61904764,0.7823129,0.7414966,0.82312924,0.7755102,0.85714287,0.72789115,0.7482993,0.7755102,0.71428573,0.93877554,0.79591835,0.8979592,0.877551,0.82312924,0.81632656,0.93877554,0.93877554,0.99319726,0.95238096,0.95238096,0.9319728,0.9591837,0.93877554,0.99319726,1.0,0.9727891,0.9863946,1.0,1.0
+TabM (default),0.14965986,0.14285715,0.15646258,0.19727892,0.18367347,0.21088435,0.2993197,0.26530612,0.18367347,0.1292517,0.19727892,0.292517,0.2857143,0.2857143,0.36734694,0.44217688,0.12244898,0.2993197,0.33333334,0.3809524,0.2585034,0.3537415,0.45578232,0.36734694,0.45578232,0.4489796,0.46258503,0.46258503,0.40816328,0.5102041,0.4217687,0.5170068,0.48979592,,0.46258503,0.46258503,0.5510204,0.64625853,0.64625853,0.56462586,0.52380955,0.5782313,0.61904764,0.52380955,0.49659863,0.7210884,0.7346939,0.65986395,0.5782313,0.7755102,0.707483,0.7619048,0.8027211,0.8435374,0.64625853,0.7346939,0.7346939,0.68707484,0.91156465,0.7619048,0.81632656,0.8367347,0.8095238,0.8027211,0.91156465,0.9319728,0.9863946,0.92517006,0.9319728,0.93877554,0.9319728,0.92517006,0.97959185,1.0,0.9591837,0.97959185,1.0,1.0
+ExtraTrees (tuned + ensembled),0.07482993,0.13605443,0.15646258,0.10204082,0.19727892,0.15646258,0.0952381,0.18367347,0.21088435,0.21088435,0.21768707,0.11564626,0.2244898,0.27891156,0.26530612,0.24489796,0.4217687,0.292517,0.40816328,0.3265306,0.42857143,0.36734694,0.27210885,0.3877551,0.3537415,0.4829932,0.50340134,0.3197279,0.50340134,0.3809524,0.52380955,0.47619048,0.49659863,0.53741497,,0.47619048,0.5170068,0.4489796,0.53061223,0.53061223,0.7176871,0.5578231,0.6122449,0.76870745,0.67346936,0.60544217,0.53061223,0.6122449,0.8503401,0.59183675,0.70068026,0.70068026,0.6122449,0.6666667,0.78911567,0.8095238,0.81632656,0.8095238,0.7755102,0.95238096,0.8707483,0.9455782,0.9659864,1.0,0.7619048,0.7619048,0.91156465,0.91156465,0.91156465,0.92517006,0.95238096,0.93877554,0.8979592,0.97959185,1.0,0.9319728,0.9455782,1.0
+SAP-RPT-OSS (default),0.24489796,0.25170067,0.2585034,0.37414965,0.2857143,0.3469388,0.42857143,0.48979592,0.3809524,0.3469388,0.3605442,0.414966,0.48979592,0.45578232,0.5102041,0.50340134,0.3605442,0.49659863,0.48979592,0.52380955,0.42857143,0.48979592,0.46938777,0.5102041,0.46258503,0.44217688,0.45578232,0.59183675,0.46938777,0.48979592,0.4829932,0.50340134,0.5578231,0.53741497,0.52380955,,0.52380955,0.68707484,0.5782313,0.53061223,0.53061223,0.5714286,0.5442177,0.52380955,0.5578231,0.60544217,0.6530612,0.5782313,0.5510204,0.70068026,0.53741497,0.63265306,0.7210884,0.7414966,0.59183675,0.53061223,0.5102041,0.585034,0.71428573,0.56462586,0.6394558,0.6530612,0.6530612,0.63265306,0.8027211,0.8027211,0.7210884,0.7346939,0.7210884,0.75510204,0.7482993,0.7482993,0.78911567,0.7482993,0.7619048,0.9455782,0.8095238,1.0
+TabDPT (default),0.0952381,0.20408164,0.2244898,0.1632653,0.2993197,0.2244898,0.18367347,0.25170067,0.3197279,0.39455783,0.3809524,0.39455783,0.3197279,0.34013605,0.25170067,0.21088435,0.40136054,0.4489796,0.40136054,0.26530612,0.39455783,0.42857143,0.4489796,0.44217688,0.45578232,0.4829932,0.48979592,0.39455783,0.44217688,0.46938777,0.43537414,0.42857143,0.42857143,0.4489796,0.4829932,0.47619048,,0.45578232,0.46938777,0.52380955,0.5170068,0.49659863,0.45578232,0.5714286,0.5986394,0.49659863,0.5578231,0.5782313,0.6258503,0.52380955,0.5578231,0.50340134,0.6122449,0.5714286,0.7346939,0.65986395,0.63265306,0.6938776,0.63265306,0.8503401,0.82312924,0.8095238,0.8027211,0.9183673,0.5782313,0.64625853,0.9183673,0.7619048,0.7482993,0.8503401,0.8707483,0.88435376,0.9047619,1.0,0.877551,0.89115644,0.9319728,1.0
+TorchMLP (tuned + ensembled),0.07482993,0.054421768,0.06122449,0.0952381,0.10204082,0.08163265,0.12244898,0.17687075,0.12244898,0.19727892,0.14965986,0.2993197,0.1632653,0.17006803,0.3197279,0.2993197,0.2244898,0.292517,0.2585034,0.36734694,0.25170067,0.3605442,0.40816328,0.3469388,0.42857143,0.3537415,0.3537415,0.52380955,0.39455783,0.47619048,0.36734694,0.40816328,0.2857143,0.3537415,0.5510204,0.31292516,0.5442177,,0.3809524,0.5782313,0.5782313,0.5170068,0.5170068,0.585034,0.5170068,0.4829932,0.6938776,0.6394558,0.59183675,0.81292516,0.61904764,0.5714286,0.79591835,0.6938776,0.6394558,0.6530612,0.6530612,0.6530612,0.89115644,0.72789115,0.68707484,0.7619048,0.7414966,0.78911567,0.8707483,0.9183673,0.91156465,0.8979592,0.9047619,0.8367347,0.877551,0.8707483,0.9591837,0.99319726,0.93877554,0.9727891,0.99319726,1.0
+EBM (tuned + ensembled),0.0952381,0.04761905,0.06802721,0.08163265,0.0952381,0.13605443,0.17006803,0.17006803,0.12244898,0.08163265,0.10204082,0.17687075,0.18367347,0.18367347,0.26530612,0.33333334,0.13605443,0.21088435,0.2244898,0.37414965,0.19727892,0.2585034,0.30612245,0.27210885,0.3265306,0.26530612,0.23809524,0.4489796,0.2857143,0.3537415,0.30612245,0.48979592,0.2993197,0.3537415,0.46938777,0.4217687,0.53061223,0.61904764,,0.5102041,0.5102041,0.5986394,0.47619048,0.5102041,0.5170068,0.7585034,0.5578231,0.61904764,0.5510204,0.68707484,0.5986394,0.8435374,0.67346936,0.7346939,0.59183675,0.6938776,0.707483,0.7346939,0.9183673,0.7482993,0.829932,0.8707483,0.78911567,0.8027211,0.93877554,0.9727891,0.9455782,0.9319728,0.93877554,0.9183673,0.9319728,0.9319728,0.97959185,0.99319726,0.93877554,0.9863946,0.99319726,1.0
+iLTM (tuned),0.06802721,0.04761905,0.04761905,0.06122449,0.10204082,0.054421768,0.088435374,0.12244898,0.10204082,0.0952381,0.11564626,0.17006803,0.12244898,0.17006803,0.21088435,0.19727892,0.2585034,0.30612245,0.2585034,0.3197279,0.3197279,0.3265306,0.3265306,0.3537415,0.40136054,0.3809524,0.37414965,0.3537415,0.40816328,0.3265306,0.4217687,0.36734694,0.40816328,0.43537414,0.46938777,0.46938777,0.47619048,0.4217687,0.48979592,,0.5714286,0.53061223,0.5442177,0.59183675,0.5170068,0.59183675,0.6258503,0.64625853,0.71428573,0.6122449,0.67346936,0.6666667,0.67346936,0.72789115,0.707483,0.7755102,0.7346939,0.72789115,0.8503401,0.7619048,0.8367347,0.8639456,0.8639456,0.92517006,0.76870745,0.829932,0.97959185,0.8435374,0.8435374,0.89115644,0.91156465,0.9047619,0.93877554,0.9727891,0.88435376,0.9455782,0.9727891,1.0
+RandomForest (tuned + ensembled),0.0952381,0.12244898,0.1292517,0.0952381,0.17687075,0.17687075,0.12244898,0.1292517,0.1904762,0.13605443,0.1632653,0.0952381,0.18367347,0.26530612,0.20408164,0.21768707,0.3605442,0.2857143,0.3469388,0.2993197,0.3537415,0.30612245,0.19727892,0.34013605,0.2857143,0.40816328,0.40136054,0.25170067,0.414966,0.3265306,0.45578232,0.40136054,0.43537414,0.47619048,0.28231293,0.46938777,0.4829932,0.4217687,0.48979592,0.42857143,,0.52380955,0.5170068,0.5510204,0.5986394,0.5442177,0.47619048,0.53061223,0.79931974,0.53061223,0.63265306,0.64625853,0.5170068,0.59183675,0.72789115,0.76870745,0.7619048,0.7482993,0.72789115,0.9047619,0.8503401,0.8979592,0.9863946,0.95238096,0.707483,0.71428573,0.8435374,0.92517006,0.92517006,0.91156465,0.9319728,0.9319728,0.88435376,0.9455782,0.99319726,0.93877554,0.9183673,1.0
+xRFM (tuned),0.07482993,0.08163265,0.08163265,0.06122449,0.1292517,0.14965986,0.0952381,0.18367347,0.1632653,0.14965986,0.17006803,0.10204082,0.1904762,0.1904762,0.2585034,0.3265306,0.2993197,0.21768707,0.27210885,0.3265306,0.30612245,0.23809524,0.20408164,0.25170067,0.27891156,0.3197279,0.3197279,0.36734694,0.31292516,0.3197279,0.34013605,0.3265306,0.3809524,0.4217687,0.44217688,0.42857143,0.50340134,0.4829932,0.40136054,0.46938777,0.47619048,,0.44217688,0.53741497,0.6258503,0.5102041,0.49659863,0.53061223,0.585034,0.59183675,0.56462586,0.61904764,0.53741497,0.5578231,0.63265306,0.7414966,0.71428573,0.9183673,0.79591835,0.8707483,0.88435376,0.9455782,0.9183673,0.9455782,0.8707483,0.877551,0.85714287,0.92517006,0.9319728,0.89115644,0.9183673,0.9047619,0.92517006,0.99319726,0.9727891,0.97959185,0.99319726,1.0
+CatBoost (default),0.06802721,0.034013607,0.06122449,0.054421768,0.088435374,0.04761905,0.11564626,0.11564626,0.06802721,0.06802721,0.108843535,0.12244898,0.08163265,0.17687075,0.21088435,0.2585034,0.17006803,0.19727892,0.18367347,0.2993197,0.23809524,0.17687075,0.2857143,0.23129252,0.30612245,0.25170067,0.23129252,0.3537415,0.27210885,0.33333334,0.2585034,0.34013605,0.3265306,0.3809524,0.3877551,0.45578232,0.5442177,0.4829932,0.52380955,0.45578232,0.4829932,0.5578231,,0.48979592,0.48979592,0.6666667,0.56462586,0.60544217,0.60544217,0.63265306,0.67346936,0.70068026,0.6666667,0.707483,0.72789115,0.78911567,0.8027211,0.70068026,0.85714287,0.79591835,0.93877554,0.89115644,0.8435374,0.88435376,0.8367347,0.8639456,1.0,0.9183673,0.91156465,0.9183673,0.9455782,0.9183673,0.9659864,1.0,0.9455782,0.99319726,1.0,1.0
+ExtraTrees (tuned),0.08163265,0.108843535,0.1292517,0.10204082,0.15646258,0.1632653,0.088435374,0.1632653,0.1632653,0.14285715,0.17687075,0.0952381,0.1904762,0.23129252,0.21088435,0.23129252,0.36734694,0.26530612,0.3197279,0.27891156,0.3877551,0.27891156,0.19727892,0.31292516,0.2857143,0.3809524,0.3877551,0.25170067,0.43537414,0.2993197,0.4217687,0.3605442,0.42857143,0.47619048,0.23129252,0.47619048,0.42857143,0.414966,0.48979592,0.40816328,0.4489796,0.46258503,0.5102041,,0.6122449,0.56462586,0.48979592,0.4829932,0.6360544,0.5170068,0.5510204,0.60544217,0.53741497,0.56462586,0.7346939,0.6802721,0.68707484,0.71428573,0.70068026,0.9319728,0.8503401,0.91156465,0.89115644,0.88435376,0.707483,0.68707484,0.8503401,0.91156465,0.9047619,0.91156465,0.91156465,0.9183673,0.88435376,0.9659864,1.0,0.9319728,0.92517006,1.0
+TabSwift (default),0.08163265,0.027210884,0.027210884,0.013605442,0.06802721,0.04761905,0.054421768,0.08163265,0.10204082,0.21088435,0.21088435,0.1904762,0.12244898,0.15646258,0.13605443,0.17687075,0.31292516,0.21768707,0.292517,0.17687075,0.33333334,0.27210885,0.3469388,0.3197279,0.3197279,0.42857143,0.414966,0.19727892,0.3809524,0.3809524,0.3877551,0.27210885,0.42857143,0.50340134,0.3265306,0.44217688,0.40136054,0.4829932,0.4829932,0.4829932,0.40136054,0.37414965,0.5102041,0.3877551,,0.5442177,0.6394558,0.52380955,0.56462586,0.5986394,0.5442177,0.5782313,0.65986395,0.65986395,0.7482993,0.6122449,0.60544217,0.5986394,0.6802721,0.8435374,0.7823129,0.92517006,0.85714287,0.7619048,0.65986395,0.6938776,0.8367347,0.8707483,0.85714287,0.9319728,0.9455782,0.9455782,0.8707483,0.95238096,0.95238096,0.9319728,0.9659864,1.0
+EBM (tuned),0.07482993,0.040816326,0.06122449,0.054421768,0.088435374,0.0952381,0.14965986,0.14285715,0.108843535,0.07482993,0.06122449,0.12244898,0.14285715,0.17006803,0.23129252,0.27210885,0.108843535,0.18367347,0.19727892,0.30612245,0.14285715,0.17006803,0.23809524,0.2244898,0.2585034,0.21768707,0.1904762,0.3809524,0.2244898,0.292517,0.21768707,0.3537415,0.1904762,0.27891156,0.39455783,0.39455783,0.50340134,0.5170068,0.2414966,0.40816328,0.45578232,0.48979592,0.33333334,0.43537414,0.45578232,,0.5102041,0.5170068,0.52380955,0.63265306,0.5102041,0.70068026,0.6258503,0.64625853,0.49659863,0.63265306,0.65986395,0.68707484,0.877551,0.6938776,0.7823129,0.81632656,0.75510204,0.7755102,0.89115644,0.92517006,0.9047619,0.9183673,0.92517006,0.8979592,0.91156465,0.9047619,0.9659864,0.99319726,0.93877554,0.97959185,0.99319726,1.0
+FastaiMLP (tuned + ensembled),0.10204082,0.07482993,0.06802721,0.1292517,0.08163265,0.054421768,0.17006803,0.18367347,0.12244898,0.0952381,0.1292517,0.23129252,0.17687075,0.1904762,0.23129252,0.2244898,0.18367347,0.30612245,0.17687075,0.30612245,0.21768707,0.31292516,0.33333334,0.3265306,0.3809524,0.33333334,0.3265306,0.3877551,0.3537415,0.34013605,0.36734694,0.43537414,0.2857143,0.26530612,0.46938777,0.3469388,0.44217688,0.30612245,0.44217688,0.37414965,0.52380955,0.50340134,0.43537414,0.5102041,0.3605442,0.48979592,,0.5986394,0.53741497,0.50340134,0.5578231,0.5102041,0.75510204,0.59183675,0.585034,0.68707484,0.67346936,0.5578231,0.67346936,0.63265306,0.72789115,0.63265306,0.64625853,0.7823129,0.8095238,0.85714287,0.78911567,0.76870745,0.76870745,0.6802721,0.7210884,0.7210884,0.99319726,0.9183673,0.89115644,0.95238096,1.0,0.99319726
+ModernNCA (default),0.020408163,0.054421768,0.06802721,0.034013607,0.0952381,0.06802721,0.034013607,0.04761905,0.13605443,0.054421768,0.08163265,0.06122449,0.12244898,0.19727892,0.14965986,0.17006803,0.1904762,0.2244898,0.18367347,0.23129252,0.2244898,0.17687075,0.0952381,0.21088435,0.1904762,0.21768707,0.21768707,0.2993197,0.24489796,0.20408164,0.26530612,0.3265306,0.2857143,0.34013605,0.3877551,0.4217687,0.4217687,0.3605442,0.3809524,0.3537415,0.46938777,0.46938777,0.39455783,0.5170068,0.47619048,0.4829932,0.40136054,,0.63265306,0.4829932,0.50340134,0.5714286,0.4829932,0.50340134,0.6530612,0.7414966,0.7619048,0.6666667,0.72789115,0.81632656,0.8503401,0.81632656,0.8503401,0.8367347,0.76870745,0.8095238,0.8367347,0.7823129,0.7823129,0.877551,0.9047619,0.8979592,0.85714287,0.92517006,0.8979592,0.8639456,0.97959185,1.0
+RandomForest (tuned),0.0952381,0.0952381,0.0952381,0.0952381,0.088435374,0.13605443,0.10204082,0.1292517,0.15646258,0.088435374,0.14285715,0.10204082,0.15646258,0.17687075,0.1632653,0.21088435,0.24489796,0.19727892,0.24489796,0.25170067,0.27891156,0.17687075,0.18367347,0.20408164,0.23129252,0.26530612,0.292517,0.21768707,0.2857143,0.23809524,0.30612245,0.2993197,0.3809524,0.4217687,0.14965986,0.4489796,0.37414965,0.40816328,0.4489796,0.2857143,0.20068027,0.414966,0.39455783,0.36394557,0.43537414,0.47619048,0.46258503,0.36734694,,0.46258503,0.47619048,0.585034,0.5102041,0.5170068,0.64625853,0.5986394,0.60544217,0.61904764,0.65986395,0.8435374,0.79591835,0.85714287,0.8197279,0.78911567,0.67346936,0.6666667,0.79591835,0.877551,0.8707483,0.89115644,0.9183673,0.91156465,0.8707483,0.92517006,0.9727891,0.9183673,0.8979592,1.0
+TorchMLP (tuned),0.06802721,0.054421768,0.054421768,0.04761905,0.10204082,0.08163265,0.108843535,0.12244898,0.108843535,0.14285715,0.10204082,0.18367347,0.12244898,0.1292517,0.1904762,0.23129252,0.1292517,0.23129252,0.17006803,0.27210885,0.1632653,0.21088435,0.23809524,0.23129252,0.24489796,0.27210885,0.26530612,0.3265306,0.27210885,0.3197279,0.27210885,0.30612245,0.21768707,0.2244898,0.40816328,0.2993197,0.47619048,0.18707483,0.31292516,0.3877551,0.46938777,0.40816328,0.36734694,0.4829932,0.40136054,0.36734694,0.49659863,0.5170068,0.53741497,,0.49659863,0.3809524,0.6394558,0.5102041,0.5102041,0.5578231,0.5578231,0.5986394,0.8095238,0.68707484,0.65986395,0.6802721,0.707483,0.7346939,0.7346939,0.78911567,0.85714287,0.8639456,0.85714287,0.7823129,0.829932,0.81632656,0.92517006,0.9659864,0.93877554,0.95238096,0.9863946,1.0
+XGBoost (default),0.06802721,0.020408163,0.013605442,0.006802721,0.054421768,0.020408163,0.020408163,0.07482993,0.08163265,0.006802721,0.040816326,0.054421768,0.06802721,0.14285715,0.10204082,0.13605443,0.11564626,0.17006803,0.1632653,0.19727892,0.15646258,0.054421768,0.1292517,0.11564626,0.19727892,0.06122449,0.06802721,0.23809524,0.15646258,0.17006803,0.14965986,0.25170067,0.2585034,0.292517,0.2993197,0.46258503,0.44217688,0.3809524,0.40136054,0.3265306,0.36734694,0.43537414,0.3265306,0.4489796,0.45578232,0.48979592,0.44217688,0.49659863,0.52380955,0.50340134,,0.6122449,0.52380955,0.5782313,0.6802721,0.64625853,0.6802721,0.67346936,0.75510204,0.8027211,0.8639456,0.88435376,0.8639456,0.8639456,0.7210884,0.7619048,0.9591837,0.877551,0.8639456,0.89115644,0.92517006,0.9183673,0.9455782,0.9863946,0.92517006,0.9863946,1.0,1.0
+EBM (default),0.07482993,0.054421768,0.06802721,0.06122449,0.07482993,0.07482993,0.08163265,0.0952381,0.08163265,0.06802721,0.08163265,0.10204082,0.0952381,0.1292517,0.20408164,0.23129252,0.108843535,0.14965986,0.13605443,0.27210885,0.1292517,0.1292517,0.20408164,0.15646258,0.23809524,0.15646258,0.15646258,0.292517,0.1904762,0.23129252,0.18367347,0.24489796,0.17687075,0.23809524,0.2993197,0.36734694,0.49659863,0.42857143,0.15646258,0.33333334,0.3537415,0.3809524,0.2993197,0.39455783,0.4217687,0.2993197,0.48979592,0.42857143,0.414966,0.61904764,0.3877551,,0.5714286,0.52380955,0.47619048,0.60544217,0.6258503,0.61904764,0.76870745,0.75510204,0.7482993,0.78911567,0.7210884,0.75510204,0.7823129,0.85714287,0.8639456,0.9047619,0.91156465,0.877551,0.8979592,0.8707483,0.9659864,0.97959185,0.9727891,0.9727891,1.0,1.0
+FastaiMLP (tuned),0.06122449,0.034013607,0.04761905,0.06122449,0.06122449,0.020408163,0.06122449,0.10204082,0.10204082,0.088435374,0.06802721,0.11564626,0.11564626,0.13605443,0.18367347,0.13605443,0.12244898,0.1904762,0.10204082,0.2244898,0.1632653,0.17687075,0.19727892,0.21768707,0.26530612,0.26530612,0.2585034,0.31292516,0.21768707,0.2244898,0.23809524,0.34013605,0.2244898,0.19727892,0.3877551,0.27891156,0.3877551,0.20408164,0.3265306,0.3265306,0.4829932,0.46258503,0.33333334,0.46258503,0.34013605,0.37414965,0.24489796,0.5170068,0.48979592,0.3605442,0.47619048,0.42857143,,0.4829932,0.53061223,0.6258503,0.5986394,0.56462586,0.64625853,0.65986395,0.70068026,0.61904764,0.63265306,0.75510204,0.7482993,0.79591835,0.78911567,0.7346939,0.7482993,0.65986395,0.70068026,0.707483,0.9863946,0.829932,0.89115644,0.93877554,1.0,0.99319726
+RealMLP (default),0.020408163,0.006802721,0.013605442,0.020408163,0.006802721,0.034013607,0.013605442,0.07482993,0.08163265,0.006802721,0.020408163,0.020408163,0.12244898,0.088435374,0.14285715,0.15646258,0.04761905,0.14965986,0.1292517,0.25170067,0.088435374,0.054421768,0.11564626,0.0952381,0.1904762,0.15646258,0.15646258,0.3197279,0.17006803,0.18367347,0.18367347,0.292517,0.14285715,0.15646258,0.33333334,0.2585034,0.42857143,0.30612245,0.26530612,0.27210885,0.40816328,0.44217688,0.292517,0.43537414,0.34013605,0.3537415,0.40816328,0.49659863,0.4829932,0.48979592,0.4217687,0.47619048,0.5170068,,0.5102041,0.5442177,0.56462586,0.5578231,0.7414966,0.68707484,0.6530612,0.6666667,0.72789115,0.7482993,0.7823129,0.8435374,0.829932,0.8367347,0.8367347,0.829932,0.89115644,0.8435374,0.9319728,0.92517006,0.92517006,0.93877554,0.99319726,1.0
+LightGBM (default),0.07482993,0.054421768,0.06802721,0.027210884,0.088435374,0.06802721,0.07482993,0.07482993,0.07482993,0.07482993,0.08163265,0.06122449,0.06122449,0.14285715,0.14285715,0.14965986,0.11564626,0.13605443,0.14285715,0.15646258,0.21088435,0.11564626,0.19727892,0.15646258,0.20408164,0.12244898,0.13605443,0.14965986,0.06122449,0.17006803,0.08163265,0.292517,0.27210885,0.3537415,0.21088435,0.40816328,0.26530612,0.3605442,0.40816328,0.292517,0.27210885,0.36734694,0.27210885,0.26530612,0.25170067,0.50340134,0.414966,0.3469388,0.3537415,0.48979592,0.3197279,0.52380955,0.46938777,0.48979592,,0.47619048,0.48979592,0.5102041,0.64625853,0.7619048,0.7619048,0.7414966,0.6530612,0.6258503,0.5986394,0.64625853,0.82312924,0.8027211,0.7755102,0.8707483,0.91156465,0.877551,0.88435376,0.88435376,0.92517006,0.93877554,0.97959185,1.0
+ChimeraBoost (tuned),0.07482993,0.04761905,0.06122449,0.027210884,0.088435374,0.06802721,0.0952381,0.08163265,0.06802721,0.06122449,0.08163265,0.088435374,0.07482993,0.13605443,0.108843535,0.11564626,0.14285715,0.1632653,0.12244898,0.14285715,0.20408164,0.1292517,0.12244898,0.15646258,0.17687075,0.11564626,0.12244898,0.1904762,0.054421768,0.108843535,0.06802721,0.13605443,0.25170067,0.26530612,0.1904762,0.46938777,0.34013605,0.3469388,0.30612245,0.2244898,0.23129252,0.2585034,0.21088435,0.3197279,0.3877551,0.36734694,0.31292516,0.2585034,0.40136054,0.44217688,0.3537415,0.39455783,0.37414965,0.45578232,0.52380955,,0.4727891,0.6394558,0.5986394,0.63265306,0.92517006,0.7414966,0.61904764,0.68707484,0.6802721,0.70068026,0.82312924,0.7823129,0.78911567,0.71428573,0.7755102,0.7619048,0.9455782,0.9319728,0.85714287,0.91156465,0.9727891,1.0
+ChimeraBoost (tuned + ensembled),0.07482993,0.034013607,0.04761905,0.013605442,0.08163265,0.054421768,0.07482993,0.07482993,0.07482993,0.054421768,0.06122449,0.06122449,0.08163265,0.13605443,0.08163265,0.11564626,0.11564626,0.15646258,0.11564626,0.12244898,0.18367347,0.088435374,0.13605443,0.15646258,0.17006803,0.11564626,0.108843535,0.17687075,0.013605442,0.0952381,0.054421768,0.11564626,0.2244898,0.26530612,0.18367347,0.48979592,0.36734694,0.3469388,0.292517,0.26530612,0.23809524,0.2857143,0.19727892,0.31292516,0.39455783,0.34013605,0.3265306,0.23809524,0.39455783,0.44217688,0.3197279,0.37414965,0.40136054,0.43537414,0.5102041,0.5272109,,0.65986395,0.59183675,0.63265306,0.92517006,0.7482993,0.61904764,0.70068026,0.6666667,0.707483,0.829932,0.78911567,0.79591835,0.71428573,0.79591835,0.7755102,0.9659864,0.9659864,0.85714287,0.93877554,0.97959185,1.0
+xRFM (default),0.07482993,0.020408163,0.034013607,0.013605442,0.06802721,0.06802721,0.020408163,0.108843535,0.108843535,0.06802721,0.0952381,0.027210884,0.17006803,0.14965986,0.14965986,0.2244898,0.19727892,0.17006803,0.21088435,0.18367347,0.2244898,0.10204082,0.108843535,0.1292517,0.14285715,0.21768707,0.21768707,0.2585034,0.24489796,0.2244898,0.23129252,0.040816326,0.2857143,0.31292516,0.1904762,0.414966,0.30612245,0.3469388,0.26530612,0.27210885,0.25170067,0.08163265,0.2993197,0.2857143,0.40136054,0.31292516,0.44217688,0.33333334,0.3809524,0.40136054,0.3265306,0.3809524,0.43537414,0.44217688,0.48979592,0.3605442,0.34013605,,0.49659863,0.6258503,0.5578231,0.65986395,0.71428573,0.53061223,0.65986395,0.6666667,0.585034,0.8095238,0.8027211,0.7414966,0.7755102,0.7755102,0.72789115,0.7482993,0.9183673,0.97959185,0.9319728,1.0
+TorchMLP (default),0.06802721,0.006802721,0.013605442,0.013605442,0.040816326,0.054421768,0.034013607,0.108843535,0.0952381,0.027210884,0.034013607,0.040816326,0.12244898,0.10204082,0.108843535,0.15646258,0.054421768,0.13605443,0.088435374,0.14285715,0.06802721,0.040816326,0.07482993,0.06802721,0.088435374,0.08163265,0.08163265,0.2244898,0.11564626,0.14285715,0.13605443,0.14965986,0.06122449,0.088435374,0.2244898,0.2857143,0.36734694,0.108843535,0.08163265,0.14965986,0.27210885,0.20408164,0.14285715,0.2993197,0.3197279,0.12244898,0.3265306,0.27210885,0.34013605,0.1904762,0.24489796,0.23129252,0.3537415,0.2585034,0.3537415,0.40136054,0.40816328,0.50340134,,0.61904764,0.5578231,0.5986394,0.6122449,0.6394558,0.6666667,0.71428573,0.52380955,0.75510204,0.7619048,0.6258503,0.6938776,0.6666667,0.8503401,0.8027211,0.92517006,0.89115644,0.93877554,1.0
+KNN (tuned + ensembled),0.06802721,0.06802721,0.07482993,0.020408163,0.12244898,0.04761905,0.054421768,0.06802721,0.108843535,0.08163265,0.07482993,0.054421768,0.06122449,0.12244898,0.034013607,0.08163265,0.11564626,0.14285715,0.17687075,0.040816326,0.15646258,0.13605443,0.088435374,0.15646258,0.08163265,0.15646258,0.15646258,0.06802721,0.14965986,0.14965986,0.14285715,0.088435374,0.20408164,0.23809524,0.04761905,0.43537414,0.14965986,0.27210885,0.25170067,0.23809524,0.0952381,0.1292517,0.20408164,0.06802721,0.15646258,0.30612245,0.36734694,0.18367347,0.15646258,0.31292516,0.19727892,0.24489796,0.34013605,0.31292516,0.23809524,0.36734694,0.36734694,0.37414965,0.3809524,,0.5714286,0.45578232,0.414966,0.44217688,0.48979592,0.49659863,0.4829932,0.4829932,0.49659863,0.68707484,0.72789115,0.71428573,0.63265306,0.585034,0.9319728,0.6802721,0.7482993,0.99319726
+ChimeraBoost (default),0.06122449,0.040816326,0.06122449,0.013605442,0.06802721,0.054421768,0.06802721,0.06802721,0.06802721,0.06122449,0.054421768,0.054421768,0.06122449,0.0952381,0.088435374,0.11564626,0.04761905,0.11564626,0.07482993,0.08163265,0.088435374,0.08163265,0.11564626,0.08163265,0.088435374,0.07482993,0.07482993,0.13605443,0.020408163,0.06802721,0.034013607,0.06122449,0.10204082,0.18367347,0.1292517,0.3605442,0.17687075,0.31292516,0.17006803,0.1632653,0.14965986,0.11564626,0.06122449,0.14965986,0.21768707,0.21768707,0.27210885,0.14965986,0.20408164,0.34013605,0.13605443,0.25170067,0.2993197,0.3469388,0.23809524,0.07482993,0.07482993,0.44217688,0.44217688,0.42857143,,0.53061223,0.46938777,0.27210885,0.585034,0.59183675,0.67346936,0.70068026,0.7210884,0.6258503,0.6802721,0.65986395,0.7346939,0.64625853,0.8435374,0.8095238,0.9047619,1.0
+OrionMSP (default),0.06802721,0.020408163,0.013605442,0.006802721,0.013605442,0.034013607,0.006802721,0.088435374,0.088435374,0.013605442,0.020408163,0.027210884,0.0952381,0.108843535,0.040816326,0.06122449,0.06122449,0.10204082,0.0952381,0.040816326,0.07482993,0.027210884,0.04761905,0.027210884,0.040816326,0.040816326,0.04761905,0.10204082,0.08163265,0.13605443,0.06802721,0.027210884,0.12244898,0.1632653,0.054421768,0.3469388,0.1904762,0.23809524,0.1292517,0.13605443,0.10204082,0.054421768,0.108843535,0.088435374,0.07482993,0.18367347,0.36734694,0.18367347,0.14285715,0.3197279,0.11564626,0.21088435,0.3809524,0.33333334,0.2585034,0.2585034,0.25170067,0.34013605,0.40136054,0.5442177,0.46938777,,0.56462586,0.44217688,0.5442177,0.56462586,0.5170068,0.68707484,0.6666667,0.65986395,0.75510204,0.7346939,0.7346939,0.7482993,0.91156465,0.89115644,0.89115644,1.0
+RandomForest (default),0.006802721,0.013605442,0.013605442,0.006802721,0.040816326,0.0,0.0,0.006802721,0.10204082,0.027210884,0.027210884,0.0,0.037414964,0.108843535,0.020408163,0.0,0.108843535,0.11904762,0.10204082,0.040816326,0.11564626,0.08163265,0.054421768,0.088435374,0.08163265,0.088435374,0.0952381,0.05782313,0.10204082,0.08163265,0.11564626,0.034013607,0.17687075,0.1904762,0.034013607,0.3469388,0.19727892,0.2585034,0.21088435,0.13605443,0.013605442,0.08163265,0.15646258,0.108843535,0.14285715,0.24489796,0.3537415,0.14965986,0.1802721,0.292517,0.13605443,0.27891156,0.36734694,0.27210885,0.3469388,0.3809524,0.3809524,0.2857143,0.3877551,0.585034,0.53061223,0.43537414,,0.49659863,0.49659863,0.5578231,0.45578232,0.5442177,0.5442177,0.6938776,0.72789115,0.70068026,0.65986395,0.6394558,0.88435376,0.82312924,0.8639456,1.0
+ExtraTrees (default),0.0,0.006802721,0.006802721,0.006802721,0.006802721,0.0,0.0,0.027210884,0.06122449,0.0,0.020408163,0.0,0.06122449,0.07482993,0.0,0.0,0.088435374,0.14285715,0.07482993,0.020408163,0.14965986,0.054421768,0.013605442,0.07482993,0.034013607,0.07482993,0.088435374,0.06802721,0.054421768,0.006802721,0.054421768,0.006802721,0.18367347,0.19727892,0.0,0.36734694,0.08163265,0.21088435,0.19727892,0.07482993,0.04761905,0.054421768,0.11564626,0.11564626,0.23809524,0.2244898,0.21768707,0.1632653,0.21088435,0.26530612,0.13605443,0.24489796,0.24489796,0.25170067,0.37414965,0.31292516,0.2993197,0.46938777,0.3605442,0.5578231,0.72789115,0.5578231,0.50340134,,0.53061223,0.53061223,0.59183675,0.6122449,0.61904764,0.6394558,0.72789115,0.6938776,0.7755102,0.70068026,0.8367347,0.7755102,0.8503401,1.0
+Linear (tuned + ensembled),0.006802721,0.006802721,0.006802721,0.027210884,0.0,0.020408163,0.013605442,0.013605442,0.013605442,0.006802721,0.013605442,0.013605442,0.020408163,0.034013607,0.14285715,0.13605443,0.040816326,0.06122449,0.054421768,0.1904762,0.027210884,0.040816326,0.088435374,0.04761905,0.11564626,0.054421768,0.054421768,0.1904762,0.11564626,0.18367347,0.108843535,0.040816326,0.06122449,0.088435374,0.23809524,0.19727892,0.4217687,0.1292517,0.06122449,0.23129252,0.292517,0.1292517,0.1632653,0.292517,0.34013605,0.108843535,0.1904762,0.23129252,0.3265306,0.26530612,0.27891156,0.21768707,0.25170067,0.21768707,0.40136054,0.3197279,0.33333334,0.34013605,0.33333334,0.5102041,0.414966,0.45578232,0.50340134,0.46938777,,0.74489796,0.45578232,0.45578232,0.4829932,0.47619048,0.48979592,0.5170068,0.45578232,0.53061223,0.75510204,0.6666667,0.9455782,0.9319728
+Linear (tuned),0.006802721,0.0,0.006802721,0.020408163,0.006802721,0.013605442,0.013605442,0.020408163,0.006802721,0.013605442,0.006802721,0.006802721,0.034013607,0.034013607,0.1292517,0.14965986,0.020408163,0.07482993,0.040816326,0.19727892,0.027210884,0.034013607,0.040816326,0.034013607,0.06802721,0.040816326,0.040816326,0.17687075,0.07482993,0.11564626,0.07482993,0.034013607,0.06122449,0.06802721,0.23809524,0.19727892,0.3537415,0.08163265,0.027210884,0.17006803,0.2857143,0.12244898,0.13605443,0.31292516,0.30612245,0.07482993,0.14285715,0.1904762,0.33333334,0.21088435,0.23809524,0.14285715,0.20408164,0.15646258,0.3537415,0.2993197,0.292517,0.33333334,0.2857143,0.50340134,0.40816328,0.43537414,0.44217688,0.46938777,0.25510204,,0.4489796,0.40136054,0.414966,0.4489796,0.47619048,0.50340134,0.43537414,0.52380955,0.7414966,0.6530612,0.85714287,0.9319728
+iLTM (default),0.06122449,0.0,0.0,0.0,0.013605442,0.0,0.006802721,0.06122449,0.06122449,0.0,0.006802721,0.027210884,0.06122449,0.06122449,0.06122449,0.06122449,0.0,0.108843535,0.020408163,0.06802721,0.0,0.0,0.020408163,0.0,0.027210884,0.006802721,0.006802721,0.108843535,0.0,0.006802721,0.0,0.10204082,0.006802721,0.013605442,0.088435374,0.27891156,0.08163265,0.088435374,0.054421768,0.020408163,0.15646258,0.14285715,0.0,0.14965986,0.1632653,0.0952381,0.21088435,0.1632653,0.20408164,0.14285715,0.040816326,0.13605443,0.21088435,0.17006803,0.17687075,0.17687075,0.17006803,0.414966,0.47619048,0.5170068,0.3265306,0.4829932,0.5442177,0.40816328,0.5442177,0.5510204,,0.68707484,0.6802721,0.6530612,0.6802721,0.6802721,0.65986395,0.6938776,0.75510204,0.7755102,0.76870745,0.9319728
+TabSTAR (tuned),0.06122449,0.013605442,0.006802721,0.006802721,0.013605442,0.006802721,0.0,0.06802721,0.07482993,0.027210884,0.027210884,0.020408163,0.07482993,0.0952381,0.0952381,0.12244898,0.06122449,0.088435374,0.06122449,0.0952381,0.07482993,0.020408163,0.088435374,0.034013607,0.07482993,0.04761905,0.040816326,0.14965986,0.054421768,0.13605443,0.04761905,0.034013607,0.04761905,0.07482993,0.088435374,0.26530612,0.23809524,0.10204082,0.06802721,0.15646258,0.07482993,0.07482993,0.08163265,0.088435374,0.1292517,0.08163265,0.23129252,0.21768707,0.12244898,0.13605443,0.12244898,0.0952381,0.26530612,0.1632653,0.19727892,0.21768707,0.21088435,0.1904762,0.24489796,0.5170068,0.2993197,0.31292516,0.45578232,0.3877551,0.5442177,0.5986394,0.31292516,,0.53741497,0.5102041,0.5510204,0.53061223,0.585034,0.42857143,0.8503401,1.0,0.9047619,0.9863946
+TabSTAR (tuned + ensembled),0.06802721,0.013605442,0.006802721,0.006802721,0.020408163,0.006802721,0.020408163,0.06802721,0.07482993,0.034013607,0.040816326,0.034013607,0.07482993,0.108843535,0.0952381,0.12244898,0.04761905,0.0952381,0.06122449,0.0952381,0.04761905,0.027210884,0.0952381,0.040816326,0.06802721,0.054421768,0.04761905,0.14965986,0.054421768,0.14285715,0.040816326,0.034013607,0.04761905,0.06802721,0.088435374,0.27891156,0.25170067,0.0952381,0.06122449,0.15646258,0.07482993,0.06802721,0.088435374,0.0952381,0.14285715,0.07482993,0.23129252,0.21768707,0.1292517,0.14285715,0.13605443,0.088435374,0.25170067,0.1632653,0.2244898,0.21088435,0.20408164,0.19727892,0.23809524,0.50340134,0.27891156,0.33333334,0.45578232,0.3809524,0.5170068,0.585034,0.3197279,0.46258503,,0.49659863,0.53741497,0.53061223,0.5714286,0.4217687,0.8503401,1.0,0.88435376,0.9863946
+PerpetualBooster (tuned + ensembled),0.006802721,0.013605442,0.013605442,0.006802721,0.013605442,0.034013607,0.027210884,0.027210884,0.027210884,0.020408163,0.013605442,0.020408163,0.04761905,0.034013607,0.040816326,0.06122449,0.04761905,0.020408163,0.06122449,0.06802721,0.040816326,0.027210884,0.020408163,0.027210884,0.034013607,0.040816326,0.040816326,0.054421768,0.06122449,0.108843535,0.06122449,0.088435374,0.06802721,0.06122449,0.07482993,0.24489796,0.14965986,0.1632653,0.08163265,0.108843535,0.088435374,0.108843535,0.08163265,0.088435374,0.06802721,0.10204082,0.3197279,0.12244898,0.108843535,0.21768707,0.108843535,0.12244898,0.34013605,0.17006803,0.1292517,0.2857143,0.2857143,0.2585034,0.37414965,0.31292516,0.37414965,0.34013605,0.30612245,0.3605442,0.52380955,0.5510204,0.3469388,0.48979592,0.50340134,,0.7755102,0.8027211,0.585034,0.63265306,0.6258503,0.5510204,0.65986395,0.93877554
+PerpetualBooster (tuned),0.027210884,0.027210884,0.027210884,0.027210884,0.027210884,0.04761905,0.027210884,0.034013607,0.034013607,0.027210884,0.027210884,0.027210884,0.04761905,0.040816326,0.040816326,0.06122449,0.034013607,0.034013607,0.054421768,0.054421768,0.04761905,0.027210884,0.027210884,0.027210884,0.034013607,0.040816326,0.04761905,0.06122449,0.04761905,0.088435374,0.04761905,0.108843535,0.040816326,0.06802721,0.04761905,0.25170067,0.1292517,0.12244898,0.06802721,0.088435374,0.06802721,0.08163265,0.054421768,0.088435374,0.054421768,0.088435374,0.27891156,0.0952381,0.08163265,0.17006803,0.07482993,0.10204082,0.2993197,0.108843535,0.088435374,0.2244898,0.20408164,0.2244898,0.30612245,0.27210885,0.3197279,0.24489796,0.27210885,0.27210885,0.5102041,0.52380955,0.3197279,0.4489796,0.46258503,0.2244898,,0.6496599,0.5578231,0.5170068,0.61904764,0.50340134,0.6394558,0.9319728
+PerpetualBooster (default),0.034013607,0.034013607,0.027210884,0.034013607,0.027210884,0.034013607,0.034013607,0.034013607,0.034013607,0.034013607,0.040816326,0.034013607,0.040816326,0.027210884,0.04761905,0.054421768,0.04761905,0.034013607,0.04761905,0.04761905,0.054421768,0.034013607,0.04761905,0.034013607,0.054421768,0.034013607,0.040816326,0.054421768,0.06122449,0.08163265,0.06122449,0.108843535,0.06122449,0.07482993,0.06122449,0.25170067,0.11564626,0.1292517,0.06802721,0.0952381,0.06802721,0.0952381,0.08163265,0.08163265,0.054421768,0.0952381,0.27891156,0.10204082,0.088435374,0.18367347,0.08163265,0.1292517,0.292517,0.15646258,0.12244898,0.23809524,0.2244898,0.2244898,0.33333334,0.2857143,0.34013605,0.26530612,0.2993197,0.30612245,0.4829932,0.49659863,0.3197279,0.46938777,0.46938777,0.19727892,0.35034013,,0.53741497,0.53741497,0.59183675,0.53741497,0.6122449,0.7414966
+FastaiMLP (default),0.0,0.0,0.006802721,0.006802721,0.006802721,0.0,0.006802721,0.013605442,0.06122449,0.0,0.0,0.0,0.06802721,0.08163265,0.06122449,0.06122449,0.020408163,0.0952381,0.020408163,0.06122449,0.027210884,0.020408163,0.054421768,0.027210884,0.06122449,0.013605442,0.013605442,0.12244898,0.0,0.006802721,0.0,0.034013607,0.006802721,0.020408163,0.10204082,0.21088435,0.0952381,0.040816326,0.020408163,0.06122449,0.11564626,0.07482993,0.034013607,0.11564626,0.1292517,0.034013607,0.006802721,0.14285715,0.1292517,0.07482993,0.054421768,0.034013607,0.013605442,0.06802721,0.11564626,0.054421768,0.034013607,0.27210885,0.14965986,0.36734694,0.26530612,0.26530612,0.34013605,0.2244898,0.5442177,0.56462586,0.34013605,0.414966,0.42857143,0.414966,0.44217688,0.46258503,,0.3469388,0.7823129,0.7619048,0.8979592,0.99319726
+TabFlex (default),0.034013607,0.0,0.0,0.0,0.006802721,0.0,0.0,0.054421768,0.06122449,0.0,0.0,0.006802721,0.06122449,0.06122449,0.0,0.0,0.0,0.088435374,0.020408163,0.0,0.0,0.0,0.006802721,0.0,0.013605442,0.006802721,0.006802721,0.06122449,0.0,0.0,0.0,0.006802721,0.0,0.0,0.020408163,0.25170067,0.0,0.006802721,0.006802721,0.027210884,0.054421768,0.006802721,0.0,0.034013607,0.04761905,0.006802721,0.08163265,0.07482993,0.07482993,0.034013607,0.013605442,0.020408163,0.17006803,0.07482993,0.11564626,0.06802721,0.034013607,0.25170067,0.19727892,0.414966,0.3537415,0.25170067,0.3605442,0.2993197,0.46938777,0.47619048,0.30612245,0.5714286,0.5782313,0.36734694,0.4829932,0.46258503,0.6530612,,0.79591835,0.76870745,0.81632656,1.0
+KNN (tuned),0.06802721,0.034013607,0.04761905,0.020408163,0.054421768,0.040816326,0.054421768,0.06122449,0.06122449,0.04761905,0.027210884,0.04761905,0.06122449,0.06122449,0.027210884,0.054421768,0.034013607,0.06802721,0.04761905,0.040816326,0.034013607,0.04761905,0.06802721,0.04761905,0.027210884,0.06802721,0.06802721,0.06122449,0.04761905,0.10204082,0.04761905,0.013605442,0.027210884,0.040816326,0.0,0.23809524,0.12244898,0.06122449,0.06122449,0.11564626,0.006802721,0.027210884,0.054421768,0.0,0.04761905,0.06122449,0.108843535,0.10204082,0.027210884,0.06122449,0.07482993,0.027210884,0.108843535,0.07482993,0.07482993,0.14285715,0.14285715,0.08163265,0.07482993,0.06802721,0.15646258,0.088435374,0.11564626,0.1632653,0.24489796,0.2585034,0.24489796,0.14965986,0.14965986,0.37414965,0.3809524,0.40816328,0.21768707,0.20408164,,0.37414965,0.37414965,0.92517006
+TabSTAR (default),0.006802721,0.013605442,0.006802721,0.006802721,0.0,0.006802721,0.0,0.006802721,0.06802721,0.006802721,0.013605442,0.0,0.06802721,0.06802721,0.06802721,0.06122449,0.006802721,0.07482993,0.034013607,0.07482993,0.013605442,0.006802721,0.034013607,0.006802721,0.034013607,0.013605442,0.013605442,0.11564626,0.006802721,0.027210884,0.006802721,0.013605442,0.013605442,0.020408163,0.06802721,0.054421768,0.108843535,0.027210884,0.013605442,0.054421768,0.06122449,0.020408163,0.006802721,0.06802721,0.06802721,0.020408163,0.04761905,0.13605443,0.08163265,0.04761905,0.013605442,0.027210884,0.06122449,0.06122449,0.06122449,0.088435374,0.06122449,0.020408163,0.108843535,0.3197279,0.1904762,0.108843535,0.17687075,0.2244898,0.33333334,0.3469388,0.2244898,0.0,0.0,0.4489796,0.49659863,0.46258503,0.23809524,0.23129252,0.6258503,,0.6802721,0.9591837
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.04761905,0.034013607,0.0,0.006802721,0.006802721,0.06122449,0.0,0.0,0.006802721,0.0,0.006802721,0.0,0.0,0.0,0.0,0.006802721,0.0,0.0,0.0,0.0,0.054421768,0.1904762,0.06802721,0.006802721,0.006802721,0.027210884,0.08163265,0.006802721,0.0,0.07482993,0.034013607,0.006802721,0.0,0.020408163,0.10204082,0.013605442,0.0,0.0,0.0,0.006802721,0.020408163,0.027210884,0.020408163,0.06802721,0.06122449,0.25170067,0.0952381,0.108843535,0.13605443,0.14965986,0.054421768,0.14285715,0.23129252,0.0952381,0.11564626,0.34013605,0.3605442,0.3877551,0.10204082,0.18367347,0.6258503,0.3197279,,0.91156465
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.006802721,0.0,0.0,0.0,0.0,0.0,0.006802721,0.0,0.0,0.0,0.0,0.0,0.0,0.006802721,0.0,0.0,0.0,0.0,0.06802721,0.06802721,0.06802721,0.013605442,0.013605442,0.06122449,0.06802721,0.2585034,0.006802721,0.0,0.07482993,0.040816326,0.088435374,
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_all/leaderboard_overview_explorer.html
index acdb7695bb6b9ed7d6d837600c4d7eb40b8d5b64..37512c4bd06d0c9a44e7693b0e598eabe26364ee 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Regression | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_regression/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..6eabb515d0789effc38d4363216a93191544672e
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_regression/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2004.0,"score":0.918,"rank":3.96,"hrank":1.96,"imp":1.557,"train_time":39.42,"predict_time":6.476,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-110","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1792.0,"score":0.743,"rank":9.3,"hrank":3.44,"imp":2.326,"train_time":3.66,"predict_time":0.399,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+222\/-126","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1770.0,"score":0.687,"rank":10.09,"hrank":4.4,"imp":3.851,"train_time":335.03,"predict_time":4.334,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+133\/-90","imputed":false},{"position":3,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1753.0,"score":0.685,"rank":10.74,"hrank":3.57,"imp":3.349,"train_time":11.58,"predict_time":2.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+117\/-76","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1740.0,"score":0.639,"rank":11.22,"hrank":8.06,"imp":3.94,"train_time":8.52,"predict_time":0.696,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-52","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1739.0,"score":0.627,"rank":11.29,"hrank":6.36,"imp":3.304,"train_time":1709.05,"predict_time":8.119,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+151\/-104","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1716.0,"score":0.626,"rank":12.26,"hrank":4.21,"imp":4.169,"train_time":4786.6,"predict_time":239.304,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-88","imputed":false},{"position":7,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1674.0,"score":0.656,"rank":14.1,"hrank":4.55,"imp":3.875,"train_time":1.32,"predict_time":0.122,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+248\/-142","imputed":false},{"position":8,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1672.0,"score":0.601,"rank":14.2,"hrank":7.62,"imp":4.446,"train_time":6.93,"predict_time":1.389,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+133\/-83","imputed":false},{"position":9,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1663.0,"score":0.574,"rank":14.63,"hrank":7.81,"imp":4.611,"train_time":4786.6,"predict_time":38.497,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-67","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1646.0,"score":0.506,"rank":15.5,"hrank":10.77,"imp":4.973,"train_time":3995.01,"predict_time":10.048,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-64","imputed":false},{"position":11,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1627.0,"score":0.521,"rank":16.46,"hrank":8.59,"imp":4.049,"train_time":1709.05,"predict_time":0.811,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-113","imputed":false},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1597.0,"score":0.495,"rank":18.03,"hrank":9.06,"imp":5.128,"train_time":2.08,"predict_time":0.182,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+176\/-97","imputed":false},{"position":13,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1590.0,"score":0.41,"rank":18.41,"hrank":12.02,"imp":6.369,"train_time":1866.35,"predict_time":6.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+107\/-90","imputed":false},{"position":14,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1575.0,"score":0.474,"rank":19.28,"hrank":9.16,"imp":5.528,"train_time":46.62,"predict_time":39.212,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-65","imputed":false},{"position":15,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1541.0,"score":0.367,"rank":21.23,"hrank":14.24,"imp":5.938,"train_time":3995.01,"predict_time":0.844,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+111\/-80","imputed":false},{"position":16,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1532.0,"score":0.42,"rank":21.73,"hrank":14.47,"imp":5.524,"train_time":7.04,"predict_time":0.508,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-99","imputed":false},{"position":17,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1528.0,"score":0.376,"rank":21.94,"hrank":8.51,"imp":7.234,"train_time":3779.7,"predict_time":7.69,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-122","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1482.0,"score":0.243,"rank":24.77,"hrank":16.57,"imp":7.911,"train_time":2229.21,"predict_time":0.38,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-65","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1474.0,"score":0.244,"rank":25.31,"hrank":19.54,"imp":8.368,"train_time":700.19,"predict_time":9.322,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-83","imputed":false},{"position":20,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1458.0,"score":0.219,"rank":26.29,"hrank":17.68,"imp":8.049,"train_time":2229.21,"predict_time":0.055,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-71","imputed":false},{"position":21,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1458.0,"score":0.274,"rank":26.32,"hrank":18.54,"imp":7.418,"train_time":714.5,"predict_time":1.376,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-90","imputed":false},{"position":22,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1443.0,"score":0.28,"rank":27.31,"hrank":21.15,"imp":6.819,"train_time":3013.86,"predict_time":2.092,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-81","imputed":false},{"position":23,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1431.0,"score":0.221,"rank":28.12,"hrank":21.8,"imp":9.267,"train_time":12685.08,"predict_time":321.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-46","imputed":false},{"position":24,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1411.0,"score":0.18,"rank":29.5,"hrank":23.32,"imp":8.946,"train_time":700.19,"predict_time":0.968,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-78","imputed":false},{"position":25,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1400.0,"score":0.158,"rank":30.26,"hrank":27.27,"imp":8.889,"train_time":834.93,"predict_time":2.614,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-44","imputed":false},{"position":26,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1391.0,"score":0.265,"rank":30.9,"hrank":20.22,"imp":7.186,"train_time":0.91,"predict_time":0.075,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+151\/-108","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1380.0,"score":0.137,"rank":31.71,"hrank":28.11,"imp":9.034,"train_time":834.93,"predict_time":0.388,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-53","imputed":false},{"position":28,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1375.0,"score":0.149,"rank":32.01,"hrank":23.59,"imp":9.538,"train_time":8.73,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-93","imputed":false},{"position":29,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1375.0,"score":0.192,"rank":32.03,"hrank":27.43,"imp":8.08,"train_time":714.5,"predict_time":0.099,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-83","imputed":false},{"position":30,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1370.0,"score":0.193,"rank":32.4,"hrank":24.37,"imp":7.73,"train_time":3013.86,"predict_time":0.181,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-92","imputed":false},{"position":31,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1368.0,"score":0.206,"rank":32.56,"hrank":21.4,"imp":9.192,"train_time":3779.7,"predict_time":0.395,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-87","imputed":false},{"position":32,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1325.0,"score":0.106,"rank":35.58,"hrank":31.06,"imp":9.222,"train_time":12685.08,"predict_time":59.099,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-57","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1320.0,"score":0.118,"rank":35.95,"hrank":26.88,"imp":11.237,"train_time":420.1,"predict_time":0.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-112","imputed":false},{"position":34,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1317.0,"score":0.242,"rank":36.14,"hrank":18.01,"imp":8.59,"train_time":4223.87,"predict_time":27.543,"imputed_pct":46.15,"hardware":"GPU","elo_ci":"+156\/-113","imputed":true},{"position":35,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1285.0,"score":0.094,"rank":38.41,"hrank":30.15,"imp":11.522,"train_time":420.1,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-107","imputed":false},{"position":36,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1273.0,"score":0.109,"rank":39.23,"hrank":32.27,"imp":9.355,"train_time":9.19,"predict_time":0.132,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-105","imputed":false},{"position":37,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1268.0,"score":0.102,"rank":39.57,"hrank":31.46,"imp":10.702,"train_time":15.5,"predict_time":0.298,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-68","imputed":false},{"position":38,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1260.0,"score":0.154,"rank":40.12,"hrank":21.3,"imp":10.766,"train_time":20.24,"predict_time":6.621,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-127","imputed":false},{"position":39,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1260.0,"score":0.151,"rank":40.14,"hrank":22.42,"imp":10.212,"train_time":71.06,"predict_time":1.848,"imputed_pct":46.15,"hardware":"GPU","elo_ci":"+123\/-97","imputed":true},{"position":40,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1251.0,"score":0.138,"rank":40.76,"hrank":18.41,"imp":10.306,"train_time":74.68,"predict_time":19.758,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+161\/-155","imputed":false},{"position":41,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1232.0,"score":0.151,"rank":42.14,"hrank":24.72,"imp":9.635,"train_time":4223.87,"predict_time":0.451,"imputed_pct":46.15,"hardware":"GPU","elo_ci":"+150\/-125","imputed":true},{"position":42,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1230.0,"score":0.095,"rank":42.27,"hrank":30.79,"imp":10.893,"train_time":4608.59,"predict_time":1.233,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-96","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1220.0,"score":0.095,"rank":42.95,"hrank":33.67,"imp":12.881,"train_time":263.04,"predict_time":0.808,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-99","imputed":false},{"position":44,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1218.0,"score":0.077,"rank":43.09,"hrank":37.35,"imp":10.443,"train_time":8.9,"predict_time":1.644,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-84","imputed":false},{"position":45,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1218.0,"score":0.054,"rank":43.12,"hrank":38.31,"imp":12.5,"train_time":2.68,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-119","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1199.0,"score":0.08,"rank":44.46,"hrank":33.03,"imp":13.105,"train_time":263.04,"predict_time":0.066,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-98","imputed":false},{"position":47,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1198.0,"score":0.009,"rank":44.54,"hrank":42.82,"imp":11.327,"train_time":2.11,"predict_time":0.275,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-43","imputed":false},{"position":48,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1183.0,"score":0.12,"rank":45.61,"hrank":33.56,"imp":11.004,"train_time":2.8,"predict_time":0.313,"imputed_pct":46.15,"hardware":"GPU","elo_ci":"+129\/-134","imputed":true},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1182.0,"score":0.034,"rank":45.69,"hrank":40.19,"imp":11.981,"train_time":2.24,"predict_time":0.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-83","imputed":false},{"position":50,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1178.0,"score":0.061,"rank":45.98,"hrank":38.97,"imp":11.671,"train_time":4608.59,"predict_time":0.097,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-98","imputed":false},{"position":51,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1172.0,"score":0.027,"rank":46.34,"hrank":43.07,"imp":13.161,"train_time":162.38,"predict_time":0.363,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-91","imputed":false},{"position":52,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1152.0,"score":0.024,"rank":47.75,"hrank":44.78,"imp":13.914,"train_time":515.75,"predict_time":0.771,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-68","imputed":false},{"position":53,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1146.0,"score":0.036,"rank":48.17,"hrank":41.55,"imp":13.686,"train_time":2.45,"predict_time":0.742,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-113","imputed":false},{"position":54,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1144.0,"score":0.063,"rank":48.33,"hrank":38.09,"imp":14.43,"train_time":1865.27,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-160","imputed":false},{"position":55,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1120.0,"score":0.01,"rank":49.92,"hrank":47.31,"imp":14.225,"train_time":162.38,"predict_time":0.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-84","imputed":false},{"position":56,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1115.0,"score":0.01,"rank":50.2,"hrank":47.17,"imp":14.394,"train_time":515.75,"predict_time":0.124,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-89","imputed":false},{"position":57,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1100.0,"score":0.047,"rank":51.16,"hrank":40.59,"imp":14.958,"train_time":1865.27,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+121\/-163","imputed":false},{"position":58,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1053.0,"score":0.009,"rank":54.03,"hrank":46.88,"imp":15.541,"train_time":2.54,"predict_time":0.24,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-120","imputed":false},{"position":59,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1042.0,"score":0.025,"rank":54.69,"hrank":47.21,"imp":15.875,"train_time":7.87,"predict_time":0.011,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-166","imputed":false},{"position":60,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1028.0,"score":0.002,"rank":55.49,"hrank":52.41,"imp":15.037,"train_time":20.49,"predict_time":0.08,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-111","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1024.0,"score":0.023,"rank":55.71,"hrank":51.95,"imp":15.281,"train_time":540.14,"predict_time":2.672,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-112","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":57.04,"hrank":55.31,"imp":15.887,"train_time":0.53,"predict_time":0.062,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-75","imputed":false},{"position":63,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":979.0,"score":0.014,"rank":58.19,"hrank":55.29,"imp":15.759,"train_time":540.14,"predict_time":0.325,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-113","imputed":false},{"position":64,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":962.0,"score":0.113,"rank":59.07,"hrank":22.3,"imp":23.453,"train_time":28729.74,"predict_time":19.901,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+216\/-299","imputed":false},{"position":65,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":948.0,"score":0.107,"rank":59.76,"hrank":24.71,"imp":23.618,"train_time":28729.74,"predict_time":5.236,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+217\/-300","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":940.0,"score":0.002,"rank":60.14,"hrank":58.46,"imp":17.688,"train_time":27.32,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-116","imputed":false},{"position":67,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":889.0,"score":0.0,"rank":62.51,"hrank":55.75,"imp":20.913,"train_time":92.55,"predict_time":0.897,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+138\/-148","imputed":false},{"position":68,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":872.0,"score":0.0,"rank":63.25,"hrank":62.23,"imp":18.896,"train_time":357.02,"predict_time":81.613,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-116","imputed":false},{"position":69,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":864.0,"score":0.0,"rank":63.57,"hrank":61.91,"imp":19.929,"train_time":2.6,"predict_time":0.392,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-155","imputed":false},{"position":70,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":843.0,"score":0.05,"rank":64.4,"hrank":33.75,"imp":26.547,"train_time":323.52,"predict_time":4.929,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+230\/-344","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":787.0,"score":0.0,"rank":66.44,"hrank":62.36,"imp":23.286,"train_time":92.55,"predict_time":0.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+138\/-167","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":686.0,"score":0.0,"rank":69.34,"hrank":56.9,"imp":30.341,"train_time":0.19,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+167\/-237","imputed":false},{"position":73,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":501.0,"score":0.0,"rank":72.77,"hrank":72.52,"imp":37.422,"train_time":193.98,"predict_time":0.173,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-367","imputed":false},{"position":74,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":467.0,"score":0.0,"rank":73.2,"hrank":72.93,"imp":37.593,"train_time":193.98,"predict_time":0.074,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+140\/-447","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":298.0,"score":0.0,"rank":74.89,"hrank":74.85,"imp":39.985,"train_time":0.95,"predict_time":0.097,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+145\/-401","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_all/pareto_front_explorer.html
index 1dd8be500b88a6d382a8cf6d5d97a6b2ca7ab521..d18403dad9065f967590b3fddf5d8178d5327f5b 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.016935661,"imp":9.5378434703,"elo":1375.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0551291653,"imp":8.0488180904,"elo":1458.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3802116897,"imp":7.9112641225,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0297627373,"imp":12.5000900797,"elo":1217.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0341858448,"imp":11.5221658925,"elo":1284.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"imp":11.2373809843,"elo":1320.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0108193564,"imp":15.8748135928,"elo":1041.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0116341148,"imp":14.9579207126,"elo":1100.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1165303532,"imp":14.4303563856,"elo":1143.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.24026221,"imp":15.5405757602,"elo":1052.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0658904991,"imp":13.1051718626,"elo":1198.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8077239936,"imp":12.8807484442,"elo":1220.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3915661558,"imp":19.9290138007,"elo":864.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.324857112,"imp":15.7585901151,"elo":978.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.6723835679,"imp":15.2805874958,"elo":1023.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2745146333,"imp":11.3266771495,"elo":1197.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9682498776,"imp":8.9458975327,"elo":1410.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":9.3218182128,"imp":8.3676917171,"elo":1473.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":19.7582527317,"imp":10.3056469019,"elo":1251.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":1.848279835,"imp":10.2120891669,"elo":1260.0,"imputed":true,"imputed_pct":46.1538461538,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2983006451,"imp":10.7021818552,"elo":1268.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3953581979,"imp":9.1922472156,"elo":1367.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.6904227381,"imp":7.2343739368,"elo":1528.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.3890423048,"imp":4.4459713547,"elo":1672.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0226031244,"imp":3.3492403256,"elo":1752.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.020234932,"imp":17.6884202929,"elo":940.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1695058591,"imp":14.2248312499,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3627650406,"imp":13.1606855861,"elo":1172.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0621551845,"imp":15.8865082935,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1235618305,"imp":14.3940660144,"elo":1115.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7709478321,"imp":13.9141114822,"elo":1152.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.644117959,"imp":10.4433294011,"elo":1218.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.8438670931,"imp":5.9378331413,"elo":1540.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":10.0484057771,"imp":4.9730851056,"elo":1645.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"imp":5.5239227792,"elo":1532.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8108737399,"imp":4.0488895491,"elo":1626.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1192270005,"imp":3.3038411005,"elo":1739.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.620625058,"imp":10.7664752895,"elo":1260.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"imp":5.5283691171,"elo":1574.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.4972730809,"imp":4.6110596122,"elo":1663.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":239.3040135069,"imp":4.1688637663,"elo":1715.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"imp":5.127593625,"elo":1597.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4760486585,"imp":1.5572886698,"elo":2004.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1222431699,"imp":3.8749858511,"elo":1674.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1324525523,"imp":9.3551363844,"elo":1273.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1814320495,"imp":7.7301423167,"elo":1369.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.0923554564,"imp":6.8194097728,"elo":1442.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.696318366,"imp":3.9399238497,"elo":1740.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"imp":2.325985518,"elo":1791.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3129231924,"imp":11.0039225884,"elo":1182.9,"imputed":true,"imputed_pct":46.1538461538,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.451160218,"imp":9.635148448,"elo":1231.6,"imputed":true,"imputed_pct":46.1538461538,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":27.5427954877,"imp":8.5900286965,"elo":1317.4,"imputed":true,"imputed_pct":46.1538461538,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9291777504,"imp":26.5470868186,"elo":843.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.23600059,"imp":23.6177601403,"elo":948.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.9007182558,"imp":23.4531065734,"elo":961.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0752279828,"imp":7.1861494306,"elo":1390.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0796563802,"imp":15.0373165529,"elo":1027.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0968974085,"imp":11.6709471148,"elo":1177.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.232537078,"imp":10.8925056509,"elo":1229.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.242470131,"imp":11.9809667662,"elo":1181.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3882893337,"imp":9.0340846296,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.6142653492,"imp":8.8886521695,"elo":1399.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":81.6126739219,"imp":18.8962293987,"elo":872.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":59.0992991738,"imp":9.2221582665,"elo":1325.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":321.7409586602,"imp":9.2665052414,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7424024527,"imp":13.6861569247,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0990665936,"imp":8.0801831737,"elo":1375.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.3755494169,"imp":7.4176531706,"elo":1457.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.016935661,"x_train":8.7329235239,"imp":9.5378434703,"elo":1375.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0551291653,"x_train":2229.2144624726,"imp":8.0488180904,"elo":1458.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3802116897,"x_train":2229.2144624726,"imp":7.9112641225,"elo":1482.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0297627373,"x_train":2.6789664191,"imp":12.5000900797,"elo":1217.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0341858448,"x_train":420.0975247537,"imp":11.5221658925,"elo":1284.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"x_train":420.0975247537,"imp":11.2373809843,"elo":1320.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0108193564,"x_train":7.8692134542,"imp":15.8748135928,"elo":1041.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0116341148,"x_train":1865.2663509929,"imp":14.9579207126,"elo":1100.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1165303532,"x_train":1865.2663509929,"imp":14.4303563856,"elo":1143.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.24026221,"x_train":2.5445212109,"imp":15.5405757602,"elo":1052.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0658904991,"x_train":263.0379305179,"imp":13.1051718626,"elo":1198.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8077239936,"x_train":263.0379305179,"imp":12.8807484442,"elo":1220.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3915661558,"x_train":2.6039043157,"imp":19.9290138007,"elo":864.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.324857112,"x_train":540.1381360311,"imp":15.7585901151,"elo":978.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.6723835679,"x_train":540.1381360311,"imp":15.2805874958,"elo":1023.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2745146333,"x_train":2.1108344816,"imp":11.3266771495,"elo":1197.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9682498776,"x_train":700.1867028829,"imp":8.9458975327,"elo":1410.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":9.3218182128,"x_train":700.1867028829,"imp":8.3676917171,"elo":1473.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":19.7582527317,"x_train":74.6790742113,"imp":10.3056469019,"elo":1251.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":1.848279835,"x_train":71.0611073496,"imp":10.2120891669,"elo":1260.0,"imputed":true,"imputed_pct":46.1538461538,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2983006451,"x_train":15.5008512826,"imp":10.7021818552,"elo":1268.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3953581979,"x_train":3779.7025911156,"imp":9.1922472156,"elo":1367.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.6904227381,"x_train":3779.7025911156,"imp":7.2343739368,"elo":1528.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.3890423048,"x_train":6.9283541331,"imp":4.4459713547,"elo":1672.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0226031244,"x_train":11.5842534555,"imp":3.3492403256,"elo":1752.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.020234932,"x_train":27.322832323,"imp":17.6884202929,"elo":940.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1695058591,"x_train":162.3794942955,"imp":14.2248312499,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3627650406,"x_train":162.3794942955,"imp":13.1606855861,"elo":1172.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0621551845,"x_train":0.5271703851,"imp":15.8865082935,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1235618305,"x_train":515.7544728433,"imp":14.3940660144,"elo":1115.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7709478321,"x_train":515.7544728433,"imp":13.9141114822,"elo":1152.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.644117959,"x_train":8.9020711389,"imp":10.4433294011,"elo":1218.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.8438670931,"x_train":3995.0078687545,"imp":5.9378331413,"elo":1540.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":10.0484057771,"x_train":3995.0078687545,"imp":4.9730851056,"elo":1645.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"x_train":7.0435968115,"imp":5.5239227792,"elo":1532.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8108737399,"x_train":1709.0533523899,"imp":4.0488895491,"elo":1626.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1192270005,"x_train":1709.0533523899,"imp":3.3038411005,"elo":1739.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.620625058,"x_train":20.2378618082,"imp":10.7664752895,"elo":1260.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"x_train":46.6209023207,"imp":5.5283691171,"elo":1574.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.4972730809,"x_train":4786.5989331939,"imp":4.6110596122,"elo":1663.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":239.3040135069,"x_train":4786.5989331939,"imp":4.1688637663,"elo":1715.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"x_train":2.0755087942,"imp":5.127593625,"elo":1597.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4760486585,"x_train":39.4205185389,"imp":1.5572886698,"elo":2004.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1222431699,"x_train":1.3204095155,"imp":3.8749858511,"elo":1674.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1324525523,"x_train":9.1928711765,"imp":9.3551363844,"elo":1273.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1814320495,"x_train":3013.8595115999,"imp":7.7301423167,"elo":1369.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.0923554564,"x_train":3013.8595115999,"imp":6.8194097728,"elo":1442.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.696318366,"x_train":8.5238425595,"imp":3.9399238497,"elo":1740.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"x_train":3.65846503,"imp":2.325985518,"elo":1791.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3129231924,"x_train":2.8034680169,"imp":11.0039225884,"elo":1182.9,"imputed":true,"imputed_pct":46.1538461538,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.451160218,"x_train":4223.8673583406,"imp":9.635148448,"elo":1231.6,"imputed":true,"imputed_pct":46.1538461538,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":27.5427954877,"x_train":4223.8673583406,"imp":8.5900286965,"elo":1317.4,"imputed":true,"imputed_pct":46.1538461538,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9291777504,"x_train":323.5247663683,"imp":26.5470868186,"elo":843.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.23600059,"x_train":28729.7373655693,"imp":23.6177601403,"elo":948.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.9007182558,"x_train":28729.7373655693,"imp":23.4531065734,"elo":961.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0752279828,"x_train":0.9081877562,"imp":7.1861494306,"elo":1390.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0796563802,"x_train":20.4886632638,"imp":15.0373165529,"elo":1027.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0968974085,"x_train":4608.594420269,"imp":11.6709471148,"elo":1177.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.232537078,"x_train":4608.594420269,"imp":10.8925056509,"elo":1229.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.242470131,"x_train":2.2442581404,"imp":11.9809667662,"elo":1181.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3882893337,"x_train":834.9300717203,"imp":9.0340846296,"elo":1379.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.6142653492,"x_train":834.9300717203,"imp":8.8886521695,"elo":1399.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":81.6126739219,"x_train":357.0213835173,"imp":18.8962293987,"elo":872.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":59.0992991738,"x_train":12685.0815501834,"imp":9.2221582665,"elo":1325.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":321.7409586602,"x_train":12685.0815501834,"imp":9.2665052414,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7424024527,"x_train":2.4487016432,"imp":13.6861569247,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0990665936,"x_train":714.4999525058,"imp":8.0801831737,"elo":1375.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.3755494169,"x_train":714.4999525058,"imp":7.4176531706,"elo":1457.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_all/tuning_trajectories_explorer.html
index 9cf9c5e6f5f20ec00fc068adaee20b62f1fa68a9..b4aeaf1897c977fafe2f5a778f4ade3a52185c6c 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":8.7329235239,"imp":9.5415883502,"elo":1382.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":21.046557619,"imp":9.052424581,"elo":1416.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":52.5864800365,"imp":8.5280669311,"elo":1458.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":107.3625813458,"imp":8.2811460153,"elo":1473.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":273.3208727607,"imp":8.0370843646,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":552.7941001427,"imp":7.9794953002,"elo":1494.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1107.2542254236,"imp":7.9405455656,"elo":1500.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1651.763231487,"imp":7.9312142319,"elo":1501.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":2229.2144624726,"imp":7.9154003667,"elo":1503.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.6789664191,"imp":12.5036926258,"elo":1224.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.4383850203,"imp":12.3857227824,"elo":1231.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.0701848574,"imp":12.1997093691,"elo":1251.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":20.464525374,"imp":11.975821331,"elo":1273.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":51.9588660046,"imp":11.717285953,"elo":1295.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":104.2658709225,"imp":11.5340310349,"elo":1311.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":208.787997457,"imp":11.3634298481,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":313.2556024507,"imp":11.2852117883,"elo":1328.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":420.0975247537,"imp":11.2407392686,"elo":1334.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.8692134542,"imp":15.8782160518,"elo":1046.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":17.0945668224,"imp":15.5100688048,"elo":1068.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":52.1405202142,"imp":15.1506225175,"elo":1099.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":97.8428503791,"imp":14.9368902307,"elo":1117.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":240.4577474664,"imp":14.647861129,"elo":1133.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":484.5851613774,"imp":14.5613896114,"elo":1141.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":933.0688585006,"imp":14.4906006741,"elo":1148.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1378.9985712807,"imp":14.4579611505,"elo":1153.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1865.2663509929,"imp":14.4339245001,"elo":1157.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.5445212109,"imp":15.5439575949,"elo":1055.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1259596548,"imp":14.983644089,"elo":1100.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.7721912257,"imp":14.1482425125,"elo":1157.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":13.288092342,"imp":13.7395918191,"elo":1180.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":32.8419339595,"imp":13.1240234841,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":65.4938634865,"imp":12.9969229612,"elo":1225.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":130.8369715186,"imp":12.9208860623,"elo":1230.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":196.2244804971,"imp":12.8961762886,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":263.0379305179,"imp":12.8848736462,"elo":1236.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.6039043157,"imp":19.9319675262,"elo":862.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.1690100082,"imp":19.6669845016,"elo":873.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":13.4266251032,"imp":18.8188902497,"elo":907.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":27.320730914,"imp":18.3312991755,"elo":929.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":66.1669393228,"imp":17.3656780816,"elo":964.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":132.6885087308,"imp":16.7925171587,"elo":982.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":267.8675184086,"imp":16.1260392716,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":402.5490610457,"imp":15.651950622,"elo":1023.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":540.1381360311,"imp":15.2842935381,"elo":1039.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.1108344816,"imp":11.3305865257,"elo":1203.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":5.9546664661,"imp":10.7626486133,"elo":1247.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":16.8872918174,"imp":9.8079172803,"elo":1333.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":32.7521125678,"imp":9.3177639193,"elo":1382.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":84.7959741874,"imp":8.7914270214,"elo":1446.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":173.2180148137,"imp":8.6327898415,"elo":1470.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":347.0816109705,"imp":8.4940484014,"elo":1487.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":519.953899606,"imp":8.4259537033,"elo":1495.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":700.1867028829,"imp":8.372106182,"elo":1502.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":74.6790742113,"imp":10.3100013499,"elo":1264.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":71.0611073496,"imp":10.2162762354,"elo":1269.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":15.5008512826,"imp":10.7059081481,"elo":1281.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":40.5171687704,"imp":9.9795507524,"elo":1327.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":107.503248256,"imp":8.8047615949,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":200.5362465049,"imp":8.2787850335,"elo":1458.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":460.9985605524,"imp":7.6960903275,"elo":1507.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":947.159574971,"imp":7.5052846299,"elo":1524.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1857.8046474259,"imp":7.3358223067,"elo":1538.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":2770.8108835678,"imp":7.257115895,"elo":1545.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":3779.7025911156,"imp":7.2380750646,"elo":1549.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":6.9283541331,"imp":4.4505981083,"elo":1696.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":11.5842534555,"imp":3.3541042043,"elo":1778.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":27.322832323,"imp":17.6914554913,"elo":951.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":54.4506451141,"imp":15.2951073539,"elo":1059.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":135.4944823225,"imp":13.2967858437,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":162.3794942955,"imp":13.1639160845,"elo":1173.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.5271703851,"imp":15.8898163328,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.6581130708,"imp":15.2491669429,"elo":1057.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":13.0531670853,"imp":14.7645678344,"elo":1091.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":26.0784291852,"imp":14.3293136053,"elo":1122.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":63.4089966233,"imp":14.0804924274,"elo":1142.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":127.7253484811,"imp":13.9970695471,"elo":1152.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":254.9677280133,"imp":13.9444501915,"elo":1161.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":383.936400338,"imp":13.9265769145,"elo":1165.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":515.7544728433,"imp":13.9175588028,"elo":1167.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":8.9020711389,"imp":10.4477296125,"elo":1228.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.3173471418,"imp":8.6146299817,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":77.6709970732,"imp":6.6995678981,"elo":1489.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":189.6529598329,"imp":6.0358973507,"elo":1552.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":487.3244648114,"imp":5.5456040675,"elo":1607.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":988.6461264381,"imp":5.2966277928,"elo":1633.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1985.9304497408,"imp":5.132364452,"elo":1652.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2976.2042051285,"imp":5.0408622888,"elo":1667.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3995.0078687545,"imp":4.9778613777,"elo":1679.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":7.0435968115,"imp":5.5285268429,"elo":1545.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.015104703,"imp":5.1272895786,"elo":1568.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":41.0105415705,"imp":4.4279944122,"elo":1611.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":83.3999540139,"imp":4.0445187551,"elo":1656.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":210.1679172678,"imp":3.6770677837,"elo":1711.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":423.4200904906,"imp":3.5017457878,"elo":1737.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":852.1950752527,"imp":3.3961930438,"elo":1754.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1277.5788085201,"imp":3.3440764638,"elo":1764.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1709.0533523899,"imp":3.3088373688,"elo":1770.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":20.2378618082,"imp":10.7703527232,"elo":1278.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.6209023207,"imp":5.5330374627,"elo":1584.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.0936255942,"imp":5.13258681,"elo":1618.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":132.9785409439,"imp":4.7917052234,"elo":1656.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":240.939305723,"imp":4.5484747413,"elo":1684.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":589.3561296929,"imp":4.3381835956,"elo":1710.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1183.6501039466,"imp":4.2414216133,"elo":1724.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2376.2279314343,"imp":4.1883513242,"elo":1734.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3574.1919132683,"imp":4.1762746047,"elo":1735.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4786.5989331939,"imp":4.17309193,"elo":1736.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0755087942,"imp":5.1321082718,"elo":1611.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":39.4205185389,"imp":1.561914241,"elo":2036.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.3204095155,"imp":3.8795600526,"elo":1691.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.1928711765,"imp":9.3594275294,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":22.5165548441,"imp":8.9510801796,"elo":1308.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":69.1089159979,"imp":8.6375112137,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":146.4525878247,"imp":8.1382527442,"elo":1373.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":371.1147116855,"imp":7.6770509463,"elo":1407.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":743.1660518453,"imp":7.390413591,"elo":1430.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1486.16387375,"imp":7.1830506209,"elo":1447.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2240.0758838289,"imp":6.8778441115,"elo":1458.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3013.8595115999,"imp":6.8232160269,"elo":1467.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":8.5238425595,"imp":3.9444773837,"elo":1770.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.65846503,"imp":2.3376619879,"elo":1805.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2.8034680169,"imp":11.0084903049,"elo":1188.2,"imputed":true,"imputed_pct":46.1538461538,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":28.6961544367,"imp":10.3084833707,"elo":1214.9,"imputed":true,"imputed_pct":46.1538461538,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":88.2106097476,"imp":9.6821280327,"elo":1249.8,"imputed":true,"imputed_pct":46.1538461538,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":197.7484589912,"imp":9.4111419316,"elo":1269.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":491.4205479278,"imp":9.0260600487,"elo":1298.2,"imputed":true,"imputed_pct":46.1538461538,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1003.8148779411,"imp":8.8434607207,"elo":1311.9,"imputed":true,"imputed_pct":46.1538461538,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2033.7269652077,"imp":8.6871488387,"elo":1323.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3090.6047962378,"imp":8.6365262203,"elo":1327.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4223.8673583406,"imp":8.594801241,"elo":1329.5,"imputed":true,"imputed_pct":46.1538461538,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":323.5247663683,"imp":26.5499230448,"elo":843.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":458.087256476,"imp":26.4570546319,"elo":860.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":775.9282960298,"imp":26.2535366776,"elo":874.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1467.4534767341,"imp":25.7863425152,"elo":900.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3523.6508469471,"imp":25.0583934582,"elo":923.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":7117.6518843327,"imp":24.3662102546,"elo":941.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":14227.2413776887,"imp":23.9341587962,"elo":955.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":21385.328840253,"imp":23.6594587218,"elo":967.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":28729.7373655693,"imp":23.4561231832,"elo":981.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.9081877562,"imp":7.1908302664,"elo":1404.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":20.4886632638,"imp":15.0412606211,"elo":1039.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":41.7266327108,"imp":14.4129095555,"elo":1067.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":106.9107806776,"imp":13.4060491315,"elo":1114.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":227.2145545781,"imp":12.8789217848,"elo":1144.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":570.0093805222,"imp":12.1576678872,"elo":1182.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1173.9046077839,"imp":11.7068687677,"elo":1210.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2290.0065707371,"imp":11.2432796352,"elo":1230.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3456.708959715,"imp":11.0328367377,"elo":1240.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4608.594420269,"imp":10.8963815769,"elo":1249.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.2442581404,"imp":11.9846414175,"elo":1186.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":8.0761527298,"imp":10.5472618454,"elo":1269.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":20.7267313474,"imp":9.856030755,"elo":1325.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":40.9776832255,"imp":9.496296632,"elo":1358.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":103.9269515729,"imp":9.2184819504,"elo":1384.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":207.504207675,"imp":9.0898342445,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":414.7237545011,"imp":8.994063751,"elo":1408.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":622.7467847485,"imp":8.9291999843,"elo":1415.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":834.9300717203,"imp":8.892633565,"elo":1421.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":357.0213835173,"imp":18.8993993134,"elo":883.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":826.6192885105,"imp":14.7816911953,"elo":1029.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2323.5909795411,"imp":15.7127542665,"elo":1203.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4948.4562156001,"imp":14.1768212258,"elo":1295.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12120.2560802151,"imp":9.3058102035,"elo":1443.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12685.0815501834,"imp":9.2711150315,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.4487016432,"imp":13.6898369866,"elo":1156.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.9610558826,"imp":12.3708492404,"elo":1219.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.0080432107,"imp":10.6408050397,"elo":1292.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":34.8874663118,"imp":9.3227972422,"elo":1359.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.7983697442,"imp":8.2038969548,"elo":1417.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":177.9391500768,"imp":7.8036442088,"elo":1445.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":356.5372096884,"imp":7.559794805,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":533.621477288,"imp":7.46789742,"elo":1471.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":714.4999525058,"imp":7.4219334723,"elo":1478.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.016935661,"x_train":8.7329235239,"imp":9.5415883502,"elo":1382.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0753245982,"x_train":21.046557619,"imp":9.052424581,"elo":1416.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1609118157,"x_train":52.5864800365,"imp":8.5280669311,"elo":1458.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2251450506,"x_train":107.3625813458,"imp":8.2811460153,"elo":1473.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2353196357,"x_train":273.3208727607,"imp":8.0370843646,"elo":1488.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2694149278,"x_train":552.7941001427,"imp":7.9794953002,"elo":1494.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3197063318,"x_train":1107.2542254236,"imp":7.9405455656,"elo":1500.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3499660795,"x_train":1651.763231487,"imp":7.9312142319,"elo":1501.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3802116897,"x_train":2229.2144624726,"imp":7.9154003667,"elo":1503.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0297627373,"x_train":2.6789664191,"imp":12.5036926258,"elo":1224.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0556929991,"x_train":4.4383850203,"imp":12.3857227824,"elo":1231.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0988043477,"x_train":10.0701848574,"imp":12.1997093691,"elo":1251.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1363240631,"x_train":20.464525374,"imp":11.975821331,"elo":1273.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1837219602,"x_train":51.9588660046,"imp":11.717285953,"elo":1295.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2043019279,"x_train":104.2658709225,"imp":11.5340310349,"elo":1311.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2249081343,"x_train":208.787997457,"imp":11.3634298481,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2311065475,"x_train":313.2556024507,"imp":11.2852117883,"elo":1328.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"x_train":420.0975247537,"imp":11.2407392686,"elo":1334.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0108193564,"x_train":7.8692134542,"imp":15.8782160518,"elo":1046.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0214757223,"x_train":17.0945668224,"imp":15.5100688048,"elo":1068.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0403341023,"x_train":52.1405202142,"imp":15.1506225175,"elo":1099.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0568329264,"x_train":97.8428503791,"imp":14.9368902307,"elo":1117.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0777112708,"x_train":240.4577474664,"imp":14.647861129,"elo":1133.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0931559756,"x_train":484.5851613774,"imp":14.5613896114,"elo":1141.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1063228048,"x_train":933.0688585006,"imp":14.4906006741,"elo":1148.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1134174096,"x_train":1378.9985712807,"imp":14.4579611505,"elo":1153.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.116575014,"x_train":1865.2663509929,"imp":14.4339245001,"elo":1157.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.24026221,"x_train":2.5445212109,"imp":15.5439575949,"elo":1055.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2792114359,"x_train":3.1259596548,"imp":14.983644089,"elo":1100.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3158110254,"x_train":6.7721912257,"imp":14.1482425125,"elo":1157.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3028704861,"x_train":13.288092342,"imp":13.7395918191,"elo":1180.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4105206818,"x_train":32.8419339595,"imp":13.1240234841,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5221999466,"x_train":65.4938634865,"imp":12.9969229612,"elo":1225.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.660487887,"x_train":130.8369715186,"imp":12.9208860623,"elo":1230.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7291312705,"x_train":196.2244804971,"imp":12.8961762886,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8077239936,"x_train":263.0379305179,"imp":12.8848736462,"elo":1236.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3915661558,"x_train":2.6039043157,"imp":19.9319675262,"elo":862.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6643959346,"x_train":5.1690100082,"imp":19.6669845016,"elo":873.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1863927783,"x_train":13.4266251032,"imp":18.8188902497,"elo":907.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6398576272,"x_train":27.320730914,"imp":18.3312991755,"elo":929.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0095610562,"x_train":66.1669393228,"imp":17.3656780816,"elo":964.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4439102689,"x_train":132.6885087308,"imp":16.7925171587,"elo":982.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1002617128,"x_train":267.8675184086,"imp":16.1260392716,"elo":1003.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8287572918,"x_train":402.5490610457,"imp":15.651950622,"elo":1023.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.6723835679,"x_train":540.1381360311,"imp":15.2842935381,"elo":1039.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2745146333,"x_train":2.1108344816,"imp":11.3305865257,"elo":1203.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4016545585,"x_train":5.9546664661,"imp":10.7626486133,"elo":1247.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.8377517363,"x_train":16.8872918174,"imp":9.8079172803,"elo":1333.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.8851780262,"x_train":32.7521125678,"imp":9.3177639193,"elo":1382.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.8791663559,"x_train":84.7959741874,"imp":8.7914270214,"elo":1446.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.5304245512,"x_train":173.2180148137,"imp":8.6327898415,"elo":1470.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.6947623735,"x_train":347.0816109705,"imp":8.4940484014,"elo":1487.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":9.0305331912,"x_train":519.953899606,"imp":8.4259537033,"elo":1495.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":9.3218182128,"x_train":700.1867028829,"imp":8.372106182,"elo":1502.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":19.7582527317,"x_train":74.6790742113,"imp":10.3100013499,"elo":1264.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":1.848279835,"x_train":71.0611073496,"imp":10.2162762354,"elo":1269.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2983006451,"x_train":15.5008512826,"imp":10.7059081481,"elo":1281.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7075904089,"x_train":40.5171687704,"imp":9.9795507524,"elo":1327.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.5287946275,"x_train":107.503248256,"imp":8.8047615949,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2651751376,"x_train":200.5362465049,"imp":8.2787850335,"elo":1458.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.3092160521,"x_train":460.9985605524,"imp":7.6960903275,"elo":1507.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.6686809631,"x_train":947.159574971,"imp":7.5052846299,"elo":1524.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2095781123,"x_train":1857.8046474259,"imp":7.3358223067,"elo":1538.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.1496874361,"x_train":2770.8108835678,"imp":7.257115895,"elo":1545.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.6904227381,"x_train":3779.7025911156,"imp":7.2380750646,"elo":1549.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":1.3890423048,"x_train":6.9283541331,"imp":4.4505981083,"elo":1696.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0226031244,"x_train":11.5842534555,"imp":3.3541042043,"elo":1778.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.020234932,"x_train":27.322832323,"imp":17.6914554913,"elo":951.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1408989609,"x_train":54.4506451141,"imp":15.2951073539,"elo":1059.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3464615835,"x_train":135.4944823225,"imp":13.2967858437,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3627650406,"x_train":162.3794942955,"imp":13.1639160845,"elo":1173.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0621551845,"x_train":0.5271703851,"imp":15.8898163328,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1542795056,"x_train":3.6581130708,"imp":15.2491669429,"elo":1057.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2520439254,"x_train":13.0531670853,"imp":14.7645678344,"elo":1091.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3406296664,"x_train":26.0784291852,"imp":14.3293136053,"elo":1122.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5181009628,"x_train":63.4089966233,"imp":14.0804924274,"elo":1142.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5887365969,"x_train":127.7253484811,"imp":13.9970695471,"elo":1152.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7109300332,"x_train":254.9677280133,"imp":13.9444501915,"elo":1161.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7595438385,"x_train":383.936400338,"imp":13.9265769145,"elo":1165.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7784693874,"x_train":515.7544728433,"imp":13.9175588028,"elo":1167.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.644117959,"x_train":8.9020711389,"imp":10.4477296125,"elo":1228.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.1298601441,"x_train":24.3173471418,"imp":8.6146299817,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.3924065732,"x_train":77.6709970732,"imp":6.6995678981,"elo":1489.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.0155793542,"x_train":189.6529598329,"imp":6.0358973507,"elo":1552.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.2080767736,"x_train":487.3244648114,"imp":5.5456040675,"elo":1607.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.640857572,"x_train":988.6461264381,"imp":5.2966277928,"elo":1633.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.6390543881,"x_train":1985.9304497408,"imp":5.132364452,"elo":1652.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.478029918,"x_train":2976.2042051285,"imp":5.0408622888,"elo":1667.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.0840513338,"x_train":3995.0078687545,"imp":4.9778613777,"elo":1679.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"x_train":7.0435968115,"imp":5.5285268429,"elo":1545.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4031929556,"x_train":16.015104703,"imp":5.1272895786,"elo":1568.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8742272233,"x_train":41.0105415705,"imp":4.4279944122,"elo":1611.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.8222925994,"x_train":83.3999540139,"imp":4.0445187551,"elo":1656.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7321639458,"x_train":210.1679172678,"imp":3.6770677837,"elo":1711.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.7691704179,"x_train":423.4200904906,"imp":3.5017457878,"elo":1737.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.1518842984,"x_train":852.1950752527,"imp":3.3961930438,"elo":1754.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8504991295,"x_train":1277.5788085201,"imp":3.3440764638,"elo":1764.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1651715053,"x_train":1709.0533523899,"imp":3.3088373688,"elo":1770.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.620625058,"x_train":20.2378618082,"imp":10.7703527232,"elo":1278.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"x_train":46.6209023207,"imp":5.5330374627,"elo":1584.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":55.2611832093,"x_train":70.0936255942,"imp":5.13258681,"elo":1618.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":81.5128039491,"x_train":132.9785409439,"imp":4.7917052234,"elo":1656.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":105.6461170065,"x_train":240.939305723,"imp":4.5484747413,"elo":1684.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":122.0577162857,"x_train":589.3561296929,"imp":4.3381835956,"elo":1710.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":163.7123037519,"x_train":1183.6501039466,"imp":4.2414216133,"elo":1724.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":204.6580402482,"x_train":2376.2279314343,"imp":4.1883513242,"elo":1734.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":216.1656039799,"x_train":3574.1919132683,"imp":4.1762746047,"elo":1735.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":239.3040135069,"x_train":4786.5989331939,"imp":4.17309193,"elo":1736.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1824194029,"x_train":2.0755087942,"imp":5.1321082718,"elo":1611.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.4760486585,"x_train":39.4205185389,"imp":1.561914241,"elo":2036.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1222431699,"x_train":1.3204095155,"imp":3.8795600526,"elo":1691.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1324525523,"x_train":9.1928711765,"imp":9.3594275294,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2760220834,"x_train":22.5165548441,"imp":8.9510801796,"elo":1308.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.5604210779,"x_train":69.1089159979,"imp":8.6375112137,"elo":1339.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.7916548768,"x_train":146.4525878247,"imp":8.1382527442,"elo":1373.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.0818885965,"x_train":371.1147116855,"imp":7.6770509463,"elo":1407.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.3749946791,"x_train":743.1660518453,"imp":7.390413591,"elo":1430.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.6910948789,"x_train":1486.16387375,"imp":7.1830506209,"elo":1447.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.9072725195,"x_train":2240.0758838289,"imp":6.8778441115,"elo":1458.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.0923554564,"x_train":3013.8595115999,"imp":6.8232160269,"elo":1467.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.696318366,"x_train":8.5238425595,"imp":3.9444773837,"elo":1770.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3988194969,"x_train":3.65846503,"imp":2.3376619879,"elo":1805.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3129231924,"x_train":2.8034680169,"imp":11.0084903049,"elo":1188.2,"imputed":true,"imputed_pct":46.1538461538,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":5.0129796195,"x_train":28.6961544367,"imp":10.3084833707,"elo":1214.9,"imputed":true,"imputed_pct":46.1538461538,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.0008361411,"x_train":88.2106097476,"imp":9.6821280327,"elo":1249.8,"imputed":true,"imputed_pct":46.1538461538,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":16.1155274179,"x_train":197.7484589912,"imp":9.4111419316,"elo":1269.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.5762988303,"x_train":491.4205479278,"imp":9.0260600487,"elo":1298.2,"imputed":true,"imputed_pct":46.1538461538,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":20.7587025099,"x_train":1003.8148779411,"imp":8.8434607207,"elo":1311.9,"imputed":true,"imputed_pct":46.1538461538,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":24.7353926128,"x_train":2033.7269652077,"imp":8.6871488387,"elo":1323.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":26.1184500506,"x_train":3090.6047962378,"imp":8.6365262203,"elo":1327.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":27.5427954877,"x_train":4223.8673583406,"imp":8.594801241,"elo":1329.5,"imputed":true,"imputed_pct":46.1538461538,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9291777504,"x_train":323.5247663683,"imp":26.5499230448,"elo":843.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.9691116129,"x_train":458.087256476,"imp":26.4570546319,"elo":860.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.4073606358,"x_train":775.9282960298,"imp":26.2535366776,"elo":874.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.6930338556,"x_train":1467.4534767341,"imp":25.7863425152,"elo":900.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.0890319964,"x_train":3523.6508469471,"imp":25.0583934582,"elo":923.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.4354221138,"x_train":7117.6518843327,"imp":24.3662102546,"elo":941.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.0213237631,"x_train":14227.2413776887,"imp":23.9341587962,"elo":955.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3998340801,"x_train":21385.328840253,"imp":23.6594587218,"elo":967.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.9007182558,"x_train":28729.7373655693,"imp":23.4561231832,"elo":981.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0752279828,"x_train":0.9081877562,"imp":7.1908302664,"elo":1404.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0796563802,"x_train":20.4886632638,"imp":15.0412606211,"elo":1039.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1719706556,"x_train":41.7266327108,"imp":14.4129095555,"elo":1067.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.408364249,"x_train":106.9107806776,"imp":13.4060491315,"elo":1114.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5543275658,"x_train":227.2145545781,"imp":12.8789217848,"elo":1144.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7496106434,"x_train":570.0093805222,"imp":12.1576678872,"elo":1182.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9621284433,"x_train":1173.9046077839,"imp":11.7068687677,"elo":1210.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.075322816,"x_train":2290.0065707371,"imp":11.2432796352,"elo":1230.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1499754889,"x_train":3456.708959715,"imp":11.0328367377,"elo":1240.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.232537078,"x_train":4608.594420269,"imp":10.8963815769,"elo":1249.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.242470131,"x_train":2.2442581404,"imp":11.9846414175,"elo":1186.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6395763256,"x_train":8.0761527298,"imp":10.5472618454,"elo":1269.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1210539326,"x_train":20.7267313474,"imp":9.856030755,"elo":1325.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2386472307,"x_train":40.9776832255,"imp":9.496296632,"elo":1358.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8103396721,"x_train":103.9269515729,"imp":9.2184819504,"elo":1384.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.1270291973,"x_train":207.504207675,"imp":9.0898342445,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3814568805,"x_train":414.7237545011,"imp":8.994063751,"elo":1408.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.578940555,"x_train":622.7467847485,"imp":8.9291999843,"elo":1415.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.6107066669,"x_train":834.9300717203,"imp":8.892633565,"elo":1421.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":81.6126739219,"x_train":357.0213835173,"imp":18.8993993134,"elo":883.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":133.585345468,"x_train":826.6192885105,"imp":14.7816911953,"elo":1029.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":216.0194313928,"x_train":2323.5909795411,"imp":15.7127542665,"elo":1203.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":301.6106544607,"x_train":4948.4562156001,"imp":14.1768212258,"elo":1295.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":315.3980497637,"x_train":12120.2560802151,"imp":9.3058102035,"elo":1443.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":321.7409586602,"x_train":12685.0815501834,"imp":9.2711150315,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.7424024527,"x_train":2.4487016432,"imp":13.6898369866,"elo":1156.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8510374477,"x_train":5.9610558826,"imp":12.3708492404,"elo":1219.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.0266701319,"x_train":17.0080432107,"imp":10.6408050397,"elo":1292.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.2520438362,"x_train":34.8874663118,"imp":9.3227972422,"elo":1359.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.5268809255,"x_train":88.7983697442,"imp":8.2038969548,"elo":1417.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.802084336,"x_train":177.9391500768,"imp":7.8036442088,"elo":1445.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.4216240621,"x_train":356.5372096884,"imp":7.559794805,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.3580011157,"x_train":533.621477288,"imp":7.46789742,"elo":1471.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.3755494169,"x_train":714.4999525058,"imp":7.4219334723,"elo":1478.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..b91be88b9f15b0b75d9374f111658c971cc82ff8
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_regression/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "Nori-30M (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "Nori (default)", "TabDPT (tuned + ensembled)", "RealMLP (tuned + ensembled)", "TabDPT (tuned)", "RealTabPFN-2.5 (tuned)", "TabDPT (default)", "RealTabPFN-2.5 (default)", "TabDPT-Turbo (default)", "RealMLP (tuned)", "AutoGluon 1.4 (best, 4h)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "LightGBM (tuned + ensembled)", "TabM (tuned + ensembled)", "iLTM (tuned + ensembled)", "TabPFNv2 (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "xRFM (tuned + ensembled)", "TabSwift (default)", "XGBoost (tuned + ensembled)", "CatBoost (default)", "LightGBM (tuned)", "Mitra (default)", "XGBoost (tuned)", "SAP-RPT-OSS (default)", "TabM (tuned)", "xRFM (tuned)", "ChimeraBoost (tuned + ensembled)", "TabPFNv2 (tuned)", "iLTM (tuned)", "TabPFNv2 (default)", "ChimeraBoost (tuned)", "TabM (default)", "ModernNCA (tuned)", "RealMLP (default)", "ExtraTrees (tuned + ensembled)", "ChimeraBoost (default)", "TorchMLP (tuned + ensembled)", "ModernNCA (default)", "ExtraTrees (tuned)", "LimiX (default)", "EBM (tuned + ensembled)", "LightGBM (default)", "TorchMLP (tuned)", "EBM (tuned)", "RandomForest (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "XGBoost (default)", "RandomForest (tuned)", "xRFM (default)", "EBM (default)", "TabSTAR (tuned + ensembled)", "PerpetualBooster (tuned)", "TabSTAR (tuned)", "ExtraTrees (default)", "FastaiMLP (tuned + ensembled)", "TorchMLP (default)", "FastaiMLP (tuned)", "RandomForest (default)", "TabSTAR (default)", "PerpetualBooster (default)", "KNN (tuned + ensembled)", "FastaiMLP (default)", "iLTM (default)", "KNN (tuned)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 0.6126126050949097, 0.6981981992721558, 0.7297297120094299, 0.815315306186676, 0.7432432174682617, 0.792792797088623, 0.7972972989082336, 0.837837815284729, 0.9009009003639221, 0.9054054021835327, 0.815315306186676, 0.8873873949050903, 0.9324324131011963, 0.954954981803894, 0.9459459185600281, 0.9594594836235046, 0.9639639854431152, 0.977477490901947, 0.9729729890823364, 0.954954981803894, 0.9639639854431152, 0.8738738894462585, 0.9594594836235046, 0.9504504799842834, 0.8963963985443115, 0.9639639854431152, 0.9684684872627258, 0.977477490901947, 0.9459459185600281, 0.9729729890823364, 0.9324324131011963, 0.9684684872627258, 0.9684684872627258, 0.9729729890823364, 0.9054054021835327, 0.977477490901947, 0.9684684872627258, 0.9684684872627258, 0.977477490901947, 0.9909909963607788, 0.9909909963607788, 0.9729729890823364, 0.9954954981803894, 0.9909909963607788, 0.9909909963607788, 0.977477490901947, 0.9864864945411682, 0.9729729890823364, 1.0, 0.9954954981803894, 0.9819819927215576, 0.9954954981803894, 0.9909909963607788, 1.0, 0.9954954981803894, 0.9909909963607788, 0.9864864945411682, 0.9234234094619751, 0.9954954981803894, 0.9279279112815857, 0.9819819927215576, 1.0, 1.0, 1.0, 0.9954954981803894, 0.9414414167404175, 1.0, 0.9864864945411682, 1.0, 1.0, 0.9954954981803894, 0.9909909963607788, 1.0, 1.0, 1.0], [0.38738739490509033, NaN, 0.477477490901947, 0.6531531810760498, 0.6801801919937134, 0.545045018196106, 0.6216216087341309, 0.7297297120094299, 0.630630612373352, 0.7207207083702087, 0.7162162065505981, 0.6981981992721558, 0.7387387156486511, 0.792792797088623, 0.7747747898101807, 0.792792797088623, 0.8288288116455078, 0.8918918967247009, 0.9189189076423645, 0.9144144058227539, 0.9234234094619751, 0.9009009003639221, 0.8963963985443115, 0.792792797088623, 0.8783783912658691, 0.869369387626648, 0.9279279112815857, 0.9189189076423645, 0.9234234094619751, 0.954954981803894, 0.9369369149208069, 0.8918918967247009, 0.9459459185600281, 0.9279279112815857, 0.9414414167404175, 0.9459459185600281, 0.977477490901947, 0.9459459185600281, 0.954954981803894, 0.9684684872627258, 0.8963963985443115, 0.9639639854431152, 0.9594594836235046, 0.9729729890823364, 0.9684684872627258, 0.9504504799842834, 0.9459459185600281, 0.9054054021835327, 0.977477490901947, 0.9909909963607788, 0.9819819927215576, 0.9864864945411682, 0.9819819927215576, 0.9864864945411682, 0.9729729890823364, 0.9819819927215576, 0.9684684872627258, 0.9909909963607788, 0.9144144058227539, 0.9909909963607788, 0.9234234094619751, 0.9639639854431152, 0.9954954981803894, 1.0, 1.0, 0.9864864945411682, 0.9279279112815857, 1.0, 0.9819819927215576, 1.0, 0.9954954981803894, 0.9909909963607788, 0.9819819927215576, 1.0, 1.0, 1.0], [0.30180180072784424, 0.522522509098053, NaN, 0.6441441178321838, 0.6396396160125732, 0.6126126050949097, 0.6216216087341309, 0.6396396160125732, 0.6486486196517944, 0.792792797088623, 0.6891891956329346, 0.7522522807121277, 0.7882882952690125, 0.8783783912658691, 0.8783783912658691, 0.8243243098258972, 0.8423423171043396, 0.8558558821678162, 0.8603603839874268, 0.8873873949050903, 0.8918918967247009, 0.8963963985443115, 0.8648648858070374, 0.8558558821678162, 0.8783783912658691, 0.9369369149208069, 0.8873873949050903, 0.8738738894462585, 0.9099099040031433, 0.8333333134651184, 0.8828828930854797, 0.8738738894462585, 0.9234234094619751, 0.9279279112815857, 0.8783783912658691, 0.9279279112815857, 0.9729729890823364, 0.9639639854431152, 0.8828828930854797, 0.9369369149208069, 0.8963963985443115, 0.9684684872627258, 0.9459459185600281, 0.9189189076423645, 0.9729729890823364, 0.9684684872627258, 0.954954981803894, 0.9279279112815857, 0.9009009003639221, 0.9909909963607788, 0.977477490901947, 0.9144144058227539, 0.977477490901947, 0.9504504799842834, 0.9504504799842834, 0.977477490901947, 0.9639639854431152, 0.9414414167404175, 0.8603603839874268, 0.9819819927215576, 0.8783783912658691, 0.9684684872627258, 1.0, 1.0, 1.0, 0.9954954981803894, 0.9279279112815857, 1.0, 0.9684684872627258, 1.0, 1.0, 0.9819819927215576, 0.9909909963607788, 1.0, 1.0, 1.0], [0.2702702581882477, 0.3468468487262726, 0.3558558523654938, NaN, 0.5630630850791931, 0.44144144654273987, 0.45945945382118225, 0.522522509098053, 0.5990990996360779, 0.7297297120094299, 0.6216216087341309, 0.6081081032752991, 0.6891891956329346, 0.7297297120094299, 0.7432432174682617, 0.8108108043670654, 0.8603603839874268, 0.8648648858070374, 0.8873873949050903, 0.8873873949050903, 0.9099099040031433, 0.9054054021835327, 0.7432432174682617, 0.8018018007278442, 0.8648648858070374, 0.8648648858070374, 0.9234234094619751, 0.9009009003639221, 0.9324324131011963, 0.8603603839874268, 0.9369369149208069, 0.7972972989082336, 0.9234234094619751, 0.9099099040031433, 0.9234234094619751, 0.8783783912658691, 0.9414414167404175, 0.9234234094619751, 0.9414414167404175, 0.9414414167404175, 0.954954981803894, 0.977477490901947, 0.9369369149208069, 0.9459459185600281, 0.9864864945411682, 0.9639639854431152, 0.9414414167404175, 0.8603603839874268, 0.9639639854431152, 0.977477490901947, 0.9864864945411682, 0.9639639854431152, 0.9684684872627258, 0.9909909963607788, 0.9954954981803894, 0.9729729890823364, 0.9729729890823364, 0.954954981803894, 0.8963963985443115, 0.9954954981803894, 0.9009009003639221, 0.9504504799842834, 0.9954954981803894, 1.0, 1.0, 1.0, 0.9324324131011963, 1.0, 0.977477490901947, 1.0, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.18468467891216278, 0.3198198080062866, 0.36036035418510437, 0.4369369447231293, NaN, 0.38738739490509033, 0.47297295928001404, 0.47297295928001404, 0.5945945978164673, 0.6576576828956604, 0.6351351141929626, 0.6261261105537415, 0.6801801919937134, 0.792792797088623, 0.684684693813324, 0.8018018007278442, 0.8018018007278442, 0.8783783912658691, 0.9009009003639221, 0.8738738894462585, 0.9144144058227539, 0.8738738894462585, 0.7297297120094299, 0.8108108043670654, 0.8783783912658691, 0.8558558821678162, 0.9369369149208069, 0.9324324131011963, 0.9054054021835327, 0.8873873949050903, 0.9414414167404175, 0.8468468189239502, 0.9234234094619751, 0.9504504799842834, 0.9459459185600281, 0.8558558821678162, 0.9729729890823364, 0.8918918967247009, 0.9594594836235046, 0.9594594836235046, 0.9459459185600281, 0.9729729890823364, 0.9369369149208069, 0.9864864945411682, 0.9594594836235046, 0.9864864945411682, 0.9459459185600281, 0.9414414167404175, 0.9729729890823364, 0.9954954981803894, 0.9864864945411682, 0.9819819927215576, 0.9909909963607788, 0.9954954981803894, 1.0, 0.9909909963607788, 0.977477490901947, 0.9819819927215576, 0.9009009003639221, 0.9954954981803894, 0.9099099040031433, 0.9639639854431152, 0.9819819927215576, 1.0, 0.9909909963607788, 0.9954954981803894, 0.9369369149208069, 1.0, 0.9864864945411682, 0.9954954981803894, 1.0, 0.9864864945411682, 0.9909909963607788, 1.0, 1.0, 1.0], [0.2567567527294159, 0.45495495200157166, 0.38738739490509033, 0.5585585832595825, 0.6126126050949097, NaN, 0.5855855941772461, 0.5675675868988037, 0.684684693813324, 0.7522522807121277, 0.6936936974525452, 0.6981981992721558, 0.7207207083702087, 0.8333333134651184, 0.7837837934494019, 0.815315306186676, 0.8468468189239502, 0.8783783912658691, 0.8738738894462585, 0.8738738894462585, 0.9144144058227539, 0.8738738894462585, 0.7837837934494019, 0.8513513803482056, 0.8603603839874268, 0.8333333134651184, 0.8828828930854797, 0.8828828930854797, 0.8918918967247009, 0.8513513803482056, 0.8738738894462585, 0.9099099040031433, 0.9189189076423645, 0.8738738894462585, 0.869369387626648, 0.8828828930854797, 0.9009009003639221, 0.9369369149208069, 0.8738738894462585, 0.9234234094619751, 0.8963963985443115, 0.9459459185600281, 0.9189189076423645, 0.9099099040031433, 0.9144144058227539, 0.9009009003639221, 0.9144144058227539, 0.9324324131011963, 0.9099099040031433, 0.9504504799842834, 0.9054054021835327, 0.9054054021835327, 0.9369369149208069, 0.9324324131011963, 0.9189189076423645, 0.9459459185600281, 0.9639639854431152, 0.9414414167404175, 0.869369387626648, 0.9324324131011963, 0.8828828930854797, 0.9594594836235046, 0.9594594836235046, 0.9594594836235046, 0.9594594836235046, 0.9729729890823364, 0.9324324131011963, 0.9594594836235046, 0.9819819927215576, 0.9594594836235046, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.20720720291137695, 0.37837839126586914, 0.37837839126586914, 0.5405405163764954, 0.5270270109176636, 0.4144144058227539, NaN, 0.5360360145568848, 0.6081081032752991, 0.6981981992721558, 0.6351351141929626, 0.8198198080062866, 0.684684693813324, 0.8828828930854797, 0.7477477192878723, 0.7792792916297913, 0.7972972989082336, 0.8063063025474548, 0.8468468189239502, 0.8513513803482056, 0.9054054021835327, 0.837837815284729, 0.7837837934494019, 0.7882882952690125, 0.8468468189239502, 0.8603603839874268, 0.8918918967247009, 0.8918918967247009, 0.869369387626648, 0.8333333134651184, 0.8963963985443115, 0.815315306186676, 0.9324324131011963, 0.8873873949050903, 0.8963963985443115, 0.9234234094619751, 0.9234234094619751, 0.9144144058227539, 0.9324324131011963, 0.9594594836235046, 0.9279279112815857, 0.9504504799842834, 0.8828828930854797, 0.9729729890823364, 0.977477490901947, 0.9459459185600281, 0.8873873949050903, 0.9144144058227539, 0.954954981803894, 0.9864864945411682, 0.9819819927215576, 0.9684684872627258, 0.954954981803894, 0.9864864945411682, 0.9729729890823364, 0.9639639854431152, 0.9324324131011963, 0.9729729890823364, 0.8648648858070374, 0.9864864945411682, 0.8783783912658691, 0.954954981803894, 1.0, 1.0, 1.0, 0.9909909963607788, 0.9189189076423645, 1.0, 0.9684684872627258, 1.0, 1.0, 0.977477490901947, 0.9864864945411682, 1.0, 1.0, 1.0], [0.20270270109176636, 0.2702702581882477, 0.36036035418510437, 0.477477490901947, 0.5270270109176636, 0.4324324429035187, 0.46396395564079285, NaN, 0.5540540814399719, 0.5855855941772461, 0.5900900959968567, 0.545045018196106, 0.6351351141929626, 0.7657657861709595, 0.6891891956329346, 0.7477477192878723, 0.7522522807121277, 0.9189189076423645, 0.9144144058227539, 0.9099099040031433, 0.8423423171043396, 0.8513513803482056, 0.7837837934494019, 0.7612612843513489, 0.8423423171043396, 0.7702702879905701, 0.8963963985443115, 0.9144144058227539, 0.9144144058227539, 0.8963963985443115, 0.9099099040031433, 0.8468468189239502, 0.8963963985443115, 0.9279279112815857, 0.9279279112815857, 0.8828828930854797, 0.9819819927215576, 0.8963963985443115, 0.9369369149208069, 0.9594594836235046, 0.8288288116455078, 0.9369369149208069, 0.9504504799842834, 0.9459459185600281, 0.9639639854431152, 0.9144144058227539, 0.9504504799842834, 0.9054054021835327, 0.9414414167404175, 1.0, 0.9864864945411682, 0.9324324131011963, 0.9909909963607788, 0.9504504799842834, 0.954954981803894, 0.9909909963607788, 0.9684684872627258, 0.9459459185600281, 0.9009009003639221, 0.9729729890823364, 0.9009009003639221, 0.977477490901947, 0.9864864945411682, 1.0, 0.9909909963607788, 0.9954954981803894, 0.9234234094619751, 0.9954954981803894, 0.977477490901947, 1.0, 1.0, 0.9864864945411682, 0.9819819927215576, 1.0, 1.0, 1.0], [0.1621621549129486, 0.36936935782432556, 0.3513513505458832, 0.4009009003639221, 0.4054054021835327, 0.315315306186676, 0.3918918967247009, 0.44594594836235046, NaN, 0.5495495200157166, 0.684684693813324, 0.5405405163764954, 0.7747747898101807, 0.6936936974525452, 0.7792792916297913, 0.6126126050949097, 0.7747747898101807, 0.7882882952690125, 0.8063063025474548, 0.815315306186676, 0.792792797088623, 0.8828828930854797, 0.6711711883544922, 0.7657657861709595, 0.869369387626648, 0.7612612843513489, 0.8558558821678162, 0.8513513803482056, 0.837837815284729, 0.792792797088623, 0.8783783912658691, 0.7972972989082336, 0.8513513803482056, 0.9009009003639221, 0.8468468189239502, 0.7972972989082336, 0.9504504799842834, 0.8333333134651184, 0.869369387626648, 0.9099099040031433, 0.9189189076423645, 0.9459459185600281, 0.9504504799842834, 0.9189189076423645, 0.9369369149208069, 0.9504504799842834, 0.9414414167404175, 0.8828828930854797, 0.8873873949050903, 0.977477490901947, 0.9594594836235046, 0.8963963985443115, 0.9819819927215576, 0.9819819927215576, 0.9594594836235046, 0.9864864945411682, 0.9594594836235046, 0.9504504799842834, 0.8288288116455078, 0.9864864945411682, 0.8288288116455078, 0.9639639854431152, 1.0, 1.0, 1.0, 1.0, 0.9144144058227539, 1.0, 0.9819819927215576, 1.0, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.09909909963607788, 0.27927929162979126, 0.20720720291137695, 0.2702702581882477, 0.342342346906662, 0.2477477490901947, 0.30180180072784424, 0.4144144058227539, 0.45045045018196106, NaN, 0.5135135054588318, 0.4189189076423645, 0.5855855941772461, 0.6036036014556885, 0.6126126050949097, 0.7387387156486511, 0.6891891956329346, 0.7837837934494019, 0.7792792916297913, 0.7837837934494019, 0.7792792916297913, 0.8558558821678162, 0.662162184715271, 0.7522522807121277, 0.8783783912658691, 0.7207207083702087, 0.8648648858070374, 0.8288288116455078, 0.8513513803482056, 0.8018018007278442, 0.869369387626648, 0.792792797088623, 0.8738738894462585, 0.9279279112815857, 0.8108108043670654, 0.792792797088623, 0.9414414167404175, 0.8963963985443115, 0.8243243098258972, 0.9144144058227539, 0.8963963985443115, 0.9864864945411682, 0.9414414167404175, 0.9099099040031433, 0.9054054021835327, 0.9684684872627258, 0.9504504799842834, 0.9144144058227539, 0.8738738894462585, 0.9864864945411682, 0.9369369149208069, 0.9009009003639221, 0.9819819927215576, 0.977477490901947, 0.9684684872627258, 0.9864864945411682, 0.9684684872627258, 0.9729729890823364, 0.815315306186676, 0.9864864945411682, 0.8198198080062866, 0.977477490901947, 1.0, 0.9954954981803894, 0.9954954981803894, 0.9954954981803894, 0.9009009003639221, 1.0, 0.9684684872627258, 1.0, 1.0, 0.9819819927215576, 0.9909909963607788, 1.0, 1.0, 1.0], [0.09459459781646729, 0.28378379344940186, 0.31081080436706543, 0.37837839126586914, 0.36486485600471497, 0.30630630254745483, 0.36486485600471497, 0.4099099040031433, 0.315315306186676, 0.4864864945411682, NaN, 0.5090090036392212, 0.6756756901741028, 0.5945945978164673, 0.684684693813324, 0.5765765905380249, 0.7297297120094299, 0.7612612843513489, 0.7792792916297913, 0.7837837934494019, 0.7432432174682617, 0.8333333134651184, 0.6171171069145203, 0.7522522807121277, 0.837837815284729, 0.7162162065505981, 0.8333333134651184, 0.815315306186676, 0.8423423171043396, 0.7972972989082336, 0.8468468189239502, 0.7837837934494019, 0.8108108043670654, 0.8873873949050903, 0.8243243098258972, 0.7567567825317383, 0.954954981803894, 0.7882882952690125, 0.8468468189239502, 0.8738738894462585, 0.8963963985443115, 0.9279279112815857, 0.9369369149208069, 0.9099099040031433, 0.9369369149208069, 0.9459459185600281, 0.9324324131011963, 0.8828828930854797, 0.8603603839874268, 0.977477490901947, 0.9504504799842834, 0.8963963985443115, 0.9684684872627258, 0.9729729890823364, 0.9504504799842834, 0.9729729890823364, 0.9639639854431152, 0.9414414167404175, 0.8198198080062866, 0.977477490901947, 0.8243243098258972, 0.9594594836235046, 0.9954954981803894, 1.0, 1.0, 0.9909909963607788, 0.9189189076423645, 1.0, 0.9819819927215576, 1.0, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.18468467891216278, 0.30180180072784424, 0.2477477490901947, 0.3918918967247009, 0.37387385964393616, 0.30180180072784424, 0.18018017709255219, 0.45495495200157166, 0.45945945382118225, 0.5810810923576355, 0.4909909963607788, NaN, 0.5765765905380249, 0.6869369149208069, 0.6351351141929626, 0.6531531810760498, 0.6441441178321838, 0.6981981992721558, 0.7297297120094299, 0.7207207083702087, 0.792792797088623, 0.7432432174682617, 0.6981981992721558, 0.684684693813324, 0.792792797088623, 0.7702702879905701, 0.8108108043670654, 0.7702702879905701, 0.7792792916297913, 0.7207207083702087, 0.8108108043670654, 0.7297297120094299, 0.8603603839874268, 0.8603603839874268, 0.7972972989082336, 0.8603603839874268, 0.8828828930854797, 0.8288288116455078, 0.8333333134651184, 0.8963963985443115, 0.8513513803482056, 0.9144144058227539, 0.8603603839874268, 0.9099099040031433, 0.9144144058227539, 0.9009009003639221, 0.8648648858070374, 0.869369387626648, 0.8828828930854797, 0.9594594836235046, 0.9279279112815857, 0.9054054021835327, 0.8963963985443115, 0.954954981803894, 0.9234234094619751, 0.9099099040031433, 0.9009009003639221, 0.9459459185600281, 0.8063063025474548, 0.9639639854431152, 0.8108108043670654, 0.9099099040031433, 0.977477490901947, 0.9954954981803894, 1.0, 0.9639639854431152, 0.8918918967247009, 1.0, 0.9324324131011963, 0.9954954981803894, 1.0, 0.9729729890823364, 0.9684684872627258, 1.0, 1.0, 1.0], [0.11261261254549026, 0.2612612545490265, 0.21171170473098755, 0.31081080436706543, 0.3198198080062866, 0.27927929162979126, 0.315315306186676, 0.36486485600471497, 0.22522522509098053, 0.4144144058227539, 0.3243243098258972, 0.4234234094619751, NaN, 0.5045045018196106, 0.5765765905380249, 0.47297295928001404, 0.5855855941772461, 0.7027027010917664, 0.7162162065505981, 0.6936936974525452, 0.6081081032752991, 0.7792792916297913, 0.6081081032752991, 0.6891891956329346, 0.7297297120094299, 0.6576576828956604, 0.7477477192878723, 0.7522522807121277, 0.7702702879905701, 0.7342342138290405, 0.7747747898101807, 0.7297297120094299, 0.6936936974525452, 0.8108108043670654, 0.7702702879905701, 0.707207202911377, 0.8963963985443115, 0.7657657861709595, 0.7792792916297913, 0.8243243098258972, 0.8063063025474548, 0.869369387626648, 0.9324324131011963, 0.8198198080062866, 0.9009009003639221, 0.9144144058227539, 0.9369369149208069, 0.8558558821678162, 0.8108108043670654, 0.9414414167404175, 0.9279279112815857, 0.8243243098258972, 0.9729729890823364, 0.9144144058227539, 0.9459459185600281, 0.977477490901947, 0.954954981803894, 0.9099099040031433, 0.8288288116455078, 0.9729729890823364, 0.8243243098258972, 0.954954981803894, 0.9864864945411682, 0.9909909963607788, 0.9864864945411682, 0.9954954981803894, 0.9234234094619751, 1.0, 0.9819819927215576, 1.0, 1.0, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.06756756454706192, 0.20720720291137695, 0.12162162363529205, 0.2702702581882477, 0.20720720291137695, 0.1666666716337204, 0.11711711436510086, 0.23423422873020172, 0.30630630254745483, 0.3963963985443115, 0.4054054021835327, 0.3130630552768707, 0.4954954981803894, NaN, 0.545045018196106, 0.545045018196106, 0.6171171069145203, 0.7162162065505981, 0.7612612843513489, 0.6981981992721558, 0.7207207083702087, 0.7027027010917664, 0.5855855941772461, 0.6576576828956604, 0.7342342138290405, 0.6891891956329346, 0.7657657861709595, 0.7837837934494019, 0.7972972989082336, 0.7432432174682617, 0.792792797088623, 0.7117117047309875, 0.7657657861709595, 0.8198198080062866, 0.8018018007278442, 0.7387387156486511, 0.8513513803482056, 0.7432432174682617, 0.815315306186676, 0.8783783912658691, 0.837837815284729, 0.8333333134651184, 0.8648648858070374, 0.8828828930854797, 0.9099099040031433, 0.8738738894462585, 0.8738738894462585, 0.8423423171043396, 0.8468468189239502, 0.9324324131011963, 0.9189189076423645, 0.8648648858070374, 0.9054054021835327, 0.9054054021835327, 0.9189189076423645, 0.9054054021835327, 0.9144144058227539, 0.9279279112815857, 0.8198198080062866, 0.9144144058227539, 0.8243243098258972, 0.9009009003639221, 0.9729729890823364, 1.0, 0.9819819927215576, 0.9459459185600281, 0.8963963985443115, 0.9639639854431152, 0.9234234094619751, 0.9954954981803894, 1.0, 0.9504504799842834, 0.9594594836235046, 1.0, 1.0, 1.0], [0.045045044273138046, 0.22522522509098053, 0.12162162363529205, 0.2567567527294159, 0.315315306186676, 0.21621622145175934, 0.2522522509098053, 0.31081080436706543, 0.22072072327136993, 0.38738739490509033, 0.315315306186676, 0.36486485600471497, 0.4234234094619751, 0.45495495200157166, NaN, 0.4909909963607788, 0.630630612373352, 0.6981981992721558, 0.7117117047309875, 0.6801801919937134, 0.6756756901741028, 0.7252252101898193, 0.545045018196106, 0.6981981992721558, 0.7387387156486511, 0.6576576828956604, 0.7522522807121277, 0.7432432174682617, 0.7837837934494019, 0.7117117047309875, 0.7702702879905701, 0.7207207083702087, 0.7477477192878723, 0.8198198080062866, 0.7612612843513489, 0.6801801919937134, 0.8648648858070374, 0.6801801919937134, 0.7702702879905701, 0.8468468189239502, 0.8333333134651184, 0.8783783912658691, 0.9099099040031433, 0.8333333134651184, 0.9099099040031433, 0.9144144058227539, 0.9189189076423645, 0.837837815284729, 0.792792797088623, 0.9414414167404175, 0.9279279112815857, 0.8198198080062866, 0.9414414167404175, 0.9144144058227539, 0.9324324131011963, 0.9414414167404175, 0.9459459185600281, 0.9009009003639221, 0.7837837934494019, 0.9639639854431152, 0.792792797088623, 0.9369369149208069, 0.9819819927215576, 0.9909909963607788, 0.9909909963607788, 0.9864864945411682, 0.8873873949050903, 0.9909909963607788, 0.9684684872627258, 1.0, 1.0, 0.9729729890823364, 0.977477490901947, 1.0, 1.0, 1.0], [0.054054055362939835, 0.20720720291137695, 0.1756756752729416, 0.18918919563293457, 0.19819819927215576, 0.18468467891216278, 0.22072072327136993, 0.2522522509098053, 0.38738739490509033, 0.2612612545490265, 0.4234234094619751, 0.3468468487262726, 0.5270270109176636, 0.45495495200157166, 0.5090090036392212, NaN, 0.4819819927215576, 0.6531531810760498, 0.662162184715271, 0.6576576828956604, 0.6891891956329346, 0.6981981992721558, 0.5675675868988037, 0.5990990996360779, 0.6711711883544922, 0.5855855941772461, 0.7477477192878723, 0.6981981992721558, 0.7657657861709595, 0.7027027010917664, 0.7567567825317383, 0.7297297120094299, 0.7702702879905701, 0.7837837934494019, 0.7297297120094299, 0.684684693813324, 0.8288288116455078, 0.7342342138290405, 0.7882882952690125, 0.8558558821678162, 0.8288288116455078, 0.8918918967247009, 0.8198198080062866, 0.8468468189239502, 0.8873873949050903, 0.9369369149208069, 0.8513513803482056, 0.815315306186676, 0.8423423171043396, 0.9144144058227539, 0.9369369149208069, 0.8648648858070374, 0.9279279112815857, 0.9459459185600281, 0.9414414167404175, 0.9504504799842834, 0.8873873949050903, 0.9414414167404175, 0.7972972989082336, 0.9729729890823364, 0.8063063025474548, 0.9054054021835327, 0.977477490901947, 0.9819819927215576, 0.9909909963607788, 0.9909909963607788, 0.8738738894462585, 0.9909909963607788, 0.9504504799842834, 0.9909909963607788, 1.0, 0.9639639854431152, 0.977477490901947, 0.9954954981803894, 0.9954954981803894, 0.9954954981803894], [0.04054053872823715, 0.171171173453331, 0.157657653093338, 0.13963964581489563, 0.19819819927215576, 0.15315315127372742, 0.20270270109176636, 0.2477477490901947, 0.22522522509098053, 0.31081080436706543, 0.2702702581882477, 0.3558558523654938, 0.4144144058227539, 0.38288289308547974, 0.36936935782432556, 0.5180180072784424, NaN, 0.6711711883544922, 0.6711711883544922, 0.7837837934494019, 0.6261261105537415, 0.6891891956329346, 0.5585585832595825, 0.6711711883544922, 0.6936936974525452, 0.6081081032752991, 0.8288288116455078, 0.7747747898101807, 0.8603603839874268, 0.6036036014556885, 0.8513513803482056, 0.6756756901741028, 0.7252252101898193, 0.792792797088623, 0.8333333134651184, 0.6396396160125732, 0.869369387626648, 0.6261261105537415, 0.837837815284729, 0.792792797088623, 0.8198198080062866, 0.7837837934494019, 0.869369387626648, 0.9054054021835327, 0.9324324131011963, 0.9189189076423645, 0.8783783912658691, 0.815315306186676, 0.8513513803482056, 0.9639639854431152, 0.9279279112815857, 0.8783783912658691, 0.9144144058227539, 0.9684684872627258, 0.9819819927215576, 0.9189189076423645, 0.9279279112815857, 0.9234234094619751, 0.7972972989082336, 0.9639639854431152, 0.8063063025474548, 0.9189189076423645, 0.9954954981803894, 0.9954954981803894, 1.0, 0.977477490901947, 0.8783783912658691, 0.9909909963607788, 0.9369369149208069, 1.0, 1.0, 0.9639639854431152, 0.9729729890823364, 1.0, 1.0, 1.0], [0.036036036908626556, 0.10810811072587967, 0.14414414763450623, 0.13513512909412384, 0.12162162363529205, 0.12162162363529205, 0.19369369745254517, 0.0810810774564743, 0.21171170473098755, 0.21621622145175934, 0.2387387454509735, 0.30180180072784424, 0.29729729890823364, 0.28378379344940186, 0.30180180072784424, 0.3468468487262726, 0.3288288414478302, NaN, 0.6396396160125732, 0.6036036014556885, 0.4279279410839081, 0.5135135054588318, 0.4954954981803894, 0.5990990996360779, 0.4864864945411682, 0.5405405163764954, 0.7657657861709595, 0.7972972989082336, 0.7297297120094299, 0.5855855941772461, 0.7837837934494019, 0.6036036014556885, 0.5360360145568848, 0.6531531810760498, 0.7792792916297913, 0.6081081032752991, 0.7657657861709595, 0.6126126050949097, 0.8423423171043396, 0.6801801919937134, 0.7567567825317383, 0.7567567825317383, 0.8018018007278442, 0.9369369149208069, 0.7882882952690125, 0.8243243098258972, 0.837837815284729, 0.7882882952690125, 0.837837815284729, 0.9504504799842834, 0.8063063025474548, 0.8738738894462585, 0.9189189076423645, 0.9639639854431152, 0.9909909963607788, 0.9234234094619751, 0.8738738894462585, 0.9369369149208069, 0.7972972989082336, 0.9819819927215576, 0.7972972989082336, 0.8918918967247009, 0.9369369149208069, 0.9279279112815857, 0.9684684872627258, 0.9864864945411682, 0.869369387626648, 1.0, 0.9459459185600281, 0.9954954981803894, 1.0, 0.9684684872627258, 0.9729729890823364, 1.0, 1.0, 1.0], [0.022522522136569023, 0.0810810774564743, 0.13963964581489563, 0.11261261254549026, 0.09909909963607788, 0.12612612545490265, 0.15315315127372742, 0.0855855867266655, 0.19369369745254517, 0.22072072327136993, 0.22072072327136993, 0.2702702581882477, 0.28378379344940186, 0.2387387454509735, 0.28828829526901245, 0.3378378450870514, 0.3288288414478302, 0.36036035418510437, NaN, 0.522522509098053, 0.4189189076423645, 0.4909909963607788, 0.4864864945411682, 0.5630630850791931, 0.45495495200157166, 0.4954954981803894, 0.6756756901741028, 0.684684693813324, 0.684684693813324, 0.5675675868988037, 0.7297297120094299, 0.6036036014556885, 0.5270270109176636, 0.6216216087341309, 0.7342342138290405, 0.5855855941772461, 0.7207207083702087, 0.6126126050949097, 0.8063063025474548, 0.6666666865348816, 0.7342342138290405, 0.7432432174682617, 0.7882882952690125, 0.9144144058227539, 0.7702702879905701, 0.8108108043670654, 0.8018018007278442, 0.7837837934494019, 0.7972972989082336, 0.9279279112815857, 0.792792797088623, 0.8603603839874268, 0.8873873949050903, 0.954954981803894, 0.9909909963607788, 0.9144144058227539, 0.8738738894462585, 0.9234234094619751, 0.792792797088623, 0.9729729890823364, 0.8018018007278442, 0.8738738894462585, 0.9369369149208069, 0.9234234094619751, 0.9594594836235046, 0.977477490901947, 0.8603603839874268, 0.9954954981803894, 0.9369369149208069, 0.9864864945411682, 1.0, 0.9639639854431152, 0.9684684872627258, 1.0, 1.0, 1.0], [0.027027027681469917, 0.0855855867266655, 0.11261261254549026, 0.11261261254549026, 0.12612612545490265, 0.12612612545490265, 0.14864864945411682, 0.09009008854627609, 0.18468467891216278, 0.21621622145175934, 0.21621622145175934, 0.27927929162979126, 0.30630630254745483, 0.30180180072784424, 0.3198198080062866, 0.342342346906662, 0.21621622145175934, 0.3963963985443115, 0.477477490901947, NaN, 0.46846845746040344, 0.4909909963607788, 0.4864864945411682, 0.5, 0.4819819927215576, 0.5405405163764954, 0.6756756901741028, 0.7117117047309875, 0.7792792916297913, 0.5540540814399719, 0.7252252101898193, 0.4909909963607788, 0.5675675868988037, 0.5630630850791931, 0.7657657861709595, 0.5945945978164673, 0.6666666865348816, 0.6171171069145203, 0.8333333134651184, 0.7252252101898193, 0.7117117047309875, 0.7612612843513489, 0.815315306186676, 0.8963963985443115, 0.7117117047309875, 0.792792797088623, 0.815315306186676, 0.7117117047309875, 0.8108108043670654, 0.9819819927215576, 0.7432432174682617, 0.837837815284729, 0.8918918967247009, 0.9729729890823364, 0.977477490901947, 0.8963963985443115, 0.8468468189239502, 0.9414414167404175, 0.7747747898101807, 0.9729729890823364, 0.7792792916297913, 0.8873873949050903, 0.9729729890823364, 0.8963963985443115, 0.9909909963607788, 0.9729729890823364, 0.8468468189239502, 1.0, 0.9234234094619751, 0.9909909963607788, 1.0, 0.9459459185600281, 0.9684684872627258, 1.0, 1.0, 1.0], [0.045045044273138046, 0.07657657563686371, 0.10810811072587967, 0.09009008854627609, 0.0855855867266655, 0.0855855867266655, 0.09459459781646729, 0.157657653093338, 0.20720720291137695, 0.22072072327136993, 0.2567567527294159, 0.20720720291137695, 0.3918918967247009, 0.27927929162979126, 0.3243243098258972, 0.31081080436706543, 0.37387385964393616, 0.5720720887184143, 0.5810810923576355, 0.5315315127372742, NaN, 0.5585585832595825, 0.46396395564079285, 0.5090090036392212, 0.5495495200157166, 0.4324324429035187, 0.6261261105537415, 0.6486486196517944, 0.6396396160125732, 0.5945945978164673, 0.6711711883544922, 0.5990990996360779, 0.8603603839874268, 0.6441441178321838, 0.6576576828956604, 0.5945945978164673, 0.7117117047309875, 0.6126126050949097, 0.707207202911377, 0.9054054021835327, 0.6441441178321838, 0.837837815284729, 0.7342342138290405, 0.7882882952690125, 0.8198198080062866, 0.7522522807121277, 0.7522522807121277, 0.7162162065505981, 0.6936936974525452, 0.8333333134651184, 0.8648648858070374, 0.7522522807121277, 0.7612612843513489, 0.8333333134651184, 0.8828828930854797, 0.7702702879905701, 0.837837815284729, 0.8063063025474548, 0.7792792916297913, 0.8513513803482056, 0.7882882952690125, 0.815315306186676, 0.9414414167404175, 0.9684684872627258, 0.9819819927215576, 0.9369369149208069, 0.8648648858070374, 0.9369369149208069, 0.9099099040031433, 0.9639639854431152, 0.9909909963607788, 0.9369369149208069, 0.9459459185600281, 1.0, 1.0, 1.0], [0.036036036908626556, 0.09909909963607788, 0.10360360145568848, 0.09459459781646729, 0.12612612545490265, 0.12612612545490265, 0.1621621549129486, 0.14864864945411682, 0.11711711436510086, 0.14414414763450623, 0.1666666716337204, 0.2567567527294159, 0.22072072327136993, 0.29729729890823364, 0.2747747600078583, 0.30180180072784424, 0.31081080436706543, 0.4864864945411682, 0.5090090036392212, 0.5090090036392212, 0.44144144654273987, NaN, 0.46396395564079285, 0.5, 0.47297295928001404, 0.522522509098053, 0.6351351141929626, 0.5765765905380249, 0.630630612373352, 0.5630630850791931, 0.6666666865348816, 0.5855855941772461, 0.5540540814399719, 0.5990990996360779, 0.6441441178321838, 0.5855855941772461, 0.8198198080062866, 0.5810810923576355, 0.684684693813324, 0.5855855941772461, 0.7162162065505981, 0.7837837934494019, 0.837837815284729, 0.7297297120094299, 0.8468468189239502, 0.8198198080062866, 0.8468468189239502, 0.7882882952690125, 0.7297297120094299, 0.9144144058227539, 0.8558558821678162, 0.7612612843513489, 0.9369369149208069, 0.8603603839874268, 0.9054054021835327, 0.9414414167404175, 0.9369369149208069, 0.815315306186676, 0.7882882952690125, 0.9324324131011963, 0.7882882952690125, 0.9099099040031433, 0.9819819927215576, 0.9729729890823364, 0.977477490901947, 0.977477490901947, 0.8603603839874268, 0.9819819927215576, 0.9594594836235046, 0.9909909963607788, 0.9864864945411682, 0.9684684872627258, 0.9864864945411682, 0.9909909963607788, 0.9909909963607788, 0.9909909963607788], [0.12612612545490265, 0.10360360145568848, 0.13513512909412384, 0.2567567527294159, 0.2702702581882477, 0.21621622145175934, 0.21621622145175934, 0.21621622145175934, 0.3288288414478302, 0.3378378450870514, 0.38288289308547974, 0.30180180072784424, 0.3918918967247009, 0.4144144058227539, 0.45495495200157166, 0.4324324429035187, 0.44144144654273987, 0.5045045018196106, 0.5135135054588318, 0.5135135054588318, 0.5360360145568848, 0.5360360145568848, NaN, 0.5765765905380249, 0.5810810923576355, 0.5585585832595825, 0.5540540814399719, 0.5720720887184143, 0.5720720887184143, 0.6418918967247009, 0.5855855941772461, 0.6486486196517944, 0.5810810923576355, 0.6216216087341309, 0.6126126050949097, 0.707207202911377, 0.6531531810760498, 0.75, 0.6261261105537415, 0.6036036014556885, 0.6486486196517944, 0.7162162065505981, 0.662162184715271, 0.6936936974525452, 0.6711711883544922, 0.6801801919937134, 0.6666666865348816, 0.7207207083702087, 0.7252252101898193, 0.7117117047309875, 0.7117117047309875, 0.7702702879905701, 0.7117117047309875, 0.7342342138290405, 0.7297297120094299, 0.7882882952690125, 0.7522522807121277, 0.8513513803482056, 0.792792797088623, 0.7657657861709595, 0.7972972989082336, 0.8963963985443115, 0.9054054021835327, 0.8423423171043396, 0.9234234094619751, 0.8716216087341309, 0.8648648858070374, 0.9279279112815857, 0.9729729890823364, 0.9504504799842834, 0.9279279112815857, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.04054053872823715, 0.20720720291137695, 0.14414414763450623, 0.19819819927215576, 0.18918919563293457, 0.14864864945411682, 0.21171170473098755, 0.2387387454509735, 0.23423422873020172, 0.2477477490901947, 0.2477477490901947, 0.315315306186676, 0.31081080436706543, 0.342342346906662, 0.30180180072784424, 0.4009009003639221, 0.3288288414478302, 0.4009009003639221, 0.4369369447231293, 0.5, 0.4909909963607788, 0.5, 0.4234234094619751, NaN, 0.477477490901947, 0.5270270109176636, 0.5900900959968567, 0.6126126050949097, 0.5810810923576355, 0.4909909963607788, 0.5990990996360779, 0.4909909963607788, 0.5765765905380249, 0.5900900959968567, 0.6396396160125732, 0.5270270109176636, 0.6216216087341309, 0.5540540814399719, 0.6981981992721558, 0.6126126050949097, 0.8738738894462585, 0.7252252101898193, 0.7522522807121277, 0.7252252101898193, 0.7477477192878723, 0.8783783912658691, 0.7702702879905701, 0.7297297120094299, 0.7117117047309875, 0.8018018007278442, 0.7477477192878723, 0.7477477192878723, 0.8558558821678162, 0.869369387626648, 0.9054054021835327, 0.8828828930854797, 0.9234234094619751, 0.792792797088623, 0.7657657861709595, 0.8963963985443115, 0.7792792916297913, 0.8963963985443115, 0.9279279112815857, 0.9099099040031433, 0.9369369149208069, 0.9639639854431152, 0.8198198080062866, 0.954954981803894, 0.954954981803894, 0.9819819927215576, 1.0, 0.977477490901947, 0.9819819927215576, 0.9954954981803894, 0.9954954981803894, 1.0], [0.04954954981803894, 0.12162162363529205, 0.12162162363529205, 0.13513512909412384, 0.12162162363529205, 0.13963964581489563, 0.15315315127372742, 0.157657653093338, 0.13063062727451324, 0.12162162363529205, 0.1621621549129486, 0.20720720291137695, 0.2702702581882477, 0.2657657563686371, 0.2612612545490265, 0.3288288414478302, 0.30630630254745483, 0.5135135054588318, 0.545045018196106, 0.5180180072784424, 0.45045045018196106, 0.5270270109176636, 0.4189189076423645, 0.522522509098053, NaN, 0.5090090036392212, 0.6081081032752991, 0.6171171069145203, 0.6351351141929626, 0.6036036014556885, 0.6666666865348816, 0.5630630850791931, 0.5495495200157166, 0.7342342138290405, 0.6576576828956604, 0.5765765905380249, 0.7657657861709595, 0.6081081032752991, 0.6981981992721558, 0.6396396160125732, 0.6936936974525452, 0.6756756901741028, 0.7567567825317383, 0.7387387156486511, 0.7747747898101807, 0.8018018007278442, 0.7837837934494019, 0.8423423171043396, 0.7657657861709595, 0.8783783912658691, 0.7972972989082336, 0.7837837934494019, 0.837837815284729, 0.8198198080062866, 0.9054054021835327, 0.8468468189239502, 0.9729729890823364, 0.815315306186676, 0.7657657861709595, 0.8513513803482056, 0.7702702879905701, 0.9234234094619751, 0.8783783912658691, 0.9369369149208069, 0.9054054021835327, 0.977477490901947, 0.8198198080062866, 0.954954981803894, 0.9684684872627258, 0.9729729890823364, 0.9819819927215576, 0.9819819927215576, 0.9819819927215576, 0.9954954981803894, 0.9954954981803894, 0.9954954981803894], [0.10360360145568848, 0.13063062727451324, 0.06306306272745132, 0.13513512909412384, 0.14414414763450623, 0.1666666716337204, 0.13963964581489563, 0.22972972691059113, 0.2387387454509735, 0.27927929162979126, 0.28378379344940186, 0.22972972691059113, 0.342342346906662, 0.31081080436706543, 0.342342346906662, 0.4144144058227539, 0.3918918967247009, 0.45945945382118225, 0.5045045018196106, 0.45945945382118225, 0.5675675868988037, 0.477477490901947, 0.44144144654273987, 0.47297295928001404, 0.4909909963607788, NaN, 0.5180180072784424, 0.522522509098053, 0.5720720887184143, 0.6171171069145203, 0.5810810923576355, 0.5765765905380249, 0.6351351141929626, 0.6261261105537415, 0.5720720887184143, 0.5900900959968567, 0.6801801919937134, 0.6441441178321838, 0.6036036014556885, 0.6351351141929626, 0.662162184715271, 0.8738738894462585, 0.7162162065505981, 0.6666666865348816, 0.792792797088623, 0.8108108043670654, 0.7387387156486511, 0.7342342138290405, 0.7207207083702087, 0.7882882952690125, 0.815315306186676, 0.7657657861709595, 0.7837837934494019, 0.7747747898101807, 0.8558558821678162, 0.7882882952690125, 0.8738738894462585, 0.8198198080062866, 0.7792792916297913, 0.8333333134651184, 0.7882882952690125, 0.8738738894462585, 0.9639639854431152, 0.9459459185600281, 0.9684684872627258, 0.8963963985443115, 0.8558558821678162, 0.9369369149208069, 0.9639639854431152, 1.0, 0.954954981803894, 0.9819819927215576, 0.9909909963607788, 1.0, 1.0, 1.0], [0.036036036908626556, 0.07207207381725311, 0.11261261254549026, 0.07657657563686371, 0.06306306272745132, 0.11711711436510086, 0.10810811072587967, 0.10360360145568848, 0.14414414763450623, 0.13513512909412384, 0.1666666716337204, 0.18918919563293457, 0.2522522509098053, 0.23423422873020172, 0.2477477490901947, 0.2522522509098053, 0.171171173453331, 0.23423422873020172, 0.3243243098258972, 0.3243243098258972, 0.37387385964393616, 0.36486485600471497, 0.44594594836235046, 0.4099099040031433, 0.3918918967247009, 0.4819819927215576, NaN, 0.5405405163764954, 0.5090090036392212, 0.5270270109176636, 0.6396396160125732, 0.4864864945411682, 0.5090090036392212, 0.5495495200157166, 0.6801801919937134, 0.5585585832595825, 0.6351351141929626, 0.5810810923576355, 0.707207202911377, 0.6081081032752991, 0.6666666865348816, 0.7297297120094299, 0.7567567825317383, 0.869369387626648, 0.7027027010917664, 0.7522522807121277, 0.7837837934494019, 0.7207207083702087, 0.792792797088623, 0.9729729890823364, 0.7477477192878723, 0.8243243098258972, 0.8738738894462585, 0.9234234094619751, 0.9459459185600281, 0.8918918967247009, 0.837837815284729, 0.9234234094619751, 0.7702702879905701, 0.9594594836235046, 0.7837837934494019, 0.8603603839874268, 0.9459459185600281, 0.8873873949050903, 0.9819819927215576, 0.9819819927215576, 0.837837815284729, 0.9864864945411682, 0.9279279112815857, 0.9909909963607788, 1.0, 0.9639639854431152, 0.9639639854431152, 1.0, 1.0, 1.0], [0.03153153136372566, 0.0810810774564743, 0.12612612545490265, 0.09909909963607788, 0.06756756454706192, 0.11711711436510086, 0.10810811072587967, 0.0855855867266655, 0.14864864945411682, 0.171171173453331, 0.18468467891216278, 0.22972972691059113, 0.2477477490901947, 0.21621622145175934, 0.2567567527294159, 0.30180180072784424, 0.22522522509098053, 0.20270270109176636, 0.315315306186676, 0.28828829526901245, 0.3513513505458832, 0.4234234094619751, 0.4279279410839081, 0.38738739490509033, 0.38288289308547974, 0.477477490901947, 0.45945945382118225, NaN, 0.46846845746040344, 0.5135135054588318, 0.5090090036392212, 0.4234234094619751, 0.47297295928001404, 0.545045018196106, 0.6351351141929626, 0.5630630850791931, 0.5945945978164673, 0.5990990996360779, 0.6936936974525452, 0.5900900959968567, 0.6126126050949097, 0.6801801919937134, 0.707207202911377, 0.8648648858070374, 0.6711711883544922, 0.6666666865348816, 0.7162162065505981, 0.6261261105537415, 0.8018018007278442, 0.7972972989082336, 0.7027027010917664, 0.8423423171043396, 0.8063063025474548, 0.9189189076423645, 0.8198198080062866, 0.8198198080062866, 0.7747747898101807, 0.9144144058227539, 0.7792792916297913, 0.9414414167404175, 0.7837837934494019, 0.8333333134651184, 0.9189189076423645, 0.8018018007278442, 0.9324324131011963, 0.9414414167404175, 0.8603603839874268, 1.0, 0.8918918967247009, 0.9864864945411682, 1.0, 0.9594594836235046, 0.9684684872627258, 1.0, 1.0, 1.0], [0.022522522136569023, 0.07657657563686371, 0.09009008854627609, 0.06756756454706192, 0.09459459781646729, 0.10810811072587967, 0.13063062727451324, 0.0855855867266655, 0.1621621549129486, 0.14864864945411682, 0.157657653093338, 0.22072072327136993, 0.22972972691059113, 0.20270270109176636, 0.21621622145175934, 0.23423422873020172, 0.13963964581489563, 0.2702702581882477, 0.315315306186676, 0.22072072327136993, 0.36036035418510437, 0.36936935782432556, 0.4279279410839081, 0.4189189076423645, 0.36486485600471497, 0.4279279410839081, 0.4909909963607788, 0.5315315127372742, NaN, 0.5, 0.5495495200157166, 0.45495495200157166, 0.4864864945411682, 0.45945945382118225, 0.6261261105537415, 0.5270270109176636, 0.5585585832595825, 0.5675675868988037, 0.6801801919937134, 0.6441441178321838, 0.6261261105537415, 0.707207202911377, 0.7522522807121277, 0.7882882952690125, 0.6711711883544922, 0.707207202911377, 0.7567567825317383, 0.6666666865348816, 0.7477477192878723, 0.9144144058227539, 0.6936936974525452, 0.8108108043670654, 0.8288288116455078, 0.8828828930854797, 0.9459459185600281, 0.837837815284729, 0.837837815284729, 0.8558558821678162, 0.7657657861709595, 0.9099099040031433, 0.7747747898101807, 0.8288288116455078, 0.9189189076423645, 0.869369387626648, 0.954954981803894, 0.9504504799842834, 0.8288288116455078, 0.9864864945411682, 0.9099099040031433, 0.977477490901947, 1.0, 0.9414414167404175, 0.9459459185600281, 0.9909909963607788, 0.9909909963607788, 0.9954954981803894], [0.054054055362939835, 0.045045044273138046, 0.1666666716337204, 0.13963964581489563, 0.11261261254549026, 0.14864864945411682, 0.1666666716337204, 0.10360360145568848, 0.20720720291137695, 0.19819819927215576, 0.20270270109176636, 0.27927929162979126, 0.2657657563686371, 0.2567567527294159, 0.28828829526901245, 0.29729729890823364, 0.3963963985443115, 0.4144144058227539, 0.4324324429035187, 0.44594594836235046, 0.4054054021835327, 0.4369369447231293, 0.3581081032752991, 0.5090090036392212, 0.3963963985443115, 0.38288289308547974, 0.47297295928001404, 0.4864864945411682, 0.5, NaN, 0.5180180072784424, 0.5585585832595825, 0.4954954981803894, 0.4909909963607788, 0.5675675868988037, 0.45270270109176636, 0.545045018196106, 0.4864864945411682, 0.5945945978164673, 0.5720720887184143, 0.6396396160125732, 0.6396396160125732, 0.6036036014556885, 0.630630612373352, 0.7567567825317383, 0.707207202911377, 0.6126126050949097, 0.6936936974525452, 0.7657657861709595, 0.6756756901741028, 0.7837837934494019, 0.8198198080062866, 0.6216216087341309, 0.7162162065505981, 0.7252252101898193, 0.7162162065505981, 0.7387387156486511, 0.8198198080062866, 0.8828828930854797, 0.7387387156486511, 0.9009009003639221, 0.7972972989082336, 0.9144144058227539, 0.8513513803482056, 0.9414414167404175, 0.837837815284729, 0.9189189076423645, 0.9324324131011963, 0.9414414167404175, 0.9369369149208069, 0.9279279112815857, 0.9729729890823364, 0.9864864945411682, 0.9909909963607788, 0.9909909963607788, 1.0], [0.027027027681469917, 0.06306306272745132, 0.11711711436510086, 0.06306306272745132, 0.05855855718255043, 0.12612612545490265, 0.10360360145568848, 0.09009008854627609, 0.12162162363529205, 0.13063062727451324, 0.15315315127372742, 0.18918919563293457, 0.22522522509098053, 0.20720720291137695, 0.22972972691059113, 0.2432432472705841, 0.14864864945411682, 0.21621622145175934, 0.2702702581882477, 0.2747747600078583, 0.3288288414478302, 0.3333333432674408, 0.4144144058227539, 0.4009009003639221, 0.3333333432674408, 0.4189189076423645, 0.36036035418510437, 0.4909909963607788, 0.45045045018196106, 0.4819819927215576, NaN, 0.4819819927215576, 0.46846845746040344, 0.5090090036392212, 0.6036036014556885, 0.5405405163764954, 0.5675675868988037, 0.5675675868988037, 0.6396396160125732, 0.5585585832595825, 0.630630612373352, 0.6981981992721558, 0.7297297120094299, 0.8243243098258972, 0.6936936974525452, 0.7207207083702087, 0.7522522807121277, 0.7117117047309875, 0.7522522807121277, 0.9099099040031433, 0.7387387156486511, 0.8198198080062866, 0.8288288116455078, 0.8918918967247009, 0.9279279112815857, 0.837837815284729, 0.8288288116455078, 0.8963963985443115, 0.7702702879905701, 0.9234234094619751, 0.7702702879905701, 0.8468468189239502, 0.9369369149208069, 0.8963963985443115, 0.9684684872627258, 0.9639639854431152, 0.837837815284729, 0.9864864945411682, 0.9144144058227539, 0.9819819927215576, 1.0, 0.954954981803894, 0.9639639854431152, 1.0, 1.0, 1.0], [0.06756756454706192, 0.10810811072587967, 0.12612612545490265, 0.20270270109176636, 0.15315315127372742, 0.09009008854627609, 0.18468467891216278, 0.15315315127372742, 0.20270270109176636, 0.20720720291137695, 0.21621622145175934, 0.2702702581882477, 0.2702702581882477, 0.28828829526901245, 0.27927929162979126, 0.2702702581882477, 0.3243243098258972, 0.3963963985443115, 0.3963963985443115, 0.5090090036392212, 0.4009009003639221, 0.4144144058227539, 0.3513513505458832, 0.5090090036392212, 0.4369369447231293, 0.4234234094619751, 0.5135135054588318, 0.5765765905380249, 0.545045018196106, 0.44144144654273987, 0.5180180072784424, NaN, 0.477477490901947, 0.4819819927215576, 0.630630612373352, 0.5270270109176636, 0.5495495200157166, 0.5495495200157166, 0.630630612373352, 0.46846845746040344, 0.6126126050949097, 0.5900900959968567, 0.6576576828956604, 0.7117117047309875, 0.5990990996360779, 0.6936936974525452, 0.6576576828956604, 0.6756756901741028, 0.7252252101898193, 0.7297297120094299, 0.5900900959968567, 0.7702702879905701, 0.7432432174682617, 0.7522522807121277, 0.7882882952690125, 0.7432432174682617, 0.7342342138290405, 0.8288288116455078, 0.7432432174682617, 0.7657657861709595, 0.7522522807121277, 0.8198198080062866, 0.8828828930854797, 0.8063063025474548, 0.8918918967247009, 0.837837815284729, 0.8783783912658691, 0.9054054021835327, 0.9189189076423645, 0.9369369149208069, 0.9054054021835327, 0.9864864945411682, 0.9864864945411682, 0.9954954981803894, 0.9954954981803894, 1.0], [0.03153153136372566, 0.054054055362939835, 0.07657657563686371, 0.07657657563686371, 0.07657657563686371, 0.0810810774564743, 0.06756756454706192, 0.10360360145568848, 0.14864864945411682, 0.12612612545490265, 0.18918919563293457, 0.13963964581489563, 0.30630630254745483, 0.23423422873020172, 0.2522522509098053, 0.22972972691059113, 0.2747747600078583, 0.46396395564079285, 0.47297295928001404, 0.4324324429035187, 0.13963964581489563, 0.44594594836235046, 0.4189189076423645, 0.4234234094619751, 0.45045045018196106, 0.36486485600471497, 0.4909909963607788, 0.5270270109176636, 0.5135135054588318, 0.5045045018196106, 0.5315315127372742, 0.522522509098053, NaN, 0.5495495200157166, 0.5675675868988037, 0.5495495200157166, 0.6351351141929626, 0.5630630850791931, 0.5630630850791931, 0.7162162065505981, 0.5405405163764954, 0.7522522807121277, 0.684684693813324, 0.662162184715271, 0.7027027010917664, 0.6576576828956604, 0.7027027010917664, 0.662162184715271, 0.6261261105537415, 0.7342342138290405, 0.7702702879905701, 0.6531531810760498, 0.6891891956329346, 0.7612612843513489, 0.8063063025474548, 0.7117117047309875, 0.792792797088623, 0.7432432174682617, 0.7612612843513489, 0.7747747898101807, 0.7702702879905701, 0.7567567825317383, 0.869369387626648, 0.9414414167404175, 0.9324324131011963, 0.8873873949050903, 0.837837815284729, 0.8918918967247009, 0.8828828930854797, 0.9144144058227539, 0.9819819927215576, 0.9144144058227539, 0.9369369149208069, 0.9729729890823364, 0.9729729890823364, 0.9954954981803894], [0.03153153136372566, 0.07207207381725311, 0.07207207381725311, 0.09009008854627609, 0.04954954981803894, 0.12612612545490265, 0.11261261254549026, 0.07207207381725311, 0.09909909963607788, 0.07207207381725311, 0.11261261254549026, 0.13963964581489563, 0.18918919563293457, 0.18018017709255219, 0.18018017709255219, 0.21621622145175934, 0.20720720291137695, 0.3468468487262726, 0.37837839126586914, 0.4369369447231293, 0.3558558523654938, 0.4009009003639221, 0.37837839126586914, 0.4099099040031433, 0.2657657563686371, 0.37387385964393616, 0.45045045018196106, 0.45495495200157166, 0.5405405163764954, 0.5090090036392212, 0.4909909963607788, 0.5180180072784424, 0.45045045018196106, NaN, 0.5090090036392212, 0.5315315127372742, 0.6351351141929626, 0.5585585832595825, 0.6036036014556885, 0.5540540814399719, 0.6171171069145203, 0.6486486196517944, 0.630630612373352, 0.6486486196517944, 0.7387387156486511, 0.7342342138290405, 0.684684693813324, 0.7657657861709595, 0.7162162065505981, 0.7882882952690125, 0.7342342138290405, 0.7297297120094299, 0.7837837934494019, 0.7432432174682617, 0.8558558821678162, 0.7882882952690125, 0.8963963985443115, 0.7522522807121277, 0.7477477192878723, 0.792792797088623, 0.7522522807121277, 0.8243243098258972, 0.8468468189239502, 0.9234234094619751, 0.8828828930854797, 0.9594594836235046, 0.8018018007278442, 0.9324324131011963, 0.9504504799842834, 0.9504504799842834, 0.9909909963607788, 0.9639639854431152, 0.9729729890823364, 0.9864864945411682, 0.9864864945411682, 1.0], [0.027027027681469917, 0.05855855718255043, 0.12162162363529205, 0.07657657563686371, 0.054054055362939835, 0.13063062727451324, 0.10360360145568848, 0.07207207381725311, 0.15315315127372742, 0.18918919563293457, 0.1756756752729416, 0.20270270109176636, 0.22972972691059113, 0.19819819927215576, 0.2387387454509735, 0.2702702581882477, 0.1666666716337204, 0.22072072327136993, 0.2657657563686371, 0.23423422873020172, 0.342342346906662, 0.3558558523654938, 0.38738739490509033, 0.36036035418510437, 0.342342346906662, 0.4279279410839081, 0.3198198080062866, 0.36486485600471497, 0.37387385964393616, 0.4324324429035187, 0.3963963985443115, 0.36936935782432556, 0.4324324429035187, 0.4909909963607788, NaN, 0.5135135054588318, 0.5090090036392212, 0.5315315127372742, 0.6396396160125732, 0.5810810923576355, 0.5270270109176636, 0.6531531810760498, 0.6216216087341309, 0.8558558821678162, 0.6441441178321838, 0.5855855941772461, 0.6441441178321838, 0.5495495200157166, 0.7477477192878723, 0.7342342138290405, 0.6261261105537415, 0.7972972989082336, 0.7567567825317383, 0.8873873949050903, 0.7657657861709595, 0.7747747898101807, 0.6801801919937134, 0.8828828930854797, 0.7297297120094299, 0.8873873949050903, 0.7477477192878723, 0.7432432174682617, 0.8198198080062866, 0.7612612843513489, 0.8243243098258972, 0.9009009003639221, 0.8243243098258972, 0.9504504799842834, 0.8648648858070374, 0.9369369149208069, 0.9594594836235046, 0.9459459185600281, 0.9594594836235046, 0.9954954981803894, 0.9954954981803894, 0.9954954981803894], [0.09459459781646729, 0.054054055362939835, 0.07207207381725311, 0.12162162363529205, 0.14414414763450623, 0.11711711436510086, 0.07657657563686371, 0.11711711436510086, 0.20270270109176636, 0.20720720291137695, 0.2432432472705841, 0.13963964581489563, 0.29279279708862305, 0.2612612545490265, 0.3198198080062866, 0.315315306186676, 0.36036035418510437, 0.3918918967247009, 0.4144144058227539, 0.4054054021835327, 0.4054054021835327, 0.4144144058227539, 0.29279279708862305, 0.47297295928001404, 0.4234234094619751, 0.4099099040031433, 0.44144144654273987, 0.4369369447231293, 0.47297295928001404, 0.5472972989082336, 0.45945945382118225, 0.47297295928001404, 0.45045045018196106, 0.46846845746040344, 0.4864864945411682, NaN, 0.5135135054588318, 0.5382882952690125, 0.522522509098053, 0.5270270109176636, 0.5630630850791931, 0.5945945978164673, 0.5270270109176636, 0.5630630850791931, 0.6081081032752991, 0.5810810923576355, 0.5315315127372742, 0.662162184715271, 0.684684693813324, 0.5855855941772461, 0.6531531810760498, 0.7342342138290405, 0.5585585832595825, 0.5900900959968567, 0.6396396160125732, 0.6441441178321838, 0.707207202911377, 0.7612612843513489, 0.7567567825317383, 0.6396396160125732, 0.7567567825317383, 0.7657657861709595, 0.8198198080062866, 0.792792797088623, 0.8648648858070374, 0.8130630850791931, 0.8018018007278442, 0.869369387626648, 0.9009009003639221, 0.9144144058227539, 0.8963963985443115, 0.9369369149208069, 0.9459459185600281, 0.9909909963607788, 0.9909909963607788, 1.0], [0.022522522136569023, 0.022522522136569023, 0.027027027681469917, 0.05855855718255043, 0.027027027681469917, 0.09909909963607788, 0.07657657563686371, 0.018018018454313278, 0.04954954981803894, 0.05855855718255043, 0.045045044273138046, 0.11711711436510086, 0.10360360145568848, 0.14864864945411682, 0.13513512909412384, 0.171171173453331, 0.13063062727451324, 0.23423422873020172, 0.27927929162979126, 0.3333333432674408, 0.28828829526901245, 0.18018017709255219, 0.3468468487262726, 0.37837839126586914, 0.23423422873020172, 0.3198198080062866, 0.36486485600471497, 0.4054054021835327, 0.44144144654273987, 0.45495495200157166, 0.4324324429035187, 0.45045045018196106, 0.36486485600471497, 0.36486485600471497, 0.4909909963607788, 0.4864864945411682, NaN, 0.5, 0.5405405163764954, 0.46396395564079285, 0.5495495200157166, 0.5945945978164673, 0.6576576828956604, 0.6036036014556885, 0.6216216087341309, 0.6351351141929626, 0.684684693813324, 0.7027027010917664, 0.662162184715271, 0.7522522807121277, 0.6936936974525452, 0.707207202911377, 0.7792792916297913, 0.6801801919937134, 0.8018018007278442, 0.7792792916297913, 0.869369387626648, 0.7252252101898193, 0.7342342138290405, 0.7477477192878723, 0.7297297120094299, 0.8288288116455078, 0.869369387626648, 0.8828828930854797, 0.8873873949050903, 0.9369369149208069, 0.7882882952690125, 0.8963963985443115, 0.9054054021835327, 0.9639639854431152, 0.9819819927215576, 0.9459459185600281, 0.9414414167404175, 1.0, 1.0, 1.0], [0.03153153136372566, 0.054054055362939835, 0.036036036908626556, 0.07657657563686371, 0.10810811072587967, 0.06306306272745132, 0.0855855867266655, 0.10360360145568848, 0.1666666716337204, 0.10360360145568848, 0.21171170473098755, 0.171171173453331, 0.23423422873020172, 0.2567567527294159, 0.3198198080062866, 0.2657657563686371, 0.37387385964393616, 0.38738739490509033, 0.38738739490509033, 0.38288289308547974, 0.38738739490509033, 0.4189189076423645, 0.25, 0.44594594836235046, 0.3918918967247009, 0.3558558523654938, 0.4189189076423645, 0.4009009003639221, 0.4324324429035187, 0.5135135054588318, 0.4324324429035187, 0.45045045018196106, 0.4369369447231293, 0.44144144654273987, 0.46846845746040344, 0.46171170473098755, 0.5, NaN, 0.4864864945411682, 0.477477490901947, 0.5495495200157166, 0.5495495200157166, 0.5585585832595825, 0.5, 0.5765765905380249, 0.5855855941772461, 0.5630630850791931, 0.6531531810760498, 0.630630612373352, 0.545045018196106, 0.6216216087341309, 0.684684693813324, 0.5675675868988037, 0.5855855941772461, 0.6036036014556885, 0.6576576828956604, 0.707207202911377, 0.7027027010917664, 0.7342342138290405, 0.630630612373352, 0.7342342138290405, 0.7837837934494019, 0.7522522807121277, 0.7657657861709595, 0.8063063025474548, 0.8243243098258972, 0.7747747898101807, 0.815315306186676, 0.9144144058227539, 0.8738738894462585, 0.8603603839874268, 0.9369369149208069, 0.9459459185600281, 1.0, 1.0, 1.0], [0.03153153136372566, 0.045045044273138046, 0.11711711436510086, 0.05855855718255043, 0.04054053872823715, 0.12612612545490265, 0.06756756454706192, 0.06306306272745132, 0.13063062727451324, 0.1756756752729416, 0.15315315127372742, 0.1666666716337204, 0.22072072327136993, 0.18468467891216278, 0.22972972691059113, 0.21171170473098755, 0.1621621549129486, 0.157657653093338, 0.19369369745254517, 0.1666666716337204, 0.29279279708862305, 0.315315306186676, 0.37387385964393616, 0.30180180072784424, 0.30180180072784424, 0.3963963985443115, 0.29279279708862305, 0.30630630254745483, 0.3198198080062866, 0.4054054021835327, 0.36036035418510437, 0.36936935782432556, 0.4369369447231293, 0.3963963985443115, 0.36036035418510437, 0.477477490901947, 0.45945945382118225, 0.5135135054588318, NaN, 0.545045018196106, 0.5, 0.6351351141929626, 0.5990990996360779, 0.7702702879905701, 0.6081081032752991, 0.5675675868988037, 0.6216216087341309, 0.5180180072784424, 0.684684693813324, 0.6711711883544922, 0.5990990996360779, 0.7612612843513489, 0.7162162065505981, 0.8423423171043396, 0.7117117047309875, 0.7477477192878723, 0.6666666865348816, 0.8198198080062866, 0.7342342138290405, 0.8558558821678162, 0.7387387156486511, 0.7297297120094299, 0.792792797088623, 0.7432432174682617, 0.8108108043670654, 0.8468468189239502, 0.8018018007278442, 0.9459459185600281, 0.8513513803482056, 0.9009009003639221, 0.9594594836235046, 0.9234234094619751, 0.9414414167404175, 0.9909909963607788, 0.9864864945411682, 0.9909909963607788], [0.022522522136569023, 0.03153153136372566, 0.06306306272745132, 0.05855855718255043, 0.04054053872823715, 0.07657657563686371, 0.04054053872823715, 0.04054053872823715, 0.09009008854627609, 0.0855855867266655, 0.12612612545490265, 0.10360360145568848, 0.1756756752729416, 0.12162162363529205, 0.15315315127372742, 0.14414414763450623, 0.20720720291137695, 0.3198198080062866, 0.3333333432674408, 0.2747747600078583, 0.09459459781646729, 0.4144144058227539, 0.3963963985443115, 0.38738739490509033, 0.36036035418510437, 0.36486485600471497, 0.3918918967247009, 0.4099099040031433, 0.3558558523654938, 0.4279279410839081, 0.44144144654273987, 0.5315315127372742, 0.28378379344940186, 0.44594594836235046, 0.4189189076423645, 0.47297295928001404, 0.5360360145568848, 0.522522509098053, 0.45495495200157166, NaN, 0.4819819927215576, 0.6081081032752991, 0.6261261105537415, 0.545045018196106, 0.6666666865348816, 0.6171171069145203, 0.6351351141929626, 0.6351351141929626, 0.5900900959968567, 0.6666666865348816, 0.7387387156486511, 0.6441441178321838, 0.6441441178321838, 0.6891891956329346, 0.7657657861709595, 0.6531531810760498, 0.6936936974525452, 0.7297297120094299, 0.7567567825317383, 0.7162162065505981, 0.7657657861709595, 0.6891891956329346, 0.8423423171043396, 0.837837815284729, 0.9189189076423645, 0.8423423171043396, 0.8333333134651184, 0.8873873949050903, 0.837837815284729, 0.9054054021835327, 0.9459459185600281, 0.8783783912658691, 0.8873873949050903, 0.9819819927215576, 0.9819819927215576, 0.9954954981803894], [0.009009009227156639, 0.10360360145568848, 0.10360360145568848, 0.045045044273138046, 0.054054055362939835, 0.10360360145568848, 0.07207207381725311, 0.171171173453331, 0.0810810774564743, 0.10360360145568848, 0.10360360145568848, 0.14864864945411682, 0.19369369745254517, 0.1621621549129486, 0.1666666716337204, 0.171171173453331, 0.18018017709255219, 0.2432432472705841, 0.2657657563686371, 0.28828829526901245, 0.3558558523654938, 0.28378379344940186, 0.3513513505458832, 0.12612612545490265, 0.30630630254745483, 0.3378378450870514, 0.3333333432674408, 0.38738739490509033, 0.37387385964393616, 0.36036035418510437, 0.36936935782432556, 0.38738739490509033, 0.45945945382118225, 0.38288289308547974, 0.47297295928001404, 0.4369369447231293, 0.45045045018196106, 0.45045045018196106, 0.5, 0.5180180072784424, NaN, 0.5810810923576355, 0.5630630850791931, 0.5360360145568848, 0.6036036014556885, 0.7387387156486511, 0.5765765905380249, 0.5630630850791931, 0.6036036014556885, 0.6261261105537415, 0.6576576828956604, 0.6486486196517944, 0.7342342138290405, 0.684684693813324, 0.707207202911377, 0.7477477192878723, 0.815315306186676, 0.7027027010917664, 0.7387387156486511, 0.7612612843513489, 0.7432432174682617, 0.7567567825317383, 0.8513513803482056, 0.8108108043670654, 0.8873873949050903, 0.9054054021835327, 0.7792792916297913, 0.8828828930854797, 0.9234234094619751, 0.9459459185600281, 0.9639639854431152, 0.954954981803894, 0.9639639854431152, 0.9819819927215576, 0.9819819927215576, 1.0], [0.009009009227156639, 0.036036036908626556, 0.03153153136372566, 0.022522522136569023, 0.027027027681469917, 0.054054055362939835, 0.04954954981803894, 0.06306306272745132, 0.054054055362939835, 0.013513513840734959, 0.07207207381725311, 0.0855855867266655, 0.13063062727451324, 0.1666666716337204, 0.12162162363529205, 0.10810811072587967, 0.21621622145175934, 0.2432432472705841, 0.2567567527294159, 0.2387387454509735, 0.1621621549129486, 0.21621622145175934, 0.28378379344940186, 0.2747747600078583, 0.3243243098258972, 0.12612612545490265, 0.2702702581882477, 0.3198198080062866, 0.29279279708862305, 0.36036035418510437, 0.30180180072784424, 0.4099099040031433, 0.2477477490901947, 0.3513513505458832, 0.3468468487262726, 0.4054054021835327, 0.4054054021835327, 0.45045045018196106, 0.36486485600471497, 0.3918918967247009, 0.4189189076423645, NaN, 0.5900900959968567, 0.47297295928001404, 0.5630630850791931, 0.5630630850791931, 0.6081081032752991, 0.5540540814399719, 0.5630630850791931, 0.545045018196106, 0.6486486196517944, 0.5945945978164673, 0.6891891956329346, 0.5495495200157166, 0.6756756901741028, 0.7117117047309875, 0.6936936974525452, 0.6486486196517944, 0.6756756901741028, 0.6441441178321838, 0.684684693813324, 0.7477477192878723, 0.8333333134651184, 0.8333333134651184, 0.8783783912658691, 0.8198198080062866, 0.8108108043670654, 0.8288288116455078, 0.8828828930854797, 0.9414414167404175, 0.9234234094619751, 0.9594594836235046, 0.9729729890823364, 0.9729729890823364, 0.9729729890823364, 1.0], [0.027027027681469917, 0.04054053872823715, 0.054054055362939835, 0.06306306272745132, 0.06306306272745132, 0.0810810774564743, 0.11711711436510086, 0.04954954981803894, 0.04954954981803894, 0.05855855718255043, 0.06306306272745132, 0.13963964581489563, 0.06756756454706192, 0.13513512909412384, 0.09009008854627609, 0.18018017709255219, 0.13063062727451324, 0.19819819927215576, 0.21171170473098755, 0.18468467891216278, 0.2657657563686371, 0.1621621549129486, 0.3378378450870514, 0.2477477490901947, 0.2432432472705841, 0.28378379344940186, 0.2432432472705841, 0.29279279708862305, 0.2477477490901947, 0.3963963985443115, 0.2702702581882477, 0.342342346906662, 0.315315306186676, 0.36936935782432556, 0.37837839126586914, 0.47297295928001404, 0.342342346906662, 0.44144144654273987, 0.4009009003639221, 0.37387385964393616, 0.4369369447231293, 0.4099099040031433, NaN, 0.522522509098053, 0.4054054021835327, 0.45045045018196106, 0.704954981803894, 0.5540540814399719, 0.5630630850791931, 0.45045045018196106, 0.4369369447231293, 0.5945945978164673, 0.7657657861709595, 0.5990990996360779, 0.5765765905380249, 0.7882882952690125, 0.6486486196517944, 0.6441441178321838, 0.6891891956329346, 0.662162184715271, 0.6936936974525452, 0.954954981803894, 0.7432432174682617, 0.5945945978164673, 0.7342342138290405, 0.9234234094619751, 0.8063063025474548, 0.7747747898101807, 0.9729729890823364, 0.9369369149208069, 0.954954981803894, 0.9909909963607788, 0.9909909963607788, 1.0, 1.0, 1.0], [0.0045045046135783195, 0.027027027681469917, 0.0810810774564743, 0.054054055362939835, 0.013513513840734959, 0.09009008854627609, 0.027027027681469917, 0.054054055362939835, 0.0810810774564743, 0.09009008854627609, 0.09009008854627609, 0.09009008854627609, 0.18018017709255219, 0.11711711436510086, 0.1666666716337204, 0.15315315127372742, 0.09459459781646729, 0.06306306272745132, 0.0855855867266655, 0.10360360145568848, 0.21171170473098755, 0.2702702581882477, 0.30630630254745483, 0.2747747600078583, 0.2612612545490265, 0.3333333432674408, 0.13063062727451324, 0.13513512909412384, 0.21171170473098755, 0.36936935782432556, 0.1756756752729416, 0.28828829526901245, 0.3378378450870514, 0.3513513505458832, 0.14414414763450623, 0.4369369447231293, 0.3963963985443115, 0.5, 0.22972972691059113, 0.45495495200157166, 0.46396395564079285, 0.5270270109176636, 0.477477490901947, NaN, 0.5900900959968567, 0.5135135054588318, 0.5, 0.5045045018196106, 0.6036036014556885, 0.5405405163764954, 0.6081081032752991, 0.684684693813324, 0.6666666865348816, 0.7477477192878723, 0.6576576828956604, 0.6756756901741028, 0.630630612373352, 0.7387387156486511, 0.7117117047309875, 0.8108108043670654, 0.7162162065505981, 0.7207207083702087, 0.7162162065505981, 0.7162162065505981, 0.7477477192878723, 0.837837815284729, 0.7837837934494019, 0.9369369149208069, 0.837837815284729, 0.8918918967247009, 0.9504504799842834, 0.8963963985443115, 0.9234234094619751, 1.0, 1.0, 1.0], [0.009009009227156639, 0.03153153136372566, 0.027027027681469917, 0.013513513840734959, 0.04054053872823715, 0.0855855867266655, 0.022522522136569023, 0.036036036908626556, 0.06306306272745132, 0.09459459781646729, 0.06306306272745132, 0.0855855867266655, 0.09909909963607788, 0.09009008854627609, 0.09009008854627609, 0.11261261254549026, 0.06756756454706192, 0.21171170473098755, 0.22972972691059113, 0.28828829526901245, 0.18018017709255219, 0.15315315127372742, 0.3288288414478302, 0.2522522509098053, 0.22522522509098053, 0.20720720291137695, 0.29729729890823364, 0.3288288414478302, 0.3288288414478302, 0.2432432472705841, 0.30630630254745483, 0.4009009003639221, 0.29729729890823364, 0.2612612545490265, 0.3558558523654938, 0.3918918967247009, 0.37837839126586914, 0.4234234094619751, 0.3918918967247009, 0.3333333432674408, 0.3963963985443115, 0.4369369447231293, 0.5945945978164673, 0.4099099040031433, NaN, 0.5180180072784424, 0.6126126050949097, 0.5, 0.4909909963607788, 0.5315315127372742, 0.7207207083702087, 0.5630630850791931, 0.662162184715271, 0.4954954981803894, 0.5945945978164673, 0.684684693813324, 0.7252252101898193, 0.6216216087341309, 0.7342342138290405, 0.5810810923576355, 0.7297297120094299, 0.662162184715271, 0.8018018007278442, 0.9414414167404175, 0.8918918967247009, 0.7837837934494019, 0.8333333134651184, 0.837837815284729, 0.8603603839874268, 0.8963963985443115, 0.8918918967247009, 0.8963963985443115, 0.9054054021835327, 0.9279279112815857, 0.9279279112815857, 1.0], [0.009009009227156639, 0.04954954981803894, 0.03153153136372566, 0.036036036908626556, 0.013513513840734959, 0.09909909963607788, 0.054054055362939835, 0.0855855867266655, 0.04954954981803894, 0.03153153136372566, 0.054054055362939835, 0.09909909963607788, 0.0855855867266655, 0.12612612545490265, 0.0855855867266655, 0.06306306272745132, 0.0810810774564743, 0.1756756752729416, 0.18918919563293457, 0.20720720291137695, 0.2477477490901947, 0.18018017709255219, 0.3198198080062866, 0.12162162363529205, 0.19819819927215576, 0.18918919563293457, 0.2477477490901947, 0.3333333432674408, 0.29279279708862305, 0.29279279708862305, 0.27927929162979126, 0.30630630254745483, 0.342342346906662, 0.2657657563686371, 0.4144144058227539, 0.4189189076423645, 0.36486485600471497, 0.4144144058227539, 0.4324324429035187, 0.38288289308547974, 0.2612612545490265, 0.4369369447231293, 0.5495495200157166, 0.4864864945411682, 0.4819819927215576, NaN, 0.5540540814399719, 0.4909909963607788, 0.522522509098053, 0.5, 0.5405405163764954, 0.5405405163764954, 0.6891891956329346, 0.5630630850791931, 0.6126126050949097, 0.7342342138290405, 0.7837837934494019, 0.5990990996360779, 0.7162162065505981, 0.6756756901741028, 0.7162162065505981, 0.7522522807121277, 0.7747747898101807, 0.7567567825317383, 0.8243243098258972, 0.9144144058227539, 0.792792797088623, 0.8558558821678162, 0.9189189076423645, 0.9189189076423645, 0.9594594836235046, 0.9414414167404175, 0.9639639854431152, 0.9594594836235046, 0.9594594836235046, 1.0], [0.022522522136569023, 0.054054055362939835, 0.045045044273138046, 0.05855855718255043, 0.054054055362939835, 0.0855855867266655, 0.11261261254549026, 0.04954954981803894, 0.05855855718255043, 0.04954954981803894, 0.06756756454706192, 0.13513512909412384, 0.06306306272745132, 0.12612612545490265, 0.0810810774564743, 0.14864864945411682, 0.12162162363529205, 0.1621621549129486, 0.19819819927215576, 0.18468467891216278, 0.2477477490901947, 0.15315315127372742, 0.3333333432674408, 0.22972972691059113, 0.21621622145175934, 0.2612612545490265, 0.21621622145175934, 0.28378379344940186, 0.2432432472705841, 0.38738739490509033, 0.2477477490901947, 0.342342346906662, 0.29729729890823364, 0.315315306186676, 0.3558558523654938, 0.46846845746040344, 0.315315306186676, 0.4369369447231293, 0.37837839126586914, 0.36486485600471497, 0.4234234094619751, 0.3918918967247009, 0.29504504799842834, 0.5, 0.38738739490509033, 0.44594594836235046, NaN, 0.5495495200157166, 0.5405405163764954, 0.4279279410839081, 0.4279279410839081, 0.5720720887184143, 0.6936936974525452, 0.5900900959968567, 0.5315315127372742, 0.7342342138290405, 0.6351351141929626, 0.6261261105537415, 0.684684693813324, 0.6441441178321838, 0.6936936974525452, 0.9099099040031433, 0.707207202911377, 0.5765765905380249, 0.7252252101898193, 0.9099099040031433, 0.8018018007278442, 0.7837837934494019, 0.9684684872627258, 0.9234234094619751, 0.954954981803894, 0.9954954981803894, 0.9909909963607788, 1.0, 1.0, 1.0], [0.013513513840734959, 0.09459459781646729, 0.07207207381725311, 0.13963964581489563, 0.05855855718255043, 0.06756756454706192, 0.0855855867266655, 0.09459459781646729, 0.11711711436510086, 0.0855855867266655, 0.11711711436510086, 0.13063062727451324, 0.14414414763450623, 0.157657653093338, 0.1621621549129486, 0.18468467891216278, 0.18468467891216278, 0.21171170473098755, 0.21621622145175934, 0.28828829526901245, 0.28378379344940186, 0.21171170473098755, 0.27927929162979126, 0.2702702581882477, 0.157657653093338, 0.2657657563686371, 0.27927929162979126, 0.37387385964393616, 0.3333333432674408, 0.30630630254745483, 0.28828829526901245, 0.3243243098258972, 0.3378378450870514, 0.23423422873020172, 0.45045045018196106, 0.3378378450870514, 0.29729729890823364, 0.3468468487262726, 0.4819819927215576, 0.36486485600471497, 0.4369369447231293, 0.44594594836235046, 0.44594594836235046, 0.4954954981803894, 0.5, 0.5090090036392212, 0.45045045018196106, NaN, 0.5585585832595825, 0.5, 0.5315315127372742, 0.5945945978164673, 0.5405405163764954, 0.5585585832595825, 0.5720720887184143, 0.5720720887184143, 0.6171171069145203, 0.6036036014556885, 0.7297297120094299, 0.5720720887184143, 0.7297297120094299, 0.5540540814399719, 0.7162162065505981, 0.6576576828956604, 0.7432432174682617, 0.6891891956329346, 0.7657657861709595, 0.7567567825317383, 0.815315306186676, 0.7477477192878723, 0.7522522807121277, 0.869369387626648, 0.9504504799842834, 0.9324324131011963, 0.9324324131011963, 0.9864864945411682], [0.027027027681469917, 0.022522522136569023, 0.09909909963607788, 0.036036036908626556, 0.027027027681469917, 0.09009008854627609, 0.045045044273138046, 0.05855855718255043, 0.11261261254549026, 0.12612612545490265, 0.13963964581489563, 0.11711711436510086, 0.18918919563293457, 0.15315315127372742, 0.20720720291137695, 0.157657653093338, 0.14864864945411682, 0.1621621549129486, 0.20270270109176636, 0.18918919563293457, 0.30630630254745483, 0.2702702581882477, 0.2747747600078583, 0.28828829526901245, 0.23423422873020172, 0.27927929162979126, 0.20720720291137695, 0.19819819927215576, 0.2522522509098053, 0.23423422873020172, 0.2477477490901947, 0.2747747600078583, 0.37387385964393616, 0.28378379344940186, 0.2522522509098053, 0.315315306186676, 0.3378378450870514, 0.36936935782432556, 0.315315306186676, 0.4099099040031433, 0.3963963985443115, 0.4369369447231293, 0.4369369447231293, 0.3963963985443115, 0.5090090036392212, 0.477477490901947, 0.45945945382118225, 0.44144144654273987, NaN, 0.44144144654273987, 0.5540540814399719, 0.8063063025474548, 0.5180180072784424, 0.6351351141929626, 0.5045045018196106, 0.5540540814399719, 0.5585585832595825, 0.8603603839874268, 0.6981981992721558, 0.6531531810760498, 0.7297297120094299, 0.5360360145568848, 0.6396396160125732, 0.707207202911377, 0.7027027010917664, 0.6711711883544922, 0.8468468189239502, 0.8513513803482056, 0.7972972989082336, 0.8738738894462585, 0.8243243098258972, 0.8288288116455078, 0.869369387626648, 0.9864864945411682, 0.9864864945411682, 0.9954954981803894], [0.0, 0.009009009227156639, 0.009009009227156639, 0.022522522136569023, 0.0045045046135783195, 0.04954954981803894, 0.013513513840734959, 0.0, 0.022522522136569023, 0.013513513840734959, 0.022522522136569023, 0.04054053872823715, 0.05855855718255043, 0.06756756454706192, 0.05855855718255043, 0.0855855867266655, 0.036036036908626556, 0.04954954981803894, 0.07207207381725311, 0.018018018454313278, 0.1666666716337204, 0.0855855867266655, 0.28828829526901245, 0.19819819927215576, 0.12162162363529205, 0.21171170473098755, 0.027027027681469917, 0.20270270109176636, 0.0855855867266655, 0.3243243098258972, 0.09009008854627609, 0.2702702581882477, 0.2657657563686371, 0.21171170473098755, 0.2657657563686371, 0.4144144058227539, 0.2477477490901947, 0.45495495200157166, 0.3288288414478302, 0.3333333432674408, 0.37387385964393616, 0.45495495200157166, 0.5495495200157166, 0.45945945382118225, 0.46846845746040344, 0.5, 0.5720720887184143, 0.5, 0.5585585832595825, NaN, 0.5135135054588318, 0.6171171069145203, 0.6081081032752991, 0.5945945978164673, 0.7117117047309875, 0.6531531810760498, 0.7117117047309875, 0.6711711883544922, 0.7477477192878723, 0.6576576828956604, 0.7477477192878723, 0.7477477192878723, 0.7792792916297913, 0.7567567825317383, 0.8333333134651184, 0.8828828930854797, 0.7792792916297913, 0.8918918967247009, 0.8558558821678162, 0.9279279112815857, 0.9909909963607788, 0.9009009003639221, 0.9189189076423645, 0.9864864945411682, 0.9864864945411682, 0.9954954981803894], [0.0045045046135783195, 0.018018018454313278, 0.022522522136569023, 0.013513513840734959, 0.013513513840734959, 0.09459459781646729, 0.018018018454313278, 0.013513513840734959, 0.04054053872823715, 0.06306306272745132, 0.04954954981803894, 0.07207207381725311, 0.07207207381725311, 0.0810810774564743, 0.07207207381725311, 0.06306306272745132, 0.07207207381725311, 0.19369369745254517, 0.20720720291137695, 0.2567567527294159, 0.13513512909412384, 0.14414414763450623, 0.28828829526901245, 0.2522522509098053, 0.20270270109176636, 0.18468467891216278, 0.2522522509098053, 0.29729729890823364, 0.30630630254745483, 0.21621622145175934, 0.2612612545490265, 0.4099099040031433, 0.22972972691059113, 0.2657657563686371, 0.37387385964393616, 0.3468468487262726, 0.30630630254745483, 0.37837839126586914, 0.4009009003639221, 0.2612612545490265, 0.342342346906662, 0.3513513505458832, 0.5630630850791931, 0.3918918967247009, 0.27927929162979126, 0.45945945382118225, 0.5720720887184143, 0.46846845746040344, 0.44594594836235046, 0.4864864945411682, NaN, 0.4819819927215576, 0.6081081032752991, 0.46846845746040344, 0.5270270109176636, 0.6216216087341309, 0.6666666865348816, 0.5855855941772461, 0.7297297120094299, 0.545045018196106, 0.7297297120094299, 0.6261261105537415, 0.7027027010917664, 0.8783783912658691, 0.8108108043670654, 0.7207207083702087, 0.8288288116455078, 0.7747747898101807, 0.837837815284729, 0.8333333134651184, 0.8783783912658691, 0.8873873949050903, 0.8828828930854797, 0.9009009003639221, 0.9009009003639221, 1.0], [0.018018018454313278, 0.013513513840734959, 0.0855855867266655, 0.036036036908626556, 0.018018018454313278, 0.09459459781646729, 0.03153153136372566, 0.06756756454706192, 0.10360360145568848, 0.09909909963607788, 0.10360360145568848, 0.09459459781646729, 0.1756756752729416, 0.13513512909412384, 0.18018017709255219, 0.13513512909412384, 0.12162162363529205, 0.12612612545490265, 0.13963964581489563, 0.1621621549129486, 0.2477477490901947, 0.2387387454509735, 0.22972972691059113, 0.2522522509098053, 0.21621622145175934, 0.23423422873020172, 0.1756756752729416, 0.157657653093338, 0.18918919563293457, 0.18018017709255219, 0.18018017709255219, 0.22972972691059113, 0.3468468487262726, 0.2702702581882477, 0.20270270109176636, 0.2657657563686371, 0.29279279708862305, 0.315315306186676, 0.2387387454509735, 0.3558558523654938, 0.3513513505458832, 0.4054054021835327, 0.4054054021835327, 0.315315306186676, 0.4369369447231293, 0.45945945382118225, 0.4279279410839081, 0.4054054021835327, 0.19369369745254517, 0.38288289308547974, 0.5180180072784424, NaN, 0.46846845746040344, 0.5720720887184143, 0.45945945382118225, 0.5135135054588318, 0.5405405163764954, 0.7162162065505981, 0.6801801919937134, 0.6171171069145203, 0.707207202911377, 0.5135135054588318, 0.5945945978164673, 0.6666666865348816, 0.6711711883544922, 0.6261261105537415, 0.837837815284729, 0.8243243098258972, 0.7702702879905701, 0.8468468189239502, 0.8018018007278442, 0.8198198080062866, 0.8468468189239502, 0.9819819927215576, 0.9819819927215576, 0.9954954981803894], [0.0045045046135783195, 0.018018018454313278, 0.022522522136569023, 0.03153153136372566, 0.009009009227156639, 0.06306306272745132, 0.045045044273138046, 0.009009009227156639, 0.018018018454313278, 0.018018018454313278, 0.03153153136372566, 0.10360360145568848, 0.027027027681469917, 0.09459459781646729, 0.05855855718255043, 0.07207207381725311, 0.0855855867266655, 0.0810810774564743, 0.11261261254549026, 0.10810811072587967, 0.2387387454509735, 0.06306306272745132, 0.28828829526901245, 0.14414414763450623, 0.1621621549129486, 0.21621622145175934, 0.12612612545490265, 0.19369369745254517, 0.171171173453331, 0.37837839126586914, 0.171171173453331, 0.2567567527294159, 0.31081080436706543, 0.21621622145175934, 0.2432432472705841, 0.44144144654273987, 0.22072072327136993, 0.4324324429035187, 0.28378379344940186, 0.3558558523654938, 0.2657657563686371, 0.31081080436706543, 0.23423422873020172, 0.3333333432674408, 0.3378378450870514, 0.31081080436706543, 0.30630630254745483, 0.45945945382118225, 0.4819819927215576, 0.3918918967247009, 0.3918918967247009, 0.5315315127372742, NaN, 0.5180180072784424, 0.46396395564079285, 0.6824324131011963, 0.5405405163764954, 0.5945945978164673, 0.6351351141929626, 0.5855855941772461, 0.6486486196517944, 0.7207207083702087, 0.6171171069145203, 0.5675675868988037, 0.6666666865348816, 0.9864864945411682, 0.7612612843513489, 0.7747747898101807, 0.9234234094619751, 0.9054054021835327, 0.9459459185600281, 0.9504504799842834, 0.9684684872627258, 1.0, 1.0, 1.0], [0.009009009227156639, 0.013513513840734959, 0.04954954981803894, 0.009009009227156639, 0.0045045046135783195, 0.06756756454706192, 0.013513513840734959, 0.04954954981803894, 0.018018018454313278, 0.022522522136569023, 0.027027027681469917, 0.045045044273138046, 0.0855855867266655, 0.09459459781646729, 0.0855855867266655, 0.054054055362939835, 0.03153153136372566, 0.036036036908626556, 0.045045044273138046, 0.027027027681469917, 0.1666666716337204, 0.13963964581489563, 0.2657657563686371, 0.13063062727451324, 0.18018017709255219, 0.22522522509098053, 0.07657657563686371, 0.0810810774564743, 0.11711711436510086, 0.28378379344940186, 0.10810811072587967, 0.2477477490901947, 0.2387387454509735, 0.2567567527294159, 0.11261261254549026, 0.4099099040031433, 0.3198198080062866, 0.4144144058227539, 0.157657653093338, 0.31081080436706543, 0.315315306186676, 0.45045045018196106, 0.4009009003639221, 0.2522522509098053, 0.5045045018196106, 0.4369369447231293, 0.4099099040031433, 0.44144144654273987, 0.36486485600471497, 0.4054054021835327, 0.5315315127372742, 0.4279279410839081, 0.4819819927215576, NaN, 0.545045018196106, 0.5135135054588318, 0.5990990996360779, 0.5315315127372742, 0.5945945978164673, 0.8108108043670654, 0.5945945978164673, 0.5900900959968567, 0.6891891956329346, 0.6486486196517944, 0.7162162065505981, 0.707207202911377, 0.6711711883544922, 0.9639639854431152, 0.7657657861709595, 0.8603603839874268, 0.8828828930854797, 0.8963963985443115, 0.9144144058227539, 0.9819819927215576, 0.9819819927215576, 0.9954954981803894], [0.0, 0.027027027681469917, 0.04954954981803894, 0.0045045046135783195, 0.0, 0.0810810774564743, 0.027027027681469917, 0.045045044273138046, 0.04054053872823715, 0.03153153136372566, 0.04954954981803894, 0.07657657563686371, 0.054054055362939835, 0.0810810774564743, 0.06756756454706192, 0.05855855718255043, 0.018018018454313278, 0.009009009227156639, 0.009009009227156639, 0.022522522136569023, 0.11711711436510086, 0.09459459781646729, 0.2702702581882477, 0.09459459781646729, 0.09459459781646729, 0.14414414763450623, 0.054054055362939835, 0.18018017709255219, 0.054054055362939835, 0.2747747600078583, 0.07207207381725311, 0.21171170473098755, 0.19369369745254517, 0.14414414763450623, 0.23423422873020172, 0.36036035418510437, 0.19819819927215576, 0.3963963985443115, 0.28828829526901245, 0.23423422873020172, 0.29279279708862305, 0.3243243098258972, 0.4234234094619751, 0.342342346906662, 0.4054054021835327, 0.38738739490509033, 0.46846845746040344, 0.4279279410839081, 0.4954954981803894, 0.28828829526901245, 0.47297295928001404, 0.5405405163764954, 0.5360360145568848, 0.45495495200157166, NaN, 0.5810810923576355, 0.684684693813324, 0.5720720887184143, 0.707207202911377, 0.6036036014556885, 0.707207202911377, 0.6576576828956604, 0.6486486196517944, 0.6171171069145203, 0.7342342138290405, 0.8423423171043396, 0.7477477192878723, 0.8018018007278442, 0.8423423171043396, 0.815315306186676, 0.9504504799842834, 0.9009009003639221, 0.9189189076423645, 0.9504504799842834, 0.9504504799842834, 0.9864864945411682], [0.0045045046135783195, 0.018018018454313278, 0.022522522136569023, 0.027027027681469917, 0.009009009227156639, 0.054054055362939835, 0.036036036908626556, 0.009009009227156639, 0.013513513840734959, 0.013513513840734959, 0.027027027681469917, 0.09009008854627609, 0.022522522136569023, 0.09459459781646729, 0.05855855718255043, 0.04954954981803894, 0.0810810774564743, 0.07657657563686371, 0.0855855867266655, 0.10360360145568848, 0.22972972691059113, 0.05855855718255043, 0.21171170473098755, 0.11711711436510086, 0.15315315127372742, 0.21171170473098755, 0.10810811072587967, 0.18018017709255219, 0.1621621549129486, 0.28378379344940186, 0.1621621549129486, 0.2567567527294159, 0.28828829526901245, 0.21171170473098755, 0.22522522509098053, 0.3558558523654938, 0.22072072327136993, 0.342342346906662, 0.2522522509098053, 0.3468468487262726, 0.2522522509098053, 0.28828829526901245, 0.21171170473098755, 0.3243243098258972, 0.315315306186676, 0.2657657563686371, 0.2657657563686371, 0.4279279410839081, 0.44594594836235046, 0.3468468487262726, 0.37837839126586914, 0.4864864945411682, 0.3175675570964813, 0.4864864945411682, 0.4189189076423645, NaN, 0.5045045018196106, 0.5585585832595825, 0.5810810923576355, 0.5360360145568848, 0.5990990996360779, 0.6216216087341309, 0.5810810923576355, 0.5495495200157166, 0.6351351141929626, 0.7747747898101807, 0.7567567825317383, 0.7522522807121277, 0.8963963985443115, 0.9009009003639221, 0.9459459185600281, 0.9504504799842834, 0.9594594836235046, 1.0, 1.0, 1.0], [0.009009009227156639, 0.03153153136372566, 0.036036036908626556, 0.027027027681469917, 0.022522522136569023, 0.036036036908626556, 0.06756756454706192, 0.03153153136372566, 0.04054053872823715, 0.03153153136372566, 0.036036036908626556, 0.09909909963607788, 0.045045044273138046, 0.0855855867266655, 0.054054055362939835, 0.11261261254549026, 0.07207207381725311, 0.12612612545490265, 0.12612612545490265, 0.15315315127372742, 0.1621621549129486, 0.06306306272745132, 0.2477477490901947, 0.07657657563686371, 0.027027027681469917, 0.12612612545490265, 0.1621621549129486, 0.22522522509098053, 0.1621621549129486, 0.2612612545490265, 0.171171173453331, 0.2657657563686371, 0.20720720291137695, 0.10360360145568848, 0.3198198080062866, 0.29279279708862305, 0.13063062727451324, 0.29279279708862305, 0.3333333432674408, 0.30630630254745483, 0.18468467891216278, 0.30630630254745483, 0.3513513505458832, 0.36936935782432556, 0.2747747600078583, 0.21621622145175934, 0.36486485600471497, 0.38288289308547974, 0.44144144654273987, 0.28828829526901245, 0.3333333432674408, 0.45945945382118225, 0.45945945382118225, 0.4009009003639221, 0.315315306186676, 0.4954954981803894, NaN, 0.5315315127372742, 0.6351351141929626, 0.44594594836235046, 0.6351351141929626, 0.5090090036392212, 0.5135135054588318, 0.5135135054588318, 0.5585585832595825, 0.6441441178321838, 0.7207207083702087, 0.6216216087341309, 0.8738738894462585, 0.7252252101898193, 0.8198198080062866, 0.9234234094619751, 0.977477490901947, 0.9234234094619751, 0.9144144058227539, 0.9909909963607788], [0.013513513840734959, 0.009009009227156639, 0.05855855718255043, 0.045045044273138046, 0.018018018454313278, 0.05855855718255043, 0.027027027681469917, 0.054054055362939835, 0.04954954981803894, 0.027027027681469917, 0.05855855718255043, 0.054054055362939835, 0.09009008854627609, 0.07207207381725311, 0.09909909963607788, 0.05855855718255043, 0.07657657563686371, 0.06306306272745132, 0.07657657563686371, 0.05855855718255043, 0.19369369745254517, 0.18468467891216278, 0.14864864945411682, 0.20720720291137695, 0.18468467891216278, 0.18018017709255219, 0.07657657563686371, 0.0855855867266655, 0.14414414763450623, 0.18018017709255219, 0.10360360145568848, 0.171171173453331, 0.2567567527294159, 0.2477477490901947, 0.11711711436510086, 0.2387387454509735, 0.2747747600078583, 0.29729729890823364, 0.18018017709255219, 0.2702702581882477, 0.29729729890823364, 0.3513513505458832, 0.3558558523654938, 0.2612612545490265, 0.37837839126586914, 0.4009009003639221, 0.37387385964393616, 0.3963963985443115, 0.13963964581489563, 0.3288288414478302, 0.4144144058227539, 0.28378379344940186, 0.4054054021835327, 0.46846845746040344, 0.4279279410839081, 0.44144144654273987, 0.46846845746040344, NaN, 0.6261261105537415, 0.5270270109176636, 0.630630612373352, 0.4954954981803894, 0.5540540814399719, 0.6486486196517944, 0.6126126050949097, 0.5855855941772461, 0.7837837934494019, 0.7837837934494019, 0.7657657861709595, 0.792792797088623, 0.7837837934494019, 0.792792797088623, 0.8603603839874268, 0.9864864945411682, 0.9864864945411682, 0.9909909963607788], [0.07657657563686371, 0.0855855867266655, 0.13963964581489563, 0.10360360145568848, 0.09909909963607788, 0.13063062727451324, 0.13513512909412384, 0.09909909963607788, 0.171171173453331, 0.18468467891216278, 0.18018017709255219, 0.19369369745254517, 0.171171173453331, 0.18018017709255219, 0.21621622145175934, 0.20270270109176636, 0.20270270109176636, 0.20270270109176636, 0.20720720291137695, 0.22522522509098053, 0.22072072327136993, 0.21171170473098755, 0.20720720291137695, 0.23423422873020172, 0.23423422873020172, 0.22072072327136993, 0.22972972691059113, 0.22072072327136993, 0.23423422873020172, 0.11711711436510086, 0.22972972691059113, 0.2567567527294159, 0.2387387454509735, 0.2522522509098053, 0.2702702581882477, 0.2432432472705841, 0.2657657563686371, 0.2657657563686371, 0.2657657563686371, 0.2432432472705841, 0.2612612545490265, 0.3243243098258972, 0.31081080436706543, 0.28828829526901245, 0.2657657563686371, 0.28378379344940186, 0.315315306186676, 0.2702702581882477, 0.30180180072784424, 0.2522522509098053, 0.2702702581882477, 0.3198198080062866, 0.36486485600471497, 0.4054054021835327, 0.29279279708862305, 0.4189189076423645, 0.36486485600471497, 0.37387385964393616, NaN, 0.44594594836235046, 0.630630612373352, 0.46846845746040344, 0.4054054021835327, 0.3918918967247009, 0.4054054021835327, 0.4234234094619751, 0.9819819927215576, 0.477477490901947, 0.6801801919937134, 0.5990990996360779, 0.5540540814399719, 0.7567567825317383, 0.7882882952690125, 0.8288288116455078, 0.8288288116455078, 0.9144144058227539], [0.0045045046135783195, 0.009009009227156639, 0.018018018454313278, 0.0045045046135783195, 0.0045045046135783195, 0.06756756454706192, 0.013513513840734959, 0.027027027681469917, 0.013513513840734959, 0.013513513840734959, 0.022522522136569023, 0.036036036908626556, 0.027027027681469917, 0.0855855867266655, 0.036036036908626556, 0.027027027681469917, 0.036036036908626556, 0.018018018454313278, 0.027027027681469917, 0.027027027681469917, 0.14864864945411682, 0.06756756454706192, 0.23423422873020172, 0.10360360145568848, 0.14864864945411682, 0.1666666716337204, 0.04054053872823715, 0.05855855718255043, 0.09009008854627609, 0.2612612545490265, 0.07657657563686371, 0.23423422873020172, 0.22522522509098053, 0.20720720291137695, 0.11261261254549026, 0.36036035418510437, 0.2522522509098053, 0.36936935782432556, 0.14414414763450623, 0.28378379344940186, 0.2387387454509735, 0.3558558523654938, 0.3378378450870514, 0.18918919563293457, 0.4189189076423645, 0.3243243098258972, 0.3558558523654938, 0.4279279410839081, 0.3468468487262726, 0.342342346906662, 0.45495495200157166, 0.38288289308547974, 0.4144144058227539, 0.18918919563293457, 0.3963963985443115, 0.46396395564079285, 0.5540540814399719, 0.47297295928001404, 0.5540540814399719, NaN, 0.5540540814399719, 0.545045018196106, 0.6396396160125732, 0.6171171069145203, 0.6711711883544922, 0.6711711883544922, 0.6396396160125732, 0.8265765905380249, 0.7207207083702087, 0.792792797088623, 0.8468468189239502, 0.8738738894462585, 0.8963963985443115, 0.977477490901947, 0.977477490901947, 0.9909909963607788], [0.07207207381725311, 0.07657657563686371, 0.12162162363529205, 0.09909909963607788, 0.09009008854627609, 0.11711711436510086, 0.12162162363529205, 0.09909909963607788, 0.171171173453331, 0.18018017709255219, 0.1756756752729416, 0.18918919563293457, 0.1756756752729416, 0.1756756752729416, 0.20720720291137695, 0.19369369745254517, 0.19369369745254517, 0.20270270109176636, 0.19819819927215576, 0.22072072327136993, 0.21171170473098755, 0.21171170473098755, 0.20270270109176636, 0.22072072327136993, 0.22972972691059113, 0.21171170473098755, 0.21621622145175934, 0.21621622145175934, 0.22522522509098053, 0.09909909963607788, 0.22972972691059113, 0.2477477490901947, 0.22972972691059113, 0.2477477490901947, 0.2522522509098053, 0.2432432472705841, 0.2702702581882477, 0.2657657563686371, 0.2612612545490265, 0.23423422873020172, 0.2567567527294159, 0.315315306186676, 0.30630630254745483, 0.28378379344940186, 0.2702702581882477, 0.28378379344940186, 0.30630630254745483, 0.2702702581882477, 0.2702702581882477, 0.2522522509098053, 0.2702702581882477, 0.29279279708862305, 0.3513513505458832, 0.4054054021835327, 0.29279279708862305, 0.4009009003639221, 0.36486485600471497, 0.36936935782432556, 0.36936935782432556, 0.44594594836235046, NaN, 0.44594594836235046, 0.4054054021835327, 0.38738739490509033, 0.4009009003639221, 0.4054054021835327, 0.9684684872627258, 0.477477490901947, 0.6801801919937134, 0.5990990996360779, 0.5540540814399719, 0.7522522807121277, 0.7882882952690125, 0.8288288116455078, 0.8288288116455078, 0.8963963985443115], [0.018018018454313278, 0.036036036908626556, 0.03153153136372566, 0.04954954981803894, 0.036036036908626556, 0.04054053872823715, 0.045045044273138046, 0.022522522136569023, 0.036036036908626556, 0.022522522136569023, 0.04054053872823715, 0.09009008854627609, 0.045045044273138046, 0.09909909963607788, 0.06306306272745132, 0.09459459781646729, 0.0810810774564743, 0.10810811072587967, 0.12612612545490265, 0.11261261254549026, 0.18468467891216278, 0.09009008854627609, 0.10360360145568848, 0.10360360145568848, 0.07657657563686371, 0.12612612545490265, 0.13963964581489563, 0.1666666716337204, 0.171171173453331, 0.20270270109176636, 0.15315315127372742, 0.18018017709255219, 0.2432432472705841, 0.1756756752729416, 0.2567567527294159, 0.23423422873020172, 0.171171173453331, 0.21621622145175934, 0.2702702581882477, 0.31081080436706543, 0.2432432472705841, 0.2522522509098053, 0.045045044273138046, 0.27927929162979126, 0.3378378450870514, 0.2477477490901947, 0.09009008854627609, 0.44594594836235046, 0.46396395564079285, 0.2522522509098053, 0.37387385964393616, 0.4864864945411682, 0.27927929162979126, 0.4099099040031433, 0.342342346906662, 0.37837839126586914, 0.4909909963607788, 0.5045045018196106, 0.5315315127372742, 0.45495495200157166, 0.5540540814399719, NaN, 0.47297295928001404, 0.4054054021835327, 0.477477490901947, 0.6396396160125732, 0.6981981992721558, 0.6351351141929626, 0.8783783912658691, 0.7657657861709595, 0.8423423171043396, 0.9234234094619751, 0.9819819927215576, 0.9954954981803894, 0.9954954981803894, 1.0], [0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.018018018454313278, 0.04054053872823715, 0.0, 0.013513513840734959, 0.0, 0.0, 0.0045045046135783195, 0.022522522136569023, 0.013513513840734959, 0.027027027681469917, 0.018018018454313278, 0.022522522136569023, 0.0045045046135783195, 0.06306306272745132, 0.06306306272745132, 0.027027027681469917, 0.05855855718255043, 0.018018018454313278, 0.09459459781646729, 0.07207207381725311, 0.12162162363529205, 0.036036036908626556, 0.054054055362939835, 0.0810810774564743, 0.0810810774564743, 0.0855855867266655, 0.06306306272745132, 0.11711711436510086, 0.13063062727451324, 0.15315315127372742, 0.18018017709255219, 0.18018017709255219, 0.13063062727451324, 0.2477477490901947, 0.20720720291137695, 0.157657653093338, 0.14864864945411682, 0.1666666716337204, 0.2567567527294159, 0.28378379344940186, 0.19819819927215576, 0.22522522509098053, 0.29279279708862305, 0.28378379344940186, 0.36036035418510437, 0.22072072327136993, 0.29729729890823364, 0.4054054021835327, 0.38288289308547974, 0.31081080436706543, 0.3513513505458832, 0.4189189076423645, 0.4864864945411682, 0.44594594836235046, 0.5945945978164673, 0.36036035418510437, 0.5945945978164673, 0.5270270109176636, NaN, 0.5180180072784424, 0.6216216087341309, 0.5720720887184143, 0.6891891956329346, 0.6396396160125732, 0.7342342138290405, 0.9504504799842834, 0.8288288116455078, 0.8018018007278442, 0.8873873949050903, 0.977477490901947, 0.977477490901947, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.04054053872823715, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.009009009227156639, 0.0, 0.009009009227156639, 0.018018018454313278, 0.0045045046135783195, 0.07207207381725311, 0.07657657563686371, 0.10360360145568848, 0.03153153136372566, 0.027027027681469917, 0.157657653093338, 0.09009008854627609, 0.06306306272745132, 0.054054055362939835, 0.11261261254549026, 0.19819819927215576, 0.13063062727451324, 0.14864864945411682, 0.10360360145568848, 0.19369369745254517, 0.05855855718255043, 0.07657657563686371, 0.2387387454509735, 0.20720720291137695, 0.11711711436510086, 0.23423422873020172, 0.2567567527294159, 0.1621621549129486, 0.18918919563293457, 0.1666666716337204, 0.4054054021835327, 0.28378379344940186, 0.05855855718255043, 0.2432432472705841, 0.4234234094619751, 0.342342346906662, 0.29279279708862305, 0.2432432472705841, 0.12162162363529205, 0.3333333432674408, 0.4324324429035187, 0.3513513505458832, 0.38288289308547974, 0.45045045018196106, 0.4864864945411682, 0.3513513505458832, 0.6081081032752991, 0.38288289308547974, 0.6126126050949097, 0.5945945978164673, 0.4819819927215576, NaN, 0.5405405163764954, 0.6441441178321838, 0.707207202911377, 0.5405405163764954, 0.7747747898101807, 0.6351351141929626, 0.7657657861709595, 0.8513513803482056, 0.8603603839874268, 0.8873873949050903, 0.8873873949050903, 0.9864864945411682], [0.0, 0.0, 0.0, 0.0, 0.009009009227156639, 0.04054053872823715, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.013513513840734959, 0.018018018454313278, 0.009009009227156639, 0.009009009227156639, 0.0, 0.03153153136372566, 0.04054053872823715, 0.009009009227156639, 0.018018018454313278, 0.022522522136569023, 0.07657657563686371, 0.06306306272745132, 0.09459459781646729, 0.03153153136372566, 0.018018018454313278, 0.06756756454706192, 0.045045044273138046, 0.05855855718255043, 0.03153153136372566, 0.10810811072587967, 0.06756756454706192, 0.11711711436510086, 0.1756756752729416, 0.13513512909412384, 0.11261261254549026, 0.19369369745254517, 0.18918919563293457, 0.0810810774564743, 0.11261261254549026, 0.12162162363529205, 0.2657657563686371, 0.2522522509098053, 0.10810811072587967, 0.1756756752729416, 0.2747747600078583, 0.2567567527294159, 0.29729729890823364, 0.1666666716337204, 0.18918919563293457, 0.3288288414478302, 0.3333333432674408, 0.28378379344940186, 0.2657657563686371, 0.36486485600471497, 0.44144144654273987, 0.38738739490509033, 0.5945945978164673, 0.3288288414478302, 0.5990990996360779, 0.522522509098053, 0.37837839126586914, 0.45945945382118225, NaN, 0.5585585832595825, 0.6936936974525452, 0.5720720887184143, 0.707207202911377, 0.8198198080062866, 0.792792797088623, 0.7702702879905701, 0.8738738894462585, 0.954954981803894, 0.9504504799842834, 1.0], [0.0045045046135783195, 0.013513513840734959, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.027027027681469917, 0.009009009227156639, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.009009009227156639, 0.036036036908626556, 0.0045045046135783195, 0.054054055362939835, 0.013513513840734959, 0.009009009227156639, 0.022522522136569023, 0.013513513840734959, 0.022522522136569023, 0.027027027681469917, 0.06306306272745132, 0.022522522136569023, 0.12837837636470795, 0.036036036908626556, 0.022522522136569023, 0.10360360145568848, 0.018018018454313278, 0.05855855718255043, 0.04954954981803894, 0.1621621549129486, 0.036036036908626556, 0.1621621549129486, 0.11261261254549026, 0.04054053872823715, 0.09909909963607788, 0.18693692982196808, 0.06306306272745132, 0.1756756752729416, 0.15315315127372742, 0.157657653093338, 0.09459459781646729, 0.18018017709255219, 0.07657657563686371, 0.1621621549129486, 0.21621622145175934, 0.0855855867266655, 0.09009008854627609, 0.31081080436706543, 0.3288288414478302, 0.11711711436510086, 0.27927929162979126, 0.37387385964393616, 0.013513513840734959, 0.29279279708862305, 0.157657653093338, 0.22522522509098053, 0.3558558523654938, 0.4144144058227539, 0.5765765905380249, 0.3288288414478302, 0.5945945978164673, 0.36036035418510437, 0.4279279410839081, 0.3558558523654938, 0.44144144654273987, NaN, 0.6936936974525452, 0.5855855941772461, 0.7972972989082336, 0.6801801919937134, 0.7567567825317383, 0.8738738894462585, 0.9189189076423645, 0.954954981803894, 0.954954981803894, 0.9819819927215576], [0.05855855718255043, 0.07207207381725311, 0.07207207381725311, 0.06756756454706192, 0.06306306272745132, 0.06756756454706192, 0.0810810774564743, 0.07657657563686371, 0.0855855867266655, 0.09909909963607788, 0.0810810774564743, 0.10810811072587967, 0.07657657563686371, 0.10360360145568848, 0.11261261254549026, 0.12612612545490265, 0.12162162363529205, 0.13063062727451324, 0.13963964581489563, 0.15315315127372742, 0.13513512909412384, 0.13963964581489563, 0.13513512909412384, 0.18018017709255219, 0.18018017709255219, 0.14414414763450623, 0.1621621549129486, 0.13963964581489563, 0.171171173453331, 0.0810810774564743, 0.1621621549129486, 0.12162162363529205, 0.1621621549129486, 0.19819819927215576, 0.1756756752729416, 0.19819819927215576, 0.21171170473098755, 0.22522522509098053, 0.19819819927215576, 0.1666666716337204, 0.22072072327136993, 0.18918919563293457, 0.19369369745254517, 0.21621622145175934, 0.1666666716337204, 0.20720720291137695, 0.19819819927215576, 0.23423422873020172, 0.15315315127372742, 0.22072072327136993, 0.171171173453331, 0.1621621549129486, 0.2387387454509735, 0.3288288414478302, 0.2522522509098053, 0.2432432472705841, 0.27927929162979126, 0.21621622145175934, 0.018018018454313278, 0.36036035418510437, 0.03153153136372566, 0.30180180072784424, 0.31081080436706543, 0.29279279708862305, 0.30630630254745483, 0.30630630254745483, NaN, 0.37837839126586914, 0.5855855941772461, 0.5675675868988037, 0.44144144654273987, 0.6531531810760498, 0.7117117047309875, 0.8243243098258972, 0.8243243098258972, 0.8333333134651184], [0.0, 0.0, 0.0, 0.0, 0.0, 0.04054053872823715, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.036036036908626556, 0.009009009227156639, 0.009009009227156639, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0, 0.06306306272745132, 0.018018018454313278, 0.07207207381725311, 0.045045044273138046, 0.045045044273138046, 0.06306306272745132, 0.013513513840734959, 0.0, 0.013513513840734959, 0.06756756454706192, 0.013513513840734959, 0.09459459781646729, 0.10810811072587967, 0.06756756454706192, 0.04954954981803894, 0.13063062727451324, 0.10360360145568848, 0.18468467891216278, 0.054054055362939835, 0.11261261254549026, 0.11711711436510086, 0.171171173453331, 0.22522522509098053, 0.06306306272745132, 0.1621621549129486, 0.14414414763450623, 0.21621622145175934, 0.2432432472705841, 0.14864864945411682, 0.10810811072587967, 0.22522522509098053, 0.1756756752729416, 0.22522522509098053, 0.036036036908626556, 0.19819819927215576, 0.2477477490901947, 0.37837839126586914, 0.21621622145175934, 0.522522509098053, 0.1734234243631363, 0.522522509098053, 0.36486485600471497, 0.36036035418510437, 0.45945945382118225, 0.4279279410839081, 0.4144144058227539, 0.6216216087341309, NaN, 0.5900900959968567, 0.5945945978164673, 0.6486486196517944, 0.7117117047309875, 0.8513513803482056, 0.9144144058227539, 0.9144144058227539, 0.9729729890823364], [0.013513513840734959, 0.018018018454313278, 0.03153153136372566, 0.022522522136569023, 0.013513513840734959, 0.018018018454313278, 0.03153153136372566, 0.022522522136569023, 0.018018018454313278, 0.03153153136372566, 0.018018018454313278, 0.06756756454706192, 0.018018018454313278, 0.07657657563686371, 0.03153153136372566, 0.04954954981803894, 0.06306306272745132, 0.054054055362939835, 0.06306306272745132, 0.07657657563686371, 0.09009008854627609, 0.04054053872823715, 0.027027027681469917, 0.045045044273138046, 0.03153153136372566, 0.036036036908626556, 0.07207207381725311, 0.10810811072587967, 0.09009008854627609, 0.05855855718255043, 0.0855855867266655, 0.0810810774564743, 0.11711711436510086, 0.04954954981803894, 0.13513512909412384, 0.09909909963607788, 0.09459459781646729, 0.0855855867266655, 0.14864864945411682, 0.1621621549129486, 0.07657657563686371, 0.11711711436510086, 0.027027027681469917, 0.1621621549129486, 0.13963964581489563, 0.0810810774564743, 0.03153153136372566, 0.18468467891216278, 0.20270270109176636, 0.14414414763450623, 0.1621621549129486, 0.22972972691059113, 0.07657657563686371, 0.23423422873020172, 0.157657653093338, 0.10360360145568848, 0.12612612545490265, 0.23423422873020172, 0.3198198080062866, 0.27927929162979126, 0.3198198080062866, 0.12162162363529205, 0.2657657563686371, 0.22522522509098053, 0.29279279708862305, 0.20270270109176636, 0.4144144058227539, 0.4099099040031433, NaN, 0.44144144654273987, 0.3333333432674408, 0.9324324131011963, 0.9729729890823364, 0.8783783912658691, 0.8783783912658691, 0.9729729890823364], [0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.04054053872823715, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.013513513840734959, 0.009009009227156639, 0.036036036908626556, 0.009009009227156639, 0.04954954981803894, 0.018018018454313278, 0.027027027681469917, 0.0, 0.009009009227156639, 0.013513513840734959, 0.022522522136569023, 0.06306306272745132, 0.018018018454313278, 0.06306306272745132, 0.0855855867266655, 0.04954954981803894, 0.06306306272745132, 0.0855855867266655, 0.036036036908626556, 0.12612612545490265, 0.09909909963607788, 0.09459459781646729, 0.054054055362939835, 0.05855855718255043, 0.06306306272745132, 0.10810811072587967, 0.10360360145568848, 0.0810810774564743, 0.07657657563686371, 0.2522522509098053, 0.12612612545490265, 0.07207207381725311, 0.1666666716337204, 0.15315315127372742, 0.09459459781646729, 0.13963964581489563, 0.18468467891216278, 0.09909909963607788, 0.2747747600078583, 0.20720720291137695, 0.4009009003639221, 0.20720720291137695, 0.4009009003639221, 0.23423422873020172, 0.04954954981803894, 0.36486485600471497, 0.18018017709255219, 0.3198198080062866, 0.4324324429035187, 0.4054054021835327, 0.5585585832595825, NaN, 0.45045045018196106, 0.7207207083702087, 0.7972972989082336, 0.9414414167404175, 0.9414414167404175, 0.9954954981803894], [0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.009009009227156639, 0.013513513840734959, 0.07207207381725311, 0.0, 0.018018018454313278, 0.045045044273138046, 0.0, 0.0, 0.0, 0.07207207381725311, 0.0, 0.09459459781646729, 0.018018018454313278, 0.009009009227156639, 0.04054053872823715, 0.10360360145568848, 0.018018018454313278, 0.13963964581489563, 0.04054053872823715, 0.054054055362939835, 0.036036036908626556, 0.07657657563686371, 0.045045044273138046, 0.04954954981803894, 0.10810811072587967, 0.04054053872823715, 0.045045044273138046, 0.2477477490901947, 0.1756756752729416, 0.009009009227156639, 0.12162162363529205, 0.19819819927215576, 0.054054055362939835, 0.11711711436510086, 0.04954954981803894, 0.054054055362939835, 0.18018017709255219, 0.21621622145175934, 0.44594594836235046, 0.15315315127372742, 0.44594594836235046, 0.157657653093338, 0.171171173453331, 0.23423422873020172, 0.20720720291137695, 0.2432432472705841, 0.5585585832595825, 0.3513513505458832, 0.6666666865348816, 0.5495495200157166, NaN, 0.7612612843513489, 0.792792797088623, 0.8243243098258972, 0.8198198080062866, 0.8783783912658691], [0.0045045046135783195, 0.009009009227156639, 0.018018018454313278, 0.009009009227156639, 0.013513513840734959, 0.009009009227156639, 0.022522522136569023, 0.013513513840734959, 0.009009009227156639, 0.018018018454313278, 0.009009009227156639, 0.027027027681469917, 0.009009009227156639, 0.04954954981803894, 0.027027027681469917, 0.036036036908626556, 0.036036036908626556, 0.03153153136372566, 0.036036036908626556, 0.054054055362939835, 0.06306306272745132, 0.03153153136372566, 0.009009009227156639, 0.022522522136569023, 0.018018018454313278, 0.018018018454313278, 0.036036036908626556, 0.04054053872823715, 0.05855855718255043, 0.027027027681469917, 0.045045044273138046, 0.013513513840734959, 0.0855855867266655, 0.036036036908626556, 0.054054055362939835, 0.06306306272745132, 0.054054055362939835, 0.06306306272745132, 0.07657657563686371, 0.12162162363529205, 0.045045044273138046, 0.04054053872823715, 0.009009009227156639, 0.10360360145568848, 0.10360360145568848, 0.05855855718255043, 0.0045045046135783195, 0.13063062727451324, 0.171171173453331, 0.09909909963607788, 0.11261261254549026, 0.18018017709255219, 0.04954954981803894, 0.10360360145568848, 0.09909909963607788, 0.04954954981803894, 0.07657657563686371, 0.20720720291137695, 0.2432432472705841, 0.12612612545490265, 0.2477477490901947, 0.07657657563686371, 0.19819819927215576, 0.14864864945411682, 0.22972972691059113, 0.12612612545490265, 0.3468468487262726, 0.28828829526901245, 0.06756756454706192, 0.27927929162979126, 0.2387387454509735, NaN, 0.9189189076423645, 0.8603603839874268, 0.8603603839874268, 0.9144144058227539], [0.009009009227156639, 0.018018018454313278, 0.009009009227156639, 0.009009009227156639, 0.009009009227156639, 0.009009009227156639, 0.013513513840734959, 0.018018018454313278, 0.009009009227156639, 0.009009009227156639, 0.009009009227156639, 0.03153153136372566, 0.009009009227156639, 0.04054053872823715, 0.022522522136569023, 0.022522522136569023, 0.027027027681469917, 0.027027027681469917, 0.03153153136372566, 0.03153153136372566, 0.054054055362939835, 0.013513513840734959, 0.009009009227156639, 0.018018018454313278, 0.018018018454313278, 0.009009009227156639, 0.036036036908626556, 0.03153153136372566, 0.054054055362939835, 0.013513513840734959, 0.036036036908626556, 0.013513513840734959, 0.06306306272745132, 0.027027027681469917, 0.04054053872823715, 0.054054055362939835, 0.05855855718255043, 0.054054055362939835, 0.05855855718255043, 0.11261261254549026, 0.036036036908626556, 0.027027027681469917, 0.009009009227156639, 0.07657657563686371, 0.09459459781646729, 0.036036036908626556, 0.009009009227156639, 0.04954954981803894, 0.13063062727451324, 0.0810810774564743, 0.11711711436510086, 0.15315315127372742, 0.03153153136372566, 0.0855855867266655, 0.0810810774564743, 0.04054053872823715, 0.022522522136569023, 0.13963964581489563, 0.21171170473098755, 0.10360360145568848, 0.21171170473098755, 0.018018018454313278, 0.11261261254549026, 0.13963964581489563, 0.12612612545490265, 0.0810810774564743, 0.28828829526901245, 0.14864864945411682, 0.027027027681469917, 0.20270270109176636, 0.20720720291137695, 0.0810810774564743, NaN, 0.630630612373352, 0.6351351141929626, 0.6891891956329346], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.009009009227156639, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.027027027681469917, 0.013513513840734959, 0.0045045046135783195, 0.009009009227156639, 0.0, 0.0, 0.009009009227156639, 0.018018018454313278, 0.018018018454313278, 0.027027027681469917, 0.0, 0.0, 0.07207207381725311, 0.04054053872823715, 0.0, 0.06756756454706192, 0.013513513840734959, 0.013513513840734959, 0.09909909963607788, 0.018018018454313278, 0.0, 0.018018018454313278, 0.04954954981803894, 0.0, 0.07657657563686371, 0.013513513840734959, 0.171171173453331, 0.022522522136569023, 0.171171173453331, 0.0045045046135783195, 0.022522522136569023, 0.11261261254549026, 0.045045044273138046, 0.045045044273138046, 0.1756756752729416, 0.0855855867266655, 0.12162162363529205, 0.05855855718255043, 0.1756756752729416, 0.13963964581489563, 0.36936935782432556, NaN, 0.6013513803482056, 0.8243243098258972], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.009009009227156639, 0.009009009227156639, 0.0, 0.0045045046135783195, 0.027027027681469917, 0.013513513840734959, 0.0045045046135783195, 0.009009009227156639, 0.0, 0.0, 0.013513513840734959, 0.018018018454313278, 0.018018018454313278, 0.027027027681469917, 0.0, 0.0, 0.07207207381725311, 0.04054053872823715, 0.0, 0.06756756454706192, 0.013513513840734959, 0.013513513840734959, 0.09909909963607788, 0.018018018454313278, 0.0, 0.018018018454313278, 0.04954954981803894, 0.0, 0.0855855867266655, 0.013513513840734959, 0.171171173453331, 0.022522522136569023, 0.171171173453331, 0.0045045046135783195, 0.022522522136569023, 0.11261261254549026, 0.04954954981803894, 0.045045044273138046, 0.1756756752729416, 0.0855855867266655, 0.12162162363529205, 0.05855855718255043, 0.18018017709255219, 0.13963964581489563, 0.36486485600471497, 0.3986486494541168, NaN, 0.7972972989082336], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.009009009227156639, 0.0045045046135783195, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.013513513840734959, 0.0045045046135783195, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.0, 0.0045045046135783195, 0.013513513840734959, 0.0, 0.009009009227156639, 0.009009009227156639, 0.0855855867266655, 0.009009009227156639, 0.10360360145568848, 0.0, 0.0, 0.013513513840734959, 0.0, 0.018018018454313278, 0.1666666716337204, 0.027027027681469917, 0.027027027681469917, 0.0045045046135783195, 0.12162162363529205, 0.0855855867266655, 0.31081080436706543, 0.1756756752729416, 0.20270270109176636, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9447447491},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.892972974},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8844444474},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8634834842},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8629429442},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.858978982},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8585585602},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.848408409},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8246846883},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8028828859},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.7979579584},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7829129144},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7505105097},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7373273253},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7351351328},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7256456473},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7229429443},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6683483495},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6458258277},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6448648658},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6357357349},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6355555558},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.632582584},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6297297308},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6284084086},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6254654677},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5989789785},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5781981971},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5761561552},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.5730930913},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5722522514},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5703303276},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.5595195196},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5559759755},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5324924911},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5300600601},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5094294287},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5064864865},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5001801773},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4980780749},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.4978378382},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4503903889},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4454354362},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4375375346},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.433453453},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4322522536},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4268768761},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4224624636},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4162762765},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4020420415},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3965765754},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.3748948931},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.36987988},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3587987986},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3539339324},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3414714716},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3240240213},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.322822823},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3218017995},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.3142042046},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3120720716},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.3060060057},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.2723723722},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2709909917},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2413813832},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2292492508},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2243243246},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1973273269},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1753153149},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1569969961},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.144504505},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1225825832},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.083663663},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0506606609},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0478378379},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0194594595}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_regression/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..97336b222e5622a0751547ecab2a82f3299e5008
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_all/winrate_matrix.csv
@@ -0,0 +1,77 @@
+,TabFM (default),Nori-30M (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),Nori (default),TabDPT (tuned + ensembled),RealMLP (tuned + ensembled),TabDPT (tuned),RealTabPFN-2.5 (tuned),TabDPT (default),RealTabPFN-2.5 (default),TabDPT-Turbo (default),RealMLP (tuned),"AutoGluon 1.4 (best, 4h)",CatBoost (tuned + ensembled),CatBoost (tuned),LightGBM (tuned + ensembled),TabM (tuned + ensembled),iLTM (tuned + ensembled),TabPFNv2 (tuned + ensembled),ModernNCA (tuned + ensembled),xRFM (tuned + ensembled),TabSwift (default),XGBoost (tuned + ensembled),CatBoost (default),LightGBM (tuned),Mitra (default),XGBoost (tuned),SAP-RPT-OSS (default),TabM (tuned),xRFM (tuned),ChimeraBoost (tuned + ensembled),TabPFNv2 (tuned),iLTM (tuned),TabPFNv2 (default),ChimeraBoost (tuned),TabM (default),ModernNCA (tuned),RealMLP (default),ExtraTrees (tuned + ensembled),ChimeraBoost (default),TorchMLP (tuned + ensembled),ModernNCA (default),ExtraTrees (tuned),LimiX (default),EBM (tuned + ensembled),LightGBM (default),TorchMLP (tuned),EBM (tuned),RandomForest (tuned + ensembled),PerpetualBooster (tuned + ensembled),XGBoost (default),RandomForest (tuned),xRFM (default),EBM (default),TabSTAR (tuned + ensembled),PerpetualBooster (tuned),TabSTAR (tuned),ExtraTrees (default),FastaiMLP (tuned + ensembled),TorchMLP (default),FastaiMLP (tuned),RandomForest (default),TabSTAR (default),PerpetualBooster (default),KNN (tuned + ensembled),FastaiMLP (default),iLTM (default),KNN (tuned),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,0.6126126,0.6981982,0.7297297,0.8153153,0.7432432,0.7927928,0.7972973,0.8378378,0.9009009,0.9054054,0.8153153,0.8873874,0.9324324,0.954955,0.9459459,0.9594595,0.963964,0.9774775,0.972973,0.954955,0.963964,0.8738739,0.9594595,0.9504505,0.8963964,0.963964,0.9684685,0.9774775,0.9459459,0.972973,0.9324324,0.9684685,0.9684685,0.972973,0.9054054,0.9774775,0.9684685,0.9684685,0.9774775,0.990991,0.990991,0.972973,0.9954955,0.990991,0.990991,0.9774775,0.9864865,0.972973,1.0,0.9954955,0.981982,0.9954955,0.990991,1.0,0.9954955,0.990991,0.9864865,0.9234234,0.9954955,0.9279279,0.981982,1.0,1.0,1.0,0.9954955,0.9414414,1.0,0.9864865,1.0,1.0,0.9954955,0.990991,1.0,1.0,1.0
+Nori-30M (default),0.3873874,,0.4774775,0.6531532,0.6801802,0.545045,0.6216216,0.7297297,0.6306306,0.7207207,0.7162162,0.6981982,0.7387387,0.7927928,0.7747748,0.7927928,0.8288288,0.8918919,0.9189189,0.9144144,0.9234234,0.9009009,0.8963964,0.7927928,0.8783784,0.8693694,0.9279279,0.9189189,0.9234234,0.954955,0.9369369,0.8918919,0.9459459,0.9279279,0.9414414,0.9459459,0.9774775,0.9459459,0.954955,0.9684685,0.8963964,0.963964,0.9594595,0.972973,0.9684685,0.9504505,0.9459459,0.9054054,0.9774775,0.990991,0.981982,0.9864865,0.981982,0.9864865,0.972973,0.981982,0.9684685,0.990991,0.9144144,0.990991,0.9234234,0.963964,0.9954955,1.0,1.0,0.9864865,0.9279279,1.0,0.981982,1.0,0.9954955,0.990991,0.981982,1.0,1.0,1.0
+TabPFN-3 (default),0.3018018,0.5225225,,0.6441441,0.6396396,0.6126126,0.6216216,0.6396396,0.6486486,0.7927928,0.6891892,0.7522523,0.7882883,0.8783784,0.8783784,0.8243243,0.8423423,0.8558559,0.8603604,0.8873874,0.8918919,0.8963964,0.8648649,0.8558559,0.8783784,0.9369369,0.8873874,0.8738739,0.9099099,0.8333333,0.8828829,0.8738739,0.9234234,0.9279279,0.8783784,0.9279279,0.972973,0.963964,0.8828829,0.9369369,0.8963964,0.9684685,0.9459459,0.9189189,0.972973,0.9684685,0.954955,0.9279279,0.9009009,0.990991,0.9774775,0.9144144,0.9774775,0.9504505,0.9504505,0.9774775,0.963964,0.9414414,0.8603604,0.981982,0.8783784,0.9684685,1.0,1.0,1.0,0.9954955,0.9279279,1.0,0.9684685,1.0,1.0,0.981982,0.990991,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.27027026,0.34684685,0.35585585,,0.5630631,0.44144145,0.45945945,0.5225225,0.5990991,0.7297297,0.6216216,0.6081081,0.6891892,0.7297297,0.7432432,0.8108108,0.8603604,0.8648649,0.8873874,0.8873874,0.9099099,0.9054054,0.7432432,0.8018018,0.8648649,0.8648649,0.9234234,0.9009009,0.9324324,0.8603604,0.9369369,0.7972973,0.9234234,0.9099099,0.9234234,0.8783784,0.9414414,0.9234234,0.9414414,0.9414414,0.954955,0.9774775,0.9369369,0.9459459,0.9864865,0.963964,0.9414414,0.8603604,0.963964,0.9774775,0.9864865,0.963964,0.9684685,0.990991,0.9954955,0.972973,0.972973,0.954955,0.8963964,0.9954955,0.9009009,0.9504505,0.9954955,1.0,1.0,1.0,0.9324324,1.0,0.9774775,1.0,1.0,0.990991,0.990991,1.0,1.0,1.0
+TabPFN-2.6 (default),0.18468468,0.3198198,0.36036035,0.43693694,,0.3873874,0.47297296,0.47297296,0.5945946,0.6576577,0.6351351,0.6261261,0.6801802,0.7927928,0.6846847,0.8018018,0.8018018,0.8783784,0.9009009,0.8738739,0.9144144,0.8738739,0.7297297,0.8108108,0.8783784,0.8558559,0.9369369,0.9324324,0.9054054,0.8873874,0.9414414,0.8468468,0.9234234,0.9504505,0.9459459,0.8558559,0.972973,0.8918919,0.9594595,0.9594595,0.9459459,0.972973,0.9369369,0.9864865,0.9594595,0.9864865,0.9459459,0.9414414,0.972973,0.9954955,0.9864865,0.981982,0.990991,0.9954955,1.0,0.990991,0.9774775,0.981982,0.9009009,0.9954955,0.9099099,0.963964,0.981982,1.0,0.990991,0.9954955,0.9369369,1.0,0.9864865,0.9954955,1.0,0.9864865,0.990991,1.0,1.0,1.0
+TabICLv2 (default),0.25675675,0.45495495,0.3873874,0.5585586,0.6126126,,0.5855856,0.5675676,0.6846847,0.7522523,0.6936937,0.6981982,0.7207207,0.8333333,0.7837838,0.8153153,0.8468468,0.8783784,0.8738739,0.8738739,0.9144144,0.8738739,0.7837838,0.8513514,0.8603604,0.8333333,0.8828829,0.8828829,0.8918919,0.8513514,0.8738739,0.9099099,0.9189189,0.8738739,0.8693694,0.8828829,0.9009009,0.9369369,0.8738739,0.9234234,0.8963964,0.9459459,0.9189189,0.9099099,0.9144144,0.9009009,0.9144144,0.9324324,0.9099099,0.9504505,0.9054054,0.9054054,0.9369369,0.9324324,0.9189189,0.9459459,0.963964,0.9414414,0.8693694,0.9324324,0.8828829,0.9594595,0.9594595,0.9594595,0.9594595,0.972973,0.9324324,0.9594595,0.981982,0.9594595,1.0,0.990991,0.990991,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.2072072,0.3783784,0.3783784,0.5405405,0.527027,0.4144144,,0.536036,0.6081081,0.6981982,0.6351351,0.8198198,0.6846847,0.8828829,0.7477477,0.7792793,0.7972973,0.8063063,0.8468468,0.8513514,0.9054054,0.8378378,0.7837838,0.7882883,0.8468468,0.8603604,0.8918919,0.8918919,0.8693694,0.8333333,0.8963964,0.8153153,0.9324324,0.8873874,0.8963964,0.9234234,0.9234234,0.9144144,0.9324324,0.9594595,0.9279279,0.9504505,0.8828829,0.972973,0.9774775,0.9459459,0.8873874,0.9144144,0.954955,0.9864865,0.981982,0.9684685,0.954955,0.9864865,0.972973,0.963964,0.9324324,0.972973,0.8648649,0.9864865,0.8783784,0.954955,1.0,1.0,1.0,0.990991,0.9189189,1.0,0.9684685,1.0,1.0,0.9774775,0.9864865,1.0,1.0,1.0
+Nori (default),0.2027027,0.27027026,0.36036035,0.4774775,0.527027,0.43243244,0.46396396,,0.5540541,0.5855856,0.5900901,0.545045,0.6351351,0.7657658,0.6891892,0.7477477,0.7522523,0.9189189,0.9144144,0.9099099,0.8423423,0.8513514,0.7837838,0.7612613,0.8423423,0.7702703,0.8963964,0.9144144,0.9144144,0.8963964,0.9099099,0.8468468,0.8963964,0.9279279,0.9279279,0.8828829,0.981982,0.8963964,0.9369369,0.9594595,0.8288288,0.9369369,0.9504505,0.9459459,0.963964,0.9144144,0.9504505,0.9054054,0.9414414,1.0,0.9864865,0.9324324,0.990991,0.9504505,0.954955,0.990991,0.9684685,0.9459459,0.9009009,0.972973,0.9009009,0.9774775,0.9864865,1.0,0.990991,0.9954955,0.9234234,0.9954955,0.9774775,1.0,1.0,0.9864865,0.981982,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.16216215,0.36936936,0.35135135,0.4009009,0.4054054,0.3153153,0.3918919,0.44594595,,0.5495495,0.6846847,0.5405405,0.7747748,0.6936937,0.7792793,0.6126126,0.7747748,0.7882883,0.8063063,0.8153153,0.7927928,0.8828829,0.6711712,0.7657658,0.8693694,0.7612613,0.8558559,0.8513514,0.8378378,0.7927928,0.8783784,0.7972973,0.8513514,0.9009009,0.8468468,0.7972973,0.9504505,0.8333333,0.8693694,0.9099099,0.9189189,0.9459459,0.9504505,0.9189189,0.9369369,0.9504505,0.9414414,0.8828829,0.8873874,0.9774775,0.9594595,0.8963964,0.981982,0.981982,0.9594595,0.9864865,0.9594595,0.9504505,0.8288288,0.9864865,0.8288288,0.963964,1.0,1.0,1.0,1.0,0.9144144,1.0,0.981982,1.0,1.0,0.990991,0.990991,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0990991,0.2792793,0.2072072,0.27027026,0.34234235,0.24774775,0.3018018,0.4144144,0.45045045,,0.5135135,0.4189189,0.5855856,0.6036036,0.6126126,0.7387387,0.6891892,0.7837838,0.7792793,0.7837838,0.7792793,0.8558559,0.6621622,0.7522523,0.8783784,0.7207207,0.8648649,0.8288288,0.8513514,0.8018018,0.8693694,0.7927928,0.8738739,0.9279279,0.8108108,0.7927928,0.9414414,0.8963964,0.8243243,0.9144144,0.8963964,0.9864865,0.9414414,0.9099099,0.9054054,0.9684685,0.9504505,0.9144144,0.8738739,0.9864865,0.9369369,0.9009009,0.981982,0.9774775,0.9684685,0.9864865,0.9684685,0.972973,0.8153153,0.9864865,0.8198198,0.9774775,1.0,0.9954955,0.9954955,0.9954955,0.9009009,1.0,0.9684685,1.0,1.0,0.981982,0.990991,1.0,1.0,1.0
+TabDPT (tuned),0.0945946,0.2837838,0.3108108,0.3783784,0.36486486,0.3063063,0.36486486,0.4099099,0.3153153,0.4864865,,0.509009,0.6756757,0.5945946,0.6846847,0.5765766,0.7297297,0.7612613,0.7792793,0.7837838,0.7432432,0.8333333,0.6171171,0.7522523,0.8378378,0.7162162,0.8333333,0.8153153,0.8423423,0.7972973,0.8468468,0.7837838,0.8108108,0.8873874,0.8243243,0.7567568,0.954955,0.7882883,0.8468468,0.8738739,0.8963964,0.9279279,0.9369369,0.9099099,0.9369369,0.9459459,0.9324324,0.8828829,0.8603604,0.9774775,0.9504505,0.8963964,0.9684685,0.972973,0.9504505,0.972973,0.963964,0.9414414,0.8198198,0.9774775,0.8243243,0.9594595,0.9954955,1.0,1.0,0.990991,0.9189189,1.0,0.981982,1.0,1.0,0.990991,0.990991,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.18468468,0.3018018,0.24774775,0.3918919,0.37387386,0.3018018,0.18018018,0.45495495,0.45945945,0.5810811,0.490991,,0.5765766,0.6869369,0.6351351,0.6531532,0.6441441,0.6981982,0.7297297,0.7207207,0.7927928,0.7432432,0.6981982,0.6846847,0.7927928,0.7702703,0.8108108,0.7702703,0.7792793,0.7207207,0.8108108,0.7297297,0.8603604,0.8603604,0.7972973,0.8603604,0.8828829,0.8288288,0.8333333,0.8963964,0.8513514,0.9144144,0.8603604,0.9099099,0.9144144,0.9009009,0.8648649,0.8693694,0.8828829,0.9594595,0.9279279,0.9054054,0.8963964,0.954955,0.9234234,0.9099099,0.9009009,0.9459459,0.8063063,0.963964,0.8108108,0.9099099,0.9774775,0.9954955,1.0,0.963964,0.8918919,1.0,0.9324324,0.9954955,1.0,0.972973,0.9684685,1.0,1.0,1.0
+TabDPT (default),0.11261261,0.26126125,0.2117117,0.3108108,0.3198198,0.2792793,0.3153153,0.36486486,0.22522523,0.4144144,0.3243243,0.4234234,,0.5045045,0.5765766,0.47297296,0.5855856,0.7027027,0.7162162,0.6936937,0.6081081,0.7792793,0.6081081,0.6891892,0.7297297,0.6576577,0.7477477,0.7522523,0.7702703,0.7342342,0.7747748,0.7297297,0.6936937,0.8108108,0.7702703,0.7072072,0.8963964,0.7657658,0.7792793,0.8243243,0.8063063,0.8693694,0.9324324,0.8198198,0.9009009,0.9144144,0.9369369,0.8558559,0.8108108,0.9414414,0.9279279,0.8243243,0.972973,0.9144144,0.9459459,0.9774775,0.954955,0.9099099,0.8288288,0.972973,0.8243243,0.954955,0.9864865,0.990991,0.9864865,0.9954955,0.9234234,1.0,0.981982,1.0,1.0,0.990991,0.990991,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.067567565,0.2072072,0.12162162,0.27027026,0.2072072,0.16666667,0.117117114,0.23423423,0.3063063,0.3963964,0.4054054,0.31306306,0.4954955,,0.545045,0.545045,0.6171171,0.7162162,0.7612613,0.6981982,0.7207207,0.7027027,0.5855856,0.6576577,0.7342342,0.6891892,0.7657658,0.7837838,0.7972973,0.7432432,0.7927928,0.7117117,0.7657658,0.8198198,0.8018018,0.7387387,0.8513514,0.7432432,0.8153153,0.8783784,0.8378378,0.8333333,0.8648649,0.8828829,0.9099099,0.8738739,0.8738739,0.8423423,0.8468468,0.9324324,0.9189189,0.8648649,0.9054054,0.9054054,0.9189189,0.9054054,0.9144144,0.9279279,0.8198198,0.9144144,0.8243243,0.9009009,0.972973,1.0,0.981982,0.9459459,0.8963964,0.963964,0.9234234,0.9954955,1.0,0.9504505,0.9594595,1.0,1.0,1.0
+TabDPT-Turbo (default),0.045045044,0.22522523,0.12162162,0.25675675,0.3153153,0.21621622,0.25225225,0.3108108,0.22072072,0.3873874,0.3153153,0.36486486,0.4234234,0.45495495,,0.490991,0.6306306,0.6981982,0.7117117,0.6801802,0.6756757,0.7252252,0.545045,0.6981982,0.7387387,0.6576577,0.7522523,0.7432432,0.7837838,0.7117117,0.7702703,0.7207207,0.7477477,0.8198198,0.7612613,0.6801802,0.8648649,0.6801802,0.7702703,0.8468468,0.8333333,0.8783784,0.9099099,0.8333333,0.9099099,0.9144144,0.9189189,0.8378378,0.7927928,0.9414414,0.9279279,0.8198198,0.9414414,0.9144144,0.9324324,0.9414414,0.9459459,0.9009009,0.7837838,0.963964,0.7927928,0.9369369,0.981982,0.990991,0.990991,0.9864865,0.8873874,0.990991,0.9684685,1.0,1.0,0.972973,0.9774775,1.0,1.0,1.0
+RealMLP (tuned),0.054054055,0.2072072,0.17567568,0.1891892,0.1981982,0.18468468,0.22072072,0.25225225,0.3873874,0.26126125,0.4234234,0.34684685,0.527027,0.45495495,0.509009,,0.481982,0.6531532,0.6621622,0.6576577,0.6891892,0.6981982,0.5675676,0.5990991,0.6711712,0.5855856,0.7477477,0.6981982,0.7657658,0.7027027,0.7567568,0.7297297,0.7702703,0.7837838,0.7297297,0.6846847,0.8288288,0.7342342,0.7882883,0.8558559,0.8288288,0.8918919,0.8198198,0.8468468,0.8873874,0.9369369,0.8513514,0.8153153,0.8423423,0.9144144,0.9369369,0.8648649,0.9279279,0.9459459,0.9414414,0.9504505,0.8873874,0.9414414,0.7972973,0.972973,0.8063063,0.9054054,0.9774775,0.981982,0.990991,0.990991,0.8738739,0.990991,0.9504505,0.990991,1.0,0.963964,0.9774775,0.9954955,0.9954955,0.9954955
+"AutoGluon 1.4 (best, 4h)",0.04054054,0.17117117,0.15765765,0.13963965,0.1981982,0.15315315,0.2027027,0.24774775,0.22522523,0.3108108,0.27027026,0.35585585,0.4144144,0.3828829,0.36936936,0.518018,,0.6711712,0.6711712,0.7837838,0.6261261,0.6891892,0.5585586,0.6711712,0.6936937,0.6081081,0.8288288,0.7747748,0.8603604,0.6036036,0.8513514,0.6756757,0.7252252,0.7927928,0.8333333,0.6396396,0.8693694,0.6261261,0.8378378,0.7927928,0.8198198,0.7837838,0.8693694,0.9054054,0.9324324,0.9189189,0.8783784,0.8153153,0.8513514,0.963964,0.9279279,0.8783784,0.9144144,0.9684685,0.981982,0.9189189,0.9279279,0.9234234,0.7972973,0.963964,0.8063063,0.9189189,0.9954955,0.9954955,1.0,0.9774775,0.8783784,0.990991,0.9369369,1.0,1.0,0.963964,0.972973,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.036036037,0.10810811,0.14414415,0.13513513,0.12162162,0.12162162,0.1936937,0.08108108,0.2117117,0.21621622,0.23873875,0.3018018,0.2972973,0.2837838,0.3018018,0.34684685,0.32882884,,0.6396396,0.6036036,0.42792794,0.5135135,0.4954955,0.5990991,0.4864865,0.5405405,0.7657658,0.7972973,0.7297297,0.5855856,0.7837838,0.6036036,0.536036,0.6531532,0.7792793,0.6081081,0.7657658,0.6126126,0.8423423,0.6801802,0.7567568,0.7567568,0.8018018,0.9369369,0.7882883,0.8243243,0.8378378,0.7882883,0.8378378,0.9504505,0.8063063,0.8738739,0.9189189,0.963964,0.990991,0.9234234,0.8738739,0.9369369,0.7972973,0.981982,0.7972973,0.8918919,0.9369369,0.9279279,0.9684685,0.9864865,0.8693694,1.0,0.9459459,0.9954955,1.0,0.9684685,0.972973,1.0,1.0,1.0
+CatBoost (tuned),0.022522522,0.08108108,0.13963965,0.11261261,0.0990991,0.12612613,0.15315315,0.08558559,0.1936937,0.22072072,0.22072072,0.27027026,0.2837838,0.23873875,0.2882883,0.33783785,0.32882884,0.36036035,,0.5225225,0.4189189,0.490991,0.4864865,0.5630631,0.45495495,0.4954955,0.6756757,0.6846847,0.6846847,0.5675676,0.7297297,0.6036036,0.527027,0.6216216,0.7342342,0.5855856,0.7207207,0.6126126,0.8063063,0.6666667,0.7342342,0.7432432,0.7882883,0.9144144,0.7702703,0.8108108,0.8018018,0.7837838,0.7972973,0.9279279,0.7927928,0.8603604,0.8873874,0.954955,0.990991,0.9144144,0.8738739,0.9234234,0.7927928,0.972973,0.8018018,0.8738739,0.9369369,0.9234234,0.9594595,0.9774775,0.8603604,0.9954955,0.9369369,0.9864865,1.0,0.963964,0.9684685,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.027027028,0.08558559,0.11261261,0.11261261,0.12612613,0.12612613,0.14864865,0.09009009,0.18468468,0.21621622,0.21621622,0.2792793,0.3063063,0.3018018,0.3198198,0.34234235,0.21621622,0.3963964,0.4774775,,0.46846846,0.490991,0.4864865,0.5,0.481982,0.5405405,0.6756757,0.7117117,0.7792793,0.5540541,0.7252252,0.490991,0.5675676,0.5630631,0.7657658,0.5945946,0.6666667,0.6171171,0.8333333,0.7252252,0.7117117,0.7612613,0.8153153,0.8963964,0.7117117,0.7927928,0.8153153,0.7117117,0.8108108,0.981982,0.7432432,0.8378378,0.8918919,0.972973,0.9774775,0.8963964,0.8468468,0.9414414,0.7747748,0.972973,0.7792793,0.8873874,0.972973,0.8963964,0.990991,0.972973,0.8468468,1.0,0.9234234,0.990991,1.0,0.9459459,0.9684685,1.0,1.0,1.0
+TabM (tuned + ensembled),0.045045044,0.076576576,0.10810811,0.09009009,0.08558559,0.08558559,0.0945946,0.15765765,0.2072072,0.22072072,0.25675675,0.2072072,0.3918919,0.2792793,0.3243243,0.3108108,0.37387386,0.5720721,0.5810811,0.5315315,,0.5585586,0.46396396,0.509009,0.5495495,0.43243244,0.6261261,0.6486486,0.6396396,0.5945946,0.6711712,0.5990991,0.8603604,0.6441441,0.6576577,0.5945946,0.7117117,0.6126126,0.7072072,0.9054054,0.6441441,0.8378378,0.7342342,0.7882883,0.8198198,0.7522523,0.7522523,0.7162162,0.6936937,0.8333333,0.8648649,0.7522523,0.7612613,0.8333333,0.8828829,0.7702703,0.8378378,0.8063063,0.7792793,0.8513514,0.7882883,0.8153153,0.9414414,0.9684685,0.981982,0.9369369,0.8648649,0.9369369,0.9099099,0.963964,0.990991,0.9369369,0.9459459,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.036036037,0.0990991,0.1036036,0.0945946,0.12612613,0.12612613,0.16216215,0.14864865,0.117117114,0.14414415,0.16666667,0.25675675,0.22072072,0.2972973,0.27477476,0.3018018,0.3108108,0.4864865,0.509009,0.509009,0.44144145,,0.46396396,0.5,0.47297296,0.5225225,0.6351351,0.5765766,0.6306306,0.5630631,0.6666667,0.5855856,0.5540541,0.5990991,0.6441441,0.5855856,0.8198198,0.5810811,0.6846847,0.5855856,0.7162162,0.7837838,0.8378378,0.7297297,0.8468468,0.8198198,0.8468468,0.7882883,0.7297297,0.9144144,0.8558559,0.7612613,0.9369369,0.8603604,0.9054054,0.9414414,0.9369369,0.8153153,0.7882883,0.9324324,0.7882883,0.9099099,0.981982,0.972973,0.9774775,0.9774775,0.8603604,0.981982,0.9594595,0.990991,0.9864865,0.9684685,0.9864865,0.990991,0.990991,0.990991
+TabPFNv2 (tuned + ensembled),0.12612613,0.1036036,0.13513513,0.25675675,0.27027026,0.21621622,0.21621622,0.21621622,0.32882884,0.33783785,0.3828829,0.3018018,0.3918919,0.4144144,0.45495495,0.43243244,0.44144145,0.5045045,0.5135135,0.5135135,0.536036,0.536036,,0.5765766,0.5810811,0.5585586,0.5540541,0.5720721,0.5720721,0.6418919,0.5855856,0.6486486,0.5810811,0.6216216,0.6126126,0.7072072,0.6531532,0.75,0.6261261,0.6036036,0.6486486,0.7162162,0.6621622,0.6936937,0.6711712,0.6801802,0.6666667,0.7207207,0.7252252,0.7117117,0.7117117,0.7702703,0.7117117,0.7342342,0.7297297,0.7882883,0.7522523,0.8513514,0.7927928,0.7657658,0.7972973,0.8963964,0.9054054,0.8423423,0.9234234,0.8716216,0.8648649,0.9279279,0.972973,0.9504505,0.9279279,0.990991,0.990991,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.04054054,0.2072072,0.14414415,0.1981982,0.1891892,0.14864865,0.2117117,0.23873875,0.23423423,0.24774775,0.24774775,0.3153153,0.3108108,0.34234235,0.3018018,0.4009009,0.32882884,0.4009009,0.43693694,0.5,0.490991,0.5,0.4234234,,0.4774775,0.527027,0.5900901,0.6126126,0.5810811,0.490991,0.5990991,0.490991,0.5765766,0.5900901,0.6396396,0.527027,0.6216216,0.5540541,0.6981982,0.6126126,0.8738739,0.7252252,0.7522523,0.7252252,0.7477477,0.8783784,0.7702703,0.7297297,0.7117117,0.8018018,0.7477477,0.7477477,0.8558559,0.8693694,0.9054054,0.8828829,0.9234234,0.7927928,0.7657658,0.8963964,0.7792793,0.8963964,0.9279279,0.9099099,0.9369369,0.963964,0.8198198,0.954955,0.954955,0.981982,1.0,0.9774775,0.981982,0.9954955,0.9954955,1.0
+xRFM (tuned + ensembled),0.04954955,0.12162162,0.12162162,0.13513513,0.12162162,0.13963965,0.15315315,0.15765765,0.13063063,0.12162162,0.16216215,0.2072072,0.27027026,0.26576576,0.26126125,0.32882884,0.3063063,0.5135135,0.545045,0.518018,0.45045045,0.527027,0.4189189,0.5225225,,0.509009,0.6081081,0.6171171,0.6351351,0.6036036,0.6666667,0.5630631,0.5495495,0.7342342,0.6576577,0.5765766,0.7657658,0.6081081,0.6981982,0.6396396,0.6936937,0.6756757,0.7567568,0.7387387,0.7747748,0.8018018,0.7837838,0.8423423,0.7657658,0.8783784,0.7972973,0.7837838,0.8378378,0.8198198,0.9054054,0.8468468,0.972973,0.8153153,0.7657658,0.8513514,0.7702703,0.9234234,0.8783784,0.9369369,0.9054054,0.9774775,0.8198198,0.954955,0.9684685,0.972973,0.981982,0.981982,0.981982,0.9954955,0.9954955,0.9954955
+TabSwift (default),0.1036036,0.13063063,0.06306306,0.13513513,0.14414415,0.16666667,0.13963965,0.22972973,0.23873875,0.2792793,0.2837838,0.22972973,0.34234235,0.3108108,0.34234235,0.4144144,0.3918919,0.45945945,0.5045045,0.45945945,0.5675676,0.4774775,0.44144145,0.47297296,0.490991,,0.518018,0.5225225,0.5720721,0.6171171,0.5810811,0.5765766,0.6351351,0.6261261,0.5720721,0.5900901,0.6801802,0.6441441,0.6036036,0.6351351,0.6621622,0.8738739,0.7162162,0.6666667,0.7927928,0.8108108,0.7387387,0.7342342,0.7207207,0.7882883,0.8153153,0.7657658,0.7837838,0.7747748,0.8558559,0.7882883,0.8738739,0.8198198,0.7792793,0.8333333,0.7882883,0.8738739,0.963964,0.9459459,0.9684685,0.8963964,0.8558559,0.9369369,0.963964,1.0,0.954955,0.981982,0.990991,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.036036037,0.072072074,0.11261261,0.076576576,0.06306306,0.117117114,0.10810811,0.1036036,0.14414415,0.13513513,0.16666667,0.1891892,0.25225225,0.23423423,0.24774775,0.25225225,0.17117117,0.23423423,0.3243243,0.3243243,0.37387386,0.36486486,0.44594595,0.4099099,0.3918919,0.481982,,0.5405405,0.509009,0.527027,0.6396396,0.4864865,0.509009,0.5495495,0.6801802,0.5585586,0.6351351,0.5810811,0.7072072,0.6081081,0.6666667,0.7297297,0.7567568,0.8693694,0.7027027,0.7522523,0.7837838,0.7207207,0.7927928,0.972973,0.7477477,0.8243243,0.8738739,0.9234234,0.9459459,0.8918919,0.8378378,0.9234234,0.7702703,0.9594595,0.7837838,0.8603604,0.9459459,0.8873874,0.981982,0.981982,0.8378378,0.9864865,0.9279279,0.990991,1.0,0.963964,0.963964,1.0,1.0,1.0
+CatBoost (default),0.03153153,0.08108108,0.12612613,0.0990991,0.067567565,0.117117114,0.10810811,0.08558559,0.14864865,0.17117117,0.18468468,0.22972973,0.24774775,0.21621622,0.25675675,0.3018018,0.22522523,0.2027027,0.3153153,0.2882883,0.35135135,0.4234234,0.42792794,0.3873874,0.3828829,0.4774775,0.45945945,,0.46846846,0.5135135,0.509009,0.4234234,0.47297296,0.545045,0.6351351,0.5630631,0.5945946,0.5990991,0.6936937,0.5900901,0.6126126,0.6801802,0.7072072,0.8648649,0.6711712,0.6666667,0.7162162,0.6261261,0.8018018,0.7972973,0.7027027,0.8423423,0.8063063,0.9189189,0.8198198,0.8198198,0.7747748,0.9144144,0.7792793,0.9414414,0.7837838,0.8333333,0.9189189,0.8018018,0.9324324,0.9414414,0.8603604,1.0,0.8918919,0.9864865,1.0,0.9594595,0.9684685,1.0,1.0,1.0
+LightGBM (tuned),0.022522522,0.076576576,0.09009009,0.067567565,0.0945946,0.10810811,0.13063063,0.08558559,0.16216215,0.14864865,0.15765765,0.22072072,0.22972973,0.2027027,0.21621622,0.23423423,0.13963965,0.27027026,0.3153153,0.22072072,0.36036035,0.36936936,0.42792794,0.4189189,0.36486486,0.42792794,0.490991,0.5315315,,0.5,0.5495495,0.45495495,0.4864865,0.45945945,0.6261261,0.527027,0.5585586,0.5675676,0.6801802,0.6441441,0.6261261,0.7072072,0.7522523,0.7882883,0.6711712,0.7072072,0.7567568,0.6666667,0.7477477,0.9144144,0.6936937,0.8108108,0.8288288,0.8828829,0.9459459,0.8378378,0.8378378,0.8558559,0.7657658,0.9099099,0.7747748,0.8288288,0.9189189,0.8693694,0.954955,0.9504505,0.8288288,0.9864865,0.9099099,0.9774775,1.0,0.9414414,0.9459459,0.990991,0.990991,0.9954955
+Mitra (default),0.054054055,0.045045044,0.16666667,0.13963965,0.11261261,0.14864865,0.16666667,0.1036036,0.2072072,0.1981982,0.2027027,0.2792793,0.26576576,0.25675675,0.2882883,0.2972973,0.3963964,0.4144144,0.43243244,0.44594595,0.4054054,0.43693694,0.3581081,0.509009,0.3963964,0.3828829,0.47297296,0.4864865,0.5,,0.518018,0.5585586,0.4954955,0.490991,0.5675676,0.4527027,0.545045,0.4864865,0.5945946,0.5720721,0.6396396,0.6396396,0.6036036,0.6306306,0.7567568,0.7072072,0.6126126,0.6936937,0.7657658,0.6756757,0.7837838,0.8198198,0.6216216,0.7162162,0.7252252,0.7162162,0.7387387,0.8198198,0.8828829,0.7387387,0.9009009,0.7972973,0.9144144,0.8513514,0.9414414,0.8378378,0.9189189,0.9324324,0.9414414,0.9369369,0.9279279,0.972973,0.9864865,0.990991,0.990991,1.0
+XGBoost (tuned),0.027027028,0.06306306,0.117117114,0.06306306,0.058558557,0.12612613,0.1036036,0.09009009,0.12162162,0.13063063,0.15315315,0.1891892,0.22522523,0.2072072,0.22972973,0.24324325,0.14864865,0.21621622,0.27027026,0.27477476,0.32882884,0.33333334,0.4144144,0.4009009,0.33333334,0.4189189,0.36036035,0.490991,0.45045045,0.481982,,0.481982,0.46846846,0.509009,0.6036036,0.5405405,0.5675676,0.5675676,0.6396396,0.5585586,0.6306306,0.6981982,0.7297297,0.8243243,0.6936937,0.7207207,0.7522523,0.7117117,0.7522523,0.9099099,0.7387387,0.8198198,0.8288288,0.8918919,0.9279279,0.8378378,0.8288288,0.8963964,0.7702703,0.9234234,0.7702703,0.8468468,0.9369369,0.8963964,0.9684685,0.963964,0.8378378,0.9864865,0.9144144,0.981982,1.0,0.954955,0.963964,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.067567565,0.10810811,0.12612613,0.2027027,0.15315315,0.09009009,0.18468468,0.15315315,0.2027027,0.2072072,0.21621622,0.27027026,0.27027026,0.2882883,0.2792793,0.27027026,0.3243243,0.3963964,0.3963964,0.509009,0.4009009,0.4144144,0.35135135,0.509009,0.43693694,0.4234234,0.5135135,0.5765766,0.545045,0.44144145,0.518018,,0.4774775,0.481982,0.6306306,0.527027,0.5495495,0.5495495,0.6306306,0.46846846,0.6126126,0.5900901,0.6576577,0.7117117,0.5990991,0.6936937,0.6576577,0.6756757,0.7252252,0.7297297,0.5900901,0.7702703,0.7432432,0.7522523,0.7882883,0.7432432,0.7342342,0.8288288,0.7432432,0.7657658,0.7522523,0.8198198,0.8828829,0.8063063,0.8918919,0.8378378,0.8783784,0.9054054,0.9189189,0.9369369,0.9054054,0.9864865,0.9864865,0.9954955,0.9954955,1.0
+TabM (tuned),0.03153153,0.054054055,0.076576576,0.076576576,0.076576576,0.08108108,0.067567565,0.1036036,0.14864865,0.12612613,0.1891892,0.13963965,0.3063063,0.23423423,0.25225225,0.22972973,0.27477476,0.46396396,0.47297296,0.43243244,0.13963965,0.44594595,0.4189189,0.4234234,0.45045045,0.36486486,0.490991,0.527027,0.5135135,0.5045045,0.5315315,0.5225225,,0.5495495,0.5675676,0.5495495,0.6351351,0.5630631,0.5630631,0.7162162,0.5405405,0.7522523,0.6846847,0.6621622,0.7027027,0.6576577,0.7027027,0.6621622,0.6261261,0.7342342,0.7702703,0.6531532,0.6891892,0.7612613,0.8063063,0.7117117,0.7927928,0.7432432,0.7612613,0.7747748,0.7702703,0.7567568,0.8693694,0.9414414,0.9324324,0.8873874,0.8378378,0.8918919,0.8828829,0.9144144,0.981982,0.9144144,0.9369369,0.972973,0.972973,0.9954955
+xRFM (tuned),0.03153153,0.072072074,0.072072074,0.09009009,0.04954955,0.12612613,0.11261261,0.072072074,0.0990991,0.072072074,0.11261261,0.13963965,0.1891892,0.18018018,0.18018018,0.21621622,0.2072072,0.34684685,0.3783784,0.43693694,0.35585585,0.4009009,0.3783784,0.4099099,0.26576576,0.37387386,0.45045045,0.45495495,0.5405405,0.509009,0.490991,0.518018,0.45045045,,0.509009,0.5315315,0.6351351,0.5585586,0.6036036,0.5540541,0.6171171,0.6486486,0.6306306,0.6486486,0.7387387,0.7342342,0.6846847,0.7657658,0.7162162,0.7882883,0.7342342,0.7297297,0.7837838,0.7432432,0.8558559,0.7882883,0.8963964,0.7522523,0.7477477,0.7927928,0.7522523,0.8243243,0.8468468,0.9234234,0.8828829,0.9594595,0.8018018,0.9324324,0.9504505,0.9504505,0.990991,0.963964,0.972973,0.9864865,0.9864865,1.0
+ChimeraBoost (tuned + ensembled),0.027027028,0.058558557,0.12162162,0.076576576,0.054054055,0.13063063,0.1036036,0.072072074,0.15315315,0.1891892,0.17567568,0.2027027,0.22972973,0.1981982,0.23873875,0.27027026,0.16666667,0.22072072,0.26576576,0.23423423,0.34234235,0.35585585,0.3873874,0.36036035,0.34234235,0.42792794,0.3198198,0.36486486,0.37387386,0.43243244,0.3963964,0.36936936,0.43243244,0.490991,,0.5135135,0.509009,0.5315315,0.6396396,0.5810811,0.527027,0.6531532,0.6216216,0.8558559,0.6441441,0.5855856,0.6441441,0.5495495,0.7477477,0.7342342,0.6261261,0.7972973,0.7567568,0.8873874,0.7657658,0.7747748,0.6801802,0.8828829,0.7297297,0.8873874,0.7477477,0.7432432,0.8198198,0.7612613,0.8243243,0.9009009,0.8243243,0.9504505,0.8648649,0.9369369,0.9594595,0.9459459,0.9594595,0.9954955,0.9954955,0.9954955
+TabPFNv2 (tuned),0.0945946,0.054054055,0.072072074,0.12162162,0.14414415,0.117117114,0.076576576,0.117117114,0.2027027,0.2072072,0.24324325,0.13963965,0.2927928,0.26126125,0.3198198,0.3153153,0.36036035,0.3918919,0.4144144,0.4054054,0.4054054,0.4144144,0.2927928,0.47297296,0.4234234,0.4099099,0.44144145,0.43693694,0.47297296,0.5472973,0.45945945,0.47297296,0.45045045,0.46846846,0.4864865,,0.5135135,0.5382883,0.5225225,0.527027,0.5630631,0.5945946,0.527027,0.5630631,0.6081081,0.5810811,0.5315315,0.6621622,0.6846847,0.5855856,0.6531532,0.7342342,0.5585586,0.5900901,0.6396396,0.6441441,0.7072072,0.7612613,0.7567568,0.6396396,0.7567568,0.7657658,0.8198198,0.7927928,0.8648649,0.8130631,0.8018018,0.8693694,0.9009009,0.9144144,0.8963964,0.9369369,0.9459459,0.990991,0.990991,1.0
+iLTM (tuned),0.022522522,0.022522522,0.027027028,0.058558557,0.027027028,0.0990991,0.076576576,0.018018018,0.04954955,0.058558557,0.045045044,0.117117114,0.1036036,0.14864865,0.13513513,0.17117117,0.13063063,0.23423423,0.2792793,0.33333334,0.2882883,0.18018018,0.34684685,0.3783784,0.23423423,0.3198198,0.36486486,0.4054054,0.44144145,0.45495495,0.43243244,0.45045045,0.36486486,0.36486486,0.490991,0.4864865,,0.5,0.5405405,0.46396396,0.5495495,0.5945946,0.6576577,0.6036036,0.6216216,0.6351351,0.6846847,0.7027027,0.6621622,0.7522523,0.6936937,0.7072072,0.7792793,0.6801802,0.8018018,0.7792793,0.8693694,0.7252252,0.7342342,0.7477477,0.7297297,0.8288288,0.8693694,0.8828829,0.8873874,0.9369369,0.7882883,0.8963964,0.9054054,0.963964,0.981982,0.9459459,0.9414414,1.0,1.0,1.0
+TabPFNv2 (default),0.03153153,0.054054055,0.036036037,0.076576576,0.10810811,0.06306306,0.08558559,0.1036036,0.16666667,0.1036036,0.2117117,0.17117117,0.23423423,0.25675675,0.3198198,0.26576576,0.37387386,0.3873874,0.3873874,0.3828829,0.3873874,0.4189189,0.25,0.44594595,0.3918919,0.35585585,0.4189189,0.4009009,0.43243244,0.5135135,0.43243244,0.45045045,0.43693694,0.44144145,0.46846846,0.4617117,0.5,,0.4864865,0.4774775,0.5495495,0.5495495,0.5585586,0.5,0.5765766,0.5855856,0.5630631,0.6531532,0.6306306,0.545045,0.6216216,0.6846847,0.5675676,0.5855856,0.6036036,0.6576577,0.7072072,0.7027027,0.7342342,0.6306306,0.7342342,0.7837838,0.7522523,0.7657658,0.8063063,0.8243243,0.7747748,0.8153153,0.9144144,0.8738739,0.8603604,0.9369369,0.9459459,1.0,1.0,1.0
+ChimeraBoost (tuned),0.03153153,0.045045044,0.117117114,0.058558557,0.04054054,0.12612613,0.067567565,0.06306306,0.13063063,0.17567568,0.15315315,0.16666667,0.22072072,0.18468468,0.22972973,0.2117117,0.16216215,0.15765765,0.1936937,0.16666667,0.2927928,0.3153153,0.37387386,0.3018018,0.3018018,0.3963964,0.2927928,0.3063063,0.3198198,0.4054054,0.36036035,0.36936936,0.43693694,0.3963964,0.36036035,0.4774775,0.45945945,0.5135135,,0.545045,0.5,0.6351351,0.5990991,0.7702703,0.6081081,0.5675676,0.6216216,0.518018,0.6846847,0.6711712,0.5990991,0.7612613,0.7162162,0.8423423,0.7117117,0.7477477,0.6666667,0.8198198,0.7342342,0.8558559,0.7387387,0.7297297,0.7927928,0.7432432,0.8108108,0.8468468,0.8018018,0.9459459,0.8513514,0.9009009,0.9594595,0.9234234,0.9414414,0.990991,0.9864865,0.990991
+TabM (default),0.022522522,0.03153153,0.06306306,0.058558557,0.04054054,0.076576576,0.04054054,0.04054054,0.09009009,0.08558559,0.12612613,0.1036036,0.17567568,0.12162162,0.15315315,0.14414415,0.2072072,0.3198198,0.33333334,0.27477476,0.0945946,0.4144144,0.3963964,0.3873874,0.36036035,0.36486486,0.3918919,0.4099099,0.35585585,0.42792794,0.44144145,0.5315315,0.2837838,0.44594595,0.4189189,0.47297296,0.536036,0.5225225,0.45495495,,0.481982,0.6081081,0.6261261,0.545045,0.6666667,0.6171171,0.6351351,0.6351351,0.5900901,0.6666667,0.7387387,0.6441441,0.6441441,0.6891892,0.7657658,0.6531532,0.6936937,0.7297297,0.7567568,0.7162162,0.7657658,0.6891892,0.8423423,0.8378378,0.9189189,0.8423423,0.8333333,0.8873874,0.8378378,0.9054054,0.9459459,0.8783784,0.8873874,0.981982,0.981982,0.9954955
+ModernNCA (tuned),0.009009009,0.1036036,0.1036036,0.045045044,0.054054055,0.1036036,0.072072074,0.17117117,0.08108108,0.1036036,0.1036036,0.14864865,0.1936937,0.16216215,0.16666667,0.17117117,0.18018018,0.24324325,0.26576576,0.2882883,0.35585585,0.2837838,0.35135135,0.12612613,0.3063063,0.33783785,0.33333334,0.3873874,0.37387386,0.36036035,0.36936936,0.3873874,0.45945945,0.3828829,0.47297296,0.43693694,0.45045045,0.45045045,0.5,0.518018,,0.5810811,0.5630631,0.536036,0.6036036,0.7387387,0.5765766,0.5630631,0.6036036,0.6261261,0.6576577,0.6486486,0.7342342,0.6846847,0.7072072,0.7477477,0.8153153,0.7027027,0.7387387,0.7612613,0.7432432,0.7567568,0.8513514,0.8108108,0.8873874,0.9054054,0.7792793,0.8828829,0.9234234,0.9459459,0.963964,0.954955,0.963964,0.981982,0.981982,1.0
+RealMLP (default),0.009009009,0.036036037,0.03153153,0.022522522,0.027027028,0.054054055,0.04954955,0.06306306,0.054054055,0.013513514,0.072072074,0.08558559,0.13063063,0.16666667,0.12162162,0.10810811,0.21621622,0.24324325,0.25675675,0.23873875,0.16216215,0.21621622,0.2837838,0.27477476,0.3243243,0.12612613,0.27027026,0.3198198,0.2927928,0.36036035,0.3018018,0.4099099,0.24774775,0.35135135,0.34684685,0.4054054,0.4054054,0.45045045,0.36486486,0.3918919,0.4189189,,0.5900901,0.47297296,0.5630631,0.5630631,0.6081081,0.5540541,0.5630631,0.545045,0.6486486,0.5945946,0.6891892,0.5495495,0.6756757,0.7117117,0.6936937,0.6486486,0.6756757,0.6441441,0.6846847,0.7477477,0.8333333,0.8333333,0.8783784,0.8198198,0.8108108,0.8288288,0.8828829,0.9414414,0.9234234,0.9594595,0.972973,0.972973,0.972973,1.0
+ExtraTrees (tuned + ensembled),0.027027028,0.04054054,0.054054055,0.06306306,0.06306306,0.08108108,0.117117114,0.04954955,0.04954955,0.058558557,0.06306306,0.13963965,0.067567565,0.13513513,0.09009009,0.18018018,0.13063063,0.1981982,0.2117117,0.18468468,0.26576576,0.16216215,0.33783785,0.24774775,0.24324325,0.2837838,0.24324325,0.2927928,0.24774775,0.3963964,0.27027026,0.34234235,0.3153153,0.36936936,0.3783784,0.47297296,0.34234235,0.44144145,0.4009009,0.37387386,0.43693694,0.4099099,,0.5225225,0.4054054,0.45045045,0.704955,0.5540541,0.5630631,0.45045045,0.43693694,0.5945946,0.7657658,0.5990991,0.5765766,0.7882883,0.6486486,0.6441441,0.6891892,0.6621622,0.6936937,0.954955,0.7432432,0.5945946,0.7342342,0.9234234,0.8063063,0.7747748,0.972973,0.9369369,0.954955,0.990991,0.990991,1.0,1.0,1.0
+ChimeraBoost (default),0.0045045046,0.027027028,0.08108108,0.054054055,0.013513514,0.09009009,0.027027028,0.054054055,0.08108108,0.09009009,0.09009009,0.09009009,0.18018018,0.117117114,0.16666667,0.15315315,0.0945946,0.06306306,0.08558559,0.1036036,0.2117117,0.27027026,0.3063063,0.27477476,0.26126125,0.33333334,0.13063063,0.13513513,0.2117117,0.36936936,0.17567568,0.2882883,0.33783785,0.35135135,0.14414415,0.43693694,0.3963964,0.5,0.22972973,0.45495495,0.46396396,0.527027,0.4774775,,0.5900901,0.5135135,0.5,0.5045045,0.6036036,0.5405405,0.6081081,0.6846847,0.6666667,0.7477477,0.6576577,0.6756757,0.6306306,0.7387387,0.7117117,0.8108108,0.7162162,0.7207207,0.7162162,0.7162162,0.7477477,0.8378378,0.7837838,0.9369369,0.8378378,0.8918919,0.9504505,0.8963964,0.9234234,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.009009009,0.03153153,0.027027028,0.013513514,0.04054054,0.08558559,0.022522522,0.036036037,0.06306306,0.0945946,0.06306306,0.08558559,0.0990991,0.09009009,0.09009009,0.11261261,0.067567565,0.2117117,0.22972973,0.2882883,0.18018018,0.15315315,0.32882884,0.25225225,0.22522523,0.2072072,0.2972973,0.32882884,0.32882884,0.24324325,0.3063063,0.4009009,0.2972973,0.26126125,0.35585585,0.3918919,0.3783784,0.4234234,0.3918919,0.33333334,0.3963964,0.43693694,0.5945946,0.4099099,,0.518018,0.6126126,0.5,0.490991,0.5315315,0.7207207,0.5630631,0.6621622,0.4954955,0.5945946,0.6846847,0.7252252,0.6216216,0.7342342,0.5810811,0.7297297,0.6621622,0.8018018,0.9414414,0.8918919,0.7837838,0.8333333,0.8378378,0.8603604,0.8963964,0.8918919,0.8963964,0.9054054,0.9279279,0.9279279,1.0
+ModernNCA (default),0.009009009,0.04954955,0.03153153,0.036036037,0.013513514,0.0990991,0.054054055,0.08558559,0.04954955,0.03153153,0.054054055,0.0990991,0.08558559,0.12612613,0.08558559,0.06306306,0.08108108,0.17567568,0.1891892,0.2072072,0.24774775,0.18018018,0.3198198,0.12162162,0.1981982,0.1891892,0.24774775,0.33333334,0.2927928,0.2927928,0.2792793,0.3063063,0.34234235,0.26576576,0.4144144,0.4189189,0.36486486,0.4144144,0.43243244,0.3828829,0.26126125,0.43693694,0.5495495,0.4864865,0.481982,,0.5540541,0.490991,0.5225225,0.5,0.5405405,0.5405405,0.6891892,0.5630631,0.6126126,0.7342342,0.7837838,0.5990991,0.7162162,0.6756757,0.7162162,0.7522523,0.7747748,0.7567568,0.8243243,0.9144144,0.7927928,0.8558559,0.9189189,0.9189189,0.9594595,0.9414414,0.963964,0.9594595,0.9594595,1.0
+ExtraTrees (tuned),0.022522522,0.054054055,0.045045044,0.058558557,0.054054055,0.08558559,0.11261261,0.04954955,0.058558557,0.04954955,0.067567565,0.13513513,0.06306306,0.12612613,0.08108108,0.14864865,0.12162162,0.16216215,0.1981982,0.18468468,0.24774775,0.15315315,0.33333334,0.22972973,0.21621622,0.26126125,0.21621622,0.2837838,0.24324325,0.3873874,0.24774775,0.34234235,0.2972973,0.3153153,0.35585585,0.46846846,0.3153153,0.43693694,0.3783784,0.36486486,0.4234234,0.3918919,0.29504505,0.5,0.3873874,0.44594595,,0.5495495,0.5405405,0.42792794,0.42792794,0.5720721,0.6936937,0.5900901,0.5315315,0.7342342,0.6351351,0.6261261,0.6846847,0.6441441,0.6936937,0.9099099,0.7072072,0.5765766,0.7252252,0.9099099,0.8018018,0.7837838,0.9684685,0.9234234,0.954955,0.9954955,0.990991,1.0,1.0,1.0
+LimiX (default),0.013513514,0.0945946,0.072072074,0.13963965,0.058558557,0.067567565,0.08558559,0.0945946,0.117117114,0.08558559,0.117117114,0.13063063,0.14414415,0.15765765,0.16216215,0.18468468,0.18468468,0.2117117,0.21621622,0.2882883,0.2837838,0.2117117,0.2792793,0.27027026,0.15765765,0.26576576,0.2792793,0.37387386,0.33333334,0.3063063,0.2882883,0.3243243,0.33783785,0.23423423,0.45045045,0.33783785,0.2972973,0.34684685,0.481982,0.36486486,0.43693694,0.44594595,0.44594595,0.4954955,0.5,0.509009,0.45045045,,0.5585586,0.5,0.5315315,0.5945946,0.5405405,0.5585586,0.5720721,0.5720721,0.6171171,0.6036036,0.7297297,0.5720721,0.7297297,0.5540541,0.7162162,0.6576577,0.7432432,0.6891892,0.7657658,0.7567568,0.8153153,0.7477477,0.7522523,0.8693694,0.9504505,0.9324324,0.9324324,0.9864865
+EBM (tuned + ensembled),0.027027028,0.022522522,0.0990991,0.036036037,0.027027028,0.09009009,0.045045044,0.058558557,0.11261261,0.12612613,0.13963965,0.117117114,0.1891892,0.15315315,0.2072072,0.15765765,0.14864865,0.16216215,0.2027027,0.1891892,0.3063063,0.27027026,0.27477476,0.2882883,0.23423423,0.2792793,0.2072072,0.1981982,0.25225225,0.23423423,0.24774775,0.27477476,0.37387386,0.2837838,0.25225225,0.3153153,0.33783785,0.36936936,0.3153153,0.4099099,0.3963964,0.43693694,0.43693694,0.3963964,0.509009,0.4774775,0.45945945,0.44144145,,0.44144145,0.5540541,0.8063063,0.518018,0.6351351,0.5045045,0.5540541,0.5585586,0.8603604,0.6981982,0.6531532,0.7297297,0.536036,0.6396396,0.7072072,0.7027027,0.6711712,0.8468468,0.8513514,0.7972973,0.8738739,0.8243243,0.8288288,0.8693694,0.9864865,0.9864865,0.9954955
+LightGBM (default),0.0,0.009009009,0.009009009,0.022522522,0.0045045046,0.04954955,0.013513514,0.0,0.022522522,0.013513514,0.022522522,0.04054054,0.058558557,0.067567565,0.058558557,0.08558559,0.036036037,0.04954955,0.072072074,0.018018018,0.16666667,0.08558559,0.2882883,0.1981982,0.12162162,0.2117117,0.027027028,0.2027027,0.08558559,0.3243243,0.09009009,0.27027026,0.26576576,0.2117117,0.26576576,0.4144144,0.24774775,0.45495495,0.32882884,0.33333334,0.37387386,0.45495495,0.5495495,0.45945945,0.46846846,0.5,0.5720721,0.5,0.5585586,,0.5135135,0.6171171,0.6081081,0.5945946,0.7117117,0.6531532,0.7117117,0.6711712,0.7477477,0.6576577,0.7477477,0.7477477,0.7792793,0.7567568,0.8333333,0.8828829,0.7792793,0.8918919,0.8558559,0.9279279,0.990991,0.9009009,0.9189189,0.9864865,0.9864865,0.9954955
+TorchMLP (tuned),0.0045045046,0.018018018,0.022522522,0.013513514,0.013513514,0.0945946,0.018018018,0.013513514,0.04054054,0.06306306,0.04954955,0.072072074,0.072072074,0.08108108,0.072072074,0.06306306,0.072072074,0.1936937,0.2072072,0.25675675,0.13513513,0.14414415,0.2882883,0.25225225,0.2027027,0.18468468,0.25225225,0.2972973,0.3063063,0.21621622,0.26126125,0.4099099,0.22972973,0.26576576,0.37387386,0.34684685,0.3063063,0.3783784,0.4009009,0.26126125,0.34234235,0.35135135,0.5630631,0.3918919,0.2792793,0.45945945,0.5720721,0.46846846,0.44594595,0.4864865,,0.481982,0.6081081,0.46846846,0.527027,0.6216216,0.6666667,0.5855856,0.7297297,0.545045,0.7297297,0.6261261,0.7027027,0.8783784,0.8108108,0.7207207,0.8288288,0.7747748,0.8378378,0.8333333,0.8783784,0.8873874,0.8828829,0.9009009,0.9009009,1.0
+EBM (tuned),0.018018018,0.013513514,0.08558559,0.036036037,0.018018018,0.0945946,0.03153153,0.067567565,0.1036036,0.0990991,0.1036036,0.0945946,0.17567568,0.13513513,0.18018018,0.13513513,0.12162162,0.12612613,0.13963965,0.16216215,0.24774775,0.23873875,0.22972973,0.25225225,0.21621622,0.23423423,0.17567568,0.15765765,0.1891892,0.18018018,0.18018018,0.22972973,0.34684685,0.27027026,0.2027027,0.26576576,0.2927928,0.3153153,0.23873875,0.35585585,0.35135135,0.4054054,0.4054054,0.3153153,0.43693694,0.45945945,0.42792794,0.4054054,0.1936937,0.3828829,0.518018,,0.46846846,0.5720721,0.45945945,0.5135135,0.5405405,0.7162162,0.6801802,0.6171171,0.7072072,0.5135135,0.5945946,0.6666667,0.6711712,0.6261261,0.8378378,0.8243243,0.7702703,0.8468468,0.8018018,0.8198198,0.8468468,0.981982,0.981982,0.9954955
+RandomForest (tuned + ensembled),0.0045045046,0.018018018,0.022522522,0.03153153,0.009009009,0.06306306,0.045045044,0.009009009,0.018018018,0.018018018,0.03153153,0.1036036,0.027027028,0.0945946,0.058558557,0.072072074,0.08558559,0.08108108,0.11261261,0.10810811,0.23873875,0.06306306,0.2882883,0.14414415,0.16216215,0.21621622,0.12612613,0.1936937,0.17117117,0.3783784,0.17117117,0.25675675,0.3108108,0.21621622,0.24324325,0.44144145,0.22072072,0.43243244,0.2837838,0.35585585,0.26576576,0.3108108,0.23423423,0.33333334,0.33783785,0.3108108,0.3063063,0.45945945,0.481982,0.3918919,0.3918919,0.5315315,,0.518018,0.46396396,0.6824324,0.5405405,0.5945946,0.6351351,0.5855856,0.6486486,0.7207207,0.6171171,0.5675676,0.6666667,0.9864865,0.7612613,0.7747748,0.9234234,0.9054054,0.9459459,0.9504505,0.9684685,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.009009009,0.013513514,0.04954955,0.009009009,0.0045045046,0.067567565,0.013513514,0.04954955,0.018018018,0.022522522,0.027027028,0.045045044,0.08558559,0.0945946,0.08558559,0.054054055,0.03153153,0.036036037,0.045045044,0.027027028,0.16666667,0.13963965,0.26576576,0.13063063,0.18018018,0.22522523,0.076576576,0.08108108,0.117117114,0.2837838,0.10810811,0.24774775,0.23873875,0.25675675,0.11261261,0.4099099,0.3198198,0.4144144,0.15765765,0.3108108,0.3153153,0.45045045,0.4009009,0.25225225,0.5045045,0.43693694,0.4099099,0.44144145,0.36486486,0.4054054,0.5315315,0.42792794,0.481982,,0.545045,0.5135135,0.5990991,0.5315315,0.5945946,0.8108108,0.5945946,0.5900901,0.6891892,0.6486486,0.7162162,0.7072072,0.6711712,0.963964,0.7657658,0.8603604,0.8828829,0.8963964,0.9144144,0.981982,0.981982,0.9954955
+XGBoost (default),0.0,0.027027028,0.04954955,0.0045045046,0.0,0.08108108,0.027027028,0.045045044,0.04054054,0.03153153,0.04954955,0.076576576,0.054054055,0.08108108,0.067567565,0.058558557,0.018018018,0.009009009,0.009009009,0.022522522,0.117117114,0.0945946,0.27027026,0.0945946,0.0945946,0.14414415,0.054054055,0.18018018,0.054054055,0.27477476,0.072072074,0.2117117,0.1936937,0.14414415,0.23423423,0.36036035,0.1981982,0.3963964,0.2882883,0.23423423,0.2927928,0.3243243,0.4234234,0.34234235,0.4054054,0.3873874,0.46846846,0.42792794,0.4954955,0.2882883,0.47297296,0.5405405,0.536036,0.45495495,,0.5810811,0.6846847,0.5720721,0.7072072,0.6036036,0.7072072,0.6576577,0.6486486,0.6171171,0.7342342,0.8423423,0.7477477,0.8018018,0.8423423,0.8153153,0.9504505,0.9009009,0.9189189,0.9504505,0.9504505,0.9864865
+RandomForest (tuned),0.0045045046,0.018018018,0.022522522,0.027027028,0.009009009,0.054054055,0.036036037,0.009009009,0.013513514,0.013513514,0.027027028,0.09009009,0.022522522,0.0945946,0.058558557,0.04954955,0.08108108,0.076576576,0.08558559,0.1036036,0.22972973,0.058558557,0.2117117,0.117117114,0.15315315,0.2117117,0.10810811,0.18018018,0.16216215,0.2837838,0.16216215,0.25675675,0.2882883,0.2117117,0.22522523,0.35585585,0.22072072,0.34234235,0.25225225,0.34684685,0.25225225,0.2882883,0.2117117,0.3243243,0.3153153,0.26576576,0.26576576,0.42792794,0.44594595,0.34684685,0.3783784,0.4864865,0.31756756,0.4864865,0.4189189,,0.5045045,0.5585586,0.5810811,0.536036,0.5990991,0.6216216,0.5810811,0.5495495,0.6351351,0.7747748,0.7567568,0.7522523,0.8963964,0.9009009,0.9459459,0.9504505,0.9594595,1.0,1.0,1.0
+xRFM (default),0.009009009,0.03153153,0.036036037,0.027027028,0.022522522,0.036036037,0.067567565,0.03153153,0.04054054,0.03153153,0.036036037,0.0990991,0.045045044,0.08558559,0.054054055,0.11261261,0.072072074,0.12612613,0.12612613,0.15315315,0.16216215,0.06306306,0.24774775,0.076576576,0.027027028,0.12612613,0.16216215,0.22522523,0.16216215,0.26126125,0.17117117,0.26576576,0.2072072,0.1036036,0.3198198,0.2927928,0.13063063,0.2927928,0.33333334,0.3063063,0.18468468,0.3063063,0.35135135,0.36936936,0.27477476,0.21621622,0.36486486,0.3828829,0.44144145,0.2882883,0.33333334,0.45945945,0.45945945,0.4009009,0.3153153,0.4954955,,0.5315315,0.6351351,0.44594595,0.6351351,0.509009,0.5135135,0.5135135,0.5585586,0.6441441,0.7207207,0.6216216,0.8738739,0.7252252,0.8198198,0.9234234,0.9774775,0.9234234,0.9144144,0.990991
+EBM (default),0.013513514,0.009009009,0.058558557,0.045045044,0.018018018,0.058558557,0.027027028,0.054054055,0.04954955,0.027027028,0.058558557,0.054054055,0.09009009,0.072072074,0.0990991,0.058558557,0.076576576,0.06306306,0.076576576,0.058558557,0.1936937,0.18468468,0.14864865,0.2072072,0.18468468,0.18018018,0.076576576,0.08558559,0.14414415,0.18018018,0.1036036,0.17117117,0.25675675,0.24774775,0.117117114,0.23873875,0.27477476,0.2972973,0.18018018,0.27027026,0.2972973,0.35135135,0.35585585,0.26126125,0.3783784,0.4009009,0.37387386,0.3963964,0.13963965,0.32882884,0.4144144,0.2837838,0.4054054,0.46846846,0.42792794,0.44144145,0.46846846,,0.6261261,0.527027,0.6306306,0.4954955,0.5540541,0.6486486,0.6126126,0.5855856,0.7837838,0.7837838,0.7657658,0.7927928,0.7837838,0.7927928,0.8603604,0.9864865,0.9864865,0.990991
+TabSTAR (tuned + ensembled),0.076576576,0.08558559,0.13963965,0.1036036,0.0990991,0.13063063,0.13513513,0.0990991,0.17117117,0.18468468,0.18018018,0.1936937,0.17117117,0.18018018,0.21621622,0.2027027,0.2027027,0.2027027,0.2072072,0.22522523,0.22072072,0.2117117,0.2072072,0.23423423,0.23423423,0.22072072,0.22972973,0.22072072,0.23423423,0.117117114,0.22972973,0.25675675,0.23873875,0.25225225,0.27027026,0.24324325,0.26576576,0.26576576,0.26576576,0.24324325,0.26126125,0.3243243,0.3108108,0.2882883,0.26576576,0.2837838,0.3153153,0.27027026,0.3018018,0.25225225,0.27027026,0.3198198,0.36486486,0.4054054,0.2927928,0.4189189,0.36486486,0.37387386,,0.44594595,0.6306306,0.46846846,0.4054054,0.3918919,0.4054054,0.4234234,0.981982,0.4774775,0.6801802,0.5990991,0.5540541,0.7567568,0.7882883,0.8288288,0.8288288,0.9144144
+PerpetualBooster (tuned),0.0045045046,0.009009009,0.018018018,0.0045045046,0.0045045046,0.067567565,0.013513514,0.027027028,0.013513514,0.013513514,0.022522522,0.036036037,0.027027028,0.08558559,0.036036037,0.027027028,0.036036037,0.018018018,0.027027028,0.027027028,0.14864865,0.067567565,0.23423423,0.1036036,0.14864865,0.16666667,0.04054054,0.058558557,0.09009009,0.26126125,0.076576576,0.23423423,0.22522523,0.2072072,0.11261261,0.36036035,0.25225225,0.36936936,0.14414415,0.2837838,0.23873875,0.35585585,0.33783785,0.1891892,0.4189189,0.3243243,0.35585585,0.42792794,0.34684685,0.34234235,0.45495495,0.3828829,0.4144144,0.1891892,0.3963964,0.46396396,0.5540541,0.47297296,0.5540541,,0.5540541,0.545045,0.6396396,0.6171171,0.6711712,0.6711712,0.6396396,0.8265766,0.7207207,0.7927928,0.8468468,0.8738739,0.8963964,0.9774775,0.9774775,0.990991
+TabSTAR (tuned),0.072072074,0.076576576,0.12162162,0.0990991,0.09009009,0.117117114,0.12162162,0.0990991,0.17117117,0.18018018,0.17567568,0.1891892,0.17567568,0.17567568,0.2072072,0.1936937,0.1936937,0.2027027,0.1981982,0.22072072,0.2117117,0.2117117,0.2027027,0.22072072,0.22972973,0.2117117,0.21621622,0.21621622,0.22522523,0.0990991,0.22972973,0.24774775,0.22972973,0.24774775,0.25225225,0.24324325,0.27027026,0.26576576,0.26126125,0.23423423,0.25675675,0.3153153,0.3063063,0.2837838,0.27027026,0.2837838,0.3063063,0.27027026,0.27027026,0.25225225,0.27027026,0.2927928,0.35135135,0.4054054,0.2927928,0.4009009,0.36486486,0.36936936,0.36936936,0.44594595,,0.44594595,0.4054054,0.3873874,0.4009009,0.4054054,0.9684685,0.4774775,0.6801802,0.5990991,0.5540541,0.7522523,0.7882883,0.8288288,0.8288288,0.8963964
+ExtraTrees (default),0.018018018,0.036036037,0.03153153,0.04954955,0.036036037,0.04054054,0.045045044,0.022522522,0.036036037,0.022522522,0.04054054,0.09009009,0.045045044,0.0990991,0.06306306,0.0945946,0.08108108,0.10810811,0.12612613,0.11261261,0.18468468,0.09009009,0.1036036,0.1036036,0.076576576,0.12612613,0.13963965,0.16666667,0.17117117,0.2027027,0.15315315,0.18018018,0.24324325,0.17567568,0.25675675,0.23423423,0.17117117,0.21621622,0.27027026,0.3108108,0.24324325,0.25225225,0.045045044,0.2792793,0.33783785,0.24774775,0.09009009,0.44594595,0.46396396,0.25225225,0.37387386,0.4864865,0.2792793,0.4099099,0.34234235,0.3783784,0.490991,0.5045045,0.5315315,0.45495495,0.5540541,,0.47297296,0.4054054,0.4774775,0.6396396,0.6981982,0.6351351,0.8783784,0.7657658,0.8423423,0.9234234,0.981982,0.9954955,0.9954955,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0045045046,0.0,0.0045045046,0.018018018,0.04054054,0.0,0.013513514,0.0,0.0,0.0045045046,0.022522522,0.013513514,0.027027028,0.018018018,0.022522522,0.0045045046,0.06306306,0.06306306,0.027027028,0.058558557,0.018018018,0.0945946,0.072072074,0.12162162,0.036036037,0.054054055,0.08108108,0.08108108,0.08558559,0.06306306,0.117117114,0.13063063,0.15315315,0.18018018,0.18018018,0.13063063,0.24774775,0.2072072,0.15765765,0.14864865,0.16666667,0.25675675,0.2837838,0.1981982,0.22522523,0.2927928,0.2837838,0.36036035,0.22072072,0.2972973,0.4054054,0.3828829,0.3108108,0.35135135,0.4189189,0.4864865,0.44594595,0.5945946,0.36036035,0.5945946,0.527027,,0.518018,0.6216216,0.5720721,0.6891892,0.6396396,0.7342342,0.9504505,0.8288288,0.8018018,0.8873874,0.9774775,0.9774775,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.04054054,0.0,0.0,0.0,0.0045045046,0.0,0.0045045046,0.009009009,0.0,0.009009009,0.018018018,0.0045045046,0.072072074,0.076576576,0.1036036,0.03153153,0.027027028,0.15765765,0.09009009,0.06306306,0.054054055,0.11261261,0.1981982,0.13063063,0.14864865,0.1036036,0.1936937,0.058558557,0.076576576,0.23873875,0.2072072,0.117117114,0.23423423,0.25675675,0.16216215,0.1891892,0.16666667,0.4054054,0.2837838,0.058558557,0.24324325,0.4234234,0.34234235,0.2927928,0.24324325,0.12162162,0.33333334,0.43243244,0.35135135,0.3828829,0.45045045,0.4864865,0.35135135,0.6081081,0.3828829,0.6126126,0.5945946,0.481982,,0.5405405,0.6441441,0.7072072,0.5405405,0.7747748,0.6351351,0.7657658,0.8513514,0.8603604,0.8873874,0.8873874,0.9864865
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.009009009,0.04054054,0.0,0.009009009,0.0,0.0045045046,0.0,0.0,0.013513514,0.018018018,0.009009009,0.009009009,0.0,0.03153153,0.04054054,0.009009009,0.018018018,0.022522522,0.076576576,0.06306306,0.0945946,0.03153153,0.018018018,0.067567565,0.045045044,0.058558557,0.03153153,0.10810811,0.067567565,0.117117114,0.17567568,0.13513513,0.11261261,0.1936937,0.1891892,0.08108108,0.11261261,0.12162162,0.26576576,0.25225225,0.10810811,0.17567568,0.27477476,0.25675675,0.2972973,0.16666667,0.1891892,0.32882884,0.33333334,0.2837838,0.26576576,0.36486486,0.44144145,0.3873874,0.5945946,0.32882884,0.5990991,0.5225225,0.3783784,0.45945945,,0.5585586,0.6936937,0.5720721,0.7072072,0.8198198,0.7927928,0.7702703,0.8738739,0.954955,0.9504505,1.0
+RandomForest (default),0.0045045046,0.013513514,0.0045045046,0.0,0.0045045046,0.027027028,0.009009009,0.0045045046,0.0,0.0045045046,0.009009009,0.036036037,0.0045045046,0.054054055,0.013513514,0.009009009,0.022522522,0.013513514,0.022522522,0.027027028,0.06306306,0.022522522,0.12837838,0.036036037,0.022522522,0.1036036,0.018018018,0.058558557,0.04954955,0.16216215,0.036036037,0.16216215,0.11261261,0.04054054,0.0990991,0.18693693,0.06306306,0.17567568,0.15315315,0.15765765,0.0945946,0.18018018,0.076576576,0.16216215,0.21621622,0.08558559,0.09009009,0.3108108,0.32882884,0.117117114,0.2792793,0.37387386,0.013513514,0.2927928,0.15765765,0.22522523,0.35585585,0.4144144,0.5765766,0.32882884,0.5945946,0.36036035,0.42792794,0.35585585,0.44144145,,0.6936937,0.5855856,0.7972973,0.6801802,0.7567568,0.8738739,0.9189189,0.954955,0.954955,0.981982
+TabSTAR (default),0.058558557,0.072072074,0.072072074,0.067567565,0.06306306,0.067567565,0.08108108,0.076576576,0.08558559,0.0990991,0.08108108,0.10810811,0.076576576,0.1036036,0.11261261,0.12612613,0.12162162,0.13063063,0.13963965,0.15315315,0.13513513,0.13963965,0.13513513,0.18018018,0.18018018,0.14414415,0.16216215,0.13963965,0.17117117,0.08108108,0.16216215,0.12162162,0.16216215,0.1981982,0.17567568,0.1981982,0.2117117,0.22522523,0.1981982,0.16666667,0.22072072,0.1891892,0.1936937,0.21621622,0.16666667,0.2072072,0.1981982,0.23423423,0.15315315,0.22072072,0.17117117,0.16216215,0.23873875,0.32882884,0.25225225,0.24324325,0.2792793,0.21621622,0.018018018,0.36036035,0.03153153,0.3018018,0.3108108,0.2927928,0.3063063,0.3063063,,0.3783784,0.5855856,0.5675676,0.44144145,0.6531532,0.7117117,0.8243243,0.8243243,0.8333333
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.04054054,0.0,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.036036037,0.009009009,0.009009009,0.009009009,0.0,0.0045045046,0.0,0.06306306,0.018018018,0.072072074,0.045045044,0.045045044,0.06306306,0.013513514,0.0,0.013513514,0.067567565,0.013513514,0.0945946,0.10810811,0.067567565,0.04954955,0.13063063,0.1036036,0.18468468,0.054054055,0.11261261,0.117117114,0.17117117,0.22522523,0.06306306,0.16216215,0.14414415,0.21621622,0.24324325,0.14864865,0.10810811,0.22522523,0.17567568,0.22522523,0.036036037,0.1981982,0.24774775,0.3783784,0.21621622,0.5225225,0.17342342,0.5225225,0.36486486,0.36036035,0.45945945,0.42792794,0.4144144,0.6216216,,0.5900901,0.5945946,0.6486486,0.7117117,0.8513514,0.9144144,0.9144144,0.972973
+KNN (tuned + ensembled),0.013513514,0.018018018,0.03153153,0.022522522,0.013513514,0.018018018,0.03153153,0.022522522,0.018018018,0.03153153,0.018018018,0.067567565,0.018018018,0.076576576,0.03153153,0.04954955,0.06306306,0.054054055,0.06306306,0.076576576,0.09009009,0.04054054,0.027027028,0.045045044,0.03153153,0.036036037,0.072072074,0.10810811,0.09009009,0.058558557,0.08558559,0.08108108,0.117117114,0.04954955,0.13513513,0.0990991,0.0945946,0.08558559,0.14864865,0.16216215,0.076576576,0.117117114,0.027027028,0.16216215,0.13963965,0.08108108,0.03153153,0.18468468,0.2027027,0.14414415,0.16216215,0.22972973,0.076576576,0.23423423,0.15765765,0.1036036,0.12612613,0.23423423,0.3198198,0.2792793,0.3198198,0.12162162,0.26576576,0.22522523,0.2927928,0.2027027,0.4144144,0.4099099,,0.44144145,0.33333334,0.9324324,0.972973,0.8783784,0.8783784,0.972973
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0045045046,0.04054054,0.0,0.0,0.0,0.0,0.0,0.0045045046,0.0,0.0045045046,0.0,0.009009009,0.0,0.0045045046,0.013513514,0.009009009,0.036036037,0.009009009,0.04954955,0.018018018,0.027027028,0.0,0.009009009,0.013513514,0.022522522,0.06306306,0.018018018,0.06306306,0.08558559,0.04954955,0.06306306,0.08558559,0.036036037,0.12612613,0.0990991,0.0945946,0.054054055,0.058558557,0.06306306,0.10810811,0.1036036,0.08108108,0.076576576,0.25225225,0.12612613,0.072072074,0.16666667,0.15315315,0.0945946,0.13963965,0.18468468,0.0990991,0.27477476,0.2072072,0.4009009,0.2072072,0.4009009,0.23423423,0.04954955,0.36486486,0.18018018,0.3198198,0.43243244,0.4054054,0.5585586,,0.45045045,0.7207207,0.7972973,0.9414414,0.9414414,0.9954955
+iLTM (default),0.0,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.009009009,0.013513514,0.072072074,0.0,0.018018018,0.045045044,0.0,0.0,0.0,0.072072074,0.0,0.0945946,0.018018018,0.009009009,0.04054054,0.1036036,0.018018018,0.13963965,0.04054054,0.054054055,0.036036037,0.076576576,0.045045044,0.04954955,0.10810811,0.04054054,0.045045044,0.24774775,0.17567568,0.009009009,0.12162162,0.1981982,0.054054055,0.117117114,0.04954955,0.054054055,0.18018018,0.21621622,0.44594595,0.15315315,0.44594595,0.15765765,0.17117117,0.23423423,0.2072072,0.24324325,0.5585586,0.35135135,0.6666667,0.5495495,,0.7612613,0.7927928,0.8243243,0.8198198,0.8783784
+KNN (tuned),0.0045045046,0.009009009,0.018018018,0.009009009,0.013513514,0.009009009,0.022522522,0.013513514,0.009009009,0.018018018,0.009009009,0.027027028,0.009009009,0.04954955,0.027027028,0.036036037,0.036036037,0.03153153,0.036036037,0.054054055,0.06306306,0.03153153,0.009009009,0.022522522,0.018018018,0.018018018,0.036036037,0.04054054,0.058558557,0.027027028,0.045045044,0.013513514,0.08558559,0.036036037,0.054054055,0.06306306,0.054054055,0.06306306,0.076576576,0.12162162,0.045045044,0.04054054,0.009009009,0.1036036,0.1036036,0.058558557,0.0045045046,0.13063063,0.17117117,0.0990991,0.11261261,0.18018018,0.04954955,0.1036036,0.0990991,0.04954955,0.076576576,0.2072072,0.24324325,0.12612613,0.24774775,0.076576576,0.1981982,0.14864865,0.22972973,0.12612613,0.34684685,0.2882883,0.067567565,0.2792793,0.23873875,,0.9189189,0.8603604,0.8603604,0.9144144
+KNN (default),0.009009009,0.018018018,0.009009009,0.009009009,0.009009009,0.009009009,0.013513514,0.018018018,0.009009009,0.009009009,0.009009009,0.03153153,0.009009009,0.04054054,0.022522522,0.022522522,0.027027028,0.027027028,0.03153153,0.03153153,0.054054055,0.013513514,0.009009009,0.018018018,0.018018018,0.009009009,0.036036037,0.03153153,0.054054055,0.013513514,0.036036037,0.013513514,0.06306306,0.027027028,0.04054054,0.054054055,0.058558557,0.054054055,0.058558557,0.11261261,0.036036037,0.027027028,0.009009009,0.076576576,0.0945946,0.036036037,0.009009009,0.04954955,0.13063063,0.08108108,0.117117114,0.15315315,0.03153153,0.08558559,0.08108108,0.04054054,0.022522522,0.13963965,0.2117117,0.1036036,0.2117117,0.018018018,0.11261261,0.13963965,0.12612613,0.08108108,0.2882883,0.14864865,0.027027028,0.2027027,0.2072072,0.08108108,,0.6306306,0.6351351,0.6891892
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.009009009,0.0,0.0045045046,0.0045045046,0.0,0.0,0.0,0.009009009,0.009009009,0.0,0.0045045046,0.027027028,0.013513514,0.0045045046,0.009009009,0.0,0.0,0.009009009,0.018018018,0.018018018,0.027027028,0.0,0.0,0.072072074,0.04054054,0.0,0.067567565,0.013513514,0.013513514,0.0990991,0.018018018,0.0,0.018018018,0.04954955,0.0,0.076576576,0.013513514,0.17117117,0.022522522,0.17117117,0.0045045046,0.022522522,0.11261261,0.045045044,0.045045044,0.17567568,0.08558559,0.12162162,0.058558557,0.17567568,0.13963965,0.36936936,,0.6013514,0.8243243
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.009009009,0.0,0.0045045046,0.0045045046,0.0,0.0,0.0,0.009009009,0.009009009,0.0,0.0045045046,0.027027028,0.013513514,0.0045045046,0.009009009,0.0,0.0,0.013513514,0.018018018,0.018018018,0.027027028,0.0,0.0,0.072072074,0.04054054,0.0,0.067567565,0.013513514,0.013513514,0.0990991,0.018018018,0.0,0.018018018,0.04954955,0.0,0.08558559,0.013513514,0.17117117,0.022522522,0.17117117,0.0045045046,0.022522522,0.11261261,0.04954955,0.045045044,0.17567568,0.08558559,0.12162162,0.058558557,0.18018018,0.13963965,0.36486486,0.39864865,,0.7972973
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.009009009,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.0045045046,0.0,0.0,0.0,0.0045045046,0.0,0.0045045046,0.0,0.0,0.0,0.009009009,0.0045045046,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.013513514,0.0045045046,0.0045045046,0.0,0.0045045046,0.0,0.0045045046,0.013513514,0.0,0.009009009,0.009009009,0.08558559,0.009009009,0.1036036,0.0,0.0,0.013513514,0.0,0.018018018,0.16666667,0.027027028,0.027027028,0.0045045046,0.12162162,0.08558559,0.3108108,0.17567568,0.2027027,
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
index 5d50bb023d6ad87b58ee7358dc3083da66ba6857..0ae6842515ec400a815782b0fa1b7605fa275c9d 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Regression | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..5463c299284809586f22ec6b39545e67d0e72234
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_regression/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2313.0,"score":0.9,"rank":2.22,"hrank":1.72,"imp":0.604,"train_time":25.25,"predict_time":11.645,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+991\/-101","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1956.0,"score":0.634,"rank":7.93,"hrank":6.59,"imp":3.379,"train_time":735.58,"predict_time":9.557,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+686\/-97","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1928.0,"score":0.734,"rank":8.73,"hrank":2.63,"imp":1.883,"train_time":0.91,"predict_time":0.204,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1164\/-274","imputed":false},{"position":3,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1900.0,"score":0.675,"rank":9.62,"hrank":3.63,"imp":4.076,"train_time":366.93,"predict_time":2.982,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+408\/-7","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1886.0,"score":0.561,"rank":10.07,"hrank":8.38,"imp":4.068,"train_time":2.76,"predict_time":0.701,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+580\/-58","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1874.0,"score":0.587,"rank":10.44,"hrank":8.59,"imp":3.855,"train_time":735.58,"predict_time":0.714,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+636\/-92","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1858.0,"score":0.612,"rank":11.0,"hrank":2.78,"imp":3.417,"train_time":4786.6,"predict_time":361.183,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+869\/-152","imputed":false},{"position":7,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1829.0,"score":0.509,"rank":12.07,"hrank":10.29,"imp":4.444,"train_time":2526.31,"predict_time":3.717,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+632\/-37","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1823.0,"score":0.67,"rank":12.27,"hrank":3.56,"imp":3.118,"train_time":0.46,"predict_time":0.104,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1194\/-336","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1795.0,"score":0.488,"rank":13.36,"hrank":11.18,"imp":4.851,"train_time":2028.29,"predict_time":1.675,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+496\/-88","imputed":false},{"position":10,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1793.0,"score":0.469,"rank":13.42,"hrank":9.24,"imp":5.443,"train_time":473.02,"predict_time":8.73,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+352\/-0","imputed":false},{"position":11,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1788.0,"score":0.563,"rank":13.64,"hrank":8.13,"imp":3.716,"train_time":0.69,"predict_time":0.178,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+731\/-186","imputed":false},{"position":12,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1786.0,"score":0.56,"rank":13.73,"hrank":6.09,"imp":3.824,"train_time":4786.6,"predict_time":41.599,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+825\/-140","imputed":false},{"position":13,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1735.0,"score":0.452,"rank":15.82,"hrank":12.31,"imp":5.676,"train_time":4.87,"predict_time":1.868,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+324\/-78","imputed":false},{"position":14,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1700.0,"score":0.461,"rank":17.36,"hrank":12.62,"imp":4.759,"train_time":1.6,"predict_time":0.508,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+699\/-204","imputed":false},{"position":15,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1686.0,"score":0.48,"rank":17.98,"hrank":7.95,"imp":4.446,"train_time":45.77,"predict_time":39.212,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+722\/-147","imputed":false},{"position":16,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1645.0,"score":0.349,"rank":19.89,"hrank":14.6,"imp":5.88,"train_time":2028.29,"predict_time":0.081,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+472\/-154","imputed":false},{"position":17,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1630.0,"score":0.367,"rank":20.62,"hrank":17.01,"imp":6.056,"train_time":1.95,"predict_time":0.641,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+399\/-108","imputed":false},{"position":18,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1605.0,"score":0.333,"rank":21.84,"hrank":18.14,"imp":6.375,"train_time":425.61,"predict_time":0.615,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+319\/-59","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1603.0,"score":0.247,"rank":21.93,"hrank":15.56,"imp":7.997,"train_time":222.05,"predict_time":11.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+318\/-123","imputed":false},{"position":20,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1581.0,"score":0.213,"rank":23.0,"hrank":12.59,"imp":8.164,"train_time":912.76,"predict_time":0.345,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+290\/-92","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1561.0,"score":0.21,"rank":24.02,"hrank":13.2,"imp":8.244,"train_time":912.76,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+306\/-131","imputed":false},{"position":22,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1560.0,"score":0.201,"rank":24.07,"hrank":18.5,"imp":8.292,"train_time":222.05,"predict_time":0.968,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+311\/-127","imputed":false},{"position":23,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1551.0,"score":0.298,"rank":24.53,"hrank":20.99,"imp":6.209,"train_time":2526.31,"predict_time":0.166,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+421\/-174","imputed":false},{"position":24,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1517.0,"score":0.275,"rank":26.31,"hrank":22.6,"imp":7.176,"train_time":2406.36,"predict_time":0.713,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+314\/-182","imputed":false},{"position":25,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1480.0,"score":0.192,"rank":28.24,"hrank":19.56,"imp":8.929,"train_time":2.88,"predict_time":0.01,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+302\/-289","imputed":false},{"position":26,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1479.0,"score":0.199,"rank":28.27,"hrank":26.16,"imp":7.306,"train_time":425.61,"predict_time":0.04,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+242\/-83","imputed":false},{"position":27,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1452.0,"score":0.205,"rank":29.69,"hrank":26.64,"imp":7.924,"train_time":2406.36,"predict_time":0.05,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+233\/-164","imputed":false},{"position":28,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1447.0,"score":0.121,"rank":29.96,"hrank":27.57,"imp":8.863,"train_time":282.13,"predict_time":1.74,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+229\/-141","imputed":false},{"position":29,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1440.0,"score":0.181,"rank":30.36,"hrank":27.44,"imp":7.364,"train_time":6.88,"predict_time":0.094,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+277\/-143","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1434.0,"score":0.107,"rank":30.64,"hrank":27.3,"imp":8.983,"train_time":282.13,"predict_time":0.388,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+254\/-148","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1418.0,"score":0.102,"rank":31.49,"hrank":29.69,"imp":9.603,"train_time":2173.01,"predict_time":65.027,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-29","imputed":false},{"position":32,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1354.0,"score":0.123,"rank":34.96,"hrank":24.59,"imp":12.624,"train_time":123.72,"predict_time":0.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+407\/-458","imputed":false},{"position":33,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1336.0,"score":0.069,"rank":35.96,"hrank":34.77,"imp":9.163,"train_time":2173.01,"predict_time":16.096,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-89","imputed":false},{"position":34,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1333.0,"score":0.176,"rank":36.11,"hrank":23.55,"imp":8.642,"train_time":24.78,"predict_time":8.888,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+490\/-328","imputed":false},{"position":35,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1314.0,"score":0.068,"rank":37.11,"hrank":30.88,"imp":10.089,"train_time":0.6,"predict_time":0.08,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+295\/-282","imputed":false},{"position":36,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1312.0,"score":0.102,"rank":37.22,"hrank":28.94,"imp":12.82,"train_time":123.72,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+360\/-458","imputed":false},{"position":37,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1310.0,"score":0.195,"rank":37.31,"hrank":28.49,"imp":8.967,"train_time":0.48,"predict_time":0.126,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+374\/-338","imputed":false},{"position":38,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1299.0,"score":0.043,"rank":37.93,"hrank":36.19,"imp":10.003,"train_time":17.56,"predict_time":0.39,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+153\/-251","imputed":false},{"position":39,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1264.0,"score":0.04,"rank":39.8,"hrank":37.58,"imp":10.105,"train_time":3.13,"predict_time":0.038,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+159\/-196","imputed":false},{"position":40,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1249.0,"score":0.011,"rank":40.58,"hrank":39.17,"imp":10.726,"train_time":2.28,"predict_time":0.13,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+154\/-149","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1243.0,"score":0.061,"rank":40.87,"hrank":38.93,"imp":10.193,"train_time":4348.61,"predict_time":0.528,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+156\/-178","imputed":false},{"position":42,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1236.0,"score":0.014,"rank":41.24,"hrank":40.31,"imp":10.313,"train_time":0.55,"predict_time":0.233,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-156","imputed":false},{"position":43,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1227.0,"score":0.079,"rank":41.69,"hrank":35.9,"imp":13.84,"train_time":0.73,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+331\/-484","imputed":false},{"position":44,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1202.0,"score":0.0,"rank":42.98,"hrank":41.78,"imp":10.696,"train_time":17.56,"predict_time":0.192,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+139\/-254","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1182.0,"score":0.038,"rank":44.02,"hrank":41.99,"imp":11.165,"train_time":4348.61,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+156\/-219","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1174.0,"score":0.034,"rank":44.42,"hrank":41.51,"imp":12.042,"train_time":49.61,"predict_time":0.231,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+202\/-324","imputed":false},{"position":47,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1146.0,"score":0.001,"rank":45.84,"hrank":43.92,"imp":11.652,"train_time":6.75,"predict_time":0.235,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-249","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1143.0,"score":0.023,"rank":46.0,"hrank":43.56,"imp":12.233,"train_time":49.61,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+172\/-344","imputed":false},{"position":49,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1132.0,"score":0.0,"rank":46.51,"hrank":46.18,"imp":11.918,"train_time":79.08,"predict_time":0.546,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-192","imputed":false},{"position":50,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1119.0,"score":0.067,"rank":47.18,"hrank":37.36,"imp":11.931,"train_time":8.79,"predict_time":7.636,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+361\/-370","imputed":false},{"position":51,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1108.0,"score":0.0,"rank":47.69,"hrank":47.29,"imp":12.214,"train_time":79.08,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-202","imputed":false},{"position":52,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1076.0,"score":0.059,"rank":49.2,"hrank":41.4,"imp":15.75,"train_time":1213.97,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+347\/-554","imputed":false},{"position":53,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1018.0,"score":0.001,"rank":51.71,"hrank":49.72,"imp":15.117,"train_time":15.08,"predict_time":0.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+165\/-330","imputed":false},{"position":54,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1016.0,"score":0.041,"rank":51.82,"hrank":46.32,"imp":16.308,"train_time":1213.97,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+322\/-571","imputed":false},{"position":55,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1002.0,"score":0.0,"rank":52.4,"hrank":51.0,"imp":14.379,"train_time":1.5,"predict_time":0.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+173\/-491","imputed":false},{"position":56,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":52.49,"hrank":52.24,"imp":13.351,"train_time":0.43,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+22\/-248","imputed":false},{"position":57,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":978.0,"score":0.0,"rank":53.38,"hrank":51.19,"imp":14.05,"train_time":0.78,"predict_time":0.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+168\/-520","imputed":false},{"position":58,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":975.0,"score":0.034,"rank":53.51,"hrank":48.32,"imp":16.982,"train_time":2.36,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+323\/-615","imputed":false},{"position":59,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":902.0,"score":0.0,"rank":56.29,"hrank":55.48,"imp":16.112,"train_time":93.33,"predict_time":21.745,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-616","imputed":false},{"position":60,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":842.0,"score":0.0,"rank":58.27,"hrank":57.71,"imp":16.191,"train_time":318.98,"predict_time":0.827,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-506","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":800.0,"score":0.0,"rank":59.56,"hrank":59.13,"imp":16.639,"train_time":318.98,"predict_time":0.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-471","imputed":false},{"position":62,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":763.0,"score":0.0,"rank":60.6,"hrank":60.19,"imp":19.755,"train_time":142.58,"predict_time":1.184,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-387","imputed":false},{"position":63,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":672.0,"score":0.0,"rank":62.84,"hrank":62.57,"imp":23.616,"train_time":4632.6,"predict_time":5.283,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-663","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":662.0,"score":0.0,"rank":63.07,"hrank":62.66,"imp":20.189,"train_time":1.36,"predict_time":0.111,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-719","imputed":false},{"position":65,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":637.0,"score":0.0,"rank":63.6,"hrank":63.39,"imp":23.844,"train_time":4632.6,"predict_time":1.724,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+123\/-664","imputed":false},{"position":66,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":627.0,"score":0.0,"rank":63.8,"hrank":63.55,"imp":22.259,"train_time":142.58,"predict_time":0.064,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-429","imputed":false},{"position":67,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":504.0,"score":0.0,"rank":65.93,"hrank":65.82,"imp":27.081,"train_time":112.83,"predict_time":1.964,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-753","imputed":false},{"position":68,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":408.0,"score":0.0,"rank":67.16,"hrank":66.95,"imp":29.6,"train_time":0.06,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-1120","imputed":false},{"position":69,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":33.0,"score":0.0,"rank":69.9,"hrank":69.9,"imp":41.19,"train_time":44.3,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-1380","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-138.0,"score":0.0,"rank":70.74,"hrank":70.74,"imp":41.565,"train_time":44.3,"predict_time":0.011,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+176\/-1402","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-400.0,"score":0.0,"rank":71.78,"hrank":71.78,"imp":44.679,"train_time":0.13,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+195\/-1626","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/pareto_front_explorer.html
index 20dbf83a4e2552bf3692d34e14f9a31c8a0e2d25..d2dd94e879897792cb1acd05549a48ead6d41ab2 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0095047043,"imp":8.9289217905,"elo":1479.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0151739618,"imp":8.2440800428,"elo":1561.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3448570028,"imp":8.1635768776,"elo":1581.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0132050202,"imp":13.8402266011,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0405844599,"imp":12.820361936,"elo":1312.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"imp":12.6244692317,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0053932693,"imp":16.9821360779,"elo":975.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0053282856,"imp":16.307855987,"elo":1015.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0697951177,"imp":15.7503986055,"elo":1075.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0781872102,"imp":14.0502699508,"elo":978.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0196168127,"imp":12.2328759419,"elo":1142.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2313039056,"imp":12.0422079013,"elo":1174.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"imp":20.1888549154,"elo":661.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1141755222,"imp":16.6393490521,"elo":799.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8269429369,"imp":16.1913776362,"elo":842.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2328706766,"imp":10.3133865423,"elo":1235.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9682498776,"imp":8.2922840605,"elo":1560.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":11.2363457974,"imp":7.9966042912,"elo":1602.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.8875279848,"imp":8.6418127076,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.0940273851,"imp":7.3642181904,"elo":1439.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.1656582564,"imp":6.2094250691,"elo":1551.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":3.7166074672,"imp":4.4441658088,"elo":1828.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.6406884541,"imp":6.0560180179,"elo":1629.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8676058389,"imp":5.675536215,"elo":1735.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.009266306,"imp":14.3785932579,"elo":1002.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1920034411,"imp":10.6961018894,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3903416157,"imp":10.0026128041,"elo":1298.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0529855097,"imp":13.3509676333,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0565808637,"imp":12.2137889713,"elo":1108.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5464279632,"imp":11.9179109649,"elo":1132.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"imp":11.6524728164,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0807033173,"imp":5.8804255611,"elo":1645.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.6746674551,"imp":4.851092681,"elo":1795.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"imp":4.7589446613,"elo":1700.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7137361675,"imp":3.8547006757,"elo":1874.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5569470263,"imp":3.3793076451,"elo":1956.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.6364561632,"imp":11.9310816279,"elo":1118.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"imp":4.4457870824,"elo":1686.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":41.5989132411,"imp":3.8237840689,"elo":1785.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":361.1825850771,"imp":3.4173495548,"elo":1858.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1784260866,"imp":3.7162273027,"elo":1787.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":11.6453720048,"imp":0.6036434685,"elo":2313.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1038916814,"imp":3.1179522944,"elo":1823.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0383065968,"imp":10.1054237502,"elo":1263.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0498907511,"imp":7.9241643499,"elo":1452.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.7132695165,"imp":7.1759292759,"elo":1516.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"imp":4.0683965406,"elo":1885.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042898407,"imp":1.8828107499,"elo":1928.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"imp":27.0811357382,"elo":504.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.723789106,"imp":23.8439307085,"elo":636.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"imp":23.6160952188,"elo":671.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1260951636,"imp":8.9674037362,"elo":1310.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0279916773,"imp":15.1168656964,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0335416412,"imp":11.1647618802,"elo":1182.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5275799068,"imp":10.1930533871,"elo":1243.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0799084859,"imp":10.0891414742,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3882893337,"imp":8.9825719868,"elo":1434.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.7397156138,"imp":8.8631737523,"elo":1447.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":21.7449541194,"imp":16.1121022022,"elo":901.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":16.0959669691,"imp":9.1633459226,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.026662372,"imp":9.6026560483,"elo":1418.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1303369377,"imp":10.7259140587,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.040253721,"imp":7.3060870345,"elo":1479.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.6149915399,"imp":6.375444867,"elo":1604.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0095047043,"x_train":2.8843766069,"imp":8.9289217905,"elo":1479.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0151739618,"x_train":912.7561702144,"imp":8.2440800428,"elo":1561.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3448570028,"x_train":912.7561702144,"imp":8.1635768776,"elo":1581.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0132050202,"x_train":0.7327990592,"imp":13.8402266011,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0405844599,"x_train":123.7245358586,"imp":12.820361936,"elo":1312.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"x_train":123.7245358586,"imp":12.6244692317,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0053932693,"x_train":2.363335295,"imp":16.9821360779,"elo":975.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0053282856,"x_train":1213.9718626455,"imp":16.307855987,"elo":1015.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0697951177,"x_train":1213.9718626455,"imp":15.7503986055,"elo":1075.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0781872102,"x_train":0.7788435912,"imp":14.0502699508,"elo":978.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0196168127,"x_train":49.6127891307,"imp":12.2328759419,"elo":1142.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2313039056,"x_train":49.6127891307,"imp":12.0422079013,"elo":1174.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"x_train":1.3590539265,"imp":20.1888549154,"elo":661.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1141755222,"x_train":318.981772136,"imp":16.6393490521,"elo":799.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8269429369,"x_train":318.981772136,"imp":16.1913776362,"elo":842.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2328706766,"x_train":0.5527468904,"imp":10.3133865423,"elo":1235.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9682498776,"x_train":222.0508632426,"imp":8.2922840605,"elo":1560.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":11.2363457974,"x_train":222.0508632426,"imp":7.9966042912,"elo":1602.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.8875279848,"x_train":24.7794415376,"imp":8.6418127076,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.0940273851,"x_train":6.8840236669,"imp":7.3642181904,"elo":1439.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.1656582564,"x_train":2526.3066806002,"imp":6.2094250691,"elo":1551.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":3.7166074672,"x_train":2526.3066806002,"imp":4.4441658088,"elo":1828.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.6406884541,"x_train":1.9532897692,"imp":6.0560180179,"elo":1629.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8676058389,"x_train":4.8736992926,"imp":5.675536215,"elo":1735.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.009266306,"x_train":1.4985629713,"imp":14.3785932579,"elo":1002.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1920034411,"x_train":17.5566235787,"imp":10.6961018894,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3903416157,"x_train":17.5566235787,"imp":10.0026128041,"elo":1298.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0529855097,"x_train":0.4305837825,"imp":13.3509676333,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0565808637,"x_train":79.0783741735,"imp":12.2137889713,"elo":1108.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5464279632,"x_train":79.0783741735,"imp":11.9179109649,"elo":1132.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"x_train":6.7532462417,"imp":11.6524728164,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0807033173,"x_train":2028.289449394,"imp":5.8804255611,"elo":1645.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.6746674551,"x_train":2028.289449394,"imp":4.851092681,"elo":1795.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"x_train":1.6000651298,"imp":4.7589446613,"elo":1700.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7137361675,"x_train":735.5848736649,"imp":3.8547006757,"elo":1874.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5569470263,"x_train":735.5848736649,"imp":3.3793076451,"elo":1956.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.6364561632,"x_train":8.7889189928,"imp":11.9310816279,"elo":1118.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"x_train":45.7738111436,"imp":4.4457870824,"elo":1686.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":41.5989132411,"x_train":4786.5989331939,"imp":3.8237840689,"elo":1785.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":361.1825850771,"x_train":4786.5989331939,"imp":3.4173495548,"elo":1858.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1784260866,"x_train":0.6904298262,"imp":3.7162273027,"elo":1787.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":11.6453720048,"x_train":25.251811811,"imp":0.6036434685,"elo":2313.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1038916814,"x_train":0.4638004915,"imp":3.1179522944,"elo":1823.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0383065968,"x_train":3.1341264164,"imp":10.1054237502,"elo":1263.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0498907511,"x_train":2406.360319586,"imp":7.9241643499,"elo":1452.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.7132695165,"x_train":2406.360319586,"imp":7.1759292759,"elo":1516.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"x_train":2.763002179,"imp":4.0683965406,"elo":1885.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042898407,"x_train":0.9134642862,"imp":1.8828107499,"elo":1928.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"x_train":112.8279660853,"imp":27.0811357382,"elo":504.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.723789106,"x_train":4632.5950923332,"imp":23.8439307085,"elo":636.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"x_train":4632.5950923332,"imp":23.6160952188,"elo":671.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1260951636,"x_train":0.4819098389,"imp":8.9674037362,"elo":1310.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0279916773,"x_train":15.0791065291,"imp":15.1168656964,"elo":1018.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0335416412,"x_train":4348.6118964812,"imp":11.1647618802,"elo":1182.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5275799068,"x_train":4348.6118964812,"imp":10.1930533871,"elo":1243.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0799084859,"x_train":0.597628695,"imp":10.0891414742,"elo":1314.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3882893337,"x_train":282.1320041804,"imp":8.9825719868,"elo":1434.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.7397156138,"x_train":282.1320041804,"imp":8.8631737523,"elo":1447.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":21.7449541194,"x_train":93.3311418241,"imp":16.1121022022,"elo":901.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":16.0959669691,"x_train":2173.0087869337,"imp":9.1633459226,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.026662372,"x_train":2173.0087869337,"imp":9.6026560483,"elo":1418.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1303369377,"x_train":2.275318476,"imp":10.7259140587,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.040253721,"x_train":425.6136485357,"imp":7.3060870345,"elo":1479.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.6149915399,"x_train":425.6136485357,"imp":6.375444867,"elo":1604.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
index 459d2a7d19447eb58717b8159ed5cb8c01be4dda..832738602a9c6a5d9ff061569a3c63e0a6537a9e 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":2.8843766069,"imp":8.9299549058,"elo":1494.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":8.2528193911,"imp":8.7105054199,"elo":1515.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":21.6626665741,"imp":8.4887309209,"elo":1553.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":44.1604809025,"imp":8.3923746931,"elo":1565.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":109.0816998293,"imp":8.2570654578,"elo":1584.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":221.6369492763,"imp":8.2079108097,"elo":1594.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":453.4873826918,"imp":8.1769147819,"elo":1605.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":679.1817937831,"imp":8.1709441403,"elo":1605.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":912.7561702144,"imp":8.1646332008,"elo":1606.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.7327990592,"imp":13.8412569008,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.3862926068,"imp":13.8758796411,"elo":1226.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.0643858395,"imp":13.7477732407,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.9372791069,"imp":13.4747905089,"elo":1280.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":15.3821406623,"imp":13.1467714941,"elo":1319.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":31.1470442744,"imp":12.9349519149,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":61.7023905687,"imp":12.7559455414,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":92.1424404788,"imp":12.6633921819,"elo":1374.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":123.7245358586,"imp":12.6257470877,"elo":1382.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.363335295,"imp":16.9830509773,"elo":978.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":10.5715947461,"imp":16.6384137163,"elo":989.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":33.6247878081,"imp":16.3265767301,"elo":1011.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":62.9357438641,"imp":16.1826188225,"elo":1031.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":154.7347742278,"imp":15.94979051,"elo":1053.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":312.6511755596,"imp":15.8658173525,"elo":1071.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":608.0045919755,"imp":15.7966651015,"elo":1083.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":896.9079641691,"imp":15.7709114778,"elo":1093.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1213.9718626455,"imp":15.752285937,"elo":1102.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.7788435912,"imp":14.0512924699,"elo":974.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.0217265688,"imp":13.4121426201,"elo":1040.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.7428029518,"imp":12.7467383474,"elo":1108.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9291798058,"imp":12.5238198046,"elo":1130.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.6147364724,"imp":12.1897531097,"elo":1167.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.7108167547,"imp":12.1313110609,"elo":1177.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.9416823722,"imp":12.0738639893,"elo":1186.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.1384318226,"imp":12.0536734972,"elo":1191.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":49.6127891307,"imp":12.043520821,"elo":1197.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.3590539265,"imp":20.1897089635,"elo":652.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.8414619709,"imp":20.0123167582,"elo":673.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":7.8616883329,"imp":19.5912779375,"elo":699.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":16.2112927486,"imp":19.2530853425,"elo":717.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":38.3780166466,"imp":18.1212969579,"elo":751.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":77.3536256265,"imp":17.3890213875,"elo":775.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":157.4362726765,"imp":16.8092188229,"elo":800.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":237.2917035986,"imp":16.4215969874,"elo":829.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":318.981772136,"imp":16.1924281318,"elo":851.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.5527468904,"imp":10.3144188427,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":1.764654979,"imp":10.1439744184,"elo":1273.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":5.1230989868,"imp":9.3779852563,"elo":1386.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":10.4740202796,"imp":8.9461937198,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":27.1713091927,"imp":8.3796482578,"elo":1551.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":54.7938156246,"imp":8.2125157706,"elo":1592.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":109.9587992843,"imp":8.0892161778,"elo":1617.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":165.4413385162,"imp":8.0466483632,"elo":1630.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":222.0508632426,"imp":7.9979976356,"elo":1647.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":24.7794415376,"imp":8.6432598339,"elo":1349.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":6.8840236669,"imp":7.3652532414,"elo":1465.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":21.1491685049,"imp":6.8490090606,"elo":1512.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.2869505149,"imp":6.025229913,"elo":1606.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":121.8592858048,"imp":5.5681201505,"elo":1684.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":292.3137710756,"imp":4.9895084388,"elo":1774.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":629.8900400305,"imp":4.7518111757,"elo":1812.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1243.5591974675,"imp":4.5697026849,"elo":1843.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1862.5519447854,"imp":4.4826479243,"elo":1862.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2526.3066806002,"imp":4.4446564679,"elo":1878.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":1.9532897692,"imp":6.0570910871,"elo":1670.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":4.8736992926,"imp":5.6766219205,"elo":1785.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.4985629713,"imp":14.3795592436,"elo":1014.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":5.1141485965,"imp":11.8232633719,"elo":1166.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":14.778241098,"imp":10.0724475671,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":17.5566235787,"imp":10.003641549,"elo":1305.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4305837825,"imp":13.3519737021,"elo":999.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.815213214,"imp":12.9091076456,"elo":1052.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":2.0588185718,"imp":12.5338637204,"elo":1081.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":4.0449726807,"imp":12.2144596854,"elo":1116.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":9.8632350349,"imp":12.0433557673,"elo":1137.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":19.655091039,"imp":11.9759550183,"elo":1149.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":39.2307694386,"imp":11.9396715877,"elo":1159.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":58.9645108517,"imp":11.9252718852,"elo":1164.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":79.0783741735,"imp":11.9189325871,"elo":1167.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.7532462417,"imp":11.6535004345,"elo":1168.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.9212846988,"imp":9.3001805229,"elo":1335.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":45.3026550705,"imp":7.0868286661,"elo":1553.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":98.5294547231,"imp":6.2737962747,"elo":1637.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":250.4641319867,"imp":5.6133750729,"elo":1740.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":506.2044753501,"imp":5.2815085909,"elo":1776.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1007.9516707303,"imp":5.0497861022,"elo":1813.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1512.7513669099,"imp":4.944648337,"elo":1832.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2028.289449394,"imp":4.8525237555,"elo":1853.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.6000651298,"imp":4.76003587,"elo":1731.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.1278647155,"imp":4.6479612017,"elo":1750.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.2069927888,"imp":4.4088830446,"elo":1772.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":34.2536042268,"imp":4.1041097996,"elo":1826.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":90.6790780885,"imp":3.7645520907,"elo":1919.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":181.3952243454,"imp":3.6005496212,"elo":1960.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":363.9172361826,"imp":3.4795975726,"elo":1991.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":547.7650992279,"imp":3.4134922755,"elo":2012.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":735.5848736649,"imp":3.3801975505,"elo":2022.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":8.7889189928,"imp":11.932087857,"elo":1142.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":45.7738111436,"imp":4.4469104839,"elo":1703.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":70.0936255942,"imp":4.3204912586,"elo":1727.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":132.9785409439,"imp":4.1649385631,"elo":1750.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":233.9204718624,"imp":4.0433518212,"elo":1770.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":585.2609706035,"imp":3.8074108911,"elo":1806.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1183.6501039466,"imp":3.6469702608,"elo":1831.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2376.2279314343,"imp":3.5169662809,"elo":1859.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3574.1919132683,"imp":3.4472605818,"elo":1873.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4786.5989331939,"imp":3.4184218295,"elo":1882.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.6904298262,"imp":3.7173335385,"elo":1816.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":25.251811811,"imp":0.6047647202,"elo":2377.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.4638004915,"imp":3.1190648248,"elo":1848.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":3.1341264164,"imp":10.1064497634,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":11.6448003172,"imp":9.496118583,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":46.7227871137,"imp":8.708612053,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":111.5296374455,"imp":8.2026296087,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":297.5742071652,"imp":7.7725618975,"elo":1487.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":594.9119312701,"imp":7.5192422695,"elo":1513.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1160.4787998549,"imp":7.3290046159,"elo":1534.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1766.291540139,"imp":7.2426921608,"elo":1546.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2406.360319586,"imp":7.1769941527,"elo":1560.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.763002179,"imp":4.0694795565,"elo":1933.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":0.9134642862,"imp":1.8854650044,"elo":1932.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":112.8279660853,"imp":27.0818393232,"elo":479.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":136.6220329381,"imp":26.9998231049,"elo":511.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":192.3169267284,"imp":26.8524843999,"elo":522.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":268.6819124436,"imp":26.4817153546,"elo":543.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":587.894611724,"imp":25.7481757445,"elo":569.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1166.9417941917,"imp":24.8517487919,"elo":596.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2300.4888270054,"imp":24.2935720809,"elo":616.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3411.7622050693,"imp":23.902805457,"elo":647.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4632.5950923332,"imp":23.61683989,"elo":682.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.4819098389,"imp":8.9684319131,"elo":1327.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":15.0791065291,"imp":15.1178597201,"elo":1033.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":37.5490244182,"imp":14.1906901618,"elo":1058.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":99.0264645811,"imp":12.8390046101,"elo":1120.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":209.9682919696,"imp":12.1128246668,"elo":1165.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":502.6360049647,"imp":11.2660448555,"elo":1204.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1067.3549756812,"imp":10.8935667918,"elo":1227.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2143.4901830021,"imp":10.5085358073,"elo":1245.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3266.8165720174,"imp":10.3122475245,"elo":1257.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4348.6118964812,"imp":10.1939206839,"elo":1272.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.597628695,"imp":10.090168709,"elo":1327.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.4341782987,"imp":9.9737458625,"elo":1337.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":7.341922179,"imp":9.5555612023,"elo":1393.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":13.4253918933,"imp":9.3535205866,"elo":1412.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":35.1370362218,"imp":9.1242230507,"elo":1437.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":69.8388019133,"imp":8.9984966808,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":138.6843072833,"imp":8.9098779875,"elo":1466.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":209.6003170598,"imp":8.8821110462,"elo":1473.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":282.1320041804,"imp":8.8639971487,"elo":1480.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":93.3311418241,"imp":16.1130673079,"elo":908.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":176.98147887,"imp":14.0309647611,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":425.9076940912,"imp":18.8066937887,"elo":1120.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":844.3867407395,"imp":14.9020596719,"elo":1230.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":2092.7290649042,"imp":9.6291821749,"elo":1439.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":2173.0087869337,"imp":9.6037114724,"elo":1446.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.275318476,"imp":10.7269469123,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":4.3886320534,"imp":9.9095442967,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":10.7096165053,"imp":8.8150655878,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":21.3971746465,"imp":7.8221470759,"elo":1487.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":53.3414186748,"imp":7.014236119,"elo":1560.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":106.1300468914,"imp":6.6825183636,"elo":1590.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":212.0315758458,"imp":6.4961289435,"elo":1614.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":317.9586579883,"imp":6.4098620166,"elo":1629.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":425.6136485357,"imp":6.3770966809,"elo":1642.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0095047043,"x_train":2.8843766069,"imp":8.9299549058,"elo":1494.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0481663313,"x_train":8.2528193911,"imp":8.7105054199,"elo":1515.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1163670459,"x_train":21.6626665741,"imp":8.4887309209,"elo":1553.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1729198824,"x_train":44.1604809025,"imp":8.3923746931,"elo":1565.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2175739337,"x_train":109.0816998293,"imp":8.2570654578,"elo":1584.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2694149278,"x_train":221.6369492763,"imp":8.2079108097,"elo":1594.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3197063318,"x_train":453.4873826918,"imp":8.1769147819,"elo":1605.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3407350551,"x_train":679.1817937831,"imp":8.1709441403,"elo":1605.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3448570028,"x_train":912.7561702144,"imp":8.1646332008,"elo":1606.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0132050202,"x_train":0.7327990592,"imp":13.8412569008,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.027612708,"x_train":1.3862926068,"imp":13.8758796411,"elo":1226.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0666759456,"x_train":3.0643858395,"imp":13.7477732407,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1073929028,"x_train":5.9372791069,"imp":13.4747905089,"elo":1280.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1481068526,"x_train":15.3821406623,"imp":13.1467714941,"elo":1319.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1786997869,"x_train":31.1470442744,"imp":12.9349519149,"elo":1345.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2109984061,"x_train":61.7023905687,"imp":12.7559455414,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2269570142,"x_train":92.1424404788,"imp":12.6633921819,"elo":1374.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2421474108,"x_train":123.7245358586,"imp":12.6257470877,"elo":1382.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0053932693,"x_train":2.363335295,"imp":16.9830509773,"elo":978.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.010637589,"x_train":10.5715947461,"imp":16.6384137163,"elo":989.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0211515534,"x_train":33.6247878081,"imp":16.3265767301,"elo":1011.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0306648453,"x_train":62.9357438641,"imp":16.1826188225,"elo":1031.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0433861149,"x_train":154.7347742278,"imp":15.94979051,"elo":1053.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0522152152,"x_train":312.6511755596,"imp":15.8658173525,"elo":1071.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0627955752,"x_train":608.0045919755,"imp":15.7966651015,"elo":1083.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0676616143,"x_train":896.9079641691,"imp":15.7709114778,"elo":1093.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.070209085,"x_train":1213.9718626455,"imp":15.752285937,"elo":1102.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0781872102,"x_train":0.7788435912,"imp":14.0512924699,"elo":974.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0889630222,"x_train":1.0217265688,"imp":13.4121426201,"elo":1040.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0926460136,"x_train":1.7428029518,"imp":12.7467383474,"elo":1108.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1127201263,"x_train":2.9291798058,"imp":12.5238198046,"elo":1130.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1338232875,"x_train":6.6147364724,"imp":12.1897531097,"elo":1167.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1601021681,"x_train":12.7108167547,"imp":12.1313110609,"elo":1177.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739802168,"x_train":24.9416823722,"imp":12.0738639893,"elo":1186.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1953387622,"x_train":37.1384318226,"imp":12.0536734972,"elo":1191.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2313039056,"x_train":49.6127891307,"imp":12.043520821,"elo":1197.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1111866275,"x_train":1.3590539265,"imp":20.1897089635,"elo":652.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2081543614,"x_train":2.8414619709,"imp":20.0123167582,"elo":673.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3420605892,"x_train":7.8616883329,"imp":19.5912779375,"elo":699.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4546707256,"x_train":16.2112927486,"imp":19.2530853425,"elo":717.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4701854923,"x_train":38.3780166466,"imp":18.1212969579,"elo":751.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5743045365,"x_train":77.3536256265,"imp":17.3890213875,"elo":775.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7208507445,"x_train":157.4362726765,"imp":16.8092188229,"elo":800.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7541111543,"x_train":237.2917035986,"imp":16.4215969874,"elo":829.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8269429369,"x_train":318.981772136,"imp":16.1924281318,"elo":851.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2328706766,"x_train":0.5527468904,"imp":10.3144188427,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4016545585,"x_train":1.764654979,"imp":10.1439744184,"elo":1273.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2355387213,"x_train":5.1230989868,"imp":9.3779852563,"elo":1386.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.8851780262,"x_train":10.4740202796,"imp":8.9461937198,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.4690263501,"x_train":27.1713091927,"imp":8.3796482578,"elo":1551.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.1421527385,"x_train":54.7938156246,"imp":8.2125157706,"elo":1592.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.9341743249,"x_train":109.9587992843,"imp":8.0892161778,"elo":1617.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":11.0753701273,"x_train":165.4413385162,"imp":8.0466483632,"elo":1630.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":11.2363457974,"x_train":222.0508632426,"imp":7.9979976356,"elo":1647.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":8.8875279848,"x_train":24.7794415376,"imp":8.6432598339,"elo":1349.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.0940273851,"x_train":6.8840236669,"imp":7.3652532414,"elo":1465.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2557002328,"x_train":21.1491685049,"imp":6.8490090606,"elo":1512.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.578211498,"x_train":60.2869505149,"imp":6.025229913,"elo":1606.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9818556188,"x_train":121.8592858048,"imp":5.5681201505,"elo":1684.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.6674120378,"x_train":292.3137710756,"imp":4.9895084388,"elo":1774.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2999785133,"x_train":629.8900400305,"imp":4.7518111757,"elo":1812.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.975897112,"x_train":1243.5591974675,"imp":4.5697026849,"elo":1843.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.4550985564,"x_train":1862.5519447854,"imp":4.4826479243,"elo":1862.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.7166074672,"x_train":2526.3066806002,"imp":4.4446564679,"elo":1878.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.6406884541,"x_train":1.9532897692,"imp":6.0570910871,"elo":1670.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8676058389,"x_train":4.8736992926,"imp":5.6766219205,"elo":1785.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.009266306,"x_train":1.4985629713,"imp":14.3795592436,"elo":1014.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1022784942,"x_train":5.1141485965,"imp":11.8232633719,"elo":1166.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3309819924,"x_train":14.778241098,"imp":10.0724475671,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3903416157,"x_train":17.5566235787,"imp":10.003641549,"elo":1305.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0529855097,"x_train":0.4305837825,"imp":13.3519737021,"elo":999.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1039804779,"x_train":0.815213214,"imp":12.9091076456,"elo":1052.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1684108594,"x_train":2.0588185718,"imp":12.5338637204,"elo":1081.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2358696304,"x_train":4.0449726807,"imp":12.2144596854,"elo":1116.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.356186403,"x_train":9.8632350349,"imp":12.0433557673,"elo":1137.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4195441878,"x_train":19.655091039,"imp":11.9759550183,"elo":1149.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4998627172,"x_train":39.2307694386,"imp":11.9396715877,"elo":1159.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.53892783,"x_train":58.9645108517,"imp":11.9252718852,"elo":1164.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5464279632,"x_train":79.0783741735,"imp":11.9189325871,"elo":1167.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.234981848,"x_train":6.7532462417,"imp":11.6535004345,"elo":1168.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3080259999,"x_train":15.9212846988,"imp":9.3001805229,"elo":1335.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3619920454,"x_train":45.3026550705,"imp":7.0868286661,"elo":1553.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5857834951,"x_train":98.5294547231,"imp":6.2737962747,"elo":1637.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.897596221,"x_train":250.4641319867,"imp":5.6133750729,"elo":1740.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1621653964,"x_train":506.2044753501,"imp":5.2815085909,"elo":1776.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4580786339,"x_train":1007.9516707303,"imp":5.0497861022,"elo":1813.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5889939537,"x_train":1512.7513669099,"imp":4.944648337,"elo":1832.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.6682080328,"x_train":2028.289449394,"imp":4.8525237555,"elo":1853.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5079303228,"x_train":1.6000651298,"imp":4.76003587,"elo":1731.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4031929556,"x_train":5.1278647155,"imp":4.6479612017,"elo":1750.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2513268047,"x_train":16.2069927888,"imp":4.4088830446,"elo":1772.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.9918356458,"x_train":34.2536042268,"imp":4.1041097996,"elo":1826.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4620925312,"x_train":90.6790780885,"imp":3.7645520907,"elo":1919.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5018360728,"x_train":181.3952243454,"imp":3.6005496212,"elo":1960.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.7259397593,"x_train":363.9172361826,"imp":3.4795975726,"elo":1991.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.0781180135,"x_train":547.7650992279,"imp":3.4134922755,"elo":2012.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.6037629654,"x_train":735.5848736649,"imp":3.3801975505,"elo":2022.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.6364561632,"x_train":8.7889189928,"imp":11.932087857,"elo":1142.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2116871068,"x_train":45.7738111436,"imp":4.4469104839,"elo":1703.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":59.6608119519,"x_train":70.0936255942,"imp":4.3204912586,"elo":1727.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":83.8163328946,"x_train":132.9785409439,"imp":4.1649385631,"elo":1750.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":117.6114176494,"x_train":233.9204718624,"imp":4.0433518212,"elo":1770.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":205.7656427606,"x_train":585.2609706035,"imp":3.8074108911,"elo":1806.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":266.8254661426,"x_train":1183.6501039466,"imp":3.6469702608,"elo":1831.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":326.9571428165,"x_train":2376.2279314343,"imp":3.5169662809,"elo":1859.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":356.2280399404,"x_train":3574.1919132683,"imp":3.4472605818,"elo":1873.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":361.1825850771,"x_train":4786.5989331939,"imp":3.4184218295,"elo":1882.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1784260866,"x_train":0.6904298262,"imp":3.7173335385,"elo":1816.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":11.6453720048,"x_train":25.251811811,"imp":0.6047647202,"elo":2377.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1038916814,"x_train":0.4638004915,"imp":3.1190648248,"elo":1848.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0383065968,"x_train":3.1341264164,"imp":10.1064497634,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0804358946,"x_train":11.6448003172,"imp":9.496118583,"elo":1349.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.147897522,"x_train":46.7227871137,"imp":8.708612053,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2244672749,"x_train":111.5296374455,"imp":8.2026296087,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3405422347,"x_train":297.5742071652,"imp":7.7725618975,"elo":1487.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4659120865,"x_train":594.9119312701,"imp":7.5192422695,"elo":1513.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.6581520332,"x_train":1160.4787998549,"imp":7.3290046159,"elo":1534.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.7117167229,"x_train":1766.291540139,"imp":7.2426921608,"elo":1546.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.7132695165,"x_train":2406.360319586,"imp":7.1769941527,"elo":1560.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7008750147,"x_train":2.763002179,"imp":4.0694795565,"elo":1933.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042898407,"x_train":0.9134642862,"imp":1.8854650044,"elo":1932.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9637818715,"x_train":112.8279660853,"imp":27.0818393232,"elo":479.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.9334778523,"x_train":136.6220329381,"imp":26.9998231049,"elo":511.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1406475162,"x_train":192.3169267284,"imp":26.8524843999,"elo":522.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.4886555763,"x_train":268.6819124436,"imp":26.4817153546,"elo":543.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.967901931,"x_train":587.894611724,"imp":25.7481757445,"elo":569.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0057237984,"x_train":1166.9417941917,"imp":24.8517487919,"elo":596.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.7178375205,"x_train":2300.4888270054,"imp":24.2935720809,"elo":616.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9435530344,"x_train":3411.7622050693,"imp":23.902805457,"elo":647.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.282666633,"x_train":4632.5950923332,"imp":23.61683989,"elo":682.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1260951636,"x_train":0.4819098389,"imp":8.9684319131,"elo":1327.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0279916773,"x_train":15.0791065291,"imp":15.1178597201,"elo":1033.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0762465224,"x_train":37.5490244182,"imp":14.1906901618,"elo":1058.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1408050768,"x_train":99.0264645811,"imp":12.8390046101,"elo":1120.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1800817267,"x_train":209.9682919696,"imp":12.1128246668,"elo":1165.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2447433002,"x_train":502.6360049647,"imp":11.2660448555,"elo":1204.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3486057789,"x_train":1067.3549756812,"imp":10.8935667918,"elo":1227.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4465492625,"x_train":2143.4901830021,"imp":10.5085358073,"elo":1245.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5425320243,"x_train":3266.8165720174,"imp":10.3122475245,"elo":1257.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5275799068,"x_train":4348.6118964812,"imp":10.1939206839,"elo":1272.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0799084859,"x_train":0.597628695,"imp":10.090168709,"elo":1327.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3337766093,"x_train":2.4341782987,"imp":9.9737458625,"elo":1337.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8523423396,"x_train":7.341922179,"imp":9.5555612023,"elo":1393.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1380843561,"x_train":13.4253918933,"imp":9.3535205866,"elo":1412.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.5157666044,"x_train":35.1370362218,"imp":9.1242230507,"elo":1437.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7011934543,"x_train":69.8388019133,"imp":8.9984966808,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.839001453,"x_train":138.6843072833,"imp":8.9098779875,"elo":1466.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8309092575,"x_train":209.6003170598,"imp":8.8821110462,"elo":1473.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7397156138,"x_train":282.1320041804,"imp":8.8639971487,"elo":1480.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":21.7449541194,"x_train":93.3311418241,"imp":16.1130673079,"elo":908.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":27.6110229171,"x_train":176.98147887,"imp":14.0309647611,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.033754283,"x_train":425.9076940912,"imp":18.8066937887,"elo":1120.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":46.9242331846,"x_train":844.3867407395,"imp":14.9020596719,"elo":1230.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":66.5740410026,"x_train":2092.7290649042,"imp":9.6291821749,"elo":1439.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":65.026662372,"x_train":2173.0087869337,"imp":9.6037114724,"elo":1446.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1303369377,"x_train":2.275318476,"imp":10.7269469123,"elo":1263.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.1626576583,"x_train":4.3886320534,"imp":9.9095442967,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.175716895,"x_train":10.7096165053,"imp":8.8150655878,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.2654850598,"x_train":21.3971746465,"imp":7.8221470759,"elo":1487.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.3507721627,"x_train":53.3414186748,"imp":7.014236119,"elo":1560.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.4554854246,"x_train":106.1300468914,"imp":6.6825183636,"elo":1590.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.5400047224,"x_train":212.0315758458,"imp":6.4961289435,"elo":1614.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":0.5704772449,"x_train":317.9586579883,"imp":6.4098620166,"elo":1629.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":0.6149915399,"x_train":425.6136485357,"imp":6.3770966809,"elo":1642.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..3cf027a473a3d720e15dfd143630940c84f93b5c
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_regression/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned)", "TabDPT (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "TabICLv2 (default)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabDPT-Turbo (default)", "TabDPT (tuned)", "Nori-30M (default)", "RealTabPFN-2.5 (default)", "TabDPT (default)", "RealMLP (tuned)", "Nori (default)", "xRFM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "LightGBM (tuned)", "ModernNCA (tuned)", "TabM (tuned + ensembled)", "CatBoost (default)", "xRFM (tuned)", "TabM (tuned)", "XGBoost (tuned + ensembled)", "ModernNCA (default)", "XGBoost (tuned)", "iLTM (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "LimiX (default)", "XGBoost (default)", "ChimeraBoost (tuned)", "TabSwift (default)", "PerpetualBooster (tuned + ensembled)", "TabM (default)", "xRFM (default)", "TorchMLP (tuned + ensembled)", "LightGBM (default)", "ChimeraBoost (default)", "PerpetualBooster (tuned)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "RealMLP (default)", "ExtraTrees (tuned)", "RandomForest (tuned + ensembled)", "SAP-RPT-OSS (default)", "RandomForest (tuned)", "EBM (tuned + ensembled)", "TorchMLP (default)", "EBM (tuned)", "PerpetualBooster (default)", "RandomForest (default)", "ExtraTrees (default)", "EBM (default)", "iLTM (default)", "FastaiMLP (tuned + ensembled)", "FastaiMLP (tuned)", "KNN (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "FastaiMLP (default)", "TabSTAR (tuned)", "KNN (tuned)", "TabSTAR (default)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 1.0, 0.7333333492279053, 0.800000011920929, 1.0, 1.0, 0.6888889074325562, 1.0, 0.8888888955116272, 1.0, 1.0, 0.9555555582046509, 0.8444444537162781, 1.0, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, NaN, 0.2222222238779068, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.6000000238418579, 0.800000011920929, 0.2666666805744171, 0.9111111164093018, 0.7777777910232544, 0.6000000238418579, 0.6000000238418579, 0.8222222328186035, 1.0, 0.6000000238418579, 0.9555555582046509, 0.9555555582046509, 0.9777777791023254, 0.8222222328186035, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.9777777791023254, 0.8444444537162781, 1.0, 0.9777777791023254, 0.9777777791023254, 1.0, 0.9555555582046509, 1.0, 0.8444444537162781, 1.0, 1.0, 0.9555555582046509, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2666666805744171, 0.7777777910232544, NaN, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.42222222685813904, 0.800000011920929, 0.8222222328186035, 0.800000011920929, 0.800000011920929, 0.8444444537162781, 0.5333333611488342, 0.800000011920929, 1.0, 0.6888889074325562, 0.800000011920929, 0.9777777791023254, 0.8222222328186035, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.3333333432674408, 0.20000000298023224, NaN, 0.5555555820465088, 0.46666666865348816, 0.35555556416511536, 0.5555555820465088, 0.20000000298023224, 0.7333333492279053, 0.9777777791023254, 0.35555556416511536, 0.42222222685813904, 0.6666666865348816, 0.5111111402511597, 0.6000000238418579, 0.800000011920929, 0.8888888955116272, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2666666805744171, 0.20000000298023224, 0.4444444477558136, NaN, 0.5333333611488342, 0.5111111402511597, 0.6222222447395325, 0.24444444477558136, 0.8222222328186035, 0.644444465637207, 0.4444444477558136, 0.5777778029441833, 0.6666666865348816, 0.6222222447395325, 0.6000000238418579, 0.9555555582046509, 0.8222222328186035, 0.8888888955116272, 0.8222222328186035, 0.8222222328186035, 0.8222222328186035, 0.8444444537162781, 1.0, 0.9555555582046509, 0.9111111164093018, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.13333334028720856, 0.2222222238779068, 0.5333333611488342, 0.46666666865348816, NaN, 0.5333333611488342, 0.6000000238418579, 0.24444444477558136, 0.800000011920929, 0.6888889074325562, 0.5555555820465088, 0.5777778029441833, 0.7555555701255798, 0.800000011920929, 0.5777778029441833, 0.8444444537162781, 0.9111111164093018, 0.9333333373069763, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 0.8222222328186035, 1.0, 1.0, 0.8222222328186035, 1.0, 1.0, 0.9333333373069763, 1.0, 0.8444444537162781, 1.0, 0.8444444537162781, 1.0, 1.0, 0.8444444537162781, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.31111112236976624, 0.4000000059604645, 0.5777778029441833, 0.644444465637207, 0.4888888895511627, 0.46666666865348816, NaN, 0.46666666865348816, 0.4888888895511627, 0.4888888895511627, 0.7111111283302307, 0.7777777910232544, 0.9555555582046509, 0.6666666865348816, 0.7111111283302307, 0.9777777791023254, 0.5111111402511597, 0.8666666746139526, 0.8888888955116272, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.8222222328186035, 0.9111111164093018, 0.7333333492279053, 0.8888888955116272, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.7777777910232544, 0.9555555582046509, 0.800000011920929, 1.0, 0.8444444537162781, 0.8222222328186035, 0.8222222328186035, 0.9111111164093018, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.4444444477558136, 0.3777777850627899, 0.4000000059604645, 0.5333333611488342, NaN, 0.20000000298023224, 0.46666666865348816, 0.5777778029441833, 0.42222222685813904, 0.5333333611488342, 0.644444465637207, 0.6666666865348816, 0.6000000238418579, 0.8444444537162781, 0.7111111283302307, 0.8444444537162781, 0.7555555701255798, 0.7555555701255798, 0.7555555701255798, 0.8222222328186035, 1.0, 0.8444444537162781, 0.8666666746139526, 0.9777777791023254, 0.9555555582046509, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9777777791023254, 0.9111111164093018, 0.9777777791023254, 0.9333333373069763, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.7333333492279053, 0.17777778208255768, 0.800000011920929, 0.7555555701255798, 0.7555555701255798, 0.5111111402511597, 0.800000011920929, NaN, 0.7777777910232544, 0.800000011920929, 0.6000000238418579, 0.5777778029441833, 0.800000011920929, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.9555555582046509, 0.800000011920929, 0.9555555582046509, 0.800000011920929, 0.800000011920929, 0.8222222328186035, 1.0, 1.0, 1.0, 0.8444444537162781, 1.0, 0.8444444537162781, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.08888889104127884, 0.20000000298023224, 0.2666666805744171, 0.17777778208255768, 0.20000000298023224, 0.5111111402511597, 0.5333333611488342, 0.2222222238779068, NaN, 0.6222222447395325, 0.4444444477558136, 0.5333333611488342, 0.5111111402511597, 0.5555555820465088, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.8888888955116272, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.7555555701255798, 0.7777777910232544, 0.9333333373069763, 0.800000011920929, 0.9777777791023254, 0.9777777791023254, 0.9111111164093018, 1.0, 0.8666666746139526, 1.0, 0.800000011920929, 1.0, 0.9777777791023254, 0.8666666746139526, 0.8666666746139526, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2222222238779068, 0.20000000298023224, 0.02222222276031971, 0.35555556416511536, 0.31111112236976624, 0.2888889014720917, 0.42222222685813904, 0.20000000298023224, 0.3777777850627899, NaN, 0.2666666805744171, 0.3777777850627899, 0.4888888895511627, 0.4000000059604645, 0.5777778029441833, 0.7777777910232544, 0.800000011920929, 0.8444444537162781, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.7111111283302307, 0.8888888955116272, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 1.0, 0.9111111164093018, 1.0, 0.9777777791023254, 1.0, 0.9777777791023254, 0.800000011920929, 1.0, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.04444444552063942, 0.4000000059604645, 0.15555556118488312, 0.644444465637207, 0.5555555820465088, 0.4444444477558136, 0.2222222238779068, 0.5777778029441833, 0.4000000059604645, 0.5555555820465088, 0.7333333492279053, NaN, 0.31111112236976624, 0.5777778029441833, 0.7555555701255798, 0.5555555820465088, 0.6000000238418579, 0.8888888955116272, 0.7333333492279053, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.7777777910232544, 0.9777777791023254, 0.800000011920929, 0.9111111164093018, 0.9777777791023254, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.15555556118488312, 0.4000000059604645, 0.46666666865348816, 0.5777778029441833, 0.42222222685813904, 0.42222222685813904, 0.04444444552063942, 0.46666666865348816, 0.42222222685813904, 0.46666666865348816, 0.6222222447395325, 0.6888889074325562, NaN, 0.5777778029441833, 0.644444465637207, 0.8222222328186035, 0.46666666865348816, 0.8444444537162781, 0.8666666746139526, 0.7111111283302307, 0.7111111283302307, 0.7111111283302307, 0.7333333492279053, 0.7555555701255798, 0.9111111164093018, 0.7111111283302307, 0.8888888955116272, 0.9111111164093018, 0.7777777910232544, 0.9111111164093018, 0.7555555701255798, 0.9333333373069763, 0.800000011920929, 1.0, 0.800000011920929, 0.7777777910232544, 0.8222222328186035, 0.8888888955116272, 0.9555555582046509, 0.9111111164093018, 1.0, 1.0, 0.9777777791023254, 0.8888888955116272, 0.9777777791023254, 1.0, 0.9777777791023254, 0.9555555582046509, 1.0, 1.0, 0.9333333373069763, 1.0, 0.8444444537162781, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.17777778208255768, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.24444444477558136, 0.3333333432674408, 0.35555556416511536, 0.20000000298023224, 0.4888888895511627, 0.5111111402511597, 0.42222222685813904, 0.42222222685813904, NaN, 0.4888888895511627, 0.5333333611488342, 0.4888888895511627, 0.7555555701255798, 0.7333333492279053, 0.7555555701255798, 0.7555555701255798, 0.7555555701255798, 0.7555555701255798, 0.6888889074325562, 0.9333333373069763, 0.7555555701255798, 0.8666666746139526, 0.9555555582046509, 0.8888888955116272, 0.8444444537162781, 0.8888888955116272, 0.9555555582046509, 0.800000011920929, 0.9777777791023254, 0.6222222447395325, 0.8888888955116272, 0.8444444537162781, 0.7333333492279053, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 0.9333333373069763, 0.9777777791023254, 0.9555555582046509, 1.0, 1.0, 1.0, 0.9777777791023254, 0.9777777791023254, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.4888888895511627, 0.3777777850627899, 0.20000000298023224, 0.2888889014720917, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.6000000238418579, 0.24444444477558136, 0.35555556416511536, 0.5111111402511597, NaN, 0.3777777850627899, 0.5777778029441833, 0.6000000238418579, 0.6888889074325562, 0.7555555701255798, 0.800000011920929, 0.800000011920929, 0.7555555701255798, 0.800000011920929, 0.9555555582046509, 0.800000011920929, 0.7777777910232544, 0.9555555582046509, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8888888955116272, 0.8444444537162781, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.9777777791023254, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.8222222328186035, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.4000000059604645, 0.31111112236976624, 0.4000000059604645, 0.4000000059604645, 0.42222222685813904, 0.02222222276031971, 0.4000000059604645, 0.4444444477558136, 0.4000000059604645, 0.42222222685813904, 0.4444444477558136, 0.17777778208255768, 0.46666666865348816, 0.6222222447395325, NaN, 0.42222222685813904, 0.7555555701255798, 0.5777778029441833, 0.6666666865348816, 0.7111111283302307, 0.7111111283302307, 0.7333333492279053, 0.5111111402511597, 0.7111111283302307, 0.7111111283302307, 0.8444444537162781, 0.8444444537162781, 0.7333333492279053, 0.7555555701255798, 0.7333333492279053, 0.8888888955116272, 0.800000011920929, 0.9777777791023254, 0.800000011920929, 0.7555555701255798, 0.800000011920929, 0.8444444537162781, 0.7333333492279053, 0.9111111164093018, 0.9333333373069763, 1.0, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.8222222328186035, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.9777777791023254, 0.8222222328186035, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.20000000298023224, 0.20000000298023224, 0.04444444552063942, 0.15555556118488312, 0.4888888895511627, 0.15555556118488312, 0.2222222238779068, 0.20000000298023224, 0.2222222238779068, 0.4000000059604645, 0.5333333611488342, 0.5111111402511597, 0.42222222685813904, 0.5777778029441833, NaN, 0.42222222685813904, 0.5777778029441833, 0.5555555820465088, 0.5777778029441833, 0.5777778029441833, 0.6000000238418579, 0.7111111283302307, 0.7777777910232544, 0.6000000238418579, 0.800000011920929, 0.7777777910232544, 0.7333333492279053, 0.9777777791023254, 0.7333333492279053, 0.8222222328186035, 0.6222222447395325, 0.800000011920929, 0.9111111164093018, 0.8222222328186035, 0.7333333492279053, 0.7333333492279053, 0.9111111164093018, 0.9111111164093018, 0.8888888955116272, 0.9777777791023254, 0.8888888955116272, 0.8222222328186035, 1.0, 1.0, 0.8444444537162781, 0.9555555582046509, 0.8666666746139526, 0.9777777791023254, 1.0, 1.0, 0.9555555582046509, 1.0, 0.9777777791023254, 1.0, 1.0, 0.9111111164093018, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.02222222276031971, 0.1111111119389534, 0.17777778208255768, 0.08888889104127884, 0.13333334028720856, 0.2888889014720917, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.1111111119389534, 0.15555556118488312, 0.24444444477558136, 0.4000000059604645, 0.24444444477558136, 0.5777778029441833, NaN, 0.6000000238418579, 0.7555555701255798, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.3777777850627899, 0.7555555701255798, 0.7777777910232544, 0.8444444537162781, 0.8666666746139526, 0.800000011920929, 0.6888889074325562, 0.800000011920929, 0.8444444537162781, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.9777777791023254, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.02222222276031971, 0.17777778208255768, 0.06666667014360428, 0.1111111119389534, 0.06666667014360428, 0.1111111119389534, 0.15555556118488312, 0.20000000298023224, 0.1111111119389534, 0.15555556118488312, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.31111112236976624, 0.42222222685813904, 0.42222222685813904, 0.4000000059604645, NaN, 0.6000000238418579, 0.6888889074325562, 0.6888889074325562, 0.6222222447395325, 0.4000000059604645, 0.6222222447395325, 0.7111111283302307, 1.0, 0.7555555701255798, 0.7111111283302307, 0.7777777910232544, 0.7111111283302307, 0.8888888955116272, 0.6888889074325562, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7777777910232544, 0.8888888955116272, 0.9777777791023254, 1.0, 1.0, 0.9777777791023254, 0.7777777910232544, 1.0, 1.0, 0.9555555582046509, 1.0, 0.9777777791023254, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.17777778208255768, 0.20000000298023224, 0.0, 0.17777778208255768, 0.20000000298023224, 0.2666666805744171, 0.24444444477558136, 0.20000000298023224, 0.2222222238779068, 0.06666667014360428, 0.2222222238779068, 0.2888889014720917, 0.24444444477558136, 0.24444444477558136, 0.3333333432674408, 0.4444444477558136, 0.24444444477558136, 0.4000000059604645, NaN, 0.46666666865348816, 0.6000000238418579, 0.7333333492279053, 0.5333333611488342, 0.35555556416511536, 0.8222222328186035, 0.4888888895511627, 0.5111111402511597, 0.9777777791023254, 0.5777778029441833, 0.9555555582046509, 0.7555555701255798, 0.9777777791023254, 0.7777777910232544, 0.6000000238418579, 1.0, 1.0, 0.6888889074325562, 1.0, 0.9555555582046509, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.17777778208255768, 0.2222222238779068, 0.2666666805744171, 0.24444444477558136, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.2888889014720917, 0.24444444477558136, 0.20000000298023224, 0.2888889014720917, 0.42222222685813904, 0.2222222238779068, 0.31111112236976624, 0.5333333611488342, NaN, 0.6666666865348816, 0.644444465637207, 0.5333333611488342, 0.2666666805744171, 0.8222222328186035, 0.4888888895511627, 0.3777777850627899, 0.8444444537162781, 0.5555555820465088, 0.9111111164093018, 0.644444465637207, 0.9111111164093018, 0.7555555701255798, 0.6000000238418579, 0.9777777791023254, 0.9555555582046509, 0.6888889074325562, 0.9777777791023254, 0.9333333373069763, 0.7777777910232544, 0.800000011920929, 0.9777777791023254, 1.0, 1.0, 0.800000011920929, 0.9333333373069763, 1.0, 0.9555555582046509, 1.0, 0.800000011920929, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.17777778208255768, 0.20000000298023224, 0.2666666805744171, 0.24444444477558136, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.2888889014720917, 0.24444444477558136, 0.20000000298023224, 0.2888889014720917, 0.42222222685813904, 0.2222222238779068, 0.31111112236976624, 0.4000000059604645, 0.3333333432674408, NaN, 0.5777778029441833, 0.5333333611488342, 0.2666666805744171, 0.7111111283302307, 0.4888888895511627, 0.3777777850627899, 0.7333333492279053, 0.5555555820465088, 0.7555555701255798, 0.644444465637207, 0.9111111164093018, 0.6888889074325562, 0.6000000238418579, 0.9777777791023254, 0.9555555582046509, 0.6888889074325562, 0.9777777791023254, 0.9333333373069763, 0.800000011920929, 0.800000011920929, 0.9777777791023254, 0.9777777791023254, 1.0, 0.800000011920929, 0.9333333373069763, 1.0, 0.9555555582046509, 0.9777777791023254, 0.800000011920929, 1.0, 1.0, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.15555556118488312, 0.17777778208255768, 0.2666666805744171, 0.17777778208255768, 0.20000000298023224, 0.24444444477558136, 0.06666667014360428, 0.2222222238779068, 0.2666666805744171, 0.24444444477558136, 0.24444444477558136, 0.2666666805744171, 0.4000000059604645, 0.2222222238779068, 0.3777777850627899, 0.2666666805744171, 0.35555556416511536, 0.42222222685813904, NaN, 0.46666666865348816, 0.24444444477558136, 0.5555555820465088, 0.42222222685813904, 0.3333333432674408, 0.9333333373069763, 0.4888888895511627, 1.0, 0.6222222447395325, 0.9333333373069763, 0.800000011920929, 0.6000000238418579, 1.0, 0.9777777791023254, 0.6000000238418579, 0.9777777791023254, 0.9111111164093018, 0.7777777910232544, 0.800000011920929, 1.0, 0.9777777791023254, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.04444444552063942, 0.0, 0.0, 0.17777778208255768, 0.0, 0.20000000298023224, 0.2222222238779068, 0.2888889014720917, 0.2222222238779068, 0.24444444477558136, 0.31111112236976624, 0.20000000298023224, 0.4888888895511627, 0.2888889014720917, 0.6222222447395325, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, NaN, 0.6222222447395325, 0.46666666865348816, 0.6666666865348816, 0.7111111283302307, 0.6000000238418579, 0.8666666746139526, 0.6000000238418579, 0.7555555701255798, 0.6000000238418579, 0.800000011920929, 0.7555555701255798, 0.6888889074325562, 0.6666666865348816, 0.8222222328186035, 0.8222222328186035, 0.8888888955116272, 0.8444444537162781, 0.9333333373069763, 0.8222222328186035, 0.6666666865348816, 0.9333333373069763, 1.0, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 0.9111111164093018, 0.8888888955116272, 0.9777777791023254, 0.8222222328186035, 1.0, 0.9333333373069763, 1.0, 0.9777777791023254, 0.8888888955116272, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.02222222276031971, 0.0, 0.0, 0.04444444552063942, 0.0, 0.08888889104127884, 0.15555556118488312, 0.0, 0.06666667014360428, 0.1111111119389534, 0.02222222276031971, 0.08888889104127884, 0.06666667014360428, 0.04444444552063942, 0.2888889014720917, 0.2222222238779068, 0.24444444477558136, 0.3777777850627899, 0.644444465637207, 0.7333333492279053, 0.7333333492279053, 0.7555555701255798, 0.3777777850627899, NaN, 0.7333333492279053, 0.4444444477558136, 0.9555555582046509, 0.7777777910232544, 0.5111111402511597, 0.7777777910232544, 0.6888889074325562, 0.7777777910232544, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 1.0, 0.8444444537162781, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 0.8444444537162781, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.15555556118488312, 0.20000000298023224, 0.0, 0.08888889104127884, 0.17777778208255768, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.20000000298023224, 0.2888889014720917, 0.24444444477558136, 0.20000000298023224, 0.2888889014720917, 0.4000000059604645, 0.2222222238779068, 0.2888889014720917, 0.17777778208255768, 0.17777778208255768, 0.2888889014720917, 0.4444444477558136, 0.5333333611488342, 0.2666666805744171, NaN, 0.4888888895511627, 0.4000000059604645, 0.6666666865348816, 0.5555555820465088, 0.7111111283302307, 0.644444465637207, 0.6222222447395325, 0.6888889074325562, 0.6000000238418579, 0.800000011920929, 0.7555555701255798, 0.6888889074325562, 0.8888888955116272, 0.7333333492279053, 0.7777777910232544, 0.800000011920929, 0.7555555701255798, 0.7777777910232544, 0.9333333373069763, 0.800000011920929, 0.7111111283302307, 0.7777777910232544, 0.7111111283302307, 0.800000011920929, 0.6222222447395325, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.1111111119389534, 0.02222222276031971, 0.20000000298023224, 0.02222222276031971, 0.02222222276031971, 0.08888889104127884, 0.1111111119389534, 0.13333334028720856, 0.2222222238779068, 0.15555556118488312, 0.20000000298023224, 0.15555556118488312, 0.0, 0.5111111402511597, 0.5111111402511597, 0.5111111402511597, 0.5777778029441833, 0.3333333432674408, 0.5555555820465088, 0.5111111402511597, NaN, 0.5777778029441833, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7111111283302307, 0.6222222447395325, 0.800000011920929, 0.6000000238418579, 0.7111111283302307, 0.644444465637207, 0.7111111283302307, 0.7333333492279053, 0.9111111164093018, 0.9111111164093018, 0.9555555582046509, 0.8666666746139526, 0.644444465637207, 0.8666666746139526, 0.9777777791023254, 0.800000011920929, 0.9555555582046509, 0.8666666746139526, 0.9555555582046509, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.9777777791023254, 1.0, 0.9555555582046509, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.02222222276031971, 0.0, 0.0, 0.02222222276031971, 0.0, 0.06666667014360428, 0.04444444552063942, 0.0, 0.02222222276031971, 0.04444444552063942, 0.02222222276031971, 0.08888889104127884, 0.04444444552063942, 0.04444444552063942, 0.15555556118488312, 0.2222222238779068, 0.13333334028720856, 0.24444444477558136, 0.4888888895511627, 0.6222222447395325, 0.6222222447395325, 0.6666666865348816, 0.2888889014720917, 0.04444444552063942, 0.6000000238418579, 0.42222222685813904, NaN, 0.7333333492279053, 0.4000000059604645, 0.7333333492279053, 0.46666666865348816, 0.7333333492279053, 0.7777777910232544, 0.6000000238418579, 0.7555555701255798, 0.7333333492279053, 0.6222222447395325, 0.800000011920929, 0.9777777791023254, 0.8444444537162781, 0.6000000238418579, 0.7555555701255798, 0.7555555701255798, 0.800000011920929, 0.7111111283302307, 0.9555555582046509, 1.0, 1.0, 0.800000011920929, 0.8222222328186035, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.08888889104127884, 0.0, 0.20000000298023224, 0.2222222238779068, 0.1111111119389534, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2888889014720917, 0.02222222276031971, 0.15555556118488312, 0.2666666805744171, 0.06666667014360428, 0.4000000059604645, 0.2222222238779068, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, NaN, 0.4444444477558136, 0.5333333611488342, 0.4444444477558136, 0.5555555820465088, 0.7333333492279053, 0.6000000238418579, 0.7777777910232544, 0.5777778029441833, 0.5777778029441833, 0.8666666746139526, 0.6888889074325562, 0.7555555701255798, 0.800000011920929, 0.9777777791023254, 0.8444444537162781, 0.9555555582046509, 0.800000011920929, 0.8666666746139526, 0.9111111164093018, 0.9111111164093018, 1.0, 0.800000011920929, 1.0, 0.9777777791023254, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.06666667014360428, 0.0, 0.20000000298023224, 0.0, 0.08888889104127884, 0.06666667014360428, 0.08888889104127884, 0.15555556118488312, 0.20000000298023224, 0.24444444477558136, 0.02222222276031971, 0.31111112236976624, 0.2222222238779068, 0.42222222685813904, 0.4444444477558136, 0.4444444477558136, 0.5111111402511597, 0.13333334028720856, 0.4888888895511627, 0.4444444477558136, 0.4000000059604645, 0.6000000238418579, 0.5555555820465088, NaN, 0.5777778029441833, 0.5333333611488342, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6222222447395325, 0.6222222447395325, 0.7555555701255798, 0.800000011920929, 0.8888888955116272, 0.7777777910232544, 0.6222222447395325, 0.800000011920929, 0.9777777791023254, 0.800000011920929, 0.8444444537162781, 0.800000011920929, 0.8888888955116272, 0.8666666746139526, 0.9555555582046509, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.9555555582046509, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.20000000298023224, 0.0, 0.0, 0.15555556118488312, 0.2222222238779068, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.0, 0.20000000298023224, 0.24444444477558136, 0.1111111119389534, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2888889014720917, 0.04444444552063942, 0.08888889104127884, 0.24444444477558136, 0.0, 0.4000000059604645, 0.2222222238779068, 0.2888889014720917, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.42222222685813904, NaN, 0.3777777850627899, 0.5333333611488342, 0.644444465637207, 0.6000000238418579, 0.800000011920929, 0.5555555820465088, 0.5555555820465088, 0.7555555701255798, 0.644444465637207, 0.6888889074325562, 0.7777777910232544, 0.8666666746139526, 0.8222222328186035, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.8888888955116272, 0.9111111164093018, 1.0, 0.800000011920929, 1.0, 0.9777777791023254, 0.9111111164093018, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.04444444552063942, 0.02222222276031971, 0.20000000298023224, 0.0, 0.02222222276031971, 0.08888889104127884, 0.06666667014360428, 0.04444444552063942, 0.20000000298023224, 0.1111111119389534, 0.17777778208255768, 0.15555556118488312, 0.1111111119389534, 0.24444444477558136, 0.35555556416511536, 0.35555556416511536, 0.3777777850627899, 0.24444444477558136, 0.31111112236976624, 0.35555556416511536, 0.2888889014720917, 0.5333333611488342, 0.5555555820465088, 0.46666666865348816, 0.6222222447395325, NaN, 0.5111111402511597, 0.8444444537162781, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5777778029441833, 0.644444465637207, 0.7111111283302307, 0.9333333373069763, 0.9111111164093018, 0.8888888955116272, 0.5777778029441833, 0.8666666746139526, 0.9555555582046509, 0.8888888955116272, 0.9111111164093018, 0.8888888955116272, 0.9777777791023254, 0.7777777910232544, 0.9777777791023254, 0.7777777910232544, 0.9777777791023254, 0.7777777910232544, 0.9777777791023254, 0.9777777791023254, 0.9111111164093018, 0.7777777910232544, 0.9555555582046509, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254], [0.0, 0.15555556118488312, 0.20000000298023224, 0.0, 0.04444444552063942, 0.15555556118488312, 0.20000000298023224, 0.08888889104127884, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.20000000298023224, 0.31111112236976624, 0.02222222276031971, 0.08888889104127884, 0.08888889104127884, 0.06666667014360428, 0.4000000059604645, 0.2222222238779068, 0.3777777850627899, 0.3777777850627899, 0.2666666805744171, 0.4444444477558136, 0.4000000059604645, 0.46666666865348816, 0.4888888895511627, NaN, 0.5111111402511597, 0.4000000059604645, 0.7777777910232544, 0.8222222328186035, 0.6000000238418579, 0.7777777910232544, 0.5777778029441833, 0.5777778029441833, 0.6000000238418579, 0.7777777910232544, 0.9555555582046509, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.7555555701255798, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.644444465637207, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.02222222276031971, 0.0, 0.0, 0.02222222276031971, 0.1111111119389534, 0.02222222276031971, 0.20000000298023224, 0.0, 0.06666667014360428, 0.2222222238779068, 0.24444444477558136, 0.31111112236976624, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.31111112236976624, 0.20000000298023224, 0.2222222238779068, 0.2666666805744171, 0.20000000298023224, 0.35555556416511536, 0.15555556118488312, 0.4888888895511627, NaN, 0.6000000238418579, 0.4888888895511627, 0.4888888895511627, 0.4888888895511627, 0.4444444477558136, 0.6000000238418579, 0.8444444537162781, 0.7111111283302307, 0.7555555701255798, 0.5777778029441833, 0.6222222447395325, 0.8222222328186035, 0.7777777910232544, 0.8666666746139526, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.8222222328186035, 1.0, 0.9111111164093018, 0.800000011920929, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.15555556118488312, 0.06666667014360428, 0.0, 0.02222222276031971, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.15555556118488312, 0.20000000298023224, 0.08888889104127884, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.24444444477558136, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4444444477558136, 0.6000000238418579, 0.4000000059604645, NaN, 0.46666666865348816, 0.6000000238418579, 0.5777778029441833, 0.42222222685813904, 0.4444444477558136, 0.4444444477558136, 0.5333333611488342, 0.42222222685813904, 0.6000000238418579, 0.46666666865348816, 0.644444465637207, 0.4000000059604645, 0.6888889074325562, 0.4000000059604645, 0.6000000238418579, 0.9777777791023254, 0.644444465637207, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.7111111283302307, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.20000000298023224, 0.0, 0.0, 0.15555556118488312, 0.17777778208255768, 0.02222222276031971, 0.20000000298023224, 0.13333334028720856, 0.0, 0.20000000298023224, 0.2222222238779068, 0.1111111119389534, 0.20000000298023224, 0.24444444477558136, 0.17777778208255768, 0.20000000298023224, 0.20000000298023224, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.31111112236976624, 0.20000000298023224, 0.20000000298023224, 0.2888889014720917, 0.24444444477558136, 0.2222222238779068, 0.4000000059604645, 0.20000000298023224, 0.3333333432674408, 0.2222222238779068, 0.5111111402511597, 0.5333333611488342, NaN, 0.3333333432674408, 0.4888888895511627, 0.4888888895511627, 0.35555556416511536, 0.5333333611488342, 0.7111111283302307, 0.46666666865348816, 0.5333333611488342, 0.7777777910232544, 0.7555555701255798, 0.4444444477558136, 0.6000000238418579, 0.46666666865348816, 0.7777777910232544, 0.6222222447395325, 0.800000011920929, 0.9777777791023254, 0.800000011920929, 1.0, 1.0, 1.0, 0.7777777910232544, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.04444444552063942, 0.20000000298023224, 0.0, 0.02222222276031971, 0.1111111119389534, 0.17777778208255768, 0.02222222276031971, 0.20000000298023224, 0.13333334028720856, 0.0, 0.20000000298023224, 0.17777778208255768, 0.15555556118488312, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.0, 0.04444444552063942, 0.04444444552063942, 0.02222222276031971, 0.3333333432674408, 0.20000000298023224, 0.24444444477558136, 0.35555556416511536, 0.2666666805744171, 0.42222222685813904, 0.4000000059604645, 0.4444444477558136, 0.4444444477558136, 0.17777778208255768, 0.5111111402511597, 0.4000000059604645, 0.6666666865348816, NaN, 0.6000000238418579, 0.7333333492279053, 0.5333333611488342, 0.5555555820465088, 0.6000000238418579, 0.7333333492279053, 0.8444444537162781, 0.7777777910232544, 0.6000000238418579, 0.6888889074325562, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.7555555701255798, 0.6222222447395325, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8888888955116272, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.8444444537162781, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.04444444552063942, 0.0, 0.04444444552063942, 0.15555556118488312, 0.0, 0.1111111119389534, 0.2666666805744171, 0.0, 0.15555556118488312, 0.2666666805744171, 0.2222222238779068, 0.2222222238779068, 0.31111112236976624, 0.31111112236976624, 0.31111112236976624, 0.4000000059604645, 0.17777778208255768, 0.3333333432674408, 0.31111112236976624, 0.2888889014720917, 0.3777777850627899, 0.42222222685813904, 0.3777777850627899, 0.4444444477558136, 0.42222222685813904, 0.4000000059604645, 0.5111111402511597, 0.42222222685813904, 0.5111111402511597, 0.4000000059604645, NaN, 0.46666666865348816, 0.3777777850627899, 0.6000000238418579, 0.6222222447395325, 0.5333333611488342, 0.42222222685813904, 0.5333333611488342, 0.6222222447395325, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6222222447395325, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.6888889074325562, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.04444444552063942, 0.02222222276031971, 0.20000000298023224, 0.2666666805744171, 0.08888889104127884, 0.20000000298023224, 0.1111111119389534, 0.0, 0.02222222276031971, 0.02222222276031971, 0.02222222276031971, 0.17777778208255768, 0.20000000298023224, 0.1111111119389534, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.3777777850627899, 0.24444444477558136, 0.35555556416511536, 0.2222222238779068, 0.5555555820465088, 0.5777778029441833, 0.5111111402511597, 0.2666666805744171, 0.5333333611488342, NaN, 0.3777777850627899, 0.6222222447395325, 0.7555555701255798, 0.5555555820465088, 0.3777777850627899, 0.9777777791023254, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.7555555701255798, 0.644444465637207, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8444444537162781, 1.0, 0.9555555582046509, 0.9333333373069763, 0.8888888955116272, 1.0, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.02222222276031971, 0.0, 0.08888889104127884, 0.08888889104127884, 0.0, 0.02222222276031971, 0.04444444552063942, 0.06666667014360428, 0.06666667014360428, 0.08888889104127884, 0.1111111119389534, 0.0, 0.2666666805744171, 0.08888889104127884, 0.02222222276031971, 0.31111112236976624, 0.24444444477558136, 0.35555556416511536, 0.2888889014720917, 0.42222222685813904, 0.4000000059604645, 0.5555555820465088, 0.644444465637207, 0.46666666865348816, 0.6222222447395325, 0.6222222447395325, NaN, 0.42222222685813904, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.6888889074325562, 0.6000000238418579, 0.6222222447395325, 0.8222222328186035, 0.6222222447395325, 0.7333333492279053, 0.800000011920929, 0.7555555701255798, 0.800000011920929, 0.6222222447395325, 0.800000011920929, 0.800000011920929, 0.9555555582046509, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.06666667014360428, 0.1111111119389534, 0.0, 0.0, 0.20000000298023224, 0.2222222238779068, 0.20000000298023224, 0.2222222238779068, 0.15555556118488312, 0.15555556118488312, 0.2222222238779068, 0.08888889104127884, 0.15555556118488312, 0.24444444477558136, 0.20000000298023224, 0.31111112236976624, 0.06666667014360428, 0.42222222685813904, 0.15555556118488312, 0.5555555820465088, 0.46666666865348816, 0.4444444477558136, 0.4000000059604645, 0.3777777850627899, 0.5777778029441833, NaN, 0.3333333432674408, 0.42222222685813904, 0.4444444477558136, 0.5333333611488342, 0.46666666865348816, 0.7777777910232544, 0.6888889074325562, 0.8222222328186035, 0.7333333492279053, 0.800000011920929, 0.7555555701255798, 0.6222222447395325, 0.9555555582046509, 0.644444465637207, 0.7777777910232544, 0.9555555582046509, 0.8888888955116272, 0.7333333492279053, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.02222222276031971, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.2222222238779068, 0.20000000298023224, 0.04444444552063942, 0.4000000059604645, 0.20000000298023224, 0.1111111119389534, 0.2222222238779068, 0.08888889104127884, 0.4000000059604645, 0.2888889014720917, 0.46666666865348816, 0.2888889014720917, 0.4000000059604645, 0.3777777850627899, 0.24444444477558136, 0.46666666865348816, 0.6666666865348816, NaN, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.9777777791023254, 0.6666666865348816, 0.5777778029441833, 0.6666666865348816, 0.7777777910232544, 0.800000011920929, 0.8444444537162781, 0.7111111283302307, 1.0, 0.7777777910232544, 0.7777777910232544, 0.8444444537162781, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.9111111164093018, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.04444444552063942, 0.0, 0.0, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.06666667014360428, 0.1111111119389534, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.17777778208255768, 0.20000000298023224, 0.24444444477558136, 0.13333334028720856, 0.24444444477558136, 0.02222222276031971, 0.2222222238779068, 0.13333334028720856, 0.1111111119389534, 0.2222222238779068, 0.24444444477558136, 0.5777778029441833, 0.5333333611488342, 0.2666666805744171, 0.46666666865348816, 0.4444444477558136, 0.4000000059604645, 0.5777778029441833, 0.6000000238418579, NaN, 0.5111111402511597, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.7111111283302307, 0.6222222447395325, 0.6666666865348816, 0.7555555701255798, 0.644444465637207, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.06666667014360428, 0.02222222276031971, 0.20000000298023224, 0.04444444552063942, 0.0, 0.20000000298023224, 0.1111111119389534, 0.04444444552063942, 0.20000000298023224, 0.20000000298023224, 0.17777778208255768, 0.20000000298023224, 0.2222222238779068, 0.0, 0.0, 0.02222222276031971, 0.02222222276031971, 0.3333333432674408, 0.20000000298023224, 0.2222222238779068, 0.35555556416511536, 0.24444444477558136, 0.15555556118488312, 0.3777777850627899, 0.17777778208255768, 0.42222222685813904, 0.04444444552063942, 0.42222222685813904, 0.4000000059604645, 0.46666666865348816, 0.15555556118488312, 0.5777778029441833, 0.6222222447395325, 0.46666666865348816, 0.5555555820465088, 0.6000000238418579, 0.4888888895511627, NaN, 0.7555555701255798, 0.6000000238418579, 0.4444444477558136, 0.5777778029441833, 0.46666666865348816, 0.5555555820465088, 0.42222222685813904, 0.6000000238418579, 0.6888889074325562, 0.6222222447395325, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7777777910232544, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8444444537162781, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.04444444552063942, 0.02222222276031971, 0.02222222276031971, 0.20000000298023224, 0.06666667014360428, 0.0, 0.08888889104127884, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.04444444552063942, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.3777777850627899, 0.5333333611488342, 0.2222222238779068, 0.2222222238779068, 0.46666666865348816, 0.02222222276031971, 0.31111112236976624, 0.46666666865348816, 0.6666666865348816, 0.46666666865348816, 0.24444444477558136, NaN, 0.7111111283302307, 0.5333333611488342, 0.5111111402511597, 0.5555555820465088, 0.6222222447395325, 0.6222222447395325, 0.7555555701255798, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9777777791023254, 0.8888888955116272, 0.7555555701255798, 0.800000011920929, 1.0, 0.9777777791023254, 0.9777777791023254, 0.9555555582046509, 1.0, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.17777778208255768, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.2888889014720917, 0.20000000298023224, 0.02222222276031971, 0.20000000298023224, 0.04444444552063942, 0.4000000059604645, 0.17777778208255768, 0.35555556416511536, 0.24444444477558136, 0.4000000059604645, 0.3777777850627899, 0.20000000298023224, 0.4000000059604645, 0.5333333611488342, 0.02222222276031971, 0.4000000059604645, 0.4000000059604645, 0.2888889014720917, NaN, 0.6222222447395325, 0.46666666865348816, 0.6222222447395325, 0.6222222447395325, 0.800000011920929, 0.644444465637207, 0.644444465637207, 1.0, 0.6666666865348816, 0.7111111283302307, 0.7111111283302307, 0.644444465637207, 0.7555555701255798, 0.8888888955116272, 0.8888888955116272, 0.9555555582046509, 1.0, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.02222222276031971, 0.0, 0.06666667014360428, 0.15555556118488312, 0.0, 0.04444444552063942, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.20000000298023224, 0.0, 0.2888889014720917, 0.20000000298023224, 0.04444444552063942, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.1111111119389534, 0.2666666805744171, 0.2222222238779068, 0.6000000238418579, 0.5555555820465088, 0.31111112236976624, 0.4000000059604645, 0.4000000059604645, 0.3777777850627899, 0.2222222238779068, 0.3333333432674408, 0.4000000059604645, 0.5555555820465088, 0.46666666865348816, 0.3777777850627899, NaN, 0.5111111402511597, 0.9111111164093018, 0.6000000238418579, 0.800000011920929, 0.4888888895511627, 0.6000000238418579, 0.42222222685813904, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.04444444552063942, 0.0, 0.06666667014360428, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.0, 0.2222222238779068, 0.04444444552063942, 0.0, 0.08888889104127884, 0.15555556118488312, 0.1111111119389534, 0.08888889104127884, 0.2666666805744171, 0.13333334028720856, 0.31111112236976624, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.17777778208255768, 0.31111112236976624, 0.42222222685813904, 0.2888889014720917, 0.42222222685813904, 0.4888888895511627, 0.5333333611488342, 0.4888888895511627, NaN, 0.5333333611488342, 0.644444465637207, 0.644444465637207, 0.6888889074325562, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.7555555701255798, 0.8444444537162781, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.8444444537162781, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.02222222276031971, 0.0, 0.04444444552063942, 0.04444444552063942, 0.0, 0.20000000298023224, 0.0, 0.2888889014720917, 0.13333334028720856, 0.0, 0.08888889104127884, 0.20000000298023224, 0.08888889104127884, 0.1111111119389534, 0.24444444477558136, 0.20000000298023224, 0.6000000238418579, 0.5333333611488342, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.3777777850627899, 0.17777778208255768, 0.3333333432674408, 0.3777777850627899, 0.5333333611488342, 0.4444444477558136, 0.3777777850627899, 0.08888889104127884, 0.46666666865348816, NaN, 0.46666666865348816, 0.800000011920929, 0.4444444477558136, 0.6000000238418579, 0.42222222685813904, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.08888889104127884, 0.20000000298023224, 0.20000000298023224, 0.04444444552063942, 0.20000000298023224, 0.0, 0.1111111119389534, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.4000000059604645, 0.2222222238779068, 0.20000000298023224, 0.4000000059604645, 0.24444444477558136, 0.2666666805744171, 0.2666666805744171, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3777777850627899, 0.3777777850627899, 0.4000000059604645, 0.35555556416511536, 0.5333333611488342, NaN, 0.7777777910232544, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.9777777791023254, 0.644444465637207, 0.6666666865348816, 0.8888888955116272, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.08888889104127884, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.04444444552063942, 0.06666667014360428, 0.06666667014360428, 0.0, 0.17777778208255768, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1111111119389534, 0.15555556118488312, 0.3777777850627899, 0.20000000298023224, 0.17777778208255768, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.2222222238779068, 0.4000000059604645, 0.20000000298023224, 0.02222222276031971, 0.3777777850627899, 0.4000000059604645, 0.2666666805744171, 0.35555556416511536, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.24444444477558136, 0.5777778029441833, 0.3777777850627899, 0.20000000298023224, 0.20000000298023224, 0.35555556416511536, 0.20000000298023224, 0.2222222238779068, NaN, 0.2222222238779068, 0.7555555701255798, 0.5111111402511597, 0.800000011920929, 0.6000000238418579, 0.42222222685813904, 0.6000000238418579, 0.800000011920929, 0.5777778029441833, 0.8222222328186035, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.04444444552063942, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.35555556416511536, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.20000000298023224, 0.24444444477558136, 0.24444444477558136, 0.15555556118488312, 0.35555556416511536, 0.4000000059604645, 0.24444444477558136, 0.35555556416511536, 0.5111111402511597, 0.31111112236976624, 0.5555555820465088, 0.4000000059604645, 0.7777777910232544, NaN, 0.6000000238418579, 0.7111111283302307, 0.6666666865348816, 0.7111111283302307, 0.7111111283302307, 0.6222222447395325, 0.6666666865348816, 0.8888888955116272, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.20000000298023224, 0.04444444552063942, 0.0, 0.20000000298023224, 0.15555556118488312, 0.04444444552063942, 0.20000000298023224, 0.20000000298023224, 0.04444444552063942, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.17777778208255768, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.20000000298023224, 0.02222222276031971, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.24444444477558136, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.2888889014720917, 0.20000000298023224, 0.31111112236976624, 0.20000000298023224, 0.35555556416511536, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.24444444477558136, 0.4000000059604645, NaN, 0.6666666865348816, 1.0, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 1.0, 0.4888888895511627, 0.6000000238418579, 0.6000000238418579, 0.6888889074325562, 0.8888888955116272, 0.7555555701255798, 0.9777777791023254, 0.7777777910232544, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.08888889104127884, 0.08888889104127884, 0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.08888889104127884, 0.0, 0.08888889104127884, 0.02222222276031971, 0.35555556416511536, 0.0, 0.20000000298023224, 0.20000000298023224, 0.3777777850627899, 0.2222222238779068, 0.20000000298023224, 0.3777777850627899, 0.04444444552063942, 0.0, 0.20000000298023224, 0.3777777850627899, 0.20000000298023224, 0.0, 0.5777778029441833, 0.4000000059604645, 0.5777778029441833, 0.2666666805744171, 0.4888888895511627, 0.2888889014720917, 0.3333333432674408, NaN, 0.46666666865348816, 0.4000000059604645, 0.4888888895511627, 0.6000000238418579, 0.46666666865348816, 0.5777778029441833, 0.7555555701255798, 0.7777777910232544, 0.9111111164093018, 0.8222222328186035, 0.800000011920929, 0.8444444537162781, 1.0, 0.9555555582046509, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.02222222276031971, 0.0, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.20000000298023224, 0.15555556118488312, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.15555556118488312, 0.20000000298023224, 0.35555556416511536, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.0, 0.5333333611488342, NaN, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.9333333373069763, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.8666666746139526, 0.7555555701255798, 0.9555555582046509, 0.7555555701255798, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.17777778208255768, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.17777778208255768, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.2222222238779068, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.02222222276031971, 0.2888889014720917, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.4000000059604645, 0.2888889014720917, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, NaN, 0.2888889014720917, 0.4000000059604645, 0.6222222447395325, 0.7333333492279053, 0.5777778029441833, 0.7111111283302307, 0.6888889074325562, 1.0, 0.9333333373069763, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.31111112236976624, 0.04444444552063942, 0.04444444552063942, 0.04444444552063942, 0.15555556118488312, 0.0, 0.20000000298023224, 0.1111111119389534, 0.2888889014720917, 0.20000000298023224, 0.24444444477558136, 0.20000000298023224, 0.02222222276031971, 0.5777778029441833, 0.2888889014720917, 0.6000000238418579, 0.5111111402511597, 0.6666666865348816, 0.7111111283302307, NaN, 0.5555555820465088, 0.6666666865348816, 0.6888889074325562, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9777777791023254, 0.9777777791023254, 0.9777777791023254, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.02222222276031971, 0.08888889104127884, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.02222222276031971, 0.04444444552063942, 0.0, 0.0, 0.20000000298023224, 0.0, 0.08888889104127884, 0.20000000298023224, 0.08888889104127884, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.1111111119389534, 0.2666666805744171, 0.06666667014360428, 0.2222222238779068, 0.24444444477558136, 0.35555556416511536, 0.0, 0.15555556118488312, 0.0, 0.35555556416511536, 0.4000000059604645, 0.3777777850627899, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4444444477558136, NaN, 0.6000000238418579, 0.7777777910232544, 0.6000000238418579, 0.644444465637207, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.15555556118488312, 0.02222222276031971, 0.0, 0.20000000298023224, 0.17777778208255768, 0.02222222276031971, 0.20000000298023224, 0.08888889104127884, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.2222222238779068, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.1111111119389534, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.24444444477558136, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.0, 0.5333333611488342, 0.06666667014360428, 0.3777777850627899, 0.3333333432674408, 0.4000000059604645, NaN, 0.4000000059604645, 0.5555555820465088, 0.6000000238418579, 0.6000000238418579, 0.8444444537162781, 0.6666666865348816, 0.8888888955116272, 0.644444465637207, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.20000000298023224, 0.02222222276031971, 0.2888889014720917, 0.02222222276031971, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.02222222276031971, 0.06666667014360428, 0.0, 0.20000000298023224, 0.0, 0.1111111119389534, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1111111119389534, 0.42222222685813904, 0.1111111119389534, 0.5111111402511597, 0.42222222685813904, 0.6000000238418579, 0.2666666805744171, 0.31111112236976624, 0.2222222238779068, 0.6000000238418579, NaN, 0.5333333611488342, 0.5777778029441833, 0.644444465637207, 0.8444444537162781, 0.8222222328186035, 0.8444444537162781, 0.800000011920929, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.0, 0.20000000298023224, 0.02222222276031971, 0.1111111119389534, 0.20000000298023224, 0.0, 0.20000000298023224, 0.06666667014360428, 0.17777778208255768, 0.13333334028720856, 0.4000000059604645, 0.24444444477558136, 0.4000000059604645, 0.42222222685813904, 0.13333334028720856, 0.4000000059604645, 0.4444444477558136, 0.46666666865348816, NaN, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 1.0, 0.800000011920929, 0.5777778029441833, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.20000000298023224, 0.02222222276031971, 0.04444444552063942, 0.20000000298023224, 0.0, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.4000000059604645, 0.2222222238779068, 0.4000000059604645, 0.2888889014720917, 0.06666667014360428, 0.35555556416511536, 0.4000000059604645, 0.42222222685813904, 0.20000000298023224, NaN, 0.4000000059604645, 0.800000011920929, 0.9111111164093018, 0.800000011920929, 0.5777778029441833, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.08888889104127884, 0.0, 0.0, 0.0, 0.1111111119389534, 0.02222222276031971, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.15555556118488312, 0.04444444552063942, 0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.20000000298023224, 0.0, 0.31111112236976624, 0.08888889104127884, 0.4000000059604645, 0.31111112236976624, 0.06666667014360428, 0.2666666805744171, 0.4000000059604645, 0.35555556416511536, 0.6000000238418579, 0.6000000238418579, NaN, 0.4888888895511627, 0.6000000238418579, 0.4888888895511627, 1.0, 0.5777778029441833, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.1111111119389534, 0.17777778208255768, 0.13333334028720856, 0.0, 0.02222222276031971, 0.2666666805744171, 0.15555556118488312, 0.15555556118488312, 0.20000000298023224, 0.20000000298023224, 0.5111111402511597, NaN, 0.24444444477558136, 0.7555555701255798, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.0, 0.08888889104127884, 0.02222222276031971, 0.24444444477558136, 0.20000000298023224, 0.24444444477558136, 0.06666667014360428, 0.02222222276031971, 0.06666667014360428, 0.3333333432674408, 0.17777778208255768, 0.0, 0.08888889104127884, 0.4000000059604645, 0.7555555701255798, NaN, 0.7555555701255798, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.02222222276031971, 0.15555556118488312, 0.04444444552063942, 0.0, 0.02222222276031971, 0.2666666805744171, 0.1111111119389534, 0.15555556118488312, 0.20000000298023224, 0.20000000298023224, 0.5111111402511597, 0.24444444477558136, 0.24444444477558136, NaN, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.0, 0.24444444477558136, 0.2222222238779068, 0.0, 0.20000000298023224, 0.35555556416511536, 0.20000000298023224, 0.42222222685813904, 0.42222222685813904, 0.0, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, NaN, 0.4000000059604645, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.15555556118488312, 0.0, 0.0, 0.0, 0.15555556118488312, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.04444444552063942, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.42222222685813904, 0.0, 0.20000000298023224, 0.0, 0.6000000238418579, NaN, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.13333334028720856, 0.0, 0.0, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, NaN, 0.800000011920929, 0.800000011920929, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, NaN, 0.8777777552604675, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.12222222238779068, NaN, 0.9111111164093018], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02222222276031971, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.08888889104127884, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9827856035},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9023474218},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8910798179},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8785602536},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8723004737},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8669796605},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.859154935},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8441314599},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8413145621},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8259780962},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8250391267},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.8219092401},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.8206572829},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.791236312},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.769640069},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7608763765},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7339593188},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.7236306787},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.7064162815},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7051643241},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6901408508},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6757433548},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6751173753},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6685446079},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6435054839},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6162754383},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.6159624495},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.595931149},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5921752797},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5865414786},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5824726192},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5705790367},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5217527473},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5076682363},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.5054773151},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4913928066},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4898278652},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.4885759082},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4798122131},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4535211337},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.4425665151},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4384976571},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4331768448},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4269170652},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.4087636984},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3940532134},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3884194113},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.3683881119},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3661971882},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3589984394},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3496087691},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3424100201},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3211267646},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2857590025},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.2841940576},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2760563425},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2748043856},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2622848248},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.2604068904},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2212832591},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.1934272321},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.1752738676},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1605633828},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.1289514888},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1258215979},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.1183098609},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1154929588},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.0854460106},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0682316129},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0295774645},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0176838812},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0031298905}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..1edad74d2600c0362051b2a0519b19b9032d8962
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,73 @@
+,TabFM (default),RealTabPFN-2.5 (tuned + ensembled),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),RealTabPFN-2.5 (tuned),TabDPT (tuned + ensembled),ModernNCA (tuned + ensembled),TabICLv2 (default),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabDPT-Turbo (default),TabDPT (tuned),Nori-30M (default),RealTabPFN-2.5 (default),TabDPT (default),RealMLP (tuned),Nori (default),xRFM (tuned + ensembled),LightGBM (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),LightGBM (tuned),ModernNCA (tuned),TabM (tuned + ensembled),CatBoost (default),xRFM (tuned),TabM (tuned),XGBoost (tuned + ensembled),ModernNCA (default),XGBoost (tuned),iLTM (tuned + ensembled),ChimeraBoost (tuned + ensembled),iLTM (tuned),LimiX (default),XGBoost (default),ChimeraBoost (tuned),TabSwift (default),PerpetualBooster (tuned + ensembled),TabM (default),xRFM (default),TorchMLP (tuned + ensembled),LightGBM (default),ChimeraBoost (default),PerpetualBooster (tuned),TorchMLP (tuned),ExtraTrees (tuned + ensembled),RealMLP (default),ExtraTrees (tuned),RandomForest (tuned + ensembled),SAP-RPT-OSS (default),RandomForest (tuned),EBM (tuned + ensembled),TorchMLP (default),EBM (tuned),PerpetualBooster (default),RandomForest (default),ExtraTrees (default),EBM (default),iLTM (default),FastaiMLP (tuned + ensembled),FastaiMLP (tuned),KNN (tuned + ensembled),TabSTAR (tuned + ensembled),FastaiMLP (default),TabSTAR (tuned),KNN (tuned),TabSTAR (default),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,1.0,0.73333335,0.8,1.0,1.0,0.6888889,1.0,0.8888889,1.0,1.0,0.95555556,0.84444445,1.0,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,,0.22222222,0.6666667,0.73333335,0.8666667,0.6,0.8,0.26666668,0.9111111,0.7777778,0.6,0.6,0.82222223,1.0,0.6,0.95555556,0.95555556,0.9777778,0.82222223,0.8,0.8,0.8,1.0,0.9777778,0.84444445,1.0,0.9777778,0.9777778,1.0,0.95555556,1.0,0.84444445,1.0,1.0,0.95555556,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.26666668,0.7777778,,0.8,0.8,0.7777778,0.42222223,0.8,0.82222223,0.8,0.8,0.84444445,0.53333336,0.8,1.0,0.6888889,0.8,0.9777778,0.82222223,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,0.8,1.0,0.8,0.93333334,0.8,1.0,1.0,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,0.8,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,0.84444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2,0.33333334,0.2,,0.5555556,0.46666667,0.35555556,0.5555556,0.2,0.73333335,0.9777778,0.35555556,0.42222223,0.6666667,0.51111114,0.6,0.8,0.8888889,0.93333334,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.0,0.26666668,0.2,0.44444445,,0.53333336,0.51111114,0.62222224,0.24444444,0.82222223,0.64444447,0.44444445,0.5777778,0.6666667,0.62222224,0.6,0.95555556,0.82222223,0.8888889,0.82222223,0.82222223,0.82222223,0.84444445,1.0,0.95555556,0.9111111,0.9777778,0.9777778,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.13333334,0.22222222,0.53333336,0.46666667,,0.53333336,0.6,0.24444444,0.8,0.6888889,0.5555556,0.5777778,0.75555557,0.8,0.5777778,0.84444445,0.9111111,0.93333334,0.8,0.7777778,0.8,0.82222223,1.0,1.0,0.82222223,1.0,1.0,0.93333334,1.0,0.84444445,1.0,0.84444445,1.0,1.0,0.84444445,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.31111112,0.4,0.5777778,0.64444447,0.4888889,0.46666667,,0.46666667,0.4888889,0.4888889,0.7111111,0.7777778,0.95555556,0.6666667,0.7111111,0.9777778,0.51111114,0.8666667,0.8888889,0.73333335,0.73333335,0.73333335,0.73333335,0.82222223,0.9111111,0.73333335,0.8888889,0.93333334,0.8,0.93333334,0.7777778,0.95555556,0.8,1.0,0.84444445,0.82222223,0.82222223,0.9111111,0.9777778,0.9777778,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,0.8888889,1.0,0.9777778,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.2,0.2,0.44444445,0.37777779,0.4,0.53333336,,0.2,0.46666667,0.5777778,0.42222223,0.53333336,0.64444447,0.6666667,0.6,0.84444445,0.7111111,0.84444445,0.75555557,0.75555557,0.75555557,0.82222223,1.0,0.84444445,0.8666667,0.9777778,0.95555556,0.93333334,1.0,0.93333334,0.9777778,0.9111111,0.9777778,0.93333334,0.9777778,0.9777778,0.95555556,1.0,1.0,1.0,1.0,0.9777778,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.11111111,0.73333335,0.17777778,0.8,0.75555557,0.75555557,0.51111114,0.8,,0.7777778,0.8,0.6,0.5777778,0.8,0.7777778,0.5555556,0.7777778,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,1.0,0.95555556,0.8,0.95555556,0.8,0.8,0.82222223,1.0,1.0,1.0,0.84444445,1.0,0.84444445,0.8,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.08888889,0.2,0.26666668,0.17777778,0.2,0.51111114,0.53333336,0.22222222,,0.62222224,0.44444445,0.53333336,0.51111114,0.5555556,0.6,0.8,0.73333335,0.8888889,0.7777778,0.8,0.8,0.75555557,0.7777778,0.93333334,0.8,0.9777778,0.9777778,0.9111111,1.0,0.8666667,1.0,0.8,1.0,0.9777778,0.8666667,0.8666667,0.95555556,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.22222222,0.2,0.022222223,0.35555556,0.31111112,0.2888889,0.42222223,0.2,0.37777779,,0.26666668,0.37777779,0.4888889,0.4,0.5777778,0.7777778,0.8,0.84444445,0.93333334,0.93333334,0.93333334,0.93333334,0.7111111,0.8888889,0.9777778,0.9777778,0.95555556,1.0,0.9111111,1.0,0.9777778,1.0,0.9777778,0.8,1.0,1.0,0.84444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.044444446,0.4,0.15555556,0.64444447,0.5555556,0.44444445,0.22222222,0.5777778,0.4,0.5555556,0.73333335,,0.31111112,0.5777778,0.75555557,0.5555556,0.6,0.8888889,0.73333335,0.7777778,0.8,0.8,0.7777778,0.7777778,0.9777778,0.8,0.9111111,0.9777778,0.8,0.93333334,0.8,0.9111111,0.8,1.0,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,0.8,0.95555556,1.0,1.0,1.0,1.0,1.0,0.95555556,1.0,0.8,1.0,0.8,1.0,1.0,1.0,0.84444445,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.15555556,0.4,0.46666667,0.5777778,0.42222223,0.42222223,0.044444446,0.46666667,0.42222223,0.46666667,0.62222224,0.6888889,,0.5777778,0.64444447,0.82222223,0.46666667,0.84444445,0.8666667,0.7111111,0.7111111,0.7111111,0.73333335,0.75555557,0.9111111,0.7111111,0.8888889,0.9111111,0.7777778,0.9111111,0.75555557,0.93333334,0.8,1.0,0.8,0.7777778,0.82222223,0.8888889,0.95555556,0.9111111,1.0,1.0,0.9777778,0.8888889,0.9777778,1.0,0.9777778,0.95555556,1.0,1.0,0.93333334,1.0,0.84444445,1.0,0.9777778,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori-30M (default),0.0,0.17777778,0.2,0.33333334,0.33333334,0.24444444,0.33333334,0.35555556,0.2,0.4888889,0.51111114,0.42222223,0.42222223,,0.4888889,0.53333336,0.4888889,0.75555557,0.73333335,0.75555557,0.75555557,0.75555557,0.75555557,0.6888889,0.93333334,0.75555557,0.8666667,0.95555556,0.8888889,0.84444445,0.8888889,0.95555556,0.8,0.9777778,0.62222224,0.8888889,0.84444445,0.73333335,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.95555556,0.9777778,0.9777778,0.9777778,0.95555556,0.9777778,0.9777778,0.93333334,0.9777778,0.95555556,1.0,1.0,1.0,0.9777778,0.9777778,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.0,0.0,0.4888889,0.37777779,0.2,0.2888889,0.33333334,0.22222222,0.44444445,0.6,0.24444444,0.35555556,0.51111114,,0.37777779,0.5777778,0.6,0.6888889,0.75555557,0.8,0.8,0.75555557,0.8,0.95555556,0.8,0.7777778,0.95555556,0.8,0.8,0.8,0.8,0.8,0.8888889,0.84444445,0.8,0.8,1.0,0.8,1.0,1.0,0.9777778,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,0.8,0.82222223,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.13333334,0.4,0.31111112,0.4,0.4,0.42222223,0.022222223,0.4,0.44444445,0.4,0.42222223,0.44444445,0.17777778,0.46666667,0.62222224,,0.42222223,0.75555557,0.5777778,0.6666667,0.7111111,0.7111111,0.73333335,0.51111114,0.7111111,0.7111111,0.84444445,0.84444445,0.73333335,0.75555557,0.73333335,0.8888889,0.8,0.9777778,0.8,0.75555557,0.8,0.84444445,0.73333335,0.9111111,0.93333334,1.0,0.93333334,0.8,0.93333334,1.0,0.93333334,0.93333334,0.93333334,1.0,0.82222223,1.0,0.8,1.0,0.8,1.0,1.0,0.9777778,0.82222223,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.044444446,0.2,0.2,0.044444446,0.15555556,0.4888889,0.15555556,0.22222222,0.2,0.22222222,0.4,0.53333336,0.51111114,0.42222223,0.5777778,,0.42222223,0.5777778,0.5555556,0.5777778,0.5777778,0.6,0.7111111,0.7777778,0.6,0.8,0.7777778,0.73333335,0.9777778,0.73333335,0.82222223,0.62222224,0.8,0.9111111,0.82222223,0.73333335,0.73333335,0.9111111,0.9111111,0.8888889,0.9777778,0.8888889,0.82222223,1.0,1.0,0.84444445,0.95555556,0.8666667,0.9777778,1.0,1.0,0.95555556,1.0,0.9777778,1.0,1.0,0.9111111,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori (default),0.0,0.044444446,0.022222223,0.11111111,0.17777778,0.08888889,0.13333334,0.2888889,0.2,0.26666668,0.2,0.11111111,0.15555556,0.24444444,0.4,0.24444444,0.5777778,,0.6,0.75555557,0.7777778,0.7777778,0.7777778,0.37777779,0.75555557,0.7777778,0.84444445,0.8666667,0.8,0.6888889,0.8,0.84444445,0.8,1.0,0.8,0.8,0.8,0.7777778,0.8,1.0,1.0,1.0,1.0,0.8,0.9111111,1.0,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,0.8,0.9777778,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.022222223,0.17777778,0.06666667,0.11111111,0.06666667,0.11111111,0.15555556,0.2,0.11111111,0.15555556,0.26666668,0.13333334,0.26666668,0.31111112,0.42222223,0.42222223,0.4,,0.6,0.6888889,0.6888889,0.62222224,0.4,0.62222224,0.7111111,1.0,0.75555557,0.7111111,0.7777778,0.7111111,0.8888889,0.6888889,0.93333334,0.73333335,0.8,0.73333335,0.7777778,0.8888889,0.9777778,1.0,1.0,0.9777778,0.7777778,1.0,1.0,0.95555556,1.0,0.9777778,1.0,0.8,1.0,0.8,1.0,0.84444445,1.0,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.17777778,0.2,0.0,0.17777778,0.2,0.26666668,0.24444444,0.2,0.22222222,0.06666667,0.22222222,0.2888889,0.24444444,0.24444444,0.33333334,0.44444445,0.24444444,0.4,,0.46666667,0.6,0.73333335,0.53333336,0.35555556,0.82222223,0.4888889,0.51111114,0.9777778,0.5777778,0.95555556,0.75555557,0.9777778,0.7777778,0.6,1.0,1.0,0.6888889,1.0,0.95555556,0.8,0.8,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.2,0.2,0.0,0.17777778,0.22222222,0.26666668,0.24444444,0.2,0.2,0.06666667,0.2,0.2888889,0.24444444,0.2,0.2888889,0.42222223,0.22222222,0.31111112,0.53333336,,0.6666667,0.64444447,0.53333336,0.26666668,0.82222223,0.4888889,0.37777779,0.84444445,0.5555556,0.9111111,0.64444447,0.9111111,0.75555557,0.6,0.9777778,0.95555556,0.6888889,0.9777778,0.93333334,0.7777778,0.8,0.9777778,1.0,1.0,0.8,0.93333334,1.0,0.95555556,1.0,0.8,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.2,0.2,0.0,0.17777778,0.2,0.26666668,0.24444444,0.2,0.2,0.06666667,0.2,0.2888889,0.24444444,0.2,0.2888889,0.42222223,0.22222222,0.31111112,0.4,0.33333334,,0.5777778,0.53333336,0.26666668,0.7111111,0.4888889,0.37777779,0.73333335,0.5555556,0.75555557,0.64444447,0.9111111,0.6888889,0.6,0.9777778,0.95555556,0.6888889,0.9777778,0.93333334,0.8,0.8,0.9777778,0.9777778,1.0,0.8,0.93333334,1.0,0.95555556,0.9777778,0.8,1.0,1.0,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.2,0.2,0.0,0.15555556,0.17777778,0.26666668,0.17777778,0.2,0.24444444,0.06666667,0.22222222,0.26666668,0.24444444,0.24444444,0.26666668,0.4,0.22222222,0.37777779,0.26666668,0.35555556,0.42222223,,0.46666667,0.24444444,0.5555556,0.42222223,0.33333334,0.93333334,0.4888889,1.0,0.62222224,0.93333334,0.8,0.6,1.0,0.9777778,0.6,0.9777778,0.9111111,0.7777778,0.8,1.0,0.9777778,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.2,0.044444446,0.0,0.0,0.17777778,0.0,0.2,0.22222222,0.2888889,0.22222222,0.24444444,0.31111112,0.2,0.4888889,0.2888889,0.62222224,0.6,0.46666667,0.46666667,0.46666667,0.53333336,,0.62222224,0.46666667,0.6666667,0.7111111,0.6,0.8666667,0.6,0.75555557,0.6,0.8,0.75555557,0.6888889,0.6666667,0.82222223,0.82222223,0.8888889,0.84444445,0.93333334,0.82222223,0.6666667,0.93333334,1.0,0.8,0.9111111,0.8,0.9111111,0.8888889,0.9777778,0.82222223,1.0,0.93333334,1.0,0.9777778,0.8888889,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.022222223,0.0,0.0,0.044444446,0.0,0.08888889,0.15555556,0.0,0.06666667,0.11111111,0.022222223,0.08888889,0.06666667,0.044444446,0.2888889,0.22222222,0.24444444,0.37777779,0.64444447,0.73333335,0.73333335,0.75555557,0.37777779,,0.73333335,0.44444445,0.95555556,0.7777778,0.51111114,0.7777778,0.6888889,0.7777778,0.8,0.6,0.8,0.8,0.6666667,0.8,1.0,0.84444445,0.7777778,0.8,0.8,0.8,0.8,1.0,1.0,1.0,0.8,0.84444445,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.15555556,0.2,0.0,0.08888889,0.17777778,0.26666668,0.13333334,0.2,0.2,0.022222223,0.2,0.2888889,0.24444444,0.2,0.2888889,0.4,0.22222222,0.2888889,0.17777778,0.17777778,0.2888889,0.44444445,0.53333336,0.26666668,,0.4888889,0.4,0.6666667,0.5555556,0.7111111,0.64444447,0.62222224,0.6888889,0.6,0.8,0.75555557,0.6888889,0.8888889,0.73333335,0.7777778,0.8,0.75555557,0.7777778,0.93333334,0.8,0.7111111,0.7777778,0.7111111,0.8,0.62222224,0.8,1.0,0.8,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.022222223,0.0,0.11111111,0.022222223,0.2,0.022222223,0.022222223,0.08888889,0.11111111,0.13333334,0.22222222,0.15555556,0.2,0.15555556,0.0,0.51111114,0.51111114,0.51111114,0.5777778,0.33333334,0.5555556,0.51111114,,0.5777778,0.6,0.6,0.6,0.7111111,0.62222224,0.8,0.6,0.7111111,0.64444447,0.7111111,0.73333335,0.9111111,0.9111111,0.95555556,0.8666667,0.64444447,0.8666667,0.9777778,0.8,0.95555556,0.8666667,0.95555556,0.8,1.0,0.8,1.0,0.8,0.9777778,1.0,0.95555556,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.022222223,0.0,0.0,0.022222223,0.0,0.06666667,0.044444446,0.0,0.022222223,0.044444446,0.022222223,0.08888889,0.044444446,0.044444446,0.15555556,0.22222222,0.13333334,0.24444444,0.4888889,0.62222224,0.62222224,0.6666667,0.2888889,0.044444446,0.6,0.42222223,,0.73333335,0.4,0.73333335,0.46666667,0.73333335,0.7777778,0.6,0.75555557,0.73333335,0.62222224,0.8,0.9777778,0.84444445,0.6,0.75555557,0.75555557,0.8,0.7111111,0.95555556,1.0,1.0,0.8,0.82222223,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.022222223,0.2,0.0,0.0,0.06666667,0.2,0.06666667,0.2,0.08888889,0.0,0.2,0.22222222,0.11111111,0.2,0.26666668,0.26666668,0.2,0.2888889,0.022222223,0.15555556,0.26666668,0.06666667,0.4,0.22222222,0.33333334,0.4,0.26666668,,0.44444445,0.53333336,0.44444445,0.5555556,0.73333335,0.6,0.7777778,0.5777778,0.5777778,0.8666667,0.6888889,0.75555557,0.8,0.9777778,0.84444445,0.95555556,0.8,0.8666667,0.9111111,0.9111111,1.0,0.8,1.0,0.9777778,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.044444446,0.0,0.0,0.06666667,0.0,0.2,0.0,0.08888889,0.06666667,0.08888889,0.15555556,0.2,0.24444444,0.022222223,0.31111112,0.22222222,0.42222223,0.44444445,0.44444445,0.51111114,0.13333334,0.4888889,0.44444445,0.4,0.6,0.5555556,,0.5777778,0.53333336,0.6,0.8,0.6,0.6,0.6,0.62222224,0.62222224,0.75555557,0.8,0.8888889,0.7777778,0.62222224,0.8,0.9777778,0.8,0.84444445,0.8,0.8888889,0.8666667,0.95555556,0.8,1.0,0.8,1.0,0.95555556,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.044444446,0.2,0.0,0.0,0.15555556,0.22222222,0.06666667,0.2,0.13333334,0.0,0.2,0.24444444,0.11111111,0.2,0.26666668,0.26666668,0.2,0.2888889,0.044444446,0.08888889,0.24444444,0.0,0.4,0.22222222,0.2888889,0.4,0.26666668,0.46666667,0.42222223,,0.37777779,0.53333336,0.64444447,0.6,0.8,0.5555556,0.5555556,0.75555557,0.64444447,0.6888889,0.7777778,0.8666667,0.82222223,0.93333334,0.8,0.8666667,0.8888889,0.9111111,1.0,0.8,1.0,0.9777778,0.9111111,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.06666667,0.0,0.0,0.0,0.044444446,0.022222223,0.2,0.0,0.022222223,0.08888889,0.06666667,0.044444446,0.2,0.11111111,0.17777778,0.15555556,0.11111111,0.24444444,0.35555556,0.35555556,0.37777779,0.24444444,0.31111112,0.35555556,0.2888889,0.53333336,0.5555556,0.46666667,0.62222224,,0.51111114,0.84444445,0.5555556,0.6666667,0.5555556,0.5777778,0.64444447,0.7111111,0.93333334,0.9111111,0.8888889,0.5777778,0.8666667,0.95555556,0.8888889,0.9111111,0.8888889,0.9777778,0.7777778,0.9777778,0.7777778,0.9777778,0.7777778,0.9777778,0.9777778,0.9111111,0.7777778,0.95555556,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778,0.9777778
+ChimeraBoost (tuned + ensembled),0.0,0.15555556,0.2,0.0,0.044444446,0.15555556,0.2,0.08888889,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.37777779,0.2,0.31111112,0.022222223,0.08888889,0.08888889,0.06666667,0.4,0.22222222,0.37777779,0.37777779,0.26666668,0.44444445,0.4,0.46666667,0.4888889,,0.51111114,0.4,0.7777778,0.82222223,0.6,0.7777778,0.5777778,0.5777778,0.6,0.7777778,0.95555556,0.8,0.6,0.73333335,0.73333335,0.75555557,0.8,0.6,0.8,0.8,0.64444447,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.9111111,0.8,0.8,0.8,1.0,0.84444445,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.022222223,0.0,0.0,0.022222223,0.11111111,0.022222223,0.2,0.0,0.06666667,0.22222222,0.24444444,0.31111112,0.2,0.2,0.2,0.31111112,0.2,0.22222222,0.26666668,0.2,0.35555556,0.15555556,0.4888889,,0.6,0.4888889,0.4888889,0.4888889,0.44444445,0.6,0.84444445,0.7111111,0.75555557,0.5777778,0.62222224,0.82222223,0.7777778,0.8666667,0.8,1.0,0.8,1.0,0.8,1.0,0.8,0.82222223,1.0,0.9111111,0.8,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.0,0.0,0.2,0.0,0.0,0.15555556,0.06666667,0.0,0.022222223,0.2,0.2,0.2,0.37777779,0.15555556,0.2,0.08888889,0.2,0.26666668,0.4,0.4,0.4,0.4,0.24444444,0.4,0.4,0.4,0.4,0.4,0.4,0.4,0.44444445,0.6,0.4,,0.46666667,0.6,0.5777778,0.42222223,0.44444445,0.44444445,0.53333336,0.42222223,0.6,0.46666667,0.64444447,0.4,0.6888889,0.4,0.6,0.9777778,0.64444447,0.8,0.8,0.8,0.8,0.8,0.6666667,0.8,0.7111111,0.9777778,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.0,0.044444446,0.2,0.0,0.0,0.15555556,0.17777778,0.022222223,0.2,0.13333334,0.0,0.2,0.22222222,0.11111111,0.2,0.24444444,0.17777778,0.2,0.2,0.0,0.022222223,0.022222223,0.0,0.31111112,0.2,0.2,0.2888889,0.24444444,0.22222222,0.4,0.2,0.33333334,0.22222222,0.51111114,0.53333336,,0.33333334,0.4888889,0.4888889,0.35555556,0.53333336,0.7111111,0.46666667,0.53333336,0.7777778,0.75555557,0.44444445,0.6,0.46666667,0.7777778,0.62222224,0.8,0.9777778,0.8,1.0,1.0,1.0,0.7777778,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.044444446,0.2,0.0,0.022222223,0.11111111,0.17777778,0.022222223,0.2,0.13333334,0.0,0.2,0.17777778,0.15555556,0.2,0.2,0.26666668,0.2,0.26666668,0.0,0.044444446,0.044444446,0.022222223,0.33333334,0.2,0.24444444,0.35555556,0.26666668,0.42222223,0.4,0.44444445,0.44444445,0.17777778,0.51111114,0.4,0.6666667,,0.6,0.73333335,0.53333336,0.5555556,0.6,0.73333335,0.84444445,0.7777778,0.6,0.6888889,0.73333335,0.73333335,0.8,0.6,0.8,0.75555557,0.62222224,0.8,0.8,0.8,0.7777778,0.8,0.8,0.8,0.8,0.8888889,0.8,0.8,0.8,1.0,0.84444445,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.0,0.0,0.0,0.0,0.08888889,0.044444446,0.0,0.044444446,0.15555556,0.0,0.11111111,0.26666668,0.0,0.15555556,0.26666668,0.22222222,0.22222222,0.31111112,0.31111112,0.31111112,0.4,0.17777778,0.33333334,0.31111112,0.2888889,0.37777779,0.42222223,0.37777779,0.44444445,0.42222223,0.4,0.51111114,0.42222223,0.51111114,0.4,,0.46666667,0.37777779,0.6,0.62222224,0.53333336,0.42222223,0.53333336,0.62222224,0.6,0.8,0.6,0.6,0.73333335,0.62222224,0.8,0.7777778,0.8,0.8,0.6888889,0.8,0.8,0.8,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.2,0.0,0.0,0.0,0.022222223,0.0,0.2,0.0,0.0,0.2,0.044444446,0.022222223,0.2,0.26666668,0.08888889,0.2,0.11111111,0.0,0.022222223,0.022222223,0.022222223,0.17777778,0.2,0.11111111,0.26666668,0.2,0.13333334,0.37777779,0.24444444,0.35555556,0.22222222,0.5555556,0.5777778,0.51111114,0.26666668,0.53333336,,0.37777779,0.62222224,0.75555557,0.5555556,0.37777779,0.9777778,0.8,0.6,0.6666667,0.6,0.75555557,0.64444447,0.8,0.8,0.8,0.84444445,1.0,0.95555556,0.93333334,0.8888889,1.0,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.08888889,0.022222223,0.0,0.08888889,0.08888889,0.0,0.022222223,0.044444446,0.06666667,0.06666667,0.08888889,0.11111111,0.0,0.26666668,0.08888889,0.022222223,0.31111112,0.24444444,0.35555556,0.2888889,0.42222223,0.4,0.5555556,0.64444447,0.46666667,0.62222224,0.62222224,,0.42222223,0.53333336,0.6,0.53333336,0.6888889,0.6,0.62222224,0.82222223,0.62222224,0.73333335,0.8,0.75555557,0.8,0.62222224,0.8,0.8,0.95555556,0.8666667,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.0,0.0,0.0,0.022222223,0.0,0.06666667,0.11111111,0.0,0.0,0.2,0.22222222,0.2,0.22222222,0.15555556,0.15555556,0.22222222,0.08888889,0.15555556,0.24444444,0.2,0.31111112,0.06666667,0.42222223,0.15555556,0.5555556,0.46666667,0.44444445,0.4,0.37777779,0.5777778,,0.33333334,0.42222223,0.44444445,0.53333336,0.46666667,0.7777778,0.6888889,0.82222223,0.73333335,0.8,0.75555557,0.62222224,0.95555556,0.64444447,0.7777778,0.95555556,0.8888889,0.73333335,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.022222223,0.022222223,0.0,0.022222223,0.0,0.0,0.2,0.2,0.2,0.2,0.06666667,0.22222222,0.2,0.044444446,0.4,0.2,0.11111111,0.22222222,0.08888889,0.4,0.2888889,0.46666667,0.2888889,0.4,0.37777779,0.24444444,0.46666667,0.6666667,,0.4,0.4,0.33333334,0.9777778,0.6666667,0.5777778,0.6666667,0.7777778,0.8,0.84444445,0.7111111,1.0,0.7777778,0.7777778,0.84444445,0.73333335,0.8,0.93333334,0.9111111,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.044444446,0.0,0.0,0.0,0.022222223,0.022222223,0.0,0.06666667,0.11111111,0.0,0.022222223,0.0,0.022222223,0.022222223,0.0,0.17777778,0.2,0.24444444,0.13333334,0.24444444,0.022222223,0.22222222,0.13333334,0.11111111,0.22222222,0.24444444,0.5777778,0.53333336,0.26666668,0.46666667,0.44444445,0.4,0.5777778,0.6,,0.51111114,0.53333336,0.6,0.6,0.7111111,0.62222224,0.6666667,0.75555557,0.64444447,0.8,0.8,0.8,0.8,1.0,0.93333334,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.2,0.0,0.0,0.0,0.06666667,0.022222223,0.2,0.044444446,0.0,0.2,0.11111111,0.044444446,0.2,0.2,0.17777778,0.2,0.22222222,0.0,0.0,0.022222223,0.022222223,0.33333334,0.2,0.22222222,0.35555556,0.24444444,0.15555556,0.37777779,0.17777778,0.42222223,0.044444446,0.42222223,0.4,0.46666667,0.15555556,0.5777778,0.62222224,0.46666667,0.5555556,0.6,0.4888889,,0.75555557,0.6,0.44444445,0.5777778,0.46666667,0.5555556,0.42222223,0.6,0.6888889,0.62222224,0.8,0.8,0.8,0.7777778,0.8,0.8,0.8,0.8,0.84444445,0.8,0.8,0.8,0.93333334,0.8,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.044444446,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.044444446,0.022222223,0.022222223,0.2,0.06666667,0.0,0.08888889,0.0,0.0,0.0,0.0,0.0,0.06666667,0.2,0.06666667,0.13333334,0.2,0.044444446,0.2,0.06666667,0.13333334,0.2,0.37777779,0.53333336,0.22222222,0.22222222,0.46666667,0.022222223,0.31111112,0.46666667,0.6666667,0.46666667,0.24444444,,0.7111111,0.53333336,0.51111114,0.5555556,0.62222224,0.62222224,0.75555557,0.8,0.8,0.8,0.9777778,0.8888889,0.75555557,0.8,1.0,0.9777778,0.9777778,0.95555556,1.0,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.17777778,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.022222223,0.2888889,0.2,0.022222223,0.2,0.044444446,0.4,0.17777778,0.35555556,0.24444444,0.4,0.37777779,0.2,0.4,0.53333336,0.022222223,0.4,0.4,0.2888889,,0.62222224,0.46666667,0.62222224,0.62222224,0.8,0.64444447,0.64444447,1.0,0.6666667,0.7111111,0.7111111,0.64444447,0.75555557,0.8888889,0.8888889,0.95555556,1.0,1.0,0.95555556,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.022222223,0.0,0.06666667,0.15555556,0.0,0.044444446,0.0,0.06666667,0.06666667,0.0,0.2,0.0,0.2888889,0.2,0.044444446,0.13333334,0.2,0.13333334,0.11111111,0.26666668,0.22222222,0.6,0.5555556,0.31111112,0.4,0.4,0.37777779,0.22222222,0.33333334,0.4,0.5555556,0.46666667,0.37777779,,0.51111114,0.9111111,0.6,0.8,0.4888889,0.6,0.42222223,0.6,0.6,0.8,1.0,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.044444446,0.044444446,0.0,0.06666667,0.044444446,0.0,0.0,0.0,0.0,0.0,0.0,0.08888889,0.0,0.22222222,0.044444446,0.0,0.08888889,0.15555556,0.11111111,0.08888889,0.26666668,0.13333334,0.31111112,0.4,0.26666668,0.2,0.33333334,0.17777778,0.31111112,0.42222223,0.2888889,0.42222223,0.4888889,0.53333336,0.4888889,,0.53333336,0.64444447,0.64444447,0.6888889,0.8,0.6,0.8,0.8,0.75555557,0.84444445,0.8,0.8,1.0,1.0,0.84444445,0.8,1.0,0.8,1.0,0.93333334,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.06666667,0.13333334,0.0,0.022222223,0.0,0.044444446,0.044444446,0.0,0.2,0.0,0.2888889,0.13333334,0.0,0.08888889,0.2,0.08888889,0.11111111,0.24444444,0.2,0.6,0.53333336,0.26666668,0.4,0.4,0.37777779,0.17777778,0.33333334,0.37777779,0.53333336,0.44444445,0.37777779,0.08888889,0.46666667,,0.46666667,0.8,0.44444445,0.6,0.42222223,0.6,0.6,0.8,1.0,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15555556,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.022222223,0.0,0.08888889,0.2,0.2,0.044444446,0.2,0.0,0.11111111,0.0,0.022222223,0.2,0.0,0.4,0.22222222,0.2,0.4,0.24444444,0.26666668,0.26666668,0.22222222,0.33333334,0.44444445,0.37777779,0.37777779,0.4,0.35555556,0.53333336,,0.7777778,0.6,0.6,0.73333335,0.6666667,0.73333335,0.9777778,0.64444447,0.6666667,0.8888889,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.08888889,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.044444446,0.06666667,0.06666667,0.0,0.17777778,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.11111111,0.15555556,0.37777779,0.2,0.17777778,0.2,0.13333334,0.2,0.22222222,0.4,0.2,0.022222223,0.37777779,0.4,0.26666668,0.35555556,0.2,0.2,0.2,0.24444444,0.5777778,0.37777779,0.2,0.2,0.35555556,0.2,0.22222222,,0.22222222,0.75555557,0.51111114,0.8,0.6,0.42222223,0.6,0.8,0.5777778,0.82222223,0.8666667,0.8,0.8,0.9111111,0.8,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15555556,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.2,0.0,0.2,0.0,0.044444446,0.0,0.022222223,0.2,0.0,0.35555556,0.2,0.2,0.37777779,0.2,0.24444444,0.24444444,0.15555556,0.35555556,0.4,0.24444444,0.35555556,0.51111114,0.31111112,0.5555556,0.4,0.7777778,,0.6,0.7111111,0.6666667,0.7111111,0.7111111,0.62222224,0.6666667,0.8888889,0.8666667,0.93333334,1.0,1.0,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.2,0.0,0.0,0.0,0.11111111,0.0,0.2,0.044444446,0.0,0.2,0.15555556,0.044444446,0.2,0.2,0.044444446,0.2,0.2,0.0,0.0,0.0,0.0,0.17777778,0.2,0.0,0.2,0.2,0.022222223,0.2,0.022222223,0.22222222,0.2,0.2,0.2,0.022222223,0.24444444,0.2,0.2,0.2,0.37777779,0.2888889,0.2,0.31111112,0.2,0.35555556,0.4,0.2,0.4,0.4,0.24444444,0.4,,0.6666667,1.0,0.46666667,0.4,0.4,1.0,0.4888889,0.6,0.6,0.6888889,0.8888889,0.75555557,0.9777778,0.7777778,1.0,0.8,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.08888889,0.08888889,0.08888889,0.06666667,0.0,0.0,0.2,0.0,0.0,0.08888889,0.0,0.08888889,0.022222223,0.35555556,0.0,0.2,0.2,0.37777779,0.22222222,0.2,0.37777779,0.044444446,0.0,0.2,0.37777779,0.2,0.0,0.5777778,0.4,0.5777778,0.26666668,0.4888889,0.2888889,0.33333334,,0.46666667,0.4,0.4888889,0.6,0.46666667,0.5777778,0.75555557,0.7777778,0.9111111,0.82222223,0.8,0.84444445,1.0,0.95555556,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.2,0.0,0.0,0.0,0.022222223,0.0,0.2,0.0,0.0,0.2,0.022222223,0.0,0.2,0.2,0.022222223,0.2,0.15555556,0.0,0.0,0.0,0.0,0.06666667,0.2,0.0,0.2,0.2,0.0,0.2,0.0,0.22222222,0.2,0.2,0.2,0.0,0.2,0.2,0.15555556,0.2,0.35555556,0.22222222,0.2,0.2,0.2,0.33333334,0.4,0.2,0.4,0.33333334,0.2,0.33333334,0.0,0.53333336,,0.4,0.33333334,0.4,0.93333334,0.4,0.6,0.6,0.6,0.8666667,0.75555557,0.95555556,0.75555557,1.0,0.8,1.0,1.0,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.17777778,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.022222223,0.2,0.0,0.0,0.0,0.022222223,0.2,0.17777778,0.2,0.0,0.2,0.2,0.0,0.2,0.22222222,0.22222222,0.2,0.2,0.022222223,0.2888889,0.4,0.2,0.4,0.26666668,0.4,0.2888889,0.53333336,0.6,0.6,,0.2888889,0.4,0.62222224,0.73333335,0.5777778,0.7111111,0.6888889,1.0,0.93333334,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.022222223,0.2,0.0,0.2,0.0,0.2,0.31111112,0.044444446,0.044444446,0.044444446,0.15555556,0.0,0.2,0.11111111,0.2888889,0.2,0.24444444,0.2,0.022222223,0.5777778,0.2888889,0.6,0.51111114,0.6666667,0.7111111,,0.5555556,0.6666667,0.6888889,0.8666667,0.93333334,0.93333334,0.9777778,0.9777778,0.9777778,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.022222223,0.08888889,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.022222223,0.044444446,0.0,0.0,0.2,0.0,0.08888889,0.2,0.08888889,0.33333334,0.22222222,0.22222222,0.2,0.06666667,0.13333334,0.11111111,0.26666668,0.06666667,0.22222222,0.24444444,0.35555556,0.0,0.15555556,0.0,0.35555556,0.4,0.37777779,0.6,0.4,0.6,0.6,0.44444445,,0.6,0.7777778,0.6,0.64444447,0.73333335,0.73333335,0.93333334,0.73333335,0.8,0.8,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.15555556,0.0,0.0,0.0,0.022222223,0.0,0.2,0.0,0.0,0.15555556,0.022222223,0.0,0.2,0.17777778,0.022222223,0.2,0.08888889,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.2,0.2,0.0,0.2,0.0,0.22222222,0.2,0.2,0.2,0.0,0.2,0.2,0.11111111,0.2,0.26666668,0.2,0.2,0.2,0.2,0.24444444,0.4,0.2,0.4,0.33333334,0.2,0.33333334,0.0,0.53333336,0.06666667,0.37777779,0.33333334,0.4,,0.4,0.5555556,0.6,0.6,0.84444445,0.6666667,0.8888889,0.64444447,1.0,0.8,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.044444446,0.2,0.022222223,0.2888889,0.022222223,0.2,0.2,0.0,0.2,0.022222223,0.06666667,0.0,0.2,0.0,0.11111111,0.2,0.2,0.2,0.11111111,0.42222223,0.11111111,0.51111114,0.42222223,0.6,0.26666668,0.31111112,0.22222222,0.6,,0.53333336,0.5777778,0.64444447,0.84444445,0.82222223,0.84444445,0.8,1.0,0.8666667,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.022222223,0.0,0.2,0.0,0.0,0.0,0.0,0.08888889,0.0,0.2,0.022222223,0.11111111,0.2,0.0,0.2,0.06666667,0.17777778,0.13333334,0.4,0.24444444,0.4,0.42222223,0.13333334,0.4,0.44444445,0.46666667,,0.8,0.4,0.8,1.0,0.8,0.5777778,0.8,1.0,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.022222223,0.0,0.2,0.0,0.0,0.0,0.0,0.022222223,0.0,0.2,0.022222223,0.044444446,0.2,0.0,0.2,0.06666667,0.13333334,0.06666667,0.4,0.22222222,0.4,0.2888889,0.06666667,0.35555556,0.4,0.42222223,0.2,,0.4,0.8,0.9111111,0.8,0.5777778,0.8,1.0,1.0,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.08888889,0.0,0.0,0.0,0.11111111,0.022222223,0.044444446,0.0,0.0,0.0,0.0,0.15555556,0.044444446,0.0,0.0,0.15555556,0.0,0.0,0.2,0.0,0.31111112,0.08888889,0.4,0.31111112,0.06666667,0.26666668,0.4,0.35555556,0.6,0.6,,0.4888889,0.6,0.4888889,1.0,0.5777778,1.0,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.11111111,0.17777778,0.13333334,0.0,0.022222223,0.26666668,0.15555556,0.15555556,0.2,0.2,0.51111114,,0.24444444,0.75555557,0.6,1.0,0.6,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.044444446,0.0,0.0,0.0,0.0,0.08888889,0.022222223,0.24444444,0.2,0.24444444,0.06666667,0.022222223,0.06666667,0.33333334,0.17777778,0.0,0.08888889,0.4,0.75555557,,0.75555557,0.4,0.8,0.6,1.0,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.022222223,0.15555556,0.044444446,0.0,0.022222223,0.26666668,0.11111111,0.15555556,0.2,0.2,0.51111114,0.24444444,0.24444444,,0.6,1.0,0.6,1.0,1.0,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.0,0.24444444,0.22222222,0.0,0.2,0.35555556,0.2,0.42222223,0.42222223,0.0,0.4,0.6,0.4,,0.4,1.0,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.15555556,0.0,0.0,0.0,0.15555556,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.044444446,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.2,0.42222223,0.0,0.2,0.0,0.6,,0.6,1.0,1.0,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.2,0.13333334,0.0,0.0,0.0,0.4,0.4,0.4,0.0,0.4,,0.8,0.8,0.8888889
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,,0.87777776,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.12222222,,0.9111111
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.022222223,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.08888889,
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_small/leaderboard_overview_explorer.html
index 8fe6f3f40f2846f5a05f63679260aafc30852566..3a8bd9e65736944662b2e993c4d08aadbb68a3d0 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Regression | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_all/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_all/tasks_regression/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..2e96dbb7116071c0d9fa9a306f7ebca7572479d7
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_all/tasks_regression/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1949.0,"score":0.922,"rank":5.05,"hrank":2.14,"imp":2.153,"train_time":49.29,"predict_time":4.094,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+173\/-99","imputed":false},{"position":1,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1850.0,"score":0.813,"rank":7.51,"hrank":2.47,"imp":1.895,"train_time":12.09,"predict_time":2.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+255\/-123","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1785.0,"score":0.732,"rank":9.66,"hrank":4.27,"imp":2.603,"train_time":14.09,"predict_time":0.439,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+192\/-107","imputed":false},{"position":3,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1770.0,"score":0.725,"rank":10.18,"hrank":5.66,"imp":3.44,"train_time":12.37,"predict_time":1.566,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+192\/-103","imputed":false},{"position":4,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1765.0,"score":0.69,"rank":10.39,"hrank":5.06,"imp":3.71,"train_time":322.87,"predict_time":5.571,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+192\/-94","imputed":false},{"position":5,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1727.0,"score":0.673,"rank":11.95,"hrank":7.87,"imp":3.86,"train_time":9.2,"predict_time":0.585,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+155\/-68","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1702.0,"score":0.619,"rank":13.05,"hrank":6.2,"imp":4.639,"train_time":4823.37,"predict_time":140.737,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+159\/-70","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1694.0,"score":0.609,"rank":13.38,"hrank":6.23,"imp":3.257,"train_time":2307.57,"predict_time":7.226,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+241\/-122","imputed":false},{"position":8,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1657.0,"score":0.566,"rank":15.19,"hrank":9.49,"imp":5.103,"train_time":4823.37,"predict_time":22.312,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+126\/-47","imputed":false},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1656.0,"score":0.645,"rank":15.25,"hrank":5.51,"imp":4.348,"train_time":5.15,"predict_time":0.14,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+238\/-195","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1625.0,"score":0.501,"rank":16.84,"hrank":10.53,"imp":5.049,"train_time":4560.01,"predict_time":33.831,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-62","imputed":false},{"position":11,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1568.0,"score":0.45,"rank":20.09,"hrank":10.13,"imp":6.205,"train_time":48.14,"predict_time":42.001,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-61","imputed":false},{"position":12,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1565.0,"score":0.464,"rank":20.22,"hrank":8.6,"imp":4.17,"train_time":2307.57,"predict_time":0.902,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+240\/-134","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1556.0,"score":0.433,"rank":20.78,"hrank":9.75,"imp":6.01,"train_time":7.93,"predict_time":0.264,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+144\/-73","imputed":false},{"position":14,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1544.0,"score":0.359,"rank":21.53,"hrank":14.81,"imp":6.948,"train_time":5807.07,"predict_time":4.31,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-72","imputed":false},{"position":15,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1535.0,"score":0.359,"rank":22.06,"hrank":14.03,"imp":5.974,"train_time":4560.01,"predict_time":2.233,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-71","imputed":false},{"position":16,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1498.0,"score":0.389,"rank":24.47,"hrank":15.92,"imp":6.002,"train_time":7.67,"predict_time":0.547,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+184\/-123","imputed":false},{"position":17,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1495.0,"score":0.393,"rank":24.65,"hrank":12.7,"imp":5.614,"train_time":6559.81,"predict_time":56.287,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+244\/-174","imputed":true},{"position":18,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1477.0,"score":0.241,"rank":25.87,"hrank":20.65,"imp":7.754,"train_time":4283.1,"predict_time":0.444,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-68","imputed":false},{"position":19,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1476.0,"score":0.273,"rank":25.96,"hrank":18.69,"imp":9.056,"train_time":39519.07,"predict_time":1519.479,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+122\/-61","imputed":false},{"position":20,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1473.0,"score":0.306,"rank":26.12,"hrank":17.04,"imp":6.073,"train_time":3.53,"predict_time":0.064,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+146\/-89","imputed":false},{"position":21,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1454.0,"score":0.224,"rank":27.42,"hrank":23.25,"imp":8.6,"train_time":1203.24,"predict_time":7.106,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-75","imputed":false},{"position":22,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1450.0,"score":0.207,"rank":27.7,"hrank":22.45,"imp":7.927,"train_time":4283.1,"predict_time":0.063,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-66","imputed":false},{"position":23,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1446.0,"score":0.276,"rank":27.94,"hrank":20.34,"imp":6.597,"train_time":3660.26,"predict_time":4.755,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+166\/-98","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1444.0,"score":0.278,"rank":28.11,"hrank":7.68,"imp":8.978,"train_time":8982.74,"predict_time":8.012,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+183\/-107","imputed":false},{"position":25,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1430.0,"score":0.217,"rank":29.12,"hrank":18.8,"imp":8.069,"train_time":2039.7,"predict_time":2.142,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-104","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1411.0,"score":0.158,"rank":30.45,"hrank":27.08,"imp":8.905,"train_time":1266.57,"predict_time":3.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-42","imputed":false},{"position":27,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1400.0,"score":0.245,"rank":31.19,"hrank":16.41,"imp":8.25,"train_time":93.39,"predict_time":2.396,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+150\/-113","imputed":true},{"position":28,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1384.0,"score":0.134,"rank":32.37,"hrank":28.65,"imp":9.066,"train_time":1266.57,"predict_time":0.475,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-47","imputed":false},{"position":29,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1377.0,"score":0.147,"rank":32.89,"hrank":27.86,"imp":9.354,"train_time":1203.24,"predict_time":0.956,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-79","imputed":false},{"position":30,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1361.0,"score":0.176,"rank":34.09,"hrank":23.14,"imp":7.609,"train_time":3660.26,"predict_time":0.492,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+168\/-107","imputed":false},{"position":31,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1358.0,"score":0.207,"rank":34.27,"hrank":16.67,"imp":10.039,"train_time":22.46,"predict_time":1.038,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-159","imputed":false},{"position":32,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1357.0,"score":0.105,"rank":34.36,"hrank":27.07,"imp":9.918,"train_time":13.63,"predict_time":0.049,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-85","imputed":false},{"position":33,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1357.0,"score":0.163,"rank":34.38,"hrank":28.29,"imp":8.564,"train_time":2039.7,"predict_time":0.109,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-82","imputed":false},{"position":34,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1356.0,"score":0.245,"rank":34.41,"hrank":18.44,"imp":7.313,"train_time":6559.81,"predict_time":0.485,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+267\/-182","imputed":true},{"position":35,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1344.0,"score":0.123,"rank":35.35,"hrank":29.12,"imp":9.259,"train_time":39519.07,"predict_time":167.037,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-70","imputed":false},{"position":36,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1334.0,"score":0.101,"rank":36.08,"hrank":28.45,"imp":10.37,"train_time":1846.43,"predict_time":0.349,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-92","imputed":false},{"position":37,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1314.0,"score":0.13,"rank":37.52,"hrank":21.65,"imp":11.057,"train_time":8982.74,"predict_time":0.421,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+163\/-101","imputed":false},{"position":38,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1299.0,"score":0.075,"rank":38.65,"hrank":30.85,"imp":10.711,"train_time":1846.43,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-88","imputed":false},{"position":39,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1298.0,"score":0.147,"rank":38.77,"hrank":29.61,"imp":8.886,"train_time":13.53,"predict_time":0.467,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+157\/-136","imputed":false},{"position":40,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1280.0,"score":0.196,"rank":40.07,"hrank":27.1,"imp":9.537,"train_time":8.74,"predict_time":0.773,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+222\/-225","imputed":true},{"position":41,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1271.0,"score":0.12,"rank":40.76,"hrank":33.97,"imp":9.688,"train_time":21.52,"predict_time":8.35,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-103","imputed":false},{"position":42,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1261.0,"score":0.13,"rank":41.54,"hrank":30.0,"imp":13.405,"train_time":1163.69,"predict_time":2.098,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-158","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.114,"rank":42.77,"hrank":27.13,"imp":11.33,"train_time":5151.88,"predict_time":1.591,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-162","imputed":false},{"position":44,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1241.0,"score":0.113,"rank":43.0,"hrank":28.59,"imp":13.65,"train_time":1163.69,"predict_time":0.288,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-159","imputed":false},{"position":45,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1238.0,"score":0.11,"rank":43.17,"hrank":16.15,"imp":11.346,"train_time":121.77,"predict_time":21.803,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+209\/-206","imputed":false},{"position":46,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1234.0,"score":0.022,"rank":43.51,"hrank":39.83,"imp":11.663,"train_time":8.29,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-84","imputed":false},{"position":47,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1210.0,"score":0.029,"rank":45.21,"hrank":34.59,"imp":12.788,"train_time":26.38,"predict_time":0.302,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-88","imputed":false},{"position":48,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1195.0,"score":0.055,"rank":46.28,"hrank":35.76,"imp":13.605,"train_time":3084.49,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+154\/-165","imputed":false},{"position":49,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1192.0,"score":0.073,"rank":46.48,"hrank":36.96,"imp":11.987,"train_time":5151.88,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-153","imputed":false},{"position":50,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1191.0,"score":0.003,"rank":46.6,"hrank":44.55,"imp":11.96,"train_time":4.3,"predict_time":0.317,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-44","imputed":false},{"position":51,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1164.0,"score":0.039,"rank":48.46,"hrank":43.9,"imp":15.162,"train_time":675.17,"predict_time":1.54,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-146","imputed":false},{"position":52,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1155.0,"score":0.038,"rank":49.08,"hrank":37.07,"imp":14.114,"train_time":3084.49,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+147\/-165","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1126.0,"score":0.016,"rank":51.05,"hrank":46.5,"imp":15.757,"train_time":675.17,"predict_time":0.206,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-146","imputed":false},{"position":54,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1126.0,"score":0.0,"rank":51.06,"hrank":49.52,"imp":13.163,"train_time":4.06,"predict_time":0.299,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-54","imputed":false},{"position":55,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1119.0,"score":0.0,"rank":51.48,"hrank":48.78,"imp":15.134,"train_time":195.75,"predict_time":0.36,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-122","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1112.0,"score":0.038,"rank":51.94,"hrank":47.55,"imp":14.711,"train_time":2215.61,"predict_time":7.791,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-127","imputed":false},{"position":57,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1099.0,"score":0.044,"rank":52.81,"hrank":43.13,"imp":15.536,"train_time":10.2,"predict_time":2.716,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+142\/-209","imputed":false},{"position":58,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1095.0,"score":0.015,"rank":53.05,"hrank":43.84,"imp":16.472,"train_time":7.47,"predict_time":1.008,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-192","imputed":false},{"position":59,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1084.0,"score":0.006,"rank":53.76,"hrank":45.65,"imp":15.183,"train_time":12.26,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-150","imputed":false},{"position":60,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1080.0,"score":0.0,"rank":53.99,"hrank":51.31,"imp":16.43,"train_time":195.75,"predict_time":0.165,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-126","imputed":false},{"position":61,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1075.0,"score":0.183,"rank":54.27,"hrank":15.75,"imp":23.351,"train_time":52983.71,"predict_time":36.311,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+220\/-421","imputed":false},{"position":62,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1065.0,"score":0.174,"rank":54.91,"hrank":17.71,"imp":23.476,"train_time":52983.71,"predict_time":5.585,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+219\/-430","imputed":false},{"position":63,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1064.0,"score":0.022,"rank":55.0,"hrank":51.46,"imp":15.208,"train_time":2215.61,"predict_time":0.838,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-135","imputed":false},{"position":64,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1037.0,"score":0.0,"rank":56.58,"hrank":53.21,"imp":14.988,"train_time":22.4,"predict_time":0.134,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-164","imputed":false},{"position":65,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":58.65,"hrank":56.02,"imp":17.471,"train_time":0.63,"predict_time":0.12,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-146","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":956.0,"score":0.081,"rank":60.94,"hrank":25.51,"imp":26.213,"train_time":394.68,"predict_time":6.52,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+231\/-467","imputed":false},{"position":67,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":947.0,"score":0.0,"rank":61.38,"hrank":51.65,"imp":21.637,"train_time":88.43,"predict_time":0.512,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+155\/-341","imputed":false},{"position":68,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":945.0,"score":0.0,"rank":61.44,"hrank":59.33,"imp":19.767,"train_time":6.97,"predict_time":0.818,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-184","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":907.0,"score":0.0,"rank":63.21,"hrank":62.25,"imp":19.757,"train_time":32.6,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-148","imputed":false},{"position":70,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":849.0,"score":0.0,"rank":65.58,"hrank":59.49,"imp":23.927,"train_time":88.43,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-329","imputed":false},{"position":71,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":841.0,"score":0.0,"rank":65.88,"hrank":65.34,"imp":20.636,"train_time":725.45,"predict_time":268.761,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-129","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":772.0,"score":0.0,"rank":68.2,"hrank":50.63,"imp":30.804,"train_time":0.42,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+170\/-439","imputed":false},{"position":73,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":606.0,"score":0.0,"rank":72.07,"hrank":71.69,"imp":35.068,"train_time":777.82,"predict_time":0.312,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-382","imputed":false},{"position":74,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":596.0,"score":0.0,"rank":72.24,"hrank":71.84,"imp":35.111,"train_time":777.82,"predict_time":0.108,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-420","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":447.0,"score":0.0,"rank":74.33,"hrank":74.28,"imp":37.052,"train_time":3.91,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-399","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_small/pareto_front_explorer.html
index b973872154cd6959821d05c065c808ce4d42fbaa..2dc2eaa353ea77015554efd7cc9a096cf22b3592 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0489086629,"imp":9.9184195202,"elo":1356.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0627456348,"imp":7.9267793701,"elo":1449.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4436888454,"imp":7.7535686505,"elo":1476.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0368319183,"imp":11.6625047538,"elo":1233.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0290955235,"imp":10.7107933653,"elo":1299.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.348849021,"imp":10.3704508297,"elo":1333.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0130181103,"imp":15.1827370397,"elo":1083.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0122824328,"imp":14.1142111661,"elo":1154.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1263863053,"imp":13.6053299983,"elo":1195.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0076902476,"imp":16.472016891,"elo":1094.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2883404835,"imp":13.650356813,"elo":1240.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":2.0978537644,"imp":13.4048362835,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8180850143,"imp":19.7666131041,"elo":945.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.837804851,"imp":15.2081157794,"elo":1063.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7913188557,"imp":14.7113436581,"elo":1112.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.3174685594,"imp":11.9599837789,"elo":1190.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9564936689,"imp":9.3544059529,"elo":1376.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":7.1056715587,"imp":8.5996213583,"elo":1454.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":21.8030250918,"imp":11.3455432733,"elo":1238.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.3955680904,"imp":8.2502901254,"elo":1400.4,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3020207606,"imp":12.7884091458,"elo":1210.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.420913606,"imp":11.0565110572,"elo":1314.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0116272048,"imp":8.9782540168,"elo":1444.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.5662902939,"imp":3.4396921903,"elo":1770.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0528542696,"imp":1.8953053947,"elo":1850.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.021441561,"imp":19.7570621899,"elo":907.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1652630526,"imp":16.4302871002,"elo":1079.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3602092205,"imp":15.1344810749,"elo":1119.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1203913506,"imp":17.4712212062,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2060693731,"imp":15.7567391664,"elo":1125.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.5403575271,"imp":15.1617368055,"elo":1163.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":8.3504595229,"imp":9.6876147665,"elo":1271.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":2.2331214774,"imp":5.9737128789,"elo":1535.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":33.8307336671,"imp":5.0493303711,"elo":1625.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5470075607,"imp":6.002034103,"elo":1497.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9019759916,"imp":4.170257595,"elo":1565.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.2264004569,"imp":3.2566745101,"elo":1694.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.0380986551,"imp":10.038596328,"elo":1358.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.001452678,"imp":6.2049828888,"elo":1567.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":22.311628541,"imp":5.1031068267,"elo":1656.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":140.7371116271,"imp":4.6385601486,"elo":1701.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2636901296,"imp":6.0096975764,"elo":1556.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":4.0942168713,"imp":2.1533169207,"elo":1948.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1402308582,"imp":4.3481318241,"elo":1655.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4666328992,"imp":8.8862067808,"elo":1297.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4918252514,"imp":7.6088785459,"elo":1360.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.7546477722,"imp":6.5965850834,"elo":1446.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5852250627,"imp":3.8596284179,"elo":1726.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4388252423,"imp":2.6029697481,"elo":1784.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.7725507791,"imp":9.5370194353,"elo":1280.4,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4849193201,"imp":7.3127614207,"elo":1356.3,"imputed":true,"imputed_pct":12.5,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":56.2866799416,"imp":5.6144418246,"elo":1495.0,"imputed":true,"imputed_pct":12.5,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5202853799,"imp":26.2133062439,"elo":955.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5850454428,"imp":23.4764035351,"elo":1064.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.3108208411,"imp":23.3512386699,"elo":1075.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0639725065,"imp":6.0728654896,"elo":1473.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1340751587,"imp":14.9875983383,"elo":1036.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1363485289,"imp":11.9873128865,"elo":1192.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5906165471,"imp":11.3296633158,"elo":1244.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2989233835,"imp":13.1633575736,"elo":1125.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4754931932,"imp":9.0662800314,"elo":1384.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.1465188849,"imp":8.9045761802,"elo":1410.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":268.7612898715,"imp":20.6363088964,"elo":841.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":167.0374934194,"imp":9.2589159814,"elo":1343.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1519.4791766157,"imp":9.0564109871,"elo":1475.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":2.7158548006,"imp":15.5363087159,"elo":1098.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1085198178,"imp":8.5639932607,"elo":1356.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.1419404045,"imp":8.0690333604,"elo":1429.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0489086629,"x_train":13.6316706327,"imp":9.9184195202,"elo":1356.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0627456348,"x_train":4283.1043461266,"imp":7.9267793701,"elo":1449.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4436888454,"x_train":4283.1043461266,"imp":7.7535686505,"elo":1476.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0368319183,"x_train":8.2896427598,"imp":11.6625047538,"elo":1233.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0290955235,"x_train":1846.4337432747,"imp":10.7107933653,"elo":1299.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.348849021,"x_train":1846.4337432747,"imp":10.3704508297,"elo":1333.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0130181103,"x_train":12.2643063605,"imp":15.1827370397,"elo":1083.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0122824328,"x_train":3084.4850399813,"imp":14.1142111661,"elo":1154.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1263863053,"x_train":3084.4850399813,"imp":13.6053299983,"elo":1195.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0076902476,"x_train":7.4740218671,"imp":16.472016891,"elo":1094.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2883404835,"x_train":1163.6859454819,"imp":13.650356813,"elo":1240.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":2.0978537644,"x_train":1163.6859454819,"imp":13.4048362835,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8180850143,"x_train":6.9730562426,"imp":19.7666131041,"elo":945.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.837804851,"x_train":2215.6113217439,"imp":15.2081157794,"elo":1063.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7913188557,"x_train":2215.6113217439,"imp":14.7113436581,"elo":1112.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.3174685594,"x_train":4.301698659,"imp":11.9599837789,"elo":1190.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.9564936689,"x_train":1203.2369028783,"imp":9.3544059529,"elo":1376.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":7.1056715587,"x_train":1203.2369028783,"imp":8.5996213583,"elo":1454.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":21.8030250918,"x_train":121.7676060282,"imp":11.3455432733,"elo":1238.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.3955680904,"x_train":93.3902149907,"imp":8.2502901254,"elo":1400.4,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3020207606,"x_train":26.3794580707,"imp":12.7884091458,"elo":1210.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.420913606,"x_train":8982.7355846534,"imp":11.0565110572,"elo":1314.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0116272048,"x_train":8982.7355846534,"imp":8.9782540168,"elo":1444.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.5662902939,"x_train":12.3703933426,"imp":3.4396921903,"elo":1770.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0528542696,"x_train":12.0914930116,"imp":1.8953053947,"elo":1850.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.021441561,"x_train":32.5994632136,"imp":19.7570621899,"elo":907.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1652630526,"x_train":195.7517175714,"imp":16.4302871002,"elo":1079.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3602092205,"x_train":195.7517175714,"imp":15.1344810749,"elo":1119.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1203913506,"x_train":0.6268352454,"imp":17.4712212062,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2060693731,"x_train":675.1680482075,"imp":15.7567391664,"elo":1125.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.5403575271,"x_train":675.1680482075,"imp":15.1617368055,"elo":1163.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":8.3504595229,"x_train":21.5151248338,"imp":9.6876147665,"elo":1271.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":2.2331214774,"x_train":4560.0060476784,"imp":5.9737128789,"elo":1535.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":33.8307336671,"x_train":4560.0060476784,"imp":5.0493303711,"elo":1625.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5470075607,"x_train":7.6690681838,"imp":6.002034103,"elo":1497.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9019759916,"x_train":2307.574171405,"imp":4.170257595,"elo":1565.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.2264004569,"x_train":2307.574171405,"imp":3.2566745101,"elo":1694.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.0380986551,"x_train":22.4635991568,"imp":10.038596328,"elo":1358.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":42.001452678,"x_train":48.1376226066,"imp":6.2049828888,"elo":1567.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":22.311628541,"x_train":4823.3684232511,"imp":5.1031068267,"elo":1656.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":140.7371116271,"x_train":4823.3684232511,"imp":4.6385601486,"elo":1701.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2636901296,"x_train":7.9333050678,"imp":6.0096975764,"elo":1556.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":4.0942168713,"x_train":49.2939082066,"imp":2.1533169207,"elo":1948.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1402308582,"x_train":5.1543015637,"imp":4.3481318241,"elo":1655.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4666328992,"x_train":13.5301358652,"imp":8.8862067808,"elo":1297.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4918252514,"x_train":3660.2557617825,"imp":7.6088785459,"elo":1360.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.7546477722,"x_train":3660.2557617825,"imp":6.5965850834,"elo":1446.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5852250627,"x_train":9.2031778761,"imp":3.8596284179,"elo":1726.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4388252423,"x_train":14.0862798048,"imp":2.6029697481,"elo":1784.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.7725507791,"x_train":8.7397288053,"imp":9.5370194353,"elo":1280.4,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4849193201,"x_train":6559.811990291,"imp":7.3127614207,"elo":1356.3,"imputed":true,"imputed_pct":12.5,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":56.2866799416,"x_train":6559.811990291,"imp":5.6144418246,"elo":1495.0,"imputed":true,"imputed_pct":12.5,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5202853799,"x_train":394.6813525384,"imp":26.2133062439,"elo":955.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5850454428,"x_train":52983.7100836136,"imp":23.4764035351,"elo":1064.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.3108208411,"x_train":52983.7100836136,"imp":23.3512386699,"elo":1075.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0639725065,"x_train":3.5256005858,"imp":6.0728654896,"elo":1473.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1340751587,"x_train":22.3971569167,"imp":14.9875983383,"elo":1036.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1363485289,"x_train":5151.8777257105,"imp":11.9873128865,"elo":1192.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5906165471,"x_train":5151.8777257105,"imp":11.3296633158,"elo":1244.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2989233835,"x_train":4.0580642834,"imp":13.1633575736,"elo":1125.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4754931932,"x_train":1266.5674390493,"imp":9.0662800314,"elo":1384.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.1465188849,"x_train":1266.5674390493,"imp":8.9045761802,"elo":1410.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":268.7612898715,"x_train":725.4510762602,"imp":20.6363088964,"elo":841.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":167.0374934194,"x_train":39519.0698998597,"imp":9.2589159814,"elo":1343.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1519.4791766157,"x_train":39519.0698998597,"imp":9.0564109871,"elo":1475.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":2.7158548006,"x_train":10.1998611956,"imp":15.5363087159,"elo":1098.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1085198178,"x_train":2039.6992814536,"imp":8.5639932607,"elo":1356.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.1419404045,"x_train":2039.6992814536,"imp":8.0690333604,"elo":1429.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_small/tuning_trajectories_explorer.html
index 1474ec8b779620415b71d37642f16d7f133993cc..303f9bbfb3e69204984799c06992f50dc113f6b8 100644
--- a/data/imputation_yes/splits_all/tasks_regression/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_all/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":13.6316706327,"imp":9.9238592529,"elo":1364.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":36.2728602647,"imp":9.2661240567,"elo":1407.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":97.6908201342,"imp":8.5526519375,"elo":1454.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":201.9357530623,"imp":8.2116280916,"elo":1472.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":526.2527558948,"imp":7.8995961814,"elo":1485.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":1057.486905704,"imp":7.8367356068,"elo":1490.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":2119.3859898824,"imp":7.7928148055,"elo":1495.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":3172.1186602631,"imp":7.7813830392,"elo":1495.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":4283.1043461266,"imp":7.7596298454,"elo":1499.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":8.2896427598,"imp":11.667714954,"elo":1239.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":17.6287133887,"imp":11.4543747457,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":44.8911284914,"imp":11.2321694494,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":90.9274231145,"imp":11.0389655949,"elo":1296.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":228.5740056066,"imp":10.8238574899,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":458.5111602981,"imp":10.6584554849,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":918.1582493662,"imp":10.4931075398,"elo":1335.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1376.6746665085,"imp":10.4238490422,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1846.4337432747,"imp":10.3751093817,"elo":1346.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":12.2643063605,"imp":15.1876942233,"elo":1086.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":29.8736292865,"imp":14.8048532351,"elo":1115.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":80.8890311763,"imp":14.4156511347,"elo":1152.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":157.9967250155,"imp":14.1583098609,"elo":1171.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":390.307349424,"imp":13.8341552659,"elo":1187.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":781.3223148424,"imp":13.7461222732,"elo":1193.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1537.3695296789,"imp":13.674310407,"elo":1198.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":2294.6159909916,"imp":13.637367196,"elo":1203.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":3084.4850399813,"imp":13.6099486021,"elo":1206.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.4740218671,"imp":16.476873298,"elo":1098.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.043565096,"imp":15.9658325071,"elo":1138.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":33.9390995628,"imp":15.0241826156,"elo":1193.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":59.6650516119,"imp":14.4994493282,"elo":1220.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":145.2899663332,"imp":13.7079424681,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":289.9500533509,"imp":13.5379303989,"elo":1268.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":579.1158947796,"imp":13.4502748579,"elo":1273.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":868.4436513583,"imp":13.4227405332,"elo":1276.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1163.6859454819,"imp":13.410719162,"elo":1277.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.9730562426,"imp":19.7708791278,"elo":943.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":17.4702969526,"imp":19.4511518411,"elo":952.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":51.586202932,"imp":18.3361479448,"elo":992.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":104.4045259859,"imp":17.7551828211,"elo":1017.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":268.392593962,"imp":16.893416284,"elo":1056.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":544.3008720644,"imp":16.4197020156,"elo":1074.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1091.6618186196,"imp":15.6990520521,"elo":1094.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1644.5558204956,"imp":15.1709216436,"elo":1111.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2215.6113217439,"imp":14.7167094171,"elo":1127.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":4.301698659,"imp":11.9656913275,"elo":1194.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":10.0806694886,"imp":11.1493199851,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":27.1783376768,"imp":10.0766247952,"elo":1337.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":56.2958523673,"imp":9.549995294,"elo":1385.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":147.4003920877,"imp":9.0487887487,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":297.8210784396,"imp":8.8954611357,"elo":1455.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":595.7506287926,"imp":8.7470685411,"elo":1470.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":891.9636616374,"imp":8.6630195408,"elo":1477.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1203.2369028783,"imp":8.6059240236,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":121.7676060282,"imp":11.3517147974,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":93.3902149907,"imp":8.2564653552,"elo":1411.5,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":26.3794580707,"imp":12.7938174649,"elo":1222.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":74.5462848654,"imp":11.9361393097,"elo":1273.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":211.6307745864,"imp":10.5419688961,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":427.0209555269,"imp":9.9729505854,"elo":1401.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":1036.8544848991,"imp":9.3877040079,"elo":1440.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":2221.9585559949,"imp":9.2262055387,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":4422.440532253,"imp":9.0646470703,"elo":1459.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":6612.381411833,"imp":8.9911583767,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":8982.7355846534,"imp":8.9839616876,"elo":1462.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":12.3703933426,"imp":3.4465399965,"elo":1792.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":12.0914930116,"imp":1.9025306316,"elo":1866.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":32.5994632136,"imp":19.7613906462,"elo":917.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":65.3760935075,"imp":17.4650098427,"elo":1012.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":163.2396739605,"imp":15.3119972665,"elo":1109.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":195.7517175714,"imp":15.1390876691,"elo":1120.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.6268352454,"imp":17.4759679769,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.8932399565,"imp":16.7117040037,"elo":1054.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":13.9716136863,"imp":16.1587579057,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":30.8300578438,"imp":15.6510973053,"elo":1130.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":81.3676986662,"imp":15.3537028399,"elo":1153.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":165.6231513559,"imp":15.2602661276,"elo":1163.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":334.2669121033,"imp":15.1974368189,"elo":1172.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":502.9575285794,"imp":15.1773925578,"elo":1176.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":675.1680482075,"imp":15.1667001875,"elo":1178.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":21.5151248338,"imp":9.6941228488,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":44.1500541097,"imp":8.1861608935,"elo":1367.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":110.9865188464,"imp":6.4575299181,"elo":1500.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":225.999805441,"imp":5.8872105232,"elo":1559.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":564.497352775,"imp":5.5032471892,"elo":1598.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1133.451985238,"imp":5.306077294,"elo":1621.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2267.2333864899,"imp":5.1839759206,"elo":1631.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3401.9727496793,"imp":5.1009960087,"elo":1644.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4560.0060476784,"imp":5.0561973916,"elo":1652.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":7.6690681838,"imp":6.0088337009,"elo":1510.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":19.8560696815,"imp":5.4268698142,"elo":1535.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":55.2048577583,"imp":4.4399390169,"elo":1588.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":111.3159815209,"imp":4.0072743523,"elo":1632.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":289.4641949987,"imp":3.6223900919,"elo":1671.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":572.0729326007,"imp":3.4399933919,"elo":1691.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1145.5080925273,"imp":3.3440652133,"elo":1704.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1725.0191110982,"imp":3.3006915814,"elo":1710.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2307.574171405,"imp":3.2642372553,"elo":1716.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":22.4635991568,"imp":10.0442682646,"elo":1376.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":48.1376226066,"imp":6.2118668245,"elo":1580.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":69.1348663267,"imp":5.6401465297,"elo":1621.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":132.3528960267,"imp":5.1834343861,"elo":1668.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":250.1973072075,"imp":4.8641765664,"elo":1703.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":607.0781754894,"imp":4.669916536,"elo":1725.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1214.0044600928,"imp":4.6129537086,"elo":1732.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2403.8929291273,"imp":4.6079669763,"elo":1732.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3586.3826643416,"imp":4.6319083691,"elo":1728.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4823.3684232511,"imp":4.6447607428,"elo":1724.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":7.9333050678,"imp":6.0163424802,"elo":1569.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":49.2939082066,"imp":2.1601326915,"elo":1971.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":5.1543015637,"imp":4.35486957,"elo":1673.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":13.5301358652,"imp":8.8925386331,"elo":1304.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":30.4154141023,"imp":8.6104311775,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":85.0731093685,"imp":8.5930731892,"elo":1337.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":178.9262812137,"imp":8.0980172039,"elo":1370.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":451.7833076823,"imp":7.6173566018,"elo":1406.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":903.7701842028,"imp":7.309895667,"elo":1429.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1809.2442565983,"imp":7.091829374,"elo":1445.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2725.2737601417,"imp":6.6498140807,"elo":1457.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3660.2557617825,"imp":6.6021046982,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":9.2031778761,"imp":3.8663510257,"elo":1751.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":14.0862798048,"imp":2.6202851026,"elo":1806.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":8.7397288053,"imp":9.5438132182,"elo":1286.3,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":52.3185483507,"imp":8.40630195,"elo":1326.6,"imputed":true,"imputed_pct":12.5,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":144.3008163018,"imp":7.3884745258,"elo":1380.5,"imputed":true,"imputed_pct":12.5,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":314.2173893111,"imp":6.9481221115,"elo":1411.5,"imputed":true,"imputed_pct":12.5,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":770.3625024964,"imp":6.3223640517,"elo":1458.9,"imputed":true,"imputed_pct":12.5,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1584.8818989347,"imp":6.0256401438,"elo":1482.1,"imputed":true,"imputed_pct":12.5,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3211.2353391815,"imp":5.7716333356,"elo":1501.7,"imputed":true,"imputed_pct":12.5,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4841.0485061893,"imp":5.6893715806,"elo":1508.5,"imputed":true,"imputed_pct":12.5,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":6559.811990291,"imp":5.6215684893,"elo":1512.5,"imputed":true,"imputed_pct":12.5,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":394.6813525384,"imp":26.2174753709,"elo":961.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":831.1606530777,"imp":26.1178243363,"elo":976.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1632.8165586183,"imp":25.8791943512,"elo":992.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2696.2641572437,"imp":25.3517344905,"elo":1020.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":6500.6961192202,"imp":24.6272795292,"elo":1045.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":13195.5862950654,"imp":24.0627486688,"elo":1062.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":26552.7976576922,"imp":23.7095254932,"elo":1076.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":39668.7876205984,"imp":23.5073670123,"elo":1085.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":52983.7100836136,"imp":23.3556752415,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":3.5256005858,"imp":6.0798292372,"elo":1485.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":22.3971569167,"imp":14.9933861843,"elo":1049.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":48.9969305386,"imp":14.5517966766,"elo":1082.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":122.7996474091,"imp":13.7604519574,"elo":1126.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":260.4618741099,"imp":13.3577324835,"elo":1152.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":644.8021976959,"imp":12.7149322821,"elo":1190.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1311.6428242239,"imp":12.2151825026,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2562.5226347774,"imp":11.7024945276,"elo":1245.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3859.2194758298,"imp":11.483204996,"elo":1254.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":5151.8777257105,"imp":11.335419635,"elo":1261.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":4.0580642834,"imp":13.1686868603,"elo":1129.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":10.6096996819,"imp":10.9057093348,"elo":1258.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":30.177444805,"imp":10.0438242254,"elo":1322.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":60.5203899854,"imp":9.5855316604,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":156.9516526756,"imp":9.2773937627,"elo":1394.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":314.3467912249,"imp":9.1469202219,"elo":1407.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":629.7467608773,"imp":9.0466798531,"elo":1418.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":943.5999846765,"imp":8.9586305705,"elo":1425.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":1266.5674390493,"imp":8.9105313253,"elo":1432.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":725.4510762602,"imp":20.6408568168,"elo":853.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":2170.0049801645,"imp":15.2508952167,"elo":1054.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":6794.2091341246,"imp":13.7790420651,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":15326.7493172037,"imp":13.723547197,"elo":1354.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":37614.2771727845,"imp":9.1037027213,"elo":1487.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":39519.0698998597,"imp":9.063242256,"elo":1494.1,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":10.1998611956,"imp":15.541643283,"elo":1113.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":20.3286221704,"imp":13.9091648303,"elo":1184.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":50.8497537215,"imp":11.781892197,"elo":1275.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":101.4913596974,"imp":10.2607035962,"elo":1332.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":253.6820748947,"imp":8.9474349773,"elo":1389.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":507.7670435057,"imp":8.5043478621,"elo":1418.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":1015.1060714549,"imp":8.2245859685,"elo":1433.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1522.261390547,"imp":8.129169547,"elo":1442.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2039.6992814536,"imp":8.074956467,"elo":1447.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0489086629,"x_train":13.6316706327,"imp":9.9238592529,"elo":1364.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1144842333,"x_train":36.2728602647,"imp":9.2661240567,"elo":1407.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2272990226,"x_train":97.6908201342,"imp":8.5526519375,"elo":1454.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2840630283,"x_train":201.9357530623,"imp":8.2116280916,"elo":1472.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3167883742,"x_train":526.2527558948,"imp":7.8995961814,"elo":1485.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3510472721,"x_train":1057.486905704,"imp":7.8367356068,"elo":1490.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3908307212,"x_train":2119.3859898824,"imp":7.7928148055,"elo":1495.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4190199844,"x_train":3172.1186602631,"imp":7.7813830392,"elo":1495.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4436888454,"x_train":4283.1043461266,"imp":7.7596298454,"elo":1499.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0368319183,"x_train":8.2896427598,"imp":11.667714954,"elo":1239.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0688276247,"x_train":17.6287133887,"imp":11.4543747457,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1219067073,"x_train":44.8911284914,"imp":11.2321694494,"elo":1278.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1714117234,"x_train":90.9274231145,"imp":11.0389655949,"elo":1296.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2379383628,"x_train":228.5740056066,"imp":10.8238574899,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2766475086,"x_train":458.5111602981,"imp":10.6584554849,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3155848723,"x_train":918.1582493662,"imp":10.4931075398,"elo":1335.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3354039341,"x_train":1376.6746665085,"imp":10.4238490422,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.348849021,"x_train":1846.4337432747,"imp":10.3751093817,"elo":1346.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0130181103,"x_train":12.2643063605,"imp":15.1876942233,"elo":1086.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0248159619,"x_train":29.8736292865,"imp":14.8048532351,"elo":1115.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0457946914,"x_train":80.8890311763,"imp":14.4156511347,"elo":1152.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0624903018,"x_train":157.9967250155,"imp":14.1583098609,"elo":1171.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0826354068,"x_train":390.307349424,"imp":13.8341552659,"elo":1187.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0975020907,"x_train":781.3223148424,"imp":13.7461222732,"elo":1193.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1099002096,"x_train":1537.3695296789,"imp":13.674310407,"elo":1198.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1183419318,"x_train":2294.6159909916,"imp":13.637367196,"elo":1203.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1269231171,"x_train":3084.4850399813,"imp":13.6099486021,"elo":1206.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0076902476,"x_train":7.4740218671,"imp":16.476873298,"elo":1098.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1498964698,"x_train":14.043565096,"imp":15.9658325071,"elo":1138.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1014576612,"x_train":33.9390995628,"imp":15.0241826156,"elo":1193.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0082008847,"x_train":59.6650516119,"imp":14.4994493282,"elo":1220.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0626226355,"x_train":145.2899663332,"imp":13.7079424681,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3379772641,"x_train":289.9500533509,"imp":13.5379303989,"elo":1268.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.6772475427,"x_train":579.1158947796,"imp":13.4502748579,"elo":1273.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.953467119,"x_train":868.4436513583,"imp":13.4227405332,"elo":1276.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":2.0978537644,"x_train":1163.6859454819,"imp":13.410719162,"elo":1277.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8180850143,"x_train":6.9730562426,"imp":19.7708791278,"elo":943.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.579872527,"x_train":17.4702969526,"imp":19.4511518411,"elo":952.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8040014596,"x_train":51.586202932,"imp":18.3361479448,"elo":992.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.9064906779,"x_train":104.4045259859,"imp":17.7551828211,"elo":1017.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6070665305,"x_train":268.392593962,"imp":16.893416284,"elo":1056.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.9054304219,"x_train":544.3008720644,"imp":16.4197020156,"elo":1074.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.5132807751,"x_train":1091.6618186196,"imp":15.6990520521,"elo":1094.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7654538542,"x_train":1644.5558204956,"imp":15.1709216436,"elo":1111.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.7913188557,"x_train":2215.6113217439,"imp":14.7167094171,"elo":1127.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.3174685594,"x_train":4.301698659,"imp":11.9656913275,"elo":1194.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5015390954,"x_train":10.0806694886,"imp":11.1493199851,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.8415814505,"x_train":27.1783376768,"imp":10.0766247952,"elo":1337.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.7464507056,"x_train":56.2958523673,"imp":9.549995294,"elo":1385.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.9217323461,"x_train":147.4003920877,"imp":9.0487887487,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.166466432,"x_train":297.8210784396,"imp":8.8954611357,"elo":1455.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.9020363863,"x_train":595.7506287926,"imp":8.7470685411,"elo":1470.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.08197779,"x_train":891.9636616374,"imp":8.6630195408,"elo":1477.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.1056715587,"x_train":1203.2369028783,"imp":8.6059240236,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":21.8030250918,"x_train":121.7676060282,"imp":11.3517147974,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.3955680904,"x_train":93.3902149907,"imp":8.2564653552,"elo":1411.5,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3020207606,"x_train":26.3794580707,"imp":12.7938174649,"elo":1222.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7507962656,"x_train":74.5462848654,"imp":11.9361393097,"elo":1273.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.6316502282,"x_train":211.6307745864,"imp":10.5419688961,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.4921562567,"x_train":427.0209555269,"imp":9.9729505854,"elo":1401.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.017217154,"x_train":1036.8544848991,"imp":9.3877040079,"elo":1440.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.2418672637,"x_train":2221.9585559949,"imp":9.2262055387,"elo":1450.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.6456169825,"x_train":4422.440532253,"imp":9.0646470703,"elo":1459.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5039594902,"x_train":6612.381411833,"imp":8.9911583767,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0116272048,"x_train":8982.7355846534,"imp":8.9839616876,"elo":1462.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":1.5662902939,"x_train":12.3703933426,"imp":3.4465399965,"elo":1792.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0528542696,"x_train":12.0914930116,"imp":1.9025306316,"elo":1866.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.021441561,"x_train":32.5994632136,"imp":19.7613906462,"elo":917.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1447058964,"x_train":65.3760935075,"imp":17.4650098427,"elo":1012.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3551781139,"x_train":163.2396739605,"imp":15.3119972665,"elo":1109.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3602092205,"x_train":195.7517175714,"imp":15.1390876691,"elo":1120.4,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1203913506,"x_train":0.6268352454,"imp":17.4759679769,"elo":999.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3503086463,"x_train":3.8932399565,"imp":16.7117040037,"elo":1054.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5821895355,"x_train":13.9716136863,"imp":16.1587579057,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.806509101,"x_train":30.8300578438,"imp":15.6510973053,"elo":1130.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0727860502,"x_train":81.3676986662,"imp":15.3537028399,"elo":1153.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2227674298,"x_train":165.6231513559,"imp":15.2602661276,"elo":1163.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4040008059,"x_train":334.2669121033,"imp":15.1974368189,"elo":1172.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4924073803,"x_train":502.9575285794,"imp":15.1773925578,"elo":1176.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.5448750192,"x_train":675.1680482075,"imp":15.1667001875,"elo":1178.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.3504595229,"x_train":21.5151248338,"imp":9.6941228488,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.3689378053,"x_train":44.1500541097,"imp":8.1861608935,"elo":1367.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.2357930967,"x_train":110.9865188464,"imp":6.4575299181,"elo":1500.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":18.2146470478,"x_train":225.999805441,"imp":5.8872105232,"elo":1559.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.1100006203,"x_train":564.497352775,"imp":5.5032471892,"elo":1598.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":24.4652884983,"x_train":1133.451985238,"imp":5.306077294,"elo":1621.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":28.5926981827,"x_train":2267.2333864899,"imp":5.1839759206,"elo":1631.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":31.4173043227,"x_train":3401.9727496793,"imp":5.1009960087,"elo":1644.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":33.8307336671,"x_train":4560.0060476784,"imp":5.0561973916,"elo":1652.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5470075607,"x_train":7.6690681838,"imp":6.0088337009,"elo":1510.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4422544088,"x_train":19.8560696815,"imp":5.4268698142,"elo":1535.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.7604907751,"x_train":55.2048577583,"imp":4.4399390169,"elo":1588.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.6627807362,"x_train":111.3159815209,"imp":4.0072743523,"elo":1632.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.2418923926,"x_train":289.4641949987,"imp":3.6223900919,"elo":1671.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.0570210776,"x_train":572.0729326007,"imp":3.4399933919,"elo":1691.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4887817681,"x_train":1145.5080925273,"imp":3.3440652133,"elo":1704.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.9791618808,"x_train":1725.0191110982,"imp":3.3006915814,"elo":1710.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.2648971866,"x_train":2307.574171405,"imp":3.2642372553,"elo":1716.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.0380986551,"x_train":22.4635991568,"imp":10.0442682646,"elo":1376.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":42.001452678,"x_train":48.1376226066,"imp":6.2118668245,"elo":1580.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":48.0919926375,"x_train":69.1348663267,"imp":5.6401465297,"elo":1621.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":55.5080414247,"x_train":132.3528960267,"imp":5.1834343861,"elo":1668.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":69.0757561523,"x_train":250.1973072075,"imp":4.8641765664,"elo":1703.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":89.4729687406,"x_train":607.0781754894,"imp":4.669916536,"elo":1725.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":114.5105718552,"x_train":1214.0044600928,"imp":4.6129537086,"elo":1732.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":141.2613310565,"x_train":2403.8929291273,"imp":4.6079669763,"elo":1732.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":143.7240278852,"x_train":3586.3826643416,"imp":4.6319083691,"elo":1728.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":140.7371116271,"x_train":4823.3684232511,"imp":4.6447607428,"elo":1724.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2636901296,"x_train":7.9333050678,"imp":6.0163424802,"elo":1569.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":4.0942168713,"x_train":49.2939082066,"imp":2.1601326915,"elo":1971.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1402308582,"x_train":5.1543015637,"imp":4.35486957,"elo":1673.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4666328992,"x_train":13.5301358652,"imp":8.8925386331,"elo":1304.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.9441490344,"x_train":30.4154141023,"imp":8.6104311775,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.8546554659,"x_train":85.0731093685,"imp":8.5930731892,"elo":1337.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.4234961622,"x_train":178.9262812137,"imp":8.0980172039,"elo":1370.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":3.1588401041,"x_train":451.7833076823,"imp":7.6173566018,"elo":1406.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.744707909,"x_train":903.7701842028,"imp":7.309895667,"elo":1429.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":4.3265393551,"x_train":1809.2442565983,"imp":7.091829374,"elo":1445.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.5964870629,"x_train":2725.2737601417,"imp":6.6498140807,"elo":1457.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.7546477722,"x_train":3660.2557617825,"imp":6.6021046982,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5852250627,"x_train":9.2031778761,"imp":3.8663510257,"elo":1751.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4388252423,"x_train":14.0862798048,"imp":2.6202851026,"elo":1806.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.7725507791,"x_train":8.7397288053,"imp":9.5438132182,"elo":1286.3,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":7.2956294278,"x_train":52.3185483507,"imp":8.40630195,"elo":1326.6,"imputed":true,"imputed_pct":12.5,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":16.9606282311,"x_train":144.3008163018,"imp":7.3884745258,"elo":1380.5,"imputed":true,"imputed_pct":12.5,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":27.0774429419,"x_train":314.2173893111,"imp":6.9481221115,"elo":1411.5,"imputed":true,"imputed_pct":12.5,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":35.9358742992,"x_train":770.3625024964,"imp":6.3223640517,"elo":1458.9,"imputed":true,"imputed_pct":12.5,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":42.8882096793,"x_train":1584.8818989347,"imp":6.0256401438,"elo":1482.1,"imputed":true,"imputed_pct":12.5,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":50.041743122,"x_train":3211.2353391815,"imp":5.7716333356,"elo":1501.7,"imputed":true,"imputed_pct":12.5,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":53.9207044786,"x_train":4841.0485061893,"imp":5.6893715806,"elo":1508.5,"imputed":true,"imputed_pct":12.5,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":56.2866799416,"x_train":6559.811990291,"imp":5.6215684893,"elo":1512.5,"imputed":true,"imputed_pct":12.5,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5202853799,"x_train":394.6813525384,"imp":26.2174753709,"elo":961.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3982818311,"x_train":831.1606530777,"imp":26.1178243363,"elo":976.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0787209898,"x_train":1632.8165586183,"imp":25.8791943512,"elo":992.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.3600489151,"x_train":2696.2641572437,"imp":25.3517344905,"elo":1020.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":23.7183592261,"x_train":6500.6961192202,"imp":24.6272795292,"elo":1045.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":28.12157704,"x_train":13195.5862950654,"imp":24.0627486688,"elo":1062.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":32.840614162,"x_train":26552.7976576922,"imp":23.7095254932,"elo":1076.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":34.8841403507,"x_train":39668.7876205984,"imp":23.5073670123,"elo":1085.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.3108208411,"x_train":52983.7100836136,"imp":23.3556752415,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0639725065,"x_train":3.5256005858,"imp":6.0798292372,"elo":1485.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1340751587,"x_train":22.3971569167,"imp":14.9933861843,"elo":1049.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2671714395,"x_train":48.9969305386,"imp":14.5517966766,"elo":1082.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5043831924,"x_train":122.7996474091,"imp":13.7604519574,"elo":1126.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7072384242,"x_train":260.4618741099,"imp":13.3577324835,"elo":1152.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.979629298,"x_train":644.8021976959,"imp":12.7149322821,"elo":1190.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2101455751,"x_train":1311.6428242239,"imp":12.2151825026,"elo":1222.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3941877423,"x_train":2562.5226347774,"imp":11.7024945276,"elo":1245.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5127765948,"x_train":3859.2194758298,"imp":11.483204996,"elo":1254.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5906165471,"x_train":5151.8777257105,"imp":11.335419635,"elo":1261.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2989233835,"x_train":4.0580642834,"imp":13.1686868603,"elo":1129.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8790913968,"x_train":10.6096996819,"imp":10.9057093348,"elo":1258.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.6245949041,"x_train":30.177444805,"imp":10.0438242254,"elo":1322.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0221234831,"x_train":60.5203899854,"imp":9.5855316604,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.4398238016,"x_train":156.9516526756,"imp":9.2773937627,"elo":1394.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.655128881,"x_train":314.3467912249,"imp":9.1469202219,"elo":1407.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.9193143966,"x_train":629.7467608773,"imp":9.0466798531,"elo":1418.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.0384704369,"x_train":943.5999846765,"imp":8.9586305705,"elo":1425.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.1465188849,"x_train":1266.5674390493,"imp":8.9105313253,"elo":1432.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":268.7612898715,"x_train":725.4510762602,"imp":20.6408568168,"elo":853.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":565.8509098848,"x_train":2170.0049801645,"imp":15.2508952167,"elo":1054.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":1025.1468177864,"x_train":6794.2091341246,"imp":13.7790420651,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":1403.2757745602,"x_train":15326.7493172037,"imp":13.723547197,"elo":1354.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":1492.6468598574,"x_train":37614.2771727845,"imp":9.1037027213,"elo":1487.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":1519.4791766157,"x_train":39519.0698998597,"imp":9.063242256,"elo":1494.1,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":2.7158548006,"x_train":10.1998611956,"imp":15.541643283,"elo":1113.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":2.8570323643,"x_train":20.3286221704,"imp":13.9091648303,"elo":1184.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.0885835567,"x_train":50.8497537215,"imp":11.781892197,"elo":1275.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":2.1746682726,"x_train":101.4913596974,"imp":10.2607035962,"elo":1332.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.1322181978,"x_train":253.6820748947,"imp":8.9474349773,"elo":1389.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.9099808138,"x_train":507.7670435057,"imp":8.5043478621,"elo":1418.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.6891571227,"x_train":1015.1060714549,"imp":8.2245859685,"elo":1433.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.443541988,"x_train":1522.261390547,"imp":8.129169547,"elo":1442.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.1468082511,"x_train":2039.6992814536,"imp":8.074956467,"elo":1447.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_all/tasks_regression/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..25136b527bbfe13ab318c0b87c40890bf3b4ac78
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_all/tasks_regression/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "Nori-30M (default)", "TabPFN-3 (default)", "Nori (default)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "TabDPT (tuned + ensembled)", "RealMLP (tuned + ensembled)", "TabDPT (tuned)", "RealTabPFN-2.5 (tuned)", "TabDPT (default)", "RealTabPFN-2.5 (default)", "TabPFNv2 (tuned + ensembled)", "RealMLP (tuned)", "TabDPT-Turbo (default)", "AutoGluon 1.4 (best, 4h)", "CatBoost (tuned + ensembled)", "TabSwift (default)", "iLTM (tuned + ensembled)", "Mitra (default)", "CatBoost (tuned)", "TabM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "SAP-RPT-OSS (default)", "xRFM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabPFNv2 (tuned)", "ModernNCA (tuned + ensembled)", "XGBoost (tuned)", "CatBoost (default)", "TabPFNv2 (default)", "LightGBM (tuned)", "TabM (tuned)", "xRFM (tuned)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "TabM (default)", "ChimeraBoost (tuned)", "RealMLP (default)", "ExtraTrees (tuned + ensembled)", "ModernNCA (tuned)", "EBM (tuned + ensembled)", "ExtraTrees (tuned)", "ChimeraBoost (default)", "TorchMLP (tuned + ensembled)", "LimiX (default)", "EBM (tuned)", "TorchMLP (tuned)", "ModernNCA (default)", "LightGBM (default)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned + ensembled)", "TabSTAR (tuned)", "EBM (default)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "XGBoost (default)", "FastaiMLP (tuned + ensembled)", "xRFM (default)", "PerpetualBooster (tuned)", "TorchMLP (default)", "TabSTAR (default)", "FastaiMLP (tuned)", "RandomForest (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "FastaiMLP (default)", "KNN (tuned)", "iLTM (default)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 0.5141242742538452, 0.6892655491828918, 0.7457627058029175, 0.7062146663665771, 0.7683615684509277, 0.7118644118309021, 0.7401130199432373, 0.8757061958312988, 0.8757061958312988, 0.9209039807319641, 0.7683615684509277, 0.8926553726196289, 0.9152542352676392, 0.8418079018592834, 0.9322034120559692, 0.9548022747039795, 0.9491525292396545, 0.9548022747039795, 0.8700565099716187, 0.9548022747039795, 0.9322034120559692, 0.9717513918876648, 0.9435028433799744, 0.9661017060279846, 0.9152542352676392, 0.9378530979156494, 0.9548022747039795, 0.8813559412956238, 0.9491525292396545, 0.9661017060279846, 0.9604519605636597, 0.9604519605636597, 0.9717513918876648, 0.9604519605636597, 0.9604519605636597, 0.9661017060279846, 0.9717513918876648, 0.9717513918876648, 0.9604519605636597, 0.9887005686759949, 0.9661017060279846, 0.9887005686759949, 0.9661017060279846, 0.9717513918876648, 0.994350254535675, 0.9887005686759949, 0.9830508232116699, 0.9774011373519897, 0.994350254535675, 0.9887005686759949, 1.0, 0.903954803943634, 0.994350254535675, 0.909604549407959, 0.9830508232116699, 0.994350254535675, 0.9887005686759949, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 0.994350254535675, 1.0, 0.9265536665916443, 1.0, 0.994350254535675, 0.9830508232116699, 1.0, 1.0, 0.994350254535675, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.4858756959438324, NaN, 0.5480226278305054, 0.7231638431549072, 0.6327683329582214, 0.7683615684509277, 0.7344632744789124, 0.7344632744789124, 0.7062146663665771, 0.7796609997749329, 0.790960431098938, 0.8135592937469482, 0.790960431098938, 0.8700565099716187, 0.8757061958312988, 0.8700565099716187, 0.8644067645072937, 0.909604549407959, 0.9265536665916443, 0.903954803943634, 0.887005627155304, 0.9491525292396545, 0.9604519605636597, 0.9209039807319641, 0.9548022747039795, 0.8813559412956238, 0.9152542352676392, 0.9378530979156494, 0.9378530979156494, 0.903954803943634, 0.9491525292396545, 0.9604519605636597, 0.9378530979156494, 0.9661017060279846, 0.9435028433799744, 0.9435028433799744, 0.9774011373519897, 0.9774011373519897, 0.9661017060279846, 0.9830508232116699, 0.9661017060279846, 0.9548022747039795, 0.9491525292396545, 0.9830508232116699, 0.9378530979156494, 0.9774011373519897, 0.9661017060279846, 0.9774011373519897, 0.9830508232116699, 0.9830508232116699, 0.9774011373519897, 0.994350254535675, 0.8926553726196289, 0.9830508232116699, 0.903954803943634, 0.9887005686759949, 0.9830508232116699, 0.9887005686759949, 0.9604519605636597, 0.994350254535675, 0.994350254535675, 0.9661017060279846, 0.994350254535675, 1.0, 0.909604549407959, 1.0, 0.9887005686759949, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9774011373519897, 1.0, 1.0, 1.0], [0.31073445081710815, 0.451977401971817, NaN, 0.5536723136901855, 0.5593220591545105, 0.598870038986206, 0.604519784450531, 0.5819209218025208, 0.7062146663665771, 0.790960431098938, 0.7288135886192322, 0.7457627058029175, 0.8135592937469482, 0.8474576473236084, 0.8305084705352783, 0.8305084705352783, 0.887005627155304, 0.8531073331832886, 0.8700565099716187, 0.9209039807319641, 0.887005627155304, 0.790960431098938, 0.8757061958312988, 0.8644067645072937, 0.909604549407959, 0.8418079018592834, 0.8926553726196289, 0.909604549407959, 0.909604549407959, 0.8700565099716187, 0.903954803943634, 0.8926553726196289, 0.9548022747039795, 0.9378530979156494, 0.903954803943634, 0.909604549407959, 0.8983050584793091, 0.9661017060279846, 0.9209039807319641, 0.903954803943634, 0.9604519605636597, 0.9322034120559692, 0.9209039807319641, 0.9265536665916443, 0.9435028433799744, 0.9491525292396545, 0.9661017060279846, 0.909604549407959, 0.9435028433799744, 0.9717513918876648, 0.9604519605636597, 0.9887005686759949, 0.8248587846755981, 0.9717513918876648, 0.8474576473236084, 0.9661017060279846, 0.9717513918876648, 0.9887005686759949, 0.9604519605636597, 0.9887005686759949, 1.0, 0.9548022747039795, 0.9887005686759949, 1.0, 0.909604549407959, 1.0, 0.994350254535675, 0.9604519605636597, 1.0, 1.0, 0.9774011373519897, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.2542372941970825, 0.2768361568450928, 0.44632768630981445, NaN, 0.49152541160583496, 0.6158192157745361, 0.5706214904785156, 0.5706214904785156, 0.6610169410705566, 0.6666666865348816, 0.700564980506897, 0.6610169410705566, 0.7344632744789124, 0.8587570786476135, 0.7288135886192322, 0.790960431098938, 0.8361582159996033, 0.8926553726196289, 0.9548022747039795, 0.7683615684509277, 0.8531073331832886, 0.8700565099716187, 0.9491525292396545, 0.8644067645072937, 0.9491525292396545, 0.8587570786476135, 0.903954803943634, 0.9209039807319641, 0.8531073331832886, 0.8813559412956238, 0.9378530979156494, 0.9491525292396545, 0.8700565099716187, 0.9491525292396545, 0.903954803943634, 0.9491525292396545, 0.9604519605636597, 0.9774011373519897, 0.9491525292396545, 0.9717513918876648, 0.9209039807319641, 0.9378530979156494, 0.9435028433799744, 0.9774011373519897, 0.9378530979156494, 0.9830508232116699, 0.9548022747039795, 0.9322034120559692, 0.9661017060279846, 0.9830508232116699, 0.9717513918876648, 1.0, 0.8757061958312988, 0.9887005686759949, 0.8757061958312988, 0.9830508232116699, 0.9887005686759949, 0.9887005686759949, 0.9717513918876648, 0.994350254535675, 0.9830508232116699, 0.9604519605636597, 0.9887005686759949, 1.0, 0.903954803943634, 0.9887005686759949, 0.994350254535675, 0.9717513918876648, 1.0, 1.0, 0.9830508232116699, 1.0, 0.9774011373519897, 1.0, 1.0, 1.0], [0.29378530383110046, 0.3672316372394562, 0.4406779706478119, 0.508474588394165, NaN, 0.5762711763381958, 0.4971751272678375, 0.5480226278305054, 0.7288135886192322, 0.7457627058029175, 0.7231638431549072, 0.6836158037185669, 0.7627118825912476, 0.8474576473236084, 0.7288135886192322, 0.8248587846755981, 0.8305084705352783, 0.8587570786476135, 0.8983050584793091, 0.790960431098938, 0.8926553726196289, 0.8135592937469482, 0.8926553726196289, 0.8926553726196289, 0.8926553726196289, 0.887005627155304, 0.8757061958312988, 0.903954803943634, 0.8531073331832886, 0.8644067645072937, 0.8926553726196289, 0.903954803943634, 0.9209039807319641, 0.9152542352676392, 0.8983050584793091, 0.8926553726196289, 0.887005627155304, 0.9265536665916443, 0.903954803943634, 0.8926553726196289, 0.9322034120559692, 0.8983050584793091, 0.9209039807319641, 0.9378530979156494, 0.8926553726196289, 0.9378530979156494, 0.9435028433799744, 0.9152542352676392, 0.9322034120559692, 0.9265536665916443, 0.9265536665916443, 0.9491525292396545, 0.8361582159996033, 0.9604519605636597, 0.8531073331832886, 0.9774011373519897, 0.9717513918876648, 0.9661017060279846, 0.9491525292396545, 0.9491525292396545, 0.9491525292396545, 0.9661017060279846, 0.9661017060279846, 0.9491525292396545, 0.9152542352676392, 0.9491525292396545, 0.9661017060279846, 0.9774011373519897, 1.0, 0.9491525292396545, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.23163841664791107, 0.23163841664791107, 0.40112993121147156, 0.38418078422546387, 0.4237288236618042, NaN, 0.4350282549858093, 0.5254237055778503, 0.6158192157745361, 0.6158192157745361, 0.6497175097465515, 0.6497175097465515, 0.700564980506897, 0.8361582159996033, 0.6610169410705566, 0.7627118825912476, 0.7457627058029175, 0.8418079018592834, 0.8926553726196289, 0.8192090392112732, 0.8418079018592834, 0.8587570786476135, 0.9209039807319641, 0.903954803943634, 0.887005627155304, 0.8079096078872681, 0.8757061958312988, 0.9209039807319641, 0.8192090392112732, 0.8587570786476135, 0.9265536665916443, 0.9378530979156494, 0.8644067645072937, 0.9209039807319641, 0.909604549407959, 0.9435028433799744, 0.9435028433799744, 0.9661017060279846, 0.9491525292396545, 0.9548022747039795, 0.9661017060279846, 0.9209039807319641, 0.9322034120559692, 0.9661017060279846, 0.9322034120559692, 0.9830508232116699, 0.9491525292396545, 0.9265536665916443, 0.9774011373519897, 0.9830508232116699, 0.9830508232116699, 0.994350254535675, 0.8757061958312988, 0.9887005686759949, 0.887005627155304, 0.9774011373519897, 0.9887005686759949, 0.994350254535675, 0.9548022747039795, 1.0, 0.9774011373519897, 0.9717513918876648, 0.994350254535675, 1.0, 0.9209039807319641, 0.9887005686759949, 0.994350254535675, 0.9830508232116699, 1.0, 0.994350254535675, 0.9830508232116699, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.2881355881690979, 0.26553672552108765, 0.395480215549469, 0.42937853932380676, 0.5028248429298401, 0.5649717450141907, NaN, 0.49152541160583496, 0.6610169410705566, 0.7288135886192322, 0.6723163723945618, 0.6440678238868713, 0.7118644118309021, 0.7853107452392578, 0.6779661178588867, 0.8135592937469482, 0.8418079018592834, 0.8305084705352783, 0.8305084705352783, 0.8305084705352783, 0.8813559412956238, 0.8248587846755981, 0.8587570786476135, 0.887005627155304, 0.8587570786476135, 0.7683615684509277, 0.8474576473236084, 0.903954803943634, 0.8474576473236084, 0.8644067645072937, 0.9209039807319641, 0.8757061958312988, 0.903954803943634, 0.9152542352676392, 0.903954803943634, 0.887005627155304, 0.903954803943634, 0.9265536665916443, 0.9265536665916443, 0.9265536665916443, 0.9717513918876648, 0.9209039807319641, 0.9548022747039795, 0.9548022747039795, 0.9265536665916443, 0.9322034120559692, 0.9830508232116699, 0.8757061958312988, 0.9548022747039795, 0.9830508232116699, 0.9661017060279846, 0.9717513918876648, 0.8700565099716187, 0.9604519605636597, 0.8757061958312988, 0.9435028433799744, 0.9661017060279846, 0.9887005686759949, 0.9378530979156494, 0.994350254535675, 0.994350254535675, 0.9661017060279846, 0.994350254535675, 1.0, 0.9152542352676392, 1.0, 1.0, 0.9717513918876648, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.2598870098590851, 0.26553672552108765, 0.41807910799980164, 0.42937853932380676, 0.451977401971817, 0.47457626461982727, 0.508474588394165, NaN, 0.6101694703102112, 0.6440678238868713, 0.6440678238868713, 0.8079096078872681, 0.7062146663665771, 0.8531073331832886, 0.7288135886192322, 0.7344632744789124, 0.7853107452392578, 0.8022598624229431, 0.8079096078872681, 0.8248587846755981, 0.7966101765632629, 0.790960431098938, 0.8587570786476135, 0.887005627155304, 0.8587570786476135, 0.7683615684509277, 0.8135592937469482, 0.8700565099716187, 0.903954803943634, 0.7853107452392578, 0.8813559412956238, 0.903954803943634, 0.8926553726196289, 0.887005627155304, 0.9209039807319641, 0.8587570786476135, 0.909604549407959, 0.903954803943634, 0.9491525292396545, 0.9265536665916443, 0.9378530979156494, 0.8531073331832886, 0.909604549407959, 0.9435028433799744, 0.8587570786476135, 0.9661017060279846, 0.9717513918876648, 0.8926553726196289, 0.9604519605636597, 0.9774011373519897, 0.9322034120559692, 0.9830508232116699, 0.8305084705352783, 0.9435028433799744, 0.8474576473236084, 0.9661017060279846, 0.9548022747039795, 0.9830508232116699, 0.9435028433799744, 0.9774011373519897, 1.0, 0.9152542352676392, 0.9830508232116699, 1.0, 0.8983050584793091, 1.0, 0.9887005686759949, 0.9604519605636597, 1.0, 1.0, 0.9717513918876648, 1.0, 0.9830508232116699, 1.0, 1.0, 1.0], [0.12429378181695938, 0.29378530383110046, 0.29378530383110046, 0.33898305892944336, 0.2711864411830902, 0.38418078422546387, 0.33898305892944336, 0.38983049988746643, NaN, 0.5649717450141907, 0.6158192157745361, 0.5593220591545105, 0.7231638431549072, 0.6892655491828918, 0.5875706076622009, 0.6384180784225464, 0.7796609997749329, 0.790960431098938, 0.8022598624229431, 0.7231638431549072, 0.8644067645072937, 0.7401130199432373, 0.8248587846755981, 0.7627118825912476, 0.8361582159996033, 0.7457627058029175, 0.8644067645072937, 0.8700565099716187, 0.7457627058029175, 0.8418079018592834, 0.903954803943634, 0.8813559412956238, 0.790960431098938, 0.8644067645072937, 0.8305084705352783, 0.903954803943634, 0.8587570786476135, 0.9378530979156494, 0.8926553726196289, 0.8813559412956238, 0.9378530979156494, 0.9378530979156494, 0.9435028433799744, 0.887005627155304, 0.9265536665916443, 0.9152542352676392, 0.9209039807319641, 0.8926553726196289, 0.8757061958312988, 0.9491525292396545, 0.9548022747039795, 0.9717513918876648, 0.7853107452392578, 0.9774011373519897, 0.7853107452392578, 0.9435028433799744, 0.9830508232116699, 0.9830508232116699, 0.9548022747039795, 0.994350254535675, 1.0, 0.9491525292396545, 0.9830508232116699, 1.0, 0.8926553726196289, 1.0, 1.0, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.12429378181695938, 0.22033898532390594, 0.20903955399990082, 0.3333333432674408, 0.2542372941970825, 0.38418078422546387, 0.2711864411830902, 0.35593220591545105, 0.4350282549858093, NaN, 0.508474588394165, 0.47457626461982727, 0.5819209218025208, 0.6158192157745361, 0.5762711763381958, 0.7231638431549072, 0.6553672552108765, 0.7062146663665771, 0.7796609997749329, 0.6610169410705566, 0.8192090392112732, 0.7514124512672424, 0.7740113139152527, 0.7401130199432373, 0.7853107452392578, 0.7401130199432373, 0.8757061958312988, 0.8531073331832886, 0.7401130199432373, 0.8079096078872681, 0.8700565099716187, 0.8361582159996033, 0.8700565099716187, 0.8757061958312988, 0.8474576473236084, 0.9152542352676392, 0.8135592937469482, 0.9265536665916443, 0.8926553726196289, 0.8135592937469482, 0.9830508232116699, 0.9265536665916443, 0.9265536665916443, 0.8531073331832886, 0.9378530979156494, 0.8983050584793091, 0.8813559412956238, 0.8983050584793091, 0.8757061958312988, 0.9209039807319641, 0.9604519605636597, 0.9830508232116699, 0.7683615684509277, 0.9774011373519897, 0.7740113139152527, 0.9661017060279846, 0.9830508232116699, 0.9717513918876648, 0.9717513918876648, 0.994350254535675, 1.0, 0.9604519605636597, 0.9830508232116699, 0.994350254535675, 0.8757061958312988, 0.994350254535675, 0.994350254535675, 0.9604519605636597, 1.0, 1.0, 0.9774011373519897, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.07909604161977768, 0.20903955399990082, 0.2711864411830902, 0.299435019493103, 0.2768361568450928, 0.3502824902534485, 0.32768362760543823, 0.35593220591545105, 0.38418078422546387, 0.49152541160583496, NaN, 0.5310734510421753, 0.6384180784225464, 0.5819209218025208, 0.5197740197181702, 0.604519784450531, 0.6836158037185669, 0.7570621371269226, 0.7740113139152527, 0.6723163723945618, 0.8079096078872681, 0.7457627058029175, 0.7966101765632629, 0.700564980506897, 0.8022598624229431, 0.7457627058029175, 0.8305084705352783, 0.8474576473236084, 0.694915235042572, 0.8248587846755981, 0.8700565099716187, 0.8418079018592834, 0.7344632744789124, 0.8700565099716187, 0.7853107452392578, 0.887005627155304, 0.8305084705352783, 0.9435028433799744, 0.8644067645072937, 0.8531073331832886, 0.9209039807319641, 0.9265536665916443, 0.9322034120559692, 0.8644067645072937, 0.9152542352676392, 0.9152542352676392, 0.9209039807319641, 0.903954803943634, 0.8757061958312988, 0.9378530979156494, 0.9548022747039795, 0.9774011373519897, 0.7740113139152527, 0.9604519605636597, 0.7796609997749329, 0.9322034120559692, 0.9661017060279846, 0.9774011373519897, 0.9491525292396545, 0.994350254535675, 0.994350254535675, 0.9548022747039795, 0.9774011373519897, 1.0, 0.8983050584793091, 1.0, 0.9887005686759949, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.23163841664791107, 0.18644067645072937, 0.2542372941970825, 0.33898305892944336, 0.3163841664791107, 0.3502824902534485, 0.35593220591545105, 0.19209039211273193, 0.4406779706478119, 0.5254237055778503, 0.4689265489578247, NaN, 0.5762711763381958, 0.6581920981407166, 0.6214689016342163, 0.604519784450531, 0.6553672552108765, 0.6327683329582214, 0.6779661178588867, 0.7118644118309021, 0.6779661178588867, 0.6497175097465515, 0.7118644118309021, 0.7401130199432373, 0.700564980506897, 0.6610169410705566, 0.7570621371269226, 0.7796609997749329, 0.8248587846755981, 0.7062146663665771, 0.8022598624229431, 0.7570621371269226, 0.7853107452392578, 0.7683615684509277, 0.8248587846755981, 0.8248587846755981, 0.7853107452392578, 0.8531073331832886, 0.8700565099716187, 0.8192090392112732, 0.8926553726196289, 0.8248587846755981, 0.8135592937469482, 0.8531073331832886, 0.8305084705352783, 0.887005627155304, 0.8926553726196289, 0.8361582159996033, 0.8813559412956238, 0.909604549407959, 0.8757061958312988, 0.9491525292396545, 0.7570621371269226, 0.8700565099716187, 0.7627118825912476, 0.9322034120559692, 0.887005627155304, 0.9435028433799744, 0.887005627155304, 0.9435028433799744, 0.9717513918876648, 0.8757061958312988, 0.9548022747039795, 0.994350254535675, 0.8644067645072937, 1.0, 0.9548022747039795, 0.9152542352676392, 1.0, 0.994350254535675, 0.9661017060279846, 1.0, 0.9604519605636597, 1.0, 1.0, 1.0], [0.10734463483095169, 0.20903955399990082, 0.18644067645072937, 0.26553672552108765, 0.23728813230991364, 0.299435019493103, 0.2881355881690979, 0.29378530383110046, 0.2768361568450928, 0.41807910799980164, 0.3615819215774536, 0.4237288236618042, NaN, 0.47457626461982727, 0.508474588394165, 0.4858756959438324, 0.6101694703102112, 0.6271186470985413, 0.700564980506897, 0.6101694703102112, 0.7514124512672424, 0.6666666865348816, 0.7175140976905823, 0.5819209218025208, 0.700564980506897, 0.7062146663665771, 0.7683615684509277, 0.7514124512672424, 0.6327683329582214, 0.7627118825912476, 0.7853107452392578, 0.7627118825912476, 0.7062146663665771, 0.7796609997749329, 0.6553672552108765, 0.8022598624229431, 0.7627118825912476, 0.8757061958312988, 0.8022598624229431, 0.7740113139152527, 0.8531073331832886, 0.9322034120559692, 0.8813559412956238, 0.8135592937469482, 0.9378530979156494, 0.8248587846755981, 0.8757061958312988, 0.8700565099716187, 0.8305084705352783, 0.909604549407959, 0.9548022747039795, 0.9435028433799744, 0.7853107452392578, 0.9661017060279846, 0.7796609997749329, 0.9322034120559692, 0.9717513918876648, 0.9604519605636597, 0.9491525292396545, 0.994350254535675, 0.9830508232116699, 0.9604519605636597, 0.9830508232116699, 0.9887005686759949, 0.903954803943634, 0.9830508232116699, 0.994350254535675, 0.9774011373519897, 1.0, 1.0, 0.9887005686759949, 1.0, 0.9887005686759949, 1.0, 1.0, 1.0], [0.08474576473236084, 0.12994350492954254, 0.1525423675775528, 0.14124293625354767, 0.1525423675775528, 0.16384181380271912, 0.21468926966190338, 0.14689265191555023, 0.31073445081710815, 0.38418078422546387, 0.41807910799980164, 0.34180790185928345, 0.5254237055778503, NaN, 0.48022598028182983, 0.5367231369018555, 0.6214689016342163, 0.6214689016342163, 0.694915235042572, 0.6101694703102112, 0.6779661178588867, 0.6779661178588867, 0.7514124512672424, 0.6610169410705566, 0.6836158037185669, 0.6384180784225464, 0.7457627058029175, 0.7570621371269226, 0.6723163723945618, 0.7401130199432373, 0.790960431098938, 0.7796609997749329, 0.6779661178588867, 0.8079096078872681, 0.7175140976905823, 0.8305084705352783, 0.8022598624229431, 0.8418079018592834, 0.8474576473236084, 0.8192090392112732, 0.790960431098938, 0.8305084705352783, 0.8474576473236084, 0.8587570786476135, 0.8418079018592834, 0.903954803943634, 0.8926553726196289, 0.8418079018592834, 0.8813559412956238, 0.8983050584793091, 0.8926553726196289, 0.9152542352676392, 0.7740113139152527, 0.8813559412956238, 0.7796609997749329, 0.9604519605636597, 0.8813559412956238, 0.9322034120559692, 0.8757061958312988, 0.9491525292396545, 0.9661017060279846, 0.8926553726196289, 0.9435028433799744, 1.0, 0.8700565099716187, 0.9774011373519897, 0.9322034120559692, 0.903954803943634, 1.0, 0.994350254535675, 0.9378530979156494, 1.0, 0.9491525292396545, 1.0, 1.0, 1.0], [0.15819208323955536, 0.12429378181695938, 0.16949152946472168, 0.2711864411830902, 0.2711864411830902, 0.33898305892944336, 0.32203391194343567, 0.2711864411830902, 0.4124293923377991, 0.4237288236618042, 0.48022598028182983, 0.3785310685634613, 0.49152541160583496, 0.5197740197181702, NaN, 0.5423728823661804, 0.5706214904785156, 0.5536723136901855, 0.6327683329582214, 0.6214689016342163, 0.6666666865348816, 0.6807909607887268, 0.6440678238868713, 0.6723163723945618, 0.6440678238868713, 0.6666666865348816, 0.7288135886192322, 0.694915235042572, 0.7598869800567627, 0.7231638431549072, 0.7344632744789124, 0.7175140976905823, 0.8163841962814331, 0.7175140976905823, 0.7288135886192322, 0.7796609997749329, 0.7175140976905823, 0.8192090392112732, 0.7457627058029175, 0.7344632744789124, 0.8361582159996033, 0.7796609997749329, 0.8079096078872681, 0.7570621371269226, 0.7853107452392578, 0.8192090392112732, 0.8022598624229431, 0.8531073331832886, 0.7966101765632629, 0.8192090392112732, 0.8418079018592834, 0.8926553726196289, 0.7457627058029175, 0.887005627155304, 0.7514124512672424, 0.8983050584793091, 0.9152542352676392, 0.909604549407959, 0.9830508232116699, 0.9152542352676392, 0.9152542352676392, 0.9322034120559692, 0.9322034120559692, 0.9265536665916443, 0.8305084705352783, 0.9209039807319641, 0.9689265489578247, 0.9830508232116699, 0.9830508232116699, 0.9435028433799744, 0.9887005686759949, 0.9887005686759949, 0.9887005686759949, 1.0, 1.0, 1.0], [0.06779661029577255, 0.12994350492954254, 0.16949152946472168, 0.20903955399990082, 0.17514124512672424, 0.23728813230991364, 0.18644067645072937, 0.26553672552108765, 0.3615819215774536, 0.2768361568450928, 0.395480215549469, 0.395480215549469, 0.5141242742538452, 0.46327683329582214, 0.4576271176338196, NaN, 0.5367231369018555, 0.5480226278305054, 0.6723163723945618, 0.5480226278305054, 0.6666666865348816, 0.6271186470985413, 0.6836158037185669, 0.6666666865348816, 0.6836158037185669, 0.6610169410705566, 0.694915235042572, 0.7514124512672424, 0.604519784450531, 0.7118644118309021, 0.7627118825912476, 0.7231638431549072, 0.6666666865348816, 0.8079096078872681, 0.7683615684509277, 0.7796609997749329, 0.7570621371269226, 0.8361582159996033, 0.8418079018592834, 0.8022598624229431, 0.8757061958312988, 0.8135592937469482, 0.8587570786476135, 0.8135592937469482, 0.8474576473236084, 0.8531073331832886, 0.8644067645072937, 0.790960431098938, 0.8361582159996033, 0.9209039807319641, 0.9265536665916443, 0.9209039807319641, 0.7457627058029175, 0.9152542352676392, 0.7570621371269226, 0.9322034120559692, 0.9378530979156494, 0.9548022747039795, 0.903954803943634, 0.9717513918876648, 0.9717513918876648, 0.887005627155304, 0.9661017060279846, 0.9774011373519897, 0.8418079018592834, 0.9887005686759949, 0.9887005686759949, 0.9378530979156494, 0.9887005686759949, 0.9887005686759949, 0.9548022747039795, 1.0, 0.9717513918876648, 0.994350254535675, 0.994350254535675, 0.994350254535675], [0.0451977401971817, 0.1355932205915451, 0.11299435049295425, 0.16384181380271912, 0.16949152946472168, 0.2542372941970825, 0.15819208323955536, 0.21468926966190338, 0.22033898532390594, 0.3446327745914459, 0.3163841664791107, 0.3446327745914459, 0.38983049988746643, 0.3785310685634613, 0.42937853932380676, 0.46327683329582214, NaN, 0.604519784450531, 0.6723163723945618, 0.5706214904785156, 0.6779661178588867, 0.6384180784225464, 0.6892655491828918, 0.598870038986206, 0.6553672552108765, 0.6610169410705566, 0.7401130199432373, 0.7401130199432373, 0.598870038986206, 0.7288135886192322, 0.7627118825912476, 0.7288135886192322, 0.598870038986206, 0.7853107452392578, 0.6892655491828918, 0.7966101765632629, 0.7514124512672424, 0.8305084705352783, 0.8079096078872681, 0.7627118825912476, 0.8474576473236084, 0.887005627155304, 0.8474576473236084, 0.790960431098938, 0.8983050584793091, 0.8418079018592834, 0.887005627155304, 0.8474576473236084, 0.8248587846755981, 0.909604549407959, 0.909604549407959, 0.9265536665916443, 0.7288135886192322, 0.9265536665916443, 0.7401130199432373, 0.9152542352676392, 0.9265536665916443, 0.9435028433799744, 0.9209039807319641, 0.9661017060279846, 0.9774011373519897, 0.9322034120559692, 0.9661017060279846, 0.9887005686759949, 0.8587570786476135, 0.9887005686759949, 0.9830508232116699, 0.9604519605636597, 0.9887005686759949, 1.0, 0.9661017060279846, 1.0, 0.9717513918876648, 1.0, 1.0, 1.0], [0.050847455859184265, 0.0903954803943634, 0.14689265191555023, 0.10734463483095169, 0.14124293625354767, 0.15819208323955536, 0.16949152946472168, 0.1977401077747345, 0.20903955399990082, 0.29378530383110046, 0.2429378479719162, 0.3672316372394562, 0.37288135290145874, 0.3785310685634613, 0.44632768630981445, 0.451977401971817, 0.395480215549469, NaN, 0.604519784450531, 0.5480226278305054, 0.6158192157745361, 0.5028248429298401, 0.604519784450531, 0.5593220591545105, 0.7457627058029175, 0.6214689016342163, 0.6553672552108765, 0.7853107452392578, 0.5480226278305054, 0.7344632744789124, 0.8135592937469482, 0.7231638431549072, 0.5310734510421753, 0.8418079018592834, 0.6666666865348816, 0.7457627058029175, 0.790960431098938, 0.8418079018592834, 0.7401130199432373, 0.7966101765632629, 0.7288135886192322, 0.8361582159996033, 0.8474576473236084, 0.8135592937469482, 0.8474576473236084, 0.8813559412956238, 0.9152542352676392, 0.8192090392112732, 0.8474576473236084, 0.909604549407959, 0.9209039807319641, 0.9548022747039795, 0.7457627058029175, 0.8926553726196289, 0.7570621371269226, 0.903954803943634, 0.8983050584793091, 0.9604519605636597, 0.8983050584793091, 0.9774011373519897, 0.994350254535675, 0.909604549407959, 0.9548022747039795, 0.994350254535675, 0.8474576473236084, 1.0, 0.9717513918876648, 0.9209039807319641, 0.9887005686759949, 1.0, 0.9548022747039795, 1.0, 0.9661017060279846, 1.0, 1.0, 1.0], [0.0451977401971817, 0.07344632595777512, 0.12994350492954254, 0.0451977401971817, 0.10169491171836853, 0.10734463483095169, 0.16949152946472168, 0.19209039211273193, 0.1977401077747345, 0.22033898532390594, 0.2259887009859085, 0.32203391194343567, 0.299435019493103, 0.3050847351551056, 0.3672316372394562, 0.32768362760543823, 0.32768362760543823, 0.395480215549469, NaN, 0.5028248429298401, 0.48022598028182983, 0.48022598028182983, 0.6327683329582214, 0.4689265489578247, 0.6214689016342163, 0.5536723136901855, 0.5310734510421753, 0.7457627058029175, 0.508474588394165, 0.6892655491828918, 0.7514124512672424, 0.790960431098938, 0.5141242742538452, 0.7514124512672424, 0.5762711763381958, 0.694915235042572, 0.7457627058029175, 0.7683615684509277, 0.6158192157745361, 0.8135592937469482, 0.694915235042572, 0.7683615684509277, 0.8135592937469482, 0.7966101765632629, 0.8079096078872681, 0.9209039807319641, 0.7853107452392578, 0.8361582159996033, 0.8418079018592834, 0.8079096078872681, 0.8926553726196289, 0.9435028433799744, 0.7457627058029175, 0.8983050584793091, 0.7457627058029175, 0.9209039807319641, 0.903954803943634, 0.9604519605636597, 0.8644067645072937, 0.994350254535675, 0.9209039807319641, 0.8983050584793091, 0.9774011373519897, 0.9322034120559692, 0.8361582159996033, 0.9604519605636597, 0.9830508232116699, 0.9322034120559692, 1.0, 0.994350254535675, 0.9604519605636597, 1.0, 0.9661017060279846, 1.0, 1.0, 1.0], [0.12994350492954254, 0.09604519605636597, 0.07909604161977768, 0.23163841664791107, 0.20903955399990082, 0.1807909607887268, 0.16949152946472168, 0.17514124512672424, 0.2768361568450928, 0.33898305892944336, 0.32768362760543823, 0.2881355881690979, 0.38983049988746643, 0.38983049988746643, 0.3785310685634613, 0.451977401971817, 0.42937853932380676, 0.451977401971817, 0.4971751272678375, NaN, 0.49152541160583496, 0.598870038986206, 0.5536723136901855, 0.6271186470985413, 0.4971751272678375, 0.5367231369018555, 0.5593220591545105, 0.5423728823661804, 0.5649717450141907, 0.5819209218025208, 0.6158192157745361, 0.5762711763381958, 0.6327683329582214, 0.6158192157745361, 0.700564980506897, 0.7118644118309021, 0.6158192157745361, 0.7231638431549072, 0.700564980506897, 0.6553672552108765, 0.8926553726196289, 0.7457627058029175, 0.7853107452392578, 0.700564980506897, 0.7740113139152527, 0.7288135886192322, 0.8361582159996033, 0.8135592937469482, 0.7570621371269226, 0.8644067645072937, 0.9209039807319641, 0.8531073331832886, 0.7231638431549072, 0.8305084705352783, 0.7344632744789124, 0.8248587846755981, 0.8305084705352783, 0.8531073331832886, 0.8926553726196289, 0.9435028433799744, 0.9548022747039795, 0.9435028433799744, 0.909604549407959, 0.9887005686759949, 0.8192090392112732, 0.9604519605636597, 0.9491525292396545, 0.9604519605636597, 0.9717513918876648, 1.0, 0.9774011373519897, 0.994350254535675, 0.9887005686759949, 1.0, 1.0, 1.0], [0.0451977401971817, 0.11299435049295425, 0.11299435049295425, 0.14689265191555023, 0.10734463483095169, 0.15819208323955536, 0.11864406615495682, 0.20338982343673706, 0.1355932205915451, 0.1807909607887268, 0.19209039211273193, 0.32203391194343567, 0.24858756363391876, 0.32203391194343567, 0.3333333432674408, 0.3333333432674408, 0.32203391194343567, 0.38418078422546387, 0.5197740197181702, 0.508474588394165, NaN, 0.4576271176338196, 0.5480226278305054, 0.47457626461982727, 0.5762711763381958, 0.5367231369018555, 0.5649717450141907, 0.6553672552108765, 0.4858756959438324, 0.6214689016342163, 0.6779661178588867, 0.6327683329582214, 0.48022598028182983, 0.694915235042572, 0.5593220591545105, 0.6779661178588867, 0.6779661178588867, 0.8135592937469482, 0.5536723136901855, 0.7175140976905823, 0.7514124512672424, 0.8248587846755981, 0.8361582159996033, 0.7175140976905823, 0.8361582159996033, 0.7683615684509277, 0.8305084705352783, 0.8474576473236084, 0.7570621371269226, 0.8305084705352783, 0.909604549407959, 0.9209039807319641, 0.7401130199432373, 0.9265536665916443, 0.7401130199432373, 0.8248587846755981, 0.9322034120559692, 0.9152542352676392, 0.909604549407959, 0.9661017060279846, 0.9830508232116699, 0.9378530979156494, 0.9491525292396545, 0.9717513918876648, 0.8305084705352783, 0.9774011373519897, 0.9774011373519897, 0.9548022747039795, 0.9830508232116699, 0.994350254535675, 0.9661017060279846, 0.994350254535675, 0.9887005686759949, 0.994350254535675, 0.994350254535675, 0.994350254535675], [0.06779661029577255, 0.050847455859184265, 0.20903955399990082, 0.12994350492954254, 0.18644067645072937, 0.14124293625354767, 0.17514124512672424, 0.20903955399990082, 0.2598870098590851, 0.24858756363391876, 0.2542372941970825, 0.3502824902534485, 0.3333333432674408, 0.32203391194343567, 0.3192090392112732, 0.37288135290145874, 0.3615819215774536, 0.4971751272678375, 0.5197740197181702, 0.40112993121147156, 0.5423728823661804, NaN, 0.5423728823661804, 0.508474588394165, 0.5593220591545105, 0.5536723136901855, 0.4971751272678375, 0.5932203531265259, 0.4378530979156494, 0.6384180784225464, 0.6497175097465515, 0.6101694703102112, 0.4830508530139923, 0.6271186470985413, 0.6214689016342163, 0.6158192157745361, 0.6610169410705566, 0.6836158037185669, 0.7062146663665771, 0.694915235042572, 0.7401130199432373, 0.7062146663665771, 0.7966101765632629, 0.8079096078872681, 0.7175140976905823, 0.7401130199432373, 0.909604549407959, 0.8192090392112732, 0.8587570786476135, 0.909604549407959, 0.8757061958312988, 0.8474576473236084, 0.8587570786476135, 0.7740113139152527, 0.8813559412956238, 0.8587570786476135, 0.8248587846755981, 0.887005627155304, 0.8587570786476135, 0.909604549407959, 0.9265536665916443, 0.9152542352676392, 0.8983050584793091, 0.9378530979156494, 0.8983050584793091, 0.9435028433799744, 0.9237288236618042, 0.9435028433799744, 0.9887005686759949, 0.9265536665916443, 0.9661017060279846, 0.9887005686759949, 0.9830508232116699, 0.9887005686759949, 0.9887005686759949, 1.0], [0.028248587623238564, 0.03954802080988884, 0.12429378181695938, 0.050847455859184265, 0.10734463483095169, 0.07909604161977768, 0.14124293625354767, 0.14124293625354767, 0.17514124512672424, 0.2259887009859085, 0.20338982343673706, 0.2881355881690979, 0.28248587250709534, 0.24858756363391876, 0.35593220591545105, 0.3163841664791107, 0.31073445081710815, 0.395480215549469, 0.3672316372394562, 0.44632768630981445, 0.451977401971817, 0.4576271176338196, NaN, 0.4576271176338196, 0.5536723136901855, 0.5536723136901855, 0.49152541160583496, 0.6610169410705566, 0.48022598028182983, 0.6440678238868713, 0.7231638431549072, 0.6779661178588867, 0.5141242742538452, 0.7118644118309021, 0.5649717450141907, 0.6553672552108765, 0.6892655491828918, 0.7288135886192322, 0.598870038986206, 0.7683615684509277, 0.6779661178588867, 0.7514124512672424, 0.7853107452392578, 0.7457627058029175, 0.7627118825912476, 0.8983050584793091, 0.7627118825912476, 0.8305084705352783, 0.8248587846755981, 0.790960431098938, 0.8757061958312988, 0.9152542352676392, 0.7401130199432373, 0.8644067645072937, 0.7514124512672424, 0.903954803943634, 0.8926553726196289, 0.9491525292396545, 0.8418079018592834, 0.994350254535675, 0.9209039807319641, 0.8926553726196289, 0.9661017060279846, 0.9265536665916443, 0.8248587846755981, 0.9491525292396545, 0.9717513918876648, 0.9209039807319641, 0.994350254535675, 0.9830508232116699, 0.9548022747039795, 1.0, 0.9604519605636597, 1.0, 1.0, 1.0], [0.05649717524647713, 0.07909604161977768, 0.1355932205915451, 0.1355932205915451, 0.10734463483095169, 0.09604519605636597, 0.11299435049295425, 0.11299435049295425, 0.23728813230991364, 0.2598870098590851, 0.299435019493103, 0.2598870098590851, 0.41807910799980164, 0.33898305892944336, 0.32768362760543823, 0.3333333432674408, 0.40112993121147156, 0.4406779706478119, 0.5310734510421753, 0.37288135290145874, 0.5254237055778503, 0.49152541160583496, 0.5423728823661804, NaN, 0.5028248429298401, 0.5367231369018555, 0.5932203531265259, 0.5875706076622009, 0.49152541160583496, 0.598870038986206, 0.6440678238868713, 0.6271186470985413, 0.5141242742538452, 0.6101694703102112, 0.8361582159996033, 0.694915235042572, 0.6271186470985413, 0.6892655491828918, 0.8813559412956238, 0.6836158037185669, 0.7966101765632629, 0.6666666865348816, 0.7118644118309021, 0.6666666865348816, 0.6892655491828918, 0.7853107452392578, 0.8305084705352783, 0.7457627058029175, 0.7401130199432373, 0.8813559412956238, 0.8135592937469482, 0.8418079018592834, 0.7231638431549072, 0.7514124512672424, 0.7344632744789124, 0.8079096078872681, 0.7627118825912476, 0.8418079018592834, 0.7683615684509277, 0.903954803943634, 0.9265536665916443, 0.8361582159996033, 0.8644067645072937, 0.9604519605636597, 0.8305084705352783, 0.9774011373519897, 0.9209039807319641, 0.887005627155304, 0.9717513918876648, 0.9548022747039795, 0.9209039807319641, 0.9887005686759949, 0.9322034120559692, 1.0, 1.0, 1.0], [0.033898305147886276, 0.0451977401971817, 0.0903954803943634, 0.050847455859184265, 0.10734463483095169, 0.11299435049295425, 0.14124293625354767, 0.14124293625354767, 0.16384181380271912, 0.21468926966190338, 0.1977401077747345, 0.299435019493103, 0.299435019493103, 0.3163841664791107, 0.35593220591545105, 0.3163841664791107, 0.3446327745914459, 0.2542372941970825, 0.3785310685634613, 0.5028248429298401, 0.4237288236618042, 0.4406779706478119, 0.44632768630981445, 0.4971751272678375, NaN, 0.4124293923377991, 0.5028248429298401, 0.598870038986206, 0.49152541160583496, 0.5649717450141907, 0.6666666865348816, 0.6836158037185669, 0.5197740197181702, 0.790960431098938, 0.5819209218025208, 0.5819209218025208, 0.7118644118309021, 0.6384180784225464, 0.6666666865348816, 0.790960431098938, 0.700564980506897, 0.7683615684509277, 0.7570621371269226, 0.7627118825912476, 0.7683615684509277, 0.8700565099716187, 0.6892655491828918, 0.7401130199432373, 0.7966101765632629, 0.7288135886192322, 0.8474576473236084, 0.9774011373519897, 0.7175140976905823, 0.8644067645072937, 0.7231638431549072, 0.9265536665916443, 0.8700565099716187, 0.9661017060279846, 0.8587570786476135, 0.9717513918876648, 0.9661017060279846, 0.8587570786476135, 0.9661017060279846, 0.8926553726196289, 0.8079096078872681, 0.9887005686759949, 0.9661017060279846, 0.903954803943634, 1.0, 0.9887005686759949, 0.9322034120559692, 1.0, 0.9604519605636597, 1.0, 1.0, 1.0], [0.08474576473236084, 0.11864406615495682, 0.15819208323955536, 0.14124293625354767, 0.11299435049295425, 0.19209039211273193, 0.23163841664791107, 0.23163841664791107, 0.2542372941970825, 0.2598870098590851, 0.2542372941970825, 0.33898305892944336, 0.29378530383110046, 0.3615819215774536, 0.3333333432674408, 0.33898305892944336, 0.33898305892944336, 0.3785310685634613, 0.44632768630981445, 0.46327683329582214, 0.46327683329582214, 0.44632768630981445, 0.44632768630981445, 0.46327683329582214, 0.5875706076622009, NaN, 0.4971751272678375, 0.5932203531265259, 0.5536723136901855, 0.6384180784225464, 0.598870038986206, 0.6271186470985413, 0.5819209218025208, 0.6327683329582214, 0.5536723136901855, 0.5536723136901855, 0.6892655491828918, 0.6384180784225464, 0.5367231369018555, 0.6892655491828918, 0.6497175097465515, 0.7740113139152527, 0.7401130199432373, 0.7175140976905823, 0.7740113139152527, 0.7457627058029175, 0.700564980506897, 0.8418079018592834, 0.7627118825912476, 0.6892655491828918, 0.8361582159996033, 0.8531073331832886, 0.7288135886192322, 0.8757061958312988, 0.7401130199432373, 0.8361582159996033, 0.8757061958312988, 0.8531073331832886, 0.8757061958312988, 0.8926553726196289, 0.8983050584793091, 0.8700565099716187, 0.8644067645072937, 0.8813559412956238, 0.8474576473236084, 0.8983050584793091, 0.9435028433799744, 0.9491525292396545, 0.9830508232116699, 0.9435028433799744, 0.9830508232116699, 0.9887005686759949, 0.9830508232116699, 0.994350254535675, 0.994350254535675, 1.0], [0.06214689090847969, 0.08474576473236084, 0.10734463483095169, 0.09604519605636597, 0.12429378181695938, 0.12429378181695938, 0.1525423675775528, 0.18644067645072937, 0.1355932205915451, 0.12429378181695938, 0.16949152946472168, 0.2429378479719162, 0.23163841664791107, 0.2542372941970825, 0.2711864411830902, 0.3050847351551056, 0.2598870098590851, 0.3446327745914459, 0.4689265489578247, 0.4406779706478119, 0.4350282549858093, 0.5028248429298401, 0.508474588394165, 0.4067796468734741, 0.4971751272678375, 0.5028248429298401, NaN, 0.5819209218025208, 0.4689265489578247, 0.6158192157745361, 0.6553672552108765, 0.5932203531265259, 0.508474588394165, 0.6384180784225464, 0.4971751272678375, 0.6666666865348816, 0.6497175097465515, 0.7231638431549072, 0.5536723136901855, 0.6892655491828918, 0.5932203531265259, 0.7062146663665771, 0.7683615684509277, 0.7570621371269226, 0.7344632744789124, 0.7288135886192322, 0.7175140976905823, 0.8700565099716187, 0.7683615684509277, 0.7457627058029175, 0.8079096078872681, 0.8531073331832886, 0.7062146663665771, 0.7966101765632629, 0.7118644118309021, 0.790960431098938, 0.8079096078872681, 0.8022598624229431, 0.903954803943634, 0.9322034120559692, 0.8474576473236084, 0.9661017060279846, 0.8135592937469482, 0.9209039807319641, 0.7740113139152527, 0.8813559412956238, 0.9717513918876648, 0.9604519605636597, 0.9435028433799744, 0.9661017060279846, 0.9774011373519897, 0.9774011373519897, 0.9774011373519897, 0.994350254535675, 0.994350254535675, 0.994350254535675], [0.0451977401971817, 0.06214689090847969, 0.0903954803943634, 0.07909604161977768, 0.09604519605636597, 0.07909604161977768, 0.09604519605636597, 0.12994350492954254, 0.12994350492954254, 0.14689265191555023, 0.1525423675775528, 0.22033898532390594, 0.24858756363391876, 0.2429378479719162, 0.3050847351551056, 0.24858756363391876, 0.2598870098590851, 0.21468926966190338, 0.2542372941970825, 0.4576271176338196, 0.3446327745914459, 0.4067796468734741, 0.33898305892944336, 0.4124293923377991, 0.40112993121147156, 0.4067796468734741, 0.41807910799980164, NaN, 0.44632768630981445, 0.4971751272678375, 0.6666666865348816, 0.5932203531265259, 0.47457626461982727, 0.6214689016342163, 0.5706214904785156, 0.5875706076622009, 0.7118644118309021, 0.6101694703102112, 0.5875706076622009, 0.7401130199432373, 0.6836158037185669, 0.7288135886192322, 0.7344632744789124, 0.7457627058029175, 0.7514124512672424, 0.8757061958312988, 0.6779661178588867, 0.7514124512672424, 0.7796609997749329, 0.7344632744789124, 0.8305084705352783, 0.9717513918876648, 0.7118644118309021, 0.8418079018592834, 0.7288135886192322, 0.903954803943634, 0.8644067645072937, 0.9378530979156494, 0.8248587846755981, 0.9887005686759949, 0.9322034120559692, 0.8587570786476135, 0.9604519605636597, 0.8813559412956238, 0.7966101765632629, 0.9774011373519897, 0.9774011373519897, 0.909604549407959, 0.9830508232116699, 0.9887005686759949, 0.9548022747039795, 1.0, 0.9548022747039795, 1.0, 1.0, 1.0], [0.11864406615495682, 0.06214689090847969, 0.0903954803943634, 0.14689265191555023, 0.14689265191555023, 0.1807909607887268, 0.1525423675775528, 0.09604519605636597, 0.2542372941970825, 0.2598870098590851, 0.3050847351551056, 0.17514124512672424, 0.3672316372394562, 0.32768362760543823, 0.24011299014091492, 0.395480215549469, 0.40112993121147156, 0.451977401971817, 0.49152541160583496, 0.4350282549858093, 0.5141242742538452, 0.5621469020843506, 0.5197740197181702, 0.508474588394165, 0.508474588394165, 0.44632768630981445, 0.5310734510421753, 0.5536723136901855, NaN, 0.5932203531265259, 0.5762711763381958, 0.5480226278305054, 0.5508474707603455, 0.5932203531265259, 0.5649717450141907, 0.5875706076622009, 0.5593220591545105, 0.6440678238868713, 0.6497175097465515, 0.604519784450531, 0.6836158037185669, 0.6101694703102112, 0.700564980506897, 0.7062146663665771, 0.6158192157745361, 0.6553672552108765, 0.7231638431549072, 0.7796609997749329, 0.7514124512672424, 0.7457627058029175, 0.7175140976905823, 0.7344632744789124, 0.700564980506897, 0.694915235042572, 0.700564980506897, 0.7853107452392578, 0.7344632744789124, 0.7288135886192322, 0.8192090392112732, 0.8022598624229431, 0.8079096078872681, 0.8757061958312988, 0.7740113139152527, 0.8644067645072937, 0.7514124512672424, 0.8474576473236084, 0.895480215549469, 0.8926553726196289, 0.909604549407959, 0.8983050584793091, 0.9209039807319641, 0.9491525292396545, 0.9322034120559692, 0.9887005686759949, 0.9887005686759949, 1.0], [0.050847455859184265, 0.09604519605636597, 0.12994350492954254, 0.11864406615495682, 0.1355932205915451, 0.14124293625354767, 0.1355932205915451, 0.21468926966190338, 0.15819208323955536, 0.19209039211273193, 0.17514124512672424, 0.29378530383110046, 0.23728813230991364, 0.2598870098590851, 0.2768361568450928, 0.2881355881690979, 0.2711864411830902, 0.26553672552108765, 0.31073445081710815, 0.41807910799980164, 0.3785310685634613, 0.3615819215774536, 0.35593220591545105, 0.40112993121147156, 0.4350282549858093, 0.3615819215774536, 0.38418078422546387, 0.5028248429298401, 0.4067796468734741, NaN, 0.5141242742538452, 0.5480226278305054, 0.4406779706478119, 0.5197740197181702, 0.48022598028182983, 0.49152541160583496, 0.5706214904785156, 0.5310734510421753, 0.5141242742538452, 0.6271186470985413, 0.6553672552108765, 0.6892655491828918, 0.8418079018592834, 0.6384180784225464, 0.7118644118309021, 0.6610169410705566, 0.6836158037185669, 0.6779661178588867, 0.6836158037185669, 0.6836158037185669, 0.8474576473236084, 0.7570621371269226, 0.7062146663665771, 0.8192090392112732, 0.7231638431549072, 0.7401130199432373, 0.8531073331832886, 0.8361582159996033, 0.8700565099716187, 0.887005627155304, 0.909604549407959, 0.903954803943634, 0.8700565099716187, 0.887005627155304, 0.7740113139152527, 0.9209039807319641, 0.9548022747039795, 0.9435028433799744, 0.9435028433799744, 0.9774011373519897, 0.9717513918876648, 1.0, 0.9774011373519897, 0.994350254535675, 0.994350254535675, 1.0], [0.033898305147886276, 0.050847455859184265, 0.09604519605636597, 0.06214689090847969, 0.10734463483095169, 0.07344632595777512, 0.07909604161977768, 0.11864406615495682, 0.09604519605636597, 0.12994350492954254, 0.12994350492954254, 0.1977401077747345, 0.21468926966190338, 0.20903955399990082, 0.26553672552108765, 0.23728813230991364, 0.23728813230991364, 0.18644067645072937, 0.24858756363391876, 0.38418078422546387, 0.32203391194343567, 0.3502824902534485, 0.2768361568450928, 0.35593220591545105, 0.3333333432674408, 0.40112993121147156, 0.3446327745914459, 0.3333333432674408, 0.4237288236618042, 0.4858756959438324, NaN, 0.5423728823661804, 0.4576271176338196, 0.5649717450141907, 0.5197740197181702, 0.5367231369018555, 0.6214689016342163, 0.5480226278305054, 0.5367231369018555, 0.6610169410705566, 0.6497175097465515, 0.694915235042572, 0.6892655491828918, 0.694915235042572, 0.7118644118309021, 0.8248587846755981, 0.6723163723945618, 0.7401130199432373, 0.7740113139152527, 0.7231638431549072, 0.7966101765632629, 0.9209039807319641, 0.7118644118309021, 0.7853107452392578, 0.7118644118309021, 0.8700565099716187, 0.7966101765632629, 0.9265536665916443, 0.8079096078872681, 0.9604519605636597, 0.9209039807319641, 0.8644067645072937, 0.9209039807319641, 0.8926553726196289, 0.7966101765632629, 0.9604519605636597, 0.9548022747039795, 0.8926553726196289, 0.9830508232116699, 0.9774011373519897, 0.9435028433799744, 1.0, 0.9548022747039795, 1.0, 1.0, 1.0], [0.03954802080988884, 0.03954802080988884, 0.10734463483095169, 0.050847455859184265, 0.09604519605636597, 0.06214689090847969, 0.12429378181695938, 0.09604519605636597, 0.11864406615495682, 0.16384181380271912, 0.15819208323955536, 0.2429378479719162, 0.23728813230991364, 0.22033898532390594, 0.28248587250709534, 0.2768361568450928, 0.2711864411830902, 0.2768361568450928, 0.20903955399990082, 0.4237288236618042, 0.3672316372394562, 0.38983049988746643, 0.32203391194343567, 0.37288135290145874, 0.3163841664791107, 0.37288135290145874, 0.4067796468734741, 0.4067796468734741, 0.451977401971817, 0.451977401971817, 0.4576271176338196, NaN, 0.4971751272678375, 0.47457626461982727, 0.49152541160583496, 0.5593220591545105, 0.6384180784225464, 0.5706214904785156, 0.5536723136901855, 0.6779661178588867, 0.6553672552108765, 0.7062146663665771, 0.6327683329582214, 0.7514124512672424, 0.7175140976905823, 0.887005627155304, 0.6384180784225464, 0.6327683329582214, 0.8022598624229431, 0.6779661178588867, 0.694915235042572, 0.8079096078872681, 0.7231638431549072, 0.8079096078872681, 0.7288135886192322, 0.8926553726196289, 0.8248587846755981, 0.9265536665916443, 0.7966101765632629, 0.8248587846755981, 0.8983050584793091, 0.7740113139152527, 0.9435028433799744, 0.8022598624229431, 0.8248587846755981, 0.9152542352676392, 0.9265536665916443, 0.8644067645072937, 1.0, 0.9830508232116699, 0.9491525292396545, 1.0, 0.9604519605636597, 1.0, 1.0, 1.0], [0.03954802080988884, 0.06214689090847969, 0.0451977401971817, 0.12994350492954254, 0.07909604161977768, 0.1355932205915451, 0.09604519605636597, 0.10734463483095169, 0.20903955399990082, 0.12994350492954254, 0.26553672552108765, 0.21468926966190338, 0.29378530383110046, 0.32203391194343567, 0.1836158186197281, 0.3333333432674408, 0.40112993121147156, 0.4689265489578247, 0.4858756959438324, 0.3672316372394562, 0.5197740197181702, 0.5169491767883301, 0.4858756959438324, 0.4858756959438324, 0.48022598028182983, 0.41807910799980164, 0.49152541160583496, 0.5254237055778503, 0.44915252923965454, 0.5593220591545105, 0.5423728823661804, 0.5028248429298401, NaN, 0.5423728823661804, 0.5480226278305054, 0.5536723136901855, 0.5367231369018555, 0.6271186470985413, 0.5875706076622009, 0.5593220591545105, 0.6271186470985413, 0.6497175097465515, 0.6836158037185669, 0.6384180784225464, 0.6553672552108765, 0.5762711763381958, 0.6836158037185669, 0.7683615684509277, 0.6892655491828918, 0.7062146663665771, 0.7231638431549072, 0.6836158037185669, 0.6723163723945618, 0.7062146663665771, 0.6723163723945618, 0.7118644118309021, 0.7514124512672424, 0.7231638431549072, 0.8418079018592834, 0.7570621371269226, 0.7231638431549072, 0.8757061958312988, 0.7627118825912476, 0.8305084705352783, 0.7175140976905823, 0.7740113139152527, 0.9067796468734741, 0.909604549407959, 0.8418079018592834, 0.8474576473236084, 0.9209039807319641, 0.903954803943634, 0.9322034120559692, 1.0, 1.0, 1.0], [0.028248587623238564, 0.033898305147886276, 0.06214689090847969, 0.050847455859184265, 0.08474576473236084, 0.07909604161977768, 0.08474576473236084, 0.11299435049295425, 0.1355932205915451, 0.12429378181695938, 0.12994350492954254, 0.23163841664791107, 0.22033898532390594, 0.19209039211273193, 0.28248587250709534, 0.19209039211273193, 0.21468926966190338, 0.15819208323955536, 0.24858756363391876, 0.38418078422546387, 0.3050847351551056, 0.37288135290145874, 0.2881355881690979, 0.38983049988746643, 0.20903955399990082, 0.3672316372394562, 0.3615819215774536, 0.3785310685634613, 0.4067796468734741, 0.48022598028182983, 0.4350282549858093, 0.5254237055778503, 0.4576271176338196, NaN, 0.5254237055778503, 0.4689265489578247, 0.5480226278305054, 0.4971751272678375, 0.5762711763381958, 0.604519784450531, 0.6327683329582214, 0.6892655491828918, 0.6666666865348816, 0.6836158037185669, 0.694915235042572, 0.7401130199432373, 0.6384180784225464, 0.6836158037185669, 0.7627118825912476, 0.6666666865348816, 0.7627118825912476, 0.8926553726196289, 0.7062146663665771, 0.7853107452392578, 0.7175140976905823, 0.8192090392112732, 0.7966101765632629, 0.8587570786476135, 0.7853107452392578, 0.9322034120559692, 0.8983050584793091, 0.8531073331832886, 0.887005627155304, 0.8531073331832886, 0.7853107452392578, 0.9435028433799744, 0.9378530979156494, 0.887005627155304, 0.9830508232116699, 0.9717513918876648, 0.9265536665916443, 1.0, 0.9322034120559692, 0.9887005686759949, 0.9887005686759949, 0.994350254535675], [0.03954802080988884, 0.05649717524647713, 0.09604519605636597, 0.09604519605636597, 0.10169491171836853, 0.0903954803943634, 0.09604519605636597, 0.07909604161977768, 0.16949152946472168, 0.1525423675775528, 0.21468926966190338, 0.17514124512672424, 0.3446327745914459, 0.28248587250709534, 0.2711864411830902, 0.23163841664791107, 0.31073445081710815, 0.3333333432674408, 0.4237288236618042, 0.299435019493103, 0.4406779706478119, 0.3785310685634613, 0.4350282549858093, 0.16384181380271912, 0.41807910799980164, 0.44632768630981445, 0.5028248429298401, 0.42937853932380676, 0.4350282549858093, 0.5197740197181702, 0.48022598028182983, 0.508474588394165, 0.451977401971817, 0.47457626461982727, NaN, 0.5819209218025208, 0.5254237055778503, 0.598870038986206, 0.6497175097465515, 0.5197740197181702, 0.6892655491828918, 0.6158192157745361, 0.604519784450531, 0.5819209218025208, 0.6271186470985413, 0.6384180784225464, 0.7288135886192322, 0.6779661178588867, 0.6158192157745361, 0.7853107452392578, 0.7231638431549072, 0.7288135886192322, 0.700564980506897, 0.6610169410705566, 0.7118644118309021, 0.7288135886192322, 0.6892655491828918, 0.7514124512672424, 0.694915235042572, 0.8192090392112732, 0.8361582159996033, 0.7796609997749329, 0.7683615684509277, 0.9265536665916443, 0.7966101765632629, 0.9152542352676392, 0.8587570786476135, 0.8531073331832886, 0.9152542352676392, 0.8926553726196289, 0.8926553726196289, 0.9774011373519897, 0.9209039807319641, 0.9661017060279846, 0.9661017060279846, 0.994350254535675], [0.03954802080988884, 0.05649717524647713, 0.0903954803943634, 0.050847455859184265, 0.10734463483095169, 0.05649717524647713, 0.11299435049295425, 0.14124293625354767, 0.09604519605636597, 0.08474576473236084, 0.11299435049295425, 0.17514124512672424, 0.1977401077747345, 0.16949152946472168, 0.22033898532390594, 0.22033898532390594, 0.20338982343673706, 0.2542372941970825, 0.3050847351551056, 0.2881355881690979, 0.32203391194343567, 0.38418078422546387, 0.3446327745914459, 0.3050847351551056, 0.41807910799980164, 0.44632768630981445, 0.3333333432674408, 0.4124293923377991, 0.4124293923377991, 0.508474588394165, 0.46327683329582214, 0.4406779706478119, 0.44632768630981445, 0.5310734510421753, 0.41807910799980164, NaN, 0.48022598028182983, 0.5932203531265259, 0.46327683329582214, 0.5932203531265259, 0.5706214904785156, 0.5875706076622009, 0.6892655491828918, 0.694915235042572, 0.6384180784225464, 0.6497175097465515, 0.6836158037185669, 0.8079096078872681, 0.7118644118309021, 0.6723163723945618, 0.7683615684509277, 0.7683615684509277, 0.6836158037185669, 0.7401130199432373, 0.6892655491828918, 0.7401130199432373, 0.7344632744789124, 0.7457627058029175, 0.790960431098938, 0.8926553726196289, 0.8079096078872681, 0.8926553726196289, 0.7740113139152527, 0.903954803943634, 0.7514124512672424, 0.8531073331832886, 0.9491525292396545, 0.9378530979156494, 0.9209039807319641, 0.9378530979156494, 0.9548022747039795, 0.9887005686759949, 0.9661017060279846, 0.9830508232116699, 0.9830508232116699, 1.0], [0.033898305147886276, 0.02259887009859085, 0.10169491171836853, 0.03954802080988884, 0.11299435049295425, 0.05649717524647713, 0.09604519605636597, 0.0903954803943634, 0.14124293625354767, 0.18644067645072937, 0.16949152946472168, 0.21468926966190338, 0.23728813230991364, 0.1977401077747345, 0.28248587250709534, 0.2429378479719162, 0.24858756363391876, 0.20903955399990082, 0.2542372941970825, 0.38418078422546387, 0.32203391194343567, 0.33898305892944336, 0.31073445081710815, 0.37288135290145874, 0.2881355881690979, 0.31073445081710815, 0.3502824902534485, 0.2881355881690979, 0.4406779706478119, 0.42937853932380676, 0.3785310685634613, 0.3615819215774536, 0.46327683329582214, 0.451977401971817, 0.47457626461982727, 0.5197740197181702, NaN, 0.508474588394165, 0.5819209218025208, 0.5932203531265259, 0.6327683329582214, 0.5932203531265259, 0.5593220591545105, 0.7344632744789124, 0.6158192157745361, 0.8305084705352783, 0.6553672552108765, 0.5875706076622009, 0.7966101765632629, 0.6327683329582214, 0.6327683329582214, 0.7231638431549072, 0.7118644118309021, 0.7457627058029175, 0.7344632744789124, 0.903954803943634, 0.7683615684509277, 0.9152542352676392, 0.7288135886192322, 0.7627118825912476, 0.8248587846755981, 0.7062146663665771, 0.909604549407959, 0.790960431098938, 0.8192090392112732, 0.8305084705352783, 0.9265536665916443, 0.8531073331832886, 0.9887005686759949, 0.9717513918876648, 0.9322034120559692, 1.0, 0.9491525292396545, 0.994350254535675, 0.994350254535675, 0.994350254535675], [0.028248587623238564, 0.02259887009859085, 0.033898305147886276, 0.02259887009859085, 0.07344632595777512, 0.033898305147886276, 0.07344632595777512, 0.09604519605636597, 0.06214689090847969, 0.07344632595777512, 0.05649717524647713, 0.14689265191555023, 0.12429378181695938, 0.15819208323955536, 0.1807909607887268, 0.16384181380271912, 0.16949152946472168, 0.15819208323955536, 0.23163841664791107, 0.2768361568450928, 0.18644067645072937, 0.3163841664791107, 0.2711864411830902, 0.31073445081710815, 0.3615819215774536, 0.3615819215774536, 0.2768361568450928, 0.38983049988746643, 0.35593220591545105, 0.4689265489578247, 0.451977401971817, 0.42937853932380676, 0.37288135290145874, 0.5028248429298401, 0.40112993121147156, 0.4067796468734741, 0.49152541160583496, NaN, 0.42937853932380676, 0.5536723136901855, 0.5254237055778503, 0.6271186470985413, 0.6384180784225464, 0.6271186470985413, 0.6553672552108765, 0.6101694703102112, 0.598870038986206, 0.7288135886192322, 0.6836158037185669, 0.6610169410705566, 0.7457627058029175, 0.7514124512672424, 0.6666666865348816, 0.7231638431549072, 0.6610169410705566, 0.7062146663665771, 0.7231638431549072, 0.7401130199432373, 0.8079096078872681, 0.8813559412956238, 0.8361582159996033, 0.8757061958312988, 0.7796609997749329, 0.8531073331832886, 0.7344632744789124, 0.8587570786476135, 0.9209039807319641, 0.8813559412956238, 0.9152542352676392, 0.9548022747039795, 0.9322034120559692, 0.9830508232116699, 0.9265536665916443, 1.0, 1.0, 1.0], [0.028248587623238564, 0.033898305147886276, 0.07909604161977768, 0.050847455859184265, 0.09604519605636597, 0.050847455859184265, 0.07344632595777512, 0.050847455859184265, 0.10734463483095169, 0.10734463483095169, 0.1355932205915451, 0.12994350492954254, 0.1977401077747345, 0.1525423675775528, 0.2542372941970825, 0.15819208323955536, 0.19209039211273193, 0.2598870098590851, 0.38418078422546387, 0.299435019493103, 0.44632768630981445, 0.29378530383110046, 0.40112993121147156, 0.11864406615495682, 0.3333333432674408, 0.46327683329582214, 0.44632768630981445, 0.4124293923377991, 0.3502824902534485, 0.4858756959438324, 0.46327683329582214, 0.44632768630981445, 0.4124293923377991, 0.4237288236618042, 0.3502824902534485, 0.5367231369018555, 0.41807910799980164, 0.5706214904785156, NaN, 0.451977401971817, 0.5536723136901855, 0.6271186470985413, 0.5762711763381958, 0.5367231369018555, 0.6384180784225464, 0.5480226278305054, 0.700564980506897, 0.6553672552108765, 0.604519784450531, 0.7740113139152527, 0.7118644118309021, 0.6836158037185669, 0.694915235042572, 0.6214689016342163, 0.7062146663665771, 0.7118644118309021, 0.6271186470985413, 0.7062146663665771, 0.6440678238868713, 0.7966101765632629, 0.8022598624229431, 0.7627118825912476, 0.7231638431549072, 0.8926553726196289, 0.790960431098938, 0.8983050584793091, 0.8135592937469482, 0.7966101765632629, 0.909604549407959, 0.8813559412956238, 0.8474576473236084, 0.9830508232116699, 0.8587570786476135, 0.9774011373519897, 0.9774011373519897, 0.994350254535675], [0.03954802080988884, 0.016949152573943138, 0.09604519605636597, 0.028248587623238564, 0.10734463483095169, 0.0451977401971817, 0.07344632595777512, 0.07344632595777512, 0.11864406615495682, 0.18644067645072937, 0.14689265191555023, 0.1807909607887268, 0.2259887009859085, 0.1807909607887268, 0.26553672552108765, 0.1977401077747345, 0.23728813230991364, 0.20338982343673706, 0.18644067645072937, 0.3446327745914459, 0.28248587250709534, 0.3050847351551056, 0.23163841664791107, 0.3163841664791107, 0.20903955399990082, 0.31073445081710815, 0.31073445081710815, 0.2598870098590851, 0.395480215549469, 0.37288135290145874, 0.33898305892944336, 0.32203391194343567, 0.4406779706478119, 0.395480215549469, 0.48022598028182983, 0.4067796468734741, 0.4067796468734741, 0.44632768630981445, 0.5480226278305054, NaN, 0.6101694703102112, 0.5762711763381958, 0.5423728823661804, 0.6666666865348816, 0.5932203531265259, 0.7514124512672424, 0.6101694703102112, 0.5480226278305054, 0.7514124512672424, 0.598870038986206, 0.6101694703102112, 0.6553672552108765, 0.7175140976905823, 0.694915235042572, 0.7231638431549072, 0.8248587846755981, 0.7344632744789124, 0.8700565099716187, 0.7175140976905823, 0.7231638431549072, 0.790960431098938, 0.694915235042572, 0.8757061958312988, 0.7740113139152527, 0.790960431098938, 0.8135592937469482, 0.8587570786476135, 0.8418079018592834, 0.9830508232116699, 0.9265536665916443, 0.903954803943634, 1.0, 0.9265536665916443, 0.9887005686759949, 0.9830508232116699, 0.9887005686759949], [0.011299435049295425, 0.033898305147886276, 0.03954802080988884, 0.07909604161977768, 0.06779661029577255, 0.033898305147886276, 0.028248587623238564, 0.06214689090847969, 0.06214689090847969, 0.016949152573943138, 0.07909604161977768, 0.10734463483095169, 0.14689265191555023, 0.20903955399990082, 0.16384181380271912, 0.12429378181695938, 0.1525423675775528, 0.2711864411830902, 0.3050847351551056, 0.10734463483095169, 0.24858756363391876, 0.2598870098590851, 0.32203391194343567, 0.20338982343673706, 0.299435019493103, 0.3502824902534485, 0.4067796468734741, 0.3163841664791107, 0.3163841664791107, 0.3446327745914459, 0.3502824902534485, 0.3446327745914459, 0.37288135290145874, 0.3672316372394562, 0.31073445081710815, 0.42937853932380676, 0.3672316372394562, 0.47457626461982727, 0.44632768630981445, 0.38983049988746643, NaN, 0.6158192157745361, 0.5028248429298401, 0.5028248429298401, 0.6271186470985413, 0.4858756959438324, 0.598870038986206, 0.6158192157745361, 0.5423728823661804, 0.6779661178588867, 0.6666666865348816, 0.6101694703102112, 0.6440678238868713, 0.700564980506897, 0.6553672552108765, 0.6101694703102112, 0.7175140976905823, 0.604519784450531, 0.7231638431549072, 0.7457627058029175, 0.790960431098938, 0.790960431098938, 0.6836158037185669, 0.8926553726196289, 0.7796609997749329, 0.8474576473236084, 0.8361582159996033, 0.8926553726196289, 0.8361582159996033, 0.9265536665916443, 0.9491525292396545, 0.9548022747039795, 0.9661017060279846, 0.9661017060279846, 0.9661017060279846, 1.0], [0.033898305147886276, 0.0451977401971817, 0.06779661029577255, 0.06214689090847969, 0.10169491171836853, 0.07909604161977768, 0.07909604161977768, 0.14689265191555023, 0.06214689090847969, 0.07344632595777512, 0.07344632595777512, 0.17514124512672424, 0.06779661029577255, 0.16949152946472168, 0.22033898532390594, 0.18644067645072937, 0.11299435049295425, 0.16384181380271912, 0.23163841664791107, 0.2542372941970825, 0.17514124512672424, 0.29378530383110046, 0.24858756363391876, 0.3333333432674408, 0.23163841664791107, 0.2259887009859085, 0.29378530383110046, 0.2711864411830902, 0.38983049988746643, 0.31073445081710815, 0.3050847351551056, 0.29378530383110046, 0.3502824902534485, 0.31073445081710815, 0.38418078422546387, 0.4124293923377991, 0.4067796468734741, 0.37288135290145874, 0.37288135290145874, 0.4237288236618042, 0.38418078422546387, NaN, 0.4971751272678375, 0.5536723136901855, 0.6525423526763916, 0.5141242742538452, 0.4237288236618042, 0.5423728823661804, 0.5932203531265259, 0.451977401971817, 0.5141242742538452, 0.46327683329582214, 0.6101694703102112, 0.8079096078872681, 0.6158192157745361, 0.6553672552108765, 0.8644067645072937, 0.6497175097465515, 0.9435028433799744, 0.5819209218025208, 0.7288135886192322, 0.7570621371269226, 0.7118644118309021, 0.6384180784225464, 0.7570621371269226, 0.7175140976905823, 0.9548022747039795, 0.9661017060279846, 0.8192090392112732, 0.9209039807319641, 0.9887005686759949, 0.994350254535675, 0.9887005686759949, 1.0, 1.0, 1.0], [0.011299435049295425, 0.050847455859184265, 0.07909604161977768, 0.05649717524647713, 0.07909604161977768, 0.06779661029577255, 0.0451977401971817, 0.0903954803943634, 0.05649717524647713, 0.07344632595777512, 0.06779661029577255, 0.18644067645072937, 0.11864406615495682, 0.1525423675775528, 0.19209039211273193, 0.14124293625354767, 0.1525423675775528, 0.1525423675775528, 0.18644067645072937, 0.21468926966190338, 0.16384181380271912, 0.20338982343673706, 0.21468926966190338, 0.2881355881690979, 0.2429378479719162, 0.2598870098590851, 0.23163841664791107, 0.26553672552108765, 0.299435019493103, 0.15819208323955536, 0.31073445081710815, 0.3672316372394562, 0.3163841664791107, 0.3333333432674408, 0.395480215549469, 0.31073445081710815, 0.4406779706478119, 0.3615819215774536, 0.4237288236618042, 0.4576271176338196, 0.4971751272678375, 0.5028248429298401, NaN, 0.5480226278305054, 0.5197740197181702, 0.5028248429298401, 0.5197740197181702, 0.5141242742538452, 0.5762711763381958, 0.5706214904785156, 0.7062146663665771, 0.5762711763381958, 0.6723163723945618, 0.6892655491828918, 0.6779661178588867, 0.6327683329582214, 0.6892655491828918, 0.6497175097465515, 0.7231638431549072, 0.7118644118309021, 0.8135592937469482, 0.8079096078872681, 0.7175140976905823, 0.7627118825912476, 0.7231638431549072, 0.8587570786476135, 0.887005627155304, 0.903954803943634, 0.8531073331832886, 0.9322034120559692, 0.9435028433799744, 0.9548022747039795, 0.9548022747039795, 0.9774011373519897, 0.9774011373519897, 1.0], [0.033898305147886276, 0.016949152573943138, 0.07344632595777512, 0.02259887009859085, 0.06214689090847969, 0.033898305147886276, 0.0451977401971817, 0.05649717524647713, 0.11299435049295425, 0.14689265191555023, 0.1355932205915451, 0.14689265191555023, 0.18644067645072937, 0.14124293625354767, 0.2429378479719162, 0.18644067645072937, 0.20903955399990082, 0.18644067645072937, 0.20338982343673706, 0.299435019493103, 0.28248587250709534, 0.19209039211273193, 0.2542372941970825, 0.3333333432674408, 0.23728813230991364, 0.28248587250709534, 0.2429378479719162, 0.2542372941970825, 0.29378530383110046, 0.3615819215774536, 0.3050847351551056, 0.24858756363391876, 0.3615819215774536, 0.3163841664791107, 0.41807910799980164, 0.3050847351551056, 0.26553672552108765, 0.37288135290145874, 0.46327683329582214, 0.3333333432674408, 0.4971751272678375, 0.44632768630981445, 0.451977401971817, NaN, 0.47457626461982727, 0.41807910799980164, 0.5649717450141907, 0.5028248429298401, 0.7570621371269226, 0.604519784450531, 0.5480226278305054, 0.5028248429298401, 0.6497175097465515, 0.5480226278305054, 0.6666666865348816, 0.8248587846755981, 0.5932203531265259, 0.7457627058029175, 0.5706214904785156, 0.6271186470985413, 0.6497175097465515, 0.604519784450531, 0.7683615684509277, 0.7175140976905823, 0.8079096078872681, 0.7288135886192322, 0.7401130199432373, 0.8248587846755981, 0.9491525292396545, 0.903954803943634, 0.8418079018592834, 0.909604549407959, 0.887005627155304, 0.9830508232116699, 0.9830508232116699, 0.994350254535675], [0.028248587623238564, 0.06214689090847969, 0.05649717524647713, 0.06214689090847969, 0.10734463483095169, 0.06779661029577255, 0.07344632595777512, 0.14124293625354767, 0.07344632595777512, 0.06214689090847969, 0.08474576473236084, 0.16949152946472168, 0.06214689090847969, 0.15819208323955536, 0.21468926966190338, 0.1525423675775528, 0.10169491171836853, 0.1525423675775528, 0.19209039211273193, 0.2259887009859085, 0.16384181380271912, 0.28248587250709534, 0.23728813230991364, 0.31073445081710815, 0.23163841664791107, 0.2259887009859085, 0.26553672552108765, 0.24858756363391876, 0.38418078422546387, 0.2881355881690979, 0.2881355881690979, 0.28248587250709534, 0.3446327745914459, 0.3050847351551056, 0.37288135290145874, 0.3615819215774536, 0.38418078422546387, 0.3446327745914459, 0.3615819215774536, 0.4067796468734741, 0.37288135290145874, 0.347457617521286, 0.48022598028182983, 0.5254237055778503, NaN, 0.49152541160583496, 0.40112993121147156, 0.5367231369018555, 0.5649717450141907, 0.4406779706478119, 0.508474588394165, 0.4406779706478119, 0.604519784450531, 0.7514124512672424, 0.6158192157745361, 0.6327683329582214, 0.8079096078872681, 0.6384180784225464, 0.887005627155304, 0.5310734510421753, 0.6836158037185669, 0.7514124512672424, 0.694915235042572, 0.6158192157745361, 0.7514124512672424, 0.7062146663665771, 0.9378530979156494, 0.9604519605636597, 0.8305084705352783, 0.903954803943634, 0.994350254535675, 0.994350254535675, 0.9887005686759949, 1.0, 1.0, 1.0], [0.005649717524647713, 0.02259887009859085, 0.050847455859184265, 0.016949152573943138, 0.06214689090847969, 0.016949152573943138, 0.06779661029577255, 0.033898305147886276, 0.08474576473236084, 0.10169491171836853, 0.08474576473236084, 0.11299435049295425, 0.17514124512672424, 0.09604519605636597, 0.1807909607887268, 0.14689265191555023, 0.15819208323955536, 0.11864406615495682, 0.07909604161977768, 0.2711864411830902, 0.23163841664791107, 0.2598870098590851, 0.10169491171836853, 0.21468926966190338, 0.12994350492954254, 0.2542372941970825, 0.2711864411830902, 0.12429378181695938, 0.3446327745914459, 0.33898305892944336, 0.17514124512672424, 0.11299435049295425, 0.4237288236618042, 0.2598870098590851, 0.3615819215774536, 0.3502824902534485, 0.16949152946472168, 0.38983049988746643, 0.451977401971817, 0.24858756363391876, 0.5141242742538452, 0.4858756959438324, 0.4971751272678375, 0.5819209218025208, 0.508474588394165, NaN, 0.5875706076622009, 0.5310734510421753, 0.6553672552108765, 0.6101694703102112, 0.5480226278305054, 0.5536723136901855, 0.6892655491828918, 0.694915235042572, 0.694915235042572, 0.7231638431549072, 0.694915235042572, 0.7796609997749329, 0.7062146663665771, 0.7062146663665771, 0.694915235042572, 0.6497175097465515, 0.8248587846755981, 0.7401130199432373, 0.7796609997749329, 0.7344632744789124, 0.8474576473236084, 0.8361582159996033, 0.9717513918876648, 0.9152542352676392, 0.887005627155304, 0.9887005686759949, 0.903954803943634, 1.0, 1.0, 1.0], [0.011299435049295425, 0.033898305147886276, 0.033898305147886276, 0.0451977401971817, 0.05649717524647713, 0.050847455859184265, 0.016949152573943138, 0.028248587623238564, 0.07909604161977768, 0.11864406615495682, 0.07909604161977768, 0.10734463483095169, 0.12429378181695938, 0.10734463483095169, 0.1977401077747345, 0.1355932205915451, 0.11299435049295425, 0.08474576473236084, 0.21468926966190338, 0.16384181380271912, 0.16949152946472168, 0.0903954803943634, 0.23728813230991364, 0.16949152946472168, 0.31073445081710815, 0.299435019493103, 0.28248587250709534, 0.32203391194343567, 0.2768361568450928, 0.3163841664791107, 0.32768362760543823, 0.3615819215774536, 0.3163841664791107, 0.3615819215774536, 0.2711864411830902, 0.3163841664791107, 0.3446327745914459, 0.40112993121147156, 0.299435019493103, 0.38983049988746643, 0.40112993121147156, 0.5762711763381958, 0.48022598028182983, 0.4350282549858093, 0.598870038986206, 0.4124293923377991, NaN, 0.508474588394165, 0.508474588394165, 0.6553672552108765, 0.6214689016342163, 0.5649717450141907, 0.6666666865348816, 0.6327683329582214, 0.6610169410705566, 0.5762711763381958, 0.6440678238868713, 0.5593220591545105, 0.6440678238868713, 0.6723163723945618, 0.7740113139152527, 0.7401130199432373, 0.6440678238868713, 0.9265536665916443, 0.790960431098938, 0.8700565099716187, 0.7683615684509277, 0.8248587846755981, 0.8531073331832886, 0.8700565099716187, 0.8700565099716187, 0.8813559412956238, 0.8813559412956238, 0.909604549407959, 0.909604549407959, 1.0], [0.016949152573943138, 0.02259887009859085, 0.0903954803943634, 0.06779661029577255, 0.08474576473236084, 0.07344632595777512, 0.12429378181695938, 0.10734463483095169, 0.10734463483095169, 0.10169491171836853, 0.09604519605636597, 0.16384181380271912, 0.12994350492954254, 0.15819208323955536, 0.14689265191555023, 0.20903955399990082, 0.1525423675775528, 0.1807909607887268, 0.16384181380271912, 0.18644067645072937, 0.1525423675775528, 0.1807909607887268, 0.16949152946472168, 0.2542372941970825, 0.2598870098590851, 0.15819208323955536, 0.12994350492954254, 0.24858756363391876, 0.22033898532390594, 0.32203391194343567, 0.2598870098590851, 0.3672316372394562, 0.23163841664791107, 0.3163841664791107, 0.32203391194343567, 0.19209039211273193, 0.4124293923377991, 0.2711864411830902, 0.3446327745914459, 0.451977401971817, 0.38418078422546387, 0.4576271176338196, 0.4858756959438324, 0.4971751272678375, 0.46327683329582214, 0.4689265489578247, 0.49152541160583496, NaN, 0.5423728823661804, 0.5028248429298401, 0.5367231369018555, 0.5197740197181702, 0.6610169410705566, 0.5254237055778503, 0.6610169410705566, 0.5536723136901855, 0.5536723136901855, 0.5932203531265259, 0.5254237055778503, 0.598870038986206, 0.6497175097465515, 0.6610169410705566, 0.598870038986206, 0.6214689016342163, 0.7062146663665771, 0.6836158037185669, 0.6610169410705566, 0.7683615684509277, 0.7457627058029175, 0.6836158037185669, 0.8361582159996033, 0.7627118825912476, 0.9378530979156494, 0.9152542352676392, 0.9152542352676392, 0.9830508232116699], [0.02259887009859085, 0.016949152573943138, 0.05649717524647713, 0.033898305147886276, 0.06779661029577255, 0.02259887009859085, 0.0451977401971817, 0.03954802080988884, 0.12429378181695938, 0.12429378181695938, 0.12429378181695938, 0.11864406615495682, 0.16949152946472168, 0.11864406615495682, 0.20338982343673706, 0.16384181380271912, 0.17514124512672424, 0.1525423675775528, 0.15819208323955536, 0.2429378479719162, 0.2429378479719162, 0.14124293625354767, 0.17514124512672424, 0.2598870098590851, 0.20338982343673706, 0.23728813230991364, 0.23163841664791107, 0.22033898532390594, 0.24858756363391876, 0.3163841664791107, 0.2259887009859085, 0.1977401077747345, 0.31073445081710815, 0.23728813230991364, 0.38418078422546387, 0.2881355881690979, 0.20338982343673706, 0.3163841664791107, 0.395480215549469, 0.24858756363391876, 0.4576271176338196, 0.4067796468734741, 0.4237288236618042, 0.2429378479719162, 0.4350282549858093, 0.3446327745914459, 0.49152541160583496, 0.4576271176338196, NaN, 0.5649717450141907, 0.5254237055778503, 0.42937853932380676, 0.6327683329582214, 0.5028248429298401, 0.6440678238868713, 0.6610169410705566, 0.5593220591545105, 0.6779661178588867, 0.5423728823661804, 0.5762711763381958, 0.5932203531265259, 0.5875706076622009, 0.7231638431549072, 0.700564980506897, 0.7966101765632629, 0.6892655491828918, 0.700564980506897, 0.8135592937469482, 0.9322034120559692, 0.8700565099716187, 0.8361582159996033, 0.903954803943634, 0.8587570786476135, 0.9774011373519897, 0.9774011373519897, 0.994350254535675], [0.005649717524647713, 0.016949152573943138, 0.028248587623238564, 0.016949152573943138, 0.07344632595777512, 0.016949152573943138, 0.016949152573943138, 0.02259887009859085, 0.050847455859184265, 0.07909604161977768, 0.06214689090847969, 0.0903954803943634, 0.0903954803943634, 0.10169491171836853, 0.1807909607887268, 0.07909604161977768, 0.0903954803943634, 0.0903954803943634, 0.19209039211273193, 0.1355932205915451, 0.16949152946472168, 0.0903954803943634, 0.20903955399990082, 0.11864406615495682, 0.2711864411830902, 0.31073445081710815, 0.2542372941970825, 0.26553672552108765, 0.2542372941970825, 0.3163841664791107, 0.2768361568450928, 0.32203391194343567, 0.29378530383110046, 0.3333333432674408, 0.21468926966190338, 0.32768362760543823, 0.3672316372394562, 0.33898305892944336, 0.2259887009859085, 0.40112993121147156, 0.32203391194343567, 0.5480226278305054, 0.42937853932380676, 0.395480215549469, 0.5593220591545105, 0.38983049988746643, 0.3446327745914459, 0.4971751272678375, 0.4350282549858093, NaN, 0.5706214904785156, 0.508474588394165, 0.6610169410705566, 0.604519784450531, 0.6610169410705566, 0.5423728823661804, 0.6158192157745361, 0.5367231369018555, 0.6214689016342163, 0.598870038986206, 0.6553672552108765, 0.700564980506897, 0.6101694703102112, 0.8474576473236084, 0.7853107452392578, 0.7740113139152527, 0.7231638431549072, 0.7966101765632629, 0.790960431098938, 0.8022598624229431, 0.8587570786476135, 0.8757061958312988, 0.8531073331832886, 0.8757061958312988, 0.8757061958312988, 1.0], [0.011299435049295425, 0.02259887009859085, 0.03954802080988884, 0.028248587623238564, 0.07344632595777512, 0.016949152573943138, 0.033898305147886276, 0.06779661029577255, 0.0451977401971817, 0.03954802080988884, 0.0451977401971817, 0.12429378181695938, 0.0451977401971817, 0.10734463483095169, 0.15819208323955536, 0.07344632595777512, 0.0903954803943634, 0.07909604161977768, 0.10734463483095169, 0.07909604161977768, 0.0903954803943634, 0.12429378181695938, 0.12429378181695938, 0.18644067645072937, 0.1525423675775528, 0.16384181380271912, 0.19209039211273193, 0.16949152946472168, 0.28248587250709534, 0.1525423675775528, 0.20338982343673706, 0.3050847351551056, 0.2768361568450928, 0.23728813230991364, 0.2768361568450928, 0.23163841664791107, 0.3672316372394562, 0.2542372941970825, 0.2881355881690979, 0.38983049988746643, 0.3333333432674408, 0.4858756959438324, 0.29378530383110046, 0.451977401971817, 0.49152541160583496, 0.451977401971817, 0.3785310685634613, 0.46327683329582214, 0.47457626461982727, 0.42937853932380676, NaN, 0.42937853932380676, 0.6440678238868713, 0.6384180784225464, 0.6440678238868713, 0.5480226278305054, 0.6779661178588867, 0.5480226278305054, 0.7401130199432373, 0.6158192157745361, 0.7175140976905823, 0.7796609997749329, 0.6440678238868713, 0.694915235042572, 0.7401130199432373, 0.7796609997749329, 0.903954803943634, 0.8983050584793091, 0.8192090392112732, 0.8983050584793091, 0.9265536665916443, 0.9491525292396545, 0.9548022747039795, 0.9491525292396545, 0.9491525292396545, 1.0], [0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.050847455859184265, 0.005649717524647713, 0.028248587623238564, 0.016949152573943138, 0.028248587623238564, 0.016949152573943138, 0.02259887009859085, 0.050847455859184265, 0.05649717524647713, 0.08474576473236084, 0.10734463483095169, 0.07909604161977768, 0.07344632595777512, 0.0451977401971817, 0.05649717524647713, 0.14689265191555023, 0.07909604161977768, 0.1525423675775528, 0.08474576473236084, 0.15819208323955536, 0.02259887009859085, 0.14689265191555023, 0.14689265191555023, 0.028248587623238564, 0.26553672552108765, 0.2429378479719162, 0.07909604161977768, 0.19209039211273193, 0.3163841664791107, 0.10734463483095169, 0.2711864411830902, 0.23163841664791107, 0.2768361568450928, 0.24858756363391876, 0.3163841664791107, 0.3446327745914459, 0.38983049988746643, 0.5367231369018555, 0.4237288236618042, 0.4971751272678375, 0.5593220591545105, 0.44632768630981445, 0.4350282549858093, 0.48022598028182983, 0.5706214904785156, 0.49152541160583496, 0.5706214904785156, NaN, 0.6836158037185669, 0.5932203531265259, 0.6836158037185669, 0.6384180784225464, 0.6553672552108765, 0.6327683329582214, 0.700564980506897, 0.7570621371269226, 0.7231638431549072, 0.7457627058029175, 0.6892655491828918, 0.7457627058029175, 0.7231638431549072, 0.790960431098938, 0.8531073331832886, 0.8192090392112732, 0.9152542352676392, 0.909604549407959, 0.8757061958312988, 0.9887005686759949, 0.8983050584793091, 0.9830508232116699, 0.9830508232116699, 0.994350254535675], [0.09604519605636597, 0.10734463483095169, 0.17514124512672424, 0.12429378181695938, 0.16384181380271912, 0.12429378181695938, 0.12994350492954254, 0.16949152946472168, 0.21468926966190338, 0.23163841664791107, 0.2259887009859085, 0.2429378479719162, 0.21468926966190338, 0.2259887009859085, 0.2542372941970825, 0.2542372941970825, 0.2711864411830902, 0.2542372941970825, 0.2542372941970825, 0.2768361568450928, 0.2598870098590851, 0.14124293625354767, 0.2598870098590851, 0.2768361568450928, 0.28248587250709534, 0.2711864411830902, 0.29378530383110046, 0.2881355881690979, 0.299435019493103, 0.29378530383110046, 0.2881355881690979, 0.2768361568450928, 0.32768362760543823, 0.29378530383110046, 0.299435019493103, 0.3163841664791107, 0.2881355881690979, 0.3333333432674408, 0.3050847351551056, 0.28248587250709534, 0.35593220591545105, 0.38983049988746643, 0.32768362760543823, 0.3502824902534485, 0.395480215549469, 0.31073445081710815, 0.3333333432674408, 0.33898305892944336, 0.3672316372394562, 0.33898305892944336, 0.35593220591545105, 0.3163841664791107, NaN, 0.4576271176338196, 0.598870038986206, 0.42937853932380676, 0.5254237055778503, 0.508474588394165, 0.5197740197181702, 0.3672316372394562, 0.4576271176338196, 0.4576271176338196, 0.5593220591545105, 0.44632768630981445, 0.9774011373519897, 0.4576271176338196, 0.5254237055778503, 0.7231638431549072, 0.598870038986206, 0.6892655491828918, 0.7966101765632629, 0.6553672552108765, 0.8361582159996033, 0.7853107452392578, 0.7853107452392578, 0.8926553726196289], [0.005649717524647713, 0.016949152573943138, 0.028248587623238564, 0.011299435049295425, 0.03954802080988884, 0.011299435049295425, 0.03954802080988884, 0.05649717524647713, 0.02259887009859085, 0.02259887009859085, 0.03954802080988884, 0.12994350492954254, 0.033898305147886276, 0.11864406615495682, 0.11299435049295425, 0.08474576473236084, 0.07344632595777512, 0.10734463483095169, 0.10169491171836853, 0.16949152946472168, 0.07344632595777512, 0.2259887009859085, 0.1355932205915451, 0.24858756363391876, 0.1355932205915451, 0.12429378181695938, 0.20338982343673706, 0.15819208323955536, 0.3050847351551056, 0.1807909607887268, 0.21468926966190338, 0.19209039211273193, 0.29378530383110046, 0.21468926966190338, 0.33898305892944336, 0.2598870098590851, 0.2542372941970825, 0.2768361568450928, 0.3785310685634613, 0.3050847351551056, 0.299435019493103, 0.19209039211273193, 0.31073445081710815, 0.451977401971817, 0.24858756363391876, 0.3050847351551056, 0.3672316372394562, 0.47457626461982727, 0.4971751272678375, 0.395480215549469, 0.3615819215774536, 0.4067796468734741, 0.5423728823661804, NaN, 0.5593220591545105, 0.5762711763381958, 0.7033898234367371, 0.5875706076622009, 0.7401130199432373, 0.5254237055778503, 0.5367231369018555, 0.6101694703102112, 0.6384180784225464, 0.5254237055778503, 0.700564980506897, 0.598870038986206, 0.9887005686759949, 0.903954803943634, 0.7853107452392578, 0.8813559412956238, 0.9378530979156494, 0.9604519605636597, 0.9604519605636597, 1.0, 1.0, 1.0], [0.0903954803943634, 0.09604519605636597, 0.1525423675775528, 0.12429378181695938, 0.14689265191555023, 0.11299435049295425, 0.12429378181695938, 0.1525423675775528, 0.21468926966190338, 0.2259887009859085, 0.22033898532390594, 0.23728813230991364, 0.22033898532390594, 0.22033898532390594, 0.24858756363391876, 0.2429378479719162, 0.2598870098590851, 0.2429378479719162, 0.2542372941970825, 0.26553672552108765, 0.2598870098590851, 0.11864406615495682, 0.24858756363391876, 0.26553672552108765, 0.2768361568450928, 0.2598870098590851, 0.2881355881690979, 0.2711864411830902, 0.299435019493103, 0.2768361568450928, 0.2881355881690979, 0.2711864411830902, 0.32768362760543823, 0.28248587250709534, 0.2881355881690979, 0.31073445081710815, 0.26553672552108765, 0.33898305892944336, 0.29378530383110046, 0.2768361568450928, 0.3446327745914459, 0.38418078422546387, 0.32203391194343567, 0.3333333432674408, 0.38418078422546387, 0.3050847351551056, 0.33898305892944336, 0.33898305892944336, 0.35593220591545105, 0.33898305892944336, 0.35593220591545105, 0.3163841664791107, 0.40112993121147156, 0.4406779706478119, NaN, 0.4350282549858093, 0.5028248429298401, 0.508474588394165, 0.49152541160583496, 0.3672316372394562, 0.4576271176338196, 0.4576271176338196, 0.5593220591545105, 0.44632768630981445, 0.9604519605636597, 0.451977401971817, 0.5028248429298401, 0.7231638431549072, 0.598870038986206, 0.6892655491828918, 0.790960431098938, 0.6553672552108765, 0.8361582159996033, 0.7853107452392578, 0.7853107452392578, 0.8700565099716187], [0.016949152573943138, 0.011299435049295425, 0.033898305147886276, 0.016949152573943138, 0.02259887009859085, 0.02259887009859085, 0.05649717524647713, 0.033898305147886276, 0.05649717524647713, 0.033898305147886276, 0.06779661029577255, 0.06779661029577255, 0.06779661029577255, 0.03954802080988884, 0.10169491171836853, 0.06779661029577255, 0.08474576473236084, 0.09604519605636597, 0.07909604161977768, 0.17514124512672424, 0.17514124512672424, 0.14124293625354767, 0.09604519605636597, 0.19209039211273193, 0.07344632595777512, 0.16384181380271912, 0.20903955399990082, 0.09604519605636597, 0.21468926966190338, 0.2598870098590851, 0.12994350492954254, 0.10734463483095169, 0.2881355881690979, 0.1807909607887268, 0.2711864411830902, 0.2598870098590851, 0.09604519605636597, 0.29378530383110046, 0.2881355881690979, 0.17514124512672424, 0.38983049988746643, 0.3446327745914459, 0.3672316372394562, 0.17514124512672424, 0.3672316372394562, 0.2768361568450928, 0.4237288236618042, 0.44632768630981445, 0.33898305892944336, 0.4576271176338196, 0.451977401971817, 0.3615819215774536, 0.5706214904785156, 0.4237288236618042, 0.5649717450141907, NaN, 0.4689265489578247, 0.5593220591545105, 0.5197740197181702, 0.5367231369018555, 0.5536723136901855, 0.5197740197181702, 0.6101694703102112, 0.6779661178588867, 0.7288135886192322, 0.6158192157745361, 0.6497175097465515, 0.8079096078872681, 0.887005627155304, 0.8248587846755981, 0.8305084705352783, 0.8813559412956238, 0.8757061958312988, 0.9830508232116699, 0.9830508232116699, 0.9887005686759949], [0.005649717524647713, 0.016949152573943138, 0.028248587623238564, 0.011299435049295425, 0.028248587623238564, 0.011299435049295425, 0.033898305147886276, 0.0451977401971817, 0.016949152573943138, 0.016949152573943138, 0.033898305147886276, 0.11299435049295425, 0.028248587623238564, 0.11864406615495682, 0.08474576473236084, 0.06214689090847969, 0.07344632595777512, 0.10169491171836853, 0.09604519605636597, 0.16949152946472168, 0.06779661029577255, 0.17514124512672424, 0.10734463483095169, 0.23728813230991364, 0.12994350492954254, 0.12429378181695938, 0.19209039211273193, 0.1355932205915451, 0.26553672552108765, 0.14689265191555023, 0.20338982343673706, 0.17514124512672424, 0.24858756363391876, 0.20338982343673706, 0.31073445081710815, 0.26553672552108765, 0.23163841664791107, 0.2768361568450928, 0.37288135290145874, 0.26553672552108765, 0.28248587250709534, 0.1355932205915451, 0.31073445081710815, 0.4067796468734741, 0.19209039211273193, 0.3050847351551056, 0.35593220591545105, 0.44632768630981445, 0.4406779706478119, 0.38418078422546387, 0.32203391194343567, 0.3446327745914459, 0.47457626461982727, 0.29661017656326294, 0.4971751272678375, 0.5310734510421753, NaN, 0.5593220591545105, 0.6214689016342163, 0.47457626461982727, 0.508474588394165, 0.5706214904785156, 0.6101694703102112, 0.508474588394165, 0.694915235042572, 0.5593220591545105, 0.790960431098938, 0.8700565099716187, 0.7627118825912476, 0.8813559412956238, 0.9378530979156494, 0.9604519605636597, 0.9491525292396545, 1.0, 1.0, 1.0], [0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.033898305147886276, 0.005649717524647713, 0.011299435049295425, 0.016949152573943138, 0.016949152573943138, 0.028248587623238564, 0.02259887009859085, 0.05649717524647713, 0.03954802080988884, 0.06779661029577255, 0.0903954803943634, 0.0451977401971817, 0.05649717524647713, 0.03954802080988884, 0.03954802080988884, 0.14689265191555023, 0.08474576473236084, 0.11299435049295425, 0.050847455859184265, 0.15819208323955536, 0.033898305147886276, 0.14689265191555023, 0.1977401077747345, 0.06214689090847969, 0.2711864411830902, 0.16384181380271912, 0.07344632595777512, 0.07344632595777512, 0.2768361568450928, 0.14124293625354767, 0.24858756363391876, 0.2542372941970825, 0.08474576473236084, 0.2598870098590851, 0.29378530383110046, 0.12994350492954254, 0.395480215549469, 0.3502824902534485, 0.3502824902534485, 0.2542372941970825, 0.3615819215774536, 0.22033898532390594, 0.4406779706478119, 0.4067796468734741, 0.32203391194343567, 0.46327683329582214, 0.451977401971817, 0.3672316372394562, 0.49152541160583496, 0.4124293923377991, 0.49152541160583496, 0.4406779706478119, 0.4406779706478119, NaN, 0.5028248429298401, 0.5536723136901855, 0.6101694703102112, 0.5932203531265259, 0.7683615684509277, 0.6101694703102112, 0.5875706076622009, 0.6440678238868713, 0.6440678238868713, 0.7175140976905823, 0.9548022747039795, 0.8248587846755981, 0.8700565099716187, 0.8531073331832886, 0.8926553726196289, 0.9774011373519897, 0.9774011373519897, 0.994350254535675], [0.02259887009859085, 0.03954802080988884, 0.03954802080988884, 0.028248587623238564, 0.050847455859184265, 0.0451977401971817, 0.06214689090847969, 0.05649717524647713, 0.0451977401971817, 0.028248587623238564, 0.050847455859184265, 0.11299435049295425, 0.050847455859184265, 0.12429378181695938, 0.016949152573943138, 0.09604519605636597, 0.07909604161977768, 0.10169491171836853, 0.1355932205915451, 0.10734463483095169, 0.0903954803943634, 0.14124293625354767, 0.15819208323955536, 0.23163841664791107, 0.14124293625354767, 0.12429378181695938, 0.09604519605636597, 0.17514124512672424, 0.1807909607887268, 0.12994350492954254, 0.19209039211273193, 0.20338982343673706, 0.15819208323955536, 0.21468926966190338, 0.3050847351551056, 0.20903955399990082, 0.2711864411830902, 0.19209039211273193, 0.35593220591545105, 0.28248587250709534, 0.2768361568450928, 0.05649717524647713, 0.2768361568450928, 0.42937853932380676, 0.11299435049295425, 0.29378530383110046, 0.35593220591545105, 0.47457626461982727, 0.4576271176338196, 0.3785310685634613, 0.2598870098590851, 0.299435019493103, 0.48022598028182983, 0.2598870098590851, 0.508474588394165, 0.48022598028182983, 0.3785310685634613, 0.4971751272678375, NaN, 0.37288135290145874, 0.4406779706478119, 0.5875706076622009, 0.508474588394165, 0.4067796468734741, 0.6723163723945618, 0.4350282549858093, 0.6892655491828918, 0.9152542352676392, 0.6440678238868713, 0.7231638431549072, 0.9548022747039795, 0.8587570786476135, 0.9774011373519897, 0.994350254535675, 0.994350254535675, 1.0], [0.0, 0.005649717524647713, 0.011299435049295425, 0.005649717524647713, 0.050847455859184265, 0.0, 0.005649717524647713, 0.02259887009859085, 0.005649717524647713, 0.005649717524647713, 0.005649717524647713, 0.05649717524647713, 0.005649717524647713, 0.050847455859184265, 0.08474576473236084, 0.028248587623238564, 0.033898305147886276, 0.02259887009859085, 0.005649717524647713, 0.05649717524647713, 0.033898305147886276, 0.0903954803943634, 0.005649717524647713, 0.09604519605636597, 0.028248587623238564, 0.10734463483095169, 0.06779661029577255, 0.011299435049295425, 0.1977401077747345, 0.11299435049295425, 0.03954802080988884, 0.17514124512672424, 0.2429378479719162, 0.06779661029577255, 0.1807909607887268, 0.10734463483095169, 0.23728813230991364, 0.11864406615495682, 0.20338982343673706, 0.2768361568450928, 0.2542372941970825, 0.41807910799980164, 0.2881355881690979, 0.37288135290145874, 0.4689265489578247, 0.29378530383110046, 0.32768362760543823, 0.40112993121147156, 0.4237288236618042, 0.40112993121147156, 0.38418078422546387, 0.2429378479719162, 0.6327683329582214, 0.47457626461982727, 0.6327683329582214, 0.46327683329582214, 0.5254237055778503, 0.44632768630981445, 0.6271186470985413, NaN, 0.5593220591545105, 0.7231638431549072, 0.5593220591545105, 0.5706214904785156, 0.6836158037185669, 0.6666666865348816, 0.8022598624229431, 0.8022598624229431, 0.7514124512672424, 0.7683615684509277, 0.8757061958312988, 0.9435028433799744, 0.8983050584793091, 0.9378530979156494, 0.9378530979156494, 0.9830508232116699], [0.0, 0.005649717524647713, 0.0, 0.016949152573943138, 0.050847455859184265, 0.02259887009859085, 0.005649717524647713, 0.0, 0.0, 0.0, 0.005649717524647713, 0.028248587623238564, 0.016949152573943138, 0.033898305147886276, 0.08474576473236084, 0.028248587623238564, 0.02259887009859085, 0.005649717524647713, 0.07909604161977768, 0.0451977401971817, 0.016949152573943138, 0.07344632595777512, 0.07909604161977768, 0.07344632595777512, 0.033898305147886276, 0.10169491171836853, 0.1525423675775528, 0.06779661029577255, 0.19209039211273193, 0.0903954803943634, 0.07909604161977768, 0.10169491171836853, 0.2768361568450928, 0.10169491171836853, 0.16384181380271912, 0.19209039211273193, 0.17514124512672424, 0.16384181380271912, 0.1977401077747345, 0.20903955399990082, 0.20903955399990082, 0.2711864411830902, 0.18644067645072937, 0.3502824902534485, 0.3163841664791107, 0.3050847351551056, 0.2259887009859085, 0.3502824902534485, 0.4067796468734741, 0.3446327745914459, 0.28248587250709534, 0.2768361568450928, 0.5423728823661804, 0.46327683329582214, 0.5423728823661804, 0.44632768630981445, 0.49152541160583496, 0.38983049988746643, 0.5593220591545105, 0.4406779706478119, NaN, 0.6101694703102112, 0.44632768630981445, 0.5875706076622009, 0.6610169410705566, 0.5762711763381958, 0.6836158037185669, 0.8192090392112732, 0.694915235042572, 0.9378530979156494, 0.8587570786476135, 0.9209039807319641, 0.8587570786476135, 0.9717513918876648, 0.9717513918876648, 1.0], [0.011299435049295425, 0.033898305147886276, 0.0451977401971817, 0.03954802080988884, 0.033898305147886276, 0.028248587623238564, 0.033898305147886276, 0.08474576473236084, 0.050847455859184265, 0.03954802080988884, 0.0451977401971817, 0.12429378181695938, 0.03954802080988884, 0.10734463483095169, 0.06779661029577255, 0.11299435049295425, 0.06779661029577255, 0.0903954803943634, 0.10169491171836853, 0.05649717524647713, 0.06214689090847969, 0.08474576473236084, 0.10734463483095169, 0.16384181380271912, 0.14124293625354767, 0.12994350492954254, 0.033898305147886276, 0.14124293625354767, 0.12429378181695938, 0.09604519605636597, 0.1355932205915451, 0.2259887009859085, 0.12429378181695938, 0.14689265191555023, 0.22033898532390594, 0.10734463483095169, 0.29378530383110046, 0.12429378181695938, 0.23728813230991364, 0.3050847351551056, 0.20903955399990082, 0.2429378479719162, 0.19209039211273193, 0.395480215549469, 0.24858756363391876, 0.3502824902534485, 0.2598870098590851, 0.33898305892944336, 0.4124293923377991, 0.299435019493103, 0.22033898532390594, 0.2542372941970825, 0.5423728823661804, 0.38983049988746643, 0.5423728823661804, 0.48022598028182983, 0.42937853932380676, 0.4067796468734741, 0.4124293923377991, 0.2768361568450928, 0.38983049988746643, NaN, 0.4237288236618042, 0.40112993121147156, 0.6497175097465515, 0.44632768630981445, 0.5649717450141907, 0.8418079018592834, 0.5819209218025208, 0.6553672552108765, 0.903954803943634, 0.7796609997749329, 0.9717513918876648, 0.903954803943634, 0.8926553726196289, 0.9887005686759949], [0.005649717524647713, 0.005649717524647713, 0.011299435049295425, 0.011299435049295425, 0.033898305147886276, 0.005649717524647713, 0.005649717524647713, 0.016949152573943138, 0.016949152573943138, 0.016949152573943138, 0.02259887009859085, 0.0451977401971817, 0.016949152573943138, 0.05649717524647713, 0.06779661029577255, 0.033898305147886276, 0.033898305147886276, 0.0451977401971817, 0.02259887009859085, 0.0903954803943634, 0.050847455859184265, 0.10169491171836853, 0.033898305147886276, 0.1355932205915451, 0.033898305147886276, 0.1355932205915451, 0.18644067645072937, 0.03954802080988884, 0.2259887009859085, 0.12994350492954254, 0.07909604161977768, 0.05649717524647713, 0.23728813230991364, 0.11299435049295425, 0.23163841664791107, 0.2259887009859085, 0.0903954803943634, 0.22033898532390594, 0.2768361568450928, 0.12429378181695938, 0.3163841664791107, 0.2881355881690979, 0.28248587250709534, 0.23163841664791107, 0.3050847351551056, 0.17514124512672424, 0.35593220591545105, 0.40112993121147156, 0.2768361568450928, 0.38983049988746643, 0.35593220591545105, 0.31073445081710815, 0.4406779706478119, 0.3615819215774536, 0.4406779706478119, 0.38983049988746643, 0.38983049988746643, 0.23163841664791107, 0.49152541160583496, 0.4406779706478119, 0.5536723136901855, 0.5762711763381958, NaN, 0.5706214904785156, 0.5480226278305054, 0.5932203531265259, 0.6158192157745361, 0.6610169410705566, 0.7881355881690979, 0.7401130199432373, 0.8474576473236084, 0.8079096078872681, 0.8700565099716187, 0.9717513918876648, 0.9717513918876648, 0.9887005686759949], [0.0, 0.0, 0.0, 0.0, 0.050847455859184265, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.07344632595777512, 0.02259887009859085, 0.011299435049295425, 0.005649717524647713, 0.06779661029577255, 0.011299435049295425, 0.028248587623238564, 0.06214689090847969, 0.07344632595777512, 0.03954802080988884, 0.10734463483095169, 0.11864406615495682, 0.07909604161977768, 0.11864406615495682, 0.1355932205915451, 0.11299435049295425, 0.10734463483095169, 0.1977401077747345, 0.16949152946472168, 0.14689265191555023, 0.07344632595777512, 0.09604519605636597, 0.20903955399990082, 0.14689265191555023, 0.10734463483095169, 0.2259887009859085, 0.10734463483095169, 0.3615819215774536, 0.23728813230991364, 0.28248587250709534, 0.38418078422546387, 0.2598870098590851, 0.07344632595777512, 0.3785310685634613, 0.299435019493103, 0.1525423675775528, 0.3050847351551056, 0.2542372941970825, 0.5536723136901855, 0.47457626461982727, 0.5536723136901855, 0.32203391194343567, 0.49152541160583496, 0.38983049988746643, 0.5932203531265259, 0.42937853932380676, 0.4124293923377991, 0.598870038986206, 0.42937853932380676, NaN, 0.6440678238868713, 0.48022598028182983, 0.6836158037185669, 0.7401130199432373, 0.5762711763381958, 0.5932203531265259, 0.8135592937469482, 0.8135592937469482, 0.8248587846755981, 0.8587570786476135, 0.8587570786476135, 0.9830508232116699], [0.07344632595777512, 0.0903954803943634, 0.0903954803943634, 0.09604519605636597, 0.08474576473236084, 0.07909604161977768, 0.08474576473236084, 0.10169491171836853, 0.10734463483095169, 0.12429378181695938, 0.10169491171836853, 0.1355932205915451, 0.09604519605636597, 0.12994350492954254, 0.16949152946472168, 0.15819208323955536, 0.14124293625354767, 0.1525423675775528, 0.16384181380271912, 0.1807909607887268, 0.16949152946472168, 0.10169491171836853, 0.17514124512672424, 0.16949152946472168, 0.19209039211273193, 0.1525423675775528, 0.2259887009859085, 0.20338982343673706, 0.24858756363391876, 0.2259887009859085, 0.20338982343673706, 0.17514124512672424, 0.28248587250709534, 0.21468926966190338, 0.20338982343673706, 0.24858756363391876, 0.1807909607887268, 0.26553672552108765, 0.20903955399990082, 0.20903955399990082, 0.22033898532390594, 0.2429378479719162, 0.2768361568450928, 0.19209039211273193, 0.24858756363391876, 0.22033898532390594, 0.20903955399990082, 0.29378530383110046, 0.20338982343673706, 0.21468926966190338, 0.2598870098590851, 0.2768361568450928, 0.02259887009859085, 0.299435019493103, 0.03954802080988884, 0.2711864411830902, 0.3050847351551056, 0.4124293923377991, 0.32768362760543823, 0.3163841664791107, 0.33898305892944336, 0.3502824902534485, 0.451977401971817, 0.35593220591545105, NaN, 0.3333333432674408, 0.38418078422546387, 0.6271186470985413, 0.47457626461982727, 0.6610169410705566, 0.6666666865348816, 0.5536723136901855, 0.7401130199432373, 0.7796609997749329, 0.7796609997749329, 0.790960431098938], [0.0, 0.0, 0.0, 0.011299435049295425, 0.050847455859184265, 0.011299435049295425, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.016949152573943138, 0.02259887009859085, 0.07909604161977768, 0.011299435049295425, 0.011299435049295425, 0.0, 0.03954802080988884, 0.03954802080988884, 0.02259887009859085, 0.05649717524647713, 0.050847455859184265, 0.02259887009859085, 0.011299435049295425, 0.10169491171836853, 0.11864406615495682, 0.02259887009859085, 0.1525423675775528, 0.07909604161977768, 0.03954802080988884, 0.08474576473236084, 0.2259887009859085, 0.05649717524647713, 0.08474576473236084, 0.14689265191555023, 0.16949152946472168, 0.14124293625354767, 0.10169491171836853, 0.18644067645072937, 0.1525423675775528, 0.28248587250709534, 0.14124293625354767, 0.2711864411830902, 0.29378530383110046, 0.26553672552108765, 0.12994350492954254, 0.3163841664791107, 0.31073445081710815, 0.2259887009859085, 0.22033898532390594, 0.20903955399990082, 0.5423728823661804, 0.40112993121147156, 0.5480226278305054, 0.38418078422546387, 0.4406779706478119, 0.35593220591545105, 0.5649717450141907, 0.3333333432674408, 0.4237288236618042, 0.5536723136901855, 0.4067796468734741, 0.5197740197181702, 0.6666666865348816, NaN, 0.6836158037185669, 0.7853107452392578, 0.6440678238868713, 0.7966101765632629, 0.8192090392112732, 0.887005627155304, 0.8418079018592834, 0.9435028433799744, 0.9378530979156494, 1.0], [0.005649717524647713, 0.011299435049295425, 0.005649717524647713, 0.005649717524647713, 0.033898305147886276, 0.005649717524647713, 0.0, 0.011299435049295425, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0451977401971817, 0.005649717524647713, 0.06779661029577255, 0.031073445454239845, 0.011299435049295425, 0.016949152573943138, 0.028248587623238564, 0.016949152573943138, 0.050847455859184265, 0.02259887009859085, 0.0762711837887764, 0.028248587623238564, 0.07909604161977768, 0.033898305147886276, 0.05649717524647713, 0.028248587623238564, 0.02259887009859085, 0.10451977699995041, 0.0451977401971817, 0.0451977401971817, 0.07344632595777512, 0.09322033822536469, 0.06214689090847969, 0.14124293625354767, 0.050847455859184265, 0.07344632595777512, 0.07909604161977768, 0.18644067645072937, 0.14124293625354767, 0.16384181380271912, 0.0451977401971817, 0.11299435049295425, 0.2598870098590851, 0.06214689090847969, 0.1525423675775528, 0.23163841664791107, 0.33898305892944336, 0.299435019493103, 0.2768361568450928, 0.09604519605636597, 0.14689265191555023, 0.47457626461982727, 0.011299435049295425, 0.4971751272678375, 0.3502824902534485, 0.20903955399990082, 0.35593220591545105, 0.31073445081710815, 0.1977401077747345, 0.3163841664791107, 0.4350282549858093, 0.38418078422546387, 0.3163841664791107, 0.6158192157745361, 0.3163841664791107, NaN, 0.7627118825912476, 0.5536723136901855, 0.604519784450531, 0.8418079018592834, 0.7740113139152527, 0.8983050584793091, 0.9435028433799744, 0.9435028433799744, 0.9774011373519897], [0.016949152573943138, 0.02259887009859085, 0.03954802080988884, 0.028248587623238564, 0.02259887009859085, 0.016949152573943138, 0.028248587623238564, 0.03954802080988884, 0.02259887009859085, 0.03954802080988884, 0.02259887009859085, 0.08474576473236084, 0.02259887009859085, 0.09604519605636597, 0.016949152573943138, 0.06214689090847969, 0.03954802080988884, 0.07909604161977768, 0.06779661029577255, 0.03954802080988884, 0.0451977401971817, 0.05649717524647713, 0.07909604161977768, 0.11299435049295425, 0.09604519605636597, 0.050847455859184265, 0.03954802080988884, 0.0903954803943634, 0.10734463483095169, 0.05649717524647713, 0.10734463483095169, 0.1355932205915451, 0.0903954803943634, 0.11299435049295425, 0.14689265191555023, 0.06214689090847969, 0.14689265191555023, 0.11864406615495682, 0.20338982343673706, 0.15819208323955536, 0.10734463483095169, 0.033898305147886276, 0.09604519605636597, 0.17514124512672424, 0.03954802080988884, 0.16384181380271912, 0.17514124512672424, 0.23163841664791107, 0.18644067645072937, 0.20338982343673706, 0.10169491171836853, 0.1807909607887268, 0.2768361568450928, 0.09604519605636597, 0.2768361568450928, 0.19209039211273193, 0.12994350492954254, 0.28248587250709534, 0.08474576473236084, 0.1977401077747345, 0.1807909607887268, 0.15819208323955536, 0.33898305892944336, 0.2598870098590851, 0.37288135290145874, 0.21468926966190338, 0.23728813230991364, NaN, 0.4350282549858093, 0.40112993121147156, 0.9152542352676392, 0.32768362760543823, 0.9661017060279846, 0.8474576473236084, 0.8474576473236084, 0.9661017060279846], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.016949152573943138, 0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.0, 0.028248587623238564, 0.016949152573943138, 0.011299435049295425, 0.005649717524647713, 0.028248587623238564, 0.0, 0.016949152573943138, 0.05649717524647713, 0.016949152573943138, 0.0903954803943634, 0.05649717524647713, 0.016949152573943138, 0.0, 0.15819208323955536, 0.016949152573943138, 0.08474576473236084, 0.07909604161977768, 0.011299435049295425, 0.08474576473236084, 0.0903954803943634, 0.016949152573943138, 0.16384181380271912, 0.1807909607887268, 0.14689265191555023, 0.050847455859184265, 0.16949152946472168, 0.028248587623238564, 0.14689265191555023, 0.2542372941970825, 0.06779661029577255, 0.20903955399990082, 0.1807909607887268, 0.08474576473236084, 0.40112993121147156, 0.21468926966190338, 0.40112993121147156, 0.11299435049295425, 0.23728813230991364, 0.0451977401971817, 0.35593220591545105, 0.24858756363391876, 0.3050847351551056, 0.41807910799980164, 0.2118644118309021, 0.4237288236618042, 0.5254237055778503, 0.35593220591545105, 0.44632768630981445, 0.5649717450141907, NaN, 0.508474588394165, 0.694915235042572, 0.6271186470985413, 0.8135592937469482, 0.8926553726196289, 0.8926553726196289, 0.9661017060279846], [0.0, 0.0, 0.0, 0.0, 0.050847455859184265, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.005649717524647713, 0.05649717524647713, 0.011299435049295425, 0.0, 0.0, 0.005649717524647713, 0.0, 0.005649717524647713, 0.07344632595777512, 0.016949152573943138, 0.0451977401971817, 0.011299435049295425, 0.05649717524647713, 0.033898305147886276, 0.011299435049295425, 0.10169491171836853, 0.02259887009859085, 0.02259887009859085, 0.016949152573943138, 0.1525423675775528, 0.028248587623238564, 0.10734463483095169, 0.06214689090847969, 0.028248587623238564, 0.0451977401971817, 0.11864406615495682, 0.07344632595777512, 0.07344632595777512, 0.07909604161977768, 0.06779661029577255, 0.09604519605636597, 0.09604519605636597, 0.08474576473236084, 0.12994350492954254, 0.3163841664791107, 0.12994350492954254, 0.1977401077747345, 0.10169491171836853, 0.0903954803943634, 0.31073445081710815, 0.11864406615495682, 0.31073445081710815, 0.17514124512672424, 0.11864406615495682, 0.17514124512672424, 0.2768361568450928, 0.23163841664791107, 0.06214689090847969, 0.3446327745914459, 0.2598870098590851, 0.4067796468734741, 0.33898305892944336, 0.20338982343673706, 0.395480215549469, 0.598870038986206, 0.49152541160583496, NaN, 0.8022598624229431, 0.5197740197181702, 0.8474576473236084, 0.9265536665916443, 0.9265536665916443, 0.994350254535675], [0.005649717524647713, 0.011299435049295425, 0.02259887009859085, 0.016949152573943138, 0.011299435049295425, 0.016949152573943138, 0.011299435049295425, 0.028248587623238564, 0.011299435049295425, 0.02259887009859085, 0.011299435049295425, 0.033898305147886276, 0.011299435049295425, 0.06214689090847969, 0.011299435049295425, 0.0451977401971817, 0.033898305147886276, 0.0451977401971817, 0.03954802080988884, 0.02259887009859085, 0.033898305147886276, 0.033898305147886276, 0.0451977401971817, 0.07909604161977768, 0.06779661029577255, 0.016949152573943138, 0.02259887009859085, 0.0451977401971817, 0.07909604161977768, 0.028248587623238564, 0.05649717524647713, 0.050847455859184265, 0.07909604161977768, 0.07344632595777512, 0.10734463483095169, 0.0451977401971817, 0.06779661029577255, 0.06779661029577255, 0.1525423675775528, 0.09604519605636597, 0.050847455859184265, 0.011299435049295425, 0.05649717524647713, 0.15819208323955536, 0.005649717524647713, 0.11299435049295425, 0.12994350492954254, 0.16384181380271912, 0.16384181380271912, 0.14124293625354767, 0.07344632595777512, 0.12429378181695938, 0.20338982343673706, 0.06214689090847969, 0.20903955399990082, 0.16949152946472168, 0.06214689090847969, 0.12994350492954254, 0.0451977401971817, 0.12429378181695938, 0.14124293625354767, 0.09604519605636597, 0.1525423675775528, 0.18644067645072937, 0.3333333432674408, 0.1807909607887268, 0.15819208323955536, 0.08474576473236084, 0.3050847351551056, 0.1977401077747345, NaN, 0.24858756363391876, 0.8983050584793091, 0.8248587846755981, 0.8248587846755981, 0.8926553726196289], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.011299435049295425, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.005649717524647713, 0.011299435049295425, 0.0, 0.011299435049295425, 0.0, 0.011299435049295425, 0.02259887009859085, 0.0, 0.050847455859184265, 0.0, 0.0, 0.0, 0.09604519605636597, 0.0, 0.02259887009859085, 0.011299435049295425, 0.0, 0.016949152573943138, 0.016949152573943138, 0.0, 0.0451977401971817, 0.005649717524647713, 0.0451977401971817, 0.0903954803943634, 0.005649717524647713, 0.011299435049295425, 0.11864406615495682, 0.23728813230991364, 0.09604519605636597, 0.12429378181695938, 0.050847455859184265, 0.011299435049295425, 0.3446327745914459, 0.03954802080988884, 0.3446327745914459, 0.11864406615495682, 0.03954802080988884, 0.14689265191555023, 0.14124293625354767, 0.05649717524647713, 0.07909604161977768, 0.22033898532390594, 0.19209039211273193, 0.18644067645072937, 0.44632768630981445, 0.11299435049295425, 0.2259887009859085, 0.6723163723945618, 0.37288135290145874, 0.48022598028182983, 0.7514124512672424, NaN, 0.7740113139152527, 0.7796609997749329, 0.7740113139152527, 0.8474576473236084], [0.011299435049295425, 0.02259887009859085, 0.011299435049295425, 0.02259887009859085, 0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.016949152573943138, 0.011299435049295425, 0.011299435049295425, 0.011299435049295425, 0.03954802080988884, 0.011299435049295425, 0.050847455859184265, 0.011299435049295425, 0.028248587623238564, 0.028248587623238564, 0.033898305147886276, 0.033898305147886276, 0.011299435049295425, 0.011299435049295425, 0.016949152573943138, 0.03954802080988884, 0.06779661029577255, 0.03954802080988884, 0.016949152573943138, 0.02259887009859085, 0.0451977401971817, 0.06779661029577255, 0.02259887009859085, 0.0451977401971817, 0.03954802080988884, 0.06779661029577255, 0.06779661029577255, 0.07909604161977768, 0.033898305147886276, 0.050847455859184265, 0.07344632595777512, 0.14124293625354767, 0.07344632595777512, 0.033898305147886276, 0.011299435049295425, 0.0451977401971817, 0.11299435049295425, 0.011299435049295425, 0.09604519605636597, 0.11864406615495682, 0.06214689090847969, 0.14124293625354767, 0.14689265191555023, 0.0451977401971817, 0.10169491171836853, 0.16384181380271912, 0.03954802080988884, 0.16384181380271912, 0.12429378181695938, 0.050847455859184265, 0.10734463483095169, 0.02259887009859085, 0.10169491171836853, 0.14124293625354767, 0.028248587623238564, 0.12994350492954254, 0.17514124512672424, 0.2598870098590851, 0.15819208323955536, 0.10169491171836853, 0.033898305147886276, 0.18644067645072937, 0.1525423675775528, 0.10169491171836853, 0.2259887009859085, NaN, 0.5875706076622009, 0.5932203531265259, 0.6384180784225464], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.0, 0.0, 0.005649717524647713, 0.005649717524647713, 0.0, 0.011299435049295425, 0.005649717524647713, 0.0, 0.0, 0.0, 0.011299435049295425, 0.033898305147886276, 0.016949152573943138, 0.005649717524647713, 0.0, 0.02259887009859085, 0.011299435049295425, 0.033898305147886276, 0.0, 0.02259887009859085, 0.016949152573943138, 0.0, 0.0, 0.0903954803943634, 0.08474576473236084, 0.02259887009859085, 0.12429378181695938, 0.050847455859184265, 0.016949152573943138, 0.21468926966190338, 0.0, 0.21468926966190338, 0.016949152573943138, 0.0, 0.02259887009859085, 0.005649717524647713, 0.06214689090847969, 0.028248587623238564, 0.09604519605636597, 0.028248587623238564, 0.14124293625354767, 0.22033898532390594, 0.05649717524647713, 0.05649717524647713, 0.1525423675775528, 0.10734463483095169, 0.07344632595777512, 0.17514124512672424, 0.22033898532390594, 0.4124293923377991, NaN, 0.5310734510421753, 0.7796609997749329], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.0, 0.0, 0.005649717524647713, 0.005649717524647713, 0.0, 0.011299435049295425, 0.005649717524647713, 0.0, 0.0, 0.0, 0.011299435049295425, 0.033898305147886276, 0.016949152573943138, 0.005649717524647713, 0.0, 0.02259887009859085, 0.016949152573943138, 0.033898305147886276, 0.0, 0.02259887009859085, 0.016949152573943138, 0.0, 0.0, 0.0903954803943634, 0.08474576473236084, 0.02259887009859085, 0.12429378181695938, 0.050847455859184265, 0.016949152573943138, 0.21468926966190338, 0.0, 0.21468926966190338, 0.016949152573943138, 0.0, 0.02259887009859085, 0.005649717524647713, 0.06214689090847969, 0.028248587623238564, 0.10734463483095169, 0.028248587623238564, 0.14124293625354767, 0.22033898532390594, 0.06214689090847969, 0.05649717524647713, 0.1525423675775528, 0.10734463483095169, 0.07344632595777512, 0.17514124512672424, 0.2259887009859085, 0.4067796468734741, 0.4689265489578247, NaN, 0.7683615684509277], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.005649717524647713, 0.005649717524647713, 0.0, 0.005649717524647713, 0.0, 0.005649717524647713, 0.011299435049295425, 0.0, 0.0, 0.0, 0.005649717524647713, 0.0, 0.0, 0.0, 0.016949152573943138, 0.005649717524647713, 0.0, 0.0, 0.005649717524647713, 0.10734463483095169, 0.0, 0.12994350492954254, 0.011299435049295425, 0.0, 0.005649717524647713, 0.0, 0.016949152573943138, 0.0, 0.011299435049295425, 0.011299435049295425, 0.016949152573943138, 0.20903955399990082, 0.0, 0.02259887009859085, 0.033898305147886276, 0.033898305147886276, 0.005649717524647713, 0.10734463483095169, 0.1525423675775528, 0.3615819215774536, 0.22033898532390594, 0.23163841664791107, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.934839921},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.9163088477},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8812806046},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.8763841788},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8613182672},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8588323911},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8580037653},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8461016961},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8140112985},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7946515974},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.7897551792},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7597363462},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7446327667},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.725988698},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7206026362},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7199246686},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7106591387},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.6946139384},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6586064007},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6575517898},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6465536728},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.6458003782},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6338229753},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6289265544},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6255367245},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.6254613926},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6045951027},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.595178909},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5920150675},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5731073448},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5639924685},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5633144995},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5622598865},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5465913348},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.5447834306},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5355178918},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5314500931},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5032015059},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5025988696},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4986064004},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4659887},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4543502815},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.4502448201},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4393973634},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4364218422},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4351789068},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4266666684},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.3973634617},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.3972881345},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3929190207},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3880225984},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3864406754},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.37227872},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3642561175},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.362636535},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.3377024486},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3362335208},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3215819205},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.3146515996},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3120903934},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.2932580023},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2869303196},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2836534837},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2644821089},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2607909596},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2596610169},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2145386059},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1803389832},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.177137476},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1663276817},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1259510355},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.1237664785},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0885122405},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0564218456},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0557438791},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0236534841}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_all/tasks_regression/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_all/tasks_regression/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab564c21c1bbdb232258afc211f193500544f61a
--- /dev/null
+++ b/data/imputation_yes/splits_all/tasks_regression/datasets_small/winrate_matrix.csv
@@ -0,0 +1,77 @@
+,TabFM (default),Nori-30M (default),TabPFN-3 (default),Nori (default),TabICLv2 (default),TabPFN-2.6 (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),TabDPT (tuned + ensembled),RealMLP (tuned + ensembled),TabDPT (tuned),RealTabPFN-2.5 (tuned),TabDPT (default),RealTabPFN-2.5 (default),TabPFNv2 (tuned + ensembled),RealMLP (tuned),TabDPT-Turbo (default),"AutoGluon 1.4 (best, 4h)",CatBoost (tuned + ensembled),TabSwift (default),iLTM (tuned + ensembled),Mitra (default),CatBoost (tuned),TabM (tuned + ensembled),LightGBM (tuned + ensembled),SAP-RPT-OSS (default),xRFM (tuned + ensembled),XGBoost (tuned + ensembled),TabPFNv2 (tuned),ModernNCA (tuned + ensembled),XGBoost (tuned),CatBoost (default),TabPFNv2 (default),LightGBM (tuned),TabM (tuned),xRFM (tuned),ChimeraBoost (tuned + ensembled),iLTM (tuned),TabM (default),ChimeraBoost (tuned),RealMLP (default),ExtraTrees (tuned + ensembled),ModernNCA (tuned),EBM (tuned + ensembled),ExtraTrees (tuned),ChimeraBoost (default),TorchMLP (tuned + ensembled),LimiX (default),EBM (tuned),TorchMLP (tuned),ModernNCA (default),LightGBM (default),TabSTAR (tuned + ensembled),RandomForest (tuned + ensembled),TabSTAR (tuned),EBM (default),RandomForest (tuned),PerpetualBooster (tuned + ensembled),ExtraTrees (default),XGBoost (default),FastaiMLP (tuned + ensembled),xRFM (default),PerpetualBooster (tuned),TorchMLP (default),TabSTAR (default),FastaiMLP (tuned),RandomForest (default),KNN (tuned + ensembled),PerpetualBooster (default),FastaiMLP (default),KNN (tuned),iLTM (default),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,0.5141243,0.68926555,0.7457627,0.70621467,0.76836157,0.7118644,0.740113,0.8757062,0.8757062,0.920904,0.76836157,0.8926554,0.91525424,0.8418079,0.9322034,0.9548023,0.9491525,0.9548023,0.8700565,0.9548023,0.9322034,0.9717514,0.94350284,0.9661017,0.91525424,0.9378531,0.9548023,0.88135594,0.9491525,0.9661017,0.96045196,0.96045196,0.9717514,0.96045196,0.96045196,0.9661017,0.9717514,0.9717514,0.96045196,0.98870057,0.9661017,0.98870057,0.9661017,0.9717514,0.99435025,0.98870057,0.9830508,0.97740114,0.99435025,0.98870057,1.0,0.9039548,0.99435025,0.90960455,0.9830508,0.99435025,0.98870057,0.97740114,1.0,1.0,0.98870057,0.99435025,1.0,0.92655367,1.0,0.99435025,0.9830508,1.0,1.0,0.99435025,1.0,0.98870057,1.0,1.0,1.0
+Nori-30M (default),0.4858757,,0.5480226,0.72316384,0.63276833,0.76836157,0.7344633,0.7344633,0.70621467,0.779661,0.79096043,0.8135593,0.79096043,0.8700565,0.8757062,0.8700565,0.86440676,0.90960455,0.92655367,0.9039548,0.8870056,0.9491525,0.96045196,0.920904,0.9548023,0.88135594,0.91525424,0.9378531,0.9378531,0.9039548,0.9491525,0.96045196,0.9378531,0.9661017,0.94350284,0.94350284,0.97740114,0.97740114,0.9661017,0.9830508,0.9661017,0.9548023,0.9491525,0.9830508,0.9378531,0.97740114,0.9661017,0.97740114,0.9830508,0.9830508,0.97740114,0.99435025,0.8926554,0.9830508,0.9039548,0.98870057,0.9830508,0.98870057,0.96045196,0.99435025,0.99435025,0.9661017,0.99435025,1.0,0.90960455,1.0,0.98870057,0.97740114,1.0,1.0,0.98870057,1.0,0.97740114,1.0,1.0,1.0
+TabPFN-3 (default),0.31073445,0.4519774,,0.5536723,0.55932206,0.59887004,0.6045198,0.5819209,0.70621467,0.79096043,0.7288136,0.7457627,0.8135593,0.84745765,0.8305085,0.8305085,0.8870056,0.85310733,0.8700565,0.920904,0.8870056,0.79096043,0.8757062,0.86440676,0.90960455,0.8418079,0.8926554,0.90960455,0.90960455,0.8700565,0.9039548,0.8926554,0.9548023,0.9378531,0.9039548,0.90960455,0.89830506,0.9661017,0.920904,0.9039548,0.96045196,0.9322034,0.920904,0.92655367,0.94350284,0.9491525,0.9661017,0.90960455,0.94350284,0.9717514,0.96045196,0.98870057,0.8248588,0.9717514,0.84745765,0.9661017,0.9717514,0.98870057,0.96045196,0.98870057,1.0,0.9548023,0.98870057,1.0,0.90960455,1.0,0.99435025,0.96045196,1.0,1.0,0.97740114,1.0,0.98870057,1.0,1.0,1.0
+Nori (default),0.2542373,0.27683616,0.4463277,,0.4915254,0.6158192,0.5706215,0.5706215,0.66101694,0.6666667,0.700565,0.66101694,0.7344633,0.8587571,0.7288136,0.79096043,0.8361582,0.8926554,0.9548023,0.76836157,0.85310733,0.8700565,0.9491525,0.86440676,0.9491525,0.8587571,0.9039548,0.920904,0.85310733,0.88135594,0.9378531,0.9491525,0.8700565,0.9491525,0.9039548,0.9491525,0.96045196,0.97740114,0.9491525,0.9717514,0.920904,0.9378531,0.94350284,0.97740114,0.9378531,0.9830508,0.9548023,0.9322034,0.9661017,0.9830508,0.9717514,1.0,0.8757062,0.98870057,0.8757062,0.9830508,0.98870057,0.98870057,0.9717514,0.99435025,0.9830508,0.96045196,0.98870057,1.0,0.9039548,0.98870057,0.99435025,0.9717514,1.0,1.0,0.9830508,1.0,0.97740114,1.0,1.0,1.0
+TabICLv2 (default),0.2937853,0.36723164,0.44067797,0.5084746,,0.5762712,0.49717513,0.5480226,0.7288136,0.7457627,0.72316384,0.6836158,0.7627119,0.84745765,0.7288136,0.8248588,0.8305085,0.8587571,0.89830506,0.79096043,0.8926554,0.8135593,0.8926554,0.8926554,0.8926554,0.8870056,0.8757062,0.9039548,0.85310733,0.86440676,0.8926554,0.9039548,0.920904,0.91525424,0.89830506,0.8926554,0.8870056,0.92655367,0.9039548,0.8926554,0.9322034,0.89830506,0.920904,0.9378531,0.8926554,0.9378531,0.94350284,0.91525424,0.9322034,0.92655367,0.92655367,0.9491525,0.8361582,0.96045196,0.85310733,0.97740114,0.9717514,0.9661017,0.9491525,0.9491525,0.9491525,0.9661017,0.9661017,0.9491525,0.91525424,0.9491525,0.9661017,0.97740114,1.0,0.9491525,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+TabPFN-2.6 (default),0.23163842,0.23163842,0.40112993,0.38418078,0.42372882,,0.43502825,0.5254237,0.6158192,0.6158192,0.6497175,0.6497175,0.700565,0.8361582,0.66101694,0.7627119,0.7457627,0.8418079,0.8926554,0.81920904,0.8418079,0.8587571,0.920904,0.9039548,0.8870056,0.8079096,0.8757062,0.920904,0.81920904,0.8587571,0.92655367,0.9378531,0.86440676,0.920904,0.90960455,0.94350284,0.94350284,0.9661017,0.9491525,0.9548023,0.9661017,0.920904,0.9322034,0.9661017,0.9322034,0.9830508,0.9491525,0.92655367,0.97740114,0.9830508,0.9830508,0.99435025,0.8757062,0.98870057,0.8870056,0.97740114,0.98870057,0.99435025,0.9548023,1.0,0.97740114,0.9717514,0.99435025,1.0,0.920904,0.98870057,0.99435025,0.9830508,1.0,0.99435025,0.9830508,1.0,0.98870057,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2881356,0.26553673,0.39548022,0.42937854,0.50282484,0.56497175,,0.4915254,0.66101694,0.7288136,0.6723164,0.6440678,0.7118644,0.78531075,0.6779661,0.8135593,0.8418079,0.8305085,0.8305085,0.8305085,0.88135594,0.8248588,0.8587571,0.8870056,0.8587571,0.76836157,0.84745765,0.9039548,0.84745765,0.86440676,0.920904,0.8757062,0.9039548,0.91525424,0.9039548,0.8870056,0.9039548,0.92655367,0.92655367,0.92655367,0.9717514,0.920904,0.9548023,0.9548023,0.92655367,0.9322034,0.9830508,0.8757062,0.9548023,0.9830508,0.9661017,0.9717514,0.8700565,0.96045196,0.8757062,0.94350284,0.9661017,0.98870057,0.9378531,0.99435025,0.99435025,0.9661017,0.99435025,1.0,0.91525424,1.0,1.0,0.9717514,1.0,1.0,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.259887,0.26553673,0.4180791,0.42937854,0.4519774,0.47457626,0.5084746,,0.6101695,0.6440678,0.6440678,0.8079096,0.70621467,0.85310733,0.7288136,0.7344633,0.78531075,0.80225986,0.8079096,0.8248588,0.7966102,0.79096043,0.8587571,0.8870056,0.8587571,0.76836157,0.8135593,0.8700565,0.9039548,0.78531075,0.88135594,0.9039548,0.8926554,0.8870056,0.920904,0.8587571,0.90960455,0.9039548,0.9491525,0.92655367,0.9378531,0.85310733,0.90960455,0.94350284,0.8587571,0.9661017,0.9717514,0.8926554,0.96045196,0.97740114,0.9322034,0.9830508,0.8305085,0.94350284,0.84745765,0.9661017,0.9548023,0.9830508,0.94350284,0.97740114,1.0,0.91525424,0.9830508,1.0,0.89830506,1.0,0.98870057,0.96045196,1.0,1.0,0.9717514,1.0,0.9830508,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.12429378,0.2937853,0.2937853,0.33898306,0.27118644,0.38418078,0.33898306,0.3898305,,0.56497175,0.6158192,0.55932206,0.72316384,0.68926555,0.5875706,0.6384181,0.779661,0.79096043,0.80225986,0.72316384,0.86440676,0.740113,0.8248588,0.7627119,0.8361582,0.7457627,0.86440676,0.8700565,0.7457627,0.8418079,0.9039548,0.88135594,0.79096043,0.86440676,0.8305085,0.9039548,0.8587571,0.9378531,0.8926554,0.88135594,0.9378531,0.9378531,0.94350284,0.8870056,0.92655367,0.91525424,0.920904,0.8926554,0.8757062,0.9491525,0.9548023,0.9717514,0.78531075,0.97740114,0.78531075,0.94350284,0.9830508,0.9830508,0.9548023,0.99435025,1.0,0.9491525,0.9830508,1.0,0.8926554,1.0,1.0,0.97740114,1.0,1.0,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.12429378,0.22033899,0.20903955,0.33333334,0.2542373,0.38418078,0.27118644,0.3559322,0.43502825,,0.5084746,0.47457626,0.5819209,0.6158192,0.5762712,0.72316384,0.65536726,0.70621467,0.779661,0.66101694,0.81920904,0.75141245,0.7740113,0.740113,0.78531075,0.740113,0.8757062,0.85310733,0.740113,0.8079096,0.8700565,0.8361582,0.8700565,0.8757062,0.84745765,0.91525424,0.8135593,0.92655367,0.8926554,0.8135593,0.9830508,0.92655367,0.92655367,0.85310733,0.9378531,0.89830506,0.88135594,0.89830506,0.8757062,0.920904,0.96045196,0.9830508,0.76836157,0.97740114,0.7740113,0.9661017,0.9830508,0.9717514,0.9717514,0.99435025,1.0,0.96045196,0.9830508,0.99435025,0.8757062,0.99435025,0.99435025,0.96045196,1.0,1.0,0.97740114,1.0,0.98870057,1.0,1.0,1.0
+TabDPT (tuned),0.07909604,0.20903955,0.27118644,0.29943502,0.27683616,0.3502825,0.32768363,0.3559322,0.38418078,0.4915254,,0.53107345,0.6384181,0.5819209,0.519774,0.6045198,0.6836158,0.75706214,0.7740113,0.6723164,0.8079096,0.7457627,0.7966102,0.700565,0.80225986,0.7457627,0.8305085,0.84745765,0.69491524,0.8248588,0.8700565,0.8418079,0.7344633,0.8700565,0.78531075,0.8870056,0.8305085,0.94350284,0.86440676,0.85310733,0.920904,0.92655367,0.9322034,0.86440676,0.91525424,0.91525424,0.920904,0.9039548,0.8757062,0.9378531,0.9548023,0.97740114,0.7740113,0.96045196,0.779661,0.9322034,0.9661017,0.97740114,0.9491525,0.99435025,0.99435025,0.9548023,0.97740114,1.0,0.89830506,1.0,0.98870057,0.97740114,1.0,1.0,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.23163842,0.18644068,0.2542373,0.33898306,0.31638417,0.3502825,0.3559322,0.19209039,0.44067797,0.5254237,0.46892655,,0.5762712,0.6581921,0.6214689,0.6045198,0.65536726,0.63276833,0.6779661,0.7118644,0.6779661,0.6497175,0.7118644,0.740113,0.700565,0.66101694,0.75706214,0.779661,0.8248588,0.70621467,0.80225986,0.75706214,0.78531075,0.76836157,0.8248588,0.8248588,0.78531075,0.85310733,0.8700565,0.81920904,0.8926554,0.8248588,0.8135593,0.85310733,0.8305085,0.8870056,0.8926554,0.8361582,0.88135594,0.90960455,0.8757062,0.9491525,0.75706214,0.8700565,0.7627119,0.9322034,0.8870056,0.94350284,0.8870056,0.94350284,0.9717514,0.8757062,0.9548023,0.99435025,0.86440676,1.0,0.9548023,0.91525424,1.0,0.99435025,0.9661017,1.0,0.96045196,1.0,1.0,1.0
+TabDPT (default),0.107344635,0.20903955,0.18644068,0.26553673,0.23728813,0.29943502,0.2881356,0.2937853,0.27683616,0.4180791,0.36158192,0.42372882,,0.47457626,0.5084746,0.4858757,0.6101695,0.62711865,0.700565,0.6101695,0.75141245,0.6666667,0.7175141,0.5819209,0.700565,0.70621467,0.76836157,0.75141245,0.63276833,0.7627119,0.78531075,0.7627119,0.70621467,0.779661,0.65536726,0.80225986,0.7627119,0.8757062,0.80225986,0.7740113,0.85310733,0.9322034,0.88135594,0.8135593,0.9378531,0.8248588,0.8757062,0.8700565,0.8305085,0.90960455,0.9548023,0.94350284,0.78531075,0.9661017,0.779661,0.9322034,0.9717514,0.96045196,0.9491525,0.99435025,0.9830508,0.96045196,0.9830508,0.98870057,0.9039548,0.9830508,0.99435025,0.97740114,1.0,1.0,0.98870057,1.0,0.98870057,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.084745765,0.1299435,0.15254237,0.14124294,0.15254237,0.16384181,0.21468927,0.14689265,0.31073445,0.38418078,0.4180791,0.3418079,0.5254237,,0.48022598,0.53672314,0.6214689,0.6214689,0.69491524,0.6101695,0.6779661,0.6779661,0.75141245,0.66101694,0.6836158,0.6384181,0.7457627,0.75706214,0.6723164,0.740113,0.79096043,0.779661,0.6779661,0.8079096,0.7175141,0.8305085,0.80225986,0.8418079,0.84745765,0.81920904,0.79096043,0.8305085,0.84745765,0.8587571,0.8418079,0.9039548,0.8926554,0.8418079,0.88135594,0.89830506,0.8926554,0.91525424,0.7740113,0.88135594,0.779661,0.96045196,0.88135594,0.9322034,0.8757062,0.9491525,0.9661017,0.8926554,0.94350284,1.0,0.8700565,0.97740114,0.9322034,0.9039548,1.0,0.99435025,0.9378531,1.0,0.9491525,1.0,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.15819208,0.12429378,0.16949153,0.27118644,0.27118644,0.33898306,0.3220339,0.27118644,0.4124294,0.42372882,0.48022598,0.37853107,0.4915254,0.519774,,0.5423729,0.5706215,0.5536723,0.63276833,0.6214689,0.6666667,0.68079096,0.6440678,0.6723164,0.6440678,0.6666667,0.7288136,0.69491524,0.759887,0.72316384,0.7344633,0.7175141,0.8163842,0.7175141,0.7288136,0.779661,0.7175141,0.81920904,0.7457627,0.7344633,0.8361582,0.779661,0.8079096,0.75706214,0.78531075,0.81920904,0.80225986,0.85310733,0.7966102,0.81920904,0.8418079,0.8926554,0.7457627,0.8870056,0.75141245,0.89830506,0.91525424,0.90960455,0.9830508,0.91525424,0.91525424,0.9322034,0.9322034,0.92655367,0.8305085,0.920904,0.96892655,0.9830508,0.9830508,0.94350284,0.98870057,0.98870057,0.98870057,1.0,1.0,1.0
+RealMLP (tuned),0.06779661,0.1299435,0.16949153,0.20903955,0.17514125,0.23728813,0.18644068,0.26553673,0.36158192,0.27683616,0.39548022,0.39548022,0.5141243,0.46327683,0.45762712,,0.53672314,0.5480226,0.6723164,0.5480226,0.6666667,0.62711865,0.6836158,0.6666667,0.6836158,0.66101694,0.69491524,0.75141245,0.6045198,0.7118644,0.7627119,0.72316384,0.6666667,0.8079096,0.76836157,0.779661,0.75706214,0.8361582,0.8418079,0.80225986,0.8757062,0.8135593,0.8587571,0.8135593,0.84745765,0.85310733,0.86440676,0.79096043,0.8361582,0.920904,0.92655367,0.920904,0.7457627,0.91525424,0.75706214,0.9322034,0.9378531,0.9548023,0.9039548,0.9717514,0.9717514,0.8870056,0.9661017,0.97740114,0.8418079,0.98870057,0.98870057,0.9378531,0.98870057,0.98870057,0.9548023,1.0,0.9717514,0.99435025,0.99435025,0.99435025
+TabDPT-Turbo (default),0.04519774,0.13559322,0.11299435,0.16384181,0.16949153,0.2542373,0.15819208,0.21468927,0.22033899,0.34463277,0.31638417,0.34463277,0.3898305,0.37853107,0.42937854,0.46327683,,0.6045198,0.6723164,0.5706215,0.6779661,0.6384181,0.68926555,0.59887004,0.65536726,0.66101694,0.740113,0.740113,0.59887004,0.7288136,0.7627119,0.7288136,0.59887004,0.78531075,0.68926555,0.7966102,0.75141245,0.8305085,0.8079096,0.7627119,0.84745765,0.8870056,0.84745765,0.79096043,0.89830506,0.8418079,0.8870056,0.84745765,0.8248588,0.90960455,0.90960455,0.92655367,0.7288136,0.92655367,0.740113,0.91525424,0.92655367,0.94350284,0.920904,0.9661017,0.97740114,0.9322034,0.9661017,0.98870057,0.8587571,0.98870057,0.9830508,0.96045196,0.98870057,1.0,0.9661017,1.0,0.9717514,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.050847456,0.09039548,0.14689265,0.107344635,0.14124294,0.15819208,0.16949153,0.19774011,0.20903955,0.2937853,0.24293785,0.36723164,0.37288135,0.37853107,0.4463277,0.4519774,0.39548022,,0.6045198,0.5480226,0.6158192,0.50282484,0.6045198,0.55932206,0.7457627,0.6214689,0.65536726,0.78531075,0.5480226,0.7344633,0.8135593,0.72316384,0.53107345,0.8418079,0.6666667,0.7457627,0.79096043,0.8418079,0.740113,0.7966102,0.7288136,0.8361582,0.84745765,0.8135593,0.84745765,0.88135594,0.91525424,0.81920904,0.84745765,0.90960455,0.920904,0.9548023,0.7457627,0.8926554,0.75706214,0.9039548,0.89830506,0.96045196,0.89830506,0.97740114,0.99435025,0.90960455,0.9548023,0.99435025,0.84745765,1.0,0.9717514,0.920904,0.98870057,1.0,0.9548023,1.0,0.9661017,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.04519774,0.073446326,0.1299435,0.04519774,0.10169491,0.107344635,0.16949153,0.19209039,0.19774011,0.22033899,0.2259887,0.3220339,0.29943502,0.30508474,0.36723164,0.32768363,0.32768363,0.39548022,,0.50282484,0.48022598,0.48022598,0.63276833,0.46892655,0.6214689,0.5536723,0.53107345,0.7457627,0.5084746,0.68926555,0.75141245,0.79096043,0.5141243,0.75141245,0.5762712,0.69491524,0.7457627,0.76836157,0.6158192,0.8135593,0.69491524,0.76836157,0.8135593,0.7966102,0.8079096,0.920904,0.78531075,0.8361582,0.8418079,0.8079096,0.8926554,0.94350284,0.7457627,0.89830506,0.7457627,0.920904,0.9039548,0.96045196,0.86440676,0.99435025,0.920904,0.89830506,0.97740114,0.9322034,0.8361582,0.96045196,0.9830508,0.9322034,1.0,0.99435025,0.96045196,1.0,0.9661017,1.0,1.0,1.0
+TabSwift (default),0.1299435,0.096045196,0.07909604,0.23163842,0.20903955,0.18079096,0.16949153,0.17514125,0.27683616,0.33898306,0.32768363,0.2881356,0.3898305,0.3898305,0.37853107,0.4519774,0.42937854,0.4519774,0.49717513,,0.4915254,0.59887004,0.5536723,0.62711865,0.49717513,0.53672314,0.55932206,0.5423729,0.56497175,0.5819209,0.6158192,0.5762712,0.63276833,0.6158192,0.700565,0.7118644,0.6158192,0.72316384,0.700565,0.65536726,0.8926554,0.7457627,0.78531075,0.700565,0.7740113,0.7288136,0.8361582,0.8135593,0.75706214,0.86440676,0.920904,0.85310733,0.72316384,0.8305085,0.7344633,0.8248588,0.8305085,0.85310733,0.8926554,0.94350284,0.9548023,0.94350284,0.90960455,0.98870057,0.81920904,0.96045196,0.9491525,0.96045196,0.9717514,1.0,0.97740114,0.99435025,0.98870057,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.04519774,0.11299435,0.11299435,0.14689265,0.107344635,0.15819208,0.118644066,0.20338982,0.13559322,0.18079096,0.19209039,0.3220339,0.24858756,0.3220339,0.33333334,0.33333334,0.3220339,0.38418078,0.519774,0.5084746,,0.45762712,0.5480226,0.47457626,0.5762712,0.53672314,0.56497175,0.65536726,0.4858757,0.6214689,0.6779661,0.63276833,0.48022598,0.69491524,0.55932206,0.6779661,0.6779661,0.8135593,0.5536723,0.7175141,0.75141245,0.8248588,0.8361582,0.7175141,0.8361582,0.76836157,0.8305085,0.84745765,0.75706214,0.8305085,0.90960455,0.920904,0.740113,0.92655367,0.740113,0.8248588,0.9322034,0.91525424,0.90960455,0.9661017,0.9830508,0.9378531,0.9491525,0.9717514,0.8305085,0.97740114,0.97740114,0.9548023,0.9830508,0.99435025,0.9661017,0.99435025,0.98870057,0.99435025,0.99435025,0.99435025
+Mitra (default),0.06779661,0.050847456,0.20903955,0.1299435,0.18644068,0.14124294,0.17514125,0.20903955,0.259887,0.24858756,0.2542373,0.3502825,0.33333334,0.3220339,0.31920904,0.37288135,0.36158192,0.49717513,0.519774,0.40112993,0.5423729,,0.5423729,0.5084746,0.55932206,0.5536723,0.49717513,0.59322035,0.4378531,0.6384181,0.6497175,0.6101695,0.48305085,0.62711865,0.6214689,0.6158192,0.66101694,0.6836158,0.70621467,0.69491524,0.740113,0.70621467,0.7966102,0.8079096,0.7175141,0.740113,0.90960455,0.81920904,0.8587571,0.90960455,0.8757062,0.84745765,0.8587571,0.7740113,0.88135594,0.8587571,0.8248588,0.8870056,0.8587571,0.90960455,0.92655367,0.91525424,0.89830506,0.9378531,0.89830506,0.94350284,0.9237288,0.94350284,0.98870057,0.92655367,0.9661017,0.98870057,0.9830508,0.98870057,0.98870057,1.0
+CatBoost (tuned),0.028248588,0.03954802,0.12429378,0.050847456,0.107344635,0.07909604,0.14124294,0.14124294,0.17514125,0.2259887,0.20338982,0.2881356,0.28248587,0.24858756,0.3559322,0.31638417,0.31073445,0.39548022,0.36723164,0.4463277,0.4519774,0.45762712,,0.45762712,0.5536723,0.5536723,0.4915254,0.66101694,0.48022598,0.6440678,0.72316384,0.6779661,0.5141243,0.7118644,0.56497175,0.65536726,0.68926555,0.7288136,0.59887004,0.76836157,0.6779661,0.75141245,0.78531075,0.7457627,0.7627119,0.89830506,0.7627119,0.8305085,0.8248588,0.79096043,0.8757062,0.91525424,0.740113,0.86440676,0.75141245,0.9039548,0.8926554,0.9491525,0.8418079,0.99435025,0.920904,0.8926554,0.9661017,0.92655367,0.8248588,0.9491525,0.9717514,0.920904,0.99435025,0.9830508,0.9548023,1.0,0.96045196,1.0,1.0,1.0
+TabM (tuned + ensembled),0.056497175,0.07909604,0.13559322,0.13559322,0.107344635,0.096045196,0.11299435,0.11299435,0.23728813,0.259887,0.29943502,0.259887,0.4180791,0.33898306,0.32768363,0.33333334,0.40112993,0.44067797,0.53107345,0.37288135,0.5254237,0.4915254,0.5423729,,0.50282484,0.53672314,0.59322035,0.5875706,0.4915254,0.59887004,0.6440678,0.62711865,0.5141243,0.6101695,0.8361582,0.69491524,0.62711865,0.68926555,0.88135594,0.6836158,0.7966102,0.6666667,0.7118644,0.6666667,0.68926555,0.78531075,0.8305085,0.7457627,0.740113,0.88135594,0.8135593,0.8418079,0.72316384,0.75141245,0.7344633,0.8079096,0.7627119,0.8418079,0.76836157,0.9039548,0.92655367,0.8361582,0.86440676,0.96045196,0.8305085,0.97740114,0.920904,0.8870056,0.9717514,0.9548023,0.920904,0.98870057,0.9322034,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.033898305,0.04519774,0.09039548,0.050847456,0.107344635,0.11299435,0.14124294,0.14124294,0.16384181,0.21468927,0.19774011,0.29943502,0.29943502,0.31638417,0.3559322,0.31638417,0.34463277,0.2542373,0.37853107,0.50282484,0.42372882,0.44067797,0.4463277,0.49717513,,0.4124294,0.50282484,0.59887004,0.4915254,0.56497175,0.6666667,0.6836158,0.519774,0.79096043,0.5819209,0.5819209,0.7118644,0.6384181,0.6666667,0.79096043,0.700565,0.76836157,0.75706214,0.7627119,0.76836157,0.8700565,0.68926555,0.740113,0.7966102,0.7288136,0.84745765,0.97740114,0.7175141,0.86440676,0.72316384,0.92655367,0.8700565,0.9661017,0.8587571,0.9717514,0.9661017,0.8587571,0.9661017,0.8926554,0.8079096,0.98870057,0.9661017,0.9039548,1.0,0.98870057,0.9322034,1.0,0.96045196,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.084745765,0.118644066,0.15819208,0.14124294,0.11299435,0.19209039,0.23163842,0.23163842,0.2542373,0.259887,0.2542373,0.33898306,0.2937853,0.36158192,0.33333334,0.33898306,0.33898306,0.37853107,0.4463277,0.46327683,0.46327683,0.4463277,0.4463277,0.46327683,0.5875706,,0.49717513,0.59322035,0.5536723,0.6384181,0.59887004,0.62711865,0.5819209,0.63276833,0.5536723,0.5536723,0.68926555,0.6384181,0.53672314,0.68926555,0.6497175,0.7740113,0.740113,0.7175141,0.7740113,0.7457627,0.700565,0.8418079,0.7627119,0.68926555,0.8361582,0.85310733,0.7288136,0.8757062,0.740113,0.8361582,0.8757062,0.85310733,0.8757062,0.8926554,0.89830506,0.8700565,0.86440676,0.88135594,0.84745765,0.89830506,0.94350284,0.9491525,0.9830508,0.94350284,0.9830508,0.98870057,0.9830508,0.99435025,0.99435025,1.0
+xRFM (tuned + ensembled),0.06214689,0.084745765,0.107344635,0.096045196,0.12429378,0.12429378,0.15254237,0.18644068,0.13559322,0.12429378,0.16949153,0.24293785,0.23163842,0.2542373,0.27118644,0.30508474,0.259887,0.34463277,0.46892655,0.44067797,0.43502825,0.50282484,0.5084746,0.40677965,0.49717513,0.50282484,,0.5819209,0.46892655,0.6158192,0.65536726,0.59322035,0.5084746,0.6384181,0.49717513,0.6666667,0.6497175,0.72316384,0.5536723,0.68926555,0.59322035,0.70621467,0.76836157,0.75706214,0.7344633,0.7288136,0.7175141,0.8700565,0.76836157,0.7457627,0.8079096,0.85310733,0.70621467,0.7966102,0.7118644,0.79096043,0.8079096,0.80225986,0.9039548,0.9322034,0.84745765,0.9661017,0.8135593,0.920904,0.7740113,0.88135594,0.9717514,0.96045196,0.94350284,0.9661017,0.97740114,0.97740114,0.97740114,0.99435025,0.99435025,0.99435025
+XGBoost (tuned + ensembled),0.04519774,0.06214689,0.09039548,0.07909604,0.096045196,0.07909604,0.096045196,0.1299435,0.1299435,0.14689265,0.15254237,0.22033899,0.24858756,0.24293785,0.30508474,0.24858756,0.259887,0.21468927,0.2542373,0.45762712,0.34463277,0.40677965,0.33898306,0.4124294,0.40112993,0.40677965,0.4180791,,0.4463277,0.49717513,0.6666667,0.59322035,0.47457626,0.6214689,0.5706215,0.5875706,0.7118644,0.6101695,0.5875706,0.740113,0.6836158,0.7288136,0.7344633,0.7457627,0.75141245,0.8757062,0.6779661,0.75141245,0.779661,0.7344633,0.8305085,0.9717514,0.7118644,0.8418079,0.7288136,0.9039548,0.86440676,0.9378531,0.8248588,0.98870057,0.9322034,0.8587571,0.96045196,0.88135594,0.7966102,0.97740114,0.97740114,0.90960455,0.9830508,0.98870057,0.9548023,1.0,0.9548023,1.0,1.0,1.0
+TabPFNv2 (tuned),0.118644066,0.06214689,0.09039548,0.14689265,0.14689265,0.18079096,0.15254237,0.096045196,0.2542373,0.259887,0.30508474,0.17514125,0.36723164,0.32768363,0.24011299,0.39548022,0.40112993,0.4519774,0.4915254,0.43502825,0.5141243,0.5621469,0.519774,0.5084746,0.5084746,0.4463277,0.53107345,0.5536723,,0.59322035,0.5762712,0.5480226,0.5508475,0.59322035,0.56497175,0.5875706,0.55932206,0.6440678,0.6497175,0.6045198,0.6836158,0.6101695,0.700565,0.70621467,0.6158192,0.65536726,0.72316384,0.779661,0.75141245,0.7457627,0.7175141,0.7344633,0.700565,0.69491524,0.700565,0.78531075,0.7344633,0.7288136,0.81920904,0.80225986,0.8079096,0.8757062,0.7740113,0.86440676,0.75141245,0.84745765,0.8954802,0.8926554,0.90960455,0.89830506,0.920904,0.9491525,0.9322034,0.98870057,0.98870057,1.0
+ModernNCA (tuned + ensembled),0.050847456,0.096045196,0.1299435,0.118644066,0.13559322,0.14124294,0.13559322,0.21468927,0.15819208,0.19209039,0.17514125,0.2937853,0.23728813,0.259887,0.27683616,0.2881356,0.27118644,0.26553673,0.31073445,0.4180791,0.37853107,0.36158192,0.3559322,0.40112993,0.43502825,0.36158192,0.38418078,0.50282484,0.40677965,,0.5141243,0.5480226,0.44067797,0.519774,0.48022598,0.4915254,0.5706215,0.53107345,0.5141243,0.62711865,0.65536726,0.68926555,0.8418079,0.6384181,0.7118644,0.66101694,0.6836158,0.6779661,0.6836158,0.6836158,0.84745765,0.75706214,0.70621467,0.81920904,0.72316384,0.740113,0.85310733,0.8361582,0.8700565,0.8870056,0.90960455,0.9039548,0.8700565,0.8870056,0.7740113,0.920904,0.9548023,0.94350284,0.94350284,0.97740114,0.9717514,1.0,0.97740114,0.99435025,0.99435025,1.0
+XGBoost (tuned),0.033898305,0.050847456,0.096045196,0.06214689,0.107344635,0.073446326,0.07909604,0.118644066,0.096045196,0.1299435,0.1299435,0.19774011,0.21468927,0.20903955,0.26553673,0.23728813,0.23728813,0.18644068,0.24858756,0.38418078,0.3220339,0.3502825,0.27683616,0.3559322,0.33333334,0.40112993,0.34463277,0.33333334,0.42372882,0.4858757,,0.5423729,0.45762712,0.56497175,0.519774,0.53672314,0.6214689,0.5480226,0.53672314,0.66101694,0.6497175,0.69491524,0.68926555,0.69491524,0.7118644,0.8248588,0.6723164,0.740113,0.7740113,0.72316384,0.7966102,0.920904,0.7118644,0.78531075,0.7118644,0.8700565,0.7966102,0.92655367,0.8079096,0.96045196,0.920904,0.86440676,0.920904,0.8926554,0.7966102,0.96045196,0.9548023,0.8926554,0.9830508,0.97740114,0.94350284,1.0,0.9548023,1.0,1.0,1.0
+CatBoost (default),0.03954802,0.03954802,0.107344635,0.050847456,0.096045196,0.06214689,0.12429378,0.096045196,0.118644066,0.16384181,0.15819208,0.24293785,0.23728813,0.22033899,0.28248587,0.27683616,0.27118644,0.27683616,0.20903955,0.42372882,0.36723164,0.3898305,0.3220339,0.37288135,0.31638417,0.37288135,0.40677965,0.40677965,0.4519774,0.4519774,0.45762712,,0.49717513,0.47457626,0.4915254,0.55932206,0.6384181,0.5706215,0.5536723,0.6779661,0.65536726,0.70621467,0.63276833,0.75141245,0.7175141,0.8870056,0.6384181,0.63276833,0.80225986,0.6779661,0.69491524,0.8079096,0.72316384,0.8079096,0.7288136,0.8926554,0.8248588,0.92655367,0.7966102,0.8248588,0.89830506,0.7740113,0.94350284,0.80225986,0.8248588,0.91525424,0.92655367,0.86440676,1.0,0.9830508,0.9491525,1.0,0.96045196,1.0,1.0,1.0
+TabPFNv2 (default),0.03954802,0.06214689,0.04519774,0.1299435,0.07909604,0.13559322,0.096045196,0.107344635,0.20903955,0.1299435,0.26553673,0.21468927,0.2937853,0.3220339,0.18361582,0.33333334,0.40112993,0.46892655,0.4858757,0.36723164,0.519774,0.5169492,0.4858757,0.4858757,0.48022598,0.4180791,0.4915254,0.5254237,0.44915253,0.55932206,0.5423729,0.50282484,,0.5423729,0.5480226,0.5536723,0.53672314,0.62711865,0.5875706,0.55932206,0.62711865,0.6497175,0.6836158,0.6384181,0.65536726,0.5762712,0.6836158,0.76836157,0.68926555,0.70621467,0.72316384,0.6836158,0.6723164,0.70621467,0.6723164,0.7118644,0.75141245,0.72316384,0.8418079,0.75706214,0.72316384,0.8757062,0.7627119,0.8305085,0.7175141,0.7740113,0.90677965,0.90960455,0.8418079,0.84745765,0.920904,0.9039548,0.9322034,1.0,1.0,1.0
+LightGBM (tuned),0.028248588,0.033898305,0.06214689,0.050847456,0.084745765,0.07909604,0.084745765,0.11299435,0.13559322,0.12429378,0.1299435,0.23163842,0.22033899,0.19209039,0.28248587,0.19209039,0.21468927,0.15819208,0.24858756,0.38418078,0.30508474,0.37288135,0.2881356,0.3898305,0.20903955,0.36723164,0.36158192,0.37853107,0.40677965,0.48022598,0.43502825,0.5254237,0.45762712,,0.5254237,0.46892655,0.5480226,0.49717513,0.5762712,0.6045198,0.63276833,0.68926555,0.6666667,0.6836158,0.69491524,0.740113,0.6384181,0.6836158,0.7627119,0.6666667,0.7627119,0.8926554,0.70621467,0.78531075,0.7175141,0.81920904,0.7966102,0.8587571,0.78531075,0.9322034,0.89830506,0.85310733,0.8870056,0.85310733,0.78531075,0.94350284,0.9378531,0.8870056,0.9830508,0.9717514,0.92655367,1.0,0.9322034,0.98870057,0.98870057,0.99435025
+TabM (tuned),0.03954802,0.056497175,0.096045196,0.096045196,0.10169491,0.09039548,0.096045196,0.07909604,0.16949153,0.15254237,0.21468927,0.17514125,0.34463277,0.28248587,0.27118644,0.23163842,0.31073445,0.33333334,0.42372882,0.29943502,0.44067797,0.37853107,0.43502825,0.16384181,0.4180791,0.4463277,0.50282484,0.42937854,0.43502825,0.519774,0.48022598,0.5084746,0.4519774,0.47457626,,0.5819209,0.5254237,0.59887004,0.6497175,0.519774,0.68926555,0.6158192,0.6045198,0.5819209,0.62711865,0.6384181,0.7288136,0.6779661,0.6158192,0.78531075,0.72316384,0.7288136,0.700565,0.66101694,0.7118644,0.7288136,0.68926555,0.75141245,0.69491524,0.81920904,0.8361582,0.779661,0.76836157,0.92655367,0.7966102,0.91525424,0.8587571,0.85310733,0.91525424,0.8926554,0.8926554,0.97740114,0.920904,0.9661017,0.9661017,0.99435025
+xRFM (tuned),0.03954802,0.056497175,0.09039548,0.050847456,0.107344635,0.056497175,0.11299435,0.14124294,0.096045196,0.084745765,0.11299435,0.17514125,0.19774011,0.16949153,0.22033899,0.22033899,0.20338982,0.2542373,0.30508474,0.2881356,0.3220339,0.38418078,0.34463277,0.30508474,0.4180791,0.4463277,0.33333334,0.4124294,0.4124294,0.5084746,0.46327683,0.44067797,0.4463277,0.53107345,0.4180791,,0.48022598,0.59322035,0.46327683,0.59322035,0.5706215,0.5875706,0.68926555,0.69491524,0.6384181,0.6497175,0.6836158,0.8079096,0.7118644,0.6723164,0.76836157,0.76836157,0.6836158,0.740113,0.68926555,0.740113,0.7344633,0.7457627,0.79096043,0.8926554,0.8079096,0.8926554,0.7740113,0.9039548,0.75141245,0.85310733,0.9491525,0.9378531,0.920904,0.9378531,0.9548023,0.98870057,0.9661017,0.9830508,0.9830508,1.0
+ChimeraBoost (tuned + ensembled),0.033898305,0.02259887,0.10169491,0.03954802,0.11299435,0.056497175,0.096045196,0.09039548,0.14124294,0.18644068,0.16949153,0.21468927,0.23728813,0.19774011,0.28248587,0.24293785,0.24858756,0.20903955,0.2542373,0.38418078,0.3220339,0.33898306,0.31073445,0.37288135,0.2881356,0.31073445,0.3502825,0.2881356,0.44067797,0.42937854,0.37853107,0.36158192,0.46327683,0.4519774,0.47457626,0.519774,,0.5084746,0.5819209,0.59322035,0.63276833,0.59322035,0.55932206,0.7344633,0.6158192,0.8305085,0.65536726,0.5875706,0.7966102,0.63276833,0.63276833,0.72316384,0.7118644,0.7457627,0.7344633,0.9039548,0.76836157,0.91525424,0.7288136,0.7627119,0.8248588,0.70621467,0.90960455,0.79096043,0.81920904,0.8305085,0.92655367,0.85310733,0.98870057,0.9717514,0.9322034,1.0,0.9491525,0.99435025,0.99435025,0.99435025
+iLTM (tuned),0.028248588,0.02259887,0.033898305,0.02259887,0.073446326,0.033898305,0.073446326,0.096045196,0.06214689,0.073446326,0.056497175,0.14689265,0.12429378,0.15819208,0.18079096,0.16384181,0.16949153,0.15819208,0.23163842,0.27683616,0.18644068,0.31638417,0.27118644,0.31073445,0.36158192,0.36158192,0.27683616,0.3898305,0.3559322,0.46892655,0.4519774,0.42937854,0.37288135,0.50282484,0.40112993,0.40677965,0.4915254,,0.42937854,0.5536723,0.5254237,0.62711865,0.6384181,0.62711865,0.65536726,0.6101695,0.59887004,0.7288136,0.6836158,0.66101694,0.7457627,0.75141245,0.6666667,0.72316384,0.66101694,0.70621467,0.72316384,0.740113,0.8079096,0.88135594,0.8361582,0.8757062,0.779661,0.85310733,0.7344633,0.8587571,0.920904,0.88135594,0.91525424,0.9548023,0.9322034,0.9830508,0.92655367,1.0,1.0,1.0
+TabM (default),0.028248588,0.033898305,0.07909604,0.050847456,0.096045196,0.050847456,0.073446326,0.050847456,0.107344635,0.107344635,0.13559322,0.1299435,0.19774011,0.15254237,0.2542373,0.15819208,0.19209039,0.259887,0.38418078,0.29943502,0.4463277,0.2937853,0.40112993,0.118644066,0.33333334,0.46327683,0.4463277,0.4124294,0.3502825,0.4858757,0.46327683,0.4463277,0.4124294,0.42372882,0.3502825,0.53672314,0.4180791,0.5706215,,0.4519774,0.5536723,0.62711865,0.5762712,0.53672314,0.6384181,0.5480226,0.700565,0.65536726,0.6045198,0.7740113,0.7118644,0.6836158,0.69491524,0.6214689,0.70621467,0.7118644,0.62711865,0.70621467,0.6440678,0.7966102,0.80225986,0.7627119,0.72316384,0.8926554,0.79096043,0.89830506,0.8135593,0.7966102,0.90960455,0.88135594,0.84745765,0.9830508,0.8587571,0.97740114,0.97740114,0.99435025
+ChimeraBoost (tuned),0.03954802,0.016949153,0.096045196,0.028248588,0.107344635,0.04519774,0.073446326,0.073446326,0.118644066,0.18644068,0.14689265,0.18079096,0.2259887,0.18079096,0.26553673,0.19774011,0.23728813,0.20338982,0.18644068,0.34463277,0.28248587,0.30508474,0.23163842,0.31638417,0.20903955,0.31073445,0.31073445,0.259887,0.39548022,0.37288135,0.33898306,0.3220339,0.44067797,0.39548022,0.48022598,0.40677965,0.40677965,0.4463277,0.5480226,,0.6101695,0.5762712,0.5423729,0.6666667,0.59322035,0.75141245,0.6101695,0.5480226,0.75141245,0.59887004,0.6101695,0.65536726,0.7175141,0.69491524,0.72316384,0.8248588,0.7344633,0.8700565,0.7175141,0.72316384,0.79096043,0.69491524,0.8757062,0.7740113,0.79096043,0.8135593,0.8587571,0.8418079,0.9830508,0.92655367,0.9039548,1.0,0.92655367,0.98870057,0.9830508,0.98870057
+RealMLP (default),0.011299435,0.033898305,0.03954802,0.07909604,0.06779661,0.033898305,0.028248588,0.06214689,0.06214689,0.016949153,0.07909604,0.107344635,0.14689265,0.20903955,0.16384181,0.12429378,0.15254237,0.27118644,0.30508474,0.107344635,0.24858756,0.259887,0.3220339,0.20338982,0.29943502,0.3502825,0.40677965,0.31638417,0.31638417,0.34463277,0.3502825,0.34463277,0.37288135,0.36723164,0.31073445,0.42937854,0.36723164,0.47457626,0.4463277,0.3898305,,0.6158192,0.50282484,0.50282484,0.62711865,0.4858757,0.59887004,0.6158192,0.5423729,0.6779661,0.6666667,0.6101695,0.6440678,0.700565,0.65536726,0.6101695,0.7175141,0.6045198,0.72316384,0.7457627,0.79096043,0.79096043,0.6836158,0.8926554,0.779661,0.84745765,0.8361582,0.8926554,0.8361582,0.92655367,0.9491525,0.9548023,0.9661017,0.9661017,0.9661017,1.0
+ExtraTrees (tuned + ensembled),0.033898305,0.04519774,0.06779661,0.06214689,0.10169491,0.07909604,0.07909604,0.14689265,0.06214689,0.073446326,0.073446326,0.17514125,0.06779661,0.16949153,0.22033899,0.18644068,0.11299435,0.16384181,0.23163842,0.2542373,0.17514125,0.2937853,0.24858756,0.33333334,0.23163842,0.2259887,0.2937853,0.27118644,0.3898305,0.31073445,0.30508474,0.2937853,0.3502825,0.31073445,0.38418078,0.4124294,0.40677965,0.37288135,0.37288135,0.42372882,0.38418078,,0.49717513,0.5536723,0.65254235,0.5141243,0.42372882,0.5423729,0.59322035,0.4519774,0.5141243,0.46327683,0.6101695,0.8079096,0.6158192,0.65536726,0.86440676,0.6497175,0.94350284,0.5819209,0.7288136,0.75706214,0.7118644,0.6384181,0.75706214,0.7175141,0.9548023,0.9661017,0.81920904,0.920904,0.98870057,0.99435025,0.98870057,1.0,1.0,1.0
+ModernNCA (tuned),0.011299435,0.050847456,0.07909604,0.056497175,0.07909604,0.06779661,0.04519774,0.09039548,0.056497175,0.073446326,0.06779661,0.18644068,0.118644066,0.15254237,0.19209039,0.14124294,0.15254237,0.15254237,0.18644068,0.21468927,0.16384181,0.20338982,0.21468927,0.2881356,0.24293785,0.259887,0.23163842,0.26553673,0.29943502,0.15819208,0.31073445,0.36723164,0.31638417,0.33333334,0.39548022,0.31073445,0.44067797,0.36158192,0.42372882,0.45762712,0.49717513,0.50282484,,0.5480226,0.519774,0.50282484,0.519774,0.5141243,0.5762712,0.5706215,0.70621467,0.5762712,0.6723164,0.68926555,0.6779661,0.63276833,0.68926555,0.6497175,0.72316384,0.7118644,0.8135593,0.8079096,0.7175141,0.7627119,0.72316384,0.8587571,0.8870056,0.9039548,0.85310733,0.9322034,0.94350284,0.9548023,0.9548023,0.97740114,0.97740114,1.0
+EBM (tuned + ensembled),0.033898305,0.016949153,0.073446326,0.02259887,0.06214689,0.033898305,0.04519774,0.056497175,0.11299435,0.14689265,0.13559322,0.14689265,0.18644068,0.14124294,0.24293785,0.18644068,0.20903955,0.18644068,0.20338982,0.29943502,0.28248587,0.19209039,0.2542373,0.33333334,0.23728813,0.28248587,0.24293785,0.2542373,0.2937853,0.36158192,0.30508474,0.24858756,0.36158192,0.31638417,0.4180791,0.30508474,0.26553673,0.37288135,0.46327683,0.33333334,0.49717513,0.4463277,0.4519774,,0.47457626,0.4180791,0.56497175,0.50282484,0.75706214,0.6045198,0.5480226,0.50282484,0.6497175,0.5480226,0.6666667,0.8248588,0.59322035,0.7457627,0.5706215,0.62711865,0.6497175,0.6045198,0.76836157,0.7175141,0.8079096,0.7288136,0.740113,0.8248588,0.9491525,0.9039548,0.8418079,0.90960455,0.8870056,0.9830508,0.9830508,0.99435025
+ExtraTrees (tuned),0.028248588,0.06214689,0.056497175,0.06214689,0.107344635,0.06779661,0.073446326,0.14124294,0.073446326,0.06214689,0.084745765,0.16949153,0.06214689,0.15819208,0.21468927,0.15254237,0.10169491,0.15254237,0.19209039,0.2259887,0.16384181,0.28248587,0.23728813,0.31073445,0.23163842,0.2259887,0.26553673,0.24858756,0.38418078,0.2881356,0.2881356,0.28248587,0.34463277,0.30508474,0.37288135,0.36158192,0.38418078,0.34463277,0.36158192,0.40677965,0.37288135,0.34745762,0.48022598,0.5254237,,0.4915254,0.40112993,0.53672314,0.56497175,0.44067797,0.5084746,0.44067797,0.6045198,0.75141245,0.6158192,0.63276833,0.8079096,0.6384181,0.8870056,0.53107345,0.6836158,0.75141245,0.69491524,0.6158192,0.75141245,0.70621467,0.9378531,0.96045196,0.8305085,0.9039548,0.99435025,0.99435025,0.98870057,1.0,1.0,1.0
+ChimeraBoost (default),0.0056497175,0.02259887,0.050847456,0.016949153,0.06214689,0.016949153,0.06779661,0.033898305,0.084745765,0.10169491,0.084745765,0.11299435,0.17514125,0.096045196,0.18079096,0.14689265,0.15819208,0.118644066,0.07909604,0.27118644,0.23163842,0.259887,0.10169491,0.21468927,0.1299435,0.2542373,0.27118644,0.12429378,0.34463277,0.33898306,0.17514125,0.11299435,0.42372882,0.259887,0.36158192,0.3502825,0.16949153,0.3898305,0.4519774,0.24858756,0.5141243,0.4858757,0.49717513,0.5819209,0.5084746,,0.5875706,0.53107345,0.65536726,0.6101695,0.5480226,0.5536723,0.68926555,0.69491524,0.69491524,0.72316384,0.69491524,0.779661,0.70621467,0.70621467,0.69491524,0.6497175,0.8248588,0.740113,0.779661,0.7344633,0.84745765,0.8361582,0.9717514,0.91525424,0.8870056,0.98870057,0.9039548,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.011299435,0.033898305,0.033898305,0.04519774,0.056497175,0.050847456,0.016949153,0.028248588,0.07909604,0.118644066,0.07909604,0.107344635,0.12429378,0.107344635,0.19774011,0.13559322,0.11299435,0.084745765,0.21468927,0.16384181,0.16949153,0.09039548,0.23728813,0.16949153,0.31073445,0.29943502,0.28248587,0.3220339,0.27683616,0.31638417,0.32768363,0.36158192,0.31638417,0.36158192,0.27118644,0.31638417,0.34463277,0.40112993,0.29943502,0.3898305,0.40112993,0.5762712,0.48022598,0.43502825,0.59887004,0.4124294,,0.5084746,0.5084746,0.65536726,0.6214689,0.56497175,0.6666667,0.63276833,0.66101694,0.5762712,0.6440678,0.55932206,0.6440678,0.6723164,0.7740113,0.740113,0.6440678,0.92655367,0.79096043,0.8700565,0.76836157,0.8248588,0.85310733,0.8700565,0.8700565,0.88135594,0.88135594,0.90960455,0.90960455,1.0
+LimiX (default),0.016949153,0.02259887,0.09039548,0.06779661,0.084745765,0.073446326,0.12429378,0.107344635,0.107344635,0.10169491,0.096045196,0.16384181,0.1299435,0.15819208,0.14689265,0.20903955,0.15254237,0.18079096,0.16384181,0.18644068,0.15254237,0.18079096,0.16949153,0.2542373,0.259887,0.15819208,0.1299435,0.24858756,0.22033899,0.3220339,0.259887,0.36723164,0.23163842,0.31638417,0.3220339,0.19209039,0.4124294,0.27118644,0.34463277,0.4519774,0.38418078,0.45762712,0.4858757,0.49717513,0.46327683,0.46892655,0.4915254,,0.5423729,0.50282484,0.53672314,0.519774,0.66101694,0.5254237,0.66101694,0.5536723,0.5536723,0.59322035,0.5254237,0.59887004,0.6497175,0.66101694,0.59887004,0.6214689,0.70621467,0.6836158,0.66101694,0.76836157,0.7457627,0.6836158,0.8361582,0.7627119,0.9378531,0.91525424,0.91525424,0.9830508
+EBM (tuned),0.02259887,0.016949153,0.056497175,0.033898305,0.06779661,0.02259887,0.04519774,0.03954802,0.12429378,0.12429378,0.12429378,0.118644066,0.16949153,0.118644066,0.20338982,0.16384181,0.17514125,0.15254237,0.15819208,0.24293785,0.24293785,0.14124294,0.17514125,0.259887,0.20338982,0.23728813,0.23163842,0.22033899,0.24858756,0.31638417,0.2259887,0.19774011,0.31073445,0.23728813,0.38418078,0.2881356,0.20338982,0.31638417,0.39548022,0.24858756,0.45762712,0.40677965,0.42372882,0.24293785,0.43502825,0.34463277,0.4915254,0.45762712,,0.56497175,0.5254237,0.42937854,0.63276833,0.50282484,0.6440678,0.66101694,0.55932206,0.6779661,0.5423729,0.5762712,0.59322035,0.5875706,0.72316384,0.700565,0.7966102,0.68926555,0.700565,0.8135593,0.9322034,0.8700565,0.8361582,0.9039548,0.8587571,0.97740114,0.97740114,0.99435025
+TorchMLP (tuned),0.0056497175,0.016949153,0.028248588,0.016949153,0.073446326,0.016949153,0.016949153,0.02259887,0.050847456,0.07909604,0.06214689,0.09039548,0.09039548,0.10169491,0.18079096,0.07909604,0.09039548,0.09039548,0.19209039,0.13559322,0.16949153,0.09039548,0.20903955,0.118644066,0.27118644,0.31073445,0.2542373,0.26553673,0.2542373,0.31638417,0.27683616,0.3220339,0.2937853,0.33333334,0.21468927,0.32768363,0.36723164,0.33898306,0.2259887,0.40112993,0.3220339,0.5480226,0.42937854,0.39548022,0.55932206,0.3898305,0.34463277,0.49717513,0.43502825,,0.5706215,0.5084746,0.66101694,0.6045198,0.66101694,0.5423729,0.6158192,0.53672314,0.6214689,0.59887004,0.65536726,0.700565,0.6101695,0.84745765,0.78531075,0.7740113,0.72316384,0.7966102,0.79096043,0.80225986,0.8587571,0.8757062,0.85310733,0.8757062,0.8757062,1.0
+ModernNCA (default),0.011299435,0.02259887,0.03954802,0.028248588,0.073446326,0.016949153,0.033898305,0.06779661,0.04519774,0.03954802,0.04519774,0.12429378,0.04519774,0.107344635,0.15819208,0.073446326,0.09039548,0.07909604,0.107344635,0.07909604,0.09039548,0.12429378,0.12429378,0.18644068,0.15254237,0.16384181,0.19209039,0.16949153,0.28248587,0.15254237,0.20338982,0.30508474,0.27683616,0.23728813,0.27683616,0.23163842,0.36723164,0.2542373,0.2881356,0.3898305,0.33333334,0.4858757,0.2937853,0.4519774,0.4915254,0.4519774,0.37853107,0.46327683,0.47457626,0.42937854,,0.42937854,0.6440678,0.6384181,0.6440678,0.5480226,0.6779661,0.5480226,0.740113,0.6158192,0.7175141,0.779661,0.6440678,0.69491524,0.740113,0.779661,0.9039548,0.89830506,0.81920904,0.89830506,0.92655367,0.9491525,0.9548023,0.9491525,0.9491525,1.0
+LightGBM (default),0.0,0.0056497175,0.011299435,0.0,0.050847456,0.0056497175,0.028248588,0.016949153,0.028248588,0.016949153,0.02259887,0.050847456,0.056497175,0.084745765,0.107344635,0.07909604,0.073446326,0.04519774,0.056497175,0.14689265,0.07909604,0.15254237,0.084745765,0.15819208,0.02259887,0.14689265,0.14689265,0.028248588,0.26553673,0.24293785,0.07909604,0.19209039,0.31638417,0.107344635,0.27118644,0.23163842,0.27683616,0.24858756,0.31638417,0.34463277,0.3898305,0.53672314,0.42372882,0.49717513,0.55932206,0.4463277,0.43502825,0.48022598,0.5706215,0.4915254,0.5706215,,0.6836158,0.59322035,0.6836158,0.6384181,0.65536726,0.63276833,0.700565,0.75706214,0.72316384,0.7457627,0.68926555,0.7457627,0.72316384,0.79096043,0.85310733,0.81920904,0.91525424,0.90960455,0.8757062,0.98870057,0.89830506,0.9830508,0.9830508,0.99435025
+TabSTAR (tuned + ensembled),0.096045196,0.107344635,0.17514125,0.12429378,0.16384181,0.12429378,0.1299435,0.16949153,0.21468927,0.23163842,0.2259887,0.24293785,0.21468927,0.2259887,0.2542373,0.2542373,0.27118644,0.2542373,0.2542373,0.27683616,0.259887,0.14124294,0.259887,0.27683616,0.28248587,0.27118644,0.2937853,0.2881356,0.29943502,0.2937853,0.2881356,0.27683616,0.32768363,0.2937853,0.29943502,0.31638417,0.2881356,0.33333334,0.30508474,0.28248587,0.3559322,0.3898305,0.32768363,0.3502825,0.39548022,0.31073445,0.33333334,0.33898306,0.36723164,0.33898306,0.3559322,0.31638417,,0.45762712,0.59887004,0.42937854,0.5254237,0.5084746,0.519774,0.36723164,0.45762712,0.45762712,0.55932206,0.4463277,0.97740114,0.45762712,0.5254237,0.72316384,0.59887004,0.68926555,0.7966102,0.65536726,0.8361582,0.78531075,0.78531075,0.8926554
+RandomForest (tuned + ensembled),0.0056497175,0.016949153,0.028248588,0.011299435,0.03954802,0.011299435,0.03954802,0.056497175,0.02259887,0.02259887,0.03954802,0.1299435,0.033898305,0.118644066,0.11299435,0.084745765,0.073446326,0.107344635,0.10169491,0.16949153,0.073446326,0.2259887,0.13559322,0.24858756,0.13559322,0.12429378,0.20338982,0.15819208,0.30508474,0.18079096,0.21468927,0.19209039,0.2937853,0.21468927,0.33898306,0.259887,0.2542373,0.27683616,0.37853107,0.30508474,0.29943502,0.19209039,0.31073445,0.4519774,0.24858756,0.30508474,0.36723164,0.47457626,0.49717513,0.39548022,0.36158192,0.40677965,0.5423729,,0.55932206,0.5762712,0.7033898,0.5875706,0.740113,0.5254237,0.53672314,0.6101695,0.6384181,0.5254237,0.700565,0.59887004,0.98870057,0.9039548,0.78531075,0.88135594,0.9378531,0.96045196,0.96045196,1.0,1.0,1.0
+TabSTAR (tuned),0.09039548,0.096045196,0.15254237,0.12429378,0.14689265,0.11299435,0.12429378,0.15254237,0.21468927,0.2259887,0.22033899,0.23728813,0.22033899,0.22033899,0.24858756,0.24293785,0.259887,0.24293785,0.2542373,0.26553673,0.259887,0.118644066,0.24858756,0.26553673,0.27683616,0.259887,0.2881356,0.27118644,0.29943502,0.27683616,0.2881356,0.27118644,0.32768363,0.28248587,0.2881356,0.31073445,0.26553673,0.33898306,0.2937853,0.27683616,0.34463277,0.38418078,0.3220339,0.33333334,0.38418078,0.30508474,0.33898306,0.33898306,0.3559322,0.33898306,0.3559322,0.31638417,0.40112993,0.44067797,,0.43502825,0.50282484,0.5084746,0.4915254,0.36723164,0.45762712,0.45762712,0.55932206,0.4463277,0.96045196,0.4519774,0.50282484,0.72316384,0.59887004,0.68926555,0.79096043,0.65536726,0.8361582,0.78531075,0.78531075,0.8700565
+EBM (default),0.016949153,0.011299435,0.033898305,0.016949153,0.02259887,0.02259887,0.056497175,0.033898305,0.056497175,0.033898305,0.06779661,0.06779661,0.06779661,0.03954802,0.10169491,0.06779661,0.084745765,0.096045196,0.07909604,0.17514125,0.17514125,0.14124294,0.096045196,0.19209039,0.073446326,0.16384181,0.20903955,0.096045196,0.21468927,0.259887,0.1299435,0.107344635,0.2881356,0.18079096,0.27118644,0.259887,0.096045196,0.2937853,0.2881356,0.17514125,0.3898305,0.34463277,0.36723164,0.17514125,0.36723164,0.27683616,0.42372882,0.4463277,0.33898306,0.45762712,0.4519774,0.36158192,0.5706215,0.42372882,0.56497175,,0.46892655,0.55932206,0.519774,0.53672314,0.5536723,0.519774,0.6101695,0.6779661,0.7288136,0.6158192,0.6497175,0.8079096,0.8870056,0.8248588,0.8305085,0.88135594,0.8757062,0.9830508,0.9830508,0.98870057
+RandomForest (tuned),0.0056497175,0.016949153,0.028248588,0.011299435,0.028248588,0.011299435,0.033898305,0.04519774,0.016949153,0.016949153,0.033898305,0.11299435,0.028248588,0.118644066,0.084745765,0.06214689,0.073446326,0.10169491,0.096045196,0.16949153,0.06779661,0.17514125,0.107344635,0.23728813,0.1299435,0.12429378,0.19209039,0.13559322,0.26553673,0.14689265,0.20338982,0.17514125,0.24858756,0.20338982,0.31073445,0.26553673,0.23163842,0.27683616,0.37288135,0.26553673,0.28248587,0.13559322,0.31073445,0.40677965,0.19209039,0.30508474,0.3559322,0.4463277,0.44067797,0.38418078,0.3220339,0.34463277,0.47457626,0.29661018,0.49717513,0.53107345,,0.55932206,0.6214689,0.47457626,0.5084746,0.5706215,0.6101695,0.5084746,0.69491524,0.55932206,0.79096043,0.8700565,0.7627119,0.88135594,0.9378531,0.96045196,0.9491525,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.011299435,0.011299435,0.011299435,0.011299435,0.033898305,0.0056497175,0.011299435,0.016949153,0.016949153,0.028248588,0.02259887,0.056497175,0.03954802,0.06779661,0.09039548,0.04519774,0.056497175,0.03954802,0.03954802,0.14689265,0.084745765,0.11299435,0.050847456,0.15819208,0.033898305,0.14689265,0.19774011,0.06214689,0.27118644,0.16384181,0.073446326,0.073446326,0.27683616,0.14124294,0.24858756,0.2542373,0.084745765,0.259887,0.2937853,0.1299435,0.39548022,0.3502825,0.3502825,0.2542373,0.36158192,0.22033899,0.44067797,0.40677965,0.3220339,0.46327683,0.4519774,0.36723164,0.4915254,0.4124294,0.4915254,0.44067797,0.44067797,,0.50282484,0.5536723,0.6101695,0.59322035,0.76836157,0.6101695,0.5875706,0.6440678,0.6440678,0.7175141,0.9548023,0.8248588,0.8700565,0.85310733,0.8926554,0.97740114,0.97740114,0.99435025
+ExtraTrees (default),0.02259887,0.03954802,0.03954802,0.028248588,0.050847456,0.04519774,0.06214689,0.056497175,0.04519774,0.028248588,0.050847456,0.11299435,0.050847456,0.12429378,0.016949153,0.096045196,0.07909604,0.10169491,0.13559322,0.107344635,0.09039548,0.14124294,0.15819208,0.23163842,0.14124294,0.12429378,0.096045196,0.17514125,0.18079096,0.1299435,0.19209039,0.20338982,0.15819208,0.21468927,0.30508474,0.20903955,0.27118644,0.19209039,0.3559322,0.28248587,0.27683616,0.056497175,0.27683616,0.42937854,0.11299435,0.2937853,0.3559322,0.47457626,0.45762712,0.37853107,0.259887,0.29943502,0.48022598,0.259887,0.5084746,0.48022598,0.37853107,0.49717513,,0.37288135,0.44067797,0.5875706,0.5084746,0.40677965,0.6723164,0.43502825,0.68926555,0.91525424,0.6440678,0.72316384,0.9548023,0.8587571,0.97740114,0.99435025,0.99435025,1.0
+XGBoost (default),0.0,0.0056497175,0.011299435,0.0056497175,0.050847456,0.0,0.0056497175,0.02259887,0.0056497175,0.0056497175,0.0056497175,0.056497175,0.0056497175,0.050847456,0.084745765,0.028248588,0.033898305,0.02259887,0.0056497175,0.056497175,0.033898305,0.09039548,0.0056497175,0.096045196,0.028248588,0.107344635,0.06779661,0.011299435,0.19774011,0.11299435,0.03954802,0.17514125,0.24293785,0.06779661,0.18079096,0.107344635,0.23728813,0.118644066,0.20338982,0.27683616,0.2542373,0.4180791,0.2881356,0.37288135,0.46892655,0.2937853,0.32768363,0.40112993,0.42372882,0.40112993,0.38418078,0.24293785,0.63276833,0.47457626,0.63276833,0.46327683,0.5254237,0.4463277,0.62711865,,0.55932206,0.72316384,0.55932206,0.5706215,0.6836158,0.6666667,0.80225986,0.80225986,0.75141245,0.76836157,0.8757062,0.94350284,0.89830506,0.9378531,0.9378531,0.9830508
+FastaiMLP (tuned + ensembled),0.0,0.0056497175,0.0,0.016949153,0.050847456,0.02259887,0.0056497175,0.0,0.0,0.0,0.0056497175,0.028248588,0.016949153,0.033898305,0.084745765,0.028248588,0.02259887,0.0056497175,0.07909604,0.04519774,0.016949153,0.073446326,0.07909604,0.073446326,0.033898305,0.10169491,0.15254237,0.06779661,0.19209039,0.09039548,0.07909604,0.10169491,0.27683616,0.10169491,0.16384181,0.19209039,0.17514125,0.16384181,0.19774011,0.20903955,0.20903955,0.27118644,0.18644068,0.3502825,0.31638417,0.30508474,0.2259887,0.3502825,0.40677965,0.34463277,0.28248587,0.27683616,0.5423729,0.46327683,0.5423729,0.4463277,0.4915254,0.3898305,0.55932206,0.44067797,,0.6101695,0.4463277,0.5875706,0.66101694,0.5762712,0.6836158,0.81920904,0.69491524,0.9378531,0.8587571,0.920904,0.8587571,0.9717514,0.9717514,1.0
+xRFM (default),0.011299435,0.033898305,0.04519774,0.03954802,0.033898305,0.028248588,0.033898305,0.084745765,0.050847456,0.03954802,0.04519774,0.12429378,0.03954802,0.107344635,0.06779661,0.11299435,0.06779661,0.09039548,0.10169491,0.056497175,0.06214689,0.084745765,0.107344635,0.16384181,0.14124294,0.1299435,0.033898305,0.14124294,0.12429378,0.096045196,0.13559322,0.2259887,0.12429378,0.14689265,0.22033899,0.107344635,0.2937853,0.12429378,0.23728813,0.30508474,0.20903955,0.24293785,0.19209039,0.39548022,0.24858756,0.3502825,0.259887,0.33898306,0.4124294,0.29943502,0.22033899,0.2542373,0.5423729,0.3898305,0.5423729,0.48022598,0.42937854,0.40677965,0.4124294,0.27683616,0.3898305,,0.42372882,0.40112993,0.6497175,0.4463277,0.56497175,0.8418079,0.5819209,0.65536726,0.9039548,0.779661,0.9717514,0.9039548,0.8926554,0.98870057
+PerpetualBooster (tuned),0.0056497175,0.0056497175,0.011299435,0.011299435,0.033898305,0.0056497175,0.0056497175,0.016949153,0.016949153,0.016949153,0.02259887,0.04519774,0.016949153,0.056497175,0.06779661,0.033898305,0.033898305,0.04519774,0.02259887,0.09039548,0.050847456,0.10169491,0.033898305,0.13559322,0.033898305,0.13559322,0.18644068,0.03954802,0.2259887,0.1299435,0.07909604,0.056497175,0.23728813,0.11299435,0.23163842,0.2259887,0.09039548,0.22033899,0.27683616,0.12429378,0.31638417,0.2881356,0.28248587,0.23163842,0.30508474,0.17514125,0.3559322,0.40112993,0.27683616,0.3898305,0.3559322,0.31073445,0.44067797,0.36158192,0.44067797,0.3898305,0.3898305,0.23163842,0.4915254,0.44067797,0.5536723,0.5762712,,0.5706215,0.5480226,0.59322035,0.6158192,0.66101694,0.7881356,0.740113,0.84745765,0.8079096,0.8700565,0.9717514,0.9717514,0.98870057
+TorchMLP (default),0.0,0.0,0.0,0.0,0.050847456,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0056497175,0.011299435,0.0,0.073446326,0.02259887,0.011299435,0.0056497175,0.06779661,0.011299435,0.028248588,0.06214689,0.073446326,0.03954802,0.107344635,0.118644066,0.07909604,0.118644066,0.13559322,0.11299435,0.107344635,0.19774011,0.16949153,0.14689265,0.073446326,0.096045196,0.20903955,0.14689265,0.107344635,0.2259887,0.107344635,0.36158192,0.23728813,0.28248587,0.38418078,0.259887,0.073446326,0.37853107,0.29943502,0.15254237,0.30508474,0.2542373,0.5536723,0.47457626,0.5536723,0.3220339,0.4915254,0.3898305,0.59322035,0.42937854,0.4124294,0.59887004,0.42937854,,0.6440678,0.48022598,0.6836158,0.740113,0.5762712,0.59322035,0.8135593,0.8135593,0.8248588,0.8587571,0.8587571,0.9830508
+TabSTAR (default),0.073446326,0.09039548,0.09039548,0.096045196,0.084745765,0.07909604,0.084745765,0.10169491,0.107344635,0.12429378,0.10169491,0.13559322,0.096045196,0.1299435,0.16949153,0.15819208,0.14124294,0.15254237,0.16384181,0.18079096,0.16949153,0.10169491,0.17514125,0.16949153,0.19209039,0.15254237,0.2259887,0.20338982,0.24858756,0.2259887,0.20338982,0.17514125,0.28248587,0.21468927,0.20338982,0.24858756,0.18079096,0.26553673,0.20903955,0.20903955,0.22033899,0.24293785,0.27683616,0.19209039,0.24858756,0.22033899,0.20903955,0.2937853,0.20338982,0.21468927,0.259887,0.27683616,0.02259887,0.29943502,0.03954802,0.27118644,0.30508474,0.4124294,0.32768363,0.31638417,0.33898306,0.3502825,0.4519774,0.3559322,,0.33333334,0.38418078,0.62711865,0.47457626,0.66101694,0.6666667,0.5536723,0.740113,0.779661,0.779661,0.79096043
+FastaiMLP (tuned),0.0,0.0,0.0,0.011299435,0.050847456,0.011299435,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.016949153,0.02259887,0.07909604,0.011299435,0.011299435,0.0,0.03954802,0.03954802,0.02259887,0.056497175,0.050847456,0.02259887,0.011299435,0.10169491,0.118644066,0.02259887,0.15254237,0.07909604,0.03954802,0.084745765,0.2259887,0.056497175,0.084745765,0.14689265,0.16949153,0.14124294,0.10169491,0.18644068,0.15254237,0.28248587,0.14124294,0.27118644,0.2937853,0.26553673,0.1299435,0.31638417,0.31073445,0.2259887,0.22033899,0.20903955,0.5423729,0.40112993,0.5480226,0.38418078,0.44067797,0.3559322,0.56497175,0.33333334,0.42372882,0.5536723,0.40677965,0.519774,0.6666667,,0.6836158,0.78531075,0.6440678,0.7966102,0.81920904,0.8870056,0.8418079,0.94350284,0.9378531,1.0
+RandomForest (default),0.0056497175,0.011299435,0.0056497175,0.0056497175,0.033898305,0.0056497175,0.0,0.011299435,0.0,0.0056497175,0.011299435,0.04519774,0.0056497175,0.06779661,0.031073445,0.011299435,0.016949153,0.028248588,0.016949153,0.050847456,0.02259887,0.076271184,0.028248588,0.07909604,0.033898305,0.056497175,0.028248588,0.02259887,0.10451978,0.04519774,0.04519774,0.073446326,0.09322034,0.06214689,0.14124294,0.050847456,0.073446326,0.07909604,0.18644068,0.14124294,0.16384181,0.04519774,0.11299435,0.259887,0.06214689,0.15254237,0.23163842,0.33898306,0.29943502,0.27683616,0.096045196,0.14689265,0.47457626,0.011299435,0.49717513,0.3502825,0.20903955,0.3559322,0.31073445,0.19774011,0.31638417,0.43502825,0.38418078,0.31638417,0.6158192,0.31638417,,0.7627119,0.5536723,0.6045198,0.8418079,0.7740113,0.89830506,0.94350284,0.94350284,0.97740114
+KNN (tuned + ensembled),0.016949153,0.02259887,0.03954802,0.028248588,0.02259887,0.016949153,0.028248588,0.03954802,0.02259887,0.03954802,0.02259887,0.084745765,0.02259887,0.096045196,0.016949153,0.06214689,0.03954802,0.07909604,0.06779661,0.03954802,0.04519774,0.056497175,0.07909604,0.11299435,0.096045196,0.050847456,0.03954802,0.09039548,0.107344635,0.056497175,0.107344635,0.13559322,0.09039548,0.11299435,0.14689265,0.06214689,0.14689265,0.118644066,0.20338982,0.15819208,0.107344635,0.033898305,0.096045196,0.17514125,0.03954802,0.16384181,0.17514125,0.23163842,0.18644068,0.20338982,0.10169491,0.18079096,0.27683616,0.096045196,0.27683616,0.19209039,0.1299435,0.28248587,0.084745765,0.19774011,0.18079096,0.15819208,0.33898306,0.259887,0.37288135,0.21468927,0.23728813,,0.43502825,0.40112993,0.91525424,0.32768363,0.9661017,0.84745765,0.84745765,0.9661017
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.016949153,0.011299435,0.011299435,0.011299435,0.0,0.028248588,0.016949153,0.011299435,0.0056497175,0.028248588,0.0,0.016949153,0.056497175,0.016949153,0.09039548,0.056497175,0.016949153,0.0,0.15819208,0.016949153,0.084745765,0.07909604,0.011299435,0.084745765,0.09039548,0.016949153,0.16384181,0.18079096,0.14689265,0.050847456,0.16949153,0.028248588,0.14689265,0.2542373,0.06779661,0.20903955,0.18079096,0.084745765,0.40112993,0.21468927,0.40112993,0.11299435,0.23728813,0.04519774,0.3559322,0.24858756,0.30508474,0.4180791,0.21186441,0.42372882,0.5254237,0.3559322,0.4463277,0.56497175,,0.5084746,0.69491524,0.62711865,0.8135593,0.8926554,0.8926554,0.9661017
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.050847456,0.0056497175,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0056497175,0.056497175,0.011299435,0.0,0.0,0.0056497175,0.0,0.0056497175,0.073446326,0.016949153,0.04519774,0.011299435,0.056497175,0.033898305,0.011299435,0.10169491,0.02259887,0.02259887,0.016949153,0.15254237,0.028248588,0.107344635,0.06214689,0.028248588,0.04519774,0.118644066,0.073446326,0.073446326,0.07909604,0.06779661,0.096045196,0.096045196,0.084745765,0.1299435,0.31638417,0.1299435,0.19774011,0.10169491,0.09039548,0.31073445,0.118644066,0.31073445,0.17514125,0.118644066,0.17514125,0.27683616,0.23163842,0.06214689,0.34463277,0.259887,0.40677965,0.33898306,0.20338982,0.39548022,0.59887004,0.4915254,,0.80225986,0.519774,0.84745765,0.92655367,0.92655367,0.99435025
+KNN (tuned),0.0056497175,0.011299435,0.02259887,0.016949153,0.011299435,0.016949153,0.011299435,0.028248588,0.011299435,0.02259887,0.011299435,0.033898305,0.011299435,0.06214689,0.011299435,0.04519774,0.033898305,0.04519774,0.03954802,0.02259887,0.033898305,0.033898305,0.04519774,0.07909604,0.06779661,0.016949153,0.02259887,0.04519774,0.07909604,0.028248588,0.056497175,0.050847456,0.07909604,0.073446326,0.107344635,0.04519774,0.06779661,0.06779661,0.15254237,0.096045196,0.050847456,0.011299435,0.056497175,0.15819208,0.0056497175,0.11299435,0.1299435,0.16384181,0.16384181,0.14124294,0.073446326,0.12429378,0.20338982,0.06214689,0.20903955,0.16949153,0.06214689,0.1299435,0.04519774,0.12429378,0.14124294,0.096045196,0.15254237,0.18644068,0.33333334,0.18079096,0.15819208,0.084745765,0.30508474,0.19774011,,0.24858756,0.89830506,0.8248588,0.8248588,0.8926554
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.011299435,0.0,0.0,0.0,0.0,0.0056497175,0.0056497175,0.011299435,0.0,0.011299435,0.0,0.011299435,0.02259887,0.0,0.050847456,0.0,0.0,0.0,0.096045196,0.0,0.02259887,0.011299435,0.0,0.016949153,0.016949153,0.0,0.04519774,0.0056497175,0.04519774,0.09039548,0.0056497175,0.011299435,0.118644066,0.23728813,0.096045196,0.12429378,0.050847456,0.011299435,0.34463277,0.03954802,0.34463277,0.118644066,0.03954802,0.14689265,0.14124294,0.056497175,0.07909604,0.22033899,0.19209039,0.18644068,0.4463277,0.11299435,0.2259887,0.6723164,0.37288135,0.48022598,0.75141245,,0.7740113,0.779661,0.7740113,0.84745765
+KNN (default),0.011299435,0.02259887,0.011299435,0.02259887,0.011299435,0.011299435,0.011299435,0.016949153,0.011299435,0.011299435,0.011299435,0.03954802,0.011299435,0.050847456,0.011299435,0.028248588,0.028248588,0.033898305,0.033898305,0.011299435,0.011299435,0.016949153,0.03954802,0.06779661,0.03954802,0.016949153,0.02259887,0.04519774,0.06779661,0.02259887,0.04519774,0.03954802,0.06779661,0.06779661,0.07909604,0.033898305,0.050847456,0.073446326,0.14124294,0.073446326,0.033898305,0.011299435,0.04519774,0.11299435,0.011299435,0.096045196,0.118644066,0.06214689,0.14124294,0.14689265,0.04519774,0.10169491,0.16384181,0.03954802,0.16384181,0.12429378,0.050847456,0.107344635,0.02259887,0.10169491,0.14124294,0.028248588,0.1299435,0.17514125,0.259887,0.15819208,0.10169491,0.033898305,0.18644068,0.15254237,0.10169491,0.2259887,,0.5875706,0.59322035,0.6384181
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0056497175,0.011299435,0.0,0.0,0.0,0.0056497175,0.0056497175,0.0,0.011299435,0.0056497175,0.0,0.0,0.0,0.011299435,0.033898305,0.016949153,0.0056497175,0.0,0.02259887,0.011299435,0.033898305,0.0,0.02259887,0.016949153,0.0,0.0,0.09039548,0.084745765,0.02259887,0.12429378,0.050847456,0.016949153,0.21468927,0.0,0.21468927,0.016949153,0.0,0.02259887,0.0056497175,0.06214689,0.028248588,0.096045196,0.028248588,0.14124294,0.22033899,0.056497175,0.056497175,0.15254237,0.107344635,0.073446326,0.17514125,0.22033899,0.4124294,,0.53107345,0.779661
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0056497175,0.011299435,0.0,0.0,0.0,0.0056497175,0.0056497175,0.0,0.011299435,0.0056497175,0.0,0.0,0.0,0.011299435,0.033898305,0.016949153,0.0056497175,0.0,0.02259887,0.016949153,0.033898305,0.0,0.02259887,0.016949153,0.0,0.0,0.09039548,0.084745765,0.02259887,0.12429378,0.050847456,0.016949153,0.21468927,0.0,0.21468927,0.016949153,0.0,0.02259887,0.0056497175,0.06214689,0.028248588,0.107344635,0.028248588,0.14124294,0.22033899,0.06214689,0.056497175,0.15254237,0.107344635,0.073446326,0.17514125,0.2259887,0.40677965,0.46892655,,0.76836157
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.0,0.0,0.0,0.0056497175,0.0056497175,0.0,0.0056497175,0.0,0.0056497175,0.011299435,0.0,0.0,0.0,0.0056497175,0.0,0.0,0.0,0.016949153,0.0056497175,0.0,0.0,0.0056497175,0.107344635,0.0,0.1299435,0.011299435,0.0,0.0056497175,0.0,0.016949153,0.0,0.011299435,0.011299435,0.016949153,0.20903955,0.0,0.02259887,0.033898305,0.033898305,0.0056497175,0.107344635,0.15254237,0.36158192,0.22033899,0.23163842,
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_all/leaderboard_overview_explorer.html
index 2225404d6230b776483cf98a90d1176e855815c1..1777a7bd59240ce8ad48a6b55b2ea96a0fb58bfe 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | All Tasks | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_all/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..5414b3fe190610de360d513263df07404f9f3945
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_all/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1792.0,"score":0.826,"rank":6.0,"hrank":1.64,"imp":4.493,"train_time":40.79,"predict_time":6.991,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-121","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1648.0,"score":0.641,"rank":10.98,"hrank":4.12,"imp":7.231,"train_time":293.65,"predict_time":4.358,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-86","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1642.0,"score":0.624,"rank":11.26,"hrank":4.16,"imp":8.415,"train_time":3.59,"predict_time":0.395,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-64","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1611.0,"score":0.574,"rank":12.73,"hrank":7.17,"imp":9.926,"train_time":5.75,"predict_time":0.6,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-60","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1601.0,"score":0.56,"rank":13.26,"hrank":5.89,"imp":8.809,"train_time":2059.94,"predict_time":9.785,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-58","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1558.0,"score":0.537,"rank":15.57,"hrank":6.36,"imp":10.372,"train_time":1.96,"predict_time":0.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-75","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1546.0,"score":0.506,"rank":16.31,"hrank":9.1,"imp":10.735,"train_time":2059.94,"predict_time":1.03,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-60","imputed":false},{"position":7,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1525.0,"score":0.466,"rank":17.63,"hrank":7.92,"imp":11.558,"train_time":1754.94,"predict_time":1.766,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-58","imputed":false},{"position":8,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1522.0,"score":0.464,"rank":17.76,"hrank":10.79,"imp":10.918,"train_time":5.72,"predict_time":0.611,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-48","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1486.0,"score":0.422,"rank":20.18,"hrank":13.01,"imp":12.456,"train_time":2791.97,"predict_time":13.886,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-40","imputed":false},{"position":10,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1421.0,"score":0.374,"rank":25.0,"hrank":8.59,"imp":13.442,"train_time":6155.04,"predict_time":386.158,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-56","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1414.0,"score":0.348,"rank":25.6,"hrank":14.79,"imp":13.806,"train_time":2791.97,"predict_time":0.373,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-57","imputed":false},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1413.0,"score":0.351,"rank":25.66,"hrank":13.07,"imp":14.119,"train_time":2.02,"predict_time":0.183,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-45","imputed":false},{"position":13,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1399.0,"score":0.329,"rank":26.76,"hrank":13.74,"imp":14.115,"train_time":2462.02,"predict_time":1.988,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-45","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1394.0,"score":0.267,"rank":27.14,"hrank":20.62,"imp":14.979,"train_time":416.63,"predict_time":2.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-42","imputed":false},{"position":15,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1389.0,"score":0.287,"rank":27.57,"hrank":15.21,"imp":13.991,"train_time":1347.32,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-42","imputed":false},{"position":16,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1388.0,"score":0.289,"rank":27.63,"hrank":17.63,"imp":14.535,"train_time":12741.09,"predict_time":397.882,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-48","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1387.0,"score":0.277,"rank":27.74,"hrank":17.94,"imp":14.256,"train_time":1347.32,"predict_time":0.364,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-37","imputed":false},{"position":18,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1377.0,"score":0.326,"rank":28.56,"hrank":13.92,"imp":14.855,"train_time":6155.04,"predict_time":39.452,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-53","imputed":false},{"position":19,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1366.0,"score":0.31,"rank":29.51,"hrank":13.27,"imp":15.246,"train_time":4618.5,"predict_time":7.735,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-59","imputed":false},{"position":20,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1364.0,"score":0.287,"rank":29.72,"hrank":16.48,"imp":14.801,"train_time":2462.02,"predict_time":0.231,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-51","imputed":false},{"position":21,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1355.0,"score":0.315,"rank":30.47,"hrank":9.03,"imp":14.641,"train_time":27.33,"predict_time":6.11,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-66","imputed":false},{"position":22,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1353.0,"score":0.231,"rank":30.6,"hrank":22.02,"imp":15.543,"train_time":700.96,"predict_time":1.438,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-45","imputed":false},{"position":23,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1339.0,"score":0.218,"rank":31.78,"hrank":24.94,"imp":15.852,"train_time":416.63,"predict_time":0.381,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-47","imputed":false},{"position":24,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1337.0,"score":0.222,"rank":31.98,"hrank":21.31,"imp":15.619,"train_time":5.81,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-43","imputed":false},{"position":25,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1335.0,"score":0.251,"rank":32.15,"hrank":18.51,"imp":15.843,"train_time":4618.5,"predict_time":0.47,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-57","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1332.0,"score":0.213,"rank":32.41,"hrank":17.49,"imp":15.697,"train_time":700.96,"predict_time":0.213,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-48","imputed":false},{"position":27,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1330.0,"score":0.247,"rank":32.56,"hrank":20.02,"imp":15.811,"train_time":1.19,"predict_time":0.07,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-51","imputed":false},{"position":28,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1324.0,"score":0.243,"rank":33.16,"hrank":20.72,"imp":15.793,"train_time":866.12,"predict_time":2.007,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-47","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1322.0,"score":0.227,"rank":33.28,"hrank":22.94,"imp":17.012,"train_time":463.3,"predict_time":0.366,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-63","imputed":false},{"position":30,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1314.0,"score":0.277,"rank":34.02,"hrank":15.68,"imp":16.606,"train_time":2943.39,"predict_time":17.365,"imputed_pct":35.29,"hardware":"GPU","elo_ci":"+69\/-70","imputed":true},{"position":31,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1294.0,"score":0.174,"rank":35.76,"hrank":26.52,"imp":16.167,"train_time":12741.09,"predict_time":60.101,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+46\/-43","imputed":false},{"position":32,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1292.0,"score":0.203,"rank":36.0,"hrank":22.09,"imp":17.402,"train_time":463.3,"predict_time":0.031,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-63","imputed":false},{"position":33,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1291.0,"score":0.233,"rank":36.04,"hrank":18.19,"imp":17.376,"train_time":87.36,"predict_time":2.432,"imputed_pct":35.29,"hardware":"GPU","elo_ci":"+53\/-64","imputed":true},{"position":34,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1279.0,"score":0.179,"rank":37.12,"hrank":19.02,"imp":17.09,"train_time":866.12,"predict_time":0.097,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-43","imputed":false},{"position":35,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1275.0,"score":0.229,"rank":37.46,"hrank":19.13,"imp":17.37,"train_time":45.42,"predict_time":39.405,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-69","imputed":false},{"position":36,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1273.0,"score":0.179,"rank":37.68,"hrank":24.83,"imp":17.086,"train_time":7.81,"predict_time":0.237,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+46\/-52","imputed":false},{"position":37,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1272.0,"score":0.208,"rank":37.71,"hrank":14.57,"imp":17.26,"train_time":6.86,"predict_time":1.52,"imputed_pct":29.41,"hardware":"GPU","elo_ci":"+56\/-56","imputed":true},{"position":38,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1256.0,"score":0.146,"rank":39.18,"hrank":26.72,"imp":18.296,"train_time":1877.68,"predict_time":0.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-49","imputed":false},{"position":39,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1252.0,"score":0.193,"rank":39.53,"hrank":22.76,"imp":18.086,"train_time":2943.39,"predict_time":0.262,"imputed_pct":35.29,"hardware":"GPU","elo_ci":"+69\/-56","imputed":true},{"position":40,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1248.0,"score":0.138,"rank":39.92,"hrank":32.37,"imp":17.188,"train_time":2832.93,"predict_time":1.801,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-53","imputed":false},{"position":41,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1242.0,"score":0.118,"rank":40.45,"hrank":30.24,"imp":17.875,"train_time":10.44,"predict_time":1.714,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+42\/-50","imputed":false},{"position":42,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1238.0,"score":0.181,"rank":40.79,"hrank":20.64,"imp":18.865,"train_time":13.96,"predict_time":2.081,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-60","imputed":false},{"position":43,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1237.0,"score":0.169,"rank":40.89,"hrank":26.8,"imp":18.888,"train_time":203.0,"predict_time":1.155,"imputed_pct":25.49,"hardware":"GPU","elo_ci":"+62\/-49","imputed":true},{"position":44,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1236.0,"score":0.144,"rank":41.02,"hrank":30.47,"imp":18.852,"train_time":2.07,"predict_time":0.033,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-58","imputed":false},{"position":45,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1220.0,"score":0.171,"rank":42.44,"hrank":19.79,"imp":18.555,"train_time":3.27,"predict_time":0.315,"imputed_pct":35.29,"hardware":"GPU","elo_ci":"+64\/-74","imputed":true},{"position":46,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1214.0,"score":0.117,"rank":42.93,"hrank":19.81,"imp":19.857,"train_time":13.74,"predict_time":0.316,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-48","imputed":false},{"position":47,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1207.0,"score":0.109,"rank":43.59,"hrank":30.81,"imp":19.145,"train_time":1877.68,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-51","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1196.0,"score":0.1,"rank":44.53,"hrank":32.22,"imp":19.624,"train_time":257.64,"predict_time":0.846,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-47","imputed":false},{"position":49,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1186.0,"score":0.106,"rank":45.5,"hrank":22.23,"imp":19.856,"train_time":6.9,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-57","imputed":false},{"position":50,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1185.0,"score":0.107,"rank":45.56,"hrank":34.62,"imp":18.978,"train_time":2832.93,"predict_time":0.112,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-59","imputed":false},{"position":51,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1184.0,"score":0.1,"rank":45.67,"hrank":22.5,"imp":18.885,"train_time":2.06,"predict_time":0.122,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-55","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1167.0,"score":0.08,"rank":47.17,"hrank":33.1,"imp":20.833,"train_time":257.64,"predict_time":0.068,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-59","imputed":false},{"position":53,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1166.0,"score":0.093,"rank":47.25,"hrank":35.24,"imp":20.53,"train_time":594.95,"predict_time":4.649,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-68","imputed":false},{"position":54,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1165.0,"score":0.189,"rank":47.3,"hrank":12.78,"imp":22.73,"train_time":0.53,"predict_time":0.077,"imputed_pct":74.51,"hardware":"GPU","elo_ci":"+90\/-84","imputed":true},{"position":55,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1160.0,"score":0.078,"rank":47.73,"hrank":38.61,"imp":20.735,"train_time":377.14,"predict_time":0.747,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-53","imputed":false},{"position":56,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1155.0,"score":0.171,"rank":48.17,"hrank":17.79,"imp":22.999,"train_time":0.53,"predict_time":0.077,"imputed_pct":74.51,"hardware":"GPU","elo_ci":"+92\/-81","imputed":true},{"position":57,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1153.0,"score":0.073,"rank":48.32,"hrank":41.38,"imp":19.567,"train_time":2.2,"predict_time":0.171,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-49","imputed":false},{"position":58,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1121.0,"score":0.045,"rank":51.09,"hrank":44.86,"imp":21.491,"train_time":377.14,"predict_time":0.091,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-45","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1102.0,"score":0.056,"rank":52.67,"hrank":39.13,"imp":22.073,"train_time":594.95,"predict_time":0.336,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-71","imputed":false},{"position":60,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1080.0,"score":0.063,"rank":54.46,"hrank":33.04,"imp":23.091,"train_time":301.0,"predict_time":65.827,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-67","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1080.0,"score":0.067,"rank":54.48,"hrank":44.73,"imp":26.914,"train_time":176.26,"predict_time":0.499,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-64","imputed":false},{"position":62,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1074.0,"score":0.104,"rank":54.96,"hrank":22.06,"imp":25.607,"train_time":39935.25,"predict_time":4.288,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-85","imputed":false},{"position":63,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1072.0,"score":0.1,"rank":55.13,"hrank":23.37,"imp":25.72,"train_time":39935.25,"predict_time":20.166,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-92","imputed":false},{"position":64,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1048.0,"score":0.028,"rank":56.94,"hrank":50.54,"imp":23.884,"train_time":12.4,"predict_time":2.558,"imputed_pct":25.49,"hardware":"GPU","elo_ci":"+51\/-51","imputed":true},{"position":65,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1038.0,"score":0.023,"rank":57.69,"hrank":52.22,"imp":23.585,"train_time":8.96,"predict_time":0.129,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-60","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1035.0,"score":0.035,"rank":57.94,"hrank":52.93,"imp":28.237,"train_time":176.26,"predict_time":0.191,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-60","imputed":false},{"position":67,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1028.0,"score":0.053,"rank":58.45,"hrank":48.0,"imp":26.514,"train_time":3.14,"predict_time":0.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-74","imputed":false},{"position":68,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1005.0,"score":0.027,"rank":60.1,"hrank":51.77,"imp":26.027,"train_time":1.97,"predict_time":0.251,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-68","imputed":false},{"position":69,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.012,"rank":60.43,"hrank":53.29,"imp":26.194,"train_time":0.43,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-54","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":980.0,"score":0.028,"rank":61.81,"hrank":54.04,"imp":28.4,"train_time":0.8,"predict_time":0.119,"imputed_pct":25.49,"hardware":"GPU","elo_ci":"+69\/-76","imputed":true},{"position":71,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":977.0,"score":0.02,"rank":61.96,"hrank":57.5,"imp":25.757,"train_time":3.12,"predict_time":0.312,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-74","imputed":false},{"position":72,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":975.0,"score":0.073,"rank":62.1,"hrank":22.12,"imp":31.188,"train_time":398.79,"predict_time":4.645,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-108","imputed":false},{"position":73,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":968.0,"score":0.025,"rank":62.56,"hrank":54.23,"imp":27.723,"train_time":129.17,"predict_time":1.627,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-80","imputed":false},{"position":74,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":938.0,"score":0.026,"rank":64.4,"hrank":44.52,"imp":33.278,"train_time":22.76,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-86","imputed":false},{"position":75,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":901.0,"score":0.025,"rank":66.55,"hrank":28.76,"imp":34.284,"train_time":240.78,"predict_time":0.308,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-102","imputed":false},{"position":76,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":878.0,"score":0.018,"rank":67.72,"hrank":40.7,"imp":34.833,"train_time":240.78,"predict_time":0.068,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-105","imputed":false},{"position":77,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":819.0,"score":0.014,"rank":70.47,"hrank":66.66,"imp":33.594,"train_time":129.17,"predict_time":0.103,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-85","imputed":false},{"position":78,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":817.0,"score":0.006,"rank":70.55,"hrank":66.61,"imp":37.335,"train_time":1.23,"predict_time":0.115,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-107","imputed":false},{"position":79,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":603.0,"score":0.0,"rank":76.88,"hrank":76.48,"imp":45.902,"train_time":0.19,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-103","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_all/pareto_front_explorer.html
index 4508af24c1511d64940ef99c9e849971d3db2589..3016675d15fcac330cb0d06c9160aa6cef85bebf 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.1550205733,"imp":18.8876343208,"elo":1237.1,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"imp":15.6194075121,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0367019752,"imp":13.9914099321,"elo":1389.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3641930223,"imp":14.2555340647,"elo":1387.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0331429177,"imp":18.8522253844,"elo":1235.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306783583,"imp":17.4020216494,"elo":1291.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"imp":17.0122029193,"elo":1322.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0148501396,"imp":19.855862201,"elo":1185.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0139622688,"imp":19.145448228,"elo":1207.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1396506436,"imp":18.2964633633,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2510258327,"imp":26.0270390708,"elo":1004.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0677914174,"imp":20.8330990233,"elo":1166.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8460764958,"imp":19.624332108,"elo":1196.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.311955429,"imp":25.7573864963,"elo":977.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3360688368,"imp":22.072750532,"elo":1102.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.6489525795,"imp":20.5299873145,"elo":1165.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1711427769,"imp":19.5668776787,"elo":1153.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3811195827,"imp":15.8518389049,"elo":1339.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.2357814014,"imp":14.9790313591,"elo":1394.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.109529595,"imp":14.6408751857,"elo":1354.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.4321623466,"imp":17.3758218583,"elo":1291.1,"imputed":true,"imputed_pct":35.2941176471,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3157602832,"imp":19.8567761429,"elo":1214.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4695293307,"imp":15.8434491121,"elo":1335.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.7345001255,"imp":15.2459076632,"elo":1366.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.0774082329,"imp":22.9988553982,"elo":1155.2,"imputed":true,"imputed_pct":74.5098039216,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0774082329,"imp":22.729974018,"elo":1165.1,"imputed":true,"imputed_pct":74.5098039216,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5580293662,"imp":23.8840016642,"elo":1048.1,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0248147768,"imp":33.2776711455,"elo":938.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.191170079,"imp":28.2368808154,"elo":1034.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.499326448,"imp":26.9142672963,"elo":1079.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"imp":26.1937438859,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0914129487,"imp":21.4906341641,"elo":1121.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7469116355,"imp":20.7351011577,"elo":1160.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7143460095,"imp":17.8752414114,"elo":1242.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3727607921,"imp":13.8058109888,"elo":1413.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":13.8860562126,"imp":12.4562280645,"elo":1486.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6105561478,"imp":10.9177336683,"elo":1522.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0296972468,"imp":10.7353852445,"elo":1545.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7853653133,"imp":8.80858319,"elo":1600.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0805747376,"imp":18.8647866237,"elo":1238.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"imp":17.3704817754,"elo":1275.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"imp":14.8552991246,"elo":1377.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":386.1584680055,"imp":13.4419394329,"elo":1421.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"imp":14.1192694637,"elo":1412.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.9905568694,"imp":4.49265048,"elo":1791.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.118779525,"imp":28.4000103255,"elo":979.6,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.5195481527,"imp":17.2596666726,"elo":1272.5,"imputed":true,"imputed_pct":29.4117647059,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1463864614,"imp":10.3717098275,"elo":1558.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2369687803,"imp":17.0863242178,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2313083312,"imp":14.8014608188,"elo":1363.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.9881175398,"imp":14.1148939312,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6004352741,"imp":9.9257506719,"elo":1611.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.394910574,"imp":8.4152952196,"elo":1642.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3150187952,"imp":18.5545741771,"elo":1219.9,"imputed":true,"imputed_pct":35.2941176471,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2619314173,"imp":18.0860645344,"elo":1252.2,"imputed":true,"imputed_pct":35.2941176471,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.3650169889,"imp":16.6061454671,"elo":1313.8,"imputed":true,"imputed_pct":35.2941176471,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.6449420594,"imp":31.1884740696,"elo":975.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2876390828,"imp":25.6074091524,"elo":1073.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.1657011965,"imp":25.7200316315,"elo":1071.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0695707729,"imp":15.8110836345,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1288563013,"imp":23.5850289886,"elo":1038.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1115292311,"imp":18.9780169078,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8008738756,"imp":17.1879342911,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"imp":18.8846861164,"elo":1183.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2127694574,"imp":15.6967764254,"elo":1332.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.4381002715,"imp":15.5426569196,"elo":1353.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":65.8265809722,"imp":23.0909171752,"elo":1079.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":60.1008510867,"imp":16.1671093544,"elo":1294.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":397.8816919076,"imp":14.5350218759,"elo":1388.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7405088776,"imp":26.5142713324,"elo":1027.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0970074274,"imp":17.0901508832,"elo":1279.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0069414377,"imp":15.7930925499,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.1550205733,"x_train":202.9993903946,"imp":18.8876343208,"elo":1237.1,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"x_train":5.8052200156,"imp":15.6194075121,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0367019752,"x_train":1347.3189819557,"imp":13.9914099321,"elo":1389.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3641930223,"x_train":1347.3189819557,"imp":14.2555340647,"elo":1387.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0331429177,"x_train":2.0744471986,"imp":18.8522253844,"elo":1235.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306783583,"x_train":463.3015284407,"imp":17.4020216494,"elo":1291.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"x_train":463.3015284407,"imp":17.0122029193,"elo":1322.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0148501396,"x_train":6.9024058688,"imp":19.855862201,"elo":1185.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0139622688,"x_train":1877.6824729228,"imp":19.145448228,"elo":1207.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1396506436,"x_train":1877.6824729228,"imp":18.2964633633,"elo":1256.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2510258327,"x_train":1.9744557778,"imp":26.0270390708,"elo":1004.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0677914174,"x_train":257.6385593165,"imp":20.8330990233,"elo":1166.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8460764958,"x_train":257.6385593165,"imp":19.624332108,"elo":1196.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.311955429,"x_train":3.1153293529,"imp":25.7573864963,"elo":977.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3360688368,"x_train":594.9528585231,"imp":22.072750532,"elo":1102.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.6489525795,"x_train":594.9528585231,"imp":20.5299873145,"elo":1165.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1711427769,"x_train":2.2021467311,"imp":19.5668776787,"elo":1153.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.3811195827,"x_train":416.6281847875,"imp":15.8518389049,"elo":1339.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.2357814014,"x_train":416.6281847875,"imp":14.9790313591,"elo":1394.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":6.109529595,"x_train":27.3254634149,"imp":14.6408751857,"elo":1354.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.4321623466,"x_train":87.3574675883,"imp":17.3758218583,"elo":1291.1,"imputed":true,"imputed_pct":35.2941176471,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3157602832,"x_train":13.7418320826,"imp":19.8567761429,"elo":1214.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4695293307,"x_train":4618.5023417328,"imp":15.8434491121,"elo":1335.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.7345001255,"x_train":4618.5023417328,"imp":15.2459076632,"elo":1366.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.0774082329,"x_train":0.5286672115,"imp":22.9988553982,"elo":1155.2,"imputed":true,"imputed_pct":74.5098039216,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0774082329,"x_train":0.5286672115,"imp":22.729974018,"elo":1165.1,"imputed":true,"imputed_pct":74.5098039216,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5580293662,"x_train":12.4012404367,"imp":23.8840016642,"elo":1048.1,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0248147768,"x_train":22.7552921112,"imp":33.2776711455,"elo":938.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.191170079,"x_train":176.2617571354,"imp":28.2368808154,"elo":1034.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.499326448,"x_train":176.2617571354,"imp":26.9142672963,"elo":1079.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":0.431661138,"imp":26.1937438859,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0914129487,"x_train":377.1421152596,"imp":21.4906341641,"elo":1121.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7469116355,"x_train":377.1421152596,"imp":20.7351011577,"elo":1160.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7143460095,"x_train":10.4435312748,"imp":17.8752414114,"elo":1242.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3727607921,"x_train":2791.9683399256,"imp":13.8058109888,"elo":1413.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":13.8860562126,"x_train":2791.9683399256,"imp":12.4562280645,"elo":1486.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6105561478,"x_train":5.7153047813,"imp":10.9177336683,"elo":1522.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0296972468,"x_train":2059.9443792676,"imp":10.7353852445,"elo":1545.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7853653133,"x_train":2059.9443792676,"imp":8.80858319,"elo":1600.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0805747376,"x_train":13.9576562813,"imp":18.8647866237,"elo":1238.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"x_train":45.4161088134,"imp":17.3704817754,"elo":1275.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"x_train":6155.0377202868,"imp":14.8552991246,"elo":1377.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":386.1584680055,"x_train":6155.0377202868,"imp":13.4419394329,"elo":1421.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"x_train":2.0183513985,"imp":14.1192694637,"elo":1412.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.9905568694,"x_train":40.7926562599,"imp":4.49265048,"elo":1791.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.118779525,"x_train":0.8007418442,"imp":28.4000103255,"elo":979.6,"imputed":true,"imputed_pct":25.4901960784,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.5195481527,"x_train":6.8566983121,"imp":17.2596666726,"elo":1272.5,"imputed":true,"imputed_pct":29.4117647059,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1463864614,"x_train":1.9632436902,"imp":10.3717098275,"elo":1558.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2369687803,"x_train":7.8050600165,"imp":17.0863242178,"elo":1272.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2313083312,"x_train":2462.0152230263,"imp":14.8014608188,"elo":1363.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.9881175398,"x_train":2462.0152230263,"imp":14.1148939312,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6004352741,"x_train":5.7520413762,"imp":9.9257506719,"elo":1611.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.394910574,"x_train":3.5877237797,"imp":8.4152952196,"elo":1642.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3150187952,"x_train":3.2658532122,"imp":18.5545741771,"elo":1219.9,"imputed":true,"imputed_pct":35.2941176471,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2619314173,"x_train":2943.3924917132,"imp":18.0860645344,"elo":1252.2,"imputed":true,"imputed_pct":35.2941176471,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.3650169889,"x_train":2943.3924917132,"imp":16.6061454671,"elo":1313.8,"imputed":true,"imputed_pct":35.2941176471,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.6449420594,"x_train":398.7869028722,"imp":31.1884740696,"elo":975.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2876390828,"x_train":39935.2451367739,"imp":25.6074091524,"elo":1073.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.1657011965,"x_train":39935.2451367739,"imp":25.7200316315,"elo":1071.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0695707729,"x_train":1.1859471423,"imp":15.8110836345,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1288563013,"x_train":8.9576391549,"imp":23.5850289886,"elo":1038.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1115292311,"x_train":2832.9262089828,"imp":18.9780169078,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8008738756,"x_train":2832.9262089828,"imp":17.1879342911,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"x_train":2.0592641478,"imp":18.8846861164,"elo":1183.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2127694574,"x_train":700.9607998747,"imp":15.6967764254,"elo":1332.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.4381002715,"x_train":700.9607998747,"imp":15.5426569196,"elo":1353.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":65.8265809722,"x_train":300.998886807,"imp":23.0909171752,"elo":1079.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":60.1008510867,"x_train":12741.0939157936,"imp":16.1671093544,"elo":1294.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":397.8816919076,"x_train":12741.0939157936,"imp":14.5350218759,"elo":1388.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7405088776,"x_train":3.1442091702,"imp":26.5142713324,"elo":1027.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0970074274,"x_train":866.1169883185,"imp":17.0901508832,"elo":1279.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0069414377,"x_train":866.1169883185,"imp":15.7930925499,"elo":1323.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_all/tuning_trajectories_explorer.html
index cbc9c2896ce5d20d503021e6f420b99a5e2b4e0f..7657d6d0bbf9c1ca58509b6d6f860ba1fe307366 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_all/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":202.9993903946,"imp":18.9182060614,"elo":1251.6,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":5.8052200156,"imp":15.6465469683,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":11.6250818675,"imp":15.1249440765,"elo":1375.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":30.7197244584,"imp":14.7585670862,"elo":1395.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":60.1926505566,"imp":14.5746677832,"elo":1403.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":159.5146818194,"imp":14.4512537257,"elo":1408.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":318.9223225532,"imp":14.4418775922,"elo":1406.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":662.5706287402,"imp":14.3366421221,"elo":1404.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":998.7349671254,"imp":14.3380320849,"elo":1400.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1347.3189819557,"imp":14.2727326515,"elo":1404.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.0744471986,"imp":18.881761217,"elo":1247.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.099236637,"imp":18.3013566752,"elo":1281.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.8289007778,"imp":17.8983629942,"elo":1304.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.1099632833,"imp":17.654680276,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":57.3124085491,"imp":17.341388682,"elo":1328.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":115.9554885729,"imp":17.1603030396,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":231.6358590441,"imp":17.0648259018,"elo":1341.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":346.5027373053,"imp":17.0206764957,"elo":1344.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":463.3015284407,"imp":17.0361398634,"elo":1345.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":6.9024058688,"imp":19.881088258,"elo":1192.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":16.7298961864,"imp":19.3942755583,"elo":1212.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":47.9305716086,"imp":19.0014805865,"elo":1235.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":98.1075414046,"imp":18.8546190738,"elo":1242.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":242.3014674264,"imp":18.6770210947,"elo":1253.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":485.7841387908,"imp":18.5386980706,"elo":1260.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":938.1843487506,"imp":18.3941001625,"elo":1268.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1390.5258657356,"imp":18.3551475433,"elo":1272.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1877.6824729228,"imp":18.3162246395,"elo":1273.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.9744557778,"imp":26.0492293327,"elo":1009.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1741914103,"imp":24.2843992181,"elo":1049.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.7533015627,"imp":21.6227649006,"elo":1116.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":13.1877719851,"imp":20.7297206666,"elo":1150.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":32.4660623442,"imp":20.0308571538,"elo":1177.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":64.6497454444,"imp":19.8058067482,"elo":1188.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":128.5728546231,"imp":19.6323964618,"elo":1197.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":192.4203093077,"imp":19.5365639827,"elo":1203.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":257.6385593165,"imp":19.6756636178,"elo":1203.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.1153293529,"imp":25.7771875463,"elo":980.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.1374994683,"imp":24.876756423,"elo":1026.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":15.2752739191,"imp":23.1456861482,"elo":1090.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":30.4682284412,"imp":22.3925614597,"elo":1117.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":73.2357604794,"imp":21.4581896548,"elo":1139.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":146.7179006005,"imp":21.0275364314,"elo":1154.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":295.2364780788,"imp":20.742587909,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":443.050159936,"imp":20.5808957716,"elo":1169.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":594.9528585231,"imp":20.5609253512,"elo":1175.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.2021467311,"imp":19.5941493563,"elo":1156.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":4.7004585926,"imp":17.8899738637,"elo":1241.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":10.0203601171,"imp":16.6036769991,"elo":1324.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":19.9125130873,"imp":15.9867838326,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":50.9228061571,"imp":15.4180491992,"elo":1389.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":102.592415423,"imp":15.2316218238,"elo":1403.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":206.3698310558,"imp":15.1011915051,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":309.4092129514,"imp":15.0186734919,"elo":1415.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":416.6281847875,"imp":15.0059838127,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":27.3254634149,"imp":14.7874216629,"elo":1372.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":87.3574675883,"imp":17.399881616,"elo":1310.3,"imputed":true,"imputed_pct":35.2941176471,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":13.7418320826,"imp":19.8826245808,"elo":1223.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":40.2965934723,"imp":18.4871232731,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":115.9419475491,"imp":16.3972267359,"elo":1341.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":237.5304229618,"imp":15.8865961157,"elo":1366.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":577.047955689,"imp":15.3810958072,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1164.1941811379,"imp":15.3385354816,"elo":1388.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2285.7499058023,"imp":15.4823428051,"elo":1381.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3416.2270282822,"imp":15.3837854078,"elo":1385.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4618.5023417328,"imp":15.2891807581,"elo":1386.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":0.5286672115,"imp":23.0228189989,"elo":1159.4,"imputed":true,"imputed_pct":74.5098039216,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":0.5286672115,"imp":22.753998208,"elo":1169.3,"imputed":true,"imputed_pct":74.5098039216,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_train":12.4012404367,"imp":23.7880581414,"elo":1061.0,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":22.7552921112,"imp":33.3074772071,"elo":947.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":51.1368372531,"imp":30.2900412798,"elo":1012.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":147.0385313693,"imp":27.0810195816,"elo":1080.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":176.2617571354,"imp":26.9179045669,"elo":1086.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.431661138,"imp":26.227750944,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.2883199842,"imp":23.5363108451,"elo":1066.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":7.9454413682,"imp":22.1253378349,"elo":1105.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":17.3814840956,"imp":21.4146723267,"elo":1125.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":45.2922595201,"imp":20.8867925277,"elo":1154.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":91.6104975483,"imp":20.7685716253,"elo":1161.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":185.2817035188,"imp":20.7593791652,"elo":1165.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":279.7219448516,"imp":20.7714489618,"elo":1165.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":377.1421152596,"imp":20.7586444281,"elo":1165.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.4435312748,"imp":17.8971758422,"elo":1248.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.7804646251,"imp":15.6216855555,"elo":1348.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":65.9805881846,"imp":13.980886962,"elo":1432.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":135.870476352,"imp":13.3384610939,"elo":1456.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":350.3376696757,"imp":12.7919049544,"elo":1485.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":701.5287510764,"imp":12.5237519135,"elo":1502.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1392.1340578984,"imp":12.4280702709,"elo":1507.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2087.8411572447,"imp":12.4691295389,"elo":1512.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2791.9683399256,"imp":12.4879188143,"elo":1514.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.7153047813,"imp":10.9500292678,"elo":1536.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":17.0263004767,"imp":10.8431774194,"elo":1541.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":47.4652638957,"imp":10.577852835,"elo":1554.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":98.3276706028,"imp":10.4847043524,"elo":1573.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":249.0037143861,"imp":10.2616081896,"elo":1592.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":505.0675823061,"imp":10.118360411,"elo":1596.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1022.195993405,"imp":9.6606861106,"elo":1611.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1532.1033434601,"imp":9.3909922704,"elo":1618.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2059.9443792676,"imp":8.858293049,"elo":1621.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.9576562813,"imp":18.8880527947,"elo":1251.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":45.4161088134,"imp":17.3918101211,"elo":1289.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":79.4617813141,"imp":16.4208969649,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":160.0616335809,"imp":15.5193530321,"elo":1367.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":301.1906399516,"imp":15.1221468536,"elo":1384.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":750.5793401849,"imp":14.4277805507,"elo":1403.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1518.8392119437,"imp":13.959828143,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3050.4213050061,"imp":13.7357066119,"elo":1429.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":4599.9264282578,"imp":13.5915488253,"elo":1433.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":6155.0377202868,"imp":13.48028436,"elo":1438.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0183513985,"imp":14.1442730464,"elo":1430.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":40.7926562599,"imp":4.5226678587,"elo":1811.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.8007418442,"imp":28.4164769475,"elo":989.3,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":6.8566983121,"imp":17.2886133715,"elo":1286.3,"imputed":true,"imputed_pct":29.4117647059,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.9632436902,"imp":10.4032316807,"elo":1579.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":7.8050600165,"imp":17.1130725794,"elo":1287.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":18.1240740517,"imp":16.6424250938,"elo":1309.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":58.9610972546,"imp":15.9844944369,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":118.8818809271,"imp":15.4001592093,"elo":1370.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":303.352703028,"imp":14.9397708906,"elo":1390.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":608.2074565649,"imp":14.6327422125,"elo":1407.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1220.3134501457,"imp":14.341329658,"elo":1417.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1834.4356979489,"imp":14.2393340233,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2462.0152230263,"imp":14.1464902517,"elo":1425.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.7520413762,"imp":9.9561341223,"elo":1639.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.5877237797,"imp":8.4098317476,"elo":1658.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3.2658532122,"imp":18.5780430326,"elo":1229.6,"imputed":true,"imputed_pct":35.2941176471,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":23.1992334667,"imp":18.1260893345,"elo":1257.1,"imputed":true,"imputed_pct":35.2941176471,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":63.4230955303,"imp":17.5529908583,"elo":1290.2,"imputed":true,"imputed_pct":35.2941176471,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":140.6313481987,"imp":17.1557673958,"elo":1305.8,"imputed":true,"imputed_pct":35.2941176471,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":348.9467316628,"imp":16.704835099,"elo":1328.9,"imputed":true,"imputed_pct":35.2941176471,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":710.5040634902,"imp":16.5175267856,"elo":1339.2,"imputed":true,"imputed_pct":35.2941176471,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1446.907380806,"imp":16.4798979148,"elo":1339.8,"imputed":true,"imputed_pct":35.2941176471,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2180.6569876176,"imp":16.5528775198,"elo":1336.4,"imputed":true,"imputed_pct":35.2941176471,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2943.3924917132,"imp":16.6395566907,"elo":1328.9,"imputed":true,"imputed_pct":35.2941176471,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":398.7869028722,"imp":31.2030628488,"elo":982.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":634.250223124,"imp":31.0556732029,"elo":992.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1221.7141616423,"imp":30.6541210546,"elo":1002.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2110.4203991936,"imp":29.8409344122,"elo":1018.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5146.9875310979,"imp":28.3244288538,"elo":1041.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":10543.907094521,"imp":27.0209182939,"elo":1059.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":19963.1285723042,"imp":26.4784409815,"elo":1070.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":29831.9301413401,"imp":26.1087331647,"elo":1080.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":39935.2451367739,"imp":25.7429181154,"elo":1092.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1859471423,"imp":15.8418345309,"elo":1351.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":8.9576391549,"imp":23.6014659413,"elo":1044.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":22.6590123875,"imp":21.4720166086,"elo":1103.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":65.5045736145,"imp":19.6242101382,"elo":1159.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":143.026937766,"imp":18.8561951171,"elo":1195.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":340.3179128497,"imp":18.1568769857,"elo":1224.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":710.8916252159,"imp":17.7750180092,"elo":1240.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1385.1316080566,"imp":17.5263503874,"elo":1249.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2123.8601198211,"imp":17.329141063,"elo":1259.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2832.9262089828,"imp":17.2121890207,"elo":1264.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.0592641478,"imp":18.9130825235,"elo":1189.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.635768988,"imp":17.1075757375,"elo":1282.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":17.5495862649,"imp":16.5287609183,"elo":1320.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":33.8580856565,"imp":16.1650880109,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.7445970232,"imp":15.8785433209,"elo":1355.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":173.5566472908,"imp":15.7656953435,"elo":1364.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":346.7594994462,"imp":15.6708900588,"elo":1371.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":520.510860917,"imp":15.5925892546,"elo":1375.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":700.9607998747,"imp":15.5659679182,"elo":1376.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":300.998886807,"imp":23.1137358616,"elo":1093.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":757.2419712235,"imp":20.0983627483,"elo":1172.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2344.8516752833,"imp":18.2976688332,"elo":1281.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4981.7847745692,"imp":16.9278957609,"elo":1338.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12170.8681485011,"imp":14.5443390737,"elo":1408.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12741.0939157936,"imp":14.5705715956,"elo":1408.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.1442091702,"imp":26.5318377702,"elo":1038.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":7.5774846269,"imp":23.2161937325,"elo":1104.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":20.6232394062,"imp":19.181375086,"elo":1216.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":42.5478154764,"imp":17.40302461,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":107.8487563604,"imp":16.5656082959,"elo":1319.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":214.7680889693,"imp":16.1981624387,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":431.3100440381,"imp":15.964231707,"elo":1337.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":645.974194479,"imp":15.8613673923,"elo":1340.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":866.1169883185,"imp":15.7565168682,"elo":1343.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.1550205733,"x_train":202.9993903946,"imp":18.9182060614,"elo":1251.6,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"x_train":5.8052200156,"imp":15.6465469683,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0601795968,"x_train":11.6250818675,"imp":15.1249440765,"elo":1375.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1181383737,"x_train":30.7197244584,"imp":14.7585670862,"elo":1395.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1643410418,"x_train":60.1926505566,"imp":14.5746677832,"elo":1403.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2285526113,"x_train":159.5146818194,"imp":14.4512537257,"elo":1408.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2603250146,"x_train":318.9223225532,"imp":14.4418775922,"elo":1406.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3044187588,"x_train":662.5706287402,"imp":14.3366421221,"elo":1404.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3313311934,"x_train":998.7349671254,"imp":14.3380320849,"elo":1400.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3641930223,"x_train":1347.3189819557,"imp":14.2727326515,"elo":1404.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0331429177,"x_train":2.0744471986,"imp":18.881761217,"elo":1247.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0646948782,"x_train":4.099236637,"imp":18.3013566752,"elo":1281.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1097672871,"x_train":10.8289007778,"imp":17.8983629942,"elo":1304.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1712030423,"x_train":22.1099632833,"imp":17.654680276,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2532193674,"x_train":57.3124085491,"imp":17.341388682,"elo":1328.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3030166676,"x_train":115.9554885729,"imp":17.1603030396,"elo":1337.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3373890654,"x_train":231.6358590441,"imp":17.0648259018,"elo":1341.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3569505185,"x_train":346.5027373053,"imp":17.0206764957,"elo":1344.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"x_train":463.3015284407,"imp":17.0361398634,"elo":1345.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0148501396,"x_train":6.9024058688,"imp":19.881088258,"elo":1192.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0303468055,"x_train":16.7298961864,"imp":19.3942755583,"elo":1212.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0551923837,"x_train":47.9305716086,"imp":19.0014805865,"elo":1235.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0698156568,"x_train":98.1075414046,"imp":18.8546190738,"elo":1242.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0981563113,"x_train":242.3014674264,"imp":18.6770210947,"elo":1253.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1132271626,"x_train":485.7841387908,"imp":18.5386980706,"elo":1260.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1244815695,"x_train":938.1843487506,"imp":18.3941001625,"elo":1268.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1306053559,"x_train":1390.5258657356,"imp":18.3551475433,"elo":1272.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1402144831,"x_train":1877.6824729228,"imp":18.3162246395,"elo":1273.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2510258327,"x_train":1.9744557778,"imp":26.0492293327,"elo":1009.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2910788221,"x_train":3.1741914103,"imp":24.2843992181,"elo":1049.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3725395347,"x_train":6.7533015627,"imp":21.6227649006,"elo":1116.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4132860205,"x_train":13.1877719851,"imp":20.7297206666,"elo":1150.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5777882664,"x_train":32.4660623442,"imp":20.0308571538,"elo":1177.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7270465219,"x_train":64.6497454444,"imp":19.8058067482,"elo":1188.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.83253734,"x_train":128.5728546231,"imp":19.6323964618,"elo":1197.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8229876046,"x_train":192.4203093077,"imp":19.5365639827,"elo":1203.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8460764958,"x_train":257.6385593165,"imp":19.6756636178,"elo":1203.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.311955429,"x_train":3.1153293529,"imp":25.7771875463,"elo":980.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.599233021,"x_train":6.1374994683,"imp":24.876756423,"elo":1026.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1109333389,"x_train":15.2752739191,"imp":23.1456861482,"elo":1090.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7115436462,"x_train":30.4682284412,"imp":22.3925614597,"elo":1117.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4877540535,"x_train":73.2357604794,"imp":21.4581896548,"elo":1139.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.2290986651,"x_train":146.7179006005,"imp":21.0275364314,"elo":1154.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.9684851932,"x_train":295.2364780788,"imp":20.742587909,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.1837657267,"x_train":443.050159936,"imp":20.5808957716,"elo":1169.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.6489525795,"x_train":594.9528585231,"imp":20.5609253512,"elo":1175.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1711427769,"x_train":2.2021467311,"imp":19.5941493563,"elo":1156.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6899284291,"x_train":4.7004585926,"imp":17.8899738637,"elo":1241.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9547904471,"x_train":10.0203601171,"imp":16.6036769991,"elo":1324.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.415717788,"x_train":19.9125130873,"imp":15.9867838326,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5465698671,"x_train":50.9228061571,"imp":15.4180491992,"elo":1389.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6454746717,"x_train":102.592415423,"imp":15.2316218238,"elo":1403.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.9962776452,"x_train":206.3698310558,"imp":15.1011915051,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1744811378,"x_train":309.4092129514,"imp":15.0186734919,"elo":1415.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2357814014,"x_train":416.6281847875,"imp":15.0059838127,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":6.109529595,"x_train":27.3254634149,"imp":14.7874216629,"elo":1372.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.4321623466,"x_train":87.3574675883,"imp":17.399881616,"elo":1310.3,"imputed":true,"imputed_pct":35.2941176471,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3157602832,"x_train":13.7418320826,"imp":19.8826245808,"elo":1223.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8658037861,"x_train":40.2965934723,"imp":18.4871232731,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7616896461,"x_train":115.9419475491,"imp":16.3972267359,"elo":1341.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7495673171,"x_train":237.5304229618,"imp":15.8865961157,"elo":1366.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.9629243753,"x_train":577.047955689,"imp":15.3810958072,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1800785058,"x_train":1164.1941811379,"imp":15.3385354816,"elo":1388.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.7233625262,"x_train":2285.7499058023,"imp":15.4823428051,"elo":1381.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.6697936774,"x_train":3416.2270282822,"imp":15.3837854078,"elo":1385.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.7345001255,"x_train":4618.5023417328,"imp":15.2891807581,"elo":1386.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.0774082329,"x_train":0.5286672115,"imp":23.0228189989,"elo":1159.4,"imputed":true,"imputed_pct":74.5098039216,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0774082329,"x_train":0.5286672115,"imp":22.753998208,"elo":1169.3,"imputed":true,"imputed_pct":74.5098039216,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5580293662,"x_train":12.4012404367,"imp":23.7880581414,"elo":1061.0,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0248147768,"x_train":22.7552921112,"imp":33.3074772071,"elo":947.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1956514835,"x_train":51.1368372531,"imp":30.2900412798,"elo":1012.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4510510662,"x_train":147.0385313693,"imp":27.0810195816,"elo":1080.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.499326448,"x_train":176.2617571354,"imp":26.9179045669,"elo":1086.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":0.431661138,"imp":26.227750944,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1195246673,"x_train":2.2883199842,"imp":23.5363108451,"elo":1066.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2293352158,"x_train":7.9454413682,"imp":22.1253378349,"elo":1105.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3451275073,"x_train":17.3814840956,"imp":21.4146723267,"elo":1125.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4910539573,"x_train":45.2922595201,"imp":20.8867925277,"elo":1154.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5816243107,"x_train":91.6104975483,"imp":20.7685716253,"elo":1161.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.664675779,"x_train":185.2817035188,"imp":20.7593791652,"elo":1165.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7215840165,"x_train":279.7219448516,"imp":20.7714489618,"elo":1165.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7462800839,"x_train":377.1421152596,"imp":20.7586444281,"elo":1165.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7143460095,"x_train":10.4435312748,"imp":17.8971758422,"elo":1248.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.7127272606,"x_train":24.7804646251,"imp":15.6216855555,"elo":1348.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.9943069673,"x_train":65.9805881846,"imp":13.980886962,"elo":1432.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.4327546965,"x_train":135.870476352,"imp":13.3384610939,"elo":1456.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.3277160444,"x_train":350.3376696757,"imp":12.7919049544,"elo":1485.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.3527013663,"x_train":701.5287510764,"imp":12.5237519135,"elo":1502.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.7064755441,"x_train":1392.1340578984,"imp":12.4280702709,"elo":1507.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.274656038,"x_train":2087.8411572447,"imp":12.4691295389,"elo":1512.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.8860562126,"x_train":2791.9683399256,"imp":12.4879188143,"elo":1514.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6105561478,"x_train":5.7153047813,"imp":10.9500292678,"elo":1536.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3877879479,"x_train":17.0263004767,"imp":10.8431774194,"elo":1541.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8738328207,"x_train":47.4652638957,"imp":10.577852835,"elo":1554.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1588746477,"x_train":98.3276706028,"imp":10.4847043524,"elo":1573.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.8281839931,"x_train":249.0037143861,"imp":10.2616081896,"elo":1592.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4738495327,"x_train":505.0675823061,"imp":10.118360411,"elo":1596.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6074456326,"x_train":1022.195993405,"imp":9.6606861106,"elo":1611.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.8784737047,"x_train":1532.1033434601,"imp":9.3909922704,"elo":1618.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7107599974,"x_train":2059.9443792676,"imp":8.858293049,"elo":1621.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":2.0805747376,"x_train":13.9576562813,"imp":18.8880527947,"elo":1251.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"x_train":45.4161088134,"imp":17.3918101211,"elo":1289.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":56.7537152129,"x_train":79.4617813141,"imp":16.4208969649,"elo":1324.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":89.1044883635,"x_train":160.0616335809,"imp":15.5193530321,"elo":1367.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":119.8919914946,"x_train":301.1906399516,"imp":15.1221468536,"elo":1384.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":187.2342742423,"x_train":750.5793401849,"imp":14.4277805507,"elo":1403.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":245.0072603318,"x_train":1518.8392119437,"imp":13.959828143,"elo":1419.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":313.8851452232,"x_train":3050.4213050061,"imp":13.7357066119,"elo":1429.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":354.6149052654,"x_train":4599.9264282578,"imp":13.5915488253,"elo":1433.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":387.8903208002,"x_train":6155.0377202868,"imp":13.48028436,"elo":1438.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"x_train":2.0183513985,"imp":14.1442730464,"elo":1430.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.9905568694,"x_train":40.7926562599,"imp":4.5226678587,"elo":1811.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.118779525,"x_train":0.8007418442,"imp":28.4164769475,"elo":989.3,"imputed":true,"imputed_pct":25.4901960784,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.5195481527,"x_train":6.8566983121,"imp":17.2886133715,"elo":1286.3,"imputed":true,"imputed_pct":29.4117647059,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1463864614,"x_train":1.9632436902,"imp":10.4032316807,"elo":1579.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2369687803,"x_train":7.8050600165,"imp":17.1130725794,"elo":1287.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.439953166,"x_train":18.1240740517,"imp":16.6424250938,"elo":1309.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9304964542,"x_train":58.9610972546,"imp":15.9844944369,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.898947224,"x_train":118.8818809271,"imp":15.4001592093,"elo":1370.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.1135689245,"x_train":303.352703028,"imp":14.9397708906,"elo":1390.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.5576850953,"x_train":608.2074565649,"imp":14.6327422125,"elo":1407.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.7400707788,"x_train":1220.3134501457,"imp":14.341329658,"elo":1417.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.9026564546,"x_train":1834.4356979489,"imp":14.2393340233,"elo":1423.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":1.9881175398,"x_train":2462.0152230263,"imp":14.1464902517,"elo":1425.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6004352741,"x_train":5.7520413762,"imp":9.9561341223,"elo":1639.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.394910574,"x_train":3.5877237797,"imp":8.4098317476,"elo":1658.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3150187952,"x_train":3.2658532122,"imp":18.5780430326,"elo":1229.6,"imputed":true,"imputed_pct":35.2941176471,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":2.8573639225,"x_train":23.1992334667,"imp":18.1260893345,"elo":1257.1,"imputed":true,"imputed_pct":35.2941176471,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.4265055798,"x_train":63.4230955303,"imp":17.5529908583,"elo":1290.2,"imputed":true,"imputed_pct":35.2941176471,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":7.5843389189,"x_train":140.6313481987,"imp":17.1557673958,"elo":1305.8,"imputed":true,"imputed_pct":35.2941176471,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":11.6381793283,"x_train":348.9467316628,"imp":16.704835099,"elo":1328.9,"imputed":true,"imputed_pct":35.2941176471,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":15.3774900763,"x_train":710.5040634902,"imp":16.5175267856,"elo":1339.2,"imputed":true,"imputed_pct":35.2941176471,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.7203326823,"x_train":1446.907380806,"imp":16.4798979148,"elo":1339.8,"imputed":true,"imputed_pct":35.2941176471,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.5386193489,"x_train":2180.6569876176,"imp":16.5528775198,"elo":1336.4,"imputed":true,"imputed_pct":35.2941176471,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.8212823247,"x_train":2943.3924917132,"imp":16.6395566907,"elo":1328.9,"imputed":true,"imputed_pct":35.2941176471,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.6449420594,"x_train":398.7869028722,"imp":31.2030628488,"elo":982.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.0436884998,"x_train":634.250223124,"imp":31.0556732029,"elo":992.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.2115756111,"x_train":1221.7141616423,"imp":30.6541210546,"elo":1002.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.7979635353,"x_train":2110.4203991936,"imp":29.8409344122,"elo":1018.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.002547622,"x_train":5146.9875310979,"imp":28.3244288538,"elo":1041.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.7064222686,"x_train":10543.907094521,"imp":27.0209182939,"elo":1059.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.6908292471,"x_train":19963.1285723042,"imp":26.4784409815,"elo":1070.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.7650487513,"x_train":29831.9301413401,"imp":26.1087331647,"elo":1080.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.1657011965,"x_train":39935.2451367739,"imp":25.7429181154,"elo":1092.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0695707729,"x_train":1.1859471423,"imp":15.8418345309,"elo":1351.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1288563013,"x_train":8.9576391549,"imp":23.6014659413,"elo":1044.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2470592409,"x_train":22.6590123875,"imp":21.4720166086,"elo":1103.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4546102434,"x_train":65.5045736145,"imp":19.6242101382,"elo":1159.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7104475927,"x_train":143.026937766,"imp":18.8561951171,"elo":1195.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0905794242,"x_train":340.3179128497,"imp":18.1568769857,"elo":1224.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3321372143,"x_train":710.8916252159,"imp":17.7750180092,"elo":1240.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6375611361,"x_train":1385.1316080566,"imp":17.5263503874,"elo":1249.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.8070112948,"x_train":2123.8601198211,"imp":17.329141063,"elo":1259.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.9084608704,"x_train":2832.9262089828,"imp":17.2121890207,"elo":1264.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"x_train":2.0592641478,"imp":18.9130825235,"elo":1189.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.560565097,"x_train":5.635768988,"imp":17.1075757375,"elo":1282.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0675882643,"x_train":17.5495862649,"imp":16.5287609183,"elo":1320.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1769638293,"x_train":33.8580856565,"imp":16.1650880109,"elo":1341.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4785625458,"x_train":86.7445970232,"imp":15.8785433209,"elo":1355.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.6035368187,"x_train":173.5566472908,"imp":15.7656953435,"elo":1364.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7255766262,"x_train":346.7594994462,"imp":15.6708900588,"elo":1371.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.5600405676,"x_train":520.510860917,"imp":15.5925892546,"elo":1375.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4381002715,"x_train":700.9607998747,"imp":15.5659679182,"elo":1376.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":65.8265809722,"x_train":300.998886807,"imp":23.1137358616,"elo":1093.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":133.265500669,"x_train":757.2419712235,"imp":20.0983627483,"elo":1172.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":232.2933397079,"x_train":2344.8516752833,"imp":18.2976688332,"elo":1281.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":345.85504867,"x_train":4981.7847745692,"imp":16.9278957609,"elo":1338.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":412.1802530188,"x_train":12170.8681485011,"imp":14.5443390737,"elo":1408.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":397.8816919076,"x_train":12741.0939157936,"imp":14.5705715956,"elo":1408.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.7405088776,"x_train":3.1442091702,"imp":26.5318377702,"elo":1038.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8845279741,"x_train":7.5774846269,"imp":23.2161937325,"elo":1104.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1069120467,"x_train":20.6232394062,"imp":19.181375086,"elo":1216.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.5245468853,"x_train":42.5478154764,"imp":17.40302461,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.6856549267,"x_train":107.8487563604,"imp":16.5656082959,"elo":1319.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.157086811,"x_train":214.7680889693,"imp":16.1981624387,"elo":1334.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.8867738381,"x_train":431.3100440381,"imp":15.964231707,"elo":1337.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.1655042904,"x_train":645.974194479,"imp":15.8613673923,"elo":1340.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9734199806,"x_train":866.1169883185,"imp":15.7565168682,"elo":1343.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..c6ef5d1cf4308a4dff0a98cfd1491bc8fb34fc57
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_all/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (default)", "RealMLP (tuned + ensembled)", "TabDPT (tuned + ensembled)", "RealMLP (tuned)", "TabDPT-Turbo (default)", "TabM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "CatBoost (tuned)", "iLTM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "TabDPT (tuned)", "ModernNCA (tuned + ensembled)", "TabM (tuned)", "LimiX (default)", "XGBoost (tuned + ensembled)", "LightGBM (tuned)", "CatBoost (default)", "ModernNCA (tuned)", "XGBoost (tuned)", "TabSwift (default)", "xRFM (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "TabPFNv2 (tuned + ensembled)", "iLTM (tuned)", "ChimeraBoost (tuned)", "Mitra (default)", "xRFM (tuned)", "TabDPT (default)", "TabM (default)", "TabICL (default)", "EBM (tuned + ensembled)", "TabPFNv2 (tuned)", "TorchMLP (tuned + ensembled)", "RealMLP (default)", "SAP-RPT-OSS (default)", "BetaTabPFN (default)", "ChimeraBoost (default)", "TabPFNv2 (default)", "ModernNCA (default)", "EBM (tuned)", "ExtraTrees (tuned + ensembled)", "EBM (default)", "TorchMLP (tuned)", "XGBoost (default)", "ExtraTrees (tuned)", "FastaiMLP (tuned + ensembled)", "Nori-30M (default)", "RandomForest (tuned + ensembled)", "Nori (default)", "LightGBM (default)", "RandomForest (tuned)", "FastaiMLP (tuned)", "iLTM (default)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "OrionMSP (default)", "TorchMLP (default)", "PerpetualBooster (tuned)", "xRFM (default)", "ExtraTrees (default)", "RandomForest (default)", "TabFlex (default)", "FastaiMLP (default)", "TabSTAR (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "KNN (tuned)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.7254902124404907, 0.8039215803146362, 0.843137264251709, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.9019607901573181, 0.8823529481887817, 0.9411764740943909, 0.9411764740943909, 0.9411764740943909, 0.9215686321258545, 0.8627451062202454, 0.9411764740943909, 0.9607843160629272, 0.8823529481887817, 0.9607843160629272, 0.9019607901573181, 0.9215686321258545, 0.9607843160629272, 0.9019607901573181, 0.9215686321258545, 0.8823529481887817, 0.8823529481887817, 0.9411764740943909, 0.9215686321258545, 0.9607843160629272, 0.9411764740943909, 0.9019607901573181, 1.0, 0.9411764740943909, 0.9803921580314636, 0.9411764740943909, 0.9411764740943909, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.9215686321258545, 0.9215686321258545, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.9607843160629272, 0.9215686321258545, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9411764740943909, 1.0, 0.9411764740943909, 0.9215686321258545, 0.9803921580314636, 0.9803921580314636, 1.0, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0, 0.9607843160629272, 0.9411764740943909, 1.0, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0], [0.27450981736183167, NaN, 0.45098039507865906, 0.6078431606292725, 0.5686274766921997, 0.5098039507865906, 0.6666666865348816, 0.7450980544090271, 0.6078431606292725, 0.8235294222831726, 0.7450980544090271, 0.843137264251709, 0.7450980544090271, 0.8627451062202454, 0.9215686321258545, 0.8823529481887817, 0.9019607901573181, 0.9019607901573181, 0.7647058963775635, 0.8627451062202454, 0.843137264251709, 0.7647058963775635, 0.9411764740943909, 0.9411764740943909, 0.9411764740943909, 0.9215686321258545, 0.9411764740943909, 0.843137264251709, 0.9215686321258545, 0.9411764740943909, 0.7843137383460999, 0.9215686321258545, 0.9411764740943909, 0.843137264251709, 0.9019607901573181, 0.8235294222831726, 0.9019607901573181, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.9411764740943909, 0.9019607901573181, 0.8627451062202454, 0.8823529481887817, 0.9607843160629272, 0.843137264251709, 0.9607843160629272, 0.9019607901573181, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.9411764740943909, 0.9411764740943909, 0.9019607901573181, 0.8039215803146362, 0.9607843160629272, 0.8627451062202454, 0.9803921580314636, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9019607901573181, 0.8823529481887817, 1.0, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9019607901573181, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 0.9803921580314636, 1.0], [0.19607843458652496, 0.5490196347236633, NaN, 0.5686274766921997, 0.5588235259056091, 0.6274510025978088, 0.686274528503418, 0.6274510025978088, 0.6470588445663452, 0.7450980544090271, 0.7058823704719543, 0.7647058963775635, 0.843137264251709, 0.843137264251709, 0.8823529481887817, 0.7647058963775635, 0.843137264251709, 0.8039215803146362, 0.8235294222831726, 0.9019607901573181, 0.9019607901573181, 0.7647058963775635, 0.8627451062202454, 0.9411764740943909, 0.8627451062202454, 0.843137264251709, 0.843137264251709, 0.8823529481887817, 0.8627451062202454, 0.8823529481887817, 0.8235294222831726, 0.9215686321258545, 0.8627451062202454, 0.8823529481887817, 0.9215686321258545, 0.9215686321258545, 0.8823529481887817, 0.9215686321258545, 0.8235294222831726, 0.9215686321258545, 0.9019607901573181, 0.9411764740943909, 0.843137264251709, 0.9411764740943909, 0.9019607901573181, 0.843137264251709, 0.9607843160629272, 0.8823529481887817, 0.9607843160629272, 0.9019607901573181, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9019607901573181, 0.8823529481887817, 0.9411764740943909, 0.9019607901573181, 0.9607843160629272, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9607843160629272, 0.9019607901573181, 0.9019607901573181, 1.0, 0.9607843160629272, 1.0, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9607843160629272, 1.0, 0.9019607901573181, 0.9607843160629272, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0], [0.1568627506494522, 0.3921568691730499, 0.4313725531101227, NaN, 0.47058823704719543, 0.5098039507865906, 0.5490196347236633, 0.5490196347236633, 0.686274528503418, 0.7254902124404907, 0.7058823704719543, 0.8039215803146362, 0.7647058963775635, 0.7843137383460999, 0.8823529481887817, 0.8039215803146362, 0.843137264251709, 0.843137264251709, 0.7254902124404907, 0.8627451062202454, 0.8235294222831726, 0.8039215803146362, 0.8235294222831726, 0.9411764740943909, 0.9019607901573181, 0.843137264251709, 0.8627451062202454, 0.8235294222831726, 0.843137264251709, 0.8627451062202454, 0.8039215803146362, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.8627451062202454, 0.7843137383460999, 0.8823529481887817, 0.9019607901573181, 0.9019607901573181, 0.9019607901573181, 0.9215686321258545, 0.9215686321258545, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.8235294222831726, 0.9607843160629272, 0.9411764740943909, 0.9411764740943909, 0.9019607901573181, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9215686321258545, 0.8235294222831726, 0.9607843160629272, 0.8627451062202454, 0.9607843160629272, 0.9411764740943909, 0.9215686321258545, 0.9607843160629272, 0.9803921580314636, 0.8823529481887817, 0.8823529481887817, 0.9803921580314636, 0.9803921580314636, 1.0, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9411764740943909, 0.9803921580314636, 0.8823529481887817, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0], [0.13725490868091583, 0.4313725531101227, 0.44117647409439087, 0.529411792755127, NaN, 0.5490196347236633, 0.7254902124404907, 0.6078431606292725, 0.7058823704719543, 0.7450980544090271, 0.6666666865348816, 0.7647058963775635, 0.7058823704719543, 0.7254902124404907, 0.843137264251709, 0.8235294222831726, 0.7843137383460999, 0.8627451062202454, 0.7254902124404907, 0.7647058963775635, 0.8039215803146362, 0.7450980544090271, 0.8627451062202454, 0.8823529481887817, 0.9019607901573181, 0.8627451062202454, 0.8627451062202454, 0.8235294222831726, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.843137264251709, 0.8823529481887817, 0.8823529481887817, 0.8823529481887817, 0.7843137383460999, 0.8823529481887817, 0.8039215803146362, 0.8823529481887817, 0.9607843160629272, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.8823529481887817, 0.8823529481887817, 0.9411764740943909, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 0.9411764740943909, 0.8823529481887817, 0.843137264251709, 0.9215686321258545, 0.8823529481887817, 0.9803921580314636, 0.9411764740943909, 0.9411764740943909, 0.9215686321258545, 0.9607843160629272, 0.8823529481887817, 0.8823529481887817, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9411764740943909, 0.9803921580314636, 0.9803921580314636, 0.9215686321258545, 0.9803921580314636, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 1.0], [0.11764705926179886, 0.4901960790157318, 0.37254902720451355, 0.4901960790157318, 0.45098039507865906, NaN, 0.6274510025978088, 0.6666666865348816, 0.5882353186607361, 0.6666666865348816, 0.7254902124404907, 0.686274528503418, 0.7647058963775635, 0.7254902124404907, 0.8235294222831726, 0.7647058963775635, 0.7843137383460999, 0.7647058963775635, 0.7450980544090271, 0.8235294222831726, 0.7450980544090271, 0.7450980544090271, 0.7843137383460999, 0.843137264251709, 0.8039215803146362, 0.8039215803146362, 0.7843137383460999, 0.8235294222831726, 0.7843137383460999, 0.8235294222831726, 0.7450980544090271, 0.7843137383460999, 0.8039215803146362, 0.8039215803146362, 0.8235294222831726, 0.7843137383460999, 0.8627451062202454, 0.8235294222831726, 0.7843137383460999, 0.8039215803146362, 0.8627451062202454, 0.843137264251709, 0.8235294222831726, 0.9019607901573181, 0.8235294222831726, 0.8627451062202454, 0.9019607901573181, 0.843137264251709, 0.8823529481887817, 0.9019607901573181, 0.8823529481887817, 0.8627451062202454, 0.9215686321258545, 0.9411764740943909, 0.8039215803146362, 0.8823529481887817, 0.8039215803146362, 0.9215686321258545, 0.8823529481887817, 0.9215686321258545, 0.8823529481887817, 0.9215686321258545, 0.9019607901573181, 0.9019607901573181, 0.9215686321258545, 0.9607843160629272, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.9411764740943909, 0.9607843160629272, 0.9803921580314636, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0], [0.11764705926179886, 0.3333333432674408, 0.3137255012989044, 0.45098039507865906, 0.27450981736183167, 0.37254902720451355, NaN, 0.5490196347236633, 0.6176470518112183, 0.6470588445663452, 0.6274510025978088, 0.7450980544090271, 0.686274528503418, 0.7647058963775635, 0.7450980544090271, 0.7450980544090271, 0.686274528503418, 0.7647058963775635, 0.686274528503418, 0.7058823704719543, 0.8235294222831726, 0.6470588445663452, 0.8039215803146362, 0.8627451062202454, 0.8627451062202454, 0.8235294222831726, 0.8235294222831726, 0.7450980544090271, 0.7843137383460999, 0.7450980544090271, 0.7156862616539001, 0.8627451062202454, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.7647058963775635, 0.8627451062202454, 0.8235294222831726, 0.8627451062202454, 0.9215686321258545, 0.8627451062202454, 0.8627451062202454, 0.843137264251709, 0.8627451062202454, 0.843137264251709, 0.843137264251709, 0.9019607901573181, 0.8627451062202454, 0.9607843160629272, 0.8823529481887817, 0.9215686321258545, 0.9411764740943909, 0.9019607901573181, 0.8627451062202454, 0.843137264251709, 0.9215686321258545, 0.8627451062202454, 0.9607843160629272, 0.9215686321258545, 0.9215686321258545, 0.9019607901573181, 0.9607843160629272, 0.8627451062202454, 0.843137264251709, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9019607901573181, 0.9607843160629272, 0.9019607901573181, 0.9215686321258545, 0.9607843160629272, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 1.0], [0.1764705926179886, 0.2549019753932953, 0.37254902720451355, 0.45098039507865906, 0.3921568691730499, 0.3333333432674408, 0.45098039507865906, NaN, 0.45098039507865906, 0.5490196347236633, 0.529411792755127, 0.7058823704719543, 0.6078431606292725, 0.6666666865348816, 0.8235294222831726, 0.7647058963775635, 0.7843137383460999, 0.8039215803146362, 0.5882353186607361, 0.6470588445663452, 0.7450980544090271, 0.6470588445663452, 0.8627451062202454, 0.8627451062202454, 0.8235294222831726, 0.7058823704719543, 0.8235294222831726, 0.7058823704719543, 0.8039215803146362, 0.843137264251709, 0.5882353186607361, 0.8235294222831726, 0.8627451062202454, 0.686274528503418, 0.8235294222831726, 0.7450980544090271, 0.8235294222831726, 0.7647058963775635, 0.8039215803146362, 0.686274528503418, 0.9215686321258545, 0.8235294222831726, 0.7647058963775635, 0.8235294222831726, 0.8823529481887817, 0.7647058963775635, 0.9215686321258545, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9411764740943909, 0.9215686321258545, 0.8627451062202454, 0.9215686321258545, 0.7647058963775635, 0.9019607901573181, 0.8235294222831726, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272, 0.8235294222831726, 0.8235294222831726, 0.9411764740943909, 0.9803921580314636, 0.9803921580314636, 0.9215686321258545, 0.9803921580314636, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 0.8823529481887817, 0.9411764740943909, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0, 1.0, 1.0], [0.13725490868091583, 0.3921568691730499, 0.3529411852359772, 0.3137255012989044, 0.29411765933036804, 0.4117647111415863, 0.38235294818878174, 0.5490196347236633, NaN, 0.5686274766921997, 0.6470588445663452, 0.6666666865348816, 0.6470588445663452, 0.7058823704719543, 0.7450980544090271, 0.7058823704719543, 0.6274510025978088, 0.7058823704719543, 0.6666666865348816, 0.6470588445663452, 0.7450980544090271, 0.7058823704719543, 0.7450980544090271, 0.7647058963775635, 0.7450980544090271, 0.7058823704719543, 0.7843137383460999, 0.843137264251709, 0.8039215803146362, 0.7450980544090271, 0.6470588445663452, 0.8529411554336548, 0.7450980544090271, 0.8039215803146362, 0.843137264251709, 0.7647058963775635, 0.843137264251709, 0.7450980544090271, 0.8039215803146362, 0.8627451062202454, 0.843137264251709, 0.8235294222831726, 0.843137264251709, 0.8627451062202454, 0.843137264251709, 0.9215686321258545, 0.8823529481887817, 0.8823529481887817, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.8823529481887817, 0.8627451062202454, 0.9215686321258545, 0.7843137383460999, 0.8627451062202454, 0.8235294222831726, 0.9607843160629272, 0.9215686321258545, 0.9803921580314636, 0.9215686321258545, 0.9411764740943909, 0.9019607901573181, 0.9019607901573181, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9215686321258545, 0.9803921580314636, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.9215686321258545, 0.9215686321258545, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636], [0.13725490868091583, 0.1764705926179886, 0.2549019753932953, 0.27450981736183167, 0.2549019753932953, 0.3333333432674408, 0.3529411852359772, 0.45098039507865906, 0.4313725531101227, NaN, 0.5098039507865906, 0.686274528503418, 0.5882353186607361, 0.6274510025978088, 0.7058823704719543, 0.7254902124404907, 0.7450980544090271, 0.7843137383460999, 0.5490196347236633, 0.6470588445663452, 0.7058823704719543, 0.5686274766921997, 0.7647058963775635, 0.7647058963775635, 0.7843137383460999, 0.686274528503418, 0.8039215803146362, 0.6666666865348816, 0.7450980544090271, 0.7058823704719543, 0.5882353186607361, 0.8235294222831726, 0.7647058963775635, 0.6666666865348816, 0.8039215803146362, 0.7058823704719543, 0.8235294222831726, 0.7058823704719543, 0.8039215803146362, 0.7254902124404907, 0.843137264251709, 0.9215686321258545, 0.8039215803146362, 0.7843137383460999, 0.8039215803146362, 0.7058823704719543, 0.9019607901573181, 0.843137264251709, 0.8627451062202454, 0.8627451062202454, 0.843137264251709, 0.9019607901573181, 0.8627451062202454, 0.9215686321258545, 0.7647058963775635, 0.843137264251709, 0.7843137383460999, 0.9411764740943909, 0.9019607901573181, 0.9019607901573181, 0.9215686321258545, 0.9607843160629272, 0.843137264251709, 0.843137264251709, 0.9607843160629272, 0.9411764740943909, 0.9803921580314636, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.9215686321258545, 0.9803921580314636, 0.8823529481887817, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0], [0.09803921729326248, 0.2549019753932953, 0.29411765933036804, 0.29411765933036804, 0.3333333432674408, 0.27450981736183167, 0.37254902720451355, 0.47058823704719543, 0.3529411852359772, 0.4901960790157318, NaN, 0.5686274766921997, 0.4901960790157318, 0.529411792755127, 0.5882353186607361, 0.5490196347236633, 0.6470588445663452, 0.5882353186607361, 0.7450980544090271, 0.529411792755127, 0.5490196347236633, 0.6274510025978088, 0.6274510025978088, 0.6078431606292725, 0.6078431606292725, 0.5686274766921997, 0.6274510025978088, 0.6274510025978088, 0.6666666865348816, 0.6274510025978088, 0.6274510025978088, 0.7450980544090271, 0.6274510025978088, 0.686274528503418, 0.7254902124404907, 0.8039215803146362, 0.686274528503418, 0.5882353186607361, 0.6470588445663452, 0.7450980544090271, 0.7450980544090271, 0.7058823704719543, 0.7254902124404907, 0.6666666865348816, 0.6470588445663452, 0.7058823704719543, 0.8235294222831726, 0.7254902124404907, 0.7843137383460999, 0.7843137383460999, 0.7647058963775635, 0.7450980544090271, 0.8627451062202454, 0.7450980544090271, 0.8235294222831726, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.8039215803146362, 0.843137264251709, 0.8039215803146362, 0.843137264251709, 0.8235294222831726, 0.7843137383460999, 0.9607843160629272, 0.8627451062202454, 0.9019607901573181, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.8823529481887817, 0.9215686321258545, 0.9215686321258545, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 0.9215686321258545, 1.0], [0.11764705926179886, 0.1568627506494522, 0.23529411852359772, 0.19607843458652496, 0.23529411852359772, 0.3137255012989044, 0.2549019753932953, 0.29411765933036804, 0.3333333432674408, 0.3137255012989044, 0.4313725531101227, NaN, 0.4117647111415863, 0.5098039507865906, 0.5882353186607361, 0.5490196347236633, 0.529411792755127, 0.5490196347236633, 0.5098039507865906, 0.529411792755127, 0.6078431606292725, 0.5, 0.6666666865348816, 0.6666666865348816, 0.6274510025978088, 0.686274528503418, 0.7058823704719543, 0.5686274766921997, 0.6470588445663452, 0.5686274766921997, 0.5686274766921997, 0.7254902124404907, 0.6274510025978088, 0.5882353186607361, 0.7058823704719543, 0.6274510025978088, 0.7647058963775635, 0.686274528503418, 0.7647058963775635, 0.6274510025978088, 0.8039215803146362, 0.8235294222831726, 0.686274528503418, 0.7647058963775635, 0.686274528503418, 0.686274528503418, 0.7647058963775635, 0.7843137383460999, 0.7647058963775635, 0.8039215803146362, 0.8627451062202454, 0.843137264251709, 0.8235294222831726, 0.843137264251709, 0.7450980544090271, 0.843137264251709, 0.7254902124404907, 0.843137264251709, 0.8823529481887817, 0.8823529481887817, 0.9019607901573181, 0.9019607901573181, 0.843137264251709, 0.8627451062202454, 0.8725489974021912, 0.9607843160629272, 0.9215686321258545, 0.9215686321258545, 0.9019607901573181, 0.9411764740943909, 0.9019607901573181, 0.9607843160629272, 0.8627451062202454, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272, 0.970588207244873, 0.9803921580314636, 0.9607843160629272, 1.0], [0.05882352963089943, 0.2549019753932953, 0.1568627506494522, 0.23529411852359772, 0.29411765933036804, 0.23529411852359772, 0.3137255012989044, 0.3921568691730499, 0.3529411852359772, 0.4117647111415863, 0.5098039507865906, 0.5882353186607361, NaN, 0.47058823704719543, 0.5490196347236633, 0.5490196347236633, 0.6078431606292725, 0.5882353186607361, 0.5490196347236633, 0.5098039507865906, 0.529411792755127, 0.6078431606292725, 0.6666666865348816, 0.5882353186607361, 0.6470588445663452, 0.5882353186607361, 0.686274528503418, 0.6960784196853638, 0.6666666865348816, 0.6666666865348816, 0.6274510025978088, 0.7450980544090271, 0.6666666865348816, 0.6666666865348816, 0.7450980544090271, 0.7254902124404907, 0.7058823704719543, 0.6274510025978088, 0.6666666865348816, 0.7058823704719543, 0.7450980544090271, 0.7058823704719543, 0.7058823704719543, 0.6470588445663452, 0.686274528503418, 0.6666666865348816, 0.7254902124404907, 0.7058823704719543, 0.8235294222831726, 0.7647058963775635, 0.8235294222831726, 0.7843137383460999, 0.843137264251709, 0.7843137383460999, 0.7058823704719543, 0.8235294222831726, 0.7647058963775635, 0.8627451062202454, 0.8627451062202454, 0.9019607901573181, 0.843137264251709, 0.843137264251709, 0.8039215803146362, 0.7843137383460999, 0.9411764740943909, 0.8823529481887817, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.8627451062202454, 0.9411764740943909, 0.9607843160629272, 0.8823529481887817, 0.8823529481887817, 0.9803921580314636, 0.9215686321258545, 1.0], [0.05882352963089943, 0.13725490868091583, 0.1568627506494522, 0.21568627655506134, 0.27450981736183167, 0.27450981736183167, 0.23529411852359772, 0.3333333432674408, 0.29411765933036804, 0.37254902720451355, 0.47058823704719543, 0.4901960790157318, 0.529411792755127, NaN, 0.5882353186607361, 0.4901960790157318, 0.529411792755127, 0.5490196347236633, 0.5098039507865906, 0.5490196347236633, 0.6274510025978088, 0.4901960790157318, 0.5882353186607361, 0.6274510025978088, 0.6470588445663452, 0.6078431606292725, 0.686274528503418, 0.5882353186607361, 0.5686274766921997, 0.6078431606292725, 0.529411792755127, 0.6470588445663452, 0.7254902124404907, 0.5098039507865906, 0.5686274766921997, 0.6470588445663452, 0.8823529481887817, 0.7058823704719543, 0.7843137383460999, 0.5686274766921997, 0.7450980544090271, 0.8235294222831726, 0.7058823704719543, 0.7647058963775635, 0.7450980544090271, 0.6470588445663452, 0.7450980544090271, 0.7647058963775635, 0.7450980544090271, 0.8039215803146362, 0.8039215803146362, 0.7843137383460999, 0.7450980544090271, 0.8235294222831726, 0.6274510025978088, 0.8039215803146362, 0.6666666865348816, 0.8627451062202454, 0.8039215803146362, 0.8823529481887817, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.8627451062202454, 0.9019607901573181, 0.9803921580314636, 0.9019607901573181, 0.843137264251709, 0.8627451062202454, 0.8823529481887817, 0.9411764740943909, 0.9411764740943909, 0.8823529481887817, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 1.0, 0.9607843160629272, 1.0], [0.05882352963089943, 0.0784313753247261, 0.11764705926179886, 0.11764705926179886, 0.1568627506494522, 0.1764705926179886, 0.2549019753932953, 0.1764705926179886, 0.2549019753932953, 0.29411765933036804, 0.4117647111415863, 0.4117647111415863, 0.45098039507865906, 0.4117647111415863, NaN, 0.5490196347236633, 0.5098039507865906, 0.5686274766921997, 0.45098039507865906, 0.47058823704719543, 0.4901960790157318, 0.45098039507865906, 0.5686274766921997, 0.7450980544090271, 0.7254902124404907, 0.5490196347236633, 0.6666666865348816, 0.5490196347236633, 0.5882353186607361, 0.6470588445663452, 0.5098039507865906, 0.6666666865348816, 0.7450980544090271, 0.529411792755127, 0.686274528503418, 0.6078431606292725, 0.7254902124404907, 0.6470588445663452, 0.7450980544090271, 0.5686274766921997, 0.7647058963775635, 0.7254902124404907, 0.6666666865348816, 0.6666666865348816, 0.8039215803146362, 0.6666666865348816, 0.8039215803146362, 0.8235294222831726, 0.843137264251709, 0.8627451062202454, 0.8039215803146362, 0.9019607901573181, 0.8039215803146362, 0.8235294222831726, 0.7450980544090271, 0.8627451062202454, 0.7647058963775635, 0.9607843160629272, 0.9411764740943909, 0.8627451062202454, 0.9215686321258545, 0.9019607901573181, 0.843137264251709, 0.8235294222831726, 0.8823529481887817, 0.9215686321258545, 0.9215686321258545, 0.843137264251709, 0.9215686321258545, 0.9803921580314636, 0.9019607901573181, 0.9215686321258545, 0.8627451062202454, 0.9411764740943909, 0.9215686321258545, 0.9803921580314636, 0.9803921580314636, 1.0, 0.9607843160629272, 1.0], [0.0784313753247261, 0.11764705926179886, 0.23529411852359772, 0.19607843458652496, 0.1764705926179886, 0.23529411852359772, 0.2549019753932953, 0.23529411852359772, 0.29411765933036804, 0.27450981736183167, 0.45098039507865906, 0.45098039507865906, 0.45098039507865906, 0.5098039507865906, 0.45098039507865906, NaN, 0.4901960790157318, 0.4117647111415863, 0.47058823704719543, 0.5490196347236633, 0.529411792755127, 0.529411792755127, 0.6078431606292725, 0.5490196347236633, 0.686274528503418, 0.5686274766921997, 0.6666666865348816, 0.529411792755127, 0.5882353186607361, 0.5686274766921997, 0.529411792755127, 0.7450980544090271, 0.6470588445663452, 0.5882353186607361, 0.686274528503418, 0.5882353186607361, 0.686274528503418, 0.6470588445663452, 0.7254902124404907, 0.6078431606292725, 0.7647058963775635, 0.7647058963775635, 0.7058823704719543, 0.686274528503418, 0.7843137383460999, 0.6470588445663452, 0.7254902124404907, 0.7647058963775635, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.8627451062202454, 0.7647058963775635, 0.7843137383460999, 0.7058823704719543, 0.843137264251709, 0.7058823704719543, 0.843137264251709, 0.9019607901573181, 0.843137264251709, 0.8823529481887817, 0.8823529481887817, 0.8235294222831726, 0.8235294222831726, 0.8627451062202454, 0.9411764740943909, 0.9411764740943909, 0.843137264251709, 0.8823529481887817, 0.9411764740943909, 0.9019607901573181, 0.9215686321258545, 0.843137264251709, 0.9215686321258545, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0], [0.13725490868091583, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.21568627655506134, 0.21568627655506134, 0.3137255012989044, 0.21568627655506134, 0.37254902720451355, 0.2549019753932953, 0.3529411852359772, 0.47058823704719543, 0.3921568691730499, 0.47058823704719543, 0.4901960790157318, 0.5098039507865906, NaN, 0.47058823704719543, 0.4117647111415863, 0.47058823704719543, 0.529411792755127, 0.4803921580314636, 0.5882353186607361, 0.5686274766921997, 0.529411792755127, 0.6078431606292725, 0.6666666865348816, 0.529411792755127, 0.5686274766921997, 0.6470588445663452, 0.5490196347236633, 0.7058823704719543, 0.6666666865348816, 0.529411792755127, 0.6666666865348816, 0.5882353186607361, 0.6470588445663452, 0.5980392098426819, 0.6666666865348816, 0.6470588445663452, 0.7647058963775635, 0.7254902124404907, 0.7254902124404907, 0.6274510025978088, 0.7450980544090271, 0.7254902124404907, 0.7647058963775635, 0.686274528503418, 0.843137264251709, 0.7647058963775635, 0.843137264251709, 0.8235294222831726, 0.8235294222831726, 0.8627451062202454, 0.7058823704719543, 0.9019607901573181, 0.7058823704719543, 0.8627451062202454, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.8039215803146362, 0.8039215803146362, 0.8823529481887817, 0.9607843160629272, 0.9411764740943909, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.8627451062202454, 0.9607843160629272, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9607843160629272, 1.0], [0.05882352963089943, 0.09803921729326248, 0.19607843458652496, 0.1568627506494522, 0.13725490868091583, 0.23529411852359772, 0.23529411852359772, 0.19607843458652496, 0.29411765933036804, 0.21568627655506134, 0.4117647111415863, 0.45098039507865906, 0.4117647111415863, 0.45098039507865906, 0.4313725531101227, 0.5882353186607361, 0.529411792755127, NaN, 0.45098039507865906, 0.5490196347236633, 0.47058823704719543, 0.4901960790157318, 0.5882353186607361, 0.5686274766921997, 0.7058823704719543, 0.5588235259056091, 0.6470588445663452, 0.5098039507865906, 0.5490196347236633, 0.5882353186607361, 0.529411792755127, 0.7450980544090271, 0.686274528503418, 0.5686274766921997, 0.686274528503418, 0.5686274766921997, 0.6274510025978088, 0.686274528503418, 0.686274528503418, 0.6274510025978088, 0.7450980544090271, 0.7647058963775635, 0.7450980544090271, 0.6470588445663452, 0.8627451062202454, 0.6274510025978088, 0.7058823704719543, 0.7450980544090271, 0.8235294222831726, 0.8235294222831726, 0.8235294222831726, 0.8627451062202454, 0.8235294222831726, 0.7843137383460999, 0.7254902124404907, 0.8823529481887817, 0.7254902124404907, 0.843137264251709, 0.9215686321258545, 0.8823529481887817, 0.8823529481887817, 0.9215686321258545, 0.8235294222831726, 0.8235294222831726, 0.843137264251709, 0.9411764740943909, 0.9215686321258545, 0.8823529481887817, 0.9019607901573181, 0.9607843160629272, 0.8823529481887817, 0.9215686321258545, 0.843137264251709, 0.9019607901573181, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636, 1.0], [0.03921568766236305, 0.23529411852359772, 0.1764705926179886, 0.27450981736183167, 0.27450981736183167, 0.2549019753932953, 0.3137255012989044, 0.4117647111415863, 0.3333333432674408, 0.45098039507865906, 0.2549019753932953, 0.4901960790157318, 0.45098039507865906, 0.4901960790157318, 0.5490196347236633, 0.529411792755127, 0.5882353186607361, 0.5490196347236633, NaN, 0.5098039507865906, 0.5098039507865906, 0.5490196347236633, 0.6078431606292725, 0.5882353186607361, 0.5882353186607361, 0.5098039507865906, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.5686274766921997, 0.7058823704719543, 0.6470588445663452, 0.6470588445663452, 0.6666666865348816, 0.813725471496582, 0.6274510025978088, 0.5686274766921997, 0.6274510025978088, 0.7058823704719543, 0.6470588445663452, 0.6470588445663452, 0.6470588445663452, 0.5882353186607361, 0.686274528503418, 0.6470588445663452, 0.7647058963775635, 0.6470588445663452, 0.7254902124404907, 0.7254902124404907, 0.7450980544090271, 0.686274528503418, 0.8235294222831726, 0.6666666865348816, 0.6666666865348816, 0.686274528503418, 0.7058823704719543, 0.7254902124404907, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.843137264251709, 0.8039215803146362, 0.7843137383460999, 0.9215686321258545, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.9411764740943909, 0.8823529481887817, 0.9019607901573181, 0.9607843160629272, 0.8627451062202454, 0.8823529481887817, 0.9411764740943909, 0.8627451062202454, 0.8627451062202454, 0.9607843160629272, 0.9019607901573181, 0.9803921580314636], [0.11764705926179886, 0.13725490868091583, 0.09803921729326248, 0.13725490868091583, 0.23529411852359772, 0.1764705926179886, 0.29411765933036804, 0.3529411852359772, 0.3529411852359772, 0.3529411852359772, 0.47058823704719543, 0.47058823704719543, 0.4901960790157318, 0.45098039507865906, 0.529411792755127, 0.45098039507865906, 0.529411792755127, 0.45098039507865906, 0.4901960790157318, NaN, 0.4901960790157318, 0.47058823704719543, 0.5882353186607361, 0.6666666865348816, 0.5882353186607361, 0.6764705777168274, 0.5882353186607361, 0.5686274766921997, 0.5882353186607361, 0.6078431606292725, 0.47058823704719543, 0.6078431606292725, 0.6274510025978088, 0.47058823704719543, 0.6666666865348816, 0.6078431606292725, 0.6078431606292725, 0.6666666865348816, 0.6666666865348816, 0.5490196347236633, 0.6470588445663452, 0.7254902124404907, 0.6274510025978088, 0.686274528503418, 0.6666666865348816, 0.6470588445663452, 0.8039215803146362, 0.6666666865348816, 0.7647058963775635, 0.7254902124404907, 0.7843137383460999, 0.7843137383460999, 0.7647058963775635, 0.8039215803146362, 0.7254902124404907, 0.7843137383460999, 0.7450980544090271, 0.843137264251709, 0.7647058963775635, 0.8627451062202454, 0.8235294222831726, 0.843137264251709, 0.7450980544090271, 0.7647058963775635, 0.8627451062202454, 0.843137264251709, 0.843137264251709, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.8627451062202454, 0.8823529481887817, 0.8627451062202454, 0.8921568393707275, 0.9019607901573181, 0.8627451062202454, 0.8823529481887817, 0.9019607901573181, 0.8823529481887817, 0.9803921580314636], [0.03921568766236305, 0.1568627506494522, 0.09803921729326248, 0.1764705926179886, 0.19607843458652496, 0.2549019753932953, 0.1764705926179886, 0.2549019753932953, 0.2549019753932953, 0.29411765933036804, 0.45098039507865906, 0.3921568691730499, 0.47058823704719543, 0.37254902720451355, 0.5098039507865906, 0.47058823704719543, 0.47058823704719543, 0.529411792755127, 0.4901960790157318, 0.5098039507865906, NaN, 0.45098039507865906, 0.5490196347236633, 0.5490196347236633, 0.6274510025978088, 0.5686274766921997, 0.6078431606292725, 0.5490196347236633, 0.5098039507865906, 0.6764705777168274, 0.5098039507865906, 0.6274510025978088, 0.6666666865348816, 0.4901960790157318, 0.5490196347236633, 0.6078431606292725, 0.7647058963775635, 0.7058823704719543, 0.686274528503418, 0.5490196347236633, 0.6666666865348816, 0.7058823704719543, 0.6470588445663452, 0.7254902124404907, 0.6666666865348816, 0.6078431606292725, 0.7254902124404907, 0.7254902124404907, 0.7254902124404907, 0.7254902124404907, 0.8039215803146362, 0.7843137383460999, 0.7254902124404907, 0.7450980544090271, 0.6470588445663452, 0.7450980544090271, 0.6666666865348816, 0.8235294222831726, 0.7843137383460999, 0.8627451062202454, 0.7843137383460999, 0.8823529481887817, 0.8235294222831726, 0.8235294222831726, 0.8823529481887817, 0.9803921580314636, 0.9019607901573181, 0.8235294222831726, 0.843137264251709, 0.9019607901573181, 0.9411764740943909, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636], [0.09803921729326248, 0.23529411852359772, 0.23529411852359772, 0.19607843458652496, 0.2549019753932953, 0.2549019753932953, 0.3529411852359772, 0.3529411852359772, 0.29411765933036804, 0.4313725531101227, 0.37254902720451355, 0.5, 0.3921568691730499, 0.5098039507865906, 0.5490196347236633, 0.47058823704719543, 0.5196078419685364, 0.5098039507865906, 0.45098039507865906, 0.529411792755127, 0.5490196347236633, NaN, 0.5686274766921997, 0.5686274766921997, 0.5686274766921997, 0.5686274766921997, 0.5882353186607361, 0.4901960790157318, 0.529411792755127, 0.6078431606292725, 0.6078431606292725, 0.6078431606292725, 0.6470588445663452, 0.5882353186607361, 0.6470588445663452, 0.5490196347236633, 0.6078431606292725, 0.5882353186607361, 0.686274528503418, 0.686274528503418, 0.6666666865348816, 0.7058823704719543, 0.6274510025978088, 0.6470588445663452, 0.6666666865348816, 0.6470588445663452, 0.686274528503418, 0.7254902124404907, 0.6666666865348816, 0.7450980544090271, 0.6666666865348816, 0.7058823704719543, 0.6666666865348816, 0.686274528503418, 0.7254902124404907, 0.7058823704719543, 0.7254902124404907, 0.686274528503418, 0.7450980544090271, 0.7843137383460999, 0.7647058963775635, 0.7450980544090271, 0.7745097875595093, 0.7647058963775635, 0.813725471496582, 0.8039215803146362, 0.7647058963775635, 0.8235294222831726, 0.8823529481887817, 0.8823529481887817, 0.8235294222831726, 0.8627451062202454, 0.843137264251709, 0.8823529481887817, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.8823529481887817, 0.9215686321258545, 0.9803921580314636], [0.0784313753247261, 0.05882352963089943, 0.13725490868091583, 0.1764705926179886, 0.13725490868091583, 0.21568627655506134, 0.19607843458652496, 0.13725490868091583, 0.2549019753932953, 0.23529411852359772, 0.37254902720451355, 0.3333333432674408, 0.3333333432674408, 0.4117647111415863, 0.4313725531101227, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.3921568691730499, 0.4117647111415863, 0.45098039507865906, 0.4313725531101227, NaN, 0.5098039507865906, 0.529411792755127, 0.4901960790157318, 0.6274510025978088, 0.47058823704719543, 0.529411792755127, 0.529411792755127, 0.4313725531101227, 0.6274510025978088, 0.6078431606292725, 0.529411792755127, 0.6274510025978088, 0.5686274766921997, 0.6470588445663452, 0.6470588445663452, 0.686274528503418, 0.5490196347236633, 0.7254902124404907, 0.7254902124404907, 0.6078431606292725, 0.6078431606292725, 0.7450980544090271, 0.6470588445663452, 0.7058823704719543, 0.7450980544090271, 0.8235294222831726, 0.7843137383460999, 0.7352941036224365, 0.8823529481887817, 0.8039215803146362, 0.7843137383460999, 0.7058823704719543, 0.8627451062202454, 0.7254902124404907, 0.9803921580314636, 0.9019607901573181, 0.8235294222831726, 0.8823529481887817, 0.843137264251709, 0.843137264251709, 0.843137264251709, 0.8823529481887817, 0.9215686321258545, 0.8823529481887817, 0.8235294222831726, 0.9215686321258545, 0.9607843160629272, 0.8823529481887817, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 0.9607843160629272, 1.0], [0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.23529411852359772, 0.23529411852359772, 0.3921568691730499, 0.3333333432674408, 0.4117647111415863, 0.37254902720451355, 0.2549019753932953, 0.45098039507865906, 0.4313725531101227, 0.4313725531101227, 0.4117647111415863, 0.3333333432674408, 0.45098039507865906, 0.4313725531101227, 0.4901960790157318, NaN, 0.5490196347236633, 0.5098039507865906, 0.529411792755127, 0.47058823704719543, 0.5490196347236633, 0.5490196347236633, 0.4901960790157318, 0.5686274766921997, 0.6470588445663452, 0.45098039507865906, 0.6078431606292725, 0.5686274766921997, 0.7254902124404907, 0.6078431606292725, 0.7254902124404907, 0.5490196347236633, 0.6666666865348816, 0.686274528503418, 0.6274510025978088, 0.686274528503418, 0.7058823704719543, 0.6078431606292725, 0.6470588445663452, 0.7450980544090271, 0.7254902124404907, 0.7843137383460999, 0.7647058963775635, 0.8235294222831726, 0.7647058963775635, 0.7843137383460999, 0.7058823704719543, 0.7843137383460999, 0.7254902124404907, 0.8627451062202454, 0.8823529481887817, 0.843137264251709, 0.8627451062202454, 0.8823529481887817, 0.8039215803146362, 0.8235294222831726, 0.8627451062202454, 0.8823529481887817, 0.9019607901573181, 0.843137264251709, 0.843137264251709, 0.9411764740943909, 0.8823529481887817, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9215686321258545, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 0.9411764740943909, 1.0], [0.09803921729326248, 0.05882352963089943, 0.13725490868091583, 0.09803921729326248, 0.09803921729326248, 0.19607843458652496, 0.13725490868091583, 0.1764705926179886, 0.2549019753932953, 0.21568627655506134, 0.3921568691730499, 0.37254902720451355, 0.3529411852359772, 0.3529411852359772, 0.27450981736183167, 0.3137255012989044, 0.47058823704719543, 0.29411765933036804, 0.4117647111415863, 0.4117647111415863, 0.37254902720451355, 0.4313725531101227, 0.47058823704719543, 0.45098039507865906, NaN, 0.45098039507865906, 0.4901960790157318, 0.5098039507865906, 0.5686274766921997, 0.47058823704719543, 0.5098039507865906, 0.6470588445663452, 0.6078431606292725, 0.47058823704719543, 0.6274510025978088, 0.5882353186607361, 0.5882353186607361, 0.5882353186607361, 0.6470588445663452, 0.5882353186607361, 0.7058823704719543, 0.7450980544090271, 0.6274510025978088, 0.5490196347236633, 0.7843137383460999, 0.6274510025978088, 0.6274510025978088, 0.686274528503418, 0.7254902124404907, 0.7647058963775635, 0.7647058963775635, 0.7647058963775635, 0.7254902124404907, 0.8039215803146362, 0.7058823704719543, 0.8039215803146362, 0.7058823704719543, 0.7843137383460999, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.843137264251709, 0.8235294222831726, 0.843137264251709, 0.9019607901573181, 0.9411764740943909, 0.8627451062202454, 0.8627451062202454, 0.9215686321258545, 0.8823529481887817, 0.9411764740943909, 0.8627451062202454, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9803921580314636, 1.0], [0.0784313753247261, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.13725490868091583, 0.19607843458652496, 0.1764705926179886, 0.29411765933036804, 0.29411765933036804, 0.3137255012989044, 0.4313725531101227, 0.3137255012989044, 0.4117647111415863, 0.3921568691730499, 0.45098039507865906, 0.4313725531101227, 0.3921568691730499, 0.44117647409439087, 0.4901960790157318, 0.3235294222831726, 0.4313725531101227, 0.4313725531101227, 0.5098039507865906, 0.4901960790157318, 0.5490196347236633, NaN, 0.4901960790157318, 0.5098039507865906, 0.529411792755127, 0.5098039507865906, 0.5098039507865906, 0.5686274766921997, 0.5490196347236633, 0.4901960790157318, 0.6274510025978088, 0.5882353186607361, 0.6078431606292725, 0.5686274766921997, 0.5686274766921997, 0.5490196347236633, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6470588445663452, 0.6470588445663452, 0.6666666865348816, 0.7647058963775635, 0.686274528503418, 0.6666666865348816, 0.7058823704719543, 0.7647058963775635, 0.7450980544090271, 0.686274528503418, 0.7156862616539001, 0.7254902124404907, 0.7254902124404907, 0.7450980544090271, 0.7843137383460999, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.8235294222831726, 0.7843137383460999, 0.7843137383460999, 0.843137264251709, 0.8627451062202454, 0.843137264251709, 0.843137264251709, 0.8235294222831726, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.8823529481887817, 0.9215686321258545, 0.8823529481887817, 0.9215686321258545, 0.9411764740943909, 0.9411764740943909, 0.9411764740943909, 1.0], [0.11764705926179886, 0.05882352963089943, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.21568627655506134, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.19607843458652496, 0.37254902720451355, 0.29411765933036804, 0.3137255012989044, 0.3137255012989044, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3529411852359772, 0.37254902720451355, 0.4117647111415863, 0.3921568691730499, 0.4117647111415863, 0.37254902720451355, 0.47058823704719543, 0.5098039507865906, 0.5098039507865906, NaN, 0.45098039507865906, 0.5686274766921997, 0.529411792755127, 0.47058823704719543, 0.5882353186607361, 0.5490196347236633, 0.5098039507865906, 0.5882353186607361, 0.529411792755127, 0.5882353186607361, 0.6078431606292725, 0.6274510025978088, 0.5490196347236633, 0.6470588445663452, 0.6666666865348816, 0.6078431606292725, 0.6274510025978088, 0.686274528503418, 0.6470588445663452, 0.686274528503418, 0.7450980544090271, 0.8235294222831726, 0.7647058963775635, 0.7254902124404907, 0.8627451062202454, 0.8039215803146362, 0.7647058963775635, 0.686274528503418, 0.8235294222831726, 0.7058823704719543, 0.9019607901573181, 0.8627451062202454, 0.843137264251709, 0.8627451062202454, 0.8039215803146362, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.8823529481887817, 0.7843137383460999, 0.9019607901573181, 0.9411764740943909, 0.8823529481887817, 0.9019607901573181, 0.8627451062202454, 0.9215686321258545, 0.9019607901573181, 0.9215686321258545, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272, 1.0], [0.11764705926179886, 0.1568627506494522, 0.11764705926179886, 0.1764705926179886, 0.1764705926179886, 0.1764705926179886, 0.2549019753932953, 0.29411765933036804, 0.1568627506494522, 0.3333333432674408, 0.37254902720451355, 0.4313725531101227, 0.30392158031463623, 0.4117647111415863, 0.45098039507865906, 0.47058823704719543, 0.47058823704719543, 0.4901960790157318, 0.37254902720451355, 0.4313725531101227, 0.45098039507865906, 0.5098039507865906, 0.529411792755127, 0.529411792755127, 0.4901960790157318, 0.4901960790157318, 0.5490196347236633, NaN, 0.5686274766921997, 0.4901960790157318, 0.4901960790157318, 0.5686274766921997, 0.529411792755127, 0.529411792755127, 0.6274510025978088, 0.5686274766921997, 0.5882353186607361, 0.5098039507865906, 0.6470588445663452, 0.5686274766921997, 0.5686274766921997, 0.6470588445663452, 0.7058823704719543, 0.6078431606292725, 0.7058823704719543, 0.5882353186607361, 0.6666666865348816, 0.6470588445663452, 0.6666666865348816, 0.7254902124404907, 0.6666666865348816, 0.6666666865348816, 0.7058823704719543, 0.8235294222831726, 0.6470588445663452, 0.686274528503418, 0.6274510025978088, 0.7647058963775635, 0.7058823704719543, 0.8039215803146362, 0.7843137383460999, 0.8039215803146362, 0.7843137383460999, 0.7843137383460999, 0.8823529481887817, 0.8235294222831726, 0.843137264251709, 0.8235294222831726, 0.8823529481887817, 0.8235294222831726, 0.9411764740943909, 0.9019607901573181, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.8823529481887817, 0.9019607901573181, 0.9803921580314636, 0.9019607901573181, 1.0], [0.05882352963089943, 0.0784313753247261, 0.13725490868091583, 0.1568627506494522, 0.1568627506494522, 0.21568627655506134, 0.21568627655506134, 0.19607843458652496, 0.19607843458652496, 0.2549019753932953, 0.3333333432674408, 0.3529411852359772, 0.3333333432674408, 0.4313725531101227, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.45098039507865906, 0.37254902720451355, 0.4117647111415863, 0.4901960790157318, 0.47058823704719543, 0.47058823704719543, 0.45098039507865906, 0.4313725531101227, 0.47058823704719543, 0.4313725531101227, 0.4313725531101227, NaN, 0.45098039507865906, 0.47058823704719543, 0.529411792755127, 0.529411792755127, 0.5098039507865906, 0.686274528503418, 0.47058823704719543, 0.6078431606292725, 0.5882353186607361, 0.5882353186607361, 0.5490196347236633, 0.6078431606292725, 0.6470588445663452, 0.5882353186607361, 0.6078431606292725, 0.529411792755127, 0.6078431606292725, 0.7450980544090271, 0.6274510025978088, 0.7450980544090271, 0.6470588445663452, 0.6666666865348816, 0.686274528503418, 0.7843137383460999, 0.7254902124404907, 0.686274528503418, 0.8039215803146362, 0.686274528503418, 0.7450980544090271, 0.8235294222831726, 0.7843137383460999, 0.7647058963775635, 0.8039215803146362, 0.7647058963775635, 0.7647058963775635, 0.9019607901573181, 0.8823529481887817, 0.843137264251709, 0.9215686321258545, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.9215686321258545, 0.8235294222831726, 0.9411764740943909, 0.8823529481887817, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636, 0.9607843160629272, 0.9803921580314636], [0.0784313753247261, 0.05882352963089943, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.2549019753932953, 0.1568627506494522, 0.2549019753932953, 0.29411765933036804, 0.37254902720451355, 0.4313725531101227, 0.3333333432674408, 0.3921568691730499, 0.3529411852359772, 0.4313725531101227, 0.3529411852359772, 0.4117647111415863, 0.37254902720451355, 0.3921568691730499, 0.3235294222831726, 0.3921568691730499, 0.47058823704719543, 0.45098039507865906, 0.529411792755127, 0.4901960790157318, 0.47058823704719543, 0.5098039507865906, 0.5490196347236633, NaN, 0.47058823704719543, 0.5882353186607361, 0.6078431606292725, 0.47058823704719543, 0.6470588445663452, 0.5490196347236633, 0.6078431606292725, 0.6274510025978088, 0.686274528503418, 0.5098039507865906, 0.6274510025978088, 0.7058823704719543, 0.6078431606292725, 0.529411792755127, 0.843137264251709, 0.6078431606292725, 0.6078431606292725, 0.7254902124404907, 0.6666666865348816, 0.7254902124404907, 0.7058823704719543, 0.7254902124404907, 0.686274528503418, 0.7647058963775635, 0.6470588445663452, 0.7647058963775635, 0.6470588445663452, 0.7450980544090271, 0.8039215803146362, 0.7450980544090271, 0.7843137383460999, 0.8235294222831726, 0.8235294222831726, 0.8039215803146362, 0.8235294222831726, 0.8235294222831726, 0.8627451062202454, 0.8039215803146362, 0.7843137383460999, 0.8627451062202454, 0.8823529481887817, 0.8627451062202454, 0.843137264251709, 0.8627451062202454, 0.8627451062202454, 0.9411764740943909, 0.9607843160629272, 0.9607843160629272, 0.9803921580314636, 1.0], [0.03921568766236305, 0.21568627655506134, 0.1764705926179886, 0.19607843458652496, 0.13725490868091583, 0.2549019753932953, 0.28431373834609985, 0.4117647111415863, 0.3529411852359772, 0.4117647111415863, 0.37254902720451355, 0.4313725531101227, 0.37254902720451355, 0.47058823704719543, 0.4901960790157318, 0.47058823704719543, 0.45098039507865906, 0.47058823704719543, 0.4313725531101227, 0.529411792755127, 0.4901960790157318, 0.3921568691730499, 0.5686274766921997, 0.5098039507865906, 0.4901960790157318, 0.4901960790157318, 0.529411792755127, 0.5098039507865906, 0.529411792755127, 0.529411792755127, NaN, 0.4901960790157318, 0.5686274766921997, 0.5784313678741455, 0.5882353186607361, 0.4901960790157318, 0.5490196347236633, 0.5490196347236633, 0.6470588445663452, 0.686274528503418, 0.5686274766921997, 0.6274510025978088, 0.5882353186607361, 0.656862735748291, 0.6078431606292725, 0.656862735748291, 0.5490196347236633, 0.6666666865348816, 0.5490196347236633, 0.686274528503418, 0.686274528503418, 0.6470588445663452, 0.6078431606292725, 0.6666666865348816, 0.6274510025978088, 0.6470588445663452, 0.6470588445663452, 0.686274528503418, 0.6764705777168274, 0.7254902124404907, 0.7450980544090271, 0.686274528503418, 0.7058823704719543, 0.686274528503418, 0.7745097875595093, 0.7254902124404907, 0.7450980544090271, 0.7647058963775635, 0.843137264251709, 0.7941176295280457, 0.7745097875595093, 0.843137264251709, 0.7647058963775635, 0.8039215803146362, 0.8823529481887817, 0.8823529481887817, 0.8823529481887817, 0.8627451062202454, 0.8823529481887817, 1.0], [0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.09803921729326248, 0.1568627506494522, 0.21568627655506134, 0.13725490868091583, 0.1764705926179886, 0.14705882966518402, 0.1764705926179886, 0.2549019753932953, 0.27450981736183167, 0.2549019753932953, 0.3529411852359772, 0.3333333432674408, 0.2549019753932953, 0.29411765933036804, 0.2549019753932953, 0.29411765933036804, 0.3921568691730499, 0.37254902720451355, 0.3921568691730499, 0.37254902720451355, 0.4313725531101227, 0.3529411852359772, 0.4313725531101227, 0.4117647111415863, 0.4313725531101227, 0.47058823704719543, 0.4117647111415863, 0.5098039507865906, NaN, 0.529411792755127, 0.5098039507865906, 0.529411792755127, 0.4901960790157318, 0.529411792755127, 0.529411792755127, 0.5490196347236633, 0.5490196347236633, 0.5882353186607361, 0.6078431606292725, 0.6470588445663452, 0.6078431606292725, 0.529411792755127, 0.686274528503418, 0.6274510025978088, 0.6470588445663452, 0.686274528503418, 0.7450980544090271, 0.6470588445663452, 0.6078431606292725, 0.7450980544090271, 0.7843137383460999, 0.6470588445663452, 0.8039215803146362, 0.6470588445663452, 0.7450980544090271, 0.8039215803146362, 0.8039215803146362, 0.7745097875595093, 0.7647058963775635, 0.7843137383460999, 0.7843137383460999, 0.8039215803146362, 0.9215686321258545, 0.843137264251709, 0.8823529481887817, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.9019607901573181, 0.8627451062202454, 0.9411764740943909, 0.9019607901573181, 0.8823529481887817, 0.9215686321258545, 0.9803921580314636, 0.9215686321258545, 1.0], [0.09803921729326248, 0.05882352963089943, 0.13725490868091583, 0.13725490868091583, 0.11764705926179886, 0.19607843458652496, 0.1764705926179886, 0.13725490868091583, 0.2549019753932953, 0.23529411852359772, 0.37254902720451355, 0.37254902720451355, 0.3333333432674408, 0.27450981736183167, 0.2549019753932953, 0.3529411852359772, 0.3333333432674408, 0.3137255012989044, 0.3529411852359772, 0.37254902720451355, 0.3333333432674408, 0.3529411852359772, 0.3921568691730499, 0.3529411852359772, 0.3921568691730499, 0.45098039507865906, 0.45098039507865906, 0.47058823704719543, 0.47058823704719543, 0.3921568691730499, 0.4313725531101227, 0.47058823704719543, NaN, 0.4117647111415863, 0.5882353186607361, 0.5098039507865906, 0.529411792755127, 0.5686274766921997, 0.6274510025978088, 0.4901960790157318, 0.5882353186607361, 0.7058823704719543, 0.5882353186607361, 0.5686274766921997, 0.7254902124404907, 0.6078431606292725, 0.5686274766921997, 0.6666666865348816, 0.6274510025978088, 0.7450980544090271, 0.6274510025978088, 0.7450980544090271, 0.6666666865348816, 0.7254902124404907, 0.6078431606292725, 0.7254902124404907, 0.6078431606292725, 0.7647058963775635, 0.8039215803146362, 0.7647058963775635, 0.7843137383460999, 0.8039215803146362, 0.7843137383460999, 0.7843137383460999, 0.8039215803146362, 0.8235294222831726, 0.8235294222831726, 0.8039215803146362, 0.8039215803146362, 0.8039215803146362, 0.8823529481887817, 0.843137264251709, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.9019607901573181, 0.9411764740943909, 0.9607843160629272, 1.0], [0.0, 0.1568627506494522, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.19607843458652496, 0.13725490868091583, 0.3137255012989044, 0.19607843458652496, 0.3333333432674408, 0.3137255012989044, 0.4117647111415863, 0.3333333432674408, 0.4901960790157318, 0.47058823704719543, 0.4117647111415863, 0.47058823704719543, 0.4313725531101227, 0.3529411852359772, 0.529411792755127, 0.5098039507865906, 0.4117647111415863, 0.47058823704719543, 0.5490196347236633, 0.529411792755127, 0.5098039507865906, 0.4901960790157318, 0.47058823704719543, 0.4901960790157318, 0.529411792755127, 0.4215686321258545, 0.4901960790157318, 0.5882353186607361, NaN, 0.529411792755127, 0.5098039507865906, 0.6274510025978088, 0.5490196347236633, 0.6274510025978088, 0.5588235259056091, 0.5490196347236633, 0.5882353186607361, 0.529411792755127, 0.6078431606292725, 0.5882353186607361, 0.5882353186607361, 0.6078431606292725, 0.6470588445663452, 0.529411792755127, 0.6666666865348816, 0.7058823704719543, 0.6078431606292725, 0.5882353186607361, 0.6666666865348816, 0.5882353186607361, 0.5686274766921997, 0.5882353186607361, 0.6274510025978088, 0.6372548937797546, 0.686274528503418, 0.7254902124404907, 0.686274528503418, 0.7254902124404907, 0.7254902124404907, 0.7745097875595093, 0.7058823704719543, 0.7058823704719543, 0.8039215803146362, 0.7843137383460999, 0.7843137383460999, 0.7843137383460999, 0.8235294222831726, 0.7843137383460999, 0.8039215803146362, 0.8823529481887817, 0.8823529481887817, 0.8823529481887817, 0.8823529481887817, 0.9019607901573181, 1.0], [0.05882352963089943, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.11764705926179886, 0.1764705926179886, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.19607843458652496, 0.27450981736183167, 0.29411765933036804, 0.2549019753932953, 0.4313725531101227, 0.3137255012989044, 0.3137255012989044, 0.3333333432674408, 0.3137255012989044, 0.3333333432674408, 0.3333333432674408, 0.45098039507865906, 0.3529411852359772, 0.37254902720451355, 0.3921568691730499, 0.37254902720451355, 0.37254902720451355, 0.4117647111415863, 0.37254902720451355, 0.3137255012989044, 0.3529411852359772, 0.4117647111415863, 0.47058823704719543, 0.4117647111415863, 0.47058823704719543, NaN, 0.4313725531101227, 0.5098039507865906, 0.529411792755127, 0.5098039507865906, 0.5098039507865906, 0.5490196347236633, 0.5882353186607361, 0.6078431606292725, 0.5882353186607361, 0.5098039507865906, 0.5490196347236633, 0.6274510025978088, 0.5882353186607361, 0.7254902124404907, 0.5686274766921997, 0.6666666865348816, 0.6078431606292725, 0.7450980544090271, 0.686274528503418, 0.6274510025978088, 0.7450980544090271, 0.6274510025978088, 0.686274528503418, 0.7843137383460999, 0.7450980544090271, 0.7647058963775635, 0.7450980544090271, 0.7450980544090271, 0.7647058963775635, 0.843137264251709, 0.9019607901573181, 0.8235294222831726, 0.9607843160629272, 0.9215686321258545, 0.8627451062202454, 0.8823529481887817, 0.8823529481887817, 0.8235294222831726, 0.9411764740943909, 0.9215686321258545, 0.9215686321258545, 0.9019607901573181, 0.9803921580314636, 0.9215686321258545, 1.0], [0.019607843831181526, 0.1764705926179886, 0.0784313753247261, 0.21568627655506134, 0.21568627655506134, 0.21568627655506134, 0.23529411852359772, 0.2549019753932953, 0.23529411852359772, 0.29411765933036804, 0.19607843458652496, 0.37254902720451355, 0.27450981736183167, 0.3529411852359772, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.18627451360225677, 0.3921568691730499, 0.3921568691730499, 0.45098039507865906, 0.4313725531101227, 0.4313725531101227, 0.4117647111415863, 0.4117647111415863, 0.47058823704719543, 0.4313725531101227, 0.529411792755127, 0.45098039507865906, 0.5098039507865906, 0.5098039507865906, 0.4901960790157318, 0.4901960790157318, 0.5686274766921997, NaN, 0.5098039507865906, 0.4313725531101227, 0.45098039507865906, 0.5490196347236633, 0.5098039507865906, 0.4901960790157318, 0.529411792755127, 0.5098039507865906, 0.5686274766921997, 0.5490196347236633, 0.6470588445663452, 0.5098039507865906, 0.6078431606292725, 0.5686274766921997, 0.5686274766921997, 0.5882353186607361, 0.6666666865348816, 0.6470588445663452, 0.5882353186607361, 0.6666666865348816, 0.6078431606292725, 0.686274528503418, 0.7058823704719543, 0.7843137383460999, 0.686274528503418, 0.7450980544090271, 0.7058823704719543, 0.7254902124404907, 0.8235294222831726, 0.7843137383460999, 0.8235294222831726, 0.8039215803146362, 0.843137264251709, 0.7843137383460999, 0.8823529481887817, 0.8627451062202454, 0.7647058963775635, 0.8235294222831726, 0.8627451062202454, 0.7843137383460999, 0.8039215803146362, 0.9019607901573181, 0.8627451062202454, 0.9803921580314636], [0.05882352963089943, 0.09803921729326248, 0.11764705926179886, 0.11764705926179886, 0.11764705926179886, 0.13725490868091583, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.1764705926179886, 0.3137255012989044, 0.23529411852359772, 0.29411765933036804, 0.11764705926179886, 0.27450981736183167, 0.3137255012989044, 0.3529411852359772, 0.37254902720451355, 0.37254902720451355, 0.3921568691730499, 0.23529411852359772, 0.3921568691730499, 0.3529411852359772, 0.27450981736183167, 0.4117647111415863, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.3921568691730499, 0.3921568691730499, 0.45098039507865906, 0.47058823704719543, 0.47058823704719543, 0.37254902720451355, 0.4901960790157318, 0.4901960790157318, NaN, 0.5686274766921997, 0.5490196347236633, 0.47058823704719543, 0.5686274766921997, 0.5882353186607361, 0.5686274766921997, 0.6274510025978088, 0.5686274766921997, 0.5686274766921997, 0.6078431606292725, 0.6078431606292725, 0.6274510025978088, 0.5980392098426819, 0.686274528503418, 0.7254902124404907, 0.6666666865348816, 0.7450980544090271, 0.5882353186607361, 0.7058823704719543, 0.5882353186607361, 0.7254902124404907, 0.7254902124404907, 0.843137264251709, 0.7254902124404907, 0.843137264251709, 0.7647058963775635, 0.7843137383460999, 0.8235294222831726, 0.8823529481887817, 0.843137264251709, 0.7450980544090271, 0.8039215803146362, 0.8235294222831726, 0.8823529481887817, 0.843137264251709, 0.8627451062202454, 0.843137264251709, 0.9019607901573181, 0.9215686321258545, 0.9019607901573181, 0.9803921580314636, 0.9215686321258545, 0.9607843160629272], [0.05882352963089943, 0.13725490868091583, 0.0784313753247261, 0.09803921729326248, 0.19607843458652496, 0.1764705926179886, 0.1764705926179886, 0.23529411852359772, 0.2549019753932953, 0.29411765933036804, 0.4117647111415863, 0.3137255012989044, 0.37254902720451355, 0.29411765933036804, 0.3529411852359772, 0.3529411852359772, 0.4019607901573181, 0.3137255012989044, 0.4313725531101227, 0.3333333432674408, 0.29411765933036804, 0.4117647111415863, 0.3529411852359772, 0.3921568691730499, 0.4117647111415863, 0.4313725531101227, 0.3921568691730499, 0.4901960790157318, 0.4117647111415863, 0.37254902720451355, 0.45098039507865906, 0.47058823704719543, 0.4313725531101227, 0.45098039507865906, 0.47058823704719543, 0.5686274766921997, 0.4313725531101227, NaN, 0.529411792755127, 0.5490196347236633, 0.5686274766921997, 0.529411792755127, 0.5686274766921997, 0.5392156839370728, 0.529411792755127, 0.5686274766921997, 0.5784313678741455, 0.5686274766921997, 0.5882353186607361, 0.6274510025978088, 0.6274510025978088, 0.6078431606292725, 0.6470588445663452, 0.6274510025978088, 0.686274528503418, 0.5882353186607361, 0.686274528503418, 0.6470588445663452, 0.7058823704719543, 0.7058823704719543, 0.7647058963775635, 0.7058823704719543, 0.7647058963775635, 0.7843137383460999, 0.7745097875595093, 0.7450980544090271, 0.7450980544090271, 0.6666666865348816, 0.7647058963775635, 0.813725471496582, 0.7941176295280457, 0.8235294222831726, 0.8235294222831726, 0.8823529481887817, 0.8627451062202454, 0.9215686321258545, 0.9019607901573181, 0.9803921580314636, 0.9803921580314636, 1.0], [0.05882352963089943, 0.11764705926179886, 0.1764705926179886, 0.09803921729326248, 0.11764705926179886, 0.21568627655506134, 0.13725490868091583, 0.19607843458652496, 0.19607843458652496, 0.19607843458652496, 0.3529411852359772, 0.23529411852359772, 0.3333333432674408, 0.21568627655506134, 0.2549019753932953, 0.27450981736183167, 0.3333333432674408, 0.3137255012989044, 0.37254902720451355, 0.3333333432674408, 0.3137255012989044, 0.3137255012989044, 0.3137255012989044, 0.27450981736183167, 0.3529411852359772, 0.4313725531101227, 0.37254902720451355, 0.3529411852359772, 0.4117647111415863, 0.3137255012989044, 0.3529411852359772, 0.45098039507865906, 0.37254902720451355, 0.37254902720451355, 0.4901960790157318, 0.5490196347236633, 0.45098039507865906, 0.47058823704719543, NaN, 0.4117647111415863, 0.5098039507865906, 0.5490196347236633, 0.5882353186607361, 0.47058823704719543, 0.45098039507865906, 0.4901960790157318, 0.6666666865348816, 0.7843137383460999, 0.6470588445663452, 0.843137264251709, 0.6078431606292725, 0.6078431606292725, 0.6078431606292725, 0.686274528503418, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.5882353186607361, 0.6470588445663452, 0.7450980544090271, 0.7058823704719543, 0.6666666865348816, 0.8039215803146362, 0.8039215803146362, 0.7058823704719543, 0.843137264251709, 0.7254902124404907, 0.7647058963775635, 0.7254902124404907, 0.7450980544090271, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.8823529481887817, 0.843137264251709, 0.9607843160629272, 0.9607843160629272, 0.9215686321258545, 0.9803921580314636, 0.9607843160629272], [0.019607843831181526, 0.11764705926179886, 0.0784313753247261, 0.09803921729326248, 0.03921568766236305, 0.19607843458652496, 0.0784313753247261, 0.3137255012989044, 0.13725490868091583, 0.27450981736183167, 0.2549019753932953, 0.37254902720451355, 0.29411765933036804, 0.4313725531101227, 0.4313725531101227, 0.3921568691730499, 0.3529411852359772, 0.37254902720451355, 0.29411765933036804, 0.45098039507865906, 0.45098039507865906, 0.3137255012989044, 0.45098039507865906, 0.45098039507865906, 0.4117647111415863, 0.45098039507865906, 0.45098039507865906, 0.4313725531101227, 0.45098039507865906, 0.4901960790157318, 0.3137255012989044, 0.45098039507865906, 0.5098039507865906, 0.44117647409439087, 0.4901960790157318, 0.45098039507865906, 0.529411792755127, 0.45098039507865906, 0.5882353186607361, NaN, 0.5490196347236633, 0.5098039507865906, 0.5196078419685364, 0.5588235259056091, 0.5490196347236633, 0.5392156839370728, 0.4901960790157318, 0.5882353186607361, 0.5098039507865906, 0.6470588445663452, 0.6470588445663452, 0.5882353186607361, 0.529411792755127, 0.6078431606292725, 0.5490196347236633, 0.529411792755127, 0.5490196347236633, 0.5882353186607361, 0.6176470518112183, 0.6666666865348816, 0.7254902124404907, 0.6274510025978088, 0.686274528503418, 0.686274528503418, 0.7352941036224365, 0.686274528503418, 0.686274528503418, 0.7450980544090271, 0.8235294222831726, 0.7352941036224365, 0.7745097875595093, 0.7843137383460999, 0.7450980544090271, 0.7843137383460999, 0.843137264251709, 0.8823529481887817, 0.8823529481887817, 0.8627451062202454, 0.8823529481887817, 0.9803921580314636], [0.03921568766236305, 0.05882352963089943, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.2549019753932953, 0.19607843458652496, 0.2549019753932953, 0.2549019753932953, 0.23529411852359772, 0.23529411852359772, 0.23529411852359772, 0.2549019753932953, 0.3529411852359772, 0.3529411852359772, 0.3333333432674408, 0.3333333432674408, 0.27450981736183167, 0.3333333432674408, 0.29411765933036804, 0.3333333432674408, 0.3529411852359772, 0.4313725531101227, 0.3921568691730499, 0.37254902720451355, 0.4313725531101227, 0.4117647111415863, 0.4117647111415863, 0.45098039507865906, 0.45098039507865906, 0.4901960790157318, 0.4313725531101227, 0.4313725531101227, 0.4901960790157318, 0.45098039507865906, NaN, 0.47058823704719543, 0.5098039507865906, 0.529411792755127, 0.45098039507865906, 0.5686274766921997, 0.5098039507865906, 0.529411792755127, 0.6274510025978088, 0.5490196347236633, 0.7450980544090271, 0.6470588445663452, 0.6470588445663452, 0.6666666865348816, 0.5882353186607361, 0.686274528503418, 0.5882353186607361, 0.686274528503418, 0.7450980544090271, 0.7843137383460999, 0.7647058963775635, 0.7058823704719543, 0.7450980544090271, 0.7647058963775635, 0.7843137383460999, 0.9607843160629272, 0.7647058963775635, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.9019607901573181, 0.8823529481887817, 0.8039215803146362, 0.8823529481887817, 0.9019607901573181, 0.9215686321258545, 0.9411764740943909, 0.9411764740943909, 0.9803921580314636, 0.9607843160629272], [0.0784313753247261, 0.09803921729326248, 0.05882352963089943, 0.0784313753247261, 0.11764705926179886, 0.1568627506494522, 0.13725490868091583, 0.1764705926179886, 0.1764705926179886, 0.0784313753247261, 0.29411765933036804, 0.1764705926179886, 0.29411765933036804, 0.1764705926179886, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.23529411852359772, 0.3529411852359772, 0.27450981736183167, 0.29411765933036804, 0.29411765933036804, 0.27450981736183167, 0.3137255012989044, 0.2549019753932953, 0.3333333432674408, 0.3333333432674408, 0.3529411852359772, 0.3529411852359772, 0.29411765933036804, 0.37254902720451355, 0.3921568691730499, 0.29411765933036804, 0.4117647111415863, 0.4117647111415863, 0.5098039507865906, 0.4117647111415863, 0.47058823704719543, 0.45098039507865906, 0.4901960790157318, 0.529411792755127, NaN, 0.4901960790157318, 0.5098039507865906, 0.47058823704719543, 0.5686274766921997, 0.5882353186607361, 0.5882353186607361, 0.6078431606292725, 0.5882353186607361, 0.6078431606292725, 0.6666666865348816, 0.6666666865348816, 0.6470588445663452, 0.5882353186607361, 0.686274528503418, 0.5882353186607361, 0.6078431606292725, 0.7058823704719543, 0.7450980544090271, 0.7254902124404907, 0.7058823704719543, 0.7254902124404907, 0.7647058963775635, 0.7450980544090271, 0.8627451062202454, 0.7647058963775635, 0.7647058963775635, 0.8235294222831726, 0.8039215803146362, 0.9215686321258545, 0.9411764740943909, 0.8823529481887817, 0.8823529481887817, 0.9215686321258545, 0.9215686321258545, 0.9411764740943909, 0.9607843160629272, 0.9803921580314636, 1.0], [0.0784313753247261, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.11764705926179886, 0.1764705926179886, 0.1568627506494522, 0.23529411852359772, 0.1568627506494522, 0.19607843458652496, 0.27450981736183167, 0.3137255012989044, 0.29411765933036804, 0.29411765933036804, 0.3333333432674408, 0.29411765933036804, 0.27450981736183167, 0.2549019753932953, 0.3529411852359772, 0.37254902720451355, 0.3529411852359772, 0.37254902720451355, 0.3921568691730499, 0.37254902720451355, 0.37254902720451355, 0.3333333432674408, 0.3921568691730499, 0.29411765933036804, 0.4117647111415863, 0.3921568691730499, 0.4117647111415863, 0.3529411852359772, 0.4117647111415863, 0.47058823704719543, 0.3921568691730499, 0.47058823704719543, 0.4313725531101227, 0.4313725531101227, 0.4117647111415863, 0.4803921580314636, 0.4901960790157318, 0.5098039507865906, NaN, 0.47058823704719543, 0.47058823704719543, 0.529411792755127, 0.5490196347236633, 0.5098039507865906, 0.5490196347236633, 0.5686274766921997, 0.5490196347236633, 0.529411792755127, 0.6274510025978088, 0.6470588445663452, 0.5686274766921997, 0.5686274766921997, 0.5882353186607361, 0.6078431606292725, 0.5882353186607361, 0.7254902124404907, 0.6666666865348816, 0.7450980544090271, 0.686274528503418, 0.686274528503418, 0.686274528503418, 0.7450980544090271, 0.7450980544090271, 0.7254902124404907, 0.7843137383460999, 0.7450980544090271, 0.8235294222831726, 0.843137264251709, 0.8823529481887817, 0.8039215803146362, 0.8235294222831726, 0.9019607901573181, 0.8823529481887817, 0.9215686321258545, 0.9019607901573181, 1.0], [0.03921568766236305, 0.11764705926179886, 0.05882352963089943, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.13725490868091583, 0.1764705926179886, 0.13725490868091583, 0.21568627655506134, 0.3333333432674408, 0.23529411852359772, 0.3529411852359772, 0.23529411852359772, 0.3333333432674408, 0.3137255012989044, 0.37254902720451355, 0.3529411852359772, 0.4117647111415863, 0.3137255012989044, 0.27450981736183167, 0.3529411852359772, 0.3921568691730499, 0.3137255012989044, 0.45098039507865906, 0.3529411852359772, 0.37254902720451355, 0.3921568691730499, 0.3921568691730499, 0.47058823704719543, 0.343137264251709, 0.3921568691730499, 0.4313725531101227, 0.3921568691730499, 0.4117647111415863, 0.4901960790157318, 0.37254902720451355, 0.46078431606292725, 0.529411792755127, 0.44117647409439087, 0.47058823704719543, 0.4901960790157318, 0.529411792755127, NaN, 0.5098039507865906, 0.4901960790157318, 0.4901960790157318, 0.6666666865348816, 0.5098039507865906, 0.5882353186607361, 0.6274510025978088, 0.5686274766921997, 0.5098039507865906, 0.5686274766921997, 0.5882353186607361, 0.529411792755127, 0.5882353186607361, 0.5686274766921997, 0.5980392098426819, 0.6666666865348816, 0.7647058963775635, 0.6666666865348816, 0.7647058963775635, 0.7647058963775635, 0.7156862616539001, 0.7254902124404907, 0.686274528503418, 0.6470588445663452, 0.6666666865348816, 0.7156862616539001, 0.7941176295280457, 0.8235294222831726, 0.8627451062202454, 0.8823529481887817, 0.8039215803146362, 0.9215686321258545, 0.9215686321258545, 0.9803921580314636, 0.9607843160629272, 1.0], [0.05882352963089943, 0.03921568766236305, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.11764705926179886, 0.1568627506494522, 0.19607843458652496, 0.3529411852359772, 0.3137255012989044, 0.3137255012989044, 0.2549019753932953, 0.19607843458652496, 0.21568627655506134, 0.2549019753932953, 0.13725490868091583, 0.3137255012989044, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2549019753932953, 0.29411765933036804, 0.21568627655506134, 0.3529411852359772, 0.3137255012989044, 0.29411765933036804, 0.47058823704719543, 0.1568627506494522, 0.3921568691730499, 0.47058823704719543, 0.27450981736183167, 0.4117647111415863, 0.4901960790157318, 0.4313725531101227, 0.4313725531101227, 0.47058823704719543, 0.5490196347236633, 0.45098039507865906, 0.5490196347236633, 0.529411792755127, 0.529411792755127, 0.4901960790157318, NaN, 0.5686274766921997, 0.5490196347236633, 0.6274510025978088, 0.5098039507865906, 0.6078431606292725, 0.5490196347236633, 0.6078431606292725, 0.5490196347236633, 0.6666666865348816, 0.5490196347236633, 0.6666666865348816, 0.5686274766921997, 0.6078431606292725, 0.7254902124404907, 0.686274528503418, 0.7450980544090271, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.7647058963775635, 0.7450980544090271, 0.7843137383460999, 0.7450980544090271, 0.7647058963775635, 0.7647058963775635, 0.843137264251709, 0.8039215803146362, 0.843137264251709, 0.8235294222831726, 0.8235294222831726, 0.9019607901573181, 0.9019607901573181, 0.9607843160629272, 0.9803921580314636, 0.9803921580314636], [0.03921568766236305, 0.1568627506494522, 0.1568627506494522, 0.1764705926179886, 0.09803921729326248, 0.13725490868091583, 0.1568627506494522, 0.23529411852359772, 0.0784313753247261, 0.29411765933036804, 0.29411765933036804, 0.3137255012989044, 0.3333333432674408, 0.3529411852359772, 0.3333333432674408, 0.3529411852359772, 0.27450981736183167, 0.37254902720451355, 0.3529411852359772, 0.3529411852359772, 0.3921568691730499, 0.3529411852359772, 0.3529411852359772, 0.3921568691730499, 0.37254902720451355, 0.3333333432674408, 0.3529411852359772, 0.4117647111415863, 0.3921568691730499, 0.3921568691730499, 0.343137264251709, 0.3137255012989044, 0.3921568691730499, 0.4117647111415863, 0.45098039507865906, 0.45098039507865906, 0.4313725531101227, 0.4313725531101227, 0.5098039507865906, 0.46078431606292725, 0.4313725531101227, 0.4313725531101227, 0.47058823704719543, 0.5098039507865906, 0.4313725531101227, NaN, 0.4313725531101227, 0.5098039507865906, 0.47058823704719543, 0.5490196347236633, 0.5490196347236633, 0.47058823704719543, 0.5098039507865906, 0.5686274766921997, 0.529411792755127, 0.529411792755127, 0.5490196347236633, 0.5098039507865906, 0.5784313678741455, 0.6274510025978088, 0.6666666865348816, 0.6274510025978088, 0.6666666865348816, 0.686274528503418, 0.7058823704719543, 0.6470588445663452, 0.6470588445663452, 0.7450980544090271, 0.7450980544090271, 0.7058823704719543, 0.7647058963775635, 0.7647058963775635, 0.7254902124404907, 0.7254902124404907, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.843137264251709, 0.9019607901573181, 0.9803921580314636], [0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.1764705926179886, 0.23529411852359772, 0.27450981736183167, 0.2549019753932953, 0.19607843458652496, 0.27450981736183167, 0.23529411852359772, 0.29411765933036804, 0.23529411852359772, 0.19607843458652496, 0.27450981736183167, 0.3137255012989044, 0.29411765933036804, 0.3529411852359772, 0.37254902720451355, 0.23529411852359772, 0.3137255012989044, 0.3333333432674408, 0.2549019753932953, 0.3921568691730499, 0.45098039507865906, 0.37254902720451355, 0.4313725531101227, 0.3921568691730499, 0.37254902720451355, 0.3529411852359772, 0.3921568691730499, 0.4215686321258545, 0.3333333432674408, 0.5098039507865906, 0.4901960790157318, 0.4117647111415863, 0.45098039507865906, 0.5098039507865906, 0.45098039507865906, 0.5686274766921997, NaN, 0.4313725531101227, 0.5686274766921997, 0.45098039507865906, 0.47058823704719543, 0.47058823704719543, 0.5490196347236633, 0.529411792755127, 0.6078431606292725, 0.6666666865348816, 0.6470588445663452, 0.5882353186607361, 0.7058823704719543, 0.6470588445663452, 0.686274528503418, 0.7058823704719543, 0.7843137383460999, 0.7647058963775635, 0.7843137383460999, 0.843137264251709, 0.7843137383460999, 0.7254902124404907, 0.8235294222831726, 0.8627451062202454, 0.8235294222831726, 0.843137264251709, 0.8235294222831726, 0.8627451062202454, 0.8823529481887817, 0.8627451062202454, 0.8627451062202454, 0.9411764740943909, 0.9019607901573181, 0.9803921580314636], [0.05882352963089943, 0.09803921729326248, 0.11764705926179886, 0.05882352963089943, 0.11764705926179886, 0.1568627506494522, 0.13725490868091583, 0.09803921729326248, 0.11764705926179886, 0.1568627506494522, 0.27450981736183167, 0.21568627655506134, 0.29411765933036804, 0.23529411852359772, 0.1764705926179886, 0.23529411852359772, 0.3137255012989044, 0.2549019753932953, 0.3529411852359772, 0.3333333432674408, 0.27450981736183167, 0.27450981736183167, 0.2549019753932953, 0.2549019753932953, 0.3137255012989044, 0.3137255012989044, 0.2549019753932953, 0.3529411852359772, 0.37254902720451355, 0.27450981736183167, 0.3333333432674408, 0.3529411852359772, 0.3333333432674408, 0.3529411852359772, 0.4117647111415863, 0.4901960790157318, 0.3921568691730499, 0.4313725531101227, 0.21568627655506134, 0.4117647111415863, 0.47058823704719543, 0.4117647111415863, 0.4901960790157318, 0.3333333432674408, 0.37254902720451355, 0.4901960790157318, 0.5686274766921997, NaN, 0.5490196347236633, 0.5882353186607361, 0.5098039507865906, 0.4901960790157318, 0.5882353186607361, 0.5490196347236633, 0.5882353186607361, 0.5686274766921997, 0.6078431606292725, 0.529411792755127, 0.5882353186607361, 0.686274528503418, 0.686274528503418, 0.6666666865348816, 0.7647058963775635, 0.7647058963775635, 0.686274528503418, 0.7647058963775635, 0.6666666865348816, 0.7647058963775635, 0.7058823704719543, 0.7254902124404907, 0.8039215803146362, 0.8235294222831726, 0.8627451062202454, 0.8627451062202454, 0.8039215803146362, 0.8823529481887817, 0.9215686321258545, 0.9215686321258545, 0.9215686321258545, 0.9607843160629272], [0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.03921568766236305, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.21568627655506134, 0.23529411852359772, 0.1764705926179886, 0.2549019753932953, 0.1568627506494522, 0.21568627655506134, 0.1568627506494522, 0.1764705926179886, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.3333333432674408, 0.1764705926179886, 0.27450981736183167, 0.27450981736183167, 0.3333333432674408, 0.1764705926179886, 0.3333333432674408, 0.2549019753932953, 0.3333333432674408, 0.45098039507865906, 0.3137255012989044, 0.37254902720451355, 0.47058823704719543, 0.27450981736183167, 0.3921568691730499, 0.37254902720451355, 0.4117647111415863, 0.3529411852359772, 0.4901960790157318, 0.37254902720451355, 0.3921568691730499, 0.45098039507865906, 0.4901960790157318, 0.4901960790157318, 0.529411792755127, 0.4313725531101227, 0.45098039507865906, NaN, 0.4313725531101227, 0.46078431606292725, 0.4313725531101227, 0.7156862616539001, 0.5098039507865906, 0.6470588445663452, 0.6470588445663452, 0.6470588445663452, 0.5098039507865906, 0.7058823704719543, 0.5490196347236633, 0.6470588445663452, 0.6470588445663452, 0.7058823704719543, 0.7254902124404907, 0.7843137383460999, 0.6078431606292725, 0.7254902124404907, 0.7254902124404907, 0.9411764740943909, 0.8823529481887817, 0.843137264251709, 0.7647058963775635, 0.7843137383460999, 0.9215686321258545, 0.7843137383460999, 0.8235294222831726, 0.8823529481887817, 0.9411764740943909, 0.8823529481887817, 1.0], [0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.09803921729326248, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.21568627655506134, 0.19607843458652496, 0.23529411852359772, 0.19607843458652496, 0.13725490868091583, 0.1764705926179886, 0.23529411852359772, 0.1764705926179886, 0.27450981736183167, 0.27450981736183167, 0.27450981736183167, 0.2549019753932953, 0.21568627655506134, 0.21568627655506134, 0.23529411852359772, 0.29411765933036804, 0.23529411852359772, 0.27450981736183167, 0.3529411852359772, 0.27450981736183167, 0.3137255012989044, 0.2549019753932953, 0.2549019753932953, 0.3333333432674408, 0.4313725531101227, 0.4313725531101227, 0.4019607901573181, 0.37254902720451355, 0.1568627506494522, 0.3529411852359772, 0.45098039507865906, 0.4117647111415863, 0.4313725531101227, 0.4117647111415863, 0.3921568691730499, 0.45098039507865906, 0.5490196347236633, 0.4117647111415863, 0.5686274766921997, NaN, 0.529411792755127, 0.4117647111415863, 0.5882353186607361, 0.5686274766921997, 0.5490196347236633, 0.5882353186607361, 0.5686274766921997, 0.5098039507865906, 0.6274510025978088, 0.6666666865348816, 0.686274528503418, 0.686274528503418, 0.7254902124404907, 0.7254902124404907, 0.686274528503418, 0.7843137383460999, 0.686274528503418, 0.6666666865348816, 0.7058823704719543, 0.6666666865348816, 0.7843137383460999, 0.7843137383460999, 0.843137264251709, 0.843137264251709, 0.8235294222831726, 0.9019607901573181, 0.9215686321258545, 0.9215686321258545, 0.9607843160629272, 0.9607843160629272], [0.019607843831181526, 0.0784313753247261, 0.03921568766236305, 0.03921568766236305, 0.09803921729326248, 0.11764705926179886, 0.0784313753247261, 0.05882352963089943, 0.11764705926179886, 0.1568627506494522, 0.23529411852359772, 0.13725490868091583, 0.1764705926179886, 0.19607843458652496, 0.19607843458652496, 0.1764705926179886, 0.1568627506494522, 0.1764705926179886, 0.2549019753932953, 0.21568627655506134, 0.19607843458652496, 0.3333333432674408, 0.2647058963775635, 0.23529411852359772, 0.23529411852359772, 0.23529411852359772, 0.27450981736183167, 0.3333333432674408, 0.3333333432674408, 0.29411765933036804, 0.3137255012989044, 0.3529411852359772, 0.37254902720451355, 0.29411765933036804, 0.3333333432674408, 0.4313725531101227, 0.3137255012989044, 0.37254902720451355, 0.3921568691730499, 0.3529411852359772, 0.2549019753932953, 0.3921568691730499, 0.45098039507865906, 0.37254902720451355, 0.45098039507865906, 0.45098039507865906, 0.529411792755127, 0.4901960790157318, 0.5392156839370728, 0.47058823704719543, NaN, 0.4901960790157318, 0.5882353186607361, 0.5098039507865906, 0.5098039507865906, 0.5882353186607361, 0.5098039507865906, 0.5588235259056091, 0.5882353186607361, 0.6274510025978088, 0.6666666865348816, 0.7058823704719543, 0.7647058963775635, 0.7450980544090271, 0.686274528503418, 0.9215686321258545, 0.7254902124404907, 0.7450980544090271, 0.7254902124404907, 0.6666666865348816, 0.843137264251709, 0.8039215803146362, 0.843137264251709, 0.843137264251709, 0.843137264251709, 0.843137264251709, 0.8627451062202454, 0.9411764740943909, 0.9411764740943909, 0.9607843160629272], [0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.03921568766236305, 0.03921568766236305, 0.13725490868091583, 0.05882352963089943, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.2549019753932953, 0.1568627506494522, 0.21568627655506134, 0.21568627655506134, 0.09803921729326248, 0.13725490868091583, 0.1764705926179886, 0.13725490868091583, 0.3137255012989044, 0.21568627655506134, 0.21568627655506134, 0.29411765933036804, 0.11764705926179886, 0.1764705926179886, 0.23529411852359772, 0.2549019753932953, 0.13725490868091583, 0.3333333432674408, 0.3137255012989044, 0.27450981736183167, 0.3529411852359772, 0.3921568691730499, 0.2549019753932953, 0.3921568691730499, 0.3921568691730499, 0.4117647111415863, 0.27450981736183167, 0.3921568691730499, 0.3921568691730499, 0.4117647111415863, 0.3529411852359772, 0.3333333432674408, 0.47058823704719543, 0.4313725531101227, 0.3921568691730499, 0.529411792755127, 0.529411792755127, 0.5098039507865906, 0.5686274766921997, 0.5882353186607361, 0.5098039507865906, NaN, 0.5882353186607361, 0.5686274766921997, 0.5882353186607361, 0.5882353186607361, 0.6078431606292725, 0.529411792755127, 0.6470588445663452, 0.6274510025978088, 0.7843137383460999, 0.6274510025978088, 0.7254902124404907, 0.7254902124404907, 0.7058823704719543, 0.7450980544090271, 0.7254902124404907, 0.7450980544090271, 0.7254902124404907, 0.7843137383460999, 0.7647058963775635, 0.7647058963775635, 0.7843137383460999, 0.7843137383460999, 0.8627451062202454, 0.843137264251709, 0.8627451062202454, 0.9019607901573181, 0.8823529481887817, 0.9607843160629272], [0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.2549019753932953, 0.19607843458652496, 0.23529411852359772, 0.1764705926179886, 0.1764705926179886, 0.1764705926179886, 0.23529411852359772, 0.27450981736183167, 0.3333333432674408, 0.19607843458652496, 0.23529411852359772, 0.27450981736183167, 0.3137255012989044, 0.19607843458652496, 0.29411765933036804, 0.21568627655506134, 0.3137255012989044, 0.3921568691730499, 0.2549019753932953, 0.3333333432674408, 0.4117647111415863, 0.2549019753932953, 0.3333333432674408, 0.3333333432674408, 0.3529411852359772, 0.3921568691730499, 0.47058823704719543, 0.3529411852359772, 0.3333333432674408, 0.37254902720451355, 0.4901960790157318, 0.45098039507865906, 0.4901960790157318, 0.45098039507865906, 0.4117647111415863, 0.28431373834609985, 0.4117647111415863, 0.4117647111415863, 0.4117647111415863, NaN, 0.5098039507865906, 0.6078431606292725, 0.47058823704719543, 0.6078431606292725, 0.4313725531101227, 0.6078431606292725, 0.529411792755127, 0.5882353186607361, 0.6470588445663452, 0.686274528503418, 0.686274528503418, 0.7647058963775635, 0.5490196347236633, 0.7058823704719543, 0.6666666865348816, 0.8235294222831726, 0.843137264251709, 0.8235294222831726, 0.686274528503418, 0.8039215803146362, 0.843137264251709, 0.7647058963775635, 0.7647058963775635, 0.7647058963775635, 0.9803921580314636, 0.8823529481887817, 0.9803921580314636], [0.0784313753247261, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.11764705926179886, 0.05882352963089943, 0.13725490868091583, 0.0784313753247261, 0.0784313753247261, 0.0784313753247261, 0.2549019753932953, 0.1568627506494522, 0.21568627655506134, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.13725490868091583, 0.21568627655506134, 0.3333333432674408, 0.19607843458652496, 0.2549019753932953, 0.3137255012989044, 0.21568627655506134, 0.21568627655506134, 0.19607843458652496, 0.28431373834609985, 0.23529411852359772, 0.1764705926179886, 0.27450981736183167, 0.23529411852359772, 0.3333333432674408, 0.21568627655506134, 0.27450981736183167, 0.3333333432674408, 0.3137255012989044, 0.3529411852359772, 0.2549019753932953, 0.37254902720451355, 0.3137255012989044, 0.3921568691730499, 0.3333333432674408, 0.3529411852359772, 0.3529411852359772, 0.4313725531101227, 0.3333333432674408, 0.4313725531101227, 0.47058823704719543, 0.45098039507865906, 0.4901960790157318, 0.4313725531101227, 0.4901960790157318, 0.4313725531101227, 0.4901960790157318, NaN, 0.5490196347236633, 0.529411792755127, 0.5490196347236633, 0.45098039507865906, 0.5686274766921997, 0.7058823704719543, 0.6470588445663452, 0.6078431606292725, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6274510025978088, 0.7254902124404907, 0.7254902124404907, 0.7058823704719543, 0.8039215803146362, 0.9803921580314636, 0.7647058963775635, 0.7843137383460999, 0.843137264251709, 0.8627451062202454, 0.8823529481887817, 0.8627451062202454, 0.9215686321258545, 0.9607843160629272], [0.05882352963089943, 0.19607843458652496, 0.11764705926179886, 0.1764705926179886, 0.1568627506494522, 0.19607843458652496, 0.1568627506494522, 0.23529411852359772, 0.21568627655506134, 0.23529411852359772, 0.1764705926179886, 0.2549019753932953, 0.29411765933036804, 0.37254902720451355, 0.2549019753932953, 0.29411765933036804, 0.29411765933036804, 0.27450981736183167, 0.3333333432674408, 0.27450981736183167, 0.3529411852359772, 0.27450981736183167, 0.29411765933036804, 0.29411765933036804, 0.29411765933036804, 0.27450981736183167, 0.3137255012989044, 0.3529411852359772, 0.3137255012989044, 0.3529411852359772, 0.37254902720451355, 0.3529411852359772, 0.3921568691730499, 0.4117647111415863, 0.37254902720451355, 0.4117647111415863, 0.4117647111415863, 0.3137255012989044, 0.37254902720451355, 0.45098039507865906, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.4117647111415863, 0.45098039507865906, 0.47058823704719543, 0.3921568691730499, 0.4117647111415863, 0.3529411852359772, 0.45098039507865906, 0.4901960790157318, 0.4117647111415863, 0.3921568691730499, 0.45098039507865906, NaN, 0.3529411852359772, 0.5490196347236633, 0.47058823704719543, 0.4019607901573181, 0.5098039507865906, 0.5098039507865906, 0.5490196347236633, 0.5098039507865906, 0.529411792755127, 0.5490196347236633, 0.5686274766921997, 0.6078431606292725, 0.686274528503418, 0.6470588445663452, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.6470588445663452, 0.686274528503418, 0.7058823704719543, 0.7058823704719543, 0.7843137383460999, 0.7450980544090271, 0.9803921580314636], [0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.0784313753247261, 0.11764705926179886, 0.0784313753247261, 0.09803921729326248, 0.13725490868091583, 0.1568627506494522, 0.21568627655506134, 0.1568627506494522, 0.1764705926179886, 0.19607843458652496, 0.13725490868091583, 0.1568627506494522, 0.09803921729326248, 0.11764705926179886, 0.3137255012989044, 0.21568627655506134, 0.2549019753932953, 0.29411765933036804, 0.13725490868091583, 0.21568627655506134, 0.19607843458652496, 0.27450981736183167, 0.1764705926179886, 0.3137255012989044, 0.19607843458652496, 0.23529411852359772, 0.3529411852359772, 0.19607843458652496, 0.27450981736183167, 0.4313725531101227, 0.2549019753932953, 0.3333333432674408, 0.29411765933036804, 0.4117647111415863, 0.37254902720451355, 0.47058823704719543, 0.3137255012989044, 0.3137255012989044, 0.4313725531101227, 0.47058823704719543, 0.3333333432674408, 0.47058823704719543, 0.3333333432674408, 0.4313725531101227, 0.3529411852359772, 0.4117647111415863, 0.4117647111415863, 0.4117647111415863, 0.529411792755127, 0.47058823704719543, 0.6470588445663452, NaN, 0.6470588445663452, 0.4901960790157318, 0.7058823704719543, 0.5490196347236633, 0.6274510025978088, 0.5686274766921997, 0.686274528503418, 0.7058823704719543, 0.7058823704719543, 0.686274528503418, 0.6666666865348816, 0.686274528503418, 0.8039215803146362, 0.8823529481887817, 0.7647058963775635, 0.7450980544090271, 0.8039215803146362, 0.843137264251709, 0.686274528503418, 0.8039215803146362, 0.7843137383460999, 0.9411764740943909, 0.8627451062202454, 0.9803921580314636], [0.03921568766236305, 0.13725490868091583, 0.09803921729326248, 0.13725490868091583, 0.11764705926179886, 0.19607843458652496, 0.13725490868091583, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.1764705926179886, 0.27450981736183167, 0.23529411852359772, 0.3333333432674408, 0.23529411852359772, 0.29411765933036804, 0.29411765933036804, 0.27450981736183167, 0.29411765933036804, 0.2549019753932953, 0.3333333432674408, 0.27450981736183167, 0.27450981736183167, 0.27450981736183167, 0.29411765933036804, 0.2549019753932953, 0.29411765933036804, 0.37254902720451355, 0.3137255012989044, 0.3529411852359772, 0.3529411852359772, 0.3529411852359772, 0.3921568691730499, 0.4117647111415863, 0.37254902720451355, 0.3921568691730499, 0.4117647111415863, 0.3137255012989044, 0.37254902720451355, 0.45098039507865906, 0.4117647111415863, 0.4117647111415863, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.45098039507865906, 0.3529411852359772, 0.3921568691730499, 0.3529411852359772, 0.4313725531101227, 0.4901960790157318, 0.3921568691730499, 0.3921568691730499, 0.45098039507865906, 0.45098039507865906, 0.3529411852359772, NaN, 0.47058823704719543, 0.4019607901573181, 0.5098039507865906, 0.5098039507865906, 0.529411792755127, 0.5098039507865906, 0.529411792755127, 0.5490196347236633, 0.5686274766921997, 0.6078431606292725, 0.686274528503418, 0.6470588445663452, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.6274510025978088, 0.6470588445663452, 0.686274528503418, 0.7058823704719543, 0.7058823704719543, 0.7843137383460999, 0.7450980544090271, 0.9803921580314636], [0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.0784313753247261, 0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.1764705926179886, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.03921568766236305, 0.1568627506494522, 0.13725490868091583, 0.1568627506494522, 0.27450981736183167, 0.1568627506494522, 0.1764705926179886, 0.3137255012989044, 0.019607843831181526, 0.13725490868091583, 0.21568627655506134, 0.21568627655506134, 0.09803921729326248, 0.23529411852359772, 0.2549019753932953, 0.2549019753932953, 0.3137255012989044, 0.2549019753932953, 0.23529411852359772, 0.37254902720451355, 0.3137255012989044, 0.3137255012989044, 0.27450981736183167, 0.3529411852359772, 0.4117647111415863, 0.4117647111415863, 0.3137255012989044, 0.3921568691730499, 0.3921568691730499, 0.4313725531101227, 0.3921568691730499, 0.4901960790157318, 0.4117647111415863, 0.47058823704719543, 0.4901960790157318, 0.4901960790157318, 0.44117647409439087, 0.47058823704719543, 0.5686274766921997, 0.5490196347236633, 0.529411792755127, 0.5098039507865906, 0.529411792755127, NaN, 0.5490196347236633, 0.6078431606292725, 0.7254902124404907, 0.6666666865348816, 0.7058823704719543, 0.7254902124404907, 0.7058823704719543, 0.7058823704719543, 0.7058823704719543, 0.7254902124404907, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.7843137383460999, 0.8039215803146362, 0.7647058963775635, 0.8039215803146362, 0.7843137383460999, 0.7843137383460999, 0.9411764740943909, 0.9019607901573181, 0.9411764740943909], [0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.09803921729326248, 0.19607843458652496, 0.11764705926179886, 0.13725490868091583, 0.19607843458652496, 0.05882352963089943, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.21568627655506134, 0.23529411852359772, 0.21568627655506134, 0.2549019753932953, 0.09803921729326248, 0.11764705926179886, 0.1568627506494522, 0.21568627655506134, 0.13725490868091583, 0.29411765933036804, 0.1764705926179886, 0.19607843458652496, 0.3235294222831726, 0.19607843458652496, 0.19607843458652496, 0.36274510622024536, 0.21568627655506134, 0.29411765933036804, 0.27450981736183167, 0.29411765933036804, 0.3529411852359772, 0.38235294818878174, 0.2549019753932953, 0.29411765933036804, 0.4117647111415863, 0.4019607901573181, 0.27450981736183167, 0.4215686321258545, 0.29411765933036804, 0.4117647111415863, 0.29411765933036804, 0.37254902720451355, 0.4117647111415863, 0.3529411852359772, 0.3921568691730499, 0.4313725531101227, 0.5980392098426819, 0.29411765933036804, 0.5980392098426819, 0.45098039507865906, NaN, 0.5098039507865906, 0.6274510025978088, 0.5490196347236633, 0.6666666865348816, 0.6666666865348816, 0.6764705777168274, 0.5882353186607361, 0.5882353186607361, 0.6666666865348816, 0.7254902124404907, 0.7843137383460999, 0.7549019455909729, 0.7647058963775635, 0.7647058963775635, 0.843137264251709, 0.6666666865348816, 0.7843137383460999, 0.7647058963775635, 0.9215686321258545, 0.843137264251709, 0.9803921580314636], [0.05882352963089943, 0.05882352963089943, 0.03921568766236305, 0.0784313753247261, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.05882352963089943, 0.019607843831181526, 0.09803921729326248, 0.1568627506494522, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.11764705926179886, 0.1764705926179886, 0.13725490868091583, 0.13725490868091583, 0.21568627655506134, 0.1764705926179886, 0.1568627506494522, 0.1568627506494522, 0.1764705926179886, 0.1568627506494522, 0.19607843458652496, 0.21568627655506134, 0.2549019753932953, 0.27450981736183167, 0.19607843458652496, 0.23529411852359772, 0.3137255012989044, 0.2549019753932953, 0.21568627655506134, 0.1568627506494522, 0.29411765933036804, 0.2549019753932953, 0.3333333432674408, 0.21568627655506134, 0.2549019753932953, 0.27450981736183167, 0.3333333432674408, 0.3137255012989044, 0.37254902720451355, 0.3529411852359772, 0.3137255012989044, 0.45098039507865906, 0.3333333432674408, 0.37254902720451355, 0.37254902720451355, 0.47058823704719543, 0.29411765933036804, 0.4901960790157318, 0.45098039507865906, 0.4901960790157318, 0.3921568691730499, 0.4901960790157318, NaN, 0.5686274766921997, 0.529411792755127, 0.6274510025978088, 0.6274510025978088, 0.5686274766921997, 0.5882353186607361, 0.5686274766921997, 0.6470588445663452, 0.686274528503418, 0.6274510025978088, 0.7254902124404907, 0.7843137383460999, 0.7450980544090271, 0.686274528503418, 0.7254902124404907, 0.7843137383460999, 0.7843137383460999, 0.843137264251709, 0.8823529481887817, 0.9411764740943909], [0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.19607843458652496, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.23529411852359772, 0.11764705926179886, 0.13725490868091583, 0.13725490868091583, 0.1764705926179886, 0.13725490868091583, 0.21568627655506134, 0.23529411852359772, 0.21568627655506134, 0.2549019753932953, 0.22549019753932953, 0.21568627655506134, 0.27450981736183167, 0.23529411852359772, 0.3137255012989044, 0.27450981736183167, 0.23529411852359772, 0.29411765933036804, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.3333333432674408, 0.23529411852359772, 0.2549019753932953, 0.3333333432674408, 0.3137255012989044, 0.3137255012989044, 0.3529411852359772, 0.3137255012989044, 0.3333333432674408, 0.21568627655506134, 0.4117647111415863, 0.3529411852359772, 0.4901960790157318, 0.37254902720451355, 0.4901960790157318, 0.27450981736183167, 0.37254902720451355, 0.4313725531101227, NaN, 0.47058823704719543, 0.6666666865348816, 0.6470588445663452, 0.5098039507865906, 0.5686274766921997, 0.45098039507865906, 0.529411792755127, 0.529411792755127, 0.5490196347236633, 0.5882353186607361, 0.6274510025978088, 0.7647058963775635, 0.7647058963775635, 0.5882353186607361, 0.7647058963775635, 0.7647058963775635, 0.8235294222831726, 0.843137264251709, 0.9215686321258545], [0.0, 0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.0784313753247261, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.03921568766236305, 0.1568627506494522, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.11764705926179886, 0.2549019753932953, 0.1568627506494522, 0.11764705926179886, 0.09803921729326248, 0.1764705926179886, 0.19607843458652496, 0.19607843458652496, 0.19607843458652496, 0.1764705926179886, 0.3137255012989044, 0.23529411852359772, 0.19607843458652496, 0.3137255012989044, 0.2549019753932953, 0.2549019753932953, 0.1568627506494522, 0.29411765933036804, 0.3333333432674408, 0.37254902720451355, 0.29411765933036804, 0.29411765933036804, 0.2549019753932953, 0.3333333432674408, 0.23529411852359772, 0.37254902720451355, 0.29411765933036804, 0.3333333432674408, 0.3529411852359772, 0.3137255012989044, 0.29411765933036804, 0.37254902720451355, 0.3529411852359772, 0.3921568691730499, 0.45098039507865906, 0.4313725531101227, 0.47058823704719543, 0.3333333432674408, 0.45098039507865906, 0.47058823704719543, 0.529411792755127, NaN, 0.47058823704719543, 0.47058823704719543, 0.529411792755127, 0.4901960790157318, 0.813725471496582, 0.6078431606292725, 0.5882353186607361, 0.6470588445663452, 0.6470588445663452, 0.6078431606292725, 0.4901960790157318, 0.6078431606292725, 0.9019607901573181, 0.7450980544090271, 0.7450980544090271, 0.7450980544090271, 0.7647058963775635, 0.9215686321258545], [0.05882352963089943, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.11764705926179886, 0.09803921729326248, 0.13725490868091583, 0.1764705926179886, 0.09803921729326248, 0.1568627506494522, 0.1764705926179886, 0.1568627506494522, 0.19607843458652496, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.19607843458652496, 0.1764705926179886, 0.19607843458652496, 0.2549019753932953, 0.1764705926179886, 0.22549019753932953, 0.1568627506494522, 0.19607843458652496, 0.1568627506494522, 0.21568627655506134, 0.1568627506494522, 0.21568627655506134, 0.23529411852359772, 0.1764705926179886, 0.29411765933036804, 0.21568627655506134, 0.21568627655506134, 0.27450981736183167, 0.2549019753932953, 0.29411765933036804, 0.23529411852359772, 0.23529411852359772, 0.19607843458652496, 0.3137255012989044, 0.2549019753932953, 0.27450981736183167, 0.3137255012989044, 0.23529411852359772, 0.23529411852359772, 0.3333333432674408, 0.21568627655506134, 0.23529411852359772, 0.29411765933036804, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.3137255012989044, 0.3333333432674408, 0.4901960790157318, 0.3137255012989044, 0.4901960790157318, 0.29411765933036804, 0.3333333432674408, 0.37254902720451355, 0.3333333432674408, 0.529411792755127, NaN, 0.5196078419685364, 0.4313725531101227, 0.4313725531101227, 0.529411792755127, 0.529411792755127, 0.5686274766921997, 0.5490196347236633, 0.5686274766921997, 0.5490196347236633, 0.9607843160629272, 0.5882353186607361, 0.5490196347236633, 0.7058823704719543, 0.7254902124404907, 0.7843137383460999, 0.8039215803146362, 0.843137264251709], [0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.11764705926179886, 0.09803921729326248, 0.1568627506494522, 0.1764705926179886, 0.09803921729326248, 0.1568627506494522, 0.21568627655506134, 0.13725490868091583, 0.21568627655506134, 0.13725490868091583, 0.1764705926179886, 0.1764705926179886, 0.19607843458652496, 0.1764705926179886, 0.21568627655506134, 0.23529411852359772, 0.1764705926179886, 0.23529411852359772, 0.1568627506494522, 0.1764705926179886, 0.1764705926179886, 0.21568627655506134, 0.1568627506494522, 0.21568627655506134, 0.23529411852359772, 0.19607843458652496, 0.3137255012989044, 0.21568627655506134, 0.21568627655506134, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.21568627655506134, 0.21568627655506134, 0.19607843458652496, 0.3137255012989044, 0.23529411852359772, 0.23529411852359772, 0.3137255012989044, 0.23529411852359772, 0.19607843458652496, 0.3137255012989044, 0.23529411852359772, 0.23529411852359772, 0.27450981736183167, 0.27450981736183167, 0.2549019753932953, 0.27450981736183167, 0.3137255012989044, 0.3333333432674408, 0.47058823704719543, 0.29411765933036804, 0.47058823704719543, 0.27450981736183167, 0.3333333432674408, 0.37254902720451355, 0.3529411852359772, 0.529411792755127, 0.4803921580314636, NaN, 0.45098039507865906, 0.45098039507865906, 0.529411792755127, 0.529411792755127, 0.5490196347236633, 0.529411792755127, 0.5490196347236633, 0.529411792755127, 0.9803921580314636, 0.5686274766921997, 0.5490196347236633, 0.7058823704719543, 0.7058823704719543, 0.7843137383460999, 0.8039215803146362, 0.843137264251709], [0.019607843831181526, 0.0, 0.0, 0.019607843831181526, 0.019607843831181526, 0.0784313753247261, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.12745098769664764, 0.05882352963089943, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.11764705926179886, 0.1568627506494522, 0.0784313753247261, 0.13725490868091583, 0.11764705926179886, 0.18627451360225677, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.1568627506494522, 0.13725490868091583, 0.11764705926179886, 0.09803921729326248, 0.1764705926179886, 0.22549019753932953, 0.19607843458652496, 0.19607843458652496, 0.22549019753932953, 0.1568627506494522, 0.1764705926179886, 0.1764705926179886, 0.22549019753932953, 0.29411765933036804, 0.2647058963775635, 0.21568627655506134, 0.2549019753932953, 0.3137255012989044, 0.28431373834609985, 0.23529411852359772, 0.29411765933036804, 0.21568627655506134, 0.3137255012989044, 0.21568627655506134, 0.3137255012989044, 0.3137255012989044, 0.29411765933036804, 0.23529411852359772, 0.3333333432674408, 0.45098039507865906, 0.29411765933036804, 0.45098039507865906, 0.29411765933036804, 0.3235294222831726, 0.4313725531101227, 0.4901960790157318, 0.47058823704719543, 0.5686274766921997, 0.5490196347236633, NaN, 0.45098039507865906, 0.47058823704719543, 0.5098039507865906, 0.6078431606292725, 0.5784313678741455, 0.656862735748291, 0.5686274766921997, 0.7254902124404907, 0.6666666865348816, 0.6666666865348816, 0.7647058963775635, 0.7647058963775635, 0.8039215803146362, 0.8039215803146362, 0.9803921580314636], [0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.05882352963089943, 0.13725490868091583, 0.03921568766236305, 0.11764705926179886, 0.019607843831181526, 0.0784313753247261, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.13725490868091583, 0.1568627506494522, 0.019607843831181526, 0.19607843458652496, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.13725490868091583, 0.09803921729326248, 0.1764705926179886, 0.11764705926179886, 0.1764705926179886, 0.27450981736183167, 0.0784313753247261, 0.1764705926179886, 0.29411765933036804, 0.09803921729326248, 0.21568627655506134, 0.11764705926179886, 0.2549019753932953, 0.1568627506494522, 0.3137255012989044, 0.03921568766236305, 0.13725490868091583, 0.2549019753932953, 0.27450981736183167, 0.2549019753932953, 0.3529411852359772, 0.1568627506494522, 0.23529411852359772, 0.3921568691730499, 0.21568627655506134, 0.0784313753247261, 0.2549019753932953, 0.45098039507865906, 0.3333333432674408, 0.4313725531101227, 0.3137255012989044, 0.4313725531101227, 0.29411765933036804, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.5098039507865906, 0.5686274766921997, 0.5490196347236633, 0.5490196347236633, NaN, 0.529411792755127, 0.5490196347236633, 0.6470588445663452, 0.5882353186607361, 0.6666666865348816, 0.5490196347236633, 0.7058823704719543, 0.6666666865348816, 0.6274510025978088, 0.7058823704719543, 0.7647058963775635, 0.843137264251709, 0.843137264251709, 0.9215686321258545], [0.0, 0.03921568766236305, 0.0, 0.0, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.09803921729326248, 0.0784313753247261, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.05882352963089943, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.1568627506494522, 0.09803921729326248, 0.23529411852359772, 0.11764705926179886, 0.09803921729326248, 0.05882352963089943, 0.1568627506494522, 0.11764705926179886, 0.1568627506494522, 0.1568627506494522, 0.13725490868091583, 0.2549019753932953, 0.1568627506494522, 0.1764705926179886, 0.29411765933036804, 0.1764705926179886, 0.1764705926179886, 0.1568627506494522, 0.2549019753932953, 0.27450981736183167, 0.3137255012989044, 0.23529411852359772, 0.23529411852359772, 0.2549019753932953, 0.3137255012989044, 0.21568627655506134, 0.3529411852359772, 0.21568627655506134, 0.3333333432674408, 0.27450981736183167, 0.3137255012989044, 0.27450981736183167, 0.27450981736183167, 0.29411765933036804, 0.37254902720451355, 0.3921568691730499, 0.3333333432674408, 0.3921568691730499, 0.29411765933036804, 0.4117647111415863, 0.4313725531101227, 0.5490196347236633, 0.18627451360225677, 0.47058823704719543, 0.47058823704719543, 0.529411792755127, 0.47058823704719543, NaN, 0.5686274766921997, 0.5098039507865906, 0.5882353186607361, 0.6274510025978088, 0.5882353186607361, 0.4901960790157318, 0.5490196347236633, 0.813725471496582, 0.7254902124404907, 0.7058823704719543, 0.6666666865348816, 0.7450980544090271, 0.9215686321258545], [0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.0784313753247261, 0.0784313753247261, 0.05882352963089943, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.1568627506494522, 0.0784313753247261, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.1764705926179886, 0.1764705926179886, 0.1764705926179886, 0.1568627506494522, 0.13725490868091583, 0.1568627506494522, 0.21568627655506134, 0.1764705926179886, 0.0784313753247261, 0.19607843458652496, 0.23529411852359772, 0.11764705926179886, 0.19607843458652496, 0.19607843458652496, 0.03921568766236305, 0.19607843458652496, 0.2549019753932953, 0.3333333432674408, 0.23529411852359772, 0.2549019753932953, 0.23529411852359772, 0.23529411852359772, 0.27450981736183167, 0.3529411852359772, 0.2549019753932953, 0.2549019753932953, 0.27450981736183167, 0.23529411852359772, 0.27450981736183167, 0.3333333432674408, 0.2549019753932953, 0.2549019753932953, 0.3333333432674408, 0.27450981736183167, 0.3137255012989044, 0.3137255012989044, 0.3137255012989044, 0.27450981736183167, 0.3333333432674408, 0.3529411852359772, 0.47058823704719543, 0.3921568691730499, 0.47058823704719543, 0.47058823704719543, 0.4901960790157318, 0.45098039507865906, 0.4313725531101227, NaN, 0.4901960790157318, 0.529411792755127, 0.5882353186607361, 0.4901960790157318, 0.6078431606292725, 0.529411792755127, 0.5490196347236633, 0.6078431606292725, 0.6078431606292725, 0.7254902124404907, 0.6470588445663452, 0.8823529481887817], [0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.0784313753247261, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.09803921729326248, 0.03921568766236305, 0.13725490868091583, 0.0784313753247261, 0.11764705926179886, 0.03921568766236305, 0.09803921729326248, 0.05882352963089943, 0.13725490868091583, 0.1568627506494522, 0.11764705926179886, 0.0784313753247261, 0.1568627506494522, 0.13725490868091583, 0.1764705926179886, 0.09803921729326248, 0.11764705926179886, 0.05882352963089943, 0.21568627655506134, 0.1568627506494522, 0.13725490868091583, 0.19607843458652496, 0.21568627655506134, 0.0784313753247261, 0.1568627506494522, 0.19607843458652496, 0.23529411852359772, 0.27450981736183167, 0.1764705926179886, 0.23529411852359772, 0.1764705926179886, 0.21568627655506134, 0.3333333432674408, 0.23529411852359772, 0.2549019753932953, 0.1764705926179886, 0.29411765933036804, 0.05882352963089943, 0.29411765933036804, 0.27450981736183167, 0.27450981736183167, 0.1764705926179886, 0.27450981736183167, 0.3529411852359772, 0.19607843458652496, 0.3529411852359772, 0.23529411852359772, 0.27450981736183167, 0.3137255012989044, 0.47058823704719543, 0.4117647111415863, 0.4313725531101227, 0.45098039507865906, 0.3921568691730499, 0.3529411852359772, 0.4901960790157318, 0.5098039507865906, NaN, 0.5098039507865906, 0.5490196347236633, 0.5490196347236633, 0.5882353186607361, 0.5686274766921997, 0.6078431606292725, 0.6666666865348816, 0.6666666865348816, 0.7647058963775635, 0.7058823704719543, 0.9607843160629272], [0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.03921568766236305, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.03921568766236305, 0.05882352963089943, 0.0784313753247261, 0.09803921729326248, 0.05882352963089943, 0.1764705926179886, 0.0784313753247261, 0.13725490868091583, 0.20588235557079315, 0.09803921729326248, 0.19607843458652496, 0.21568627655506134, 0.13725490868091583, 0.21568627655506134, 0.1764705926179886, 0.18627451360225677, 0.2549019753932953, 0.2647058963775635, 0.19607843458652496, 0.19607843458652496, 0.2549019753932953, 0.28431373834609985, 0.23529411852359772, 0.29411765933036804, 0.13725490868091583, 0.27450981736183167, 0.11764705926179886, 0.3333333432674408, 0.3333333432674408, 0.21568627655506134, 0.1568627506494522, 0.29411765933036804, 0.37254902720451355, 0.11764705926179886, 0.37254902720451355, 0.23529411852359772, 0.21568627655506134, 0.37254902720451355, 0.45098039507865906, 0.3529411852359772, 0.45098039507865906, 0.47058823704719543, 0.4215686321258545, 0.4117647111415863, 0.4117647111415863, 0.47058823704719543, 0.4901960790157318, NaN, 0.5, 0.5490196347236633, 0.5882353186607361, 0.6078431606292725, 0.5882353186607361, 0.7058823704719543, 0.7058823704719543, 0.7843137383460999, 0.7450980544090271, 0.9803921580314636], [0.0, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.0784313753247261, 0.03921568766236305, 0.09803921729326248, 0.03921568766236305, 0.019607843831181526, 0.0784313753247261, 0.05882352963089943, 0.09803921729326248, 0.03921568766236305, 0.05882352963089943, 0.09803921729326248, 0.09803921729326248, 0.05882352963089943, 0.11764705926179886, 0.09803921729326248, 0.13725490868091583, 0.05882352963089943, 0.1764705926179886, 0.11764705926179886, 0.11764705926179886, 0.11764705926179886, 0.13725490868091583, 0.11764705926179886, 0.05882352963089943, 0.0784313753247261, 0.11764705926179886, 0.22549019753932953, 0.09803921729326248, 0.11764705926179886, 0.21568627655506134, 0.11764705926179886, 0.11764705926179886, 0.11764705926179886, 0.20588235557079315, 0.1764705926179886, 0.22549019753932953, 0.09803921729326248, 0.0784313753247261, 0.1764705926179886, 0.20588235557079315, 0.1568627506494522, 0.23529411852359772, 0.1764705926179886, 0.19607843458652496, 0.1568627506494522, 0.21568627655506134, 0.1568627506494522, 0.23529411852359772, 0.1764705926179886, 0.19607843458652496, 0.37254902720451355, 0.23529411852359772, 0.37254902720451355, 0.19607843458652496, 0.2450980395078659, 0.27450981736183167, 0.4117647111415863, 0.3529411852359772, 0.4313725531101227, 0.45098039507865906, 0.343137264251709, 0.3333333432674408, 0.37254902720451355, 0.4117647111415863, 0.45098039507865906, 0.5, NaN, 0.4901960790157318, 0.5686274766921997, 0.5490196347236633, 0.5686274766921997, 0.6666666865348816, 0.6666666865348816, 0.7647058963775635, 0.7254902124404907, 0.9019607901573181], [0.0, 0.019607843831181526, 0.0, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.03921568766236305, 0.11764705926179886, 0.09803921729326248, 0.13725490868091583, 0.09803921729326248, 0.09803921729326248, 0.05882352963089943, 0.09803921729326248, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.1568627506494522, 0.09803921729326248, 0.1568627506494522, 0.1764705926179886, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.1764705926179886, 0.13725490868091583, 0.21568627655506134, 0.11764705926179886, 0.05882352963089943, 0.1568627506494522, 0.1764705926179886, 0.19607843458652496, 0.23529411852359772, 0.1568627506494522, 0.1764705926179886, 0.23529411852359772, 0.21568627655506134, 0.19607843458652496, 0.23529411852359772, 0.3137255012989044, 0.019607843831181526, 0.37254902720451355, 0.2549019753932953, 0.37254902720451355, 0.21568627655506134, 0.23529411852359772, 0.21568627655506134, 0.37254902720451355, 0.3921568691730499, 0.45098039507865906, 0.47058823704719543, 0.4313725531101227, 0.45098039507865906, 0.4117647111415863, 0.5098039507865906, 0.45098039507865906, 0.45098039507865906, 0.5098039507865906, NaN, 0.6666666865348816, 0.529411792755127, 0.6666666865348816, 0.686274528503418, 0.7058823704719543, 0.7450980544090271, 0.7450980544090271, 0.9019607901573181], [0.03921568766236305, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.0784313753247261, 0.09803921729326248, 0.11764705926179886, 0.0784313753247261, 0.11764705926179886, 0.11764705926179886, 0.13725490868091583, 0.13725490868091583, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.1568627506494522, 0.13725490868091583, 0.13725490868091583, 0.13725490868091583, 0.11764705926179886, 0.13725490868091583, 0.13725490868091583, 0.1764705926179886, 0.1568627506494522, 0.23529411852359772, 0.13725490868091583, 0.1568627506494522, 0.21568627655506134, 0.1764705926179886, 0.23529411852359772, 0.13725490868091583, 0.1764705926179886, 0.13725490868091583, 0.2549019753932953, 0.19607843458652496, 0.11764705926179886, 0.11764705926179886, 0.13725490868091583, 0.1568627506494522, 0.27450981736183167, 0.1764705926179886, 0.13725490868091583, 0.21568627655506134, 0.1568627506494522, 0.1568627506494522, 0.21568627655506134, 0.19607843458652496, 0.23529411852359772, 0.37254902720451355, 0.19607843458652496, 0.37254902720451355, 0.19607843458652496, 0.23529411852359772, 0.2549019753932953, 0.23529411852359772, 0.5098039507865906, 0.03921568766236305, 0.019607843831181526, 0.27450981736183167, 0.29411765933036804, 0.5098039507865906, 0.3921568691730499, 0.4117647111415863, 0.4117647111415863, 0.4313725531101227, 0.3333333432674408, NaN, 0.5098039507865906, 0.529411792755127, 0.5882353186607361, 0.5882353186607361, 0.6470588445663452, 0.6470588445663452, 0.8039215803146362], [0.05882352963089943, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.0784313753247261, 0.05882352963089943, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.05882352963089943, 0.05882352963089943, 0.09803921729326248, 0.05882352963089943, 0.0784313753247261, 0.03921568766236305, 0.09803921729326248, 0.11764705926179886, 0.10784313827753067, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.0784313753247261, 0.0784313753247261, 0.09803921729326248, 0.05882352963089943, 0.13725490868091583, 0.19607843458652496, 0.05882352963089943, 0.1568627506494522, 0.19607843458652496, 0.05882352963089943, 0.1764705926179886, 0.1568627506494522, 0.11764705926179886, 0.11764705926179886, 0.21568627655506134, 0.11764705926179886, 0.11764705926179886, 0.19607843458652496, 0.11764705926179886, 0.1764705926179886, 0.27450981736183167, 0.13725490868091583, 0.13725490868091583, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.21568627655506134, 0.1568627506494522, 0.21568627655506134, 0.3529411852359772, 0.1568627506494522, 0.3529411852359772, 0.23529411852359772, 0.1568627506494522, 0.3137255012989044, 0.23529411852359772, 0.3921568691730499, 0.4117647111415863, 0.4313725531101227, 0.3333333432674408, 0.3333333432674408, 0.45098039507865906, 0.47058823704719543, 0.4313725531101227, 0.3921568691730499, 0.45098039507865906, 0.47058823704719543, 0.4901960790157318, NaN, 0.5490196347236633, 0.686274528503418, 0.686274528503418, 0.9803921580314636, 0.7647058963775635, 0.9803921580314636], [0.0, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.0784313753247261, 0.019607843831181526, 0.03921568766236305, 0.05882352963089943, 0.0784313753247261, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.05882352963089943, 0.09803921729326248, 0.05882352963089943, 0.13725490868091583, 0.09803921729326248, 0.0784313753247261, 0.03921568766236305, 0.11764705926179886, 0.09803921729326248, 0.09803921729326248, 0.11764705926179886, 0.13725490868091583, 0.11764705926179886, 0.09803921729326248, 0.13725490868091583, 0.11764705926179886, 0.0784313753247261, 0.13725490868091583, 0.09803921729326248, 0.13725490868091583, 0.1568627506494522, 0.1568627506494522, 0.09803921729326248, 0.0784313753247261, 0.1764705926179886, 0.19607843458652496, 0.1764705926179886, 0.1764705926179886, 0.11764705926179886, 0.19607843458652496, 0.21568627655506134, 0.1764705926179886, 0.1568627506494522, 0.13725490868091583, 0.23529411852359772, 0.1568627506494522, 0.3137255012989044, 0.3137255012989044, 0.3137255012989044, 0.19607843458652496, 0.3333333432674408, 0.27450981736183167, 0.4117647111415863, 0.09803921729326248, 0.45098039507865906, 0.45098039507865906, 0.3333333432674408, 0.37254902720451355, 0.18627451360225677, 0.45098039507865906, 0.3921568691730499, 0.4117647111415863, 0.4313725531101227, 0.3333333432674408, 0.47058823704719543, 0.45098039507865906, NaN, 0.6078431606292725, 0.6078431606292725, 0.5490196347236633, 0.6470588445663452, 0.843137264251709], [0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.05882352963089943, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.03921568766236305, 0.11764705926179886, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.13725490868091583, 0.13725490868091583, 0.0784313753247261, 0.09803921729326248, 0.05882352963089943, 0.09803921729326248, 0.03921568766236305, 0.0784313753247261, 0.0784313753247261, 0.11764705926179886, 0.03921568766236305, 0.05882352963089943, 0.11764705926179886, 0.11764705926179886, 0.09803921729326248, 0.11764705926179886, 0.0784313753247261, 0.21568627655506134, 0.0784313753247261, 0.0784313753247261, 0.03921568766236305, 0.11764705926179886, 0.0784313753247261, 0.0784313753247261, 0.09803921729326248, 0.0784313753247261, 0.09803921729326248, 0.13725490868091583, 0.13725490868091583, 0.11764705926179886, 0.1764705926179886, 0.09803921729326248, 0.1568627506494522, 0.1568627506494522, 0.23529411852359772, 0.13725490868091583, 0.29411765933036804, 0.19607843458652496, 0.29411765933036804, 0.21568627655506134, 0.21568627655506134, 0.21568627655506134, 0.23529411852359772, 0.2549019753932953, 0.29411765933036804, 0.29411765933036804, 0.23529411852359772, 0.29411765933036804, 0.27450981736183167, 0.3921568691730499, 0.3333333432674408, 0.29411765933036804, 0.3333333432674408, 0.3137255012989044, 0.4117647111415863, 0.3137255012989044, 0.3921568691730499, NaN, 0.7647058963775635, 0.4901960790157318, 0.7647058963775635, 0.7254902124404907], [0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.029411764815449715, 0.11764705926179886, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.13725490868091583, 0.11764705926179886, 0.0784313753247261, 0.09803921729326248, 0.03921568766236305, 0.05882352963089943, 0.019607843831181526, 0.05882352963089943, 0.05882352963089943, 0.09803921729326248, 0.019607843831181526, 0.03921568766236305, 0.11764705926179886, 0.0784313753247261, 0.09803921729326248, 0.11764705926179886, 0.09803921729326248, 0.19607843458652496, 0.09803921729326248, 0.09803921729326248, 0.03921568766236305, 0.11764705926179886, 0.05882352963089943, 0.05882352963089943, 0.11764705926179886, 0.0784313753247261, 0.09803921729326248, 0.13725490868091583, 0.13725490868091583, 0.0784313753247261, 0.11764705926179886, 0.0784313753247261, 0.13725490868091583, 0.13725490868091583, 0.23529411852359772, 0.11764705926179886, 0.29411765933036804, 0.21568627655506134, 0.29411765933036804, 0.21568627655506134, 0.23529411852359772, 0.21568627655506134, 0.23529411852359772, 0.2549019753932953, 0.27450981736183167, 0.29411765933036804, 0.23529411852359772, 0.23529411852359772, 0.29411765933036804, 0.3921568691730499, 0.3333333432674408, 0.29411765933036804, 0.3333333432674408, 0.29411765933036804, 0.4117647111415863, 0.3137255012989044, 0.3921568691730499, 0.23529411852359772, NaN, 0.47058823704719543, 0.686274528503418, 0.7058823704719543], [0.019607843831181526, 0.0, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.0, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.0, 0.0, 0.019607843831181526, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.09803921729326248, 0.019607843831181526, 0.11764705926179886, 0.05882352963089943, 0.03921568766236305, 0.019607843831181526, 0.05882352963089943, 0.05882352963089943, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.13725490868091583, 0.019607843831181526, 0.05882352963089943, 0.11764705926179886, 0.019607843831181526, 0.09803921729326248, 0.019607843831181526, 0.019607843831181526, 0.0784313753247261, 0.13725490868091583, 0.05882352963089943, 0.03921568766236305, 0.0784313753247261, 0.019607843831181526, 0.03921568766236305, 0.1568627506494522, 0.05882352963089943, 0.0784313753247261, 0.05882352963089943, 0.0784313753247261, 0.05882352963089943, 0.09803921729326248, 0.019607843831181526, 0.13725490868091583, 0.21568627655506134, 0.05882352963089943, 0.21568627655506134, 0.05882352963089943, 0.0784313753247261, 0.1568627506494522, 0.1764705926179886, 0.2549019753932953, 0.21568627655506134, 0.21568627655506134, 0.19607843458652496, 0.1568627506494522, 0.3333333432674408, 0.27450981736183167, 0.23529411852359772, 0.21568627655506134, 0.23529411852359772, 0.2549019753932953, 0.3529411852359772, 0.019607843831181526, 0.45098039507865906, 0.5098039507865906, 0.529411792755127, NaN, 0.5490196347236633, 0.9215686321258545], [0.0, 0.019607843831181526, 0.0, 0.019607843831181526, 0.03921568766236305, 0.0, 0.03921568766236305, 0.0, 0.019607843831181526, 0.019607843831181526, 0.0784313753247261, 0.03921568766236305, 0.0784313753247261, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.11764705926179886, 0.03921568766236305, 0.0784313753247261, 0.03921568766236305, 0.05882352963089943, 0.019607843831181526, 0.05882352963089943, 0.03921568766236305, 0.09803921729326248, 0.03921568766236305, 0.019607843831181526, 0.11764705926179886, 0.0784313753247261, 0.03921568766236305, 0.09803921729326248, 0.0784313753247261, 0.13725490868091583, 0.0784313753247261, 0.019607843831181526, 0.019607843831181526, 0.11764705926179886, 0.019607843831181526, 0.019607843831181526, 0.09803921729326248, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.09803921729326248, 0.0784313753247261, 0.11764705926179886, 0.03921568766236305, 0.05882352963089943, 0.11764705926179886, 0.11764705926179886, 0.0784313753247261, 0.2549019753932953, 0.13725490868091583, 0.2549019753932953, 0.09803921729326248, 0.1568627506494522, 0.11764705926179886, 0.1568627506494522, 0.23529411852359772, 0.19607843458652496, 0.19607843458652496, 0.19607843458652496, 0.1568627506494522, 0.2549019753932953, 0.3529411852359772, 0.29411765933036804, 0.2549019753932953, 0.27450981736183167, 0.2549019753932953, 0.3529411852359772, 0.23529411852359772, 0.3529411852359772, 0.23529411852359772, 0.3137255012989044, 0.45098039507865906, NaN, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.019607843831181526, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.019607843831181526, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.019607843831181526, 0.03921568766236305, 0.0, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.0, 0.0, 0.0, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.03921568766236305, 0.0, 0.03921568766236305, 0.03921568766236305, 0.03921568766236305, 0.019607843831181526, 0.03921568766236305, 0.019607843831181526, 0.019607843831181526, 0.019607843831181526, 0.05882352963089943, 0.019607843831181526, 0.05882352963089943, 0.0784313753247261, 0.0784313753247261, 0.1568627506494522, 0.1568627506494522, 0.019607843831181526, 0.0784313753247261, 0.0784313753247261, 0.11764705926179886, 0.03921568766236305, 0.019607843831181526, 0.09803921729326248, 0.09803921729326248, 0.19607843458652496, 0.019607843831181526, 0.1568627506494522, 0.27450981736183167, 0.29411765933036804, 0.0784313753247261, 0.3333333432674408, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9367088645},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8736659288},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8700670209},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8515760806},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8447505664},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8155870031},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8061553818},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7895259463},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7877885426},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7572598767},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6962025453},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6886324269},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6878878265},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6738644956},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6691486833},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6636882722},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6629436696},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6615785665},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.651154147},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6391164176},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6365103131},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6269545935},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6253412863},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6103251536},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6078431479},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6057334448},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.602382737},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.600521234},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5929511151},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5913378131},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5820302922},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5599404437},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5569620369},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.5564656377},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5428146059},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5384710972},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5357408889},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.5353685882},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5167535475},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5122859373},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5073219264},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5006205124},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4962770035},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.4950359985},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4934226972},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.4754281575},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4692231419},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4609084248},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4489947967},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.436708871},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4359642681},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4345991672},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4156118236},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4146190211},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.4138744216},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4085381076},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.4029535974},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4009679913},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3659717125},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3459915699},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3232812194},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3230330181},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3169521058},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3148424008},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2918838487},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2824522296},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2792256217},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2727724076},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2519235612},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2477041503},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2302060109},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2283445075},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2266071055},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2207743896},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.197443539},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1702655787},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1554976457},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1206254687},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1196326666},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0394638881}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_all/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..99d0f42139f1483199776fbf2ab94b92e51a164b
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_all/winrate_matrix.csv
@@ -0,0 +1,81 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),TabICLv2 (default),RealTabPFN-2.5 (tuned),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (default),RealMLP (tuned + ensembled),TabDPT (tuned + ensembled),RealMLP (tuned),TabDPT-Turbo (default),TabM (tuned + ensembled),LightGBM (tuned + ensembled),CatBoost (tuned),iLTM (tuned + ensembled),CatBoost (tuned + ensembled),TabDPT (tuned),ModernNCA (tuned + ensembled),TabM (tuned),LimiX (default),XGBoost (tuned + ensembled),LightGBM (tuned),CatBoost (default),ModernNCA (tuned),XGBoost (tuned),TabSwift (default),xRFM (tuned + ensembled),ChimeraBoost (tuned + ensembled),TabPFNv2 (tuned + ensembled),iLTM (tuned),ChimeraBoost (tuned),Mitra (default),xRFM (tuned),TabDPT (default),TabM (default),TabICL (default),EBM (tuned + ensembled),TabPFNv2 (tuned),TorchMLP (tuned + ensembled),RealMLP (default),SAP-RPT-OSS (default),BetaTabPFN (default),ChimeraBoost (default),TabPFNv2 (default),ModernNCA (default),EBM (tuned),ExtraTrees (tuned + ensembled),EBM (default),TorchMLP (tuned),XGBoost (default),ExtraTrees (tuned),FastaiMLP (tuned + ensembled),Nori-30M (default),RandomForest (tuned + ensembled),Nori (default),LightGBM (default),RandomForest (tuned),FastaiMLP (tuned),iLTM (default),PerpetualBooster (tuned + ensembled),TabSTAR (tuned),TabSTAR (tuned + ensembled),OrionMSP (default),TorchMLP (default),PerpetualBooster (tuned),xRFM (default),ExtraTrees (default),RandomForest (default),TabFlex (default),FastaiMLP (default),TabSTAR (default),KNN (tuned + ensembled),PerpetualBooster (default),Linear (tuned + ensembled),Linear (tuned),KNN (tuned),Linear (default),KNN (default)
+TabFM (default),,0.7254902,0.8039216,0.84313726,0.8627451,0.88235295,0.88235295,0.8235294,0.8627451,0.8627451,0.9019608,0.88235295,0.9411765,0.9411765,0.9411765,0.92156863,0.8627451,0.9411765,0.9607843,0.88235295,0.9607843,0.9019608,0.92156863,0.9607843,0.9019608,0.92156863,0.88235295,0.88235295,0.9411765,0.92156863,0.9607843,0.9411765,0.9019608,1.0,0.9411765,0.98039216,0.9411765,0.9411765,0.9411765,0.98039216,0.9607843,0.92156863,0.92156863,0.9607843,0.9411765,0.9607843,0.9607843,0.9411765,0.9607843,0.9411765,0.98039216,0.9607843,0.9607843,0.92156863,0.9411765,0.9607843,0.9607843,0.9607843,0.9607843,0.9411765,0.9411765,1.0,0.9411765,0.92156863,0.98039216,0.98039216,1.0,0.9607843,0.98039216,0.98039216,1.0,1.0,0.9607843,0.9411765,1.0,0.98039216,0.98039216,0.98039216,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.27450982,,0.4509804,0.60784316,0.5686275,0.50980395,0.6666667,0.74509805,0.60784316,0.8235294,0.74509805,0.84313726,0.74509805,0.8627451,0.92156863,0.88235295,0.9019608,0.9019608,0.7647059,0.8627451,0.84313726,0.7647059,0.9411765,0.9411765,0.9411765,0.92156863,0.9411765,0.84313726,0.92156863,0.9411765,0.78431374,0.92156863,0.9411765,0.84313726,0.9019608,0.8235294,0.9019608,0.8627451,0.88235295,0.88235295,0.9411765,0.9019608,0.8627451,0.88235295,0.9607843,0.84313726,0.9607843,0.9019608,0.9411765,0.92156863,0.92156863,0.9411765,0.9411765,0.9019608,0.8039216,0.9607843,0.8627451,0.98039216,0.9607843,0.9411765,0.9607843,0.9411765,0.9019608,0.88235295,1.0,0.9607843,0.9607843,0.98039216,0.98039216,0.9607843,0.9607843,0.98039216,0.9019608,0.98039216,0.9607843,0.98039216,0.98039216,1.0,0.98039216,1.0
+TabPFN-3 (default),0.19607843,0.54901963,,0.5686275,0.5588235,0.627451,0.6862745,0.627451,0.64705884,0.74509805,0.7058824,0.7647059,0.84313726,0.84313726,0.88235295,0.7647059,0.84313726,0.8039216,0.8235294,0.9019608,0.9019608,0.7647059,0.8627451,0.9411765,0.8627451,0.84313726,0.84313726,0.88235295,0.8627451,0.88235295,0.8235294,0.92156863,0.8627451,0.88235295,0.92156863,0.92156863,0.88235295,0.92156863,0.8235294,0.92156863,0.9019608,0.9411765,0.84313726,0.9411765,0.9019608,0.84313726,0.9607843,0.88235295,0.9607843,0.9019608,0.9607843,0.9411765,0.9607843,0.9019608,0.88235295,0.9411765,0.9019608,0.9607843,0.98039216,0.9607843,0.98039216,0.9607843,0.9019608,0.9019608,1.0,0.9607843,1.0,0.9607843,0.98039216,0.98039216,0.9607843,1.0,0.9019608,0.9607843,0.98039216,0.9607843,0.98039216,0.98039216,1.0,1.0
+TabPFN-2.6 (default),0.15686275,0.39215687,0.43137255,,0.47058824,0.50980395,0.54901963,0.54901963,0.6862745,0.7254902,0.7058824,0.8039216,0.7647059,0.78431374,0.88235295,0.8039216,0.84313726,0.84313726,0.7254902,0.8627451,0.8235294,0.8039216,0.8235294,0.9411765,0.9019608,0.84313726,0.8627451,0.8235294,0.84313726,0.8627451,0.8039216,0.9019608,0.8627451,0.9019608,0.8627451,0.78431374,0.88235295,0.9019608,0.9019608,0.9019608,0.92156863,0.92156863,0.8627451,0.88235295,0.88235295,0.8235294,0.9607843,0.9411765,0.9411765,0.9019608,0.9607843,0.9607843,0.9411765,0.92156863,0.8235294,0.9607843,0.8627451,0.9607843,0.9411765,0.92156863,0.9607843,0.98039216,0.88235295,0.88235295,0.98039216,0.98039216,1.0,0.9607843,0.9607843,0.98039216,0.9411765,0.98039216,0.88235295,0.98039216,0.98039216,0.98039216,0.98039216,0.98039216,0.98039216,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.13725491,0.43137255,0.44117647,0.5294118,,0.54901963,0.7254902,0.60784316,0.7058824,0.74509805,0.6666667,0.7647059,0.7058824,0.7254902,0.84313726,0.8235294,0.78431374,0.8627451,0.7254902,0.7647059,0.8039216,0.74509805,0.8627451,0.88235295,0.9019608,0.8627451,0.8627451,0.8235294,0.84313726,0.84313726,0.8627451,0.84313726,0.88235295,0.88235295,0.88235295,0.78431374,0.88235295,0.8039216,0.88235295,0.9607843,0.8627451,0.88235295,0.88235295,0.9019608,0.8627451,0.9019608,0.88235295,0.88235295,0.9411765,0.9019608,0.9019608,0.9607843,0.9411765,0.88235295,0.84313726,0.92156863,0.88235295,0.98039216,0.9411765,0.9411765,0.92156863,0.9607843,0.88235295,0.88235295,0.98039216,0.9607843,0.98039216,0.9411765,0.98039216,0.98039216,0.92156863,0.98039216,0.9019608,0.9411765,0.9607843,0.9411765,0.9607843,0.9607843,0.9607843,1.0
+TabICLv2 (default),0.11764706,0.49019608,0.37254903,0.49019608,0.4509804,,0.627451,0.6666667,0.5882353,0.6666667,0.7254902,0.6862745,0.7647059,0.7254902,0.8235294,0.7647059,0.78431374,0.7647059,0.74509805,0.8235294,0.74509805,0.74509805,0.78431374,0.84313726,0.8039216,0.8039216,0.78431374,0.8235294,0.78431374,0.8235294,0.74509805,0.78431374,0.8039216,0.8039216,0.8235294,0.78431374,0.8627451,0.8235294,0.78431374,0.8039216,0.8627451,0.84313726,0.8235294,0.9019608,0.8235294,0.8627451,0.9019608,0.84313726,0.88235295,0.9019608,0.88235295,0.8627451,0.92156863,0.9411765,0.8039216,0.88235295,0.8039216,0.92156863,0.88235295,0.92156863,0.88235295,0.92156863,0.9019608,0.9019608,0.92156863,0.9607843,0.9411765,0.92156863,0.92156863,0.9411765,0.9607843,0.98039216,0.92156863,0.9607843,0.9607843,0.9607843,0.98039216,0.98039216,1.0,1.0
+RealTabPFN-2.5 (tuned),0.11764706,0.33333334,0.3137255,0.4509804,0.27450982,0.37254903,,0.54901963,0.61764705,0.64705884,0.627451,0.74509805,0.6862745,0.7647059,0.74509805,0.74509805,0.6862745,0.7647059,0.6862745,0.7058824,0.8235294,0.64705884,0.8039216,0.8627451,0.8627451,0.8235294,0.8235294,0.74509805,0.78431374,0.74509805,0.71568626,0.8627451,0.8235294,0.8627451,0.8627451,0.7647059,0.8627451,0.8235294,0.8627451,0.92156863,0.8627451,0.8627451,0.84313726,0.8627451,0.84313726,0.84313726,0.9019608,0.8627451,0.9607843,0.88235295,0.92156863,0.9411765,0.9019608,0.8627451,0.84313726,0.92156863,0.8627451,0.9607843,0.92156863,0.92156863,0.9019608,0.9607843,0.8627451,0.84313726,0.98039216,0.9607843,0.98039216,0.92156863,0.9607843,0.9607843,0.9019608,0.9607843,0.9019608,0.92156863,0.9607843,0.9411765,0.9411765,0.9607843,0.9607843,1.0
+"AutoGluon 1.4 (best, 4h)",0.1764706,0.25490198,0.37254903,0.4509804,0.39215687,0.33333334,0.4509804,,0.4509804,0.54901963,0.5294118,0.7058824,0.60784316,0.6666667,0.8235294,0.7647059,0.78431374,0.8039216,0.5882353,0.64705884,0.74509805,0.64705884,0.8627451,0.8627451,0.8235294,0.7058824,0.8235294,0.7058824,0.8039216,0.84313726,0.5882353,0.8235294,0.8627451,0.6862745,0.8235294,0.74509805,0.8235294,0.7647059,0.8039216,0.6862745,0.92156863,0.8235294,0.7647059,0.8235294,0.88235295,0.7647059,0.92156863,0.9019608,0.8627451,0.9019608,0.9411765,0.92156863,0.8627451,0.92156863,0.7647059,0.9019608,0.8235294,0.9411765,0.9607843,0.9411765,0.9411765,0.9607843,0.8235294,0.8235294,0.9411765,0.98039216,0.98039216,0.92156863,0.98039216,0.98039216,0.9607843,0.98039216,0.88235295,0.9411765,0.98039216,0.98039216,0.98039216,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.13725491,0.39215687,0.3529412,0.3137255,0.29411766,0.4117647,0.38235295,0.54901963,,0.5686275,0.64705884,0.6666667,0.64705884,0.7058824,0.74509805,0.7058824,0.627451,0.7058824,0.6666667,0.64705884,0.74509805,0.7058824,0.74509805,0.7647059,0.74509805,0.7058824,0.78431374,0.84313726,0.8039216,0.74509805,0.64705884,0.85294116,0.74509805,0.8039216,0.84313726,0.7647059,0.84313726,0.74509805,0.8039216,0.8627451,0.84313726,0.8235294,0.84313726,0.8627451,0.84313726,0.92156863,0.88235295,0.88235295,0.8627451,0.88235295,0.88235295,0.88235295,0.8627451,0.92156863,0.78431374,0.8627451,0.8235294,0.9607843,0.92156863,0.98039216,0.92156863,0.9411765,0.9019608,0.9019608,0.98039216,0.98039216,0.98039216,0.92156863,0.98039216,0.9411765,0.98039216,0.9607843,0.92156863,0.92156863,0.9411765,0.9607843,0.9607843,0.98039216,0.98039216,0.98039216
+RealMLP (tuned + ensembled),0.13725491,0.1764706,0.25490198,0.27450982,0.25490198,0.33333334,0.3529412,0.4509804,0.43137255,,0.50980395,0.6862745,0.5882353,0.627451,0.7058824,0.7254902,0.74509805,0.78431374,0.54901963,0.64705884,0.7058824,0.5686275,0.7647059,0.7647059,0.78431374,0.6862745,0.8039216,0.6666667,0.74509805,0.7058824,0.5882353,0.8235294,0.7647059,0.6666667,0.8039216,0.7058824,0.8235294,0.7058824,0.8039216,0.7254902,0.84313726,0.92156863,0.8039216,0.78431374,0.8039216,0.7058824,0.9019608,0.84313726,0.8627451,0.8627451,0.84313726,0.9019608,0.8627451,0.92156863,0.7647059,0.84313726,0.78431374,0.9411765,0.9019608,0.9019608,0.92156863,0.9607843,0.84313726,0.84313726,0.9607843,0.9411765,0.98039216,0.9411765,0.98039216,0.9607843,0.92156863,0.98039216,0.88235295,0.9607843,0.98039216,0.98039216,0.98039216,0.98039216,0.98039216,1.0
+TabDPT (tuned + ensembled),0.09803922,0.25490198,0.29411766,0.29411766,0.33333334,0.27450982,0.37254903,0.47058824,0.3529412,0.49019608,,0.5686275,0.49019608,0.5294118,0.5882353,0.54901963,0.64705884,0.5882353,0.74509805,0.5294118,0.54901963,0.627451,0.627451,0.60784316,0.60784316,0.5686275,0.627451,0.627451,0.6666667,0.627451,0.627451,0.74509805,0.627451,0.6862745,0.7254902,0.8039216,0.6862745,0.5882353,0.64705884,0.74509805,0.74509805,0.7058824,0.7254902,0.6666667,0.64705884,0.7058824,0.8235294,0.7254902,0.78431374,0.78431374,0.7647059,0.74509805,0.8627451,0.74509805,0.8235294,0.78431374,0.8235294,0.8235294,0.8039216,0.84313726,0.8039216,0.84313726,0.8235294,0.78431374,0.9607843,0.8627451,0.9019608,0.9411765,0.98039216,0.9607843,0.9411765,0.9607843,0.88235295,0.92156863,0.92156863,0.9019608,0.9019608,0.9607843,0.92156863,1.0
+RealMLP (tuned),0.11764706,0.15686275,0.23529412,0.19607843,0.23529412,0.3137255,0.25490198,0.29411766,0.33333334,0.3137255,0.43137255,,0.4117647,0.50980395,0.5882353,0.54901963,0.5294118,0.54901963,0.50980395,0.5294118,0.60784316,0.5,0.6666667,0.6666667,0.627451,0.6862745,0.7058824,0.5686275,0.64705884,0.5686275,0.5686275,0.7254902,0.627451,0.5882353,0.7058824,0.627451,0.7647059,0.6862745,0.7647059,0.627451,0.8039216,0.8235294,0.6862745,0.7647059,0.6862745,0.6862745,0.7647059,0.78431374,0.7647059,0.8039216,0.8627451,0.84313726,0.8235294,0.84313726,0.74509805,0.84313726,0.7254902,0.84313726,0.88235295,0.88235295,0.9019608,0.9019608,0.84313726,0.8627451,0.872549,0.9607843,0.92156863,0.92156863,0.9019608,0.9411765,0.9019608,0.9607843,0.8627451,0.9411765,0.98039216,0.9607843,0.9705882,0.98039216,0.9607843,1.0
+TabDPT-Turbo (default),0.05882353,0.25490198,0.15686275,0.23529412,0.29411766,0.23529412,0.3137255,0.39215687,0.3529412,0.4117647,0.50980395,0.5882353,,0.47058824,0.54901963,0.54901963,0.60784316,0.5882353,0.54901963,0.50980395,0.5294118,0.60784316,0.6666667,0.5882353,0.64705884,0.5882353,0.6862745,0.6960784,0.6666667,0.6666667,0.627451,0.74509805,0.6666667,0.6666667,0.74509805,0.7254902,0.7058824,0.627451,0.6666667,0.7058824,0.74509805,0.7058824,0.7058824,0.64705884,0.6862745,0.6666667,0.7254902,0.7058824,0.8235294,0.7647059,0.8235294,0.78431374,0.84313726,0.78431374,0.7058824,0.8235294,0.7647059,0.8627451,0.8627451,0.9019608,0.84313726,0.84313726,0.8039216,0.78431374,0.9411765,0.88235295,0.9019608,0.9019608,0.9607843,0.9411765,0.9607843,0.9607843,0.8627451,0.9411765,0.9607843,0.88235295,0.88235295,0.98039216,0.92156863,1.0
+TabM (tuned + ensembled),0.05882353,0.13725491,0.15686275,0.21568628,0.27450982,0.27450982,0.23529412,0.33333334,0.29411766,0.37254903,0.47058824,0.49019608,0.5294118,,0.5882353,0.49019608,0.5294118,0.54901963,0.50980395,0.54901963,0.627451,0.49019608,0.5882353,0.627451,0.64705884,0.60784316,0.6862745,0.5882353,0.5686275,0.60784316,0.5294118,0.64705884,0.7254902,0.50980395,0.5686275,0.64705884,0.88235295,0.7058824,0.78431374,0.5686275,0.74509805,0.8235294,0.7058824,0.7647059,0.74509805,0.64705884,0.74509805,0.7647059,0.74509805,0.8039216,0.8039216,0.78431374,0.74509805,0.8235294,0.627451,0.8039216,0.6666667,0.8627451,0.8039216,0.88235295,0.84313726,0.84313726,0.8627451,0.8627451,0.9019608,0.98039216,0.9019608,0.84313726,0.8627451,0.88235295,0.9411765,0.9411765,0.88235295,0.9019608,0.9411765,0.9607843,0.9607843,1.0,0.9607843,1.0
+LightGBM (tuned + ensembled),0.05882353,0.078431375,0.11764706,0.11764706,0.15686275,0.1764706,0.25490198,0.1764706,0.25490198,0.29411766,0.4117647,0.4117647,0.4509804,0.4117647,,0.54901963,0.50980395,0.5686275,0.4509804,0.47058824,0.49019608,0.4509804,0.5686275,0.74509805,0.7254902,0.54901963,0.6666667,0.54901963,0.5882353,0.64705884,0.50980395,0.6666667,0.74509805,0.5294118,0.6862745,0.60784316,0.7254902,0.64705884,0.74509805,0.5686275,0.7647059,0.7254902,0.6666667,0.6666667,0.8039216,0.6666667,0.8039216,0.8235294,0.84313726,0.8627451,0.8039216,0.9019608,0.8039216,0.8235294,0.74509805,0.8627451,0.7647059,0.9607843,0.9411765,0.8627451,0.92156863,0.9019608,0.84313726,0.8235294,0.88235295,0.92156863,0.92156863,0.84313726,0.92156863,0.98039216,0.9019608,0.92156863,0.8627451,0.9411765,0.92156863,0.98039216,0.98039216,1.0,0.9607843,1.0
+CatBoost (tuned),0.078431375,0.11764706,0.23529412,0.19607843,0.1764706,0.23529412,0.25490198,0.23529412,0.29411766,0.27450982,0.4509804,0.4509804,0.4509804,0.50980395,0.4509804,,0.49019608,0.4117647,0.47058824,0.54901963,0.5294118,0.5294118,0.60784316,0.54901963,0.6862745,0.5686275,0.6666667,0.5294118,0.5882353,0.5686275,0.5294118,0.74509805,0.64705884,0.5882353,0.6862745,0.5882353,0.6862745,0.64705884,0.7254902,0.60784316,0.7647059,0.7647059,0.7058824,0.6862745,0.78431374,0.64705884,0.7254902,0.7647059,0.78431374,0.8235294,0.8235294,0.8627451,0.7647059,0.78431374,0.7058824,0.84313726,0.7058824,0.84313726,0.9019608,0.84313726,0.88235295,0.88235295,0.8235294,0.8235294,0.8627451,0.9411765,0.9411765,0.84313726,0.88235295,0.9411765,0.9019608,0.92156863,0.84313726,0.92156863,0.9607843,0.98039216,0.98039216,0.98039216,0.98039216,1.0
+iLTM (tuned + ensembled),0.13725491,0.09803922,0.15686275,0.15686275,0.21568628,0.21568628,0.3137255,0.21568628,0.37254903,0.25490198,0.3529412,0.47058824,0.39215687,0.47058824,0.49019608,0.50980395,,0.47058824,0.4117647,0.47058824,0.5294118,0.48039216,0.5882353,0.5686275,0.5294118,0.60784316,0.6666667,0.5294118,0.5686275,0.64705884,0.54901963,0.7058824,0.6666667,0.5294118,0.6666667,0.5882353,0.64705884,0.5980392,0.6666667,0.64705884,0.7647059,0.7254902,0.7254902,0.627451,0.74509805,0.7254902,0.7647059,0.6862745,0.84313726,0.7647059,0.84313726,0.8235294,0.8235294,0.8627451,0.7058824,0.9019608,0.7058824,0.8627451,0.9019608,0.8627451,0.9019608,0.9019608,0.8039216,0.8039216,0.88235295,0.9607843,0.9411765,0.92156863,0.9607843,0.9607843,0.9411765,0.9607843,0.8627451,0.9607843,0.9607843,0.9411765,0.9607843,0.9607843,0.9607843,1.0
+CatBoost (tuned + ensembled),0.05882353,0.09803922,0.19607843,0.15686275,0.13725491,0.23529412,0.23529412,0.19607843,0.29411766,0.21568628,0.4117647,0.4509804,0.4117647,0.4509804,0.43137255,0.5882353,0.5294118,,0.4509804,0.54901963,0.47058824,0.49019608,0.5882353,0.5686275,0.7058824,0.5588235,0.64705884,0.50980395,0.54901963,0.5882353,0.5294118,0.74509805,0.6862745,0.5686275,0.6862745,0.5686275,0.627451,0.6862745,0.6862745,0.627451,0.74509805,0.7647059,0.74509805,0.64705884,0.8627451,0.627451,0.7058824,0.74509805,0.8235294,0.8235294,0.8235294,0.8627451,0.8235294,0.78431374,0.7254902,0.88235295,0.7254902,0.84313726,0.92156863,0.88235295,0.88235295,0.92156863,0.8235294,0.8235294,0.84313726,0.9411765,0.92156863,0.88235295,0.9019608,0.9607843,0.88235295,0.92156863,0.84313726,0.9019608,0.9607843,0.98039216,0.98039216,0.9607843,0.98039216,1.0
+TabDPT (tuned),0.039215688,0.23529412,0.1764706,0.27450982,0.27450982,0.25490198,0.3137255,0.4117647,0.33333334,0.4509804,0.25490198,0.49019608,0.4509804,0.49019608,0.54901963,0.5294118,0.5882353,0.54901963,,0.50980395,0.50980395,0.54901963,0.60784316,0.5882353,0.5882353,0.50980395,0.627451,0.627451,0.627451,0.627451,0.5686275,0.7058824,0.64705884,0.64705884,0.6666667,0.8137255,0.627451,0.5686275,0.627451,0.7058824,0.64705884,0.64705884,0.64705884,0.5882353,0.6862745,0.64705884,0.7647059,0.64705884,0.7254902,0.7254902,0.74509805,0.6862745,0.8235294,0.6666667,0.6666667,0.6862745,0.7058824,0.7254902,0.78431374,0.8235294,0.8235294,0.84313726,0.8039216,0.78431374,0.92156863,0.8627451,0.9019608,0.9019608,0.9411765,0.88235295,0.9019608,0.9607843,0.8627451,0.88235295,0.9411765,0.8627451,0.8627451,0.9607843,0.9019608,0.98039216
+ModernNCA (tuned + ensembled),0.11764706,0.13725491,0.09803922,0.13725491,0.23529412,0.1764706,0.29411766,0.3529412,0.3529412,0.3529412,0.47058824,0.47058824,0.49019608,0.4509804,0.5294118,0.4509804,0.5294118,0.4509804,0.49019608,,0.49019608,0.47058824,0.5882353,0.6666667,0.5882353,0.6764706,0.5882353,0.5686275,0.5882353,0.60784316,0.47058824,0.60784316,0.627451,0.47058824,0.6666667,0.60784316,0.60784316,0.6666667,0.6666667,0.54901963,0.64705884,0.7254902,0.627451,0.6862745,0.6666667,0.64705884,0.8039216,0.6666667,0.7647059,0.7254902,0.78431374,0.78431374,0.7647059,0.8039216,0.7254902,0.78431374,0.74509805,0.84313726,0.7647059,0.8627451,0.8235294,0.84313726,0.74509805,0.7647059,0.8627451,0.84313726,0.84313726,0.9019608,0.8627451,0.9019608,0.8627451,0.88235295,0.8627451,0.89215684,0.9019608,0.8627451,0.88235295,0.9019608,0.88235295,0.98039216
+TabM (tuned),0.039215688,0.15686275,0.09803922,0.1764706,0.19607843,0.25490198,0.1764706,0.25490198,0.25490198,0.29411766,0.4509804,0.39215687,0.47058824,0.37254903,0.50980395,0.47058824,0.47058824,0.5294118,0.49019608,0.50980395,,0.4509804,0.54901963,0.54901963,0.627451,0.5686275,0.60784316,0.54901963,0.50980395,0.6764706,0.50980395,0.627451,0.6666667,0.49019608,0.54901963,0.60784316,0.7647059,0.7058824,0.6862745,0.54901963,0.6666667,0.7058824,0.64705884,0.7254902,0.6666667,0.60784316,0.7254902,0.7254902,0.7254902,0.7254902,0.8039216,0.78431374,0.7254902,0.74509805,0.64705884,0.74509805,0.6666667,0.8235294,0.78431374,0.8627451,0.78431374,0.88235295,0.8235294,0.8235294,0.88235295,0.98039216,0.9019608,0.8235294,0.84313726,0.9019608,0.9411765,0.9019608,0.8627451,0.9019608,0.9411765,0.92156863,0.92156863,0.98039216,0.9607843,0.98039216
+LimiX (default),0.09803922,0.23529412,0.23529412,0.19607843,0.25490198,0.25490198,0.3529412,0.3529412,0.29411766,0.43137255,0.37254903,0.5,0.39215687,0.50980395,0.54901963,0.47058824,0.51960784,0.50980395,0.4509804,0.5294118,0.54901963,,0.5686275,0.5686275,0.5686275,0.5686275,0.5882353,0.49019608,0.5294118,0.60784316,0.60784316,0.60784316,0.64705884,0.5882353,0.64705884,0.54901963,0.60784316,0.5882353,0.6862745,0.6862745,0.6666667,0.7058824,0.627451,0.64705884,0.6666667,0.64705884,0.6862745,0.7254902,0.6666667,0.74509805,0.6666667,0.7058824,0.6666667,0.6862745,0.7254902,0.7058824,0.7254902,0.6862745,0.74509805,0.78431374,0.7647059,0.74509805,0.7745098,0.7647059,0.8137255,0.8039216,0.7647059,0.8235294,0.88235295,0.88235295,0.8235294,0.8627451,0.84313726,0.88235295,0.8627451,0.9019608,0.9019608,0.88235295,0.92156863,0.98039216
+XGBoost (tuned + ensembled),0.078431375,0.05882353,0.13725491,0.1764706,0.13725491,0.21568628,0.19607843,0.13725491,0.25490198,0.23529412,0.37254903,0.33333334,0.33333334,0.4117647,0.43137255,0.39215687,0.4117647,0.4117647,0.39215687,0.4117647,0.4509804,0.43137255,,0.50980395,0.5294118,0.49019608,0.627451,0.47058824,0.5294118,0.5294118,0.43137255,0.627451,0.60784316,0.5294118,0.627451,0.5686275,0.64705884,0.64705884,0.6862745,0.54901963,0.7254902,0.7254902,0.60784316,0.60784316,0.74509805,0.64705884,0.7058824,0.74509805,0.8235294,0.78431374,0.7352941,0.88235295,0.8039216,0.78431374,0.7058824,0.8627451,0.7254902,0.98039216,0.9019608,0.8235294,0.88235295,0.84313726,0.84313726,0.84313726,0.88235295,0.92156863,0.88235295,0.8235294,0.92156863,0.9607843,0.88235295,0.9019608,0.8627451,0.9019608,0.9019608,0.9411765,0.9607843,0.9411765,0.9607843,1.0
+LightGBM (tuned),0.039215688,0.05882353,0.05882353,0.05882353,0.11764706,0.15686275,0.13725491,0.13725491,0.23529412,0.23529412,0.39215687,0.33333334,0.4117647,0.37254903,0.25490198,0.4509804,0.43137255,0.43137255,0.4117647,0.33333334,0.4509804,0.43137255,0.49019608,,0.54901963,0.50980395,0.5294118,0.47058824,0.54901963,0.54901963,0.49019608,0.5686275,0.64705884,0.4509804,0.60784316,0.5686275,0.7254902,0.60784316,0.7254902,0.54901963,0.6666667,0.6862745,0.627451,0.6862745,0.7058824,0.60784316,0.64705884,0.74509805,0.7254902,0.78431374,0.7647059,0.8235294,0.7647059,0.78431374,0.7058824,0.78431374,0.7254902,0.8627451,0.88235295,0.84313726,0.8627451,0.88235295,0.8039216,0.8235294,0.8627451,0.88235295,0.9019608,0.84313726,0.84313726,0.9411765,0.88235295,0.9019608,0.8627451,0.9019608,0.92156863,0.9019608,0.9411765,0.9607843,0.9411765,1.0
+CatBoost (default),0.09803922,0.05882353,0.13725491,0.09803922,0.09803922,0.19607843,0.13725491,0.1764706,0.25490198,0.21568628,0.39215687,0.37254903,0.3529412,0.3529412,0.27450982,0.3137255,0.47058824,0.29411766,0.4117647,0.4117647,0.37254903,0.43137255,0.47058824,0.4509804,,0.4509804,0.49019608,0.50980395,0.5686275,0.47058824,0.50980395,0.64705884,0.60784316,0.47058824,0.627451,0.5882353,0.5882353,0.5882353,0.64705884,0.5882353,0.7058824,0.74509805,0.627451,0.54901963,0.78431374,0.627451,0.627451,0.6862745,0.7254902,0.7647059,0.7647059,0.7647059,0.7254902,0.8039216,0.7058824,0.8039216,0.7058824,0.78431374,0.84313726,0.84313726,0.8627451,0.9019608,0.84313726,0.8235294,0.84313726,0.9019608,0.9411765,0.8627451,0.8627451,0.92156863,0.88235295,0.9411765,0.8627451,0.92156863,0.9607843,0.9607843,0.98039216,0.98039216,0.98039216,1.0
+ModernNCA (tuned),0.078431375,0.078431375,0.15686275,0.15686275,0.13725491,0.19607843,0.1764706,0.29411766,0.29411766,0.3137255,0.43137255,0.3137255,0.4117647,0.39215687,0.4509804,0.43137255,0.39215687,0.44117647,0.49019608,0.32352942,0.43137255,0.43137255,0.50980395,0.49019608,0.54901963,,0.49019608,0.50980395,0.5294118,0.50980395,0.50980395,0.5686275,0.54901963,0.49019608,0.627451,0.5882353,0.60784316,0.5686275,0.5686275,0.54901963,0.6666667,0.6666667,0.6666667,0.64705884,0.64705884,0.6666667,0.7647059,0.6862745,0.6666667,0.7058824,0.7647059,0.74509805,0.6862745,0.71568626,0.7254902,0.7254902,0.74509805,0.78431374,0.78431374,0.8235294,0.8235294,0.8235294,0.78431374,0.78431374,0.84313726,0.8627451,0.84313726,0.84313726,0.8235294,0.9019608,0.8627451,0.9019608,0.88235295,0.92156863,0.88235295,0.92156863,0.9411765,0.9411765,0.9411765,1.0
+XGBoost (tuned),0.11764706,0.05882353,0.15686275,0.13725491,0.13725491,0.21568628,0.1764706,0.1764706,0.21568628,0.19607843,0.37254903,0.29411766,0.3137255,0.3137255,0.33333334,0.33333334,0.33333334,0.3529412,0.37254903,0.4117647,0.39215687,0.4117647,0.37254903,0.47058824,0.50980395,0.50980395,,0.4509804,0.5686275,0.5294118,0.47058824,0.5882353,0.54901963,0.50980395,0.5882353,0.5294118,0.5882353,0.60784316,0.627451,0.54901963,0.64705884,0.6666667,0.60784316,0.627451,0.6862745,0.64705884,0.6862745,0.74509805,0.8235294,0.7647059,0.7254902,0.8627451,0.8039216,0.7647059,0.6862745,0.8235294,0.7058824,0.9019608,0.8627451,0.84313726,0.8627451,0.8039216,0.84313726,0.84313726,0.8627451,0.9019608,0.88235295,0.78431374,0.9019608,0.9411765,0.88235295,0.9019608,0.8627451,0.92156863,0.9019608,0.92156863,0.9411765,0.9411765,0.9607843,1.0
+TabSwift (default),0.11764706,0.15686275,0.11764706,0.1764706,0.1764706,0.1764706,0.25490198,0.29411766,0.15686275,0.33333334,0.37254903,0.43137255,0.30392158,0.4117647,0.4509804,0.47058824,0.47058824,0.49019608,0.37254903,0.43137255,0.4509804,0.50980395,0.5294118,0.5294118,0.49019608,0.49019608,0.54901963,,0.5686275,0.49019608,0.49019608,0.5686275,0.5294118,0.5294118,0.627451,0.5686275,0.5882353,0.50980395,0.64705884,0.5686275,0.5686275,0.64705884,0.7058824,0.60784316,0.7058824,0.5882353,0.6666667,0.64705884,0.6666667,0.7254902,0.6666667,0.6666667,0.7058824,0.8235294,0.64705884,0.6862745,0.627451,0.7647059,0.7058824,0.8039216,0.78431374,0.8039216,0.78431374,0.78431374,0.88235295,0.8235294,0.84313726,0.8235294,0.88235295,0.8235294,0.9411765,0.9019608,0.8627451,0.9019608,0.9019608,0.88235295,0.9019608,0.98039216,0.9019608,1.0
+xRFM (tuned + ensembled),0.05882353,0.078431375,0.13725491,0.15686275,0.15686275,0.21568628,0.21568628,0.19607843,0.19607843,0.25490198,0.33333334,0.3529412,0.33333334,0.43137255,0.4117647,0.4117647,0.43137255,0.4509804,0.37254903,0.4117647,0.49019608,0.47058824,0.47058824,0.4509804,0.43137255,0.47058824,0.43137255,0.43137255,,0.4509804,0.47058824,0.5294118,0.5294118,0.50980395,0.6862745,0.47058824,0.60784316,0.5882353,0.5882353,0.54901963,0.60784316,0.64705884,0.5882353,0.60784316,0.5294118,0.60784316,0.74509805,0.627451,0.74509805,0.64705884,0.6666667,0.6862745,0.78431374,0.7254902,0.6862745,0.8039216,0.6862745,0.74509805,0.8235294,0.78431374,0.7647059,0.8039216,0.7647059,0.7647059,0.9019608,0.88235295,0.84313726,0.92156863,0.9411765,0.92156863,0.92156863,0.92156863,0.8235294,0.9411765,0.88235295,0.9607843,0.98039216,0.98039216,0.9607843,0.98039216
+ChimeraBoost (tuned + ensembled),0.078431375,0.05882353,0.11764706,0.13725491,0.15686275,0.1764706,0.25490198,0.15686275,0.25490198,0.29411766,0.37254903,0.43137255,0.33333334,0.39215687,0.3529412,0.43137255,0.3529412,0.4117647,0.37254903,0.39215687,0.32352942,0.39215687,0.47058824,0.4509804,0.5294118,0.49019608,0.47058824,0.50980395,0.54901963,,0.47058824,0.5882353,0.60784316,0.47058824,0.64705884,0.54901963,0.60784316,0.627451,0.6862745,0.50980395,0.627451,0.7058824,0.60784316,0.5294118,0.84313726,0.60784316,0.60784316,0.7254902,0.6666667,0.7254902,0.7058824,0.7254902,0.6862745,0.7647059,0.64705884,0.7647059,0.64705884,0.74509805,0.8039216,0.74509805,0.78431374,0.8235294,0.8235294,0.8039216,0.8235294,0.8235294,0.8627451,0.8039216,0.78431374,0.8627451,0.88235295,0.8627451,0.84313726,0.8627451,0.8627451,0.9411765,0.9607843,0.9607843,0.98039216,1.0
+TabPFNv2 (tuned + ensembled),0.039215688,0.21568628,0.1764706,0.19607843,0.13725491,0.25490198,0.28431374,0.4117647,0.3529412,0.4117647,0.37254903,0.43137255,0.37254903,0.47058824,0.49019608,0.47058824,0.4509804,0.47058824,0.43137255,0.5294118,0.49019608,0.39215687,0.5686275,0.50980395,0.49019608,0.49019608,0.5294118,0.50980395,0.5294118,0.5294118,,0.49019608,0.5686275,0.57843137,0.5882353,0.49019608,0.54901963,0.54901963,0.64705884,0.6862745,0.5686275,0.627451,0.5882353,0.65686274,0.60784316,0.65686274,0.54901963,0.6666667,0.54901963,0.6862745,0.6862745,0.64705884,0.60784316,0.6666667,0.627451,0.64705884,0.64705884,0.6862745,0.6764706,0.7254902,0.74509805,0.6862745,0.7058824,0.6862745,0.7745098,0.7254902,0.74509805,0.7647059,0.84313726,0.7941176,0.7745098,0.84313726,0.7647059,0.8039216,0.88235295,0.88235295,0.88235295,0.8627451,0.88235295,1.0
+iLTM (tuned),0.05882353,0.078431375,0.078431375,0.09803922,0.15686275,0.21568628,0.13725491,0.1764706,0.14705883,0.1764706,0.25490198,0.27450982,0.25490198,0.3529412,0.33333334,0.25490198,0.29411766,0.25490198,0.29411766,0.39215687,0.37254903,0.39215687,0.37254903,0.43137255,0.3529412,0.43137255,0.4117647,0.43137255,0.47058824,0.4117647,0.50980395,,0.5294118,0.50980395,0.5294118,0.49019608,0.5294118,0.5294118,0.54901963,0.54901963,0.5882353,0.60784316,0.64705884,0.60784316,0.5294118,0.6862745,0.627451,0.64705884,0.6862745,0.74509805,0.64705884,0.60784316,0.74509805,0.78431374,0.64705884,0.8039216,0.64705884,0.74509805,0.8039216,0.8039216,0.7745098,0.7647059,0.78431374,0.78431374,0.8039216,0.92156863,0.84313726,0.88235295,0.8627451,0.9019608,0.9019608,0.9019608,0.8627451,0.9411765,0.9019608,0.88235295,0.92156863,0.98039216,0.92156863,1.0
+ChimeraBoost (tuned),0.09803922,0.05882353,0.13725491,0.13725491,0.11764706,0.19607843,0.1764706,0.13725491,0.25490198,0.23529412,0.37254903,0.37254903,0.33333334,0.27450982,0.25490198,0.3529412,0.33333334,0.3137255,0.3529412,0.37254903,0.33333334,0.3529412,0.39215687,0.3529412,0.39215687,0.4509804,0.4509804,0.47058824,0.47058824,0.39215687,0.43137255,0.47058824,,0.4117647,0.5882353,0.50980395,0.5294118,0.5686275,0.627451,0.49019608,0.5882353,0.7058824,0.5882353,0.5686275,0.7254902,0.60784316,0.5686275,0.6666667,0.627451,0.74509805,0.627451,0.74509805,0.6666667,0.7254902,0.60784316,0.7254902,0.60784316,0.7647059,0.8039216,0.7647059,0.78431374,0.8039216,0.78431374,0.78431374,0.8039216,0.8235294,0.8235294,0.8039216,0.8039216,0.8039216,0.88235295,0.84313726,0.84313726,0.84313726,0.8627451,0.9019608,0.9019608,0.9411765,0.9607843,1.0
+Mitra (default),0.0,0.15686275,0.11764706,0.09803922,0.11764706,0.19607843,0.13725491,0.3137255,0.19607843,0.33333334,0.3137255,0.4117647,0.33333334,0.49019608,0.47058824,0.4117647,0.47058824,0.43137255,0.3529412,0.5294118,0.50980395,0.4117647,0.47058824,0.54901963,0.5294118,0.50980395,0.49019608,0.47058824,0.49019608,0.5294118,0.42156863,0.49019608,0.5882353,,0.5294118,0.50980395,0.627451,0.54901963,0.627451,0.5588235,0.54901963,0.5882353,0.5294118,0.60784316,0.5882353,0.5882353,0.60784316,0.64705884,0.5294118,0.6666667,0.7058824,0.60784316,0.5882353,0.6666667,0.5882353,0.5686275,0.5882353,0.627451,0.6372549,0.6862745,0.7254902,0.6862745,0.7254902,0.7254902,0.7745098,0.7058824,0.7058824,0.8039216,0.78431374,0.78431374,0.78431374,0.8235294,0.78431374,0.8039216,0.88235295,0.88235295,0.88235295,0.88235295,0.9019608,1.0
+xRFM (tuned),0.05882353,0.09803922,0.078431375,0.13725491,0.11764706,0.1764706,0.13725491,0.1764706,0.15686275,0.19607843,0.27450982,0.29411766,0.25490198,0.43137255,0.3137255,0.3137255,0.33333334,0.3137255,0.33333334,0.33333334,0.4509804,0.3529412,0.37254903,0.39215687,0.37254903,0.37254903,0.4117647,0.37254903,0.3137255,0.3529412,0.4117647,0.47058824,0.4117647,0.47058824,,0.43137255,0.50980395,0.5294118,0.50980395,0.50980395,0.54901963,0.5882353,0.60784316,0.5882353,0.50980395,0.54901963,0.627451,0.5882353,0.7254902,0.5686275,0.6666667,0.60784316,0.74509805,0.6862745,0.627451,0.74509805,0.627451,0.6862745,0.78431374,0.74509805,0.7647059,0.74509805,0.74509805,0.7647059,0.84313726,0.9019608,0.8235294,0.9607843,0.92156863,0.8627451,0.88235295,0.88235295,0.8235294,0.9411765,0.92156863,0.92156863,0.9019608,0.98039216,0.92156863,1.0
+TabDPT (default),0.019607844,0.1764706,0.078431375,0.21568628,0.21568628,0.21568628,0.23529412,0.25490198,0.23529412,0.29411766,0.19607843,0.37254903,0.27450982,0.3529412,0.39215687,0.4117647,0.4117647,0.43137255,0.18627451,0.39215687,0.39215687,0.4509804,0.43137255,0.43137255,0.4117647,0.4117647,0.47058824,0.43137255,0.5294118,0.4509804,0.50980395,0.50980395,0.49019608,0.49019608,0.5686275,,0.50980395,0.43137255,0.4509804,0.54901963,0.50980395,0.49019608,0.5294118,0.50980395,0.5686275,0.54901963,0.64705884,0.50980395,0.60784316,0.5686275,0.5686275,0.5882353,0.6666667,0.64705884,0.5882353,0.6666667,0.60784316,0.6862745,0.7058824,0.78431374,0.6862745,0.74509805,0.7058824,0.7254902,0.8235294,0.78431374,0.8235294,0.8039216,0.84313726,0.78431374,0.88235295,0.8627451,0.7647059,0.8235294,0.8627451,0.78431374,0.8039216,0.9019608,0.8627451,0.98039216
+TabM (default),0.05882353,0.09803922,0.11764706,0.11764706,0.11764706,0.13725491,0.13725491,0.1764706,0.15686275,0.1764706,0.3137255,0.23529412,0.29411766,0.11764706,0.27450982,0.3137255,0.3529412,0.37254903,0.37254903,0.39215687,0.23529412,0.39215687,0.3529412,0.27450982,0.4117647,0.39215687,0.4117647,0.4117647,0.39215687,0.39215687,0.4509804,0.47058824,0.47058824,0.37254903,0.49019608,0.49019608,,0.5686275,0.54901963,0.47058824,0.5686275,0.5882353,0.5686275,0.627451,0.5686275,0.5686275,0.60784316,0.60784316,0.627451,0.5980392,0.6862745,0.7254902,0.6666667,0.74509805,0.5882353,0.7058824,0.5882353,0.7254902,0.7254902,0.84313726,0.7254902,0.84313726,0.7647059,0.78431374,0.8235294,0.88235295,0.84313726,0.74509805,0.8039216,0.8235294,0.88235295,0.84313726,0.8627451,0.84313726,0.9019608,0.92156863,0.9019608,0.98039216,0.92156863,0.9607843
+TabICL (default),0.05882353,0.13725491,0.078431375,0.09803922,0.19607843,0.1764706,0.1764706,0.23529412,0.25490198,0.29411766,0.4117647,0.3137255,0.37254903,0.29411766,0.3529412,0.3529412,0.4019608,0.3137255,0.43137255,0.33333334,0.29411766,0.4117647,0.3529412,0.39215687,0.4117647,0.43137255,0.39215687,0.49019608,0.4117647,0.37254903,0.4509804,0.47058824,0.43137255,0.4509804,0.47058824,0.5686275,0.43137255,,0.5294118,0.54901963,0.5686275,0.5294118,0.5686275,0.5392157,0.5294118,0.5686275,0.57843137,0.5686275,0.5882353,0.627451,0.627451,0.60784316,0.64705884,0.627451,0.6862745,0.5882353,0.6862745,0.64705884,0.7058824,0.7058824,0.7647059,0.7058824,0.7647059,0.78431374,0.7745098,0.74509805,0.74509805,0.6666667,0.7647059,0.8137255,0.7941176,0.8235294,0.8235294,0.88235295,0.8627451,0.92156863,0.9019608,0.98039216,0.98039216,1.0
+EBM (tuned + ensembled),0.05882353,0.11764706,0.1764706,0.09803922,0.11764706,0.21568628,0.13725491,0.19607843,0.19607843,0.19607843,0.3529412,0.23529412,0.33333334,0.21568628,0.25490198,0.27450982,0.33333334,0.3137255,0.37254903,0.33333334,0.3137255,0.3137255,0.3137255,0.27450982,0.3529412,0.43137255,0.37254903,0.3529412,0.4117647,0.3137255,0.3529412,0.4509804,0.37254903,0.37254903,0.49019608,0.54901963,0.4509804,0.47058824,,0.4117647,0.50980395,0.54901963,0.5882353,0.47058824,0.4509804,0.49019608,0.6666667,0.78431374,0.64705884,0.84313726,0.60784316,0.60784316,0.60784316,0.6862745,0.627451,0.627451,0.627451,0.5882353,0.64705884,0.74509805,0.7058824,0.6666667,0.8039216,0.8039216,0.7058824,0.84313726,0.7254902,0.7647059,0.7254902,0.74509805,0.8235294,0.8627451,0.8627451,0.88235295,0.84313726,0.9607843,0.9607843,0.92156863,0.98039216,0.9607843
+TabPFNv2 (tuned),0.019607844,0.11764706,0.078431375,0.09803922,0.039215688,0.19607843,0.078431375,0.3137255,0.13725491,0.27450982,0.25490198,0.37254903,0.29411766,0.43137255,0.43137255,0.39215687,0.3529412,0.37254903,0.29411766,0.4509804,0.4509804,0.3137255,0.4509804,0.4509804,0.4117647,0.4509804,0.4509804,0.43137255,0.4509804,0.49019608,0.3137255,0.4509804,0.50980395,0.44117647,0.49019608,0.4509804,0.5294118,0.4509804,0.5882353,,0.54901963,0.50980395,0.51960784,0.5588235,0.54901963,0.5392157,0.49019608,0.5882353,0.50980395,0.64705884,0.64705884,0.5882353,0.5294118,0.60784316,0.54901963,0.5294118,0.54901963,0.5882353,0.61764705,0.6666667,0.7254902,0.627451,0.6862745,0.6862745,0.7352941,0.6862745,0.6862745,0.74509805,0.8235294,0.7352941,0.7745098,0.78431374,0.74509805,0.78431374,0.84313726,0.88235295,0.88235295,0.8627451,0.88235295,0.98039216
+TorchMLP (tuned + ensembled),0.039215688,0.05882353,0.09803922,0.078431375,0.13725491,0.13725491,0.13725491,0.078431375,0.15686275,0.15686275,0.25490198,0.19607843,0.25490198,0.25490198,0.23529412,0.23529412,0.23529412,0.25490198,0.3529412,0.3529412,0.33333334,0.33333334,0.27450982,0.33333334,0.29411766,0.33333334,0.3529412,0.43137255,0.39215687,0.37254903,0.43137255,0.4117647,0.4117647,0.4509804,0.4509804,0.49019608,0.43137255,0.43137255,0.49019608,0.4509804,,0.47058824,0.50980395,0.5294118,0.4509804,0.5686275,0.50980395,0.5294118,0.627451,0.54901963,0.74509805,0.64705884,0.64705884,0.6666667,0.5882353,0.6862745,0.5882353,0.6862745,0.74509805,0.78431374,0.7647059,0.7058824,0.74509805,0.7647059,0.78431374,0.9607843,0.7647059,0.7647059,0.7647059,0.8039216,0.9019608,0.88235295,0.8039216,0.88235295,0.9019608,0.92156863,0.9411765,0.9411765,0.98039216,0.9607843
+RealMLP (default),0.078431375,0.09803922,0.05882353,0.078431375,0.11764706,0.15686275,0.13725491,0.1764706,0.1764706,0.078431375,0.29411766,0.1764706,0.29411766,0.1764706,0.27450982,0.23529412,0.27450982,0.23529412,0.3529412,0.27450982,0.29411766,0.29411766,0.27450982,0.3137255,0.25490198,0.33333334,0.33333334,0.3529412,0.3529412,0.29411766,0.37254903,0.39215687,0.29411766,0.4117647,0.4117647,0.50980395,0.4117647,0.47058824,0.4509804,0.49019608,0.5294118,,0.49019608,0.50980395,0.47058824,0.5686275,0.5882353,0.5882353,0.60784316,0.5882353,0.60784316,0.6666667,0.6666667,0.64705884,0.5882353,0.6862745,0.5882353,0.60784316,0.7058824,0.74509805,0.7254902,0.7058824,0.7254902,0.7647059,0.74509805,0.8627451,0.7647059,0.7647059,0.8235294,0.8039216,0.92156863,0.9411765,0.88235295,0.88235295,0.92156863,0.92156863,0.9411765,0.9607843,0.98039216,1.0
+SAP-RPT-OSS (default),0.078431375,0.13725491,0.15686275,0.13725491,0.11764706,0.1764706,0.15686275,0.23529412,0.15686275,0.19607843,0.27450982,0.3137255,0.29411766,0.29411766,0.33333334,0.29411766,0.27450982,0.25490198,0.3529412,0.37254903,0.3529412,0.37254903,0.39215687,0.37254903,0.37254903,0.33333334,0.39215687,0.29411766,0.4117647,0.39215687,0.4117647,0.3529412,0.4117647,0.47058824,0.39215687,0.47058824,0.43137255,0.43137255,0.4117647,0.48039216,0.49019608,0.50980395,,0.47058824,0.47058824,0.5294118,0.54901963,0.50980395,0.54901963,0.5686275,0.54901963,0.5294118,0.627451,0.64705884,0.5686275,0.5686275,0.5882353,0.60784316,0.5882353,0.7254902,0.6666667,0.74509805,0.6862745,0.6862745,0.6862745,0.74509805,0.74509805,0.7254902,0.78431374,0.74509805,0.8235294,0.84313726,0.88235295,0.8039216,0.8235294,0.9019608,0.88235295,0.92156863,0.9019608,1.0
+BetaTabPFN (default),0.039215688,0.11764706,0.05882353,0.11764706,0.09803922,0.09803922,0.13725491,0.1764706,0.13725491,0.21568628,0.33333334,0.23529412,0.3529412,0.23529412,0.33333334,0.3137255,0.37254903,0.3529412,0.4117647,0.3137255,0.27450982,0.3529412,0.39215687,0.3137255,0.4509804,0.3529412,0.37254903,0.39215687,0.39215687,0.47058824,0.34313726,0.39215687,0.43137255,0.39215687,0.4117647,0.49019608,0.37254903,0.46078432,0.5294118,0.44117647,0.47058824,0.49019608,0.5294118,,0.50980395,0.49019608,0.49019608,0.6666667,0.50980395,0.5882353,0.627451,0.5686275,0.50980395,0.5686275,0.5882353,0.5294118,0.5882353,0.5686275,0.5980392,0.6666667,0.7647059,0.6666667,0.7647059,0.7647059,0.71568626,0.7254902,0.6862745,0.64705884,0.6666667,0.71568626,0.7941176,0.8235294,0.8627451,0.88235295,0.8039216,0.92156863,0.92156863,0.98039216,0.9607843,1.0
+ChimeraBoost (default),0.05882353,0.039215688,0.09803922,0.11764706,0.13725491,0.1764706,0.15686275,0.11764706,0.15686275,0.19607843,0.3529412,0.3137255,0.3137255,0.25490198,0.19607843,0.21568628,0.25490198,0.13725491,0.3137255,0.33333334,0.33333334,0.33333334,0.25490198,0.29411766,0.21568628,0.3529412,0.3137255,0.29411766,0.47058824,0.15686275,0.39215687,0.47058824,0.27450982,0.4117647,0.49019608,0.43137255,0.43137255,0.47058824,0.54901963,0.4509804,0.54901963,0.5294118,0.5294118,0.49019608,,0.5686275,0.54901963,0.627451,0.50980395,0.60784316,0.54901963,0.60784316,0.54901963,0.6666667,0.54901963,0.6666667,0.5686275,0.60784316,0.7254902,0.6862745,0.74509805,0.7647059,0.7647059,0.8039216,0.7647059,0.74509805,0.78431374,0.74509805,0.7647059,0.7647059,0.84313726,0.8039216,0.84313726,0.8235294,0.8235294,0.9019608,0.9019608,0.9607843,0.98039216,0.98039216
+TabPFNv2 (default),0.039215688,0.15686275,0.15686275,0.1764706,0.09803922,0.13725491,0.15686275,0.23529412,0.078431375,0.29411766,0.29411766,0.3137255,0.33333334,0.3529412,0.33333334,0.3529412,0.27450982,0.37254903,0.3529412,0.3529412,0.39215687,0.3529412,0.3529412,0.39215687,0.37254903,0.33333334,0.3529412,0.4117647,0.39215687,0.39215687,0.34313726,0.3137255,0.39215687,0.4117647,0.4509804,0.4509804,0.43137255,0.43137255,0.50980395,0.46078432,0.43137255,0.43137255,0.47058824,0.50980395,0.43137255,,0.43137255,0.50980395,0.47058824,0.54901963,0.54901963,0.47058824,0.50980395,0.5686275,0.5294118,0.5294118,0.54901963,0.50980395,0.57843137,0.627451,0.6666667,0.627451,0.6666667,0.6862745,0.7058824,0.64705884,0.64705884,0.74509805,0.74509805,0.7058824,0.7647059,0.7647059,0.7254902,0.7254902,0.8235294,0.8627451,0.8627451,0.84313726,0.9019608,0.98039216
+ModernNCA (default),0.039215688,0.039215688,0.039215688,0.039215688,0.11764706,0.09803922,0.09803922,0.078431375,0.11764706,0.09803922,0.1764706,0.23529412,0.27450982,0.25490198,0.19607843,0.27450982,0.23529412,0.29411766,0.23529412,0.19607843,0.27450982,0.3137255,0.29411766,0.3529412,0.37254903,0.23529412,0.3137255,0.33333334,0.25490198,0.39215687,0.4509804,0.37254903,0.43137255,0.39215687,0.37254903,0.3529412,0.39215687,0.42156863,0.33333334,0.50980395,0.49019608,0.4117647,0.4509804,0.50980395,0.4509804,0.5686275,,0.43137255,0.5686275,0.4509804,0.47058824,0.47058824,0.54901963,0.5294118,0.60784316,0.6666667,0.64705884,0.5882353,0.7058824,0.64705884,0.6862745,0.7058824,0.78431374,0.7647059,0.78431374,0.84313726,0.78431374,0.7254902,0.8235294,0.8627451,0.8235294,0.84313726,0.8235294,0.8627451,0.88235295,0.8627451,0.8627451,0.9411765,0.9019608,0.98039216
+EBM (tuned),0.05882353,0.09803922,0.11764706,0.05882353,0.11764706,0.15686275,0.13725491,0.09803922,0.11764706,0.15686275,0.27450982,0.21568628,0.29411766,0.23529412,0.1764706,0.23529412,0.3137255,0.25490198,0.3529412,0.33333334,0.27450982,0.27450982,0.25490198,0.25490198,0.3137255,0.3137255,0.25490198,0.3529412,0.37254903,0.27450982,0.33333334,0.3529412,0.33333334,0.3529412,0.4117647,0.49019608,0.39215687,0.43137255,0.21568628,0.4117647,0.47058824,0.4117647,0.49019608,0.33333334,0.37254903,0.49019608,0.5686275,,0.54901963,0.5882353,0.50980395,0.49019608,0.5882353,0.54901963,0.5882353,0.5686275,0.60784316,0.5294118,0.5882353,0.6862745,0.6862745,0.6666667,0.7647059,0.7647059,0.6862745,0.7647059,0.6666667,0.7647059,0.7058824,0.7254902,0.8039216,0.8235294,0.8627451,0.8627451,0.8039216,0.88235295,0.92156863,0.92156863,0.92156863,0.9607843
+ExtraTrees (tuned + ensembled),0.039215688,0.05882353,0.039215688,0.05882353,0.05882353,0.11764706,0.039215688,0.13725491,0.13725491,0.13725491,0.21568628,0.23529412,0.1764706,0.25490198,0.15686275,0.21568628,0.15686275,0.1764706,0.27450982,0.23529412,0.27450982,0.33333334,0.1764706,0.27450982,0.27450982,0.33333334,0.1764706,0.33333334,0.25490198,0.33333334,0.4509804,0.3137255,0.37254903,0.47058824,0.27450982,0.39215687,0.37254903,0.4117647,0.3529412,0.49019608,0.37254903,0.39215687,0.4509804,0.49019608,0.49019608,0.5294118,0.43137255,0.4509804,,0.43137255,0.46078432,0.43137255,0.71568626,0.50980395,0.64705884,0.64705884,0.64705884,0.50980395,0.7058824,0.54901963,0.64705884,0.64705884,0.7058824,0.7254902,0.78431374,0.60784316,0.7254902,0.7254902,0.9411765,0.88235295,0.84313726,0.7647059,0.78431374,0.92156863,0.78431374,0.8235294,0.88235295,0.9411765,0.88235295,1.0
+EBM (default),0.05882353,0.078431375,0.09803922,0.09803922,0.09803922,0.09803922,0.11764706,0.09803922,0.11764706,0.13725491,0.21568628,0.19607843,0.23529412,0.19607843,0.13725491,0.1764706,0.23529412,0.1764706,0.27450982,0.27450982,0.27450982,0.25490198,0.21568628,0.21568628,0.23529412,0.29411766,0.23529412,0.27450982,0.3529412,0.27450982,0.3137255,0.25490198,0.25490198,0.33333334,0.43137255,0.43137255,0.4019608,0.37254903,0.15686275,0.3529412,0.4509804,0.4117647,0.43137255,0.4117647,0.39215687,0.4509804,0.54901963,0.4117647,0.5686275,,0.5294118,0.4117647,0.5882353,0.5686275,0.54901963,0.5882353,0.5686275,0.50980395,0.627451,0.6666667,0.6862745,0.6862745,0.7254902,0.7254902,0.6862745,0.78431374,0.6862745,0.6666667,0.7058824,0.6666667,0.78431374,0.78431374,0.84313726,0.84313726,0.8235294,0.9019608,0.92156863,0.92156863,0.9607843,0.9607843
+TorchMLP (tuned),0.019607844,0.078431375,0.039215688,0.039215688,0.09803922,0.11764706,0.078431375,0.05882353,0.11764706,0.15686275,0.23529412,0.13725491,0.1764706,0.19607843,0.19607843,0.1764706,0.15686275,0.1764706,0.25490198,0.21568628,0.19607843,0.33333334,0.2647059,0.23529412,0.23529412,0.23529412,0.27450982,0.33333334,0.33333334,0.29411766,0.3137255,0.3529412,0.37254903,0.29411766,0.33333334,0.43137255,0.3137255,0.37254903,0.39215687,0.3529412,0.25490198,0.39215687,0.4509804,0.37254903,0.4509804,0.4509804,0.5294118,0.49019608,0.5392157,0.47058824,,0.49019608,0.5882353,0.50980395,0.50980395,0.5882353,0.50980395,0.5588235,0.5882353,0.627451,0.6666667,0.7058824,0.7647059,0.74509805,0.6862745,0.92156863,0.7254902,0.74509805,0.7254902,0.6666667,0.84313726,0.8039216,0.84313726,0.84313726,0.84313726,0.84313726,0.8627451,0.9411765,0.9411765,0.9607843
+XGBoost (default),0.039215688,0.05882353,0.05882353,0.039215688,0.039215688,0.13725491,0.05882353,0.078431375,0.11764706,0.09803922,0.25490198,0.15686275,0.21568628,0.21568628,0.09803922,0.13725491,0.1764706,0.13725491,0.3137255,0.21568628,0.21568628,0.29411766,0.11764706,0.1764706,0.23529412,0.25490198,0.13725491,0.33333334,0.3137255,0.27450982,0.3529412,0.39215687,0.25490198,0.39215687,0.39215687,0.4117647,0.27450982,0.39215687,0.39215687,0.4117647,0.3529412,0.33333334,0.47058824,0.43137255,0.39215687,0.5294118,0.5294118,0.50980395,0.5686275,0.5882353,0.50980395,,0.5882353,0.5686275,0.5882353,0.5882353,0.60784316,0.5294118,0.64705884,0.627451,0.78431374,0.627451,0.7254902,0.7254902,0.7058824,0.74509805,0.7254902,0.74509805,0.7254902,0.78431374,0.7647059,0.7647059,0.78431374,0.78431374,0.8627451,0.84313726,0.8627451,0.9019608,0.88235295,0.9607843
+ExtraTrees (tuned),0.039215688,0.05882353,0.039215688,0.05882353,0.05882353,0.078431375,0.09803922,0.13725491,0.13725491,0.13725491,0.13725491,0.1764706,0.15686275,0.25490198,0.19607843,0.23529412,0.1764706,0.1764706,0.1764706,0.23529412,0.27450982,0.33333334,0.19607843,0.23529412,0.27450982,0.3137255,0.19607843,0.29411766,0.21568628,0.3137255,0.39215687,0.25490198,0.33333334,0.4117647,0.25490198,0.33333334,0.33333334,0.3529412,0.39215687,0.47058824,0.3529412,0.33333334,0.37254903,0.49019608,0.4509804,0.49019608,0.4509804,0.4117647,0.28431374,0.4117647,0.4117647,0.4117647,,0.50980395,0.60784316,0.47058824,0.60784316,0.43137255,0.60784316,0.5294118,0.5882353,0.64705884,0.6862745,0.6862745,0.7647059,0.54901963,0.7058824,0.6666667,0.8235294,0.84313726,0.8235294,0.6862745,0.8039216,0.84313726,0.7647059,0.7647059,0.7647059,0.98039216,0.88235295,0.98039216
+FastaiMLP (tuned + ensembled),0.078431375,0.09803922,0.09803922,0.078431375,0.11764706,0.05882353,0.13725491,0.078431375,0.078431375,0.078431375,0.25490198,0.15686275,0.21568628,0.1764706,0.1764706,0.21568628,0.13725491,0.21568628,0.33333334,0.19607843,0.25490198,0.3137255,0.21568628,0.21568628,0.19607843,0.28431374,0.23529412,0.1764706,0.27450982,0.23529412,0.33333334,0.21568628,0.27450982,0.33333334,0.3137255,0.3529412,0.25490198,0.37254903,0.3137255,0.39215687,0.33333334,0.3529412,0.3529412,0.43137255,0.33333334,0.43137255,0.47058824,0.4509804,0.49019608,0.43137255,0.49019608,0.43137255,0.49019608,,0.54901963,0.5294118,0.54901963,0.4509804,0.5686275,0.7058824,0.64705884,0.60784316,0.6666667,0.6666667,0.6666667,0.6666667,0.627451,0.7254902,0.7254902,0.7058824,0.8039216,0.98039216,0.7647059,0.78431374,0.84313726,0.8627451,0.88235295,0.8627451,0.92156863,0.9607843
+Nori-30M (default),0.05882353,0.19607843,0.11764706,0.1764706,0.15686275,0.19607843,0.15686275,0.23529412,0.21568628,0.23529412,0.1764706,0.25490198,0.29411766,0.37254903,0.25490198,0.29411766,0.29411766,0.27450982,0.33333334,0.27450982,0.3529412,0.27450982,0.29411766,0.29411766,0.29411766,0.27450982,0.3137255,0.3529412,0.3137255,0.3529412,0.37254903,0.3529412,0.39215687,0.4117647,0.37254903,0.4117647,0.4117647,0.3137255,0.37254903,0.4509804,0.4117647,0.4117647,0.43137255,0.4117647,0.4509804,0.47058824,0.39215687,0.4117647,0.3529412,0.4509804,0.49019608,0.4117647,0.39215687,0.4509804,,0.3529412,0.54901963,0.47058824,0.4019608,0.50980395,0.50980395,0.54901963,0.50980395,0.5294118,0.54901963,0.5686275,0.60784316,0.6862745,0.64705884,0.627451,0.627451,0.627451,0.627451,0.64705884,0.6862745,0.7058824,0.7058824,0.78431374,0.74509805,0.98039216
+RandomForest (tuned + ensembled),0.039215688,0.039215688,0.05882353,0.039215688,0.078431375,0.11764706,0.078431375,0.09803922,0.13725491,0.15686275,0.21568628,0.15686275,0.1764706,0.19607843,0.13725491,0.15686275,0.09803922,0.11764706,0.3137255,0.21568628,0.25490198,0.29411766,0.13725491,0.21568628,0.19607843,0.27450982,0.1764706,0.3137255,0.19607843,0.23529412,0.3529412,0.19607843,0.27450982,0.43137255,0.25490198,0.33333334,0.29411766,0.4117647,0.37254903,0.47058824,0.3137255,0.3137255,0.43137255,0.47058824,0.33333334,0.47058824,0.33333334,0.43137255,0.3529412,0.4117647,0.4117647,0.4117647,0.5294118,0.47058824,0.64705884,,0.64705884,0.49019608,0.7058824,0.54901963,0.627451,0.5686275,0.6862745,0.7058824,0.7058824,0.6862745,0.6666667,0.6862745,0.8039216,0.88235295,0.7647059,0.74509805,0.8039216,0.84313726,0.6862745,0.8039216,0.78431374,0.9411765,0.8627451,0.98039216
+Nori (default),0.039215688,0.13725491,0.09803922,0.13725491,0.11764706,0.19607843,0.13725491,0.1764706,0.1764706,0.21568628,0.1764706,0.27450982,0.23529412,0.33333334,0.23529412,0.29411766,0.29411766,0.27450982,0.29411766,0.25490198,0.33333334,0.27450982,0.27450982,0.27450982,0.29411766,0.25490198,0.29411766,0.37254903,0.3137255,0.3529412,0.3529412,0.3529412,0.39215687,0.4117647,0.37254903,0.39215687,0.4117647,0.3137255,0.37254903,0.4509804,0.4117647,0.4117647,0.4117647,0.4117647,0.43137255,0.4509804,0.3529412,0.39215687,0.3529412,0.43137255,0.49019608,0.39215687,0.39215687,0.4509804,0.4509804,0.3529412,,0.47058824,0.4019608,0.50980395,0.50980395,0.5294118,0.50980395,0.5294118,0.54901963,0.5686275,0.60784316,0.6862745,0.64705884,0.627451,0.627451,0.627451,0.627451,0.64705884,0.6862745,0.7058824,0.7058824,0.78431374,0.74509805,0.98039216
+LightGBM (default),0.039215688,0.019607844,0.039215688,0.039215688,0.019607844,0.078431375,0.039215688,0.05882353,0.039215688,0.05882353,0.1764706,0.15686275,0.13725491,0.13725491,0.039215688,0.15686275,0.13725491,0.15686275,0.27450982,0.15686275,0.1764706,0.3137255,0.019607844,0.13725491,0.21568628,0.21568628,0.09803922,0.23529412,0.25490198,0.25490198,0.3137255,0.25490198,0.23529412,0.37254903,0.3137255,0.3137255,0.27450982,0.3529412,0.4117647,0.4117647,0.3137255,0.39215687,0.39215687,0.43137255,0.39215687,0.49019608,0.4117647,0.47058824,0.49019608,0.49019608,0.44117647,0.47058824,0.5686275,0.54901963,0.5294118,0.50980395,0.5294118,,0.54901963,0.60784316,0.7254902,0.6666667,0.7058824,0.7254902,0.7058824,0.7058824,0.7058824,0.7254902,0.7647059,0.7647059,0.8039216,0.78431374,0.8039216,0.7647059,0.8039216,0.78431374,0.78431374,0.9411765,0.9019608,0.9411765
+RandomForest (tuned),0.039215688,0.039215688,0.019607844,0.05882353,0.05882353,0.11764706,0.078431375,0.039215688,0.078431375,0.09803922,0.19607843,0.11764706,0.13725491,0.19607843,0.05882353,0.09803922,0.09803922,0.078431375,0.21568628,0.23529412,0.21568628,0.25490198,0.09803922,0.11764706,0.15686275,0.21568628,0.13725491,0.29411766,0.1764706,0.19607843,0.32352942,0.19607843,0.19607843,0.3627451,0.21568628,0.29411766,0.27450982,0.29411766,0.3529412,0.38235295,0.25490198,0.29411766,0.4117647,0.4019608,0.27450982,0.42156863,0.29411766,0.4117647,0.29411766,0.37254903,0.4117647,0.3529412,0.39215687,0.43137255,0.5980392,0.29411766,0.5980392,0.4509804,,0.50980395,0.627451,0.54901963,0.6666667,0.6666667,0.6764706,0.5882353,0.5882353,0.6666667,0.7254902,0.78431374,0.75490195,0.7647059,0.7647059,0.84313726,0.6666667,0.78431374,0.7647059,0.92156863,0.84313726,0.98039216
+FastaiMLP (tuned),0.05882353,0.05882353,0.039215688,0.078431375,0.05882353,0.078431375,0.078431375,0.05882353,0.019607844,0.09803922,0.15686275,0.11764706,0.09803922,0.11764706,0.13725491,0.15686275,0.13725491,0.11764706,0.1764706,0.13725491,0.13725491,0.21568628,0.1764706,0.15686275,0.15686275,0.1764706,0.15686275,0.19607843,0.21568628,0.25490198,0.27450982,0.19607843,0.23529412,0.3137255,0.25490198,0.21568628,0.15686275,0.29411766,0.25490198,0.33333334,0.21568628,0.25490198,0.27450982,0.33333334,0.3137255,0.37254903,0.3529412,0.3137255,0.4509804,0.33333334,0.37254903,0.37254903,0.47058824,0.29411766,0.49019608,0.4509804,0.49019608,0.39215687,0.49019608,,0.5686275,0.5294118,0.627451,0.627451,0.5686275,0.5882353,0.5686275,0.64705884,0.6862745,0.627451,0.7254902,0.78431374,0.74509805,0.6862745,0.7254902,0.78431374,0.78431374,0.84313726,0.88235295,0.9411765
+iLTM (default),0.05882353,0.039215688,0.019607844,0.039215688,0.078431375,0.11764706,0.09803922,0.05882353,0.078431375,0.078431375,0.19607843,0.09803922,0.15686275,0.15686275,0.078431375,0.11764706,0.09803922,0.11764706,0.1764706,0.1764706,0.21568628,0.23529412,0.11764706,0.13725491,0.13725491,0.1764706,0.13725491,0.21568628,0.23529412,0.21568628,0.25490198,0.2254902,0.21568628,0.27450982,0.23529412,0.3137255,0.27450982,0.23529412,0.29411766,0.27450982,0.23529412,0.27450982,0.33333334,0.23529412,0.25490198,0.33333334,0.3137255,0.3137255,0.3529412,0.3137255,0.33333334,0.21568628,0.4117647,0.3529412,0.49019608,0.37254903,0.49019608,0.27450982,0.37254903,0.43137255,,0.47058824,0.6666667,0.64705884,0.50980395,0.5686275,0.4509804,0.5294118,0.5294118,0.54901963,0.5882353,0.627451,0.7647059,0.7647059,0.5882353,0.7647059,0.7647059,0.8235294,0.84313726,0.92156863
+PerpetualBooster (tuned + ensembled),0.0,0.05882353,0.039215688,0.019607844,0.039215688,0.078431375,0.039215688,0.039215688,0.05882353,0.039215688,0.15686275,0.09803922,0.15686275,0.15686275,0.09803922,0.11764706,0.09803922,0.078431375,0.15686275,0.15686275,0.11764706,0.25490198,0.15686275,0.11764706,0.09803922,0.1764706,0.19607843,0.19607843,0.19607843,0.1764706,0.3137255,0.23529412,0.19607843,0.3137255,0.25490198,0.25490198,0.15686275,0.29411766,0.33333334,0.37254903,0.29411766,0.29411766,0.25490198,0.33333334,0.23529412,0.37254903,0.29411766,0.33333334,0.3529412,0.3137255,0.29411766,0.37254903,0.3529412,0.39215687,0.4509804,0.43137255,0.47058824,0.33333334,0.4509804,0.47058824,0.5294118,,0.47058824,0.47058824,0.5294118,0.49019608,0.8137255,0.60784316,0.5882353,0.64705884,0.64705884,0.60784316,0.49019608,0.60784316,0.9019608,0.74509805,0.74509805,0.74509805,0.7647059,0.92156863
+TabSTAR (tuned),0.05882353,0.09803922,0.09803922,0.11764706,0.11764706,0.09803922,0.13725491,0.1764706,0.09803922,0.15686275,0.1764706,0.15686275,0.19607843,0.13725491,0.15686275,0.1764706,0.19607843,0.1764706,0.19607843,0.25490198,0.1764706,0.2254902,0.15686275,0.19607843,0.15686275,0.21568628,0.15686275,0.21568628,0.23529412,0.1764706,0.29411766,0.21568628,0.21568628,0.27450982,0.25490198,0.29411766,0.23529412,0.23529412,0.19607843,0.3137255,0.25490198,0.27450982,0.3137255,0.23529412,0.23529412,0.33333334,0.21568628,0.23529412,0.29411766,0.27450982,0.23529412,0.27450982,0.3137255,0.33333334,0.49019608,0.3137255,0.49019608,0.29411766,0.33333334,0.37254903,0.33333334,0.5294118,,0.51960784,0.43137255,0.43137255,0.5294118,0.5294118,0.5686275,0.54901963,0.5686275,0.54901963,0.9607843,0.5882353,0.54901963,0.7058824,0.7254902,0.78431374,0.8039216,0.84313726
+TabSTAR (tuned + ensembled),0.078431375,0.11764706,0.09803922,0.11764706,0.11764706,0.09803922,0.15686275,0.1764706,0.09803922,0.15686275,0.21568628,0.13725491,0.21568628,0.13725491,0.1764706,0.1764706,0.19607843,0.1764706,0.21568628,0.23529412,0.1764706,0.23529412,0.15686275,0.1764706,0.1764706,0.21568628,0.15686275,0.21568628,0.23529412,0.19607843,0.3137255,0.21568628,0.21568628,0.27450982,0.23529412,0.27450982,0.21568628,0.21568628,0.19607843,0.3137255,0.23529412,0.23529412,0.3137255,0.23529412,0.19607843,0.3137255,0.23529412,0.23529412,0.27450982,0.27450982,0.25490198,0.27450982,0.3137255,0.33333334,0.47058824,0.29411766,0.47058824,0.27450982,0.33333334,0.37254903,0.3529412,0.5294118,0.48039216,,0.4509804,0.4509804,0.5294118,0.5294118,0.54901963,0.5294118,0.54901963,0.5294118,0.98039216,0.5686275,0.54901963,0.7058824,0.7058824,0.78431374,0.8039216,0.84313726
+OrionMSP (default),0.019607844,0.0,0.0,0.019607844,0.019607844,0.078431375,0.019607844,0.05882353,0.019607844,0.039215688,0.039215688,0.12745099,0.05882353,0.09803922,0.11764706,0.13725491,0.11764706,0.15686275,0.078431375,0.13725491,0.11764706,0.18627451,0.11764706,0.13725491,0.15686275,0.15686275,0.13725491,0.11764706,0.09803922,0.1764706,0.2254902,0.19607843,0.19607843,0.2254902,0.15686275,0.1764706,0.1764706,0.2254902,0.29411766,0.2647059,0.21568628,0.25490198,0.3137255,0.28431374,0.23529412,0.29411766,0.21568628,0.3137255,0.21568628,0.3137255,0.3137255,0.29411766,0.23529412,0.33333334,0.4509804,0.29411766,0.4509804,0.29411766,0.32352942,0.43137255,0.49019608,0.47058824,0.5686275,0.54901963,,0.4509804,0.47058824,0.50980395,0.60784316,0.57843137,0.65686274,0.5686275,0.7254902,0.6666667,0.6666667,0.7647059,0.7647059,0.8039216,0.8039216,0.98039216
+TorchMLP (default),0.019607844,0.039215688,0.039215688,0.019607844,0.039215688,0.039215688,0.039215688,0.019607844,0.019607844,0.05882353,0.13725491,0.039215688,0.11764706,0.019607844,0.078431375,0.05882353,0.039215688,0.05882353,0.13725491,0.15686275,0.019607844,0.19607843,0.078431375,0.11764706,0.09803922,0.13725491,0.09803922,0.1764706,0.11764706,0.1764706,0.27450982,0.078431375,0.1764706,0.29411766,0.09803922,0.21568628,0.11764706,0.25490198,0.15686275,0.3137255,0.039215688,0.13725491,0.25490198,0.27450982,0.25490198,0.3529412,0.15686275,0.23529412,0.39215687,0.21568628,0.078431375,0.25490198,0.4509804,0.33333334,0.43137255,0.3137255,0.43137255,0.29411766,0.4117647,0.4117647,0.43137255,0.50980395,0.5686275,0.54901963,0.54901963,,0.5294118,0.54901963,0.64705884,0.5882353,0.6666667,0.54901963,0.7058824,0.6666667,0.627451,0.7058824,0.7647059,0.84313726,0.84313726,0.92156863
+PerpetualBooster (tuned),0.0,0.039215688,0.0,0.0,0.019607844,0.05882353,0.019607844,0.019607844,0.019607844,0.019607844,0.09803922,0.078431375,0.09803922,0.09803922,0.078431375,0.05882353,0.05882353,0.078431375,0.09803922,0.15686275,0.09803922,0.23529412,0.11764706,0.09803922,0.05882353,0.15686275,0.11764706,0.15686275,0.15686275,0.13725491,0.25490198,0.15686275,0.1764706,0.29411766,0.1764706,0.1764706,0.15686275,0.25490198,0.27450982,0.3137255,0.23529412,0.23529412,0.25490198,0.3137255,0.21568628,0.3529412,0.21568628,0.33333334,0.27450982,0.3137255,0.27450982,0.27450982,0.29411766,0.37254903,0.39215687,0.33333334,0.39215687,0.29411766,0.4117647,0.43137255,0.54901963,0.18627451,0.47058824,0.47058824,0.5294118,0.47058824,,0.5686275,0.50980395,0.5882353,0.627451,0.5882353,0.49019608,0.54901963,0.8137255,0.7254902,0.7058824,0.6666667,0.74509805,0.92156863
+xRFM (default),0.039215688,0.019607844,0.039215688,0.039215688,0.05882353,0.078431375,0.078431375,0.078431375,0.078431375,0.05882353,0.05882353,0.078431375,0.09803922,0.15686275,0.15686275,0.15686275,0.078431375,0.11764706,0.09803922,0.09803922,0.1764706,0.1764706,0.1764706,0.15686275,0.13725491,0.15686275,0.21568628,0.1764706,0.078431375,0.19607843,0.23529412,0.11764706,0.19607843,0.19607843,0.039215688,0.19607843,0.25490198,0.33333334,0.23529412,0.25490198,0.23529412,0.23529412,0.27450982,0.3529412,0.25490198,0.25490198,0.27450982,0.23529412,0.27450982,0.33333334,0.25490198,0.25490198,0.33333334,0.27450982,0.3137255,0.3137255,0.3137255,0.27450982,0.33333334,0.3529412,0.47058824,0.39215687,0.47058824,0.47058824,0.49019608,0.4509804,0.43137255,,0.49019608,0.5294118,0.5882353,0.49019608,0.60784316,0.5294118,0.54901963,0.60784316,0.60784316,0.7254902,0.64705884,0.88235295
+ExtraTrees (default),0.019607844,0.019607844,0.019607844,0.039215688,0.019607844,0.078431375,0.039215688,0.019607844,0.019607844,0.019607844,0.019607844,0.09803922,0.039215688,0.13725491,0.078431375,0.11764706,0.039215688,0.09803922,0.05882353,0.13725491,0.15686275,0.11764706,0.078431375,0.15686275,0.13725491,0.1764706,0.09803922,0.11764706,0.05882353,0.21568628,0.15686275,0.13725491,0.19607843,0.21568628,0.078431375,0.15686275,0.19607843,0.23529412,0.27450982,0.1764706,0.23529412,0.1764706,0.21568628,0.33333334,0.23529412,0.25490198,0.1764706,0.29411766,0.05882353,0.29411766,0.27450982,0.27450982,0.1764706,0.27450982,0.3529412,0.19607843,0.3529412,0.23529412,0.27450982,0.3137255,0.47058824,0.4117647,0.43137255,0.4509804,0.39215687,0.3529412,0.49019608,0.50980395,,0.50980395,0.54901963,0.54901963,0.5882353,0.5686275,0.60784316,0.6666667,0.6666667,0.7647059,0.7058824,0.9607843
+RandomForest (default),0.019607844,0.039215688,0.019607844,0.019607844,0.019607844,0.05882353,0.039215688,0.019607844,0.05882353,0.039215688,0.039215688,0.05882353,0.05882353,0.11764706,0.019607844,0.05882353,0.039215688,0.039215688,0.11764706,0.09803922,0.09803922,0.11764706,0.039215688,0.05882353,0.078431375,0.09803922,0.05882353,0.1764706,0.078431375,0.13725491,0.20588236,0.09803922,0.19607843,0.21568628,0.13725491,0.21568628,0.1764706,0.18627451,0.25490198,0.2647059,0.19607843,0.19607843,0.25490198,0.28431374,0.23529412,0.29411766,0.13725491,0.27450982,0.11764706,0.33333334,0.33333334,0.21568628,0.15686275,0.29411766,0.37254903,0.11764706,0.37254903,0.23529412,0.21568628,0.37254903,0.4509804,0.3529412,0.4509804,0.47058824,0.42156863,0.4117647,0.4117647,0.47058824,0.49019608,,0.5,0.54901963,0.5882353,0.60784316,0.5882353,0.7058824,0.7058824,0.78431374,0.74509805,0.98039216
+TabFlex (default),0.0,0.039215688,0.039215688,0.05882353,0.078431375,0.039215688,0.09803922,0.039215688,0.019607844,0.078431375,0.05882353,0.09803922,0.039215688,0.05882353,0.09803922,0.09803922,0.05882353,0.11764706,0.09803922,0.13725491,0.05882353,0.1764706,0.11764706,0.11764706,0.11764706,0.13725491,0.11764706,0.05882353,0.078431375,0.11764706,0.2254902,0.09803922,0.11764706,0.21568628,0.11764706,0.11764706,0.11764706,0.20588236,0.1764706,0.2254902,0.09803922,0.078431375,0.1764706,0.20588236,0.15686275,0.23529412,0.1764706,0.19607843,0.15686275,0.21568628,0.15686275,0.23529412,0.1764706,0.19607843,0.37254903,0.23529412,0.37254903,0.19607843,0.24509804,0.27450982,0.4117647,0.3529412,0.43137255,0.4509804,0.34313726,0.33333334,0.37254903,0.4117647,0.4509804,0.5,,0.49019608,0.5686275,0.54901963,0.5686275,0.6666667,0.6666667,0.7647059,0.7254902,0.9019608
+FastaiMLP (default),0.0,0.019607844,0.0,0.019607844,0.019607844,0.019607844,0.039215688,0.019607844,0.039215688,0.019607844,0.039215688,0.039215688,0.039215688,0.05882353,0.078431375,0.078431375,0.039215688,0.078431375,0.039215688,0.11764706,0.09803922,0.13725491,0.09803922,0.09803922,0.05882353,0.09803922,0.09803922,0.09803922,0.078431375,0.13725491,0.15686275,0.09803922,0.15686275,0.1764706,0.11764706,0.13725491,0.15686275,0.1764706,0.13725491,0.21568628,0.11764706,0.05882353,0.15686275,0.1764706,0.19607843,0.23529412,0.15686275,0.1764706,0.23529412,0.21568628,0.19607843,0.23529412,0.3137255,0.019607844,0.37254903,0.25490198,0.37254903,0.21568628,0.23529412,0.21568628,0.37254903,0.39215687,0.4509804,0.47058824,0.43137255,0.4509804,0.4117647,0.50980395,0.4509804,0.4509804,0.50980395,,0.6666667,0.5294118,0.6666667,0.6862745,0.7058824,0.74509805,0.74509805,0.9019608
+TabSTAR (default),0.039215688,0.09803922,0.09803922,0.11764706,0.09803922,0.078431375,0.09803922,0.11764706,0.078431375,0.11764706,0.11764706,0.13725491,0.13725491,0.11764706,0.13725491,0.15686275,0.13725491,0.15686275,0.13725491,0.13725491,0.13725491,0.15686275,0.13725491,0.13725491,0.13725491,0.11764706,0.13725491,0.13725491,0.1764706,0.15686275,0.23529412,0.13725491,0.15686275,0.21568628,0.1764706,0.23529412,0.13725491,0.1764706,0.13725491,0.25490198,0.19607843,0.11764706,0.11764706,0.13725491,0.15686275,0.27450982,0.1764706,0.13725491,0.21568628,0.15686275,0.15686275,0.21568628,0.19607843,0.23529412,0.37254903,0.19607843,0.37254903,0.19607843,0.23529412,0.25490198,0.23529412,0.50980395,0.039215688,0.019607844,0.27450982,0.29411766,0.50980395,0.39215687,0.4117647,0.4117647,0.43137255,0.33333334,,0.50980395,0.5294118,0.5882353,0.5882353,0.64705884,0.64705884,0.8039216
+KNN (tuned + ensembled),0.05882353,0.019607844,0.039215688,0.019607844,0.05882353,0.039215688,0.078431375,0.05882353,0.078431375,0.039215688,0.078431375,0.05882353,0.05882353,0.09803922,0.05882353,0.078431375,0.039215688,0.09803922,0.11764706,0.10784314,0.09803922,0.11764706,0.09803922,0.09803922,0.078431375,0.078431375,0.078431375,0.09803922,0.05882353,0.13725491,0.19607843,0.05882353,0.15686275,0.19607843,0.05882353,0.1764706,0.15686275,0.11764706,0.11764706,0.21568628,0.11764706,0.11764706,0.19607843,0.11764706,0.1764706,0.27450982,0.13725491,0.13725491,0.078431375,0.15686275,0.15686275,0.21568628,0.15686275,0.21568628,0.3529412,0.15686275,0.3529412,0.23529412,0.15686275,0.3137255,0.23529412,0.39215687,0.4117647,0.43137255,0.33333334,0.33333334,0.4509804,0.47058824,0.43137255,0.39215687,0.4509804,0.47058824,0.49019608,,0.54901963,0.6862745,0.6862745,0.98039216,0.7647059,0.98039216
+PerpetualBooster (default),0.0,0.039215688,0.019607844,0.019607844,0.039215688,0.039215688,0.039215688,0.019607844,0.05882353,0.019607844,0.078431375,0.019607844,0.039215688,0.05882353,0.078431375,0.039215688,0.039215688,0.039215688,0.05882353,0.09803922,0.05882353,0.13725491,0.09803922,0.078431375,0.039215688,0.11764706,0.09803922,0.09803922,0.11764706,0.13725491,0.11764706,0.09803922,0.13725491,0.11764706,0.078431375,0.13725491,0.09803922,0.13725491,0.15686275,0.15686275,0.09803922,0.078431375,0.1764706,0.19607843,0.1764706,0.1764706,0.11764706,0.19607843,0.21568628,0.1764706,0.15686275,0.13725491,0.23529412,0.15686275,0.3137255,0.3137255,0.3137255,0.19607843,0.33333334,0.27450982,0.4117647,0.09803922,0.4509804,0.4509804,0.33333334,0.37254903,0.18627451,0.4509804,0.39215687,0.4117647,0.43137255,0.33333334,0.47058824,0.4509804,,0.60784316,0.60784316,0.54901963,0.64705884,0.84313726
+Linear (tuned + ensembled),0.019607844,0.019607844,0.039215688,0.019607844,0.05882353,0.039215688,0.05882353,0.019607844,0.039215688,0.019607844,0.09803922,0.039215688,0.11764706,0.039215688,0.019607844,0.019607844,0.05882353,0.019607844,0.13725491,0.13725491,0.078431375,0.09803922,0.05882353,0.09803922,0.039215688,0.078431375,0.078431375,0.11764706,0.039215688,0.05882353,0.11764706,0.11764706,0.09803922,0.11764706,0.078431375,0.21568628,0.078431375,0.078431375,0.039215688,0.11764706,0.078431375,0.078431375,0.09803922,0.078431375,0.09803922,0.13725491,0.13725491,0.11764706,0.1764706,0.09803922,0.15686275,0.15686275,0.23529412,0.13725491,0.29411766,0.19607843,0.29411766,0.21568628,0.21568628,0.21568628,0.23529412,0.25490198,0.29411766,0.29411766,0.23529412,0.29411766,0.27450982,0.39215687,0.33333334,0.29411766,0.33333334,0.3137255,0.4117647,0.3137255,0.39215687,,0.7647059,0.49019608,0.7647059,0.7254902
+Linear (tuned),0.019607844,0.019607844,0.019607844,0.019607844,0.039215688,0.019607844,0.05882353,0.019607844,0.039215688,0.019607844,0.09803922,0.029411765,0.11764706,0.039215688,0.019607844,0.019607844,0.039215688,0.019607844,0.13725491,0.11764706,0.078431375,0.09803922,0.039215688,0.05882353,0.019607844,0.05882353,0.05882353,0.09803922,0.019607844,0.039215688,0.11764706,0.078431375,0.09803922,0.11764706,0.09803922,0.19607843,0.09803922,0.09803922,0.039215688,0.11764706,0.05882353,0.05882353,0.11764706,0.078431375,0.09803922,0.13725491,0.13725491,0.078431375,0.11764706,0.078431375,0.13725491,0.13725491,0.23529412,0.11764706,0.29411766,0.21568628,0.29411766,0.21568628,0.23529412,0.21568628,0.23529412,0.25490198,0.27450982,0.29411766,0.23529412,0.23529412,0.29411766,0.39215687,0.33333334,0.29411766,0.33333334,0.29411766,0.4117647,0.3137255,0.39215687,0.23529412,,0.47058824,0.6862745,0.7058824
+KNN (tuned),0.019607844,0.0,0.019607844,0.019607844,0.039215688,0.019607844,0.039215688,0.0,0.019607844,0.019607844,0.039215688,0.019607844,0.019607844,0.0,0.0,0.019607844,0.039215688,0.039215688,0.039215688,0.09803922,0.019607844,0.11764706,0.05882353,0.039215688,0.019607844,0.05882353,0.05882353,0.019607844,0.019607844,0.039215688,0.13725491,0.019607844,0.05882353,0.11764706,0.019607844,0.09803922,0.019607844,0.019607844,0.078431375,0.13725491,0.05882353,0.039215688,0.078431375,0.019607844,0.039215688,0.15686275,0.05882353,0.078431375,0.05882353,0.078431375,0.05882353,0.09803922,0.019607844,0.13725491,0.21568628,0.05882353,0.21568628,0.05882353,0.078431375,0.15686275,0.1764706,0.25490198,0.21568628,0.21568628,0.19607843,0.15686275,0.33333334,0.27450982,0.23529412,0.21568628,0.23529412,0.25490198,0.3529412,0.019607844,0.4509804,0.50980395,0.5294118,,0.54901963,0.92156863
+Linear (default),0.0,0.019607844,0.0,0.019607844,0.039215688,0.0,0.039215688,0.0,0.019607844,0.019607844,0.078431375,0.039215688,0.078431375,0.039215688,0.039215688,0.019607844,0.039215688,0.019607844,0.09803922,0.11764706,0.039215688,0.078431375,0.039215688,0.05882353,0.019607844,0.05882353,0.039215688,0.09803922,0.039215688,0.019607844,0.11764706,0.078431375,0.039215688,0.09803922,0.078431375,0.13725491,0.078431375,0.019607844,0.019607844,0.11764706,0.019607844,0.019607844,0.09803922,0.039215688,0.019607844,0.09803922,0.09803922,0.078431375,0.11764706,0.039215688,0.05882353,0.11764706,0.11764706,0.078431375,0.25490198,0.13725491,0.25490198,0.09803922,0.15686275,0.11764706,0.15686275,0.23529412,0.19607843,0.19607843,0.19607843,0.15686275,0.25490198,0.3529412,0.29411766,0.25490198,0.27450982,0.25490198,0.3529412,0.23529412,0.3529412,0.23529412,0.3137255,0.4509804,,0.6666667
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.019607844,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.019607844,0.019607844,0.019607844,0.019607844,0.0,0.0,0.0,0.0,0.0,0.0,0.019607844,0.0,0.0,0.0,0.0,0.0,0.0,0.019607844,0.039215688,0.0,0.039215688,0.019607844,0.039215688,0.0,0.0,0.0,0.019607844,0.019607844,0.019607844,0.039215688,0.0,0.039215688,0.039215688,0.039215688,0.019607844,0.039215688,0.019607844,0.019607844,0.019607844,0.05882353,0.019607844,0.05882353,0.078431375,0.078431375,0.15686275,0.15686275,0.019607844,0.078431375,0.078431375,0.11764706,0.039215688,0.019607844,0.09803922,0.09803922,0.19607843,0.019607844,0.15686275,0.27450982,0.29411766,0.078431375,0.33333334,
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/leaderboard_overview_explorer.html
index d631030b23935c1ec9b67e4b9c215a2bc267b9dd..2cbed86b40826e33503f9fbbbc929644bd1d8387 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | All Tasks | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..54cc1974ff04c3c6b7dcc5262ce3f34a0c7ce66e
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_all/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1922.0,"score":0.784,"rank":5.67,"hrank":1.97,"imp":4.458,"train_time":27.18,"predict_time":13.118,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+236\/-130","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1914.0,"score":0.736,"rank":5.87,"hrank":2.57,"imp":3.418,"train_time":182.61,"predict_time":2.229,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-67","imputed":false},{"position":2,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1847.0,"score":0.663,"rank":7.67,"hrank":3.7,"imp":4.717,"train_time":432.15,"predict_time":4.214,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+144\/-68","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1784.0,"score":0.68,"rank":9.73,"hrank":2.7,"imp":4.59,"train_time":1.02,"predict_time":0.214,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+236\/-133","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1746.0,"score":0.542,"rank":11.2,"hrank":8.43,"imp":7.097,"train_time":2.64,"predict_time":0.702,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-67","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1700.0,"score":0.513,"rank":13.2,"hrank":7.24,"imp":6.49,"train_time":739.1,"predict_time":11.65,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+157\/-100","imputed":false},{"position":6,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1695.0,"score":0.535,"rank":13.4,"hrank":5.61,"imp":7.508,"train_time":0.56,"predict_time":0.118,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+204\/-117","imputed":false},{"position":7,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1653.0,"score":0.496,"rank":15.47,"hrank":10.13,"imp":6.901,"train_time":2522.59,"predict_time":6.234,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-116","imputed":false},{"position":8,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1650.0,"score":0.474,"rank":15.6,"hrank":9.55,"imp":7.814,"train_time":739.1,"predict_time":1.134,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-95","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1598.0,"score":0.445,"rank":18.4,"hrank":12.86,"imp":8.359,"train_time":1714.29,"predict_time":1.743,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-81","imputed":false},{"position":10,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1563.0,"score":0.365,"rank":20.43,"hrank":11.24,"imp":9.211,"train_time":477.59,"predict_time":0.189,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-75","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1563.0,"score":0.35,"rank":20.47,"hrank":15.78,"imp":10.072,"train_time":129.25,"predict_time":2.621,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-57","imputed":false},{"position":12,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1552.0,"score":0.351,"rank":21.13,"hrank":11.9,"imp":9.407,"train_time":477.59,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+107\/-75","imputed":false},{"position":13,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1536.0,"score":0.341,"rank":22.07,"hrank":14.6,"imp":9.628,"train_time":1176.25,"predict_time":0.556,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-84","imputed":false},{"position":14,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1533.0,"score":0.352,"rank":22.3,"hrank":15.59,"imp":10.198,"train_time":2522.59,"predict_time":0.318,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-107","imputed":false},{"position":15,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1532.0,"score":0.319,"rank":22.33,"hrank":16.47,"imp":9.808,"train_time":289.35,"predict_time":0.706,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-61","imputed":false},{"position":16,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1516.0,"score":0.334,"rank":23.33,"hrank":15.42,"imp":9.974,"train_time":2045.59,"predict_time":69.586,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-89","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1515.0,"score":0.291,"rank":23.4,"hrank":19.26,"imp":10.68,"train_time":129.25,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-74","imputed":false},{"position":18,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1515.0,"score":0.353,"rank":23.4,"hrank":12.56,"imp":9.245,"train_time":1.88,"predict_time":0.011,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-89","imputed":false},{"position":19,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1509.0,"score":0.342,"rank":23.8,"hrank":14.02,"imp":8.951,"train_time":1.91,"predict_time":0.668,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-102","imputed":false},{"position":20,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1501.0,"score":0.337,"rank":24.33,"hrank":14.13,"imp":10.256,"train_time":1714.29,"predict_time":0.079,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-110","imputed":false},{"position":21,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1500.0,"score":0.344,"rank":24.4,"hrank":15.43,"imp":11.301,"train_time":92.46,"predict_time":0.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-118","imputed":false},{"position":22,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1484.0,"score":0.281,"rank":25.47,"hrank":19.65,"imp":10.049,"train_time":289.35,"predict_time":0.08,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-84","imputed":false},{"position":23,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1480.0,"score":0.324,"rank":25.73,"hrank":18.02,"imp":11.401,"train_time":92.46,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+132\/-108","imputed":false},{"position":24,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1476.0,"score":0.29,"rank":26.0,"hrank":14.82,"imp":10.481,"train_time":1176.25,"predict_time":0.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+165\/-79","imputed":false},{"position":25,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1460.0,"score":0.306,"rank":27.07,"hrank":17.02,"imp":10.969,"train_time":0.76,"predict_time":0.178,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-141","imputed":false},{"position":26,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1453.0,"score":0.29,"rank":27.53,"hrank":8.16,"imp":10.481,"train_time":4532.19,"predict_time":387.89,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-111","imputed":false},{"position":27,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1403.0,"score":0.257,"rank":31.0,"hrank":23.28,"imp":11.377,"train_time":559.64,"predict_time":0.88,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-112","imputed":false},{"position":28,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1397.0,"score":0.245,"rank":31.43,"hrank":13.89,"imp":12.916,"train_time":4532.19,"predict_time":39.452,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-128","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1393.0,"score":0.246,"rank":31.73,"hrank":21.34,"imp":12.413,"train_time":0.48,"predict_time":0.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-123","imputed":false},{"position":30,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1379.0,"score":0.165,"rank":32.73,"hrank":21.76,"imp":11.952,"train_time":2045.59,"predict_time":6.231,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-86","imputed":false},{"position":31,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1371.0,"score":0.172,"rank":33.33,"hrank":27.25,"imp":13.54,"train_time":559.64,"predict_time":0.051,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+94\/-101","imputed":false},{"position":32,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1368.0,"score":0.213,"rank":33.53,"hrank":10.36,"imp":11.836,"train_time":0.48,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-121","imputed":false},{"position":33,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1362.0,"score":0.208,"rank":33.93,"hrank":19.47,"imp":12.151,"train_time":0.51,"predict_time":0.126,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+139\/-118","imputed":false},{"position":34,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1336.0,"score":0.133,"rank":35.8,"hrank":29.16,"imp":12.29,"train_time":1803.23,"predict_time":1.199,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-67","imputed":false},{"position":35,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1333.0,"score":0.123,"rank":36.07,"hrank":32.25,"imp":13.285,"train_time":1.94,"predict_time":0.046,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-77","imputed":false},{"position":36,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1320.0,"score":0.18,"rank":36.97,"hrank":16.43,"imp":12.809,"train_time":3.98,"predict_time":1.593,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+132\/-113","imputed":true},{"position":37,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1310.0,"score":0.172,"rank":37.73,"hrank":27.05,"imp":15.154,"train_time":662.24,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-133","imputed":false},{"position":38,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1285.0,"score":0.206,"rank":39.53,"hrank":9.67,"imp":15.204,"train_time":6.51,"predict_time":5.361,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-146","imputed":false},{"position":39,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1273.0,"score":0.104,"rank":40.4,"hrank":36.29,"imp":13.775,"train_time":9.07,"predict_time":0.224,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-85","imputed":false},{"position":40,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1271.0,"score":0.162,"rank":40.5,"hrank":23.03,"imp":15.733,"train_time":41.51,"predict_time":39.094,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+116\/-134","imputed":false},{"position":41,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1268.0,"score":0.074,"rank":40.73,"hrank":37.13,"imp":15.022,"train_time":6.54,"predict_time":0.294,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-80","imputed":false},{"position":42,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1237.0,"score":0.042,"rank":42.93,"hrank":36.82,"imp":14.895,"train_time":65.73,"predict_time":0.23,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-82","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1228.0,"score":0.063,"rank":43.6,"hrank":39.98,"imp":14.649,"train_time":1803.23,"predict_time":0.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-81","imputed":false},{"position":44,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.043,"rank":43.93,"hrank":41.31,"imp":15.006,"train_time":50.36,"predict_time":0.183,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-109","imputed":false},{"position":45,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.052,"rank":43.93,"hrank":41.77,"imp":14.26,"train_time":0.34,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-83","imputed":false},{"position":46,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1213.0,"score":0.103,"rank":44.67,"hrank":36.12,"imp":16.389,"train_time":662.24,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-134","imputed":false},{"position":47,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1213.0,"score":0.166,"rank":44.67,"hrank":25.03,"imp":18.165,"train_time":0.24,"predict_time":0.022,"imputed_pct":66.67,"hardware":"GPU","elo_ci":"+148\/-164","imputed":true},{"position":48,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1206.0,"score":0.031,"rank":45.1,"hrank":40.88,"imp":15.23,"train_time":65.73,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-74","imputed":false},{"position":49,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1198.0,"score":0.094,"rank":45.67,"hrank":40.03,"imp":17.381,"train_time":2.13,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-129","imputed":false},{"position":50,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1184.0,"score":0.074,"rank":46.7,"hrank":36.43,"imp":16.145,"train_time":66.06,"predict_time":0.61,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+106\/-98","imputed":true},{"position":51,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1181.0,"score":0.125,"rank":46.87,"hrank":31.61,"imp":18.381,"train_time":0.24,"predict_time":0.022,"imputed_pct":66.67,"hardware":"GPU","elo_ci":"+130\/-159","imputed":true},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1179.0,"score":0.041,"rank":47.0,"hrank":42.79,"imp":16.592,"train_time":50.36,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-123","imputed":false},{"position":53,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1168.0,"score":0.063,"rank":47.73,"hrank":39.57,"imp":17.18,"train_time":354.91,"predict_time":1.636,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-138","imputed":false},{"position":54,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1162.0,"score":0.041,"rank":48.13,"hrank":41.24,"imp":17.152,"train_time":11.42,"predict_time":10.384,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-120","imputed":false},{"position":55,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1159.0,"score":0.082,"rank":48.4,"hrank":41.69,"imp":18.421,"train_time":86.71,"predict_time":9.515,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-126","imputed":false},{"position":56,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1089.0,"score":0.01,"rank":52.93,"hrank":50.94,"imp":18.293,"train_time":2.58,"predict_time":0.039,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-80","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1058.0,"score":0.022,"rank":54.87,"hrank":50.35,"imp":22.822,"train_time":20.27,"predict_time":0.443,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-154","imputed":false},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1052.0,"score":0.01,"rank":55.2,"hrank":52.02,"imp":19.635,"train_time":354.91,"predict_time":0.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-106","imputed":false},{"position":59,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1005.0,"score":0.021,"rank":57.93,"hrank":50.06,"imp":26.948,"train_time":2.55,"predict_time":0.157,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+155\/-195","imputed":false},{"position":60,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":58.23,"hrank":56.85,"imp":20.772,"train_time":0.24,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-104","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":998.0,"score":0.008,"rank":58.33,"hrank":55.07,"imp":23.969,"train_time":20.27,"predict_time":0.225,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-156","imputed":false},{"position":62,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":974.0,"score":0.0,"rank":59.63,"hrank":58.6,"imp":21.557,"train_time":10.15,"predict_time":3.706,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+65\/-86","imputed":true},{"position":63,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":970.0,"score":0.011,"rank":59.87,"hrank":56.98,"imp":24.143,"train_time":150.58,"predict_time":4.952,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-124","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":962.0,"score":0.012,"rank":60.27,"hrank":56.27,"imp":23.089,"train_time":1.49,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-158","imputed":false},{"position":65,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":958.0,"score":0.0,"rank":60.47,"hrank":59.0,"imp":25.585,"train_time":5745.33,"predict_time":2.139,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-94","imputed":false},{"position":66,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":951.0,"score":0.0,"rank":60.87,"hrank":59.57,"imp":25.57,"train_time":5745.33,"predict_time":8.567,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-93","imputed":false},{"position":67,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":930.0,"score":0.0,"rank":61.87,"hrank":60.11,"imp":23.258,"train_time":0.47,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-128","imputed":false},{"position":68,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":905.0,"score":0.013,"rank":63.1,"hrank":57.79,"imp":33.239,"train_time":43.96,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+157\/-228","imputed":false},{"position":69,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":894.0,"score":0.0,"rank":63.57,"hrank":62.3,"imp":28.866,"train_time":0.27,"predict_time":0.043,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+96\/-122","imputed":true},{"position":70,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":880.0,"score":0.002,"rank":64.2,"hrank":61.78,"imp":30.681,"train_time":1.53,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-183","imputed":false},{"position":71,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":869.0,"score":0.002,"rank":64.7,"hrank":60.56,"imp":33.973,"train_time":43.96,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+154\/-236","imputed":false},{"position":72,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":830.0,"score":0.0,"rank":66.27,"hrank":65.6,"imp":30.996,"train_time":113.17,"predict_time":2.156,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+46\/-93","imputed":false},{"position":73,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":805.0,"score":0.0,"rank":67.2,"hrank":66.87,"imp":30.322,"train_time":150.58,"predict_time":0.546,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-115","imputed":false},{"position":74,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":756.0,"score":0.0,"rank":68.87,"hrank":68.02,"imp":36.011,"train_time":0.2,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-187","imputed":false},{"position":75,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":557.0,"score":0.0,"rank":73.47,"hrank":73.35,"imp":43.744,"train_time":0.1,"predict_time":0.068,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-144","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/pareto_front_explorer.html
index 2a8b08d7d70486118bfb43d0b8c0c10796852b04..37f75ccb56421106f3fde20c3e52be3be451b22f 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.6103903646,"imp":16.1449348399,"elo":1183.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.010890969,"imp":9.2446264705,"elo":1515.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0161908394,"imp":9.4068447854,"elo":1551.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1892457458,"imp":9.2108328349,"elo":1563.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0232162798,"imp":12.4133118638,"elo":1393.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.025881505,"imp":11.4009229884,"elo":1479.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"imp":11.3011884406,"elo":1499.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"imp":17.3805259993,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0061305091,"imp":16.389359794,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0928598604,"imp":15.154145312,"elo":1309.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721186209,"imp":23.2582690565,"elo":930.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0167167359,"imp":16.5918385982,"elo":1179.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1833493946,"imp":15.0064476194,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"imp":23.0885250093,"elo":962.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0941673994,"imp":19.6350538202,"elo":1052.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"imp":17.180063053,"elo":1168.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.06997296,"imp":14.2603049425,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1174373907,"imp":10.6800957475,"elo":1515.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.6207744311,"imp":10.0718129419,"elo":1562.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.3613072104,"imp":15.2036623307,"elo":1284.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"imp":13.7754313429,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3182658815,"imp":10.1975534028,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2338623423,"imp":6.9011250205,"elo":1652.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.0219328376,"imp":18.3805782296,"elo":1181.1,"imputed":true,"imputed_pct":66.6666666667,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0219328376,"imp":18.1653741011,"elo":1212.7,"imputed":true,"imputed_pct":66.6666666667,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.7064250231,"imp":21.5573717754,"elo":974.3,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"imp":30.6806116306,"elo":880.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2251402597,"imp":23.9690418598,"elo":998.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.442693885,"imp":22.8219634277,"elo":1057.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0219328376,"imp":20.7718668,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0250207663,"imp":15.2303427145,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2298050149,"imp":14.8953173136,"elo":1237.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2940400342,"imp":15.0224816412,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0785368436,"imp":10.2564835391,"elo":1501.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.7432247239,"imp":8.3592702667,"elo":1598.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6684256529,"imp":8.9508162247,"elo":1509.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1340282338,"imp":7.813683812,"elo":1650.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"imp":6.4896903028,"elo":1699.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.3844403101,"imp":17.1519949807,"elo":1162.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.0937641424,"imp":15.7325841932,"elo":1271.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"imp":12.9156193656,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":387.8903208002,"imp":10.4809941856,"elo":1453.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1777276152,"imp":10.9694457456,"elo":1459.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":13.118037366,"imp":4.457579922,"elo":1922.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0429529767,"imp":28.8663375667,"elo":894.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.5931837749,"imp":12.8092865353,"elo":1320.2,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1177498102,"imp":7.5082256688,"elo":1695.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.045727832,"imp":13.2846352058,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0566464663,"imp":10.4806951601,"elo":1475.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5563493456,"imp":9.6276945538,"elo":1536.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"imp":7.0972177851,"elo":1746.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2136890429,"imp":4.5903736993,"elo":1784.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1560825621,"imp":30.9959928616,"elo":830.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.139339258,"imp":25.5852650516,"elo":958.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"imp":25.570357389,"elo":950.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1258468951,"imp":12.1514539075,"elo":1362.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0390374572,"imp":18.2928174392,"elo":1089.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0507240116,"imp":14.6486117194,"elo":1227.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1994296217,"imp":12.2903298919,"elo":1336.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"imp":11.8360684126,"elo":1367.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0795849342,"imp":10.0490986518,"elo":1483.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.7060451863,"imp":9.8078613915,"elo":1532.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":9.5146310091,"imp":18.4211458676,"elo":1158.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":6.2309459511,"imp":11.9518940132,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":69.5857864072,"imp":9.9738007966,"elo":1516.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"imp":26.9481191606,"elo":1005.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0505891669,"imp":13.5401190315,"elo":1370.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.8800015259,"imp":11.3769395133,"elo":1403.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.6103903646,"x_train":66.0588444215,"imp":16.1449348399,"elo":1183.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.010890969,"x_train":1.8802002354,"imp":9.2446264705,"elo":1515.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0161908394,"x_train":477.5939557065,"imp":9.4068447854,"elo":1551.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1892457458,"x_train":477.5939557065,"imp":9.2108328349,"elo":1563.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0232162798,"x_train":0.4836414313,"imp":12.4133118638,"elo":1393.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.025881505,"x_train":92.4632546012,"imp":11.4009229884,"elo":1479.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":92.4632546012,"imp":11.3011884406,"elo":1499.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"x_train":2.1343897895,"imp":17.3805259993,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0061305091,"x_train":662.2364442662,"imp":16.389359794,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0928598604,"x_train":662.2364442662,"imp":15.154145312,"elo":1309.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721186209,"x_train":0.4655147195,"imp":23.2582690565,"elo":930.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0167167359,"x_train":50.361853205,"imp":16.5918385982,"elo":1179.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1833493946,"x_train":50.361853205,"imp":15.0064476194,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"x_train":1.4936466441,"imp":23.0885250093,"elo":962.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0941673994,"x_train":354.9067354826,"imp":19.6350538202,"elo":1052.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"x_train":354.9067354826,"imp":17.180063053,"elo":1168.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.06997296,"x_train":0.3380289469,"imp":14.2603049425,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1174373907,"x_train":129.246647918,"imp":10.6800957475,"elo":1515.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.6207744311,"x_train":129.246647918,"imp":10.0718129419,"elo":1562.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.3613072104,"x_train":6.5096502566,"imp":15.2036623307,"elo":1284.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"x_train":9.0683107381,"imp":13.7754313429,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3182658815,"x_train":2522.590218997,"imp":10.1975534028,"elo":1532.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2338623423,"x_train":2522.590218997,"imp":6.9011250205,"elo":1652.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.0219328376,"x_train":0.239812243,"imp":18.3805782296,"elo":1181.1,"imputed":true,"imputed_pct":66.6666666667,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0219328376,"x_train":0.239812243,"imp":18.1653741011,"elo":1212.7,"imputed":true,"imputed_pct":66.6666666667,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.7064250231,"x_train":10.1475376964,"imp":21.5573717754,"elo":974.3,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"x_train":1.5311965098,"imp":30.6806116306,"elo":880.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2251402597,"x_train":20.2694783344,"imp":23.9690418598,"elo":998.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.442693885,"x_train":20.2694783344,"imp":22.8219634277,"elo":1057.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0219328376,"x_train":0.239812243,"imp":20.7718668,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0250207663,"x_train":65.72538338,"imp":15.2303427145,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2298050149,"x_train":65.72538338,"imp":14.8953173136,"elo":1237.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2940400342,"x_train":6.5430745829,"imp":15.0224816412,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0785368436,"x_train":1714.2893053174,"imp":10.2564835391,"elo":1501.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.7432247239,"x_train":1714.2893053174,"imp":8.3592702667,"elo":1598.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6684256529,"x_train":1.9087879061,"imp":8.9508162247,"elo":1509.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1340282338,"x_train":739.1047147274,"imp":7.813683812,"elo":1650.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"x_train":739.1047147274,"imp":6.4896903028,"elo":1699.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.3844403101,"x_train":11.4229329799,"imp":17.1519949807,"elo":1162.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.0937641424,"x_train":41.5053698771,"imp":15.7325841932,"elo":1271.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"x_train":4532.1904663458,"imp":12.9156193656,"elo":1397.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":387.8903208002,"x_train":4532.1904663458,"imp":10.4809941856,"elo":1453.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1777276152,"x_train":0.755625227,"imp":10.9694457456,"elo":1459.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":13.118037366,"x_train":27.1820477768,"imp":4.457579922,"elo":1922.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0429529767,"x_train":0.2749370239,"imp":28.8663375667,"elo":894.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.5931837749,"x_train":3.9759380126,"imp":12.8092865353,"elo":1320.2,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1177498102,"x_train":0.5630642002,"imp":7.5082256688,"elo":1695.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.045727832,"x_train":1.9359607519,"imp":13.2846352058,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0566464663,"x_train":1176.2542230269,"imp":10.4806951601,"elo":1475.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5563493456,"x_train":1176.2542230269,"imp":9.6276945538,"elo":1536.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"x_train":2.6386678711,"imp":7.0972177851,"elo":1746.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2136890429,"x_train":1.0196284569,"imp":4.5903736993,"elo":1784.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1560825621,"x_train":113.1657286335,"imp":30.9959928616,"elo":830.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.139339258,"x_train":5745.3315346214,"imp":25.5852650516,"elo":958.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"x_train":5745.3315346214,"imp":25.570357389,"elo":950.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1258468951,"x_train":0.5121427198,"imp":12.1514539075,"elo":1362.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0390374572,"x_train":2.5795466495,"imp":18.2928174392,"elo":1089.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0507240116,"x_train":1803.2313354459,"imp":14.6486117194,"elo":1227.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1994296217,"x_train":1803.2313354459,"imp":12.2903298919,"elo":1336.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"x_train":0.4833722601,"imp":11.8360684126,"elo":1367.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0795849342,"x_train":289.3497533723,"imp":10.0490986518,"elo":1483.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.7060451863,"x_train":289.3497533723,"imp":9.8078613915,"elo":1532.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":9.5146310091,"x_train":86.7109854477,"imp":18.4211458676,"elo":1158.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":6.2309459511,"x_train":2045.5905262122,"imp":11.9518940132,"elo":1379.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":69.5857864072,"x_train":2045.5905262122,"imp":9.9738007966,"elo":1516.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"x_train":2.5457229657,"imp":26.9481191606,"elo":1005.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0505891669,"x_train":559.6449092182,"imp":13.5401190315,"elo":1370.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.8800015259,"x_train":559.6449092182,"imp":11.3769395133,"elo":1403.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/tuning_trajectories_explorer.html
index a495b78e4ddeae26d800c5a84472f0e44d1b988c..68754ed86bacd7440cb1189b834138317ba73fe6 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_all/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":66.0588444215,"imp":16.2219098886,"elo":1204.5,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.8802002354,"imp":9.3101786703,"elo":1542.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":4.3390813562,"imp":9.3668455506,"elo":1548.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.9107908196,"imp":9.267595022,"elo":1576.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":24.8738378919,"imp":9.2402319461,"elo":1583.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":59.791099159,"imp":9.2572564585,"elo":1597.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":118.7049807087,"imp":9.2515510859,"elo":1600.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":236.2186933654,"imp":9.2783419621,"elo":1597.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":351.7154510353,"imp":9.3010200485,"elo":1596.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":477.5939557065,"imp":9.2664909215,"elo":1596.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4836414313,"imp":12.4888788997,"elo":1417.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.0174591425,"imp":12.1684586985,"elo":1456.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.3395204881,"imp":11.9575651564,"elo":1476.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.5724929214,"imp":11.6909990873,"elo":1492.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.475979466,"imp":11.5322907251,"elo":1506.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.9675728072,"imp":11.3632338481,"elo":1519.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":46.0351331804,"imp":11.2982349858,"elo":1530.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":69.0484132695,"imp":11.3142188082,"elo":1535.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":92.4632546012,"imp":11.3673653853,"elo":1540.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.1343897895,"imp":17.4415958555,"elo":1212.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":6.5030474549,"imp":16.5576069156,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":15.551340727,"imp":16.0469153438,"elo":1275.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":32.7988382693,"imp":15.8315388028,"elo":1288.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":82.4434921155,"imp":15.5556478456,"elo":1309.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":165.9098913248,"imp":15.449401335,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":325.9150394035,"imp":15.3061276456,"elo":1329.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":495.2952240262,"imp":15.261890739,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":662.2364442662,"imp":15.2135933208,"elo":1334.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.4655147195,"imp":23.3078515058,"elo":926.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.8039431531,"imp":21.0918921598,"elo":996.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.5507866238,"imp":17.5283944157,"elo":1102.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9159741389,"imp":16.0430645763,"elo":1157.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.8290142883,"imp":15.3266460542,"elo":1187.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.9805618038,"imp":14.9844657736,"elo":1211.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":25.3410754681,"imp":14.6296352436,"elo":1235.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.7076806608,"imp":14.6000430678,"elo":1237.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":50.361853205,"imp":15.0595931989,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.4936466441,"imp":23.1338588854,"elo":969.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.1783708385,"imp":21.9290550691,"elo":1031.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.5233821255,"imp":20.266979712,"elo":1099.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":17.7581257529,"imp":19.5010284137,"elo":1124.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":42.7590536162,"imp":18.5238258057,"elo":1147.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":86.3142265197,"imp":18.0598399025,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":175.409142407,"imp":17.551164854,"elo":1172.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":264.0183986663,"imp":17.3246074232,"elo":1187.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":354.9067354826,"imp":17.2365017671,"elo":1189.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.3380289469,"imp":14.3277897919,"elo":1242.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.9616222835,"imp":13.6117994537,"elo":1303.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":2.6147261705,"imp":12.4484880806,"elo":1415.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":6.009789211,"imp":11.6154619585,"elo":1483.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":16.2074876554,"imp":10.6946997732,"elo":1547.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":31.9858466992,"imp":10.3197118265,"elo":1579.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":64.004281301,"imp":10.1824969771,"elo":1592.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":96.5406969888,"imp":10.1481925679,"elo":1598.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":129.246647918,"imp":10.144476849,"elo":1602.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":6.5096502566,"imp":15.2673017486,"elo":1308.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":9.0683107381,"imp":13.8345397258,"elo":1287.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":22.7816772308,"imp":12.3574421813,"elo":1376.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.9509960123,"imp":10.5279683206,"elo":1494.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":121.2878502068,"imp":10.0115247351,"elo":1533.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":291.6044227147,"imp":9.4501715832,"elo":1588.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":628.8473422899,"imp":8.7744306685,"elo":1619.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1243.1111273612,"imp":8.1696685386,"elo":1651.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1857.216089218,"imp":7.4629553504,"elo":1677.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2522.590218997,"imp":6.9899541425,"elo":1693.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":0.239812243,"imp":18.4350448023,"elo":1188.7,"imputed":true,"imputed_pct":66.6666666667,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":0.239812243,"imp":18.2198399527,"elo":1224.3,"imputed":true,"imputed_pct":66.6666666667,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_train":10.1475376964,"imp":21.2101781908,"elo":995.6,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.5311965098,"imp":30.6847464252,"elo":900.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.1703965898,"imp":27.5932505743,"elo":981.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":17.0551085935,"imp":23.1659318506,"elo":1057.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.2694783344,"imp":22.8406144223,"elo":1068.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.239812243,"imp":20.8316944461,"elo":1004.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4901066084,"imp":17.3008310184,"elo":1117.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.4763660732,"imp":15.9727945524,"elo":1174.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":3.0456620821,"imp":15.4757041149,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":7.9306942372,"imp":15.1219825125,"elo":1239.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":16.1953879054,"imp":15.0062206669,"elo":1244.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":32.6380042897,"imp":14.9252458929,"elo":1251.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":48.9781692218,"imp":14.9213810044,"elo":1252.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":65.72538338,"imp":14.9488517824,"elo":1256.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.5430745829,"imp":15.0709468125,"elo":1276.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.4482837345,"imp":12.7764398981,"elo":1401.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.4902474287,"imp":11.0401568285,"elo":1501.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":81.1768457292,"imp":10.0828756939,"elo":1538.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":211.6867497619,"imp":9.3638496992,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":425.9393992898,"imp":8.9914600467,"elo":1601.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":844.9984797236,"imp":8.6564499876,"elo":1612.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1273.8908426812,"imp":8.5179970894,"elo":1629.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1714.2893053174,"imp":8.4329287591,"elo":1638.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.9087879061,"imp":9.0318979829,"elo":1521.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.1930136195,"imp":8.9282202861,"elo":1538.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.3030487246,"imp":8.6970393181,"elo":1559.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":34.3657315438,"imp":8.4975753927,"elo":1585.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":89.1338776753,"imp":8.0943433787,"elo":1647.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":182.7153391231,"imp":7.6911791455,"elo":1669.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":366.5547593939,"imp":7.2073441384,"elo":1700.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":550.802457324,"imp":6.9240564882,"elo":1714.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":739.1047147274,"imp":6.5642947033,"elo":1728.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":11.4229329799,"imp":17.2052766922,"elo":1177.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.5053698771,"imp":15.77888031,"elo":1287.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.019350737,"imp":14.4694915963,"elo":1326.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":119.7137023169,"imp":13.3885047556,"elo":1376.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":214.9883934306,"imp":12.9181225664,"elo":1397.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":547.8027265576,"imp":11.9376853893,"elo":1433.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1117.1957402472,"imp":11.324402862,"elo":1455.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2246.8238340239,"imp":10.9679290078,"elo":1472.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3382.9355985564,"imp":10.6711842762,"elo":1472.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4532.1904663458,"imp":10.5200659712,"elo":1477.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.755625227,"imp":11.0261391187,"elo":1486.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":27.1820477768,"imp":4.5302588187,"elo":1951.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.2749370239,"imp":28.8993751931,"elo":906.4,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":3.9759380126,"imp":12.8818263097,"elo":1347.3,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.5630642002,"imp":7.5867160473,"elo":1721.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.9359607519,"imp":13.348608439,"elo":1354.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.4292557844,"imp":12.665274073,"elo":1404.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":25.9857811464,"imp":11.7175055653,"elo":1455.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":55.2138352308,"imp":10.9937623926,"elo":1499.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":144.8913305375,"imp":10.4219450187,"elo":1526.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":291.446310356,"imp":10.1316770987,"elo":1539.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":569.7562297642,"imp":9.8613946529,"elo":1553.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":866.467617543,"imp":9.7713848819,"elo":1565.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1176.2542230269,"imp":9.7068285077,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.6386678711,"imp":7.170909029,"elo":1785.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.0196284569,"imp":4.53762243,"elo":1787.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":113.1657286335,"imp":31.0261692308,"elo":826.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":141.3356612104,"imp":30.904567666,"elo":843.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":220.2044612008,"imp":30.5734501802,"elo":851.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":327.1108834516,"imp":29.8700168851,"elo":869.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":732.5344880833,"imp":28.5319891784,"elo":889.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1475.7671634624,"imp":27.2140949709,"elo":918.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2923.1088462252,"imp":26.660294051,"elo":933.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4314.2921057007,"imp":26.0667075541,"elo":954.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5745.3315346214,"imp":25.6051802421,"elo":978.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.5121427198,"imp":12.2299705968,"elo":1389.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":2.5795466495,"imp":18.3372127798,"elo":1102.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.5303184982,"imp":16.8976741311,"elo":1157.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":42.5186103306,"imp":15.3546222457,"elo":1216.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":94.7530519512,"imp":14.4400149526,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":222.7613331852,"imp":13.4934825661,"elo":1295.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":460.6794919634,"imp":13.0332265716,"elo":1320.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":882.9304835882,"imp":12.6771146149,"elo":1347.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1357.9255836308,"imp":12.4785421595,"elo":1359.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1803.2313354459,"imp":12.3521741362,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4833722601,"imp":11.9078835843,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.4574329869,"imp":11.1335409236,"elo":1460.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":7.3610615249,"imp":10.6634417872,"elo":1505.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":13.6386574902,"imp":10.3447754875,"elo":1528.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":35.9388283515,"imp":10.069810101,"elo":1545.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":71.5452674162,"imp":9.9458136244,"elo":1558.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":142.0559469286,"imp":9.8960743136,"elo":1566.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":215.0273705137,"imp":9.8849729562,"elo":1569.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":289.3497533723,"imp":9.9010563696,"elo":1569.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":86.7109854477,"imp":18.473075499,"elo":1181.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":165.5352279836,"imp":15.9866627301,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":405.7546652544,"imp":17.7353404452,"elo":1339.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":797.4916792797,"imp":15.6812231729,"elo":1427.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1969.6298609549,"imp":9.9324356592,"elo":1550.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":2045.5905262122,"imp":10.0290284619,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.5457229657,"imp":26.9867576335,"elo":1020.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.2856913484,"imp":21.7555602931,"elo":1106.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":13.6127137293,"imp":15.9862923705,"elo":1244.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":27.6956293748,"imp":13.5693447461,"elo":1347.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":69.488323154,"imp":12.6572114896,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":138.8404994793,"imp":12.1949160482,"elo":1408.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":278.0036441284,"imp":11.8122767858,"elo":1416.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":417.6192129824,"imp":11.5271614578,"elo":1422.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":559.6449092182,"imp":11.2085821154,"elo":1433.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.6103903646,"x_train":66.0588444215,"imp":16.2219098886,"elo":1204.5,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.010890969,"x_train":1.8802002354,"imp":9.3101786703,"elo":1542.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0416544912,"x_train":4.3390813562,"imp":9.3668455506,"elo":1548.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0663288817,"x_train":12.9107908196,"imp":9.267595022,"elo":1576.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0905500613,"x_train":24.8738378919,"imp":9.2402319461,"elo":1583.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1417099551,"x_train":59.791099159,"imp":9.2572564585,"elo":1597.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1656029978,"x_train":118.7049807087,"imp":9.2515510859,"elo":1600.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1778183572,"x_train":236.2186933654,"imp":9.2783419621,"elo":1597.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1723703847,"x_train":351.7154510353,"imp":9.3010200485,"elo":1596.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1892457458,"x_train":477.5939557065,"imp":9.2664909215,"elo":1596.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0232162798,"x_train":0.4836414313,"imp":12.4888788997,"elo":1417.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0460251725,"x_train":1.0174591425,"imp":12.1684586985,"elo":1456.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1045606687,"x_train":2.3395204881,"imp":11.9575651564,"elo":1476.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1437162313,"x_train":4.5724929214,"imp":11.6909990873,"elo":1492.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2215730015,"x_train":11.475979466,"imp":11.5322907251,"elo":1506.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2317567073,"x_train":22.9675728072,"imp":11.3632338481,"elo":1519.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2465009094,"x_train":46.0351331804,"imp":11.2982349858,"elo":1530.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2772856864,"x_train":69.0484132695,"imp":11.3142188082,"elo":1535.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":92.4632546012,"imp":11.3673653853,"elo":1540.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"x_train":2.1343897895,"imp":17.4415958555,"elo":1212.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0116188579,"x_train":6.5030474549,"imp":16.5576069156,"elo":1252.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0227492458,"x_train":15.551340727,"imp":16.0469153438,"elo":1275.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333486726,"x_train":32.7988382693,"imp":15.8315388028,"elo":1288.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0515992981,"x_train":82.4434921155,"imp":15.5556478456,"elo":1309.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0585840954,"x_train":165.9098913248,"imp":15.449401335,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0740425479,"x_train":325.9150394035,"imp":15.3061276456,"elo":1329.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0820593882,"x_train":495.2952240262,"imp":15.261890739,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0923119309,"x_train":662.2364442662,"imp":15.2135933208,"elo":1334.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721186209,"x_train":0.4655147195,"imp":23.3078515058,"elo":926.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.079982782,"x_train":0.8039431531,"imp":21.0918921598,"elo":996.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.084787187,"x_train":1.5507866238,"imp":17.5283944157,"elo":1102.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.078755296,"x_train":2.9159741389,"imp":16.0430645763,"elo":1157.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1006473264,"x_train":6.8290142883,"imp":15.3266460542,"elo":1187.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1257749856,"x_train":12.9805618038,"imp":14.9844657736,"elo":1211.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1512704682,"x_train":25.3410754681,"imp":14.6296352436,"elo":1235.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1728496468,"x_train":37.7076806608,"imp":14.6000430678,"elo":1237.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739987392,"x_train":50.361853205,"imp":15.0595931989,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"x_train":1.4936466441,"imp":23.1338588854,"elo":969.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2247585805,"x_train":3.1783708385,"imp":21.9290550691,"elo":1031.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4166305562,"x_train":8.5233821255,"imp":20.266979712,"elo":1099.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6283891459,"x_train":17.7581257529,"imp":19.5010284137,"elo":1124.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9164506571,"x_train":42.7590536162,"imp":18.5238258057,"elo":1147.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2468060948,"x_train":86.3142265197,"imp":18.0598399025,"elo":1162.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.6512385514,"x_train":175.409142407,"imp":17.551164854,"elo":1172.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7760120815,"x_train":264.0183986663,"imp":17.3246074232,"elo":1187.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"x_train":354.9067354826,"imp":17.2365017671,"elo":1189.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.06997296,"x_train":0.3380289469,"imp":14.3277897919,"elo":1242.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4182187449,"x_train":0.9616222835,"imp":13.6117994537,"elo":1303.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7644003426,"x_train":2.6147261705,"imp":12.4484880806,"elo":1415.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0019478197,"x_train":6.009789211,"imp":11.6154619585,"elo":1483.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1773854817,"x_train":16.2074876554,"imp":10.6946997732,"elo":1547.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4649015268,"x_train":31.9858466992,"imp":10.3197118265,"elo":1579.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1045193014,"x_train":64.004281301,"imp":10.1824969771,"elo":1592.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1744811378,"x_train":96.5406969888,"imp":10.1481925679,"elo":1598.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.6207744311,"x_train":129.246647918,"imp":10.144476849,"elo":1602.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":5.3613072104,"x_train":6.5096502566,"imp":15.2673017486,"elo":1308.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"x_train":9.0683107381,"imp":13.8345397258,"elo":1287.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5307544296,"x_train":22.7816772308,"imp":12.3574421813,"elo":1376.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.3312012868,"x_train":60.9509960123,"imp":10.5279683206,"elo":1494.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1424478891,"x_train":121.2878502068,"imp":10.0115247351,"elo":1533.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.7588304305,"x_train":291.6044227147,"imp":9.4501715832,"elo":1588.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.9692086024,"x_train":628.8473422899,"imp":8.7744306685,"elo":1619.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.3479391024,"x_train":1243.1111273612,"imp":8.1696685386,"elo":1651.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2360012984,"x_train":1857.216089218,"imp":7.4629553504,"elo":1677.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2338623423,"x_train":2522.590218997,"imp":6.9899541425,"elo":1693.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.0219328376,"x_train":0.239812243,"imp":18.4350448023,"elo":1188.7,"imputed":true,"imputed_pct":66.6666666667,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.0219328376,"x_train":0.239812243,"imp":18.2198399527,"elo":1224.3,"imputed":true,"imputed_pct":66.6666666667,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.7064250231,"x_train":10.1475376964,"imp":21.2101781908,"elo":995.6,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"x_train":1.5311965098,"imp":30.6847464252,"elo":900.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1165097969,"x_train":6.1703965898,"imp":27.5932505743,"elo":981.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3758127641,"x_train":17.0551085935,"imp":23.1659318506,"elo":1057.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.442693885,"x_train":20.2694783344,"imp":22.8406144223,"elo":1068.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0219328376,"x_train":0.239812243,"imp":20.8316944461,"elo":1004.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0440982753,"x_train":0.4901066084,"imp":17.3008310184,"elo":1117.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1060575328,"x_train":1.4763660732,"imp":15.9727945524,"elo":1174.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1361499657,"x_train":3.0456620821,"imp":15.4757041149,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2062124655,"x_train":7.9306942372,"imp":15.1219825125,"elo":1239.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2414176328,"x_train":16.1953879054,"imp":15.0062206669,"elo":1244.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2481870329,"x_train":32.6380042897,"imp":14.9252458929,"elo":1251.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2529555406,"x_train":48.9781692218,"imp":14.9213810044,"elo":1252.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2298050149,"x_train":65.72538338,"imp":14.9488517824,"elo":1256.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2940400342,"x_train":6.5430745829,"imp":15.0709468125,"elo":1276.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4024728498,"x_train":15.4482837345,"imp":12.7764398981,"elo":1401.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.6227888714,"x_train":39.4902474287,"imp":11.0401568285,"elo":1501.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.8897271358,"x_train":81.1768457292,"imp":10.0828756939,"elo":1538.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.2749551545,"x_train":211.6867497619,"imp":9.3638496992,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4994174796,"x_train":425.9393992898,"imp":8.9914600467,"elo":1601.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7662329972,"x_train":844.9984797236,"imp":8.6564499876,"elo":1612.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5220276492,"x_train":1273.8908426812,"imp":8.5179970894,"elo":1629.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7432247239,"x_train":1714.2893053174,"imp":8.4329287591,"elo":1638.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6684256529,"x_train":1.9087879061,"imp":9.0318979829,"elo":1521.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4990238167,"x_train":5.1930136195,"imp":8.9282202861,"elo":1538.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.3467414164,"x_train":16.3030487246,"imp":8.6970393181,"elo":1559.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.8781041597,"x_train":34.3657315438,"imp":8.4975753927,"elo":1585.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8044658046,"x_train":89.1338776753,"imp":8.0943433787,"elo":1647.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.8326114376,"x_train":182.7153391231,"imp":7.6911791455,"elo":1669.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.1219292846,"x_train":366.5547593939,"imp":7.2073441384,"elo":1700.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.4798418546,"x_train":550.802457324,"imp":6.9240564882,"elo":1714.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"x_train":739.1047147274,"imp":6.5642947033,"elo":1728.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":10.3844403101,"x_train":11.4229329799,"imp":17.2052766922,"elo":1177.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.0937641424,"x_train":41.5053698771,"imp":15.77888031,"elo":1287.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":55.1055828632,"x_train":63.019350737,"imp":14.4694915963,"elo":1326.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":84.5945717618,"x_train":119.7137023169,"imp":13.3885047556,"elo":1376.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":115.9268466871,"x_train":214.9883934306,"imp":12.9181225664,"elo":1397.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":182.0790373328,"x_train":547.8027265576,"imp":11.9376853893,"elo":1433.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":248.1191218836,"x_train":1117.1957402472,"imp":11.324402862,"elo":1455.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":319.6845628759,"x_train":2246.8238340239,"imp":10.9679290078,"elo":1472.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":359.660599797,"x_train":3382.9355985564,"imp":10.6711842762,"elo":1472.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":390.2528105033,"x_train":4532.1904663458,"imp":10.5200659712,"elo":1477.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1777276152,"x_train":0.755625227,"imp":11.0261391187,"elo":1486.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":13.118037366,"x_train":27.1820477768,"imp":4.5302588187,"elo":1951.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0429529767,"x_train":0.2749370239,"imp":28.8993751931,"elo":906.4,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.5931837749,"x_train":3.9759380126,"imp":12.8818263097,"elo":1347.3,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1177498102,"x_train":0.5630642002,"imp":7.5867160473,"elo":1721.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.045727832,"x_train":1.9359607519,"imp":13.348608439,"elo":1354.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1008784954,"x_train":6.4292557844,"imp":12.665274073,"elo":1404.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.1981532601,"x_train":25.9857811464,"imp":11.7175055653,"elo":1455.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2598475914,"x_train":55.2138352308,"imp":10.9937623926,"elo":1499.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3596014701,"x_train":144.8913305375,"imp":10.4219450187,"elo":1526.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4349645047,"x_train":291.446310356,"imp":10.1316770987,"elo":1539.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4765747411,"x_train":569.7562297642,"imp":9.8613946529,"elo":1553.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.4821521088,"x_train":866.467617543,"imp":9.7713848819,"elo":1565.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5419846432,"x_train":1176.2542230269,"imp":9.7068285077,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"x_train":2.6386678711,"imp":7.170909029,"elo":1785.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2136890429,"x_train":1.0196284569,"imp":4.53762243,"elo":1787.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1560825621,"x_train":113.1657286335,"imp":31.0261692308,"elo":826.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.2636388537,"x_train":141.3356612104,"imp":30.904567666,"elo":843.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.3912942358,"x_train":220.2044612008,"imp":30.5734501802,"elo":851.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.8322499442,"x_train":327.1108834516,"imp":29.8700168851,"elo":869.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.9995876557,"x_train":732.5344880833,"imp":28.5319891784,"elo":889.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5174296389,"x_train":1475.7671634624,"imp":27.2140949709,"elo":918.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.8892404898,"x_train":2923.1088462252,"imp":26.660294051,"elo":933.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.0618917211,"x_train":4314.2921057007,"imp":26.0667075541,"elo":954.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"x_train":5745.3315346214,"imp":25.6051802421,"elo":978.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1258468951,"x_train":0.5121427198,"imp":12.2299705968,"elo":1389.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0390374572,"x_train":2.5795466495,"imp":18.3372127798,"elo":1102.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0911383405,"x_train":9.5303184982,"imp":16.8976741311,"elo":1157.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2406739756,"x_train":42.5186103306,"imp":15.3546222457,"elo":1216.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3974574041,"x_train":94.7530519512,"imp":14.4400149526,"elo":1261.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5700088429,"x_train":222.7613331852,"imp":13.4934825661,"elo":1295.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7281380887,"x_train":460.6794919634,"imp":13.0332265716,"elo":1320.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9742904808,"x_train":882.9304835882,"imp":12.6771146149,"elo":1347.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9743613689,"x_train":1357.9255836308,"imp":12.4785421595,"elo":1359.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2024249833,"x_train":1803.2313354459,"imp":12.3521741362,"elo":1368.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"x_train":0.4833722601,"imp":11.9078835843,"elo":1394.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.176191836,"x_train":2.4574329869,"imp":11.1335409236,"elo":1460.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2515770525,"x_train":7.3610615249,"imp":10.6634417872,"elo":1505.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3262966161,"x_train":13.6386574902,"imp":10.3447754875,"elo":1528.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3865979705,"x_train":35.9388283515,"imp":10.069810101,"elo":1545.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4060971572,"x_train":71.5452674162,"imp":9.9458136244,"elo":1558.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4323510597,"x_train":142.0559469286,"imp":9.8960743136,"elo":1566.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4840201507,"x_train":215.0273705137,"imp":9.8849729562,"elo":1569.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7060451863,"x_train":289.3497533723,"imp":9.9010563696,"elo":1569.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":9.5146310091,"x_train":86.7109854477,"imp":18.473075499,"elo":1181.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":18.8009106596,"x_train":165.5352279836,"imp":15.9866627301,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":28.6343734044,"x_train":405.7546652544,"imp":17.7353404452,"elo":1339.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.779029417,"x_train":797.4916792797,"imp":15.6812231729,"elo":1427.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":67.7213907223,"x_train":1969.6298609549,"imp":9.9324356592,"elo":1550.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":69.5857864072,"x_train":2045.5905262122,"imp":10.0290284619,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"x_train":2.5457229657,"imp":26.9867576335,"elo":1020.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.2277922302,"x_train":5.2856913484,"imp":21.7555602931,"elo":1106.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.2497110384,"x_train":13.6127137293,"imp":15.9862923705,"elo":1244.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.3680809292,"x_train":27.6956293748,"imp":13.5693447461,"elo":1347.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.5701916292,"x_train":69.488323154,"imp":12.6572114896,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.6682251127,"x_train":138.8404994793,"imp":12.1949160482,"elo":1408.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.7937638305,"x_train":278.0036441284,"imp":11.8122767858,"elo":1416.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":0.8299964002,"x_train":417.6192129824,"imp":11.5271614578,"elo":1422.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":0.8800015259,"x_train":559.6449092182,"imp":11.2085821154,"elo":1433.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..8b1abb65b0ab725884aba72e3df58fb07074fc4c
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_all/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "AutoGluon 1.4 (best, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabICLv2 (default)", "ModernNCA (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "CatBoost (tuned + ensembled)", "LightGBM (tuned + ensembled)", "CatBoost (tuned)", "TabM (tuned + ensembled)", "ModernNCA (tuned)", "XGBoost (tuned + ensembled)", "iLTM (tuned + ensembled)", "LightGBM (tuned)", "CatBoost (default)", "RealTabPFN-2.5 (default)", "RealMLP (tuned)", "ChimeraBoost (tuned + ensembled)", "XGBoost (tuned)", "ChimeraBoost (tuned)", "TabM (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "xRFM (tuned + ensembled)", "TabDPT (tuned)", "ChimeraBoost (default)", "iLTM (tuned)", "xRFM (tuned)", "XGBoost (default)", "TabSwift (default)", "TorchMLP (tuned + ensembled)", "TabM (default)", "TabICL (default)", "EBM (tuned + ensembled)", "LimiX (default)", "ModernNCA (default)", "TabDPT (default)", "RealMLP (default)", "RandomForest (tuned + ensembled)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "LightGBM (default)", "Nori-30M (default)", "EBM (tuned)", "RandomForest (tuned)", "EBM (default)", "BetaTabPFN (default)", "Nori (default)", "ExtraTrees (tuned)", "FastaiMLP (tuned + ensembled)", "SAP-RPT-OSS (default)", "iLTM (default)", "TorchMLP (default)", "PerpetualBooster (tuned + ensembled)", "FastaiMLP (tuned)", "xRFM (default)", "RandomForest (default)", "PerpetualBooster (tuned)", "OrionMSP (default)", "KNN (tuned + ensembled)", "FastaiMLP (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "ExtraTrees (default)", "Linear (tuned + ensembled)", "TabFlex (default)", "PerpetualBooster (default)", "Linear (tuned)", "TabSTAR (default)", "KNN (tuned)", "Linear (default)", "KNN (default)"], "matrix": [[NaN, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 1.0, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 1.0, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4000000059604645, NaN, 0.6666666865348816, 0.4000000059604645, 0.6666666865348816, 0.800000011920929, 0.5333333611488342, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.7333333492279053, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4000000059604645, 0.3333333432674408, NaN, 0.4000000059604645, 0.7333333492279053, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 1.0, 0.9333333373069763, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.6666666865348816, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2666666805744171, 0.6000000238418579, 0.6000000238418579, NaN, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 1.0, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, NaN, 0.46666666865348816, 0.5333333611488342, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.20000000298023224, 0.46666666865348816, 0.3333333432674408, 0.5333333611488342, NaN, 0.4000000059604645, 0.5333333611488342, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.46666666865348816, 0.46666666865348816, 0.2666666805744171, 0.46666666865348816, 0.6000000238418579, NaN, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 1.0, 0.8666666746139526, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.20000000298023224, 0.4000000059604645, 0.13333334028720856, 0.2666666805744171, 0.46666666865348816, 0.2666666805744171, NaN, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 1.0, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.46666666865348816, 0.20000000298023224, 0.2666666805744171, 0.46666666865348816, NaN, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, NaN, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.0, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, NaN, 0.4000000059604645, 0.6666666865348816, 0.4000000059604645, 0.4333333373069763, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.6000000238418579, NaN, 0.6666666865348816, 0.3333333432674408, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.6000000238418579, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 0.7333333492279053, 1.0, 1.0, 1.0, 0.8666666746139526, 0.800000011920929, 1.0, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.0, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, NaN, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, NaN, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 1.0, 0.7333333492279053, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 1.0, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.6666666865348816, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 1.0, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.46666666865348816, 0.5666666626930237, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, NaN, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.0, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.7333333492279053, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.5333333611488342, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 0.6666666865348816, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, NaN, 0.4000000059604645, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 0.8666666746139526, 0.6666666865348816, 0.6666666865348816, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.6666666865348816, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.4000000059604645, 0.5333333611488342, 0.2666666805744171, 0.5333333611488342, 0.3333333432674408, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, NaN, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.46666666865348816, 0.6666666865348816, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 1.0, 0.7333333492279053, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, NaN, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, NaN, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 1.0, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 1.0, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0], [0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, NaN, 0.3333333432674408, 0.6000000238418579, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.5333333611488342, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, NaN, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.20000000298023224, 0.0, 0.0, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, NaN, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.8666666746139526, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.6666666865348816, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.13333334028720856, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.2666666805744171, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.4000000059604645, 0.5333333611488342, NaN, 0.46666666865348816, 0.46666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, NaN, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 1.0, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.5333333611488342, 0.3333333432674408, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, NaN, 0.3333333432674408, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.6666666865348816, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.800000011920929, 0.9333333373069763, 1.0, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5333333611488342, 0.6666666865348816, NaN, 0.6666666865348816, 0.800000011920929, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.800000011920929, 0.6666666865348816, 0.4000000059604645, 0.5333333611488342, 0.6666666865348816, 0.8666666746139526, 0.9333333373069763, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.06666667014360428, 0.0, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.06666667014360428, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.5333333611488342, 0.4000000059604645, 0.3333333432674408, NaN, 0.4000000059604645, 0.3333333432674408, 0.5333333611488342, 0.6666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.20000000298023224, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.20000000298023224, 0.6000000238418579, NaN, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.8999999761581421, 0.5333333611488342, 0.6000000238418579, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 1.0, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.800000011920929, 0.9333333373069763, 1.0, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.46666666865348816, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, NaN, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.46666666865348816, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0], [0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, NaN, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.0, 0.0, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.46666666865348816, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.46666666865348816, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5333333611488342, NaN, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.6666666865348816, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.4000000059604645, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.6000000238418579, 0.5333333611488342, NaN, 0.46666666865348816, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.5333333611488342, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.5333333611488342, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 0.9333333373069763, 1.0], [0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.13333334028720856, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.3333333432674408, 0.13333334028720856, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, NaN, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0], [0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.13333334028720856, 0.3333333432674408, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, NaN, 0.4000000059604645, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 1.0, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 1.0, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.0, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.8666666746139526, 0.5333333611488342, 0.7333333492279053, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 0.8666666746139526, 1.0, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.46666666865348816, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.4000000059604645, NaN, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.6000000238418579, 0.6333333253860474, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 0.8333333134651184, 0.6666666865348816, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.13333334028720856, 0.0, 0.06666667014360428, 0.2666666805744171, 0.13333334028720856, 0.06666667014360428, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, NaN, 0.46666666865348816, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.9333333373069763, 0.6666666865348816, 1.0, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 1.0, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763], [0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, NaN, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.9333333373069763], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.5333333611488342, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.5333333611488342, NaN, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.4000000059604645, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.9333333373069763, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 1.0, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.0, 0.13333334028720856, 0.13333334028720856, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.13333334028720856, 0.06666667014360428, 0.46666666865348816, 0.10000000149011612, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, NaN, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.9333333373069763, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.0, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.46666666865348816, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, NaN, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 1.0, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, NaN, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0, 1.0, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.0, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, NaN, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 0.6666666865348816, 0.9333333373069763, 0.7333333492279053, 0.6000000238418579, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, NaN, 0.46666666865348816, 0.6666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 1.0, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.0, 0.13333334028720856, 0.0, 0.13333334028720856, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.46666666865348816, 0.2666666805744171, 0.46666666865348816, 0.20000000298023224, 0.4000000059604645, 0.46666666865348816, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, NaN, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.0, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.0, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.13333334028720856, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.46666666865348816, NaN, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.9333333373069763, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.8666666746139526, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.0, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.3333333432674408, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.06666667014360428, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, NaN, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.36666667461395264, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, NaN, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.8333333134651184, 0.6666666865348816, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.0, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.4000000059604645, 0.0, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, NaN, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.6666666865348816, 0.8666666746139526, 1.0, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763], [0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.0, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.13333334028720856, 0.36666667461395264, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.4333333373069763, 0.46666666865348816, NaN, 0.46666666865348816, 0.4000000059604645, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.5333333611488342, 0.6333333253860474, 0.6000000238418579, 0.7666666507720947, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.9333333373069763, 1.0], [0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.06666667014360428, 0.06666667014360428, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, NaN, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.9333333373069763, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.8666666746139526, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, NaN, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 1.0, 1.0, 1.0], [0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.13333334028720856, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, NaN, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.7333333492279053, 1.0, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.9333333373069763, 0.6666666865348816, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 1.0, 1.0], [0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, NaN, 0.46666666865348816, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5333333611488342, 0.46666666865348816, 0.2666666805744171, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, NaN, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.6666666865348816, 0.8666666746139526, 0.7333333492279053, 0.6000000238418579, 0.8666666746139526, 1.0, 0.8666666746139526, 0.9333333373069763, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.0, 0.20000000298023224, 0.3333333432674408, 0.13333334028720856, 0.3333333432674408, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.06666667014360428, 0.46666666865348816, 0.20000000298023224, 0.46666666865348816, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.4000000059604645, NaN, 0.4000000059604645, 0.6666666865348816, 0.6666666865348816, 1.0, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 1.0, 0.6666666865348816, 0.46666666865348816, 0.7333333492279053, 0.6666666865348816, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.0, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.46666666865348816, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.46666666865348816, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.5333333611488342, 0.8666666746139526, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 1.0], [0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, NaN, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, 0.6666666865348816, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6000000238418579, 0.46666666865348816, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.2666666805744171, 0.13333334028720856, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.06666667014360428, 0.13333334028720856, 0.3333333432674408, 0.3333333432674408, 0.1666666716337204, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.13333334028720856, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.6000000238418579, NaN, 0.4000000059604645, 0.5666666626930237, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.5666666626930237, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.13333334028720856, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.06666667014360428, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.46666666865348816, 0.3333333432674408, 0.0, 0.4000000059604645, 0.5333333611488342, 0.6000000238418579, NaN, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.9333333373069763, 0.6666666865348816, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.0, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.2666666805744171, 0.10000000149011612, 0.2666666805744171, 0.23333333432674408, 0.2666666805744171, 0.13333334028720856, 0.46666666865348816, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.5333333611488342, 0.5333333611488342, 0.4333333373069763, 0.3333333432674408, NaN, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 1.0], [0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.3333333432674408, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.46666666865348816, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.06666667014360428, 0.46666666865348816, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.3333333432674408, NaN, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.800000011920929, 0.5333333611488342, 0.6666666865348816, 0.800000011920929, 0.6000000238418579, 1.0, 0.8666666746139526, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.06666667014360428, 0.2666666805744171, 0.13333334028720856, 0.3333333432674408, 0.3333333432674408, 0.0, 0.13333334028720856, 0.20000000298023224, 0.4000000059604645, 0.46666666865348816, 0.13333334028720856, 0.5333333611488342, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, NaN, 0.6000000238418579, 0.6000000238418579, 0.46666666865348816, 0.7333333492279053, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.8666666746139526], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.0, 0.06666667014360428, 0.0, 0.3333333432674408, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.0, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.4000000059604645, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.5333333611488342, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.4000000059604645, NaN, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 1.0, 0.800000011920929, 0.8666666746139526, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.0, 0.0, 0.0, 0.06666667014360428, 0.3333333432674408, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.0, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.4000000059604645, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.5333333611488342, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.4000000059604645, 0.3333333432674408, NaN, 0.5333333611488342, 0.7333333492279053, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 1.0, 0.800000011920929, 0.8666666746139526, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.0, 0.13333334028720856, 0.2666666805744171, 0.13333334028720856, 0.06666667014360428, 0.2666666805744171, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.0, 0.06666667014360428, 0.13333334028720856, 0.3333333432674408, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.13333334028720856, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.6000000238418579, 0.2666666805744171, 0.46666666865348816, 0.3333333432674408, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.46666666865348816, NaN, 0.6666666865348816, 0.5333333611488342, 0.7333333492279053, 0.6666666865348816, 0.46666666865348816, 0.6666666865348816, 0.6666666865348816, 1.0], [0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.13333334028720856, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.0, 0.20000000298023224, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.20000000298023224, 0.0, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.5333333611488342, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, NaN, 0.46666666865348816, 0.46666666865348816, 0.8999999761581421, 0.3333333432674408, 0.4000000059604645, 0.8666666746139526, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.13333334028720856, 0.06666667014360428, 0.0, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.06666667014360428, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.3333333432674408, 0.4333333373069763, 0.4000000059604645, 0.36666667461395264, 0.46666666865348816, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.5333333611488342, NaN, 0.6000000238418579, 0.5333333611488342, 0.5333333611488342, 0.6666666865348816, 0.6000000238418579, 0.8666666746139526], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.0, 0.3333333432674408, 0.46666666865348816, 0.2666666805744171, 0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, 0.2666666805744171, 0.5333333611488342, 0.4000000059604645, NaN, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.8666666746139526], [0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.13333334028720856, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.20000000298023224, 0.3333333432674408, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.5333333611488342, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.10000000149011612, 0.46666666865348816, 0.46666666865348816, NaN, 0.3333333432674408, 0.4000000059604645, 0.800000011920929, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.13333334028720856, 0.0, 0.3333333432674408, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.13333334028720856, 0.13333334028720856, 0.0, 0.0, 0.06666667014360428, 0.5333333611488342, 0.13333334028720856, 0.46666666865348816, 0.3333333432674408, 0.5333333611488342, 0.2666666805744171, 0.4000000059604645, 0.2666666805744171, 0.0, 0.0, 0.5333333611488342, 0.6666666865348816, 0.46666666865348816, 0.5333333611488342, 0.6666666865348816, NaN, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.3333333432674408, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.3333333432674408, 0.0, 0.20000000298023224, 0.0, 0.13333334028720856, 0.0, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.3333333432674408, 0.0, 0.3333333432674408, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.6000000238418579, 0.3333333432674408, 0.6000000238418579, 0.6000000238418579, 0.3333333432674408, NaN, 0.6000000238418579, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.06666667014360428, 0.13333334028720856, 0.0, 0.06666667014360428, 0.0, 0.0, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.2666666805744171, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.2666666805744171, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.3333333432674408, 0.06666667014360428, 0.5333333611488342, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.13333334028720856, 0.4000000059604645, 0.46666666865348816, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, NaN, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.0, 0.3333333432674408, 0.13333334028720856, 0.13333334028720856, 0.3333333432674408, 0.2666666805744171, 0.0, 0.3333333432674408, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9377777815},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9351111142},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9111111152},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8835555621},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8640000059},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8373333406},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8346666743},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8071111192},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8053333409},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7680000093},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.740888897},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7404444536},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7315555638},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7191111218},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.7160000097},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.715555564},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7022222316},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.7013333425},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7013333417},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6960000086},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6888888981},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6880000106},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6737777859},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.6702222318},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.666666677},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6524444564},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6462222318},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6000000098},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5942222316},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5902222336},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5768888981},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5688888985},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.566222233},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5608889013},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5360000085},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5324444535},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.5204444537},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5102222329},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4862222356},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4746666758},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4733333445},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4702222313},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.440888898},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4320000086},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4275555634},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4275555631},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.4177777885},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4177777876},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4120000069},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4044444551},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.3906666739},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.3884444544},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3866666746},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3768888974},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3715555641},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3680000093},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3075555621},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2817777845},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.277333341},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2408888954},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2368888953},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.235555563},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.218222229},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2151111164},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2097777836},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2071111171},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2017777836},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1884444502},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1720000043},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1657777829},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1573333382},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1506666712},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1297777822},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1173333365},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0951111145},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0337777789}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..058806ea40561e4b9204ff873dd4506ced9f98b9
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,77 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)","AutoGluon 1.4 (best, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),TabICLv2 (default),ModernNCA (tuned + ensembled),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),CatBoost (tuned + ensembled),LightGBM (tuned + ensembled),CatBoost (tuned),TabM (tuned + ensembled),ModernNCA (tuned),XGBoost (tuned + ensembled),iLTM (tuned + ensembled),LightGBM (tuned),CatBoost (default),RealTabPFN-2.5 (default),RealMLP (tuned),ChimeraBoost (tuned + ensembled),XGBoost (tuned),ChimeraBoost (tuned),TabM (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),xRFM (tuned + ensembled),TabDPT (tuned),ChimeraBoost (default),iLTM (tuned),xRFM (tuned),XGBoost (default),TabSwift (default),TorchMLP (tuned + ensembled),TabM (default),TabICL (default),EBM (tuned + ensembled),LimiX (default),ModernNCA (default),TabDPT (default),RealMLP (default),RandomForest (tuned + ensembled),TorchMLP (tuned),ExtraTrees (tuned + ensembled),LightGBM (default),Nori-30M (default),EBM (tuned),RandomForest (tuned),EBM (default),BetaTabPFN (default),Nori (default),ExtraTrees (tuned),FastaiMLP (tuned + ensembled),SAP-RPT-OSS (default),iLTM (default),TorchMLP (default),PerpetualBooster (tuned + ensembled),FastaiMLP (tuned),xRFM (default),RandomForest (default),PerpetualBooster (tuned),OrionMSP (default),KNN (tuned + ensembled),FastaiMLP (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),ExtraTrees (default),Linear (tuned + ensembled),TabFlex (default),PerpetualBooster (default),Linear (tuned),TabSTAR (default),KNN (tuned),Linear (default),KNN (default)
+TabFM (default),,0.6,0.6,0.73333335,0.8666667,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.93333334,0.8666667,1.0,0.8666667,0.8666667,0.8,1.0,0.8666667,0.93333334,0.8666667,0.8666667,0.8,0.8666667,1.0,1.0,0.8666667,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.8666667,0.93333334,1.0,0.8666667,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.4,,0.6666667,0.4,0.6666667,0.8,0.53333336,0.8,0.8,0.8666667,1.0,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,0.8,0.93333334,1.0,1.0,1.0,0.93333334,0.93333334,0.8,0.93333334,0.73333335,1.0,0.93333334,1.0,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,0.8666667,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,0.8666667,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.4,0.33333334,,0.4,0.73333335,0.53333336,0.53333336,0.6,0.6666667,0.8,1.0,0.93333334,1.0,0.8,0.8,1.0,1.0,0.93333334,0.93333334,0.6666667,0.93333334,0.93333334,1.0,0.93333334,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,0.93333334,1.0,0.8666667,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.26666668,0.6,0.6,,0.6666667,0.6666667,0.73333335,0.8666667,0.8,0.73333335,0.73333335,0.8,0.73333335,0.8666667,0.8,0.73333335,0.8,0.93333334,0.73333335,0.93333334,0.73333335,0.8,0.73333335,0.8,0.93333334,0.93333334,0.73333335,0.8,0.8,0.8,0.93333334,0.8666667,0.8666667,0.93333334,0.8,1.0,0.8666667,0.8666667,0.8666667,1.0,1.0,1.0,0.93333334,1.0,0.93333334,1.0,0.93333334,0.93333334,1.0,0.93333334,1.0,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.13333334,0.33333334,0.26666668,0.33333334,,0.46666667,0.53333336,0.73333335,0.53333336,0.73333335,0.73333335,0.8666667,0.73333335,0.8,0.8,0.6666667,0.73333335,0.93333334,0.8,0.8,0.8,0.8666667,0.8,0.8666667,0.8666667,0.8666667,0.8666667,0.73333335,0.8,0.8,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,1.0,0.93333334,1.0,0.8666667,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,0.8666667,1.0,1.0,1.0,0.93333334,0.93333334,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.13333334,0.2,0.46666667,0.33333334,0.53333336,,0.4,0.53333336,0.8,0.73333335,0.73333335,0.73333335,0.6666667,0.6,0.8,0.73333335,0.6666667,0.8,0.8,0.8666667,0.73333335,0.73333335,0.8,0.8,0.73333335,0.8,0.73333335,0.8,0.8,0.8,0.8,0.8,0.8666667,0.93333334,0.8,0.93333334,0.8666667,0.93333334,0.73333335,0.93333334,0.8666667,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,0.93333334,1.0,0.93333334,1.0,1.0,1.0,0.8666667,0.93333334,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.13333334,0.46666667,0.46666667,0.26666668,0.46666667,0.6,,0.73333335,0.73333335,0.6666667,0.73333335,0.8,0.73333335,0.73333335,0.8,0.73333335,0.73333335,0.8666667,0.73333335,0.73333335,0.73333335,0.73333335,0.73333335,0.73333335,0.73333335,0.8666667,0.8,0.73333335,0.8,0.73333335,0.8,0.73333335,0.73333335,0.8666667,0.8,1.0,0.8666667,0.73333335,0.8666667,0.93333334,0.8666667,0.93333334,0.93333334,0.8,1.0,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,1.0,0.93333334,1.0,0.93333334,0.93333334,0.93333334,1.0,0.93333334,1.0,0.8666667,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.13333334,0.2,0.4,0.13333334,0.26666668,0.46666667,0.26666668,,0.53333336,0.6,0.6,0.6,0.6,0.6666667,0.8,0.73333335,0.6666667,0.6666667,0.73333335,0.73333335,0.8,0.73333335,0.73333335,0.73333335,0.73333335,0.6666667,0.8666667,0.93333334,0.8,0.8666667,0.8,0.8666667,0.73333335,0.8,0.8666667,0.8666667,0.8666667,0.8666667,0.8,1.0,0.8666667,0.93333334,0.8666667,0.93333334,0.93333334,1.0,0.8666667,0.8666667,0.8666667,0.93333334,0.93333334,0.93333334,0.93333334,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.93333334,0.93333334,1.0,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.06666667,0.2,0.33333334,0.2,0.46666667,0.2,0.26666668,0.46666667,,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.73333335,0.6,0.6,0.8,0.73333335,0.73333335,0.73333335,0.6,0.73333335,0.73333335,0.8,0.8,0.73333335,0.73333335,0.8,0.73333335,0.8,0.8,0.8666667,0.8666667,0.8,0.93333334,0.8666667,0.93333334,0.73333335,0.93333334,0.8666667,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,0.93333334,1.0,0.93333334,0.93333334,1.0,0.8666667,0.8666667,0.93333334,0.8666667,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.13333334,0.13333334,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.4,0.33333334,,0.6,0.53333336,0.6,0.53333336,0.53333336,0.53333336,0.6666667,0.6,0.6,0.6666667,0.8,0.46666667,0.6666667,0.6,0.6666667,0.6666667,0.6666667,0.8666667,0.6666667,0.6666667,0.8,0.93333334,0.8,0.73333335,0.8666667,0.93333334,0.6666667,0.73333335,0.8,0.93333334,0.8,1.0,0.93333334,0.93333334,0.93333334,1.0,0.8,0.8666667,0.93333334,0.8666667,0.8666667,0.8666667,0.8666667,1.0,1.0,0.8666667,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.13333334,0.0,0.0,0.26666668,0.26666668,0.26666668,0.26666668,0.4,0.33333334,0.4,,0.4,0.6666667,0.4,0.43333334,0.53333336,0.6,0.46666667,0.8,0.6,0.6,0.6666667,0.6666667,0.8,0.46666667,0.46666667,0.6,0.73333335,0.6666667,0.8666667,0.8666667,0.8666667,0.8666667,0.6666667,0.8,0.8,0.8,0.73333335,0.6666667,0.73333335,0.73333335,0.8666667,0.93333334,0.93333334,1.0,0.93333334,0.73333335,0.73333335,0.93333334,0.93333334,0.8666667,0.73333335,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.06666667,0.0,0.06666667,0.2,0.13333334,0.26666668,0.2,0.4,0.33333334,0.46666667,0.6,,0.6666667,0.33333334,0.46666667,0.6,0.6666667,0.73333335,0.73333335,0.53333336,0.53333336,0.6,0.6,0.73333335,0.53333336,0.6666667,0.53333336,0.6,0.6,0.73333335,0.8,0.73333335,0.8666667,0.6,0.73333335,0.8666667,0.73333335,0.8,0.6666667,0.8666667,0.8,0.8666667,0.93333334,0.8666667,1.0,1.0,0.73333335,1.0,1.0,1.0,0.8666667,0.8,1.0,0.8666667,0.8666667,1.0,1.0,1.0,0.93333334,0.8666667,1.0,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.13333334,0.0,0.0,0.26666668,0.26666668,0.33333334,0.26666668,0.4,0.33333334,0.4,0.33333334,0.33333334,,0.46666667,0.46666667,0.53333336,0.6,0.46666667,0.6666667,0.53333336,0.6,0.6666667,0.6666667,0.6666667,0.46666667,0.46666667,0.6,0.73333335,0.6,0.8666667,0.8666667,0.8,0.8666667,0.6666667,0.8,0.8666667,0.8,0.8,0.6666667,0.73333335,0.73333335,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.73333335,0.73333335,0.93333334,0.93333334,0.8666667,0.73333335,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.06666667,0.2,0.13333334,0.2,0.4,0.26666668,0.33333334,0.33333334,0.46666667,0.6,0.6666667,0.53333336,,0.53333336,0.46666667,0.53333336,0.6666667,0.6,0.46666667,0.53333336,0.53333336,0.6666667,0.53333336,0.6666667,0.6,0.53333336,0.53333336,0.6666667,0.6,0.6666667,0.53333336,0.6666667,0.6666667,0.73333335,1.0,0.73333335,0.8666667,0.6666667,0.8,0.73333335,1.0,0.8666667,0.8,0.8666667,0.93333334,0.6666667,0.8666667,0.8666667,0.8666667,0.93333334,0.8,0.8666667,0.93333334,0.8666667,0.8,1.0,0.93333334,1.0,0.8666667,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.13333334,0.06666667,0.2,0.2,0.2,0.2,0.2,0.2,0.26666668,0.46666667,0.56666666,0.53333336,0.53333336,0.46666667,,0.6,0.53333336,0.53333336,0.53333336,0.53333336,0.53333336,0.46666667,0.6,0.46666667,0.53333336,0.53333336,0.6666667,0.6,0.8,0.6666667,0.6666667,0.8,0.6,0.8666667,0.8666667,0.73333335,0.8,0.6666667,0.73333335,0.93333334,0.8666667,0.8,0.8,0.8666667,0.8666667,0.8666667,0.8,0.8,0.93333334,0.8666667,0.93333334,0.8666667,0.8666667,0.8666667,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.13333334,0.0,0.0,0.26666668,0.33333334,0.26666668,0.26666668,0.26666668,0.4,0.46666667,0.46666667,0.4,0.46666667,0.53333336,0.4,,0.4,0.4,0.53333336,0.53333336,0.46666667,0.46666667,0.73333335,0.53333336,0.53333336,0.46666667,0.6,0.6,0.53333336,0.73333335,0.73333335,0.6666667,0.8666667,0.53333336,0.8666667,0.8,0.8,0.8,0.6,0.8,0.8,0.8666667,0.93333334,0.8666667,1.0,1.0,0.6666667,0.8666667,0.93333334,1.0,0.8666667,0.73333335,0.93333334,0.8666667,0.93333334,0.93333334,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.2,0.0,0.0,0.2,0.26666668,0.33333334,0.26666668,0.33333334,0.4,0.33333334,0.4,0.33333334,0.4,0.46666667,0.46666667,0.6,,0.4,0.46666667,0.6,0.53333336,0.53333336,0.6666667,0.53333336,0.53333336,0.4,0.53333336,0.6,0.6,0.6,0.73333335,0.6,0.73333335,0.6,0.8666667,0.73333335,0.8,0.73333335,0.73333335,0.6666667,0.73333335,0.8666667,0.93333334,1.0,0.8666667,0.8666667,0.6666667,0.6666667,0.93333334,0.8,0.8666667,0.6666667,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.06666667,0.06666667,0.06666667,0.2,0.13333334,0.33333334,0.2,0.4,0.53333336,0.26666668,0.53333336,0.33333334,0.46666667,0.6,0.6,,0.53333336,0.53333336,0.46666667,0.6,0.6,0.73333335,0.46666667,0.6666667,0.6,0.6,0.6,0.73333335,0.6666667,0.6666667,0.8,0.46666667,0.6666667,0.8666667,0.6666667,0.8,0.6666667,0.8,0.8,0.8666667,0.8666667,0.8,0.8666667,1.0,0.73333335,0.93333334,1.0,0.93333334,0.8666667,0.8,0.93333334,0.8666667,0.8666667,1.0,0.93333334,1.0,0.93333334,0.8666667,1.0,1.0,1.0,0.93333334,0.93333334,1.0,1.0,1.0,0.8666667,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.13333334,0.0,0.06666667,0.26666668,0.2,0.2,0.26666668,0.26666668,0.26666668,0.4,0.2,0.26666668,0.33333334,0.4,0.46666667,0.46666667,0.53333336,0.46666667,,0.6,0.6,0.4,0.53333336,0.53333336,0.46666667,0.46666667,0.6,0.73333335,0.6,0.73333335,0.8,0.73333335,0.73333335,0.6666667,0.8666667,0.73333335,0.8,0.8,0.6666667,0.8,0.73333335,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.73333335,0.8,0.8666667,0.93333334,0.8666667,0.73333335,0.8666667,0.93333334,0.8666667,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.06666667,0.2,0.33333334,0.06666667,0.2,0.13333334,0.26666668,0.26666668,0.26666668,0.33333334,0.4,0.46666667,0.46666667,0.53333336,0.46666667,0.46666667,0.4,0.46666667,0.4,,0.46666667,0.46666667,0.6,0.46666667,0.6,0.53333336,0.6666667,0.73333335,0.6,0.6666667,0.73333335,0.6666667,0.73333335,0.8666667,0.6666667,0.8,0.8,0.6666667,0.6666667,0.8666667,0.6666667,0.8,0.8,0.8666667,0.8666667,1.0,0.8,0.8,0.8666667,0.8,1.0,0.8666667,0.8666667,0.8,0.93333334,0.8,1.0,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,1.0,0.93333334,0.8666667,1.0,1.0,1.0,0.93333334,1.0,0.93333334,0.93333334,1.0,1.0,0.93333334,1.0
+RealMLP (tuned),0.13333334,0.06666667,0.06666667,0.26666668,0.2,0.26666668,0.26666668,0.2,0.26666668,0.2,0.4,0.46666667,0.4,0.46666667,0.46666667,0.53333336,0.46666667,0.53333336,0.4,0.53333336,,0.33333334,0.6,0.4,0.53333336,0.46666667,0.6666667,0.8,0.6666667,0.53333336,0.73333335,0.73333335,0.73333335,0.6,0.8666667,0.73333335,0.6666667,0.6666667,0.73333335,0.8,0.73333335,0.93333334,0.8666667,0.93333334,0.73333335,0.8666667,0.73333335,0.8,0.8666667,0.8,0.8666667,0.6666667,0.8,0.93333334,0.8666667,0.8,1.0,0.93333334,1.0,0.93333334,0.8666667,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.13333334,0.0,0.06666667,0.2,0.13333334,0.26666668,0.26666668,0.26666668,0.4,0.53333336,0.33333334,0.4,0.33333334,0.46666667,0.53333336,0.53333336,0.46666667,0.4,0.6,0.53333336,0.6666667,,0.53333336,0.6,0.46666667,0.46666667,0.6666667,0.73333335,0.6,0.8666667,0.73333335,0.8,0.73333335,0.53333336,0.73333335,0.73333335,0.8,0.8,0.6,0.73333335,0.6666667,0.8666667,0.8666667,0.8,0.8666667,0.8666667,0.73333335,0.8666667,0.8666667,0.93333334,0.8,0.73333335,0.8666667,0.8666667,0.8666667,0.8666667,0.8,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.93333334,1.0,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.93333334,0.93333334,1.0,0.93333334,1.0,1.0,1.0
+XGBoost (tuned),0.2,0.0,0.0,0.26666668,0.2,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.33333334,0.4,0.33333334,0.33333334,0.4,0.26666668,0.33333334,0.4,0.46666667,0.4,0.4,0.46666667,,0.46666667,0.46666667,0.46666667,0.6666667,0.6666667,0.53333336,0.6666667,0.6666667,0.6,0.8666667,0.53333336,0.73333335,0.73333335,0.73333335,0.6666667,0.6,0.8,0.73333335,0.8,0.8666667,0.8666667,0.93333334,0.8666667,0.6666667,0.8666667,0.8666667,0.93333334,0.93333334,0.73333335,0.93333334,0.8,0.8666667,0.93333334,1.0,0.8666667,0.93333334,0.8,1.0,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.13333334,0.0,0.06666667,0.2,0.13333334,0.2,0.26666668,0.26666668,0.26666668,0.4,0.2,0.26666668,0.33333334,0.46666667,0.53333336,0.46666667,0.46666667,0.26666668,0.46666667,0.53333336,0.6,0.4,0.53333336,,0.46666667,0.46666667,0.6666667,0.73333335,0.6,0.8,0.73333335,0.8,0.73333335,0.6,0.73333335,0.73333335,0.8,0.8,0.6,0.73333335,0.6,0.8666667,0.8666667,0.8,0.8666667,0.8666667,0.73333335,0.8666667,0.8666667,0.93333334,0.8666667,0.73333335,0.8,0.8666667,0.8666667,0.8,0.8,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.93333334,1.0,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.93333334,0.93333334,1.0,0.93333334,1.0,1.0,1.0
+TabM (tuned),0.0,0.06666667,0.2,0.06666667,0.13333334,0.26666668,0.26666668,0.26666668,0.2,0.33333334,0.53333336,0.46666667,0.53333336,0.33333334,0.46666667,0.46666667,0.46666667,0.53333336,0.53333336,0.4,0.46666667,0.53333336,0.53333336,0.53333336,,0.53333336,0.46666667,0.46666667,0.6,0.46666667,0.6,0.53333336,0.6666667,0.6,0.6,0.8666667,0.73333335,0.73333335,0.6666667,0.8,0.6,0.8666667,0.73333335,0.8,0.8,0.8,0.6666667,0.8666667,0.8,0.8,0.93333334,0.73333335,0.8,0.8666667,0.8666667,0.73333335,1.0,0.93333334,0.93333334,0.8666667,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0,0.93333334,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.06666667,0.2,0.06666667,0.13333334,0.2,0.13333334,0.33333334,0.2,0.33333334,0.53333336,0.33333334,0.53333336,0.4,0.46666667,0.53333336,0.6,0.33333334,0.53333336,0.46666667,0.53333336,0.53333336,0.53333336,0.53333336,0.46666667,,0.33333334,0.6,0.53333336,0.53333336,0.73333335,0.73333335,0.6,0.73333335,0.73333335,0.6666667,0.53333336,0.6,0.8,0.73333335,0.8666667,0.73333335,0.73333335,0.8666667,0.8,0.8,0.73333335,0.6666667,0.8666667,0.6666667,0.73333335,0.93333334,0.8,0.73333335,0.73333335,0.8,0.8666667,0.93333334,0.93333334,0.93333334,1.0,1.0,0.93333334,0.8666667,0.93333334,0.8666667,0.8666667,1.0,0.8,0.93333334,1.0,0.8,1.0,1.0,0.93333334,1.0
+TabDPT (tuned + ensembled),0.13333334,0.2,0.2,0.26666668,0.13333334,0.26666668,0.2,0.13333334,0.26666668,0.33333334,0.4,0.46666667,0.4,0.46666667,0.33333334,0.4,0.46666667,0.4,0.4,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.53333336,0.6666667,,0.6666667,0.8,0.46666667,0.6,0.6666667,0.53333336,0.53333336,0.8,0.6666667,0.4,0.53333336,0.6666667,0.8666667,0.93333334,0.73333335,0.73333335,0.8666667,0.73333335,0.8666667,0.8666667,0.73333335,0.73333335,0.73333335,0.73333335,0.93333334,0.8,0.73333335,0.8,0.6666667,0.8666667,1.0,0.93333334,0.93333334,1.0,1.0,0.93333334,0.93333334,0.93333334,0.8666667,0.8666667,1.0,0.8666667,0.93333334,1.0,0.8666667,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.06666667,0.06666667,0.0,0.2,0.26666668,0.2,0.26666668,0.06666667,0.26666668,0.13333334,0.26666668,0.4,0.26666668,0.46666667,0.4,0.4,0.4,0.4,0.26666668,0.26666668,0.2,0.26666668,0.33333334,0.26666668,0.53333336,0.4,0.33333334,,0.4,0.33333334,0.53333336,0.6666667,0.46666667,0.46666667,0.6,0.6,0.6,0.6,0.73333335,0.73333335,0.53333336,0.73333335,0.8,0.6666667,0.8,0.6666667,0.6666667,0.8,0.8,0.6666667,0.8,0.6666667,0.8,0.8,0.73333335,0.73333335,0.8666667,0.93333334,0.8666667,1.0,0.93333334,0.93333334,1.0,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,0.93333334,1.0,0.93333334,1.0,1.0,1.0
+TabDPT (tuned),0.06666667,0.26666668,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.33333334,0.33333334,0.4,0.4,0.33333334,0.2,0.46666667,0.4,0.4,0.4,0.4,0.33333334,0.4,0.46666667,0.4,0.4,0.46666667,0.2,0.6,,0.46666667,0.6,0.6666667,0.46666667,0.53333336,0.53333336,0.53333336,0.4,0.46666667,0.6666667,0.6,0.9,0.53333336,0.6,0.8,0.6666667,0.73333335,0.6666667,0.6666667,0.73333335,0.6,0.6,0.8,0.8,0.6,0.73333335,0.73333335,0.8666667,1.0,0.8666667,0.93333334,0.8666667,1.0,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,1.0,0.8,0.93333334,1.0,0.8,1.0,1.0,0.93333334,1.0
+ChimeraBoost (default),0.06666667,0.0,0.0,0.2,0.2,0.2,0.26666668,0.13333334,0.26666668,0.33333334,0.13333334,0.26666668,0.13333334,0.4,0.33333334,0.26666668,0.4,0.26666668,0.26666668,0.33333334,0.46666667,0.13333334,0.33333334,0.2,0.53333336,0.46666667,0.53333336,0.6666667,0.53333336,,0.53333336,0.6666667,0.6,0.46666667,0.6666667,0.53333336,0.6,0.6666667,0.6,0.6666667,0.6,0.6666667,0.73333335,0.8,0.6,0.6666667,0.6666667,0.8,0.73333335,0.8,0.73333335,0.73333335,0.6,0.8,0.8,0.73333335,0.8,0.8666667,0.8666667,0.8666667,0.93333334,0.8666667,0.93333334,0.93333334,0.8666667,0.93333334,0.93333334,0.93333334,1.0,0.93333334,0.8666667,1.0,0.93333334,1.0,1.0,1.0
+iLTM (tuned),0.06666667,0.06666667,0.0,0.06666667,0.13333334,0.2,0.2,0.2,0.2,0.2,0.13333334,0.2,0.13333334,0.33333334,0.33333334,0.26666668,0.26666668,0.33333334,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.26666668,0.4,0.26666668,0.4,0.46666667,0.4,0.46666667,,0.46666667,0.4,0.53333336,0.6666667,0.53333336,0.6666667,0.6,0.6666667,0.6,0.6,0.6666667,0.8,0.73333335,0.6,0.8,0.6,0.6666667,0.8666667,0.6666667,0.8,0.6,0.8,0.8666667,0.8666667,0.8,0.93333334,0.8666667,0.8666667,0.93333334,0.93333334,0.8666667,0.93333334,1.0,0.93333334,1.0,1.0,0.93333334,0.93333334,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.06666667,0.0,0.0,0.13333334,0.13333334,0.2,0.26666668,0.13333334,0.2,0.06666667,0.13333334,0.26666668,0.2,0.46666667,0.2,0.33333334,0.4,0.33333334,0.26666668,0.33333334,0.26666668,0.2,0.4,0.2,0.46666667,0.26666668,0.33333334,0.33333334,0.33333334,0.33333334,0.53333336,,0.46666667,0.46666667,0.6,0.6,0.53333336,0.6,0.6,0.6666667,0.53333336,0.73333335,0.73333335,0.8,0.73333335,0.6666667,0.6,0.6666667,0.8,0.6,0.73333335,0.6,0.73333335,0.73333335,0.8666667,0.6666667,0.93333334,0.8666667,0.8666667,1.0,0.8666667,0.93333334,0.93333334,1.0,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.13333334,0.26666668,0.26666668,0.13333334,0.2,0.13333334,0.13333334,0.13333334,0.33333334,0.4,0.13333334,0.26666668,0.2,0.26666668,0.26666668,0.26666668,0.26666668,0.13333334,0.26666668,0.33333334,0.4,0.46666667,0.53333336,0.53333336,0.4,0.6,0.53333336,,0.46666667,0.6,0.4,0.6,0.6,0.6,0.6666667,0.6666667,0.53333336,0.8,0.6666667,0.73333335,0.53333336,0.6,0.73333335,0.8,0.8,0.73333335,0.6666667,0.73333335,0.8,0.8,0.93333334,0.8666667,0.8,0.8,0.8,0.93333334,0.93333334,1.0,0.93333334,0.8666667,1.0,1.0,0.8666667,0.93333334,1.0,1.0,0.93333334,1.0,1.0,0.93333334,1.0
+TabSwift (default),0.13333334,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.2,0.13333334,0.26666668,0.33333334,0.4,0.33333334,0.33333334,0.13333334,0.46666667,0.4,0.53333336,0.33333334,0.13333334,0.4,0.46666667,0.46666667,0.4,0.4,0.26666668,0.46666667,0.53333336,0.46666667,0.53333336,0.46666667,0.53333336,0.53333336,,0.53333336,0.6,0.53333336,0.6,0.6,0.6,0.6666667,0.6,0.6,0.6,0.6,0.73333335,0.6666667,0.6666667,0.6,0.73333335,0.6666667,0.6666667,0.6666667,0.8,0.73333335,0.73333335,0.73333335,0.8666667,0.8,0.8,0.8,0.8666667,0.8666667,0.8666667,0.8,0.8666667,0.8666667,0.93333334,0.8,0.8666667,0.93333334,0.8666667,0.93333334,1.0,0.8666667,1.0
+TorchMLP (tuned + ensembled),0.06666667,0.0,0.0,0.2,0.13333334,0.2,0.2,0.13333334,0.2,0.13333334,0.2,0.26666668,0.2,0.26666668,0.13333334,0.13333334,0.13333334,0.33333334,0.13333334,0.33333334,0.13333334,0.26666668,0.26666668,0.26666668,0.4,0.26666668,0.2,0.4,0.46666667,0.33333334,0.33333334,0.4,0.4,0.46666667,,0.4,0.46666667,0.6,0.46666667,0.46666667,0.53333336,0.6,0.73333335,0.8,0.73333335,0.53333336,0.6666667,0.6666667,0.8,0.6,0.6666667,0.6666667,0.73333335,0.6666667,0.8666667,0.8,1.0,0.73333335,0.93333334,0.8,0.93333334,0.73333335,0.93333334,1.0,0.93333334,1.0,1.0,0.8666667,1.0,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.13333334,0.06666667,0.06666667,0.2,0.13333334,0.13333334,0.0,0.26666668,0.2,0.26666668,0.13333334,0.26666668,0.2,0.26666668,0.26666668,0.26666668,0.26666668,0.13333334,0.33333334,0.33333334,0.4,0.46666667,0.46666667,0.46666667,0.4,0.6,0.4,0.6,,0.6,0.46666667,0.6,0.6,0.53333336,0.73333335,0.73333335,0.6,0.6666667,0.8,0.53333336,0.6666667,0.73333335,0.73333335,0.8666667,0.53333336,0.73333335,0.8666667,0.73333335,0.73333335,0.8,0.93333334,0.93333334,0.8,0.8666667,0.93333334,1.0,0.93333334,0.93333334,1.0,1.0,1.0,0.8666667,1.0,0.93333334,0.8,1.0,1.0,0.93333334,1.0
+TabICL (default),0.13333334,0.06666667,0.06666667,0.13333334,0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.33333334,0.2,0.26666668,0.2,0.26666668,0.2,0.2,0.2,0.33333334,0.2,0.2,0.33333334,0.2,0.26666668,0.2,0.26666668,0.46666667,0.6,0.4,0.6,0.4,0.33333334,0.46666667,0.4,0.46666667,0.53333336,0.4,,0.53333336,0.46666667,0.6,0.6666667,0.53333336,0.6,0.6,0.6,0.6,0.6666667,0.6,0.6333333,0.6,0.6333333,0.6666667,0.73333335,0.73333335,0.73333335,0.8,0.6666667,0.6666667,0.8,0.6,0.8333333,0.6666667,0.8333333,0.93333334,0.8666667,0.93333334,1.0,0.8666667,0.8666667,0.8333333,0.8666667,0.8666667,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.13333334,0.0,0.06666667,0.26666668,0.13333334,0.06666667,0.26666668,0.26666668,0.2,0.2,0.13333334,0.33333334,0.2,0.26666668,0.2,0.2,0.33333334,0.33333334,0.2,0.33333334,0.2,0.26666668,0.4,0.46666667,0.4,0.53333336,0.33333334,0.4,0.4,0.4,0.4,0.4,0.53333336,0.46666667,,0.46666667,0.6666667,0.53333336,0.53333336,0.6,0.6,0.6,0.53333336,0.6666667,0.93333334,0.6666667,1.0,0.6666667,0.6666667,0.6,0.6666667,0.73333335,0.6,0.8666667,0.6666667,0.73333335,0.73333335,0.73333335,0.6666667,0.73333335,0.8666667,0.8666667,1.0,0.93333334,0.73333335,0.93333334,0.8,0.73333335,0.93333334,1.0,0.93333334,1.0,0.93333334
+LimiX (default),0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.26666668,0.13333334,0.2,0.26666668,0.2,0.33333334,0.33333334,0.33333334,0.33333334,0.26666668,0.4,0.26666668,0.33333334,0.33333334,0.33333334,0.26666668,0.4,0.4,0.4,0.33333334,0.2,0.33333334,0.26666668,0.33333334,0.4,0.33333334,0.4,0.4,0.4,0.53333336,0.4,0.53333336,0.53333336,,0.46666667,0.53333336,0.6,0.53333336,0.6,0.4,0.4,0.6666667,0.53333336,0.53333336,0.53333336,0.6,0.6,0.46666667,0.6,0.6,0.53333336,0.6666667,0.6,0.73333335,0.8,0.8666667,0.6666667,0.73333335,0.8,0.8,0.6666667,0.6666667,0.8666667,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.93333334
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.06666667,0.26666668,0.13333334,0.26666668,0.2,0.06666667,0.2,0.33333334,0.2,0.2,0.13333334,0.2,0.26666668,0.2,0.26666668,0.2,0.26666668,0.13333334,0.26666668,0.4,0.33333334,0.4,0.33333334,0.33333334,0.4,0.53333336,0.4,0.4,0.33333334,0.53333336,,0.46666667,0.46666667,0.6,0.46666667,0.53333336,0.6,0.6,0.53333336,0.6,0.4,0.6666667,0.73333335,0.6,0.53333336,0.6,0.6,0.93333334,0.8,0.73333335,0.8666667,0.93333334,0.93333334,0.93333334,0.93333334,0.93333334,0.8666667,0.8666667,0.93333334,0.8,0.93333334,1.0,0.8,1.0,1.0,0.93333334,1.0
+TabDPT (default),0.0,0.13333334,0.13333334,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.26666668,0.2,0.26666668,0.26666668,0.13333334,0.2,0.26666668,0.2,0.26666668,0.33333334,0.26666668,0.33333334,0.26666668,0.4,0.4,0.13333334,0.06666667,0.46666667,0.1,0.4,0.4,0.46666667,0.33333334,0.33333334,0.46666667,0.46666667,0.33333334,0.46666667,0.46666667,0.53333336,,0.46666667,0.6,0.6,0.53333336,0.6,0.53333336,0.53333336,0.6666667,0.46666667,0.6666667,0.6,0.53333336,0.6,0.6,0.53333336,0.8,0.73333335,0.73333335,0.8,0.73333335,0.8666667,0.8,0.6666667,0.8666667,0.8,0.8,0.73333335,0.8,0.93333334,1.0,0.8,0.8,0.8,0.8666667,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.0,0.13333334,0.13333334,0.06666667,0.0,0.2,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.06666667,0.13333334,0.2,0.13333334,0.13333334,0.26666668,0.26666668,0.26666668,0.46666667,0.33333334,0.33333334,0.26666668,0.46666667,0.4,0.4,0.26666668,0.46666667,0.46666667,0.4,0.53333336,0.53333336,,0.6666667,0.6,0.6,0.53333336,0.53333336,0.6666667,0.73333335,0.6666667,0.6666667,0.53333336,0.6666667,0.6666667,0.6666667,0.6,0.8,0.8,1.0,0.73333335,0.8,0.8666667,0.8,0.8666667,1.0,0.8666667,0.8666667,0.93333334,0.93333334,0.93333334,0.93333334,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.13333334,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.2,0.06666667,0.06666667,0.13333334,0.13333334,0.2,0.13333334,0.13333334,0.13333334,0.13333334,0.26666668,0.26666668,0.26666668,0.2,0.4,0.26666668,0.2,0.26666668,0.2,0.4,0.26666668,0.26666668,0.4,0.4,0.46666667,0.4,0.4,0.33333334,,0.53333336,0.53333336,0.53333336,0.6,0.53333336,0.53333336,0.46666667,0.6666667,0.6,0.6666667,0.53333336,0.73333335,0.8,0.73333335,0.6666667,0.8,0.6666667,0.8666667,0.73333335,0.8666667,0.93333334,0.93333334,1.0,1.0,0.8666667,0.8666667,0.93333334,0.8,0.8666667,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.06666667,0.2,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.2,0.13333334,0.13333334,0.0,0.2,0.06666667,0.13333334,0.06666667,0.2,0.13333334,0.2,0.2,0.13333334,0.13333334,0.33333334,0.2,0.2,0.26666668,0.2,0.33333334,0.4,0.2,0.4,0.4,0.4,0.4,0.53333336,0.4,0.4,0.46666667,,0.6,0.53333336,0.53333336,0.6,0.46666667,0.53333336,0.53333336,0.53333336,0.6666667,0.53333336,0.73333335,0.6666667,0.93333334,0.73333335,0.6,0.8,0.73333335,0.73333335,0.8,0.8666667,0.8,0.93333334,0.93333334,0.8,0.93333334,0.8666667,0.8666667,0.93333334,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.06666667,0.13333334,0.13333334,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.26666668,0.13333334,0.06666667,0.13333334,0.2,0.2,0.26666668,0.2,0.33333334,0.4,0.4,0.26666668,0.26666668,0.4,0.26666668,0.33333334,0.4,0.4,0.6,0.46666667,0.46666667,0.4,0.46666667,0.4,,0.46666667,0.6666667,0.46666667,0.46666667,0.46666667,0.6,0.6666667,0.8,0.53333336,0.6,0.6666667,0.53333336,0.6666667,0.6666667,0.6666667,0.93333334,0.73333335,0.93333334,0.8666667,0.8,0.8666667,0.8666667,1.0,0.8,0.93333334,0.8666667,0.93333334,0.8666667,0.93333334,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.13333334,0.0,0.13333334,0.0,0.13333334,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.2,0.13333334,0.33333334,0.26666668,0.33333334,0.2,0.33333334,0.46666667,0.26666668,0.46666667,0.2,0.4,0.46666667,0.6,0.4,0.4,0.46666667,0.46666667,0.46666667,0.53333336,,0.53333336,0.53333336,0.46666667,0.46666667,0.6,0.53333336,0.6,0.6,0.6,0.6666667,0.8,0.8,0.73333335,0.8,0.8666667,0.93333334,0.93333334,0.8666667,0.8,0.8666667,0.93333334,0.93333334,0.8,0.93333334,0.93333334,0.8,1.0,1.0,0.93333334,1.0
+Nori-30M (default),0.0,0.13333334,0.2,0.06666667,0.13333334,0.06666667,0.06666667,0.13333334,0.0,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.2,0.33333334,0.33333334,0.26666668,0.26666668,0.2,0.26666668,0.26666668,0.33333334,0.26666668,0.33333334,0.26666668,0.13333334,0.33333334,0.33333334,0.33333334,0.4,0.4,0.4,0.33333334,0.33333334,0.46666667,0.33333334,0.33333334,0.33333334,0.4,0.46666667,0.46666667,0.4,0.46666667,0.33333334,0.46666667,,0.46666667,0.4,0.46666667,0.53333336,0.6,0.4,0.46666667,0.53333336,0.46666667,0.53333336,0.6666667,0.53333336,0.93333334,0.6666667,0.73333335,0.73333335,0.53333336,0.6666667,0.6,0.6,0.8666667,0.6666667,0.73333335,0.8666667,0.6666667,0.6666667,0.6666667,0.73333335,1.0
+EBM (tuned),0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.13333334,0.06666667,0.13333334,0.26666668,0.0,0.26666668,0.13333334,0.2,0.13333334,0.33333334,0.06666667,0.2,0.2,0.2,0.13333334,0.13333334,0.13333334,0.13333334,0.33333334,0.26666668,0.2,0.33333334,0.2,0.33333334,0.33333334,0.26666668,0.33333334,0.33333334,0.33333334,0.4,0.06666667,0.46666667,0.46666667,0.46666667,0.33333334,0.46666667,0.4,0.53333336,0.46666667,0.53333336,,0.53333336,0.6,0.46666667,0.6,0.6,0.53333336,0.6666667,0.53333336,0.6666667,0.6666667,0.73333335,0.6666667,0.6666667,0.6666667,0.73333335,0.8,0.8,0.93333334,0.93333334,0.6666667,0.8,0.8,0.6666667,0.8666667,1.0,0.93333334,0.93333334,0.93333334
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.13333334,0.0,0.06666667,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.06666667,0.06666667,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.13333334,0.26666668,0.2,0.26666668,0.26666668,0.13333334,0.2,0.2,0.4,0.2,0.26666668,0.36666667,0.33333334,0.46666667,0.4,0.33333334,0.26666668,0.46666667,0.53333336,0.53333336,0.53333336,0.6,0.46666667,,0.46666667,0.56666666,0.6,0.6,0.53333336,0.73333335,0.73333335,0.73333335,0.6666667,0.73333335,0.6666667,0.8333333,0.6666667,0.9,0.93333334,0.93333334,0.93333334,0.93333334,0.8666667,0.8,0.9,0.8,0.8,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.0,0.06666667,0.13333334,0.13333334,0.0,0.2,0.06666667,0.06666667,0.2,0.2,0.06666667,0.06666667,0.06666667,0.2,0.33333334,0.26666668,0.33333334,0.4,0.2,0.33333334,0.4,0.2,0.26666668,0.4,0.26666668,0.4,0.0,0.46666667,0.6,0.53333336,0.33333334,0.53333336,0.46666667,0.53333336,0.53333336,0.53333336,0.4,0.53333336,,0.53333336,0.6,0.6,0.6,0.6,0.53333336,0.6666667,0.6666667,0.73333335,0.6,0.6666667,0.6666667,0.73333335,0.8,0.73333335,0.8666667,0.8666667,0.6666667,0.8,0.8,0.6666667,0.8666667,1.0,0.93333334,0.93333334,0.93333334
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.06666667,0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.0,0.13333334,0.2,0.06666667,0.13333334,0.06666667,0.26666668,0.26666668,0.2,0.4,0.26666668,0.2,0.26666668,0.26666668,0.33333334,0.33333334,0.13333334,0.36666667,0.33333334,0.4,0.33333334,0.33333334,0.33333334,0.33333334,0.46666667,0.4,0.4,0.46666667,0.53333336,0.43333334,0.46666667,,0.46666667,0.4,0.6666667,0.6666667,0.6,0.6,0.6,0.8,0.53333336,0.6333333,0.6,0.76666665,0.93333334,0.8666667,0.8666667,0.8666667,0.73333335,0.8,0.8333333,0.8,0.8,1.0,1.0,0.93333334,1.0
+Nori (default),0.0,0.06666667,0.06666667,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.13333334,0.26666668,0.2,0.26666668,0.2,0.13333334,0.26666668,0.33333334,0.2,0.26666668,0.13333334,0.33333334,0.26666668,0.26666668,0.26666668,0.26666668,0.06666667,0.06666667,0.33333334,0.2,0.26666668,0.4,0.4,0.33333334,0.33333334,0.33333334,0.46666667,0.33333334,0.33333334,0.4,0.26666668,0.4,0.46666667,0.4,0.46666667,0.33333334,0.46666667,0.4,0.4,0.4,0.4,0.53333336,,0.4,0.46666667,0.46666667,0.46666667,0.53333336,0.6,0.53333336,0.93333334,0.6666667,0.73333335,0.73333335,0.53333336,0.6666667,0.6,0.6,0.8666667,0.6666667,0.73333335,0.8666667,0.6666667,0.6666667,0.6666667,0.73333335,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.13333334,0.13333334,0.06666667,0.0,0.06666667,0.13333334,0.13333334,0.06666667,0.06666667,0.06666667,0.13333334,0.13333334,0.2,0.13333334,0.06666667,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.26666668,0.26666668,0.33333334,0.26666668,0.26666668,0.26666668,0.4,0.53333336,0.4,0.46666667,0.33333334,0.33333334,0.33333334,0.2,0.4,0.6,0.4,0.4,0.4,0.6,0.6,,0.53333336,0.46666667,0.46666667,0.46666667,0.6666667,0.6,0.6666667,0.8666667,0.8,0.8666667,0.73333335,0.6666667,0.73333335,0.73333335,0.93333334,0.8,0.93333334,0.8666667,0.8,0.8666667,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.06666667,0.0,0.0,0.06666667,0.06666667,0.13333334,0.06666667,0.0,0.13333334,0.0,0.06666667,0.13333334,0.06666667,0.06666667,0.13333334,0.13333334,0.06666667,0.13333334,0.06666667,0.2,0.06666667,0.13333334,0.2,0.13333334,0.13333334,0.26666668,0.26666668,0.2,0.4,0.2,0.13333334,0.26666668,0.2,0.2,0.33333334,0.13333334,0.26666668,0.33333334,0.4,0.46666667,0.4,0.33333334,0.46666667,0.46666667,0.46666667,0.4,0.53333336,0.46666667,0.46666667,0.4,0.33333334,0.53333336,0.46666667,,0.53333336,0.46666667,0.53333336,0.6,0.73333335,0.6666667,0.6,0.6,0.53333336,0.73333335,1.0,0.73333335,0.73333335,0.73333335,0.93333334,0.6666667,0.8,0.93333334,0.8666667,0.8,1.0,1.0
+SAP-RPT-OSS (default),0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.13333334,0.06666667,0.06666667,0.13333334,0.13333334,0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.26666668,0.2,0.26666668,0.26666668,0.2,0.13333334,0.13333334,0.2,0.26666668,0.13333334,0.26666668,0.26666668,0.26666668,0.4,0.4,0.4,0.33333334,0.26666668,0.26666668,0.4,0.4,0.46666667,0.33333334,0.26666668,0.4,0.33333334,0.53333336,0.53333336,0.46666667,,0.46666667,0.6,0.73333335,0.73333335,0.6666667,0.6,0.73333335,0.6666667,0.8666667,0.8666667,0.8,0.8,0.8,0.8666667,0.8,0.73333335,0.8666667,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.06666667,0.13333334,0.13333334,0.06666667,0.0,0.06666667,0.2,0.06666667,0.06666667,0.0,0.0,0.06666667,0.2,0.2,0.13333334,0.06666667,0.2,0.26666668,0.2,0.33333334,0.26666668,0.26666668,0.26666668,0.2,0.33333334,0.06666667,0.26666668,0.2,0.26666668,0.2,0.4,0.46666667,0.4,0.46666667,0.4,0.2,0.33333334,0.33333334,0.33333334,0.53333336,0.46666667,0.26666668,0.46666667,0.4,0.53333336,0.53333336,0.53333336,0.53333336,,0.6,0.53333336,0.6666667,0.6,0.6,0.53333336,0.6666667,0.8,0.8,0.8666667,0.8666667,0.6666667,0.8666667,0.73333335,0.6,0.8666667,1.0,0.8666667,0.93333334,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.06666667,0.0,0.0,0.2,0.0,0.2,0.0,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.06666667,0.06666667,0.13333334,0.26666668,0.0,0.2,0.33333334,0.13333334,0.33333334,0.06666667,0.2,0.2,0.26666668,0.06666667,0.46666667,0.2,0.46666667,0.33333334,0.26666668,0.33333334,0.4,0.46666667,0.53333336,0.46666667,0.4,0.4,,0.6,0.6,0.6666667,0.6666667,0.6666667,0.73333335,0.8,0.6,0.73333335,0.73333335,0.73333335,0.8,0.8,0.73333335,0.8666667,0.93333334,1.0,0.93333334,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.0,0.0,0.06666667,0.06666667,0.13333334,0.13333334,0.13333334,0.06666667,0.06666667,0.0,0.06666667,0.0,0.13333334,0.13333334,0.13333334,0.2,0.13333334,0.26666668,0.06666667,0.33333334,0.33333334,0.4,0.2,0.26666668,0.2,0.33333334,0.26666668,0.33333334,0.2,0.33333334,0.33333334,0.33333334,0.33333334,0.4,0.4,0.33333334,0.4,0.26666668,0.46666667,0.4,,0.4,0.6666667,0.6666667,1.0,0.6666667,0.6,0.53333336,0.46666667,0.46666667,0.73333335,0.6666667,0.73333335,1.0,0.6666667,0.46666667,0.73333335,0.6666667,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.0,0.06666667,0.06666667,0.06666667,0.0,0.06666667,0.0,0.13333334,0.06666667,0.13333334,0.06666667,0.06666667,0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.2,0.06666667,0.06666667,0.2,0.26666668,0.26666668,0.26666668,0.26666668,0.0,0.2,0.4,0.33333334,0.26666668,0.46666667,0.26666668,0.26666668,0.26666668,0.2,0.46666667,0.4,0.26666668,0.26666668,0.33333334,0.4,0.6,,0.6,0.53333336,0.6,0.46666667,0.53333336,0.8666667,0.6666667,0.6666667,0.6666667,0.73333335,0.6666667,0.6666667,0.73333335,0.8666667,0.8,0.93333334,1.0
+xRFM (default),0.06666667,0.0,0.0,0.06666667,0.06666667,0.06666667,0.13333334,0.0,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.13333334,0.06666667,0.06666667,0.0,0.13333334,0.06666667,0.06666667,0.06666667,0.13333334,0.2,0.13333334,0.13333334,0.06666667,0.06666667,0.0,0.06666667,0.13333334,0.06666667,0.0,0.2,0.2,0.2,0.2,0.4,0.26666668,0.2,0.13333334,0.2,0.26666668,0.33333334,0.2,0.33333334,0.2,0.06666667,0.33333334,0.33333334,0.4,0.46666667,0.06666667,0.33333334,0.33333334,0.33333334,0.4,0.33333334,0.33333334,0.4,,0.4,0.46666667,0.46666667,0.46666667,0.46666667,0.53333336,0.53333336,0.4,0.46666667,0.6666667,0.53333336,0.46666667,0.53333336,0.6,0.46666667,0.8
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.13333334,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.13333334,0.06666667,0.06666667,0.13333334,0.06666667,0.2,0.06666667,0.13333334,0.16666667,0.26666668,0.13333334,0.06666667,0.26666668,0.2,0.13333334,0.26666668,0.06666667,0.13333334,0.33333334,0.33333334,0.16666667,0.33333334,0.36666667,0.33333334,0.13333334,0.4,0.4,0.4,0.33333334,0.33333334,0.46666667,0.6,,0.4,0.56666666,0.53333336,0.6666667,0.6,0.6,0.73333335,0.6666667,0.56666666,0.73333335,0.6666667,0.6666667,0.6666667,0.73333335,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.13333334,0.0,0.13333334,0.06666667,0.0,0.0,0.06666667,0.0,0.13333334,0.13333334,0.06666667,0.06666667,0.13333334,0.26666668,0.06666667,0.33333334,0.33333334,0.33333334,0.06666667,0.13333334,0.13333334,0.26666668,0.26666668,0.26666668,0.06666667,0.26666668,0.33333334,0.33333334,0.33333334,0.4,0.26666668,0.2,0.4,0.26666668,0.46666667,0.33333334,0.0,0.4,0.53333336,0.6,,0.6666667,0.53333336,0.53333336,0.46666667,0.46666667,0.53333336,0.6666667,0.6,0.93333334,0.6666667,0.46666667,0.53333336,0.6666667,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.13333334,0.06666667,0.06666667,0.06666667,0.0,0.13333334,0.06666667,0.0,0.16666667,0.26666668,0.26666668,0.06666667,0.2,0.2,0.13333334,0.2,0.06666667,0.06666667,0.26666668,0.26666668,0.1,0.26666668,0.23333333,0.26666668,0.13333334,0.46666667,0.33333334,0.33333334,0.26666668,0.33333334,0.53333336,0.53333336,0.43333334,0.33333334,,0.6666667,0.6,0.53333336,0.53333336,0.6666667,0.6666667,0.6333333,0.6666667,0.6666667,0.73333335,0.6666667,0.73333335,1.0
+KNN (tuned + ensembled),0.06666667,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.13333334,0.06666667,0.0,0.13333334,0.06666667,0.0,0.0,0.06666667,0.13333334,0.0,0.06666667,0.06666667,0.13333334,0.2,0.06666667,0.33333334,0.13333334,0.06666667,0.13333334,0.13333334,0.13333334,0.46666667,0.2,0.06666667,0.2,0.06666667,0.46666667,0.26666668,0.26666668,0.13333334,0.2,0.2,0.4,0.46666667,0.53333336,0.46666667,0.46666667,0.33333334,,0.53333336,0.46666667,0.46666667,0.53333336,0.8,0.53333336,0.6666667,0.8,0.6,1.0,0.8666667,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.13333334,0.0,0.13333334,0.06666667,0.13333334,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.06666667,0.13333334,0.2,0.06666667,0.06666667,0.13333334,0.13333334,0.2,0.06666667,0.13333334,0.0,0.06666667,0.2,0.2,0.2,0.33333334,0.2,0.06666667,0.26666668,0.13333334,0.33333334,0.33333334,0.0,0.13333334,0.2,0.4,0.46666667,0.13333334,0.53333336,0.33333334,0.46666667,0.4,0.46666667,,0.6,0.6,0.46666667,0.73333335,0.6,0.6666667,0.73333335,0.73333335,0.6666667,0.8,0.8666667
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.13333334,0.13333334,0.06666667,0.13333334,0.06666667,0.0,0.06666667,0.0,0.13333334,0.0,0.0,0.06666667,0.0,0.33333334,0.13333334,0.2,0.13333334,0.0,0.06666667,0.13333334,0.13333334,0.4,0.06666667,0.06666667,0.13333334,0.13333334,0.4,0.26666668,0.26666668,0.2,0.13333334,0.26666668,0.53333336,0.33333334,0.46666667,0.4,0.53333336,0.46666667,0.53333336,0.4,,0.6666667,0.53333336,0.73333335,0.6666667,0.53333336,0.73333335,1.0,0.8,0.8666667,0.8
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.13333334,0.13333334,0.06666667,0.13333334,0.06666667,0.0,0.06666667,0.0,0.13333334,0.0,0.0,0.0,0.06666667,0.33333334,0.13333334,0.2,0.13333334,0.0,0.06666667,0.13333334,0.06666667,0.4,0.06666667,0.06666667,0.13333334,0.13333334,0.4,0.26666668,0.26666668,0.2,0.13333334,0.26666668,0.53333336,0.33333334,0.46666667,0.4,0.53333336,0.46666667,0.53333336,0.4,0.33333334,,0.53333336,0.73333335,0.6666667,0.53333336,0.73333335,1.0,0.8,0.8666667,0.8
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.13333334,0.06666667,0.13333334,0.0,0.13333334,0.26666668,0.13333334,0.06666667,0.26666668,0.06666667,0.13333334,0.2,0.0,0.06666667,0.13333334,0.33333334,0.13333334,0.33333334,0.26666668,0.13333334,0.06666667,0.26666668,0.2,0.33333334,0.26666668,0.26666668,0.33333334,0.6,0.26666668,0.46666667,0.33333334,0.46666667,0.53333336,0.46666667,0.46666667,,0.6666667,0.53333336,0.73333335,0.6666667,0.46666667,0.6666667,0.6666667,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.13333334,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.2,0.13333334,0.0,0.2,0.0,0.06666667,0.0,0.06666667,0.2,0.0,0.13333334,0.13333334,0.06666667,0.2,0.2,0.2,0.06666667,0.13333334,0.06666667,0.2,0.2,0.33333334,0.2,0.2,0.2,0.2,0.33333334,0.2,0.06666667,0.13333334,0.13333334,0.2,0.33333334,0.26666668,0.53333336,0.33333334,0.33333334,0.33333334,0.2,0.26666668,0.26666668,0.26666668,0.33333334,,0.46666667,0.46666667,0.9,0.33333334,0.4,0.8666667,0.6666667
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.0,0.13333334,0.06666667,0.0,0.16666667,0.2,0.2,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.06666667,0.06666667,0.26666668,0.2,0.1,0.2,0.16666667,0.26666668,0.06666667,0.33333334,0.2,0.26666668,0.2,0.26666668,0.33333334,0.33333334,0.43333334,0.4,0.36666667,0.46666667,0.4,0.33333334,0.33333334,0.46666667,0.53333336,,0.6,0.53333336,0.53333336,0.6666667,0.6,0.8666667
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.06666667,0.0,0.13333334,0.06666667,0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.13333334,0.26666668,0.2,0.0,0.0,0.06666667,0.2,0.13333334,0.13333334,0.06666667,0.13333334,0.33333334,0.2,0.33333334,0.2,0.13333334,0.13333334,0.2,0.26666668,0.4,0.26666668,0.0,0.33333334,0.46666667,0.26666668,0.06666667,0.33333334,0.33333334,0.33333334,0.46666667,0.46666667,0.26666668,0.53333336,0.4,,0.53333336,0.46666667,0.4,0.53333336,0.8666667
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.06666667,0.2,0.13333334,0.0,0.2,0.0,0.0,0.0,0.06666667,0.13333334,0.0,0.2,0.13333334,0.06666667,0.2,0.2,0.2,0.0,0.13333334,0.06666667,0.06666667,0.2,0.33333334,0.13333334,0.2,0.13333334,0.2,0.33333334,0.2,0.06666667,0.13333334,0.13333334,0.13333334,0.33333334,0.26666668,0.53333336,0.33333334,0.33333334,0.33333334,0.2,0.26666668,0.26666668,0.26666668,0.33333334,0.1,0.46666667,0.46666667,,0.33333334,0.4,0.8,0.6666667
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.13333334,0.0,0.33333334,0.0,0.0,0.0,0.0,0.33333334,0.13333334,0.13333334,0.0,0.0,0.06666667,0.53333336,0.13333334,0.46666667,0.33333334,0.53333336,0.26666668,0.4,0.26666668,0.0,0.0,0.53333336,0.6666667,0.46666667,0.53333336,0.6666667,,0.6666667,0.6666667,0.73333335
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.2,0.0,0.2,0.0,0.0,0.0,0.06666667,0.0,0.33333334,0.06666667,0.0,0.06666667,0.0,0.33333334,0.0,0.2,0.0,0.13333334,0.0,0.26666668,0.2,0.4,0.33333334,0.46666667,0.33333334,0.0,0.33333334,0.2,0.2,0.33333334,0.6,0.33333334,0.6,0.6,0.33333334,,0.6,1.0
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.0,0.06666667,0.0,0.0,0.0,0.06666667,0.13333334,0.0,0.06666667,0.0,0.0,0.2,0.06666667,0.13333334,0.0,0.0,0.0,0.0,0.06666667,0.26666668,0.06666667,0.0,0.06666667,0.06666667,0.26666668,0.0,0.0,0.0,0.06666667,0.06666667,0.33333334,0.06666667,0.53333336,0.26666668,0.33333334,0.26666668,0.13333334,0.2,0.13333334,0.13333334,0.33333334,0.13333334,0.4,0.46666667,0.2,0.33333334,0.4,,0.6666667
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.06666667,0.0,0.06666667,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.13333334,0.2,0.2,0.0,0.33333334,0.13333334,0.13333334,0.33333334,0.26666668,0.0,0.33333334,
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_small/leaderboard_overview_explorer.html
index 7af155f7822f08512f691a10d5b4d5eefdc8483f..f59e95477403c64a361ccc2b493835e6f46edf3f 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | All Tasks | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_all/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..99950f3bedd94b3d6e9ed2ba6b394d245c70eacd
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_all/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1779.0,"score":0.841,"rank":6.14,"hrank":1.53,"imp":4.507,"train_time":45.35,"predict_time":5.343,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+235\/-121","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1622.0,"score":0.596,"rank":11.9,"hrank":5.37,"imp":10.009,"train_time":8.57,"predict_time":0.411,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+100\/-53","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1598.0,"score":0.595,"rank":13.11,"hrank":5.5,"imp":8.82,"train_time":345.57,"predict_time":5.168,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-75","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1594.0,"score":0.571,"rank":13.29,"hrank":5.47,"imp":9.775,"train_time":2320.56,"predict_time":8.956,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-89","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1593.0,"score":0.578,"rank":13.36,"hrank":6.75,"imp":11.104,"train_time":7.07,"predict_time":0.555,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-59","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1560.0,"score":0.503,"rank":15.14,"hrank":9.84,"imp":11.737,"train_time":6.58,"predict_time":0.586,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-57","imputed":false},{"position":6,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1538.0,"score":0.531,"rank":16.47,"hrank":6.74,"imp":11.565,"train_time":3.89,"predict_time":0.15,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-88","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1536.0,"score":0.511,"rank":16.61,"hrank":8.92,"imp":11.953,"train_time":2320.56,"predict_time":0.941,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-68","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1473.0,"score":0.402,"rank":20.81,"hrank":13.07,"imp":14.163,"train_time":3855.77,"predict_time":24.914,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-50","imputed":false},{"position":9,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1459.0,"score":0.375,"rank":21.78,"hrank":15.05,"imp":14.409,"train_time":3351.52,"predict_time":1.558,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-51","imputed":false},{"position":10,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1442.0,"score":0.393,"rank":23.1,"hrank":12.01,"imp":14.868,"train_time":3560.23,"predict_time":49.196,"imputed_pct":8.33,"hardware":"GPU","elo_ci":"+101\/-72","imputed":true},{"position":11,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1431.0,"score":0.399,"rank":23.94,"hrank":8.78,"imp":14.676,"train_time":8061.74,"predict_time":364.097,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-69","imputed":false},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1416.0,"score":0.359,"rank":25.07,"hrank":11.91,"imp":15.432,"train_time":5.1,"predict_time":0.187,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-61","imputed":false},{"position":13,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1406.0,"score":0.343,"rank":25.9,"hrank":15.06,"imp":15.285,"train_time":3855.77,"predict_time":0.954,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-73","imputed":false},{"position":14,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1405.0,"score":0.33,"rank":25.96,"hrank":14.13,"imp":15.961,"train_time":128.84,"predict_time":2.999,"imputed_pct":8.33,"hardware":"GPU","elo_ci":"+82\/-48","imputed":true},{"position":15,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1399.0,"score":0.354,"rank":26.47,"hrank":8.79,"imp":14.406,"train_time":34.03,"predict_time":7.205,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-83","imputed":false},{"position":16,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1390.0,"score":0.354,"rank":27.19,"hrank":13.91,"imp":15.663,"train_time":8061.74,"predict_time":39.456,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-68","imputed":false},{"position":17,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1372.0,"score":0.315,"rank":28.72,"hrank":13.41,"imp":15.985,"train_time":2564.23,"predict_time":3.453,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-60","imputed":false},{"position":18,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1364.0,"score":0.258,"rank":29.42,"hrank":18.76,"imp":16.436,"train_time":22680.19,"predict_time":529.126,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-48","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1358.0,"score":0.22,"rank":29.92,"hrank":23.65,"imp":17.024,"train_time":833.09,"predict_time":2.182,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-42","imputed":false},{"position":20,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1354.0,"score":0.247,"rank":30.25,"hrank":17.21,"imp":15.902,"train_time":1827.95,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-48","imputed":false},{"position":21,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1348.0,"score":0.226,"rank":30.78,"hrank":23.86,"imp":16.357,"train_time":1827.95,"predict_time":0.391,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-57","imputed":false},{"position":22,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1346.0,"score":0.273,"rank":30.9,"hrank":18.12,"imp":16.965,"train_time":3560.23,"predict_time":0.519,"imputed_pct":8.33,"hardware":"GPU","elo_ci":"+78\/-66","imputed":true},{"position":23,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1342.0,"score":0.276,"rank":31.26,"hrank":17.29,"imp":16.602,"train_time":2564.23,"predict_time":0.453,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-64","imputed":false},{"position":24,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1338.0,"score":0.255,"rank":31.65,"hrank":20.23,"imp":17.336,"train_time":2.54,"predict_time":0.058,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-54","imputed":false},{"position":25,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1312.0,"score":0.23,"rank":33.94,"hrank":19.79,"imp":17.633,"train_time":1346.82,"predict_time":2.149,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-54","imputed":false},{"position":26,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1311.0,"score":0.181,"rank":34.04,"hrank":25.63,"imp":17.932,"train_time":886.81,"predict_time":1.98,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-44","imputed":false},{"position":27,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1300.0,"score":0.243,"rank":35.03,"hrank":15.5,"imp":17.631,"train_time":4.98,"predict_time":0.511,"imputed_pct":8.33,"hardware":"GPU","elo_ci":"+79\/-63","imputed":true},{"position":28,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1297.0,"score":0.174,"rank":35.28,"hrank":28.43,"imp":18.007,"train_time":833.09,"predict_time":0.435,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-48","imputed":false},{"position":29,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1297.0,"score":0.171,"rank":35.31,"hrank":16.73,"imp":18.05,"train_time":886.81,"predict_time":0.318,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-53","imputed":false},{"position":30,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1296.0,"score":0.224,"rank":35.36,"hrank":15.24,"imp":18.723,"train_time":6343.61,"predict_time":8.545,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-63","imputed":false},{"position":31,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1294.0,"score":0.154,"rank":35.56,"hrank":30.02,"imp":18.276,"train_time":9.04,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-40","imputed":false},{"position":32,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1292.0,"score":0.251,"rank":35.75,"hrank":17.83,"imp":18.053,"train_time":51.79,"predict_time":39.921,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-77","imputed":false},{"position":33,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1286.0,"score":0.199,"rank":36.25,"hrank":20.08,"imp":18.196,"train_time":6343.61,"predict_time":0.501,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-59","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1279.0,"score":0.169,"rank":36.88,"hrank":28.76,"imp":19.392,"train_time":1180.0,"predict_time":0.457,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-54","imputed":false},{"position":35,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1279.0,"score":0.166,"rank":36.92,"hrank":29.16,"imp":17.923,"train_time":22680.19,"predict_time":126.301,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-58","imputed":false},{"position":36,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1277.0,"score":0.239,"rank":37.07,"hrank":17.03,"imp":19.578,"train_time":16.54,"predict_time":1.254,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-74","imputed":false},{"position":37,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1270.0,"score":0.211,"rank":37.71,"hrank":13.9,"imp":19.114,"train_time":7.56,"predict_time":1.493,"imputed_pct":27.78,"hardware":"GPU","elo_ci":"+85\/-98","imputed":true},{"position":38,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1269.0,"score":0.203,"rank":37.83,"hrank":24.03,"imp":20.03,"train_time":269.21,"predict_time":1.881,"imputed_pct":22.22,"hardware":"GPU","elo_ci":"+65\/-68","imputed":true},{"position":39,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1266.0,"score":0.19,"rank":38.12,"hrank":22.63,"imp":18.67,"train_time":11.22,"predict_time":0.393,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-62","imputed":false},{"position":40,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1262.0,"score":0.173,"rank":38.47,"hrank":16.88,"imp":18.569,"train_time":1346.82,"predict_time":0.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-56","imputed":false},{"position":41,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1252.0,"score":0.128,"rank":39.33,"hrank":26.52,"imp":19.606,"train_time":2241.86,"predict_time":0.152,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-56","imputed":false},{"position":42,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1245.0,"score":0.128,"rank":40.0,"hrank":28.0,"imp":19.064,"train_time":16.61,"predict_time":4.299,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+54\/-61","imputed":false},{"position":43,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1243.0,"score":0.143,"rank":40.17,"hrank":24.37,"imp":19.902,"train_time":1180.0,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-70","imputed":false},{"position":44,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1228.0,"score":0.133,"rank":41.53,"hrank":33.89,"imp":19.229,"train_time":3694.49,"predict_time":2.137,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-59","imputed":false},{"position":45,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1217.0,"score":0.105,"rank":42.58,"hrank":28.92,"imp":20.294,"train_time":2241.86,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-59","imputed":false},{"position":46,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1202.0,"score":0.115,"rank":43.88,"hrank":16.64,"imp":22.391,"train_time":14.91,"predict_time":0.323,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-71","imputed":false},{"position":47,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1194.0,"score":0.118,"rank":44.67,"hrank":29.49,"imp":21.548,"train_time":841.6,"predict_time":1.393,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-77","imputed":false},{"position":48,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1192.0,"score":0.091,"rank":44.78,"hrank":37.05,"imp":21.535,"train_time":5.52,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-70","imputed":false},{"position":49,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1191.0,"score":0.102,"rank":44.88,"hrank":18.71,"imp":20.887,"train_time":7.6,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-66","imputed":false},{"position":50,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1180.0,"score":0.121,"rank":45.93,"hrank":32.65,"imp":20.782,"train_time":3694.49,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-69","imputed":false},{"position":51,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1175.0,"score":0.101,"rank":46.38,"hrank":33.53,"imp":21.926,"train_time":1666.93,"predict_time":7.822,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-84","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1167.0,"score":0.093,"rank":47.01,"hrank":30.21,"imp":22.6,"train_time":841.6,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-76","imputed":false},{"position":53,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1158.0,"score":0.195,"rank":47.88,"hrank":10.6,"imp":24.632,"train_time":0.68,"predict_time":0.108,"imputed_pct":77.78,"hardware":"GPU","elo_ci":"+110\/-109","imputed":true},{"position":54,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1154.0,"score":0.185,"rank":48.18,"hrank":15.01,"imp":24.923,"train_time":0.68,"predict_time":0.108,"imputed_pct":77.78,"hardware":"GPU","elo_ci":"+106\/-107","imputed":true},{"position":55,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1139.0,"score":0.092,"rank":49.5,"hrank":39.28,"imp":23.168,"train_time":575.62,"predict_time":1.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-75","imputed":false},{"position":56,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1134.0,"score":0.074,"rank":49.93,"hrank":41.1,"imp":21.778,"train_time":3.63,"predict_time":0.218,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-62","imputed":false},{"position":57,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1126.0,"score":0.046,"rank":50.61,"hrank":43.92,"imp":21.822,"train_time":3.1,"predict_time":0.207,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-60","imputed":false},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1124.0,"score":0.072,"rank":50.83,"hrank":35.24,"imp":23.088,"train_time":1666.93,"predict_time":0.561,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-80","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1114.0,"score":0.147,"rank":51.67,"hrank":17.43,"imp":25.617,"train_time":46972.72,"predict_time":5.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-100","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1113.0,"score":0.141,"rank":51.74,"hrank":18.57,"imp":25.782,"train_time":46972.72,"predict_time":22.64,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-95","imputed":false},{"position":61,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1094.0,"score":0.05,"rank":53.31,"hrank":46.53,"imp":24.099,"train_time":575.62,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-66","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1089.0,"score":0.079,"rank":53.76,"hrank":42.55,"imp":28.619,"train_time":282.83,"predict_time":0.509,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-81","imputed":false},{"position":63,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1075.0,"score":0.039,"rank":54.85,"hrank":47.22,"imp":24.853,"train_time":13.57,"predict_time":2.436,"imputed_pct":22.22,"hardware":"GPU","elo_ci":"+73\/-66","imputed":true},{"position":64,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1057.0,"score":0.054,"rank":56.32,"hrank":30.26,"imp":25.037,"train_time":537.96,"predict_time":157.807,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-94","imputed":false},{"position":65,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1046.0,"score":0.045,"rank":57.11,"hrank":51.75,"imp":30.015,"train_time":282.83,"predict_time":0.162,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-75","imputed":false},{"position":66,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1039.0,"score":0.063,"rank":57.67,"hrank":46.78,"imp":26.334,"train_time":6.09,"predict_time":0.951,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-87","imputed":false},{"position":67,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1033.0,"score":0.039,"rank":58.14,"hrank":48.28,"imp":27.181,"train_time":4.37,"predict_time":0.629,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-86","imputed":false},{"position":68,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1020.0,"score":0.027,"rank":59.11,"hrank":52.39,"imp":25.79,"train_time":12.68,"predict_time":0.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-89","imputed":false},{"position":69,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1018.0,"score":0.103,"rank":59.25,"hrank":17.27,"imp":31.269,"train_time":459.76,"predict_time":5.723,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-140","imputed":false},{"position":70,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1006.0,"score":0.04,"rank":60.11,"hrank":50.65,"imp":28.206,"train_time":1.59,"predict_time":0.149,"imputed_pct":22.22,"hardware":"GPU","elo_ci":"+67\/-74","imputed":true},{"position":71,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.016,"rank":60.51,"hrank":51.27,"imp":28.453,"train_time":0.51,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-84","imputed":false},{"position":72,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":985.0,"score":0.021,"rank":61.56,"hrank":57.21,"imp":26.869,"train_time":5.83,"predict_time":0.571,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-82","imputed":false},{"position":73,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":966.0,"score":0.031,"rank":62.79,"hrank":52.6,"imp":29.215,"train_time":116.33,"predict_time":1.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-97","imputed":false},{"position":74,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":959.0,"score":0.034,"rank":63.26,"hrank":39.49,"imp":34.36,"train_time":38.53,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-93","imputed":false},{"position":75,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":897.0,"score":0.028,"rank":66.88,"hrank":23.67,"imp":34.719,"train_time":520.74,"predict_time":0.491,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-107","imputed":false},{"position":76,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":878.0,"score":0.023,"rank":67.86,"hrank":35.56,"imp":35.192,"train_time":520.74,"predict_time":0.097,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-99","imputed":false},{"position":77,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":832.0,"score":0.008,"rank":70.03,"hrank":65.04,"imp":37.886,"train_time":2.59,"predict_time":0.129,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-134","imputed":false},{"position":78,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":816.0,"score":0.02,"rank":70.72,"hrank":65.61,"imp":34.958,"train_time":116.33,"predict_time":0.075,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-113","imputed":false},{"position":79,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":620.0,"score":0.0,"rank":76.64,"hrank":76.12,"imp":46.801,"train_time":0.37,"predict_time":0.035,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-171","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_small/pareto_front_explorer.html
index 7598a3825eac07a48c6d62b6303a444e0ca0cd58..4b1eaa8503859a4ca897062a08b6ca644355c553 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8806983387,"imp":20.0304257711,"elo":1268.8,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0408851968,"imp":18.2755662794,"elo":1293.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0482592414,"imp":15.9016454099,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3907846602,"imp":16.3574929104,"elo":1347.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0367584655,"imp":21.5351060179,"elo":1192.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0344803412,"imp":19.9024794249,"elo":1243.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4568895686,"imp":19.3917922854,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0170112085,"imp":20.8872522851,"elo":1191.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0168225275,"imp":20.2938184088,"elo":1216.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1523084536,"imp":19.6057625514,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6288369946,"imp":27.1806932435,"elo":1032.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1925559683,"imp":22.6002908671,"elo":1167.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3932122077,"imp":21.548450645,"elo":1193.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5709841633,"imp":26.8694121158,"elo":984.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5609692074,"imp":23.0884574952,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.821672488,"imp":21.9257890902,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2175696189,"imp":21.7779496521,"elo":1134.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.4350380217,"imp":18.0067318871,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.1817652315,"imp":17.0237056996,"elo":1357.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.2046047302,"imp":14.4063805419,"elo":1398.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.9990016855,"imp":15.9608031326,"elo":1405.3,"imputed":true,"imputed_pct":8.3333333333,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3225041576,"imp":22.3906698096,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5011277894,"imp":18.1959056576,"elo":1286.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.5454388038,"imp":18.722900431,"elo":1296.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.1083048852,"imp":24.9231375518,"elo":1154.2,"imputed":true,"imputed_pct":77.7777777778,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.1083048852,"imp":24.63189065,"elo":1157.7,"imputed":true,"imputed_pct":77.7777777778,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.4356700948,"imp":24.8534307845,"elo":1075.3,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0412162752,"imp":34.3597792767,"elo":958.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1615795606,"imp":30.0151470468,"elo":1046.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5087521152,"imp":28.6193939083,"elo":1088.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852899147,"imp":28.4528593383,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468843405,"imp":24.0990889348,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"imp":23.1683444261,"elo":1139.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.2987548984,"imp":19.0638913156,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.954021458,"imp":15.2846974261,"elo":1406.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":24.9138452893,"imp":14.1632938136,"elo":1472.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5863334021,"imp":11.7372826032,"elo":1560.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9412054434,"imp":11.9527608413,"elo":1535.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9558346011,"imp":9.7747885597,"elo":1594.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2543929634,"imp":19.5784498083,"elo":1277.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"imp":18.0529391013,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.456011971,"imp":15.6634990242,"elo":1390.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":364.0974768275,"imp":14.6756666193,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1868636356,"imp":15.4316960129,"elo":1416.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.3427919745,"imp":4.5072632125,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1488925237,"imp":28.2057073084,"elo":1005.7,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.4929299615,"imp":19.1139917298,"elo":1270.2,"imputed":true,"imputed_pct":27.7777777778,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1502327458,"imp":11.5648282269,"elo":1538.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.3927969172,"imp":18.6703613062,"elo":1265.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4533957018,"imp":16.6017798432,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.4526368273,"imp":15.9845603384,"elo":1371.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5551544404,"imp":11.1043060414,"elo":1592.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4108441184,"imp":10.0090125197,"elo":1622.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5110070735,"imp":17.6307022508,"elo":1299.8,"imputed":true,"imputed_pct":8.3333333333,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.519347753,"imp":16.9647441906,"elo":1346.4,"imputed":true,"imputed_pct":8.3333333333,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":49.1960479984,"imp":14.8681921787,"elo":1441.7,"imputed":true,"imputed_pct":8.3333333333,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"imp":31.2686745729,"elo":1017.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3103758735,"imp":25.6166358611,"elo":1113.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.63963438,"imp":25.7823958993,"elo":1113.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0581109952,"imp":17.3359293541,"elo":1337.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1547654046,"imp":25.7901171342,"elo":1019.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305659975,"imp":20.7819357362,"elo":1179.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1373326275,"imp":19.2286027907,"elo":1228.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2070918304,"imp":21.8216101597,"elo":1126.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3177561745,"imp":18.0499754978,"elo":1296.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.980191462,"imp":17.9321550563,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":157.807153647,"imp":25.03665522,"elo":1056.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":126.3012190283,"imp":17.9234490799,"elo":1278.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":529.125864783,"imp":16.435530659,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9508145682,"imp":26.3335014039,"elo":1039.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1034549609,"imp":18.5693308214,"elo":1261.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.1486246717,"imp":17.6331563152,"elo":1311.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8806983387,"x_train":269.2082934936,"imp":20.0304257711,"elo":1268.8,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0408851968,"x_train":9.0359136176,"imp":18.2755662794,"elo":1293.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0482592414,"x_train":1827.947236699,"imp":15.9016454099,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3907846602,"x_train":1827.947236699,"imp":16.3574929104,"elo":1347.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0367584655,"x_train":5.5160770154,"imp":21.5351060179,"elo":1192.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0344803412,"x_train":1179.9999181447,"imp":19.9024794249,"elo":1243.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4568895686,"x_train":1179.9999181447,"imp":19.3917922854,"elo":1279.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0170112085,"x_train":7.5965245684,"imp":20.8872522851,"elo":1191.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0168225275,"x_train":2241.8559420912,"imp":20.2938184088,"elo":1216.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1523084536,"x_train":2241.8559420912,"imp":19.6057625514,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6288369946,"x_train":4.3684404027,"imp":27.1806932435,"elo":1032.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1925559683,"x_train":841.5972483589,"imp":22.6002908671,"elo":1167.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3932122077,"x_train":841.5972483589,"imp":21.548450645,"elo":1193.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5709841633,"x_train":5.8349009345,"imp":26.8694121158,"elo":984.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5609692074,"x_train":1666.9308833948,"imp":23.0884574952,"elo":1123.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.821672488,"x_train":1666.9308833948,"imp":21.9257890902,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2175696189,"x_train":3.6257108294,"imp":21.7779496521,"elo":1134.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.4350380217,"x_train":833.0924159486,"imp":18.0067318871,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":2.1817652315,"x_train":833.0924159486,"imp":17.0237056996,"elo":1357.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":7.2046047302,"x_train":34.0310873482,"imp":14.4063805419,"elo":1398.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.9990016855,"x_train":128.8352109042,"imp":15.9608031326,"elo":1405.3,"imputed":true,"imputed_pct":8.3333333333,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3225041576,"x_train":14.9096333145,"imp":22.3906698096,"elo":1202.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5011277894,"x_train":6343.6148094754,"imp":18.1959056576,"elo":1286.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.5454388038,"x_train":6343.6148094754,"imp":18.722900431,"elo":1296.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.1083048852,"x_train":0.6786165268,"imp":24.9231375518,"elo":1154.2,"imputed":true,"imputed_pct":77.7777777778,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.1083048852,"x_train":0.6786165268,"imp":24.63189065,"elo":1157.7,"imputed":true,"imputed_pct":77.7777777778,"variant":"Default"},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.4356700948,"x_train":13.5688156132,"imp":24.8534307845,"elo":1075.3,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0412162752,"x_train":38.5275966029,"imp":34.3597792767,"elo":958.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1615795606,"x_train":282.8267356163,"imp":30.0151470468,"elo":1046.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5087521152,"x_train":282.8267356163,"imp":28.6193939083,"elo":1088.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852899147,"x_train":0.5076178694,"imp":28.4528593383,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468843405,"x_train":575.6224152834,"imp":24.0990889348,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"x_train":575.6224152834,"imp":23.1683444261,"elo":1139.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.2987548984,"x_train":16.6080899164,"imp":19.0638913156,"elo":1245.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.954021458,"x_train":3855.7666751446,"imp":15.2846974261,"elo":1406.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":24.9138452893,"x_train":3855.7666751446,"imp":14.1632938136,"elo":1472.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5863334021,"x_train":6.5779030898,"imp":11.7372826032,"elo":1560.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9412054434,"x_train":2320.5586596122,"imp":11.9527608413,"elo":1535.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9558346011,"x_train":2320.5586596122,"imp":9.7747885597,"elo":1594.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2543929634,"x_train":16.5391218381,"imp":19.5784498083,"elo":1277.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"x_train":51.7864862661,"imp":18.0529391013,"elo":1291.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.456011971,"x_train":8061.743825954,"imp":15.6634990242,"elo":1390.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":364.0974768275,"x_train":8061.743825954,"imp":14.6756666193,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1868636356,"x_train":5.1020093375,"imp":15.4316960129,"elo":1416.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.3427919745,"x_train":45.3524743275,"imp":4.5072632125,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1488925237,"x_train":1.589504217,"imp":28.2057073084,"elo":1005.7,"imputed":true,"imputed_pct":22.2222222222,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.4929299615,"x_train":7.5554129923,"imp":19.1139917298,"elo":1270.2,"imputed":true,"imputed_pct":27.7777777778,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1502327458,"x_train":3.8948553652,"imp":11.5648282269,"elo":1538.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.3927969172,"x_train":11.2200428783,"imp":18.6703613062,"elo":1265.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4533957018,"x_train":2564.2291668214,"imp":16.6017798432,"elo":1342.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.4526368273,"x_train":2564.2291668214,"imp":15.9845603384,"elo":1371.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5551544404,"x_train":7.0738556742,"imp":11.1043060414,"elo":1592.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4108441184,"x_train":8.5714229456,"imp":10.0090125197,"elo":1622.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5110070735,"x_train":4.9840988627,"imp":17.6307022508,"elo":1299.8,"imputed":true,"imputed_pct":8.3333333333,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.519347753,"x_train":3560.2284047226,"imp":16.9647441906,"elo":1346.4,"imputed":true,"imputed_pct":8.3333333333,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":49.1960479984,"x_train":3560.2284047226,"imp":14.8681921787,"elo":1441.7,"imputed":true,"imputed_pct":8.3333333333,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"x_train":459.757659874,"imp":31.2686745729,"elo":1017.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.3103758735,"x_train":46972.7202570992,"imp":25.6166358611,"elo":1113.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.63963438,"x_train":46972.7202570992,"imp":25.7823958993,"elo":1113.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0581109952,"x_train":2.5372143019,"imp":17.3359293541,"elo":1337.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1547654046,"x_train":12.6783036886,"imp":25.7901171342,"elo":1019.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305659975,"x_train":3694.4938905556,"imp":20.7819357362,"elo":1179.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1373326275,"x_train":3694.4938905556,"imp":19.2286027907,"elo":1228.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2070918304,"x_train":3.0974213034,"imp":21.8216101597,"elo":1126.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3177561745,"x_train":886.8089747599,"imp":18.0499754978,"elo":1296.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.980191462,"x_train":886.8089747599,"imp":17.9321550563,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":157.807153647,"x_train":537.959331743,"imp":25.03665522,"elo":1056.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":126.3012190283,"x_train":22680.1927917466,"imp":17.9234490799,"elo":1278.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":529.125864783,"x_train":22680.1927917466,"imp":16.435530659,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9508145682,"x_train":6.0878006956,"imp":26.3335014039,"elo":1039.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1034549609,"x_train":1346.8150776785,"imp":18.5693308214,"elo":1261.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.1486246717,"x_train":1346.8150776785,"imp":17.6331563152,"elo":1311.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_small/tuning_trajectories_explorer.html
index 2e0eabc12f78c4ec6c33e9158467df79f524f754..779f677206cb18cb716a3c24535eac46dbbd82b7 100644
--- a/data/imputation_yes/splits_lite/tasks_all/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_all/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":269.2082934936,"imp":20.0416628001,"elo":1284.1,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":9.0359136176,"imp":18.2867004257,"elo":1307.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":21.9992777967,"imp":17.5241517957,"elo":1337.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":53.8141065802,"imp":17.046472113,"elo":1356.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":96.9989733937,"imp":16.797349382,"elo":1365.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":231.1646909722,"imp":16.6154192537,"elo":1368.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":468.6200877614,"imp":16.6045136365,"elo":1364.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":950.6643741645,"imp":16.4442671888,"elo":1362.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1384.6574015383,"imp":16.4367871001,"elo":1357.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1827.947236699,"imp":16.3586667057,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.5160770154,"imp":21.5454621825,"elo":1201.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.4571846403,"imp":20.8567308321,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":28.0250922704,"imp":20.3736954266,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":57.1334404694,"imp":20.139547438,"elo":1270.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":145.0536005577,"imp":19.761846164,"elo":1283.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":292.2184372302,"imp":19.5757485361,"elo":1292.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":586.8028092523,"imp":19.4675721168,"elo":1294.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":881.2145061251,"imp":19.3983671988,"elo":1297.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1179.9999181447,"imp":19.3981292293,"elo":1296.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.5965245684,"imp":20.8975434258,"elo":1197.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":20.5778802427,"imp":20.5762208261,"elo":1213.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":58.0053529444,"imp":20.2325494376,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":114.2580939201,"imp":20.1142358534,"elo":1242.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":281.7059626239,"imp":19.9775932818,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":567.1802440047,"imp":19.8259050438,"elo":1257.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1120.6313986655,"imp":19.6807553778,"elo":1264.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1665.6646846813,"imp":19.6440045451,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2241.8559420912,"imp":19.608987689,"elo":1268.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.3684404027,"imp":27.1914700939,"elo":1039.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":8.974442795,"imp":25.6146104924,"elo":1066.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":23.2753987371,"imp":23.3287526027,"elo":1123.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":45.9919048085,"imp":22.6824940376,"elo":1152.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":110.3208397341,"imp":21.990945112,"elo":1181.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":212.5071471057,"imp":21.814698821,"elo":1189.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":419.021686722,"imp":21.7168803028,"elo":1193.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":627.2476408835,"imp":21.5934476972,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":841.5972483589,"imp":21.5990262923,"elo":1201.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.8349009345,"imp":26.8785744883,"elo":985.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.4449316294,"imp":26.1049653205,"elo":1027.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":42.2187698316,"imp":24.3451471633,"elo":1094.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":78.730916991,"imp":23.5973668956,"elo":1122.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":201.7091267248,"imp":22.6808412586,"elo":1146.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":410.0601179712,"imp":22.2640766517,"elo":1163.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":823.1441900238,"imp":22.0723475153,"elo":1170.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1238.4424229224,"imp":21.9376825834,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1666.9308833948,"imp":21.9461018446,"elo":1182.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":3.6257108294,"imp":21.7884658415,"elo":1134.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":8.2503296642,"imp":19.6725465346,"elo":1233.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":20.5596206802,"imp":18.3350057151,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":40.5867329142,"imp":17.8081679468,"elo":1338.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":103.1262709089,"imp":17.38611146,"elo":1357.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":206.3615235739,"imp":17.2782509894,"elo":1366.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":413.6924902571,"imp":17.1506475584,"elo":1374.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":620.0408191726,"imp":17.0480405436,"elo":1378.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":833.0924159486,"imp":17.0316117142,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":34.0310873482,"imp":14.5874716272,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":128.8352109042,"imp":15.9721952582,"elo":1427.3,"imputed":true,"imputed_pct":8.3333333333,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.9096333145,"imp":22.402659937,"elo":1212.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":51.4995677932,"imp":21.0411570613,"elo":1230.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":153.1634884827,"imp":18.8427510755,"elo":1305.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":306.0472940151,"imp":18.3345425243,"elo":1328.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":758.7881779295,"imp":17.8523142338,"elo":1348.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1565.6822074431,"imp":18.0735791538,"elo":1334.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3151.1944700481,"imp":18.5292904162,"elo":1315.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4699.0006066566,"imp":18.684131265,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6343.6148094754,"imp":18.747191848,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":0.6786165268,"imp":24.9343915808,"elo":1159.6,"imputed":true,"imputed_pct":77.7777777778,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":0.6786165268,"imp":24.6432308144,"elo":1161.9,"imputed":true,"imputed_pct":77.7777777778,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_train":13.5688156132,"imp":24.8621747875,"elo":1084.0,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":38.5275966029,"imp":34.4002816995,"elo":964.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":93.0126953024,"imp":31.4137040737,"elo":1023.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":235.686098342,"imp":28.7123061362,"elo":1090.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":282.8267356163,"imp":28.6167754604,"elo":1094.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.5076178694,"imp":28.4761078182,"elo":997.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.3146920986,"imp":26.1344274395,"elo":1046.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":12.5906114909,"imp":24.6888975359,"elo":1083.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":27.1833109568,"imp":23.8892424149,"elo":1104.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":68.9897633789,"imp":23.2887967007,"elo":1132.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":140.3776559984,"imp":23.1695511914,"elo":1140.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":283.2223248418,"imp":23.1902680287,"elo":1144.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":428.3020784862,"imp":23.2089772774,"elo":1144.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":575.6224152834,"imp":23.1793913638,"elo":1143.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":16.6080899164,"imp":19.0747712712,"elo":1253.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":35.8566583253,"imp":16.8072045794,"elo":1348.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":91.663010689,"imp":15.2061911842,"elo":1433.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":189.0044514953,"imp":14.6949550106,"elo":1454.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":476.9907049476,"imp":14.2202613107,"elo":1481.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":961.4088522866,"imp":13.9955401914,"elo":1496.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1920.3182480183,"imp":13.9995787223,"elo":1498.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2879.4073323842,"imp":14.1154347261,"elo":1501.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3855.7666751446,"imp":14.177498004,"elo":1500.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.5779030898,"imp":11.7492506365,"elo":1583.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":19.6925140518,"imp":11.641076225,"elo":1582.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":56.5542449974,"imp":11.3615251337,"elo":1591.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":113.5361347503,"imp":11.3126747523,"elo":1607.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":293.4788372615,"imp":11.1646351942,"elo":1607.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":577.5893123253,"imp":11.1296859382,"elo":1603.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1152.3312972563,"imp":10.6829119323,"elo":1611.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1735.6880999961,"imp":10.4188821796,"elo":1615.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2320.5586596122,"imp":9.8141256931,"elo":1614.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":16.5391218381,"imp":19.589209504,"elo":1293.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.7864862661,"imp":18.063864209,"elo":1307.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":90.9755007226,"imp":17.2339825352,"elo":1341.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":208.5282027548,"imp":16.4072064806,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":397.9314678674,"imp":16.0404903066,"elo":1399.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":999.4790170647,"imp":15.4653202013,"elo":1415.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":2004.3956749143,"imp":15.0579220101,"elo":1430.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":4010.9044638578,"imp":14.8889472803,"elo":1438.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":6016.93821747,"imp":14.8083673875,"elo":1444.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":8061.743825954,"imp":14.7137086887,"elo":1450.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":5.1020093375,"imp":15.4434955162,"elo":1435.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":45.3524743275,"imp":4.5195049586,"elo":1793.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.589504217,"imp":28.2152693452,"elo":1014.6,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":7.5554129923,"imp":19.1247746472,"elo":1281.2,"imputed":true,"imputed_pct":27.7777777778,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":3.8948553652,"imp":11.5767798612,"elo":1559.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":11.2200428783,"imp":18.6815993046,"elo":1280.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":23.7397758074,"imp":18.2995713525,"elo":1293.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":61.4524391665,"imp":17.7624064667,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":127.9575000974,"imp":17.2361578829,"elo":1346.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":319.7265066945,"imp":16.8221983373,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":634.7918157013,"imp":16.5081860099,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1265.5050030954,"imp":16.2079692435,"elo":1393.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1910.2960689809,"imp":16.1009794989,"elo":1398.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2564.2291668214,"imp":15.9963493117,"elo":1397.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":7.0738556742,"imp":11.1166445779,"elo":1619.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":8.5714229456,"imp":10.0232522966,"elo":1644.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4.9840988627,"imp":17.641257265,"elo":1315.8,"imputed":true,"imputed_pct":8.3333333333,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":26.6938878603,"imp":17.000989526,"elo":1353.1,"imputed":true,"imputed_pct":8.3333333333,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":73.0172393675,"imp":16.1891000181,"elo":1398.3,"imputed":true,"imputed_pct":8.3333333333,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":164.9132339833,"imp":15.6263667796,"elo":1420.6,"imputed":true,"imputed_pct":8.3333333333,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":415.664902619,"imp":14.9875460257,"elo":1456.4,"imputed":true,"imputed_pct":8.3333333333,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":849.106399404,"imp":14.7221925817,"elo":1473.6,"imputed":true,"imputed_pct":8.3333333333,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1740.3427841345,"imp":14.6688850148,"elo":1474.6,"imputed":true,"imputed_pct":8.3333333333,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2623.8682588802,"imp":14.7722727885,"elo":1468.8,"imputed":true,"imputed_pct":8.3333333333,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3560.2284047226,"imp":14.8950682806,"elo":1456.3,"imputed":true,"imputed_pct":8.3333333333,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":459.757659874,"imp":31.2767685229,"elo":1026.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":859.3423899592,"imp":31.1186338433,"elo":1034.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1648.7767518666,"imp":30.6877339189,"elo":1045.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2593.696023839,"imp":29.8288167151,"elo":1061.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5812.689260478,"imp":28.2379453852,"elo":1085.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11696.3454938888,"imp":26.9404280118,"elo":1102.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23869.0757052958,"imp":26.4026688692,"elo":1112.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":35350.6242030506,"imp":26.1262438358,"elo":1121.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":46972.7202570992,"imp":25.800308896,"elo":1131.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":2.5372143019,"imp":17.3467778368,"elo":1358.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":12.6783036886,"imp":25.7949047586,"elo":1024.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":27.894998506,"imp":23.3779926409,"elo":1090.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":88.2967448357,"imp":21.4032050935,"elo":1149.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":185.1308765061,"imp":20.6962701856,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":457.9055405764,"imp":20.0999579938,"elo":1213.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":916.161299444,"imp":19.7507644415,"elo":1227.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1819.6211273997,"imp":19.5468652926,"elo":1227.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2759.4404936339,"imp":19.3502239394,"elo":1238.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3694.4938905556,"imp":19.2371952226,"elo":1241.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":3.0974213034,"imp":21.8319154148,"elo":1128.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":8.3851286101,"imp":19.59675691,"elo":1233.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":21.5559379962,"imp":18.9726438895,"elo":1272.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":42.3770236352,"imp":18.590218229,"elo":1295.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":109.5235780735,"imp":18.2988488292,"elo":1309.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":220.0167184889,"imp":18.1906460598,"elo":1318.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":441.8368524844,"imp":18.077063286,"elo":1325.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":661.3695455545,"imp":17.9707627122,"elo":1331.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":886.8089747599,"imp":17.9263477301,"elo":1331.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":537.959331743,"imp":25.0473443461,"elo":1068.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1311.0677158659,"imp":21.8115710892,"elo":1159.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":3930.4572134166,"imp":18.5319723283,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":8786.6645332738,"imp":17.4473426726,"elo":1325.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":21604.4108540479,"imp":16.4659654965,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":22680.1927917466,"imp":16.4628812347,"elo":1381.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":6.0878006956,"imp":26.3422878272,"elo":1049.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":12.2392299129,"imp":23.8247909989,"elo":1108.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":32.0134798368,"imp":20.5126595508,"elo":1217.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":65.1287013348,"imp":19.0003912199,"elo":1280.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":166.8282289738,"imp":18.1941069653,"elo":1311.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":335.4693508184,"imp":17.8661817681,"elo":1326.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":669.3015790718,"imp":17.6942129242,"elo":1328.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1004.6327529567,"imp":17.6672865316,"elo":1330.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1346.8150776785,"imp":17.6514896819,"elo":1330.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.8806983387,"x_train":269.2082934936,"imp":20.0416628001,"elo":1284.1,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0408851968,"x_train":9.0359136176,"imp":18.2867004257,"elo":1307.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0905042501,"x_train":21.9992777967,"imp":17.5241517957,"elo":1337.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.172973571,"x_train":53.8141065802,"imp":17.046472113,"elo":1356.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2079530855,"x_train":96.9989733937,"imp":16.797349382,"elo":1365.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2459997699,"x_train":231.1646909722,"imp":16.6154192537,"elo":1368.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.291674196,"x_train":468.6200877614,"imp":16.6045136365,"elo":1364.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.343571202,"x_train":950.6643741645,"imp":16.4442671888,"elo":1362.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3883571805,"x_train":1384.6574015383,"imp":16.4367871001,"elo":1357.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3907846602,"x_train":1827.947236699,"imp":16.3586667057,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0367584655,"x_train":5.5160770154,"imp":21.5454621825,"elo":1201.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0787805569,"x_train":11.4571846403,"imp":20.8567308321,"elo":1234.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.152959706,"x_train":28.0250922704,"imp":20.3736954266,"elo":1261.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.221696575,"x_train":57.1334404694,"imp":20.139547438,"elo":1270.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2819310273,"x_train":145.0536005577,"imp":19.761846164,"elo":1283.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.339421727,"x_train":292.2184372302,"imp":19.5757485361,"elo":1292.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3845866458,"x_train":586.8028092523,"imp":19.4675721168,"elo":1294.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3951493501,"x_train":881.2145061251,"imp":19.3983671988,"elo":1297.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4568895686,"x_train":1179.9999181447,"imp":19.3981292293,"elo":1296.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0170112085,"x_train":7.5965245684,"imp":20.8975434258,"elo":1197.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0331881177,"x_train":20.5778802427,"imp":20.5762208261,"elo":1213.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0590854757,"x_train":58.0053529444,"imp":20.2325494376,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0783042685,"x_train":114.2580939201,"imp":20.1142358534,"elo":1242.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1039353704,"x_train":281.7059626239,"imp":19.9775932818,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1218997247,"x_train":567.1802440047,"imp":19.8259050438,"elo":1257.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1354779968,"x_train":1120.6313986655,"imp":19.6807553778,"elo":1264.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1434728711,"x_train":1665.6646846813,"imp":19.6440045451,"elo":1268.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1523084536,"x_train":2241.8559420912,"imp":19.608987689,"elo":1268.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6288369946,"x_train":4.3684404027,"imp":27.1914700939,"elo":1039.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7303730135,"x_train":8.974442795,"imp":25.6146104924,"elo":1066.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6701745763,"x_train":23.2753987371,"imp":23.3287526027,"elo":1123.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7102008111,"x_train":45.9919048085,"imp":22.6824940376,"elo":1152.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9526662594,"x_train":110.3208397341,"imp":21.990945112,"elo":1181.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2214235524,"x_train":212.5071471057,"imp":21.814698821,"elo":1189.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5521012524,"x_train":419.021686722,"imp":21.7168803028,"elo":1193.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7473298834,"x_train":627.2476408835,"imp":21.5934476972,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3932122077,"x_train":841.5972483589,"imp":21.5990262923,"elo":1201.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5709841633,"x_train":5.8349009345,"imp":26.8785744883,"elo":985.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4163112584,"x_train":14.4449316294,"imp":26.1049653205,"elo":1027.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7078527827,"x_train":42.2187698316,"imp":24.3451471633,"elo":1094.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.5424617722,"x_train":78.730916991,"imp":23.5973668956,"elo":1122.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.7771560016,"x_train":201.7091267248,"imp":22.6808412586,"elo":1146.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.2313050635,"x_train":410.0601179712,"imp":22.2640766517,"elo":1163.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.1854207053,"x_train":823.1441900238,"imp":22.0723475153,"elo":1170.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.6836795815,"x_train":1238.4424229224,"imp":21.9376825834,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.821672488,"x_train":1666.9308833948,"imp":21.9461018446,"elo":1182.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2175696189,"x_train":3.6257108294,"imp":21.7884658415,"elo":1134.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7244393627,"x_train":8.2503296642,"imp":19.6725465346,"elo":1233.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2312451116,"x_train":20.5596206802,"imp":18.3350057151,"elo":1308.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5807687941,"x_train":40.5867329142,"imp":17.8081679468,"elo":1338.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.7582613878,"x_train":103.1262709089,"imp":17.38611146,"elo":1357.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.9395731779,"x_train":206.3615235739,"imp":17.2782509894,"elo":1366.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.9939153323,"x_train":413.6924902571,"imp":17.1506475584,"elo":1374.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2268088052,"x_train":620.0408191726,"imp":17.0480405436,"elo":1378.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1817652315,"x_train":833.0924159486,"imp":17.0316117142,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":7.2046047302,"x_train":34.0310873482,"imp":14.5874716272,"elo":1415.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.9990016855,"x_train":128.8352109042,"imp":15.9721952582,"elo":1427.3,"imputed":true,"imputed_pct":8.3333333333,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3225041576,"x_train":14.9096333145,"imp":22.402659937,"elo":1212.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8785929508,"x_train":51.4995677932,"imp":21.0411570613,"elo":1230.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7997608403,"x_train":153.1634884827,"imp":18.8427510755,"elo":1305.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.832599148,"x_train":306.0472940151,"imp":18.3345425243,"elo":1328.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1547858756,"x_train":758.7881779295,"imp":17.8523142338,"elo":1348.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.4651963907,"x_train":1565.6822074431,"imp":18.0735791538,"elo":1334.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.2034777482,"x_train":3151.1944700481,"imp":18.5292904162,"elo":1315.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.1674130588,"x_train":4699.0006066566,"imp":18.684131265,"elo":1314.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.5454388038,"x_train":6343.6148094754,"imp":18.747191848,"elo":1312.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.1083048852,"x_train":0.6786165268,"imp":24.9343915808,"elo":1159.6,"imputed":true,"imputed_pct":77.7777777778,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":0.1083048852,"x_train":0.6786165268,"imp":24.6432308144,"elo":1161.9,"imputed":true,"imputed_pct":77.7777777778,"n_configs":1.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.4356700948,"x_train":13.5688156132,"imp":24.8621747875,"elo":1084.0,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0412162752,"x_train":38.5275966029,"imp":34.4002816995,"elo":964.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2867661168,"x_train":93.0126953024,"imp":31.4137040737,"elo":1023.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5061660782,"x_train":235.686098342,"imp":28.7123061362,"elo":1090.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5087521152,"x_train":282.8267356163,"imp":28.6167754604,"elo":1094.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0852899147,"x_train":0.5076178694,"imp":28.4761078182,"elo":997.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2679638116,"x_train":3.3146920986,"imp":26.1344274395,"elo":1046.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4603970571,"x_train":12.5906114909,"imp":24.6888975359,"elo":1083.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6007649726,"x_train":27.1833109568,"imp":23.8892424149,"elo":1104.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7995923156,"x_train":68.9897633789,"imp":23.2887967007,"elo":1132.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9217934517,"x_train":140.3776559984,"imp":23.1695511914,"elo":1140.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0817312972,"x_train":283.2223248418,"imp":23.1902680287,"elo":1144.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2223316361,"x_train":428.3020784862,"imp":23.2089772774,"elo":1144.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"x_train":575.6224152834,"imp":23.1793913638,"elo":1143.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.2987548984,"x_train":16.6080899164,"imp":19.0747712712,"elo":1253.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.8477674709,"x_train":35.8566583253,"imp":16.8072045794,"elo":1348.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.0672087974,"x_train":91.663010689,"imp":15.2061911842,"elo":1433.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.5254352945,"x_train":189.0044514953,"imp":14.6949550106,"elo":1454.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4341466661,"x_train":476.9907049476,"imp":14.2202613107,"elo":1481.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":18.2190658863,"x_train":961.4088522866,"imp":13.9955401914,"elo":1496.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":20.0311215643,"x_train":1920.3182480183,"imp":13.9995787223,"elo":1498.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":22.0848749753,"x_train":2879.4073323842,"imp":14.1154347261,"elo":1501.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":24.9138452893,"x_train":3855.7666751446,"imp":14.177498004,"elo":1500.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5863334021,"x_train":6.5779030898,"imp":11.7492506365,"elo":1583.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.32293962,"x_train":19.6925140518,"imp":11.641076225,"elo":1582.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8419737629,"x_train":56.5542449974,"imp":11.3615251337,"elo":1591.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.9894747323,"x_train":113.5361347503,"imp":11.3126747523,"elo":1607.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.5217042751,"x_train":293.4788372615,"imp":11.1646351942,"elo":1607.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.2594554521,"x_train":577.5893123253,"imp":11.1296859382,"elo":1603.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.0497467794,"x_train":1152.3312972563,"imp":10.6829119323,"elo":1611.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8968974009,"x_train":1735.6880999961,"imp":10.4188821796,"elo":1615.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9558346011,"x_train":2320.5586596122,"imp":9.8141256931,"elo":1614.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.2543929634,"x_train":16.5391218381,"imp":19.589209504,"elo":1293.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"x_train":51.7864862661,"imp":18.063864209,"elo":1307.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":59.7687648763,"x_train":90.9755007226,"imp":17.2339825352,"elo":1341.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":91.9802288244,"x_train":208.5282027548,"imp":16.4072064806,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":132.6956612183,"x_train":397.9314678674,"imp":16.0404903066,"elo":1399.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":197.2683733908,"x_train":999.4790170647,"imp":15.4653202013,"elo":1415.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":241.2137556465,"x_train":2004.3956749143,"imp":15.0579220101,"elo":1430.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":310.7845512859,"x_train":4010.9044638578,"imp":14.8889472803,"elo":1438.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":340.4331355157,"x_train":6016.93821747,"imp":14.8083673875,"elo":1444.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":347.4963020094,"x_train":8061.743825954,"imp":14.7137086887,"elo":1450.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1868636356,"x_train":5.1020093375,"imp":15.4434955162,"elo":1435.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":5.3427919745,"x_train":45.3524743275,"imp":4.5195049586,"elo":1793.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1488925237,"x_train":1.589504217,"imp":28.2152693452,"elo":1014.6,"imputed":true,"imputed_pct":22.2222222222,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.4929299615,"x_train":7.5554129923,"imp":19.1247746472,"elo":1281.2,"imputed":true,"imputed_pct":27.7777777778,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1502327458,"x_train":3.8948553652,"imp":11.5767798612,"elo":1559.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.3927969172,"x_train":11.2200428783,"imp":18.6815993046,"elo":1280.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.8073226159,"x_train":23.7397758074,"imp":18.2995713525,"elo":1293.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.4253994355,"x_train":61.4524391665,"imp":17.7624064667,"elo":1315.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.9037000676,"x_train":127.9575000974,"imp":17.2361578829,"elo":1346.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.480365301,"x_train":319.7265066945,"imp":16.8221983373,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":2.9323777701,"x_train":634.7918157013,"imp":16.5081860099,"elo":1383.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.1758660276,"x_train":1265.5050030954,"imp":16.2079692435,"elo":1393.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":3.2134702655,"x_train":1910.2960689809,"imp":16.1009794989,"elo":1398.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":3.4526368273,"x_train":2564.2291668214,"imp":15.9963493117,"elo":1397.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5551544404,"x_train":7.0738556742,"imp":11.1166445779,"elo":1619.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4108441184,"x_train":8.5714229456,"imp":10.0232522966,"elo":1644.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5110070735,"x_train":4.9840988627,"imp":17.641257265,"elo":1315.8,"imputed":true,"imputed_pct":8.3333333333,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":5.2679052389,"x_train":26.6938878603,"imp":17.000989526,"elo":1353.1,"imputed":true,"imputed_pct":8.3333333333,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.3532799483,"x_train":73.0172393675,"imp":16.1891000181,"elo":1398.3,"imputed":true,"imputed_pct":8.3333333333,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.3324490655,"x_train":164.9132339833,"imp":15.6263667796,"elo":1420.6,"imputed":true,"imputed_pct":8.3333333333,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":25.7012755501,"x_train":415.664902619,"imp":14.9875460257,"elo":1456.4,"imputed":true,"imputed_pct":8.3333333333,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":34.4130216561,"x_train":849.106399404,"imp":14.7221925817,"elo":1473.6,"imputed":true,"imputed_pct":8.3333333333,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":41.4628183099,"x_train":1740.3427841345,"imp":14.6688850148,"elo":1474.6,"imputed":true,"imputed_pct":8.3333333333,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":46.8009333303,"x_train":2623.8682588802,"imp":14.7722727885,"elo":1468.8,"imputed":true,"imputed_pct":8.3333333333,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":49.1960479984,"x_train":3560.2284047226,"imp":14.8950682806,"elo":1456.3,"imputed":true,"imputed_pct":8.3333333333,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"x_train":459.757659874,"imp":31.2767685229,"elo":1026.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.7716558163,"x_train":859.3423899592,"imp":31.1186338433,"elo":1034.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.5636156792,"x_train":1648.7767518666,"imp":30.6877339189,"elo":1045.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.7112841968,"x_train":2593.696023839,"imp":29.8288167151,"elo":1061.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8983325102,"x_train":5812.689260478,"imp":28.2379453852,"elo":1085.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.5003951009,"x_train":11696.3454938888,"imp":26.9404280118,"elo":1102.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":24.9793991039,"x_train":23869.0757052958,"imp":26.4026688692,"elo":1112.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":25.6066199148,"x_train":35350.6242030506,"imp":26.1262438358,"elo":1121.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.63963438,"x_train":46972.7202570992,"imp":25.800308896,"elo":1131.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0581109952,"x_train":2.5372143019,"imp":17.3467778368,"elo":1358.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1547654046,"x_train":12.6783036886,"imp":25.7949047586,"elo":1024.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3184069177,"x_train":27.894998506,"imp":23.3779926409,"elo":1090.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6772463403,"x_train":88.2967448357,"imp":21.4032050935,"elo":1149.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8527607509,"x_train":185.1308765061,"imp":20.6962701856,"elo":1184.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2375187431,"x_train":457.9055405764,"imp":20.0999579938,"elo":1213.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5433408305,"x_train":916.161299444,"imp":19.7507644415,"elo":1227.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.800349055,"x_train":1819.6211273997,"imp":19.5468652926,"elo":1227.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0481960908,"x_train":2759.4404936339,"imp":19.3502239394,"elo":1238.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1373326275,"x_train":3694.4938905556,"imp":19.2371952226,"elo":1241.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2070918304,"x_train":3.0974213034,"imp":21.8319154148,"elo":1128.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7336842368,"x_train":8.3851286101,"imp":19.59675691,"elo":1233.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.3748576592,"x_train":21.5559379962,"imp":18.9726438895,"elo":1272.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.5232150506,"x_train":42.3770236352,"imp":18.590218229,"elo":1295.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8858828782,"x_train":109.5235780735,"imp":18.2988488292,"elo":1309.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.079369511,"x_train":220.0167184889,"imp":18.1906460598,"elo":1318.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3062309572,"x_train":441.8368524844,"imp":18.077063286,"elo":1325.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0953020923,"x_train":661.3695455545,"imp":17.9707627122,"elo":1331.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.854250542,"x_train":886.8089747599,"imp":17.9263477301,"elo":1331.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":157.807153647,"x_train":537.959331743,"imp":25.0473443461,"elo":1068.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":288.0546364493,"x_train":1311.0677158659,"imp":21.8115710892,"elo":1159.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":473.6713928263,"x_train":3930.4572134166,"imp":18.5319723283,"elo":1275.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":620.719533977,"x_train":8786.6645332738,"imp":17.4473426726,"elo":1325.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":506.099831026,"x_train":21604.4108540479,"imp":16.4659654965,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":529.125864783,"x_train":22680.1927917466,"imp":16.4628812347,"elo":1381.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9508145682,"x_train":6.0878006956,"imp":26.3422878272,"elo":1049.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.3492504189,"x_train":12.2392299129,"imp":23.8247909989,"elo":1108.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.629173245,"x_train":32.0134798368,"imp":20.5126595508,"elo":1217.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.6347035029,"x_train":65.1287013348,"imp":19.0003912199,"elo":1280.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.9845916387,"x_train":166.8282289738,"imp":18.1941069653,"elo":1311.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.2321888977,"x_train":335.4693508184,"imp":17.8661817681,"elo":1326.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.3191528609,"x_train":669.3015790718,"imp":17.6942129242,"elo":1328.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.2084021337,"x_train":1004.6327529567,"imp":17.6672865316,"elo":1330.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.1486246717,"x_train":1346.8150776785,"imp":17.6514896819,"elo":1330.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_all/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..3839c880a9d31026531a6711b5dcf7ef1d1bc303
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_all/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabPFNv2 (tuned + ensembled)", "TabDPT (tuned + ensembled)", "TabDPT-Turbo (default)", "RealMLP (tuned)", "Mitra (default)", "LimiX (default)", "TabDPT (tuned)", "TabM (tuned + ensembled)", "iLTM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "CatBoost (tuned)", "CatBoost (tuned + ensembled)", "TabPFNv2 (tuned)", "TabM (tuned)", "TabSwift (default)", "xRFM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabPFNv2 (default)", "LightGBM (tuned)", "XGBoost (tuned)", "ModernNCA (tuned + ensembled)", "CatBoost (default)", "TabDPT (default)", "ModernNCA (tuned)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "SAP-RPT-OSS (default)", "TabICL (default)", "BetaTabPFN (default)", "TabM (default)", "xRFM (tuned)", "EBM (tuned + ensembled)", "RealMLP (default)", "ChimeraBoost (tuned)", "TorchMLP (tuned + ensembled)", "EBM (tuned)", "ModernNCA (default)", "ExtraTrees (tuned + ensembled)", "ChimeraBoost (default)", "EBM (default)", "TorchMLP (tuned)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (tuned)", "Nori-30M (default)", "Nori (default)", "RandomForest (tuned + ensembled)", "LightGBM (default)", "XGBoost (default)", "FastaiMLP (tuned)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "OrionMSP (default)", "iLTM (default)", "PerpetualBooster (tuned)", "xRFM (default)", "ExtraTrees (default)", "TorchMLP (default)", "TabSTAR (default)", "TabFlex (default)", "RandomForest (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 1.0, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 1.0, 0.9722222089767456, 0.9166666865348816, 1.0, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 1.0, 0.9722222089767456, 1.0, 0.9444444179534912, 1.0, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 1.0], [0.1666666716337204, NaN, 0.5277777910232544, 0.5138888955116272, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.6388888955116272, 0.75, 0.6388888955116272, 0.75, 0.6944444179534912, 0.8055555820465088, 0.7777777910232544, 0.8333333134651184, 0.7222222089767456, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.9166666865348816, 0.7777777910232544, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.9166666865348816, 0.7777777910232544, 0.9444444179534912, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8055555820465088, 0.9444444179534912, 0.9166666865348816, 0.8333333134651184, 0.9444444179534912, 0.8055555820465088, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.8888888955116272, 0.9722222089767456, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.8611111044883728, 0.8611111044883728, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 1.0, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 1.0, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 1.0], [0.2222222238779068, 0.4722222089767456, NaN, 0.4722222089767456, 0.5833333134651184, 0.5277777910232544, 0.5, 0.6111111044883728, 0.8055555820465088, 0.7777777910232544, 0.6944444179534912, 0.7222222089767456, 0.6666666865348816, 0.8055555820465088, 0.7777777910232544, 0.7222222089767456, 0.7777777910232544, 0.8333333134651184, 0.8611111044883728, 0.8888888955116272, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.9166666865348816, 0.9166666865348816, 0.7777777910232544, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.8055555820465088, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8611111044883728, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.8888888955116272, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.7777777910232544, 0.8333333134651184, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.9166666865348816, 0.8611111044883728, 0.8333333134651184, 0.9444444179534912, 0.9166666865348816, 1.0, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 1.0], [0.1388888955116272, 0.4861111044883728, 0.5277777910232544, NaN, 0.5277777910232544, 0.6388888955116272, 0.6111111044883728, 0.6944444179534912, 0.75, 0.6666666865348816, 0.8055555820465088, 0.6388888955116272, 0.6666666865348816, 0.7777777910232544, 0.8333333134651184, 0.75, 0.6944444179534912, 0.7777777910232544, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.9444444179534912, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.9444444179534912, 0.75, 0.8888888955116272, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.7777777910232544, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.8055555820465088, 0.8333333134651184, 0.9166666865348816, 0.9722222089767456, 1.0, 0.9166666865348816, 0.8333333134651184, 0.8333333134651184, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.8611111044883728, 0.8888888955116272, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 1.0], [0.1666666716337204, 0.4722222089767456, 0.4166666567325592, 0.4722222089767456, NaN, 0.6388888955116272, 0.5, 0.5555555820465088, 0.7222222089767456, 0.6666666865348816, 0.7222222089767456, 0.6388888955116272, 0.7222222089767456, 0.8055555820465088, 0.8611111044883728, 0.7777777910232544, 0.6944444179534912, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272, 0.8888888955116272, 0.75, 0.9444444179534912, 0.8888888955116272, 0.9166666865348816, 0.9444444179534912, 0.75, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.8611111044883728, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.8611111044883728, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.8055555820465088, 0.8333333134651184, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.8888888955116272, 0.8333333134651184, 0.8333333134651184, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 1.0, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.8333333134651184, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.1666666716337204, 0.4722222089767456, 0.4722222089767456, 0.3611111044883728, 0.3611111044883728, NaN, 0.4722222089767456, 0.4305555522441864, 0.6666666865348816, 0.6388888955116272, 0.5277777910232544, 0.6388888955116272, 0.6944444179534912, 0.75, 0.75, 0.7222222089767456, 0.6944444179534912, 0.7777777910232544, 0.7222222089767456, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8055555820465088, 0.8888888955116272, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.9027777910232544, 0.8055555820465088, 0.7222222089767456, 0.8055555820465088, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9722222089767456, 0.8611111044883728, 0.7777777910232544, 0.8055555820465088, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 1.0, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 0.8888888955116272, 0.9722222089767456, 0.9444444179534912, 1.0, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 0.9722222089767456], [0.1111111119389534, 0.4166666567325592, 0.5, 0.3888888955116272, 0.5, 0.5277777910232544, NaN, 0.5833333134651184, 0.6666666865348816, 0.75, 0.6388888955116272, 0.6944444179534912, 0.7222222089767456, 0.6666666865348816, 0.7222222089767456, 0.6944444179534912, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.7222222089767456, 0.75, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.75, 0.8055555820465088, 0.8611111044883728, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.8055555820465088, 0.8611111044883728, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8888888955116272, 0.8055555820465088, 0.8888888955116272, 0.8333333134651184, 0.8611111044883728, 0.8888888955116272, 0.9166666865348816, 0.9444444179534912, 0.8888888955116272, 0.75, 0.75, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 0.9722222089767456, 1.0], [0.1388888955116272, 0.3611111044883728, 0.3888888955116272, 0.3055555522441864, 0.4444444477558136, 0.5694444179534912, 0.4166666567325592, NaN, 0.6388888955116272, 0.6388888955116272, 0.5972222089767456, 0.5833333134651184, 0.6388888955116272, 0.75, 0.8055555820465088, 0.6111111044883728, 0.6388888955116272, 0.8055555820465088, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8888888955116272, 0.8333333134651184, 0.6944444179534912, 0.8055555820465088, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8611111044883728, 0.8055555820465088, 0.9166666865348816, 0.7222222089767456, 0.8611111044883728, 0.8055555820465088, 0.8888888955116272, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.8888888955116272, 0.8333333134651184, 0.8888888955116272, 0.9444444179534912, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.8611111044883728, 0.9166666865348816, 0.7777777910232544, 0.8055555820465088, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.8888888955116272, 0.8055555820465088, 0.7777777910232544, 0.8888888955116272, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 1.0], [0.1388888955116272, 0.25, 0.1944444477558136, 0.25, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6388888955116272, 0.5555555820465088, 0.4722222089767456, 0.5, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728, 0.6388888955116272, 0.7222222089767456, 0.6388888955116272, 0.6944444179534912, 0.8611111044883728, 0.6111111044883728, 0.8333333134651184, 0.8611111044883728, 0.75, 0.8611111044883728, 0.6666666865348816, 0.75, 0.8055555820465088, 0.8333333134651184, 0.7222222089767456, 0.7222222089767456, 0.75, 0.7777777910232544, 0.75, 0.8333333134651184, 0.8888888955116272, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.8055555820465088, 0.8888888955116272, 0.8611111044883728, 0.75, 0.75, 0.8055555820465088, 0.9166666865348816, 0.9444444179534912, 0.8611111044883728, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.8333333134651184, 0.8888888955116272, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.0833333358168602, 0.3611111044883728, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, 0.25, 0.3611111044883728, 0.4444444477558136, NaN, 0.4166666567325592, 0.4166666567325592, 0.5277777910232544, 0.6111111044883728, 0.5555555820465088, 0.5555555820465088, 0.5, 0.6111111044883728, 0.6944444179534912, 0.7777777910232544, 0.6666666865348816, 0.7222222089767456, 0.5555555820465088, 0.7222222089767456, 0.5833333134651184, 0.7222222089767456, 0.8055555820465088, 0.6666666865348816, 0.8333333134651184, 0.75, 0.6666666865348816, 0.7777777910232544, 0.6944444179534912, 0.6666666865348816, 0.8055555820465088, 0.75, 0.6944444179534912, 0.6944444179534912, 0.75, 0.75, 0.75, 0.7222222089767456, 0.75, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.8055555820465088, 0.8333333134651184, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8055555820465088, 0.75, 0.7777777910232544, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.75, 0.75, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456, 0.8888888955116272, 0.9722222089767456, 0.9722222089767456, 0.8333333134651184, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0, 1.0, 1.0], [0.0555555559694767, 0.25, 0.3055555522441864, 0.1944444477558136, 0.2777777910232544, 0.4722222089767456, 0.3611111044883728, 0.4027777910232544, 0.5555555820465088, 0.5833333134651184, NaN, 0.5277777910232544, 0.5277777910232544, 0.5555555820465088, 0.6111111044883728, 0.5, 0.5833333134651184, 0.6666666865348816, 0.6388888955116272, 0.6944444179534912, 0.6666666865348816, 0.6666666865348816, 0.7638888955116272, 0.6944444179534912, 0.6388888955116272, 0.7222222089767456, 0.8055555820465088, 0.7222222089767456, 0.7222222089767456, 0.75, 0.75, 0.6944444179534912, 0.5833333134651184, 0.6944444179534912, 0.7222222089767456, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.7638888955116272, 0.7222222089767456, 0.7777777910232544, 0.8055555820465088, 0.8055555820465088, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.75, 0.75, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.7777777910232544, 0.8055555820465088, 0.7638888955116272, 0.7916666865348816, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8333333134651184, 0.75, 0.7222222089767456, 0.8888888955116272, 0.8333333134651184, 0.8472222089767456, 0.8888888955116272, 0.8888888955116272, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.8055555820465088, 0.8472222089767456, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 1.0], [0.0833333358168602, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3611111044883728, 0.3611111044883728, 0.3055555522441864, 0.4166666567325592, 0.5555555820465088, 0.5833333134651184, 0.4722222089767456, NaN, 0.4166666567325592, 0.6666666865348816, 0.5555555820465088, 0.6111111044883728, 0.7222222089767456, 0.5555555820465088, 0.7222222089767456, 0.6388888955116272, 0.6111111044883728, 0.6666666865348816, 0.6388888955116272, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7222222089767456, 0.5833333134651184, 0.6944444179534912, 0.75, 0.6944444179534912, 0.6944444179534912, 0.75, 0.6666666865348816, 0.75, 0.8055555820465088, 0.6944444179534912, 0.6666666865348816, 0.6388888955116272, 0.6944444179534912, 0.75, 0.6944444179534912, 0.6944444179534912, 0.75, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.7222222089767456, 0.8055555820465088, 0.7222222089767456, 0.75, 0.8888888955116272, 0.8055555820465088, 0.7777777910232544, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.75, 0.8333333134651184, 0.7777777910232544, 0.9722222089767456, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.8611111044883728, 0.8333333134651184, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 1.0], [0.0833333358168602, 0.1944444477558136, 0.3333333432674408, 0.3333333432674408, 0.2777777910232544, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, 0.5833333134651184, NaN, 0.6111111044883728, 0.5277777910232544, 0.5277777910232544, 0.5555555820465088, 0.5, 0.6111111044883728, 0.6388888955116272, 0.5555555820465088, 0.6111111044883728, 0.5833333134651184, 0.5555555820465088, 0.6805555820465088, 0.6944444179534912, 0.7222222089767456, 0.5277777910232544, 0.6944444179534912, 0.75, 0.5833333134651184, 0.6944444179534912, 0.6666666865348816, 0.6388888955116272, 0.7222222089767456, 0.75, 0.6944444179534912, 0.6666666865348816, 0.6111111044883728, 0.7222222089767456, 0.75, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.75, 0.7222222089767456, 0.7222222089767456, 0.8333333134651184, 0.75, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.6944444179534912, 0.6944444179534912, 0.8611111044883728, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.7777777910232544, 0.75, 0.8611111044883728, 0.8055555820465088, 0.9444444179534912, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.9444444179534912, 0.8888888955116272, 0.8055555820465088, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456, 1.0], [0.1111111119389534, 0.2222222238779068, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.25, 0.3333333432674408, 0.25, 0.3611111044883728, 0.3888888955116272, 0.4444444477558136, 0.3333333432674408, 0.3888888955116272, NaN, 0.4722222089767456, 0.4027777910232544, 0.4444444477558136, 0.5277777910232544, 0.5555555820465088, 0.6388888955116272, 0.6111111044883728, 0.6111111044883728, 0.5277777910232544, 0.6388888955116272, 0.5555555820465088, 0.5833333134651184, 0.7222222089767456, 0.6111111044883728, 0.7222222089767456, 0.75, 0.6666666865348816, 0.7222222089767456, 0.5833333134651184, 0.7777777910232544, 0.6666666865348816, 0.7222222089767456, 0.6111111044883728, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.6944444179534912, 0.8055555820465088, 0.7777777910232544, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.75, 0.7777777910232544, 0.75, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.75, 0.75, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8333333134651184, 0.7777777910232544, 0.8055555820465088, 0.8888888955116272, 0.8888888955116272, 0.8472222089767456, 0.9444444179534912, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.8055555820465088, 0.8611111044883728, 0.9722222089767456, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 0.9583333134651184, 0.9444444179534912, 0.9722222089767456, 1.0], [0.0, 0.1666666716337204, 0.2222222238779068, 0.1666666716337204, 0.1388888955116272, 0.25, 0.2777777910232544, 0.1944444477558136, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.4444444477558136, 0.4722222089767456, 0.5277777910232544, NaN, 0.5277777910232544, 0.4444444477558136, 0.6944444179534912, 0.6666666865348816, 0.6666666865348816, 0.5833333134651184, 0.6111111044883728, 0.5833333134651184, 0.7222222089767456, 0.5833333134651184, 0.6666666865348816, 0.6666666865348816, 0.625, 0.7777777910232544, 0.6944444179534912, 0.75, 0.75, 0.6111111044883728, 0.7222222089767456, 0.7222222089767456, 0.6666666865348816, 0.5833333134651184, 0.7083333134651184, 0.7083333134651184, 0.8333333134651184, 0.6944444179534912, 0.7777777910232544, 0.75, 0.8055555820465088, 0.75, 0.7777777910232544, 0.8333333134651184, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6944444179534912, 0.6944444179534912, 0.75, 0.8333333134651184, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.8888888955116272, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.875, 0.9027777910232544, 0.8888888955116272, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.0833333358168602, 0.2777777910232544, 0.2777777910232544, 0.25, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.3888888955116272, 0.5277777910232544, 0.4444444477558136, 0.5, 0.3888888955116272, 0.4722222089767456, 0.5972222089767456, 0.4722222089767456, NaN, 0.5, 0.5833333134651184, 0.625, 0.6388888955116272, 0.5277777910232544, 0.5833333134651184, 0.6111111044883728, 0.6388888955116272, 0.5277777910232544, 0.6388888955116272, 0.6388888955116272, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.6388888955116272, 0.6111111044883728, 0.6666666865348816, 0.6944444179534912, 0.75, 0.75, 0.75, 0.75, 0.7222222089767456, 0.8055555820465088, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8333333134651184, 0.6944444179534912, 0.7222222089767456, 0.75, 0.75, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8194444179534912, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8472222089767456, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 1.0], [0.02777777798473835, 0.1666666716337204, 0.2222222238779068, 0.3055555522441864, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.5, 0.5, 0.4166666567325592, 0.2777777910232544, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5, NaN, 0.5555555820465088, 0.6666666865348816, 0.6111111044883728, 0.5833333134651184, 0.6388888955116272, 0.6111111044883728, 0.5555555820465088, 0.6666666865348816, 0.6388888955116272, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6944444179534912, 0.6388888955116272, 0.6666666865348816, 0.7777777910232544, 0.6388888955116272, 0.7222222089767456, 0.75, 0.6111111044883728, 0.6388888955116272, 0.5833333134651184, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.75, 0.6944444179534912, 0.6388888955116272, 0.8333333134651184, 0.75, 0.7777777910232544, 0.7777777910232544, 0.7222222089767456, 0.6944444179534912, 0.8333333134651184, 0.6666666865348816, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.75, 0.8055555820465088, 0.7777777910232544, 0.9444444179534912, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.9166666865348816, 0.8611111044883728, 0.8055555820465088, 0.8888888955116272, 0.8888888955116272, 0.9722222089767456, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.9444444179534912, 0.9722222089767456], [0.0833333358168602, 0.1666666716337204, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2777777910232544, 0.1944444477558136, 0.3333333432674408, 0.3888888955116272, 0.3333333432674408, 0.4444444477558136, 0.5, 0.4722222089767456, 0.3055555522441864, 0.4166666567325592, 0.4444444477558136, NaN, 0.5277777910232544, 0.5555555820465088, 0.4722222089767456, 0.5277777910232544, 0.3888888955116272, 0.6111111044883728, 0.5555555820465088, 0.5833333134651184, 0.6388888955116272, 0.5, 0.6111111044883728, 0.6944444179534912, 0.6388888955116272, 0.6666666865348816, 0.6111111044883728, 0.6388888955116272, 0.6388888955116272, 0.6388888955116272, 0.6388888955116272, 0.6944444179534912, 0.6944444179534912, 0.8333333134651184, 0.5833333134651184, 0.75, 0.75, 0.8055555820465088, 0.75, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.8055555820465088, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.6944444179534912, 0.6111111044883728, 0.6111111044883728, 0.7777777910232544, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.8333333134651184, 0.8055555820465088, 0.9722222089767456, 0.8333333134651184, 0.9166666865348816, 0.8333333134651184, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 1.0, 1.0], [0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1111111119389534, 0.2777777910232544, 0.1944444477558136, 0.2777777910232544, 0.2222222238779068, 0.3055555522441864, 0.3611111044883728, 0.2777777910232544, 0.3888888955116272, 0.4444444477558136, 0.3333333432674408, 0.375, 0.3333333432674408, 0.4722222089767456, NaN, 0.5555555820465088, 0.5555555820465088, 0.5, 0.5, 0.5277777910232544, 0.5, 0.5555555820465088, 0.5833333134651184, 0.6111111044883728, 0.6388888955116272, 0.6666666865348816, 0.5277777910232544, 0.5555555820465088, 0.5277777910232544, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.6388888955116272, 0.5138888955116272, 0.5277777910232544, 0.6111111044883728, 0.6944444179534912, 0.6388888955116272, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.75, 0.7777777910232544, 0.8333333134651184, 0.7777777910232544, 0.7222222089767456, 0.7222222089767456, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.7222222089767456, 0.7222222089767456, 0.8888888955116272, 0.8888888955116272, 0.8333333134651184, 0.8611111044883728, 0.9444444179534912, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.8055555820465088, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9444444179534912, 1.0], [0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3055555522441864, 0.3611111044883728, 0.3611111044883728, 0.3611111044883728, 0.3333333432674408, 0.3611111044883728, 0.3888888955116272, 0.4444444477558136, 0.4444444477558136, NaN, 0.5, 0.5555555820465088, 0.3888888955116272, 0.4722222089767456, 0.5277777910232544, 0.5833333134651184, 0.5555555820465088, 0.5277777910232544, 0.75, 0.6944444179534912, 0.5, 0.7222222089767456, 0.5277777910232544, 0.5833333134651184, 0.6666666865348816, 0.6111111044883728, 0.5833333134651184, 0.6111111044883728, 0.5833333134651184, 0.6666666865348816, 0.6666666865348816, 0.7222222089767456, 0.6666666865348816, 0.75, 0.7777777910232544, 0.75, 0.7777777910232544, 0.7777777910232544, 0.8333333134651184, 0.8055555820465088, 0.7777777910232544, 0.8055555820465088, 0.7222222089767456, 0.75, 0.75, 0.8333333134651184, 0.9444444179534912, 0.9166666865348816, 0.8333333134651184, 0.7777777910232544, 0.75, 0.9166666865348816, 0.8611111044883728, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.8055555820465088, 0.8611111044883728, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912, 0.8888888955116272, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 1.0, 1.0], [0.0555555559694767, 0.2222222238779068, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3888888955116272, 0.4444444477558136, 0.3888888955116272, 0.4166666567325592, 0.4722222089767456, 0.4166666567325592, 0.5277777910232544, 0.4444444477558136, 0.5, NaN, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4722222089767456, 0.5277777910232544, 0.6388888955116272, 0.5, 0.5833333134651184, 0.6666666865348816, 0.6111111044883728, 0.6944444179534912, 0.5277777910232544, 0.6111111044883728, 0.5277777910232544, 0.6944444179534912, 0.6111111044883728, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.6388888955116272, 0.6944444179534912, 0.6944444179534912, 0.6388888955116272, 0.75, 0.7777777910232544, 0.7222222089767456, 0.7222222089767456, 0.75, 0.7777777910232544, 0.7777777910232544, 0.7222222089767456, 0.6944444179534912, 0.6944444179534912, 0.6944444179534912, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8055555820465088, 0.75, 0.75, 0.8888888955116272, 0.8333333134651184, 0.8055555820465088, 0.8611111044883728, 0.9166666865348816, 0.8055555820465088, 0.8333333134651184, 0.9166666865348816, 0.7777777910232544, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.02777777798473835, 0.1666666716337204, 0.1388888955116272, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.2222222238779068, 0.1944444477558136, 0.1388888955116272, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.3888888955116272, 0.3888888955116272, 0.3888888955116272, 0.4166666567325592, 0.3611111044883728, 0.4722222089767456, 0.5, 0.4444444477558136, 0.5555555820465088, NaN, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.6111111044883728, 0.4722222089767456, 0.6111111044883728, 0.6388888955116272, 0.6111111044883728, 0.6666666865348816, 0.5, 0.6111111044883728, 0.5555555820465088, 0.6944444179534912, 0.6666666865348816, 0.6388888955116272, 0.5555555820465088, 0.5555555820465088, 0.6111111044883728, 0.6666666865348816, 0.7222222089767456, 0.6388888955116272, 0.7222222089767456, 0.75, 0.6944444179534912, 0.75, 0.8611111044883728, 0.7777777910232544, 0.7777777910232544, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.7222222089767456, 0.8611111044883728, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.75, 0.75, 0.9166666865348816, 0.8888888955116272, 0.7777777910232544, 0.8611111044883728, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.7777777910232544, 0.8333333134651184, 0.9444444179534912, 0.8888888955116272, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 1.0], [0.02777777798473835, 0.1111111119389534, 0.1666666716337204, 0.0555555559694767, 0.1388888955116272, 0.1666666716337204, 0.2777777910232544, 0.1111111119389534, 0.3611111044883728, 0.4444444477558136, 0.2361111044883728, 0.3611111044883728, 0.4166666567325592, 0.4722222089767456, 0.4166666567325592, 0.3888888955116272, 0.3888888955116272, 0.6111111044883728, 0.5, 0.6111111044883728, 0.5555555820465088, 0.5277777910232544, NaN, 0.6388888955116272, 0.5277777910232544, 0.6111111044883728, 0.6388888955116272, 0.5555555820465088, 0.6388888955116272, 0.6388888955116272, 0.6388888955116272, 0.5833333134651184, 0.5277777910232544, 0.6388888955116272, 0.6666666865348816, 0.6111111044883728, 0.5694444179534912, 0.5555555820465088, 0.625, 0.6944444179534912, 0.6388888955116272, 0.7222222089767456, 0.6388888955116272, 0.6944444179534912, 0.75, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.75, 0.7777777910232544, 0.8055555820465088, 0.6944444179534912, 0.6944444179534912, 0.6527777910232544, 0.6527777910232544, 0.6944444179534912, 0.7777777910232544, 0.8055555820465088, 0.75, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.75, 0.7916666865348816, 0.8611111044883728, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.7777777910232544, 0.8472222089767456, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.9722222089767456, 0.9722222089767456, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456], [0.0555555559694767, 0.1111111119389534, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.25, 0.1666666716337204, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.4444444477558136, 0.4444444477558136, 0.3611111044883728, 0.2777777910232544, 0.3611111044883728, 0.4444444477558136, 0.3888888955116272, 0.4722222089767456, 0.5277777910232544, 0.4444444477558136, 0.5277777910232544, 0.3611111044883728, NaN, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.4444444477558136, 0.5555555820465088, 0.6388888955116272, 0.6111111044883728, 0.6666666865348816, 0.6111111044883728, 0.6111111044883728, 0.7361111044883728, 0.6388888955116272, 0.5555555820465088, 0.6944444179534912, 0.6388888955116272, 0.7222222089767456, 0.5555555820465088, 0.6666666865348816, 0.6388888955116272, 0.7222222089767456, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.75, 0.6944444179534912, 0.8055555820465088, 0.6944444179534912, 0.6944444179534912, 0.6388888955116272, 0.6388888955116272, 0.75, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.75, 0.75, 0.7777777910232544, 0.8611111044883728, 0.8333333134651184, 0.8055555820465088, 0.8888888955116272, 0.8055555820465088, 0.7777777910232544, 0.9722222089767456, 0.8055555820465088, 0.9166666865348816, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456], [0.1111111119389534, 0.1388888955116272, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.3055555522441864, 0.3611111044883728, 0.4166666567325592, 0.3611111044883728, 0.3333333432674408, 0.3194444477558136, 0.4444444477558136, 0.4166666567325592, 0.4722222089767456, 0.3333333432674408, 0.4444444477558136, 0.5, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.4722222089767456, 0.4722222089767456, NaN, 0.5833333134651184, 0.5555555820465088, 0.5, 0.5277777910232544, 0.5833333134651184, 0.5277777910232544, 0.5555555820465088, 0.5277777910232544, 0.6388888955116272, 0.5, 0.6111111044883728, 0.6944444179534912, 0.5, 0.5833333134651184, 0.5833333134651184, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5833333134651184, 0.5833333134651184, 0.6388888955116272, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.7222222089767456, 0.6944444179534912, 0.8333333134651184, 0.7222222089767456, 0.6388888955116272, 0.6111111044883728, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.8055555820465088, 0.75, 0.75, 0.75, 0.7777777910232544, 0.8888888955116272, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.8333333134651184, 0.9722222089767456, 0.8333333134651184, 0.9444444179534912, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456, 1.0], [0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.3055555522441864, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3055555522441864, 0.4166666567325592, 0.3333333432674408, 0.3611111044883728, 0.3611111044883728, 0.4166666567325592, 0.4444444477558136, 0.4166666567325592, 0.4722222089767456, 0.5277777910232544, 0.3888888955116272, 0.4722222089767456, 0.4166666567325592, NaN, 0.5, 0.4722222089767456, 0.4722222089767456, 0.4722222089767456, 0.5555555820465088, 0.5, 0.4444444477558136, 0.5, 0.5277777910232544, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.5277777910232544, 0.6111111044883728, 0.6944444179534912, 0.5833333134651184, 0.6111111044883728, 0.6388888955116272, 0.6111111044883728, 0.5555555820465088, 0.75, 0.7222222089767456, 0.6111111044883728, 0.6388888955116272, 0.6666666865348816, 0.6944444179534912, 0.7777777910232544, 0.6944444179534912, 0.6944444179534912, 0.8055555820465088, 0.7777777910232544, 0.7777777910232544, 0.75, 0.6944444179534912, 0.6944444179534912, 0.8333333134651184, 0.75, 0.8611111044883728, 0.7777777910232544, 0.8055555820465088, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456], [0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1944444477558136, 0.1111111119389534, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.2777777910232544, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3611111044883728, 0.3333333432674408, 0.3611111044883728, 0.4166666567325592, 0.4444444477558136, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.4166666567325592, 0.4444444477558136, 0.5, NaN, 0.5, 0.5555555820465088, 0.5833333134651184, 0.4722222089767456, 0.5277777910232544, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.5833333134651184, 0.4722222089767456, 0.5833333134651184, 0.5, 0.5833333134651184, 0.6111111044883728, 0.6388888955116272, 0.6666666865348816, 0.6388888955116272, 0.6666666865348816, 0.6944444179534912, 0.6666666865348816, 0.75, 0.75, 0.6944444179534912, 0.6805555820465088, 0.75, 0.75, 0.7222222089767456, 0.7222222089767456, 0.8333333134651184, 0.9722222089767456, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8055555820465088, 0.8333333134651184, 0.9444444179534912, 0.8888888955116272, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 1.0], [0.0555555559694767, 0.2222222238779068, 0.2222222238779068, 0.1388888955116272, 0.25, 0.0833333358168602, 0.1944444477558136, 0.2222222238779068, 0.3888888955116272, 0.3333333432674408, 0.2777777910232544, 0.4166666567325592, 0.4722222089767456, 0.3888888955116272, 0.375, 0.4444444477558136, 0.4444444477558136, 0.5, 0.3888888955116272, 0.4722222089767456, 0.5, 0.5277777910232544, 0.4444444477558136, 0.5555555820465088, 0.5, 0.5277777910232544, 0.5, NaN, 0.5555555820465088, 0.5, 0.5, 0.5277777910232544, 0.5277777910232544, 0.4722222089767456, 0.5277777910232544, 0.4166666567325592, 0.5, 0.5416666865348816, 0.5694444179534912, 0.5555555820465088, 0.5833333134651184, 0.6111111044883728, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.5833333134651184, 0.5833333134651184, 0.6388888955116272, 0.5833333134651184, 0.6388888955116272, 0.6666666865348816, 0.6388888955116272, 0.6666666865348816, 0.6111111044883728, 0.6388888955116272, 0.6944444179534912, 0.6666666865348816, 0.6388888955116272, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.75, 0.75, 0.7638888955116272, 0.7777777910232544, 0.75, 0.8055555820465088, 0.75, 0.7777777910232544, 0.75, 0.8472222089767456, 0.7916666865348816, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456], [0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.1666666716337204, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3888888955116272, 0.3611111044883728, 0.25, 0.4166666567325592, 0.3888888955116272, 0.3611111044883728, 0.4444444477558136, 0.4722222089767456, 0.5277777910232544, 0.4444444477558136, 0.4444444477558136, NaN, 0.5, 0.3333333432674408, 0.5555555820465088, 0.4722222089767456, 0.5277777910232544, 0.5277777910232544, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.6111111044883728, 0.6666666865348816, 0.5833333134651184, 0.6944444179534912, 0.6111111044883728, 0.6111111044883728, 0.6666666865348816, 0.6666666865348816, 0.5833333134651184, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.75, 0.75, 0.6944444179534912, 0.6944444179534912, 0.6944444179534912, 0.75, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.7222222089767456, 0.75, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912, 1.0], [0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.1944444477558136, 0.1388888955116272, 0.1388888955116272, 0.25, 0.25, 0.25, 0.25, 0.25, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.3611111044883728, 0.3611111044883728, 0.4166666567325592, 0.5277777910232544, 0.4166666567325592, 0.5, 0.5, NaN, 0.4722222089767456, 0.5277777910232544, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5, 0.5555555820465088, 0.5, 0.5277777910232544, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.5833333134651184, 0.6111111044883728, 0.6944444179534912, 0.6388888955116272, 0.7777777910232544, 0.6944444179534912, 0.6944444179534912, 0.6666666865348816, 0.75, 0.75, 0.6944444179534912, 0.6944444179534912, 0.8055555820465088, 0.9166666865348816, 0.8611111044883728, 0.8055555820465088, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728, 0.7777777910232544, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.9166666865348816, 0.8888888955116272, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 1.0], [0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.0833333358168602, 0.1944444477558136, 0.1388888955116272, 0.1944444477558136, 0.25, 0.3333333432674408, 0.25, 0.3055555522441864, 0.4166666567325592, 0.3333333432674408, 0.25, 0.3333333432674408, 0.3611111044883728, 0.3611111044883728, 0.4722222089767456, 0.5, 0.3888888955116272, 0.3888888955116272, 0.3611111044883728, 0.3888888955116272, 0.4722222089767456, 0.4444444477558136, 0.5277777910232544, 0.5, 0.6666666865348816, 0.5277777910232544, NaN, 0.5277777910232544, 0.5, 0.625, 0.5555555820465088, 0.5277777910232544, 0.5, 0.5833333134651184, 0.5833333134651184, 0.5, 0.5833333134651184, 0.5833333134651184, 0.6388888955116272, 0.5833333134651184, 0.5555555820465088, 0.5833333134651184, 0.7222222089767456, 0.6944444179534912, 0.5833333134651184, 0.6388888955116272, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.6666666865348816, 0.6666666865348816, 0.75, 0.7777777910232544, 0.8055555820465088, 0.8055555820465088, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.8472222089767456, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.9722222089767456], [0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.1666666716337204, 0.0833333358168602, 0.1388888955116272, 0.2222222238779068, 0.3055555522441864, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.25, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.4166666567325592, 0.3333333432674408, 0.4444444477558136, 0.5, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, NaN, 0.5277777910232544, 0.4444444477558136, 0.5, 0.5833333134651184, 0.5277777910232544, 0.5, 0.4166666567325592, 0.5277777910232544, 0.5833333134651184, 0.5833333134651184, 0.6944444179534912, 0.6388888955116272, 0.6388888955116272, 0.6388888955116272, 0.5555555820465088, 0.6666666865348816, 0.8055555820465088, 0.6944444179534912, 0.6944444179534912, 0.75, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.75, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.75, 0.8333333134651184, 0.8611111044883728, 0.7777777910232544, 0.8333333134651184, 0.9166666865348816, 0.8333333134651184, 0.8055555820465088, 0.8888888955116272, 0.8055555820465088, 0.8333333134651184, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.02777777798473835, 0.1111111119389534, 0.1944444477558136, 0.25, 0.25, 0.1944444477558136, 0.25, 0.2777777910232544, 0.3333333432674408, 0.3055555522441864, 0.4166666567325592, 0.25, 0.3333333432674408, 0.4166666567325592, 0.3888888955116272, 0.4444444477558136, 0.2222222238779068, 0.3888888955116272, 0.4722222089767456, 0.4722222089767456, 0.4722222089767456, 0.5, 0.4722222089767456, 0.3888888955116272, 0.4722222089767456, 0.5555555820465088, 0.5277777910232544, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.5, 0.4722222089767456, NaN, 0.5277777910232544, 0.5, 0.5555555820465088, 0.5, 0.4722222089767456, 0.4444444477558136, 0.5277777910232544, 0.6111111044883728, 0.4444444477558136, 0.5, 0.5277777910232544, 0.5277777910232544, 0.5, 0.6944444179534912, 0.6388888955116272, 0.6388888955116272, 0.6111111044883728, 0.5555555820465088, 0.6666666865348816, 0.7222222089767456, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.7222222089767456, 0.6944444179534912, 0.8055555820465088, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.75, 0.8333333134651184, 0.75, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272, 0.7777777910232544, 0.75, 0.8611111044883728, 0.8055555820465088, 0.8611111044883728, 0.8888888955116272, 0.8055555820465088, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.9444444179534912, 0.9722222089767456], [0.0555555559694767, 0.1388888955116272, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.25, 0.3333333432674408, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.3611111044883728, 0.3611111044883728, 0.3333333432674408, 0.4166666567325592, 0.3888888955116272, 0.3888888955116272, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.5, 0.4722222089767456, 0.5277777910232544, 0.4722222089767456, 0.4444444477558136, 0.375, 0.5555555820465088, 0.4722222089767456, NaN, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.5555555820465088, 0.5277777910232544, 0.6111111044883728, 0.5833333134651184, 0.5833333134651184, 0.6388888955116272, 0.6944444179534912, 0.5833333134651184, 0.6388888955116272, 0.6388888955116272, 0.7222222089767456, 0.6527777910232544, 0.6111111044883728, 0.6944444179534912, 0.6944444179534912, 0.6944444179534912, 0.75, 0.8055555820465088, 0.75, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8055555820465088, 0.7777777910232544, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.8888888955116272, 0.8333333134651184, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 1.0], [0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.2777777910232544, 0.25, 0.2777777910232544, 0.3333333432674408, 0.2777777910232544, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3055555522441864, 0.3333333432674408, 0.4722222089767456, 0.4444444477558136, 0.3333333432674408, 0.2638888955116272, 0.5, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4444444477558136, 0.5, 0.5, 0.4722222089767456, NaN, 0.5277777910232544, 0.5, 0.5555555820465088, 0.4166666567325592, 0.5555555820465088, 0.5833333134651184, 0.6388888955116272, 0.6388888955116272, 0.6111111044883728, 0.5833333134651184, 0.6666666865348816, 0.5555555820465088, 0.5833333134651184, 0.8333333134651184, 0.6388888955116272, 0.6666666865348816, 0.7222222089767456, 0.6111111044883728, 0.6111111044883728, 0.6111111044883728, 0.7222222089767456, 0.6944444179534912, 0.7222222089767456, 0.6944444179534912, 0.7777777910232544, 0.75, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.75, 0.8611111044883728, 0.7777777910232544, 0.7222222089767456, 0.8333333134651184, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.8055555820465088, 0.8333333134651184, 0.9166666865348816, 0.9444444179534912, 0.9722222089767456, 0.9444444179534912, 1.0], [0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.0833333358168602, 0.0972222238779068, 0.2222222238779068, 0.1111111119389534, 0.1666666716337204, 0.25, 0.3333333432674408, 0.1944444477558136, 0.25, 0.2777777910232544, 0.3333333432674408, 0.2777777910232544, 0.25, 0.3611111044883728, 0.3055555522441864, 0.3888888955116272, 0.3055555522441864, 0.3055555522441864, 0.3888888955116272, 0.3611111044883728, 0.3888888955116272, 0.4722222089767456, 0.4166666567325592, 0.5833333134651184, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.4166666567325592, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, NaN, 0.5555555820465088, 0.4722222089767456, 0.5277777910232544, 0.5277777910232544, 0.5555555820465088, 0.5277777910232544, 0.5833333134651184, 0.6388888955116272, 0.5555555820465088, 0.6388888955116272, 0.6388888955116272, 0.7222222089767456, 0.5555555820465088, 0.7777777910232544, 0.6111111044883728, 0.75, 0.7222222089767456, 0.6666666865348816, 0.6666666865348816, 0.8055555820465088, 0.7222222089767456, 0.6944444179534912, 0.7777777910232544, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.7222222089767456, 0.75, 0.7638888955116272, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.9166666865348816, 0.8055555820465088, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9722222089767456, 1.0], [0.0833333358168602, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3055555522441864, 0.3888888955116272, 0.4166666567325592, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.3611111044883728, 0.4166666567325592, 0.3888888955116272, 0.3333333432674408, 0.4305555522441864, 0.4444444477558136, 0.3055555522441864, 0.4722222089767456, 0.5277777910232544, 0.5, 0.4722222089767456, 0.5, 0.5, 0.4722222089767456, 0.5, 0.4166666567325592, 0.5, 0.4444444477558136, NaN, 0.5, 0.5277777910232544, 0.5, 0.5, 0.4722222089767456, 0.5833333134651184, 0.5277777910232544, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.5833333134651184, 0.6388888955116272, 0.6666666865348816, 0.7222222089767456, 0.6666666865348816, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.6944444179534912, 0.6666666865348816, 0.7222222089767456, 0.6388888955116272, 0.6388888955116272, 0.7222222089767456, 0.75, 0.6944444179534912, 0.75, 0.75, 0.75, 0.7777777910232544, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.7777777910232544, 0.8611111044883728, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 1.0], [0.02777777798473835, 0.0555555559694767, 0.1666666716337204, 0.2222222238779068, 0.1111111119389534, 0.2777777910232544, 0.1944444477558136, 0.1944444477558136, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.3333333432674408, 0.3333333432674408, 0.3055555522441864, 0.2916666567325592, 0.3888888955116272, 0.3611111044883728, 0.3055555522441864, 0.4861111044883728, 0.3888888955116272, 0.4166666567325592, 0.3611111044883728, 0.4444444477558136, 0.3055555522441864, 0.5, 0.4166666567325592, 0.4166666567325592, 0.4583333432674408, 0.4166666567325592, 0.4444444477558136, 0.4166666567325592, 0.5, 0.5277777910232544, 0.5277777910232544, 0.4444444477558136, 0.5277777910232544, 0.5, NaN, 0.5, 0.4444444477558136, 0.4722222089767456, 0.5277777910232544, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.5555555820465088, 0.5694444179534912, 0.5833333134651184, 0.5833333134651184, 0.6388888955116272, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.6944444179534912, 0.6944444179534912, 0.5833333134651184, 0.6666666865348816, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.7361111044883728, 0.7222222089767456, 0.75, 0.75, 0.7777777910232544, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.75, 0.7777777910232544, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9722222089767456, 1.0], [0.0555555559694767, 0.0833333358168602, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.25, 0.25, 0.2361111044883728, 0.3611111044883728, 0.3888888955116272, 0.2777777910232544, 0.2916666567325592, 0.3333333432674408, 0.4166666567325592, 0.3055555522441864, 0.4722222089767456, 0.4166666567325592, 0.3888888955116272, 0.4444444477558136, 0.375, 0.3611111044883728, 0.4166666567325592, 0.4722222089767456, 0.5, 0.4305555522441864, 0.3888888955116272, 0.5, 0.4166666567325592, 0.5833333134651184, 0.5555555820465088, 0.4722222089767456, 0.5833333134651184, 0.4722222089767456, 0.4722222089767456, 0.5, NaN, 0.4722222089767456, 0.4722222089767456, 0.6111111044883728, 0.5555555820465088, 0.5555555820465088, 0.5277777910232544, 0.7222222089767456, 0.5555555820465088, 0.5555555820465088, 0.6111111044883728, 0.6388888955116272, 0.6944444179534912, 0.5277777910232544, 0.5555555820465088, 0.6388888955116272, 0.6388888955116272, 0.6111111044883728, 0.6388888955116272, 0.6944444179534912, 0.6111111044883728, 0.7222222089767456, 0.7222222089767456, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.8333333134651184, 0.7222222089767456, 0.6944444179534912, 0.6388888955116272, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.75, 0.8055555820465088, 0.8611111044883728, 0.8055555820465088, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 1.0], [0.0833333358168602, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1388888955116272, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.25, 0.2777777910232544, 0.3055555522441864, 0.2777777910232544, 0.2222222238779068, 0.1666666716337204, 0.3055555522441864, 0.3333333432674408, 0.1666666716337204, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, 0.4444444477558136, 0.3055555522441864, 0.2777777910232544, 0.4166666567325592, 0.3888888955116272, 0.4166666567325592, 0.4444444477558136, 0.3333333432674408, 0.4722222089767456, 0.5, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4444444477558136, 0.4722222089767456, 0.5, 0.5555555820465088, 0.5277777910232544, NaN, 0.5277777910232544, 0.5833333134651184, 0.5277777910232544, 0.5555555820465088, 0.5555555820465088, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.6111111044883728, 0.5416666865348816, 0.7222222089767456, 0.6944444179534912, 0.6388888955116272, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.8055555820465088, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.8055555820465088, 0.75, 0.7222222089767456, 0.8055555820465088, 0.7222222089767456, 0.7222222089767456, 0.9166666865348816, 0.8055555820465088, 0.8333333134651184, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912], [0.0555555559694767, 0.0555555559694767, 0.1388888955116272, 0.0833333358168602, 0.1388888955116272, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.25, 0.25, 0.2222222238779068, 0.25, 0.25, 0.3055555522441864, 0.3055555522441864, 0.25, 0.3333333432674408, 0.4166666567325592, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.4444444477558136, 0.3333333432674408, 0.3055555522441864, 0.3888888955116272, 0.4166666567325592, 0.4166666567325592, 0.4166666567325592, 0.4166666567325592, 0.4166666567325592, 0.3888888955116272, 0.4444444477558136, 0.4166666567325592, 0.4444444477558136, 0.5, 0.5277777910232544, 0.5277777910232544, 0.4722222089767456, NaN, 0.4722222089767456, 0.5277777910232544, 0.5, 0.5277777910232544, 0.5555555820465088, 0.6111111044883728, 0.7222222089767456, 0.5833333134651184, 0.5555555820465088, 0.6111111044883728, 0.6666666865348816, 0.75, 0.6388888955116272, 0.6388888955116272, 0.75, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.7777777910232544, 0.6944444179534912, 0.8055555820465088, 0.8055555820465088, 0.7777777910232544, 0.9444444179534912, 0.8888888955116272, 0.8888888955116272, 0.75, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816, 0.9166666865348816, 0.8888888955116272, 0.8611111044883728, 0.8888888955116272, 0.9722222089767456, 1.0], [0.0833333358168602, 0.1944444477558136, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.2777777910232544, 0.1944444477558136, 0.3055555522441864, 0.3055555522441864, 0.1944444477558136, 0.2222222238779068, 0.25, 0.3055555522441864, 0.25, 0.3611111044883728, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.4166666567325592, 0.3611111044883728, 0.3888888955116272, 0.3055555522441864, 0.3888888955116272, 0.4166666567325592, 0.4166666567325592, 0.5555555820465088, 0.4722222089767456, 0.3611111044883728, 0.4722222089767456, 0.5277777910232544, 0.4722222089767456, 0.3888888955116272, 0.4166666567325592, 0.5277777910232544, NaN, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.7222222089767456, 0.6666666865348816, 0.6666666865348816, 0.5, 0.7777777910232544, 0.6111111044883728, 0.6944444179534912, 0.6111111044883728, 0.6111111044883728, 0.6111111044883728, 0.6388888955116272, 0.6111111044883728, 0.6944444179534912, 0.75, 0.7222222089767456, 0.75, 0.6388888955116272, 0.6666666865348816, 0.6944444179534912, 0.75, 0.75, 0.7777777910232544, 0.7222222089767456, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.75, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9722222089767456, 0.9722222089767456, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456], [0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1666666716337204, 0.1111111119389534, 0.25, 0.1944444477558136, 0.3055555522441864, 0.3055555522441864, 0.2222222238779068, 0.25, 0.25, 0.3055555522441864, 0.25, 0.3333333432674408, 0.3333333432674408, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3611111044883728, 0.3333333432674408, 0.3888888955116272, 0.3333333432674408, 0.4722222089767456, 0.3888888955116272, 0.3888888955116272, 0.3611111044883728, 0.3055555522441864, 0.5, 0.3888888955116272, 0.3611111044883728, 0.4166666567325592, 0.4166666567325592, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, NaN, 0.3611111044883728, 0.5833333134651184, 0.5555555820465088, 0.6111111044883728, 0.6111111044883728, 0.5277777910232544, 0.5555555820465088, 0.6111111044883728, 0.6388888955116272, 0.6666666865348816, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.6388888955116272, 0.75, 0.6388888955116272, 0.6666666865348816, 0.7222222089767456, 0.6944444179534912, 0.6666666865348816, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8333333134651184, 0.9166666865348816, 0.8055555820465088, 0.9166666865348816, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456, 0.9444444179534912, 1.0], [0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.2222222238779068, 0.25, 0.2777777910232544, 0.2777777910232544, 0.1944444477558136, 0.25, 0.25, 0.1944444477558136, 0.2777777910232544, 0.25, 0.3611111044883728, 0.3611111044883728, 0.3055555522441864, 0.2777777910232544, 0.4166666567325592, 0.3611111044883728, 0.3611111044883728, 0.4722222089767456, 0.3888888955116272, 0.4166666567325592, 0.4166666567325592, 0.3611111044883728, 0.4722222089767456, 0.4166666567325592, 0.3888888955116272, 0.3611111044883728, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4444444477558136, 0.5, 0.5555555820465088, 0.6388888955116272, NaN, 0.5277777910232544, 0.5833333134651184, 0.5, 0.5277777910232544, 0.6944444179534912, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6111111044883728, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7222222089767456, 0.75, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.75, 0.7777777910232544, 0.8055555820465088, 0.7777777910232544, 0.75, 0.8333333134651184, 0.8055555820465088, 0.8611111044883728, 0.75, 0.8333333134651184, 0.7777777910232544, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912, 0.9166666865348816, 1.0], [0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.2222222238779068, 0.2777777910232544, 0.25, 0.2222222238779068, 0.25, 0.2777777910232544, 0.3055555522441864, 0.25, 0.2777777910232544, 0.2222222238779068, 0.25, 0.2777777910232544, 0.25, 0.3055555522441864, 0.4166666567325592, 0.3888888955116272, 0.3333333432674408, 0.4166666567325592, 0.3333333432674408, 0.3888888955116272, 0.4444444477558136, 0.3611111044883728, 0.4722222089767456, 0.4166666567325592, 0.4166666567325592, 0.4444444477558136, 0.3611111044883728, 0.4166666567325592, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.4444444477558136, 0.4166666567325592, 0.4722222089767456, NaN, 0.4722222089767456, 0.5277777910232544, 0.5833333134651184, 0.5, 0.5277777910232544, 0.7222222089767456, 0.6666666865348816, 0.6111111044883728, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.75, 0.75, 0.7222222089767456, 0.6388888955116272, 0.6666666865348816, 0.7222222089767456, 0.6944444179534912, 0.7222222089767456, 0.75, 0.7777777910232544, 0.75, 0.7222222089767456, 0.9444444179534912, 0.7222222089767456, 0.8888888955116272, 0.75, 0.8611111044883728, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456, 0.9166666865348816, 0.9444444179534912], [0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.0833333358168602, 0.0833333358168602, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.1388888955116272, 0.1944444477558136, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.3611111044883728, 0.2777777910232544, 0.3055555522441864, 0.25, 0.2222222238779068, 0.25, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.4444444477558136, 0.3055555522441864, 0.4166666567325592, 0.3333333432674408, 0.3055555522441864, 0.4166666567325592, 0.3611111044883728, 0.5, 0.3611111044883728, 0.3333333432674408, 0.3611111044883728, 0.4166666567325592, 0.4444444477558136, 0.2777777910232544, 0.4166666567325592, 0.4444444477558136, 0.2777777910232544, 0.4444444477558136, 0.4166666567325592, 0.5277777910232544, NaN, 0.6111111044883728, 0.5555555820465088, 0.4444444477558136, 0.5833333134651184, 0.5555555820465088, 0.5555555820465088, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.6111111044883728, 0.5555555820465088, 0.5833333134651184, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.6111111044883728, 0.6666666865348816, 0.6666666865348816, 0.75, 0.6666666865348816, 0.8055555820465088, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.8055555820465088, 0.75, 0.8333333134651184, 0.8888888955116272, 0.8611111044883728, 0.9166666865348816, 0.9444444179534912, 0.9166666865348816, 0.9166666865348816, 0.9722222089767456], [0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.1388888955116272, 0.0555555559694767, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.25, 0.1944444477558136, 0.2777777910232544, 0.25, 0.1666666716337204, 0.2222222238779068, 0.1666666716337204, 0.2777777910232544, 0.1944444477558136, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3055555522441864, 0.25, 0.3333333432674408, 0.4166666567325592, 0.4166666567325592, 0.3611111044883728, 0.2777777910232544, 0.4444444477558136, 0.3055555522441864, 0.3055555522441864, 0.4444444477558136, 0.3611111044883728, 0.3888888955116272, 0.4305555522441864, 0.4444444477558136, 0.3888888955116272, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, 0.5, 0.4722222089767456, 0.3888888955116272, NaN, 0.5833333134651184, 0.5, 0.4722222089767456, 0.4722222089767456, 0.5277777910232544, 0.5277777910232544, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.5833333134651184, 0.5277777910232544, 0.6111111044883728, 0.75, 0.7222222089767456, 0.75, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.6666666865348816, 0.7777777910232544, 0.8055555820465088, 0.75, 0.7777777910232544, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.9166666865348816, 0.9722222089767456], [0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1666666716337204, 0.0555555559694767, 0.1666666716337204, 0.1944444477558136, 0.25, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.2777777910232544, 0.2222222238779068, 0.25, 0.3055555522441864, 0.1666666716337204, 0.2222222238779068, 0.2777777910232544, 0.25, 0.3055555522441864, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.25, 0.3611111044883728, 0.3333333432674408, 0.2222222238779068, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.4166666567325592, 0.4166666567325592, 0.2777777910232544, 0.3888888955116272, 0.4166666567325592, 0.4444444477558136, 0.3888888955116272, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.4722222089767456, 0.4166666567325592, 0.4444444477558136, 0.4166666567325592, NaN, 0.5277777910232544, 0.4166666567325592, 0.4861111044883728, 0.5, 0.6805555820465088, 0.6388888955116272, 0.6388888955116272, 0.7222222089767456, 0.5277777910232544, 0.5, 0.5, 0.6388888955116272, 0.6666666865348816, 0.8055555820465088, 0.6388888955116272, 0.7222222089767456, 0.6388888955116272, 0.7222222089767456, 0.75, 0.9166666865348816, 0.6388888955116272, 0.75, 0.8055555820465088, 0.8611111044883728, 0.75, 0.9444444179534912, 0.75, 0.8333333134651184, 0.8611111044883728, 0.8333333134651184, 0.9444444179534912, 1.0], [0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.2777777910232544, 0.25, 0.25, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.25, 0.1388888955116272, 0.25, 0.25, 0.2222222238779068, 0.3888888955116272, 0.25, 0.4166666567325592, 0.3055555522441864, 0.3055555522441864, 0.4166666567325592, 0.1944444477558136, 0.3611111044883728, 0.3611111044883728, 0.1666666716337204, 0.4444444477558136, 0.4166666567325592, 0.4166666567325592, 0.3888888955116272, 0.3888888955116272, 0.4166666567325592, 0.5, 0.4722222089767456, 0.3055555522441864, 0.5, 0.5555555820465088, 0.5, 0.4722222089767456, NaN, 0.5277777910232544, 0.4444444477558136, 0.6111111044883728, 0.5277777910232544, 0.5, 0.5, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.75, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.75, 0.75, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.6944444179534912, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.9722222089767456, 0.9444444179534912, 0.9722222089767456], [0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.25, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3055555522441864, 0.3611111044883728, 0.2777777910232544, 0.3055555522441864, 0.3611111044883728, 0.3055555522441864, 0.3888888955116272, 0.3611111044883728, 0.3611111044883728, 0.2222222238779068, 0.3611111044883728, 0.3611111044883728, 0.3611111044883728, 0.4583333432674408, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4722222089767456, 0.4166666567325592, 0.5277777910232544, 0.5833333134651184, 0.4722222089767456, NaN, 0.5555555820465088, 0.5555555820465088, 0.5833333134651184, 0.5555555820465088, 0.5555555820465088, 0.6111111044883728, 0.5, 0.5, 0.6388888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.6666666865348816, 0.75, 0.6944444179534912, 0.6944444179534912, 0.7222222089767456, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8055555820465088, 0.8611111044883728, 0.8888888955116272, 0.9444444179534912, 0.9444444179534912, 0.9722222089767456, 0.9166666865348816, 0.9722222089767456], [0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.1111111119389534, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.0833333358168602, 0.1944444477558136, 0.0833333358168602, 0.1388888955116272, 0.2777777910232544, 0.1944444477558136, 0.1666666716337204, 0.1111111119389534, 0.3055555522441864, 0.2777777910232544, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.3055555522441864, 0.3333333432674408, 0.3194444477558136, 0.3333333432674408, 0.25, 0.3333333432674408, 0.2777777910232544, 0.3055555522441864, 0.4444444477558136, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.3333333432674408, 0.3611111044883728, 0.3055555522441864, 0.2777777910232544, 0.3888888955116272, 0.3888888955116272, 0.3888888955116272, 0.4444444477558136, 0.2777777910232544, 0.4444444477558136, 0.5277777910232544, 0.5138888955116272, 0.5555555820465088, 0.4444444477558136, NaN, 0.5, 0.5555555820465088, 0.5, 0.5, 0.6388888955116272, 0.5694444179534912, 0.5555555820465088, 0.6388888955116272, 0.6944444179534912, 0.6666666865348816, 0.6388888955116272, 0.6944444179534912, 0.6388888955116272, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.6944444179534912, 0.9166666865348816, 0.7777777910232544, 0.8333333134651184, 0.6388888955116272, 0.8055555820465088, 0.8333333134651184, 0.8333333134651184, 0.8055555820465088, 0.8333333134651184, 0.9166666865348816, 0.9166666865348816, 0.9444444179534912], [0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.25, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.2222222238779068, 0.1666666716337204, 0.1944444477558136, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.1666666716337204, 0.3055555522441864, 0.25, 0.3611111044883728, 0.25, 0.25, 0.2777777910232544, 0.25, 0.3333333432674408, 0.3472222089767456, 0.2777777910232544, 0.25, 0.2777777910232544, 0.4166666567325592, 0.4722222089767456, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3611111044883728, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4722222089767456, 0.5, 0.3888888955116272, 0.4444444477558136, 0.5, NaN, 0.5, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4722222089767456, 0.5277777910232544, 0.6944444179534912, 0.6388888955116272, 0.6388888955116272, 0.6111111044883728, 0.6111111044883728, 0.7222222089767456, 0.7222222089767456, 0.6388888955116272, 0.75, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.8611111044883728, 0.75, 0.9722222089767456, 0.8055555820465088, 0.8611111044883728, 0.8333333134651184, 0.8611111044883728, 0.8888888955116272, 0.8888888955116272, 0.9444444179534912], [0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.2222222238779068, 0.25, 0.1666666716337204, 0.3055555522441864, 0.2222222238779068, 0.2777777910232544, 0.3055555522441864, 0.2222222238779068, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.2222222238779068, 0.25, 0.3333333432674408, 0.3055555522441864, 0.25, 0.3055555522441864, 0.3333333432674408, 0.2777777910232544, 0.3888888955116272, 0.3888888955116272, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.4444444477558136, 0.3611111044883728, 0.25, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, 0.3888888955116272, 0.4166666567325592, 0.4722222089767456, 0.3194444477558136, 0.4722222089767456, 0.4166666567325592, 0.4444444477558136, 0.5, NaN, 0.6111111044883728, 0.6111111044883728, 0.5277777910232544, 0.4444444477558136, 0.4722222089767456, 0.5, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.6388888955116272, 0.7222222089767456, 0.6388888955116272, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5833333134651184, 0.7777777910232544, 0.7777777910232544, 0.8333333134651184, 0.6944444179534912, 0.8888888955116272, 0.7222222089767456, 0.75, 0.75, 0.8333333134651184, 0.9722222089767456, 0.9722222089767456], [0.0833333358168602, 0.1388888955116272, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.25, 0.2222222238779068, 0.25, 0.25, 0.2361111044883728, 0.1944444477558136, 0.3055555522441864, 0.25, 0.3055555522441864, 0.25, 0.3333333432674408, 0.3888888955116272, 0.2777777910232544, 0.25, 0.3055555522441864, 0.2777777910232544, 0.3472222089767456, 0.3611111044883728, 0.3611111044883728, 0.3055555522441864, 0.2777777910232544, 0.3888888955116272, 0.3055555522441864, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3888888955116272, 0.3055555522441864, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, 0.3055555522441864, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.3888888955116272, 0.3888888955116272, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.3888888955116272, 0.3611111044883728, 0.5, 0.4444444477558136, 0.5, 0.4444444477558136, 0.3888888955116272, NaN, 0.5277777910232544, 0.3333333432674408, 0.4722222089767456, 0.4166666567325592, 0.5, 0.4722222089767456, 0.5, 0.4027777910232544, 0.5, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.5833333134651184, 0.5555555820465088, 0.5833333134651184, 0.6111111044883728, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.6111111044883728, 0.7222222089767456, 0.7222222089767456, 0.75, 0.8333333134651184, 0.9722222089767456], [0.0555555559694767, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.25, 0.1944444477558136, 0.25, 0.2222222238779068, 0.2083333283662796, 0.2222222238779068, 0.3055555522441864, 0.25, 0.3055555522441864, 0.2222222238779068, 0.3333333432674408, 0.3888888955116272, 0.2777777910232544, 0.25, 0.3055555522441864, 0.2777777910232544, 0.3472222089767456, 0.3611111044883728, 0.3888888955116272, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3055555522441864, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3888888955116272, 0.3055555522441864, 0.3888888955116272, 0.3333333432674408, 0.3888888955116272, 0.3055555522441864, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.3888888955116272, 0.3888888955116272, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.3888888955116272, 0.3611111044883728, 0.5, 0.4444444477558136, 0.5, 0.4444444477558136, 0.3888888955116272, 0.4722222089767456, NaN, 0.3333333432674408, 0.4722222089767456, 0.4166666567325592, 0.5, 0.4722222089767456, 0.5, 0.4027777910232544, 0.5, 0.4722222089767456, 0.5277777910232544, 0.5555555820465088, 0.5833333134651184, 0.5555555820465088, 0.5833333134651184, 0.6111111044883728, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.6944444179534912, 0.6111111044883728, 0.7222222089767456, 0.7222222089767456, 0.75, 0.8333333134651184, 0.9722222089767456], [0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.1111111119389534, 0.1388888955116272, 0.0833333358168602, 0.1944444477558136, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.25, 0.2222222238779068, 0.2777777910232544, 0.2222222238779068, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1388888955116272, 0.3055555522441864, 0.25, 0.2777777910232544, 0.1944444477558136, 0.1666666716337204, 0.3055555522441864, 0.25, 0.1944444477558136, 0.25, 0.2222222238779068, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.1944444477558136, 0.3055555522441864, 0.4166666567325592, 0.3888888955116272, 0.3055555522441864, 0.25, 0.3611111044883728, 0.3055555522441864, 0.3333333432674408, 0.3333333432674408, 0.3888888955116272, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.4444444477558136, 0.4722222089767456, 0.6666666865348816, 0.6666666865348816, NaN, 0.4722222089767456, 0.5, 0.4444444477558136, 0.5555555820465088, 0.5833333134651184, 0.7777777910232544, 0.5277777910232544, 0.6388888955116272, 0.5555555820465088, 0.6388888955116272, 0.6944444179534912, 0.7777777910232544, 0.6666666865348816, 0.7222222089767456, 0.6944444179534912, 0.8888888955116272, 0.6666666865348816, 0.8055555820465088, 0.6388888955116272, 0.7777777910232544, 0.75, 0.8055555820465088, 0.9166666865348816, 0.9722222089767456], [0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1944444477558136, 0.1111111119389534, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.2777777910232544, 0.1666666716337204, 0.1388888955116272, 0.0555555559694767, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.02777777798473835, 0.3333333432674408, 0.1944444477558136, 0.0833333358168602, 0.2222222238779068, 0.25, 0.2777777910232544, 0.25, 0.3055555522441864, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.3055555522441864, 0.3055555522441864, 0.3888888955116272, 0.3611111044883728, 0.2777777910232544, 0.25, 0.4444444477558136, 0.4166666567325592, 0.4722222089767456, 0.4166666567325592, 0.5, 0.4305555522441864, 0.5277777910232544, 0.5555555820465088, 0.5277777910232544, 0.5277777910232544, 0.5277777910232544, NaN, 0.4722222089767456, 0.5555555820465088, 0.6388888955116272, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.6111111044883728, 0.75, 0.6111111044883728, 0.6944444179534912, 0.6944444179534912, 0.6666666865348816, 0.7222222089767456, 0.75, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.75, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.9166666865348816, 0.9166666865348816], [0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.1666666716337204, 0.2222222238779068, 0.1666666716337204, 0.1388888955116272, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1666666716337204, 0.2777777910232544, 0.2222222238779068, 0.1111111119389534, 0.3611111044883728, 0.1666666716337204, 0.1388888955116272, 0.1944444477558136, 0.2222222238779068, 0.3055555522441864, 0.1944444477558136, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3055555522441864, 0.2222222238779068, 0.3333333432674408, 0.3055555522441864, 0.25, 0.25, 0.25, 0.4166666567325592, 0.4722222089767456, 0.5, 0.3888888955116272, 0.5, 0.4444444477558136, 0.4722222089767456, 0.5277777910232544, 0.5833333134651184, 0.5833333134651184, 0.5, 0.5277777910232544, NaN, 0.5555555820465088, 0.6111111044883728, 0.6111111044883728, 0.5833333134651184, 0.5555555820465088, 0.5833333134651184, 0.7222222089767456, 0.6388888955116272, 0.7222222089767456, 0.6666666865348816, 0.6944444179534912, 0.6944444179534912, 0.6666666865348816, 0.7222222089767456, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8611111044883728, 0.8611111044883728, 0.9444444179534912], [0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.0833333358168602, 0.1666666716337204, 0.1944444477558136, 0.1111111119389534, 0.1666666716337204, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.1666666716337204, 0.25, 0.1666666716337204, 0.1944444477558136, 0.25, 0.2222222238779068, 0.3055555522441864, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.25, 0.3055555522441864, 0.2222222238779068, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.1944444477558136, 0.3055555522441864, 0.25, 0.3611111044883728, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.5, 0.3888888955116272, 0.3611111044883728, 0.3611111044883728, 0.3055555522441864, 0.5, 0.5, 0.5, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, NaN, 0.6111111044883728, 0.6111111044883728, 0.5833333134651184, 0.5, 0.6111111044883728, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6944444179534912, 0.6666666865348816, 0.6944444179534912, 0.75, 0.6666666865348816, 0.75, 0.75, 0.75, 0.8055555820465088, 0.8055555820465088, 0.8611111044883728, 0.8611111044883728, 0.9166666865348816], [0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.2222222238779068, 0.25, 0.25, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1805555522441864, 0.2222222238779068, 0.1944444477558136, 0.2777777910232544, 0.2222222238779068, 0.25, 0.25, 0.2777777910232544, 0.25, 0.25, 0.3055555522441864, 0.2222222238779068, 0.3055555522441864, 0.2777777910232544, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.2777777910232544, 0.2222222238779068, 0.3055555522441864, 0.3611111044883728, 0.3055555522441864, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.2777777910232544, 0.3333333432674408, 0.2777777910232544, 0.3611111044883728, 0.3055555522441864, 0.25, 0.3611111044883728, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.3611111044883728, 0.3333333432674408, 0.5277777910232544, 0.5277777910232544, 0.4444444477558136, 0.3611111044883728, 0.3888888955116272, 0.3888888955116272, NaN, 0.4583333432674408, 0.4444444477558136, 0.5277777910232544, 0.4166666567325592, 0.4166666567325592, 0.5277777910232544, 0.5555555820465088, 0.5833333134651184, 0.5, 0.9444444179534912, 0.5277777910232544, 0.6111111044883728, 0.6111111044883728, 0.6111111044883728, 0.5555555820465088, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728], [0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.2222222238779068, 0.2222222238779068, 0.25, 0.2777777910232544, 0.25, 0.25, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.25, 0.1944444477558136, 0.2777777910232544, 0.25, 0.25, 0.25, 0.2777777910232544, 0.25, 0.25, 0.3055555522441864, 0.2222222238779068, 0.2777777910232544, 0.25, 0.2222222238779068, 0.3055555522441864, 0.25, 0.3055555522441864, 0.2777777910232544, 0.25, 0.3055555522441864, 0.3611111044883728, 0.3055555522441864, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.25, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3055555522441864, 0.2777777910232544, 0.3333333432674408, 0.25, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, 0.3333333432674408, 0.5, 0.5, 0.4166666567325592, 0.3611111044883728, 0.3888888955116272, 0.3888888955116272, 0.5416666865348816, NaN, 0.4444444477558136, 0.5277777910232544, 0.4444444477558136, 0.4444444477558136, 0.5277777910232544, 0.5555555820465088, 0.5555555820465088, 0.5277777910232544, 0.9722222089767456, 0.5, 0.5833333134651184, 0.5833333134651184, 0.5833333134651184, 0.5555555820465088, 0.6944444179534912, 0.6944444179534912, 0.7777777910232544, 0.7777777910232544, 0.8611111044883728], [0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.0555555559694767, 0.1111111119389534, 0.1666666716337204, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1944444477558136, 0.2222222238779068, 0.25, 0.1666666716337204, 0.1111111119389534, 0.25, 0.1666666716337204, 0.1388888955116272, 0.2777777910232544, 0.1666666716337204, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.2222222238779068, 0.2777777910232544, 0.2638888955116272, 0.3333333432674408, 0.2777777910232544, 0.2222222238779068, 0.3611111044883728, 0.3055555522441864, 0.2222222238779068, 0.2777777910232544, 0.3888888955116272, 0.25, 0.1944444477558136, 0.2777777910232544, 0.3333333432674408, 0.3611111044883728, 0.3888888955116272, 0.3055555522441864, 0.5972222089767456, 0.5972222089767456, 0.2222222238779068, 0.4166666567325592, 0.4166666567325592, 0.4166666567325592, 0.5555555820465088, 0.5555555820465088, NaN, 0.5, 0.5833333134651184, 0.5833333134651184, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5277777910232544, 0.6666666865348816, 0.6944444179534912, 0.7638888955116272, 0.6944444179534912, 0.8055555820465088, 0.6111111044883728, 0.7777777910232544, 0.75, 0.7777777910232544, 0.8888888955116272, 0.9722222089767456], [0.0, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.1666666716337204, 0.2222222238779068, 0.1944444477558136, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1111111119389534, 0.25, 0.1388888955116272, 0.2222222238779068, 0.25, 0.2222222238779068, 0.25, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.1388888955116272, 0.25, 0.2222222238779068, 0.1944444477558136, 0.2777777910232544, 0.25, 0.2777777910232544, 0.3055555522441864, 0.1944444477558136, 0.3055555522441864, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3055555522441864, 0.3333333432674408, 0.3333333432674408, 0.3611111044883728, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.3888888955116272, 0.3611111044883728, 0.5, 0.5, 0.4722222089767456, 0.3888888955116272, 0.4444444477558136, 0.5, 0.4722222089767456, 0.4722222089767456, 0.5, NaN, 0.4722222089767456, 0.5555555820465088, 0.7361111044883728, 0.5833333134651184, 0.5277777910232544, 0.5277777910232544, 0.5, 0.6111111044883728, 0.6388888955116272, 0.6388888955116272, 0.6111111044883728, 0.8611111044883728, 0.7777777910232544, 0.7777777910232544, 0.8055555820465088, 0.75, 0.8888888955116272], [0.02777777798473835, 0.0, 0.0, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.1111111119389534, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.1527777761220932, 0.02777777798473835, 0.0555555559694767, 0.1527777761220932, 0.1388888955116272, 0.1527777761220932, 0.0555555559694767, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.2083333283662796, 0.1666666716337204, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.2361111044883728, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.25, 0.3055555522441864, 0.25, 0.3055555522441864, 0.25, 0.1944444477558136, 0.3055555522441864, 0.2777777910232544, 0.25, 0.2777777910232544, 0.3333333432674408, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.2777777910232544, 0.2777777910232544, 0.5277777910232544, 0.5277777910232544, 0.3611111044883728, 0.3888888955116272, 0.4166666567325592, 0.3888888955116272, 0.5833333134651184, 0.5555555820465088, 0.4166666567325592, 0.5277777910232544, NaN, 0.5555555820465088, 0.5277777910232544, 0.5, 0.5833333134651184, 0.5277777910232544, 0.7222222089767456, 0.6666666865348816, 0.6388888955116272, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.8055555820465088, 0.8055555820465088, 0.8333333134651184, 0.8611111044883728, 0.9722222089767456], [0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.1388888955116272, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.0555555559694767, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1388888955116272, 0.2222222238779068, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.1666666716337204, 0.25, 0.2222222238779068, 0.25, 0.2361111044883728, 0.25, 0.25, 0.1666666716337204, 0.2777777910232544, 0.1944444477558136, 0.25, 0.2222222238779068, 0.2222222238779068, 0.25, 0.25, 0.2777777910232544, 0.3611111044883728, 0.25, 0.25, 0.3333333432674408, 0.2777777910232544, 0.3611111044883728, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.25, 0.2777777910232544, 0.3333333432674408, 0.5833333134651184, 0.5555555820465088, 0.4166666567325592, 0.4444444477558136, 0.4444444477558136, NaN, 0.4166666567325592, 0.5, 0.4722222089767456, 0.5555555820465088, 0.6666666865348816, 0.5277777910232544, 0.5277777910232544, 0.5555555820465088, 0.75, 0.5833333134651184, 0.7222222089767456, 0.7222222089767456, 0.8055555820465088, 0.8055555820465088, 0.8888888955116272], [0.0, 0.0, 0.0555555559694767, 0.02777777798473835, 0.0, 0.0, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1388888955116272, 0.1111111119389534, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1944444477558136, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1666666716337204, 0.25, 0.1388888955116272, 0.1666666716337204, 0.2222222238779068, 0.0833333358168602, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.25, 0.2222222238779068, 0.2777777910232544, 0.1944444477558136, 0.2222222238779068, 0.25, 0.2777777910232544, 0.1944444477558136, 0.2222222238779068, 0.3333333432674408, 0.2777777910232544, 0.2777777910232544, 0.25, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3611111044883728, 0.3888888955116272, 0.3611111044883728, 0.4444444477558136, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.2638888955116272, 0.4722222089767456, 0.5833333134651184, NaN, 0.5833333134651184, 0.5, 0.5277777910232544, 0.5, 0.6388888955116272, 0.5833333134651184, 0.6111111044883728, 0.5555555820465088, 0.7638888955116272, 0.75, 0.7222222089767456, 0.7777777910232544, 0.7222222089767456, 0.8888888955116272], [0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.1111111119389534, 0.2222222238779068, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1388888955116272, 0.25, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.0555555559694767, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.25, 0.1944444477558136, 0.3333333432674408, 0.25, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.25, 0.3333333432674408, 0.4166666567325592, 0.4166666567325592, 0.3055555522441864, 0.3055555522441864, 0.2777777910232544, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4166666567325592, 0.5, 0.5, 0.4166666567325592, NaN, 0.5277777910232544, 0.5, 0.6388888955116272, 0.5555555820465088, 0.5833333134651184, 0.5, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7222222089767456, 0.7777777910232544, 0.9166666865348816], [0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.1111111119389534, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.1111111119389534, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.1944444477558136, 0.1111111119389534, 0.0833333358168602, 0.1944444477558136, 0.0555555559694767, 0.1111111119389534, 0.1666666716337204, 0.1388888955116272, 0.1388888955116272, 0.2222222238779068, 0.1388888955116272, 0.0833333358168602, 0.1111111119389534, 0.25, 0.2222222238779068, 0.1388888955116272, 0.1944444477558136, 0.1944444477558136, 0.1111111119389534, 0.2222222238779068, 0.2777777910232544, 0.1666666716337204, 0.2222222238779068, 0.2777777910232544, 0.3611111044883728, 0.2777777910232544, 0.1111111119389534, 0.2777777910232544, 0.2222222238779068, 0.25, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.0833333358168602, 0.3055555522441864, 0.2777777910232544, 0.3055555522441864, 0.2777777910232544, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.2222222238779068, 0.3055555522441864, 0.3333333432674408, 0.3055555522441864, 0.4166666567325592, 0.4444444477558136, 0.3333333432674408, 0.4722222089767456, 0.4166666567325592, 0.5277777910232544, 0.5, 0.4722222089767456, NaN, 0.3888888955116272, 0.6388888955116272, 0.5555555820465088, 0.6111111044883728, 0.5555555820465088, 0.6111111044883728, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7222222089767456, 0.8055555820465088, 0.9444444179534912], [0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.1111111119389534, 0.1388888955116272, 0.1111111119389534, 0.0555555559694767, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.0833333358168602, 0.1666666716337204, 0.02777777798473835, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1111111119389534, 0.2222222238779068, 0.1666666716337204, 0.1666666716337204, 0.0833333358168602, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.0555555559694767, 0.1944444477558136, 0.1944444477558136, 0.3611111044883728, 0.2777777910232544, 0.1666666716337204, 0.0833333358168602, 0.2777777910232544, 0.4166666567325592, 0.4166666567325592, 0.4166666567325592, 0.3333333432674408, 0.3333333432674408, 0.3055555522441864, 0.3333333432674408, 0.5, 0.4722222089767456, 0.4722222089767456, 0.4722222089767456, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.5, 0.6111111044883728, NaN, 0.6111111044883728, 0.6111111044883728, 0.5555555820465088, 0.5277777910232544, 0.6111111044883728, 0.5833333134651184, 0.6666666865348816, 0.7222222089767456, 0.8055555820465088, 0.7777777910232544, 0.8888888955116272], [0.0555555559694767, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1111111119389534, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.1388888955116272, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.1944444477558136, 0.25, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.25, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.1666666716337204, 0.25, 0.1944444477558136, 0.1944444477558136, 0.25, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.2777777910232544, 0.1944444477558136, 0.25, 0.25, 0.1944444477558136, 0.2222222238779068, 0.2222222238779068, 0.2777777910232544, 0.2222222238779068, 0.3888888955116272, 0.3888888955116272, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.3055555522441864, 0.0555555559694767, 0.02777777798473835, 0.3333333432674408, 0.5, 0.2777777910232544, 0.3333333432674408, 0.5, 0.3611111044883728, 0.3611111044883728, 0.3888888955116272, NaN, 0.4166666567325592, 0.4444444477558136, 0.3611111044883728, 0.5555555820465088, 0.5277777910232544, 0.5555555820465088, 0.5555555820465088, 0.6388888955116272, 0.6388888955116272, 0.8333333134651184], [0.0, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.1388888955116272, 0.1111111119389534, 0.0555555559694767, 0.1527777761220932, 0.0555555559694767, 0.02777777798473835, 0.1388888955116272, 0.125, 0.1666666716337204, 0.1111111119389534, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.1527777761220932, 0.0833333358168602, 0.02777777798473835, 0.1111111119389534, 0.1666666716337204, 0.1527777761220932, 0.1666666716337204, 0.1666666716337204, 0.1944444477558136, 0.1666666716337204, 0.1388888955116272, 0.1944444477558136, 0.1388888955116272, 0.1388888955116272, 0.1666666716337204, 0.2222222238779068, 0.2222222238779068, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1944444477558136, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.2222222238779068, 0.1666666716337204, 0.1388888955116272, 0.2222222238779068, 0.4166666567325592, 0.4166666567325592, 0.3055555522441864, 0.25, 0.3333333432674408, 0.25, 0.4722222089767456, 0.5, 0.3055555522441864, 0.3888888955116272, 0.3333333432674408, 0.4722222089767456, 0.3611111044883728, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.5833333134651184, NaN, 0.5277777910232544, 0.5277777910232544, 0.5833333134651184, 0.5555555820465088, 0.7222222089767456, 0.7222222089767456, 0.7777777910232544, 0.8055555820465088, 0.9166666865348816], [0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.0972222238779068, 0.1111111119389534, 0.1111111119389534, 0.1666666716337204, 0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.1666666716337204, 0.1388888955116272, 0.1666666716337204, 0.0833333358168602, 0.0555555559694767, 0.2083333283662796, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.1944444477558136, 0.1388888955116272, 0.1666666716337204, 0.1111111119389534, 0.1944444477558136, 0.1944444477558136, 0.25, 0.1944444477558136, 0.1388888955116272, 0.25, 0.1944444477558136, 0.25, 0.25, 0.25, 0.1666666716337204, 0.1388888955116272, 0.3055555522441864, 0.3333333432674408, 0.3611111044883728, 0.25, 0.1666666716337204, 0.3888888955116272, 0.3888888955116272, 0.1111111119389534, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3888888955116272, 0.4166666567325592, 0.2361111044883728, 0.3611111044883728, 0.3611111044883728, 0.4722222089767456, 0.4166666567325592, 0.4166666567325592, 0.3888888955116272, 0.4444444477558136, 0.5555555820465088, 0.4722222089767456, NaN, 0.5, 0.6388888955116272, 0.5277777910232544, 0.7222222089767456, 0.7222222089767456, 0.75, 0.8333333134651184, 0.9722222089767456], [0.0, 0.0, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.1111111119389534, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.1388888955116272, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1944444477558136, 0.1111111119389534, 0.1111111119389534, 0.1666666716337204, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1666666716337204, 0.1944444477558136, 0.1944444477558136, 0.1944444477558136, 0.1388888955116272, 0.1388888955116272, 0.0833333358168602, 0.1666666716337204, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.25, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.02777777798473835, 0.3055555522441864, 0.3888888955116272, 0.3888888955116272, 0.3333333432674408, 0.2222222238779068, 0.2777777910232544, 0.25, 0.3888888955116272, 0.4166666567325592, 0.3055555522441864, 0.3611111044883728, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.5, 0.4444444477558136, 0.4722222089767456, 0.6388888955116272, 0.4722222089767456, 0.5, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6944444179534912, 0.7222222089767456, 0.7777777910232544, 0.9166666865348816], [0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.1388888955116272, 0.1111111119389534, 0.1111111119389534, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1944444477558136, 0.1111111119389534, 0.1111111119389534, 0.1527777761220932, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1944444477558136, 0.0833333358168602, 0.2222222238779068, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.0833333358168602, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1666666716337204, 0.1111111119389534, 0.1666666716337204, 0.0555555559694767, 0.2222222238779068, 0.1388888955116272, 0.1666666716337204, 0.1944444477558136, 0.1111111119389534, 0.3055555522441864, 0.3055555522441864, 0.1944444477558136, 0.2777777910232544, 0.2777777910232544, 0.25, 0.3888888955116272, 0.4166666567325592, 0.1944444477558136, 0.3888888955116272, 0.3333333432674408, 0.25, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.3888888955116272, 0.4444444477558136, 0.4166666567325592, 0.3611111044883728, 0.4444444477558136, NaN, 0.5, 0.6388888955116272, 0.6388888955116272, 0.7222222089767456, 0.9722222089767456, 0.9722222089767456], [0.0, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1388888955116272, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.0555555559694767, 0.1944444477558136, 0.1666666716337204, 0.1666666716337204, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.1944444477558136, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.0833333358168602, 0.1666666716337204, 0.1111111119389534, 0.1388888955116272, 0.1666666716337204, 0.25, 0.1944444477558136, 0.1111111119389534, 0.1666666716337204, 0.1388888955116272, 0.2777777910232544, 0.3888888955116272, 0.3888888955116272, 0.3611111044883728, 0.25, 0.1944444477558136, 0.25, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.1388888955116272, 0.3333333432674408, 0.4166666567325592, 0.2361111044883728, 0.4444444477558136, 0.4444444477558136, 0.4166666567325592, 0.4722222089767456, 0.4444444477558136, 0.4722222089767456, 0.3333333432674408, 0.5, NaN, 0.6388888955116272, 0.6388888955116272, 0.6944444179534912, 0.6111111044883728, 0.8333333134651184], [0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.0833333358168602, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.0555555559694767, 0.2222222238779068, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.02777777798473835, 0.0833333358168602, 0.1388888955116272, 0.1111111119389534, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.1388888955116272, 0.0555555559694767, 0.1944444477558136, 0.1666666716337204, 0.25, 0.2777777910232544, 0.2777777910232544, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1944444477558136, 0.3055555522441864, 0.3055555522441864, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.2777777910232544, 0.25, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3611111044883728, 0.3611111044883728, NaN, 0.7083333134651184, 0.7222222089767456, 0.5277777910232544, 0.75], [0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0416666679084301, 0.02777777798473835, 0.0555555559694767, 0.1111111119389534, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.02777777798473835, 0.1944444477558136, 0.0555555559694767, 0.0555555559694767, 0.1111111119389534, 0.1111111119389534, 0.0833333358168602, 0.02777777798473835, 0.0555555559694767, 0.1388888955116272, 0.02777777798473835, 0.0833333358168602, 0.1388888955116272, 0.0833333358168602, 0.0555555559694767, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.0555555559694767, 0.1666666716337204, 0.1388888955116272, 0.25, 0.2777777910232544, 0.2777777910232544, 0.25, 0.2222222238779068, 0.1666666716337204, 0.1944444477558136, 0.2777777910232544, 0.3055555522441864, 0.25, 0.2222222238779068, 0.1944444477558136, 0.2777777910232544, 0.2777777910232544, 0.3333333432674408, 0.3333333432674408, 0.2777777910232544, 0.4444444477558136, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.3611111044883728, 0.3611111044883728, 0.2916666567325592, NaN, 0.6388888955116272, 0.5, 0.7222222089767456], [0.0, 0.0, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0, 0.0, 0.0555555559694767, 0.02777777798473835, 0.0, 0.0555555559694767, 0.1111111119389534, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.1111111119389534, 0.0555555559694767, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.1666666716337204, 0.02777777798473835, 0.1388888955116272, 0.0833333358168602, 0.02777777798473835, 0.1111111119389534, 0.1388888955116272, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.1111111119389534, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.1111111119389534, 0.1666666716337204, 0.25, 0.25, 0.1944444477558136, 0.1111111119389534, 0.1388888955116272, 0.1388888955116272, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1944444477558136, 0.1666666716337204, 0.1944444477558136, 0.2222222238779068, 0.2777777910232544, 0.2777777910232544, 0.1944444477558136, 0.3611111044883728, 0.2222222238779068, 0.25, 0.2777777910232544, 0.2777777910232544, 0.3055555522441864, 0.2777777910232544, 0.3611111044883728, NaN, 0.4722222089767456, 0.6666666865348816], [0.02777777798473835, 0.02777777798473835, 0.0, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.02777777798473835, 0.0, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0, 0.0555555559694767, 0.0, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.1388888955116272, 0.02777777798473835, 0.0555555559694767, 0.0833333358168602, 0.0555555559694767, 0.02777777798473835, 0.1111111119389534, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.0833333358168602, 0.0555555559694767, 0.0555555559694767, 0.0833333358168602, 0.0833333358168602, 0.1111111119389534, 0.02777777798473835, 0.1666666716337204, 0.1666666716337204, 0.0833333358168602, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.25, 0.1388888955116272, 0.1944444477558136, 0.2777777910232544, 0.2222222238779068, 0.1944444477558136, 0.2222222238779068, 0.3611111044883728, 0.1944444477558136, 0.1666666716337204, 0.2222222238779068, 0.02777777798473835, 0.3888888955116272, 0.4722222089767456, 0.5, 0.5277777910232544, NaN, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.02777777798473835, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02777777798473835, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02777777798473835, 0.02777777798473835, 0.0, 0.02777777798473835, 0.0, 0.02777777798473835, 0.0, 0.0, 0.02777777798473835, 0.0, 0.02777777798473835, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0555555559694767, 0.0, 0.02777777798473835, 0.0, 0.0, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.0, 0.02777777798473835, 0.02777777798473835, 0.0555555559694767, 0.0555555559694767, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.02777777798473835, 0.0833333358168602, 0.0555555559694767, 0.0833333358168602, 0.1388888955116272, 0.1388888955116272, 0.02777777798473835, 0.1111111119389534, 0.02777777798473835, 0.1111111119389534, 0.1111111119389534, 0.0833333358168602, 0.0555555559694767, 0.1111111119389534, 0.1666666716337204, 0.0833333358168602, 0.02777777798473835, 0.0833333358168602, 0.02777777798473835, 0.1666666716337204, 0.25, 0.2777777910232544, 0.3333333432674408, 0.1111111119389534, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9349507674},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8619901506},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8466947941},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8444092809},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8435302356},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8210267215},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8041490864},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8023909975},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7492967637},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7369901546},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7202883269},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7095639942},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6953234868},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6847749644},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.6840717281},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6775668087},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6684247567},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6490857942},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6402953575},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6339662451},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6297468352},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6230661039},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.6214838251},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6169127993},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6119901535},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5829817154},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5817510549},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5692686372},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5661040799},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.565752463},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5650492273},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5625879042},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.5601265839},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.553797469},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5458860759},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5453586511},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.543424753},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.5353375511},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5337552733},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5300632927},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5256680739},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.514767933},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5063291154},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5042194107},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4869901543},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4736286948},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4572784817},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4472573852},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4458509129},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4446202543},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4312587937},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.425632912},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4175457114},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.4066455693},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.4027777773},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.386075953},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3806258811},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3720112517},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3691983154},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3586497917},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3577707482},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3379043632},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3321026745},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3183895959},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.299753871},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2897327724},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2827004254},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2767229289},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2644163166},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2626582319},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2517580895},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2466596353},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2334739829},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2178270059},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.211849509},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1661392426},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1536568238},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.126230663},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1174402266},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0425457112}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_all/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_all/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..3ca8e7c6b65388415c10e93ac7611a63ec112fe3
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_all/datasets_small/winrate_matrix.csv
@@ -0,0 +1,81 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),TabPFN-2.6 (default),RealTabPFN-2.5 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabPFNv2 (tuned + ensembled),TabDPT (tuned + ensembled),TabDPT-Turbo (default),RealMLP (tuned),Mitra (default),LimiX (default),TabDPT (tuned),TabM (tuned + ensembled),iLTM (tuned + ensembled),LightGBM (tuned + ensembled),CatBoost (tuned),CatBoost (tuned + ensembled),TabPFNv2 (tuned),TabM (tuned),TabSwift (default),xRFM (tuned + ensembled),XGBoost (tuned + ensembled),TabPFNv2 (default),LightGBM (tuned),XGBoost (tuned),ModernNCA (tuned + ensembled),CatBoost (default),TabDPT (default),ModernNCA (tuned),ChimeraBoost (tuned + ensembled),iLTM (tuned),SAP-RPT-OSS (default),TabICL (default),BetaTabPFN (default),TabM (default),xRFM (tuned),EBM (tuned + ensembled),RealMLP (default),ChimeraBoost (tuned),TorchMLP (tuned + ensembled),EBM (tuned),ModernNCA (default),ExtraTrees (tuned + ensembled),ChimeraBoost (default),EBM (default),TorchMLP (tuned),FastaiMLP (tuned + ensembled),ExtraTrees (tuned),Nori-30M (default),Nori (default),RandomForest (tuned + ensembled),LightGBM (default),XGBoost (default),FastaiMLP (tuned),TabSTAR (tuned),TabSTAR (tuned + ensembled),RandomForest (tuned),PerpetualBooster (tuned + ensembled),OrionMSP (default),iLTM (default),PerpetualBooster (tuned),xRFM (default),ExtraTrees (default),TorchMLP (default),TabSTAR (default),TabFlex (default),RandomForest (default),FastaiMLP (default),KNN (tuned + ensembled),PerpetualBooster (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.8333333,0.7777778,0.8611111,0.8333333,0.8333333,0.8888889,0.8611111,0.8611111,0.9166667,0.9444444,0.9166667,0.9166667,0.8888889,1.0,0.9166667,0.9722222,0.9166667,0.8888889,0.9444444,0.9444444,0.9722222,0.9722222,0.9444444,0.8888889,0.9444444,0.9444444,0.9444444,0.9444444,0.9166667,0.8888889,0.9166667,0.9722222,0.9444444,0.9444444,0.9444444,0.9166667,0.9722222,0.9444444,0.9166667,0.9444444,0.9166667,0.8888889,0.9166667,0.9722222,0.9166667,0.9444444,0.9444444,0.9444444,0.9166667,0.9722222,0.9166667,0.9444444,0.9166667,0.9444444,0.9444444,0.9444444,0.9722222,0.9166667,0.9166667,0.8888889,0.9444444,1.0,0.9722222,0.9166667,1.0,0.9722222,0.9722222,0.9722222,0.9444444,1.0,0.9722222,1.0,0.9444444,1.0,0.9722222,0.9722222,1.0,0.9722222,1.0
+TabPFN-3 (default),0.16666667,,0.5277778,0.5138889,0.5277778,0.5277778,0.5833333,0.6388889,0.75,0.6388889,0.75,0.6944444,0.8055556,0.7777778,0.8333333,0.7222222,0.8333333,0.8333333,0.8611111,0.9166667,0.7777778,0.8333333,0.8888889,0.8888889,0.8611111,0.8888889,0.9166667,0.7777778,0.9444444,0.8888889,0.9166667,0.9166667,0.8888889,0.8611111,0.9166667,0.9166667,0.8055556,0.9444444,0.9166667,0.8333333,0.9444444,0.8055556,0.9166667,0.8888889,0.9444444,0.8611111,0.9444444,0.9722222,0.9444444,0.8888889,0.9444444,0.8888889,0.9722222,0.8611111,0.8611111,0.9444444,0.9444444,0.9722222,0.9444444,0.8611111,0.8611111,0.9722222,0.9722222,1.0,0.9722222,1.0,0.9722222,0.9722222,0.9444444,0.8611111,0.9444444,0.9722222,1.0,0.9722222,0.9722222,0.9722222,0.9722222,1.0,0.9722222,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.22222222,0.4722222,,0.4722222,0.5833333,0.5277778,0.5,0.6111111,0.8055556,0.7777778,0.6944444,0.7222222,0.6666667,0.8055556,0.7777778,0.7222222,0.7777778,0.8333333,0.8611111,0.8888889,0.8333333,0.8611111,0.8333333,0.8055556,0.8055556,0.9166667,0.9166667,0.7777778,0.9166667,0.9166667,0.8888889,0.9166667,0.8055556,0.9166667,0.9166667,0.9166667,0.8055556,0.8333333,0.8333333,0.8611111,0.8611111,0.8333333,0.8611111,0.9166667,0.9166667,0.8611111,0.9444444,0.9166667,0.9444444,0.8888889,0.8888889,0.8611111,0.9166667,0.7777778,0.8333333,0.9444444,0.9722222,0.9444444,0.9166667,0.8611111,0.8333333,0.9444444,0.9166667,1.0,0.9444444,0.9444444,0.9722222,0.9722222,0.9444444,0.8611111,0.9444444,0.9444444,0.9722222,0.9722222,0.9444444,0.9722222,0.9722222,0.9722222,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.1388889,0.4861111,0.5277778,,0.5277778,0.6388889,0.6111111,0.6944444,0.75,0.6666667,0.8055556,0.6388889,0.6666667,0.7777778,0.8333333,0.75,0.6944444,0.7777778,0.8333333,0.8888889,0.8888889,0.9166667,0.9444444,0.8333333,0.7777778,0.8611111,0.9166667,0.8611111,0.9166667,0.8888889,0.8611111,0.9444444,0.75,0.8888889,0.8888889,0.8611111,0.8611111,0.7777778,0.8611111,0.8611111,0.9166667,0.8611111,0.8611111,0.9166667,0.8888889,0.8611111,0.8611111,0.9166667,0.8888889,0.8888889,0.8888889,0.8888889,0.9166667,0.8055556,0.8333333,0.9166667,0.9722222,1.0,0.9166667,0.8333333,0.8333333,0.9166667,0.9444444,0.9722222,0.9166667,0.9722222,0.9444444,0.9722222,0.9444444,0.8611111,0.8888889,0.9722222,0.9722222,0.9444444,0.9444444,0.9444444,0.9722222,0.9444444,0.9444444,1.0
+TabPFN-2.6 (default),0.16666667,0.4722222,0.41666666,0.4722222,,0.6388889,0.5,0.5555556,0.7222222,0.6666667,0.7222222,0.6388889,0.7222222,0.8055556,0.8611111,0.7777778,0.6944444,0.7777778,0.8888889,0.8888889,0.8333333,0.8888889,0.8611111,0.8055556,0.8055556,0.8888889,0.8888889,0.75,0.9444444,0.8888889,0.9166667,0.9444444,0.75,0.8611111,0.8611111,0.9166667,0.8333333,0.8888889,0.8611111,0.8333333,0.8611111,0.8611111,0.8888889,0.8611111,0.9444444,0.9166667,0.9444444,0.9166667,0.9166667,0.8611111,0.9444444,0.9166667,0.9166667,0.8055556,0.8333333,0.9444444,0.9444444,0.9722222,0.8888889,0.8333333,0.8333333,0.9166667,0.9722222,0.9722222,0.9444444,1.0,0.9722222,0.9444444,0.9722222,0.8333333,0.9166667,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,1.0
+RealTabPFN-2.5 (default),0.16666667,0.4722222,0.4722222,0.3611111,0.3611111,,0.4722222,0.43055555,0.6666667,0.6388889,0.5277778,0.6388889,0.6944444,0.75,0.75,0.7222222,0.6944444,0.7777778,0.7222222,0.8611111,0.8055556,0.8333333,0.8333333,0.8055556,0.8333333,0.8333333,0.8611111,0.9166667,0.8888889,0.8611111,0.8055556,0.8888889,0.8055556,0.8055556,0.8611111,0.9027778,0.8055556,0.7222222,0.8055556,0.8611111,0.9166667,0.8611111,0.8333333,0.8611111,0.9166667,0.9166667,0.8888889,0.8611111,0.9166667,0.9166667,0.8888889,0.9722222,0.8611111,0.7777778,0.8055556,0.8888889,0.9444444,0.9444444,1.0,0.8611111,0.8611111,0.9444444,0.9444444,0.9722222,0.9722222,1.0,0.9166667,0.9722222,0.9722222,0.8888889,0.9722222,0.9444444,1.0,0.9166667,0.9444444,0.9722222,0.9722222,1.0,0.9722222,0.9722222
+TabICLv2 (default),0.11111111,0.41666666,0.5,0.3888889,0.5,0.5277778,,0.5833333,0.6666667,0.75,0.6388889,0.6944444,0.7222222,0.6666667,0.7222222,0.6944444,0.7222222,0.7222222,0.8055556,0.8333333,0.7777778,0.7777778,0.7222222,0.75,0.8055556,0.8055556,0.8055556,0.8055556,0.8333333,0.8055556,0.8611111,0.8333333,0.75,0.8055556,0.8611111,0.7777778,0.7777778,0.8055556,0.8611111,0.8055556,0.8611111,0.8055556,0.8055556,0.8333333,0.8888889,0.8055556,0.8888889,0.8333333,0.8611111,0.8888889,0.9166667,0.9444444,0.8888889,0.75,0.75,0.8611111,0.9166667,0.9166667,0.8888889,0.8611111,0.8611111,0.8611111,0.9166667,0.8888889,0.8611111,0.9444444,0.9444444,0.8888889,0.9444444,0.8888889,0.9444444,0.9166667,0.9722222,0.9444444,0.9722222,0.9722222,0.9722222,1.0,0.9722222,1.0
+RealTabPFN-2.5 (tuned),0.1388889,0.3611111,0.3888889,0.30555555,0.44444445,0.5694444,0.41666666,,0.6388889,0.6388889,0.5972222,0.5833333,0.6388889,0.75,0.8055556,0.6111111,0.6388889,0.8055556,0.7222222,0.7777778,0.7777778,0.8055556,0.8888889,0.8333333,0.6944444,0.8055556,0.8888889,0.7777778,0.8888889,0.8611111,0.8055556,0.9166667,0.7222222,0.8611111,0.8055556,0.8888889,0.8055556,0.8055556,0.8333333,0.8333333,0.8888889,0.8333333,0.8333333,0.8611111,0.8888889,0.8333333,0.8888889,0.9444444,0.8888889,0.8611111,0.9166667,0.8611111,0.9166667,0.7777778,0.8055556,0.9166667,0.9444444,0.9722222,0.8888889,0.8055556,0.7777778,0.8888889,0.9444444,0.9722222,0.9166667,0.9722222,0.9166667,0.9444444,0.9444444,0.8611111,0.8611111,0.9444444,0.9444444,0.9166667,0.9444444,0.9444444,0.9444444,0.9444444,0.9444444,1.0
+RealMLP (tuned + ensembled),0.1388889,0.25,0.19444445,0.25,0.2777778,0.33333334,0.33333334,0.3611111,,0.5555556,0.44444445,0.44444445,0.5555556,0.6388889,0.5555556,0.4722222,0.5,0.6666667,0.7777778,0.7777778,0.7777778,0.8611111,0.6388889,0.7222222,0.6388889,0.6944444,0.8611111,0.6111111,0.8333333,0.8611111,0.75,0.8611111,0.6666667,0.75,0.8055556,0.8333333,0.7222222,0.7222222,0.75,0.7777778,0.75,0.8333333,0.8888889,0.8333333,0.8333333,0.8333333,0.8888889,0.8333333,0.8611111,0.8611111,0.8055556,0.8888889,0.8611111,0.75,0.75,0.8055556,0.9166667,0.9444444,0.8611111,0.7777778,0.7777778,0.8888889,0.9444444,0.9444444,0.9444444,0.9722222,0.9166667,0.9722222,0.9166667,0.8333333,0.8888889,0.9722222,0.9722222,0.9444444,0.9722222,0.9722222,0.9722222,0.9722222,0.9722222,1.0
+"AutoGluon 1.4 (best, 4h)",0.083333336,0.3611111,0.22222222,0.33333334,0.33333334,0.3611111,0.25,0.3611111,0.44444445,,0.41666666,0.41666666,0.5277778,0.6111111,0.5555556,0.5555556,0.5,0.6111111,0.6944444,0.7777778,0.6666667,0.7222222,0.5555556,0.7222222,0.5833333,0.7222222,0.8055556,0.6666667,0.8333333,0.75,0.6666667,0.7777778,0.6944444,0.6666667,0.8055556,0.75,0.6944444,0.6944444,0.75,0.75,0.75,0.7222222,0.75,0.8333333,0.8888889,0.8611111,0.8888889,0.8055556,0.8333333,0.8611111,0.9166667,0.8888889,0.8055556,0.75,0.7777778,0.8611111,0.9166667,0.9166667,0.9166667,0.75,0.75,0.9444444,0.9444444,0.9166667,0.9166667,0.9722222,0.8888889,0.9722222,0.9722222,0.8333333,0.9444444,0.9722222,0.9722222,0.9166667,0.9722222,0.9722222,0.9722222,1.0,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.055555556,0.25,0.30555555,0.19444445,0.2777778,0.4722222,0.3611111,0.4027778,0.5555556,0.5833333,,0.5277778,0.5277778,0.5555556,0.6111111,0.5,0.5833333,0.6666667,0.6388889,0.6944444,0.6666667,0.6666667,0.7638889,0.6944444,0.6388889,0.7222222,0.8055556,0.7222222,0.7222222,0.75,0.75,0.6944444,0.5833333,0.6944444,0.7222222,0.6666667,0.6666667,0.6944444,0.7638889,0.7222222,0.7777778,0.8055556,0.8055556,0.7777778,0.7777778,0.8055556,0.75,0.75,0.8333333,0.8333333,0.8611111,0.7777778,0.8055556,0.7638889,0.7916667,0.8611111,0.9166667,0.8888889,0.8333333,0.75,0.7222222,0.8888889,0.8333333,0.8472222,0.8888889,0.8888889,0.8333333,0.8888889,0.8888889,0.8055556,0.8472222,0.9166667,0.9166667,0.9166667,0.9444444,0.9722222,0.9722222,0.9444444,0.9444444,1.0
+TabDPT (tuned + ensembled),0.083333336,0.30555555,0.2777778,0.3611111,0.3611111,0.3611111,0.30555555,0.41666666,0.5555556,0.5833333,0.4722222,,0.41666666,0.6666667,0.5555556,0.6111111,0.7222222,0.5555556,0.7222222,0.6388889,0.6111111,0.6666667,0.6388889,0.5555556,0.6666667,0.6666667,0.7222222,0.5833333,0.6944444,0.75,0.6944444,0.6944444,0.75,0.6666667,0.75,0.8055556,0.6944444,0.6666667,0.6388889,0.6944444,0.75,0.6944444,0.6944444,0.75,0.7222222,0.7222222,0.8055556,0.8055556,0.7222222,0.8055556,0.7222222,0.75,0.8888889,0.8055556,0.7777778,0.8055556,0.8055556,0.8333333,0.8055556,0.8055556,0.75,0.8333333,0.7777778,0.9722222,0.8611111,0.8611111,0.9444444,0.9722222,0.8611111,0.8333333,0.9444444,0.9444444,0.9722222,0.9166667,0.8888889,0.9166667,0.9166667,0.8888889,0.9444444,1.0
+TabDPT-Turbo (default),0.083333336,0.19444445,0.33333334,0.33333334,0.2777778,0.30555555,0.2777778,0.3611111,0.44444445,0.4722222,0.4722222,0.5833333,,0.6111111,0.5277778,0.5277778,0.5555556,0.5,0.6111111,0.6388889,0.5555556,0.6111111,0.5833333,0.5555556,0.6805556,0.6944444,0.7222222,0.5277778,0.6944444,0.75,0.5833333,0.6944444,0.6666667,0.6388889,0.7222222,0.75,0.6944444,0.6666667,0.6111111,0.7222222,0.75,0.6944444,0.6944444,0.7222222,0.75,0.7222222,0.7222222,0.8333333,0.75,0.8055556,0.8055556,0.8055556,0.8611111,0.6944444,0.6944444,0.8611111,0.8888889,0.8611111,0.8888889,0.7777778,0.75,0.8611111,0.8055556,0.9444444,0.8611111,0.8611111,0.8888889,0.9444444,0.8888889,0.8055556,0.9722222,0.9166667,0.9722222,0.9722222,0.9444444,0.9166667,0.9166667,0.9166667,0.9722222,1.0
+RealMLP (tuned),0.11111111,0.22222222,0.19444445,0.22222222,0.19444445,0.25,0.33333334,0.25,0.3611111,0.3888889,0.44444445,0.33333334,0.3888889,,0.4722222,0.4027778,0.44444445,0.5277778,0.5555556,0.6388889,0.6111111,0.6111111,0.5277778,0.6388889,0.5555556,0.5833333,0.7222222,0.6111111,0.7222222,0.75,0.6666667,0.7222222,0.5833333,0.7777778,0.6666667,0.7222222,0.6111111,0.6944444,0.7222222,0.7777778,0.6944444,0.8055556,0.7777778,0.7222222,0.7777778,0.7777778,0.75,0.7777778,0.75,0.8055556,0.8333333,0.8055556,0.8333333,0.75,0.75,0.8333333,0.8333333,0.8888889,0.8333333,0.7777778,0.8055556,0.8888889,0.8888889,0.8472222,0.9444444,0.8888889,0.9166667,0.8888889,0.9444444,0.8055556,0.8611111,0.9722222,0.9444444,0.9166667,0.9722222,0.9444444,0.9583333,0.9444444,0.9722222,1.0
+Mitra (default),0.0,0.16666667,0.22222222,0.16666667,0.1388889,0.25,0.2777778,0.19444445,0.44444445,0.44444445,0.3888889,0.44444445,0.4722222,0.5277778,,0.5277778,0.44444445,0.6944444,0.6666667,0.6666667,0.5833333,0.6111111,0.5833333,0.7222222,0.5833333,0.6666667,0.6666667,0.625,0.7777778,0.6944444,0.75,0.75,0.6111111,0.7222222,0.7222222,0.6666667,0.5833333,0.7083333,0.7083333,0.8333333,0.6944444,0.7777778,0.75,0.8055556,0.75,0.7777778,0.8333333,0.7222222,0.8055556,0.8055556,0.8888889,0.7777778,0.7777778,0.6944444,0.6944444,0.75,0.8333333,0.8333333,0.7777778,0.7777778,0.7777778,0.8333333,0.8333333,0.8611111,0.8611111,0.8333333,0.8888889,0.8055556,0.8611111,0.8333333,0.875,0.9027778,0.8888889,0.9166667,0.9444444,0.9722222,0.9722222,0.9722222,0.9722222,1.0
+LimiX (default),0.083333336,0.2777778,0.2777778,0.25,0.22222222,0.2777778,0.30555555,0.3888889,0.5277778,0.44444445,0.5,0.3888889,0.4722222,0.5972222,0.4722222,,0.5,0.5833333,0.625,0.6388889,0.5277778,0.5833333,0.6111111,0.6388889,0.5277778,0.6388889,0.6388889,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.5555556,0.6944444,0.6944444,0.7222222,0.6388889,0.6111111,0.6666667,0.6944444,0.75,0.75,0.75,0.75,0.7222222,0.8055556,0.7777778,0.7777778,0.7777778,0.8333333,0.6944444,0.7222222,0.75,0.75,0.7777778,0.7777778,0.8055556,0.8333333,0.8055556,0.8194444,0.8055556,0.8333333,0.8055556,0.8472222,0.8611111,0.8055556,0.8333333,0.8888889,0.8611111,0.8611111,0.8333333,0.8888889,0.8888889,0.9166667,0.8888889,0.9444444,0.9444444,0.9722222,0.9166667,1.0
+TabDPT (tuned),0.027777778,0.16666667,0.22222222,0.30555555,0.30555555,0.30555555,0.2777778,0.3611111,0.5,0.5,0.41666666,0.2777778,0.44444445,0.5555556,0.5555556,0.5,,0.5555556,0.6666667,0.6111111,0.5833333,0.6388889,0.6111111,0.5555556,0.6666667,0.6388889,0.6666667,0.5555556,0.6666667,0.6944444,0.6388889,0.6666667,0.7777778,0.6388889,0.7222222,0.75,0.6111111,0.6388889,0.5833333,0.6666667,0.6666667,0.6944444,0.6944444,0.75,0.6944444,0.6388889,0.8333333,0.75,0.7777778,0.7777778,0.7222222,0.6944444,0.8333333,0.6666667,0.6666667,0.7222222,0.7222222,0.7777778,0.8055556,0.7777778,0.75,0.8055556,0.7777778,0.9444444,0.8611111,0.8611111,0.8888889,0.9166667,0.8611111,0.8055556,0.8888889,0.8888889,0.9722222,0.8888889,0.9166667,0.8888889,0.8888889,0.8888889,0.9444444,0.9722222
+TabM (tuned + ensembled),0.083333336,0.16666667,0.16666667,0.22222222,0.22222222,0.22222222,0.2777778,0.19444445,0.33333334,0.3888889,0.33333334,0.44444445,0.5,0.4722222,0.30555555,0.41666666,0.44444445,,0.5277778,0.5555556,0.4722222,0.5277778,0.3888889,0.6111111,0.5555556,0.5833333,0.6388889,0.5,0.6111111,0.6944444,0.6388889,0.6666667,0.6111111,0.6388889,0.6388889,0.6388889,0.6388889,0.6944444,0.6944444,0.8333333,0.5833333,0.75,0.75,0.8055556,0.75,0.7222222,0.7222222,0.6944444,0.8055556,0.7777778,0.8055556,0.7777778,0.6944444,0.6111111,0.6111111,0.7777778,0.8333333,0.8333333,0.8333333,0.8055556,0.8055556,0.7777778,0.8055556,0.8611111,0.8611111,0.8888889,0.8333333,0.8055556,0.9722222,0.8333333,0.9166667,0.8333333,0.9166667,0.8888889,0.9444444,0.9444444,0.9444444,0.9444444,1.0,1.0
+iLTM (tuned + ensembled),0.11111111,0.1388889,0.1388889,0.16666667,0.11111111,0.2777778,0.19444445,0.2777778,0.22222222,0.30555555,0.3611111,0.2777778,0.3888889,0.44444445,0.33333334,0.375,0.33333334,0.4722222,,0.5555556,0.5555556,0.5,0.5,0.5277778,0.5,0.5555556,0.5833333,0.6111111,0.6388889,0.6666667,0.5277778,0.5555556,0.5277778,0.6666667,0.6944444,0.6944444,0.6388889,0.5138889,0.5277778,0.6111111,0.6944444,0.6388889,0.6666667,0.7222222,0.7222222,0.6944444,0.8055556,0.8333333,0.8055556,0.75,0.7777778,0.8333333,0.7777778,0.7222222,0.7222222,0.8888889,0.8611111,0.8611111,0.8333333,0.7222222,0.7222222,0.8888889,0.8888889,0.8333333,0.8611111,0.9444444,0.8888889,0.9444444,0.9444444,0.8055556,0.9166667,0.9444444,0.9444444,0.9444444,0.9444444,0.9166667,0.9444444,0.9444444,0.9444444,1.0
+LightGBM (tuned + ensembled),0.055555556,0.083333336,0.11111111,0.11111111,0.11111111,0.1388889,0.16666667,0.22222222,0.22222222,0.22222222,0.30555555,0.3611111,0.3611111,0.3611111,0.33333334,0.3611111,0.3888889,0.44444445,0.44444445,,0.5,0.5555556,0.3888889,0.4722222,0.5277778,0.5833333,0.5555556,0.5277778,0.75,0.6944444,0.5,0.7222222,0.5277778,0.5833333,0.6666667,0.6111111,0.5833333,0.6111111,0.5833333,0.6666667,0.6666667,0.7222222,0.6666667,0.75,0.7777778,0.75,0.7777778,0.7777778,0.8333333,0.8055556,0.7777778,0.8055556,0.7222222,0.75,0.75,0.8333333,0.9444444,0.9166667,0.8333333,0.7777778,0.75,0.9166667,0.8611111,0.8333333,0.8888889,0.8888889,0.8333333,0.8888889,0.8888889,0.8055556,0.8611111,0.9722222,0.9166667,0.9444444,0.8888889,0.9722222,0.9722222,0.9444444,1.0,1.0
+CatBoost (tuned),0.055555556,0.22222222,0.16666667,0.11111111,0.16666667,0.19444445,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.3888889,0.44444445,0.3888889,0.41666666,0.4722222,0.41666666,0.5277778,0.44444445,0.5,,0.44444445,0.44444445,0.5555556,0.4722222,0.5277778,0.6388889,0.5,0.5833333,0.6666667,0.6111111,0.6944444,0.5277778,0.6111111,0.5277778,0.6944444,0.6111111,0.5833333,0.6111111,0.6111111,0.6388889,0.6944444,0.6944444,0.6388889,0.75,0.7777778,0.7222222,0.7222222,0.75,0.7777778,0.7777778,0.7222222,0.6944444,0.6944444,0.6944444,0.8055556,0.8055556,0.8611111,0.8055556,0.75,0.75,0.8888889,0.8333333,0.8055556,0.8611111,0.9166667,0.8055556,0.8333333,0.9166667,0.7777778,0.8611111,0.9166667,0.9166667,0.8888889,0.9444444,0.9722222,0.9722222,0.9722222,0.9722222,1.0
+CatBoost (tuned + ensembled),0.027777778,0.16666667,0.1388889,0.083333336,0.11111111,0.16666667,0.22222222,0.19444445,0.1388889,0.2777778,0.33333334,0.33333334,0.3888889,0.3888889,0.3888889,0.41666666,0.3611111,0.4722222,0.5,0.44444445,0.5555556,,0.4722222,0.4722222,0.44444445,0.4722222,0.6111111,0.4722222,0.6111111,0.6388889,0.6111111,0.6666667,0.5,0.6111111,0.5555556,0.6944444,0.6666667,0.6388889,0.5555556,0.5555556,0.6111111,0.6666667,0.7222222,0.6388889,0.7222222,0.75,0.6944444,0.75,0.8611111,0.7777778,0.7777778,0.7222222,0.7777778,0.7222222,0.7222222,0.8611111,0.8055556,0.8611111,0.8333333,0.75,0.75,0.9166667,0.8888889,0.7777778,0.8611111,0.8888889,0.8611111,0.8611111,0.9166667,0.7777778,0.8333333,0.9444444,0.8888889,0.8611111,0.9444444,0.9722222,0.9722222,0.9722222,0.9444444,1.0
+TabPFNv2 (tuned),0.027777778,0.11111111,0.16666667,0.055555556,0.1388889,0.16666667,0.2777778,0.11111111,0.3611111,0.44444445,0.2361111,0.3611111,0.41666666,0.4722222,0.41666666,0.3888889,0.3888889,0.6111111,0.5,0.6111111,0.5555556,0.5277778,,0.6388889,0.5277778,0.6111111,0.6388889,0.5555556,0.6388889,0.6388889,0.6388889,0.5833333,0.5277778,0.6388889,0.6666667,0.6111111,0.5694444,0.5555556,0.625,0.6944444,0.6388889,0.7222222,0.6388889,0.6944444,0.75,0.6944444,0.6666667,0.6944444,0.75,0.7777778,0.8055556,0.6944444,0.6944444,0.6527778,0.6527778,0.6944444,0.7777778,0.8055556,0.75,0.7222222,0.7222222,0.8055556,0.75,0.7916667,0.8611111,0.8055556,0.8055556,0.8611111,0.8333333,0.7777778,0.8472222,0.8333333,0.8333333,0.8888889,0.8888889,0.9722222,0.9722222,0.9444444,0.9444444,0.9722222
+TabM (tuned),0.055555556,0.11111111,0.19444445,0.16666667,0.19444445,0.19444445,0.25,0.16666667,0.2777778,0.2777778,0.30555555,0.44444445,0.44444445,0.3611111,0.2777778,0.3611111,0.44444445,0.3888889,0.4722222,0.5277778,0.44444445,0.5277778,0.3611111,,0.5277778,0.5277778,0.5833333,0.44444445,0.5555556,0.6388889,0.6111111,0.6666667,0.6111111,0.6111111,0.7361111,0.6388889,0.5555556,0.6944444,0.6388889,0.7222222,0.5555556,0.6666667,0.6388889,0.7222222,0.6944444,0.6666667,0.6944444,0.6944444,0.75,0.6944444,0.8055556,0.6944444,0.6944444,0.6388889,0.6388889,0.75,0.8333333,0.8333333,0.8333333,0.75,0.75,0.7777778,0.8611111,0.8333333,0.8055556,0.8888889,0.8055556,0.7777778,0.9722222,0.8055556,0.9166667,0.8611111,0.8611111,0.8888889,0.9444444,0.9166667,0.9166667,0.9444444,0.9722222,0.9722222
+TabSwift (default),0.11111111,0.1388889,0.19444445,0.22222222,0.19444445,0.16666667,0.19444445,0.30555555,0.3611111,0.41666666,0.3611111,0.33333334,0.31944445,0.44444445,0.41666666,0.4722222,0.33333334,0.44444445,0.5,0.4722222,0.5277778,0.5555556,0.4722222,0.4722222,,0.5833333,0.5555556,0.5,0.5277778,0.5833333,0.5277778,0.5555556,0.5277778,0.6388889,0.5,0.6111111,0.6944444,0.5,0.5833333,0.5833333,0.6666667,0.6666667,0.6666667,0.5833333,0.5833333,0.6388889,0.6944444,0.6944444,0.7777778,0.7222222,0.6944444,0.8333333,0.7222222,0.6388889,0.6111111,0.7222222,0.7777778,0.7222222,0.8055556,0.75,0.75,0.75,0.7777778,0.8888889,0.8055556,0.8333333,0.8333333,0.8611111,0.8611111,0.8333333,0.9722222,0.8333333,0.9444444,0.9166667,0.8888889,0.9166667,0.9166667,0.9166667,0.9722222,1.0
+xRFM (tuned + ensembled),0.055555556,0.11111111,0.083333336,0.1388889,0.11111111,0.16666667,0.19444445,0.19444445,0.30555555,0.2777778,0.2777778,0.33333334,0.30555555,0.41666666,0.33333334,0.3611111,0.3611111,0.41666666,0.44444445,0.41666666,0.4722222,0.5277778,0.3888889,0.4722222,0.41666666,,0.5,0.4722222,0.4722222,0.4722222,0.5555556,0.5,0.44444445,0.5,0.5277778,0.5277778,0.5277778,0.5833333,0.5277778,0.6111111,0.6944444,0.5833333,0.6111111,0.6388889,0.6111111,0.5555556,0.75,0.7222222,0.6111111,0.6388889,0.6666667,0.6944444,0.7777778,0.6944444,0.6944444,0.8055556,0.7777778,0.7777778,0.75,0.6944444,0.6944444,0.8333333,0.75,0.8611111,0.7777778,0.8055556,0.8888889,0.9166667,0.8888889,0.7777778,0.8888889,0.9166667,0.9166667,0.9166667,0.8611111,0.9444444,0.9722222,0.9444444,0.9722222,0.9722222
+XGBoost (tuned + ensembled),0.055555556,0.083333336,0.083333336,0.083333336,0.11111111,0.1388889,0.19444445,0.11111111,0.1388889,0.19444445,0.19444445,0.2777778,0.2777778,0.2777778,0.33333334,0.3611111,0.33333334,0.3611111,0.41666666,0.44444445,0.3611111,0.3888889,0.3611111,0.41666666,0.44444445,0.5,,0.5,0.5555556,0.5833333,0.4722222,0.5277778,0.4722222,0.5277778,0.5555556,0.5833333,0.4722222,0.5833333,0.5,0.5833333,0.6111111,0.6388889,0.6666667,0.6388889,0.6666667,0.6944444,0.6666667,0.75,0.75,0.6944444,0.6805556,0.75,0.75,0.7222222,0.7222222,0.8333333,0.9722222,0.8888889,0.7777778,0.7777778,0.7777778,0.8888889,0.7777778,0.8333333,0.8611111,0.8333333,0.7777778,0.8888889,0.8888889,0.8055556,0.8333333,0.9444444,0.8888889,0.8611111,0.8611111,0.9166667,0.9444444,0.9444444,0.9166667,1.0
+TabPFNv2 (default),0.055555556,0.22222222,0.22222222,0.1388889,0.25,0.083333336,0.19444445,0.22222222,0.3888889,0.33333334,0.2777778,0.41666666,0.4722222,0.3888889,0.375,0.44444445,0.44444445,0.5,0.3888889,0.4722222,0.5,0.5277778,0.44444445,0.5555556,0.5,0.5277778,0.5,,0.5555556,0.5,0.5,0.5277778,0.5277778,0.4722222,0.5277778,0.41666666,0.5,0.5416667,0.5694444,0.5555556,0.5833333,0.6111111,0.5277778,0.5277778,0.5833333,0.5833333,0.5833333,0.6388889,0.5833333,0.6388889,0.6666667,0.6388889,0.6666667,0.6111111,0.6388889,0.6944444,0.6666667,0.6388889,0.6944444,0.6944444,0.7222222,0.75,0.75,0.7638889,0.7777778,0.75,0.8055556,0.75,0.7777778,0.75,0.8472222,0.7916667,0.8055556,0.8055556,0.8611111,0.9444444,0.9444444,0.9722222,0.9166667,0.9722222
+LightGBM (tuned),0.055555556,0.055555556,0.083333336,0.083333336,0.055555556,0.11111111,0.16666667,0.11111111,0.16666667,0.16666667,0.2777778,0.30555555,0.30555555,0.2777778,0.22222222,0.33333334,0.33333334,0.3888889,0.3611111,0.25,0.41666666,0.3888889,0.3611111,0.44444445,0.4722222,0.5277778,0.44444445,0.44444445,,0.5,0.33333334,0.5555556,0.4722222,0.5277778,0.5277778,0.5277778,0.5277778,0.5833333,0.6111111,0.6666667,0.5833333,0.6944444,0.6111111,0.6111111,0.6666667,0.6666667,0.5833333,0.6666667,0.6944444,0.7222222,0.75,0.75,0.6944444,0.6944444,0.6944444,0.75,0.8055556,0.8333333,0.8055556,0.7222222,0.75,0.8333333,0.8333333,0.8055556,0.8055556,0.8611111,0.8333333,0.7777778,0.8611111,0.8055556,0.8333333,0.9166667,0.8888889,0.8888889,0.8888889,0.9166667,0.9166667,0.9166667,0.9444444,1.0
+XGBoost (tuned),0.083333336,0.11111111,0.083333336,0.11111111,0.11111111,0.1388889,0.19444445,0.1388889,0.1388889,0.25,0.25,0.25,0.25,0.25,0.30555555,0.33333334,0.30555555,0.30555555,0.33333334,0.30555555,0.33333334,0.3611111,0.3611111,0.3611111,0.41666666,0.5277778,0.41666666,0.5,0.5,,0.4722222,0.5277778,0.44444445,0.5555556,0.5555556,0.5555556,0.5,0.5555556,0.5,0.5277778,0.5833333,0.6111111,0.6111111,0.5833333,0.6111111,0.6944444,0.6388889,0.7777778,0.6944444,0.6944444,0.6666667,0.75,0.75,0.6944444,0.6944444,0.8055556,0.9166667,0.8611111,0.8055556,0.7777778,0.7777778,0.8611111,0.7777778,0.8055556,0.8333333,0.8333333,0.7777778,0.8611111,0.8611111,0.8055556,0.8333333,0.9166667,0.8888889,0.8888889,0.8611111,0.8888889,0.9166667,0.9444444,0.9166667,1.0
+ModernNCA (tuned + ensembled),0.11111111,0.083333336,0.11111111,0.1388889,0.083333336,0.19444445,0.1388889,0.19444445,0.25,0.33333334,0.25,0.30555555,0.41666666,0.33333334,0.25,0.33333334,0.3611111,0.3611111,0.4722222,0.5,0.3888889,0.3888889,0.3611111,0.3888889,0.4722222,0.44444445,0.5277778,0.5,0.6666667,0.5277778,,0.5277778,0.5,0.625,0.5555556,0.5277778,0.5,0.5833333,0.5833333,0.5,0.5833333,0.5833333,0.6388889,0.5833333,0.5555556,0.5833333,0.7222222,0.6944444,0.5833333,0.6388889,0.7222222,0.7222222,0.6944444,0.6666667,0.6666667,0.75,0.7777778,0.8055556,0.8055556,0.6666667,0.6944444,0.7222222,0.7777778,0.8055556,0.7777778,0.7777778,0.8611111,0.8055556,0.8055556,0.8055556,0.8055556,0.8611111,0.8333333,0.8472222,0.8611111,0.8333333,0.8611111,0.8333333,0.8611111,0.9722222
+CatBoost (default),0.083333336,0.083333336,0.083333336,0.055555556,0.055555556,0.11111111,0.16666667,0.083333336,0.1388889,0.22222222,0.30555555,0.30555555,0.30555555,0.2777778,0.25,0.33333334,0.33333334,0.33333334,0.44444445,0.2777778,0.30555555,0.33333334,0.41666666,0.33333334,0.44444445,0.5,0.4722222,0.4722222,0.44444445,0.4722222,0.4722222,,0.5277778,0.44444445,0.5,0.5833333,0.5277778,0.5,0.41666666,0.5277778,0.5833333,0.5833333,0.6944444,0.6388889,0.6388889,0.6388889,0.5555556,0.6666667,0.8055556,0.6944444,0.6944444,0.75,0.6666667,0.6944444,0.6944444,0.7777778,0.75,0.7777778,0.7777778,0.7777778,0.75,0.8333333,0.8611111,0.7777778,0.8333333,0.9166667,0.8333333,0.8055556,0.8888889,0.8055556,0.8333333,0.8888889,0.9166667,0.8888889,0.9444444,0.9444444,0.9722222,0.9722222,0.9722222,1.0
+TabDPT (default),0.027777778,0.11111111,0.19444445,0.25,0.25,0.19444445,0.25,0.2777778,0.33333334,0.30555555,0.41666666,0.25,0.33333334,0.41666666,0.3888889,0.44444445,0.22222222,0.3888889,0.4722222,0.4722222,0.4722222,0.5,0.4722222,0.3888889,0.4722222,0.5555556,0.5277778,0.4722222,0.5277778,0.5555556,0.5,0.4722222,,0.5277778,0.5,0.5555556,0.5,0.4722222,0.44444445,0.5277778,0.6111111,0.44444445,0.5,0.5277778,0.5277778,0.5,0.6944444,0.6388889,0.6388889,0.6111111,0.5555556,0.6666667,0.7222222,0.6111111,0.6111111,0.6944444,0.7222222,0.6944444,0.8055556,0.6666667,0.6944444,0.7222222,0.75,0.8333333,0.75,0.8055556,0.8055556,0.8888889,0.7777778,0.75,0.8611111,0.8055556,0.8611111,0.8888889,0.8055556,0.7777778,0.8055556,0.8611111,0.9444444,0.9722222
+ModernNCA (tuned),0.055555556,0.1388889,0.083333336,0.11111111,0.1388889,0.19444445,0.19444445,0.1388889,0.25,0.33333334,0.30555555,0.33333334,0.3611111,0.22222222,0.2777778,0.30555555,0.3611111,0.3611111,0.33333334,0.41666666,0.3888889,0.3888889,0.3611111,0.3888889,0.3611111,0.5,0.4722222,0.5277778,0.4722222,0.44444445,0.375,0.5555556,0.4722222,,0.5277778,0.5277778,0.5833333,0.4722222,0.5277778,0.5555556,0.5555556,0.5277778,0.6111111,0.5833333,0.5833333,0.6388889,0.6944444,0.5833333,0.6388889,0.6388889,0.7222222,0.6527778,0.6111111,0.6944444,0.6944444,0.6944444,0.75,0.8055556,0.75,0.7222222,0.7222222,0.7222222,0.7777778,0.7777778,0.7777778,0.8055556,0.8055556,0.7777778,0.8333333,0.8333333,0.8055556,0.8611111,0.8611111,0.8888889,0.8333333,0.9166667,0.9444444,0.9166667,0.9166667,1.0
+ChimeraBoost (tuned + ensembled),0.055555556,0.083333336,0.083333336,0.11111111,0.1388889,0.1388889,0.1388889,0.19444445,0.19444445,0.19444445,0.2777778,0.25,0.2777778,0.33333334,0.2777778,0.30555555,0.2777778,0.3611111,0.30555555,0.33333334,0.4722222,0.44444445,0.33333334,0.2638889,0.5,0.4722222,0.44444445,0.4722222,0.4722222,0.44444445,0.44444445,0.5,0.5,0.4722222,,0.5277778,0.5,0.5555556,0.41666666,0.5555556,0.5833333,0.6388889,0.6388889,0.6111111,0.5833333,0.6666667,0.5555556,0.5833333,0.8333333,0.6388889,0.6666667,0.7222222,0.6111111,0.6111111,0.6111111,0.7222222,0.6944444,0.7222222,0.6944444,0.7777778,0.75,0.7777778,0.8055556,0.7777778,0.75,0.8611111,0.7777778,0.7222222,0.8333333,0.8055556,0.8611111,0.8333333,0.8611111,0.8055556,0.8333333,0.9166667,0.9444444,0.9722222,0.9444444,1.0
+iLTM (tuned),0.055555556,0.083333336,0.083333336,0.1388889,0.083333336,0.097222224,0.22222222,0.11111111,0.16666667,0.25,0.33333334,0.19444445,0.25,0.2777778,0.33333334,0.2777778,0.25,0.3611111,0.30555555,0.3888889,0.30555555,0.30555555,0.3888889,0.3611111,0.3888889,0.4722222,0.41666666,0.5833333,0.4722222,0.44444445,0.4722222,0.41666666,0.44444445,0.4722222,0.4722222,,0.5555556,0.4722222,0.5277778,0.5277778,0.5555556,0.5277778,0.5833333,0.6388889,0.5555556,0.6388889,0.6388889,0.7222222,0.5555556,0.7777778,0.6111111,0.75,0.7222222,0.6666667,0.6666667,0.8055556,0.7222222,0.6944444,0.7777778,0.6944444,0.6944444,0.7777778,0.7222222,0.75,0.7638889,0.8333333,0.8611111,0.8333333,0.9166667,0.8055556,0.8611111,0.8888889,0.8888889,0.9166667,0.8888889,0.8611111,0.8888889,0.8888889,0.9722222,1.0
+SAP-RPT-OSS (default),0.083333336,0.19444445,0.19444445,0.1388889,0.16666667,0.19444445,0.22222222,0.19444445,0.2777778,0.30555555,0.33333334,0.30555555,0.30555555,0.3888889,0.41666666,0.3611111,0.3888889,0.3611111,0.3611111,0.41666666,0.3888889,0.33333334,0.43055555,0.44444445,0.30555555,0.4722222,0.5277778,0.5,0.4722222,0.5,0.5,0.4722222,0.5,0.41666666,0.5,0.44444445,,0.5,0.5277778,0.5,0.5,0.4722222,0.5833333,0.5277778,0.6388889,0.5833333,0.6111111,0.6111111,0.5833333,0.6388889,0.6666667,0.7222222,0.6666667,0.6111111,0.6111111,0.6944444,0.6944444,0.6666667,0.7222222,0.6388889,0.6388889,0.7222222,0.75,0.6944444,0.75,0.75,0.75,0.7777778,0.8055556,0.8333333,0.8333333,0.8055556,0.8333333,0.7777778,0.8611111,0.9166667,0.8888889,0.8611111,0.8888889,1.0
+TabICL (default),0.027777778,0.055555556,0.16666667,0.22222222,0.11111111,0.2777778,0.19444445,0.19444445,0.2777778,0.30555555,0.30555555,0.33333334,0.33333334,0.30555555,0.29166666,0.3888889,0.3611111,0.30555555,0.4861111,0.3888889,0.41666666,0.3611111,0.44444445,0.30555555,0.5,0.41666666,0.41666666,0.45833334,0.41666666,0.44444445,0.41666666,0.5,0.5277778,0.5277778,0.44444445,0.5277778,0.5,,0.5,0.44444445,0.4722222,0.5277778,0.5277778,0.5277778,0.5833333,0.5555556,0.5694444,0.5833333,0.5833333,0.6388889,0.6388889,0.5833333,0.6111111,0.6944444,0.6944444,0.5833333,0.6666667,0.6944444,0.6666667,0.6944444,0.6944444,0.7361111,0.7222222,0.75,0.75,0.7777778,0.6944444,0.7222222,0.7777778,0.75,0.7777778,0.8055556,0.8055556,0.8611111,0.8611111,0.9444444,0.9166667,0.9722222,0.9722222,1.0
+BetaTabPFN (default),0.055555556,0.083333336,0.16666667,0.1388889,0.1388889,0.19444445,0.1388889,0.16666667,0.25,0.25,0.2361111,0.3611111,0.3888889,0.2777778,0.29166666,0.33333334,0.41666666,0.30555555,0.4722222,0.41666666,0.3888889,0.44444445,0.375,0.3611111,0.41666666,0.4722222,0.5,0.43055555,0.3888889,0.5,0.41666666,0.5833333,0.5555556,0.4722222,0.5833333,0.4722222,0.4722222,0.5,,0.4722222,0.4722222,0.6111111,0.5555556,0.5555556,0.5277778,0.7222222,0.5555556,0.5555556,0.6111111,0.6388889,0.6944444,0.5277778,0.5555556,0.6388889,0.6388889,0.6111111,0.6388889,0.6944444,0.6111111,0.7222222,0.7222222,0.6666667,0.6944444,0.6944444,0.8333333,0.7222222,0.6944444,0.6388889,0.7777778,0.8055556,0.7777778,0.75,0.8055556,0.8611111,0.8055556,0.9722222,0.9722222,0.9722222,0.9722222,1.0
+TabM (default),0.083333336,0.16666667,0.1388889,0.1388889,0.16666667,0.1388889,0.19444445,0.16666667,0.22222222,0.25,0.2777778,0.30555555,0.2777778,0.22222222,0.16666667,0.30555555,0.33333334,0.16666667,0.3888889,0.33333334,0.3888889,0.44444445,0.30555555,0.2777778,0.41666666,0.3888889,0.41666666,0.44444445,0.33333334,0.4722222,0.5,0.4722222,0.4722222,0.44444445,0.44444445,0.4722222,0.5,0.5555556,0.5277778,,0.5277778,0.5833333,0.5277778,0.5555556,0.5555556,0.5833333,0.6111111,0.6111111,0.6111111,0.5416667,0.7222222,0.6944444,0.6388889,0.6111111,0.6111111,0.6944444,0.6944444,0.7777778,0.8055556,0.6666667,0.6944444,0.7222222,0.8055556,0.75,0.7222222,0.8055556,0.7222222,0.7222222,0.9166667,0.8055556,0.8333333,0.8055556,0.8055556,0.8055556,0.8888889,0.9444444,0.9444444,0.9166667,0.9722222,0.9444444
+xRFM (tuned),0.055555556,0.055555556,0.1388889,0.083333336,0.1388889,0.083333336,0.1388889,0.11111111,0.25,0.25,0.22222222,0.25,0.25,0.30555555,0.30555555,0.25,0.33333334,0.41666666,0.30555555,0.33333334,0.3611111,0.3888889,0.3611111,0.44444445,0.33333334,0.30555555,0.3888889,0.41666666,0.41666666,0.41666666,0.41666666,0.41666666,0.3888889,0.44444445,0.41666666,0.44444445,0.5,0.5277778,0.5277778,0.4722222,,0.4722222,0.5277778,0.5,0.5277778,0.5555556,0.6111111,0.7222222,0.5833333,0.5555556,0.6111111,0.6666667,0.75,0.6388889,0.6388889,0.75,0.6944444,0.6666667,0.6944444,0.6666667,0.6944444,0.7777778,0.6944444,0.8055556,0.8055556,0.7777778,0.9444444,0.8888889,0.8888889,0.75,0.8333333,0.8611111,0.8611111,0.9166667,0.9166667,0.8888889,0.8611111,0.8888889,0.9722222,1.0
+EBM (tuned + ensembled),0.083333336,0.19444445,0.16666667,0.1388889,0.1388889,0.1388889,0.19444445,0.16666667,0.16666667,0.2777778,0.19444445,0.30555555,0.30555555,0.19444445,0.22222222,0.25,0.30555555,0.25,0.3611111,0.2777778,0.30555555,0.33333334,0.2777778,0.33333334,0.33333334,0.41666666,0.3611111,0.3888889,0.30555555,0.3888889,0.41666666,0.41666666,0.5555556,0.4722222,0.3611111,0.4722222,0.5277778,0.4722222,0.3888889,0.41666666,0.5277778,,0.5555556,0.44444445,0.5555556,0.7222222,0.6666667,0.6666667,0.5,0.7777778,0.6111111,0.6944444,0.6111111,0.6111111,0.6111111,0.6388889,0.6111111,0.6944444,0.75,0.7222222,0.75,0.6388889,0.6666667,0.6944444,0.75,0.75,0.7777778,0.7222222,0.8333333,0.8055556,0.8333333,0.75,0.8611111,0.8888889,0.8888889,0.9722222,0.9722222,0.9722222,0.9166667,0.9722222
+RealMLP (default),0.11111111,0.083333336,0.1388889,0.1388889,0.11111111,0.16666667,0.19444445,0.16666667,0.11111111,0.25,0.19444445,0.30555555,0.30555555,0.22222222,0.25,0.25,0.30555555,0.25,0.33333334,0.33333334,0.30555555,0.2777778,0.3611111,0.3611111,0.33333334,0.3888889,0.33333334,0.4722222,0.3888889,0.3888889,0.3611111,0.30555555,0.5,0.3888889,0.3611111,0.41666666,0.41666666,0.4722222,0.44444445,0.4722222,0.4722222,0.44444445,,0.3611111,0.5833333,0.5555556,0.6111111,0.6111111,0.5277778,0.5555556,0.6111111,0.6388889,0.6666667,0.6111111,0.6111111,0.6944444,0.6388889,0.75,0.6388889,0.6666667,0.7222222,0.6944444,0.6666667,0.7222222,0.7777778,0.7222222,0.7777778,0.7777778,0.8888889,0.8333333,0.9166667,0.8055556,0.9166667,0.8888889,0.9166667,0.9166667,0.9166667,0.9722222,0.9444444,1.0
+ChimeraBoost (tuned),0.083333336,0.11111111,0.083333336,0.083333336,0.1388889,0.1388889,0.16666667,0.1388889,0.16666667,0.16666667,0.22222222,0.25,0.2777778,0.2777778,0.19444445,0.25,0.25,0.19444445,0.2777778,0.25,0.3611111,0.3611111,0.30555555,0.2777778,0.41666666,0.3611111,0.3611111,0.4722222,0.3888889,0.41666666,0.41666666,0.3611111,0.4722222,0.41666666,0.3888889,0.3611111,0.4722222,0.4722222,0.44444445,0.44444445,0.5,0.5555556,0.6388889,,0.5277778,0.5833333,0.5,0.5277778,0.6944444,0.6666667,0.5555556,0.6666667,0.6111111,0.5555556,0.5555556,0.6666667,0.7222222,0.75,0.7222222,0.7222222,0.7222222,0.7777778,0.7777778,0.75,0.7777778,0.8055556,0.7777778,0.75,0.8333333,0.8055556,0.8611111,0.75,0.8333333,0.7777778,0.8333333,0.8611111,0.8611111,0.9444444,0.9166667,1.0
+TorchMLP (tuned + ensembled),0.027777778,0.055555556,0.083333336,0.11111111,0.055555556,0.083333336,0.11111111,0.11111111,0.16666667,0.11111111,0.22222222,0.2777778,0.25,0.22222222,0.25,0.2777778,0.30555555,0.25,0.2777778,0.22222222,0.25,0.2777778,0.25,0.30555555,0.41666666,0.3888889,0.33333334,0.41666666,0.33333334,0.3888889,0.44444445,0.3611111,0.4722222,0.41666666,0.41666666,0.44444445,0.3611111,0.41666666,0.4722222,0.44444445,0.4722222,0.44444445,0.41666666,0.4722222,,0.4722222,0.5277778,0.5833333,0.5,0.5277778,0.7222222,0.6666667,0.6111111,0.5555556,0.5555556,0.6666667,0.75,0.75,0.7222222,0.6388889,0.6666667,0.7222222,0.6944444,0.7222222,0.75,0.7777778,0.75,0.7222222,0.9444444,0.7222222,0.8888889,0.75,0.8611111,0.8333333,0.8888889,0.8888889,0.9166667,0.9722222,0.9166667,0.9444444
+EBM (tuned),0.083333336,0.1388889,0.1388889,0.1388889,0.083333336,0.083333336,0.19444445,0.16666667,0.16666667,0.1388889,0.19444445,0.2777778,0.2777778,0.22222222,0.22222222,0.19444445,0.3611111,0.2777778,0.30555555,0.25,0.22222222,0.25,0.30555555,0.33333334,0.3611111,0.44444445,0.30555555,0.41666666,0.33333334,0.30555555,0.41666666,0.3611111,0.5,0.3611111,0.33333334,0.3611111,0.41666666,0.44444445,0.2777778,0.41666666,0.44444445,0.2777778,0.44444445,0.41666666,0.5277778,,0.6111111,0.5555556,0.44444445,0.5833333,0.5555556,0.5555556,0.5833333,0.6111111,0.6111111,0.6111111,0.5555556,0.5833333,0.6666667,0.6944444,0.6944444,0.6111111,0.6666667,0.6666667,0.75,0.6666667,0.8055556,0.7222222,0.8055556,0.8055556,0.8055556,0.75,0.8333333,0.8888889,0.8611111,0.9166667,0.9444444,0.9166667,0.9166667,0.9722222
+ModernNCA (default),0.055555556,0.055555556,0.055555556,0.1388889,0.055555556,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.25,0.19444445,0.2777778,0.25,0.16666667,0.22222222,0.16666667,0.2777778,0.19444445,0.22222222,0.2777778,0.30555555,0.33333334,0.30555555,0.30555555,0.25,0.33333334,0.41666666,0.41666666,0.3611111,0.2777778,0.44444445,0.30555555,0.30555555,0.44444445,0.3611111,0.3888889,0.43055555,0.44444445,0.3888889,0.3888889,0.33333334,0.3888889,0.5,0.4722222,0.3888889,,0.5833333,0.5,0.4722222,0.4722222,0.5277778,0.5277778,0.6111111,0.6111111,0.6944444,0.5833333,0.5277778,0.6111111,0.75,0.7222222,0.75,0.6666667,0.7222222,0.7222222,0.7222222,0.6666667,0.7777778,0.8055556,0.75,0.7777778,0.8333333,0.8055556,0.8333333,0.8333333,0.8888889,0.8888889,0.8888889,0.9166667,0.9722222
+ExtraTrees (tuned + ensembled),0.055555556,0.027777778,0.083333336,0.083333336,0.083333336,0.1388889,0.16666667,0.055555556,0.16666667,0.19444445,0.25,0.19444445,0.16666667,0.22222222,0.2777778,0.22222222,0.25,0.30555555,0.16666667,0.22222222,0.2777778,0.25,0.30555555,0.30555555,0.30555555,0.2777778,0.25,0.3611111,0.33333334,0.22222222,0.30555555,0.33333334,0.3611111,0.41666666,0.41666666,0.2777778,0.3888889,0.41666666,0.44444445,0.3888889,0.2777778,0.33333334,0.3888889,0.4722222,0.41666666,0.44444445,0.41666666,,0.5277778,0.41666666,0.4861111,0.5,0.6805556,0.6388889,0.6388889,0.7222222,0.5277778,0.5,0.5,0.6388889,0.6666667,0.8055556,0.6388889,0.7222222,0.6388889,0.7222222,0.75,0.9166667,0.6388889,0.75,0.8055556,0.8611111,0.75,0.9444444,0.75,0.8333333,0.8611111,0.8333333,0.9444444,1.0
+ChimeraBoost (default),0.055555556,0.055555556,0.055555556,0.11111111,0.083333336,0.083333336,0.1388889,0.11111111,0.1388889,0.16666667,0.16666667,0.2777778,0.25,0.25,0.19444445,0.22222222,0.22222222,0.19444445,0.19444445,0.16666667,0.25,0.1388889,0.25,0.25,0.22222222,0.3888889,0.25,0.41666666,0.30555555,0.30555555,0.41666666,0.19444445,0.3611111,0.3611111,0.16666667,0.44444445,0.41666666,0.41666666,0.3888889,0.3888889,0.41666666,0.5,0.4722222,0.30555555,0.5,0.5555556,0.5,0.4722222,,0.5277778,0.44444445,0.6111111,0.5277778,0.5,0.5,0.6388889,0.5833333,0.6111111,0.6111111,0.6944444,0.75,0.7222222,0.7222222,0.6944444,0.75,0.75,0.6944444,0.6944444,0.7222222,0.8055556,0.8055556,0.6944444,0.7777778,0.7777778,0.8055556,0.8611111,0.8611111,0.9722222,0.9444444,0.9722222
+EBM (default),0.083333336,0.11111111,0.11111111,0.11111111,0.1388889,0.083333336,0.11111111,0.1388889,0.1388889,0.1388889,0.16666667,0.19444445,0.19444445,0.19444445,0.19444445,0.16666667,0.22222222,0.22222222,0.25,0.19444445,0.22222222,0.22222222,0.22222222,0.30555555,0.2777778,0.3611111,0.30555555,0.3611111,0.2777778,0.30555555,0.3611111,0.30555555,0.3888889,0.3611111,0.3611111,0.22222222,0.3611111,0.3611111,0.3611111,0.45833334,0.44444445,0.22222222,0.44444445,0.33333334,0.4722222,0.41666666,0.5277778,0.5833333,0.4722222,,0.5555556,0.5555556,0.5833333,0.5555556,0.5555556,0.6111111,0.5,0.5,0.6388889,0.6666667,0.6666667,0.6666667,0.6944444,0.6666667,0.75,0.6944444,0.6944444,0.7222222,0.8333333,0.7777778,0.7777778,0.6666667,0.8055556,0.8611111,0.8888889,0.9444444,0.9444444,0.9722222,0.9166667,0.9722222
+TorchMLP (tuned),0.027777778,0.055555556,0.11111111,0.11111111,0.055555556,0.11111111,0.083333336,0.083333336,0.19444445,0.083333336,0.1388889,0.2777778,0.19444445,0.16666667,0.11111111,0.30555555,0.2777778,0.19444445,0.22222222,0.22222222,0.22222222,0.22222222,0.19444445,0.19444445,0.30555555,0.33333334,0.31944445,0.33333334,0.25,0.33333334,0.2777778,0.30555555,0.44444445,0.2777778,0.33333334,0.3888889,0.33333334,0.3611111,0.30555555,0.2777778,0.3888889,0.3888889,0.3888889,0.44444445,0.2777778,0.44444445,0.5277778,0.5138889,0.5555556,0.44444445,,0.5,0.5555556,0.5,0.5,0.6388889,0.5694444,0.5555556,0.6388889,0.6944444,0.6666667,0.6388889,0.6944444,0.6388889,0.6666667,0.7222222,0.7222222,0.6944444,0.9166667,0.7777778,0.8333333,0.6388889,0.8055556,0.8333333,0.8333333,0.8055556,0.8333333,0.9166667,0.9166667,0.9444444
+FastaiMLP (tuned + ensembled),0.083333336,0.11111111,0.1388889,0.11111111,0.083333336,0.027777778,0.055555556,0.1388889,0.11111111,0.11111111,0.22222222,0.25,0.19444445,0.19444445,0.22222222,0.2777778,0.30555555,0.22222222,0.16666667,0.19444445,0.2777778,0.2777778,0.30555555,0.30555555,0.16666667,0.30555555,0.25,0.3611111,0.25,0.25,0.2777778,0.25,0.33333334,0.3472222,0.2777778,0.25,0.2777778,0.41666666,0.4722222,0.30555555,0.33333334,0.30555555,0.3611111,0.33333334,0.33333334,0.44444445,0.4722222,0.5,0.3888889,0.44444445,0.5,,0.5,0.5555556,0.5555556,0.5555556,0.4722222,0.5277778,0.6944444,0.6388889,0.6388889,0.6111111,0.6111111,0.7222222,0.7222222,0.6388889,0.75,0.7222222,0.7222222,0.7222222,0.8611111,0.75,0.9722222,0.8055556,0.8611111,0.8333333,0.8611111,0.8888889,0.8888889,0.9444444
+ExtraTrees (tuned),0.055555556,0.027777778,0.083333336,0.083333336,0.083333336,0.1388889,0.11111111,0.083333336,0.1388889,0.19444445,0.19444445,0.11111111,0.1388889,0.16666667,0.22222222,0.25,0.16666667,0.30555555,0.22222222,0.2777778,0.30555555,0.22222222,0.30555555,0.30555555,0.2777778,0.22222222,0.25,0.33333334,0.30555555,0.25,0.30555555,0.33333334,0.2777778,0.3888889,0.3888889,0.2777778,0.33333334,0.3888889,0.44444445,0.3611111,0.25,0.3888889,0.33333334,0.3888889,0.3888889,0.41666666,0.4722222,0.31944445,0.4722222,0.41666666,0.44444445,0.5,,0.6111111,0.6111111,0.5277778,0.44444445,0.4722222,0.5,0.6666667,0.6666667,0.6944444,0.6388889,0.7222222,0.6388889,0.6666667,0.6666667,0.7777778,0.5833333,0.7777778,0.7777778,0.8333333,0.6944444,0.8888889,0.7222222,0.75,0.75,0.8333333,0.9722222,0.9722222
+Nori-30M (default),0.083333336,0.1388889,0.22222222,0.19444445,0.19444445,0.22222222,0.25,0.22222222,0.25,0.25,0.2361111,0.19444445,0.30555555,0.25,0.30555555,0.25,0.33333334,0.3888889,0.2777778,0.25,0.30555555,0.2777778,0.3472222,0.3611111,0.3611111,0.30555555,0.2777778,0.3888889,0.30555555,0.30555555,0.33333334,0.30555555,0.3888889,0.30555555,0.3888889,0.33333334,0.3888889,0.30555555,0.3611111,0.3888889,0.3611111,0.3888889,0.3888889,0.44444445,0.44444445,0.3888889,0.3888889,0.3611111,0.5,0.44444445,0.5,0.44444445,0.3888889,,0.5277778,0.33333334,0.4722222,0.41666666,0.5,0.4722222,0.5,0.4027778,0.5,0.4722222,0.5277778,0.5555556,0.5833333,0.5555556,0.5833333,0.6111111,0.5833333,0.6111111,0.6111111,0.6944444,0.6111111,0.7222222,0.7222222,0.75,0.8333333,0.9722222
+Nori (default),0.055555556,0.1388889,0.16666667,0.16666667,0.16666667,0.19444445,0.25,0.19444445,0.25,0.22222222,0.20833333,0.22222222,0.30555555,0.25,0.30555555,0.22222222,0.33333334,0.3888889,0.2777778,0.25,0.30555555,0.2777778,0.3472222,0.3611111,0.3888889,0.30555555,0.2777778,0.3611111,0.30555555,0.30555555,0.33333334,0.30555555,0.3888889,0.30555555,0.3888889,0.33333334,0.3888889,0.30555555,0.3611111,0.3888889,0.3611111,0.3888889,0.3888889,0.44444445,0.44444445,0.3888889,0.3888889,0.3611111,0.5,0.44444445,0.5,0.44444445,0.3888889,0.4722222,,0.33333334,0.4722222,0.41666666,0.5,0.4722222,0.5,0.4027778,0.5,0.4722222,0.5277778,0.5555556,0.5833333,0.5555556,0.5833333,0.6111111,0.5833333,0.6111111,0.6111111,0.6944444,0.6111111,0.7222222,0.7222222,0.75,0.8333333,0.9722222
+RandomForest (tuned + ensembled),0.055555556,0.055555556,0.055555556,0.083333336,0.055555556,0.11111111,0.1388889,0.083333336,0.19444445,0.1388889,0.1388889,0.19444445,0.1388889,0.16666667,0.25,0.22222222,0.2777778,0.22222222,0.11111111,0.16666667,0.19444445,0.1388889,0.30555555,0.25,0.2777778,0.19444445,0.16666667,0.30555555,0.25,0.19444445,0.25,0.22222222,0.30555555,0.30555555,0.2777778,0.19444445,0.30555555,0.41666666,0.3888889,0.30555555,0.25,0.3611111,0.30555555,0.33333334,0.33333334,0.3888889,0.30555555,0.2777778,0.3611111,0.3888889,0.3611111,0.44444445,0.4722222,0.6666667,0.6666667,,0.4722222,0.5,0.44444445,0.5555556,0.5833333,0.7777778,0.5277778,0.6388889,0.5555556,0.6388889,0.6944444,0.7777778,0.6666667,0.7222222,0.6944444,0.8888889,0.6666667,0.8055556,0.6388889,0.7777778,0.75,0.8055556,0.9166667,0.9722222
+LightGBM (default),0.055555556,0.055555556,0.027777778,0.027777778,0.055555556,0.055555556,0.083333336,0.055555556,0.083333336,0.083333336,0.083333336,0.19444445,0.11111111,0.16666667,0.16666667,0.19444445,0.2777778,0.16666667,0.1388889,0.055555556,0.19444445,0.19444445,0.22222222,0.16666667,0.22222222,0.22222222,0.027777778,0.33333334,0.19444445,0.083333336,0.22222222,0.25,0.2777778,0.25,0.30555555,0.2777778,0.30555555,0.33333334,0.3611111,0.30555555,0.30555555,0.3888889,0.3611111,0.2777778,0.25,0.44444445,0.41666666,0.4722222,0.41666666,0.5,0.43055555,0.5277778,0.5555556,0.5277778,0.5277778,0.5277778,,0.4722222,0.5555556,0.6388889,0.6388889,0.5833333,0.6111111,0.6111111,0.75,0.6111111,0.6944444,0.6944444,0.6666667,0.7222222,0.75,0.7222222,0.7777778,0.7222222,0.75,0.7777778,0.7777778,0.8888889,0.9166667,0.9166667
+XGBoost (default),0.027777778,0.027777778,0.055555556,0.0,0.027777778,0.055555556,0.083333336,0.027777778,0.055555556,0.083333336,0.11111111,0.16666667,0.1388889,0.11111111,0.16666667,0.16666667,0.22222222,0.16666667,0.1388889,0.083333336,0.1388889,0.1388889,0.19444445,0.16666667,0.2777778,0.22222222,0.11111111,0.3611111,0.16666667,0.1388889,0.19444445,0.22222222,0.30555555,0.19444445,0.2777778,0.30555555,0.33333334,0.30555555,0.30555555,0.22222222,0.33333334,0.30555555,0.25,0.25,0.25,0.41666666,0.4722222,0.5,0.3888889,0.5,0.44444445,0.4722222,0.5277778,0.5833333,0.5833333,0.5,0.5277778,,0.5555556,0.6111111,0.6111111,0.5833333,0.5555556,0.5833333,0.7222222,0.6388889,0.7222222,0.6666667,0.6944444,0.6944444,0.6666667,0.7222222,0.7222222,0.7222222,0.8055556,0.8055556,0.8333333,0.8611111,0.8611111,0.9444444
+FastaiMLP (tuned),0.083333336,0.055555556,0.083333336,0.083333336,0.11111111,0.0,0.11111111,0.11111111,0.1388889,0.083333336,0.16666667,0.19444445,0.11111111,0.16666667,0.22222222,0.19444445,0.19444445,0.16666667,0.16666667,0.16666667,0.19444445,0.16666667,0.25,0.16666667,0.19444445,0.25,0.22222222,0.30555555,0.19444445,0.19444445,0.19444445,0.22222222,0.19444445,0.25,0.30555555,0.22222222,0.2777778,0.33333334,0.3888889,0.19444445,0.30555555,0.25,0.3611111,0.2777778,0.2777778,0.33333334,0.3888889,0.5,0.3888889,0.3611111,0.3611111,0.30555555,0.5,0.5,0.5,0.5555556,0.44444445,0.44444445,,0.6111111,0.6111111,0.5833333,0.5,0.6111111,0.6666667,0.5555556,0.6666667,0.6944444,0.6666667,0.6944444,0.75,0.6666667,0.75,0.75,0.75,0.8055556,0.8055556,0.8611111,0.8611111,0.9166667
+TabSTAR (tuned),0.083333336,0.1388889,0.1388889,0.16666667,0.16666667,0.1388889,0.1388889,0.19444445,0.22222222,0.25,0.25,0.19444445,0.22222222,0.22222222,0.22222222,0.18055555,0.22222222,0.19444445,0.2777778,0.22222222,0.25,0.25,0.2777778,0.25,0.25,0.30555555,0.22222222,0.30555555,0.2777778,0.22222222,0.33333334,0.22222222,0.33333334,0.2777778,0.22222222,0.30555555,0.3611111,0.30555555,0.2777778,0.33333334,0.33333334,0.2777778,0.33333334,0.2777778,0.3611111,0.30555555,0.25,0.3611111,0.30555555,0.33333334,0.30555555,0.3611111,0.33333334,0.5277778,0.5277778,0.44444445,0.3611111,0.3888889,0.3888889,,0.45833334,0.44444445,0.5277778,0.41666666,0.41666666,0.5277778,0.5555556,0.5833333,0.5,0.9444444,0.5277778,0.6111111,0.6111111,0.6111111,0.5555556,0.6944444,0.7222222,0.7777778,0.7777778,0.8611111
+TabSTAR (tuned + ensembled),0.11111111,0.1388889,0.16666667,0.16666667,0.16666667,0.1388889,0.1388889,0.22222222,0.22222222,0.25,0.2777778,0.25,0.25,0.19444445,0.22222222,0.19444445,0.25,0.19444445,0.2777778,0.25,0.25,0.25,0.2777778,0.25,0.25,0.30555555,0.22222222,0.2777778,0.25,0.22222222,0.30555555,0.25,0.30555555,0.2777778,0.25,0.30555555,0.3611111,0.30555555,0.2777778,0.30555555,0.30555555,0.25,0.2777778,0.2777778,0.33333334,0.30555555,0.2777778,0.33333334,0.25,0.33333334,0.33333334,0.3611111,0.33333334,0.5,0.5,0.41666666,0.3611111,0.3888889,0.3888889,0.5416667,,0.44444445,0.5277778,0.44444445,0.44444445,0.5277778,0.5555556,0.5555556,0.5277778,0.9722222,0.5,0.5833333,0.5833333,0.5833333,0.5555556,0.6944444,0.6944444,0.7777778,0.7777778,0.8611111
+RandomForest (tuned),0.055555556,0.027777778,0.055555556,0.083333336,0.083333336,0.055555556,0.1388889,0.11111111,0.11111111,0.055555556,0.11111111,0.16666667,0.1388889,0.11111111,0.16666667,0.16666667,0.19444445,0.22222222,0.11111111,0.083333336,0.11111111,0.083333336,0.19444445,0.22222222,0.25,0.16666667,0.11111111,0.25,0.16666667,0.1388889,0.2777778,0.16666667,0.2777778,0.2777778,0.22222222,0.22222222,0.2777778,0.2638889,0.33333334,0.2777778,0.22222222,0.3611111,0.30555555,0.22222222,0.2777778,0.3888889,0.25,0.19444445,0.2777778,0.33333334,0.3611111,0.3888889,0.30555555,0.5972222,0.5972222,0.22222222,0.41666666,0.41666666,0.41666666,0.5555556,0.5555556,,0.5,0.5833333,0.5833333,0.5555556,0.6666667,0.6666667,0.5277778,0.6666667,0.6944444,0.7638889,0.6944444,0.8055556,0.6111111,0.7777778,0.75,0.7777778,0.8888889,0.9722222
+PerpetualBooster (tuned + ensembled),0.0,0.027777778,0.083333336,0.055555556,0.027777778,0.055555556,0.083333336,0.055555556,0.055555556,0.055555556,0.16666667,0.22222222,0.19444445,0.11111111,0.16666667,0.19444445,0.22222222,0.19444445,0.11111111,0.1388889,0.16666667,0.11111111,0.25,0.1388889,0.22222222,0.25,0.22222222,0.25,0.16666667,0.22222222,0.22222222,0.1388889,0.25,0.22222222,0.19444445,0.2777778,0.25,0.2777778,0.30555555,0.19444445,0.30555555,0.33333334,0.33333334,0.22222222,0.30555555,0.33333334,0.33333334,0.3611111,0.2777778,0.30555555,0.30555555,0.3888889,0.3611111,0.5,0.5,0.4722222,0.3888889,0.44444445,0.5,0.4722222,0.4722222,0.5,,0.4722222,0.5555556,0.7361111,0.5833333,0.5277778,0.5277778,0.5,0.6111111,0.6388889,0.6388889,0.6111111,0.8611111,0.7777778,0.7777778,0.8055556,0.75,0.8888889
+OrionMSP (default),0.027777778,0.0,0.0,0.027777778,0.027777778,0.027777778,0.11111111,0.027777778,0.055555556,0.083333336,0.15277778,0.027777778,0.055555556,0.15277778,0.1388889,0.15277778,0.055555556,0.1388889,0.16666667,0.16666667,0.19444445,0.22222222,0.20833333,0.16666667,0.11111111,0.1388889,0.16666667,0.2361111,0.19444445,0.19444445,0.19444445,0.22222222,0.16666667,0.22222222,0.22222222,0.25,0.30555555,0.25,0.30555555,0.25,0.19444445,0.30555555,0.2777778,0.25,0.2777778,0.33333334,0.2777778,0.2777778,0.30555555,0.33333334,0.3611111,0.2777778,0.2777778,0.5277778,0.5277778,0.3611111,0.3888889,0.41666666,0.3888889,0.5833333,0.5555556,0.41666666,0.5277778,,0.5555556,0.5277778,0.5,0.5833333,0.5277778,0.7222222,0.6666667,0.6388889,0.5555556,0.6666667,0.6666667,0.8055556,0.8055556,0.8333333,0.8611111,0.9722222
+iLTM (default),0.083333336,0.027777778,0.055555556,0.083333336,0.055555556,0.027777778,0.1388889,0.083333336,0.055555556,0.083333336,0.11111111,0.1388889,0.1388889,0.055555556,0.1388889,0.1388889,0.1388889,0.1388889,0.1388889,0.11111111,0.1388889,0.1388889,0.1388889,0.19444445,0.19444445,0.22222222,0.1388889,0.22222222,0.19444445,0.16666667,0.22222222,0.16666667,0.25,0.22222222,0.25,0.2361111,0.25,0.25,0.16666667,0.2777778,0.19444445,0.25,0.22222222,0.22222222,0.25,0.25,0.2777778,0.3611111,0.25,0.25,0.33333334,0.2777778,0.3611111,0.4722222,0.4722222,0.44444445,0.25,0.2777778,0.33333334,0.5833333,0.5555556,0.41666666,0.44444445,0.44444445,,0.41666666,0.5,0.4722222,0.5555556,0.6666667,0.5277778,0.5277778,0.5555556,0.75,0.5833333,0.7222222,0.7222222,0.8055556,0.8055556,0.8888889
+PerpetualBooster (tuned),0.0,0.0,0.055555556,0.027777778,0.0,0.0,0.055555556,0.027777778,0.027777778,0.027777778,0.11111111,0.1388889,0.1388889,0.11111111,0.16666667,0.19444445,0.1388889,0.11111111,0.055555556,0.11111111,0.083333336,0.11111111,0.19444445,0.11111111,0.16666667,0.19444445,0.16666667,0.25,0.1388889,0.16666667,0.22222222,0.083333336,0.19444445,0.19444445,0.1388889,0.16666667,0.25,0.22222222,0.2777778,0.19444445,0.22222222,0.25,0.2777778,0.19444445,0.22222222,0.33333334,0.2777778,0.2777778,0.25,0.30555555,0.2777778,0.3611111,0.33333334,0.44444445,0.44444445,0.3611111,0.3888889,0.3611111,0.44444445,0.4722222,0.4722222,0.44444445,0.2638889,0.4722222,0.5833333,,0.5833333,0.5,0.5277778,0.5,0.6388889,0.5833333,0.6111111,0.5555556,0.7638889,0.75,0.7222222,0.7777778,0.7222222,0.8888889
+xRFM (default),0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.083333336,0.055555556,0.083333336,0.083333336,0.11111111,0.16666667,0.055555556,0.11111111,0.083333336,0.11111111,0.16666667,0.11111111,0.16666667,0.11111111,0.16666667,0.19444445,0.1388889,0.19444445,0.19444445,0.16666667,0.11111111,0.22222222,0.19444445,0.16666667,0.22222222,0.1388889,0.16666667,0.19444445,0.19444445,0.22222222,0.1388889,0.25,0.30555555,0.30555555,0.2777778,0.055555556,0.22222222,0.22222222,0.22222222,0.25,0.19444445,0.33333334,0.25,0.30555555,0.30555555,0.2777778,0.25,0.33333334,0.41666666,0.41666666,0.30555555,0.30555555,0.2777778,0.33333334,0.44444445,0.44444445,0.33333334,0.41666666,0.5,0.5,0.41666666,,0.5277778,0.5,0.6388889,0.5555556,0.5833333,0.5,0.5555556,0.5555556,0.6666667,0.6666667,0.7222222,0.7777778,0.9166667
+ExtraTrees (default),0.027777778,0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.11111111,0.055555556,0.027777778,0.027777778,0.11111111,0.027777778,0.055555556,0.11111111,0.19444445,0.11111111,0.083333336,0.19444445,0.055555556,0.11111111,0.16666667,0.1388889,0.1388889,0.22222222,0.1388889,0.083333336,0.11111111,0.25,0.22222222,0.1388889,0.19444445,0.19444445,0.11111111,0.22222222,0.2777778,0.16666667,0.22222222,0.2777778,0.3611111,0.2777778,0.11111111,0.2777778,0.22222222,0.25,0.2777778,0.2777778,0.22222222,0.083333336,0.30555555,0.2777778,0.30555555,0.2777778,0.22222222,0.44444445,0.44444445,0.22222222,0.30555555,0.33333334,0.30555555,0.41666666,0.44444445,0.33333334,0.4722222,0.41666666,0.5277778,0.5,0.4722222,,0.3888889,0.6388889,0.5555556,0.6111111,0.5555556,0.6111111,0.5555556,0.6666667,0.6666667,0.7222222,0.8055556,0.9444444
+TorchMLP (default),0.027777778,0.055555556,0.055555556,0.055555556,0.027777778,0.027777778,0.055555556,0.055555556,0.083333336,0.027777778,0.11111111,0.1388889,0.11111111,0.055555556,0.1388889,0.1388889,0.1388889,0.027777778,0.055555556,0.11111111,0.083333336,0.083333336,0.16666667,0.027777778,0.1388889,0.11111111,0.11111111,0.22222222,0.1388889,0.1388889,0.19444445,0.11111111,0.22222222,0.16666667,0.16666667,0.083333336,0.19444445,0.22222222,0.22222222,0.083333336,0.11111111,0.16666667,0.11111111,0.16666667,0.055555556,0.19444445,0.19444445,0.3611111,0.2777778,0.16666667,0.083333336,0.2777778,0.41666666,0.41666666,0.41666666,0.33333334,0.33333334,0.30555555,0.33333334,0.5,0.4722222,0.4722222,0.4722222,0.4722222,0.44444445,0.4722222,0.5,0.6111111,,0.6111111,0.6111111,0.5555556,0.5277778,0.6111111,0.5833333,0.6666667,0.7222222,0.8055556,0.7777778,0.8888889
+TabSTAR (default),0.055555556,0.1388889,0.1388889,0.1388889,0.16666667,0.11111111,0.11111111,0.1388889,0.16666667,0.16666667,0.19444445,0.16666667,0.19444445,0.19444445,0.16666667,0.1388889,0.19444445,0.16666667,0.19444445,0.19444445,0.22222222,0.22222222,0.22222222,0.19444445,0.16666667,0.22222222,0.19444445,0.25,0.19444445,0.19444445,0.19444445,0.19444445,0.25,0.16666667,0.19444445,0.19444445,0.16666667,0.25,0.19444445,0.19444445,0.25,0.19444445,0.16666667,0.19444445,0.2777778,0.19444445,0.25,0.25,0.19444445,0.22222222,0.22222222,0.2777778,0.22222222,0.3888889,0.3888889,0.2777778,0.2777778,0.30555555,0.30555555,0.055555556,0.027777778,0.33333334,0.5,0.2777778,0.33333334,0.5,0.3611111,0.3611111,0.3888889,,0.41666666,0.44444445,0.3611111,0.5555556,0.5277778,0.5555556,0.5555556,0.6388889,0.6388889,0.8333333
+TabFlex (default),0.0,0.055555556,0.055555556,0.11111111,0.083333336,0.027777778,0.055555556,0.1388889,0.11111111,0.055555556,0.15277778,0.055555556,0.027777778,0.1388889,0.125,0.16666667,0.11111111,0.083333336,0.083333336,0.1388889,0.1388889,0.16666667,0.15277778,0.083333336,0.027777778,0.11111111,0.16666667,0.15277778,0.16666667,0.16666667,0.19444445,0.16666667,0.1388889,0.19444445,0.1388889,0.1388889,0.16666667,0.22222222,0.22222222,0.16666667,0.16666667,0.16666667,0.083333336,0.1388889,0.11111111,0.19444445,0.22222222,0.19444445,0.19444445,0.22222222,0.16666667,0.1388889,0.22222222,0.41666666,0.41666666,0.30555555,0.25,0.33333334,0.25,0.4722222,0.5,0.30555555,0.3888889,0.33333334,0.4722222,0.3611111,0.44444445,0.44444445,0.3888889,0.5833333,,0.5277778,0.5277778,0.5833333,0.5555556,0.7222222,0.7222222,0.7777778,0.8055556,0.9166667
+RandomForest (default),0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.055555556,0.083333336,0.055555556,0.027777778,0.027777778,0.083333336,0.055555556,0.083333336,0.027777778,0.097222224,0.11111111,0.11111111,0.16666667,0.055555556,0.027777778,0.083333336,0.055555556,0.16666667,0.1388889,0.16666667,0.083333336,0.055555556,0.20833333,0.083333336,0.083333336,0.1388889,0.11111111,0.19444445,0.1388889,0.16666667,0.11111111,0.19444445,0.19444445,0.25,0.19444445,0.1388889,0.25,0.19444445,0.25,0.25,0.25,0.16666667,0.1388889,0.30555555,0.33333334,0.3611111,0.25,0.16666667,0.3888889,0.3888889,0.11111111,0.2777778,0.2777778,0.33333334,0.3888889,0.41666666,0.2361111,0.3611111,0.3611111,0.4722222,0.41666666,0.41666666,0.3888889,0.44444445,0.5555556,0.4722222,,0.5,0.6388889,0.5277778,0.7222222,0.7222222,0.75,0.8333333,0.9722222
+FastaiMLP (default),0.0,0.0,0.027777778,0.027777778,0.027777778,0.0,0.027777778,0.055555556,0.027777778,0.027777778,0.083333336,0.027777778,0.027777778,0.055555556,0.11111111,0.11111111,0.027777778,0.083333336,0.055555556,0.083333336,0.083333336,0.11111111,0.16666667,0.1388889,0.055555556,0.083333336,0.11111111,0.19444445,0.11111111,0.11111111,0.16666667,0.083333336,0.1388889,0.1388889,0.1388889,0.11111111,0.16666667,0.19444445,0.19444445,0.19444445,0.1388889,0.1388889,0.083333336,0.16666667,0.1388889,0.16666667,0.19444445,0.25,0.22222222,0.19444445,0.19444445,0.027777778,0.30555555,0.3888889,0.3888889,0.33333334,0.22222222,0.2777778,0.25,0.3888889,0.41666666,0.30555555,0.3611111,0.44444445,0.44444445,0.3888889,0.5,0.44444445,0.4722222,0.6388889,0.4722222,0.5,,0.5555556,0.6666667,0.6666667,0.6944444,0.7222222,0.7777778,0.9166667
+KNN (tuned + ensembled),0.055555556,0.027777778,0.027777778,0.055555556,0.027777778,0.083333336,0.055555556,0.083333336,0.055555556,0.083333336,0.083333336,0.083333336,0.027777778,0.083333336,0.083333336,0.083333336,0.11111111,0.11111111,0.055555556,0.055555556,0.11111111,0.1388889,0.11111111,0.11111111,0.083333336,0.083333336,0.1388889,0.19444445,0.11111111,0.11111111,0.15277778,0.11111111,0.11111111,0.11111111,0.19444445,0.083333336,0.22222222,0.1388889,0.1388889,0.19444445,0.083333336,0.11111111,0.11111111,0.22222222,0.16666667,0.11111111,0.16666667,0.055555556,0.22222222,0.1388889,0.16666667,0.19444445,0.11111111,0.30555555,0.30555555,0.19444445,0.2777778,0.2777778,0.25,0.3888889,0.41666666,0.19444445,0.3888889,0.33333334,0.25,0.44444445,0.44444445,0.3888889,0.3888889,0.44444445,0.41666666,0.3611111,0.44444445,,0.5,0.6388889,0.6388889,0.7222222,0.9722222,0.9722222
+PerpetualBooster (default),0.0,0.027777778,0.055555556,0.055555556,0.027777778,0.055555556,0.027777778,0.055555556,0.027777778,0.027777778,0.055555556,0.11111111,0.055555556,0.027777778,0.055555556,0.11111111,0.083333336,0.055555556,0.055555556,0.11111111,0.055555556,0.055555556,0.11111111,0.055555556,0.11111111,0.1388889,0.1388889,0.1388889,0.11111111,0.1388889,0.1388889,0.055555556,0.19444445,0.16666667,0.16666667,0.11111111,0.1388889,0.1388889,0.19444445,0.11111111,0.083333336,0.11111111,0.083333336,0.16666667,0.11111111,0.1388889,0.16666667,0.25,0.19444445,0.11111111,0.16666667,0.1388889,0.2777778,0.3888889,0.3888889,0.3611111,0.25,0.19444445,0.25,0.44444445,0.44444445,0.3888889,0.1388889,0.33333334,0.41666666,0.2361111,0.44444445,0.44444445,0.41666666,0.4722222,0.44444445,0.4722222,0.33333334,0.5,,0.6388889,0.6388889,0.6944444,0.6111111,0.8333333
+Linear (tuned + ensembled),0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.027777778,0.083333336,0.083333336,0.055555556,0.027777778,0.055555556,0.11111111,0.055555556,0.083333336,0.027777778,0.027777778,0.027777778,0.027777778,0.083333336,0.083333336,0.055555556,0.083333336,0.055555556,0.083333336,0.11111111,0.16666667,0.055555556,0.22222222,0.083333336,0.083333336,0.1388889,0.083333336,0.055555556,0.027777778,0.055555556,0.11111111,0.027777778,0.083333336,0.1388889,0.11111111,0.083333336,0.11111111,0.16666667,0.1388889,0.055555556,0.19444445,0.16666667,0.25,0.2777778,0.2777778,0.22222222,0.22222222,0.19444445,0.19444445,0.30555555,0.30555555,0.22222222,0.22222222,0.19444445,0.2777778,0.25,0.33333334,0.33333334,0.33333334,0.44444445,0.2777778,0.2777778,0.33333334,0.3611111,0.3611111,,0.7083333,0.7222222,0.5277778,0.75
+Linear (tuned),0.027777778,0.027777778,0.027777778,0.027777778,0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.027777778,0.027777778,0.083333336,0.083333336,0.041666668,0.027777778,0.055555556,0.11111111,0.055555556,0.055555556,0.027777778,0.027777778,0.027777778,0.027777778,0.083333336,0.083333336,0.027777778,0.055555556,0.055555556,0.083333336,0.083333336,0.1388889,0.027777778,0.19444445,0.055555556,0.055555556,0.11111111,0.11111111,0.083333336,0.027777778,0.055555556,0.1388889,0.027777778,0.083333336,0.1388889,0.083333336,0.055555556,0.11111111,0.1388889,0.1388889,0.055555556,0.16666667,0.1388889,0.25,0.2777778,0.2777778,0.25,0.22222222,0.16666667,0.19444445,0.2777778,0.30555555,0.25,0.22222222,0.19444445,0.2777778,0.2777778,0.33333334,0.33333334,0.2777778,0.44444445,0.2777778,0.2777778,0.30555555,0.3611111,0.3611111,0.29166666,,0.6388889,0.5,0.7222222
+Linear (default),0.0,0.0,0.027777778,0.055555556,0.027777778,0.0,0.0,0.055555556,0.027777778,0.0,0.055555556,0.11111111,0.083333336,0.055555556,0.027777778,0.027777778,0.11111111,0.055555556,0.055555556,0.055555556,0.027777778,0.027777778,0.055555556,0.055555556,0.083333336,0.055555556,0.055555556,0.027777778,0.083333336,0.055555556,0.16666667,0.027777778,0.1388889,0.083333336,0.027777778,0.11111111,0.1388889,0.027777778,0.027777778,0.083333336,0.11111111,0.027777778,0.027777778,0.055555556,0.027777778,0.083333336,0.11111111,0.16666667,0.027777778,0.027777778,0.083333336,0.11111111,0.16666667,0.25,0.25,0.19444445,0.11111111,0.1388889,0.1388889,0.22222222,0.22222222,0.22222222,0.19444445,0.16666667,0.19444445,0.22222222,0.2777778,0.2777778,0.19444445,0.3611111,0.22222222,0.25,0.2777778,0.2777778,0.30555555,0.2777778,0.3611111,,0.4722222,0.6666667
+KNN (tuned),0.027777778,0.027777778,0.0,0.055555556,0.027777778,0.027777778,0.027777778,0.055555556,0.027777778,0.0,0.055555556,0.055555556,0.027777778,0.027777778,0.027777778,0.083333336,0.055555556,0.0,0.055555556,0.0,0.027777778,0.055555556,0.055555556,0.027777778,0.027777778,0.027777778,0.083333336,0.083333336,0.055555556,0.083333336,0.1388889,0.027777778,0.055555556,0.083333336,0.055555556,0.027777778,0.11111111,0.027777778,0.027777778,0.027777778,0.027777778,0.083333336,0.055555556,0.083333336,0.083333336,0.083333336,0.083333336,0.055555556,0.055555556,0.083333336,0.083333336,0.11111111,0.027777778,0.16666667,0.16666667,0.083333336,0.083333336,0.1388889,0.1388889,0.22222222,0.22222222,0.11111111,0.25,0.1388889,0.19444445,0.2777778,0.22222222,0.19444445,0.22222222,0.3611111,0.19444445,0.16666667,0.22222222,0.027777778,0.3888889,0.4722222,0.5,0.5277778,,0.8888889
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.027777778,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.027777778,0.0,0.0,0.0,0.0,0.0,0.027777778,0.027777778,0.0,0.027777778,0.0,0.027777778,0.0,0.0,0.027777778,0.0,0.027777778,0.0,0.0,0.0,0.0,0.0,0.0,0.055555556,0.0,0.027777778,0.0,0.0,0.055555556,0.027777778,0.027777778,0.0,0.027777778,0.027777778,0.055555556,0.055555556,0.027777778,0.027777778,0.027777778,0.027777778,0.083333336,0.055555556,0.083333336,0.1388889,0.1388889,0.027777778,0.11111111,0.027777778,0.11111111,0.11111111,0.083333336,0.055555556,0.11111111,0.16666667,0.083333336,0.027777778,0.083333336,0.027777778,0.16666667,0.25,0.2777778,0.33333334,0.11111111,
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/leaderboard_overview_explorer.html
index c443561dbb9934d00ed1df3a2b96627a0a9fc870..10e41ddae414344f206e7e451e48ba66be2dcd7d 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Binary | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..02649da49d593d441082f501a66f1a02a65b6720
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_binary/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1774.0,"score":0.796,"rank":6.47,"hrank":1.74,"imp":5.151,"train_time":37.56,"predict_time":7.056,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+165\/-123","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1632.0,"score":0.599,"rank":11.75,"hrank":4.45,"imp":10.298,"train_time":2.9,"predict_time":0.353,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-72","imputed":false},{"position":2,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1607.0,"score":0.596,"rank":13.0,"hrank":3.98,"imp":7.753,"train_time":270.87,"predict_time":4.72,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+144\/-105","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1584.0,"score":0.526,"rank":14.23,"hrank":7.63,"imp":11.898,"train_time":4.82,"predict_time":0.467,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-72","imputed":false},{"position":4,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1577.0,"score":0.516,"rank":14.6,"hrank":7.33,"imp":11.656,"train_time":1.77,"predict_time":0.16,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-84","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1547.0,"score":0.465,"rank":16.38,"hrank":7.89,"imp":12.33,"train_time":1878.36,"predict_time":10.744,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-67","imputed":false},{"position":6,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1539.0,"score":0.469,"rank":16.87,"hrank":6.61,"imp":12.238,"train_time":1231.24,"predict_time":1.359,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-75","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1537.0,"score":0.455,"rank":16.98,"hrank":10.62,"imp":12.649,"train_time":4.79,"predict_time":0.708,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-52","imputed":false},{"position":8,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1515.0,"score":0.444,"rank":18.42,"hrank":10.07,"imp":13.076,"train_time":1878.36,"predict_time":1.035,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-62","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1487.0,"score":0.398,"rank":20.37,"hrank":13.26,"imp":13.691,"train_time":2722.13,"predict_time":9.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+65\/-57","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1416.0,"score":0.338,"rank":25.68,"hrank":12.74,"imp":15.368,"train_time":2722.13,"predict_time":0.265,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-72","imputed":false},{"position":11,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1408.0,"score":0.316,"rank":26.4,"hrank":14.4,"imp":15.5,"train_time":8274.67,"predict_time":361.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-72","imputed":false},{"position":12,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1406.0,"score":0.324,"rank":26.5,"hrank":11.37,"imp":15.515,"train_time":1641.74,"predict_time":1.602,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-68","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1392.0,"score":0.297,"rank":27.68,"hrank":11.84,"imp":15.662,"train_time":1.48,"predict_time":0.167,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-70","imputed":false},{"position":14,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1385.0,"score":0.285,"rank":28.3,"hrank":10.61,"imp":15.127,"train_time":8.05,"predict_time":2.007,"imputed_pct":3.33,"hardware":"GPU","elo_ci":"+70\/-59","imputed":true},{"position":15,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1376.0,"score":0.279,"rank":29.0,"hrank":14.01,"imp":15.816,"train_time":735.46,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-61","imputed":false},{"position":16,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1376.0,"score":0.242,"rank":29.07,"hrank":21.47,"imp":16.405,"train_time":328.64,"predict_time":0.773,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-50","imputed":false},{"position":17,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1374.0,"score":0.325,"rank":29.2,"hrank":8.75,"imp":16.819,"train_time":14.27,"predict_time":4.304,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-90","imputed":false},{"position":18,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1368.0,"score":0.263,"rank":29.68,"hrank":17.34,"imp":16.07,"train_time":735.46,"predict_time":0.301,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-73","imputed":false},{"position":19,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1368.0,"score":0.258,"rank":29.73,"hrank":17.79,"imp":15.978,"train_time":3.18,"predict_time":0.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-49","imputed":false},{"position":20,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1367.0,"score":0.277,"rank":29.78,"hrank":19.15,"imp":17.07,"train_time":275.93,"predict_time":0.306,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-71","imputed":false},{"position":21,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1367.0,"score":0.266,"rank":29.83,"hrank":12.04,"imp":15.181,"train_time":5891.3,"predict_time":409.932,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-60","imputed":false},{"position":22,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1362.0,"score":0.279,"rank":30.18,"hrank":15.77,"imp":16.506,"train_time":1641.74,"predict_time":0.219,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-79","imputed":false},{"position":23,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1362.0,"score":0.255,"rank":30.18,"hrank":18.66,"imp":16.896,"train_time":463.01,"predict_time":0.613,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-68","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1361.0,"score":0.27,"rank":30.33,"hrank":15.63,"imp":16.551,"train_time":3436.74,"predict_time":0.405,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-74","imputed":false},{"position":25,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1354.0,"score":0.247,"rank":30.88,"hrank":18.0,"imp":16.508,"train_time":1.11,"predict_time":0.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-73","imputed":false},{"position":26,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1342.0,"score":0.279,"rank":31.98,"hrank":16.97,"imp":16.706,"train_time":3436.74,"predict_time":8.476,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-93","imputed":false},{"position":27,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1335.0,"score":0.242,"rank":32.57,"hrank":17.5,"imp":17.7,"train_time":275.93,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-79","imputed":false},{"position":28,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1329.0,"score":0.223,"rank":33.08,"hrank":17.97,"imp":17.023,"train_time":5891.3,"predict_time":39.564,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-70","imputed":false},{"position":29,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1328.0,"score":0.223,"rank":33.17,"hrank":13.64,"imp":17.223,"train_time":463.01,"predict_time":0.108,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-77","imputed":false},{"position":30,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1327.0,"score":0.224,"rank":33.3,"hrank":21.67,"imp":18.222,"train_time":248.52,"predict_time":1.699,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-61","imputed":false},{"position":31,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1316.0,"score":0.191,"rank":34.23,"hrank":28.09,"imp":17.527,"train_time":328.64,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-59","imputed":false},{"position":32,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1314.0,"score":0.183,"rank":34.37,"hrank":23.03,"imp":17.269,"train_time":8274.67,"predict_time":40.268,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+55\/-52","imputed":false},{"position":33,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1302.0,"score":0.244,"rank":35.52,"hrank":17.6,"imp":19.116,"train_time":2916.21,"predict_time":17.907,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+87\/-80","imputed":true},{"position":34,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1301.0,"score":0.148,"rank":35.57,"hrank":25.48,"imp":18.459,"train_time":976.68,"predict_time":0.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-58","imputed":false},{"position":35,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1293.0,"score":0.134,"rank":36.27,"hrank":25.49,"imp":17.915,"train_time":9.04,"predict_time":1.162,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-53","imputed":false},{"position":36,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1293.0,"score":0.193,"rank":36.3,"hrank":26.49,"imp":17.435,"train_time":559.77,"predict_time":1.513,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-83","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1291.0,"score":0.183,"rank":36.45,"hrank":23.87,"imp":18.056,"train_time":5.88,"predict_time":0.202,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-75","imputed":false},{"position":38,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1291.0,"score":0.216,"rank":36.45,"hrank":15.63,"imp":19.295,"train_time":106.36,"predict_time":2.667,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+100\/-76","imputed":true},{"position":39,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1287.0,"score":0.177,"rank":36.8,"hrank":26.57,"imp":18.483,"train_time":1.52,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-72","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1274.0,"score":0.155,"rank":38.0,"hrank":29.21,"imp":17.913,"train_time":2206.64,"predict_time":2.301,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-51","imputed":false},{"position":41,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1256.0,"score":0.134,"rank":39.63,"hrank":30.34,"imp":19.008,"train_time":559.77,"predict_time":0.086,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-69","imputed":false},{"position":42,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1254.0,"score":0.177,"rank":39.75,"hrank":21.71,"imp":20.622,"train_time":2916.21,"predict_time":0.225,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+83\/-66","imputed":true},{"position":43,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1248.0,"score":0.089,"rank":40.3,"hrank":28.61,"imp":19.285,"train_time":976.68,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-54","imputed":false},{"position":44,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1248.0,"score":0.12,"rank":40.32,"hrank":16.3,"imp":19.877,"train_time":2.84,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-59","imputed":false},{"position":45,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1240.0,"score":0.152,"rank":41.05,"hrank":24.9,"imp":21.003,"train_time":13.07,"predict_time":3.169,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-77","imputed":false},{"position":46,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1219.0,"score":0.134,"rank":42.92,"hrank":26.13,"imp":19.959,"train_time":47.89,"predict_time":39.444,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-87","imputed":false},{"position":47,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1213.0,"score":0.148,"rank":43.4,"hrank":17.85,"imp":21.306,"train_time":2.61,"predict_time":0.256,"imputed_pct":33.33,"hardware":"GPU","elo_ci":"+76\/-76","imputed":true},{"position":48,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1213.0,"score":0.111,"rank":43.42,"hrank":32.22,"imp":20.466,"train_time":561.29,"predict_time":4.462,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-93","imputed":false},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1202.0,"score":0.133,"rank":44.37,"hrank":16.87,"imp":20.132,"train_time":1.42,"predict_time":0.118,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-76","imputed":false},{"position":50,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1200.0,"score":0.104,"rank":44.58,"hrank":15.21,"imp":21.067,"train_time":11.15,"predict_time":0.312,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-62","imputed":false},{"position":51,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1198.0,"score":0.11,"rank":44.75,"hrank":36.52,"imp":20.229,"train_time":2206.64,"predict_time":0.107,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-66","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1168.0,"score":0.065,"rank":47.35,"hrank":40.23,"imp":21.275,"train_time":180.03,"predict_time":0.779,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-65","imputed":false},{"position":53,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1165.0,"score":0.099,"rank":47.62,"hrank":24.86,"imp":22.833,"train_time":202.68,"predict_time":44.667,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-77","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1145.0,"score":0.06,"rank":49.33,"hrank":38.78,"imp":22.501,"train_time":180.03,"predict_time":0.056,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-90","imputed":false},{"position":55,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1142.0,"score":0.141,"rank":49.52,"hrank":15.84,"imp":24.202,"train_time":34925.3,"predict_time":3.663,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-106","imputed":false},{"position":56,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1141.0,"score":0.069,"rank":49.62,"hrank":42.87,"imp":20.95,"train_time":0.93,"predict_time":0.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-66","imputed":false},{"position":57,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1141.0,"score":0.074,"rank":49.63,"hrank":33.71,"imp":22.47,"train_time":561.29,"predict_time":0.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-92","imputed":false},{"position":58,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1140.0,"score":0.052,"rank":49.73,"hrank":43.45,"imp":22.667,"train_time":171.61,"predict_time":0.552,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-78","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1133.0,"score":0.134,"rank":50.33,"hrank":17.61,"imp":24.391,"train_time":34925.3,"predict_time":16.932,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+99\/-115","imputed":false},{"position":60,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1108.0,"score":0.013,"rank":52.37,"hrank":48.64,"imp":23.243,"train_time":171.61,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-65","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1106.0,"score":0.096,"rank":52.47,"hrank":40.26,"imp":25.325,"train_time":130.48,"predict_time":0.498,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-105","imputed":false},{"position":62,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1086.0,"score":0.045,"rank":54.05,"hrank":45.56,"imp":23.983,"train_time":14.37,"predict_time":2.74,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-85","imputed":false},{"position":63,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1059.0,"score":0.111,"rank":56.13,"hrank":15.13,"imp":29.368,"train_time":387.37,"predict_time":4.09,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-142","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1056.0,"score":0.051,"rank":56.33,"hrank":50.47,"imp":26.271,"train_time":130.48,"predict_time":0.143,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-91","imputed":false},{"position":65,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1051.0,"score":0.026,"rank":56.73,"hrank":49.97,"imp":24.57,"train_time":4.97,"predict_time":0.092,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-72","imputed":false},{"position":66,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1019.0,"score":0.027,"rank":59.0,"hrank":54.01,"imp":25.143,"train_time":2.79,"predict_time":0.259,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-91","imputed":false},{"position":67,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1009.0,"score":0.046,"rank":59.67,"hrank":48.95,"imp":29.565,"train_time":0.81,"predict_time":0.115,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+99\/-114","imputed":false},{"position":68,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.02,"rank":60.27,"hrank":48.7,"imp":27.283,"train_time":0.29,"predict_time":0.031,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-94","imputed":false},{"position":69,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":996.0,"score":0.04,"rank":60.53,"hrank":49.39,"imp":27.074,"train_time":1.15,"predict_time":0.22,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-111","imputed":false},{"position":70,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":993.0,"score":0.007,"rank":60.72,"hrank":58.1,"imp":27.035,"train_time":140.69,"predict_time":3.209,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-81","imputed":false},{"position":71,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":979.0,"score":0.04,"rank":61.63,"hrank":20.25,"imp":30.562,"train_time":155.53,"predict_time":0.277,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-121","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":977.0,"score":0.041,"rank":61.73,"hrank":36.32,"imp":32.015,"train_time":12.44,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-126","imputed":false},{"position":73,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":956.0,"score":0.029,"rank":63.02,"hrank":31.1,"imp":31.169,"train_time":155.53,"predict_time":0.05,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-124","imputed":false},{"position":74,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":937.0,"score":0.03,"rank":64.07,"hrank":57.21,"imp":30.471,"train_time":2.71,"predict_time":0.392,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-105","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":915.0,"score":0.009,"rank":65.27,"hrank":60.47,"imp":32.833,"train_time":0.8,"predict_time":0.088,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-99","imputed":false},{"position":76,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":845.0,"score":0.0,"rank":68.57,"hrank":67.79,"imp":33.03,"train_time":140.69,"predict_time":0.188,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-80","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":609.0,"score":0.0,"rank":75.27,"hrank":74.95,"imp":45.764,"train_time":0.15,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-158","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/pareto_front_explorer.html
index 887c15f8b75029e100c7c9833ecf8e6ccd276f5d..e4e07c314ebc7779aa6688648cde937dab393d5b 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.6993924491,"imp":18.2218497556,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0209339609,"imp":15.9779788966,"elo":1367.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0271393798,"imp":15.8159745017,"elo":1376.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3007363462,"imp":16.0702318848,"elo":1368.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0303915587,"imp":18.4828200758,"elo":1287.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0257224855,"imp":17.6997719652,"elo":1334.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3055895504,"imp":17.0701459198,"elo":1367.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0149256706,"imp":19.8774931383,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0149447918,"imp":19.2845150837,"elo":1248.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1399325633,"imp":18.4586133843,"elo":1301.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2195070167,"imp":27.0739588045,"elo":995.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0555212696,"imp":22.5006991233,"elo":1144.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7789141111,"imp":21.2746888276,"elo":1168.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2592580681,"imp":25.1425500853,"elo":1018.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2364230333,"imp":22.470090687,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4616886326,"imp":20.4661706065,"elo":1213.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.094055721,"imp":20.9498074447,"elo":1141.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0890724278,"imp":17.5271029046,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7734248689,"imp":16.4047314294,"elo":1375.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.3036225459,"imp":16.8194767689,"elo":1374.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.6670530253,"imp":19.2946547315,"elo":1291.1,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3117928723,"imp":21.0669867384,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4051957243,"imp":16.5506564353,"elo":1360.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4763406294,"imp":16.7056605028,"elo":1341.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7399372917,"imp":23.982542653,"elo":1086.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0268138653,"imp":32.0145424844,"elo":977.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1426565145,"imp":26.271001098,"elo":1056.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4978820663,"imp":25.3250881541,"elo":1106.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0312580102,"imp":27.2834817371,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0526216987,"imp":23.2427649093,"elo":1107.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5516567504,"imp":22.6670872677,"elo":1139.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.1617619189,"imp":17.9146222582,"elo":1293.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.2645206496,"imp":15.3680033109,"elo":1416.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":9.3100869738,"imp":13.6912952333,"elo":1486.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7081268775,"imp":12.6489007703,"elo":1537.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0352207543,"imp":13.0760951992,"elo":1515.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.7444053479,"imp":12.3299578304,"elo":1546.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.1690613568,"imp":21.002766606,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"imp":19.9588241591,"elo":1218.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.5640473457,"imp":17.0227783378,"elo":1329.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":409.9318208736,"imp":15.1807845083,"elo":1366.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1668648195,"imp":15.6622109692,"elo":1392.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"imp":5.1511046509,"elo":1774.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.115458016,"imp":29.565351036,"elo":1009.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":2.0066462138,"imp":15.1274433273,"elo":1384.7,"imputed":true,"imputed_pct":3.3333333333,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1600398508,"imp":11.6560007761,"elo":1577.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2024528313,"imp":18.0557224187,"elo":1291.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.218906887,"imp":16.5055297598,"elo":1362.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.6019358162,"imp":15.5150758039,"elo":1406.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4673275113,"imp":11.897805388,"elo":1583.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.353062543,"imp":10.2983957644,"elo":1632.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2562998879,"imp":21.3062268554,"elo":1213.3,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2246817989,"imp":20.6218769339,"elo":1254.2,"imputed":true,"imputed_pct":33.3333333333,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.9065990706,"imp":19.1155261256,"elo":1301.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0895012297,"imp":29.3675795038,"elo":1059.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6626595929,"imp":24.2024990954,"elo":1142.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.9317836231,"imp":24.3905591101,"elo":1132.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622971538,"imp":16.5083848914,"elo":1354.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0923536337,"imp":24.5695821262,"elo":1050.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1067953784,"imp":20.2292654487,"elo":1198.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.3011998096,"imp":17.9131586959,"elo":1273.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1176002559,"imp":20.1318460972,"elo":1202.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1079666958,"imp":17.2225951889,"elo":1328.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6132840031,"imp":16.8960348336,"elo":1362.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":44.6673844457,"imp":22.8325807189,"elo":1164.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":40.2677406188,"imp":17.2689825081,"elo":1314.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":361.1026011106,"imp":15.5000705017,"elo":1407.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.3922735327,"imp":30.4706736225,"elo":937.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0860477633,"imp":19.0084894359,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.5127555025,"imp":17.4351077716,"elo":1292.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.6993924491,"x_train":248.5156197629,"imp":18.2218497556,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0209339609,"x_train":3.1841367928,"imp":15.9779788966,"elo":1367.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0271393798,"x_train":735.4606570949,"imp":15.8159745017,"elo":1376.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3007363462,"x_train":735.4606570949,"imp":16.0702318848,"elo":1368.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0303915587,"x_train":1.5249213808,"imp":18.4828200758,"elo":1287.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0257224855,"x_train":275.9265613682,"imp":17.6997719652,"elo":1334.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3055895504,"x_train":275.9265613682,"imp":17.0701459198,"elo":1367.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0149256706,"x_train":2.8412925593,"imp":19.8774931383,"elo":1247.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0149447918,"x_train":976.6816362244,"imp":19.2845150837,"elo":1248.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1399325633,"x_train":976.6816362244,"imp":18.4586133843,"elo":1301.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2195070167,"x_train":1.1488986663,"imp":27.0739588045,"elo":995.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0555212696,"x_train":180.0329734893,"imp":22.5006991233,"elo":1144.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7789141111,"x_train":180.0329734893,"imp":21.2746888276,"elo":1168.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2592580681,"x_train":2.7868984981,"imp":25.1425500853,"elo":1018.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2364230333,"x_train":561.2940883157,"imp":22.470090687,"elo":1141.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.4616886326,"x_train":561.2940883157,"imp":20.4661706065,"elo":1213.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.094055721,"x_train":0.9329196027,"imp":20.9498074447,"elo":1141.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0890724278,"x_train":328.641625286,"imp":17.5271029046,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7734248689,"x_train":328.641625286,"imp":16.4047314294,"elo":1375.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.3036225459,"x_train":14.2669009183,"imp":16.8194767689,"elo":1374.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.6670530253,"x_train":106.3572012411,"imp":19.2946547315,"elo":1291.1,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3117928723,"x_train":11.1478855018,"imp":21.0669867384,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4051957243,"x_train":3436.7433273226,"imp":16.5506564353,"elo":1360.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4763406294,"x_train":3436.7433273226,"imp":16.7056605028,"elo":1341.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7399372917,"x_train":14.3691534377,"imp":23.982542653,"elo":1086.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0268138653,"x_train":12.4370351532,"imp":32.0145424844,"elo":977.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1426565145,"x_train":130.4829960164,"imp":26.271001098,"elo":1056.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4978820663,"x_train":130.4829960164,"imp":25.3250881541,"elo":1106.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0312580102,"x_train":0.2940996202,"imp":27.2834817371,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0526216987,"x_train":171.6065412889,"imp":23.2427649093,"elo":1107.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5516567504,"x_train":171.6065412889,"imp":22.6670872677,"elo":1139.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.1617619189,"x_train":9.0446607596,"imp":17.9146222582,"elo":1293.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.2645206496,"x_train":2722.1286706272,"imp":15.3680033109,"elo":1416.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":9.3100869738,"x_train":2722.1286706272,"imp":13.6912952333,"elo":1486.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7081268775,"x_train":4.7880152698,"imp":12.6489007703,"elo":1537.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0352207543,"x_train":1878.3590992619,"imp":13.0760951992,"elo":1515.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.7444053479,"x_train":1878.3590992619,"imp":12.3299578304,"elo":1546.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.1690613568,"x_train":13.0741284928,"imp":21.002766606,"elo":1239.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"x_train":47.8871994,"imp":19.9588241591,"elo":1218.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.5640473457,"x_train":5891.3014300686,"imp":17.0227783378,"elo":1329.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":409.9318208736,"x_train":5891.3014300686,"imp":15.1807845083,"elo":1366.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1668648195,"x_train":1.4816805767,"imp":15.6622109692,"elo":1392.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"x_train":37.5608528654,"imp":5.1511046509,"elo":1774.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.115458016,"x_train":0.8137451985,"imp":29.565351036,"elo":1009.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":2.0066462138,"x_train":8.0513022526,"imp":15.1274433273,"elo":1384.7,"imputed":true,"imputed_pct":3.3333333333,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1600398508,"x_train":1.7694169027,"imp":11.6560007761,"elo":1577.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2024528313,"x_train":5.8795969926,"imp":18.0557224187,"elo":1291.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.218906887,"x_train":1641.7401040643,"imp":16.5055297598,"elo":1362.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.6019358162,"x_train":1641.7401040643,"imp":15.5150758039,"elo":1406.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4673275113,"x_train":4.8183344075,"imp":11.897805388,"elo":1583.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.353062543,"x_train":2.9028568606,"imp":10.2983957644,"elo":1632.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2562998879,"x_train":2.6092164782,"imp":21.3062268554,"elo":1213.3,"imputed":true,"imputed_pct":33.3333333333,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2246817989,"x_train":2916.211932309,"imp":20.6218769339,"elo":1254.2,"imputed":true,"imputed_pct":33.3333333333,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.9065990706,"x_train":2916.211932309,"imp":19.1155261256,"elo":1301.5,"imputed":true,"imputed_pct":33.3333333333,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0895012297,"x_train":387.3713684902,"imp":29.3675795038,"elo":1059.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6626595929,"x_train":34925.3007391894,"imp":24.2024990954,"elo":1142.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.9317836231,"x_train":34925.3007391894,"imp":24.3905591101,"elo":1132.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622971538,"x_train":1.1067907259,"imp":16.5083848914,"elo":1354.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0923536337,"x_train":4.9663458274,"imp":24.5695821262,"elo":1050.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1067953784,"x_train":2206.6390156442,"imp":20.2292654487,"elo":1198.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.3011998096,"x_train":2206.6390156442,"imp":17.9131586959,"elo":1273.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1176002559,"x_train":1.4234367091,"imp":20.1318460972,"elo":1202.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1079666958,"x_train":463.012192935,"imp":17.2225951889,"elo":1328.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6132840031,"x_train":463.012192935,"imp":16.8960348336,"elo":1362.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":44.6673844457,"x_train":202.6812316751,"imp":22.8325807189,"elo":1164.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":40.2677406188,"x_train":8274.6664503747,"imp":17.2689825081,"elo":1314.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":361.1026011106,"x_train":8274.6664503747,"imp":15.5000705017,"elo":1407.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.3922735327,"x_train":2.7144283593,"imp":30.4706736225,"elo":937.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0860477633,"x_train":559.7696517925,"imp":19.0084894359,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.5127555025,"x_train":559.7696517925,"imp":17.4351077716,"elo":1292.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/tuning_trajectories_explorer.html
index e353a707dce72ca656e5af49194ce55e37eae73d..725cc8358fe3b7874b60ef1183beb951036b017f 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_binary/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":248.5156197629,"imp":18.2677043989,"elo":1350.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.1841367928,"imp":16.0178183653,"elo":1390.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":7.6931563125,"imp":15.987956061,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":18.2301485936,"imp":15.9531640847,"elo":1394.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":36.373457126,"imp":15.9468063623,"elo":1398.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":93.5688559171,"imp":16.0074444249,"elo":1400.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":183.1491958089,"imp":16.0324997291,"elo":1396.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":366.895128245,"imp":16.0600273697,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":546.7577277937,"imp":16.1283708816,"elo":1385.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":735.4606570949,"imp":16.092828704,"elo":1389.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.5249213808,"imp":18.5272765343,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.9231596776,"imp":17.8485305735,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":6.9435649233,"imp":17.435293103,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":13.7562217845,"imp":17.2721315724,"elo":1382.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":34.4671973421,"imp":17.1093824899,"elo":1393.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":68.8507318629,"imp":16.9897257409,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":137.5904273763,"imp":17.0254191543,"elo":1397.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":206.0245082163,"imp":17.0332899257,"elo":1398.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":275.9265613682,"imp":17.1051813975,"elo":1395.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.8412925593,"imp":19.9145027791,"elo":1258.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":8.2691673892,"imp":19.4002332647,"elo":1281.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":21.1083282561,"imp":18.9945473229,"elo":1298.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":45.3683328739,"imp":18.8557700351,"elo":1303.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":119.8161001058,"imp":18.7172958877,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":243.8194356466,"imp":18.5795740654,"elo":1321.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":486.6455896663,"imp":18.4917172729,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":730.1621818496,"imp":18.4984983838,"elo":1324.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":976.6816362244,"imp":18.4858814608,"elo":1323.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.1488986663,"imp":27.1040814061,"elo":1008.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.0419356023,"imp":25.4385513423,"elo":1038.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":4.7437750172,"imp":22.9698736592,"elo":1099.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.1794740447,"imp":22.1045006433,"elo":1130.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":22.6140856772,"imp":21.6965636808,"elo":1147.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":44.8789786871,"imp":21.4218170868,"elo":1160.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":89.703830225,"imp":21.2101789491,"elo":1172.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":134.3807181269,"imp":21.1038018687,"elo":1180.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":180.0329734893,"imp":21.3504159422,"elo":1178.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.7868984981,"imp":25.1710224144,"elo":1029.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.47397211,"imp":24.108151502,"elo":1086.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":14.1482286155,"imp":22.3537231105,"elo":1153.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":28.5676512334,"imp":21.6150311381,"elo":1178.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":68.1068582413,"imp":20.8519784371,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":137.1828565728,"imp":20.5764844812,"elo":1211.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":277.9218393643,"imp":20.4695306001,"elo":1214.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":418.1539061688,"imp":20.4079072987,"elo":1217.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":561.2940883157,"imp":20.5125265792,"elo":1224.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.9329196027,"imp":20.9905024372,"elo":1149.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":2.8974631618,"imp":18.717252783,"elo":1261.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":7.4728798139,"imp":17.3595244812,"elo":1343.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":15.0938578715,"imp":16.8301057051,"elo":1374.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":39.6915504803,"imp":16.5324825538,"elo":1391.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":80.8366710185,"imp":16.4624284574,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":163.2701936469,"imp":16.4200557594,"elo":1400.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":243.2687421878,"imp":16.3655590811,"elo":1402.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":328.641625286,"imp":16.444256379,"elo":1401.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":14.2669009183,"imp":17.0167171748,"elo":1389.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":106.3572012411,"imp":19.3289663591,"elo":1308.6,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":11.1478855018,"imp":21.1029724976,"elo":1208.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":27.3629676777,"imp":19.6968265883,"elo":1243.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":84.0873327564,"imp":17.3088764222,"elo":1342.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":169.5069286032,"imp":16.8266087961,"elo":1363.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":413.8881530141,"imp":16.3056234208,"elo":1391.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":856.7378065106,"imp":16.4512831666,"elo":1381.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1708.0482807143,"imp":16.7773139814,"elo":1368.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":2544.0473436332,"imp":16.8183196783,"elo":1366.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":3436.7433273226,"imp":16.7728736116,"elo":1366.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":14.3691534377,"imp":23.8149294637,"elo":1107.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":12.4370351532,"imp":32.0502146336,"elo":991.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":39.6579418967,"imp":29.3371060358,"elo":1043.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":109.2939161538,"imp":25.5465503401,"elo":1116.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":130.4829960164,"imp":25.3286594254,"elo":1120.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.2940996202,"imp":27.3176315227,"elo":1000.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":1.108342455,"imp":25.0269678712,"elo":1057.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":3.7342981222,"imp":23.9344191638,"elo":1094.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":8.0549426531,"imp":23.4579126288,"elo":1113.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":20.7189526054,"imp":22.9726526352,"elo":1136.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":41.766842473,"imp":22.8599782966,"elo":1144.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":84.379284379,"imp":22.7747309581,"elo":1148.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":127.399603328,"imp":22.7599101855,"elo":1147.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":171.6065412889,"imp":22.7006117634,"elo":1150.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":9.0446607596,"imp":17.9459120344,"elo":1303.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":23.323034735,"imp":16.227921693,"elo":1390.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":62.8372296721,"imp":14.9411124084,"elo":1459.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":131.5115032503,"imp":14.557801128,"elo":1471.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":336.6014034599,"imp":14.0927367932,"elo":1494.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":679.3497815723,"imp":13.8846309228,"elo":1505.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1355.7488156461,"imp":13.8318473136,"elo":1505.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2029.6864482209,"imp":13.7843124657,"elo":1515.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2722.1286706272,"imp":13.7391598332,"elo":1519.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":4.7880152698,"imp":12.6970229414,"elo":1551.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.4527651431,"imp":12.7654155819,"elo":1550.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":44.0076788597,"imp":12.7488030637,"elo":1543.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":89.9272230443,"imp":12.7842798182,"elo":1544.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":233.2858308743,"imp":12.7177636975,"elo":1547.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":464.3616718274,"imp":12.8344913092,"elo":1541.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":937.3819334882,"imp":12.5476762639,"elo":1554.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1406.5304544301,"imp":12.4887662192,"elo":1563.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1878.3590992619,"imp":12.4069726077,"elo":1564.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.0741284928,"imp":21.0358242144,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":47.8871994,"imp":19.9887390237,"elo":1229.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":76.8228235044,"imp":18.6349645959,"elo":1273.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":154.1734218688,"imp":17.4871836502,"elo":1325.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":288.3811437278,"imp":17.0625132855,"elo":1339.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":734.9046229398,"imp":16.2535972245,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1475.2401534104,"imp":15.6783377602,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2935.4601233793,"imp":15.4094123414,"elo":1383.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":4395.9302046427,"imp":15.2860293711,"elo":1385.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":5891.3014300686,"imp":15.2421217005,"elo":1389.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":1.4816805767,"imp":15.6981760746,"elo":1410.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":37.5608528654,"imp":5.195074238,"elo":1797.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.8137451985,"imp":29.588055863,"elo":1027.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":8.0513022526,"imp":15.1707582392,"elo":1403.9,"imputed":true,"imputed_pct":3.3333333333,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.7694169027,"imp":11.7029291028,"elo":1599.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":5.8795969926,"imp":18.0944856252,"elo":1311.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":13.3369957806,"imp":17.7749861989,"elo":1322.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":38.7568509497,"imp":17.1717274262,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":78.6946304538,"imp":16.5979496827,"elo":1382.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":201.4443768551,"imp":16.1258714479,"elo":1403.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":405.7219066472,"imp":15.8553978863,"elo":1418.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":807.6777264234,"imp":15.6543662568,"elo":1426.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1220.4156808351,"imp":15.5901216306,"elo":1434.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1641.7401040643,"imp":15.561933124,"elo":1435.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":4.8183344075,"imp":11.9424971927,"elo":1611.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":2.9028568606,"imp":10.2772421387,"elo":1649.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2.6092164782,"imp":21.3395688074,"elo":1222.3,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":22.8672371568,"imp":20.6977308651,"elo":1251.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":62.9137172928,"imp":20.0800780954,"elo":1282.9,"imputed":true,"imputed_pct":33.3333333333,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":140.2609154123,"imp":19.6589836045,"elo":1298.4,"imputed":true,"imputed_pct":33.3333333333,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":347.4920082066,"imp":19.1890847234,"elo":1320.5,"imputed":true,"imputed_pct":33.3333333333,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":706.1972040222,"imp":19.015710344,"elo":1331.1,"imputed":true,"imputed_pct":33.3333333333,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1438.110967546,"imp":19.0008011295,"elo":1330.1,"imputed":true,"imputed_pct":33.3333333333,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2168.8573846957,"imp":19.0922957383,"elo":1324.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2916.211932309,"imp":19.1779850452,"elo":1313.5,"imputed":true,"imputed_pct":33.3333333333,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":387.3713684902,"imp":29.3875976863,"elo":1073.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":619.0654872096,"imp":29.2699684092,"elo":1080.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1126.9710593518,"imp":28.9452903243,"elo":1085.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1799.6745277943,"imp":28.250897063,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4218.8944407454,"imp":26.9033265918,"elo":1111.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":8667.7234636881,"imp":25.6818413182,"elo":1125.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":17722.5911655635,"imp":25.1929972913,"elo":1133.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":26217.887217876,"imp":24.793983631,"elo":1143.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34925.3007391894,"imp":24.4242199422,"elo":1155.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1067907259,"imp":16.554354774,"elo":1378.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":4.9663458274,"imp":24.5854535013,"elo":1062.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.872258751,"imp":22.0843362149,"elo":1132.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":43.9083852063,"imp":20.0317547645,"elo":1187.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":107.4432911334,"imp":19.2360452664,"elo":1226.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":267.8896031045,"imp":18.5859012039,"elo":1261.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":552.8424768764,"imp":18.2627562112,"elo":1277.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1101.0843510949,"imp":18.1797163399,"elo":1278.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1649.8652006608,"imp":18.032575652,"elo":1288.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2206.6390156442,"imp":17.9473130882,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.4234367091,"imp":20.1741412388,"elo":1214.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":4.1993942212,"imp":18.2089294006,"elo":1311.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":11.4672731767,"imp":17.6658989971,"elo":1344.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":22.2498464852,"imp":17.3593397325,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":57.7997562863,"imp":17.1192287025,"elo":1370.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":114.5664011196,"imp":17.0644712234,"elo":1378.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":229.0218261459,"imp":16.9859238323,"elo":1385.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":343.7649500386,"imp":16.9058672707,"elo":1389.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":463.012192935,"imp":16.9294193488,"elo":1389.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":202.6812316751,"imp":22.8658447838,"elo":1180.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":512.7516159338,"imp":20.3009736911,"elo":1251.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":1466.1655464244,"imp":17.472040179,"elo":1337.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":3194.8705446357,"imp":16.4000617135,"elo":1384.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":7908.2276811319,"imp":15.5054242237,"elo":1430.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":8274.6664503747,"imp":15.5539976857,"elo":1430.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.7144283593,"imp":30.4948770081,"elo":953.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.4835528309,"imp":25.913159969,"elo":1040.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":13.8167338689,"imp":20.5297244826,"elo":1189.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":27.7848683857,"imp":18.4868051449,"elo":1278.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":69.6362065708,"imp":17.9776134007,"elo":1302.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":138.9025571316,"imp":17.7572096471,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":278.1866797416,"imp":17.5939282366,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":417.6393154275,"imp":17.5124840659,"elo":1309.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":559.7696517925,"imp":17.3620070536,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.6993924491,"x_train":248.5156197629,"imp":18.2677043989,"elo":1350.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0209339609,"x_train":3.1841367928,"imp":16.0178183653,"elo":1390.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0431790436,"x_train":7.6931563125,"imp":15.987956061,"elo":1391.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0794519096,"x_train":18.2301485936,"imp":15.9531640847,"elo":1394.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1149153864,"x_train":36.373457126,"imp":15.9468063623,"elo":1398.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1611888807,"x_train":93.5688559171,"imp":16.0074444249,"elo":1400.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1821892961,"x_train":183.1491958089,"imp":16.0324997291,"elo":1396.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2211955566,"x_train":366.895128245,"imp":16.0600273697,"elo":1392.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2469426441,"x_train":546.7577277937,"imp":16.1283708816,"elo":1385.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3090917862,"x_train":735.4606570949,"imp":16.092828704,"elo":1389.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0303915587,"x_train":1.5249213808,"imp":18.5272765343,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0624141331,"x_train":2.9231596776,"imp":17.8485305735,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1045674981,"x_train":6.9435649233,"imp":17.435293103,"elo":1375.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.167681727,"x_train":13.7562217845,"imp":17.2721315724,"elo":1382.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2384325101,"x_train":34.4671973421,"imp":17.1093824899,"elo":1393.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2910163995,"x_train":68.8507318629,"imp":16.9897257409,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3160449363,"x_train":137.5904273763,"imp":17.0254191543,"elo":1397.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3090326441,"x_train":206.0245082163,"imp":17.0332899257,"elo":1398.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3055895504,"x_train":275.9265613682,"imp":17.1051813975,"elo":1395.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0149256706,"x_train":2.8412925593,"imp":19.9145027791,"elo":1258.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0294172839,"x_train":8.2691673892,"imp":19.4002332647,"elo":1281.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0524661348,"x_train":21.1083282561,"imp":18.9945473229,"elo":1298.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0721220241,"x_train":45.3683328739,"imp":18.8557700351,"elo":1303.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0973613753,"x_train":119.8161001058,"imp":18.7172958877,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1172786529,"x_train":243.8194356466,"imp":18.5795740654,"elo":1321.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1285427493,"x_train":486.6455896663,"imp":18.4917172729,"elo":1323.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1300971595,"x_train":730.1621818496,"imp":18.4984983838,"elo":1324.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1415209075,"x_train":976.6816362244,"imp":18.4858814608,"elo":1323.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2195070167,"x_train":1.1488986663,"imp":27.1040814061,"elo":1008.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2579008861,"x_train":2.0419356023,"imp":25.4385513423,"elo":1038.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3206822077,"x_train":4.7437750172,"imp":22.9698736592,"elo":1099.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3866075277,"x_train":9.1794740447,"imp":22.1045006433,"elo":1130.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.496862742,"x_train":22.6140856772,"imp":21.6965636808,"elo":1147.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6116316629,"x_train":44.8789786871,"imp":21.4218170868,"elo":1160.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6264150113,"x_train":89.703830225,"imp":21.2101789491,"elo":1172.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6944098681,"x_train":134.3807181269,"imp":21.1038018687,"elo":1180.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7789141111,"x_train":180.0329734893,"imp":21.3504159422,"elo":1178.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2592580681,"x_train":2.7868984981,"imp":25.1710224144,"elo":1029.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.470732767,"x_train":5.47397211,"imp":24.108151502,"elo":1086.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.0377657426,"x_train":14.1482286155,"imp":22.3537231105,"elo":1153.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.5322826304,"x_train":28.5676512334,"imp":21.6150311381,"elo":1178.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.246588778,"x_train":68.1068582413,"imp":20.8519784371,"elo":1198.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.9237399683,"x_train":137.1828565728,"imp":20.5764844812,"elo":1211.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.6333332925,"x_train":277.9218393643,"imp":20.4695306001,"elo":1214.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.0459379226,"x_train":418.1539061688,"imp":20.4079072987,"elo":1217.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.5532282308,"x_train":561.2940883157,"imp":20.5125265792,"elo":1224.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.094055721,"x_train":0.9329196027,"imp":20.9905024372,"elo":1149.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4273281397,"x_train":2.8974631618,"imp":18.717252783,"elo":1261.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7023827628,"x_train":7.4728798139,"imp":17.3595244812,"elo":1343.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9043924684,"x_train":15.0938578715,"imp":16.8301057051,"elo":1374.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.098568899,"x_train":39.6915504803,"imp":16.5324825538,"elo":1391.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.147173026,"x_train":80.8366710185,"imp":16.4624284574,"elo":1397.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.084673009,"x_train":163.2701936469,"imp":16.4200557594,"elo":1400.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0402490715,"x_train":243.2687421878,"imp":16.3655590811,"elo":1402.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0244823629,"x_train":328.641625286,"imp":16.444256379,"elo":1401.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.3036225459,"x_train":14.2669009183,"imp":17.0167171748,"elo":1389.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.6670530253,"x_train":106.3572012411,"imp":19.3289663591,"elo":1308.6,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3117928723,"x_train":11.1478855018,"imp":21.1029724976,"elo":1208.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8385654835,"x_train":27.3629676777,"imp":19.6968265883,"elo":1243.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7997608403,"x_train":84.0873327564,"imp":17.3088764222,"elo":1342.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9001397479,"x_train":169.5069286032,"imp":16.8266087961,"elo":1363.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.0611635517,"x_train":413.8881530141,"imp":16.3056234208,"elo":1391.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1854995844,"x_train":856.7378065106,"imp":16.4512831666,"elo":1381.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.5251211404,"x_train":1708.0482807143,"imp":16.7773139814,"elo":1368.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8121612888,"x_train":2544.0473436332,"imp":16.8183196783,"elo":1366.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4763406294,"x_train":3436.7433273226,"imp":16.7728736116,"elo":1366.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.7399372917,"x_train":14.3691534377,"imp":23.8149294637,"elo":1107.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0268138653,"x_train":12.4370351532,"imp":32.0502146336,"elo":991.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2097117411,"x_train":39.6579418967,"imp":29.3371060358,"elo":1043.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4237356511,"x_train":109.2939161538,"imp":25.5465503401,"elo":1116.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4978820663,"x_train":130.4829960164,"imp":25.3286594254,"elo":1120.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0312580102,"x_train":0.2940996202,"imp":27.3176315227,"elo":1000.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0906042357,"x_train":1.108342455,"imp":25.0269678712,"elo":1057.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1840220874,"x_train":3.7342981222,"imp":23.9344191638,"elo":1094.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2757254745,"x_train":8.0549426531,"imp":23.4579126288,"elo":1113.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3625016031,"x_train":20.7189526054,"imp":22.9726526352,"elo":1136.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4725802008,"x_train":41.766842473,"imp":22.8599782966,"elo":1144.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5376208643,"x_train":84.379284379,"imp":22.7747309581,"elo":1148.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5574391287,"x_train":127.399603328,"imp":22.7599101855,"elo":1147.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5606060557,"x_train":171.6065412889,"imp":22.7006117634,"elo":1150.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1617619189,"x_train":9.0446607596,"imp":17.9459120344,"elo":1303.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7002960265,"x_train":23.323034735,"imp":16.227921693,"elo":1390.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.6925437923,"x_train":62.8372296721,"imp":14.9411124084,"elo":1459.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.2033439648,"x_train":131.5115032503,"imp":14.557801128,"elo":1471.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.9399740849,"x_train":336.6014034599,"imp":14.0927367932,"elo":1494.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.9896970288,"x_train":679.3497815723,"imp":13.8846309228,"elo":1505.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.8216592508,"x_train":1355.7488156461,"imp":13.8318473136,"elo":1505.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.4687086151,"x_train":2029.6864482209,"imp":13.7843124657,"elo":1515.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.3100869738,"x_train":2722.1286706272,"imp":13.7391598332,"elo":1519.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7081268775,"x_train":4.7880152698,"imp":12.6970229414,"elo":1551.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.6939241991,"x_train":16.4527651431,"imp":12.7654155819,"elo":1550.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.1722189812,"x_train":44.0076788597,"imp":12.7488030637,"elo":1543.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.2123187912,"x_train":89.9272230443,"imp":12.7842798182,"elo":1544.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.1074574709,"x_train":233.2858308743,"imp":12.7177636975,"elo":1547.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3342760367,"x_train":464.3616718274,"imp":12.8344913092,"elo":1541.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.530318776,"x_train":937.3819334882,"imp":12.5476762639,"elo":1554.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.3639010218,"x_train":1406.5304544301,"imp":12.4887662192,"elo":1563.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":10.8083515158,"x_train":1878.3590992619,"imp":12.4069726077,"elo":1564.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":3.1690613568,"x_train":13.0741284928,"imp":21.0358242144,"elo":1252.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"x_train":47.8871994,"imp":19.9887390237,"elo":1229.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":57.0488035322,"x_train":76.8228235044,"imp":18.6349645959,"elo":1273.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":89.3976769749,"x_train":154.1734218688,"imp":17.4871836502,"elo":1325.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":120.7151402618,"x_train":288.3811437278,"imp":17.0625132855,"elo":1339.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":191.9678402321,"x_train":734.9046229398,"imp":16.2535972245,"elo":1356.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":255.2028837353,"x_train":1475.2401534104,"imp":15.6783377602,"elo":1373.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":327.3661339693,"x_train":2935.4601233793,"imp":15.4094123414,"elo":1383.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":361.0134335556,"x_train":4395.9302046427,"imp":15.2860293711,"elo":1385.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":410.7225790808,"x_train":5891.3014300686,"imp":15.2421217005,"elo":1389.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1668648195,"x_train":1.4816805767,"imp":15.6981760746,"elo":1410.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"x_train":37.5608528654,"imp":5.195074238,"elo":1797.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.115458016,"x_train":0.8137451985,"imp":29.588055863,"elo":1027.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":2.0066462138,"x_train":8.0513022526,"imp":15.1707582392,"elo":1403.9,"imputed":true,"imputed_pct":3.3333333333,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1600398508,"x_train":1.7694169027,"imp":11.7029291028,"elo":1599.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2024528313,"x_train":5.8795969926,"imp":18.0944856252,"elo":1311.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.3732754571,"x_train":13.3369957806,"imp":17.7749861989,"elo":1322.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.5461575366,"x_train":38.7568509497,"imp":17.1717274262,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.731945022,"x_train":78.6946304538,"imp":16.5979496827,"elo":1382.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.0226182454,"x_train":201.4443768551,"imp":16.1258714479,"elo":1403.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.1935240832,"x_train":405.7219066472,"imp":15.8553978863,"elo":1418.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.3541917099,"x_train":807.6777264234,"imp":15.6543662568,"elo":1426.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.5669239288,"x_train":1220.4156808351,"imp":15.5901216306,"elo":1434.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":1.5118634561,"x_train":1641.7401040643,"imp":15.561933124,"elo":1435.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4673275113,"x_train":4.8183344075,"imp":11.9424971927,"elo":1611.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.353062543,"x_train":2.9028568606,"imp":10.2772421387,"elo":1649.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2562998879,"x_train":2.6092164782,"imp":21.3395688074,"elo":1222.3,"imputed":true,"imputed_pct":33.3333333333,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":2.8468480497,"x_train":22.8672371568,"imp":20.6977308651,"elo":1251.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.1940748338,"x_train":62.9137172928,"imp":20.0800780954,"elo":1282.9,"imputed":true,"imputed_pct":33.3333333333,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":7.8193427161,"x_train":140.2609154123,"imp":19.6589836045,"elo":1298.4,"imputed":true,"imputed_pct":33.3333333333,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":13.8886048577,"x_train":347.4920082066,"imp":19.1890847234,"elo":1320.5,"imputed":true,"imputed_pct":33.3333333333,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":16.0381103916,"x_train":706.1972040222,"imp":19.015710344,"elo":1331.1,"imputed":true,"imputed_pct":33.3333333333,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":19.582466557,"x_train":1438.110967546,"imp":19.0008011295,"elo":1330.1,"imputed":true,"imputed_pct":33.3333333333,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":20.9489332922,"x_train":2168.8573846957,"imp":19.0922957383,"elo":1324.7,"imputed":true,"imputed_pct":33.3333333333,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.4120581858,"x_train":2916.211932309,"imp":19.1779850452,"elo":1313.5,"imputed":true,"imputed_pct":33.3333333333,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0895012297,"x_train":387.3713684902,"imp":29.3875976863,"elo":1073.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.1246177871,"x_train":619.0654872096,"imp":29.2699684092,"elo":1080.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9933319757,"x_train":1126.9710593518,"imp":28.9452903243,"elo":1085.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.945280813,"x_train":1799.6745277943,"imp":28.250897063,"elo":1095.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.9921647692,"x_train":4218.8944407454,"imp":26.9033265918,"elo":1111.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.187584759,"x_train":8667.7234636881,"imp":25.6818413182,"elo":1125.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":10.7769486168,"x_train":17722.5911655635,"imp":25.1929972913,"elo":1133.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.719108285,"x_train":26217.887217876,"imp":24.793983631,"elo":1143.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.9317836231,"x_train":34925.3007391894,"imp":24.4242199422,"elo":1155.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622971538,"x_train":1.1067907259,"imp":16.554354774,"elo":1378.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0923536337,"x_train":4.9663458274,"imp":24.5854535013,"elo":1062.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1965690156,"x_train":13.872258751,"imp":22.0843362149,"elo":1132.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4325805708,"x_train":43.9083852063,"imp":20.0317547645,"elo":1187.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6854277095,"x_train":107.4432911334,"imp":19.2360452664,"elo":1226.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1471383289,"x_train":267.8896031045,"imp":18.5859012039,"elo":1261.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5876850499,"x_train":552.8424768764,"imp":18.2627562112,"elo":1277.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.9574603872,"x_train":1101.0843510949,"imp":18.1797163399,"elo":1278.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1875558499,"x_train":1649.8652006608,"imp":18.032575652,"elo":1288.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.3011998096,"x_train":2206.6390156442,"imp":17.9473130882,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1176002559,"x_train":1.4234367091,"imp":20.1741412388,"elo":1214.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.291370012,"x_train":4.1993942212,"imp":18.2089294006,"elo":1311.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4007037631,"x_train":11.4672731767,"imp":17.6658989971,"elo":1344.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4312521607,"x_train":22.2498464852,"imp":17.3593397325,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4858065918,"x_train":57.7997562863,"imp":17.1192287025,"elo":1370.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4981474911,"x_train":114.5664011196,"imp":17.0644712234,"elo":1378.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5038664892,"x_train":229.0218261459,"imp":16.9859238323,"elo":1385.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6132491042,"x_train":343.7649500386,"imp":16.9058672707,"elo":1389.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6132840031,"x_train":463.012192935,"imp":16.9294193488,"elo":1389.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":44.6673844457,"x_train":202.6812316751,"imp":22.8658447838,"elo":1180.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":94.7353822956,"x_train":512.7516159338,"imp":20.3009736911,"elo":1251.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":154.9782910324,"x_train":1466.1655464244,"imp":17.472040179,"elo":1337.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":248.7212562072,"x_train":3194.8705446357,"imp":16.4000617135,"elo":1384.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":357.2213513445,"x_train":7908.2276811319,"imp":15.5054242237,"elo":1430.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":361.1026011106,"x_train":8274.6664503747,"imp":15.5539976857,"elo":1430.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.3922735327,"x_train":2.7144283593,"imp":30.4948770081,"elo":953.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.5728509945,"x_train":5.4835528309,"imp":25.913159969,"elo":1040.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.9190160514,"x_train":13.8167338689,"imp":20.5297244826,"elo":1189.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.2127267231,"x_train":27.7848683857,"imp":18.4868051449,"elo":1278.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.4609202066,"x_train":69.6362065708,"imp":17.9776134007,"elo":1302.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.965003678,"x_train":138.9025571316,"imp":17.7572096471,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.6090145696,"x_train":278.1866797416,"imp":17.5939282366,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.626504492,"x_train":417.6393154275,"imp":17.5124840659,"elo":1309.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.4524874978,"x_train":559.7696517925,"imp":17.3620070536,"elo":1310.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..922806625dde2b807d93c48c4b3050e5b235bada
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_binary/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "iLTM (tuned + ensembled)", "TabM (tuned + ensembled)", "TabDPT-Turbo (default)", "TabICL (default)", "CatBoost (tuned)", "LightGBM (tuned + ensembled)", "LimiX (default)", "CatBoost (tuned + ensembled)", "CatBoost (default)", "ChimeraBoost (tuned + ensembled)", "TabDPT (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabM (tuned)", "ModernNCA (tuned)", "TabSwift (default)", "ModernNCA (tuned + ensembled)", "ChimeraBoost (tuned)", "TabDPT (tuned)", "XGBoost (tuned)", "BetaTabPFN (default)", "LightGBM (tuned)", "iLTM (tuned)", "TabPFNv2 (tuned + ensembled)", "EBM (tuned + ensembled)", "RealMLP (default)", "xRFM (tuned + ensembled)", "TabM (default)", "Mitra (default)", "ChimeraBoost (default)", "TorchMLP (tuned + ensembled)", "xRFM (tuned)", "TabPFNv2 (tuned)", "EBM (tuned)", "EBM (default)", "SAP-RPT-OSS (default)", "TabDPT (default)", "TabPFNv2 (default)", "FastaiMLP (tuned + ensembled)", "XGBoost (default)", "ModernNCA (default)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "iLTM (default)", "ExtraTrees (tuned)", "TabSTAR (tuned)", "LightGBM (default)", "FastaiMLP (tuned)", "RandomForest (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "OrionMSP (default)", "TabSTAR (default)", "PerpetualBooster (tuned)", "TorchMLP (default)", "FastaiMLP (default)", "TabFlex (default)", "RandomForest (default)", "ExtraTrees (default)", "KNN (tuned + ensembled)", "Linear (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned)", "xRFM (default)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.8333333134651184, 0.7333333492279053, 0.8666666746139526, 0.8333333134651184, 0.9333333373069763, 0.7666666507720947, 0.8333333134651184, 0.9333333373069763, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.9666666388511658, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9666666388511658, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 1.0, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.8999999761581421, 1.0, 0.9333333373069763, 0.8999999761581421, 0.9666666388511658, 0.9333333373069763, 1.0, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 1.0, 1.0, 0.9333333373069763, 1.0, 1.0, 1.0, 1.0, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 1.0, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0, 1.0], [0.1666666716337204, NaN, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.6166666746139526, 0.5666666626930237, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.7666666507720947, 0.8999999761581421, 0.7333333492279053, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 1.0, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.8333333134651184, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.8333333134651184, 1.0, 0.9666666388511658, 1.0, 0.8333333134651184, 1.0, 0.9666666388511658, 1.0, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0, 1.0], [0.2666666805744171, 0.4000000059604645, NaN, 0.6000000238418579, 0.46666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.6333333253860474, 0.800000011920929, 0.8666666746139526, 0.8999999761581421, 0.699999988079071, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.7666666507720947, 0.8999999761581421, 0.7333333492279053, 0.8999999761581421, 0.8999999761581421, 0.7666666507720947, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.7666666507720947, 0.8333333134651184, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.9333333373069763, 0.8999999761581421, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.7666666507720947, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8333333134651184, 0.9666666388511658, 0.8999999761581421, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.8999999761581421, 1.0, 0.8333333134651184, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0], [0.13333334028720856, 0.4333333373069763, 0.4000000059604645, NaN, 0.5, 0.5333333611488342, 0.5333333611488342, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.8999999761581421, 0.7333333492279053, 0.8999999761581421, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.7333333492279053, 0.699999988079071, 0.800000011920929, 0.7666666507720947, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 1.0, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.7666666507720947, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.8666666746139526, 1.0, 0.800000011920929, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.800000011920929, 1.0, 0.9666666388511658, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 0.9333333373069763, 1.0, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 1.0, 1.0], [0.1666666716337204, 0.4333333373069763, 0.5333333611488342, 0.5, NaN, 0.5333333611488342, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.7666666507720947, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.7666666507720947, 0.8333333134651184, 0.7666666507720947, 0.9333333373069763, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.9666666388511658, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9666666388511658, 1.0, 1.0, 1.0, 0.9333333373069763, 0.8999999761581421, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 1.0, 1.0, 1.0], [0.06666667014360428, 0.38333332538604736, 0.3333333432674408, 0.46666666865348816, 0.46666666865348816, NaN, 0.5, 0.5333333611488342, 0.699999988079071, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.6666666865348816, 0.8333333134651184, 0.8333333134651184, 0.7666666507720947, 0.6666666865348816, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.699999988079071, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.8999999761581421, 0.7666666507720947, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9333333373069763, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.9333333373069763, 0.8333333134651184, 0.8666666746139526, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.800000011920929, 1.0, 0.8999999761581421, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.8333333134651184, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 1.0], [0.23333333432674408, 0.4333333373069763, 0.36666667461395264, 0.46666666865348816, 0.36666667461395264, 0.5, NaN, 0.4333333373069763, 0.5333333611488342, 0.6000000238418579, 0.7666666507720947, 0.8333333134651184, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.6333333253860474, 0.8333333134651184, 0.7666666507720947, 0.7333333492279053, 0.6666666865348816, 0.8333333134651184, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.7666666507720947, 0.8666666746139526, 0.8333333134651184, 0.6333333253860474, 0.7666666507720947, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.699999988079071, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.699999988079071, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.7666666507720947, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.7666666507720947, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0, 1.0], [0.1666666716337204, 0.4000000059604645, 0.4000000059604645, 0.36666667461395264, 0.4000000059604645, 0.46666666865348816, 0.5666666626930237, NaN, 0.46666666865348816, 0.5666666626930237, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.800000011920929, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8833333253860474, 0.699999988079071, 0.8333333134651184, 0.7666666507720947, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.9333333373069763, 0.8999999761581421, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 1.0, 0.9666666388511658, 0.8999999761581421, 0.8666666746139526, 0.9666666388511658, 0.9333333373069763, 1.0, 0.8666666746139526, 1.0, 1.0, 0.9333333373069763, 1.0, 0.9333333373069763, 1.0, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0], [0.06666667014360428, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.30000001192092896, 0.46666666865348816, 0.5333333611488342, NaN, 0.5333333611488342, 0.6333333253860474, 0.5666666626930237, 0.699999988079071, 0.6333333253860474, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.8333333134651184, 0.6333333253860474, 0.6333333253860474, 0.699999988079071, 0.800000011920929, 0.7666666507720947, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.8999999761581421, 0.800000011920929, 0.7166666388511658, 0.8333333134651184, 0.800000011920929, 0.7333333492279053, 0.8666666746139526, 0.8666666746139526, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.9333333373069763, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 1.0, 0.8333333134651184, 0.8999999761581421, 0.800000011920929, 0.9666666388511658, 0.8666666746139526, 0.9333333373069763, 0.800000011920929, 0.9333333373069763, 0.9333333373069763, 1.0, 0.8333333134651184, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 1.0], [0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.46666666865348816, NaN, 0.6666666865348816, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.5333333611488342, 0.800000011920929, 0.7666666507720947, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.699999988079071, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.7666666507720947, 0.6333333253860474, 0.800000011920929, 0.8666666746139526, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.7333333492279053, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.7333333492279053, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.7666666507720947, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.7666666507720947, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658, 1.0, 1.0], [0.1666666716337204, 0.30000001192092896, 0.1666666716337204, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.23333333432674408, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, NaN, 0.46666666865348816, 0.5, 0.5, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.44999998807907104, 0.5666666626930237, 0.6333333253860474, 0.5, 0.46666666865348816, 0.6333333253860474, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.6000000238418579, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.5666666626930237, 0.6333333253860474, 0.800000011920929, 0.6666666865348816, 0.6333333253860474, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.7666666507720947, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.7833333611488342, 0.800000011920929, 0.8666666746139526, 0.9666666388511658, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.949999988079071, 0.9666666388511658, 0.9333333373069763, 1.0, 1.0], [0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.23333333432674408, 0.2666666805744171, 0.3333333432674408, 0.1666666716337204, 0.46666666865348816, 0.4333333373069763, 0.30000001192092896, 0.5333333611488342, NaN, 0.46666666865348816, 0.5, 0.44999998807907104, 0.5666666626930237, 0.5666666626930237, 0.44999998807907104, 0.5, 0.5, 0.6333333253860474, 0.46666666865348816, 0.6000000238418579, 0.5, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.5, 0.699999988079071, 0.5, 0.6000000238418579, 0.699999988079071, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.6333333253860474, 0.5666666626930237, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.7333333492279053, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 1.0], [0.06666667014360428, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5333333611488342, NaN, 0.5666666626930237, 0.6000000238418579, 0.5, 0.6666666865348816, 0.46666666865348816, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.6000000238418579, 0.6333333253860474, 0.6000000238418579, 0.699999988079071, 0.5666666626930237, 0.699999988079071, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.5333333611488342, 0.7666666507720947, 0.7666666507720947, 0.6000000238418579, 0.8333333134651184, 0.46666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5666666626930237, 0.5666666626930237, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.7333333492279053, 0.8333333134651184, 0.800000011920929, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8999999761581421, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0], [0.06666667014360428, 0.20000000298023224, 0.36666667461395264, 0.23333333432674408, 0.23333333432674408, 0.4000000059604645, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.4000000059604645, 0.5, 0.5, 0.4333333373069763, NaN, 0.46666666865348816, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.5, 0.5333333611488342, 0.5666666626930237, 0.6333333253860474, 0.5666666626930237, 0.4333333373069763, 0.46666666865348816, 0.5833333134651184, 0.46666666865348816, 0.5666666626930237, 0.699999988079071, 0.6000000238418579, 0.5333333611488342, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.9666666388511658, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.8999999761581421, 1.0, 1.0], [0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.20000000298023224, 0.4000000059604645, 0.4333333373069763, 0.550000011920929, 0.4000000059604645, 0.5333333611488342, NaN, 0.5333333611488342, 0.5, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, 0.6000000238418579, 0.5333333611488342, 0.4000000059604645, 0.5666666626930237, 0.5666666626930237, 0.5, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.5666666626930237, 0.6333333253860474, 0.550000011920929, 0.5666666626930237, 0.6666666865348816, 0.6000000238418579, 0.5666666626930237, 0.550000011920929, 0.6333333253860474, 0.7666666507720947, 0.6666666865348816, 0.6833333373069763, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.7166666388511658, 0.7666666507720947, 0.7666666507720947, 0.8166666626930237, 0.699999988079071, 0.800000011920929, 0.7666666507720947, 0.8666666746139526, 0.7666666507720947, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.9166666865348816, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.949999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.9666666388511658, 1.0, 1.0], [0.10000000149011612, 0.23333333432674408, 0.13333334028720856, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.30000001192092896, 0.46666666865348816, 0.4333333373069763, 0.5, 0.5333333611488342, 0.46666666865348816, NaN, 0.4333333373069763, 0.46666666865348816, 0.46666666865348816, 0.6333333253860474, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.5, 0.46666666865348816, 0.5, 0.5, 0.5666666626930237, 0.5, 0.6333333253860474, 0.5666666626930237, 0.5333333611488342, 0.699999988079071, 0.5333333611488342, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.5666666626930237, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.6333333253860474, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.7666666507720947, 0.8999999761581421, 0.800000011920929, 0.7666666507720947, 0.7666666507720947, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 1.0, 1.0], [0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.2666666805744171, 0.36666667461395264, 0.4333333373069763, 0.3333333432674408, 0.4333333373069763, 0.5, 0.5666666626930237, NaN, 0.36666667461395264, 0.5666666626930237, 0.6000000238418579, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5, 0.5, 0.5333333611488342, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.46666666865348816, 0.7666666507720947, 0.6666666865348816, 0.5, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.5, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.5333333611488342, 0.7666666507720947, 0.800000011920929, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.9666666388511658, 0.7666666507720947, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.8666666746139526, 0.9333333373069763, 1.0, 1.0], [0.10000000149011612, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.4000000059604645, 0.46666666865348816, 0.550000011920929, 0.550000011920929, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, 0.5333333611488342, 0.6333333253860474, NaN, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.4333333373069763, 0.6333333253860474, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, 0.5666666626930237, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.5666666626930237, 0.6000000238418579, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6833333373069763, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.7666666507720947, 0.7833333611488342, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.9666666388511658], [0.06666667014360428, 0.1666666716337204, 0.06666667014360428, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.1666666716337204, 0.3333333432674408, 0.2666666805744171, 0.20000000298023224, 0.4333333373069763, 0.5, 0.4000000059604645, 0.5, 0.5333333611488342, 0.5333333611488342, 0.4333333373069763, 0.4333333373069763, NaN, 0.6000000238418579, 0.5, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.44999998807907104, 0.46666666865348816, 0.5, 0.5666666626930237, 0.5, 0.5666666626930237, 0.5, 0.5666666626930237, 0.699999988079071, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.6333333253860474, 0.5, 0.5, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.6000000238418579, 0.6666666865348816, 0.7333333492279053, 0.800000011920929, 0.6333333253860474, 0.6333333253860474, 0.7333333492279053, 0.7666666507720947, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.7666666507720947, 0.8333333134651184, 0.8333333134651184, 0.8999999761581421, 0.7666666507720947, 0.9333333373069763, 0.8666666746139526, 0.7333333492279053, 0.7666666507720947, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8999999761581421, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 0.9666666388511658, 1.0], [0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.30000001192092896, 0.1666666716337204, 0.23333333432674408, 0.36666667461395264, 0.5, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.36666667461395264, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, NaN, 0.4333333373069763, 0.5333333611488342, 0.5, 0.4333333373069763, 0.4333333373069763, 0.5, 0.5, 0.5666666626930237, 0.5333333611488342, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6333333253860474, 0.5666666626930237, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.6000000238418579, 0.46666666865348816, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8333333134651184, 0.800000011920929, 0.7666666507720947, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 0.9333333373069763, 0.9666666388511658, 1.0, 1.0], [0.10000000149011612, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.5, 0.36666667461395264, 0.4333333373069763, 0.4333333373069763, 0.5333333611488342, 0.5333333611488342, 0.5333333611488342, 0.4333333373069763, 0.5, 0.5666666626930237, NaN, 0.5333333611488342, 0.5333333611488342, 0.38333332538604736, 0.5, 0.5333333611488342, 0.5, 0.6333333253860474, 0.5, 0.5666666626930237, 0.4000000059604645, 0.6000000238418579, 0.6333333253860474, 0.5333333611488342, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.6333333253860474, 0.5, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.5333333611488342, 0.7333333492279053, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.7666666507720947, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.7333333492279053, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 1.0], [0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.3333333432674408, 0.3333333432674408, 0.23333333432674408, 0.36666667461395264, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.5333333611488342, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, 0.46666666865348816, NaN, 0.5, 0.46666666865348816, 0.4333333373069763, 0.5, 0.4333333373069763, 0.5, 0.6666666865348816, 0.5, 0.5, 0.5333333611488342, 0.6333333253860474, 0.6666666865348816, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6333333253860474, 0.5, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.9333333373069763, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.9666666388511658, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9666666388511658, 0.8999999761581421, 0.9666666388511658, 1.0], [0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.30000001192092896, 0.23333333432674408, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.46666666865348816, 0.46666666865348816, 0.5333333611488342, 0.36666667461395264, 0.5333333611488342, 0.5, 0.46666666865348816, 0.5, NaN, 0.46666666865348816, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.5666666626930237, 0.5, 0.699999988079071, 0.4333333373069763, 0.6000000238418579, 0.6000000238418579, 0.4333333373069763, 0.6333333253860474, 0.699999988079071, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.5, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.7333333492279053, 0.8333333134651184, 0.7333333492279053, 0.6833333373069763, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.7333333492279053, 0.9333333373069763, 0.800000011920929, 0.9666666388511658, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.9333333373069763, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 0.9333333373069763, 1.0], [0.03333333507180214, 0.10000000149011612, 0.23333333432674408, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4333333373069763, 0.5666666626930237, 0.6000000238418579, 0.5, 0.5333333611488342, 0.4333333373069763, 0.5666666626930237, 0.5666666626930237, 0.6166666746139526, 0.5333333611488342, 0.5333333611488342, NaN, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6333333253860474, 0.5, 0.6333333253860474, 0.4333333373069763, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.8333333134651184, 0.7666666507720947, 0.7333333492279053, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.9666666388511658, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658], [0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.23333333432674408, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.5333333611488342, 0.4333333373069763, 0.5333333611488342, 0.5, 0.4333333373069763, 0.550000011920929, 0.5666666626930237, 0.5, 0.5666666626930237, 0.5666666626930237, 0.4333333373069763, NaN, 0.5333333611488342, 0.46666666865348816, 0.5333333611488342, 0.6333333253860474, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.5333333611488342, 0.46666666865348816, 0.5666666626930237, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.5, 0.6333333253860474, 0.699999988079071, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6833333373069763, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.7666666507720947, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.9333333373069763, 0.8999999761581421, 0.9333333373069763, 1.0], [0.13333334028720856, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.30000001192092896, 0.20000000298023224, 0.3333333432674408, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.36666667461395264, 0.4166666567325592, 0.4333333373069763, 0.5, 0.5, 0.5333333611488342, 0.5333333611488342, 0.5, 0.46666666865348816, 0.5, 0.5333333611488342, 0.4333333373069763, 0.46666666865348816, NaN, 0.5333333611488342, 0.5333333611488342, 0.4333333373069763, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.6333333253860474, 0.5333333611488342, 0.699999988079071, 0.5666666626930237, 0.5666666626930237, 0.699999988079071, 0.5, 0.6666666865348816, 0.6333333253860474, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.7666666507720947, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9666666388511658, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 0.8666666746139526, 1.0, 1.0], [0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.3333333432674408, 0.4333333373069763, 0.46666666865348816, 0.4000000059604645, 0.5333333611488342, 0.5, 0.5, 0.46666666865348816, 0.4333333373069763, 0.5, 0.5, 0.5, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, 0.5333333611488342, 0.46666666865348816, NaN, 0.5333333611488342, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.6333333253860474, 0.5, 0.4333333373069763, 0.6333333253860474, 0.699999988079071, 0.6000000238418579, 0.5333333611488342, 0.4333333373069763, 0.5666666626930237, 0.6000000238418579, 0.6333333253860474, 0.5, 0.6333333253860474, 0.6333333253860474, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.7666666507720947, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.8500000238418579, 0.7666666507720947, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.800000011920929, 0.8666666746139526, 0.9666666388511658], [0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.4333333373069763, 0.36666667461395264, 0.30000001192092896, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.36666667461395264, 0.5, 0.4333333373069763, 0.36666667461395264, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, NaN, 0.46666666865348816, 0.5333333611488342, 0.5, 0.5, 0.46666666865348816, 0.46666666865348816, 0.699999988079071, 0.6666666865348816, 0.6000000238418579, 0.5333333611488342, 0.4333333373069763, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.5, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.6333333253860474, 0.699999988079071, 0.7666666507720947, 0.7666666507720947, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.7666666507720947, 0.8333333134651184, 0.7666666507720947, 0.8666666746139526, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0], [0.03333333507180214, 0.06666667014360428, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.23333333432674408, 0.2666666805744171, 0.2666666805744171, 0.23333333432674408, 0.3333333432674408, 0.4333333373069763, 0.5, 0.4333333373069763, 0.30000001192092896, 0.4000000059604645, 0.5, 0.46666666865348816, 0.46666666865348816, 0.5, 0.46666666865348816, 0.5, 0.3333333432674408, 0.5, 0.4333333373069763, 0.36666667461395264, 0.5666666626930237, 0.4333333373069763, 0.5333333611488342, NaN, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.5, 0.5666666626930237, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.6333333253860474, 0.5666666626930237, 0.7833333611488342, 0.6333333253860474, 0.5333333611488342, 0.6000000238418579, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.6333333253860474, 0.7666666507720947, 0.6333333253860474, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.9666666388511658, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658], [0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.3333333432674408, 0.4333333373069763, 0.46666666865348816, 0.4333333373069763, 0.5, 0.30000001192092896, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, NaN, 0.46666666865348816, 0.5333333611488342, 0.5333333611488342, 0.4333333373069763, 0.5333333611488342, 0.6000000238418579, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.5666666626930237, 0.5333333611488342, 0.6666666865348816, 0.5, 0.699999988079071, 0.6666666865348816, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.699999988079071, 0.800000011920929, 0.6666666865348816, 0.6666666865348816, 0.8666666746139526, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.7666666507720947, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8999999761581421, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 1.0], [0.03333333507180214, 0.03333333507180214, 0.1666666716337204, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.23333333432674408, 0.13333334028720856, 0.1666666716337204, 0.30000001192092896, 0.3333333432674408, 0.5, 0.36666667461395264, 0.46666666865348816, 0.46666666865348816, 0.4333333373069763, 0.5333333611488342, 0.4333333373069763, 0.5, 0.5333333611488342, 0.6000000238418579, 0.5, 0.5666666626930237, 0.4333333373069763, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.5, 0.6000000238418579, 0.5333333611488342, NaN, 0.5, 0.5, 0.4333333373069763, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.5, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.699999988079071, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.7333333492279053, 0.6333333253860474, 0.7666666507720947, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.7333333492279053, 0.9333333373069763, 0.8666666746139526, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.9333333373069763, 1.0, 1.0], [0.03333333507180214, 0.0, 0.06666667014360428, 0.0, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.36666667461395264, 0.4333333373069763, 0.4333333373069763, 0.46666666865348816, 0.23333333432674408, 0.4000000059604645, 0.4333333373069763, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.4000000059604645, 0.36666667461395264, 0.5, 0.46666666865348816, 0.46666666865348816, 0.5, NaN, 0.5, 0.46666666865348816, 0.6333333253860474, 0.6000000238418579, 0.6333333253860474, 0.6333333253860474, 0.4333333373069763, 0.6333333253860474, 0.5666666626930237, 0.6666666865348816, 0.5333333611488342, 0.6333333253860474, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6333333253860474, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.7666666507720947, 0.8333333134651184, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.8999999761581421, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.9333333373069763, 1.0], [0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.11666666716337204, 0.20000000298023224, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.5, 0.5333333611488342, 0.4000000059604645, 0.46666666865348816, 0.5, 0.5, NaN, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.5333333611488342, 0.5, 0.4333333373069763, 0.5666666626930237, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.7333333492279053, 0.699999988079071, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.5666666626930237, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.6166666746139526, 0.8333333134651184, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.8666666746139526, 0.7666666507720947, 0.8333333134651184, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.8999999761581421, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.9666666388511658, 0.8666666746139526, 1.0, 1.0], [0.03333333507180214, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.36666667461395264, 0.30000001192092896, 0.28333333134651184, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.46666666865348816, 0.3333333432674408, 0.44999998807907104, 0.46666666865348816, 0.5, 0.3333333432674408, 0.46666666865348816, 0.4333333373069763, 0.46666666865348816, 0.3333333432674408, 0.5666666626930237, 0.46666666865348816, 0.5333333611488342, 0.4333333373069763, 0.5666666626930237, 0.5333333611488342, 0.4000000059604645, 0.5666666626930237, 0.5666666626930237, 0.5333333611488342, 0.46666666865348816, NaN, 0.6000000238418579, 0.5666666626930237, 0.5, 0.5333333611488342, 0.5, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.6333333253860474, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.5, 0.6000000238418579, 0.5666666626930237, 0.6666666865348816, 0.5666666626930237, 0.6666666865348816, 0.5666666626930237, 0.6666666865348816, 0.6333333253860474, 0.5666666626930237, 0.699999988079071, 0.6333333253860474, 0.699999988079071, 0.5666666626930237, 0.6833333373069763, 0.6666666865348816, 0.7333333492279053, 0.6333333253860474, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.7333333492279053, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 1.0], [0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.30000001192092896, 0.3333333432674408, 0.23333333432674408, 0.36666667461395264, 0.4333333373069763, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.30000001192092896, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.36666667461395264, 0.36666667461395264, 0.30000001192092896, 0.46666666865348816, 0.46666666865348816, 0.4333333373069763, 0.36666667461395264, 0.4333333373069763, 0.4000000059604645, NaN, 0.6333333253860474, 0.5, 0.5666666626930237, 0.46666666865348816, 0.4333333373069763, 0.6000000238418579, 0.5666666626930237, 0.5, 0.7333333492279053, 0.8333333134651184, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.7666666507720947, 0.6000000238418579, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.9333333373069763, 0.9333333373069763, 0.8333333134651184, 0.9333333373069763, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 1.0], [0.10000000149011612, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.13333334028720856, 0.23333333432674408, 0.3333333432674408, 0.23333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.23333333432674408, 0.3333333432674408, 0.4000000059604645, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.46666666865348816, 0.30000001192092896, 0.3333333432674408, 0.5, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.36666667461395264, NaN, 0.4333333373069763, 0.46666666865348816, 0.5, 0.46666666865348816, 0.6333333253860474, 0.5666666626930237, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.699999988079071, 0.6666666865348816, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.7333333492279053, 0.7333333492279053, 0.6666666865348816, 0.800000011920929, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.9666666388511658, 0.9666666388511658, 0.9333333373069763, 0.8333333134651184, 0.8666666746139526, 0.9333333373069763, 0.9333333373069763, 0.9333333373069763, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658, 1.0], [0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4333333373069763, 0.5, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4333333373069763, 0.3333333432674408, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.5, 0.5, 0.5666666626930237, NaN, 0.5333333611488342, 0.5333333611488342, 0.36666667461395264, 0.5, 0.6333333253860474, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.6333333253860474, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.6000000238418579, 0.699999988079071, 0.6666666865348816, 0.7333333492279053, 0.699999988079071, 0.6666666865348816, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.7666666507720947, 0.699999988079071, 0.8333333134651184, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.9333333373069763, 0.8333333134651184, 0.9666666388511658, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658, 0.9666666388511658], [0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.2666666805744171, 0.30000001192092896, 0.36666667461395264, 0.1666666716337204, 0.36666667461395264, 0.4333333373069763, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.5, 0.4000000059604645, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.36666667461395264, 0.36666667461395264, 0.4333333373069763, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.36666667461395264, 0.46666666865348816, 0.46666666865348816, 0.4333333373069763, 0.5333333611488342, 0.46666666865348816, NaN, 0.36666667461395264, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, 0.5333333611488342, 0.5166666507720947, 0.6000000238418579, 0.6333333253860474, 0.6000000238418579, 0.7333333492279053, 0.6333333253860474, 0.699999988079071, 0.699999988079071, 0.6333333253860474, 0.6333333253860474, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.800000011920929, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.9666666388511658, 0.9333333373069763], [0.0, 0.10000000149011612, 0.2666666805744171, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.30000001192092896, 0.20000000298023224, 0.13333334028720856, 0.30000001192092896, 0.4333333373069763, 0.4333333373069763, 0.5333333611488342, 0.3333333432674408, 0.44999998807907104, 0.46666666865348816, 0.5, 0.36666667461395264, 0.5, 0.5333333611488342, 0.5, 0.36666667461395264, 0.4333333373069763, 0.5666666626930237, 0.5, 0.4333333373069763, 0.5666666626930237, 0.5666666626930237, 0.4000000059604645, 0.46666666865348816, 0.5, 0.5666666626930237, 0.5, 0.5, 0.5333333611488342, 0.5, 0.46666666865348816, 0.6333333253860474, NaN, 0.5333333611488342, 0.5, 0.5, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.5333333611488342, 0.6333333253860474, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.6499999761581421, 0.6666666865348816, 0.75, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.7666666507720947, 0.7666666507720947, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.9333333373069763, 0.8333333134651184, 0.8333333134651184, 1.0], [0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.5, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.30000001192092896, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.36666667461395264, 0.5666666626930237, 0.4333333373069763, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.4333333373069763, 0.46666666865348816, NaN, 0.5666666626930237, 0.6333333253860474, 0.46666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.5666666626930237, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.800000011920929, 0.6666666865348816, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.6666666865348816, 0.7666666507720947, 0.800000011920929, 0.7333333492279053, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.8999999761581421, 0.7666666507720947, 0.8999999761581421, 0.8666666746139526, 0.9666666388511658, 0.9333333373069763, 1.0], [0.03333333507180214, 0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.3333333432674408, 0.36666667461395264, 0.23333333432674408, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4333333373069763, 0.4333333373069763, 0.4333333373069763, 0.46666666865348816, 0.4000000059604645, 0.36666667461395264, 0.5, 0.4333333373069763, 0.5, 0.4333333373069763, NaN, 0.5333333611488342, 0.5, 0.46666666865348816, 0.4333333373069763, 0.5333333611488342, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6000000238418579, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.8333333134651184, 0.7333333492279053, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658, 0.9666666388511658], [0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.3333333432674408, 0.2666666805744171, 0.4333333373069763, 0.30000001192092896, 0.3333333432674408, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.30000001192092896, 0.4333333373069763, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4333333373069763, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.4333333373069763, 0.4333333373069763, 0.36666667461395264, 0.4333333373069763, 0.5, 0.36666667461395264, 0.46666666865348816, NaN, 0.5, 0.5666666626930237, 0.46666666865348816, 0.6333333253860474, 0.5, 0.5666666626930237, 0.6000000238418579, 0.5, 0.5666666626930237, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.7333333492279053, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.6000000238418579, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8666666746139526, 0.8999999761581421, 0.9666666388511658, 0.8666666746139526, 0.9666666388511658, 1.0], [0.03333333507180214, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.1666666716337204, 0.06666667014360428, 0.30000001192092896, 0.1666666716337204, 0.06666667014360428, 0.2666666805744171, 0.36666667461395264, 0.36666667461395264, 0.4333333373069763, 0.2666666805744171, 0.3166666626930237, 0.4333333373069763, 0.46666666865348816, 0.2666666805744171, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.30000001192092896, 0.5, 0.4333333373069763, 0.46666666865348816, 0.36666667461395264, 0.5, 0.5, 0.3333333432674408, 0.5, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.36666667461395264, 0.5, 0.46666666865348816, 0.4333333373069763, 0.5333333611488342, 0.4000000059604645, 0.5333333611488342, 0.5, 0.5, NaN, 0.5, 0.5666666626930237, 0.5166666507720947, 0.5, 0.5333333611488342, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.6333333253860474, 0.5, 0.6333333253860474, 0.5333333611488342, 0.5666666626930237, 0.6333333253860474, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.5833333134651184, 0.6333333253860474, 0.699999988079071, 0.6333333253860474, 0.699999988079071, 0.6333333253860474, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 1.0], [0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.06666667014360428, 0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.2666666805744171, 0.23333333432674408, 0.2666666805744171, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.4000000059604645, 0.36666667461395264, 0.30000001192092896, 0.46666666865348816, 0.30000001192092896, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.4000000059604645, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.46666666865348816, 0.4333333373069763, 0.3333333432674408, 0.5333333611488342, 0.4333333373069763, 0.5, NaN, 0.5, 0.6333333253860474, 0.6000000238418579, 0.6000000238418579, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.6333333253860474, 0.6000000238418579, 0.699999988079071, 0.7666666507720947, 0.6333333253860474, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.9333333373069763, 0.8333333134651184, 0.800000011920929, 0.8666666746139526, 0.8666666746139526, 0.8666666746139526, 0.9666666388511658, 1.0], [0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.36666667461395264, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.1666666716337204, 0.4333333373069763, 0.4333333373069763, 0.4833333194255829, 0.4333333373069763, 0.36666667461395264, 0.5666666626930237, 0.5333333611488342, 0.4333333373069763, 0.5, NaN, 0.5333333611488342, 0.5333333611488342, 0.5666666626930237, 0.6333333253860474, 0.5, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.6333333253860474, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.7666666507720947, 0.7666666507720947, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.7666666507720947, 0.800000011920929, 0.9333333373069763, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.800000011920929, 0.9333333373069763, 0.9666666388511658, 1.0], [0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.30000001192092896, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.23333333432674408, 0.3333333432674408, 0.3333333432674408, 0.20000000298023224, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.4333333373069763, 0.36666667461395264, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.46666666865348816, 0.3333333432674408, 0.4333333373069763, 0.4000000059604645, 0.4000000059604645, 0.4333333373069763, 0.3333333432674408, 0.46666666865348816, 0.36666667461395264, 0.4833333194255829, 0.36666667461395264, 0.46666666865348816, NaN, 0.5666666626930237, 0.5666666626930237, 0.6000000238418579, 0.46666666865348816, 0.5666666626930237, 0.5666666626930237, 0.6333333253860474, 0.6000000238418579, 0.7333333492279053, 0.6333333253860474, 0.6000000238418579, 0.699999988079071, 0.5666666626930237, 0.6333333253860474, 0.6000000238418579, 0.7666666507720947, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.7333333492279053, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.8333333134651184, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.9333333373069763, 1.0], [0.0, 0.06666667014360428, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.21666666865348816, 0.36666667461395264, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.5, 0.3333333432674408, 0.30000001192092896, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, NaN, 0.46666666865348816, 0.5, 0.5, 0.6333333253860474, 0.46666666865348816, 0.5666666626930237, 0.5333333611488342, 0.6333333253860474, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.7666666507720947, 0.6666666865348816, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.800000011920929, 0.7333333492279053, 0.7333333492279053, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8666666746139526, 0.9666666388511658], [0.06666667014360428, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.10000000149011612, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.30000001192092896, 0.28333333134651184, 0.36666667461395264, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.3333333432674408, 0.4000000059604645, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.46666666865348816, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.5333333611488342, NaN, 0.5, 0.4333333373069763, 0.4333333373069763, 0.5333333611488342, 0.4333333373069763, 0.5666666626930237, 0.5, 0.5666666626930237, 0.46666666865348816, 0.5666666626930237, 0.5333333611488342, 0.6000000238418579, 0.550000011920929, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.7333333492279053, 0.7666666507720947, 0.6666666865348816, 0.7333333492279053, 0.6333333253860474, 0.7666666507720947, 0.8333333134651184, 0.7666666507720947, 0.8666666746139526, 0.8333333134651184, 0.800000011920929, 1.0], [0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.03333333507180214, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.36666667461395264, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.36666667461395264, 0.2666666805744171, 0.30000001192092896, 0.3166666626930237, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.46666666865348816, 0.30000001192092896, 0.36666667461395264, 0.30000001192092896, 0.2666666805744171, 0.4333333373069763, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.36666667461395264, 0.4000000059604645, 0.5, 0.5, NaN, 0.4333333373069763, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.5333333611488342, 0.7666666507720947, 0.5666666626930237, 0.6000000238418579, 0.6333333253860474, 0.7333333492279053, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.7666666507720947, 0.9666666388511658, 0.8333333134651184, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.800000011920929, 0.8999999761581421, 0.8333333134651184, 0.8666666746139526, 0.8666666746139526, 0.8999999761581421, 0.9666666388511658], [0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.3333333432674408, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.3333333432674408, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.2666666805744171, 0.23333333432674408, 0.4000000059604645, 0.20000000298023224, 0.2666666805744171, 0.30000001192092896, 0.4333333373069763, 0.3333333432674408, 0.3333333432674408, 0.30000001192092896, 0.4000000059604645, 0.36666667461395264, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.5, 0.4000000059604645, 0.4333333373069763, 0.5, 0.5333333611488342, 0.5, 0.5666666626930237, 0.5666666626930237, NaN, 0.5666666626930237, 0.5, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6000000238418579, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.7666666507720947, 0.800000011920929, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.8666666746139526, 0.9333333373069763], [0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.10000000149011612, 0.2666666805744171, 0.23333333432674408, 0.1666666716337204, 0.3333333432674408, 0.18333333730697632, 0.30000001192092896, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.20000000298023224, 0.3333333432674408, 0.23333333432674408, 0.3333333432674408, 0.2666666805744171, 0.3333333432674408, 0.36666667461395264, 0.36666667461395264, 0.3333333432674408, 0.4333333373069763, 0.23333333432674408, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.4000000059604645, 0.3333333432674408, 0.4333333373069763, 0.36666667461395264, 0.5666666626930237, 0.4000000059604645, 0.4333333373069763, NaN, 0.4000000059604645, 0.5333333611488342, 0.6000000238418579, 0.5333333611488342, 0.7666666507720947, 0.5333333611488342, 0.6000000238418579, 0.6333333253860474, 0.7666666507720947, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.8333333134651184, 0.699999988079071, 0.8333333134651184, 0.800000011920929, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.8666666746139526, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.7666666507720947, 0.8666666746139526, 0.9666666388511658, 0.9666666388511658], [0.0, 0.03333333507180214, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.3333333432674408, 0.3166666626930237, 0.23333333432674408, 0.20000000298023224, 0.4000000059604645, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.23333333432674408, 0.23333333432674408, 0.3333333432674408, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.2666666805744171, 0.3333333432674408, 0.36666667461395264, 0.4333333373069763, 0.3333333432674408, 0.4333333373069763, 0.5333333611488342, 0.46666666865348816, 0.4333333373069763, 0.5, 0.6000000238418579, NaN, 0.5833333134651184, 0.6000000238418579, 0.6333333253860474, 0.7333333492279053, 0.550000011920929, 0.5333333611488342, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.7333333492279053, 0.8999999761581421, 0.7666666507720947, 0.8999999761581421, 0.699999988079071, 0.7666666507720947, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.9333333373069763, 0.9666666388511658], [0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.13333334028720856, 0.0, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.13333334028720856, 0.30000001192092896, 0.1666666716337204, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.13333334028720856, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.13333334028720856, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.4333333373069763, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.46666666865348816, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.5, 0.36666667461395264, 0.30000001192092896, 0.36666667461395264, 0.4333333373069763, 0.5666666626930237, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.4166666567325592, NaN, 0.46666666865348816, 0.6666666865348816, 0.6000000238418579, 0.46666666865348816, 0.4333333373069763, 0.5333333611488342, 0.6333333253860474, 0.6666666865348816, 0.6000000238418579, 0.6666666865348816, 0.699999988079071, 0.6333333253860474, 0.5666666626930237, 0.699999988079071, 0.7666666507720947, 0.8666666746139526, 0.8999999761581421, 0.8666666746139526, 0.7666666507720947, 0.7333333492279053, 0.8666666746139526, 0.7666666507720947, 0.8333333134651184, 0.8999999761581421, 1.0], [0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.13333334028720856, 0.30000001192092896, 0.20000000298023224, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.38333332538604736, 0.3333333432674408, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.36666667461395264, 0.36666667461395264, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, NaN, 0.6000000238418579, 0.6666666865348816, 0.4000000059604645, 0.5333333611488342, 0.5666666626930237, 0.6666666865348816, 0.5333333611488342, 0.5333333611488342, 0.6333333253860474, 0.7333333492279053, 0.5, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.8333333134651184, 0.7333333492279053, 0.5666666626930237, 0.7333333492279053, 0.7666666507720947, 0.8333333134651184, 0.8666666746139526, 0.9333333373069763], [0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.2666666805744171, 0.2666666805744171, 0.1666666716337204, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.1666666716337204, 0.36666667461395264, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.30000001192092896, 0.4000000059604645, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.46666666865348816, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.36666667461395264, 0.5, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.36666667461395264, 0.3333333432674408, 0.4000000059604645, NaN, 0.5666666626930237, 0.36666667461395264, 0.4333333373069763, 0.4333333373069763, 0.5666666626930237, 0.6000000238418579, 0.6000000238418579, 0.6333333253860474, 0.7333333492279053, 0.6000000238418579, 0.5, 0.6000000238418579, 0.7333333492279053, 0.800000011920929, 0.800000011920929, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.8333333134651184, 0.9666666388511658, 0.9666666388511658], [0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.3166666626930237, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.3333333432674408, 0.23333333432674408, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.4333333373069763, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.3333333432674408, 0.4333333373069763, 0.23333333432674408, 0.2666666805744171, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.3333333432674408, 0.23333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.4333333373069763, NaN, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.699999988079071, 0.4333333373069763, 0.6000000238418579, 0.5333333611488342, 0.9333333373069763, 0.6000000238418579, 0.5333333611488342, 0.6000000238418579, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.7666666507720947, 0.7333333492279053, 0.8333333134651184, 0.8999999761581421], [0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.0, 0.06666667014360428, 0.0, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.03333333507180214, 0.30000001192092896, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.0, 0.23333333432674408, 0.1666666716337204, 0.2666666805744171, 0.13333334028720856, 0.1666666716337204, 0.36666667461395264, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.5333333611488342, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, 0.44999998807907104, 0.5333333611488342, 0.6000000238418579, 0.6333333253860474, 0.6000000238418579, NaN, 0.5333333611488342, 0.5, 0.6333333253860474, 0.5333333611488342, 0.6000000238418579, 0.5666666626930237, 0.699999988079071, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.699999988079071, 0.7666666507720947, 0.7666666507720947, 0.699999988079071, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.9333333373069763], [0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.03333333507180214, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.2666666805744171, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.2666666805744171, 0.23333333432674408, 0.36666667461395264, 0.23333333432674408, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.4000000059604645, 0.2666666805744171, 0.2666666805744171, 0.3333333432674408, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4333333373069763, 0.23333333432674408, 0.4000000059604645, 0.4000000059604645, 0.46666666865348816, 0.5666666626930237, 0.46666666865348816, 0.5666666626930237, 0.5666666626930237, 0.46666666865348816, NaN, 0.46666666865348816, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.5666666626930237, 0.6666666865348816, 0.6333333253860474, 0.6333333253860474, 0.7666666507720947, 0.7666666507720947, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.7333333492279053, 0.699999988079071, 0.7666666507720947, 0.800000011920929, 0.8666666746139526, 0.9333333373069763], [0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.0, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.06666667014360428, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.13333334028720856, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.2666666805744171, 0.06666667014360428, 0.23333333432674408, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.1666666716337204, 0.36666667461395264, 0.13333334028720856, 0.2666666805744171, 0.20000000298023224, 0.13333334028720856, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.4333333373069763, 0.23333333432674408, 0.23333333432674408, 0.3333333432674408, 0.4333333373069763, 0.3333333432674408, 0.30000001192092896, 0.4333333373069763, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.3333333432674408, 0.36666667461395264, 0.4000000059604645, 0.46666666865348816, 0.4333333373069763, 0.5666666626930237, 0.6000000238418579, 0.5, 0.5333333611488342, NaN, 0.6333333253860474, 0.6666666865348816, 0.5333333611488342, 0.5666666626930237, 0.7333333492279053, 0.6333333253860474, 0.699999988079071, 0.699999988079071, 0.6666666865348816, 0.8666666746139526, 0.800000011920929, 0.7666666507720947, 0.7333333492279053, 0.6333333253860474, 0.7333333492279053, 0.800000011920929, 0.8333333134651184, 0.8999999761581421, 0.9666666388511658], [0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.3333333432674408, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.4333333373069763, 0.23333333432674408, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4333333373069763, 0.23333333432674408, 0.2666666805744171, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.23333333432674408, 0.30000001192092896, 0.36666667461395264, 0.3333333432674408, 0.4333333373069763, 0.30000001192092896, 0.36666667461395264, 0.4333333373069763, 0.36666667461395264, NaN, 0.4333333373069763, 0.6000000238418579, 0.5333333611488342, 0.9666666388511658, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.6000000238418579, 0.6333333253860474, 0.6333333253860474, 0.6333333253860474, 0.7333333492279053, 0.7333333492279053, 0.8333333134651184, 0.8999999761581421], [0.03333333507180214, 0.0, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.0833333358168602, 0.10000000149011612, 0.06666667014360428, 0.23333333432674408, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.23333333432674408, 0.03333333507180214, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.13333334028720856, 0.2666666805744171, 0.10000000149011612, 0.2666666805744171, 0.10000000149011612, 0.1666666716337204, 0.3166666626930237, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.3499999940395355, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.4166666567325592, 0.30000001192092896, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.44999998807907104, 0.36666667461395264, 0.3333333432674408, 0.36666667461395264, 0.4000000059604645, 0.3333333432674408, 0.46666666865348816, 0.4000000059604645, 0.5666666626930237, 0.46666666865348816, 0.46666666865348816, 0.3333333432674408, 0.5666666626930237, NaN, 0.5333333611488342, 0.6000000238418579, 0.699999988079071, 0.5666666626930237, 0.5666666626930237, 0.7666666507720947, 0.7333333492279053, 0.7833333611488342, 0.7666666507720947, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.8333333134651184, 0.800000011920929, 0.8999999761581421, 0.9666666388511658], [0.0, 0.03333333507180214, 0.10000000149011612, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.20000000298023224, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.36666667461395264, 0.30000001192092896, 0.23333333432674408, 0.23333333432674408, 0.3333333432674408, 0.36666667461395264, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.46666666865348816, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4333333373069763, 0.46666666865348816, 0.4000000059604645, 0.46666666865348816, NaN, 0.5, 0.4000000059604645, 0.800000011920929, 0.5, 0.5333333611488342, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6333333253860474, 0.699999988079071, 0.8333333134651184, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.9333333373069763], [0.0, 0.0, 0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.0, 0.0, 0.06666667014360428, 0.21666666865348816, 0.1666666716337204, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.23333333432674408, 0.20000000298023224, 0.21666666865348816, 0.2666666805744171, 0.20000000298023224, 0.20000000298023224, 0.06666667014360428, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.23333333432674408, 0.23333333432674408, 0.13333334028720856, 0.23333333432674408, 0.23333333432674408, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.23333333432674408, 0.2666666805744171, 0.1666666716337204, 0.23333333432674408, 0.25, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.30000001192092896, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.36666667461395264, 0.36666667461395264, 0.46666666865348816, 0.4333333373069763, 0.4333333373069763, 0.4333333373069763, 0.46666666865348816, 0.4000000059604645, 0.5, NaN, 0.6333333253860474, 0.5, 0.5, 0.5, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.6666666865348816, 0.699999988079071, 0.7333333492279053, 0.7333333492279053, 0.699999988079071, 0.9666666388511658], [0.06666667014360428, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.36666667461395264, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.36666667461395264, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.1666666716337204, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.2666666805744171, 0.06666667014360428, 0.30000001192092896, 0.3333333432674408, 0.2666666805744171, 0.03333333507180214, 0.30000001192092896, 0.6000000238418579, 0.36666667461395264, NaN, 0.6000000238418579, 0.4000000059604645, 0.3333333432674408, 0.5666666626930237, 0.5666666626930237, 0.5, 0.5, 0.5, 0.6333333253860474, 0.5, 0.6000000238418579, 0.5333333611488342, 0.6666666865348816, 0.8666666746139526], [0.0, 0.0, 0.06666667014360428, 0.0, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.0, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.23333333432674408, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.1666666716337204, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.1666666716337204, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.23333333432674408, 0.3333333432674408, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.36666667461395264, 0.5, 0.4000000059604645, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.4333333373069763, 0.20000000298023224, 0.5, 0.4000000059604645, NaN, 0.46666666865348816, 0.5, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.5333333611488342, 0.6666666865348816, 0.800000011920929, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6000000238418579, 0.9333333373069763], [0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.0, 0.06666667014360428, 0.03333333507180214, 0.0, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.1666666716337204, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.03333333507180214, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.1666666716337204, 0.03333333507180214, 0.20000000298023224, 0.10000000149011612, 0.3333333432674408, 0.1666666716337204, 0.06666667014360428, 0.1666666716337204, 0.36666667461395264, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.36666667461395264, 0.23333333432674408, 0.30000001192092896, 0.1666666716337204, 0.10000000149011612, 0.4333333373069763, 0.3333333432674408, 0.5, 0.46666666865348816, 0.3333333432674408, 0.36666667461395264, 0.30000001192092896, 0.4333333373069763, 0.4333333373069763, 0.5, 0.5, 0.6000000238418579, 0.5333333611488342, NaN, 0.5333333611488342, 0.6333333253860474, 0.6333333253860474, 0.6666666865348816, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.699999988079071, 0.6666666865348816, 0.7666666507720947, 0.8333333134651184, 0.9333333373069763], [0.0, 0.0, 0.03333333507180214, 0.03333333507180214, 0.0, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.03333333507180214, 0.1666666716337204, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.03333333507180214, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.03333333507180214, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.30000001192092896, 0.20000000298023224, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.03333333507180214, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.30000001192092896, 0.3333333432674408, 0.4000000059604645, 0.4000000059604645, 0.3333333432674408, 0.23333333432674408, 0.30000001192092896, 0.4000000059604645, 0.23333333432674408, 0.46666666865348816, 0.5, 0.6666666865348816, 0.5, 0.46666666865348816, NaN, 0.5666666626930237, 0.5, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.699999988079071, 0.5666666626930237, 0.6666666865348816, 0.6000000238418579, 0.800000011920929, 0.9333333373069763], [0.0, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.0, 0.13333334028720856, 0.06666667014360428, 0.0, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.10000000149011612, 0.06666667014360428, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.1666666716337204, 0.03333333507180214, 0.23333333432674408, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.30000001192092896, 0.2666666805744171, 0.10000000149011612, 0.23333333432674408, 0.30000001192092896, 0.2666666805744171, 0.30000001192092896, 0.30000001192092896, 0.23333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.30000001192092896, 0.4333333373069763, 0.4333333373069763, 0.36666667461395264, 0.4333333373069763, NaN, 0.5333333611488342, 0.5, 0.46666666865348816, 0.6000000238418579, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.6333333253860474, 0.6666666865348816, 0.8333333134651184], [0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.1666666716337204, 0.10000000149011612, 0.05000000074505806, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.2666666805744171, 0.3333333432674408, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.30000001192092896, 0.13333334028720856, 0.3333333432674408, 0.20000000298023224, 0.2666666805744171, 0.23333333432674408, 0.30000001192092896, 0.13333334028720856, 0.30000001192092896, 0.21666666865348816, 0.30000001192092896, 0.4000000059604645, 0.4333333373069763, 0.4000000059604645, 0.36666667461395264, 0.5, 0.46666666865348816, NaN, 0.5, 0.5, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.6666666865348816, 0.6333333253860474, 0.6666666865348816, 0.9666666388511658], [0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.03333333507180214, 0.0, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.06666667014360428, 0.10000000149011612, 0.1666666716337204, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.03333333507180214, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.2666666805744171, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.1666666716337204, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.23333333432674408, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.1666666716337204, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.2666666805744171, 0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.23333333432674408, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.23333333432674408, 0.2666666805744171, 0.20000000298023224, 0.3333333432674408, 0.23333333432674408, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4333333373069763, 0.3333333432674408, 0.4333333373069763, 0.5, 0.5, NaN, 0.4333333373069763, 0.5666666626930237, 0.5333333611488342, 0.5666666626930237, 0.6000000238418579, 0.5666666626930237, 0.6666666865348816, 0.9333333373069763], [0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.03333333507180214, 0.13333334028720856, 0.10000000149011612, 0.03333333507180214, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.15000000596046448, 0.13333334028720856, 0.1666666716337204, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.2666666805744171, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.1666666716337204, 0.30000001192092896, 0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.2666666805744171, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.2666666805744171, 0.36666667461395264, 0.13333334028720856, 0.23333333432674408, 0.13333334028720856, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.23333333432674408, 0.36666667461395264, 0.30000001192092896, 0.2666666805744171, 0.23333333432674408, 0.4000000059604645, 0.20000000298023224, 0.36666667461395264, 0.4000000059604645, 0.5, 0.46666666865348816, 0.3333333432674408, 0.4333333373069763, 0.5333333611488342, 0.5, 0.5666666626930237, NaN, 0.6666666865348816, 0.5333333611488342, 0.6666666865348816, 0.699999988079071, 0.699999988079071, 0.9666666388511658, 0.9666666388511658], [0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.23333333432674408, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.13333334028720856, 0.23333333432674408, 0.23333333432674408, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.23333333432674408, 0.2666666805744171, 0.30000001192092896, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.5, 0.3333333432674408, 0.36666667461395264, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.3333333432674408, NaN, 0.4333333373069763, 0.7666666507720947, 0.5666666626930237, 0.6666666865348816, 0.5666666626930237, 0.8333333134651184], [0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.13333334028720856, 0.1666666716337204, 0.06666667014360428, 0.1666666716337204, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.06666667014360428, 0.1666666716337204, 0.13333334028720856, 0.13333334028720856, 0.23333333432674408, 0.13333334028720856, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.13333334028720856, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.4333333373069763, 0.30000001192092896, 0.36666667461395264, 0.2666666805744171, 0.2666666805744171, 0.36666667461395264, 0.36666667461395264, 0.4000000059604645, 0.1666666716337204, 0.3333333432674408, 0.36666667461395264, 0.20000000298023224, 0.3333333432674408, 0.30000001192092896, 0.4333333373069763, 0.46666666865348816, 0.46666666865348816, 0.46666666865348816, 0.5666666626930237, NaN, 0.5666666626930237, 0.5666666626930237, 0.5666666626930237, 0.5, 0.8666666746139526], [0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.05000000074505806, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.10000000149011612, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.1666666716337204, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.23333333432674408, 0.23333333432674408, 0.1666666716337204, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.2666666805744171, 0.30000001192092896, 0.3333333432674408, 0.30000001192092896, 0.30000001192092896, 0.2666666805744171, 0.36666667461395264, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.5, 0.36666667461395264, 0.30000001192092896, 0.4333333373069763, 0.4000000059604645, 0.4333333373069763, 0.4333333373069763, 0.3333333432674408, 0.23333333432674408, 0.4333333373069763, NaN, 0.5666666626930237, 0.6000000238418579, 0.5333333611488342, 0.800000011920929], [0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.10000000149011612, 0.13333334028720856, 0.06666667014360428, 0.10000000149011612, 0.06666667014360428, 0.13333334028720856, 0.03333333507180214, 0.10000000149011612, 0.1666666716337204, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.13333334028720856, 0.10000000149011612, 0.13333334028720856, 0.1666666716337204, 0.13333334028720856, 0.03333333507180214, 0.10000000149011612, 0.13333334028720856, 0.13333334028720856, 0.10000000149011612, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.03333333507180214, 0.13333334028720856, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.13333334028720856, 0.13333334028720856, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.23333333432674408, 0.2666666805744171, 0.23333333432674408, 0.20000000298023224, 0.23333333432674408, 0.20000000298023224, 0.2666666805744171, 0.1666666716337204, 0.3333333432674408, 0.2666666805744171, 0.4000000059604645, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4333333373069763, 0.3333333432674408, 0.4000000059604645, 0.30000001192092896, 0.4333333373069763, 0.4333333373069763, 0.4333333373069763, NaN, 0.4333333373069763, 0.5666666626930237, 0.800000011920929], [0.0, 0.0, 0.03333333507180214, 0.03333333507180214, 0.0, 0.06666667014360428, 0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.10000000149011612, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.06666667014360428, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.20000000298023224, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.1666666716337204, 0.03333333507180214, 0.03333333507180214, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.1666666716337204, 0.1666666716337204, 0.2666666805744171, 0.1666666716337204, 0.20000000298023224, 0.1666666716337204, 0.2666666805744171, 0.20000000298023224, 0.30000001192092896, 0.2666666805744171, 0.46666666865348816, 0.3333333432674408, 0.23333333432674408, 0.4000000059604645, 0.36666667461395264, 0.36666667461395264, 0.4333333373069763, 0.30000001192092896, 0.3333333432674408, 0.4333333373069763, 0.4000000059604645, 0.5666666626930237, NaN, 0.5666666626930237, 0.7333333492279053], [0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.13333334028720856, 0.03333333507180214, 0.0, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.0, 0.13333334028720856, 0.06666667014360428, 0.03333333507180214, 0.06666667014360428, 0.0, 0.06666667014360428, 0.0, 0.20000000298023224, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.03333333507180214, 0.1666666716337204, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.1666666716337204, 0.03333333507180214, 0.03333333507180214, 0.06666667014360428, 0.13333334028720856, 0.20000000298023224, 0.10000000149011612, 0.13333334028720856, 0.03333333507180214, 0.06666667014360428, 0.10000000149011612, 0.13333334028720856, 0.03333333507180214, 0.1666666716337204, 0.10000000149011612, 0.13333334028720856, 0.10000000149011612, 0.1666666716337204, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.3333333432674408, 0.4000000059604645, 0.1666666716337204, 0.20000000298023224, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.03333333507180214, 0.4333333373069763, 0.5, 0.46666666865348816, 0.4333333373069763, 0.4333333373069763, NaN, 0.9333333373069763], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.03333333507180214, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.06666667014360428, 0.0, 0.0, 0.03333333507180214, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03333333507180214, 0.0, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.03333333507180214, 0.0, 0.06666667014360428, 0.03333333507180214, 0.10000000149011612, 0.06666667014360428, 0.06666667014360428, 0.03333333507180214, 0.10000000149011612, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.13333334028720856, 0.06666667014360428, 0.06666667014360428, 0.06666667014360428, 0.1666666716337204, 0.03333333507180214, 0.06666667014360428, 0.03333333507180214, 0.1666666716337204, 0.13333334028720856, 0.20000000298023224, 0.20000000298023224, 0.2666666805744171, 0.06666667014360428, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9290043201},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8603896032},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8441558368},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8281385239},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8233766198},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8002164469},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7939393863},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7924242401},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7738095217},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7484848461},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6794372319},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6701298685},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6688311689},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6534632037},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6454545457},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6363636367},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6354978375},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6337662385},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6274891792},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6268398275},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6261904803},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6255411274},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6209956735},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6209956744},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6190476209},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6119047665},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5976190509},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5900432927},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5833333374},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5822510867},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5805194848},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5683982683},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5666666696},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5517316068},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5510822527},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5419913457},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5415584439},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5396103928},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.5396103939},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5350649389},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5194805216},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.4982684032},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.4967532497},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4896103949},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4893939439},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4798701341},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4556277114},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.4493506553},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.449134203},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.436796543},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4339826889},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4318181846},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3980519522},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3945887514},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3722943781},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3699134245},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3686147233},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3683982726},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3670995716},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3593073647},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3329004359},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3316017366},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3110389649},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2839826903},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2813852872},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2761904809},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.246753251},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2380952433},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2303030344},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2268398318},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2244588785},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2125541168},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2112554159},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1945887496},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1809523856},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1653679697},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1225108255},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.035497837}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..324426f77da72062332001d4e949d0b01116ffd7
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_all/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-2.6 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),RealMLP (tuned),iLTM (tuned + ensembled),TabM (tuned + ensembled),TabDPT-Turbo (default),TabICL (default),CatBoost (tuned),LightGBM (tuned + ensembled),LimiX (default),CatBoost (tuned + ensembled),CatBoost (default),ChimeraBoost (tuned + ensembled),TabDPT (tuned + ensembled),XGBoost (tuned + ensembled),TabM (tuned),ModernNCA (tuned),TabSwift (default),ModernNCA (tuned + ensembled),ChimeraBoost (tuned),TabDPT (tuned),XGBoost (tuned),BetaTabPFN (default),LightGBM (tuned),iLTM (tuned),TabPFNv2 (tuned + ensembled),EBM (tuned + ensembled),RealMLP (default),xRFM (tuned + ensembled),TabM (default),Mitra (default),ChimeraBoost (default),TorchMLP (tuned + ensembled),xRFM (tuned),TabPFNv2 (tuned),EBM (tuned),EBM (default),SAP-RPT-OSS (default),TabDPT (default),TabPFNv2 (default),FastaiMLP (tuned + ensembled),XGBoost (default),ModernNCA (default),TorchMLP (tuned),ExtraTrees (tuned + ensembled),iLTM (default),ExtraTrees (tuned),TabSTAR (tuned),LightGBM (default),FastaiMLP (tuned),RandomForest (tuned + ensembled),TabSTAR (tuned + ensembled),RandomForest (tuned),PerpetualBooster (tuned + ensembled),OrionMSP (default),TabSTAR (default),PerpetualBooster (tuned),TorchMLP (default),FastaiMLP (default),TabFlex (default),RandomForest (default),ExtraTrees (default),KNN (tuned + ensembled),Linear (tuned + ensembled),PerpetualBooster (default),Linear (tuned),xRFM (default),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.8333333,0.73333335,0.8666667,0.8333333,0.93333334,0.76666665,0.8333333,0.93333334,0.8,0.8333333,0.8,0.93333334,0.93333334,0.9,0.9,0.93333334,0.9,0.93333334,0.8666667,0.9,0.93333334,0.9,0.96666664,0.9,0.8666667,0.8666667,0.8666667,0.96666664,0.8333333,0.96666664,0.96666664,0.93333334,0.96666664,0.93333334,0.9,0.93333334,0.93333334,1.0,0.93333334,0.96666664,0.93333334,0.96666664,0.93333334,0.93333334,0.9,1.0,0.93333334,0.9,0.96666664,0.93333334,1.0,0.96666664,0.93333334,0.96666664,0.93333334,0.96666664,0.93333334,0.96666664,0.9,0.96666664,1.0,1.0,0.93333334,1.0,1.0,1.0,1.0,0.96666664,0.96666664,0.93333334,0.96666664,1.0,0.96666664,0.96666664,1.0,1.0,1.0
+TabPFN-3 (default),0.16666667,,0.6,0.56666666,0.56666666,0.6166667,0.56666666,0.6,0.6666667,0.73333335,0.7,0.73333335,0.8333333,0.8,0.8666667,0.76666665,0.9,0.73333335,0.8333333,0.8666667,0.8666667,0.8,0.8333333,0.9,0.8333333,0.8666667,0.9,0.8333333,0.93333334,0.8,0.96666664,1.0,0.9,0.8,0.8333333,0.9,0.8666667,0.8666667,0.9,0.9,0.8666667,0.9,0.93333334,0.93333334,0.9,0.8333333,0.93333334,0.8,0.8666667,0.93333334,0.93333334,0.96666664,0.96666664,0.96666664,0.96666664,0.8333333,0.96666664,0.93333334,0.93333334,0.8333333,1.0,0.96666664,1.0,0.8333333,1.0,0.96666664,1.0,0.93333334,0.96666664,0.96666664,0.96666664,0.93333334,0.96666664,0.96666664,0.96666664,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.26666668,0.4,,0.6,0.46666667,0.6666667,0.6333333,0.6,0.6666667,0.8,0.8333333,0.8333333,0.8,0.6333333,0.8,0.8666667,0.9,0.7,0.93333334,0.9,0.9,0.8,0.9,0.76666665,0.9,0.73333335,0.9,0.9,0.76666665,0.9,0.8333333,0.93333334,0.8666667,0.76666665,0.8333333,0.8333333,0.93333334,0.8666667,0.73333335,0.93333334,0.9,0.8666667,0.8333333,0.8333333,0.8666667,0.8333333,0.8333333,0.76666665,0.8666667,0.9,0.93333334,0.9,0.93333334,0.93333334,0.93333334,0.8333333,0.96666664,0.9,0.93333334,0.8333333,0.93333334,0.9,1.0,0.8333333,0.93333334,0.93333334,0.96666664,0.93333334,0.93333334,0.96666664,0.96666664,0.96666664,0.93333334,0.96666664,0.96666664,0.96666664,1.0,1.0
+TabPFN-2.6 (default),0.13333334,0.43333334,0.4,,0.5,0.53333336,0.53333336,0.6333333,0.6,0.6666667,0.7,0.76666665,0.73333335,0.76666665,0.9,0.73333335,0.9,0.76666665,0.8,0.8333333,0.8,0.73333335,0.7,0.8,0.76666665,0.73333335,0.8666667,0.8,0.76666665,0.8,0.8666667,1.0,0.8333333,0.8,0.9,0.8666667,0.8333333,0.8666667,0.8333333,0.8333333,0.9,0.8666667,0.9,0.93333334,0.8666667,0.8333333,0.8,0.76666665,0.8666667,0.93333334,0.93333334,0.96666664,0.93333334,0.93333334,0.93333334,0.8,0.96666664,0.8666667,1.0,0.8,0.96666664,0.96666664,0.96666664,0.8,1.0,0.96666664,0.96666664,0.9,0.96666664,0.93333334,1.0,0.96666664,0.96666664,0.96666664,0.93333334,0.96666664,1.0,1.0
+TabICLv2 (default),0.16666667,0.43333334,0.53333336,0.5,,0.53333336,0.6333333,0.6,0.6666667,0.6666667,0.6333333,0.73333335,0.7,0.76666665,0.73333335,0.76666665,0.8,0.7,0.76666665,0.8,0.8,0.8333333,0.76666665,0.73333335,0.8,0.8,0.8333333,0.76666665,0.8333333,0.76666665,0.93333334,0.8333333,0.76666665,0.8,0.8,0.8333333,0.8333333,0.8666667,0.8,0.8,0.8666667,0.8666667,0.8333333,0.9,0.9,0.8333333,0.9,0.8666667,0.96666664,0.8666667,0.9,0.9,0.9,0.8333333,0.93333334,0.8666667,0.96666664,0.93333334,0.93333334,0.8666667,0.93333334,0.93333334,0.93333334,0.8666667,0.96666664,1.0,1.0,1.0,0.93333334,0.9,0.96666664,0.93333334,0.96666664,0.96666664,0.93333334,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.06666667,0.38333333,0.33333334,0.46666667,0.46666667,,0.5,0.53333336,0.7,0.6333333,0.6333333,0.6666667,0.6666667,0.6,0.7,0.76666665,0.8,0.6666667,0.8333333,0.8333333,0.76666665,0.6666667,0.8,0.76666665,0.8,0.76666665,0.7,0.8333333,0.76666665,0.8,0.8666667,0.9,0.76666665,0.8666667,0.8333333,0.8,0.8,0.8666667,0.8666667,0.8,0.8,0.8666667,0.93333334,0.8333333,0.8666667,0.8333333,0.8,0.8333333,0.8,0.93333334,0.8333333,0.8666667,0.96666664,0.8666667,0.96666664,0.8,1.0,0.9,0.93333334,0.8,0.96666664,0.93333334,0.96666664,0.8333333,0.96666664,0.93333334,0.96666664,0.8666667,0.96666664,0.96666664,0.93333334,0.9,0.93333334,0.93333334,0.93333334,0.93333334,0.96666664,1.0
+"AutoGluon 1.4 (best, 4h)",0.23333333,0.43333334,0.36666667,0.46666667,0.36666667,0.5,,0.43333334,0.53333336,0.6,0.76666665,0.8333333,0.7,0.6666667,0.7,0.8,0.8,0.6333333,0.8333333,0.76666665,0.73333335,0.6666667,0.8333333,0.76666665,0.7,0.7,0.6666667,0.76666665,0.73333335,0.8333333,0.76666665,0.8666667,0.8333333,0.6333333,0.76666665,0.76666665,0.8333333,0.8,0.7,0.8333333,0.9,0.8666667,0.7,0.9,0.8333333,0.8,0.8666667,0.76666665,0.8666667,0.9,0.9,0.93333334,0.9,0.9,0.8666667,0.76666665,0.93333334,0.9,0.93333334,0.8,0.96666664,0.93333334,0.93333334,0.8333333,0.96666664,0.96666664,0.96666664,0.93333334,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.16666667,0.4,0.4,0.36666667,0.4,0.46666667,0.56666666,,0.46666667,0.56666666,0.6666667,0.53333336,0.6666667,0.7,0.6666667,0.6666667,0.76666665,0.7,0.6666667,0.7,0.7,0.76666665,0.73333335,0.73333335,0.7,0.8,0.6333333,0.7,0.73333335,0.8,0.8666667,0.8,0.8833333,0.7,0.8333333,0.76666665,0.8333333,0.8333333,0.8,0.8333333,0.8333333,0.8666667,0.8333333,0.93333334,0.9,0.8666667,0.8333333,0.9,0.8666667,0.9,0.9,0.8666667,0.8666667,0.8666667,0.8666667,0.8666667,1.0,0.96666664,0.9,0.8666667,0.96666664,0.93333334,1.0,0.8666667,1.0,1.0,0.93333334,1.0,0.93333334,1.0,0.96666664,0.93333334,0.93333334,0.93333334,0.96666664,0.96666664,1.0,1.0
+RealTabPFN-2.5 (tuned),0.06666667,0.33333334,0.33333334,0.4,0.33333334,0.3,0.46666667,0.53333336,,0.53333336,0.6333333,0.56666666,0.7,0.6333333,0.8,0.73333335,0.73333335,0.6,0.73333335,0.8333333,0.6333333,0.6333333,0.7,0.8,0.76666665,0.6666667,0.6666667,0.76666665,0.76666665,0.73333335,0.8333333,0.9,0.8,0.71666664,0.8333333,0.8,0.73333335,0.8666667,0.8666667,0.76666665,0.8,0.8333333,0.93333334,0.8333333,0.8333333,0.8333333,0.8,0.8,0.8,0.9,0.8666667,0.9,1.0,0.8333333,0.9,0.8,0.96666664,0.8666667,0.93333334,0.8,0.93333334,0.93333334,1.0,0.8333333,0.96666664,0.93333334,0.93333334,0.8666667,0.96666664,0.96666664,0.93333334,0.9,0.93333334,0.9,0.93333334,0.93333334,0.96666664,1.0
+RealMLP (tuned + ensembled),0.2,0.26666668,0.2,0.33333334,0.33333334,0.36666667,0.4,0.43333334,0.46666667,,0.6666667,0.7,0.6,0.6,0.6,0.7,0.73333335,0.53333336,0.8,0.76666665,0.6333333,0.6,0.6666667,0.7,0.6333333,0.6,0.6666667,0.6666667,0.6666667,0.76666665,0.7,0.8,0.76666665,0.6333333,0.8,0.8666667,0.73333335,0.73333335,0.7,0.76666665,0.8333333,0.8,0.73333335,0.8333333,0.8333333,0.8333333,0.8333333,0.73333335,0.9,0.8666667,0.9,0.8333333,0.9,0.9,0.8333333,0.76666665,0.93333334,0.8666667,0.8666667,0.76666665,0.93333334,0.93333334,0.93333334,0.8,0.96666664,0.93333334,0.96666664,0.8666667,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,0.96666664,1.0,1.0
+RealMLP (tuned),0.16666667,0.3,0.16666667,0.3,0.36666667,0.36666667,0.23333333,0.33333334,0.36666667,0.33333334,,0.46666667,0.5,0.5,0.56666666,0.53333336,0.6333333,0.45,0.56666666,0.6333333,0.5,0.46666667,0.6333333,0.6,0.6,0.53333336,0.56666666,0.56666666,0.56666666,0.6666667,0.6666667,0.7,0.6666667,0.6,0.7,0.76666665,0.7,0.7,0.56666666,0.6333333,0.8,0.6666667,0.6333333,0.73333335,0.73333335,0.7,0.73333335,0.73333335,0.8,0.8,0.73333335,0.8333333,0.8,0.8666667,0.8666667,0.76666665,0.8666667,0.8333333,0.9,0.8,0.93333334,0.8666667,0.78333336,0.8,0.8666667,0.96666664,0.93333334,0.8333333,0.93333334,0.93333334,0.93333334,0.93333334,0.96666664,0.95,0.96666664,0.93333334,1.0,1.0
+iLTM (tuned + ensembled),0.2,0.26666668,0.16666667,0.23333333,0.26666668,0.33333334,0.16666667,0.46666667,0.43333334,0.3,0.53333336,,0.46666667,0.5,0.45,0.56666666,0.56666666,0.45,0.5,0.5,0.6333333,0.46666667,0.6,0.5,0.56666666,0.56666666,0.53333336,0.6333333,0.5,0.7,0.5,0.6,0.7,0.56666666,0.6666667,0.6666667,0.7,0.6333333,0.56666666,0.7,0.76666665,0.73333335,0.6333333,0.6666667,0.7,0.76666665,0.7,0.76666665,0.8333333,0.8,0.76666665,0.8,0.8333333,0.8333333,0.8,0.73333335,0.9,0.8333333,0.93333334,0.73333335,0.93333334,0.8666667,0.8333333,0.8,0.93333334,0.96666664,0.93333334,0.9,0.96666664,0.93333334,0.96666664,0.93333334,0.93333334,0.93333334,0.96666664,0.93333334,0.96666664,1.0
+TabM (tuned + ensembled),0.06666667,0.16666667,0.2,0.26666668,0.3,0.33333334,0.3,0.33333334,0.3,0.4,0.5,0.53333336,,0.56666666,0.6,0.5,0.6666667,0.46666667,0.6,0.6,0.56666666,0.56666666,0.53333336,0.56666666,0.6,0.6333333,0.6,0.7,0.56666666,0.7,0.6333333,0.6333333,0.6333333,0.53333336,0.76666665,0.76666665,0.6,0.8333333,0.46666667,0.6666667,0.6666667,0.56666666,0.56666666,0.7,0.76666665,0.73333335,0.76666665,0.7,0.8,0.73333335,0.8333333,0.8,0.76666665,0.76666665,0.73333335,0.8333333,0.8666667,0.8333333,0.8333333,0.8333333,0.8333333,0.8,0.8333333,0.8333333,0.9,0.96666664,0.93333334,0.9,0.8333333,0.8333333,0.8666667,0.93333334,0.93333334,0.93333334,0.8666667,0.93333334,1.0,1.0
+TabDPT-Turbo (default),0.06666667,0.2,0.36666667,0.23333333,0.23333333,0.4,0.33333334,0.3,0.36666667,0.4,0.5,0.5,0.43333334,,0.46666667,0.46666667,0.56666666,0.6,0.5,0.53333336,0.56666666,0.6333333,0.56666666,0.43333334,0.46666667,0.5833333,0.46666667,0.56666666,0.7,0.6,0.53333336,0.56666666,0.6666667,0.6666667,0.6333333,0.6,0.6666667,0.6333333,0.6666667,0.6,0.6333333,0.7,0.73333335,0.6666667,0.7,0.7,0.8,0.7,0.73333335,0.7,0.6666667,0.76666665,0.8,0.76666665,0.8,0.76666665,0.8333333,0.8666667,0.8,0.76666665,0.8333333,0.8,0.9,0.8333333,0.8333333,0.8333333,0.96666664,0.93333334,0.9,0.93333334,0.93333334,0.8333333,0.93333334,0.8333333,0.93333334,0.9,1.0,1.0
+TabICL (default),0.1,0.13333334,0.2,0.1,0.26666668,0.3,0.3,0.33333334,0.2,0.4,0.43333334,0.55,0.4,0.53333336,,0.53333336,0.5,0.53333336,0.46666667,0.53333336,0.46666667,0.6,0.53333336,0.4,0.56666666,0.56666666,0.5,0.53333336,0.6,0.6,0.53333336,0.56666666,0.6333333,0.55,0.56666666,0.6666667,0.6,0.56666666,0.55,0.6333333,0.76666665,0.6666667,0.68333334,0.6333333,0.7,0.73333335,0.76666665,0.71666664,0.76666665,0.76666665,0.81666666,0.7,0.8,0.76666665,0.8666667,0.76666665,0.8666667,0.8333333,0.8333333,0.8,0.9166667,0.8333333,0.8666667,0.8333333,0.9,0.9,0.9,0.9,0.95,0.9,0.9,0.9,0.93333334,0.9,0.9,0.96666664,1.0,1.0
+CatBoost (tuned),0.1,0.23333333,0.13333334,0.26666668,0.23333333,0.23333333,0.2,0.33333334,0.26666668,0.3,0.46666667,0.43333334,0.5,0.53333336,0.46666667,,0.43333334,0.46666667,0.46666667,0.6333333,0.46666667,0.46666667,0.53333336,0.5,0.46666667,0.5,0.5,0.56666666,0.5,0.6333333,0.56666666,0.53333336,0.7,0.53333336,0.6666667,0.6666667,0.6333333,0.6,0.53333336,0.6666667,0.7,0.73333335,0.56666666,0.73333335,0.73333335,0.76666665,0.7,0.6333333,0.73333335,0.76666665,0.7,0.8,0.76666665,0.8,0.73333335,0.76666665,0.8333333,0.8,0.8666667,0.76666665,0.9,0.8,0.76666665,0.76666665,0.9,0.93333334,0.8666667,0.8333333,0.9,0.8333333,0.93333334,0.96666664,0.93333334,0.96666664,0.9,0.96666664,1.0,1.0
+LightGBM (tuned + ensembled),0.06666667,0.1,0.1,0.1,0.2,0.2,0.2,0.23333333,0.26666668,0.26666668,0.36666667,0.43333334,0.33333334,0.43333334,0.5,0.56666666,,0.36666667,0.56666666,0.6,0.46666667,0.46666667,0.46666667,0.46666667,0.5,0.5,0.53333336,0.56666666,0.53333336,0.6333333,0.46666667,0.76666665,0.6666667,0.5,0.6666667,0.6666667,0.6666667,0.6333333,0.5,0.7,0.7,0.73333335,0.53333336,0.76666665,0.8,0.6666667,0.7,0.6666667,0.76666665,0.8333333,0.76666665,0.8,0.8666667,0.8666667,0.8,0.8,0.96666664,0.76666665,0.8666667,0.8,0.93333334,0.8333333,0.8,0.8,0.8666667,0.8666667,0.8666667,0.8333333,0.96666664,0.8666667,0.96666664,0.96666664,0.8666667,0.96666664,0.8666667,0.93333334,1.0,1.0
+LimiX (default),0.1,0.26666668,0.3,0.23333333,0.3,0.33333334,0.36666667,0.3,0.4,0.46666667,0.55,0.55,0.53333336,0.4,0.46666667,0.53333336,0.6333333,,0.56666666,0.53333336,0.56666666,0.43333334,0.6333333,0.56666666,0.56666666,0.46666667,0.56666666,0.6,0.53333336,0.6666667,0.56666666,0.6,0.7,0.6666667,0.6666667,0.6666667,0.6,0.6333333,0.6333333,0.6333333,0.6666667,0.7,0.73333335,0.73333335,0.7,0.6666667,0.6666667,0.6666667,0.6333333,0.6666667,0.7,0.6666667,0.7,0.7,0.7,0.68333334,0.7,0.73333335,0.7,0.6666667,0.76666665,0.76666665,0.78333336,0.76666665,0.76666665,0.8,0.8666667,0.8,0.9,0.93333334,0.8666667,0.8666667,0.8333333,0.8666667,0.93333334,0.8666667,0.8666667,0.96666664
+CatBoost (tuned + ensembled),0.06666667,0.16666667,0.06666667,0.2,0.23333333,0.16666667,0.16666667,0.33333334,0.26666668,0.2,0.43333334,0.5,0.4,0.5,0.53333336,0.53333336,0.43333334,0.43333334,,0.6,0.5,0.43333334,0.46666667,0.43333334,0.45,0.46666667,0.5,0.56666666,0.5,0.56666666,0.5,0.56666666,0.7,0.53333336,0.6,0.6666667,0.6333333,0.5,0.5,0.8,0.6666667,0.73333335,0.6,0.6666667,0.73333335,0.8,0.6333333,0.6333333,0.73333335,0.76666665,0.73333335,0.8,0.8333333,0.8,0.8,0.76666665,0.8333333,0.8333333,0.9,0.76666665,0.93333334,0.8666667,0.73333335,0.76666665,0.8666667,0.93333334,0.8666667,0.8,0.93333334,0.8666667,0.9,0.96666664,0.93333334,0.96666664,0.9,0.96666664,0.96666664,1.0
+CatBoost (default),0.13333334,0.13333334,0.1,0.16666667,0.2,0.16666667,0.23333333,0.3,0.16666667,0.23333333,0.36666667,0.5,0.4,0.46666667,0.46666667,0.36666667,0.4,0.46666667,0.4,,0.43333334,0.53333336,0.5,0.43333334,0.43333334,0.5,0.5,0.56666666,0.53333336,0.53333336,0.46666667,0.53333336,0.6333333,0.56666666,0.6666667,0.76666665,0.7,0.6,0.46666667,0.73333335,0.73333335,0.73333335,0.6,0.7,0.73333335,0.7,0.7,0.6666667,0.8,0.8,0.7,0.8333333,0.8,0.76666665,0.73333335,0.8,0.8,0.8333333,0.9,0.8,0.9,0.8333333,0.8,0.8,0.9,0.93333334,0.9,0.8333333,0.9,0.8333333,0.96666664,0.96666664,0.93333334,0.96666664,0.93333334,0.96666664,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.1,0.13333334,0.1,0.2,0.2,0.23333333,0.26666668,0.3,0.36666667,0.36666667,0.5,0.36666667,0.43333334,0.43333334,0.53333336,0.53333336,0.53333336,0.43333334,0.5,0.56666666,,0.53333336,0.53333336,0.38333333,0.5,0.53333336,0.5,0.6333333,0.5,0.56666666,0.4,0.6,0.6333333,0.53333336,0.7,0.6666667,0.7,0.6333333,0.5,0.8,0.6666667,0.73333335,0.53333336,0.73333335,0.7,0.6666667,0.6666667,0.7,0.76666665,0.76666665,0.7,0.73333335,0.73333335,0.7,0.73333335,0.8,0.8333333,0.73333335,0.8666667,0.8,0.8666667,0.76666665,0.8,0.8,0.8333333,0.8666667,0.8,0.8666667,0.8666667,0.8333333,0.9,0.93333334,0.8333333,0.93333334,0.8666667,0.96666664,0.96666664,1.0
+TabDPT (tuned + ensembled),0.06666667,0.2,0.2,0.26666668,0.16666667,0.33333334,0.33333334,0.23333333,0.36666667,0.4,0.53333336,0.53333336,0.43333334,0.36666667,0.4,0.53333336,0.53333336,0.56666666,0.56666666,0.46666667,0.46666667,,0.5,0.46666667,0.43333334,0.5,0.43333334,0.5,0.6666667,0.5,0.5,0.53333336,0.6333333,0.6666667,0.56666666,0.6,0.6,0.6,0.6333333,0.5,0.6333333,0.6666667,0.7,0.6666667,0.7,0.7,0.73333335,0.6333333,0.6333333,0.6666667,0.8,0.6666667,0.76666665,0.73333335,0.8333333,0.76666665,0.76666665,0.8,0.73333335,0.73333335,0.76666665,0.76666665,0.93333334,0.8333333,0.8333333,0.8,0.96666664,0.9,0.93333334,0.96666664,0.9,0.8666667,0.8666667,0.8666667,0.96666664,0.9,0.96666664,1.0
+XGBoost (tuned + ensembled),0.1,0.16666667,0.1,0.3,0.23333333,0.2,0.16666667,0.26666668,0.3,0.33333334,0.36666667,0.4,0.46666667,0.43333334,0.46666667,0.46666667,0.53333336,0.36666667,0.53333336,0.5,0.46666667,0.5,,0.46666667,0.43333334,0.46666667,0.43333334,0.56666666,0.5,0.7,0.43333334,0.6,0.6,0.43333334,0.6333333,0.7,0.56666666,0.56666666,0.56666666,0.6666667,0.6666667,0.7,0.5,0.7,0.7,0.6666667,0.7,0.6666667,0.73333335,0.8333333,0.73333335,0.68333334,0.8666667,0.8,0.8,0.8,1.0,0.73333335,0.93333334,0.8,0.96666664,0.73333335,0.8,0.8,0.8,0.9,0.8333333,0.8,0.93333334,0.9,0.9,0.9,0.8333333,0.93333334,0.9,0.93333334,0.93333334,1.0
+TabM (tuned),0.033333335,0.1,0.23333333,0.2,0.26666668,0.23333333,0.23333333,0.26666668,0.2,0.3,0.4,0.5,0.43333334,0.56666666,0.6,0.5,0.53333336,0.43333334,0.56666666,0.56666666,0.6166667,0.53333336,0.53333336,,0.56666666,0.56666666,0.53333336,0.6333333,0.56666666,0.6,0.56666666,0.53333336,0.6,0.53333336,0.6,0.6333333,0.5,0.6333333,0.43333334,0.6,0.6,0.56666666,0.56666666,0.6666667,0.6666667,0.6666667,0.73333335,0.6333333,0.7,0.73333335,0.8333333,0.76666665,0.73333335,0.6666667,0.7,0.76666665,0.76666665,0.8,0.76666665,0.76666665,0.8,0.8666667,0.8,0.8,0.9,0.96666664,0.8666667,0.9,0.8333333,0.8,0.8666667,0.8666667,0.93333334,0.8666667,0.8333333,0.93333334,0.96666664,0.96666664
+ModernNCA (tuned),0.1,0.16666667,0.1,0.23333333,0.2,0.2,0.3,0.3,0.23333333,0.36666667,0.4,0.43333334,0.4,0.53333336,0.43333334,0.53333336,0.5,0.43333334,0.55,0.56666666,0.5,0.56666666,0.56666666,0.43333334,,0.53333336,0.46666667,0.53333336,0.6333333,0.53333336,0.6,0.56666666,0.53333336,0.46666667,0.56666666,0.6666667,0.6,0.6333333,0.5,0.6333333,0.7,0.6666667,0.53333336,0.6666667,0.7,0.7,0.7,0.7,0.68333334,0.73333335,0.8,0.8,0.7,0.7,0.73333335,0.76666665,0.8333333,0.8,0.76666665,0.76666665,0.8333333,0.8,0.8,0.8666667,0.8333333,0.8333333,0.8666667,0.8333333,0.9,0.8333333,0.93333334,0.8666667,0.8333333,0.9,0.93333334,0.9,0.93333334,1.0
+TabSwift (default),0.13333334,0.13333334,0.26666668,0.26666668,0.2,0.23333333,0.3,0.2,0.33333334,0.4,0.46666667,0.43333334,0.36666667,0.41666666,0.43333334,0.5,0.5,0.53333336,0.53333336,0.5,0.46666667,0.5,0.53333336,0.43333334,0.46666667,,0.53333336,0.53333336,0.43333334,0.53333336,0.53333336,0.6,0.56666666,0.56666666,0.6333333,0.53333336,0.7,0.56666666,0.56666666,0.7,0.5,0.6666667,0.6333333,0.6,0.7,0.7,0.6666667,0.6666667,0.8,0.6666667,0.6666667,0.6,0.76666665,0.73333335,0.8,0.76666665,0.73333335,0.73333335,0.7,0.76666665,0.73333335,0.76666665,0.9,0.8333333,0.8,0.8,0.8666667,0.96666664,0.8,0.9,0.9,0.8666667,0.8666667,0.9,0.93333334,0.8666667,1.0,1.0
+ModernNCA (tuned + ensembled),0.13333334,0.1,0.1,0.13333334,0.16666667,0.3,0.33333334,0.36666667,0.33333334,0.33333334,0.43333334,0.46666667,0.4,0.53333336,0.5,0.5,0.46666667,0.43333334,0.5,0.5,0.5,0.56666666,0.56666666,0.46666667,0.53333336,0.46666667,,0.53333336,0.56666666,0.6,0.56666666,0.6333333,0.5,0.43333334,0.6333333,0.7,0.6,0.53333336,0.43333334,0.56666666,0.6,0.6333333,0.5,0.6333333,0.6333333,0.6,0.7,0.6,0.73333335,0.73333335,0.76666665,0.76666665,0.73333335,0.73333335,0.7,0.6666667,0.8666667,0.8,0.76666665,0.7,0.73333335,0.76666665,0.76666665,0.8,0.76666665,0.8,0.8333333,0.76666665,0.8333333,0.8,0.85,0.76666665,0.8666667,0.8,0.93333334,0.8,0.8666667,0.96666664
+ChimeraBoost (tuned),0.13333334,0.16666667,0.1,0.2,0.23333333,0.16666667,0.23333333,0.3,0.23333333,0.33333334,0.43333334,0.36666667,0.3,0.43333334,0.46666667,0.43333334,0.43333334,0.4,0.43333334,0.43333334,0.36666667,0.5,0.43333334,0.36666667,0.46666667,0.46666667,0.46666667,,0.46666667,0.53333336,0.5,0.5,0.46666667,0.46666667,0.7,0.6666667,0.6,0.53333336,0.43333334,0.6666667,0.6,0.73333335,0.5,0.7,0.7,0.6666667,0.6333333,0.7,0.76666665,0.76666665,0.6666667,0.6666667,0.7,0.7,0.73333335,0.76666665,0.8333333,0.76666665,0.8333333,0.76666665,0.8666667,0.73333335,0.76666665,0.8,0.8,0.8666667,0.8,0.8666667,0.8,0.8666667,0.8666667,0.9,0.8333333,0.9,0.8666667,0.93333334,0.93333334,1.0
+TabDPT (tuned),0.033333335,0.06666667,0.23333333,0.23333333,0.16666667,0.23333333,0.26666668,0.26666668,0.23333333,0.33333334,0.43333334,0.5,0.43333334,0.3,0.4,0.5,0.46666667,0.46666667,0.5,0.46666667,0.5,0.33333334,0.5,0.43333334,0.36666667,0.56666666,0.43333334,0.53333336,,0.53333336,0.4,0.53333336,0.6,0.6,0.53333336,0.5,0.56666666,0.53333336,0.6,0.6,0.53333336,0.6,0.6666667,0.53333336,0.6333333,0.56666666,0.78333336,0.6333333,0.53333336,0.6,0.73333335,0.7,0.7,0.76666665,0.8,0.73333335,0.6333333,0.76666665,0.6333333,0.73333335,0.73333335,0.76666665,0.8666667,0.8,0.8333333,0.8,0.96666664,0.8333333,0.8666667,0.9,0.8333333,0.8,0.9,0.8333333,0.9,0.8666667,0.96666664,0.96666664
+XGBoost (tuned),0.16666667,0.2,0.1,0.2,0.23333333,0.2,0.16666667,0.2,0.26666668,0.23333333,0.33333334,0.3,0.3,0.4,0.4,0.36666667,0.36666667,0.33333334,0.43333334,0.46666667,0.43333334,0.5,0.3,0.4,0.46666667,0.46666667,0.4,0.46666667,0.46666667,,0.46666667,0.53333336,0.53333336,0.43333334,0.53333336,0.6,0.6666667,0.53333336,0.53333336,0.56666666,0.53333336,0.6666667,0.5,0.7,0.6666667,0.6333333,0.6333333,0.6333333,0.7,0.8,0.6666667,0.6666667,0.8666667,0.76666665,0.8,0.8,0.9,0.76666665,0.8666667,0.8,0.9,0.7,0.76666665,0.8,0.8,0.8666667,0.8333333,0.8,0.9,0.8666667,0.93333334,0.8666667,0.8333333,0.9,0.8666667,0.93333334,0.93333334,1.0
+BetaTabPFN (default),0.033333335,0.033333335,0.16666667,0.13333334,0.06666667,0.13333334,0.23333333,0.13333334,0.16666667,0.3,0.33333334,0.5,0.36666667,0.46666667,0.46666667,0.43333334,0.53333336,0.43333334,0.5,0.53333336,0.6,0.5,0.56666666,0.43333334,0.4,0.46666667,0.43333334,0.5,0.6,0.53333336,,0.5,0.5,0.43333334,0.56666666,0.56666666,0.56666666,0.53333336,0.5,0.6,0.56666666,0.56666666,0.53333336,0.7,0.6333333,0.6666667,0.6666667,0.6,0.6333333,0.73333335,0.6333333,0.76666665,0.73333335,0.73333335,0.73333335,0.76666665,0.8,0.73333335,0.73333335,0.76666665,0.73333335,0.76666665,0.76666665,0.8333333,0.8,0.8333333,0.8333333,0.8666667,0.8,0.73333335,0.93333334,0.8666667,0.8333333,0.8666667,0.8333333,0.93333334,1.0,1.0
+LightGBM (tuned),0.033333335,0.0,0.06666667,0.0,0.16666667,0.1,0.13333334,0.2,0.1,0.2,0.3,0.4,0.36666667,0.43333334,0.43333334,0.46666667,0.23333333,0.4,0.43333334,0.46666667,0.4,0.46666667,0.4,0.46666667,0.43333334,0.4,0.36666667,0.5,0.46666667,0.46666667,0.5,,0.5,0.46666667,0.6333333,0.6,0.6333333,0.6333333,0.43333334,0.6333333,0.56666666,0.6666667,0.53333336,0.6333333,0.6666667,0.6333333,0.6666667,0.6,0.7,0.7,0.6333333,0.76666665,0.73333335,0.76666665,0.76666665,0.76666665,0.8333333,0.73333335,0.8,0.8,0.9,0.8,0.76666665,0.8,0.8333333,0.8,0.8333333,0.8,0.9,0.8333333,0.9,0.8333333,0.8666667,0.9,0.8666667,0.9,0.93333334,1.0
+iLTM (tuned),0.06666667,0.1,0.13333334,0.16666667,0.23333333,0.23333333,0.16666667,0.11666667,0.2,0.23333333,0.33333334,0.3,0.36666667,0.33333334,0.36666667,0.3,0.33333334,0.3,0.3,0.36666667,0.36666667,0.36666667,0.4,0.4,0.46666667,0.43333334,0.5,0.53333336,0.4,0.46666667,0.5,0.5,,0.53333336,0.56666666,0.56666666,0.6,0.53333336,0.5,0.43333334,0.56666666,0.6666667,0.53333336,0.6666667,0.73333335,0.7,0.6333333,0.7,0.73333335,0.56666666,0.6666667,0.6666667,0.8,0.6166667,0.8333333,0.76666665,0.73333335,0.76666665,0.8666667,0.76666665,0.8333333,0.7,0.7,0.8,0.8333333,0.9,0.9,0.8333333,0.8666667,0.8666667,0.93333334,0.8666667,0.8666667,0.9,0.96666664,0.8666667,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.033333335,0.2,0.23333333,0.2,0.2,0.13333334,0.36666667,0.3,0.28333333,0.36666667,0.4,0.43333334,0.46666667,0.33333334,0.45,0.46666667,0.5,0.33333334,0.46666667,0.43333334,0.46666667,0.33333334,0.56666666,0.46666667,0.53333336,0.43333334,0.56666666,0.53333336,0.4,0.56666666,0.56666666,0.53333336,0.46666667,,0.6,0.56666666,0.5,0.53333336,0.5,0.56666666,0.53333336,0.6333333,0.6333333,0.6,0.6,0.53333336,0.5,0.6,0.56666666,0.6666667,0.56666666,0.6666667,0.56666666,0.6666667,0.6333333,0.56666666,0.7,0.6333333,0.7,0.56666666,0.68333334,0.6666667,0.73333335,0.6333333,0.76666665,0.7,0.8,0.73333335,0.8,0.8333333,0.73333335,0.8,0.8666667,0.8,0.9,0.8,0.8,1.0
+EBM (tuned + ensembled),0.06666667,0.16666667,0.16666667,0.1,0.2,0.16666667,0.23333333,0.16666667,0.16666667,0.2,0.3,0.33333334,0.23333333,0.36666667,0.43333334,0.33333334,0.33333334,0.33333334,0.4,0.33333334,0.3,0.43333334,0.36666667,0.4,0.43333334,0.36666667,0.36666667,0.3,0.46666667,0.46666667,0.43333334,0.36666667,0.43333334,0.4,,0.6333333,0.5,0.56666666,0.46666667,0.43333334,0.6,0.56666666,0.5,0.73333335,0.8333333,0.6666667,0.6666667,0.6,0.73333335,0.6666667,0.76666665,0.7,0.76666665,0.6,0.7,0.76666665,0.7,0.76666665,0.73333335,0.76666665,0.76666665,0.7,0.76666665,0.8,0.73333335,0.8333333,0.8666667,0.9,0.8,0.8333333,0.93333334,0.93333334,0.8333333,0.93333334,0.8666667,0.96666664,0.96666664,1.0
+RealMLP (default),0.1,0.1,0.16666667,0.13333334,0.16666667,0.2,0.23333333,0.23333333,0.2,0.13333334,0.23333333,0.33333334,0.23333333,0.4,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.23333333,0.33333334,0.4,0.3,0.36666667,0.33333334,0.46666667,0.3,0.33333334,0.5,0.4,0.43333334,0.4,0.43333334,0.43333334,0.36666667,,0.43333334,0.46666667,0.5,0.46666667,0.6333333,0.56666666,0.53333336,0.6,0.56666666,0.56666666,0.7,0.6666667,0.6,0.7,0.73333335,0.73333335,0.6666667,0.6666667,0.73333335,0.73333335,0.6666667,0.8,0.73333335,0.76666665,0.76666665,0.73333335,0.73333335,0.8,0.76666665,0.96666664,0.96666664,0.93333334,0.8333333,0.8666667,0.93333334,0.93333334,0.93333334,0.96666664,0.8666667,0.96666664,0.96666664,1.0
+xRFM (tuned + ensembled),0.06666667,0.13333334,0.06666667,0.16666667,0.16666667,0.2,0.16666667,0.16666667,0.26666668,0.26666668,0.3,0.3,0.4,0.33333334,0.4,0.36666667,0.33333334,0.4,0.36666667,0.3,0.3,0.4,0.43333334,0.5,0.4,0.3,0.4,0.4,0.43333334,0.33333334,0.43333334,0.36666667,0.4,0.5,0.5,0.56666666,,0.53333336,0.53333336,0.36666667,0.5,0.6333333,0.56666666,0.53333336,0.56666666,0.6,0.56666666,0.6333333,0.6666667,0.6,0.73333335,0.6,0.7,0.6666667,0.73333335,0.7,0.6666667,0.73333335,0.76666665,0.7,0.76666665,0.7,0.8333333,0.73333335,0.76666665,0.8,0.9,0.8666667,0.9,0.9,0.9,0.93333334,0.8333333,0.96666664,0.9,0.93333334,0.96666664,0.96666664
+TabM (default),0.06666667,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.2,0.16666667,0.13333334,0.26666668,0.3,0.36666667,0.16666667,0.36666667,0.43333334,0.4,0.36666667,0.36666667,0.5,0.4,0.36666667,0.4,0.43333334,0.36666667,0.36666667,0.43333334,0.46666667,0.46666667,0.46666667,0.46666667,0.46666667,0.36666667,0.46666667,0.46666667,0.43333334,0.53333336,0.46666667,,0.36666667,0.56666666,0.56666666,0.56666666,0.46666667,0.53333336,0.51666665,0.6,0.6333333,0.6,0.73333335,0.6333333,0.7,0.7,0.6333333,0.6333333,0.7,0.73333335,0.73333335,0.8,0.76666665,0.73333335,0.76666665,0.8,0.76666665,0.8,0.8,0.9,0.8,0.8333333,0.8,0.8,0.8333333,0.8666667,0.8666667,0.8333333,0.8,0.8666667,0.96666664,0.93333334
+Mitra (default),0.0,0.1,0.26666668,0.16666667,0.2,0.13333334,0.3,0.2,0.13333334,0.3,0.43333334,0.43333334,0.53333336,0.33333334,0.45,0.46666667,0.5,0.36666667,0.5,0.53333336,0.5,0.36666667,0.43333334,0.56666666,0.5,0.43333334,0.56666666,0.56666666,0.4,0.46666667,0.5,0.56666666,0.5,0.5,0.53333336,0.5,0.46666667,0.6333333,,0.53333336,0.5,0.5,0.6,0.56666666,0.56666666,0.56666666,0.6,0.6,0.56666666,0.6,0.6333333,0.7,0.53333336,0.6333333,0.6,0.6,0.6,0.6,0.56666666,0.6,0.65,0.6666667,0.75,0.6666667,0.7,0.6666667,0.76666665,0.76666665,0.76666665,0.76666665,0.7,0.8,0.8666667,0.8,0.93333334,0.8333333,0.8333333,1.0
+ChimeraBoost (default),0.06666667,0.1,0.06666667,0.16666667,0.2,0.2,0.16666667,0.16666667,0.23333333,0.23333333,0.36666667,0.3,0.33333334,0.4,0.36666667,0.33333334,0.3,0.36666667,0.2,0.26666668,0.2,0.5,0.33333334,0.4,0.36666667,0.3,0.43333334,0.33333334,0.4,0.43333334,0.4,0.36666667,0.56666666,0.43333334,0.56666666,0.53333336,0.6333333,0.43333334,0.46666667,,0.56666666,0.6333333,0.46666667,0.6666667,0.6333333,0.6666667,0.56666666,0.6333333,0.6666667,0.6666667,0.6333333,0.6,0.6666667,0.6666667,0.6666667,0.8,0.6666667,0.73333335,0.76666665,0.8,0.8333333,0.6666667,0.76666665,0.8,0.73333335,0.8333333,0.8,0.8333333,0.8,0.8333333,0.8666667,0.9,0.76666665,0.9,0.8666667,0.96666664,0.93333334,1.0
+TorchMLP (tuned + ensembled),0.033333335,0.13333334,0.1,0.1,0.13333334,0.2,0.1,0.16666667,0.2,0.16666667,0.2,0.23333333,0.33333334,0.36666667,0.23333333,0.3,0.3,0.33333334,0.33333334,0.26666668,0.33333334,0.36666667,0.33333334,0.4,0.3,0.5,0.4,0.4,0.46666667,0.46666667,0.43333334,0.43333334,0.43333334,0.46666667,0.4,0.36666667,0.5,0.43333334,0.5,0.43333334,,0.53333336,0.5,0.46666667,0.43333334,0.53333336,0.6,0.6,0.6,0.6666667,0.6,0.73333335,0.7,0.7,0.7,0.7,0.7,0.73333335,0.76666665,0.7,0.8333333,0.73333335,0.76666665,0.7,0.8,0.93333334,0.8666667,0.9,0.8,0.8,0.9,0.9,0.8666667,0.9,0.8333333,0.96666664,0.96666664,0.96666664
+xRFM (tuned),0.06666667,0.1,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.13333334,0.16666667,0.2,0.33333334,0.26666668,0.43333334,0.3,0.33333334,0.26666668,0.26666668,0.3,0.26666668,0.26666668,0.26666668,0.33333334,0.3,0.43333334,0.33333334,0.33333334,0.36666667,0.26666668,0.4,0.33333334,0.43333334,0.33333334,0.33333334,0.36666667,0.43333334,0.43333334,0.36666667,0.43333334,0.5,0.36666667,0.46666667,,0.5,0.56666666,0.46666667,0.6333333,0.5,0.56666666,0.6,0.5,0.56666666,0.6666667,0.73333335,0.6666667,0.73333335,0.6666667,0.6666667,0.6666667,0.6666667,0.7,0.73333335,0.6,0.73333335,0.73333335,0.7,0.8333333,0.8666667,0.8,0.8333333,0.9,0.9,0.9,0.8666667,0.9,0.96666664,0.8666667,0.96666664,1.0
+TabPFNv2 (tuned),0.033333335,0.06666667,0.16666667,0.1,0.16666667,0.06666667,0.3,0.16666667,0.06666667,0.26666668,0.36666667,0.36666667,0.43333334,0.26666668,0.31666666,0.43333334,0.46666667,0.26666668,0.4,0.4,0.46666667,0.3,0.5,0.43333334,0.46666667,0.36666667,0.5,0.5,0.33333334,0.5,0.46666667,0.46666667,0.46666667,0.36666667,0.5,0.46666667,0.43333334,0.53333336,0.4,0.53333336,0.5,0.5,,0.5,0.56666666,0.51666665,0.5,0.53333336,0.53333336,0.6,0.56666666,0.6333333,0.5,0.6333333,0.53333336,0.56666666,0.6333333,0.6,0.56666666,0.56666666,0.5833333,0.6333333,0.7,0.6333333,0.7,0.6333333,0.7,0.76666665,0.73333335,0.8333333,0.73333335,0.8,0.8333333,0.8,0.8666667,0.8,0.8333333,1.0
+EBM (tuned),0.06666667,0.06666667,0.16666667,0.06666667,0.1,0.16666667,0.1,0.06666667,0.16666667,0.16666667,0.26666668,0.33333334,0.3,0.33333334,0.36666667,0.26666668,0.23333333,0.26666668,0.33333334,0.3,0.26666668,0.33333334,0.3,0.33333334,0.33333334,0.4,0.36666667,0.3,0.46666667,0.3,0.3,0.36666667,0.33333334,0.4,0.26666668,0.4,0.46666667,0.46666667,0.43333334,0.33333334,0.53333336,0.43333334,0.5,,0.5,0.6333333,0.6,0.6,0.56666666,0.56666666,0.6,0.56666666,0.6333333,0.6,0.7,0.76666665,0.6333333,0.7,0.6666667,0.76666665,0.7,0.7,0.73333335,0.8,0.7,0.8,0.8,0.8666667,0.8,0.8,0.93333334,0.8333333,0.8,0.8666667,0.8666667,0.8666667,0.96666664,1.0
+EBM (default),0.06666667,0.1,0.13333334,0.13333334,0.1,0.13333334,0.16666667,0.1,0.16666667,0.16666667,0.26666668,0.3,0.23333333,0.3,0.3,0.26666668,0.2,0.3,0.26666668,0.26666668,0.3,0.3,0.3,0.33333334,0.3,0.3,0.36666667,0.3,0.36666667,0.33333334,0.36666667,0.33333334,0.26666668,0.4,0.16666667,0.43333334,0.43333334,0.48333332,0.43333334,0.36666667,0.56666666,0.53333336,0.43333334,0.5,,0.53333336,0.53333336,0.56666666,0.6333333,0.5,0.6666667,0.6666667,0.7,0.6,0.7,0.73333335,0.6333333,0.7,0.7,0.73333335,0.73333335,0.76666665,0.76666665,0.76666665,0.8,0.8333333,0.8333333,0.8666667,0.76666665,0.8,0.93333334,0.8666667,0.8666667,0.9,0.8,0.93333334,0.96666664,1.0
+SAP-RPT-OSS (default),0.1,0.16666667,0.16666667,0.16666667,0.16666667,0.16666667,0.2,0.13333334,0.16666667,0.16666667,0.3,0.23333333,0.26666668,0.3,0.26666668,0.23333333,0.33333334,0.33333334,0.2,0.3,0.33333334,0.3,0.33333334,0.33333334,0.3,0.3,0.4,0.33333334,0.43333334,0.36666667,0.33333334,0.36666667,0.3,0.46666667,0.33333334,0.43333334,0.4,0.4,0.43333334,0.33333334,0.46666667,0.36666667,0.48333332,0.36666667,0.46666667,,0.56666666,0.56666666,0.6,0.46666667,0.56666666,0.56666666,0.6333333,0.6,0.73333335,0.6333333,0.6,0.7,0.56666666,0.6333333,0.6,0.76666665,0.7,0.8,0.8,0.73333335,0.8333333,0.8666667,0.8,0.8333333,0.8333333,0.8666667,0.8333333,0.8333333,0.8666667,0.8666667,0.93333334,1.0
+TabDPT (default),0.0,0.06666667,0.16666667,0.2,0.1,0.2,0.13333334,0.16666667,0.2,0.16666667,0.26666668,0.3,0.23333333,0.2,0.23333333,0.3,0.3,0.33333334,0.36666667,0.3,0.33333334,0.26666668,0.3,0.26666668,0.3,0.33333334,0.3,0.36666667,0.21666667,0.36666667,0.33333334,0.33333334,0.36666667,0.5,0.33333334,0.3,0.43333334,0.36666667,0.4,0.43333334,0.4,0.5,0.5,0.4,0.46666667,0.43333334,,0.46666667,0.5,0.5,0.6333333,0.46666667,0.56666666,0.53333336,0.6333333,0.6,0.6,0.7,0.6,0.6333333,0.6666667,0.6666667,0.76666665,0.6666667,0.76666665,0.73333335,0.8333333,0.8,0.73333335,0.73333335,0.73333335,0.76666665,0.8,0.76666665,0.8,0.8,0.8666667,0.96666664
+TabPFNv2 (default),0.06666667,0.2,0.23333333,0.23333333,0.13333334,0.16666667,0.23333333,0.1,0.2,0.26666668,0.26666668,0.23333333,0.3,0.3,0.28333333,0.36666667,0.33333334,0.33333334,0.36666667,0.33333334,0.3,0.36666667,0.33333334,0.36666667,0.3,0.33333334,0.4,0.3,0.36666667,0.36666667,0.4,0.4,0.3,0.4,0.4,0.33333334,0.36666667,0.4,0.4,0.36666667,0.4,0.43333334,0.46666667,0.4,0.43333334,0.43333334,0.53333336,,0.5,0.43333334,0.43333334,0.53333336,0.43333334,0.56666666,0.5,0.56666666,0.46666667,0.56666666,0.53333336,0.6,0.55,0.6333333,0.6666667,0.6666667,0.6666667,0.6333333,0.73333335,0.76666665,0.6666667,0.73333335,0.6333333,0.76666665,0.8333333,0.76666665,0.8666667,0.8333333,0.8,1.0
+FastaiMLP (tuned + ensembled),0.1,0.13333334,0.13333334,0.13333334,0.033333335,0.2,0.13333334,0.13333334,0.2,0.1,0.2,0.16666667,0.2,0.26666668,0.23333333,0.26666668,0.23333333,0.36666667,0.26666668,0.2,0.23333333,0.36666667,0.26666668,0.3,0.31666666,0.2,0.26666668,0.23333333,0.46666667,0.3,0.36666667,0.3,0.26666668,0.43333334,0.26666668,0.4,0.33333334,0.26666668,0.43333334,0.33333334,0.4,0.4,0.46666667,0.43333334,0.36666667,0.4,0.5,0.5,,0.43333334,0.6,0.56666666,0.6,0.56666666,0.6,0.6,0.53333336,0.76666665,0.56666666,0.6,0.6333333,0.73333335,0.7,0.7,0.73333335,0.76666665,0.96666664,0.8333333,0.76666665,0.8,0.8666667,0.8,0.9,0.8333333,0.8666667,0.8666667,0.9,0.96666664
+XGBoost (default),0.033333335,0.06666667,0.1,0.06666667,0.13333334,0.06666667,0.1,0.1,0.1,0.13333334,0.2,0.2,0.26666668,0.3,0.23333333,0.23333333,0.16666667,0.33333334,0.23333333,0.2,0.23333333,0.33333334,0.16666667,0.26666668,0.26666668,0.33333334,0.26666668,0.23333333,0.4,0.2,0.26666668,0.3,0.43333334,0.33333334,0.33333334,0.3,0.4,0.36666667,0.4,0.33333334,0.33333334,0.5,0.4,0.43333334,0.5,0.53333336,0.5,0.56666666,0.56666666,,0.56666666,0.5,0.6666667,0.6666667,0.6666667,0.6666667,0.6,0.6,0.6666667,0.6666667,0.6666667,0.6,0.6333333,0.7,0.7,0.7,0.73333335,0.7,0.76666665,0.73333335,0.76666665,0.8,0.8,0.8333333,0.8666667,0.8,0.8666667,0.93333334
+ModernNCA (default),0.06666667,0.06666667,0.06666667,0.06666667,0.1,0.16666667,0.1,0.1,0.13333334,0.1,0.26666668,0.23333333,0.16666667,0.33333334,0.18333334,0.3,0.23333333,0.3,0.26666668,0.3,0.3,0.2,0.26666668,0.16666667,0.2,0.33333334,0.23333333,0.33333334,0.26666668,0.33333334,0.36666667,0.36666667,0.33333334,0.43333334,0.23333333,0.26666668,0.26666668,0.3,0.36666667,0.36666667,0.4,0.43333334,0.43333334,0.4,0.33333334,0.43333334,0.36666667,0.56666666,0.4,0.43333334,,0.4,0.53333336,0.6,0.53333336,0.76666665,0.53333336,0.6,0.6333333,0.76666665,0.6333333,0.6666667,0.7,0.8333333,0.7,0.8333333,0.8,0.73333335,0.8,0.8333333,0.8666667,0.8,0.8,0.8,0.76666665,0.8666667,0.96666664,0.96666664
+TorchMLP (tuned),0.0,0.033333335,0.1,0.033333335,0.1,0.13333334,0.06666667,0.13333334,0.1,0.16666667,0.16666667,0.2,0.2,0.23333333,0.3,0.2,0.2,0.33333334,0.2,0.16666667,0.26666668,0.33333334,0.31666666,0.23333333,0.2,0.4,0.23333333,0.33333334,0.3,0.33333334,0.23333333,0.23333333,0.33333334,0.33333334,0.3,0.26666668,0.4,0.3,0.3,0.4,0.26666668,0.33333334,0.36666667,0.43333334,0.33333334,0.43333334,0.53333336,0.46666667,0.43333334,0.5,0.6,,0.5833333,0.6,0.6333333,0.73333335,0.55,0.53333336,0.6,0.7,0.6,0.73333335,0.7,0.76666665,0.73333335,0.9,0.76666665,0.9,0.7,0.76666665,0.8333333,0.8333333,0.8333333,0.8333333,0.8,0.9,0.93333334,0.96666664
+ExtraTrees (tuned + ensembled),0.033333335,0.033333335,0.06666667,0.06666667,0.1,0.033333335,0.1,0.13333334,0.0,0.1,0.2,0.16666667,0.23333333,0.2,0.2,0.23333333,0.13333334,0.3,0.16666667,0.2,0.26666668,0.23333333,0.13333334,0.26666668,0.3,0.23333333,0.26666668,0.3,0.3,0.13333334,0.26666668,0.26666668,0.2,0.43333334,0.23333333,0.33333334,0.3,0.36666667,0.46666667,0.33333334,0.3,0.26666668,0.5,0.36666667,0.3,0.36666667,0.43333334,0.56666666,0.4,0.33333334,0.46666667,0.41666666,,0.46666667,0.6666667,0.6,0.46666667,0.43333334,0.53333336,0.6333333,0.6666667,0.6,0.6666667,0.7,0.6333333,0.56666666,0.7,0.76666665,0.8666667,0.9,0.8666667,0.76666665,0.73333335,0.8666667,0.76666665,0.8333333,0.9,1.0
+iLTM (default),0.06666667,0.033333335,0.06666667,0.06666667,0.16666667,0.13333334,0.1,0.13333334,0.16666667,0.1,0.13333334,0.16666667,0.23333333,0.23333333,0.23333333,0.2,0.13333334,0.3,0.2,0.23333333,0.3,0.26666668,0.2,0.33333334,0.3,0.26666668,0.26666668,0.3,0.23333333,0.23333333,0.26666668,0.23333333,0.38333333,0.33333334,0.4,0.33333334,0.33333334,0.36666667,0.36666667,0.33333334,0.3,0.33333334,0.36666667,0.4,0.4,0.4,0.46666667,0.43333334,0.43333334,0.33333334,0.4,0.4,0.53333336,,0.6,0.6666667,0.4,0.53333336,0.56666666,0.6666667,0.53333336,0.53333336,0.6333333,0.73333335,0.5,0.6666667,0.6666667,0.7,0.6666667,0.7,0.8333333,0.73333335,0.56666666,0.73333335,0.76666665,0.8333333,0.8666667,0.93333334
+ExtraTrees (tuned),0.033333335,0.033333335,0.06666667,0.06666667,0.06666667,0.033333335,0.13333334,0.13333334,0.1,0.16666667,0.13333334,0.2,0.26666668,0.2,0.13333334,0.26666668,0.2,0.3,0.2,0.26666668,0.26666668,0.16666667,0.2,0.3,0.26666668,0.2,0.3,0.26666668,0.2,0.2,0.26666668,0.23333333,0.16666667,0.36666667,0.3,0.26666668,0.26666668,0.3,0.4,0.33333334,0.3,0.26666668,0.46666667,0.3,0.3,0.26666668,0.36666667,0.5,0.4,0.33333334,0.46666667,0.36666667,0.33333334,0.4,,0.56666666,0.36666667,0.43333334,0.43333334,0.56666666,0.6,0.6,0.6333333,0.73333335,0.6,0.5,0.6,0.73333335,0.8,0.8,0.76666665,0.7,0.7,0.7,0.73333335,0.8333333,0.96666664,0.96666664
+TabSTAR (tuned),0.06666667,0.16666667,0.16666667,0.2,0.13333334,0.2,0.23333333,0.13333334,0.2,0.23333333,0.23333333,0.26666668,0.16666667,0.23333333,0.23333333,0.23333333,0.2,0.31666666,0.23333333,0.2,0.2,0.23333333,0.2,0.23333333,0.23333333,0.23333333,0.33333334,0.23333333,0.26666668,0.2,0.23333333,0.23333333,0.23333333,0.43333334,0.23333333,0.26666668,0.3,0.26666668,0.4,0.2,0.3,0.33333334,0.43333334,0.23333333,0.26666668,0.36666667,0.4,0.43333334,0.4,0.33333334,0.23333333,0.26666668,0.4,0.33333334,0.43333334,,0.4,0.43333334,0.4,0.7,0.43333334,0.6,0.53333336,0.93333334,0.6,0.53333336,0.6,0.7,0.73333335,0.7,0.6333333,0.6333333,0.6333333,0.6666667,0.76666665,0.73333335,0.8333333,0.9
+LightGBM (default),0.033333335,0.033333335,0.033333335,0.033333335,0.033333335,0.0,0.06666667,0.0,0.033333335,0.06666667,0.13333334,0.1,0.13333334,0.16666667,0.13333334,0.16666667,0.033333335,0.3,0.16666667,0.2,0.16666667,0.23333333,0.0,0.23333333,0.16666667,0.26666668,0.13333334,0.16666667,0.36666667,0.1,0.2,0.16666667,0.26666668,0.3,0.3,0.33333334,0.33333334,0.26666668,0.4,0.33333334,0.3,0.33333334,0.36666667,0.36666667,0.36666667,0.4,0.4,0.53333336,0.46666667,0.4,0.46666667,0.45,0.53333336,0.6,0.6333333,0.6,,0.53333336,0.5,0.6333333,0.53333336,0.6,0.56666666,0.7,0.6333333,0.6666667,0.6666667,0.7,0.76666665,0.76666665,0.7,0.7,0.73333335,0.7,0.8,0.8333333,0.9,0.93333334
+FastaiMLP (tuned),0.06666667,0.06666667,0.1,0.13333334,0.06666667,0.1,0.1,0.033333335,0.13333334,0.13333334,0.16666667,0.16666667,0.16666667,0.13333334,0.16666667,0.2,0.23333333,0.26666668,0.16666667,0.16666667,0.26666668,0.2,0.26666668,0.2,0.2,0.26666668,0.2,0.23333333,0.23333333,0.23333333,0.26666668,0.26666668,0.23333333,0.36666667,0.23333333,0.2,0.26666668,0.2,0.4,0.26666668,0.26666668,0.33333334,0.4,0.3,0.3,0.3,0.3,0.43333334,0.23333333,0.4,0.4,0.46666667,0.56666666,0.46666667,0.56666666,0.56666666,0.46666667,,0.46666667,0.56666666,0.53333336,0.56666666,0.56666666,0.6666667,0.6333333,0.6333333,0.76666665,0.76666665,0.7,0.73333335,0.73333335,0.7,0.73333335,0.7,0.76666665,0.8,0.8666667,0.93333334
+RandomForest (tuned + ensembled),0.033333335,0.06666667,0.06666667,0.0,0.06666667,0.06666667,0.06666667,0.1,0.06666667,0.13333334,0.1,0.06666667,0.16666667,0.2,0.16666667,0.13333334,0.13333334,0.3,0.1,0.1,0.13333334,0.26666668,0.06666667,0.23333333,0.23333333,0.3,0.23333333,0.16666667,0.36666667,0.13333334,0.26666668,0.2,0.13333334,0.3,0.26666668,0.26666668,0.23333333,0.23333333,0.43333334,0.23333333,0.23333333,0.33333334,0.43333334,0.33333334,0.3,0.43333334,0.4,0.46666667,0.43333334,0.33333334,0.36666667,0.4,0.46666667,0.43333334,0.56666666,0.6,0.5,0.53333336,,0.6333333,0.6666667,0.53333336,0.56666666,0.73333335,0.6333333,0.7,0.7,0.6666667,0.8666667,0.8,0.76666665,0.73333335,0.6333333,0.73333335,0.8,0.8333333,0.9,0.96666664
+TabSTAR (tuned + ensembled),0.1,0.16666667,0.16666667,0.2,0.13333334,0.2,0.2,0.13333334,0.2,0.23333333,0.2,0.26666668,0.16666667,0.23333333,0.2,0.23333333,0.2,0.33333334,0.23333333,0.2,0.2,0.26666668,0.2,0.23333333,0.23333333,0.23333333,0.3,0.23333333,0.26666668,0.2,0.23333333,0.2,0.23333333,0.43333334,0.23333333,0.23333333,0.3,0.26666668,0.4,0.2,0.3,0.3,0.43333334,0.23333333,0.26666668,0.36666667,0.36666667,0.4,0.4,0.33333334,0.23333333,0.3,0.36666667,0.33333334,0.43333334,0.3,0.36666667,0.43333334,0.36666667,,0.43333334,0.6,0.53333336,0.96666664,0.6,0.56666666,0.6,0.6666667,0.7,0.6666667,0.6,0.6333333,0.6333333,0.6333333,0.73333335,0.73333335,0.8333333,0.9
+RandomForest (tuned),0.033333335,0.0,0.06666667,0.033333335,0.06666667,0.033333335,0.033333335,0.033333335,0.06666667,0.06666667,0.06666667,0.06666667,0.16666667,0.16666667,0.083333336,0.1,0.06666667,0.23333333,0.06666667,0.1,0.13333334,0.23333333,0.033333335,0.2,0.16666667,0.26666668,0.26666668,0.13333334,0.26666668,0.1,0.26666668,0.1,0.16666667,0.31666666,0.23333333,0.23333333,0.23333333,0.23333333,0.35,0.16666667,0.16666667,0.26666668,0.41666666,0.3,0.26666668,0.4,0.33333334,0.45,0.36666667,0.33333334,0.36666667,0.4,0.33333334,0.46666667,0.4,0.56666666,0.46666667,0.46666667,0.33333334,0.56666666,,0.53333336,0.6,0.7,0.56666666,0.56666666,0.76666665,0.73333335,0.78333336,0.76666665,0.8,0.7,0.6,0.7,0.8333333,0.8,0.9,0.96666664
+PerpetualBooster (tuned + ensembled),0.0,0.033333335,0.1,0.033333335,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.13333334,0.2,0.2,0.16666667,0.2,0.16666667,0.23333333,0.13333334,0.16666667,0.23333333,0.23333333,0.26666668,0.13333334,0.2,0.23333333,0.23333333,0.26666668,0.23333333,0.3,0.23333333,0.2,0.3,0.33333334,0.3,0.26666668,0.3,0.2,0.33333334,0.33333334,0.26666668,0.4,0.36666667,0.3,0.23333333,0.23333333,0.33333334,0.36666667,0.26666668,0.4,0.33333334,0.26666668,0.4,0.46666667,0.4,0.4,0.4,0.43333334,0.46666667,0.4,0.46666667,,0.5,0.4,0.8,0.5,0.53333336,0.6,0.7,0.6,0.6333333,0.7,0.8333333,0.7,0.6666667,0.7,0.7,0.93333334
+OrionMSP (default),0.0,0.0,0.0,0.033333335,0.06666667,0.033333335,0.06666667,0.0,0.0,0.06666667,0.21666667,0.16666667,0.16666667,0.1,0.13333334,0.23333333,0.2,0.21666667,0.26666668,0.2,0.2,0.06666667,0.2,0.2,0.2,0.1,0.23333333,0.23333333,0.13333334,0.23333333,0.23333333,0.23333333,0.3,0.26666668,0.23333333,0.26666668,0.16666667,0.23333333,0.25,0.23333333,0.23333333,0.26666668,0.3,0.26666668,0.23333333,0.3,0.23333333,0.33333334,0.3,0.36666667,0.3,0.3,0.33333334,0.36666667,0.36666667,0.46666667,0.43333334,0.43333334,0.43333334,0.46666667,0.4,0.5,,0.6333333,0.5,0.5,0.5,0.7,0.6,0.7,0.6,0.7,0.6666667,0.7,0.73333335,0.73333335,0.7,0.96666664
+TabSTAR (default),0.06666667,0.16666667,0.16666667,0.2,0.13333334,0.16666667,0.16666667,0.13333334,0.16666667,0.2,0.2,0.2,0.16666667,0.16666667,0.16666667,0.23333333,0.2,0.23333333,0.23333333,0.2,0.2,0.16666667,0.2,0.2,0.13333334,0.16666667,0.2,0.2,0.2,0.2,0.16666667,0.2,0.2,0.36666667,0.2,0.2,0.26666668,0.2,0.33333334,0.2,0.3,0.26666668,0.36666667,0.2,0.23333333,0.2,0.33333334,0.33333334,0.3,0.3,0.16666667,0.23333333,0.3,0.26666668,0.26666668,0.06666667,0.3,0.33333334,0.26666668,0.033333335,0.3,0.6,0.36666667,,0.6,0.4,0.33333334,0.56666666,0.56666666,0.5,0.5,0.5,0.6333333,0.5,0.6,0.53333336,0.6666667,0.8666667
+PerpetualBooster (tuned),0.0,0.0,0.06666667,0.0,0.033333335,0.033333335,0.033333335,0.0,0.033333335,0.033333335,0.13333334,0.06666667,0.1,0.16666667,0.1,0.1,0.13333334,0.23333333,0.13333334,0.1,0.16666667,0.16666667,0.2,0.1,0.16666667,0.2,0.23333333,0.2,0.16666667,0.2,0.2,0.16666667,0.16666667,0.23333333,0.26666668,0.23333333,0.23333333,0.2,0.3,0.26666668,0.2,0.3,0.3,0.3,0.2,0.2,0.23333333,0.33333334,0.26666668,0.3,0.3,0.26666668,0.36666667,0.5,0.4,0.4,0.36666667,0.36666667,0.36666667,0.4,0.43333334,0.2,0.5,0.4,,0.46666667,0.5,0.56666666,0.6,0.56666666,0.53333336,0.6666667,0.8,0.6333333,0.6666667,0.6666667,0.6,0.93333334
+TorchMLP (default),0.0,0.033333335,0.06666667,0.033333335,0.0,0.06666667,0.033333335,0.0,0.06666667,0.06666667,0.033333335,0.033333335,0.033333335,0.16666667,0.1,0.06666667,0.13333334,0.2,0.06666667,0.06666667,0.13333334,0.2,0.1,0.033333335,0.16666667,0.2,0.2,0.13333334,0.2,0.13333334,0.16666667,0.2,0.1,0.3,0.16666667,0.033333335,0.2,0.1,0.33333334,0.16666667,0.06666667,0.16666667,0.36666667,0.2,0.16666667,0.26666668,0.26666668,0.36666667,0.23333333,0.3,0.16666667,0.1,0.43333334,0.33333334,0.5,0.46666667,0.33333334,0.36666667,0.3,0.43333334,0.43333334,0.5,0.5,0.6,0.53333336,,0.53333336,0.6333333,0.6333333,0.6666667,0.6666667,0.6333333,0.6666667,0.7,0.6666667,0.76666665,0.8333333,0.93333334
+FastaiMLP (default),0.0,0.0,0.033333335,0.033333335,0.0,0.033333335,0.033333335,0.06666667,0.06666667,0.033333335,0.06666667,0.06666667,0.06666667,0.033333335,0.1,0.13333334,0.13333334,0.13333334,0.13333334,0.1,0.2,0.033333335,0.16666667,0.13333334,0.13333334,0.13333334,0.16666667,0.2,0.033333335,0.16666667,0.16666667,0.16666667,0.1,0.2,0.13333334,0.033333335,0.1,0.2,0.23333333,0.2,0.13333334,0.13333334,0.3,0.2,0.16666667,0.16666667,0.16666667,0.26666668,0.033333335,0.26666668,0.2,0.23333333,0.3,0.33333334,0.4,0.4,0.33333334,0.23333333,0.3,0.4,0.23333333,0.46666667,0.5,0.6666667,0.5,0.46666667,,0.56666666,0.5,0.56666666,0.56666666,0.56666666,0.7,0.56666666,0.6666667,0.6,0.8,0.93333334
+TabFlex (default),0.0,0.06666667,0.06666667,0.1,0.0,0.13333334,0.06666667,0.0,0.13333334,0.13333334,0.16666667,0.1,0.1,0.06666667,0.1,0.16666667,0.16666667,0.2,0.2,0.16666667,0.13333334,0.1,0.2,0.1,0.16666667,0.033333335,0.23333333,0.13333334,0.16666667,0.2,0.13333334,0.2,0.16666667,0.26666668,0.1,0.06666667,0.13333334,0.16666667,0.23333333,0.16666667,0.1,0.2,0.23333333,0.13333334,0.13333334,0.13333334,0.2,0.23333333,0.16666667,0.3,0.26666668,0.1,0.23333333,0.3,0.26666668,0.3,0.3,0.23333333,0.33333334,0.33333334,0.26666668,0.4,0.3,0.43333334,0.43333334,0.36666667,0.43333334,,0.53333336,0.5,0.46666667,0.6,0.56666666,0.6,0.56666666,0.6333333,0.6666667,0.8333333
+RandomForest (default),0.033333335,0.033333335,0.06666667,0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.033333335,0.16666667,0.1,0.05,0.1,0.033333335,0.1,0.06666667,0.1,0.13333334,0.06666667,0.06666667,0.16666667,0.1,0.2,0.16666667,0.2,0.13333334,0.1,0.2,0.1,0.13333334,0.2,0.2,0.16666667,0.1,0.2,0.23333333,0.2,0.2,0.16666667,0.26666668,0.2,0.23333333,0.2,0.26666668,0.33333334,0.23333333,0.23333333,0.2,0.3,0.13333334,0.33333334,0.2,0.26666668,0.23333333,0.3,0.13333334,0.3,0.21666667,0.3,0.4,0.43333334,0.4,0.36666667,0.5,0.46666667,,0.5,0.5,0.56666666,0.53333336,0.56666666,0.6666667,0.6333333,0.6666667,0.96666664
+ExtraTrees (default),0.033333335,0.033333335,0.033333335,0.06666667,0.1,0.033333335,0.033333335,0.0,0.033333335,0.033333335,0.06666667,0.06666667,0.16666667,0.06666667,0.1,0.16666667,0.13333334,0.06666667,0.13333334,0.16666667,0.16666667,0.033333335,0.1,0.2,0.16666667,0.1,0.2,0.13333334,0.1,0.13333334,0.26666668,0.16666667,0.13333334,0.16666667,0.16666667,0.13333334,0.1,0.2,0.23333333,0.16666667,0.2,0.1,0.16666667,0.2,0.2,0.16666667,0.26666668,0.26666668,0.2,0.26666668,0.16666667,0.23333333,0.1,0.3,0.2,0.3,0.23333333,0.26666668,0.2,0.33333334,0.23333333,0.4,0.3,0.5,0.43333334,0.33333334,0.43333334,0.5,0.5,,0.43333334,0.56666666,0.53333336,0.56666666,0.6,0.56666666,0.6666667,0.93333334
+KNN (tuned + ensembled),0.06666667,0.033333335,0.033333335,0.0,0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.033333335,0.06666667,0.033333335,0.13333334,0.06666667,0.1,0.06666667,0.033333335,0.13333334,0.1,0.033333335,0.1,0.1,0.1,0.13333334,0.06666667,0.1,0.15,0.13333334,0.16666667,0.06666667,0.06666667,0.1,0.06666667,0.26666668,0.06666667,0.06666667,0.1,0.16666667,0.3,0.13333334,0.1,0.1,0.26666668,0.06666667,0.06666667,0.16666667,0.26666668,0.36666667,0.13333334,0.23333333,0.13333334,0.16666667,0.13333334,0.16666667,0.23333333,0.36666667,0.3,0.26666668,0.23333333,0.4,0.2,0.36666667,0.4,0.5,0.46666667,0.33333334,0.43333334,0.53333336,0.5,0.56666666,,0.6666667,0.53333336,0.6666667,0.7,0.7,0.96666664,0.96666664
+Linear (tuned + ensembled),0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.1,0.033333335,0.06666667,0.1,0.033333335,0.06666667,0.06666667,0.06666667,0.16666667,0.1,0.033333335,0.033333335,0.13333334,0.033333335,0.033333335,0.06666667,0.13333334,0.1,0.13333334,0.13333334,0.13333334,0.23333333,0.1,0.2,0.13333334,0.13333334,0.16666667,0.13333334,0.2,0.06666667,0.06666667,0.06666667,0.13333334,0.2,0.1,0.1,0.1,0.2,0.16666667,0.13333334,0.13333334,0.23333333,0.23333333,0.2,0.2,0.2,0.16666667,0.23333333,0.26666668,0.3,0.36666667,0.3,0.3,0.26666668,0.36666667,0.3,0.3,0.3,0.5,0.33333334,0.36666667,0.43333334,0.4,0.43333334,0.43333334,0.33333334,,0.43333334,0.76666665,0.56666666,0.6666667,0.56666666,0.8333333
+PerpetualBooster (default),0.0,0.033333335,0.06666667,0.033333335,0.033333335,0.06666667,0.033333335,0.06666667,0.06666667,0.033333335,0.033333335,0.06666667,0.06666667,0.06666667,0.06666667,0.06666667,0.13333334,0.16666667,0.06666667,0.06666667,0.16666667,0.13333334,0.16666667,0.06666667,0.16666667,0.13333334,0.13333334,0.16666667,0.1,0.16666667,0.16666667,0.13333334,0.13333334,0.13333334,0.16666667,0.06666667,0.16666667,0.13333334,0.13333334,0.23333333,0.13333334,0.13333334,0.16666667,0.2,0.13333334,0.16666667,0.2,0.16666667,0.1,0.2,0.2,0.16666667,0.26666668,0.43333334,0.3,0.36666667,0.26666668,0.26666668,0.36666667,0.36666667,0.4,0.16666667,0.33333334,0.36666667,0.2,0.33333334,0.3,0.43333334,0.46666667,0.46666667,0.46666667,0.56666666,,0.56666666,0.56666666,0.56666666,0.5,0.8666667
+Linear (tuned),0.033333335,0.033333335,0.033333335,0.033333335,0.033333335,0.06666667,0.033333335,0.06666667,0.1,0.033333335,0.05,0.06666667,0.06666667,0.16666667,0.1,0.033333335,0.033333335,0.13333334,0.033333335,0.033333335,0.06666667,0.13333334,0.06666667,0.13333334,0.1,0.1,0.2,0.1,0.16666667,0.1,0.13333334,0.1,0.1,0.2,0.06666667,0.033333335,0.033333335,0.16666667,0.2,0.1,0.1,0.1,0.2,0.13333334,0.1,0.16666667,0.23333333,0.23333333,0.16666667,0.16666667,0.2,0.16666667,0.13333334,0.26666668,0.3,0.33333334,0.3,0.3,0.26666668,0.36666667,0.3,0.3,0.3,0.5,0.36666667,0.3,0.43333334,0.4,0.43333334,0.43333334,0.33333334,0.23333333,0.43333334,,0.56666666,0.6,0.53333336,0.8
+xRFM (default),0.033333335,0.033333335,0.033333335,0.06666667,0.06666667,0.06666667,0.033333335,0.033333335,0.06666667,0.033333335,0.033333335,0.033333335,0.13333334,0.06666667,0.1,0.1,0.13333334,0.06666667,0.1,0.06666667,0.13333334,0.033333335,0.1,0.16666667,0.06666667,0.06666667,0.06666667,0.13333334,0.1,0.13333334,0.16666667,0.13333334,0.033333335,0.1,0.13333334,0.13333334,0.1,0.2,0.06666667,0.13333334,0.16666667,0.033333335,0.13333334,0.13333334,0.2,0.13333334,0.2,0.13333334,0.13333334,0.13333334,0.23333333,0.2,0.23333333,0.23333333,0.26666668,0.23333333,0.2,0.23333333,0.2,0.26666668,0.16666667,0.33333334,0.26666668,0.4,0.33333334,0.33333334,0.33333334,0.43333334,0.33333334,0.4,0.3,0.43333334,0.43333334,0.43333334,,0.43333334,0.56666666,0.8
+Linear (default),0.0,0.0,0.033333335,0.033333335,0.0,0.06666667,0.0,0.033333335,0.06666667,0.033333335,0.06666667,0.06666667,0.06666667,0.1,0.033333335,0.033333335,0.06666667,0.13333334,0.033333335,0.033333335,0.033333335,0.1,0.06666667,0.06666667,0.1,0.13333334,0.2,0.06666667,0.13333334,0.06666667,0.06666667,0.1,0.13333334,0.2,0.033333335,0.033333335,0.06666667,0.13333334,0.16666667,0.033333335,0.033333335,0.13333334,0.2,0.13333334,0.06666667,0.13333334,0.2,0.16666667,0.13333334,0.2,0.13333334,0.1,0.16666667,0.16666667,0.16666667,0.26666668,0.16666667,0.2,0.16666667,0.26666668,0.2,0.3,0.26666668,0.46666667,0.33333334,0.23333333,0.4,0.36666667,0.36666667,0.43333334,0.3,0.33333334,0.43333334,0.4,0.56666666,,0.56666666,0.73333335
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.033333335,0.0,0.0,0.033333335,0.0,0.0,0.033333335,0.0,0.0,0.0,0.0,0.0,0.13333334,0.033333335,0.0,0.033333335,0.033333335,0.06666667,0.033333335,0.06666667,0.0,0.13333334,0.06666667,0.033333335,0.06666667,0.0,0.06666667,0.0,0.2,0.033333335,0.033333335,0.033333335,0.033333335,0.16666667,0.06666667,0.033333335,0.033333335,0.16666667,0.033333335,0.033333335,0.06666667,0.13333334,0.2,0.1,0.13333334,0.033333335,0.06666667,0.1,0.13333334,0.033333335,0.16666667,0.1,0.13333334,0.1,0.16666667,0.1,0.3,0.3,0.33333334,0.4,0.16666667,0.2,0.33333334,0.33333334,0.33333334,0.033333335,0.43333334,0.5,0.46666667,0.43333334,0.43333334,,0.93333334
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.033333335,0.0,0.0,0.0,0.0,0.0,0.033333335,0.0,0.0,0.033333335,0.0,0.033333335,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.033333335,0.06666667,0.0,0.0,0.033333335,0.0,0.0,0.0,0.0,0.0,0.033333335,0.0,0.033333335,0.06666667,0.033333335,0.033333335,0.0,0.06666667,0.033333335,0.1,0.06666667,0.06666667,0.033333335,0.1,0.033333335,0.06666667,0.033333335,0.13333334,0.06666667,0.06666667,0.06666667,0.16666667,0.033333335,0.06666667,0.033333335,0.16666667,0.13333334,0.2,0.2,0.26666668,0.06666667,
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
index 59bbea002300f81136f306053edf85c170c239b9..2293fff576aa999cd9f19b1634b8503815e01338 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Binary | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..852aeaf64750ea9d2becc9465f7958f0e161c132
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_binary/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2130.0,"score":0.759,"rank":4.11,"hrank":2.26,"imp":2.931,"train_time":124.09,"predict_time":2.117,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+248\/-71","imputed":false},{"position":1,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2070.0,"score":0.726,"rank":5.22,"hrank":2.78,"imp":4.307,"train_time":310.56,"predict_time":1.824,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+216\/-59","imputed":false},{"position":2,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1972.0,"score":0.733,"rank":7.67,"hrank":2.25,"imp":5.662,"train_time":42.22,"predict_time":19.82,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+364\/-147","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1871.0,"score":0.611,"rank":11.11,"hrank":2.76,"imp":6.163,"train_time":1.19,"predict_time":0.309,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+329\/-188","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1858.0,"score":0.516,"rank":11.67,"hrank":8.24,"imp":8.445,"train_time":2.62,"predict_time":0.882,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+177\/-76","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1807.0,"score":0.475,"rank":13.89,"hrank":6.3,"imp":8.984,"train_time":0.64,"predict_time":0.201,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+232\/-143","imputed":false},{"position":6,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1746.0,"score":0.473,"rank":16.89,"hrank":9.66,"imp":7.213,"train_time":2149.06,"predict_time":6.373,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+211\/-151","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1731.0,"score":0.384,"rank":17.67,"hrank":10.07,"imp":8.547,"train_time":841.84,"predict_time":11.882,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+198\/-152","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1727.0,"score":0.449,"rank":17.89,"hrank":12.22,"imp":8.133,"train_time":1655.56,"predict_time":2.483,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-48","imputed":false},{"position":9,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1725.0,"score":0.428,"rank":18.0,"hrank":11.63,"imp":10.268,"train_time":1878.64,"predict_time":70.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+204\/-118","imputed":false},{"position":10,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1725.0,"score":0.392,"rank":18.0,"hrank":13.06,"imp":9.778,"train_time":114.48,"predict_time":0.404,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+200\/-98","imputed":false},{"position":11,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1717.0,"score":0.39,"rank":18.44,"hrank":12.03,"imp":9.401,"train_time":607.67,"predict_time":0.556,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+188\/-100","imputed":false},{"position":12,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1713.0,"score":0.437,"rank":18.67,"hrank":12.25,"imp":9.845,"train_time":92.46,"predict_time":0.366,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+183\/-128","imputed":false},{"position":13,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1700.0,"score":0.425,"rank":19.39,"hrank":9.72,"imp":9.155,"train_time":339.63,"predict_time":0.154,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+178\/-106","imputed":false},{"position":14,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1697.0,"score":0.412,"rank":19.56,"hrank":11.0,"imp":9.377,"train_time":339.63,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+196\/-118","imputed":false},{"position":15,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1691.0,"score":0.408,"rank":19.89,"hrank":14.43,"imp":10.033,"train_time":92.46,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+196\/-134","imputed":false},{"position":16,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1683.0,"score":0.43,"rank":20.33,"hrank":9.75,"imp":8.51,"train_time":1.49,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-82","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1673.0,"score":0.356,"rank":20.89,"hrank":15.69,"imp":10.891,"train_time":87.3,"predict_time":0.533,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-95","imputed":false},{"position":18,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1672.0,"score":0.325,"rank":21.0,"hrank":15.75,"imp":10.379,"train_time":841.84,"predict_time":1.263,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+185\/-109","imputed":false},{"position":19,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1661.0,"score":0.336,"rank":21.61,"hrank":13.77,"imp":11.664,"train_time":2149.06,"predict_time":0.318,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+158\/-112","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1642.0,"score":0.335,"rank":22.78,"hrank":16.49,"imp":10.113,"train_time":114.48,"predict_time":0.058,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+186\/-112","imputed":false},{"position":21,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1636.0,"score":0.346,"rank":23.11,"hrank":12.3,"imp":10.54,"train_time":1655.56,"predict_time":0.086,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+188\/-106","imputed":false},{"position":22,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1620.0,"score":0.331,"rank":24.11,"hrank":11.99,"imp":10.471,"train_time":607.67,"predict_time":0.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+231\/-130","imputed":false},{"position":23,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1602.0,"score":0.274,"rank":25.22,"hrank":21.26,"imp":11.808,"train_time":87.3,"predict_time":0.069,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+173\/-105","imputed":false},{"position":24,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1600.0,"score":0.326,"rank":25.33,"hrank":16.67,"imp":10.54,"train_time":0.48,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+191\/-142","imputed":false},{"position":25,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1579.0,"score":0.258,"rank":26.67,"hrank":16.02,"imp":10.311,"train_time":2.32,"predict_time":0.889,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+175\/-129","imputed":false},{"position":26,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1562.0,"score":0.258,"rank":27.78,"hrank":11.42,"imp":11.884,"train_time":9.96,"predict_time":4.847,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-137","imputed":false},{"position":27,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1527.0,"score":0.213,"rank":30.0,"hrank":17.22,"imp":12.466,"train_time":1878.64,"predict_time":5.934,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+193\/-112","imputed":false},{"position":28,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1510.0,"score":0.173,"rank":31.11,"hrank":24.56,"imp":12.201,"train_time":1128.19,"predict_time":1.674,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-79","imputed":false},{"position":29,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1503.0,"score":0.231,"rank":31.56,"hrank":15.91,"imp":12.578,"train_time":0.68,"predict_time":0.155,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+226\/-208","imputed":false},{"position":30,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1497.0,"score":0.255,"rank":32.0,"hrank":7.16,"imp":12.483,"train_time":0.48,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+175\/-194","imputed":false},{"position":31,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1483.0,"score":0.155,"rank":32.89,"hrank":29.17,"imp":12.872,"train_time":1.55,"predict_time":0.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+140\/-124","imputed":false},{"position":32,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1478.0,"score":0.232,"rank":33.22,"hrank":23.61,"imp":11.801,"train_time":559.89,"predict_time":1.156,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+161\/-159","imputed":false},{"position":33,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1468.0,"score":0.173,"rank":33.89,"hrank":29.63,"imp":15.118,"train_time":180.21,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+176\/-127","imputed":false},{"position":34,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1466.0,"score":0.164,"rank":34.0,"hrank":26.25,"imp":14.695,"train_time":559.89,"predict_time":0.056,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-136","imputed":false},{"position":35,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1465.0,"score":0.154,"rank":34.11,"hrank":28.25,"imp":12.591,"train_time":4560.09,"predict_time":421.951,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-130","imputed":false},{"position":36,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1461.0,"score":0.163,"rank":34.33,"hrank":23.17,"imp":14.292,"train_time":0.76,"predict_time":0.204,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+204\/-161","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1438.0,"score":0.092,"rank":35.89,"hrank":33.48,"imp":14.084,"train_time":6.55,"predict_time":0.295,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-78","imputed":false},{"position":38,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1402.0,"score":0.114,"rank":38.28,"hrank":23.51,"imp":16.054,"train_time":4560.09,"predict_time":39.452,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+191\/-177","imputed":false},{"position":39,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1399.0,"score":0.097,"rank":38.44,"hrank":32.05,"imp":14.678,"train_time":368.88,"predict_time":2.507,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-172","imputed":false},{"position":40,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1368.0,"score":0.112,"rank":40.44,"hrank":29.6,"imp":15.503,"train_time":92.9,"predict_time":0.786,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+181\/-176","imputed":false},{"position":41,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1349.0,"score":0.065,"rank":41.67,"hrank":38.09,"imp":17.837,"train_time":0.67,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+166\/-128","imputed":false},{"position":42,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1349.0,"score":0.076,"rank":41.67,"hrank":35.63,"imp":16.766,"train_time":180.21,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+197\/-144","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1339.0,"score":0.08,"rank":42.33,"hrank":38.32,"imp":15.57,"train_time":1128.19,"predict_time":0.055,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-126","imputed":false},{"position":44,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1328.0,"score":0.113,"rank":43.0,"hrank":36.47,"imp":19.25,"train_time":82.69,"predict_time":7.996,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-165","imputed":false},{"position":45,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1327.0,"score":0.015,"rank":43.11,"hrank":42.14,"imp":16.668,"train_time":43.93,"predict_time":0.204,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-74","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1305.0,"score":0.024,"rank":44.44,"hrank":41.96,"imp":16.14,"train_time":58.36,"predict_time":0.123,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-166","imputed":false},{"position":47,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1304.0,"score":0.06,"rank":44.56,"hrank":41.71,"imp":14.86,"train_time":0.34,"predict_time":0.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-141","imputed":false},{"position":48,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1282.0,"score":0.0,"rank":45.89,"hrank":45.4,"imp":17.091,"train_time":43.93,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-60","imputed":false},{"position":49,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1273.0,"score":0.034,"rank":46.44,"hrank":44.46,"imp":16.211,"train_time":9.93,"predict_time":0.224,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-131","imputed":false},{"position":50,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1271.0,"score":0.026,"rank":46.56,"hrank":42.28,"imp":18.665,"train_time":16.58,"predict_time":16.469,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-133","imputed":false},{"position":51,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1267.0,"score":0.124,"rank":46.78,"hrank":30.45,"imp":20.926,"train_time":5.13,"predict_time":2.705,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+218\/-419","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1241.0,"score":0.031,"rank":48.33,"hrank":43.17,"imp":18.59,"train_time":58.36,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+138\/-218","imputed":false},{"position":53,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1223.0,"score":0.023,"rank":49.39,"hrank":44.87,"imp":20.298,"train_time":41.8,"predict_time":39.131,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+150\/-246","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1209.0,"score":0.015,"rank":50.22,"hrank":46.93,"imp":18.638,"train_time":368.88,"predict_time":0.084,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-217","imputed":false},{"position":55,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1177.0,"score":0.013,"rank":52.0,"hrank":49.96,"imp":18.151,"train_time":1.81,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-130","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1116.0,"score":0.019,"rank":55.22,"hrank":50.63,"imp":21.386,"train_time":1.61,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-268","imputed":false},{"position":57,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1110.0,"score":0.017,"rank":55.56,"hrank":52.21,"imp":22.435,"train_time":514.21,"predict_time":52.341,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-203","imputed":false},{"position":58,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1106.0,"score":0.016,"rank":55.72,"hrank":50.33,"imp":26.623,"train_time":43.96,"predict_time":0.099,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-246","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1094.0,"score":0.0,"rank":56.33,"hrank":55.06,"imp":25.215,"train_time":3623.0,"predict_time":2.734,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-127","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1073.0,"score":0.0,"rank":57.33,"hrank":56.1,"imp":25.289,"train_time":3623.0,"predict_time":9.843,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-131","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1068.0,"score":0.002,"rank":57.61,"hrank":53.52,"imp":27.327,"train_time":43.96,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+170\/-261","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":60.67,"hrank":59.15,"imp":24.29,"train_time":0.11,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+134\/-288","imputed":false},{"position":63,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":987.0,"score":0.0,"rank":61.22,"hrank":59.95,"imp":24.224,"train_time":17.73,"predict_time":7.022,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+123\/-216","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":976.0,"score":0.0,"rank":61.67,"hrank":59.41,"imp":26.768,"train_time":20.27,"predict_time":0.572,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+149\/-210","imputed":false},{"position":65,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":934.0,"score":0.0,"rank":63.33,"hrank":62.83,"imp":25.19,"train_time":0.33,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-260","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":925.0,"score":0.0,"rank":63.67,"hrank":62.98,"imp":31.34,"train_time":99.35,"predict_time":2.573,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+52\/-135","imputed":false},{"position":67,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":922.0,"score":0.024,"rank":63.78,"hrank":54.89,"imp":32.339,"train_time":2.77,"predict_time":0.157,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+225\/-491","imputed":false},{"position":68,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":919.0,"score":0.0,"rank":63.89,"hrank":63.18,"imp":29.063,"train_time":0.22,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-222","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":906.0,"score":0.0,"rank":64.33,"hrank":62.69,"imp":27.991,"train_time":20.27,"predict_time":0.247,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-251","imputed":false},{"position":70,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":890.0,"score":0.0,"rank":64.89,"hrank":64.67,"imp":29.981,"train_time":514.21,"predict_time":2.489,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-136","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":853.0,"score":0.0,"rank":66.11,"hrank":65.39,"imp":34.192,"train_time":0.2,"predict_time":0.035,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-194","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":790.0,"score":0.0,"rank":67.89,"hrank":66.9,"imp":36.787,"train_time":1.32,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+174\/-312","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":541.0,"score":0.0,"rank":72.33,"hrank":72.3,"imp":47.009,"train_time":0.11,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-312","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/pareto_front_explorer.html
index d8811d094d34014605a5e161f3876d65094d6488..837ed32dbae4ca385aef9a64717ed0a392fc9e6f 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.7858204126,"imp":15.5034551235,"elo":1368.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.019012743,"imp":8.5100101051,"elo":1683.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0144557546,"imp":9.3767023248,"elo":1696.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1540111592,"imp":9.1553395514,"elo":1699.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0245734202,"imp":10.5404571105,"elo":1600.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.025881505,"imp":10.0327959053,"elo":1690.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"imp":9.8448535884,"elo":1712.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0055651519,"imp":17.8374386075,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0063239748,"imp":16.7662499865,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0928598604,"imp":15.1179814359,"elo":1468.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0568493697,"imp":25.190452996,"elo":933.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0146798524,"imp":18.5903487894,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1233644034,"imp":16.1401114871,"elo":1305.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"imp":21.3859848317,"elo":1116.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0840284294,"imp":18.6378211225,"elo":1208.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.5074280623,"imp":14.6778588264,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0523242217,"imp":14.8604464106,"elo":1303.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0693419145,"imp":11.8077329112,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.533338997,"imp":10.8910789344,"elo":1673.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.7046252775,"imp":20.9257711492,"elo":1267.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"imp":16.2109312274,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3182658815,"imp":11.6636815593,"elo":1661.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3731434107,"imp":7.2130516693,"elo":1745.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":7.0220909982,"imp":24.2240911066,"elo":986.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0203313112,"imp":36.787229785,"elo":790.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2469408751,"imp":27.9905565021,"elo":906.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.571547743,"imp":26.7680731711,"elo":976.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0124785185,"imp":24.2904330857,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0180054122,"imp":17.0912787012,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2037872055,"imp":16.6677683402,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2945442984,"imp":14.0840458159,"elo":1437.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0857486963,"imp":10.5402822992,"elo":1636.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.4828000197,"imp":8.1328353033,"elo":1727.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8891223145,"imp":10.3108739427,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2634274244,"imp":10.3791191378,"elo":1671.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.8819383621,"imp":8.5465175306,"elo":1731.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.4685760401,"imp":18.6653800737,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.1313967486,"imp":20.2984613004,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"imp":16.0539258759,"elo":1401.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":421.9512729692,"imp":12.5907618928,"elo":1464.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2037859869,"imp":14.2920595104,"elo":1461.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":19.8203012705,"imp":5.6624539657,"elo":1971.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.034500585,"imp":34.1923792442,"elo":852.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":4.8471187111,"imp":11.8836633452,"elo":1561.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2005539263,"imp":8.9841380587,"elo":1806.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0532307625,"imp":12.8722729759,"elo":1483.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0566464663,"imp":10.4705308008,"elo":1619.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5563493456,"imp":9.4012773101,"elo":1716.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8822174168,"imp":8.4446221801,"elo":1857.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3092961409,"imp":6.163322271,"elo":1871.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5734866142,"imp":31.3402831495,"elo":924.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.7336975574,"imp":25.2149811049,"elo":1093.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.8426087184,"imp":25.2886840599,"elo":1073.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1545280266,"imp":12.5776482388,"elo":1503.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.048205775,"imp":18.1506174111,"elo":1176.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0547828197,"imp":15.5701709413,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6737046957,"imp":12.2009644998,"elo":1510.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"imp":12.4832418523,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0584702913,"imp":10.1133985191,"elo":1641.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4036472816,"imp":9.7776638033,"elo":1724.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.9962696925,"imp":19.2502851352,"elo":1328.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":5.9343927935,"imp":12.4657984196,"elo":1527.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":70.7413293281,"imp":10.2681129547,"elo":1724.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"imp":32.3394427831,"elo":921.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0563121316,"imp":14.6952543561,"elo":1466.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.156329922,"imp":11.8014817948,"elo":1478.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.7858204126,"x_train":92.9022516918,"imp":15.5034551235,"elo":1368.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.019012743,"x_train":1.4897650506,"imp":8.5100101051,"elo":1683.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0144557546,"x_train":339.6336294327,"imp":9.3767023248,"elo":1696.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1540111592,"x_train":339.6336294327,"imp":9.1553395514,"elo":1699.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0245734202,"x_train":0.4836414313,"imp":10.5404571105,"elo":1600.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.025881505,"x_train":92.4632546012,"imp":10.0327959053,"elo":1690.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"x_train":92.4632546012,"imp":9.8448535884,"elo":1712.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0055651519,"x_train":0.6682646338,"imp":17.8374386075,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0063239748,"x_train":180.2108414054,"imp":16.7662499865,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0928598604,"x_train":180.2108414054,"imp":15.1179814359,"elo":1468.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0568493697,"x_train":0.3303836838,"imp":25.190452996,"elo":933.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0146798524,"x_train":58.3611515474,"imp":18.5903487894,"elo":1241.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1233644034,"x_train":58.3611515474,"imp":16.1401114871,"elo":1305.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"x_train":1.6147778992,"imp":21.3859848317,"elo":1116.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0840284294,"x_train":368.8801690096,"imp":18.6378211225,"elo":1208.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.5074280623,"x_train":368.8801690096,"imp":14.6778588264,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0523242217,"x_train":0.3380289469,"imp":14.8604464106,"elo":1303.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0693419145,"x_train":87.2996876502,"imp":11.8077329112,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.533338997,"x_train":87.2996876502,"imp":10.8910789344,"elo":1673.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.7046252775,"x_train":5.1314871311,"imp":20.9257711492,"elo":1267.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"x_train":9.925993635,"imp":16.2109312274,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.3182658815,"x_train":2149.0612249684,"imp":11.6636815593,"elo":1661.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3731434107,"x_train":2149.0612249684,"imp":7.2130516693,"elo":1745.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":7.0220909982,"x_train":17.7296447292,"imp":24.2240911066,"elo":986.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0203313112,"x_train":1.3180150875,"imp":36.787229785,"elo":790.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2469408751,"x_train":20.2694783344,"imp":27.9905565021,"elo":906.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.571547743,"x_train":20.2694783344,"imp":26.7680731711,"elo":976.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0124785185,"x_train":0.112647328,"imp":24.2904330857,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0180054122,"x_train":43.9333129713,"imp":17.0912787012,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2037872055,"x_train":43.9333129713,"imp":16.6677683402,"elo":1326.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2945442984,"x_train":6.549410715,"imp":14.0840458159,"elo":1437.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0857486963,"x_train":1655.5627332568,"imp":10.5402822992,"elo":1636.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.4828000197,"x_train":1655.5627332568,"imp":8.1328353033,"elo":1727.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8891223145,"x_train":2.3192637563,"imp":10.3108739427,"elo":1578.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2634274244,"x_train":841.8438595298,"imp":10.3791191378,"elo":1671.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.8819383621,"x_train":841.8438595298,"imp":8.5465175306,"elo":1731.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.4685760401,"x_train":16.5759739993,"imp":18.6653800737,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.1313967486,"x_train":41.8036776423,"imp":20.2984613004,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4515153408,"x_train":4560.0938951404,"imp":16.0539258759,"elo":1401.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":421.9512729692,"x_train":4560.0938951404,"imp":12.5907618928,"elo":1464.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2037859869,"x_train":0.755625227,"imp":14.2920595104,"elo":1461.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":19.8203012705,"x_train":42.2213895265,"imp":5.6624539657,"elo":1971.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.034500585,"x_train":0.1964259267,"imp":34.1923792442,"elo":852.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":4.8471187111,"x_train":9.9621880719,"imp":11.8836633452,"elo":1561.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2005539263,"x_train":0.6380837584,"imp":8.9841380587,"elo":1806.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0532307625,"x_train":1.5496904139,"imp":12.8722729759,"elo":1483.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0566464663,"x_train":607.6744130104,"imp":10.4705308008,"elo":1619.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5563493456,"x_train":607.6744130104,"imp":9.4012773101,"elo":1716.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8822174168,"x_train":2.6215608168,"imp":8.4446221801,"elo":1857.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3092961409,"x_train":1.1913903519,"imp":6.163322271,"elo":1871.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5734866142,"x_train":99.3538077736,"imp":31.3402831495,"elo":924.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.7336975574,"x_train":3623.0018158721,"imp":25.2149811049,"elo":1093.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.8426087184,"x_train":3623.0018158721,"imp":25.2886840599,"elo":1073.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1545280266,"x_train":0.6841801368,"imp":12.5776482388,"elo":1503.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.048205775,"x_train":1.8087875724,"imp":18.1506174111,"elo":1176.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0547828197,"x_train":1128.1946608067,"imp":15.5701709413,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6737046957,"x_train":1128.1946608067,"imp":12.2009644998,"elo":1510.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"x_train":0.4833722601,"imp":12.4832418523,"elo":1496.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0584702913,"x_train":114.4782771397,"imp":10.1133985191,"elo":1641.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4036472816,"x_train":114.4782771397,"imp":9.7776638033,"elo":1724.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.9962696925,"x_train":82.6909536839,"imp":19.2502851352,"elo":1328.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":5.9343927935,"x_train":1878.6350381184,"imp":12.4657984196,"elo":1527.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":70.7413293281,"x_train":1878.6350381184,"imp":10.2681129547,"elo":1724.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"x_train":2.766166532,"imp":32.3394427831,"elo":921.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0563121316,"x_train":559.8943943667,"imp":14.6952543561,"elo":1466.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.156329922,"x_train":559.8943943667,"imp":11.8014817948,"elo":1478.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
index 833552716815880708c40e112d839c2f1ec15bb4..576776e9d70822b60e420330b28df66b189c58fa 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_binary/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":92.9022516918,"imp":15.6304054648,"elo":1417.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.4897650506,"imp":8.6175174681,"elo":1748.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.6051687222,"imp":8.9090497492,"elo":1741.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.532113363,"imp":8.9584311887,"elo":1759.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":23.4136507344,"imp":9.0073795291,"elo":1766.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":48.9298256458,"imp":9.1405422262,"elo":1779.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":89.6438214128,"imp":9.1678098295,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":172.2716550113,"imp":9.2461851945,"elo":1769.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":253.7610304456,"imp":9.2949339601,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":339.6336294327,"imp":9.2466010383,"elo":1759.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4836414313,"imp":10.6647570108,"elo":1654.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.0174591425,"imp":10.2624047538,"elo":1731.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.3395204881,"imp":10.152082235,"elo":1748.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.5724929214,"imp":9.9439906289,"elo":1762.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.475979466,"imp":9.9069243724,"elo":1768.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.9675728072,"imp":9.7840294945,"elo":1773.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":46.0351331804,"imp":9.7880840369,"elo":1780.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":69.0484132695,"imp":9.8588736329,"elo":1782.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":92.4632546012,"imp":9.9546748098,"elo":1790.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":0.6682646338,"imp":17.9374937959,"elo":1381.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":1.7357034029,"imp":16.7485885801,"elo":1442.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":4.534589395,"imp":16.1813561541,"elo":1468.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":8.6492410666,"imp":15.9262941062,"elo":1484.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":22.0622979635,"imp":15.634017391,"elo":1503.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":45.0093118346,"imp":15.5216476661,"elo":1510.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":89.7036949426,"imp":15.3378053723,"elo":1515.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":134.7272460967,"imp":15.2818799778,"elo":1511.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":180.2108414054,"imp":15.2148659224,"elo":1512.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3303836838,"imp":25.2720625045,"elo":934.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.5821560801,"imp":22.7342112793,"elo":1018.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.5274874744,"imp":18.7683910282,"elo":1168.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.9159741389,"imp":17.0732380081,"elo":1242.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.250384706,"imp":16.4973246499,"elo":1272.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.4926603773,"imp":16.0164339132,"elo":1310.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":29.01329069,"imp":15.4801085211,"elo":1348.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.4906512987,"imp":15.4505921437,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":58.3611515474,"imp":16.2265867491,"elo":1339.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.6147778992,"imp":21.4605604621,"elo":1147.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.3105413306,"imp":19.5493514352,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.7063156272,"imp":17.2285226049,"elo":1338.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":18.3992783327,"imp":16.3771081549,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":44.132927448,"imp":15.4921196998,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":89.5658730903,"imp":15.2294822358,"elo":1426.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":182.7601700398,"imp":14.8000127545,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":274.9982902211,"imp":14.7253165202,"elo":1447.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":368.8801690096,"imp":14.7707568952,"elo":1443.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.3380289469,"imp":14.9714279867,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.8082590402,"imp":13.8770524654,"elo":1454.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":1.9749981399,"imp":12.6793074379,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":4.0486763087,"imp":11.8511537137,"elo":1666.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":10.7215076455,"imp":11.3125487279,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":21.3864383794,"imp":11.1109997449,"elo":1733.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":42.7548465236,"imp":11.0041379428,"elo":1742.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":64.660287192,"imp":10.9794984841,"elo":1746.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":87.2996876502,"imp":11.0104346726,"elo":1738.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":5.1314871311,"imp":21.0307677001,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":9.925993635,"imp":16.3078410227,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.0390242902,"imp":14.2660907298,"elo":1438.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.9509960123,"imp":11.8490302933,"elo":1614.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":113.4600835723,"imp":11.3846883486,"elo":1650.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":261.7792952647,"imp":10.929286301,"elo":1697.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":534.5397963686,"imp":10.0322866144,"elo":1736.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1054.256664045,"imp":9.1909133551,"elo":1774.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1575.3072324116,"imp":8.0985875663,"elo":1803.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2149.0612249684,"imp":7.359343456,"elo":1824.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":17.7296447292,"imp":23.650676519,"elo":1030.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.3180150875,"imp":36.7902034003,"elo":824.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.1703965898,"imp":33.4478287347,"elo":900.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":17.0551085935,"imp":27.259488321,"elo":992.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.2694783344,"imp":26.7981128702,"elo":1007.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.112647328,"imp":24.3796637993,"elo":1002.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3309222318,"imp":19.5589979146,"elo":1167.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.0383939984,"imp":17.8883505387,"elo":1250.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":2.1243719464,"imp":17.3396423226,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":5.3583380815,"imp":16.924771508,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":10.7016437794,"imp":16.7951307758,"elo":1352.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":21.6483655058,"imp":16.6996630136,"elo":1358.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":32.660208208,"imp":16.7037001297,"elo":1355.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":43.9333129713,"imp":16.7551826166,"elo":1362.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.549410715,"imp":14.1635695209,"elo":1466.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.3225140655,"imp":12.0458058992,"elo":1599.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.3256521543,"imp":10.6591161217,"elo":1686.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":79.4731674816,"imp":9.7218016261,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":207.8738621521,"imp":9.0370268671,"elo":1755.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":419.1813968527,"imp":8.6634115511,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":824.3242193836,"imp":8.4126113257,"elo":1782.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1235.2478685981,"imp":8.288629776,"elo":1797.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1655.5627332568,"imp":8.2540581451,"elo":1805.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2.3192637563,"imp":10.4442783914,"elo":1615.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.6540988882,"imp":10.3587014855,"elo":1640.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":21.4768228985,"imp":10.1758730325,"elo":1664.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":39.3241398729,"imp":10.0612192928,"elo":1683.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":89.1338776753,"imp":9.8567254649,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":188.9339441617,"imp":9.8531462616,"elo":1734.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":396.8907515036,"imp":9.4465158291,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":604.5921677407,"imp":9.186088088,"elo":1776.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":841.8438595298,"imp":8.6686696416,"elo":1790.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":16.5759739993,"imp":18.752656523,"elo":1301.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.8036776423,"imp":20.3740740053,"elo":1256.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":64.005785751,"imp":18.2998325203,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":119.9651969318,"imp":16.6080427324,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":214.9883934306,"imp":15.9206822461,"elo":1428.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":546.7601156085,"imp":14.5477934303,"elo":1476.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1120.3502125295,"imp":13.6911932241,"elo":1504.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2258.5773171118,"imp":13.2463437214,"elo":1523.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3411.9202161954,"imp":12.8657222174,"elo":1517.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4560.0938951404,"imp":12.6550523838,"elo":1521.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.755625227,"imp":14.3847318698,"elo":1511.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":42.2213895265,"imp":5.7817499917,"elo":2042.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1964259267,"imp":34.246563302,"elo":880.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":9.9621880719,"imp":12.0028644466,"elo":1617.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6380837584,"imp":9.1131354242,"elo":1863.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.5496904139,"imp":12.9775029644,"elo":1534.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":4.0437474833,"imp":12.4030184403,"elo":1570.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":13.0885329817,"imp":11.5965488724,"elo":1640.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":28.6386949065,"imp":10.8420153944,"elo":1689.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":73.9114278495,"imp":10.2149575839,"elo":1724.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":149.8736667915,"imp":9.9359304829,"elo":1742.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":297.7197032026,"imp":9.6522318264,"elo":1758.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":450.2716217096,"imp":9.5765885634,"elo":1774.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":607.6744130104,"imp":9.5315611642,"elo":1785.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.6215608168,"imp":8.5655234821,"elo":1931.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.1913903519,"imp":6.0659943447,"elo":1911.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":99.3538077736,"imp":31.3894385909,"elo":939.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":109.4145314705,"imp":31.2869146228,"elo":955.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":139.9023655107,"imp":30.974960341,"elo":964.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":205.6659012229,"imp":30.2636319035,"elo":988.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":461.8625479973,"imp":28.8724797652,"elo":1012.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":930.6476255381,"imp":27.4482185724,"elo":1053.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1843.2021584131,"imp":26.7907491687,"elo":1073.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2720.2054158222,"imp":26.0025523356,"elo":1100.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3623.0018158721,"imp":25.3453862524,"elo":1130.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.6841801368,"imp":12.7068829234,"elo":1552.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":1.8087875724,"imp":18.2232203292,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.3361236019,"imp":16.3802668296,"elo":1307.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":27.0568282205,"imp":14.6929540758,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":59.5222844332,"imp":13.8903221845,"elo":1440.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":141.9573130138,"imp":13.096623762,"elo":1480.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":277.9811204407,"imp":12.6971443003,"elo":1512.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":543.4134565854,"imp":12.4788870399,"elo":1538.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":841.3480660167,"imp":12.3353095196,"elo":1556.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1128.1946608067,"imp":12.3024351581,"elo":1562.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4833722601,"imp":12.6011537279,"elo":1551.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":1.1643639044,"imp":11.4004661213,"elo":1662.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":3.1384428828,"imp":10.8599877747,"elo":1716.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":5.5373139842,"imp":10.434780576,"elo":1752.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":14.4678529679,"imp":10.0903272106,"elo":1771.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":28.6143879135,"imp":9.9494308173,"elo":1786.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":57.062520432,"imp":9.9085698836,"elo":1794.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":85.1163434587,"imp":9.8976928184,"elo":1796.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":114.4782771397,"imp":9.9314116209,"elo":1796.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":82.6909536839,"imp":19.3351619487,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":155.5869584216,"imp":16.8717041741,"elo":1450.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":374.5843390199,"imp":13.8493671046,"elo":1561.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":733.4208142097,"imp":11.8980579202,"elo":1681.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1807.7783820047,"imp":10.1727090797,"elo":1798.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1878.6350381184,"imp":10.3583880688,"elo":1794.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.766166532,"imp":32.4029572942,"elo":943.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.6121672096,"imp":24.7720666516,"elo":1078.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":14.0207540085,"imp":16.7736184787,"elo":1314.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":27.8741073966,"imp":14.0163454027,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":69.7840899876,"imp":13.2110038325,"elo":1508.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":138.964614784,"imp":12.7230863393,"elo":1520.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":278.3697153548,"imp":12.2774774892,"elo":1527.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":417.6594178725,"imp":11.9525494205,"elo":1529.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":559.8943943667,"imp":11.5190627788,"elo":1539.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.7858204126,"x_train":92.9022516918,"imp":15.6304054648,"elo":1417.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.019012743,"x_train":1.4897650506,"imp":8.6175174681,"elo":1748.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0371172646,"x_train":3.6051687222,"imp":8.9090497492,"elo":1741.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0610065729,"x_train":12.532113363,"imp":8.9584311887,"elo":1759.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0760820847,"x_train":23.4136507344,"imp":9.0073795291,"elo":1766.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0945525499,"x_train":48.9298256458,"imp":9.1405422262,"elo":1779.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1099725948,"x_train":89.6438214128,"imp":9.1678098295,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1316544584,"x_train":172.2716550113,"imp":9.2461851945,"elo":1769.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1470337609,"x_train":253.7610304456,"imp":9.2949339601,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1540111592,"x_train":339.6336294327,"imp":9.2466010383,"elo":1759.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0245734202,"x_train":0.4836414313,"imp":10.6647570108,"elo":1654.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0497602673,"x_train":1.0174591425,"imp":10.2624047538,"elo":1731.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1045743275,"x_train":2.3395204881,"imp":10.152082235,"elo":1748.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1712030423,"x_train":4.5724929214,"imp":9.9439906289,"elo":1762.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2592282785,"x_train":11.475979466,"imp":9.9069243724,"elo":1768.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3139296897,"x_train":22.9675728072,"imp":9.7840294945,"elo":1773.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3547635864,"x_train":46.0351331804,"imp":9.7880840369,"elo":1780.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3569505185,"x_train":69.0484132695,"imp":9.8588736329,"elo":1782.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3659493967,"x_train":92.4632546012,"imp":9.9546748098,"elo":1790.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0055651519,"x_train":0.6682646338,"imp":17.9374937959,"elo":1381.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0116188579,"x_train":1.7357034029,"imp":16.7485885801,"elo":1442.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0227492458,"x_train":4.534589395,"imp":16.1813561541,"elo":1468.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0333486726,"x_train":8.6492410666,"imp":15.9262941062,"elo":1484.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.05377229,"x_train":22.0622979635,"imp":15.634017391,"elo":1503.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0655447367,"x_train":45.0093118346,"imp":15.5216476661,"elo":1510.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0784040733,"x_train":89.7036949426,"imp":15.3378053723,"elo":1515.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0899934006,"x_train":134.7272460967,"imp":15.2818799778,"elo":1511.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0923119309,"x_train":180.2108414054,"imp":15.2148659224,"elo":1512.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0568493697,"x_train":0.3303836838,"imp":25.2720625045,"elo":934.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0701976016,"x_train":0.5821560801,"imp":22.7342112793,"elo":1018.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.084787187,"x_train":1.5274874744,"imp":18.7683910282,"elo":1168.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0744434252,"x_train":2.9159741389,"imp":17.0732380081,"elo":1242.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1006473264,"x_train":7.250384706,"imp":16.4973246499,"elo":1272.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1108145286,"x_train":14.4926603773,"imp":16.0164339132,"elo":1310.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.12746224,"x_train":29.01329069,"imp":15.4801085211,"elo":1348.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1444622794,"x_train":43.4906512987,"imp":15.4505921437,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1233644034,"x_train":58.3611515474,"imp":16.2265867491,"elo":1339.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1128458591,"x_train":1.6147778992,"imp":21.4605604621,"elo":1147.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.229817567,"x_train":3.3105413306,"imp":19.5493514352,"elo":1237.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5679873118,"x_train":8.7063156272,"imp":17.2285226049,"elo":1338.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9361192036,"x_train":18.3992783327,"imp":16.3771081549,"elo":1374.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2238362098,"x_train":44.132927448,"imp":15.4921196998,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4486181927,"x_train":89.5658730903,"imp":15.2294822358,"elo":1426.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.813496387,"x_train":182.7601700398,"imp":14.8000127545,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.9956219304,"x_train":274.9982902211,"imp":14.7253165202,"elo":1447.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.3842646122,"x_train":368.8801690096,"imp":14.7707568952,"elo":1443.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0523242217,"x_train":0.3380289469,"imp":14.9714279867,"elo":1353.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.314019146,"x_train":0.8082590402,"imp":13.8770524654,"elo":1454.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.507868109,"x_train":1.9749981399,"imp":12.6793074379,"elo":1579.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6446859686,"x_train":4.0486763087,"imp":11.8511537137,"elo":1666.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7299947112,"x_train":10.7215076455,"imp":11.3125487279,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7313868749,"x_train":21.3864383794,"imp":11.1109997449,"elo":1733.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5834680373,"x_train":42.7548465236,"imp":11.0041379428,"elo":1742.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5831249824,"x_train":64.660287192,"imp":10.9794984841,"elo":1746.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.611337261,"x_train":87.2996876502,"imp":11.0104346726,"elo":1738.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":2.7046252775,"x_train":5.1314871311,"imp":21.0307677001,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2236735964,"x_train":9.925993635,"imp":16.3078410227,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5307544296,"x_train":25.0390242902,"imp":14.2660907298,"elo":1438.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.3312012868,"x_train":60.9509960123,"imp":11.8490302933,"elo":1614.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1424478891,"x_train":113.4600835723,"imp":11.3846883486,"elo":1650.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.1266315403,"x_train":261.7792952647,"imp":10.929286301,"elo":1697.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1746450553,"x_train":534.5397963686,"imp":10.0322866144,"elo":1736.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5719781161,"x_train":1054.256664045,"imp":9.1909133551,"elo":1774.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3837794763,"x_train":1575.3072324116,"imp":8.0985875663,"elo":1803.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.3731434107,"x_train":2149.0612249684,"imp":7.359343456,"elo":1824.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":7.0220909982,"x_train":17.7296447292,"imp":23.650676519,"elo":1030.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0203313112,"x_train":1.3180150875,"imp":36.7902034003,"elo":824.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1531889724,"x_train":6.1703965898,"imp":33.4478287347,"elo":900.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4887140664,"x_train":17.0551085935,"imp":27.259488321,"elo":992.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.571547743,"x_train":20.2694783344,"imp":26.7981128702,"elo":1007.0,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0124785185,"x_train":0.112647328,"imp":24.3796637993,"elo":1002.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.038943275,"x_train":0.3309222318,"imp":19.5589979146,"elo":1167.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0835714888,"x_train":1.0383939984,"imp":17.8883505387,"elo":1250.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1355199317,"x_train":2.1243719464,"imp":17.3396423226,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1452933374,"x_train":5.3583380815,"imp":16.924771508,"elo":1347.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1986293863,"x_train":10.7016437794,"imp":16.7951307758,"elo":1352.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2366835006,"x_train":21.6483655058,"imp":16.6996630136,"elo":1358.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2524117136,"x_train":32.660208208,"imp":16.7037001297,"elo":1355.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2089988036,"x_train":43.9333129713,"imp":16.7551826166,"elo":1362.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2945442984,"x_train":6.549410715,"imp":14.1635695209,"elo":1466.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5181830946,"x_train":15.3225140655,"imp":12.0458058992,"elo":1599.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.8861217734,"x_train":39.3256521543,"imp":10.6591161217,"elo":1686.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.1358686586,"x_train":79.4731674816,"imp":9.7218016261,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.3941285921,"x_train":207.8738621521,"imp":9.0370268671,"elo":1755.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.841674859,"x_train":419.1813968527,"imp":8.6634115511,"elo":1778.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.304113458,"x_train":824.3242193836,"imp":8.4126113257,"elo":1782.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.6517056168,"x_train":1235.2478685981,"imp":8.288629776,"elo":1797.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.4828000197,"x_train":1655.5627332568,"imp":8.2540581451,"elo":1805.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8891223145,"x_train":2.3192637563,"imp":10.4442783914,"elo":1615.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7570224004,"x_train":6.6540988882,"imp":10.3587014855,"elo":1640.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.5554651954,"x_train":21.4768228985,"imp":10.1758730325,"elo":1664.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.8781041597,"x_train":39.3241398729,"imp":10.0612192928,"elo":1683.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.8489456775,"x_train":89.1338776753,"imp":9.8567254649,"elo":1719.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.2088546073,"x_train":188.9339441617,"imp":9.8531462616,"elo":1734.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.4189615616,"x_train":396.8907515036,"imp":9.4465158291,"elo":1763.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.8211715812,"x_train":604.5921677407,"imp":9.186088088,"elo":1776.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.8819383621,"x_train":841.8438595298,"imp":8.6686696416,"elo":1790.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":16.4685760401,"x_train":16.5759739993,"imp":18.752656523,"elo":1301.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.1313967486,"x_train":41.8036776423,"imp":20.3740740053,"elo":1256.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":56.7537152129,"x_train":64.005785751,"imp":18.2998325203,"elo":1314.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":88.8320015192,"x_train":119.9651969318,"imp":16.6080427324,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":118.3350772121,"x_train":214.9883934306,"imp":15.9206822461,"elo":1428.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":192.322295595,"x_train":546.7601156085,"imp":14.5477934303,"elo":1476.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":288.6025785079,"x_train":1120.3502125295,"imp":13.6911932241,"elo":1504.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":387.0015785179,"x_train":2258.5773171118,"imp":13.2463437214,"elo":1523.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":401.5222034651,"x_train":3411.9202161954,"imp":12.8657222174,"elo":1517.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":415.8360217452,"x_train":4560.0938951404,"imp":12.6550523838,"elo":1521.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2037859869,"x_train":0.755625227,"imp":14.3847318698,"elo":1511.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":19.8203012705,"x_train":42.2213895265,"imp":5.7817499917,"elo":2042.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.034500585,"x_train":0.1964259267,"imp":34.246563302,"elo":880.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":4.8471187111,"x_train":9.9621880719,"imp":12.0028644466,"elo":1617.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2005539263,"x_train":0.6380837584,"imp":9.1131354242,"elo":1863.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0532307625,"x_train":1.5496904139,"imp":12.9775029644,"elo":1534.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1151376498,"x_train":4.0437474833,"imp":12.4030184403,"elo":1570.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2040125941,"x_train":13.0885329817,"imp":11.5965488724,"elo":1640.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2639855015,"x_train":28.6386949065,"imp":10.8420153944,"elo":1689.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3596014701,"x_train":73.9114278495,"imp":10.2149575839,"elo":1724.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4038726776,"x_train":149.8736667915,"imp":9.9359304829,"elo":1742.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4765747411,"x_train":297.7197032026,"imp":9.6522318264,"elo":1758.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.479443369,"x_train":450.2716217096,"imp":9.5765885634,"elo":1774.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.4521463095,"x_train":607.6744130104,"imp":9.5315611642,"elo":1785.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.8822174168,"x_train":2.6215608168,"imp":8.5655234821,"elo":1931.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3092961409,"x_train":1.1913903519,"imp":6.0659943447,"elo":1911.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5734866142,"x_train":99.3538077736,"imp":31.3894385909,"elo":939.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5734924126,"x_train":109.4145314705,"imp":31.2869146228,"elo":955.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.5646489644,"x_train":139.9023655107,"imp":30.974960341,"elo":964.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.8322499442,"x_train":205.6659012229,"imp":30.2636319035,"elo":988.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.7583190775,"x_train":461.8625479973,"imp":28.8724797652,"elo":1012.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2467243907,"x_train":930.6476255381,"imp":27.4482185724,"elo":1053.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7537273236,"x_train":1843.2021584131,"imp":26.7907491687,"elo":1073.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.0618917211,"x_train":2720.2054158222,"imp":26.0025523356,"elo":1100.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.8426087184,"x_train":3623.0018158721,"imp":25.3453862524,"elo":1130.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1545280266,"x_train":0.6841801368,"imp":12.7068829234,"elo":1552.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.048205775,"x_train":1.8087875724,"imp":18.2232203292,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1068515144,"x_train":6.3361236019,"imp":16.3802668296,"elo":1307.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2859445433,"x_train":27.0568282205,"imp":14.6929540758,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4165762351,"x_train":59.5222844332,"imp":13.8903221845,"elo":1440.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5967947147,"x_train":141.9573130138,"imp":13.096623762,"elo":1480.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8317571104,"x_train":277.9811204407,"imp":12.6971443003,"elo":1512.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1610815918,"x_train":543.4134565854,"imp":12.4788870399,"elo":1538.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.325976181,"x_train":841.3480660167,"imp":12.3353095196,"elo":1556.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6737046957,"x_train":1128.1946608067,"imp":12.3024351581,"elo":1562.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0474258719,"x_train":0.4833722601,"imp":12.6011537279,"elo":1551.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1130083333,"x_train":1.1643639044,"imp":11.4004661213,"elo":1662.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2216013349,"x_train":3.1384428828,"imp":10.8599877747,"elo":1716.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2660467618,"x_train":5.5373139842,"imp":10.434780576,"elo":1752.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3235442016,"x_train":14.4678529679,"imp":10.0903272106,"elo":1771.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3097698815,"x_train":28.6143879135,"imp":9.9494308173,"elo":1786.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3428691139,"x_train":57.062520432,"imp":9.9085698836,"elo":1794.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.384116264,"x_train":85.1163434587,"imp":9.8976928184,"elo":1796.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4036472816,"x_train":114.4782771397,"imp":9.9314116209,"elo":1796.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":7.9962696925,"x_train":82.6909536839,"imp":19.3351619487,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":17.6685334779,"x_train":155.5869584216,"imp":16.8717041741,"elo":1450.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":28.6343734044,"x_train":374.5843390199,"imp":13.8493671046,"elo":1561.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.779029417,"x_train":733.4208142097,"imp":11.8980579202,"elo":1681.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":71.1344199188,"x_train":1807.7783820047,"imp":10.1727090797,"elo":1798.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":70.7413293281,"x_train":1878.6350381184,"imp":10.3583880688,"elo":1794.0,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1572778596,"x_train":2.766166532,"imp":32.4029572942,"elo":943.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.2277922302,"x_train":5.6121672096,"imp":24.7720666516,"elo":1078.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.2497110384,"x_train":14.0207540085,"imp":16.7736184787,"elo":1314.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.3680809292,"x_train":27.8741073966,"imp":14.0163454027,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.5836572817,"x_train":69.7840899876,"imp":13.2110038325,"elo":1508.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.7285042482,"x_train":138.964614784,"imp":12.7230863393,"elo":1520.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.8340835198,"x_train":278.3697153548,"imp":12.2774774892,"elo":1527.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.0359081604,"x_train":417.6594178725,"imp":11.9525494205,"elo":1529.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.1682693315,"x_train":559.8943943667,"imp":11.5190627788,"elo":1539.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..3d98ec5728bbee7bdf971f6d95981806c1aac7ff
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_binary/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "AutoGluon 1.4 (best, 4h)", "TabFM (default)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "ModernNCA (tuned + ensembled)", "RealTabPFN-2.5 (tuned + ensembled)", "RealMLP (tuned + ensembled)", "iLTM (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabM (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "ChimeraBoost (tuned)", "CatBoost (default)", "LightGBM (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "ModernNCA (tuned)", "XGBoost (tuned)", "RealMLP (tuned)", "TabM (tuned)", "LightGBM (tuned)", "ChimeraBoost (default)", "RealTabPFN-2.5 (default)", "TabICL (default)", "iLTM (tuned)", "TorchMLP (tuned + ensembled)", "TabSwift (default)", "XGBoost (default)", "TabM (default)", "xRFM (tuned + ensembled)", "EBM (tuned + ensembled)", "xRFM (tuned)", "TabDPT (tuned + ensembled)", "TabDPT-Turbo (default)", "RealMLP (default)", "TabDPT (tuned)", "FastaiMLP (tuned + ensembled)", "BetaTabPFN (default)", "EBM (default)", "EBM (tuned)", "TorchMLP (tuned)", "iLTM (default)", "RandomForest (tuned + ensembled)", "ExtraTrees (tuned + ensembled)", "LightGBM (default)", "RandomForest (tuned)", "ModernNCA (default)", "SAP-RPT-OSS (default)", "LimiX (default)", "ExtraTrees (tuned)", "TabDPT (default)", "FastaiMLP (tuned)", "TorchMLP (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "Linear (tuned + ensembled)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "Linear (tuned)", "RandomForest (default)", "OrionMSP (default)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "TabSTAR (default)", "xRFM (default)", "Linear (default)", "PerpetualBooster (tuned)", "KNN (tuned)", "TabFlex (default)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.7777777910232544, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5555555820465088, 0.4444444477558136, NaN, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5555555820465088, 0.4444444477558136, 0.2222222238779068, NaN, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.4444444477558136, NaN, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.4444444477558136, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 1.0, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.6666666865348816, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, NaN, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 1.0, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, NaN, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, NaN, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, NaN, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 1.0, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, NaN, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.6666666865348816, 0.3333333432674408, NaN, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.4444444477558136, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.5555555820465088, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, NaN, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.5555555820465088, 1.0, 0.5555555820465088, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, NaN, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.4444444477558136, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.4444444477558136, 0.4444444477558136, 0.3888888955116272, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 1.0, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, NaN, 0.5555555820465088, 0.6666666865348816, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, NaN, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.5555555820465088, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.2222222238779068, 0.4444444477558136, 0.6111111044883728, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.8888888955116272, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 1.0, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.4444444477558136, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, NaN, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, NaN, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, NaN, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, NaN, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.5555555820465088, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, 0.5555555820465088, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.5555555820465088, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, 0.5555555820465088, NaN, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 0.6666666865348816, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.2222222238779068, 0.1111111119389534, 0.5555555820465088, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.0, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 1.0, 0.8888888955116272, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, NaN, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, NaN, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, NaN, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, NaN, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.0, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, NaN, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.8333333134651184, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, NaN, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.0, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, 0.5555555820465088, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.0, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.0, 0.0, 0.3333333432674408, 0.3333333432674408, 0.0, 0.2222222238779068, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 1.0, 1.0, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, NaN, 0.6666666865348816, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.4444444477558136, 0.7777777910232544, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.4444444477558136, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.0, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, NaN, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 1.0, 1.0, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.0, 0.3333333432674408, 0.0, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.0, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, NaN, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.8888888955116272, 1.0, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 0.8888888955116272, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, NaN, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.2222222238779068, 0.5555555820465088, NaN, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1666666716337204, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.6666666865348816, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.5555555820465088, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, NaN, 0.5555555820465088, 0.8888888955116272, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.6666666865348816, 0.8888888955116272, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.0, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.3333333432674408, 0.4444444477558136, NaN, 0.4444444477558136, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 0.8888888955116272, 1.0, 0.8888888955116272, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.5555555820465088, 0.2222222238779068, 0.1111111119389534, 0.5555555820465088, NaN, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 0.8888888955116272, 1.0], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.2222222238779068, 0.5555555820465088, NaN, 0.7777777910232544, 0.4444444477558136, 0.4444444477558136, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 0.7777777910232544, 0.8888888955116272, 0.7777777910232544, 1.0, 0.7777777910232544, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.0, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, NaN, 0.4444444477558136, 0.4444444477558136, 0.8333333134651184, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, NaN, 0.8888888955116272, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.0, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.1111111119389534, NaN, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.6666666865348816, 1.0, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544, 0.8888888955116272, 1.0, 0.7777777910232544, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.0, 0.1111111119389534, 0.0, 0.2222222238779068, 0.3333333432674408, 0.0, 0.3333333432674408, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.1666666716337204, 0.4444444477558136, 0.4444444477558136, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 0.7777777910232544, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.6666666865348816, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, NaN, 0.5555555820465088, 0.4444444477558136, 0.7777777910232544, 0.4444444477558136, 0.8888888955116272, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.5555555820465088, 0.7777777910232544, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.1111111119389534, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, NaN, 0.4444444477558136, 0.6666666865348816, 0.5555555820465088, 0.7777777910232544, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.6666666865348816, 0.6666666865348816, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, NaN, 0.5555555820465088, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, 1.0, 0.5555555820465088, 0.5555555820465088, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.3333333432674408, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, NaN, 0.3333333432674408, 0.7777777910232544, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.4444444477558136, 0.0, 0.0, 0.4444444477558136, 0.5555555820465088, 0.4444444477558136, 0.7777777910232544, 0.6666666865348816, NaN, 0.6666666865348816, 0.4444444477558136, 0.7777777910232544, 0.6666666865348816, 0.6666666865348816, 0.7777777910232544, 0.7777777910232544], [0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.3333333432674408, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.3333333432674408, NaN, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.5555555820465088, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.3333333432674408, 0.0, 0.2222222238779068, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.7777777910232544, NaN, 0.5555555820465088, 0.5555555820465088, 0.5555555820465088, 0.6666666865348816, 0.8888888955116272], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.2222222238779068, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.4444444477558136, 0.5555555820465088, 0.0, 0.4444444477558136, 0.2222222238779068, 0.5555555820465088, 0.4444444477558136, NaN, 0.2222222238779068, 0.3333333432674408, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.3333333432674408, 0.0, 0.3333333432674408, 0.1111111119389534, 0.0, 0.2222222238779068, 0.0, 0.4444444477558136, 0.1111111119389534, 0.1111111119389534, 0.4444444477558136, 0.5555555820465088, 0.5555555820465088, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.6666666865348816, 0.4444444477558136, 0.7777777910232544, NaN, 0.5555555820465088, 0.8888888955116272, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.0, 0.3333333432674408, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.2222222238779068, 0.3333333432674408, 0.0, 0.0, 0.3333333432674408, 0.4444444477558136, 0.3333333432674408, 0.4444444477558136, 0.4444444477558136, 0.3333333432674408, 0.5555555820465088, 0.4444444477558136, 0.6666666865348816, 0.4444444477558136, NaN, 0.6666666865348816, 0.7777777910232544], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.2222222238779068, 0.1111111119389534, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.2222222238779068, 0.0, 0.0, 0.1111111119389534, 0.0, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, 0.2222222238779068, 0.2222222238779068, 0.3333333432674408, 0.2222222238779068, 0.3333333432674408, 0.0, 0.2222222238779068, 0.2222222238779068, 0.4444444477558136, 0.3333333432674408, 0.1111111119389534, 0.1111111119389534, 0.3333333432674408, NaN, 0.7777777910232544], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.1111111119389534, 0.0, 0.0, 0.0, 0.0, 0.2222222238779068, 0.3333333432674408, 0.1111111119389534, 0.0, 0.0, 0.2222222238779068, 0.2222222238779068, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9573820413},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9421613425},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9086758041},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8614916359},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8538812847},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8234398865},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7823439972},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7716895065},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7686453654},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7671232971},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7671232964},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7610350156},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7579908758},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7480974204},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7458143155},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.741248105},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7351598241},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7275494755},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7260274061},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.717656021},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.7016742854},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6971080742},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6834094448},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6681887441},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.6666666746},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6484018351},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6331811346},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6027397347},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5875190335},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.581430757},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5753424738},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5631659135},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5585997035},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5494672823},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.547945213},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5464231438},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.5433790059},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5220700215},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4893455179},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4870624118},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.4596651527},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4429223824},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4429223817},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4337899595},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.4246575406},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4231354711},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4048706305},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4033485605},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3850837193},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3774733677},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3759512989},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.3729071621},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3515981787},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.3371385149},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3257229887},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3013698668},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2572298381},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2526636263},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2503805232},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2420091368},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2283105065},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2245053327},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.1826484048},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.1750380548},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1689497747},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1461187241},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1415525143},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1400304434},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1385083748},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.132420093},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1248097436},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1080669732},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0837138521},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0228310505}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..b99e981fc58a99bbc4a1cf8b80237ea40d95fc5f
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,"AutoGluon 1.5 (extreme, 4h)","AutoGluon 1.4 (best, 4h)",TabFM (default),TabPFN-3 (default),TabPFN-2.6 (default),TabICLv2 (default),ModernNCA (tuned + ensembled),RealTabPFN-2.5 (tuned + ensembled),RealMLP (tuned + ensembled),iLTM (tuned + ensembled),XGBoost (tuned + ensembled),TabM (tuned + ensembled),ChimeraBoost (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),ChimeraBoost (tuned),CatBoost (default),LightGBM (tuned + ensembled),RealTabPFN-2.5 (tuned),ModernNCA (tuned),XGBoost (tuned),RealMLP (tuned),TabM (tuned),LightGBM (tuned),ChimeraBoost (default),RealTabPFN-2.5 (default),TabICL (default),iLTM (tuned),TorchMLP (tuned + ensembled),TabSwift (default),XGBoost (default),TabM (default),xRFM (tuned + ensembled),EBM (tuned + ensembled),xRFM (tuned),TabDPT (tuned + ensembled),TabDPT-Turbo (default),RealMLP (default),TabDPT (tuned),FastaiMLP (tuned + ensembled),BetaTabPFN (default),EBM (default),EBM (tuned),TorchMLP (tuned),iLTM (default),RandomForest (tuned + ensembled),ExtraTrees (tuned + ensembled),LightGBM (default),RandomForest (tuned),ModernNCA (default),SAP-RPT-OSS (default),LimiX (default),ExtraTrees (tuned),TabDPT (default),FastaiMLP (tuned),TorchMLP (default),FastaiMLP (default),KNN (tuned + ensembled),Linear (tuned + ensembled),TabSTAR (tuned),TabSTAR (tuned + ensembled),Linear (tuned),RandomForest (default),OrionMSP (default),PerpetualBooster (tuned + ensembled),ExtraTrees (default),TabSTAR (default),xRFM (default),Linear (default),PerpetualBooster (tuned),KNN (tuned),TabFlex (default),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.5555556,0.44444445,0.44444445,0.6666667,0.6666667,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.44444445,,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.7777778,1.0,1.0,1.0,0.7777778,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,0.7777778,1.0,1.0,0.7777778,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.5555556,0.44444445,,0.7777778,0.8888889,0.7777778,0.7777778,0.8888889,0.7777778,0.6666667,0.7777778,1.0,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.8888889,1.0,0.7777778,0.6666667,0.7777778,1.0,1.0,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.5555556,0.44444445,0.22222222,,0.5555556,0.6666667,0.8888889,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.6666667,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.7777778,0.6666667,0.6666667,0.8888889,1.0,0.7777778,0.8888889,0.7777778,0.8888889,0.6666667,0.8888889,0.8888889,1.0,0.7777778,0.8888889,0.7777778,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.33333334,0.22222222,0.11111111,0.44444445,,0.6666667,0.7777778,0.6666667,0.7777778,0.5555556,0.44444445,0.7777778,0.7777778,0.6666667,0.6666667,0.7777778,0.6666667,0.8888889,0.7777778,0.6666667,0.6666667,0.6666667,0.8888889,1.0,0.6666667,0.8888889,0.8888889,0.7777778,0.7777778,0.7777778,0.8888889,1.0,0.6666667,1.0,0.7777778,1.0,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,,0.6666667,0.7777778,0.5555556,0.6666667,0.6666667,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.7777778,0.6666667,0.6666667,0.5555556,0.8888889,0.6666667,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.6666667,1.0,0.6666667,0.7777778,0.6666667,0.8888889,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,0.7777778,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.33333334,0.22222222,0.11111111,0.22222222,0.33333334,,0.6666667,0.5555556,0.5555556,0.6666667,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.6666667,0.7777778,0.6666667,0.6666667,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,0.22222222,0.11111111,0.33333334,0.33333334,0.22222222,0.33333334,,0.5555556,0.44444445,0.5555556,0.44444445,0.5555556,0.6666667,0.5555556,0.6666667,0.6666667,0.5555556,0.8888889,0.6666667,0.6666667,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.6666667,0.8888889,0.7777778,0.8888889,0.6666667,0.8888889,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,0.6666667,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.22222222,0.33333334,0.22222222,0.44444445,0.44444445,0.44444445,,0.5555556,0.33333334,0.44444445,0.33333334,0.5555556,0.5555556,0.44444445,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.5555556,0.7777778,0.5555556,0.7777778,0.8888889,0.6666667,0.7777778,0.8888889,0.7777778,0.7777778,0.8888889,0.7777778,0.7777778,1.0,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.5555556,0.44444445,,0.6666667,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.44444445,0.6666667,0.5555556,0.7777778,0.6666667,0.5555556,0.5555556,0.6666667,0.7777778,0.6666667,0.6666667,0.8888889,0.6666667,0.7777778,0.7777778,0.8888889,0.7777778,0.7777778,0.6666667,0.6666667,0.8888889,0.7777778,0.8888889,0.7777778,0.7777778,0.6666667,1.0,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.22222222,0.33333334,0.5555556,0.33333334,0.33333334,0.44444445,0.6666667,0.33333334,,0.6666667,0.44444445,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.44444445,0.7777778,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.8888889,0.5555556,0.8888889,0.8888889,0.6666667,0.8888889,0.7777778,0.7777778,0.6666667,0.8888889,0.6666667,0.7777778,0.7777778,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,0.6666667,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.11111111,0.22222222,0.0,0.22222222,0.22222222,0.44444445,0.44444445,0.5555556,0.5555556,0.44444445,0.33333334,,0.44444445,0.5555556,0.44444445,0.44444445,0.5555556,0.7777778,0.5555556,0.7777778,0.6666667,0.6666667,0.5555556,0.6666667,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,1.0,0.5555556,1.0,0.5555556,0.7777778,0.8888889,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.7777778,1.0,0.8888889,1.0,1.0,1.0,0.8888889,0.7777778,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.44444445,0.6666667,0.44444445,0.5555556,0.5555556,,0.44444445,0.44444445,0.6666667,0.5555556,0.5555556,0.6666667,0.5555556,0.5555556,0.7777778,0.5555556,0.5555556,0.7777778,0.6666667,0.7777778,0.7777778,0.7777778,0.44444445,0.7777778,0.7777778,0.8888889,0.8888889,1.0,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,0.7777778,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,0.8888889,1.0,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.44444445,0.5555556,,0.5555556,0.6666667,0.6666667,0.44444445,0.44444445,0.3888889,0.5555556,0.6666667,0.5555556,0.5555556,0.7777778,0.7777778,0.6666667,0.8888889,0.7777778,0.6666667,0.7777778,0.6666667,0.8888889,0.6666667,1.0,0.6666667,0.6666667,0.7777778,0.7777778,0.8888889,0.7777778,0.8888889,0.6666667,1.0,0.8888889,1.0,1.0,0.8888889,1.0,0.8888889,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.44444445,,0.5555556,0.6666667,0.33333334,0.44444445,0.44444445,0.6666667,0.6666667,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,0.7777778,0.8888889,0.6666667,0.6666667,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,0.6666667,1.0,0.8888889,1.0,0.8888889,0.8888889,1.0,0.8888889,1.0,0.7777778,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.5555556,0.44444445,0.44444445,0.5555556,0.33333334,0.33333334,0.44444445,,0.5555556,0.44444445,0.44444445,0.6666667,0.5555556,0.7777778,0.5555556,0.44444445,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.5555556,0.7777778,0.7777778,0.8888889,1.0,1.0,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,0.7777778,1.0,0.8888889,0.8888889,1.0,0.8888889,1.0,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.11111111,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.33333334,0.44444445,,0.44444445,0.33333334,0.44444445,0.44444445,0.6666667,0.5555556,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.6666667,0.7777778,0.6666667,0.8888889,0.7777778,0.8888889,0.6666667,0.6666667,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,0.7777778,1.0,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.44444445,0.44444445,0.44444445,0.5555556,0.33333334,0.22222222,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.7777778,0.5555556,0.5555556,0.5555556,0.7777778,0.6666667,0.5555556,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.5555556,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.7777778,1.0,1.0,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.11111111,0.0,0.22222222,0.22222222,0.11111111,0.33333334,0.11111111,0.44444445,0.33333334,0.33333334,0.44444445,0.33333334,0.5555556,0.5555556,0.5555556,0.6666667,0.5555556,,0.5555556,0.5555556,0.5555556,0.6666667,0.7777778,0.5555556,0.6666667,0.7777778,0.6666667,0.6666667,0.7777778,0.7777778,0.8888889,0.5555556,0.8888889,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.7777778,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,0.6666667,0.7777778,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.11111111,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.44444445,0.44444445,0.5555556,0.22222222,0.44444445,0.6111111,0.5555556,0.33333334,0.5555556,0.5555556,0.44444445,,0.5555556,0.5555556,0.33333334,0.5555556,0.5555556,0.6666667,0.6666667,0.5555556,0.7777778,0.7777778,0.5555556,0.6666667,0.5555556,0.6666667,0.7777778,0.7777778,0.7777778,0.7777778,1.0,0.7777778,0.8888889,0.8888889,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.44444445,0.22222222,0.22222222,0.33333334,0.44444445,0.44444445,0.33333334,0.44444445,0.5555556,0.5555556,0.44444445,0.44444445,,0.44444445,0.5555556,0.6666667,0.5555556,0.44444445,0.5555556,0.5555556,0.6666667,0.5555556,0.8888889,0.7777778,0.7777778,0.6666667,0.6666667,0.8888889,0.6666667,0.7777778,0.6666667,0.6666667,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,0.6666667,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.22222222,0.33333334,0.33333334,0.33333334,0.22222222,0.44444445,0.22222222,0.33333334,0.44444445,0.33333334,0.22222222,0.33333334,0.33333334,0.22222222,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,,0.44444445,0.5555556,0.44444445,0.6666667,0.5555556,0.7777778,0.8888889,0.5555556,0.7777778,0.6666667,0.8888889,0.6666667,0.6666667,0.7777778,0.5555556,1.0,0.7777778,0.8888889,0.7777778,0.7777778,0.7777778,1.0,0.7777778,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.11111111,0.22222222,0.0,0.11111111,0.11111111,0.44444445,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.33333334,0.6666667,0.44444445,0.5555556,,0.44444445,0.33333334,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.7777778,0.44444445,0.7777778,0.5555556,0.6666667,0.7777778,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,1.0,0.8888889,0.6666667,0.7777778,0.7777778,0.7777778,0.8888889,1.0,0.8888889,0.7777778,0.7777778,0.7777778,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.11111111,0.33333334,0.22222222,0.33333334,0.44444445,0.33333334,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,0.44444445,0.22222222,0.22222222,0.44444445,0.33333334,0.44444445,0.5555556,,0.5555556,0.5555556,0.44444445,0.5555556,0.5555556,0.33333334,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.8888889,0.7777778,1.0,1.0,1.0,0.8888889,0.7777778,0.8888889,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.11111111,0.22222222,0.33333334,0.33333334,0.22222222,0.33333334,0.44444445,0.33333334,0.33333334,0.44444445,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,0.5555556,0.6666667,0.44444445,,0.44444445,0.5555556,0.6666667,0.6666667,0.44444445,0.6666667,0.5555556,0.8888889,0.7777778,0.7777778,0.7777778,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,0.8888889,0.8888889,0.8888889,0.7777778,0.8888889,0.7777778,0.6666667,0.8888889,0.7777778,1.0,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+RealTabPFN-2.5 (default),0.0,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.22222222,0.44444445,0.33333334,0.33333334,0.5555556,0.33333334,0.44444445,0.44444445,0.5555556,,0.6666667,0.6666667,0.5555556,0.7777778,0.7777778,0.6666667,0.6666667,0.6666667,0.5555556,0.7777778,0.6666667,0.6666667,0.6666667,0.6666667,1.0,0.8888889,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,1.0,0.8888889,0.8888889,0.8888889,0.6666667,0.7777778,0.7777778,0.8888889,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0
+TabICL (default),0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.22222222,0.44444445,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.22222222,0.33333334,0.44444445,0.22222222,0.33333334,0.44444445,0.44444445,0.33333334,0.5555556,0.44444445,0.33333334,,0.44444445,0.6666667,0.44444445,0.6666667,0.5555556,0.5555556,0.5555556,0.6666667,0.8888889,0.7777778,0.6666667,0.8888889,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,0.7777778,0.7777778,1.0,1.0,0.7777778,0.7777778,0.7777778,0.8888889,0.7777778,0.8888889,1.0,0.7777778,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.11111111,0.0,0.11111111,0.11111111,0.22222222,0.22222222,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.44444445,0.44444445,0.22222222,0.44444445,0.44444445,0.33333334,0.33333334,0.5555556,,0.5555556,0.5555556,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,0.5555556,0.44444445,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.7777778,0.8888889,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.11111111,0.33333334,0.22222222,0.22222222,0.22222222,0.33333334,0.11111111,0.11111111,0.11111111,0.33333334,0.22222222,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.22222222,0.33333334,0.11111111,0.44444445,0.44444445,0.33333334,0.44444445,0.33333334,0.44444445,,0.5555556,0.5555556,0.33333334,0.5555556,0.5555556,0.5555556,0.33333334,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,0.5555556,0.6666667,0.8888889,0.7777778,0.8888889,0.8888889,0.5555556,0.8888889,0.7777778,0.8888889,0.6666667,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.11111111,0.0,0.22222222,0.11111111,0.22222222,0.22222222,0.33333334,0.11111111,0.33333334,0.33333334,0.44444445,0.22222222,0.5555556,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.22222222,0.22222222,0.44444445,0.44444445,0.33333334,0.6666667,0.5555556,0.22222222,0.5555556,0.44444445,0.44444445,,0.5555556,0.5555556,0.6666667,0.5555556,0.5555556,0.5555556,0.44444445,0.44444445,0.5555556,0.6666667,0.6666667,0.7777778,0.6666667,0.5555556,0.7777778,0.6666667,0.6666667,0.7777778,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.8888889,0.6666667,0.6666667,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.7777778,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,0.7777778,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.33333334,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.44444445,0.11111111,0.22222222,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.5555556,0.44444445,0.44444445,,0.44444445,0.6666667,0.5555556,0.6666667,0.5555556,0.5555556,0.44444445,0.6666667,0.6666667,0.5555556,0.6666667,0.6666667,0.5555556,0.8888889,0.8888889,0.8888889,0.5555556,0.8888889,0.7777778,0.8888889,0.7777778,0.8888889,0.7777778,0.6666667,0.8888889,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.11111111,0.33333334,0.22222222,0.33333334,0.22222222,0.22222222,0.44444445,0.33333334,0.44444445,0.44444445,0.6666667,0.44444445,0.5555556,,0.5555556,0.33333334,0.5555556,0.5555556,0.5555556,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.6666667,0.6666667,0.7777778,0.7777778,0.6666667,0.8888889,0.7777778,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.7777778,1.0,1.0,0.6666667,0.8888889,1.0,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.11111111,0.22222222,0.33333334,0.33333334,0.11111111,0.33333334,0.22222222,0.11111111,0.33333334,0.44444445,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.44444445,0.44444445,0.22222222,0.11111111,0.5555556,0.33333334,0.11111111,0.33333334,0.44444445,0.33333334,0.44444445,0.33333334,0.33333334,0.44444445,,0.5555556,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.7777778,0.7777778,0.5555556,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.11111111,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.0,0.11111111,0.33333334,0.22222222,0.0,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.33333334,0.22222222,0.22222222,0.22222222,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,0.6666667,0.44444445,,0.44444445,0.5555556,0.44444445,0.6666667,0.6666667,0.6666667,0.7777778,1.0,0.8888889,0.7777778,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.8888889,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+xRFM (tuned),0.0,0.0,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.11111111,0.22222222,0.22222222,0.44444445,0.0,0.0,0.11111111,0.0,0.11111111,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.44444445,0.22222222,0.22222222,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.33333334,0.44444445,0.5555556,0.5555556,,0.44444445,0.44444445,0.5555556,0.44444445,0.5555556,0.5555556,0.5555556,0.6666667,0.7777778,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.6666667,1.0,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0,1.0,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+TabDPT (tuned + ensembled),0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.11111111,0.33333334,0.44444445,0.22222222,0.22222222,0.11111111,0.22222222,0.33333334,0.33333334,0.22222222,0.22222222,0.11111111,0.44444445,0.6666667,0.44444445,0.44444445,0.44444445,0.5555556,0.44444445,0.5555556,,0.6666667,0.6666667,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.6666667,0.7777778,0.6666667,0.8888889,0.6666667,0.6666667,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,0.7777778,0.7777778,0.7777778,0.7777778,1.0,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0
+TabDPT-Turbo (default),0.0,0.11111111,0.0,0.0,0.0,0.0,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.11111111,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.22222222,0.22222222,0.22222222,0.33333334,0.44444445,0.22222222,0.22222222,0.33333334,0.33333334,0.22222222,0.5555556,0.5555556,0.5555556,0.44444445,0.44444445,0.5555556,0.5555556,0.5555556,0.33333334,,0.5555556,0.6666667,0.5555556,0.5555556,0.5555556,0.5555556,0.7777778,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.8888889,0.6666667,0.8888889,0.8888889,0.7777778,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,1.0,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.22222222,0.0,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.44444445,0.5555556,0.5555556,0.33333334,0.44444445,0.33333334,0.44444445,0.33333334,0.44444445,,0.6666667,0.44444445,0.6666667,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.5555556,0.7777778,0.7777778,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.33333334,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.11111111,0.0,0.33333334,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.11111111,0.44444445,0.33333334,0.44444445,0.33333334,0.33333334,0.44444445,0.33333334,0.5555556,0.33333334,0.33333334,0.33333334,,0.33333334,0.33333334,0.44444445,0.5555556,0.7777778,0.6666667,0.44444445,0.5555556,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.7777778,0.8333333,0.7777778,0.7777778,0.8888889,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.8888889,0.7777778,1.0,1.0,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.0,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.22222222,0.44444445,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,0.6666667,,0.44444445,0.44444445,0.5555556,0.6666667,0.44444445,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.6666667,0.6666667,0.6666667,0.8888889,0.7777778,1.0,0.8888889,0.8888889,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.22222222,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.33333334,0.0,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.22222222,0.33333334,0.22222222,0.44444445,0.44444445,0.44444445,0.33333334,0.6666667,0.5555556,,0.44444445,0.5555556,0.5555556,0.5555556,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,0.7777778,0.6666667,0.5555556,0.5555556,0.7777778,0.7777778,0.7777778,0.8888889,0.6666667,0.8888889,0.8888889,0.6666667,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.33333334,0.44444445,0.22222222,0.33333334,0.22222222,0.33333334,0.0,0.44444445,0.33333334,0.44444445,0.33333334,0.5555556,0.5555556,0.5555556,,0.5555556,0.5555556,0.5555556,0.6666667,0.5555556,0.6666667,0.6666667,0.7777778,0.7777778,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.8888889,0.6666667,0.8888889,0.8888889,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.11111111,0.0,0.0,0.33333334,0.33333334,0.0,0.22222222,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.22222222,0.11111111,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.44444445,,0.44444445,0.5555556,0.5555556,0.5555556,0.5555556,0.6666667,0.5555556,0.8888889,0.6666667,0.6666667,0.5555556,0.7777778,0.7777778,0.7777778,0.8888889,0.6666667,1.0,1.0,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,0.7777778,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.22222222,0.11111111,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.33333334,0.33333334,0.11111111,0.44444445,0.44444445,0.33333334,0.44444445,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.44444445,0.44444445,0.5555556,,0.6666667,0.44444445,0.6666667,0.5555556,0.44444445,0.7777778,0.6666667,0.5555556,0.7777778,0.5555556,0.44444445,0.8888889,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.33333334,0.0,0.22222222,0.33333334,0.22222222,0.33333334,0.22222222,0.22222222,0.11111111,0.22222222,0.33333334,0.44444445,0.44444445,0.44444445,0.33333334,0.44444445,0.33333334,0.5555556,0.44444445,0.44444445,0.44444445,0.33333334,,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889,0.7777778,0.8888889,0.8888889,0.7777778,0.7777778,0.8888889,0.7777778,0.8888889,1.0,0.8888889,0.8888889,0.7777778,0.8888889,1.0,0.7777778,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.5555556,0.33333334,0.5555556,0.33333334,0.44444445,0.5555556,0.6666667,,0.5555556,0.5555556,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.5555556,0.7777778,0.7777778,0.8888889,0.8888889,0.7777778,1.0,1.0,0.7777778,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.11111111,0.33333334,0.11111111,0.33333334,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.33333334,0.5555556,0.44444445,,0.44444445,0.5555556,0.5555556,0.44444445,0.6666667,0.6666667,0.5555556,0.5555556,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,1.0,1.0,0.7777778,1.0,0.7777778,0.8888889,1.0,0.7777778,0.8888889,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.0,0.33333334,0.0,0.11111111,0.22222222,0.44444445,0.22222222,0.44444445,0.11111111,0.44444445,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.44444445,0.33333334,0.33333334,0.33333334,0.44444445,0.44444445,0.5555556,0.44444445,0.5555556,,0.44444445,0.5555556,0.5555556,0.6666667,0.6666667,0.5555556,0.5555556,0.7777778,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.6666667,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,0.8888889,1.0,1.0,0.8888889,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.33333334,0.44444445,0.33333334,0.33333334,0.5555556,0.5555556,0.44444445,0.44444445,0.5555556,,0.5555556,0.6666667,0.6666667,0.5555556,0.44444445,0.6666667,0.7777778,0.8888889,0.8888889,0.6666667,0.8888889,0.8888889,0.6666667,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.11111111,0.22222222,0.11111111,0.33333334,0.11111111,0.44444445,0.22222222,0.44444445,0.22222222,0.44444445,0.22222222,0.44444445,0.22222222,0.44444445,0.44444445,0.44444445,0.44444445,0.44444445,,0.44444445,0.6666667,0.5555556,0.44444445,0.5555556,0.8888889,0.8888889,0.8888889,0.6666667,0.7777778,0.7777778,0.6666667,0.8888889,0.8888889,0.8888889,1.0,1.0,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.11111111,0.0,0.33333334,0.33333334,0.22222222,0.44444445,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.44444445,0.33333334,0.5555556,0.44444445,0.33333334,0.5555556,,0.6666667,0.7777778,0.5555556,0.6666667,0.5555556,0.8888889,1.0,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,1.0,0.8888889,1.0,0.8888889,1.0,1.0,0.8888889,1.0
+LimiX (default),0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.11111111,0.11111111,0.33333334,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.33333334,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.11111111,0.44444445,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,,0.44444445,0.6666667,0.5555556,0.5555556,0.6666667,0.6666667,0.6666667,0.44444445,0.44444445,0.6666667,0.7777778,0.5555556,0.6666667,0.8888889,0.6666667,0.8888889,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.8888889
+ExtraTrees (tuned),0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.22222222,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.22222222,0.0,0.11111111,0.11111111,0.22222222,0.11111111,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.22222222,0.33333334,0.44444445,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.33333334,0.44444445,0.44444445,0.22222222,0.5555556,,0.5555556,0.44444445,0.44444445,0.44444445,0.5555556,0.6666667,0.5555556,0.5555556,0.6666667,0.8888889,0.8888889,0.7777778,0.8888889,0.7777778,0.8888889,1.0,0.8888889,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.0,0.0,0.0,0.11111111,0.11111111,0.22222222,0.0,0.22222222,0.22222222,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.11111111,0.11111111,0.22222222,0.16666667,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.44444445,0.44444445,0.44444445,0.5555556,0.5555556,0.44444445,0.33333334,0.44444445,,0.5555556,0.6666667,0.7777778,0.44444445,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.6666667,0.7777778,0.5555556,0.6666667,0.7777778,0.7777778,0.8888889,0.6666667,0.8888889,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.22222222,0.22222222,0.22222222,0.5555556,0.33333334,0.22222222,0.44444445,0.44444445,0.33333334,0.44444445,0.33333334,0.44444445,0.5555556,0.44444445,,0.5555556,0.8888889,0.5555556,0.5555556,0.6666667,0.6666667,0.5555556,0.7777778,0.6666667,0.8888889,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,0.8888889,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.0,0.33333334,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.0,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.22222222,0.44444445,0.22222222,0.22222222,0.11111111,0.44444445,0.44444445,0.5555556,0.33333334,0.44444445,,0.44444445,0.7777778,0.6666667,0.6666667,0.6666667,0.7777778,0.7777778,0.7777778,0.7777778,0.7777778,0.8888889,0.8888889,0.8888889,0.8888889,1.0,0.8888889,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.11111111,0.33333334,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.11111111,0.11111111,0.11111111,0.33333334,0.5555556,0.22222222,0.11111111,0.5555556,,0.44444445,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.7777778,0.6666667,0.7777778,0.7777778,0.8888889,0.8888889,1.0
+KNN (tuned + ensembled),0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.0,0.22222222,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.22222222,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.33333334,0.44444445,0.5555556,0.44444445,0.22222222,0.5555556,,0.7777778,0.44444445,0.44444445,0.7777778,0.7777778,0.5555556,0.6666667,0.7777778,0.6666667,0.7777778,0.8888889,0.7777778,1.0,0.7777778,0.8888889,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.11111111,0.22222222,0.0,0.11111111,0.22222222,0.11111111,0.0,0.33333334,0.11111111,0.22222222,0.0,0.11111111,0.0,0.22222222,0.33333334,0.11111111,0.33333334,0.11111111,0.33333334,0.33333334,0.33333334,0.11111111,0.22222222,0.22222222,0.33333334,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.33333334,0.44444445,0.22222222,,0.44444445,0.44444445,0.8333333,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.7777778,0.7777778,0.5555556,0.5555556,0.6666667,0.6666667,0.8888889
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.22222222,0.0,0.0,0.11111111,0.0,0.11111111,0.22222222,0.22222222,0.0,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.5555556,0.44444445,0.33333334,0.33333334,0.33333334,0.44444445,0.5555556,0.5555556,,0.8888889,0.5555556,0.6666667,0.6666667,0.7777778,0.6666667,1.0,0.6666667,0.7777778,0.7777778,0.8888889,1.0,0.7777778,0.8888889
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.0,0.0,0.11111111,0.0,0.11111111,0.22222222,0.22222222,0.0,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.5555556,0.44444445,0.33333334,0.33333334,0.33333334,0.44444445,0.5555556,0.5555556,0.11111111,,0.5555556,0.6666667,0.6666667,0.7777778,0.6666667,1.0,0.6666667,0.7777778,0.7777778,0.8888889,1.0,0.7777778,0.8888889
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.22222222,0.0,0.0,0.22222222,0.11111111,0.33333334,0.0,0.11111111,0.0,0.22222222,0.33333334,0.0,0.33333334,0.11111111,0.33333334,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.11111111,0.33333334,0.33333334,0.33333334,0.22222222,0.33333334,0.33333334,0.33333334,0.44444445,0.22222222,0.44444445,0.22222222,0.16666667,0.44444445,0.44444445,,0.5555556,0.5555556,0.5555556,0.5555556,0.5555556,0.7777778,0.7777778,0.5555556,0.5555556,0.6666667,0.6666667,0.8888889
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.22222222,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.33333334,0.22222222,0.22222222,0.44444445,0.22222222,0.44444445,0.33333334,0.33333334,0.44444445,,0.5555556,0.44444445,0.7777778,0.44444445,0.8888889,0.5555556,0.5555556,0.44444445,0.5555556,0.7777778,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.22222222,0.44444445,0.11111111,0.33333334,0.33333334,0.22222222,0.33333334,0.44444445,0.44444445,0.33333334,0.33333334,0.44444445,0.44444445,,0.44444445,0.6666667,0.5555556,0.7777778,0.5555556,0.44444445,0.44444445,0.6666667,0.6666667,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.33333334,0.22222222,0.22222222,0.11111111,0.22222222,0.22222222,0.33333334,0.44444445,0.22222222,0.22222222,0.44444445,0.5555556,0.5555556,,0.5555556,0.22222222,0.5555556,0.44444445,1.0,0.5555556,0.5555556,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.22222222,0.22222222,0.22222222,0.11111111,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.11111111,0.11111111,0.44444445,0.22222222,0.22222222,0.33333334,0.22222222,0.44444445,0.33333334,0.33333334,0.44444445,0.22222222,0.33333334,0.44444445,,0.33333334,0.7777778,0.44444445,0.5555556,0.5555556,0.5555556,0.7777778,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.22222222,0.0,0.0,0.0,0.0,0.33333334,0.22222222,0.33333334,0.11111111,0.11111111,0.22222222,0.33333334,0.44444445,0.0,0.0,0.44444445,0.5555556,0.44444445,0.7777778,0.6666667,,0.6666667,0.44444445,0.7777778,0.6666667,0.6666667,0.7777778,0.7777778
+xRFM (default),0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.22222222,0.22222222,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.11111111,0.11111111,0.22222222,0.22222222,0.22222222,0.33333334,0.33333334,0.22222222,0.11111111,0.22222222,0.44444445,0.22222222,0.33333334,,0.22222222,0.44444445,0.33333334,0.44444445,0.5555556,0.6666667
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.22222222,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.11111111,0.11111111,0.11111111,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.33333334,0.0,0.22222222,0.11111111,0.11111111,0.33333334,0.11111111,0.22222222,0.22222222,0.22222222,0.22222222,0.44444445,0.44444445,0.5555556,0.5555556,0.5555556,0.7777778,,0.5555556,0.5555556,0.5555556,0.6666667,0.8888889
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.22222222,0.0,0.11111111,0.11111111,0.11111111,0.33333334,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.22222222,0.44444445,0.22222222,0.22222222,0.44444445,0.44444445,0.5555556,0.0,0.44444445,0.22222222,0.5555556,0.44444445,,0.22222222,0.33333334,0.8888889,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.0,0.0,0.0,0.33333334,0.0,0.33333334,0.11111111,0.0,0.22222222,0.0,0.44444445,0.11111111,0.11111111,0.44444445,0.5555556,0.5555556,0.44444445,0.44444445,0.33333334,0.6666667,0.44444445,0.7777778,,0.5555556,0.8888889,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.0,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.11111111,0.0,0.33333334,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.22222222,0.33333334,0.0,0.0,0.33333334,0.44444445,0.33333334,0.44444445,0.44444445,0.33333334,0.5555556,0.44444445,0.6666667,0.44444445,,0.6666667,0.7777778
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.0,0.22222222,0.11111111,0.0,0.0,0.11111111,0.0,0.11111111,0.22222222,0.0,0.0,0.11111111,0.0,0.11111111,0.11111111,0.33333334,0.22222222,0.22222222,0.33333334,0.22222222,0.33333334,0.0,0.22222222,0.22222222,0.44444445,0.33333334,0.11111111,0.11111111,0.33333334,,0.7777778
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.0,0.0,0.0,0.0,0.0,0.0,0.11111111,0.11111111,0.11111111,0.11111111,0.0,0.0,0.0,0.0,0.22222222,0.33333334,0.11111111,0.0,0.0,0.22222222,0.22222222,
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/leaderboard_overview_explorer.html
index fcd421dc52eb805d2d1476dbf1cccff72bf8511f..1893723cecf3e8572f3ce55492dfb53ccead4346 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Binary | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..6ac3052f70bbabe1bcf7b571cfd7060471ce6946
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_binary/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1772.0,"score":0.819,"rank":5.95,"hrank":1.59,"imp":4.932,"train_time":37.34,"predict_time":5.423,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+242\/-155","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1606.0,"score":0.59,"rank":12.02,"hrank":6.04,"imp":12.071,"train_time":4.66,"predict_time":0.361,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+100\/-68","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1593.0,"score":0.526,"rank":12.64,"hrank":9.28,"imp":13.651,"train_time":5.54,"predict_time":0.55,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-50","imputed":false},{"position":3,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1551.0,"score":0.524,"rank":14.9,"hrank":7.88,"imp":12.801,"train_time":1.96,"predict_time":0.151,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-91","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1544.0,"score":0.521,"rank":15.33,"hrank":7.4,"imp":13.378,"train_time":5.75,"predict_time":0.434,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-87","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1536.0,"score":0.49,"rank":15.83,"hrank":7.22,"imp":13.951,"train_time":2125.17,"predict_time":9.785,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+139\/-82","imputed":false},{"position":6,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1520.0,"score":0.521,"rank":16.81,"hrank":5.91,"imp":9.82,"train_time":369.86,"predict_time":6.202,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-104","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1512.0,"score":0.484,"rank":17.31,"hrank":8.72,"imp":14.232,"train_time":2125.17,"predict_time":0.881,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-93","imputed":false},{"position":8,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1454.0,"score":0.409,"rank":21.29,"hrank":6.7,"imp":15.06,"train_time":19.33,"predict_time":4.519,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+145\/-105","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1452.0,"score":0.365,"rank":21.43,"hrank":13.76,"imp":16.073,"train_time":2791.97,"predict_time":23.933,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-83","imputed":false},{"position":10,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1446.0,"score":0.351,"rank":21.86,"hrank":16.17,"imp":15.636,"train_time":1886.83,"predict_time":1.333,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-79","imputed":false},{"position":11,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1419.0,"score":0.349,"rank":23.88,"hrank":13.48,"imp":16.898,"train_time":3277.29,"predict_time":52.281,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+130\/-78","imputed":true},{"position":12,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1407.0,"score":0.345,"rank":24.83,"hrank":9.79,"imp":16.249,"train_time":2.4,"predict_time":0.116,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-79","imputed":false},{"position":13,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1402.0,"score":0.309,"rank":25.21,"hrank":11.85,"imp":17.154,"train_time":145.42,"predict_time":3.953,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+135\/-67","imputed":true},{"position":14,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1385.0,"score":0.324,"rank":26.6,"hrank":12.93,"imp":17.437,"train_time":2791.97,"predict_time":0.543,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-77","imputed":false},{"position":15,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1368.0,"score":0.301,"rank":28.0,"hrank":9.66,"imp":16.291,"train_time":6619.83,"predict_time":408.772,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+99\/-78","imputed":false},{"position":16,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1362.0,"score":0.285,"rank":28.52,"hrank":10.29,"imp":16.518,"train_time":7.68,"predict_time":1.811,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+72\/-81","imputed":true},{"position":17,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1347.0,"score":0.264,"rank":29.86,"hrank":19.4,"imp":19.387,"train_time":298.47,"predict_time":2.036,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-84","imputed":false},{"position":18,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1346.0,"score":0.253,"rank":29.93,"hrank":17.01,"imp":19.05,"train_time":3277.29,"predict_time":0.512,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+101\/-69","imputed":true},{"position":19,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1346.0,"score":0.286,"rank":29.95,"hrank":11.1,"imp":18.135,"train_time":2289.51,"predict_time":2.042,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-88","imputed":false},{"position":20,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1345.0,"score":0.261,"rank":30.0,"hrank":16.04,"imp":17.742,"train_time":15318.17,"predict_time":500.783,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+82\/-78","imputed":false},{"position":21,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1340.0,"score":0.246,"rank":30.4,"hrank":19.04,"imp":18.193,"train_time":1.21,"predict_time":0.052,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-79","imputed":false},{"position":22,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1337.0,"score":0.267,"rank":30.67,"hrank":16.3,"imp":17.438,"train_time":6619.83,"predict_time":39.71,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-83","imputed":false},{"position":23,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1316.0,"score":0.181,"rank":32.57,"hrank":25.49,"imp":18.768,"train_time":512.21,"predict_time":1.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-54","imputed":false},{"position":24,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1313.0,"score":0.248,"rank":32.79,"hrank":18.23,"imp":19.092,"train_time":2289.51,"predict_time":0.259,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+116\/-98","imputed":false},{"position":25,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1310.0,"score":0.209,"rank":33.05,"hrank":15.87,"imp":18.576,"train_time":1228.69,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-81","imputed":false},{"position":26,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1302.0,"score":0.172,"rank":33.76,"hrank":27.52,"imp":19.179,"train_time":3.78,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-57","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1299.0,"score":0.229,"rank":34.07,"hrank":16.58,"imp":18.645,"train_time":4618.5,"predict_time":0.426,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-85","imputed":false},{"position":28,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1298.0,"score":0.18,"rank":34.1,"hrank":26.1,"imp":19.034,"train_time":1228.69,"predict_time":0.333,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-78","imputed":false},{"position":29,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1293.0,"score":0.2,"rank":34.55,"hrank":25.24,"imp":20.167,"train_time":571.47,"predict_time":0.264,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-76","imputed":false},{"position":30,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1287.0,"score":0.212,"rank":35.14,"hrank":13.69,"imp":20.027,"train_time":4.16,"predict_time":0.543,"imputed_pct":4.76,"hardware":"GPU","elo_ci":"+101\/-88","imputed":true},{"position":31,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1284.0,"score":0.185,"rank":35.4,"hrank":22.87,"imp":19.947,"train_time":700.96,"predict_time":0.743,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-78","imputed":false},{"position":32,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1277.0,"score":0.163,"rank":36.05,"hrank":26.87,"imp":19.327,"train_time":15318.17,"predict_time":75.138,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-74","imputed":false},{"position":33,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1276.0,"score":0.126,"rank":36.1,"hrank":24.01,"imp":19.89,"train_time":1364.2,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-80","imputed":false},{"position":34,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1274.0,"score":0.142,"rank":36.24,"hrank":23.09,"imp":19.556,"train_time":11.09,"predict_time":2.46,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-64","imputed":false},{"position":35,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1259.0,"score":0.163,"rank":37.62,"hrank":12.7,"imp":20.269,"train_time":700.96,"predict_time":0.203,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-72","imputed":false},{"position":36,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1259.0,"score":0.172,"rank":37.62,"hrank":27.95,"imp":19.85,"train_time":536.34,"predict_time":1.738,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-86","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1257.0,"score":0.182,"rank":37.79,"hrank":22.12,"imp":20.277,"train_time":6.8,"predict_time":0.262,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+99\/-97","imputed":false},{"position":38,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1255.0,"score":0.162,"rank":38.0,"hrank":19.26,"imp":20.986,"train_time":571.47,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-81","imputed":false},{"position":39,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1254.0,"score":0.144,"rank":38.1,"hrank":32.59,"imp":19.978,"train_time":512.21,"predict_time":0.122,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-65","imputed":false},{"position":40,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1252.0,"score":0.206,"rank":38.31,"hrank":21.12,"imp":22.005,"train_time":8.99,"predict_time":1.724,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-110","imputed":false},{"position":41,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1250.0,"score":0.188,"rank":38.45,"hrank":25.11,"imp":20.774,"train_time":4618.5,"predict_time":9.218,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-91","imputed":false},{"position":42,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1240.0,"score":0.085,"rank":39.33,"hrank":26.32,"imp":20.364,"train_time":1364.2,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-65","imputed":false},{"position":43,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1240.0,"score":0.131,"rank":39.36,"hrank":13.07,"imp":20.752,"train_time":6.11,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-82","imputed":false},{"position":44,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1238.0,"score":0.179,"rank":39.57,"hrank":22.09,"imp":19.813,"train_time":52.04,"predict_time":39.81,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-85","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1222.0,"score":0.14,"rank":40.95,"hrank":31.79,"imp":20.361,"train_time":2535.73,"predict_time":2.546,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-83","imputed":false},{"position":46,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1214.0,"score":0.102,"rank":41.71,"hrank":35.65,"imp":21.887,"train_time":3.24,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-67","imputed":false},{"position":47,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1212.0,"score":0.118,"rank":41.86,"hrank":32.46,"imp":20.857,"train_time":536.34,"predict_time":0.097,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-91","imputed":false},{"position":48,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1193.0,"score":0.124,"rank":43.6,"hrank":11.86,"imp":23.148,"train_time":13.4,"predict_time":0.329,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-105","imputed":false},{"position":49,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1173.0,"score":0.11,"rank":45.36,"hrank":32.25,"imp":22.947,"train_time":650.76,"predict_time":5.616,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-110","imputed":false},{"position":50,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1173.0,"score":0.118,"rank":45.4,"hrank":35.65,"imp":22.226,"train_time":2535.73,"predict_time":0.112,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-86","imputed":false},{"position":51,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1166.0,"score":0.201,"rank":46.02,"hrank":12.11,"imp":23.769,"train_time":44524.42,"predict_time":4.557,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-119","imputed":false},{"position":52,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1157.0,"score":0.192,"rank":46.76,"hrank":13.59,"imp":24.006,"train_time":44524.42,"predict_time":20.33,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-130","imputed":false},{"position":53,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1146.0,"score":0.134,"rank":47.76,"hrank":35.18,"imp":24.707,"train_time":196.18,"predict_time":0.35,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-111","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1136.0,"score":0.08,"rank":48.6,"hrank":39.53,"imp":23.475,"train_time":349.07,"predict_time":0.941,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-97","imputed":false},{"position":55,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1132.0,"score":0.097,"rank":49.0,"hrank":30.0,"imp":24.112,"train_time":650.76,"predict_time":0.441,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-144","imputed":false},{"position":56,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1129.0,"score":0.091,"rank":49.21,"hrank":21.82,"imp":24.368,"train_time":335.8,"predict_time":85.392,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-121","imputed":false},{"position":57,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1126.0,"score":0.072,"rank":49.48,"hrank":40.24,"imp":23.41,"train_time":1.98,"predict_time":0.122,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-77","imputed":false},{"position":58,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1125.0,"score":0.071,"rank":49.57,"hrank":37.09,"imp":24.177,"train_time":349.07,"predict_time":0.091,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-101","imputed":false},{"position":59,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1120.0,"score":0.064,"rank":50.02,"hrank":40.9,"imp":23.879,"train_time":13.99,"predict_time":2.578,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-80","imputed":false},{"position":60,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1101.0,"score":0.065,"rank":51.6,"hrank":43.3,"imp":23.56,"train_time":2.47,"predict_time":0.168,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-95","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1096.0,"score":0.073,"rank":51.95,"hrank":46.18,"imp":25.534,"train_time":196.18,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-94","imputed":false},{"position":62,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1094.0,"score":0.159,"rank":52.14,"hrank":11.39,"imp":28.522,"train_time":439.9,"predict_time":5.445,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-148","imputed":false},{"position":63,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1091.0,"score":0.065,"rank":52.38,"hrank":43.87,"imp":25.238,"train_time":494.7,"predict_time":0.936,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-101","imputed":false},{"position":64,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1059.0,"score":0.018,"rank":54.95,"hrank":49.96,"imp":25.879,"train_time":494.7,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-100","imputed":false},{"position":65,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1046.0,"score":0.065,"rank":55.95,"hrank":43.84,"imp":27.582,"train_time":1.29,"predict_time":0.149,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+109\/-146","imputed":false},{"position":66,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1021.0,"score":0.058,"rank":57.76,"hrank":30.14,"imp":29.969,"train_time":27.03,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-141","imputed":false},{"position":67,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1018.0,"score":0.058,"rank":58.0,"hrank":44.58,"imp":27.881,"train_time":1.98,"predict_time":0.374,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-158","imputed":false},{"position":68,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1012.0,"score":0.03,"rank":58.38,"hrank":49.73,"imp":27.321,"train_time":7.76,"predict_time":0.146,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-123","imputed":false},{"position":69,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.029,"rank":59.24,"hrank":44.77,"imp":28.566,"train_time":0.39,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-150","imputed":false},{"position":70,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":991.0,"score":0.028,"rank":59.86,"hrank":55.03,"imp":26.753,"train_time":3.48,"predict_time":0.427,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-114","imputed":false},{"position":71,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":949.0,"score":0.002,"rank":62.55,"hrank":60.7,"imp":29.007,"train_time":89.64,"predict_time":1.627,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-112","imputed":false},{"position":72,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":947.0,"score":0.032,"rank":62.67,"hrank":57.25,"imp":29.67,"train_time":2.66,"predict_time":0.702,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-130","imputed":false},{"position":73,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":934.0,"score":0.048,"rank":63.4,"hrank":16.08,"imp":32.25,"train_time":280.42,"predict_time":0.382,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-151","imputed":false},{"position":74,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":914.0,"score":0.04,"rank":64.57,"hrank":26.26,"imp":32.815,"train_time":280.42,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+115\/-150","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":908.0,"score":0.014,"rank":64.9,"hrank":58.61,"imp":34.449,"train_time":1.48,"predict_time":0.127,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+109\/-165","imputed":false},{"position":76,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":811.0,"score":0.0,"rank":69.38,"hrank":68.48,"imp":34.336,"train_time":89.64,"predict_time":0.08,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-134","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":622.0,"score":0.0,"rank":74.81,"hrank":74.39,"imp":45.23,"train_time":0.24,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-167","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/pareto_front_explorer.html
index 4debfa59da23002b4251bb1ebb68982f4843bf0e..f2213181f12cbc3e8119e18d445af243e4553542 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0361740111,"imp":19.3868760265,"elo":1346.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"imp":19.17853695,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0362429023,"imp":18.5756625775,"elo":1310.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3334156391,"imp":19.0337571705,"elo":1298.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0364337543,"imp":21.8866899181,"elo":1214.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240617752,"imp":20.985618848,"elo":1255.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2636490914,"imp":20.1666997761,"elo":1293.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.016650456,"imp":20.751802223,"elo":1240.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0169525138,"imp":20.3637715539,"elo":1240.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1473350525,"imp":19.8903127907,"elo":1276.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3735107819,"imp":27.8811755796,"elo":1017.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0910270033,"imp":24.1765635522,"elo":1124.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9410076245,"imp":23.4752219736,"elo":1136.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4267781724,"imp":26.7525066226,"elo":990.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4413511276,"imp":24.1124919289,"elo":1131.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6157797575,"imp":22.9468756551,"elo":1173.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1682254962,"imp":23.5595336021,"elo":1100.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1220181584,"imp":19.9782614732,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1696974604,"imp":18.7677253558,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.5187065444,"imp":15.0596363203,"elo":1453.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.9527919841,"imp":17.1536068655,"elo":1402.4,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.329248032,"imp":23.1481533859,"elo":1193.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4255186878,"imp":18.6450742392,"elo":1298.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2181811333,"imp":20.7739214314,"elo":1250.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5777056813,"imp":23.8790218872,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0303599261,"imp":29.9691050699,"elo":1021.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1211754978,"imp":25.534048782,"elo":1096.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3495945281,"imp":24.7066660039,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0405290578,"imp":28.5662168734,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0927957297,"imp":25.8791161413,"elo":1058.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.9355489781,"imp":25.238223951,"elo":1091.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":2.4598647569,"imp":19.5562978763,"elo":1274.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5434324627,"imp":17.4370266017,"elo":1385.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":23.9328857361,"imp":16.0734923462,"elo":1451.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5499649048,"imp":13.6509122679,"elo":1593.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8814794063,"imp":14.2319420827,"elo":1511.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7853653133,"imp":13.9514322446,"elo":1535.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7241354212,"imp":22.0045036913,"elo":1251.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8096498633,"imp":19.8132653843,"elo":1237.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.7095967461,"imp":17.4380008214,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":408.7718504964,"imp":16.2907942006,"elo":1368.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1160046101,"imp":16.2494187373,"elo":1407.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.4226160049,"imp":4.9319549445,"elo":1771.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1485826053,"imp":27.5823389467,"elo":1045.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.8107806366,"imp":16.5176347481,"elo":1362.3,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1509533846,"imp":12.8010847978,"elo":1551.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2622436544,"imp":20.2772007513,"elo":1257.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.258500598,"imp":19.0919578851,"elo":1313.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.0417782615,"imp":18.1352751584,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4340112036,"imp":13.3777410485,"elo":1543.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3608180836,"imp":12.0705701187,"elo":1605.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5429839611,"imp":20.0272813282,"elo":1286.7,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5115386633,"imp":19.0496385831,"elo":1345.8,"imputed":true,"imputed_pct":4.7619047619,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":52.2806053162,"imp":16.897708857,"elo":1419.3,"imputed":true,"imputed_pct":4.7619047619,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.4449928547,"imp":28.5221350842,"elo":1093.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5569457535,"imp":23.7685782342,"elo":1165.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.3301528826,"imp":24.0056484173,"elo":1157.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0524064596,"imp":18.1929863139,"elo":1340.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1464537212,"imp":27.3205670041,"elo":1012.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1115292311,"imp":22.2260202376,"elo":1172.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.545560207,"imp":20.3612419227,"elo":1222.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"imp":23.409819345,"elo":1125.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2027324806,"imp":20.2693937617,"elo":1259.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.7428396647,"imp":19.9467652751,"elo":1283.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":85.3917069912,"imp":24.3678502547,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":75.1381347608,"imp":19.3274899746,"elo":1276.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":500.7827657461,"imp":17.7423380218,"elo":1345.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.701764822,"imp":29.6697725536,"elo":946.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0969305634,"imp":20.8570187558,"elo":1212.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.7378187563,"imp":19.8495189045,"elo":1259.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0361740111,"x_train":298.4715930877,"imp":19.3868760265,"elo":1346.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"x_train":3.776461578,"imp":19.17853695,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0362429023,"x_train":1228.6932168834,"imp":18.5756625775,"elo":1310.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3334156391,"x_train":1228.6932168834,"imp":19.0337571705,"elo":1298.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0364337543,"x_train":3.2394562483,"imp":21.8866899181,"elo":1214.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240617752,"x_train":571.4736016874,"imp":20.985618848,"elo":1255.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2636490914,"x_train":571.4736016874,"imp":20.1666997761,"elo":1293.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.016650456,"x_train":6.1072816838,"imp":20.751802223,"elo":1240.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0169525138,"x_train":1364.2004288288,"imp":20.3637715539,"elo":1240.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1473350525,"x_train":1364.2004288288,"imp":19.8903127907,"elo":1276.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3735107819,"x_train":1.983283782,"imp":27.8811755796,"elo":1017.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0910270033,"x_train":349.0651424646,"imp":24.1765635522,"elo":1124.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9410076245,"x_train":349.0651424646,"imp":23.4752219736,"elo":1136.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4267781724,"x_train":3.4776029853,"imp":26.7525066226,"elo":990.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4413511276,"x_train":650.7592051506,"imp":24.1124919289,"elo":1131.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6157797575,"x_train":650.7592051506,"imp":22.9468756551,"elo":1173.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1682254962,"x_train":2.4736320925,"imp":23.5595336021,"elo":1100.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1220181584,"x_train":512.2070722628,"imp":19.9782614732,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.1696974604,"x_train":512.2070722628,"imp":18.7677253558,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.5187065444,"x_train":19.3292393435,"imp":15.0596363203,"elo":1453.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.9527919841,"x_train":145.4200846867,"imp":17.1536068655,"elo":1402.4,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.329248032,"x_train":13.4010979517,"imp":23.1481533859,"elo":1193.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4255186878,"x_train":4618.5023417328,"imp":18.6450742392,"elo":1298.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2181811333,"x_train":4618.5023417328,"imp":20.7739214314,"elo":1250.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5777056813,"x_train":13.9928451766,"imp":23.8790218872,"elo":1119.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0303599261,"x_train":27.0284046233,"imp":29.9691050699,"elo":1021.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1211754978,"x_train":196.1786690245,"imp":25.534048782,"elo":1096.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3495945281,"x_train":196.1786690245,"imp":24.7066660039,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0405290578,"x_train":0.3945714661,"imp":28.5662168734,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0927957297,"x_train":494.6995565402,"imp":25.8791161413,"elo":1058.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.9355489781,"x_train":494.6995565402,"imp":25.238223951,"elo":1091.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":2.4598647569,"x_train":11.0934085905,"imp":19.5562978763,"elo":1274.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5434324627,"x_train":2791.9683399256,"imp":17.4370266017,"elo":1385.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":23.9328857361,"x_train":2791.9683399256,"imp":16.0734923462,"elo":1451.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5499649048,"x_train":5.5361255471,"imp":13.6509122679,"elo":1593.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8814794063,"x_train":2125.1735517383,"imp":14.2319420827,"elo":1511.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7853653133,"x_train":2125.1735517383,"imp":13.9514322446,"elo":1535.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7241354212,"x_train":8.9906883617,"imp":22.0045036913,"elo":1251.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8096498633,"x_train":52.0419884348,"imp":19.8132653843,"elo":1237.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.7095967461,"x_train":6619.8287889361,"imp":17.4380008214,"elo":1337.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":408.7718504964,"x_train":6619.8287889361,"imp":16.2907942006,"elo":1368.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1160046101,"x_train":2.397422576,"imp":16.2494187373,"elo":1407.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":5.4226160049,"x_train":37.343177154,"imp":4.9319549445,"elo":1771.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1485826053,"x_train":1.2915962934,"imp":27.5823389467,"elo":1045.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.8107806366,"x_train":7.6821993292,"imp":16.5176347481,"elo":1362.3,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1509533846,"x_train":1.9632436902,"imp":12.8010847978,"elo":1551.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2622436544,"x_train":6.8040758648,"imp":20.2772007513,"elo":1257.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.258500598,"x_train":2289.5124290727,"imp":19.0919578851,"elo":1313.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.0417782615,"x_train":2289.5124290727,"imp":18.1352751584,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4340112036,"x_train":5.7520413762,"imp":13.3777410485,"elo":1543.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3608180836,"x_train":4.6649008838,"imp":12.0705701187,"elo":1605.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5429839611,"x_train":4.1577622978,"imp":20.0272813282,"elo":1286.7,"imputed":true,"imputed_pct":4.7619047619,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5115386633,"x_train":3277.2937851341,"imp":19.0496385831,"elo":1345.8,"imputed":true,"imputed_pct":4.7619047619,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":52.2806053162,"x_train":3277.2937851341,"imp":16.897708857,"elo":1419.3,"imputed":true,"imputed_pct":4.7619047619,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.4449928547,"x_train":439.9020625591,"imp":28.5221350842,"elo":1093.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5569457535,"x_train":44524.4248051345,"imp":23.7685782342,"elo":1165.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.3301528826,"x_train":44524.4248051345,"imp":24.0056484173,"elo":1157.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0524064596,"x_train":1.2131979048,"imp":18.1929863139,"elo":1340.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1464537212,"x_train":7.7572571145,"imp":27.3205670041,"elo":1012.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1115292311,"x_train":2535.7310668162,"imp":22.2260202376,"elo":1172.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.545560207,"x_train":2535.7310668162,"imp":20.3612419227,"elo":1222.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"x_train":1.9818637578,"imp":23.409819345,"elo":1125.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2027324806,"x_train":700.9607998747,"imp":20.2693937617,"elo":1259.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.7428396647,"x_train":700.9607998747,"imp":19.9467652751,"elo":1283.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":85.3917069912,"x_train":335.8028740883,"imp":24.3678502547,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":75.1381347608,"x_train":15318.1713297129,"imp":19.3274899746,"elo":1276.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":500.7827657461,"x_train":15318.1713297129,"imp":17.7423380218,"elo":1345.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.701764822,"x_train":2.6626901865,"imp":29.6697725536,"elo":946.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0969305634,"x_train":536.3421035528,"imp":20.8570187558,"elo":1212.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.7378187563,"x_train":536.3421035528,"imp":19.8495189045,"elo":1259.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/tuning_trajectories_explorer.html
index 5cbd345657ca5a794af047d633724d50163fd8e1..1b3b6a9c97d1f0e6203f653909c69d4f7f3d27b7 100644
--- a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_binary/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":298.4715930877,"imp":19.3979753706,"elo":1367.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.776461578,"imp":19.1893758927,"elo":1318.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":11.3675639805,"imp":19.0217730518,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":27.7361311763,"imp":18.9509067544,"elo":1320.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":56.7638986603,"imp":18.9208464337,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":152.5302644531,"imp":18.95040251,"elo":1322.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":302.4049671528,"imp":18.974509686,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":611.2316361016,"imp":18.9802454449,"elo":1313.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":916.6605962686,"imp":19.0569867051,"elo":1306.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1228.6932168834,"imp":19.026926275,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.2394562483,"imp":21.8969277586,"elo":1226.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":5.9235675502,"imp":21.0997273534,"elo":1269.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":13.5672030509,"imp":20.5566691893,"elo":1295.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":26.8798672275,"imp":20.4127634054,"elo":1301.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":69.6756367181,"imp":20.1961502546,"elo":1316.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":141.1714518585,"imp":20.077881275,"elo":1321.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":284.6537506459,"imp":20.1271342047,"elo":1318.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":426.257320709,"imp":20.1080397655,"elo":1319.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":571.4736016874,"imp":20.1696842208,"elo":1313.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":6.1072816838,"imp":20.7617923433,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":13.1289398573,"imp":20.5366524152,"elo":1262.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":33.6623952718,"imp":20.2002006809,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":67.1158913983,"imp":20.1112597189,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":169.6245147956,"imp":20.0387009577,"elo":1288.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":340.6955797003,"imp":19.8901139509,"elo":1297.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":679.8713931322,"imp":19.8433938018,"elo":1297.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1018.1194397335,"imp":19.8770491292,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1364.2004288288,"imp":19.887745263,"elo":1297.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.983283782,"imp":27.8892323639,"elo":1030.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.8106834519,"imp":26.5975542265,"elo":1043.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.1319737136,"imp":24.7705090724,"elo":1084.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":17.6833597648,"imp":24.2607560584,"elo":1105.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":43.7139412379,"imp":23.9248089798,"elo":1121.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":87.1516779041,"imp":23.7384098755,"elo":1129.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":173.8653607249,"imp":23.6659234182,"elo":1134.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":260.0380199862,"imp":23.5266060366,"elo":1145.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":349.0651424646,"imp":23.5463427393,"elo":1146.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.4776029853,"imp":26.761220394,"elo":997.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.3198175557,"imp":26.0619229592,"elo":1048.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":16.004059149,"imp":24.5502376129,"elo":1112.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":32.5922652125,"imp":23.8598552738,"elo":1137.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":79.5473090208,"imp":23.1490607531,"elo":1156.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":159.3682421601,"imp":22.8680568721,"elo":1167.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":321.9650866878,"imp":22.8993239625,"elo":1168.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":484.324679178,"imp":22.8433033466,"elo":1170.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":650.7592051506,"imp":22.9732850152,"elo":1181.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.4736320925,"imp":23.5701057732,"elo":1102.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":6.1022691557,"imp":20.7916243477,"elo":1229.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":14.2595602286,"imp":19.3653317855,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":26.7224200508,"imp":18.9639422729,"elo":1319.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":64.7711179435,"imp":18.7695970506,"elo":1327.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":128.979983649,"imp":18.7558979056,"elo":1332.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":256.6446480506,"imp":18.7411633951,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":382.3781018725,"imp":18.6738707656,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":512.2070722628,"imp":18.7730371103,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":19.3292393435,"imp":15.2964098068,"elo":1457.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":145.4200846867,"imp":17.1643817419,"elo":1420.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":13.4010979517,"imp":23.1580288439,"elo":1201.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":46.9223002438,"imp":22.0242848134,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":130.5934927336,"imp":19.6488104775,"elo":1289.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":243.5354599493,"imp":19.1588604165,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":579.0165854136,"imp":18.6097679007,"elo":1333.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1164.1941811379,"imp":19.202281689,"elo":1308.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2285.7499058023,"imp":20.0286285355,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3416.2270282822,"imp":20.5553477263,"elo":1267.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4618.5023417328,"imp":20.8072436783,"elo":1261.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":13.9928451766,"imp":23.8853235829,"elo":1131.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":27.0284046233,"imp":30.0187908764,"elo":1029.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":65.0087037442,"imp":27.5753677363,"elo":1078.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":163.7081734538,"imp":24.8124340626,"elo":1150.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":196.1786690245,"imp":24.6988936633,"elo":1153.3,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3945714661,"imp":28.576760547,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.8045226314,"imp":27.3703835668,"elo":1025.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":10.2666426077,"imp":26.5255914317,"elo":1053.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":22.5942287967,"imp":26.0800284744,"elo":1068.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":59.458872306,"imp":25.5646016896,"elo":1085.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":121.2365649401,"imp":25.4591986626,"elo":1095.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":244.8550522727,"imp":25.3783315058,"elo":1099.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":368.7590966134,"imp":25.3554287809,"elo":1098.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":494.6995565402,"imp":25.2486528263,"elo":1100.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":11.0934085905,"imp":19.5669159688,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.9886873017,"imp":18.0202570333,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":67.1873109421,"imp":16.7762536741,"elo":1434.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":135.870476352,"imp":16.630372343,"elo":1440.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":350.3376696757,"imp":16.2594696186,"elo":1460.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":701.5287510764,"imp":16.1222963678,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1392.1340578984,"imp":16.1543770227,"elo":1466.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2087.8411572447,"imp":16.1396050469,"elo":1474.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2791.9683399256,"imp":16.0899176995,"elo":1477.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.5361255471,"imp":13.6624848914,"elo":1616.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.6926284807,"imp":13.7968644804,"elo":1599.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":47.4652638957,"imp":13.8514873628,"elo":1572.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":100.4462202862,"imp":13.9513057576,"elo":1564.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":257.9252062887,"imp":13.94392294,"elo":1552.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":526.5698153079,"imp":14.1122106153,"elo":1537.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1055.8495282233,"imp":13.8767450217,"elo":1543.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1596.2092361301,"imp":13.904199704,"elo":1550.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2125.1735517383,"imp":14.0091024503,"elo":1547.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":8.9906883617,"imp":22.0143246536,"elo":1265.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":52.0419884348,"imp":19.8235954602,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":82.5176074755,"imp":18.7785926283,"elo":1292.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":165.2724746986,"imp":17.8639583293,"elo":1336.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":324.4987032773,"imp":17.5518694453,"elo":1344.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":823.5187233903,"imp":16.9846559934,"elo":1353.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1657.5635357281,"imp":16.5299711329,"elo":1369.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3300.3444632835,"imp":16.33644175,"elo":1377.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":4936.7697365795,"imp":16.3233038655,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":6619.8287889361,"imp":16.3508656933,"elo":1387.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.397422576,"imp":16.2610807338,"elo":1422.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":37.343177154,"imp":4.9436417722,"elo":1782.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.2915962934,"imp":27.5915526748,"elo":1060.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":7.6821993292,"imp":16.5284270074,"elo":1376.7,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.9632436902,"imp":12.8128406793,"elo":1568.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":6.8040758648,"imp":20.287478194,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":15.0911403676,"imp":20.0772580954,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":49.7796487814,"imp":19.5610896636,"elo":1295.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":110.2249442572,"imp":19.0647786633,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":278.8447574852,"imp":18.6591202467,"elo":1343.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":562.6102815387,"imp":18.3923124877,"elo":1358.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1109.0426689601,"imp":18.2267095841,"elo":1366.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1688.1872952436,"imp":18.1673500879,"elo":1372.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2289.5124290727,"imp":18.1463782496,"elo":1371.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.7520413762,"imp":13.3897716402,"elo":1563.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":4.6649008838,"imp":12.0820626219,"elo":1622.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4.1577622978,"imp":20.0366709537,"elo":1300.4,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":25.5112428814,"imp":19.1197596076,"elo":1340.9,"imputed":true,"imputed_pct":4.7619047619,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":67.8224489171,"imp":18.2373985081,"elo":1383.2,"imputed":true,"imputed_pct":4.7619047619,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":153.1128985945,"imp":17.6358349496,"elo":1405.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":391.2360497059,"imp":16.9645508338,"elo":1438.3,"imputed":true,"imputed_pct":4.7619047619,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":783.4909131959,"imp":16.7168731489,"elo":1455.5,"imputed":true,"imputed_pct":4.7619047619,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1617.0529463655,"imp":16.695574271,"elo":1453.9,"imputed":true,"imputed_pct":4.7619047619,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2443.7577834731,"imp":16.826280855,"elo":1445.1,"imputed":true,"imputed_pct":4.7619047619,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3277.2937851341,"imp":16.9486941506,"elo":1427.5,"imputed":true,"imputed_pct":4.7619047619,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":439.9020625591,"imp":28.5296658701,"elo":1104.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":801.2621451867,"imp":28.4055628891,"elo":1109.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1413.7529870408,"imp":28.0754317457,"elo":1113.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2244.4819077532,"imp":27.3882964171,"elo":1122.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5605.4218651295,"imp":26.0594038032,"elo":1138.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11324.5526346468,"imp":24.9248224949,"elo":1149.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":22976.1739807235,"imp":24.5082464868,"elo":1155.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":33851.8786642142,"imp":24.2760256148,"elo":1164.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":44524.4248051345,"imp":24.0294343806,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.2131979048,"imp":18.2032712814,"elo":1361.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":7.7572571145,"imp":27.3121248608,"elo":1020.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":19.6428307806,"imp":24.5289373801,"elo":1093.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":58.6478087255,"imp":22.3198122025,"elo":1148.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":126.8672414768,"imp":21.5270694444,"elo":1183.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":319.7555992434,"imp":20.9384486791,"elo":1219.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":628.8366489884,"imp":20.6480184587,"elo":1231.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1222.8237025273,"imp":20.6229288971,"elo":1223.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1880.3608960104,"imp":20.4742611373,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2535.7310668162,"imp":20.3665464868,"elo":1233.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.9818637578,"imp":23.4197073149,"elo":1131.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.635768988,"imp":21.1268422346,"elo":1228.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":16.6218131925,"imp":20.5827180924,"elo":1259.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":31.8397504377,"imp":20.3270079424,"elo":1274.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.7445970232,"imp":20.1316150561,"elo":1282.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":173.5566472908,"imp":20.1137742546,"elo":1288.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":346.7594994462,"imp":20.0190755247,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":520.510860917,"imp":19.9093706074,"elo":1303.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":700.9607998747,"imp":19.9285655179,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":335.8028740883,"imp":24.3789945704,"elo":1139.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":885.3987291777,"imp":21.7706606269,"elo":1216.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2675.7253682148,"imp":19.0246143537,"elo":1300.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":5956.5440310824,"imp":18.3294919106,"elo":1329.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":14642.2890769422,"imp":17.7908735711,"elo":1360.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":15318.1713297129,"imp":17.7806875215,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.6626901865,"imp":29.677128314,"elo":960.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":5.3549384522,"imp":26.4021999621,"elo":1033.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":13.526226486,"imp":22.1394841986,"elo":1171.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":26.9163621783,"imp":20.402716463,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":67.3146528947,"imp":20.0204460728,"elo":1267.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":133.4916186547,"imp":19.9146910647,"elo":1275.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":266.8571042097,"imp":19.8724071284,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":400.0742214596,"imp":19.8953131997,"elo":1269.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":536.3421035528,"imp":19.8661260285,"elo":1268.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0361740111,"x_train":298.4715930877,"imp":19.3979753706,"elo":1367.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0247887611,"x_train":3.776461578,"imp":19.1893758927,"elo":1318.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.051816687,"x_train":11.3675639805,"imp":19.0217730518,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0997115139,"x_train":27.7361311763,"imp":18.9509067544,"elo":1320.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1513344731,"x_train":56.7638986603,"imp":18.9208464337,"elo":1324.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1887199469,"x_train":152.5302644531,"imp":18.95040251,"elo":1322.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2103626636,"x_train":302.4049671528,"imp":18.974509686,"elo":1317.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2622354828,"x_train":611.2316361016,"imp":18.9802454449,"elo":1313.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3145487088,"x_train":916.6605962686,"imp":19.0569867051,"elo":1306.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3334156391,"x_train":1228.6932168834,"imp":19.026926275,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0364337543,"x_train":3.2394562483,"imp":21.8969277586,"elo":1226.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0646948782,"x_train":5.9235675502,"imp":21.0997273534,"elo":1269.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0881631024,"x_train":13.5672030509,"imp":20.5566691893,"elo":1295.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1641604118,"x_train":26.8798672275,"imp":20.4127634054,"elo":1301.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1371985301,"x_train":69.6756367181,"imp":20.1961502546,"elo":1316.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1571534493,"x_train":141.1714518585,"imp":20.077881275,"elo":1321.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2082282212,"x_train":284.6537506459,"imp":20.1271342047,"elo":1318.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2292933372,"x_train":426.257320709,"imp":20.1080397655,"elo":1319.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2636490914,"x_train":571.4736016874,"imp":20.1696842208,"elo":1313.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.016650456,"x_train":6.1072816838,"imp":20.7617923433,"elo":1251.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0349337669,"x_train":13.1289398573,"imp":20.5366524152,"elo":1262.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0579169347,"x_train":33.6623952718,"imp":20.2002006809,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0799795713,"x_train":67.1158913983,"imp":20.1112597189,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1070850335,"x_train":169.6245147956,"imp":20.0387009577,"elo":1288.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1311981574,"x_train":340.6955797003,"imp":19.8901139509,"elo":1297.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1410621911,"x_train":679.8713931322,"imp":19.8433938018,"elo":1297.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1315950546,"x_train":1018.1194397335,"imp":19.8770491292,"elo":1299.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1473350525,"x_train":1364.2004288288,"imp":19.887745263,"elo":1297.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3735107819,"x_train":1.983283782,"imp":27.8892323639,"elo":1030.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4387177066,"x_train":3.8106834519,"imp":26.5975542265,"elo":1043.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4048503292,"x_train":9.1319737136,"imp":24.7705090724,"elo":1084.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4948762569,"x_train":17.6833597648,"imp":24.2607560584,"elo":1105.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6654595574,"x_train":43.7139412379,"imp":23.9248089798,"elo":1121.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7853524883,"x_train":87.1516779041,"imp":23.7384098755,"elo":1129.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9217284909,"x_train":173.8653607249,"imp":23.6659234182,"elo":1134.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0260469415,"x_train":260.0380199862,"imp":23.5266060366,"elo":1145.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9968010118,"x_train":349.0651424646,"imp":23.5463427393,"elo":1146.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4267781724,"x_train":3.4776029853,"imp":26.761220394,"elo":997.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8858037829,"x_train":6.3198175557,"imp":26.0619229592,"elo":1048.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2888728302,"x_train":16.004059149,"imp":24.5502376129,"elo":1112.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0867775237,"x_train":32.5922652125,"imp":23.8598552738,"elo":1137.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.369188339,"x_train":79.5473090208,"imp":23.1490607531,"elo":1156.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.8656965125,"x_train":159.3682421601,"imp":22.8680568721,"elo":1167.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.8893865347,"x_train":321.9650866878,"imp":22.8993239625,"elo":1168.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.8776147813,"x_train":484.324679178,"imp":22.8433033466,"elo":1170.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.6157797575,"x_train":650.7592051506,"imp":22.9732850152,"elo":1181.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1682254962,"x_train":2.4736320925,"imp":23.5701057732,"elo":1102.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5110863698,"x_train":6.1022691557,"imp":20.7916243477,"elo":1229.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7399382766,"x_train":14.2595602286,"imp":19.3653317855,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9104777545,"x_train":26.7224200508,"imp":18.9639422729,"elo":1319.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1034172812,"x_train":64.7711179435,"imp":18.7695970506,"elo":1327.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1824620436,"x_train":128.979983649,"imp":18.7558979056,"elo":1332.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3415381857,"x_train":256.6446480506,"imp":18.7411633951,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3940963769,"x_train":382.3781018725,"imp":18.6738707656,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2769181111,"x_train":512.2070722628,"imp":18.7730371103,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.5187065444,"x_train":19.3292393435,"imp":15.2964098068,"elo":1457.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":3.9527919841,"x_train":145.4200846867,"imp":17.1643817419,"elo":1420.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.329248032,"x_train":13.4010979517,"imp":23.1580288439,"elo":1201.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.8658037861,"x_train":46.9223002438,"imp":22.0242848134,"elo":1208.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8062580347,"x_train":130.5934927336,"imp":19.6488104775,"elo":1289.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9476753212,"x_train":243.5354599493,"imp":19.1588604165,"elo":1308.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.1594027281,"x_train":579.0165854136,"imp":18.6097679007,"elo":1333.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1909206629,"x_train":1164.1941811379,"imp":19.202281689,"elo":1308.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.7233625262,"x_train":2285.7499058023,"imp":20.0286285355,"elo":1278.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.3337898401,"x_train":3416.2270282822,"imp":20.5553477263,"elo":1267.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.2181811333,"x_train":4618.5023417328,"imp":20.8072436783,"elo":1261.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.5777056813,"x_train":13.9928451766,"imp":23.8853235829,"elo":1131.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0303599261,"x_train":27.0284046233,"imp":30.0187908764,"elo":1029.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2237719987,"x_train":65.0087037442,"imp":27.5753677363,"elo":1078.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3253204999,"x_train":163.7081734538,"imp":24.8124340626,"elo":1150.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3495945281,"x_train":196.1786690245,"imp":24.6988936633,"elo":1153.3,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0405290578,"x_train":0.3945714661,"imp":28.576760547,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1142508754,"x_train":2.8045226314,"imp":27.3703835668,"elo":1025.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2873847788,"x_train":10.2666426077,"imp":26.5255914317,"elo":1053.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4467472843,"x_train":22.5942287967,"imp":26.0800284744,"elo":1068.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6179766911,"x_train":59.458872306,"imp":25.5646016896,"elo":1085.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7267800191,"x_train":121.2365649401,"imp":25.4591986626,"elo":1095.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7605122217,"x_train":244.8550522727,"imp":25.3783315058,"elo":1099.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8870651701,"x_train":368.7590966134,"imp":25.3554287809,"elo":1098.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9355489781,"x_train":494.6995565402,"imp":25.2486528263,"elo":1100.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.4598647569,"x_train":11.0934085905,"imp":19.5669159688,"elo":1284.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.3602889236,"x_train":24.9886873017,"imp":18.0202570333,"elo":1363.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.2554138788,"x_train":67.1873109421,"imp":16.7762536741,"elo":1434.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.1157993328,"x_train":135.870476352,"imp":16.630372343,"elo":1440.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.7725018708,"x_train":350.3376696757,"imp":16.2594696186,"elo":1460.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.6120198051,"x_train":701.5287510764,"imp":16.1222963678,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.3758571791,"x_train":1392.1340578984,"imp":16.1543770227,"elo":1466.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.5072646854,"x_train":2087.8411572447,"imp":16.1396050469,"elo":1474.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":23.9328857361,"x_train":2791.9683399256,"imp":16.0899176995,"elo":1477.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5499649048,"x_train":5.5361255471,"imp":13.6624848914,"elo":1616.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3877879479,"x_train":16.6926284807,"imp":13.7968644804,"elo":1599.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.0362039712,"x_train":47.4652638957,"imp":13.8514873628,"elo":1572.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1588746477,"x_train":100.4462202862,"imp":13.9513057576,"elo":1564.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.6152050014,"x_train":257.9252062887,"imp":13.94392294,"elo":1552.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.2170693651,"x_train":526.5698153079,"imp":14.1122106153,"elo":1537.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.1535813458,"x_train":1055.8495282233,"imp":13.8767450217,"elo":1543.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0359551071,"x_train":1596.2092361301,"imp":13.904199704,"elo":1550.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.7107599974,"x_train":2125.1735517383,"imp":14.0091024503,"elo":1547.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7241354212,"x_train":8.9906883617,"imp":22.0143246536,"elo":1265.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8096498633,"x_train":52.0419884348,"imp":19.8235954602,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":58.7248120485,"x_train":82.5176074755,"imp":18.7785926283,"elo":1292.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":90.6781979203,"x_train":165.2724746986,"imp":17.8639583293,"elo":1336.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":125.590395152,"x_train":324.4987032773,"imp":17.5518694453,"elo":1344.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":191.6133848691,"x_train":823.5187233903,"imp":16.9846559934,"elo":1353.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":254.4244251102,"x_train":1657.5635357281,"imp":16.5299711329,"elo":1369.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":323.4247300441,"x_train":3300.3444632835,"imp":16.33644175,"elo":1377.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":360.6190160112,"x_train":4936.7697365795,"imp":16.3233038655,"elo":1382.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":408.7718504964,"x_train":6619.8287889361,"imp":16.3508656933,"elo":1387.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1160046101,"x_train":2.397422576,"imp":16.2610807338,"elo":1422.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":5.4226160049,"x_train":37.343177154,"imp":4.9436417722,"elo":1782.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1485826053,"x_train":1.2915962934,"imp":27.5915526748,"elo":1060.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.8107806366,"x_train":7.6821993292,"imp":16.5284270074,"elo":1376.7,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1509533846,"x_train":1.9632436902,"imp":12.8128406793,"elo":1568.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2622436544,"x_train":6.8040758648,"imp":20.287478194,"elo":1272.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4930239494,"x_train":15.0911403676,"imp":20.0772580954,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9344890524,"x_train":49.7796487814,"imp":19.5610896636,"elo":1295.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.1910121728,"x_train":110.2249442572,"imp":19.0647786633,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.7229810256,"x_train":278.8447574852,"imp":18.6591202467,"elo":1343.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.8482256852,"x_train":562.6102815387,"imp":18.3923124877,"elo":1358.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":2.0517870766,"x_train":1109.0426689601,"imp":18.2267095841,"elo":1366.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.108383211,"x_train":1688.1872952436,"imp":18.1673500879,"elo":1372.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.0417782615,"x_train":2289.5124290727,"imp":18.1463782496,"elo":1371.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.4340112036,"x_train":5.7520413762,"imp":13.3897716402,"elo":1563.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3608180836,"x_train":4.6649008838,"imp":12.0820626219,"elo":1622.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5429839611,"x_train":4.1577622978,"imp":20.0366709537,"elo":1300.4,"imputed":true,"imputed_pct":4.7619047619,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.1730533624,"x_train":25.5112428814,"imp":19.1197596076,"elo":1340.9,"imputed":true,"imputed_pct":4.7619047619,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.6579367369,"x_train":67.8224489171,"imp":18.2373985081,"elo":1383.2,"imputed":true,"imputed_pct":4.7619047619,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":15.8511539698,"x_train":153.1128985945,"imp":17.6358349496,"elo":1405.0,"imputed":true,"imputed_pct":4.7619047619,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":24.461206162,"x_train":391.2360497059,"imp":16.9645508338,"elo":1438.3,"imputed":true,"imputed_pct":4.7619047619,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":32.2082359314,"x_train":783.4909131959,"imp":16.7168731489,"elo":1455.5,"imputed":true,"imputed_pct":4.7619047619,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":42.5602613211,"x_train":1617.0529463655,"imp":16.695574271,"elo":1453.9,"imputed":true,"imputed_pct":4.7619047619,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":48.9306371927,"x_train":2443.7577834731,"imp":16.826280855,"elo":1445.1,"imputed":true,"imputed_pct":4.7619047619,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":52.1193903923,"x_train":3277.2937851341,"imp":16.9486941506,"elo":1427.5,"imputed":true,"imputed_pct":4.7619047619,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.4449928547,"x_train":439.9020625591,"imp":28.5296658701,"elo":1104.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7828189872,"x_train":801.2621451867,"imp":28.4055628891,"elo":1109.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.6843898688,"x_train":1413.7529870408,"imp":28.0754317457,"elo":1113.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5679815934,"x_train":2244.4819077532,"imp":27.3882964171,"elo":1122.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.9965249174,"x_train":5605.4218651295,"imp":26.0594038032,"elo":1138.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.7064222686,"x_train":11324.5526346468,"imp":24.9248224949,"elo":1149.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.8749951615,"x_train":22976.1739807235,"imp":24.5082464868,"elo":1155.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":23.2068554068,"x_train":33851.8786642142,"imp":24.2760256148,"elo":1164.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.3301528826,"x_train":44524.4248051345,"imp":24.0294343806,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0524064596,"x_train":1.2131979048,"imp":18.2032712814,"elo":1361.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1464537212,"x_train":7.7572571145,"imp":27.3121248608,"elo":1020.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2861450331,"x_train":19.6428307806,"imp":24.5289373801,"elo":1093.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7112520661,"x_train":58.6478087255,"imp":22.3198122025,"elo":1148.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0077171512,"x_train":126.8672414768,"imp":21.5270694444,"elo":1183.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2404297158,"x_train":319.7555992434,"imp":20.9384486791,"elo":1219.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6141725337,"x_train":628.8366489884,"imp":20.6480184587,"elo":1231.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1193572199,"x_train":1222.8237025273,"imp":20.6229288971,"elo":1223.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.2621370554,"x_train":1880.3608960104,"imp":20.4742611373,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.545560207,"x_train":2535.7310668162,"imp":20.3665464868,"elo":1233.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1218385777,"x_train":1.9818637578,"imp":23.4197073149,"elo":1131.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.465082175,"x_train":5.635768988,"imp":21.1268422346,"elo":1228.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5859087742,"x_train":16.6218131925,"imp":20.5827180924,"elo":1259.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6695677292,"x_train":31.8397504377,"imp":20.3270079424,"elo":1274.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7535860957,"x_train":86.7445970232,"imp":20.1316150561,"elo":1282.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8503171682,"x_train":173.5566472908,"imp":20.1137742546,"elo":1288.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9856110848,"x_train":346.7594994462,"imp":20.0190755247,"elo":1297.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.238113448,"x_train":520.510860917,"imp":19.9093706074,"elo":1303.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.7548289838,"x_train":700.9607998747,"imp":19.9285655179,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":85.3917069912,"x_train":335.8028740883,"imp":24.3789945704,"elo":1139.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":175.3477427959,"x_train":885.3987291777,"imp":21.7706606269,"elo":1216.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":278.6108939781,"x_train":2675.7253682148,"imp":19.0246143537,"elo":1300.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":396.3850032557,"x_train":5956.5440310824,"imp":18.3294919106,"elo":1329.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":478.1228243423,"x_train":14642.2890769422,"imp":17.7908735711,"elo":1360.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":500.7827657461,"x_train":15318.1713297129,"imp":17.7806875215,"elo":1361.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.701764822,"x_train":2.6626901865,"imp":29.677128314,"elo":960.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8119673878,"x_train":5.3549384522,"imp":26.4021999621,"elo":1033.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.0771486904,"x_train":13.526226486,"imp":22.1394841986,"elo":1171.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.5044555898,"x_train":26.9163621783,"imp":20.402716463,"elo":1248.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.4747611582,"x_train":67.3146528947,"imp":20.0204460728,"elo":1267.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.1791495409,"x_train":133.4916186547,"imp":19.9146910647,"elo":1275.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.7604902387,"x_train":266.8571042097,"imp":19.8724071284,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.6481693302,"x_train":400.0742214596,"imp":19.8953131997,"elo":1269.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.6495357454,"x_train":536.3421035528,"imp":19.8661260285,"elo":1268.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..27f44cd9f965850b13c0b71d748f60921aaac996
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_binary/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "RealTabPFN-2.5 (default)", "TabICLv2 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned)", "LimiX (default)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabPFNv2 (tuned + ensembled)", "TabDPT-Turbo (default)", "Mitra (default)", "RealMLP (tuned)", "TabDPT (tuned + ensembled)", "TabICL (default)", "BetaTabPFN (default)", "TabPFNv2 (tuned)", "TabM (tuned + ensembled)", "iLTM (tuned + ensembled)", "TabSwift (default)", "TabDPT (tuned)", "LightGBM (tuned + ensembled)", "TabM (tuned)", "CatBoost (tuned)", "CatBoost (default)", "ModernNCA (tuned)", "CatBoost (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "TabPFNv2 (default)", "XGBoost (tuned + ensembled)", "iLTM (tuned)", "EBM (tuned + ensembled)", "RealMLP (default)", "XGBoost (tuned)", "xRFM (tuned + ensembled)", "TabM (default)", "ChimeraBoost (tuned)", "LightGBM (tuned)", "SAP-RPT-OSS (default)", "ModernNCA (tuned + ensembled)", "EBM (tuned)", "EBM (default)", "TabDPT (default)", "TorchMLP (tuned + ensembled)", "ChimeraBoost (default)", "xRFM (tuned)", "ModernNCA (default)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (tuned + ensembled)", "FastaiMLP (tuned)", "iLTM (default)", "XGBoost (default)", "ExtraTrees (tuned)", "OrionMSP (default)", "LightGBM (default)", "PerpetualBooster (tuned)", "TabSTAR (default)", "RandomForest (tuned + ensembled)", "RandomForest (tuned)", "TabFlex (default)", "PerpetualBooster (default)", "ExtraTrees (default)", "TorchMLP (default)", "RandomForest (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "xRFM (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 1.0, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 1.0, 0.9047619104385376, 0.8571428656578064, 1.0, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 1.0, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.1428571492433548, NaN, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.5952380895614624, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.761904776096344, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.7142857313156128, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 1.0, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.761904776096344, 0.761904776096344, 0.9523809552192688, 1.0, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 0.9523809552192688, 1.0, 0.761904776096344, 0.9523809552192688, 1.0, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.190476194024086, 0.523809552192688, NaN, 0.4761904776096344, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.6666666865348816, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.976190447807312, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.8095238208770752, 0.9523809552192688, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 1.0, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 1.0, 1.0, 1.0, 0.8095238208770752, 0.9523809552192688, 1.0, 1.0, 0.9047619104385376, 1.0, 1.0, 0.9523809552192688, 1.0, 1.0, 1.0, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.1428571492433548, 0.4761904776096344, 0.523809552192688, NaN, 0.5714285969734192, 0.4285714328289032, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 1.0, 0.9523809552192688, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 1.0, 0.9523809552192688, 0.8571428656578064, 1.0, 0.9047619104385376, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.1428571492433548, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, NaN, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.6190476417541504, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 0.761904776096344, 0.761904776096344, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.6666666865348816, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.8095238208770752, 1.0, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.7142857313156128, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.7142857313156128, 0.7142857313156128, 0.9523809552192688, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 1.0, 0.7142857313156128, 1.0, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.0476190485060215, 0.4047619104385376, 0.4285714328289032, 0.5714285969734192, 0.523809552192688, NaN, 0.4761904776096344, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.8095238208770752, 0.523809552192688, 0.8095238208770752, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.8095238208770752, 0.9047619104385376, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.761904776096344, 0.9047619104385376, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.8571428656578064, 1.0, 0.9523809552192688, 0.9523809552192688, 1.0, 0.9523809552192688, 0.761904776096344, 0.9523809552192688, 0.9523809552192688, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 1.0], [0.190476194024086, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.5714285969734192, 0.523809552192688, NaN, 0.523809552192688, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.4761904776096344, 0.6190476417541504, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.6666666865348816, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0, 0.9523809552192688, 0.9047619104385376, 0.761904776096344, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.095238097012043, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, NaN, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.5952380895614624, 0.5714285969734192, 0.8095238208770752, 0.6666666865348816, 0.5714285969734192, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 1.0, 0.8095238208770752, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 1.0, 0.9523809552192688, 0.9523809552192688, 0.761904776096344, 0.9523809552192688, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 1.0], [0.095238097012043, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, NaN, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.738095223903656, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.738095223903656, 0.5714285969734192, 0.6190476417541504, 0.8095238208770752, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.9047619104385376, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.6666666865348816, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.7857142686843872, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8809523582458496, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.380952388048172, NaN, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.5714285969734192, 0.6190476417541504, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.6666666865348816, 0.9047619104385376, 0.761904776096344, 0.6190476417541504, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.7142857313156128, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.095238097012043, 0.380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, NaN, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.6666666865348816, 0.5714285969734192, 0.6666666865348816, 0.761904776096344, 0.5714285969734192, 0.6190476417541504, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.7142857313156128, 0.6666666865348816, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.6666666865348816, 0.7142857313156128, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.761904776096344, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0, 1.0], [0.0476190485060215, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.4047619104385376, 0.380952388048172, 0.523809552192688, 0.523809552192688, NaN, 0.4761904776096344, 0.5, 0.523809552192688, 0.4761904776096344, 0.6428571343421936, 0.7142857313156128, 0.6904761791229248, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.6428571343421936, 0.8095238208770752, 0.6190476417541504, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.5714285969734192, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.9523809552192688, 0.8571428656578064, 0.7142857313156128, 0.9523809552192688, 0.9285714030265808, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9285714030265808, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 1.0], [0.095238097012043, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.4761904776096344, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, NaN, 0.523809552192688, 0.523809552192688, 0.761904776096344, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.5952380895614624, 0.7142857313156128, 0.7142857313156128, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0, 1.0], [0.0, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.380952388048172, 0.190476194024086, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5, 0.4761904776096344, NaN, 0.5714285969734192, 0.523809552192688, 0.6428571343421936, 0.6190476417541504, 0.6428571343421936, 0.761904776096344, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.7142857313156128, 0.8095238208770752, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6428571343421936, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.8571428656578064, 0.761904776096344, 0.9047619104385376, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8333333134651184, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8809523582458496, 0.8571428656578064, 0.9047619104385376, 0.761904776096344, 0.8571428656578064, 0.8809523582458496, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.1428571492433548, 0.2857142984867096, 0.190476194024086, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.261904776096344, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, NaN, 0.3333333432674408, 0.5714285969734192, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.738095223903656, 0.8571428656578064, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.9523809552192688, 0.761904776096344, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9285714030265808, 0.9047619104385376, 1.0, 1.0], [0.0476190485060215, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.2380952388048172, 0.4761904776096344, 0.6666666865348816, NaN, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6666666865348816, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.523809552192688, 0.6666666865348816, 0.6190476417541504, 0.4761904776096344, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.8095238208770752, 0.9047619104385376, 1.0, 0.9047619104385376, 1.0, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 1.0], [0.0476190485060215, 0.095238097012043, 0.3333333432674408, 0.2857142984867096, 0.095238097012043, 0.3333333432674408, 0.2380952388048172, 0.190476194024086, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.4761904776096344, NaN, 0.4761904776096344, 0.5476190447807312, 0.4285714328289032, 0.6428571343421936, 0.6190476417541504, 0.4761904776096344, 0.523809552192688, 0.4285714328289032, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.523809552192688, 0.5714285969734192, 0.5952380895614624, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.8095238208770752, 0.7142857313156128, 0.6666666865348816, 0.7857142686843872, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.9285714030265808, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9285714030265808, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.0476190485060215, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.523809552192688, 0.523809552192688, NaN, 0.4285714328289032, 0.4761904776096344, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.761904776096344, 0.523809552192688, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.7142857313156128, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.095238097012043, 0.2380952388048172, 0.095238097012043, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.3095238208770752, 0.380952388048172, 0.3571428656578064, 0.4761904776096344, 0.4285714328289032, 0.4523809552192688, 0.5714285969734192, NaN, 0.6190476417541504, 0.523809552192688, 0.4761904776096344, 0.4285714328289032, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.6666666865348816, 0.5476190447807312, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6428571343421936, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8333333134651184, 0.8095238208770752, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 1.0, 1.0], [0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.2380952388048172, 0.4285714328289032, 0.4761904776096344, 0.5714285969734192, 0.523809552192688, 0.380952388048172, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.8095238208770752, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.5714285969734192, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.761904776096344, 0.9523809552192688, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0, 1.0], [0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.261904776096344, 0.2380952388048172, 0.2380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.3571428656578064, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, NaN, 0.523809552192688, 0.380952388048172, 0.6190476417541504, 0.4761904776096344, 0.5714285969734192, 0.4761904776096344, 0.5714285969734192, 0.4761904776096344, 0.6666666865348816, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.7142857313156128, 0.6190476417541504, 0.6190476417541504, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.9047619104385376, 0.9047619104385376, 0.7142857313156128, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 1.0], [0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4047619104385376, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, NaN, 0.380952388048172, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.523809552192688, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.8571428656578064, 0.6190476417541504, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.9523809552192688, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 1.0, 1.0], [0.0, 0.0476190485060215, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.2857142984867096, 0.4761904776096344, 0.523809552192688, 0.3333333432674408, 0.523809552192688, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, NaN, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.7142857313156128, 0.761904776096344, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.8095238208770752, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.9047619104385376, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 1.0, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688], [0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4761904776096344, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, NaN, 0.4285714328289032, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.5714285969734192, 0.5714285969734192, 0.761904776096344, 0.5714285969734192, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.9523809552192688, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.9047619104385376, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9523809552192688, 0.8571428656578064, 1.0, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9047619104385376, 1.0, 1.0], [0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.4761904776096344, 0.190476194024086, 0.3333333432674408, 0.4761904776096344, 0.5714285969734192, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, NaN, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.6904761791229248, 0.4761904776096344, 0.6190476417541504, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.523809552192688, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.7142857313156128, 0.9523809552192688, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688], [0.0476190485060215, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4761904776096344, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, NaN, 0.6190476417541504, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.523809552192688, 0.761904776096344, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.8571428656578064, 0.6666666865348816, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.9047619104385376, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.3333333432674408, 0.1428571492433548, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.2380952388048172, 0.2380952388048172, 0.4761904776096344, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.380952388048172, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.5714285969734192, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.761904776096344, 0.9047619104385376, 0.761904776096344, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0, 1.0], [0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.2380952388048172, 0.4285714328289032, 0.2857142984867096, 0.3333333432674408, 0.4761904776096344, 0.3333333432674408, 0.4761904776096344, 0.3333333432674408, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, NaN, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.6428571343421936, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 1.0], [0.0, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.2857142984867096, 0.095238097012043, 0.2380952388048172, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.523809552192688, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, NaN, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.7142857313156128, 0.523809552192688, 0.6666666865348816, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.8095238208770752, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.6190476417541504, 0.8095238208770752, 0.8095238208770752, 0.6666666865348816, 0.8571428656578064, 0.9047619104385376, 0.7142857313156128, 0.9047619104385376, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.5714285969734192, 0.380952388048172, 0.523809552192688, 0.3095238208770752, 0.5714285969734192, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, NaN, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.8095238208770752, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 1.0], [0.095238097012043, 0.2857142984867096, 0.095238097012043, 0.190476194024086, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.3333333432674408, 0.523809552192688, 0.4047619104385376, 0.4761904776096344, 0.4523809552192688, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, NaN, 0.4761904776096344, 0.380952388048172, 0.523809552192688, 0.4285714328289032, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4285714328289032, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.738095223903656, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8095238208770752, 0.738095223903656, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688, 1.0], [0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, NaN, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.6666666865348816, 0.523809552192688, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.5952380895614624, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.8095238208770752, 0.6666666865348816, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 0.9047619104385376, 0.9523809552192688, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 1.0], [0.0476190485060215, 0.095238097012043, 0.02380952425301075, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.6190476417541504, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.4761904776096344, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6666666865348816, 0.761904776096344, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.5714285969734192, 0.8571428656578064, 0.761904776096344, 0.5952380895614624, 0.6190476417541504, 0.8095238208770752, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.9047619104385376, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 0.9523809552192688, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 1.0, 1.0], [0.095238097012043, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.3333333432674408, 0.190476194024086, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, NaN, 0.6190476417541504, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4761904776096344, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.523809552192688, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.8095238208770752, 0.761904776096344, 0.6190476417541504, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.5714285969734192, 0.380952388048172, 0.4285714328289032, 0.380952388048172, NaN, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.380952388048172, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.7142857313156128, 0.523809552192688, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.9047619104385376, 0.9047619104385376, 0.8095238208770752, 0.9523809552192688, 0.8095238208770752, 0.9523809552192688, 0.9523809552192688, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.3333333432674408, 0.523809552192688, 0.4761904776096344, 0.523809552192688, NaN, 0.6190476417541504, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.4285714328289032, 0.6190476417541504, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.523809552192688, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.6666666865348816, 0.9047619104385376, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.9047619104385376, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 1.0], [0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.5714285969734192, 0.380952388048172, NaN, 0.5714285969734192, 0.523809552192688, 0.380952388048172, 0.523809552192688, 0.523809552192688, 0.4285714328289032, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688, 0.9047619104385376, 0.9523809552192688, 0.9523809552192688], [0.095238097012043, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.380952388048172, 0.5714285969734192, 0.4285714328289032, 0.4761904776096344, 0.5714285969734192, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.4761904776096344, 0.4047619104385376, 0.5714285969734192, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.761904776096344, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376], [0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.5714285969734192, 0.4285714328289032, 0.3333333432674408, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.4285714328289032, NaN, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 0.9047619104385376, 1.0], [0.0476190485060215, 0.0, 0.0476190485060215, 0.190476194024086, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.2380952388048172, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.4761904776096344, NaN, 0.523809552192688, 0.380952388048172, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.6190476417541504, 0.4761904776096344, 0.6666666865348816, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9047619104385376, 1.0], [0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.3571428656578064, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, NaN, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.761904776096344, 0.6666666865348816, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.8571428656578064, 0.9047619104385376, 0.8571428656578064, 0.8095238208770752, 0.9047619104385376, 1.0], [0.095238097012043, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.190476194024086, 0.4285714328289032, 0.190476194024086, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.6190476417541504, 0.4761904776096344, NaN, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.761904776096344, 0.6190476417541504, 0.6666666865348816, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.7857142686843872, 0.9047619104385376, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.9523809552192688], [0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.2380952388048172, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.3333333432674408, 0.4761904776096344, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.523809552192688, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.5714285969734192, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.523809552192688, 0.4761904776096344, NaN, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.4285714328289032, 0.4761904776096344, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9523809552192688, 0.9047619104385376, 0.9047619104385376, 0.9047619104385376, 0.8571428656578064, 0.9523809552192688, 1.0], [0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.2380952388048172, 0.2380952388048172, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.5952380895614624, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, NaN, 0.4761904776096344, 0.6190476417541504, 0.4761904776096344, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.8571428656578064, 0.9523809552192688, 0.8095238208770752, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688, 1.0], [0.0, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.190476194024086, 0.4285714328289032, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2380952388048172, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.4761904776096344, 0.3333333432674408, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.523809552192688, NaN, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.4761904776096344, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.761904776096344, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9523809552192688, 0.9523809552192688], [0.0, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.380952388048172, 0.190476194024086, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.4761904776096344, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.2857142984867096, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.523809552192688, NaN, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.6666666865348816, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688, 0.9523809552192688], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.190476194024086, 0.190476194024086, 0.4761904776096344, 0.3333333432674408, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.523809552192688, NaN, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.4761904776096344, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9047619104385376, 1.0], [0.0476190485060215, 0.0476190485060215, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, NaN, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.4761904776096344, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.8571428656578064, 0.8571428656578064, 0.8095238208770752, 0.9523809552192688, 1.0], [0.095238097012043, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.1428571492433548, 0.3333333432674408, 0.2380952388048172, 0.2142857164144516, 0.3333333432674408, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.190476194024086, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.4761904776096344, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, NaN, 0.4761904776096344, 0.4761904776096344, 0.761904776096344, 0.761904776096344, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9523809552192688, 0.9523809552192688], [0.095238097012043, 0.1428571492433548, 0.0476190485060215, 0.0, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.190476194024086, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3571428656578064, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.523809552192688, NaN, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.4761904776096344, 0.6666666865348816, 0.6190476417541504, 0.523809552192688, 0.6190476417541504, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.761904776096344, 0.7142857313156128, 0.9523809552192688, 0.8571428656578064, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.9047619104385376, 0.9523809552192688], [0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.095238097012043, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.380952388048172, 0.2857142984867096, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.4047619104385376, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.523809552192688, 0.3333333432674408, 0.523809552192688, 0.380952388048172, 0.523809552192688, 0.4761904776096344, NaN, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.5476190447807312, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.5476190447807312, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.8571428656578064, 0.761904776096344, 0.7142857313156128, 0.9047619104385376, 0.6666666865348816, 0.761904776096344, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376, 0.9523809552192688], [0.095238097012043, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2142857164144516, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2380952388048172, 0.4761904776096344, 0.3333333432674408, NaN, 0.6190476417541504, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.9047619104385376, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.8095238208770752, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.9047619104385376], [0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.380952388048172, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.4761904776096344, 0.380952388048172, 0.380952388048172, NaN, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.9523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.9047619104385376], [0.0, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.3333333432674408, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, NaN, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.7142857313156128, 0.4761904776096344, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.761904776096344, 0.6190476417541504, 0.6190476417541504, 0.761904776096344, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.761904776096344, 0.9047619104385376], [0.0476190485060215, 0.0, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.3333333432674408, 0.1428571492433548, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4523809552192688, 0.523809552192688, 0.5714285969734192, 0.523809552192688, NaN, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.6666666865348816, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.8571428656578064, 0.5714285969734192, 0.8095238208770752, 0.7142857313156128, 0.9047619104385376, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.761904776096344, 0.9047619104385376, 1.0], [0.095238097012043, 0.095238097012043, 0.0, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.1428571492433548, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.523809552192688, 0.523809552192688, 0.4285714328289032, 0.6190476417541504, NaN, 0.523809552192688, 0.4285714328289032, 0.5714285969734192, 0.523809552192688, 0.4761904776096344, 0.523809552192688, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.761904776096344, 0.8571428656578064, 0.9047619104385376], [0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.380952388048172, 0.2857142984867096, 0.2380952388048172, 0.4047619104385376, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4761904776096344, NaN, 0.4285714328289032, 0.5714285969734192, 0.523809552192688, 0.380952388048172, 0.380952388048172, 0.6190476417541504, 0.6666666865348816, 0.5714285969734192, 0.5714285969734192, 0.4761904776096344, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376], [0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.4285714328289032, 0.190476194024086, 0.380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.523809552192688, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.4761904776096344, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.6190476417541504, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.761904776096344, 0.8095238208770752, 0.761904776096344, 0.8095238208770752, 0.9047619104385376], [0.0476190485060215, 0.0, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.380952388048172, 0.2857142984867096, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.523809552192688, 0.380952388048172, 0.4761904776096344, 0.7142857313156128, 0.4761904776096344, 0.6666666865348816, 0.6190476417541504, 0.5714285969734192, 0.761904776096344, 0.523809552192688, 0.761904776096344, 0.6666666865348816, 0.8571428656578064, 0.6666666865348816, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.9523809552192688, 0.9523809552192688], [0.0, 0.0, 0.0, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0, 0.0, 0.1190476194024086, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.1666666716337204, 0.261904776096344, 0.0476190485060215, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.1428571492433548, 0.095238097012043, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.4761904776096344, 0.4761904776096344, 0.523809552192688, 0.4761904776096344, NaN, 0.5714285969734192, 0.523809552192688, 0.6666666865348816, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.6666666865348816, 0.7142857313156128, 0.6190476417541504, 0.6666666865348816, 0.5714285969734192, 0.6666666865348816, 0.7142857313156128, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.8095238208770752, 0.9523809552192688], [0.0476190485060215, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.2857142984867096, 0.3333333432674408, 0.0476190485060215, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.380952388048172, 0.0, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.095238097012043, 0.2857142984867096, 0.3333333432674408, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.190476194024086, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.523809552192688, 0.4523809552192688, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.380952388048172, 0.6190476417541504, 0.4285714328289032, NaN, 0.4761904776096344, 0.5714285969734192, 0.523809552192688, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.6190476417541504, 0.7142857313156128, 0.6666666865348816, 0.6666666865348816, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.8095238208770752, 0.8571428656578064, 0.9047619104385376], [0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.1428571492433548, 0.095238097012043, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.4761904776096344, 0.4761904776096344, 0.2857142984867096, 0.4285714328289032, 0.4761904776096344, 0.6190476417541504, 0.4285714328289032, 0.523809552192688, 0.4761904776096344, 0.523809552192688, NaN, 0.4761904776096344, 0.4761904776096344, 0.5714285969734192, 0.6666666865348816, 0.761904776096344, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.761904776096344, 0.761904776096344, 0.9047619104385376], [0.095238097012043, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.380952388048172, 0.2380952388048172, 0.380952388048172, 0.3333333432674408, 0.095238097012043, 0.0476190485060215, 0.523809552192688, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.523809552192688, NaN, 0.380952388048172, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.4285714328289032, 0.523809552192688, 0.5714285969734192, 0.380952388048172, 0.5714285969734192, 0.5714285969734192, 0.523809552192688, 0.523809552192688, 0.5714285969734192, 0.6666666865348816, 0.9047619104385376], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.095238097012043, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.190476194024086, 0.2857142984867096, 0.095238097012043, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.4761904776096344, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, 0.523809552192688, 0.6190476417541504, NaN, 0.7142857313156128, 0.523809552192688, 0.523809552192688, 0.7142857313156128, 0.6666666865348816, 0.8571428656578064, 0.6190476417541504, 0.7142857313156128, 0.761904776096344, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8571428656578064, 0.9523809552192688], [0.0476190485060215, 0.0, 0.0, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0714285746216774, 0.1428571492433548, 0.1190476194024086, 0.0476190485060215, 0.190476194024086, 0.0714285746216774, 0.190476194024086, 0.2142857164144516, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.190476194024086, 0.261904776096344, 0.0476190485060215, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.2857142984867096, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.4285714328289032, 0.4285714328289032, 0.3333333432674408, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, NaN, 0.6190476417541504, 0.4761904776096344, 0.6666666865348816, 0.4761904776096344, 0.738095223903656, 0.7142857313156128, 0.761904776096344, 0.8095238208770752, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.9523809552192688], [0.0, 0.095238097012043, 0.0, 0.0, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.0476190485060215, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.1428571492433548, 0.2857142984867096, 0.2380952388048172, 0.1428571492433548, 0.095238097012043, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.190476194024086, 0.3333333432674408, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.190476194024086, 0.1428571492433548, 0.4285714328289032, 0.4761904776096344, 0.380952388048172, 0.3333333432674408, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.4761904776096344, 0.4761904776096344, 0.380952388048172, NaN, 0.523809552192688, 0.523809552192688, 0.4761904776096344, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.761904776096344, 0.8571428656578064], [0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.2380952388048172, 0.095238097012043, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2380952388048172, 0.4285714328289032, 0.4285714328289032, 0.2380952388048172, 0.380952388048172, 0.3333333432674408, 0.523809552192688, 0.2857142984867096, 0.4285714328289032, 0.3333333432674408, 0.380952388048172, 0.2380952388048172, 0.4285714328289032, 0.4761904776096344, 0.523809552192688, 0.4761904776096344, NaN, 0.5714285969734192, 0.4761904776096344, 0.5714285969734192, 0.380952388048172, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.6666666865348816, 0.9047619104385376], [0.0476190485060215, 0.0476190485060215, 0.0, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.2380952388048172, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.1428571492433548, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2857142984867096, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.1428571492433548, 0.2857142984867096, 0.380952388048172, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.5714285969734192, 0.2857142984867096, 0.3333333432674408, 0.4761904776096344, 0.4285714328289032, NaN, 0.380952388048172, 0.6190476417541504, 0.4761904776096344, 0.523809552192688, 0.523809552192688, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.7142857313156128, 0.9047619104385376], [0.0, 0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.190476194024086, 0.0476190485060215, 0.1428571492433548, 0.190476194024086, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.0476190485060215, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.095238097012043, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.3333333432674408, 0.3333333432674408, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4761904776096344, 0.3333333432674408, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.6190476417541504, NaN, 0.5714285969734192, 0.5714285969734192, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6666666865348816, 0.7142857313156128, 0.761904776096344, 0.9047619104385376], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0714285746216774, 0.1428571492433548, 0.1190476194024086, 0.0476190485060215, 0.095238097012043, 0.0714285746216774, 0.190476194024086, 0.1666666716337204, 0.2380952388048172, 0.0476190485060215, 0.2380952388048172, 0.1428571492433548, 0.0476190485060215, 0.2380952388048172, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.095238097012043, 0.190476194024086, 0.261904776096344, 0.095238097012043, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.3333333432674408, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.1428571492433548, 0.261904776096344, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.4285714328289032, NaN, 0.523809552192688, 0.6190476417541504, 0.5714285969734192, 0.6190476417541504, 0.6190476417541504, 0.6666666865348816, 0.761904776096344, 0.9523809552192688], [0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.095238097012043, 0.0, 0.0476190485060215, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.1428571492433548, 0.190476194024086, 0.095238097012043, 0.1428571492433548, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.1428571492433548, 0.2380952388048172, 0.0476190485060215, 0.2380952388048172, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.2857142984867096, 0.2857142984867096, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.3333333432674408, 0.380952388048172, 0.6190476417541504, 0.380952388048172, 0.2857142984867096, 0.4285714328289032, 0.6190476417541504, 0.523809552192688, 0.4285714328289032, 0.4761904776096344, NaN, 0.6190476417541504, 0.6190476417541504, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8095238208770752, 0.9047619104385376], [0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.095238097012043, 0.2380952388048172, 0.2142857164144516, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.3333333432674408, 0.380952388048172, 0.2380952388048172, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.3333333432674408, 0.3333333432674408, 0.3333333432674408, 0.4285714328289032, 0.2857142984867096, 0.2380952388048172, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.380952388048172, 0.380952388048172, NaN, 0.6666666865348816, 0.6190476417541504, 0.6190476417541504, 0.6190476417541504, 0.9523809552192688, 0.9523809552192688], [0.0, 0.0, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.190476194024086, 0.0476190485060215, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.4285714328289032, 0.2380952388048172, 0.190476194024086, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.4285714328289032, 0.4285714328289032, 0.380952388048172, 0.3333333432674408, NaN, 0.523809552192688, 0.523809552192688, 0.523809552192688, 0.6666666865348816, 0.8571428656578064], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.190476194024086, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.095238097012043, 0.2857142984867096, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.190476194024086, 0.3333333432674408, 0.3333333432674408, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.2857142984867096, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.2380952388048172, 0.4761904776096344, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.380952388048172, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, NaN, 0.738095223903656, 0.6190476417541504, 0.5714285969734192, 0.8095238208770752], [0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0714285746216774, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.095238097012043, 0.0476190485060215, 0.190476194024086, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.190476194024086, 0.2857142984867096, 0.3333333432674408, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.190476194024086, 0.2857142984867096, 0.2857142984867096, 0.4761904776096344, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.3333333432674408, 0.380952388048172, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.261904776096344, NaN, 0.523809552192688, 0.523809552192688, 0.761904776096344], [0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.190476194024086, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.1428571492433548, 0.095238097012043, 0.1428571492433548, 0.190476194024086, 0.2857142984867096, 0.1428571492433548, 0.0476190485060215, 0.190476194024086, 0.0476190485060215, 0.0476190485060215, 0.190476194024086, 0.1428571492433548, 0.190476194024086, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.4285714328289032, 0.2380952388048172, 0.2857142984867096, 0.2857142984867096, 0.3333333432674408, 0.380952388048172, 0.2857142984867096, 0.3333333432674408, 0.4285714328289032, 0.380952388048172, 0.4761904776096344, 0.380952388048172, 0.4761904776096344, NaN, 0.5714285969734192, 0.6666666865348816], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0, 0.0, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.190476194024086, 0.190476194024086, 0.2380952388048172, 0.095238097012043, 0.1428571492433548, 0.1428571492433548, 0.190476194024086, 0.0476190485060215, 0.190476194024086, 0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.1428571492433548, 0.1428571492433548, 0.2380952388048172, 0.3333333432674408, 0.2857142984867096, 0.2380952388048172, 0.2380952388048172, 0.190476194024086, 0.0476190485060215, 0.3333333432674408, 0.4285714328289032, 0.4761904776096344, 0.4285714328289032, NaN, 0.9047619104385376], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0476190485060215, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0476190485060215, 0.095238097012043, 0.0, 0.0, 0.0, 0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0, 0.0, 0.0476190485060215, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.0476190485060215, 0.1428571492433548, 0.095238097012043, 0.095238097012043, 0.095238097012043, 0.0476190485060215, 0.095238097012043, 0.0476190485060215, 0.1428571492433548, 0.190476194024086, 0.2380952388048172, 0.3333333432674408, 0.095238097012043, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9356833681},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8568336497},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8487940704},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.819418686},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8138528228},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.807359317},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.7946815191},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7881880117},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7365491755},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7346938891},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7291280262},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7028447849},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6904762044},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.685528767},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6675943227},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6493506637},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6425479382},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6252319248},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.6243042787},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6239950664},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.623376635},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.6181199883},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.6147186304},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5899814592},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.5871985299},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.5837971669},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5745207291},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.570500941},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5701917253},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5643166502},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5565862836},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5531849214},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5448361276},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.544217698},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5423624096},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5244279641},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5244279641},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5222634627},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5194805323},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5182436719},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5154607407},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5136054538},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5021645131},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.5018552967},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4990723666},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4811379202},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4712430544},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.4693877651},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4468150995},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4239332197},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4233147906},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.4152752112},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.4056895592},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3927025451},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3818800337},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3766233866},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3738404547},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.370439094},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3692022358},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3633271593},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3429189952},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.3382807748},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.3358070605},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.332714912},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.2993197335},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.286332722},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2628324119},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2597402668},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2547928329},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.243661107},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2356215275},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2006802767},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.1991342045},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1895485511},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1743970365},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1700680312},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1119356854},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0414347566}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_binary/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9c5a548e8ba4d1963ceb4b9cf17ffc4369f460e6
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_binary/datasets_small/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),TabPFN-3 (default),RealTabPFN-2.5 (default),TabICLv2 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned),LimiX (default),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabPFNv2 (tuned + ensembled),TabDPT-Turbo (default),Mitra (default),RealMLP (tuned),TabDPT (tuned + ensembled),TabICL (default),BetaTabPFN (default),TabPFNv2 (tuned),TabM (tuned + ensembled),iLTM (tuned + ensembled),TabSwift (default),TabDPT (tuned),LightGBM (tuned + ensembled),TabM (tuned),CatBoost (tuned),CatBoost (default),ModernNCA (tuned),CatBoost (tuned + ensembled),ChimeraBoost (tuned + ensembled),TabPFNv2 (default),XGBoost (tuned + ensembled),iLTM (tuned),EBM (tuned + ensembled),RealMLP (default),XGBoost (tuned),xRFM (tuned + ensembled),TabM (default),ChimeraBoost (tuned),LightGBM (tuned),SAP-RPT-OSS (default),ModernNCA (tuned + ensembled),EBM (tuned),EBM (default),TabDPT (default),TorchMLP (tuned + ensembled),ChimeraBoost (default),xRFM (tuned),ModernNCA (default),FastaiMLP (tuned + ensembled),TorchMLP (tuned),TabSTAR (tuned),TabSTAR (tuned + ensembled),PerpetualBooster (tuned + ensembled),ExtraTrees (tuned + ensembled),FastaiMLP (tuned),iLTM (default),XGBoost (default),ExtraTrees (tuned),OrionMSP (default),LightGBM (default),PerpetualBooster (tuned),TabSTAR (default),RandomForest (tuned + ensembled),RandomForest (tuned),TabFlex (default),PerpetualBooster (default),ExtraTrees (default),TorchMLP (default),RandomForest (default),FastaiMLP (default),KNN (tuned + ensembled),xRFM (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.85714287,0.8095238,0.85714287,0.85714287,0.95238096,0.8095238,0.9047619,0.9047619,0.8095238,0.9047619,0.95238096,0.9047619,1.0,0.85714287,0.95238096,0.95238096,0.95238096,0.95238096,0.9047619,0.85714287,0.9047619,1.0,0.95238096,0.95238096,0.95238096,0.9047619,0.95238096,1.0,0.95238096,0.9047619,0.95238096,0.95238096,0.9047619,0.85714287,0.9047619,0.95238096,0.9047619,0.9047619,0.95238096,0.9047619,0.9047619,0.9047619,0.9047619,1.0,1.0,0.95238096,0.95238096,0.9047619,0.9047619,1.0,0.9047619,0.85714287,1.0,0.95238096,0.9047619,0.9047619,1.0,0.95238096,1.0,0.95238096,1.0,0.9047619,0.95238096,0.95238096,1.0,1.0,0.95238096,1.0,0.95238096,1.0,0.95238096,1.0,0.95238096,0.95238096,1.0,1.0,1.0
+TabPFN-3 (default),0.14285715,,0.47619048,0.52380955,0.5714286,0.5952381,0.61904764,0.61904764,0.6666667,0.7619048,0.61904764,0.71428573,0.71428573,0.85714287,0.71428573,0.7619048,0.9047619,0.95238096,0.9047619,0.85714287,0.7619048,0.85714287,0.95238096,0.95238096,0.9047619,0.8095238,0.95238096,0.85714287,0.9047619,0.9047619,0.71428573,0.9047619,0.9047619,0.8095238,0.85714287,0.85714287,0.9047619,0.8095238,0.85714287,1.0,0.8095238,0.9047619,0.9047619,0.85714287,0.9047619,0.95238096,0.95238096,0.95238096,0.9047619,0.85714287,0.95238096,0.7619048,0.7619048,0.95238096,1.0,0.9047619,0.95238096,0.95238096,1.0,1.0,0.95238096,1.0,0.7619048,0.95238096,1.0,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0,1.0,0.95238096,0.95238096,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.1904762,0.52380955,,0.47619048,0.47619048,0.5714286,0.5714286,0.52380955,0.71428573,0.71428573,0.71428573,0.61904764,0.71428573,0.7619048,0.8095238,0.7619048,0.6666667,0.8095238,0.8095238,0.8095238,0.6666667,0.8095238,0.7619048,0.9047619,0.85714287,0.8095238,0.9047619,0.85714287,0.85714287,0.85714287,0.9047619,0.9047619,0.97619045,0.9047619,0.8095238,0.9047619,0.9047619,0.9047619,0.85714287,0.95238096,0.85714287,0.8095238,0.95238096,0.9047619,0.85714287,0.95238096,0.95238096,1.0,0.9047619,0.95238096,0.9047619,0.8095238,0.8095238,0.9047619,0.9047619,1.0,0.95238096,0.95238096,0.9047619,1.0,1.0,1.0,0.8095238,0.95238096,1.0,1.0,0.9047619,1.0,1.0,0.95238096,1.0,1.0,1.0,0.95238096,0.95238096,1.0,1.0,1.0
+TabICLv2 (default),0.14285715,0.47619048,0.52380955,,0.5714286,0.42857143,0.61904764,0.5714286,0.61904764,0.71428573,0.7619048,0.71428573,0.6666667,0.71428573,0.61904764,0.8095238,0.71428573,0.9047619,0.7619048,0.7619048,0.7619048,0.8095238,0.8095238,0.8095238,0.8095238,0.8095238,0.85714287,0.8095238,0.8095238,0.85714287,0.8095238,0.8095238,0.7619048,0.8095238,0.8095238,0.8095238,0.9047619,0.8095238,0.8095238,0.8095238,0.8095238,0.9047619,0.85714287,0.85714287,0.85714287,0.9047619,0.85714287,0.95238096,0.85714287,1.0,0.95238096,0.8095238,0.8095238,0.9047619,0.85714287,0.9047619,0.8095238,0.95238096,0.9047619,0.9047619,0.95238096,0.95238096,0.8095238,0.9047619,0.9047619,1.0,0.95238096,0.85714287,1.0,0.9047619,1.0,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0,1.0
+TabPFN-2.6 (default),0.14285715,0.42857143,0.52380955,0.42857143,,0.47619048,0.42857143,0.52380955,0.71428573,0.61904764,0.6666667,0.71428573,0.6666667,0.7619048,0.71428573,0.61904764,0.9047619,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.9047619,0.7619048,0.7619048,0.9047619,0.8095238,0.85714287,0.8095238,0.6666667,0.8095238,0.85714287,0.85714287,0.8095238,0.85714287,0.9047619,0.8095238,0.8095238,1.0,0.8095238,0.9047619,0.9047619,0.8095238,0.71428573,0.95238096,0.9047619,0.9047619,0.9047619,0.85714287,0.95238096,0.71428573,0.71428573,0.95238096,0.9047619,0.8095238,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096,1.0,0.71428573,1.0,0.95238096,0.85714287,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,1.0,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.04761905,0.4047619,0.42857143,0.5714286,0.52380955,,0.47619048,0.61904764,0.6666667,0.6666667,0.61904764,0.8095238,0.52380955,0.8095238,0.6666667,0.61904764,0.6666667,0.8095238,0.9047619,0.7619048,0.7619048,0.71428573,0.71428573,0.9047619,0.8095238,0.85714287,0.9047619,0.85714287,0.9047619,0.85714287,0.7619048,0.9047619,0.8095238,0.8095238,0.7619048,0.85714287,0.85714287,0.85714287,0.9047619,0.95238096,0.8095238,0.85714287,0.8095238,0.85714287,0.71428573,0.85714287,0.85714287,0.95238096,0.8095238,0.8095238,0.85714287,0.71428573,0.71428573,0.9047619,0.95238096,0.85714287,0.85714287,1.0,0.95238096,0.95238096,1.0,0.95238096,0.7619048,0.95238096,0.95238096,0.8095238,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.9047619,0.95238096,0.9047619,0.95238096,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.1904762,0.3809524,0.42857143,0.3809524,0.5714286,0.52380955,,0.52380955,0.5714286,0.71428573,0.6666667,0.6666667,0.47619048,0.61904764,0.7619048,0.71428573,0.7619048,0.7619048,0.7619048,0.7619048,0.7619048,0.6666667,0.71428573,0.85714287,0.71428573,0.8095238,0.85714287,0.9047619,0.9047619,0.85714287,0.6666667,0.85714287,0.85714287,0.7619048,0.7619048,0.85714287,0.9047619,0.8095238,0.85714287,0.9047619,0.7619048,0.85714287,0.7619048,0.8095238,0.7619048,0.85714287,0.9047619,0.8095238,0.9047619,0.8095238,0.85714287,0.7619048,0.7619048,0.85714287,0.9047619,0.85714287,0.9047619,0.9047619,0.9047619,1.0,0.95238096,0.9047619,0.7619048,0.9047619,0.9047619,0.9047619,0.9047619,0.95238096,0.9047619,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0952381,0.3809524,0.47619048,0.42857143,0.47619048,0.3809524,0.47619048,,0.5714286,0.5714286,0.61904764,0.5952381,0.5714286,0.8095238,0.6666667,0.5714286,0.8095238,0.8095238,0.9047619,0.8095238,0.6666667,0.61904764,0.71428573,0.8095238,0.85714287,0.8095238,0.9047619,0.85714287,0.8095238,0.7619048,0.71428573,0.85714287,0.85714287,0.8095238,0.7619048,0.8095238,0.8095238,0.85714287,0.85714287,0.95238096,0.8095238,0.8095238,0.8095238,0.8095238,0.71428573,0.85714287,0.85714287,0.9047619,0.85714287,0.8095238,0.9047619,0.71428573,0.71428573,0.9047619,1.0,0.8095238,0.85714287,0.95238096,0.95238096,1.0,0.95238096,0.95238096,0.7619048,0.95238096,0.9047619,0.8095238,0.9047619,0.95238096,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096,0.9047619,0.9047619,0.9047619,0.95238096,1.0
+LimiX (default),0.0952381,0.33333334,0.2857143,0.3809524,0.2857143,0.33333334,0.42857143,0.42857143,,0.61904764,0.52380955,0.61904764,0.52380955,0.5714286,0.7380952,0.47619048,0.5714286,0.6666667,0.71428573,0.6666667,0.7380952,0.5714286,0.61904764,0.8095238,0.71428573,0.6666667,0.6666667,0.7619048,0.71428573,0.71428573,0.61904764,0.7619048,0.9047619,0.8095238,0.7619048,0.8095238,0.7619048,0.7619048,0.7619048,0.7619048,0.8095238,0.7619048,0.9047619,0.85714287,0.6666667,0.8095238,0.8095238,0.85714287,0.85714287,0.7619048,0.7619048,0.78571427,0.7619048,0.8095238,0.85714287,0.8095238,0.85714287,0.85714287,0.8095238,0.88095236,0.85714287,0.8095238,0.8095238,0.85714287,0.95238096,0.85714287,0.85714287,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+RealMLP (tuned + ensembled),0.1904762,0.23809524,0.2857143,0.2857143,0.3809524,0.33333334,0.2857143,0.42857143,0.3809524,,0.5714286,0.47619048,0.52380955,0.5714286,0.61904764,0.52380955,0.61904764,0.6666667,0.61904764,0.6666667,0.7619048,0.5714286,0.61904764,0.8095238,0.71428573,0.7619048,0.85714287,0.6666667,0.9047619,0.7619048,0.61904764,0.8095238,0.7619048,0.8095238,0.8095238,0.85714287,0.71428573,0.6666667,0.7619048,0.85714287,0.7619048,0.7619048,0.8095238,0.8095238,0.7619048,0.8095238,0.85714287,0.7619048,0.85714287,0.85714287,0.7619048,0.6666667,0.6666667,0.9047619,0.85714287,0.8095238,0.9047619,0.9047619,0.8095238,0.9047619,0.9047619,0.95238096,0.71428573,0.8095238,0.9047619,0.8095238,0.95238096,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0952381,0.3809524,0.2857143,0.23809524,0.33333334,0.3809524,0.33333334,0.3809524,0.47619048,0.42857143,,0.47619048,0.5714286,0.5714286,0.6666667,0.52380955,0.61904764,0.6666667,0.5714286,0.6666667,0.7619048,0.5714286,0.61904764,0.7619048,0.7619048,0.71428573,0.71428573,0.61904764,0.7619048,0.6666667,0.6666667,0.7619048,0.7619048,0.6666667,0.6666667,0.7619048,0.7619048,0.71428573,0.71428573,0.8095238,0.71428573,0.6666667,0.85714287,0.7619048,0.8095238,0.85714287,0.7619048,0.8095238,0.85714287,0.8095238,0.9047619,0.6666667,0.71428573,0.9047619,0.85714287,0.85714287,0.85714287,0.9047619,0.8095238,0.9047619,0.9047619,0.95238096,0.7619048,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.04761905,0.2857143,0.3809524,0.2857143,0.2857143,0.1904762,0.33333334,0.4047619,0.3809524,0.52380955,0.52380955,,0.47619048,0.5,0.52380955,0.47619048,0.64285713,0.71428573,0.6904762,0.6666667,0.61904764,0.5714286,0.52380955,0.71428573,0.6666667,0.6666667,0.61904764,0.7619048,0.6666667,0.6666667,0.64285713,0.8095238,0.61904764,0.8095238,0.7619048,0.8095238,0.71428573,0.71428573,0.7619048,0.7619048,0.6666667,0.8095238,0.7619048,0.7619048,0.5714286,0.7619048,0.8095238,0.85714287,0.7619048,0.7619048,0.85714287,0.6666667,0.6666667,0.7619048,0.8095238,0.8095238,0.85714287,0.9047619,0.85714287,0.8095238,0.95238096,0.85714287,0.71428573,0.95238096,0.9285714,0.8095238,0.9047619,0.85714287,0.9047619,0.9285714,0.95238096,0.95238096,0.9047619,0.95238096,0.95238096,0.9047619,0.95238096,1.0
+TabDPT-Turbo (default),0.0952381,0.2857143,0.2857143,0.33333334,0.33333334,0.47619048,0.52380955,0.42857143,0.47619048,0.47619048,0.42857143,0.52380955,,0.52380955,0.52380955,0.7619048,0.5714286,0.52380955,0.61904764,0.5714286,0.5714286,0.5952381,0.71428573,0.71428573,0.52380955,0.52380955,0.61904764,0.5714286,0.5714286,0.6666667,0.5714286,0.6666667,0.71428573,0.6666667,0.61904764,0.71428573,0.71428573,0.71428573,0.6666667,0.71428573,0.71428573,0.5714286,0.71428573,0.7619048,0.7619048,0.6666667,0.71428573,0.7619048,0.71428573,0.8095238,0.7619048,0.7619048,0.7619048,0.71428573,0.85714287,0.85714287,0.8095238,0.8095238,0.85714287,0.9047619,0.9047619,0.7619048,0.7619048,0.85714287,0.85714287,0.95238096,0.9047619,0.9047619,0.85714287,0.85714287,1.0,1.0,0.95238096,0.9047619,0.9047619,0.9047619,1.0,1.0
+Mitra (default),0.0,0.14285715,0.23809524,0.2857143,0.23809524,0.1904762,0.3809524,0.1904762,0.42857143,0.42857143,0.42857143,0.5,0.47619048,,0.5714286,0.52380955,0.64285713,0.61904764,0.64285713,0.7619048,0.61904764,0.5714286,0.52380955,0.71428573,0.8095238,0.6666667,0.7619048,0.71428573,0.71428573,0.71428573,0.64285713,0.61904764,0.6666667,0.71428573,0.6666667,0.6666667,0.6666667,0.85714287,0.8095238,0.8095238,0.71428573,0.8095238,0.71428573,0.71428573,0.71428573,0.71428573,0.7619048,0.6666667,0.85714287,0.7619048,0.9047619,0.71428573,0.71428573,0.7619048,0.7619048,0.7619048,0.8095238,0.8095238,0.8095238,0.8333333,0.8095238,0.7619048,0.7619048,0.7619048,0.88095236,0.85714287,0.9047619,0.7619048,0.85714287,0.88095236,0.9047619,0.9047619,0.95238096,0.95238096,0.95238096,0.95238096,1.0,1.0
+RealMLP (tuned),0.14285715,0.2857143,0.1904762,0.3809524,0.2857143,0.33333334,0.23809524,0.33333334,0.26190478,0.3809524,0.33333334,0.47619048,0.47619048,0.42857143,,0.33333334,0.5714286,0.61904764,0.52380955,0.5714286,0.52380955,0.52380955,0.47619048,0.71428573,0.6666667,0.61904764,0.7619048,0.6666667,0.6666667,0.61904764,0.6666667,0.71428573,0.61904764,0.71428573,0.6666667,0.71428573,0.61904764,0.71428573,0.71428573,0.7619048,0.61904764,0.71428573,0.71428573,0.71428573,0.6666667,0.7619048,0.71428573,0.6666667,0.6666667,0.7619048,0.7619048,0.6666667,0.71428573,0.8095238,0.8095238,0.7619048,0.9047619,0.8095238,0.85714287,0.7380952,0.85714287,0.8095238,0.71428573,0.9047619,0.95238096,0.7619048,0.95238096,0.9047619,0.95238096,0.95238096,0.9047619,0.9047619,0.95238096,0.9047619,0.9285714,0.9047619,1.0,1.0
+TabDPT (tuned + ensembled),0.04761905,0.23809524,0.23809524,0.1904762,0.3809524,0.3809524,0.2857143,0.42857143,0.52380955,0.47619048,0.47619048,0.52380955,0.23809524,0.47619048,0.6666667,,0.52380955,0.47619048,0.5714286,0.52380955,0.61904764,0.52380955,0.6666667,0.5714286,0.52380955,0.61904764,0.52380955,0.52380955,0.6666667,0.61904764,0.47619048,0.61904764,0.71428573,0.61904764,0.5714286,0.6666667,0.61904764,0.6666667,0.6666667,0.61904764,0.71428573,0.61904764,0.6666667,0.71428573,0.6666667,0.61904764,0.61904764,0.71428573,0.7619048,0.6666667,0.61904764,0.7619048,0.71428573,0.6666667,0.8095238,0.7619048,0.8095238,0.7619048,0.85714287,0.95238096,0.7619048,0.7619048,0.7619048,0.7619048,0.8095238,0.9047619,0.8095238,0.95238096,0.8095238,0.9047619,1.0,0.9047619,1.0,0.9047619,0.9047619,0.85714287,0.95238096,1.0
+TabICL (default),0.04761905,0.0952381,0.33333334,0.2857143,0.0952381,0.33333334,0.23809524,0.1904762,0.42857143,0.3809524,0.3809524,0.35714287,0.42857143,0.35714287,0.42857143,0.47619048,,0.47619048,0.54761904,0.42857143,0.64285713,0.61904764,0.47619048,0.52380955,0.42857143,0.61904764,0.61904764,0.6666667,0.52380955,0.5714286,0.5952381,0.61904764,0.71428573,0.5714286,0.6666667,0.6666667,0.61904764,0.5714286,0.6666667,0.5714286,0.71428573,0.61904764,0.61904764,0.71428573,0.6666667,0.8095238,0.71428573,0.6666667,0.78571427,0.8095238,0.71428573,0.71428573,0.71428573,0.7619048,0.8095238,0.85714287,0.7619048,0.8095238,0.8095238,0.8095238,0.85714287,0.85714287,0.7619048,0.8095238,0.9285714,0.85714287,0.9047619,0.85714287,0.95238096,0.9285714,0.95238096,0.9047619,0.9047619,0.95238096,0.95238096,0.95238096,1.0,1.0
+BetaTabPFN (default),0.04761905,0.04761905,0.1904762,0.0952381,0.14285715,0.1904762,0.23809524,0.1904762,0.33333334,0.33333334,0.33333334,0.2857143,0.47619048,0.3809524,0.3809524,0.52380955,0.52380955,,0.42857143,0.47619048,0.61904764,0.52380955,0.5714286,0.6666667,0.5714286,0.52380955,0.6666667,0.52380955,0.61904764,0.7619048,0.52380955,0.71428573,0.5714286,0.71428573,0.6666667,0.71428573,0.6666667,0.6666667,0.6666667,0.61904764,0.61904764,0.5714286,0.7619048,0.71428573,0.71428573,0.61904764,0.71428573,0.61904764,0.6666667,0.6666667,0.85714287,0.71428573,0.71428573,0.71428573,0.8095238,0.71428573,0.8095238,0.85714287,0.8095238,0.71428573,0.85714287,0.7619048,0.7619048,0.85714287,0.8095238,0.8095238,0.8095238,0.71428573,0.85714287,0.8095238,0.85714287,0.95238096,0.85714287,0.95238096,0.95238096,0.95238096,1.0,1.0
+TabPFNv2 (tuned),0.04761905,0.0952381,0.1904762,0.23809524,0.14285715,0.0952381,0.23809524,0.0952381,0.2857143,0.3809524,0.42857143,0.30952382,0.3809524,0.35714287,0.47619048,0.42857143,0.45238096,0.5714286,,0.61904764,0.52380955,0.47619048,0.42857143,0.6666667,0.61904764,0.61904764,0.5714286,0.6666667,0.5714286,0.6666667,0.54761904,0.71428573,0.61904764,0.6666667,0.61904764,0.71428573,0.61904764,0.71428573,0.71428573,0.6666667,0.64285713,0.71428573,0.61904764,0.71428573,0.5714286,0.71428573,0.7619048,0.6666667,0.7619048,0.71428573,0.8095238,0.6666667,0.6666667,0.71428573,0.71428573,0.7619048,0.8095238,0.8095238,0.71428573,0.7619048,0.85714287,0.7619048,0.71428573,0.7619048,0.78571427,0.85714287,0.85714287,0.85714287,0.8095238,0.8333333,0.8095238,0.95238096,0.85714287,0.95238096,0.95238096,0.9047619,1.0,1.0
+TabM (tuned + ensembled),0.0952381,0.14285715,0.1904762,0.23809524,0.2857143,0.23809524,0.23809524,0.1904762,0.33333334,0.33333334,0.33333334,0.33333334,0.42857143,0.23809524,0.42857143,0.47619048,0.5714286,0.52380955,0.3809524,,0.5714286,0.5714286,0.42857143,0.61904764,0.5714286,0.52380955,0.61904764,0.52380955,0.61904764,0.61904764,0.5714286,0.61904764,0.61904764,0.6666667,0.6666667,0.71428573,0.61904764,0.7619048,0.8095238,0.61904764,0.6666667,0.6666667,0.5714286,0.6666667,0.71428573,0.6666667,0.71428573,0.5714286,0.7619048,0.7619048,0.7619048,0.7619048,0.7619048,0.71428573,0.71428573,0.7619048,0.7619048,0.7619048,0.6666667,0.7619048,0.8095238,0.85714287,0.7619048,0.7619048,0.7619048,0.85714287,0.9047619,0.7619048,0.95238096,0.7619048,0.9047619,0.85714287,0.85714287,0.9047619,0.9047619,0.9047619,1.0,1.0
+iLTM (tuned + ensembled),0.14285715,0.23809524,0.33333334,0.23809524,0.14285715,0.23809524,0.23809524,0.33333334,0.26190478,0.23809524,0.23809524,0.3809524,0.42857143,0.3809524,0.47619048,0.3809524,0.35714287,0.3809524,0.47619048,0.42857143,,0.52380955,0.3809524,0.61904764,0.47619048,0.5714286,0.47619048,0.5714286,0.47619048,0.6666667,0.6666667,0.5714286,0.71428573,0.61904764,0.5714286,0.6666667,0.61904764,0.5714286,0.6666667,0.61904764,0.6666667,0.5714286,0.6666667,0.6666667,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.8095238,0.71428573,0.61904764,0.61904764,0.8095238,0.8095238,0.8095238,0.7619048,0.8095238,0.71428573,0.7619048,0.9047619,0.9047619,0.71428573,0.9047619,0.9047619,0.85714287,0.9047619,0.9047619,0.95238096,0.95238096,0.9047619,0.95238096,0.95238096,0.9047619,0.9047619,0.9047619,0.95238096,1.0
+TabSwift (default),0.0952381,0.14285715,0.1904762,0.1904762,0.2857143,0.2857143,0.33333334,0.3809524,0.42857143,0.42857143,0.42857143,0.42857143,0.4047619,0.42857143,0.47619048,0.47619048,0.3809524,0.47619048,0.52380955,0.42857143,0.47619048,,0.3809524,0.52380955,0.47619048,0.5714286,0.5714286,0.5714286,0.61904764,0.42857143,0.5714286,0.5714286,0.61904764,0.6666667,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.61904764,0.5714286,0.6666667,0.5714286,0.52380955,0.7619048,0.71428573,0.6666667,0.85714287,0.61904764,0.7619048,0.7619048,0.6666667,0.8095238,0.7619048,0.71428573,0.71428573,0.8095238,0.85714287,0.71428573,0.71428573,0.8095238,0.71428573,0.7619048,0.95238096,0.8095238,0.85714287,0.85714287,0.7619048,0.95238096,0.95238096,0.95238096,0.95238096,0.95238096,0.9047619,1.0,1.0
+TabDPT (tuned),0.0,0.04761905,0.23809524,0.1904762,0.2857143,0.2857143,0.2857143,0.2857143,0.3809524,0.3809524,0.3809524,0.47619048,0.2857143,0.47619048,0.52380955,0.33333334,0.52380955,0.42857143,0.5714286,0.5714286,0.61904764,0.61904764,,0.52380955,0.52380955,0.5714286,0.52380955,0.52380955,0.61904764,0.61904764,0.52380955,0.5714286,0.6666667,0.61904764,0.5714286,0.61904764,0.61904764,0.61904764,0.6666667,0.61904764,0.5714286,0.5714286,0.52380955,0.71428573,0.7619048,0.61904764,0.71428573,0.61904764,0.8095238,0.61904764,0.6666667,0.71428573,0.71428573,0.6666667,0.7619048,0.7619048,0.8095238,0.71428573,0.8095238,0.9047619,0.6666667,0.7619048,0.71428573,0.71428573,0.7619048,0.8095238,0.85714287,0.85714287,0.8095238,0.85714287,1.0,0.85714287,0.9047619,0.85714287,0.9047619,0.85714287,0.95238096,0.95238096
+LightGBM (tuned + ensembled),0.04761905,0.04761905,0.0952381,0.1904762,0.0952381,0.0952381,0.14285715,0.1904762,0.1904762,0.1904762,0.23809524,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.47619048,0.33333334,0.33333334,0.3809524,0.3809524,0.47619048,0.47619048,,0.42857143,0.52380955,0.61904764,0.52380955,0.5714286,0.47619048,0.52380955,0.52380955,0.61904764,0.61904764,0.61904764,0.71428573,0.6666667,0.5714286,0.5714286,0.7619048,0.5714286,0.5714286,0.6666667,0.71428573,0.5714286,0.71428573,0.7619048,0.71428573,0.6666667,0.7619048,0.7619048,0.71428573,0.71428573,0.7619048,0.8095238,0.71428573,0.8095238,0.85714287,0.71428573,0.71428573,0.95238096,0.8095238,0.71428573,0.8095238,0.9047619,0.7619048,0.8095238,0.8095238,0.8095238,0.95238096,0.85714287,1.0,0.85714287,0.95238096,0.95238096,0.9047619,1.0,1.0
+TabM (tuned),0.04761905,0.0952381,0.14285715,0.1904762,0.23809524,0.1904762,0.2857143,0.14285715,0.2857143,0.2857143,0.23809524,0.33333334,0.47619048,0.1904762,0.33333334,0.47619048,0.5714286,0.42857143,0.3809524,0.42857143,0.52380955,0.52380955,0.47619048,0.5714286,,0.52380955,0.61904764,0.52380955,0.61904764,0.6904762,0.47619048,0.61904764,0.61904764,0.52380955,0.5714286,0.6666667,0.52380955,0.5714286,0.71428573,0.5714286,0.5714286,0.61904764,0.52380955,0.5714286,0.71428573,0.61904764,0.71428573,0.5714286,0.7619048,0.6666667,0.71428573,0.6666667,0.6666667,0.8095238,0.71428573,0.7619048,0.6666667,0.7619048,0.6666667,0.71428573,0.7619048,0.85714287,0.71428573,0.7619048,0.7619048,0.85714287,0.9047619,0.71428573,0.95238096,0.7619048,0.8095238,0.85714287,0.8095238,0.85714287,0.85714287,0.9047619,0.95238096,0.95238096
+CatBoost (tuned),0.04761905,0.1904762,0.1904762,0.1904762,0.23809524,0.14285715,0.1904762,0.1904762,0.33333334,0.23809524,0.2857143,0.33333334,0.47619048,0.33333334,0.3809524,0.3809524,0.3809524,0.47619048,0.3809524,0.47619048,0.42857143,0.42857143,0.42857143,0.47619048,0.47619048,,0.61904764,0.47619048,0.47619048,0.42857143,0.47619048,0.5714286,0.61904764,0.61904764,0.5714286,0.61904764,0.52380955,0.52380955,0.5714286,0.52380955,0.6666667,0.52380955,0.7619048,0.6666667,0.61904764,0.6666667,0.61904764,0.6666667,0.61904764,0.6666667,0.71428573,0.6666667,0.6666667,0.71428573,0.71428573,0.7619048,0.7619048,0.7619048,0.6666667,0.6666667,0.8095238,0.85714287,0.6666667,0.8095238,0.85714287,0.7619048,0.9047619,0.7619048,0.9047619,0.85714287,0.85714287,0.9047619,0.85714287,0.95238096,0.95238096,0.95238096,1.0,1.0
+CatBoost (default),0.0952381,0.04761905,0.0952381,0.14285715,0.0952381,0.0952381,0.14285715,0.0952381,0.33333334,0.14285715,0.2857143,0.3809524,0.3809524,0.23809524,0.23809524,0.47619048,0.3809524,0.33333334,0.42857143,0.3809524,0.52380955,0.42857143,0.47619048,0.3809524,0.3809524,0.3809524,,0.42857143,0.42857143,0.42857143,0.52380955,0.5714286,0.5714286,0.61904764,0.71428573,0.5714286,0.61904764,0.5714286,0.61904764,0.52380955,0.5714286,0.52380955,0.6666667,0.6666667,0.61904764,0.6666667,0.7619048,0.6666667,0.5714286,0.7619048,0.7619048,0.71428573,0.71428573,0.7619048,0.7619048,0.7619048,0.71428573,0.8095238,0.6666667,0.71428573,0.7619048,0.85714287,0.71428573,0.85714287,0.85714287,0.7619048,0.9047619,0.7619048,0.9047619,0.85714287,0.85714287,0.95238096,0.9047619,0.95238096,0.95238096,0.95238096,1.0,1.0
+ModernNCA (tuned),0.04761905,0.14285715,0.14285715,0.1904762,0.1904762,0.14285715,0.0952381,0.14285715,0.23809524,0.33333334,0.3809524,0.23809524,0.42857143,0.2857143,0.33333334,0.47619048,0.33333334,0.47619048,0.33333334,0.47619048,0.42857143,0.42857143,0.47619048,0.47619048,0.47619048,0.52380955,0.5714286,,0.52380955,0.52380955,0.5714286,0.5714286,0.52380955,0.52380955,0.61904764,0.52380955,0.61904764,0.61904764,0.61904764,0.5714286,0.61904764,0.52380955,0.61904764,0.61904764,0.5714286,0.6666667,0.6666667,0.61904764,0.71428573,0.64285713,0.8095238,0.71428573,0.71428573,0.71428573,0.61904764,0.71428573,0.61904764,0.8095238,0.6666667,0.71428573,0.8095238,0.7619048,0.8095238,0.71428573,0.7619048,0.7619048,0.7619048,0.7619048,0.8095238,0.85714287,0.8095238,0.9047619,0.9047619,0.85714287,0.9047619,0.85714287,0.9047619,1.0
+CatBoost (tuned + ensembled),0.0,0.0952381,0.14285715,0.1904762,0.14285715,0.0952381,0.0952381,0.1904762,0.2857143,0.0952381,0.23809524,0.33333334,0.42857143,0.2857143,0.33333334,0.33333334,0.47619048,0.3809524,0.42857143,0.3809524,0.52380955,0.3809524,0.3809524,0.42857143,0.3809524,0.52380955,0.5714286,0.47619048,,0.47619048,0.47619048,0.52380955,0.61904764,0.5714286,0.61904764,0.5714286,0.52380955,0.42857143,0.52380955,0.5714286,0.71428573,0.52380955,0.6666667,0.6666667,0.52380955,0.61904764,0.8095238,0.61904764,0.6666667,0.6666667,0.71428573,0.6666667,0.6666667,0.8095238,0.7619048,0.7619048,0.7619048,0.7619048,0.7619048,0.61904764,0.8095238,0.8095238,0.6666667,0.85714287,0.9047619,0.71428573,0.9047619,0.8095238,0.9047619,0.9047619,0.8095238,0.85714287,0.85714287,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+ChimeraBoost (tuned + ensembled),0.04761905,0.0952381,0.14285715,0.14285715,0.1904762,0.14285715,0.14285715,0.23809524,0.2857143,0.23809524,0.33333334,0.33333334,0.33333334,0.2857143,0.3809524,0.3809524,0.42857143,0.23809524,0.33333334,0.3809524,0.33333334,0.5714286,0.3809524,0.52380955,0.30952382,0.5714286,0.5714286,0.47619048,0.52380955,,0.5714286,0.52380955,0.5714286,0.61904764,0.5714286,0.5714286,0.61904764,0.5714286,0.61904764,0.61904764,0.52380955,0.5714286,0.61904764,0.5714286,0.5714286,0.61904764,0.8095238,0.61904764,0.61904764,0.71428573,0.6666667,0.71428573,0.71428573,0.71428573,0.6666667,0.6666667,0.61904764,0.7619048,0.6666667,0.71428573,0.8095238,0.8095238,0.71428573,0.8095238,0.8095238,0.8095238,0.7619048,0.7619048,0.85714287,0.8095238,0.7619048,0.85714287,0.8095238,0.9047619,0.9047619,0.95238096,0.95238096,1.0
+TabPFNv2 (default),0.0952381,0.2857143,0.0952381,0.1904762,0.33333334,0.23809524,0.33333334,0.2857143,0.3809524,0.3809524,0.33333334,0.35714287,0.42857143,0.35714287,0.33333334,0.52380955,0.4047619,0.47619048,0.45238096,0.42857143,0.33333334,0.42857143,0.47619048,0.47619048,0.52380955,0.52380955,0.47619048,0.42857143,0.52380955,0.42857143,,0.47619048,0.3809524,0.52380955,0.42857143,0.52380955,0.52380955,0.52380955,0.42857143,0.5714286,0.52380955,0.5714286,0.47619048,0.52380955,0.61904764,0.5714286,0.52380955,0.5714286,0.5714286,0.6666667,0.6666667,0.6666667,0.71428573,0.71428573,0.61904764,0.71428573,0.71428573,0.5714286,0.6666667,0.71428573,0.61904764,0.71428573,0.7619048,0.71428573,0.7380952,0.85714287,0.85714287,0.71428573,0.8095238,0.7380952,0.85714287,0.8095238,0.85714287,0.9047619,0.9047619,0.95238096,0.95238096,1.0
+XGBoost (tuned + ensembled),0.04761905,0.0952381,0.0952381,0.1904762,0.1904762,0.0952381,0.14285715,0.14285715,0.23809524,0.1904762,0.23809524,0.1904762,0.33333334,0.3809524,0.2857143,0.3809524,0.3809524,0.2857143,0.2857143,0.3809524,0.42857143,0.42857143,0.42857143,0.47619048,0.3809524,0.42857143,0.42857143,0.42857143,0.47619048,0.47619048,0.52380955,,0.5714286,0.52380955,0.61904764,0.6666667,0.52380955,0.42857143,0.5714286,0.5714286,0.52380955,0.47619048,0.61904764,0.5714286,0.5714286,0.5714286,0.6666667,0.6666667,0.61904764,0.71428573,0.5952381,0.71428573,0.71428573,0.61904764,0.8095238,0.6666667,0.7619048,0.8095238,0.7619048,0.71428573,1.0,0.71428573,0.71428573,0.9047619,0.95238096,0.71428573,0.7619048,0.85714287,0.85714287,0.9047619,0.8095238,0.85714287,0.85714287,0.85714287,0.9047619,0.9047619,0.9047619,1.0
+iLTM (tuned),0.04761905,0.0952381,0.023809524,0.23809524,0.14285715,0.1904762,0.14285715,0.14285715,0.0952381,0.23809524,0.23809524,0.3809524,0.2857143,0.33333334,0.3809524,0.2857143,0.2857143,0.42857143,0.3809524,0.3809524,0.2857143,0.3809524,0.33333334,0.3809524,0.3809524,0.3809524,0.42857143,0.47619048,0.3809524,0.42857143,0.61904764,0.42857143,,0.5714286,0.5714286,0.47619048,0.5714286,0.52380955,0.6666667,0.52380955,0.61904764,0.5714286,0.6666667,0.7619048,0.52380955,0.5714286,0.47619048,0.71428573,0.61904764,0.71428573,0.6666667,0.6666667,0.6666667,0.5714286,0.85714287,0.7619048,0.5952381,0.61904764,0.8095238,0.61904764,0.71428573,0.7619048,0.71428573,0.9047619,0.8095238,0.7619048,0.8095238,0.8095238,0.9047619,0.85714287,0.9047619,0.9047619,0.95238096,0.85714287,0.85714287,0.8095238,1.0,1.0
+EBM (tuned + ensembled),0.0952381,0.1904762,0.0952381,0.1904762,0.14285715,0.1904762,0.23809524,0.1904762,0.1904762,0.1904762,0.33333334,0.1904762,0.33333334,0.2857143,0.2857143,0.3809524,0.42857143,0.2857143,0.33333334,0.33333334,0.3809524,0.33333334,0.3809524,0.3809524,0.47619048,0.3809524,0.3809524,0.47619048,0.42857143,0.3809524,0.47619048,0.47619048,0.42857143,,0.61904764,0.52380955,0.52380955,0.52380955,0.42857143,0.42857143,0.5714286,0.47619048,0.6666667,0.7619048,0.6666667,0.6666667,0.52380955,0.61904764,0.71428573,0.7619048,0.6666667,0.6666667,0.6666667,0.61904764,0.8095238,0.7619048,0.61904764,0.7619048,0.71428573,0.71428573,0.71428573,0.6666667,0.71428573,0.7619048,0.7619048,0.85714287,0.8095238,0.8095238,0.8095238,0.7619048,0.85714287,0.95238096,0.85714287,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+RealMLP (default),0.14285715,0.14285715,0.1904762,0.1904762,0.1904762,0.23809524,0.23809524,0.23809524,0.23809524,0.1904762,0.33333334,0.23809524,0.3809524,0.33333334,0.33333334,0.42857143,0.33333334,0.33333334,0.3809524,0.33333334,0.42857143,0.42857143,0.42857143,0.3809524,0.42857143,0.42857143,0.2857143,0.3809524,0.3809524,0.42857143,0.5714286,0.3809524,0.42857143,0.3809524,,0.47619048,0.42857143,0.52380955,0.42857143,0.47619048,0.47619048,0.3809524,0.5714286,0.52380955,0.6666667,0.71428573,0.52380955,0.61904764,0.71428573,0.6666667,0.71428573,0.61904764,0.6666667,0.61904764,0.6666667,0.71428573,0.71428573,0.7619048,0.71428573,0.6666667,0.6666667,0.6666667,0.71428573,0.7619048,0.7619048,0.9047619,0.9047619,0.8095238,0.95238096,0.8095238,0.95238096,0.95238096,0.85714287,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+XGBoost (tuned),0.0952381,0.14285715,0.0952381,0.1904762,0.14285715,0.14285715,0.14285715,0.1904762,0.1904762,0.14285715,0.23809524,0.1904762,0.2857143,0.33333334,0.2857143,0.33333334,0.33333334,0.2857143,0.2857143,0.2857143,0.33333334,0.42857143,0.3809524,0.2857143,0.33333334,0.3809524,0.42857143,0.47619048,0.42857143,0.42857143,0.47619048,0.33333334,0.52380955,0.47619048,0.52380955,,0.61904764,0.42857143,0.47619048,0.47619048,0.52380955,0.42857143,0.61904764,0.5714286,0.52380955,0.47619048,0.5714286,0.6666667,0.52380955,0.71428573,0.5714286,0.71428573,0.71428573,0.61904764,0.85714287,0.71428573,0.71428573,0.7619048,0.7619048,0.6666667,0.9047619,0.71428573,0.71428573,0.85714287,0.9047619,0.71428573,0.7619048,0.8095238,0.8095238,0.85714287,0.8095238,0.9047619,0.85714287,0.8095238,0.85714287,0.9047619,0.9047619,1.0
+xRFM (tuned + ensembled),0.04761905,0.0952381,0.0952381,0.0952381,0.0952381,0.14285715,0.0952381,0.1904762,0.23809524,0.2857143,0.23809524,0.2857143,0.2857143,0.33333334,0.3809524,0.3809524,0.3809524,0.33333334,0.3809524,0.3809524,0.3809524,0.2857143,0.3809524,0.33333334,0.47619048,0.47619048,0.3809524,0.3809524,0.47619048,0.3809524,0.47619048,0.47619048,0.42857143,0.47619048,0.5714286,0.3809524,,0.5714286,0.52380955,0.3809524,0.52380955,0.52380955,0.42857143,0.52380955,0.52380955,0.52380955,0.47619048,0.71428573,0.71428573,0.6666667,0.61904764,0.61904764,0.61904764,0.61904764,0.6666667,0.71428573,0.6666667,0.71428573,0.71428573,0.7619048,0.71428573,0.71428573,0.6666667,0.7619048,0.7619048,0.8095238,0.8095238,0.85714287,0.8095238,0.85714287,0.9047619,0.85714287,0.85714287,0.9047619,0.95238096,0.9047619,0.95238096,0.95238096
+TabM (default),0.0952381,0.1904762,0.0952381,0.1904762,0.1904762,0.14285715,0.1904762,0.14285715,0.23809524,0.33333334,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.33333334,0.42857143,0.33333334,0.2857143,0.23809524,0.42857143,0.42857143,0.3809524,0.42857143,0.42857143,0.47619048,0.42857143,0.3809524,0.5714286,0.42857143,0.47619048,0.5714286,0.47619048,0.47619048,0.47619048,0.5714286,0.42857143,,0.5714286,0.42857143,0.52380955,0.5714286,0.47619048,0.4047619,0.5714286,0.52380955,0.61904764,0.5714286,0.61904764,0.71428573,0.71428573,0.61904764,0.61904764,0.71428573,0.61904764,0.7619048,0.5714286,0.6666667,0.6666667,0.6666667,0.6666667,0.71428573,0.71428573,0.7619048,0.7619048,0.7619048,0.8095238,0.71428573,0.9047619,0.7619048,0.7619048,0.8095238,0.7619048,0.9047619,0.9047619,0.85714287,0.95238096,0.9047619
+ChimeraBoost (tuned),0.0952381,0.14285715,0.14285715,0.1904762,0.1904762,0.0952381,0.14285715,0.14285715,0.23809524,0.23809524,0.2857143,0.23809524,0.33333334,0.1904762,0.2857143,0.33333334,0.33333334,0.33333334,0.2857143,0.1904762,0.33333334,0.42857143,0.33333334,0.42857143,0.2857143,0.42857143,0.3809524,0.3809524,0.47619048,0.3809524,0.5714286,0.42857143,0.33333334,0.5714286,0.5714286,0.52380955,0.47619048,0.42857143,,0.52380955,0.52380955,0.52380955,0.5714286,0.5714286,0.5714286,0.52380955,0.6666667,0.61904764,0.5714286,0.71428573,0.5714286,0.6666667,0.6666667,0.6666667,0.61904764,0.71428573,0.6666667,0.7619048,0.71428573,0.6666667,0.8095238,0.7619048,0.71428573,0.7619048,0.8095238,0.8095238,0.7619048,0.8095238,0.85714287,0.71428573,0.7619048,0.8095238,0.8095238,0.85714287,0.85714287,0.9047619,0.9047619,1.0
+LightGBM (tuned),0.04761905,0.0,0.04761905,0.1904762,0.0,0.04761905,0.0952381,0.04761905,0.23809524,0.14285715,0.1904762,0.23809524,0.2857143,0.1904762,0.23809524,0.3809524,0.42857143,0.3809524,0.33333334,0.3809524,0.3809524,0.42857143,0.3809524,0.23809524,0.42857143,0.47619048,0.47619048,0.42857143,0.42857143,0.3809524,0.42857143,0.42857143,0.47619048,0.5714286,0.52380955,0.52380955,0.61904764,0.5714286,0.47619048,,0.52380955,0.3809524,0.52380955,0.5714286,0.52380955,0.5714286,0.6666667,0.61904764,0.47619048,0.6666667,0.7619048,0.6666667,0.71428573,0.71428573,0.71428573,0.6666667,0.6666667,0.71428573,0.71428573,0.6666667,0.7619048,0.7619048,0.71428573,0.7619048,0.85714287,0.71428573,0.8095238,0.7619048,0.7619048,0.85714287,0.8095238,0.9047619,0.85714287,0.85714287,0.85714287,0.85714287,0.9047619,1.0
+SAP-RPT-OSS (default),0.0952381,0.1904762,0.14285715,0.1904762,0.1904762,0.1904762,0.23809524,0.1904762,0.1904762,0.23809524,0.2857143,0.33333334,0.2857143,0.2857143,0.3809524,0.2857143,0.2857143,0.3809524,0.35714287,0.33333334,0.33333334,0.2857143,0.42857143,0.42857143,0.42857143,0.33333334,0.42857143,0.3809524,0.2857143,0.47619048,0.47619048,0.47619048,0.3809524,0.42857143,0.52380955,0.47619048,0.47619048,0.47619048,0.47619048,0.47619048,,0.52380955,0.47619048,0.5714286,0.5714286,0.61904764,0.47619048,0.52380955,0.5714286,0.7619048,0.6666667,0.5714286,0.5714286,0.71428573,0.6666667,0.71428573,0.6666667,0.61904764,0.71428573,0.6666667,0.6666667,0.7619048,0.71428573,0.6666667,0.71428573,0.8095238,0.8095238,0.8095238,0.8095238,0.8095238,0.8095238,0.7619048,0.85714287,0.9047619,0.85714287,0.8095238,0.9047619,1.0
+ModernNCA (tuned + ensembled),0.0952381,0.0952381,0.1904762,0.0952381,0.0952381,0.14285715,0.14285715,0.1904762,0.23809524,0.23809524,0.33333334,0.1904762,0.42857143,0.1904762,0.2857143,0.3809524,0.3809524,0.42857143,0.2857143,0.33333334,0.42857143,0.3809524,0.42857143,0.42857143,0.3809524,0.47619048,0.47619048,0.47619048,0.47619048,0.42857143,0.42857143,0.52380955,0.42857143,0.52380955,0.61904764,0.5714286,0.47619048,0.42857143,0.47619048,0.61904764,0.47619048,,0.52380955,0.52380955,0.5714286,0.52380955,0.47619048,0.5714286,0.6666667,0.61904764,0.71428573,0.5714286,0.61904764,0.6666667,0.6666667,0.71428573,0.6666667,0.7619048,0.61904764,0.6666667,0.8095238,0.6666667,0.71428573,0.71428573,0.6666667,0.6666667,0.8095238,0.71428573,0.7619048,0.7619048,0.7619048,0.78571427,0.9047619,0.71428573,0.7619048,0.71428573,0.8095238,0.95238096
+EBM (tuned),0.0952381,0.0952381,0.04761905,0.14285715,0.0952381,0.1904762,0.23809524,0.1904762,0.0952381,0.1904762,0.14285715,0.23809524,0.2857143,0.2857143,0.2857143,0.33333334,0.3809524,0.23809524,0.3809524,0.42857143,0.33333334,0.42857143,0.47619048,0.33333334,0.47619048,0.23809524,0.33333334,0.3809524,0.33333334,0.3809524,0.52380955,0.3809524,0.33333334,0.33333334,0.42857143,0.3809524,0.5714286,0.52380955,0.42857143,0.47619048,0.52380955,0.47619048,,0.52380955,0.61904764,0.61904764,0.42857143,0.47619048,0.61904764,0.61904764,0.61904764,0.6666667,0.6666667,0.61904764,0.6666667,0.6666667,0.61904764,0.6666667,0.71428573,0.6666667,0.6666667,0.61904764,0.71428573,0.71428573,0.71428573,0.8095238,0.7619048,0.8095238,0.8095238,0.8095238,0.8095238,0.95238096,0.9047619,0.9047619,0.9047619,0.85714287,0.95238096,1.0
+EBM (default),0.0952381,0.14285715,0.0952381,0.14285715,0.1904762,0.14285715,0.1904762,0.1904762,0.14285715,0.1904762,0.23809524,0.23809524,0.23809524,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.33333334,0.33333334,0.33333334,0.2857143,0.2857143,0.42857143,0.33333334,0.33333334,0.3809524,0.33333334,0.42857143,0.47619048,0.42857143,0.23809524,0.23809524,0.47619048,0.42857143,0.47619048,0.5952381,0.42857143,0.42857143,0.42857143,0.47619048,0.47619048,,0.47619048,0.61904764,0.47619048,0.5714286,0.61904764,0.6666667,0.71428573,0.6666667,0.6666667,0.71428573,0.7619048,0.6666667,0.61904764,0.5714286,0.71428573,0.71428573,0.61904764,0.7619048,0.6666667,0.71428573,0.7619048,0.8095238,0.85714287,0.8095238,0.85714287,0.7619048,0.85714287,0.95238096,0.8095238,0.95238096,0.95238096,0.95238096,0.95238096,1.0
+TabDPT (default),0.0,0.0952381,0.14285715,0.14285715,0.2857143,0.2857143,0.23809524,0.2857143,0.33333334,0.23809524,0.1904762,0.42857143,0.23809524,0.2857143,0.33333334,0.33333334,0.33333334,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.23809524,0.42857143,0.2857143,0.3809524,0.3809524,0.42857143,0.47619048,0.42857143,0.3809524,0.42857143,0.47619048,0.33333334,0.33333334,0.47619048,0.47619048,0.42857143,0.42857143,0.47619048,0.42857143,0.42857143,0.3809524,0.52380955,,0.47619048,0.52380955,0.5714286,0.6666667,0.5714286,0.47619048,0.5714286,0.61904764,0.61904764,0.61904764,0.7619048,0.61904764,0.61904764,0.71428573,0.8095238,0.6666667,0.71428573,0.6666667,0.6666667,0.71428573,0.7619048,0.71428573,0.8095238,0.7619048,0.7619048,0.85714287,0.85714287,0.8095238,0.8095238,0.8095238,0.8095238,0.95238096,0.95238096
+TorchMLP (tuned + ensembled),0.0,0.04761905,0.04761905,0.0952381,0.04761905,0.14285715,0.14285715,0.14285715,0.1904762,0.1904762,0.14285715,0.23809524,0.33333334,0.2857143,0.23809524,0.3809524,0.1904762,0.3809524,0.2857143,0.33333334,0.2857143,0.47619048,0.3809524,0.2857143,0.3809524,0.33333334,0.33333334,0.33333334,0.3809524,0.3809524,0.42857143,0.42857143,0.42857143,0.33333334,0.2857143,0.52380955,0.47619048,0.47619048,0.47619048,0.42857143,0.3809524,0.47619048,0.3809524,0.3809524,0.52380955,,0.47619048,0.52380955,0.52380955,0.61904764,0.6666667,0.5714286,0.5714286,0.61904764,0.61904764,0.6666667,0.6666667,0.71428573,0.61904764,0.6666667,0.7619048,0.71428573,0.5714286,0.71428573,0.8095238,0.85714287,0.8095238,0.71428573,0.9047619,0.71428573,0.85714287,0.85714287,0.8095238,0.85714287,0.85714287,0.95238096,0.95238096,0.95238096
+ChimeraBoost (default),0.04761905,0.04761905,0.04761905,0.14285715,0.0952381,0.14285715,0.0952381,0.14285715,0.1904762,0.14285715,0.23809524,0.1904762,0.2857143,0.23809524,0.2857143,0.3809524,0.2857143,0.2857143,0.23809524,0.2857143,0.2857143,0.23809524,0.2857143,0.23809524,0.2857143,0.3809524,0.23809524,0.33333334,0.1904762,0.1904762,0.47619048,0.33333334,0.52380955,0.47619048,0.47619048,0.42857143,0.52380955,0.3809524,0.33333334,0.33333334,0.52380955,0.52380955,0.5714286,0.52380955,0.47619048,0.52380955,,0.5714286,0.5714286,0.61904764,0.47619048,0.71428573,0.71428573,0.5714286,0.61904764,0.6666667,0.61904764,0.6666667,0.61904764,0.6666667,0.6666667,0.6666667,0.71428573,0.71428573,0.8095238,0.7619048,0.71428573,0.7619048,0.8095238,0.71428573,0.7619048,0.8095238,0.8095238,0.85714287,0.85714287,0.95238096,0.9047619,1.0
+xRFM (tuned),0.04761905,0.04761905,0.0,0.04761905,0.0952381,0.04761905,0.1904762,0.0952381,0.14285715,0.23809524,0.1904762,0.14285715,0.23809524,0.33333334,0.33333334,0.2857143,0.33333334,0.3809524,0.33333334,0.42857143,0.2857143,0.2857143,0.3809524,0.2857143,0.42857143,0.33333334,0.33333334,0.3809524,0.3809524,0.3809524,0.42857143,0.33333334,0.2857143,0.3809524,0.3809524,0.33333334,0.2857143,0.42857143,0.3809524,0.3809524,0.47619048,0.42857143,0.52380955,0.42857143,0.42857143,0.47619048,0.42857143,,0.52380955,0.61904764,0.61904764,0.5714286,0.61904764,0.47619048,0.71428573,0.61904764,0.71428573,0.5714286,0.71428573,0.6666667,0.6666667,0.61904764,0.61904764,0.6666667,0.71428573,0.71428573,0.85714287,0.85714287,0.8095238,0.8095238,0.85714287,0.85714287,0.95238096,0.85714287,0.85714287,0.8095238,0.95238096,1.0
+ModernNCA (default),0.0952381,0.0952381,0.0952381,0.14285715,0.0952381,0.1904762,0.0952381,0.14285715,0.14285715,0.14285715,0.14285715,0.23809524,0.2857143,0.14285715,0.33333334,0.23809524,0.21428572,0.33333334,0.23809524,0.23809524,0.2857143,0.33333334,0.1904762,0.33333334,0.23809524,0.3809524,0.42857143,0.2857143,0.33333334,0.3809524,0.42857143,0.3809524,0.3809524,0.2857143,0.2857143,0.47619048,0.2857143,0.3809524,0.42857143,0.52380955,0.42857143,0.33333334,0.3809524,0.3809524,0.33333334,0.47619048,0.42857143,0.47619048,,0.47619048,0.47619048,0.7619048,0.7619048,0.5714286,0.5714286,0.61904764,0.6666667,0.52380955,0.52380955,0.61904764,0.5714286,0.61904764,0.7619048,0.71428573,0.71428573,0.6666667,0.71428573,0.7619048,0.8095238,0.7619048,0.7619048,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.95238096,0.95238096
+FastaiMLP (tuned + ensembled),0.0952381,0.14285715,0.04761905,0.0,0.14285715,0.1904762,0.1904762,0.1904762,0.23809524,0.14285715,0.1904762,0.23809524,0.1904762,0.23809524,0.23809524,0.33333334,0.1904762,0.33333334,0.2857143,0.23809524,0.1904762,0.14285715,0.3809524,0.23809524,0.33333334,0.33333334,0.23809524,0.35714287,0.33333334,0.2857143,0.33333334,0.2857143,0.2857143,0.23809524,0.33333334,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.23809524,0.3809524,0.3809524,0.33333334,0.42857143,0.3809524,0.3809524,0.3809524,0.52380955,,0.52380955,0.52380955,0.52380955,0.6666667,0.5714286,0.71428573,0.61904764,0.47619048,0.5714286,0.6666667,0.47619048,0.6666667,0.61904764,0.52380955,0.61904764,0.8095238,0.85714287,0.7619048,0.7619048,0.71428573,0.95238096,0.85714287,0.8095238,0.7619048,0.8095238,0.8095238,0.9047619,0.95238096
+TorchMLP (tuned),0.0,0.04761905,0.0952381,0.04761905,0.04761905,0.14285715,0.14285715,0.0952381,0.23809524,0.23809524,0.0952381,0.14285715,0.23809524,0.0952381,0.23809524,0.3809524,0.2857143,0.14285715,0.1904762,0.23809524,0.2857143,0.3809524,0.33333334,0.23809524,0.2857143,0.2857143,0.23809524,0.1904762,0.2857143,0.33333334,0.33333334,0.4047619,0.33333334,0.33333334,0.2857143,0.42857143,0.3809524,0.2857143,0.42857143,0.23809524,0.33333334,0.2857143,0.3809524,0.2857143,0.52380955,0.33333334,0.52380955,0.3809524,0.52380955,0.47619048,,0.6666667,0.61904764,0.61904764,0.54761904,0.5714286,0.5714286,0.52380955,0.5714286,0.61904764,0.54761904,0.61904764,0.6666667,0.6666667,0.6666667,0.85714287,0.7619048,0.71428573,0.9047619,0.6666667,0.7619048,0.85714287,0.7619048,0.8095238,0.8095238,0.85714287,0.9047619,0.95238096
+TabSTAR (tuned),0.0952381,0.23809524,0.1904762,0.1904762,0.2857143,0.2857143,0.23809524,0.2857143,0.21428572,0.33333334,0.33333334,0.33333334,0.23809524,0.2857143,0.33333334,0.23809524,0.2857143,0.2857143,0.33333334,0.23809524,0.3809524,0.23809524,0.2857143,0.2857143,0.33333334,0.33333334,0.2857143,0.2857143,0.33333334,0.2857143,0.33333334,0.2857143,0.33333334,0.33333334,0.3809524,0.2857143,0.3809524,0.3809524,0.33333334,0.33333334,0.42857143,0.42857143,0.33333334,0.33333334,0.42857143,0.42857143,0.2857143,0.42857143,0.23809524,0.47619048,0.33333334,,0.61904764,0.52380955,0.47619048,0.47619048,0.42857143,0.47619048,0.42857143,0.47619048,0.47619048,0.52380955,0.9047619,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.61904764,0.7619048,0.6666667,0.6666667,0.8095238,0.6666667,0.71428573,0.71428573,0.8095238,0.9047619
+TabSTAR (tuned + ensembled),0.14285715,0.23809524,0.1904762,0.1904762,0.2857143,0.2857143,0.23809524,0.2857143,0.23809524,0.33333334,0.2857143,0.33333334,0.23809524,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.33333334,0.23809524,0.3809524,0.23809524,0.2857143,0.2857143,0.33333334,0.33333334,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.2857143,0.33333334,0.33333334,0.33333334,0.2857143,0.3809524,0.3809524,0.33333334,0.2857143,0.42857143,0.3809524,0.33333334,0.33333334,0.3809524,0.42857143,0.2857143,0.3809524,0.23809524,0.47619048,0.3809524,0.3809524,,0.52380955,0.42857143,0.47619048,0.42857143,0.47619048,0.42857143,0.47619048,0.47619048,0.52380955,0.95238096,0.52380955,0.5714286,0.52380955,0.5714286,0.6666667,0.6666667,0.71428573,0.6666667,0.61904764,0.7619048,0.6666667,0.6666667,0.71428573,0.8095238,0.9047619
+PerpetualBooster (tuned + ensembled),0.0,0.04761905,0.0952381,0.0952381,0.04761905,0.0952381,0.14285715,0.0952381,0.1904762,0.0952381,0.0952381,0.23809524,0.2857143,0.23809524,0.1904762,0.33333334,0.23809524,0.2857143,0.2857143,0.2857143,0.1904762,0.33333334,0.33333334,0.23809524,0.1904762,0.2857143,0.23809524,0.2857143,0.1904762,0.2857143,0.2857143,0.3809524,0.42857143,0.3809524,0.3809524,0.3809524,0.3809524,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.3809524,0.2857143,0.3809524,0.3809524,0.42857143,0.52380955,0.42857143,0.33333334,0.3809524,0.47619048,0.47619048,,0.47619048,0.5714286,0.5714286,0.52380955,0.47619048,0.47619048,0.52380955,0.71428573,0.47619048,0.61904764,0.61904764,0.61904764,0.7619048,0.61904764,0.61904764,0.7619048,0.6666667,0.7619048,0.71428573,0.8095238,0.8095238,0.8095238,0.7619048,0.9047619
+ExtraTrees (tuned + ensembled),0.04761905,0.0,0.0952381,0.14285715,0.0952381,0.04761905,0.0952381,0.0,0.14285715,0.14285715,0.14285715,0.1904762,0.14285715,0.23809524,0.1904762,0.1904762,0.1904762,0.1904762,0.2857143,0.2857143,0.1904762,0.1904762,0.23809524,0.1904762,0.2857143,0.2857143,0.23809524,0.3809524,0.23809524,0.33333334,0.3809524,0.1904762,0.14285715,0.1904762,0.33333334,0.14285715,0.33333334,0.3809524,0.3809524,0.2857143,0.33333334,0.33333334,0.33333334,0.23809524,0.3809524,0.3809524,0.3809524,0.2857143,0.42857143,0.42857143,0.45238096,0.52380955,0.5714286,0.52380955,,0.3809524,0.42857143,0.42857143,0.6666667,0.52380955,0.47619048,0.5714286,0.6666667,0.5714286,0.71428573,0.6666667,0.61904764,0.85714287,0.5714286,0.8095238,0.71428573,0.9047619,0.71428573,0.8095238,0.85714287,0.7619048,0.9047619,1.0
+FastaiMLP (tuned),0.0952381,0.0952381,0.0,0.0952381,0.1904762,0.14285715,0.14285715,0.1904762,0.1904762,0.1904762,0.14285715,0.1904762,0.14285715,0.23809524,0.23809524,0.23809524,0.14285715,0.2857143,0.23809524,0.23809524,0.1904762,0.23809524,0.23809524,0.2857143,0.23809524,0.23809524,0.23809524,0.2857143,0.23809524,0.33333334,0.2857143,0.33333334,0.23809524,0.23809524,0.2857143,0.2857143,0.2857143,0.23809524,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.33333334,0.23809524,0.33333334,0.33333334,0.3809524,0.3809524,0.2857143,0.42857143,0.52380955,0.52380955,0.42857143,0.61904764,,0.52380955,0.42857143,0.5714286,0.52380955,0.47619048,0.52380955,0.5714286,0.5714286,0.61904764,0.71428573,0.6666667,0.71428573,0.6666667,0.6666667,0.71428573,0.8095238,0.71428573,0.7619048,0.7619048,0.7619048,0.85714287,0.9047619
+iLTM (default),0.0952381,0.04761905,0.04761905,0.1904762,0.0952381,0.14285715,0.0952381,0.14285715,0.14285715,0.0952381,0.14285715,0.14285715,0.1904762,0.1904762,0.0952381,0.1904762,0.23809524,0.1904762,0.1904762,0.23809524,0.23809524,0.2857143,0.1904762,0.1904762,0.33333334,0.23809524,0.2857143,0.3809524,0.23809524,0.3809524,0.2857143,0.23809524,0.4047619,0.3809524,0.2857143,0.2857143,0.33333334,0.42857143,0.33333334,0.33333334,0.33333334,0.33333334,0.3809524,0.3809524,0.3809524,0.33333334,0.3809524,0.2857143,0.33333334,0.3809524,0.42857143,0.5714286,0.5714286,0.42857143,0.5714286,0.47619048,,0.42857143,0.5714286,0.52380955,0.3809524,0.3809524,0.61904764,0.6666667,0.5714286,0.5714286,0.47619048,0.61904764,0.6666667,0.61904764,0.61904764,0.8095238,0.71428573,0.71428573,0.71428573,0.8095238,0.85714287,0.9047619
+XGBoost (default),0.0,0.04761905,0.04761905,0.04761905,0.04761905,0.0,0.0952381,0.04761905,0.14285715,0.0952381,0.0952381,0.0952381,0.1904762,0.1904762,0.1904762,0.23809524,0.1904762,0.14285715,0.1904762,0.23809524,0.1904762,0.2857143,0.2857143,0.14285715,0.23809524,0.23809524,0.1904762,0.1904762,0.23809524,0.23809524,0.42857143,0.1904762,0.3809524,0.23809524,0.23809524,0.23809524,0.2857143,0.33333334,0.23809524,0.2857143,0.3809524,0.23809524,0.33333334,0.42857143,0.3809524,0.2857143,0.33333334,0.42857143,0.47619048,0.52380955,0.47619048,0.52380955,0.52380955,0.47619048,0.5714286,0.5714286,0.5714286,,0.5714286,0.47619048,0.61904764,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.6666667,0.61904764,0.71428573,0.71428573,0.71428573,0.85714287,0.7619048,0.8095238,0.7619048,0.8095238,0.9047619
+ExtraTrees (tuned),0.04761905,0.0,0.0952381,0.0952381,0.0952381,0.04761905,0.0952381,0.04761905,0.1904762,0.1904762,0.1904762,0.14285715,0.14285715,0.1904762,0.14285715,0.14285715,0.1904762,0.1904762,0.2857143,0.33333334,0.2857143,0.1904762,0.1904762,0.2857143,0.33333334,0.33333334,0.33333334,0.33333334,0.23809524,0.33333334,0.33333334,0.23809524,0.1904762,0.2857143,0.2857143,0.23809524,0.2857143,0.33333334,0.2857143,0.2857143,0.2857143,0.3809524,0.2857143,0.2857143,0.2857143,0.3809524,0.3809524,0.2857143,0.47619048,0.42857143,0.42857143,0.5714286,0.5714286,0.52380955,0.33333334,0.42857143,0.42857143,0.42857143,,0.52380955,0.3809524,0.47619048,0.71428573,0.47619048,0.6666667,0.61904764,0.5714286,0.7619048,0.52380955,0.7619048,0.6666667,0.85714287,0.6666667,0.71428573,0.71428573,0.7619048,0.95238096,0.95238096
+OrionMSP (default),0.0,0.0,0.0,0.0952381,0.04761905,0.04761905,0.0,0.0,0.11904762,0.0952381,0.0952381,0.1904762,0.0952381,0.16666667,0.26190478,0.04761905,0.1904762,0.2857143,0.23809524,0.23809524,0.23809524,0.14285715,0.0952381,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.3809524,0.2857143,0.2857143,0.2857143,0.3809524,0.2857143,0.33333334,0.33333334,0.23809524,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.33333334,0.2857143,0.1904762,0.33333334,0.33333334,0.33333334,0.3809524,0.33333334,0.3809524,0.52380955,0.52380955,0.52380955,0.47619048,0.47619048,0.47619048,0.52380955,0.47619048,,0.5714286,0.52380955,0.6666667,0.5714286,0.5714286,0.71428573,0.6666667,0.71428573,0.61904764,0.6666667,0.5714286,0.6666667,0.71428573,0.8095238,0.8095238,0.8095238,0.8095238,0.95238096
+LightGBM (default),0.04761905,0.04761905,0.0,0.04761905,0.04761905,0.0,0.04761905,0.04761905,0.14285715,0.0952381,0.0952381,0.04761905,0.0952381,0.1904762,0.14285715,0.23809524,0.14285715,0.14285715,0.14285715,0.1904762,0.0952381,0.2857143,0.33333334,0.04761905,0.23809524,0.1904762,0.23809524,0.1904762,0.1904762,0.1904762,0.3809524,0.0,0.2857143,0.2857143,0.33333334,0.0952381,0.2857143,0.33333334,0.1904762,0.23809524,0.33333334,0.1904762,0.33333334,0.3809524,0.33333334,0.23809524,0.33333334,0.33333334,0.42857143,0.52380955,0.45238096,0.52380955,0.52380955,0.47619048,0.52380955,0.52380955,0.61904764,0.3809524,0.61904764,0.42857143,,0.47619048,0.5714286,0.52380955,0.5714286,0.61904764,0.61904764,0.71428573,0.61904764,0.71428573,0.6666667,0.6666667,0.7619048,0.71428573,0.71428573,0.8095238,0.85714287,0.9047619
+PerpetualBooster (tuned),0.0,0.0,0.0,0.04761905,0.0,0.04761905,0.0952381,0.04761905,0.1904762,0.04761905,0.04761905,0.14285715,0.23809524,0.23809524,0.1904762,0.23809524,0.14285715,0.23809524,0.23809524,0.14285715,0.0952381,0.2857143,0.23809524,0.1904762,0.14285715,0.14285715,0.14285715,0.23809524,0.1904762,0.1904762,0.2857143,0.2857143,0.23809524,0.33333334,0.33333334,0.2857143,0.2857143,0.2857143,0.23809524,0.23809524,0.23809524,0.33333334,0.3809524,0.23809524,0.2857143,0.2857143,0.33333334,0.3809524,0.3809524,0.33333334,0.3809524,0.47619048,0.47619048,0.2857143,0.42857143,0.47619048,0.61904764,0.42857143,0.52380955,0.47619048,0.52380955,,0.47619048,0.47619048,0.5714286,0.6666667,0.7619048,0.61904764,0.61904764,0.6666667,0.61904764,0.6666667,0.71428573,0.7619048,0.71428573,0.7619048,0.7619048,0.9047619
+TabSTAR (default),0.0952381,0.23809524,0.1904762,0.1904762,0.2857143,0.23809524,0.23809524,0.23809524,0.1904762,0.2857143,0.23809524,0.2857143,0.23809524,0.23809524,0.2857143,0.23809524,0.23809524,0.23809524,0.2857143,0.23809524,0.2857143,0.1904762,0.2857143,0.2857143,0.2857143,0.33333334,0.2857143,0.1904762,0.33333334,0.2857143,0.23809524,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.33333334,0.33333334,0.42857143,0.2857143,0.3809524,0.23809524,0.3809524,0.33333334,0.0952381,0.04761905,0.52380955,0.33333334,0.42857143,0.3809524,0.42857143,0.2857143,0.33333334,0.42857143,0.52380955,,0.3809524,0.42857143,0.52380955,0.5714286,0.42857143,0.52380955,0.5714286,0.3809524,0.5714286,0.5714286,0.52380955,0.52380955,0.5714286,0.6666667,0.9047619
+RandomForest (tuned + ensembled),0.04761905,0.04761905,0.04761905,0.0952381,0.0,0.04761905,0.0952381,0.04761905,0.14285715,0.1904762,0.0952381,0.04761905,0.14285715,0.23809524,0.0952381,0.23809524,0.1904762,0.14285715,0.23809524,0.23809524,0.0952381,0.2857143,0.2857143,0.1904762,0.23809524,0.1904762,0.14285715,0.2857143,0.14285715,0.1904762,0.2857143,0.0952381,0.0952381,0.23809524,0.23809524,0.14285715,0.23809524,0.23809524,0.23809524,0.23809524,0.33333334,0.2857143,0.2857143,0.2857143,0.33333334,0.2857143,0.2857143,0.33333334,0.2857143,0.47619048,0.33333334,0.42857143,0.47619048,0.3809524,0.42857143,0.42857143,0.33333334,0.42857143,0.52380955,0.42857143,0.47619048,0.52380955,0.61904764,,0.71428573,0.52380955,0.52380955,0.71428573,0.6666667,0.85714287,0.61904764,0.71428573,0.7619048,0.71428573,0.71428573,0.7619048,0.85714287,0.95238096
+RandomForest (tuned),0.04761905,0.0,0.0,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.04761905,0.0952381,0.04761905,0.071428575,0.14285715,0.11904762,0.04761905,0.1904762,0.071428575,0.1904762,0.21428572,0.23809524,0.0952381,0.23809524,0.23809524,0.0952381,0.23809524,0.14285715,0.14285715,0.23809524,0.0952381,0.1904762,0.26190478,0.04761905,0.1904762,0.23809524,0.23809524,0.0952381,0.23809524,0.23809524,0.1904762,0.14285715,0.2857143,0.33333334,0.2857143,0.23809524,0.2857143,0.1904762,0.1904762,0.2857143,0.2857143,0.3809524,0.33333334,0.42857143,0.42857143,0.3809524,0.2857143,0.3809524,0.42857143,0.42857143,0.33333334,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,,0.61904764,0.47619048,0.6666667,0.47619048,0.7380952,0.71428573,0.7619048,0.8095238,0.71428573,0.71428573,0.71428573,0.85714287,0.95238096
+TabFlex (default),0.0,0.0952381,0.0,0.0,0.14285715,0.1904762,0.0952381,0.1904762,0.14285715,0.1904762,0.0952381,0.1904762,0.04761905,0.14285715,0.23809524,0.0952381,0.14285715,0.1904762,0.14285715,0.14285715,0.14285715,0.04761905,0.1904762,0.23809524,0.14285715,0.23809524,0.23809524,0.23809524,0.2857143,0.1904762,0.14285715,0.2857143,0.23809524,0.14285715,0.0952381,0.2857143,0.1904762,0.23809524,0.1904762,0.2857143,0.1904762,0.33333334,0.1904762,0.1904762,0.23809524,0.14285715,0.23809524,0.2857143,0.33333334,0.1904762,0.14285715,0.42857143,0.47619048,0.3809524,0.33333334,0.2857143,0.42857143,0.42857143,0.3809524,0.2857143,0.3809524,0.33333334,0.47619048,0.47619048,0.3809524,,0.52380955,0.52380955,0.47619048,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.7619048,0.85714287
+PerpetualBooster (default),0.0,0.04761905,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.14285715,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.04761905,0.1904762,0.0952381,0.1904762,0.14285715,0.0952381,0.0952381,0.1904762,0.14285715,0.1904762,0.0952381,0.0952381,0.0952381,0.23809524,0.0952381,0.23809524,0.14285715,0.23809524,0.1904762,0.1904762,0.0952381,0.23809524,0.1904762,0.1904762,0.23809524,0.1904762,0.1904762,0.1904762,0.23809524,0.14285715,0.2857143,0.1904762,0.2857143,0.14285715,0.2857143,0.14285715,0.23809524,0.42857143,0.42857143,0.23809524,0.3809524,0.33333334,0.52380955,0.2857143,0.42857143,0.33333334,0.3809524,0.23809524,0.42857143,0.47619048,0.52380955,0.47619048,,0.5714286,0.47619048,0.5714286,0.3809524,0.61904764,0.61904764,0.6666667,0.6666667,0.6666667,0.6666667,0.9047619
+ExtraTrees (default),0.04761905,0.04761905,0.0,0.14285715,0.0952381,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.14285715,0.0952381,0.23809524,0.0952381,0.04761905,0.14285715,0.2857143,0.14285715,0.23809524,0.0952381,0.14285715,0.14285715,0.1904762,0.2857143,0.23809524,0.23809524,0.23809524,0.1904762,0.23809524,0.2857143,0.14285715,0.1904762,0.1904762,0.1904762,0.1904762,0.14285715,0.2857143,0.1904762,0.23809524,0.1904762,0.2857143,0.1904762,0.1904762,0.1904762,0.2857143,0.23809524,0.14285715,0.23809524,0.23809524,0.2857143,0.2857143,0.33333334,0.3809524,0.14285715,0.2857143,0.3809524,0.33333334,0.23809524,0.2857143,0.2857143,0.3809524,0.5714286,0.2857143,0.33333334,0.47619048,0.42857143,,0.3809524,0.61904764,0.47619048,0.52380955,0.52380955,0.61904764,0.61904764,0.61904764,0.71428573,0.9047619
+TorchMLP (default),0.0,0.04761905,0.0,0.0,0.04761905,0.0952381,0.0952381,0.0952381,0.0952381,0.0952381,0.04761905,0.0952381,0.14285715,0.14285715,0.04761905,0.1904762,0.04761905,0.14285715,0.1904762,0.04761905,0.04761905,0.14285715,0.1904762,0.1904762,0.04761905,0.0952381,0.0952381,0.1904762,0.0952381,0.14285715,0.1904762,0.14285715,0.0952381,0.1904762,0.04761905,0.1904762,0.1904762,0.0952381,0.14285715,0.23809524,0.1904762,0.23809524,0.1904762,0.14285715,0.23809524,0.0952381,0.1904762,0.1904762,0.1904762,0.23809524,0.0952381,0.3809524,0.33333334,0.3809524,0.42857143,0.33333334,0.33333334,0.3809524,0.47619048,0.3809524,0.3809524,0.3809524,0.47619048,0.33333334,0.52380955,0.52380955,0.52380955,0.61904764,,0.5714286,0.5714286,0.61904764,0.5714286,0.61904764,0.6666667,0.71428573,0.7619048,0.9047619
+RandomForest (default),0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.04761905,0.071428575,0.14285715,0.11904762,0.04761905,0.0952381,0.071428575,0.1904762,0.16666667,0.23809524,0.04761905,0.23809524,0.14285715,0.04761905,0.23809524,0.14285715,0.14285715,0.14285715,0.0952381,0.1904762,0.26190478,0.0952381,0.14285715,0.23809524,0.1904762,0.14285715,0.14285715,0.23809524,0.2857143,0.14285715,0.1904762,0.23809524,0.1904762,0.23809524,0.23809524,0.2857143,0.2857143,0.1904762,0.23809524,0.2857143,0.33333334,0.23809524,0.2857143,0.23809524,0.1904762,0.33333334,0.3809524,0.2857143,0.23809524,0.33333334,0.2857143,0.33333334,0.42857143,0.14285715,0.26190478,0.42857143,0.42857143,0.3809524,0.42857143,,0.52380955,0.61904764,0.5714286,0.61904764,0.61904764,0.6666667,0.7619048,0.95238096
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.04761905,0.0,0.0952381,0.0952381,0.0,0.04761905,0.14285715,0.1904762,0.0952381,0.0952381,0.04761905,0.0,0.14285715,0.1904762,0.14285715,0.14285715,0.1904762,0.1904762,0.23809524,0.14285715,0.1904762,0.0952381,0.14285715,0.04761905,0.1904762,0.0952381,0.23809524,0.23809524,0.1904762,0.1904762,0.23809524,0.1904762,0.14285715,0.14285715,0.14285715,0.23809524,0.14285715,0.23809524,0.04761905,0.23809524,0.33333334,0.33333334,0.33333334,0.2857143,0.2857143,0.3809524,0.2857143,0.33333334,0.42857143,0.33333334,0.3809524,0.61904764,0.3809524,0.2857143,0.42857143,0.61904764,0.52380955,0.42857143,0.47619048,,0.61904764,0.61904764,0.5714286,0.5714286,0.5714286,0.8095238,0.9047619
+KNN (tuned + ensembled),0.04761905,0.0,0.0,0.04761905,0.0,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0,0.0952381,0.0952381,0.0952381,0.0952381,0.04761905,0.04761905,0.14285715,0.04761905,0.04761905,0.14285715,0.0,0.14285715,0.0952381,0.04761905,0.0952381,0.14285715,0.14285715,0.1904762,0.14285715,0.0952381,0.04761905,0.04761905,0.0952381,0.14285715,0.1904762,0.1904762,0.0952381,0.23809524,0.21428572,0.04761905,0.04761905,0.14285715,0.14285715,0.1904762,0.14285715,0.14285715,0.14285715,0.14285715,0.33333334,0.3809524,0.23809524,0.0952381,0.1904762,0.1904762,0.2857143,0.14285715,0.33333334,0.33333334,0.33333334,0.42857143,0.2857143,0.23809524,0.42857143,0.3809524,0.47619048,0.3809524,0.3809524,0.3809524,,0.6666667,0.61904764,0.61904764,0.61904764,0.95238096,0.95238096
+xRFM (default),0.0,0.0,0.0,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.0,0.0952381,0.14285715,0.14285715,0.14285715,0.04761905,0.04761905,0.0952381,0.14285715,0.1904762,0.14285715,0.0952381,0.0952381,0.14285715,0.1904762,0.14285715,0.14285715,0.04761905,0.14285715,0.14285715,0.14285715,0.14285715,0.23809524,0.1904762,0.14285715,0.14285715,0.0952381,0.0952381,0.1904762,0.1904762,0.1904762,0.1904762,0.04761905,0.2857143,0.1904762,0.23809524,0.1904762,0.23809524,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.33333334,0.2857143,0.23809524,0.2857143,0.42857143,0.23809524,0.1904762,0.42857143,0.3809524,0.47619048,0.42857143,0.42857143,0.3809524,0.33333334,,0.52380955,0.52380955,0.52380955,0.6666667,0.85714287
+Linear (tuned + ensembled),0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.0952381,0.0952381,0.04761905,0.04761905,0.04761905,0.0952381,0.0952381,0.04761905,0.14285715,0.04761905,0.14285715,0.04761905,0.04761905,0.14285715,0.04761905,0.0952381,0.0952381,0.14285715,0.14285715,0.04761905,0.04761905,0.1904762,0.0952381,0.0952381,0.14285715,0.14285715,0.0952381,0.2857143,0.0952381,0.04761905,0.1904762,0.14285715,0.14285715,0.14285715,0.14285715,0.23809524,0.1904762,0.33333334,0.33333334,0.1904762,0.1904762,0.23809524,0.2857143,0.23809524,0.2857143,0.1904762,0.2857143,0.23809524,0.47619048,0.2857143,0.2857143,0.2857143,0.33333334,0.3809524,0.3809524,0.3809524,0.42857143,0.3809524,0.47619048,,0.7380952,0.61904764,0.5714286,0.8095238
+Linear (tuned),0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.071428575,0.0952381,0.04761905,0.04761905,0.04761905,0.0952381,0.0952381,0.04761905,0.0952381,0.04761905,0.14285715,0.04761905,0.04761905,0.0952381,0.04761905,0.0952381,0.0952381,0.0952381,0.14285715,0.04761905,0.04761905,0.14285715,0.04761905,0.0952381,0.14285715,0.14285715,0.14285715,0.23809524,0.0952381,0.04761905,0.1904762,0.14285715,0.14285715,0.14285715,0.14285715,0.1904762,0.1904762,0.2857143,0.33333334,0.1904762,0.14285715,0.23809524,0.2857143,0.1904762,0.2857143,0.1904762,0.2857143,0.2857143,0.47619048,0.2857143,0.2857143,0.2857143,0.33333334,0.3809524,0.33333334,0.3809524,0.42857143,0.3809524,0.47619048,0.26190478,,0.52380955,0.52380955,0.7619048
+Linear (default),0.0,0.0,0.0,0.0,0.04761905,0.0952381,0.04761905,0.0952381,0.04761905,0.04761905,0.0,0.0952381,0.0952381,0.04761905,0.0952381,0.14285715,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.0952381,0.14285715,0.0952381,0.0952381,0.04761905,0.04761905,0.14285715,0.04761905,0.04761905,0.04761905,0.0952381,0.1904762,0.04761905,0.04761905,0.0952381,0.0952381,0.14285715,0.0952381,0.14285715,0.1904762,0.2857143,0.14285715,0.04761905,0.1904762,0.04761905,0.04761905,0.1904762,0.14285715,0.1904762,0.14285715,0.2857143,0.2857143,0.1904762,0.23809524,0.23809524,0.1904762,0.23809524,0.23809524,0.1904762,0.1904762,0.23809524,0.42857143,0.23809524,0.2857143,0.2857143,0.33333334,0.3809524,0.2857143,0.33333334,0.42857143,0.3809524,0.47619048,0.3809524,0.47619048,,0.5714286,0.6666667
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.04761905,0.0,0.04761905,0.04761905,0.0,0.0,0.04761905,0.0,0.0,0.0,0.04761905,0.0,0.0,0.0,0.0,0.04761905,0.0,0.04761905,0.0,0.04761905,0.0,0.0,0.0952381,0.04761905,0.04761905,0.04761905,0.0952381,0.0,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.1904762,0.04761905,0.04761905,0.04761905,0.04761905,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.1904762,0.1904762,0.23809524,0.0952381,0.14285715,0.14285715,0.1904762,0.04761905,0.1904762,0.14285715,0.23809524,0.33333334,0.14285715,0.14285715,0.23809524,0.33333334,0.2857143,0.23809524,0.23809524,0.1904762,0.04761905,0.33333334,0.42857143,0.47619048,0.42857143,,0.9047619
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.04761905,0.0,0.04761905,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.04761905,0.0952381,0.0,0.0,0.0,0.04761905,0.0,0.0,0.04761905,0.04761905,0.0,0.0,0.04761905,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.0,0.0952381,0.0952381,0.0952381,0.04761905,0.04761905,0.0952381,0.0952381,0.0952381,0.04761905,0.04761905,0.14285715,0.0952381,0.0952381,0.0952381,0.04761905,0.0952381,0.04761905,0.14285715,0.1904762,0.23809524,0.33333334,0.0952381,
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/leaderboard_overview_explorer.html
index a76b4971e27e2e0ca441b1251466edc37523d892..bbde1604bac83b551d187baea2fee4f70d2ed59f 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Classification | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..5accfb1217257bfd7a80b72cd348885b46fe1f6e
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_classification/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1750.0,"score":0.795,"rank":7.05,"hrank":1.67,"imp":5.548,"train_time":40.79,"predict_time":7.056,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+160\/-138","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1622.0,"score":0.613,"rank":11.95,"hrank":4.05,"imp":8.564,"train_time":268.44,"predict_time":4.668,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-105","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1615.0,"score":0.587,"rank":12.3,"hrank":4.68,"imp":10.518,"train_time":3.52,"predict_time":0.385,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-75","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1587.0,"score":0.538,"rank":13.76,"hrank":7.82,"imp":12.162,"train_time":5.39,"predict_time":0.531,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-65","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1571.0,"score":0.513,"rank":14.62,"hrank":5.95,"imp":10.891,"train_time":2063.62,"predict_time":9.938,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-73","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1558.0,"score":0.511,"rank":15.39,"hrank":7.16,"imp":12.424,"train_time":2.04,"predict_time":0.182,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-86","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1532.0,"score":0.477,"rank":16.97,"hrank":9.74,"imp":13.159,"train_time":2063.62,"predict_time":1.086,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-70","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1530.0,"score":0.454,"rank":17.05,"hrank":10.42,"imp":12.906,"train_time":5.63,"predict_time":0.771,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-61","imputed":false},{"position":8,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1521.0,"score":0.462,"rank":17.66,"hrank":7.08,"imp":13.384,"train_time":1815.89,"predict_time":1.455,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-59","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1471.0,"score":0.391,"rank":21.08,"hrank":12.4,"imp":14.77,"train_time":2774.17,"predict_time":16.363,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-49","imputed":false},{"position":10,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1408.0,"score":0.333,"rank":25.83,"hrank":13.74,"imp":16.189,"train_time":2774.17,"predict_time":0.39,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-53","imputed":false},{"position":11,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1402.0,"score":0.33,"rank":26.37,"hrank":12.11,"imp":16.493,"train_time":2207.17,"predict_time":1.894,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-61","imputed":false},{"position":12,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1391.0,"score":0.349,"rank":27.24,"hrank":7.79,"imp":16.096,"train_time":18.2,"predict_time":4.692,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-70","imputed":false},{"position":13,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1388.0,"score":0.293,"rank":27.47,"hrank":16.02,"imp":16.589,"train_time":12550.1,"predict_time":427.59,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-65","imputed":false},{"position":14,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1386.0,"score":0.293,"rank":27.63,"hrank":14.3,"imp":15.846,"train_time":1087.73,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-52","imputed":false},{"position":15,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1384.0,"score":0.296,"rank":27.8,"hrank":12.54,"imp":17.092,"train_time":2.1,"predict_time":0.186,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-61","imputed":false},{"position":16,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1381.0,"score":0.253,"rank":28.08,"hrank":20.62,"imp":17.096,"train_time":391.27,"predict_time":1.243,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-44","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1380.0,"score":0.279,"rank":28.17,"hrank":17.47,"imp":16.213,"train_time":1087.73,"predict_time":0.348,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-48","imputed":false},{"position":18,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1365.0,"score":0.291,"rank":29.41,"hrank":14.78,"imp":17.286,"train_time":2207.17,"predict_time":0.234,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-60","imputed":false},{"position":19,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1365.0,"score":0.282,"rank":29.45,"hrank":13.09,"imp":16.671,"train_time":6067.42,"predict_time":404.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-58","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1359.0,"score":0.245,"rank":29.91,"hrank":19.98,"imp":17.607,"train_time":682.62,"predict_time":0.971,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-64","imputed":false},{"position":21,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1359.0,"score":0.269,"rank":29.96,"hrank":11.58,"imp":17.469,"train_time":9.08,"predict_time":2.007,"imputed_pct":5.26,"hardware":"GPU","elo_ci":"+74\/-69","imputed":true},{"position":22,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1348.0,"score":0.251,"rank":30.91,"hrank":17.15,"imp":17.618,"train_time":4853.47,"predict_time":0.473,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-68","imputed":false},{"position":23,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1346.0,"score":0.24,"rank":31.08,"hrank":19.67,"imp":17.315,"train_time":4.89,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-54","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1335.0,"score":0.263,"rank":32.0,"hrank":18.53,"imp":17.669,"train_time":4853.47,"predict_time":9.385,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+93\/-80","imputed":false},{"position":25,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1333.0,"score":0.249,"rank":32.14,"hrank":21.22,"imp":18.741,"train_time":500.4,"predict_time":0.416,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-69","imputed":false},{"position":26,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1332.0,"score":0.21,"rank":32.29,"hrank":25.1,"imp":17.998,"train_time":391.27,"predict_time":0.186,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-49","imputed":false},{"position":27,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1331.0,"score":0.221,"rank":32.32,"hrank":15.22,"imp":17.82,"train_time":682.62,"predict_time":0.183,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-64","imputed":false},{"position":28,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1320.0,"score":0.235,"rank":33.25,"hrank":17.05,"imp":18.396,"train_time":6067.42,"predict_time":39.564,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-69","imputed":false},{"position":29,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1319.0,"score":0.213,"rank":33.38,"hrank":20.28,"imp":18.777,"train_time":1.2,"predict_time":0.064,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-63","imputed":false},{"position":30,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1319.0,"score":0.272,"rank":33.41,"hrank":16.18,"imp":19.63,"train_time":2916.21,"predict_time":17.593,"imputed_pct":31.58,"hardware":"GPU","elo_ci":"+96\/-80","imputed":true},{"position":31,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1312.0,"score":0.217,"rank":33.97,"hrank":22.49,"imp":19.654,"train_time":250.99,"predict_time":2.035,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-57","imputed":false},{"position":32,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1308.0,"score":0.226,"rank":34.32,"hrank":19.47,"imp":19.133,"train_time":500.4,"predict_time":0.031,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-69","imputed":false},{"position":33,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1306.0,"score":0.24,"rank":34.47,"hrank":16.35,"imp":19.926,"train_time":118.42,"predict_time":2.667,"imputed_pct":31.58,"hardware":"GPU","elo_ci":"+85\/-75","imputed":true},{"position":34,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1300.0,"score":0.216,"rank":35.0,"hrank":20.13,"imp":18.472,"train_time":1053.8,"predict_time":2.067,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-57","imputed":false},{"position":35,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1300.0,"score":0.18,"rank":35.03,"hrank":24.55,"imp":18.336,"train_time":12550.1,"predict_time":61.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-53","imputed":false},{"position":36,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1300.0,"score":0.158,"rank":35.08,"hrank":24.51,"imp":19.478,"train_time":1449.14,"predict_time":0.149,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-50","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1288.0,"score":0.195,"rank":36.12,"hrank":22.12,"imp":19.542,"train_time":7.26,"predict_time":0.25,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-58","imputed":false},{"position":38,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1264.0,"score":0.193,"rank":38.22,"hrank":21.59,"imp":21.105,"train_time":2916.21,"predict_time":0.271,"imputed_pct":31.58,"hardware":"GPU","elo_ci":"+78\/-74","imputed":true},{"position":39,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1264.0,"score":0.149,"rank":38.24,"hrank":30.29,"imp":19.356,"train_time":2480.8,"predict_time":2.175,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-51","imputed":false},{"position":40,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1256.0,"score":0.155,"rank":38.95,"hrank":17.24,"imp":20.024,"train_time":1053.8,"predict_time":0.105,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-57","imputed":false},{"position":41,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1256.0,"score":0.121,"rank":38.97,"hrank":27.93,"imp":20.063,"train_time":10.77,"predict_time":1.786,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-48","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1255.0,"score":0.164,"rank":39.05,"hrank":27.72,"imp":20.678,"train_time":1.92,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-77","imputed":false},{"position":43,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1248.0,"score":0.188,"rank":39.64,"hrank":18.41,"imp":21.54,"train_time":13.62,"predict_time":1.984,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-94","imputed":false},{"position":44,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1245.0,"score":0.108,"rank":39.95,"hrank":29.05,"imp":20.428,"train_time":1449.14,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-55","imputed":false},{"position":45,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1240.0,"score":0.124,"rank":40.38,"hrank":18.43,"imp":21.011,"train_time":5.48,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-50","imputed":false},{"position":46,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1235.0,"score":0.17,"rank":40.83,"hrank":17.68,"imp":21.345,"train_time":3.3,"predict_time":0.305,"imputed_pct":31.58,"hardware":"GPU","elo_ci":"+82\/-77","imputed":true},{"position":47,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1210.0,"score":0.142,"rank":43.01,"hrank":24.0,"imp":21.378,"train_time":47.89,"predict_time":39.444,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-63","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1206.0,"score":0.117,"rank":43.36,"hrank":31.21,"imp":22.197,"train_time":622.86,"predict_time":5.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-85","imputed":false},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1199.0,"score":0.117,"rank":43.95,"hrank":19.17,"imp":20.881,"train_time":1.88,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-68","imputed":false},{"position":50,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1196.0,"score":0.112,"rank":44.2,"hrank":32.32,"imp":21.548,"train_time":2480.8,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-62","imputed":false},{"position":51,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1191.0,"score":0.103,"rank":44.68,"hrank":30.47,"imp":21.773,"train_time":282.36,"predict_time":0.86,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-65","imputed":false},{"position":52,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1178.0,"score":0.087,"rank":45.83,"hrank":17.75,"imp":22.961,"train_time":13.35,"predict_time":0.352,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+60\/-60","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1166.0,"score":0.096,"rank":46.84,"hrank":36.09,"imp":22.829,"train_time":327.37,"predict_time":0.708,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-72","imputed":false},{"position":54,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1158.0,"score":0.082,"rank":47.49,"hrank":30.99,"imp":23.348,"train_time":282.36,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-68","imputed":false},{"position":55,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1149.0,"score":0.082,"rank":48.28,"hrank":39.95,"imp":22.038,"train_time":2.25,"predict_time":0.159,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-60","imputed":false},{"position":56,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1134.0,"score":0.083,"rank":49.49,"hrank":28.01,"imp":24.406,"train_time":267.89,"predict_time":62.492,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+70\/-67","imputed":false},{"position":57,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1131.0,"score":0.072,"rank":49.79,"hrank":34.96,"imp":24.105,"train_time":622.86,"predict_time":0.345,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-76","imputed":false},{"position":58,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1127.0,"score":0.052,"rank":50.13,"hrank":42.84,"imp":23.592,"train_time":327.37,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-61","imputed":false},{"position":59,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1113.0,"score":0.116,"rank":51.21,"hrank":18.57,"imp":26.057,"train_time":38846.7,"predict_time":4.269,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-90","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1103.0,"score":0.107,"rank":52.04,"hrank":20.56,"imp":26.27,"train_time":38846.7,"predict_time":19.475,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+91\/-95","imputed":false},{"position":61,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1063.0,"score":0.035,"rank":55.14,"hrank":47.68,"imp":26.359,"train_time":16.07,"predict_time":2.876,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+56\/-70","imputed":false},{"position":62,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1058.0,"score":0.075,"rank":55.53,"hrank":43.89,"imp":31.215,"train_time":175.23,"predict_time":0.654,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-84","imputed":false},{"position":63,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1042.0,"score":0.026,"rank":56.68,"hrank":50.77,"imp":26.35,"train_time":5.73,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-73","imputed":false},{"position":64,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1017.0,"score":0.088,"rank":58.5,"hrank":18.06,"imp":32.474,"train_time":403.92,"predict_time":4.972,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-122","imputed":false},{"position":65,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1007.0,"score":0.04,"rank":59.17,"hrank":53.54,"imp":32.74,"train_time":175.23,"predict_time":0.2,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-91","imputed":false},{"position":66,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1000.0,"score":0.025,"rank":59.61,"hrank":54.82,"imp":27.678,"train_time":3.13,"predict_time":0.334,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-66","imputed":false},{"position":67,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.016,"rank":59.62,"hrank":49.84,"imp":29.459,"train_time":0.37,"predict_time":0.038,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-85","imputed":false},{"position":68,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":998.0,"score":0.034,"rank":59.78,"hrank":50.43,"imp":29.84,"train_time":140.69,"predict_time":2.876,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-95","imputed":false},{"position":69,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":984.0,"score":0.033,"rank":60.66,"hrank":51.08,"imp":29.507,"train_time":1.75,"predict_time":0.268,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-88","imputed":false},{"position":70,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":976.0,"score":0.038,"rank":61.18,"hrank":53.57,"imp":30.695,"train_time":3.48,"predict_time":0.677,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-87","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":971.0,"score":0.036,"rank":61.47,"hrank":51.98,"imp":32.42,"train_time":1.04,"predict_time":0.192,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-122","imputed":false},{"position":72,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":966.0,"score":0.032,"rank":61.78,"hrank":23.55,"imp":33.28,"train_time":256.18,"predict_time":0.36,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-99","imputed":false},{"position":73,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":944.0,"score":0.023,"rank":63.11,"hrank":34.71,"imp":33.955,"train_time":256.18,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-108","imputed":false},{"position":74,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":930.0,"score":0.032,"rank":63.86,"hrank":40.5,"imp":38.404,"train_time":20.44,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-111","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":882.0,"score":0.007,"rank":66.37,"hrank":62.24,"imp":36.482,"train_time":1.32,"predict_time":0.117,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-99","imputed":false},{"position":76,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":838.0,"score":0.018,"rank":68.37,"hrank":63.87,"imp":36.995,"train_time":140.69,"predict_time":0.188,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-105","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":565.0,"score":0.0,"rank":75.74,"hrank":75.47,"imp":51.26,"train_time":0.18,"predict_time":0.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-124","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/pareto_front_explorer.html
index 9c724e74a63fd27241bcd70040f3715856cf65f2..22bf7f90bd5d150d2d6dc0eab7d81f9a39842a6f 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0351934444,"imp":19.6537536854,"elo":1312.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.024809806,"imp":17.3150528678,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0341120659,"imp":15.8464457377,"elo":1386.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3481993779,"imp":16.2125121758,"elo":1379.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0396842484,"imp":20.6775753505,"elo":1254.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306719592,"imp":19.1331421267,"elo":1308.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4161492489,"imp":18.7410428806,"elo":1333.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0162731597,"imp":21.0107924786,"elo":1239.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0167928172,"imp":20.4281215893,"elo":1244.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1493309259,"imp":19.4776411566,"elo":1299.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2675026243,"imp":29.5066303657,"elo":984.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721048132,"imp":23.3476574197,"elo":1158.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8604796191,"imp":21.7734951095,"elo":1190.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3340852103,"imp":27.6775315034,"elo":1000.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3452999313,"imp":24.1045086163,"elo":1130.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0368770459,"imp":22.1968928669,"elo":1206.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1585029211,"imp":22.0381919082,"elo":1148.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1855840981,"imp":17.998245484,"elo":1331.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.2430091096,"imp":17.0958003435,"elo":1381.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.6915866654,"imp":16.0959184486,"elo":1391.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.6670530253,"imp":19.925959121,"elo":1306.5,"imputed":true,"imputed_pct":31.5789473684,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"imp":22.9612918562,"elo":1177.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4731015689,"imp":17.6184824655,"elo":1347.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"imp":17.6689551215,"elo":1334.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.876262961,"imp":26.3594045937,"elo":1063.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0443776494,"imp":38.403551102,"elo":930.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1999302717,"imp":32.7395847294,"elo":1006.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6542696803,"imp":31.2151516083,"elo":1058.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0376682658,"imp":29.459321786,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0845096703,"imp":23.591835267,"elo":1126.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7082779811,"imp":22.8291563924,"elo":1165.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7864396959,"imp":20.0631203141,"elo":1255.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3898971973,"imp":16.1890919334,"elo":1408.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":16.3626514212,"imp":14.7695848003,"elo":1470.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7713317233,"imp":12.9062996177,"elo":1530.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0864910732,"imp":13.1588726501,"elo":1531.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.9380601197,"imp":10.8909589894,"elo":1571.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.9837292889,"imp":21.5403298343,"elo":1248.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"imp":21.3779360936,"elo":1210.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.5640473457,"imp":18.3956873376,"elo":1320.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":404.0529881985,"imp":16.6705413456,"elo":1364.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1861515215,"imp":17.0923762106,"elo":1384.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"imp":5.5483540114,"elo":1749.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1916685529,"imp":32.4203635865,"elo":971.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":2.0066462138,"imp":17.4688497365,"elo":1358.6,"imputed":true,"imputed_pct":5.2631578947,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"imp":12.4241131571,"elo":1557.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2496062173,"imp":19.542360491,"elo":1287.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2337420667,"imp":17.2856932714,"elo":1365.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.8937013268,"imp":16.4930559746,"elo":1401.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"imp":12.1615024256,"elo":1586.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3846051827,"imp":10.5179792428,"elo":1615.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3052741537,"imp":21.3452090401,"elo":1234.8,"imputed":true,"imputed_pct":31.5789473684,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2712952691,"imp":21.1050043397,"elo":1264.2,"imputed":true,"imputed_pct":31.5789473684,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.5931496568,"imp":19.6295950754,"elo":1318.7,"imputed":true,"imputed_pct":31.5789473684,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9719997826,"imp":32.4744091093,"elo":1016.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2687972891,"imp":26.0568333519,"elo":1113.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.4748027028,"imp":26.2699222413,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0643226705,"imp":18.7766103518,"elo":1319.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1483357715,"imp":26.3504187059,"elo":1042.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1166104296,"imp":21.5484213473,"elo":1196.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1753468186,"imp":19.3559768486,"elo":1264.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209262941,"imp":20.8813545514,"elo":1199.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1829513734,"imp":17.8195759499,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.9713359475,"imp":17.6065215295,"elo":1359.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":62.4915524975,"imp":24.4055566951,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":61.3096906599,"imp":18.3357174571,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":427.5904998868,"imp":16.5890768836,"elo":1388.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.6769281514,"imp":30.6951398147,"elo":975.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1053014255,"imp":20.0239827209,"elo":1256.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0671352109,"imp":18.4717714757,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0351934444,"x_train":250.9922109033,"imp":19.6537536854,"elo":1312.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.024809806,"x_train":4.8861388456,"imp":17.3150528678,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0341120659,"x_train":1087.7269134101,"imp":15.8464457377,"elo":1386.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3481993779,"x_train":1087.7269134101,"imp":16.2125121758,"elo":1379.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0396842484,"x_train":1.9155810093,"imp":20.6775753505,"elo":1254.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0306719592,"x_train":500.3975293451,"imp":19.1331421267,"elo":1308.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4161492489,"x_train":500.3975293451,"imp":18.7410428806,"elo":1333.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0162731597,"x_train":5.4796099733,"imp":21.0107924786,"elo":1239.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0167928172,"x_train":1449.1413756847,"imp":20.4281215893,"elo":1244.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1493309259,"x_train":1449.1413756847,"imp":19.4776411566,"elo":1299.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2675026243,"x_train":1.7477789606,"imp":29.5066303657,"elo":984.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0721048132,"x_train":282.3556156822,"imp":23.3476574197,"elo":1158.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8604796191,"x_train":282.3556156822,"imp":21.7734951095,"elo":1190.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3340852103,"x_train":3.1251204631,"imp":27.6775315034,"elo":1000.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3452999313,"x_train":622.8560318368,"imp":24.1045086163,"elo":1130.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0368770459,"x_train":622.8560318368,"imp":22.1968928669,"elo":1206.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1585029211,"x_train":2.2519280723,"imp":22.0381919082,"elo":1148.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.1855840981,"x_train":391.2703614285,"imp":17.998245484,"elo":1331.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.2430091096,"x_train":391.2703614285,"imp":17.0958003435,"elo":1381.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":4.6915866654,"x_train":18.1964917538,"imp":16.0959184486,"elo":1391.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.6670530253,"x_train":118.4223166146,"imp":19.925959121,"elo":1306.5,"imputed":true,"imputed_pct":31.5789473684,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"x_train":13.3530116618,"imp":22.9612918562,"elo":1177.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4731015689,"x_train":4853.4730504877,"imp":17.6184824655,"elo":1347.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"x_train":4853.4730504877,"imp":17.6689551215,"elo":1334.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.876262961,"x_train":16.0704711344,"imp":26.3594045937,"elo":1063.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0443776494,"x_train":20.4435114187,"imp":38.403551102,"elo":930.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1999302717,"x_train":175.2300293672,"imp":32.7395847294,"elo":1006.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6542696803,"x_train":175.2300293672,"imp":31.2151516083,"elo":1058.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0376682658,"x_train":0.3738939375,"imp":29.459321786,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0845096703,"x_train":327.3662383914,"imp":23.591835267,"elo":1126.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7082779811,"x_train":327.3662383914,"imp":22.8291563924,"elo":1165.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.7864396959,"x_train":10.7684699326,"imp":20.0631203141,"elo":1255.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3898971973,"x_train":2774.1715903227,"imp":16.1890919334,"elo":1408.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":16.3626514212,"x_train":2774.1715903227,"imp":14.7695848003,"elo":1470.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7713317233,"x_train":5.6257151642,"imp":12.9062996177,"elo":1530.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0864910732,"x_train":2063.6226141883,"imp":13.1588726501,"elo":1531.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.9380601197,"x_train":2063.6226141883,"imp":10.8909589894,"elo":1571.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.9837292889,"x_train":13.6228654067,"imp":21.5403298343,"elo":1248.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"x_train":47.8871994,"imp":21.3779360936,"elo":1210.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.5640473457,"x_train":6067.4246083224,"imp":18.3956873376,"elo":1320.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":404.0529881985,"x_train":6067.4246083224,"imp":16.6705413456,"elo":1364.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1861515215,"x_train":2.0990306801,"imp":17.0923762106,"elo":1384.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"x_train":40.7899647401,"imp":5.5483540114,"elo":1749.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1916685529,"x_train":1.0350786389,"imp":32.4203635865,"elo":971.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":2.0066462138,"x_train":9.0755702975,"imp":17.4688497365,"elo":1358.6,"imputed":true,"imputed_pct":5.2631578947,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"x_train":2.0391419632,"imp":12.4241131571,"elo":1557.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2496062173,"x_train":7.2637875629,"imp":19.542360491,"elo":1287.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.2337420667,"x_train":2207.1695731088,"imp":17.2856932714,"elo":1365.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":1.8937013268,"x_train":2207.1695731088,"imp":16.4930559746,"elo":1401.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"x_train":5.3935958382,"imp":12.1615024256,"elo":1586.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3846051827,"x_train":3.5186684216,"imp":10.5179792428,"elo":1615.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3052741537,"x_train":3.2956748919,"imp":21.3452090401,"elo":1234.8,"imputed":true,"imputed_pct":31.5789473684,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2712952691,"x_train":2916.211932309,"imp":21.1050043397,"elo":1264.2,"imputed":true,"imputed_pct":31.5789473684,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.5931496568,"x_train":2916.211932309,"imp":19.6295950754,"elo":1318.7,"imputed":true,"imputed_pct":31.5789473684,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9719997826,"x_train":403.917691534,"imp":32.4744091093,"elo":1016.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.2687972891,"x_train":38846.7008402763,"imp":26.0568333519,"elo":1113.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.4748027028,"x_train":38846.7008402763,"imp":26.2699222413,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0643226705,"x_train":1.1995725235,"imp":18.7766103518,"elo":1319.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1483357715,"x_train":5.7268864593,"imp":26.3504187059,"elo":1042.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1166104296,"x_train":2480.8031686863,"imp":21.5484213473,"elo":1196.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1753468186,"x_train":2480.8031686863,"imp":19.3559768486,"elo":1264.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209262941,"x_train":1.8759235268,"imp":20.8813545514,"elo":1199.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.1829513734,"x_train":682.6224605816,"imp":17.8195759499,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.9713359475,"x_train":682.6224605816,"imp":17.6065215295,"elo":1359.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":62.4915524975,"x_train":267.891665628,"imp":24.4055566951,"elo":1134.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":61.3096906599,"x_train":12550.0998837068,"imp":18.3357174571,"elo":1300.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":427.5904998868,"x_train":12550.0998837068,"imp":16.5890768836,"elo":1388.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.6769281514,"x_train":3.4779953241,"imp":30.6951398147,"elo":975.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1053014255,"x_train":1053.8003845303,"imp":20.0239827209,"elo":1256.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0671352109,"x_train":1053.8003845303,"imp":18.4717714757,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/tuning_trajectories_explorer.html
index 304bbad919dad87483c72b99ee8c7097855aedcb..2dfa7c612cff964461f48714bb49ea91574fd23c 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_classification/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":250.9922109033,"imp":19.6941900294,"elo":1330.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":4.8861388456,"imp":17.3507653956,"elo":1363.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":10.6750136791,"imp":16.8641766357,"elo":1380.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":25.2669903942,"imp":16.6115028744,"elo":1390.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":49.0638838885,"imp":16.4693487311,"elo":1397.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":133.206456981,"imp":16.4083750635,"elo":1402.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":264.5793957685,"imp":16.4272207116,"elo":1398.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":539.0252673695,"imp":16.3039335537,"elo":1396.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":810.5253697201,"imp":16.3166939099,"elo":1390.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1087.7269134101,"imp":16.2346962417,"elo":1395.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.9155810093,"imp":20.7165680113,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.4587928667,"imp":20.0603874145,"elo":1306.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":11.7223762673,"imp":19.6362149625,"elo":1326.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":24.3697423433,"imp":19.3788877323,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":61.9355959829,"imp":19.0415332537,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":124.7079952924,"imp":18.8531591174,"elo":1354.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":249.4958722159,"imp":18.7803121087,"elo":1355.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":373.7478601635,"imp":18.743666544,"elo":1356.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":500.3975293451,"imp":18.7727958545,"elo":1355.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":5.4796099733,"imp":21.044099446,"elo":1244.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":13.9490355599,"imp":20.5616937087,"elo":1264.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":38.1339214312,"imp":20.1883366972,"elo":1281.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":74.7850861491,"imp":20.0673987229,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":183.3974305409,"imp":19.9080389592,"elo":1297.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":366.1485755963,"imp":19.7474525784,"elo":1304.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":725.871285674,"imp":19.580579071,"elo":1312.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1084.12220491,"imp":19.5443956678,"elo":1315.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1449.1413756847,"imp":19.5039237042,"elo":1315.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.7477789606,"imp":29.5358260729,"elo":991.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1400651467,"imp":27.3537167562,"elo":1030.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.3396935938,"imp":24.0532143367,"elo":1100.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.4314608681,"imp":22.9857689489,"elo":1139.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":35.4438180004,"imp":22.2488583178,"elo":1168.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":70.3987810435,"imp":21.9885948974,"elo":1179.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":140.7929692058,"imp":21.7800936112,"elo":1190.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":211.1596497818,"imp":21.6556229605,"elo":1198.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":282.3556156822,"imp":21.8416133194,"elo":1197.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.1251204631,"imp":27.7036492778,"elo":1007.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.228658512,"imp":26.5962905781,"elo":1061.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":15.6396665341,"imp":24.5818525711,"elo":1132.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":31.5302468269,"imp":23.728822502,"elo":1162.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":76.3915347501,"imp":22.8054368602,"elo":1184.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":153.0430713803,"imp":22.4289582063,"elo":1199.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":308.6007823833,"imp":22.2493670824,"elo":1205.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":463.687419557,"imp":22.1714881727,"elo":1208.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":622.8560318368,"imp":22.2377412244,"elo":1214.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.2519280723,"imp":22.0740859134,"elo":1153.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":4.5917771836,"imp":20.0760955801,"elo":1249.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":9.9026612043,"imp":18.7237210856,"elo":1328.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":18.7813971613,"imp":18.0763418387,"elo":1362.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":48.1169020642,"imp":17.4980263496,"elo":1381.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":97.0735394653,"imp":17.3033751678,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":194.547566338,"imp":17.1929075924,"elo":1399.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":290.5602659885,"imp":17.1165386988,"elo":1401.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":391.2703614285,"imp":17.1311906392,"elo":1402.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":18.1964917538,"imp":16.2916492222,"elo":1403.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":118.4223166146,"imp":19.957263849,"elo":1323.1,"imputed":true,"imputed_pct":31.5789473684,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":13.3530116618,"imp":22.9952601478,"elo":1182.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":43.609446858,"imp":21.3213986232,"elo":1220.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":123.2677201414,"imp":18.7855395799,"elo":1317.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":240.5329414556,"imp":18.2333157971,"elo":1341.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":578.0322705513,"imp":17.6819035974,"elo":1368.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1205.8399034278,"imp":17.6946223505,"elo":1362.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2388.27785775,"imp":17.9394275128,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3575.4198584316,"imp":17.8432362289,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4853.4730504877,"imp":17.7262295326,"elo":1355.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":16.0704711344,"imp":26.2300441368,"elo":1079.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":20.4435114187,"imp":38.4431278401,"elo":939.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":50.7468311596,"imp":35.1420498515,"elo":995.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":146.088248762,"imp":31.4303757853,"elo":1064.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":175.2300293672,"imp":31.2195904793,"elo":1068.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3738939375,"imp":29.5043688368,"elo":998.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":2.1861266183,"imp":26.156918325,"elo":1063.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":7.0625783305,"imp":24.4292526257,"elo":1106.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":15.2157298212,"imp":23.6111235602,"elo":1127.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":39.5598749075,"imp":22.9769532811,"elo":1162.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":80.1299502447,"imp":22.8480462676,"elo":1169.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":161.6248068008,"imp":22.8533002163,"elo":1172.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":243.343699307,"imp":22.8754934947,"elo":1171.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":327.3662383914,"imp":22.8601519272,"elo":1172.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":10.7684699326,"imp":20.0916304434,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":24.8845759634,"imp":17.7546360003,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":65.188545506,"imp":16.2485999039,"elo":1430.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":135.7029218302,"imp":15.6336913346,"elo":1450.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":345.9238137878,"imp":15.0598393532,"elo":1476.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":694.3474135019,"imp":14.7878788049,"elo":1488.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1382.0545973419,"imp":14.7009422463,"elo":1490.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2071.1098936624,"imp":14.7729273816,"elo":1494.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2774.1715903227,"imp":14.8117502872,"elo":1496.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.6257151642,"imp":12.9485738533,"elo":1540.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.8594644787,"imp":12.9661880271,"elo":1539.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":46.8848802687,"imp":12.8755818833,"elo":1542.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":98.2945586224,"imp":12.9007224386,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":251.1485792262,"imp":12.7565954746,"elo":1562.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":507.8298692924,"imp":12.6238088111,"elo":1562.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1024.2615862152,"imp":12.041357481,"elo":1574.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1537.2603273066,"imp":11.6693540549,"elo":1583.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2063.6226141883,"imp":10.9566747432,"elo":1585.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.6228654067,"imp":21.5706920947,"elo":1257.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":47.8871994,"imp":21.4055424731,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":78.2444408904,"imp":20.2440694751,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":157.8326162312,"imp":19.1697186978,"elo":1309.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":299.7184949711,"imp":18.7441852966,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":746.8187102503,"imp":17.8989738569,"elo":1347.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1502.6287683848,"imp":17.3061287921,"elo":1364.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3013.6303119073,"imp":17.0388632039,"elo":1374.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":4537.799646108,"imp":16.8634697335,"elo":1377.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":6067.4246083224,"imp":16.7208156017,"elo":1383.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0990306801,"imp":17.1249222242,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":40.7899647401,"imp":5.5875134042,"elo":1766.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":1.0350786389,"imp":32.4418693765,"elo":985.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":9.0755702975,"imp":17.5071051034,"elo":1372.7,"imputed":true,"imputed_pct":5.2631578947,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":2.0391419632,"imp":12.4653888677,"elo":1575.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":7.2637875629,"imp":19.5772992171,"elo":1303.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":15.7913063985,"imp":19.1349596148,"elo":1318.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":48.8806618329,"imp":18.4715102008,"elo":1345.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":105.9612963967,"imp":17.8694747512,"elo":1375.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":268.8269097705,"imp":17.4048250163,"elo":1393.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":542.3321694006,"imp":17.095761429,"elo":1407.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1076.1837280763,"imp":16.7683531601,"elo":1416.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1633.7378956367,"imp":16.6461127566,"elo":1423.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2207.1695731088,"imp":16.5343936719,"elo":1425.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":5.3935958382,"imp":12.2011515198,"elo":1611.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":3.5186684216,"imp":10.5055868089,"elo":1629.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3.2956748919,"imp":21.3756301402,"elo":1242.1,"imputed":true,"imputed_pct":31.5789473684,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":22.8672371568,"imp":21.0436009702,"elo":1268.3,"imputed":true,"imputed_pct":31.5789473684,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":62.9137172928,"imp":20.5242498462,"elo":1300.4,"imputed":true,"imputed_pct":31.5789473684,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":140.2609154123,"imp":20.083197098,"elo":1316.8,"imputed":true,"imputed_pct":31.5789473684,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":347.4920082066,"imp":19.6221243623,"elo":1337.4,"imputed":true,"imputed_pct":31.5789473684,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":706.1972040222,"imp":19.4232924647,"elo":1348.5,"imputed":true,"imputed_pct":31.5789473684,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1438.110967546,"imp":19.4360405983,"elo":1346.9,"imputed":true,"imputed_pct":31.5789473684,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2168.8573846957,"imp":19.5453303925,"elo":1341.3,"imputed":true,"imputed_pct":31.5789473684,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2916.211932309,"imp":19.6732972304,"elo":1330.6,"imputed":true,"imputed_pct":31.5789473684,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":403.917691534,"imp":32.493477385,"elo":1025.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":655.6361879164,"imp":32.3446434367,"elo":1034.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1251.8865438921,"imp":31.9298236109,"elo":1040.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2072.8542359813,"imp":31.0529272589,"elo":1052.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5070.1005510261,"imp":29.3150593657,"elo":1072.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":10272.9118282161,"imp":27.7904789136,"elo":1089.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":19607.2944133693,"imp":27.1827696577,"elo":1098.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":29100.9871266142,"imp":26.7467459397,"elo":1109.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":38846.7008402763,"imp":26.3001453638,"elo":1123.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":1.1995725235,"imp":18.8168161208,"elo":1337.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":5.7268864593,"imp":26.3717660694,"elo":1049.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":16.2195091536,"imp":23.7366873395,"elo":1119.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":55.3202394018,"imp":21.6539094708,"elo":1177.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":125.651096343,"imp":20.8262781464,"elo":1218.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":306.1276418894,"imp":20.1696490504,"elo":1247.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":615.7158666145,"imp":19.8226381548,"elo":1261.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1199.8979174815,"imp":19.6679103673,"elo":1265.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1834.8903114269,"imp":19.4851144281,"elo":1275.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2480.8031686863,"imp":19.3876914292,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":1.8759235268,"imp":20.9187305948,"elo":1206.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":5.5595037251,"imp":19.1021793944,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":17.0856997287,"imp":18.5794202315,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":32.8489180471,"imp":18.2250099103,"elo":1349.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":86.2976754817,"imp":17.9438017356,"elo":1360.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":171.0781299242,"imp":17.8411705978,"elo":1368.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":341.1269358042,"imp":17.7472637508,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":508.913221887,"imp":17.6641603926,"elo":1380.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":682.6224605816,"imp":17.63697572,"elo":1380.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":267.891665628,"imp":24.4356350107,"elo":1145.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":693.7919545668,"imp":21.8748042649,"elo":1211.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2176.9330069286,"imp":18.9916330813,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4855.081405543,"imp":17.7431457596,"elo":1353.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":11992.0699103258,"imp":16.5946188983,"elo":1406.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12550.0998837068,"imp":16.6358496653,"elo":1406.1,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.4779953241,"imp":30.7180424778,"elo":986.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":10.0220449928,"imp":26.7407819438,"elo":1065.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":26.9153595064,"imp":21.9420219617,"elo":1195.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":52.7158440595,"imp":20.00142602,"elo":1271.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":131.9095207406,"imp":19.259715784,"elo":1300.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":263.979102644,"imp":18.8819267553,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":526.7568299014,"imp":18.6532171667,"elo":1313.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":787.1603694099,"imp":18.5534356886,"elo":1315.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1053.8003845303,"imp":18.4217031021,"elo":1317.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.0351934444,"x_train":250.9922109033,"imp":19.6941900294,"elo":1330.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.024809806,"x_train":4.8861388456,"imp":17.3507653956,"elo":1363.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0523082602,"x_train":10.6750136791,"imp":16.8641766357,"elo":1380.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1051923231,"x_train":25.2669903942,"imp":16.6115028744,"elo":1390.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1507927975,"x_train":49.0638838885,"imp":16.4693487311,"elo":1397.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.198237584,"x_train":133.206456981,"imp":16.4083750635,"elo":1402.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2413983074,"x_train":264.5793957685,"imp":16.4272207116,"elo":1398.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2948037083,"x_train":539.0252673695,"imp":16.3039335537,"elo":1396.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.317649589,"x_train":810.5253697201,"imp":16.3166939099,"elo":1390.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3481993779,"x_train":1087.7269134101,"imp":16.2346962417,"elo":1395.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0396842484,"x_train":1.9155810093,"imp":20.7165680113,"elo":1267.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0753273204,"x_train":4.4587928667,"imp":20.0603874145,"elo":1306.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1295300095,"x_train":11.7223762673,"imp":19.6362149625,"elo":1326.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1888398904,"x_train":24.3697423433,"imp":19.3788877323,"elo":1335.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2593943388,"x_train":61.9355959829,"imp":19.0415332537,"elo":1347.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3225450949,"x_train":124.7079952924,"imp":18.8531591174,"elo":1354.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3870003342,"x_train":249.4958722159,"imp":18.7803121087,"elo":1355.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4035085653,"x_train":373.7478601635,"imp":18.743666544,"elo":1356.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4161492489,"x_train":500.3975293451,"imp":18.7727958545,"elo":1355.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0162731597,"x_train":5.4796099733,"imp":21.044099446,"elo":1244.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0330339112,"x_train":13.9490355599,"imp":20.5616937087,"elo":1264.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0589358123,"x_train":38.1339214312,"imp":20.1883366972,"elo":1281.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0759361533,"x_train":74.7850861491,"imp":20.0673987229,"elo":1286.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1039353704,"x_train":183.3974305409,"imp":19.9080389592,"elo":1297.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1239174887,"x_train":366.1485755963,"imp":19.7474525784,"elo":1304.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1375888064,"x_train":725.871285674,"imp":19.580579071,"elo":1312.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.135225794,"x_train":1084.12220491,"imp":19.5443956678,"elo":1315.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1523084536,"x_train":1449.1413756847,"imp":19.5039237042,"elo":1315.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2675026243,"x_train":1.7477789606,"imp":29.5358260729,"elo":991.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2918606873,"x_train":3.1400651467,"imp":27.3537167562,"elo":1030.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3821908459,"x_train":7.3396935938,"imp":24.0532143367,"elo":1100.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4421545506,"x_train":14.4314608681,"imp":22.9857689489,"elo":1139.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5921825198,"x_train":35.4438180004,"imp":22.2488583178,"elo":1168.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.729487764,"x_train":70.3987810435,"imp":21.9885948974,"elo":1179.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8457914876,"x_train":140.7929692058,"imp":21.7800936112,"elo":1190.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8457827933,"x_train":211.1596497818,"imp":21.6556229605,"elo":1198.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8643331318,"x_train":282.3556156822,"imp":21.8416133194,"elo":1197.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3340852103,"x_train":3.1251204631,"imp":27.7036492778,"elo":1007.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6312937965,"x_train":6.228658512,"imp":26.5962905781,"elo":1061.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1801924809,"x_train":15.6396665341,"imp":24.5818525711,"elo":1132.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7736202367,"x_train":31.5302468269,"imp":23.728822502,"elo":1162.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.5336467582,"x_train":76.3915347501,"imp":22.8054368602,"elo":1184.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.419046747,"x_train":153.0430713803,"imp":22.4289582063,"elo":1199.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.1140070815,"x_train":308.6007823833,"imp":22.2493670824,"elo":1205.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.5060075639,"x_train":463.687419557,"imp":22.1714881727,"elo":1208.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.0469843044,"x_train":622.8560318368,"imp":22.2377412244,"elo":1214.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1585029211,"x_train":2.2519280723,"imp":22.0740859134,"elo":1153.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5387304774,"x_train":4.5917771836,"imp":20.0760955801,"elo":1249.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7747491802,"x_train":9.9026612043,"imp":18.7237210856,"elo":1328.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.9746279953,"x_train":18.7813971613,"imp":18.0763418387,"elo":1362.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.1744648066,"x_train":48.1169020642,"imp":17.4980263496,"elo":1381.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2950339303,"x_train":97.0735394653,"imp":17.3033751678,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4013389398,"x_train":194.547566338,"imp":17.1929075924,"elo":1399.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5863655126,"x_train":290.5602659885,"imp":17.1165386988,"elo":1401.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.2823687024,"x_train":391.2703614285,"imp":17.1311906392,"elo":1402.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":4.6915866654,"x_train":18.1964917538,"imp":16.2916492222,"elo":1403.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.6670530253,"x_train":118.4223166146,"imp":19.957263849,"elo":1323.1,"imputed":true,"imputed_pct":31.5789473684,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"x_train":13.3530116618,"imp":22.9952601478,"elo":1182.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9081956315,"x_train":43.609446858,"imp":21.3213986232,"elo":1220.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8094695931,"x_train":123.2677201414,"imp":18.7855395799,"elo":1317.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9001397479,"x_train":240.5329414556,"imp":18.2333157971,"elo":1341.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.3070047996,"x_train":578.0322705513,"imp":17.6819035974,"elo":1368.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5713626606,"x_train":1205.8399034278,"imp":17.6946223505,"elo":1362.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.0463659943,"x_train":2388.27785775,"imp":17.9394275128,"elo":1351.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4271738524,"x_train":3575.4198584316,"imp":17.8432362289,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"x_train":4853.4730504877,"imp":17.7262295326,"elo":1355.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.876262961,"x_train":16.0704711344,"imp":26.2300441368,"elo":1079.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0443776494,"x_train":20.4435114187,"imp":38.4431278401,"elo":939.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3417975297,"x_train":50.7468311596,"imp":35.1420498515,"elo":995.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5993854901,"x_train":146.088248762,"imp":31.4303757853,"elo":1064.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6542696803,"x_train":175.2300293672,"imp":31.2195904793,"elo":1068.6,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0376682658,"x_train":0.3738939375,"imp":29.5043688368,"elo":998.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1044595646,"x_train":2.1861266183,"imp":26.156918325,"elo":1063.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2235814001,"x_train":7.0625783305,"imp":24.4292526257,"elo":1106.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3083081492,"x_train":15.2157298212,"imp":23.6111235602,"elo":1127.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.471766492,"x_train":39.5598749075,"imp":22.9769532811,"elo":1162.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5721305279,"x_train":80.1299502447,"imp":22.8480462676,"elo":1169.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.659224899,"x_train":161.6248068008,"imp":22.8533002163,"elo":1172.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6901438372,"x_train":243.343699307,"imp":22.8754934947,"elo":1171.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6730171196,"x_train":327.3662383914,"imp":22.8601519272,"elo":1172.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7864396959,"x_train":10.7684699326,"imp":20.0916304434,"elo":1260.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.7130532123,"x_train":24.8845759634,"imp":17.7546360003,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.0407156134,"x_train":65.188545506,"imp":16.2485999039,"elo":1430.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.5693353243,"x_train":135.7029218302,"imp":15.6336913346,"elo":1450.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.5979957571,"x_train":345.9238137878,"imp":15.0598393532,"elo":1476.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.5139570657,"x_train":694.3474135019,"imp":14.7878788049,"elo":1488.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.7224848724,"x_train":1382.0545973419,"imp":14.7009422463,"elo":1490.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.7145191322,"x_train":2071.1098936624,"imp":14.7729273816,"elo":1494.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.3626514212,"x_train":2774.1715903227,"imp":14.8117502872,"elo":1496.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7713317233,"x_train":5.6257151642,"imp":12.9485738533,"elo":1540.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3657976323,"x_train":16.8594644787,"imp":12.9661880271,"elo":1539.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8885842429,"x_train":46.8848802687,"imp":12.8755818833,"elo":1542.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.9894747323,"x_train":98.2945586224,"imp":12.9007224386,"elo":1551.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.8519951234,"x_train":251.1485792262,"imp":12.7565954746,"elo":1562.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6072809051,"x_train":507.8298692924,"imp":12.6238088111,"elo":1562.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1136602495,"x_train":1024.2615862152,"imp":12.041357481,"elo":1574.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.10648955,"x_train":1537.2603273066,"imp":11.6693540549,"elo":1583.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.9647036297,"x_train":2063.6226141883,"imp":10.9566747432,"elo":1585.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.9837292889,"x_train":13.6228654067,"imp":21.5706920947,"elo":1257.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4439438782,"x_train":47.8871994,"imp":21.4055424731,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":57.0488035322,"x_train":78.2444408904,"imp":20.2440694751,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":89.3976769749,"x_train":157.8326162312,"imp":19.1697186978,"elo":1309.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":120.7151402618,"x_train":299.7184949711,"imp":18.7441852966,"elo":1326.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":191.9678402321,"x_train":746.8187102503,"imp":17.8989738569,"elo":1347.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":252.9847402475,"x_train":1502.6287683848,"imp":17.3061287921,"elo":1364.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":320.0227255786,"x_train":3013.6303119073,"imp":17.0388632039,"elo":1374.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":359.8655501595,"x_train":4537.799646108,"imp":16.8634697335,"elo":1377.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":405.618626345,"x_train":6067.4246083224,"imp":16.7208156017,"elo":1383.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1861515215,"x_train":2.0990306801,"imp":17.1249222242,"elo":1399.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.0561792583,"x_train":40.7899647401,"imp":5.5875134042,"elo":1766.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.1916685529,"x_train":1.0350786389,"imp":32.4418693765,"elo":985.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":2.0066462138,"x_train":9.0755702975,"imp":17.5071051034,"elo":1372.7,"imputed":true,"imputed_pct":5.2631578947,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"x_train":2.0391419632,"imp":12.4653888677,"elo":1575.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2496062173,"x_train":7.2637875629,"imp":19.5772992171,"elo":1303.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4588827195,"x_train":15.7913063985,"imp":19.1349596148,"elo":1318.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.9324927533,"x_train":48.8806618329,"imp":18.4715102008,"elo":1345.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.0449796984,"x_train":105.9612963967,"imp":17.8694747512,"elo":1375.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.418274975,"x_train":268.8269097705,"imp":17.4048250163,"elo":1393.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.7029553903,"x_train":542.3321694006,"imp":17.095761429,"elo":1407.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.8501005988,"x_train":1076.1837280763,"imp":16.7683531601,"elo":1416.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.8904873538,"x_train":1633.7378956367,"imp":16.6461127566,"elo":1423.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":1.8937013268,"x_train":2207.1695731088,"imp":16.5343936719,"elo":1425.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"x_train":5.3935958382,"imp":12.2011515198,"elo":1611.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3846051827,"x_train":3.5186684216,"imp":10.5055868089,"elo":1629.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3052741537,"x_train":3.2956748919,"imp":21.3756301402,"elo":1242.1,"imputed":true,"imputed_pct":31.5789473684,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":2.8468480497,"x_train":22.8672371568,"imp":21.0436009702,"elo":1268.3,"imputed":true,"imputed_pct":31.5789473684,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.3678602786,"x_train":62.9137172928,"imp":20.5242498462,"elo":1300.4,"imputed":true,"imputed_pct":31.5789473684,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":7.4693303184,"x_train":140.2609154123,"imp":20.083197098,"elo":1316.8,"imputed":true,"imputed_pct":31.5789473684,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.1384233264,"x_train":347.4920082066,"imp":19.6221243623,"elo":1337.4,"imputed":true,"imputed_pct":31.5789473684,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":14.0187998736,"x_train":706.1972040222,"imp":19.4232924647,"elo":1348.5,"imputed":true,"imputed_pct":31.5789473684,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.5186392666,"x_train":1438.110967546,"imp":19.4360405983,"elo":1346.9,"imputed":true,"imputed_pct":31.5789473684,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.0115372266,"x_train":2168.8573846957,"imp":19.5453303925,"elo":1341.3,"imputed":true,"imputed_pct":31.5789473684,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.098608772,"x_train":2916.211932309,"imp":19.6732972304,"elo":1330.6,"imputed":true,"imputed_pct":31.5789473684,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.9719997826,"x_train":403.917691534,"imp":32.493477385,"elo":1025.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.9132537435,"x_train":655.6361879164,"imp":32.3446434367,"elo":1034.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.44798274,"x_train":1251.8865438921,"imp":31.9298236109,"elo":1040.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.6829725644,"x_train":2072.8542359813,"imp":31.0529272589,"elo":1052.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.9995362697,"x_train":5070.1005510261,"imp":29.3150593657,"elo":1072.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.0786013215,"x_train":10272.9118282161,"imp":27.7904789136,"elo":1089.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.8375490304,"x_train":19607.2944133693,"imp":27.1827696577,"elo":1098.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.5577174656,"x_train":29100.9871266142,"imp":26.7467459397,"elo":1109.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.4748027028,"x_train":38846.7008402763,"imp":26.3001453638,"elo":1123.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0643226705,"x_train":1.1995725235,"imp":18.8168161208,"elo":1337.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1483357715,"x_train":5.7268864593,"imp":26.3717660694,"elo":1049.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3175868529,"x_train":16.2195091536,"imp":23.7366873395,"elo":1119.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6704028705,"x_train":55.3202394018,"imp":21.6539094708,"elo":1177.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8151812596,"x_train":125.651096343,"imp":20.8262781464,"elo":1218.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.208494529,"x_train":306.1276418894,"imp":20.1696490504,"elo":1247.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5100756369,"x_train":615.7158666145,"imp":19.8226381548,"elo":1261.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.9011606861,"x_train":1199.8979174815,"imp":19.6679103673,"elo":1265.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0966186606,"x_train":1834.8903114269,"imp":19.4851144281,"elo":1275.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1753468186,"x_train":2480.8031686863,"imp":19.3876914292,"elo":1277.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1209262941,"x_train":1.8759235268,"imp":20.9187305948,"elo":1206.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4702992151,"x_train":5.5595037251,"imp":19.1021793944,"elo":1298.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8389111922,"x_train":17.0856997287,"imp":18.5794202315,"elo":1330.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8064791193,"x_train":32.8489180471,"imp":18.2250099103,"elo":1349.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8473214228,"x_train":86.2976754817,"imp":17.9438017356,"elo":1360.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0012724874,"x_train":171.0781299242,"imp":17.8411705978,"elo":1368.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9989995929,"x_train":341.1269358042,"imp":17.7472637508,"elo":1375.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2150238563,"x_train":508.913221887,"imp":17.6641603926,"elo":1380.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9713359475,"x_train":682.6224605816,"imp":17.63697572,"elo":1380.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":62.4915524975,"x_train":267.891665628,"imp":24.4356350107,"elo":1145.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":128.140815719,"x_train":693.7919545668,"imp":21.8748042649,"elo":1211.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":235.7496154858,"x_train":2176.9330069286,"imp":18.9916330813,"elo":1302.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":355.4432898564,"x_train":4855.081405543,"imp":17.7431457596,"elo":1353.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":432.5178282259,"x_train":11992.0699103258,"imp":16.5946188983,"elo":1406.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":427.5904998868,"x_train":12550.0998837068,"imp":16.6358496653,"elo":1406.1,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.6769281514,"x_train":3.4779953241,"imp":30.7180424778,"elo":986.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8523326353,"x_train":10.0220449928,"imp":26.7407819438,"elo":1065.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.1689874481,"x_train":26.9153595064,"imp":21.9420219617,"elo":1195.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.525739026,"x_train":52.7158440595,"imp":20.00142602,"elo":1271.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.8428620046,"x_train":131.9095207406,"imp":19.259715784,"elo":1300.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.1886115821,"x_train":263.979102644,"imp":18.8819267553,"elo":1312.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.033782756,"x_train":526.7568299014,"imp":18.6532171667,"elo":1313.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.9700745263,"x_train":787.1603694099,"imp":18.5534356886,"elo":1315.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9325738702,"x_train":1053.8003845303,"imp":18.4217031021,"elo":1317.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..b514984fd324db01592afbad1c04ca60b638e19f
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_classification/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned)", "RealTabPFN-2.5 (default)", "AutoGluon 1.4 (best, 4h)", "RealMLP (tuned + ensembled)", "RealMLP (tuned)", "TabM (tuned + ensembled)", "LimiX (default)", "iLTM (tuned + ensembled)", "CatBoost (tuned)", "TabDPT-Turbo (default)", "LightGBM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "TabM (tuned)", "TabDPT (tuned + ensembled)", "XGBoost (tuned + ensembled)", "TabICL (default)", "ModernNCA (tuned)", "CatBoost (default)", "ModernNCA (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "LightGBM (tuned)", "XGBoost (tuned)", "TabDPT (tuned)", "TabSwift (default)", "TabPFNv2 (tuned + ensembled)", "BetaTabPFN (default)", "ChimeraBoost (tuned)", "Mitra (default)", "xRFM (tuned + ensembled)", "iLTM (tuned)", "EBM (tuned + ensembled)", "TabM (default)", "TabPFNv2 (tuned)", "TorchMLP (tuned + ensembled)", "xRFM (tuned)", "RealMLP (default)", "ChimeraBoost (default)", "SAP-RPT-OSS (default)", "EBM (tuned)", "EBM (default)", "TabPFNv2 (default)", "TabDPT (default)", "FastaiMLP (tuned + ensembled)", "XGBoost (default)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "ModernNCA (default)", "RandomForest (tuned + ensembled)", "ExtraTrees (tuned)", "LightGBM (default)", "iLTM (default)", "FastaiMLP (tuned)", "RandomForest (tuned)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "OrionMSP (default)", "PerpetualBooster (tuned + ensembled)", "TorchMLP (default)", "TabSTAR (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "RandomForest (default)", "KNN (tuned + ensembled)", "ExtraTrees (default)", "xRFM (default)", "TabFlex (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (default)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.7368420958518982, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.8947368264198303, 0.8421052694320679, 0.7631579041481018, 0.8157894611358643, 0.8421052694320679, 0.9210526347160339, 0.8684210777282715, 0.8157894611358643, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.8947368264198303, 0.9473684430122375, 0.8421052694320679, 0.9473684430122375, 0.8684210777282715, 0.9736841917037964, 0.9473684430122375, 0.8684210777282715, 1.0, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.8947368264198303, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9210526347160339, 0.8947368264198303, 0.9736841917037964, 1.0, 0.9736841917037964, 0.9473684430122375, 1.0, 1.0, 0.9736841917037964, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 1.0, 0.9736841917037964, 0.9736841917037964, 1.0, 1.0, 0.9736841917037964, 1.0], [0.2631579041481018, NaN, 0.4736842215061188, 0.6052631735801697, 0.6052631735801697, 0.5263158082962036, 0.6842105388641357, 0.5789473652839661, 0.6842105388641357, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7368420958518982, 0.8684210777282715, 0.8421052694320679, 0.6842105388641357, 0.8947368264198303, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.9210526347160339, 0.8157894611358643, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.8157894611358643, 0.7894737124443054, 0.7631579041481018, 0.8421052694320679, 0.9210526347160339, 0.7894737124443054, 0.9210526347160339, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.8684210777282715, 0.8684210777282715, 0.9473684430122375, 0.8157894611358643, 0.8684210777282715, 0.8947368264198303, 0.7894737124443054, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.9210526347160339, 0.9473684430122375, 0.8684210777282715, 0.8684210777282715, 1.0, 0.9210526347160339, 0.9473684430122375, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 1.0, 1.0], [0.18421052396297455, 0.5263158082962036, NaN, 0.5526315569877625, 0.5657894611358643, 0.6052631735801697, 0.6578947305679321, 0.5789473652839661, 0.5526315569877625, 0.7105262875556946, 0.7368420958518982, 0.7894737124443054, 0.7105262875556946, 0.7894737124443054, 0.7368420958518982, 0.8157894611358643, 0.8684210777282715, 0.7894737124443054, 0.8684210777282715, 0.7894737124443054, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.8684210777282715, 0.9473684430122375, 0.8157894611358643, 0.8947368264198303, 0.8684210777282715, 0.7894737124443054, 0.9210526347160339, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8947368264198303, 0.8157894611358643, 0.8421052694320679, 0.9210526347160339, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.7894737124443054, 0.8947368264198303, 0.8947368264198303, 0.7894737124443054, 0.8947368264198303, 0.8684210777282715, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.8684210777282715, 0.8684210777282715, 1.0, 0.9736841917037964, 0.9473684430122375, 0.8684210777282715, 1.0, 1.0, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0, 0.9736841917037964, 1.0], [0.15789473056793213, 0.3947368562221527, 0.44736841320991516, NaN, 0.4736842215061188, 0.5, 0.5526315569877625, 0.6578947305679321, 0.5263158082962036, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.8157894611358643, 0.7631579041481018, 0.7894737124443054, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.7368420958518982, 0.7631579041481018, 0.8684210777282715, 0.8157894611358643, 0.8684210777282715, 0.8947368264198303, 0.8157894611358643, 0.9736841917037964, 0.8421052694320679, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8157894611358643, 0.9210526347160339, 0.8684210777282715, 0.7631579041481018, 0.8157894611358643, 0.8947368264198303, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.8947368264198303, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.8421052694320679, 1.0, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 1.0], [0.1315789520740509, 0.3947368562221527, 0.43421053886413574, 0.5263158082962036, NaN, 0.5263158082962036, 0.7105262875556946, 0.6315789222717285, 0.5526315569877625, 0.6842105388641357, 0.7105262875556946, 0.6842105388641357, 0.6842105388641357, 0.7368420958518982, 0.8157894611358643, 0.6578947305679321, 0.8157894611358643, 0.8684210777282715, 0.7631579041481018, 0.6842105388641357, 0.8421052694320679, 0.7368420958518982, 0.8421052694320679, 0.8684210777282715, 0.7631579041481018, 0.7894737124443054, 0.8947368264198303, 0.8421052694320679, 0.7631579041481018, 0.8157894611358643, 0.8947368264198303, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.9473684430122375, 0.8157894611358643, 0.8684210777282715, 0.8421052694320679, 0.8157894611358643, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.8157894611358643, 0.8421052694320679, 0.9473684430122375, 0.8684210777282715, 0.9473684430122375, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 0.8421052694320679, 0.8421052694320679, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 1.0], [0.1315789520740509, 0.4736842215061188, 0.3947368562221527, 0.5, 0.4736842215061188, NaN, 0.6052631735801697, 0.5789473652839661, 0.6315789222717285, 0.6578947305679321, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.7894737124443054, 0.8157894611358643, 0.7368420958518982, 0.7105262875556946, 0.8157894611358643, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.7631579041481018, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.8947368264198303, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.7631579041481018, 0.7631579041481018, 0.8421052694320679, 0.8157894611358643, 0.8684210777282715, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.9473684430122375, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 0.8421052694320679, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.8947368264198303, 0.9736841917037964, 1.0, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0, 0.9736841917037964, 1.0], [0.10526315867900848, 0.31578946113586426, 0.34210526943206787, 0.44736841320991516, 0.28947368264198303, 0.3947368562221527, NaN, 0.5657894611358643, 0.5, 0.6052631735801697, 0.7105262875556946, 0.7368420958518982, 0.5789473652839661, 0.6578947305679321, 0.7368420958518982, 0.6578947305679321, 0.7368420958518982, 0.7631579041481018, 0.7894737124443054, 0.6578947305679321, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.8684210777282715, 0.7368420958518982, 0.6842105388641357, 0.8684210777282715, 0.7894737124443054, 0.7631579041481018, 0.7368420958518982, 0.7236841917037964, 0.8421052694320679, 0.7894737124443054, 0.8684210777282715, 0.7631579041481018, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.9473684430122375, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.9736841917037964, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.9473684430122375, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 1.0, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 1.0], [0.15789473056793213, 0.42105263471603394, 0.42105263471603394, 0.34210526943206787, 0.3684210479259491, 0.42105263471603394, 0.43421053886413574, NaN, 0.5526315569877625, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.7105262875556946, 0.5789473652839661, 0.6842105388641357, 0.6578947305679321, 0.7631579041481018, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.7368420958518982, 0.6842105388641357, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.7105262875556946, 0.7894737124443054, 0.7894737124443054, 0.7368420958518982, 0.8421052694320679, 0.6578947305679321, 0.8421052694320679, 0.7105262875556946, 0.8157894611358643, 0.8157894611358643, 0.8815789222717285, 0.7894737124443054, 0.8157894611358643, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.9210526347160339, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.8421052694320679, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 0.9736841917037964, 0.8947368264198303, 0.9736841917037964, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 1.0, 0.9473684430122375, 0.9736841917037964, 0.8947368264198303, 1.0, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 1.0, 0.9473684430122375, 1.0, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0], [0.2368421107530594, 0.31578946113586426, 0.44736841320991516, 0.4736842215061188, 0.44736841320991516, 0.3684210479259491, 0.5, 0.44736841320991516, NaN, 0.5789473652839661, 0.7105262875556946, 0.6578947305679321, 0.6052631735801697, 0.8157894611358643, 0.7631579041481018, 0.6578947305679321, 0.7894737124443054, 0.7894737124443054, 0.7368420958518982, 0.6315789222717285, 0.8157894611358643, 0.6842105388641357, 0.6842105388641357, 0.7631579041481018, 0.6578947305679321, 0.7894737124443054, 0.8421052694320679, 0.7894737124443054, 0.7105262875556946, 0.7105262875556946, 0.5789473652839661, 0.7631579041481018, 0.8157894611358643, 0.6578947305679321, 0.7894737124443054, 0.7894737124443054, 0.7631579041481018, 0.7894737124443054, 0.6842105388641357, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.8421052694320679, 0.7631579041481018, 0.8947368264198303, 0.8684210777282715, 0.7631579041481018, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.9210526347160339, 0.8421052694320679, 0.9210526347160339, 0.8684210777282715, 0.8421052694320679, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.7894737124443054, 0.8157894611358643, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.8684210777282715, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9210526347160339, 0.9736841917037964, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 1.0, 1.0, 1.0], [0.18421052396297455, 0.18421052396297455, 0.28947368264198303, 0.28947368264198303, 0.31578946113586426, 0.34210526943206787, 0.3947368562221527, 0.3947368562221527, 0.42105263471603394, NaN, 0.6578947305679321, 0.6052631735801697, 0.5, 0.7105262875556946, 0.6842105388641357, 0.5789473652839661, 0.7105262875556946, 0.7631579041481018, 0.6842105388641357, 0.5789473652839661, 0.7105262875556946, 0.6052631735801697, 0.6578947305679321, 0.7631579041481018, 0.6842105388641357, 0.6578947305679321, 0.7631579041481018, 0.7631579041481018, 0.6315789222717285, 0.6315789222717285, 0.5789473652839661, 0.7105262875556946, 0.6842105388641357, 0.6315789222717285, 0.7105262875556946, 0.7894737124443054, 0.7894737124443054, 0.7631579041481018, 0.7105262875556946, 0.8157894611358643, 0.7631579041481018, 0.8947368264198303, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.6842105388641357, 0.7894737124443054, 0.8947368264198303, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.8421052694320679, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.8947368264198303, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 0.9736841917037964, 1.0], [0.15789473056793213, 0.18421052396297455, 0.2631579041481018, 0.2631579041481018, 0.28947368264198303, 0.3684210479259491, 0.28947368264198303, 0.31578946113586426, 0.28947368264198303, 0.34210526943206787, NaN, 0.4736842215061188, 0.43421053886413574, 0.5, 0.5, 0.4736842215061188, 0.6052631735801697, 0.5263158082962036, 0.5789473652839661, 0.4736842215061188, 0.6315789222717285, 0.5789473652839661, 0.6578947305679321, 0.6315789222717285, 0.6052631735801697, 0.5526315569877625, 0.6578947305679321, 0.6842105388641357, 0.5789473652839661, 0.5789473652839661, 0.5526315569877625, 0.6842105388641357, 0.5789473652839661, 0.5263158082962036, 0.6578947305679321, 0.7105262875556946, 0.7105262875556946, 0.7105262875556946, 0.6315789222717285, 0.7894737124443054, 0.6842105388641357, 0.8157894611358643, 0.6578947305679321, 0.6578947305679321, 0.7368420958518982, 0.7368420958518982, 0.6842105388641357, 0.6842105388641357, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.7631579041481018, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.8157894611358643, 0.8421052694320679, 0.8289473652839661, 0.8947368264198303, 0.9473684430122375, 0.8421052694320679, 0.8947368264198303, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9473684430122375, 0.9605262875556946, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 1.0], [0.07894736528396606, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.31578946113586426, 0.31578946113586426, 0.2631579041481018, 0.31578946113586426, 0.34210526943206787, 0.3947368562221527, 0.5263158082962036, NaN, 0.42105263471603394, 0.5263158082962036, 0.4736842215061188, 0.5789473652839661, 0.6315789222717285, 0.5263158082962036, 0.5526315569877625, 0.5789473652839661, 0.5526315569877625, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.6052631735801697, 0.5789473652839661, 0.6052631735801697, 0.6842105388641357, 0.6052631735801697, 0.6315789222717285, 0.5, 0.6842105388641357, 0.6842105388641357, 0.4736842215061188, 0.5789473652839661, 0.6315789222717285, 0.7894737124443054, 0.8421052694320679, 0.5263158082962036, 0.7105262875556946, 0.5789473652839661, 0.8157894611358643, 0.7105262875556946, 0.7105262875556946, 0.7368420958518982, 0.7894737124443054, 0.6315789222717285, 0.7631579041481018, 0.7894737124443054, 0.7368420958518982, 0.7894737124443054, 0.7105262875556946, 0.8421052694320679, 0.7894737124443054, 0.7105262875556946, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.9736841917037964, 0.8684210777282715, 0.9210526347160339, 0.9210526347160339, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 1.0, 1.0], [0.1315789520740509, 0.2631579041481018, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.31578946113586426, 0.42105263471603394, 0.28947368264198303, 0.3947368562221527, 0.5, 0.5657894611358643, 0.5789473652839661, NaN, 0.5657894611358643, 0.5263158082962036, 0.44736841320991516, 0.6315789222717285, 0.5789473652839661, 0.6052631735801697, 0.44736841320991516, 0.6578947305679321, 0.5263158082962036, 0.6052631735801697, 0.5789473652839661, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.6842105388641357, 0.5526315569877625, 0.5263158082962036, 0.6578947305679321, 0.6052631735801697, 0.6578947305679321, 0.6052631735801697, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.6578947305679321, 0.7105262875556946, 0.6842105388641357, 0.7368420958518982, 0.7368420958518982, 0.6842105388641357, 0.6578947305679321, 0.7368420958518982, 0.7368420958518982, 0.6578947305679321, 0.6578947305679321, 0.6578947305679321, 0.7368420958518982, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.75, 0.7368420958518982, 0.8289473652839661, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8947368264198303, 0.9210526347160339, 0.8684210777282715, 0.9473684430122375, 0.9210526347160339, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.9736841917037964], [0.18421052396297455, 0.1315789520740509, 0.21052631735801697, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.34210526943206787, 0.42105263471603394, 0.18421052396297455, 0.28947368264198303, 0.5, 0.4736842215061188, 0.43421053886413574, NaN, 0.5, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, 0.5, 0.42105263471603394, 0.5263158082962036, 0.46052631735801697, 0.6052631735801697, 0.5, 0.5263158082962036, 0.6578947305679321, 0.5526315569877625, 0.6315789222717285, 0.5, 0.5789473652839661, 0.5263158082962036, 0.5, 0.6578947305679321, 0.5, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6052631735801697, 0.6315789222717285, 0.7368420958518982, 0.7368420958518982, 0.7105262875556946, 0.7368420958518982, 0.7105262875556946, 0.6578947305679321, 0.7368420958518982, 0.7368420958518982, 0.6842105388641357, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.7894737124443054, 0.8421052694320679, 0.8684210777282715, 0.8157894611358643, 0.8684210777282715, 0.7631579041481018, 0.7631579041481018, 0.8421052694320679, 0.8947368264198303, 0.9473684430122375, 0.8421052694320679, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 0.9473684430122375, 1.0], [0.10526315867900848, 0.15789473056793213, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.2631579041481018, 0.2631579041481018, 0.31578946113586426, 0.2368421107530594, 0.31578946113586426, 0.5, 0.5263158082962036, 0.4736842215061188, 0.5, NaN, 0.5263158082962036, 0.44736841320991516, 0.44736841320991516, 0.5526315569877625, 0.5263158082962036, 0.5789473652839661, 0.5263158082962036, 0.5263158082962036, 0.6578947305679321, 0.5789473652839661, 0.5263158082962036, 0.5263158082962036, 0.6578947305679321, 0.5526315569877625, 0.5789473652839661, 0.5, 0.5789473652839661, 0.6052631735801697, 0.5526315569877625, 0.6315789222717285, 0.7368420958518982, 0.6578947305679321, 0.6315789222717285, 0.6052631735801697, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.6315789222717285, 0.7105262875556946, 0.7368420958518982, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.7631579041481018, 0.8157894611358643, 0.7368420958518982, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8684210777282715, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.9210526347160339, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.8684210777282715, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0], [0.07894736528396606, 0.31578946113586426, 0.18421052396297455, 0.21052631735801697, 0.34210526943206787, 0.21052631735801697, 0.34210526943206787, 0.34210526943206787, 0.34210526943206787, 0.42105263471603394, 0.5263158082962036, 0.42105263471603394, 0.5526315569877625, 0.5526315569877625, 0.4736842215061188, NaN, 0.5, 0.5, 0.44736841320991516, 0.6315789222717285, 0.5789473652839661, 0.5, 0.5, 0.5789473652839661, 0.5, 0.6052631735801697, 0.5, 0.6052631735801697, 0.6842105388641357, 0.6710526347160339, 0.6315789222717285, 0.5263158082962036, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.6842105388641357, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.7105262875556946, 0.6578947305679321, 0.8157894611358643, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.7631579041481018, 0.6842105388641357, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.7894737124443054, 0.8684210777282715, 0.8157894611358643, 0.7631579041481018, 0.7631579041481018, 0.9210526347160339, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375, 0.9210526347160339, 0.9210526347160339, 0.9473684430122375, 0.8947368264198303, 0.9473684430122375, 0.8421052694320679, 0.8421052694320679, 0.9473684430122375, 0.8947368264198303, 0.9736841917037964, 1.0], [0.07894736528396606, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.21052631735801697, 0.28947368264198303, 0.3947368562221527, 0.3684210479259491, 0.3684210479259491, 0.4736842215061188, 0.5526315569877625, 0.5, NaN, 0.5789473652839661, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, 0.5263158082962036, 0.5, 0.6578947305679321, 0.5263158082962036, 0.5789473652839661, 0.7105262875556946, 0.6315789222717285, 0.5526315569877625, 0.5789473652839661, 0.4736842215061188, 0.5526315569877625, 0.6578947305679321, 0.4736842215061188, 0.6315789222717285, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.5526315569877625, 0.7368420958518982, 0.7105262875556946, 0.6842105388641357, 0.7368420958518982, 0.6578947305679321, 0.7894737124443054, 0.8157894611358643, 0.6578947305679321, 0.7105262875556946, 0.7631579041481018, 0.8684210777282715, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.9473684430122375, 0.8947368264198303, 0.8157894611358643, 0.9210526347160339, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964, 0.9210526347160339, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.9736841917037964, 0.9736841917037964, 0.8947368264198303, 0.9473684430122375, 1.0, 1.0], [0.07894736528396606, 0.10526315867900848, 0.21052631735801697, 0.18421052396297455, 0.1315789520740509, 0.2631579041481018, 0.2368421107530594, 0.31578946113586426, 0.21052631735801697, 0.2368421107530594, 0.4736842215061188, 0.4736842215061188, 0.42105263471603394, 0.5526315569877625, 0.5526315569877625, 0.5, 0.42105263471603394, NaN, 0.5, 0.5, 0.5263158082962036, 0.5789473652839661, 0.5131579041481018, 0.6315789222717285, 0.5789473652839661, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.5526315569877625, 0.5526315569877625, 0.5, 0.5263158082962036, 0.6052631735801697, 0.5263158082962036, 0.6315789222717285, 0.7368420958518982, 0.6052631735801697, 0.5526315569877625, 0.6315789222717285, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.8157894611358643, 0.7368420958518982, 0.6842105388641357, 0.7631579041481018, 0.6052631735801697, 0.6842105388641357, 0.7368420958518982, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7631579041481018, 0.8421052694320679, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.9210526347160339, 0.8157894611358643, 0.8947368264198303, 0.8947368264198303, 0.9473684430122375, 0.8684210777282715, 0.8684210777282715, 0.8947368264198303, 0.8421052694320679, 0.9736841917037964, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 1.0], [0.05263157933950424, 0.21052631735801697, 0.1315789520740509, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.21052631735801697, 0.2631579041481018, 0.2631579041481018, 0.31578946113586426, 0.42105263471603394, 0.44736841320991516, 0.3947368562221527, 0.5, 0.44736841320991516, 0.5526315569877625, 0.5263158082962036, 0.5, NaN, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.6052631735801697, 0.5789473652839661, 0.5526315569877625, 0.6447368264198303, 0.5263158082962036, 0.6052631735801697, 0.6052631735801697, 0.6052631735801697, 0.5, 0.6315789222717285, 0.6315789222717285, 0.44736841320991516, 0.5263158082962036, 0.6052631735801697, 0.6578947305679321, 0.6842105388641357, 0.5263158082962036, 0.6315789222717285, 0.5789473652839661, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.7105262875556946, 0.6842105388641357, 0.5789473652839661, 0.7368420958518982, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.6842105388641357, 0.8421052694320679, 0.7368420958518982, 0.6842105388641357, 0.7894737124443054, 0.7105262875556946, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.9736841917037964, 0.8421052694320679, 0.9210526347160339, 0.8684210777282715, 0.8684210777282715, 0.8684210777282715, 0.8157894611358643, 0.8421052694320679, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 0.9473684430122375, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964], [0.07894736528396606, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.31578946113586426, 0.18421052396297455, 0.34210526943206787, 0.2368421107530594, 0.3684210479259491, 0.42105263471603394, 0.5263158082962036, 0.42105263471603394, 0.5526315569877625, 0.5789473652839661, 0.4736842215061188, 0.3684210479259491, 0.5263158082962036, 0.5, 0.44736841320991516, NaN, 0.5263158082962036, 0.44736841320991516, 0.4736842215061188, 0.5263158082962036, 0.4736842215061188, 0.5263158082962036, 0.5263158082962036, 0.5263158082962036, 0.7105262875556946, 0.5263158082962036, 0.6052631735801697, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.5789473652839661, 0.6578947305679321, 0.5789473652839661, 0.5789473652839661, 0.6842105388641357, 0.6578947305679321, 0.6578947305679321, 0.6315789222717285, 0.5263158082962036, 0.6578947305679321, 0.6578947305679321, 0.7105262875556946, 0.6315789222717285, 0.7631579041481018, 0.6578947305679321, 0.6842105388641357, 0.6842105388641357, 0.7105262875556946, 0.8157894611358643, 0.7105262875556946, 0.8157894611358643, 0.7631579041481018, 0.7368420958518982, 0.7894737124443054, 0.7368420958518982, 0.7894737124443054, 0.7631579041481018, 0.9473684430122375, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375, 0.9473684430122375, 0.8947368264198303, 0.9736841917037964, 0.9473684430122375, 0.9210526347160339, 0.8684210777282715, 0.8684210777282715, 0.8947368264198303, 0.8947368264198303, 0.9473684430122375, 1.0], [0.10526315867900848, 0.07894736528396606, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.2368421107530594, 0.2368421107530594, 0.2631579041481018, 0.18421052396297455, 0.28947368264198303, 0.3684210479259491, 0.44736841320991516, 0.34210526943206787, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, 0.5263158082962036, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, NaN, 0.5263158082962036, 0.4736842215061188, 0.5526315569877625, 0.5, 0.5526315569877625, 0.5789473652839661, 0.6842105388641357, 0.5, 0.5263158082962036, 0.3947368562221527, 0.4736842215061188, 0.6315789222717285, 0.5, 0.5526315569877625, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.5263158082962036, 0.6842105388641357, 0.6842105388641357, 0.7105262875556946, 0.7105262875556946, 0.6315789222717285, 0.6842105388641357, 0.7105262875556946, 0.6315789222717285, 0.6842105388641357, 0.7105262875556946, 0.8684210777282715, 0.6973684430122375, 0.8157894611358643, 0.7894737124443054, 0.8421052694320679, 0.7894737124443054, 0.9736841917037964, 0.8421052694320679, 0.7631579041481018, 0.8947368264198303, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.7894737124443054, 0.8947368264198303, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375, 0.8684210777282715, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.9473684430122375, 0.8684210777282715, 0.9473684430122375, 0.9210526347160339, 1.0], [0.07894736528396606, 0.18421052396297455, 0.10526315867900848, 0.1315789520740509, 0.2631579041481018, 0.21052631735801697, 0.21052631735801697, 0.31578946113586426, 0.31578946113586426, 0.3947368562221527, 0.42105263471603394, 0.3947368562221527, 0.4736842215061188, 0.5394737124443054, 0.4736842215061188, 0.5, 0.4736842215061188, 0.42105263471603394, 0.3947368562221527, 0.5526315569877625, 0.4736842215061188, NaN, 0.5263158082962036, 0.5263158082962036, 0.44736841320991516, 0.4736842215061188, 0.5263158082962036, 0.5263158082962036, 0.5789473652839661, 0.6052631735801697, 0.5394737124443054, 0.5526315569877625, 0.5263158082962036, 0.5263158082962036, 0.5526315569877625, 0.6315789222717285, 0.5526315569877625, 0.5526315569877625, 0.6447368264198303, 0.7105262875556946, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.6578947305679321, 0.6052631735801697, 0.6578947305679321, 0.6578947305679321, 0.7631579041481018, 0.7105262875556946, 0.7368420958518982, 0.7105262875556946, 0.7631579041481018, 0.7763158082962036, 0.7631579041481018, 0.8421052694320679, 0.8421052694320679, 0.7631579041481018, 0.7894737124443054, 0.8552631735801697, 0.7631579041481018, 0.7894737124443054, 0.8684210777282715, 0.8684210777282715, 0.8684210777282715, 0.8157894611358643, 0.9210526347160339, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0], [0.10526315867900848, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.18421052396297455, 0.31578946113586426, 0.31578946113586426, 0.34210526943206787, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.4736842215061188, 0.5, 0.5, 0.4868420958518982, 0.3947368562221527, 0.5263158082962036, 0.5263158082962036, 0.4736842215061188, NaN, 0.5789473652839661, 0.43421053886413574, 0.5263158082962036, 0.5526315569877625, 0.5, 0.5789473652839661, 0.5526315569877625, 0.4736842215061188, 0.5789473652839661, 0.5526315569877625, 0.4736842215061188, 0.5789473652839661, 0.5789473652839661, 0.5263158082962036, 0.6052631735801697, 0.5263158082962036, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.6842105388641357, 0.6710526347160339, 0.7105262875556946, 0.7631579041481018, 0.7105262875556946, 0.8421052694320679, 0.7368420958518982, 0.7368420958518982, 0.7894737124443054, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.9210526347160339, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.8684210777282715, 0.9210526347160339, 0.9210526347160339, 1.0], [0.1315789520740509, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.21052631735801697, 0.1315789520740509, 0.28947368264198303, 0.2368421107530594, 0.2368421107530594, 0.3684210479259491, 0.3947368562221527, 0.42105263471603394, 0.5, 0.34210526943206787, 0.42105263471603394, 0.34210526943206787, 0.3684210479259491, 0.42105263471603394, 0.4736842215061188, 0.44736841320991516, 0.4736842215061188, 0.42105263471603394, NaN, 0.4736842215061188, 0.4736842215061188, 0.44736841320991516, 0.4736842215061188, 0.5, 0.5526315569877625, 0.5, 0.42105263471603394, 0.5789473652839661, 0.42105263471603394, 0.6578947305679321, 0.6578947305679321, 0.5789473652839661, 0.5526315569877625, 0.5789473652839661, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.7368420958518982, 0.6578947305679321, 0.6315789222717285, 0.7105262875556946, 0.6052631735801697, 0.7105262875556946, 0.7631579041481018, 0.7368420958518982, 0.7894737124443054, 0.7368420958518982, 0.7105262875556946, 0.8157894611358643, 0.6842105388641357, 0.7631579041481018, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.9473684430122375, 0.8421052694320679, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964, 1.0], [0.1315789520740509, 0.07894736528396606, 0.07894736528396606, 0.10526315867900848, 0.2368421107530594, 0.15789473056793213, 0.2631579041481018, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.4736842215061188, 0.42105263471603394, 0.5, 0.4736842215061188, 0.42105263471603394, 0.44736841320991516, 0.5263158082962036, 0.5, 0.5526315569877625, 0.5657894611358643, 0.5263158082962036, NaN, 0.5263158082962036, 0.6052631735801697, 0.5263158082962036, 0.5526315569877625, 0.5, 0.42105263471603394, 0.5789473652839661, 0.5526315569877625, 0.3947368562221527, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.5263158082962036, 0.5, 0.6052631735801697, 0.6578947305679321, 0.7105262875556946, 0.6052631735801697, 0.5789473652839661, 0.6052631735801697, 0.6578947305679321, 0.6052631735801697, 0.6842105388641357, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.7368420958518982, 0.8157894611358643, 0.7368420958518982, 0.7105262875556946, 0.8421052694320679, 0.7631579041481018, 0.8157894611358643, 0.7105262875556946, 0.7105262875556946, 0.7368420958518982, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.8552631735801697, 0.8421052694320679, 0.8947368264198303, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.9736841917037964], [0.10526315867900848, 0.07894736528396606, 0.1315789520740509, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.31578946113586426, 0.28947368264198303, 0.21052631735801697, 0.34210526943206787, 0.44736841320991516, 0.42105263471603394, 0.3684210479259491, 0.34210526943206787, 0.4736842215061188, 0.3947368562221527, 0.42105263471603394, 0.44736841320991516, 0.3552631437778473, 0.4736842215061188, 0.44736841320991516, 0.5263158082962036, 0.4736842215061188, 0.5263158082962036, 0.4736842215061188, NaN, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, 0.5526315569877625, 0.4736842215061188, 0.3947368562221527, 0.5789473652839661, 0.44736841320991516, 0.6052631735801697, 0.6052631735801697, 0.6315789222717285, 0.5789473652839661, 0.5, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.8157894611358643, 0.6315789222717285, 0.6842105388641357, 0.6578947305679321, 0.6052631735801697, 0.6842105388641357, 0.7105262875556946, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.7631579041481018, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.7105262875556946, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375, 0.8421052694320679, 0.9736841917037964, 0.9473684430122375, 1.0], [0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.21052631735801697, 0.15789473056793213, 0.2368421107530594, 0.34210526943206787, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.4736842215061188, 0.5, 0.28947368264198303, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, 0.42105263471603394, 0.4736842215061188, 0.44736841320991516, 0.5526315569877625, 0.3947368562221527, 0.5263158082962036, NaN, 0.4736842215061188, 0.5, 0.5, 0.44736841320991516, 0.5789473652839661, 0.6052631735801697, 0.42105263471603394, 0.6052631735801697, 0.5263158082962036, 0.6578947305679321, 0.6578947305679321, 0.5263158082962036, 0.6315789222717285, 0.6578947305679321, 0.6315789222717285, 0.6842105388641357, 0.6315789222717285, 0.6842105388641357, 0.7105262875556946, 0.5789473652839661, 0.6842105388641357, 0.7105262875556946, 0.7631579041481018, 0.7631579041481018, 0.7105262875556946, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.8421052694320679, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.9473684430122375, 1.0], [0.15789473056793213, 0.07894736528396606, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.21052631735801697, 0.21052631735801697, 0.2368421107530594, 0.31578946113586426, 0.31578946113586426, 0.31578946113586426, 0.3684210479259491, 0.34210526943206787, 0.3947368562221527, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.4736842215061188, 0.31578946113586426, 0.4736842215061188, 0.5, 0.5263158082962036, 0.4736842215061188, 0.5263158082962036, 0.5263158082962036, NaN, 0.4736842215061188, 0.5263158082962036, 0.42105263471603394, 0.5, 0.5526315569877625, 0.4736842215061188, 0.6315789222717285, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.5263158082962036, 0.5789473652839661, 0.6578947305679321, 0.6315789222717285, 0.6315789222717285, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.6315789222717285, 0.6315789222717285, 0.6842105388641357, 0.8421052694320679, 0.6842105388641357, 0.8157894611358643, 0.7368420958518982, 0.7894737124443054, 0.8157894611358643, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.7631579041481018, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.9210526347160339, 0.8684210777282715, 0.9473684430122375, 0.9210526347160339, 1.0], [0.05263157933950424, 0.18421052396297455, 0.10526315867900848, 0.2368421107530594, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.2631579041481018, 0.28947368264198303, 0.3684210479259491, 0.42105263471603394, 0.3947368562221527, 0.44736841320991516, 0.5, 0.44736841320991516, 0.31578946113586426, 0.44736841320991516, 0.44736841320991516, 0.3947368562221527, 0.28947368264198303, 0.5, 0.42105263471603394, 0.42105263471603394, 0.5, 0.44736841320991516, 0.5263158082962036, 0.5, 0.5263158082962036, NaN, 0.5789473652839661, 0.5526315569877625, 0.44736841320991516, 0.5526315569877625, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.5526315569877625, 0.5263158082962036, 0.6315789222717285, 0.5263158082962036, 0.5789473652839661, 0.5526315569877625, 0.6052631735801697, 0.5526315569877625, 0.5526315569877625, 0.6315789222717285, 0.5789473652839661, 0.8026315569877625, 0.5526315569877625, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.7105262875556946, 0.5789473652839661, 0.7631579041481018, 0.6315789222717285, 0.7631579041481018, 0.7631579041481018, 0.7105262875556946, 0.7631579041481018, 0.7631579041481018, 0.8947368264198303, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.9473684430122375, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.8947368264198303, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.9210526347160339, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964], [0.1315789520740509, 0.21052631735801697, 0.1315789520740509, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.2631579041481018, 0.15789473056793213, 0.28947368264198303, 0.3684210479259491, 0.42105263471603394, 0.3684210479259491, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, 0.32894736528396606, 0.42105263471603394, 0.44736841320991516, 0.3947368562221527, 0.4736842215061188, 0.4736842215061188, 0.3947368562221527, 0.44736841320991516, 0.44736841320991516, 0.5, 0.44736841320991516, 0.5, 0.4736842215061188, 0.42105263471603394, NaN, 0.5, 0.5263158082962036, 0.5, 0.5, 0.6052631735801697, 0.5263158082962036, 0.5789473652839661, 0.5263158082962036, 0.5526315569877625, 0.5, 0.6052631735801697, 0.5789473652839661, 0.6842105388641357, 0.6578947305679321, 0.5789473652839661, 0.6578947305679321, 0.6052631735801697, 0.6578947305679321, 0.7631579041481018, 0.6052631735801697, 0.6052631735801697, 0.6578947305679321, 0.6578947305679321, 0.6315789222717285, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.6578947305679321, 0.7631579041481018, 0.7631579041481018, 0.8947368264198303, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8157894611358643, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.9736841917037964, 0.8421052694320679, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.9736841917037964, 1.0], [0.02631578966975212, 0.2368421107530594, 0.21052631735801697, 0.18421052396297455, 0.10526315867900848, 0.2368421107530594, 0.2763157784938812, 0.34210526943206787, 0.42105263471603394, 0.42105263471603394, 0.44736841320991516, 0.5, 0.34210526943206787, 0.4736842215061188, 0.5, 0.3684210479259491, 0.5263158082962036, 0.5, 0.5, 0.3947368562221527, 0.6052631735801697, 0.46052631735801697, 0.5263158082962036, 0.5, 0.5789473652839661, 0.5263158082962036, 0.5526315569877625, 0.5789473652839661, 0.44736841320991516, 0.5, NaN, 0.6052631735801697, 0.5789473652839661, 0.5263158082962036, 0.5526315569877625, 0.5, 0.6052631735801697, 0.5789473652839661, 0.6578947305679321, 0.5789473652839661, 0.6315789222717285, 0.6315789222717285, 0.6052631735801697, 0.5263158082962036, 0.6315789222717285, 0.6315789222717285, 0.6052631735801697, 0.5526315569877625, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.5789473652839661, 0.6052631735801697, 0.6578947305679321, 0.6578947305679321, 0.7368420958518982, 0.6842105388641357, 0.6842105388641357, 0.6710526347160339, 0.6315789222717285, 0.6315789222717285, 0.7631579041481018, 0.7368420958518982, 0.7368420958518982, 0.6842105388641357, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.7368420958518982, 0.8421052694320679, 0.8684210777282715, 0.7631579041481018, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8157894611358643, 1.0], [0.05263157933950424, 0.15789473056793213, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.2368421107530594, 0.28947368264198303, 0.31578946113586426, 0.31578946113586426, 0.3947368562221527, 0.5, 0.42105263471603394, 0.4736842215061188, 0.44736841320991516, 0.4736842215061188, 0.3684210479259491, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, 0.42105263471603394, 0.5789473652839661, 0.42105263471603394, 0.6052631735801697, 0.42105263471603394, 0.5, 0.5526315569877625, 0.4736842215061188, 0.3947368562221527, NaN, 0.5263158082962036, 0.44736841320991516, 0.5263158082962036, 0.5263158082962036, 0.5526315569877625, 0.4736842215061188, 0.5, 0.5789473652839661, 0.5526315569877625, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.7368420958518982, 0.6052631735801697, 0.5526315569877625, 0.6578947305679321, 0.6315789222717285, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.7368420958518982, 0.7631579041481018, 0.7368420958518982, 0.7105262875556946, 0.7631579041481018, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8684210777282715, 0.7894737124443054, 0.8947368264198303, 0.7631579041481018, 0.8157894611358643, 0.8947368264198303, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 1.0], [0.1315789520740509, 0.07894736528396606, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.21052631735801697, 0.28947368264198303, 0.18421052396297455, 0.31578946113586426, 0.42105263471603394, 0.31578946113586426, 0.34210526943206787, 0.34210526943206787, 0.3947368562221527, 0.3947368562221527, 0.34210526943206787, 0.3947368562221527, 0.3684210479259491, 0.4736842215061188, 0.3684210479259491, 0.4736842215061188, 0.44736841320991516, 0.42105263471603394, 0.44736841320991516, 0.42105263471603394, 0.3947368562221527, 0.44736841320991516, 0.44736841320991516, 0.5, 0.42105263471603394, 0.4736842215061188, NaN, 0.3947368562221527, 0.5263158082962036, 0.4736842215061188, 0.6315789222717285, 0.5, 0.4736842215061188, 0.5789473652839661, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.6315789222717285, 0.6578947305679321, 0.6842105388641357, 0.6052631735801697, 0.6578947305679321, 0.7105262875556946, 0.7105262875556946, 0.6578947305679321, 0.6315789222717285, 0.6315789222717285, 0.7105262875556946, 0.6842105388641357, 0.7894737124443054, 0.7368420958518982, 0.7368420958518982, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.7631579041481018, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.9473684430122375, 0.9210526347160339, 1.0], [0.0, 0.21052631735801697, 0.1315789520740509, 0.1315789520740509, 0.1315789520740509, 0.21052631735801697, 0.1315789520740509, 0.18421052396297455, 0.34210526943206787, 0.3684210479259491, 0.4736842215061188, 0.5263158082962036, 0.3947368562221527, 0.5, 0.44736841320991516, 0.3684210479259491, 0.5263158082962036, 0.4736842215061188, 0.5526315569877625, 0.3947368562221527, 0.5, 0.4736842215061188, 0.5263158082962036, 0.5789473652839661, 0.6052631735801697, 0.5526315569877625, 0.5789473652839661, 0.5263158082962036, 0.44736841320991516, 0.5, 0.4736842215061188, 0.5526315569877625, 0.6052631735801697, NaN, 0.5263158082962036, 0.5263158082962036, 0.5526315569877625, 0.6578947305679321, 0.6052631735801697, 0.5263158082962036, 0.5526315569877625, 0.5789473652839661, 0.5789473652839661, 0.5263158082962036, 0.5789473652839661, 0.6052631735801697, 0.6052631735801697, 0.6052631735801697, 0.6052631735801697, 0.6315789222717285, 0.7105262875556946, 0.5526315569877625, 0.6578947305679321, 0.5789473652839661, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.6315789222717285, 0.6447368264198303, 0.6578947305679321, 0.6578947305679321, 0.7763158082962036, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.7368420958518982, 0.7894737124443054, 0.9210526347160339, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.8421052694320679, 1.0], [0.07894736528396606, 0.07894736528396606, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.2368421107530594, 0.18421052396297455, 0.21052631735801697, 0.28947368264198303, 0.34210526943206787, 0.42105263471603394, 0.3684210479259491, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.3684210479259491, 0.3684210479259491, 0.4736842215061188, 0.42105263471603394, 0.44736841320991516, 0.44736841320991516, 0.42105263471603394, 0.34210526943206787, 0.44736841320991516, 0.3947368562221527, 0.3947368562221527, 0.3684210479259491, 0.4736842215061188, 0.3947368562221527, 0.44736841320991516, 0.4736842215061188, 0.4736842215061188, 0.4736842215061188, NaN, 0.42105263471603394, 0.5, 0.5526315569877625, 0.5263158082962036, 0.5263158082962036, 0.6315789222717285, 0.6052631735801697, 0.44736841320991516, 0.5526315569877625, 0.5263158082962036, 0.5789473652839661, 0.5789473652839661, 0.5789473652839661, 0.6578947305679321, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.7368420958518982, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.7368420958518982, 0.7368420958518982, 0.8684210777282715, 0.7631579041481018, 0.8421052694320679, 0.7894737124443054, 0.8157894611358643, 0.8947368264198303, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9210526347160339, 0.8947368264198303, 0.9473684430122375, 0.9736841917037964, 0.8684210777282715, 0.9473684430122375, 0.9736841917037964, 0.9736841917037964], [0.07894736528396606, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.1184210553765297, 0.21052631735801697, 0.21052631735801697, 0.28947368264198303, 0.3684210479259491, 0.31578946113586426, 0.31578946113586426, 0.2631579041481018, 0.31578946113586426, 0.34210526943206787, 0.2631579041481018, 0.3947368562221527, 0.34210526943206787, 0.3947368562221527, 0.3684210479259491, 0.42105263471603394, 0.34210526943206787, 0.4736842215061188, 0.3947368562221527, 0.4736842215061188, 0.44736841320991516, 0.3947368562221527, 0.4736842215061188, 0.5, 0.4736842215061188, 0.5263158082962036, 0.4736842215061188, 0.5789473652839661, NaN, 0.5, 0.5, 0.5263158082962036, 0.5789473652839661, 0.6315789222717285, 0.5789473652839661, 0.4736842215061188, 0.6578947305679321, 0.6052631735801697, 0.7105262875556946, 0.6842105388641357, 0.6315789222717285, 0.7368420958518982, 0.5526315569877625, 0.6578947305679321, 0.7368420958518982, 0.7105262875556946, 0.7894737124443054, 0.8157894611358643, 0.7105262875556946, 0.6973684430122375, 0.7631579041481018, 0.7894737124443054, 0.7631579041481018, 0.7631579041481018, 0.7368420958518982, 0.7631579041481018, 0.8947368264198303, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.8684210777282715, 0.9210526347160339, 0.8947368264198303, 0.8947368264198303, 0.8684210777282715, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964, 1.0], [0.07894736528396606, 0.15789473056793213, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.21052631735801697, 0.2368421107530594, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.31578946113586426, 0.3684210479259491, 0.34210526943206787, 0.3947368562221527, 0.31578946113586426, 0.3947368562221527, 0.34210526943206787, 0.42105263471603394, 0.3684210479259491, 0.44736841320991516, 0.4736842215061188, 0.42105263471603394, 0.3947368562221527, 0.3684210479259491, 0.34210526943206787, 0.44736841320991516, 0.44736841320991516, 0.42105263471603394, 0.3947368562221527, 0.44736841320991516, 0.3684210479259491, 0.44736841320991516, 0.5, 0.5, NaN, 0.5, 0.4736842215061188, 0.5789473652839661, 0.5789473652839661, 0.6315789222717285, 0.5, 0.6578947305679321, 0.7631579041481018, 0.7894737124443054, 0.5526315569877625, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.7894737124443054, 0.7105262875556946, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.7631579041481018, 0.7894737124443054, 0.7631579041481018, 0.7631579041481018, 0.8684210777282715, 0.8421052694320679, 0.7894737124443054, 0.8684210777282715, 0.8157894611358643, 0.8947368264198303, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.9473684430122375, 0.9473684430122375, 0.8684210777282715, 0.9736841917037964, 0.9473684430122375, 1.0], [0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.15789473056793213, 0.34210526943206787, 0.3947368562221527, 0.3684210479259491, 0.3947368562221527, 0.34210526943206787, 0.44736841320991516, 0.31578946113586426, 0.42105263471603394, 0.3947368562221527, 0.44736841320991516, 0.3947368562221527, 0.44736841320991516, 0.4736842215061188, 0.42105263471603394, 0.34210526943206787, 0.44736841320991516, 0.4736842215061188, 0.4736842215061188, 0.42105263471603394, 0.5263158082962036, 0.5, 0.34210526943206787, 0.44736841320991516, 0.5, 0.5, NaN, 0.42105263471603394, 0.5789473652839661, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.5789473652839661, 0.5394737124443054, 0.5263158082962036, 0.6315789222717285, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.6052631735801697, 0.7105262875556946, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.8157894611358643, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.7894737124443054, 0.8421052694320679, 0.9210526347160339, 0.8421052694320679, 0.8421052694320679, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.7894737124443054, 0.8684210777282715, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964, 0.9473684430122375], [0.02631578966975212, 0.1315789520740509, 0.07894736528396606, 0.1315789520740509, 0.05263157933950424, 0.18421052396297455, 0.05263157933950424, 0.1315789520740509, 0.31578946113586426, 0.28947368264198303, 0.3684210479259491, 0.4736842215061188, 0.28947368264198303, 0.3684210479259491, 0.3947368562221527, 0.31578946113586426, 0.44736841320991516, 0.3684210479259491, 0.4736842215061188, 0.31578946113586426, 0.4736842215061188, 0.3552631437778473, 0.4736842215061188, 0.42105263471603394, 0.5, 0.5, 0.4736842215061188, 0.4736842215061188, 0.3684210479259491, 0.44736841320991516, 0.34210526943206787, 0.5, 0.5263158082962036, 0.3947368562221527, 0.4736842215061188, 0.4736842215061188, 0.5263158082962036, 0.5789473652839661, NaN, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.5789473652839661, 0.4868420958518982, 0.5263158082962036, 0.6052631735801697, 0.5263158082962036, 0.5263158082962036, 0.5526315569877625, 0.6315789222717285, 0.6578947305679321, 0.5263158082962036, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.6578947305679321, 0.6578947305679321, 0.6315789222717285, 0.6184210777282715, 0.6315789222717285, 0.6315789222717285, 0.7368420958518982, 0.7105262875556946, 0.6842105388641357, 0.6842105388641357, 0.7631579041481018, 0.7368420958518982, 0.7368420958518982, 0.7368420958518982, 0.8421052694320679, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 1.0], [0.05263157933950424, 0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.18421052396297455, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.21052631735801697, 0.28947368264198303, 0.31578946113586426, 0.2631579041481018, 0.2631579041481018, 0.34210526943206787, 0.2631579041481018, 0.28947368264198303, 0.3684210479259491, 0.34210526943206787, 0.31578946113586426, 0.28947368264198303, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.3684210479259491, 0.3684210479259491, 0.42105263471603394, 0.4736842215061188, 0.5, 0.42105263471603394, 0.42105263471603394, 0.42105263471603394, 0.4736842215061188, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, 0.42105263471603394, 0.44736841320991516, NaN, 0.5526315569877625, 0.44736841320991516, 0.44736841320991516, 0.5, 0.4736842215061188, 0.4736842215061188, 0.5526315569877625, 0.6315789222717285, 0.6052631735801697, 0.6578947305679321, 0.7368420958518982, 0.6578947305679321, 0.6052631735801697, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.7631579041481018, 0.9473684430122375, 0.7631579041481018, 0.8421052694320679, 0.8684210777282715, 0.7894737124443054, 0.8684210777282715, 0.8157894611358643, 0.8157894611358643, 0.9210526347160339, 0.8947368264198303, 0.9210526347160339, 0.8947368264198303, 0.9736841917037964, 0.9473684430122375, 0.9736841917037964], [0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.2368421107530594, 0.31578946113586426, 0.42105263471603394, 0.2631579041481018, 0.2631579041481018, 0.2631579041481018, 0.31578946113586426, 0.28947368264198303, 0.2631579041481018, 0.42105263471603394, 0.34210526943206787, 0.31578946113586426, 0.3684210479259491, 0.31578946113586426, 0.28947368264198303, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.34210526943206787, 0.42105263471603394, 0.3947368562221527, 0.3684210479259491, 0.44736841320991516, 0.3684210479259491, 0.44736841320991516, 0.3684210479259491, 0.3684210479259491, 0.42105263471603394, 0.44736841320991516, 0.44736841320991516, 0.44736841320991516, NaN, 0.5, 0.42105263471603394, 0.5789473652839661, 0.5263158082962036, 0.44736841320991516, 0.5, 0.5263158082962036, 0.5789473652839661, 0.5, 0.6315789222717285, 0.7105262875556946, 0.5789473652839661, 0.6842105388641357, 0.7368420958518982, 0.6315789222717285, 0.6842105388641357, 0.6578947305679321, 0.7368420958518982, 0.7105262875556946, 0.7368420958518982, 0.7894737124443054, 0.6842105388641357, 0.8684210777282715, 0.7894737124443054, 0.7631579041481018, 0.8421052694320679, 0.8157894611358643, 0.9210526347160339, 0.9210526347160339, 0.9736841917037964, 0.8421052694320679, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964, 1.0], [0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.2631579041481018, 0.28947368264198303, 0.2631579041481018, 0.3684210479259491, 0.31578946113586426, 0.2631579041481018, 0.34210526943206787, 0.3684210479259491, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.2368421107530594, 0.28947368264198303, 0.31578946113586426, 0.3684210479259491, 0.3684210479259491, 0.44736841320991516, 0.42105263471603394, 0.3684210479259491, 0.3947368562221527, 0.31578946113586426, 0.42105263471603394, 0.3947368562221527, 0.42105263471603394, 0.3684210479259491, 0.44736841320991516, 0.44736841320991516, 0.5526315569877625, 0.5, NaN, 0.44736841320991516, 0.5, 0.5526315569877625, 0.5, 0.5526315569877625, 0.6315789222717285, 0.5789473652839661, 0.6315789222717285, 0.6578947305679321, 0.6052631735801697, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.6052631735801697, 0.6578947305679321, 0.7368420958518982, 0.6842105388641357, 0.7105262875556946, 0.7631579041481018, 0.7105262875556946, 0.7631579041481018, 0.8947368264198303, 0.8421052694320679, 0.7894737124443054, 0.9210526347160339, 0.7894737124443054, 0.8947368264198303, 0.8421052694320679, 0.8157894611358643, 0.9473684430122375, 0.8947368264198303, 0.9210526347160339, 0.9210526347160339, 0.9736841917037964, 0.9473684430122375, 1.0], [0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.15789473056793213, 0.2368421107530594, 0.34210526943206787, 0.28947368264198303, 0.31578946113586426, 0.2631579041481018, 0.28947368264198303, 0.3684210479259491, 0.2631579041481018, 0.18421052396297455, 0.3684210479259491, 0.4736842215061188, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.2631579041481018, 0.3947368562221527, 0.18421052396297455, 0.31578946113586426, 0.3684210479259491, 0.3947368562221527, 0.31578946113586426, 0.3947368562221527, 0.3684210479259491, 0.31578946113586426, 0.42105263471603394, 0.5526315569877625, 0.5263158082962036, 0.5, 0.44736841320991516, 0.42105263471603394, 0.5526315569877625, 0.5789473652839661, 0.5526315569877625, NaN, 0.6052631735801697, 0.5789473652839661, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.6315789222717285, 0.5789473652839661, 0.5526315569877625, 0.5789473652839661, 0.6315789222717285, 0.6842105388641357, 0.5789473652839661, 0.6052631735801697, 0.6842105388641357, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.8157894611358643, 0.7368420958518982, 0.7105262875556946, 0.7894737124443054, 0.8421052694320679, 0.7631579041481018, 0.7894737124443054, 0.7368420958518982, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.8684210777282715, 0.8684210777282715, 0.7894737124443054, 0.9736841917037964, 0.9473684430122375, 1.0], [0.10526315867900848, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.1315789520740509, 0.2368421107530594, 0.15789473056793213, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.34210526943206787, 0.28947368264198303, 0.34210526943206787, 0.28947368264198303, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.2631579041481018, 0.34210526943206787, 0.34210526943206787, 0.3684210479259491, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.42105263471603394, 0.3684210479259491, 0.3684210479259491, 0.3947368562221527, 0.44736841320991516, 0.34210526943206787, 0.4736842215061188, 0.3947368562221527, 0.3684210479259491, 0.4736842215061188, 0.44736841320991516, 0.34210526943206787, 0.34210526943206787, 0.44736841320991516, 0.5131579041481018, 0.5, 0.42105263471603394, 0.5, 0.3947368562221527, NaN, 0.42105263471603394, 0.5, 0.5789473652839661, 0.5526315569877625, 0.6052631735801697, 0.4736842215061188, 0.5789473652839661, 0.6052631735801697, 0.5789473652839661, 0.5526315569877625, 0.6842105388641357, 0.6052631735801697, 0.6052631735801697, 0.7105262875556946, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.6842105388641357, 0.7894737124443054, 0.7368420958518982, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.7631579041481018, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.8684210777282715, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.8947368264198303, 1.0], [0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.10526315867900848, 0.18421052396297455, 0.2631579041481018, 0.2631579041481018, 0.2631579041481018, 0.34210526943206787, 0.28947368264198303, 0.34210526943206787, 0.21052631735801697, 0.31578946113586426, 0.28947368264198303, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.31578946113586426, 0.31578946113586426, 0.31578946113586426, 0.44736841320991516, 0.42105263471603394, 0.3684210479259491, 0.2631579041481018, 0.34210526943206787, 0.42105263471603394, 0.4736842215061188, 0.3947368562221527, 0.2368421107530594, 0.42105263471603394, 0.4736842215061188, 0.5263158082962036, 0.4736842215061188, 0.44736841320991516, 0.42105263471603394, 0.5789473652839661, NaN, 0.5, 0.5526315569877625, 0.6052631735801697, 0.5789473652839661, 0.5263158082962036, 0.5526315569877625, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.5789473652839661, 0.6842105388641357, 0.7105262875556946, 0.6578947305679321, 0.7631579041481018, 0.7631579041481018, 0.7368420958518982, 0.7631579041481018, 0.8157894611358643, 0.8421052694320679, 0.7631579041481018, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.8421052694320679, 0.8947368264198303, 0.9473684430122375, 1.0], [0.07894736528396606, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.18421052396297455, 0.2631579041481018, 0.21052631735801697, 0.2631579041481018, 0.2631579041481018, 0.2368421107530594, 0.28947368264198303, 0.18421052396297455, 0.2368421107530594, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.34210526943206787, 0.3684210479259491, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.28947368264198303, 0.31578946113586426, 0.3684210479259491, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.31578946113586426, 0.3947368562221527, 0.42105263471603394, 0.28947368264198303, 0.21052631735801697, 0.46052631735801697, 0.3947368562221527, 0.5263158082962036, 0.5526315569877625, 0.5, 0.44736841320991516, 0.5, 0.5, NaN, 0.5, 0.5526315569877625, 0.6315789222717285, 0.4736842215061188, 0.6315789222717285, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.5789473652839661, 0.6842105388641357, 0.7105262875556946, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7631579041481018, 0.8157894611358643, 0.8421052694320679, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.7368420958518982, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.8947368264198303, 0.8684210777282715, 0.8947368264198303, 0.8947368264198303, 0.9473684430122375, 0.9473684430122375, 1.0], [0.05263157933950424, 0.21052631735801697, 0.21052631735801697, 0.2368421107530594, 0.1315789520740509, 0.1315789520740509, 0.18421052396297455, 0.07894736528396606, 0.2368421107530594, 0.31578946113586426, 0.31578946113586426, 0.3684210479259491, 0.34210526943206787, 0.2631579041481018, 0.3684210479259491, 0.34210526943206787, 0.34210526943206787, 0.3947368562221527, 0.42105263471603394, 0.3684210479259491, 0.3684210479259491, 0.34210526943206787, 0.34210526943206787, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.42105263471603394, 0.3684210479259491, 0.42105263471603394, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.3947368562221527, 0.3947368562221527, 0.42105263471603394, 0.31578946113586426, 0.44736841320991516, 0.4736842215061188, 0.4736842215061188, 0.44736841320991516, 0.5, 0.44736841320991516, 0.44736841320991516, 0.42105263471603394, 0.44736841320991516, 0.5, NaN, 0.5526315569877625, 0.5263158082962036, 0.5, 0.5789473652839661, 0.4736842215061188, 0.44736841320991516, 0.5526315569877625, 0.5263158082962036, 0.5526315569877625, 0.6052631735801697, 0.6052631735801697, 0.5921052694320679, 0.6315789222717285, 0.6578947305679321, 0.7105262875556946, 0.7105262875556946, 0.6842105388641357, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.7105262875556946, 0.6578947305679321, 0.7368420958518982, 0.8421052694320679, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.8684210777282715, 0.8157894611358643, 1.0], [0.02631578966975212, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.21052631735801697, 0.31578946113586426, 0.2368421107530594, 0.34210526943206787, 0.31578946113586426, 0.28947368264198303, 0.18421052396297455, 0.28947368264198303, 0.31578946113586426, 0.2631579041481018, 0.2368421107530594, 0.31578946113586426, 0.2368421107530594, 0.31578946113586426, 0.28947368264198303, 0.31578946113586426, 0.31578946113586426, 0.31578946113586426, 0.3684210479259491, 0.19736842811107635, 0.34210526943206787, 0.44736841320991516, 0.34210526943206787, 0.34210526943206787, 0.3947368562221527, 0.42105263471603394, 0.3684210479259491, 0.31578946113586426, 0.3684210479259491, 0.4736842215061188, 0.3684210479259491, 0.4736842215061188, 0.3684210479259491, 0.44736841320991516, 0.44736841320991516, 0.3947368562221527, 0.44736841320991516, 0.44736841320991516, NaN, 0.5263158082962036, 0.5, 0.44736841320991516, 0.5, 0.6052631735801697, 0.5526315569877625, 0.5789473652839661, 0.6052631735801697, 0.5789473652839661, 0.7105262875556946, 0.6052631735801697, 0.6315789222717285, 0.6578947305679321, 0.7631579041481018, 0.7105262875556946, 0.7105262875556946, 0.7105262875556946, 0.7894737124443054, 0.8157894611358643, 0.7105262875556946, 0.7631579041481018, 0.7894737124443054, 0.7894737124443054, 0.8421052694320679, 0.7105262875556946, 0.7368420958518982, 0.8157894611358643, 0.8157894611358643, 0.8684210777282715, 0.9736841917037964], [0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.05263157933950424, 0.18421052396297455, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.21052631735801697, 0.21052631735801697, 0.34210526943206787, 0.18421052396297455, 0.2631579041481018, 0.28947368264198303, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.34210526943206787, 0.28947368264198303, 0.28947368264198303, 0.32894736528396606, 0.2368421107530594, 0.2631579041481018, 0.28947368264198303, 0.28947368264198303, 0.31578946113586426, 0.44736841320991516, 0.2368421107530594, 0.3947368562221527, 0.3684210479259491, 0.28947368264198303, 0.3947368562221527, 0.34210526943206787, 0.2631579041481018, 0.28947368264198303, 0.28947368264198303, 0.44736841320991516, 0.3947368562221527, 0.42105263471603394, 0.42105263471603394, 0.3684210479259491, 0.3947368562221527, 0.42105263471603394, 0.3684210479259491, 0.4736842215061188, 0.4736842215061188, NaN, 0.44736841320991516, 0.5526315569877625, 0.5789473652839661, 0.6052631735801697, 0.5526315569877625, 0.5789473652839661, 0.5263158082962036, 0.6052631735801697, 0.7368420958518982, 0.6052631735801697, 0.6315789222717285, 0.6315789222717285, 0.6842105388641357, 0.7368420958518982, 0.7105262875556946, 0.7368420958518982, 0.7631579041481018, 0.9736841917037964, 0.7368420958518982, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.8684210777282715, 0.8157894611358643, 0.8421052694320679, 0.8947368264198303, 0.8947368264198303, 0.8947368264198303, 0.9736841917037964], [0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.18421052396297455, 0.2631579041481018, 0.2631579041481018, 0.21052631735801697, 0.18421052396297455, 0.2631579041481018, 0.1315789520740509, 0.18421052396297455, 0.2631579041481018, 0.31578946113586426, 0.1315789520740509, 0.2631579041481018, 0.28947368264198303, 0.2631579041481018, 0.2631579041481018, 0.31578946113586426, 0.2368421107530594, 0.15789473056793213, 0.3947368562221527, 0.3947368562221527, 0.31578946113586426, 0.31578946113586426, 0.28947368264198303, 0.3684210479259491, 0.3947368562221527, 0.44736841320991516, 0.34210526943206787, 0.34210526943206787, 0.3684210479259491, 0.34210526943206787, 0.5, 0.3684210479259491, 0.42105263471603394, 0.5263158082962036, 0.4736842215061188, 0.5263158082962036, 0.5, 0.5, 0.5526315569877625, NaN, 0.5263158082962036, 0.6315789222717285, 0.6052631735801697, 0.6052631735801697, 0.6315789222717285, 0.6052631735801697, 0.7368420958518982, 0.6052631735801697, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.6842105388641357, 0.6842105388641357, 0.7368420958518982, 0.7631579041481018, 0.7631579041481018, 0.7631579041481018, 0.7894737124443054, 0.7631579041481018, 0.7368420958518982, 0.7894737124443054, 0.7631579041481018, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375], [0.02631578966975212, 0.10526315867900848, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.28947368264198303, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.21052631735801697, 0.18421052396297455, 0.2368421107530594, 0.31578946113586426, 0.30263158679008484, 0.28947368264198303, 0.2368421107530594, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.34210526943206787, 0.3947368562221527, 0.31578946113586426, 0.28947368264198303, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.34210526943206787, 0.31578946113586426, 0.31578946113586426, 0.34210526943206787, 0.2631579041481018, 0.3684210479259491, 0.34210526943206787, 0.44736841320991516, 0.42105263471603394, 0.44736841320991516, 0.3684210479259491, 0.42105263471603394, 0.5526315569877625, 0.44736841320991516, 0.4736842215061188, NaN, 0.5394737124443054, 0.6315789222717285, 0.5789473652839661, 0.6052631735801697, 0.5657894611358643, 0.6315789222717285, 0.5526315569877625, 0.5789473652839661, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.8947368264198303, 0.8157894611358643, 0.7894737124443054, 0.7894737124443054, 0.6842105388641357, 0.8421052694320679, 0.7631579041481018, 0.7894737124443054, 0.9210526347160339, 0.8157894611358643, 0.8421052694320679, 0.8421052694320679, 0.9210526347160339, 0.9473684430122375, 0.9736841917037964], [0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.1315789520740509, 0.02631578966975212, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.2368421107530594, 0.28947368264198303, 0.2631579041481018, 0.18421052396297455, 0.2368421107530594, 0.2368421107530594, 0.18421052396297455, 0.18421052396297455, 0.31578946113586426, 0.28947368264198303, 0.18421052396297455, 0.2368421107530594, 0.28947368264198303, 0.2631579041481018, 0.2631579041481018, 0.34210526943206787, 0.28947368264198303, 0.18421052396297455, 0.3684210479259491, 0.34210526943206787, 0.42105263471603394, 0.34210526943206787, 0.3684210479259491, 0.44736841320991516, 0.31578946113586426, 0.2631579041481018, 0.2631579041481018, 0.3947368562221527, 0.4736842215061188, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.42105263471603394, 0.3947368562221527, 0.3947368562221527, 0.34210526943206787, 0.5263158082962036, 0.5, 0.42105263471603394, 0.3684210479259491, 0.46052631735801697, NaN, 0.5, 0.5789473652839661, 0.6973684430122375, 0.5263158082962036, 0.5526315569877625, 0.5, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.7368420958518982, 0.6578947305679321, 0.6315789222717285, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.8684210777282715, 0.8947368264198303, 0.9210526347160339, 0.7894737124443054, 0.8157894611358643, 0.7631579041481018, 0.8421052694320679, 0.7894737124443054, 0.8421052694320679, 0.9210526347160339, 1.0], [0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.07894736528396606, 0.07894736528396606, 0.10526315867900848, 0.2368421107530594, 0.15789473056793213, 0.2631579041481018, 0.18421052396297455, 0.2368421107530594, 0.31578946113586426, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.22368420660495758, 0.15789473056793213, 0.28947368264198303, 0.18421052396297455, 0.31578946113586426, 0.31578946113586426, 0.2631579041481018, 0.28947368264198303, 0.34210526943206787, 0.3947368562221527, 0.34210526943206787, 0.3684210479259491, 0.34210526943206787, 0.2368421107530594, 0.28947368264198303, 0.21052631735801697, 0.28947368264198303, 0.44736841320991516, 0.3947368562221527, 0.42105263471603394, 0.31578946113586426, 0.3684210479259491, 0.42105263471603394, 0.34210526943206787, 0.31578946113586426, 0.5526315569877625, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.3684210479259491, 0.5, NaN, 0.5789473652839661, 0.5, 0.5, 0.5789473652839661, 0.5526315569877625, 0.6315789222717285, 0.7368420958518982, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.8157894611358643, 0.7894737124443054, 0.7368420958518982, 0.7894737124443054, 0.8157894611358643, 0.8157894611358643, 0.8157894611358643, 0.7105262875556946, 0.7631579041481018, 0.8157894611358643, 0.8157894611358643, 0.8421052694320679, 0.8684210777282715, 0.9210526347160339, 0.9736841917037964], [0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.10526315867900848, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.21052631735801697, 0.15789473056793213, 0.21052631735801697, 0.2631579041481018, 0.1315789520740509, 0.18421052396297455, 0.2368421107530594, 0.18421052396297455, 0.15789473056793213, 0.2631579041481018, 0.28947368264198303, 0.15789473056793213, 0.2368421107530594, 0.2631579041481018, 0.18421052396297455, 0.2631579041481018, 0.2368421107530594, 0.2368421107530594, 0.21052631735801697, 0.42105263471603394, 0.3684210479259491, 0.34210526943206787, 0.31578946113586426, 0.28947368264198303, 0.42105263471603394, 0.2631579041481018, 0.21052631735801697, 0.28947368264198303, 0.28947368264198303, 0.44736841320991516, 0.28947368264198303, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.44736841320991516, 0.3684210479259491, 0.34210526943206787, 0.44736841320991516, 0.44736841320991516, 0.44736841320991516, 0.3947368562221527, 0.42105263471603394, 0.42105263471603394, 0.42105263471603394, NaN, 0.5789473652839661, 0.5, 0.5, 0.5526315569877625, 0.7105262875556946, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6052631735801697, 0.7368420958518982, 0.7631579041481018, 0.7105262875556946, 0.7105262875556946, 0.8684210777282715, 0.7894737124443054, 0.8421052694320679, 0.7894737124443054, 0.7105262875556946, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.8157894611358643, 0.9210526347160339, 0.9736841917037964], [0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.28947368264198303, 0.2631579041481018, 0.21052631735801697, 0.2631579041481018, 0.21052631735801697, 0.2368421107530594, 0.21052631735801697, 0.31578946113586426, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.2631579041481018, 0.31578946113586426, 0.28947368264198303, 0.31578946113586426, 0.2631579041481018, 0.18421052396297455, 0.2368421107530594, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.3684210479259491, 0.2631579041481018, 0.18421052396297455, 0.31578946113586426, 0.34210526943206787, 0.44736841320991516, 0.34210526943206787, 0.2631579041481018, 0.31578946113586426, 0.42105263471603394, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, 0.3684210479259491, 0.3947368562221527, 0.30263158679008484, 0.5, 0.42105263471603394, NaN, 0.44736841320991516, 0.4736842215061188, 0.4736842215061188, 0.5789473652839661, 0.6315789222717285, 0.6315789222717285, 0.7105262875556946, 0.6578947305679321, 0.5789473652839661, 0.7631579041481018, 0.6842105388641357, 0.6315789222717285, 0.8157894611358643, 0.7894737124443054, 0.8157894611358643, 0.7105262875556946, 0.7894737124443054, 0.6842105388641357, 0.6842105388641357, 0.7631579041481018, 0.8421052694320679, 0.9736841917037964, 0.9736841917037964], [0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.07894736528396606, 0.15789473056793213, 0.15789473056793213, 0.2631579041481018, 0.15789473056793213, 0.18421052396297455, 0.18421052396297455, 0.05263157933950424, 0.18421052396297455, 0.21052631735801697, 0.2368421107530594, 0.02631578966975212, 0.15789473056793213, 0.21052631735801697, 0.2368421107530594, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.10526315867900848, 0.3684210479259491, 0.2631579041481018, 0.2631579041481018, 0.2631579041481018, 0.21052631735801697, 0.34210526943206787, 0.34210526943206787, 0.28947368264198303, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.42105263471603394, 0.42105263471603394, 0.44736841320991516, 0.3947368562221527, 0.4736842215061188, 0.3947368562221527, 0.43421053886413574, 0.4736842215061188, 0.5, 0.5, 0.5526315569877625, NaN, 0.6315789222717285, 0.5526315569877625, 0.5263158082962036, 0.6578947305679321, 0.6842105388641357, 0.6315789222717285, 0.6842105388641357, 0.6842105388641357, 0.7631579041481018, 0.7105262875556946, 0.7368420958518982, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.7631579041481018, 0.7631579041481018, 0.7105262875556946, 0.7105262875556946, 0.7894737124443054, 0.8684210777282715, 0.9210526347160339, 0.9473684430122375], [0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.07894736528396606, 0.10526315867900848, 0.1315789520740509, 0.21052631735801697, 0.2368421107530594, 0.1315789520740509, 0.15789473056793213, 0.21052631735801697, 0.10526315867900848, 0.15789473056793213, 0.28947368264198303, 0.2631579041481018, 0.15789473056793213, 0.2368421107530594, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.2631579041481018, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.2368421107530594, 0.2631579041481018, 0.34210526943206787, 0.31578946113586426, 0.30263158679008484, 0.31578946113586426, 0.34210526943206787, 0.34210526943206787, 0.28947368264198303, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.31578946113586426, 0.31578946113586426, 0.3947368562221527, 0.42105263471603394, 0.3947368562221527, 0.2631579041481018, 0.3684210479259491, 0.44736841320991516, 0.42105263471603394, 0.5, 0.5263158082962036, 0.3684210479259491, NaN, 0.4736842215061188, 0.4736842215061188, 0.6842105388641357, 0.6578947305679321, 0.6052631735801697, 0.6052631735801697, 0.6578947305679321, 0.7631579041481018, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.7894737124443054, 0.6842105388641357, 0.6842105388641357, 0.7105262875556946, 0.7105262875556946, 0.7105262875556946, 0.6578947305679321, 0.8157894611358643, 0.8421052694320679, 0.9473684430122375], [0.07894736528396606, 0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.07894736528396606, 0.07894736528396606, 0.10526315867900848, 0.02631578966975212, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.2368421107530594, 0.21052631735801697, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.28947368264198303, 0.21052631735801697, 0.21052631735801697, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.2631579041481018, 0.2631579041481018, 0.3684210479259491, 0.2631579041481018, 0.2368421107530594, 0.2368421107530594, 0.18421052396297455, 0.3684210479259491, 0.2631579041481018, 0.34210526943206787, 0.2631579041481018, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.3947368562221527, 0.28947368264198303, 0.2631579041481018, 0.3947368562221527, 0.44736841320991516, 0.5, 0.44736841320991516, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, 0.5263158082962036, NaN, 0.5, 0.5789473652839661, 0.5789473652839661, 0.5789473652839661, 0.6315789222717285, 0.6052631735801697, 0.7105262875556946, 0.6842105388641357, 0.8157894611358643, 0.6578947305679321, 0.7105262875556946, 0.7368420958518982, 0.7368420958518982, 0.7894737124443054, 0.7105262875556946, 0.7105262875556946, 0.7631579041481018, 0.8421052694320679, 0.8684210777282715, 0.9473684430122375], [0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.07894736528396606, 0.07894736528396606, 0.05263157933950424, 0.1315789520740509, 0.1315789520740509, 0.21052631735801697, 0.21052631735801697, 0.1315789520740509, 0.1315789520740509, 0.18421052396297455, 0.07894736528396606, 0.10526315867900848, 0.21052631735801697, 0.2631579041481018, 0.10526315867900848, 0.14473684132099152, 0.21052631735801697, 0.15789473056793213, 0.28947368264198303, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.28947368264198303, 0.34210526943206787, 0.32894736528396606, 0.28947368264198303, 0.18421052396297455, 0.3552631437778473, 0.2368421107530594, 0.21052631735801697, 0.2631579041481018, 0.2631579041481018, 0.3815789520740509, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.2631579041481018, 0.42105263471603394, 0.34210526943206787, 0.28947368264198303, 0.40789473056793213, 0.3947368562221527, 0.3947368562221527, 0.3684210479259491, 0.42105263471603394, 0.31578946113586426, 0.3684210479259491, 0.28947368264198303, 0.42105263471603394, 0.4736842215061188, 0.5263158082962036, 0.5, NaN, 0.6315789222717285, 0.6315789222717285, 0.6578947305679321, 0.6052631735801697, 0.6315789222717285, 0.7368420958518982, 0.6578947305679321, 0.7631579041481018, 0.8026315569877625, 0.7894737124443054, 0.7894737124443054, 0.7894737124443054, 0.7631579041481018, 0.7105262875556946, 0.6842105388641357, 0.6842105388641357, 0.7894737124443054, 0.8947368264198303, 0.9736841917037964], [0.07894736528396606, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.15789473056793213, 0.25, 0.2368421107530594, 0.21052631735801697, 0.2368421107530594, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.21052631735801697, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.18421052396297455, 0.28947368264198303, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.2368421107530594, 0.2368421107530594, 0.3684210479259491, 0.2368421107530594, 0.21052631735801697, 0.34210526943206787, 0.2631579041481018, 0.2368421107530594, 0.2368421107530594, 0.2631579041481018, 0.3684210479259491, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.2368421107530594, 0.34210526943206787, 0.2368421107530594, 0.2631579041481018, 0.3684210479259491, 0.3684210479259491, 0.3684210479259491, 0.31578946113586426, 0.2631579041481018, 0.34210526943206787, 0.2631579041481018, 0.34210526943206787, 0.3684210479259491, 0.34210526943206787, 0.31578946113586426, 0.42105263471603394, 0.3684210479259491, NaN, 0.6447368264198303, 0.5, 0.6315789222717285, 0.4736842215061188, 0.9473684430122375, 0.6315789222717285, 0.6052631735801697, 0.6578947305679321, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.6842105388641357, 0.6578947305679321, 0.7631579041481018, 0.8421052694320679, 0.9210526347160339], [0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.18421052396297455, 0.15789473056793213, 0.1315789520740509, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.18421052396297455, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.21052631735801697, 0.21052631735801697, 0.18421052396297455, 0.2631579041481018, 0.18421052396297455, 0.18421052396297455, 0.15789473056793213, 0.2368421107530594, 0.2368421107530594, 0.3684210479259491, 0.2368421107530594, 0.21052631735801697, 0.34210526943206787, 0.2631579041481018, 0.2368421107530594, 0.21052631735801697, 0.2368421107530594, 0.3684210479259491, 0.2631579041481018, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.34210526943206787, 0.2368421107530594, 0.2631579041481018, 0.34210526943206787, 0.34210526943206787, 0.3684210479259491, 0.31578946113586426, 0.28947368264198303, 0.31578946113586426, 0.2631579041481018, 0.31578946113586426, 0.3684210479259491, 0.31578946113586426, 0.34210526943206787, 0.42105263471603394, 0.3684210479259491, 0.3552631437778473, NaN, 0.5263158082962036, 0.6315789222717285, 0.5, 0.9736841917037964, 0.6315789222717285, 0.5789473652839661, 0.6315789222717285, 0.6052631735801697, 0.6315789222717285, 0.6315789222717285, 0.6578947305679321, 0.6578947305679321, 0.6578947305679321, 0.6578947305679321, 0.7631579041481018, 0.8421052694320679, 0.9210526347160339], [0.02631578966975212, 0.0, 0.0, 0.02631578966975212, 0.02631578966975212, 0.07894736528396606, 0.0, 0.0, 0.07894736528396606, 0.05263157933950424, 0.17105263471603394, 0.1315789520740509, 0.17105263471603394, 0.15789473056793213, 0.18421052396297455, 0.07894736528396606, 0.15789473056793213, 0.21052631735801697, 0.15789473056793213, 0.05263157933950424, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.18421052396297455, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.10526315867900848, 0.10526315867900848, 0.2368421107530594, 0.21052631735801697, 0.21052631735801697, 0.22368420660495758, 0.1315789520740509, 0.2631579041481018, 0.2368421107530594, 0.21052631735801697, 0.2631579041481018, 0.2368421107530594, 0.21052631735801697, 0.28947368264198303, 0.2631579041481018, 0.31578946113586426, 0.2631579041481018, 0.2368421107530594, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.31578946113586426, 0.28947368264198303, 0.2631579041481018, 0.28947368264198303, 0.3684210479259491, 0.28947368264198303, 0.3684210479259491, 0.3947368562221527, 0.42105263471603394, 0.34210526943206787, 0.5, 0.4736842215061188, NaN, 0.5526315569877625, 0.4736842215061188, 0.6842105388641357, 0.5526315569877625, 0.5263158082962036, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.6315789222717285, 0.7105262875556946, 0.6842105388641357, 0.6842105388641357, 0.6842105388641357, 0.7368420958518982, 0.7368420958518982, 0.9736841917037964], [0.0, 0.07894736528396606, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.21052631735801697, 0.21052631735801697, 0.1315789520740509, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.21052631735801697, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.21052631735801697, 0.2631579041481018, 0.18421052396297455, 0.2368421107530594, 0.2631579041481018, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.15789473056793213, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.2368421107530594, 0.28947368264198303, 0.21052631735801697, 0.2368421107530594, 0.18421052396297455, 0.28947368264198303, 0.28947368264198303, 0.2631579041481018, 0.31578946113586426, 0.2368421107530594, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.3684210479259491, 0.3947368562221527, 0.3684210479259491, 0.3684210479259491, 0.44736841320991516, NaN, 0.42105263471603394, 0.3684210479259491, 0.8157894611358643, 0.5263158082962036, 0.6052631735801697, 0.5526315569877625, 0.5526315569877625, 0.5789473652839661, 0.6052631735801697, 0.6578947305679321, 0.6578947305679321, 0.8684210777282715, 0.6842105388641357, 0.6842105388641357, 0.9210526347160339], [0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.18421052396297455, 0.05263157933950424, 0.07894736528396606, 0.15789473056793213, 0.10526315867900848, 0.07894736528396606, 0.02631578966975212, 0.18421052396297455, 0.10526315867900848, 0.1315789520740509, 0.15789473056793213, 0.05263157933950424, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.15789473056793213, 0.15789473056793213, 0.28947368264198303, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.31578946113586426, 0.05263157933950424, 0.1315789520740509, 0.10526315867900848, 0.21052631735801697, 0.2631579041481018, 0.18421052396297455, 0.15789473056793213, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.2631579041481018, 0.10526315867900848, 0.3684210479259491, 0.18421052396297455, 0.2631579041481018, 0.42105263471603394, 0.31578946113586426, 0.34210526943206787, 0.3947368562221527, 0.3684210479259491, 0.5263158082962036, 0.5, 0.5263158082962036, 0.5789473652839661, NaN, 0.6578947305679321, 0.6052631735801697, 0.5789473652839661, 0.5789473652839661, 0.6578947305679321, 0.6578947305679321, 0.6052631735801697, 0.6842105388641357, 0.6315789222717285, 0.7105262875556946, 0.7105262875556946, 0.7894737124443054, 0.8421052694320679, 0.9473684430122375], [0.05263157933950424, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.31578946113586426, 0.1315789520740509, 0.15789473056793213, 0.28947368264198303, 0.21052631735801697, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.31578946113586426, 0.2368421107530594, 0.21052631735801697, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.28947368264198303, 0.28947368264198303, 0.2631579041481018, 0.2368421107530594, 0.18421052396297455, 0.2631579041481018, 0.21052631735801697, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.2368421107530594, 0.28947368264198303, 0.2631579041481018, 0.05263157933950424, 0.02631578966975212, 0.31578946113586426, 0.6315789222717285, 0.34210526943206787, NaN, 0.6315789222717285, 0.31578946113586426, 0.5, 0.5263158082962036, 0.4736842215061188, 0.5, 0.5263158082962036, 0.5, 0.5, 0.6578947305679321, 0.5789473652839661, 0.6842105388641357, 0.8947368264198303], [0.0, 0.05263157933950424, 0.0, 0.0, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.10526315867900848, 0.07894736528396606, 0.18421052396297455, 0.05263157933950424, 0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.07894736528396606, 0.1315789520740509, 0.07894736528396606, 0.18421052396297455, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.1315789520740509, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.2368421107530594, 0.18421052396297455, 0.1315789520740509, 0.21052631735801697, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.2368421107530594, 0.21052631735801697, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.2631579041481018, 0.21052631735801697, 0.2368421107530594, 0.2368421107530594, 0.21052631735801697, 0.28947368264198303, 0.2631579041481018, 0.28947368264198303, 0.31578946113586426, 0.28947368264198303, 0.42105263471603394, 0.31578946113586426, 0.34210526943206787, 0.3684210479259491, 0.3684210479259491, 0.44736841320991516, 0.18421052396297455, 0.3947368562221527, 0.3684210479259491, NaN, 0.5, 0.5263158082962036, 0.4736842215061188, 0.4736842215061188, 0.5789473652839661, 0.5789473652839661, 0.6315789222717285, 0.6052631735801697, 0.7763158082962036, 0.6578947305679321, 0.5789473652839661, 0.9210526347160339], [0.0, 0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.0, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.07894736528396606, 0.10526315867900848, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.1315789520740509, 0.1315789520740509, 0.1315789520740509, 0.07894736528396606, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.05263157933950424, 0.1315789520740509, 0.18421052396297455, 0.1315789520740509, 0.18421052396297455, 0.21052631735801697, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.18421052396297455, 0.2631579041481018, 0.1315789520740509, 0.15789473056793213, 0.07894736528396606, 0.21052631735801697, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.2368421107530594, 0.18421052396297455, 0.02631578966975212, 0.2368421107530594, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.28947368264198303, 0.3684210479259491, 0.2631579041481018, 0.34210526943206787, 0.18421052396297455, 0.2368421107530594, 0.3947368562221527, 0.42105263471603394, 0.4736842215061188, 0.4736842215061188, 0.42105263471603394, 0.6842105388641357, 0.5, NaN, 0.5, 0.5526315569877625, 0.5, 0.6052631735801697, 0.5789473652839661, 0.5789473652839661, 0.6052631735801697, 0.7105262875556946, 0.6578947305679321, 0.7894737124443054, 0.9473684430122375], [0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.07894736528396606, 0.15789473056793213, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.02631578966975212, 0.05263157933950424, 0.1315789520740509, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.07894736528396606, 0.07894736528396606, 0.15789473056793213, 0.18421052396297455, 0.21052631735801697, 0.21052631735801697, 0.21052631735801697, 0.21052631735801697, 0.10526315867900848, 0.1315789520740509, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.21052631735801697, 0.18421052396297455, 0.21052631735801697, 0.2631579041481018, 0.2368421107530594, 0.21052631735801697, 0.2631579041481018, 0.28947368264198303, 0.28947368264198303, 0.2631579041481018, 0.21052631735801697, 0.31578946113586426, 0.1315789520740509, 0.18421052396297455, 0.1315789520740509, 0.18421052396297455, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.19736842811107635, 0.34210526943206787, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.42105263471603394, 0.5, 0.4736842215061188, 0.5, NaN, 0.5263158082962036, 0.5263158082962036, 0.5789473652839661, 0.5, 0.6052631735801697, 0.6052631735801697, 0.5789473652839661, 0.6578947305679321, 0.7105262875556946, 0.9736841917037964], [0.07894736528396606, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.1315789520740509, 0.1315789520740509, 0.05263157933950424, 0.10526315867900848, 0.07894736528396606, 0.07894736528396606, 0.1315789520740509, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.07894736528396606, 0.07894736528396606, 0.14473684132099152, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.2631579041481018, 0.10526315867900848, 0.18421052396297455, 0.2631579041481018, 0.07894736528396606, 0.07894736528396606, 0.10526315867900848, 0.18421052396297455, 0.2631579041481018, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.18421052396297455, 0.21052631735801697, 0.10526315867900848, 0.10526315867900848, 0.34210526943206787, 0.2368421107530594, 0.18421052396297455, 0.2368421107530594, 0.15789473056793213, 0.10526315867900848, 0.18421052396297455, 0.21052631735801697, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.3684210479259491, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.34210526943206787, 0.4736842215061188, 0.5263158082962036, 0.44736841320991516, 0.4736842215061188, NaN, 0.5526315569877625, 0.6315789222717285, 0.5526315569877625, 0.6052631735801697, 0.6052631735801697, 0.6052631735801697, 0.6842105388641357, 0.9736841917037964, 0.9736841917037964], [0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.15789473056793213, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.05263157933950424, 0.10526315867900848, 0.1315789520740509, 0.18421052396297455, 0.02631578966975212, 0.07894736528396606, 0.10526315867900848, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.10526315867900848, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.2368421107530594, 0.15789473056793213, 0.21052631735801697, 0.07894736528396606, 0.10526315867900848, 0.15789473056793213, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.07894736528396606, 0.15789473056793213, 0.21052631735801697, 0.18421052396297455, 0.18421052396297455, 0.18421052396297455, 0.2631579041481018, 0.21052631735801697, 0.21052631735801697, 0.2631579041481018, 0.2368421107530594, 0.07894736528396606, 0.18421052396297455, 0.15789473056793213, 0.18421052396297455, 0.2368421107530594, 0.31578946113586426, 0.2631579041481018, 0.21052631735801697, 0.34210526943206787, 0.3684210479259491, 0.31578946113586426, 0.44736841320991516, 0.34210526943206787, 0.5263158082962036, 0.5263158082962036, 0.5, 0.4736842215061188, 0.44736841320991516, NaN, 0.5526315569877625, 0.5263158082962036, 0.5526315569877625, 0.5526315569877625, 0.6052631735801697, 0.6052631735801697, 0.7105262875556946, 0.9473684430122375], [0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.07894736528396606, 0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.10526315867900848, 0.15789473056793213, 0.05263157933950424, 0.15789473056793213, 0.15789473056793213, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.07894736528396606, 0.07894736528396606, 0.10526315867900848, 0.15789473056793213, 0.21052631735801697, 0.15789473056793213, 0.18421052396297455, 0.02631578966975212, 0.18421052396297455, 0.21052631735801697, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.15789473056793213, 0.21052631735801697, 0.18421052396297455, 0.21052631735801697, 0.21052631735801697, 0.21052631735801697, 0.28947368264198303, 0.21052631735801697, 0.28947368264198303, 0.2368421107530594, 0.31578946113586426, 0.2631579041481018, 0.21052631735801697, 0.34210526943206787, 0.3684210479259491, 0.3684210479259491, 0.42105263471603394, 0.3947368562221527, 0.5, 0.42105263471603394, 0.3947368562221527, 0.42105263471603394, 0.3684210479259491, 0.44736841320991516, NaN, 0.5, 0.4736842215061188, 0.4736842215061188, 0.5, 0.5263158082962036, 0.6315789222717285, 0.8421052694320679], [0.0, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.10526315867900848, 0.02631578966975212, 0.10526315867900848, 0.0, 0.05263157933950424, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.15789473056793213, 0.07894736528396606, 0.1315789520740509, 0.05263157933950424, 0.1315789520740509, 0.15789473056793213, 0.07894736528396606, 0.07894736528396606, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.1315789520740509, 0.02631578966975212, 0.2368421107530594, 0.10526315867900848, 0.10526315867900848, 0.21052631735801697, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.1315789520740509, 0.21052631735801697, 0.07894736528396606, 0.15789473056793213, 0.05263157933950424, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.21052631735801697, 0.15789473056793213, 0.1315789520740509, 0.2368421107530594, 0.07894736528396606, 0.18421052396297455, 0.2368421107530594, 0.28947368264198303, 0.21052631735801697, 0.2368421107530594, 0.28947368264198303, 0.21052631735801697, 0.2368421107530594, 0.31578946113586426, 0.34210526943206787, 0.28947368264198303, 0.3947368562221527, 0.31578946113586426, 0.4736842215061188, 0.42105263471603394, 0.42105263471603394, 0.5, 0.44736841320991516, 0.4736842215061188, 0.5, NaN, 0.5526315569877625, 0.5526315569877625, 0.5526315569877625, 0.6315789222717285, 0.6842105388641357, 0.8684210777282715], [0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.07894736528396606, 0.02631578966975212, 0.15789473056793213, 0.02631578966975212, 0.02631578966975212, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.10526315867900848, 0.05263157933950424, 0.18421052396297455, 0.07894736528396606, 0.1315789520740509, 0.10526315867900848, 0.18421052396297455, 0.15789473056793213, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.15789473056793213, 0.05263157933950424, 0.15789473056793213, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.10526315867900848, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.28947368264198303, 0.18421052396297455, 0.18421052396297455, 0.18421052396297455, 0.2368421107530594, 0.18421052396297455, 0.2631579041481018, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.3684210479259491, 0.5, 0.3684210479259491, 0.42105263471603394, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, NaN, 0.75, 0.5, 0.7368420958518982, 0.6315789222717285, 0.8684210777282715], [0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.03947368264198303, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.02631578966975212, 0.15789473056793213, 0.02631578966975212, 0.02631578966975212, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.1315789520740509, 0.07894736528396606, 0.02631578966975212, 0.15789473056793213, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.18421052396297455, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.15789473056793213, 0.02631578966975212, 0.10526315867900848, 0.05263157933950424, 0.1315789520740509, 0.15789473056793213, 0.07894736528396606, 0.1315789520740509, 0.07894736528396606, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.10526315867900848, 0.18421052396297455, 0.2631579041481018, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.15789473056793213, 0.18421052396297455, 0.28947368264198303, 0.31578946113586426, 0.28947368264198303, 0.28947368264198303, 0.28947368264198303, 0.31578946113586426, 0.31578946113586426, 0.34210526943206787, 0.31578946113586426, 0.34210526943206787, 0.28947368264198303, 0.5, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.3947368562221527, 0.44736841320991516, 0.5263158082962036, 0.44736841320991516, 0.25, NaN, 0.5, 0.6578947305679321, 0.6052631735801697, 0.8421052694320679], [0.0, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.1315789520740509, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.1315789520740509, 0.05263157933950424, 0.10526315867900848, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.07894736528396606, 0.10526315867900848, 0.10526315867900848, 0.1315789520740509, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.10526315867900848, 0.10526315867900848, 0.07894736528396606, 0.21052631735801697, 0.15789473056793213, 0.15789473056793213, 0.10526315867900848, 0.1315789520740509, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.21052631735801697, 0.15789473056793213, 0.28947368264198303, 0.2368421107530594, 0.21052631735801697, 0.34210526943206787, 0.2368421107530594, 0.31578946113586426, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.1315789520740509, 0.28947368264198303, 0.34210526943206787, 0.22368420660495758, 0.28947368264198303, 0.42105263471603394, 0.3947368562221527, 0.3947368562221527, 0.5, 0.44736841320991516, 0.5, 0.5, NaN, 0.5263158082962036, 0.5, 0.8421052694320679], [0.0, 0.02631578966975212, 0.0, 0.02631578966975212, 0.05263157933950424, 0.0, 0.05263157933950424, 0.02631578966975212, 0.0, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.02631578966975212, 0.10526315867900848, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.02631578966975212, 0.15789473056793213, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.05263157933950424, 0.05263157933950424, 0.1315789520740509, 0.05263157933950424, 0.10526315867900848, 0.02631578966975212, 0.10526315867900848, 0.15789473056793213, 0.02631578966975212, 0.10526315867900848, 0.02631578966975212, 0.02631578966975212, 0.1315789520740509, 0.10526315867900848, 0.05263157933950424, 0.1315789520740509, 0.18421052396297455, 0.10526315867900848, 0.15789473056793213, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.1315789520740509, 0.18421052396297455, 0.15789473056793213, 0.21052631735801697, 0.2368421107530594, 0.2368421107530594, 0.2631579041481018, 0.31578946113586426, 0.21052631735801697, 0.42105263471603394, 0.34210526943206787, 0.34210526943206787, 0.34210526943206787, 0.31578946113586426, 0.3947368562221527, 0.4736842215061188, 0.3684210479259491, 0.2631579041481018, 0.34210526943206787, 0.4736842215061188, NaN, 0.6052631735801697, 0.7894737124443054], [0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.0, 0.1315789520740509, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.0, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.07894736528396606, 0.02631578966975212, 0.07894736528396606, 0.02631578966975212, 0.1315789520740509, 0.05263157933950424, 0.05263157933950424, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.18421052396297455, 0.02631578966975212, 0.07894736528396606, 0.15789473056793213, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.15789473056793213, 0.05263157933950424, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.10526315867900848, 0.05263157933950424, 0.05263157933950424, 0.18421052396297455, 0.1315789520740509, 0.10526315867900848, 0.1315789520740509, 0.05263157933950424, 0.07894736528396606, 0.07894736528396606, 0.07894736528396606, 0.02631578966975212, 0.07894736528396606, 0.15789473056793213, 0.1315789520740509, 0.10526315867900848, 0.15789473056793213, 0.15789473056793213, 0.2631579041481018, 0.31578946113586426, 0.15789473056793213, 0.31578946113586426, 0.42105263471603394, 0.21052631735801697, 0.28947368264198303, 0.02631578966975212, 0.28947368264198303, 0.3684210479259491, 0.31578946113586426, 0.3684210479259491, 0.3947368562221527, 0.5, 0.3947368562221527, NaN, 0.9473684430122375], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.0, 0.0, 0.05263157933950424, 0.0, 0.02631578966975212, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.02631578966975212, 0.0, 0.02631578966975212, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.05263157933950424, 0.05263157933950424, 0.02631578966975212, 0.07894736528396606, 0.07894736528396606, 0.02631578966975212, 0.07894736528396606, 0.05263157933950424, 0.10526315867900848, 0.07894736528396606, 0.05263157933950424, 0.02631578966975212, 0.02631578966975212, 0.05263157933950424, 0.15789473056793213, 0.1315789520740509, 0.1315789520740509, 0.15789473056793213, 0.15789473056793213, 0.21052631735801697, 0.05263157933950424, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9213943977},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8578263897},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8532125816},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8342447037},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.823137394},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8130553698},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7925495592},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7915242659},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7836637062},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7392344486},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6775461391},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6705399905},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.6592617896},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6561859227},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.654135336},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6519138772},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6483253568},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6471291854},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6310663031},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6305536595},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.624572799},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6238892709},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6115857862},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6093643188},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5974025974},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5955229001},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5936432016},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5933014362},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5811688295},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5794600163},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5791182508},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5717703351},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5673274108},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.56527683},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5584415598},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.558099796},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5574162705},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5439166094},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5165755293},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5164046478},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5071770322},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5068352687},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5058099797},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4981203035},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4941900208},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4885509221},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.4827409434},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4543745703},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4499316466},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4422419721},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4389952152},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4326725922},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4178058762},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4046479817},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3962747767},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3860218732},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3703007529},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3663704715},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.361927547},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3479152428},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3371496913},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2968215997},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2918660297},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2768284346},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2532467517},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2445317843},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2388926878},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2387218056},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2366712239},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2252221459},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2183868759},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2146274774},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2106971973},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1934381407},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1836978819},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1510594681},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1250854406},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0293916608}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9a9a30bbe2225a7efd3273c1a4519cc5871a35e7
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_all/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),TabICLv2 (default),RealTabPFN-2.5 (tuned),RealTabPFN-2.5 (default),"AutoGluon 1.4 (best, 4h)",RealMLP (tuned + ensembled),RealMLP (tuned),TabM (tuned + ensembled),LimiX (default),iLTM (tuned + ensembled),CatBoost (tuned),TabDPT-Turbo (default),LightGBM (tuned + ensembled),CatBoost (tuned + ensembled),TabM (tuned),TabDPT (tuned + ensembled),XGBoost (tuned + ensembled),TabICL (default),ModernNCA (tuned),CatBoost (default),ModernNCA (tuned + ensembled),ChimeraBoost (tuned + ensembled),LightGBM (tuned),XGBoost (tuned),TabDPT (tuned),TabSwift (default),TabPFNv2 (tuned + ensembled),BetaTabPFN (default),ChimeraBoost (tuned),Mitra (default),xRFM (tuned + ensembled),iLTM (tuned),EBM (tuned + ensembled),TabM (default),TabPFNv2 (tuned),TorchMLP (tuned + ensembled),xRFM (tuned),RealMLP (default),ChimeraBoost (default),SAP-RPT-OSS (default),EBM (tuned),EBM (default),TabPFNv2 (default),TabDPT (default),FastaiMLP (tuned + ensembled),XGBoost (default),TorchMLP (tuned),ExtraTrees (tuned + ensembled),ModernNCA (default),RandomForest (tuned + ensembled),ExtraTrees (tuned),LightGBM (default),iLTM (default),FastaiMLP (tuned),RandomForest (tuned),TabSTAR (tuned),TabSTAR (tuned + ensembled),OrionMSP (default),PerpetualBooster (tuned + ensembled),TorchMLP (default),TabSTAR (default),PerpetualBooster (tuned),FastaiMLP (default),RandomForest (default),KNN (tuned + ensembled),ExtraTrees (default),xRFM (default),TabFlex (default),Linear (tuned + ensembled),Linear (tuned),PerpetualBooster (default),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.7368421,0.81578946,0.84210527,0.8684211,0.8684211,0.8947368,0.84210527,0.7631579,0.81578946,0.84210527,0.92105263,0.8684211,0.81578946,0.8947368,0.92105263,0.92105263,0.92105263,0.94736844,0.92105263,0.8947368,0.92105263,0.8947368,0.8684211,0.8684211,0.8947368,0.94736844,0.84210527,0.94736844,0.8684211,0.9736842,0.94736844,0.8684211,1.0,0.92105263,0.92105263,0.92105263,0.92105263,0.9736842,0.94736844,0.92105263,0.8947368,0.92105263,0.8947368,0.92105263,0.92105263,0.94736844,0.9736842,0.8947368,0.94736844,0.9736842,0.94736844,0.94736844,0.94736844,0.94736844,0.94736844,0.92105263,0.92105263,0.94736844,0.92105263,0.8947368,0.9736842,1.0,0.9736842,0.94736844,1.0,1.0,0.9736842,0.92105263,0.9736842,0.94736844,1.0,0.9736842,0.9736842,1.0,1.0,0.9736842,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2631579,,0.47368422,0.6052632,0.6052632,0.5263158,0.68421054,0.57894737,0.68421054,0.81578946,0.81578946,0.81578946,0.7368421,0.8684211,0.84210527,0.68421054,0.8947368,0.8947368,0.7894737,0.81578946,0.92105263,0.81578946,0.8947368,0.92105263,0.92105263,0.92105263,0.92105263,0.92105263,0.81578946,0.7894737,0.7631579,0.84210527,0.92105263,0.7894737,0.92105263,0.8947368,0.84210527,0.8684211,0.8684211,0.92105263,0.8684211,0.8684211,0.94736844,0.81578946,0.8684211,0.8947368,0.7894737,0.84210527,0.8684211,0.92105263,0.8947368,0.92105263,0.94736844,0.94736844,0.92105263,0.9736842,0.94736844,0.92105263,0.94736844,0.8684211,0.8684211,1.0,0.92105263,0.94736844,0.8684211,0.94736844,0.9736842,0.94736844,0.9736842,0.9736842,0.9736842,0.94736844,0.9736842,0.9736842,0.94736844,0.9736842,1.0,1.0
+TabPFN-3 (default),0.18421052,0.5263158,,0.55263156,0.56578946,0.6052632,0.65789473,0.57894737,0.55263156,0.7105263,0.7368421,0.7894737,0.7105263,0.7894737,0.7368421,0.81578946,0.8684211,0.7894737,0.8684211,0.7894737,0.84210527,0.8947368,0.84210527,0.84210527,0.92105263,0.8684211,0.94736844,0.81578946,0.8947368,0.8684211,0.7894737,0.92105263,0.84210527,0.8684211,0.84210527,0.8947368,0.81578946,0.84210527,0.92105263,0.8684211,0.8947368,0.92105263,0.8947368,0.7894737,0.8947368,0.8947368,0.7894737,0.8947368,0.8684211,0.94736844,0.94736844,0.94736844,0.94736844,0.92105263,0.94736844,0.94736844,0.9736842,0.94736844,0.9736842,0.8684211,0.8684211,1.0,0.9736842,0.94736844,0.8684211,1.0,1.0,0.9736842,0.94736844,0.9736842,0.94736844,0.94736844,0.94736844,0.9736842,0.9736842,1.0,0.9736842,1.0
+TabPFN-2.6 (default),0.15789473,0.39473686,0.4473684,,0.47368422,0.5,0.55263156,0.65789473,0.5263158,0.7105263,0.7368421,0.7368421,0.7631579,0.81578946,0.7631579,0.7894737,0.8947368,0.81578946,0.7894737,0.7368421,0.7631579,0.8684211,0.81578946,0.8684211,0.8947368,0.81578946,0.9736842,0.84210527,0.7631579,0.7894737,0.81578946,0.84210527,0.81578946,0.8684211,0.84210527,0.8684211,0.8684211,0.84210527,0.8684211,0.8947368,0.84210527,0.8947368,0.84210527,0.81578946,0.92105263,0.8684211,0.7631579,0.81578946,0.8947368,0.94736844,0.94736844,0.92105263,0.94736844,0.94736844,0.92105263,0.94736844,0.94736844,0.8947368,0.92105263,0.84210527,0.84210527,0.9736842,0.9736842,0.9736842,0.84210527,1.0,0.9736842,0.9736842,0.9736842,0.94736844,0.94736844,0.92105263,0.9736842,0.9736842,0.9736842,0.9736842,0.9736842,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.13157895,0.39473686,0.43421054,0.5263158,,0.5263158,0.7105263,0.6315789,0.55263156,0.68421054,0.7105263,0.68421054,0.68421054,0.7368421,0.81578946,0.65789473,0.81578946,0.8684211,0.7631579,0.68421054,0.84210527,0.7368421,0.84210527,0.8684211,0.7631579,0.7894737,0.8947368,0.84210527,0.7631579,0.81578946,0.8947368,0.8684211,0.84210527,0.8684211,0.81578946,0.81578946,0.84210527,0.84210527,0.94736844,0.81578946,0.8684211,0.84210527,0.81578946,0.8684211,0.84210527,0.8684211,0.8684211,0.81578946,0.84210527,0.94736844,0.8684211,0.94736844,0.8684211,0.92105263,0.94736844,0.9736842,0.8947368,0.92105263,0.94736844,0.84210527,0.84210527,0.9736842,0.94736844,0.94736844,0.8684211,0.9736842,0.9736842,0.9736842,0.92105263,0.9736842,0.94736844,0.8947368,0.92105263,0.94736844,0.94736844,0.94736844,0.94736844,1.0
+TabICLv2 (default),0.13157895,0.47368422,0.39473686,0.5,0.47368422,,0.6052632,0.57894737,0.6315789,0.65789473,0.6315789,0.68421054,0.68421054,0.7631579,0.7368421,0.7894737,0.81578946,0.7368421,0.7105263,0.81578946,0.7631579,0.7894737,0.7894737,0.7894737,0.84210527,0.81578946,0.84210527,0.7631579,0.81578946,0.81578946,0.7631579,0.8947368,0.7894737,0.7894737,0.7894737,0.7631579,0.7631579,0.84210527,0.81578946,0.8684211,0.84210527,0.81578946,0.81578946,0.7631579,0.8684211,0.8947368,0.8684211,0.8684211,0.94736844,0.8684211,0.8947368,0.8684211,0.92105263,0.8947368,0.92105263,0.94736844,0.84210527,0.92105263,0.8947368,0.8684211,0.8684211,0.92105263,0.94736844,0.9736842,0.8947368,0.9736842,1.0,0.94736844,0.94736844,0.92105263,0.92105263,0.9736842,0.94736844,0.9736842,0.9736842,1.0,0.9736842,1.0
+RealTabPFN-2.5 (tuned),0.10526316,0.31578946,0.34210527,0.4473684,0.28947368,0.39473686,,0.56578946,0.5,0.6052632,0.7105263,0.7368421,0.57894737,0.65789473,0.7368421,0.65789473,0.7368421,0.7631579,0.7894737,0.65789473,0.7631579,0.7894737,0.81578946,0.8684211,0.7368421,0.68421054,0.8684211,0.7894737,0.7631579,0.7368421,0.7236842,0.84210527,0.7894737,0.8684211,0.7631579,0.84210527,0.84210527,0.84210527,0.94736844,0.81578946,0.84210527,0.84210527,0.7894737,0.84210527,0.84210527,0.84210527,0.81578946,0.81578946,0.81578946,0.92105263,0.8947368,0.9736842,0.8947368,0.92105263,0.8947368,0.94736844,0.8684211,0.8947368,0.92105263,0.84210527,0.84210527,1.0,0.94736844,0.94736844,0.8684211,0.9736842,0.94736844,0.9736842,0.92105263,0.9736842,0.94736844,0.8947368,0.92105263,0.92105263,0.94736844,0.94736844,0.94736844,1.0
+RealTabPFN-2.5 (default),0.15789473,0.42105263,0.42105263,0.34210527,0.36842105,0.42105263,0.43421054,,0.55263156,0.6052632,0.68421054,0.68421054,0.7105263,0.57894737,0.68421054,0.65789473,0.7631579,0.68421054,0.7368421,0.7631579,0.7368421,0.68421054,0.68421054,0.7105263,0.65789473,0.7105263,0.7894737,0.7894737,0.7368421,0.84210527,0.65789473,0.84210527,0.7105263,0.81578946,0.81578946,0.8815789,0.7894737,0.81578946,0.8684211,0.84210527,0.84210527,0.81578946,0.84210527,0.84210527,0.8947368,0.8684211,0.92105263,0.84210527,0.8947368,0.8947368,0.8684211,0.84210527,0.92105263,0.84210527,0.84210527,0.9736842,0.8947368,0.9736842,0.92105263,0.8947368,0.8947368,1.0,0.94736844,0.9736842,0.8947368,1.0,0.94736844,0.94736844,0.92105263,1.0,0.94736844,1.0,0.94736844,0.94736844,0.94736844,0.9736842,0.9736842,1.0
+"AutoGluon 1.4 (best, 4h)",0.23684211,0.31578946,0.4473684,0.47368422,0.4473684,0.36842105,0.5,0.4473684,,0.57894737,0.7105263,0.65789473,0.6052632,0.81578946,0.7631579,0.65789473,0.7894737,0.7894737,0.7368421,0.6315789,0.81578946,0.68421054,0.68421054,0.7631579,0.65789473,0.7894737,0.84210527,0.7894737,0.7105263,0.7105263,0.57894737,0.7631579,0.81578946,0.65789473,0.7894737,0.7894737,0.7631579,0.7894737,0.68421054,0.8947368,0.81578946,0.7894737,0.84210527,0.7631579,0.8947368,0.8684211,0.7631579,0.84210527,0.8947368,0.8947368,0.92105263,0.84210527,0.92105263,0.8684211,0.84210527,0.92105263,0.92105263,0.92105263,0.94736844,0.7894737,0.81578946,0.92105263,0.94736844,0.9736842,0.8684211,0.9736842,0.9736842,0.9736842,0.92105263,0.9736842,0.92105263,0.94736844,0.9736842,0.9736842,0.9736842,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.18421052,0.18421052,0.28947368,0.28947368,0.31578946,0.34210527,0.39473686,0.39473686,0.42105263,,0.65789473,0.6052632,0.5,0.7105263,0.68421054,0.57894737,0.7105263,0.7631579,0.68421054,0.57894737,0.7105263,0.6052632,0.65789473,0.7631579,0.68421054,0.65789473,0.7631579,0.7631579,0.6315789,0.6315789,0.57894737,0.7105263,0.68421054,0.6315789,0.7105263,0.7894737,0.7894737,0.7631579,0.7105263,0.81578946,0.7631579,0.8947368,0.7631579,0.7894737,0.81578946,0.81578946,0.68421054,0.7894737,0.8947368,0.8684211,0.81578946,0.81578946,0.8947368,0.7894737,0.81578946,0.92105263,0.8947368,0.8684211,0.8684211,0.81578946,0.81578946,0.94736844,0.94736844,0.92105263,0.84210527,0.9736842,0.9736842,0.94736844,0.94736844,0.9736842,0.94736844,0.8947368,0.9736842,0.9736842,0.9736842,0.9736842,0.9736842,1.0
+RealMLP (tuned),0.15789473,0.18421052,0.2631579,0.2631579,0.28947368,0.36842105,0.28947368,0.31578946,0.28947368,0.34210527,,0.47368422,0.43421054,0.5,0.5,0.47368422,0.6052632,0.5263158,0.57894737,0.47368422,0.6315789,0.57894737,0.65789473,0.6315789,0.6052632,0.55263156,0.65789473,0.68421054,0.57894737,0.57894737,0.55263156,0.68421054,0.57894737,0.5263158,0.65789473,0.7105263,0.7105263,0.7105263,0.6315789,0.7894737,0.68421054,0.81578946,0.65789473,0.65789473,0.7368421,0.7368421,0.68421054,0.68421054,0.7894737,0.81578946,0.81578946,0.7631579,0.7631579,0.84210527,0.84210527,0.84210527,0.8684211,0.84210527,0.8684211,0.81578946,0.84210527,0.82894737,0.8947368,0.94736844,0.84210527,0.8947368,0.94736844,0.92105263,0.92105263,0.94736844,0.94736844,0.8684211,0.94736844,0.9605263,0.9736842,0.94736844,0.9736842,1.0
+TabM (tuned + ensembled),0.078947365,0.18421052,0.21052632,0.2631579,0.31578946,0.31578946,0.2631579,0.31578946,0.34210527,0.39473686,0.5263158,,0.42105263,0.5263158,0.47368422,0.57894737,0.6315789,0.5263158,0.55263156,0.57894737,0.55263156,0.6052632,0.6315789,0.6052632,0.6052632,0.57894737,0.6052632,0.68421054,0.6052632,0.6315789,0.5,0.68421054,0.68421054,0.47368422,0.57894737,0.6315789,0.7894737,0.84210527,0.5263158,0.7105263,0.57894737,0.81578946,0.7105263,0.7105263,0.7368421,0.7894737,0.6315789,0.7631579,0.7894737,0.7368421,0.7894737,0.7105263,0.84210527,0.7894737,0.7105263,0.84210527,0.7894737,0.84210527,0.7894737,0.84210527,0.8684211,0.8684211,0.84210527,0.9736842,0.8684211,0.92105263,0.92105263,0.84210527,0.8684211,0.84210527,0.8684211,0.92105263,0.94736844,0.94736844,0.94736844,0.94736844,1.0,1.0
+LimiX (default),0.13157895,0.2631579,0.28947368,0.23684211,0.31578946,0.31578946,0.42105263,0.28947368,0.39473686,0.5,0.56578946,0.57894737,,0.56578946,0.5263158,0.4473684,0.6315789,0.57894737,0.6052632,0.4473684,0.65789473,0.5263158,0.6052632,0.57894737,0.6052632,0.6315789,0.6052632,0.68421054,0.55263156,0.5263158,0.65789473,0.6052632,0.65789473,0.6052632,0.6315789,0.68421054,0.68421054,0.65789473,0.7105263,0.68421054,0.7368421,0.7368421,0.68421054,0.65789473,0.7368421,0.7368421,0.65789473,0.65789473,0.65789473,0.7368421,0.7105263,0.7368421,0.7368421,0.7368421,0.7368421,0.7368421,0.7631579,0.7894737,0.7894737,0.75,0.7368421,0.82894737,0.81578946,0.81578946,0.81578946,0.81578946,0.8947368,0.92105263,0.8684211,0.94736844,0.92105263,0.84210527,0.8947368,0.8947368,0.8684211,0.8947368,0.8684211,0.9736842
+iLTM (tuned + ensembled),0.18421052,0.13157895,0.21052632,0.18421052,0.2631579,0.23684211,0.34210527,0.42105263,0.18421052,0.28947368,0.5,0.47368422,0.43421054,,0.5,0.4473684,0.5263158,0.4473684,0.5,0.42105263,0.5263158,0.46052632,0.6052632,0.5,0.5263158,0.65789473,0.55263156,0.6315789,0.5,0.57894737,0.5263158,0.5,0.65789473,0.5,0.65789473,0.68421054,0.6315789,0.6052632,0.6315789,0.7368421,0.7368421,0.7105263,0.7368421,0.7105263,0.65789473,0.7368421,0.7368421,0.68421054,0.81578946,0.7894737,0.81578946,0.81578946,0.81578946,0.8684211,0.7894737,0.84210527,0.8684211,0.81578946,0.8684211,0.7631579,0.7631579,0.84210527,0.8947368,0.94736844,0.84210527,0.94736844,0.94736844,0.94736844,0.94736844,0.94736844,0.92105263,0.92105263,0.92105263,0.94736844,0.94736844,0.94736844,0.94736844,1.0
+CatBoost (tuned),0.10526316,0.15789473,0.2631579,0.23684211,0.18421052,0.2631579,0.2631579,0.31578946,0.23684211,0.31578946,0.5,0.5263158,0.47368422,0.5,,0.5263158,0.4473684,0.4473684,0.55263156,0.5263158,0.57894737,0.5263158,0.5263158,0.65789473,0.57894737,0.5263158,0.5263158,0.65789473,0.55263156,0.57894737,0.5,0.57894737,0.6052632,0.55263156,0.6315789,0.7368421,0.65789473,0.6315789,0.6052632,0.7368421,0.7368421,0.7368421,0.7105263,0.7105263,0.7105263,0.7631579,0.6315789,0.7105263,0.7368421,0.81578946,0.81578946,0.7631579,0.7631579,0.81578946,0.7368421,0.81578946,0.84210527,0.81578946,0.8684211,0.7894737,0.81578946,0.81578946,0.84210527,0.92105263,0.81578946,0.92105263,0.8947368,0.92105263,0.8947368,0.8684211,0.8684211,0.8684211,0.9736842,0.9736842,0.94736844,0.9736842,0.9736842,1.0
+TabDPT-Turbo (default),0.078947365,0.31578946,0.18421052,0.21052632,0.34210527,0.21052632,0.34210527,0.34210527,0.34210527,0.42105263,0.5263158,0.42105263,0.55263156,0.55263156,0.47368422,,0.5,0.5,0.4473684,0.6315789,0.57894737,0.5,0.5,0.57894737,0.5,0.6052632,0.5,0.6052632,0.68421054,0.67105263,0.6315789,0.5263158,0.6052632,0.6315789,0.6052632,0.68421054,0.6052632,0.6052632,0.68421054,0.65789473,0.68421054,0.6315789,0.6315789,0.65789473,0.65789473,0.7105263,0.65789473,0.81578946,0.7105263,0.7368421,0.7631579,0.7631579,0.68421054,0.7631579,0.7894737,0.81578946,0.7894737,0.8684211,0.81578946,0.7631579,0.7631579,0.92105263,0.81578946,0.84210527,0.84210527,0.8684211,0.94736844,0.92105263,0.92105263,0.94736844,0.8947368,0.94736844,0.84210527,0.84210527,0.94736844,0.8947368,0.9736842,1.0
+LightGBM (tuned + ensembled),0.078947365,0.10526316,0.13157895,0.10526316,0.18421052,0.18421052,0.2631579,0.23684211,0.21052632,0.28947368,0.39473686,0.36842105,0.36842105,0.47368422,0.55263156,0.5,,0.57894737,0.47368422,0.47368422,0.47368422,0.5263158,0.5,0.65789473,0.5263158,0.57894737,0.7105263,0.6315789,0.55263156,0.57894737,0.47368422,0.55263156,0.65789473,0.47368422,0.6315789,0.65789473,0.68421054,0.65789473,0.55263156,0.7368421,0.7105263,0.68421054,0.7368421,0.65789473,0.7894737,0.81578946,0.65789473,0.7105263,0.7631579,0.8684211,0.7894737,0.81578946,0.81578946,0.81578946,0.7631579,0.94736844,0.8947368,0.81578946,0.92105263,0.81578946,0.81578946,0.84210527,0.8684211,0.8947368,0.84210527,0.8947368,0.8947368,0.9736842,0.92105263,0.8947368,0.84210527,0.8684211,0.9736842,0.9736842,0.8947368,0.94736844,1.0,1.0
+CatBoost (tuned + ensembled),0.078947365,0.10526316,0.21052632,0.18421052,0.13157895,0.2631579,0.23684211,0.31578946,0.21052632,0.23684211,0.47368422,0.47368422,0.42105263,0.55263156,0.55263156,0.5,0.42105263,,0.5,0.5,0.5263158,0.57894737,0.5131579,0.6315789,0.57894737,0.55263156,0.5263158,0.6052632,0.55263156,0.55263156,0.5,0.5263158,0.6052632,0.5263158,0.6315789,0.7368421,0.6052632,0.55263156,0.6315789,0.7105263,0.7368421,0.7368421,0.81578946,0.7368421,0.68421054,0.7631579,0.6052632,0.68421054,0.7368421,0.81578946,0.81578946,0.81578946,0.7631579,0.84210527,0.7894737,0.81578946,0.84210527,0.84210527,0.8947368,0.81578946,0.81578946,0.7894737,0.8947368,0.92105263,0.81578946,0.8947368,0.8947368,0.94736844,0.8684211,0.8684211,0.8947368,0.84210527,0.9736842,0.9736842,0.94736844,0.9736842,0.94736844,1.0
+TabM (tuned),0.05263158,0.21052632,0.13157895,0.21052632,0.23684211,0.28947368,0.21052632,0.2631579,0.2631579,0.31578946,0.42105263,0.4473684,0.39473686,0.5,0.4473684,0.55263156,0.5263158,0.5,,0.55263156,0.5263158,0.6052632,0.6052632,0.57894737,0.55263156,0.6447368,0.5263158,0.6052632,0.6052632,0.6052632,0.5,0.6315789,0.6315789,0.4473684,0.5263158,0.6052632,0.65789473,0.68421054,0.5263158,0.6315789,0.57894737,0.65789473,0.6315789,0.65789473,0.7105263,0.68421054,0.57894737,0.7368421,0.68421054,0.7368421,0.7631579,0.68421054,0.84210527,0.7368421,0.68421054,0.7894737,0.7105263,0.81578946,0.7894737,0.7894737,0.81578946,0.84210527,0.8947368,0.9736842,0.84210527,0.92105263,0.8684211,0.8684211,0.8684211,0.81578946,0.84210527,0.92105263,0.8947368,0.8947368,0.94736844,0.94736844,0.9736842,0.9736842
+TabDPT (tuned + ensembled),0.078947365,0.18421052,0.21052632,0.2631579,0.31578946,0.18421052,0.34210527,0.23684211,0.36842105,0.42105263,0.5263158,0.42105263,0.55263156,0.57894737,0.47368422,0.36842105,0.5263158,0.5,0.4473684,,0.5263158,0.4473684,0.47368422,0.5263158,0.47368422,0.5263158,0.5263158,0.5263158,0.7105263,0.5263158,0.6052632,0.55263156,0.5263158,0.6052632,0.57894737,0.65789473,0.57894737,0.57894737,0.68421054,0.65789473,0.65789473,0.6315789,0.5263158,0.65789473,0.65789473,0.7105263,0.6315789,0.7631579,0.65789473,0.68421054,0.68421054,0.7105263,0.81578946,0.7105263,0.81578946,0.7631579,0.7368421,0.7894737,0.7368421,0.7894737,0.7631579,0.94736844,0.7894737,0.81578946,0.84210527,0.8684211,0.94736844,0.94736844,0.8947368,0.9736842,0.94736844,0.92105263,0.8684211,0.8684211,0.8947368,0.8947368,0.94736844,1.0
+XGBoost (tuned + ensembled),0.10526316,0.078947365,0.15789473,0.23684211,0.15789473,0.23684211,0.23684211,0.2631579,0.18421052,0.28947368,0.36842105,0.4473684,0.34210527,0.47368422,0.42105263,0.42105263,0.5263158,0.47368422,0.47368422,0.47368422,,0.5263158,0.47368422,0.55263156,0.5,0.55263156,0.57894737,0.68421054,0.5,0.5263158,0.39473686,0.47368422,0.6315789,0.5,0.55263156,0.6052632,0.6315789,0.6052632,0.5263158,0.68421054,0.68421054,0.7105263,0.7105263,0.6315789,0.68421054,0.7105263,0.6315789,0.68421054,0.7105263,0.8684211,0.69736844,0.81578946,0.7894737,0.84210527,0.7894737,0.9736842,0.84210527,0.7631579,0.8947368,0.84210527,0.84210527,0.84210527,0.7894737,0.8947368,0.84210527,0.84210527,0.8684211,0.94736844,0.8684211,0.92105263,0.84210527,0.84210527,0.92105263,0.94736844,0.8684211,0.94736844,0.92105263,1.0
+TabICL (default),0.078947365,0.18421052,0.10526316,0.13157895,0.2631579,0.21052632,0.21052632,0.31578946,0.31578946,0.39473686,0.42105263,0.39473686,0.47368422,0.5394737,0.47368422,0.5,0.47368422,0.42105263,0.39473686,0.55263156,0.47368422,,0.5263158,0.5263158,0.4473684,0.47368422,0.5263158,0.5263158,0.57894737,0.6052632,0.5394737,0.55263156,0.5263158,0.5263158,0.55263156,0.6315789,0.55263156,0.55263156,0.6447368,0.7105263,0.6315789,0.65789473,0.65789473,0.65789473,0.6052632,0.65789473,0.65789473,0.7631579,0.7105263,0.7368421,0.7105263,0.7631579,0.7763158,0.7631579,0.84210527,0.84210527,0.7631579,0.7894737,0.8552632,0.7631579,0.7894737,0.8684211,0.8684211,0.8684211,0.81578946,0.92105263,0.8684211,0.92105263,0.8947368,0.8947368,0.84210527,0.8947368,0.8947368,0.8684211,0.94736844,0.9736842,0.9736842,1.0
+ModernNCA (tuned),0.10526316,0.10526316,0.15789473,0.18421052,0.15789473,0.21052632,0.18421052,0.31578946,0.31578946,0.34210527,0.34210527,0.36842105,0.39473686,0.39473686,0.47368422,0.5,0.5,0.4868421,0.39473686,0.5263158,0.5263158,0.47368422,,0.57894737,0.43421054,0.5263158,0.55263156,0.5,0.57894737,0.55263156,0.47368422,0.57894737,0.55263156,0.47368422,0.57894737,0.57894737,0.5263158,0.6052632,0.5263158,0.65789473,0.68421054,0.65789473,0.65789473,0.68421054,0.65789473,0.6315789,0.65789473,0.68421054,0.67105263,0.7105263,0.7631579,0.7105263,0.84210527,0.7368421,0.7368421,0.7894737,0.7631579,0.7894737,0.7894737,0.7894737,0.7894737,0.84210527,0.84210527,0.84210527,0.8947368,0.8684211,0.8684211,0.92105263,0.92105263,0.8684211,0.8947368,0.8684211,0.8947368,0.92105263,0.8684211,0.92105263,0.92105263,1.0
+CatBoost (default),0.13157895,0.078947365,0.15789473,0.13157895,0.13157895,0.21052632,0.13157895,0.28947368,0.23684211,0.23684211,0.36842105,0.39473686,0.42105263,0.5,0.34210527,0.42105263,0.34210527,0.36842105,0.42105263,0.47368422,0.4473684,0.47368422,0.42105263,,0.47368422,0.47368422,0.4473684,0.47368422,0.5,0.55263156,0.5,0.42105263,0.57894737,0.42105263,0.65789473,0.65789473,0.57894737,0.55263156,0.57894737,0.7105263,0.7105263,0.7631579,0.7368421,0.65789473,0.6315789,0.7105263,0.6052632,0.7105263,0.7631579,0.7368421,0.7894737,0.7368421,0.7105263,0.81578946,0.68421054,0.7631579,0.81578946,0.81578946,0.84210527,0.81578946,0.81578946,0.81578946,0.8684211,0.94736844,0.84210527,0.92105263,0.92105263,0.92105263,0.92105263,0.84210527,0.8947368,0.8684211,0.94736844,0.9736842,0.94736844,0.9736842,0.9736842,1.0
+ModernNCA (tuned + ensembled),0.13157895,0.078947365,0.078947365,0.10526316,0.23684211,0.15789473,0.2631579,0.34210527,0.34210527,0.31578946,0.39473686,0.39473686,0.39473686,0.47368422,0.42105263,0.5,0.47368422,0.42105263,0.4473684,0.5263158,0.5,0.55263156,0.56578946,0.5263158,,0.5263158,0.6052632,0.5263158,0.55263156,0.5,0.42105263,0.57894737,0.55263156,0.39473686,0.55263156,0.5263158,0.6052632,0.5263158,0.5,0.6052632,0.65789473,0.7105263,0.6052632,0.57894737,0.6052632,0.65789473,0.6052632,0.68421054,0.7368421,0.7368421,0.7631579,0.7368421,0.81578946,0.7368421,0.7105263,0.84210527,0.7631579,0.81578946,0.7105263,0.7105263,0.7368421,0.81578946,0.81578946,0.81578946,0.84210527,0.81578946,0.84210527,0.8684211,0.8552632,0.84210527,0.8947368,0.81578946,0.81578946,0.84210527,0.8947368,0.84210527,0.8684211,0.9736842
+ChimeraBoost (tuned + ensembled),0.10526316,0.078947365,0.13157895,0.18421052,0.21052632,0.18421052,0.31578946,0.28947368,0.21052632,0.34210527,0.4473684,0.42105263,0.36842105,0.34210527,0.47368422,0.39473686,0.42105263,0.4473684,0.35526314,0.47368422,0.4473684,0.5263158,0.47368422,0.5263158,0.47368422,,0.47368422,0.47368422,0.47368422,0.55263156,0.47368422,0.39473686,0.57894737,0.4473684,0.6052632,0.6052632,0.6315789,0.57894737,0.5,0.6315789,0.68421054,0.68421054,0.81578946,0.6315789,0.68421054,0.65789473,0.6052632,0.68421054,0.7105263,0.68421054,0.7105263,0.65789473,0.68421054,0.7631579,0.68421054,0.7631579,0.7368421,0.7105263,0.81578946,0.81578946,0.81578946,0.7894737,0.7894737,0.84210527,0.84210527,0.84210527,0.84210527,0.84210527,0.84210527,0.81578946,0.84210527,0.8684211,0.92105263,0.94736844,0.84210527,0.9736842,0.94736844,1.0
+LightGBM (tuned),0.05263158,0.078947365,0.05263158,0.02631579,0.10526316,0.15789473,0.13157895,0.21052632,0.15789473,0.23684211,0.34210527,0.39473686,0.39473686,0.4473684,0.47368422,0.5,0.28947368,0.47368422,0.47368422,0.47368422,0.42105263,0.47368422,0.4473684,0.55263156,0.39473686,0.5263158,,0.47368422,0.5,0.5,0.4473684,0.57894737,0.6052632,0.42105263,0.6052632,0.5263158,0.65789473,0.65789473,0.5263158,0.6315789,0.65789473,0.6315789,0.68421054,0.6315789,0.68421054,0.7105263,0.57894737,0.68421054,0.7105263,0.7631579,0.7631579,0.7105263,0.68421054,0.7631579,0.7368421,0.84210527,0.81578946,0.7894737,0.8947368,0.7894737,0.81578946,0.81578946,0.84210527,0.84210527,0.84210527,0.8684211,0.8684211,0.92105263,0.8684211,0.84210527,0.84210527,0.84210527,0.8684211,0.92105263,0.8947368,0.92105263,0.94736844,1.0
+XGBoost (tuned),0.15789473,0.078947365,0.18421052,0.15789473,0.15789473,0.23684211,0.21052632,0.21052632,0.21052632,0.23684211,0.31578946,0.31578946,0.31578946,0.36842105,0.34210527,0.39473686,0.36842105,0.39473686,0.39473686,0.47368422,0.31578946,0.47368422,0.5,0.5263158,0.47368422,0.5263158,0.5263158,,0.47368422,0.5263158,0.42105263,0.5,0.55263156,0.47368422,0.6315789,0.55263156,0.55263156,0.55263156,0.5263158,0.57894737,0.65789473,0.6315789,0.6315789,0.6052632,0.68421054,0.68421054,0.6315789,0.6315789,0.68421054,0.84210527,0.68421054,0.81578946,0.7368421,0.7894737,0.81578946,0.8947368,0.81578946,0.7894737,0.84210527,0.84210527,0.84210527,0.81578946,0.7631579,0.8684211,0.84210527,0.84210527,0.8684211,0.92105263,0.8947368,0.8947368,0.81578946,0.84210527,0.8947368,0.92105263,0.8684211,0.94736844,0.92105263,1.0
+TabDPT (tuned),0.05263158,0.18421052,0.10526316,0.23684211,0.23684211,0.18421052,0.23684211,0.2631579,0.28947368,0.36842105,0.42105263,0.39473686,0.4473684,0.5,0.4473684,0.31578946,0.4473684,0.4473684,0.39473686,0.28947368,0.5,0.42105263,0.42105263,0.5,0.4473684,0.5263158,0.5,0.5263158,,0.57894737,0.55263156,0.4473684,0.55263156,0.55263156,0.5263158,0.6052632,0.55263156,0.5263158,0.6315789,0.5263158,0.57894737,0.55263156,0.6052632,0.55263156,0.55263156,0.6315789,0.57894737,0.80263156,0.55263156,0.6052632,0.65789473,0.6315789,0.7105263,0.57894737,0.7631579,0.6315789,0.7631579,0.7631579,0.7105263,0.7631579,0.7631579,0.8947368,0.7894737,0.81578946,0.81578946,0.8684211,0.94736844,0.84210527,0.84210527,0.92105263,0.8947368,0.8684211,0.81578946,0.81578946,0.92105263,0.8684211,0.94736844,0.9736842
+TabSwift (default),0.13157895,0.21052632,0.13157895,0.21052632,0.18421052,0.18421052,0.2631579,0.15789473,0.28947368,0.36842105,0.42105263,0.36842105,0.47368422,0.42105263,0.42105263,0.32894737,0.42105263,0.4473684,0.39473686,0.47368422,0.47368422,0.39473686,0.4473684,0.4473684,0.5,0.4473684,0.5,0.47368422,0.42105263,,0.5,0.5263158,0.5,0.5,0.6052632,0.5263158,0.57894737,0.5263158,0.55263156,0.5,0.6052632,0.57894737,0.68421054,0.65789473,0.57894737,0.65789473,0.6052632,0.65789473,0.7631579,0.6052632,0.6052632,0.65789473,0.65789473,0.6315789,0.7105263,0.7368421,0.7368421,0.7368421,0.65789473,0.7631579,0.7631579,0.8947368,0.7894737,0.7894737,0.84210527,0.84210527,0.8684211,0.81578946,0.8684211,0.8684211,0.84210527,0.9736842,0.84210527,0.8684211,0.8947368,0.8684211,0.9736842,1.0
+TabPFNv2 (tuned + ensembled),0.02631579,0.23684211,0.21052632,0.18421052,0.10526316,0.23684211,0.27631578,0.34210527,0.42105263,0.42105263,0.4473684,0.5,0.34210527,0.47368422,0.5,0.36842105,0.5263158,0.5,0.5,0.39473686,0.6052632,0.46052632,0.5263158,0.5,0.57894737,0.5263158,0.55263156,0.57894737,0.4473684,0.5,,0.6052632,0.57894737,0.5263158,0.55263156,0.5,0.6052632,0.57894737,0.65789473,0.57894737,0.6315789,0.6315789,0.6052632,0.5263158,0.6315789,0.6315789,0.6052632,0.55263156,0.6052632,0.68421054,0.68421054,0.57894737,0.6052632,0.65789473,0.65789473,0.7368421,0.68421054,0.68421054,0.67105263,0.6315789,0.6315789,0.7631579,0.7368421,0.7368421,0.68421054,0.81578946,0.81578946,0.7894737,0.7368421,0.84210527,0.8684211,0.7631579,0.84210527,0.84210527,0.8947368,0.84210527,0.81578946,1.0
+BetaTabPFN (default),0.05263158,0.15789473,0.078947365,0.15789473,0.13157895,0.10526316,0.15789473,0.15789473,0.23684211,0.28947368,0.31578946,0.31578946,0.39473686,0.5,0.42105263,0.47368422,0.4473684,0.47368422,0.36842105,0.4473684,0.5263158,0.4473684,0.42105263,0.57894737,0.42105263,0.6052632,0.42105263,0.5,0.55263156,0.47368422,0.39473686,,0.5263158,0.4473684,0.5263158,0.5263158,0.55263156,0.47368422,0.5,0.57894737,0.55263156,0.6052632,0.6315789,0.6052632,0.7368421,0.6052632,0.55263156,0.65789473,0.6315789,0.68421054,0.7105263,0.65789473,0.65789473,0.68421054,0.65789473,0.7368421,0.7631579,0.7368421,0.7105263,0.7631579,0.7631579,0.7894737,0.81578946,0.84210527,0.8684211,0.84210527,0.8684211,0.7894737,0.8947368,0.7631579,0.81578946,0.8947368,0.8947368,0.8947368,0.8684211,0.94736844,0.9736842,1.0
+ChimeraBoost (tuned),0.13157895,0.078947365,0.15789473,0.18421052,0.15789473,0.21052632,0.21052632,0.28947368,0.18421052,0.31578946,0.42105263,0.31578946,0.34210527,0.34210527,0.39473686,0.39473686,0.34210527,0.39473686,0.36842105,0.47368422,0.36842105,0.47368422,0.4473684,0.42105263,0.4473684,0.42105263,0.39473686,0.4473684,0.4473684,0.5,0.42105263,0.47368422,,0.39473686,0.5263158,0.47368422,0.6315789,0.5,0.47368422,0.57894737,0.6315789,0.68421054,0.68421054,0.6315789,0.65789473,0.68421054,0.6052632,0.65789473,0.7105263,0.7105263,0.65789473,0.6315789,0.6315789,0.7105263,0.68421054,0.7894737,0.7368421,0.7368421,0.81578946,0.7894737,0.7894737,0.7894737,0.7631579,0.84210527,0.84210527,0.81578946,0.81578946,0.7894737,0.81578946,0.84210527,0.84210527,0.8947368,0.8684211,0.8684211,0.84210527,0.94736844,0.92105263,1.0
+Mitra (default),0.0,0.21052632,0.13157895,0.13157895,0.13157895,0.21052632,0.13157895,0.18421052,0.34210527,0.36842105,0.47368422,0.5263158,0.39473686,0.5,0.4473684,0.36842105,0.5263158,0.47368422,0.55263156,0.39473686,0.5,0.47368422,0.5263158,0.57894737,0.6052632,0.55263156,0.57894737,0.5263158,0.4473684,0.5,0.47368422,0.55263156,0.6052632,,0.5263158,0.5263158,0.55263156,0.65789473,0.6052632,0.5263158,0.55263156,0.57894737,0.57894737,0.5263158,0.57894737,0.6052632,0.6052632,0.6052632,0.6052632,0.6315789,0.7105263,0.55263156,0.65789473,0.57894737,0.6315789,0.65789473,0.65789473,0.6315789,0.6447368,0.65789473,0.65789473,0.7763158,0.7368421,0.7105263,0.7105263,0.7631579,0.7894737,0.7894737,0.7368421,0.7894737,0.92105263,0.7894737,0.84210527,0.84210527,0.8947368,0.8684211,0.84210527,1.0
+xRFM (tuned + ensembled),0.078947365,0.078947365,0.15789473,0.15789473,0.18421052,0.21052632,0.23684211,0.18421052,0.21052632,0.28947368,0.34210527,0.42105263,0.36842105,0.34210527,0.36842105,0.39473686,0.36842105,0.36842105,0.47368422,0.42105263,0.4473684,0.4473684,0.42105263,0.34210527,0.4473684,0.39473686,0.39473686,0.36842105,0.47368422,0.39473686,0.4473684,0.47368422,0.47368422,0.47368422,,0.42105263,0.5,0.55263156,0.5263158,0.5263158,0.6315789,0.6052632,0.4473684,0.55263156,0.5263158,0.57894737,0.57894737,0.57894737,0.65789473,0.6052632,0.6052632,0.68421054,0.7631579,0.7368421,0.7368421,0.65789473,0.68421054,0.7368421,0.7631579,0.7368421,0.7368421,0.8684211,0.7631579,0.84210527,0.7894737,0.81578946,0.8947368,0.8947368,0.92105263,0.92105263,0.92105263,0.8947368,0.94736844,0.9736842,0.8684211,0.94736844,0.9736842,0.9736842
+iLTM (tuned),0.078947365,0.10526316,0.10526316,0.13157895,0.18421052,0.23684211,0.15789473,0.118421055,0.21052632,0.21052632,0.28947368,0.36842105,0.31578946,0.31578946,0.2631579,0.31578946,0.34210527,0.2631579,0.39473686,0.34210527,0.39473686,0.36842105,0.42105263,0.34210527,0.47368422,0.39473686,0.47368422,0.4473684,0.39473686,0.47368422,0.5,0.47368422,0.5263158,0.47368422,0.57894737,,0.5,0.5,0.5263158,0.57894737,0.6315789,0.57894737,0.47368422,0.65789473,0.6052632,0.7105263,0.68421054,0.6315789,0.7368421,0.55263156,0.65789473,0.7368421,0.7105263,0.7894737,0.81578946,0.7105263,0.69736844,0.7631579,0.7894737,0.7631579,0.7631579,0.7368421,0.7631579,0.8947368,0.84210527,0.8684211,0.8684211,0.8684211,0.92105263,0.8947368,0.8947368,0.8684211,0.84210527,0.8947368,0.8947368,0.8947368,0.9736842,1.0
+EBM (tuned + ensembled),0.078947365,0.15789473,0.18421052,0.13157895,0.15789473,0.23684211,0.15789473,0.21052632,0.23684211,0.21052632,0.28947368,0.21052632,0.31578946,0.36842105,0.34210527,0.39473686,0.31578946,0.39473686,0.34210527,0.42105263,0.36842105,0.4473684,0.47368422,0.42105263,0.39473686,0.36842105,0.34210527,0.4473684,0.4473684,0.42105263,0.39473686,0.4473684,0.36842105,0.4473684,0.5,0.5,,0.5,0.47368422,0.57894737,0.57894737,0.6315789,0.5,0.65789473,0.7631579,0.7894737,0.55263156,0.68421054,0.7105263,0.65789473,0.68421054,0.7368421,0.7894737,0.7105263,0.68421054,0.65789473,0.68421054,0.7631579,0.7368421,0.7631579,0.7894737,0.7631579,0.7631579,0.8684211,0.84210527,0.7894737,0.8684211,0.81578946,0.8947368,0.84210527,0.84210527,0.92105263,0.94736844,0.94736844,0.8684211,0.9736842,0.94736844,1.0
+TabM (default),0.078947365,0.13157895,0.15789473,0.15789473,0.15789473,0.15789473,0.15789473,0.18421052,0.21052632,0.23684211,0.28947368,0.15789473,0.34210527,0.39473686,0.36842105,0.39473686,0.34210527,0.4473684,0.31578946,0.42105263,0.39473686,0.4473684,0.39473686,0.4473684,0.47368422,0.42105263,0.34210527,0.4473684,0.47368422,0.47368422,0.42105263,0.5263158,0.5,0.34210527,0.4473684,0.5,0.5,,0.42105263,0.57894737,0.55263156,0.55263156,0.55263156,0.55263156,0.57894737,0.5394737,0.5263158,0.6315789,0.7105263,0.65789473,0.68421054,0.6052632,0.7105263,0.7105263,0.65789473,0.68421054,0.65789473,0.81578946,0.7368421,0.7368421,0.7631579,0.7894737,0.84210527,0.92105263,0.84210527,0.84210527,0.81578946,0.7894737,0.81578946,0.81578946,0.7894737,0.8684211,0.8947368,0.8684211,0.8947368,0.8947368,0.9736842,0.94736844
+TabPFNv2 (tuned),0.02631579,0.13157895,0.078947365,0.13157895,0.05263158,0.18421052,0.05263158,0.13157895,0.31578946,0.28947368,0.36842105,0.47368422,0.28947368,0.36842105,0.39473686,0.31578946,0.4473684,0.36842105,0.47368422,0.31578946,0.47368422,0.35526314,0.47368422,0.42105263,0.5,0.5,0.47368422,0.47368422,0.36842105,0.4473684,0.34210527,0.5,0.5263158,0.39473686,0.47368422,0.47368422,0.5263158,0.57894737,,0.55263156,0.55263156,0.55263156,0.57894737,0.4868421,0.5263158,0.6052632,0.5263158,0.5263158,0.55263156,0.6315789,0.65789473,0.5263158,0.55263156,0.55263156,0.55263156,0.65789473,0.65789473,0.6315789,0.6184211,0.6315789,0.6315789,0.7368421,0.7105263,0.68421054,0.68421054,0.7631579,0.7368421,0.7368421,0.7368421,0.84210527,0.84210527,0.7894737,0.84210527,0.84210527,0.8684211,0.84210527,0.84210527,1.0
+TorchMLP (tuned + ensembled),0.05263158,0.078947365,0.13157895,0.10526316,0.18421052,0.13157895,0.18421052,0.15789473,0.10526316,0.18421052,0.21052632,0.28947368,0.31578946,0.2631579,0.2631579,0.34210527,0.2631579,0.28947368,0.36842105,0.34210527,0.31578946,0.28947368,0.34210527,0.28947368,0.39473686,0.36842105,0.36842105,0.42105263,0.47368422,0.5,0.42105263,0.42105263,0.42105263,0.47368422,0.47368422,0.42105263,0.42105263,0.42105263,0.4473684,,0.55263156,0.4473684,0.4473684,0.5,0.47368422,0.47368422,0.55263156,0.6315789,0.6052632,0.65789473,0.7368421,0.65789473,0.6052632,0.7105263,0.65789473,0.68421054,0.7105263,0.7368421,0.7631579,0.7105263,0.7368421,0.7631579,0.7631579,0.94736844,0.7631579,0.84210527,0.8684211,0.7894737,0.8684211,0.81578946,0.81578946,0.92105263,0.8947368,0.92105263,0.8947368,0.9736842,0.94736844,0.9736842
+xRFM (tuned),0.078947365,0.13157895,0.10526316,0.15789473,0.13157895,0.15789473,0.15789473,0.15789473,0.18421052,0.23684211,0.31578946,0.42105263,0.2631579,0.2631579,0.2631579,0.31578946,0.28947368,0.2631579,0.42105263,0.34210527,0.31578946,0.36842105,0.31578946,0.28947368,0.34210527,0.31578946,0.34210527,0.34210527,0.42105263,0.39473686,0.36842105,0.4473684,0.36842105,0.4473684,0.36842105,0.36842105,0.42105263,0.4473684,0.4473684,0.4473684,,0.5,0.42105263,0.57894737,0.5263158,0.4473684,0.5,0.5263158,0.57894737,0.5,0.6315789,0.7105263,0.57894737,0.68421054,0.7368421,0.6315789,0.68421054,0.65789473,0.7368421,0.7105263,0.7368421,0.7894737,0.68421054,0.8684211,0.7894737,0.7631579,0.84210527,0.81578946,0.92105263,0.92105263,0.9736842,0.84210527,0.8947368,0.8684211,0.8947368,0.8947368,0.9736842,1.0
+RealMLP (default),0.10526316,0.13157895,0.078947365,0.10526316,0.15789473,0.18421052,0.15789473,0.18421052,0.21052632,0.10526316,0.18421052,0.18421052,0.2631579,0.28947368,0.2631579,0.36842105,0.31578946,0.2631579,0.34210527,0.36842105,0.28947368,0.34210527,0.34210527,0.23684211,0.28947368,0.31578946,0.36842105,0.36842105,0.4473684,0.42105263,0.36842105,0.39473686,0.31578946,0.42105263,0.39473686,0.42105263,0.36842105,0.4473684,0.4473684,0.55263156,0.5,,0.4473684,0.5,0.55263156,0.5,0.55263156,0.6315789,0.57894737,0.6315789,0.65789473,0.6052632,0.68421054,0.65789473,0.68421054,0.6052632,0.65789473,0.7368421,0.68421054,0.7105263,0.7631579,0.7105263,0.7631579,0.8947368,0.84210527,0.7894737,0.92105263,0.7894737,0.8947368,0.84210527,0.81578946,0.94736844,0.8947368,0.92105263,0.92105263,0.9736842,0.94736844,1.0
+ChimeraBoost (default),0.078947365,0.05263158,0.10526316,0.15789473,0.18421052,0.18421052,0.21052632,0.15789473,0.15789473,0.23684211,0.34210527,0.28947368,0.31578946,0.2631579,0.28947368,0.36842105,0.2631579,0.18421052,0.36842105,0.47368422,0.28947368,0.34210527,0.34210527,0.2631579,0.39473686,0.18421052,0.31578946,0.36842105,0.39473686,0.31578946,0.39473686,0.36842105,0.31578946,0.42105263,0.55263156,0.5263158,0.5,0.4473684,0.42105263,0.55263156,0.57894737,0.55263156,,0.6052632,0.57894737,0.55263156,0.55263156,0.55263156,0.6315789,0.57894737,0.55263156,0.57894737,0.6315789,0.68421054,0.57894737,0.6052632,0.68421054,0.68421054,0.7368421,0.7631579,0.81578946,0.7368421,0.7105263,0.7894737,0.84210527,0.7631579,0.7894737,0.7368421,0.81578946,0.7894737,0.7894737,0.84210527,0.8684211,0.8684211,0.7894737,0.9736842,0.94736844,1.0
+SAP-RPT-OSS (default),0.10526316,0.18421052,0.21052632,0.18421052,0.13157895,0.23684211,0.15789473,0.15789473,0.23684211,0.21052632,0.34210527,0.28947368,0.34210527,0.28947368,0.28947368,0.34210527,0.34210527,0.2631579,0.34210527,0.34210527,0.36842105,0.34210527,0.31578946,0.34210527,0.42105263,0.36842105,0.36842105,0.39473686,0.4473684,0.34210527,0.47368422,0.39473686,0.36842105,0.47368422,0.4473684,0.34210527,0.34210527,0.4473684,0.5131579,0.5,0.42105263,0.5,0.39473686,,0.42105263,0.5,0.57894737,0.55263156,0.6052632,0.47368422,0.57894737,0.6052632,0.57894737,0.55263156,0.68421054,0.6052632,0.6052632,0.7105263,0.57894737,0.65789473,0.65789473,0.68421054,0.7894737,0.7368421,0.84210527,0.81578946,0.84210527,0.7631579,0.7894737,0.81578946,0.81578946,0.8684211,0.8684211,0.84210527,0.84210527,0.8684211,0.8947368,1.0
+EBM (tuned),0.078947365,0.13157895,0.10526316,0.078947365,0.15789473,0.13157895,0.15789473,0.10526316,0.10526316,0.18421052,0.2631579,0.2631579,0.2631579,0.34210527,0.28947368,0.34210527,0.21052632,0.31578946,0.28947368,0.34210527,0.31578946,0.39473686,0.34210527,0.36842105,0.39473686,0.31578946,0.31578946,0.31578946,0.4473684,0.42105263,0.36842105,0.2631579,0.34210527,0.42105263,0.47368422,0.39473686,0.23684211,0.42105263,0.47368422,0.5263158,0.47368422,0.4473684,0.42105263,0.57894737,,0.5,0.55263156,0.6052632,0.57894737,0.5263158,0.55263156,0.6052632,0.65789473,0.6315789,0.65789473,0.57894737,0.68421054,0.7105263,0.65789473,0.7631579,0.7631579,0.7368421,0.7631579,0.81578946,0.84210527,0.7631579,0.81578946,0.7894737,0.8947368,0.81578946,0.84210527,0.8947368,0.84210527,0.8947368,0.84210527,0.8947368,0.94736844,1.0
+EBM (default),0.078947365,0.10526316,0.10526316,0.13157895,0.13157895,0.10526316,0.15789473,0.13157895,0.13157895,0.18421052,0.2631579,0.21052632,0.2631579,0.2631579,0.23684211,0.28947368,0.18421052,0.23684211,0.31578946,0.28947368,0.28947368,0.34210527,0.36842105,0.28947368,0.34210527,0.34210527,0.28947368,0.31578946,0.36842105,0.34210527,0.36842105,0.39473686,0.31578946,0.39473686,0.42105263,0.28947368,0.21052632,0.46052632,0.39473686,0.5263158,0.55263156,0.5,0.4473684,0.5,0.5,,0.5,0.55263156,0.6315789,0.47368422,0.6315789,0.65789473,0.68421054,0.65789473,0.68421054,0.57894737,0.68421054,0.7105263,0.7105263,0.7368421,0.7368421,0.7631579,0.81578946,0.84210527,0.81578946,0.84210527,0.84210527,0.7368421,0.8947368,0.81578946,0.7894737,0.8947368,0.8684211,0.8947368,0.8947368,0.94736844,0.94736844,1.0
+TabPFNv2 (default),0.05263158,0.21052632,0.21052632,0.23684211,0.13157895,0.13157895,0.18421052,0.078947365,0.23684211,0.31578946,0.31578946,0.36842105,0.34210527,0.2631579,0.36842105,0.34210527,0.34210527,0.39473686,0.42105263,0.36842105,0.36842105,0.34210527,0.34210527,0.39473686,0.39473686,0.39473686,0.42105263,0.36842105,0.42105263,0.39473686,0.39473686,0.4473684,0.39473686,0.39473686,0.42105263,0.31578946,0.4473684,0.47368422,0.47368422,0.4473684,0.5,0.4473684,0.4473684,0.42105263,0.4473684,0.5,,0.55263156,0.5263158,0.5,0.57894737,0.47368422,0.4473684,0.55263156,0.5263158,0.55263156,0.6052632,0.6052632,0.59210527,0.6315789,0.65789473,0.7105263,0.7105263,0.68421054,0.7105263,0.7368421,0.7631579,0.7105263,0.65789473,0.7368421,0.84210527,0.7894737,0.81578946,0.81578946,0.8684211,0.8684211,0.81578946,1.0
+TabDPT (default),0.02631579,0.15789473,0.10526316,0.18421052,0.18421052,0.13157895,0.18421052,0.15789473,0.15789473,0.21052632,0.31578946,0.23684211,0.34210527,0.31578946,0.28947368,0.18421052,0.28947368,0.31578946,0.2631579,0.23684211,0.31578946,0.23684211,0.31578946,0.28947368,0.31578946,0.31578946,0.31578946,0.36842105,0.19736843,0.34210527,0.4473684,0.34210527,0.34210527,0.39473686,0.42105263,0.36842105,0.31578946,0.36842105,0.47368422,0.36842105,0.47368422,0.36842105,0.4473684,0.4473684,0.39473686,0.4473684,0.4473684,,0.5263158,0.5,0.4473684,0.5,0.6052632,0.55263156,0.57894737,0.6052632,0.57894737,0.7105263,0.6052632,0.6315789,0.65789473,0.7631579,0.7105263,0.7105263,0.7105263,0.7894737,0.81578946,0.7105263,0.7631579,0.7894737,0.7894737,0.84210527,0.7105263,0.7368421,0.81578946,0.81578946,0.8684211,0.9736842
+FastaiMLP (tuned + ensembled),0.10526316,0.13157895,0.13157895,0.10526316,0.15789473,0.05263158,0.18421052,0.10526316,0.10526316,0.10526316,0.21052632,0.21052632,0.34210527,0.18421052,0.2631579,0.28947368,0.23684211,0.2631579,0.31578946,0.34210527,0.28947368,0.28947368,0.32894737,0.23684211,0.2631579,0.28947368,0.28947368,0.31578946,0.4473684,0.23684211,0.39473686,0.36842105,0.28947368,0.39473686,0.34210527,0.2631579,0.28947368,0.28947368,0.4473684,0.39473686,0.42105263,0.42105263,0.36842105,0.39473686,0.42105263,0.36842105,0.47368422,0.47368422,,0.4473684,0.55263156,0.57894737,0.6052632,0.55263156,0.57894737,0.5263158,0.6052632,0.7368421,0.6052632,0.6315789,0.6315789,0.68421054,0.7368421,0.7105263,0.7368421,0.7631579,0.9736842,0.7368421,0.81578946,0.7894737,0.81578946,0.8684211,0.81578946,0.84210527,0.8947368,0.8947368,0.8947368,0.9736842
+XGBoost (default),0.05263158,0.078947365,0.05263158,0.05263158,0.05263158,0.13157895,0.078947365,0.10526316,0.10526316,0.13157895,0.18421052,0.2631579,0.2631579,0.21052632,0.18421052,0.2631579,0.13157895,0.18421052,0.2631579,0.31578946,0.13157895,0.2631579,0.28947368,0.2631579,0.2631579,0.31578946,0.23684211,0.15789473,0.39473686,0.39473686,0.31578946,0.31578946,0.28947368,0.36842105,0.39473686,0.4473684,0.34210527,0.34210527,0.36842105,0.34210527,0.5,0.36842105,0.42105263,0.5263158,0.47368422,0.5263158,0.5,0.5,0.55263156,,0.5263158,0.6315789,0.6052632,0.6052632,0.6315789,0.6052632,0.7368421,0.6052632,0.6315789,0.68421054,0.68421054,0.68421054,0.68421054,0.7368421,0.7631579,0.7631579,0.7631579,0.7894737,0.7631579,0.7368421,0.7894737,0.7631579,0.81578946,0.84210527,0.84210527,0.84210527,0.8684211,0.94736844
+TorchMLP (tuned),0.02631579,0.10526316,0.05263158,0.05263158,0.13157895,0.10526316,0.10526316,0.13157895,0.078947365,0.18421052,0.18421052,0.21052632,0.28947368,0.18421052,0.18421052,0.23684211,0.21052632,0.18421052,0.23684211,0.31578946,0.3026316,0.28947368,0.23684211,0.21052632,0.23684211,0.28947368,0.23684211,0.31578946,0.34210527,0.39473686,0.31578946,0.28947368,0.34210527,0.28947368,0.39473686,0.34210527,0.31578946,0.31578946,0.34210527,0.2631579,0.36842105,0.34210527,0.4473684,0.42105263,0.4473684,0.36842105,0.42105263,0.55263156,0.4473684,0.47368422,,0.5394737,0.6315789,0.57894737,0.6052632,0.56578946,0.6315789,0.55263156,0.57894737,0.7368421,0.7105263,0.7105263,0.7631579,0.8947368,0.81578946,0.7894737,0.7894737,0.68421054,0.84210527,0.7631579,0.7894737,0.92105263,0.81578946,0.84210527,0.84210527,0.92105263,0.94736844,0.9736842
+ExtraTrees (tuned + ensembled),0.05263158,0.078947365,0.05263158,0.078947365,0.05263158,0.13157895,0.02631579,0.15789473,0.15789473,0.18421052,0.23684211,0.28947368,0.2631579,0.18421052,0.23684211,0.23684211,0.18421052,0.18421052,0.31578946,0.28947368,0.18421052,0.23684211,0.28947368,0.2631579,0.2631579,0.34210527,0.28947368,0.18421052,0.36842105,0.34210527,0.42105263,0.34210527,0.36842105,0.4473684,0.31578946,0.2631579,0.2631579,0.39473686,0.47368422,0.34210527,0.28947368,0.39473686,0.42105263,0.39473686,0.39473686,0.34210527,0.5263158,0.5,0.42105263,0.36842105,0.46052632,,0.5,0.57894737,0.69736844,0.5263158,0.55263156,0.5,0.68421054,0.65789473,0.68421054,0.7368421,0.65789473,0.6315789,0.7368421,0.7105263,0.7105263,0.8684211,0.8947368,0.92105263,0.7894737,0.81578946,0.7631579,0.84210527,0.7894737,0.84210527,0.92105263,1.0
+ModernNCA (default),0.05263158,0.05263158,0.05263158,0.05263158,0.13157895,0.078947365,0.10526316,0.078947365,0.078947365,0.10526316,0.23684211,0.15789473,0.2631579,0.18421052,0.23684211,0.31578946,0.18421052,0.23684211,0.15789473,0.18421052,0.21052632,0.2236842,0.15789473,0.28947368,0.18421052,0.31578946,0.31578946,0.2631579,0.28947368,0.34210527,0.39473686,0.34210527,0.36842105,0.34210527,0.23684211,0.28947368,0.21052632,0.28947368,0.4473684,0.39473686,0.42105263,0.31578946,0.36842105,0.42105263,0.34210527,0.31578946,0.55263156,0.39473686,0.39473686,0.39473686,0.36842105,0.5,,0.57894737,0.5,0.5,0.57894737,0.55263156,0.6315789,0.7368421,0.7368421,0.7105263,0.7105263,0.81578946,0.7894737,0.7368421,0.7894737,0.81578946,0.81578946,0.81578946,0.7105263,0.7631579,0.81578946,0.81578946,0.84210527,0.8684211,0.92105263,0.9736842
+RandomForest (tuned + ensembled),0.05263158,0.05263158,0.078947365,0.05263158,0.078947365,0.10526316,0.078947365,0.15789473,0.13157895,0.21052632,0.15789473,0.21052632,0.2631579,0.13157895,0.18421052,0.23684211,0.18421052,0.15789473,0.2631579,0.28947368,0.15789473,0.23684211,0.2631579,0.18421052,0.2631579,0.23684211,0.23684211,0.21052632,0.42105263,0.36842105,0.34210527,0.31578946,0.28947368,0.42105263,0.2631579,0.21052632,0.28947368,0.28947368,0.4473684,0.28947368,0.31578946,0.34210527,0.31578946,0.4473684,0.36842105,0.34210527,0.4473684,0.4473684,0.4473684,0.39473686,0.42105263,0.42105263,0.42105263,,0.57894737,0.5,0.5,0.55263156,0.7105263,0.65789473,0.68421054,0.6315789,0.6052632,0.7368421,0.7631579,0.7105263,0.7105263,0.8684211,0.7894737,0.84210527,0.7894737,0.7105263,0.7368421,0.7105263,0.7105263,0.81578946,0.92105263,0.9736842
+ExtraTrees (tuned),0.05263158,0.078947365,0.05263158,0.078947365,0.05263158,0.078947365,0.10526316,0.15789473,0.15789473,0.18421052,0.15789473,0.28947368,0.2631579,0.21052632,0.2631579,0.21052632,0.23684211,0.21052632,0.31578946,0.18421052,0.21052632,0.15789473,0.2631579,0.31578946,0.28947368,0.31578946,0.2631579,0.18421052,0.23684211,0.28947368,0.34210527,0.34210527,0.31578946,0.36842105,0.2631579,0.18421052,0.31578946,0.34210527,0.4473684,0.34210527,0.2631579,0.31578946,0.42105263,0.31578946,0.34210527,0.31578946,0.47368422,0.42105263,0.42105263,0.36842105,0.39473686,0.3026316,0.5,0.42105263,,0.4473684,0.47368422,0.47368422,0.57894737,0.6315789,0.6315789,0.7105263,0.65789473,0.57894737,0.7631579,0.68421054,0.6315789,0.81578946,0.7894737,0.81578946,0.7105263,0.7894737,0.68421054,0.68421054,0.7631579,0.84210527,0.9736842,0.9736842
+LightGBM (default),0.05263158,0.02631579,0.05263158,0.05263158,0.02631579,0.05263158,0.05263158,0.02631579,0.078947365,0.078947365,0.15789473,0.15789473,0.2631579,0.15789473,0.18421052,0.18421052,0.05263158,0.18421052,0.21052632,0.23684211,0.02631579,0.15789473,0.21052632,0.23684211,0.15789473,0.23684211,0.15789473,0.10526316,0.36842105,0.2631579,0.2631579,0.2631579,0.21052632,0.34210527,0.34210527,0.28947368,0.34210527,0.31578946,0.34210527,0.31578946,0.36842105,0.39473686,0.39473686,0.39473686,0.42105263,0.42105263,0.4473684,0.39473686,0.47368422,0.39473686,0.43421054,0.47368422,0.5,0.5,0.55263156,,0.6315789,0.55263156,0.5263158,0.65789473,0.68421054,0.6315789,0.68421054,0.68421054,0.7631579,0.7105263,0.7368421,0.7105263,0.7105263,0.7631579,0.7631579,0.7631579,0.7105263,0.7105263,0.7894737,0.8684211,0.92105263,0.94736844
+iLTM (default),0.078947365,0.05263158,0.02631579,0.05263158,0.10526316,0.15789473,0.13157895,0.10526316,0.078947365,0.10526316,0.13157895,0.21052632,0.23684211,0.13157895,0.15789473,0.21052632,0.10526316,0.15789473,0.28947368,0.2631579,0.15789473,0.23684211,0.23684211,0.18421052,0.23684211,0.2631579,0.18421052,0.18421052,0.23684211,0.2631579,0.31578946,0.23684211,0.2631579,0.34210527,0.31578946,0.3026316,0.31578946,0.34210527,0.34210527,0.28947368,0.31578946,0.34210527,0.31578946,0.39473686,0.31578946,0.31578946,0.39473686,0.42105263,0.39473686,0.2631579,0.36842105,0.4473684,0.42105263,0.5,0.5263158,0.36842105,,0.47368422,0.47368422,0.68421054,0.65789473,0.6052632,0.6052632,0.65789473,0.7631579,0.57894737,0.65789473,0.65789473,0.7894737,0.68421054,0.68421054,0.7105263,0.7105263,0.7105263,0.65789473,0.81578946,0.84210527,0.94736844
+FastaiMLP (tuned),0.078947365,0.078947365,0.05263158,0.10526316,0.078947365,0.078947365,0.10526316,0.02631579,0.078947365,0.13157895,0.15789473,0.15789473,0.21052632,0.18421052,0.18421052,0.13157895,0.18421052,0.15789473,0.18421052,0.21052632,0.23684211,0.21052632,0.21052632,0.18421052,0.18421052,0.28947368,0.21052632,0.21052632,0.23684211,0.2631579,0.31578946,0.2631579,0.2631579,0.36842105,0.2631579,0.23684211,0.23684211,0.18421052,0.36842105,0.2631579,0.34210527,0.2631579,0.31578946,0.28947368,0.28947368,0.28947368,0.39473686,0.28947368,0.2631579,0.39473686,0.4473684,0.5,0.4473684,0.4473684,0.5263158,0.4473684,0.5263158,,0.5,0.57894737,0.57894737,0.57894737,0.6315789,0.6052632,0.7105263,0.68421054,0.81578946,0.65789473,0.7105263,0.7368421,0.7368421,0.7894737,0.7105263,0.7105263,0.7631579,0.84210527,0.8684211,0.94736844
+RandomForest (tuned),0.05263158,0.05263158,0.02631579,0.078947365,0.05263158,0.10526316,0.078947365,0.078947365,0.05263158,0.13157895,0.13157895,0.21052632,0.21052632,0.13157895,0.13157895,0.18421052,0.078947365,0.10526316,0.21052632,0.2631579,0.10526316,0.14473684,0.21052632,0.15789473,0.28947368,0.18421052,0.10526316,0.15789473,0.28947368,0.34210527,0.32894737,0.28947368,0.18421052,0.35526314,0.23684211,0.21052632,0.2631579,0.2631579,0.38157895,0.23684211,0.2631579,0.31578946,0.2631579,0.42105263,0.34210527,0.28947368,0.40789473,0.39473686,0.39473686,0.36842105,0.42105263,0.31578946,0.36842105,0.28947368,0.42105263,0.47368422,0.5263158,0.5,,0.6315789,0.6315789,0.65789473,0.6052632,0.6315789,0.7368421,0.65789473,0.7631579,0.80263156,0.7894737,0.7894737,0.7894737,0.7631579,0.7105263,0.68421054,0.68421054,0.7894737,0.8947368,0.9736842
+TabSTAR (tuned),0.078947365,0.13157895,0.13157895,0.15789473,0.15789473,0.13157895,0.15789473,0.10526316,0.21052632,0.18421052,0.18421052,0.15789473,0.25,0.23684211,0.21052632,0.23684211,0.18421052,0.18421052,0.21052632,0.21052632,0.15789473,0.23684211,0.21052632,0.18421052,0.28947368,0.18421052,0.21052632,0.15789473,0.23684211,0.23684211,0.36842105,0.23684211,0.21052632,0.34210527,0.2631579,0.23684211,0.23684211,0.2631579,0.36842105,0.28947368,0.28947368,0.28947368,0.23684211,0.34210527,0.23684211,0.2631579,0.36842105,0.36842105,0.36842105,0.31578946,0.2631579,0.34210527,0.2631579,0.34210527,0.36842105,0.34210527,0.31578946,0.42105263,0.36842105,,0.6447368,0.5,0.6315789,0.47368422,0.94736844,0.6315789,0.6052632,0.65789473,0.6315789,0.65789473,0.65789473,0.68421054,0.65789473,0.68421054,0.65789473,0.7631579,0.84210527,0.92105263
+TabSTAR (tuned + ensembled),0.10526316,0.13157895,0.13157895,0.15789473,0.15789473,0.13157895,0.15789473,0.10526316,0.18421052,0.18421052,0.15789473,0.13157895,0.2631579,0.23684211,0.18421052,0.23684211,0.18421052,0.18421052,0.18421052,0.23684211,0.15789473,0.21052632,0.21052632,0.18421052,0.2631579,0.18421052,0.18421052,0.15789473,0.23684211,0.23684211,0.36842105,0.23684211,0.21052632,0.34210527,0.2631579,0.23684211,0.21052632,0.23684211,0.36842105,0.2631579,0.2631579,0.23684211,0.18421052,0.34210527,0.23684211,0.2631579,0.34210527,0.34210527,0.36842105,0.31578946,0.28947368,0.31578946,0.2631579,0.31578946,0.36842105,0.31578946,0.34210527,0.42105263,0.36842105,0.35526314,,0.5263158,0.6315789,0.5,0.9736842,0.6315789,0.57894737,0.6315789,0.6052632,0.6315789,0.6315789,0.65789473,0.65789473,0.65789473,0.65789473,0.7631579,0.84210527,0.92105263
+OrionMSP (default),0.02631579,0.0,0.0,0.02631579,0.02631579,0.078947365,0.0,0.0,0.078947365,0.05263158,0.17105263,0.13157895,0.17105263,0.15789473,0.18421052,0.078947365,0.15789473,0.21052632,0.15789473,0.05263158,0.15789473,0.13157895,0.15789473,0.18421052,0.18421052,0.21052632,0.18421052,0.18421052,0.10526316,0.10526316,0.23684211,0.21052632,0.21052632,0.2236842,0.13157895,0.2631579,0.23684211,0.21052632,0.2631579,0.23684211,0.21052632,0.28947368,0.2631579,0.31578946,0.2631579,0.23684211,0.28947368,0.23684211,0.31578946,0.31578946,0.28947368,0.2631579,0.28947368,0.36842105,0.28947368,0.36842105,0.39473686,0.42105263,0.34210527,0.5,0.47368422,,0.55263156,0.47368422,0.68421054,0.55263156,0.5263158,0.6052632,0.6052632,0.68421054,0.6315789,0.7105263,0.68421054,0.68421054,0.68421054,0.7368421,0.7368421,0.9736842
+PerpetualBooster (tuned + ensembled),0.0,0.078947365,0.02631579,0.02631579,0.05263158,0.05263158,0.05263158,0.05263158,0.05263158,0.05263158,0.10526316,0.15789473,0.18421052,0.10526316,0.15789473,0.18421052,0.13157895,0.10526316,0.10526316,0.21052632,0.21052632,0.13157895,0.15789473,0.13157895,0.18421052,0.21052632,0.15789473,0.23684211,0.21052632,0.21052632,0.2631579,0.18421052,0.23684211,0.2631579,0.23684211,0.23684211,0.23684211,0.15789473,0.28947368,0.23684211,0.31578946,0.23684211,0.28947368,0.21052632,0.23684211,0.18421052,0.28947368,0.28947368,0.2631579,0.31578946,0.23684211,0.34210527,0.28947368,0.39473686,0.34210527,0.31578946,0.39473686,0.36842105,0.39473686,0.36842105,0.36842105,0.4473684,,0.42105263,0.36842105,0.81578946,0.5263158,0.6052632,0.55263156,0.55263156,0.57894737,0.6052632,0.65789473,0.65789473,0.8684211,0.68421054,0.68421054,0.92105263
+TorchMLP (default),0.02631579,0.05263158,0.05263158,0.02631579,0.05263158,0.02631579,0.05263158,0.02631579,0.02631579,0.078947365,0.05263158,0.02631579,0.18421052,0.05263158,0.078947365,0.15789473,0.10526316,0.078947365,0.02631579,0.18421052,0.10526316,0.13157895,0.15789473,0.05263158,0.18421052,0.15789473,0.15789473,0.13157895,0.18421052,0.21052632,0.2631579,0.15789473,0.15789473,0.28947368,0.15789473,0.10526316,0.13157895,0.078947365,0.31578946,0.05263158,0.13157895,0.10526316,0.21052632,0.2631579,0.18421052,0.15789473,0.31578946,0.28947368,0.28947368,0.2631579,0.10526316,0.36842105,0.18421052,0.2631579,0.42105263,0.31578946,0.34210527,0.39473686,0.36842105,0.5263158,0.5,0.5263158,0.57894737,,0.65789473,0.6052632,0.57894737,0.57894737,0.65789473,0.65789473,0.6052632,0.68421054,0.6315789,0.7105263,0.7105263,0.7894737,0.84210527,0.94736844
+TabSTAR (default),0.05263158,0.13157895,0.13157895,0.15789473,0.13157895,0.10526316,0.13157895,0.10526316,0.13157895,0.15789473,0.15789473,0.13157895,0.18421052,0.15789473,0.18421052,0.15789473,0.15789473,0.18421052,0.15789473,0.15789473,0.15789473,0.18421052,0.10526316,0.15789473,0.15789473,0.15789473,0.15789473,0.15789473,0.18421052,0.15789473,0.31578946,0.13157895,0.15789473,0.28947368,0.21052632,0.15789473,0.15789473,0.15789473,0.31578946,0.23684211,0.21052632,0.15789473,0.15789473,0.15789473,0.15789473,0.18421052,0.28947368,0.28947368,0.2631579,0.23684211,0.18421052,0.2631579,0.21052632,0.23684211,0.23684211,0.23684211,0.23684211,0.28947368,0.2631579,0.05263158,0.02631579,0.31578946,0.6315789,0.34210527,,0.6315789,0.31578946,0.5,0.5263158,0.47368422,0.5,0.5263158,0.5,0.5,0.65789473,0.57894737,0.68421054,0.8947368
+PerpetualBooster (tuned),0.0,0.05263158,0.0,0.0,0.02631579,0.02631579,0.02631579,0.0,0.02631579,0.02631579,0.10526316,0.078947365,0.18421052,0.05263158,0.078947365,0.13157895,0.10526316,0.10526316,0.078947365,0.13157895,0.15789473,0.078947365,0.13157895,0.078947365,0.18421052,0.15789473,0.13157895,0.15789473,0.13157895,0.15789473,0.18421052,0.15789473,0.18421052,0.23684211,0.18421052,0.13157895,0.21052632,0.15789473,0.23684211,0.15789473,0.23684211,0.21052632,0.23684211,0.18421052,0.23684211,0.15789473,0.2631579,0.21052632,0.23684211,0.23684211,0.21052632,0.28947368,0.2631579,0.28947368,0.31578946,0.28947368,0.42105263,0.31578946,0.34210527,0.36842105,0.36842105,0.4473684,0.18421052,0.39473686,0.36842105,,0.5,0.5263158,0.47368422,0.47368422,0.57894737,0.57894737,0.6315789,0.6052632,0.7763158,0.65789473,0.57894737,0.92105263
+FastaiMLP (default),0.0,0.02631579,0.0,0.02631579,0.02631579,0.0,0.05263158,0.05263158,0.02631579,0.02631579,0.05263158,0.078947365,0.10526316,0.05263158,0.10526316,0.05263158,0.10526316,0.10526316,0.13157895,0.05263158,0.13157895,0.13157895,0.13157895,0.078947365,0.15789473,0.15789473,0.13157895,0.13157895,0.05263158,0.13157895,0.18421052,0.13157895,0.18421052,0.21052632,0.10526316,0.13157895,0.13157895,0.18421052,0.2631579,0.13157895,0.15789473,0.078947365,0.21052632,0.15789473,0.18421052,0.15789473,0.23684211,0.18421052,0.02631579,0.23684211,0.21052632,0.28947368,0.21052632,0.28947368,0.36842105,0.2631579,0.34210527,0.18421052,0.23684211,0.39473686,0.42105263,0.47368422,0.47368422,0.42105263,0.68421054,0.5,,0.5,0.55263156,0.5,0.6052632,0.57894737,0.57894737,0.6052632,0.7105263,0.65789473,0.7894737,0.94736844
+RandomForest (default),0.02631579,0.05263158,0.02631579,0.02631579,0.02631579,0.05263158,0.02631579,0.05263158,0.02631579,0.05263158,0.078947365,0.15789473,0.078947365,0.05263158,0.078947365,0.078947365,0.02631579,0.05263158,0.13157895,0.05263158,0.05263158,0.078947365,0.078947365,0.078947365,0.13157895,0.15789473,0.078947365,0.078947365,0.15789473,0.18421052,0.21052632,0.21052632,0.21052632,0.21052632,0.10526316,0.13157895,0.18421052,0.21052632,0.2631579,0.21052632,0.18421052,0.21052632,0.2631579,0.23684211,0.21052632,0.2631579,0.28947368,0.28947368,0.2631579,0.21052632,0.31578946,0.13157895,0.18421052,0.13157895,0.18421052,0.28947368,0.34210527,0.34210527,0.19736843,0.34210527,0.36842105,0.39473686,0.39473686,0.42105263,0.5,0.47368422,0.5,,0.5263158,0.5263158,0.57894737,0.5,0.6052632,0.6052632,0.57894737,0.65789473,0.7105263,0.9736842
+KNN (tuned + ensembled),0.078947365,0.02631579,0.05263158,0.02631579,0.078947365,0.05263158,0.078947365,0.078947365,0.078947365,0.05263158,0.078947365,0.13157895,0.13157895,0.05263158,0.10526316,0.078947365,0.078947365,0.13157895,0.13157895,0.10526316,0.13157895,0.10526316,0.078947365,0.078947365,0.14473684,0.15789473,0.13157895,0.10526316,0.15789473,0.13157895,0.2631579,0.10526316,0.18421052,0.2631579,0.078947365,0.078947365,0.10526316,0.18421052,0.2631579,0.13157895,0.078947365,0.10526316,0.18421052,0.21052632,0.10526316,0.10526316,0.34210527,0.23684211,0.18421052,0.23684211,0.15789473,0.10526316,0.18421052,0.21052632,0.21052632,0.28947368,0.21052632,0.28947368,0.21052632,0.36842105,0.39473686,0.39473686,0.4473684,0.34210527,0.47368422,0.5263158,0.4473684,0.47368422,,0.55263156,0.6315789,0.55263156,0.6052632,0.6052632,0.6052632,0.68421054,0.9736842,0.9736842
+ExtraTrees (default),0.02631579,0.02631579,0.02631579,0.05263158,0.02631579,0.078947365,0.02631579,0.0,0.02631579,0.02631579,0.05263158,0.15789473,0.05263158,0.05263158,0.13157895,0.05263158,0.10526316,0.13157895,0.18421052,0.02631579,0.078947365,0.10526316,0.13157895,0.15789473,0.15789473,0.18421052,0.15789473,0.10526316,0.078947365,0.13157895,0.15789473,0.23684211,0.15789473,0.21052632,0.078947365,0.10526316,0.15789473,0.18421052,0.15789473,0.18421052,0.078947365,0.15789473,0.21052632,0.18421052,0.18421052,0.18421052,0.2631579,0.21052632,0.21052632,0.2631579,0.23684211,0.078947365,0.18421052,0.15789473,0.18421052,0.23684211,0.31578946,0.2631579,0.21052632,0.34210527,0.36842105,0.31578946,0.4473684,0.34210527,0.5263158,0.5263158,0.5,0.47368422,0.4473684,,0.55263156,0.5263158,0.55263156,0.55263156,0.6052632,0.6052632,0.7105263,0.94736844
+xRFM (default),0.05263158,0.02631579,0.05263158,0.05263158,0.05263158,0.078947365,0.05263158,0.05263158,0.078947365,0.05263158,0.05263158,0.13157895,0.078947365,0.078947365,0.13157895,0.10526316,0.15789473,0.10526316,0.15789473,0.05263158,0.15789473,0.15789473,0.10526316,0.10526316,0.10526316,0.15789473,0.15789473,0.18421052,0.10526316,0.15789473,0.13157895,0.18421052,0.15789473,0.078947365,0.078947365,0.10526316,0.15789473,0.21052632,0.15789473,0.18421052,0.02631579,0.18421052,0.21052632,0.18421052,0.15789473,0.21052632,0.15789473,0.21052632,0.18421052,0.21052632,0.21052632,0.21052632,0.28947368,0.21052632,0.28947368,0.23684211,0.31578946,0.2631579,0.21052632,0.34210527,0.36842105,0.36842105,0.42105263,0.39473686,0.5,0.42105263,0.39473686,0.42105263,0.36842105,0.4473684,,0.5,0.47368422,0.47368422,0.5,0.5263158,0.6315789,0.84210527
+TabFlex (default),0.0,0.05263158,0.05263158,0.078947365,0.10526316,0.02631579,0.10526316,0.0,0.05263158,0.10526316,0.13157895,0.078947365,0.15789473,0.078947365,0.13157895,0.05263158,0.13157895,0.15789473,0.078947365,0.078947365,0.15789473,0.10526316,0.13157895,0.13157895,0.18421052,0.13157895,0.15789473,0.15789473,0.13157895,0.02631579,0.23684211,0.10526316,0.10526316,0.21052632,0.10526316,0.13157895,0.078947365,0.13157895,0.21052632,0.078947365,0.15789473,0.05263158,0.15789473,0.13157895,0.10526316,0.10526316,0.21052632,0.15789473,0.13157895,0.23684211,0.078947365,0.18421052,0.23684211,0.28947368,0.21052632,0.23684211,0.28947368,0.21052632,0.23684211,0.31578946,0.34210527,0.28947368,0.39473686,0.31578946,0.47368422,0.42105263,0.42105263,0.5,0.4473684,0.47368422,0.5,,0.55263156,0.55263156,0.55263156,0.6315789,0.68421054,0.8684211
+Linear (tuned + ensembled),0.02631579,0.02631579,0.05263158,0.02631579,0.078947365,0.05263158,0.078947365,0.05263158,0.02631579,0.02631579,0.05263158,0.05263158,0.10526316,0.078947365,0.02631579,0.15789473,0.02631579,0.02631579,0.10526316,0.13157895,0.078947365,0.10526316,0.10526316,0.05263158,0.18421052,0.078947365,0.13157895,0.10526316,0.18421052,0.15789473,0.15789473,0.10526316,0.13157895,0.15789473,0.05263158,0.15789473,0.05263158,0.10526316,0.15789473,0.10526316,0.10526316,0.10526316,0.13157895,0.13157895,0.15789473,0.13157895,0.18421052,0.28947368,0.18421052,0.18421052,0.18421052,0.23684211,0.18421052,0.2631579,0.31578946,0.28947368,0.28947368,0.28947368,0.28947368,0.34210527,0.34210527,0.31578946,0.34210527,0.36842105,0.5,0.36842105,0.42105263,0.39473686,0.39473686,0.4473684,0.5263158,0.4473684,,0.75,0.5,0.7368421,0.6315789,0.8684211
+Linear (tuned),0.02631579,0.02631579,0.02631579,0.02631579,0.05263158,0.02631579,0.078947365,0.05263158,0.02631579,0.02631579,0.039473683,0.05263158,0.10526316,0.05263158,0.02631579,0.15789473,0.02631579,0.02631579,0.10526316,0.13157895,0.05263158,0.13157895,0.078947365,0.02631579,0.15789473,0.05263158,0.078947365,0.078947365,0.18421052,0.13157895,0.15789473,0.10526316,0.13157895,0.15789473,0.02631579,0.10526316,0.05263158,0.13157895,0.15789473,0.078947365,0.13157895,0.078947365,0.13157895,0.15789473,0.10526316,0.10526316,0.18421052,0.2631579,0.15789473,0.15789473,0.15789473,0.15789473,0.18421052,0.28947368,0.31578946,0.28947368,0.28947368,0.28947368,0.31578946,0.31578946,0.34210527,0.31578946,0.34210527,0.28947368,0.5,0.39473686,0.39473686,0.39473686,0.39473686,0.4473684,0.5263158,0.4473684,0.25,,0.5,0.65789473,0.6052632,0.84210527
+PerpetualBooster (default),0.0,0.05263158,0.02631579,0.02631579,0.05263158,0.02631579,0.05263158,0.05263158,0.02631579,0.02631579,0.02631579,0.05263158,0.13157895,0.05263158,0.05263158,0.05263158,0.10526316,0.05263158,0.05263158,0.10526316,0.13157895,0.05263158,0.13157895,0.05263158,0.10526316,0.15789473,0.10526316,0.13157895,0.078947365,0.10526316,0.10526316,0.13157895,0.15789473,0.10526316,0.13157895,0.10526316,0.13157895,0.10526316,0.13157895,0.10526316,0.10526316,0.078947365,0.21052632,0.15789473,0.15789473,0.10526316,0.13157895,0.18421052,0.10526316,0.15789473,0.15789473,0.21052632,0.15789473,0.28947368,0.23684211,0.21052632,0.34210527,0.23684211,0.31578946,0.34210527,0.34210527,0.31578946,0.13157895,0.28947368,0.34210527,0.2236842,0.28947368,0.42105263,0.39473686,0.39473686,0.5,0.4473684,0.5,0.5,,0.5263158,0.5,0.84210527
+Linear (default),0.0,0.02631579,0.0,0.02631579,0.05263158,0.0,0.05263158,0.02631579,0.0,0.02631579,0.05263158,0.05263158,0.10526316,0.05263158,0.02631579,0.10526316,0.05263158,0.02631579,0.05263158,0.10526316,0.05263158,0.02631579,0.078947365,0.02631579,0.15789473,0.02631579,0.078947365,0.05263158,0.13157895,0.13157895,0.15789473,0.05263158,0.05263158,0.13157895,0.05263158,0.10526316,0.02631579,0.10526316,0.15789473,0.02631579,0.10526316,0.02631579,0.02631579,0.13157895,0.10526316,0.05263158,0.13157895,0.18421052,0.10526316,0.15789473,0.078947365,0.15789473,0.13157895,0.18421052,0.15789473,0.13157895,0.18421052,0.15789473,0.21052632,0.23684211,0.23684211,0.2631579,0.31578946,0.21052632,0.42105263,0.34210527,0.34210527,0.34210527,0.31578946,0.39473686,0.47368422,0.36842105,0.2631579,0.34210527,0.47368422,,0.6052632,0.7894737
+KNN (tuned),0.02631579,0.0,0.02631579,0.02631579,0.05263158,0.02631579,0.05263158,0.02631579,0.0,0.02631579,0.02631579,0.0,0.13157895,0.05263158,0.02631579,0.02631579,0.0,0.05263158,0.02631579,0.05263158,0.078947365,0.02631579,0.078947365,0.02631579,0.13157895,0.05263158,0.05263158,0.078947365,0.05263158,0.02631579,0.18421052,0.02631579,0.078947365,0.15789473,0.02631579,0.02631579,0.05263158,0.02631579,0.15789473,0.05263158,0.02631579,0.05263158,0.05263158,0.10526316,0.05263158,0.05263158,0.18421052,0.13157895,0.10526316,0.13157895,0.05263158,0.078947365,0.078947365,0.078947365,0.02631579,0.078947365,0.15789473,0.13157895,0.10526316,0.15789473,0.15789473,0.2631579,0.31578946,0.15789473,0.31578946,0.42105263,0.21052632,0.28947368,0.02631579,0.28947368,0.36842105,0.31578946,0.36842105,0.39473686,0.5,0.39473686,,0.94736844
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.02631579,0.0,0.0,0.0,0.0,0.0,0.02631579,0.0,0.0,0.0,0.0,0.0,0.02631579,0.0,0.0,0.0,0.02631579,0.0,0.0,0.0,0.0,0.0,0.02631579,0.0,0.0,0.05263158,0.0,0.02631579,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.02631579,0.02631579,0.05263158,0.02631579,0.0,0.02631579,0.02631579,0.02631579,0.05263158,0.05263158,0.05263158,0.02631579,0.078947365,0.078947365,0.02631579,0.078947365,0.05263158,0.10526316,0.078947365,0.05263158,0.02631579,0.02631579,0.05263158,0.15789473,0.13157895,0.13157895,0.15789473,0.15789473,0.21052632,0.05263158,
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
index 4b553671787bddd15aed3d1cfbe83d004d1a4bb0..f4d1257070746d47e1f8feee2e0ee9c7181b4a3f 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Classification | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..b3b4a2398ec32760fef2724874fd72783d7fe1a4
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_classification/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2102.0,"score":0.769,"rank":3.9,"hrank":2.23,"imp":3.082,"train_time":145.29,"predict_time":1.858,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+238\/-74","imputed":false},{"position":1,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2030.0,"score":0.732,"rank":5.2,"hrank":2.91,"imp":4.491,"train_time":307.34,"predict_time":2.186,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-65","imputed":false},{"position":2,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1929.0,"score":0.71,"rank":7.7,"hrank":2.42,"imp":6.449,"train_time":38.94,"predict_time":18.762,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+324\/-148","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1847.0,"score":0.635,"rank":10.4,"hrank":2.84,"imp":6.0,"train_time":1.11,"predict_time":0.291,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+404\/-179","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1827.0,"score":0.532,"rank":11.2,"hrank":8.1,"imp":8.539,"train_time":2.33,"predict_time":0.699,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+162\/-77","imputed":false},{"position":5,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1773.0,"score":0.475,"rank":13.5,"hrank":6.55,"imp":9.506,"train_time":0.62,"predict_time":0.172,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+238\/-136","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1717.0,"score":0.431,"rank":16.2,"hrank":8.15,"imp":8.142,"train_time":780.95,"predict_time":11.64,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+240\/-126","imputed":false},{"position":7,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1688.0,"score":0.459,"rank":17.7,"hrank":10.29,"imp":8.22,"train_time":2335.83,"predict_time":14.145,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+195\/-148","imputed":false},{"position":8,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1680.0,"score":0.395,"rank":18.1,"hrank":13.48,"imp":10.342,"train_time":202.54,"predict_time":0.43,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-101","imputed":false},{"position":9,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1675.0,"score":0.423,"rank":18.4,"hrank":12.2,"imp":10.878,"train_time":1598.12,"predict_time":70.164,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+162\/-118","imputed":false},{"position":10,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1662.0,"score":0.426,"rank":19.15,"hrank":10.16,"imp":9.741,"train_time":384.95,"predict_time":0.163,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+168\/-91","imputed":false},{"position":11,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1661.0,"score":0.428,"rank":19.2,"hrank":12.88,"imp":10.555,"train_time":89.02,"predict_time":0.317,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-127","imputed":false},{"position":12,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1657.0,"score":0.414,"rank":19.4,"hrank":11.45,"imp":9.959,"train_time":384.95,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+178\/-96","imputed":false},{"position":13,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1655.0,"score":0.367,"rank":19.5,"hrank":13.55,"imp":10.08,"train_time":780.95,"predict_time":1.493,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+212\/-105","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1646.0,"score":0.367,"rank":20.0,"hrank":15.22,"imp":11.236,"train_time":91.02,"predict_time":0.568,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+172\/-82","imputed":false},{"position":15,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1643.0,"score":0.404,"rank":20.2,"hrank":14.99,"imp":10.686,"train_time":89.02,"predict_time":0.026,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+143\/-120","imputed":false},{"position":16,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1638.0,"score":0.351,"rank":20.5,"hrank":12.92,"imp":10.904,"train_time":875.01,"predict_time":0.549,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+200\/-104","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1634.0,"score":0.404,"rank":20.7,"hrank":13.19,"imp":10.118,"train_time":1635.74,"predict_time":2.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-86","imputed":false},{"position":18,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1631.0,"score":0.42,"rank":20.9,"hrank":10.4,"imp":9.393,"train_time":1.53,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-67","imputed":false},{"position":19,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1608.0,"score":0.333,"rank":22.25,"hrank":14.51,"imp":12.289,"train_time":2335.83,"predict_time":0.502,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+156\/-106","imputed":false},{"position":20,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1604.0,"score":0.342,"rank":22.5,"hrank":16.78,"imp":10.671,"train_time":202.54,"predict_time":0.055,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+164\/-101","imputed":false},{"position":21,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1580.0,"score":0.293,"rank":24.0,"hrank":19.99,"imp":12.064,"train_time":91.02,"predict_time":0.074,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-99","imputed":false},{"position":22,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1554.0,"score":0.311,"rank":25.6,"hrank":13.29,"imp":12.341,"train_time":1635.74,"predict_time":0.082,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+165\/-115","imputed":false},{"position":23,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1552.0,"score":0.298,"rank":25.7,"hrank":12.89,"imp":11.869,"train_time":875.01,"predict_time":0.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+187\/-138","imputed":false},{"position":24,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1543.0,"score":0.302,"rank":26.3,"hrank":17.59,"imp":11.715,"train_time":0.41,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-119","imputed":false},{"position":25,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1534.0,"score":0.26,"rank":26.9,"hrank":16.77,"imp":11.128,"train_time":2.3,"predict_time":0.818,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+167\/-109","imputed":false},{"position":26,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1517.0,"score":0.254,"rank":28.0,"hrank":12.17,"imp":12.659,"train_time":8.46,"predict_time":3.879,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+178\/-121","imputed":false},{"position":27,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1483.0,"score":0.209,"rank":30.2,"hrank":18.06,"imp":13.283,"train_time":1598.12,"predict_time":6.083,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+137\/-105","imputed":false},{"position":28,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1473.0,"score":0.27,"rank":30.9,"hrank":7.66,"imp":12.804,"train_time":0.47,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+233\/-150","imputed":false},{"position":29,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1465.0,"score":0.203,"rank":31.4,"hrank":24.1,"imp":15.025,"train_time":421.22,"predict_time":0.1,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+194\/-123","imputed":false},{"position":30,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1456.0,"score":0.211,"rank":32.0,"hrank":16.85,"imp":13.666,"train_time":0.63,"predict_time":0.171,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+235\/-178","imputed":false},{"position":31,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1453.0,"score":0.155,"rank":32.2,"hrank":25.62,"imp":13.439,"train_time":1128.74,"predict_time":1.573,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-84","imputed":false},{"position":32,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1452.0,"score":0.193,"rank":32.3,"hrank":21.75,"imp":14.305,"train_time":0.79,"predict_time":0.232,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+192\/-163","imputed":false},{"position":33,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1420.0,"score":0.14,"rank":34.4,"hrank":28.93,"imp":13.694,"train_time":4546.14,"predict_time":406.102,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-125","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1415.0,"score":0.139,"rank":34.8,"hrank":30.51,"imp":15.061,"train_time":1.61,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-111","imputed":false},{"position":35,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1413.0,"score":0.208,"rank":34.9,"hrank":24.92,"imp":13.861,"train_time":713.01,"predict_time":2.317,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+148\/-164","imputed":false},{"position":36,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1401.0,"score":0.148,"rank":35.7,"hrank":27.59,"imp":16.621,"train_time":713.01,"predict_time":0.155,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-162","imputed":false},{"position":37,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1365.0,"score":0.082,"rank":38.1,"hrank":34.95,"imp":16.833,"train_time":6.53,"predict_time":0.294,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-114","imputed":false},{"position":38,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1354.0,"score":0.102,"rank":38.85,"hrank":24.66,"imp":17.196,"train_time":4546.14,"predict_time":39.323,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+166\/-155","imputed":false},{"position":39,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1350.0,"score":0.113,"rank":39.1,"hrank":31.74,"imp":16.571,"train_time":421.22,"predict_time":0.007,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+181\/-127","imputed":false},{"position":40,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1338.0,"score":0.061,"rank":39.9,"hrank":32.84,"imp":16.429,"train_time":52.7,"predict_time":0.202,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-107","imputed":false},{"position":41,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1334.0,"score":0.09,"rank":40.2,"hrank":36.59,"imp":17.843,"train_time":1.4,"predict_time":0.007,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+137\/-125","imputed":false},{"position":42,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1327.0,"score":0.087,"rank":40.6,"hrank":33.62,"imp":17.764,"train_time":387.73,"predict_time":2.446,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-176","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1307.0,"score":0.072,"rank":41.9,"hrank":38.29,"imp":16.432,"train_time":1128.74,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+111\/-98","imputed":false},{"position":44,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1306.0,"score":0.101,"rank":42.0,"hrank":31.07,"imp":17.663,"train_time":82.52,"predict_time":0.97,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+176\/-165","imputed":false},{"position":45,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1306.0,"score":0.119,"rank":42.0,"hrank":36.09,"imp":19.39,"train_time":68.71,"predict_time":7.89,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+184\/-156","imputed":false},{"position":46,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1303.0,"score":0.212,"rank":42.2,"hrank":7.72,"imp":18.833,"train_time":5.05,"predict_time":2.948,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+252\/-285","imputed":false},{"position":47,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1293.0,"score":0.045,"rank":42.8,"hrank":37.75,"imp":16.851,"train_time":52.7,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-100","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1289.0,"score":0.041,"rank":43.1,"hrank":40.53,"imp":16.551,"train_time":49.21,"predict_time":0.149,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+113\/-139","imputed":false},{"position":49,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1258.0,"score":0.054,"rank":45.0,"hrank":42.34,"imp":16.349,"train_time":0.33,"predict_time":0.05,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-132","imputed":false},{"position":50,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.03,"rank":45.9,"hrank":44.09,"imp":17.041,"train_time":11.03,"predict_time":0.489,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-109","imputed":false},{"position":51,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1232.0,"score":0.024,"rank":46.6,"hrank":42.71,"imp":19.606,"train_time":14.93,"predict_time":13.661,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-115","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1227.0,"score":0.042,"rank":46.9,"hrank":42.04,"imp":18.856,"train_time":49.21,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+145\/-194","imputed":false},{"position":53,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1193.0,"score":0.021,"rank":48.95,"hrank":44.89,"imp":21.028,"train_time":41.63,"predict_time":39.259,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-163","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1155.0,"score":0.013,"rank":51.1,"hrank":47.91,"imp":21.17,"train_time":387.73,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-170","imputed":false},{"position":55,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1137.0,"score":0.011,"rank":52.1,"hrank":50.25,"imp":19.824,"train_time":1.84,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-123","imputed":false},{"position":56,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1060.0,"score":0.0,"rank":56.1,"hrank":54.95,"imp":26.384,"train_time":3486.11,"predict_time":2.194,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-103","imputed":false},{"position":57,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1058.0,"score":0.017,"rank":56.2,"hrank":51.77,"imp":24.692,"train_time":1.63,"predict_time":0.11,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+153\/-235","imputed":false},{"position":58,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1046.0,"score":0.015,"rank":56.75,"hrank":51.56,"imp":29.46,"train_time":43.41,"predict_time":0.111,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+185\/-264","imputed":false},{"position":59,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1043.0,"score":0.016,"rank":56.9,"hrank":53.51,"imp":26.105,"train_time":332.4,"predict_time":33.481,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+121\/-226","imputed":false},{"position":60,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1039.0,"score":0.0,"rank":57.1,"hrank":55.99,"imp":26.456,"train_time":3486.11,"predict_time":8.148,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-109","imputed":false},{"position":61,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1008.0,"score":0.002,"rank":58.55,"hrank":54.61,"imp":30.389,"train_time":43.41,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+165\/-280","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":58.9,"hrank":57.01,"imp":24.603,"train_time":0.11,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-242","imputed":false},{"position":63,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":956.0,"score":0.0,"rank":60.8,"hrank":59.64,"imp":25.782,"train_time":17.26,"predict_time":6.893,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-210","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":934.0,"score":0.0,"rank":61.7,"hrank":59.66,"imp":29.251,"train_time":22.23,"predict_time":0.567,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+145\/-196","imputed":false},{"position":65,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":893.0,"score":0.0,"rank":63.3,"hrank":62.85,"imp":27.911,"train_time":0.32,"predict_time":0.056,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-236","imputed":false},{"position":66,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":890.0,"score":0.0,"rank":63.4,"hrank":62.78,"imp":32.855,"train_time":85.23,"predict_time":2.947,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+48\/-120","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":865.0,"score":0.0,"rank":64.3,"hrank":63.63,"imp":31.979,"train_time":0.21,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-243","imputed":false},{"position":68,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":865.0,"score":0.0,"rank":64.3,"hrank":62.82,"imp":30.588,"train_time":22.23,"predict_time":0.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-251","imputed":false},{"position":69,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":859.0,"score":0.022,"rank":64.5,"hrank":56.16,"imp":35.072,"train_time":2.96,"predict_time":0.183,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+236\/-497","imputed":false},{"position":70,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":823.0,"score":0.0,"rank":65.7,"hrank":65.42,"imp":34.185,"train_time":332.4,"predict_time":1.631,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-172","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":790.0,"score":0.0,"rank":66.7,"hrank":66.0,"imp":36.745,"train_time":0.18,"predict_time":0.032,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-192","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":737.0,"score":0.0,"rank":68.1,"hrank":67.2,"imp":39.058,"train_time":1.67,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+163\/-360","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":478.0,"score":0.0,"rank":72.5,"hrank":72.47,"imp":50.755,"train_time":0.12,"predict_time":0.093,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-349","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/pareto_front_explorer.html
index ed7844a6df3c1f6bb3b526a56934fbcbd5548b4e..90dc00d59411dc9e2d362f20cfd5186af0a70dab 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.970420493,"imp":17.6629337184,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0198727775,"imp":9.3929627125,"elo":1630.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0167342488,"imp":9.9587335282,"elo":1657.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1627138271,"imp":9.7410518279,"elo":1661.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.02389485,"imp":11.715466652,"elo":1543.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0257224855,"imp":10.6863195805,"elo":1643.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3166644071,"imp":10.5546263337,"elo":1660.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0068099474,"imp":17.8429705318,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.006754717,"imp":16.5713605893,"elo":1350.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0998897869,"imp":15.0254091034,"elo":1465.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0559129084,"imp":27.9111955894,"elo":892.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0143949688,"imp":18.8561350573,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1486815713,"imp":16.5513343714,"elo":1288.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.109744233,"imp":24.6915792924,"elo":1057.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0890979144,"imp":21.1700554595,"elo":1155.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4458463373,"imp":17.7641123217,"elo":1327.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0504837885,"imp":16.3491381425,"elo":1258.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0742212397,"imp":12.0635574748,"elo":1579.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.5681105528,"imp":11.2357370861,"elo":1646.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.948262606,"imp":18.8331940342,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4891016512,"imp":17.0408331207,"elo":1243.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5024822125,"imp":12.2885548487,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":14.1450303793,"imp":8.2196331689,"elo":1687.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8931236803,"imp":25.7815891216,"elo":956.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0175726795,"imp":39.058107621,"elo":737.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2360405674,"imp":30.5884063132,"elo":865.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5669505531,"imp":29.250530009,"elo":934.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0120761251,"imp":24.6033316585,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0173683474,"imp":16.8508789793,"elo":1293.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2020648674,"imp":16.4285035282,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2942921663,"imp":16.8326922638,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.08214277,"imp":12.3410565047,"elo":1554.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.1463462325,"imp":10.1178278972,"elo":1634.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8184752083,"imp":11.1276358993,"elo":1533.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4934856828,"imp":10.0797482352,"elo":1655.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6399970659,"imp":8.1416301567,"elo":1716.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.6609678586,"imp":19.6059845316,"elo":1232.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2594035993,"imp":21.0282927544,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.3227901343,"imp":17.1961294701,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":406.1020417363,"imp":13.6941328483,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2319547305,"imp":14.3049973813,"elo":1451.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":18.7619482789,"imp":6.4485292983,"elo":1928.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0317264099,"imp":36.7450378085,"elo":789.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.8791795848,"imp":12.6594612614,"elo":1516.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.171595863,"imp":9.5061806177,"elo":1772.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0553381931,"imp":15.0605116327,"elo":1414.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0570275282,"imp":11.8692752948,"elo":1552.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5491669944,"imp":10.9044317748,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6991145285,"imp":8.5386327366,"elo":1826.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2909668033,"imp":5.9995997481,"elo":1847.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9466920315,"imp":32.8545637836,"elo":890.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1940808855,"imp":26.3835533588,"elo":1059.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1480810234,"imp":26.4556542463,"elo":1038.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1710376878,"imp":13.6658034437,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0468974441,"imp":19.8235103881,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0536547712,"imp":16.4323136139,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5732102138,"imp":13.4385258165,"elo":1453.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0471622003,"imp":12.8041808129,"elo":1472.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0546297999,"imp":10.6713217964,"elo":1604.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4302741489,"imp":10.3423013795,"elo":1680.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.8904270487,"imp":19.3895045825,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":6.0826693723,"imp":13.2834660869,"elo":1483.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":70.1635578676,"imp":10.8780238272,"elo":1675.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1830265708,"imp":35.0723495798,"elo":859.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1546216562,"imp":16.6207806431,"elo":1401.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.3169671728,"imp":13.8612441205,"elo":1413.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.970420493,"x_train":82.518930664,"imp":17.6629337184,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0198727775,"x_train":1.5291414753,"imp":9.3929627125,"elo":1630.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0167342488,"x_train":384.9547373775,"imp":9.9587335282,"elo":1657.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1627138271,"x_train":384.9547373775,"imp":9.7410518279,"elo":1661.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.02389485,"x_train":0.4117206383,"imp":11.715466652,"elo":1543.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0257224855,"x_train":89.0178815303,"imp":10.6863195805,"elo":1643.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3166644071,"x_train":89.0178815303,"imp":10.5546263337,"elo":1660.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0068099474,"x_train":1.4013272116,"imp":17.8429705318,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.006754717,"x_train":421.2236428358,"imp":16.5713605893,"elo":1350.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0998897869,"x_train":421.2236428358,"imp":15.0254091034,"elo":1465.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0559129084,"x_train":0.3214016003,"imp":27.9111955894,"elo":892.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0143949688,"x_train":49.2121405424,"imp":18.8561350573,"elo":1227.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1486815713,"x_train":49.2121405424,"imp":16.5513343714,"elo":1288.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.109744233,"x_train":1.6301605991,"imp":24.6915792924,"elo":1057.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0890979144,"x_train":387.7306724242,"imp":21.1700554595,"elo":1155.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.4458463373,"x_train":387.7306724242,"imp":17.7641123217,"elo":1327.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0504837885,"x_train":0.3317135,"imp":16.3491381425,"elo":1258.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.0742212397,"x_train":91.0184357178,"imp":12.0635574748,"elo":1579.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":0.5681105528,"x_train":91.0184357178,"imp":11.2357370861,"elo":1646.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":2.948262606,"x_train":5.0473841082,"imp":18.8331940342,"elo":1302.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4891016512,"x_train":11.026497565,"imp":17.0408331207,"elo":1243.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5024822125,"x_train":2335.8257219827,"imp":12.2885548487,"elo":1608.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":14.1450303793,"x_train":2335.8257219827,"imp":8.2196331689,"elo":1687.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8931236803,"x_train":17.2564525143,"imp":25.7815891216,"elo":956.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0175726795,"x_train":1.6656394605,"imp":39.058107621,"elo":737.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.2360405674,"x_train":22.233335416,"imp":30.5884063132,"elo":865.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5669505531,"x_train":22.233335416,"imp":29.250530009,"elo":934.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0120761251,"x_train":0.1136960958,"imp":24.6033316585,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0173683474,"x_train":52.7035618074,"imp":16.8508789793,"elo":1293.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.2020648674,"x_train":52.7035618074,"imp":16.4285035282,"elo":1338.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2942921663,"x_train":6.5297785644,"imp":16.8326922638,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.08214277,"x_train":1635.7415319042,"imp":12.3410565047,"elo":1554.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":2.1463462325,"x_train":1635.7415319042,"imp":10.1178278972,"elo":1634.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8184752083,"x_train":2.2966540372,"imp":11.1276358993,"elo":1533.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4934856828,"x_train":780.948577715,"imp":10.0797482352,"elo":1655.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6399970659,"x_train":780.948577715,"imp":8.1416301567,"elo":1716.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.6609678586,"x_train":14.9320242657,"imp":19.6059845316,"elo":1232.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2594035993,"x_train":41.6325123071,"imp":21.0282927544,"elo":1192.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.3227901343,"x_train":4546.1421807431,"imp":17.1961294701,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":406.1020417363,"x_train":4546.1421807431,"imp":13.6941328483,"elo":1420.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2319547305,"x_train":0.7863945199,"imp":14.3049973813,"elo":1451.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":18.7619482789,"x_train":38.9361051757,"imp":6.4485292983,"elo":1928.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0317264099,"x_train":0.1750892902,"imp":36.7450378085,"elo":789.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":3.8791795848,"x_train":8.464564148,"imp":12.6594612614,"elo":1516.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.171595863,"x_train":0.6194744057,"imp":9.5061806177,"elo":1772.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0553381931,"x_train":1.6076337396,"imp":15.0605116327,"elo":1414.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0570275282,"x_train":875.0118329531,"imp":11.8692752948,"elo":1552.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5491669944,"x_train":875.0118329531,"imp":10.9044317748,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6991145285,"x_train":2.330497005,"imp":8.5386327366,"elo":1826.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2909668033,"x_train":1.1055094044,"imp":5.9995997481,"elo":1847.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9466920315,"x_train":85.2279597522,"imp":32.8545637836,"elo":890.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.1940808855,"x_train":3486.1100818653,"imp":26.3835533588,"elo":1059.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1480810234,"x_train":3486.1100818653,"imp":26.4556542463,"elo":1038.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1710376878,"x_train":0.6314611374,"imp":13.6658034437,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0468974441,"x_train":1.8436799882,"imp":19.8235103881,"elo":1136.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0536547712,"x_train":1128.7390334238,"imp":16.4323136139,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5732102138,"x_train":1128.7390334238,"imp":13.4385258165,"elo":1453.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0471622003,"x_train":0.4669717504,"imp":12.8041808129,"elo":1472.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0546297999,"x_train":202.5364616477,"imp":10.6713217964,"elo":1604.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4302741489,"x_train":202.5364616477,"imp":10.3423013795,"elo":1680.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.8904270487,"x_train":68.7091297861,"imp":19.3895045825,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":6.0826693723,"x_train":1598.1220946317,"imp":13.2834660869,"elo":1483.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":70.1635578676,"x_train":1598.1220946317,"imp":10.8780238272,"elo":1675.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.1830265708,"x_train":2.9551878511,"imp":35.0723495798,"elo":859.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1546216562,"x_train":713.0056913426,"imp":16.6207806431,"elo":1401.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.3169671728,"x_train":713.0056913426,"imp":13.8612441205,"elo":1413.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
index 3208ccde4b089e5840b542c59cbe2cd242e67b43..c7bf18fd5d616d0f99026626829bb105cee22931 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_classification/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":82.518930664,"imp":17.7783471369,"elo":1344.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":1.5291414753,"imp":9.491241243,"elo":1683.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.9721250392,"imp":9.6999483994,"elo":1681.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":12.7214520913,"imp":9.6728929545,"elo":1701.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":24.1437443132,"imp":9.6761930535,"elo":1710.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":53.2602822109,"imp":9.7647789612,"elo":1727.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":99.398536178,"imp":9.7773216807,"elo":1727.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":193.3194459181,"imp":9.8341522645,"elo":1721.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":286.5879298672,"imp":9.8717448659,"elo":1715.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":384.9547373775,"imp":9.824439582,"elo":1712.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.4117206383,"imp":11.8287672651,"elo":1585.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.8690160818,"imp":11.3284736317,"elo":1654.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.0802815735,"imp":11.0507653547,"elo":1676.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.1754502095,"imp":10.7601558928,"elo":1692.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.940122673,"imp":10.6749383989,"elo":1699.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":22.2455560723,"imp":10.5350564433,"elo":1707.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":44.3873766321,"imp":10.5110964613,"elo":1715.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":66.3664018643,"imp":10.5653157475,"elo":1718.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":89.0178815303,"imp":10.6549946425,"elo":1724.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":1.4013272116,"imp":17.9345318934,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":4.1193754289,"imp":16.7471474255,"elo":1418.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":10.042965061,"imp":16.1012884683,"elo":1450.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":20.7240396679,"imp":15.8481919199,"elo":1465.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":52.2528950395,"imp":15.5347747171,"elo":1486.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":105.4596015797,"imp":15.4125820364,"elo":1495.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":207.809367173,"imp":15.232787978,"elo":1502.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":315.0112350614,"imp":15.1784841885,"elo":1499.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":421.2236428358,"imp":15.1141849883,"elo":1501.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3214016003,"imp":27.98552044,"elo":887.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.5493187284,"imp":24.9759763005,"elo":974.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.3228210017,"imp":19.9577975472,"elo":1132.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.5095259777,"imp":17.8375606124,"elo":1217.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.1644908467,"imp":16.9416737546,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.2587333157,"imp":16.4669656718,"elo":1289.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.4863048451,"imp":15.9705548137,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":36.7034709861,"imp":15.9380732553,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":49.2121405424,"imp":16.6306303792,"elo":1315.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.6301605991,"imp":24.7595361474,"elo":1079.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.3926690631,"imp":23.0477806361,"elo":1163.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":9.3189904508,"imp":20.7635168043,"elo":1259.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":19.4603799719,"imp":19.7613730569,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":46.679624122,"imp":18.8448843712,"elo":1322.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":94.3885780511,"imp":18.5005763241,"elo":1341.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":191.9205953467,"imp":18.0209777098,"elo":1353.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":288.7632596624,"imp":17.8684816306,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":387.7306724242,"imp":17.8487232653,"elo":1360.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.3317135,"imp":16.450314925,"elo":1299.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":0.8849406618,"imp":15.4488333983,"elo":1388.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":2.143134634,"imp":14.0917188516,"elo":1508.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":4.3187074851,"imp":13.0693815002,"elo":1592.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":11.2464284633,"imp":11.9726028643,"elo":1654.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":22.4893847368,"imp":11.4942207921,"elo":1689.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":44.9141349299,"imp":11.3534039222,"elo":1700.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":67.4875589082,"imp":11.3257978613,"elo":1705.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":91.0184357178,"imp":11.344681261,"elo":1701.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":5.0473841082,"imp":18.9276909301,"elo":1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":11.026497565,"imp":17.1294418406,"elo":1269.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.1149019147,"imp":15.1706285991,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":63.584802766,"imp":12.8525382121,"elo":1550.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":122.2265200742,"imp":12.3287993704,"elo":1583.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":281.6186890352,"imp":11.782681671,"elo":1630.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":581.6935693293,"imp":10.889492571,"elo":1666.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1150.8300045772,"imp":10.065987363,"elo":1703.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1716.2616608148,"imp":9.0450371026,"elo":1731.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2335.8257219827,"imp":8.352818795,"elo":1752.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":17.2564525143,"imp":25.2607495901,"elo":991.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.6656394605,"imp":39.0642635035,"elo":761.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":6.4059982085,"imp":35.591731487,"elo":849.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":18.5028948612,"imp":29.7259121704,"elo":940.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":22.233335416,"imp":29.278456989,"elo":955.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1136960958,"imp":24.6930241703,"elo":999.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.3871933784,"imp":19.5833547824,"elo":1169.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.2235890503,"imp":17.7609340623,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":2.5806805069,"imp":17.1613052185,"elo":1295.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":6.4224304798,"imp":16.7142897375,"elo":1350.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":12.9124255888,"imp":16.5753022845,"elo":1356.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":25.9873657979,"imp":16.4674560346,"elo":1365.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":39.2207943229,"imp":16.4652364548,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":52.7035618074,"imp":16.5087546984,"elo":1370.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.5297785644,"imp":16.9053393771,"elo":1383.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":15.3163332372,"imp":14.6210264384,"elo":1504.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":39.3133997603,"imp":13.1111128566,"elo":1582.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":79.4447895462,"imp":12.0724836057,"elo":1614.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":204.9296802966,"imp":11.2915866822,"elo":1647.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":412.61016262,"imp":10.881660853,"elo":1668.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":813.6980979059,"imp":10.4752223495,"elo":1674.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1221.0196908737,"imp":10.3145721226,"elo":1687.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1635.7415319042,"imp":10.2282592278,"elo":1696.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2.2966540372,"imp":11.2492008343,"elo":1560.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.7344117695,"imp":11.1445835898,"elo":1583.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":18.4049918241,"imp":10.9046971829,"elo":1608.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":36.2506481168,"imp":10.7569347224,"elo":1627.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":88.2736709813,"imp":10.3198889772,"elo":1679.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":182.8828378906,"imp":9.8152496316,"elo":1703.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":376.7342623721,"imp":9.1635361732,"elo":1732.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":570.2939458958,"imp":8.7762828203,"elo":1746.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":780.948577715,"imp":8.2533253958,"elo":1761.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":14.9320242657,"imp":19.6858576472,"elo":1256.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.6325123071,"imp":21.0976772389,"elo":1218.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.3901218357,"imp":19.2028304181,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":119.1912414891,"imp":17.6639004573,"elo":1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":214.6945851817,"imp":17.0159710418,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":545.8855951454,"imp":15.668596096,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1118.7729763883,"imp":14.8368783417,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2249.1619866125,"imp":14.3706389903,"elo":1465.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3393.4090688327,"imp":13.9629897147,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4546.1421807431,"imp":13.7527405267,"elo":1466.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.7863945199,"imp":14.3899781323,"elo":1495.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":38.9361051757,"imp":6.557487887,"elo":1990.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1750892902,"imp":36.7945450937,"elo":809.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":8.464564148,"imp":12.7682217685,"elo":1559.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.6194744057,"imp":9.6238578672,"elo":1821.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":1.6076337396,"imp":15.1564198567,"elo":1452.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":5.2365016338,"imp":14.4005222476,"elo":1487.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":19.5371570641,"imp":13.3518807492,"elo":1552.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":41.9262650686,"imp":12.4994851212,"elo":1601.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":107.9205498469,"imp":11.8438984455,"elo":1634.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":217.2325339637,"imp":11.5183191167,"elo":1650.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":430.1308094699,"imp":11.1900788346,"elo":1666.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":649.8093981365,"imp":11.0914491011,"elo":1681.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":875.0118329531,"imp":11.0230785467,"elo":1695.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.330497005,"imp":8.6491122691,"elo":1889.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":1.1055094044,"imp":5.9137624298,"elo":1875.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":85.2279597522,"imp":32.899788994,"elo":899.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":99.1248084332,"imp":32.7796115249,"elo":915.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":139.1755162215,"imp":32.4329755298,"elo":924.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":198.0940117736,"imp":31.6399865293,"elo":949.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":444.3646878451,"imp":30.0949731269,"elo":974.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":895.5564058915,"imp":28.5399652438,"elo":1013.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1773.8913331235,"imp":27.8873277974,"elo":1031.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2617.7743320608,"imp":27.1154322071,"elo":1057.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3486.1100818653,"imp":26.5078469087,"elo":1085.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.6314611374,"imp":13.7835229791,"elo":1495.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":1.8436799882,"imp":19.8900519711,"elo":1164.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":6.4007633821,"imp":18.1795243011,"elo":1251.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":27.9733904091,"imp":16.5495848047,"elo":1319.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":59.9007241046,"imp":15.6091450562,"elo":1373.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":143.3365257329,"imp":14.6701208982,"elo":1412.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":278.2745551506,"imp":14.183180672,"elo":1444.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":543.6674489023,"imp":13.8593137304,"elo":1473.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":841.4117357913,"imp":13.6597843306,"elo":1490.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":1128.7390334238,"imp":13.5312380654,"elo":1497.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.4669717504,"imp":12.9118537073,"elo":1518.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.0321466005,"imp":11.8212367436,"elo":1616.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":5.5778533182,"imp":11.3229193666,"elo":1666.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":9.7317062539,"imp":10.9403239771,"elo":1698.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":25.7016530469,"imp":10.628774182,"elo":1717.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":51.0311253263,"imp":10.5022373116,"elo":1730.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":101.303854312,"imp":10.4647468749,"elo":1739.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":150.6411976385,"imp":10.4529428699,"elo":1741.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":202.5364616477,"imp":10.4820421716,"elo":1742.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":68.7091297861,"imp":19.4673543521,"elo":1350.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":129.9001536616,"imp":17.1822083639,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":315.8079093334,"imp":14.4126589357,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":621.987734561,"imp":12.5570854498,"elo":1618.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1537.7988263416,"imp":10.8050873617,"elo":1735.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1598.1220946317,"imp":10.9608112326,"elo":1732.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.9551878511,"imp":35.1302552748,"elo":871.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":6.5948259183,"imp":27.689365732,"elo":1016.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.3219967074,"imp":19.5874877747,"elo":1246.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":35.2109614365,"imp":16.4609945688,"elo":1381.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.816423174,"imp":15.4820902855,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":176.8663518766,"imp":14.9281733183,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":354.8398796964,"imp":14.4325741349,"elo":1447.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":531.8168061758,"imp":14.0604231438,"elo":1450.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":713.0056913426,"imp":13.6083145675,"elo":1460.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":0.970420493,"x_train":82.518930664,"imp":17.7783471369,"elo":1344.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0198727775,"x_train":1.5291414753,"imp":9.491241243,"elo":1683.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0381500095,"x_train":3.9721250392,"imp":9.6999483994,"elo":1681.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0610258172,"x_train":12.7214520913,"imp":9.6728929545,"elo":1701.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0803301633,"x_train":24.1437443132,"imp":9.6761930535,"elo":1710.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0971069508,"x_train":53.2602822109,"imp":9.7647789612,"elo":1727.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1155626332,"x_train":99.398536178,"imp":9.7773216807,"elo":1727.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1393798126,"x_train":193.3194459181,"imp":9.8341522645,"elo":1721.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1506167429,"x_train":286.5879298672,"imp":9.8717448659,"elo":1715.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1627138271,"x_train":384.9547373775,"imp":9.824439582,"elo":1712.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.02389485,"x_train":0.4117206383,"imp":11.8287672651,"elo":1585.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0478927199,"x_train":0.8690160818,"imp":11.3284736317,"elo":1654.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1045674981,"x_train":2.0802815735,"imp":11.0507653547,"elo":1676.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1647539973,"x_train":4.1754502095,"imp":10.7601558928,"elo":1692.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2414369657,"x_train":10.940122673,"imp":10.6749383989,"elo":1699.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2964729106,"x_train":22.2455560723,"imp":10.5350564433,"elo":1707.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3247321968,"x_train":44.3873766321,"imp":10.5110964613,"elo":1715.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3090326441,"x_train":66.3664018643,"imp":10.5653157475,"elo":1718.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3166644071,"x_train":89.0178815303,"imp":10.6549946425,"elo":1724.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0068099474,"x_train":1.4013272116,"imp":17.9345318934,"elo":1359.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0177962725,"x_train":4.1193754289,"imp":16.7471474255,"elo":1418.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0331174137,"x_train":10.042965061,"imp":16.1012884683,"elo":1450.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0439779117,"x_train":20.7240396679,"imp":15.8481919199,"elo":1465.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0691665167,"x_train":52.2528950395,"imp":15.5347747171,"elo":1486.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0836217017,"x_train":105.4596015797,"imp":15.4125820364,"elo":1495.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0948483349,"x_train":207.809367173,"imp":15.232787978,"elo":1502.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1018479444,"x_train":315.0112350614,"imp":15.1784841885,"elo":1499.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0996158221,"x_train":421.2236428358,"imp":15.1141849883,"elo":1501.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0559129084,"x_train":0.3214016003,"imp":27.98552044,"elo":887.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.064531699,"x_train":0.5493187284,"imp":24.9759763005,"elo":974.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0703221149,"x_train":1.3228210017,"imp":19.9577975472,"elo":1132.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.073653813,"x_train":2.5095259777,"imp":17.8375606124,"elo":1217.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.086629059,"x_train":6.1644908467,"imp":16.9416737546,"elo":1254.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0999308324,"x_train":12.2587333157,"imp":16.4669656718,"elo":1289.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1177984553,"x_train":24.4863048451,"imp":15.9705548137,"elo":1321.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1410295133,"x_train":36.7034709861,"imp":15.9380732553,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1422071534,"x_train":49.2121405424,"imp":16.6306303792,"elo":1315.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.109744233,"x_train":1.6301605991,"imp":24.7595361474,"elo":1079.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2272880738,"x_train":3.3926690631,"imp":23.0477806361,"elo":1163.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5474602559,"x_train":9.3189904508,"imp":20.7635168043,"elo":1259.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8604236954,"x_train":19.4603799719,"imp":19.7613730569,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.1405220728,"x_train":46.679624122,"imp":18.8448843712,"elo":1322.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3925732088,"x_train":94.3885780511,"imp":18.5005763241,"elo":1341.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7973939159,"x_train":191.9205953467,"imp":18.0209777098,"elo":1353.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.9804848892,"x_train":288.7632596624,"imp":17.8684816306,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.3433373669,"x_train":387.7306724242,"imp":17.8487232653,"elo":1360.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0504837885,"x_train":0.3317135,"imp":16.450314925,"elo":1299.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.330362467,"x_train":0.8849406618,"imp":15.4488333983,"elo":1388.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5377965122,"x_train":2.143134634,"imp":14.0917188516,"elo":1508.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7884842205,"x_train":4.3187074851,"imp":13.0693815002,"elo":1592.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7687781828,"x_train":11.2464284633,"imp":11.9726028643,"elo":1654.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.755201967,"x_train":22.4893847368,"imp":11.4942207921,"elo":1689.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7276316945,"x_train":44.9141349299,"imp":11.3534039222,"elo":1700.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6450965711,"x_train":67.4875589082,"imp":11.3257978613,"elo":1705.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6136389605,"x_train":91.0184357178,"imp":11.344681261,"elo":1701.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":2.948262606,"x_train":5.0473841082,"imp":18.9276909301,"elo":1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4891016512,"x_train":11.026497565,"imp":17.1294418406,"elo":1269.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.0456997204,"x_train":25.1149019147,"imp":15.1706285991,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.5533251388,"x_train":63.584802766,"imp":12.8525382121,"elo":1550.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.2314257133,"x_train":122.2265200742,"imp":12.3287993704,"elo":1583.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.994864421,"x_train":281.6186890352,"imp":11.782681671,"elo":1630.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.3596868906,"x_train":581.6935693293,"imp":10.889492571,"elo":1666.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":12.3832071121,"x_train":1150.8300045772,"imp":10.065987363,"elo":1703.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":13.3833030376,"x_train":1716.2616608148,"imp":9.0450371026,"elo":1731.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":14.1450303793,"x_train":2335.8257219827,"imp":8.352818795,"elo":1752.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.8931236803,"x_train":17.2564525143,"imp":25.2607495901,"elo":991.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0175726795,"x_train":1.6656394605,"imp":39.0642635035,"elo":761.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1527010552,"x_train":6.4059982085,"imp":35.591731487,"elo":849.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4812091674,"x_train":18.5028948612,"imp":29.7259121704,"elo":940.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5669505531,"x_train":22.233335416,"imp":29.278456989,"elo":955.5,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0120761251,"x_train":0.1136960958,"imp":24.6930241703,"elo":999.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0375156837,"x_train":0.3871933784,"imp":19.5833547824,"elo":1169.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0764464768,"x_train":1.2235890503,"imp":17.7609340623,"elo":1255.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1157837036,"x_train":2.5806805069,"imp":17.1613052185,"elo":1295.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1326680035,"x_train":6.4224304798,"imp":16.7142897375,"elo":1350.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1710393828,"x_train":12.9124255888,"imp":16.5753022845,"elo":1356.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1991880198,"x_train":25.9873657979,"imp":16.4674560346,"elo":1365.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2137395788,"x_train":39.2207943229,"imp":16.4652364548,"elo":1363.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.193317742,"x_train":52.7035618074,"imp":16.5087546984,"elo":1370.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2942921663,"x_train":6.5297785644,"imp":16.9053393771,"elo":1383.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4603279722,"x_train":15.3163332372,"imp":14.6210264384,"elo":1504.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.7544553224,"x_train":39.3133997603,"imp":13.1111128566,"elo":1582.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.0304844995,"x_train":79.4447895462,"imp":12.0724836057,"elo":1614.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.3434994293,"x_train":204.9296802966,"imp":11.2915866822,"elo":1647.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.7588377133,"x_train":412.61016262,"imp":10.881660853,"elo":1668.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.0398213365,"x_train":813.6980979059,"imp":10.4752223495,"elo":1674.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.1897696555,"x_train":1221.0196908737,"imp":10.3145721226,"elo":1687.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.1463462325,"x_train":1635.7415319042,"imp":10.2282592278,"elo":1696.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8184752083,"x_train":2.2966540372,"imp":11.2492008343,"elo":1560.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.6939241991,"x_train":5.7344117695,"imp":11.1445835898,"elo":1583.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2294004302,"x_train":18.4049918241,"imp":10.9046971829,"elo":1608.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.1368098652,"x_train":36.2506481168,"imp":10.7569347224,"elo":1627.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.826705741,"x_train":88.2736709813,"imp":10.3198889772,"elo":1679.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.0207330225,"x_train":182.8828378906,"imp":9.8152496316,"elo":1703.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.2704454231,"x_train":376.7342623721,"imp":9.1635361732,"elo":1732.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":12.3919513821,"x_train":570.2939458958,"imp":8.7762828203,"elo":1746.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6399970659,"x_train":780.948577715,"imp":8.2533253958,"elo":1761.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":13.6609678586,"x_train":14.9320242657,"imp":19.6858576472,"elo":1256.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.2594035993,"x_train":41.6325123071,"imp":21.0976772389,"elo":1218.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":56.6400616397,"x_train":63.3901218357,"imp":19.2028304181,"elo":1274.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":88.5407748208,"x_train":119.1912414891,"imp":17.6639004573,"elo":1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":118.0359236254,"x_train":214.6945851817,"imp":17.0159710418,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":189.7782849186,"x_train":545.8855951454,"imp":15.668596096,"elo":1419.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":271.3697151109,"x_train":1118.7729763883,"imp":14.8368783417,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":353.6812333996,"x_train":2249.1619866125,"imp":14.3706389903,"elo":1465.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":380.7963519935,"x_train":3393.4090688327,"imp":13.9629897147,"elo":1462.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":414.7537971415,"x_train":4546.1421807431,"imp":13.7527405267,"elo":1466.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2319547305,"x_train":0.7863945199,"imp":14.3899781323,"elo":1495.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":18.7619482789,"x_train":38.9361051757,"imp":6.557487887,"elo":1990.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0317264099,"x_train":0.1750892902,"imp":36.7945450937,"elo":809.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":3.8791795848,"x_train":8.464564148,"imp":12.7682217685,"elo":1559.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.171595863,"x_train":0.6194744057,"imp":9.6238578672,"elo":1821.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0553381931,"x_train":1.6076337396,"imp":15.1564198567,"elo":1452.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1172004971,"x_train":5.2365016338,"imp":14.4005222476,"elo":1487.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2044016087,"x_train":19.5371570641,"imp":13.3518807492,"elo":1552.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2619165464,"x_train":41.9262650686,"imp":12.4994851212,"elo":1601.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3522960448,"x_train":107.9205498469,"imp":11.8438984455,"elo":1634.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4098866808,"x_train":217.2325339637,"imp":11.5183191167,"elo":1650.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4587122968,"x_train":430.1308094699,"imp":11.1900788346,"elo":1666.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.4807977389,"x_train":649.8093981365,"imp":11.0914491011,"elo":1681.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.4970654764,"x_train":875.0118329531,"imp":11.0230785467,"elo":1695.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6991145285,"x_train":2.330497005,"imp":8.6491122691,"elo":1889.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2909668033,"x_train":1.1055094044,"imp":5.9137624298,"elo":1875.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9466920315,"x_train":85.2279597522,"imp":32.899788994,"elo":899.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9477715354,"x_train":99.1248084332,"imp":32.7796115249,"elo":915.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.9426321268,"x_train":139.1755162215,"imp":32.4329755298,"elo":924.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.8633854515,"x_train":198.0940117736,"imp":31.6399865293,"elo":949.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.8789533666,"x_train":444.3646878451,"imp":30.0949731269,"elo":974.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.1163669829,"x_train":895.5564058915,"imp":28.5399652438,"elo":1013.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5458134689,"x_train":1773.8913331235,"imp":27.8873277974,"elo":1031.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.78350456,"x_train":2617.7743320608,"imp":27.1154322071,"elo":1057.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1480810234,"x_train":3486.1100818653,"imp":26.5078469087,"elo":1085.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1710376878,"x_train":0.6314611374,"imp":13.7835229791,"elo":1495.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0468974441,"x_train":1.8436799882,"imp":19.8900519711,"elo":1164.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1038006698,"x_train":6.4007633821,"imp":18.1795243011,"elo":1251.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2635867973,"x_train":27.9733904091,"imp":16.5495848047,"elo":1319.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4134180888,"x_train":59.9007241046,"imp":15.6091450562,"elo":1373.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.593126281,"x_train":143.3365257329,"imp":14.6701208982,"elo":1412.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8268268446,"x_train":278.2745551506,"imp":14.183180672,"elo":1444.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0814556315,"x_train":543.6674489023,"imp":13.8593137304,"elo":1473.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.3191008047,"x_train":841.4117357913,"imp":13.6597843306,"elo":1490.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.5861131346,"x_train":1128.7390334238,"imp":13.5312380654,"elo":1497.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0471622003,"x_train":0.4669717504,"imp":12.9118537073,"elo":1518.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1406300688,"x_train":2.0321466005,"imp":11.8212367436,"elo":1616.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2332870616,"x_train":5.5778533182,"imp":11.3229193666,"elo":1666.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2722786888,"x_train":9.7317062539,"imp":10.9403239771,"elo":1698.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3510039423,"x_train":25.7016530469,"imp":10.628774182,"elo":1717.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3533147712,"x_train":51.0311253263,"imp":10.5022373116,"elo":1730.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3673660669,"x_train":101.303854312,"imp":10.4647468749,"elo":1739.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.417887152,"x_train":150.6411976385,"imp":10.4529428699,"elo":1741.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.4302741489,"x_train":202.5364616477,"imp":10.4820421716,"elo":1742.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":7.8904270487,"x_train":68.7091297861,"imp":19.4673543521,"elo":1350.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":15.1962130277,"x_train":129.9001536616,"imp":17.1822083639,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":26.5841757971,"x_train":315.8079093334,"imp":14.4126589357,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.6468979155,"x_train":621.987734561,"imp":12.5570854498,"elo":1618.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":69.4279053206,"x_train":1537.7988263416,"imp":10.8050873617,"elo":1735.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":70.1635578676,"x_train":1598.1220946317,"imp":10.9608112326,"elo":1732.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.1830265708,"x_train":2.9551878511,"imp":35.1302552748,"elo":871.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.3237118849,"x_train":6.5948259183,"imp":27.689365732,"elo":1016.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.5569193127,"x_train":17.3219967074,"imp":19.5874877747,"elo":1246.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.8277646092,"x_train":35.2109614365,"imp":16.4609945688,"elo":1381.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.1468593548,"x_train":88.816423174,"imp":15.4820902855,"elo":1426.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.4856238345,"x_train":176.8663518766,"imp":14.9281733183,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.8509761365,"x_train":354.8398796964,"imp":14.4325741349,"elo":1447.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.9863302204,"x_train":531.8168061758,"imp":14.0604231438,"elo":1450.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.170390818,"x_train":713.0056913426,"imp":13.6083145675,"elo":1460.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..960f6bae527e78b2b67f349740b680a603f26e1e
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_classification/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "AutoGluon 1.4 (best, 4h)", "TabFM (default)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "XGBoost (tuned + ensembled)", "iLTM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "ChimeraBoost (tuned + ensembled)", "CatBoost (tuned)", "RealTabPFN-2.5 (tuned)", "LightGBM (tuned + ensembled)", "ChimeraBoost (tuned)", "TabM (tuned + ensembled)", "RealMLP (tuned + ensembled)", "CatBoost (default)", "ModernNCA (tuned)", "XGBoost (tuned)", "LightGBM (tuned)", "RealMLP (tuned)", "TabM (tuned)", "ChimeraBoost (default)", "RealTabPFN-2.5 (default)", "TabICL (default)", "iLTM (tuned)", "XGBoost (default)", "EBM (tuned + ensembled)", "TabSwift (default)", "TorchMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "TabM (default)", "xRFM (tuned + ensembled)", "xRFM (tuned)", "RealMLP (default)", "TabDPT (tuned)", "EBM (tuned)", "RandomForest (tuned + ensembled)", "EBM (default)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned)", "BetaTabPFN (default)", "iLTM (default)", "LimiX (default)", "RandomForest (tuned)", "ExtraTrees (tuned + ensembled)", "LightGBM (default)", "ModernNCA (default)", "SAP-RPT-OSS (default)", "ExtraTrees (tuned)", "TabDPT (default)", "FastaiMLP (tuned)", "TorchMLP (default)", "TabSTAR (tuned)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "KNN (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "Linear (tuned)", "RandomForest (default)", "OrionMSP (default)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "TabSTAR (default)", "PerpetualBooster (tuned)", "Linear (default)", "xRFM (default)", "KNN (tuned)", "TabFlex (default)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.6000000238418579, 0.5, 0.5, 0.699999988079071, 0.699999988079071, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4000000059604645, NaN, 0.6000000238418579, 0.5, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5, 0.4000000059604645, NaN, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 1.0, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.5, 0.5, 0.30000001192092896, NaN, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, NaN, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.5, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 1.0, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.699999988079071, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, NaN, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, NaN, 0.4000000059604645, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.5, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.6000000238418579, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.6000000238418579, NaN, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.5, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.5, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.6000000238418579, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.5, 0.5, 0.6000000238418579, NaN, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.5, 0.4000000059604645, 0.5, NaN, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.699999988079071, 0.5, 0.5, 0.699999988079071, 0.44999998807907104, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.5, 0.5, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.5, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.5, 0.4000000059604645, 0.6000000238418579, NaN, 0.4000000059604645, 0.30000001192092896, 0.6000000238418579, 0.6000000238418579, 0.5, 0.699999988079071, 0.5, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.6000000238418579, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.5, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.5, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.5, 0.699999988079071, 0.4000000059604645, NaN, 0.6000000238418579, 0.30000001192092896, 0.5, 0.6000000238418579, 0.5, 0.5, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 1.0, 0.699999988079071, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.5, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.5, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, 0.5, 0.699999988079071, 0.4000000059604645, NaN, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.6000000238418579, 0.6000000238418579, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.5, 0.5, 0.30000001192092896, 0.5, 0.5, 0.5, 0.4000000059604645, 0.4000000059604645, NaN, 0.5, 0.5, 0.5, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.5, 0.699999988079071, 0.5, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, NaN, 0.5, 0.4000000059604645, 0.5, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.5, 0.4000000059604645, 0.550000011920929, 0.4000000059604645, 0.5, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.5, 0.5, NaN, 0.5, 0.5, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.5, 0.6000000238418579, 0.5, NaN, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, 0.5, 0.5, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, NaN, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.699999988079071, 1.0, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.5, 0.800000011920929, 0.5, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 1.0, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.30000001192092896, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.5, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 1.0, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.5, 0.5, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.699999988079071, NaN, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 1.0, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.5, 0.6000000238418579, NaN, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 1.0, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.5, 0.30000001192092896, NaN, 0.5, 0.6000000238418579, 0.5, 0.5, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, 0.30000001192092896, 0.5, NaN, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.5, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, NaN, 0.5, 0.5, 0.5, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.5, NaN, 0.5, 0.5, 0.5, 0.6000000238418579, 0.699999988079071, 0.5, 0.5, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 1.0, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0], [0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.5, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.5, 0.4000000059604645, 0.5, 0.20000000298023224, 0.5, 0.4000000059604645, 0.5, 0.5, NaN, 0.5, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.5, NaN, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.5, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.5, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.6000000238418579, NaN, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.699999988079071, 0.6000000238418579, NaN, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.0, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, 0.5, 0.5, NaN, 0.5, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.699999988079071, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.10000000149011612, 0.4000000059604645, 0.20000000298023224, 0.10000000149011612, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.5, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.5, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, NaN, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.5, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.0, 0.4000000059604645, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.5, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.0, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.6000000238418579, 0.4000000059604645, NaN, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.699999988079071, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.8500000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.10000000149011612, 0.4000000059604645, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.10000000149011612, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, NaN, 0.5, 0.5, 0.5, 0.5, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.0, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4000000059604645, 0.6000000238418579, 0.5, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.4000000059604645, NaN, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 1.0, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.5, 0.6000000238418579, 0.5, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.5, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.0, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, NaN, 0.5, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.5, 0.5, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.5, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.5, NaN, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.800000011920929, 1.0], [0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.5, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.5, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.8999999761581421, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.5, 0.6000000238418579, 0.6000000238418579, NaN, 0.5, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.5, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.6000000238418579, 0.6000000238418579, 0.5, NaN, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.0, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.5, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.5, 0.6000000238418579, 0.4000000059604645, 0.5, NaN, 0.5, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.5, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.5, NaN, 0.5, 0.5, 0.5, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.6000000238418579, 0.30000001192092896, 0.5, 0.5, 0.5, NaN, 0.699999988079071, 0.5, 0.699999988079071, 0.6000000238418579, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.5, 0.30000001192092896, 0.5, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.5, 0.30000001192092896, NaN, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.5, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.15000000596046448, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.5, 0.4000000059604645, 0.5, 0.5, 0.5, 0.4000000059604645, NaN, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.5, 0.699999988079071, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.8999999761581421, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.5, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.5, 0.4000000059604645, NaN, 0.5, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.8999999761581421, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.0, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.10000000149011612, 0.4000000059604645, 0.5, 0.30000001192092896, 0.5, NaN, 0.699999988079071, 0.5, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 0.8999999761581421, 1.0, 0.8999999761581421, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.5, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, NaN, 0.5, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.5, 0.20000000298023224, 0.10000000149011612, 0.5, 0.5, NaN, 0.6000000238418579, 0.5, 0.5, 0.6000000238418579, 0.5, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929, 0.8999999761581421, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.0, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, NaN, 0.30000001192092896, 0.4000000059604645, 0.8500000238418579, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.0, 0.4000000059604645, 0.5, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.5, 0.699999988079071, NaN, 0.4000000059604645, 0.699999988079071, 0.699999988079071, 0.5, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.5, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.10000000149011612, 0.5, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.800000011920929, 0.699999988079071, 1.0, 0.800000011920929, 0.800000011920929, 0.699999988079071, 0.8999999761581421, 1.0, 0.800000011920929, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.0, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.0, 0.0, 0.0, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.10000000149011612, 0.30000001192092896, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.15000000596046448, 0.30000001192092896, 0.4000000059604645, NaN, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.699999988079071, 0.699999988079071, 0.8999999761581421], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.4000000059604645, 0.30000001192092896, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.30000001192092896, 0.4000000059604645, 0.5, NaN, 0.6000000238418579, 0.5, 0.800000011920929, 0.5, 0.6000000238418579, 0.6000000238418579, 0.8999999761581421, 0.5, 0.6000000238418579, 0.800000011920929, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.4000000059604645, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.5, 0.30000001192092896, 0.5, 0.4000000059604645, NaN, 0.5, 0.699999988079071, 0.6000000238418579, 0.5, 0.6000000238418579, 0.800000011920929, 0.5, 0.699999988079071, 0.699999988079071, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.5, 0.4000000059604645, 0.20000000298023224, 0.5, 0.5, 0.5, NaN, 0.6000000238418579, 0.20000000298023224, 1.0, 0.5, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.5, 0.30000001192092896, 0.30000001192092896, 0.5, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, NaN, 0.30000001192092896, 0.6000000238418579, 0.5, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.30000001192092896, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.10000000149011612, 0.0, 0.30000001192092896, 0.5, 0.4000000059604645, 0.0, 0.5, 0.5, 0.4000000059604645, 0.800000011920929, 0.699999988079071, NaN, 0.800000011920929, 0.5, 0.699999988079071, 0.699999988079071, 0.699999988079071, 0.800000011920929, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.5, 0.30000001192092896, 0.20000000298023224, 0.5, 0.4000000059604645, 0.5, 0.0, 0.4000000059604645, 0.20000000298023224, NaN, 0.5, 0.6000000238418579, 0.30000001192092896, 0.4000000059604645, 0.8999999761581421, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.5, 0.5, 0.5, 0.5, NaN, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.8999999761581421], [0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.0, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.20000000298023224, 0.10000000149011612, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.30000001192092896, 0.4000000059604645, 0.20000000298023224, NaN, 0.4000000059604645, 0.5, 0.5, 0.699999988079071], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.30000001192092896, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.30000001192092896, 0.10000000149011612, 0.0, 0.10000000149011612, 0.20000000298023224, 0.4000000059604645, 0.0, 0.10000000149011612, 0.4000000059604645, 0.5, 0.5, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.699999988079071, 0.4000000059604645, 0.6000000238418579, NaN, 0.5, 0.800000011920929, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.30000001192092896, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.10000000149011612, 0.30000001192092896, 0.20000000298023224, 0.0, 0.30000001192092896, 0.4000000059604645, 0.30000001192092896, 0.4000000059604645, 0.4000000059604645, 0.30000001192092896, 0.6000000238418579, 0.4000000059604645, 0.5, 0.5, NaN, 0.6000000238418579, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.10000000149011612, 0.0, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.10000000149011612, 0.20000000298023224, 0.30000001192092896, 0.20000000298023224, 0.30000001192092896, 0.0, 0.20000000298023224, 0.20000000298023224, 0.10000000149011612, 0.30000001192092896, 0.5, 0.20000000298023224, 0.4000000059604645, NaN, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.0, 0.10000000149011612, 0.10000000149011612, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.10000000149011612, 0.30000001192092896, 0.0, 0.20000000298023224, 0.20000000298023224, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9602739697},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9424657507},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9082191754},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8712328721},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8602739703},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.828767121},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7917808186},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.771232877},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7657534241},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7616438359},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7513698623},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7506849291},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7479452023},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.746575342},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7397260297},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.7369862994},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7328767118},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7301369856},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.7273972572},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.7089041096},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.7054794523},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6849315072},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6630136997},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6616438383},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.6534246581},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6452054803},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6301369878},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.6000000027},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.5904109585},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5835616412},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5753424685},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.572602744},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.5712328801},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5424657569},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5369863026},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5356164383},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5246575362},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4917808234},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4815068543},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4780821947},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4671232903},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4630137013},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4575342483},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4397260284},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.438356166},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.4383561663},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4356164416},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4273972607},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4232876751},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3972602748},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3849315062},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3753424673},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3712328801},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.3431506868},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3136986311},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.3000000003},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2452054812},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2438356182},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2363013735},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2342465766},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2315068513},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2116438384},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2068493182},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.1808219196},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1684931531},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1465753453},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1452054803},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1328767138},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1328767137},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.130136988},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1136986315},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1000000022},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0808219193},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0205479456}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..22da9d5a886c4b239dfb1843d70bd4b90bb8b509
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,"AutoGluon 1.5 (extreme, 4h)","AutoGluon 1.4 (best, 4h)",TabFM (default),TabPFN-3 (default),TabPFN-2.6 (default),TabICLv2 (default),RealTabPFN-2.5 (tuned + ensembled),ModernNCA (tuned + ensembled),XGBoost (tuned + ensembled),iLTM (tuned + ensembled),CatBoost (tuned + ensembled),ChimeraBoost (tuned + ensembled),CatBoost (tuned),RealTabPFN-2.5 (tuned),LightGBM (tuned + ensembled),ChimeraBoost (tuned),TabM (tuned + ensembled),RealMLP (tuned + ensembled),CatBoost (default),ModernNCA (tuned),XGBoost (tuned),LightGBM (tuned),RealMLP (tuned),TabM (tuned),ChimeraBoost (default),RealTabPFN-2.5 (default),TabICL (default),iLTM (tuned),XGBoost (default),EBM (tuned + ensembled),TabSwift (default),TorchMLP (tuned + ensembled),TabDPT-Turbo (default),TabDPT (tuned + ensembled),TabM (default),xRFM (tuned + ensembled),xRFM (tuned),RealMLP (default),TabDPT (tuned),EBM (tuned),RandomForest (tuned + ensembled),EBM (default),FastaiMLP (tuned + ensembled),TorchMLP (tuned),BetaTabPFN (default),iLTM (default),LimiX (default),RandomForest (tuned),ExtraTrees (tuned + ensembled),LightGBM (default),ModernNCA (default),SAP-RPT-OSS (default),ExtraTrees (tuned),TabDPT (default),FastaiMLP (tuned),TorchMLP (default),TabSTAR (tuned),FastaiMLP (default),Linear (tuned + ensembled),KNN (tuned + ensembled),TabSTAR (tuned + ensembled),Linear (tuned),RandomForest (default),OrionMSP (default),PerpetualBooster (tuned + ensembled),ExtraTrees (default),TabSTAR (default),PerpetualBooster (tuned),Linear (default),xRFM (default),KNN (tuned),TabFlex (default),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.6,0.5,0.5,0.7,0.7,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,0.9,1.0,1.0,0.9,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.4,,0.6,0.5,0.8,0.7,0.7,0.7,1.0,1.0,1.0,0.9,1.0,0.9,0.9,0.9,0.8,1.0,0.9,0.9,1.0,1.0,1.0,0.8,1.0,0.8,0.9,1.0,0.9,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.5,0.4,,0.7,0.8,0.8,0.8,0.8,0.8,0.7,0.8,0.8,0.8,0.9,0.9,0.8,1.0,0.8,0.8,0.8,0.7,1.0,0.8,1.0,0.9,0.9,0.8,0.9,0.9,1.0,0.8,0.9,1.0,0.9,1.0,0.9,0.9,1.0,0.9,1.0,1.0,1.0,0.9,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.5,0.5,0.3,,0.6,0.7,0.6,0.9,0.7,0.7,0.7,0.8,0.7,0.8,0.8,0.8,0.8,0.7,0.7,0.8,0.7,1.0,0.7,0.9,0.8,0.9,0.8,0.9,0.9,0.9,0.9,0.7,1.0,0.9,1.0,0.8,0.8,1.0,0.9,1.0,0.9,1.0,0.9,1.0,1.0,1.0,0.8,1.0,0.9,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.3,0.2,0.2,0.4,,0.7,0.6,0.8,0.5,0.6,0.7,0.8,0.7,0.7,0.9,0.8,0.8,0.8,0.7,0.7,0.7,1.0,0.7,0.9,0.7,0.9,0.9,0.8,0.9,1.0,0.8,0.8,1.0,1.0,1.0,0.7,0.8,1.0,0.9,1.0,1.0,1.0,0.9,1.0,0.9,1.0,0.8,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.3,0.3,0.2,0.3,0.3,,0.7,0.7,0.7,0.7,0.7,0.7,0.7,0.8,0.8,0.7,0.6,0.6,0.7,0.8,0.7,0.9,0.7,0.6,0.7,0.8,0.8,0.8,0.7,0.7,0.8,0.8,1.0,0.9,1.0,0.7,0.7,0.9,0.9,1.0,1.0,1.0,0.9,0.8,1.0,0.9,0.8,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,0.3,0.2,0.4,0.4,0.3,,0.4,0.6,0.5,0.7,0.6,0.6,0.9,0.6,0.7,0.5,0.6,0.7,0.7,0.7,0.8,0.6,0.7,0.7,0.8,0.8,0.7,0.8,0.9,0.9,0.7,0.8,0.8,0.9,0.7,0.7,0.9,0.9,0.9,0.9,0.9,0.8,0.9,1.0,0.9,0.6,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.3,0.2,0.1,0.2,0.3,0.6,,0.6,0.5,0.5,0.6,0.5,0.6,0.5,0.6,0.6,0.6,0.6,0.7,0.6,0.6,0.8,0.7,0.8,0.8,0.8,0.7,0.6,0.8,0.7,0.8,0.7,1.0,0.8,0.9,0.8,0.9,0.9,0.8,0.8,0.9,1.0,0.9,0.9,0.9,0.8,0.8,0.9,1.0,1.0,0.9,0.9,1.0,1.0,0.9,0.9,1.0,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.2,0.3,0.5,0.3,0.4,0.4,,0.4,0.6,0.5,0.5,0.6,0.6,0.6,0.7,0.7,0.7,0.5,0.8,0.6,0.6,0.7,0.7,0.7,0.7,0.7,0.9,0.8,0.6,0.9,0.6,0.8,0.9,0.7,0.8,0.9,0.7,0.8,0.9,1.0,0.8,0.9,0.8,0.9,0.6,0.9,1.0,1.0,1.0,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.3,0.3,0.4,0.3,0.5,0.5,0.6,,0.5,0.6,0.5,0.6,0.4,0.6,0.6,0.5,0.6,0.6,0.7,0.5,0.7,0.6,0.7,0.8,0.7,0.7,0.7,0.7,0.7,0.9,0.6,0.7,0.8,0.9,0.8,0.9,0.8,0.6,0.9,0.8,0.9,1.0,0.8,1.0,0.8,0.9,0.9,0.9,0.9,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.2,0.3,0.3,0.3,0.3,0.5,0.4,0.5,,0.6,0.6,0.4,0.4,0.7,0.5,0.5,0.7,0.45,0.6,0.5,0.7,0.6,0.8,0.8,0.7,0.9,0.8,0.6,0.7,0.8,0.6,0.7,0.7,0.9,1.0,0.8,0.8,0.6,0.9,0.9,0.9,1.0,0.8,0.9,0.7,0.9,1.0,0.9,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.1,0.2,0.2,0.2,0.3,0.4,0.4,0.5,0.4,0.4,,0.4,0.6,0.5,0.6,0.6,0.7,0.6,0.6,0.5,0.5,0.8,0.6,0.8,0.7,0.8,0.8,0.7,0.8,0.5,0.8,0.6,0.9,0.8,0.9,1.0,0.9,0.8,0.9,0.9,1.0,0.9,0.9,0.8,0.9,0.7,0.9,0.9,0.9,0.9,1.0,0.9,0.9,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.2,0.3,0.3,0.3,0.4,0.5,0.5,0.5,0.4,0.6,,0.4,0.3,0.6,0.6,0.5,0.7,0.5,0.7,0.5,0.7,0.6,0.8,0.7,0.7,0.9,0.8,0.7,0.7,0.8,0.6,0.7,0.8,0.9,0.9,0.9,0.7,0.6,0.9,0.9,0.9,1.0,0.8,0.9,0.7,0.9,0.9,0.9,0.9,1.0,0.9,0.9,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.1,0.1,0.2,0.3,0.2,0.1,0.4,0.4,0.4,0.6,0.4,0.6,,0.6,0.6,0.5,0.5,0.7,0.6,0.6,0.8,0.6,0.7,0.6,0.7,0.8,0.7,0.8,0.9,0.8,0.7,0.8,0.8,0.9,0.6,0.7,0.9,0.9,0.9,0.9,0.9,0.8,0.9,0.9,0.8,0.6,1.0,1.0,1.0,0.9,0.9,0.8,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.1,0.1,0.2,0.1,0.2,0.4,0.5,0.4,0.6,0.6,0.5,0.7,0.4,,0.6,0.3,0.5,0.6,0.5,0.5,0.8,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.7,0.6,0.7,0.8,0.6,0.8,0.7,0.8,0.8,0.7,1.0,0.9,1.0,0.8,0.9,0.8,1.0,0.7,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.1,0.2,0.2,0.2,0.3,0.3,0.4,0.4,0.4,0.3,0.4,0.4,0.4,0.4,,0.6,0.6,0.6,0.7,0.5,0.4,0.8,0.6,0.7,0.7,0.8,0.8,0.7,0.9,0.6,0.8,0.6,0.9,0.8,0.9,1.0,0.9,0.8,0.9,0.9,1.0,0.9,0.9,0.9,0.8,0.7,0.9,0.9,0.9,0.9,1.0,0.8,0.8,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.1,0.2,0.0,0.2,0.2,0.4,0.5,0.4,0.3,0.4,0.5,0.4,0.4,0.5,0.7,0.4,,0.6,0.5,0.7,0.6,0.6,0.7,0.6,0.5,0.6,0.6,0.6,0.6,0.9,0.7,0.7,0.8,0.7,1.0,0.6,0.6,1.0,0.9,0.9,0.9,0.9,0.9,0.8,0.9,0.7,0.7,0.9,0.8,1.0,1.0,0.9,0.8,0.9,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.2,0.3,0.2,0.4,0.4,0.4,0.3,0.5,0.5,0.3,0.5,0.5,0.5,0.4,0.4,,0.5,0.5,0.5,0.6,0.8,0.6,0.5,0.7,0.5,0.7,0.7,0.7,0.6,0.8,0.7,0.7,0.9,0.8,0.9,1.0,0.7,0.8,0.9,0.8,1.0,0.9,0.8,0.8,0.8,0.9,0.9,1.0,0.9,1.0,0.8,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.1,0.2,0.3,0.3,0.3,0.3,0.4,0.3,0.4,0.3,0.4,0.3,0.3,0.4,0.4,0.5,0.5,,0.5,0.4,0.5,0.7,0.6,0.7,0.8,0.7,0.8,0.7,0.7,0.7,0.9,0.6,0.7,0.7,0.9,0.9,0.9,0.7,0.7,0.9,0.9,0.9,1.0,0.8,0.9,0.7,0.9,0.9,0.9,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.1,0.1,0.2,0.2,0.3,0.2,0.3,0.3,0.5,0.4,0.55,0.4,0.5,0.4,0.5,0.3,0.3,0.5,0.5,,0.5,0.5,0.6,0.4,0.6,0.7,0.7,0.6,0.5,0.6,0.8,0.8,0.7,0.8,0.7,0.6,0.8,0.8,1.0,0.7,0.8,0.8,0.8,0.8,0.9,0.9,0.8,0.9,0.9,0.9,1.0,0.9,0.9,1.0,1.0,0.9,0.9,1.0,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.3,0.3,0.3,0.3,0.3,0.4,0.2,0.3,0.4,0.5,0.3,0.4,0.5,0.5,0.4,0.5,0.6,0.5,,0.6,0.5,0.6,0.6,0.5,0.6,0.6,0.9,0.6,0.6,0.7,0.6,0.9,0.8,0.8,0.7,0.8,0.7,0.8,0.8,0.9,0.7,0.9,0.9,0.9,0.6,0.8,0.9,0.9,1.0,0.9,0.9,0.9,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.1,0.2,0.4,0.4,0.5,0.5,0.5,0.5,0.2,0.2,0.6,0.4,0.4,0.5,0.5,0.4,,0.5,0.6,0.6,0.6,0.5,0.6,0.7,0.7,0.4,0.6,0.8,0.7,0.8,0.7,0.8,0.8,0.7,0.9,0.8,0.9,0.8,0.8,0.8,1.0,0.7,1.0,0.8,1.0,1.0,0.9,0.9,1.0,0.9,0.9,1.0,0.9,0.8,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.2,0.3,0.3,0.3,0.4,0.2,0.4,0.3,0.3,0.2,0.3,0.4,0.4,0.2,0.3,0.2,0.3,0.4,0.5,0.5,,0.4,0.4,0.6,0.5,0.7,0.7,0.6,0.5,0.8,0.5,0.7,0.7,0.9,0.7,1.0,0.7,0.7,0.8,0.7,0.9,0.9,0.8,0.7,0.8,0.8,0.7,0.9,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.1,0.2,0.0,0.1,0.1,0.4,0.3,0.3,0.3,0.4,0.4,0.4,0.4,0.3,0.4,0.4,0.4,0.4,0.4,0.6,0.4,0.4,0.6,,0.3,0.5,0.6,0.5,0.6,0.7,0.6,0.6,0.7,0.6,0.8,0.5,0.6,0.8,0.8,0.9,0.7,0.8,0.8,0.8,0.9,0.6,0.7,0.8,0.7,0.8,1.0,0.9,0.7,0.8,0.9,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.1,0.2,0.3,0.3,0.3,0.2,0.3,0.3,0.2,0.2,0.2,0.4,0.4,0.3,0.5,0.5,0.3,0.4,0.4,0.4,0.6,0.7,,0.4,0.5,0.6,0.6,0.7,0.5,0.7,0.6,0.8,0.6,0.9,0.8,0.7,0.7,0.8,0.8,0.8,0.8,0.9,0.7,0.7,0.7,0.8,0.7,0.7,0.8,1.0,0.7,0.8,0.9,0.9,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,0.9,1.0
+RealTabPFN-2.5 (default),0.0,0.2,0.1,0.1,0.1,0.2,0.2,0.2,0.3,0.2,0.2,0.3,0.3,0.3,0.4,0.3,0.4,0.3,0.2,0.3,0.5,0.4,0.4,0.5,0.6,,0.7,0.7,0.7,0.6,0.8,0.6,0.6,0.8,0.7,0.7,0.6,0.7,0.7,0.8,0.7,0.8,0.7,0.8,1.0,0.7,0.6,0.8,0.8,1.0,0.9,0.9,0.8,0.8,0.9,1.0,1.0,0.8,0.9,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0
+TabICL (default),0.1,0.1,0.2,0.2,0.1,0.2,0.2,0.2,0.3,0.3,0.3,0.2,0.3,0.2,0.4,0.2,0.4,0.5,0.3,0.3,0.4,0.5,0.5,0.4,0.5,0.3,,0.5,0.6,0.5,0.5,0.7,0.7,0.9,0.6,0.6,0.7,0.7,0.9,0.6,0.8,0.6,0.7,0.7,0.7,0.8,0.7,0.8,0.8,0.9,0.9,0.8,1.0,1.0,0.8,0.8,0.9,0.8,0.8,0.9,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+iLTM (tuned),0.1,0.0,0.1,0.1,0.2,0.2,0.3,0.3,0.3,0.3,0.1,0.2,0.1,0.3,0.2,0.2,0.4,0.3,0.2,0.4,0.4,0.4,0.3,0.5,0.4,0.3,0.5,,0.4,0.5,0.6,0.6,0.4,0.6,0.6,0.7,0.6,0.6,0.6,0.6,0.7,0.6,0.8,0.7,0.7,0.7,0.7,0.8,0.6,0.8,0.8,0.9,0.9,0.9,0.8,0.9,1.0,0.9,0.9,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.1,0.1,0.1,0.1,0.1,0.3,0.2,0.4,0.1,0.3,0.2,0.3,0.2,0.2,0.2,0.3,0.4,0.3,0.3,0.5,0.1,0.3,0.3,0.4,0.4,0.3,0.4,0.6,,0.5,0.5,0.5,0.5,0.6,0.5,0.7,0.7,0.5,0.7,0.6,0.8,0.7,0.7,0.6,0.6,0.9,0.7,0.8,0.9,0.6,0.8,0.9,0.9,0.8,0.7,0.9,1.0,0.8,0.9,0.9,1.0,0.9,0.9,1.0,0.9,0.9,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.1,0.0,0.3,0.1,0.2,0.2,0.3,0.4,0.2,0.3,0.1,0.3,0.1,0.1,0.3,0.3,0.4,0.4,0.3,0.4,0.3,0.3,0.4,0.5,0.5,0.5,,0.5,0.5,0.5,0.6,0.7,0.5,0.5,0.7,0.7,0.9,0.7,1.0,0.7,0.8,0.8,0.6,0.6,0.8,0.7,0.7,0.9,0.9,0.7,0.7,0.8,0.9,1.0,0.9,0.9,0.9,1.0,0.9,0.9,0.9,0.9,0.9,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0
+TabSwift (default),0.1,0.0,0.2,0.1,0.2,0.2,0.1,0.3,0.4,0.3,0.3,0.5,0.3,0.2,0.4,0.4,0.3,0.4,0.3,0.2,0.4,0.6,0.5,0.4,0.5,0.2,0.5,0.4,0.5,0.5,,0.5,0.4,0.6,0.6,0.7,0.6,0.5,0.6,0.6,0.6,0.7,0.7,0.6,0.7,0.7,0.7,0.6,0.6,0.8,0.7,0.7,0.7,0.9,0.7,0.7,0.8,0.7,0.7,0.8,0.8,0.8,0.9,1.0,1.0,1.0,0.9,1.0,0.8,0.9,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.1,0.3,0.2,0.2,0.3,0.2,0.1,0.1,0.2,0.2,0.2,0.3,0.3,0.2,0.3,0.2,0.1,0.2,0.3,0.4,0.2,0.4,0.3,0.4,0.3,0.4,0.5,0.5,0.5,,0.4,0.3,0.4,0.6,0.6,0.6,0.7,0.6,0.8,0.5,0.6,0.8,0.6,0.7,0.6,0.8,0.8,0.6,0.7,0.9,0.8,0.8,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.1,0.0,0.0,0.0,0.0,0.2,0.3,0.4,0.4,0.4,0.4,0.4,0.2,0.2,0.4,0.2,0.3,0.4,0.3,0.4,0.2,0.5,0.3,0.4,0.4,0.3,0.6,0.5,0.5,0.6,0.6,,0.4,0.5,0.6,0.6,0.6,0.7,0.6,0.6,0.6,0.6,0.8,0.6,0.7,0.8,0.8,0.7,0.7,0.6,0.7,0.7,0.9,0.9,0.8,0.8,0.9,0.7,0.8,0.8,0.7,1.0,0.9,1.0,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0
+TabDPT (tuned + ensembled),0.0,0.0,0.1,0.1,0.0,0.1,0.2,0.0,0.2,0.3,0.3,0.1,0.3,0.2,0.4,0.1,0.3,0.3,0.3,0.2,0.1,0.3,0.3,0.4,0.2,0.2,0.1,0.4,0.4,0.4,0.4,0.7,0.6,,0.5,0.6,0.6,0.7,0.7,0.6,0.6,0.6,0.6,0.8,0.6,0.5,0.6,0.6,0.6,0.8,0.9,0.7,0.7,0.9,0.9,0.8,0.8,0.9,0.8,0.9,0.8,0.8,1.0,0.9,1.0,1.0,1.0,1.0,1.0,0.9,1.0,0.9,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.2,0.1,0.2,0.3,0.2,0.2,0.1,0.2,0.2,0.0,0.1,0.3,0.3,0.2,0.2,0.3,0.2,0.4,0.3,0.4,0.4,0.5,0.3,0.4,0.6,0.5,0.5,,0.5,0.5,0.7,0.6,0.6,0.7,0.7,0.8,0.6,0.8,0.7,0.6,0.7,0.6,0.8,0.8,0.7,0.7,0.7,0.9,0.9,1.0,0.9,0.8,0.9,1.0,0.7,0.8,1.0,1.0,1.0,1.0,1.0,0.9,0.9,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.1,0.2,0.3,0.3,0.3,0.1,0.3,0.1,0.1,0.1,0.1,0.4,0.3,0.1,0.4,0.2,0.1,0.4,0.2,0.3,0.1,0.5,0.1,0.3,0.4,0.3,0.3,0.5,0.3,0.4,0.4,0.4,0.5,,0.5,0.6,0.5,0.7,0.7,0.6,0.7,0.5,0.7,0.6,0.7,0.7,0.7,0.5,0.7,0.7,0.7,0.6,0.8,0.8,0.9,0.9,1.0,1.0,0.9,1.0,0.9,1.0,0.9,1.0,0.9,0.9,1.0,1.0,1.0,1.0,0.9,1.0
+xRFM (tuned),0.0,0.0,0.1,0.2,0.2,0.3,0.3,0.2,0.2,0.2,0.0,0.0,0.1,0.3,0.2,0.0,0.4,0.1,0.1,0.2,0.3,0.2,0.3,0.4,0.2,0.4,0.3,0.4,0.3,0.5,0.4,0.4,0.4,0.4,0.5,0.5,,0.6,0.4,0.6,0.6,0.5,0.6,0.7,0.6,0.5,0.6,0.7,0.7,0.6,0.6,0.9,0.7,0.6,0.8,0.9,0.9,0.9,1.0,1.0,0.9,1.0,0.8,0.9,0.9,1.0,1.0,0.9,1.0,1.0,1.0,1.0,0.9,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.2,0.1,0.1,0.1,0.2,0.1,0.0,0.0,0.1,0.2,0.2,0.2,0.0,0.2,0.3,0.3,0.3,0.4,0.5,0.3,0.5,0.4,0.4,0.3,0.3,0.4,0.4,,0.6,0.6,0.6,0.6,0.5,0.7,0.6,0.5,0.5,0.7,0.6,0.6,0.7,0.7,0.7,0.7,1.0,0.9,0.9,1.0,0.9,0.9,0.9,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.1,0.0,0.1,0.1,0.1,0.1,0.1,0.1,0.3,0.2,0.2,0.2,0.3,0.1,0.3,0.2,0.1,0.3,0.3,0.0,0.3,0.3,0.3,0.2,0.3,0.3,0.1,0.4,0.3,0.3,0.4,0.3,0.3,0.3,0.4,0.5,0.6,0.4,,0.5,0.4,0.4,0.4,0.7,0.4,0.6,0.6,0.6,0.5,0.6,0.5,0.6,0.7,0.85,0.8,0.8,0.8,0.9,0.7,0.8,0.8,0.7,0.8,0.8,1.0,1.0,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.2,0.2,0.4,0.4,0.1,0.4,0.1,0.0,0.1,0.1,0.2,0.3,0.3,0.2,0.1,0.3,0.1,0.2,0.2,0.4,0.4,0.4,0.1,0.4,0.4,0.4,0.4,0.4,0.3,0.4,0.4,0.5,,0.5,0.5,0.5,0.5,0.5,0.6,0.6,0.6,0.6,0.6,0.6,0.9,0.7,0.6,0.8,0.8,1.0,0.8,0.7,0.9,1.0,0.8,0.8,0.9,0.9,0.8,1.0,0.9,0.9,0.8,1.0,1.0,0.9,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.2,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.2,0.2,0.2,0.2,0.3,0.2,0.3,0.2,0.3,0.2,0.3,0.4,0.2,0.4,0.4,0.3,0.3,0.4,0.4,0.6,0.5,,0.4,0.4,0.6,0.6,0.7,0.6,0.6,0.6,0.6,0.6,0.7,0.7,0.6,0.8,0.8,1.0,0.9,0.8,0.9,1.0,0.8,0.9,0.9,0.9,1.0,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.2,0.1,0.0,0.1,0.1,0.0,0.0,0.1,0.2,0.1,0.2,0.1,0.1,0.3,0.2,0.2,0.2,0.4,0.4,0.3,0.0,0.3,0.5,0.4,0.4,0.3,0.4,0.5,0.4,0.6,0.5,0.6,,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.7,0.8,0.8,0.7,0.7,0.8,0.8,0.9,0.8,0.7,0.9,0.9,0.8,0.8,0.9,0.9,0.8,1.0,0.9,0.9,0.8,1.0,1.0,0.9,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.1,0.1,0.1,0.1,0.2,0.0,0.2,0.1,0.1,0.1,0.1,0.2,0.2,0.1,0.1,0.0,0.1,0.2,0.3,0.2,0.1,0.2,0.2,0.3,0.3,0.2,0.3,0.3,0.3,0.4,0.4,0.4,0.2,0.3,0.4,0.5,0.6,0.5,0.6,0.4,,0.6,0.4,0.4,0.6,0.6,0.6,0.6,0.7,0.7,0.6,0.6,0.8,0.7,0.7,1.0,0.9,0.9,0.7,0.9,0.8,0.7,0.9,0.9,0.9,0.9,1.0,1.0,0.9,0.9,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.2,0.1,0.1,0.1,0.0,0.0,0.1,0.0,0.1,0.1,0.1,0.2,0.1,0.0,0.2,0.1,0.2,0.1,0.2,0.1,0.2,0.3,0.3,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.5,0.3,0.3,0.3,0.5,0.4,0.4,0.4,,0.5,0.6,0.5,0.4,0.6,0.6,0.8,0.7,0.7,0.6,0.5,0.9,0.9,0.8,0.9,0.8,0.9,0.9,0.8,0.9,1.0,0.9,1.0,1.0,1.0,0.9,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.2,0.2,0.2,0.2,0.2,0.1,0.2,0.1,0.1,0.2,0.2,0.1,0.1,0.2,0.2,0.1,0.3,0.0,0.3,0.3,0.4,0.2,0.3,0.4,0.4,0.4,0.2,0.3,0.4,0.4,0.6,0.5,0.4,0.4,0.6,0.5,,0.5,0.6,0.5,0.5,0.6,0.5,0.7,0.5,0.5,0.8,0.7,0.8,0.8,0.7,0.9,0.8,0.7,0.7,0.9,0.9,0.8,1.0,0.9,0.9,0.8,1.0,1.0,0.9,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.1,0.0,0.1,0.1,0.1,0.2,0.0,0.2,0.3,0.2,0.1,0.1,0.1,0.0,0.3,0.4,0.3,0.3,0.2,0.3,0.1,0.4,0.3,0.3,0.3,0.5,0.3,0.4,0.5,0.5,0.4,0.4,0.3,0.4,0.6,0.4,0.5,,0.6,0.4,0.4,0.5,0.6,0.6,0.7,0.7,0.7,0.7,0.9,0.8,0.8,0.9,0.9,0.8,0.8,0.9,0.8,0.9,1.0,0.8,0.9,0.9,0.9,1.0,0.8,1.0
+LimiX (default),0.1,0.1,0.2,0.2,0.2,0.2,0.4,0.2,0.4,0.2,0.3,0.3,0.3,0.4,0.3,0.3,0.3,0.2,0.3,0.2,0.4,0.3,0.2,0.3,0.3,0.4,0.3,0.3,0.3,0.4,0.3,0.4,0.2,0.4,0.4,0.3,0.4,0.5,0.4,0.4,0.4,0.4,0.4,0.5,0.4,0.4,,0.4,0.4,0.4,0.4,0.4,0.5,0.7,0.6,0.6,0.5,0.7,0.7,0.7,0.5,0.7,0.8,0.6,0.7,0.9,0.7,0.7,0.7,0.9,0.7,0.7,0.8,0.9
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.1,0.1,0.1,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.4,0.3,0.3,0.3,0.3,0.4,0.4,0.4,0.4,0.4,0.6,0.5,0.6,0.6,,0.5,0.6,0.6,0.7,0.6,0.5,0.7,0.8,0.9,0.9,0.7,0.9,0.9,0.7,0.9,1.0,0.9,1.0,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.1,0.1,0.0,0.0,0.1,0.2,0.1,0.1,0.1,0.1,0.2,0.3,0.3,0.3,0.2,0.2,0.4,0.1,0.3,0.4,0.2,0.3,0.4,0.4,0.3,0.3,0.4,0.5,0.4,0.4,0.4,0.4,0.4,0.5,0.6,0.6,0.5,,0.5,0.6,0.5,0.7,0.6,0.6,0.6,0.8,0.7,0.7,0.8,0.8,0.9,1.0,1.0,0.8,1.0,0.8,0.8,1.0,0.9,0.9,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.1,0.0,0.0,0.1,0.0,0.0,0.1,0.1,0.1,0.0,0.1,0.2,0.3,0.0,0.1,0.2,0.4,0.3,0.2,0.4,0.3,0.2,0.2,0.5,0.4,0.4,0.4,0.4,0.4,0.3,0.4,0.4,0.4,0.5,0.6,0.4,0.5,,0.5,0.5,0.6,0.5,0.6,0.8,0.8,0.7,0.7,0.8,0.9,0.7,0.8,0.9,0.9,0.9,1.0,1.0,0.9,0.9,1.0,0.9,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.0,0.1,0.0,0.1,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.1,0.1,0.2,0.2,0.1,0.3,0.3,0.4,0.1,0.2,0.3,0.4,0.3,0.5,0.4,0.4,0.2,0.3,0.2,0.5,0.4,0.6,0.4,0.4,0.5,,0.5,0.5,0.5,0.6,0.9,0.8,0.9,0.7,0.9,0.8,0.7,0.9,0.9,0.9,1.0,1.0,0.9,0.9,0.9,1.0,0.9,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.0,0.1,0.1,0.1,0.2,0.1,0.0,0.1,0.2,0.1,0.1,0.1,0.3,0.1,0.3,0.3,0.3,0.3,0.1,0.3,0.4,0.1,0.3,0.2,0.3,0.3,0.3,0.4,0.6,0.3,0.5,0.5,0.5,,0.7,0.5,0.7,0.6,0.8,0.9,0.8,1.0,0.8,0.8,0.7,0.8,0.9,0.9,1.0,0.9,1.0,0.9,1.0,1.0,0.9,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.1,0.1,0.2,0.0,0.2,0.2,0.2,0.1,0.1,0.1,0.1,0.2,0.3,0.3,0.2,0.0,0.1,0.1,0.3,0.3,0.2,0.3,0.3,0.3,0.3,0.3,0.3,0.3,0.3,0.3,0.3,0.4,0.3,0.5,0.3,0.5,0.4,0.3,0.4,0.5,0.3,,0.6,0.5,0.5,0.6,0.5,0.7,0.6,0.6,0.7,0.9,0.9,0.8,0.9,0.8,0.9,1.0,0.9,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.1,0.0,0.0,0.2,0.1,0.1,0.1,0.0,0.1,0.0,0.2,0.2,0.2,0.2,0.0,0.1,0.2,0.3,0.1,0.2,0.1,0.1,0.3,0.4,0.4,0.3,0.15,0.4,0.4,0.3,0.4,0.4,0.5,0.3,0.3,0.5,0.4,0.5,0.5,0.5,0.4,,0.6,0.7,0.7,0.8,0.7,0.5,0.7,0.7,0.6,0.7,0.8,0.6,0.7,0.9,0.8,0.8,0.7,0.9,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.1,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.1,0.1,0.2,0.2,0.3,0.2,0.3,0.1,0.1,0.1,0.1,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.5,0.2,0.3,0.4,0.3,0.4,0.4,0.4,0.3,0.5,0.4,,0.5,0.6,0.9,0.6,0.6,0.6,0.6,0.7,0.6,0.9,0.8,0.9,0.9,0.9,0.9,0.9,0.9,0.9,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.0,0.1,0.0,0.2,0.1,0.1,0.1,0.3,0.0,0.2,0.2,0.1,0.2,0.1,0.1,0.2,0.2,0.2,0.2,0.3,0.1,0.3,0.3,0.4,0.2,0.4,0.2,0.1,0.4,0.5,0.3,0.5,,0.7,0.5,0.7,0.8,0.7,0.8,0.7,0.8,0.8,0.8,0.9,0.9,0.9,0.9,1.0,0.9,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.0,0.1,0.1,0.1,0.2,0.0,0.0,0.1,0.3,0.1,0.2,0.1,0.5,0.1,0.2,0.2,0.2,0.2,0.4,0.3,0.4,0.3,,0.5,0.6,0.6,0.9,0.6,0.6,0.7,0.8,0.7,1.0,0.8,0.8,0.7,0.9,1.0,0.8,0.9
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.1,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.2,0.2,0.1,0.2,0.1,0.3,0.1,0.1,0.1,0.1,0.1,0.1,0.0,0.1,0.2,0.1,0.2,0.0,0.2,0.2,0.2,0.3,0.1,0.3,0.3,0.1,0.1,0.5,0.2,0.1,0.5,0.5,,0.6,0.5,0.5,0.6,0.5,0.6,0.7,0.6,0.7,0.7,0.7,0.8,0.8,0.9,0.9,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.1,0.0,0.1,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.2,0.0,0.1,0.0,0.1,0.2,0.1,0.1,0.1,0.3,0.0,0.3,0.2,0.2,0.0,0.0,0.1,0.3,0.3,0.2,0.3,0.1,0.1,0.3,0.2,0.3,0.3,0.3,0.3,0.3,0.2,0.3,0.3,0.4,0.3,0.4,0.4,,0.3,0.4,0.85,0.5,0.5,0.5,0.5,0.5,0.5,0.8,0.8,0.6,0.7,0.7,0.9
+KNN (tuned + ensembled),0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.1,0.1,0.0,0.1,0.1,0.2,0.0,0.2,0.1,0.1,0.0,0.0,0.1,0.2,0.1,0.1,0.1,0.1,0.2,0.1,0.1,0.3,0.1,0.2,0.2,0.1,0.0,0.4,0.5,0.4,0.2,0.4,0.5,0.7,,0.4,0.7,0.7,0.5,0.6,0.7,0.6,0.7,0.8,0.8,1.0,0.8,0.9,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.0,0.1,0.1,0.1,0.2,0.0,0.0,0.1,0.3,0.1,0.2,0.1,0.5,0.1,0.2,0.1,0.2,0.2,0.4,0.3,0.4,0.3,0.1,0.5,0.6,0.6,,0.6,0.6,0.7,0.8,0.7,1.0,0.8,0.8,0.7,0.9,1.0,0.8,0.9
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.1,0.2,0.0,0.1,0.1,0.2,0.0,0.3,0.2,0.3,0.0,0.0,0.0,0.3,0.2,0.2,0.2,0.1,0.1,0.3,0.2,0.3,0.3,0.1,0.3,0.3,0.2,0.3,0.3,0.4,0.2,0.4,0.4,0.15,0.3,0.4,,0.5,0.5,0.5,0.5,0.5,0.5,0.8,0.8,0.6,0.7,0.7,0.9
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.1,0.0,0.1,0.1,0.1,0.1,0.0,0.0,0.0,0.2,0.1,0.2,0.2,0.2,0.2,0.1,0.2,0.2,0.2,0.3,0.2,0.2,0.1,0.0,0.2,0.1,0.3,0.1,0.4,0.3,0.3,0.4,0.5,0.5,0.3,0.4,0.5,,0.6,0.5,0.8,0.5,0.6,0.6,0.9,0.5,0.6,0.8,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.2,0.2,0.1,0.1,0.1,0.3,0.1,0.1,0.1,0.4,0.0,0.0,0.1,0.1,0.2,0.1,0.3,0.4,0.2,0.3,0.4,0.5,0.5,0.3,0.5,0.4,,0.5,0.7,0.6,0.5,0.6,0.8,0.5,0.7,0.7,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.1,0.0,0.1,0.2,0.3,0.1,0.2,0.1,0.1,0.1,0.2,0.2,0.1,0.2,0.2,0.3,0.5,0.4,0.2,0.5,0.5,0.5,,0.6,0.2,1.0,0.5,0.6,0.6,0.6,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.1,0.1,0.2,0.1,0.1,0.0,0.0,0.1,0.0,0.1,0.1,0.4,0.2,0.2,0.3,0.4,0.5,0.3,0.3,0.5,0.2,0.3,0.4,,0.3,0.6,0.5,0.8,0.6,0.6,0.8,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.3,0.0,0.2,0.0,0.0,0.0,0.2,0.3,0.1,0.1,0.0,0.3,0.5,0.4,0.0,0.5,0.5,0.4,0.8,0.7,,0.8,0.5,0.7,0.7,0.7,0.8,0.8
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.1,0.0,0.1,0.2,0.3,0.1,0.2,0.0,0.1,0.1,0.1,0.1,0.1,0.1,0.2,0.3,0.5,0.3,0.2,0.5,0.4,0.5,0.0,0.4,0.2,,0.5,0.6,0.3,0.4,0.9,1.0
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.0,0.2,0.0,0.1,0.0,0.1,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.0,0.1,0.1,0.3,0.0,0.0,0.1,0.1,0.0,0.0,0.2,0.1,0.1,0.2,0.3,0.2,0.2,0.2,0.2,0.4,0.4,0.5,0.5,0.5,0.5,,0.8,0.6,0.6,0.7,0.9
+xRFM (default),0.0,0.0,0.1,0.1,0.1,0.1,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.0,0.0,0.1,0.1,0.0,0.1,0.0,0.1,0.1,0.0,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.0,0.0,0.1,0.1,0.2,0.1,0.2,0.0,0.1,0.2,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.2,0.1,0.1,0.3,0.2,0.2,0.2,0.3,0.2,0.1,0.2,0.4,0.2,0.3,0.4,0.2,,0.4,0.5,0.5,0.7
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.3,0.0,0.1,0.0,0.0,0.0,0.0,0.3,0.1,0.0,0.1,0.2,0.4,0.0,0.1,0.4,0.5,0.5,0.4,0.4,0.3,0.7,0.4,0.6,,0.5,0.8,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.3,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.0,0.1,0.3,0.2,0.0,0.3,0.4,0.3,0.4,0.4,0.3,0.6,0.4,0.5,0.5,,0.6,0.8
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.1,0.1,0.0,0.0,0.1,0.1,0.1,0.0,0.0,0.1,0.2,0.2,0.1,0.0,0.0,0.0,0.1,0.0,0.0,0.1,0.0,0.2,0.1,0.3,0.1,0.2,0.3,0.2,0.3,0.0,0.2,0.2,0.1,0.3,0.5,0.2,0.4,,0.8
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.1,0.0,0.1,0.0,0.1,0.1,0.0,0.0,0.0,0.0,0.2,0.0,0.1,0.3,0.0,0.2,0.2,
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/leaderboard_overview_explorer.html
index 72974a1fa5e9d535f844f4de6647672972882f66..156400857351417f7f0411e1cf099f56a301f842 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Classification | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..917719950a252796a334c6976aa74853b6bd1ba7
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_classification/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1745.0,"score":0.822,"rank":6.82,"hrank":1.51,"imp":5.227,"train_time":40.79,"predict_time":6.31,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+277\/-165","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1590.0,"score":0.568,"rank":12.98,"hrank":6.09,"imp":12.132,"train_time":7.56,"predict_time":0.399,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-76","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1582.0,"score":0.513,"rank":13.39,"hrank":9.17,"imp":13.542,"train_time":6.03,"predict_time":0.703,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-61","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1570.0,"score":0.536,"rank":14.05,"hrank":5.42,"imp":11.873,"train_time":2280.19,"predict_time":9.666,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-86","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1559.0,"score":0.533,"rank":14.68,"hrank":7.72,"imp":13.455,"train_time":6.28,"predict_time":0.531,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-79","imputed":false},{"position":5,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1556.0,"score":0.554,"rank":14.82,"hrank":5.7,"imp":10.522,"train_time":360.81,"predict_time":5.696,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+110\/-101","imputed":false},{"position":6,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1535.0,"score":0.516,"rank":16.07,"hrank":7.41,"imp":13.466,"train_time":3.45,"predict_time":0.182,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-96","imputed":false},{"position":7,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1535.0,"score":0.508,"rank":16.07,"hrank":8.85,"imp":14.259,"train_time":2280.19,"predict_time":0.968,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-74","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1460.0,"score":0.378,"rank":21.07,"hrank":12.14,"imp":16.431,"train_time":3487.23,"predict_time":22.857,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-64","imputed":false},{"position":9,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1452.0,"score":0.396,"rank":21.61,"hrank":7.81,"imp":15.118,"train_time":28.08,"predict_time":5.033,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+92\/-82","imputed":false},{"position":10,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1445.0,"score":0.36,"rank":22.11,"hrank":14.53,"imp":16.56,"train_time":2993.95,"predict_time":1.4,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-65","imputed":false},{"position":11,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1427.0,"score":0.37,"rank":23.5,"hrank":12.85,"imp":17.85,"train_time":3227.57,"predict_time":49.196,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+116\/-84","imputed":true},{"position":12,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1407.0,"score":0.326,"rank":25.07,"hrank":12.99,"imp":18.255,"train_time":145.75,"predict_time":4.066,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+102\/-80","imputed":true},{"position":13,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1400.0,"score":0.334,"rank":25.62,"hrank":13.89,"imp":17.563,"train_time":3487.23,"predict_time":1.001,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-70","imputed":false},{"position":14,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1393.0,"score":0.325,"rank":26.2,"hrank":10.89,"imp":18.088,"train_time":4.86,"predict_time":0.17,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-81","imputed":false},{"position":15,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1375.0,"score":0.323,"rank":27.68,"hrank":10.95,"imp":17.734,"train_time":7432.95,"predict_time":404.053,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-69","imputed":false},{"position":16,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1366.0,"score":0.314,"rank":28.46,"hrank":11.85,"imp":18.489,"train_time":2478.69,"predict_time":3.189,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-78","imputed":false},{"position":17,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1347.0,"score":0.261,"rank":30.04,"hrank":17.59,"imp":19.853,"train_time":3227.57,"predict_time":0.494,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+93\/-82","imputed":true},{"position":18,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1341.0,"score":0.239,"rank":30.57,"hrank":15.7,"imp":17.949,"train_time":1580.61,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-50","imputed":false},{"position":19,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1340.0,"score":0.265,"rank":30.66,"hrank":11.38,"imp":19.186,"train_time":9.08,"predict_time":1.864,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+68\/-62","imputed":true},{"position":20,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1340.0,"score":0.252,"rank":30.68,"hrank":20.42,"imp":20.365,"train_time":303.59,"predict_time":2.323,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-70","imputed":false},{"position":21,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1339.0,"score":0.239,"rank":30.71,"hrank":18.03,"imp":18.629,"train_time":21108.98,"predict_time":510.049,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-64","imputed":false},{"position":22,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1339.0,"score":0.281,"rank":30.73,"hrank":15.59,"imp":19.22,"train_time":2478.69,"predict_time":0.447,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-79","imputed":false},{"position":23,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1336.0,"score":0.202,"rank":30.96,"hrank":23.62,"imp":19.189,"train_time":774.79,"predict_time":1.455,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-52","imputed":false},{"position":24,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1335.0,"score":0.28,"rank":31.04,"hrank":15.33,"imp":18.824,"train_time":7432.95,"predict_time":39.789,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-92","imputed":false},{"position":25,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1331.0,"score":0.216,"rank":31.39,"hrank":23.51,"imp":18.524,"train_time":1580.61,"predict_time":0.381,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-62","imputed":false},{"position":26,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1305.0,"score":0.231,"rank":33.7,"hrank":14.14,"imp":20.182,"train_time":4.32,"predict_time":0.468,"imputed_pct":7.14,"hardware":"GPU","elo_ci":"+88\/-72","imputed":true},{"position":27,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1304.0,"score":0.207,"rank":33.73,"hrank":21.85,"imp":20.602,"train_time":2.25,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-68","imputed":false},{"position":28,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1301.0,"score":0.212,"rank":34.0,"hrank":18.34,"imp":19.522,"train_time":6288.36,"predict_time":0.473,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-73","imputed":false},{"position":29,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1300.0,"score":0.183,"rank":34.12,"hrank":24.14,"imp":20.201,"train_time":829.5,"predict_time":1.642,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-60","imputed":false},{"position":30,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1293.0,"score":0.166,"rank":34.71,"hrank":28.88,"imp":20.144,"train_time":7.32,"predict_time":0.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-47","imputed":false},{"position":31,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1291.0,"score":0.215,"rank":34.89,"hrank":18.81,"imp":20.118,"train_time":1165.31,"predict_time":2.067,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-69","imputed":false},{"position":32,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1287.0,"score":0.172,"rank":35.25,"hrank":27.63,"imp":20.118,"train_time":774.79,"predict_time":0.282,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-60","imputed":false},{"position":33,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1281.0,"score":0.169,"rank":35.82,"hrank":14.74,"imp":20.373,"train_time":829.5,"predict_time":0.241,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-58","imputed":false},{"position":34,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1276.0,"score":0.134,"rank":36.25,"hrank":24.63,"imp":21.068,"train_time":2068.2,"predict_time":0.165,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-63","imputed":false},{"position":35,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1276.0,"score":0.205,"rank":36.3,"hrank":20.1,"imp":21.143,"train_time":9.94,"predict_time":0.367,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+75\/-82","imputed":false},{"position":36,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1272.0,"score":0.163,"rank":36.61,"hrank":28.13,"imp":20.14,"train_time":21108.98,"predict_time":120.297,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-61","imputed":false},{"position":37,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1271.0,"score":0.246,"rank":36.73,"hrank":15.26,"imp":22.231,"train_time":13.41,"predict_time":1.64,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-91","imputed":false},{"position":38,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1270.0,"score":0.178,"rank":36.77,"hrank":27.59,"imp":21.665,"train_time":1110.83,"predict_time":0.553,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-63","imputed":false},{"position":39,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1267.0,"score":0.187,"rank":37.11,"hrank":25.96,"imp":21.044,"train_time":6288.36,"predict_time":9.385,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-78","imputed":false},{"position":40,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1246.0,"score":0.126,"rank":39.0,"hrank":26.0,"imp":21.217,"train_time":14.59,"predict_time":4.033,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-73","imputed":false},{"position":41,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1242.0,"score":0.154,"rank":39.36,"hrank":21.79,"imp":22.15,"train_time":1110.83,"predict_time":0.043,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-68","imputed":false},{"position":42,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1236.0,"score":0.156,"rank":39.82,"hrank":15.19,"imp":21.239,"train_time":1165.31,"predict_time":0.105,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-85","imputed":false},{"position":43,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1235.0,"score":0.099,"rank":39.96,"hrank":28.14,"imp":21.806,"train_time":2068.2,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-70","imputed":false},{"position":44,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1233.0,"score":0.127,"rank":40.16,"hrank":15.64,"imp":22.142,"train_time":7.27,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-66","imputed":false},{"position":45,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1231.0,"score":0.183,"rank":40.32,"hrank":20.51,"imp":21.503,"train_time":51.79,"predict_time":39.921,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-102","imputed":false},{"position":46,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1230.0,"score":0.141,"rank":40.39,"hrank":32.39,"imp":21.469,"train_time":2895.01,"predict_time":2.42,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-79","imputed":false},{"position":47,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1194.0,"score":0.105,"rank":43.61,"hrank":34.89,"imp":23.878,"train_time":4.88,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-71","imputed":false},{"position":48,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1189.0,"score":0.122,"rank":44.05,"hrank":30.37,"imp":23.78,"train_time":1425.59,"predict_time":7.352,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-84","imputed":false},{"position":49,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1181.0,"score":0.123,"rank":44.73,"hrank":30.53,"imp":23.376,"train_time":2895.01,"predict_time":0.145,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-79","imputed":false},{"position":50,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1175.0,"score":0.122,"rank":45.25,"hrank":27.98,"imp":23.639,"train_time":669.3,"predict_time":1.187,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-87","imputed":false},{"position":51,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1171.0,"score":0.101,"rank":45.66,"hrank":14.61,"imp":25.076,"train_time":14.22,"predict_time":0.352,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-88","imputed":false},{"position":52,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1149.0,"score":0.094,"rank":47.55,"hrank":28.3,"imp":24.952,"train_time":669.3,"predict_time":0.166,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-77","imputed":false},{"position":53,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1138.0,"score":0.054,"rank":48.46,"hrank":41.25,"imp":23.766,"train_time":2.85,"predict_time":0.19,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-59","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1133.0,"score":0.09,"rank":48.89,"hrank":31.77,"imp":25.153,"train_time":1425.59,"predict_time":0.561,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-92","imputed":false},{"position":55,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1133.0,"score":0.158,"rank":48.89,"hrank":14.99,"imp":25.94,"train_time":45012.0,"predict_time":5.273,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+111\/-118","imputed":false},{"position":56,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1130.0,"score":0.086,"rank":49.16,"hrank":39.02,"imp":24.07,"train_time":3.27,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-81","imputed":false},{"position":57,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1130.0,"score":0.107,"rank":49.18,"hrank":37.33,"imp":25.115,"train_time":562.08,"predict_time":1.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-92","imputed":false},{"position":58,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1124.0,"score":0.145,"rank":49.66,"hrank":16.73,"imp":26.204,"train_time":45012.0,"predict_time":21.34,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-121","imputed":false},{"position":59,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1097.0,"score":0.068,"rank":51.88,"hrank":25.88,"imp":26.197,"train_time":454.83,"predict_time":146.788,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+103\/-94","imputed":false},{"position":60,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1092.0,"score":0.048,"rank":52.27,"hrank":44.06,"imp":26.566,"train_time":15.58,"predict_time":2.601,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-77","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1088.0,"score":0.1,"rank":52.61,"hrank":39.86,"imp":31.917,"train_time":399.28,"predict_time":0.696,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-108","imputed":false},{"position":62,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1088.0,"score":0.054,"rank":52.61,"hrank":44.87,"imp":25.999,"train_time":562.08,"predict_time":0.138,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-86","imputed":false},{"position":63,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1044.0,"score":0.119,"rank":56.04,"hrank":14.37,"imp":32.339,"train_time":461.86,"predict_time":5.723,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-157","imputed":false},{"position":64,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1038.0,"score":0.054,"rank":56.48,"hrank":50.42,"imp":33.508,"train_time":399.28,"predict_time":0.184,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-98","imputed":false},{"position":65,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1019.0,"score":0.03,"rank":57.89,"hrank":50.64,"imp":28.681,"train_time":9.06,"predict_time":0.195,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+79\/-86","imputed":false},{"position":66,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1009.0,"score":0.045,"rank":58.57,"hrank":47.24,"imp":30.076,"train_time":3.76,"predict_time":0.597,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-114","imputed":false},{"position":67,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1007.0,"score":0.043,"rank":58.71,"hrank":52.0,"imp":29.132,"train_time":5.43,"predict_time":0.835,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-125","imputed":false},{"position":68,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1006.0,"score":0.049,"rank":58.75,"hrank":47.94,"imp":30.876,"train_time":2.45,"predict_time":0.519,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-113","imputed":false},{"position":69,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.022,"rank":59.2,"hrank":47.23,"imp":31.194,"train_time":0.46,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-109","imputed":false},{"position":70,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":986.0,"score":0.026,"rank":60.11,"hrank":55.46,"imp":28.744,"train_time":5.03,"predict_time":0.601,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-91","imputed":false},{"position":71,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":982.0,"score":0.04,"rank":60.38,"hrank":49.16,"imp":31.174,"train_time":123.01,"predict_time":1.578,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-119","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":970.0,"score":0.043,"rank":61.2,"hrank":35.19,"imp":38.17,"train_time":48.1,"predict_time":0.05,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-128","imputed":false},{"position":73,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":943.0,"score":0.036,"rank":62.86,"hrank":19.66,"imp":34.644,"train_time":479.33,"predict_time":0.495,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-104","imputed":false},{"position":74,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":922.0,"score":0.03,"rank":64.02,"hrank":30.58,"imp":35.229,"train_time":479.33,"predict_time":0.098,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-112","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":880.0,"score":0.01,"rank":66.25,"hrank":61.03,"imp":38.09,"train_time":2.34,"predict_time":0.132,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-122","imputed":false},{"position":76,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":828.0,"score":0.024,"rank":68.61,"hrank":62.76,"imp":37.998,"train_time":123.01,"predict_time":0.092,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-136","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":581.0,"score":0.0,"rank":75.46,"hrank":75.13,"imp":51.44,"train_time":0.28,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-148","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/pareto_front_explorer.html
index 555bbf0de6b74ec7edeec720cb8cb90443cd91b1..6217dfc7f235d700fe3cd725a54421adcf89d218 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3228316175,"imp":20.3647608165,"elo":1339.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404950151,"imp":20.1443707805,"elo":1293.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0468267403,"imp":17.9492000982,"elo":1340.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806426698,"imp":18.5237480144,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0464942884,"imp":23.8783284571,"elo":1194.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0428145039,"imp":22.1498644646,"elo":1241.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5527550553,"imp":21.6647630759,"elo":1270.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0218718174,"imp":22.1421574595,"elo":1232.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0200835992,"imp":21.8055362322,"elo":1234.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1652902771,"imp":21.0677240327,"elo":1276.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5968426684,"imp":30.0764285001,"elo":1009.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1664465955,"imp":24.9517725491,"elo":1148.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1866680852,"imp":23.6385525159,"elo":1175.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6011123119,"imp":28.7439430074,"elo":986.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5609692074,"imp":25.1525276008,"elo":1133.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.3522009809,"imp":23.7800287759,"elo":1188.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1925833078,"imp":24.0699968245,"elo":1130.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.2819017032,"imp":20.1177769158,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.4553717527,"imp":19.1886800782,"elo":1336.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.0331218603,"imp":15.1183200252,"elo":1452.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":4.0663954294,"imp":18.2554689291,"elo":1407.0,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"imp":25.0757414046,"elo":1170.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4731015689,"imp":19.522028043,"elo":1301.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"imp":21.0437129617,"elo":1266.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6014690178,"imp":26.5657672623,"elo":1092.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0497782518,"imp":38.1697809167,"elo":969.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1838950332,"imp":33.5078627352,"elo":1038.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6959068701,"imp":31.9168021795,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0698340289,"imp":31.1936039744,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.138131658,"imp":25.9993196555,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"imp":25.1151038439,"elo":1129.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.0326517405,"imp":21.2168446178,"elo":1245.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.0010951073,"imp":17.5633903008,"elo":1400.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":22.857077649,"imp":16.4309265514,"elo":1459.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7026957462,"imp":13.54153666,"elo":1582.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9678506248,"imp":14.2585599411,"elo":1534.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.6655104794,"imp":11.8728621439,"elo":1569.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6402183396,"imp":22.2311674425,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"imp":21.5028087148,"elo":1230.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.7894649716,"imp":18.8241008618,"elo":1335.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":404.0529881985,"imp":17.7335443803,"elo":1374.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1696165976,"imp":18.0878686497,"elo":1393.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.3100890028,"imp":5.2268628375,"elo":1744.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5190992166,"imp":30.8758370787,"elo":1006.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.8635665416,"imp":19.1864884776,"elo":1339.7,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"imp":13.4662319212,"elo":1534.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.3671853745,"imp":21.1430207976,"elo":1275.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4472563167,"imp":19.220128263,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.1888600496,"imp":18.4889931888,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"imp":13.4553844574,"elo":1558.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3985329866,"imp":12.1316862052,"elo":1590.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4683450152,"imp":20.1815938192,"elo":1304.9,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4942135315,"imp":19.8527267129,"elo":1347.0,"imputed":true,"imputed_pct":7.1428571429,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":49.1960479984,"imp":17.8503855685,"elo":1427.0,"imputed":true,"imputed_pct":7.1428571429,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"imp":32.3386395828,"elo":1044.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.2730552405,"imp":25.9401476351,"elo":1133.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.3395164105,"imp":26.2035893824,"elo":1124.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0547712076,"imp":20.6018985333,"elo":1304.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.195480498,"imp":28.6814573908,"elo":1018.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.144904019,"imp":23.3756026807,"elo":1181.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.419707216,"imp":21.4693522172,"elo":1230.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1903629661,"imp":23.766059458,"elo":1138.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2410402043,"imp":20.3725238619,"elo":1281.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.641714504,"imp":20.2008858688,"elo":1300.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":146.7883490527,"imp":26.1970038782,"elo":1097.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":120.2974149859,"imp":20.1400929464,"elo":1272.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":510.0490782849,"imp":18.6287386895,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.8352682059,"imp":29.1318506129,"elo":1007.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1053014255,"imp":21.2394120344,"elo":1236.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0671352109,"imp":20.1183883883,"elo":1291.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3228316175,"x_train":303.5874437498,"imp":20.3647608165,"elo":1339.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404950151,"x_train":7.3166111387,"imp":20.1443707805,"elo":1293.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0468267403,"x_train":1580.6128837326,"imp":17.9492000982,"elo":1340.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806426698,"x_train":1580.6128837326,"imp":18.5237480144,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0464942884,"x_train":4.8835669578,"imp":23.8783284571,"elo":1194.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0428145039,"x_train":1110.8290906413,"imp":22.1498644646,"elo":1241.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5527550553,"x_train":1110.8290906413,"imp":21.6647630759,"elo":1270.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0218718174,"x_train":7.2693312731,"imp":22.1421574595,"elo":1232.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0200835992,"x_train":2068.2009167512,"imp":21.8055362322,"elo":1234.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1652902771,"x_train":2068.2009167512,"imp":21.0677240327,"elo":1276.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5968426684,"x_train":3.7608843708,"imp":30.0764285001,"elo":1009.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1664465955,"x_train":669.3008326264,"imp":24.9517725491,"elo":1148.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1866680852,"x_train":669.3008326264,"imp":23.6385525159,"elo":1175.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6011123119,"x_train":5.0325246315,"imp":28.7439430074,"elo":986.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5609692074,"x_train":1425.5943698968,"imp":25.1525276008,"elo":1133.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.3522009809,"x_train":1425.5943698968,"imp":23.7800287759,"elo":1188.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.1925833078,"x_train":3.2714234862,"imp":24.0699968245,"elo":1130.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.2819017032,"x_train":774.7922679504,"imp":20.1177769158,"elo":1287.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":1.4553717527,"x_train":774.7922679504,"imp":19.1886800782,"elo":1336.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":5.0331218603,"x_train":28.0828822259,"imp":15.1183200252,"elo":1452.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":4.0663954294,"x_train":145.748631087,"imp":18.2554689291,"elo":1407.0,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"x_train":14.219089197,"imp":25.0757414046,"elo":1170.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4731015689,"x_train":6288.3552384481,"imp":19.522028043,"elo":1301.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"x_train":6288.3552384481,"imp":21.0437129617,"elo":1266.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6014690178,"x_train":15.5816092874,"imp":26.5657672623,"elo":1092.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0497782518,"x_train":48.1038174593,"imp":38.1697809167,"elo":969.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1838950332,"x_train":399.2771714041,"imp":33.5078627352,"elo":1038.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6959068701,"x_train":399.2771714041,"imp":31.9168021795,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0698340289,"x_train":0.4597788004,"imp":31.1936039744,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.138131658,"x_train":562.0823628399,"imp":25.9993196555,"elo":1088.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"x_train":562.0823628399,"imp":25.1151038439,"elo":1129.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":4.0326517405,"x_train":14.5881086057,"imp":21.2168446178,"elo":1245.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":1.0010951073,"x_train":3487.2314440046,"imp":17.5633903008,"elo":1400.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":22.857077649,"x_train":3487.2314440046,"imp":16.4309265514,"elo":1459.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7026957462,"x_train":6.0254254253,"imp":13.54153666,"elo":1582.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.9678506248,"x_train":2280.1896584743,"imp":14.2585599411,"elo":1534.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.6655104794,"x_train":2280.1896584743,"imp":11.8728621439,"elo":1569.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6402183396,"x_train":13.4089193675,"imp":22.2311674425,"elo":1270.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"x_train":51.7864862661,"imp":21.5028087148,"elo":1230.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.7894649716,"x_train":7432.9537119468,"imp":18.8241008618,"elo":1335.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":404.0529881985,"x_train":7432.9537119468,"imp":17.7335443803,"elo":1374.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1696165976,"x_train":4.8617027155,"imp":18.0878686497,"elo":1393.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.3100890028,"x_train":40.7899647401,"imp":5.2268628375,"elo":1744.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5190992166,"x_train":2.4480622255,"imp":30.8758370787,"elo":1006.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.8635665416,"x_train":9.0755702975,"imp":19.1864884776,"elo":1339.7,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"x_train":3.4521869623,"imp":13.4662319212,"elo":1534.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.3671853745,"x_train":9.93671076,"imp":21.1430207976,"elo":1275.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4472563167,"x_train":2478.69202837,"imp":19.220128263,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":3.1888600496,"x_train":2478.69202837,"imp":18.4889931888,"elo":1365.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"x_train":6.2811404231,"imp":13.4553844574,"elo":1558.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3985329866,"x_train":7.5625787904,"imp":12.1316862052,"elo":1590.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4683450152,"x_train":4.3178493542,"imp":20.1815938192,"elo":1304.9,"imputed":true,"imputed_pct":7.1428571429,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4942135315,"x_train":3227.5659724907,"imp":19.8527267129,"elo":1347.0,"imputed":true,"imputed_pct":7.1428571429,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":49.1960479984,"x_train":3227.5659724907,"imp":17.8503855685,"elo":1427.0,"imputed":true,"imputed_pct":7.1428571429,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"x_train":461.8626497906,"imp":32.3386395828,"elo":1044.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.2730552405,"x_train":45012.0033435545,"imp":25.9401476351,"elo":1133.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.3395164105,"x_train":45012.0033435545,"imp":26.2035893824,"elo":1124.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0547712076,"x_train":2.2471782685,"imp":20.6018985333,"elo":1304.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.195480498,"x_train":9.0610314235,"imp":28.6814573908,"elo":1018.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.144904019,"x_train":2895.0057162096,"imp":23.3756026807,"elo":1181.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.419707216,"x_train":2895.0057162096,"imp":21.4693522172,"elo":1230.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1903629661,"x_train":2.8478128159,"imp":23.766059458,"elo":1138.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.2410402043,"x_train":829.4997787901,"imp":20.3725238619,"elo":1281.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.641714504,"x_train":829.4997787901,"imp":20.2008858688,"elo":1300.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":146.7883490527,"x_train":454.8291985924,"imp":26.1970038782,"elo":1097.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":120.2974149859,"x_train":21108.9759033856,"imp":20.1400929464,"elo":1272.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":510.0490782849,"x_train":21108.9759033856,"imp":18.6287386895,"elo":1339.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.8352682059,"x_train":5.4303046313,"imp":29.1318506129,"elo":1007.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1053014255,"x_train":1165.3088146236,"imp":21.2394120344,"elo":1236.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.0671352109,"x_train":1165.3088146236,"imp":20.1183883883,"elo":1291.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/tuning_trajectories_explorer.html
index bbaa5238828f5f0883c4a71708c6b1a2f835c890..e942852956ef2d8bfb00147bd835142b62cb3a3b 100644
--- a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_classification/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":303.5874437498,"imp":20.3784196339,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":7.3166111387,"imp":20.1577383072,"elo":1305.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":13.0021013751,"imp":19.4228295773,"elo":1328.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":34.9838507458,"imp":19.0895778458,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":71.4491054101,"imp":18.8954757588,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":197.3421561911,"imp":18.7810879572,"elo":1347.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":391.4036568625,"imp":18.8021846512,"elo":1342.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":788.0795112228,"imp":18.6145697284,"elo":1341.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1179.7013801715,"imp":18.6184614256,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1580.6128837326,"imp":18.5240736201,"elo":1341.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.8835669578,"imp":23.8907825634,"elo":1199.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":10.414093577,"imp":23.1789280512,"elo":1233.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":25.923774854,"imp":22.7024469653,"elo":1254.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":52.8409015956,"imp":22.4570062464,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":137.1347812527,"imp":22.0296028447,"elo":1277.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":276.1566202862,"imp":21.8239100724,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":553.6509294091,"imp":21.7336034114,"elo":1285.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":829.2384033907,"imp":21.6645061141,"elo":1286.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1110.8290906413,"imp":21.6720105731,"elo":1283.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.2693312731,"imp":22.1546592862,"elo":1235.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":17.3786155791,"imp":21.924031667,"elo":1244.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":52.8479572348,"imp":21.647996779,"elo":1259.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":105.3309657113,"imp":21.5742582954,"elo":1261.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":259.310462097,"imp":21.4699190457,"elo":1269.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":524.5599999251,"imp":21.2956206292,"elo":1277.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1025.3924289592,"imp":21.1333616043,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1544.9631429961,"imp":21.1036497676,"elo":1289.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2068.2009167512,"imp":21.0716875312,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.7608843708,"imp":30.0895066561,"elo":1016.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.2336179369,"imp":28.2029097761,"elo":1043.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":17.356104837,"imp":25.5158631901,"elo":1096.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":34.0850647785,"imp":24.8244147834,"elo":1127.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":84.0342963818,"imp":24.1442813761,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":167.1827755321,"imp":23.9606053351,"elo":1164.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":333.2449706903,"imp":23.854928896,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":499.1329867675,"imp":23.6976049981,"elo":1180.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":669.3008326264,"imp":23.7026786551,"elo":1181.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.0325246315,"imp":28.755118253,"elo":989.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":12.7079451787,"imp":27.8636155574,"elo":1038.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":35.5292055013,"imp":25.9455439164,"elo":1110.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":67.8432178902,"imp":25.1457687324,"elo":1142.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":172.6394168737,"imp":24.219919892,"elo":1164.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":350.4018274029,"imp":23.8319517356,"elo":1179.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":703.8401639882,"imp":23.7595061441,"elo":1184.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1060.0265172973,"imp":23.7082762235,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1425.5943698968,"imp":23.8052476384,"elo":1194.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":3.2714234862,"imp":24.0825755521,"elo":1128.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":7.187975863,"imp":21.7286892164,"elo":1233.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":18.491534641,"imp":20.3780075977,"elo":1302.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":37.6314766741,"imp":19.8645419596,"elo":1326.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":95.7218520577,"imp":19.4713918801,"elo":1336.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":192.3000546929,"imp":19.3780731591,"elo":1343.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":385.4240248592,"imp":19.2784446174,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":577.4328596419,"imp":19.1846604265,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":774.7922679504,"imp":19.1978011314,"elo":1353.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":28.0828822259,"imp":15.3502057551,"elo":1457.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":145.748631087,"imp":18.2687953607,"elo":1423.5,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.219089197,"imp":25.0901952576,"elo":1174.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":51.0552417446,"imp":23.5181022033,"elo":1190.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":152.0656634461,"imp":20.9044686398,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":301.2283204956,"imp":20.3420716637,"elo":1299.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":742.3216508123,"imp":19.7887685711,"elo":1325.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1548.7724245944,"imp":20.1250258432,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3115.4832472775,"imp":20.7513704234,"elo":1283.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4642.1963843119,"imp":20.9854502025,"elo":1279.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6288.3552384481,"imp":21.0738762246,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":15.5816092874,"imp":26.5762207606,"elo":1101.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":48.1038174593,"imp":38.2212936747,"elo":973.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":125.2282972555,"imp":34.9814492674,"elo":1025.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":333.0389251457,"imp":32.0391127906,"elo":1091.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":399.2771714041,"imp":31.9128524401,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4597788004,"imp":31.2227062178,"elo":997.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.1539887452,"imp":28.5046195902,"elo":1036.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":11.7755523341,"imp":26.8107949698,"elo":1074.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":25.7323507846,"imp":25.9146301109,"elo":1093.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":67.6419649919,"imp":25.2136188323,"elo":1125.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":137.8532509555,"imp":25.0883119759,"elo":1133.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":278.1383106963,"imp":25.1339588526,"elo":1135.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":418.6187333473,"imp":25.1648710089,"elo":1135.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":562.0823628399,"imp":25.1285080803,"elo":1134.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":14.5881086057,"imp":21.2295915386,"elo":1250.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":30.5807961877,"imp":18.8737822725,"elo":1350.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":79.1587952873,"imp":17.3691309922,"elo":1424.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":166.3929204249,"imp":16.9055512378,"elo":1443.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":428.3936464406,"imp":16.4056438785,"elo":1469.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":868.2323938389,"imp":16.1829566449,"elo":1479.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1742.3462491724,"imp":16.2101279238,"elo":1479.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":2600.7236668564,"imp":16.365197117,"elo":1481.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":3487.2314440046,"imp":16.4487113798,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.0254254253,"imp":13.5554927886,"elo":1599.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":19.6925140518,"imp":13.6167610405,"elo":1587.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":54.027252147,"imp":13.5794692762,"elo":1578.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":109.1627008491,"imp":13.6663609087,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":282.2546622885,"imp":13.6268477951,"elo":1578.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":564.5394139879,"imp":13.6268656609,"elo":1569.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1133.3178139172,"imp":13.0691508052,"elo":1575.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1704.1213724065,"imp":12.7025937815,"elo":1582.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2280.1896584743,"imp":11.922156653,"elo":1581.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":13.4089193675,"imp":22.2438472545,"elo":1281.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.7864862661,"imp":21.5154943425,"elo":1234.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":88.6021525086,"imp":20.6159405669,"elo":1274.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":190.7023074438,"imp":19.7075109266,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":357.7608890189,"imp":19.3614046734,"elo":1337.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":906.5893386893,"imp":18.695537343,"elo":1353.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1833.1395916312,"imp":18.1880039529,"elo":1369.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3679.7001768605,"imp":17.991800423,"elo":1378.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":5552.0614127314,"imp":17.8993554545,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":7432.9537119468,"imp":17.7808424142,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":4.8617027155,"imp":18.1016879713,"elo":1404.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":40.7899647401,"imp":5.2410939461,"elo":1750.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":2.4480622255,"imp":30.8873423347,"elo":1018.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":9.0755702975,"imp":19.1995634373,"elo":1349.4,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":3.4521869623,"imp":13.4802213679,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.93671076,"imp":21.156184703,"elo":1288.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":21.7784528479,"imp":20.8258301031,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":59.121616311,"imp":20.2999492906,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":120.4004748358,"imp":19.7873281905,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":303.5800929631,"imp":19.3908702201,"elo":1357.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":609.3448213855,"imp":19.087705112,"elo":1372.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1226.0184259582,"imp":18.7605939907,"elo":1380.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1845.3925750472,"imp":18.629921205,"elo":1386.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2478.69202837,"imp":18.5027205023,"elo":1385.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":6.2811404231,"imp":13.4697369664,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":7.5625787904,"imp":12.1455240872,"elo":1603.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4.3178493542,"imp":20.1937210416,"elo":1316.7,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":25.5388834084,"imp":19.7431100252,"elo":1350.4,"imputed":true,"imputed_pct":7.1428571429,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":68.1485215536,"imp":19.038276357,"elo":1392.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":152.9753311952,"imp":18.43970477,"elo":1414.5,"imputed":true,"imputed_pct":7.1428571429,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":382.6473174812,"imp":17.8139632001,"elo":1445.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":773.9943558933,"imp":17.5441199106,"elo":1463.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1584.6944383903,"imp":17.561420949,"elo":1460.4,"imputed":true,"imputed_pct":7.1428571429,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2385.8103893693,"imp":17.7097428125,"elo":1451.2,"imputed":true,"imputed_pct":7.1428571429,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3227.5659724907,"imp":17.8834120926,"elo":1434.6,"imputed":true,"imputed_pct":7.1428571429,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":461.8626497906,"imp":32.3483660961,"elo":1051.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":812.6941385093,"imp":32.1892976909,"elo":1057.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1430.8811369497,"imp":31.7501264969,"elo":1063.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2384.5896366526,"imp":30.8432632338,"elo":1073.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5535.519999844,"imp":29.0365187367,"elo":1094.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11245.833348223,"imp":27.5228052242,"elo":1108.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23102.6961039552,"imp":26.9311417507,"elo":1117.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34138.663197007,"imp":26.6150722727,"elo":1126.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":45012.0033435545,"imp":26.2259662407,"elo":1138.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":2.2471782685,"imp":20.6144208142,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":9.0610314235,"imp":28.6866639617,"elo":1021.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":23.1218764356,"imp":25.7213884246,"elo":1096.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":66.1742587244,"imp":23.4768825658,"elo":1155.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":145.1501764392,"imp":22.6895399643,"elo":1194.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":352.024745135,"imp":22.1337662477,"elo":1222.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":719.83310274,"imp":21.8367301129,"elo":1232.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1418.0597495196,"imp":21.7424091663,"elo":1228.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2155.6849806435,"imp":21.5655894629,"elo":1237.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2895.0057162096,"imp":21.4792819163,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.8478128159,"imp":23.7783294832,"elo":1138.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":7.2765847065,"imp":21.7025160554,"elo":1231.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":20.3573867716,"imp":21.1710276833,"elo":1262.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":39.9575239687,"imp":20.8266834579,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":102.7267296917,"imp":20.5563115761,"elo":1291.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":205.8681515072,"imp":20.4622182,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":411.951005552,"imp":20.348162635,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":617.2039633301,"imp":20.2395952222,"elo":1314.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":829.4997787901,"imp":20.1923091302,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":454.8291985924,"imp":26.2100209601,"elo":1104.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1266.5525763884,"imp":23.5507313725,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":3732.2979901517,"imp":20.6269809904,"elo":1266.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":8203.1308672787,"imp":19.595310156,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":20166.4238376739,"imp":18.6623087328,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":21108.9759033856,"imp":18.6626491055,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":5.4303046313,"imp":29.1422521931,"elo":1014.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":11.776121563,"imp":26.4020020194,"elo":1079.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":28.7928271844,"imp":22.7829270285,"elo":1199.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":57.3403812154,"imp":21.265865824,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":145.2967129053,"imp":20.6088677477,"elo":1289.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":291.5311600969,"imp":20.2939815542,"elo":1302.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":582.7415169755,"imp":20.1605896781,"elo":1301.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":872.0774356109,"imp":20.158083026,"elo":1302.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1165.3088146236,"imp":20.1407704358,"elo":1302.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":2.3228316175,"x_train":303.5874437498,"imp":20.3784196339,"elo":1354.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0404950151,"x_train":7.3166111387,"imp":20.1577383072,"elo":1305.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0764260122,"x_train":13.0021013751,"imp":19.4228295773,"elo":1328.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1646517485,"x_train":34.9838507458,"imp":19.0895778458,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2059671983,"x_train":71.4491054101,"imp":18.8954757588,"elo":1345.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2459997699,"x_train":197.3421561911,"imp":18.7810879572,"elo":1347.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2734896112,"x_train":391.4036568625,"imp":18.8021846512,"elo":1342.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3310336248,"x_train":788.0795112228,"imp":18.6145697284,"elo":1341.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3663053836,"x_train":1179.7013801715,"imp":18.6184614256,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3806426698,"x_train":1580.6128837326,"imp":18.5240736201,"elo":1341.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0464942884,"x_train":4.8835669578,"imp":23.8907825634,"elo":1199.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0848784521,"x_train":10.414093577,"imp":23.1789280512,"elo":1233.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1691167697,"x_train":25.923774854,"imp":22.7024469653,"elo":1254.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2269108909,"x_train":52.8409015956,"imp":22.4570062464,"elo":1263.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3076887709,"x_train":137.1347812527,"imp":22.0296028447,"elo":1277.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3654650581,"x_train":276.1566202862,"imp":21.8239100724,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.462284001,"x_train":553.6509294091,"imp":21.7336034114,"elo":1285.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4730067641,"x_train":829.2384033907,"imp":21.6645061141,"elo":1286.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5527550553,"x_train":1110.8290906413,"imp":21.6720105731,"elo":1283.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0218718174,"x_train":7.2693312731,"imp":22.1546592862,"elo":1235.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0508456884,"x_train":17.3786155791,"imp":21.924031667,"elo":1244.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0739410125,"x_train":52.8479572348,"imp":21.647996779,"elo":1259.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0914252829,"x_train":105.3309657113,"imp":21.5742582954,"elo":1261.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1248677418,"x_train":259.310462097,"imp":21.4699190457,"elo":1269.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1350318723,"x_train":524.5599999251,"imp":21.2956206292,"elo":1277.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1560473551,"x_train":1025.3924289592,"imp":21.1333616043,"elo":1285.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1586418791,"x_train":1544.9631429961,"imp":21.1036497676,"elo":1289.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1652902771,"x_train":2068.2009167512,"imp":21.0716875312,"elo":1289.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.5968426684,"x_train":3.7608843708,"imp":30.0895066561,"elo":1016.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6853995207,"x_train":7.2336179369,"imp":28.2029097761,"elo":1043.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6085460645,"x_train":17.356104837,"imp":25.5158631901,"elo":1096.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6559709961,"x_train":34.0850647785,"imp":24.8244147834,"elo":1127.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9058839094,"x_train":84.0342963818,"imp":24.1442813761,"elo":1157.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.196621892,"x_train":167.1827755321,"imp":23.9606053351,"elo":1164.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4547771678,"x_train":333.2449706903,"imp":23.854928896,"elo":1169.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.5815845677,"x_train":499.1329867675,"imp":23.6976049981,"elo":1180.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1866680852,"x_train":669.3008326264,"imp":23.7026786551,"elo":1181.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6011123119,"x_train":5.0325246315,"imp":28.755118253,"elo":989.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4502158609,"x_train":12.7079451787,"imp":27.8636155574,"elo":1038.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7078527827,"x_train":35.5292055013,"imp":25.9455439164,"elo":1110.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.3184880832,"x_train":67.8432178902,"imp":25.1457687324,"elo":1142.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.5303897769,"x_train":172.6394168737,"imp":24.219919892,"elo":1164.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.7545080726,"x_train":350.4018274029,"imp":23.8319517356,"elo":1179.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.1854207053,"x_train":703.8401639882,"imp":23.7595061441,"elo":1184.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.6864231297,"x_train":1060.0265172973,"imp":23.7082762235,"elo":1186.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.3010312699,"x_train":1425.5943698968,"imp":23.8052476384,"elo":1194.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.1925833078,"x_train":3.2714234862,"imp":24.0825755521,"elo":1128.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.6370937782,"x_train":7.187975863,"imp":21.7286892164,"elo":1233.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8585219059,"x_train":18.491534641,"imp":20.3780075977,"elo":1302.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.0687137304,"x_train":37.6314766741,"imp":19.8645419596,"elo":1326.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3923655359,"x_train":95.7218520577,"imp":19.4713918801,"elo":1336.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.480046567,"x_train":192.3000546929,"imp":19.3780731591,"elo":1343.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6191358781,"x_train":385.4240248592,"imp":19.2784446174,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.8652145084,"x_train":577.4328596419,"imp":19.1846604265,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4553717527,"x_train":774.7922679504,"imp":19.1978011314,"elo":1353.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":5.0331218603,"x_train":28.0828822259,"imp":15.3502057551,"elo":1457.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":4.0663954294,"x_train":145.748631087,"imp":18.2687953607,"elo":1423.5,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.352275142,"x_train":14.219089197,"imp":25.0901952576,"elo":1174.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9081956315,"x_train":51.0552417446,"imp":23.5181022033,"elo":1190.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8094695931,"x_train":152.0656634461,"imp":20.9044686398,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9001397479,"x_train":301.2283204956,"imp":20.3420716637,"elo":1299.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.3070047996,"x_train":742.3216508123,"imp":19.7887685711,"elo":1325.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.5713626606,"x_train":1548.7724245944,"imp":20.1250258432,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.0463659943,"x_train":3115.4832472775,"imp":20.7513704234,"elo":1283.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.4271738524,"x_train":4642.1963843119,"imp":20.9854502025,"elo":1279.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.384570834,"x_train":6288.3552384481,"imp":21.0738762246,"elo":1277.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":2.6014690178,"x_train":15.5816092874,"imp":26.5762207606,"elo":1101.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0497782518,"x_train":48.1038174593,"imp":38.2212936747,"elo":973.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4010049235,"x_train":125.2282972555,"imp":34.9814492674,"elo":1025.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.7337631815,"x_train":333.0389251457,"imp":32.0391127906,"elo":1091.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.6959068701,"x_train":399.2771714041,"imp":31.9128524401,"elo":1094.2,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0698340289,"x_train":0.4597788004,"imp":31.2227062178,"elo":997.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2614383057,"x_train":3.1539887452,"imp":28.5046195902,"elo":1036.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.45693222,"x_train":11.7755523341,"imp":26.8107949698,"elo":1074.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5950629764,"x_train":25.7323507846,"imp":25.9146301109,"elo":1093.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7642726368,"x_train":67.6419649919,"imp":25.2136188323,"elo":1125.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9217934517,"x_train":137.8532509555,"imp":25.0883119759,"elo":1133.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0817312972,"x_train":278.1383106963,"imp":25.1339588526,"elo":1135.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2223316361,"x_train":418.6187333473,"imp":25.1648710089,"elo":1135.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.1552648936,"x_train":562.0823628399,"imp":25.1285080803,"elo":1134.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":4.0326517405,"x_train":14.5881086057,"imp":21.2295915386,"elo":1250.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.5458694726,"x_train":30.5807961877,"imp":18.8737822725,"elo":1350.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":8.0822974518,"x_train":79.1587952873,"imp":17.3691309922,"elo":1424.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.3031022636,"x_train":166.3929204249,"imp":16.9055512378,"elo":1443.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.3882183239,"x_train":428.3936464406,"imp":16.4056438785,"elo":1469.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4616303712,"x_train":868.2323938389,"imp":16.1829566449,"elo":1479.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.6029747084,"x_train":1742.3462491724,"imp":16.2101279238,"elo":1479.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.4107968916,"x_train":2600.7236668564,"imp":16.365197117,"elo":1481.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":22.857077649,"x_train":3487.2314440046,"imp":16.4487113798,"elo":1480.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7026957462,"x_train":6.0254254253,"imp":13.5554927886,"elo":1599.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.32293962,"x_train":19.6925140518,"imp":13.6167610405,"elo":1587.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8631014568,"x_train":54.027252147,"imp":13.5794692762,"elo":1578.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.9894747323,"x_train":109.1627008491,"imp":13.6663609087,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.5217042751,"x_train":282.2546622885,"imp":13.6268477951,"elo":1578.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.0109718207,"x_train":564.5394139879,"imp":13.6268656609,"elo":1569.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.1730834001,"x_train":1133.3178139172,"imp":13.0691508052,"elo":1575.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.0817714469,"x_train":1704.1213724065,"imp":12.7025937815,"elo":1582.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.6282078215,"x_train":2280.1896584743,"imp":11.922156653,"elo":1581.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.6402183396,"x_train":13.4089193675,"imp":22.2438472545,"elo":1281.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.9214989303,"x_train":51.7864862661,"imp":21.5154943425,"elo":1234.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":59.2095298832,"x_train":88.6021525086,"imp":20.6159405669,"elo":1274.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":91.3130194728,"x_train":190.7023074438,"imp":19.7075109266,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":128.7090086601,"x_train":357.7608890189,"imp":19.3614046734,"elo":1337.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":197.3779151734,"x_train":906.5893386893,"imp":18.695537343,"elo":1353.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":248.1558123661,"x_train":1833.1395916312,"imp":18.1880039529,"elo":1369.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":314.8436401055,"x_train":3679.7001768605,"imp":17.991800423,"elo":1378.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":344.2915338009,"x_train":5552.0614127314,"imp":17.8993554545,"elo":1383.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":400.899764047,"x_train":7432.9537119468,"imp":17.7808424142,"elo":1389.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1696165976,"x_train":4.8617027155,"imp":18.1016879713,"elo":1404.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.3100890028,"x_train":40.7899647401,"imp":5.2410939461,"elo":1750.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.5190992166,"x_train":2.4480622255,"imp":30.8873423347,"elo":1018.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.8635665416,"x_train":9.0755702975,"imp":19.1995634373,"elo":1349.4,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1820042357,"x_train":3.4521869623,"imp":13.4802213679,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.3671853745,"x_train":9.93671076,"imp":21.156184703,"elo":1288.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.7792451672,"x_train":21.7784528479,"imp":20.8258301031,"elo":1298.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.3835716074,"x_train":59.121616311,"imp":20.2999492906,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":1.7666598519,"x_train":120.4004748358,"imp":19.7873281905,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.2463538995,"x_train":303.5800929631,"imp":19.3908702201,"elo":1357.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":2.282918805,"x_train":609.3448213855,"imp":19.087705112,"elo":1372.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":2.6967059891,"x_train":1226.0184259582,"imp":18.7605939907,"elo":1380.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":2.9081786218,"x_train":1845.3925750472,"imp":18.629921205,"elo":1386.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":3.1888600496,"x_train":2478.69202837,"imp":18.5027205023,"elo":1385.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.530737751,"x_train":6.2811404231,"imp":13.4697369664,"elo":1576.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3985329866,"x_train":7.5625787904,"imp":12.1455240872,"elo":1603.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4683450152,"x_train":4.3178493542,"imp":20.1937210416,"elo":1316.7,"imputed":true,"imputed_pct":7.1428571429,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.1540163946,"x_train":25.5388834084,"imp":19.7431100252,"elo":1350.4,"imputed":true,"imputed_pct":7.1428571429,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":9.8028846905,"x_train":68.1485215536,"imp":19.038276357,"elo":1392.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":14.9803553745,"x_train":152.9753311952,"imp":18.43970477,"elo":1414.5,"imputed":true,"imputed_pct":7.1428571429,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":23.970685412,"x_train":382.6473174812,"imp":17.8139632001,"elo":1445.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":30.0452683926,"x_train":773.9943558933,"imp":17.5441199106,"elo":1463.0,"imputed":true,"imputed_pct":7.1428571429,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":40.2654929937,"x_train":1584.6944383903,"imp":17.561420949,"elo":1460.4,"imputed":true,"imputed_pct":7.1428571429,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":45.5687948607,"x_train":2385.8103893693,"imp":17.7097428125,"elo":1451.2,"imputed":true,"imputed_pct":7.1428571429,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":49.1960479984,"x_train":3227.5659724907,"imp":17.8834120926,"elo":1434.6,"imputed":true,"imputed_pct":7.1428571429,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.722691649,"x_train":461.8626497906,"imp":32.3483660961,"elo":1051.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.1148611269,"x_train":812.6941385093,"imp":32.1892976909,"elo":1057.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3665562879,"x_train":1430.8811369497,"imp":31.7501264969,"elo":1063.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.8811964581,"x_train":2384.5896366526,"imp":30.8432632338,"elo":1073.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.7428459792,"x_train":5535.519999844,"imp":29.0365187367,"elo":1094.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.4043789983,"x_train":11245.833348223,"imp":27.5228052242,"elo":1108.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":23.5017727083,"x_train":23102.6961039552,"imp":26.9311417507,"elo":1117.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":24.5387683339,"x_train":34138.663197007,"imp":26.6150722727,"elo":1126.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.3395164105,"x_train":45012.0033435545,"imp":26.2259662407,"elo":1138.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0547712076,"x_train":2.2471782685,"imp":20.6144208142,"elo":1320.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.195480498,"x_train":9.0610314235,"imp":28.6866639617,"elo":1021.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.385980123,"x_train":23.1218764356,"imp":25.7213884246,"elo":1096.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7451473275,"x_train":66.1742587244,"imp":23.4768825658,"elo":1155.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0088740142,"x_train":145.1501764392,"imp":22.6895399643,"elo":1194.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2591398849,"x_train":352.024745135,"imp":22.1337662477,"elo":1222.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6135463766,"x_train":719.83310274,"imp":21.8367301129,"elo":1232.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0639002358,"x_train":1418.0597495196,"imp":21.7424091663,"elo":1228.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1878184457,"x_train":2155.6849806435,"imp":21.5655894629,"elo":1237.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.419707216,"x_train":2895.0057162096,"imp":21.4792819163,"elo":1237.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1903629661,"x_train":2.8478128159,"imp":23.7783294832,"elo":1138.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.664697711,"x_train":7.2765847065,"imp":21.7025160554,"elo":1231.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.24520154,"x_train":20.3573867716,"imp":21.1710276833,"elo":1262.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4521450072,"x_train":39.9575239687,"imp":20.8266834579,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8858828782,"x_train":102.7267296917,"imp":20.5563115761,"elo":1291.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.9134288142,"x_train":205.8681515072,"imp":20.4622182,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0984566815,"x_train":411.951005552,"imp":20.348162635,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.8473752296,"x_train":617.2039633301,"imp":20.2395952222,"elo":1314.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.503647059,"x_train":829.4997787901,"imp":20.1923091302,"elo":1315.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":146.7883490527,"x_train":454.8291985924,"imp":26.2100209601,"elo":1104.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":268.9452732023,"x_train":1266.5525763884,"imp":23.5507313725,"elo":1174.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":453.5002018107,"x_train":3732.2979901517,"imp":20.6269809904,"elo":1266.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":588.0968491642,"x_train":8203.1308672787,"imp":19.595310156,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":491.6490605571,"x_train":20166.4238376739,"imp":18.6623087328,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":510.0490782849,"x_train":21108.9759033856,"imp":18.6626491055,"elo":1351.6,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.8352682059,"x_train":5.4303046313,"imp":29.1422521931,"elo":1014.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.0296880071,"x_train":11.776121563,"imp":26.4020020194,"elo":1079.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.2425439486,"x_train":28.7928271844,"imp":22.7829270285,"elo":1199.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.5881834196,"x_train":57.3403812154,"imp":21.265865824,"elo":1263.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.9845916387,"x_train":145.2967129053,"imp":20.6088677477,"elo":1289.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.1886115821,"x_train":291.5311600969,"imp":20.2939815542,"elo":1302.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.033782756,"x_train":582.7415169755,"imp":20.1605896781,"elo":1301.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":1.9700745263,"x_train":872.0774356109,"imp":20.158083026,"elo":1302.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9325738702,"x_train":1165.3088146236,"imp":20.1407704358,"elo":1302.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..ab158b27937e7d5bbda1d203d90f872890198c1b
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_classification/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "TabPFN-3 (default)", "RealTabPFN-2.5 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-2.6 (default)", "AutoGluon 1.5 (extreme, 4h)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned)", "RealMLP (tuned + ensembled)", "LimiX (default)", "AutoGluon 1.4 (best, 4h)", "TabPFNv2 (tuned + ensembled)", "Mitra (default)", "RealMLP (tuned)", "TabDPT-Turbo (default)", "TabDPT (tuned + ensembled)", "TabM (tuned + ensembled)", "TabPFNv2 (tuned)", "CatBoost (tuned)", "TabICL (default)", "BetaTabPFN (default)", "iLTM (tuned + ensembled)", "TabM (tuned)", "LightGBM (tuned + ensembled)", "TabDPT (tuned)", "CatBoost (tuned + ensembled)", "TabPFNv2 (default)", "TabSwift (default)", "ModernNCA (tuned)", "XGBoost (tuned + ensembled)", "CatBoost (default)", "xRFM (tuned + ensembled)", "LightGBM (tuned)", "XGBoost (tuned)", "EBM (tuned + ensembled)", "TabM (default)", "iLTM (tuned)", "SAP-RPT-OSS (default)", "ChimeraBoost (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "RealMLP (default)", "ChimeraBoost (tuned)", "xRFM (tuned)", "EBM (tuned)", "EBM (default)", "TabDPT (default)", "TorchMLP (tuned + ensembled)", "ChimeraBoost (default)", "FastaiMLP (tuned + ensembled)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "ModernNCA (default)", "ExtraTrees (tuned)", "XGBoost (default)", "TabSTAR (tuned)", "FastaiMLP (tuned)", "LightGBM (default)", "RandomForest (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "iLTM (default)", "OrionMSP (default)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (default)", "PerpetualBooster (tuned)", "TorchMLP (default)", "ExtraTrees (default)", "xRFM (default)", "TabFlex (default)", "RandomForest (default)", "FastaiMLP (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 1.0, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 1.0, 0.9285714030265808, 1.0, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0, 0.9285714030265808, 1.0, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0], [0.1428571492433548, NaN, 0.4642857015132904, 0.5535714030265808, 0.5357142686843872, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.5714285969734192, 0.7142857313156128, 0.8214285969734192, 0.75, 0.75, 0.75, 0.7857142686843872, 0.8928571343421936, 0.75, 0.9285714030265808, 0.8928571343421936, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8214285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8928571343421936, 0.75, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.8214285969734192, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 1.0, 0.9642857313156128, 0.9642857313156128, 0.8214285969734192, 1.0, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 1.0, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0], [0.1785714328289032, 0.5357142686843872, NaN, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5, 0.4821428656578064, 0.7142857313156128, 0.75, 0.6785714030265808, 0.5714285969734192, 0.7857142686843872, 0.7857142686843872, 0.6785714030265808, 0.75, 0.7857142686843872, 0.8571428656578064, 0.8214285969734192, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.8928571343421936, 0.75, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.9464285969734192, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.9642857313156128, 0.8571428656578064, 1.0, 0.9642857313156128, 0.8928571343421936, 0.8571428656578064, 0.9642857313156128, 1.0, 0.9642857313156128, 0.9285714030265808, 0.8571428656578064, 1.0, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0], [0.1071428582072258, 0.4464285671710968, 0.5714285969734192, NaN, 0.5714285969734192, 0.5357142686843872, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.6428571343421936, 0.8571428656578064, 0.8214285969734192, 0.75, 0.6071428656578064, 0.6428571343421936, 0.75, 0.9285714030265808, 0.8928571343421936, 0.7142857313156128, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.7142857313156128, 0.9285714030265808, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.8571428656578064, 0.75, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 1.0, 0.7857142686843872, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.7857142686843872, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 1.0], [0.1428571492433548, 0.4642857015132904, 0.5714285969734192, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.5, 0.6785714030265808, 0.75, 0.6428571343421936, 0.75, 0.8214285969734192, 0.75, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.75, 0.8928571343421936, 0.7142857313156128, 0.8571428656578064, 0.6785714030265808, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.8928571343421936, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.7857142686843872, 0.8214285969734192, 0.9285714030265808, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.8214285969734192, 0.75, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.7857142686843872, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.7857142686843872, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.7857142686843872, 1.0, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.1785714328289032, 0.4642857015132904, 0.4285714328289032, 0.4642857015132904, 0.5714285969734192, NaN, 0.4642857015132904, 0.5714285969734192, 0.75, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.75, 0.5714285969734192, 0.75, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.75, 0.8571428656578064, 0.7857142686843872, 0.8571428656578064, 0.7142857313156128, 0.75, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.75, 0.8928571343421936, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.8214285969734192, 0.9285714030265808, 1.0, 0.9285714030265808, 0.8928571343421936, 0.8214285969734192, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 1.0], [0.1071428582072258, 0.4285714328289032, 0.5, 0.3928571343421936, 0.5714285969734192, 0.5357142686843872, NaN, 0.5357142686843872, 0.6785714030265808, 0.6428571343421936, 0.75, 0.6785714030265808, 0.7142857313156128, 0.6071428656578064, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.75, 0.75, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.75, 0.8214285969734192, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.75, 0.7142857313156128, 0.8571428656578064, 0.8928571343421936, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.9642857313156128, 0.9285714030265808, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 1.0, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 0.9642857313156128, 1.0], [0.1071428582072258, 0.3928571343421936, 0.5178571343421936, 0.3571428656578064, 0.5, 0.4285714328289032, 0.4642857015132904, NaN, 0.6428571343421936, 0.5714285969734192, 0.6428571343421936, 0.625, 0.8214285969734192, 0.75, 0.6071428656578064, 0.6071428656578064, 0.8214285969734192, 0.9285714030265808, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.75, 0.8214285969734192, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.75, 0.7142857313156128, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.75, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.7857142686843872, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.7857142686843872, 0.8928571343421936, 1.0, 0.8928571343421936, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.8571428656578064, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 1.0], [0.1785714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.25, 0.3214285671710968, 0.3571428656578064, NaN, 0.3928571343421936, 0.5714285969734192, 0.4642857015132904, 0.5357142686843872, 0.6071428656578064, 0.5357142686843872, 0.5357142686843872, 0.6785714030265808, 0.6428571343421936, 0.75, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.6071428656578064, 0.8571428656578064, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.6785714030265808, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.7142857313156128, 0.8214285969734192, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.8214285969734192, 0.75, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8928571343421936, 0.8214285969734192, 0.9285714030265808, 0.75, 0.8214285969734192, 0.8928571343421936, 0.75, 0.75, 0.9285714030265808, 0.9285714030265808, 0.8571428656578064, 0.9285714030265808, 0.7857142686843872, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.1071428582072258, 0.3214285671710968, 0.25, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3571428656578064, 0.4285714328289032, 0.6071428656578064, NaN, 0.5, 0.6071428656578064, 0.5357142686843872, 0.6964285969734192, 0.5357142686843872, 0.4642857015132904, 0.6785714030265808, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.6964285969734192, 0.7142857313156128, 0.75, 0.6071428656578064, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.75, 0.75, 0.6785714030265808, 0.75, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8214285969734192, 0.75, 0.75, 0.75, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.6428571343421936, 0.7857142686843872, 0.8214285969734192, 0.75, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8392857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.9107142686843872, 0.9285714030265808, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.4285714328289032, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.25, 0.3571428656578064, 0.4285714328289032, 0.5, NaN, 0.4285714328289032, 0.5357142686843872, 0.6071428656578064, 0.5714285969734192, 0.5, 0.6071428656578064, 0.5714285969734192, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.75, 0.7142857313156128, 0.75, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.75, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6428571343421936, 0.7142857313156128, 0.7857142686843872, 0.75, 0.8571428656578064, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.7857142686843872, 0.8928571343421936, 0.7857142686843872, 0.8928571343421936, 0.7142857313156128, 0.8928571343421936, 0.8928571343421936, 0.8214285969734192, 0.75, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0, 1.0, 1.0], [0.0357142873108387, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.25, 0.3214285671710968, 0.3214285671710968, 0.375, 0.5357142686843872, 0.3928571343421936, 0.5714285969734192, NaN, 0.5535714030265808, 0.5357142686843872, 0.5, 0.5357142686843872, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.625, 0.7142857313156128, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.5714285969734192, 0.6785714030265808, 0.6607142686843872, 0.6428571343421936, 0.7142857313156128, 0.8214285969734192, 0.6785714030265808, 0.7142857313156128, 0.75, 0.7857142686843872, 0.7857142686843872, 0.7142857313156128, 0.6428571343421936, 0.6071428656578064, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.6071428656578064, 0.7857142686843872, 0.8214285969734192, 0.75, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.7142857313156128, 0.8214285969734192, 0.9285714030265808, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8214285969734192, 0.875, 0.8214285969734192, 0.75, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.9107142686843872, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 1.0], [0.0, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.2857142984867096, 0.2857142984867096, 0.1785714328289032, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.4464285671710968, NaN, 0.5714285969734192, 0.5, 0.5357142686843872, 0.7142857313156128, 0.625, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.75, 0.7142857313156128, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.7142857313156128, 0.8214285969734192, 0.6785714030265808, 0.6071428656578064, 0.75, 0.8214285969734192, 0.7142857313156128, 0.8214285969734192, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6785714030265808, 0.7142857313156128, 0.7857142686843872, 0.75, 0.8928571343421936, 0.75, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.75, 0.75, 0.8214285969734192, 0.75, 0.75, 0.8214285969734192, 0.8392857313156128, 0.8392857313156128, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.9285714030265808, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.1428571492433548, 0.25, 0.2142857164144516, 0.25, 0.25, 0.25, 0.3928571343421936, 0.25, 0.3928571343421936, 0.3035714328289032, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, NaN, 0.4642857015132904, 0.3928571343421936, 0.5357142686843872, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.5714285969734192, 0.6428571343421936, 0.6785714030265808, 0.5357142686843872, 0.6071428656578064, 0.6428571343421936, 0.6071428656578064, 0.75, 0.7142857313156128, 0.75, 0.5714285969734192, 0.7142857313156128, 0.75, 0.75, 0.7142857313156128, 0.7142857313156128, 0.6071428656578064, 0.6785714030265808, 0.75, 0.75, 0.7142857313156128, 0.6785714030265808, 0.75, 0.75, 0.6428571343421936, 0.7857142686843872, 0.75, 0.75, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8571428656578064, 0.8571428656578064, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.9285714030265808, 0.8035714030265808, 0.8928571343421936, 0.8571428656578064, 0.7857142686843872, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.8214285969734192, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9464285969734192, 0.9285714030265808, 0.9642857313156128, 1.0], [0.1071428582072258, 0.25, 0.3214285671710968, 0.3928571343421936, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.5, 0.5, 0.5357142686843872, NaN, 0.7142857313156128, 0.5, 0.5714285969734192, 0.5, 0.5714285969734192, 0.5, 0.6071428656578064, 0.5, 0.6071428656578064, 0.6785714030265808, 0.5357142686843872, 0.5357142686843872, 0.6964285969734192, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.6785714030265808, 0.5714285969734192, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.75, 0.7857142686843872, 0.6785714030265808, 0.7142857313156128, 0.75, 0.75, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.8214285969734192, 0.75, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.75, 0.8214285969734192, 0.9285714030265808, 0.8214285969734192, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 1.0], [0.0714285746216774, 0.25, 0.25, 0.3571428656578064, 0.3571428656578064, 0.25, 0.2142857164144516, 0.3928571343421936, 0.4642857015132904, 0.5357142686843872, 0.5, 0.4642857015132904, 0.4642857015132904, 0.6071428656578064, 0.2857142984867096, NaN, 0.4642857015132904, 0.5714285969734192, 0.5357142686843872, 0.5714285969734192, 0.5357142686843872, 0.6785714030265808, 0.4642857015132904, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.6071428656578064, 0.75, 0.6428571343421936, 0.6785714030265808, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.75, 0.7142857313156128, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.6428571343421936, 0.75, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.75, 0.75, 0.75, 0.75, 0.8214285969734192, 0.9642857313156128, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 1.0], [0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.2857142984867096, 0.4642857015132904, 0.5, 0.5357142686843872, NaN, 0.3571428656578064, 0.5, 0.6071428656578064, 0.6071428656578064, 0.5714285969734192, 0.5357142686843872, 0.6071428656578064, 0.5, 0.5357142686843872, 0.5, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.75, 0.7857142686843872, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.75, 0.7857142686843872, 0.5714285969734192, 0.6785714030265808, 0.75, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.9642857313156128, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.7857142686843872, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 1.0, 1.0], [0.0357142873108387, 0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.1785714328289032, 0.25, 0.0714285746216774, 0.3571428656578064, 0.3214285671710968, 0.4285714328289032, 0.2857142984867096, 0.375, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.6428571343421936, NaN, 0.5357142686843872, 0.4821428656578064, 0.5714285969734192, 0.5, 0.6428571343421936, 0.6071428656578064, 0.4642857015132904, 0.5, 0.5535714030265808, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.6071428656578064, 0.5535714030265808, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.6428571343421936, 0.75, 0.5714285969734192, 0.75, 0.7857142686843872, 0.6785714030265808, 0.8214285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8214285969734192, 0.7142857313156128, 0.75, 0.8214285969734192, 0.7142857313156128, 0.7142857313156128, 0.8214285969734192, 0.7857142686843872, 0.8035714030265808, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8392857313156128, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 1.0], [0.0714285746216774, 0.25, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.25, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.4285714328289032, 0.5, 0.4642857015132904, 0.5, 0.4642857015132904, NaN, 0.4642857015132904, 0.5, 0.5, 0.5357142686843872, 0.5, 0.5, 0.4642857015132904, 0.5, 0.5357142686843872, 0.5357142686843872, 0.6071428656578064, 0.6428571343421936, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6785714030265808, 0.6071428656578064, 0.5, 0.6071428656578064, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.75, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.6785714030265808, 0.75, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.8214285969734192, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8214285969734192, 0.75, 0.8571428656578064, 0.7857142686843872, 0.75, 0.8928571343421936, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.0357142873108387, 0.0714285746216774, 0.3214285671710968, 0.2857142984867096, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.375, 0.3571428656578064, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.5178571343421936, 0.5357142686843872, NaN, 0.5, 0.625, 0.3928571343421936, 0.5, 0.4642857015132904, 0.4642857015132904, 0.5357142686843872, 0.6428571343421936, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.5357142686843872, 0.5357142686843872, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.6785714030265808, 0.6071428656578064, 0.5714285969734192, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.75, 0.7321428656578064, 0.7857142686843872, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.8214285969734192, 0.75, 0.7142857313156128, 0.75, 0.8214285969734192, 0.875, 0.8214285969734192, 0.75, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9642857313156128, 1.0], [0.0714285746216774, 0.1071428582072258, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.3571428656578064, 0.5, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, 0.5, 0.5, NaN, 0.6071428656578064, 0.4642857015132904, 0.5357142686843872, 0.5357142686843872, 0.5714285969734192, 0.5, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.7142857313156128, 0.6071428656578064, 0.5, 0.6428571343421936, 0.6785714030265808, 0.5714285969734192, 0.6071428656578064, 0.5714285969734192, 0.75, 0.5357142686843872, 0.6785714030265808, 0.6785714030265808, 0.6071428656578064, 0.8214285969734192, 0.6785714030265808, 0.7142857313156128, 0.6428571343421936, 0.75, 0.6428571343421936, 0.7857142686843872, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8571428656578064, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.75, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.1428571492433548, 0.1785714328289032, 0.2857142984867096, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.25, 0.2142857164144516, 0.3035714328289032, 0.25, 0.3571428656578064, 0.3214285671710968, 0.4285714328289032, 0.3928571343421936, 0.3214285671710968, 0.4285714328289032, 0.5, 0.5, 0.375, 0.3928571343421936, NaN, 0.4642857015132904, 0.5714285969734192, 0.3928571343421936, 0.4285714328289032, 0.6428571343421936, 0.5357142686843872, 0.6071428656578064, 0.5, 0.4642857015132904, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.5357142686843872, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.5357142686843872, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.5714285969734192, 0.6785714030265808, 0.75, 0.7857142686843872, 0.75, 0.7857142686843872, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.6785714030265808, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.6785714030265808, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.7857142686843872, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9285714030265808, 1.0], [0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.25, 0.25, 0.25, 0.1785714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.25, 0.3571428656578064, 0.5, 0.5357142686843872, 0.4642857015132904, 0.3571428656578064, 0.4642857015132904, 0.6071428656578064, 0.5357142686843872, 0.5357142686843872, NaN, 0.5714285969734192, 0.5357142686843872, 0.5357142686843872, 0.4285714328289032, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.5357142686843872, 0.5714285969734192, 0.6785714030265808, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.7321428656578064, 0.6428571343421936, 0.6071428656578064, 0.7142857313156128, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.75, 0.6428571343421936, 0.75, 0.6785714030265808, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.75, 0.75, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.9642857313156128, 0.75, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128], [0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.25, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3928571343421936, 0.4285714328289032, 0.3928571343421936, 0.3928571343421936, 0.5, 0.5, 0.4642857015132904, 0.4285714328289032, 0.4285714328289032, NaN, 0.5, 0.5714285969734192, 0.5357142686843872, 0.5714285969734192, 0.5, 0.5, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.5357142686843872, 0.6428571343421936, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6071428656578064, 0.75, 0.7857142686843872, 0.75, 0.75, 0.75, 0.75, 0.6785714030265808, 0.8928571343421936, 0.75, 0.7857142686843872, 0.9285714030265808, 0.7857142686843872, 0.75, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.9642857313156128, 0.8928571343421936, 0.9285714030265808, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 1.0, 1.0], [0.0357142873108387, 0.1071428582072258, 0.25, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.3928571343421936, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.3214285671710968, 0.2857142984867096, 0.5, 0.5357142686843872, 0.5, 0.5357142686843872, 0.4642857015132904, 0.6071428656578064, 0.4642857015132904, 0.5, NaN, 0.5357142686843872, 0.5, 0.6428571343421936, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.5714285969734192, 0.6785714030265808, 0.5357142686843872, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7857142686843872, 0.6071428656578064, 0.7142857313156128, 0.6071428656578064, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.7857142686843872, 0.7142857313156128, 0.75, 0.75, 0.6428571343421936, 0.6428571343421936, 0.75, 0.8214285969734192, 0.9285714030265808, 0.75, 0.7142857313156128, 0.75, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.9642857313156128, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128], [0.0357142873108387, 0.1785714328289032, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.1428571492433548, 0.25, 0.1785714328289032, 0.1428571492433548, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, 0.4642857015132904, 0.5, 0.5357142686843872, 0.5357142686843872, 0.4285714328289032, 0.5714285969734192, 0.4642857015132904, 0.4285714328289032, 0.4642857015132904, NaN, 0.4642857015132904, 0.5, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.5357142686843872, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.5, 0.6785714030265808, 0.6428571343421936, 0.5357142686843872, 0.6071428656578064, 0.7142857313156128, 0.5714285969734192, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.6071428656578064, 0.6785714030265808, 0.8214285969734192, 0.6785714030265808, 0.75, 0.75, 0.7142857313156128, 0.75, 0.8214285969734192, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.75, 0.8214285969734192, 0.7142857313156128, 0.8928571343421936, 0.8571428656578064, 0.75, 0.8571428656578064, 0.8928571343421936, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.9285714030265808, 0.8571428656578064, 0.8214285969734192, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 1.0], [0.0714285746216774, 0.2857142984867096, 0.0714285746216774, 0.1785714328289032, 0.3214285671710968, 0.2857142984867096, 0.1785714328289032, 0.25, 0.3928571343421936, 0.3928571343421936, 0.3214285671710968, 0.3392857015132904, 0.3571428656578064, 0.3571428656578064, 0.4642857015132904, 0.5, 0.5, 0.4464285671710968, 0.5, 0.4642857015132904, 0.5, 0.3571428656578064, 0.5714285969734192, 0.4642857015132904, 0.5, 0.5357142686843872, NaN, 0.5, 0.4642857015132904, 0.5, 0.5357142686843872, 0.5357142686843872, 0.5714285969734192, 0.5, 0.5714285969734192, 0.5714285969734192, 0.3928571343421936, 0.4642857015132904, 0.5357142686843872, 0.5357142686843872, 0.5357142686843872, 0.5357142686843872, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.5714285969734192, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.75, 0.75, 0.75, 0.7678571343421936, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.7142857313156128, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.25, 0.1785714328289032, 0.2857142984867096, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.3035714328289032, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.3571428656578064, 0.4642857015132904, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, 0.3571428656578064, 0.5, 0.5, NaN, 0.5357142686843872, 0.5, 0.5, 0.5714285969734192, 0.4642857015132904, 0.5, 0.6071428656578064, 0.5, 0.5714285969734192, 0.6428571343421936, 0.4285714328289032, 0.5714285969734192, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.5714285969734192, 0.6428571343421936, 0.5714285969734192, 0.5, 0.75, 0.7857142686843872, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.75, 0.75, 0.7142857313156128, 0.6428571343421936, 0.75, 0.75, 0.8571428656578064, 0.6785714030265808, 0.7142857313156128, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.9642857313156128, 0.7857142686843872, 0.9285714030265808, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 1.0], [0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.1071428582072258, 0.2857142984867096, 0.25, 0.3928571343421936, 0.2857142984867096, 0.2857142984867096, 0.25, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.3571428656578064, 0.4642857015132904, 0.3928571343421936, 0.4642857015132904, 0.3928571343421936, 0.3928571343421936, 0.5, 0.4285714328289032, 0.4642857015132904, 0.5357142686843872, 0.4642857015132904, NaN, 0.5357142686843872, 0.6071428656578064, 0.5714285969734192, 0.5714285969734192, 0.5, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.5714285969734192, 0.4821428656578064, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.625, 0.75, 0.6428571343421936, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.75, 0.7142857313156128, 0.75, 0.7142857313156128, 0.75, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.8928571343421936, 1.0], [0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.1071428582072258, 0.1428571492433548, 0.25, 0.25, 0.1785714328289032, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.4642857015132904, 0.3571428656578064, 0.5, 0.3928571343421936, 0.5, 0.4285714328289032, 0.4285714328289032, 0.5, 0.5, 0.4642857015132904, NaN, 0.5, 0.5, 0.5714285969734192, 0.6428571343421936, 0.5714285969734192, 0.5, 0.5714285969734192, 0.5, 0.5714285969734192, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.625, 0.75, 0.7142857313156128, 0.75, 0.8571428656578064, 0.7857142686843872, 0.7142857313156128, 0.9642857313156128, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.7857142686843872, 0.7857142686843872, 0.9285714030265808, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 1.0], [0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1785714328289032, 0.0714285746216774, 0.1428571492433548, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.2142857164144516, 0.25, 0.3571428656578064, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.3928571343421936, 0.2857142984867096, 0.5357142686843872, 0.3571428656578064, 0.3214285671710968, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.5, 0.3928571343421936, 0.5, NaN, 0.5714285969734192, 0.4285714328289032, 0.5, 0.5357142686843872, 0.5, 0.6071428656578064, 0.5357142686843872, 0.5, 0.5, 0.7142857313156128, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.75, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.75, 0.75, 0.75, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.9285714030265808, 0.7857142686843872, 0.8571428656578064, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 1.0], [0.0714285746216774, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.3214285671710968, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.4285714328289032, 0.5, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.5, 0.5714285969734192, 0.4642857015132904, 0.5, 0.5, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.4642857015132904, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.75, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.75, 0.6785714030265808, 0.7142857313156128, 0.8214285969734192, 0.7857142686843872, 0.7142857313156128, 0.75, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128, 0.9642857313156128], [0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.2857142984867096, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2857142984867096, 0.3928571343421936, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.4642857015132904, 0.4642857015132904, 0.5, 0.4285714328289032, 0.4285714328289032, 0.3214285671710968, 0.4285714328289032, 0.4642857015132904, 0.4285714328289032, 0.5357142686843872, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5, 0.6428571343421936, 0.6071428656578064, 0.5, 0.5357142686843872, 0.5357142686843872, 0.3928571343421936, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6428571343421936, 0.5714285969734192, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6785714030265808, 0.5714285969734192, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.75, 0.7857142686843872, 0.75, 0.75, 0.75, 0.75, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.5, 0.5, 0.5, 0.3571428656578064, 0.5, 0.5714285969734192, 0.5, NaN, 0.5357142686843872, 0.4642857015132904, 0.5357142686843872, 0.5, 0.5357142686843872, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.6785714030265808, 0.6071428656578064, 0.7857142686843872, 0.6428571343421936, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.75, 0.8928571343421936, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.75, 0.8571428656578064, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 1.0], [0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.3214285671710968, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3571428656578064, 0.3571428656578064, 0.25, 0.3214285671710968, 0.3571428656578064, 0.4285714328289032, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.4285714328289032, 0.3928571343421936, 0.5, 0.4285714328289032, 0.4642857015132904, 0.5, 0.3571428656578064, 0.4642857015132904, NaN, 0.4285714328289032, 0.5, 0.5714285969734192, 0.4285714328289032, 0.4642857015132904, 0.6071428656578064, 0.4642857015132904, 0.6071428656578064, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.6071428656578064, 0.5714285969734192, 0.7142857313156128, 0.6428571343421936, 0.75, 0.75, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.75, 0.8571428656578064, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.7857142686843872, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.9642857313156128, 0.9642857313156128, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.2857142984867096, 0.25, 0.2857142984867096, 0.2857142984867096, 0.1785714328289032, 0.2857142984867096, 0.3571428656578064, 0.3928571343421936, 0.2142857164144516, 0.2857142984867096, 0.4285714328289032, 0.4642857015132904, 0.4285714328289032, 0.4642857015132904, 0.3571428656578064, 0.3928571343421936, 0.4285714328289032, 0.5, 0.4285714328289032, 0.5, 0.4285714328289032, 0.5, 0.5, 0.4285714328289032, 0.3928571343421936, 0.5357142686843872, 0.5714285969734192, NaN, 0.5357142686843872, 0.5, 0.5, 0.5714285969734192, 0.5, 0.6071428656578064, 0.5714285969734192, 0.5714285969734192, 0.4821428656578064, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.7142857313156128, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.7857142686843872, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.75, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.9285714030265808, 0.75, 0.75, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.9285714030265808, 0.9285714030265808, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808], [0.0714285746216774, 0.1071428582072258, 0.0535714291036129, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.25, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.25, 0.3571428656578064, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.6071428656578064, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.5357142686843872, 0.5, 0.4642857015132904, 0.5, 0.4642857015132904, NaN, 0.5714285969734192, 0.4642857015132904, 0.5357142686843872, 0.5714285969734192, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.75, 0.5357142686843872, 0.5714285969734192, 0.5, 0.7142857313156128, 0.6428571343421936, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.6071428656578064, 0.6785714030265808, 0.75, 0.6785714030265808, 0.8214285969734192, 0.6785714030265808, 0.6964285969734192, 0.6785714030265808, 0.7857142686843872, 0.6785714030265808, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.9642857313156128, 1.0], [0.1071428582072258, 0.25, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.25, 0.2857142984867096, 0.1785714328289032, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.4464285671710968, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.4642857015132904, 0.3571428656578064, 0.5357142686843872, 0.3571428656578064, 0.3928571343421936, 0.5, 0.4642857015132904, 0.5, 0.4642857015132904, 0.5, 0.4285714328289032, 0.5, 0.4285714328289032, NaN, 0.5, 0.5357142686843872, 0.5714285969734192, 0.5, 0.5357142686843872, 0.5357142686843872, 0.6071428656578064, 0.5714285969734192, 0.6428571343421936, 0.5357142686843872, 0.7142857313156128, 0.6785714030265808, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.6071428656578064, 0.6071428656578064, 0.7142857313156128, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.6785714030265808, 0.75, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.7142857313156128, 0.8214285969734192, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 1.0], [0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.25, 0.25, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.3214285671710968, 0.5, 0.4285714328289032, 0.25, 0.3214285671710968, 0.2678571343421936, 0.3928571343421936, 0.3571428656578064, 0.4642857015132904, 0.4642857015132904, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5, 0.5, 0.4642857015132904, 0.4642857015132904, 0.5714285969734192, 0.5, 0.5357142686843872, 0.5, NaN, 0.5, 0.6071428656578064, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.5714285969734192, 0.8214285969734192, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.75, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.75, 0.6785714030265808, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.75, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.2142857164144516, 0.25, 0.3571428656578064, 0.2142857164144516, 0.1785714328289032, 0.25, 0.4285714328289032, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.3571428656578064, 0.4642857015132904, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, 0.5178571343421936, 0.4642857015132904, 0.5, 0.4285714328289032, 0.6071428656578064, 0.5, 0.5357142686843872, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.5, NaN, 0.6428571343421936, 0.5357142686843872, 0.6071428656578064, 0.5357142686843872, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6428571343421936, 0.7857142686843872, 0.6428571343421936, 0.75, 0.7857142686843872, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.75, 0.6785714030265808, 0.75, 0.7857142686843872, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.75, 0.8214285969734192, 0.7857142686843872, 0.8035714030265808, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.8214285969734192, 0.9642857313156128], [0.1428571492433548, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3571428656578064, 0.3928571343421936, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.3571428656578064, 0.3928571343421936, 0.2857142984867096, 0.4642857015132904, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.2857142984867096, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.5, 0.4285714328289032, 0.4285714328289032, 0.3928571343421936, 0.3571428656578064, NaN, 0.3928571343421936, 0.5357142686843872, 0.5357142686843872, 0.4642857015132904, 0.6071428656578064, 0.6071428656578064, 0.5, 0.6071428656578064, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.7142857313156128, 0.8928571343421936, 0.7857142686843872, 0.7857142686843872, 0.9285714030265808, 0.7857142686843872, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.2857142984867096, 0.3928571343421936, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.5357142686843872, 0.3928571343421936, 0.3571428656578064, 0.5, 0.4285714328289032, 0.4642857015132904, 0.6071428656578064, NaN, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.5, 0.6785714030265808, 0.6428571343421936, 0.5714285969734192, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.9285714030265808, 0.8928571343421936, 1.0], [0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.2857142984867096, 0.1428571492433548, 0.25, 0.2142857164144516, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.4285714328289032, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.2857142984867096, 0.4285714328289032, 0.3214285671710968, 0.4285714328289032, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.3928571343421936, 0.4285714328289032, 0.3571428656578064, 0.4642857015132904, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.5, NaN, 0.5, 0.4285714328289032, 0.5, 0.4642857015132904, 0.5, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.5714285969734192, 0.75, 0.5714285969734192, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.75, 0.75, 0.6071428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8928571343421936, 0.8928571343421936, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.9642857313156128, 1.0], [0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.25, 0.3928571343421936, 0.1785714328289032, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4642857015132904, 0.4285714328289032, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.5357142686843872, 0.3928571343421936, 0.3571428656578064, 0.2857142984867096, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.3928571343421936, 0.4642857015132904, 0.4642857015132904, 0.4285714328289032, 0.5, NaN, 0.5, 0.6071428656578064, 0.5714285969734192, 0.5, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.6428571343421936, 0.5714285969734192, 0.6785714030265808, 0.6785714030265808, 0.5714285969734192, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.7857142686843872, 0.7142857313156128, 0.8214285969734192, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.8928571343421936, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.3928571343421936, 0.2857142984867096, 0.5178571343421936, 0.25, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, 0.5357142686843872, 0.4285714328289032, 0.5714285969734192, 0.5, NaN, 0.5, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.6428571343421936, 0.6785714030265808, 0.5357142686843872, 0.6785714030265808, 0.6785714030265808, 0.5357142686843872, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.75, 0.7857142686843872, 0.75, 0.8214285969734192, 0.8571428656578064, 0.8214285969734192, 0.7857142686843872, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8928571343421936, 0.8928571343421936, 0.9285714030265808, 0.9285714030265808, 0.9642857313156128, 0.9285714030265808, 1.0], [0.0357142873108387, 0.1428571492433548, 0.1428571492433548, 0.25, 0.25, 0.2142857164144516, 0.1785714328289032, 0.25, 0.25, 0.3571428656578064, 0.2142857164144516, 0.3928571343421936, 0.3214285671710968, 0.3571428656578064, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.3571428656578064, 0.3214285671710968, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.3928571343421936, 0.2142857164144516, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4642857015132904, 0.3214285671710968, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.3928571343421936, 0.3928571343421936, 0.5, 0.3928571343421936, 0.5, NaN, 0.4285714328289032, 0.5357142686843872, 0.5714285969734192, 0.4642857015132904, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.75, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.6428571343421936, 0.6785714030265808, 0.7142857313156128, 0.75, 0.7142857313156128, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.75, 0.8214285969734192, 0.8571428656578064, 0.75, 0.7142857313156128, 0.75, 0.8214285969734192, 0.9285714030265808, 0.9642857313156128], [0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.25, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.5, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3571428656578064, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.4642857015132904, 0.3928571343421936, 0.5, 0.5357142686843872, 0.4285714328289032, 0.4642857015132904, 0.5714285969734192, NaN, 0.5, 0.6071428656578064, 0.7142857313156128, 0.6071428656578064, 0.5714285969734192, 0.6071428656578064, 0.7142857313156128, 0.6071428656578064, 0.6785714030265808, 0.7142857313156128, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.9285714030265808, 0.75, 0.7857142686843872, 0.8928571343421936, 0.7142857313156128, 0.8571428656578064, 0.8214285969734192, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128, 0.9285714030265808, 0.9642857313156128], [0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.25, 0.2857142984867096, 0.3571428656578064, 0.2142857164144516, 0.2142857164144516, 0.3214285671710968, 0.2857142984867096, 0.25, 0.25, 0.25, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.3928571343421936, 0.25, 0.3214285671710968, 0.2857142984867096, 0.25, 0.4285714328289032, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.5, 0.4642857015132904, 0.1785714328289032, 0.4642857015132904, 0.5, 0.3214285671710968, 0.5, 0.5, 0.4642857015132904, 0.4642857015132904, 0.5, NaN, 0.5714285969734192, 0.4285714328289032, 0.5357142686843872, 0.5714285969734192, 0.5357142686843872, 0.5714285969734192, 0.6785714030265808, 0.6071428656578064, 0.5714285969734192, 0.6428571343421936, 0.75, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.7857142686843872, 0.7142857313156128, 0.75, 0.7142857313156128, 0.7142857313156128, 0.7857142686843872, 0.6428571343421936, 0.75, 0.75, 0.75, 0.8214285969734192, 0.8214285969734192, 0.9642857313156128, 0.9285714030265808, 1.0], [0.1071428582072258, 0.1428571492433548, 0.0357142873108387, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.0357142873108387, 0.1785714328289032, 0.1428571492433548, 0.25, 0.1428571492433548, 0.25, 0.25, 0.25, 0.25, 0.3214285671710968, 0.25, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.2142857164144516, 0.3571428656578064, 0.25, 0.3928571343421936, 0.3214285671710968, 0.3571428656578064, 0.2142857164144516, 0.375, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, NaN, 0.5357142686843872, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5, 0.6071428656578064, 0.7142857313156128, 0.5, 0.5357142686843872, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.75, 0.75, 0.8571428656578064, 0.7142857313156128, 0.9642857313156128, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.8214285969734192, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128], [0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.25, 0.3571428656578064, 0.2142857164144516, 0.1785714328289032, 0.25, 0.2857142984867096, 0.2142857164144516, 0.25, 0.25, 0.25, 0.3571428656578064, 0.25, 0.2857142984867096, 0.3928571343421936, 0.25, 0.375, 0.2857142984867096, 0.3571428656578064, 0.25, 0.3928571343421936, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.3571428656578064, 0.5357142686843872, 0.2857142984867096, 0.5714285969734192, 0.4642857015132904, NaN, 0.5178571343421936, 0.5714285969734192, 0.5714285969734192, 0.5, 0.6785714030265808, 0.5714285969734192, 0.5535714030265808, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.75, 0.7142857313156128, 0.8928571343421936, 0.7142857313156128, 0.75, 0.8928571343421936, 0.6428571343421936, 0.7857142686843872, 0.8571428656578064, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8928571343421936, 0.9285714030265808, 0.9642857313156128], [0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.0357142873108387, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2142857164144516, 0.25, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.25, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.25, 0.3214285671710968, 0.25, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.25, 0.3928571343421936, 0.2142857164144516, 0.3571428656578064, 0.4285714328289032, 0.3214285671710968, 0.3928571343421936, 0.4642857015132904, 0.2857142984867096, 0.3928571343421936, 0.3214285671710968, 0.4642857015132904, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, 0.4821428656578064, NaN, 0.4642857015132904, 0.6964285969734192, 0.4642857015132904, 0.6071428656578064, 0.4642857015132904, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.5357142686843872, 0.6428571343421936, 0.75, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.6428571343421936, 0.8928571343421936, 0.75, 0.75, 0.8214285969734192, 0.7142857313156128, 0.9285714030265808, 0.7142857313156128, 0.7857142686843872, 0.8214285969734192, 0.7857142686843872, 0.9285714030265808, 1.0], [0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.25, 0.2857142984867096, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.2678571343421936, 0.2857142984867096, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2857142984867096, 0.4285714328289032, 0.3571428656578064, 0.2142857164144516, 0.2857142984867096, 0.3928571343421936, 0.2142857164144516, 0.4285714328289032, 0.3571428656578064, 0.25, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.25, 0.3214285671710968, 0.4642857015132904, 0.4285714328289032, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5357142686843872, NaN, 0.5, 0.4642857015132904, 0.7142857313156128, 0.5357142686843872, 0.5, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.7857142686843872, 0.75, 0.6428571343421936, 0.7142857313156128, 0.7857142686843872, 0.75, 0.7857142686843872, 0.7857142686843872, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8928571343421936, 0.9642857313156128], [0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.25, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.25, 0.3928571343421936, 0.25, 0.3214285671710968, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.2142857164144516, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.25, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.3928571343421936, 0.4642857015132904, 0.4285714328289032, 0.4285714328289032, 0.3035714328289032, 0.5, NaN, 0.4642857015132904, 0.6428571343421936, 0.4642857015132904, 0.4642857015132904, 0.4642857015132904, 0.6428571343421936, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.75, 0.6071428656578064, 0.6071428656578064, 0.7857142686843872, 0.6428571343421936, 0.7142857313156128, 0.7857142686843872, 0.6785714030265808, 0.8571428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.9642857313156128, 0.9642857313156128], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1071428582072258, 0.2857142984867096, 0.1785714328289032, 0.3571428656578064, 0.3571428656578064, 0.2142857164144516, 0.1428571492433548, 0.25, 0.2857142984867096, 0.2142857164144516, 0.1785714328289032, 0.2857142984867096, 0.2857142984867096, 0.3928571343421936, 0.3928571343421936, 0.3214285671710968, 0.2142857164144516, 0.3214285671710968, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4642857015132904, 0.3928571343421936, 0.2857142984867096, 0.4285714328289032, 0.5, 0.5, 0.5357142686843872, 0.5357142686843872, 0.5357142686843872, NaN, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.5357142686843872, 0.5714285969734192, 0.6785714030265808, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.6785714030265808, 0.75, 0.6785714030265808, 0.75, 0.75, 0.7142857313156128, 0.7857142686843872, 0.7857142686843872, 0.8214285969734192, 0.8214285969734192, 0.8214285969734192, 0.9285714030265808], [0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.25, 0.1607142835855484, 0.2857142984867096, 0.2857142984867096, 0.25, 0.25, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.25, 0.3214285671710968, 0.2857142984867096, 0.25, 0.25, 0.25, 0.2857142984867096, 0.25, 0.25, 0.2142857164144516, 0.25, 0.3214285671710968, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.3214285671710968, 0.3928571343421936, 0.25, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, NaN, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.5535714030265808, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.5714285969734192, 0.9285714030265808, 0.5714285969734192, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.5714285969734192, 0.6785714030265808, 0.6428571343421936, 0.6428571343421936, 0.6071428656578064, 0.6785714030265808, 0.7142857313156128, 0.75, 0.8214285969734192, 0.9285714030265808], [0.1071428582072258, 0.0714285746216774, 0.0, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.25, 0.2142857164144516, 0.1428571492433548, 0.25, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2857142984867096, 0.25, 0.2857142984867096, 0.2857142984867096, 0.25, 0.2857142984867096, 0.25, 0.25, 0.25, 0.2142857164144516, 0.25, 0.2857142984867096, 0.3571428656578064, 0.25, 0.3571428656578064, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3214285671710968, 0.25, 0.3214285671710968, 0.3928571343421936, 0.2857142984867096, 0.4285714328289032, 0.5357142686843872, 0.4642857015132904, 0.5357142686843872, 0.4285714328289032, 0.5714285969734192, NaN, 0.4642857015132904, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.6428571343421936, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6428571343421936, 0.7857142686843872, 0.75, 0.7142857313156128, 0.75, 0.75, 0.8214285969734192, 0.8571428656578064, 0.9285714030265808], [0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.25, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.0714285746216774, 0.3571428656578064, 0.2142857164144516, 0.3214285671710968, 0.2857142984867096, 0.25, 0.0357142873108387, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.1071428582072258, 0.3571428656578064, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.2142857164144516, 0.3928571343421936, 0.25, 0.3571428656578064, 0.4285714328289032, 0.4642857015132904, 0.3571428656578064, 0.2857142984867096, 0.4285714328289032, 0.5, 0.4464285671710968, 0.4642857015132904, 0.5, 0.5357142686843872, 0.3928571343421936, 0.6071428656578064, 0.5357142686843872, NaN, 0.5357142686843872, 0.6071428656578064, 0.6785714030265808, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.6785714030265808, 0.6071428656578064, 0.6428571343421936, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6785714030265808, 0.75, 0.6785714030265808, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8928571343421936, 0.9285714030265808], [0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.25, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.25, 0.1428571492433548, 0.1785714328289032, 0.25, 0.25, 0.2857142984867096, 0.2142857164144516, 0.25, 0.2142857164144516, 0.1428571492433548, 0.25, 0.2142857164144516, 0.3571428656578064, 0.1785714328289032, 0.2857142984867096, 0.3571428656578064, 0.2857142984867096, 0.1785714328289032, 0.2142857164144516, 0.25, 0.25, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.1785714328289032, 0.3571428656578064, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.3571428656578064, 0.4642857015132904, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.5357142686843872, 0.4642857015132904, 0.5357142686843872, 0.4642857015132904, 0.4642857015132904, NaN, 0.5714285969734192, 0.4285714328289032, 0.5357142686843872, 0.75, 0.5, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.7857142686843872, 0.75, 0.6071428656578064, 0.8571428656578064, 0.6428571343421936, 0.75, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.75, 0.8928571343421936, 0.9642857313156128], [0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.25, 0.1785714328289032, 0.25, 0.2857142984867096, 0.25, 0.2142857164144516, 0.25, 0.25, 0.1785714328289032, 0.2857142984867096, 0.25, 0.2857142984867096, 0.25, 0.3214285671710968, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.2142857164144516, 0.25, 0.3214285671710968, 0.25, 0.2142857164144516, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.25, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.25, 0.3928571343421936, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.4464285671710968, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, NaN, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.5714285969734192, 0.9642857313156128, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.5357142686843872, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.75, 0.8214285969734192, 0.9285714030265808], [0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.0714285746216774, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.25, 0.1428571492433548, 0.1785714328289032, 0.25, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.25, 0.25, 0.2857142984867096, 0.1785714328289032, 0.2142857164144516, 0.2857142984867096, 0.25, 0.2142857164144516, 0.2857142984867096, 0.3571428656578064, 0.3035714328289032, 0.3214285671710968, 0.3214285671710968, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.4642857015132904, 0.3571428656578064, 0.4642857015132904, 0.3214285671710968, 0.6071428656578064, 0.3928571343421936, 0.3214285671710968, 0.5714285969734192, 0.5714285969734192, NaN, 0.5, 0.5, 0.5357142686843872, 0.6785714030265808, 0.5, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.75, 0.6071428656578064, 0.6785714030265808, 0.6785714030265808, 0.7857142686843872, 0.8214285969734192, 0.9285714030265808], [0.0357142873108387, 0.0, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0, 0.1071428582072258, 0.0, 0.0714285746216774, 0.0892857164144516, 0.1071428582072258, 0.1785714328289032, 0.1607142835855484, 0.1964285671710968, 0.0714285746216774, 0.0357142873108387, 0.1785714328289032, 0.2142857164144516, 0.25, 0.1785714328289032, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.0714285746216774, 0.2857142984867096, 0.25, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.25, 0.1785714328289032, 0.25, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.25, 0.3214285671710968, 0.2857142984867096, 0.25, 0.3214285671710968, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.3571428656578064, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, 0.5357142686843872, 0.5, NaN, 0.4642857015132904, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.6785714030265808, 0.5714285969734192, 0.7142857313156128, 0.6785714030265808, 0.5714285969734192, 0.6428571343421936, 0.6785714030265808, 0.75, 0.75, 0.7857142686843872, 0.8214285969734192, 0.9642857313156128], [0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.1428571492433548, 0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.125, 0.1607142835855484, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.25, 0.1964285671710968, 0.1428571492433548, 0.125, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.1071428582072258, 0.25, 0.1071428582072258, 0.2321428507566452, 0.3214285671710968, 0.25, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.25, 0.2142857164144516, 0.3214285671710968, 0.2142857164144516, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.25, 0.3214285671710968, 0.25, 0.3571428656578064, 0.25, 0.2857142984867096, 0.3928571343421936, 0.3571428656578064, 0.25, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.5357142686843872, 0.4285714328289032, 0.4285714328289032, 0.25, 0.5357142686843872, 0.5, 0.5357142686843872, NaN, 0.5, 0.6428571343421936, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.75, 0.6785714030265808, 0.7678571343421936, 0.7142857313156128, 0.75, 0.6071428656578064, 0.7142857313156128, 0.6785714030265808, 0.7142857313156128, 0.8571428656578064, 0.9642857313156128], [0.0, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.25, 0.2857142984867096, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.2857142984867096, 0.1428571492433548, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.25, 0.25, 0.25, 0.3214285671710968, 0.2857142984867096, 0.3928571343421936, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.3214285671710968, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.3928571343421936, 0.5, 0.4285714328289032, 0.4642857015132904, 0.4285714328289032, 0.5, NaN, 0.4285714328289032, 0.75, 0.5, 0.5357142686843872, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.6071428656578064, 0.6071428656578064, 0.8214285969734192, 0.7142857313156128, 0.7142857313156128, 0.75, 0.7142857313156128, 0.8928571343421936], [0.0714285746216774, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.25, 0.25, 0.25, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.25, 0.25, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.25, 0.2857142984867096, 0.3214285671710968, 0.2142857164144516, 0.3214285671710968, 0.25, 0.2857142984867096, 0.2857142984867096, 0.25, 0.3214285671710968, 0.0714285746216774, 0.3571428656578064, 0.3214285671710968, 0.3214285671710968, 0.0357142873108387, 0.3214285671710968, 0.2857142984867096, 0.3571428656578064, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.5, 0.3214285671710968, 0.5714285969734192, 0.6071428656578064, 0.5, 0.5, 0.6071428656578064, 0.6785714030265808, 0.9285714030265808], [0.0, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.1428571492433548, 0.0357142873108387, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1071428582072258, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.25, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.25, 0.2857142984867096, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.1785714328289032, 0.25, 0.2142857164144516, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3928571343421936, 0.3214285671710968, 0.4285714328289032, 0.3928571343421936, 0.3928571343421936, 0.3571428656578064, 0.4285714328289032, 0.5, 0.4285714328289032, 0.4285714328289032, 0.25, 0.4285714328289032, NaN, 0.5, 0.5, 0.5714285969734192, 0.6428571343421936, 0.5714285969734192, 0.5714285969734192, 0.5357142686843872, 0.7321428656578064, 0.6785714030265808, 0.6428571343421936, 0.7142857313156128, 0.6785714030265808, 0.8928571343421936], [0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.0357142873108387, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.2857142984867096, 0.0714285746216774, 0.25, 0.2857142984867096, 0.1071428582072258, 0.3571428656578064, 0.2142857164144516, 0.3928571343421936, 0.3214285671710968, 0.4642857015132904, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.5, 0.5714285969734192, 0.5, NaN, 0.6071428656578064, 0.5, 0.6071428656578064, 0.5357142686843872, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6071428656578064, 0.6785714030265808, 0.75, 0.7857142686843872, 0.9285714030265808], [0.0357142873108387, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.1428571492433548, 0.2142857164144516, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.25, 0.0714285746216774, 0.25, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.2857142984867096, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.25, 0.1428571492433548, 0.2142857164144516, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.25, 0.2857142984867096, 0.25, 0.2857142984867096, 0.1071428582072258, 0.25, 0.2142857164144516, 0.3214285671710968, 0.3571428656578064, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.3928571343421936, 0.3928571343421936, 0.3214285671710968, 0.2857142984867096, 0.4642857015132904, 0.6071428656578064, 0.5, 0.3928571343421936, NaN, 0.4642857015132904, 0.5, 0.5714285969734192, 0.5357142686843872, 0.5, 0.5357142686843872, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.75, 0.9285714030265808], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.1428571492433548, 0.1071428582072258, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.25, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.0357142873108387, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.25, 0.25, 0.25, 0.3571428656578064, 0.3571428656578064, 0.25, 0.3571428656578064, 0.3214285671710968, 0.2857142984867096, 0.25, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.25, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5, 0.5357142686843872, NaN, 0.5, 0.5357142686843872, 0.4642857015132904, 0.4285714328289032, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.7142857313156128, 0.8928571343421936], [0.0, 0.0714285746216774, 0.0, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.1428571492433548, 0.0357142873108387, 0.1785714328289032, 0.2142857164144516, 0.1785714328289032, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.25, 0.0714285746216774, 0.1428571492433548, 0.2142857164144516, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.1071428582072258, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.4285714328289032, 0.25, 0.2857142984867096, 0.3928571343421936, 0.4642857015132904, 0.3928571343421936, 0.2857142984867096, 0.3214285671710968, 0.3928571343421936, 0.5357142686843872, 0.3571428656578064, 0.3928571343421936, 0.5, 0.5, NaN, 0.5357142686843872, 0.5357142686843872, 0.5357142686843872, 0.5357142686843872, 0.6428571343421936, 0.6428571343421936, 0.7142857313156128, 0.75, 0.8928571343421936], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0892857164144516, 0.1071428582072258, 0.0357142873108387, 0.1071428582072258, 0.0714285746216774, 0.2142857164144516, 0.1607142835855484, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.0714285746216774, 0.1785714328289032, 0.0357142873108387, 0.1428571492433548, 0.0714285746216774, 0.2142857164144516, 0.2142857164144516, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.2142857164144516, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.2142857164144516, 0.1785714328289032, 0.2142857164144516, 0.2857142984867096, 0.1785714328289032, 0.2142857164144516, 0.2857142984867096, 0.25, 0.2857142984867096, 0.3571428656578064, 0.2857142984867096, 0.3571428656578064, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.25, 0.3214285671710968, 0.3571428656578064, 0.3214285671710968, 0.1428571492433548, 0.3571428656578064, 0.3928571343421936, 0.3214285671710968, 0.2321428507566452, 0.3571428656578064, 0.5, 0.4285714328289032, 0.4642857015132904, 0.4285714328289032, 0.4642857015132904, 0.4642857015132904, NaN, 0.5, 0.6071428656578064, 0.5, 0.6428571343421936, 0.6428571343421936, 0.6785714030265808, 0.7857142686843872, 0.9642857313156128], [0.0, 0.0, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.2142857164144516, 0.1071428582072258, 0.2142857164144516, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.25, 0.0357142873108387, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3214285671710968, 0.25, 0.3571428656578064, 0.2142857164144516, 0.25, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.4285714328289032, 0.2857142984867096, 0.3928571343421936, 0.6785714030265808, 0.4285714328289032, 0.3928571343421936, 0.4642857015132904, 0.5357142686843872, 0.4642857015132904, 0.5, NaN, 0.5714285969734192, 0.6428571343421936, 0.5714285969734192, 0.6071428656578064, 0.6428571343421936, 0.7857142686843872, 0.9285714030265808], [0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.1071428582072258, 0.2857142984867096, 0.2142857164144516, 0.1964285671710968, 0.1071428582072258, 0.25, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.25, 0.2142857164144516, 0.1428571492433548, 0.0714285746216774, 0.2142857164144516, 0.1428571492433548, 0.2857142984867096, 0.3571428656578064, 0.25, 0.3214285671710968, 0.25, 0.3928571343421936, 0.25, 0.3571428656578064, 0.25, 0.3928571343421936, 0.4285714328289032, 0.4642857015132904, 0.3928571343421936, 0.5, 0.5714285969734192, 0.4642857015132904, 0.3928571343421936, 0.4285714328289032, NaN, 0.5, 0.5714285969734192, 0.5714285969734192, 0.6428571343421936, 0.9642857313156128, 0.9642857313156128], [0.0, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.1785714328289032, 0.0714285746216774, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.1071428582072258, 0.1785714328289032, 0.1071428582072258, 0.25, 0.1428571492433548, 0.25, 0.1428571492433548, 0.2142857164144516, 0.2857142984867096, 0.2142857164144516, 0.3214285671710968, 0.2142857164144516, 0.3928571343421936, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, 0.3214285671710968, 0.3928571343421936, 0.1785714328289032, 0.3928571343421936, 0.2678571343421936, 0.3928571343421936, 0.4642857015132904, 0.5, 0.4642857015132904, 0.5, 0.3571428656578064, 0.5, NaN, 0.5714285969734192, 0.5714285969734192, 0.6071428656578064, 0.6071428656578064, 0.8571428656578064], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.0357142873108387, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.1071428582072258, 0.2142857164144516, 0.1071428582072258, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.0714285746216774, 0.2857142984867096, 0.1428571492433548, 0.1785714328289032, 0.2142857164144516, 0.2142857164144516, 0.2142857164144516, 0.1428571492433548, 0.3214285671710968, 0.2142857164144516, 0.3214285671710968, 0.25, 0.2857142984867096, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.25, 0.2857142984867096, 0.2857142984867096, 0.5, 0.3214285671710968, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.3571428656578064, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.7142857313156128, 0.7142857313156128, 0.6428571343421936, 0.8571428656578064], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0535714291036129, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.1428571492433548, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.1428571492433548, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.0714285746216774, 0.0714285746216774, 0.25, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.3214285671710968, 0.1785714328289032, 0.2857142984867096, 0.25, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.3214285671710968, 0.25, 0.3214285671710968, 0.2857142984867096, 0.5, 0.3571428656578064, 0.3214285671710968, 0.4285714328289032, 0.4285714328289032, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, NaN, 0.6071428656578064, 0.6071428656578064, 0.8214285969734192], [0.0, 0.0, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.1428571492433548, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1428571492433548, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.1428571492433548, 0.1785714328289032, 0.0357142873108387, 0.2142857164144516, 0.0357142873108387, 0.0714285746216774, 0.1428571492433548, 0.1071428582072258, 0.0357142873108387, 0.1785714328289032, 0.0357142873108387, 0.0357142873108387, 0.1428571492433548, 0.1071428582072258, 0.2142857164144516, 0.1428571492433548, 0.2142857164144516, 0.1785714328289032, 0.25, 0.1785714328289032, 0.1428571492433548, 0.25, 0.25, 0.2142857164144516, 0.2142857164144516, 0.2857142984867096, 0.25, 0.3928571343421936, 0.2857142984867096, 0.25, 0.3571428656578064, 0.3571428656578064, 0.2857142984867096, 0.3214285671710968, 0.3571428656578064, 0.3571428656578064, 0.3928571343421936, 0.2857142984867096, 0.3928571343421936, NaN, 0.6071428656578064, 0.75], [0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.0, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.1428571492433548, 0.0714285746216774, 0.1785714328289032, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.0357142873108387, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.1071428582072258, 0.1071428582072258, 0.1785714328289032, 0.1785714328289032, 0.1785714328289032, 0.1428571492433548, 0.2857142984867096, 0.3214285671710968, 0.3214285671710968, 0.2142857164144516, 0.25, 0.2857142984867096, 0.25, 0.2142857164144516, 0.2142857164144516, 0.0357142873108387, 0.3928571343421936, 0.3571428656578064, 0.3928571343421936, 0.3928571343421936, NaN, 0.9285714030265808], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0, 0.0, 0.0714285746216774, 0.0, 0.0, 0.0, 0.0357142873108387, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0, 0.0357142873108387, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0714285746216774, 0.0714285746216774, 0.0357142873108387, 0.0357142873108387, 0.1071428582072258, 0.0714285746216774, 0.1071428582072258, 0.0714285746216774, 0.0714285746216774, 0.1071428582072258, 0.1071428582072258, 0.0357142873108387, 0.0714285746216774, 0.0357142873108387, 0.1428571492433548, 0.1428571492433548, 0.1785714328289032, 0.25, 0.0714285746216774, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9243970273},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8443877556},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8390538046},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8304730982},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8223562157},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8205009269},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8042671629},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8042671618},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7393320981},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.732374769},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7258812633},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7077922059},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.6873840463},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6801948036},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6727736556},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6535250455},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6433209623},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.6229128042},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6159554718},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.6147959182},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.6145640087},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6141001811},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6138682738},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6108534312},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.60992579},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6052875708},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5753710562},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5749072373},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5714285738},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.569805196},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5621521336},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.5598330247},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5551948052},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5477736537},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5422077952},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5415120594},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5375695734},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5359461957},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5354823761},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5310760644},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.5064935033},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5018552883},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.4958256061},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4939703159},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4914192935},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4893320972},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4884044535},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4466604843},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4408627113},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.432050094},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4253246746},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4199907245},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3954081621},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3835807063},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.3780148429},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3780148442},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3745361795},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3743042688},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.3680426738},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.3392857164},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.3341836766},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3297773675},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3297773687},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.2852504664},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2794526926},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2611317281},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2523191117},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.2504638248},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.2500000012},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2442022278},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2323747692},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2288961068},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2182282031},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1966604854},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1815862739},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1525974054},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1219851594},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0329313554}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_classification/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..53b8fb595009609278a76cde283de77c2f24eb84
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_classification/datasets_small/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,TabFM (default),TabPFN-3 (default),RealTabPFN-2.5 (default),RealTabPFN-2.5 (tuned + ensembled),TabPFN-2.6 (default),"AutoGluon 1.5 (extreme, 4h)",TabICLv2 (default),RealTabPFN-2.5 (tuned),RealMLP (tuned + ensembled),LimiX (default),"AutoGluon 1.4 (best, 4h)",TabPFNv2 (tuned + ensembled),Mitra (default),RealMLP (tuned),TabDPT-Turbo (default),TabDPT (tuned + ensembled),TabM (tuned + ensembled),TabPFNv2 (tuned),CatBoost (tuned),TabICL (default),BetaTabPFN (default),iLTM (tuned + ensembled),TabM (tuned),LightGBM (tuned + ensembled),TabDPT (tuned),CatBoost (tuned + ensembled),TabPFNv2 (default),TabSwift (default),ModernNCA (tuned),XGBoost (tuned + ensembled),CatBoost (default),xRFM (tuned + ensembled),LightGBM (tuned),XGBoost (tuned),EBM (tuned + ensembled),TabM (default),iLTM (tuned),SAP-RPT-OSS (default),ChimeraBoost (tuned + ensembled),ModernNCA (tuned + ensembled),RealMLP (default),ChimeraBoost (tuned),xRFM (tuned),EBM (tuned),EBM (default),TabDPT (default),TorchMLP (tuned + ensembled),ChimeraBoost (default),FastaiMLP (tuned + ensembled),TorchMLP (tuned),ExtraTrees (tuned + ensembled),ModernNCA (default),ExtraTrees (tuned),XGBoost (default),TabSTAR (tuned),FastaiMLP (tuned),LightGBM (default),RandomForest (tuned + ensembled),TabSTAR (tuned + ensembled),iLTM (default),OrionMSP (default),RandomForest (tuned),PerpetualBooster (tuned + ensembled),TabSTAR (default),PerpetualBooster (tuned),TorchMLP (default),ExtraTrees (default),xRFM (default),TabFlex (default),RandomForest (default),FastaiMLP (default),KNN (tuned + ensembled),PerpetualBooster (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),KNN (tuned),KNN (default)
+TabFM (default),,0.85714287,0.8214286,0.89285713,0.85714287,0.8214286,0.89285713,0.89285713,0.8214286,0.89285713,0.89285713,0.96428573,1.0,0.85714287,0.89285713,0.9285714,0.89285713,0.96428573,0.9285714,0.96428573,0.9285714,0.85714287,0.9285714,0.9285714,0.96428573,0.96428573,0.9285714,0.89285713,0.9285714,0.9285714,0.89285713,0.9285714,0.9285714,0.89285713,0.89285713,0.89285713,0.9285714,0.89285713,0.9285714,0.89285713,0.85714287,0.89285713,0.9285714,0.89285713,0.89285713,0.96428573,0.96428573,0.9285714,0.89285713,0.96428573,0.9285714,0.9285714,0.9285714,0.96428573,0.89285713,0.89285713,0.9285714,0.9285714,0.85714287,0.89285713,0.96428573,0.9285714,1.0,0.9285714,1.0,0.96428573,0.96428573,0.96428573,1.0,0.96428573,1.0,0.9285714,1.0,0.96428573,0.96428573,1.0,0.96428573,1.0
+TabPFN-3 (default),0.14285715,,0.4642857,0.5535714,0.53571427,0.53571427,0.5714286,0.60714287,0.71428573,0.6785714,0.5714286,0.71428573,0.8214286,0.75,0.75,0.75,0.78571427,0.89285713,0.75,0.9285714,0.89285713,0.8214286,0.85714287,0.89285713,0.89285713,0.8214286,0.71428573,0.85714287,0.85714287,0.89285713,0.89285713,0.85714287,0.9285714,0.85714287,0.78571427,0.78571427,0.89285713,0.75,0.89285713,0.9285714,0.89285713,0.85714287,0.9285714,0.85714287,0.85714287,0.85714287,0.9285714,0.9285714,0.85714287,0.9285714,0.96428573,0.9285714,0.96428573,0.96428573,0.8214286,0.9285714,0.9285714,0.9285714,0.8214286,0.96428573,1.0,0.96428573,0.96428573,0.8214286,1.0,0.9285714,0.96428573,0.96428573,0.9285714,0.96428573,1.0,0.96428573,0.96428573,0.96428573,0.96428573,1.0,0.96428573,1.0
+RealTabPFN-2.5 (default),0.17857143,0.53571427,,0.42857143,0.42857143,0.5714286,0.5,0.48214287,0.71428573,0.75,0.6785714,0.5714286,0.78571427,0.78571427,0.6785714,0.75,0.78571427,0.85714287,0.8214286,0.6785714,0.78571427,0.71428573,0.8214286,0.89285713,0.75,0.85714287,0.9285714,0.85714287,0.8214286,0.89285713,0.89285713,0.85714287,0.9285714,0.89285713,0.85714287,0.85714287,0.9464286,0.8214286,0.85714287,0.8214286,0.85714287,0.85714287,0.9285714,0.9285714,0.89285713,0.85714287,0.9285714,0.9285714,0.96428573,0.89285713,0.85714287,0.9285714,0.85714287,0.96428573,0.85714287,1.0,0.96428573,0.89285713,0.85714287,0.96428573,1.0,0.96428573,0.9285714,0.85714287,1.0,0.96428573,1.0,0.96428573,1.0,0.96428573,1.0,0.9285714,0.9285714,0.96428573,0.96428573,1.0,0.96428573,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.10714286,0.44642857,0.5714286,,0.5714286,0.53571427,0.60714287,0.64285713,0.71428573,0.71428573,0.64285713,0.85714287,0.8214286,0.75,0.60714287,0.64285713,0.75,0.9285714,0.89285713,0.71428573,0.8214286,0.8214286,0.78571427,0.89285713,0.71428573,0.9285714,0.8214286,0.78571427,0.89285713,0.9285714,0.9285714,0.85714287,0.9285714,0.89285713,0.8214286,0.8214286,0.85714287,0.85714287,0.85714287,0.89285713,0.8214286,0.89285713,0.9285714,0.8214286,0.85714287,0.75,0.85714287,0.85714287,0.85714287,0.85714287,0.9285714,0.85714287,0.9285714,1.0,0.78571427,0.89285713,0.96428573,0.9285714,0.78571427,0.89285713,0.96428573,0.9285714,0.9285714,0.8214286,0.96428573,0.9285714,0.96428573,0.96428573,0.85714287,0.96428573,0.96428573,0.9285714,0.9285714,0.9285714,0.96428573,0.9285714,0.9285714,1.0
+TabPFN-2.6 (default),0.14285715,0.4642857,0.5714286,0.42857143,,0.42857143,0.42857143,0.5,0.6785714,0.75,0.64285713,0.75,0.8214286,0.75,0.71428573,0.64285713,0.71428573,0.8214286,0.78571427,0.85714287,0.8214286,0.89285713,0.75,0.89285713,0.71428573,0.85714287,0.6785714,0.78571427,0.85714287,0.85714287,0.9285714,0.89285713,0.96428573,0.89285713,0.8214286,0.78571427,0.89285713,0.78571427,0.8214286,0.9285714,0.85714287,0.8214286,0.85714287,0.89285713,0.8214286,0.75,0.9285714,0.89285713,0.89285713,0.9285714,0.89285713,0.9285714,0.89285713,0.96428573,0.78571427,0.85714287,0.9285714,0.9285714,0.78571427,0.9285714,0.96428573,0.89285713,0.96428573,0.78571427,1.0,0.96428573,0.9285714,0.96428573,0.89285713,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.17857143,0.4642857,0.42857143,0.4642857,0.5714286,,0.4642857,0.5714286,0.75,0.6785714,0.71428573,0.6785714,0.71428573,0.75,0.5714286,0.75,0.78571427,0.8214286,0.78571427,0.78571427,0.78571427,0.8214286,0.75,0.85714287,0.78571427,0.85714287,0.71428573,0.75,0.89285713,0.89285713,0.89285713,0.89285713,0.89285713,0.89285713,0.78571427,0.8214286,0.89285713,0.75,0.89285713,0.89285713,0.8214286,0.89285713,0.8214286,0.8214286,0.85714287,0.78571427,0.89285713,0.9285714,0.8214286,0.85714287,0.89285713,0.9285714,0.89285713,0.9285714,0.8214286,0.89285713,0.96428573,0.9285714,0.8214286,0.9285714,1.0,0.9285714,0.89285713,0.8214286,0.9285714,0.9285714,0.96428573,0.96428573,0.9285714,0.9285714,0.96428573,0.96428573,0.9285714,0.96428573,0.96428573,0.96428573,1.0,1.0
+TabICLv2 (default),0.10714286,0.42857143,0.5,0.39285713,0.5714286,0.53571427,,0.53571427,0.6785714,0.64285713,0.75,0.6785714,0.71428573,0.60714287,0.71428573,0.78571427,0.71428573,0.75,0.75,0.78571427,0.85714287,0.78571427,0.75,0.8214286,0.78571427,0.75,0.8214286,0.8214286,0.78571427,0.78571427,0.8214286,0.8214286,0.8214286,0.78571427,0.78571427,0.78571427,0.75,0.71428573,0.85714287,0.89285713,0.78571427,0.8214286,0.89285713,0.8214286,0.85714287,0.8214286,0.89285713,0.85714287,0.96428573,0.9285714,0.8214286,0.89285713,0.89285713,0.9285714,0.8214286,0.89285713,0.9285714,0.85714287,0.8214286,0.8214286,0.89285713,0.85714287,0.9285714,0.85714287,0.96428573,0.96428573,0.89285713,0.9285714,0.96428573,0.9285714,1.0,0.9285714,0.96428573,0.96428573,0.96428573,1.0,0.96428573,1.0
+RealTabPFN-2.5 (tuned),0.10714286,0.39285713,0.51785713,0.35714287,0.5,0.42857143,0.4642857,,0.64285713,0.5714286,0.64285713,0.625,0.8214286,0.75,0.60714287,0.60714287,0.8214286,0.9285714,0.78571427,0.78571427,0.8214286,0.75,0.8214286,0.78571427,0.71428573,0.8214286,0.75,0.71428573,0.89285713,0.89285713,0.9285714,0.8214286,0.89285713,0.85714287,0.8214286,0.8214286,0.89285713,0.8214286,0.78571427,0.85714287,0.8214286,0.85714287,0.89285713,0.8214286,0.8214286,0.75,0.85714287,0.85714287,0.8214286,0.89285713,0.96428573,0.89285713,0.9285714,0.96428573,0.78571427,0.85714287,0.9285714,0.9285714,0.78571427,0.89285713,1.0,0.89285713,0.9285714,0.8214286,0.96428573,0.9285714,0.96428573,0.96428573,0.85714287,0.96428573,0.9285714,0.9285714,0.9285714,0.9285714,0.9285714,0.9285714,0.9285714,1.0
+RealMLP (tuned + ensembled),0.17857143,0.2857143,0.2857143,0.2857143,0.32142857,0.25,0.32142857,0.35714287,,0.39285713,0.5714286,0.4642857,0.53571427,0.60714287,0.53571427,0.53571427,0.6785714,0.64285713,0.75,0.64285713,0.6785714,0.78571427,0.71428573,0.78571427,0.60714287,0.85714287,0.60714287,0.64285713,0.71428573,0.85714287,0.85714287,0.6785714,0.8214286,0.85714287,0.8214286,0.71428573,0.8214286,0.71428573,0.78571427,0.78571427,0.85714287,0.78571427,0.71428573,0.8214286,0.8214286,0.75,0.8214286,0.85714287,0.85714287,0.78571427,0.78571427,0.89285713,0.8214286,0.9285714,0.75,0.8214286,0.89285713,0.75,0.75,0.9285714,0.9285714,0.85714287,0.9285714,0.78571427,0.96428573,0.89285713,0.96428573,0.9285714,0.85714287,0.96428573,0.96428573,0.9285714,0.96428573,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+LimiX (default),0.10714286,0.32142857,0.25,0.2857143,0.25,0.32142857,0.35714287,0.42857143,0.60714287,,0.5,0.60714287,0.53571427,0.6964286,0.53571427,0.4642857,0.6785714,0.6785714,0.60714287,0.60714287,0.6785714,0.6964286,0.71428573,0.75,0.60714287,0.6785714,0.60714287,0.60714287,0.75,0.75,0.6785714,0.75,0.71428573,0.78571427,0.78571427,0.75,0.8214286,0.75,0.75,0.75,0.8214286,0.78571427,0.85714287,0.85714287,0.85714287,0.64285713,0.78571427,0.8214286,0.75,0.78571427,0.85714287,0.85714287,0.8214286,0.89285713,0.83928573,0.85714287,0.85714287,0.85714287,0.8214286,0.89285713,0.91071427,0.9285714,0.85714287,0.85714287,0.85714287,0.89285713,0.96428573,0.9285714,0.89285713,0.96428573,0.96428573,0.9285714,0.89285713,0.96428573,0.96428573,0.96428573,0.9285714,1.0
+"AutoGluon 1.4 (best, 4h)",0.10714286,0.42857143,0.32142857,0.35714287,0.35714287,0.2857143,0.25,0.35714287,0.42857143,0.5,,0.42857143,0.53571427,0.60714287,0.5714286,0.5,0.60714287,0.5714286,0.6785714,0.60714287,0.6785714,0.75,0.71428573,0.75,0.60714287,0.71428573,0.6785714,0.60714287,0.60714287,0.75,0.71428573,0.71428573,0.78571427,0.71428573,0.6785714,0.71428573,0.71428573,0.6785714,0.75,0.64285713,0.71428573,0.78571427,0.75,0.85714287,0.8214286,0.78571427,0.85714287,0.78571427,0.85714287,0.89285713,0.78571427,0.89285713,0.78571427,0.89285713,0.71428573,0.89285713,0.89285713,0.8214286,0.75,0.89285713,0.89285713,0.9285714,0.9285714,0.8214286,0.96428573,0.96428573,0.96428573,0.89285713,0.9285714,0.96428573,0.96428573,0.89285713,0.96428573,0.96428573,0.96428573,1.0,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.035714287,0.2857143,0.42857143,0.14285715,0.25,0.32142857,0.32142857,0.375,0.53571427,0.39285713,0.5714286,,0.5535714,0.53571427,0.5,0.53571427,0.6785714,0.71428573,0.6785714,0.625,0.71428573,0.64285713,0.6785714,0.71428573,0.5714286,0.6785714,0.66071427,0.64285713,0.71428573,0.8214286,0.6785714,0.71428573,0.75,0.78571427,0.78571427,0.71428573,0.64285713,0.60714287,0.71428573,0.78571427,0.78571427,0.78571427,0.78571427,0.78571427,0.78571427,0.60714287,0.78571427,0.8214286,0.75,0.85714287,0.78571427,0.78571427,0.85714287,0.89285713,0.71428573,0.8214286,0.9285714,0.85714287,0.71428573,0.85714287,0.8214286,0.875,0.8214286,0.75,0.89285713,0.89285713,0.85714287,0.85714287,0.8214286,0.91071427,0.9285714,0.89285713,0.9285714,0.96428573,0.96428573,0.9285714,0.9285714,1.0
+Mitra (default),0.0,0.17857143,0.21428572,0.17857143,0.17857143,0.2857143,0.2857143,0.17857143,0.4642857,0.4642857,0.4642857,0.44642857,,0.5714286,0.5,0.53571427,0.71428573,0.625,0.60714287,0.64285713,0.64285713,0.6785714,0.75,0.71428573,0.53571427,0.64285713,0.64285713,0.64285713,0.71428573,0.6785714,0.78571427,0.6785714,0.78571427,0.71428573,0.71428573,0.8214286,0.6785714,0.60714287,0.75,0.8214286,0.71428573,0.8214286,0.6785714,0.71428573,0.75,0.6785714,0.71428573,0.78571427,0.75,0.89285713,0.75,0.85714287,0.8214286,0.8214286,0.75,0.75,0.8214286,0.75,0.75,0.8214286,0.83928573,0.83928573,0.8214286,0.78571427,0.8214286,0.85714287,0.78571427,0.9285714,0.85714287,0.89285713,0.89285713,0.89285713,0.9285714,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+RealMLP (tuned),0.14285715,0.25,0.21428572,0.25,0.25,0.25,0.39285713,0.25,0.39285713,0.30357143,0.39285713,0.4642857,0.42857143,,0.4642857,0.39285713,0.53571427,0.5714286,0.5714286,0.60714287,0.64285713,0.5714286,0.64285713,0.6785714,0.53571427,0.60714287,0.64285713,0.60714287,0.75,0.71428573,0.75,0.5714286,0.71428573,0.75,0.75,0.71428573,0.71428573,0.60714287,0.6785714,0.75,0.75,0.71428573,0.6785714,0.75,0.75,0.64285713,0.78571427,0.75,0.75,0.78571427,0.78571427,0.75,0.85714287,0.85714287,0.75,0.78571427,0.8214286,0.85714287,0.78571427,0.9285714,0.8035714,0.89285713,0.85714287,0.78571427,0.85714287,0.9285714,0.9285714,0.9285714,0.8214286,0.96428573,0.9285714,0.89285713,0.96428573,0.9285714,0.9464286,0.9285714,0.96428573,1.0
+TabDPT-Turbo (default),0.10714286,0.25,0.32142857,0.39285713,0.2857143,0.42857143,0.2857143,0.39285713,0.4642857,0.4642857,0.42857143,0.5,0.5,0.53571427,,0.71428573,0.5,0.5714286,0.5,0.5714286,0.5,0.60714287,0.5,0.60714287,0.6785714,0.53571427,0.53571427,0.6964286,0.5714286,0.64285713,0.64285713,0.60714287,0.60714287,0.6785714,0.64285713,0.64285713,0.71428573,0.64285713,0.6785714,0.5714286,0.64285713,0.6785714,0.71428573,0.6785714,0.75,0.78571427,0.6785714,0.71428573,0.75,0.75,0.78571427,0.71428573,0.8214286,0.8214286,0.75,0.85714287,0.85714287,0.8214286,0.75,0.8214286,0.9285714,0.8214286,0.75,0.78571427,0.8214286,0.85714287,0.9285714,0.89285713,0.96428573,0.89285713,0.96428573,0.96428573,0.9285714,0.89285713,0.89285713,0.89285713,0.96428573,1.0
+TabDPT (tuned + ensembled),0.071428575,0.25,0.25,0.35714287,0.35714287,0.25,0.21428572,0.39285713,0.4642857,0.53571427,0.5,0.4642857,0.4642857,0.60714287,0.2857143,,0.4642857,0.5714286,0.53571427,0.5714286,0.53571427,0.6785714,0.4642857,0.5714286,0.71428573,0.5714286,0.5,0.5714286,0.5714286,0.64285713,0.60714287,0.5714286,0.60714287,0.6785714,0.64285713,0.60714287,0.75,0.64285713,0.6785714,0.64285713,0.60714287,0.6785714,0.6785714,0.6785714,0.75,0.71428573,0.64285713,0.64285713,0.6785714,0.64285713,0.75,0.78571427,0.85714287,0.78571427,0.78571427,0.75,0.75,0.75,0.75,0.8214286,0.96428573,0.78571427,0.71428573,0.78571427,0.8214286,0.8214286,0.96428573,0.96428573,0.9285714,0.9285714,0.96428573,0.89285713,0.85714287,0.89285713,0.89285713,0.85714287,0.9285714,1.0
+TabM (tuned + ensembled),0.10714286,0.21428572,0.21428572,0.25,0.2857143,0.21428572,0.2857143,0.17857143,0.32142857,0.32142857,0.39285713,0.32142857,0.2857143,0.4642857,0.5,0.53571427,,0.35714287,0.5,0.60714287,0.60714287,0.5714286,0.53571427,0.60714287,0.5,0.53571427,0.5,0.60714287,0.60714287,0.64285713,0.64285713,0.5714286,0.60714287,0.71428573,0.75,0.78571427,0.64285713,0.64285713,0.64285713,0.6785714,0.75,0.78571427,0.5714286,0.6785714,0.75,0.71428573,0.71428573,0.78571427,0.75,0.78571427,0.6785714,0.78571427,0.6785714,0.78571427,0.78571427,0.78571427,0.78571427,0.75,0.8214286,0.8214286,0.8214286,0.75,0.78571427,0.8214286,0.89285713,0.96428573,0.78571427,0.85714287,0.89285713,0.78571427,0.89285713,0.85714287,0.9285714,0.9285714,0.9285714,0.9285714,1.0,1.0
+TabPFNv2 (tuned),0.035714287,0.10714286,0.14285715,0.071428575,0.17857143,0.17857143,0.25,0.071428575,0.35714287,0.32142857,0.42857143,0.2857143,0.375,0.42857143,0.42857143,0.42857143,0.64285713,,0.53571427,0.48214287,0.5714286,0.5,0.64285713,0.60714287,0.4642857,0.5,0.5535714,0.5714286,0.64285713,0.64285713,0.5714286,0.60714287,0.64285713,0.64285713,0.6785714,0.71428573,0.60714287,0.5535714,0.6785714,0.6785714,0.6785714,0.71428573,0.6785714,0.64285713,0.75,0.5714286,0.75,0.78571427,0.6785714,0.8214286,0.71428573,0.71428573,0.71428573,0.8214286,0.71428573,0.75,0.8214286,0.71428573,0.71428573,0.8214286,0.78571427,0.8035714,0.78571427,0.75,0.8214286,0.8214286,0.85714287,0.8214286,0.85714287,0.83928573,0.8214286,0.89285713,0.89285713,0.96428573,0.96428573,0.9285714,0.96428573,1.0
+CatBoost (tuned),0.071428575,0.25,0.17857143,0.10714286,0.21428572,0.21428572,0.25,0.21428572,0.25,0.39285713,0.32142857,0.32142857,0.39285713,0.42857143,0.5,0.4642857,0.5,0.4642857,,0.4642857,0.5,0.5,0.53571427,0.5,0.5,0.4642857,0.5,0.53571427,0.53571427,0.60714287,0.64285713,0.53571427,0.53571427,0.64285713,0.64285713,0.5714286,0.6785714,0.60714287,0.5,0.60714287,0.6785714,0.60714287,0.6785714,0.75,0.71428573,0.64285713,0.71428573,0.6785714,0.6785714,0.75,0.71428573,0.71428573,0.6785714,0.8214286,0.71428573,0.78571427,0.78571427,0.78571427,0.75,0.8214286,0.75,0.85714287,0.78571427,0.75,0.89285713,0.89285713,0.8214286,0.8214286,0.8214286,0.89285713,0.89285713,0.85714287,0.9285714,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+TabICL (default),0.035714287,0.071428575,0.32142857,0.2857143,0.14285715,0.21428572,0.21428572,0.21428572,0.35714287,0.39285713,0.39285713,0.375,0.35714287,0.39285713,0.42857143,0.42857143,0.39285713,0.51785713,0.53571427,,0.5,0.625,0.39285713,0.5,0.4642857,0.4642857,0.53571427,0.64285713,0.60714287,0.53571427,0.60714287,0.53571427,0.53571427,0.5714286,0.5714286,0.53571427,0.6785714,0.60714287,0.5714286,0.53571427,0.64285713,0.64285713,0.60714287,0.60714287,0.6785714,0.6785714,0.71428573,0.71428573,0.71428573,0.71428573,0.75,0.73214287,0.78571427,0.78571427,0.71428573,0.78571427,0.8214286,0.75,0.71428573,0.75,0.8214286,0.875,0.8214286,0.75,0.89285713,0.89285713,0.85714287,0.8214286,0.85714287,0.89285713,0.89285713,0.89285713,0.9285714,0.9285714,0.89285713,0.96428573,0.96428573,1.0
+BetaTabPFN (default),0.071428575,0.10714286,0.21428572,0.17857143,0.17857143,0.21428572,0.14285715,0.17857143,0.32142857,0.32142857,0.32142857,0.2857143,0.35714287,0.35714287,0.5,0.4642857,0.39285713,0.42857143,0.5,0.5,,0.60714287,0.4642857,0.53571427,0.53571427,0.5714286,0.5,0.53571427,0.53571427,0.64285713,0.71428573,0.60714287,0.5,0.64285713,0.6785714,0.5714286,0.60714287,0.5714286,0.75,0.53571427,0.6785714,0.6785714,0.60714287,0.8214286,0.6785714,0.71428573,0.64285713,0.75,0.64285713,0.78571427,0.71428573,0.71428573,0.71428573,0.78571427,0.75,0.71428573,0.78571427,0.78571427,0.75,0.85714287,0.75,0.78571427,0.78571427,0.8214286,0.8214286,0.89285713,0.75,0.8214286,0.85714287,0.8214286,0.89285713,0.89285713,0.85714287,0.96428573,0.96428573,0.96428573,0.96428573,1.0
+iLTM (tuned + ensembled),0.14285715,0.17857143,0.2857143,0.17857143,0.10714286,0.17857143,0.21428572,0.25,0.21428572,0.30357143,0.25,0.35714287,0.32142857,0.42857143,0.39285713,0.32142857,0.42857143,0.5,0.5,0.375,0.39285713,,0.4642857,0.5714286,0.39285713,0.42857143,0.64285713,0.53571427,0.60714287,0.5,0.4642857,0.5714286,0.5714286,0.60714287,0.60714287,0.53571427,0.6785714,0.60714287,0.6785714,0.53571427,0.64285713,0.6785714,0.71428573,0.6785714,0.71428573,0.5714286,0.6785714,0.75,0.78571427,0.75,0.78571427,0.78571427,0.71428573,0.8214286,0.6785714,0.78571427,0.8214286,0.85714287,0.6785714,0.8214286,0.78571427,0.85714287,0.85714287,0.78571427,0.9285714,0.9285714,0.9285714,0.89285713,0.89285713,0.9285714,0.9285714,0.9285714,0.9285714,0.89285713,0.9285714,0.9285714,0.9285714,1.0
+TabM (tuned),0.071428575,0.14285715,0.17857143,0.21428572,0.25,0.25,0.25,0.17857143,0.2857143,0.2857143,0.2857143,0.32142857,0.25,0.35714287,0.5,0.53571427,0.4642857,0.35714287,0.4642857,0.60714287,0.53571427,0.53571427,,0.5714286,0.53571427,0.53571427,0.42857143,0.60714287,0.60714287,0.60714287,0.64285713,0.53571427,0.5714286,0.6785714,0.64285713,0.64285713,0.64285713,0.5714286,0.73214287,0.64285713,0.60714287,0.71428573,0.5714286,0.64285713,0.64285713,0.71428573,0.64285713,0.75,0.64285713,0.75,0.6785714,0.78571427,0.6785714,0.78571427,0.71428573,0.78571427,0.78571427,0.75,0.75,0.75,0.78571427,0.78571427,0.85714287,0.78571427,0.89285713,0.96428573,0.75,0.8214286,0.89285713,0.8214286,0.8214286,0.85714287,0.9285714,0.89285713,0.89285713,0.9285714,0.96428573,0.96428573
+LightGBM (tuned + ensembled),0.071428575,0.10714286,0.10714286,0.10714286,0.10714286,0.14285715,0.17857143,0.21428572,0.21428572,0.25,0.25,0.2857143,0.2857143,0.32142857,0.39285713,0.42857143,0.39285713,0.39285713,0.5,0.5,0.4642857,0.42857143,0.42857143,,0.5,0.5714286,0.53571427,0.5714286,0.5,0.5,0.6785714,0.60714287,0.6785714,0.6785714,0.6785714,0.60714287,0.60714287,0.5714286,0.60714287,0.53571427,0.64285713,0.6785714,0.6785714,0.71428573,0.75,0.60714287,0.75,0.78571427,0.75,0.75,0.75,0.75,0.6785714,0.89285713,0.75,0.78571427,0.9285714,0.78571427,0.75,0.85714287,0.78571427,0.89285713,0.8214286,0.78571427,0.85714287,0.85714287,0.85714287,0.8214286,0.8214286,0.96428573,0.89285713,0.9285714,0.85714287,0.96428573,0.96428573,0.9285714,1.0,1.0
+TabDPT (tuned),0.035714287,0.10714286,0.25,0.2857143,0.2857143,0.21428572,0.21428572,0.2857143,0.39285713,0.39285713,0.39285713,0.42857143,0.4642857,0.4642857,0.32142857,0.2857143,0.5,0.53571427,0.5,0.53571427,0.4642857,0.60714287,0.4642857,0.5,,0.53571427,0.5,0.64285713,0.5714286,0.5714286,0.5714286,0.53571427,0.5714286,0.60714287,0.64285713,0.5714286,0.6785714,0.53571427,0.64285713,0.5714286,0.60714287,0.6785714,0.5714286,0.5714286,0.71428573,0.78571427,0.60714287,0.71428573,0.60714287,0.64285713,0.6785714,0.78571427,0.78571427,0.71428573,0.75,0.75,0.64285713,0.64285713,0.75,0.8214286,0.9285714,0.75,0.71428573,0.75,0.8214286,0.8214286,0.89285713,0.89285713,0.85714287,0.85714287,0.96428573,0.85714287,0.89285713,0.85714287,0.85714287,0.85714287,0.9285714,0.96428573
+CatBoost (tuned + ensembled),0.035714287,0.17857143,0.14285715,0.071428575,0.14285715,0.14285715,0.25,0.17857143,0.14285715,0.32142857,0.2857143,0.32142857,0.35714287,0.39285713,0.4642857,0.42857143,0.4642857,0.5,0.53571427,0.53571427,0.42857143,0.5714286,0.4642857,0.42857143,0.4642857,,0.4642857,0.5,0.53571427,0.5714286,0.60714287,0.53571427,0.53571427,0.60714287,0.60714287,0.5,0.6785714,0.64285713,0.53571427,0.60714287,0.71428573,0.5714286,0.64285713,0.71428573,0.71428573,0.60714287,0.6785714,0.8214286,0.6785714,0.75,0.75,0.71428573,0.75,0.8214286,0.75,0.78571427,0.78571427,0.8214286,0.75,0.8214286,0.71428573,0.89285713,0.85714287,0.75,0.85714287,0.89285713,0.8214286,0.85714287,0.78571427,0.9285714,0.85714287,0.8214286,0.9285714,0.96428573,0.96428573,0.96428573,0.9285714,1.0
+TabPFNv2 (default),0.071428575,0.2857143,0.071428575,0.17857143,0.32142857,0.2857143,0.17857143,0.25,0.39285713,0.39285713,0.32142857,0.3392857,0.35714287,0.35714287,0.4642857,0.5,0.5,0.44642857,0.5,0.4642857,0.5,0.35714287,0.5714286,0.4642857,0.5,0.53571427,,0.5,0.4642857,0.5,0.53571427,0.53571427,0.5714286,0.5,0.5714286,0.5714286,0.39285713,0.4642857,0.53571427,0.53571427,0.53571427,0.53571427,0.60714287,0.53571427,0.60714287,0.60714287,0.60714287,0.60714287,0.64285713,0.71428573,0.64285713,0.5714286,0.6785714,0.64285713,0.71428573,0.71428573,0.6785714,0.71428573,0.75,0.75,0.75,0.76785713,0.78571427,0.78571427,0.78571427,0.8214286,0.71428573,0.8214286,0.85714287,0.78571427,0.85714287,0.78571427,0.89285713,0.9285714,0.9285714,0.96428573,0.9285714,1.0
+TabSwift (default),0.10714286,0.14285715,0.14285715,0.21428572,0.21428572,0.25,0.17857143,0.2857143,0.35714287,0.39285713,0.39285713,0.35714287,0.35714287,0.39285713,0.30357143,0.42857143,0.39285713,0.42857143,0.4642857,0.35714287,0.4642857,0.4642857,0.39285713,0.42857143,0.35714287,0.5,0.5,,0.53571427,0.5,0.5,0.5714286,0.4642857,0.5,0.60714287,0.5,0.5714286,0.64285713,0.42857143,0.5714286,0.60714287,0.53571427,0.60714287,0.5714286,0.64285713,0.5714286,0.5,0.75,0.78571427,0.60714287,0.6785714,0.64285713,0.71428573,0.64285713,0.75,0.75,0.71428573,0.64285713,0.75,0.75,0.85714287,0.6785714,0.71428573,0.8214286,0.78571427,0.8214286,0.8214286,0.8214286,0.96428573,0.78571427,0.9285714,0.89285713,0.85714287,0.89285713,0.89285713,0.89285713,0.96428573,1.0
+ModernNCA (tuned),0.071428575,0.14285715,0.17857143,0.10714286,0.14285715,0.10714286,0.21428572,0.10714286,0.2857143,0.25,0.39285713,0.2857143,0.2857143,0.25,0.42857143,0.42857143,0.39285713,0.35714287,0.4642857,0.39285713,0.4642857,0.39285713,0.39285713,0.5,0.42857143,0.4642857,0.53571427,0.4642857,,0.53571427,0.60714287,0.5714286,0.5714286,0.5,0.5,0.5714286,0.5714286,0.60714287,0.5714286,0.48214287,0.60714287,0.64285713,0.64285713,0.64285713,0.5714286,0.5714286,0.60714287,0.6785714,0.625,0.75,0.64285713,0.78571427,0.6785714,0.78571427,0.75,0.71428573,0.75,0.71428573,0.75,0.71428573,0.78571427,0.75,0.78571427,0.85714287,0.8214286,0.8214286,0.8214286,0.85714287,0.8214286,0.89285713,0.8214286,0.89285713,0.8214286,0.89285713,0.9285714,0.89285713,0.89285713,1.0
+XGBoost (tuned + ensembled),0.071428575,0.10714286,0.10714286,0.071428575,0.14285715,0.10714286,0.21428572,0.10714286,0.14285715,0.25,0.25,0.17857143,0.32142857,0.2857143,0.35714287,0.35714287,0.35714287,0.35714287,0.39285713,0.4642857,0.35714287,0.5,0.39285713,0.5,0.42857143,0.42857143,0.5,0.5,0.4642857,,0.5,0.5,0.5714286,0.64285713,0.5714286,0.5,0.5714286,0.5,0.5714286,0.53571427,0.64285713,0.64285713,0.64285713,0.64285713,0.60714287,0.5714286,0.60714287,0.71428573,0.6785714,0.625,0.75,0.71428573,0.75,0.85714287,0.78571427,0.71428573,0.96428573,0.8214286,0.78571427,0.8214286,0.78571427,0.89285713,0.71428573,0.78571427,0.78571427,0.85714287,0.89285713,0.78571427,0.78571427,0.9285714,0.85714287,0.8214286,0.8214286,0.89285713,0.9285714,0.9285714,0.89285713,1.0
+CatBoost (default),0.10714286,0.10714286,0.10714286,0.071428575,0.071428575,0.10714286,0.17857143,0.071428575,0.14285715,0.32142857,0.2857143,0.32142857,0.21428572,0.25,0.35714287,0.39285713,0.35714287,0.42857143,0.35714287,0.39285713,0.2857143,0.53571427,0.35714287,0.32142857,0.42857143,0.39285713,0.4642857,0.5,0.39285713,0.5,,0.5714286,0.42857143,0.5,0.53571427,0.5,0.60714287,0.53571427,0.5,0.5,0.71428573,0.64285713,0.64285713,0.60714287,0.64285713,0.64285713,0.64285713,0.75,0.71428573,0.71428573,0.6785714,0.60714287,0.60714287,0.75,0.75,0.75,0.71428573,0.78571427,0.75,0.78571427,0.75,0.8214286,0.8214286,0.78571427,0.89285713,0.9285714,0.78571427,0.85714287,0.8214286,0.89285713,0.89285713,0.89285713,0.9285714,0.9285714,0.96428573,0.96428573,0.96428573,1.0
+xRFM (tuned + ensembled),0.071428575,0.14285715,0.14285715,0.14285715,0.10714286,0.10714286,0.17857143,0.17857143,0.32142857,0.25,0.2857143,0.2857143,0.32142857,0.42857143,0.39285713,0.42857143,0.42857143,0.39285713,0.4642857,0.4642857,0.39285713,0.42857143,0.4642857,0.39285713,0.4642857,0.4642857,0.4642857,0.42857143,0.42857143,0.5,0.42857143,,0.42857143,0.42857143,0.5,0.5714286,0.4642857,0.5,0.5,0.5714286,0.60714287,0.60714287,0.6785714,0.4642857,0.5714286,0.5714286,0.5714286,0.5714286,0.64285713,0.64285713,0.6785714,0.78571427,0.75,0.71428573,0.6785714,0.71428573,0.71428573,0.75,0.6785714,0.71428573,0.8214286,0.78571427,0.71428573,0.75,0.78571427,0.85714287,0.89285713,0.89285713,0.85714287,0.89285713,0.89285713,0.89285713,0.85714287,0.9285714,0.96428573,0.9285714,0.96428573,0.96428573
+LightGBM (tuned),0.071428575,0.071428575,0.071428575,0.071428575,0.035714287,0.10714286,0.17857143,0.10714286,0.17857143,0.2857143,0.21428572,0.25,0.21428572,0.2857143,0.39285713,0.39285713,0.39285713,0.35714287,0.4642857,0.4642857,0.5,0.42857143,0.42857143,0.32142857,0.42857143,0.4642857,0.42857143,0.53571427,0.42857143,0.42857143,0.5714286,0.5714286,,0.5,0.64285713,0.60714287,0.5,0.53571427,0.53571427,0.39285713,0.5714286,0.60714287,0.60714287,0.60714287,0.64285713,0.5714286,0.64285713,0.71428573,0.6785714,0.75,0.6785714,0.5714286,0.6785714,0.78571427,0.71428573,0.75,0.78571427,0.75,0.75,0.75,0.75,0.85714287,0.78571427,0.78571427,0.8214286,0.8214286,0.78571427,0.8214286,0.78571427,0.89285713,0.85714287,0.85714287,0.85714287,0.89285713,0.89285713,0.89285713,0.9285714,1.0
+XGBoost (tuned),0.10714286,0.14285715,0.10714286,0.10714286,0.10714286,0.10714286,0.21428572,0.14285715,0.14285715,0.21428572,0.2857143,0.21428572,0.2857143,0.25,0.32142857,0.32142857,0.2857143,0.35714287,0.35714287,0.42857143,0.35714287,0.39285713,0.32142857,0.32142857,0.39285713,0.39285713,0.5,0.5,0.5,0.35714287,0.5,0.5714286,0.5,,0.53571427,0.4642857,0.53571427,0.5,0.53571427,0.5,0.5714286,0.5714286,0.64285713,0.64285713,0.60714287,0.53571427,0.53571427,0.64285713,0.6785714,0.60714287,0.78571427,0.64285713,0.78571427,0.8214286,0.78571427,0.75,0.89285713,0.78571427,0.78571427,0.78571427,0.75,0.85714287,0.71428573,0.78571427,0.78571427,0.8214286,0.85714287,0.78571427,0.78571427,0.89285713,0.85714287,0.85714287,0.8214286,0.85714287,0.89285713,0.9285714,0.89285713,1.0
+EBM (tuned + ensembled),0.10714286,0.21428572,0.14285715,0.17857143,0.17857143,0.21428572,0.21428572,0.17857143,0.17857143,0.21428572,0.32142857,0.21428572,0.2857143,0.25,0.35714287,0.35714287,0.25,0.32142857,0.35714287,0.42857143,0.32142857,0.39285713,0.35714287,0.32142857,0.35714287,0.39285713,0.42857143,0.39285713,0.5,0.42857143,0.4642857,0.5,0.35714287,0.4642857,,0.42857143,0.5,0.5714286,0.42857143,0.4642857,0.60714287,0.4642857,0.60714287,0.71428573,0.71428573,0.6785714,0.60714287,0.5714286,0.71428573,0.64285713,0.75,0.75,0.6785714,0.71428573,0.6785714,0.75,0.64285713,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.78571427,0.75,0.85714287,0.8214286,0.8214286,0.89285713,0.78571427,0.85714287,0.89285713,0.85714287,0.96428573,0.96428573,0.96428573,0.9285714,1.0
+TabM (default),0.10714286,0.21428572,0.14285715,0.17857143,0.21428572,0.17857143,0.21428572,0.17857143,0.2857143,0.25,0.2857143,0.2857143,0.17857143,0.2857143,0.35714287,0.39285713,0.21428572,0.2857143,0.42857143,0.4642857,0.42857143,0.4642857,0.35714287,0.39285713,0.42857143,0.5,0.42857143,0.5,0.42857143,0.5,0.5,0.42857143,0.39285713,0.53571427,0.5714286,,0.53571427,0.5,0.5,0.5714286,0.5,0.60714287,0.5714286,0.5714286,0.48214287,0.60714287,0.5714286,0.60714287,0.6785714,0.71428573,0.60714287,0.6785714,0.64285713,0.71428573,0.64285713,0.78571427,0.64285713,0.71428573,0.6785714,0.64285713,0.71428573,0.75,0.78571427,0.78571427,0.78571427,0.9285714,0.75,0.75,0.8214286,0.78571427,0.78571427,0.78571427,0.85714287,0.9285714,0.9285714,0.89285713,0.96428573,0.9285714
+iLTM (tuned),0.071428575,0.10714286,0.05357143,0.14285715,0.10714286,0.10714286,0.25,0.10714286,0.17857143,0.17857143,0.2857143,0.35714287,0.32142857,0.2857143,0.2857143,0.25,0.35714287,0.39285713,0.32142857,0.32142857,0.39285713,0.32142857,0.35714287,0.39285713,0.32142857,0.32142857,0.60714287,0.42857143,0.42857143,0.42857143,0.39285713,0.53571427,0.5,0.4642857,0.5,0.4642857,,0.5714286,0.4642857,0.53571427,0.5714286,0.64285713,0.64285713,0.60714287,0.75,0.53571427,0.5714286,0.5,0.71428573,0.64285713,0.78571427,0.6785714,0.78571427,0.60714287,0.6785714,0.75,0.6785714,0.8214286,0.6785714,0.6964286,0.6785714,0.78571427,0.6785714,0.78571427,0.8214286,0.89285713,0.85714287,0.85714287,0.8214286,0.85714287,0.85714287,0.89285713,0.85714287,0.8214286,0.85714287,0.85714287,0.96428573,1.0
+SAP-RPT-OSS (default),0.10714286,0.25,0.17857143,0.14285715,0.21428572,0.25,0.2857143,0.17857143,0.2857143,0.25,0.32142857,0.39285713,0.39285713,0.39285713,0.35714287,0.35714287,0.35714287,0.44642857,0.39285713,0.39285713,0.42857143,0.39285713,0.42857143,0.42857143,0.4642857,0.35714287,0.53571427,0.35714287,0.39285713,0.5,0.4642857,0.5,0.4642857,0.5,0.42857143,0.5,0.42857143,,0.5,0.53571427,0.5714286,0.5,0.53571427,0.53571427,0.60714287,0.5714286,0.64285713,0.53571427,0.71428573,0.6785714,0.64285713,0.60714287,0.6785714,0.60714287,0.60714287,0.71428573,0.64285713,0.64285713,0.60714287,0.6785714,0.64285713,0.6785714,0.75,0.78571427,0.78571427,0.78571427,0.78571427,0.78571427,0.8214286,0.78571427,0.8214286,0.71428573,0.8214286,0.89285713,0.85714287,0.8214286,0.85714287,1.0
+ChimeraBoost (tuned + ensembled),0.071428575,0.10714286,0.14285715,0.14285715,0.17857143,0.10714286,0.14285715,0.21428572,0.21428572,0.25,0.25,0.2857143,0.25,0.32142857,0.32142857,0.32142857,0.35714287,0.32142857,0.5,0.42857143,0.25,0.32142857,0.26785713,0.39285713,0.35714287,0.4642857,0.4642857,0.5714286,0.42857143,0.42857143,0.5,0.5,0.4642857,0.4642857,0.5714286,0.5,0.53571427,0.5,,0.5,0.60714287,0.5714286,0.5714286,0.60714287,0.53571427,0.60714287,0.5714286,0.8214286,0.64285713,0.64285713,0.5714286,0.60714287,0.60714287,0.6785714,0.75,0.64285713,0.71428573,0.71428573,0.75,0.6785714,0.71428573,0.78571427,0.75,0.78571427,0.8214286,0.8214286,0.75,0.78571427,0.8214286,0.78571427,0.8214286,0.78571427,0.78571427,0.89285713,0.9285714,0.96428573,0.9285714,1.0
+ModernNCA (tuned + ensembled),0.10714286,0.071428575,0.17857143,0.10714286,0.071428575,0.10714286,0.10714286,0.14285715,0.21428572,0.25,0.35714287,0.21428572,0.17857143,0.25,0.42857143,0.35714287,0.32142857,0.32142857,0.39285713,0.4642857,0.4642857,0.4642857,0.35714287,0.4642857,0.42857143,0.39285713,0.4642857,0.42857143,0.51785713,0.4642857,0.5,0.42857143,0.60714287,0.5,0.53571427,0.42857143,0.4642857,0.4642857,0.5,,0.64285713,0.53571427,0.60714287,0.53571427,0.5714286,0.5714286,0.53571427,0.53571427,0.64285713,0.71428573,0.6785714,0.75,0.64285713,0.78571427,0.64285713,0.75,0.78571427,0.71428573,0.6785714,0.71428573,0.75,0.6785714,0.75,0.78571427,0.75,0.78571427,0.78571427,0.85714287,0.75,0.8214286,0.78571427,0.8035714,0.85714287,0.78571427,0.8214286,0.78571427,0.8214286,0.96428573
+RealMLP (default),0.14285715,0.10714286,0.14285715,0.17857143,0.14285715,0.17857143,0.21428572,0.17857143,0.14285715,0.17857143,0.2857143,0.21428572,0.2857143,0.25,0.35714287,0.39285713,0.25,0.32142857,0.32142857,0.35714287,0.32142857,0.35714287,0.39285713,0.35714287,0.39285713,0.2857143,0.4642857,0.39285713,0.39285713,0.35714287,0.2857143,0.39285713,0.42857143,0.42857143,0.39285713,0.5,0.42857143,0.42857143,0.39285713,0.35714287,,0.39285713,0.53571427,0.53571427,0.4642857,0.60714287,0.60714287,0.5,0.60714287,0.64285713,0.60714287,0.6785714,0.6785714,0.6785714,0.64285713,0.64285713,0.60714287,0.6785714,0.71428573,0.71428573,0.6785714,0.6785714,0.6785714,0.78571427,0.71428573,0.89285713,0.78571427,0.78571427,0.9285714,0.78571427,0.89285713,0.89285713,0.89285713,0.89285713,0.89285713,0.96428573,0.9285714,1.0
+ChimeraBoost (tuned),0.10714286,0.14285715,0.14285715,0.10714286,0.17857143,0.10714286,0.17857143,0.14285715,0.21428572,0.21428572,0.21428572,0.21428572,0.17857143,0.2857143,0.32142857,0.32142857,0.21428572,0.2857143,0.39285713,0.35714287,0.32142857,0.32142857,0.2857143,0.32142857,0.32142857,0.42857143,0.4642857,0.4642857,0.35714287,0.35714287,0.35714287,0.39285713,0.39285713,0.42857143,0.53571427,0.39285713,0.35714287,0.5,0.42857143,0.4642857,0.60714287,,0.5,0.5714286,0.5714286,0.60714287,0.5,0.6785714,0.64285713,0.5714286,0.53571427,0.53571427,0.64285713,0.71428573,0.71428573,0.6785714,0.75,0.64285713,0.71428573,0.71428573,0.71428573,0.78571427,0.71428573,0.78571427,0.78571427,0.8214286,0.78571427,0.78571427,0.85714287,0.71428573,0.78571427,0.75,0.78571427,0.8214286,0.8214286,0.9285714,0.89285713,1.0
+xRFM (tuned),0.071428575,0.071428575,0.071428575,0.071428575,0.14285715,0.17857143,0.10714286,0.10714286,0.2857143,0.14285715,0.25,0.21428572,0.32142857,0.32142857,0.2857143,0.32142857,0.42857143,0.32142857,0.32142857,0.39285713,0.39285713,0.2857143,0.42857143,0.32142857,0.42857143,0.35714287,0.39285713,0.39285713,0.35714287,0.35714287,0.35714287,0.32142857,0.39285713,0.35714287,0.39285713,0.42857143,0.35714287,0.4642857,0.42857143,0.39285713,0.4642857,0.5,,0.5,0.42857143,0.5,0.4642857,0.5,0.5714286,0.60714287,0.71428573,0.5714286,0.75,0.5714286,0.64285713,0.60714287,0.64285713,0.71428573,0.6785714,0.75,0.75,0.75,0.60714287,0.71428573,0.71428573,0.85714287,0.89285713,0.96428573,0.78571427,0.8214286,0.8214286,0.89285713,0.89285713,0.85714287,0.8214286,0.85714287,0.96428573,1.0
+EBM (tuned),0.10714286,0.14285715,0.071428575,0.17857143,0.10714286,0.17857143,0.17857143,0.17857143,0.17857143,0.14285715,0.14285715,0.21428572,0.2857143,0.25,0.32142857,0.32142857,0.32142857,0.35714287,0.25,0.39285713,0.17857143,0.32142857,0.35714287,0.2857143,0.42857143,0.2857143,0.4642857,0.42857143,0.35714287,0.35714287,0.39285713,0.53571427,0.39285713,0.35714287,0.2857143,0.42857143,0.39285713,0.4642857,0.39285713,0.4642857,0.4642857,0.42857143,0.5,,0.5,0.60714287,0.5714286,0.5,0.60714287,0.5714286,0.60714287,0.6785714,0.64285713,0.5714286,0.6785714,0.6785714,0.5714286,0.6785714,0.6785714,0.71428573,0.6785714,0.6785714,0.71428573,0.78571427,0.71428573,0.8214286,0.8214286,0.85714287,0.85714287,0.78571427,0.8214286,0.89285713,0.8214286,0.89285713,0.9285714,0.89285713,0.9285714,1.0
+EBM (default),0.10714286,0.14285715,0.10714286,0.14285715,0.17857143,0.14285715,0.14285715,0.17857143,0.17857143,0.14285715,0.17857143,0.21428572,0.25,0.25,0.25,0.25,0.25,0.25,0.2857143,0.32142857,0.32142857,0.2857143,0.35714287,0.25,0.2857143,0.2857143,0.39285713,0.35714287,0.42857143,0.39285713,0.35714287,0.42857143,0.35714287,0.39285713,0.2857143,0.51785713,0.25,0.39285713,0.4642857,0.42857143,0.53571427,0.42857143,0.5714286,0.5,,0.5,0.53571427,0.53571427,0.64285713,0.64285713,0.6785714,0.64285713,0.6785714,0.53571427,0.6785714,0.6785714,0.53571427,0.6785714,0.6785714,0.71428573,0.71428573,0.75,0.78571427,0.75,0.8214286,0.85714287,0.8214286,0.78571427,0.85714287,0.71428573,0.85714287,0.89285713,0.89285713,0.9285714,0.9285714,0.96428573,0.9285714,1.0
+TabDPT (default),0.035714287,0.14285715,0.14285715,0.25,0.25,0.21428572,0.17857143,0.25,0.25,0.35714287,0.21428572,0.39285713,0.32142857,0.35714287,0.21428572,0.2857143,0.2857143,0.42857143,0.35714287,0.32142857,0.2857143,0.42857143,0.2857143,0.39285713,0.21428572,0.39285713,0.39285713,0.42857143,0.42857143,0.42857143,0.35714287,0.42857143,0.42857143,0.4642857,0.32142857,0.39285713,0.4642857,0.42857143,0.39285713,0.42857143,0.39285713,0.39285713,0.5,0.39285713,0.5,,0.42857143,0.53571427,0.5714286,0.4642857,0.53571427,0.64285713,0.64285713,0.60714287,0.60714287,0.75,0.64285713,0.60714287,0.64285713,0.6785714,0.78571427,0.64285713,0.6785714,0.71428573,0.75,0.71428573,0.85714287,0.78571427,0.8214286,0.75,0.8214286,0.85714287,0.75,0.71428573,0.75,0.8214286,0.9285714,0.96428573
+TorchMLP (tuned + ensembled),0.035714287,0.071428575,0.071428575,0.14285715,0.071428575,0.10714286,0.10714286,0.14285715,0.17857143,0.21428572,0.14285715,0.21428572,0.2857143,0.21428572,0.32142857,0.35714287,0.2857143,0.25,0.2857143,0.2857143,0.35714287,0.32142857,0.35714287,0.25,0.39285713,0.32142857,0.39285713,0.5,0.39285713,0.39285713,0.35714287,0.42857143,0.35714287,0.4642857,0.39285713,0.42857143,0.42857143,0.35714287,0.42857143,0.4642857,0.39285713,0.5,0.53571427,0.42857143,0.4642857,0.5714286,,0.5,0.60714287,0.71428573,0.60714287,0.5714286,0.60714287,0.71428573,0.60714287,0.6785714,0.71428573,0.6785714,0.64285713,0.71428573,0.6785714,0.75,0.6785714,0.6785714,0.78571427,0.9285714,0.75,0.78571427,0.89285713,0.71428573,0.85714287,0.8214286,0.85714287,0.85714287,0.89285713,0.96428573,0.9285714,0.96428573
+ChimeraBoost (default),0.071428575,0.071428575,0.071428575,0.14285715,0.10714286,0.071428575,0.14285715,0.14285715,0.14285715,0.17857143,0.21428572,0.17857143,0.21428572,0.25,0.2857143,0.35714287,0.21428572,0.21428572,0.32142857,0.2857143,0.25,0.25,0.25,0.21428572,0.2857143,0.17857143,0.39285713,0.25,0.32142857,0.2857143,0.25,0.42857143,0.2857143,0.35714287,0.42857143,0.39285713,0.5,0.4642857,0.17857143,0.4642857,0.5,0.32142857,0.5,0.5,0.4642857,0.4642857,0.5,,0.5714286,0.42857143,0.53571427,0.5714286,0.53571427,0.5714286,0.6785714,0.60714287,0.5714286,0.64285713,0.75,0.6785714,0.64285713,0.71428573,0.64285713,0.78571427,0.71428573,0.75,0.71428573,0.71428573,0.78571427,0.64285713,0.75,0.75,0.75,0.8214286,0.8214286,0.96428573,0.9285714,1.0
+FastaiMLP (tuned + ensembled),0.10714286,0.14285715,0.035714287,0.14285715,0.10714286,0.17857143,0.035714287,0.17857143,0.14285715,0.25,0.14285715,0.25,0.25,0.25,0.25,0.32142857,0.25,0.32142857,0.32142857,0.2857143,0.35714287,0.21428572,0.35714287,0.25,0.39285713,0.32142857,0.35714287,0.21428572,0.375,0.32142857,0.2857143,0.35714287,0.32142857,0.32142857,0.2857143,0.32142857,0.2857143,0.2857143,0.35714287,0.35714287,0.39285713,0.35714287,0.42857143,0.39285713,0.35714287,0.42857143,0.39285713,0.42857143,,0.53571427,0.5714286,0.5714286,0.5714286,0.5,0.60714287,0.71428573,0.5,0.53571427,0.60714287,0.6785714,0.6785714,0.60714287,0.6785714,0.6785714,0.71428573,0.71428573,0.75,0.75,0.85714287,0.71428573,0.96428573,0.78571427,0.85714287,0.78571427,0.8214286,0.85714287,0.89285713,0.96428573
+TorchMLP (tuned),0.035714287,0.071428575,0.10714286,0.14285715,0.071428575,0.14285715,0.071428575,0.10714286,0.21428572,0.21428572,0.10714286,0.14285715,0.10714286,0.21428572,0.25,0.35714287,0.21428572,0.17857143,0.25,0.2857143,0.21428572,0.25,0.25,0.25,0.35714287,0.25,0.2857143,0.39285713,0.25,0.375,0.2857143,0.35714287,0.25,0.39285713,0.35714287,0.2857143,0.35714287,0.32142857,0.35714287,0.2857143,0.35714287,0.42857143,0.39285713,0.42857143,0.35714287,0.53571427,0.2857143,0.5714286,0.4642857,,0.51785713,0.5714286,0.5714286,0.5,0.6785714,0.5714286,0.5535714,0.64285713,0.64285713,0.64285713,0.64285713,0.64285713,0.6785714,0.75,0.71428573,0.89285713,0.71428573,0.75,0.89285713,0.64285713,0.78571427,0.85714287,0.78571427,0.78571427,0.8214286,0.89285713,0.9285714,0.96428573
+ExtraTrees (tuned + ensembled),0.071428575,0.035714287,0.14285715,0.071428575,0.10714286,0.10714286,0.17857143,0.035714287,0.21428572,0.14285715,0.21428572,0.21428572,0.25,0.21428572,0.21428572,0.25,0.32142857,0.2857143,0.2857143,0.25,0.2857143,0.21428572,0.32142857,0.25,0.32142857,0.25,0.35714287,0.32142857,0.35714287,0.25,0.32142857,0.32142857,0.32142857,0.21428572,0.25,0.39285713,0.21428572,0.35714287,0.42857143,0.32142857,0.39285713,0.4642857,0.2857143,0.39285713,0.32142857,0.4642857,0.39285713,0.4642857,0.42857143,0.48214287,,0.4642857,0.6964286,0.4642857,0.60714287,0.4642857,0.53571427,0.64285713,0.64285713,0.53571427,0.64285713,0.75,0.60714287,0.71428573,0.6785714,0.64285713,0.89285713,0.75,0.75,0.8214286,0.71428573,0.9285714,0.71428573,0.78571427,0.8214286,0.78571427,0.9285714,1.0
+ModernNCA (default),0.071428575,0.071428575,0.071428575,0.14285715,0.071428575,0.071428575,0.10714286,0.10714286,0.10714286,0.14285715,0.10714286,0.21428572,0.14285715,0.25,0.2857143,0.21428572,0.21428572,0.2857143,0.2857143,0.26785713,0.2857143,0.21428572,0.21428572,0.25,0.21428572,0.2857143,0.42857143,0.35714287,0.21428572,0.2857143,0.39285713,0.21428572,0.42857143,0.35714287,0.25,0.32142857,0.32142857,0.39285713,0.39285713,0.25,0.32142857,0.4642857,0.42857143,0.32142857,0.35714287,0.35714287,0.42857143,0.42857143,0.42857143,0.42857143,0.53571427,,0.5,0.4642857,0.71428573,0.53571427,0.5,0.64285713,0.71428573,0.64285713,0.64285713,0.71428573,0.64285713,0.71428573,0.6785714,0.78571427,0.75,0.64285713,0.71428573,0.78571427,0.75,0.78571427,0.78571427,0.85714287,0.85714287,0.85714287,0.89285713,0.96428573
+ExtraTrees (tuned),0.071428575,0.035714287,0.14285715,0.071428575,0.10714286,0.10714286,0.10714286,0.071428575,0.17857143,0.17857143,0.21428572,0.14285715,0.17857143,0.14285715,0.17857143,0.14285715,0.32142857,0.2857143,0.32142857,0.21428572,0.2857143,0.2857143,0.32142857,0.32142857,0.21428572,0.25,0.32142857,0.2857143,0.32142857,0.25,0.39285713,0.25,0.32142857,0.21428572,0.32142857,0.35714287,0.21428572,0.32142857,0.39285713,0.35714287,0.32142857,0.35714287,0.25,0.35714287,0.32142857,0.35714287,0.39285713,0.4642857,0.42857143,0.42857143,0.30357143,0.5,,0.4642857,0.64285713,0.4642857,0.4642857,0.4642857,0.64285713,0.53571427,0.64285713,0.64285713,0.60714287,0.75,0.60714287,0.60714287,0.78571427,0.64285713,0.71428573,0.78571427,0.6785714,0.85714287,0.6785714,0.6785714,0.6785714,0.78571427,0.96428573,0.96428573
+XGBoost (default),0.035714287,0.035714287,0.035714287,0.0,0.035714287,0.071428575,0.071428575,0.035714287,0.071428575,0.10714286,0.10714286,0.10714286,0.17857143,0.14285715,0.17857143,0.21428572,0.21428572,0.17857143,0.17857143,0.21428572,0.21428572,0.17857143,0.21428572,0.10714286,0.2857143,0.17857143,0.35714287,0.35714287,0.21428572,0.14285715,0.25,0.2857143,0.21428572,0.17857143,0.2857143,0.2857143,0.39285713,0.39285713,0.32142857,0.21428572,0.32142857,0.2857143,0.42857143,0.42857143,0.4642857,0.39285713,0.2857143,0.42857143,0.5,0.5,0.53571427,0.53571427,0.53571427,,0.5714286,0.5714286,0.60714287,0.53571427,0.5714286,0.6785714,0.5714286,0.5714286,0.60714287,0.6785714,0.6785714,0.6785714,0.6785714,0.75,0.6785714,0.75,0.75,0.71428573,0.78571427,0.78571427,0.8214286,0.8214286,0.8214286,0.9285714
+TabSTAR (tuned),0.10714286,0.17857143,0.14285715,0.21428572,0.21428572,0.17857143,0.17857143,0.21428572,0.25,0.16071428,0.2857143,0.2857143,0.25,0.25,0.25,0.21428572,0.21428572,0.2857143,0.2857143,0.2857143,0.25,0.32142857,0.2857143,0.25,0.25,0.25,0.2857143,0.25,0.25,0.21428572,0.25,0.32142857,0.2857143,0.21428572,0.32142857,0.35714287,0.32142857,0.39285713,0.25,0.35714287,0.35714287,0.2857143,0.35714287,0.32142857,0.32142857,0.39285713,0.39285713,0.32142857,0.39285713,0.32142857,0.39285713,0.2857143,0.35714287,0.42857143,,0.42857143,0.39285713,0.4642857,0.5535714,0.39285713,0.42857143,0.4642857,0.5714286,0.9285714,0.5714286,0.53571427,0.64285713,0.64285713,0.5714286,0.6785714,0.64285713,0.64285713,0.60714287,0.6785714,0.71428573,0.75,0.8214286,0.9285714
+FastaiMLP (tuned),0.10714286,0.071428575,0.0,0.10714286,0.14285715,0.10714286,0.10714286,0.14285715,0.17857143,0.14285715,0.10714286,0.17857143,0.25,0.21428572,0.14285715,0.25,0.21428572,0.25,0.21428572,0.21428572,0.2857143,0.21428572,0.21428572,0.21428572,0.25,0.21428572,0.2857143,0.25,0.2857143,0.2857143,0.25,0.2857143,0.25,0.25,0.25,0.21428572,0.25,0.2857143,0.35714287,0.25,0.35714287,0.32142857,0.39285713,0.32142857,0.32142857,0.25,0.32142857,0.39285713,0.2857143,0.42857143,0.53571427,0.4642857,0.53571427,0.42857143,0.5714286,,0.4642857,0.53571427,0.5714286,0.60714287,0.5714286,0.5714286,0.53571427,0.64285713,0.60714287,0.64285713,0.71428573,0.6785714,0.75,0.64285713,0.78571427,0.75,0.71428573,0.75,0.75,0.8214286,0.85714287,0.9285714
+LightGBM (default),0.071428575,0.071428575,0.035714287,0.035714287,0.071428575,0.035714287,0.071428575,0.071428575,0.10714286,0.14285715,0.10714286,0.071428575,0.17857143,0.17857143,0.14285715,0.25,0.21428572,0.17857143,0.21428572,0.17857143,0.21428572,0.17857143,0.21428572,0.071428575,0.35714287,0.21428572,0.32142857,0.2857143,0.25,0.035714287,0.2857143,0.2857143,0.21428572,0.10714286,0.35714287,0.35714287,0.32142857,0.35714287,0.2857143,0.21428572,0.39285713,0.25,0.35714287,0.42857143,0.4642857,0.35714287,0.2857143,0.42857143,0.5,0.44642857,0.4642857,0.5,0.53571427,0.39285713,0.60714287,0.53571427,,0.53571427,0.60714287,0.6785714,0.53571427,0.5714286,0.60714287,0.6785714,0.60714287,0.64285713,0.71428573,0.71428573,0.71428573,0.6785714,0.75,0.6785714,0.71428573,0.71428573,0.71428573,0.85714287,0.89285713,0.9285714
+RandomForest (tuned + ensembled),0.071428575,0.071428575,0.10714286,0.071428575,0.071428575,0.071428575,0.14285715,0.071428575,0.25,0.14285715,0.17857143,0.14285715,0.25,0.14285715,0.17857143,0.25,0.25,0.2857143,0.21428572,0.25,0.21428572,0.14285715,0.25,0.21428572,0.35714287,0.17857143,0.2857143,0.35714287,0.2857143,0.17857143,0.21428572,0.25,0.25,0.21428572,0.2857143,0.2857143,0.17857143,0.35714287,0.2857143,0.2857143,0.32142857,0.35714287,0.2857143,0.32142857,0.32142857,0.39285713,0.32142857,0.35714287,0.4642857,0.35714287,0.35714287,0.35714287,0.53571427,0.4642857,0.53571427,0.4642857,0.4642857,,0.5714286,0.42857143,0.53571427,0.75,0.5,0.6785714,0.64285713,0.71428573,0.78571427,0.75,0.60714287,0.85714287,0.64285713,0.75,0.64285713,0.71428573,0.6785714,0.75,0.89285713,0.96428573
+TabSTAR (tuned + ensembled),0.14285715,0.17857143,0.14285715,0.21428572,0.21428572,0.17857143,0.17857143,0.21428572,0.25,0.17857143,0.25,0.2857143,0.25,0.21428572,0.25,0.25,0.17857143,0.2857143,0.25,0.2857143,0.25,0.32142857,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.21428572,0.25,0.32142857,0.25,0.21428572,0.2857143,0.32142857,0.32142857,0.39285713,0.25,0.32142857,0.2857143,0.2857143,0.32142857,0.32142857,0.32142857,0.35714287,0.35714287,0.25,0.39285713,0.35714287,0.35714287,0.2857143,0.35714287,0.42857143,0.44642857,0.42857143,0.39285713,0.42857143,,0.42857143,0.4642857,0.4642857,0.5714286,0.96428573,0.5714286,0.5714286,0.60714287,0.60714287,0.53571427,0.64285713,0.60714287,0.60714287,0.60714287,0.6785714,0.6785714,0.75,0.8214286,0.9285714
+iLTM (default),0.10714286,0.035714287,0.035714287,0.10714286,0.071428575,0.071428575,0.17857143,0.10714286,0.071428575,0.10714286,0.10714286,0.14285715,0.17857143,0.071428575,0.17857143,0.17857143,0.17857143,0.17857143,0.17857143,0.25,0.14285715,0.17857143,0.25,0.14285715,0.17857143,0.17857143,0.25,0.25,0.2857143,0.17857143,0.21428572,0.2857143,0.25,0.21428572,0.2857143,0.35714287,0.30357143,0.32142857,0.32142857,0.2857143,0.2857143,0.2857143,0.25,0.2857143,0.2857143,0.32142857,0.2857143,0.32142857,0.32142857,0.35714287,0.4642857,0.35714287,0.4642857,0.32142857,0.60714287,0.39285713,0.32142857,0.5714286,0.5714286,,0.5,0.5,0.53571427,0.6785714,0.5,0.64285713,0.60714287,0.60714287,0.60714287,0.60714287,0.60714287,0.75,0.60714287,0.6785714,0.6785714,0.78571427,0.8214286,0.9285714
+OrionMSP (default),0.035714287,0.0,0.0,0.035714287,0.035714287,0.0,0.10714286,0.0,0.071428575,0.08928572,0.10714286,0.17857143,0.16071428,0.19642857,0.071428575,0.035714287,0.17857143,0.21428572,0.25,0.17857143,0.25,0.21428572,0.21428572,0.21428572,0.071428575,0.2857143,0.25,0.14285715,0.21428572,0.21428572,0.25,0.17857143,0.25,0.25,0.2857143,0.2857143,0.32142857,0.35714287,0.2857143,0.25,0.32142857,0.2857143,0.25,0.32142857,0.2857143,0.21428572,0.32142857,0.35714287,0.32142857,0.35714287,0.35714287,0.35714287,0.35714287,0.42857143,0.5714286,0.42857143,0.4642857,0.4642857,0.53571427,0.5,,0.4642857,0.5714286,0.71428573,0.5714286,0.5714286,0.6785714,0.5714286,0.71428573,0.6785714,0.5714286,0.64285713,0.6785714,0.75,0.75,0.78571427,0.8214286,0.96428573
+RandomForest (tuned),0.071428575,0.035714287,0.035714287,0.071428575,0.10714286,0.071428575,0.14285715,0.10714286,0.14285715,0.071428575,0.071428575,0.125,0.16071428,0.10714286,0.17857143,0.21428572,0.25,0.19642857,0.14285715,0.125,0.21428572,0.14285715,0.21428572,0.10714286,0.25,0.10714286,0.23214285,0.32142857,0.25,0.10714286,0.17857143,0.21428572,0.14285715,0.14285715,0.2857143,0.25,0.21428572,0.32142857,0.21428572,0.32142857,0.32142857,0.21428572,0.25,0.32142857,0.25,0.35714287,0.25,0.2857143,0.39285713,0.35714287,0.25,0.2857143,0.35714287,0.42857143,0.53571427,0.42857143,0.42857143,0.25,0.53571427,0.5,0.53571427,,0.5,0.64285713,0.5714286,0.5714286,0.71428573,0.75,0.6785714,0.76785713,0.71428573,0.75,0.60714287,0.71428573,0.6785714,0.71428573,0.85714287,0.96428573
+PerpetualBooster (tuned + ensembled),0.0,0.035714287,0.071428575,0.071428575,0.035714287,0.10714286,0.071428575,0.071428575,0.071428575,0.14285715,0.071428575,0.17857143,0.17857143,0.14285715,0.25,0.2857143,0.21428572,0.21428572,0.21428572,0.17857143,0.21428572,0.14285715,0.14285715,0.17857143,0.2857143,0.14285715,0.21428572,0.2857143,0.21428572,0.2857143,0.17857143,0.2857143,0.21428572,0.2857143,0.2857143,0.21428572,0.32142857,0.25,0.25,0.25,0.32142857,0.2857143,0.39285713,0.2857143,0.21428572,0.32142857,0.32142857,0.35714287,0.32142857,0.32142857,0.39285713,0.35714287,0.39285713,0.39285713,0.42857143,0.4642857,0.39285713,0.5,0.42857143,0.4642857,0.42857143,0.5,,0.42857143,0.75,0.5,0.53571427,0.5714286,0.60714287,0.64285713,0.60714287,0.60714287,0.8214286,0.71428573,0.71428573,0.75,0.71428573,0.89285713
+TabSTAR (default),0.071428575,0.17857143,0.14285715,0.17857143,0.21428572,0.17857143,0.14285715,0.17857143,0.21428572,0.14285715,0.17857143,0.25,0.21428572,0.21428572,0.21428572,0.21428572,0.17857143,0.25,0.25,0.25,0.17857143,0.21428572,0.21428572,0.21428572,0.25,0.25,0.21428572,0.17857143,0.14285715,0.21428572,0.21428572,0.25,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.21428572,0.2857143,0.21428572,0.25,0.2857143,0.32142857,0.21428572,0.32142857,0.25,0.2857143,0.2857143,0.25,0.32142857,0.071428575,0.35714287,0.32142857,0.32142857,0.035714287,0.32142857,0.2857143,0.35714287,0.5714286,,0.5714286,0.42857143,0.39285713,0.42857143,0.4642857,0.5,0.32142857,0.5714286,0.60714287,0.5,0.5,0.60714287,0.6785714,0.9285714
+PerpetualBooster (tuned),0.0,0.0,0.0,0.035714287,0.0,0.071428575,0.035714287,0.035714287,0.035714287,0.14285715,0.035714287,0.10714286,0.17857143,0.14285715,0.17857143,0.17857143,0.10714286,0.17857143,0.10714286,0.10714286,0.17857143,0.071428575,0.10714286,0.14285715,0.17857143,0.14285715,0.21428572,0.21428572,0.17857143,0.21428572,0.10714286,0.21428572,0.17857143,0.21428572,0.25,0.21428572,0.17857143,0.21428572,0.17857143,0.25,0.2857143,0.21428572,0.2857143,0.2857143,0.17857143,0.25,0.21428572,0.2857143,0.2857143,0.2857143,0.32142857,0.32142857,0.39285713,0.32142857,0.42857143,0.39285713,0.39285713,0.35714287,0.42857143,0.5,0.42857143,0.42857143,0.25,0.42857143,,0.5,0.5,0.5714286,0.64285713,0.5714286,0.5714286,0.53571427,0.73214287,0.6785714,0.64285713,0.71428573,0.6785714,0.89285713
+TorchMLP (default),0.035714287,0.071428575,0.035714287,0.071428575,0.035714287,0.071428575,0.035714287,0.071428575,0.10714286,0.10714286,0.035714287,0.10714286,0.14285715,0.071428575,0.14285715,0.17857143,0.035714287,0.17857143,0.10714286,0.10714286,0.10714286,0.071428575,0.035714287,0.14285715,0.17857143,0.10714286,0.17857143,0.17857143,0.17857143,0.14285715,0.071428575,0.14285715,0.17857143,0.17857143,0.14285715,0.071428575,0.10714286,0.21428572,0.17857143,0.21428572,0.10714286,0.17857143,0.14285715,0.17857143,0.14285715,0.2857143,0.071428575,0.25,0.2857143,0.10714286,0.35714287,0.21428572,0.39285713,0.32142857,0.4642857,0.35714287,0.35714287,0.2857143,0.42857143,0.35714287,0.42857143,0.42857143,0.5,0.5714286,0.5,,0.60714287,0.5,0.60714287,0.53571427,0.60714287,0.60714287,0.60714287,0.60714287,0.6785714,0.75,0.78571427,0.9285714
+ExtraTrees (default),0.035714287,0.035714287,0.0,0.035714287,0.071428575,0.035714287,0.10714286,0.035714287,0.035714287,0.035714287,0.035714287,0.14285715,0.21428572,0.071428575,0.071428575,0.035714287,0.21428572,0.14285715,0.17857143,0.14285715,0.25,0.071428575,0.25,0.14285715,0.10714286,0.17857143,0.2857143,0.17857143,0.17857143,0.10714286,0.21428572,0.10714286,0.21428572,0.14285715,0.17857143,0.25,0.14285715,0.21428572,0.25,0.21428572,0.21428572,0.21428572,0.10714286,0.17857143,0.17857143,0.14285715,0.25,0.2857143,0.25,0.2857143,0.10714286,0.25,0.21428572,0.32142857,0.35714287,0.2857143,0.2857143,0.21428572,0.39285713,0.39285713,0.32142857,0.2857143,0.4642857,0.60714287,0.5,0.39285713,,0.4642857,0.5,0.5714286,0.53571427,0.5,0.53571427,0.5714286,0.5714286,0.64285713,0.75,0.9285714
+xRFM (default),0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.071428575,0.071428575,0.10714286,0.14285715,0.071428575,0.071428575,0.10714286,0.035714287,0.14285715,0.17857143,0.17857143,0.17857143,0.17857143,0.10714286,0.17857143,0.17857143,0.10714286,0.14285715,0.17857143,0.17857143,0.14285715,0.21428572,0.14285715,0.10714286,0.17857143,0.21428572,0.17857143,0.25,0.14285715,0.21428572,0.21428572,0.14285715,0.21428572,0.21428572,0.035714287,0.14285715,0.21428572,0.21428572,0.21428572,0.2857143,0.25,0.25,0.25,0.35714287,0.35714287,0.25,0.35714287,0.32142857,0.2857143,0.25,0.39285713,0.39285713,0.42857143,0.25,0.42857143,0.5714286,0.42857143,0.5,0.53571427,,0.5,0.53571427,0.4642857,0.42857143,0.5,0.5714286,0.5714286,0.64285713,0.71428573,0.89285713
+TabFlex (default),0.0,0.071428575,0.0,0.14285715,0.10714286,0.071428575,0.035714287,0.14285715,0.14285715,0.10714286,0.071428575,0.17857143,0.14285715,0.17857143,0.035714287,0.071428575,0.10714286,0.14285715,0.17857143,0.14285715,0.14285715,0.10714286,0.10714286,0.17857143,0.14285715,0.21428572,0.14285715,0.035714287,0.17857143,0.21428572,0.17857143,0.14285715,0.21428572,0.21428572,0.10714286,0.17857143,0.17857143,0.17857143,0.17857143,0.25,0.071428575,0.14285715,0.21428572,0.14285715,0.14285715,0.17857143,0.10714286,0.21428572,0.14285715,0.10714286,0.25,0.2857143,0.2857143,0.32142857,0.42857143,0.25,0.2857143,0.39285713,0.4642857,0.39285713,0.2857143,0.32142857,0.39285713,0.53571427,0.35714287,0.39285713,0.5,0.5,,0.53571427,0.53571427,0.53571427,0.53571427,0.64285713,0.64285713,0.71428573,0.75,0.89285713
+RandomForest (default),0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.071428575,0.035714287,0.035714287,0.035714287,0.035714287,0.08928572,0.10714286,0.035714287,0.10714286,0.071428575,0.21428572,0.16071428,0.10714286,0.10714286,0.17857143,0.071428575,0.17857143,0.035714287,0.14285715,0.071428575,0.21428572,0.21428572,0.10714286,0.071428575,0.10714286,0.10714286,0.10714286,0.10714286,0.21428572,0.21428572,0.14285715,0.21428572,0.21428572,0.17857143,0.21428572,0.2857143,0.17857143,0.21428572,0.2857143,0.25,0.2857143,0.35714287,0.2857143,0.35714287,0.17857143,0.21428572,0.21428572,0.25,0.32142857,0.35714287,0.32142857,0.14285715,0.35714287,0.39285713,0.32142857,0.23214285,0.35714287,0.5,0.42857143,0.4642857,0.42857143,0.4642857,0.4642857,,0.5,0.60714287,0.5,0.64285713,0.64285713,0.6785714,0.78571427,0.96428573
+FastaiMLP (default),0.0,0.0,0.0,0.035714287,0.035714287,0.035714287,0.0,0.071428575,0.035714287,0.035714287,0.035714287,0.071428575,0.10714286,0.071428575,0.035714287,0.035714287,0.10714286,0.17857143,0.10714286,0.10714286,0.10714286,0.071428575,0.17857143,0.10714286,0.035714287,0.14285715,0.14285715,0.071428575,0.17857143,0.14285715,0.10714286,0.10714286,0.14285715,0.14285715,0.14285715,0.21428572,0.14285715,0.17857143,0.17857143,0.21428572,0.10714286,0.21428572,0.17857143,0.17857143,0.14285715,0.17857143,0.14285715,0.25,0.035714287,0.21428572,0.2857143,0.25,0.32142857,0.25,0.35714287,0.21428572,0.25,0.35714287,0.39285713,0.39285713,0.42857143,0.2857143,0.39285713,0.6785714,0.42857143,0.39285713,0.4642857,0.53571427,0.4642857,0.5,,0.5714286,0.64285713,0.5714286,0.60714287,0.64285713,0.78571427,0.9285714
+KNN (tuned + ensembled),0.071428575,0.035714287,0.071428575,0.071428575,0.035714287,0.035714287,0.071428575,0.071428575,0.071428575,0.071428575,0.10714286,0.10714286,0.10714286,0.10714286,0.035714287,0.10714286,0.14285715,0.10714286,0.14285715,0.10714286,0.10714286,0.071428575,0.14285715,0.071428575,0.14285715,0.17857143,0.21428572,0.10714286,0.10714286,0.17857143,0.10714286,0.10714286,0.14285715,0.14285715,0.10714286,0.21428572,0.10714286,0.2857143,0.21428572,0.19642857,0.10714286,0.25,0.10714286,0.10714286,0.10714286,0.14285715,0.17857143,0.25,0.21428572,0.14285715,0.071428575,0.21428572,0.14285715,0.2857143,0.35714287,0.25,0.32142857,0.25,0.39285713,0.25,0.35714287,0.25,0.39285713,0.42857143,0.4642857,0.39285713,0.5,0.5714286,0.4642857,0.39285713,0.42857143,,0.5,0.5714286,0.5714286,0.64285713,0.96428573,0.96428573
+PerpetualBooster (default),0.0,0.035714287,0.071428575,0.071428575,0.035714287,0.071428575,0.035714287,0.071428575,0.035714287,0.10714286,0.035714287,0.071428575,0.071428575,0.035714287,0.071428575,0.14285715,0.071428575,0.10714286,0.071428575,0.071428575,0.14285715,0.071428575,0.071428575,0.14285715,0.10714286,0.071428575,0.10714286,0.14285715,0.17857143,0.17857143,0.071428575,0.14285715,0.14285715,0.17857143,0.14285715,0.14285715,0.14285715,0.17857143,0.21428572,0.14285715,0.10714286,0.21428572,0.10714286,0.17857143,0.10714286,0.25,0.14285715,0.25,0.14285715,0.21428572,0.2857143,0.21428572,0.32142857,0.21428572,0.39285713,0.2857143,0.2857143,0.35714287,0.39285713,0.39285713,0.32142857,0.39285713,0.17857143,0.39285713,0.26785713,0.39285713,0.4642857,0.5,0.4642857,0.5,0.35714287,0.5,,0.5714286,0.5714286,0.60714287,0.60714287,0.85714287
+Linear (tuned + ensembled),0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.10714286,0.10714286,0.071428575,0.035714287,0.035714287,0.071428575,0.035714287,0.10714286,0.10714286,0.035714287,0.14285715,0.035714287,0.071428575,0.10714286,0.10714286,0.10714286,0.071428575,0.071428575,0.10714286,0.14285715,0.035714287,0.071428575,0.17857143,0.10714286,0.10714286,0.21428572,0.10714286,0.17857143,0.14285715,0.10714286,0.071428575,0.2857143,0.14285715,0.17857143,0.21428572,0.21428572,0.21428572,0.14285715,0.32142857,0.21428572,0.32142857,0.25,0.2857143,0.2857143,0.32142857,0.32142857,0.25,0.2857143,0.2857143,0.5,0.32142857,0.39285713,0.42857143,0.42857143,0.35714287,0.35714287,0.42857143,0.42857143,0.42857143,,0.71428573,0.71428573,0.64285713,0.85714287
+Linear (tuned),0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.035714287,0.035714287,0.035714287,0.035714287,0.05357143,0.10714286,0.10714286,0.071428575,0.035714287,0.035714287,0.10714286,0.035714287,0.071428575,0.10714286,0.035714287,0.14285715,0.035714287,0.071428575,0.10714286,0.071428575,0.071428575,0.035714287,0.035714287,0.10714286,0.10714286,0.035714287,0.071428575,0.14285715,0.14285715,0.071428575,0.17857143,0.10714286,0.17857143,0.17857143,0.071428575,0.071428575,0.25,0.10714286,0.17857143,0.17857143,0.17857143,0.17857143,0.14285715,0.32142857,0.17857143,0.2857143,0.25,0.2857143,0.32142857,0.32142857,0.32142857,0.25,0.32142857,0.2857143,0.5,0.35714287,0.32142857,0.42857143,0.42857143,0.35714287,0.35714287,0.39285713,0.42857143,0.42857143,0.2857143,,0.60714287,0.60714287,0.8214286
+Linear (default),0.0,0.0,0.0,0.071428575,0.035714287,0.035714287,0.0,0.071428575,0.035714287,0.035714287,0.0,0.071428575,0.035714287,0.071428575,0.10714286,0.14285715,0.071428575,0.071428575,0.035714287,0.035714287,0.035714287,0.071428575,0.071428575,0.071428575,0.14285715,0.035714287,0.035714287,0.10714286,0.10714286,0.071428575,0.035714287,0.071428575,0.10714286,0.071428575,0.035714287,0.10714286,0.14285715,0.17857143,0.035714287,0.21428572,0.035714287,0.071428575,0.14285715,0.10714286,0.035714287,0.17857143,0.035714287,0.035714287,0.14285715,0.10714286,0.21428572,0.14285715,0.21428572,0.17857143,0.25,0.17857143,0.14285715,0.25,0.25,0.21428572,0.21428572,0.2857143,0.25,0.39285713,0.2857143,0.25,0.35714287,0.35714287,0.2857143,0.32142857,0.35714287,0.35714287,0.39285713,0.2857143,0.39285713,,0.60714287,0.75
+KNN (tuned),0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.0,0.035714287,0.071428575,0.035714287,0.071428575,0.0,0.071428575,0.035714287,0.035714287,0.035714287,0.071428575,0.0,0.035714287,0.035714287,0.035714287,0.035714287,0.071428575,0.035714287,0.0,0.071428575,0.071428575,0.071428575,0.035714287,0.10714286,0.10714286,0.035714287,0.035714287,0.071428575,0.10714286,0.071428575,0.035714287,0.035714287,0.14285715,0.071428575,0.17857143,0.071428575,0.10714286,0.035714287,0.071428575,0.071428575,0.071428575,0.071428575,0.071428575,0.10714286,0.071428575,0.071428575,0.10714286,0.035714287,0.17857143,0.17857143,0.14285715,0.10714286,0.10714286,0.17857143,0.17857143,0.17857143,0.14285715,0.2857143,0.32142857,0.32142857,0.21428572,0.25,0.2857143,0.25,0.21428572,0.21428572,0.035714287,0.39285713,0.35714287,0.39285713,0.39285713,,0.9285714
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.035714287,0.0,0.035714287,0.0,0.0,0.0,0.0,0.0,0.0,0.035714287,0.0,0.0,0.0,0.071428575,0.0,0.0,0.0,0.035714287,0.0,0.0,0.0,0.0,0.0,0.035714287,0.035714287,0.0,0.035714287,0.035714287,0.0,0.035714287,0.035714287,0.071428575,0.071428575,0.071428575,0.071428575,0.035714287,0.071428575,0.071428575,0.035714287,0.035714287,0.10714286,0.071428575,0.10714286,0.071428575,0.071428575,0.10714286,0.10714286,0.035714287,0.071428575,0.035714287,0.14285715,0.14285715,0.17857143,0.25,0.071428575,
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
index e3603b3127af8b1a76ed3f583a476560cc62357f..948453f1c4eb5f5c34020c2b92aebeb496949d71 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Multiclass | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..664709e513d1f8a413d61a5f53b592b35d189589
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_multiclass/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1736.0,"score":0.678,"rank":8.0,"hrank":4.31,"imp":11.605,"train_time":253.44,"predict_time":3.242,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+227\/-75","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1736.0,"score":0.696,"rank":8.0,"hrank":3.09,"imp":5.495,"train_time":2793.87,"predict_time":8.928,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+412\/-158","imputed":false},{"position":2,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1701.0,"score":0.793,"rank":9.25,"hrank":1.46,"imp":7.038,"train_time":49.97,"predict_time":7.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+740\/-260","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1646.0,"score":0.602,"rank":11.56,"hrank":8.66,"imp":13.469,"train_time":2793.87,"predict_time":1.512,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+193\/-120","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1636.0,"score":0.584,"rank":12.0,"hrank":8.61,"imp":13.15,"train_time":9.9,"predict_time":0.591,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+230\/-126","imputed":false},{"position":5,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1588.0,"score":0.545,"rank":14.38,"hrank":5.85,"imp":11.341,"train_time":8.93,"predict_time":0.49,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+331\/-161","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1537.0,"score":0.453,"rank":17.31,"hrank":9.72,"imp":13.872,"train_time":9.88,"predict_time":0.825,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+240\/-106","imputed":false},{"position":7,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1520.0,"score":0.49,"rank":18.38,"hrank":6.57,"imp":15.305,"train_time":4.73,"predict_time":0.225,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+355\/-191","imputed":false},{"position":8,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1497.0,"score":0.439,"rank":19.88,"hrank":5.52,"imp":13.383,"train_time":41.12,"predict_time":8.609,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+274\/-156","imputed":false},{"position":9,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1486.0,"score":0.436,"rank":20.62,"hrank":9.66,"imp":17.683,"train_time":5547.02,"predict_time":3.267,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+164\/-104","imputed":false},{"position":10,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1458.0,"score":0.344,"rank":22.5,"hrank":15.5,"imp":15.961,"train_time":2807.98,"predict_time":0.179,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+207\/-95","imputed":false},{"position":11,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1458.0,"score":0.339,"rank":22.5,"hrank":17.96,"imp":16.746,"train_time":2807.98,"predict_time":1.223,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+184\/-98","imputed":false},{"position":12,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1440.0,"score":0.365,"rank":23.75,"hrank":9.99,"imp":18.813,"train_time":4335.39,"predict_time":21.143,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+176\/-103","imputed":false},{"position":13,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1432.0,"score":0.294,"rank":24.38,"hrank":17.97,"imp":19.687,"train_time":1280.18,"predict_time":4.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+129\/-89","imputed":false},{"position":14,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1423.0,"score":0.284,"rank":25.0,"hrank":17.94,"imp":19.765,"train_time":1280.18,"predict_time":1.051,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+128\/-100","imputed":false},{"position":15,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1416.0,"score":0.378,"rank":25.5,"hrank":12.44,"imp":21.557,"train_time":2584.96,"predict_time":12.368,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+283\/-194","imputed":true},{"position":16,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1411.0,"score":0.35,"rank":25.88,"hrank":16.07,"imp":20.16,"train_time":3796.93,"predict_time":4.545,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-120","imputed":false},{"position":17,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1404.0,"score":0.317,"rank":26.38,"hrank":19.46,"imp":19.268,"train_time":4335.39,"predict_time":3.527,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+182\/-107","imputed":false},{"position":18,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1403.0,"score":0.333,"rank":26.5,"hrank":11.96,"imp":20.211,"train_time":3796.93,"predict_time":0.767,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+198\/-123","imputed":false},{"position":19,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1395.0,"score":0.329,"rank":27.06,"hrank":19.75,"imp":22.293,"train_time":159.48,"predict_time":3.544,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+202\/-178","imputed":true},{"position":20,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1383.0,"score":0.345,"rank":28.0,"hrank":19.46,"imp":22.257,"train_time":9413.13,"predict_time":343.731,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+207\/-150","imputed":false},{"position":21,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1380.0,"score":0.291,"rank":28.25,"hrank":16.11,"imp":22.455,"train_time":5.38,"predict_time":0.274,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+236\/-140","imputed":false},{"position":22,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1372.0,"score":0.209,"rank":28.88,"hrank":27.18,"imp":20.271,"train_time":2029.77,"predict_time":4.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-63","imputed":false},{"position":23,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1369.0,"score":0.215,"rank":29.12,"hrank":27.02,"imp":20.058,"train_time":2029.77,"predict_time":1.04,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-75","imputed":false},{"position":24,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1357.0,"score":0.299,"rank":30.12,"hrank":10.59,"imp":22.359,"train_time":2079.22,"predict_time":3.347,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+210\/-187","imputed":false},{"position":25,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1344.0,"score":0.25,"rank":31.19,"hrank":17.06,"imp":21.491,"train_time":5.48,"predict_time":0.354,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+198\/-145","imputed":true},{"position":26,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1340.0,"score":0.204,"rank":31.5,"hrank":27.69,"imp":20.673,"train_time":27893.23,"predict_time":752.506,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+129\/-90","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1334.0,"score":0.204,"rank":32.06,"hrank":28.34,"imp":21.281,"train_time":6686.5,"predict_time":9.587,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-92","imputed":false},{"position":28,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1328.0,"score":0.25,"rank":32.5,"hrank":21.13,"imp":22.917,"train_time":2584.96,"predict_time":0.407,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+198\/-148","imputed":true},{"position":29,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1322.0,"score":0.181,"rank":33.06,"hrank":27.01,"imp":21.623,"train_time":6686.5,"predict_time":0.752,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-113","imputed":false},{"position":30,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1319.0,"score":0.198,"rank":33.25,"hrank":21.43,"imp":23.299,"train_time":17528.83,"predict_time":0.282,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+140\/-158","imputed":false},{"position":31,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1312.0,"score":0.278,"rank":33.88,"hrank":14.3,"imp":23.544,"train_time":9413.13,"predict_time":39.488,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+222\/-156","imputed":false},{"position":32,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1306.0,"score":0.321,"rank":34.38,"hrank":9.3,"imp":23.556,"train_time":21.85,"predict_time":1.702,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+220\/-260","imputed":false},{"position":33,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1302.0,"score":0.244,"rank":34.69,"hrank":15.95,"imp":23.644,"train_time":968.86,"predict_time":1.864,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+198\/-166","imputed":false},{"position":34,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1299.0,"score":0.239,"rank":34.88,"hrank":17.34,"imp":25.117,"train_time":14.97,"predict_time":0.624,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+193\/-159","imputed":false},{"position":35,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1285.0,"score":0.263,"rank":36.0,"hrank":22.07,"imp":23.437,"train_time":729.27,"predict_time":1.833,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+207\/-217","imputed":false},{"position":36,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1284.0,"score":0.172,"rank":36.12,"hrank":32.65,"imp":22.329,"train_time":26.32,"predict_time":0.254,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-93","imputed":false},{"position":37,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1283.0,"score":0.21,"rank":36.19,"hrank":17.63,"imp":26.249,"train_time":11.51,"predict_time":1.948,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+213\/-170","imputed":true},{"position":38,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1281.0,"score":0.234,"rank":36.38,"hrank":6.58,"imp":23.832,"train_time":2079.22,"predict_time":0.319,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+236\/-198","imputed":false},{"position":39,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1279.0,"score":0.192,"rank":36.5,"hrank":26.19,"imp":25.023,"train_time":283.65,"predict_time":4.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+148\/-155","imputed":false},{"position":40,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1267.0,"score":0.166,"rank":37.5,"hrank":32.62,"imp":22.336,"train_time":27893.23,"predict_time":206.719,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+160\/-115","imputed":false},{"position":41,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1252.0,"score":0.182,"rank":38.62,"hrank":30.81,"imp":24.717,"train_time":17528.83,"predict_time":0.025,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-171","imputed":false},{"position":42,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1246.0,"score":0.127,"rank":39.12,"hrank":35.17,"imp":24.767,"train_time":3647.6,"predict_time":2.161,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-150","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1228.0,"score":0.164,"rank":40.56,"hrank":17.67,"imp":26.524,"train_time":968.86,"predict_time":0.284,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+197\/-170","imputed":false},{"position":44,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1228.0,"score":0.139,"rank":40.62,"hrank":36.13,"imp":25.261,"train_time":12.19,"predict_time":0.042,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-142","imputed":false},{"position":45,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1225.0,"score":0.164,"rank":40.88,"hrank":33.61,"imp":24.508,"train_time":1560.18,"predict_time":0.201,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+131\/-183","imputed":false},{"position":46,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.143,"rank":41.0,"hrank":35.65,"imp":25.007,"train_time":1560.18,"predict_time":1.226,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-178","imputed":false},{"position":47,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1214.0,"score":0.199,"rank":41.75,"hrank":29.6,"imp":24.901,"train_time":729.27,"predict_time":0.325,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+199\/-188","imputed":false},{"position":48,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1209.0,"score":0.122,"rank":42.12,"hrank":22.57,"imp":26.495,"train_time":3647.6,"predict_time":0.194,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+164\/-188","imputed":false},{"position":49,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1206.0,"score":0.055,"rank":42.38,"hrank":39.12,"imp":23.692,"train_time":4.93,"predict_time":0.592,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-90","imputed":false},{"position":50,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1202.0,"score":0.085,"rank":42.75,"hrank":38.75,"imp":27.282,"train_time":3.02,"predict_time":0.076,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+124\/-127","imputed":false},{"position":51,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1197.0,"score":0.138,"rank":43.12,"hrank":27.91,"imp":28.687,"train_time":2721.87,"predict_time":12.59,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+174\/-198","imputed":false},{"position":52,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1196.0,"score":0.13,"rank":43.25,"hrank":31.81,"imp":26.12,"train_time":5.12,"predict_time":0.444,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+162\/-157","imputed":false},{"position":53,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1194.0,"score":0.17,"rank":43.38,"hrank":18.39,"imp":26.7,"train_time":46.36,"predict_time":39.803,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+250\/-184","imputed":false},{"position":54,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1142.0,"score":0.114,"rank":47.5,"hrank":33.06,"imp":28.908,"train_time":7.42,"predict_time":0.203,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+178\/-241","imputed":false},{"position":55,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1121.0,"score":0.071,"rank":49.12,"hrank":43.62,"imp":28.12,"train_time":19.13,"predict_time":6.92,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-137","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1103.0,"score":0.063,"rank":50.38,"hrank":40.6,"imp":30.234,"train_time":2721.87,"predict_time":1.082,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+140\/-174","imputed":false},{"position":57,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1103.0,"score":0.067,"rank":50.38,"hrank":43.25,"imp":31.537,"train_time":8.99,"predict_time":1.372,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+162\/-248","imputed":false},{"position":58,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1102.0,"score":0.025,"rank":50.5,"hrank":47.25,"imp":30.065,"train_time":17.24,"predict_time":0.565,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+101\/-143","imputed":false},{"position":59,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1015.0,"score":0.135,"rank":56.25,"hrank":33.72,"imp":40.36,"train_time":155.59,"predict_time":2.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+270\/-364","imputed":false},{"position":60,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1011.0,"score":0.025,"rank":56.5,"hrank":53.45,"imp":30.304,"train_time":495.65,"predict_time":205.513,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+107\/-193","imputed":false},{"position":61,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1011.0,"score":0.026,"rank":56.5,"hrank":54.01,"imp":33.029,"train_time":14.37,"predict_time":0.362,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+131\/-168","imputed":false},{"position":62,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":57.19,"hrank":54.61,"imp":37.619,"train_time":0.74,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-187","imputed":false},{"position":63,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":994.0,"score":0.023,"rank":57.56,"hrank":52.59,"imp":33.011,"train_time":44782.6,"predict_time":7.958,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-251","imputed":false},{"position":64,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":980.0,"score":0.005,"rank":58.44,"hrank":55.16,"imp":33.318,"train_time":44782.6,"predict_time":36.102,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-224","imputed":false},{"position":65,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":966.0,"score":0.0,"rank":59.25,"hrank":57.78,"imp":35.273,"train_time":20.95,"predict_time":3.326,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-171","imputed":false},{"position":66,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":932.0,"score":0.005,"rank":61.12,"hrank":58.6,"imp":38.629,"train_time":5.16,"predict_time":0.787,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-187","imputed":false},{"position":67,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":918.0,"score":0.017,"rank":61.88,"hrank":58.07,"imp":37.184,"train_time":8.38,"predict_time":0.659,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+144\/-238","imputed":false},{"position":68,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":910.0,"score":0.0,"rank":62.31,"hrank":60.42,"imp":43.471,"train_time":909.8,"predict_time":0.609,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-239","imputed":false},{"position":69,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":888.0,"score":0.0,"rank":63.44,"hrank":61.44,"imp":44.403,"train_time":909.8,"predict_time":0.107,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-233","imputed":false},{"position":70,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":808.0,"score":0.0,"rank":67.0,"hrank":66.32,"imp":53.303,"train_time":1308.45,"predict_time":9.593,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-184","imputed":false},{"position":71,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":798.0,"score":0.0,"rank":67.38,"hrank":66.58,"imp":44.125,"train_time":666.72,"predict_time":7.823,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+100\/-242","imputed":false},{"position":72,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":792.0,"score":0.085,"rank":67.62,"hrank":52.48,"imp":51.863,"train_time":155.59,"predict_time":0.191,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+300\/-499","imputed":false},{"position":73,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":776.0,"score":0.0,"rank":68.25,"hrank":67.7,"imp":43.127,"train_time":4.67,"predict_time":0.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-239","imputed":false},{"position":74,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":731.0,"score":0.0,"rank":69.81,"hrank":69.38,"imp":56.997,"train_time":1308.45,"predict_time":1.071,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-183","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":709.0,"score":0.0,"rank":70.5,"hrank":69.94,"imp":50.164,"train_time":4.62,"predict_time":0.135,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-271","imputed":false},{"position":76,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":664.0,"score":0.0,"rank":71.81,"hrank":71.17,"imp":62.362,"train_time":218.36,"predict_time":0.46,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+109\/-330","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":205.0,"score":0.0,"rank":77.5,"hrank":77.49,"imp":71.869,"train_time":0.52,"predict_time":0.075,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-642","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/pareto_front_explorer.html
index 33ab5dc156f2aef4abc210258cc27d2afecdaacc..eabf83b6d05bc03d2b801efa44ab7ec888397d42 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.1027625141,"imp":25.0233934223,"elo":1279.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2536073213,"imp":22.3290802602,"elo":1283.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1791910888,"imp":15.9607128727,"elo":1458.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.2234800966,"imp":16.7460632673,"elo":1458.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2033347873,"imp":28.9079076305,"elo":1142.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2010198493,"imp":24.5082802321,"elo":1224.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2260580575,"imp":25.0069064835,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0424445763,"imp":25.2606650045,"elo":1227.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0252302724,"imp":24.7166459855,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.2817546014,"imp":23.2989953026,"elo":1319.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7866568202,"imp":38.6291487203,"elo":932.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2836586346,"imp":26.5237510309,"elo":1228.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.8636375274,"imp":23.6440186663,"elo":1301.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6590449294,"imp":37.1837118215,"elo":918.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.0818469447,"imp":30.2335758512,"elo":1103.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5900916415,"imp":28.6871013436,"elo":1197.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.4437307263,"imp":26.1196336464,"elo":1195.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.0508933307,"imp":19.7650301564,"elo":1423.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.0776987398,"imp":19.6873087712,"elo":1431.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.6091982733,"imp":13.3825747475,"elo":1497.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.5435441032,"imp":22.2933505815,"elo":1395.4,"imputed":true,"imputed_pct":25.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5648398113,"imp":30.064936048,"elo":1101.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.751659725,"imp":21.6228300791,"elo":1321.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5865347302,"imp":21.2813099415,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3262335461,"imp":35.2726368712,"elo":965.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4602862659,"imp":62.362333418,"elo":663.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0713018657,"imp":56.9967733472,"elo":731.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":9.5934077851,"imp":53.3028895617,"elo":807.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.147930342,"imp":37.6187219692,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3251277774,"imp":24.9008491087,"elo":1213.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.8330056127,"imp":23.4369156098,"elo":1285.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":6.920194808,"imp":28.1199880238,"elo":1120.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.5270444068,"imp":19.2681742676,"elo":1404.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.142857233,"imp":18.8131706763,"elo":1440.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8250564093,"imp":13.8715452955,"elo":1537.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.5121585317,"imp":13.4692880907,"elo":1645.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9280086995,"imp":5.4947133355,"elo":1735.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7021367127,"imp":23.5561919407,"elo":1305.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8027161329,"imp":26.6996058482,"elo":1194.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4876798649,"imp":23.5440960871,"elo":1311.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":343.7310101457,"imp":22.2571294857,"elo":1383.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2735262831,"imp":22.4554958659,"elo":1380.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.6889691672,"imp":7.0380391136,"elo":1701.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.8545350274,"imp":43.1266606509,"elo":775.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.9479990042,"imp":26.2491237713,"elo":1283.0,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2245693736,"imp":15.304534586,"elo":1520.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.624076982,"imp":25.1172532623,"elo":1299.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7671600452,"imp":20.2113064398,"elo":1402.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.5447590803,"imp":20.1604816148,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5909006176,"imp":13.1503663168,"elo":1636.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4902017202,"imp":11.3414172871,"elo":1588.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3535753061,"imp":21.4913922326,"elo":1344.1,"imputed":true,"imputed_pct":25.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4074152559,"imp":22.9167321116,"elo":1328.2,"imputed":true,"imputed_pct":25.0,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.3680288933,"imp":21.5573536372,"elo":1416.3,"imputed":true,"imputed_pct":25.0,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8225789867,"imp":44.12502013,"elo":798.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.9584829717,"imp":33.0105868135,"elo":993.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.1023197407,"imp":33.3175339836,"elo":979.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0760162666,"imp":27.2824558285,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3624794907,"imp":33.0285558796,"elo":1011.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.194090358,"imp":26.495255967,"elo":1209.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1614416305,"imp":24.7665449214,"elo":1246.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.5916710541,"imp":23.6920112547,"elo":1206.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0396563215,"imp":20.0582538038,"elo":1369.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":4.1126713529,"imp":20.270846639,"elo":1372.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":205.5125020069,"imp":30.304216606,"elo":1011.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":206.7186712774,"imp":22.3359735156,"elo":1266.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":752.505846815,"imp":20.672850816,"elo":1340.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.371587587,"imp":31.5368880354,"elo":1103.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.3186223488,"imp":23.8320825397,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.3466778136,"imp":22.3592603663,"elo":1356.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.1027625141,"x_train":283.646286296,"imp":25.0233934223,"elo":1279.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2536073213,"x_train":26.3194957247,"imp":22.3290802602,"elo":1283.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.1791910888,"x_train":2807.9829916593,"imp":15.9607128727,"elo":1458.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.2234800966,"x_train":2807.9829916593,"imp":16.7460632673,"elo":1458.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2033347873,"x_train":7.4221463904,"imp":28.9079076305,"elo":1142.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2010198493,"x_train":1560.1761733922,"imp":24.5082802321,"elo":1224.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2260580575,"x_train":1560.1761733922,"imp":25.0069064835,"elo":1223.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0424445763,"x_train":12.1942564441,"imp":25.2606650045,"elo":1227.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0252302724,"x_train":17528.8314911942,"imp":24.7166459855,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.2817546014,"x_train":17528.8314911942,"imp":23.2989953026,"elo":1319.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7866568202,"x_train":5.1558379712,"imp":38.6291487203,"elo":932.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2836586346,"x_train":968.8626049308,"imp":26.5237510309,"elo":1228.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.8636375274,"x_train":968.8626049308,"imp":23.6440186663,"elo":1301.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6590449294,"x_train":8.3755198931,"imp":37.1837118215,"elo":918.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.0818469447,"x_train":2721.8722746731,"imp":30.2335758512,"elo":1103.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5900916415,"x_train":2721.8722746731,"imp":28.6871013436,"elo":1197.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.4437307263,"x_train":5.1200069793,"imp":26.1196336464,"elo":1195.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.0508933307,"x_train":1280.184939635,"imp":19.7650301564,"elo":1423.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.0776987398,"x_train":1280.184939635,"imp":19.6873087712,"elo":1431.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.6091982733,"x_train":41.1234544852,"imp":13.3825747475,"elo":1497.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":3.5435441032,"x_train":159.4828696115,"imp":22.2933505815,"elo":1395.4,"imputed":true,"imputed_pct":25.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5648398113,"x_train":17.2441843169,"imp":30.064936048,"elo":1101.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.751659725,"x_train":6686.5004259785,"imp":21.6228300791,"elo":1321.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5865347302,"x_train":6686.5004259785,"imp":21.2813099415,"elo":1333.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3262335461,"x_train":20.9501855302,"imp":35.2726368712,"elo":965.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4602862659,"x_train":218.361445605,"imp":62.362333418,"elo":663.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.0713018657,"x_train":1308.4507704502,"imp":56.9967733472,"elo":731.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":9.5934077851,"x_train":1308.4507704502,"imp":53.3028895617,"elo":807.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.147930342,"x_train":0.7350368805,"imp":37.6187219692,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3251277774,"x_train":729.2747129799,"imp":24.9008491087,"elo":1213.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.8330056127,"x_train":729.2747129799,"imp":23.4369156098,"elo":1285.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":6.920194808,"x_train":19.1299431712,"imp":28.1199880238,"elo":1120.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":3.5270444068,"x_train":4335.3941369138,"imp":19.2681742676,"elo":1404.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.142857233,"x_train":4335.3941369138,"imp":18.8131706763,"elo":1440.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8250564093,"x_train":9.8834251377,"imp":13.8715452955,"elo":1537.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.5121585317,"x_train":2793.8656494383,"imp":13.4692880907,"elo":1645.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9280086995,"x_train":2793.8656494383,"imp":5.4947133355,"elo":1735.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7021367127,"x_train":21.8451817987,"imp":23.5561919407,"elo":1305.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8027161329,"x_train":46.3634612718,"imp":26.6996058482,"elo":1194.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4876798649,"x_train":9413.1269943826,"imp":23.5440960871,"elo":1311.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":343.7310101457,"x_train":9413.1269943826,"imp":22.2571294857,"elo":1383.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2735262831,"x_train":5.378689382,"imp":22.4554958659,"elo":1380.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":7.6889691672,"x_train":49.9745503327,"imp":7.0380391136,"elo":1701.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.8545350274,"x_train":4.6668652982,"imp":43.1266606509,"elo":775.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.9479990042,"x_train":11.5082153319,"imp":26.2491237713,"elo":1283.0,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2245693736,"x_train":4.72524967,"imp":15.304534586,"elo":1520.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.624076982,"x_train":14.968259658,"imp":25.1172532623,"elo":1299.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7671600452,"x_train":3796.9310139525,"imp":20.2113064398,"elo":1402.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.5447590803,"x_train":3796.9310139525,"imp":20.1604816148,"elo":1411.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5909006176,"x_train":9.9022651232,"imp":13.1503663168,"elo":1636.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4902017202,"x_train":8.9308378882,"imp":11.3414172871,"elo":1588.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3535753061,"x_train":5.4845158339,"imp":21.4913922326,"elo":1344.1,"imputed":true,"imputed_pct":25.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4074152559,"x_train":2584.9581516158,"imp":22.9167321116,"elo":1328.2,"imputed":true,"imputed_pct":25.0,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.3680288933,"x_train":2584.9581516158,"imp":21.5573536372,"elo":1416.3,"imputed":true,"imputed_pct":25.0,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8225789867,"x_train":666.7207045489,"imp":44.12502013,"elo":798.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.9584829717,"x_train":44782.5951977839,"imp":33.0105868135,"elo":993.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.1023197407,"x_train":44782.5951977839,"imp":33.3175339836,"elo":979.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0760162666,"x_train":3.0219759465,"imp":27.2824558285,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3624794907,"x_train":14.370479118,"imp":33.0285558796,"elo":1011.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.194090358,"x_train":3647.6027397269,"imp":26.495255967,"elo":1209.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1614416305,"x_train":3647.6027397269,"imp":24.7665449214,"elo":1246.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.5916710541,"x_train":4.9305373226,"imp":23.6920112547,"elo":1206.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0396563215,"x_train":2029.7698407236,"imp":20.0582538038,"elo":1369.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":4.1126713529,"x_train":2029.7698407236,"imp":20.270846639,"elo":1372.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":205.5125020069,"x_train":495.6488461748,"imp":30.304216606,"elo":1011.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":206.7186712774,"x_train":27893.2292728524,"imp":22.3359735156,"elo":1266.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":752.505846815,"x_train":27893.2292728524,"imp":20.672850816,"elo":1340.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.371587587,"x_train":8.9914037311,"imp":31.5368880354,"elo":1103.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.3186223488,"x_train":2079.2181272278,"imp":23.8320825397,"elo":1280.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.3466778136,"x_train":2079.2181272278,"imp":22.3592603663,"elo":1356.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
index 7f704e44695b872e908a07307d74b014f05bd904..8adb1023a75bff3a38266af3074880655434003d 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_multiclass/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":283.646286296,"imp":25.0435111439,"elo":1273.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":26.3194957247,"imp":22.3493167591,"elo":1280.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":52.2760454584,"imp":20.1500037908,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":120.7421051485,"imp":19.080273336,"elo":1404.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":216.6422936563,"imp":18.4288826138,"elo":1429.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":415.1322548777,"imp":17.9118649582,"elo":1446.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":727.7924472465,"imp":17.907424396,"elo":1442.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1402.2322064531,"imp":17.2185817434,"elo":1452.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2094.5870262909,"imp":17.022905266,"elo":1445.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":2807.9829916593,"imp":16.766699508,"elo":1456.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":7.4221463904,"imp":28.9264110498,"elo":1134.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":15.2214794665,"imp":28.3548505682,"elo":1144.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":38.0924279868,"imp":27.8896719359,"elo":1157.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":76.6924688657,"imp":27.2792233321,"elo":1172.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":193.4343472071,"imp":26.2870986179,"elo":1187.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":388.7016596181,"imp":25.8410342794,"elo":1203.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":776.6628709802,"imp":25.3611606878,"elo":1211.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1163.7786661849,"imp":25.1575788624,"elo":1215.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1560.1761733922,"imp":25.0263500682,"elo":1222.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":12.1942564441,"imp":25.280086947,"elo":1214.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":81.2230381354,"imp":24.9171703738,"elo":1222.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":314.4258454848,"imp":24.665046851,"elo":1239.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":700.8132948146,"imp":24.6110063021,"elo":1244.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":1955.0904519336,"imp":24.3733254772,"elo":1254.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":4550.043096608,"imp":24.1269970022,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":8811.0765244269,"imp":23.6638108139,"elo":1293.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":13221.5244172938,"imp":23.4665104829,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":17528.8314911942,"imp":23.3215821166,"elo":1309.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":5.1558379712,"imp":38.6548685735,"elo":922.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":9.9671223858,"imp":34.5355870581,"elo":1003.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":24.6192287496,"imp":28.1157418775,"elo":1120.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":48.7817729595,"imp":26.290525095,"elo":1198.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":120.4785637723,"imp":24.3199632065,"elo":1275.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":240.3865094023,"imp":24.1140116869,"elo":1280.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":480.7654959148,"imp":23.917273594,"elo":1285.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":722.1627706934,"imp":23.7249520548,"elo":1295.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":968.8626049308,"imp":23.6836034837,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.3755198931,"imp":37.2010000155,"elo":914.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":21.7646234862,"imp":35.9268121135,"elo":957.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":63.386888105,"imp":32.9373380486,"elo":1060.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":129.352553403,"imp":31.6555401166,"elo":1119.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":331.7596087189,"imp":30.130905947,"elo":1155.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":670.337180032,"imp":29.3757346754,"elo":1178.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1343.6187037298,"imp":28.9237538913,"elo":1196.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2023.8043409809,"imp":28.7849164504,"elo":1199.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2721.8722746731,"imp":28.707296144,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":5.1200069793,"imp":26.1375239488,"elo":1192.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":12.2253394924,"imp":25.1717560691,"elo":1228.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":30.3802629809,"imp":23.8394583519,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":62.1068025438,"imp":22.7497273395,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":158.8472939274,"imp":21.1188155841,"elo":1370.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":317.66743218,"imp":20.4569253318,"elo":1404.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":636.5825631214,"imp":20.0911019659,"elo":1427.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":952.9418266181,"imp":19.9327122653,"elo":1431.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1280.184939635,"imp":19.7071941149,"elo":1442.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":41.1234544852,"imp":13.5726443998,"elo":1503.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":159.4828696115,"imp":22.3133794361,"elo":1405.6,"imputed":true,"imputed_pct":25.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":17.2441843169,"imp":30.0913388364,"elo":1100.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":52.8627222227,"imp":27.4135437542,"elo":1153.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":160.6082009531,"imp":24.3230264211,"elo":1246.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":327.2456661478,"imp":23.5084670506,"elo":1274.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":806.1467207879,"imp":22.8429542598,"elo":1303.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1661.0512370873,"imp":22.3571442902,"elo":1310.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3324.8557036777,"imp":22.2973532555,"elo":1307.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":4975.6687289296,"imp":21.6866732933,"elo":1327.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6686.5004259785,"imp":21.3013142362,"elo":1336.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":20.9501855302,"imp":35.2867241607,"elo":965.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":218.361445605,"imp":62.4165523647,"elo":634.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":445.0214813344,"imp":56.9105891605,"elo":739.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1099.0649504843,"imp":53.4947212047,"elo":798.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1308.4507704502,"imp":53.3105819315,"elo":797.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.7350368805,"imp":37.704633765,"elo":988.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.4865086941,"imp":30.3942325268,"elo":1092.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":15.4344714131,"imp":26.2848781078,"elo":1169.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":33.707023879,"imp":24.185664553,"elo":1204.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":88.3093327016,"imp":22.9930807033,"elo":1284.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":179.7459885669,"imp":22.8033011591,"elo":1290.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":361.9168504232,"imp":23.1479349346,"elo":1289.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":543.5646193801,"imp":23.3089309041,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":729.2747129799,"imp":23.4584275413,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":19.1299431712,"imp":28.1380744773,"elo":1116.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":40.534251589,"imp":23.4798146527,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":105.762419468,"imp":21.1516780122,"elo":1351.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":213.4238147529,"imp":19.6682796097,"elo":1404.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":537.3854228471,"imp":18.6864739532,"elo":1444.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1076.4891299391,"imp":18.1750583629,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2156.0193004102,"imp":17.9600482441,"elo":1469.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3232.7208354751,"imp":18.4802333164,"elo":1454.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4335.3941369138,"imp":18.8339644895,"elo":1447.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":9.8834251377,"imp":13.8918897729,"elo":1535.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":23.9402834756,"imp":13.7190846966,"elo":1536.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":65.6941343642,"imp":13.3510024565,"elo":1578.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":134.7721717608,"imp":13.3373822654,"elo":1625.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":345.3716668375,"imp":12.9022146387,"elo":1677.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":691.4097037424,"imp":11.8337494433,"elo":1716.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1395.1578235853,"imp":10.1426620452,"elo":1727.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2094.6976308922,"imp":8.5965584386,"elo":1735.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2793.8656494383,"imp":5.5180577513,"elo":1750.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":21.8451817987,"imp":23.5764466457,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":46.3634612718,"imp":26.7185554083,"elo":1182.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":95.1581936188,"imp":26.2782127721,"elo":1216.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":234.8668414501,"imp":25.4792251263,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":480.2032798425,"imp":25.0504553383,"elo":1298.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":1187.414451069,"imp":24.0691362287,"elo":1335.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":2358.7610971892,"imp":23.4103451615,"elo":1355.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":4693.5593417073,"imp":23.149303938,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":7018.4549761562,"imp":22.7788710925,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":9413.1269943826,"imp":22.2659177313,"elo":1387.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":5.378689382,"imp":22.475220285,"elo":1382.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":49.9745503327,"imp":7.0591602774,"elo":1697.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":4.6668652982,"imp":43.1436700524,"elo":773.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":11.5082153319,"imp":26.2684058444,"elo":1273.4,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":4.72524967,"imp":15.3246129861,"elo":1525.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":14.968259658,"imp":25.137850187,"elo":1297.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":32.4104572417,"imp":24.2348599246,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":92.1851660275,"imp":23.3456956053,"elo":1344.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":187.6481584341,"imp":22.6376937582,"elo":1375.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":467.1995115985,"imp":22.2009008978,"elo":1382.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":936.7829918111,"imp":21.7471247143,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1869.6886183547,"imp":20.9458040476,"elo":1405.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2822.7067282211,"imp":20.6060794792,"elo":1410.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3796.9310139525,"imp":20.1811207266,"elo":1417.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":9.9022651232,"imp":13.1711052461,"elo":1652.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":8.9308378882,"imp":11.3618793221,"elo":1591.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":5.4845158339,"imp":21.5108601382,"elo":1342.0,"imputed":true,"imputed_pct":25.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":22.8051747724,"imp":22.3406138644,"elo":1354.7,"imputed":true,"imputed_pct":25.0,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":59.3809251643,"imp":22.1898939118,"elo":1392.1,"imputed":true,"imputed_pct":25.0,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":126.9601741408,"imp":21.6739976986,"elo":1415.0,"imputed":true,"imputed_pct":25.0,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":316.7601299191,"imp":21.2460230079,"elo":1433.6,"imputed":true,"imputed_pct":25.0,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":634.2403265363,"imp":20.9517254175,"elo":1450.8,"imputed":true,"imputed_pct":25.0,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1281.8033275426,"imp":21.0681886061,"elo":1445.9,"imputed":true,"imputed_pct":25.0,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1922.320018713,"imp":21.2442103456,"elo":1437.0,"imputed":true,"imputed_pct":25.0,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2584.9581516158,"imp":21.5307179248,"elo":1425.7,"imputed":true,"imputed_pct":25.0,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":666.7207045489,"imp":44.140526255,"elo":780.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1026.6183216352,"imp":43.8746747897,"elo":801.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1652.5474732053,"imp":43.1218234355,"elo":820.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2473.0165808756,"imp":41.5605404935,"elo":843.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5380.2088519459,"imp":38.3590572676,"elo":895.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11008.2551380325,"imp":35.6978698962,"elo":927.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":22866.390108276,"imp":34.6444160317,"elo":951.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":33810.8321985267,"imp":34.0696045971,"elo":966.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":44782.5951977839,"imp":33.3348656952,"elo":990.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":3.0219759465,"imp":27.3010461712,"elo":1201.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.370479118,"imp":33.0704382,"elo":1000.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":28.5804934274,"imp":29.9330040565,"elo":1085.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":89.37412395,"imp":27.7369896196,"elo":1160.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":187.12724454,"imp":26.7896514464,"elo":1210.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":451.1590289841,"imp":26.1087034748,"elo":1217.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":875.4622563657,"imp":25.6721954433,"elo":1224.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1726.0460158647,"imp":25.2486379702,"elo":1238.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2669.9686934389,"imp":24.9321348385,"elo":1250.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3647.6027397269,"imp":24.7891102082,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":4.9305373226,"imp":23.7109406798,"elo":1200.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":15.4118115252,"imp":22.4518668711,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":47.0344042271,"imp":22.0051248606,"elo":1299.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":94.1830833987,"imp":21.4712730772,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":248.7881121837,"imp":21.0359506096,"elo":1344.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":502.7403096728,"imp":20.753793252,"elo":1357.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":1010.0144540287,"imp":20.6022884448,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":1510.9325947178,"imp":20.5077595998,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":2029.7698407236,"imp":20.2903121122,"elo":1375.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":495.6488461748,"imp":30.3223483612,"elo":1006.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1563.1416754751,"imp":27.776668917,"elo":1068.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":4713.7549099867,"imp":24.690106465,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":10547.3540062779,"imp":22.7797109326,"elo":1254.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":26588.4297619676,"imp":20.6790989279,"elo":1339.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":27893.2292728524,"imp":20.6927945888,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":8.9914037311,"imp":31.554912989,"elo":1091.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":19.0589824805,"imp":29.8443643494,"elo":1161.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":49.701827647,"imp":27.2381375084,"elo":1246.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":101.4654106495,"imp":25.6812543014,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":258.6773776274,"imp":24.0675997212,"elo":1313.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":515.1758887421,"imp":23.0996159111,"elo":1342.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":1034.8791027689,"imp":22.6255506547,"elo":1355.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1547.9692908758,"imp":22.4570042735,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2079.2181272278,"imp":22.3955632839,"elo":1364.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.1027625141,"x_train":283.646286296,"imp":25.0435111439,"elo":1273.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2536073213,"x_train":26.3194957247,"imp":22.3493167591,"elo":1280.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4231601839,"x_train":52.2760454584,"imp":20.1500037908,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.6051567129,"x_train":120.7421051485,"imp":19.080273336,"elo":1404.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.7512952022,"x_train":216.6422936563,"imp":18.4288826138,"elo":1429.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.8418166173,"x_train":415.1322548777,"imp":17.9118649582,"elo":1446.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.007557247,"x_train":727.7924472465,"imp":17.907424396,"elo":1442.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1300458847,"x_train":1402.2322064531,"imp":17.2185817434,"elo":1452.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2201190611,"x_train":2094.5870262909,"imp":17.022905266,"elo":1445.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2219998697,"x_train":2807.9829916593,"imp":16.766699508,"elo":1456.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2033347873,"x_train":7.4221463904,"imp":28.9264110498,"elo":1134.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3168559826,"x_train":15.2214794665,"imp":28.3548505682,"elo":1144.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.4637258396,"x_train":38.0924279868,"imp":27.8896719359,"elo":1157.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.6360313896,"x_train":76.6924688657,"imp":27.2792233321,"elo":1172.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.8576608402,"x_train":193.4343472071,"imp":26.2870986179,"elo":1187.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.9940308037,"x_train":388.7016596181,"imp":25.8410342794,"elo":1203.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1389353212,"x_train":776.6628709802,"imp":25.3611606878,"elo":1211.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1914455458,"x_train":1163.7786661849,"imp":25.1575788624,"elo":1215.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2260580575,"x_train":1560.1761733922,"imp":25.0263500682,"elo":1222.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0424445763,"x_train":12.1942564441,"imp":25.280086947,"elo":1214.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.1131096225,"x_train":81.2230381354,"imp":24.9171703738,"elo":1222.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.1833513515,"x_train":314.4258454848,"imp":24.665046851,"elo":1239.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.1834859617,"x_train":700.8132948146,"imp":24.6110063021,"elo":1244.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.1958839273,"x_train":1955.0904519336,"imp":24.3733254772,"elo":1254.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1993753888,"x_train":4550.043096608,"imp":24.1269970022,"elo":1262.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.2146537242,"x_train":8811.0765244269,"imp":23.6638108139,"elo":1293.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.2386362077,"x_train":13221.5244172938,"imp":23.4665104829,"elo":1304.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.2817546014,"x_train":17528.8314911942,"imp":23.3215821166,"elo":1309.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7866568202,"x_train":5.1558379712,"imp":38.6548685735,"elo":922.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9513567957,"x_train":9.9671223858,"imp":34.5355870581,"elo":1003.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7969717524,"x_train":24.6192287496,"imp":28.1157418775,"elo":1120.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0328003281,"x_train":48.7817729595,"imp":26.290525095,"elo":1198.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4148344838,"x_train":120.4785637723,"imp":24.3199632065,"elo":1275.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4054403817,"x_train":240.3865094023,"imp":24.1140116869,"elo":1280.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.6379815682,"x_train":480.7654959148,"imp":23.917273594,"elo":1285.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7473298834,"x_train":722.1627706934,"imp":23.7249520548,"elo":1295.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.8636375274,"x_train":968.8626049308,"imp":23.6836034837,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6590449294,"x_train":8.3755198931,"imp":37.2010000155,"elo":914.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.4924031952,"x_train":21.7646234862,"imp":35.9268121135,"elo":957.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1390542652,"x_train":63.386888105,"imp":32.9373380486,"elo":1060.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.194877341,"x_train":129.352553403,"imp":31.6555401166,"elo":1119.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.8927130955,"x_train":331.7596087189,"imp":30.130905947,"elo":1155.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.3270562101,"x_train":670.337180032,"imp":29.3757346754,"elo":1178.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":9.5813386245,"x_train":1343.6187037298,"imp":28.9237538913,"elo":1196.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":11.4452770419,"x_train":2023.8043409809,"imp":28.7849164504,"elo":1199.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":12.5900916415,"x_train":2721.8722746731,"imp":28.707296144,"elo":1199.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.4437307263,"x_train":5.1200069793,"imp":26.1375239488,"elo":1192.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.7310412682,"x_train":12.2253394924,"imp":25.1717560691,"elo":1228.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.3562691131,"x_train":30.3802629809,"imp":23.8394583519,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.953724397,"x_train":62.1068025438,"imp":22.7497273395,"elo":1335.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.4130882094,"x_train":158.8472939274,"imp":21.1188155841,"elo":1370.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.0010341081,"x_train":317.66743218,"imp":20.4569253318,"elo":1404.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.5773453268,"x_train":636.5825631214,"imp":20.0911019659,"elo":1427.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.2557034144,"x_train":952.9418266181,"imp":19.9327122653,"elo":1431.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.0776987398,"x_train":1280.184939635,"imp":19.7071941149,"elo":1442.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":8.6091982733,"x_train":41.1234544852,"imp":13.5726443998,"elo":1503.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":3.5435441032,"x_train":159.4828696115,"imp":22.3133794361,"elo":1405.6,"imputed":true,"imputed_pct":25.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5648398113,"x_train":17.2441843169,"imp":30.0913388364,"elo":1100.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.1086986149,"x_train":52.8627222227,"imp":27.4135437542,"elo":1153.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.9275358111,"x_train":160.6082009531,"imp":24.3230264211,"elo":1246.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.9365839976,"x_train":327.2456661478,"imp":23.5084670506,"elo":1274.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.7350733756,"x_train":806.1467207879,"imp":22.8429542598,"elo":1303.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":6.2166850654,"x_train":1661.0512370873,"imp":22.3571442902,"elo":1310.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5594018842,"x_train":3324.8557036777,"imp":22.2973532555,"elo":1307.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.7492172244,"x_train":4975.6687289296,"imp":21.6866732933,"elo":1327.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5865347302,"x_train":6686.5004259785,"imp":21.3013142362,"elo":1336.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3262335461,"x_train":20.9501855302,"imp":35.2867241607,"elo":965.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4602862659,"x_train":218.361445605,"imp":62.4165523647,"elo":634.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":5.3441524978,"x_train":445.0214813344,"imp":56.9105891605,"elo":739.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":10.9553867171,"x_train":1099.0649504843,"imp":53.4947212047,"elo":798.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":9.5934077851,"x_train":1308.4507704502,"imp":53.3105819315,"elo":797.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.147930342,"x_train":0.7350368805,"imp":37.704633765,"elo":988.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4717318903,"x_train":4.4865086941,"imp":30.3942325268,"elo":1092.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8456747425,"x_train":15.4344714131,"imp":26.2848781078,"elo":1169.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2491970937,"x_train":33.707023879,"imp":24.185664553,"elo":1204.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.3358363812,"x_train":88.3093327016,"imp":22.9930807033,"elo":1284.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.453941186,"x_train":179.7459885669,"imp":22.8033011591,"elo":1290.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.4155319356,"x_train":361.9168504232,"imp":23.1479349346,"elo":1289.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.5473644093,"x_train":543.5646193801,"imp":23.3089309041,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.8330056127,"x_train":729.2747129799,"imp":23.4584275413,"elo":1282.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.920194808,"x_train":19.1299431712,"imp":28.1380744773,"elo":1116.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.2086544334,"x_train":40.534251589,"imp":23.4798146527,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.1368730282,"x_train":105.762419468,"imp":21.1516780122,"elo":1351.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":11.7650894271,"x_train":213.4238147529,"imp":19.6682796097,"elo":1404.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.4912076922,"x_train":537.3854228471,"imp":18.6864739532,"elo":1444.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.4616303712,"x_train":1076.4891299391,"imp":18.1750583629,"elo":1462.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":16.6029747084,"x_train":2156.0193004102,"imp":17.9600482441,"elo":1469.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":19.7682775511,"x_train":3232.7208354751,"imp":18.4802333164,"elo":1454.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.142857233,"x_train":4335.3941369138,"imp":18.8339644895,"elo":1447.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8250564093,"x_train":9.8834251377,"imp":13.8918897729,"elo":1535.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2057905305,"x_train":23.9402834756,"imp":13.7190846966,"elo":1536.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.4337944176,"x_train":65.6941343642,"imp":13.3510024565,"elo":1578.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.2295930552,"x_train":134.7721717608,"imp":13.3373822654,"elo":1625.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.2854006353,"x_train":345.3716668375,"imp":12.9022146387,"elo":1677.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.7906340457,"x_train":691.4097037424,"imp":11.8337494433,"elo":1716.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4633585633,"x_train":1395.1578235853,"imp":10.1426620452,"elo":1727.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.8898273316,"x_train":2094.6976308922,"imp":8.5965584386,"elo":1735.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.9280086995,"x_train":2793.8656494383,"imp":5.5180577513,"elo":1750.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.7021367127,"x_train":21.8451817987,"imp":23.5764466457,"elo":1300.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.8027161329,"x_train":46.3634612718,"imp":26.7185554083,"elo":1182.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":61.1894558461,"x_train":95.1581936188,"imp":26.2782127721,"elo":1216.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":96.0468231173,"x_train":234.8668414501,"imp":25.4792251263,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":131.2394515616,"x_train":480.2032798425,"imp":25.0504553383,"elo":1298.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":196.2246230598,"x_train":1187.414451069,"imp":24.0691362287,"elo":1335.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":222.0364795428,"x_train":2358.7610971892,"imp":23.4103451615,"elo":1355.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":282.5497243141,"x_train":4693.5593417073,"imp":23.149303938,"elo":1368.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":310.9391931013,"x_train":7018.4549761562,"imp":22.7788710925,"elo":1374.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":343.7310101457,"x_train":9413.1269943826,"imp":22.2659177313,"elo":1387.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2735262831,"x_train":5.378689382,"imp":22.475220285,"elo":1382.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":7.6889691672,"x_train":49.9745503327,"imp":7.0591602774,"elo":1697.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.8545350274,"x_train":4.6668652982,"imp":43.1436700524,"elo":773.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.9479990042,"x_train":11.5082153319,"imp":26.2684058444,"elo":1273.4,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2245693736,"x_train":4.72524967,"imp":15.3246129861,"elo":1525.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.624076982,"x_train":14.968259658,"imp":25.137850187,"elo":1297.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":1.0912900685,"x_train":32.4104572417,"imp":24.2348599246,"elo":1324.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.7054134697,"x_train":92.1851660275,"imp":23.3456956053,"elo":1344.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.3761032339,"x_train":187.6481584341,"imp":22.6376937582,"elo":1375.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":2.9257260459,"x_train":467.1995115985,"imp":22.2009008978,"elo":1382.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.3279465065,"x_train":936.7829918111,"imp":21.7471247143,"elo":1394.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.9820772589,"x_train":1869.6886183547,"imp":20.9458040476,"elo":1405.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.3321042202,"x_train":2822.7067282211,"imp":20.6060794792,"elo":1410.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.5447590803,"x_train":3796.9310139525,"imp":20.1811207266,"elo":1417.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5909006176,"x_train":9.9022651232,"imp":13.1711052461,"elo":1652.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4902017202,"x_train":8.9308378882,"imp":11.3618793221,"elo":1591.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3535753061,"x_train":5.4845158339,"imp":21.5108601382,"elo":1342.0,"imputed":true,"imputed_pct":25.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":2.8664922731,"x_train":22.8051747724,"imp":22.3406138644,"elo":1354.7,"imputed":true,"imputed_pct":25.0,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.9751066064,"x_train":59.3809251643,"imp":22.1898939118,"elo":1392.1,"imputed":true,"imputed_pct":25.0,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":5.8704527559,"x_train":126.9601741408,"imp":21.6739976986,"elo":1415.0,"imputed":true,"imputed_pct":25.0,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":7.1318272574,"x_train":316.7601299191,"imp":21.2460230079,"elo":1433.6,"imputed":true,"imputed_pct":25.0,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":8.820094637,"x_train":634.2403265363,"imp":20.9517254175,"elo":1450.8,"imputed":true,"imputed_pct":25.0,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.6696076637,"x_train":1281.8033275426,"imp":21.0681886061,"elo":1445.9,"imputed":true,"imputed_pct":25.0,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.0910547594,"x_train":1922.320018713,"imp":21.2442103456,"elo":1437.0,"imputed":true,"imputed_pct":25.0,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.3680288933,"x_train":2584.9581516158,"imp":21.5307179248,"elo":1425.7,"imputed":true,"imputed_pct":25.0,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.8225789867,"x_train":666.7207045489,"imp":44.140526255,"elo":780.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.7876620486,"x_train":1026.6183216352,"imp":43.8746747897,"elo":801.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":13.2339480691,"x_train":1652.5474732053,"imp":43.1218234355,"elo":820.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":15.9767784961,"x_train":2473.0165808756,"imp":41.5605404935,"elo":843.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.8983325102,"x_train":5380.2088519459,"imp":38.3590572676,"elo":895.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.7325471822,"x_train":11008.2551380325,"imp":35.6978698962,"elo":927.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":24.6022552378,"x_train":22866.390108276,"imp":34.6444160317,"elo":951.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":28.6345093899,"x_train":33810.8321985267,"imp":34.0696045971,"elo":966.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.1023197407,"x_train":44782.5951977839,"imp":33.3348656952,"elo":990.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0760162666,"x_train":3.0219759465,"imp":27.3010461712,"elo":1201.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3624794907,"x_train":14.370479118,"imp":33.0704382,"elo":1000.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6934737284,"x_train":28.5804934274,"imp":29.9330040565,"elo":1085.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9854209931,"x_train":89.37412395,"imp":27.7369896196,"elo":1160.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.0396624232,"x_train":187.12724454,"imp":26.7896514464,"elo":1210.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2562289121,"x_train":451.1590289841,"imp":26.1087034748,"elo":1217.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.445176011,"x_train":875.4622563657,"imp":25.6721954433,"elo":1224.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6802253472,"x_train":1726.0460158647,"imp":25.2486379702,"elo":1238.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.9118329956,"x_train":2669.9686934389,"imp":24.9321348385,"elo":1250.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.1614416305,"x_train":3647.6027397269,"imp":24.7891102082,"elo":1248.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5916710541,"x_train":4.9305373226,"imp":23.7109406798,"elo":1200.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.9620057603,"x_train":15.4118115252,"imp":22.4518668711,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.5632518606,"x_train":47.0344042271,"imp":22.0051248606,"elo":1299.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.7564565665,"x_train":94.1830833987,"imp":21.4712730772,"elo":1321.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.5049134408,"x_train":248.7881121837,"imp":21.0359506096,"elo":1344.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.9073614098,"x_train":502.7403096728,"imp":20.753793252,"elo":1357.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.4289445362,"x_train":1010.0144540287,"imp":20.6022884448,"elo":1363.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.0567675025,"x_train":1510.9325947178,"imp":20.5077595998,"elo":1371.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.1126713529,"x_train":2029.7698407236,"imp":20.2903121122,"elo":1375.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":205.5125020069,"x_train":495.6488461748,"imp":30.3223483612,"elo":1006.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":447.1825554268,"x_train":1563.1416754751,"imp":27.776668917,"elo":1068.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":729.916383431,"x_train":4713.7549099867,"imp":24.690106465,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":828.8918571484,"x_train":10547.3540062779,"imp":22.7797109326,"elo":1254.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":777.6640795695,"x_train":26588.4297619676,"imp":20.6790989279,"elo":1339.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":752.505846815,"x_train":27893.2292728524,"imp":20.6927945888,"elo":1338.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.371587587,"x_train":8.9914037311,"imp":31.554912989,"elo":1091.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.7371554717,"x_train":19.0589824805,"imp":29.8443643494,"elo":1161.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.3176035748,"x_train":49.701827647,"imp":27.2381375084,"elo":1246.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":2.7650973713,"x_train":101.4654106495,"imp":25.6812543014,"elo":1268.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.4463880936,"x_train":258.6773776274,"imp":24.0675997212,"elo":1313.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.6826514001,"x_train":515.1758887421,"imp":23.0996159111,"elo":1342.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":3.024314406,"x_train":1034.8791027689,"imp":22.6255506547,"elo":1355.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":3.3820709216,"x_train":1547.9692908758,"imp":22.4570042735,"elo":1362.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.3466778136,"x_train":2079.2181272278,"imp":22.3955632839,"elo":1364.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..e81a51a043d9408b22a3336dd5e9adad281ba0a2
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_multiclass/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "TabFM (default)", "RealTabPFN-2.5 (tuned)", "TabPFN-2.6 (default)", "TabPFN-3 (default)", "RealTabPFN-2.5 (default)", "TabICLv2 (default)", "LimiX (default)", "AutoGluon 1.4 (best, 4h)", "CatBoost (tuned)", "CatBoost (tuned + ensembled)", "RealMLP (tuned + ensembled)", "LightGBM (tuned + ensembled)", "LightGBM (tuned)", "TabPFNv2 (tuned + ensembled)", "TabM (tuned + ensembled)", "RealMLP (tuned)", "TabM (tuned)", "Mitra (default)", "TabDPT (tuned + ensembled)", "TabDPT-Turbo (default)", "XGBoost (tuned + ensembled)", "XGBoost (tuned)", "xRFM (tuned + ensembled)", "TabPFNv2 (default)", "iLTM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "TabPFNv2 (tuned)", "ModernNCA (tuned)", "EBM (tuned + ensembled)", "TabDPT (tuned)", "SAP-RPT-OSS (default)", "ExtraTrees (tuned + ensembled)", "TabM (default)", "RandomForest (tuned + ensembled)", "CatBoost (default)", "TabICL (default)", "xRFM (tuned)", "BetaTabPFN (default)", "iLTM (tuned)", "EBM (tuned)", "TorchMLP (tuned + ensembled)", "ExtraTrees (tuned)", "EBM (default)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "RandomForest (tuned)", "TorchMLP (tuned)", "XGBoost (default)", "TabSwift (default)", "FastaiMLP (tuned + ensembled)", "LightGBM (default)", "TabDPT (default)", "ChimeraBoost (default)", "RealMLP (default)", "FastaiMLP (tuned)", "xRFM (default)", "ModernNCA (default)", "KNN (tuned + ensembled)", "iLTM (default)", "TorchMLP (default)", "RandomForest (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "OrionMSP (default)", "ExtraTrees (default)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (default)", "KNN (tuned)", "TabFlex (default)", "PerpetualBooster (tuned)", "Linear (default)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.375, 0.25, 0.75, 0.625, 0.75, 0.5, 0.75, 0.875, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 0.75, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.625, NaN, 0.375, 0.75, 0.75, 0.625, 1.0, 0.75, 0.75, 0.75, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 0.75, 1.0, 0.75, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.75, 1.0, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.75, 0.625, NaN, 0.75, 0.75, 0.75, 0.875, 1.0, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.25, 0.25, NaN, 0.625, 0.375, 0.6875, 0.625, 0.5, 0.625, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.875, 1.0, 0.75, 0.875, 0.75, 0.75, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 0.75, 0.875, 0.875, 0.75, 0.875, 1.0, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.375, 0.25, 0.25, 0.375, NaN, 0.5, 0.75, 0.5, 0.75, 0.5, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.75, 1.0, 0.75, 0.875, 1.0, 1.0, 0.875, 0.75, 1.0, 1.0, 0.75, 1.0, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 1.0, 0.75, 0.75, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.375, 0.25, 0.625, 0.5, NaN, 0.5, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.625, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.75, 0.75, 1.0, 1.0, 0.875, 0.875, 0.75, 0.75, 0.625, 0.875, 0.75, 0.875, 0.75, 1.0, 0.875, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.5, 0.0, 0.125, 0.3125, 0.25, 0.5, NaN, 0.5, 0.75, 0.5, 0.75, 0.75, 0.75, 0.75, 0.75, 0.5, 0.75, 0.75, 0.75, 0.875, 0.75, 0.5, 0.75, 0.75, 0.75, 1.0, 0.75, 0.75, 1.0, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 0.75, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.25, 0.0, 0.375, 0.5, 0.25, 0.5, NaN, 0.625, 0.625, 0.625, 0.625, 0.625, 0.875, 0.875, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.875, 0.75, 0.75, 0.625, 0.875, 0.875, 0.875, 0.75, 0.75, 0.625, 0.75, 0.5, 0.75, 0.75, 0.75, 0.75, 1.0, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.25, 0.5, 0.25, 0.375, 0.25, 0.375, NaN, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 0.5, 0.5, 0.625, 0.75, 0.75, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.75, 0.625, 0.625, 0.875, 0.75, 0.875, 0.75, 0.75, 0.875, 0.75, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 0.75, 0.875, 0.625, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.25, 0.375, 0.5, 0.5, 0.5, 0.375, 0.5, NaN, 0.625, 0.625, 0.5, 0.75, 0.75, 0.375, 0.5, 0.5, 0.625, 0.5, 0.5, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.625, 0.625, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 0.625, 0.625, 0.75, 0.625, 0.875, 0.875, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.0, 0.125, 0.25, 0.125, 0.375, 0.25, 0.375, 0.5, 0.375, NaN, 0.375, 0.375, 0.5, 0.5, 0.375, 0.625, 0.625, 0.75, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 0.625, 0.75, 0.875, 0.75, 0.75, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.625, 0.875, 0.625, 0.875, 0.75, 0.875, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.75, 1.0, 0.75, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.0, 0.125, 0.125, 0.125, 0.375, 0.25, 0.375, 0.375, 0.375, 0.625, NaN, 0.375, 0.375, 0.375, 0.375, 0.75, 0.625, 0.75, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 0.5, 0.75, 0.875, 0.75, 0.75, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.625, 0.875, 0.75, 0.875, 0.75, 0.875, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.75, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.375, 0.25, 0.375, 0.375, 0.5, 0.625, 0.625, NaN, 0.625, 0.625, 0.375, 0.625, 0.625, 0.625, 0.375, 0.5, 0.5, 0.875, 0.75, 0.625, 0.5, 0.75, 0.75, 0.625, 0.75, 0.75, 0.5, 0.625, 0.5, 0.875, 0.5, 0.75, 0.625, 0.625, 0.75, 0.875, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.875, 0.75, 0.875, 0.875, 0.625, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.5, 0.625, 0.375, NaN, 0.5, 0.375, 0.5, 0.5, 0.5, 0.375, 0.5, 0.75, 0.5, 0.625, 0.5, 0.625, 0.625, 0.5, 0.625, 0.5, 0.75, 0.625, 0.625, 0.625, 0.75, 0.625, 0.875, 0.625, 0.625, 0.875, 0.625, 0.875, 0.875, 0.625, 0.875, 1.0, 1.0, 0.875, 0.75, 1.0, 0.875, 0.75, 0.875, 0.75, 0.875, 0.75, 1.0, 0.75, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.5, 0.625, 0.375, 0.5, NaN, 0.375, 0.5, 0.5, 0.5, 0.375, 0.5, 0.75, 0.5, 0.5, 0.5, 0.5, 0.625, 0.5, 0.5, 0.5, 0.75, 0.625, 0.625, 0.625, 0.75, 0.625, 0.875, 0.625, 0.625, 0.875, 0.625, 0.875, 0.875, 0.625, 0.875, 1.0, 1.0, 0.875, 0.75, 1.0, 0.875, 0.75, 0.875, 0.75, 0.875, 0.75, 1.0, 0.75, 0.875, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.0, 0.0, 0.25, 0.125, 0.25, 0.5, 0.375, 0.375, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, NaN, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.75, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.5, 0.625, 0.625, 0.5, 0.625, 0.75, 0.5, 0.75, 0.5, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.125, 0.125, 0.25, 0.375, 0.25, 0.375, 0.25, 0.5, 0.375, 0.25, 0.375, 0.5, 0.5, 0.375, NaN, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.5, 0.375, 0.5, 0.625, 0.375, 0.75, 0.875, 0.75, 0.625, 0.5, 0.875, 0.625, 0.625, 0.625, 0.625, 0.875, 0.625, 0.875, 0.875, 0.625, 0.875, 0.625, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.0, 0.125, 0.0, 0.125, 0.125, 0.25, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, 0.375, NaN, 0.5, 0.375, 0.5, 0.375, 0.625, 0.75, 0.5, 0.5, 0.625, 0.75, 0.625, 0.875, 0.75, 0.625, 0.5, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.875, 0.75, 0.75, 0.75, 0.5, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.375, 0.5, 0.5, 0.375, 0.5, 0.5, NaN, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.375, 0.5, 0.625, 0.375, 0.75, 0.875, 0.75, 0.625, 0.5, 0.875, 0.625, 0.625, 0.625, 0.625, 0.875, 0.625, 0.875, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.0, 0.125, 0.0, 0.25, 0.125, 0.25, 0.5, 0.5, 0.375, 0.375, 0.625, 0.625, 0.625, 0.375, 0.5, 0.625, 0.5, NaN, 0.5, 0.5, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.625, 0.625, 0.625, 0.625, 0.375, 0.625, 0.75, 0.625, 0.75, 0.5625, 0.75, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 0.625, 0.75, 0.875, 0.75, 0.875, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0], [0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.5, 0.5, 0.25, 0.25, 0.5, 0.5, 0.5, 0.375, 0.375, 0.5, 0.375, 0.5, NaN, 0.375, 0.625, 0.625, 0.5, 0.625, 0.75, 0.625, 0.625, 0.625, 0.625, 0.875, 0.5, 0.5, 0.5, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.875, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0], [0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.5, 0.125, 0.375, 0.375, 0.5, 0.5, 0.5, 0.25, 0.25, 0.5, 0.375, 0.625, 0.5, 0.5, 0.625, NaN, 0.625, 0.625, 0.375, 0.5, 0.75, 0.625, 0.5, 0.625, 0.5, 0.625, 0.5, 0.625, 0.5, 0.625, 0.75, 0.625, 0.625, 0.5, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 1.0, 0.625, 0.75, 0.875, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 0.75, 0.75, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.5, 0.5, 0.25, 0.375, 0.375, 0.5, 0.25, 0.375, 0.375, NaN, 0.625, 0.5, 0.5, 0.75, 0.75, 0.625, 0.625, 0.625, 0.5, 0.5, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.875, 0.875, 0.625, 0.75, 1.0, 0.75, 0.625, 0.875, 0.625, 0.875, 0.75, 0.875, 0.625, 1.0, 0.75, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.5, 0.375, 0.375, 0.25, 0.375, 0.25, 0.375, 0.375, 0.375, NaN, 0.5, 0.625, 0.625, 0.75, 0.625, 0.625, 0.625, 0.5, 0.5, 0.625, 0.625, 0.5, 0.75, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.875, 0.75, 0.875, 0.875, 0.625, 0.75, 1.0, 0.75, 0.625, 0.875, 0.625, 0.875, 0.75, 0.875, 0.625, 1.0, 0.75, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.25, 0.5, 0.5, 0.375, 0.25, 0.5, 0.625, 0.5, 0.5, NaN, 0.375, 0.5, 0.625, 0.375, 0.5, 0.5, 0.625, 0.375, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.625, 0.5, 0.5, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.625, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 1.0, 0.875, 1.0, 0.75, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.0, 0.0, 0.125, 0.25, 0.25, 0.0, 0.125, 0.375, 0.25, 0.375, 0.5, 0.5, 0.375, 0.5, 0.375, 0.625, 0.5, 0.625, 0.375, 0.375, 0.5, 0.5, 0.375, 0.625, NaN, 0.375, 0.375, 0.5, 0.5, 0.625, 0.625, 0.375, 0.625, 0.75, 0.625, 0.625, 0.5625, 0.75, 0.625, 0.375, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.625, 0.875, 0.625, 0.75, 0.875, 0.75, 0.75, 0.5, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.125, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.25, 0.25, 0.25, 0.25, 0.375, 0.5, 0.625, NaN, 0.5, 0.625, 0.75, 0.5, 0.5, 0.5, 0.75, 0.5, 0.625, 0.5, 0.5, 0.75, 0.5, 0.625, 0.625, 0.625, 0.75, 0.875, 0.75, 0.75, 0.625, 0.875, 0.75, 0.625, 0.75, 0.625, 0.625, 0.875, 0.875, 0.75, 0.75, 1.0, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.125, 0.25, 0.375, 0.125, 0.125, 0.25, 0.5, 0.5, 0.375, 0.375, 0.25, 0.375, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.625, 0.5, NaN, 0.5, 0.6875, 0.5, 0.5, 0.5, 0.75, 0.5, 0.625, 0.625, 0.75, 0.75, 0.625, 0.625, 0.5, 0.625, 0.75, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.25, 0.125, 0.0, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.5, 0.25, 0.625, 0.375, 0.625, 0.375, 0.375, 0.5, 0.375, 0.375, 0.625, 0.5, 0.375, 0.5, NaN, 0.5, 0.625, 0.5, 0.375, 0.625, 0.75, 0.5, 0.5, 0.5, 0.75, 0.625, 0.5, 0.625, 0.75, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.875, 0.75, 0.75, 0.5, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0], [0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.375, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.5, 0.5, 0.5, 0.25, 0.125, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.25, 0.3125, 0.5, NaN, 0.375, 0.375, 0.625, 0.75, 0.5, 0.625, 0.625, 0.625, 0.75, 0.5, 0.75, 0.625, 0.5, 0.75, 0.375, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 0.75, 1.0, 0.875, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.25, 0.375, 0.125, 0.25, 0.125, 0.375, 0.375, 0.5, 0.375, 0.375, 0.5, 0.375, 0.5, 0.5, 0.375, 0.625, NaN, 0.375, 0.625, 0.625, 0.25, 0.625, 0.75, 0.5, 0.625, 0.5, 0.75, 0.875, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.75, 0.75, 0.625, 0.75, 0.75, 0.875, 0.75, 1.0, 1.0, 0.875, 0.75, 0.875, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.5, 0.375, 0.375, 0.375, 0.25, 0.375, 0.25, 0.375, 0.125, 0.375, 0.5, 0.5, 0.375, 0.375, 0.5, 0.5, 0.5, 0.625, 0.625, NaN, 0.5, 0.375, 0.5, 0.375, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.875, 0.625, 0.75, 0.75, 0.875, 0.625, 0.875, 0.75, 0.875, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0], [0.25, 0.0, 0.125, 0.125, 0.25, 0.375, 0.25, 0.5, 0.375, 0.375, 0.5, 0.5, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.625, 0.5, 0.5, 0.5, 0.5, 0.625, 0.625, 0.5, 0.5, 0.625, 0.375, 0.375, 0.5, NaN, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.5, 0.625, 0.5, 0.5, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.5, 0.625, 0.75, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.5, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.5, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.25, 0.375, 0.25, 0.375, 0.625, 0.5, NaN, 0.5, 0.75, 0.5, 0.375, 0.375, 0.625, 0.5, 0.5, 0.5, 0.8125, 0.5, 0.625, 0.625, 0.75, 0.625, 0.5, 0.75, 0.5, 0.75, 0.75, 0.75, 0.625, 0.75, 0.875, 0.625, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.75, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0], [0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.125, 0.25, 0.5, 0.5, 0.25, 0.375, 0.375, 0.25, 0.5, 0.5, 0.25, 0.5, 0.75, 0.5, 0.375, 0.5, NaN, 0.5, 0.625, 0.5, 0.5, 0.75, 0.625, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.625, 0.5, 0.625, 0.5, 0.625, 0.875, 0.75, 0.75, 0.75, 0.75, 1.0, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.375, 0.25, 0.125, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.625, 0.5, 0.25, 0.5, NaN, 0.5, 0.5, 0.25, 0.5, 0.5, 0.5, 0.5, 0.625, 0.5, 0.75, 0.625, 0.875, 0.5, 0.625, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 0.75, 0.625, 0.875, 0.875, 1.0, 0.875, 1.0, 0.75, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.375, 0.375, 0.375, 0.25, 0.25, 0.25, 0.25, 0.25, 0.5, 0.375, 0.5, 0.375, 0.5, 0.375, 0.25, 0.375, 0.5, 0.5, 0.375, 0.5, NaN, 0.5, 0.625, 0.25, 0.75, 0.375, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.75, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.0, 0.25, 0.25, 0.0, 0.25, 0.0, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.4375, 0.375, 0.375, 0.25, 0.25, 0.375, 0.4375, 0.5, 0.25, 0.5, 0.375, 0.5, 0.5, 0.375, 0.625, 0.5, 0.5, 0.5, NaN, 0.5, 0.625, 0.625, 0.5, 0.5, 0.75, 0.5, 0.5, 0.5, 0.625, 0.75, 0.625, 0.75, 0.5, 0.75, 0.75, 0.75, 0.625, 0.625, 0.625, 0.625, 0.875, 0.75, 0.75, 0.8125, 0.75, 0.75, 0.875, 0.875, 0.75, 0.875, 0.75, 1.0, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.5, 0.375, 0.625, 0.5, 0.75, 0.375, 0.5, NaN, 0.5, 0.5, 0.375, 0.375, 0.75, 0.375, 0.75, 0.5, 0.75, 0.5, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.75, 0.625, 1.0, 0.625, 1.0, 0.75, 1.0, 0.75, 0.875, 0.875, 1.0, 1.0, 0.75, 0.875, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.5, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.5, 0.5, 0.375, 0.375, 0.375, 0.25, 0.5, 0.75, 0.375, 0.5, NaN, 0.625, 0.875, 0.625, 0.375, 0.5, 0.625, 0.625, 0.625, 0.5, 0.5, 0.5, 0.625, 0.5, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.25, 0.375, 0.375, 0.125, 0.125, 0.125, 0.375, 0.375, 0.375, 0.375, 0.125, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.5, 0.625, 0.375, 0.375, 0.5, 0.25, 0.25, 0.375, 0.5, 0.5, 0.375, 0.5, 0.25, 0.375, 0.5, 0.375, NaN, 0.375, 0.625, 0.75, 0.625, 0.5, 0.5, 0.625, 0.625, 0.5, 0.625, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.875, 0.875, 1.0, 0.875, 0.875, 0.75, 0.75, 0.875, 1.0, 0.75, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.5, 0.375, 0.375, 0.125, 0.375, 0.375, 0.5, 0.25, 0.5, 0.625, 0.5, 0.625, 0.125, 0.625, NaN, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.375, 0.5, 0.625, 0.375, 0.625, 0.75, 0.625, 0.75, 0.75, 0.875, 0.75, 1.0, 0.875, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.25, 0.5, 0.5, 0.5, 0.375, 0.5, 0.375, 0.5, 0.375, 0.5, 0.625, 0.375, 0.375, 0.5, NaN, 0.5, 0.625, 0.5, 0.5, 0.5, 0.75, 0.625, 0.5, 0.625, 0.625, 0.75, 0.5, 0.75, 0.75, 0.75, 0.625, 0.75, 0.75, 1.0, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.25, 0.25, 0.375, 0.25, 0.375, 0.375, 0.5, 0.1875, 0.5, 0.375, 0.5, 0.25, 0.25, 0.625, 0.25, 0.5, 0.5, NaN, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.625, 0.5, 0.75, 0.625, 0.75, 0.5, 0.625, 0.625, 0.625, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.75, 0.625, 0.625, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.125, 0.25, 0.25, 0.625, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.5, 0.625, 0.5, 0.375, 0.5, 0.375, 0.625, NaN, 0.375, 0.5, 0.625, 0.5, 0.375, 0.5, 0.625, 0.375, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 1.0, 0.875, 0.625, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.0, 0.25, 0.25, 0.25, 0.0, 0.0, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.25, 0.375, 0.5, 0.25, 0.375, 0.5, 0.5, 0.5, 0.5, 0.625, NaN, 0.625, 0.625, 0.625, 0.5, 0.625, 0.5, 0.625, 0.75, 0.75, 0.75, 0.625, 0.75, 0.5, 0.625, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 0.75, 0.875, 0.75, 0.75, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.0, 0.25, 0.25, 0.25, 0.0, 0.0, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.375, NaN, 0.625, 0.625, 0.375, 0.625, 0.5, 0.5, 0.75, 0.875, 0.75, 0.625, 0.75, 0.625, 0.625, 0.875, 0.75, 0.75, 0.875, 0.875, 0.75, 0.75, 1.0, 0.875, 1.0, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0], [0.0, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.125, 0.125, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.5, 0.25, 0.375, 0.125, 0.375, 0.375, 0.25, 0.375, 0.375, 0.5, 0.5, 0.5, 0.375, 0.375, 0.375, NaN, 0.5, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.375, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.75, 0.625, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.125, 0.25, 0.25, 0.375, 0.375, 0.5, 0.375, 0.375, 0.375, 0.5, 0.375, 0.25, 0.5, 0.5, 0.375, 0.5, 0.25, 0.5, 0.5, 0.375, 0.375, 0.5, NaN, 0.375, 0.375, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.875, 0.75, 0.875, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.125, 0.0, 0.0, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.5, 0.25, 0.375, 0.5, 0.375, 0.5, 0.5, 0.5, 0.625, 0.375, 0.5, 0.625, 0.5, 0.625, 0.5, 0.625, NaN, 0.625, 0.5, 0.625, 0.5, 0.75, 0.625, 0.625, 0.5, 0.75, 0.75, 1.0, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.0, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.5, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.5, 0.375, 0.5, 0.5, 0.375, 0.5, 0.375, 0.375, 0.375, 0.625, 0.375, NaN, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.5, 0.625, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.75, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0], [0.125, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.125, 0.25, 0.0, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.5, 0.5, 0.375, 0.25, 0.375, 0.375, 0.5, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.375, NaN, 0.5, 0.375, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.75, 0.5, 0.625, 0.75, 0.75, 0.625, 0.75, 1.0, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.375, 0.125, 0.375, 0.25, 0.25, 0.375, 0.5, 0.375, 0.375, 0.25, 0.5, 0.5, 0.375, 0.25, 0.5, 0.5, 0.375, 0.625, 0.375, 0.25, 0.625, 0.375, 0.5, 0.5, 0.375, 0.375, 0.25, 0.5, NaN, 0.375, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.5, 0.875, 0.875, 0.875, 0.75, 1.0, 0.75, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.375, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.25, 0.25, 0.5, 0.25, 0.375, 0.125, 0.125, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.125, 0.5, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.5, 0.375, 0.625, 0.625, NaN, 0.5, 0.625, 0.75, 0.75, 0.5, 0.875, 0.75, 0.625, 0.625, 0.75, 0.75, 0.75, 1.0, 0.75, 0.5, 0.625, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.375, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.5, 0.375, 0.25, 0.25, 0.375, 0.25, 0.375, 0.25, 0.5, 0.25, 0.25, 0.25, 0.125, 0.375, 0.375, 0.25, 0.375, 0.5, 0.375, 0.5, NaN, 0.625, 0.5, 0.5, 0.625, 0.625, 0.75, 0.625, 0.5, 0.625, 0.875, 0.625, 0.625, 0.75, 0.75, 0.75, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.25, 0.125, 0.0, 0.0, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.25, 0.5, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.5, 0.375, 0.375, 0.375, NaN, 0.5, 0.625, 0.5, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.375, 0.25, 0.375, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.25, 0.5, 0.5, NaN, 0.625, 0.625, 0.625, 0.75, 0.5, 0.5, 0.625, 0.625, 0.625, 0.75, 1.0, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.375, 0.375, 0.0, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.375, 0.125, 0.25, 0.25, 0.25, 0.375, 0.0, 0.25, 0.375, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.25, 0.5, 0.5, 0.375, 0.375, 0.25, 0.5, 0.375, 0.375, NaN, 0.5, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.625, 0.625, 0.625, 0.625, 0.75, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.0, 0.0, 0.125, 0.5, 0.0, 0.0, 0.5, 0.0, 0.125, 0.375, 0.375, 0.375, 0.25, 0.375, 0.25, 0.375, 0.375, 0.25, 0.125, 0.125, 0.375, 0.375, 0.25, 0.5, 0.375, 0.625, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.5, NaN, 0.625, 0.5, 0.75, 0.875, 0.625, 0.625, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.625, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.0, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.375, 0.0, 0.25, 0.125, 0.25, 0.125, 0.0, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.375, 0.375, 0.25, 0.125, 0.25, 0.25, 0.375, 0.25, 0.125, 0.375, 0.25, 0.375, 0.375, 0.375, NaN, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, 0.375, 0.75, 0.375, 1.0, 0.625, 0.75, 0.625, 0.875, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.0, 0.0, 0.25, 0.25, 0.0, 0.125, 0.25, 0.0, 0.0, 0.25, 0.375, 0.125, 0.25, 0.25, 0.0, 0.25, 0.25, 0.125, 0.0, 0.0, 0.25, 0.25, 0.0, 0.125, 0.125, 0.25, 0.25, 0.0, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.375, 0.5, 0.625, NaN, 0.625, 0.625, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.125, 0.0, 0.125, 0.0, 0.25, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.25, 0.5, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.25, 0.625, 0.375, NaN, 0.375, 0.75, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.0, 0.125, 0.25, 0.375, 0.125, 0.25, 0.125, 0.0, 0.1875, 0.25, 0.25, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.125, 0.375, 0.125, 0.125, 0.375, 0.5, 0.375, 0.5, 0.375, 0.5, 0.25, 0.125, 0.625, 0.375, 0.625, NaN, 0.625, 0.625, 0.375, 0.625, 0.5, 0.75, 0.75, 0.75, 0.75, 0.875, 0.625, 0.75, 0.75, 0.75, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.25, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.375, 0.375, 0.25, 0.375, 0.625, 0.25, 0.25, 0.375, NaN, 0.4375, 0.375, 0.5, 0.625, 0.75, 0.75, 0.75, 1.0, 0.875, 0.625, 0.75, 0.875, 0.75, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.25, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.125, 0.25, 0.375, 0.25, 0.375, 0.625, 0.375, 0.25, 0.375, 0.5625, NaN, 0.5, 0.5, 0.5, 0.75, 0.75, 0.75, 1.0, 0.875, 0.625, 0.75, 0.875, 0.75, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.25, 0.0, 0.375, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.25, 0.0, 0.25, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.375, 0.125, 0.25, 0.375, 0.375, 0.375, 0.25, 0.25, 0.625, 0.5, 0.375, 0.625, 0.625, 0.5, NaN, 0.625, 0.625, 0.625, 0.625, 0.75, 0.875, 0.875, 0.75, 0.75, 0.75, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.25, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.0, 0.375, 0.25, 0.25, 0.375, 0.25, 0.5, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, NaN, 0.75, 0.5, 0.5, 0.625, 0.625, 0.875, 0.625, 0.875, 0.75, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.0, 0.25, 0.25, 0.0, 0.25, 0.125, 0.125, 0.25, 0.125, 0.125, 0.0, 0.25, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.25, 0.0, 0.375, 0.25, 0.5, 0.375, 0.25, 0.5, 0.375, 0.5, 0.375, 0.25, NaN, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 0.5, 0.875, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.25, 0.0, 0.25, 0.125, 0.125, 0.125, 0.0, 0.25, 0.125, 0.125, 0.375, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.125, 0.25, 0.5, 0.25, 0.25, 0.25, 0.375, 0.125, 0.625, 0.375, 0.375, 0.25, 0.25, 0.25, 0.375, 0.5, 0.375, NaN, 0.6875, 0.5, 0.5, 0.875, 0.625, 0.5, 1.0, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.125, 0.25, 0.0, 0.375, 0.0, 0.25, 0.25, 0.0, 0.125, 0.0, 0.0, 0.375, 0.125, 0.25, 0.0, 0.375, 0.125, 0.125, 0.25, 0.125, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.125, 0.625, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.5, 0.25, 0.3125, NaN, 0.5, 0.5, 0.875, 0.625, 0.5, 0.875, 0.75, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.5, 0.5, 0.5, NaN, 0.25, 0.625, 0.625, 0.875, 0.625, 1.0, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.25, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.375, 0.625, 0.125, 0.125, 0.25, 0.0, 0.0, 0.125, 0.375, 0.25, 0.5, 0.5, 0.75, NaN, 0.75, 0.375, 0.75, 0.75, 0.75, 1.0], [0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.25, 0.0, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.375, 0.25, NaN, 0.25, 0.5, 0.25, 0.5, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.125, 0.375, 0.25, 0.125, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.625, 0.75, NaN, 0.375, 0.625, 0.5, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.5, 0.5, 0.5, 0.125, 0.25, 0.5, 0.625, NaN, 0.625, 0.6875, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.375, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.0, 0.125, 0.375, 0.25, 0.75, 0.375, 0.375, NaN, 0.625, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.0, 0.125, 0.0, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.0, 0.25, 0.5, 0.5, 0.3125, 0.375, NaN, 0.75], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.25, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9090909091},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9090909091},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.8928571429},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8628246753},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8571428571},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8262987013},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7881493506},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.7743506494},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7548701299},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7451298701},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7207792208},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7207792208},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7045454545},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6964285714},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6883116883},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6818181818},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6769480519},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6704545455},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6688311688},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.661525974},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6493506494},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6461038961},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.637987013},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6347402597},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6217532468},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.6079545455},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6038961039},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5965909091},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5909090909},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5836038961},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5811688312},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5730519481},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.5665584416},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5625},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5600649351},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5454545455},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5438311688},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.5430194805},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5405844156},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.538961039},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.525974026},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.5113636364},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5048701299},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4862012987},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4853896104},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4821428571},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4805194805},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4707792208},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4659090909},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4626623377},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.4577922078},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4529220779},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4512987013},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4496753247},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.3961038961},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.375},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3587662338},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3587662338},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3571428571},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.2824675325},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2792207792},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2792207792},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2702922078},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2654220779},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2540584416},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2435064935},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2191558442},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2094155844},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.2037337662},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.1891233766},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1428571429},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.137987013},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1347402597},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1266233766},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1063311688},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0974025974},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0803571429},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0064935065}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..4b6ff330acc1bd66cd3213864db647ae6933f917
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_all/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),TabFM (default),RealTabPFN-2.5 (tuned),TabPFN-2.6 (default),TabPFN-3 (default),RealTabPFN-2.5 (default),TabICLv2 (default),LimiX (default),"AutoGluon 1.4 (best, 4h)",CatBoost (tuned),CatBoost (tuned + ensembled),RealMLP (tuned + ensembled),LightGBM (tuned + ensembled),LightGBM (tuned),TabPFNv2 (tuned + ensembled),TabM (tuned + ensembled),RealMLP (tuned),TabM (tuned),Mitra (default),TabDPT (tuned + ensembled),TabDPT-Turbo (default),XGBoost (tuned + ensembled),XGBoost (tuned),xRFM (tuned + ensembled),TabPFNv2 (default),iLTM (tuned + ensembled),ModernNCA (tuned + ensembled),TabPFNv2 (tuned),ModernNCA (tuned),EBM (tuned + ensembled),TabDPT (tuned),SAP-RPT-OSS (default),ExtraTrees (tuned + ensembled),TabM (default),RandomForest (tuned + ensembled),CatBoost (default),TabICL (default),xRFM (tuned),BetaTabPFN (default),iLTM (tuned),EBM (tuned),TorchMLP (tuned + ensembled),ExtraTrees (tuned),EBM (default),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),RandomForest (tuned),TorchMLP (tuned),XGBoost (default),TabSwift (default),FastaiMLP (tuned + ensembled),LightGBM (default),TabDPT (default),ChimeraBoost (default),RealMLP (default),FastaiMLP (tuned),xRFM (default),ModernNCA (default),KNN (tuned + ensembled),iLTM (default),TorchMLP (default),RandomForest (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),OrionMSP (default),ExtraTrees (default),FastaiMLP (default),Linear (tuned + ensembled),Linear (tuned),PerpetualBooster (tuned + ensembled),TabSTAR (default),KNN (tuned),TabFlex (default),PerpetualBooster (tuned),Linear (default),PerpetualBooster (default),KNN (default)
+"AutoGluon 1.5 (extreme, 4h)",,0.375,0.25,0.75,0.625,0.75,0.5,0.75,0.875,0.875,0.75,0.75,0.875,0.875,0.875,0.75,0.875,0.75,0.875,1.0,0.875,0.875,1.0,1.0,0.875,0.875,1.0,1.0,1.0,0.875,0.875,1.0,0.75,0.875,0.875,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.625,,0.375,0.75,0.75,0.625,1.0,0.75,0.75,0.75,1.0,1.0,0.875,0.875,0.875,1.0,0.75,1.0,0.75,0.875,0.75,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,0.75,1.0,0.875,0.75,0.875,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.75,0.625,,0.75,0.75,0.75,0.875,1.0,0.75,0.75,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.25,0.25,0.25,,0.625,0.375,0.6875,0.625,0.5,0.625,0.75,0.875,0.875,0.75,0.75,0.75,0.875,1.0,0.75,0.875,0.75,0.75,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,0.875,0.75,0.875,0.875,0.75,0.875,1.0,0.75,0.875,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.375,0.25,0.25,0.375,,0.5,0.75,0.5,0.75,0.5,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.75,1.0,0.75,0.875,1.0,1.0,0.875,0.75,1.0,1.0,0.75,1.0,0.75,0.75,0.75,0.875,0.75,0.75,1.0,0.75,0.75,0.75,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.25,0.375,0.25,0.625,0.5,,0.5,0.75,0.625,0.5,0.625,0.625,0.625,0.75,0.75,0.75,0.625,0.875,0.75,0.75,0.75,0.875,0.875,0.875,0.75,0.75,1.0,1.0,0.875,0.875,0.75,0.75,0.625,0.875,0.75,0.875,0.75,1.0,0.875,0.75,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,0.75,0.875,1.0,1.0,0.875,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.5,0.0,0.125,0.3125,0.25,0.5,,0.5,0.75,0.5,0.75,0.75,0.75,0.75,0.75,0.5,0.75,0.75,0.75,0.875,0.75,0.5,0.75,0.75,0.75,1.0,0.75,0.75,1.0,0.625,0.625,0.75,0.75,0.75,0.75,0.625,0.75,0.75,0.75,0.75,0.875,0.75,0.875,0.75,0.75,0.75,0.75,0.75,0.875,0.875,1.0,1.0,0.875,0.875,0.875,1.0,1.0,0.875,1.0,0.75,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.25,0.25,0.0,0.375,0.5,0.25,0.5,,0.625,0.625,0.625,0.625,0.625,0.875,0.875,0.625,0.625,0.625,0.625,0.75,0.75,0.875,0.75,0.75,0.625,0.875,0.875,0.875,0.75,0.75,0.625,0.75,0.5,0.75,0.75,0.75,0.75,1.0,0.75,0.75,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.75,0.875,0.875,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0
+LimiX (default),0.125,0.25,0.25,0.5,0.25,0.375,0.25,0.375,,0.5,0.5,0.625,0.625,0.625,0.625,0.625,0.75,0.625,0.75,0.5,0.5,0.625,0.75,0.75,0.75,0.625,0.625,0.75,0.625,0.75,0.75,0.625,0.625,0.875,0.75,0.875,0.75,0.75,0.875,0.75,0.625,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.75,0.75,0.875,0.625,0.875,1.0,1.0,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.125,0.25,0.25,0.375,0.5,0.5,0.5,0.375,0.5,,0.625,0.625,0.5,0.75,0.75,0.375,0.5,0.5,0.625,0.5,0.5,0.625,0.75,0.625,0.625,0.75,0.75,0.625,0.625,0.625,0.75,0.625,0.625,0.625,0.75,0.625,0.75,0.625,0.625,0.75,0.625,0.875,0.875,0.75,1.0,1.0,1.0,0.875,0.875,0.875,0.75,1.0,0.875,0.75,0.875,0.875,1.0,0.75,1.0,0.75,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.25,0.0,0.125,0.25,0.125,0.375,0.25,0.375,0.5,0.375,,0.375,0.375,0.5,0.5,0.375,0.625,0.625,0.75,0.625,0.75,0.5,0.75,0.75,0.625,0.625,0.75,0.875,0.75,0.75,0.625,0.75,0.5,0.75,0.75,0.625,0.75,0.75,0.75,0.625,0.875,0.625,0.875,0.75,0.875,0.75,0.75,0.75,0.875,1.0,0.875,0.75,0.75,0.75,0.875,1.0,0.875,0.75,1.0,0.75,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.25,0.0,0.125,0.125,0.125,0.375,0.25,0.375,0.375,0.375,0.625,,0.375,0.375,0.375,0.375,0.75,0.625,0.75,0.625,0.75,0.5,0.75,0.75,0.625,0.5,0.75,0.875,0.75,0.75,0.625,0.75,0.5,0.75,0.75,0.625,0.75,0.75,0.75,0.625,0.875,0.75,0.875,0.75,0.875,0.75,0.75,0.75,0.875,1.0,0.875,0.75,0.75,0.875,0.875,1.0,0.875,0.875,0.875,0.75,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.125,0.125,0.125,0.125,0.125,0.375,0.25,0.375,0.375,0.5,0.625,0.625,,0.625,0.625,0.375,0.625,0.625,0.625,0.375,0.5,0.5,0.875,0.75,0.625,0.5,0.75,0.75,0.625,0.75,0.75,0.5,0.625,0.5,0.875,0.5,0.75,0.625,0.625,0.75,0.875,0.75,0.75,0.75,0.75,0.75,0.75,0.625,0.75,0.875,0.75,0.875,0.875,0.625,0.75,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.375,0.25,0.5,0.625,0.375,,0.5,0.375,0.5,0.5,0.5,0.375,0.5,0.75,0.5,0.625,0.5,0.625,0.625,0.5,0.625,0.5,0.75,0.625,0.625,0.625,0.75,0.625,0.875,0.625,0.625,0.875,0.625,0.875,0.875,0.625,0.875,1.0,1.0,0.875,0.75,1.0,0.875,0.75,0.875,0.75,0.875,0.75,1.0,0.75,1.0,0.75,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.375,0.25,0.5,0.625,0.375,0.5,,0.375,0.5,0.5,0.5,0.375,0.5,0.75,0.5,0.5,0.5,0.5,0.625,0.5,0.5,0.5,0.75,0.625,0.625,0.625,0.75,0.625,0.875,0.625,0.625,0.875,0.625,0.875,0.875,0.625,0.875,1.0,1.0,0.875,0.75,1.0,0.875,0.75,0.875,0.75,0.875,0.75,1.0,0.75,0.875,0.75,1.0,1.0,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.25,0.0,0.0,0.25,0.125,0.25,0.5,0.375,0.375,0.625,0.625,0.625,0.625,0.625,0.625,,0.625,0.625,0.625,0.625,0.625,0.5,0.75,0.625,0.75,0.625,0.625,0.625,0.75,0.5,0.625,0.625,0.5,0.625,0.75,0.5,0.75,0.5,0.625,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.625,0.75,0.75,0.75,0.75,0.875,0.75,0.75,0.875,0.875,0.75,0.75,0.75,0.75,0.875,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.125,0.25,0.125,0.125,0.25,0.375,0.25,0.375,0.25,0.5,0.375,0.25,0.375,0.5,0.5,0.375,,0.625,0.5,0.5,0.625,0.625,0.625,0.625,0.5,0.375,0.5,0.625,0.375,0.75,0.875,0.75,0.625,0.5,0.875,0.625,0.625,0.625,0.625,0.875,0.625,0.875,0.875,0.625,0.875,0.625,0.625,0.625,0.75,0.75,0.625,0.75,0.75,0.75,0.875,1.0,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.25,0.0,0.125,0.0,0.125,0.125,0.25,0.375,0.375,0.5,0.375,0.375,0.375,0.5,0.5,0.375,0.375,,0.5,0.375,0.5,0.375,0.625,0.75,0.5,0.5,0.625,0.75,0.625,0.875,0.75,0.625,0.5,0.625,0.75,0.625,0.625,0.625,0.75,0.75,0.875,0.75,0.75,0.75,0.75,0.625,0.75,0.625,0.75,0.875,0.75,0.75,0.75,0.5,0.75,1.0,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.375,0.25,0.375,0.25,0.25,0.375,0.5,0.5,0.375,0.5,0.5,,0.5,0.625,0.5,0.5,0.625,0.625,0.375,0.5,0.625,0.375,0.75,0.875,0.75,0.625,0.5,0.875,0.625,0.625,0.625,0.625,0.875,0.625,0.875,0.75,0.625,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.625,0.875,0.75,0.75,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Mitra (default),0.0,0.125,0.0,0.125,0.0,0.25,0.125,0.25,0.5,0.5,0.375,0.375,0.625,0.625,0.625,0.375,0.5,0.625,0.5,,0.5,0.5,0.75,0.75,0.75,0.625,0.75,0.75,0.625,0.625,0.625,0.625,0.375,0.625,0.75,0.625,0.75,0.5625,0.75,0.75,0.625,0.625,0.625,0.75,0.75,0.75,0.75,0.625,0.75,0.75,0.75,0.75,0.875,0.625,0.75,0.875,0.75,0.875,0.75,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.5,0.5,0.25,0.25,0.5,0.5,0.5,0.375,0.375,0.5,0.375,0.5,,0.375,0.625,0.625,0.5,0.625,0.75,0.625,0.625,0.625,0.625,0.875,0.5,0.5,0.5,0.625,0.75,0.625,0.625,0.75,0.75,0.625,0.75,0.75,0.75,0.625,0.75,0.625,0.75,0.75,0.625,0.75,0.75,0.875,0.625,0.75,0.75,0.875,0.875,0.875,0.75,0.875,1.0,0.875,0.875,1.0,1.0,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0
+TabDPT-Turbo (default),0.125,0.125,0.125,0.25,0.125,0.125,0.5,0.125,0.375,0.375,0.5,0.5,0.5,0.25,0.25,0.5,0.375,0.625,0.5,0.5,0.625,,0.625,0.625,0.375,0.5,0.75,0.625,0.5,0.625,0.5,0.625,0.5,0.625,0.5,0.625,0.75,0.625,0.625,0.5,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.875,1.0,0.625,0.75,0.875,0.75,0.75,0.875,0.75,0.75,0.875,0.875,0.875,1.0,0.75,0.75,1.0,1.0,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.125,0.0,0.0,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.125,0.5,0.5,0.25,0.375,0.375,0.5,0.25,0.375,0.375,,0.625,0.5,0.5,0.75,0.75,0.625,0.625,0.625,0.5,0.5,0.625,0.75,0.5,0.75,0.75,0.625,0.625,0.625,0.625,0.75,0.75,0.75,0.875,0.875,0.625,0.75,1.0,0.75,0.625,0.875,0.625,0.875,0.75,0.875,0.625,1.0,0.75,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.125,0.0,0.0,0.125,0.25,0.25,0.25,0.375,0.25,0.25,0.25,0.375,0.5,0.375,0.375,0.25,0.375,0.25,0.375,0.375,0.375,,0.5,0.625,0.625,0.75,0.625,0.625,0.625,0.5,0.5,0.625,0.625,0.5,0.75,0.75,0.625,0.625,0.625,0.625,0.75,0.875,0.75,0.875,0.875,0.625,0.75,1.0,0.75,0.625,0.875,0.625,0.875,0.75,0.875,0.625,1.0,0.75,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.375,0.25,0.375,0.375,0.375,0.375,0.5,0.5,0.25,0.5,0.5,0.375,0.25,0.5,0.625,0.5,0.5,,0.375,0.5,0.625,0.375,0.5,0.5,0.625,0.375,0.625,0.625,0.625,0.5,0.625,0.625,0.625,0.5,0.5,0.625,0.75,0.625,0.75,0.75,0.75,0.625,0.625,0.75,0.625,0.625,0.625,0.75,0.75,0.75,1.0,0.875,1.0,0.75,1.0,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (default),0.125,0.0,0.0,0.125,0.25,0.25,0.0,0.125,0.375,0.25,0.375,0.5,0.5,0.375,0.5,0.375,0.625,0.5,0.625,0.375,0.375,0.5,0.5,0.375,0.625,,0.375,0.375,0.5,0.5,0.625,0.625,0.375,0.625,0.75,0.625,0.625,0.5625,0.75,0.625,0.375,0.625,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.625,0.625,0.875,0.625,0.75,0.875,0.75,0.75,0.5,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.75,0.875,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.125,0.375,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.5,0.375,0.5,0.25,0.25,0.25,0.25,0.375,0.5,0.625,,0.5,0.625,0.75,0.5,0.5,0.5,0.75,0.5,0.625,0.5,0.5,0.75,0.5,0.625,0.625,0.625,0.75,0.875,0.75,0.75,0.625,0.875,0.75,0.625,0.75,0.625,0.625,0.875,0.875,0.75,0.75,1.0,0.875,1.0,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.125,0.25,0.375,0.125,0.125,0.25,0.5,0.5,0.375,0.375,0.25,0.375,0.25,0.375,0.375,0.25,0.25,0.375,0.625,0.5,,0.5,0.6875,0.5,0.5,0.5,0.75,0.5,0.625,0.625,0.75,0.75,0.625,0.625,0.5,0.625,0.75,0.75,0.625,0.625,0.625,0.75,0.75,0.625,0.75,0.75,0.625,0.75,0.75,0.875,0.75,1.0,0.875,0.875,0.875,1.0,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (tuned),0.0,0.0,0.0,0.0,0.25,0.125,0.0,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.5,0.25,0.625,0.375,0.625,0.375,0.375,0.5,0.375,0.375,0.625,0.5,0.375,0.5,,0.5,0.625,0.5,0.375,0.625,0.75,0.5,0.5,0.5,0.75,0.625,0.5,0.625,0.75,0.625,0.75,0.625,0.625,0.75,0.75,0.75,0.75,0.625,0.75,0.625,0.75,0.875,0.75,0.75,0.5,0.75,0.75,0.875,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.125,0.0,0.125,0.0,0.0,0.125,0.375,0.25,0.25,0.375,0.25,0.25,0.25,0.5,0.5,0.5,0.25,0.125,0.25,0.375,0.375,0.375,0.375,0.375,0.5,0.5,0.25,0.3125,0.5,,0.375,0.375,0.625,0.75,0.5,0.625,0.625,0.625,0.75,0.5,0.75,0.625,0.5,0.75,0.375,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.75,0.625,0.75,0.75,1.0,0.875,1.0,0.875,1.0,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.125,0.125,0.125,0.125,0.25,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.25,0.25,0.25,0.375,0.125,0.25,0.125,0.375,0.375,0.5,0.375,0.375,0.5,0.375,0.5,0.5,0.375,0.625,,0.375,0.625,0.625,0.25,0.625,0.75,0.5,0.625,0.5,0.75,0.875,0.5,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.625,0.5,0.75,0.75,0.625,0.75,0.75,0.875,0.75,1.0,1.0,0.875,0.75,0.875,0.75,0.875,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.25,0.25,0.5,0.375,0.375,0.375,0.25,0.375,0.25,0.375,0.125,0.375,0.5,0.5,0.375,0.375,0.5,0.5,0.5,0.625,0.625,,0.5,0.375,0.5,0.375,0.625,0.5,0.5,0.625,0.625,0.625,0.5,0.625,0.625,0.625,0.625,0.625,0.5,0.625,0.625,0.625,0.625,0.875,0.625,0.75,0.75,0.875,0.625,0.875,0.75,0.875,0.75,0.875,0.875,1.0,1.0,0.875,0.875,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0
+SAP-RPT-OSS (default),0.25,0.0,0.125,0.125,0.25,0.375,0.25,0.5,0.375,0.375,0.5,0.5,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.625,0.5,0.5,0.5,0.5,0.625,0.625,0.5,0.5,0.625,0.375,0.375,0.5,,0.5,0.625,0.5,0.5,0.625,0.625,0.625,0.5,0.625,0.625,0.5,0.625,0.5,0.5,0.5,0.625,0.5,0.5,0.625,0.625,0.5,0.625,0.75,0.75,0.625,0.625,0.625,0.625,0.75,0.625,0.75,0.75,0.625,0.75,0.875,0.875,0.875,0.875,1.0,0.75,0.875,0.875,0.875,0.875,1.0
+ExtraTrees (tuned + ensembled),0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.375,0.25,0.25,0.5,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.5,0.375,0.375,0.375,0.375,0.375,0.25,0.25,0.375,0.25,0.375,0.625,0.5,,0.5,0.75,0.5,0.375,0.375,0.625,0.5,0.5,0.5,0.8125,0.5,0.625,0.625,0.75,0.625,0.5,0.75,0.5,0.75,0.75,0.75,0.625,0.75,0.875,0.625,1.0,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.75,0.75,0.75,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0
+TabM (default),0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.125,0.25,0.125,0.25,0.5,0.5,0.25,0.375,0.375,0.25,0.5,0.5,0.25,0.5,0.75,0.5,0.375,0.5,,0.5,0.625,0.5,0.5,0.75,0.625,0.75,0.625,0.5,0.625,0.625,0.625,0.625,0.625,0.75,0.625,0.625,0.5,0.625,0.5,0.625,0.875,0.75,0.75,0.75,0.75,1.0,0.75,0.75,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.125,0.125,0.125,0.25,0.125,0.375,0.25,0.125,0.375,0.375,0.375,0.5,0.375,0.375,0.5,0.375,0.375,0.375,0.375,0.375,0.375,0.5,0.5,0.375,0.375,0.375,0.375,0.5,0.375,0.375,0.625,0.5,0.25,0.5,,0.5,0.5,0.25,0.5,0.5,0.5,0.5,0.625,0.5,0.75,0.625,0.875,0.5,0.625,0.625,0.5,0.5,0.625,0.625,0.625,0.625,0.75,0.625,0.875,0.75,0.875,0.875,0.875,0.875,0.875,1.0,0.75,0.75,0.625,0.875,0.875,1.0,0.875,1.0,0.75,1.0,1.0
+CatBoost (default),0.0,0.0,0.125,0.0,0.0,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.125,0.125,0.25,0.375,0.375,0.375,0.25,0.25,0.25,0.25,0.25,0.5,0.375,0.5,0.375,0.5,0.375,0.25,0.375,0.5,0.5,0.375,0.5,,0.5,0.625,0.25,0.75,0.375,0.625,0.5,0.625,0.625,0.625,0.625,0.625,0.5,0.75,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.75,1.0,1.0,1.0,0.875,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabICL (default),0.125,0.125,0.0,0.25,0.25,0.0,0.25,0.0,0.25,0.375,0.25,0.25,0.375,0.375,0.375,0.5,0.375,0.375,0.375,0.4375,0.375,0.375,0.25,0.25,0.375,0.4375,0.5,0.25,0.5,0.375,0.5,0.5,0.375,0.625,0.5,0.5,0.5,,0.5,0.625,0.625,0.5,0.5,0.75,0.5,0.5,0.5,0.625,0.75,0.625,0.75,0.5,0.75,0.75,0.75,0.625,0.625,0.625,0.625,0.875,0.75,0.75,0.8125,0.75,0.75,0.875,0.875,0.75,0.875,0.75,1.0,0.75,0.875,0.875,1.0,1.0,1.0,1.0
+xRFM (tuned),0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.375,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.25,0.25,0.25,0.25,0.375,0.5,0.375,0.625,0.5,0.75,0.375,0.5,,0.5,0.5,0.375,0.375,0.75,0.375,0.75,0.5,0.75,0.5,0.5,0.625,0.5,0.5,0.625,0.625,0.75,0.625,1.0,0.625,1.0,0.75,1.0,0.75,0.875,0.875,1.0,1.0,0.75,0.875,0.75,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.25,0.125,0.125,0.25,0.125,0.25,0.125,0.25,0.25,0.5,0.375,0.375,0.375,0.375,0.5,0.375,0.375,0.5,0.5,0.375,0.375,0.375,0.25,0.5,0.75,0.375,0.5,,0.625,0.875,0.625,0.375,0.5,0.625,0.625,0.625,0.5,0.5,0.5,0.625,0.5,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.875,0.875,0.75,0.75,0.75,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.25,0.375,0.375,0.125,0.125,0.125,0.375,0.375,0.375,0.375,0.125,0.375,0.375,0.25,0.25,0.375,0.375,0.5,0.625,0.375,0.375,0.5,0.25,0.25,0.375,0.5,0.5,0.375,0.5,0.25,0.375,0.5,0.375,,0.375,0.625,0.75,0.625,0.5,0.5,0.625,0.625,0.5,0.625,0.75,0.625,0.625,0.625,0.625,0.75,0.625,0.875,0.875,1.0,0.875,0.875,0.75,0.75,0.875,1.0,0.75,0.75,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.125,0.375,0.25,0.25,0.125,0.125,0.25,0.125,0.25,0.125,0.375,0.375,0.375,0.375,0.375,0.5,0.375,0.375,0.5,0.375,0.375,0.125,0.375,0.375,0.5,0.25,0.5,0.625,0.5,0.625,0.125,0.625,,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.375,0.5,0.625,0.375,0.625,0.75,0.625,0.75,0.75,0.875,0.75,1.0,0.875,0.75,0.75,0.75,0.75,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.375,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.25,0.5,0.5,0.5,0.375,0.5,0.375,0.5,0.375,0.5,0.625,0.375,0.375,0.5,,0.5,0.625,0.5,0.5,0.5,0.75,0.625,0.5,0.625,0.625,0.75,0.5,0.75,0.75,0.75,0.625,0.75,0.75,1.0,0.75,0.75,0.875,0.75,0.875,0.875,0.875,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.25,0.25,0.25,0.375,0.375,0.25,0.375,0.25,0.375,0.25,0.25,0.25,0.25,0.125,0.25,0.375,0.25,0.25,0.375,0.25,0.375,0.375,0.5,0.1875,0.5,0.375,0.5,0.25,0.25,0.625,0.25,0.5,0.5,,0.375,0.5,0.5,0.5,0.5,0.5,0.625,0.5,0.75,0.625,0.75,0.5,0.625,0.625,0.625,0.875,0.75,0.875,0.875,0.875,0.875,1.0,0.875,0.75,0.625,0.625,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0
+EBM (default),0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.0,0.125,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.125,0.25,0.25,0.625,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.5,0.625,0.5,0.375,0.5,0.375,0.625,,0.375,0.5,0.625,0.5,0.375,0.5,0.625,0.375,0.625,0.75,0.75,0.75,0.75,0.75,0.75,1.0,0.875,0.625,0.75,0.75,0.75,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.0,0.25,0.25,0.25,0.0,0.0,0.25,0.375,0.375,0.375,0.25,0.375,0.25,0.125,0.125,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.5,0.375,0.375,0.25,0.375,0.5,0.25,0.375,0.5,0.5,0.5,0.5,0.625,,0.625,0.625,0.625,0.5,0.625,0.5,0.625,0.75,0.75,0.75,0.625,0.75,0.5,0.625,0.875,0.75,0.75,0.875,0.875,0.875,0.75,0.875,0.75,0.75,0.875,1.0,0.875,1.0,0.875,1.0,0.875,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.0,0.25,0.25,0.25,0.0,0.0,0.25,0.375,0.25,0.25,0.25,0.25,0.25,0.125,0.125,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.5,0.375,0.375,0.375,0.375,0.5,0.5,0.375,0.5,0.5,0.5,0.5,0.5,0.375,,0.625,0.625,0.375,0.625,0.5,0.5,0.75,0.875,0.75,0.625,0.75,0.625,0.625,0.875,0.75,0.75,0.875,0.875,0.75,0.75,1.0,0.875,1.0,0.875,1.0,0.875,0.875,0.875,1.0,0.875,1.0
+RandomForest (tuned),0.0,0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.125,0.25,0.25,0.375,0.125,0.125,0.375,0.375,0.375,0.25,0.375,0.375,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.25,0.375,0.375,0.375,0.5,0.25,0.375,0.125,0.375,0.375,0.25,0.375,0.375,0.5,0.5,0.5,0.375,0.375,0.375,,0.5,0.5,0.625,0.5,0.5,0.625,0.625,0.625,0.625,0.625,0.375,0.75,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.75,0.625,0.875,0.875,0.875,0.875,1.0,0.75,1.0,1.0
+TorchMLP (tuned),0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.125,0.25,0.25,0.375,0.375,0.5,0.375,0.375,0.375,0.5,0.375,0.25,0.5,0.5,0.375,0.5,0.25,0.5,0.5,0.375,0.375,0.5,,0.375,0.375,0.5,0.625,0.625,0.625,0.625,0.625,0.75,0.75,0.875,0.75,0.875,0.625,0.75,0.75,0.75,0.75,0.875,0.75,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0
+XGBoost (default),0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.0,0.125,0.0,0.0,0.25,0.25,0.125,0.25,0.25,0.25,0.125,0.0,0.0,0.375,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.5,0.5,0.25,0.375,0.5,0.375,0.5,0.5,0.5,0.625,0.375,0.5,0.625,0.5,0.625,0.5,0.625,,0.625,0.5,0.625,0.5,0.75,0.625,0.625,0.5,0.75,0.75,1.0,0.875,0.875,0.75,0.75,0.875,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.25,0.25,0.125,0.125,0.25,0.125,0.125,0.25,0.375,0.25,0.25,0.25,0.375,0.0,0.25,0.25,0.25,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.5,0.25,0.375,0.375,0.25,0.25,0.375,0.5,0.375,0.5,0.5,0.375,0.5,0.375,0.375,0.375,0.625,0.375,,0.625,0.75,0.625,0.625,0.75,0.75,0.5,0.625,0.75,0.75,0.75,0.875,0.75,0.75,0.875,0.75,0.875,0.75,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0
+FastaiMLP (tuned + ensembled),0.125,0.0,0.125,0.125,0.0,0.125,0.0,0.125,0.25,0.0,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.5,0.5,0.375,0.25,0.375,0.375,0.5,0.375,0.5,0.5,0.5,0.5,0.5,0.375,,0.5,0.375,0.5,0.5,0.625,0.625,0.625,0.625,0.75,0.5,0.625,0.75,0.75,0.625,0.75,1.0,0.875,0.875,0.75,0.875,0.875,1.0,0.875,1.0,0.875,1.0
+LightGBM (default),0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.25,0.25,0.125,0.125,0.375,0.125,0.375,0.25,0.25,0.375,0.5,0.375,0.375,0.25,0.5,0.5,0.375,0.25,0.5,0.5,0.375,0.625,0.375,0.25,0.625,0.375,0.5,0.5,0.375,0.375,0.25,0.5,,0.375,0.625,0.625,0.625,0.625,0.625,0.75,0.75,0.75,0.5,0.875,0.875,0.875,0.75,1.0,0.75,0.75,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.375,0.25,0.25,0.125,0.375,0.25,0.25,0.25,0.25,0.5,0.25,0.375,0.125,0.125,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.25,0.125,0.5,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.375,0.375,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.5,0.375,0.625,0.625,,0.5,0.625,0.75,0.75,0.5,0.875,0.75,0.625,0.625,0.75,0.75,0.75,1.0,0.75,0.5,0.625,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0
+ChimeraBoost (default),0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.25,0.375,0.25,0.125,0.125,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.375,0.375,0.25,0.5,0.375,0.25,0.25,0.375,0.25,0.375,0.25,0.5,0.25,0.25,0.25,0.125,0.375,0.375,0.25,0.375,0.5,0.375,0.5,,0.625,0.5,0.5,0.625,0.625,0.75,0.625,0.5,0.625,0.875,0.625,0.625,0.75,0.75,0.75,0.875,1.0,1.0,0.875,0.875,1.0,0.875,1.0
+RealMLP (default),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.25,0.0,0.125,0.0,0.0,0.0,0.25,0.25,0.125,0.0,0.0,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.125,0.125,0.25,0.125,0.375,0.375,0.25,0.25,0.375,0.375,0.375,0.25,0.375,0.25,0.25,0.375,0.375,0.25,0.5,0.25,0.25,0.25,0.375,0.375,0.375,0.25,0.5,0.375,0.375,0.375,,0.5,0.625,0.5,0.75,0.625,0.625,0.625,0.625,0.75,0.625,0.75,0.75,0.75,0.75,0.875,1.0,0.875,1.0,0.875,1.0,0.875,1.0
+FastaiMLP (tuned),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.125,0.375,0.25,0.375,0.375,0.25,0.25,0.25,0.25,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.375,0.375,0.25,0.5,0.5,,0.625,0.625,0.625,0.75,0.5,0.5,0.625,0.625,0.625,0.75,1.0,0.75,0.75,0.875,0.875,0.875,0.875,0.875,1.0,0.875,1.0
+xRFM (default),0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.25,0.25,0.25,0.125,0.125,0.125,0.125,0.125,0.25,0.375,0.375,0.0,0.25,0.25,0.25,0.25,0.25,0.25,0.125,0.375,0.125,0.25,0.25,0.25,0.375,0.0,0.25,0.375,0.25,0.25,0.375,0.25,0.25,0.25,0.375,0.25,0.5,0.5,0.375,0.375,0.25,0.5,0.375,0.375,,0.5,0.625,0.625,0.625,0.75,0.75,0.75,0.75,0.625,0.625,0.625,0.625,0.75,0.875,0.875,0.75,0.75,0.875,0.75,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.125,0.25,0.125,0.125,0.125,0.25,0.125,0.25,0.0,0.0,0.125,0.5,0.0,0.0,0.5,0.0,0.125,0.375,0.375,0.375,0.25,0.375,0.25,0.375,0.375,0.25,0.125,0.125,0.375,0.375,0.25,0.5,0.375,0.625,0.25,0.25,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.5,,0.625,0.5,0.75,0.875,0.625,0.625,0.75,0.75,0.75,0.875,0.875,0.875,0.625,0.75,0.875,0.875,0.875,1.0,1.0
+KNN (tuned + ensembled),0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.0,0.25,0.125,0.125,0.25,0.125,0.25,0.125,0.375,0.0,0.25,0.125,0.25,0.125,0.0,0.25,0.125,0.25,0.25,0.125,0.25,0.375,0.375,0.25,0.125,0.25,0.25,0.375,0.25,0.125,0.375,0.25,0.375,0.375,0.375,,0.375,0.375,0.375,0.375,0.375,0.375,0.5,0.5,0.375,0.375,0.75,0.375,1.0,0.625,0.75,0.625,0.875,1.0
+iLTM (default),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.25,0.125,0.125,0.125,0.25,0.25,0.125,0.0,0.0,0.25,0.25,0.0,0.125,0.25,0.0,0.0,0.25,0.375,0.125,0.25,0.25,0.0,0.25,0.25,0.125,0.0,0.0,0.25,0.25,0.0,0.125,0.125,0.25,0.25,0.0,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.375,0.5,0.625,,0.625,0.625,0.75,0.625,0.5,0.625,0.625,0.625,0.625,0.875,0.875,0.75,0.75,0.875,0.75,1.0,1.0
+TorchMLP (default),0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.25,0.125,0.0,0.125,0.0,0.25,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.25,0.5,0.25,0.375,0.375,0.375,0.5,0.375,0.25,0.625,0.375,,0.375,0.75,0.75,0.625,0.625,0.75,0.625,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.25,0.125,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.25,0.0,0.125,0.25,0.375,0.125,0.25,0.125,0.0,0.1875,0.25,0.25,0.125,0.25,0.25,0.125,0.375,0.25,0.25,0.125,0.375,0.125,0.125,0.375,0.5,0.375,0.5,0.375,0.5,0.25,0.125,0.625,0.375,0.625,,0.625,0.625,0.375,0.625,0.5,0.75,0.75,0.75,0.75,0.875,0.625,0.75,0.75,0.75,1.0
+TabSTAR (tuned),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.25,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.125,0.25,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.125,0.25,0.375,0.375,0.375,0.25,0.375,0.625,0.25,0.25,0.375,,0.4375,0.375,0.5,0.625,0.75,0.75,0.75,1.0,0.875,0.625,0.75,0.875,0.75,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.125,0.125,0.25,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.25,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.125,0.25,0.125,0.25,0.375,0.25,0.375,0.625,0.375,0.25,0.375,0.5625,,0.5,0.5,0.5,0.75,0.75,0.75,1.0,0.875,0.625,0.75,0.875,0.75,1.0
+OrionMSP (default),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.25,0.0,0.375,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.25,0.25,0.0,0.25,0.125,0.25,0.125,0.25,0.125,0.125,0.375,0.125,0.25,0.375,0.375,0.375,0.25,0.25,0.625,0.5,0.375,0.625,0.625,0.5,,0.625,0.625,0.625,0.625,0.75,0.875,0.875,0.75,0.75,0.75,0.75,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.0,0.125,0.0,0.125,0.0,0.25,0.0,0.125,0.0,0.125,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.0,0.375,0.25,0.25,0.375,0.25,0.5,0.375,0.375,0.375,0.5,0.5,0.375,,0.75,0.5,0.5,0.625,0.625,0.875,0.625,0.875,0.75,0.875,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.0,0.25,0.25,0.0,0.25,0.125,0.125,0.25,0.125,0.125,0.0,0.25,0.125,0.125,0.125,0.0,0.0,0.25,0.25,0.25,0.0,0.375,0.25,0.5,0.375,0.25,0.5,0.375,0.5,0.375,0.25,,0.625,0.75,0.5,0.75,0.75,0.625,0.5,0.875,0.75,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.25,0.0,0.25,0.125,0.125,0.125,0.0,0.25,0.125,0.125,0.375,0.125,0.25,0.125,0.25,0.25,0.125,0.25,0.125,0.25,0.5,0.25,0.25,0.25,0.375,0.125,0.625,0.375,0.375,0.25,0.25,0.25,0.375,0.5,0.375,,0.6875,0.5,0.5,0.875,0.625,0.5,1.0,0.75,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.25,0.125,0.25,0.0,0.375,0.0,0.25,0.25,0.0,0.125,0.0,0.0,0.375,0.125,0.25,0.0,0.375,0.125,0.125,0.25,0.125,0.25,0.375,0.25,0.25,0.25,0.375,0.125,0.625,0.375,0.25,0.25,0.25,0.25,0.375,0.5,0.25,0.3125,,0.5,0.5,0.875,0.625,0.5,0.875,0.75,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.25,0.0,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.125,0.125,0.25,0.25,0.25,0.25,0.375,0.5,0.5,0.5,,0.25,0.625,0.625,0.875,0.625,1.0,0.875
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.0,0.25,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.0,0.125,0.125,0.375,0.625,0.125,0.125,0.25,0.0,0.0,0.125,0.375,0.25,0.5,0.5,0.75,,0.75,0.375,0.75,0.75,0.75,1.0
+KNN (tuned),0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.25,0.0,0.25,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.375,0.25,,0.25,0.5,0.25,0.5,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.25,0.125,0.375,0.25,0.125,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.625,0.75,,0.375,0.625,0.5,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.125,0.125,0.25,0.25,0.25,0.25,0.125,0.5,0.5,0.5,0.125,0.25,0.5,0.625,,0.625,0.6875,0.875
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.25,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.375,0.25,0.125,0.25,0.125,0.125,0.25,0.25,0.125,0.0,0.125,0.375,0.25,0.75,0.375,0.375,,0.625,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.25,0.0,0.125,0.0,0.125,0.25,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.0,0.25,0.5,0.5,0.3125,0.375,,0.75
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.25,
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
index 7bdd5a39ce1bebb4e8c3c8d1584ca52acfa61525..65f94d8cc05cf9d2c30db5f68509311c2af97e0d 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Multiclass | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..4d2c03a1a1ae4efa89bb03b42e2558e5c3297531
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":34554.0,"score":1.0,"rank":1.0,"hrank":1.0,"imp":0.0,"train_time":3.94,"predict_time":3.192,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":33174.0,"score":0.852,"rank":2.0,"hrank":2.0,"imp":4.445,"train_time":259.67,"predict_time":0.61,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":31548.0,"score":0.85,"rank":3.0,"hrank":3.0,"imp":4.498,"train_time":134.59,"predict_time":8.308,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":29818.0,"score":0.849,"rank":4.0,"hrank":4.0,"imp":4.526,"train_time":0.93,"predict_time":0.273,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":4,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":28012.0,"score":0.791,"rank":5.0,"hrank":5.0,"imp":6.149,"train_time":276.98,"predict_time":4.897,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":26991.0,"score":0.746,"rank":6.0,"hrank":6.0,"imp":7.385,"train_time":134.59,"predict_time":2.818,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":6,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":25644.0,"score":0.67,"rank":7.0,"hrank":7.0,"imp":9.385,"train_time":1.84,"predict_time":0.516,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":7,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":24744.0,"score":0.502,"rank":8.0,"hrank":8.0,"imp":13.523,"train_time":35.65,"predict_time":17.704,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":8,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":24030.0,"score":0.474,"rank":9.0,"hrank":9.0,"imp":14.192,"train_time":6601.46,"predict_time":0.107,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":23243.0,"score":0.473,"rank":10.0,"hrank":10.0,"imp":14.205,"train_time":0.48,"predict_time":0.143,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":10,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":22438.0,"score":0.47,"rank":11.0,"hrank":11.0,"imp":14.275,"train_time":61.47,"predict_time":0.154,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":21690.0,"score":0.467,"rank":12.0,"hrank":12.0,"imp":14.338,"train_time":157.4,"predict_time":4.579,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":21043.0,"score":0.466,"rank":13.0,"hrank":13.0,"imp":14.366,"train_time":157.4,"predict_time":0.401,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":20253.0,"score":0.464,"rank":14.0,"hrank":14.0,"imp":14.421,"train_time":0.82,"predict_time":0.278,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":14,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":19480.0,"score":0.452,"rank":15.0,"hrank":15.0,"imp":14.687,"train_time":61.47,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":15,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":18724.0,"score":0.446,"rank":16.0,"hrank":16.0,"imp":14.817,"train_time":6601.46,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":16,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":18005.0,"score":0.438,"rank":17.0,"hrank":17.0,"imp":15.012,"train_time":1261.24,"predict_time":0.234,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":17,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":17293.0,"score":0.43,"rank":18.0,"hrank":18.0,"imp":15.197,"train_time":1261.24,"predict_time":0.03,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":18,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":16562.0,"score":0.42,"rank":19.0,"hrank":19.0,"imp":15.424,"train_time":456.33,"predict_time":1.17,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":19,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":15859.0,"score":0.408,"rank":20.0,"hrank":20.0,"imp":15.693,"train_time":456.33,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":20,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":15134.0,"score":0.408,"rank":21.0,"hrank":21.0,"imp":15.693,"train_time":0.32,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":21,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":14414.0,"score":0.377,"rank":22.0,"hrank":22.0,"imp":16.367,"train_time":1229.42,"predict_time":21.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":22,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":13717.0,"score":0.368,"rank":23.0,"hrank":23.0,"imp":16.568,"train_time":66.23,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":23,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":13046.0,"score":0.351,"rank":24.0,"hrank":24.0,"imp":16.943,"train_time":66.23,"predict_time":0.196,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":24,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":12359.0,"score":0.335,"rank":25.0,"hrank":25.0,"imp":17.279,"train_time":4378.88,"predict_time":25.595,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":25,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":11659.0,"score":0.332,"rank":26.0,"hrank":26.0,"imp":17.34,"train_time":3.59,"predict_time":0.022,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":26,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":10927.0,"score":0.306,"rank":27.0,"hrank":27.0,"imp":17.893,"train_time":13.87,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":27,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":10208.0,"score":0.306,"rank":28.0,"hrank":28.0,"imp":17.912,"train_time":4378.88,"predict_time":2.216,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":28,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":9576.0,"score":0.279,"rank":29.0,"hrank":29.0,"imp":18.478,"train_time":1.48,"predict_time":0.535,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":29,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":8935.0,"score":0.222,"rank":30.0,"hrank":30.0,"imp":19.642,"train_time":2.79,"predict_time":1.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":30,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":8262.0,"score":0.192,"rank":31.0,"hrank":31.0,"imp":20.252,"train_time":28.44,"predict_time":0.174,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":7604.0,"score":0.172,"rank":32.0,"hrank":32.0,"imp":20.642,"train_time":1229.42,"predict_time":7.785,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":32,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":7143.0,"score":0.172,"rank":33.0,"hrank":33.0,"imp":20.642,"train_time":52.48,"predict_time":7.785,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":33,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":6522.0,"score":0.141,"rank":34.0,"hrank":34.0,"imp":21.248,"train_time":28.44,"predict_time":0.009,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":5793.0,"score":0.087,"rank":35.0,"hrank":35.0,"imp":22.291,"train_time":0.23,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":35,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":5284.0,"score":0.025,"rank":36.0,"hrank":36.0,"imp":23.459,"train_time":0.58,"predict_time":0.199,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":36,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":4593.0,"score":0.016,"rank":37.0,"hrank":37.0,"imp":23.624,"train_time":4391.01,"predict_time":275.95,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":37,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":4011.0,"score":0.0,"rank":38.0,"hrank":38.0,"imp":24.192,"train_time":2341.94,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":38,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":3320.0,"score":0.0,"rank":39.0,"hrank":39.0,"imp":24.433,"train_time":1142.35,"predict_time":0.542,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":39,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":2811.0,"score":0.0,"rank":40.0,"hrank":40.0,"imp":24.458,"train_time":1142.35,"predict_time":0.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":40,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":2082.0,"score":0.0,"rank":41.0,"hrank":41.0,"imp":24.51,"train_time":14.53,"predict_time":1.747,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":41,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1461.0,"score":0.0,"rank":42.0,"hrank":42.0,"imp":24.577,"train_time":2341.94,"predict_time":0.172,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":42,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":43.0,"hrank":43.0,"imp":27.419,"train_time":0.18,"predict_time":0.007,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":43,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":342.0,"score":0.0,"rank":44.0,"hrank":44.0,"imp":27.476,"train_time":4391.01,"predict_time":38.524,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":44,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-331.0,"score":0.0,"rank":45.0,"hrank":45.0,"imp":27.597,"train_time":41.21,"predict_time":39.405,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":45,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-972.0,"score":0.0,"rank":46.0,"hrank":46.0,"imp":27.983,"train_time":1322.5,"predict_time":0.545,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":46,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-1604.0,"score":0.0,"rank":47.0,"hrank":47.0,"imp":28.071,"train_time":9.89,"predict_time":9.237,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":47,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-2323.0,"score":0.0,"rank":48.0,"hrank":48.0,"imp":28.548,"train_time":1322.5,"predict_time":0.055,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":48,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":-3055.0,"score":0.0,"rank":49.0,"hrank":49.0,"imp":29.747,"train_time":0.21,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":49,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-3755.0,"score":0.0,"rank":50.0,"hrank":50.0,"imp":32.399,"train_time":2009.14,"predict_time":3.767,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":50,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-4442.0,"score":0.0,"rank":51.0,"hrank":51.0,"imp":33.951,"train_time":2009.14,"predict_time":0.974,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":51,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-5113.0,"score":0.0,"rank":52.0,"hrank":52.0,"imp":34.755,"train_time":4.05,"predict_time":0.074,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":52,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-5810.0,"score":0.0,"rank":53.0,"hrank":53.0,"imp":34.88,"train_time":3.34,"predict_time":0.028,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":53,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-6530.0,"score":0.0,"rank":54.0,"hrank":54.0,"imp":36.901,"train_time":3349.22,"predict_time":1.469,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":54,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-7255.0,"score":0.0,"rank":55.0,"hrank":55.0,"imp":36.958,"train_time":3349.22,"predict_time":6.037,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":55,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-7958.0,"score":0.0,"rank":56.0,"hrank":56.0,"imp":37.098,"train_time":66.06,"predict_time":1.783,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":56,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-8689.0,"score":0.0,"rank":57.0,"hrank":57.0,"imp":39.799,"train_time":16.78,"predict_time":6.764,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":57,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-9402.0,"score":0.0,"rank":58.0,"hrank":58.0,"imp":41.571,"train_time":6.51,"predict_time":0.162,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":58,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-10120.0,"score":0.0,"rank":59.0,"hrank":59.0,"imp":43.96,"train_time":553.49,"predict_time":0.174,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-10876.0,"score":0.0,"rank":60.0,"hrank":60.0,"imp":45.54,"train_time":553.49,"predict_time":1.636,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":60,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":-11649.0,"score":0.0,"rank":61.0,"hrank":61.0,"imp":46.483,"train_time":65.84,"predict_time":3.48,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":61,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-12439.0,"score":0.0,"rank":62.0,"hrank":62.0,"imp":51.593,"train_time":30.07,"predict_time":0.562,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":62,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":-13086.0,"score":0.0,"rank":63.0,"hrank":63.0,"imp":52.398,"train_time":0.23,"predict_time":0.042,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":63,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-13834.0,"score":0.0,"rank":64.0,"hrank":64.0,"imp":53.969,"train_time":30.07,"predict_time":0.124,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":64,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":-14639.0,"score":0.0,"rank":65.0,"hrank":65.0,"imp":54.442,"train_time":2.73,"predict_time":0.083,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":65,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-15426.0,"score":0.0,"rank":66.0,"hrank":66.0,"imp":54.991,"train_time":30.19,"predict_time":0.134,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":66,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-16140.0,"score":0.0,"rank":67.0,"hrank":67.0,"imp":57.946,"train_time":30.19,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":67,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-17040.0,"score":0.0,"rank":68.0,"hrank":68.0,"imp":58.227,"train_time":0.13,"predict_time":0.015,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":68,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-18387.0,"score":0.0,"rank":69.0,"hrank":69.0,"imp":59.135,"train_time":76.41,"predict_time":2.853,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":-19408.0,"score":0.0,"rank":70.0,"hrank":70.0,"imp":59.496,"train_time":2.73,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":70,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":-21214.0,"score":0.0,"rank":71.0,"hrank":71.0,"imp":59.669,"train_time":3.14,"predict_time":0.921,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":-22944.0,"score":0.0,"rank":72.0,"hrank":72.0,"imp":59.719,"train_time":0.11,"predict_time":0.023,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+0\/-0","imputed":false},{"position":72,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-24570.0,"score":0.0,"rank":73.0,"hrank":73.0,"imp":72.021,"train_time":76.41,"predict_time":0.258,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false},{"position":73,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-25950.0,"score":0.0,"rank":74.0,"hrank":74.0,"imp":84.461,"train_time":0.14,"predict_time":0.123,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+0\/-0","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/pareto_front_explorer.html
index 63ec50f5a2aae1b65b307f6295e4909fc760f8f1..12f92bad9b6f0dced76194e4e913f1d3f3aea05f 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7827549071,"imp":37.0982410724,"elo":-7958.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0218721746,"imp":17.3395361789,"elo":11658.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0300453503,"imp":15.1970143595,"elo":17292.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.234055305,"imp":15.012462316,"elo":18005.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0128154423,"imp":22.2905525261,"elo":5792.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0152880152,"imp":16.5680326571,"elo":13716.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1959219282,"imp":16.9425810411,"elo":13045.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.014064448,"imp":17.8927578513,"elo":10927.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0119814523,"imp":14.8173560152,"elo":18724.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1069197134,"imp":14.192258111,"elo":24029.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0418367287,"imp":52.3978789302,"elo":-13086.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0088855815,"imp":21.2482114687,"elo":6522.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739987392,"imp":20.2523403303,"elo":8261.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0832738002,"imp":54.4419294386,"elo":-14639.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1735651487,"imp":43.9601644927,"elo":-10120.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"imp":45.5403937799,"elo":-10875.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0474691168,"imp":29.7473637296,"elo":-3054.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.4009232048,"imp":14.3659785469,"elo":21042.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.5785391893,"imp":14.3376604523,"elo":21690.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":3.1918999344,"imp":0.0,"elo":34554.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7472229818,"imp":24.5099501596,"elo":2081.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2155277801,"imp":17.9124144533,"elo":10208.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":25.5948192523,"imp":17.2788666655,"elo":12359.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7641563624,"imp":39.799071256,"elo":-8688.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"imp":59.4960081453,"elo":-19408.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1240589189,"imp":53.9690546132,"elo":-13833.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5623533631,"imp":51.5926415496,"elo":-12438.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0069604217,"imp":27.4194188138,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0154110106,"imp":14.6872814824,"elo":19479.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1538951715,"imp":14.2751202196,"elo":22437.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.1615432661,"imp":41.5705102944,"elo":-9401.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0552248943,"imp":28.548024354,"elo":-2323.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5454919741,"imp":27.9827612422,"elo":-972.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5348496069,"imp":18.4784935089,"elo":9576.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8184836382,"imp":7.3854101119,"elo":26990.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3079343419,"imp":4.4976437918,"elo":31548.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2367423058,"imp":28.0714246531,"elo":-1604.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"imp":27.5967758404,"elo":-331.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.5244814723,"imp":27.4759618173,"elo":342.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":275.9499442925,"imp":23.6244714479,"elo":4593.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.277558181,"imp":14.4214382191,"elo":20252.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":17.7035952873,"imp":13.5232072922,"elo":24743.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0232842395,"imp":59.7189648865,"elo":-22944.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.056543605,"imp":19.6416425074,"elo":8935.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426377996,"imp":14.2045636484,"elo":23243.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0740791906,"imp":34.7546595439,"elo":-5112.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0574085901,"imp":24.4579757413,"elo":2811.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5419846432,"imp":24.4328219572,"elo":3320.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5160116403,"imp":9.3847277445,"elo":25643.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2726374657,"imp":4.5260970421,"elo":29818.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4795389483,"imp":46.4830894898,"elo":-11648.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.468865642,"imp":36.9007036438,"elo":-6530.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.0373758651,"imp":36.9583859241,"elo":-7254.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1994643012,"imp":23.4592002877,"elo":5283.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0276328914,"imp":34.8795471811,"elo":-5810.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0266472235,"imp":24.1915976671,"elo":4010.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1720376576,"imp":24.5765776663,"elo":1461.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"imp":15.6926314585,"elo":15134.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"imp":15.692631292,"elo":15858.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.1704905161,"imp":15.4240395653,"elo":16562.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"imp":20.6424796077,"elo":7142.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"imp":20.6424750924,"elo":7603.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":21.7408394399,"imp":16.3672216796,"elo":14414.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9209350283,"imp":59.6685107498,"elo":-21214.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.9740680655,"imp":33.9505172259,"elo":-4441.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.7671484184,"imp":32.399105052,"elo":-3755.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7827549071,"x_train":66.0588444215,"imp":37.0982410724,"elo":-7958.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0218721746,"x_train":3.5923477438,"imp":17.3395361789,"elo":11658.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0300453503,"x_train":1261.2440700178,"imp":15.1970143595,"elo":17292.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.234055305,"x_train":1261.2440700178,"imp":15.012462316,"elo":18005.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0128154423,"x_train":0.2281173573,"imp":22.2905525261,"elo":5792.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0152880152,"x_train":66.2292774111,"imp":16.5680326571,"elo":13716.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1959219282,"x_train":66.2292774111,"imp":16.9425810411,"elo":13045.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.014064448,"x_train":13.8733921956,"imp":17.8927578513,"elo":10927.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0119814523,"x_train":6601.4555400381,"imp":14.8173560152,"elo":18724.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1069197134,"x_train":6601.4555400381,"imp":14.192258111,"elo":24029.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0418367287,"x_train":0.2342836282,"imp":52.3978789302,"elo":-13086.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0088855815,"x_train":28.4407435042,"imp":21.2482114687,"elo":6522.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739987392,"x_train":28.4407435042,"imp":20.2523403303,"elo":8261.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0832738002,"x_train":2.7277827402,"imp":54.4419294386,"elo":-14639.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1735651487,"x_train":553.4880496571,"imp":43.9601644927,"elo":-10120.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"x_train":553.4880496571,"imp":45.5403937799,"elo":-10875.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.0474691168,"x_train":0.2060542703,"imp":29.7473637296,"elo":-3054.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.4009232048,"x_train":157.4028552082,"imp":14.3659785469,"elo":21042.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.5785391893,"x_train":157.4028552082,"imp":14.3376604523,"elo":21690.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":3.1918999344,"x_train":3.9378111103,"imp":0.0,"elo":34554.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7472229818,"x_train":14.534943689,"imp":24.5099501596,"elo":2081.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":2.2155277801,"x_train":4378.8792899745,"imp":17.9124144533,"elo":10208.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":25.5948192523,"x_train":4378.8792899745,"imp":17.2788666655,"elo":12359.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7641563624,"x_train":16.7832602994,"imp":39.799071256,"elo":-8688.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"x_train":2.7257969394,"imp":59.4960081453,"elo":-19408.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1240589189,"x_train":30.0706477892,"imp":53.9690546132,"elo":-13833.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5623533631,"x_train":30.0706477892,"imp":51.5926415496,"elo":-12438.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0069604217,"x_train":0.1776920981,"imp":27.4194188138,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0154110106,"x_train":61.4738106436,"imp":14.6872814824,"elo":19479.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1538951715,"x_train":61.4738106436,"imp":14.2751202196,"elo":22437.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.1615432661,"x_train":6.5101464138,"imp":41.5705102944,"elo":-9401.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0552248943,"x_train":1322.498714333,"imp":28.548024354,"elo":-2323.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":0.5454919741,"x_train":1322.498714333,"imp":27.9827612422,"elo":-972.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5348496069,"x_train":1.4849621492,"imp":18.4784935089,"elo":9576.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.8184836382,"x_train":134.594931747,"imp":7.3854101119,"elo":26990.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3079343419,"x_train":134.594931747,"imp":4.4976437918,"elo":31548.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2367423058,"x_train":9.8924281166,"imp":28.0714246531,"elo":-1604.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"x_train":41.2100429512,"imp":27.5967758404,"elo":-331.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.5244814723,"x_train":4391.0141404006,"imp":27.4759618173,"elo":342.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":275.9499442925,"x_train":4391.0141404006,"imp":23.6244714479,"elo":4593.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.277558181,"x_train":0.8171638129,"imp":14.4214382191,"elo":20252.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":17.7035952873,"x_train":35.6508208249,"imp":13.5232072922,"elo":24743.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0232842395,"x_train":0.1087394813,"imp":59.7189648865,"elo":-22944.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":1.056543605,"x_train":2.7860859796,"imp":19.6416425074,"elo":8935.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426377996,"x_train":0.4823068139,"imp":14.2045636484,"elo":23243.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0740791906,"x_train":4.047894089,"imp":34.7546595439,"elo":-5112.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0574085901,"x_train":1142.3492528959,"imp":24.4579757413,"elo":2811.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5419846432,"x_train":1142.3492528959,"imp":24.4328219572,"elo":3320.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5160116403,"x_train":1.8403475551,"imp":9.3847277445,"elo":25643.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2726374657,"x_train":0.92595243,"imp":4.5260970421,"elo":29818.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4795389483,"x_train":65.8419489083,"imp":46.4830894898,"elo":-11648.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.468865642,"x_train":3349.2183478585,"imp":36.9007036438,"elo":-6530.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.0373758651,"x_train":3349.2183478585,"imp":36.9583859241,"elo":-7254.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1994643012,"x_train":0.578742138,"imp":23.4592002877,"elo":5283.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0276328914,"x_train":3.3435344852,"imp":34.8795471811,"elo":-5810.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0266472235,"x_train":2341.9377908803,"imp":24.1915976671,"elo":4010.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1720376576,"x_train":2341.9377908803,"imp":24.5765776663,"elo":1461.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"x_train":0.321861902,"imp":15.6926314585,"elo":15134.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"x_train":456.3324763471,"imp":15.692631292,"elo":15858.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.1704905161,"x_train":456.3324763471,"imp":15.4240395653,"elo":16562.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"x_train":52.4793328731,"imp":20.6424796077,"elo":7142.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"x_train":1229.419892487,"imp":20.6424750924,"elo":7603.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":21.7408394399,"x_train":1229.419892487,"imp":16.3672216796,"elo":14414.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.9209350283,"x_train":3.1442091702,"imp":59.6685107498,"elo":-21214.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.9740680655,"x_train":2009.1393249946,"imp":33.9505172259,"elo":-4441.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":3.7671484184,"x_train":2009.1393249946,"imp":32.399105052,"elo":-3755.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
index cd854712342702e5ae9f009c390f5b745b0c6a6b..17fc70716f901a46cd49095fc98596e0055f2a61 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":66.0588444215,"imp":37.1098221861,"elo":-1801.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":3.5923477438,"imp":17.3547552173,"elo":4866.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":9.9824633777,"imp":16.8180362511,"elo":5512.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":28.1814882563,"imp":16.1030488468,"elo":5968.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":56.6972466524,"imp":15.6955147734,"elo":6243.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":159.5146818194,"imp":15.3829095761,"elo":6860.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":313.167560406,"imp":15.2629283415,"elo":7001.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":620.4583392781,"imp":15.1258558949,"elo":7233.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":930.7284389971,"imp":15.0630430176,"elo":7303.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":1261.2440700178,"imp":15.0249864754,"elo":7430.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.2281173573,"imp":22.3048595539,"elo":2628.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.5627936741,"imp":20.9230935328,"elo":2866.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.4596619621,"imp":19.1389134322,"elo":3963.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.0218870561,"imp":18.1056432687,"elo":4344.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":7.9773474182,"imp":17.5870646377,"elo":4695.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":16.3876808766,"imp":17.2942989824,"elo":4987.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":33.0525808274,"imp":17.0182082807,"elo":5249.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":49.4935904635,"imp":16.9232947793,"elo":5448.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":66.2292774111,"imp":16.9578731367,"elo":5349.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":13.8733921956,"imp":17.9078747707,"elo":4639.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":46.1464646082,"imp":16.7341770342,"elo":5575.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":147.52300074,"imp":15.3806792968,"elo":6927.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":306.6782807185,"imp":15.1452722438,"elo":7158.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":801.6286487563,"imp":14.6415906517,"elo":7708.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":1627.1101123839,"imp":14.430991369,"elo":8244.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":3264.2024061238,"imp":14.2876314294,"elo":8729.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":4901.8335790951,"imp":14.2479220842,"elo":8861.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":6601.4555400381,"imp":14.2080565817,"elo":9421.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.2342836282,"imp":52.4066418594,"elo":-3705.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.3800410039,"imp":45.1518614912,"elo":-2755.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.7928114455,"imp":30.662456218,"elo":-381.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.4929370392,"imp":24.7164640518,"elo":1784.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.6241622692,"imp":20.9408156964,"elo":2803.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.1550594653,"imp":20.5217514993,"elo":3207.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.2035240625,"imp":20.3845714473,"elo":3278.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":21.2866256405,"imp":20.325403259,"elo":3350.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":28.4407435042,"imp":20.2670230505,"elo":3493.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.7277827402,"imp":54.4503173153,"elo":-3981.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.2812955867,"imp":54.5336434441,"elo":-4153.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":13.7046141456,"imp":52.5784645996,"elo":-3806.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":28.1489948252,"imp":50.2197571748,"elo":-3467.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":65.9177907885,"imp":49.0197664142,"elo":-3402.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":133.9142240391,"imp":47.9404231189,"elo":-3332.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":275.0443968187,"imp":47.0096623081,"elo":-3257.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":412.7471335694,"imp":46.1569676243,"elo":-2959.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":553.4880496571,"imp":45.5504205965,"elo":-2824.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.2060542703,"imp":29.7602973693,"elo":-254.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":1.2095397384,"imp":29.5948617941,"elo":-190.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":3.4430801699,"imp":26.8034215743,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":7.3054081422,"imp":24.0334315785,"elo":2217.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":19.34969751,"imp":17.9130900921,"elo":4525.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":38.6369993724,"imp":14.9432102176,"elo":7490.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":77.1677596734,"imp":14.4967977362,"elo":7814.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":116.0761985074,"imp":14.4424922558,"elo":7964.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":157.4028552082,"imp":14.352900556,"elo":8512.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":3.9378111103,"imp":0.0,"elo":12234.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":14.534943689,"imp":24.5238492021,"elo":1912.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":37.933698739,"imp":23.3114694228,"elo":2525.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":106.8025033221,"imp":21.8841094817,"elo":2683.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":212.1550450122,"imp":20.8257985661,"elo":2932.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":514.1027685745,"imp":19.4632400006,"elo":3899.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1078.615636086,"imp":18.6043461802,"elo":4028.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":2143.6938229576,"imp":17.9416534339,"elo":4405.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":3217.2348435455,"imp":17.5630829294,"elo":4751.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":4378.8792899745,"imp":17.294096846,"elo":5115.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":16.7832602994,"imp":39.7514072301,"elo":-2273.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":2.7257969394,"imp":59.5308044317,"elo":-6061.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":8.7745991633,"imp":54.8868562582,"elo":-4248.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":25.2036626358,"imp":51.9237268152,"elo":-3648.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":30.0706477892,"imp":51.6015540581,"elo":-3559.3,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.1776920981,"imp":27.5132675094,"elo":763.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4652499352,"imp":19.8025665932,"elo":3633.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":1.4763660732,"imp":16.6141857743,"elo":5639.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":3.0369890673,"imp":15.5562712818,"elo":6378.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":7.4865228781,"imp":14.8199538034,"elo":7607.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":15.1232073981,"imp":14.5968458625,"elo":7757.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":30.3263660899,"imp":14.3775932239,"elo":8409.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":45.7813804379,"imp":14.3190633811,"elo":8543.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":61.4738106436,"imp":14.2909034341,"elo":8634.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.5101464138,"imp":41.5812680827,"elo":-2541.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":12.3980382971,"imp":37.7980112911,"elo":-2067.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":30.8974860361,"imp":35.1790834708,"elo":-1482.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":62.2345356879,"imp":33.228621422,"elo":-850.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":164.4196952291,"imp":31.5826250175,"elo":-565.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":334.0371204561,"imp":30.8459045697,"elo":-443.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":660.4294593011,"imp":29.0387215642,"elo":-61.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":990.5204223442,"imp":28.5480532422,"elo":134.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":1322.498714333,"imp":27.9960689724,"elo":338.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.4849621492,"imp":18.4935028204,"elo":4123.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.9184440201,"imp":18.2175225288,"elo":4281.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":3.2451422052,"imp":17.4641145366,"elo":4808.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.3339567122,"imp":17.0183735883,"elo":5182.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":13.7229026201,"imp":14.4883605878,"elo":7883.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":29.7246860705,"imp":9.4741799614,"elo":9957.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":63.7718881675,"imp":6.6167192702,"elo":10518.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":96.2590274265,"imp":5.0880354109,"elo":10764.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":134.594931747,"imp":4.5152271839,"elo":11279.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":9.8924281166,"imp":28.0846677645,"elo":234.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.2100429512,"imp":27.6101063417,"elo":515.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":62.7744579205,"imp":27.3298114984,"elo":1156.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":116.8910871561,"imp":27.1666199817,"elo":1215.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":207.4972060022,"imp":26.8735702028,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":532.9957633827,"imp":25.7558200876,"elo":1524.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1085.275659615,"imp":25.1480444001,"elo":1629.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2177.7649560761,"imp":24.4892964104,"elo":1987.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3284.6615266756,"imp":23.8383971903,"elo":2270.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4391.0141404006,"imp":23.6319338123,"elo":2375.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.8171638129,"imp":14.4371944943,"elo":8101.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":35.6508208249,"imp":13.539128945,"elo":9732.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":0.1087394813,"imp":59.7263812191,"elo":-6683.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":2.7860859796,"imp":19.656437666,"elo":3735.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.4823068139,"imp":14.2203598534,"elo":9097.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":4.047894089,"imp":34.7666718876,"elo":-1149.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.1210227808,"imp":32.3780565133,"elo":-624.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":26.2329613573,"imp":29.1498676397,"elo":-126.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":55.2138352308,"imp":27.4167126626,"elo":1095.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":141.9296718443,"imp":26.5043661991,"elo":1374.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":284.591401136,"imp":25.7598168207,"elo":1473.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":562.5419157372,"imp":25.0307019084,"elo":1682.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":849.3471745634,"imp":24.7251939402,"elo":1734.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":1142.3492528959,"imp":24.446734989,"elo":2116.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":1.8403475551,"imp":9.4014113524,"elo":10105.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":0.92595243,"imp":4.5436751956,"elo":10938.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":65.8419489083,"imp":46.4929426215,"elo":-3100.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":82.3201598909,"imp":46.2138836432,"elo":-3026.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":128.1588279475,"imp":45.555112229,"elo":-2900.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":190.5221223243,"imp":44.0271781614,"elo":-2657.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":426.866827693,"imp":41.0974133818,"elo":-2410.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":860.4651862449,"imp":38.3656852859,"elo":-2135.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1704.5805078339,"imp":37.7565354561,"elo":-1999.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2515.3432482994,"imp":37.1313510511,"elo":-1932.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":3349.2183478585,"imp":36.9699928149,"elo":-1706.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.578742138,"imp":23.4732834803,"elo":2475.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":3.3435344852,"imp":34.8915367475,"elo":-1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":13.743652342,"imp":34.3728415445,"elo":-1086.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":42.5186103306,"imp":33.259261365,"elo":-907.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":107.3316064659,"imp":31.0785509014,"elo":-505.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":274.494576287,"imp":28.8315951234,"elo":2.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":577.8040181111,"imp":27.5575080172,"elo":622.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1167.0902850561,"imp":26.2831539442,"elo":1424.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":1768.1386269443,"imp":25.5800576293,"elo":1576.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":2341.9377908803,"imp":24.5904642306,"elo":1830.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.321861902,"imp":15.7081535219,"elo":6139.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":3.1340720026,"imp":15.6081723436,"elo":6311.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":10.4589579493,"imp":15.4893036937,"elo":6502.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":19.8143120293,"imp":15.4902145867,"elo":6442.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":56.2043887857,"imp":15.4747969248,"elo":6609.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":113.5858543435,"imp":15.4774957603,"elo":6558.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":226.5076734886,"imp":15.4703397968,"elo":6656.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":337.6096231295,"imp":15.4501933332,"elo":6702.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":456.3324763471,"imp":15.4377171274,"elo":6801.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":52.4793328731,"imp":20.657085983,"elo":3102.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":96.7397158429,"imp":19.9767460723,"elo":3563.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":241.2633499833,"imp":19.4822854162,"elo":3834.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":479.2228112407,"imp":18.4883332163,"elo":4218.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":1184.4533790756,"imp":16.4964918995,"elo":5767.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":1229.419892487,"imp":16.3826197074,"elo":5900.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":3.1442091702,"imp":59.6759371,"elo":-6488.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":13.155466857,"imp":53.945057456,"elo":-3855.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":45.477404513,"imp":44.9123114388,"elo":-2708.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":96.1451044259,"imp":38.4628370628,"elo":-2204.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":249.6329672147,"imp":35.9218683621,"elo":-1547.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":497.3958095536,"imp":34.7739561298,"elo":-1281.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":999.5916054172,"imp":33.8284439456,"elo":-965.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1494.1081451946,"imp":33.031286653,"elo":-794.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2009.1393249946,"imp":32.4115806664,"elo":-738.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":1.7827549071,"x_train":66.0588444215,"imp":37.1098221861,"elo":-1801.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0218721746,"x_train":3.5923477438,"imp":17.3547552173,"elo":4866.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0497134013,"x_train":9.9824633777,"imp":16.8180362511,"elo":5512.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1174311067,"x_train":28.1814882563,"imp":16.1030488468,"elo":5968.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1428266464,"x_train":56.6972466524,"imp":15.6955147734,"elo":6243.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2357748832,"x_train":159.5146818194,"imp":15.3829095761,"elo":6860.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2695920408,"x_train":313.167560406,"imp":15.2629283415,"elo":7001.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3044187588,"x_train":620.4583392781,"imp":15.1258558949,"elo":7233.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2568636437,"x_train":930.7284389971,"imp":15.0630430176,"elo":7303.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.280383328,"x_train":1261.2440700178,"imp":15.0249864754,"elo":7430.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0128154423,"x_train":0.2281173573,"imp":22.3048595539,"elo":2628.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0292978626,"x_train":0.5627936741,"imp":20.9230935328,"elo":2866.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0535690992,"x_train":1.4596619621,"imp":19.1389134322,"elo":3963.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0727289501,"x_train":3.0218870561,"imp":18.1056432687,"elo":4344.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1288669352,"x_train":7.9773474182,"imp":17.5870646377,"elo":4695.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.154154967,"x_train":16.3876808766,"imp":17.2942989824,"elo":4987.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1757045439,"x_train":33.0525808274,"imp":17.0182082807,"elo":5249.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1727469812,"x_train":49.4935904635,"imp":16.9232947793,"elo":5448.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1959219282,"x_train":66.2292774111,"imp":16.9578731367,"elo":5349.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.014064448,"x_train":13.8733921956,"imp":17.9078747707,"elo":4639.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0311340555,"x_train":46.1464646082,"imp":16.7341770342,"elo":5575.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0551923837,"x_train":147.52300074,"imp":15.3806792968,"elo":6927.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0688712999,"x_train":306.6782807185,"imp":15.1452722438,"elo":7158.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0984842812,"x_train":801.6286487563,"imp":14.6415906517,"elo":7708.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1253388792,"x_train":1627.1101123839,"imp":14.430991369,"elo":8244.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1457912936,"x_train":3264.2024061238,"imp":14.2876314294,"elo":8729.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1332857555,"x_train":4901.8335790951,"imp":14.2479220842,"elo":8861.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1069197134,"x_train":6601.4555400381,"imp":14.2080565817,"elo":9421.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0418367287,"x_train":0.2342836282,"imp":52.4066418594,"elo":-3705.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0444077791,"x_train":0.3800410039,"imp":45.1518614912,"elo":-2755.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0356755189,"x_train":0.7928114455,"imp":30.662456218,"elo":-381.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0273270001,"x_train":1.4929370392,"imp":24.7164640518,"elo":1784.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0395302294,"x_train":3.6241622692,"imp":20.9408156964,"elo":2803.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0642314309,"x_train":7.1550594653,"imp":20.5217514993,"elo":3207.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1059977241,"x_train":14.2035240625,"imp":20.3845714473,"elo":3278.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1375967472,"x_train":21.2866256405,"imp":20.325403259,"elo":3350.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1739987392,"x_train":28.4407435042,"imp":20.2670230505,"elo":3493.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.0832738002,"x_train":2.7277827402,"imp":54.4503173153,"elo":-3981.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1651997039,"x_train":5.2812955867,"imp":54.5336434441,"elo":-4153.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3882144021,"x_train":13.7046141456,"imp":52.5784645996,"elo":-3806.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6283891459,"x_train":28.1489948252,"imp":50.2197571748,"elo":-3467.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.9164506571,"x_train":65.9177907885,"imp":49.0197664142,"elo":-3402.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2468060948,"x_train":133.9142240391,"imp":47.9404231189,"elo":-3332.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7812914447,"x_train":275.0443968187,"imp":47.0096623081,"elo":-3257.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7948173827,"x_train":412.7471335694,"imp":46.1569676243,"elo":-2959.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.636166665,"x_train":553.4880496571,"imp":45.5504205965,"elo":-2824.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.0474691168,"x_train":0.2060542703,"imp":29.7602973693,"elo":-254.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.7435906933,"x_train":1.2095397384,"imp":29.5948617941,"elo":-190.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.8375843537,"x_train":3.4430801699,"imp":26.8034215743,"elo":1323.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.6051740249,"x_train":7.3054081422,"imp":24.0334315785,"elo":2217.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.3230917617,"x_train":19.34969751,"imp":17.9130900921,"elo":4525.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4649015268,"x_train":38.6369993724,"imp":14.9432102176,"elo":7490.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.2768939316,"x_train":77.1677596734,"imp":14.4967977362,"elo":7814.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.0310390923,"x_train":116.0761985074,"imp":14.4424922558,"elo":7964.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.5785391893,"x_train":157.4028552082,"imp":14.352900556,"elo":8512.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":3.1918999344,"x_train":3.9378111103,"imp":0.0,"elo":12234.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7472229818,"x_train":14.534943689,"imp":24.5238492021,"elo":1912.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.6445542539,"x_train":37.933698739,"imp":23.3114694228,"elo":2525.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.9535369347,"x_train":106.8025033221,"imp":21.8841094817,"elo":2683.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":11.2275665264,"x_train":212.1550450122,"imp":20.8257985661,"elo":2932.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":16.3103609399,"x_train":514.1027685745,"imp":19.4632400006,"elo":3899.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":19.9788525201,"x_train":1078.615636086,"imp":18.6043461802,"elo":4028.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":23.1702683559,"x_train":2143.6938229576,"imp":17.9416534339,"elo":4405.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":25.2758129387,"x_train":3217.2348435455,"imp":17.5630829294,"elo":4751.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":25.5948192523,"x_train":4378.8792899745,"imp":17.294096846,"elo":5115.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":6.7641563624,"x_train":16.7832602994,"imp":39.7514072301,"elo":-2273.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0124056683,"x_train":2.7257969394,"imp":59.5308044317,"elo":-6061.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1522131379,"x_train":8.7745991633,"imp":54.8868562582,"elo":-4248.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4737042685,"x_train":25.2036626358,"imp":51.9237268152,"elo":-3648.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5623533631,"x_train":30.0706477892,"imp":51.6015540581,"elo":-3559.3,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0069604217,"x_train":0.1776920981,"imp":27.5132675094,"elo":763.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0213595479,"x_train":0.4652499352,"imp":19.8025665932,"elo":3633.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0392463929,"x_train":1.4763660732,"imp":16.6141857743,"elo":5639.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0501367115,"x_train":3.0369890673,"imp":15.5562712818,"elo":6378.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0768595462,"x_train":7.4865228781,"imp":14.8199538034,"elo":7607.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.09727726,"x_train":15.1232073981,"imp":14.5968458625,"elo":7757.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1236267341,"x_train":30.3263660899,"imp":14.3775932239,"elo":8409.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1434609645,"x_train":45.7813804379,"imp":14.3190633811,"elo":8543.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1538951715,"x_train":61.4738106436,"imp":14.2909034341,"elo":8634.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.1615432661,"x_train":6.5101464138,"imp":41.5812680827,"elo":-2541.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2453378418,"x_train":12.3980382971,"imp":37.7980112911,"elo":-2067.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3029056071,"x_train":30.8974860361,"imp":35.1790834708,"elo":-1482.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.347018823,"x_train":62.2345356879,"imp":33.228621422,"elo":-850.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4558697672,"x_train":164.4196952291,"imp":31.5826250175,"elo":-565.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5786006034,"x_train":334.0371204561,"imp":30.8459045697,"elo":-443.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.471412833,"x_train":660.4294593011,"imp":29.0387215642,"elo":-61.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4979071245,"x_train":990.5204223442,"imp":28.5480532422,"elo":134.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.5454919741,"x_train":1322.498714333,"imp":27.9960689724,"elo":338.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5348496069,"x_train":1.4849621492,"imp":18.4935028204,"elo":4123.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6897445483,"x_train":1.9184440201,"imp":18.2175225288,"elo":4281.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.0053886682,"x_train":3.2451422052,"imp":17.4641145366,"elo":4808.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.7446747876,"x_train":5.3339567122,"imp":17.0183735883,"elo":5182.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.1304312666,"x_train":13.7229026201,"imp":14.4883605878,"elo":7883.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.5017657204,"x_train":29.7246860705,"imp":9.4741799614,"elo":9957.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6074456326,"x_train":63.7718881675,"imp":6.6167192702,"elo":10518.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.6520668764,"x_train":96.2590274265,"imp":5.0880354109,"elo":10764.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3079343419,"x_train":134.594931747,"imp":4.5152271839,"elo":11279.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":9.2367423058,"x_train":9.8924281166,"imp":28.0846677645,"elo":234.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":39.4049115076,"x_train":41.2100429512,"imp":27.6101063417,"elo":515.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":50.8141493102,"x_train":62.7744579205,"imp":27.3298114984,"elo":1156.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":77.2065039497,"x_train":116.8910871561,"imp":27.1666199817,"elo":1215.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":109.5641245161,"x_train":207.4972060022,"imp":26.8735702028,"elo":1270.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":163.2920265134,"x_train":532.9957633827,"imp":25.7558200876,"elo":1524.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":213.967010973,"x_train":1085.275659615,"imp":25.1480444001,"elo":1629.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":269.7422775799,"x_train":2177.7649560761,"imp":24.4892964104,"elo":1987.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":263.3543429854,"x_train":3284.6615266756,"imp":23.8383971903,"elo":2270.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":275.9499442925,"x_train":4391.0141404006,"imp":23.6319338123,"elo":2375.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.277558181,"x_train":0.8171638129,"imp":14.4371944943,"elo":8101.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":17.7035952873,"x_train":35.6508208249,"imp":13.539128945,"elo":9732.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.0232842395,"x_train":0.1087394813,"imp":59.7263812191,"elo":-6683.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":1.056543605,"x_train":2.7860859796,"imp":19.656437666,"elo":3735.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1426377996,"x_train":0.4823068139,"imp":14.2203598534,"elo":9097.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0740791906,"x_train":4.047894089,"imp":34.7666718876,"elo":-1149.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.127108128,"x_train":9.1210227808,"imp":32.3780565133,"elo":-624.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.2047906233,"x_train":26.2329613573,"imp":29.1498676397,"elo":-126.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2449575585,"x_train":55.2138352308,"imp":27.4167126626,"elo":1095.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3319391408,"x_train":141.9296718443,"imp":26.5043661991,"elo":1374.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4159006841,"x_train":284.591401136,"imp":25.7598168207,"elo":1473.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.4198951784,"x_train":562.5419157372,"imp":25.0307019084,"elo":1682.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.4821521088,"x_train":849.3471745634,"imp":24.7251939402,"elo":1734.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5419846432,"x_train":1142.3492528959,"imp":24.446734989,"elo":2116.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.5160116403,"x_train":1.8403475551,"imp":9.4014113524,"elo":10105.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2726374657,"x_train":0.92595243,"imp":4.5436751956,"elo":10938.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.4795389483,"x_train":65.8419489083,"imp":46.4929426215,"elo":-3100.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6272217169,"x_train":82.3201598909,"imp":46.2138836432,"elo":-3026.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.0494971432,"x_train":128.1588279475,"imp":45.555112229,"elo":-2900.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.8279085887,"x_train":190.5221223243,"imp":44.0271781614,"elo":-2657.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.9995876557,"x_train":426.866827693,"imp":41.0974133818,"elo":-2410.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":3.6380882465,"x_train":860.4651862449,"imp":38.3656852859,"elo":-2135.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.4734687528,"x_train":1704.5805078339,"imp":37.7565354561,"elo":-1999.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.7675349663,"x_train":2515.3432482994,"imp":37.1313510511,"elo":-1932.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.0373758651,"x_train":3349.2183478585,"imp":36.9699928149,"elo":-1706.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1994643012,"x_train":0.578742138,"imp":23.4732834803,"elo":2475.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0276328914,"x_train":3.3435344852,"imp":34.8915367475,"elo":-1348.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0620535399,"x_train":13.743652342,"imp":34.3728415445,"elo":-1086.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1179715238,"x_train":42.5186103306,"imp":33.259261365,"elo":-907.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1651259896,"x_train":107.3316064659,"imp":31.0785509014,"elo":-505.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2389435798,"x_train":274.494576287,"imp":28.8315951234,"elo":2.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2817484073,"x_train":577.8040181111,"imp":27.5575080172,"elo":622.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2701272436,"x_train":1167.0902850561,"imp":26.2831539442,"elo":1424.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.244273469,"x_train":1768.1386269443,"imp":25.5800576293,"elo":1576.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1720376576,"x_train":2341.9377908803,"imp":24.5904642306,"elo":1830.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.0468985287,"x_train":0.321861902,"imp":15.7081535219,"elo":6139.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3465841915,"x_train":3.1340720026,"imp":15.6081723436,"elo":6311.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6102341201,"x_train":10.4589579493,"imp":15.4893036937,"elo":6502.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.5888194515,"x_train":19.8143120293,"imp":15.4902145867,"elo":6442.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6938849221,"x_train":56.2043887857,"imp":15.4747969248,"elo":6609.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6851908948,"x_train":113.5858543435,"imp":15.4774957603,"elo":6558.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.74544132,"x_train":226.5076734886,"imp":15.4703397968,"elo":6656.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.9582567497,"x_train":337.6096231295,"imp":15.4501933332,"elo":6702.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1704905161,"x_train":456.3324763471,"imp":15.4377171274,"elo":6801.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":7.7845844048,"x_train":52.4793328731,"imp":20.657085983,"elo":3102.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":11.3040848505,"x_train":96.7397158429,"imp":19.9767460723,"elo":3563.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":18.4470099318,"x_train":241.2633499833,"imp":19.4822854162,"elo":3834.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":20.7391590001,"x_train":479.2228112407,"imp":18.4883332163,"elo":4218.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":23.6267433633,"x_train":1184.4533790756,"imp":16.4964918995,"elo":5767.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":21.7408394399,"x_train":1229.419892487,"imp":16.3826197074,"elo":5900.9,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.9209350283,"x_train":3.1442091702,"imp":59.6759371,"elo":-6488.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.8238649007,"x_train":13.155466857,"imp":53.945057456,"elo":-3855.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":2.7289794426,"x_train":45.477404513,"imp":44.9123114388,"elo":-2708.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":3.6121868214,"x_train":96.1451044259,"imp":38.4628370628,"elo":-2204.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":4.9673931163,"x_train":249.6329672147,"imp":35.9218683621,"elo":-1547.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":5.8186016306,"x_train":497.3958095536,"imp":34.7739561298,"elo":-1281.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":6.1787775661,"x_train":999.5916054172,"imp":33.8284439456,"elo":-965.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":5.262793854,"x_train":1494.1081451946,"imp":33.031286653,"elo":-794.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":3.7671484184,"x_train":2009.1393249946,"imp":32.4115806664,"elo":-738.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..1af6525b44179f58fcb814fac76355b9222e1fd6
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_multiclass/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["LimiX (default)", "AutoGluon 1.5 (extreme, 4h)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-3 (default)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (tuned)", "TabPFN-2.6 (default)", "TabFM (default)", "EBM (tuned + ensembled)", "TabICLv2 (default)", "RandomForest (tuned + ensembled)", "LightGBM (tuned + ensembled)", "LightGBM (tuned)", "TabDPT-Turbo (default)", "RandomForest (tuned)", "EBM (tuned)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "XGBoost (tuned + ensembled)", "XGBoost (tuned)", "XGBoost (default)", "iLTM (tuned + ensembled)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "CatBoost (default)", "EBM (default)", "ModernNCA (tuned)", "RealTabPFN-2.5 (default)", "TabICL (default)", "ExtraTrees (tuned + ensembled)", "iLTM (tuned)", "iLTM (default)", "ExtraTrees (tuned)", "ChimeraBoost (default)", "TabSwift (default)", "TabDPT (tuned + ensembled)", "TorchMLP (tuned)", "TabM (tuned + ensembled)", "TabM (tuned)", "ModernNCA (default)", "TorchMLP (tuned + ensembled)", "RandomForest (default)", "TabDPT (tuned)", "TabDPT (default)", "RealMLP (tuned + ensembled)", "SAP-RPT-OSS (default)", "RealMLP (tuned)", "LightGBM (default)", "xRFM (tuned + ensembled)", "xRFM (tuned)", "TabM (default)", "TorchMLP (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "BetaTabPFN (default)", "OrionMSP (default)", "RealMLP (default)", "FastaiMLP (tuned)", "FastaiMLP (tuned + ensembled)", "TabSTAR (default)", "PerpetualBooster (tuned + ensembled)", "ExtraTrees (default)", "PerpetualBooster (tuned)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "xRFM (default)", "TabFlex (default)", "KNN (tuned)", "KNN (default)"], "matrix": [[NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":1.0},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9863013699},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9726027397},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.9589041096},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.9452054795},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.9315068493},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.9178082192},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.904109589},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8904109589},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8767123288},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8630136986},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.8493150685},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.8356164384},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.8219178082},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.8082191781},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.7945205479},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7808219178},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7671232877},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7534246575},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.7397260274},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.7260273973},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7123287671},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.698630137},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6849315068},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6712328767},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6575342466},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.6438356164},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6301369863},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.6164383562},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.602739726},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5890410959},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5753424658},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.5616438356},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.5479452055},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.5342465753},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5205479452},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5068493151},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4931506849},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4794520548},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.4657534247},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.4520547945},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4383561644},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.4246575342},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.4109589041},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.397260274},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.3835616438},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.3698630137},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.3561643836},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3424657534},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.3287671233},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.3150684932},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.301369863},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2876712329},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2739726027},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2602739726},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.2465753425},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2328767123},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.2191780822},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2054794521},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.1917808219},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1780821918},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1643835616},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.1506849315},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1369863014},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1232876712},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.1095890411},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.095890411},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0821917808},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.0684931507},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0547945205},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.0410958904},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.0273972603},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.0136986301},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..034bcd327fe452b3b8dab7198d23566c6f8885fc
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,75 @@
+,LimiX (default),"AutoGluon 1.5 (extreme, 4h)",RealTabPFN-2.5 (tuned + ensembled),TabPFN-3 (default),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (tuned),TabPFN-2.6 (default),TabFM (default),EBM (tuned + ensembled),TabICLv2 (default),RandomForest (tuned + ensembled),LightGBM (tuned + ensembled),LightGBM (tuned),TabDPT-Turbo (default),RandomForest (tuned),EBM (tuned),CatBoost (tuned + ensembled),CatBoost (tuned),XGBoost (tuned + ensembled),XGBoost (tuned),XGBoost (default),iLTM (tuned + ensembled),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),ModernNCA (tuned + ensembled),CatBoost (default),EBM (default),ModernNCA (tuned),RealTabPFN-2.5 (default),TabICL (default),ExtraTrees (tuned + ensembled),iLTM (tuned),iLTM (default),ExtraTrees (tuned),ChimeraBoost (default),TabSwift (default),TabDPT (tuned + ensembled),TorchMLP (tuned),TabM (tuned + ensembled),TabM (tuned),ModernNCA (default),TorchMLP (tuned + ensembled),RandomForest (default),TabDPT (tuned),TabDPT (default),RealMLP (tuned + ensembled),SAP-RPT-OSS (default),RealMLP (tuned),LightGBM (default),xRFM (tuned + ensembled),xRFM (tuned),TabM (default),TorchMLP (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),BetaTabPFN (default),OrionMSP (default),RealMLP (default),FastaiMLP (tuned),FastaiMLP (tuned + ensembled),TabSTAR (default),PerpetualBooster (tuned + ensembled),ExtraTrees (default),PerpetualBooster (tuned),FastaiMLP (default),Linear (tuned + ensembled),Linear (tuned),Linear (default),KNN (tuned + ensembled),PerpetualBooster (default),xRFM (default),TabFlex (default),KNN (tuned),KNN (default)
+LimiX (default),,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICL (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+OrionMSP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
index 8577658f5c04a3855389608ccf5d1279e6a09749..b7966aafa1cd5a5aaf6e2acbaadcbfab1bfc9f9b 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Multiclass | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..171703ce24d8f3d589df6575c435ea7deb7332c0
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_multiclass/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1743.0,"score":0.673,"rank":8.71,"hrank":3.11,"imp":5.637,"train_time":2837.96,"predict_time":9.546,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+469\/-162","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1739.0,"score":0.652,"rank":8.86,"hrank":5.17,"imp":12.628,"train_time":247.2,"predict_time":3.959,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+244\/-61","imputed":false},{"position":2,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1724.0,"score":0.832,"rank":9.43,"hrank":1.31,"imp":6.112,"train_time":52.66,"predict_time":6.434,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1290\/-280","imputed":false},{"position":3,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1657.0,"score":0.58,"rank":12.36,"hrank":9.25,"imp":14.338,"train_time":2837.96,"predict_time":1.132,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+241\/-110","imputed":false},{"position":4,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1649.0,"score":0.57,"rank":12.71,"hrank":8.9,"imp":13.688,"train_time":10.78,"predict_time":0.65,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+248\/-109","imputed":false},{"position":5,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1594.0,"score":0.475,"rank":15.64,"hrank":8.87,"imp":13.213,"train_time":10.97,"predict_time":0.855,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+310\/-126","imputed":false},{"position":6,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1590.0,"score":0.5,"rank":15.86,"hrank":6.26,"imp":12.315,"train_time":9.45,"predict_time":0.514,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+323\/-141","imputed":false},{"position":7,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1531.0,"score":0.49,"rank":19.57,"hrank":6.27,"imp":15.462,"train_time":5.36,"predict_time":0.234,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+375\/-177","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1524.0,"score":0.417,"rank":20.0,"hrank":8.96,"imp":17.503,"train_time":4655.15,"predict_time":21.781,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+191\/-93","imputed":false},{"position":9,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1490.0,"score":0.433,"rank":22.36,"hrank":11.26,"imp":20.708,"train_time":2971.32,"predict_time":17.821,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+318\/-194","imputed":true},{"position":10,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1486.0,"score":0.359,"rank":22.57,"hrank":15.56,"imp":15.294,"train_time":46.11,"predict_time":9.187,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+233\/-134","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1484.0,"score":0.362,"rank":22.71,"hrank":17.86,"imp":17.942,"train_time":4655.15,"predict_time":4.029,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+178\/-112","imputed":false},{"position":12,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1482.0,"score":0.385,"rank":22.86,"hrank":11.14,"imp":19.331,"train_time":6318.92,"predict_time":3.195,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+236\/-120","imputed":false},{"position":13,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1478.0,"score":0.329,"rank":23.14,"hrank":15.2,"imp":16.07,"train_time":3632.1,"predict_time":0.238,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+212\/-101","imputed":false},{"position":14,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1476.0,"score":0.323,"rank":23.29,"hrank":18.11,"imp":16.994,"train_time":3632.1,"predict_time":1.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+211\/-102","imputed":false},{"position":15,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1466.0,"score":0.4,"rank":24.0,"hrank":14.83,"imp":19.55,"train_time":3829.4,"predict_time":4.64,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+163\/-125","imputed":false},{"position":16,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1458.0,"score":0.381,"rank":24.57,"hrank":10.87,"imp":19.605,"train_time":3829.4,"predict_time":0.879,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+165\/-130","imputed":false},{"position":17,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1457.0,"score":0.376,"rank":24.64,"hrank":18.31,"imp":21.561,"train_time":202.72,"predict_time":4.62,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+225\/-181","imputed":true},{"position":18,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1437.0,"score":0.267,"rank":26.14,"hrank":19.35,"imp":20.452,"train_time":1394.84,"predict_time":3.577,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+182\/-95","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1429.0,"score":0.256,"rank":26.71,"hrank":18.97,"imp":20.536,"train_time":1394.84,"predict_time":1.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+195\/-92","imputed":false},{"position":20,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1429.0,"score":0.342,"rank":26.71,"hrank":9.5,"imp":20.925,"train_time":2149.3,"predict_time":2.926,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+161\/-182","imputed":false},{"position":21,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1429.0,"score":0.388,"rank":26.71,"hrank":18.23,"imp":22.062,"train_time":10182.26,"predict_time":399.334,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+216\/-195","imputed":false},{"position":22,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1396.0,"score":0.286,"rank":29.36,"hrank":15.69,"imp":20.645,"train_time":6.0,"predict_time":0.388,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+232\/-180","imputed":true},{"position":23,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1384.0,"score":0.176,"rank":30.29,"hrank":28.96,"imp":20.963,"train_time":2342.58,"predict_time":5.129,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-55","imputed":false},{"position":24,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1384.0,"score":0.264,"rank":30.29,"hrank":16.46,"imp":23.603,"train_time":5.65,"predict_time":0.269,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+212\/-166","imputed":false},{"position":25,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1384.0,"score":0.286,"rank":30.36,"hrank":19.57,"imp":22.262,"train_time":2971.32,"predict_time":0.477,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+223\/-181","imputed":true},{"position":26,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1383.0,"score":0.185,"rank":30.43,"hrank":28.45,"imp":20.682,"train_time":2342.58,"predict_time":1.065,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-65","imputed":false},{"position":27,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1365.0,"score":0.274,"rank":31.86,"hrank":15.79,"imp":23.74,"train_time":17.34,"predict_time":0.728,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+164\/-155","imputed":false},{"position":28,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1364.0,"score":0.367,"rank":32.0,"hrank":8.33,"imp":22.911,"train_time":22.26,"predict_time":1.556,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+310\/-247","imputed":false},{"position":29,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1362.0,"score":0.318,"rank":32.14,"hrank":13.02,"imp":22.982,"train_time":10182.26,"predict_time":40.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+244\/-196","imputed":false},{"position":30,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1353.0,"score":0.176,"rank":32.86,"hrank":28.75,"imp":21.288,"train_time":34015.19,"predict_time":986.352,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+126\/-88","imputed":false},{"position":31,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1351.0,"score":0.183,"rank":33.07,"hrank":28.89,"imp":21.853,"train_time":6995.99,"predict_time":9.551,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+149\/-103","imputed":false},{"position":32,"method":"BetaTabPFN","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.02527","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1350.0,"score":0.219,"rank":33.14,"hrank":24.24,"imp":23.298,"train_time":326.32,"predict_time":4.279,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+165\/-137","imputed":false},{"position":33,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1343.0,"score":0.268,"rank":33.71,"hrank":5.85,"imp":22.387,"train_time":2149.3,"predict_time":0.146,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+192\/-197","imputed":false},{"position":34,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1342.0,"score":0.161,"rank":33.79,"hrank":26.87,"imp":22.153,"train_time":6995.99,"predict_time":0.583,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-128","imputed":false},{"position":35,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1324.0,"score":0.248,"rank":35.21,"hrank":14.91,"imp":24.129,"train_time":1042.39,"predict_time":1.873,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+187\/-214","imputed":false},{"position":36,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1306.0,"score":0.157,"rank":36.71,"hrank":26.69,"imp":24.6,"train_time":27727.7,"predict_time":0.31,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+126\/-125","imputed":false},{"position":37,"method":"TabICL","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2502.05564","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1301.0,"score":0.205,"rank":37.07,"hrank":16.65,"imp":27.193,"train_time":11.7,"predict_time":2.132,"imputed_pct":14.29,"hardware":"GPU","elo_ci":"+245\/-195","imputed":true},{"position":38,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1295.0,"score":0.146,"rank":37.57,"hrank":33.89,"imp":23.042,"train_time":29.58,"predict_time":0.343,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-94","imputed":false},{"position":39,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1286.0,"score":0.162,"rank":38.29,"hrank":32.71,"imp":22.578,"train_time":34015.19,"predict_time":240.683,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+144\/-141","imputed":false},{"position":40,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1281.0,"score":0.145,"rank":38.71,"hrank":34.37,"imp":24.794,"train_time":3834.57,"predict_time":2.294,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+125\/-145","imputed":false},{"position":41,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1270.0,"score":0.231,"rank":39.57,"hrank":25.78,"imp":24.746,"train_time":731.23,"predict_time":2.021,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+191\/-249","imputed":false},{"position":42,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1263.0,"score":0.158,"rank":40.14,"hrank":25.83,"imp":26.279,"train_time":2750.22,"predict_time":13.569,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+266\/-217","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1246.0,"score":0.164,"rank":41.5,"hrank":16.54,"imp":27.277,"train_time":1042.39,"predict_time":0.294,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+210\/-215","imputed":false},{"position":44,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1242.0,"score":0.149,"rank":41.86,"hrank":30.09,"imp":25.601,"train_time":5.15,"predict_time":0.526,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+184\/-213","imputed":false},{"position":45,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1242.0,"score":0.142,"rank":41.86,"hrank":35.51,"imp":26.131,"train_time":27727.7,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+131\/-130","imputed":false},{"position":46,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1233.0,"score":0.113,"rank":42.57,"hrank":37.96,"imp":26.313,"train_time":10.95,"predict_time":0.048,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+125\/-146","imputed":false},{"position":47,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1233.0,"score":0.194,"rank":42.57,"hrank":16.88,"imp":26.571,"train_time":51.52,"predict_time":40.201,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+249\/-250","imputed":false},{"position":48,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1231.0,"score":0.138,"rank":42.71,"hrank":21.33,"imp":26.824,"train_time":3834.57,"predict_time":0.22,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+188\/-192","imputed":false},{"position":49,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.11,"rank":43.43,"hrank":38.3,"imp":26.159,"train_time":2001.62,"predict_time":1.335,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+130\/-172","imputed":false},{"position":50,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.132,"rank":43.43,"hrank":35.99,"imp":25.643,"train_time":2001.62,"predict_time":0.274,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+153\/-202","imputed":false},{"position":51,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1219.0,"score":0.089,"rank":43.71,"hrank":39.18,"imp":27.829,"train_time":3.46,"predict_time":0.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+106\/-172","imputed":false},{"position":52,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1198.0,"score":0.003,"rank":45.43,"hrank":44.62,"imp":24.835,"train_time":5.38,"predict_time":0.624,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-77","imputed":false},{"position":53,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1196.0,"score":0.161,"rank":45.57,"hrank":34.38,"imp":26.36,"train_time":731.23,"predict_time":0.334,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+136\/-220","imputed":false},{"position":54,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1180.0,"score":0.076,"rank":46.86,"hrank":40.79,"imp":27.518,"train_time":9.4,"predict_time":1.444,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-222","imputed":false},{"position":55,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1174.0,"score":0.081,"rank":47.29,"hrank":41.85,"imp":26.198,"train_time":20.64,"predict_time":7.532,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+122\/-201","imputed":false},{"position":56,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1157.0,"score":0.071,"rank":48.57,"hrank":38.63,"imp":28.273,"train_time":2750.22,"predict_time":1.238,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+190\/-210","imputed":false},{"position":57,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1148.0,"score":0.114,"rank":49.29,"hrank":32.8,"imp":29.853,"train_time":8.73,"predict_time":0.238,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+210\/-301","imputed":false},{"position":58,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1112.0,"score":0.029,"rank":51.86,"hrank":48.3,"imp":30.859,"train_time":19.64,"predict_time":0.478,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+125\/-147","imputed":false},{"position":59,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":1082.0,"score":0.154,"rank":53.86,"hrank":31.31,"imp":37.677,"train_time":180.48,"predict_time":1.239,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+258\/-368","imputed":false},{"position":60,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1043.0,"score":0.029,"rank":56.43,"hrank":53.6,"imp":32.764,"train_time":14.88,"predict_time":0.391,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+147\/-227","imputed":false},{"position":61,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1026.0,"score":0.027,"rank":57.5,"hrank":51.9,"imp":32.455,"train_time":45499.58,"predict_time":9.024,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+172\/-259","imputed":false},{"position":62,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1012.0,"score":0.005,"rank":58.36,"hrank":54.65,"imp":32.797,"train_time":45499.58,"predict_time":43.922,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-246","imputed":false},{"position":63,"method":"OrionMSP","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.02818","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1001.0,"score":0.0,"rank":59.0,"hrank":57.35,"imp":34.626,"train_time":23.33,"predict_time":3.304,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-216","imputed":false},{"position":64,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":59.07,"hrank":56.55,"imp":39.076,"train_time":0.78,"predict_time":0.155,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-249","imputed":false},{"position":65,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":987.0,"score":0.0,"rank":59.86,"hrank":58.64,"imp":31.684,"train_time":559.69,"predict_time":266.319,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-231","imputed":false},{"position":66,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":979.0,"score":0.006,"rank":60.29,"hrank":57.57,"imp":36.662,"train_time":5.96,"predict_time":0.898,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+136\/-236","imputed":false},{"position":67,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":969.0,"score":0.019,"rank":60.86,"hrank":56.79,"imp":34.718,"train_time":8.78,"predict_time":0.688,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+167\/-260","imputed":false},{"position":68,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":962.0,"score":0.0,"rank":61.21,"hrank":59.26,"imp":41.826,"train_time":913.37,"predict_time":0.738,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-236","imputed":false},{"position":69,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":941.0,"score":0.0,"rank":62.36,"hrank":60.28,"imp":42.468,"train_time":913.37,"predict_time":0.13,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-265","imputed":false},{"position":70,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":858.0,"score":0.097,"rank":66.29,"hrank":50.2,"imp":48.983,"train_time":180.48,"predict_time":0.136,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+299\/-509","imputed":false},{"position":71,"method":"TabFlex","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.05584","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":838.0,"score":0.0,"rank":67.14,"hrank":66.66,"imp":40.756,"train_time":5.48,"predict_time":0.909,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+77\/-197","imputed":false},{"position":72,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":838.0,"score":0.0,"rank":67.14,"hrank":66.37,"imp":53.547,"train_time":1679.96,"predict_time":11.05,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+107\/-213","imputed":false},{"position":73,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":824.0,"score":0.0,"rank":67.71,"hrank":66.81,"imp":43.788,"train_time":866.21,"predict_time":9.353,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-309","imputed":false},{"position":74,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":758.0,"score":0.0,"rank":70.07,"hrank":69.58,"imp":57.429,"train_time":1679.96,"predict_time":1.824,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-259","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":751.0,"score":0.0,"rank":70.29,"hrank":69.65,"imp":49.012,"train_time":4.65,"predict_time":0.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-283","imputed":false},{"position":76,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":711.0,"score":0.0,"rank":71.5,"hrank":70.78,"imp":62.772,"train_time":273.7,"predict_time":0.508,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-357","imputed":false},{"position":77,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":261.0,"score":0.0,"rank":77.43,"hrank":77.41,"imp":70.07,"train_time":0.58,"predict_time":0.07,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-470","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/pareto_front_explorer.html
index ba1b3ee6dc11a6f64a931e70c1f453e59df63640..3d00d3a58ee7346b85648807e4ca2404c645c953 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.2793557109,"imp":23.2984151865,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3431890039,"imp":23.0418722718,"elo":1295.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2382022972,"imp":16.0698126603,"elo":1478.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.2361036908,"imp":16.993720546,"elo":1476.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2379844651,"imp":29.853244074,"elo":1147.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2739576431,"imp":25.6426013142,"elo":1222.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3347926666,"imp":26.1589529753,"elo":1222.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0483681042,"imp":26.3132231692,"elo":1233.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0286685262,"imp":26.130830267,"elo":1241.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.3103094802,"imp":24.5999577586,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8984521391,"imp":36.6621872617,"elo":979.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2943369872,"imp":27.2773995398,"elo":1246.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.872894906,"imp":24.1285441429,"elo":1324.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6878838583,"imp":34.7182521619,"elo":969.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2378249243,"imp":28.2726346167,"elo":1157.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":13.5689532413,"imp":26.2794881384,"elo":1263.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.5257108651,"imp":25.6013864916,"elo":1241.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.0866582746,"imp":20.5363232435,"elo":1429.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":3.5768582903,"imp":20.4515442453,"elo":1437.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":9.1873597509,"imp":15.29437114,"elo":1486.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":4.6200815851,"imp":21.5610551198,"elo":1457.3,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4781119128,"imp":30.8585054606,"elo":1111.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5829764758,"imp":22.1528894542,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5509605348,"imp":21.8530875524,"elo":1350.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3043972003,"imp":34.6260033877,"elo":1001.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5079486588,"imp":62.7718084569,"elo":710.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.8243625063,"imp":57.4293045949,"elo":757.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":11.050111703,"imp":53.5472107063,"elo":837.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1545608396,"imp":39.0757652771,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3337366528,"imp":26.3599301981,"elo":1196.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":2.0214591481,"imp":24.7457435227,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.5324903306,"imp":26.1984848423,"elo":1174.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":4.0293360205,"imp":17.9424813982,"elo":1484.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.781269562,"imp":17.5032291669,"elo":1524.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.855277474,"imp":13.2134098365,"elo":1594.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1322378847,"imp":14.3384135162,"elo":1656.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5456556456,"imp":5.6371518418,"elo":1742.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.556301258,"imp":22.9111586961,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":40.2005207582,"imp":26.5714387064,"elo":1233.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":40.1033683003,"imp":22.9824009828,"elo":1361.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":399.3341259005,"imp":22.0617949196,"elo":1429.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2694943851,"imp":23.6032183869,"elo":1384.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4342474097,"imp":6.1115865167,"elo":1724.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.9093813646,"imp":40.7563314744,"elo":837.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":2.1316275701,"imp":27.1930496661,"elo":1301.4,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2337945274,"imp":15.4616732914,"elo":1530.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7280240831,"imp":23.7404809364,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.8794831101,"imp":19.6046393967,"elo":1458.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.6401447093,"imp":19.5501472802,"elo":1466.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.650333935,"imp":13.6883146842,"elo":1649.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5140794313,"imp":12.3150344649,"elo":1590.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.387923994,"imp":20.6445312924,"elo":1395.8,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4768883996,"imp":22.2619911024,"elo":1383.5,"imputed":true,"imputed_pct":14.2857142857,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.8212823247,"imp":20.7084157032,"elo":1489.6,"imputed":true,"imputed_pct":14.2857142857,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3534204381,"imp":43.7881530786,"elo":823.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.0244207573,"imp":32.4548558378,"elo":1025.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":43.9219279194,"imp":32.7974122778,"elo":1011.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622328922,"imp":27.8286351915,"elo":1219.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3908251335,"imp":32.7641285508,"elo":1042.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2202658411,"imp":26.8243500098,"elo":1231.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.293854225,"imp":24.7936831007,"elo":1280.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.6237615495,"imp":24.8347797971,"elo":1197.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0653522593,"imp":20.6819141626,"elo":1382.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":5.1289419395,"imp":20.9632476496,"elo":1384.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":266.3190646898,"imp":31.6844647486,"elo":986.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":240.6831310693,"imp":22.5779018618,"elo":1286.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":986.35207523,"imp":21.2879406926,"elo":1353.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.4435953081,"imp":27.5180847905,"elo":1179.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1464664376,"imp":22.3865918703,"elo":1342.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9262072088,"imp":20.9249968398,"elo":1429.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.2793557109,"x_train":326.3219590044,"imp":23.2984151865,"elo":1349.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.3431890039,"x_train":29.5776098075,"imp":23.0418722718,"elo":1295.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.2382022972,"x_train":3632.0960320665,"imp":16.0698126603,"elo":1478.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":1.2361036908,"x_train":3632.0960320665,"imp":16.993720546,"elo":1476.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2379844651,"x_train":8.7255028029,"imp":29.853244074,"elo":1147.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2739576431,"x_train":2001.6183652265,"imp":25.6426013142,"elo":1222.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3347926666,"x_train":2001.6183652265,"imp":26.1589529753,"elo":1222.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0483681042,"x_train":10.9494219165,"imp":26.3132231692,"elo":1233.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0286685262,"x_train":27727.7037054394,"imp":26.130830267,"elo":1241.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.3103094802,"x_train":27727.7037054394,"imp":24.5999577586,"elo":1305.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8984521391,"x_train":5.955842539,"imp":36.6621872617,"elo":979.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2943369872,"x_train":1042.3872546633,"imp":27.2773995398,"elo":1246.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.872894906,"x_train":1042.3872546633,"imp":24.1285441429,"elo":1324.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6878838583,"x_train":8.7848766331,"imp":34.7182521619,"elo":969.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.2378249243,"x_train":2750.2207329287,"imp":28.2726346167,"elo":1157.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":13.5689532413,"x_train":2750.2207329287,"imp":26.2794881384,"elo":1263.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.5257108651,"x_train":5.149666856,"imp":25.6013864916,"elo":1241.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":1.0866582746,"x_train":1394.8404749323,"imp":20.5363232435,"elo":1429.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":3.5768582903,"x_train":1394.8404749323,"imp":20.4515442453,"elo":1437.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":9.1873597509,"x_train":46.1120625115,"imp":15.29437114,"elo":1486.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":4.6200815851,"x_train":202.7170180212,"imp":21.5610551198,"elo":1457.3,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4781119128,"x_train":19.6359284437,"imp":30.8585054606,"elo":1111.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5829764758,"x_train":6995.9928750372,"imp":22.1528894542,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5509605348,"x_train":6995.9928750372,"imp":21.8530875524,"elo":1350.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3043972003,"x_train":23.3325961779,"imp":34.6260033877,"elo":1001.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5079486588,"x_train":273.6950495461,"imp":62.7718084569,"elo":710.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":1.8243625063,"x_train":1679.959262267,"imp":57.4293045949,"elo":757.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":11.050111703,"x_train":1679.959262267,"imp":53.5472107063,"elo":837.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1545608396,"x_train":0.7756363005,"imp":39.0757652771,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.3337366528,"x_train":731.2261557986,"imp":26.3599301981,"elo":1196.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":2.0214591481,"x_train":731.2261557986,"imp":24.7457435227,"elo":1270.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.5324903306,"x_train":20.6393168714,"imp":26.1984848423,"elo":1174.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":4.0293360205,"x_train":4655.1453813481,"imp":17.9424813982,"elo":1484.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":21.781269562,"x_train":4655.1453813481,"imp":17.5032291669,"elo":1524.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.855277474,"x_train":10.9716481006,"imp":13.2134098365,"elo":1594.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.1322378847,"x_train":2837.9578551597,"imp":14.3384135162,"elo":1656.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5456556456,"x_train":2837.9578551597,"imp":5.6371518418,"elo":1742.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.556301258,"x_train":22.2617203941,"imp":22.9111586961,"elo":1363.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":40.2005207582,"x_train":51.5168795924,"imp":26.5714387064,"elo":1233.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":40.1033683003,"x_train":10182.2577368658,"imp":22.9824009828,"elo":1361.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":399.3341259005,"x_train":10182.2577368658,"imp":22.0617949196,"elo":1429.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2694943851,"x_train":5.6498915817,"imp":23.6032183869,"elo":1384.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4342474097,"x_train":52.6612506816,"imp":6.1115865167,"elo":1724.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFlex","family":"Foundation Model","x_infer":0.9093813646,"x_train":5.4835182137,"imp":40.7563314744,"elo":837.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICL","family":"Foundation Model","x_infer":2.1316275701,"x_train":11.7044641808,"imp":27.1930496661,"elo":1301.4,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2337945274,"x_train":5.3571462631,"imp":15.4616732914,"elo":1530.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.7280240831,"x_train":17.3386025246,"imp":23.7404809364,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.8794831101,"x_train":3829.3977883333,"imp":19.6046393967,"elo":1458.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.6401447093,"x_train":3829.3977883333,"imp":19.5501472802,"elo":1466.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.650333935,"x_train":10.7776505002,"imp":13.6883146842,"elo":1649.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5140794313,"x_train":9.4477958531,"imp":12.3150344649,"elo":1590.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.387923994,"x_train":6.0031297644,"imp":20.6445312924,"elo":1395.8,"imputed":true,"imputed_pct":14.2857142857,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.4768883996,"x_train":2971.3159884536,"imp":22.2619911024,"elo":1383.5,"imputed":true,"imputed_pct":14.2857142857,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.8212823247,"x_train":2971.3159884536,"imp":20.7084157032,"elo":1489.6,"imputed":true,"imputed_pct":14.2857142857,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3534204381,"x_train":866.2091617048,"imp":43.7881530786,"elo":823.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.0244207573,"x_train":45499.5818819746,"imp":32.4548558378,"elo":1025.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":43.9219279194,"x_train":45499.5818819746,"imp":32.7974122778,"elo":1011.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622328922,"x_train":3.4571113671,"imp":27.8286351915,"elo":1219.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3908251335,"x_train":14.8754458124,"imp":32.7641285508,"elo":1042.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2202658411,"x_train":3834.5684419838,"imp":26.8243500098,"elo":1231.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":2.293854225,"x_train":3834.5684419838,"imp":24.7936831007,"elo":1280.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.6237615495,"x_train":5.3772386715,"imp":24.8347797971,"elo":1197.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":1.0653522593,"x_train":2342.5781829117,"imp":20.6819141626,"elo":1382.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":5.1289419395,"x_train":2342.5781829117,"imp":20.9632476496,"elo":1384.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":266.3190646898,"x_train":559.6852708101,"imp":31.6844647486,"elo":986.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":240.6831310693,"x_train":34015.1930441921,"imp":22.5779018618,"elo":1286.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":986.35207523,"x_train":34015.1930441921,"imp":21.2879406926,"elo":1353.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.4435953081,"x_train":9.4039762628,"imp":27.5180847905,"elo":1179.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.1464664376,"x_train":2149.2969294609,"imp":22.3865918703,"elo":1342.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.9262072088,"x_train":2149.2969294609,"imp":20.9249968398,"elo":1429.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
index 51fe96353fa5f76aa7ae1a8060e3fd43304783ed..6bd865e50e013f4c9319022b868921453fbf8817 100644
--- a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_multiclass/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_train":326.3219590044,"imp":23.3197524236,"elo":1341.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":29.5776098075,"imp":23.0628255508,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":61.6242204097,"imp":20.6259991536,"elo":1380.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":146.2103635918,"imp":19.5055911201,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":255.2797833412,"imp":18.8193637339,"elo":1449.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":461.9229712959,"imp":18.2731442985,"elo":1463.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":901.1280977162,"imp":18.2852095466,"elo":1457.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1816.8097050761,"imp":17.5175425789,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2717.6279686964,"imp":17.3028855872,"elo":1458.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":3632.0960320665,"imp":17.0155156555,"elo":1469.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":8.7255028029,"imp":29.8723469778,"elo":1136.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":18.8295905976,"imp":29.4165301447,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":48.1908367108,"imp":29.1397802936,"elo":1153.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":97.2610688395,"imp":28.5897347697,"elo":1167.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":247.1661812847,"imp":27.5299606151,"elo":1182.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":498.9992240184,"imp":27.0619964647,"elo":1199.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":996.7852549912,"imp":26.5530110316,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1493.1381281887,"imp":26.33390516,"elo":1209.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2001.6183652265,"imp":26.1789896298,"elo":1218.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":10.9494219165,"imp":26.333260115,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":116.2996116626,"imp":26.0861694223,"elo":1217.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":481.3286902295,"imp":25.9913850731,"elo":1226.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":1094.9483089107,"imp":25.9632540248,"elo":1230.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":2930.5133530918,"imp":25.7635733094,"elo":1238.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":7066.1129355898,"imp":25.5121406641,"elo":1243.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":13790.4831017652,"imp":25.0032650117,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":20842.8750821883,"imp":24.7834516827,"elo":1288.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":27727.7037054394,"imp":24.6235143359,"elo":1292.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":5.955842539,"imp":36.6903295327,"elo":967.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":11.0977776932,"imp":33.0189764248,"elo":1046.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":26.6390299573,"imp":27.7519255431,"elo":1155.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":52.8400597922,"imp":26.5153909584,"elo":1221.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":130.2628378892,"imp":24.8026985651,"elo":1299.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":259.6675727983,"imp":24.6271917138,"elo":1303.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":518.6557637866,"imp":24.4219453292,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":777.9260343247,"imp":24.2106018828,"elo":1319.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1042.3872546633,"imp":24.1716864028,"elo":1321.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":8.7848766331,"imp":34.7368118298,"elo":963.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":23.0929542575,"imp":33.268693352,"elo":1010.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":64.8591819071,"imp":30.131462827,"elo":1119.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":131.6977426498,"imp":29.0035091083,"elo":1180.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":335.6638780889,"imp":27.4324973088,"elo":1219.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":677.7524464582,"imp":26.7236363263,"elo":1243.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1359.9623914477,"imp":26.3400526889,"elo":1263.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2046.8674083081,"imp":26.3031948541,"elo":1265.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2750.2207329287,"imp":26.3011355079,"elo":1264.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":5.149666856,"imp":25.6199848887,"elo":1236.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":12.638683475,"imp":24.5398838227,"elo":1276.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":32.1130206474,"imp":23.4160350344,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":66.7832587484,"imp":22.5663410197,"elo":1375.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":172.6401098684,"imp":21.5767763686,"elo":1396.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":345.9193991848,"imp":21.2445989196,"elo":1410.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":693.5342730413,"imp":20.8902882844,"elo":1433.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":1039.0554015758,"imp":20.7170294095,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1394.8404749323,"imp":20.4720931948,"elo":1444.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":46.1120625115,"imp":15.5115935998,"elo":1497.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":202.7170180212,"imp":21.5820362171,"elo":1473.8,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":19.6359284437,"imp":30.8866944985,"elo":1105.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":54.5342406386,"imp":27.999554373,"elo":1160.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":167.3188814038,"imp":24.6714431267,"elo":1266.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":346.6206170635,"imp":23.8917054055,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":852.383981166,"imp":23.3257705825,"elo":1324.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":1740.145368883,"imp":22.8932583059,"elo":1331.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":3492.7302851634,"imp":22.9195960872,"elo":1324.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":5232.8252409525,"imp":22.2757576311,"elo":1344.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":6995.9928750372,"imp":21.8737738633,"elo":1352.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_train":23.3325961779,"imp":34.6489122936,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":273.6950495461,"imp":62.8288020694,"elo":682.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":577.7198455713,"imp":57.1996938609,"elo":772.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1420.3840579116,"imp":53.7191489746,"elo":828.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1679.959262267,"imp":53.5547287705,"elo":826.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.7756363005,"imp":39.1605432301,"elo":993.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.64517611,"imp":31.9073276602,"elo":1077.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":15.6783968263,"imp":27.666405584,"elo":1152.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":34.2083641857,"imp":25.4184350204,"elo":1189.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":88.9128109825,"imp":24.1606702604,"elo":1274.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":180.8451833376,"imp":23.9756519157,"elo":1279.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":363.1699124457,"imp":24.4008408933,"elo":1274.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":545.3863630441,"imp":24.593197693,"elo":1275.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":731.2261557986,"imp":24.7680738423,"elo":1264.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":20.6393168714,"imp":26.217618248,"elo":1169.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":44.6022278057,"imp":21.4343579901,"elo":1339.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":114.6509525781,"imp":19.1477629466,"elo":1432.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":230.7629821174,"imp":17.7310879223,"elo":1491.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":577.5814409002,"imp":16.8441666583,"elo":1539.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1157.7764691244,"imp":16.3649374762,"elo":1561.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2306.7724282356,"imp":16.3773806269,"elo":1566.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3460.7466523769,"imp":17.0419733271,"elo":1545.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4655.1453813481,"imp":17.5250924206,"elo":1534.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":10.9716481006,"imp":13.2345164804,"elo":1593.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":27.742192784,"imp":13.0764507206,"elo":1594.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":70.0993253105,"imp":12.7634150165,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":140.813217678,"imp":12.8115263622,"elo":1692.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":354.18227158,"imp":12.6756223603,"elo":1713.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":706.681175239,"imp":12.1708307979,"elo":1734.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1422.0173619203,"imp":10.6463681559,"elo":1741.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2134.9546013675,"imp":9.097776014,"elo":1748.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2837.9578551597,"imp":5.6613192609,"elo":1758.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":22.2617203941,"imp":22.9324150573,"elo":1357.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.5168795924,"imp":26.5911909892,"elo":1219.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":99.6707673609,"imp":26.1279843826,"elo":1247.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":250.1259988665,"imp":25.2381687184,"elo":1309.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":520.4892551291,"imp":24.7900103576,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":1282.2422972143,"imp":23.8281813917,"elo":1382.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":2545.8074097094,"imp":23.1621024131,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":5076.4052782117,"imp":22.9578764419,"elo":1415.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":7591.7531128733,"imp":22.6275102214,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":10182.2577368658,"imp":22.0707725769,"elo":1433.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":5.6498915817,"imp":23.6235096837,"elo":1381.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":52.6612506816,"imp":6.1334504678,"elo":1715.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_train":5.4835182137,"imp":40.7747113143,"elo":834.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_train":11.7044641808,"imp":27.2129727271,"elo":1291.0,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":5.3571462631,"imp":15.4823634336,"elo":1533.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":17.3386025246,"imp":23.7623042298,"elo":1364.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":34.7857185326,"imp":23.0715461263,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":92.2135159049,"imp":22.5165281719,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":188.1504220603,"imp":21.9549767719,"elo":1434.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":467.3246962797,"imp":21.5861201405,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":939.5188242924,"imp":21.1738829848,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1881.8343045706,"imp":20.3622472104,"elo":1463.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2843.5213469349,"imp":20.0176345562,"elo":1469.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3829.3977883333,"imp":19.5717472605,"elo":1472.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":10.7776505002,"imp":13.7096329452,"elo":1665.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":9.4477958531,"imp":12.335908483,"elo":1595.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":6.0031297644,"imp":20.6648713052,"elo":1397.8,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":25.5665239353,"imp":21.613161278,"elo":1412.9,"imputed":true,"imputed_pct":14.2857142857,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":68.47459419,"imp":21.4409099036,"elo":1457.6,"imputed":true,"imputed_pct":14.2857142857,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":145.8870940856,"imp":20.8513142313,"elo":1485.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":368.5308235658,"imp":20.3622002991,"elo":1508.0,"imputed":true,"imputed_pct":14.2857142857,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":735.5036358061,"imp":20.0258601958,"elo":1529.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1478.9026734027,"imp":20.1589609829,"elo":1523.1,"imputed":true,"imputed_pct":14.2857142857,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2215.4355400323,"imp":20.3601286851,"elo":1512.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2971.3159884536,"imp":20.6875659185,"elo":1498.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":866.2091617048,"imp":43.8044667741,"elo":799.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1369.7270777514,"imp":43.5405020964,"elo":822.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1966.5345052574,"imp":42.7742107507,"elo":842.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2789.3976815102,"imp":41.2081636838,"elo":865.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5422.0998389164,"imp":37.967863537,"elo":921.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11167.1140617992,"imp":35.316753412,"elo":954.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23229.2182271869,"imp":34.1998275425,"elo":979.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34425.4477297998,"imp":33.6322122465,"elo":996.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":45499.5818819746,"imp":32.815561821,"elo":1020.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":3.4571113671,"imp":27.8478694128,"elo":1216.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.8754458124,"imp":32.8102812646,"elo":1028.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":30.9336251788,"imp":29.2987415582,"elo":1122.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":94.1324659798,"imp":26.948093656,"elo":1205.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":200.5394916928,"imp":26.1769515242,"elo":1258.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":461.9575980829,"imp":25.7197189536,"elo":1262.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":911.8260308778,"imp":25.4028650756,"elo":1266.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":1835.2639428382,"imp":25.1008499739,"elo":1270.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":2805.3782071109,"imp":24.8395744398,"elo":1283.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":3834.5684419838,"imp":24.817488205,"elo":1278.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":5.3772386715,"imp":24.8541959881,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":17.3939236522,"imp":23.4295375178,"elo":1268.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":53.4704779379,"imp":22.9359564559,"elo":1299.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":108.3183028113,"imp":22.3257100045,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":286.0649937532,"imp":21.830401136,"elo":1350.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":578.8859474941,"imp":21.5075500365,"elo":1365.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":1164.3479303143,"imp":21.3354239659,"elo":1370.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":1743.9656718772,"imp":21.2302690665,"elo":1379.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":2342.5781829117,"imp":20.9835399671,"elo":1382.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":559.6852708101,"imp":31.7031001295,"elo":975.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":1879.2300695283,"imp":28.8909436091,"elo":1049.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":5685.7541761762,"imp":25.4340809006,"elo":1188.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":12716.1318239357,"imp":23.392764892,"elo":1265.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":32397.1155311761,"imp":21.2766142176,"elo":1351.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":34015.1930441921,"imp":21.3085338575,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":9.4039762628,"imp":27.5376238303,"elo":1166.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":19.9957554462,"imp":26.4014081913,"elo":1230.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":52.7123918422,"imp":24.7132555183,"elo":1315.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":106.7857168731,"imp":23.855313907,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":267.72178804,"imp":22.3741327725,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":532.9559679306,"imp":21.4318530228,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":1070.1666001206,"imp":21.0251373274,"elo":1433.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1601.8304365569,"imp":20.9463925049,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":2149.2969294609,"imp":20.9647036579,"elo":1441.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"BetaTabPFN","family":"Foundation Model","x_infer":4.2793557109,"x_train":326.3219590044,"imp":23.3197524236,"elo":1341.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3431890039,"x_train":29.5776098075,"imp":23.0628255508,"elo":1290.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.5366471716,"x_train":61.6242204097,"imp":20.6259991536,"elo":1380.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.7424350969,"x_train":146.2103635918,"imp":19.5055911201,"elo":1423.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.9011323043,"x_train":255.2797833412,"imp":18.8193637339,"elo":1449.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.9752719466,"x_train":461.9229712959,"imp":18.2731442985,"elo":1463.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1210385171,"x_train":901.1280977162,"imp":18.2852095466,"elo":1457.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.1513120746,"x_train":1816.8097050761,"imp":17.5175425789,"elo":1467.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2237069108,"x_train":2717.6279686964,"imp":17.3028855872,"elo":1458.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":1.2361036908,"x_train":3632.0960320665,"imp":17.0155156555,"elo":1469.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2379844651,"x_train":8.7255028029,"imp":29.8723469778,"elo":1136.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3316836793,"x_train":18.8295905976,"imp":29.4165301447,"elo":1145.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.5417567135,"x_train":48.1908367108,"imp":29.1397802936,"elo":1153.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.7237500996,"x_train":97.2610688395,"imp":28.5897347697,"elo":1167.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.891756985,"x_train":247.1661812847,"imp":27.5299606151,"elo":1182.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.0241181517,"x_train":498.9992240184,"imp":27.0619964647,"elo":1199.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.1473050403,"x_train":996.7852549912,"imp":26.5530110316,"elo":1206.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.2443893534,"x_train":1493.1381281887,"imp":26.33390516,"elo":1209.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":1.3347926666,"x_train":2001.6183652265,"imp":26.1789896298,"elo":1218.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0483681042,"x_train":10.9494219165,"imp":26.333260115,"elo":1215.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.1342425713,"x_train":116.2996116626,"imp":26.0861694223,"elo":1217.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.2180498533,"x_train":481.3286902295,"imp":25.9913850731,"elo":1226.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.207248058,"x_train":1094.9483089107,"imp":25.9632540248,"elo":1230.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.2016961442,"x_train":2930.5133530918,"imp":25.7635733094,"elo":1238.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.2168105015,"x_train":7066.1129355898,"imp":25.5121406641,"elo":1243.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.2240727367,"x_train":13790.4831017652,"imp":25.0032650117,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.2523087712,"x_train":20842.8750821883,"imp":24.7834516827,"elo":1288.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.3103094802,"x_train":27727.7037054394,"imp":24.6235143359,"elo":1292.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8984521391,"x_train":5.955842539,"imp":36.6903295327,"elo":967.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1872124249,"x_train":11.0977776932,"imp":33.0189764248,"elo":1046.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9761079981,"x_train":26.6390299573,"imp":27.7519255431,"elo":1155.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.2785129959,"x_train":52.8400597922,"imp":26.5153909584,"elo":1221.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.703820181,"x_train":130.2628378892,"imp":24.8026985651,"elo":1299.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.4670921859,"x_train":259.6675727983,"imp":24.6271917138,"elo":1303.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7516614998,"x_train":518.6557637866,"imp":24.4219453292,"elo":1308.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7521164634,"x_train":777.9260343247,"imp":24.2106018828,"elo":1319.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.872894906,"x_train":1042.3872546633,"imp":24.1716864028,"elo":1321.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6878838583,"x_train":8.7848766331,"imp":34.7368118298,"elo":963.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.5412796889,"x_train":23.0929542575,"imp":33.268693352,"elo":1010.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.1850045276,"x_train":64.8591819071,"imp":30.131462827,"elo":1119.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":4.2907795257,"x_train":131.6977426498,"imp":29.0035091083,"elo":1180.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.7270220551,"x_train":335.6638780889,"imp":27.4324973088,"elo":1219.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":8.095832519,"x_train":677.7524464582,"imp":26.7236363263,"elo":1243.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":10.3587654477,"x_train":1359.9623914477,"imp":26.3400526889,"elo":1263.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":11.4534017479,"x_train":2046.8674083081,"imp":26.3031948541,"elo":1265.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":13.5689532413,"x_train":2750.2207329287,"imp":26.3011355079,"elo":1264.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.5257108651,"x_train":5.149666856,"imp":25.6199848887,"elo":1236.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.7506945395,"x_train":12.638683475,"imp":24.5398838227,"elo":1276.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.7230234059,"x_train":32.1130206474,"imp":23.4160350344,"elo":1332.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.8714779904,"x_train":66.7832587484,"imp":22.5663410197,"elo":1375.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.11547675,"x_train":172.6401098684,"imp":21.5767763686,"elo":1396.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.59344747,"x_train":345.9193991848,"imp":21.2445989196,"elo":1410.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.9034755619,"x_train":693.5342730413,"imp":20.8902882844,"elo":1433.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.7091890428,"x_train":1039.0554015758,"imp":20.7170294095,"elo":1436.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.5768582903,"x_train":1394.8404749323,"imp":20.4720931948,"elo":1444.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":9.1873597509,"x_train":46.1120625115,"imp":15.5115935998,"elo":1497.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":4.6200815851,"x_train":202.7170180212,"imp":21.5820362171,"elo":1473.8,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4781119128,"x_train":19.6359284437,"imp":30.8866944985,"elo":1105.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.0130189541,"x_train":54.5342406386,"imp":27.999554373,"elo":1160.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.8126811516,"x_train":167.3188814038,"imp":24.6714431267,"elo":1266.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.8125941214,"x_train":346.6206170635,"imp":23.8917054055,"elo":1297.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.4546068712,"x_train":852.383981166,"imp":23.3257705825,"elo":1324.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.9518046583,"x_train":1740.145368883,"imp":22.8932583059,"elo":1331.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.3693694624,"x_train":3492.7302851634,"imp":22.9195960872,"elo":1324.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.5205578648,"x_train":5232.8252409525,"imp":22.2757576311,"elo":1344.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":9.5509605348,"x_train":6995.9928750372,"imp":21.8737738633,"elo":1352.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"OrionMSP","family":"Foundation Model","x_infer":3.3043972003,"x_train":23.3325961779,"imp":34.6489122936,"elo":999.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.5079486588,"x_train":273.6950495461,"imp":62.8288020694,"elo":682.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":7.5114343904,"x_train":577.7198455713,"imp":57.1996938609,"elo":772.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":12.1599901492,"x_train":1420.3840579116,"imp":53.7191489746,"elo":828.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":11.050111703,"x_train":1679.959262267,"imp":53.5547287705,"elo":826.8,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1545608396,"x_train":0.7756363005,"imp":39.1605432301,"elo":993.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4758379412,"x_train":4.64517611,"imp":31.9073276602,"elo":1077.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.8964239223,"x_train":15.6783968263,"imp":27.666405584,"elo":1152.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2954751768,"x_train":34.2083641857,"imp":25.4184350204,"elo":1189.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.5928379977,"x_train":88.9128109825,"imp":24.1606702604,"elo":1274.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7104260673,"x_train":180.8451833376,"imp":23.9756519157,"elo":1279.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7486850861,"x_train":363.1699124457,"imp":24.4008408933,"elo":1274.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.7684293692,"x_train":545.3863630441,"imp":24.593197693,"elo":1275.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":2.0214591481,"x_train":731.2261557986,"imp":24.7680738423,"elo":1264.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.5324903306,"x_train":20.6393168714,"imp":26.217618248,"elo":1169.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.4467410735,"x_train":44.6022278057,"imp":21.4343579901,"elo":1339.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":12.4636747092,"x_train":114.6509525781,"imp":19.1477629466,"elo":1432.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":13.7310288169,"x_train":230.7629821174,"imp":17.7310879223,"elo":1491.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.1150100532,"x_train":577.5814409002,"imp":16.8441666583,"elo":1539.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":14.7078734544,"x_train":1157.7764691244,"imp":16.3649374762,"elo":1561.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":17.3622951828,"x_train":2306.7724282356,"imp":16.3773806269,"elo":1566.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.0724225096,"x_train":3460.7466523769,"imp":17.0419733271,"elo":1545.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":21.781269562,"x_train":4655.1453813481,"imp":17.5250924206,"elo":1534.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.855277474,"x_train":10.9716481006,"imp":13.2345164804,"elo":1593.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.2634027639,"x_train":27.742192784,"imp":13.0764507206,"elo":1594.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.5574648521,"x_train":70.0993253105,"imp":12.7634150165,"elo":1637.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.4891965121,"x_train":140.813217678,"imp":12.8115263622,"elo":1692.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.4282035488,"x_train":354.18227158,"imp":12.6756223603,"elo":1713.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.8048742763,"x_train":706.681175239,"imp":12.1708307979,"elo":1734.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.3192714939,"x_train":1422.0173619203,"imp":10.6463681559,"elo":1741.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.1275877868,"x_train":2134.9546013675,"imp":9.097776014,"elo":1748.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.5456556456,"x_train":2837.9578551597,"imp":5.6613192609,"elo":1758.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":1.556301258,"x_train":22.2617203941,"imp":22.9324150573,"elo":1357.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":40.2005207582,"x_train":51.5168795924,"imp":26.5911909892,"elo":1219.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":66.1274474954,"x_train":99.6707673609,"imp":26.1279843826,"elo":1247.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":107.1341355034,"x_train":250.1259988665,"imp":25.2381687184,"elo":1309.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":145.95289892,"x_train":520.4892551291,"imp":24.7900103576,"elo":1342.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":207.3229714868,"x_train":1282.2422972143,"imp":23.8281813917,"elo":1382.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":226.2356139625,"x_train":2545.8074097094,"imp":23.1621024131,"elo":1405.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":283.6685778956,"x_train":5076.4052782117,"imp":22.9578764419,"elo":1415.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":335.2555385893,"x_train":7591.7531128733,"imp":22.6275102214,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":399.3341259005,"x_train":10182.2577368658,"imp":22.0707725769,"elo":1433.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2694943851,"x_train":5.6498915817,"imp":23.6235096837,"elo":1381.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.4342474097,"x_train":52.6612506816,"imp":6.1334504678,"elo":1715.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFlex","family":"Foundation Model","x_infer":0.9093813646,"x_train":5.4835182137,"imp":40.7747113143,"elo":834.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICL","family":"Foundation Model","x_infer":2.1316275701,"x_train":11.7044641808,"imp":27.2129727271,"elo":1291.0,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.2337945274,"x_train":5.3571462631,"imp":15.4823634336,"elo":1533.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.7280240831,"x_train":17.3386025246,"imp":23.7623042298,"elo":1364.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":1.1443311646,"x_train":34.7857185326,"imp":23.0715461263,"elo":1392.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.9947973581,"x_train":92.2135159049,"imp":22.5165281719,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.7539019401,"x_train":188.1504220603,"imp":21.9549767719,"elo":1434.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":3.0347115346,"x_train":467.3246962797,"imp":21.5861201405,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.4643032105,"x_train":939.5188242924,"imp":21.1738829848,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":3.9883342671,"x_train":1881.8343045706,"imp":20.3622472104,"elo":1463.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.4674678667,"x_train":2843.5213469349,"imp":20.0176345562,"elo":1469.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.6401447093,"x_train":3829.3977883333,"imp":19.5717472605,"elo":1472.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.650333935,"x_train":10.7776505002,"imp":13.7096329452,"elo":1665.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.5140794313,"x_train":9.4477958531,"imp":12.335908483,"elo":1595.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.387923994,"x_train":6.0031297644,"imp":20.6648713052,"elo":1397.8,"imputed":true,"imputed_pct":14.2857142857,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":3.396345592,"x_train":25.5665239353,"imp":21.613161278,"elo":1412.9,"imputed":true,"imputed_pct":14.2857142857,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":5.523707633,"x_train":68.47459419,"imp":21.4409099036,"elo":1457.6,"imputed":true,"imputed_pct":14.2857142857,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":6.9457862857,"x_train":145.8870940856,"imp":20.8513142313,"elo":1485.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":9.8989073341,"x_train":368.5308235658,"imp":20.3622002991,"elo":1508.0,"imputed":true,"imputed_pct":14.2857142857,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.2340400543,"x_train":735.5036358061,"imp":20.0258601958,"elo":1529.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":14.317762499,"x_train":1478.9026734027,"imp":20.1589609829,"elo":1523.1,"imputed":true,"imputed_pct":14.2857142857,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":16.699154772,"x_train":2215.4355400323,"imp":20.3601286851,"elo":1512.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":17.8212823247,"x_train":2971.3159884536,"imp":20.6875659185,"elo":1498.2,"imputed":true,"imputed_pct":14.2857142857,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.3534204381,"x_train":866.2091617048,"imp":43.8044667741,"elo":799.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":11.7604054174,"x_train":1369.7270777514,"imp":43.5405020964,"elo":822.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":13.9844555509,"x_train":1966.5345052574,"imp":42.7742107507,"elo":842.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":16.4034660481,"x_train":2789.3976815102,"imp":41.2081636838,"elo":865.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":19.4052999829,"x_train":5422.0998389164,"imp":37.967863537,"elo":921.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.7782915986,"x_train":11167.1140617992,"imp":35.316753412,"elo":954.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":26.7442737345,"x_train":23229.2182271869,"imp":34.1998275425,"elo":979.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":31.7906715653,"x_train":34425.4477297998,"imp":33.6322122465,"elo":996.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":43.9219279194,"x_train":45499.5818819746,"imp":32.815561821,"elo":1020.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0622328922,"x_train":3.4571113671,"imp":27.8478694128,"elo":1216.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3908251335,"x_train":14.8754458124,"imp":32.8102812646,"elo":1028.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7508941509,"x_train":30.9336251788,"imp":29.2987415582,"elo":1122.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2080831708,"x_train":94.1324659798,"imp":26.948093656,"elo":1205.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2008541117,"x_train":200.5394916928,"imp":26.1769515242,"elo":1258.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2778500539,"x_train":461.9575980829,"imp":25.7197189536,"elo":1262.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4463341877,"x_train":911.8260308778,"imp":25.4028650756,"elo":1266.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6817828504,"x_train":1835.2639428382,"imp":25.1008499739,"elo":1270.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.0166546965,"x_train":2805.3782071109,"imp":24.8395744398,"elo":1283.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":2.293854225,"x_train":3834.5684419838,"imp":24.817488205,"elo":1278.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6237615495,"x_train":5.3772386715,"imp":24.8541959881,"elo":1186.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.8908831859,"x_train":17.3939236522,"imp":23.4295375178,"elo":1268.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.8059237272,"x_train":53.4704779379,"imp":22.9359564559,"elo":1299.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.2721762964,"x_train":108.3183028113,"imp":22.3257100045,"elo":1325.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":4.6366016842,"x_train":286.0649937532,"imp":21.830401136,"elo":1350.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":5.2825632775,"x_train":578.8859474941,"imp":21.5075500365,"elo":1365.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":6.035961193,"x_train":1164.3479303143,"imp":21.3354239659,"elo":1370.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":5.2659142989,"x_train":1743.9656718772,"imp":21.2302690665,"elo":1379.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":5.1289419395,"x_train":2342.5781829117,"imp":20.9835399671,"elo":1382.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":266.3190646898,"x_train":559.6852708101,"imp":31.7031001295,"elo":975.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":588.4423040375,"x_train":1879.2300695283,"imp":28.8909436091,"elo":1049.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":968.8511096453,"x_train":5685.7541761762,"imp":25.4340809006,"elo":1188.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":1043.4500106951,"x_train":12716.1318239357,"imp":23.392764892,"elo":1265.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":1036.6685407389,"x_train":32397.1155311761,"imp":21.2766142176,"elo":1351.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":986.35207523,"x_train":34015.1930441921,"imp":21.3085338575,"elo":1349.3,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.4435953081,"x_train":9.4039762628,"imp":27.5376238303,"elo":1166.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.6504460427,"x_train":19.9957554462,"imp":26.4014081913,"elo":1230.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.9062277071,"x_train":52.7123918422,"imp":24.7132555183,"elo":1315.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.9180079213,"x_train":106.7857168731,"imp":23.855313907,"elo":1337.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.0548434429,"x_train":267.72178804,"imp":22.3741327725,"elo":1386.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.5313546067,"x_train":532.9559679306,"imp":21.4318530228,"elo":1420.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":2.3070752733,"x_train":1070.1666001206,"imp":21.0251373274,"elo":1433.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.2264927788,"x_train":1601.8304365569,"imp":20.9463925049,"elo":1439.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.9262072088,"x_train":2149.2969294609,"imp":20.9647036579,"elo":1441.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..f2b1cb607a3ab0ded3545b31d57fec11754ab64d
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_multiclass/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["RealTabPFN-2.5 (tuned + ensembled)", "AutoGluon 1.5 (extreme, 4h)", "TabFM (default)", "RealTabPFN-2.5 (tuned)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (default)", "TabPFN-3 (default)", "TabICLv2 (default)", "RealMLP (tuned + ensembled)", "TabPFNv2 (tuned + ensembled)", "LimiX (default)", "RealMLP (tuned)", "AutoGluon 1.4 (best, 4h)", "CatBoost (tuned)", "CatBoost (tuned + ensembled)", "TabM (tuned + ensembled)", "TabM (tuned)", "Mitra (default)", "LightGBM (tuned + ensembled)", "TabDPT (tuned + ensembled)", "xRFM (tuned + ensembled)", "LightGBM (tuned)", "TabPFNv2 (default)", "TabDPT-Turbo (default)", "XGBoost (tuned + ensembled)", "TabPFNv2 (tuned)", "XGBoost (tuned)", "TabM (default)", "SAP-RPT-OSS (default)", "TabDPT (tuned)", "iLTM (tuned + ensembled)", "ModernNCA (tuned + ensembled)", "BetaTabPFN (default)", "xRFM (tuned)", "ModernNCA (tuned)", "ExtraTrees (tuned + ensembled)", "EBM (tuned + ensembled)", "TabICL (default)", "CatBoost (default)", "iLTM (tuned)", "TorchMLP (tuned + ensembled)", "RandomForest (tuned + ensembled)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (tuned)", "EBM (tuned)", "LightGBM (default)", "EBM (default)", "TabDPT (default)", "TorchMLP (tuned)", "ChimeraBoost (tuned)", "ChimeraBoost (tuned + ensembled)", "TabSwift (default)", "XGBoost (default)", "RandomForest (tuned)", "xRFM (default)", "RealMLP (default)", "FastaiMLP (tuned)", "ChimeraBoost (default)", "ModernNCA (default)", "KNN (tuned + ensembled)", "TorchMLP (default)", "TabSTAR (tuned)", "TabSTAR (tuned + ensembled)", "OrionMSP (default)", "RandomForest (default)", "iLTM (default)", "ExtraTrees (default)", "FastaiMLP (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "KNN (tuned)", "TabFlex (default)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (default)", "PerpetualBooster (tuned)", "Linear (default)", "PerpetualBooster (default)", "KNN (default)"], "matrix": [[NaN, 0.7142857313156128, 0.2857142984867096, 0.7142857313156128, 0.7142857313156128, 1.0, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.7142857313156128, 1.0, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, NaN, 0.1428571492433548, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 0.7142857313156128, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.7142857313156128, 0.8571428656578064, NaN, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.2857142984867096, 0.1428571492433548, NaN, 0.5714285969734192, 0.6428571343421936, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 1.0, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, NaN, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.7142857313156128, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.5714285969734192, 0.1428571492433548, 0.3571428656578064, 0.2857142984867096, NaN, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.5714285969734192, 0.8571428656578064, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 1.0, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.2857142984867096, 0.0, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, NaN, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, NaN, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7857142686843872, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, NaN, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, NaN, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, NaN, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.7142857313156128, 0.7142857313156128, 0.2857142984867096, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 1.0, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, NaN, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 1.0, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.6428571343421936, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.2857142984867096, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.6428571343421936, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, NaN, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.3571428656578064, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.9285714030265808, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, NaN, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2142857164144516, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, NaN, 0.4285714328289032, 0.7142857313156128, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, NaN, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0], [0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, NaN, 0.5714285969734192, 0.7142857313156128, 0.6428571343421936, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 1.0, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, NaN, 0.2857142984867096, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.4285714328289032, 0.8571428656578064, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 1.0, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.0, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.7142857313156128, NaN, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, NaN, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.8571428656578064, 0.4285714328289032, 0.7857142686843872, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, NaN, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.0, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.3571428656578064, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7857142686843872, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, NaN, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.4285714328289032, 0.0, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, NaN, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 1.0], [0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.1428571492433548, 0.2857142984867096, 0.2142857164144516, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 1.0, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.0, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, NaN, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.7142857313156128, NaN, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, NaN, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, NaN, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0], [0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.1428571492433548, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 1.0, 0.8571428656578064, 1.0, 1.0, 1.0, 0.8571428656578064, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.7142857313156128, 1.0, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, NaN, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.4285714328289032, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0, 0.8571428656578064, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0, 0.8571428656578064, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.0, 0.4285714328289032, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, NaN, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.8571428656578064, 1.0, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 1.0, 0.5714285969734192, 0.8571428656578064, 0.4285714328289032, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, NaN, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, NaN, 0.3571428656578064, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.6428571343421936, NaN, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 1.0, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.4285714328289032, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, NaN, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0714285746216774, 0.0, 0.0, 0.0, 0.0, 0.0714285746216774, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2142857164144516, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.1428571492433548, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.2857142984867096, NaN, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0], [0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, NaN, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.8571428656578064, 0.8571428656578064, 0.7142857313156128, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, NaN, 0.7142857313156128, 0.4285714328289032, 0.4285714328289032, 0.8571428656578064, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.8571428656578064, 0.7142857313156128, 0.8571428656578064, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, NaN, 0.5714285969734192, 0.7142857313156128, 0.7142857313156128, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.4285714328289032, NaN, 0.6428571343421936, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 1.0, 0.7142857313156128, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.0, 0.0, 0.2857142984867096, 0.0, 0.2857142984867096, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.0, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.2857142984867096, 0.0, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.5714285969734192, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.5714285969734192, 0.2857142984867096, 0.3571428656578064, NaN, 0.8571428656578064, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.5714285969734192, 0.8571428656578064, 0.7142857313156128, 1.0], [0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, NaN, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.5714285969734192, 0.2857142984867096, 0.5714285969734192, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, NaN, 0.4285714328289032, 0.7142857313156128, 0.4285714328289032, 0.7142857313156128, 0.5714285969734192, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.5714285969734192, NaN, 0.2857142984867096, 0.8571428656578064, 0.5714285969734192, 1.0, 0.8571428656578064], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.0, 0.4285714328289032, 0.5714285969734192, 0.1428571492433548, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.7142857313156128, 0.2857142984867096, 0.7142857313156128, NaN, 0.7142857313156128, 0.7142857313156128, 0.7142857313156128, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.4285714328289032, 0.5714285969734192, 0.1428571492433548, 0.2857142984867096, NaN, 0.5714285969734192, 0.6428571343421936, 0.8571428656578064], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.2857142984867096, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.2857142984867096, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.1428571492433548, 0.0, 0.1428571492433548, 0.7142857313156128, 0.2857142984867096, 0.4285714328289032, 0.2857142984867096, 0.4285714328289032, NaN, 0.5714285969734192, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.0, 0.0, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.1428571492433548, 0.0, 0.1428571492433548, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.0, 0.1428571492433548, 0.2857142984867096, 0.2857142984867096, 0.2857142984867096, 0.4285714328289032, 0.4285714328289032, 0.0, 0.2857142984867096, 0.3571428656578064, 0.4285714328289032, NaN, 0.7142857313156128], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1428571492433548, 0.0, 0.1428571492433548, 0.0, 0.2857142984867096, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8998144768},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.8979591884},{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.8905380399},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8525046449},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8478664262},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.8098330329},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8070501017},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.7588126272},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7532467627},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.72263452},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.7198515879},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7179962983},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7161410134},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.7124304379},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.710575149},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7012987116},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6938775619},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.692949918},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6734693972},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6660482493},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6660482491},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6660482465},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.631725428},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.6196660601},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6196660587},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.6187384163},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.6178107711},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.5992578978},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.597402614},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.5955473216},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5862708812},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5834879511},{"method":"BetaTabPFN (default)","model":"BetaTabPFN","variant":"Default","family":"Foundation Model","mean":0.5825603074},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.5751391556},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5742115127},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5556586371},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5361781192},{"method":"TabICL (default)","model":"TabICL","variant":"Default","family":"Foundation Model","mean":0.5315399011},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5250463925},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.515769955},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.5102040936},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4990723669},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4916512173},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4740259851},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.4693877647},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.4693877647},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.4601113272},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.4601113278},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.4582560399},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4489796026},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4489796011},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.4452690262},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4230055735},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.4211502888},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.4044526997},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.398886836},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.3821892487},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.372912811},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.3395176331},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.3135436077},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2801484303},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2662337744},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2551020477},{"method":"OrionMSP (default)","model":"OrionMSP","variant":"Default","family":"Foundation Model","mean":0.2467532545},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2458256104},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2356215284},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2300556648},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.2226345156},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.217996296},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.2031539951},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1521335867},{"method":"TabFlex (default)","model":"TabFlex","variant":"Default","family":"Foundation Model","mean":0.1410018585},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.1410018597},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1335807089},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.1029684631},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.1001855321},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.0844155873},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0074211506}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..b46ee56d46813169852ff4977be089b64fb03ac7
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_multiclass/datasets_small/winrate_matrix.csv
@@ -0,0 +1,79 @@
+,RealTabPFN-2.5 (tuned + ensembled),"AutoGluon 1.5 (extreme, 4h)",TabFM (default),RealTabPFN-2.5 (tuned),TabPFN-2.6 (default),RealTabPFN-2.5 (default),TabPFN-3 (default),TabICLv2 (default),RealMLP (tuned + ensembled),TabPFNv2 (tuned + ensembled),LimiX (default),RealMLP (tuned),"AutoGluon 1.4 (best, 4h)",CatBoost (tuned),CatBoost (tuned + ensembled),TabM (tuned + ensembled),TabM (tuned),Mitra (default),LightGBM (tuned + ensembled),TabDPT (tuned + ensembled),xRFM (tuned + ensembled),LightGBM (tuned),TabPFNv2 (default),TabDPT-Turbo (default),XGBoost (tuned + ensembled),TabPFNv2 (tuned),XGBoost (tuned),TabM (default),SAP-RPT-OSS (default),TabDPT (tuned),iLTM (tuned + ensembled),ModernNCA (tuned + ensembled),BetaTabPFN (default),xRFM (tuned),ModernNCA (tuned),ExtraTrees (tuned + ensembled),EBM (tuned + ensembled),TabICL (default),CatBoost (default),iLTM (tuned),TorchMLP (tuned + ensembled),RandomForest (tuned + ensembled),FastaiMLP (tuned + ensembled),ExtraTrees (tuned),EBM (tuned),LightGBM (default),EBM (default),TabDPT (default),TorchMLP (tuned),ChimeraBoost (tuned),ChimeraBoost (tuned + ensembled),TabSwift (default),XGBoost (default),RandomForest (tuned),xRFM (default),RealMLP (default),FastaiMLP (tuned),ChimeraBoost (default),ModernNCA (default),KNN (tuned + ensembled),TorchMLP (default),TabSTAR (tuned),TabSTAR (tuned + ensembled),OrionMSP (default),RandomForest (default),iLTM (default),ExtraTrees (default),FastaiMLP (default),Linear (tuned + ensembled),Linear (tuned),KNN (tuned),TabFlex (default),PerpetualBooster (tuned + ensembled),TabSTAR (default),PerpetualBooster (tuned),Linear (default),PerpetualBooster (default),KNN (default)
+RealTabPFN-2.5 (tuned + ensembled),,0.71428573,0.2857143,0.71428573,0.71428573,1.0,0.5714286,0.71428573,0.85714287,1.0,0.85714287,1.0,0.71428573,1.0,1.0,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,0.71428573,1.0,0.71428573,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2857143,,0.14285715,0.71428573,0.5714286,0.42857143,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,0.85714287,0.71428573,1.0,1.0,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,0.85714287,0.85714287,1.0,1.0,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabFM (default),0.71428573,0.85714287,,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.2857143,0.2857143,0.14285715,,0.5714286,0.64285713,0.42857143,0.5714286,0.85714287,0.71428573,0.5714286,1.0,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,1.0,1.0,1.0,0.71428573,0.85714287,0.71428573,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.71428573,1.0,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.2857143,0.42857143,0.14285715,0.42857143,,0.71428573,0.5714286,0.42857143,0.85714287,0.85714287,0.85714287,0.85714287,0.5714286,0.85714287,0.85714287,0.71428573,0.71428573,1.0,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,1.0,0.71428573,1.0,0.71428573,0.71428573,0.71428573,1.0,1.0,0.71428573,0.71428573,1.0,0.85714287,0.71428573,0.71428573,1.0,1.0,0.85714287,0.71428573,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.71428573,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.5714286,0.14285715,0.35714287,0.2857143,,0.5714286,0.5714286,0.71428573,0.42857143,0.85714287,0.71428573,0.5714286,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,1.0,0.5714286,0.85714287,1.0,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.71428573,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.42857143,0.2857143,0.14285715,0.5714286,0.42857143,0.42857143,,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.42857143,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.5714286,0.71428573,1.0,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.2857143,0.2857143,0.0,0.42857143,0.5714286,0.42857143,0.2857143,,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.85714287,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.42857143,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,1.0,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.42857143,0.42857143,,0.42857143,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.42857143,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,1.0,0.71428573,0.85714287,0.85714287,0.5714286,0.5714286,0.85714287,0.85714287,0.71428573,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,0.85714287,1.0,0.85714287,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.71428573,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.0,0.2857143,0.0,0.2857143,0.14285715,0.5714286,0.2857143,0.42857143,0.5714286,,0.42857143,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.78571427,0.71428573,0.71428573,0.42857143,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.85714287,0.5714286,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0
+LimiX (default),0.14285715,0.0,0.14285715,0.42857143,0.14285715,0.14285715,0.2857143,0.2857143,0.5714286,0.5714286,,0.5714286,0.42857143,0.42857143,0.5714286,0.71428573,0.71428573,0.42857143,0.5714286,0.42857143,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.85714287,0.85714287,1.0,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.2857143,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,,0.5714286,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.5714286,0.5714286,0.42857143,0.5714286,0.5714286,0.42857143,0.71428573,0.71428573,0.85714287,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,1.0,0.71428573,0.85714287,0.71428573,0.71428573,1.0,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.5714286,0.85714287,0.71428573,0.85714287,0.85714287,1.0,0.71428573,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.2857143,0.14285715,0.14285715,0.2857143,0.42857143,0.42857143,0.5714286,0.2857143,0.42857143,0.2857143,0.5714286,0.42857143,,0.5714286,0.5714286,0.42857143,0.5714286,0.42857143,0.71428573,0.42857143,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.85714287,0.5714286,1.0,0.71428573,0.85714287,0.85714287,1.0,0.71428573,0.85714287,1.0,1.0,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,1.0,0.85714287,1.0,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.42857143,0.42857143,0.2857143,0.2857143,0.5714286,0.5714286,0.42857143,,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.42857143,0.71428573,0.71428573,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,1.0,0.85714287,0.71428573,1.0,0.85714287,0.85714287,1.0,0.71428573,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.42857143,0.42857143,0.2857143,0.2857143,0.42857143,0.5714286,0.42857143,0.5714286,,0.71428573,0.71428573,0.5714286,0.42857143,0.71428573,0.5714286,0.42857143,0.42857143,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.42857143,0.71428573,0.71428573,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.5714286,0.5714286,0.42857143,0.2857143,,0.42857143,0.42857143,0.5714286,0.71428573,0.42857143,0.5714286,0.2857143,0.71428573,0.71428573,0.2857143,0.71428573,0.85714287,0.5714286,0.71428573,0.5714286,0.71428573,0.85714287,0.5714286,0.85714287,0.5714286,1.0,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,1.0,0.71428573,1.0,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,1.0,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.5714286,,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.2857143,0.5714286,0.5714286,0.2857143,0.71428573,0.85714287,0.5714286,0.71428573,0.5714286,0.71428573,0.85714287,0.5714286,0.85714287,0.5714286,1.0,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,1.0,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Mitra (default),0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.2857143,0.2857143,0.5714286,0.2857143,0.5714286,0.5714286,0.5714286,0.42857143,0.42857143,0.5714286,0.5714286,,0.71428573,0.5714286,0.71428573,0.71428573,0.64285713,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,0.2857143,0.5714286,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.64285713,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.9285714,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.5714286,0.42857143,0.42857143,0.2857143,,0.42857143,0.42857143,0.42857143,0.5714286,0.71428573,0.42857143,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.42857143,0.85714287,0.5714286,0.42857143,0.5714286,0.85714287,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,1.0,1.0,0.85714287,1.0,0.85714287,0.71428573,0.71428573,1.0,0.85714287,1.0,0.71428573,1.0,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.5714286,0.42857143,0.5714286,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.5714286,,0.42857143,0.5714286,0.5714286,0.42857143,0.71428573,0.5714286,0.71428573,0.42857143,0.42857143,0.85714287,0.85714287,0.71428573,0.71428573,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0
+xRFM (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.5714286,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.2857143,0.5714286,0.5714286,,0.5714286,0.42857143,0.71428573,0.5714286,0.42857143,0.5714286,0.5714286,0.42857143,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.5714286,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,1.0,0.71428573,0.71428573,0.85714287,1.0,1.0,1.0,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.5714286,0.42857143,0.42857143,0.2857143,0.5714286,0.42857143,0.42857143,,0.42857143,0.71428573,0.42857143,0.42857143,0.42857143,0.71428573,0.5714286,0.5714286,0.5714286,0.42857143,0.85714287,0.5714286,0.42857143,0.5714286,0.85714287,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,1.0,1.0,0.85714287,1.0,0.85714287,0.71428573,0.71428573,1.0,0.85714287,0.85714287,0.71428573,1.0,0.85714287,0.85714287,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (default),0.0,0.14285715,0.0,0.14285715,0.2857143,0.0,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.5714286,0.71428573,0.71428573,0.35714287,0.42857143,0.42857143,0.5714286,0.5714286,,0.5714286,0.5714286,0.42857143,0.42857143,0.71428573,0.2857143,0.5714286,0.42857143,0.42857143,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.64285713,0.71428573,0.42857143,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.5714286,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.9285714,0.85714287,0.71428573,0.85714287,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.14285715,0.14285715,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.5714286,0.2857143,0.2857143,0.42857143,,0.5714286,0.42857143,0.5714286,0.42857143,0.42857143,0.5714286,0.71428573,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,1.0,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,1.0,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.2857143,0.0,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.14285715,0.5714286,0.2857143,0.42857143,0.5714286,0.42857143,0.42857143,,0.5714286,0.5714286,0.71428573,0.42857143,0.42857143,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,0.5714286,0.71428573,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.5714286,0.71428573,0.85714287,0.85714287,1.0,0.71428573,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (tuned),0.0,0.0,0.0,0.0,0.2857143,0.0,0.14285715,0.2857143,0.2857143,0.21428572,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.71428573,0.71428573,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.42857143,,0.42857143,0.71428573,0.2857143,0.5714286,0.42857143,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.85714287,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.5714286,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.5714286,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.14285715,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.42857143,0.2857143,0.42857143,0.5714286,0.5714286,0.42857143,0.42857143,0.5714286,,0.5714286,0.42857143,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.85714287,0.71428573,0.85714287,0.71428573,0.5714286,0.71428573,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.5714286,0.71428573,0.85714287,0.85714287,1.0,0.71428573,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.5714286,0.42857143,0.2857143,0.2857143,0.5714286,0.2857143,0.2857143,0.42857143,,0.42857143,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.42857143,0.42857143,0.71428573,0.42857143,0.5714286,0.5714286,0.42857143,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.42857143,0.5714286,0.42857143,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,1.0
+TabDPT (tuned),0.2857143,0.0,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.14285715,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.42857143,,0.5714286,0.5714286,0.5714286,0.42857143,0.71428573,0.42857143,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.42857143,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.71428573,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.14285715,0.42857143,0.14285715,0.42857143,0.42857143,0.5714286,0.2857143,0.2857143,0.5714286,0.42857143,0.42857143,0.42857143,0.42857143,,0.42857143,0.42857143,0.71428573,0.71428573,0.71428573,0.5714286,0.42857143,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.42857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.5714286,0.2857143,0.2857143,0.5714286,0.5714286,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,,0.5714286,0.71428573,0.64285713,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+BetaTabPFN (default),0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.14285715,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.42857143,,0.5714286,0.5714286,0.42857143,0.5714286,0.42857143,0.85714287,0.71428573,0.71428573,0.5714286,0.5714286,0.42857143,1.0,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,0.2857143,0.42857143,,0.2857143,0.71428573,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.85714287,0.42857143,0.85714287,0.42857143,0.5714286,0.42857143,0.71428573,0.5714286,0.85714287,0.5714286,0.71428573,0.5714286,0.85714287,1.0,0.71428573,0.5714286,0.71428573,0.71428573,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.71428573,0.85714287,0.71428573,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.14285715,0.14285715,0.0,0.0,0.2857143,0.14285715,0.2857143,0.14285715,0.42857143,0.2857143,0.0,0.42857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.5714286,0.2857143,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,0.2857143,0.35714287,0.42857143,0.71428573,,0.71428573,0.42857143,0.5714286,0.71428573,0.71428573,0.42857143,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,1.0,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,0.2857143,0.5714286,0.2857143,0.2857143,,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.85714287,0.42857143,0.78571427,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.5714286,0.71428573,0.71428573,0.5714286,1.0,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,1.0,0.71428573,0.71428573,0.71428573,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0
+EBM (tuned + ensembled),0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.0,0.0,0.2857143,0.14285715,0.2857143,0.42857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.14285715,0.5714286,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,,0.42857143,0.71428573,0.71428573,0.42857143,0.5714286,0.5714286,0.5714286,0.85714287,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.71428573,0.85714287,0.71428573,0.85714287,1.0,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICL (default),0.14285715,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.0,0.0,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.35714287,0.42857143,0.2857143,0.2857143,0.42857143,0.35714287,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.2857143,0.5714286,0.42857143,0.42857143,0.5714286,0.5714286,,0.5714286,0.5714286,0.42857143,0.5714286,0.42857143,0.71428573,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.85714287,0.71428573,0.71428573,0.71428573,0.85714287,0.78571427,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.85714287,1.0,0.71428573,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.5714286,0.42857143,0.14285715,0.5714286,0.2857143,0.42857143,0.2857143,0.42857143,,0.71428573,0.5714286,0.5714286,0.5714286,0.42857143,0.42857143,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.2857143,0.0,0.2857143,0.42857143,0.0,0.42857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.14285715,0.42857143,0.42857143,0.5714286,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.5714286,0.2857143,0.42857143,0.2857143,,0.5714286,0.5714286,0.71428573,0.71428573,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,1.0,1.0,0.71428573,0.71428573,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.42857143,0.42857143,,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,1.0,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,0.2857143,0.42857143,0.14285715,0.2857143,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.71428573,0.5714286,0.5714286,0.5714286,0.85714287,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,1.0,0.71428573,0.71428573,0.5714286,1.0,0.85714287,0.85714287,0.85714287,1.0,0.71428573,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.14285715,0.14285715,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.0,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.2857143,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.2857143,0.42857143,0.5714286,,0.5714286,0.42857143,0.5714286,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,0.71428573,0.85714287,0.85714287,1.0,0.85714287,1.0
+ExtraTrees (tuned),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.42857143,0.14285715,0.14285715,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.5714286,0.14285715,0.2857143,0.21428572,0.42857143,0.2857143,0.5714286,0.2857143,0.42857143,0.42857143,0.42857143,,0.5714286,0.71428573,0.42857143,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.42857143,0.5714286,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.71428573,0.5714286,0.5714286,1.0,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0
+EBM (tuned),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.0,0.0,0.2857143,0.14285715,0.2857143,0.42857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.0,0.5714286,0.2857143,0.42857143,0.14285715,0.42857143,0.5714286,0.5714286,0.42857143,0.5714286,0.5714286,0.42857143,,0.2857143,0.42857143,0.5714286,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,0.71428573,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,1.0,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.2857143,0.71428573,,0.71428573,0.42857143,0.42857143,0.42857143,0.5714286,0.2857143,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.5714286,0.85714287,0.71428573,1.0,0.71428573,0.71428573,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (default),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.5714286,0.5714286,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.2857143,0.5714286,0.5714286,0.5714286,0.5714286,0.2857143,,0.5714286,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.14285715,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.14285715,0.42857143,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.42857143,0.42857143,0.5714286,0.42857143,,0.42857143,0.2857143,0.2857143,0.42857143,0.5714286,0.42857143,0.71428573,0.5714286,0.71428573,0.5714286,0.5714286,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,1.0,0.71428573,0.42857143,0.5714286,0.85714287,1.0,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0
+TorchMLP (tuned),0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.14285715,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.14285715,0.5714286,0.42857143,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,1.0,1.0,0.85714287,1.0,1.0,1.0,0.85714287,1.0
+ChimeraBoost (tuned),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.0,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.0,0.2857143,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.5714286,0.5714286,0.71428573,0.5714286,,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.42857143,0.5714286,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0
+ChimeraBoost (tuned + ensembled),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.71428573,0.5714286,0.42857143,,0.5714286,0.42857143,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.5714286,0.5714286,0.71428573,0.85714287,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,1.0,0.85714287,1.0,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0
+TabSwift (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.0,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.42857143,0.5714286,0.71428573,0.5714286,0.5714286,0.5714286,0.42857143,0.42857143,,0.42857143,0.42857143,0.42857143,0.71428573,0.71428573,0.71428573,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,1.0
+XGBoost (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.2857143,0.0,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.42857143,0.2857143,0.14285715,0.14285715,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.71428573,0.5714286,0.5714286,0.42857143,0.5714286,0.42857143,0.5714286,0.5714286,,0.5714286,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.85714287,0.85714287,1.0,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.14285715,0.42857143,0.2857143,0.2857143,0.2857143,0.42857143,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.5714286,0.42857143,0.2857143,0.2857143,0.5714286,0.42857143,,0.5714286,0.5714286,0.5714286,0.5714286,0.2857143,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.85714287,0.71428573,0.71428573,0.5714286,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.71428573,1.0,1.0
+xRFM (default),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.0,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.5714286,0.5714286,0.42857143,,0.42857143,0.42857143,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0
+RealMLP (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.2857143,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.0,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.5714286,,0.42857143,0.42857143,0.5714286,0.71428573,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0
+FastaiMLP (tuned),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.0,0.2857143,0.2857143,0.0,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.5714286,,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.71428573,0.71428573,0.5714286,0.85714287,0.71428573,1.0,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0,0.85714287,1.0
+ChimeraBoost (default),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.42857143,0.2857143,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.5714286,0.42857143,,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.5714286,0.42857143,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,1.0,0.85714287,0.85714287,1.0,0.85714287,1.0,0.85714287,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.42857143,0.2857143,0.0,0.42857143,0.0,0.14285715,0.2857143,0.2857143,0.0,0.0,0.14285715,0.2857143,0.0,0.42857143,0.14285715,0.42857143,0.2857143,0.14285715,0.2857143,0.42857143,0.2857143,0.42857143,0.14285715,0.2857143,0.2857143,0.42857143,0.2857143,0.5714286,0.42857143,0.42857143,0.2857143,0.71428573,0.42857143,0.42857143,0.2857143,0.42857143,,0.5714286,0.71428573,0.5714286,0.5714286,0.71428573,0.85714287,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.71428573,0.85714287,0.85714287,0.5714286,0.85714287,0.85714287,1.0,1.0
+KNN (tuned + ensembled),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.0,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.14285715,0.14285715,0.14285715,0.2857143,0.0,0.14285715,0.0,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.42857143,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.42857143,,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,0.42857143,0.42857143,1.0,0.5714286,0.85714287,0.42857143,0.85714287,0.71428573,0.85714287,1.0
+TorchMLP (default),0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.14285715,0.14285715,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.2857143,0.0,0.14285715,0.0,0.14285715,0.42857143,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.2857143,0.5714286,,0.71428573,0.71428573,0.5714286,0.42857143,0.42857143,0.5714286,0.71428573,0.5714286,0.71428573,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,0.85714287,1.0
+TabSTAR (tuned),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.0,0.14285715,0.0,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.42857143,0.5714286,0.2857143,,0.35714287,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.71428573,0.71428573,0.85714287,0.5714286,0.71428573,1.0,0.71428573,0.85714287,0.71428573,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.0,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.42857143,0.5714286,0.2857143,0.64285713,,0.42857143,0.42857143,0.42857143,0.42857143,0.42857143,0.71428573,0.71428573,0.85714287,0.5714286,0.71428573,1.0,0.71428573,0.85714287,0.71428573,1.0
+OrionMSP (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.42857143,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.0,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.0,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.42857143,0.2857143,0.5714286,0.42857143,0.71428573,0.5714286,,0.71428573,0.5714286,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.71428573,0.71428573,0.85714287,0.71428573,0.71428573,0.71428573,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.071428575,0.0,0.0,0.0,0.0,0.071428575,0.0,0.0,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.21428572,0.0,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.42857143,0.5714286,0.14285715,0.5714286,0.5714286,0.5714286,0.5714286,0.2857143,,0.42857143,0.5714286,0.42857143,0.71428573,0.71428573,0.85714287,0.5714286,0.71428573,0.71428573,0.71428573,0.71428573,0.71428573,1.0
+iLTM (default),0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.0,0.14285715,0.0,0.14285715,0.0,0.14285715,0.0,0.2857143,0.0,0.0,0.14285715,0.2857143,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.42857143,0.5714286,0.5714286,0.71428573,0.5714286,0.42857143,0.5714286,,0.5714286,0.5714286,0.5714286,0.5714286,0.71428573,0.71428573,0.85714287,0.85714287,0.85714287,0.71428573,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.2857143,0.0,0.0,0.14285715,0.0,0.14285715,0.2857143,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.0,0.2857143,0.14285715,0.14285715,0.2857143,0.14285715,0.0,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.5714286,0.5714286,0.42857143,0.42857143,0.42857143,,0.71428573,0.42857143,0.42857143,0.85714287,0.5714286,0.71428573,0.71428573,0.85714287,0.71428573,0.85714287,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.0,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.0,0.2857143,0.14285715,0.2857143,0.0,0.2857143,0.14285715,0.0,0.14285715,0.2857143,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.0,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.42857143,0.5714286,0.42857143,0.2857143,,0.5714286,0.71428573,0.71428573,0.5714286,0.5714286,0.85714287,0.5714286,0.85714287,0.71428573,1.0
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.0,0.14285715,0.0,0.2857143,0.0,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.42857143,0.14285715,0.2857143,0.14285715,0.5714286,0.2857143,0.2857143,0.14285715,0.2857143,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.5714286,0.42857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.42857143,,0.64285713,0.85714287,0.5714286,0.5714286,0.5714286,0.5714286,1.0,0.71428573,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.2857143,0.0,0.0,0.0,0.2857143,0.0,0.2857143,0.0,0.2857143,0.0,0.14285715,0.0,0.42857143,0.14285715,0.42857143,0.0,0.2857143,0.14285715,0.42857143,0.14285715,0.2857143,0.0,0.2857143,0.14285715,0.42857143,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.5714286,0.2857143,0.2857143,0.2857143,0.42857143,0.2857143,0.42857143,0.5714286,0.2857143,0.35714287,,0.85714287,0.5714286,0.5714286,0.5714286,0.5714286,0.85714287,0.71428573,1.0
+KNN (tuned),0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.0,0.14285715,0.0,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.2857143,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.14285715,,0.2857143,0.42857143,0.2857143,0.5714286,0.2857143,0.5714286,1.0
+TabFlex (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.14285715,0.2857143,0.0,0.14285715,0.14285715,0.14285715,0.42857143,0.14285715,0.42857143,0.42857143,0.2857143,0.42857143,0.2857143,0.42857143,0.42857143,0.42857143,0.42857143,0.71428573,,0.42857143,0.71428573,0.42857143,0.71428573,0.5714286,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.2857143,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.2857143,0.42857143,0.42857143,0.42857143,0.5714286,0.5714286,,0.2857143,0.85714287,0.5714286,1.0,0.85714287
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.14285715,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.2857143,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.0,0.14285715,0.0,0.42857143,0.5714286,0.14285715,0.0,0.0,0.14285715,0.2857143,0.14285715,0.2857143,0.14285715,0.42857143,0.42857143,0.71428573,0.2857143,0.71428573,,0.71428573,0.71428573,0.71428573,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.14285715,0.42857143,0.42857143,0.42857143,0.42857143,0.5714286,0.14285715,0.2857143,,0.5714286,0.64285713,0.85714287
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.2857143,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.14285715,0.0,0.2857143,0.0,0.0,0.0,0.0,0.14285715,0.2857143,0.14285715,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.14285715,0.0,0.14285715,0.71428573,0.2857143,0.42857143,0.2857143,0.42857143,,0.5714286,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.0,0.0,0.14285715,0.14285715,0.14285715,0.14285715,0.0,0.14285715,0.14285715,0.2857143,0.2857143,0.2857143,0.2857143,0.0,0.14285715,0.2857143,0.2857143,0.2857143,0.42857143,0.42857143,0.0,0.2857143,0.35714287,0.42857143,,0.71428573
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.14285715,0.0,0.14285715,0.0,0.2857143,
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/leaderboard_overview_explorer.html
index 788a71d2bb465ce8801220223ec299110abdb50f..420c96b2c15f6322901dd7656ca4320eb7e398f2 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Regression | All Datasets</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..24b7f80efab34486c5446a87496c469b2a4a0643
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_regression/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2191.0,"score":0.904,"rank":2.77,"hrank":1.54,"imp":1.407,"train_time":40.79,"predict_time":6.492,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+260\/-65","imputed":false},{"position":1,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1910.0,"score":0.686,"rank":7.85,"hrank":4.34,"imp":3.335,"train_time":339.37,"predict_time":3.244,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-39","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1902.0,"score":0.7,"rank":8.08,"hrank":3.14,"imp":2.269,"train_time":4.02,"predict_time":0.398,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+267\/-138","imputed":false},{"position":3,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1891.0,"score":0.681,"rank":8.38,"hrank":4.0,"imp":3.06,"train_time":11.81,"predict_time":2.007,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+161\/-85","imputed":false},{"position":4,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1866.0,"score":0.645,"rank":9.15,"hrank":5.7,"imp":2.722,"train_time":2043.54,"predict_time":8.366,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+211\/-140","imputed":false},{"position":5,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1854.0,"score":0.631,"rank":9.54,"hrank":5.77,"imp":3.39,"train_time":8.44,"predict_time":0.7,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+170\/-57","imputed":false},{"position":6,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1793.0,"score":0.596,"rank":11.69,"hrank":4.28,"imp":4.004,"train_time":7126.67,"predict_time":336.907,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+217\/-95","imputed":false},{"position":7,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1791.0,"score":0.604,"rank":11.77,"hrank":6.1,"imp":4.114,"train_time":6.86,"predict_time":1.346,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+167\/-106","imputed":false},{"position":8,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1738.0,"score":0.551,"rank":13.92,"hrank":9.03,"imp":4.506,"train_time":7126.67,"predict_time":38.573,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+147\/-74","imputed":false},{"position":9,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1738.0,"score":0.543,"rank":13.92,"hrank":7.62,"imp":3.651,"train_time":2043.54,"predict_time":0.78,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+169\/-130","imputed":false},{"position":10,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1703.0,"score":0.577,"rank":15.46,"hrank":4.79,"imp":4.372,"train_time":1.32,"predict_time":0.122,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+286\/-147","imputed":false},{"position":11,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1666.0,"score":0.449,"rank":17.23,"hrank":15.13,"imp":5.694,"train_time":4057.31,"predict_time":10.474,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-54","imputed":false},{"position":12,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1663.0,"score":0.416,"rank":17.38,"hrank":12.09,"imp":6.221,"train_time":1490.17,"predict_time":5.79,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+109\/-90","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1633.0,"score":0.448,"rank":18.92,"hrank":14.59,"imp":5.429,"train_time":2.02,"predict_time":0.183,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+149\/-85","imputed":false},{"position":14,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1627.0,"score":0.435,"rank":19.23,"hrank":11.89,"imp":5.105,"train_time":6.94,"predict_time":0.519,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+182\/-148","imputed":false},{"position":15,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1621.0,"score":0.438,"rank":19.54,"hrank":11.94,"imp":5.656,"train_time":44.65,"predict_time":38.297,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+177\/-84","imputed":false},{"position":16,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1585.0,"score":0.396,"rank":21.46,"hrank":7.24,"imp":8.163,"train_time":3811.61,"predict_time":7.581,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+179\/-174","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1536.0,"score":0.237,"rank":24.23,"hrank":20.57,"imp":8.792,"train_time":686.5,"predict_time":5.477,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-78","imputed":false},{"position":18,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1532.0,"score":0.336,"rank":24.46,"hrank":18.95,"imp":6.839,"train_time":4057.31,"predict_time":0.337,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-133","imputed":false},{"position":19,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1504.0,"score":0.192,"rank":26.15,"hrank":19.44,"imp":8.535,"train_time":2267.19,"predict_time":0.422,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-75","imputed":false},{"position":20,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1491.0,"score":0.192,"rank":26.92,"hrank":18.62,"imp":8.569,"train_time":2267.19,"predict_time":0.046,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-73","imputed":false},{"position":21,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1490.0,"score":0.268,"rank":27.0,"hrank":22.48,"imp":7.163,"train_time":3448.73,"predict_time":2.111,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-92","imputed":false},{"position":22,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1487.0,"score":0.273,"rank":27.15,"hrank":22.55,"imp":7.963,"train_time":708.95,"predict_time":1.973,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+138\/-93","imputed":false},{"position":23,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1477.0,"score":0.216,"rank":27.77,"hrank":24.88,"imp":8.531,"train_time":12741.09,"predict_time":387.573,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+84\/-57","imputed":false},{"position":24,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1466.0,"score":0.294,"rank":28.46,"hrank":18.96,"imp":7.143,"train_time":0.9,"predict_time":0.072,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+185\/-124","imputed":false},{"position":25,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1445.0,"score":0.222,"rank":29.85,"hrank":24.72,"imp":7.54,"train_time":3448.73,"predict_time":0.194,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-93","imputed":false},{"position":26,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1445.0,"score":0.175,"rank":29.85,"hrank":24.36,"imp":9.578,"train_time":686.5,"predict_time":0.743,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-96","imputed":false},{"position":27,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1432.0,"score":0.192,"rank":30.69,"hrank":26.92,"imp":8.514,"train_time":708.95,"predict_time":0.069,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-62","imputed":false},{"position":28,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1409.0,"score":0.114,"rank":32.23,"hrank":30.33,"imp":9.492,"train_time":848.99,"predict_time":0.475,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-59","imputed":false},{"position":29,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1408.0,"score":0.111,"rank":32.31,"hrank":30.99,"imp":9.51,"train_time":848.99,"predict_time":2.375,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-50","imputed":false},{"position":30,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1392.0,"score":0.169,"rank":33.38,"hrank":29.75,"imp":10.782,"train_time":16.07,"predict_time":0.289,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+94\/-85","imputed":false},{"position":31,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1386.0,"score":0.259,"rank":33.81,"hrank":13.94,"imp":7.768,"train_time":3805.62,"predict_time":10.413,"imputed_pct":46.15,"hardware":"GPU","elo_ci":"+210\/-144","imputed":true},{"position":32,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1386.0,"score":0.114,"rank":33.85,"hrank":27.87,"imp":10.663,"train_time":9.02,"predict_time":0.018,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+103\/-103","imputed":false},{"position":33,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1373.0,"score":0.205,"rank":34.69,"hrank":23.92,"imp":10.655,"train_time":3811.61,"predict_time":0.452,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-161","imputed":false},{"position":34,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1363.0,"score":0.116,"rank":35.38,"hrank":29.63,"imp":11.959,"train_time":402.93,"predict_time":0.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+127\/-112","imputed":false},{"position":35,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1336.0,"score":0.101,"rank":37.15,"hrank":34.12,"imp":9.828,"train_time":12741.09,"predict_time":60.101,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+111\/-77","imputed":false},{"position":36,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1317.0,"score":0.166,"rank":38.46,"hrank":26.65,"imp":9.922,"train_time":71.01,"predict_time":1.588,"imputed_pct":46.15,"hardware":"GPU","elo_ci":"+135\/-102","imputed":true},{"position":37,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1315.0,"score":0.168,"rank":38.54,"hrank":16.8,"imp":10.388,"train_time":59.38,"predict_time":14.073,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+198\/-192","imputed":false},{"position":38,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1307.0,"score":0.09,"rank":39.08,"hrank":34.64,"imp":12.342,"train_time":402.93,"predict_time":0.034,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-112","imputed":false},{"position":39,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1283.0,"score":0.079,"rank":40.69,"hrank":38.05,"imp":9.907,"train_time":9.79,"predict_time":0.134,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-85","imputed":false},{"position":40,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1282.0,"score":0.172,"rank":40.73,"hrank":26.44,"imp":9.261,"train_time":3805.62,"predict_time":0.262,"imputed_pct":46.15,"hardware":"GPU","elo_ci":"+170\/-153","imputed":true},{"position":41,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1270.0,"score":0.135,"rank":41.54,"hrank":31.29,"imp":11.044,"train_time":18.77,"predict_time":6.542,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+172\/-151","imputed":false},{"position":42,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1250.0,"score":0.071,"rank":42.92,"hrank":38.9,"imp":11.48,"train_time":9.18,"predict_time":1.415,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+88\/-97","imputed":false},{"position":43,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1249.0,"score":0.061,"rank":43.0,"hrank":38.19,"imp":13.342,"train_time":257.64,"predict_time":0.813,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+92\/-99","imputed":false},{"position":44,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1249.0,"score":0.067,"rank":43.0,"hrank":39.69,"imp":10.851,"train_time":4452.11,"predict_time":0.852,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-97","imputed":false},{"position":45,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1230.0,"score":0.156,"rank":44.23,"hrank":29.56,"imp":10.397,"train_time":2.78,"predict_time":0.315,"imputed_pct":46.15,"hardware":"GPU","elo_ci":"+170\/-173","imputed":true},{"position":46,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1225.0,"score":0.051,"rank":44.62,"hrank":41.19,"imp":13.517,"train_time":2.46,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-114","imputed":false},{"position":47,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1221.0,"score":0.049,"rank":44.85,"hrank":40.47,"imp":13.483,"train_time":257.64,"predict_time":0.06,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-97","imputed":false},{"position":48,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1202.0,"score":0.056,"rank":46.15,"hrank":42.14,"imp":11.465,"train_time":4452.11,"predict_time":0.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-122","imputed":false},{"position":49,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1197.0,"score":0.066,"rank":46.46,"hrank":35.71,"imp":14.293,"train_time":2.47,"predict_time":0.755,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+133\/-156","imputed":false},{"position":50,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1197.0,"score":0.019,"rank":46.46,"hrank":44.88,"imp":12.343,"train_time":2.03,"predict_time":0.296,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-84","imputed":false},{"position":51,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1170.0,"score":0.024,"rank":48.19,"hrank":45.75,"imp":14.342,"train_time":177.92,"predict_time":0.433,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-104","imputed":false},{"position":52,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1170.0,"score":0.08,"rank":48.23,"hrank":35.18,"imp":14.844,"train_time":2006.55,"predict_time":0.103,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+132\/-178","imputed":false},{"position":53,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1165.0,"score":0.029,"rank":48.54,"hrank":44.44,"imp":13.048,"train_time":2.15,"predict_time":0.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-120","imputed":false},{"position":54,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1154.0,"score":0.007,"rank":49.23,"hrank":47.46,"imp":14.614,"train_time":526.19,"predict_time":0.77,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-80","imputed":false},{"position":55,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1131.0,"score":0.01,"rank":50.65,"hrank":49.19,"imp":15.075,"train_time":177.92,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-109","imputed":false},{"position":56,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1122.0,"score":0.083,"rank":51.15,"hrank":36.28,"imp":15.396,"train_time":2006.55,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+144\/-192","imputed":false},{"position":57,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1114.0,"score":0.007,"rank":51.65,"hrank":49.53,"imp":15.349,"train_time":526.19,"predict_time":0.119,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-115","imputed":false},{"position":58,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1076.0,"score":0.007,"rank":53.85,"hrank":50.21,"imp":15.856,"train_time":2.57,"predict_time":0.242,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-126","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1045.0,"score":0.004,"rank":55.54,"hrank":54.18,"imp":15.657,"train_time":527.3,"predict_time":2.83,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-96","imputed":false},{"position":60,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1024.0,"score":0.001,"rank":56.62,"hrank":54.31,"imp":15.502,"train_time":20.5,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+80\/-140","imputed":false},{"position":61,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1022.0,"score":0.026,"rank":56.77,"hrank":52.95,"imp":16.48,"train_time":7.59,"predict_time":0.012,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+114\/-154","imputed":false},{"position":62,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1013.0,"score":0.0,"rank":57.23,"hrank":55.61,"imp":16.134,"train_time":527.3,"predict_time":0.311,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-108","imputed":false},{"position":63,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":57.88,"hrank":57.12,"imp":16.648,"train_time":0.63,"predict_time":0.064,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-96","imputed":false},{"position":64,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":979.0,"score":0.066,"rank":58.92,"hrank":37.36,"imp":24.113,"train_time":46109.53,"predict_time":22.149,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+225\/-314","imputed":false},{"position":65,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":953.0,"score":0.003,"rank":60.15,"hrank":58.43,"imp":18.294,"train_time":22.8,"predict_time":0.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-152","imputed":false},{"position":66,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":938.0,"score":0.053,"rank":60.85,"hrank":46.59,"imp":24.294,"train_time":46109.53,"predict_time":5.06,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+221\/-298","imputed":false},{"position":67,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":883.0,"score":0.0,"rank":63.15,"hrank":62.01,"imp":20.145,"train_time":3.08,"predict_time":0.293,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-142","imputed":false},{"position":68,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":863.0,"score":0.0,"rank":63.92,"hrank":63.26,"imp":19.248,"train_time":373.42,"predict_time":81.49,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-143","imputed":false},{"position":69,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":855.0,"score":0.0,"rank":64.23,"hrank":63.12,"imp":21.535,"train_time":90.87,"predict_time":0.906,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+111\/-157","imputed":false},{"position":70,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":793.0,"score":0.025,"rank":66.31,"hrank":59.73,"imp":27.43,"train_time":302.9,"predict_time":4.583,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+247\/-324","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":717.0,"score":0.0,"rank":68.46,"hrank":68.09,"imp":23.655,"train_time":90.87,"predict_time":0.055,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-155","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":623.0,"score":0.0,"rank":70.54,"hrank":69.83,"imp":30.241,"train_time":0.25,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+153\/-252","imputed":false},{"position":73,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":467.0,"score":0.0,"rank":72.96,"hrank":72.82,"imp":37.219,"train_time":177.06,"predict_time":0.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-354","imputed":false},{"position":74,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":407.0,"score":0.0,"rank":73.65,"hrank":73.5,"imp":37.401,"train_time":177.06,"predict_time":0.059,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+125\/-473","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":270.0,"score":0.0,"rank":74.92,"hrank":74.91,"imp":39.828,"train_time":0.88,"predict_time":0.109,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+108\/-398","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/pareto_front_explorer.html
index c6135c68f4197041331178fc1bfda02e36ac5bb6..02f23317fc67e45523e9a31071c64234d208d771 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0182719821,"imp":10.6629057031,"elo":1385.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.04636871,"imp":8.5689975774,"elo":1491.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"imp":8.5351365089,"elo":1503.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"imp":13.5165870219,"elo":1224.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0343070535,"imp":12.3418233313,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"imp":11.9586707247,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.01204424,"imp":16.479912159,"elo":1021.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0125056048,"imp":15.3960953256,"elo":1122.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1032251722,"imp":14.8437898138,"elo":1169.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2422366525,"imp":15.8559260551,"elo":1075.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0603961196,"imp":13.4828514032,"elo":1221.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8125804985,"imp":13.3421633346,"elo":1248.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2929314919,"imp":20.1446549368,"elo":883.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3110785788,"imp":16.1337653625,"elo":1012.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8297120981,"imp":15.6574941615,"elo":1044.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2962263441,"imp":12.3430360848,"elo":1197.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7428209859,"imp":9.577727366,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":5.476648655,"imp":8.7915527894,"elo":1536.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":14.072768612,"imp":10.3876718017,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":1.5880635253,"imp":9.921574475,"elo":1316.6,"imputed":true,"imputed_pct":46.1538461538,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2894827463,"imp":10.7820379041,"elo":1392.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4518951639,"imp":10.6548900789,"elo":1373.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5805507968,"imp":8.1631535544,"elo":1585.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.3455286154,"imp":4.1144151877,"elo":1790.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0067377719,"imp":3.0595728499,"elo":1891.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0171543088,"imp":18.2943297341,"elo":953.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1465391338,"imp":15.0751309127,"elo":1130.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4333863798,"imp":14.3424516152,"elo":1170.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0638540688,"imp":16.6482084857,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.119307123,"imp":15.3486617095,"elo":1113.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7700427525,"imp":14.6140166256,"elo":1154.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.4152312914,"imp":11.4799030804,"elo":1249.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3372766026,"imp":6.8392974584,"elo":1532.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":10.4744362064,"imp":5.6941083754,"elo":1665.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5191857229,"imp":5.1050024316,"elo":1626.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7797947208,"imp":3.6513451358,"elo":1737.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3660135567,"imp":2.7216385458,"elo":1865.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.5421299186,"imp":11.0439680079,"elo":1270.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"imp":5.6563845374,"elo":1620.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.5726172424,"imp":4.5064720405,"elo":1737.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":336.9071136327,"imp":4.0044876883,"elo":1792.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"imp":5.4286497419,"elo":1632.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4924694208,"imp":1.4067478497,"elo":2191.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1218380804,"imp":4.3723770177,"elo":1702.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1337247011,"imp":9.9071412654,"elo":1282.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1936377213,"imp":7.539858265,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.1105459826,"imp":7.1633433427,"elo":1489.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6996427484,"imp":3.3904763149,"elo":1854.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3981148822,"imp":2.2689880747,"elo":1901.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3150187952,"imp":10.3973338082,"elo":1230.4,"imputed":true,"imputed_pct":46.1538461538,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2619314173,"imp":9.2614712573,"elo":1282.4,"imputed":true,"imputed_pct":46.1538461538,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.4125412162,"imp":7.7683696892,"elo":1386.1,"imputed":true,"imputed_pct":46.1538461538,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5829823393,"imp":27.4295870304,"elo":793.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.0596311902,"imp":24.2937076463,"elo":938.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.148843297,"imp":24.1126590799,"elo":979.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719372146,"imp":7.1426209224,"elo":1466.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0813896645,"imp":15.5015821228,"elo":1024.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0885042776,"imp":11.4645270075,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8515826242,"imp":10.8505791228,"elo":1248.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1795378646,"imp":13.048270691,"elo":1165.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4748247737,"imp":9.4916701231,"elo":1409.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.3747946103,"imp":9.5098219062,"elo":1408.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":81.4898053701,"imp":19.2481247323,"elo":863.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":60.1008510867,"imp":9.8281010544,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":387.572909971,"imp":8.5308610842,"elo":1477.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7548401788,"imp":14.2932711533,"elo":1197.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0686603653,"imp":8.5143347422,"elo":1432.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.9734199806,"imp":7.9631079976,"elo":1487.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0182719821,"x_train":9.0226914704,"imp":10.6629057031,"elo":1385.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.04636871,"x_train":2267.1899830578,"imp":8.5689975774,"elo":1491.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"x_train":2267.1899830578,"imp":8.5351365089,"elo":1503.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"x_train":2.4559533883,"imp":13.5165870219,"elo":1224.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0343070535,"x_train":402.9323608119,"imp":12.3418233313,"elo":1307.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":402.9323608119,"imp":11.9586707247,"elo":1362.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.01204424,"x_train":7.5931963365,"imp":16.479912159,"elo":1021.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0125056048,"x_train":2006.5538981646,"imp":15.3960953256,"elo":1122.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1032251722,"x_train":2006.5538981646,"imp":14.8437898138,"elo":1169.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2422366525,"x_train":2.5746349456,"imp":15.8559260551,"elo":1075.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0603961196,"x_train":257.6385593165,"imp":13.4828514032,"elo":1221.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8125804985,"x_train":257.6385593165,"imp":13.3421633346,"elo":1248.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2929314919,"x_train":3.0840796949,"imp":20.1446549368,"elo":883.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3110785788,"x_train":527.2961316043,"imp":16.1337653625,"elo":1012.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8297120981,"x_train":527.2961316043,"imp":15.6574941615,"elo":1044.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2962263441,"x_train":2.0296176424,"imp":12.3430360848,"elo":1197.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7428209859,"x_train":686.4972156667,"imp":9.577727366,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":5.476648655,"x_train":686.4972156667,"imp":8.7915527894,"elo":1536.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":14.072768612,"x_train":59.3758334402,"imp":10.3876718017,"elo":1315.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":1.5880635253,"x_train":71.0123333261,"imp":9.921574475,"elo":1316.6,"imputed":true,"imputed_pct":46.1538461538,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2894827463,"x_train":16.0749964455,"imp":10.7820379041,"elo":1392.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.4518951639,"x_train":3811.6104621203,"imp":10.6548900789,"elo":1373.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5805507968,"x_train":3811.6104621203,"imp":8.1631535544,"elo":1585.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.3455286154,"x_train":6.8642025254,"imp":4.1144151877,"elo":1790.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0067377719,"x_train":11.8050270551,"imp":3.0595728499,"elo":1891.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0171543088,"x_train":22.8030707188,"imp":18.2943297341,"elo":953.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1465391338,"x_train":177.9195746923,"imp":15.0751309127,"elo":1130.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4333863798,"x_train":177.9195746923,"imp":14.3424516152,"elo":1170.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0638540688,"x_train":0.6331018043,"imp":16.6482084857,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.119307123,"x_train":526.192235626,"imp":15.3486617095,"elo":1113.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.7700427525,"x_train":526.192235626,"imp":14.6140166256,"elo":1154.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":1.4152312914,"x_train":9.1753045614,"imp":11.4799030804,"elo":1249.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.3372766026,"x_train":4057.3100555326,"imp":6.8392974584,"elo":1532.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":10.4744362064,"x_train":4057.3100555326,"imp":5.6941083754,"elo":1665.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5191857229,"x_train":6.9411546573,"imp":5.1050024316,"elo":1626.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.7797947208,"x_train":2043.5362680496,"imp":3.6513451358,"elo":1737.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3660135567,"x_train":2043.5362680496,"imp":2.7216385458,"elo":1865.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.5421299186,"x_train":18.769283676,"imp":11.0439680079,"elo":1270.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"x_train":44.6485648211,"imp":5.6563845374,"elo":1620.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.5726172424,"x_train":7126.6747880976,"imp":4.5064720405,"elo":1737.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":336.9071136327,"x_train":7126.6747880976,"imp":4.0044876883,"elo":1792.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"x_train":2.0183513985,"imp":5.4286497419,"elo":1632.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":6.4924694208,"x_train":40.7926562599,"imp":1.4067478497,"elo":2191.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1218380804,"x_train":1.3151215973,"imp":4.3723770177,"elo":1702.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1337247011,"x_train":9.7873678423,"imp":9.9071412654,"elo":1282.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.1936377213,"x_train":3448.7325920652,"imp":7.539858265,"elo":1445.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":2.1105459826,"x_train":3448.7325920652,"imp":7.1633433427,"elo":1489.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6996427484,"x_train":8.4368371839,"imp":3.3904763149,"elo":1854.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3981148822,"x_train":4.0220384031,"imp":2.2689880747,"elo":1901.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3150187952,"x_train":2.7780429735,"imp":10.3973338082,"elo":1230.4,"imputed":true,"imputed_pct":46.1538461538,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.2619314173,"x_train":3805.6199685881,"imp":9.2614712573,"elo":1282.4,"imputed":true,"imputed_pct":46.1538461538,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.4125412162,"x_train":3805.6199685881,"imp":7.7683696892,"elo":1386.1,"imputed":true,"imputed_pct":46.1538461538,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5829823393,"x_train":302.8991048646,"imp":27.4295870304,"elo":793.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.0596311902,"x_train":46109.5334192428,"imp":24.2937076463,"elo":938.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.148843297,"x_train":46109.5334192428,"imp":24.1126590799,"elo":979.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719372146,"x_train":0.8979740772,"imp":7.1426209224,"elo":1466.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0813896645,"x_train":20.5017216002,"imp":15.5015821228,"elo":1024.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0885042776,"x_train":4452.1085706374,"imp":11.4645270075,"elo":1201.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8515826242,"x_train":4452.1085706374,"imp":10.8505791228,"elo":1248.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.1795378646,"x_train":2.1459956511,"imp":13.048270691,"elo":1165.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.4748247737,"x_train":848.9941919866,"imp":9.4916701231,"elo":1409.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":2.3747946103,"x_train":848.9941919866,"imp":9.5098219062,"elo":1408.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":81.4898053701,"x_train":373.4221052753,"imp":19.2481247323,"elo":863.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":60.1008510867,"x_train":12741.0939157936,"imp":9.8281010544,"elo":1336.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":387.572909971,"x_train":12741.0939157936,"imp":8.5308610842,"elo":1477.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.7548401788,"x_train":2.469327894,"imp":14.2932711533,"elo":1197.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0686603653,"x_train":708.9450090283,"imp":8.5143347422,"elo":1432.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":1.9734199806,"x_train":708.9450090283,"imp":7.9631079976,"elo":1487.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/tuning_trajectories_explorer.html
index 9214ea39bc385b726b3f42eaef279965e61d2d12..e73f048259b80c9844a34f2a7fd12414fd8145d2 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_regression/datasets_all</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":9.0226914704,"imp":10.6649854115,"elo":1387.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":22.006737996,"imp":10.0410335189,"elo":1430.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":54.5936853194,"imp":9.3422932439,"elo":1493.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":108.5250024926,"imp":9.0363696279,"elo":1507.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":277.687397929,"imp":8.7304375075,"elo":1518.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":564.9102272658,"imp":8.6385669356,"elo":1520.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1124.7408503677,"imp":8.5860979377,"elo":1520.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":1677.2797709559,"imp":8.5542513657,"elo":1524.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":2267.1899830578,"imp":8.537762157,"elo":1525.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":2.4559533883,"imp":13.5184798184,"elo":1230.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":4.0653056289,"imp":13.1595745141,"elo":1250.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":9.725445997,"imp":12.8184880099,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":19.7798589835,"imp":12.6146892497,"elo":1312.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":50.0934628362,"imp":12.3717353184,"elo":1329.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":100.1940386909,"imp":12.2119545046,"elo":1348.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":200.5360400335,"imp":12.0503277586,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":300.5068144065,"imp":11.984244047,"elo":1371.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":402.9323608119,"imp":11.9597608126,"elo":1378.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":7.5931963365,"imp":16.4815170931,"elo":1032.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":19.5464706545,"imp":15.9818225032,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":52.461846709,"imp":15.5322088782,"elo":1125.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":101.1927167213,"imp":15.3095708687,"elo":1139.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":250.7607078715,"imp":15.0786611831,"elo":1158.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":505.4611049176,"imp":15.0054156631,"elo":1165.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1000.4774432421,"imp":14.9259310452,"elo":1175.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":1490.4477925659,"imp":14.8788837948,"elo":1178.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2006.5538981646,"imp":14.8444889121,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":2.5746349456,"imp":15.8576388612,"elo":1074.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1741914103,"imp":15.312547953,"elo":1115.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.7533015627,"imp":14.5183742413,"elo":1190.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":13.1877719851,"imp":14.1351179951,"elo":1214.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":32.4660623442,"imp":13.547469136,"elo":1242.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":64.6497454444,"imp":13.4253490814,"elo":1255.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":128.5728546231,"imp":13.3545124869,"elo":1262.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":192.4203093077,"imp":13.3423915858,"elo":1261.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":257.6385593165,"imp":13.3444260284,"elo":1261.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":3.0840796949,"imp":20.1459917158,"elo":866.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":5.6240636486,"imp":19.8504258159,"elo":885.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":13.6732407308,"imp":18.9476612196,"elo":928.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":27.2107956399,"imp":18.4865676439,"elo":942.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":65.2537322804,"imp":17.520082439,"elo":978.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":130.574483839,"imp":16.9310727816,"elo":1002.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":262.2064329926,"imp":16.338156479,"elo":1022.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":393.3022867336,"imp":15.93147183,"elo":1048.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":527.2961316043,"imp":15.6594635682,"elo":1061.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":2.0296176424,"imp":12.3451040358,"elo":1201.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":5.8086693996,"imp":11.4997719236,"elo":1264.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":16.4346292254,"imp":10.4066250539,"elo":1373.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":32.0193358834,"imp":9.8788450456,"elo":1427.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":82.7850514402,"imp":9.3381159903,"elo":1499.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":169.4629431685,"imp":9.1757274338,"elo":1526.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":340.4054795841,"imp":8.9869444807,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":509.4885196025,"imp":8.8864521179,"elo":1559.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":686.4972156667,"imp":8.7938407814,"elo":1574.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":59.3758334402,"imp":10.3904487974,"elo":1335.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":71.0123333261,"imp":9.9244566274,"elo":1334.5,"imputed":true,"imputed_pct":46.1538461538,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":16.0749964455,"imp":10.7841513848,"elo":1414.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":38.647009759,"imp":10.2023184034,"elo":1444.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":106.244026897,"imp":9.4160045764,"elo":1499.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":199.3707655728,"imp":9.0269539703,"elo":1542.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":462.7480963285,"imp":8.6556576511,"elo":1576.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":955.9940479091,"imp":8.4515123264,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1870.556153527,"imp":8.3000951981,"elo":1589.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":2790.7899911449,"imp":8.1946214692,"elo":1602.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":3811.6104621203,"imp":8.1654997251,"elo":1604.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":6.8642025254,"imp":4.1176969928,"elo":1809.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":11.8050270551,"imp":3.0630923517,"elo":1916.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":22.8030707188,"imp":18.2955753566,"elo":966.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":53.0597722616,"imp":16.107246993,"elo":1072.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":147.0385313693,"imp":14.3675168325,"elo":1159.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":177.9195746923,"imp":14.3437457459,"elo":1169.3,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.6331018043,"imp":16.6499447959,"elo":1008.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":3.5138045858,"imp":15.8760735961,"elo":1070.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":12.8062748979,"imp":15.3908176771,"elo":1107.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":26.5826232412,"imp":14.9942764132,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":64.673699967,"imp":14.7770918641,"elo":1145.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":130.0298950791,"imp":14.6901072863,"elo":1154.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":259.9521641476,"imp":14.638686862,"elo":1163.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":391.5234331877,"imp":14.6211649426,"elo":1167.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":526.192235626,"imp":14.6157763538,"elo":1168.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":9.1753045614,"imp":11.4826162385,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":23.6721342187,"imp":9.3869073322,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":78.6133073597,"imp":7.3521875932,"elo":1526.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":191.0765941914,"imp":6.6293265441,"elo":1573.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":503.1010684492,"imp":6.1625582504,"elo":1623.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1009.2362249093,"imp":5.9055348463,"elo":1660.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2015.6576136475,"imp":5.7842906504,"elo":1678.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3026.2086158211,"imp":5.7349512293,"elo":1692.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4057.3100555326,"imp":5.6951806629,"elo":1698.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":6.9411546573,"imp":5.1081297102,"elo":1639.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":17.2836445024,"imp":4.6374541046,"elo":1666.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":47.5797720049,"imp":3.8614140786,"elo":1719.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":98.3276706028,"imp":3.4224976388,"elo":1781.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":249.0037143861,"imp":2.9685684337,"elo":1844.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":503.1695046416,"imp":2.7947420106,"elo":1868.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1018.7089146459,"imp":2.7018005663,"elo":1898.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1526.668944465,"imp":2.7311655156,"elo":1899.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2043.5362680496,"imp":2.7245619429,"elo":1904.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":18.769283676,"imp":11.0464917637,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":44.6485648211,"imp":5.6593617074,"elo":1644.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":88.2308458205,"imp":5.2454696275,"elo":1673.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":186.0445098453,"imp":4.8490533938,"elo":1698.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":341.1376387803,"imp":4.5346498664,"elo":1733.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":866.4570381261,"imp":4.2812155017,"elo":1764.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1760.5896780945,"imp":4.1783339382,"elo":1778.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":3528.4122296034,"imp":4.0803258048,"elo":1795.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":5308.8782679903,"imp":4.0274723245,"elo":1809.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":7126.6747880976,"imp":4.0079622689,"elo":1816.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":2.0183513985,"imp":5.431606219,"elo":1661.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":40.7926562599,"imp":1.4100424178,"elo":2207.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":1.3151215973,"imp":4.3753875956,"elo":1718.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":9.7873678423,"imp":9.9099485615,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":23.4153138909,"imp":9.3565549555,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":71.5165108752,"imp":8.7147560503,"elo":1381.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":166.1346310591,"imp":8.182159933,"elo":1430.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":413.5775428251,"imp":7.7342280619,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":844.0698066634,"imp":7.4331475795,"elo":1495.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1693.1667185284,"imp":7.2469532672,"elo":1514.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2557.2442472516,"imp":7.2041346491,"elo":1517.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3448.7325920652,"imp":7.1664648695,"elo":1517.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":8.4368371839,"imp":3.393775576,"elo":1886.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":4.0220384031,"imp":2.2837784913,"elo":1912.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2.7780429735,"imp":10.4004807181,"elo":1242.7,"imputed":true,"imputed_pct":46.1538461538,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":25.3768652833,"imp":9.5979783992,"elo":1279.0,"imputed":true,"imputed_pct":46.1538461538,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":76.7795330932,"imp":8.8677722781,"elo":1321.7,"imputed":true,"imputed_pct":46.1538461538,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":175.8535512882,"imp":8.5986651896,"elo":1337.6,"imputed":true,"imputed_pct":46.1538461538,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":436.6902139142,"imp":8.1773741755,"elo":1371.8,"imputed":true,"imputed_pct":46.1538461538,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":900.8153356806,"imp":8.0237501849,"elo":1381.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1849.0970608496,"imp":7.8388654555,"elo":1390.2,"imputed":true,"imputed_pct":46.1538461538,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":2791.064560621,"imp":7.8057075844,"elo":1394.7,"imputed":true,"imputed_pct":46.1538461538,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3805.6199685881,"imp":7.7716997285,"elo":1396.9,"imputed":true,"imputed_pct":46.1538461538,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":302.8991048646,"imp":27.4310818967,"elo":788.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":632.7525144671,"imp":27.2879140579,"elo":810.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1217.0384701627,"imp":26.9251443516,"elo":845.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2404.0220528997,"imp":26.2981860908,"elo":884.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5871.8477809815,"imp":25.4287396653,"elo":924.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11843.1243713497,"imp":24.7714334055,"elo":959.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23463.5184563875,"imp":24.4196340818,"elo":977.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":34621.1415346513,"imp":24.2437727457,"elo":990.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":46109.5334192428,"imp":24.1141000046,"elo":999.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.8979740772,"imp":7.1457344991,"elo":1482.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":20.5017216002,"imp":15.5036655668,"elo":1033.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":53.7125230803,"imp":14.8522098568,"elo":1061.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":118.8503302159,"imp":13.6912428585,"elo":1126.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":237.0519073624,"imp":13.0974908775,"elo":1157.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":547.271171747,"imp":12.2733894118,"elo":1193.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1129.5925788802,"imp":11.7896668144,"elo":1220.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2199.7106433499,"imp":11.2664058306,"elo":1244.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3318.9517565565,"imp":11.0270650726,"elo":1258.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4452.1085706374,"imp":10.8530281343,"elo":1273.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":2.1459956511,"imp":13.0504189304,"elo":1166.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":8.3667450823,"imp":11.2771958175,"elo":1279.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":20.7207184201,"imp":10.5345260026,"elo":1343.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":41.7239328404,"imp":10.1437778435,"elo":1379.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":105.2854855499,"imp":9.8416341087,"elo":1406.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":210.1253434414,"imp":9.6989215232,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":421.9300367907,"imp":9.6014900362,"elo":1429.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":635.5350385867,"imp":9.5372274664,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":848.9941919866,"imp":9.5122528052,"elo":1431.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":373.4221052753,"imp":19.2497229645,"elo":871.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":839.9616197382,"imp":14.9056875458,"elo":1056.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":2344.8516752833,"imp":16.2691579543,"elo":1259.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":4981.7847745692,"imp":14.5448573032,"elo":1348.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":12170.8681485011,"imp":8.5512134328,"elo":1498.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":12741.0939157936,"imp":8.5336049305,"elo":1500.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.469327894,"imp":14.2952393944,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":6.0503740942,"imp":12.9135512688,"elo":1262.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":17.3170071504,"imp":11.1117919108,"elo":1332.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":34.6552917016,"imp":9.8076974115,"elo":1396.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":88.3965998327,"imp":8.6905248694,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":176.967409496,"imp":8.3533128978,"elo":1483.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":353.1777869696,"imp":8.1041203633,"elo":1496.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":528.7105932474,"imp":7.99224468,"elo":1508.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":708.9450090283,"imp":7.9659724924,"elo":1512.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0182719821,"x_train":9.0226914704,"imp":10.6649854115,"elo":1387.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0761788183,"x_train":22.006737996,"imp":10.0410335189,"elo":1430.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1700551281,"x_train":54.5936853194,"imp":9.3422932439,"elo":1493.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2026453565,"x_train":108.5250024926,"imp":9.0363696279,"elo":1507.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2388904639,"x_train":277.687397929,"imp":8.7304375075,"elo":1518.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3111317954,"x_train":564.9102272658,"imp":8.6385669356,"elo":1520.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.398317095,"x_train":1124.7408503677,"imp":8.5860979377,"elo":1520.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3902829889,"x_train":1677.2797709559,"imp":8.5542513657,"elo":1524.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"x_train":2267.1899830578,"imp":8.537762157,"elo":1525.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"x_train":2.4559533883,"imp":13.5184798184,"elo":1230.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0371787849,"x_train":4.0653056289,"imp":13.1595745141,"elo":1250.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0917331108,"x_train":9.725445997,"imp":12.8184880099,"elo":1292.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1170916349,"x_train":19.7798589835,"imp":12.6146892497,"elo":1312.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1607817441,"x_train":50.0934628362,"imp":12.3717353184,"elo":1329.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1954148357,"x_train":100.1940386909,"imp":12.2119545046,"elo":1348.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.246440357,"x_train":200.5360400335,"imp":12.0503277586,"elo":1361.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2772856864,"x_train":300.5068144065,"imp":11.984244047,"elo":1371.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":402.9323608119,"imp":11.9597608126,"elo":1378.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.01204424,"x_train":7.5931963365,"imp":16.4815170931,"elo":1032.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0228552037,"x_train":19.5464706545,"imp":15.9818225032,"elo":1070.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0418083956,"x_train":52.461846709,"imp":15.5322088782,"elo":1125.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0609044025,"x_train":101.1927167213,"imp":15.3095708687,"elo":1139.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0833477059,"x_train":250.7607078715,"imp":15.0786611831,"elo":1158.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.1012762431,"x_train":505.4611049176,"imp":15.0054156631,"elo":1165.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1012126631,"x_train":1000.4774432421,"imp":14.9259310452,"elo":1175.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1049257611,"x_train":1490.4477925659,"imp":14.8788837948,"elo":1178.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1032251722,"x_train":2006.5538981646,"imp":14.8444889121,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2422366525,"x_train":2.5746349456,"imp":15.8576388612,"elo":1074.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2768373111,"x_train":3.1741914103,"imp":15.312547953,"elo":1115.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3337218245,"x_train":6.7533015627,"imp":14.5183742413,"elo":1190.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2738881738,"x_train":13.1877719851,"imp":14.1351179951,"elo":1214.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.371439796,"x_train":32.4660623442,"imp":13.547469136,"elo":1242.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.4677010313,"x_train":64.6497454444,"imp":13.4253490814,"elo":1255.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.6995667618,"x_train":128.5728546231,"imp":13.3545124869,"elo":1262.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.7460530013,"x_train":192.4203093077,"imp":13.3423915858,"elo":1261.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8125804985,"x_train":257.6385593165,"imp":13.3444260284,"elo":1261.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.2929314919,"x_train":3.0840796949,"imp":20.1459917158,"elo":866.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4544855614,"x_train":5.6240636486,"imp":19.8504258159,"elo":885.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.8902529781,"x_train":13.6732407308,"imp":18.9476612196,"elo":928.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3426618483,"x_train":27.2107956399,"imp":18.4865676439,"elo":942.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.7420033115,"x_train":65.2537322804,"imp":17.520082439,"elo":978.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.0242567996,"x_train":130.574483839,"imp":16.9310727816,"elo":1002.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.5714421463,"x_train":262.2064329926,"imp":16.338156479,"elo":1022.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.7283647341,"x_train":393.3022867336,"imp":15.93147183,"elo":1048.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8297120981,"x_train":527.2961316043,"imp":15.6594635682,"elo":1061.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2962263441,"x_train":2.0296176424,"imp":12.3451040358,"elo":1201.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4089228478,"x_train":5.8086693996,"imp":11.4997719236,"elo":1264.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.8292153136,"x_train":16.4346292254,"imp":10.4066250539,"elo":1373.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.8037515446,"x_train":32.0193358834,"imp":9.8788450456,"elo":1427.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.2339770555,"x_train":82.7850514402,"imp":9.3381159903,"elo":1499.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.9246780943,"x_train":169.4629431685,"imp":9.1757274338,"elo":1526.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.5906150954,"x_train":340.4054795841,"imp":8.9869444807,"elo":1548.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.9138833177,"x_train":509.4885196025,"imp":8.8864521179,"elo":1559.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.476648655,"x_train":686.4972156667,"imp":8.7938407814,"elo":1574.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":14.072768612,"x_train":59.3758334402,"imp":10.3904487974,"elo":1335.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":1.5880635253,"x_train":71.0123333261,"imp":9.9244566274,"elo":1334.5,"imputed":true,"imputed_pct":46.1538461538,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2894827463,"x_train":16.0749964455,"imp":10.7841513848,"elo":1414.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.6821414906,"x_train":38.647009759,"imp":10.2023184034,"elo":1444.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.4325592808,"x_train":106.244026897,"imp":9.4160045764,"elo":1499.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.1934690352,"x_train":199.3707655728,"imp":9.0269539703,"elo":1542.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.1316083376,"x_train":462.7480963285,"imp":8.6556576511,"elo":1576.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.3124525847,"x_train":955.9940479091,"imp":8.4515123264,"elo":1583.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.723377164,"x_train":1870.556153527,"imp":8.3000951981,"elo":1589.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.1097813681,"x_train":2790.7899911449,"imp":8.1946214692,"elo":1602.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.5805507968,"x_train":3811.6104621203,"imp":8.1654997251,"elo":1604.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":1.3455286154,"x_train":6.8642025254,"imp":4.1176969928,"elo":1809.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0067377719,"x_train":11.8050270551,"imp":3.0630923517,"elo":1916.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0171543088,"x_train":22.8030707188,"imp":18.2955753566,"elo":966.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1298234739,"x_train":53.0597722616,"imp":16.107246993,"elo":1072.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3758127641,"x_train":147.0385313693,"imp":14.3675168325,"elo":1159.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4333863798,"x_train":177.9195746923,"imp":14.3437457459,"elo":1169.3,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0638540688,"x_train":0.6331018043,"imp":16.6499447959,"elo":1008.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468165116,"x_train":3.5138045858,"imp":15.8760735961,"elo":1070.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2412604538,"x_train":12.8062748979,"imp":15.3908176771,"elo":1107.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3470587813,"x_train":26.5826232412,"imp":14.9942764132,"elo":1129.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5360758678,"x_train":64.673699967,"imp":14.7770918641,"elo":1145.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.6230115397,"x_train":130.0298950791,"imp":14.6901072863,"elo":1154.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7069692121,"x_train":259.9521641476,"imp":14.638686862,"elo":1163.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7234915951,"x_train":391.5234331877,"imp":14.6211649426,"elo":1167.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7700427525,"x_train":526.192235626,"imp":14.6157763538,"elo":1168.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4152312914,"x_train":9.1753045614,"imp":11.4826162385,"elo":1260.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":2.1252015508,"x_train":23.6721342187,"imp":9.3869073322,"elo":1368.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":3.7976884724,"x_train":78.6133073597,"imp":7.3521875932,"elo":1526.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":5.7937010153,"x_train":191.0765941914,"imp":6.6293265441,"elo":1573.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":6.6360113436,"x_train":503.1010684492,"imp":6.1625582504,"elo":1623.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.336614684,"x_train":1009.2362249093,"imp":5.9055348463,"elo":1660.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":9.6345512743,"x_train":2015.6576136475,"imp":5.7842906504,"elo":1678.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.7498142175,"x_train":3026.2086158211,"imp":5.7349512293,"elo":1692.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.9105234185,"x_train":4057.3100555326,"imp":5.6951806629,"elo":1698.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5191857229,"x_train":6.9411546573,"imp":5.1081297102,"elo":1639.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3928986427,"x_train":17.2836445024,"imp":4.6374541046,"elo":1666.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.831577433,"x_train":47.5797720049,"imp":3.8614140786,"elo":1719.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.3479395149,"x_train":98.3276706028,"imp":3.4224976388,"elo":1781.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.8281839931,"x_train":249.0037143861,"imp":2.9685684337,"elo":1844.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.5954304941,"x_train":503.1695046416,"imp":2.7947420106,"elo":1868.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.945912213,"x_train":1018.7089146459,"imp":2.7018005663,"elo":1898.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.9331824935,"x_train":1526.668944465,"imp":2.7311655156,"elo":1899.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.3660135567,"x_train":2043.5362680496,"imp":2.7245619429,"elo":1904.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":6.5421299186,"x_train":18.769283676,"imp":11.0464917637,"elo":1291.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"x_train":44.6485648211,"imp":5.6593617074,"elo":1644.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":54.7143383617,"x_train":88.2308458205,"imp":5.2454696275,"elo":1673.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":82.6659047931,"x_train":186.0445098453,"imp":4.8490533938,"elo":1698.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":109.9030198642,"x_train":341.1376387803,"imp":4.5346498664,"elo":1733.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":170.7606880264,"x_train":866.4570381261,"imp":4.2812155017,"elo":1764.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":225.019656008,"x_train":1760.5896780945,"imp":4.1783339382,"elo":1778.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":283.327169958,"x_train":3528.4122296034,"imp":4.0803258048,"elo":1795.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":320.2042002918,"x_train":5308.8782679903,"imp":4.0274723245,"elo":1809.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":336.9071136327,"x_train":7126.6747880976,"imp":4.0079622689,"elo":1816.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1829385508,"x_train":2.0183513985,"imp":5.431606219,"elo":1661.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":6.4924694208,"x_train":40.7926562599,"imp":1.4100424178,"elo":2207.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1218380804,"x_train":1.3151215973,"imp":4.3753875956,"elo":1718.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.1337247011,"x_train":9.7873678423,"imp":9.9099485615,"elo":1291.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.2761440315,"x_train":23.4153138909,"imp":9.3565549555,"elo":1340.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.551058693,"x_train":71.5165108752,"imp":8.7147560503,"elo":1381.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.760580889,"x_train":166.1346310591,"imp":8.182159933,"elo":1430.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":1.0490418104,"x_train":413.5775428251,"imp":7.7342280619,"elo":1464.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":1.3239622098,"x_train":844.0698066634,"imp":7.4331475795,"elo":1495.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":1.6840673106,"x_train":1693.1667185284,"imp":7.2469532672,"elo":1514.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":1.9026564546,"x_train":2557.2442472516,"imp":7.2041346491,"elo":1517.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":2.1105459826,"x_train":3448.7325920652,"imp":7.1664648695,"elo":1517.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6996427484,"x_train":8.4368371839,"imp":3.393775576,"elo":1886.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.3981148822,"x_train":4.0220384031,"imp":2.2837784913,"elo":1912.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.3150187952,"x_train":2.7780429735,"imp":10.4004807181,"elo":1242.7,"imputed":true,"imputed_pct":46.1538461538,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":4.722739333,"x_train":25.3768652833,"imp":9.5979783992,"elo":1279.0,"imputed":true,"imputed_pct":46.1538461538,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":11.9792597974,"x_train":76.7795330932,"imp":8.8677722781,"elo":1321.7,"imputed":true,"imputed_pct":46.1538461538,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":12.3738520398,"x_train":175.8535512882,"imp":8.5986651896,"elo":1337.6,"imputed":true,"imputed_pct":46.1538461538,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":9.3069375156,"x_train":436.6902139142,"imp":8.1773741755,"elo":1371.8,"imputed":true,"imputed_pct":46.1538461538,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":15.9228223289,"x_train":900.8153356806,"imp":8.0237501849,"elo":1381.3,"imputed":true,"imputed_pct":46.1538461538,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":21.7068913692,"x_train":1849.0970608496,"imp":7.8388654555,"elo":1390.2,"imputed":true,"imputed_pct":46.1538461538,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.5386193489,"x_train":2791.064560621,"imp":7.8057075844,"elo":1394.7,"imputed":true,"imputed_pct":46.1538461538,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":10.4125412162,"x_train":3805.6199685881,"imp":7.7716997285,"elo":1396.9,"imputed":true,"imputed_pct":46.1538461538,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.5829823393,"x_train":302.8991048646,"imp":27.4310818967,"elo":788.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.0719468341,"x_train":632.7525144671,"imp":27.2879140579,"elo":810.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":12.0770172729,"x_train":1217.0384701627,"imp":26.9251443516,"elo":845.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":14.049836544,"x_train":2404.0220528997,"imp":26.2981860908,"elo":884.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.2612731687,"x_train":5871.8477809815,"imp":25.4287396653,"elo":924.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.7093243804,"x_train":11843.1243713497,"imp":24.7714334055,"elo":959.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":18.9694047613,"x_train":23463.5184563875,"imp":24.4196340818,"elo":977.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":20.7650487513,"x_train":34621.1415346513,"imp":24.2437727457,"elo":990.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":22.148843297,"x_train":46109.5334192428,"imp":24.1141000046,"elo":999.8,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0719372146,"x_train":0.8979740772,"imp":7.1457344991,"elo":1482.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0813896645,"x_train":20.5017216002,"imp":15.5036655668,"elo":1033.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1755690741,"x_train":53.7125230803,"imp":14.8522098568,"elo":1061.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.3675462419,"x_train":118.8503302159,"imp":13.6912428585,"elo":1126.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5785459292,"x_train":237.0519073624,"imp":13.0974908775,"elo":1157.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.7589743999,"x_train":547.271171747,"imp":12.2733894118,"elo":1193.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.982110054,"x_train":1129.5925788802,"imp":11.7896668144,"elo":1220.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.1225759388,"x_train":2199.7106433499,"imp":11.2664058306,"elo":1244.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8530829243,"x_train":3318.9517565565,"imp":11.0270650726,"elo":1258.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.8515826242,"x_train":4452.1085706374,"imp":10.8530281343,"elo":1273.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.1795378646,"x_train":2.1459956511,"imp":13.0504189304,"elo":1166.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.6639550887,"x_train":8.3667450823,"imp":11.2771958175,"elo":1279.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.1108368303,"x_train":20.7207184201,"imp":10.5345260026,"elo":1343.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.2649740084,"x_train":41.7239328404,"imp":10.1437778435,"elo":1379.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.6863086909,"x_train":105.2854855499,"imp":9.8416341087,"elo":1406.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0386947605,"x_train":210.1253434414,"imp":9.6989215232,"elo":1419.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0268020363,"x_train":421.9300367907,"imp":9.6014900362,"elo":1429.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.2593267071,"x_train":635.5350385867,"imp":9.5372274664,"elo":1431.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3747946103,"x_train":848.9941919866,"imp":9.5122528052,"elo":1431.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":81.4898053701,"x_train":373.4221052753,"imp":19.2497229645,"elo":871.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":133.265500669,"x_train":839.9616197382,"imp":14.9056875458,"elo":1056.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":212.0443313695,"x_train":2344.8516752833,"imp":16.2691579543,"elo":1259.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":304.6354469252,"x_train":4981.7847745692,"imp":14.5448573032,"elo":1348.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":380.3901708253,"x_train":12170.8681485011,"imp":8.5512134328,"elo":1498.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":387.572909971,"x_train":12741.0939157936,"imp":8.5336049305,"elo":1500.4,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.7548401788,"x_train":2.469327894,"imp":14.2952393944,"elo":1212.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.8950323165,"x_train":6.0503740942,"imp":12.9135512688,"elo":1262.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.0102057688,"x_train":17.3170071504,"imp":11.1117919108,"elo":1332.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.126261214,"x_train":34.6552917016,"imp":9.8076974115,"elo":1396.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":1.4472187823,"x_train":88.3965998327,"imp":8.6905248694,"elo":1452.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":1.7895630258,"x_train":176.967409496,"imp":8.3533128978,"elo":1483.9,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":1.8867738381,"x_train":353.1777869696,"imp":8.1041203633,"elo":1496.2,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.1903114886,"x_train":528.7105932474,"imp":7.99224468,"elo":1508.0,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":1.9734199806,"x_train":708.9450090283,"imp":7.9659724924,"elo":1512.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..5da51daacb65ea4cc59e648afd338267c8ecb317
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_regression/datasets_all</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "Nori-30M (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabPFN-2.6 (default)", "TabDPT (tuned + ensembled)", "Nori (default)", "TabDPT (tuned)", "RealTabPFN-2.5 (tuned)", "TabICLv2 (default)", "RealMLP (tuned + ensembled)", "AutoGluon 1.4 (best, 4h)", "TabDPT-Turbo (default)", "RealTabPFN-2.5 (default)", "TabDPT (default)", "ModernNCA (tuned + ensembled)", "LightGBM (tuned + ensembled)", "RealMLP (tuned)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "TabM (tuned + ensembled)", "xRFM (tuned + ensembled)", "iLTM (tuned + ensembled)", "TabSwift (default)", "LightGBM (tuned)", "TabM (tuned)", "xRFM (tuned)", "XGBoost (tuned)", "XGBoost (tuned + ensembled)", "ModernNCA (default)", "TabPFNv2 (tuned + ensembled)", "CatBoost (default)", "ModernNCA (tuned)", "ChimeraBoost (tuned + ensembled)", "iLTM (tuned)", "Mitra (default)", "LimiX (default)", "ChimeraBoost (tuned)", "TabM (default)", "TabPFNv2 (tuned)", "SAP-RPT-OSS (default)", "RealMLP (default)", "TorchMLP (tuned + ensembled)", "ExtraTrees (tuned + ensembled)", "TabPFNv2 (default)", "ChimeraBoost (default)", "ExtraTrees (tuned)", "TorchMLP (tuned)", "xRFM (default)", "LightGBM (default)", "PerpetualBooster (tuned + ensembled)", "EBM (tuned + ensembled)", "XGBoost (default)", "RandomForest (tuned + ensembled)", "PerpetualBooster (tuned)", "EBM (tuned)", "RandomForest (tuned)", "ExtraTrees (default)", "FastaiMLP (tuned + ensembled)", "TorchMLP (default)", "EBM (default)", "FastaiMLP (tuned)", "RandomForest (default)", "TabSTAR (tuned + ensembled)", "PerpetualBooster (default)", "TabSTAR (tuned)", "FastaiMLP (default)", "iLTM (default)", "KNN (tuned + ensembled)", "TabSTAR (default)", "KNN (tuned)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.3076923191547394, NaN, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.23076923191547394, 0.6153846383094788, NaN, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.6153846383094788, 0.38461539149284363, NaN, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.8461538553237915, 0.5384615659713745, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.5384615659713745, 0.4615384638309479, 0.4615384638309479, NaN, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, NaN, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.7692307829856873, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, NaN, 0.4615384638309479, 0.8461538553237915, 0.4615384638309479, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, NaN, 0.692307710647583, 0.4615384638309479, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.38461539149284363, 0.38461539149284363, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.1538461595773697, 0.3076923191547394, NaN, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.5384615659713745, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.1538461595773697, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, 0.23076923191547394, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, NaN, 0.3076923191547394, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.5384615659713745, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.5384615659713745, 0.692307710647583, NaN, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, NaN, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.7692307829856873, 1.0, 1.0, 0.7692307829856873, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 0.7692307829856873, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.38461539149284363, 0.23076923191547394, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.4615384638309479, NaN, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.9230769276618958, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.6153846383094788, 1.0, 0.7692307829856873, 1.0, 0.9230769276618958, 0.7692307829856873, 0.7692307829856873, 1.0, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 1.0, 0.9230769276618958, 0.7692307829856873, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, NaN, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 1.0, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.692307710647583, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.4615384638309479, 0.4615384638309479, 0.23076923191547394, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.6153846383094788, NaN, 0.5384615659713745, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.23076923191547394, 0.0, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, NaN, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 1.0, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, NaN, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 1.0, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.07692307978868484, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, NaN, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.4615384638309479, 0.8461538553237915, 0.5384615659713745, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.6153846383094788, 0.9230769276618958, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.692307710647583, 1.0, 0.9230769276618958, 0.6153846383094788, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.692307710647583, 1.0, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.0, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.4615384638309479, 0.3076923191547394, 0.5384615659713745, NaN, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 1.0, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 1.0, 1.0, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, NaN, 0.692307710647583, 0.38461539149284363, 0.3076923191547394, 0.4615384638309479, 0.38461539149284363, 0.692307710647583, 0.38461539149284363, 0.5384615659713745, 0.7692307829856873, 0.7692307829856873, 0.5384615659713745, 0.6153846383094788, 0.9230769276618958, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 1.0, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.3076923191547394, 0.3076923191547394, NaN, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.6153846383094788, 0.4615384638309479, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.6153846383094788, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 1.0, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.6153846383094788, 0.5384615659713745, NaN, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.8461538553237915, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.4615384638309479, 0.6153846383094788, 0.7692307829856873, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.8461538553237915, 1.0, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.5384615659713745, 0.38461539149284363, 0.692307710647583, 0.5384615659713745, 0.4615384638309479, NaN, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.8461538553237915, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.3076923191547394, NaN, 0.38461539149284363, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 1.0, 0.9230769276618958, 0.7692307829856873, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, NaN, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.5384615659713745, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, NaN, 0.38461539149284363, 0.4615384638309479, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.5384615659713745, 0.7692307829856873, 0.9230769276618958, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.38461539149284363, 0.4615384638309479, 0.3076923191547394, 0.6153846383094788, 0.5384615659713745, 0.1538461595773697, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.6153846383094788, NaN, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.38461539149284363, 0.5384615659713745, 0.7692307829856873, 0.4615384638309479, 0.7692307829856873, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 1.0, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.1538461595773697, 0.5384615659713745, 0.3076923191547394, 0.5384615659713745, 0.5384615659713745, NaN, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.7692307829856873, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 1.0, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, NaN, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.4615384638309479, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, NaN, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 1.0, 1.0, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 0.8461538553237915, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.5384615659713745, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.4615384638309479, 0.6153846383094788, 0.23076923191547394, 0.4615384638309479, 0.5384615659713745, NaN, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.9230769276618958, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, NaN, 0.4615384638309479, 0.38461539149284363, 0.5384615659713745, 0.4615384638309479, 0.7307692170143127, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.7692307829856873, 0.7692307829856873, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.807692289352417, 0.6153846383094788, 0.4615384638309479, 0.692307710647583, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.7692307829856873, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.807692289352417, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.38461539149284363, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.23076923191547394, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, 0.38461539149284363, 0.5384615659713745, NaN, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 1.0, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 1.0, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.0, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.5384615659713745, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, NaN, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 1.0, 1.0, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, NaN, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.9230769276618958, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, NaN, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 1.0, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.26923078298568726, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, NaN, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.42307692766189575, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.692307710647583, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.8461538553237915, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.8461538553237915, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.5384615659713745, NaN, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, NaN, 0.6153846383094788, 0.5384615659713745, 0.4615384638309479, 0.7692307829856873, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.0, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.0, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.5384615659713745, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, NaN, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.692307710647583, 0.9230769276618958, 0.692307710647583, 0.8461538553237915, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.5769230723381042, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, NaN, 0.6153846383094788, 0.38461539149284363, 0.5384615659713745, 0.4615384638309479, 0.5769230723381042, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.7692307829856873, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7307692170143127, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.38461539149284363, NaN, 0.5384615659713745, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.692307710647583, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.6153846383094788, 0.4615384638309479, NaN, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 1.0, 0.9230769276618958, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.4615384638309479, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, NaN, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 1.0, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.5384615659713745, 0.3076923191547394, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.38461539149284363, 0.3076923191547394, 0.5384615659713745, 0.6153846383094788, 0.38461539149284363, 0.4615384638309479, NaN, 0.5384615659713745, 0.692307710647583, 0.7692307829856873, 0.4615384638309479, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 1.0, 0.7692307829856873, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.4615384638309479, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.19230769574642181, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.42307692766189575, 0.6153846383094788, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, NaN, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.38461539149284363, 0.692307710647583, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, 0.692307710647583, 0.5384615659713745, 0.7692307829856873, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.23076923191547394, 0.0, 0.0, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.38461539149284363, 0.07692307978868484, 0.38461539149284363, 0.07692307978868484, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.1538461595773697, 0.38461539149284363, 0.5384615659713745, 0.3076923191547394, 0.4615384638309479, 0.5384615659713745, 0.3076923191547394, 0.6153846383094788, NaN, 0.4615384638309479, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.9230769276618958, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.5384615659713745, 0.1538461595773697, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, 0.5384615659713745, 0.5384615659713745, 0.38461539149284363, 0.3076923191547394, 0.5384615659713745, 0.5384615659713745, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.5384615659713745, 0.5384615659713745, NaN, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.6153846383094788, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.7692307829856873, 0.6153846383094788, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.4615384638309479, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.4615384638309479, 0.4615384638309479, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, 0.23076923191547394, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, NaN, 0.6153846383094788, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 1.0, 0.7692307829856873, 0.8461538553237915, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.5384615659713745, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.1538461595773697, 0.5384615659713745, 0.4615384638309479, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.4615384638309479, 0.38461539149284363, NaN, 0.38461539149284363, 0.3076923191547394, 0.4615384638309479, 0.38461539149284363, 0.6153846383094788, 0.4615384638309479, 0.4615384638309479, 0.692307710647583, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.1538461595773697, 0.4615384638309479, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.1538461595773697, 0.4615384638309479, 0.4615384638309479, 0.3076923191547394, 0.1538461595773697, 0.6153846383094788, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.5384615659713745, 0.6153846383094788, 0.38461539149284363, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, NaN, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.5384615659713745, 0.692307710647583, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.0, 0.3076923191547394, 0.4615384638309479, 0.0, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.4615384638309479, 0.4615384638309479, 0.07692307978868484, 0.1538461595773697, 0.6153846383094788, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.6153846383094788, 0.07692307978868484, 0.38461539149284363, 0.4615384638309479, 0.692307710647583, 0.38461539149284363, NaN, 0.6153846383094788, 0.5384615659713745, 0.5384615659713745, 0.807692289352417, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 1.0, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, 0.38461539149284363, NaN, 0.4615384638309479, 0.38461539149284363, 0.5384615659713745, 0.8461538553237915, 0.38461539149284363, 0.38461539149284363, 0.6153846383094788, 0.7692307829856873, 1.0, 0.692307710647583, 0.5384615659713745, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.1538461595773697, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.3076923191547394, 0.4615384638309479, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.1538461595773697, 0.07692307978868484, 0.5384615659713745, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.38461539149284363, 0.6153846383094788, 0.3076923191547394, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.3076923191547394, 0.4615384638309479, 0.5384615659713745, NaN, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.1538461595773697, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.4615384638309479, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.5384615659713745, 0.38461539149284363, 0.23076923191547394, 0.38461539149284363, 0.1538461595773697, 0.5384615659713745, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.4615384638309479, NaN, 0.5384615659713745, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.5384615659713745, 0.5384615659713745, 0.6153846383094788, 0.5384615659713745, 0.9230769276618958, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.8461538553237915, 1.0, 1.0, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.4615384638309479, 0.0, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.4615384638309479, 0.38461539149284363, 0.1538461595773697, 0.1538461595773697, 0.5384615659713745, 0.4615384638309479, 0.3076923191547394, 0.4615384638309479, 0.23076923191547394, 0.6153846383094788, 0.1538461595773697, 0.23076923191547394, 0.4615384638309479, 0.5384615659713745, 0.3076923191547394, 0.19230769574642181, 0.4615384638309479, 0.38461539149284363, 0.4615384638309479, NaN, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.3076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.38461539149284363, 0.38461539149284363, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, NaN, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.6153846383094788, 0.8461538553237915, 0.6153846383094788, 0.5384615659713745, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.692307710647583, 0.7692307829856873, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.4615384638309479, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.6153846383094788, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.6153846383094788, NaN, 0.5384615659713745, 0.5384615659713745, 0.4615384638309479, 0.6153846383094788, 0.5384615659713745, 0.7307692170143127, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.7692307829856873, 0.9230769276618958, 1.0, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.23076923191547394, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.38461539149284363, 0.0, 0.23076923191547394, 0.3076923191547394, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.6153846383094788, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.6153846383094788, 0.4615384638309479, NaN, 0.4615384638309479, 0.38461539149284363, 0.6153846383094788, 0.4615384638309479, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.9230769276618958, 0.7692307829856873, 0.9230769276618958, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.4615384638309479, 0.23076923191547394, 0.23076923191547394, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 0.23076923191547394, 0.5384615659713745, 0.4615384638309479, 0.5384615659713745, NaN, 0.5384615659713745, 0.6153846383094788, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 1.0, 0.7692307829856873, 0.692307710647583, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.3076923191547394, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.0, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.0, 0.4615384638309479, 0.4615384638309479, 0.38461539149284363, 0.5384615659713745, 0.0, 0.38461539149284363, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.4615384638309479, 0.3076923191547394, 0.38461539149284363, 0.5384615659713745, 0.6153846383094788, 0.4615384638309479, NaN, 0.38461539149284363, 0.4615384638309479, 0.6153846383094788, 0.692307710647583, 0.38461539149284363, 0.692307710647583, 0.4615384638309479, 0.692307710647583, 0.692307710647583, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.0, 0.1538461595773697, 0.07692307978868484, 0.0, 0.0, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.0, 0.23076923191547394, 0.38461539149284363, 0.23076923191547394, 0.1538461595773697, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.6153846383094788, NaN, 0.5384615659713745, 0.4615384638309479, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.6153846383094788, 0.692307710647583, 0.692307710647583, 0.9230769276618958, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.1538461595773697, 0.38461539149284363, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.4615384638309479, 0.23076923191547394, 0.38461539149284363, 0.4615384638309479, 0.5384615659713745, 0.38461539149284363, 0.5384615659713745, 0.4615384638309479, NaN, 0.5384615659713745, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.692307710647583, 0.6153846383094788, 0.8461538553237915, 0.7692307829856873, 0.9230769276618958, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.19230769574642181, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.0, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.26923078298568726, 0.3076923191547394, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, 0.1538461595773697, 0.07692307978868484, 0.38461539149284363, 0.1538461595773697, 0.07692307978868484, 0.23076923191547394, 0.4615384638309479, 0.23076923191547394, 0.07692307978868484, 0.23076923191547394, 0.4615384638309479, 0.26923078298568726, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.5384615659713745, 0.4615384638309479, NaN, 0.7692307829856873, 0.6153846383094788, 0.7692307829856873, 0.692307710647583, 0.7692307829856873, 0.8461538553237915, 0.8461538553237915, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.1538461595773697, 0.1538461595773697, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, NaN, 0.23076923191547394, 0.8461538553237915, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 1.0, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.3076923191547394, 0.1538461595773697, 0.0, 0.23076923191547394, 0.07692307978868484, 0.38461539149284363, 0.07692307978868484, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.3076923191547394, 0.6153846383094788, 0.38461539149284363, 0.38461539149284363, 0.38461539149284363, 0.7692307829856873, NaN, 0.7692307829856873, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.692307710647583, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.23076923191547394, 0.3076923191547394, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.23076923191547394, 0.1538461595773697, 0.23076923191547394, NaN, 0.38461539149284363, 0.38461539149284363, 0.4615384638309479, 1.0, 0.6153846383094788, 0.6153846383094788, 0.8461538553237915, 0.8461538553237915, 0.9230769276618958], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.3076923191547394, 0.0, 0.5384615659713745, 0.38461539149284363, 0.3076923191547394, 0.3076923191547394, 0.6153846383094788, 0.5384615659713745, 0.6153846383094788, NaN, 0.4615384638309479, 0.4615384638309479, 0.6153846383094788, 0.6153846383094788, 0.7692307829856873, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.0, 0.07692307978868484, 0.23076923191547394, 0.07692307978868484, 0.0, 0.07692307978868484, 0.3076923191547394, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.23076923191547394, 0.6153846383094788, 0.38461539149284363, 0.6153846383094788, 0.5384615659713745, NaN, 0.692307710647583, 0.7692307829856873, 0.7692307829856873, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 0.9230769276618958], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.1538461595773697, 0.0, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.0, 0.23076923191547394, 0.23076923191547394, 0.0, 0.07692307978868484, 0.3076923191547394, 0.3076923191547394, 0.3076923191547394, 0.38461539149284363, 0.1538461595773697, 0.5384615659713745, 0.38461539149284363, 0.5384615659713745, 0.5384615659713745, 0.3076923191547394, NaN, 0.5384615659713745, 1.0, 1.0, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.0, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.0, 0.1538461595773697, 0.0, 0.38461539149284363, 0.23076923191547394, 0.4615384638309479, NaN, 0.5384615659713745, 0.5384615659713745, 0.8461538553237915, 0.8461538553237915, 0.8461538553237915], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.23076923191547394, 0.1538461595773697, 0.1538461595773697, 0.23076923191547394, 0.0, 0.38461539149284363, 0.3076923191547394, 0.38461539149284363, 0.38461539149284363, 0.23076923191547394, 0.0, 0.4615384638309479, NaN, 0.8461538553237915, 0.9230769276618958, 0.9230769276618958, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.0, 0.0, 0.07692307978868484, 0.1538461595773697, 0.1538461595773697, 0.07692307978868484, 0.0, 0.38461539149284363, 0.1538461595773697, 0.38461539149284363, 0.23076923191547394, 0.1538461595773697, 0.0, 0.4615384638309479, 0.1538461595773697, NaN, 0.692307710647583, 0.692307710647583, 0.692307710647583], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, NaN, 0.807692289352417, 0.8461538553237915], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.0, 0.0, 0.1538461595773697, 0.07692307978868484, 0.1538461595773697, 0.0, 0.07692307978868484, 0.07692307978868484, 0.1538461595773697, 0.07692307978868484, 0.3076923191547394, 0.19230769574642181, NaN, 0.7692307829856873], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.07692307978868484, 0.0, 0.1538461595773697, 0.0, 0.3076923191547394, 0.1538461595773697, 0.23076923191547394, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9764102578},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.9087179522},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.9056410303},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.9015384664},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.8912820564},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8861538508},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.857435903},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.8564102619},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.8276923136},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.8276923142},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8071794956},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.78358975},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7815384667},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7610256476},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.7569230855},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7528205211},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.727179496},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6902564176},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6871794963},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6646153913},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6543589812},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6533333424},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6512820604},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6430769312},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.633846164},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.6153846235},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6153846234},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.6041025727},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5835897509},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5825641088},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.568205137},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5625641112},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.56205129},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.5507692411},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5415384712},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.5179487264},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.5005128292},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4994871912},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4923077001},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4707692388},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.4702564166},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.4594871875},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.4410256487},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4400000079},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4400000082},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.4235897519},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4184615462},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4153846239},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3979487271},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.3938461615},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3938461611},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.370769237},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3702564173},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.3661538532},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3569230836},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.3379487235},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.3312820578},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3246153904},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2953846208},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.2728205178},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.258461543},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.2564102617},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2502564153},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2415384654},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2276923123},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2112820555},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2020512862},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1712820552},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.1610256445},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1569230804},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.1292307728},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1005128219},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0728205147},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0405128211},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0312820522},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0143589748}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_all/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e5f3f3e0f9c4b0fa2b580f16799fb16887026c10
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_all/winrate_matrix.csv
@@ -0,0 +1,77 @@
+,TabFM (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),Nori-30M (default),RealTabPFN-2.5 (tuned + ensembled),TabPFN-2.6 (default),TabDPT (tuned + ensembled),Nori (default),TabDPT (tuned),RealTabPFN-2.5 (tuned),TabICLv2 (default),RealMLP (tuned + ensembled),"AutoGluon 1.4 (best, 4h)",TabDPT-Turbo (default),RealTabPFN-2.5 (default),TabDPT (default),ModernNCA (tuned + ensembled),LightGBM (tuned + ensembled),RealMLP (tuned),CatBoost (tuned + ensembled),CatBoost (tuned),TabM (tuned + ensembled),xRFM (tuned + ensembled),iLTM (tuned + ensembled),TabSwift (default),LightGBM (tuned),TabM (tuned),xRFM (tuned),XGBoost (tuned),XGBoost (tuned + ensembled),ModernNCA (default),TabPFNv2 (tuned + ensembled),CatBoost (default),ModernNCA (tuned),ChimeraBoost (tuned + ensembled),iLTM (tuned),Mitra (default),LimiX (default),ChimeraBoost (tuned),TabM (default),TabPFNv2 (tuned),SAP-RPT-OSS (default),RealMLP (default),TorchMLP (tuned + ensembled),ExtraTrees (tuned + ensembled),TabPFNv2 (default),ChimeraBoost (default),ExtraTrees (tuned),TorchMLP (tuned),xRFM (default),LightGBM (default),PerpetualBooster (tuned + ensembled),EBM (tuned + ensembled),XGBoost (default),RandomForest (tuned + ensembled),PerpetualBooster (tuned),EBM (tuned),RandomForest (tuned),ExtraTrees (default),FastaiMLP (tuned + ensembled),TorchMLP (default),EBM (default),FastaiMLP (tuned),RandomForest (default),TabSTAR (tuned + ensembled),PerpetualBooster (default),TabSTAR (tuned),FastaiMLP (default),iLTM (default),KNN (tuned + ensembled),TabSTAR (default),KNN (tuned),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,0.6923077,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,1.0,0.84615386,0.9230769,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.30769232,,0.3846154,0.3846154,0.46153846,0.61538464,0.53846157,0.61538464,0.61538464,0.61538464,0.46153846,0.84615386,0.9230769,0.9230769,0.6923077,0.7692308,0.6923077,1.0,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.23076923,0.61538464,,0.61538464,0.53846157,0.61538464,0.46153846,0.6923077,0.61538464,0.7692308,0.6923077,0.84615386,0.84615386,0.9230769,0.84615386,1.0,0.84615386,0.9230769,0.84615386,0.84615386,0.84615386,1.0,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,0.9230769,0.9230769,1.0,0.9230769,0.9230769,0.84615386,0.9230769,1.0,0.9230769,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,0.9230769,0.84615386,0.9230769,1.0,1.0,0.84615386,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori-30M (default),0.15384616,0.61538464,0.3846154,,0.53846157,0.61538464,0.53846157,0.6923077,0.84615386,0.53846157,0.61538464,0.7692308,0.84615386,0.9230769,0.6923077,0.7692308,0.6923077,0.9230769,0.7692308,0.9230769,0.9230769,1.0,0.9230769,1.0,0.84615386,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.84615386,0.9230769,0.84615386,0.9230769,1.0,1.0,0.84615386,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.15384616,0.53846157,0.46153846,0.46153846,,0.53846157,0.61538464,0.61538464,0.61538464,0.7692308,0.61538464,0.9230769,0.7692308,0.84615386,0.9230769,0.6923077,0.7692308,0.9230769,0.9230769,0.84615386,0.84615386,0.84615386,0.9230769,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.7692308,1.0,0.9230769,1.0,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.15384616,0.3846154,0.3846154,0.3846154,0.46153846,,0.61538464,0.53846157,0.61538464,0.53846157,0.53846157,0.7692308,0.61538464,0.6923077,0.7692308,0.6923077,0.7692308,0.84615386,1.0,0.9230769,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,1.0,1.0,0.7692308,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.15384616,0.46153846,0.53846157,0.46153846,0.3846154,0.3846154,,0.46153846,0.84615386,0.46153846,0.53846157,0.6923077,0.7692308,0.84615386,0.6923077,0.9230769,0.6923077,0.7692308,0.6923077,0.84615386,0.7692308,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.6923077,0.84615386,0.84615386,0.9230769,1.0,0.9230769,0.84615386,0.9230769,1.0,0.9230769,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,0.84615386,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori (default),0.07692308,0.3846154,0.30769232,0.30769232,0.3846154,0.46153846,0.53846157,,0.6923077,0.46153846,0.61538464,0.6923077,0.61538464,0.6923077,0.53846157,0.6923077,0.61538464,0.84615386,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,1.0,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.84615386,0.7692308,0.9230769,0.7692308,0.9230769,1.0,1.0,0.84615386,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,0.9230769,0.9230769,1.0,1.0,1.0,1.0,0.9230769,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.3846154,0.3846154,0.15384616,0.3846154,0.3846154,0.15384616,0.30769232,,0.53846157,0.46153846,0.6923077,0.7692308,0.84615386,0.53846157,0.84615386,0.6923077,0.84615386,0.6923077,0.84615386,0.7692308,0.7692308,0.9230769,0.84615386,0.7692308,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.61538464,0.84615386,0.7692308,0.9230769,1.0,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0,1.0,0.84615386,0.9230769,1.0,1.0,0.9230769,1.0,1.0,0.84615386,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.15384616,0.3846154,0.23076923,0.46153846,0.23076923,0.46153846,0.53846157,0.53846157,0.46153846,,0.30769232,0.7692308,0.6923077,0.7692308,0.7692308,0.61538464,0.61538464,0.7692308,0.84615386,0.7692308,0.7692308,0.84615386,0.84615386,0.7692308,0.7692308,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.6923077,0.84615386,0.84615386,0.9230769,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,0.84615386,0.84615386,0.9230769,1.0,0.9230769,0.9230769,1.0,0.9230769,1.0,0.84615386,1.0,1.0,0.9230769,1.0,0.9230769,1.0,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,1.0,0.9230769,0.84615386,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.07692308,0.53846157,0.30769232,0.3846154,0.3846154,0.46153846,0.46153846,0.3846154,0.53846157,0.6923077,,0.6923077,0.7692308,0.6923077,0.61538464,0.53846157,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.84615386,0.84615386,0.84615386,0.6923077,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,0.84615386,0.9230769,0.7692308,1.0,0.9230769,0.84615386,0.9230769,0.84615386,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.15384616,0.15384616,0.23076923,0.07692308,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.30769232,,0.53846157,0.61538464,0.53846157,0.46153846,0.53846157,0.6923077,0.7692308,0.84615386,0.84615386,0.6923077,0.84615386,0.84615386,0.7692308,0.7692308,0.7692308,0.9230769,0.9230769,0.9230769,0.9230769,0.61538464,0.84615386,0.7692308,0.84615386,0.9230769,0.7692308,0.7692308,1.0,1.0,0.7692308,0.84615386,1.0,0.9230769,1.0,0.7692308,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.07692308,0.15384616,0.15384616,0.23076923,0.3846154,0.23076923,0.3846154,0.23076923,0.30769232,0.23076923,0.46153846,,0.46153846,0.46153846,0.46153846,0.61538464,0.9230769,0.6923077,0.84615386,0.7692308,0.6923077,0.84615386,0.6923077,0.6923077,0.9230769,0.7692308,0.84615386,0.9230769,1.0,0.9230769,0.61538464,1.0,0.7692308,1.0,0.9230769,0.7692308,0.7692308,1.0,0.9230769,0.6923077,0.7692308,0.9230769,1.0,0.9230769,0.7692308,1.0,0.9230769,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.30769232,0.15384616,0.30769232,0.15384616,0.23076923,0.30769232,0.3846154,0.53846157,,0.3846154,0.46153846,0.53846157,0.6923077,0.7692308,0.84615386,0.7692308,0.61538464,0.84615386,0.7692308,0.7692308,0.84615386,0.7692308,0.9230769,0.9230769,0.9230769,0.84615386,0.61538464,0.84615386,0.84615386,0.84615386,0.9230769,0.7692308,0.7692308,0.84615386,1.0,0.7692308,0.84615386,0.9230769,1.0,1.0,0.6923077,0.84615386,1.0,1.0,0.9230769,1.0,0.9230769,0.84615386,0.9230769,1.0,1.0,0.84615386,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.07692308,0.30769232,0.15384616,0.30769232,0.07692308,0.23076923,0.30769232,0.46153846,0.46153846,0.23076923,0.3846154,0.46153846,0.53846157,0.61538464,,0.53846157,0.61538464,0.6923077,0.61538464,0.7692308,0.7692308,0.7692308,0.7692308,0.7692308,0.84615386,0.6923077,0.7692308,0.84615386,0.7692308,0.7692308,0.7692308,0.61538464,0.84615386,0.7692308,0.84615386,0.7692308,0.7692308,0.6923077,0.84615386,0.9230769,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.84615386,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,0.9230769,0.9230769,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0
+TabDPT (default),0.0,0.23076923,0.0,0.23076923,0.30769232,0.30769232,0.07692308,0.30769232,0.15384616,0.3846154,0.46153846,0.53846157,0.53846157,0.53846157,0.46153846,,0.61538464,0.6923077,0.53846157,0.7692308,0.7692308,0.6923077,0.84615386,0.6923077,0.6923077,0.7692308,0.7692308,0.84615386,0.7692308,0.7692308,0.7692308,0.6923077,0.7692308,0.6923077,0.84615386,0.9230769,0.7692308,0.7692308,0.9230769,0.9230769,0.7692308,0.7692308,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.9230769,0.84615386,0.84615386,0.84615386,1.0,0.9230769,0.84615386,1.0,1.0,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.07692308,0.30769232,0.15384616,0.30769232,0.23076923,0.23076923,0.30769232,0.3846154,0.30769232,0.3846154,0.23076923,0.46153846,0.3846154,0.46153846,0.3846154,0.3846154,,0.6923077,0.6923077,0.53846157,0.53846157,0.61538464,0.6923077,0.6923077,0.7692308,0.84615386,0.61538464,0.6923077,0.7692308,0.84615386,0.7692308,0.61538464,0.7692308,1.0,0.84615386,0.84615386,0.6923077,0.6923077,0.84615386,0.84615386,0.6923077,0.7692308,0.7692308,0.7692308,0.84615386,0.7692308,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,0.84615386,0.9230769,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.15384616,0.30769232,0.07692308,0.30769232,0.30769232,0.30769232,0.30769232,,0.46153846,0.53846157,0.53846157,0.53846157,0.46153846,0.61538464,0.46153846,0.84615386,0.53846157,0.61538464,0.7692308,0.84615386,0.7692308,0.61538464,0.9230769,0.6923077,0.84615386,0.6923077,0.6923077,0.6923077,1.0,0.9230769,0.61538464,0.6923077,0.84615386,0.84615386,0.9230769,0.6923077,1.0,0.9230769,0.84615386,0.84615386,1.0,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.07692308,0.15384616,0.23076923,0.07692308,0.0,0.30769232,0.15384616,0.30769232,0.15384616,0.15384616,0.23076923,0.30769232,0.23076923,0.3846154,0.46153846,0.30769232,0.53846157,,0.61538464,0.6923077,0.61538464,0.61538464,0.61538464,0.53846157,0.6923077,0.6923077,0.7692308,0.7692308,0.7692308,0.7692308,0.61538464,0.61538464,0.7692308,0.61538464,0.7692308,0.7692308,0.6923077,0.7692308,0.9230769,0.61538464,0.7692308,0.84615386,0.84615386,0.7692308,0.6923077,0.7692308,0.7692308,1.0,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,1.0,0.9230769,0.9230769,0.7692308,1.0,1.0,1.0,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.46153846,0.46153846,0.3846154,,0.6923077,0.3846154,0.30769232,0.46153846,0.3846154,0.6923077,0.3846154,0.53846157,0.7692308,0.7692308,0.53846157,0.61538464,0.9230769,0.6923077,0.6923077,0.7692308,0.6923077,0.6923077,0.9230769,0.84615386,0.61538464,0.7692308,0.84615386,0.84615386,0.84615386,0.6923077,1.0,0.9230769,0.84615386,0.84615386,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.15384616,0.07692308,0.15384616,0.07692308,0.23076923,0.07692308,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.46153846,0.46153846,0.30769232,0.30769232,,0.46153846,0.46153846,0.46153846,0.3846154,0.61538464,0.46153846,0.53846157,0.6923077,0.6923077,0.61538464,0.61538464,0.7692308,0.6923077,0.6923077,0.7692308,0.6923077,0.6923077,0.7692308,0.84615386,0.61538464,0.6923077,0.84615386,0.84615386,0.84615386,0.6923077,1.0,0.84615386,0.84615386,0.7692308,0.9230769,1.0,0.9230769,1.0,0.9230769,1.0,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.0,0.0,0.0,0.15384616,0.07692308,0.15384616,0.15384616,0.23076923,0.15384616,0.15384616,0.30769232,0.30769232,0.3846154,0.23076923,0.30769232,0.3846154,0.46153846,0.3846154,0.61538464,0.53846157,,0.53846157,0.53846157,0.46153846,0.6923077,0.84615386,0.53846157,0.6923077,0.6923077,0.46153846,0.61538464,0.7692308,0.53846157,0.6923077,0.6923077,0.61538464,0.6923077,0.84615386,1.0,0.6923077,0.6923077,0.84615386,0.84615386,0.84615386,0.6923077,0.84615386,0.84615386,0.84615386,0.7692308,0.9230769,0.84615386,0.7692308,0.9230769,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,0.9230769,1.0,0.84615386,1.0,1.0,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.15384616,0.30769232,0.53846157,0.3846154,0.6923077,0.53846157,0.46153846,,0.6923077,0.53846157,0.61538464,0.53846157,0.84615386,0.61538464,0.53846157,0.6923077,0.53846157,0.6923077,0.61538464,0.61538464,0.84615386,0.61538464,0.7692308,0.6923077,0.7692308,0.61538464,0.6923077,0.7692308,0.84615386,0.9230769,0.6923077,0.7692308,0.9230769,0.84615386,0.9230769,1.0,0.9230769,0.84615386,0.9230769,1.0,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,0.84615386,0.9230769,1.0,0.84615386,0.9230769,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.23076923,0.15384616,0.15384616,0.30769232,0.23076923,0.23076923,0.30769232,0.30769232,0.3846154,0.3846154,0.53846157,0.53846157,0.46153846,0.30769232,,0.3846154,0.61538464,0.61538464,0.46153846,0.7692308,0.7692308,0.61538464,0.61538464,0.61538464,0.61538464,0.61538464,0.7692308,0.61538464,0.61538464,0.6923077,0.7692308,0.6923077,0.7692308,0.7692308,0.84615386,0.9230769,0.6923077,0.7692308,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.7692308,0.9230769,1.0,0.9230769,0.7692308,1.0,1.0,1.0,1.0,0.84615386,1.0,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.07692308,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.07692308,0.07692308,0.23076923,0.23076923,0.15384616,0.23076923,0.30769232,0.23076923,0.15384616,0.30769232,0.23076923,0.53846157,0.46153846,0.61538464,0.61538464,0.53846157,0.46153846,0.61538464,,0.61538464,0.61538464,0.6923077,0.7692308,0.6923077,0.6923077,0.46153846,0.61538464,0.61538464,0.61538464,0.6923077,0.61538464,0.61538464,0.61538464,0.7692308,0.61538464,0.84615386,0.84615386,0.7692308,0.6923077,0.53846157,0.7692308,0.6923077,0.84615386,0.7692308,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.84615386,0.9230769,1.0,0.9230769,0.9230769,1.0,0.84615386,0.84615386,0.9230769,0.84615386,1.0,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.07692308,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.07692308,0.15384616,0.30769232,0.23076923,0.15384616,0.15384616,0.30769232,0.30769232,0.3846154,0.30769232,0.3846154,0.3846154,0.3846154,,0.3846154,0.46153846,0.6923077,0.6923077,0.53846157,0.61538464,0.53846157,0.6923077,0.61538464,0.6923077,0.53846157,0.53846157,0.7692308,0.9230769,0.61538464,0.61538464,0.84615386,0.7692308,0.7692308,0.6923077,0.7692308,0.84615386,0.7692308,0.84615386,0.9230769,1.0,0.9230769,1.0,0.84615386,1.0,0.9230769,0.84615386,0.84615386,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.3846154,0.46153846,0.30769232,0.61538464,0.53846157,0.15384616,0.46153846,0.3846154,0.3846154,0.61538464,,0.46153846,0.61538464,0.61538464,0.3846154,0.53846157,0.7692308,0.46153846,0.7692308,0.6923077,0.61538464,0.61538464,0.7692308,1.0,0.61538464,0.61538464,0.84615386,0.7692308,0.84615386,0.6923077,0.7692308,0.84615386,0.9230769,0.7692308,0.9230769,0.84615386,0.7692308,0.9230769,0.7692308,0.84615386,0.7692308,0.7692308,0.9230769,0.9230769,1.0,0.84615386,1.0,1.0,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.15384616,0.07692308,0.15384616,0.15384616,0.30769232,0.3846154,0.23076923,0.46153846,0.46153846,0.46153846,0.15384616,0.53846157,0.30769232,0.53846157,0.53846157,,0.61538464,0.53846157,0.7692308,0.53846157,0.61538464,0.53846157,0.46153846,0.7692308,0.53846157,0.61538464,0.53846157,0.6923077,0.6923077,0.6923077,0.84615386,0.84615386,0.7692308,0.6923077,0.7692308,0.7692308,0.7692308,0.9230769,0.84615386,0.9230769,0.7692308,0.9230769,0.9230769,1.0,0.7692308,0.9230769,0.9230769,1.0,1.0,0.9230769,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.30769232,0.30769232,0.3846154,0.23076923,0.23076923,0.30769232,0.3846154,0.3846154,,0.53846157,0.53846157,0.61538464,0.46153846,0.53846157,0.53846157,0.6923077,0.61538464,0.6923077,0.53846157,0.6923077,0.61538464,0.61538464,0.7692308,0.84615386,0.84615386,0.6923077,0.84615386,0.7692308,0.84615386,0.6923077,0.9230769,0.9230769,0.84615386,0.9230769,0.9230769,1.0,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.0,0.07692308,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.30769232,0.30769232,0.46153846,0.23076923,0.30769232,0.30769232,0.3846154,0.46153846,0.46153846,,0.46153846,0.53846157,0.46153846,0.53846157,0.46153846,0.6923077,0.61538464,0.6923077,0.53846157,0.7692308,0.61538464,0.53846157,0.7692308,0.84615386,0.84615386,0.6923077,0.84615386,0.84615386,0.84615386,0.7692308,1.0,1.0,0.84615386,0.9230769,0.9230769,1.0,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0,0.84615386,1.0,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.07692308,0.0,0.15384616,0.15384616,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.46153846,0.3846154,0.53846157,0.30769232,0.3846154,0.30769232,0.46153846,0.61538464,0.23076923,0.46153846,0.53846157,,0.61538464,0.61538464,0.46153846,0.61538464,0.61538464,0.53846157,0.46153846,0.61538464,0.6923077,0.6923077,0.53846157,0.6923077,0.7692308,0.7692308,0.61538464,0.6923077,0.6923077,0.7692308,0.7692308,0.84615386,0.6923077,0.6923077,0.6923077,0.9230769,0.9230769,0.6923077,0.9230769,0.84615386,0.9230769,0.9230769,0.84615386,0.9230769,1.0,0.84615386,1.0,0.9230769,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.07692308,0.15384616,0.07692308,0.15384616,0.23076923,0.23076923,0.30769232,0.23076923,0.3846154,0.30769232,0.30769232,0.3846154,0.3846154,0.3846154,0.3846154,0.30769232,0.3846154,0.3846154,0.3846154,0.3846154,0.3846154,0.3846154,0.46153846,0.3846154,0.53846157,0.3846154,0.46153846,0.46153846,0.3846154,0.46153846,0.3846154,,0.46153846,0.3846154,0.53846157,0.46153846,0.7307692,0.53846157,0.53846157,0.46153846,0.7692308,0.7692308,0.61538464,0.53846157,0.46153846,0.8076923,0.61538464,0.46153846,0.6923077,0.46153846,0.53846157,0.53846157,0.7692308,0.53846157,0.61538464,0.53846157,0.7692308,0.6923077,0.84615386,0.84615386,0.6923077,0.84615386,0.84615386,0.8076923,0.84615386,0.84615386,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.0,0.15384616,0.15384616,0.23076923,0.23076923,0.07692308,0.3846154,0.07692308,0.23076923,0.23076923,0.30769232,0.3846154,0.3846154,0.46153846,0.23076923,0.3846154,0.53846157,0.53846157,0.3846154,0.53846157,,0.53846157,0.46153846,0.61538464,0.61538464,0.46153846,0.6923077,0.6923077,0.61538464,0.53846157,0.6923077,0.6923077,0.6923077,0.6923077,0.9230769,0.84615386,0.6923077,0.7692308,0.84615386,1.0,0.84615386,0.84615386,0.7692308,1.0,0.84615386,0.84615386,0.9230769,0.9230769,0.7692308,0.9230769,0.9230769,0.9230769,0.84615386,1.0,0.9230769,1.0,1.0,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.15384616,0.15384616,0.07692308,0.07692308,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.23076923,0.30769232,0.0,0.30769232,0.23076923,0.30769232,0.30769232,0.46153846,0.3846154,0.3846154,0.3846154,0.30769232,0.53846157,0.46153846,0.46153846,0.46153846,0.53846157,0.61538464,0.46153846,,0.46153846,0.53846157,0.53846157,0.53846157,0.53846157,0.61538464,0.61538464,0.61538464,0.6923077,0.6923077,0.53846157,0.6923077,0.61538464,0.53846157,0.7692308,0.6923077,0.7692308,0.7692308,0.6923077,0.84615386,0.6923077,0.7692308,0.7692308,0.7692308,0.6923077,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,0.7692308,0.9230769,0.7692308,1.0,1.0,0.9230769,0.84615386,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.0,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.3846154,0.30769232,0.30769232,0.30769232,0.3846154,0.3846154,0.3846154,0.3846154,0.23076923,0.53846157,0.46153846,0.53846157,0.3846154,0.46153846,0.53846157,0.53846157,,0.53846157,0.53846157,0.46153846,0.6923077,0.6923077,0.53846157,0.53846157,0.7692308,0.61538464,0.6923077,0.61538464,0.9230769,0.6923077,0.6923077,0.6923077,0.6923077,0.9230769,0.84615386,0.84615386,0.7692308,0.9230769,0.84615386,0.7692308,0.6923077,0.9230769,0.7692308,0.9230769,0.84615386,0.9230769,0.7692308,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.84615386,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.15384616,0.30769232,0.23076923,0.23076923,0.23076923,0.30769232,0.15384616,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.30769232,0.30769232,0.3846154,0.53846157,0.3846154,0.46153846,0.46153846,,0.61538464,0.61538464,0.53846157,0.61538464,0.61538464,0.61538464,0.6923077,0.61538464,0.53846157,0.6923077,0.6923077,0.53846157,0.61538464,0.84615386,0.84615386,0.7692308,0.6923077,0.7692308,0.84615386,0.7692308,0.7692308,0.84615386,0.7692308,0.9230769,1.0,0.84615386,0.9230769,1.0,0.84615386,0.9230769,0.84615386,1.0,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+Mitra (default),0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.15384616,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.30769232,0.30769232,0.23076923,0.30769232,0.30769232,0.3846154,0.3846154,0.3846154,0.3846154,0.46153846,0.3846154,0.46153846,0.3846154,0.3846154,0.46153846,0.26923078,0.3846154,0.46153846,0.46153846,0.3846154,,0.46153846,0.53846157,0.53846157,0.42307693,0.53846157,0.61538464,0.61538464,0.46153846,0.53846157,0.61538464,0.46153846,0.6923077,0.46153846,0.53846157,0.53846157,0.84615386,0.53846157,0.53846157,0.53846157,0.84615386,0.61538464,0.7692308,0.84615386,0.6923077,0.84615386,0.84615386,0.7692308,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.15384616,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.23076923,0.23076923,0.23076923,0.30769232,0.23076923,0.30769232,0.30769232,0.30769232,0.30769232,0.30769232,0.30769232,0.23076923,0.3846154,0.3846154,0.46153846,0.3846154,0.3846154,0.30769232,0.30769232,0.53846157,0.46153846,0.53846157,0.46153846,0.53846157,0.3846154,0.53846157,,0.61538464,0.46153846,0.61538464,0.53846157,0.61538464,0.61538464,0.46153846,0.61538464,0.61538464,0.46153846,0.53846157,0.53846157,0.53846157,0.53846157,0.6923077,0.61538464,0.61538464,0.61538464,0.6923077,0.61538464,0.6923077,0.7692308,0.7692308,0.7692308,0.7692308,0.7692308,0.84615386,0.84615386,0.84615386,0.7692308,0.7692308,0.9230769,0.9230769,0.9230769,1.0,0.9230769,0.9230769,1.0
+ChimeraBoost (tuned),0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.0,0.0,0.15384616,0.15384616,0.07692308,0.15384616,0.0,0.23076923,0.07692308,0.23076923,0.15384616,0.30769232,0.30769232,0.3846154,0.23076923,0.23076923,0.46153846,0.46153846,0.46153846,0.3846154,0.46153846,0.30769232,0.46153846,0.30769232,0.46153846,0.46153846,0.3846154,,0.61538464,0.53846157,0.46153846,0.7692308,0.61538464,0.61538464,0.61538464,0.84615386,0.61538464,0.53846157,0.6923077,0.6923077,0.9230769,0.61538464,0.84615386,0.7692308,0.84615386,0.6923077,0.7692308,0.6923077,0.7692308,0.7692308,0.9230769,0.84615386,0.84615386,0.7692308,0.9230769,0.7692308,0.9230769,0.9230769,0.9230769,0.84615386,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.15384616,0.0,0.23076923,0.23076923,0.23076923,0.07692308,0.0,0.30769232,0.30769232,0.23076923,0.30769232,0.53846157,0.30769232,0.3846154,0.30769232,0.3846154,0.46153846,0.53846157,0.3846154,,0.61538464,0.61538464,0.6923077,0.53846157,0.6923077,0.6923077,0.61538464,0.6923077,0.6923077,0.61538464,0.84615386,0.84615386,0.6923077,0.9230769,0.6923077,0.84615386,0.6923077,0.6923077,0.7692308,0.84615386,0.7692308,0.7692308,0.9230769,0.9230769,0.84615386,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (tuned),0.0,0.07692308,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.15384616,0.23076923,0.23076923,0.30769232,0.23076923,0.15384616,0.23076923,0.30769232,0.3846154,0.3846154,0.3846154,0.3846154,0.30769232,0.3846154,0.30769232,0.3846154,0.3846154,0.3846154,0.30769232,0.3846154,0.3846154,0.30769232,0.23076923,0.3846154,0.3846154,0.46153846,0.3846154,0.5769231,0.3846154,0.46153846,0.3846154,,0.61538464,0.3846154,0.53846157,0.46153846,0.5769231,0.46153846,0.46153846,0.61538464,0.46153846,0.3846154,0.3846154,0.7692308,0.46153846,0.46153846,0.46153846,0.7692308,0.61538464,0.7692308,0.7692308,0.6923077,0.7692308,0.7692308,0.7307692,0.84615386,0.7692308,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.30769232,0.23076923,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.15384616,0.3846154,0.3846154,0.30769232,0.3846154,0.46153846,0.46153846,0.23076923,0.46153846,0.3846154,0.46153846,0.3846154,0.46153846,0.46153846,0.53846157,0.3846154,0.3846154,,0.53846157,0.46153846,0.3846154,0.3846154,0.6923077,0.46153846,0.46153846,0.46153846,0.61538464,0.61538464,0.61538464,0.6923077,0.61538464,0.53846157,0.7692308,0.61538464,0.6923077,0.7692308,0.7692308,0.7692308,0.7692308,0.6923077,0.7692308,0.7692308,0.7692308,0.84615386,0.84615386,0.84615386,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.30769232,0.3846154,0.30769232,0.30769232,0.23076923,0.30769232,0.3846154,0.3846154,0.23076923,0.30769232,0.61538464,0.46153846,,0.46153846,0.61538464,0.61538464,0.53846157,0.61538464,0.46153846,0.61538464,0.61538464,0.53846157,0.6923077,0.7692308,0.7692308,0.6923077,0.6923077,0.7692308,0.7692308,0.84615386,0.7692308,0.84615386,0.7692308,0.84615386,0.7692308,0.9230769,0.7692308,1.0,0.9230769,0.84615386,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.07692308,0.0,0.0,0.15384616,0.07692308,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.46153846,0.30769232,0.30769232,0.3846154,0.3846154,0.3846154,0.3846154,0.3846154,0.46153846,0.46153846,0.53846157,0.53846157,,0.53846157,0.61538464,0.46153846,0.61538464,0.7692308,0.61538464,0.6923077,0.53846157,0.6923077,0.61538464,0.61538464,0.53846157,0.6923077,0.6923077,0.61538464,0.84615386,1.0,0.7692308,0.9230769,0.84615386,0.84615386,0.9230769,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.23076923,0.15384616,0.15384616,0.15384616,0.07692308,0.07692308,0.30769232,0.23076923,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.53846157,0.30769232,0.46153846,0.30769232,0.46153846,0.53846157,0.53846157,0.3846154,0.30769232,0.53846157,0.61538464,0.3846154,0.46153846,,0.53846157,0.6923077,0.7692308,0.46153846,0.53846157,0.46153846,0.61538464,0.61538464,0.61538464,0.84615386,0.7692308,0.61538464,0.7692308,1.0,0.7692308,0.53846157,0.6923077,0.6923077,0.9230769,0.84615386,0.7692308,0.84615386,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (default),0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.15384616,0.23076923,0.23076923,0.30769232,0.07692308,0.15384616,0.23076923,0.30769232,0.30769232,0.30769232,0.30769232,0.30769232,0.30769232,0.30769232,0.46153846,0.30769232,0.30769232,0.30769232,0.30769232,0.30769232,0.3846154,0.1923077,0.30769232,0.30769232,0.3846154,0.30769232,0.46153846,0.3846154,0.3846154,0.30769232,0.42307693,0.61538464,0.3846154,0.3846154,0.46153846,,0.3846154,0.46153846,0.46153846,0.46153846,0.3846154,0.3846154,0.6923077,0.3846154,0.46153846,0.3846154,0.6923077,0.53846157,0.7692308,0.6923077,0.53846157,0.6923077,0.6923077,0.6923077,0.7692308,0.6923077,0.7692308,0.7692308,0.84615386,0.9230769,0.7692308,0.9230769,0.9230769,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.15384616,0.07692308,0.0,0.15384616,0.15384616,0.07692308,0.15384616,0.0,0.23076923,0.0,0.0,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.23076923,0.15384616,0.15384616,0.30769232,0.3846154,0.07692308,0.3846154,0.07692308,0.30769232,0.3846154,0.3846154,0.15384616,0.3846154,0.53846157,0.30769232,0.46153846,0.53846157,0.30769232,0.61538464,,0.46153846,0.53846157,0.61538464,0.61538464,0.9230769,0.6923077,0.6923077,0.61538464,0.84615386,0.7692308,0.6923077,0.6923077,0.7692308,0.61538464,0.7692308,0.6923077,0.84615386,0.7692308,0.9230769,0.7692308,0.84615386,0.9230769,0.84615386,0.84615386,1.0,0.9230769,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.15384616,0.15384616,0.07692308,0.07692308,0.30769232,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.30769232,0.53846157,0.15384616,0.46153846,0.30769232,0.46153846,0.53846157,0.53846157,0.3846154,0.30769232,0.53846157,0.53846157,0.3846154,0.3846154,0.23076923,0.53846157,0.53846157,,0.46153846,0.53846157,0.3846154,0.61538464,0.61538464,0.53846157,0.61538464,0.7692308,0.61538464,0.6923077,0.84615386,0.7692308,0.46153846,0.6923077,0.6923077,0.9230769,0.84615386,0.7692308,0.84615386,0.84615386,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.07692308,0.0,0.0,0.07692308,0.07692308,0.15384616,0.15384616,0.0,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.15384616,0.23076923,0.07692308,0.23076923,0.15384616,0.15384616,0.23076923,0.30769232,0.30769232,0.23076923,0.30769232,0.3846154,0.30769232,0.46153846,0.46153846,0.30769232,0.3846154,0.53846157,0.53846157,0.23076923,0.53846157,0.53846157,0.46153846,0.53846157,,0.61538464,0.53846157,0.53846157,0.61538464,0.53846157,0.61538464,0.53846157,0.61538464,0.61538464,0.61538464,0.6923077,1.0,0.7692308,0.84615386,0.61538464,0.84615386,0.84615386,0.84615386,0.84615386,0.7692308,0.84615386,0.9230769,0.9230769,0.9230769,0.9230769,0.9230769,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.07692308,0.07692308,0.23076923,0.15384616,0.23076923,0.07692308,0.30769232,0.23076923,0.23076923,0.53846157,0.23076923,0.30769232,0.30769232,0.15384616,0.53846157,0.46153846,0.30769232,0.3846154,0.53846157,0.53846157,0.3846154,0.3846154,0.46153846,0.53846157,0.3846154,0.46153846,0.3846154,,0.3846154,0.30769232,0.46153846,0.3846154,0.61538464,0.46153846,0.46153846,0.6923077,0.61538464,0.53846157,0.61538464,0.6923077,0.61538464,0.84615386,0.7692308,0.6923077,0.84615386,0.7692308,0.9230769,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.0,0.0,0.0,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.07692308,0.07692308,0.07692308,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.07692308,0.0,0.15384616,0.46153846,0.15384616,0.23076923,0.30769232,0.15384616,0.46153846,0.46153846,0.30769232,0.15384616,0.61538464,0.3846154,0.3846154,0.30769232,0.53846157,0.61538464,0.3846154,0.61538464,0.46153846,0.61538464,,0.61538464,0.61538464,0.6923077,0.53846157,0.6923077,0.61538464,0.61538464,0.7692308,0.7692308,0.7692308,0.6923077,0.7692308,0.9230769,0.84615386,0.84615386,0.84615386,0.9230769,1.0,0.9230769,0.9230769,1.0,0.9230769,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.15384616,0.0,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.0,0.07692308,0.0,0.0,0.15384616,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.07692308,0.07692308,0.0,0.30769232,0.46153846,0.0,0.23076923,0.07692308,0.23076923,0.46153846,0.46153846,0.07692308,0.15384616,0.61538464,0.3846154,0.46153846,0.46153846,0.3846154,0.61538464,0.07692308,0.3846154,0.46153846,0.6923077,0.3846154,,0.61538464,0.53846157,0.53846157,0.8076923,0.61538464,0.61538464,0.6923077,0.7692308,0.6923077,0.6923077,0.7692308,0.7692308,0.7692308,1.0,0.7692308,0.84615386,0.9230769,0.7692308,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.15384616,0.07692308,0.0,0.0,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.15384616,0.23076923,0.15384616,0.07692308,0.23076923,0.23076923,0.15384616,0.15384616,0.30769232,0.23076923,0.15384616,0.30769232,0.15384616,0.30769232,0.15384616,0.30769232,0.3846154,0.30769232,0.23076923,0.3846154,0.30769232,0.30769232,0.3846154,0.30769232,0.30769232,0.3846154,0.3846154,0.53846157,0.3846154,0.3846154,,0.46153846,0.3846154,0.53846157,0.84615386,0.3846154,0.3846154,0.61538464,0.7692308,1.0,0.6923077,0.53846157,0.84615386,0.7692308,0.9230769,0.84615386,0.7692308,0.84615386,0.9230769,0.84615386,0.84615386,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.07692308,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.0,0.15384616,0.0,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.0,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.30769232,0.46153846,0.15384616,0.15384616,0.15384616,0.23076923,0.46153846,0.3846154,0.15384616,0.07692308,0.53846157,0.30769232,0.23076923,0.3846154,0.3846154,0.61538464,0.30769232,0.46153846,0.46153846,0.61538464,0.30769232,0.46153846,0.53846157,,0.53846157,0.61538464,0.61538464,0.6923077,0.6923077,0.61538464,0.7692308,0.7692308,0.6923077,0.7692308,0.84615386,0.9230769,0.84615386,0.7692308,0.9230769,0.84615386,0.84615386,1.0,1.0,0.9230769,0.9230769,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.15384616,0.0,0.07692308,0.15384616,0.0,0.0,0.15384616,0.15384616,0.23076923,0.07692308,0.07692308,0.07692308,0.07692308,0.3846154,0.23076923,0.30769232,0.23076923,0.15384616,0.46153846,0.3846154,0.23076923,0.30769232,0.53846157,0.3846154,0.23076923,0.3846154,0.15384616,0.53846157,0.3846154,0.3846154,0.3846154,0.3846154,0.46153846,0.46153846,0.61538464,0.46153846,,0.53846157,0.61538464,0.6923077,0.6923077,0.53846157,0.53846157,0.61538464,0.53846157,0.9230769,0.7692308,0.61538464,0.7692308,0.84615386,1.0,1.0,0.9230769,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.0,0.0,0.0,0.0,0.15384616,0.07692308,0.07692308,0.15384616,0.0,0.15384616,0.0,0.0,0.0,0.07692308,0.46153846,0.0,0.23076923,0.07692308,0.23076923,0.46153846,0.3846154,0.15384616,0.15384616,0.53846157,0.46153846,0.30769232,0.46153846,0.23076923,0.61538464,0.15384616,0.23076923,0.46153846,0.53846157,0.30769232,0.1923077,0.46153846,0.3846154,0.46153846,,0.61538464,0.61538464,0.61538464,0.7692308,0.6923077,0.7692308,0.7692308,0.7692308,0.7692308,0.9230769,0.7692308,0.84615386,0.9230769,0.7692308,0.84615386,0.9230769,0.9230769,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.15384616,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.23076923,0.07692308,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.23076923,0.15384616,0.07692308,0.23076923,0.23076923,0.07692308,0.07692308,0.30769232,0.23076923,0.15384616,0.23076923,0.15384616,0.23076923,0.15384616,0.30769232,0.30769232,0.30769232,0.23076923,0.23076923,0.30769232,0.30769232,0.3846154,0.30769232,0.23076923,0.3846154,0.3846154,0.53846157,0.3846154,0.3846154,0.15384616,0.3846154,0.3846154,0.3846154,,0.3846154,0.3846154,0.46153846,0.61538464,0.84615386,0.61538464,0.53846157,0.7692308,0.6923077,0.7692308,0.84615386,0.6923077,0.7692308,0.9230769,0.84615386,0.84615386,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.0,0.0,0.15384616,0.0,0.0,0.15384616,0.15384616,0.23076923,0.07692308,0.07692308,0.07692308,0.07692308,0.30769232,0.15384616,0.23076923,0.23076923,0.15384616,0.3846154,0.3846154,0.23076923,0.30769232,0.3846154,0.3846154,0.23076923,0.30769232,0.23076923,0.46153846,0.30769232,0.30769232,0.3846154,0.30769232,0.3846154,0.3846154,0.61538464,0.30769232,0.30769232,0.3846154,0.61538464,,0.53846157,0.53846157,0.46153846,0.61538464,0.53846157,0.7307692,0.7692308,0.61538464,0.7692308,0.7692308,0.9230769,1.0,0.84615386,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.23076923,0.0,0.07692308,0.07692308,0.0,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.30769232,0.30769232,0.23076923,0.23076923,0.30769232,0.30769232,0.23076923,0.23076923,0.30769232,0.23076923,0.3846154,0.0,0.23076923,0.30769232,0.15384616,0.3846154,0.3846154,0.23076923,0.30769232,0.61538464,0.30769232,0.30769232,0.3846154,0.61538464,0.46153846,,0.46153846,0.3846154,0.61538464,0.46153846,0.61538464,0.6923077,0.61538464,0.6923077,0.6923077,0.9230769,0.9230769,0.7692308,0.9230769,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.23076923,0.23076923,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.30769232,0.23076923,0.23076923,0.30769232,0.46153846,0.23076923,0.23076923,0.3846154,0.3846154,0.46153846,0.23076923,0.53846157,0.46153846,0.53846157,,0.53846157,0.61538464,0.61538464,0.61538464,0.7692308,0.6923077,0.7692308,1.0,0.7692308,0.6923077,0.84615386,0.7692308,0.9230769,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.07692308,0.30769232,0.23076923,0.07692308,0.23076923,0.0,0.30769232,0.23076923,0.23076923,0.23076923,0.30769232,0.23076923,0.23076923,0.0,0.46153846,0.46153846,0.3846154,0.53846157,0.0,0.3846154,0.23076923,0.30769232,0.23076923,0.23076923,0.46153846,0.30769232,0.3846154,0.53846157,0.61538464,0.46153846,,0.3846154,0.46153846,0.61538464,0.6923077,0.3846154,0.6923077,0.46153846,0.6923077,0.6923077,0.84615386,0.84615386,0.84615386,0.9230769,0.9230769,1.0
+EBM (default),0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.0,0.0,0.0,0.0,0.15384616,0.15384616,0.15384616,0.07692308,0.0,0.15384616,0.07692308,0.0,0.0,0.15384616,0.15384616,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.23076923,0.07692308,0.23076923,0.23076923,0.23076923,0.15384616,0.23076923,0.30769232,0.30769232,0.23076923,0.30769232,0.23076923,0.30769232,0.30769232,0.30769232,0.0,0.23076923,0.3846154,0.23076923,0.15384616,0.3846154,0.3846154,0.3846154,0.61538464,,0.53846157,0.46153846,0.6923077,0.61538464,0.6923077,0.61538464,0.6923077,0.6923077,0.9230769,0.84615386,0.84615386,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.07692308,0.15384616,0.23076923,0.15384616,0.07692308,0.23076923,0.23076923,0.23076923,0.07692308,0.30769232,0.30769232,0.30769232,0.30769232,0.15384616,0.3846154,0.23076923,0.23076923,0.30769232,0.30769232,0.46153846,0.23076923,0.3846154,0.46153846,0.53846157,0.3846154,0.53846157,0.46153846,,0.53846157,0.7692308,0.61538464,0.7692308,0.6923077,0.6923077,0.61538464,0.84615386,0.7692308,0.9230769,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.15384616,0.0,0.0,0.0,0.0,0.0,0.0,0.1923077,0.07692308,0.15384616,0.07692308,0.0,0.23076923,0.23076923,0.15384616,0.07692308,0.26923078,0.30769232,0.15384616,0.15384616,0.07692308,0.30769232,0.15384616,0.07692308,0.3846154,0.15384616,0.07692308,0.23076923,0.46153846,0.23076923,0.07692308,0.23076923,0.46153846,0.26923078,0.3846154,0.3846154,0.3846154,0.53846157,0.46153846,,0.7692308,0.61538464,0.7692308,0.6923077,0.7692308,0.84615386,0.84615386,1.0,1.0,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.07692308,0.0,0.0,0.0,0.0,0.15384616,0.0,0.15384616,0.15384616,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.30769232,0.23076923,0.30769232,0.30769232,0.23076923,0.23076923,,0.23076923,0.84615386,0.3846154,0.3846154,0.46153846,1.0,0.61538464,0.61538464,0.84615386,0.84615386,0.9230769
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.0,0.0,0.0,0.0,0.15384616,0.0,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.07692308,0.23076923,0.23076923,0.07692308,0.07692308,0.23076923,0.30769232,0.07692308,0.23076923,0.15384616,0.30769232,0.15384616,0.0,0.23076923,0.07692308,0.3846154,0.07692308,0.30769232,0.3846154,0.3846154,0.30769232,0.61538464,0.3846154,0.3846154,0.3846154,0.7692308,,0.7692308,0.46153846,0.61538464,0.61538464,0.84615386,0.6923077,0.84615386,0.9230769,0.9230769,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.15384616,0.07692308,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.15384616,0.07692308,0.07692308,0.07692308,0.23076923,0.15384616,0.15384616,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.23076923,0.23076923,0.15384616,0.15384616,0.15384616,0.15384616,0.23076923,0.07692308,0.15384616,0.23076923,0.23076923,0.23076923,0.23076923,0.30769232,0.23076923,0.30769232,0.30769232,0.23076923,0.23076923,0.15384616,0.23076923,,0.3846154,0.3846154,0.46153846,1.0,0.61538464,0.61538464,0.84615386,0.84615386,0.9230769
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.07692308,0.23076923,0.07692308,0.07692308,0.07692308,0.15384616,0.0,0.07692308,0.07692308,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.07692308,0.15384616,0.15384616,0.23076923,0.15384616,0.15384616,0.15384616,0.23076923,0.30769232,0.0,0.53846157,0.3846154,0.30769232,0.30769232,0.61538464,0.53846157,0.61538464,,0.46153846,0.46153846,0.61538464,0.61538464,0.7692308,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.07692308,0.23076923,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.07692308,0.23076923,0.07692308,0.0,0.07692308,0.23076923,0.07692308,0.0,0.07692308,0.30769232,0.07692308,0.07692308,0.23076923,0.30769232,0.30769232,0.30769232,0.23076923,0.61538464,0.3846154,0.61538464,0.53846157,,0.6923077,0.7692308,0.7692308,0.84615386,0.9230769,0.9230769,0.9230769
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.07692308,0.0,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.15384616,0.07692308,0.0,0.07692308,0.15384616,0.0,0.15384616,0.0,0.07692308,0.23076923,0.15384616,0.15384616,0.0,0.23076923,0.23076923,0.0,0.07692308,0.30769232,0.30769232,0.30769232,0.3846154,0.15384616,0.53846157,0.3846154,0.53846157,0.53846157,0.30769232,,0.53846157,1.0,1.0,0.9230769,0.9230769,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.07692308,0.0,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.07692308,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.0,0.07692308,0.15384616,0.07692308,0.07692308,0.0,0.0,0.07692308,0.07692308,0.23076923,0.15384616,0.07692308,0.07692308,0.07692308,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.07692308,0.15384616,0.23076923,0.15384616,0.15384616,0.07692308,0.15384616,0.15384616,0.0,0.15384616,0.0,0.3846154,0.23076923,0.46153846,,0.53846157,0.53846157,0.84615386,0.84615386,0.84615386
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.15384616,0.0,0.0,0.07692308,0.15384616,0.0,0.07692308,0.23076923,0.15384616,0.15384616,0.23076923,0.0,0.3846154,0.30769232,0.3846154,0.3846154,0.23076923,0.0,0.46153846,,0.84615386,0.9230769,0.9230769,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.07692308,0.0,0.07692308,0.07692308,0.0,0.07692308,0.0,0.07692308,0.07692308,0.15384616,0.0,0.0,0.07692308,0.15384616,0.0,0.0,0.07692308,0.15384616,0.15384616,0.07692308,0.0,0.3846154,0.15384616,0.3846154,0.23076923,0.15384616,0.0,0.46153846,0.15384616,,0.6923077,0.6923077,0.6923077
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.15384616,0.07692308,0.15384616,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.30769232,,0.8076923,0.84615386
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.0,0.0,0.15384616,0.07692308,0.15384616,0.0,0.07692308,0.07692308,0.15384616,0.07692308,0.30769232,0.1923077,,0.7692308
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.07692308,0.0,0.07692308,0.0,0.07692308,0.0,0.15384616,0.0,0.30769232,0.15384616,0.23076923,
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
index 9af384e46057e8c54eb64e4fedbd9bd0390c6a5c..4075a17d839201ffd1c27bd3da41b17d16a7be49 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Regression | Medium</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..2811461d3de6a6b72fab6b6a6ec90b5f8a91a321
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_regression/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2478.0,"score":0.907,"rank":1.6,"hrank":1.43,"imp":0.476,"train_time":25.12,"predict_time":11.643,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1402\/-145","imputed":false},{"position":1,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1994.0,"score":0.63,"rank":7.2,"hrank":5.92,"imp":3.186,"train_time":739.1,"predict_time":11.65,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+776\/-70","imputed":false},{"position":2,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1971.0,"score":0.637,"rank":7.8,"hrank":6.01,"imp":3.282,"train_time":739.1,"predict_time":0.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+827\/-105","imputed":false},{"position":3,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1949.0,"score":0.74,"rank":8.4,"hrank":2.46,"imp":1.772,"train_time":0.87,"predict_time":0.204,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1538\/-285","imputed":false},{"position":4,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1902.0,"score":0.65,"rank":9.8,"hrank":3.67,"imp":4.09,"train_time":374.99,"predict_time":3.566,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+624\/-16","imputed":false},{"position":5,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1866.0,"score":0.524,"rank":11.0,"hrank":9.83,"imp":4.264,"train_time":2552.31,"predict_time":4.522,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+607\/-39","imputed":false},{"position":6,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1860.0,"score":0.515,"rank":11.2,"hrank":9.2,"imp":4.214,"train_time":2.78,"predict_time":0.702,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+658\/-45","imputed":false},{"position":7,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1822.0,"score":0.481,"rank":12.6,"hrank":8.19,"imp":5.17,"train_time":474.95,"predict_time":11.837,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+468\/-0","imputed":false},{"position":8,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1806.0,"score":0.624,"rank":13.2,"hrank":4.36,"imp":3.512,"train_time":0.47,"predict_time":0.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+1206\/-342","imputed":false},{"position":9,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1801.0,"score":0.468,"rank":13.4,"hrank":12.22,"imp":4.842,"train_time":2068.17,"predict_time":1.355,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+588\/-75","imputed":false},{"position":10,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1790.0,"score":0.528,"rank":13.8,"hrank":3.35,"imp":4.055,"train_time":4529.93,"predict_time":357.201,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+995\/-190","imputed":false},{"position":11,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1780.0,"score":0.468,"rank":14.2,"hrank":11.71,"imp":5.289,"train_time":4.86,"predict_time":1.881,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+581\/-118","imputed":false},{"position":12,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1742.0,"score":0.498,"rank":15.8,"hrank":7.4,"imp":4.355,"train_time":4529.93,"predict_time":39.54,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+911\/-180","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1723.0,"score":0.474,"rank":16.6,"hrank":11.87,"imp":4.298,"train_time":0.66,"predict_time":0.177,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+784\/-112","imputed":false},{"position":14,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1710.0,"score":0.462,"rank":17.2,"hrank":10.54,"imp":4.597,"train_time":1.59,"predict_time":0.506,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+787\/-246","imputed":false},{"position":15,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1635.0,"score":0.335,"rank":20.8,"hrank":16.55,"imp":5.935,"train_time":1.97,"predict_time":0.651,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+498\/-134","imputed":false},{"position":16,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1630.0,"score":0.337,"rank":21.0,"hrank":16.08,"imp":6.087,"train_time":2068.17,"predict_time":0.068,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+549\/-171","imputed":false},{"position":17,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1622.0,"score":0.251,"rank":21.4,"hrank":17.06,"imp":7.744,"train_time":223.05,"predict_time":9.706,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+344\/-106","imputed":false},{"position":18,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1610.0,"score":0.412,"rank":22.0,"hrank":11.6,"imp":5.141,"train_time":41.51,"predict_time":38.297,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+763\/-196","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1606.0,"score":0.227,"rank":22.2,"hrank":17.96,"imp":7.913,"train_time":223.05,"predict_time":0.793,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+376\/-112","imputed":false},{"position":20,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1602.0,"score":0.334,"rank":22.4,"hrank":18.29,"imp":6.016,"train_time":2552.31,"predict_time":0.201,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+550\/-218","imputed":false},{"position":21,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1594.0,"score":0.303,"rank":22.8,"hrank":20.52,"imp":6.408,"train_time":438.56,"predict_time":0.52,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+388\/-58","imputed":false},{"position":22,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1590.0,"score":0.176,"rank":23.0,"hrank":14.3,"imp":8.15,"train_time":917.62,"predict_time":0.422,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+321\/-79","imputed":false},{"position":23,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1557.0,"score":0.161,"rank":24.6,"hrank":12.93,"imp":8.303,"train_time":917.62,"predict_time":0.016,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+314\/-94","imputed":false},{"position":24,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1546.0,"score":0.28,"rank":25.2,"hrank":19.69,"imp":7.074,"train_time":2410.31,"predict_time":0.581,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+470\/-184","imputed":false},{"position":25,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1519.0,"score":0.234,"rank":26.6,"hrank":21.16,"imp":7.704,"train_time":2410.31,"predict_time":0.045,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+446\/-178","imputed":false},{"position":26,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1496.0,"score":0.177,"rank":27.8,"hrank":26.39,"imp":7.379,"train_time":438.56,"predict_time":0.035,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+300\/-75","imputed":false},{"position":27,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1486.0,"score":0.155,"rank":28.4,"hrank":21.55,"imp":8.948,"train_time":2.73,"predict_time":0.01,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+334\/-225","imputed":false},{"position":28,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1475.0,"score":0.191,"rank":29.0,"hrank":26.7,"imp":7.245,"train_time":6.87,"predict_time":0.098,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+391\/-136","imputed":false},{"position":29,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1442.0,"score":0.102,"rank":30.8,"hrank":29.57,"imp":8.739,"train_time":289.35,"predict_time":1.798,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+284\/-109","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1431.0,"score":0.097,"rank":31.4,"hrank":29.83,"imp":8.805,"train_time":289.35,"predict_time":0.36,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+312\/-140","imputed":false},{"position":31,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1397.0,"score":0.089,"rank":33.2,"hrank":32.6,"imp":8.165,"train_time":2150.19,"predict_time":65.934,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+202\/-41","imputed":false},{"position":32,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1393.0,"score":0.184,"rank":33.4,"hrank":19.49,"imp":7.945,"train_time":24.62,"predict_time":8.835,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+634\/-299","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1381.0,"score":0.122,"rank":34.0,"hrank":25.43,"imp":12.794,"train_time":125.2,"predict_time":0.276,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+414\/-474","imputed":false},{"position":34,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1347.0,"score":0.171,"rank":35.8,"hrank":27.75,"imp":9.123,"train_time":0.51,"predict_time":0.126,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+446\/-336","imputed":false},{"position":35,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1343.0,"score":0.108,"rank":36.0,"hrank":30.03,"imp":12.83,"train_time":125.2,"predict_time":0.041,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+384\/-461","imputed":false},{"position":36,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1316.0,"score":0.051,"rank":37.4,"hrank":36.66,"imp":9.289,"train_time":2150.19,"predict_time":18.752,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+183\/-127","imputed":false},{"position":37,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1309.0,"score":0.046,"rank":37.8,"hrank":36.05,"imp":9.733,"train_time":3.22,"predict_time":0.039,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+269\/-144","imputed":false},{"position":38,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1298.0,"score":0.059,"rank":38.4,"hrank":34.78,"imp":9.9,"train_time":0.57,"predict_time":0.073,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+238\/-247","imputed":false},{"position":39,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1283.0,"score":0.037,"rank":39.2,"hrank":37.76,"imp":9.965,"train_time":17.81,"predict_time":0.4,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+167\/-245","imputed":false},{"position":40,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1251.0,"score":0.013,"rank":41.0,"hrank":40.24,"imp":10.083,"train_time":0.66,"predict_time":0.213,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+212\/-129","imputed":false},{"position":41,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1247.0,"score":0.006,"rank":41.2,"hrank":39.9,"imp":10.7,"train_time":2.32,"predict_time":0.093,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+195\/-177","imputed":false},{"position":42,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1236.0,"score":0.068,"rank":41.8,"hrank":37.04,"imp":13.809,"train_time":0.74,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+309\/-451","imputed":false},{"position":43,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1228.0,"score":0.066,"rank":42.2,"hrank":39.81,"imp":9.994,"train_time":4108.24,"predict_time":0.479,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+192\/-211","imputed":false},{"position":44,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1193.0,"score":0.001,"rank":44.0,"hrank":43.23,"imp":10.73,"train_time":17.81,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-212","imputed":false},{"position":45,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1185.0,"score":0.026,"rank":44.4,"hrank":41.6,"imp":11.402,"train_time":6.54,"predict_time":0.284,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+296\/-295","imputed":false},{"position":46,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1181.0,"score":0.031,"rank":44.6,"hrank":42.4,"imp":11.081,"train_time":4108.24,"predict_time":0.037,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+196\/-277","imputed":false},{"position":47,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1177.0,"score":0.027,"rank":44.8,"hrank":42.56,"imp":11.917,"train_time":50.36,"predict_time":0.232,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+273\/-316","imputed":false},{"position":48,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1152.0,"score":0.026,"rank":46.0,"hrank":43.8,"imp":12.063,"train_time":50.36,"predict_time":0.019,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+272\/-335","imputed":false},{"position":49,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1118.0,"score":0.069,"rank":47.6,"hrank":37.0,"imp":12.244,"train_time":8.68,"predict_time":7.525,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+438\/-412","imputed":false},{"position":50,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1118.0,"score":0.063,"rank":47.6,"hrank":34.83,"imp":15.412,"train_time":1069.96,"predict_time":0.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+394\/-540","imputed":false},{"position":51,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1114.0,"score":0.0,"rank":47.8,"hrank":47.45,"imp":11.829,"train_time":79.96,"predict_time":0.571,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-240","imputed":false},{"position":52,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1108.0,"score":0.0,"rank":48.1,"hrank":47.52,"imp":11.989,"train_time":79.96,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-257","imputed":false},{"position":53,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1049.0,"score":0.0,"rank":50.8,"hrank":49.61,"imp":13.926,"train_time":1.53,"predict_time":0.008,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+168\/-355","imputed":false},{"position":54,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1026.0,"score":0.0,"rank":51.8,"hrank":50.36,"imp":15.231,"train_time":14.05,"predict_time":0.029,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+152\/-305","imputed":false},{"position":55,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1007.0,"score":0.036,"rank":52.6,"hrank":47.95,"imp":16.025,"train_time":1069.96,"predict_time":0.005,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+304\/-570","imputed":false},{"position":56,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":52.9,"hrank":52.59,"imp":13.109,"train_time":0.42,"predict_time":0.053,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-226","imputed":false},{"position":57,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":988.0,"score":0.048,"rank":53.4,"hrank":47.5,"imp":16.456,"train_time":2.37,"predict_time":0.006,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+336\/-644","imputed":false},{"position":58,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":988.0,"score":0.0,"rank":53.4,"hrank":51.29,"imp":13.952,"train_time":0.86,"predict_time":0.099,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+276\/-538","imputed":false},{"position":59,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":884.0,"score":0.0,"rank":57.2,"hrank":56.6,"imp":16.484,"train_time":93.51,"predict_time":19.996,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-630","imputed":false},{"position":60,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":860.0,"score":0.0,"rank":58.0,"hrank":57.41,"imp":16.012,"train_time":317.48,"predict_time":0.719,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-442","imputed":false},{"position":61,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":828.0,"score":0.0,"rank":59.0,"hrank":58.58,"imp":16.565,"train_time":317.48,"predict_time":0.127,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-470","imputed":false},{"position":62,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":772.0,"score":0.0,"rank":60.6,"hrank":60.22,"imp":20.22,"train_time":142.18,"predict_time":1.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-461","imputed":false},{"position":63,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":701.0,"score":0.0,"rank":62.4,"hrank":61.83,"imp":19.882,"train_time":1.33,"predict_time":0.126,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-726","imputed":false},{"position":64,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":683.0,"score":0.0,"rank":62.8,"hrank":62.55,"imp":23.8,"train_time":5782.12,"predict_time":8.567,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-799","imputed":false},{"position":65,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":646.0,"score":0.0,"rank":63.6,"hrank":63.34,"imp":23.989,"train_time":5782.12,"predict_time":2.139,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-861","imputed":false},{"position":66,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":617.0,"score":0.0,"rank":64.2,"hrank":64.01,"imp":22.596,"train_time":142.18,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+83\/-514","imputed":false},{"position":67,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":512.0,"score":0.0,"rank":66.0,"hrank":65.89,"imp":27.279,"train_time":113.4,"predict_time":1.615,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+132\/-900","imputed":false},{"position":68,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":402.0,"score":0.0,"rank":67.4,"hrank":67.22,"imp":29.723,"train_time":0.06,"predict_time":0.036,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-1194","imputed":false},{"position":69,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":65.0,"score":0.0,"rank":69.8,"hrank":69.8,"imp":40.798,"train_time":48.35,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+122\/-2248","imputed":false},{"position":70,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-192.0,"score":0.0,"rank":71.0,"hrank":70.99,"imp":41.14,"train_time":48.35,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+218\/-2864","imputed":false},{"position":71,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":-347.0,"score":0.0,"rank":71.6,"hrank":71.6,"imp":44.075,"train_time":0.13,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+181\/-2671","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/pareto_front_explorer.html
index e151f986b6cb776adc98d33626e7098c2bc7931e..fb83673b53e25ce8802a74d334e37bbfa9a574b1 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0103037461,"imp":8.9479539866,"elo":1485.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0161908394,"imp":8.3030672997,"elo":1557.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"imp":8.1503948491,"elo":1589.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"imp":13.8090022874,"elo":1235.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0408241222,"imp":12.8301298042,"elo":1342.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"imp":12.7943126543,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"imp":16.4556369343,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0051447125,"imp":16.0253582032,"elo":1007.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0853557077,"imp":15.4116177291,"elo":1118.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0992770116,"imp":13.9524159906,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0186999882,"imp":12.06324568,"elo":1152.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2319944846,"imp":11.9166741153,"elo":1176.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1256610016,"imp":19.8824164433,"elo":700.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1272848026,"imp":16.5650505415,"elo":828.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7187718544,"imp":16.0119645154,"elo":859.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2133657073,"imp":10.0826385426,"elo":1250.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7933065809,"imp":7.9131722928,"elo":1605.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":9.7062590878,"imp":7.7439646534,"elo":1622.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.8350020269,"imp":7.9445989236,"elo":1392.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.097838007,"imp":7.2446277873,"elo":1474.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2007087367,"imp":6.0155505108,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":4.5218047762,"imp":4.2641087237,"elo":1865.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.6513221504,"imp":5.9350713717,"elo":1634.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8805802898,"imp":5.2894589865,"elo":1780.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0078855342,"imp":13.9256196498,"elo":1049.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1934792264,"imp":10.7303129531,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4003930169,"imp":9.9648302652,"elo":1283.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"imp":13.1089370831,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"imp":11.9892701847,"elo":1107.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5707803637,"imp":11.8289448844,"elo":1114.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2839085103,"imp":11.4020603961,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0682382085,"imp":6.087337608,"elo":1630.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.3549179469,"imp":4.8421550057,"elo":1800.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5059595331,"imp":4.5971768753,"elo":1710.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6894524158,"imp":3.2815549654,"elo":1970.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"imp":3.185810595,"elo":1994.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.5250451841,"imp":12.2440158789,"elo":1118.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"imp":5.1411670708,"elo":1610.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.540415138,"imp":4.3545991568,"elo":1741.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":357.2008760846,"imp":4.0547168602,"elo":1790.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1767387784,"imp":4.2983424742,"elo":1723.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":11.6433178755,"imp":0.4756811693,"elo":2478.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1028500957,"imp":3.5123157709,"elo":1805.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.038732221,"imp":9.732882352,"elo":1309.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0448228661,"imp":7.7035348907,"elo":1518.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5807768467,"imp":7.0742201117,"elo":1545.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"imp":4.214387882,"elo":1860.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042135993,"imp":1.7719216017,"elo":1948.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6147142423,"imp":27.2788510175,"elo":511.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.139339258,"imp":23.9886884373,"elo":646.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"imp":23.7997636743,"elo":683.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1256335989,"imp":9.1227548353,"elo":1346.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0292963248,"imp":15.2314315415,"elo":1026.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0365087137,"imp":11.0812079304,"elo":1180.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4787581258,"imp":9.9939380427,"elo":1228.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.073497228,"imp":9.8998436121,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3603568941,"imp":8.8046523626,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.7976430646,"imp":8.7389814156,"elo":1441.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":19.9963063451,"imp":16.4844284378,"elo":884.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":18.7518939426,"imp":9.2887498658,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.9338217731,"imp":8.1653547355,"elo":1396.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.0932754819,"imp":10.6996583222,"elo":1246.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0348964576,"imp":7.3787958084,"elo":1496.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.5201417769,"imp":6.4083302989,"elo":1593.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0103037461,"x_train":2.7349656008,"imp":8.9479539866,"elo":1485.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0161908394,"x_train":917.6176937357,"imp":8.3030672997,"elo":1557.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"x_train":917.6176937357,"imp":8.1503948491,"elo":1589.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"x_train":0.7437448515,"imp":13.8090022874,"elo":1235.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0408241222,"x_train":125.1955265319,"imp":12.8301298042,"elo":1342.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":125.1955265319,"imp":12.7943126543,"elo":1381.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"x_train":2.3730743936,"imp":16.4556369343,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0051447125,"x_train":1069.963809484,"imp":16.0253582032,"elo":1007.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.0853557077,"x_train":1069.963809484,"imp":15.4116177291,"elo":1118.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0992770116,"x_train":0.8553953885,"imp":13.9524159906,"elo":987.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0186999882,"x_train":50.361853205,"imp":12.06324568,"elo":1152.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2319944846,"x_train":50.361853205,"imp":11.9166741153,"elo":1176.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1256610016,"x_train":1.3312352951,"imp":19.8824164433,"elo":700.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1272848026,"x_train":317.4820332824,"imp":16.5650505415,"elo":828.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7187718544,"x_train":317.4820332824,"imp":16.0119645154,"elo":859.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.2133657073,"x_train":0.6567304564,"imp":10.0826385426,"elo":1250.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.7933065809,"x_train":223.0498746731,"imp":7.9131722928,"elo":1605.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":9.7062590878,"x_train":223.0498746731,"imp":7.7439646534,"elo":1622.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":8.8350020269,"x_train":24.6223741935,"imp":7.9445989236,"elo":1392.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.097838007,"x_train":6.870566258,"imp":7.2446277873,"elo":1474.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2007087367,"x_train":2552.3091253937,"imp":6.0155505108,"elo":1601.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":4.5218047762,"x_train":2552.3091253937,"imp":4.2641087237,"elo":1865.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":0.6513221504,"x_train":1.9712126532,"imp":5.9350713717,"elo":1634.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8805802898,"x_train":4.8565548774,"imp":5.2894589865,"elo":1780.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0078855342,"x_train":1.5311965098,"imp":13.9256196498,"elo":1049.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1934792264,"x_train":17.8122979219,"imp":10.7303129531,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4003930169,"x_train":17.8122979219,"imp":9.9648302652,"elo":1283.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":0.4238705302,"imp":13.1089370831,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":79.9551879424,"imp":11.9892701847,"elo":1107.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":0.5707803637,"x_train":79.9551879424,"imp":11.8289448844,"elo":1114.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":0.2839085103,"x_train":6.5430745829,"imp":11.4020603961,"elo":1184.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.0682382085,"x_train":2068.1710397435,"imp":6.087337608,"elo":1630.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":1.3549179469,"x_train":2068.1710397435,"imp":4.8421550057,"elo":1800.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5059595331,"x_train":1.5872114628,"imp":4.5971768753,"elo":1710.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.6894524158,"x_train":739.1047147274,"imp":3.2815549654,"elo":1970.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"x_train":739.1047147274,"imp":3.185810595,"elo":1994.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.5250451841,"x_train":8.6828160405,"imp":12.2440158789,"elo":1118.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"x_train":41.5053698771,"imp":5.1411670708,"elo":1610.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":39.540415138,"x_train":4529.9305925825,"imp":4.3545991568,"elo":1741.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":357.2008760846,"x_train":4529.9305925825,"imp":4.0547168602,"elo":1790.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1767387784,"x_train":0.6611709932,"imp":4.2983424742,"elo":1723.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":11.6433178755,"x_train":25.1196552531,"imp":0.4756811693,"elo":2478.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1028500957,"x_train":0.471867489,"imp":3.5123157709,"elo":1805.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.038732221,"x_train":3.2167024044,"imp":9.732882352,"elo":1309.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.0448228661,"x_train":2410.3120769872,"imp":7.7035348907,"elo":1518.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":0.5807768467,"x_train":2410.3120769872,"imp":7.0742201117,"elo":1545.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"x_train":2.7774434997,"imp":4.214387882,"elo":1860.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042135993,"x_train":0.8699671047,"imp":1.7719216017,"elo":1948.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6147142423,"x_train":113.401275248,"imp":27.2788510175,"elo":511.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.139339258,"x_train":5782.1195021638,"imp":23.9886884373,"elo":646.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"x_train":5782.1195021638,"imp":23.7997636743,"elo":683.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1256335989,"x_train":0.5118384438,"imp":9.1227548353,"elo":1346.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0292963248,"x_train":14.0524525282,"imp":15.2314315415,"elo":1026.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0365087137,"x_train":4108.2381109512,"imp":11.0812079304,"elo":1180.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4787581258,"x_train":4108.2381109512,"imp":9.9939380427,"elo":1228.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.073497228,"x_train":0.5708322449,"imp":9.8998436121,"elo":1298.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.3603568941,"x_train":289.3497533723,"imp":8.8046523626,"elo":1430.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":1.7976430646,"x_train":289.3497533723,"imp":8.7389814156,"elo":1441.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":19.9963063451,"x_train":93.5081427033,"imp":16.4844284378,"elo":884.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":18.7518939426,"x_train":2150.1856290327,"imp":9.2887498658,"elo":1316.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":65.9338217731,"x_train":2150.1856290327,"imp":8.1653547355,"elo":1396.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":0.0932754819,"x_train":2.3191009865,"imp":10.6996583222,"elo":1246.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0348964576,"x_train":438.5586959956,"imp":7.3787958084,"elo":1496.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":0.5201417769,"x_train":438.5586959956,"imp":6.4083302989,"elo":1593.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
index 4939aca0398a631ae71fac9e9caf373f19291bfe..ec951320617bde61e666e58ea3bb182e9d63b087 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_regression/datasets_medium</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":2.7349656008,"imp":8.948053525,"elo":1483.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":8.045705742,"imp":8.7006398529,"elo":1510.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":21.5859747308,"imp":8.4569991571,"elo":1567.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":44.5235265964,"imp":8.3683097311,"elo":1576.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":109.7648010245,"imp":8.2422114531,"elo":1589.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":222.7546225359,"imp":8.2000098965,"elo":1597.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":456.5709140465,"imp":8.1667213574,"elo":1604.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":683.5034177869,"imp":8.1595704139,"elo":1611.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":917.6176937357,"imp":8.1505936003,"elo":1618.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":0.7437448515,"imp":13.8091021689,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1.4147629158,"imp":13.8484288322,"elo":1221.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":3.1548755046,"imp":13.7711647599,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":6.0549410867,"imp":13.5526854761,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":15.7085131241,"imp":13.2469953776,"elo":1314.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":31.5530586222,"imp":13.0195886576,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":62.4108808015,"imp":12.872512035,"elo":1374.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":93.2866322036,"imp":12.8120249297,"elo":1389.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":125.1955265319,"imp":12.7921068709,"elo":1395.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":2.3730743936,"imp":16.4557237796,"elo":993.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":9.4785326555,"imp":16.1785258957,"elo":1002.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":29.4810408518,"imp":15.9381690947,"elo":1013.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":54.9981205321,"imp":15.7982325684,"elo":1029.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":136.1756596671,"imp":15.5973941026,"elo":1066.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":274.1891727198,"imp":15.5230399323,"elo":1092.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":533.5366593544,"imp":15.452806981,"elo":1108.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":788.165832343,"imp":15.4287038401,"elo":1121.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":1069.963809484,"imp":15.4124099857,"elo":1131.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":0.8553953885,"imp":13.9525136374,"elo":966.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.1442763752,"imp":13.3237238784,"elo":1024.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1.877416167,"imp":12.6695881526,"elo":1091.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":3.1018975982,"imp":12.4540725039,"elo":1111.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":6.8290142883,"imp":12.0965906535,"elo":1148.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":12.9805618038,"imp":12.0194659771,"elo":1159.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":25.3410754681,"imp":11.9477961033,"elo":1177.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":37.7076806608,"imp":11.9239826929,"elo":1182.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":50.361853205,"imp":11.9175188382,"elo":1186.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1.3312352951,"imp":19.8825043615,"elo":671.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2.8523145707,"imp":19.6916039351,"elo":704.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":7.8648273493,"imp":19.2739055273,"elo":736.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":16.1612343105,"imp":18.9803391272,"elo":747.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":38.2613776387,"imp":17.8817086745,"elo":783.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":77.1343579507,"imp":17.1783670591,"elo":796.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":156.8008408086,"imp":16.6115391424,"elo":814.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":236.1441179195,"imp":16.2368590085,"elo":852.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":317.4820332824,"imp":16.0120587707,"elo":877.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":0.6567304564,"imp":10.0827395258,"elo":1253.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":1.8710214047,"imp":9.9377315646,"elo":1277.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":5.2206722708,"imp":9.1620265388,"elo":1406.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":10.6284575779,"imp":8.7076228752,"elo":1469.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":27.3782330474,"imp":8.1388935909,"elo":1569.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":54.9187089858,"imp":7.9706938952,"elo":1608.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":110.3418351857,"imp":7.8406830869,"elo":1633.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":166.090371503,"imp":7.7929819811,"elo":1646.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":223.0498746731,"imp":7.7440680251,"elo":1671.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":24.6223741935,"imp":7.9465233856,"elo":1395.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":6.870566258,"imp":7.2447354962,"elo":1497.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":22.1686321264,"imp":6.7310693457,"elo":1539.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":60.0453859865,"imp":5.8788285376,"elo":1625.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":121.2878502068,"imp":5.3769754646,"elo":1707.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":291.6044227147,"imp":4.7851514077,"elo":1816.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":631.2850224038,"imp":4.5443068635,"elo":1846.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":1243.1111273612,"imp":4.3770308899,"elo":1875.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":1862.6240988571,"imp":4.298791846,"elo":1900.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":2552.3091253937,"imp":4.2642248374,"elo":1904.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":1.9712126532,"imp":5.9351831745,"elo":1652.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":4.8565548774,"imp":5.2895686255,"elo":1824.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":1.5311965098,"imp":13.9257122688,"elo":1055.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":5.1565448569,"imp":11.5962887488,"elo":1173.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":14.9849409603,"imp":10.0459712109,"elo":1268.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":17.8122979219,"imp":9.9649292888,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.4238705302,"imp":13.1090349978,"elo":1009.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":0.8229555909,"imp":12.7357834902,"elo":1054.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":2.0559128426,"imp":12.3965155326,"elo":1083.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":4.0681999454,"imp":12.1045019075,"elo":1101.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":9.9071084465,"imp":11.9373680623,"elo":1123.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":19.6617455737,"imp":11.8680574316,"elo":1130.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":39.5175987803,"imp":11.8408256094,"elo":1137.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":59.4870987604,"imp":11.8336701035,"elo":1143.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":79.9551879424,"imp":11.8290459506,"elo":1141.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":6.5430745829,"imp":11.4021616834,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":16.387957176,"imp":9.0872668175,"elo":1364.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":45.678712897,"imp":6.8982447724,"elo":1569.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":100.5864418887,"imp":6.1036598702,"elo":1644.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":253.2316993705,"imp":5.5083757333,"elo":1730.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":513.1174708444,"imp":5.211058434,"elo":1765.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":1026.1644887093,"imp":5.0189052637,"elo":1796.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":1544.7351451859,"imp":4.9248470231,"elo":1826.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":2068.1710397435,"imp":4.8422678218,"elo":1840.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1.5872114628,"imp":4.5972922799,"elo":1726.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":5.1930136195,"imp":4.4954936788,"elo":1738.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":16.3030487246,"imp":4.2817235886,"elo":1750.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":34.3657315438,"imp":3.9788567334,"elo":1813.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":91.1547591648,"imp":3.6432521816,"elo":1940.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":182.7153391231,"imp":3.4430381734,"elo":1968.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":366.5547593939,"imp":3.2949600688,"elo":2025.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":550.802457324,"imp":3.2196038239,"elo":2053.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":739.1047147274,"imp":3.1862333181,"elo":2068.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":8.6828160405,"imp":12.2441147822,"elo":1130.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":41.5053698771,"imp":5.1412864523,"elo":1621.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":63.019350737,"imp":5.0028139527,"elo":1648.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":119.7137023169,"imp":4.8377133521,"elo":1670.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":218.048215922,"imp":4.7224256158,"elo":1691.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":550.8419093004,"imp":4.4758639758,"elo":1734.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":1115.34725308,"imp":4.2994519026,"elo":1762.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":2246.8238340239,"imp":4.1625090429,"elo":1788.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":3382.9355985564,"imp":4.0875733992,"elo":1798.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":4529.9305925825,"imp":4.0547168602,"elo":1807.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":0.6611709932,"imp":4.2984610916,"elo":1757.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":25.1196552531,"imp":0.4758006822,"elo":2505.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":0.471867489,"imp":3.5124324077,"elo":1821.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":3.2167024044,"imp":9.7329856034,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":11.7992281299,"imp":9.1947777236,"elo":1425.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":46.940037227,"imp":8.4487551977,"elo":1465.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":112.2128891191,"imp":7.9823169354,"elo":1511.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":298.6395201823,"imp":7.5780381651,"elo":1534.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":596.2505830634,"imp":7.3583930626,"elo":1548.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1163.2433543614,"imp":7.2040262897,"elo":1561.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":1770.0748555561,"imp":7.1312564436,"elo":1574.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":2410.3120769872,"imp":7.0743284296,"elo":1586.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":2.7774434997,"imp":4.2145025488,"elo":1892.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":0.8699671047,"imp":1.7853424303,"elo":1935.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":113.401275248,"imp":27.2789297045,"elo":489.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":141.7238706247,"imp":27.1544799482,"elo":523.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":221.1962991478,"imp":26.854399481,"elo":539.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":328.7844244998,"imp":26.3300775969,"elo":555.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":736.9725789769,"imp":25.4060212814,"elo":585.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":1485.4343663821,"imp":24.562354425,"elo":610.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2941.8656957704,"imp":24.2062265581,"elo":626.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4341.8172896327,"imp":23.9692582481,"elo":659.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":5782.1195021638,"imp":23.7998469089,"elo":694.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":0.5118384438,"imp":9.1228658323,"elo":1355.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":14.0524525282,"imp":15.2315343973,"elo":1027.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":35.7395534737,"imp":14.3339737912,"elo":1046.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":93.9257318856,"imp":12.9646971276,"elo":1121.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":201.2335467269,"imp":12.1017547455,"elo":1165.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":481.4580064735,"imp":11.1402059021,"elo":1201.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1015.3213795664,"imp":10.7333183709,"elo":1215.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2031.0046161997,"imp":10.312716384,"elo":1234.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3088.2903635944,"imp":10.1160578174,"elo":1241.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":4108.2381109512,"imp":9.9940462779,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":0.5708322449,"imp":9.8999433384,"elo":1302.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":2.4574329869,"imp":9.7581492837,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":7.3610615249,"imp":9.3444866285,"elo":1381.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":13.6386574902,"imp":9.1536785083,"elo":1397.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":35.9388283515,"imp":8.9518819388,"elo":1422.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":71.5452674162,"imp":8.8329662502,"elo":1441.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":142.0559469286,"imp":8.7587291911,"elo":1454.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":215.0273705137,"imp":8.7490331287,"elo":1459.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":289.3497533723,"imp":8.7390847655,"elo":1457.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":93.5081427033,"imp":16.4845177929,"elo":866.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":175.8703779457,"imp":13.5955714626,"elo":988.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":423.2568217882,"imp":24.3807034641,"elo":1102.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":836.4112409598,"imp":21.9294986191,"elo":1189.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":2070.7507295344,"imp":8.1871322543,"elo":1404.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":2150.1856290327,"imp":8.1654629203,"elo":1412.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":2.3191009865,"imp":10.699762351,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":4.6217004586,"imp":9.8879494154,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":11.2199797524,"imp":8.783901562,"elo":1383.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":22.2331145434,"imp":7.7860451008,"elo":1464.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":55.3170648503,"imp":7.0074538977,"elo":1527.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":109.9515587293,"imp":6.728401508,"elo":1558.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":218.3385295205,"imp":6.5716820878,"elo":1573.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":327.734067188,"imp":6.4606380858,"elo":1589.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":438.5586959956,"imp":6.4091172112,"elo":1611.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0103037461,"x_train":2.7349656008,"imp":8.948053525,"elo":1483.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.0484760077,"x_train":8.045705742,"imp":8.7006398529,"elo":1510.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1185677885,"x_train":21.5859747308,"imp":8.4569991571,"elo":1567.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.1738042849,"x_train":44.5235265964,"imp":8.3683097311,"elo":1576.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2388904639,"x_train":109.7648010245,"imp":8.2422114531,"elo":1589.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3111317954,"x_train":222.7546225359,"imp":8.2000098965,"elo":1597.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.398317095,"x_train":456.5709140465,"imp":8.1667213574,"elo":1604.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3902829889,"x_train":683.5034177869,"imp":8.1595704139,"elo":1611.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4221789337,"x_train":917.6176937357,"imp":8.1505936003,"elo":1618.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0135049429,"x_train":0.7437448515,"imp":13.8091021689,"elo":1231.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0286415584,"x_train":1.4147629158,"imp":13.8484288322,"elo":1221.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0683284014,"x_train":3.1548755046,"imp":13.7711647599,"elo":1247.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1120636664,"x_train":6.0549410867,"imp":13.5526854761,"elo":1276.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1607817441,"x_train":15.7085131241,"imp":13.2469953776,"elo":1314.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1954148357,"x_train":31.5530586222,"imp":13.0195886576,"elo":1352.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.246440357,"x_train":62.4108808015,"imp":12.872512035,"elo":1374.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2772856864,"x_train":93.2866322036,"imp":12.8120249297,"elo":1389.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2761852287,"x_train":125.1955265319,"imp":12.7921068709,"elo":1395.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0059270235,"x_train":2.3730743936,"imp":16.4557237796,"elo":993.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0111616749,"x_train":9.4785326555,"imp":16.1785258957,"elo":1002.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0221705298,"x_train":29.4810408518,"imp":15.9381690947,"elo":1013.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0321482347,"x_train":54.9981205321,"imp":15.7982325684,"elo":1029.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0503208887,"x_train":136.1756596671,"imp":15.5973941026,"elo":1066.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.0568078567,"x_train":274.1891727198,"imp":15.5230399323,"elo":1092.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.0668261402,"x_train":533.5366593544,"imp":15.452806981,"elo":1108.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.0718867712,"x_train":788.165832343,"imp":15.4287038401,"elo":1121.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.0853557077,"x_train":1069.963809484,"imp":15.4124099857,"elo":1131.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0992770116,"x_train":0.8553953885,"imp":13.9525136374,"elo":966.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0906794725,"x_train":1.1442763752,"imp":13.3237238784,"elo":1024.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.0932501865,"x_train":1.877416167,"imp":12.6695881526,"elo":1091.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1085131081,"x_train":3.1018975982,"imp":12.4540725039,"elo":1111.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1369942638,"x_train":6.8290142883,"imp":12.0965906535,"elo":1148.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.1923137349,"x_train":12.9805618038,"imp":12.0194659771,"elo":1159.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2461819734,"x_train":25.3410754681,"imp":11.9477961033,"elo":1177.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2514474536,"x_train":37.7076806608,"imp":11.9239826929,"elo":1182.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.2319944846,"x_train":50.361853205,"imp":11.9175188382,"elo":1186.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1256610016,"x_train":1.3312352951,"imp":19.8825043615,"elo":671.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.1869244318,"x_train":2.8523145707,"imp":19.6916039351,"elo":704.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.3616276108,"x_train":7.8648273493,"imp":19.2739055273,"elo":736.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.4682651613,"x_train":16.1612343105,"imp":18.9803391272,"elo":747.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.474521655,"x_train":38.2613776387,"imp":17.8817086745,"elo":783.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5649569566,"x_train":77.1343579507,"imp":17.1783670591,"elo":796.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.6774236469,"x_train":156.8008408086,"imp":16.6115391424,"elo":814.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7132812118,"x_train":236.1441179195,"imp":16.2368590085,"elo":852.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.7187718544,"x_train":317.4820332824,"imp":16.0120587707,"elo":877.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.2133657073,"x_train":0.6567304564,"imp":10.0827395258,"elo":1253.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.4074904298,"x_train":1.8710214047,"imp":9.9377315646,"elo":1277.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.1014164121,"x_train":5.2206722708,"imp":9.1620265388,"elo":1406.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":3.5286110142,"x_train":10.6284575779,"imp":8.7076228752,"elo":1469.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.2339770555,"x_train":27.3782330474,"imp":8.1388935909,"elo":1569.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":7.9246780943,"x_train":54.9187089858,"imp":7.9706938952,"elo":1608.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.715416536,"x_train":110.3418351857,"imp":7.8406830869,"elo":1633.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":8.9190819422,"x_train":166.090371503,"imp":7.7929819811,"elo":1646.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":9.7062590878,"x_train":223.0498746731,"imp":7.7440680251,"elo":1671.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":8.8350020269,"x_train":24.6223741935,"imp":7.9465233856,"elo":1395.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.097838007,"x_train":6.870566258,"imp":7.2447354962,"elo":1497.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.2613333852,"x_train":22.1686321264,"imp":6.7310693457,"elo":1539.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.58187982,"x_train":60.0453859865,"imp":5.8788285376,"elo":1625.0,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.9883252386,"x_train":121.2878502068,"imp":5.3769754646,"elo":1707.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.7805982465,"x_train":291.6044227147,"imp":4.7851514077,"elo":1816.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.3680852352,"x_train":631.2850224038,"imp":4.5443068635,"elo":1846.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.0820243271,"x_train":1243.1111273612,"imp":4.3770308899,"elo":1875.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.9287026871,"x_train":1862.6240988571,"imp":4.298791846,"elo":1900.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":4.5218047762,"x_train":2552.3091253937,"imp":4.2642248374,"elo":1904.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":0.6513221504,"x_train":1.9712126532,"imp":5.9351831745,"elo":1652.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":1.8805802898,"x_train":4.8565548774,"imp":5.2895686255,"elo":1824.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0078855342,"x_train":1.5311965098,"imp":13.9257122688,"elo":1055.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1032870492,"x_train":5.1565448569,"imp":11.5962887488,"elo":1173.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.3397827375,"x_train":14.9849409603,"imp":10.0459712109,"elo":1268.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4003930169,"x_train":17.8122979219,"imp":9.9649292888,"elo":1281.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.0525125134,"x_train":0.4238705302,"imp":13.1090349978,"elo":1009.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1053775796,"x_train":0.8229555909,"imp":12.7357834902,"elo":1054.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1860415756,"x_train":2.0559128426,"imp":12.3965155326,"elo":1083.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.271253998,"x_train":4.0681999454,"imp":12.1045019075,"elo":1101.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.3828553913,"x_train":9.9071084465,"imp":11.9373680623,"elo":1123.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.4327898753,"x_train":19.6617455737,"imp":11.8680574316,"elo":1130.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5237437037,"x_train":39.5175987803,"imp":11.8408256094,"elo":1137.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5507249031,"x_train":59.4870987604,"imp":11.8336701035,"elo":1143.6,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5707803637,"x_train":79.9551879424,"imp":11.8290459506,"elo":1141.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.2839085103,"x_train":6.5430745829,"imp":11.4021616834,"elo":1192.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.3853454659,"x_train":16.387957176,"imp":9.0872668175,"elo":1364.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.4768426596,"x_train":45.678712897,"imp":6.8982447724,"elo":1569.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.6715590282,"x_train":100.5864418887,"imp":6.1036598702,"elo":1644.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":0.9551220317,"x_train":253.2316993705,"imp":5.5083757333,"elo":1730.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.2841445532,"x_train":513.1174708444,"imp":5.211058434,"elo":1765.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.5404194607,"x_train":1026.1644887093,"imp":5.0189052637,"elo":1796.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.4492208192,"x_train":1544.7351451859,"imp":4.9248470231,"elo":1826.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":1.3549179469,"x_train":2068.1710397435,"imp":4.8422678218,"elo":1840.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5059595331,"x_train":1.5872114628,"imp":4.5972922799,"elo":1726.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.3928986427,"x_train":5.1930136195,"imp":4.4954936788,"elo":1738.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":3.3467414164,"x_train":16.3030487246,"imp":4.2817235886,"elo":1750.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.2833870702,"x_train":34.3657315438,"imp":3.9788567334,"elo":1813.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.2909325197,"x_train":91.1547591648,"imp":3.6432521816,"elo":1940.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.4808001914,"x_train":182.7153391231,"imp":3.4430381734,"elo":1968.0,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":8.6468940873,"x_train":366.5547593939,"imp":3.2949600688,"elo":2025.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":9.637851993,"x_train":550.802457324,"imp":3.2196038239,"elo":2053.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":11.6495886961,"x_train":739.1047147274,"imp":3.1862333181,"elo":2068.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":7.5250451841,"x_train":8.6828160405,"imp":12.2441147822,"elo":1130.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":38.2970764109,"x_train":41.5053698771,"imp":5.1412864523,"elo":1621.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":52.9928561684,"x_train":63.019350737,"imp":5.0028139527,"elo":1648.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":74.454905827,"x_train":119.7137023169,"imp":4.8377133521,"elo":1670.5,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":105.0879768186,"x_train":218.048215922,"imp":4.7224256158,"elo":1691.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":170.7606880264,"x_train":550.8419093004,"imp":4.4758639758,"elo":1734.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":225.019656008,"x_train":1115.34725308,"imp":4.2994519026,"elo":1762.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":286.7578860717,"x_train":2246.8238340239,"imp":4.1625090429,"elo":1788.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":338.6200879093,"x_train":3382.9355985564,"imp":4.0875733992,"elo":1798.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":357.2008760846,"x_train":4529.9305925825,"imp":4.0547168602,"elo":1807.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.1767387784,"x_train":0.6611709932,"imp":4.2984610916,"elo":1757.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":11.6433178755,"x_train":25.1196552531,"imp":0.4758006822,"elo":2505.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1028500957,"x_train":0.471867489,"imp":3.5124324077,"elo":1821.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.038732221,"x_train":3.2167024044,"imp":9.7329856034,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.0791734092,"x_train":11.7992281299,"imp":9.1947777236,"elo":1425.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":0.1510658648,"x_train":46.940037227,"imp":8.4487551977,"elo":1465.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":0.2267457056,"x_train":112.2128891191,"imp":7.9823169354,"elo":1511.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":0.3639606082,"x_train":298.6395201823,"imp":7.5780381651,"elo":1534.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":0.4652595841,"x_train":596.2505830634,"imp":7.3583930626,"elo":1548.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":0.6436103929,"x_train":1163.2433543614,"imp":7.2040262897,"elo":1561.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":0.6500065101,"x_train":1770.0748555561,"imp":7.1312564436,"elo":1574.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":0.5807768467,"x_train":2410.3120769872,"imp":7.0743284296,"elo":1586.2,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.7022481243,"x_train":2.7774434997,"imp":4.2145025488,"elo":1892.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.2042135993,"x_train":0.8699671047,"imp":1.7853424303,"elo":1935.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.6147142423,"x_train":113.401275248,"imp":27.2789297045,"elo":489.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":1.7368429189,"x_train":141.7238706247,"imp":27.1544799482,"elo":523.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.2568204055,"x_train":221.1962991478,"imp":26.854399481,"elo":539.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":2.7676176669,"x_train":328.7844244998,"imp":26.3300775969,"elo":555.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":4.1744464711,"x_train":736.9725789769,"imp":25.4060212814,"elo":585.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5384129217,"x_train":1485.4343663821,"imp":24.562354425,"elo":610.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":6.5382733421,"x_train":2941.8656957704,"imp":24.2062265581,"elo":626.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":7.2160549793,"x_train":4341.8172896327,"imp":23.9692582481,"elo":659.3,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":8.5672328833,"x_train":5782.1195021638,"imp":23.7998469089,"elo":694.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.1256335989,"x_train":0.5118384438,"imp":9.1228658323,"elo":1355.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0292963248,"x_train":14.0524525282,"imp":15.2315343973,"elo":1027.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.0753848671,"x_train":35.7395534737,"imp":14.3339737912,"elo":1046.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1421186744,"x_train":93.9257318856,"imp":12.9646971276,"elo":1121.1,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1772212357,"x_train":201.2335467269,"imp":12.1017547455,"elo":1165.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2513416372,"x_train":481.4580064735,"imp":11.1402059021,"elo":1201.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.39401578,"x_train":1015.3213795664,"imp":10.7333183709,"elo":1215.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5446242225,"x_train":2031.0046161997,"imp":10.312716384,"elo":1234.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.5498324906,"x_train":3088.2903635944,"imp":10.1160578174,"elo":1241.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4787581258,"x_train":4108.2381109512,"imp":9.9940462779,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.073497228,"x_train":0.5708322449,"imp":9.8999433384,"elo":1302.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.3133210319,"x_train":2.4574329869,"imp":9.7581492837,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.8172787201,"x_train":7.3610615249,"imp":9.3444866285,"elo":1381.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.0996225013,"x_train":13.6386574902,"imp":9.1536785083,"elo":1397.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.4572887885,"x_train":35.9388283515,"imp":8.9518819388,"elo":1422.0,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7915150002,"x_train":71.5452674162,"imp":8.8329662502,"elo":1441.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.9530600967,"x_train":142.0559469286,"imp":8.7587291911,"elo":1454.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0684490809,"x_train":215.0273705137,"imp":8.7490331287,"elo":1459.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":1.7976430646,"x_train":289.3497533723,"imp":8.7390847655,"elo":1457.5,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":19.9963063451,"x_train":93.5081427033,"imp":16.4845177929,"elo":866.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":27.6019600638,"x_train":175.8703779457,"imp":13.5955714626,"elo":988.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":40.3210715012,"x_train":423.2568217882,"imp":24.3807034641,"elo":1102.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":45.7301059946,"x_train":836.4112409598,"imp":21.9294986191,"elo":1189.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":67.2056620148,"x_train":2070.7507295344,"imp":8.1871322543,"elo":1404.6,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":65.9338217731,"x_train":2150.1856290327,"imp":8.1654629203,"elo":1412.2,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":0.0932754819,"x_train":2.3191009865,"imp":10.699762351,"elo":1258.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":0.122018148,"x_train":4.6217004586,"imp":9.8879494154,"elo":1316.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":0.1678454644,"x_train":11.2199797524,"imp":8.783901562,"elo":1383.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":0.2461082837,"x_train":22.2331145434,"imp":7.7860451008,"elo":1464.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":0.3494917743,"x_train":55.3170648503,"imp":7.0074538977,"elo":1527.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":0.4163168833,"x_train":109.9515587293,"imp":6.728401508,"elo":1558.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":0.4485143,"x_train":218.3385295205,"imp":6.5716820878,"elo":1573.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":0.4523893299,"x_train":327.734067188,"imp":6.4606380858,"elo":1589.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":0.5201417769,"x_train":438.5586959956,"imp":6.4091172112,"elo":1611.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..e89dc7a4c4834128341ab4f30ddc4c0dbb3f9652
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_regression/datasets_medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "RealTabPFN-2.5 (tuned + ensembled)", "RealTabPFN-2.5 (tuned)", "TabPFN-3 (default)", "AutoGluon 1.5 (extreme, 4h)", "ModernNCA (tuned + ensembled)", "TabPFN-2.6 (default)", "AutoGluon 1.4 (best, 4h)", "TabICLv2 (default)", "RealMLP (tuned + ensembled)", "TabDPT (tuned + ensembled)", "Nori-30M (default)", "TabDPT (tuned)", "TabDPT-Turbo (default)", "RealTabPFN-2.5 (default)", "Nori (default)", "RealMLP (tuned)", "LightGBM (tuned + ensembled)", "TabDPT (default)", "LightGBM (tuned)", "ModernNCA (tuned)", "xRFM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "TabM (tuned + ensembled)", "TabM (tuned)", "xRFM (tuned)", "CatBoost (default)", "ModernNCA (default)", "XGBoost (tuned + ensembled)", "XGBoost (tuned)", "iLTM (tuned + ensembled)", "LimiX (default)", "ChimeraBoost (tuned + ensembled)", "TabSwift (default)", "ChimeraBoost (tuned)", "iLTM (tuned)", "TabM (default)", "XGBoost (default)", "PerpetualBooster (tuned + ensembled)", "LightGBM (default)", "xRFM (default)", "ChimeraBoost (default)", "TorchMLP (tuned + ensembled)", "PerpetualBooster (tuned)", "RealMLP (default)", "TorchMLP (tuned)", "ExtraTrees (tuned + ensembled)", "ExtraTrees (tuned)", "SAP-RPT-OSS (default)", "EBM (tuned + ensembled)", "RandomForest (tuned + ensembled)", "RandomForest (tuned)", "PerpetualBooster (default)", "TorchMLP (default)", "EBM (tuned)", "RandomForest (default)", "EBM (default)", "ExtraTrees (default)", "iLTM (default)", "FastaiMLP (tuned + ensembled)", "FastaiMLP (tuned)", "KNN (tuned + ensembled)", "FastaiMLP (default)", "TabSTAR (tuned + ensembled)", "TabSTAR (tuned)", "KNN (tuned)", "TabSTAR (default)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, NaN, 0.6000000238418579, 0.20000000298023224, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.4000000059604645, NaN, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 1.0, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.800000011920929, 0.800000011920929, NaN, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, NaN, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 1.0, 0.6000000238418579, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, NaN, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, NaN, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, NaN, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 1.0, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, NaN, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.6000000238418579, 1.0, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, NaN, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.6000000238418579, 1.0, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, NaN, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.0, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, NaN, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 1.0, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.800000011920929, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 1.0, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, NaN, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.20000000298023224, 0.4000000059604645, 0.800000011920929, 0.800000011920929, NaN, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.800000011920929, 0.20000000298023224, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.20000000298023224, NaN, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.0, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.20000000298023224, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 0.4000000059604645, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.0, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.800000011920929, 0.20000000298023224, 0.800000011920929, 1.0, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.0, 0.4000000059604645, 0.0, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, NaN, 0.20000000298023224, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.800000011920929, NaN, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 1.0, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.0, NaN, 0.800000011920929, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, NaN, 0.6000000238418579, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.800000011920929, 1.0, 0.4000000059604645, 1.0, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, NaN, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.699999988079071, 0.6000000238418579, 0.6000000238418579, 1.0, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.0, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.20000000298023224, 0.4000000059604645, 0.0, 0.6000000238418579, 0.6000000238418579, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.0, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.800000011920929, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.6000000238418579, 0.6000000238418579, 0.20000000298023224, 0.30000001192092896, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.0, 0.4000000059604645, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, NaN, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 0.800000011920929, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.0, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, 0.4000000059604645, 0.6000000238418579, NaN, 0.800000011920929, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.0, 0.0, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.20000000298023224, 0.6000000238418579, 0.20000000298023224, NaN, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.800000011920929, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.6000000238418579, NaN, 0.6000000238418579, 0.4000000059604645, 1.0, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 0.800000011920929, 0.800000011920929, 0.800000011920929, 0.6000000238418579, 0.800000011920929, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.4000000059604645, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.0, 0.4000000059604645, 0.20000000298023224, 0.4000000059604645, 0.6000000238418579, 0.6000000238418579, NaN, 0.6000000238418579, 0.6000000238418579, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.0, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.4000000059604645, NaN, 0.800000011920929, 0.800000011920929, 0.4000000059604645, 0.800000011920929, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, NaN, 0.800000011920929, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.20000000298023224, NaN, 0.6000000238418579, 1.0, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.4000000059604645, 0.0, 0.6000000238418579, 0.4000000059604645, 0.4000000059604645, NaN, 0.4000000059604645, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.20000000298023224, 0.4000000059604645, 0.20000000298023224, 0.0, 0.0, 0.6000000238418579, NaN, 0.6000000238418579, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4000000059604645, 0.4000000059604645, 0.4000000059604645, 0.0, 0.4000000059604645, NaN, 0.800000011920929, 0.800000011920929, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, NaN, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, NaN, 0.800000011920929], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.20000000298023224, 0.0, 0.20000000298023224, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9915492963},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.9126760601},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.9042253555},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.8957746531},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.876056341},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.8591549344},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.8563380325},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.8366197215},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.8281690224},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.825352118},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.819718317},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.8140845116},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.7915493027},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7802816959},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.771830993},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.7211267661},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.7183098675},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.7126760613},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7042253611},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.7014084551},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.6985915572},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.6929577524},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6901408491},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.6676056385},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6591549358},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.6394366261},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.6225352178},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.6140845124},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.6056338104},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.580281694},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.57183099},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.5464788778},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.5436619787},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5352112768},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.5098591638},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.5070422621},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.4873239486},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4816901466},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.4732394412},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4619718354},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.436619724},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.4338028215},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4253521203},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.4197183146},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.3943662003},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.3887323993},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3859154985},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3830985966},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.3661971879},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.343661977},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3436619761},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3408450747},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.3366197226},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.2985915529},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2845070467},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.2732394405},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2690140878},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.2619718349},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.2619718353},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.2084507071},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.1971831011},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.1830985936},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.1605633829},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1352112692},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.1295774663},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.1183098605},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.1098591559},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.0845070433},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.0647887334},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0309859155},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0140845073},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.0056338029}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..8c88bbbb098105d711f86e5809fc5dd9c93c348a
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_medium/winrate_matrix.csv
@@ -0,0 +1,73 @@
+,TabFM (default),RealTabPFN-2.5 (tuned + ensembled),RealTabPFN-2.5 (tuned),TabPFN-3 (default),"AutoGluon 1.5 (extreme, 4h)",ModernNCA (tuned + ensembled),TabPFN-2.6 (default),"AutoGluon 1.4 (best, 4h)",TabICLv2 (default),RealMLP (tuned + ensembled),TabDPT (tuned + ensembled),Nori-30M (default),TabDPT (tuned),TabDPT-Turbo (default),RealTabPFN-2.5 (default),Nori (default),RealMLP (tuned),LightGBM (tuned + ensembled),TabDPT (default),LightGBM (tuned),ModernNCA (tuned),xRFM (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),TabM (tuned + ensembled),TabM (tuned),xRFM (tuned),CatBoost (default),ModernNCA (default),XGBoost (tuned + ensembled),XGBoost (tuned),iLTM (tuned + ensembled),LimiX (default),ChimeraBoost (tuned + ensembled),TabSwift (default),ChimeraBoost (tuned),iLTM (tuned),TabM (default),XGBoost (default),PerpetualBooster (tuned + ensembled),LightGBM (default),xRFM (default),ChimeraBoost (default),TorchMLP (tuned + ensembled),PerpetualBooster (tuned),RealMLP (default),TorchMLP (tuned),ExtraTrees (tuned + ensembled),ExtraTrees (tuned),SAP-RPT-OSS (default),EBM (tuned + ensembled),RandomForest (tuned + ensembled),RandomForest (tuned),PerpetualBooster (default),TorchMLP (default),EBM (tuned),RandomForest (default),EBM (default),ExtraTrees (default),iLTM (default),FastaiMLP (tuned + ensembled),FastaiMLP (tuned),KNN (tuned + ensembled),FastaiMLP (default),TabSTAR (tuned + ensembled),TabSTAR (tuned),KNN (tuned),TabSTAR (default),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.0,,0.6,0.2,0.6,0.8,0.8,0.8,0.6,1.0,0.6,0.8,0.6,0.8,1.0,1.0,1.0,1.0,0.6,0.8,1.0,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.0,0.4,,0.2,0.6,0.6,0.8,0.8,0.4,1.0,0.6,1.0,0.6,0.8,0.8,1.0,1.0,0.8,0.6,0.8,1.0,1.0,0.8,0.8,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.2,0.8,0.8,,0.8,0.8,0.8,0.8,0.8,0.8,0.4,0.8,0.6,0.8,1.0,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,0.8,1.0,0.8,0.8,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.2,0.4,0.4,0.2,,0.4,0.6,0.8,0.2,0.6,0.4,0.6,0.4,0.8,0.4,0.8,0.8,1.0,0.6,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.0,0.2,0.4,0.2,0.6,,0.4,0.6,0.2,0.6,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.6,0.8,1.0,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.0,0.2,0.2,0.2,0.4,0.6,,0.4,0.2,0.6,0.6,0.6,0.6,0.6,0.6,0.8,1.0,0.8,0.6,0.8,1.0,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.2,0.2,0.2,0.2,0.4,0.6,,0.2,0.4,0.4,0.4,0.4,0.6,0.4,0.8,0.8,1.0,0.6,0.8,0.6,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.0,0.4,0.6,0.2,0.8,0.8,0.8,0.8,,0.8,0.6,0.8,0.6,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,1.0,1.0,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.0,0.2,0.4,0.4,0.4,0.6,0.2,,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.6,0.6,0.6,0.6,1.0,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.2,0.4,0.4,0.6,0.6,0.4,0.4,0.6,0.4,0.4,,0.6,1.0,0.8,0.6,0.8,0.4,0.6,1.0,0.6,0.6,0.8,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori-30M (default),0.0,0.2,0.0,0.2,0.4,0.4,0.4,0.6,0.2,0.4,0.4,,0.6,0.8,0.4,0.8,0.4,0.8,0.6,0.8,0.6,0.8,0.8,0.8,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.4,0.4,0.4,0.6,0.4,0.4,0.6,0.4,0.4,0.0,0.4,,0.8,0.6,0.8,0.4,0.6,1.0,0.6,0.6,0.8,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.2,0.2,0.2,0.2,0.4,0.4,0.4,0.4,0.4,0.2,0.2,0.2,,0.6,0.8,0.6,0.6,0.8,0.6,0.8,0.6,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,0.8,1.0,0.8,0.8,1.0,0.8,1.0,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.0,0.0,0.2,0.0,0.6,0.4,0.4,0.6,0.4,0.4,0.4,0.6,0.4,0.4,,0.8,0.6,0.6,0.4,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori (default),0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,,0.6,0.6,0.4,0.6,0.4,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.6,0.8,0.8,1.0,0.8,0.8,0.8,0.8,1.0,1.0,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.6,0.6,0.6,0.4,0.4,0.4,,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.6,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,1.0,0.8,1.0,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.4,0.4,0.2,0.4,0.4,0.4,0.4,0.4,,0.4,0.6,0.4,0.4,0.6,0.6,0.4,0.4,0.6,1.0,0.6,1.0,0.8,0.8,0.6,0.8,0.6,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.4,0.4,0.0,0.4,0.4,0.4,0.4,0.4,0.4,0.0,0.4,0.0,0.2,0.6,0.6,0.4,0.6,,0.6,0.4,0.6,0.6,0.6,0.6,0.8,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.6,0.6,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.4,,0.4,0.4,0.6,0.6,0.2,0.2,0.4,0.6,0.4,1.0,1.0,0.8,0.6,0.8,0.6,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.4,0.2,0.4,0.4,0.4,0.4,0.2,0.2,0.6,0.4,0.6,0.6,0.6,,0.6,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.6,0.6,1.0,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.2,0.4,0.6,0.4,0.6,0.4,,0.6,0.6,0.6,0.6,1.0,0.6,0.8,0.6,0.6,1.0,0.8,0.6,0.8,0.6,1.0,0.8,0.8,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.4,0.4,0.4,0.4,0.4,,0.8,0.2,0.2,0.6,1.0,0.4,0.8,0.8,0.8,0.6,0.8,0.6,1.0,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.2,,0.2,0.2,0.6,0.6,0.4,0.6,0.6,0.8,0.6,0.8,0.6,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.6,0.4,0.8,0.2,0.4,0.8,0.8,,0.8,0.4,0.8,0.4,0.8,0.8,0.8,0.6,0.8,0.6,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.6,0.2,0.8,0.2,0.4,0.8,0.8,0.2,,0.4,0.8,0.4,0.8,0.8,0.6,0.6,0.8,0.6,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.8,0.6,0.8,1.0,0.8,1.0,1.0,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.4,0.4,0.6,0.2,0.0,0.4,0.4,0.6,0.6,,0.6,0.8,0.6,0.6,0.8,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.8,0.8,1.0,0.6,1.0,1.0,1.0,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.0,0.4,0.4,0.4,0.4,0.0,0.4,0.2,0.2,0.4,,0.4,0.8,0.8,0.8,0.6,0.4,0.6,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,0.8,0.6,1.0,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.2,0.4,0.2,0.4,0.4,0.6,0.2,0.2,0.6,0.6,0.6,0.6,0.2,0.6,,0.6,0.6,0.8,0.4,0.6,0.6,0.6,0.8,0.8,0.6,0.6,0.8,0.8,0.6,1.0,1.0,0.8,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.4,0.0,0.2,0.4,0.2,0.4,0.2,0.2,0.4,0.2,0.4,,0.6,0.4,0.6,0.4,0.4,0.4,0.8,0.6,0.8,1.0,1.0,0.8,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.0,0.2,0.4,0.2,0.4,0.2,0.2,0.4,0.2,0.4,0.4,,0.4,0.6,0.4,0.4,0.4,0.8,0.6,0.8,0.8,0.8,0.6,0.8,0.8,1.0,0.8,0.8,1.0,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.6,0.6,,0.6,0.4,0.4,0.4,0.8,0.6,0.8,0.8,0.8,1.0,0.4,0.8,0.8,0.8,1.0,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.0,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.4,0.4,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.6,0.4,0.4,0.4,,0.6,0.6,0.6,0.4,0.4,0.6,0.4,0.4,0.6,0.6,0.8,0.6,0.8,0.8,0.4,0.4,1.0,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.4,0.4,0.2,0.2,0.2,0.2,0.4,0.6,0.4,0.6,0.6,0.6,0.4,,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.6,1.0,0.6,0.8,0.8,0.6,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0
+TabSwift (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.2,0.4,0.2,0.4,0.0,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.4,0.6,0.6,0.6,0.4,0.4,,0.4,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.6,0.8,0.6,0.6,0.6,0.8,0.8,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.2,0.4,0.0,0.2,0.2,0.2,0.4,0.2,0.4,0.6,0.6,0.6,0.4,0.4,0.6,,0.6,0.6,0.8,0.8,0.8,0.6,1.0,0.6,0.8,0.8,0.6,0.8,0.8,0.6,0.6,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.6,0.4,0.4,0.4,,0.4,0.4,0.6,0.8,0.8,0.6,0.8,0.6,0.8,0.8,0.6,0.6,0.8,0.8,1.0,1.0,0.8,1.0,0.8,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.4,0.4,0.4,0.6,0.4,0.4,0.4,0.6,,0.8,0.8,0.8,0.6,0.6,0.6,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.6,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.4,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.4,0.2,0.4,0.2,0.6,0.2,,0.6,0.4,0.6,0.4,0.8,0.8,0.6,0.8,0.4,0.4,0.6,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.0,0.4,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.4,0.0,0.2,0.2,0.6,0.2,0.4,0.2,0.4,0.2,0.4,,0.4,0.8,0.2,0.8,1.0,0.6,0.8,0.6,0.6,0.6,0.8,0.8,0.8,1.0,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.6,0.2,0.4,0.2,0.2,0.2,0.6,0.6,,0.6,0.4,0.6,0.8,0.6,0.6,0.6,0.6,0.8,0.8,0.6,0.6,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.4,0.0,0.4,0.4,0.4,0.4,0.2,0.4,0.4,0.2,0.4,,0.4,0.4,0.6,0.6,0.4,0.8,0.8,0.8,0.6,0.8,0.8,0.6,0.8,0.6,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.6,0.4,0.0,0.4,0.0,0.4,0.4,0.6,0.8,0.6,0.6,,0.6,0.8,0.6,0.6,0.4,0.4,0.4,0.6,0.6,0.6,0.8,0.6,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.4,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.4,0.4,0.4,0.2,0.4,0.2,0.2,0.4,0.6,0.4,,0.2,0.6,0.8,0.6,0.6,0.8,0.8,0.6,0.8,0.8,1.0,0.8,0.8,0.8,0.6,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.4,0.2,0.4,0.2,0.4,0.2,0.2,0.0,0.2,0.4,0.2,0.8,,0.4,0.8,0.4,0.4,0.6,0.8,0.6,0.8,1.0,0.8,0.8,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.4,0.4,0.4,0.4,0.4,0.6,,0.4,0.6,0.6,0.6,0.8,0.8,0.8,0.8,0.6,0.8,0.8,0.8,0.8,0.8,1.0,1.0,0.8,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.2,0.2,0.0,0.2,0.4,0.4,0.4,0.2,0.4,0.2,0.2,0.4,0.6,0.4,0.2,0.2,0.6,,0.6,0.6,0.8,0.8,0.6,0.6,0.6,1.0,0.8,0.6,0.8,0.6,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.6,0.2,0.4,0.2,0.4,0.2,0.6,0.4,0.4,0.2,0.6,0.4,0.6,0.4,0.4,,1.0,0.8,0.6,0.6,0.4,0.6,0.4,0.6,0.8,0.6,1.0,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.6,0.2,0.4,0.2,0.4,0.2,0.6,0.4,0.4,0.2,0.6,0.4,0.6,0.4,0.4,0.0,,0.8,0.6,0.4,0.4,0.6,0.4,0.6,0.8,0.6,1.0,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.0,0.4,0.2,0.4,0.2,0.2,0.4,0.4,0.2,0.2,0.6,0.2,0.4,0.4,0.2,0.2,0.2,,0.6,0.2,0.2,0.4,0.6,0.8,0.4,0.8,0.6,0.6,0.8,0.8,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.2,0.2,0.2,0.4,0.4,0.4,,0.4,0.4,0.4,0.8,1.0,0.4,1.0,0.4,0.6,0.6,0.6,0.8,0.8,0.8,1.0,0.8,1.0,0.8,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.4,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.4,0.4,0.4,0.2,0.4,0.4,0.6,0.8,0.6,,0.4,0.6,0.6,0.6,0.8,0.6,0.6,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.4,0.2,0.0,0.2,0.2,0.2,0.4,0.2,0.4,0.2,0.2,0.2,0.4,0.6,0.6,0.8,0.6,0.6,,0.6,0.6,0.6,0.7,0.6,0.6,1.0,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.0,0.2,0.4,0.2,0.2,0.0,0.2,0.4,0.4,0.4,0.6,0.6,0.4,0.4,,0.8,0.8,0.4,0.8,0.4,0.8,0.6,0.8,0.8,0.8,1.0,1.0,0.8,1.0,1.0,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.4,0.2,0.4,0.0,0.4,0.2,0.2,0.2,0.2,0.4,0.0,0.2,0.4,0.0,0.6,0.6,0.4,0.2,0.4,0.4,0.2,,0.6,0.6,0.6,0.6,0.6,0.8,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.0,0.4,0.4,0.2,0.4,,0.4,0.8,0.4,0.4,0.6,0.6,0.6,0.8,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.4,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.4,0.2,0.2,0.6,0.6,0.2,0.3,0.6,0.4,0.6,,0.6,0.6,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+EBM (default),0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.2,0.4,0.4,0.2,0.0,0.4,0.4,0.2,0.4,0.2,0.4,,0.4,0.4,0.6,0.6,0.6,0.6,0.8,0.8,0.8,1.0,0.8,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.0,0.2,0.2,0.4,0.2,0.2,0.4,0.0,0.0,0.4,0.6,0.4,0.4,0.6,0.4,0.6,0.4,0.6,,0.8,0.6,0.6,0.8,0.8,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.4,0.4,0.0,0.0,0.2,0.4,0.6,0.2,0.6,0.2,,0.4,0.6,0.6,0.8,0.8,0.8,0.8,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.4,0.2,0.4,0.2,0.4,0.4,0.6,,0.6,0.4,1.0,0.8,0.8,0.6,0.8,1.0,1.0,1.0,1.0
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.2,0.2,0.4,0.2,0.2,0.2,0.2,0.4,0.2,0.4,0.4,0.4,0.4,,0.4,0.8,0.8,0.8,0.6,0.8,1.0,1.0,1.0,1.0
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.0,0.0,0.4,0.2,0.0,0.0,0.2,0.2,0.4,0.0,0.4,0.2,0.4,0.6,0.6,,0.6,0.6,0.6,1.0,0.6,1.0,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.2,0.0,0.0,0.2,0.2,0.2,0.0,0.4,0.2,0.2,0.0,0.2,0.4,,0.8,0.8,0.4,0.8,0.6,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.2,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.4,0.2,,0.8,0.6,1.0,0.6,1.0,1.0,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.2,0.0,0.2,0.2,0.2,0.2,0.2,0.4,0.2,0.2,,0.6,1.0,0.6,1.0,1.0,1.0
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.2,0.0,0.2,0.0,0.2,0.2,0.2,0.4,0.4,0.0,0.6,0.4,0.4,,0.4,1.0,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.2,0.4,0.2,0.0,0.0,0.6,,0.6,1.0,1.0,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.2,0.0,0.2,0.0,0.0,0.0,0.0,0.0,0.4,0.4,0.4,0.0,0.4,,0.8,0.8,0.8
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,,1.0,1.0
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,,0.8
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.2,0.0,0.2,
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/leaderboard_overview_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/leaderboard_overview_explorer.html
index 294243c16f7222169374eb135601fe1334221867..9adc940c755b9c9674ba074496675e8255ea179e 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/leaderboard_overview_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/leaderboard_overview_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena leaderboard explorer — Lite | Regression | Small</title>
+<title>TabArena leaderboard explorer — data/imputation_yes/splits_lite/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -332,6 +343,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -411,7 +439,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -448,7 +480,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1112,7 +1144,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${members.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, members.length);
       famBtn.title = `Toggle all ${members.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/leaderboard_table.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..efa6a29a19b55d179de123029388f3499ce969da
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data/imputation_yes/splits_lite/tasks_regression/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabFM","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/google-research\/tabfm","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2205.0,"score":0.896,"rank":3.5,"hrank":1.63,"imp":1.989,"train_time":56.29,"predict_time":4.031,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+308\/-71","imputed":false},{"position":1,"method":"Nori-30M","variant":"Default","family":"Foundation Model","url":"https:\/\/huggingface.co\/Synthefy\/Nori-30M","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2121.0,"score":0.795,"rank":4.75,"hrank":2.83,"imp":1.666,"train_time":14.41,"predict_time":2.084,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+297\/-68","imputed":false},{"position":2,"method":"Nori","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/Synthefy\/synthefy-nori","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":2049.0,"score":0.747,"rank":6.12,"hrank":4.38,"imp":2.977,"train_time":14.79,"predict_time":1.521,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+207\/-20","imputed":false},{"position":3,"method":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":2026.0,"score":0.701,"rank":6.62,"hrank":4.89,"imp":2.864,"train_time":323.76,"predict_time":2.906,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+174\/-49","imputed":false},{"position":4,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2605.13986","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1973.0,"score":0.66,"rank":7.88,"hrank":3.78,"imp":2.58,"train_time":14.15,"predict_time":0.487,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+346\/-116","imputed":false},{"position":5,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1950.0,"score":0.687,"rank":8.5,"hrank":4.68,"imp":2.876,"train_time":10.03,"predict_time":0.63,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+298\/-121","imputed":false},{"position":6,"method":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1886.0,"score":0.639,"rank":10.38,"hrank":5.57,"imp":2.432,"train_time":2356.36,"predict_time":7.546,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+396\/-198","imputed":false},{"position":7,"method":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1886.0,"score":0.621,"rank":10.38,"hrank":5.18,"imp":3.973,"train_time":9830.47,"predict_time":232.103,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+261\/-117","imputed":false},{"position":8,"method":"TabDPT","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1817.0,"score":0.566,"rank":12.75,"hrank":10.47,"imp":4.601,"train_time":9830.47,"predict_time":22.221,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+158\/-54","imputed":false},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2602.11139","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1717.0,"score":0.545,"rank":16.88,"hrank":5.1,"imp":4.91,"train_time":6.0,"predict_time":0.139,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+364\/-235","imputed":false},{"position":10,"method":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1698.0,"score":0.47,"rank":17.75,"hrank":9.17,"imp":3.882,"train_time":2356.36,"predict_time":0.878,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+338\/-180","imputed":false},{"position":11,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2410.18164","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1693.0,"score":0.436,"rank":18.0,"hrank":12.17,"imp":5.978,"train_time":51.28,"predict_time":43.062,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+187\/-87","imputed":false},{"position":12,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1660.0,"score":0.418,"rank":19.62,"hrank":17.79,"imp":6.227,"train_time":4918.07,"predict_time":35.587,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+118\/-54","imputed":false},{"position":13,"method":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","url":"https:\/\/openreview.net\/pdf?id=Y00pwFyrHR","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1646.0,"score":0.412,"rank":20.38,"hrank":17.03,"imp":6.135,"train_time":7.43,"predict_time":0.262,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+160\/-81","imputed":false},{"position":14,"method":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83d\udcca","verified":true,"elo":1646.0,"score":0.358,"rank":20.38,"hrank":17.21,"imp":6.878,"train_time":4921.21,"predict_time":5.366,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-100","imputed":false},{"position":15,"method":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.08667","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1643.0,"score":0.41,"rank":20.5,"hrank":12.93,"imp":5.422,"train_time":8.24,"predict_time":0.532,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+308\/-188","imputed":false},{"position":16,"method":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1637.0,"score":0.422,"rank":20.81,"hrank":9.52,"imp":4.431,"train_time":6437.84,"predict_time":56.08,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+517\/-236","imputed":true},{"position":17,"method":"TabSwift","variant":"Default","family":"Foundation Model","url":"https:\/\/github.com\/LAMDA-Tabular\/TabSwift","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1599.0,"score":0.363,"rank":22.88,"hrank":15.75,"imp":5.905,"train_time":4.08,"predict_time":0.065,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+179\/-107","imputed":false},{"position":18,"method":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1572.0,"score":0.267,"rank":24.38,"hrank":21.68,"imp":8.759,"train_time":39204.86,"predict_time":1759.104,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+122\/-84","imputed":false},{"position":19,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1545.0,"score":0.207,"rank":26.0,"hrank":23.61,"imp":9.446,"train_time":1206.7,"predict_time":4.571,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-90","imputed":false},{"position":20,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1534.0,"score":0.319,"rank":26.62,"hrank":21.33,"imp":7.309,"train_time":4918.07,"predict_time":0.949,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+159\/-128","imputed":false},{"position":21,"method":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1512.0,"score":0.299,"rank":28.0,"hrank":6.22,"imp":10.6,"train_time":9011.78,"predict_time":8.012,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+239\/-189","imputed":false},{"position":22,"method":"Mitra","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2510.21204","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1510.0,"score":0.27,"rank":28.06,"hrank":20.18,"imp":7.929,"train_time":94.16,"predict_time":2.176,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+197\/-151","imputed":true},{"position":23,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1510.0,"score":0.179,"rank":28.12,"hrank":25.09,"imp":8.776,"train_time":3127.07,"predict_time":0.459,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-76","imputed":false},{"position":24,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1510.0,"score":0.25,"rank":28.12,"hrank":24.67,"imp":7.219,"train_time":3814.24,"predict_time":4.368,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+167\/-114","imputed":false},{"position":25,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1506.0,"score":0.189,"rank":28.38,"hrank":25.7,"imp":8.735,"train_time":3127.07,"predict_time":0.057,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+154\/-82","imputed":false},{"position":26,"method":"xRFM","variant":"Tuned + Ens.","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1482.0,"score":0.231,"rank":29.88,"hrank":24.04,"imp":8.935,"train_time":1673.38,"predict_time":2.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+175\/-122","imputed":false},{"position":27,"method":"TabM","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1451.0,"score":0.203,"rank":31.88,"hrank":27.63,"imp":7.438,"train_time":3814.24,"predict_time":0.453,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+168\/-126","imputed":false},{"position":28,"method":"TabPFNv2","variant":"Tuned","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1448.0,"score":0.279,"rank":32.06,"hrank":20.02,"imp":6.857,"train_time":6437.84,"predict_time":0.557,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+383\/-242","imputed":true},{"position":29,"method":"xRFM","variant":"Tuned","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1441.0,"score":0.177,"rank":32.5,"hrank":27.26,"imp":9.224,"train_time":1673.38,"predict_time":0.097,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+126\/-85","imputed":false},{"position":30,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1438.0,"score":0.103,"rank":32.75,"hrank":30.65,"imp":9.921,"train_time":1277.57,"predict_time":0.647,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-70","imputed":false},{"position":31,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1430.0,"score":0.093,"rank":33.25,"hrank":31.96,"imp":9.992,"train_time":1277.57,"predict_time":3.744,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-61","imputed":false},{"position":32,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1409.0,"score":0.12,"rank":34.62,"hrank":31.35,"imp":10.618,"train_time":1206.7,"predict_time":0.689,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-109","imputed":false},{"position":33,"method":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1392.0,"score":0.095,"rank":35.75,"hrank":32.92,"imp":11.436,"train_time":1889.62,"predict_time":0.341,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+142\/-118","imputed":false},{"position":34,"method":"ModernNCA","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1387.0,"score":0.132,"rank":36.12,"hrank":32.05,"imp":12.993,"train_time":25.15,"predict_time":0.294,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+136\/-138","imputed":false},{"position":35,"method":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2506.10707","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1385.0,"score":0.173,"rank":36.25,"hrank":28.2,"imp":10.294,"train_time":25.28,"predict_time":0.842,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+177\/-163","imputed":false},{"position":36,"method":"iLTM","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1374.0,"score":0.125,"rank":37.0,"hrank":32.7,"imp":10.165,"train_time":39204.86,"predict_time":127.594,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+143\/-129","imputed":false},{"position":37,"method":"CatBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1706.09516","family_symbol":"\ud83c\udf33","verified":true,"elo":1370.0,"score":0.069,"rank":37.25,"hrank":34.12,"imp":11.735,"train_time":10.69,"predict_time":0.044,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+116\/-107","imputed":false},{"position":38,"method":"TabPFNv2","variant":"Default","family":"Foundation Model","url":"https:\/\/www.nature.com\/articles\/s41586-024-08328-6","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1367.0,"score":0.253,"rank":37.44,"hrank":22.95,"imp":8.703,"train_time":8.69,"predict_time":0.726,"imputed_pct":12.5,"hardware":"GPU","elo_ci":"+317\/-288","imputed":true},{"position":39,"method":"ChimeraBoost","variant":"Tuned","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1321.0,"score":0.06,"rank":40.5,"hrank":38.15,"imp":12.037,"train_time":1889.62,"predict_time":0.027,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+119\/-112","imputed":false},{"position":40,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1308.0,"score":0.077,"rank":41.38,"hrank":35.72,"imp":14.233,"train_time":1235.42,"predict_time":1.784,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+148\/-160","imputed":false},{"position":41,"method":"RealMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.04491","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1306.0,"score":0.097,"rank":41.5,"hrank":37.2,"imp":11.529,"train_time":22.42,"predict_time":7.815,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+127\/-102","imputed":false},{"position":42,"method":"LimiX","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2509.03505","family_symbol":"\ud83e\udde0\u26a1","verified":false,"elo":1302.0,"score":0.154,"rank":41.75,"hrank":15.47,"imp":11.915,"train_time":122.31,"predict_time":20.837,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+265\/-266","imputed":false},{"position":43,"method":"ModernNCA","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2407.03257","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1293.0,"score":0.108,"rank":42.38,"hrank":29.62,"imp":13.554,"train_time":9011.78,"predict_time":0.526,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+191\/-198","imputed":false},{"position":44,"method":"TabM","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2410.24210","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1291.0,"score":0.089,"rank":42.5,"hrank":39.41,"imp":10.016,"train_time":14.25,"predict_time":0.485,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+131\/-119","imputed":false},{"position":45,"method":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1284.0,"score":0.064,"rank":43.0,"hrank":39.33,"imp":11.386,"train_time":5196.37,"predict_time":1.749,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+121\/-140","imputed":false},{"position":46,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1274.0,"score":0.06,"rank":43.62,"hrank":38.44,"imp":14.37,"train_time":1235.42,"predict_time":0.317,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+138\/-164","imputed":false},{"position":47,"method":"ChimeraBoost","variant":"Default","family":"Tree-based","url":"https:\/\/github.com\/bbstats\/chimeraboost","family_symbol":"\ud83c\udf33","verified":true,"elo":1240.0,"score":0.021,"rank":45.88,"hrank":44.09,"imp":13.334,"train_time":9.1,"predict_time":0.024,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-102","imputed":false},{"position":48,"method":"TorchMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1236.0,"score":0.068,"rank":46.12,"hrank":41.41,"imp":11.704,"train_time":5196.37,"predict_time":0.121,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+164\/-174","imputed":false},{"position":49,"method":"EBM","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1221.0,"score":0.079,"rank":47.12,"hrank":34.9,"imp":14.489,"train_time":2809.99,"predict_time":0.115,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+198\/-196","imputed":false},{"position":50,"method":"EBM","variant":"Tuned","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1196.0,"score":0.1,"rank":48.75,"hrank":31.12,"imp":15.003,"train_time":2809.99,"predict_time":0.013,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+204\/-212","imputed":false},{"position":51,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1181.0,"score":0.011,"rank":49.62,"hrank":47.02,"imp":16.355,"train_time":657.99,"predict_time":1.134,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-175","imputed":false},{"position":52,"method":"xRFM","variant":"Default","family":"Other","url":"https:\/\/arxiv.org\/abs\/2508.10053","family_symbol":"\u2753","verified":true,"elo":1179.0,"score":0.096,"rank":49.75,"hrank":33.52,"imp":16.539,"train_time":7.0,"predict_time":1.362,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+200\/-276","imputed":false},{"position":53,"method":"LightGBM","variant":"Default","family":"Tree-based","url":"https:\/\/papers.nips.cc\/paper_files\/paper\/2017\/hash\/6449f44a102fde848669bdd9eb6b76fa-Abstract.html","family_symbol":"\ud83c\udf33","verified":true,"elo":1177.0,"score":0.014,"rank":49.88,"hrank":48.36,"imp":13.756,"train_time":4.54,"predict_time":0.374,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-116","imputed":false},{"position":54,"method":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1142.0,"score":0.006,"rank":52.0,"hrank":50.84,"imp":15.436,"train_time":2151.53,"predict_time":7.992,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-123","imputed":false},{"position":55,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1007\/s10994-006-6226-1","family_symbol":"\ud83c\udf33","verified":true,"elo":1131.0,"score":0.011,"rank":52.62,"hrank":48.65,"imp":17.046,"train_time":7.81,"predict_time":0.985,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+112\/-217","imputed":false},{"position":56,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1123.0,"score":0.011,"rank":53.12,"hrank":50.16,"imp":17.448,"train_time":657.99,"predict_time":0.147,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+110\/-215","imputed":false},{"position":57,"method":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1111.0,"score":0.0,"rank":53.81,"hrank":52.72,"imp":17.078,"train_time":208.16,"predict_time":0.453,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+74\/-188","imputed":false},{"position":58,"method":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1107.0,"score":0.108,"rank":54.0,"hrank":29.32,"imp":24.308,"train_time":96418.84,"predict_time":36.577,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+253\/-468","imputed":false},{"position":59,"method":"FastaiMLP","variant":"Tuned","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1105.0,"score":0.0,"rank":54.12,"hrank":52.41,"imp":15.864,"train_time":2151.53,"predict_time":0.566,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+91\/-148","imputed":false},{"position":60,"method":"PerpetualBooster","variant":"Tuned","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":1093.0,"score":0.0,"rank":54.81,"hrank":53.82,"imp":17.791,"train_time":208.16,"predict_time":0.135,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-210","imputed":false},{"position":61,"method":"XGBoost","variant":"Default","family":"Tree-based","url":"https:\/\/arxiv.org\/abs\/1603.02754","family_symbol":"\ud83c\udf33","verified":true,"elo":1092.0,"score":0.0,"rank":54.88,"hrank":53.77,"imp":15.016,"train_time":4.08,"predict_time":0.255,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-124","imputed":false},{"position":62,"method":"TabSTAR","variant":"Tuned","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1059.0,"score":0.086,"rank":56.62,"hrank":39.07,"imp":24.484,"train_time":96418.84,"predict_time":5.523,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+248\/-432","imputed":false},{"position":63,"method":"EBM","variant":"Default","family":"Tree-based","url":"https:\/\/www.cs.cornell.edu\/~yinlou\/papers\/lou-kdd13.pdf","family_symbol":"\ud83c\udf33","verified":true,"elo":1045.0,"score":0.0,"rank":57.38,"hrank":55.89,"imp":16.495,"train_time":11.1,"predict_time":0.014,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+97\/-149","imputed":false},{"position":64,"method":"TorchMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1020.0,"score":0.0,"rank":58.62,"hrank":56.27,"imp":15.67,"train_time":22.25,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+125\/-196","imputed":false},{"position":65,"method":"RandomForest","variant":"Default","family":"Tree-based","url":"https:\/\/link.springer.com\/article\/10.1023\/A:1010933404324","family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.0,"rank":59.62,"hrank":58.69,"imp":18.86,"train_time":0.69,"predict_time":0.125,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+87\/-200","imputed":false},{"position":66,"method":"FastaiMLP","variant":"Default","family":"Neural Network","url":"https:\/\/arxiv.org\/abs\/2003.06505","family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":968.0,"score":0.0,"rank":61.12,"hrank":59.82,"imp":20.309,"train_time":6.77,"predict_time":0.511,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+96\/-229","imputed":false},{"position":67,"method":"KNN","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":898.0,"score":0.0,"rank":64.0,"hrank":62.44,"imp":22.358,"train_time":88.79,"predict_time":0.571,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-302","imputed":false},{"position":68,"method":"TabSTAR","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2505.18125","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":898.0,"score":0.041,"rank":64.0,"hrank":54.76,"imp":27.524,"train_time":455.0,"predict_time":5.747,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+255\/-483","imputed":false},{"position":69,"method":"PerpetualBooster","variant":"Default","family":"Tree-based","url":"https:\/\/perpetual-ml.com\/","family_symbol":"\ud83c\udf33","verified":true,"elo":885.0,"score":0.0,"rank":64.5,"hrank":63.82,"imp":21.025,"train_time":34.18,"predict_time":0.02,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-194","imputed":false},{"position":70,"method":"iLTM","variant":"Default","family":"Foundation Model","url":"https:\/\/arxiv.org\/abs\/2511.15941","family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":839.0,"score":0.0,"rank":66.12,"hrank":65.8,"imp":20.975,"train_time":691.28,"predict_time":270.094,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+46\/-195","imputed":false},{"position":71,"method":"KNN","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":755.0,"score":0.0,"rank":68.62,"hrank":68.13,"imp":24.318,"train_time":88.79,"predict_time":0.052,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-300","imputed":false},{"position":72,"method":"KNN","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/neighbors.html","family_symbol":"\ud83d\udccf","verified":true,"elo":698.0,"score":0.0,"rank":70.0,"hrank":68.98,"imp":30.565,"train_time":0.43,"predict_time":0.047,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+162\/-419","imputed":false},{"position":73,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":568.0,"score":0.0,"rank":72.44,"hrank":72.22,"imp":34.982,"train_time":743.23,"predict_time":0.364,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-498","imputed":false},{"position":74,"method":"Linear","variant":"Tuned","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":543.0,"score":0.0,"rank":72.81,"hrank":72.6,"imp":35.064,"train_time":743.23,"predict_time":0.097,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-513","imputed":false},{"position":75,"method":"Linear","variant":"Default","family":"Baseline","url":"https:\/\/scikit-learn.org\/stable\/modules\/linear_model.html","family_symbol":"\ud83d\udccf","verified":true,"elo":401.0,"score":0.0,"rank":74.5,"hrank":74.48,"imp":37.174,"train_time":3.98,"predict_time":0.12,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+105\/-472","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/pareto_front_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/pareto_front_explorer.html
index 576347d3f285c7a615f8d4f504e5c15a9eb457a3..d97239c2c751c0328461be152f78daa005b58083 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/pareto_front_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/pareto_front_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena Pareto explorer</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}], "chipsSide": "left"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0440469671,"imp":11.7347505258,"elo":1369.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0570904815,"imp":8.7352040009,"elo":1505.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4587675243,"imp":8.7756000462,"elo":1509.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240241152,"imp":13.333827481,"elo":1240.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0270025453,"imp":12.0366317858,"elo":1321.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3406062626,"imp":11.4363945187,"elo":1392.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0136940061,"imp":16.4950841745,"elo":1045.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0126861072,"imp":15.002806027,"elo":1195.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1149138914,"imp":14.4888973668,"elo":1221.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9846715044,"imp":17.0456198454,"elo":1131.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3172224028,"imp":14.3701049802,"elo":1274.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7842242231,"imp":14.2330940967,"elo":1307.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5109617193,"imp":20.3085539952,"elo":967.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5659855728,"imp":15.8642121256,"elo":1105.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.9916224068,"imp":15.4359501903,"elo":1142.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.3736048612,"imp":13.7557845488,"elo":1177.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.6888310529,"imp":10.6180742868,"elo":1409.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.5709782007,"imp":9.4462953744,"elo":1544.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":20.8369258593,"imp":11.9145923506,"elo":1302.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.1759075978,"imp":7.9294728449,"elo":1510.5,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.29377362,"imp":12.992919227,"elo":1386.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5261711988,"imp":13.5544773089,"elo":1293.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0117732516,"imp":10.6000565736,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.5206234824,"imp":2.9765050728,"elo":2048.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0843698074,"imp":1.6658940146,"elo":2121.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0198449839,"imp":21.0247735369,"elo":885.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1347029659,"imp":17.7906421375,"elo":1092.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4528171238,"imp":17.0784649589,"elo":1110.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1250003559,"imp":18.8602531123,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468843405,"imp":17.4482814124,"elo":1122.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1335820465,"imp":16.3546864638,"elo":1181.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.8151845047,"imp":11.528554758,"elo":1306.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.9488198226,"imp":7.3092723649,"elo":1534.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":35.5866553094,"imp":6.2265792315,"elo":1660.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5321004862,"imp":5.4223934042,"elo":1643.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8781008023,"imp":3.8824639923,"elo":1698.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5456187851,"imp":2.4315310151,"elo":1886.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":0.8423700735,"imp":10.2939380886,"elo":1384.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.0616958015,"imp":5.978395454,"elo":1693.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":22.2214037327,"imp":4.6013925928,"elo":1816.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":232.1033993492,"imp":3.9730944558,"elo":1886.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2617923617,"imp":6.1350917842,"elo":1645.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":4.0307404894,"imp":1.9886645249,"elo":2204.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1387839993,"imp":4.909915297,"elo":1717.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4849712473,"imp":10.0160530863,"elo":1291.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4533957018,"imp":7.4375603739,"elo":1450.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.367582289,"imp":7.2190453621,"elo":1509.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6295234248,"imp":2.8755315855,"elo":1949.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4873524134,"imp":2.5796546204,"elo":1973.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.7259204826,"imp":8.7025817614,"elo":1367.1,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5570602208,"imp":6.8568053626,"elo":1448.0,"imputed":true,"imputed_pct":12.5,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":56.0801957423,"imp":4.4305153143,"elo":1637.3,"imputed":true,"imputed_pct":12.5,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7468382623,"imp":27.5237970384,"elo":898.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5231948052,"imp":24.4843446519,"elo":1059.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.5771589943,"imp":24.3082187083,"elo":1107.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0653075084,"imp":5.9050372269,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305890866,"imp":15.6704262361,"elo":1020.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1207478083,"imp":11.7041014307,"elo":1236.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.7488211618,"imp":11.3859797979,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2554761956,"imp":15.0160376154,"elo":1091.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6471298159,"imp":9.9210562234,"elo":1437.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.7439813565,"imp":9.9915972128,"elo":1430.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":270.0943042458,"imp":20.9754349164,"elo":838.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":127.5938263768,"imp":10.1651955473,"elo":1373.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1759.1043245943,"imp":8.7593025522,"elo":1572.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.3621042316,"imp":16.5392791726,"elo":1179.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0968773188,"imp":9.2240465758,"elo":1441.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.446364541,"imp":8.9348440593,"elo":1481.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
+  const CONFIG = {"mode": "scatter", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0440469671,"x_train":10.6906118793,"imp":11.7347505258,"elo":1369.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"CatBoost","family":"Tree-based","x_infer":0.0570904815,"x_train":3127.0650185113,"imp":8.7352040009,"elo":1505.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"CatBoost","family":"Tree-based","x_infer":0.4587675243,"x_train":3127.0650185113,"imp":8.7756000462,"elo":1509.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240241152,"x_train":9.0951433317,"imp":13.333827481,"elo":1240.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0270025453,"x_train":1889.6193046203,"imp":12.0366317858,"elo":1321.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3406062626,"x_train":1889.6193046203,"imp":11.4363945187,"elo":1392.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"EBM","family":"Tree-based","x_infer":0.0136940061,"x_train":11.1028805965,"imp":16.4950841745,"elo":1045.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"EBM","family":"Tree-based","x_infer":0.0126861072,"x_train":2809.98545611,"imp":15.002806027,"elo":1195.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"EBM","family":"Tree-based","x_infer":0.1149138914,"x_train":2809.98545611,"imp":14.4888973668,"elo":1221.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9846715044,"x_train":7.8095418615,"imp":17.0456198454,"elo":1131.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.3172224028,"x_train":1235.4175263307,"imp":14.3701049802,"elo":1274.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7842242231,"x_train":1235.4175263307,"imp":14.2330940967,"elo":1307.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5109617193,"x_train":6.7731104239,"imp":20.3085539952,"elo":967.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5659855728,"x_train":2151.5332639388,"imp":15.8642121256,"elo":1105.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.9916224068,"x_train":2151.5332639388,"imp":15.4359501903,"elo":1142.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LightGBM","family":"Tree-based","x_infer":0.3736048612,"x_train":4.544665245,"imp":13.7557845488,"elo":1177.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"LightGBM","family":"Tree-based","x_infer":0.6888310529,"x_train":1206.7027545763,"imp":10.6180742868,"elo":1409.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"LightGBM","family":"Tree-based","x_infer":4.5709782007,"x_train":1206.7027545763,"imp":9.4462953744,"elo":1544.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"LimiX","family":"Foundation Model","x_infer":20.8369258593,"x_train":122.3127076606,"imp":11.9145923506,"elo":1302.3,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Mitra","family":"Foundation Model","x_infer":2.1759075978,"x_train":94.1603678675,"imp":7.9294728449,"elo":1510.5,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.29377362,"x_train":25.1467163069,"imp":12.992919227,"elo":1386.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"ModernNCA","family":"Neural Network","x_infer":0.5261711988,"x_train":9011.7804615398,"imp":13.5544773089,"elo":1293.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0117732516,"x_train":9011.7804615398,"imp":10.6000565736,"elo":1511.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"Nori","family":"Foundation Model","x_infer":1.5206234824,"x_train":14.7870051149,"imp":2.9765050728,"elo":2048.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0843698074,"x_train":14.4142036067,"imp":1.6658940146,"elo":2121.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0198449839,"x_train":34.1835619211,"imp":21.0247735369,"elo":885.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.1347029659,"x_train":208.1564418277,"imp":17.7906421375,"elo":1092.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4528171238,"x_train":208.1564418277,"imp":17.0784649589,"elo":1110.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RandomForest","family":"Tree-based","x_infer":0.1250003559,"x_train":0.6913103896,"imp":18.8602531123,"elo":1000.0,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RandomForest","family":"Tree-based","x_infer":0.1468843405,"x_train":657.9874551063,"imp":17.4482814124,"elo":1122.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RandomForest","family":"Tree-based","x_infer":1.1335820465,"x_train":657.9874551063,"imp":16.3546864638,"elo":1181.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealMLP","family":"Neural Network","x_infer":7.8151845047,"x_train":22.4158785265,"imp":11.528554758,"elo":1306.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealMLP","family":"Neural Network","x_infer":0.9488198226,"x_train":4918.0723002043,"imp":7.3092723649,"elo":1534.1,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealMLP","family":"Neural Network","x_infer":35.5866553094,"x_train":4918.0723002043,"imp":6.2265792315,"elo":1660.2,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5321004862,"x_train":8.242142907,"imp":5.4223934042,"elo":1643.2,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.8781008023,"x_train":2356.3627464618,"imp":3.8824639923,"elo":1698.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5456187851,"x_train":2356.3627464618,"imp":2.4315310151,"elo":1886.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":0.8423700735,"x_train":25.2809733305,"imp":10.2939380886,"elo":1384.9,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":43.0616958015,"x_train":51.2843993162,"imp":5.978395454,"elo":1693.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabDPT","family":"Foundation Model","x_infer":22.2214037327,"x_train":9830.4728139885,"imp":4.6013925928,"elo":1816.9,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabDPT","family":"Foundation Model","x_infer":232.1033993492,"x_train":9830.4728139885,"imp":3.9730944558,"elo":1886.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2617923617,"x_train":7.4300775989,"imp":6.1350917842,"elo":1645.6,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabFM","family":"Foundation Model","x_infer":4.0307404894,"x_train":56.2879369055,"imp":1.9886645249,"elo":2204.8,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1387839993,"x_train":5.9977774395,"imp":4.909915297,"elo":1717.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4849712473,"x_train":14.2468083815,"imp":10.0160530863,"elo":1291.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabM","family":"Neural Network","x_infer":0.4533957018,"x_train":3814.2351446507,"imp":7.4375603739,"elo":1450.8,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabM","family":"Neural Network","x_infer":4.367582289,"x_train":3814.2351446507,"imp":7.2190453621,"elo":1509.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6295234248,"x_train":10.02817758,"imp":2.8755315855,"elo":1949.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4873524134,"x_train":14.1514284946,"imp":2.5796546204,"elo":1973.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.7259204826,"x_train":8.6865743952,"imp":8.7025817614,"elo":1367.1,"imputed":true,"imputed_pct":12.5,"variant":"Default"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.5570602208,"x_train":6437.8441109111,"imp":6.8568053626,"elo":1448.0,"imputed":true,"imputed_pct":12.5,"variant":"Tuned"},{"method":"TabPFNv2","family":"Foundation Model","x_infer":56.0801957423,"x_train":6437.8441109111,"imp":4.4305153143,"elo":1637.3,"imputed":true,"imputed_pct":12.5,"variant":"Tuned + Ens."},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7468382623,"x_train":454.9991027301,"imp":27.5237970384,"elo":898.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.5231948052,"x_train":96418.8447274511,"imp":24.4843446519,"elo":1059.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.5771589943,"x_train":96418.8447274511,"imp":24.3082187083,"elo":1107.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0653075084,"x_train":4.0841565985,"imp":5.9050372269,"elo":1599.1,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305890866,"x_train":22.2534811738,"imp":15.6704262361,"elo":1020.5,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1207478083,"x_train":5196.3714196423,"imp":11.7041014307,"elo":1236.4,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"TorchMLP","family":"Neural Network","x_infer":1.7488211618,"x_train":5196.3714196423,"imp":11.3859797979,"elo":1283.6,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"XGBoost","family":"Tree-based","x_infer":0.2554761956,"x_train":4.0843435263,"imp":15.0160376154,"elo":1091.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"XGBoost","family":"Tree-based","x_infer":0.6471298159,"x_train":1277.5662482217,"imp":9.9210562234,"elo":1437.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"XGBoost","family":"Tree-based","x_infer":3.7439813565,"x_train":1277.5662482217,"imp":9.9915972128,"elo":1430.0,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"iLTM","family":"Foundation Model","x_infer":270.0943042458,"x_train":691.2819503211,"imp":20.9754349164,"elo":838.7,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"iLTM","family":"Foundation Model","x_infer":127.5938263768,"x_train":39204.8641933754,"imp":10.1651955473,"elo":1373.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"iLTM","family":"Foundation Model","x_infer":1759.1043245943,"x_train":39204.8641933754,"imp":8.7593025522,"elo":1572.5,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."},{"method":"xRFM","family":"Other","x_infer":1.3621042316,"x_train":7.002585411,"imp":16.5392791726,"elo":1179.4,"imputed":false,"imputed_pct":0.0,"variant":"Default"},{"method":"xRFM","family":"Other","x_infer":0.0968773188,"x_train":1673.3760165809,"imp":9.2240465758,"elo":1441.3,"imputed":false,"imputed_pct":0.0,"variant":"Tuned"},{"method":"xRFM","family":"Other","x_infer":2.446364541,"x_train":1673.3760165809,"imp":8.9348440593,"elo":1481.7,"imputed":false,"imputed_pct":0.0,"variant":"Tuned + Ens."}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/tuning_trajectories_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/tuning_trajectories_explorer.html
index d6b48e6e8a2a021e736ab8abb0193a23922c7da9..0f0ce9b2b6d48ce601ec19fc10911a6a26810bb2 100644
--- a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/tuning_trajectories_explorer.html
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/tuning_trajectories_explorer.html
@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1">
-<title>TabArena tuning trajectories</title>
+<title>TabArena — data/imputation_yes/splits_lite/tasks_regression/datasets_small</title>
 <style>
   :root {
         --fam-foundation: #b07cf0;
@@ -126,9 +126,15 @@
     color-scheme: light;
   }
   html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
   body {
     color: var(--ink);
-    font-family: system-ui, -apple-system, "Segoe UI", sans-serif;
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
     line-height: 1.5;
     padding: 10px 12px 14px;
   }
@@ -167,6 +173,8 @@
   }
   .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
   .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
   .famchip:hover { border-color: var(--fam); color: var(--ink); }
   .famchip[aria-pressed="true"] {
     border: 1px solid var(--fam);
@@ -212,7 +220,10 @@
   }
   details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
 
-  svg text { font-family: system-ui, -apple-system, "Segoe UI", sans-serif; }
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
 
   @media (prefers-reduced-motion: no-preference) {
     .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
@@ -305,8 +316,8 @@
 <script>
 (function () {
   "use strict";
-  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "right"};
-  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_train":10.6906118793,"imp":11.7380678406,"elo":1383.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_train":27.5168414441,"imp":10.8787795601,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_train":74.1773887355,"imp":9.8956020481,"elo":1511.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_train":149.7230535562,"imp":9.4539070634,"elo":1529.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_train":390.8454255101,"imp":9.0355787916,"elo":1540.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_train":774.7646676657,"imp":8.9126650851,"elo":1539.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_train":1549.9544564854,"imp":8.8482083003,"elo":1535.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_train":2319.2427277232,"imp":8.8009269607,"elo":1539.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_train":3127.0650185113,"imp":8.779742505,"elo":1538.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":9.0951433317,"imp":13.3368408494,"elo":1254.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":18.0665913735,"imp":12.7290405653,"elo":1292.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":47.3379775935,"imp":12.2230650412,"elo":1348.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":95.7569831536,"imp":12.0284416083,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":237.2122715241,"imp":11.8246977814,"elo":1375.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":473.9846968668,"imp":11.7071831591,"elo":1387.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":942.9066009657,"imp":11.5364625858,"elo":1399.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1412.1394987216,"imp":11.4668809952,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_train":1889.6193046203,"imp":11.4395445262,"elo":1416.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_train":11.1028805965,"imp":16.4976379141,"elo":1061.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_train":29.3949504722,"imp":15.8588828829,"elo":1116.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_train":76.7487383079,"imp":15.2784837429,"elo":1197.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_train":146.1996217178,"imp":15.0041573064,"elo":1213.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_train":363.6322234704,"imp":14.7544531085,"elo":1228.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_train":719.7922921179,"imp":14.6819004949,"elo":1228.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_train":1405.778816208,"imp":14.5966335853,"elo":1237.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_train":2088.5183356085,"imp":14.5352462665,"elo":1237.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_train":2809.98545611,"imp":14.4895382411,"elo":1242.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_train":7.8095418615,"imp":17.0483421261,"elo":1138.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_train":14.6268296558,"imp":16.5555629996,"elo":1171.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_train":34.3331356985,"imp":15.6738655468,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_train":62.7079095608,"imp":15.1857714271,"elo":1287.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_train":155.029134045,"imp":14.4542681877,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_train":307.6069593637,"imp":14.3040260216,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_train":613.0063601899,"imp":14.2337102267,"elo":1335.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_train":922.5048472624,"imp":14.2288971439,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_train":1235.4175263307,"imp":14.2362430223,"elo":1328.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_train":6.7731104239,"imp":20.3106713122,"elo":960.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_train":17.0044463073,"imp":19.9496894914,"elo":975.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_train":50.1505869221,"imp":18.7437585272,"elo":1025.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_train":101.3841054234,"imp":18.1779604669,"elo":1041.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_train":260.7094513648,"imp":17.2940660418,"elo":1081.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_train":528.4734813547,"imp":16.7765138581,"elo":1110.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1059.681319286,"imp":16.1672923144,"elo":1131.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_train":1596.5291960327,"imp":15.7406048434,"elo":1150.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_train":2151.5332639388,"imp":15.4390915666,"elo":1160.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_train":4.544665245,"imp":13.7590818546,"elo":1186.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_train":10.636770614,"imp":12.476047148,"elo":1280.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_train":27.7973506573,"imp":11.1844991259,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_train":56.861966814,"imp":10.6108589021,"elo":1455.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_train":148.4745666134,"imp":10.0876299898,"elo":1519.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_train":299.3928042834,"imp":9.9288733954,"elo":1544.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_train":598.0451468439,"imp":9.7033578519,"elo":1566.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_train":893.892871219,"imp":9.5698709534,"elo":1579.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_train":1206.7027545763,"imp":9.4499487541,"elo":1591.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_train":122.3127076606,"imp":11.9179021798,"elo":1340.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_train":94.1603678675,"imp":7.9340948995,"elo":1539.4,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":25.1467163069,"imp":12.9962863151,"elo":1420.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_train":75.362933136,"imp":12.3718490645,"elo":1446.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_train":210.5456172414,"imp":11.6267396007,"elo":1494.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_train":424.8014044984,"imp":11.3081905363,"elo":1525.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_train":1033.2681066658,"imp":11.0747240533,"elo":1531.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_train":2222.1208631991,"imp":10.8935157408,"elo":1530.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_train":4429.3203715947,"imp":10.7520103908,"elo":1529.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_train":6624.9729135871,"imp":10.6295149838,"elo":1539.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_train":9011.7804615398,"imp":10.6037965299,"elo":1540.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_train":14.7870051149,"imp":2.9817681293,"elo":2068.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_train":14.4142036067,"imp":1.6715446805,"elo":2118.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":34.1835619211,"imp":21.0267397864,"elo":908.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":68.7178569705,"imp":18.9265958956,"elo":1014.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":173.2458309649,"imp":17.0684828461,"elo":1103.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_train":208.1564418277,"imp":17.0805060316,"elo":1113.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_train":0.6913103896,"imp":18.8630134196,"elo":1001.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_train":4.4537617156,"imp":17.8387549123,"elo":1071.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_train":14.301129027,"imp":17.2622565173,"elo":1119.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_train":30.686422522,"imp":16.8003854792,"elo":1152.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_train":79.9124717904,"imp":16.5519192401,"elo":1169.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_train":162.1056028409,"imp":16.4538884455,"elo":1180.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_train":326.1499058044,"imp":16.3873501449,"elo":1191.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_train":490.3882635115,"imp":16.3633492171,"elo":1196.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_train":657.9874551063,"imp":16.3574828558,"elo":1199.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_train":22.4158785265,"imp":11.5329003355,"elo":1323.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_train":44.4251076686,"imp":9.5741826539,"elo":1409.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_train":117.3230828329,"imp":7.6359018563,"elo":1562.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_train":253.2768449964,"imp":6.9578682152,"elo":1600.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_train":616.4798700039,"imp":6.5714223235,"elo":1638.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_train":1238.3164021687,"imp":6.339582604,"elo":1678.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_train":2457.5657747181,"imp":6.2626565171,"elo":1690.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_train":3675.6638274823,"imp":6.2412663581,"elo":1695.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_train":4918.0723002043,"imp":6.2282511886,"elo":1698.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":8.242142907,"imp":5.427403104,"elo":1664.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":19.8358235954,"imp":4.7261793707,"elo":1701.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":56.6252728837,"imp":3.5987206349,"elo":1782.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":113.5361347503,"imp":3.0747732047,"elo":1849.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":295.9584460071,"imp":2.5468910913,"elo":1878.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":584.1451401972,"imp":2.3895569087,"elo":1902.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1169.0460603376,"imp":2.3310758771,"elo":1922.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":1760.7500065581,"imp":2.4258915729,"elo":1911.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_train":2356.3627464618,"imp":2.4360173334,"elo":1914.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_train":25.2809733305,"imp":10.2979773772,"elo":1413.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":51.2843993162,"imp":5.9831587418,"elo":1730.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_train":95.8813700743,"imp":5.3971294242,"elo":1763.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_train":254.4393954229,"imp":4.8561409198,"elo":1795.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_train":502.170242762,"imp":4.4172900231,"elo":1849.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_train":1229.6581613295,"imp":4.1595602054,"elo":1877.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_train":2450.1429629138,"imp":4.1026352104,"elo":1881.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_train":4887.1069852066,"imp":4.028961281,"elo":1893.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_train":7331.038512164,"imp":3.9899091528,"elo":1912.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_train":9830.4728139885,"imp":3.9787406493,"elo":1918.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_train":7.4300775989,"imp":6.1398219236,"elo":1682.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_train":56.2879369055,"imp":1.9939435026,"elo":2193.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_train":5.9977774395,"imp":4.914734588,"elo":1739.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":14.2468083815,"imp":10.0205504104,"elo":1305.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_train":34.9108464136,"imp":9.4576657254,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_train":89.6967770714,"imp":8.8810065831,"elo":1382.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_train":190.6759613979,"imp":8.3070618065,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_train":456.3526262514,"imp":7.8318467474,"elo":1480.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_train":918.9449885462,"imp":7.4798691525,"elo":1524.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_train":1845.4338559674,"imp":7.2737826281,"elo":1547.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_train":2797.5928059,"imp":7.2496835275,"elo":1546.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_train":3814.2351446507,"imp":7.2240501445,"elo":1539.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_train":10.02817758,"imp":2.8808212181,"elo":1982.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_train":14.1514284946,"imp":2.5953010295,"elo":1996.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":8.6865743952,"imp":8.7076340469,"elo":1384.4,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":52.109381455,"imp":7.4035677786,"elo":1443.3,"imputed":true,"imputed_pct":12.5,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":144.9748468042,"imp":6.2169828319,"elo":1516.6,"imputed":true,"imputed_pct":12.5,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":312.7874942852,"imp":5.7796838131,"elo":1545.1,"imputed":true,"imputed_pct":12.5,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":754.6626382647,"imp":5.0950859152,"elo":1608.2,"imputed":true,"imputed_pct":12.5,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":1536.077969506,"imp":4.8454469305,"elo":1626.3,"imputed":true,"imputed_pct":12.5,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":3126.6373952668,"imp":4.5450092452,"elo":1643.7,"imputed":true,"imputed_pct":12.5,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":4723.4389861822,"imp":4.4911277046,"elo":1652.5,"imputed":true,"imputed_pct":12.5,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_train":6437.8441109111,"imp":4.4358649388,"elo":1657.0,"imputed":true,"imputed_pct":12.5,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_train":454.9991027301,"imp":27.5261770168,"elo":907.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_train":923.2022669306,"imp":27.3713103765,"elo":927.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_train":2264.5016798126,"imp":26.9693598956,"elo":969.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_train":4340.8142220335,"imp":26.2782538996,"elo":1017.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_train":11688.77778151,"imp":25.4429386552,"elo":1063.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_train":23707.0922342587,"imp":24.9021077683,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_train":47472.2129899786,"imp":24.5530137841,"elo":1123.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_train":71823.9094609453,"imp":24.4153443067,"elo":1132.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_train":96418.8447274511,"imp":24.3105081894,"elo":1136.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_train":4.0841565985,"imp":5.9100274158,"elo":1620.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":22.2534811738,"imp":15.6737475478,"elo":1038.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_train":54.2457268049,"imp":15.1761073979,"elo":1075.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_train":127.8066374424,"imp":14.1453339402,"elo":1146.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_train":271.1084504473,"imp":13.71982596,"elo":1175.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_train":648.9159658958,"imp":12.9816291054,"elo":1214.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_train":1311.0552340778,"imp":12.4498845916,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_train":2566.6969558991,"imp":11.8624617347,"elo":1275.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_train":3875.6410639922,"imp":11.5964446071,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_train":5196.3714196423,"imp":11.3898917946,"elo":1311.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_train":4.0843435263,"imp":15.0194661753,"elo":1096.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_train":10.4789410396,"imp":12.2265999012,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_train":29.8024094892,"imp":11.2783006114,"elo":1360.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_train":60.219465986,"imp":10.7625899279,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_train":157.148713481,"imp":10.397729215,"elo":1444.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_train":316.488046922,"imp":10.2401435688,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_train":634.705809045,"imp":10.1282155644,"elo":1465.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_train":952.8874450538,"imp":10.0298489274,"elo":1466.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_train":1277.5662482217,"imp":9.9954828299,"elo":1466.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_train":691.2819503211,"imp":20.9779761968,"elo":866.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_train":2116.0353219536,"imp":15.7245100978,"elo":1107.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_train":6702.4310187379,"imp":11.1994420107,"elo":1363.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_train":15189.3088063979,"imp":9.9294564808,"elo":1470.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_train":37312.4085130914,"imp":8.7787641694,"elo":1607.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_train":39204.8641933754,"imp":8.7636936868,"elo":1606.7,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_train":7.002585411,"imp":16.5424125465,"elo":1203.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_train":15.1443119472,"imp":14.8045524272,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_train":39.7770511549,"imp":12.5667233788,"elo":1341.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_train":81.229812475,"imp":11.0712301058,"elo":1407.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_train":207.1643683575,"imp":9.7424442267,"elo":1465.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_train":413.5799198557,"imp":9.3688825164,"elo":1500.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_train":828.216131447,"imp":9.0618942855,"elo":1513.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_train":1248.051456359,"imp":8.9494988014,"elo":1524.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_train":1673.3760165809,"imp":8.9390070431,"elo":1521.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
+  const CONFIG = {"mode": "trajectory", "title": null, "metrics": [{"key": "imp", "label": "Improvability (%)", "axisLabel": "Improvability (%) \u2014 lower is better", "lowerBetter": true, "fromZero": true, "decimals": 1, "suffix": "%"}, {"key": "elo", "label": "Elo", "axisLabel": "Elo \u2014 higher is better", "lowerBetter": false, "fromZero": false, "decimals": 0, "suffix": ""}], "xAxes": [{"key": "x_infer", "label": "Inference time", "axisLabel": "Inference time per 1K samples (s), median \u2014 log scale", "short": "Inference (s/1K, median)"}, {"key": "x_train", "label": "Train time", "axisLabel": "Train time per 1K samples (s), median \u2014 log scale", "short": "Train (s/1K, median)"}], "chipsSide": "left"};
+  const POINTS = [{"method":"CatBoost","family":"Tree-based","x_infer":0.0440469671,"x_train":10.6906118793,"imp":11.7380678406,"elo":1383.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.094198463,"x_train":27.5168414441,"imp":10.8787795601,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2094177891,"x_train":74.1773887355,"imp":9.8956020481,"elo":1511.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.2708311702,"x_train":149.7230535562,"imp":9.4539070634,"elo":1529.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3286420785,"x_train":390.8454255101,"imp":9.0355787916,"elo":1540.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.3693540345,"x_train":774.7646676657,"imp":8.9126650851,"elo":1539.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4160109308,"x_train":1549.9544564854,"imp":8.8482083003,"elo":1535.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4476529792,"x_train":2319.2427277232,"imp":8.8009269607,"elo":1539.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"CatBoost","family":"Tree-based","x_infer":0.4587675243,"x_train":3127.0650185113,"imp":8.779742505,"elo":1538.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0240241152,"x_train":9.0951433317,"imp":13.3368408494,"elo":1254.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.0530471399,"x_train":18.0665913735,"imp":12.7290405653,"elo":1292.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1229635811,"x_train":47.3379775935,"imp":12.2230650412,"elo":1348.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.1719705654,"x_train":95.7569831536,"imp":12.0284416083,"elo":1365.3,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2033909926,"x_train":237.2122715241,"imp":11.8246977814,"elo":1375.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2377216613,"x_train":473.9846968668,"imp":11.7071831591,"elo":1387.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2621009173,"x_train":942.9066009657,"imp":11.5364625858,"elo":1399.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.2795215411,"x_train":1412.1394987216,"imp":11.4668809952,"elo":1408.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ChimeraBoost","family":"Tree-based","x_infer":0.3406062626,"x_train":1889.6193046203,"imp":11.4395445262,"elo":1416.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"EBM","family":"Tree-based","x_infer":0.0136940061,"x_train":11.1028805965,"imp":16.4976379141,"elo":1061.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"EBM","family":"Tree-based","x_infer":0.0256054927,"x_train":29.3949504722,"imp":15.8588828829,"elo":1116.7,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"EBM","family":"Tree-based","x_infer":0.0485028448,"x_train":76.7487383079,"imp":15.2784837429,"elo":1197.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"EBM","family":"Tree-based","x_infer":0.0670431171,"x_train":146.1996217178,"imp":15.0041573064,"elo":1213.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"EBM","family":"Tree-based","x_infer":0.0946419883,"x_train":363.6322234704,"imp":14.7544531085,"elo":1228.8,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"EBM","family":"Tree-based","x_infer":0.10646037,"x_train":719.7922921179,"imp":14.6819004949,"elo":1228.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"EBM","family":"Tree-based","x_infer":0.1141653132,"x_train":1405.778816208,"imp":14.5966335853,"elo":1237.1,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"EBM","family":"Tree-based","x_infer":0.1073495676,"x_train":2088.5183356085,"imp":14.5352462665,"elo":1237.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"EBM","family":"Tree-based","x_infer":0.1149138914,"x_train":2809.98545611,"imp":14.4895382411,"elo":1242.1,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.9846715044,"x_train":7.8095418615,"imp":17.0483421261,"elo":1138.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.1163998088,"x_train":14.6268296558,"imp":16.5555629996,"elo":1171.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0229922831,"x_train":34.3331356985,"imp":15.6738655468,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":0.8472601246,"x_train":62.7079095608,"imp":15.1857714271,"elo":1287.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.0307532617,"x_train":155.029134045,"imp":14.4542681877,"elo":1315.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.3416498532,"x_train":307.6069593637,"imp":14.3040260216,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.624415598,"x_train":613.0063601899,"imp":14.2337102267,"elo":1335.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.8874966855,"x_train":922.5048472624,"imp":14.2288971439,"elo":1331.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ExtraTrees","family":"Tree-based","x_infer":1.7842242231,"x_train":1235.4175263307,"imp":14.2362430223,"elo":1328.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":0.5109617193,"x_train":6.7731104239,"imp":20.3106713122,"elo":960.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":1.3803532068,"x_train":17.0044463073,"imp":19.9496894914,"elo":975.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":2.8851619804,"x_train":50.1505869221,"imp":18.7437585272,"elo":1025.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":3.8252238839,"x_train":101.3841054234,"imp":18.1779604669,"elo":1041.5,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":5.4333239012,"x_train":260.7094513648,"imp":17.2940660418,"elo":1081.5,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":6.6382886698,"x_train":528.4734813547,"imp":16.7765138581,"elo":1110.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.0448649768,"x_train":1059.681319286,"imp":16.1672923144,"elo":1131.9,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.5681335639,"x_train":1596.5291960327,"imp":15.7406048434,"elo":1150.4,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"FastaiMLP","family":"Neural Network","x_infer":7.9916224068,"x_train":2151.5332639388,"imp":15.4390915666,"elo":1160.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LightGBM","family":"Tree-based","x_infer":0.3736048612,"x_train":4.544665245,"imp":13.7590818546,"elo":1186.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"LightGBM","family":"Tree-based","x_infer":1.5668632649,"x_train":10.636770614,"imp":12.476047148,"elo":1280.6,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"LightGBM","family":"Tree-based","x_infer":2.9848948133,"x_train":27.7973506573,"imp":11.1844991259,"elo":1398.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.3643462664,"x_train":56.861966814,"imp":10.6108589021,"elo":1455.2,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.4293886426,"x_train":148.4745666134,"imp":10.0876299898,"elo":1519.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.7084309099,"x_train":299.3928042834,"imp":9.9288733954,"elo":1544.8,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"LightGBM","family":"Tree-based","x_infer":6.5371993778,"x_train":598.0451468439,"imp":9.7033578519,"elo":1566.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"LightGBM","family":"Tree-based","x_infer":5.4245638467,"x_train":893.892871219,"imp":9.5698709534,"elo":1579.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"LightGBM","family":"Tree-based","x_infer":4.5709782007,"x_train":1206.7027545763,"imp":9.4499487541,"elo":1591.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"LimiX","family":"Foundation Model","x_infer":20.8369258593,"x_train":122.3127076606,"imp":11.9179021798,"elo":1340.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Mitra","family":"Foundation Model","x_infer":2.1759075978,"x_train":94.1603678675,"imp":7.9340948995,"elo":1539.4,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.29377362,"x_train":25.1467163069,"imp":12.9962863151,"elo":1420.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"ModernNCA","family":"Neural Network","x_infer":0.7419941645,"x_train":75.362933136,"imp":12.3718490645,"elo":1446.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"ModernNCA","family":"Neural Network","x_infer":1.5922413434,"x_train":210.5456172414,"imp":11.6267396007,"elo":1494.4,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"ModernNCA","family":"Neural Network","x_infer":2.4715181761,"x_train":424.8014044984,"imp":11.3081905363,"elo":1525.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"ModernNCA","family":"Neural Network","x_infer":3.7597331666,"x_train":1033.2681066658,"imp":11.0747240533,"elo":1531.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"ModernNCA","family":"Neural Network","x_infer":5.1709239995,"x_train":2222.1208631991,"imp":10.8935157408,"elo":1530.2,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.2034777482,"x_train":4429.3203715947,"imp":10.7520103908,"elo":1529.7,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"ModernNCA","family":"Neural Network","x_infer":7.8424907827,"x_train":6624.9729135871,"imp":10.6295149838,"elo":1539.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"ModernNCA","family":"Neural Network","x_infer":8.0117732516,"x_train":9011.7804615398,"imp":10.6037965299,"elo":1540.6,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"Nori","family":"Foundation Model","x_infer":1.5206234824,"x_train":14.7870051149,"imp":2.9817681293,"elo":2068.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"Nori-30M","family":"Foundation Model","x_infer":2.0843698074,"x_train":14.4142036067,"imp":1.6715446805,"elo":2118.2,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.0198449839,"x_train":34.1835619211,"imp":21.0267397864,"elo":908.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.143539937,"x_train":68.7178569705,"imp":18.9265958956,"elo":1014.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4090035871,"x_train":173.2458309649,"imp":17.0684828461,"elo":1103.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"PerpetualBooster","family":"Tree-based","x_infer":0.4528171238,"x_train":208.1564418277,"imp":17.0805060316,"elo":1113.9,"imputed":false,"imputed_pct":0.0,"n_configs":6.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.1250003559,"x_train":0.6913103896,"imp":18.8630134196,"elo":1001.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.2971430628,"x_train":4.4537617156,"imp":17.8387549123,"elo":1071.1,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.5361761201,"x_train":14.301129027,"imp":17.2622565173,"elo":1119.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.7051520621,"x_train":30.686422522,"imp":16.8003854792,"elo":1152.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.9048497109,"x_train":79.9124717904,"imp":16.5519192401,"elo":1169.7,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RandomForest","family":"Tree-based","x_infer":0.962102534,"x_train":162.1056028409,"imp":16.4538884455,"elo":1180.7,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.0937878381,"x_train":326.1499058044,"imp":16.3873501449,"elo":1191.6,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.2475282764,"x_train":490.3882635115,"imp":16.3633492171,"elo":1196.2,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RandomForest","family":"Tree-based","x_infer":1.1449696518,"x_train":657.9874551063,"imp":16.3574828558,"elo":1199.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealMLP","family":"Neural Network","x_infer":7.8151845047,"x_train":22.4158785265,"imp":11.5329003355,"elo":1323.9,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealMLP","family":"Neural Network","x_infer":10.4129743265,"x_train":44.4251076686,"imp":9.5741826539,"elo":1409.2,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealMLP","family":"Neural Network","x_infer":15.4450757134,"x_train":117.3230828329,"imp":7.6359018563,"elo":1562.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealMLP","family":"Neural Network","x_infer":20.928441831,"x_train":253.2768449964,"imp":6.9578682152,"elo":1600.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealMLP","family":"Neural Network","x_infer":24.2900358461,"x_train":616.4798700039,"imp":6.5714223235,"elo":1638.4,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealMLP","family":"Neural Network","x_infer":28.5298041944,"x_train":1238.3164021687,"imp":6.339582604,"elo":1678.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealMLP","family":"Neural Network","x_infer":33.0852161799,"x_train":2457.5657747181,"imp":6.2626565171,"elo":1690.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealMLP","family":"Neural Network","x_infer":35.1242790175,"x_train":3675.6638274823,"imp":6.2412663581,"elo":1695.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealMLP","family":"Neural Network","x_infer":35.5866553094,"x_train":4918.0723002043,"imp":6.2282511886,"elo":1698.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":0.5321004862,"x_train":8.242142907,"imp":5.427403104,"elo":1664.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":1.4099952993,"x_train":19.8358235954,"imp":4.7261793707,"elo":1701.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":2.7879588822,"x_train":56.6252728837,"imp":3.5987206349,"elo":1782.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":4.0310191127,"x_train":113.5361347503,"imp":3.0747732047,"elo":1849.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":5.4432569251,"x_train":295.9584460071,"imp":2.5468910913,"elo":1878.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.4486360166,"x_train":584.1451401972,"imp":2.3895569087,"elo":1902.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.6972941708,"x_train":1169.0460603376,"imp":2.3310758771,"elo":1922.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":6.5512603426,"x_train":1760.7500065581,"imp":2.4258915729,"elo":1911.7,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"RealTabPFN-2.5","family":"Foundation Model","x_infer":7.5456187851,"x_train":2356.3627464618,"imp":2.4360173334,"elo":1914.0,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"SAP-RPT-OSS","family":"Foundation Model","x_infer":0.8423700735,"x_train":25.2809733305,"imp":10.2979773772,"elo":1413.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":43.0616958015,"x_train":51.2843993162,"imp":5.9831587418,"elo":1730.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabDPT","family":"Foundation Model","x_infer":70.1115350575,"x_train":95.8813700743,"imp":5.3971294242,"elo":1763.8,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabDPT","family":"Foundation Model","x_infer":117.0496168126,"x_train":254.4393954229,"imp":4.8561409198,"elo":1795.7,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabDPT","family":"Foundation Model","x_infer":145.633406717,"x_train":502.170242762,"imp":4.4172900231,"elo":1849.6,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabDPT","family":"Foundation Model","x_infer":177.8219001626,"x_train":1229.6581613295,"imp":4.1595602054,"elo":1877.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabDPT","family":"Foundation Model","x_infer":195.075324367,"x_train":2450.1429629138,"imp":4.1026352104,"elo":1881.6,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabDPT","family":"Foundation Model","x_infer":207.9632508624,"x_train":4887.1069852066,"imp":4.028961281,"elo":1893.5,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabDPT","family":"Foundation Model","x_infer":227.9964108122,"x_train":7331.038512164,"imp":3.9899091528,"elo":1912.8,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabDPT","family":"Foundation Model","x_infer":232.1033993492,"x_train":9830.4728139885,"imp":3.9787406493,"elo":1918.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabDPT-Turbo","family":"Foundation Model","x_infer":0.2617923617,"x_train":7.4300775989,"imp":6.1398219236,"elo":1682.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabFM","family":"Foundation Model","x_infer":4.0307404894,"x_train":56.2879369055,"imp":1.9939435026,"elo":2193.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabICLv2","family":"Foundation Model","x_infer":0.1387839993,"x_train":5.9977774395,"imp":4.914734588,"elo":1739.1,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.4849712473,"x_train":14.2468083815,"imp":10.0205504104,"elo":1305.6,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabM","family":"Neural Network","x_infer":0.9132928236,"x_train":34.9108464136,"imp":9.4576657254,"elo":1334.3,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabM","family":"Neural Network","x_infer":1.7529208648,"x_train":89.6967770714,"imp":8.8810065831,"elo":1382.8,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabM","family":"Neural Network","x_infer":2.4556253929,"x_train":190.6759613979,"imp":8.3070618065,"elo":1438.9,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabM","family":"Neural Network","x_infer":3.1346820285,"x_train":456.3526262514,"imp":7.8318467474,"elo":1480.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabM","family":"Neural Network","x_infer":3.5972380055,"x_train":918.9449885462,"imp":7.4798691525,"elo":1524.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabM","family":"Neural Network","x_infer":4.3086986908,"x_train":1845.4338559674,"imp":7.2737826281,"elo":1547.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabM","family":"Neural Network","x_infer":4.6217168271,"x_train":2797.5928059,"imp":7.2496835275,"elo":1546.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabM","family":"Neural Network","x_infer":4.367582289,"x_train":3814.2351446507,"imp":7.2240501445,"elo":1539.7,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabPFN-2.6","family":"Foundation Model","x_infer":0.6295234248,"x_train":10.02817758,"imp":2.8808212181,"elo":1982.3,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFN-3","family":"Foundation Model","x_infer":0.4873524134,"x_train":14.1514284946,"imp":2.5953010295,"elo":1996.7,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":0.7259204826,"x_train":8.6865743952,"imp":8.7076340469,"elo":1384.4,"imputed":true,"imputed_pct":12.5,"n_configs":1.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":7.3089637658,"x_train":52.109381455,"imp":7.4035677786,"elo":1443.3,"imputed":true,"imputed_pct":12.5,"n_configs":2.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":18.0487052426,"x_train":144.9748468042,"imp":6.2169828319,"elo":1516.6,"imputed":true,"imputed_pct":12.5,"n_configs":5.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":26.9258987977,"x_train":312.7874942852,"imp":5.7796838131,"elo":1545.1,"imputed":true,"imputed_pct":12.5,"n_configs":10.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":29.661187394,"x_train":754.6626382647,"imp":5.0950859152,"elo":1608.2,"imputed":true,"imputed_pct":12.5,"n_configs":25.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":35.9775514627,"x_train":1536.077969506,"imp":4.8454469305,"elo":1626.3,"imputed":true,"imputed_pct":12.5,"n_configs":50.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":41.4628183099,"x_train":3126.6373952668,"imp":4.5450092452,"elo":1643.7,"imputed":true,"imputed_pct":12.5,"n_configs":100.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":53.4343776598,"x_train":4723.4389861822,"imp":4.4911277046,"elo":1652.5,"imputed":true,"imputed_pct":12.5,"n_configs":150.0},{"method":"TabPFNv2","family":"Foundation Model","x_infer":56.0801957423,"x_train":6437.8441109111,"imp":4.4358649388,"elo":1657.0,"imputed":true,"imputed_pct":12.5,"n_configs":201.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":5.7468382623,"x_train":454.9991027301,"imp":27.5261770168,"elo":907.4,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":9.6136847143,"x_train":923.2022669306,"imp":27.3713103765,"elo":927.5,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":17.3872419149,"x_train":2264.5016798126,"imp":26.9693598956,"elo":969.9,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":21.4500316943,"x_train":4340.8142220335,"imp":26.2782538996,"elo":1017.4,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":26.9593058501,"x_train":11688.77778151,"imp":25.4429386552,"elo":1063.2,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":33.0819178763,"x_train":23707.0922342587,"imp":24.9021077683,"elo":1103.1,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":37.9017474296,"x_train":47472.2129899786,"imp":24.5530137841,"elo":1123.4,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":38.4814086805,"x_train":71823.9094609453,"imp":24.4153443067,"elo":1132.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TabSTAR","family":"Foundation Model","x_infer":36.5771589943,"x_train":96418.8447274511,"imp":24.3105081894,"elo":1136.4,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"TabSwift","family":"Foundation Model","x_infer":0.0653075084,"x_train":4.0841565985,"imp":5.9100274158,"elo":1620.5,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.1305890866,"x_train":22.2534811738,"imp":15.6737475478,"elo":1038.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.2739656197,"x_train":54.2457268049,"imp":15.1761073979,"elo":1075.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.4747909368,"x_train":127.8066374424,"imp":14.1453339402,"elo":1146.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.6786610978,"x_train":271.1084504473,"imp":13.71982596,"elo":1175.0,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"TorchMLP","family":"Neural Network","x_infer":0.9735411513,"x_train":648.9159658958,"imp":12.9816291054,"elo":1214.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.2437344525,"x_train":1311.0552340778,"imp":12.4498845916,"elo":1252.5,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.4822354588,"x_train":2566.6969558991,"imp":11.8624617347,"elo":1275.8,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.6235068465,"x_train":3875.6410639922,"imp":11.5964446071,"elo":1294.5,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"TorchMLP","family":"Neural Network","x_infer":1.7488211618,"x_train":5196.3714196423,"imp":11.3898917946,"elo":1311.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.2554761956,"x_train":4.0843435263,"imp":15.0194661753,"elo":1096.8,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"XGBoost","family":"Tree-based","x_infer":0.957900168,"x_train":10.4789410396,"imp":12.2265999012,"elo":1286.0,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.0042194555,"x_train":29.8024094892,"imp":11.2783006114,"elo":1360.2,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.3054096434,"x_train":60.219465986,"imp":10.7625899279,"elo":1411.8,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.4008496262,"x_train":157.148713481,"imp":10.397729215,"elo":1444.1,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.6357668007,"x_train":316.488046922,"imp":10.2401435688,"elo":1456.3,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"XGBoost","family":"Tree-based","x_infer":2.6775449023,"x_train":634.705809045,"imp":10.1282155644,"elo":1465.3,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.121628747,"x_train":952.8874450538,"imp":10.0298489274,"elo":1466.9,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"XGBoost","family":"Tree-based","x_infer":3.7439813565,"x_train":1277.5662482217,"imp":9.9954828299,"elo":1466.9,"imputed":false,"imputed_pct":0.0,"n_configs":201.0},{"method":"iLTM","family":"Foundation Model","x_infer":270.0943042458,"x_train":691.2819503211,"imp":20.9779761968,"elo":866.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"iLTM","family":"Foundation Model","x_infer":546.5990400536,"x_train":2116.0353219536,"imp":15.7245100978,"elo":1107.4,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"iLTM","family":"Foundation Model","x_infer":1020.4039532515,"x_train":6702.4310187379,"imp":11.1994420107,"elo":1363.3,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"iLTM","family":"Foundation Model","x_infer":1274.0935394324,"x_train":15189.3088063979,"imp":9.9294564808,"elo":1470.7,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"iLTM","family":"Foundation Model","x_infer":1744.1344030008,"x_train":37312.4085130914,"imp":8.7787641694,"elo":1607.3,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"iLTM","family":"Foundation Model","x_infer":1759.1043245943,"x_train":39204.8641933754,"imp":8.7636936868,"elo":1606.7,"imputed":false,"imputed_pct":0.0,"n_configs":26.0},{"method":"xRFM","family":"Other","x_infer":1.3621042316,"x_train":7.002585411,"imp":16.5424125465,"elo":1203.0,"imputed":false,"imputed_pct":0.0,"n_configs":1.0},{"method":"xRFM","family":"Other","x_infer":1.7787898657,"x_train":15.1443119472,"imp":14.8045524272,"elo":1256.9,"imputed":false,"imputed_pct":0.0,"n_configs":2.0},{"method":"xRFM","family":"Other","x_infer":1.849660528,"x_train":39.7770511549,"imp":12.5667233788,"elo":1341.6,"imputed":false,"imputed_pct":0.0,"n_configs":5.0},{"method":"xRFM","family":"Other","x_infer":1.6661465754,"x_train":81.229812475,"imp":11.0712301058,"elo":1407.1,"imputed":false,"imputed_pct":0.0,"n_configs":10.0},{"method":"xRFM","family":"Other","x_infer":2.1507772428,"x_train":207.1643683575,"imp":9.7424442267,"elo":1465.9,"imputed":false,"imputed_pct":0.0,"n_configs":25.0},{"method":"xRFM","family":"Other","x_infer":2.6630094827,"x_train":413.5799198557,"imp":9.3688825164,"elo":1500.4,"imputed":false,"imputed_pct":0.0,"n_configs":50.0},{"method":"xRFM","family":"Other","x_infer":3.2283609867,"x_train":828.216131447,"imp":9.0618942855,"elo":1513.0,"imputed":false,"imputed_pct":0.0,"n_configs":100.0},{"method":"xRFM","family":"Other","x_infer":2.4759145961,"x_train":1248.051456359,"imp":8.9494988014,"elo":1524.1,"imputed":false,"imputed_pct":0.0,"n_configs":150.0},{"method":"xRFM","family":"Other","x_infer":2.446364541,"x_train":1673.3760165809,"imp":8.9390070431,"elo":1521.3,"imputed":false,"imputed_pct":0.0,"n_configs":201.0}];
 
 
   const NS = "http://www.w3.org/2000/svg";
@@ -326,6 +337,23 @@
     "Reference Pipeline": "var(--fam-reference)",
     [FAM_MERGED]: "var(--fam-baseline)",
   };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
   // The same hues stepped for use as text (see the --fam-*-ink tokens).
   const FAM_INK = {
     "Foundation Model": "var(--fam-foundation-ink)",
@@ -405,7 +433,11 @@
   // reader wants first is the figure, and it is the state worth exporting. The
   // controls, chip list and data table are one click away behind "Edit view".
   // `afterToggle` re-renders charts whose size is measured from the layout.
-  function setUpPaperView(afterToggle) {
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
     const root = document.documentElement;
     let hostTheme = null;   // the embedding page's choice, captured on entry
     const btn = document.getElementById("btn-paper");
@@ -442,7 +474,7 @@
         if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
       });
     }
-    setPaper(true);   // the figure is what opens
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
   }
 
   // --- Figure export ---------------------------------------------------------
@@ -1018,7 +1050,7 @@
       const famBtn = document.createElement("button");
       famBtn.className = "famchip";
       famBtn.style.setProperty("--fam", FAM_VAR[fam]);
-      famBtn.innerHTML = `<span class="dot"></span>${fam} <span class="count">&times;${methods.length}</span>`;
+      famBtn.innerHTML = famChipLabel(fam, methods.length);
       famBtn.title = `Toggle all ${methods.length} ${fam} methods`;
       famBtn.addEventListener("click", () => toggleFamily(fam));
       row.appendChild(famBtn);
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/winrate_explorer.html b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/winrate_explorer.html
new file mode 100644
index 0000000000000000000000000000000000000000..855a7a3d5af37ba8c9c9b3426f4f40009819de52
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/winrate_explorer.html
@@ -0,0 +1,1169 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena win-rate matrix — imputation_yes/splits_lite/tasks_regression/datasets_small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* The matrix scrolls in both directions: 40+ models will not fit a page either
+     way, and shrinking the cells to fit makes the numbers unreadable. `margin:
+     auto` centres it once the selection is small enough to fit — and collapses
+     to zero when it overflows, which `justify-content: center` would not. */
+  .wr-scroll {
+    overflow: auto; max-width: 100%;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .wr-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .wr-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  #chart { display: block; margin-inline: auto; }
+  .wr-empty { padding: 40px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, matrix only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <button class="btn toggle" id="btn-best"
+      title="Keep only each model's best-performing variant, one row and column per model — as the static figure does">One per model</button>
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <label class="metricpick">Order
+      <select id="order-select">
+        <option value="published">As published</option>
+        <option value="winrate">Mean win rate</option>
+        <option value="name">Name</option>
+      </select>
+    </label>
+    <label class="metricpick">Size
+      <select id="zoom-select" title="Fit scales the whole matrix into the panel; the percentages draw it at reading size and scroll">
+        <option value="fit">Fit panel</option>
+        <option value="0.6">60%</option>
+        <option value="0.8">80%</option>
+        <option value="1">100%</option>
+      </select>
+    </label>
+    <span class="hint">Click a row or column label to remove that model</span>
+  </div>
+  <div class="exportbar" id="exportbar" hidden>
+    <span class="hint">Export figure</span>
+    <button class="btn" id="btn-svg" title="Download as SVG — vector, keeps text selectable">Download SVG</button>
+    <button class="btn" id="btn-pdf" title="Download as a one-page PDF">Download PDF</button>
+    <button class="btn" id="btn-png" title="Download as PNG at 3x scale">Download PNG</button>
+  </div>
+  <div class="chips" id="chips"></div>
+  <div class="wr-scroll" id="wrap">
+    <svg id="chart" role="img" aria-label="Pairwise win-rate matrix"></svg>
+  </div>
+  <div class="tooltip"></div>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "methods": ["TabFM (default)", "Nori-30M (default)", "Nori (default)", "AutoGluon 1.5 (extreme, 4h)", "TabPFN-3 (default)", "TabPFN-2.6 (default)", "RealTabPFN-2.5 (tuned + ensembled)", "TabDPT (tuned + ensembled)", "TabDPT (tuned)", "TabICLv2 (default)", "RealTabPFN-2.5 (tuned)", "TabDPT (default)", "RealMLP (tuned + ensembled)", "TabDPT-Turbo (default)", "AutoGluon 1.4 (best, 4h)", "RealTabPFN-2.5 (default)", "TabPFNv2 (tuned + ensembled)", "TabSwift (default)", "iLTM (tuned + ensembled)", "LightGBM (tuned + ensembled)", "RealMLP (tuned)", "ModernNCA (tuned + ensembled)", "Mitra (default)", "TabM (tuned + ensembled)", "CatBoost (tuned + ensembled)", "CatBoost (tuned)", "xRFM (tuned + ensembled)", "TabM (tuned)", "TabPFNv2 (tuned)", "xRFM (tuned)", "XGBoost (tuned)", "XGBoost (tuned + ensembled)", "LightGBM (tuned)", "ChimeraBoost (tuned + ensembled)", "ModernNCA (default)", "SAP-RPT-OSS (default)", "iLTM (tuned)", "CatBoost (default)", "TabPFNv2 (default)", "ChimeraBoost (tuned)", "ExtraTrees (tuned + ensembled)", "RealMLP (default)", "LimiX (default)", "ModernNCA (tuned)", "TabM (default)", "TorchMLP (tuned + ensembled)", "ExtraTrees (tuned)", "ChimeraBoost (default)", "TorchMLP (tuned)", "EBM (tuned + ensembled)", "EBM (tuned)", "RandomForest (tuned + ensembled)", "xRFM (default)", "LightGBM (default)", "FastaiMLP (tuned + ensembled)", "ExtraTrees (default)", "RandomForest (tuned)", "PerpetualBooster (tuned + ensembled)", "TabSTAR (tuned + ensembled)", "FastaiMLP (tuned)", "PerpetualBooster (tuned)", "XGBoost (default)", "TabSTAR (tuned)", "EBM (default)", "TorchMLP (default)", "RandomForest (default)", "FastaiMLP (default)", "TabSTAR (default)", "KNN (tuned + ensembled)", "PerpetualBooster (default)", "iLTM (default)", "KNN (tuned)", "KNN (default)", "Linear (tuned + ensembled)", "Linear (tuned)", "Linear (default)"], "matrix": [[NaN, 0.75, 0.875, 0.625, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.75, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, NaN, 0.625, 0.75, 0.5, 0.75, 0.75, 0.625, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.375, NaN, 0.5, 0.5, 0.625, 0.625, 0.75, 1.0, 0.875, 0.75, 0.875, 1.0, 1.0, 0.875, 0.75, 0.625, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.375, 0.25, 0.5, NaN, 0.5, 0.625, 0.5, 0.625, 0.75, 0.625, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 0.75, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.5, 0.5, 0.5, NaN, 0.5, 0.375, 0.5, 0.625, 0.625, 0.75, 1.0, 0.875, 1.0, 0.875, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.25, 0.375, 0.375, 0.5, NaN, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.625, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.25, 0.375, 0.5, 0.625, 0.375, NaN, 0.625, 0.625, 0.625, 0.875, 0.75, 0.875, 0.875, 0.75, 0.875, 0.625, 0.75, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.375, 0.25, 0.375, 0.5, 0.375, 0.375, NaN, 0.75, 0.625, 0.5, 0.875, 0.875, 0.875, 0.875, 0.75, 0.5, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.25, 0.375, 0.375, 0.375, 0.25, NaN, 0.5, 0.625, 0.75, 0.875, 0.875, 0.875, 0.5, 0.375, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.375, 0.375, 0.25, 0.375, 0.375, 0.5, NaN, 0.75, 0.5, 0.625, 0.75, 0.75, 0.625, 0.5, 0.75, 0.875, 0.875, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.5, 0.375, 0.25, NaN, 0.625, 0.625, 0.75, 0.625, 0.75, 0.5, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.75, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 0.75, 1.0, 1.0, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.25, 0.125, 0.25, 0.5, 0.375, NaN, 0.625, 0.75, 0.625, 0.375, 0.5, 0.625, 0.625, 0.75, 0.625, 0.75, 0.625, 0.75, 0.875, 0.875, 1.0, 0.75, 0.625, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.75, 1.0, 1.0, 0.875, 0.75, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.375, 0.375, 0.375, NaN, 0.625, 0.5, 0.5, 0.375, 0.625, 0.75, 0.75, 0.75, 0.625, 0.625, 0.625, 0.875, 0.875, 0.75, 0.75, 0.625, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.625, 1.0, 1.0, 1.0, 0.75, 0.875, 1.0, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.375, NaN, 0.625, 0.25, 0.375, 0.625, 0.625, 0.75, 0.875, 0.625, 0.625, 0.5, 0.875, 0.75, 1.0, 0.75, 0.625, 0.875, 1.0, 1.0, 1.0, 0.875, 0.75, 0.875, 0.875, 0.875, 0.5, 0.875, 1.0, 0.875, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.375, 0.375, 0.5, 0.375, NaN, 0.5, 0.375, 0.5, 0.5, 0.875, 0.625, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.75, 0.5, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 0.75, 0.875, 1.0, 0.625, 1.0, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.5, 0.375, 0.25, 0.625, 0.5, 0.75, 0.5, NaN, 0.375, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0], [0.125, 0.25, 0.375, 0.25, 0.125, 0.375, 0.375, 0.5, 0.625, 0.5, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, NaN, 0.625, 0.625, 0.625, 0.625, 0.625, 0.8125, 0.625, 0.625, 0.625, 0.75, 0.75, 0.9375, 0.75, 0.625, 0.75, 0.625, 0.75, 0.625, 0.875, 0.75, 0.75, 0.9375, 0.75, 0.625, 0.875, 0.875, 0.625, 0.75, 0.75, 0.625, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 0.9375, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.9375, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.25, 0.0, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, 0.25, 0.375, NaN, 0.625, 0.625, 0.625, 0.375, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.625, 0.875, 0.875, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 0.75, 0.5, 0.75, 0.75, 0.875, 0.75, 1.0, 0.875, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.375, 0.375, 0.25, 0.375, 0.5, 0.25, 0.375, 0.375, NaN, 0.5, 0.5, 0.5, 0.375, 0.625, 0.75, 0.75, 0.5, 0.75, 0.5, 0.625, 0.875, 0.875, 0.875, 0.75, 0.875, 0.75, 0.75, 0.875, 0.5, 0.875, 1.0, 0.75, 0.625, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 0.75, 0.75, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.375, 0.5, NaN, 0.5, 0.375, 0.5, 0.625, 0.5, 0.5, 0.5, 0.625, 0.375, 0.625, 0.75, 0.75, 1.0, 0.875, 0.875, 0.625, 0.625, 0.875, 0.5, 1.0, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.375, 0.25, 0.125, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, NaN, 0.375, 0.625, 0.5, 0.625, 0.75, 0.625, 0.625, 0.375, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.625, 0.5, 0.75, 0.75, 0.875, 0.75, 0.875, 1.0, 0.75, 0.75, 0.75, 1.0, 1.0, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.375, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.625, 0.5, 0.625, 0.625, NaN, 0.5, 0.5, 0.375, 0.375, 0.5, 0.5, 0.5, 0.5, 0.625, 0.75, 0.875, 0.75, 0.625, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.625, 0.625, 1.0, 0.75, 0.625, 0.875, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.1875, 0.375, 0.625, 0.5, 0.375, 0.5, NaN, 0.625, 0.5, 0.5, 0.625, 0.625, 0.4375, 0.75, 0.625, 0.625, 0.75, 0.625, 0.75, 0.5, 0.625, 0.625, 0.5625, 0.75, 0.625, 0.875, 0.75, 0.75, 0.875, 0.875, 0.625, 0.875, 0.875, 1.0, 1.0, 0.75, 0.75, 0.875, 0.875, 0.875, 0.8125, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.9375, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.5, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, NaN, 0.5, 0.375, 0.625, 0.875, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.75, 0.5, 0.875, 0.75, 0.75, 0.75, 0.75, 1.0, 0.875, 0.75, 0.875, 0.875, 0.75, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.25, 0.375, 0.25, 0.25, 0.5, 0.375, 0.625, 0.5, 0.5, NaN, 0.625, 0.25, 0.5, 0.375, 0.5, 0.75, 0.75, 0.875, 0.625, 0.625, 0.75, 0.75, 0.875, 0.5, 0.875, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.5, 0.25, 0.625, 0.5, 0.625, 0.375, NaN, 0.5, 0.625, 0.375, 0.5, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.875, 0.5, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.875, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.75, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.0, 0.25, 0.0, 0.25, 0.25, 0.25, 0.375, 0.5, 0.5, 0.375, 0.5, 0.375, 0.375, 0.75, 0.5, NaN, 0.5, 0.375, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.5, 0.75, 0.875, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 0.75, 0.75, 0.875, 0.875, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.5, 0.375, 0.125, 0.5, 0.375, 0.5, NaN, 0.375, 0.5, 0.5, 0.5, 0.5, 0.75, 0.375, 0.5, 0.625, 0.75, 0.5, 0.75, 0.75, 0.75, 0.625, 0.625, 1.0, 0.875, 0.75, 0.75, 1.0, 0.75, 0.75, 0.75, 0.75, 1.0, 0.875, 0.875, 0.75, 0.875, 0.75, 1.0, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.375, 0.25, 0.375, 0.375, 0.375, 0.5, 0.25, 0.0625, 0.375, 0.5, 0.625, 0.625, 0.5, 0.5625, 0.5, 0.625, 0.625, 0.625, 0.625, NaN, 0.5, 0.625, 0.625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.625, 0.5625, 0.625, 0.625, 0.5, 0.625, 0.625, 0.625, 0.75, 0.625, 0.625, 0.75, 0.875, 0.875, 0.625, 0.75, 0.625, 0.75, 0.875, 0.8125, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.8125, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.25, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.375, 0.375, 0.25, 0.5, 0.25, 0.375, 0.5, 0.5, 0.25, 0.5, 0.5, NaN, 0.625, 0.5, 0.5, 0.375, 0.75, 0.625, 0.75, 0.625, 0.5, 0.5, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 0.75, 0.875, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.375, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.25, 0.375, 0.5, 0.375, 0.375, NaN, 0.625, 0.5, 0.625, 0.625, 0.5, 0.625, 0.625, 0.5, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.875, 0.625, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 1.0, 1.0, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.5, 0.5, 0.375, 0.5, 0.375, NaN, 0.5, 0.5, 0.5, 0.375, 0.625, 0.625, 0.5, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 0.75, 1.0, 1.0, 0.875, 0.875, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.375, 0.25, 0.125, 0.0, 0.25, 0.125, 0.25, 0.375, 0.125, 0.25, 0.375, 0.5, 0.375, 0.5, 0.5, 0.5, NaN, 0.5, 0.625, 0.5, 0.625, 0.5, 0.5, 0.625, 0.625, 0.75, 0.5, 0.875, 0.875, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 0.75, 1.0, 0.75, 1.0, 1.0, 1.0, 0.75, 1.0, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.25, 0.25, 0.125, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.625, 0.375, 0.5, 0.5, NaN, 0.375, 0.5, 0.5, 0.5, 0.5, 0.75, 0.625, 0.75, 0.5, 0.625, 0.75, 0.625, 0.625, 0.875, 0.75, 0.875, 0.875, 0.75, 0.75, 0.625, 1.0, 0.625, 0.75, 1.0, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.375, 0.125, 0.125, 0.125, 0.25, 0.375, 0.25, 0.5, 0.375, 0.25, 0.375, 0.625, 0.5, 0.25, 0.375, 0.5, 0.375, 0.625, NaN, 0.375, 0.5, 0.625, 0.375, 0.625, 0.75, 0.625, 0.5, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.625, 0.625, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 0.75, 0.75, 0.875, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.25, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.25, 0.375, 0.375, 0.5, 0.375, 0.375, 0.5, 0.625, 0.5, 0.5, 0.625, NaN, 0.5, 0.5, 0.375, 0.625, 0.5, 0.625, 0.75, 0.5, 0.5, 0.625, 0.625, 0.75, 0.625, 0.625, 0.75, 0.875, 0.625, 0.875, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.625, 0.875, 0.75, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.5, 0.5, 0.5, NaN, 0.5, 0.5, 0.625, 0.5, 0.625, 0.625, 0.625, 0.75, 0.5, 0.5, 0.75, 0.5, 0.625, 0.75, 0.75, 0.875, 0.875, 0.875, 0.75, 0.75, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.25, 0.125, 0.125, 0.375, 0.375, 0.375, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, 0.5, 0.375, 0.5, 0.5, NaN, 0.5, 0.625, 0.625, 0.625, 0.375, 0.625, 0.625, 0.625, 0.875, 1.0, 0.625, 0.75, 0.75, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.25, 0.25, 0.125, 0.25, 0.375, 0.5, 0.375, 0.125, 0.0625, 0.5, 0.5, 0.5, 0.5, 0.375, 0.4375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.4375, 0.5, 0.5, 0.5, 0.5, 0.5, 0.625, 0.625, 0.5, 0.5, NaN, 0.5, 0.625, 0.5, 0.625, 0.5, 0.5, 0.5, 0.625, 0.5, 0.5, 0.75, 0.75, 0.625, 0.75, 0.625, 0.625, 0.875, 0.6875, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.8125, 0.625, 0.625, 0.875, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.25, 0.25, 0.125, 0.0, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.25, 0.375, 0.5, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, NaN, 0.5, 0.75, 0.375, 0.625, 0.625, 0.625, 0.5, 0.75, 0.5, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 1.0, 0.75, 0.875, 0.875, 0.875, 0.75, 1.0, 0.875, 0.875, 1.0, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.375, 0.25, 0.0, 0.125, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.375, 0.25, 0.5, 0.5, 0.375, 0.375, 0.5, NaN, 0.375, 0.5, 0.625, 0.375, 0.5, 0.625, 0.75, 0.5, 0.625, 0.625, 1.0, 0.75, 0.5, 0.75, 1.0, 1.0, 0.75, 0.75, 0.625, 0.875, 0.625, 0.75, 0.75, 0.625, 1.0, 0.875, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.375, 0.125, 0.25, 0.25, 0.25, 0.375, 0.25, 0.5, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, 0.25, 0.625, NaN, 0.5, 0.375, 0.375, 0.5, 0.625, 0.625, 0.5, 0.625, 0.625, 0.75, 0.75, 0.75, 0.75, 0.75, 0.75, 0.625, 0.75, 0.625, 0.75, 1.0, 0.75, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.125, 0.25, 0.375, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.25, 0.25, 0.5, 0.5, 0.5, 0.25, 0.375, 0.625, 0.375, 0.625, 0.5, 0.5, NaN, 0.5, 0.5, 0.5, 0.5, 0.625, 0.375, 0.625, 0.625, 0.5, 0.5, 0.625, 0.625, 0.625, 0.5, 0.625, 0.75, 0.625, 0.625, 0.625, 0.75, 0.75, 0.75, 0.625, 0.625, 0.875, 0.875, 0.875, 0.75, 0.875, 1.0, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.375, 0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.25, 0.125, 0.125, 0.25, 0.375, 0.375, 0.25, 0.25, 0.25, 0.125, 0.375, 0.375, 0.5, 0.375, 0.375, 0.5, 0.375, 0.375, 0.625, 0.5, NaN, 0.5, 0.5, 0.375, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.625, 0.625, 0.75, 0.625, 0.875, 0.75, 0.875, 0.625, 0.875, 0.875, 0.75, 1.0, 0.75, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.0, 0.25, 0.125, 0.0, 0.25, 0.25, 0.25, 0.0, 0.375, 0.375, 0.25, 0.125, 0.125, 0.25, 0.375, 0.5, 0.25, 0.375, 0.5, 0.375, 0.625, 0.625, 0.5, 0.5, NaN, 0.5, 0.625, 0.625, 0.75, 0.625, 0.625, 0.625, 0.625, 0.875, 0.75, 0.625, 0.625, 0.875, 0.75, 0.875, 0.875, 1.0, 0.75, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.375, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.375, 0.375, 0.375, 0.5, 0.375, 0.5, 0.375, 0.5, 0.5, 0.5, 0.5, 0.5, NaN, 0.625, 0.5, 0.75, 0.625, 0.625, 0.625, 0.625, 0.875, 0.875, 0.625, 0.625, 0.75, 0.75, 0.875, 0.75, 0.875, 0.75, 0.75, 1.0, 0.875, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.375, 0.25, 0.0, 0.125, 0.25, 0.125, 0.375, 0.25, 0.125, 0.25, 0.125, 0.25, 0.375, 0.25, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.5, 0.125, 0.375, 0.5, 0.375, 0.375, 0.5, 0.625, 0.375, 0.375, NaN, 0.5, 0.5, 0.625, 0.625, 0.75, 0.75, 0.375, 0.75, 0.75, 0.875, 0.75, 0.75, 0.625, 0.875, 0.5, 0.75, 0.75, 0.5, 1.0, 0.75, 0.875, 1.0, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.125, 0.125, 0.0, 0.0, 0.25, 0.25, 0.375, 0.125, 0.125, 0.125, 0.375, 0.125, 0.25, 0.25, 0.25, 0.0, 0.5, 0.25, 0.25, 0.375, 0.375, 0.5, 0.375, 0.5, 0.5, NaN, 0.5, 0.75, 0.75, 0.625, 0.625, 0.625, 0.75, 0.625, 0.75, 1.0, 0.75, 0.625, 0.875, 0.75, 0.75, 0.75, 0.625, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.0, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.0, 0.25, 0.375, 0.125, 0.125, 0.25, 0.25, 0.375, 0.375, 0.5, 0.375, 0.5, 0.5, 0.5, 0.5, 0.625, 0.375, 0.25, 0.25, 0.5, 0.5, NaN, 0.5, 0.5, 0.625, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.875, 0.75, 0.75, 0.75, 0.75, 1.0, 0.625, 0.875, 0.875, 0.75, 1.0, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.0, 0.25, 0.0, 0.25, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.375, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.5, NaN, 0.75, 0.375, 0.625, 0.5, 0.625, 0.375, 0.375, 0.5, 0.875, 0.75, 0.75, 0.625, 0.875, 1.0, 0.75, 0.625, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.0, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.375, 0.25, 0.5, 0.25, NaN, 0.375, 0.625, 0.5, 0.375, 0.375, 0.375, 0.5, 0.75, 0.625, 0.5, 0.625, 0.75, 0.875, 0.75, 0.625, 0.875, 0.875, 0.875, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.25, 0.125, 0.25, 0.125, 0.0, 0.125, 0.0, 0.25, 0.375, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.25, 0.0, 0.25, 0.5, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.625, 0.625, NaN, 0.5, 0.5, 0.375, 0.75, 0.875, 0.625, 0.625, 0.375, 0.625, 0.625, 0.625, 0.625, 0.5, 1.0, 0.75, 0.875, 1.0, 0.625, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.5, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.375, 0.5, NaN, 0.375, 0.25, 0.5, 0.625, 0.375, 0.625, 0.375, 0.375, 0.375, 0.75, 0.625, 0.5, 0.75, 0.625, 0.875, 1.0, 0.75, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.25, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.375, 0.125, 0.0, 0.0, 0.125, 0.375, 0.125, 0.125, 0.125, 0.125, 0.375, 0.375, 0.5, 0.25, 0.375, 0.25, 0.125, 0.125, 0.625, 0.375, 0.375, 0.5, 0.5, 0.5, 0.625, NaN, 0.625, 0.625, 0.625, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.625, 0.75, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.25, 0.125, 0.125, 0.375, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.125, 0.25, 0.25, 0.375, 0.375, 0.625, 0.625, 0.75, 0.375, NaN, 0.625, 0.625, 0.375, 0.75, 0.625, 0.375, 0.625, 0.75, 0.75, 0.75, 0.875, 1.0, 0.875, 0.875, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.25, 0.125, 0.125, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.375, 0.125, 0.25, 0.25, 0.125, 0.125, 0.375, 0.0, 0.25, 0.375, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.625, 0.625, 0.25, 0.5, 0.375, 0.375, NaN, 0.5, 0.5, 0.625, 0.375, 0.5, 0.375, 0.625, 0.625, 0.375, 0.75, 0.625, 0.75, 1.0, 0.625, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0625, 0.0, 0.0, 0.0, 0.125, 0.125, 0.1875, 0.125, 0.0, 0.0, 0.0, 0.25, 0.1875, 0.125, 0.125, 0.125, 0.25, 0.25, 0.125, 0.125, 0.25, 0.125, 0.3125, 0.25, 0.0, 0.25, 0.5, 0.375, 0.375, 0.375, 0.125, 0.25, 0.375, 0.625, 0.625, 0.125, 0.375, 0.375, 0.375, 0.5, NaN, 0.5, 0.625, 0.375, 0.5, 0.375, 0.625, 0.625, 0.375, 0.75, 0.625, 0.75, 1.0, 0.625, 0.875, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.375, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.25, 0.125, 0.0, 0.375, 0.0, 0.25, 0.375, 0.375, 0.25, 0.125, 0.25, 0.25, 0.0, 0.25, 0.5, 0.5, 0.375, 0.625, 0.375, 0.625, 0.5, 0.5, NaN, 0.625, 0.625, 0.6875, 0.625, 0.625, 0.625, 0.625, 0.625, 0.75, 0.75, 0.625, 1.0, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.25, 0.0, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.25, 0.25, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.375, 0.25, 0.25, 0.375, 0.375, 0.375, NaN, 0.25, 0.375, 0.25, 0.875, 0.375, 0.375, 0.375, 0.5, 1.0, 0.5, 0.375, 0.5, 0.625, 0.625, 0.75, 0.75, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.25, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.375, 0.125, 0.375, 0.375, 0.375, 0.125, 0.125, 0.125, 0.375, 0.375, 0.125, 0.25, 0.375, 0.625, 0.625, 0.375, 0.375, 0.625, 0.625, 0.375, 0.75, NaN, 0.375, 0.5, 0.75, 0.5, 0.75, 0.75, 0.625, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.25, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.375, 0.125, 0.0, 0.375, 0.125, 0.125, 0.25, 0.375, 0.25, 0.125, 0.25, 0.125, 0.125, 0.25, 0.25, 0.5, 0.375, 0.625, 0.375, 0.625, 0.5, 0.5, 0.3125, 0.625, 0.625, NaN, 0.5, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.625, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.0, 0.0, 0.125, 0.25, 0.125, 0.0, 0.125, 0.375, 0.125, 0.375, 0.0, 0.375, 0.125, 0.0, 0.125, 0.5, 0.25, 0.25, 0.375, 0.375, 0.375, 0.625, 0.25, 0.375, 0.625, 0.625, 0.375, 0.75, 0.5, 0.5, NaN, 0.75, 0.625, 0.75, 0.625, 0.625, 0.75, 0.75, 0.875, 0.875, 1.0, 1.0, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.25, 0.125, 0.125, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.125, 0.25, 0.25, 0.125, 0.375, 0.25, 0.25, 0.25, 0.25, 0.375, 0.25, 0.25, 0.25, 0.25, 0.25, 0.125, 0.25, 0.375, 0.25, 0.25, 0.25, 0.375, 0.375, 0.375, 0.125, 0.25, 0.375, 0.25, NaN, 0.375, 0.375, 0.375, 0.5, 1.0, 0.5, 0.375, 0.5, 0.625, 0.625, 0.75, 0.75, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.25, 0.125, 0.125, 0.25, 0.0, 0.25, 0.125, 0.25, 0.125, 0.25, 0.25, 0.25, 0.25, 0.25, 0.0, 0.125, 0.375, 0.375, 0.375, 0.25, 0.375, 0.375, 0.375, 0.625, 0.5, 0.25, 0.375, 0.625, NaN, 0.75, 0.5, 0.625, 0.875, 0.75, 0.875, 0.875, 0.875, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.25, 0.375, 0.125, 0.375, 0.125, 0.25, 0.125, 0.125, 0.0, 0.5, 0.375, 0.0, 0.25, 0.25, 0.5, 0.5, 0.25, 0.25, 0.625, 0.625, 0.375, 0.625, 0.25, 0.375, 0.25, 0.625, 0.25, NaN, 0.625, 0.25, 0.75, 0.625, 0.5, 0.75, 0.75, 0.75, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0625, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0625, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1875, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.1875, 0.125, 0.0, 0.125, 0.375, 0.25, 0.125, 0.125, 0.0, 0.125, 0.375, 0.375, 0.375, 0.0, 0.25, 0.125, 0.125, 0.25, 0.25, 0.375, 0.625, 0.25, 0.375, 0.375, 0.625, 0.5, 0.375, NaN, 0.5, 0.75, 0.75, 0.625, 0.75, 1.0, 1.0, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.375, 0.0, 0.125, 0.0, 0.375, 0.0, 0.125, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.25, 0.375, 0.125, 0.0, 0.375, 0.375, 0.25, 0.5, 0.375, 0.25, 0.375, 0.5, 0.375, 0.75, 0.5, NaN, 0.5, 0.5, 0.75, 0.625, 0.75, 0.875, 1.0, 1.0, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.375, 0.125, 0.125, 0.0, 0.125, 0.25, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.25, 0.25, 0.25, 0.0, 0.125, 0.25, 0.25, 0.0, 0.125, 0.25, 0.25, 0.5, NaN, 0.5, 0.25, 0.375, 0.5, 0.5, 0.75, 0.75, 0.75], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.125, 0.0, 0.125, 0.25, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.375, 0.5, 0.25, 0.375, 0.25, 0.5, 0.25, 0.375, 0.25, 0.5, 0.5, NaN, 0.5, 0.25, 1.0, 1.0, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.125, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.375, 0.25, 0.125, 0.125, 0.375, 0.375, 0.0, 0.625, 0.125, 0.125, 0.125, 0.625, 0.125, 0.5, 0.375, 0.25, 0.75, 0.5, NaN, 0.5, 0.625, 0.75, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.125, 0.0, 0.0, 0.25, 0.125, 0.125, 0.0, 0.125, 0.0, 0.125, 0.0, 0.125, 0.125, 0.5, 0.125, 0.125, 0.125, 0.5, 0.125, 0.25, 0.25, 0.375, 0.625, 0.75, 0.5, NaN, 0.75, 0.75, 0.875, 0.875, 0.875], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.125, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.125, 0.375, 0.125, 0.125, 0.0, 0.375, 0.125, 0.25, 0.0, 0.25, 0.5, 0.0, 0.375, 0.25, NaN, 0.75, 0.875, 0.875, 1.0], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.125, 0.125, 0.125, 0.125, 0.0, 0.0, 0.125, 0.125, 0.0, 0.0, 0.125, 0.375, 0.125, 0.125, 0.0, 0.375, 0.125, 0.25, 0.0, 0.125, 0.5, 0.0, 0.25, 0.25, 0.25, NaN, 0.625, 0.625, 0.625], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.125, 0.25, 0.0, 0.125, 0.0, 0.0, 0.25, 0.125, 0.125, 0.125, 0.125, 0.375, NaN, 0.6875, 0.75], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.125, 0.25, 0.0, 0.125, 0.0, 0.0, 0.25, 0.125, 0.125, 0.125, 0.125, 0.375, 0.3125, NaN, 0.75], [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.125, 0.0, 0.0, 0.0, 0.0, 0.25, 0.0, 0.0, 0.125, 0.0, 0.375, 0.25, 0.25, NaN]], "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"method":"TabFM (default)","model":"TabFM","variant":"Default","family":"Foundation Model","mean":0.9666666667},{"method":"Nori-30M (default)","model":"Nori-30M","variant":"Default","family":"Foundation Model","mean":0.95},{"method":"Nori (default)","model":"Nori","variant":"Default","family":"Foundation Model","mean":0.9316666667},{"method":"AutoGluon 1.5 (extreme, 4h)","model":"AutoGluon 1.5 (extreme, 4h)","variant":"","family":"Reference Pipeline","mean":0.925},{"method":"TabPFN-3 (default)","model":"TabPFN-3","variant":"Default","family":"Foundation Model","mean":0.9083333333},{"method":"TabPFN-2.6 (default)","model":"TabPFN-2.6","variant":"Default","family":"Foundation Model","mean":0.9},{"method":"RealTabPFN-2.5 (tuned + ensembled)","model":"RealTabPFN-2.5","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.875},{"method":"TabDPT (tuned + ensembled)","model":"TabDPT","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.875},{"method":"TabDPT (tuned)","model":"TabDPT","variant":"Tuned","family":"Foundation Model","mean":0.8433333333},{"method":"TabICLv2 (default)","model":"TabICLv2","variant":"Default","family":"Foundation Model","mean":0.7883333333},{"method":"RealTabPFN-2.5 (tuned)","model":"RealTabPFN-2.5","variant":"Tuned","family":"Foundation Model","mean":0.7766666667},{"method":"TabDPT (default)","model":"TabDPT","variant":"Default","family":"Foundation Model","mean":0.7733333333},{"method":"RealMLP (tuned + ensembled)","model":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.7516666667},{"method":"TabDPT-Turbo (default)","model":"TabDPT-Turbo","variant":"Default","family":"Foundation Model","mean":0.7416666667},{"method":"AutoGluon 1.4 (best, 4h)","model":"AutoGluon 1.4 (best, 4h)","variant":"","family":"Reference Pipeline","mean":0.7416666667},{"method":"RealTabPFN-2.5 (default)","model":"RealTabPFN-2.5","variant":"Default","family":"Foundation Model","mean":0.74},{"method":"TabPFNv2 (tuned + ensembled)","model":"TabPFNv2","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.7358333333},{"method":"TabSwift (default)","model":"TabSwift","variant":"Default","family":"Foundation Model","mean":0.7083333333},{"method":"iLTM (tuned + ensembled)","model":"iLTM","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.6883333333},{"method":"LightGBM (tuned + ensembled)","model":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6666666667},{"method":"RealMLP (tuned)","model":"RealMLP","variant":"Tuned","family":"Neural Network","mean":0.6583333333},{"method":"ModernNCA (tuned + ensembled)","model":"ModernNCA","variant":"Tuned + Ens.","family":"Neural Network","mean":0.64},{"method":"Mitra (default)","model":"Mitra","variant":"Default","family":"Foundation Model","mean":0.6391666667},{"method":"TabM (tuned + ensembled)","model":"TabM","variant":"Tuned + Ens.","family":"Neural Network","mean":0.6383333333},{"method":"CatBoost (tuned + ensembled)","model":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.6383333333},{"method":"CatBoost (tuned)","model":"CatBoost","variant":"Tuned","family":"Tree-based","mean":0.635},{"method":"xRFM (tuned + ensembled)","model":"xRFM","variant":"Tuned + Ens.","family":"Other","mean":0.615},{"method":"TabM (tuned)","model":"TabM","variant":"Tuned","family":"Neural Network","mean":0.5883333333},{"method":"TabPFNv2 (tuned)","model":"TabPFNv2","variant":"Tuned","family":"Foundation Model","mean":0.5858333333},{"method":"xRFM (tuned)","model":"xRFM","variant":"Tuned","family":"Other","mean":0.58},{"method":"XGBoost (tuned)","model":"XGBoost","variant":"Tuned","family":"Tree-based","mean":0.5766666667},{"method":"XGBoost (tuned + ensembled)","model":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.57},{"method":"LightGBM (tuned)","model":"LightGBM","variant":"Tuned","family":"Tree-based","mean":0.5516666667},{"method":"ChimeraBoost (tuned + ensembled)","model":"ChimeraBoost","variant":"Tuned + Ens.","family":"Tree-based","mean":0.5366666667},{"method":"ModernNCA (default)","model":"ModernNCA","variant":"Default","family":"Neural Network","mean":0.5316666667},{"method":"SAP-RPT-OSS (default)","model":"SAP-RPT-OSS","variant":"Default","family":"Foundation Model","mean":0.53},{"method":"iLTM (tuned)","model":"iLTM","variant":"Tuned","family":"Foundation Model","mean":0.52},{"method":"CatBoost (default)","model":"CatBoost","variant":"Default","family":"Tree-based","mean":0.5166666667},{"method":"TabPFNv2 (default)","model":"TabPFNv2","variant":"Default","family":"Foundation Model","mean":0.5141666667},{"method":"ChimeraBoost (tuned)","model":"ChimeraBoost","variant":"Tuned","family":"Tree-based","mean":0.4733333333},{"method":"ExtraTrees (tuned + ensembled)","model":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","mean":0.4616666667},{"method":"RealMLP (default)","model":"RealMLP","variant":"Default","family":"Neural Network","mean":0.46},{"method":"LimiX (default)","model":"LimiX","variant":"Default","family":"Foundation Model","mean":0.4566666667},{"method":"ModernNCA (tuned)","model":"ModernNCA","variant":"Tuned","family":"Neural Network","mean":0.4483333333},{"method":"TabM (default)","model":"TabM","variant":"Default","family":"Neural Network","mean":0.4466666667},{"method":"TorchMLP (tuned + ensembled)","model":"TorchMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.44},{"method":"ExtraTrees (tuned)","model":"ExtraTrees","variant":"Tuned","family":"Tree-based","mean":0.4316666667},{"method":"ChimeraBoost (default)","model":"ChimeraBoost","variant":"Default","family":"Tree-based","mean":0.4016666667},{"method":"TorchMLP (tuned)","model":"TorchMLP","variant":"Tuned","family":"Neural Network","mean":0.3983333333},{"method":"EBM (tuned + ensembled)","model":"EBM","variant":"Tuned + Ens.","family":"Tree-based","mean":0.385},{"method":"EBM (tuned)","model":"EBM","variant":"Tuned","family":"Tree-based","mean":0.3633333333},{"method":"RandomForest (tuned + ensembled)","model":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","mean":0.3516666667},{"method":"xRFM (default)","model":"xRFM","variant":"Default","family":"Other","mean":0.35},{"method":"LightGBM (default)","model":"LightGBM","variant":"Default","family":"Tree-based","mean":0.3483333333},{"method":"FastaiMLP (tuned + ensembled)","model":"FastaiMLP","variant":"Tuned + Ens.","family":"Neural Network","mean":0.32},{"method":"ExtraTrees (default)","model":"ExtraTrees","variant":"Default","family":"Tree-based","mean":0.3116666667},{"method":"RandomForest (tuned)","model":"RandomForest","variant":"Tuned","family":"Tree-based","mean":0.305},{"method":"PerpetualBooster (tuned + ensembled)","model":"PerpetualBooster","variant":"Tuned + Ens.","family":"Tree-based","mean":0.2958333333},{"method":"TabSTAR (tuned + ensembled)","model":"TabSTAR","variant":"Tuned + Ens.","family":"Foundation Model","mean":0.2933333333},{"method":"FastaiMLP (tuned)","model":"FastaiMLP","variant":"Tuned","family":"Neural Network","mean":0.2916666667},{"method":"PerpetualBooster (tuned)","model":"PerpetualBooster","variant":"Tuned","family":"Tree-based","mean":0.2825},{"method":"XGBoost (default)","model":"XGBoost","variant":"Default","family":"Tree-based","mean":0.2816666667},{"method":"TabSTAR (tuned)","model":"TabSTAR","variant":"Tuned","family":"Foundation Model","mean":0.2583333333},{"method":"EBM (default)","model":"EBM","variant":"Default","family":"Tree-based","mean":0.2483333333},{"method":"TorchMLP (default)","model":"TorchMLP","variant":"Default","family":"Neural Network","mean":0.2316666667},{"method":"RandomForest (default)","model":"RandomForest","variant":"Default","family":"Tree-based","mean":0.2183333333},{"method":"FastaiMLP (default)","model":"FastaiMLP","variant":"Default","family":"Neural Network","mean":0.1983333333},{"method":"TabSTAR (default)","model":"TabSTAR","variant":"Default","family":"Foundation Model","mean":0.16},{"method":"KNN (tuned + ensembled)","model":"KNN","variant":"Tuned + Ens.","family":"Baseline","mean":0.16},{"method":"PerpetualBooster (default)","model":"PerpetualBooster","variant":"Default","family":"Tree-based","mean":0.1533333333},{"method":"iLTM (default)","model":"iLTM","variant":"Default","family":"Foundation Model","mean":0.1316666667},{"method":"KNN (tuned)","model":"KNN","variant":"Tuned","family":"Baseline","mean":0.0983333333},{"method":"KNN (default)","model":"KNN","variant":"Default","family":"Baseline","mean":0.08},{"method":"Linear (tuned + ensembled)","model":"Linear","variant":"Tuned + Ens.","family":"Baseline","mean":0.0475},{"method":"Linear (tuned)","model":"Linear","variant":"Tuned","family":"Baseline","mean":0.0425},{"method":"Linear (default)","model":"Linear","variant":"Default","family":"Baseline","mean":0.02}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  // POINTS is one record per matrix label (its model, variant, family and mean
+  // win rate); the matrix itself is CONFIG.matrix, row-major over CONFIG.methods.
+  const METHODS = CONFIG.methods;
+  const MATRIX = CONFIG.matrix;
+  const rowOf = new Map(METHODS.map((m, i) => [m, i]));
+  const info = new Map(POINTS.map(p => [p.method, p]));
+  const MODELS = [...new Set(POINTS.map(p => p.model))];
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+
+  // Reference pipelines are AutoML systems rather than models, so a first read of
+  // the matrix should be model against model; they are one chip click away. Falls
+  // back to everything if that would leave nothing to compare.
+  const OFF_BY_DEFAULT = new Set(["Reference Pipeline"]);
+  const familyOf = new Map(POINTS.map(p => [p.model, p.family]));
+  const DEFAULT_MODELS = MODELS.filter(m => !OFF_BY_DEFAULT.has(familyOf.get(m)));
+
+  const state = {
+    models: new Set(DEFAULT_MODELS.length > 1 ? DEFAULT_MODELS : MODELS),
+    variants: new Set(VARIANTS),
+    // One entry per model by default, as the static figure shows it: 80 rows of
+    // model-variant pairs is a wall, and the comparison people want first is
+    // between models.
+    best: true,
+    order: "published",
+    // Default to fitting the panel: at reading size the matrix is taller than a
+    // screen, and the shape of the heatmap is what most readers want first.
+    zoom: "fit",
+  };
+  // Ceiling on the fitted height. Fitting is normally width-driven; this only bites
+  // on very tall selections, and it is set where the labels are still readable
+  // rather than at half a screen, which would shrink 30 rows to 8px type.
+  const FIT_H = 1040;
+  // Nominal (drawn) size of the last render; the display size is a CSS scale of it.
+  let nomW = 0, nomH = 0;
+  const svg = document.getElementById("chart");
+  const wrap = document.getElementById("wrap");
+  const tooltip = document.querySelector(".tooltip");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByModel = new Map();
+
+  // Sized against the static figure rather than the other explorers: that one uses
+  // 16pt tick labels and 18pt axis labels on generously sized cells, and it stayed
+  // the easier of the two to read. Weight still matches the explorers' 650, so a
+  // model name looks like the same name everywhere.
+  const LABEL_SIZE = 16, LABEL_WEIGHT = 650, CAPTION_SIZE = 18, CELL = 42, VALUE_SIZE = 15;
+  // The colour key is a vertical bar to the right of the matrix, as on the static
+  // figure. It sits beside the rows rather than above them, so it is in view
+  // wherever the reader is in a tall matrix.
+  const BAR_W = 30, BAR_GAP = 32, TICK_SIZE = 15;
+  const SCALE_CAPTION = "win rate of the row over the column";
+  const SCALE_CAPTION_SHORT = "win rate";
+
+  // Purple (the column wins) through white to green (the row wins), the same
+  // diverging reading as the static figure's PRGn colormap.
+  const LOSE = [118, 42, 131], MID = [247, 247, 247], WIN = [27, 120, 55];
+  function cellColor(rate) {
+    if (rate == null || !isFinite(rate)) return "#8884";
+    const t = Math.max(0, Math.min(1, rate));
+    const [a, b, u] = t < 0.5 ? [LOSE, MID, t / 0.5] : [MID, WIN, (t - 0.5) / 0.5];
+    return "rgb(" + a.map((v, i) => Math.round(v + u * (b[i] - v))).join(",") + ")";
+  }
+
+  // Measured where possible: the labels are model names of wildly different
+  // lengths, and the margins have to clear the longest one exactly or a rotated
+  // column label runs into the caption above it. `getComputedTextLength` returns 0
+  // when the frame has not been laid out yet (it is lazily loaded, and may still be
+  // off-screen on the first paint), which would collapse both margins and push
+  // every label outside the viewBox — invisible until something forced a redraw.
+  // So fall back to an estimate, and redraw once the fonts have settled.
+  function textWidth(text, size, weight) {
+    const probe = el("text", {
+      "font-size": size || LABEL_SIZE, "font-weight": weight || LABEL_WEIGHT,
+    }, svg);
+    probe.textContent = text;
+    let width = 0;
+    try { width = probe.getComputedTextLength(); } catch (e) { width = 0; }
+    probe.remove();
+    return width > 0 ? width : text.length * (size || LABEL_SIZE) * 0.58;
+  }
+
+  // Labels always carry their variant tag — "(default)" / "(tuned + ensembled)" —
+  // including in one-per-model mode, where the tag is precisely what says which
+  // variant survived the filter. The static figure labels them the same way.
+  function labelText(label) {
+    return label;
+  }
+
+  function shown() {
+    let list = METHODS.filter(label => {
+      const p = info.get(label);
+      if (!p || !state.models.has(p.model)) return false;
+      return !p.variant || state.variants.has(p.variant);
+    });
+    if (state.best) {
+      // Best *among the variants still selected*, so the variant toggles keep
+      // meaning something in this mode.
+      const pick = new Map();
+      for (const label of list) {
+        const p = info.get(label);
+        const held = pick.get(p.model);
+        if (!held || (info.get(held).mean || 0) < (p.mean || 0)) pick.set(p.model, label);
+      }
+      list = list.filter(label => pick.get(info.get(label).model) === label);
+    }
+    if (state.order === "winrate") {
+      list.sort((a, b) => (info.get(b).mean || 0) - (info.get(a).mean || 0));
+    } else if (state.order === "name") {
+      list.sort((a, b) => labelText(a).localeCompare(labelText(b)));
+    }
+    return list;
+  }
+
+  function render() {
+    const list = shown();
+    svg.innerHTML = "";
+    if (list.length < 2) {
+      svg.setAttribute("width", 0);
+      svg.setAttribute("height", 0);
+      // Also drop the display scale, or the empty chart keeps the last figure's box.
+      svg.style.width = svg.style.height = "";
+      nomW = nomH = 0;
+      wrap.querySelector(".wr-empty") ||
+        wrap.insertAdjacentHTML("beforeend",
+          '<p class="wr-empty">Select at least two models to compare.</p>');
+      postHeight();
+      return;
+    }
+    const empty = wrap.querySelector(".wr-empty");
+    if (empty) empty.remove();
+
+    const widths = list.map(label => textWidth(labelText(label)));
+    const longest = Math.max(...widths);
+    // A label rotated -60° rises 0.866 of its length; the captions sit in the
+    // corner box rather than above the columns, where long names would cross them.
+    // Floor the header so the two corner captions always have room, even when
+    // every selected name is short.
+    const headerH = Math.max(Math.ceil(longest * 0.866) + 14, 2 * CAPTION_SIZE + 26);
+    const labelW = Math.ceil(longest) + 16;
+    const n = list.length;
+    // A -60° label anchored at (x, y) reaches (x + 0.5L, y - 0.866L): it grows up
+    // *and right*. headerH covers the rise; this covers the run, or the rightmost
+    // column's name is clipped by half its length.
+    const overhang = Math.ceil(longest * 0.5);
+    const matrixRight = labelW + n * CELL;
+    // The key occupies the band to the right of the matrix; the label overhang is
+    // above the matrix, so the two never collide and the wider of the two wins.
+    const tickW = Math.ceil(textWidth("100%", TICK_SIZE, 700));
+    const barX = matrixRight + BAR_GAP;
+    const keyRight = barX + BAR_W + 12 + tickW + 10 + CAPTION_SIZE;
+    const width = Math.max(matrixRight + overhang + 12, keyRight + 10);
+    const height = headerH + n * CELL + 12;
+    svg.setAttribute("width", width);
+    svg.setAttribute("height", height);
+    svg.setAttribute("viewBox", "0 0 " + width + " " + height);
+
+    // Captions, stacked in the corner box and bold, as on the static figure.
+    const capB = el("text", {
+      x: 4, y: headerH - 34, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capB.textContent = "Model B: loser →";
+    const capA = el("text", {
+      x: 4, y: headerH - 14, "font-size": CAPTION_SIZE, "font-weight": 700, fill: "var(--ink)",
+    }, svg);
+    capA.textContent = "Model A: winner ↓";
+
+    list.forEach((label, j) => {
+      const x = labelW + j * CELL + CELL / 2 + 4;
+      const text = el("text", {
+        x: x, y: headerH - 8, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "start", fill: FAM_INK[info.get(label).family] || "var(--ink)",
+        transform: "rotate(-60 " + x + " " + (headerH - 8) + ")", cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(label);
+      text.addEventListener("click", () => toggleModel(info.get(label).model));
+    });
+
+    list.forEach((rowLabel, i) => {
+      const y = headerH + i * CELL;
+      const text = el("text", {
+        x: labelW - 12, y: y + CELL / 2 + 6, "font-size": LABEL_SIZE, "font-weight": LABEL_WEIGHT,
+        "text-anchor": "end", fill: FAM_INK[info.get(rowLabel).family] || "var(--ink)",
+        cursor: "pointer",
+      }, svg);
+      text.textContent = labelText(rowLabel);
+      text.addEventListener("click", () => toggleModel(info.get(rowLabel).model));
+
+      list.forEach((colLabel, j) => {
+        const rate = valueAt(rowLabel, colLabel);
+        const x = labelW + j * CELL;
+        const same = rowLabel === colLabel;
+        const rect = el("rect", {
+          x: x, y: y, width: CELL - 1, height: CELL - 1, rx: 2,
+          fill: same ? "#8883" : cellColor(rate),
+        }, svg);
+        if (same) return;
+        const value = el("text", {
+          x: x + (CELL - 1) / 2, y: y + CELL / 2 + 5.5, "font-size": VALUE_SIZE,
+          "font-weight": 600, "text-anchor": "middle",
+          fill: Math.abs(rate - 0.5) > 0.3 ? "#ffffff" : "#14161a", "pointer-events": "none",
+        }, svg);
+        value.textContent = rate == null || !isFinite(rate) ? "" : Math.round(rate * 100);
+        rect.addEventListener("mousemove", ev => showTip(ev, rowLabel, colLabel, rate));
+        rect.addEventListener("mouseleave", () => { tooltip.style.display = "none"; });
+      });
+    });
+
+    drawScale(barX, headerH, n * CELL - 2, tickW);
+    nomW = width;
+    nomH = height;
+    applyZoom();
+  }
+
+  // Scale the figure for display only: the width/height *attributes* stay nominal,
+  // so the export helpers keep building full-size figures, and the CSS box drives
+  // what the page shows (the viewBox does the scaling, so it stays vector-sharp).
+  function applyZoom() {
+    if (!nomW || !nomH) return;
+    let scale = Number(state.zoom);
+    if (!(scale > 0)) {
+      const availW = Math.max(240, (wrap.clientWidth || nomW) - 4);
+      scale = Math.min(1, availW / nomW, FIT_H / nomH);
+    }
+    svg.style.width = Math.round(nomW * scale) + "px";
+    svg.style.height = Math.round(nomH * scale) + "px";
+    postHeight();
+  }
+
+  // The colour key, drawn into the figure so it is there in paper view and in an
+  // exported SVG/PNG/PDF, not only beside the live chart.
+  function drawScale(x, y, barH, tickW) {
+    const id = "wr-ramp";
+    const grad = el("linearGradient", { id: id, x1: "0", x2: "0", y1: "0", y2: "1" },
+      el("defs", {}, svg));
+    // Top of the bar is the row winning outright, the bottom is the column winning.
+    for (const [offset, rate] of [[0, 1], [0.5, 0.5], [1, 0]]) {
+      el("stop", { offset: offset, "stop-color": cellColor(rate) }, grad);
+    }
+    el("rect", {
+      x: x, y: y, width: BAR_W, height: barH, rx: 7,
+      fill: "url(#" + id + ")", stroke: "var(--muted)", "stroke-width": 1,
+    }, svg);
+    for (const [frac, label] of [[0, "100%"], [0.5, "50%"], [1, "0%"]]) {
+      const ty = y + frac * barH;
+      el("line", {
+        x1: x + BAR_W, y1: ty, x2: x + BAR_W + 6, y2: ty,
+        stroke: "var(--muted)", "stroke-width": 1.5,
+      }, svg);
+      const tick = el("text", {
+        x: x + BAR_W + 11, y: ty + TICK_SIZE * 0.36, "font-size": TICK_SIZE,
+        "font-weight": 700, "text-anchor": "start", fill: "var(--ink)",
+      }, svg);
+      tick.textContent = label;
+    }
+    // The label reads down the bar. It shortens rather than growing past the bar
+    // when only a couple of models are selected: anything above the bar would run
+    // into the rightmost column's rotated name.
+    const size = CAPTION_SIZE - 2;
+    const cx = x + BAR_W + 11 + tickW + 10 + size * 0.5;
+    const cy = y + barH / 2;
+    const caption = el("text", {
+      x: cx, y: cy, "font-size": size, "font-weight": 700,
+      "text-anchor": "middle", fill: "var(--ink)",
+      transform: "rotate(90 " + cx + " " + cy + ")",
+    }, svg);
+    caption.textContent =
+      textWidth(SCALE_CAPTION, size, 700) <= barH ? SCALE_CAPTION : SCALE_CAPTION_SHORT;
+  }
+
+  function valueAt(rowLabel, colLabel) {
+    const i = rowOf.get(rowLabel), j = rowOf.get(colLabel);
+    if (i == null || j == null) return null;
+    const row = MATRIX[i];
+    return row ? row[j] : null;
+  }
+
+  function showTip(ev, rowLabel, colLabel, rate) {
+    tooltip.style.display = "block";
+    tooltip.innerHTML =
+      '<div class="t-name">' + rowLabel + " vs " + colLabel + "</div>" +
+      "<div>" + rowLabel + " wins " + fmtNum(rate * 100, 1) + "% of tasks</div>" +
+      '<div class="t-var">' + colLabel + " wins " + fmtNum((1 - rate) * 100, 1) + "%</div>";
+    const box = document.body.getBoundingClientRect();
+    tooltip.style.left = (ev.clientX - box.left + 14) + "px";
+    tooltip.style.top = (ev.clientY - box.top + 12) + "px";
+  }
+
+  // ---------- model chips, by family ----------
+  function familyModels(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.model))];
+  }
+  function meanOfModel(model) {
+    const values = POINTS.filter(p => p.model === model && isFinite(p.mean)).map(p => p.mean);
+    return values.length ? Math.max(...values) : 0;
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models compared — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyModels(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => meanOfModel(b) - meanOfModel(a));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const model of members) {
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span></span>';
+        b.lastChild.textContent = model;
+        b.title = model + " — best mean win rate " + fmtNum(meanOfModel(model) * 100, 1) + "%";
+        b.addEventListener("click", () => toggleModel(model));
+        set.appendChild(b);
+        chipByModel.set(model, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [model, b] of chipByModel) b.setAttribute("aria-pressed", String(state.models.has(model)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyModels(fam).every(m => state.models.has(m))));
+    }
+  }
+  function toggleModel(model) {
+    if (state.models.has(model)) state.models.delete(model); else state.models.add(model);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyModels(fam);
+    const allOn = members.every(m => state.models.has(m));
+    for (const m of members) { if (allOn) state.models.delete(m); else state.models.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>';
+      // Spelled out: the matrix labels carry the long form the data uses, so the
+      // toggle should not be the only place showing the abbreviated key.
+      b.appendChild(document.createTextNode(VARIANT_BTN[v] || v));
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + (VARIANT_BTN[v] || v).toLowerCase() + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("zoom-select").addEventListener("change", ev => {
+    state.zoom = ev.target.value;
+    applyZoom();
+  });
+  window.addEventListener("resize", applyZoom);
+  // Embedded in a lazily shown panel, the first render can measure a zero-width
+  // panel; re-fit once it has a width. Width changes only, so the height applyZoom
+  // itself produces cannot feed back into another fit.
+  if (window.ResizeObserver) {
+    let lastWidth = 0;
+    new ResizeObserver(() => {
+      const w = wrap.clientWidth;
+      if (Math.abs(w - lastWidth) < 2) return;
+      lastWidth = w;
+      applyZoom();
+    }).observe(wrap);
+  }
+  document.getElementById("order-select").addEventListener("change", ev => {
+    state.order = ev.target.value;
+    render();
+  });
+  const bestBtn = document.getElementById("btn-best");
+  bestBtn.style.setProperty("--fam", "var(--accent)");
+  bestBtn.addEventListener("click", () => {
+    state.best = !state.best;
+    bestBtn.setAttribute("aria-pressed", String(state.best));
+    render();
+  });
+  bestBtn.setAttribute("aria-pressed", String(state.best));
+  setUpPaperView(render);
+  setUpExport(() => [{ svg: svg, dx: 0 }], () => slugify(CONFIG.title || document.title));
+  buildVariantBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  // The first render may have had to estimate label widths (see textWidth); redraw
+  // with real measurements as soon as the fonts are in.
+  if (document.fonts && document.fonts.ready) document.fonts.ready.then(render);
+})();
+</script>
+</body>
+</html>
diff --git a/data/imputation_yes/splits_lite/tasks_regression/datasets_small/winrate_matrix.csv b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/winrate_matrix.csv
new file mode 100644
index 0000000000000000000000000000000000000000..413c8abb2858af1668c33a5e2a5ee8bdda5669e9
--- /dev/null
+++ b/data/imputation_yes/splits_lite/tasks_regression/datasets_small/winrate_matrix.csv
@@ -0,0 +1,77 @@
+,TabFM (default),Nori-30M (default),Nori (default),"AutoGluon 1.5 (extreme, 4h)",TabPFN-3 (default),TabPFN-2.6 (default),RealTabPFN-2.5 (tuned + ensembled),TabDPT (tuned + ensembled),TabDPT (tuned),TabICLv2 (default),RealTabPFN-2.5 (tuned),TabDPT (default),RealMLP (tuned + ensembled),TabDPT-Turbo (default),"AutoGluon 1.4 (best, 4h)",RealTabPFN-2.5 (default),TabPFNv2 (tuned + ensembled),TabSwift (default),iLTM (tuned + ensembled),LightGBM (tuned + ensembled),RealMLP (tuned),ModernNCA (tuned + ensembled),Mitra (default),TabM (tuned + ensembled),CatBoost (tuned + ensembled),CatBoost (tuned),xRFM (tuned + ensembled),TabM (tuned),TabPFNv2 (tuned),xRFM (tuned),XGBoost (tuned),XGBoost (tuned + ensembled),LightGBM (tuned),ChimeraBoost (tuned + ensembled),ModernNCA (default),SAP-RPT-OSS (default),iLTM (tuned),CatBoost (default),TabPFNv2 (default),ChimeraBoost (tuned),ExtraTrees (tuned + ensembled),RealMLP (default),LimiX (default),ModernNCA (tuned),TabM (default),TorchMLP (tuned + ensembled),ExtraTrees (tuned),ChimeraBoost (default),TorchMLP (tuned),EBM (tuned + ensembled),EBM (tuned),RandomForest (tuned + ensembled),xRFM (default),LightGBM (default),FastaiMLP (tuned + ensembled),ExtraTrees (default),RandomForest (tuned),PerpetualBooster (tuned + ensembled),TabSTAR (tuned + ensembled),FastaiMLP (tuned),PerpetualBooster (tuned),XGBoost (default),TabSTAR (tuned),EBM (default),TorchMLP (default),RandomForest (default),FastaiMLP (default),TabSTAR (default),KNN (tuned + ensembled),PerpetualBooster (default),iLTM (default),KNN (tuned),KNN (default),Linear (tuned + ensembled),Linear (tuned),Linear (default)
+TabFM (default),,0.75,0.875,0.625,0.75,0.75,0.75,0.875,1.0,0.875,0.75,1.0,1.0,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori-30M (default),0.25,,0.625,0.75,0.5,0.75,0.75,0.625,1.0,0.875,0.875,0.875,1.0,1.0,1.0,0.875,0.75,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Nori (default),0.125,0.375,,0.5,0.5,0.625,0.625,0.75,1.0,0.875,0.75,0.875,1.0,1.0,0.875,0.75,0.625,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.5 (extreme, 4h)",0.375,0.25,0.5,,0.5,0.625,0.5,0.625,0.75,0.625,0.75,0.875,1.0,1.0,1.0,0.875,0.75,1.0,1.0,1.0,1.0,0.875,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-3 (default),0.25,0.5,0.5,0.5,,0.5,0.375,0.5,0.625,0.625,0.75,1.0,0.875,1.0,0.875,0.75,0.875,0.875,1.0,1.0,0.875,0.875,0.875,1.0,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFN-2.6 (default),0.25,0.25,0.375,0.375,0.5,,0.625,0.625,0.625,0.75,0.75,0.75,0.875,0.75,0.75,0.875,0.625,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned + ensembled),0.25,0.25,0.375,0.5,0.625,0.375,,0.625,0.625,0.625,0.875,0.75,0.875,0.875,0.75,0.875,0.625,0.75,0.875,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned + ensembled),0.125,0.375,0.25,0.375,0.5,0.375,0.375,,0.75,0.625,0.5,0.875,0.875,0.875,0.875,0.75,0.5,1.0,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (tuned),0.0,0.0,0.0,0.25,0.375,0.375,0.375,0.25,,0.5,0.625,0.75,0.875,0.875,0.875,0.5,0.375,0.75,0.875,1.0,0.875,0.875,0.875,0.75,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,0.75,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabICLv2 (default),0.125,0.125,0.125,0.375,0.375,0.25,0.375,0.375,0.5,,0.75,0.5,0.625,0.75,0.75,0.625,0.5,0.75,0.875,0.875,0.875,0.75,0.75,0.75,0.875,0.875,0.75,0.75,0.625,0.75,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,1.0,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (tuned),0.25,0.125,0.25,0.25,0.25,0.25,0.125,0.5,0.375,0.25,,0.625,0.625,0.75,0.625,0.75,0.5,0.625,0.625,0.75,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.75,0.875,1.0,0.875,1.0,1.0,0.875,0.875,0.875,0.75,1.0,1.0,0.875,0.875,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT (default),0.0,0.125,0.125,0.125,0.0,0.25,0.25,0.125,0.25,0.5,0.375,,0.625,0.75,0.625,0.375,0.5,0.625,0.625,0.75,0.625,0.75,0.625,0.75,0.875,0.875,1.0,0.75,0.625,1.0,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,0.75,1.0,1.0,0.875,0.75,0.875,1.0,0.875,1.0,1.0,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.375,0.375,0.375,,0.625,0.5,0.5,0.375,0.625,0.75,0.75,0.75,0.625,0.625,0.625,0.875,0.875,0.75,0.75,0.625,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,0.625,1.0,1.0,1.0,0.75,0.875,1.0,0.875,1.0,0.875,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabDPT-Turbo (default),0.0,0.0,0.0,0.0,0.0,0.25,0.125,0.125,0.125,0.25,0.25,0.25,0.375,,0.625,0.25,0.375,0.625,0.625,0.75,0.875,0.625,0.625,0.5,0.875,0.75,1.0,0.75,0.625,0.875,1.0,1.0,1.0,0.875,0.75,0.875,0.875,0.875,0.5,0.875,1.0,0.875,0.75,0.875,1.0,1.0,1.0,0.875,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+"AutoGluon 1.4 (best, 4h)",0.0,0.0,0.125,0.0,0.125,0.25,0.25,0.125,0.125,0.25,0.375,0.375,0.5,0.375,,0.5,0.375,0.5,0.5,0.875,0.625,0.75,0.625,0.625,0.75,0.625,0.75,0.75,0.5,0.75,0.875,1.0,1.0,1.0,0.875,0.75,0.875,1.0,0.625,1.0,0.875,0.875,0.75,0.875,0.875,1.0,0.875,1.0,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealTabPFN-2.5 (default),0.125,0.125,0.25,0.125,0.25,0.125,0.125,0.25,0.5,0.375,0.25,0.625,0.5,0.75,0.5,,0.375,0.75,0.75,0.75,0.625,0.75,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.875,0.75,0.75,0.75,0.875,0.75,0.75,0.75,0.875,0.875,0.875,0.875,0.75,0.625,0.75,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.875,0.75,0.875,1.0,0.875,0.875,1.0,0.875,1.0,1.0,0.875,0.875,1.0,1.0,0.875,1.0,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0
+TabPFNv2 (tuned + ensembled),0.125,0.25,0.375,0.25,0.125,0.375,0.375,0.5,0.625,0.5,0.5,0.5,0.625,0.625,0.625,0.625,,0.625,0.625,0.625,0.625,0.625,0.8125,0.625,0.625,0.625,0.75,0.75,0.9375,0.75,0.625,0.75,0.625,0.75,0.625,0.875,0.75,0.75,0.9375,0.75,0.625,0.875,0.875,0.625,0.75,0.75,0.625,0.875,0.875,0.875,0.875,0.875,0.75,0.875,0.875,1.0,0.9375,0.875,0.75,0.875,0.875,0.875,0.875,1.0,0.875,0.9375,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabSwift (default),0.125,0.125,0.0,0.0,0.125,0.125,0.25,0.0,0.25,0.25,0.375,0.375,0.375,0.375,0.5,0.25,0.375,,0.625,0.625,0.625,0.375,0.625,0.625,0.75,0.75,0.625,0.75,0.625,0.875,0.875,0.75,0.75,0.75,0.875,0.875,0.75,0.75,0.5,0.75,0.75,0.875,0.75,1.0,0.875,0.875,0.75,0.875,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.375,0.375,0.25,0.375,0.5,0.25,0.375,0.375,,0.5,0.5,0.5,0.375,0.625,0.75,0.75,0.5,0.75,0.5,0.625,0.875,0.875,0.875,0.75,0.875,0.75,0.75,0.875,0.5,0.875,1.0,0.75,0.625,0.875,0.875,0.875,1.0,1.0,0.875,0.75,0.75,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.25,0.25,0.25,0.25,0.125,0.25,0.375,0.375,0.5,,0.5,0.375,0.5,0.625,0.5,0.5,0.5,0.625,0.375,0.625,0.75,0.75,1.0,0.875,0.875,0.625,0.625,0.875,0.5,1.0,0.875,0.75,0.75,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (tuned),0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.25,0.375,0.25,0.125,0.375,0.375,0.375,0.375,0.5,0.5,,0.375,0.625,0.5,0.625,0.75,0.625,0.625,0.375,0.75,0.75,0.75,0.75,0.625,0.75,0.625,0.75,0.625,0.5,0.75,0.75,0.875,0.75,0.875,1.0,0.75,0.75,0.75,1.0,1.0,0.875,0.75,0.875,0.875,1.0,0.75,0.875,1.0,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (tuned + ensembled),0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.375,0.25,0.375,0.375,0.25,0.25,0.375,0.625,0.5,0.625,0.625,,0.5,0.5,0.375,0.375,0.5,0.5,0.5,0.5,0.625,0.75,0.875,0.75,0.625,0.625,0.75,0.625,0.625,0.75,0.75,0.625,0.625,1.0,0.75,0.625,0.875,0.75,0.75,0.75,0.75,0.875,0.875,0.75,1.0,0.875,0.875,0.875,0.75,1.0,0.875,0.875,0.75,0.875,0.875,1.0,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+Mitra (default),0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.1875,0.375,0.625,0.5,0.375,0.5,,0.625,0.5,0.5,0.625,0.625,0.4375,0.75,0.625,0.625,0.75,0.625,0.75,0.5,0.625,0.625,0.5625,0.75,0.625,0.875,0.75,0.75,0.875,0.875,0.625,0.875,0.875,1.0,1.0,0.75,0.75,0.875,0.875,0.875,0.8125,0.875,0.875,0.875,0.875,0.875,0.875,1.0,0.875,0.9375,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.25,0.25,0.25,0.25,0.375,0.5,0.375,0.25,0.375,0.375,0.375,0.375,0.5,0.5,0.375,,0.5,0.375,0.625,0.875,0.5,0.625,0.625,0.625,0.625,0.625,0.5,0.625,0.625,0.75,0.5,0.875,0.75,0.75,0.75,0.75,1.0,0.875,0.75,0.875,0.875,0.75,0.875,0.875,0.75,1.0,0.875,0.875,0.875,0.875,0.75,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned + ensembled),0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.0,0.0,0.125,0.25,0.125,0.125,0.125,0.25,0.25,0.375,0.25,0.25,0.5,0.375,0.625,0.5,0.5,,0.625,0.25,0.5,0.375,0.5,0.75,0.75,0.875,0.625,0.625,0.75,0.75,0.875,0.5,0.875,0.75,0.75,0.75,0.875,0.75,0.875,0.875,1.0,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (tuned),0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.375,0.25,0.375,0.25,0.25,0.5,0.25,0.625,0.5,0.625,0.375,,0.5,0.625,0.375,0.5,0.75,0.75,0.75,0.625,0.75,0.625,0.75,0.875,0.5,0.75,0.75,0.75,0.75,0.875,0.75,0.875,0.75,1.0,0.875,0.875,0.875,0.875,0.75,0.875,0.875,0.875,1.0,1.0,0.75,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.25,0.25,0.0,0.25,0.0,0.25,0.25,0.25,0.375,0.5,0.5,0.375,0.5,0.375,0.375,0.75,0.5,,0.5,0.375,0.75,0.625,0.5,0.625,0.625,0.625,0.625,0.75,0.75,0.5,0.75,0.875,0.625,0.75,0.75,0.75,0.75,0.875,0.75,0.75,0.875,0.875,1.0,0.875,1.0,0.875,1.0,1.0,0.875,0.75,0.875,0.875,1.0,0.75,0.875,1.0,1.0,1.0,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.5,0.375,0.125,0.5,0.375,0.5,,0.375,0.5,0.5,0.5,0.5,0.75,0.375,0.5,0.625,0.75,0.5,0.75,0.75,0.75,0.625,0.625,1.0,0.875,0.75,0.75,1.0,0.75,0.75,0.75,0.75,1.0,0.875,0.875,0.75,0.875,0.75,1.0,0.875,1.0,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (tuned),0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.125,0.125,0.375,0.25,0.375,0.375,0.375,0.5,0.25,0.0625,0.375,0.5,0.625,0.625,0.5,0.5625,0.5,0.625,0.625,0.625,0.625,,0.5,0.625,0.625,0.625,0.625,0.5,0.625,0.625,0.625,0.5625,0.625,0.625,0.5,0.625,0.625,0.625,0.75,0.625,0.625,0.75,0.875,0.875,0.625,0.75,0.625,0.75,0.875,0.8125,0.625,0.75,0.75,0.75,0.75,0.75,0.875,0.875,0.8125,0.875,0.875,0.875,0.875,1.0,0.875,0.875,1.0,1.0,1.0
+xRFM (tuned),0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.25,0.125,0.0,0.125,0.125,0.25,0.125,0.25,0.125,0.375,0.375,0.25,0.5,0.25,0.375,0.5,0.5,0.25,0.5,0.5,,0.625,0.5,0.5,0.375,0.75,0.625,0.75,0.625,0.5,0.5,0.75,0.75,0.625,0.75,0.625,0.75,0.75,0.875,0.625,0.75,0.75,0.875,0.875,0.875,1.0,0.875,0.875,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.125,0.25,0.375,0.125,0.125,0.25,0.25,0.375,0.375,0.375,0.25,0.25,0.375,0.5,0.375,0.375,,0.625,0.5,0.625,0.625,0.5,0.625,0.625,0.5,0.625,0.75,0.75,0.75,0.75,0.75,0.875,0.625,0.875,0.875,0.875,0.875,0.875,0.75,1.0,1.0,0.875,0.875,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+XGBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.0,0.25,0.25,0.25,0.125,0.25,0.25,0.25,0.375,0.375,0.25,0.25,0.5,0.5,0.375,0.5,0.375,,0.5,0.5,0.5,0.375,0.625,0.625,0.5,0.625,0.75,0.75,0.75,0.75,0.875,0.875,0.75,0.875,0.875,0.875,0.875,0.875,0.75,1.0,1.0,0.875,0.875,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LightGBM (tuned),0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.0,0.0,0.25,0.375,0.25,0.125,0.0,0.25,0.125,0.25,0.375,0.125,0.25,0.375,0.5,0.375,0.5,0.5,0.5,,0.5,0.625,0.5,0.625,0.5,0.5,0.625,0.625,0.75,0.5,0.875,0.875,0.75,0.75,0.625,0.75,0.875,0.875,0.75,0.875,0.875,1.0,0.75,0.75,1.0,0.75,1.0,1.0,1.0,0.75,1.0,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.25,0.25,0.25,0.125,0.375,0.25,0.375,0.375,0.375,0.375,0.375,0.25,0.375,0.625,0.375,0.5,0.5,,0.375,0.5,0.5,0.5,0.5,0.75,0.625,0.75,0.5,0.625,0.75,0.625,0.625,0.875,0.75,0.875,0.875,0.75,0.75,0.625,1.0,0.625,0.75,1.0,0.75,0.875,1.0,0.875,0.875,1.0,0.875,1.0,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ModernNCA (default),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.375,0.125,0.125,0.125,0.25,0.375,0.25,0.5,0.375,0.25,0.375,0.625,0.5,0.25,0.375,0.5,0.375,0.625,,0.375,0.5,0.625,0.375,0.625,0.75,0.625,0.5,0.625,0.625,0.625,0.625,0.75,0.625,0.625,0.625,0.875,0.75,0.875,0.875,0.875,0.875,0.75,0.75,0.875,0.875,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+SAP-RPT-OSS (default),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.25,0.25,0.25,0.125,0.25,0.25,0.125,0.125,0.25,0.375,0.375,0.375,0.5,0.375,0.25,0.375,0.375,0.5,0.375,0.375,0.5,0.625,0.5,0.5,0.625,,0.5,0.5,0.375,0.625,0.5,0.625,0.75,0.5,0.5,0.625,0.625,0.75,0.625,0.625,0.75,0.875,0.625,0.875,0.75,0.75,0.875,0.75,0.75,0.75,0.625,0.875,0.75,0.75,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+iLTM (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.375,0.25,0.25,0.375,0.375,0.25,0.25,0.25,0.375,0.375,0.25,0.375,0.375,0.375,0.5,0.5,0.5,,0.5,0.5,0.625,0.5,0.625,0.625,0.625,0.75,0.5,0.5,0.75,0.5,0.625,0.75,0.75,0.875,0.875,0.875,0.75,0.75,0.875,0.75,0.875,0.875,1.0,0.75,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+CatBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.25,0.25,0.125,0.125,0.375,0.375,0.375,0.25,0.125,0.125,0.25,0.25,0.375,0.375,0.375,0.375,0.5,0.5,0.375,0.5,0.5,,0.5,0.625,0.625,0.625,0.375,0.625,0.625,0.625,0.875,1.0,0.625,0.75,0.75,0.75,0.75,0.875,0.875,0.875,0.875,1.0,0.75,0.875,1.0,0.875,0.875,0.875,0.75,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TabPFNv2 (default),0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.25,0.25,0.125,0.25,0.375,0.5,0.375,0.125,0.0625,0.5,0.5,0.5,0.5,0.375,0.4375,0.5,0.5,0.5,0.5,0.5,0.4375,0.5,0.5,0.5,0.5,0.5,0.625,0.625,0.5,0.5,,0.5,0.625,0.5,0.625,0.5,0.5,0.5,0.625,0.5,0.5,0.75,0.75,0.625,0.75,0.625,0.625,0.875,0.6875,0.625,0.625,0.625,0.625,0.625,0.625,0.75,0.625,0.8125,0.625,0.625,0.875,0.75,0.875,0.875,0.875,1.0,1.0,1.0
+ChimeraBoost (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.125,0.0,0.125,0.25,0.25,0.125,0.0,0.25,0.25,0.25,0.125,0.125,0.25,0.25,0.25,0.375,0.5,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.375,0.5,,0.5,0.75,0.375,0.625,0.625,0.625,0.5,0.75,0.5,0.625,0.625,0.75,0.75,0.625,0.75,0.625,0.75,1.0,0.75,0.875,0.875,0.875,0.75,1.0,0.875,0.875,1.0,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+ExtraTrees (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.125,0.125,0.375,0.25,0.0,0.125,0.25,0.25,0.375,0.25,0.25,0.25,0.125,0.25,0.375,0.25,0.25,0.25,0.375,0.375,0.25,0.5,0.5,0.375,0.375,0.5,,0.375,0.5,0.625,0.375,0.5,0.625,0.75,0.5,0.625,0.625,1.0,0.75,0.5,0.75,1.0,1.0,0.75,0.75,0.625,0.875,0.625,0.75,0.75,0.625,1.0,0.875,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+RealMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.25,0.125,0.125,0.25,0.25,0.125,0.375,0.125,0.25,0.25,0.25,0.375,0.25,0.5,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.5,0.25,0.625,,0.5,0.375,0.375,0.5,0.625,0.625,0.5,0.625,0.625,0.75,0.75,0.75,0.75,0.75,0.75,0.625,0.75,0.625,0.75,1.0,0.75,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+LimiX (default),0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.25,0.25,0.375,0.125,0.25,0.375,0.25,0.25,0.375,0.25,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.25,0.25,0.5,0.5,0.5,0.25,0.375,0.625,0.375,0.625,0.5,0.5,,0.5,0.5,0.5,0.5,0.625,0.375,0.625,0.625,0.5,0.5,0.625,0.625,0.625,0.5,0.625,0.75,0.625,0.625,0.625,0.75,0.75,0.75,0.625,0.625,0.875,0.875,0.875,0.75,0.875,1.0,0.875,0.875,1.0
+ModernNCA (tuned),0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.25,0.125,0.125,0.125,0.125,0.25,0.375,0.0,0.125,0.125,0.125,0.0,0.25,0.25,0.125,0.125,0.25,0.375,0.375,0.25,0.25,0.25,0.125,0.375,0.375,0.5,0.375,0.375,0.5,0.375,0.375,0.625,0.5,,0.5,0.5,0.375,0.5,0.625,0.625,0.625,0.625,0.625,0.75,0.75,0.625,0.625,0.75,0.625,0.875,0.75,0.875,0.625,0.875,0.875,0.75,1.0,0.75,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+TabM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.125,0.125,0.25,0.125,0.125,0.125,0.0,0.25,0.125,0.0,0.25,0.25,0.25,0.0,0.375,0.375,0.25,0.125,0.125,0.25,0.375,0.5,0.25,0.375,0.5,0.375,0.625,0.625,0.5,0.5,,0.5,0.625,0.625,0.75,0.625,0.625,0.625,0.625,0.875,0.75,0.625,0.625,0.875,0.75,0.875,0.875,1.0,0.75,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+TorchMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.125,0.25,0.125,0.125,0.125,0.25,0.375,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.25,0.125,0.125,0.25,0.375,0.375,0.375,0.5,0.375,0.5,0.375,0.5,0.5,0.5,0.5,0.5,,0.625,0.5,0.75,0.625,0.625,0.625,0.625,0.875,0.875,0.625,0.625,0.75,0.75,0.875,0.75,0.875,0.75,0.75,1.0,0.875,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0
+ExtraTrees (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.125,0.125,0.375,0.25,0.0,0.125,0.25,0.125,0.375,0.25,0.125,0.25,0.125,0.25,0.375,0.25,0.375,0.25,0.25,0.375,0.375,0.375,0.5,0.125,0.375,0.5,0.375,0.375,0.5,0.625,0.375,0.375,,0.5,0.5,0.625,0.625,0.75,0.75,0.375,0.75,0.75,0.875,0.75,0.75,0.625,0.875,0.5,0.75,0.75,0.5,1.0,0.75,0.875,1.0,0.875,1.0,1.0,1.0,1.0,1.0,1.0
+ChimeraBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.0,0.0,0.25,0.25,0.125,0.125,0.0,0.0,0.25,0.25,0.375,0.125,0.125,0.125,0.375,0.125,0.25,0.25,0.25,0.0,0.5,0.25,0.25,0.375,0.375,0.5,0.375,0.5,0.5,,0.5,0.75,0.75,0.625,0.625,0.625,0.75,0.625,0.75,1.0,0.75,0.625,0.875,0.75,0.75,0.75,0.625,0.875,0.875,0.875,0.875,1.0,1.0,1.0,0.875,1.0,1.0,1.0
+TorchMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.0,0.25,0.125,0.125,0.125,0.125,0.25,0.0,0.25,0.375,0.125,0.125,0.25,0.25,0.375,0.375,0.5,0.375,0.5,0.5,0.5,0.5,0.625,0.375,0.25,0.25,0.5,0.5,,0.5,0.5,0.625,0.625,0.625,0.625,0.625,0.625,0.75,0.75,0.875,0.75,0.75,0.75,0.75,1.0,0.625,0.875,0.875,0.75,1.0,0.75,0.875,0.875,0.875,0.875,1.0
+EBM (tuned + ensembled),0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.0,0.25,0.0,0.25,0.125,0.125,0.125,0.25,0.125,0.25,0.125,0.125,0.125,0.125,0.375,0.375,0.375,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.25,0.5,,0.75,0.375,0.625,0.5,0.625,0.375,0.375,0.5,0.875,0.75,0.75,0.625,0.875,1.0,0.75,0.625,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0
+EBM (tuned),0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.25,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.0,0.125,0.125,0.125,0.125,0.25,0.125,0.25,0.125,0.125,0.125,0.125,0.375,0.25,0.25,0.25,0.25,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.375,0.25,0.5,0.25,,0.375,0.625,0.5,0.375,0.375,0.375,0.5,0.75,0.625,0.5,0.625,0.75,0.875,0.75,0.625,0.875,0.875,0.875,1.0,0.875,0.875,0.875,1.0,1.0,1.0
+RandomForest (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.25,0.125,0.25,0.125,0.0,0.125,0.0,0.25,0.375,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.25,0.25,0.375,0.25,0.0,0.25,0.5,0.375,0.375,0.375,0.25,0.375,0.375,0.625,0.625,,0.5,0.5,0.375,0.75,0.875,0.625,0.625,0.375,0.625,0.625,0.625,0.625,0.5,1.0,0.75,0.875,1.0,0.625,1.0,1.0,1.0,1.0,1.0,1.0
+xRFM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.25,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.125,0.25,0.125,0.25,0.25,0.125,0.25,0.25,0.125,0.25,0.25,0.375,0.125,0.25,0.25,0.25,0.25,0.25,0.5,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.375,0.5,,0.375,0.25,0.5,0.625,0.375,0.625,0.375,0.375,0.375,0.75,0.625,0.5,0.75,0.625,0.875,1.0,0.75,0.875,1.0,1.0,1.0,1.0,1.0
+LightGBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.125,0.25,0.125,0.0,0.125,0.125,0.0,0.0,0.375,0.125,0.0,0.0,0.125,0.375,0.125,0.125,0.125,0.125,0.375,0.375,0.5,0.25,0.375,0.25,0.125,0.125,0.625,0.375,0.375,0.5,0.5,0.5,0.625,,0.625,0.625,0.625,0.625,0.75,0.625,0.625,0.75,0.75,0.625,0.75,0.875,0.875,0.875,0.875,0.875,1.0,1.0,0.875,1.0,1.0,1.0
+FastaiMLP (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.0,0.0,0.0,0.0,0.0,0.125,0.25,0.125,0.125,0.375,0.25,0.25,0.25,0.375,0.25,0.25,0.125,0.25,0.25,0.375,0.375,0.625,0.625,0.75,0.375,,0.625,0.625,0.375,0.75,0.625,0.375,0.625,0.75,0.75,0.75,0.875,1.0,0.875,0.875,0.875,0.875,0.875,0.875,1.0,1.0,1.0
+ExtraTrees (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.25,0.125,0.125,0.125,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.25,0.375,0.125,0.25,0.25,0.125,0.125,0.375,0.0,0.25,0.375,0.375,0.375,0.375,0.25,0.375,0.375,0.625,0.625,0.25,0.5,0.375,0.375,,0.5,0.5,0.625,0.375,0.5,0.375,0.625,0.625,0.375,0.75,0.625,0.75,1.0,0.625,1.0,1.0,1.0,1.0,1.0,1.0
+RandomForest (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.0625,0.0,0.0,0.0,0.125,0.125,0.1875,0.125,0.0,0.0,0.0,0.25,0.1875,0.125,0.125,0.125,0.25,0.25,0.125,0.125,0.25,0.125,0.3125,0.25,0.0,0.25,0.5,0.375,0.375,0.375,0.125,0.25,0.375,0.625,0.625,0.125,0.375,0.375,0.375,0.5,,0.5,0.625,0.375,0.5,0.375,0.625,0.625,0.375,0.75,0.625,0.75,1.0,0.625,0.875,1.0,1.0,1.0,1.0,1.0
+PerpetualBooster (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.125,0.125,0.375,0.0,0.0,0.0,0.0,0.0,0.25,0.25,0.125,0.0,0.375,0.0,0.25,0.375,0.375,0.25,0.125,0.25,0.25,0.0,0.25,0.5,0.5,0.375,0.625,0.375,0.625,0.5,0.5,,0.625,0.625,0.6875,0.625,0.625,0.625,0.625,0.625,0.75,0.75,0.625,1.0,0.875,0.875,0.875,1.0,1.0,1.0
+TabSTAR (tuned + ensembled),0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.25,0.0,0.25,0.125,0.125,0.25,0.25,0.125,0.25,0.25,0.125,0.25,0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.25,0.375,0.25,0.25,0.25,0.25,0.375,0.25,0.25,0.25,0.25,0.25,0.125,0.25,0.375,0.375,0.25,0.25,0.375,0.375,0.375,,0.25,0.375,0.25,0.875,0.375,0.375,0.375,0.5,1.0,0.5,0.375,0.5,0.625,0.625,0.75,0.75,0.875
+FastaiMLP (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.0,0.25,0.0,0.0,0.0,0.0,0.125,0.125,0.25,0.125,0.125,0.375,0.125,0.375,0.375,0.375,0.125,0.125,0.125,0.375,0.375,0.125,0.25,0.375,0.625,0.625,0.375,0.375,0.625,0.625,0.375,0.75,,0.375,0.5,0.75,0.5,0.75,0.75,0.625,0.875,0.75,0.875,0.875,0.875,0.875,1.0,1.0,1.0
+PerpetualBooster (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.125,0.125,0.25,0.0,0.0,0.0,0.0,0.0,0.125,0.375,0.125,0.0,0.375,0.125,0.125,0.25,0.375,0.25,0.125,0.25,0.125,0.125,0.25,0.25,0.5,0.375,0.625,0.375,0.625,0.5,0.5,0.3125,0.625,0.625,,0.5,0.625,0.75,0.625,0.625,0.75,0.75,0.625,0.875,0.875,0.875,0.875,1.0,1.0,1.0
+XGBoost (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.0,0.0,0.0,0.125,0.25,0.125,0.0,0.125,0.375,0.125,0.375,0.0,0.375,0.125,0.0,0.125,0.5,0.25,0.25,0.375,0.375,0.375,0.625,0.25,0.375,0.625,0.625,0.375,0.75,0.5,0.5,,0.75,0.625,0.75,0.625,0.625,0.75,0.75,0.875,0.875,1.0,1.0,0.875,0.875,1.0
+TabSTAR (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.125,0.125,0.125,0.25,0.125,0.125,0.25,0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.125,0.125,0.25,0.25,0.125,0.375,0.25,0.25,0.25,0.25,0.375,0.25,0.25,0.25,0.25,0.25,0.125,0.25,0.375,0.25,0.25,0.25,0.375,0.375,0.375,0.125,0.25,0.375,0.25,,0.375,0.375,0.375,0.5,1.0,0.5,0.375,0.5,0.625,0.625,0.75,0.75,0.875
+EBM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.25,0.125,0.125,0.25,0.0,0.25,0.125,0.25,0.125,0.25,0.25,0.25,0.25,0.25,0.0,0.125,0.375,0.375,0.375,0.25,0.375,0.375,0.375,0.625,0.5,0.25,0.375,0.625,,0.75,0.5,0.625,0.875,0.75,0.875,0.875,0.875,0.875,1.0,1.0,1.0
+TorchMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.25,0.375,0.125,0.375,0.125,0.25,0.125,0.125,0.0,0.5,0.375,0.0,0.25,0.25,0.5,0.5,0.25,0.25,0.625,0.625,0.375,0.625,0.25,0.375,0.25,0.625,0.25,,0.625,0.25,0.75,0.625,0.5,0.75,0.75,0.75,0.875,0.875,1.0
+RandomForest (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.125,0.0,0.0,0.0,0.0,0.125,0.0625,0.0,0.0,0.0,0.0,0.0,0.0625,0.0,0.0,0.0,0.0,0.0,0.1875,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.1875,0.125,0.0,0.125,0.375,0.25,0.125,0.125,0.0,0.125,0.375,0.375,0.375,0.0,0.25,0.125,0.125,0.25,0.25,0.375,0.625,0.25,0.375,0.375,0.625,0.5,0.375,,0.5,0.75,0.75,0.625,0.75,1.0,1.0,1.0,1.0,1.0
+FastaiMLP (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.375,0.0,0.125,0.0,0.375,0.0,0.125,0.125,0.25,0.125,0.125,0.125,0.125,0.25,0.375,0.125,0.0,0.375,0.375,0.25,0.5,0.375,0.25,0.375,0.5,0.375,0.75,0.5,,0.5,0.5,0.75,0.625,0.75,0.875,1.0,1.0,1.0
+TabSTAR (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.125,0.375,0.125,0.125,0.0,0.125,0.25,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.125,0.25,0.25,0.25,0.0,0.125,0.25,0.25,0.0,0.125,0.25,0.25,0.5,,0.5,0.25,0.375,0.5,0.5,0.75,0.75,0.75
+KNN (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.125,0.125,0.0,0.125,0.125,0.125,0.125,0.125,0.0,0.125,0.25,0.125,0.125,0.0,0.0,0.125,0.125,0.0,0.0,0.375,0.5,0.25,0.375,0.25,0.5,0.25,0.375,0.25,0.5,0.5,,0.5,0.25,1.0,1.0,0.875,0.875,1.0
+PerpetualBooster (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.125,0.0,0.125,0.125,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.375,0.25,0.125,0.125,0.375,0.375,0.0,0.625,0.125,0.125,0.125,0.625,0.125,0.5,0.375,0.25,0.75,0.5,,0.5,0.625,0.75,0.875,0.875,1.0
+iLTM (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.25,0.0,0.0,0.125,0.0,0.0,0.25,0.125,0.125,0.0,0.125,0.0,0.125,0.0,0.125,0.125,0.5,0.125,0.125,0.125,0.5,0.125,0.25,0.25,0.375,0.625,0.75,0.5,,0.75,0.75,0.875,0.875,0.875
+KNN (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.0,0.0,0.125,0.125,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.125,0.375,0.125,0.125,0.0,0.375,0.125,0.25,0.0,0.25,0.5,0.0,0.375,0.25,,0.75,0.875,0.875,1.0
+KNN (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.125,0.125,0.125,0.125,0.0,0.0,0.125,0.125,0.0,0.0,0.125,0.375,0.125,0.125,0.0,0.375,0.125,0.25,0.0,0.125,0.5,0.0,0.25,0.25,0.25,,0.625,0.625,0.625
+Linear (tuned + ensembled),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.0,0.125,0.25,0.0,0.125,0.0,0.0,0.25,0.125,0.125,0.125,0.125,0.375,,0.6875,0.75
+Linear (tuned),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.25,0.0,0.0,0.125,0.25,0.0,0.125,0.0,0.0,0.25,0.125,0.125,0.125,0.125,0.375,0.3125,,0.75
+Linear (default),0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.125,0.0,0.0,0.0,0.0,0.25,0.0,0.0,0.125,0.0,0.375,0.25,0.25,
diff --git a/data_beyondarena/subsets/full/leaderboard_table.html b/data_beyondarena/subsets/full/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..458331513baae0454e89123a19e9d224de67dcdb
--- /dev/null
+++ b/data_beyondarena/subsets/full/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/full</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1278.0,"score":0.494,"rank":7.84,"hrank":4.31,"imp":10.996,"train_time":1303.63,"predict_time":13.741,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+31\/-27","imputed":false},{"position":1,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1276.0,"score":0.594,"rank":7.9,"hrank":2.71,"imp":8.468,"train_time":19.77,"predict_time":3.456,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+45\/-43","imputed":false},{"position":2,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1221.0,"score":0.498,"rank":9.56,"hrank":3.52,"imp":11.134,"train_time":24.91,"predict_time":2.368,"imputed_pct":14.65,"hardware":"GPU","elo_ci":"+42\/-39","imputed":true},{"position":3,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1214.0,"score":0.364,"rank":9.79,"hrank":5.55,"imp":12.417,"train_time":1079.03,"predict_time":7.011,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+30\/-27","imputed":false},{"position":4,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1207.0,"score":0.348,"rank":9.99,"hrank":6.02,"imp":12.873,"train_time":431.45,"predict_time":1.13,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+28\/-22","imputed":false},{"position":5,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1204.0,"score":0.382,"rank":10.09,"hrank":5.22,"imp":12.567,"train_time":1303.63,"predict_time":1.646,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+31\/-23","imputed":false},{"position":6,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1204.0,"score":0.504,"rank":10.11,"hrank":3.32,"imp":10.809,"train_time":11.68,"predict_time":2.38,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+43\/-41","imputed":false},{"position":7,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1191.0,"score":0.333,"rank":10.51,"hrank":6.13,"imp":13.044,"train_time":431.45,"predict_time":0.205,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+26\/-24","imputed":false},{"position":8,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1185.0,"score":0.319,"rank":10.74,"hrank":6.53,"imp":13.611,"train_time":151.08,"predict_time":2.401,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+28\/-22","imputed":false},{"position":9,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1173.0,"score":0.299,"rank":11.14,"hrank":6.01,"imp":13.173,"train_time":1079.03,"predict_time":1.012,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+28\/-30","imputed":false},{"position":10,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1145.0,"score":0.264,"rank":12.07,"hrank":7.34,"imp":14.171,"train_time":151.08,"predict_time":0.578,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+28\/-22","imputed":false},{"position":11,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1141.0,"score":0.263,"rank":12.22,"hrank":7.94,"imp":15.266,"train_time":270.68,"predict_time":6.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+31\/-24","imputed":false},{"position":12,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1133.0,"score":0.262,"rank":12.5,"hrank":8.27,"imp":14.627,"train_time":13.17,"predict_time":0.162,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+31\/-20","imputed":false},{"position":13,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1119.0,"score":0.22,"rank":12.98,"hrank":8.37,"imp":15.487,"train_time":270.68,"predict_time":1.73,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+28\/-24","imputed":false},{"position":14,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1101.0,"score":0.22,"rank":13.6,"hrank":8.69,"imp":14.797,"train_time":29.29,"predict_time":0.957,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+26\/-26","imputed":false},{"position":15,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1053.0,"score":0.161,"rank":15.26,"hrank":9.14,"imp":16.664,"train_time":44.4,"predict_time":1.567,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+21\/-26","imputed":false},{"position":16,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1049.0,"score":0.245,"rank":15.43,"hrank":6.2,"imp":18.248,"train_time":108.07,"predict_time":10.559,"imputed_pct":14.65,"hardware":"GPU","elo_ci":"+38\/-47","imputed":true},{"position":17,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1013.0,"score":0.086,"rank":16.66,"hrank":11.75,"imp":18.837,"train_time":110.05,"predict_time":1.51,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+21\/-30","imputed":false},{"position":18,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.117,"rank":17.1,"hrank":12.46,"imp":18.237,"train_time":6.69,"predict_time":1.353,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+30\/-24","imputed":false},{"position":19,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":989.0,"score":0.07,"rank":17.47,"hrank":13.42,"imp":18.409,"train_time":3.44,"predict_time":0.228,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+26\/-21","imputed":false},{"position":20,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":979.0,"score":0.069,"rank":17.82,"hrank":12.22,"imp":19.652,"train_time":110.05,"predict_time":0.272,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+28\/-34","imputed":false},{"position":21,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":962.0,"score":0.062,"rank":18.35,"hrank":13.59,"imp":20.032,"train_time":143.1,"predict_time":1.542,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+23\/-28","imputed":false},{"position":22,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":940.0,"score":0.048,"rank":19.07,"hrank":14.45,"imp":20.594,"train_time":143.1,"predict_time":0.436,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+26\/-32","imputed":false},{"position":23,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":881.0,"score":0.091,"rank":20.84,"hrank":11.84,"imp":26.942,"train_time":204.92,"predict_time":0.895,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+29\/-40","imputed":false},{"position":24,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":874.0,"score":0.042,"rank":21.02,"hrank":15.78,"imp":22.699,"train_time":9.98,"predict_time":0.746,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+30\/-34","imputed":false},{"position":25,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":866.0,"score":0.058,"rank":21.27,"hrank":14.15,"imp":23.462,"train_time":6.19,"predict_time":0.787,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+37\/-42","imputed":false},{"position":26,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":851.0,"score":0.076,"rank":21.67,"hrank":12.79,"imp":27.725,"train_time":204.92,"predict_time":0.204,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-44","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":796.0,"score":0.055,"rank":23.02,"hrank":15.57,"imp":30.204,"train_time":6.33,"predict_time":0.216,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-52","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/grouped/leaderboard_table.html b/data_beyondarena/subsets/grouped/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..e1e4505f2d74edc38aaf16849e890ba2b5c1978c
--- /dev/null
+++ b/data_beyondarena/subsets/grouped/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/grouped</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1306.0,"score":0.584,"rank":7.8,"hrank":3.93,"imp":12.426,"train_time":1872.16,"predict_time":13.565,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-63","imputed":false},{"position":1,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1302.0,"score":0.548,"rank":7.92,"hrank":4.23,"imp":11.862,"train_time":1345.55,"predict_time":7.445,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-70","imputed":false},{"position":2,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1272.0,"score":0.506,"rank":8.81,"hrank":3.89,"imp":12.349,"train_time":1345.55,"predict_time":2.136,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-77","imputed":false},{"position":3,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1257.0,"score":0.474,"rank":9.3,"hrank":3.94,"imp":15.056,"train_time":206.86,"predict_time":2.295,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-64","imputed":false},{"position":4,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1238.0,"score":0.461,"rank":9.92,"hrank":4.3,"imp":13.066,"train_time":1872.16,"predict_time":2.012,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-56","imputed":false},{"position":5,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1237.0,"score":0.443,"rank":9.95,"hrank":5.65,"imp":15.456,"train_time":206.86,"predict_time":0.542,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-70","imputed":false},{"position":6,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1236.0,"score":0.437,"rank":10.0,"hrank":5.75,"imp":16.932,"train_time":1570.46,"predict_time":0.592,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-59","imputed":false},{"position":7,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1219.0,"score":0.417,"rank":10.54,"hrank":5.24,"imp":17.104,"train_time":1570.46,"predict_time":0.213,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-57","imputed":false},{"position":8,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1176.0,"score":0.375,"rank":12.05,"hrank":6.73,"imp":18.231,"train_time":35.96,"predict_time":1.821,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+89\/-78","imputed":false},{"position":9,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1153.0,"score":0.255,"rank":12.88,"hrank":8.56,"imp":21.485,"train_time":18.0,"predict_time":0.138,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-51","imputed":false},{"position":10,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1110.0,"score":0.215,"rank":14.43,"hrank":8.54,"imp":20.961,"train_time":60.51,"predict_time":1.526,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-53","imputed":false},{"position":11,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1107.0,"score":0.23,"rank":14.53,"hrank":6.8,"imp":22.504,"train_time":45.68,"predict_time":2.913,"imputed_pct":22.22,"hardware":"GPU","elo_ci":"+86\/-89","imputed":true},{"position":12,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1107.0,"score":0.172,"rank":14.54,"hrank":10.71,"imp":23.803,"train_time":375.91,"predict_time":7.017,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-67","imputed":false},{"position":13,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1099.0,"score":0.13,"rank":14.83,"hrank":11.04,"imp":23.963,"train_time":375.91,"predict_time":1.678,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-56","imputed":false},{"position":14,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1094.0,"score":0.143,"rank":15.0,"hrank":10.31,"imp":23.132,"train_time":135.04,"predict_time":1.076,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-60","imputed":false},{"position":15,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1085.0,"score":0.132,"rank":15.36,"hrank":9.32,"imp":23.261,"train_time":135.04,"predict_time":0.341,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+64\/-66","imputed":false},{"position":16,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1075.0,"score":0.213,"rank":15.69,"hrank":5.84,"imp":21.483,"train_time":15.85,"predict_time":2.805,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+72\/-83","imputed":false},{"position":17,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1075.0,"score":0.25,"rank":15.71,"hrank":6.45,"imp":24.224,"train_time":24.5,"predict_time":4.699,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-97","imputed":false},{"position":18,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1038.0,"score":0.046,"rank":17.04,"hrank":12.62,"imp":24.695,"train_time":159.54,"predict_time":1.96,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-56","imputed":false},{"position":19,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1028.0,"score":0.079,"rank":17.38,"hrank":13.49,"imp":25.383,"train_time":5.46,"predict_time":0.229,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-58","imputed":false},{"position":20,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1022.0,"score":0.008,"rank":17.61,"hrank":14.28,"imp":24.691,"train_time":159.54,"predict_time":0.477,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-66","imputed":false},{"position":21,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1017.0,"score":0.09,"rank":17.78,"hrank":11.7,"imp":25.345,"train_time":6.49,"predict_time":0.787,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-74","imputed":false},{"position":22,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.098,"rank":18.36,"hrank":13.53,"imp":26.968,"train_time":9.39,"predict_time":1.54,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-69","imputed":false},{"position":23,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":999.0,"score":0.041,"rank":18.39,"hrank":13.34,"imp":25.67,"train_time":11.02,"predict_time":0.764,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-82","imputed":false},{"position":24,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":982.0,"score":0.155,"rank":18.98,"hrank":9.36,"imp":29.566,"train_time":208.92,"predict_time":1.069,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-124","imputed":false},{"position":25,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":962.0,"score":0.072,"rank":19.62,"hrank":9.73,"imp":27.614,"train_time":115.79,"predict_time":5.264,"imputed_pct":22.22,"hardware":"GPU","elo_ci":"+70\/-90","imputed":true},{"position":26,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":959.0,"score":0.158,"rank":19.71,"hrank":7.36,"imp":30.346,"train_time":208.92,"predict_time":0.302,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-139","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":886.0,"score":0.107,"rank":21.88,"hrank":11.57,"imp":35.304,"train_time":5.55,"predict_time":0.43,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+124\/-164","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/high-cardinality/leaderboard_table.html b/data_beyondarena/subsets/high-cardinality/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..76e17748bdec647e24e810c4f04321989d7ee7d4
--- /dev/null
+++ b/data_beyondarena/subsets/high-cardinality/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/high-cardinality</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1213.0,"score":0.647,"rank":5.78,"hrank":2.99,"imp":5.272,"train_time":610.05,"predict_time":4.078,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+120\/-97","imputed":false},{"position":1,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1200.0,"score":0.626,"rank":6.06,"hrank":3.41,"imp":5.404,"train_time":233.25,"predict_time":0.505,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-68","imputed":false},{"position":2,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1172.0,"score":0.605,"rank":6.67,"hrank":3.44,"imp":5.417,"train_time":233.25,"predict_time":0.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-75","imputed":false},{"position":3,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1153.0,"score":0.548,"rank":7.1,"hrank":5.29,"imp":6.053,"train_time":129.78,"predict_time":1.547,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-50","imputed":false},{"position":4,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1125.0,"score":0.558,"rank":7.75,"hrank":3.64,"imp":6.58,"train_time":610.05,"predict_time":0.383,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+96\/-83","imputed":false},{"position":5,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1114.0,"score":0.479,"rank":8.02,"hrank":4.53,"imp":7.121,"train_time":47.06,"predict_time":1.788,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-58","imputed":false},{"position":6,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1098.0,"score":0.53,"rank":8.43,"hrank":5.82,"imp":6.057,"train_time":5.99,"predict_time":0.081,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-68","imputed":false},{"position":7,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1080.0,"score":0.464,"rank":8.88,"hrank":4.84,"imp":6.448,"train_time":129.78,"predict_time":0.275,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-63","imputed":false},{"position":8,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1076.0,"score":0.382,"rank":8.97,"hrank":4.88,"imp":7.927,"train_time":370.9,"predict_time":3.464,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-56","imputed":false},{"position":9,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1025.0,"score":0.356,"rank":10.31,"hrank":6.2,"imp":8.214,"train_time":47.06,"predict_time":0.435,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-68","imputed":false},{"position":10,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1020.0,"score":0.369,"rank":10.44,"hrank":3.93,"imp":7.187,"train_time":8.37,"predict_time":2.944,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-67","imputed":false},{"position":11,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.315,"rank":10.98,"hrank":8.5,"imp":8.194,"train_time":1.41,"predict_time":0.225,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-40","imputed":false},{"position":12,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":980.0,"score":0.268,"rank":11.51,"hrank":7.62,"imp":9.128,"train_time":370.9,"predict_time":0.511,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+86\/-56","imputed":false},{"position":13,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":860.0,"score":0.19,"rank":14.76,"hrank":11.47,"imp":11.637,"train_time":9.97,"predict_time":0.464,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+95\/-79","imputed":false},{"position":14,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":860.0,"score":0.152,"rank":14.77,"hrank":11.19,"imp":9.95,"train_time":18.03,"predict_time":0.367,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+79\/-70","imputed":false},{"position":15,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":860.0,"score":0.251,"rank":14.78,"hrank":5.71,"imp":10.742,"train_time":6.66,"predict_time":0.469,"imputed_pct":43.33,"hardware":"GPU","elo_ci":"+117\/-112","imputed":true},{"position":16,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":848.0,"score":0.201,"rank":15.09,"hrank":6.27,"imp":11.011,"train_time":5.91,"predict_time":2.446,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-105","imputed":false},{"position":17,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":833.0,"score":0.088,"rank":15.48,"hrank":13.82,"imp":10.842,"train_time":0.49,"predict_time":0.073,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+54\/-56","imputed":false},{"position":18,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":693.0,"score":0.025,"rank":19.03,"hrank":17.5,"imp":15.41,"train_time":21.23,"predict_time":0.595,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-80","imputed":false},{"position":19,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":664.0,"score":0.012,"rank":19.74,"hrank":18.62,"imp":16.006,"train_time":32.94,"predict_time":0.391,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-64","imputed":false},{"position":20,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":610.0,"score":0.01,"rank":20.97,"hrank":19.19,"imp":16.675,"train_time":21.23,"predict_time":0.095,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-85","imputed":false},{"position":21,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":595.0,"score":0.011,"rank":21.3,"hrank":20.03,"imp":17.388,"train_time":32.94,"predict_time":0.078,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-81","imputed":false},{"position":22,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":579.0,"score":0.032,"rank":21.64,"hrank":17.73,"imp":20.086,"train_time":35.74,"predict_time":47.548,"imputed_pct":43.33,"hardware":"GPU","elo_ci":"+94\/-132","imputed":true},{"position":23,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":546.0,"score":0.009,"rank":22.33,"hrank":20.87,"imp":22.351,"train_time":36.37,"predict_time":0.533,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+78\/-94","imputed":false},{"position":24,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":507.0,"score":0.0,"rank":23.1,"hrank":22.28,"imp":19.575,"train_time":1.76,"predict_time":0.173,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+61\/-99","imputed":false},{"position":25,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":498.0,"score":0.008,"rank":23.26,"hrank":22.02,"imp":23.4,"train_time":36.37,"predict_time":0.19,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-103","imputed":false},{"position":26,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":440.0,"score":0.0,"rank":24.28,"hrank":23.85,"imp":21.088,"train_time":0.87,"predict_time":0.193,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-83","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":423.0,"score":0.008,"rank":24.55,"hrank":23.48,"imp":25.526,"train_time":1.38,"predict_time":0.199,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-122","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/high-dim/leaderboard_table.html b/data_beyondarena/subsets/high-dim/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..838f1c6599734eb55648db7c0f0bece904e21cc5
--- /dev/null
+++ b/data_beyondarena/subsets/high-dim/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/high-dim</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1262.0,"score":0.576,"rank":6.74,"hrank":3.49,"imp":8.642,"train_time":1281.84,"predict_time":13.342,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+69\/-48","imputed":false},{"position":1,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1185.0,"score":0.423,"rank":8.84,"hrank":4.89,"imp":9.632,"train_time":1178.51,"predict_time":10.819,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+51\/-41","imputed":false},{"position":2,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1181.0,"score":0.434,"rank":8.98,"hrank":4.99,"imp":10.367,"train_time":115.35,"predict_time":2.775,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-39","imputed":false},{"position":3,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1169.0,"score":0.466,"rank":9.31,"hrank":4.28,"imp":10.177,"train_time":1281.84,"predict_time":1.388,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+76\/-52","imputed":false},{"position":4,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1166.0,"score":0.389,"rank":9.4,"hrank":5.67,"imp":10.908,"train_time":542.92,"predict_time":1.14,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+46\/-44","imputed":false},{"position":5,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1159.0,"score":0.444,"rank":9.61,"hrank":3.42,"imp":8.659,"train_time":20.81,"predict_time":5.555,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-49","imputed":false},{"position":6,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1152.0,"score":0.373,"rank":9.83,"hrank":6.26,"imp":10.964,"train_time":542.92,"predict_time":0.202,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-50","imputed":false},{"position":7,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1149.0,"score":0.389,"rank":9.94,"hrank":5.97,"imp":11.663,"train_time":238.02,"predict_time":3.628,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-54","imputed":false},{"position":8,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1137.0,"score":0.371,"rank":10.29,"hrank":5.91,"imp":11.003,"train_time":115.35,"predict_time":0.655,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-41","imputed":false},{"position":9,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1132.0,"score":0.347,"rank":10.47,"hrank":5.11,"imp":10.224,"train_time":1178.51,"predict_time":1.584,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+50\/-40","imputed":false},{"position":10,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1117.0,"score":0.474,"rank":10.94,"hrank":3.11,"imp":10.806,"train_time":44.73,"predict_time":5.695,"imputed_pct":32.94,"hardware":"GPU","elo_ci":"+73\/-72","imputed":true},{"position":11,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1106.0,"score":0.325,"rank":11.3,"hrank":6.3,"imp":12.002,"train_time":238.02,"predict_time":0.715,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-50","imputed":false},{"position":12,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1091.0,"score":0.309,"rank":11.79,"hrank":8.3,"imp":11.947,"train_time":16.17,"predict_time":0.148,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+51\/-38","imputed":false},{"position":13,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1033.0,"score":0.217,"rank":13.71,"hrank":9.6,"imp":12.326,"train_time":31.98,"predict_time":2.057,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+53\/-44","imputed":false},{"position":14,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1013.0,"score":0.282,"rank":14.38,"hrank":5.22,"imp":12.872,"train_time":18.51,"predict_time":3.73,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+73\/-69","imputed":false},{"position":15,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1008.0,"score":0.158,"rank":14.56,"hrank":9.88,"imp":13.288,"train_time":30.11,"predict_time":1.419,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+44\/-41","imputed":false},{"position":16,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.163,"rank":14.82,"hrank":11.63,"imp":14.429,"train_time":5.48,"predict_time":0.682,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-46","imputed":false},{"position":17,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":970.0,"score":0.113,"rank":15.8,"hrank":12.41,"imp":14.566,"train_time":1.83,"predict_time":0.24,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-36","imputed":false},{"position":18,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":927.0,"score":0.051,"rank":17.21,"hrank":14.46,"imp":17.076,"train_time":69.28,"predict_time":1.356,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-49","imputed":false},{"position":19,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":875.0,"score":0.026,"rank":18.81,"hrank":14.73,"imp":18.374,"train_time":139.91,"predict_time":1.077,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-45","imputed":false},{"position":20,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":864.0,"score":0.027,"rank":19.17,"hrank":15.62,"imp":18.57,"train_time":69.28,"predict_time":0.202,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-55","imputed":false},{"position":21,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":847.0,"score":0.015,"rank":19.66,"hrank":16.1,"imp":19.281,"train_time":139.91,"predict_time":0.202,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-49","imputed":false},{"position":22,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":827.0,"score":0.06,"rank":20.23,"hrank":11.74,"imp":20.902,"train_time":52.83,"predict_time":8.63,"imputed_pct":32.94,"hardware":"GPU","elo_ci":"+59\/-55","imputed":true},{"position":23,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":800.0,"score":0.015,"rank":20.96,"hrank":17.43,"imp":21.048,"train_time":10.23,"predict_time":0.577,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-54","imputed":false},{"position":24,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":783.0,"score":0.067,"rank":21.4,"hrank":14.82,"imp":22.671,"train_time":100.41,"predict_time":1.617,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-63","imputed":false},{"position":25,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":774.0,"score":0.036,"rank":21.65,"hrank":15.26,"imp":21.118,"train_time":6.09,"predict_time":0.563,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+63\/-68","imputed":false},{"position":26,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":744.0,"score":0.063,"rank":22.37,"hrank":13.24,"imp":23.641,"train_time":100.41,"predict_time":0.382,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-82","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":677.0,"score":0.064,"rank":23.83,"hrank":16.49,"imp":26.865,"train_time":4.31,"predict_time":0.421,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+84\/-102","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/large/leaderboard_table.html b/data_beyondarena/subsets/large/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..0df3957391b2029356369f51b4bb6c3ccd2c719f
--- /dev/null
+++ b/data_beyondarena/subsets/large/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/large</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1383.0,"score":0.805,"rank":4.37,"hrank":2.34,"imp":1.839,"train_time":333.41,"predict_time":1.402,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+169\/-124","imputed":false},{"position":1,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1273.0,"score":0.624,"rank":6.39,"hrank":3.25,"imp":3.342,"train_time":46.72,"predict_time":2.739,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+187\/-103","imputed":false},{"position":2,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1270.0,"score":0.609,"rank":6.46,"hrank":4.34,"imp":3.694,"train_time":281.55,"predict_time":3.367,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+108\/-63","imputed":false},{"position":3,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1253.0,"score":0.702,"rank":6.81,"hrank":2.91,"imp":2.504,"train_time":333.41,"predict_time":0.211,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+144\/-124","imputed":false},{"position":4,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1248.0,"score":0.608,"rank":6.91,"hrank":3.4,"imp":3.909,"train_time":208.34,"predict_time":0.439,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+155\/-116","imputed":false},{"position":5,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1230.0,"score":0.594,"rank":7.3,"hrank":5.04,"imp":3.868,"train_time":208.34,"predict_time":0.049,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+146\/-126","imputed":false},{"position":6,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.598,"rank":7.46,"hrank":4.29,"imp":3.664,"train_time":133.01,"predict_time":1.154,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+162\/-104","imputed":false},{"position":7,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1204.0,"score":0.53,"rank":7.88,"hrank":5.46,"imp":4.288,"train_time":46.72,"predict_time":0.662,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+160\/-86","imputed":false},{"position":8,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1151.0,"score":0.485,"rank":9.11,"hrank":4.83,"imp":4.732,"train_time":281.55,"predict_time":0.497,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+109\/-72","imputed":false},{"position":9,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1145.0,"score":0.504,"rank":9.26,"hrank":4.9,"imp":3.868,"train_time":133.01,"predict_time":0.179,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+150\/-92","imputed":false},{"position":10,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1121.0,"score":0.463,"rank":9.84,"hrank":7.16,"imp":4.886,"train_time":8.03,"predict_time":0.043,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+134\/-104","imputed":false},{"position":11,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1005.0,"score":0.262,"rank":12.68,"hrank":10.43,"imp":6.24,"train_time":9.5,"predict_time":0.431,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-84","imputed":false},{"position":12,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.254,"rank":12.79,"hrank":11.84,"imp":6.672,"train_time":1.35,"predict_time":0.162,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+90\/-67","imputed":false},{"position":13,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":992.0,"score":0.207,"rank":12.99,"hrank":9.32,"imp":5.748,"train_time":12.85,"predict_time":0.217,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+98\/-71","imputed":false},{"position":14,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":945.0,"score":0.195,"rank":14.15,"hrank":6.92,"imp":9.36,"train_time":17.54,"predict_time":11.689,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+104\/-126","imputed":false},{"position":15,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":942.0,"score":0.141,"rank":14.22,"hrank":12.74,"imp":7.368,"train_time":0.49,"predict_time":0.11,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-63","imputed":false},{"position":16,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":717.0,"score":0.006,"rank":19.66,"hrank":17.86,"imp":13.772,"train_time":18.15,"predict_time":0.424,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-150","imputed":false},{"position":17,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":716.0,"score":0.015,"rank":19.67,"hrank":17.76,"imp":13.898,"train_time":41.24,"predict_time":0.222,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-139","imputed":false},{"position":18,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":702.0,"score":0.022,"rank":20.01,"hrank":15.42,"imp":14.746,"train_time":7.28,"predict_time":3.366,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-168","imputed":false},{"position":19,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":676.0,"score":0.015,"rank":20.61,"hrank":18.42,"imp":14.787,"train_time":41.24,"predict_time":0.067,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-132","imputed":false},{"position":20,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":660.0,"score":0.0,"rank":20.96,"hrank":16.86,"imp":15.587,"train_time":2.59,"predict_time":0.113,"imputed_pct":99.05,"hardware":"GPU","elo_ci":"+82\/-113","imputed":true},{"position":21,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":650.0,"score":0.0,"rank":21.18,"hrank":19.93,"imp":15.67,"train_time":2.59,"predict_time":0.113,"imputed_pct":99.05,"hardware":"GPU","elo_ci":"+68\/-108","imputed":true},{"position":22,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":650.0,"score":0.0,"rank":21.19,"hrank":19.94,"imp":15.672,"train_time":2.55,"predict_time":0.113,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-108","imputed":false},{"position":23,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":648.0,"score":0.016,"rank":21.23,"hrank":18.62,"imp":14.991,"train_time":18.15,"predict_time":0.062,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+123\/-184","imputed":false},{"position":24,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":615.0,"score":0.016,"rank":21.94,"hrank":19.16,"imp":20.395,"train_time":13.34,"predict_time":0.451,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-157","imputed":false},{"position":25,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":564.0,"score":0.015,"rank":23.0,"hrank":20.0,"imp":21.399,"train_time":13.34,"predict_time":0.092,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+120\/-177","imputed":false},{"position":26,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":538.0,"score":0.0,"rank":23.5,"hrank":18.96,"imp":17.358,"train_time":0.87,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+118\/-164","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":486.0,"score":0.0,"rank":24.41,"hrank":21.42,"imp":22.958,"train_time":0.56,"predict_time":0.118,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-170","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/low-dim/leaderboard_table.html b/data_beyondarena/subsets/low-dim/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..742e3f5f9b98f9ff0a6c6d0a70853073bd9156cf
--- /dev/null
+++ b/data_beyondarena/subsets/low-dim/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/low-dim</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1349.0,"score":0.679,"rank":6.94,"hrank":2.43,"imp":8.36,"train_time":17.99,"predict_time":2.424,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-42","imputed":false},{"position":1,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1321.0,"score":0.629,"rank":7.71,"hrank":2.76,"imp":9.654,"train_time":9.3,"predict_time":1.633,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+47\/-48","imputed":false},{"position":2,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1295.0,"score":0.448,"rank":8.46,"hrank":4.96,"imp":12.314,"train_time":1327.31,"predict_time":17.147,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+39\/-30","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1284.0,"score":0.511,"rank":8.79,"hrank":3.79,"imp":11.317,"train_time":13.91,"predict_time":1.616,"imputed_pct":4.4,"hardware":"GPU","elo_ci":"+45\/-42","imputed":true},{"position":4,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1235.0,"score":0.331,"rank":10.32,"hrank":5.99,"imp":13.978,"train_time":1011.46,"predict_time":4.537,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+38\/-35","imputed":false},{"position":5,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1235.0,"score":0.325,"rank":10.32,"hrank":6.23,"imp":13.974,"train_time":312.23,"predict_time":1.085,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+24\/-27","imputed":false},{"position":6,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1229.0,"score":0.335,"rank":10.53,"hrank":5.94,"imp":13.906,"train_time":1327.31,"predict_time":2.534,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+32\/-30","imputed":false},{"position":7,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1218.0,"score":0.31,"rank":10.9,"hrank":6.05,"imp":14.21,"train_time":312.23,"predict_time":0.224,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+25\/-28","imputed":false},{"position":8,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1200.0,"score":0.273,"rank":11.51,"hrank":6.67,"imp":14.826,"train_time":1011.46,"predict_time":0.817,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+33\/-36","imputed":false},{"position":9,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1193.0,"score":0.255,"rank":11.72,"hrank":7.91,"imp":15.429,"train_time":214.85,"predict_time":2.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+30\/-29","imputed":false},{"position":10,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1164.0,"score":0.349,"rank":12.74,"hrank":4.91,"imp":16.761,"train_time":143.53,"predict_time":10.681,"imputed_pct":4.4,"hardware":"GPU","elo_ci":"+49\/-39","imputed":true},{"position":11,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1159.0,"score":0.235,"rank":12.89,"hrank":8.25,"imp":16.129,"train_time":12.67,"predict_time":0.196,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-29","imputed":false},{"position":12,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1154.0,"score":0.203,"rank":13.06,"hrank":8.5,"imp":15.947,"train_time":214.85,"predict_time":0.47,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+35\/-29","imputed":false},{"position":13,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1142.0,"score":0.193,"rank":13.5,"hrank":9.74,"imp":17.285,"train_time":281.61,"predict_time":10.673,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+33\/-30","imputed":false},{"position":14,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1141.0,"score":0.221,"rank":13.54,"hrank":8.25,"imp":16.182,"train_time":24.96,"predict_time":0.723,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+31\/-34","imputed":false},{"position":15,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1130.0,"score":0.161,"rank":13.92,"hrank":10.26,"imp":17.44,"train_time":281.61,"predict_time":2.887,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+25\/-25","imputed":false},{"position":16,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1080.0,"score":0.162,"rank":15.66,"hrank":8.77,"imp":18.556,"train_time":60.19,"predict_time":1.859,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+35\/-34","imputed":false},{"position":17,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1060.0,"score":0.105,"rank":16.35,"hrank":10.64,"imp":19.823,"train_time":164.06,"predict_time":2.023,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+33\/-29","imputed":false},{"position":18,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1039.0,"score":0.093,"rank":17.06,"hrank":10.89,"imp":20.259,"train_time":164.06,"predict_time":0.322,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+37\/-33","imputed":false},{"position":19,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1009.0,"score":0.083,"rank":18.09,"hrank":13.03,"imp":20.961,"train_time":152.3,"predict_time":2.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+29\/-37","imputed":false},{"position":20,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.091,"rank":18.38,"hrank":12.98,"imp":20.371,"train_time":9.03,"predict_time":2.473,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-38","imputed":false},{"position":21,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":999.0,"score":0.046,"rank":18.4,"hrank":14.06,"imp":20.563,"train_time":5.43,"predict_time":0.211,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+28\/-31","imputed":false},{"position":22,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":989.0,"score":0.066,"rank":18.75,"hrank":13.67,"imp":21.33,"train_time":152.3,"predict_time":0.474,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+28\/-43","imputed":false},{"position":23,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":931.0,"score":0.104,"rank":20.52,"hrank":10.65,"imp":29.335,"train_time":244.76,"predict_time":0.444,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-47","imputed":false},{"position":24,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":912.0,"score":0.057,"rank":21.05,"hrank":14.99,"imp":23.624,"train_time":7.99,"predict_time":0.791,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-59","imputed":false},{"position":25,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":912.0,"score":0.07,"rank":21.05,"hrank":13.59,"imp":24.775,"train_time":6.3,"predict_time":1.035,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-53","imputed":false},{"position":26,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":904.0,"score":0.083,"rank":21.28,"hrank":12.54,"imp":30.014,"train_time":244.76,"predict_time":0.13,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+38\/-49","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":855.0,"score":0.049,"rank":22.56,"hrank":15.09,"imp":32.075,"train_time":10.24,"predict_time":0.16,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-52","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/medium/leaderboard_table.html b/data_beyondarena/subsets/medium/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..10bc1a2d38b273373ea40113b3294c7fbf883cfa
--- /dev/null
+++ b/data_beyondarena/subsets/medium/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/medium</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1254.0,"score":0.503,"rank":6.71,"hrank":4.5,"imp":6.611,"train_time":655.92,"predict_time":2.599,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+78\/-64","imputed":false},{"position":1,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1206.0,"score":0.447,"rank":7.91,"hrank":4.67,"imp":7.096,"train_time":108.43,"predict_time":0.384,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-48","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1203.0,"score":0.559,"rank":7.97,"hrank":2.47,"imp":5.736,"train_time":2.94,"predict_time":0.919,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+134\/-76","imputed":false},{"position":3,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1193.0,"score":0.502,"rank":8.25,"hrank":3.34,"imp":6.407,"train_time":10.65,"predict_time":3.253,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+97\/-69","imputed":false},{"position":4,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1174.0,"score":0.422,"rank":8.75,"hrank":3.82,"imp":7.248,"train_time":108.43,"predict_time":0.076,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-52","imputed":false},{"position":5,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1165.0,"score":0.396,"rank":8.99,"hrank":5.08,"imp":8.235,"train_time":655.92,"predict_time":0.238,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+81\/-58","imputed":false},{"position":6,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1163.0,"score":0.344,"rank":9.07,"hrank":4.42,"imp":8.656,"train_time":594.53,"predict_time":1.782,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-66","imputed":false},{"position":7,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1161.0,"score":0.338,"rank":9.11,"hrank":7.41,"imp":8.342,"train_time":35.72,"predict_time":1.544,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-42","imputed":false},{"position":8,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1151.0,"score":0.331,"rank":9.38,"hrank":6.22,"imp":8.964,"train_time":20.53,"predict_time":1.579,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-38","imputed":false},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1124.0,"score":0.434,"rank":10.14,"hrank":3.35,"imp":6.821,"train_time":1.89,"predict_time":0.443,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-88","imputed":false},{"position":10,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1124.0,"score":0.382,"rank":10.15,"hrank":6.45,"imp":7.685,"train_time":3.95,"predict_time":0.038,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-60","imputed":false},{"position":11,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1121.0,"score":0.294,"rank":10.24,"hrank":7.19,"imp":8.753,"train_time":35.72,"predict_time":0.277,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-55","imputed":false},{"position":12,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1098.0,"score":0.253,"rank":10.9,"hrank":7.03,"imp":9.575,"train_time":594.53,"predict_time":0.229,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+80\/-66","imputed":false},{"position":13,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1096.0,"score":0.262,"rank":10.94,"hrank":6.58,"imp":9.681,"train_time":20.53,"predict_time":0.23,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-49","imputed":false},{"position":14,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.192,"rank":13.75,"hrank":9.66,"imp":10.01,"train_time":0.82,"predict_time":0.221,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+62\/-72","imputed":false},{"position":15,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":974.0,"score":0.177,"rank":14.49,"hrank":9.02,"imp":12.531,"train_time":10.54,"predict_time":0.251,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-66","imputed":false},{"position":16,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":953.0,"score":0.16,"rank":15.09,"hrank":9.29,"imp":11.45,"train_time":17.29,"predict_time":0.178,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-66","imputed":false},{"position":17,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":918.0,"score":0.067,"rank":16.08,"hrank":14.3,"imp":11.449,"train_time":0.36,"predict_time":0.077,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-58","imputed":false},{"position":18,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":833.0,"score":0.021,"rank":18.36,"hrank":16.67,"imp":14.61,"train_time":11.82,"predict_time":0.448,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-65","imputed":false},{"position":19,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":792.0,"score":0.107,"rank":19.38,"hrank":11.79,"imp":19.527,"train_time":40.6,"predict_time":47.86,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+100\/-144","imputed":false},{"position":20,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":776.0,"score":0.012,"rank":19.78,"hrank":18.28,"imp":15.614,"train_time":15.14,"predict_time":0.338,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+57\/-70","imputed":false},{"position":21,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":752.0,"score":0.005,"rank":20.32,"hrank":18.96,"imp":15.834,"train_time":11.82,"predict_time":0.054,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-61","imputed":false},{"position":22,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":744.0,"score":0.01,"rank":20.51,"hrank":18.84,"imp":16.297,"train_time":15.14,"predict_time":0.065,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+55\/-77","imputed":false},{"position":23,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":622.0,"score":0.009,"rank":23.07,"hrank":21.63,"imp":27.124,"train_time":22.74,"predict_time":0.175,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-109","imputed":false},{"position":24,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":592.0,"score":0.0,"rank":23.62,"hrank":22.27,"imp":20.302,"train_time":1.2,"predict_time":0.325,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+75\/-89","imputed":false},{"position":25,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":580.0,"score":0.004,"rank":23.83,"hrank":22.29,"imp":21.091,"train_time":0.65,"predict_time":0.322,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-105","imputed":false},{"position":26,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":563.0,"score":0.008,"rank":24.12,"hrank":22.93,"imp":28.086,"train_time":22.74,"predict_time":0.076,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-109","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":500.0,"score":0.007,"rank":25.08,"hrank":24.1,"imp":30.432,"train_time":0.77,"predict_time":0.072,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-136","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/random/leaderboard_table.html b/data_beyondarena/subsets/random/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..ffe625580d0032f7bb1603f7897e37071e5ceb34
--- /dev/null
+++ b/data_beyondarena/subsets/random/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/random</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1383.0,"score":0.713,"rank":5.86,"hrank":2.29,"imp":5.897,"train_time":22.16,"predict_time":2.481,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+41\/-52","imputed":false},{"position":1,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1313.0,"score":0.634,"rank":7.62,"hrank":2.79,"imp":8.293,"train_time":15.4,"predict_time":1.887,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-54","imputed":false},{"position":2,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1304.0,"score":0.575,"rank":7.87,"hrank":3.25,"imp":9.301,"train_time":27.15,"predict_time":2.035,"imputed_pct":5.83,"hardware":"GPU","elo_ci":"+52\/-38","imputed":true},{"position":3,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1291.0,"score":0.447,"rank":8.23,"hrank":4.97,"imp":12.177,"train_time":1547.62,"predict_time":21.543,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+32\/-31","imputed":false},{"position":4,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1226.0,"score":0.33,"rank":10.17,"hrank":5.83,"imp":13.814,"train_time":1366.44,"predict_time":7.477,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+31\/-31","imputed":false},{"position":5,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1218.0,"score":0.292,"rank":10.42,"hrank":6.37,"imp":13.797,"train_time":477.58,"predict_time":1.29,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+25\/-24","imputed":false},{"position":6,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1212.0,"score":0.327,"rank":10.64,"hrank":6.33,"imp":14.092,"train_time":1547.62,"predict_time":2.71,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+32\/-35","imputed":false},{"position":7,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1200.0,"score":0.275,"rank":11.01,"hrank":6.61,"imp":14.066,"train_time":477.58,"predict_time":0.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+23\/-26","imputed":false},{"position":8,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1188.0,"score":0.279,"rank":11.42,"hrank":6.47,"imp":14.461,"train_time":1366.44,"predict_time":1.018,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+34\/-30","imputed":false},{"position":9,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1181.0,"score":0.243,"rank":11.66,"hrank":8.17,"imp":15.13,"train_time":214.85,"predict_time":2.515,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+31\/-27","imputed":false},{"position":10,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1149.0,"score":0.224,"rank":12.71,"hrank":8.96,"imp":15.87,"train_time":333.88,"predict_time":10.673,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+33\/-27","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1146.0,"score":0.207,"rank":12.84,"hrank":7.79,"imp":15.518,"train_time":214.85,"predict_time":0.552,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-31","imputed":false},{"position":12,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1143.0,"score":0.227,"rank":12.92,"hrank":8.35,"imp":15.278,"train_time":14.64,"predict_time":0.221,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+26\/-23","imputed":false},{"position":13,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1130.0,"score":0.187,"rank":13.35,"hrank":9.01,"imp":16.093,"train_time":333.88,"predict_time":2.535,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+34\/-28","imputed":false},{"position":14,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1122.0,"score":0.325,"rank":13.64,"hrank":5.24,"imp":16.724,"train_time":136.49,"predict_time":11.617,"imputed_pct":5.83,"hardware":"GPU","elo_ci":"+38\/-51","imputed":true},{"position":15,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1116.0,"score":0.201,"rank":13.85,"hrank":8.95,"imp":15.698,"train_time":39.03,"predict_time":0.923,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+30\/-28","imputed":false},{"position":16,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1054.0,"score":0.129,"rank":15.98,"hrank":9.5,"imp":18.057,"train_time":59.49,"predict_time":2.394,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+32\/-32","imputed":false},{"position":17,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1044.0,"score":0.092,"rank":16.33,"hrank":11.26,"imp":19.001,"train_time":171.71,"predict_time":2.094,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+30\/-28","imputed":false},{"position":18,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1004.0,"score":0.069,"rank":17.63,"hrank":12.12,"imp":19.965,"train_time":171.71,"predict_time":0.369,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+28\/-28","imputed":false},{"position":19,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.09,"rank":17.78,"hrank":12.69,"imp":18.867,"train_time":9.29,"predict_time":2.071,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+36\/-32","imputed":false},{"position":20,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":988.0,"score":0.071,"rank":18.16,"hrank":13.18,"imp":20.273,"train_time":170.27,"predict_time":2.089,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+30\/-36","imputed":false},{"position":21,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":985.0,"score":0.048,"rank":18.26,"hrank":14.31,"imp":19.325,"train_time":5.43,"predict_time":0.257,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+26\/-28","imputed":false},{"position":22,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":969.0,"score":0.058,"rank":18.76,"hrank":13.85,"imp":20.729,"train_time":170.27,"predict_time":0.485,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-33","imputed":false},{"position":23,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":891.0,"score":0.097,"rank":21.05,"hrank":11.54,"imp":28.219,"train_time":241.64,"predict_time":0.834,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-49","imputed":false},{"position":24,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":879.0,"score":0.047,"rank":21.4,"hrank":15.96,"imp":23.648,"train_time":10.32,"predict_time":0.791,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+44\/-44","imputed":false},{"position":25,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":874.0,"score":0.064,"rank":21.53,"hrank":14.12,"imp":24.558,"train_time":7.64,"predict_time":0.878,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-49","imputed":false},{"position":26,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":862.0,"score":0.077,"rank":21.84,"hrank":13.57,"imp":28.954,"train_time":241.64,"predict_time":0.195,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-57","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":811.0,"score":0.057,"rank":23.08,"hrank":15.78,"imp":31.118,"train_time":10.24,"predict_time":0.212,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+48\/-60","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/small/leaderboard_table.html b/data_beyondarena/subsets/small/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..f47e851c3a5224a219d63f82a89dbf5c13e1a0a9
--- /dev/null
+++ b/data_beyondarena/subsets/small/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/small</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1450.0,"score":0.765,"rank":4.79,"hrank":1.96,"imp":8.177,"train_time":11.07,"predict_time":1.462,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+102\/-69","imputed":false},{"position":1,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1435.0,"score":0.743,"rank":5.08,"hrank":2.37,"imp":12.145,"train_time":9.87,"predict_time":0.953,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+100\/-73","imputed":false},{"position":2,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1350.0,"score":0.676,"rank":6.9,"hrank":2.38,"imp":11.348,"train_time":5.31,"predict_time":0.725,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+105\/-90","imputed":false},{"position":3,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1312.0,"score":0.479,"rank":7.85,"hrank":4.84,"imp":15.333,"train_time":970.7,"predict_time":12.352,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+66\/-58","imputed":false},{"position":4,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1237.0,"score":0.368,"rank":9.91,"hrank":6.07,"imp":16.481,"train_time":828.24,"predict_time":3.416,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-52","imputed":false},{"position":5,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1231.0,"score":0.295,"rank":10.1,"hrank":7.71,"imp":18.464,"train_time":229.46,"predict_time":0.723,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-38","imputed":false},{"position":6,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1215.0,"score":0.299,"rank":10.57,"hrank":7.59,"imp":18.452,"train_time":121.9,"predict_time":1.213,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+43\/-42","imputed":false},{"position":7,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1207.0,"score":0.279,"rank":10.8,"hrank":7.66,"imp":18.704,"train_time":229.46,"predict_time":0.123,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-40","imputed":false},{"position":8,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1200.0,"score":0.336,"rank":11.03,"hrank":7.01,"imp":17.752,"train_time":970.7,"predict_time":1.507,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+67\/-58","imputed":false},{"position":9,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1192.0,"score":0.293,"rank":11.25,"hrank":5.54,"imp":17.669,"train_time":828.24,"predict_time":0.549,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+68\/-58","imputed":false},{"position":10,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1173.0,"score":0.217,"rank":11.85,"hrank":8.96,"imp":21.767,"train_time":161.73,"predict_time":5.716,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-37","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1158.0,"score":0.249,"rank":12.31,"hrank":7.5,"imp":19.126,"train_time":121.9,"predict_time":0.236,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-52","imputed":false},{"position":12,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1152.0,"score":0.171,"rank":12.5,"hrank":10.06,"imp":22.088,"train_time":161.73,"predict_time":1.471,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+31\/-32","imputed":false},{"position":13,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1149.0,"score":0.207,"rank":12.6,"hrank":9.72,"imp":21.344,"train_time":8.81,"predict_time":0.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-36","imputed":false},{"position":14,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1105.0,"score":0.22,"rank":13.97,"hrank":10.06,"imp":19.639,"train_time":20.6,"predict_time":0.533,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+58\/-49","imputed":false},{"position":15,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1096.0,"score":0.244,"rank":14.25,"hrank":6.39,"imp":22.281,"train_time":107.59,"predict_time":48.147,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+90\/-72","imputed":false},{"position":16,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1040.0,"score":0.138,"rank":15.99,"hrank":11.15,"imp":22.845,"train_time":41.4,"predict_time":1.328,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+62\/-49","imputed":false},{"position":17,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.072,"rank":17.23,"hrank":13.91,"imp":24.69,"train_time":5.01,"predict_time":1.253,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-53","imputed":false},{"position":18,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":998.0,"score":0.049,"rank":17.29,"hrank":12.68,"imp":24.29,"train_time":3.02,"predict_time":0.105,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+45\/-47","imputed":false},{"position":19,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":996.0,"score":0.067,"rank":17.35,"hrank":13.16,"imp":26.614,"train_time":103.94,"predict_time":1.125,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+50\/-57","imputed":false},{"position":20,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":958.0,"score":0.045,"rank":18.47,"hrank":14.31,"imp":27.312,"train_time":103.94,"predict_time":0.209,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-59","imputed":false},{"position":21,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":944.0,"score":0.059,"rank":18.88,"hrank":14.19,"imp":27.979,"train_time":105.83,"predict_time":1.077,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+66\/-79","imputed":false},{"position":22,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":912.0,"score":0.049,"rank":19.77,"hrank":14.42,"imp":28.304,"train_time":105.83,"predict_time":0.245,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-75","imputed":false},{"position":23,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":837.0,"score":0.024,"rank":21.69,"hrank":16.68,"imp":30.191,"train_time":4.82,"predict_time":0.485,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-90","imputed":false},{"position":24,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":829.0,"score":0.033,"rank":21.88,"hrank":18.76,"imp":31.344,"train_time":4.57,"predict_time":0.603,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-90","imputed":false},{"position":25,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":774.0,"score":0.007,"rank":23.09,"hrank":19.86,"imp":37.312,"train_time":132.0,"predict_time":0.392,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+65\/-93","imputed":false},{"position":26,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":738.0,"score":0.006,"rank":23.81,"hrank":20.79,"imp":37.878,"train_time":132.0,"predict_time":0.095,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-105","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":682.0,"score":0.007,"rank":24.78,"hrank":22.25,"imp":40.225,"train_time":4.52,"predict_time":0.128,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-136","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/temporal/leaderboard_table.html b/data_beyondarena/subsets/temporal/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..fabc55cc5eddad54bbbc1afa2c91870b54da22a3
--- /dev/null
+++ b/data_beyondarena/subsets/temporal/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/temporal</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1249.0,"score":0.647,"rank":5.96,"hrank":2.75,"imp":3.976,"train_time":665.87,"predict_time":5.639,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+130\/-100","imputed":false},{"position":1,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1185.0,"score":0.563,"rank":7.46,"hrank":4.6,"imp":4.923,"train_time":71.04,"predict_time":2.206,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-76","imputed":false},{"position":2,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1182.0,"score":0.582,"rank":7.55,"hrank":3.11,"imp":4.661,"train_time":665.87,"predict_time":0.644,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+115\/-97","imputed":false},{"position":3,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1169.0,"score":0.545,"rank":7.87,"hrank":4.9,"imp":4.86,"train_time":299.64,"predict_time":0.664,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-90","imputed":false},{"position":4,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1168.0,"score":0.538,"rank":7.88,"hrank":4.46,"imp":4.987,"train_time":151.78,"predict_time":2.497,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+117\/-82","imputed":false},{"position":5,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1162.0,"score":0.543,"rank":8.04,"hrank":5.06,"imp":4.556,"train_time":299.64,"predict_time":0.143,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+104\/-89","imputed":false},{"position":6,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1108.0,"score":0.372,"rank":9.52,"hrank":5.7,"imp":6.041,"train_time":348.74,"predict_time":4.406,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+74\/-71","imputed":false},{"position":7,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1107.0,"score":0.457,"rank":9.56,"hrank":5.36,"imp":5.248,"train_time":151.78,"predict_time":0.488,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-68","imputed":false},{"position":8,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1088.0,"score":0.435,"rank":10.08,"hrank":7.66,"imp":5.555,"train_time":11.61,"predict_time":0.114,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+94\/-61","imputed":false},{"position":9,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1087.0,"score":0.39,"rank":10.11,"hrank":7.18,"imp":6.465,"train_time":71.04,"predict_time":0.662,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+72\/-49","imputed":false},{"position":10,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1049.0,"score":0.306,"rank":11.22,"hrank":4.52,"imp":7.573,"train_time":16.44,"predict_time":9.788,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+85\/-78","imputed":false},{"position":11,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1031.0,"score":0.221,"rank":11.75,"hrank":6.88,"imp":7.562,"train_time":348.74,"predict_time":0.952,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+61\/-56","imputed":false},{"position":12,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1007.0,"score":0.27,"rank":12.46,"hrank":8.1,"imp":6.146,"train_time":14.95,"predict_time":0.542,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+71\/-73","imputed":false},{"position":13,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.263,"rank":12.68,"hrank":10.77,"imp":7.663,"train_time":3.15,"predict_time":0.48,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+69\/-55","imputed":false},{"position":14,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":969.0,"score":0.351,"rank":13.6,"hrank":3.47,"imp":10.378,"train_time":10.85,"predict_time":1.164,"imputed_pct":51.43,"hardware":"GPU","elo_ci":"+114\/-105","imputed":true},{"position":15,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":968.0,"score":0.172,"rank":13.64,"hrank":10.24,"imp":7.942,"train_time":0.84,"predict_time":0.173,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+52\/-38","imputed":false},{"position":16,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":966.0,"score":0.177,"rank":13.7,"hrank":9.71,"imp":7.432,"train_time":9.93,"predict_time":1.206,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+59\/-61","imputed":false},{"position":17,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":835.0,"score":0.121,"rank":17.54,"hrank":7.52,"imp":14.002,"train_time":8.33,"predict_time":4.328,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+112\/-115","imputed":false},{"position":18,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":757.0,"score":0.006,"rank":19.68,"hrank":17.69,"imp":14.35,"train_time":28.91,"predict_time":1.035,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+76\/-110","imputed":false},{"position":19,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":729.0,"score":0.035,"rank":20.4,"hrank":17.44,"imp":14.851,"train_time":67.1,"predict_time":0.595,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-101","imputed":false},{"position":20,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":720.0,"score":0.0,"rank":20.63,"hrank":15.2,"imp":17.697,"train_time":30.7,"predict_time":8.63,"imputed_pct":51.43,"hardware":"GPU","elo_ci":"+80\/-98","imputed":true},{"position":21,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":712.0,"score":0.017,"rank":20.83,"hrank":17.56,"imp":15.028,"train_time":28.91,"predict_time":0.131,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+99\/-134","imputed":false},{"position":22,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":690.0,"score":0.002,"rank":21.36,"hrank":18.38,"imp":18.427,"train_time":33.81,"predict_time":1.141,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+73\/-110","imputed":false},{"position":23,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":688.0,"score":0.016,"rank":21.4,"hrank":17.55,"imp":15.496,"train_time":2.59,"predict_time":0.415,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-80","imputed":false},{"position":24,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":669.0,"score":0.036,"rank":21.85,"hrank":18.57,"imp":16.419,"train_time":67.1,"predict_time":0.138,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+102\/-112","imputed":false},{"position":25,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":639.0,"score":0.0,"rank":22.53,"hrank":19.58,"imp":19.452,"train_time":33.81,"predict_time":0.225,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+67\/-103","imputed":false},{"position":26,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":618.0,"score":0.0,"rank":22.97,"hrank":17.45,"imp":16.474,"train_time":1.3,"predict_time":0.471,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-115","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":582.0,"score":0.0,"rank":23.7,"hrank":20.15,"imp":21.346,"train_time":1.31,"predict_time":0.208,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+82\/-137","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/text/leaderboard_table.html b/data_beyondarena/subsets/text/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..5211c5a52c9cd7ec044a3ac21b121da62e32b045
--- /dev/null
+++ b/data_beyondarena/subsets/text/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/text</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1336.0,"score":0.584,"rank":5.72,"hrank":3.03,"imp":5.708,"train_time":1299.46,"predict_time":11.038,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+152\/-85","imputed":false},{"position":1,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1245.0,"score":0.475,"rank":7.91,"hrank":3.79,"imp":7.846,"train_time":1299.46,"predict_time":0.821,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+166\/-99","imputed":false},{"position":2,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1227.0,"score":0.489,"rank":8.38,"hrank":3.07,"imp":5.925,"train_time":15.73,"predict_time":5.007,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+135\/-115","imputed":false},{"position":3,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1223.0,"score":0.378,"rank":8.51,"hrank":6.34,"imp":9.154,"train_time":448.97,"predict_time":1.487,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+86\/-56","imputed":false},{"position":4,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1214.0,"score":0.37,"rank":8.74,"hrank":6.21,"imp":9.257,"train_time":448.97,"predict_time":0.174,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+88\/-66","imputed":false},{"position":5,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1198.0,"score":0.347,"rank":9.2,"hrank":5.42,"imp":9.581,"train_time":211.87,"predict_time":3.288,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+106\/-88","imputed":false},{"position":6,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1186.0,"score":0.31,"rank":9.54,"hrank":4.88,"imp":9.482,"train_time":1234.72,"predict_time":7.865,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+121\/-89","imputed":false},{"position":7,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1184.0,"score":0.379,"rank":9.58,"hrank":7.69,"imp":8.967,"train_time":83.51,"predict_time":2.666,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+77\/-66","imputed":false},{"position":8,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1183.0,"score":0.531,"rank":9.63,"hrank":2.24,"imp":6.151,"train_time":51.86,"predict_time":3.626,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+167\/-149","imputed":true},{"position":9,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1178.0,"score":0.439,"rank":9.78,"hrank":3.46,"imp":5.675,"train_time":13.2,"predict_time":3.041,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+180\/-129","imputed":false},{"position":10,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1147.0,"score":0.318,"rank":10.69,"hrank":5.78,"imp":9.56,"train_time":83.51,"predict_time":0.659,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+95\/-82","imputed":false},{"position":11,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1128.0,"score":0.244,"rank":11.25,"hrank":6.68,"imp":10.178,"train_time":1234.72,"predict_time":1.445,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+128\/-91","imputed":false},{"position":12,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1111.0,"score":0.276,"rank":11.76,"hrank":8.28,"imp":10.224,"train_time":211.87,"predict_time":0.575,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+100\/-71","imputed":false},{"position":13,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1092.0,"score":0.283,"rank":12.34,"hrank":10.28,"imp":10.369,"train_time":12.75,"predict_time":0.177,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+81\/-86","imputed":false},{"position":14,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1046.0,"score":0.199,"rank":13.76,"hrank":10.34,"imp":11.111,"train_time":17.61,"predict_time":0.708,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+87\/-76","imputed":false},{"position":15,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1002.0,"score":0.143,"rank":15.1,"hrank":11.54,"imp":11.743,"train_time":27.54,"predict_time":1.233,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+83\/-71","imputed":false},{"position":16,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.103,"rank":15.16,"hrank":13.95,"imp":12.293,"train_time":2.84,"predict_time":0.631,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+59\/-63","imputed":false},{"position":17,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":972.0,"score":0.077,"rank":15.99,"hrank":14.4,"imp":12.637,"train_time":1.89,"predict_time":0.37,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+68\/-68","imputed":false},{"position":18,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":922.0,"score":0.047,"rank":17.46,"hrank":14.36,"imp":15.25,"train_time":48.66,"predict_time":1.503,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+89\/-116","imputed":false},{"position":19,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":874.0,"score":0.01,"rank":18.8,"hrank":16.73,"imp":16.186,"train_time":126.68,"predict_time":1.153,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+60\/-123","imputed":false},{"position":20,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":861.0,"score":0.027,"rank":19.14,"hrank":16.08,"imp":16.256,"train_time":48.66,"predict_time":0.222,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+70\/-118","imputed":false},{"position":21,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":835.0,"score":0.004,"rank":19.83,"hrank":18.07,"imp":16.576,"train_time":126.68,"predict_time":0.226,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+71\/-125","imputed":false},{"position":22,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":776.0,"score":0.014,"rank":21.28,"hrank":18.41,"imp":18.713,"train_time":12.64,"predict_time":0.396,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+101\/-152","imputed":false},{"position":23,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":774.0,"score":0.01,"rank":21.34,"hrank":17.86,"imp":21.608,"train_time":38.62,"predict_time":11.788,"imputed_pct":25.0,"hardware":"GPU","elo_ci":"+83\/-128","imputed":true},{"position":24,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":712.0,"score":0.03,"rank":22.69,"hrank":14.68,"imp":19.946,"train_time":4.09,"predict_time":0.446,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+135\/-200","imputed":false},{"position":25,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":688.0,"score":0.0,"rank":23.19,"hrank":18.9,"imp":25.067,"train_time":57.28,"predict_time":1.496,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+85\/-145","imputed":false},{"position":26,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":646.0,"score":0.003,"rank":23.98,"hrank":20.22,"imp":25.451,"train_time":57.28,"predict_time":0.329,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+93\/-182","imputed":false},{"position":27,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":565.0,"score":0.0,"rank":25.25,"hrank":24.35,"imp":27.027,"train_time":2.14,"predict_time":0.403,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+98\/-164","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_beyondarena/subsets/tiny/leaderboard_table.html b/data_beyondarena/subsets/tiny/leaderboard_table.html
new file mode 100644
index 0000000000000000000000000000000000000000..89329e8f1052ed6923b7942828f7a82a22b3d3c0
--- /dev/null
+++ b/data_beyondarena/subsets/tiny/leaderboard_table.html
@@ -0,0 +1,1172 @@
+<!doctype html>
+<html lang="en">
+<head>
+<meta charset="utf-8">
+<meta name="viewport" content="width=device-width, initial-scale=1">
+<title>TabArena leaderboard table — data_beyondarena/subsets/tiny</title>
+<style>
+  :root {
+        --fam-foundation: #b07cf0;
+    --fam-nn: #5aa9e6;
+    --fam-tree: #5cb85c;
+    --fam-reference: #f0a35a;
+    --fam-baseline: #9e9e9e;
+    --fam-other: #9e9e9e;
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  @media (prefers-color-scheme: dark) {
+    :root {
+      --paper: #131316;
+      --card: #1b1b1f;
+      --ink: #f0efea;
+      --muted: #9b9a92;
+      --line: #2e2e33;
+      --accent: #3987e5;
+      --chip-bg: #232327;
+      --pt-muted: #55555c;
+      --fam-foundation-ink: var(--fam-foundation);
+      --fam-nn-ink: var(--fam-nn);
+      --fam-tree-ink: var(--fam-tree);
+      --fam-reference-ink: var(--fam-reference);
+      --fam-baseline-ink: var(--fam-baseline);
+      --fam-other-ink: var(--fam-other);
+      --var-default: #4386d5;
+      --var-tuned: #c05f38;
+      --var-tunedens: #289972;
+      --optimal: #2ea043;
+      --tooltip-bg: #f0efea;
+      --tooltip-ink: #14161a;
+      color-scheme: dark;
+    }
+  }
+  :root[data-theme="dark"] {
+    --paper: #131316;
+    --card: #1b1b1f;
+    --ink: #f0efea;
+    --muted: #9b9a92;
+    --line: #2e2e33;
+    --accent: #3987e5;
+    --chip-bg: #232327;
+    --pt-muted: #55555c;
+    --fam-foundation-ink: var(--fam-foundation);
+    --fam-nn-ink: var(--fam-nn);
+    --fam-tree-ink: var(--fam-tree);
+    --fam-reference-ink: var(--fam-reference);
+    --fam-baseline-ink: var(--fam-baseline);
+    --fam-other-ink: var(--fam-other);
+    --var-default: #4386d5;
+    --var-tuned: #c05f38;
+    --var-tunedens: #289972;
+    --optimal: #2ea043;
+    --tooltip-bg: #f0efea;
+    --tooltip-ink: #14161a;
+    color-scheme: dark;
+  }
+  :root[data-theme="light"] {
+    --paper: #ffffff;
+    --card: #ffffff;
+    --ink: #14161a;
+    --muted: #6d6c65;
+    --line: #e4e3db;
+    --accent: #2a78d6;
+    --chip-bg: #f2f1ec;
+    --pt-muted: #b9b8b1;
+    /* The same family hues as *text*. The mark colors are tuned for fills and
+    drop below readable contrast as small labels on the light surface, so
+    light mode darkens them; dark mode reuses the mark colors as-is. */
+    --fam-foundation-ink: #7d3fc2;
+    --fam-nn-ink: #1c6fa8;
+    --fam-tree-ink: #2f7d32;
+    --fam-reference-ink: #a4600f;
+    --fam-baseline-ink: #5f5f5f;
+    --fam-other-ink: #5f5f5f;
+    /* Tuning-variant series (default / tuned / tuned + ensembled). Light mode
+    is the paper view's surface, so it uses the *static figures' own* seaborn
+    pastels — figures exported from here drop straight into a paper beside
+    them. The cost is colorblind separation: green vs. orange is 4.0 deutan
+    ΔE, well inside the band that needs secondary encoding, which here is the
+    fixed concentric bar widths plus the legend and the data table. Dark mode
+    (the website) keeps the stepped, better-separated version below. */
+    --var-default: #a1c9f4;
+    --var-tuned: #ffb482;
+    --var-tunedens: #8de5a1;
+    --optimal: #228b22;
+    --tooltip-bg: #14161a;
+    --tooltip-ink: #fbfbf9;
+    color-scheme: light;
+  }
+  html, body { margin: 0; background: var(--paper); }
+  /* The colour emoji fonts are named *before* the generic `sans-serif`. A generic
+     family matches every character through the browser's own fallback chain, so
+     anything listed after it is unreachable — and that fallback resolves emoji to
+     a monochrome font on Linux, which flattens the family symbols on the chips.
+     Latin glyphs are unaffected: the emoji fonts carry none. */
+  body {
+    color: var(--ink);
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    line-height: 1.5;
+    padding: 10px 12px 14px;
+  }
+  /* The [hidden] attribute must beat author display rules (e.g. the
+     inline-flex on .metricpick), else hidden controls render empty. */
+  [hidden] { display: none !important; }
+
+  .explorer-title { font-size: 15px; font-weight: 650; margin: 0 0 8px; }
+  .controls { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; margin-bottom: 10px; }
+  .controls .hint { font-size: 12.5px; color: var(--muted); }
+  .btnrow { display: flex; gap: 6px; flex-wrap: wrap; }
+  .btn {
+    font: 600 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 11px; cursor: pointer;
+  }
+  .btn:hover { border-color: var(--muted); }
+  .btn:focus-visible, .chip:focus-visible, .famchip:focus-visible, select:focus-visible {
+    outline: 2px solid var(--accent); outline-offset: 2px;
+  }
+  .metricpick { display: inline-flex; align-items: center; gap: 6px; font-size: 12.5px; font-weight: 600; color: var(--muted); }
+  .metricpick select {
+    font: 600 12.5px/1.2 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 5px 7px; cursor: pointer;
+  }
+
+  .chips { display: flex; flex-direction: column; gap: 9px; }
+  /* One block per family: the family toggle on top, its chips wrapping below. */
+  .chiprow { display: flex; flex-direction: column; align-items: flex-start; gap: 5px; }
+  .famchip {
+    display: inline-flex; align-items: center; gap: 6px;
+    font: 650 10.5px/1.3 system-ui, sans-serif; letter-spacing: 0.06em; text-transform: uppercase;
+    color: var(--muted); background: var(--chip-bg); border: 1px dashed var(--line);
+    border-radius: 999px; padding: 5px 10px; cursor: pointer;
+  }
+  .famchip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--fam); flex: none; }
+  .famchip .count { font-weight: 500; letter-spacing: 0; opacity: 0.75; }
+  /* The family symbol sits at text size, not the chip's small-caps size. */
+  .famchip .sym { font-size: 1.05em; letter-spacing: 0; }
+  .famchip:hover { border-color: var(--fam); color: var(--ink); }
+  .famchip[aria-pressed="true"] {
+    border: 1px solid var(--fam);
+    background: color-mix(in srgb, var(--fam) 13%, transparent);
+    color: var(--ink);
+  }
+  /* A toggle button that carries its own colour: off is faded with a neutral
+     border, on takes the colour as border and tint. Opt-in via `.toggle` so the
+     older explorers, which fade their variant buttons with inline styles, are
+     unaffected. */
+  .btn.toggle[aria-pressed] { opacity: 0.5; }
+  .btn.toggle[aria-pressed="true"] {
+    opacity: 1;
+    border-color: var(--fam);
+    background: color-mix(in srgb, var(--fam) 18%, var(--chip-bg));
+  }
+  .btn.toggle .swatch {
+    display: inline-block; width: 8px; height: 8px; border-radius: 50%;
+    background: var(--fam); margin-right: 6px; vertical-align: middle;
+  }
+  .btn.toggle[aria-pressed="false"] .swatch { background: var(--pt-muted); }
+  .grouplabel { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+  .chipset { display: flex; flex-wrap: wrap; gap: 4px; }
+  .chip {
+    display: inline-flex; align-items: center; gap: 5px;
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: none; border: 1px solid var(--line); border-radius: 999px;
+    padding: 5px 10px 5px 8px; cursor: pointer;
+  }
+  .chip .dot { width: 8px; height: 8px; border-radius: 50%; background: var(--pt-muted); flex: none; }
+  .chip .imp-mark { color: var(--muted); font-weight: 700; margin-left: -2px; }
+  .chip[aria-pressed="true"] { border-color: var(--fam); background: color-mix(in srgb, var(--fam) 13%, transparent); font-weight: 650; }
+  .chip[aria-pressed="true"] .dot { background: var(--fam); }
+  .chip:hover { border-color: var(--muted); }
+
+  .legendstrip {
+    display: flex; flex-wrap: wrap; gap: 5px 16px; align-items: center;
+    font-size: 12.5px; color: var(--muted); padding: 2px 2px 8px;
+  }
+  .legendstrip .item { display: inline-flex; align-items: center; gap: 6px; }
+
+  .tooltip {
+    position: absolute; pointer-events: none; display: none;
+    background: var(--tooltip-bg); color: var(--tooltip-ink);
+    border-radius: 8px; padding: 8px 11px; font-size: 12px; line-height: 1.45;
+    max-width: 260px; z-index: 5; font-variant-numeric: tabular-nums;
+    box-shadow: 0 4px 14px rgba(0,0,0,0.25);
+  }
+  .tooltip .t-name { font-weight: 700; font-size: 12.5px; }
+  .tooltip .t-var { opacity: 0.75; }
+  .tooltip .t-imp { opacity: 0.85; font-style: italic; }
+
+  details.datatable { margin-top: 8px; font-size: 12.5px; }
+  details.datatable summary { cursor: pointer; color: var(--muted); font-weight: 600; }
+  details.datatable .tblwrap { overflow-x: auto; margin-top: 8px; }
+  details.datatable table { border-collapse: collapse; font-variant-numeric: tabular-nums; min-width: 560px; }
+  details.datatable th, details.datatable td {
+    text-align: left; padding: 3px 12px 3px 0; border-bottom: 1px solid var(--line);
+  }
+  details.datatable th { font-size: 11px; letter-spacing: 0.05em; text-transform: uppercase; color: var(--muted); }
+
+  svg text {
+    font-family: system-ui, -apple-system, "Segoe UI",
+      "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+  }
+
+  @media (prefers-reduced-motion: no-preference) {
+    .chip, .btn, .famchip { transition: border-color 120ms ease, background-color 120ms ease; }
+  }
+
+  /* --- Paper view -----------------------------------------------------------
+     A figure-ready state for slides and papers: white surface, and only the
+     things needed to read the plot — the caption, the legend and the chart.
+     The controls, the chip list and the data table are interactive scaffolding,
+     not part of the figure. Entered via the "Paper view" button (which stamps
+     data-theme="light" on the root, see the scopes above). */
+  /* One toggle, in the same place in both states and never hidden — an exit
+     tucked into a corner of the figure was easy to miss. */
+  .viewbar { display: flex; align-items: center; gap: 10px; margin: 0 0 9px; }
+  body.paper .controls,
+  body.paper .chips,
+  body.paper .sidebox,
+  body.paper details.datatable { display: none !important; }
+  body.paper { padding: 14px 18px 18px; }
+  /* Export controls, revealed with the paper view. */
+  .exportbar { display: flex; align-items: center; gap: 8px; margin: 0 0 10px; }
+  .exportbar .hint { font-size: 12.5px; font-weight: 600; color: var(--muted); }
+
+
+  /* --- The table ------------------------------------------------------------
+     A scroll box with a sticky header, so the column names stay put through 80
+     rows. Numbers are tabular so digits line up down a column. */
+  .lbt-scroll {
+    overflow: auto; max-height: 720px; margin-top: 4px;
+    border: 1px solid var(--line); border-radius: 10px;
+    scrollbar-width: thin; scrollbar-color: var(--pt-muted) transparent;
+  }
+  .lbt-scroll::-webkit-scrollbar { width: 11px; height: 11px; }
+  .lbt-scroll::-webkit-scrollbar-thumb {
+    background: var(--pt-muted); border-radius: 8px;
+    border: 3px solid transparent; background-clip: content-box;
+  }
+  table.lbt { border-collapse: collapse; width: 100%; font-variant-numeric: tabular-nums; }
+  table.lbt th, table.lbt td { padding: 5px 9px; text-align: center; border: 1px solid var(--line); }
+  table.lbt thead th {
+    position: sticky; top: 0; z-index: 3; background: var(--card); font-size: 12px;
+    font-weight: 650; white-space: nowrap; cursor: pointer; user-select: none;
+    box-shadow: inset 0 -1px 0 var(--line);
+  }
+  /* The header is both the sort control and the definition, so it gets the
+     pointer of the former and the dotted underline of the latter. */
+  table.lbt thead th .lbl { text-decoration: underline; text-decoration-style: dotted;
+    text-decoration-color: var(--muted); text-underline-offset: 3px; }
+  table.lbt thead th:hover { color: var(--accent); }
+  table.lbt thead th::after { content: "\2195"; font-size: 0.72em; opacity: 0.3; margin-left: 5px; }
+  table.lbt thead th[aria-sort="ascending"]::after { content: "\25B2"; opacity: 0.85; }
+  table.lbt thead th[aria-sort="descending"]::after { content: "\25BC"; opacity: 0.85; }
+  table.lbt td.name { text-align: left; white-space: nowrap; }
+  table.lbt td.pos { color: var(--muted); font-variant-numeric: tabular-nums; }
+  table.lbt td.type { white-space: nowrap; width: 1%; }
+  table.lbt td.num { font-weight: 600; }
+  table.lbt td.na { color: var(--pt-muted); }
+  table.lbt tbody tr:hover td { filter: brightness(1.12); }
+  /* A fixed-width medal slot keeps the digits in their own sub-column whether or
+     not a medal is present. */
+  .cell { display: inline-flex; align-items: baseline; }
+  .medal { flex: 0 0 1.45em; width: 1.45em; text-align: left; font-size: 0.78em; }
+  .ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+  .pill { padding: 1px 7px; border-radius: 999px; font-size: 0.95em; white-space: nowrap; }
+  .verified { font-size: 0.85em; }
+  .link-icon { font-size: 0.78em; opacity: 0.65; margin-left: 2px; }
+  /* Grey rather than a dimmed inherit: dimming leaves the tag tinted with the
+     family colour, which the model name already carries, and made this table
+     disagree with the cross-subset table on the app page. */
+  .variant-tag { color: var(--muted); font-weight: 400; font-size: 0.9em; }
+  td.name .imp-mark { color: var(--muted); font-weight: 700; margin-left: 3px; }
+  td.name a { text-decoration: underline; text-decoration-style: dotted; text-underline-offset: 3px; }
+  td.name a:hover { text-decoration-style: solid; }
+
+  .lbt-search {
+    font: 500 12.5px/1 system-ui, sans-serif; color: var(--ink);
+    background: var(--chip-bg); border: 1px solid var(--line); border-radius: 7px;
+    padding: 6px 9px; width: 12em;
+  }
+  .lbt-cap { font-size: 12.5px; color: var(--muted); margin: 7px 0 0; }
+  .lbt-empty { padding: 34px 0; text-align: center; color: var(--muted); font-size: 13px; }
+  .chips-head { flex: 1 1 100%; font-size: 12.5px; color: var(--muted); font-weight: 600; }
+  /* The model selector sits above the table it filters. Families pack side by
+     side rather than stacking: one row each would leave the small ones (Baseline,
+     Other) wasting a line. */
+  .chips { margin: 2px 0 10px; flex-direction: row; flex-wrap: wrap; gap: 12px 28px; align-items: flex-start; }
+  .chiprow { flex: 0 1 auto; min-width: 0; }
+
+  /* The data export, in the same green the website's figure-export controls use,
+     so "this downloads something" reads the same across the site. */
+  .btn.export {
+    color: #063; background: #7ee0b8; border-color: #7ee0b8; font-weight: 700;
+  }
+  .btn.export:hover { background: #a6efcd; border-color: #a6efcd; }
+  /* Paper view keeps the table: here the table *is* the figure. */
+  body.paper .lbt-scroll { max-height: none; }
+</style>
+</head>
+<body>
+  <div class="viewbar">
+    <button class="btn" id="btn-paper" title="White background, table only — for slides and papers">Paper view</button>
+  </div>
+  <p class="explorer-title" id="title"></p>
+  <div class="controls">
+    <span class="grouplabel">Variants</span>
+    <div class="btnrow" id="variant-btns"></div>
+    <button class="btn" id="btn-imputed" title="Models whose score is partly imputed, marked &#8225;"></button>
+    <input class="lbt-search" id="search" type="search" placeholder="Search model or family" aria-label="Search">
+    <span class="grouplabel">Columns</span>
+    <div class="btnrow" id="col-btns"></div>
+    <button class="btn export" id="btn-csv" title="Download the rows and columns shown, in the current sort order">Download CSV</button>
+  </div>
+  <!-- The shared paper-view helper reveals this bar; a table has no figure to
+       export as SVG/PNG/PDF, and its CSV button belongs beside the filters that
+       shape the export, so it stays empty here. -->
+  <div class="exportbar" id="exportbar" hidden></div>
+  <!-- The model selector belongs above the table it filters. It doubles as the
+       family key, which is why there is no separate legend strip. -->
+  <div class="chips" id="chips"></div>
+  <div class="lbt-scroll" id="tblwrap"></div>
+  <p class="lbt-cap" id="caption"></p>
+
+<script>
+(function () {
+  "use strict";
+  const CONFIG = {"title": null, "columns": [{"key": "position", "label": "#", "always": true, "hint": "Position in this subset's published ranking."}, {"key": "family", "label": "Type", "always": true, "hint": "Model family \u2014 see the key above the table."}, {"key": "model", "label": "Model", "always": true, "hint": "The model, its tuning variant in brackets, \u2714\ufe0f when the implementation was verified, and \u2021 when some results are imputed."}, {"key": "elo", "label": "Elo", "decimals": 0, "heatmap": true, "hint": "Pairwise win-rate rating; a 400-point gap is about a 91% win rate. Higher is better."}, {"key": "score", "label": "Score", "decimals": 3, "heatmap": true, "hint": "Error rescaled per dataset to 1 (best) \u2026 0 (median), then averaged. Higher is better."}, {"key": "rank", "label": "Rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Mean rank across datasets. Lower is better."}, {"key": "hrank", "label": "Harmonic rank", "short": "H. rank", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "Harmonic mean of per-dataset ranks; rewards being excellent somewhere. Lower is better."}, {"key": "imp", "label": "Improvability (%)", "short": "Improv.", "decimals": 2, "lowerBetter": true, "heatmap": true, "hint": "How much lower the best model's error is than this one's, per dataset. Lower is better."}, {"key": "train_time", "label": "Train time (s/1K)", "short": "Train s", "decimals": 2, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to fit per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "predict_time", "label": "Predict time (s/1K)", "short": "Predict s", "decimals": 3, "lowerBetter": true, "heatmap": true, "logScale": true, "hint": "Median seconds to predict per 1000 rows. Lower is better; shaded on a log scale."}, {"key": "imputed_pct", "label": "Imputed (%)", "short": "Imputed", "decimals": 1, "lowerBetter": true, "heatmap": true, "hint": "Share of datasets whose score was imputed because the model could not run on them."}, {"key": "hardware", "label": "Hardware", "text": true, "hint": "The hardware the reported runtimes were measured on."}], "rankKey": "elo", "variants": ["Default", "Tuned", "Tuned + Ens."]};
+  const POINTS = [{"position":0,"method":"TabPFN-3","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1380.0,"score":0.654,"rank":7.55,"hrank":2.95,"imp":9.914,"train_time":191.74,"predict_time":7.508,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+63\/-48","imputed":false},{"position":1,"method":"TabICLv2","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1351.0,"score":0.617,"rank":8.4,"hrank":3.19,"imp":11.206,"train_time":89.68,"predict_time":4.467,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+64\/-44","imputed":false},{"position":2,"method":"TabPFN-2.6","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1334.0,"score":0.521,"rank":8.94,"hrank":3.73,"imp":11.427,"train_time":149.24,"predict_time":4.866,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-41","imputed":false},{"position":3,"method":"RealMLP","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1305.0,"score":0.377,"rank":9.87,"hrank":5.62,"imp":14.16,"train_time":11865.63,"predict_time":137.819,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+36\/-30","imputed":false},{"position":4,"method":"RealMLP","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1262.0,"score":0.279,"rank":11.38,"hrank":6.15,"imp":15.469,"train_time":11865.63,"predict_time":31.809,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+38\/-30","imputed":false},{"position":5,"method":"TabM","variant":"Tuned + Ens.","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1260.0,"score":0.275,"rank":11.44,"hrank":6.93,"imp":15.236,"train_time":8046.73,"predict_time":23.941,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+35\/-28","imputed":false},{"position":6,"method":"TabDPT","variant":"Default","family":"Foundation Model","url":null,"family_symbol":"\ud83e\udde0\u26a1","verified":true,"elo":1254.0,"score":0.423,"rank":11.66,"hrank":3.95,"imp":15.706,"train_time":1413.78,"predict_time":5.692,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+57\/-48","imputed":false},{"position":7,"method":"TabM","variant":"Tuned","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1244.0,"score":0.258,"rank":12.01,"hrank":6.48,"imp":15.483,"train_time":8046.73,"predict_time":7.481,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+34\/-32","imputed":false},{"position":8,"method":"CatBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1235.0,"score":0.224,"rank":12.35,"hrank":8.87,"imp":15.901,"train_time":3166.98,"predict_time":5.767,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+25\/-27","imputed":false},{"position":9,"method":"CatBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1227.0,"score":0.214,"rank":12.62,"hrank":8.74,"imp":16.119,"train_time":3166.98,"predict_time":1.319,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+22\/-26","imputed":false},{"position":10,"method":"TabM","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1212.0,"score":0.228,"rank":13.18,"hrank":7.35,"imp":16.132,"train_time":248.55,"predict_time":5.87,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+29\/-39","imputed":false},{"position":11,"method":"LightGBM","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1207.0,"score":0.206,"rank":13.37,"hrank":9.78,"imp":17.019,"train_time":1876.36,"predict_time":6.354,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+37\/-30","imputed":false},{"position":12,"method":"ExtraTrees","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1190.0,"score":0.168,"rank":14.0,"hrank":8.5,"imp":17.886,"train_time":1805.44,"predict_time":11.686,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+41\/-29","imputed":false},{"position":13,"method":"LightGBM","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1184.0,"score":0.169,"rank":14.22,"hrank":9.07,"imp":17.254,"train_time":1876.36,"predict_time":1.72,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+42\/-32","imputed":false},{"position":14,"method":"ExtraTrees","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1175.0,"score":0.145,"rank":14.54,"hrank":8.45,"imp":18.385,"train_time":1805.44,"predict_time":3.808,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+39\/-32","imputed":false},{"position":15,"method":"CatBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1167.0,"score":0.15,"rank":14.85,"hrank":9.41,"imp":17.857,"train_time":96.77,"predict_time":1.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+30\/-30","imputed":false},{"position":16,"method":"RealMLP","variant":"Default","family":"Neural Network","url":null,"family_symbol":"\ud83e\udde0\ud83d\udd01","verified":true,"elo":1143.0,"score":0.159,"rank":15.75,"hrank":7.99,"imp":19.723,"train_time":451.21,"predict_time":22.013,"imputed_pct":0.0,"hardware":"GPU","elo_ci":"+44\/-43","imputed":false},{"position":17,"method":"XGBoost","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1131.0,"score":0.12,"rank":16.2,"hrank":11.35,"imp":19.375,"train_time":2737.81,"predict_time":46.375,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+36\/-34","imputed":false},{"position":18,"method":"XGBoost","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1126.0,"score":0.097,"rank":16.39,"hrank":11.28,"imp":19.421,"train_time":2737.81,"predict_time":19.771,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+35\/-34","imputed":false},{"position":19,"method":"RandomForest","variant":"Tuned + Ens.","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1119.0,"score":0.113,"rank":16.62,"hrank":10.68,"imp":19.498,"train_time":1785.91,"predict_time":13.18,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+27\/-38","imputed":false},{"position":20,"method":"RandomForest","variant":"Tuned","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1105.0,"score":0.083,"rank":17.14,"hrank":11.84,"imp":20.025,"train_time":1785.91,"predict_time":3.785,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+32\/-39","imputed":false},{"position":21,"method":"Linear","variant":"Tuned + Ens.","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":1094.0,"score":0.226,"rank":17.51,"hrank":6.98,"imp":22.19,"train_time":1872.76,"predict_time":4.306,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+53\/-62","imputed":false},{"position":22,"method":"Linear","variant":"Tuned","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":1075.0,"score":0.188,"rank":18.22,"hrank":7.65,"imp":22.934,"train_time":1872.76,"predict_time":0.873,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+58\/-70","imputed":false},{"position":23,"method":"ExtraTrees","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1068.0,"score":0.128,"rank":18.45,"hrank":9.52,"imp":21.765,"train_time":71.91,"predict_time":9.968,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+49\/-53","imputed":false},{"position":24,"method":"RandomForest","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1053.0,"score":0.096,"rank":18.95,"hrank":12.23,"imp":21.654,"train_time":74.2,"predict_time":8.379,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+47\/-59","imputed":false},{"position":25,"method":"LightGBM","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1031.0,"score":0.059,"rank":19.68,"hrank":13.79,"imp":22.75,"train_time":55.17,"predict_time":0.887,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+33\/-48","imputed":false},{"position":26,"method":"Linear","variant":"Default","family":"Baseline","url":null,"family_symbol":"\ud83d\udccf","verified":true,"elo":1021.0,"score":0.137,"rank":20.03,"hrank":10.21,"imp":25.945,"train_time":67.78,"predict_time":1.087,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+56\/-75","imputed":false},{"position":27,"method":"XGBoost","variant":"Default","family":"Tree-based","url":null,"family_symbol":"\ud83c\udf33","verified":true,"elo":1000.0,"score":0.05,"rank":20.67,"hrank":14.13,"imp":23.126,"train_time":88.9,"predict_time":16.436,"imputed_pct":0.0,"hardware":"CPU","elo_ci":"+40\/-54","imputed":false}];
+
+
+  const NS = "http://www.w3.org/2000/svg";
+  // Baseline and Other are one bucket, as in the site's own type legend: they
+  // already share a color, and each holds only a handful of methods.
+  const FAM_MERGED = "Baseline / Other";
+  const famOf = (family) => (family === "Baseline" || family === "Other" ? FAM_MERGED : family);
+  // Normalized up front so every later lookup — colors, chips, sorting — sees
+  // the merged family. Both templates declare POINTS above this block.
+  for (const p of POINTS) p.family = famOf(p.family);
+
+  const FAM_ORDER = ["Foundation Model", "Tree-based", "Neural Network", "Reference Pipeline", FAM_MERGED];
+  const FAM_VAR = {
+    "Foundation Model": "var(--fam-foundation)",
+    "Tree-based": "var(--fam-tree)",
+    "Neural Network": "var(--fam-nn)",
+    "Reference Pipeline": "var(--fam-reference)",
+    [FAM_MERGED]: "var(--fam-baseline)",
+  };
+  // The symbol the website shows for each family, so a family chip here reads the
+  // same as the Type column on the site. Baseline and Other are one bucket, so
+  // that chip carries both symbols.
+  const FAM_SYMBOL = {
+    "Foundation Model": "🧠⚡",
+    "Tree-based": "🌳",
+    "Neural Network": "🧠🔁",
+    "Reference Pipeline": "📊",
+    [FAM_MERGED]: "📏 ❓",
+  };
+  // Tuning-variant colours, matching the --var-* tokens the charts plot with.
+  const VARIANT_VAR = {
+    "Default": "var(--var-default)",
+    "Tuned": "var(--var-tuned)",
+    "Tuned + Ens.": "var(--var-tunedens)",
+  };
+
+  // A family chip's label: its symbol, its name and how many methods it holds.
+  function famChipLabel(family, count) {
+    const symbol = FAM_SYMBOL[family];
+    return '<span class="dot"></span>' + (symbol ? '<span class="sym">' + symbol + "</span> " : "") +
+      family + ' <span class="count">&times;' + count + "</span>";
+  }
+
+  // The same hues stepped for use as text (see the --fam-*-ink tokens).
+  const FAM_INK = {
+    "Foundation Model": "var(--fam-foundation-ink)",
+    "Tree-based": "var(--fam-tree-ink)",
+    "Neural Network": "var(--fam-nn-ink)",
+    "Reference Pipeline": "var(--fam-reference-ink)",
+    [FAM_MERGED]: "var(--fam-baseline-ink)",
+  };
+
+  // Create an SVG element with attributes, optionally appended to `parent`.
+  function el(name, attrs, parent) {
+    const node = document.createElementNS(NS, name);
+    for (const k in attrs) node.setAttribute(k, attrs[k]);
+    if (parent) parent.appendChild(node);
+    return node;
+  }
+
+  // Plain, ungrouped numbers with a "." decimal separator. `toFixed` is
+  // locale-independent by definition, which is the point: `toLocaleString`
+  // would follow the *viewer's* browser locale and print 1234,5 for a German
+  // visitor, disagreeing with the figures and CSVs beside it.
+  function fmtNum(v, decimals) {
+    if (v == null || !isFinite(v)) return "—";
+    return v.toFixed(decimals);
+  }
+
+  function fmtMetric(metric, v) {
+    if (v == null || !isFinite(v)) return "—";
+    return fmtNum(v, metric.decimals) + (metric.suffix || "");
+  }
+
+  function fmtTime(v) {
+    if (v >= 100) return fmtNum(v, 0) + " s";
+    if (v >= 1) return fmtNum(v, 1) + " s";
+    if (v >= 0.1) return fmtNum(v, 2) + " s";
+    return fmtNum(v, 3) + " s";
+  }
+
+  // Smallest "nice" (1/2/2.5/5 x a power of ten) step that is at least `raw`.
+  function niceStep(raw) {
+    if (!(raw > 0)) return 1;
+    const mag = Math.pow(10, Math.floor(Math.log10(raw)));
+    for (const m of [1, 2, 2.5, 5, 10]) {
+      if (mag * m >= raw) return mag * m;
+    }
+    return mag * 10;
+  }
+
+  // ~`target` evenly spaced "nice" tick values covering [min, max]. Stepped by
+  // index rather than by accumulation so fractional steps do not drift.
+  function ticks(min, max, target) {
+    const step = niceStep((max - min) / target);
+    const first = Math.ceil(min / step);
+    const out = [];
+    for (let i = 0; first * step + i * step <= max + 1e-9; i++) out.push((first + i) * step);
+    return out.length ? out : [min];
+  }
+
+  // A hover tooltip anchored inside `box` (which must be position:relative).
+  function makeTooltip(box) {
+    const node = box.querySelector(".tooltip");
+    return {
+      show(html, ev) { node.innerHTML = html; node.style.display = "block"; this.move(ev); },
+      move(ev) {
+        const r = box.getBoundingClientRect();
+        let tx = ev.clientX - r.left + 14;
+        const ty = ev.clientY - r.top + 12;
+        if (tx > r.width - 270) tx = ev.clientX - r.left - 274;
+        node.style.left = tx + "px";
+        node.style.top = ty + "px";
+      },
+      hide() { node.style.display = "none"; },
+    };
+  }
+
+  // Paper view — white surface, legend + chart only — is the *default*: what a
+  // reader wants first is the figure, and it is the state worth exporting. The
+  // controls, chip list and data table are one click away behind "Edit view".
+  // `afterToggle` re-renders charts whose size is measured from the layout.
+  // `options.openInPaper` (default true) decides the state the page opens in. A
+  // chart opens as the figure; the leaderboard table opens with its controls,
+  // since there the interaction is the point rather than scaffolding around it.
+  function setUpPaperView(afterToggle, options) {
+    const opts = options || {};
+    const root = document.documentElement;
+    let hostTheme = null;   // the embedding page's choice, captured on entry
+    const btn = document.getElementById("btn-paper");
+    const embedded = window.parent !== window;
+
+    function setPaper(on) {
+      document.body.classList.toggle("paper", on);
+      if (on) {
+        hostTheme = root.getAttribute("data-theme");
+        root.setAttribute("data-theme", "light");
+      } else if (hostTheme) {
+        root.setAttribute("data-theme", hostTheme);
+      } else {
+        root.removeAttribute("data-theme");
+      }
+      btn.textContent = on ? "Edit view" : "Paper view";
+      document.getElementById("exportbar").hidden = !on || embedded;
+      if (afterToggle) requestAnimationFrame(afterToggle);
+      postHeight();
+    }
+    btn.addEventListener("click", () => setPaper(!document.body.classList.contains("paper")));
+    // Embedded, the host page owns these controls — they sit beside the panel's
+    // static-figure toggle and are driven from the outside. Standalone (the
+    // shareable single file) this page needs its own.
+    if (embedded) document.querySelector(".viewbar").hidden = true;
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-paper" && typeof d.on === "boolean") setPaper(d.on);
+    });
+    // Only standalone: embedded, the host owns the button and would not see the
+    // key press, so its label would fall out of step with the frame.
+    if (!embedded) {
+      document.addEventListener("keydown", ev => {
+        if (ev.key === "Escape" && !document.body.classList.contains("paper")) setPaper(true);
+      });
+    }
+    setPaper(opts.openInPaper !== false);   // for a chart, the figure is what opens
+  }
+
+  // --- Figure export ---------------------------------------------------------
+  // The chart is live SVG, so a file can be built from it directly. Three things
+  // a copy has to fix up: the colors are CSS custom properties (var(--x) means
+  // nothing outside this document), it has no background or font of its own, and
+  // the legend is HTML rather than part of the SVG.
+
+  // Rebuild the HTML legend as SVG, reusing its live layout: each item's glyph is
+  // cloned and its label re-emitted at the measured position. foreignObject would
+  // be far simpler, but Chrome refuses to rasterize it onto a canvas, which would
+  // break the PNG path.
+  // Rewrite every var(--x) in a clone's paint attributes; they resolve to nothing
+  // once the node leaves this document.
+  function resolveVars(root, resolve) {
+    for (const node of [root, ...root.querySelectorAll("*")]) {
+      for (const attr of ["fill", "stroke"]) {
+        const value = node.getAttribute(attr);
+        if (value && value.includes("var(")) node.setAttribute(attr, resolve(value));
+      }
+    }
+  }
+
+  function legendToSvg(container, resolve) {
+    const base = container.getBoundingClientRect();
+    const group = document.createElementNS(NS, "g");
+    let height = 0;
+    for (const item of container.querySelectorAll(".item")) {
+      const box = item.getBoundingClientRect();
+      if (!box.width) continue;
+      height = Math.max(height, box.bottom - base.top);
+      let textLeft = box.left - base.left;
+      const glyph = item.querySelector("svg");
+      if (glyph) {
+        const gbox = glyph.getBoundingClientRect();
+        const wrap = el("g", {
+          transform: `translate(${gbox.left - base.left} ${gbox.top - base.top})`,
+        }, group);
+        const glyphClone = glyph.cloneNode(true);
+        resolveVars(glyphClone, resolve);
+        wrap.appendChild(glyphClone);
+        textLeft = gbox.right - base.left + 5;
+      }
+      const label = item.textContent.trim();
+      if (!label) continue;
+      const colored = item.querySelector("[style*='color']");
+      const text = el("text", {
+        x: textLeft, y: box.top - base.top + box.height / 2 + 4, "font-size": 12.5,
+        fill: resolve(getComputedStyle(colored || item).color),
+      }, group);
+      text.textContent = label;
+    }
+    return { group, height: Math.ceil(height) };
+  }
+
+  // `parts` is a list of {svg, dx}, so a chart split across panes (the sticky
+  // y-axis beside the scrolling plot) still exports as one figure.
+  function buildExportSvg(parts, legendEl, pad = 10) {
+    const rootStyle = getComputedStyle(document.documentElement);
+    const resolve = value => String(value).replace(
+      /var\((--[\w-]+)\)/g, (_, name) => rootStyle.getPropertyValue(name).trim() || "none");
+    const paper = rootStyle.getPropertyValue("--paper").trim() || "#ffffff";
+
+    let chartW = 0, chartH = 0;
+    for (const part of parts) {
+      chartW = Math.max(chartW, part.dx + Number(part.svg.getAttribute("width")));
+      chartH = Math.max(chartH, Number(part.svg.getAttribute("height")));
+    }
+
+    const out = document.createElementNS(NS, "svg");
+    out.setAttribute("xmlns", NS);
+    out.setAttribute("font-family", 'system-ui, -apple-system, "Segoe UI", sans-serif');
+    let top = pad;
+    const later = [];   // built after the width is known
+    const legend = legendEl ? legendToSvg(legendEl, resolve) : null;
+    if (legend && legend.height) {
+      legend.group.setAttribute("transform", `translate(${pad} ${top})`);
+      later.push(() => out.appendChild(legend.group));
+      top += legend.height + 8;
+    }
+
+    const width = Math.max(chartW, legendEl ? legendEl.getBoundingClientRect().width : 0) + pad * 2;
+    const height = top + chartH + pad;
+    out.setAttribute("width", Math.ceil(width));
+    out.setAttribute("height", Math.ceil(height));
+    el("rect", { x: 0, y: 0, width: Math.ceil(width), height: Math.ceil(height), fill: paper }, out);
+    for (const build of later) build();
+
+    for (const part of parts) {
+      const group = el("g", { transform: `translate(${part.dx + pad} ${top})` }, out);
+      const clone = part.svg.cloneNode(true);
+      resolveVars(clone, resolve);
+      while (clone.firstChild) group.appendChild(clone.firstChild);
+    }
+    return out;
+  }
+
+  // Page title -> a safe file stem, e.g. "tabarena-leaderboard-explorer-all-tasks".
+  function slugify(text) {
+    return (text || "chart").toLowerCase().replace(/[^a-z0-9]+/g, "-").replace(/^-+|-+$/g, "").slice(0, 80);
+  }
+
+  function downloadUrl(url, filename) {
+    const a = document.createElement("a");
+    a.href = url;
+    a.download = filename;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+  }
+
+  // Rasterize the export SVG into a canvas at `scale`, then hand it to `done`.
+  function rasterize(svg, scale, done, fail) {
+    const width = Number(svg.getAttribute("width")), height = Number(svg.getAttribute("height"));
+    const img = new Image();
+    img.onload = () => {
+      const canvas = document.createElement("canvas");
+      canvas.width = Math.round(width * scale);
+      canvas.height = Math.round(height * scale);
+      const ctx = canvas.getContext("2d");
+      ctx.setTransform(scale, 0, 0, scale, 0, 0);
+      ctx.drawImage(img, 0, 0);
+      done(canvas, width, height);
+    };
+    img.onerror = fail;
+    img.src = "data:image/svg+xml;charset=utf-8,"
+      + encodeURIComponent(new XMLSerializer().serializeToString(svg));
+  }
+
+  // A one-page PDF wrapping the rendered figure, written by hand: a library would
+  // cost this page its zero-dependency, single-file property. The image is stored
+  // losslessly (raw RGB + /FlateDecode via CompressionStream) and the page is sized
+  // in points to the figure's CSS size, so it prints at the size it appears here
+  // and the pixels land at 96*scale dpi.
+  async function buildPdf(canvas, cssWidth, cssHeight) {
+    const pixels = canvas.getContext("2d").getImageData(0, 0, canvas.width, canvas.height).data;
+    const rgb = new Uint8Array((pixels.length / 4) * 3);
+    for (let i = 0, j = 0; i < pixels.length; i += 4, j += 3) {
+      rgb[j] = pixels[i];
+      rgb[j + 1] = pixels[i + 1];
+      rgb[j + 2] = pixels[i + 2];
+    }
+    const deflated = new Uint8Array(await new Response(
+      new Blob([rgb]).stream().pipeThrough(new CompressionStream("deflate"))).arrayBuffer());
+
+    const encoder = new TextEncoder();
+    const chunks = [];
+    const offsets = [];
+    let cursor = 0;
+    const put = data => {
+      const bytes = typeof data === "string" ? encoder.encode(data) : data;
+      chunks.push(bytes);
+      cursor += bytes.length;
+    };
+    const object = (id, body, stream) => {
+      offsets[id] = cursor;
+      put(`${id} 0 obj\n${body}\n`);
+      if (stream) {
+        put("stream\n");
+        put(stream);
+        put("\nendstream\n");
+      }
+      put("endobj\n");
+    };
+
+    const ptW = (cssWidth * 0.75).toFixed(2), ptH = (cssHeight * 0.75).toFixed(2);
+    const content = `q ${ptW} 0 0 ${ptH} 0 0 cm /Im0 Do Q`;
+    put("%PDF-1.4\n");
+    put(new Uint8Array([0x25, 0xe2, 0xe3, 0xcf, 0xd3, 0x0a]));   // binary marker
+    object(1, "<< /Type /Catalog /Pages 2 0 R >>");
+    object(2, "<< /Type /Pages /Kids [3 0 R] /Count 1 >>");
+    object(3, `<< /Type /Page /Parent 2 0 R /MediaBox [0 0 ${ptW} ${ptH}] `
+      + "/Resources << /XObject << /Im0 4 0 R >> >> /Contents 5 0 R >>");
+    object(4, "<< /Type /XObject /Subtype /Image "
+      + `/Width ${canvas.width} /Height ${canvas.height} /ColorSpace /DeviceRGB `
+      + `/BitsPerComponent 8 /Filter /FlateDecode /Length ${deflated.length} >>`, deflated);
+    object(5, `<< /Length ${content.length} >>`, content);
+
+    const xref = cursor;
+    let table = "xref\n0 6\n0000000000 65535 f \n";
+    for (let id = 1; id <= 5; id++) table += String(offsets[id]).padStart(10, "0") + " 00000 n \n";
+    put(table);
+    put(`trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n${xref}\n%%EOF\n`);
+    return new Blob(chunks, { type: "application/pdf" });
+  }
+
+  // Wire up the export buttons; `getParts` is called per click so the file always
+  // matches what is on screen. Returns a `run(format)` the host page can drive.
+  function setUpExport(getParts, basename) {
+    const buttons = {
+      svg: document.getElementById("btn-svg"),
+      png: document.getElementById("btn-png"),
+      pdf: document.getElementById("btn-pdf"),
+    };
+    const figure = () => buildExportSvg(getParts(), document.getElementById("legendstrip"));
+
+    // A sandboxed frame has no modals, so a failure is reported on the button.
+    function complain(format) {
+      const button = buttons[format];
+      if (!button) return;
+      const label = button.textContent;
+      button.textContent = "failed";
+      setTimeout(() => { button.textContent = label; }, 2500);
+    }
+
+    function run(format) {
+      const svg = figure();
+      const name = basename();
+      if (format === "svg") {
+        downloadUrl("data:image/svg+xml;charset=utf-8,"
+          + encodeURIComponent(new XMLSerializer().serializeToString(svg)), name + ".svg");
+        return;
+      }
+      // 3x for a screen-resolution PNG; 2x for the PDF, whose page is sized in
+      // points so the pixels already land near 200 dpi at print size.
+      rasterize(svg, format === "pdf" ? 2 : 3, (canvas, cssWidth, cssHeight) => {
+        if (format === "png") {
+          canvas.toBlob(blob => downloadUrl(URL.createObjectURL(blob), name + ".png"), "image/png");
+        } else {
+          buildPdf(canvas, cssWidth, cssHeight)
+            .then(blob => downloadUrl(URL.createObjectURL(blob), name + ".pdf"))
+            .catch(() => complain("pdf"));
+        }
+      }, () => complain(format));
+    }
+
+    for (const format of Object.keys(buttons)) {
+      if (buttons[format]) buttons[format].addEventListener("click", () => run(format));
+    }
+    // Embedded, the buttons live in the host's panel header (see main.taExport).
+    window.addEventListener("message", ev => {
+      const d = ev.data;
+      if (d && d.type === "tabarena-explorer-export" && buttons[d.format] !== undefined) run(d.format);
+    });
+  }
+
+  // When embedded, report the content height so the host page can size the
+  // iframe to fit (avoids an inner scrollbar). Works from a sandboxed frame.
+  // Measure the body (viewport-independent) — documentElement.scrollHeight is
+  // clamped to at least the iframe's current viewport, which turns the
+  // resize round-trip into a grow-forever feedback loop. The change guard
+  // stops re-posting once the height settles.
+  let lastPostedHeight = 0;
+  function postHeight() {
+    if (window.parent === window) return;
+    const height = Math.ceil(document.body.offsetHeight);
+    if (Math.abs(height - lastPostedHeight) < 3) return;
+    lastPostedHeight = height;
+    window.parent.postMessage({ type: "tabarena-explorer-height", height: height }, "*");
+  }
+
+
+  const COLUMNS = CONFIG.columns;
+  const RANK_KEY = CONFIG.rankKey;
+  const colByKey = new Map(COLUMNS.map(c => [c.key, c]));
+  const VARIANTS = CONFIG.variants.filter(v => POINTS.some(p => p.variant === v));
+
+  // How a variant is written out. The short "Tuned + Ens." form is the internal key
+  // (the charts' colour and marker tables are keyed by it), but nothing on the page
+  // should show the abbreviation: the row tag uses the spelling the data and the
+  // app's cross-subset table use, and the toggle spells it out in full.
+  const VARIANT_TEXT = {
+    "Default": "default",
+    "Tuned": "tuned",
+    "Tuned + Ens.": "tuned + ensembled",
+  };
+  const VARIANT_BTN = { "Tuned + Ens.": "Tuned + Ensembled" };
+  const variantText = v => VARIANT_TEXT[v] || v;
+  const HAS_IMPUTED = POINTS.some(p => p.imputed);
+
+  const state = {
+    methods: new Set(POINTS.map(p => p.method)),
+    variants: new Set(VARIANTS),
+    columns: new Set(COLUMNS.filter(c => c.on !== false).map(c => c.key)),
+    imputed: true,
+    search: "",
+    sortKey: null,
+    sortAsc: false,
+  };
+
+  const tblwrap = document.getElementById("tblwrap");
+  const chipsBox = document.getElementById("chips");
+  const famChips = new Map();
+  const chipByMethod = new Map();
+
+  // ---------- heat shading ----------
+  // Green (best) through olive to red (worst), matching the website's
+  // cross-subset overview so a reader moving between them reads one scale.
+  const RAMP = [[0, [28, 120, 62]], [0.5, [138, 122, 36]], [1, [160, 58, 58]]];
+  function heatColor(frac) {
+    const f = Math.max(0, Math.min(1, frac));
+    for (let i = 0; i < RAMP.length - 1; i++) {
+      const [f0, c0] = RAMP[i], [f1, c1] = RAMP[i + 1];
+      if (f <= f1) {
+        const t = f1 === f0 ? 0 : (f - f0) / (f1 - f0);
+        const mix = c0.map((v, j) => Math.round(v + t * (c1[j] - v)));
+        return "rgb(" + mix.join(",") + ")";
+      }
+    }
+    return "rgb(" + RAMP[RAMP.length - 1][1].join(",") + ")";
+  }
+  // Runtimes span orders of magnitude; shading them linearly paints every model
+  // the same green and only the slowest one red, so those normalize in log space.
+  function scaleOf(col, v) { return col.logScale ? Math.log10(v) : v; }
+  function bounds(rows, col) {
+    const values = [];
+    for (const p of rows) {
+      const v = p[col.key];
+      if (v == null || !isFinite(v)) continue;
+      if (col.logScale && !(v > 0)) continue;
+      values.push(scaleOf(col, v));
+    }
+    if (values.length < 2) return null;
+    const lo = Math.min(...values), hi = Math.max(...values);
+    return hi > lo ? [lo, hi] : null;
+  }
+
+  // ---------- cells ----------
+  function nameCell(p) {
+    const ink = FAM_INK[p.family] || "var(--muted)";
+    let inner = escapeHtml(p.method);
+    if (p.variant) inner += ' <span class="variant-tag">(' + escapeHtml(variantText(p.variant)) + ")</span>";
+    if (p.verified) inner += ' <span class="verified" title="Verified implementation">&#10004;&#65039;</span>';
+    if (p.imputed) {
+      const pct = isFinite(p.imputed_pct) ? fmtNum(p.imputed_pct, 0) + "% " : "";
+      inner += ' <span class="imp-mark" title="' + pct + 'imputed">&#8225;</span>';
+    }
+    const body = p.url
+      ? '<a href="' + escapeHtml(p.url) + '" target="_blank" rel="noopener" style="color:' + ink +
+        ';font-weight:600;">' + inner + '<span class="link-icon">&#8599;</span></a>'
+      : '<span style="color:' + ink + ';font-weight:600;">' + inner + "</span>";
+    return '<td class="name" data-export="' + escapeHtml(plainName(p)) + '">' + body + "</td>";
+  }
+  function plainName(p) {
+    return p.method + (p.variant ? " (" + p.variant + ")" : "");
+  }
+  function escapeHtml(text) {
+    return String(text == null ? "" : text).replace(/[&<>"']/g, c =>
+      ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;", "'": "&#39;" })[c]);
+  }
+
+  // ---------- rows ----------
+  function visibleRows() {
+    const term = state.search.trim().toLowerCase();
+    return POINTS.filter(p => {
+      if (!state.methods.has(p.method)) return false;
+      if (p.variant && !state.variants.has(p.variant)) return false;
+      if (!state.imputed && p.imputed) return false;
+      if (term && !(p.method + " " + (p.variant || "") + " " + p.family).toLowerCase().includes(term)) return false;
+      return true;
+    });
+  }
+  function sortRows(rows) {
+    const out = rows.slice();
+    if (!state.sortKey) return out.sort((a, b) => a.position - b.position);
+    const col = colByKey.get(state.sortKey);
+    const sign = state.sortAsc ? 1 : -1;
+    const keyOf = p => {
+      const v = p[state.sortKey];
+      if (col && col.text) return String(v == null ? "" : v).toLowerCase();
+      return v == null || !isFinite(v) ? null : v;
+    };
+    return out.sort((a, b) => {
+      const ka = keyOf(a), kb = keyOf(b);
+      if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+      return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+  }
+
+  function activeColumns() {
+    return COLUMNS.filter(c => c.always || state.columns.has(c.key));
+  }
+
+  function render() {
+    const rows = sortRows(visibleRows());
+    const cols = activeColumns();
+    if (!rows.length) {
+      tblwrap.innerHTML = '<p class="lbt-empty">Nothing selected — turn a family or variant back on.</p>';
+      document.getElementById("caption").textContent = "";
+      postHeight();
+      return;
+    }
+    // Medals mark the top three on the ranking metric; they stay on that column
+    // when the reader sorts by another, which is what they mean.
+    const ranked = rows.filter(p => isFinite(p[RANK_KEY]))
+      .sort((a, b) => (colByKey.get(RANK_KEY).lowerBetter ? a[RANK_KEY] - b[RANK_KEY] : b[RANK_KEY] - a[RANK_KEY]));
+    const medalOf = new Map(ranked.slice(0, 3).map((p, i) => [p, ["\u{1F947}", "\u{1F948}", "\u{1F949}"][i]]));
+    const heat = new Map();
+    for (const col of cols) if (col.heatmap) heat.set(col.key, bounds(rows, col));
+
+    let html = '<table class="lbt"><thead><tr>';
+    for (const col of cols) {
+      const sorted = state.sortKey === col.key
+        ? ' aria-sort="' + (state.sortAsc ? "ascending" : "descending") + '"' : "";
+      html += "<th" + sorted + ' data-key="' + col.key + '" title="' + escapeHtml(col.hint || col.label) +
+        '"><span class="lbl">' + escapeHtml(col.label) + "</span>" +
+        (col.key === "elo" ? ' <span class="ci">(95% CI)</span>' : "") + "</th>";
+    }
+    html += "</tr></thead><tbody>";
+    for (const p of rows) {
+      html += "<tr>";
+      for (const col of cols) {
+        if (col.key === "position") {
+          html += '<td class="pos num" data-sort="' + p.position + '">' + p.position + "</td>";
+        } else if (col.key === "family") {
+          const mark = FAM_VAR[p.family] || "var(--fam-baseline)";
+          html += '<td class="type" data-sort="' + escapeHtml(p.family) + '" data-export="' + escapeHtml(p.family) +
+            '"><span class="pill" style="background:color-mix(in srgb, ' + mark +
+            ' 13%, transparent);color:' + (FAM_INK[p.family] || "var(--muted)") +
+            ";border:1px solid color-mix(in srgb, " + mark + ' 40%, transparent);">' +
+            escapeHtml(p.family_symbol || "") + "</span></td>";
+        } else if (col.key === "model") {
+          html += nameCell(p);
+        } else {
+          html += valueCell(p, col, heat.get(col.key), medalOf.get(p));
+        }
+      }
+      html += "</tr>";
+    }
+    tblwrap.innerHTML = html + "</tbody></table>";
+    for (const th of tblwrap.querySelectorAll("th")) {
+      th.addEventListener("click", () => sortBy(th.dataset.key));
+    }
+    document.getElementById("caption").textContent =
+      rows.length + (rows.length === 1 ? " row" : " rows") +
+      " · click a column header to sort, hover one for what it means" +
+      (heat.size ? " · green is better, red is worse, per column" : "") +
+      (HAS_IMPUTED ? " · ‡ marks a partly imputed score" : "");
+    postHeight();
+  }
+
+  function valueCell(p, col, span, medal) {
+    const v = p[col.key];
+    if (v == null || (typeof v === "number" && !isFinite(v))) return '<td class="na">&ndash;</td>';
+    if (col.text) {
+      return '<td data-sort="' + escapeHtml(v) + '" data-export="' + escapeHtml(v) + '">' + escapeHtml(v) + "</td>";
+    }
+    let style = "";
+    if (span) {
+      let frac = (scaleOf(col, v) - span[0]) / (span[1] - span[0]);
+      if (!col.lowerBetter) frac = 1 - frac;
+      style = ' style="background:' + heatColor(frac) + ';color:#f7f7f7;"';
+    }
+    let text = fmtNum(v, col.decimals) + (col.suffix || "");
+    let ci = "";
+    if (col.key === "elo" && p.elo_ci) {
+      ci = ' <span class="ci">(' + escapeHtml(p.elo_ci) + ")</span>";
+    }
+    const inner = medal !== undefined || col.key === RANK_KEY
+      ? '<span class="cell"><span class="medal">' + (medal || "") + '</span><span>' + text + ci + "</span></span>"
+      : text + ci;
+    return '<td class="num" data-sort="' + v + '"' + (ci ? ' data-ci="' + escapeHtml(p.elo_ci) + '"' : "") +
+      style + ">" + inner + "</td>";
+  }
+
+  function sortBy(key) {
+    if (!key) return;
+    if (state.sortKey === key) state.sortAsc = !state.sortAsc;
+    else { state.sortKey = key; state.sortAsc = !!(colByKey.get(key) || {}).lowerBetter; }
+    render();
+  }
+
+  // ---------- chips (families and their methods) ----------
+  function familyMembers(fam) {
+    return [...new Set(POINTS.filter(p => p.family === fam).map(p => p.method))];
+  }
+  function methodRank(name) {
+    const col = colByKey.get(RANK_KEY);
+    const values = POINTS.filter(p => p.method === name && isFinite(p[RANK_KEY])).map(p => p[RANK_KEY]);
+    if (!values.length) return Infinity;
+    return col.lowerBetter ? Math.min(...values) : -Math.max(...values);
+  }
+  function buildChips() {
+    const head = document.createElement("div");
+    head.className = "chips-head";
+    head.textContent = "Models shown — click to remove, click a family to toggle the whole group";
+    chipsBox.appendChild(head);
+    for (const fam of FAM_ORDER) {
+      const members = familyMembers(fam);
+      if (!members.length) continue;
+      members.sort((a, b) => methodRank(a) - methodRank(b));
+      const row = document.createElement("div");
+      row.className = "chiprow";
+      const famBtn = document.createElement("button");
+      famBtn.className = "famchip";
+      famBtn.style.setProperty("--fam", FAM_VAR[fam]);
+      famBtn.innerHTML = famChipLabel(fam, members.length);
+      famBtn.title = "Toggle all " + members.length + " " + fam + " models";
+      famBtn.addEventListener("click", () => toggleFamily(fam));
+      row.appendChild(famBtn);
+      famChips.set(fam, famBtn);
+      const set = document.createElement("div");
+      set.className = "chipset";
+      for (const name of members) {
+        const imputed = POINTS.some(p => p.method === name && p.imputed);
+        const b = document.createElement("button");
+        b.className = "chip";
+        b.style.setProperty("--fam", FAM_VAR[fam]);
+        b.innerHTML = '<span class="dot"></span><span>' + escapeHtml(name) + "</span>" +
+          (imputed ? '<span class="imp-mark">&#8225;</span>' : "");
+        b.title = name + (imputed ? " — partially imputed" : "");
+        b.addEventListener("click", () => toggleMethod(name));
+        set.appendChild(b);
+        chipByMethod.set(name, b);
+      }
+      row.appendChild(set);
+      chipsBox.appendChild(row);
+    }
+  }
+  function syncChips() {
+    for (const [name, b] of chipByMethod) b.setAttribute("aria-pressed", String(state.methods.has(name)));
+    for (const [fam, b] of famChips) {
+      b.setAttribute("aria-pressed", String(familyMembers(fam).every(m => state.methods.has(m))));
+    }
+  }
+  function toggleMethod(name) {
+    if (state.methods.has(name)) state.methods.delete(name); else state.methods.add(name);
+    syncChips();
+    render();
+  }
+  function toggleFamily(fam) {
+    const members = familyMembers(fam);
+    const allOn = members.every(m => state.methods.has(m));
+    for (const m of members) { if (allOn) state.methods.delete(m); else state.methods.add(m); }
+    syncChips();
+    render();
+  }
+
+  // ---------- variant / imputed / column toggles ----------
+  const variantBtns = new Map();
+  function buildVariantBtns() {
+    const box = document.getElementById("variant-btns");
+    for (const v of VARIANTS) {
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.innerHTML = '<span class="swatch"></span>' + escapeHtml(VARIANT_BTN[v] || v);
+      b.style.setProperty("--fam", VARIANT_VAR[v] || "var(--accent)");
+      b.title = "Show or hide the " + variantText(v) + " results";
+      b.addEventListener("click", () => {
+        if (state.variants.has(v)) state.variants.delete(v); else state.variants.add(v);
+        syncVariantBtns();
+        render();
+      });
+      box.appendChild(b);
+      variantBtns.set(v, b);
+    }
+  }
+  function syncVariantBtns() {
+    for (const [v, b] of variantBtns) b.setAttribute("aria-pressed", String(state.variants.has(v)));
+    const imp = document.getElementById("btn-imputed");
+    imp.hidden = !HAS_IMPUTED;
+    imp.className = "btn toggle";
+    imp.innerHTML = '<span class="swatch"></span>&#8225; Imputed';
+    imp.setAttribute("aria-pressed", String(state.imputed));
+    imp.style.setProperty("--fam", "var(--muted)");
+  }
+  function buildColumnBtns() {
+    const box = document.getElementById("col-btns");
+    for (const col of COLUMNS) {
+      if (col.always) continue;
+      const b = document.createElement("button");
+      b.className = "btn toggle";
+      b.textContent = col.short || col.label;
+      b.title = "Show or hide " + col.label;
+      b.style.setProperty("--fam", "var(--accent)");
+      b.addEventListener("click", () => {
+        if (state.columns.has(col.key)) state.columns.delete(col.key); else state.columns.add(col.key);
+        b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+        render();
+      });
+      b.setAttribute("aria-pressed", String(state.columns.has(col.key)));
+      box.appendChild(b);
+    }
+  }
+
+  // ---------- CSV ----------
+  // Exports what the reader is looking at: the rows the filters left, the
+  // columns they picked, in the order they sorted.
+  function downloadCsv() {
+    const cols = activeColumns();
+    const rows = sortRows(visibleRows());
+    const quote = v => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const head = [];
+    for (const col of cols) {
+      head.push(col.label);
+      if (col.key === "elo" && rows.some(p => p.elo_ci)) head.push("Elo 95% CI");
+    }
+    const lines = [head.map(quote).join(",")];
+    for (const p of rows) {
+      const values = [];
+      for (const col of cols) {
+        if (col.key === "model") values.push(plainName(p));
+        else values.push(p[col.key]);
+        if (col.key === "elo" && rows.some(q => q.elo_ci)) values.push(p.elo_ci || "");
+      }
+      lines.push(values.map(quote).join(","));
+    }
+    const blob = new Blob([lines.join("\n") + "\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    downloadUrl(url, slugify(CONFIG.title || document.title || "leaderboard") + ".csv");
+    URL.revokeObjectURL(url);
+  }
+
+  // ---------- boot ----------
+  if (CONFIG.title) document.getElementById("title").textContent = CONFIG.title;
+  document.getElementById("search").addEventListener("input", ev => {
+    state.search = ev.target.value;
+    render();
+  });
+  document.getElementById("btn-imputed").addEventListener("click", () => {
+    state.imputed = !state.imputed;
+    syncVariantBtns();
+    render();
+  });
+  document.getElementById("btn-csv").addEventListener("click", downloadCsv);
+  // Embedded, the host page puts its own CSV button in the panel header (next to
+  // the title) and asks for the download over postMessage, the same way it drives
+  // the figure exports. Standalone, this page needs its own button.
+  if (window.parent !== window) {
+    document.getElementById("btn-csv").hidden = true;
+    window.addEventListener("message", ev => {
+      if (ev.data && ev.data.type === "tabarena-leaderboard-csv") downloadCsv();
+    });
+  }
+  // Opens with the controls: the filters are the point of this page, not chrome
+  // around a figure.
+  setUpPaperView(render, { openInPaper: false });
+  buildVariantBtns();
+  buildColumnBtns();
+  buildChips();
+  syncChips();
+  syncVariantBtns();
+  render();
+  window.addEventListener("resize", postHeight);
+})();
+</script>
+</body>
+</html>
diff --git a/data_loading.py b/data_loading.py
index b2ad63fa14a97a1a2e6c613190ec53dd2a23c1d1..d52ff06c1a6bed6981d2c9246e164e4816d70be3 100644
--- a/data_loading.py
+++ b/data_loading.py
@@ -99,6 +99,25 @@ def load_leaderboard_csv(path: str) -> pd.DataFrame:
     return df.rename(columns={"1#": "#"})
 
 
+VARIANT_RE = re.compile(r"\((default|tuned \+ ensembled|tuned)\)")
+
+
+def parse_model(model: str) -> tuple[str, str, str | None]:
+    """Split a Model cell into (base name, variant, url).
+
+    A cell looks like ``[TabFM (default)](https://…)``, optionally followed by an
+    ``[X% IMPUTED]`` tag. Used both for display (``views.py``) and for the JSON
+    records the API returns (``api.py``), so the two cannot disagree.
+    """
+    link = re.match(r"\[(.*?)\]\((.*?)\)", model)
+    text, url = (link.group(1), link.group(2)) if link else (model, None)
+    text = text.split("[")[0].strip()  # drop any [X% IMPUTED] tag
+    variant_match = VARIANT_RE.search(text)
+    variant = variant_match.group(1) if variant_match else ""
+    base = VARIANT_RE.sub("", text).strip()
+    return base, variant, url
+
+
 def unzip_png(base_dir: Path, img_name: str) -> str:
     """Return the path to ``base_dir/img_name``.png, unzipping the ``.png.zip`` on first access."""
     base = Path(base_dir) / img_name
diff --git a/main.py b/main.py
index 40b82df318f3c86f0c3fae8f2e72c611e6edd969..ef9c89daaea94a0f71bae17c3b5b059c9ea8d79d 100644
--- a/main.py
+++ b/main.py
@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import gradio as gr
 
+from api import register_api
 from pages import PAGES, render_page
 from website_texts import COI_HTML, TITLE
 
@@ -9,6 +10,72 @@ CSS = """
 /* Use the full page width instead of gradio's narrow centered column. */
 .gradio-container { max-width: 100% !important; }
 
+/* --- Colour emoji ----------------------------------------------------------
+   Gradio's own font stack ends "… "Noto Sans", sans-serif, "Apple Color Emoji",
+   "Segoe UI Emoji", "Segoe UI Symbol", "Noto Color Emoji"" — the colour emoji
+   fonts sit *after* the generic `sans-serif`, which matches everything, so the
+   browser never reaches them and resolves emoji through its own fallback
+   instead. On Linux that commonly lands on the monochrome Noto Emoji, which is
+   why 🧠⚡ / 🤖 / 🥇 render flat there and in colour on macOS and Windows.
+   Same stack in Gradio 5, so this is long-standing rather than new.
+   Naming the colour fonts before the generic fixes it for every emoji on the
+   page; Latin glyphs are unaffected because those fonts carry none.
+
+   Overriding the two variables rather than `font-family` directly is what makes
+   it stick for Gradio's components: its own rules set `font-family: var(--font)` /
+   `var(--font-sans)` on individual elements, which would beat a container-level
+   declaration. Raw HTML (`gr.HTML`) has no such rule and is handled by the
+   element-level block below. */
+:root, .gradio-container, html body {
+    --font: "IBM Plex Sans", ui-sans-serif, system-ui, -apple-system,
+        "Segoe UI", Roboto, "Helvetica Neue", Arial, "Noto Sans",
+        "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji", sans-serif;
+    --font-sans: var(--font);
+    font-family: var(--font);
+}
+
+/* Overriding the variables above fixes Gradio's own components (buttons, tabs),
+   which resolve `font-family: var(--font)` per element. It does not reach content
+   inside `gr.HTML`, which has no per-element rule of its own — so the elements we
+   write by hand are listed here and get the colour fonts forced.
+
+   The emoji fonts lead and `!important` settles the cascade; Latin text in these
+   elements is unaffected because emoji fonts carry no Latin glyphs, so it falls
+   through to IBM Plex Sans exactly as before. Add a selector here if an emoji
+   shows up flat somewhere new. */
+.ta-card-ico,
+.ta-path-ico,
+.ta-pill,
+.ta-lb-title,
+.ta-figbar-title,
+.ta-section-head h2,
+.ta-jump,
+.ta-viewbtn,
+.coi-badge,
+.ta-legend,
+.ta-lbtable td.ta-type-cell,
+.ta-verified,
+.ta-link-icon,
+.ta-cap,
+.coi-cta,
+.markdown-text h1, .markdown-text h2, .markdown-text h3, .markdown-text h4 {
+    font-family: "Apple Color Emoji", "Segoe UI Emoji", "Noto Color Emoji",
+        "Twemoji Mozilla", "IBM Plex Sans", ui-sans-serif, system-ui,
+        "Segoe UI", Roboto, sans-serif !important;
+}
+
+/* Gradio 6 pads every anchor inside a gr.HTML / gr.Markdown block:
+
+       .gradio-container-6-22-0 .gradio-style a { padding: 2px 8px; text-align: center }
+
+   That is a link-button style, and it lands on inline links too, which is where the
+   gaps around "(see paper)" / "(see code)" in the hero cards came from. Reset for
+   the links we write as running text — they are the ones with no class of their own
+   — at a specificity that beats Gradio's two-class selector whatever the stylesheet
+   order turns out to be. The pill-shaped links keep their own padding (below, where
+   the same Gradio rule is why they have to say `!important`). */
+.gradio-container .gradio-style a:not([class]) { padding: 0; }
+
 .markdown-text-box {
     padding: 4px;
     border-radius: 2px;
@@ -29,34 +96,108 @@ CSS = """
     color: #e6ebff;
 }
 
+/* --- Subset filter rows ----------------------------------------------------
+   The two content axes (task, dataset size) and the two view modifiers are one
+   control cluster, so they are drawn as three rows of the same panel: same frame,
+   same height, and a small uppercase label on the left saying which axis the row
+   switches. Before this the tab bars were bare text above a bordered toggle box,
+   which read as two unrelated controls.
+
+   Gradio 6 renders a Tabs as .tabs > .tab-wrapper > .tab-container > button and
+   marks the active tab .selected, styling it with a 2px underline. Here the row
+   frame lives on .tab-wrapper and the choices are flat chips with the active one
+   filled. Note .tab-wrapper holds a *second*, visually hidden .tab-container that
+   Gradio measures to decide which tabs go into its overflow menu, which is why the
+   chip rules deliberately hit every .tab-container: styling only the visible copy
+   would make it mis-measure. */
 .tab-buttons {
-    margin-top: -4px !important;
-    margin-bottom: -14px !important;
+    gap: 0 !important;
+    /* Cancel the column gap the surrounding Blocks layout adds between the rows,
+       then set the row spacing here. */
+    margin: 0 0 calc(5px - var(--layout-gap)) !important;
 }
-
-/* Top-level page tabs: size each header to its label instead of stretching one full width. */
-#top-tabs > .tab-nav { justify-content: flex-start !important; }
-#top-tabs > .tab-nav > button { flex: 0 0 auto !important; width: auto !important; }
-
-/* View toggles (imputation / Lite) below the subset tabs, set apart in a panel. */
-.view-toggles {
-    margin: 2px 0 10px 0;
-    padding: 6px 14px;
-    gap: 18px;
+.tab-buttons .tab-wrapper {
+    height: auto;
     align-items: center;
+    margin-bottom: 0;
+    padding: 5px 12px 5px 14px;
     background: #ffffff0a;
     border: 1px solid #ffffff26;
+    border-radius: 10px;
+}
+/* The rule Gradio draws under a tab bar; the row frame replaces it. */
+.tab-buttons .tab-container::after { display: none; }
+.tab-buttons .tab-container { gap: 3px; height: auto; }
+.tab-buttons .tab-container > button {
+    /* One neutral slate, and how strongly a chip is tinted says how deep in the axis
+       it sits: the row's "all" chip is the whole set, the next level is a subset of
+       it, and the lightest chips are subsets of a subset. Peers share a step, so
+       nothing in the row reads as ranked above anything else — Small and Medium are
+       the same tint, and so are Binary and Multiclass. Selection is the one place a
+       colour changes hue, and it is the app accent, the same as everywhere else. */
+    --depth: 0.62;
+    height: auto;
+    padding: 5px 12px;
+    font-size: 0.86em;
+    font-weight: 600;
+    color: #cbd5ef;
+    background: rgba(158, 176, 208, calc(0.3 * var(--depth)));
+    border: none;
     border-radius: 8px;
+    transition: background .14s ease, color .14s ease;
 }
+.tab-buttons .tab-container > button:hover:not(.selected) {
+    background: rgba(158, 176, 208, calc(0.62 * var(--depth)));
+    color: #ffffff;
+}
+.tab-buttons .tab-container > button.selected {
+    color: #ffffff;
+    background: rgba(110, 140, 245, 0.92);
+    box-shadow: 0 1px 3px #0000005c;
+}
+
+/* The depth steps. Gradio gives the tabs no per-tab class, so they are addressed by
+   position — the nth-child order is the insertion order of TASK_LABELS /
+   DATASET_LABELS: all, classification, regression, binary, multiclass and all,
+   small, medium. Chips 4 and 5 of the task row are the two classification
+   flavours, one level further in than Classification itself. */
+.axis-tasks .tab-container > button:nth-child(1),
+.axis-datasets .tab-container > button:nth-child(1) { --depth: 1; }
+.axis-tasks .tab-container > button:nth-child(n+4) { --depth: 0.34; }
+
+/* The filled chip is the selection cue; Gradio's underline would sit inside it. */
+.tab-buttons .tab-container > button.selected::after { display: none; }
+
+/* The row labels: one style, one width, so the three rows line up on the left. */
+.axis-tasks .tab-wrapper::before,
+.axis-datasets .tab-wrapper::before,
 .view-toggles::before {
-    content: "View:";
-    font-size: 0.85em;
-    font-weight: 600;
-    opacity: 0.7;
+    flex: 0 0 auto;
+    min-width: 74px;
+    font-size: 0.68em;
+    font-weight: 700;
+    letter-spacing: 0.09em;
+    text-transform: uppercase;
+    opacity: 0.5;
+    white-space: nowrap;
     align-self: center;
 }
+.axis-tasks .tab-wrapper::before { content: "Task"; }
+.axis-datasets .tab-wrapper::before { content: "Datasets"; }
+
+/* Third row of the same panel: the view modifiers (imputation / Lite). */
+.view-toggles {
+    margin: 0 0 16px 0;
+    padding: 7px 12px 7px 14px;
+    gap: 14px;
+    align-items: center;
+    background: #ffffff0a;
+    border: 1px solid #ffffff26;
+    border-radius: 10px;
+}
+.view-toggles::before { content: "View"; }
 .view-toggles label span {
-    font-size: 0.85em !important;
+    font-size: 0.86em !important;
 }
 
 /* --- Figure panel with an interactive / static view switch -----------------
@@ -150,7 +291,7 @@ CSS = """
 }
 .ta-link-btn {
     display: inline-block;
-    padding: 10px 20px;
+    padding: 10px 20px !important;
     border-radius: 8px;
     font-weight: 600;
     text-decoration: none;
@@ -197,7 +338,7 @@ CSS = """
     display: inline-block;
     margin: 0;
     font-size: 0.9em;
-    padding: 4px 14px;
+    padding: 4px 14px !important;
     border: 1px solid #ffffff2e;
     border-radius: 999px;
     text-decoration: none;
@@ -251,45 +392,202 @@ CSS = """
     background: #ffffff08;
 }
 
-/* Column key above the full table: a boxed panel with a title + hint, a chip row below,
-   and one shared reveal panel (CSS-radio toggled, one definition open at a time). */
-.ta-metric-key {
-    margin: 10px 0 12px 0;
-    padding: 11px 14px 12px;
-    background: #ffffff08;
-    border: 1px solid #ffffff1f;
-    border-radius: 10px;
+/* --- The full leaderboard table, as one card -------------------------------
+   Title strip, then the filters, then the column key, then the table itself.
+   The controls and the key used to sit outside the table (the key in its own
+   floating panel, the filters inside the old third-party widget); keeping them
+   in the same bordered card is what makes them read as belonging to it. */
+.ta-lb {
+    margin: 16px 0 10px 0;
+    padding: 0;
+    overflow: visible;
+    border: 1px solid #ffffff33;
+    border-left: 3px solid rgba(110, 140, 245, 0.55);
+    border-radius: 12px;
+    background: #ffffff10;
+    gap: 0 !important;
 }
-.ta-mk-head { display: flex; align-items: baseline; flex-wrap: wrap; gap: 3px 10px; margin-bottom: 10px; }
-.ta-mk-title { font-size: 1.05em; font-weight: 700; }
-.ta-mk-hint { font-size: 0.85em; opacity: 0.6; }
-.ta-mk-radio { display: none; }
-.ta-mk-chips { display: flex; flex-wrap: wrap; gap: 7px; }
-.ta-mk-chip {
-    display: inline-block;
+.ta-lb-bar {
+    display: flex;
+    align-items: baseline;
+    flex-wrap: wrap;
+    gap: 3px 12px;
+    padding: 11px 16px;
+    background: #ffffff21;
+    border-bottom: 1px solid #ffffff38;
+    border-radius: 9px 9px 0 0;
+}
+.ta-lb-title { font-weight: 650; font-size: 1.25em; letter-spacing: 0.01em; }
+.ta-lb-sub { font-size: 0.85em; opacity: 0.6; }
+/* One padded band for the filters, so four controls fit on a line without
+   looking cramped. Nothing here may set pointer-events, position or z-index:
+   these are live Gradio inputs and must stay clickable. */
+.ta-lb-controls {
+    padding: 12px 16px 2px 16px;
+    gap: 18px;
+    align-items: flex-start;
+}
+.ta-lb .ta-lb-imputed { padding: 0 16px 6px 16px; }
+.ta-lb .ta-legend { padding: 4px 16px 0 16px; }
+.ta-lb .ta-scroll { margin: 4px 16px 0 16px; max-height: 720px; }
+.ta-lb .ta-cap { padding: 0 16px 10px 16px; }
+
+/* --- Chip selectors, matching the plot explorers' edit view ------------------
+   The same shape, dot and pressed treatment as the method chips in the generated
+   *_explorer.html: a pill with a family-colored dot, no fill until selected, then
+   a tinted background and a matching border. Variables are the explorers' own
+   dark-theme values (the site forces dark).
+
+   These are real Gradio inputs underneath. The native box is visually hidden but
+   left in the layout and in the label, so a click on the chip still reaches it;
+   nothing here sets pointer-events, position or z-index on anything clickable.
+   Both `:has(input:checked)` and Gradio's own `.selected` drive the pressed look,
+   so a change to either survives. */
+.ta-chips, .ta-btns, .ta-famchip {
+    --line: #2e2e33;
+    --muted: #9b9a92;
+    --pt-muted: #55555c;
+    --chip-bg: #232327;
+    --fam: #9b9a92;
+}
+.ta-btns-imputed { --fam: #e6c14d; }
+.ta-chips label {
+    display: inline-flex !important;
+    align-items: center;
+    gap: 6px;
+    /* Spacing lives on the chips themselves rather than as a gap on Gradio's
+       internal container, so this does not depend on Gradio's DOM. */
+    margin: 0 5px 5px 0 !important;
+    padding: 5px 11px 5px 9px !important;
+    font: 500 12.5px/1 system-ui, sans-serif !important;
+    color: #f0efea;
+    background: none !important;
+    border: 1px solid var(--line) !important;
+    border-radius: 999px !important;
+    box-shadow: none !important;
+    cursor: pointer;
+    transition: border-color 120ms ease, background-color 120ms ease;
+}
+.ta-chips label::before {
+    content: "";
+    width: 8px;
+    height: 8px;
+    border-radius: 50%;
+    background: var(--pt-muted);
+    flex: none;
+}
+.ta-chips label:hover { border-color: var(--muted) !important; }
+.ta-chips label:has(input:checked),
+.ta-chips label.selected {
+    border-color: var(--fam) !important;
+    background: color-mix(in srgb, var(--fam) 13%, transparent) !important;
+    font-weight: 650 !important;
+}
+.ta-chips label:has(input:checked)::before,
+.ta-chips label.selected::before { background: var(--fam); }
+/* Visually hidden, still hit-testable through its label. */
+.ta-chips label input[type="checkbox"],
+.ta-chips label input[type="radio"] {
+    width: 0 !important;
+    height: 0 !important;
+    margin: 0 !important;
+    padding: 0 !important;
+    border: 0 !important;
+    opacity: 0;
+    flex: none;
+}
+/* The chip rows carry their own label above; keep Gradio's block label small. */
+.ta-chips > .block-label, .ta-chips span[data-testid="block-info"] { font-size: 0.85em; }
+
+/* The family chip that heads each row of model chips: the explorers' .famchip —
+   uppercase, dashed until selected, then solid in the family's colour. */
+.ta-famchip label {
+    display: inline-flex !important;
+    align-items: center;
+    gap: 6px;
+    margin: 0 !important;
+    padding: 5px 11px !important;
+    font: 650 10.5px/1.3 system-ui, sans-serif !important;
+    letter-spacing: 0.06em;
+    text-transform: uppercase;
+    color: var(--muted);
+    background: var(--chip-bg) !important;
+    border: 1px dashed var(--line) !important;
+    border-radius: 999px !important;
+    box-shadow: none !important;
     cursor: pointer;
-    font-size: 0.82em;
-    padding: 3px 11px;
-    border-radius: 999px;
-    border: 1px solid #ffffff24;
-    background: #ffffff0d;
     white-space: nowrap;
 }
-.ta-mk-chip:hover { border-color: #ffffff5c; background: #ffffff1a; }
-.ta-mk-def {
-    display: none;
-    margin-top: 8px;
-    max-width: 760px;
-    padding: 9px 13px;
-    border-radius: 10px;
-    background: #1d1d26;
-    border: 1px solid #ffffff33;
-    font-size: 0.82em;
-    line-height: 1.4;
-    color: #dcdce6;
-}
-.ta-mk-close { float: right; cursor: pointer; opacity: 0.55; margin-left: 12px; font-weight: 700; }
-.ta-mk-close:hover { opacity: 1; }
+.ta-famchip label::before {
+    content: "";
+    width: 8px;
+    height: 8px;
+    border-radius: 50%;
+    background: var(--fam);
+    flex: none;
+}
+.ta-famchip label:hover { border-color: var(--fam) !important; color: #f0efea; }
+.ta-famchip label:has(input:checked),
+.ta-famchip label.selected {
+    border: 1px solid var(--fam) !important;
+    background: color-mix(in srgb, var(--fam) 13%, transparent) !important;
+    color: #f0efea;
+}
+.ta-famchip label input { width: 0 !important; height: 0 !important; margin: 0 !important; opacity: 0; }
+/* Family chip left, its models filling the rest of the row. */
+.ta-chiprow { gap: 10px; align-items: flex-start; margin: 0 !important; padding: 0 16px; }
+
+/* --- Button-style toggles, as in the Leaderboard Overview explorer -----------
+   That explorer uses buttons rather than chips for the variants ("since they
+   filter the series rather than the rows"), each carrying its variant colour from
+   --var-*; see constants.variant_color. Same shape here. */
+.ta-btns label {
+    display: inline-flex !important;
+    align-items: center;
+    margin: 0 5px 5px 0 !important;
+    padding: 6px 11px !important;
+    font: 600 12.5px/1 system-ui, sans-serif !important;
+    color: #f0efea;
+    background: var(--chip-bg) !important;
+    border: 1px solid var(--line) !important;
+    border-radius: 7px !important;
+    box-shadow: none !important;
+    cursor: pointer;
+    white-space: nowrap;
+    transition: border-color 120ms ease, background-color 120ms ease;
+}
+.ta-btns label:hover { border-color: var(--muted) !important; }
+.ta-btns label:has(input:checked),
+.ta-btns label.selected {
+    border-color: var(--fam) !important;
+    background: color-mix(in srgb, var(--fam) 20%, #232327) !important;
+}
+.ta-btns label input { width: 0 !important; height: 0 !important; margin: 0 !important; opacity: 0; }
+
+/* The CSV export button sits at the right end of the card's title strip. */
+.ta-csvbtn { margin-left: auto !important; }
+
+/* Sortable headers. The overview's info headers use a help cursor; here the same
+   header is also the sort control, so pointer wins and the dotted underline is
+   what still signals "hover me for a definition". */
+.ta-lbtable thead th.ta-th-sort { cursor: pointer; user-select: none; white-space: nowrap; }
+.ta-lbtable thead th.ta-th-sort:hover { background: #262632; }
+.ta-lbtable thead th.ta-th-sort::after {
+    content: "↕";
+    font-size: 0.72em;
+    opacity: 0.28;
+    margin-left: 5px;
+}
+.ta-lbtable thead th[aria-sort="ascending"]::after { content: "▲"; opacity: 0.85; }
+.ta-lbtable thead th[aria-sort="descending"]::after { content: "▼"; opacity: 0.85; }
+.ta-lbtable th.ta-th-left { text-align: left; }
+/* Two-emoji type symbols (🧠⚡, 🧠🔁) must never break across lines, in the table
+   cell or in the legend chip that explains it. */
+.ta-pill { white-space: nowrap; }
+.ta-lbtable td.ta-type-cell { white-space: nowrap; width: 1%; }
+/* The Elo interval rides along with the value but stays visibly secondary. */
+.ta-ci { opacity: 0.5; font-weight: 400; font-size: 0.82em; }
+.ta-ci-head { opacity: 0.55; font-weight: 400; font-size: 0.82em; }
 
 /* Hero stat cards shown above the info boxes (emoji on the left of the text, compact). */
 .ta-hero {
@@ -444,7 +742,7 @@ CSS = """
     width: fit-content;
     margin: 28px auto 0;
     text-align: center;
-    padding: 12px 24px;
+    padding: 12px 24px !important;
     border-radius: 8px;
     font-weight: 600;
     font-size: 1.02em;
@@ -559,6 +857,108 @@ window.taPaperView = function (uid) {
     if (exportGroup) exportGroup.hidden = !on;
 };
 
+/* --- Leaderboard table sorting ---------------------------------------------
+   Click a header to sort the rows beneath it; click again to reverse. Numeric
+   columns are marked `data-type="num"` and every cell carries the raw value in
+   `data-sort`, so the Elo column can print its confidence interval beside the
+   number without that text taking part in the comparison. Missing values ("–")
+   always sink to the bottom, whichever direction is active. Sorting is local to
+   the browser: the server sends the rows in published order and re-sending them
+   (when a filter changes) resets the sort. See views.leaderboard_table_html. */
+window.taSortTable = function (th) {
+    const table = th.closest("table");
+    const body = table.tBodies[0];
+    if (!body) return;
+    const index = Array.prototype.indexOf.call(th.parentNode.children, th);
+    const numeric = th.dataset.type === "num";
+    const direction = th.getAttribute("aria-sort") === "ascending" ? "descending" : "ascending";
+    table.querySelectorAll("th[aria-sort]").forEach((other) => other.removeAttribute("aria-sort"));
+    th.setAttribute("aria-sort", direction);
+    const sign = direction === "ascending" ? 1 : -1;
+    const keyOf = (row) => {
+        const cell = row.children[index];
+        if (!cell) return null;
+        const raw = cell.dataset.sort !== undefined ? cell.dataset.sort : cell.textContent.trim();
+        if (raw === "" || raw === "–") return null;
+        if (!numeric) return raw.toLowerCase();
+        const value = parseFloat(raw);
+        return isNaN(value) ? null : value;
+    };
+    const rows = Array.prototype.slice.call(body.rows);
+    rows.sort((a, b) => {
+        const ka = keyOf(a), kb = keyOf(b);
+        if (ka === null || kb === null) return ka === kb ? 0 : ka === null ? 1 : -1;
+        return ka < kb ? -sign : ka > kb ? sign : 0;
+    });
+    const fragment = document.createDocumentFragment();
+    rows.forEach((row) => fragment.appendChild(row));
+    body.appendChild(fragment);
+};
+
+/* Delegated so it keeps working after the server replaces the table's HTML. */
+document.addEventListener("click", (ev) => {
+    const th = ev.target.closest && ev.target.closest("table.ta-lbtable th.ta-th-sort");
+    if (th) window.taSortTable(th);
+});
+
+/* --- Leaderboard CSV export -------------------------------------------------
+   Exports what the reader is actually looking at: the rows the filters left, the
+   columns they picked, in the order they sorted. Reading the rendered table is
+   what buys that last part — the server does not know the browser's sort. Values
+   come from `data-sort` where a cell has one (the raw unrounded number, and the
+   family name rather than its emoji), and the Elo cell's interval is split back
+   out into its own column. */
+window.taExportTable = function (tableId) {
+    const table = document.getElementById(tableId);
+    if (!table) return;
+    const quote = (v) => '"' + String(v == null ? "" : v).replace(/"/g, '""') + '"';
+    const headers = [];
+    table.querySelectorAll("thead th").forEach((th) => {
+        const label = th.cloneNode(true);
+        label.querySelectorAll(".ta-ci-head").forEach((n) => n.remove());
+        headers.push(label.textContent.trim());
+        if (th.dataset.col === "elo") headers.push("Elo 95% CI");
+    });
+    const lines = [headers.map(quote).join(",")];
+    table.tBodies[0] && Array.prototype.forEach.call(table.tBodies[0].rows, (row) => {
+        if (row.hidden) return;
+        const values = [];
+        Array.prototype.forEach.call(row.cells, (cell) => {
+            if (cell.dataset.ci !== undefined) {
+                values.push(cell.dataset.sort);
+                values.push(cell.dataset.ci);
+                return;
+            }
+            values.push(cell.dataset.sort !== undefined && cell.dataset.export !== "text"
+                ? cell.dataset.sort
+                : cell.textContent.trim());
+        });
+        lines.push(values.map(quote).join(","));
+    });
+    /* Escaped twice: this file's HEAD is a plain Python string, so a lone \n here
+       would reach the browser as a real newline inside a JS string literal and
+       break the whole script (as it once did). */
+    const blob = new Blob([lines.join("\\n") + "\\n"], { type: "text/csv;charset=utf-8" });
+    const url = URL.createObjectURL(blob);
+    const link = document.createElement("a");
+    link.href = url;
+    link.download = (table.dataset.name || tableId) + ".csv";
+    document.body.appendChild(link);
+    link.click();
+    link.remove();
+    URL.revokeObjectURL(url);
+};
+
+/* The leaderboard table's CSV export. The button lives in the panel header, in
+   the same row as the title, and the download itself happens inside the frame —
+   only it knows the current filters, columns and sort order. Same channel as the
+   figure exports below. */
+window.taLeaderboardCsv = function (button) {
+    const card = button.closest(".ta-lb");
+    const frame = card && card.querySelector("iframe.ta-explorer");
+    if (frame) frame.contentWindow.postMessage({ type: "tabarena-leaderboard-csv" }, "*");
+};
+
 /* Figure download, handed to the explorer over the same channel. */
 window.taExport = function (uid, format) {
     const frame = document.querySelector("#" + uid + "-i iframe.ta-explorer");
@@ -584,9 +984,10 @@ window.addEventListener("message", (ev) => {
 
 
 def main() -> None:
-    with gr.Blocks(
-        css=CSS, js=FORCE_DARK_THEME_JS, head=HEAD, title="TabArena", fill_width=True
-    ) as website:
+    # Gradio 6 moved css / js / head from the Blocks constructor to launch(); passing
+    # them here would be ignored with a warning, silently dropping every style on the
+    # page. They are handed to launch() at the bottom of this function.
+    with gr.Blocks(title="TabArena", fill_width=True) as website:
         gr.HTML(TITLE)
         # Conflict-of-interest corner hint (🔍 pill, top-right) + CSS-only popup.
         gr.HTML(COI_HTML)
@@ -597,7 +998,21 @@ def main() -> None:
                 with gr.TabItem(page.name):
                     render_page(page)
 
-    website.launch(show_error=True, ssr_mode=False, debug=True)
+        # The agent-facing endpoints (no UI). These are what HF's generated
+        # agents.md points a coding agent at; see api.py.
+        register_api()
+
+    website.launch(
+        css=CSS,
+        js=FORCE_DARK_THEME_JS,
+        head=HEAD,
+        show_error=True,
+        ssr_mode=False,
+        debug=True,
+        # Serves the api.py endpoints as MCP tools at /gradio_api/mcp/, so an MCP
+        # client can attach to the Space itself instead of driving the REST API.
+        mcp_server=True,
+    )
 
 
 if __name__ == "__main__":
diff --git a/pages.py b/pages.py
index 52452f9b4e6f27d678a77ae0b60ab802551285a9..4b5402cf2a8c550973f108c957ae92f4147007d2 100644
--- a/pages.py
+++ b/pages.py
@@ -38,7 +38,6 @@ from views import (
     make_cross_subset_overview,
     make_hero_stats,
     make_leaderboard,
-    make_metric_reference,
     make_overview_images,
     make_winrate_image,
 )
@@ -129,7 +128,12 @@ def _render_info_boxes(
                     scale=0,
                     min_width=10,
                 )
-                btn.click(lambda cur, k=key: "" if cur == k else k, open_topic, open_topic)
+                btn.click(
+                    lambda cur, k=key: "" if cur == k else k,
+                    open_topic,
+                    open_topic,
+                    api_visibility="private",
+                )
         if not topic:
             return
         with gr.Group(elem_classes="info-panel"):
@@ -145,6 +149,23 @@ def _render_info_boxes(
                 gr.Markdown(info_text[topic], elem_classes="markdown-text-box")
 
 
+def _axis_setter(values: list[str]):
+    """A tab-bar select handler that reports which subset the reader picked.
+
+    Wired on the bar rather than on each tab, because Gradio 6 raises the two events
+    by different routes: the bar's select comes straight out of the tab button's
+    click handler, while a TabItem's own select comes from an effect watching the
+    selected index, and that one does not fire when you come back to the tab that
+    was selected on page load. The event carries the index of the tab clicked, so
+    `values` is the axis in tab order.
+    """
+
+    def setter(evt: gr.SelectData) -> str:
+        return values[evt.index]
+
+    return setter
+
+
 def render_internal_page(page: LeaderboardPage) -> None:
     make_hero_stats(page.data_root)
     _render_info_boxes()
@@ -175,7 +196,7 @@ def render_internal_page(page: LeaderboardPage) -> None:
         )
         metric_tldr = gr.Markdown(OVERVIEW_METRIC_TLDR["Elo"], elem_classes="metric-tldr")
     overview_metric.change(
-        lambda m: OVERVIEW_METRIC_TLDR.get(m, ""), overview_metric, metric_tldr
+        lambda m: OVERVIEW_METRIC_TLDR.get(m, ""), overview_metric, metric_tldr, api_visibility="private"
     )
 
     @gr.render(inputs=[overview_metric])
@@ -196,14 +217,21 @@ def render_internal_page(page: LeaderboardPage) -> None:
     tasks_state = gr.State("all")
     datasets_state = gr.State("all")
 
-    with gr.Tabs(elem_classes="tab-buttons"):
+    # The two content axes, each labelled with the axis it switches and colour-coded
+    # per choice (see the .tab-buttons rules in main.CSS), so they carry the same
+    # weight as the view toggles underneath them.
+    with gr.Tabs(elem_classes=["tab-buttons", "axis-tasks"]) as task_bar:
         for value, label in TASK_LABELS.items():
-            with gr.TabItem(label, id=f"task_{value}") as item:
-                item.select(lambda v=value: v, outputs=tasks_state)
-    with gr.Tabs(elem_classes="tab-buttons"):
+            with gr.TabItem(label, id=f"task_{value}"):
+                pass
+    task_bar.select(_axis_setter(list(TASK_LABELS)), outputs=tasks_state, api_visibility="private")
+    with gr.Tabs(elem_classes=["tab-buttons", "axis-datasets"]) as dataset_bar:
         for value, label in DATASET_LABELS.items():
-            with gr.TabItem(label, id=f"ds_{value}") as item:
-                item.select(lambda v=value: v, outputs=datasets_state)
+            with gr.TabItem(label, id=f"ds_{value}"):
+                pass
+    dataset_bar.select(
+        _axis_setter(list(DATASET_LABELS)), outputs=datasets_state, api_visibility="private"
+    )
 
     with gr.Row(elem_classes="view-toggles"):
         impute_cb = gr.Checkbox(
@@ -213,8 +241,8 @@ def render_internal_page(page: LeaderboardPage) -> None:
             value=False, label="TabArena-Lite (single split)", scale=0, min_width=235, container=False
         )
         gr.Column(scale=1)
-    impute_cb.change(lambda v: "yes" if v else "no", impute_cb, impute_state)
-    lite_cb.change(lambda v: "lite" if v else "all", lite_cb, splits_state)
+    impute_cb.change(lambda v: "yes" if v else "no", impute_cb, impute_state, api_visibility="private")
+    lite_cb.change(lambda v: "lite" if v else "all", lite_cb, splits_state, api_visibility="private")
 
     @gr.render(inputs=[impute_state, splits_state, tasks_state, datasets_state])
     def _render_subset(imp, spl, tsk, dst):
@@ -222,11 +250,11 @@ def render_internal_page(page: LeaderboardPage) -> None:
         lb = LBContainer(data_root=page.data_root, subset=subset, name=subset_name(subset))
         gr.Markdown(subset_blurb(subset, lb.n_datasets), elem_classes="markdown-text")
         make_overview_images(lb)
-        make_metric_reference()
-        with gr.Group():
-            gr.Markdown("## ⭐ Full Leaderboard Table", elem_classes="markdown-text")
-            make_leaderboard(lb)
+        # Figures first, then the table: the win-rate matrix belongs with the other
+        # figures above it, and the table reads as the detailed reference at the end.
         make_winrate_image(lb)
+        # The table card carries its own title and column key (see make_leaderboard).
+        make_leaderboard(lb)
 
     with gr.Accordion("⚙️ Agentic Use & Interpretation", open=False):
         make_agentic_guide()
@@ -312,10 +340,13 @@ def render_beyondarena_page(page: LeaderboardPage) -> None:
     )
 
     subset_state = gr.State("full")
-    with gr.Tabs(elem_classes="tab-buttons"):
+    with gr.Tabs(elem_classes="tab-buttons") as subset_bar:
         for value, label in BEYOND_SUBSET_LABELS.items():
-            with gr.TabItem(label, id=f"beyond_{value}") as item:
-                item.select(lambda v=value: v, outputs=subset_state)
+            with gr.TabItem(label, id=f"beyond_{value}"):
+                pass
+    subset_bar.select(
+        _axis_setter(list(BEYOND_SUBSET_LABELS)), outputs=subset_state, api_visibility="private"
+    )
 
     @gr.render(inputs=[subset_state])
     def _render_beyond_subset(sub):
@@ -323,11 +354,11 @@ def render_beyondarena_page(page: LeaderboardPage) -> None:
         lb = LBContainer(data_root=page.data_root, subset=subset, name=beyond_subset_name(subset))
         gr.Markdown(beyond_subset_blurb(subset, lb.n_datasets), elem_classes="markdown-text")
         make_beyond_subset_figures(lb)
-        make_metric_reference()
-        with gr.Group():
-            gr.Markdown("## ⭐ Full Leaderboard Table", elem_classes="markdown-text")
-            make_leaderboard(lb)
-        make_winrate_image(lb)
+        # Same order as the TabArena page: figures, then the table.
+        # BeyondArena keeps the static win-rate figure; the interactive panel is
+        # TabArena's.
+        make_winrate_image(lb, interactive=False)
+        make_leaderboard(lb)
 
 
 def render_external_page(page: LeaderboardPage) -> None:
diff --git a/pyproject.toml b/pyproject.toml
index f88293104abfdb2cb24a6b598c9bb17ac20bd93f..4b5a99c44e38f75591cc270687bac48e6289bb45 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -7,9 +7,8 @@ requires-python = ">=3.12"
 dependencies = [
     "apscheduler>=3.11.0",
     "gradio-client>=1.3.0",
-    "gradio-leaderboard==0.0.13",
-    "gradio[oauth]==5.49.1",
-    "pandas>=2.2.3",
+    "gradio[mcp,oauth]==6.22.0",
+    "pandas>=2.2.3,<3",
 ]
 [project.optional-dependencies]
 dev = []
diff --git a/requirements.txt b/requirements.txt
index 8f85e02aafdc6f3a2ac410744bf26246a3e4b35a..c69980e2674685f28a34203213a4da7f36c8023d 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,6 +1,8 @@
 APScheduler
-gradio[oauth]==5.49.1
-gradio_leaderboard==0.0.13
+gradio[mcp,oauth]==6.22.0
 gradio_client
-pandas
-python-dateutil
\ No newline at end of file
+# Pinned deliberately: unpinned, a rebuild now picks up pandas 3, which changes
+# string columns from `object` to `str` dtype. Bump it on its own, not alongside
+# an unrelated change.
+pandas==2.3.3
+python-dateutil
diff --git a/views.py b/views.py
index db049dcbdbe6b9477cec0414e5cedc39a16f9c8d..aef8375afdc7c75cae4ad90db5946a11094e89b9 100644
--- a/views.py
+++ b/views.py
@@ -8,16 +8,16 @@ lives here (see ``website_texts.py``).
 from __future__ import annotations
 
 import html
+import math
 import re
 from itertools import groupby
 from pathlib import Path
 
 import gradio as gr
 import pandas as pd
-from gradio_leaderboard import ColumnFilter, Leaderboard, SelectColumns
 
 import website_texts
-from constants import Constants, model_type_color, model_type_emoji
+from constants import Constants, model_type_color, model_type_emoji, variant_color
 from data_loading import (
     BEYOND_SUBSET_LABELS,
     DATASET_SIZE_NOTE,
@@ -25,6 +25,7 @@ from data_loading import (
     LBContainer,
     Subset,
     load_leaderboard_csv,
+    parse_model,
     subset_name,
     unzip_png,
 )
@@ -43,86 +44,537 @@ _IMPUTED_INFO = (
 )
 
 
-def make_leaderboard(lb: LBContainer) -> Leaderboard:
-    df = lb.load_df()
+# Variants a model is evaluated in; all selected by default, and any combination is
+# valid. Reference pipelines (AutoGluon) carry no variant tag of their own but are
+# tuned ensembles, so they belong with those.
+VARIANT_VALUES = ["default", "tuned", "tuned + ensembled"]
+
+# Columns that drive the filters or are folded into another cell; kept on the frame,
+# never rendered as their own column.
+_INTERNAL_COLUMNS = ["TypeName", "RefModel", "Imputed", "_variant", "_base", "_search"]
+# Shown as "(+115/-110)" after the Elo value instead of as a column of its own.
+_CI_COLUMN = "Elo 95% CI"
+_ELO_COLUMN = "Elo [⬆️]"
+# Rendered as a ✔️ badge on the model name, the way the overview does it.
+_VERIFIED_COLUMN = "Verified"
+# Always shown, so they are not offered in the column picker.
+_ALWAYS_SHOWN = ["#", "Type", "Model"]
+
+# Column header -> the metric-reference entry whose text becomes its hover hint, so
+# the tooltips and the documented definitions cannot drift apart.
+_LB_TOOLTIP_METRIC = {
+    _ELO_COLUMN: "🏆 Elo (ranking aggregation)",
+    "Score [⬆️]": "📊 Score",
+    "Improvability (%) [⬇️]": "📉 Improvability (%)",
+    "Rank [⬇️]": "🔢 Average Rank",
+    "Harmonic Rank [⬇️]": "🎯 Harmonic Rank",
+    "Median Train Time (s/1K) [⬇️]": "⏱️ Train / Predict Time (s/1K)",
+    "Median Predict Time (s/1K) [⬇️]": "⏱️ Train / Predict Time (s/1K)",
+    "Imputed (%) [⬇️]": "🧩 Imputed (%)",
+}
+_LB_FIXED_TOOLTIPS = {
+    "#": "Position in this subset's Elo ranking, as published.",
+    "Type": "Model family — see the legend above the table.",
+    "Model": (
+        "The model, its configuration variant in brackets, and ✔️ when the "
+        "implementation was verified by its authors or the maintainers. Links to the "
+        "implementation."
+    ),
+    "Hardware": "The hardware the reported runtimes were measured on.",
+}
+# Which direction is good, per column, for the per-column heatmap. Columns absent
+# from this map (#, Hardware) are left unshaded.
+_LB_HIGHER_IS_BETTER = {
+    _ELO_COLUMN: True,
+    "Score [⬆️]": True,
+    "Rank [⬇️]": False,
+    "Harmonic Rank [⬇️]": False,
+    "Improvability (%) [⬇️]": False,
+    "Median Train Time (s/1K) [⬇️]": False,
+    "Median Predict Time (s/1K) [⬇️]": False,
+    "Imputed (%) [⬇️]": False,
+}
+# Runtimes span orders of magnitude, so shading them linearly paints every model
+# the same green and only the slowest one red. Normalize those in log space.
+_LB_LOG_SCALED = {
+    "Median Train Time (s/1K) [⬇️]",
+    "Median Predict Time (s/1K) [⬇️]",
+}
+# The marker the plot explorers put on an imputed method (see the generated
+# *_explorer.html); reused here so the same thing looks the same everywhere.
+IMPUTED_MARK = "‡"
+
+# Per-column value formatting; everything else falls back to _format_value.
+_LB_FORMATS = {
+    _ELO_COLUMN: lambda v: str(int(round(v))),
+    "Score [⬆️]": lambda v: f"{v:.3f}",
+    "Rank [⬇️]": lambda v: f"{v:.2f}",
+    "Harmonic Rank [⬇️]": lambda v: f"{v:.2f}",
+    "Improvability (%) [⬇️]": lambda v: f"{v:.2f}",
+    "Median Train Time (s/1K) [⬇️]": lambda v: f"{v:.2f}",
+    "Median Predict Time (s/1K) [⬇️]": lambda v: f"{v:.3f}",
+    "Imputed (%) [⬇️]": lambda v: f"{v:.1f}",
+}
 
-    # -- Derived filter columns
-    df["TypeFiler"] = df["TypeName"].apply(lambda m: f"{m} {model_type_emoji[m]}")
-    df["Only Default"] = df["Model"].str.contains("(default)", regex=False)
-    df["Only Tuned"] = df["Model"].str.contains("(tuned)", regex=False)
-    df["Only Tuned + Ensembled"] = df["Model"].str.contains(
-        "(tuned + ensembled)", regex=False
-    ) | df["Model"].str.contains("AutoGluon", regex=False)
-
-    filter_columns = [
-        ColumnFilter("TypeFiler", type="checkboxgroup", label="🤖 Model Types"),
-        ColumnFilter("Only Default", type="boolean", default=False),
-        ColumnFilter("Only Tuned", type="boolean", default=False),
-        ColumnFilter("Only Tuned + Ensembled", type="boolean", default=False),
+
+def _column_tooltip(column: str) -> str | None:
+    """Hover hint for a leaderboard column header, or None when there is nothing to add."""
+    if column in _LB_FIXED_TOOLTIPS:
+        return _LB_FIXED_TOOLTIPS[column]
+    name = _LB_TOOLTIP_METRIC.get(column)
+    if not name:
+        return None
+    for metric in website_texts.METRICS:
+        if metric["name"] == name:
+            return f"{metric['details']}  ·  Why we use it: {metric['why']}"
+    return None
+
+
+def _format_value(column: str, value) -> str:
+    """Display string for one cell; `data-sort` keeps the raw value for sorting."""
+    if value is None or (isinstance(value, float) and pd.isna(value)):
+        return "–"
+    formatter = _LB_FORMATS.get(column)
+    if formatter and isinstance(value, (int, float)) and not isinstance(value, bool):
+        return formatter(value)
+    if isinstance(value, float) and value.is_integer():
+        return str(int(value))
+    return str(value)
+
+
+def filter_leaderboard(
+    df: pd.DataFrame,
+    *,
+    models: list[str],
+    variants: list[str],
+    show_imputed: bool,
+    search: str = "",
+) -> pd.DataFrame:
+    """Apply the table's row filters.
+
+    Pure so it can be tested without a Gradio context. `df` carries the internal
+    ``_base`` / ``_variant`` / ``Imputed`` / ``_search`` columns added by
+    :func:`_prepare_leaderboard`. `models` is the selected individual models: family
+    chips are bulk selectors over that list, not a filter of their own. Column
+    selection is applied when rendering, not here.
+    """
+    sub = df[df["_base"].isin(models)]
+    # Only narrow when the reader has actually deselected something, so a row whose
+    # variant we failed to classify can never be dropped silently.
+    if variants is not None and set(variants) != set(VARIANT_VALUES):
+        sub = sub[sub["_variant"].isin(variants)]
+    if not show_imputed:
+        sub = sub[~sub["Imputed"].astype(bool)]
+    term = (search or "").strip().lower()
+    if term:
+        sub = sub[sub["_search"].str.contains(re.escape(term), regex=True)]
+    return sub
+
+
+def _prepare_leaderboard(raw: pd.DataFrame) -> pd.DataFrame:
+    """Add the internal columns the filters and the renderer need."""
+    df = raw.copy()
+    parsed = [parse_model(str(m)) for m in raw["Model"]]
+    df["_base"] = [base for base, _, _ in parsed]
+    df["_variant"] = [
+        "tuned + ensembled" if not variant and "AutoGluon" in base else variant
+        for base, variant, _ in parsed
     ]
+    df["_search"] = [
+        f"{base} {variant} {type_name}".lower()
+        for (base, variant, _), type_name in zip(parsed, raw["TypeName"], strict=True)
+    ]
+    return df
 
-    # -- Imputation column (only meaningful when some rows are imputed)
-    if df["Imputed"].any():
-        df["Imputed"] = df["Imputed"].replace({True: "Imputed", False: "Not Imputed"})
-        filter_columns.append(
-            ColumnFilter(
-                "Imputed",
-                type="checkboxgroup",
-                label="(Not) Imputed Models",
-                info=_IMPUTED_INFO,
+
+def leaderboard_families(df: pd.DataFrame) -> list[tuple[str, list[str]]]:
+    """(family, models) pairs present in `df`, in the legend's family order."""
+    families = []
+    for type_name in model_type_emoji:
+        models = sorted(set(df.loc[df["TypeName"] == type_name, "_base"]))
+        if models:
+            families.append((type_name, models))
+    return families
+
+
+def fam_chip_colors(families: list[tuple[str, list[str]]]) -> str:
+    """`--fam` per family, so each chiprow carries its family's colour."""
+    rules = [
+        f".ta-fam-{re.sub(r'[^a-z0-9]+', '-', type_name.lower())}"
+        f"{{--fam:{model_type_color.get(type_name, '#9e9e9e')};}}"
+        for type_name, _ in families
+    ]
+    # The variant toggles take their colours from the Leaderboard Overview explorer.
+    # Generated from the same list the choices come from, so the nth-of-type index
+    # cannot drift from the option order.
+    rules += [
+        f".ta-variants label:nth-of-type({i + 1}){{--fam:{variant_color[value]};}}"
+        for i, value in enumerate(VARIANT_VALUES)
+        if value in variant_color
+    ]
+    return "".join(rules)
+
+
+def _group_handler(index: int, models: list[str], render):
+    """A model chip group changed: bring its family chip in line, redraw the table."""
+
+    def handler(*values):
+        selected = values[index] or []
+        return gr.update(value=set(selected) == set(models)), render(*values)
+
+    return handler
+
+
+def _family_handler(index: int, models: list[str], render):
+    """A family chip was toggled: select or clear all of its models, redraw the table."""
+
+    def handler(checked, *values):
+        updated = list(values)
+        updated[index] = list(models) if checked else []
+        return gr.update(value=updated[index]), render(*updated)
+
+    return handler
+
+
+def _heatmap_bounds(df: pd.DataFrame, columns: list[str]) -> dict[str, tuple[float, float]]:
+    """Per-column (lo, hi) over the rendered rows, in the scale used for shading."""
+    bounds = {}
+    for column in columns:
+        if column not in _LB_HIGHER_IS_BETTER or column not in df.columns:
+            continue
+        values = pd.to_numeric(df[column], errors="coerce").dropna()
+        if column in _LB_LOG_SCALED:
+            values = values[values > 0]
+            values = values.apply(math.log10)
+        if len(values) < 2 or values.min() == values.max():
+            continue
+        bounds[column] = (float(values.min()), float(values.max()))
+    return bounds
+
+
+def _heatmap_style(column: str, value: float, bounds: dict[str, tuple[float, float]]) -> str:
+    """Inline background for one shaded cell, or "" when the column is not shaded."""
+    if column not in bounds:
+        return ""
+    lo, hi = bounds[column]
+    scaled = value
+    if column in _LB_LOG_SCALED:
+        if value <= 0:
+            return ""
+        scaled = math.log10(value)
+    frac_best = (scaled - lo) / (hi - lo)
+    if not _LB_HIGHER_IS_BETTER[column]:
+        frac_best = 1 - frac_best
+    frac_best = max(0.0, min(1.0, frac_best))
+    return f' style="background:{_interp_color(1 - frac_best)};color:#f7f7f7;"'
+
+
+def _model_cell(row: pd.Series) -> str:
+    """The Model cell, styled like the cross-subset overview's."""
+    color = model_type_color.get(row["TypeName"], "#9e9e9e")
+    _, variant, url = parse_model(str(row["Model"]))
+    name = html.escape(row["_base"])
+    if variant:
+        name += f' <span class="ta-variant">({html.escape(variant)})</span>'
+    if str(row.get(_VERIFIED_COLUMN, "")).strip() == "✔️":
+        name += ' <span class="ta-verified" title="Verified implementation">✔️</span>'
+    if bool(row.get("Imputed", False)):
+        name += (
+            f' <span class="ta-imp" title="Some results are imputed">{IMPUTED_MARK}</span>'
+        )
+    if url:
+        return (
+            f'<a href="{html.escape(url, quote=True)}" target="_blank" rel="noopener" '
+            f'style="color:{color};font-weight:600;">{name}'
+            f'<span class="ta-link-icon">↗</span></a>'
+        )
+    return f'<span style="color:{color};font-weight:600;">{name}</span>'
+
+
+def leaderboard_table_html(df: pd.DataFrame, columns: list[str], table_id: str) -> str:
+    """Render the leaderboard as a sortable HTML table in the overview's style.
+
+    Every header is clickable (``taSortTable`` in ``main.py``'s head) and carries the
+    column's definition as a hover hint. Numeric cells put the raw value in
+    ``data-sort`` and the formatted one in the text, which is what lets the Elo cell
+    show its confidence interval without that text taking part in the sort.
+    """
+    metric_columns = [c for c in columns if c not in _ALWAYS_SHOWN]
+    header = [
+        '<th class="ta-th-sort ta-th-info" data-type="num" title="'
+        + html.escape(_LB_FIXED_TOOLTIPS["#"], quote=True)
+        + '">#</th>',
+        '<th class="ta-th-sort ta-th-info" title="'
+        + html.escape(_LB_FIXED_TOOLTIPS["Type"], quote=True)
+        + '">Type</th>',
+        '<th class="ta-th-sort ta-th-info ta-th-left" title="'
+        + html.escape(_LB_FIXED_TOOLTIPS["Model"], quote=True)
+        + '">Model</th>',
+    ]
+    for column in metric_columns:
+        tooltip = _column_tooltip(column)
+        label = html.escape(column)
+        marker = ""
+        if column == _ELO_COLUMN:
+            label += ' <span class="ta-ci-head">(95% CI)</span>'
+            # Tells taExportTable to split the interval into its own CSV column.
+            marker = ' data-col="elo"'
+        classes = "ta-th-sort" + (" ta-th-info" if tooltip else "")
+        title = f' title="{html.escape(tooltip, quote=True)}"' if tooltip else ""
+        header.append(
+            f'<th class="{classes}" data-type="num"{marker}{title}>{label}</th>'
+        )
+
+    # Medals mark the top three by Elo, as in the overview. They stay attached to the
+    # Elo column when the reader sorts by something else, which is what they mean.
+    elo_rank = {}
+    if _ELO_COLUMN in df.columns:
+        ranked = df.dropna(subset=[_ELO_COLUMN]).sort_values(_ELO_COLUMN, ascending=False)
+        elo_rank = {idx: i + 1 for i, idx in enumerate(ranked.index[:3])}
+
+    bounds = _heatmap_bounds(df, metric_columns)
+
+    body = []
+    for idx, row in df.iterrows():
+        color = model_type_color.get(row["TypeName"], "#9e9e9e")
+        cells = [
+            f'<td class="ta-num" data-sort="{row["#"]}">{row["#"]}</td>',
+            f'<td class="ta-type-cell" data-sort="{html.escape(str(row["TypeName"]), quote=True)}">'
+            f'<span class="ta-pill" style="background:{color}22;color:{color};'
+            f'border:1px solid {color}66;">{row["Type"]}</span></td>',
+            # data-export="text": the CSV should carry the displayed name, not the
+            # lowercased key this cell sorts on.
+            f'<td class="ta-model-cell" data-export="text" '
+            f'data-sort="{html.escape(row["_base"].lower(), quote=True)}">'
+            f"{_model_cell(row)}</td>",
+        ]
+        for column in metric_columns:
+            value = row.get(column)
+            text = _format_value(column, value)
+            if text == "–":
+                cells.append('<td class="ta-na">–</td>')
+                continue
+            numeric = isinstance(value, (int, float)) and not isinstance(value, bool)
+            shade = _heatmap_style(column, float(value), bounds) if numeric else ""
+            if column == _ELO_COLUMN:
+                medal = _MEDALS.get(elo_rank.get(idx), "")
+                ci = row.get(_CI_COLUMN)
+                interval, ci_attr = "", ""
+                if ci is not None and not pd.isna(ci):
+                    interval = f' <span class="ta-ci">({html.escape(str(ci))})</span>'
+                    ci_attr = f' data-ci="{html.escape(str(ci), quote=True)}"'
+                cells.append(
+                    f'<td class="ta-num" data-sort="{value}"{ci_attr}{shade}>'
+                    f'<span class="ta-cell"><span class="ta-medal">{medal}</span>'
+                    f'<span class="ta-val">{text}{interval}</span></span></td>'
+                )
+                continue
+            sort_key = value if numeric else html.escape(str(value), quote=True)
+            cells.append(f'<td class="ta-num" data-sort="{sort_key}"{shade}>{text}</td>')
+        body.append(f"<tr>{''.join(cells)}</tr>")
+
+    caption = f"{len(df)} row{'' if len(df) == 1 else 's'}"
+    shaded = "green is better, red is worse, per column" if bounds else "unshaded"
+    return (
+        f'<div class="ta-scroll"><table class="ta-overview ta-lbtable" id="{table_id}">'
+        f"<thead><tr>{''.join(header)}</tr></thead>"
+        f"<tbody>{''.join(body)}</tbody></table></div>"
+        f'<p class="ta-cap">{caption} · click a column header to sort, hover one for its '
+        f"definition · {shaded} · {IMPUTED_MARK} marks a model with imputed results</p>"
+    )
+
+
+def make_leaderboard(lb: LBContainer) -> None:
+    """The full leaderboard table for one subset.
+
+    The table is a generated artifact (`leaderboard_table.html`, built by
+    `tabarena.plot.interactive.leaderboard_table`), embedded the same way as the
+    other interactive plots. It lives upstream so that it reuses the explorers'
+    family and variant colours, chip components and imputation markers rather than
+    reimplementing them here, where they drifted.
+
+    Subsets whose artifacts predate it fall back to :func:`make_leaderboard_gradio`,
+    this app's own table.
+    """
+    content = lb.html_content("leaderboard_table")
+    if content is None:
+        make_leaderboard_gradio(lb)
+        return
+    with gr.Column(elem_classes="ta-lb"):
+        gr.HTML(
+            '<div class="ta-lb-bar">'
+            '<span class="ta-lb-title">⭐ Full Leaderboard Table</span>'
+            '<span class="ta-lb-sub">every column sorts · hover a header for what it '
+            "means · filter by family, model or variant</span>"
+            # The frame owns the download: only it knows the current filters,
+            # columns and sort order (see main.taLeaderboardCsv).
+            '<button class="ta-viewbtn ta-exportbtn ta-csvbtn" '
+            'onclick="taLeaderboardCsv(this)" '
+            'title="Download the rows and columns shown, in the current sort order">'
+            "Download CSV</button></div>"
+        )
+        gr.HTML(
+            _interactive_plot_iframe(
+                content, f"Full leaderboard table — {lb.name}", height=1100
             )
         )
-    else:
+
+
+def make_leaderboard_gradio(lb: LBContainer) -> None:
+    """The fallback table, rendered by this app rather than embedded.
+
+    Kept for subsets whose artifacts were generated before `leaderboard_table.html`
+    existed. :func:`make_leaderboard` is the current path; prefer fixing the
+    upstream generator over this.
+
+    Replaces the third-party `gradio_leaderboard` widget (which pinned Gradio < 6)
+    with the same hand-rolled HTML table the cross-subset overview uses: type pills,
+    dotted-underline model links, tooltip-carrying headers, sticky header and scroll
+    box. Model selection mirrors the plot explorers' edit view — a family chip above
+    its models, in the family's colour. Sorting and CSV export are client-side
+    (`taSortTable` / `taExportTable`); the filters round-trip to rebuild the HTML.
+    """
+    raw = lb.load_df()
+    has_imputed = bool(raw["Imputed"].any())
+    df = _prepare_leaderboard(raw)
+    if not has_imputed:
         df = df.drop(columns=["Imputed (%) [⬇️]"])
 
-    # -- Color model names by type so the table shares the overview's hue.
-    def _color_model(row: pd.Series) -> str:
-        color = model_type_color.get(row["TypeName"], "#dddddd")
-        link = re.match(r"\[(.*?)\]\((.*?)\)(.*)", row["Model"])
-        if link:
-            text, url, rest = link.groups()
-            return (
-                f'<a href="{url}" target="_blank" rel="noopener" '
-                f'style="color:{color};font-weight:600;">{text}</a>{rest}'
+    # The CI and Verified columns are folded into the Elo and Model cells.
+    renderable = [
+        c
+        for c in df.columns
+        if c not in _INTERNAL_COLUMNS + [_CI_COLUMN, _VERIFIED_COLUMN]
+    ]
+    optional = [c for c in renderable if c not in _ALWAYS_SHOWN]
+    families = leaderboard_families(df)
+    table_id = f"lb-{lb.subset.rel_path.replace('/', '-')}"
+
+    def render(*values) -> str:
+        """Redraw the table from the controls' values, positional as Gradio passes them."""
+        count = len(families)
+        models = [m for group in values[:count] for m in (group or [])]
+        variants, show_imputed, columns, search = values[count : count + 4]
+        sub = filter_leaderboard(
+            df,
+            models=models,
+            variants=variants,
+            show_imputed=show_imputed or not has_imputed,
+            search=search,
+        )
+        chosen = [c for c in renderable if c in set(columns or []) or c in _ALWAYS_SHOWN]
+        return leaderboard_table_html(sub, chosen, table_id)
+
+    with gr.Column(elem_classes="ta-lb"):
+        gr.HTML(
+            # The table borrows the overview's stylesheet; inject it here rather than
+            # relying on the legend, which this card no longer draws (the family chips
+            # below say the same thing, in colour).
+            _OVERVIEW_CSS
+            + f"<style>{fam_chip_colors(families)}</style>"
+            + '<div class="ta-lb-bar">'
+            '<span class="ta-lb-title">⭐ Full Leaderboard Table</span>'
+            '<span class="ta-lb-sub">every column sorts · hover a header for what it '
+            "means</span>"
+            # Same green as the figure panels' export controls (.ta-exportbtn).
+            f'<button class="ta-viewbtn ta-exportbtn ta-csvbtn" '
+            f'onclick="taExportTable(\'{table_id}\')" '
+            'title="Download the rows and columns currently shown, in the current sort '
+            'order">⬇️ CSV</button>'
+            "</div>"
+        )
+
+        # Model selection, laid out like the Pareto explorer's chiprows: one row per
+        # family, the family chip toggling all of its models at once.
+        fam_toggles, model_groups = [], []
+        for type_name, models in families:
+            fam_class = f"ta-fam-{re.sub(r'[^a-z0-9]+', '-', type_name.lower())}"
+            with gr.Row(elem_classes="ta-chiprow"):
+                fam_toggles.append(
+                    gr.Checkbox(
+                        value=True,
+                        label=f"{model_type_emoji.get(type_name, '')} {type_name} ×{len(models)}",
+                        show_label=False,
+                        container=False,
+                        interactive=True,
+                        elem_classes=["ta-famchip", fam_class],
+                        scale=0,
+                        min_width=210,
+                    )
+                )
+                model_groups.append(
+                    gr.CheckboxGroup(
+                        choices=models,
+                        value=models,
+                        show_label=False,
+                        container=False,
+                        interactive=True,
+                        elem_classes=["ta-chips", fam_class],
+                        scale=1,
+                    )
+                )
+
+        with gr.Row(elem_classes="ta-lb-controls"):
+            variant = gr.CheckboxGroup(
+                choices=VARIANT_VALUES,
+                value=VARIANT_VALUES,
+                label="⚙️ Variants",
+                interactive=True,
+                elem_classes=["ta-btns", "ta-variants"],
+                scale=3,
+                min_width=290,
             )
-        return f'<span style="color:{color};font-weight:600;">{row["Model"]}</span>'
-
-    df["Model"] = df.apply(_color_model, axis=1)
-
-    # -- Column datatypes (read straight from dtypes; transposing collapses to
-    #    object and mistypes every numeric column as markdown).
-    datatypes = []
-    for dtype in df.dtypes:
-        if pd.api.types.is_bool_dtype(dtype):  # check bool first (it is also "numeric")
-            datatypes.append("bool")
-        elif pd.api.types.is_numeric_dtype(dtype):
-            datatypes.append("number")
-        else:
-            datatypes.append("markdown")
-
-    return Leaderboard(
-        elem_id=f"lb_{lb.subset.rel_path.replace('/', '_')}",
-        value=df,
-        datatype=datatypes,
-        select_columns=SelectColumns(
-            default_selection=list(df.columns),
-            cant_deselect=["Type", "Model"],
-            label="Select Columns to Display:",
-        ),
-        hide_columns=[
-            "TypeName",
-            "TypeFiler",
-            "RefModel",
-            "Only Default",
-            "Only Tuned",
-            "Only Tuned + Ensembled",
-            "Imputed",
-        ],
-        search_columns=["Model", "TypeName"],
-        filter_columns=filter_columns,
-        bool_checkboxgroup_label="Custom Views (exclusive, only toggle one at a time):",
-        height=800,
-    )
+            show_imputed = gr.Checkbox(
+                value=True,
+                label=f"{IMPUTED_MARK} Include imputed",
+                info=_IMPUTED_INFO,
+                interactive=True,
+                visible=has_imputed,
+                elem_classes=["ta-btns", "ta-btns-imputed"],
+                scale=1,
+                min_width=190,
+            )
+            columns = gr.Dropdown(
+                choices=optional,
+                value=optional,
+                multiselect=True,
+                label="📋 Columns",
+                info="# / Type / Model always shown",
+                interactive=True,
+                scale=2,
+                min_width=220,
+            )
+            search = gr.Textbox(
+                label="🔍 Search",
+                placeholder="model or type…",
+                interactive=True,
+                scale=1,
+                min_width=150,
+            )
+
+        state = [*model_groups, variant, show_imputed, columns, search]
+        table = gr.HTML(render(*[c.value for c in state]))
+
+        # `.input` rather than `.change`: a family chip rewrites its group's value and
+        # each group rewrites its family chip, so reacting to programmatic changes too
+        # would let the two bounce off each other.
+        for index, (models, group, fam) in enumerate(
+            zip([m for _, m in families], model_groups, fam_toggles, strict=True)
+        ):
+            group.input(
+                _group_handler(index, models, render),
+                state,
+                [fam, table],
+                api_visibility="private",
+            )
+            fam.input(
+                _family_handler(index, models, render),
+                [fam, *state],
+                [group, table],
+                api_visibility="private",
+            )
+        for control in (variant, show_imputed, columns, search):
+            control.input(render, state, table, api_visibility="private")
 
 
 # --------------------------------------------------------------------------- #
@@ -187,7 +639,6 @@ def _switchable_figure(
             label=label,
             height=height,
             show_label=True,
-            show_share_button=True,
         )
         return
 
@@ -225,7 +676,6 @@ def _switchable_figure(
             gr.Image(
                 value=lb.image_path(img_name),
                 show_label=False,
-                show_share_button=True,
             )
 
 
@@ -256,15 +706,29 @@ def make_overview_images(lb: LBContainer) -> None:
     )
 
 
-def make_winrate_image(lb: LBContainer) -> None:
-    # Uses ``lb.name`` (the caller-supplied subset name) so it is benchmark-agnostic and reusable
-    # across both the TabArena and BeyondArena tabs.
+def make_winrate_image(lb: LBContainer, *, interactive: bool = True) -> None:
+    """The win-rate matrix for one subset.
+
+    Uses ``lb.name`` (the caller-supplied subset name) so it is benchmark-agnostic and
+    reusable across the TabArena and BeyondArena tabs. With `interactive` it gets the
+    same interactive / static / paper panel as the other figures, falling back to the
+    PNG on its own for subsets whose artifacts predate ``winrate_explorer.html``.
+    BeyondArena passes ``interactive=False`` and stays on the static figure.
+    """
+    if interactive:
+        _switchable_figure(
+            lb,
+            html_name="winrate_explorer",
+            img_name="winrate_matrix",
+            label=f"Win-rate Matrix [{lb.name}]",
+            height=800,
+        )
+        return
     gr.Image(
         lb.image_path("winrate_matrix"),
         label=f"Win-rate Matrix [{lb.name}]",
         show_label=True,
         height=800,
-        show_share_button=True,
     )
 
 
@@ -297,7 +761,6 @@ _COLUMN_TOOLTIPS: dict[str, str] = {
 # Top-3 per column get a medal in a fixed-width slot reserved in every number cell, so the
 # numbers line up whether or not a medal is present (see `.ta-medal`/`.ta-val` in `_OVERVIEW_CSS`).
 _MEDALS = {1: "🥇", 2: "🥈", 3: "🥉"}
-_VARIANT_RE = re.compile(r"\((default|tuned \+ ensembled|tuned)\)")
 
 # Metrics selectable in the overview: label -> (csv column, higher_is_better, formatter).
 # Order = display order in the selector; the two headline metrics come first.
@@ -320,22 +783,11 @@ OVERVIEW_METRIC_TLDR = {
 }
 
 
-def _parse_model(model: str) -> tuple[str, str, str | None]:
-    """Split a Model cell into (base name, variant, url)."""
-    link = re.match(r"\[(.*?)\]\((.*?)\)", model)
-    text, url = (link.group(1), link.group(2)) if link else (model, None)
-    text = text.split("[")[0].strip()  # drop any [X% IMPUTED] tag
-    variant_match = _VARIANT_RE.search(text)
-    variant = variant_match.group(1) if variant_match else ""
-    base = _VARIANT_RE.sub("", text).strip()
-    return base, variant, url
-
-
 def _subset_best(df: pd.DataFrame, column: str, higher_is_better: bool) -> pd.DataFrame:
     """Best variant per model in a subset for `column`; excludes reference pipelines."""
     df = df[~df["TypeName"].isin([Constants.reference])].copy()
     df = df.dropna(subset=[column])
-    parsed = df["Model"].map(_parse_model)
+    parsed = df["Model"].map(parse_model)
     df["base"] = [p[0] for p in parsed]
     df["variant"] = [p[1] for p in parsed]
     df["url"] = [p[2] for p in parsed]
@@ -380,7 +832,9 @@ _OVERVIEW_CSS = """
 .ta-overview thead tr:nth-child(2) th { top: 33px; }
 .ta-overview .ta-group-h { border-bottom: 2px solid #ffffff33; }
 .ta-overview td.ta-model-cell { text-align: left; white-space: nowrap; }
-.ta-overview .ta-variant { opacity: 0.5; font-weight: 400; font-size: 0.9em; }
+/* Pinned to the same grey as the leaderboard table's variant tag (--muted there),
+   so a "(tuned + ensembled)" reads identically in both tables. */
+.ta-overview .ta-variant { color: #9b9a92; font-weight: 400; font-size: 0.9em; }
 /* Numbers match the rest of the table text in size; bold keeps them emphasized. */
 .ta-overview td.ta-num { font-weight: 600; }
 /* Reserve a fixed-width medal slot in every number cell so the digits line up in their
@@ -563,47 +1017,10 @@ def make_cross_subset_overview(data_root: Path, metric: str = "Elo") -> gr.HTML:
 
 
 # --------------------------------------------------------------------------- #
-# Metric reference and agentic guide
+# Agentic guide
 # --------------------------------------------------------------------------- #
 
 
-def make_metric_reference() -> None:
-    """Compact column key — pills that reveal a single shared panel (one open at a time).
-
-    Uses a CSS-only radio group (no JS): selecting a pill checks its hidden radio, which
-    reveals just that definition in the shared panel below; the ✕ rechecks a hidden
-    "none" radio to close.
-    """
-    radios = ['<input type="radio" name="ta-mk" id="ta-mk-none" class="ta-mk-radio" checked>']
-    labels, defs, rules = [], [], []
-    for i, metric in enumerate(website_texts.METRICS):
-        rid = f"ta-mk-{i}"
-        name = html.escape(metric["name"])
-        definition = html.escape(f"{metric['details']}  ·  Why we use it: {metric['why']}")
-        radios.append(f'<input type="radio" name="ta-mk" id="{rid}" class="ta-mk-radio">')
-        labels.append(f'<label for="{rid}" class="ta-mk-chip">{name}</label>')
-        defs.append(
-            f'<div class="ta-mk-def" id="{rid}-def">'
-            f'<label for="ta-mk-none" class="ta-mk-close" title="Close">✕</label>{definition}</div>'
-        )
-        rules.append(f"#{rid}:checked ~ .ta-mk-panel #{rid}-def{{display:block}}")
-        rules.append(
-            f"#{rid}:checked ~ .ta-mk-chips label[for='{rid}']"
-            "{background:rgba(110,140,245,0.22);border-color:rgba(110,140,245,0.9);color:#cdd7ff}"
-        )
-    gr.HTML(
-        "<style>" + "".join(rules) + "</style>"
-        '<div class="ta-metric-key">' + "".join(radios)
-        + '<div class="ta-mk-head">'
-        '<span class="ta-mk-title">📊 Column key</span>'
-        '<span class="ta-mk-hint">ranked by the <b>Elo</b> aggregation · click any column for its definition</span>'
-        "</div>"
-        + '<div class="ta-mk-chips">' + "".join(labels) + "</div>"
-        + '<div class="ta-mk-panel">' + "".join(defs) + "</div>"
-        + "</div>"
-    )
-
-
 def make_agentic_guide() -> None:
     gr.Markdown(website_texts.AGENTIC_GUIDE, elem_classes="markdown-text-box")
 
@@ -614,7 +1031,7 @@ def make_hero_stats(data_root: Path) -> gr.HTML:
     n_datasets = lb.n_datasets or "—"
     df = lb.load_df()
     df = df[~df["TypeName"].isin([Constants.reference])]
-    n_models = len({_parse_model(m)[0] for m in df["Model"]})
+    n_models = len({parse_model(m)[0] for m in df["Model"]})
     paper = "https://tabarena.ai/paper-tabular-ml-iid-study"
     code = "https://tabarena.ai/code"
 
@@ -681,7 +1098,7 @@ def make_beyond_hero_stats(data_root: Path) -> gr.HTML:
     n_datasets = lb.n_datasets or "—"
     df = lb.load_df()
     df = df[~df["TypeName"].isin([Constants.reference])]
-    n_models = len({_parse_model(m)[0] for m in df["Model"]})
+    n_models = len({parse_model(m)[0] for m in df["Model"]})
     # Curated subsets = every subset tab except the "full" (whole-benchmark) view.
     n_subsets = len(BEYOND_SUBSET_LABELS) - 1
     paper = "https://arxiv.org/abs/2606.30410"
@@ -731,7 +1148,6 @@ def make_beyond_overview_figure(data_root: Path) -> None:
                 label=label,
                 show_label=True,
                 height=520,
-                show_share_button=True,
             )
             shown = True
     if not shown:
@@ -750,12 +1166,10 @@ def make_beyond_subset_figures(lb: LBContainer) -> None:
         label=f"Leaderboard Overview [{name}]",
         show_label=True,
         height=320,
-        show_share_button=True,
     )
     gr.Image(
         lb.image_path("pareto_front_improvability_vs_time_infer"),
         label=f"Inference Time Pareto Front [{name}]",
         show_label=True,
         height=450,
-        show_share_button=True,
     )
diff --git a/website_texts.py b/website_texts.py
index a5d0239a959b0af923f3fbb619729441413b9268..bf9c144bf377c33c3d0bfdc0e1b185ad8bc3455a 100644
--- a/website_texts.py
+++ b/website_texts.py
@@ -274,6 +274,50 @@ accuracy (**Elo** / **Score**), robustness (**Average** & **Harmonic Rank**), an
 
 These are starting heuristics, not guarantees — always validate on your own data. These aggregations and their
 trade-offs are motivated in our [paper](https://arxiv.org/abs/2506.16791).
+
+#### Reading the numbers programmatically
+
+Every leaderboard on this page is a CSV in the Space repo, so a script or agent can read the same values the
+table shows without scraping it:
+
+```bash
+curl -sL https://huggingface.co/spaces/TabArena/leaderboard/resolve/main/data/imputation_yes/splits_all/tasks_all/datasets_all/website_leaderboard.csv
+```
+
+Swap the path segments for another subset (`imputation_{yes,no}` / `splits_{all,lite}` /
+`tasks_{all,classification,regression,binary,multiclass}` / `datasets_{all,small,medium}`); BeyondArena lives
+under `data_beyondarena/subsets/{subset}/`.
+
+#### Connecting an AI assistant (MCP)
+
+This Space is also an **MCP server**, so an assistant can query the leaderboard as a tool instead of you
+pasting numbers into a chat. Four tools are exposed:
+
+- `list_leaderboards` — the available benchmarks, their valid subset values, and the keys every record carries.
+  Worth calling first.
+- `get_tabarena_leaderboard` — ranked results for one IID subset (tasks, datasets, imputation, splits).
+- `get_beyondarena_leaderboard` — the same for a BeyondArena subset.
+- `get_pareto_frontier` — the accuracy-versus-time trade-off. Ask this rather than reading the top row: the
+  highest-Elo model is often far slower than one just behind it, and this returns the outright best, the
+  models nothing beats on both axes at once, and the best model that fits a train- or predict-time budget.
+
+All four take a `kind` of `models` (the default), `systems`, or `all` — individual models like TabPFN, whole
+AutoML systems like AutoGluon, or both.
+
+**Claude Code** — add the server once, then use it in any session:
+
+```bash
+claude mcp add --transport http tabarena https://tabarena-leaderboard.hf.space/gradio_api/mcp/
+```
+
+Add `--scope user` to make it available across all your projects. `claude mcp list` shows `✔ Connected`
+when it worked, and `/mcp` inside a session lists the tools. The trailing slash on the URL matters.
+
+**Claude Desktop and claude.ai** — add the same URL as a custom connector in settings.
+
+Any other MCP client works too; the endpoint speaks streamable HTTP. Then just ask, for example:
+*"Using TabArena, what's the best tabular model for a 5k-row regression dataset if inference has to stay
+under 0.5 s per 1000 rows?"*
 """
 
 # --- BeyondArena tab -------------------------------------------------------